From 080a2707b92282f68052de8b0cb09543e047ad79 Mon Sep 17 00:00:00 2001
From: wheresmyhair <wheresmyhair@users.noreply.github.com>
Date: Wed, 8 Jan 2025 01:19:48 +0000
Subject: [PATCH] deploy: 9a9957dc19845b0dcba0ac33f0cc6d49758f646f

---
 .buildinfo                                    |    4 +
 .doctrees/about/authors.doctree               |  Bin 0 -> 3096 bytes
 .doctrees/about/changelog.doctree             |  Bin 0 -> 3984 bytes
 .doctrees/about/index.doctree                 |  Bin 0 -> 3320 bytes
 .doctrees/autoapi/index.doctree               |  Bin 0 -> 4132 bytes
 .doctrees/autoapi/lmflow/args/index.doctree   |  Bin 0 -> 583437 bytes
 .../lmflow/datasets/dataset/index.doctree     |  Bin 0 -> 104845 bytes
 .../autoapi/lmflow/datasets/index.doctree     |  Bin 0 -> 95210 bytes
 .../multi_modal_dataset/index.doctree         |  Bin 0 -> 45539 bytes
 .doctrees/autoapi/lmflow/index.doctree        |  Bin 0 -> 8960 bytes
 .../lmflow/models/auto_model/index.doctree    |  Bin 0 -> 9093 bytes
 .../lmflow/models/base_model/index.doctree    |  Bin 0 -> 8682 bytes
 .../lmflow/models/decoder_model/index.doctree |  Bin 0 -> 9995 bytes
 .../encoder_decoder_model/index.doctree       |  Bin 0 -> 10231 bytes
 .../models/hf_decoder_model/index.doctree     |  Bin 0 -> 101597 bytes
 .../hf_encoder_decoder_model/index.doctree    |  Bin 0 -> 63023 bytes
 .../models/hf_model_mixin/index.doctree       |  Bin 0 -> 81018 bytes
 .../hf_text_regression_model/index.doctree    |  Bin 0 -> 73060 bytes
 .doctrees/autoapi/lmflow/models/index.doctree |  Bin 0 -> 3949 bytes
 .../lmflow/models/interfaces/index.doctree    |  Bin 0 -> 3409 bytes
 .../models/interfaces/tunable/index.doctree   |  Bin 0 -> 8013 bytes
 .../models/regression_model/index.doctree     |  Bin 0 -> 8992 bytes
 .../text_regression_model/index.doctree       |  Bin 0 -> 18043 bytes
 .../models/vision2seq_model/index.doctree     |  Bin 0 -> 79434 bytes
 .../vision_encoder/clip_encoder/index.doctree |  Bin 0 -> 33970 bytes
 .../models/vision_encoder/index.doctree       |  Bin 0 -> 7944 bytes
 .../lmflow/optim/adabelief/index.doctree      |  Bin 0 -> 23759 bytes
 .../lmflow/optim/adabound/index.doctree       |  Bin 0 -> 22071 bytes
 .../lmflow/optim/adadelta/index.doctree       |  Bin 0 -> 10182 bytes
 .../lmflow/optim/adagrad/index.doctree        |  Bin 0 -> 10143 bytes
 .../autoapi/lmflow/optim/adam/index.doctree   |  Bin 0 -> 10086 bytes
 .../autoapi/lmflow/optim/adamax/index.doctree |  Bin 0 -> 11854 bytes
 .../autoapi/lmflow/optim/adamp/index.doctree  |  Bin 0 -> 26361 bytes
 .../optim/adamw_schedule_free/index.doctree   |  Bin 0 -> 17928 bytes
 .../autoapi/lmflow/optim/adan/index.doctree   |  Bin 0 -> 50960 bytes
 .../autoapi/lmflow/optim/dummy/index.doctree  |  Bin 0 -> 19815 bytes
 .doctrees/autoapi/lmflow/optim/index.doctree  |  Bin 0 -> 4106 bytes
 .../autoapi/lmflow/optim/lamb/index.doctree   |  Bin 0 -> 22973 bytes
 .../autoapi/lmflow/optim/lars/index.doctree   |  Bin 0 -> 23708 bytes
 .../autoapi/lmflow/optim/nadam/index.doctree  |  Bin 0 -> 12268 bytes
 .../lmflow/optim/novograd/index.doctree       |  Bin 0 -> 12870 bytes
 .../lmflow/optim/optimizers/index.doctree     |  Bin 0 -> 3094 bytes
 .../autoapi/lmflow/optim/radam/index.doctree  |  Bin 0 -> 18078 bytes
 .../optim/sgd_schedule_free/index.doctree     |  Bin 0 -> 17307 bytes
 .../autoapi/lmflow/optim/sgdp/index.doctree   |  Bin 0 -> 28091 bytes
 .../autoapi/lmflow/optim/sophia/index.doctree |  Bin 0 -> 16883 bytes
 .../autoapi/lmflow/optim/yogi/index.doctree   |  Bin 0 -> 17001 bytes
 .../pipeline/auto_pipeline/index.doctree      |  Bin 0 -> 15937 bytes
 .../pipeline/base_aligner/index.doctree       |  Bin 0 -> 12504 bytes
 .../pipeline/base_pipeline/index.doctree      |  Bin 0 -> 8011 bytes
 .../lmflow/pipeline/base_tuner/index.doctree  |  Bin 0 -> 11914 bytes
 .../lmflow/pipeline/dpo_aligner/index.doctree |  Bin 0 -> 29332 bytes
 .../pipeline/dpov2_aligner/index.doctree      |  Bin 0 -> 74023 bytes
 .../lmflow/pipeline/evaluator/index.doctree   |  Bin 0 -> 41337 bytes
 .../lmflow/pipeline/finetuner/index.doctree   |  Bin 0 -> 29804 bytes
 .../autoapi/lmflow/pipeline/index.doctree     |  Bin 0 -> 4032 bytes
 .../lmflow/pipeline/inferencer/index.doctree  |  Bin 0 -> 112250 bytes
 .../iterative_dpo_aligner/index.doctree       |  Bin 0 -> 52740 bytes
 .../pipeline/raft_aligner/index.doctree       |  Bin 0 -> 53167 bytes
 .../pipeline/rm_inferencer/index.doctree      |  Bin 0 -> 61288 bytes
 .../lmflow/pipeline/rm_tuner/index.doctree    |  Bin 0 -> 22665 bytes
 .../utils/dpov2_dataprocessor/index.doctree   |  Bin 0 -> 47038 bytes
 .../utils/dpov2_trainer/index.doctree         |  Bin 0 -> 89960 bytes
 .../lmflow/pipeline/utils/index.doctree       |  Bin 0 -> 3819 bytes
 .../memory_safe_dpov2_align/index.doctree     |  Bin 0 -> 12414 bytes
 .../memory_safe_vllm_inference/index.doctree  |  Bin 0 -> 9704 bytes
 .../pipeline/utils/peft_trainer/index.doctree |  Bin 0 -> 28064 bytes
 .../pipeline/utils/raft_trainer/index.doctree |  Bin 0 -> 481422 bytes
 .../utils/rm_dataprocessor/index.doctree      |  Bin 0 -> 25674 bytes
 .../pipeline/utils/rm_trainer/index.doctree   |  Bin 0 -> 20525 bytes
 .../pipeline/vllm_inferencer/index.doctree    |  Bin 0 -> 92863 bytes
 .../hf_decoder_model/index.doctree            |  Bin 0 -> 30901 bytes
 .../hf_text_regression_model/index.doctree    |  Bin 0 -> 60876 bytes
 .../autoapi/lmflow/tokenization/index.doctree |  Bin 0 -> 3441 bytes
 .../autoapi/lmflow/utils/common/index.doctree |  Bin 0 -> 47565 bytes
 .../lmflow/utils/constants/index.doctree      |  Bin 0 -> 75458 bytes
 .../conversation_template/base/index.doctree  |  Bin 0 -> 199913 bytes
 .../chatglm/index.doctree                     |  Bin 0 -> 6705 bytes
 .../chatml/index.doctree                      |  Bin 0 -> 6664 bytes
 .../deepseek/index.doctree                    |  Bin 0 -> 6724 bytes
 .../conversation_template/gemma/index.doctree |  Bin 0 -> 14508 bytes
 .../conversation_template/hymba/index.doctree |  Bin 0 -> 28551 bytes
 .../utils/conversation_template/index.doctree |  Bin 0 -> 171730 bytes
 .../internlm/index.doctree                    |  Bin 0 -> 6735 bytes
 .../conversation_template/llama/index.doctree |  Bin 0 -> 42119 bytes
 .../conversation_template/phi/index.doctree   |  Bin 0 -> 6585 bytes
 .../conversation_template/qwen/index.doctree  |  Bin 0 -> 8759 bytes
 .../conversation_template/yi/index.doctree    |  Bin 0 -> 6577 bytes
 .../zephyr/index.doctree                      |  Bin 0 -> 22686 bytes
 .../lmflow/utils/data_utils/index.doctree     |  Bin 0 -> 50455 bytes
 .../bloom_flash_attention/index.doctree       |  Bin 0 -> 25636 bytes
 .../gpt2_flash_attention/index.doctree        |  Bin 0 -> 30168 bytes
 .../gpt_neo_flash_attention/index.doctree     |  Bin 0 -> 16719 bytes
 .../utils/flash_attention/index.doctree       |  Bin 0 -> 3751 bytes
 .../llama_flash_attention/index.doctree       |  Bin 0 -> 25552 bytes
 .../triton_flash_attention/index.doctree      |  Bin 0 -> 123188 bytes
 .doctrees/autoapi/lmflow/utils/index.doctree  |  Bin 0 -> 3744 bytes
 .../llava_conversation_lib/index.doctree      |  Bin 0 -> 64596 bytes
 .../autoapi/lmflow/utils/model/index.doctree  |  Bin 0 -> 10686 bytes
 .../lmflow/utils/multimodal/index.doctree     |  Bin 0 -> 12325 bytes
 .../position_interpolation/index.doctree      |  Bin 0 -> 3531 bytes
 .../index.doctree                             |  Bin 0 -> 19420 bytes
 .../lmflow/utils/versioning/index.doctree     |  Bin 0 -> 36060 bytes
 .../autoapi/lmflow/version/index.doctree      |  Bin 0 -> 6597 bytes
 .doctrees/blogs/benchmark.doctree             |  Bin 0 -> 169635 bytes
 .doctrees/blogs/index.doctree                 |  Bin 0 -> 3332 bytes
 .doctrees/environment.pickle                  |  Bin 0 -> 4282851 bytes
 .doctrees/examples/DATASETS.doctree           |  Bin 0 -> 99117 bytes
 .doctrees/examples/TASK_GUIDE.doctree         |  Bin 0 -> 20428 bytes
 .doctrees/examples/checkpoints.doctree        |  Bin 0 -> 8439 bytes
 .../customize_conversation_template.doctree   |  Bin 0 -> 18349 bytes
 .doctrees/examples/finetuning.doctree         |  Bin 0 -> 12319 bytes
 .doctrees/examples/index.doctree              |  Bin 0 -> 6059 bytes
 .doctrees/examples/medical_finetune.doctree   |  Bin 0 -> 6312 bytes
 .doctrees/examples/raft.doctree               |  Bin 0 -> 101107 bytes
 .doctrees/examples/reward_modeling.doctree    |  Bin 0 -> 33651 bytes
 .../supported_conversation_template.doctree   |  Bin 0 -> 100788 bytes
 .doctrees/index.doctree                       |  Bin 0 -> 58437 bytes
 .nojekyll                                     |    0
 _images/IT_sample1.png                        |  Bin 0 -> 523440 bytes
 _images/IT_sample2.png                        |  Bin 0 -> 282807 bytes
 _images/IT_sample3.png                        |  Bin 0 -> 396642 bytes
 _images/IT_sample4.png                        |  Bin 0 -> 334618 bytes
 _images/IT_sample5.png                        |  Bin 0 -> 632293 bytes
 _images/IT_sample6.png                        |  Bin 0 -> 365253 bytes
 _images/IT_sample7.png                        |  Bin 0 -> 578291 bytes
 _images/benchmark-1.png                       |  Bin 0 -> 62190 bytes
 _images/benchmark-2.png                       |  Bin 0 -> 142903 bytes
 _images/nll.png                               |  Bin 0 -> 55999 bytes
 _images/ppl.png                               |  Bin 0 -> 25483 bytes
 _images/raft_idea.PNG                         |  Bin 0 -> 121926 bytes
 _images/raft_reward.PNG                       |  Bin 0 -> 221269 bytes
 _modules/index.html                           |  522 ++
 _modules/lmflow.html                          |  454 ++
 _modules/lmflow/args.html                     | 2622 ++++++++++
 _modules/lmflow/datasets/dataset.html         | 1127 ++++
 .../lmflow/datasets/multi_modal_dataset.html  |  783 +++
 _modules/lmflow/models/auto_model.html        |  472 ++
 _modules/lmflow/models/base_model.html        |  456 ++
 _modules/lmflow/models/decoder_model.html     |  466 ++
 .../lmflow/models/encoder_decoder_model.html  |  466 ++
 _modules/lmflow/models/hf_decoder_model.html  | 1157 +++++
 .../models/hf_encoder_decoder_model.html      |  999 ++++
 _modules/lmflow/models/hf_model_mixin.html    | 1110 ++++
 .../models/hf_text_regression_model.html      |  958 ++++
 .../lmflow/models/interfaces/tunable.html     |  454 ++
 _modules/lmflow/models/regression_model.html  |  455 ++
 .../lmflow/models/text_regression_model.html  |  510 ++
 _modules/lmflow/models/vision2seq_model.html  |  987 ++++
 .../models/vision_encoder/clip_encoder.html   |  719 +++
 _modules/lmflow/optim/adabelief.html          |  649 +++
 _modules/lmflow/optim/adabound.html           |  611 +++
 _modules/lmflow/optim/adadelta.html           |  490 ++
 _modules/lmflow/optim/adagrad.html            |  482 ++
 _modules/lmflow/optim/adam.html               |  493 ++
 _modules/lmflow/optim/adamax.html             |  518 ++
 _modules/lmflow/optim/adamp.html              |  632 +++
 .../lmflow/optim/adamw_schedule_free.html     |  632 +++
 _modules/lmflow/optim/adan.html               |  739 +++
 _modules/lmflow/optim/dummy.html              |  527 ++
 _modules/lmflow/optim/lamb.html               |  588 +++
 _modules/lmflow/optim/lars.html               |  593 +++
 _modules/lmflow/optim/nadam.html              |  523 ++
 _modules/lmflow/optim/novograd.html           |  534 ++
 _modules/lmflow/optim/radam.html              |  623 +++
 _modules/lmflow/optim/sgd_schedule_free.html  |  611 +++
 _modules/lmflow/optim/sgdp.html               |  620 +++
 _modules/lmflow/optim/sophia.html             |  571 ++
 _modules/lmflow/optim/yogi.html               |  573 +++
 _modules/lmflow/pipeline/auto_pipeline.html   |  537 ++
 _modules/lmflow/pipeline/base_aligner.html    |  471 ++
 _modules/lmflow/pipeline/base_pipeline.html   |  453 ++
 _modules/lmflow/pipeline/base_tuner.html      |  470 ++
 _modules/lmflow/pipeline/dpo_aligner.html     |  645 +++
 _modules/lmflow/pipeline/dpov2_aligner.html   |  856 +++
 _modules/lmflow/pipeline/evaluator.html       |  979 ++++
 _modules/lmflow/pipeline/finetuner.html       | 1101 ++++
 _modules/lmflow/pipeline/inferencer.html      | 1181 +++++
 .../pipeline/iterative_dpo_aligner.html       |  746 +++
 _modules/lmflow/pipeline/raft_aligner.html    | 1176 +++++
 _modules/lmflow/pipeline/rm_inferencer.html   |  845 +++
 _modules/lmflow/pipeline/rm_tuner.html        |  683 +++
 .../pipeline/utils/dpov2_dataprocessor.html   |  684 +++
 .../lmflow/pipeline/utils/dpov2_trainer.html  |  714 +++
 .../utils/memory_safe_dpov2_align.html        |  520 ++
 .../utils/memory_safe_vllm_inference.html     |  515 ++
 .../lmflow/pipeline/utils/peft_trainer.html   |  540 ++
 .../lmflow/pipeline/utils/raft_trainer.html   | 4580 +++++++++++++++++
 .../pipeline/utils/rm_dataprocessor.html      |  515 ++
 .../lmflow/pipeline/utils/rm_trainer.html     |  501 ++
 _modules/lmflow/pipeline/vllm_inferencer.html |  866 ++++
 .../lmflow/tokenization/hf_decoder_model.html |  673 +++
 .../hf_text_regression_model.html             |  914 ++++
 _modules/lmflow/utils/common.html             |  650 +++
 _modules/lmflow/utils/constants.html          |  981 ++++
 .../lmflow/utils/conversation_template.html   |  481 ++
 .../utils/conversation_template/base.html     | 1173 +++++
 .../utils/conversation_template/chatglm.html  |  472 ++
 .../utils/conversation_template/chatml.html   |  471 ++
 .../utils/conversation_template/deepseek.html |  473 ++
 .../utils/conversation_template/gemma.html    |  501 ++
 .../utils/conversation_template/hymba.html    |  576 +++
 .../utils/conversation_template/internlm.html |  472 ++
 .../utils/conversation_template/llama.html    |  664 +++
 .../utils/conversation_template/phi.html      |  473 ++
 .../utils/conversation_template/qwen.html     |  505 ++
 .../utils/conversation_template/yi.html       |  471 ++
 .../utils/conversation_template/zephyr.html   |  529 ++
 _modules/lmflow/utils/data_utils.html         |  711 +++
 .../bloom_flash_attention.html                |  552 ++
 .../flash_attention/gpt2_flash_attention.html |  586 +++
 .../gpt_neo_flash_attention.html              |  550 ++
 .../llama_flash_attention.html                |  574 +++
 .../triton_flash_attention.html               | 1333 +++++
 .../lmflow/utils/llava_conversation_lib.html  |  927 ++++
 _modules/lmflow/utils/model.html              |  472 ++
 _modules/lmflow/utils/multimodal.html         |  500 ++
 .../llama_rope_scaled_monkey_patch.html       |  508 ++
 _modules/lmflow/utils/versioning.html         |  555 ++
 _modules/lmflow/version.html                  |  445 ++
 _sources/about/authors.md.txt                 |    4 +
 _sources/about/changelog.md.txt               |   15 +
 _sources/about/index.md.txt                   |   15 +
 _sources/autoapi/index.rst.txt                |   11 +
 _sources/autoapi/lmflow/args/index.rst.txt    | 1982 +++++++
 .../lmflow/datasets/dataset/index.rst.txt     |  646 +++
 .../autoapi/lmflow/datasets/index.rst.txt     |  651 +++
 .../multi_modal_dataset/index.rst.txt         |  177 +
 _sources/autoapi/lmflow/index.rst.txt         |   40 +
 .../lmflow/models/auto_model/index.rst.txt    |   31 +
 .../lmflow/models/base_model/index.rst.txt    |   51 +
 .../lmflow/models/decoder_model/index.rst.txt |   61 +
 .../encoder_decoder_model/index.rst.txt       |   61 +
 .../models/hf_decoder_model/index.rst.txt     |  427 ++
 .../hf_encoder_decoder_model/index.rst.txt    |  363 ++
 .../models/hf_model_mixin/index.rst.txt       |  301 ++
 .../hf_text_regression_model/index.rst.txt    |  268 +
 _sources/autoapi/lmflow/models/index.rst.txt  |   27 +
 .../lmflow/models/interfaces/index.rst.txt    |   15 +
 .../models/interfaces/tunable/index.rst.txt   |   51 +
 .../models/regression_model/index.rst.txt     |   51 +
 .../text_regression_model/index.rst.txt       |  114 +
 .../models/vision2seq_model/index.rst.txt     |  205 +
 .../vision_encoder/clip_encoder/index.rst.txt |  101 +
 .../models/vision_encoder/index.rst.txt       |   28 +
 .../lmflow/optim/adabelief/index.rst.txt      |   97 +
 .../lmflow/optim/adabound/index.rst.txt       |   80 +
 .../lmflow/optim/adadelta/index.rst.txt       |   25 +
 .../lmflow/optim/adagrad/index.rst.txt        |   25 +
 .../autoapi/lmflow/optim/adam/index.rst.txt   |   25 +
 .../autoapi/lmflow/optim/adamax/index.rst.txt |   28 +
 .../autoapi/lmflow/optim/adamp/index.rst.txt  |   93 +
 .../optim/adamw_schedule_free/index.rst.txt   |   82 +
 .../autoapi/lmflow/optim/adan/index.rst.txt   |   89 +
 .../autoapi/lmflow/optim/dummy/index.rst.txt  |   81 +
 _sources/autoapi/lmflow/optim/index.rst.txt   |   34 +
 .../autoapi/lmflow/optim/lamb/index.rst.txt   |   90 +
 .../autoapi/lmflow/optim/lars/index.rst.txt   |   98 +
 .../autoapi/lmflow/optim/nadam/index.rst.txt  |   28 +
 .../lmflow/optim/novograd/index.rst.txt       |   28 +
 .../lmflow/optim/optimizers/index.rst.txt     |   13 +
 .../autoapi/lmflow/optim/radam/index.rst.txt  |   81 +
 .../optim/sgd_schedule_free/index.rst.txt     |   82 +
 .../autoapi/lmflow/optim/sgdp/index.rst.txt   |   93 +
 .../autoapi/lmflow/optim/sophia/index.rst.txt |   53 +
 .../autoapi/lmflow/optim/yogi/index.rst.txt   |   74 +
 .../pipeline/auto_pipeline/index.rst.txt      |   67 +
 .../pipeline/base_aligner/index.rst.txt       |   58 +
 .../pipeline/base_pipeline/index.rst.txt      |   51 +
 .../lmflow/pipeline/base_tuner/index.rst.txt  |   58 +
 .../lmflow/pipeline/dpo_aligner/index.rst.txt |   92 +
 .../pipeline/dpov2_aligner/index.rst.txt      |  161 +
 .../lmflow/pipeline/evaluator/index.rst.txt   |  162 +
 .../lmflow/pipeline/finetuner/index.rst.txt   |  149 +
 .../autoapi/lmflow/pipeline/index.rst.txt     |   29 +
 .../lmflow/pipeline/inferencer/index.rst.txt  |  468 ++
 .../iterative_dpo_aligner/index.rst.txt       |   74 +
 .../pipeline/raft_aligner/index.rst.txt       |  259 +
 .../pipeline/rm_inferencer/index.rst.txt      |  104 +
 .../lmflow/pipeline/rm_tuner/index.rst.txt    |  100 +
 .../utils/dpov2_dataprocessor/index.rst.txt   |  130 +
 .../utils/dpov2_trainer/index.rst.txt         |  105 +
 .../lmflow/pipeline/utils/index.rst.txt       |   22 +
 .../memory_safe_dpov2_align/index.rst.txt     |   33 +
 .../memory_safe_vllm_inference/index.rst.txt  |   29 +
 .../pipeline/utils/peft_trainer/index.rst.txt |  135 +
 .../pipeline/utils/raft_trainer/index.rst.txt | 1560 ++++++
 .../utils/rm_dataprocessor/index.rst.txt      |   60 +
 .../pipeline/utils/rm_trainer/index.rst.txt   |   47 +
 .../pipeline/vllm_inferencer/index.rst.txt    |  268 +
 .../hf_decoder_model/index.rst.txt            |   80 +
 .../hf_text_regression_model/index.rst.txt    |  114 +
 .../autoapi/lmflow/tokenization/index.rst.txt |   16 +
 .../autoapi/lmflow/utils/common/index.rst.txt |  179 +
 .../lmflow/utils/constants/index.rst.txt      |  160 +
 .../conversation_template/base/index.rst.txt  |  580 +++
 .../chatglm/index.rst.txt                     |   19 +
 .../chatml/index.rst.txt                      |   19 +
 .../deepseek/index.rst.txt                    |   19 +
 .../conversation_template/gemma/index.rst.txt |   38 +
 .../conversation_template/hymba/index.rst.txt |   74 +
 .../utils/conversation_template/index.rst.txt |  370 ++
 .../internlm/index.rst.txt                    |   19 +
 .../conversation_template/llama/index.rst.txt |   56 +
 .../conversation_template/phi/index.rst.txt   |   19 +
 .../conversation_template/qwen/index.rst.txt  |   22 +
 .../conversation_template/yi/index.rst.txt    |   19 +
 .../zephyr/index.rst.txt                      |   38 +
 .../lmflow/utils/data_utils/index.rst.txt     |  241 +
 .../bloom_flash_attention/index.rst.txt       |   25 +
 .../gpt2_flash_attention/index.rst.txt        |   25 +
 .../gpt_neo_flash_attention/index.rst.txt     |   25 +
 .../utils/flash_attention/index.rst.txt       |   19 +
 .../llama_flash_attention/index.rst.txt       |   25 +
 .../triton_flash_attention/index.rst.txt      |  229 +
 _sources/autoapi/lmflow/utils/index.rst.txt   |   24 +
 .../llava_conversation_lib/index.rst.txt      |  187 +
 .../autoapi/lmflow/utils/model/index.rst.txt  |   29 +
 .../lmflow/utils/multimodal/index.rst.txt     |   25 +
 .../position_interpolation/index.rst.txt      |   15 +
 .../index.rst.txt                             |   44 +
 .../lmflow/utils/versioning/index.rst.txt     |   59 +
 _sources/autoapi/lmflow/version/index.rst.txt |   21 +
 _sources/blogs/benchmark.md.txt               |  253 +
 _sources/blogs/index.md.txt                   |   13 +
 _sources/examples/DATASETS.md.txt             |  385 ++
 _sources/examples/TASK_GUIDE.md.txt           |  132 +
 _sources/examples/checkpoints.md.txt          |   31 +
 .../customize_conversation_template.md.txt    |  115 +
 _sources/examples/finetuning.md.txt           |  101 +
 _sources/examples/index.md.txt                |   57 +
 _sources/examples/medical_finetune.md.txt     |   55 +
 _sources/examples/raft.md.txt                 |  342 ++
 _sources/examples/reward_modeling.md.txt      |  132 +
 .../supported_conversation_template.md.txt    |  504 ++
 _sources/index.md.txt                         |  207 +
 _sphinx_design_static/design-tabs.js          |  101 +
 _sphinx_design_static/sphinx-design.min.css   |    1 +
 _static/IT_sample1.png                        |  Bin 0 -> 523440 bytes
 _static/IT_sample2.png                        |  Bin 0 -> 282807 bytes
 _static/IT_sample3.png                        |  Bin 0 -> 396642 bytes
 _static/IT_sample4.png                        |  Bin 0 -> 334618 bytes
 _static/IT_sample5.png                        |  Bin 0 -> 632293 bytes
 _static/IT_sample6.png                        |  Bin 0 -> 365253 bytes
 _static/IT_sample7.png                        |  Bin 0 -> 578291 bytes
 _static/basic.css                             |  914 ++++
 _static/benchmark-1.png                       |  Bin 0 -> 62190 bytes
 _static/benchmark-2.png                       |  Bin 0 -> 142903 bytes
 _static/check_before_after_lora_tuning.jsonl  |  252 +
 _static/design-tabs.js                        |  101 +
 _static/doctools.js                           |  149 +
 _static/documentation_options.js              |   13 +
 _static/eq.png                                |  Bin 0 -> 79072 bytes
 _static/file.png                              |  Bin 0 -> 286 bytes
 _static/graphviz.css                          |   12 +
 _static/language_data.js                      |  192 +
 _static/logo.png                              |  Bin 0 -> 139180 bytes
 _static/logo.svg                              |    1 +
 _static/logo2.svg                             |    1 +
 _static/logo3.svg                             |    1 +
 _static/logo4.svg                             |    1 +
 _static/logo5.svg                             |    1 +
 _static/logo6.svg                             |    1 +
 _static/minus.png                             |  Bin 0 -> 90 bytes
 _static/nll.png                               |  Bin 0 -> 55999 bytes
 _static/plot_directive.css                    |   16 +
 _static/plus.png                              |  Bin 0 -> 90 bytes
 _static/ppl.png                               |  Bin 0 -> 25483 bytes
 _static/pygments.css                          |  152 +
 _static/raft-demo-examples.png                |  Bin 0 -> 324598 bytes
 _static/raft.png                              |  Bin 0 -> 181749 bytes
 _static/raft_idea.PNG                         |  Bin 0 -> 121926 bytes
 _static/raft_reward.PNG                       |  Bin 0 -> 221269 bytes
 _static/scripts/bootstrap.js                  |    3 +
 _static/scripts/bootstrap.js.LICENSE.txt      |    5 +
 _static/scripts/bootstrap.js.map              |    1 +
 _static/scripts/fontawesome.js                |    3 +
 _static/scripts/fontawesome.js.LICENSE.txt    |    5 +
 _static/scripts/fontawesome.js.map            |    1 +
 _static/scripts/pydata-sphinx-theme.js        |    2 +
 _static/scripts/pydata-sphinx-theme.js.map    |    1 +
 _static/searchtools.js                        |  632 +++
 _static/sphinx-design.min.css                 |    1 +
 _static/sphinx_highlight.js                   |  154 +
 _static/styles/pydata-sphinx-theme.css        |   32 +
 _static/styles/pydata-sphinx-theme.css.map    |    1 +
 _static/styles/theme.css                      |    2 +
 .../fontawesome/webfonts/fa-brands-400.ttf    |  Bin 0 -> 209128 bytes
 .../fontawesome/webfonts/fa-brands-400.woff2  |  Bin 0 -> 117852 bytes
 .../fontawesome/webfonts/fa-regular-400.ttf   |  Bin 0 -> 67860 bytes
 .../fontawesome/webfonts/fa-regular-400.woff2 |  Bin 0 -> 25392 bytes
 .../fontawesome/webfonts/fa-solid-900.ttf     |  Bin 0 -> 420332 bytes
 .../fontawesome/webfonts/fa-solid-900.woff2   |  Bin 0 -> 156400 bytes
 _static/webpack-macros.html                   |   24 +
 about/authors.html                            |  484 ++
 about/changelog.html                          |  513 ++
 about/index.html                              |  503 ++
 autoapi/index.html                            |  747 +++
 autoapi/lmflow/args/index.html                | 2447 +++++++++
 autoapi/lmflow/datasets/dataset/index.html    | 1053 ++++
 autoapi/lmflow/datasets/index.html            | 1024 ++++
 .../datasets/multi_modal_dataset/index.html   |  790 +++
 autoapi/lmflow/index.html                     |  674 +++
 autoapi/lmflow/models/auto_model/index.html   |  664 +++
 autoapi/lmflow/models/base_model/index.html   |  659 +++
 .../lmflow/models/decoder_model/index.html    |  668 +++
 .../models/encoder_decoder_model/index.html   |  668 +++
 .../lmflow/models/hf_decoder_model/index.html |  945 ++++
 .../hf_encoder_decoder_model/index.html       |  864 ++++
 .../lmflow/models/hf_model_mixin/index.html   |  901 ++++
 .../hf_text_regression_model/index.html       |  825 +++
 autoapi/lmflow/models/index.html              |  649 +++
 autoapi/lmflow/models/interfaces/index.html   |  640 +++
 .../models/interfaces/tunable/index.html      |  662 +++
 .../lmflow/models/regression_model/index.html |  659 +++
 .../models/text_regression_model/index.html   |  697 +++
 .../lmflow/models/vision2seq_model/index.html |  798 +++
 .../vision_encoder/clip_encoder/index.html    |  771 +++
 .../lmflow/models/vision_encoder/index.html   |  664 +++
 autoapi/lmflow/optim/adabelief/index.html     |  710 +++
 autoapi/lmflow/optim/adabound/index.html      |  689 +++
 autoapi/lmflow/optim/adadelta/index.html      |  664 +++
 autoapi/lmflow/optim/adagrad/index.html       |  664 +++
 autoapi/lmflow/optim/adam/index.html          |  664 +++
 autoapi/lmflow/optim/adamax/index.html        |  670 +++
 autoapi/lmflow/optim/adamp/index.html         |  701 +++
 .../optim/adamw_schedule_free/index.html      |  691 +++
 autoapi/lmflow/optim/adan/index.html          |  708 +++
 autoapi/lmflow/optim/dummy/index.html         |  680 +++
 autoapi/lmflow/optim/index.html               |  656 +++
 autoapi/lmflow/optim/lamb/index.html          |  695 +++
 autoapi/lmflow/optim/lars/index.html          |  705 +++
 autoapi/lmflow/optim/nadam/index.html         |  670 +++
 autoapi/lmflow/optim/novograd/index.html      |  670 +++
 autoapi/lmflow/optim/optimizers/index.html    |  621 +++
 autoapi/lmflow/optim/radam/index.html         |  687 +++
 .../lmflow/optim/sgd_schedule_free/index.html |  691 +++
 autoapi/lmflow/optim/sgdp/index.html          |  701 +++
 autoapi/lmflow/optim/sophia/index.html        |  678 +++
 autoapi/lmflow/optim/yogi/index.html          |  676 +++
 .../lmflow/pipeline/auto_pipeline/index.html  |  692 +++
 .../lmflow/pipeline/base_aligner/index.html   |  672 +++
 .../lmflow/pipeline/base_pipeline/index.html  |  659 +++
 autoapi/lmflow/pipeline/base_tuner/index.html |  672 +++
 .../lmflow/pipeline/dpo_aligner/index.html    |  740 +++
 .../lmflow/pipeline/dpov2_aligner/index.html  |  813 +++
 autoapi/lmflow/pipeline/evaluator/index.html  |  777 +++
 autoapi/lmflow/pipeline/finetuner/index.html  |  749 +++
 autoapi/lmflow/pipeline/index.html            |  651 +++
 autoapi/lmflow/pipeline/inferencer/index.html |  971 ++++
 .../pipeline/iterative_dpo_aligner/index.html |  765 +++
 .../lmflow/pipeline/raft_aligner/index.html   |  806 +++
 .../lmflow/pipeline/rm_inferencer/index.html  |  767 +++
 autoapi/lmflow/pipeline/rm_tuner/index.html   |  710 +++
 .../utils/dpov2_dataprocessor/index.html      |  771 +++
 .../pipeline/utils/dpov2_trainer/index.html   |  724 +++
 autoapi/lmflow/pipeline/utils/index.html      |  647 +++
 .../utils/memory_safe_dpov2_align/index.html  |  685 +++
 .../memory_safe_vllm_inference/index.html     |  676 +++
 .../pipeline/utils/peft_trainer/index.html    |  709 +++
 .../pipeline/utils/raft_trainer/index.html    | 1796 +++++++
 .../utils/rm_dataprocessor/index.html         |  714 +++
 .../pipeline/utils/rm_trainer/index.html      |  712 +++
 .../pipeline/vllm_inferencer/index.html       |  864 ++++
 .../tokenization/hf_decoder_model/index.html  |  702 +++
 .../hf_text_regression_model/index.html       |  740 +++
 autoapi/lmflow/tokenization/index.html        |  638 +++
 autoapi/lmflow/utils/common/index.html        |  777 +++
 autoapi/lmflow/utils/constants/index.html     |  908 ++++
 .../conversation_template/base/index.html     | 1153 +++++
 .../conversation_template/chatglm/index.html  |  658 +++
 .../conversation_template/chatml/index.html   |  658 +++
 .../conversation_template/deepseek/index.html |  658 +++
 .../conversation_template/gemma/index.html    |  694 +++
 .../conversation_template/hymba/index.html    |  713 +++
 .../utils/conversation_template/index.html    | 1086 ++++
 .../conversation_template/internlm/index.html |  658 +++
 .../conversation_template/llama/index.html    |  739 +++
 .../conversation_template/phi/index.html      |  658 +++
 .../conversation_template/qwen/index.html     |  667 +++
 .../utils/conversation_template/yi/index.html |  658 +++
 .../conversation_template/zephyr/index.html   |  694 +++
 autoapi/lmflow/utils/data_utils/index.html    |  813 +++
 .../bloom_flash_attention/index.html          |  676 +++
 .../gpt2_flash_attention/index.html           |  676 +++
 .../gpt_neo_flash_attention/index.html        |  676 +++
 .../lmflow/utils/flash_attention/index.html   |  644 +++
 .../llama_flash_attention/index.html          |  676 +++
 .../triton_flash_attention/index.html         |  892 ++++
 autoapi/lmflow/utils/index.html               |  646 +++
 .../utils/llava_conversation_lib/index.html   |  908 ++++
 autoapi/lmflow/utils/model/index.html         |  673 +++
 autoapi/lmflow/utils/multimodal/index.html    |  673 +++
 .../utils/position_interpolation/index.html   |  640 +++
 .../llama_rope_scaled_monkey_patch/index.html |  703 +++
 autoapi/lmflow/utils/versioning/index.html    |  763 +++
 autoapi/lmflow/version/index.html             |  652 +++
 blogs/benchmark.html                          | 1061 ++++
 blogs/index.html                              |  507 ++
 examples/DATASETS.html                        |  972 ++++
 examples/TASK_GUIDE.html                      |  639 +++
 examples/checkpoints.html                     |  546 ++
 examples/customize_conversation_template.html |  592 +++
 examples/finetuning.html                      |  615 +++
 examples/index.html                           |  618 +++
 examples/medical_finetune.html                |  510 ++
 examples/raft.html                            |  864 ++++
 examples/reward_modeling.html                 |  674 +++
 examples/supported_conversation_template.html |  915 ++++
 genindex.html                                 | 4001 ++++++++++++++
 index.html                                    |  794 +++
 objects.inv                                   |  Bin 0 -> 12128 bytes
 py-modindex.html                              |  921 ++++
 search.html                                   |  439 ++
 searchindex.js                                |    1 +
 515 files changed, 174494 insertions(+)
 create mode 100644 .buildinfo
 create mode 100644 .doctrees/about/authors.doctree
 create mode 100644 .doctrees/about/changelog.doctree
 create mode 100644 .doctrees/about/index.doctree
 create mode 100644 .doctrees/autoapi/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/args/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/datasets/dataset/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/datasets/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/datasets/multi_modal_dataset/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/auto_model/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/base_model/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/decoder_model/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/encoder_decoder_model/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/hf_decoder_model/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/hf_encoder_decoder_model/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/hf_model_mixin/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/hf_text_regression_model/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/interfaces/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/interfaces/tunable/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/regression_model/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/text_regression_model/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/vision2seq_model/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/vision_encoder/clip_encoder/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/models/vision_encoder/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/adabelief/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/adabound/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/adadelta/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/adagrad/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/adam/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/adamax/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/adamp/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/adamw_schedule_free/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/adan/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/dummy/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/lamb/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/lars/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/nadam/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/novograd/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/optimizers/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/radam/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/sgd_schedule_free/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/sgdp/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/sophia/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/optim/yogi/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/auto_pipeline/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/base_aligner/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/base_pipeline/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/base_tuner/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/dpo_aligner/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/dpov2_aligner/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/evaluator/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/finetuner/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/inferencer/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/iterative_dpo_aligner/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/raft_aligner/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/rm_inferencer/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/rm_tuner/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/utils/dpov2_trainer/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/utils/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/utils/peft_trainer/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/utils/raft_trainer/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/utils/rm_trainer/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/pipeline/vllm_inferencer/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/tokenization/hf_decoder_model/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/tokenization/hf_text_regression_model/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/tokenization/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/common/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/constants/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/conversation_template/base/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/conversation_template/chatglm/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/conversation_template/chatml/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/conversation_template/deepseek/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/conversation_template/gemma/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/conversation_template/hymba/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/conversation_template/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/conversation_template/internlm/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/conversation_template/llama/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/conversation_template/phi/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/conversation_template/qwen/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/conversation_template/yi/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/conversation_template/zephyr/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/data_utils/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/flash_attention/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/llava_conversation_lib/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/model/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/multimodal/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/position_interpolation/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/utils/versioning/index.doctree
 create mode 100644 .doctrees/autoapi/lmflow/version/index.doctree
 create mode 100644 .doctrees/blogs/benchmark.doctree
 create mode 100644 .doctrees/blogs/index.doctree
 create mode 100644 .doctrees/environment.pickle
 create mode 100644 .doctrees/examples/DATASETS.doctree
 create mode 100644 .doctrees/examples/TASK_GUIDE.doctree
 create mode 100644 .doctrees/examples/checkpoints.doctree
 create mode 100644 .doctrees/examples/customize_conversation_template.doctree
 create mode 100644 .doctrees/examples/finetuning.doctree
 create mode 100644 .doctrees/examples/index.doctree
 create mode 100644 .doctrees/examples/medical_finetune.doctree
 create mode 100644 .doctrees/examples/raft.doctree
 create mode 100644 .doctrees/examples/reward_modeling.doctree
 create mode 100644 .doctrees/examples/supported_conversation_template.doctree
 create mode 100644 .doctrees/index.doctree
 create mode 100644 .nojekyll
 create mode 100644 _images/IT_sample1.png
 create mode 100644 _images/IT_sample2.png
 create mode 100644 _images/IT_sample3.png
 create mode 100644 _images/IT_sample4.png
 create mode 100644 _images/IT_sample5.png
 create mode 100644 _images/IT_sample6.png
 create mode 100644 _images/IT_sample7.png
 create mode 100644 _images/benchmark-1.png
 create mode 100644 _images/benchmark-2.png
 create mode 100644 _images/nll.png
 create mode 100644 _images/ppl.png
 create mode 100644 _images/raft_idea.PNG
 create mode 100644 _images/raft_reward.PNG
 create mode 100644 _modules/index.html
 create mode 100644 _modules/lmflow.html
 create mode 100644 _modules/lmflow/args.html
 create mode 100644 _modules/lmflow/datasets/dataset.html
 create mode 100644 _modules/lmflow/datasets/multi_modal_dataset.html
 create mode 100644 _modules/lmflow/models/auto_model.html
 create mode 100644 _modules/lmflow/models/base_model.html
 create mode 100644 _modules/lmflow/models/decoder_model.html
 create mode 100644 _modules/lmflow/models/encoder_decoder_model.html
 create mode 100644 _modules/lmflow/models/hf_decoder_model.html
 create mode 100644 _modules/lmflow/models/hf_encoder_decoder_model.html
 create mode 100644 _modules/lmflow/models/hf_model_mixin.html
 create mode 100644 _modules/lmflow/models/hf_text_regression_model.html
 create mode 100644 _modules/lmflow/models/interfaces/tunable.html
 create mode 100644 _modules/lmflow/models/regression_model.html
 create mode 100644 _modules/lmflow/models/text_regression_model.html
 create mode 100644 _modules/lmflow/models/vision2seq_model.html
 create mode 100644 _modules/lmflow/models/vision_encoder/clip_encoder.html
 create mode 100644 _modules/lmflow/optim/adabelief.html
 create mode 100644 _modules/lmflow/optim/adabound.html
 create mode 100644 _modules/lmflow/optim/adadelta.html
 create mode 100644 _modules/lmflow/optim/adagrad.html
 create mode 100644 _modules/lmflow/optim/adam.html
 create mode 100644 _modules/lmflow/optim/adamax.html
 create mode 100644 _modules/lmflow/optim/adamp.html
 create mode 100644 _modules/lmflow/optim/adamw_schedule_free.html
 create mode 100644 _modules/lmflow/optim/adan.html
 create mode 100644 _modules/lmflow/optim/dummy.html
 create mode 100644 _modules/lmflow/optim/lamb.html
 create mode 100644 _modules/lmflow/optim/lars.html
 create mode 100644 _modules/lmflow/optim/nadam.html
 create mode 100644 _modules/lmflow/optim/novograd.html
 create mode 100644 _modules/lmflow/optim/radam.html
 create mode 100644 _modules/lmflow/optim/sgd_schedule_free.html
 create mode 100644 _modules/lmflow/optim/sgdp.html
 create mode 100644 _modules/lmflow/optim/sophia.html
 create mode 100644 _modules/lmflow/optim/yogi.html
 create mode 100644 _modules/lmflow/pipeline/auto_pipeline.html
 create mode 100644 _modules/lmflow/pipeline/base_aligner.html
 create mode 100644 _modules/lmflow/pipeline/base_pipeline.html
 create mode 100644 _modules/lmflow/pipeline/base_tuner.html
 create mode 100644 _modules/lmflow/pipeline/dpo_aligner.html
 create mode 100644 _modules/lmflow/pipeline/dpov2_aligner.html
 create mode 100644 _modules/lmflow/pipeline/evaluator.html
 create mode 100644 _modules/lmflow/pipeline/finetuner.html
 create mode 100644 _modules/lmflow/pipeline/inferencer.html
 create mode 100644 _modules/lmflow/pipeline/iterative_dpo_aligner.html
 create mode 100644 _modules/lmflow/pipeline/raft_aligner.html
 create mode 100644 _modules/lmflow/pipeline/rm_inferencer.html
 create mode 100644 _modules/lmflow/pipeline/rm_tuner.html
 create mode 100644 _modules/lmflow/pipeline/utils/dpov2_dataprocessor.html
 create mode 100644 _modules/lmflow/pipeline/utils/dpov2_trainer.html
 create mode 100644 _modules/lmflow/pipeline/utils/memory_safe_dpov2_align.html
 create mode 100644 _modules/lmflow/pipeline/utils/memory_safe_vllm_inference.html
 create mode 100644 _modules/lmflow/pipeline/utils/peft_trainer.html
 create mode 100644 _modules/lmflow/pipeline/utils/raft_trainer.html
 create mode 100644 _modules/lmflow/pipeline/utils/rm_dataprocessor.html
 create mode 100644 _modules/lmflow/pipeline/utils/rm_trainer.html
 create mode 100644 _modules/lmflow/pipeline/vllm_inferencer.html
 create mode 100644 _modules/lmflow/tokenization/hf_decoder_model.html
 create mode 100644 _modules/lmflow/tokenization/hf_text_regression_model.html
 create mode 100644 _modules/lmflow/utils/common.html
 create mode 100644 _modules/lmflow/utils/constants.html
 create mode 100644 _modules/lmflow/utils/conversation_template.html
 create mode 100644 _modules/lmflow/utils/conversation_template/base.html
 create mode 100644 _modules/lmflow/utils/conversation_template/chatglm.html
 create mode 100644 _modules/lmflow/utils/conversation_template/chatml.html
 create mode 100644 _modules/lmflow/utils/conversation_template/deepseek.html
 create mode 100644 _modules/lmflow/utils/conversation_template/gemma.html
 create mode 100644 _modules/lmflow/utils/conversation_template/hymba.html
 create mode 100644 _modules/lmflow/utils/conversation_template/internlm.html
 create mode 100644 _modules/lmflow/utils/conversation_template/llama.html
 create mode 100644 _modules/lmflow/utils/conversation_template/phi.html
 create mode 100644 _modules/lmflow/utils/conversation_template/qwen.html
 create mode 100644 _modules/lmflow/utils/conversation_template/yi.html
 create mode 100644 _modules/lmflow/utils/conversation_template/zephyr.html
 create mode 100644 _modules/lmflow/utils/data_utils.html
 create mode 100644 _modules/lmflow/utils/flash_attention/bloom_flash_attention.html
 create mode 100644 _modules/lmflow/utils/flash_attention/gpt2_flash_attention.html
 create mode 100644 _modules/lmflow/utils/flash_attention/gpt_neo_flash_attention.html
 create mode 100644 _modules/lmflow/utils/flash_attention/llama_flash_attention.html
 create mode 100644 _modules/lmflow/utils/flash_attention/triton_flash_attention.html
 create mode 100644 _modules/lmflow/utils/llava_conversation_lib.html
 create mode 100644 _modules/lmflow/utils/model.html
 create mode 100644 _modules/lmflow/utils/multimodal.html
 create mode 100644 _modules/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch.html
 create mode 100644 _modules/lmflow/utils/versioning.html
 create mode 100644 _modules/lmflow/version.html
 create mode 100644 _sources/about/authors.md.txt
 create mode 100644 _sources/about/changelog.md.txt
 create mode 100644 _sources/about/index.md.txt
 create mode 100644 _sources/autoapi/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/args/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/datasets/dataset/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/datasets/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/datasets/multi_modal_dataset/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/auto_model/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/base_model/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/decoder_model/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/encoder_decoder_model/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/hf_decoder_model/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/hf_encoder_decoder_model/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/hf_model_mixin/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/hf_text_regression_model/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/interfaces/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/interfaces/tunable/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/regression_model/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/text_regression_model/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/vision2seq_model/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/vision_encoder/clip_encoder/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/models/vision_encoder/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/adabelief/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/adabound/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/adadelta/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/adagrad/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/adam/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/adamax/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/adamp/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/adamw_schedule_free/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/adan/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/dummy/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/lamb/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/lars/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/nadam/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/novograd/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/optimizers/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/radam/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/sgd_schedule_free/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/sgdp/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/sophia/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/optim/yogi/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/auto_pipeline/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/base_aligner/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/base_pipeline/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/base_tuner/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/dpo_aligner/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/dpov2_aligner/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/evaluator/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/finetuner/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/inferencer/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/iterative_dpo_aligner/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/raft_aligner/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/rm_inferencer/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/rm_tuner/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/utils/dpov2_trainer/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/utils/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/utils/peft_trainer/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/utils/raft_trainer/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/utils/rm_trainer/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/pipeline/vllm_inferencer/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/tokenization/hf_decoder_model/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/tokenization/hf_text_regression_model/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/tokenization/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/common/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/constants/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/conversation_template/base/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/conversation_template/chatglm/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/conversation_template/chatml/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/conversation_template/deepseek/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/conversation_template/gemma/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/conversation_template/hymba/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/conversation_template/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/conversation_template/internlm/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/conversation_template/llama/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/conversation_template/phi/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/conversation_template/qwen/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/conversation_template/yi/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/conversation_template/zephyr/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/data_utils/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/flash_attention/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/llava_conversation_lib/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/model/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/multimodal/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/position_interpolation/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/utils/versioning/index.rst.txt
 create mode 100644 _sources/autoapi/lmflow/version/index.rst.txt
 create mode 100644 _sources/blogs/benchmark.md.txt
 create mode 100644 _sources/blogs/index.md.txt
 create mode 100644 _sources/examples/DATASETS.md.txt
 create mode 100644 _sources/examples/TASK_GUIDE.md.txt
 create mode 100644 _sources/examples/checkpoints.md.txt
 create mode 100644 _sources/examples/customize_conversation_template.md.txt
 create mode 100644 _sources/examples/finetuning.md.txt
 create mode 100644 _sources/examples/index.md.txt
 create mode 100644 _sources/examples/medical_finetune.md.txt
 create mode 100644 _sources/examples/raft.md.txt
 create mode 100644 _sources/examples/reward_modeling.md.txt
 create mode 100644 _sources/examples/supported_conversation_template.md.txt
 create mode 100644 _sources/index.md.txt
 create mode 100644 _sphinx_design_static/design-tabs.js
 create mode 100644 _sphinx_design_static/sphinx-design.min.css
 create mode 100644 _static/IT_sample1.png
 create mode 100644 _static/IT_sample2.png
 create mode 100644 _static/IT_sample3.png
 create mode 100644 _static/IT_sample4.png
 create mode 100644 _static/IT_sample5.png
 create mode 100644 _static/IT_sample6.png
 create mode 100644 _static/IT_sample7.png
 create mode 100644 _static/basic.css
 create mode 100644 _static/benchmark-1.png
 create mode 100644 _static/benchmark-2.png
 create mode 100644 _static/check_before_after_lora_tuning.jsonl
 create mode 100644 _static/design-tabs.js
 create mode 100644 _static/doctools.js
 create mode 100644 _static/documentation_options.js
 create mode 100644 _static/eq.png
 create mode 100644 _static/file.png
 create mode 100644 _static/graphviz.css
 create mode 100644 _static/language_data.js
 create mode 100644 _static/logo.png
 create mode 100644 _static/logo.svg
 create mode 100644 _static/logo2.svg
 create mode 100644 _static/logo3.svg
 create mode 100644 _static/logo4.svg
 create mode 100644 _static/logo5.svg
 create mode 100644 _static/logo6.svg
 create mode 100644 _static/minus.png
 create mode 100644 _static/nll.png
 create mode 100644 _static/plot_directive.css
 create mode 100644 _static/plus.png
 create mode 100644 _static/ppl.png
 create mode 100644 _static/pygments.css
 create mode 100644 _static/raft-demo-examples.png
 create mode 100644 _static/raft.png
 create mode 100644 _static/raft_idea.PNG
 create mode 100644 _static/raft_reward.PNG
 create mode 100644 _static/scripts/bootstrap.js
 create mode 100644 _static/scripts/bootstrap.js.LICENSE.txt
 create mode 100644 _static/scripts/bootstrap.js.map
 create mode 100644 _static/scripts/fontawesome.js
 create mode 100644 _static/scripts/fontawesome.js.LICENSE.txt
 create mode 100644 _static/scripts/fontawesome.js.map
 create mode 100644 _static/scripts/pydata-sphinx-theme.js
 create mode 100644 _static/scripts/pydata-sphinx-theme.js.map
 create mode 100644 _static/searchtools.js
 create mode 100644 _static/sphinx-design.min.css
 create mode 100644 _static/sphinx_highlight.js
 create mode 100644 _static/styles/pydata-sphinx-theme.css
 create mode 100644 _static/styles/pydata-sphinx-theme.css.map
 create mode 100644 _static/styles/theme.css
 create mode 100644 _static/vendor/fontawesome/webfonts/fa-brands-400.ttf
 create mode 100644 _static/vendor/fontawesome/webfonts/fa-brands-400.woff2
 create mode 100644 _static/vendor/fontawesome/webfonts/fa-regular-400.ttf
 create mode 100644 _static/vendor/fontawesome/webfonts/fa-regular-400.woff2
 create mode 100644 _static/vendor/fontawesome/webfonts/fa-solid-900.ttf
 create mode 100644 _static/vendor/fontawesome/webfonts/fa-solid-900.woff2
 create mode 100644 _static/webpack-macros.html
 create mode 100644 about/authors.html
 create mode 100644 about/changelog.html
 create mode 100644 about/index.html
 create mode 100644 autoapi/index.html
 create mode 100644 autoapi/lmflow/args/index.html
 create mode 100644 autoapi/lmflow/datasets/dataset/index.html
 create mode 100644 autoapi/lmflow/datasets/index.html
 create mode 100644 autoapi/lmflow/datasets/multi_modal_dataset/index.html
 create mode 100644 autoapi/lmflow/index.html
 create mode 100644 autoapi/lmflow/models/auto_model/index.html
 create mode 100644 autoapi/lmflow/models/base_model/index.html
 create mode 100644 autoapi/lmflow/models/decoder_model/index.html
 create mode 100644 autoapi/lmflow/models/encoder_decoder_model/index.html
 create mode 100644 autoapi/lmflow/models/hf_decoder_model/index.html
 create mode 100644 autoapi/lmflow/models/hf_encoder_decoder_model/index.html
 create mode 100644 autoapi/lmflow/models/hf_model_mixin/index.html
 create mode 100644 autoapi/lmflow/models/hf_text_regression_model/index.html
 create mode 100644 autoapi/lmflow/models/index.html
 create mode 100644 autoapi/lmflow/models/interfaces/index.html
 create mode 100644 autoapi/lmflow/models/interfaces/tunable/index.html
 create mode 100644 autoapi/lmflow/models/regression_model/index.html
 create mode 100644 autoapi/lmflow/models/text_regression_model/index.html
 create mode 100644 autoapi/lmflow/models/vision2seq_model/index.html
 create mode 100644 autoapi/lmflow/models/vision_encoder/clip_encoder/index.html
 create mode 100644 autoapi/lmflow/models/vision_encoder/index.html
 create mode 100644 autoapi/lmflow/optim/adabelief/index.html
 create mode 100644 autoapi/lmflow/optim/adabound/index.html
 create mode 100644 autoapi/lmflow/optim/adadelta/index.html
 create mode 100644 autoapi/lmflow/optim/adagrad/index.html
 create mode 100644 autoapi/lmflow/optim/adam/index.html
 create mode 100644 autoapi/lmflow/optim/adamax/index.html
 create mode 100644 autoapi/lmflow/optim/adamp/index.html
 create mode 100644 autoapi/lmflow/optim/adamw_schedule_free/index.html
 create mode 100644 autoapi/lmflow/optim/adan/index.html
 create mode 100644 autoapi/lmflow/optim/dummy/index.html
 create mode 100644 autoapi/lmflow/optim/index.html
 create mode 100644 autoapi/lmflow/optim/lamb/index.html
 create mode 100644 autoapi/lmflow/optim/lars/index.html
 create mode 100644 autoapi/lmflow/optim/nadam/index.html
 create mode 100644 autoapi/lmflow/optim/novograd/index.html
 create mode 100644 autoapi/lmflow/optim/optimizers/index.html
 create mode 100644 autoapi/lmflow/optim/radam/index.html
 create mode 100644 autoapi/lmflow/optim/sgd_schedule_free/index.html
 create mode 100644 autoapi/lmflow/optim/sgdp/index.html
 create mode 100644 autoapi/lmflow/optim/sophia/index.html
 create mode 100644 autoapi/lmflow/optim/yogi/index.html
 create mode 100644 autoapi/lmflow/pipeline/auto_pipeline/index.html
 create mode 100644 autoapi/lmflow/pipeline/base_aligner/index.html
 create mode 100644 autoapi/lmflow/pipeline/base_pipeline/index.html
 create mode 100644 autoapi/lmflow/pipeline/base_tuner/index.html
 create mode 100644 autoapi/lmflow/pipeline/dpo_aligner/index.html
 create mode 100644 autoapi/lmflow/pipeline/dpov2_aligner/index.html
 create mode 100644 autoapi/lmflow/pipeline/evaluator/index.html
 create mode 100644 autoapi/lmflow/pipeline/finetuner/index.html
 create mode 100644 autoapi/lmflow/pipeline/index.html
 create mode 100644 autoapi/lmflow/pipeline/inferencer/index.html
 create mode 100644 autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html
 create mode 100644 autoapi/lmflow/pipeline/raft_aligner/index.html
 create mode 100644 autoapi/lmflow/pipeline/rm_inferencer/index.html
 create mode 100644 autoapi/lmflow/pipeline/rm_tuner/index.html
 create mode 100644 autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html
 create mode 100644 autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html
 create mode 100644 autoapi/lmflow/pipeline/utils/index.html
 create mode 100644 autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.html
 create mode 100644 autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.html
 create mode 100644 autoapi/lmflow/pipeline/utils/peft_trainer/index.html
 create mode 100644 autoapi/lmflow/pipeline/utils/raft_trainer/index.html
 create mode 100644 autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html
 create mode 100644 autoapi/lmflow/pipeline/utils/rm_trainer/index.html
 create mode 100644 autoapi/lmflow/pipeline/vllm_inferencer/index.html
 create mode 100644 autoapi/lmflow/tokenization/hf_decoder_model/index.html
 create mode 100644 autoapi/lmflow/tokenization/hf_text_regression_model/index.html
 create mode 100644 autoapi/lmflow/tokenization/index.html
 create mode 100644 autoapi/lmflow/utils/common/index.html
 create mode 100644 autoapi/lmflow/utils/constants/index.html
 create mode 100644 autoapi/lmflow/utils/conversation_template/base/index.html
 create mode 100644 autoapi/lmflow/utils/conversation_template/chatglm/index.html
 create mode 100644 autoapi/lmflow/utils/conversation_template/chatml/index.html
 create mode 100644 autoapi/lmflow/utils/conversation_template/deepseek/index.html
 create mode 100644 autoapi/lmflow/utils/conversation_template/gemma/index.html
 create mode 100644 autoapi/lmflow/utils/conversation_template/hymba/index.html
 create mode 100644 autoapi/lmflow/utils/conversation_template/index.html
 create mode 100644 autoapi/lmflow/utils/conversation_template/internlm/index.html
 create mode 100644 autoapi/lmflow/utils/conversation_template/llama/index.html
 create mode 100644 autoapi/lmflow/utils/conversation_template/phi/index.html
 create mode 100644 autoapi/lmflow/utils/conversation_template/qwen/index.html
 create mode 100644 autoapi/lmflow/utils/conversation_template/yi/index.html
 create mode 100644 autoapi/lmflow/utils/conversation_template/zephyr/index.html
 create mode 100644 autoapi/lmflow/utils/data_utils/index.html
 create mode 100644 autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.html
 create mode 100644 autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.html
 create mode 100644 autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html
 create mode 100644 autoapi/lmflow/utils/flash_attention/index.html
 create mode 100644 autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.html
 create mode 100644 autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html
 create mode 100644 autoapi/lmflow/utils/index.html
 create mode 100644 autoapi/lmflow/utils/llava_conversation_lib/index.html
 create mode 100644 autoapi/lmflow/utils/model/index.html
 create mode 100644 autoapi/lmflow/utils/multimodal/index.html
 create mode 100644 autoapi/lmflow/utils/position_interpolation/index.html
 create mode 100644 autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html
 create mode 100644 autoapi/lmflow/utils/versioning/index.html
 create mode 100644 autoapi/lmflow/version/index.html
 create mode 100644 blogs/benchmark.html
 create mode 100644 blogs/index.html
 create mode 100644 examples/DATASETS.html
 create mode 100644 examples/TASK_GUIDE.html
 create mode 100644 examples/checkpoints.html
 create mode 100644 examples/customize_conversation_template.html
 create mode 100644 examples/finetuning.html
 create mode 100644 examples/index.html
 create mode 100644 examples/medical_finetune.html
 create mode 100644 examples/raft.html
 create mode 100644 examples/reward_modeling.html
 create mode 100644 examples/supported_conversation_template.html
 create mode 100644 genindex.html
 create mode 100644 index.html
 create mode 100644 objects.inv
 create mode 100644 py-modindex.html
 create mode 100644 search.html
 create mode 100644 searchindex.js

diff --git a/.buildinfo b/.buildinfo
new file mode 100644
index 000000000..43100cccc
--- /dev/null
+++ b/.buildinfo
@@ -0,0 +1,4 @@
+# Sphinx build info version 1
+# This file records the configuration used when building these files. When it is not found, a full rebuild will be done.
+config: fc6a77cd4e4d3169013d2c3d17d1fdcd
+tags: 645f666f9bcd5a90fca523b33c5a78b7
diff --git a/.doctrees/about/authors.doctree b/.doctrees/about/authors.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..9497cde932eb08dff970735c03ed5d69d7dea4b0
GIT binary patch
literal 3096
zcmcImTWcFf6pmv{wrokh<kCRmRH39vsBNWCN-2dvQc9^CQ<Bh@LRdz-BhBF5neAL+
zIbdiX0yfNBrvIXsf6)H5ezU7x$%^~ffa0AqXU=WD@0|Ii@mr^TsrcvCb<VgL?~x=C
zGNIbLl|;riqr&*N{=u*Q#NT%<MUJ$zDyH6Df=0}EnkXv#OYgct-SAZJP0lZC8XGP}
zeVjE|lP$mW_gTv|3^ysA*>deCQW(Vt)<~s&(1Y623px(|=XT%tGM;8m;N5t(=~_`8
zE?acXNt6d}DdmFt13MsYm6*~2vJ2TP@<fNom*bRZ4I#c8f@I~_N@DZrVnE_UMTbGh
zx@AK~Vav01c6)z;ZM!e}OlGvNtPoW7kEA;6fBW>SR324-5S;FpLG{T%TGJ=iFl4%y
zC7#`6ZFY+-+5u8-Aty3YBxkeLdH$U7*NpbQ<V5cDo>|`ej)<M!*F?OAP>Rt`Z=dv@
zGb_S>*-q~po>MXE{lrMv`+@S_kGKnbzQF&bxGeH@_rKVlToF?1ht$~i0lOV;fxVB?
zvpcxm5AeH--#z>upseg8wu^Euv2C_9tq83TxOO~?5K<X<M?Pb4wY;Cg;9^<g|Bj$W
z4cl>7jUqy)Bmg+dl^iLmOVl)sG$i%=m)vTlLlZ1*NMONrW2*utqZ)zjN>U=xt!Qct
z7b8$?v*yfDAvY=#@0#C})B=Jv>6$=GZjC=+>j$niqCgk{M?4k@F;p9p<yfpKDGfrU
zLq7Iyq3_*V8A1fVhfr9ZEUikh>KGYE0o7p_p)?S)I`(dB2s1(&I<tA|I)Tf=X3=;k
zRpw7!SA$A<6w54h2n~ZbO9@SBA&HK-VNt~__?sM<Q0>6A70snGQA%G?l=HG%WhAYa
z4Yx_h991;LSY0VKTqB_aJA%9J5m5qFgodIdgQSJ)h-pYE#K#3{7Euw)1X|v;tQqcp
zwyQb5Z=K(as8UizU`mowg&PQ=#>Tt#vhp)l8if__w&!-{Ofc!XrO{~s)zMq$H|9>d
zjgSC<ry1&#A|dZODL8D&h@#vH)|b(6-4#V+sgfw6;CTW!9GE$9O>kt098nIQRA^vL
z*G0RL(NyFvb6~BL*es7~Jff<FfCGw3kTe)vFC+!h!_`kkxu!Oe(Krie+MwfHY8uU+
zEFcn_=oQzz#Or#La7_kjNMsV_a>WyHu8m|gm{E`pMmrO=XvovDCtnX=aH6Ryh)G4T
zmf*hEFqv>q<#K<)Z3eOdHev)6raa@uKi#<Jx;f1wP=tb_aZ#{#kPHZ754><Um@&Ej
zyx)gd@8!qn(PB@k5$JR+XchrLGcq*8^%z{o3go+YAAW**cGWPRGan>V^B-2`!CVsX
zdPvnCr)F6AKsV?!lcjLhGGqV>C5HL^6}JsQW>L+|I)O;%0U>8@Lmvx6#u4Kqmf{ES
zE~RMn7@RsIVA;<Wl-4NJm112kqg2LZQrhj9VeZ7Z_>yd>a#2E@KD0Spg!x_!?MV+f
z1A1i4#Sj>nG=;6uCA1yPsqrjeH<hGhVwlK>9+1ffGy#2&5dovrz{j;1T89zM2u~xt
zL_*UoqTZamC4kj*ST?jXw@Ae+z~n}Dtj)AI1yp(Q@Hll_=l3)kjz6^nw^BAAupQXt
zBH>_@i40LAie`p)z$VfF77boB7ywC_{Nc&4j?tdP(h9R1zVJ?Ix%cK(xhUON9h^1X
z<&QtDrfv8vha-mEU;2WH_^uZ+J0$-I0}?rGPGiZqu;F^ypXEah9v2=~4)Im{`{zP`
z7g)N;qjg*7QY~Lc7nkMt3&Z={hRP4CgRU8-6{>iGM_3S+-|R7aq+Kfr%yEviPCR>j
zrr90*eZ<<@b!M4RobJRNd<s_ui=`wqyh^XU(pgEVhl>VtvVlx%{1mEJ=Y+L(>zu__
z<+V7pb<REp7qEs-bZy?&WC!mDwBrIxVd$*IGED(O<bY|s{3Kuu^0!F-jemyLf@dL`
z6dtD<Y$_erUHP1Ntri#XUgIK8Z4wAL3dmkf!r++n*zhEBs|87PYVjmF@piEHFVYpu
ArvLx|

literal 0
HcmV?d00001

diff --git a/.doctrees/about/changelog.doctree b/.doctrees/about/changelog.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..7e6daa23975412f6dd6d30f1ba25631e0b064b3d
GIT binary patch
literal 3984
zcmbtXTW=h<6_zaTN~_DRq{L|gOUVR<8yoTNDo!7aJft;(7AO=8MQY@!!GJr%S<>PR
zr@UCJ2LtVc0SVzl08jr<|5z^{=eC#FNrc*1Xn5xE+<3n4aQ@!;=c~b5`Ogh>&UiK(
zkvPs|Om*S5Vj0<#W~O*uJorcPvbgVhid<-ERYVK7291bu5i6P%-xaR!)eRSVRPW!^
zG%{Rf?RM5-UAFmMai8^E$8aO)!YkMNijizeg`5^%2|Bl*(V2I?`07<rMbga){F;It
zxL()>$~IiLj_=If7CfWHnVk@~O^i`|Vh!^0d2H}l`<OQ)L9~Vt&z+2<)zemN^Z8;z
zqH{$PuVdY&AyfaVU<3By={h@hUyhkf=~&q;qiTF1)%p1Ar+*akqWOd1^tcLYOeWHr
zanvL_O5=jvWxMQ%ZP*Fhsj56L(|#jxN8eJVQOn?DbTaxZxc`)>;L+oU!K0H$pFgPT
z_1@1Y)s=(46Yq~5pRo^pao9(w2>Tcn|22O1@cRV62gsIvT4s5h9kYknHmt+!xgLs7
zri$cj@c=(~#%PdmrA?6AiQrN2q6&J{1o#sg^XPv6%cJ`eXV6`V?ssj?zIaCTd0=eD
zv+2WtXWA%RHe@S9J*}oHh-rdWiG!FJqN&ks!2gelc}8b9%V_JZ8GZUB&^*lr4M-LT
z00+%9pJ4gq&wmOM=@WSY0Ocrqo5GFu%wzuR&0^kuYs?@0AxU_|fmQTIMyulpF1TUA
z*YcYuZJa-=H;?W4&0^bqYi#%a;;YFMj<ZiedIDXNQ1D8GMpl5ihjSa@FVvpLIlEb$
z{pC1m3rzPx4W!`-{(m-1i5fn7JoMNvNP}8;R4X+U0j&j!Ad9Hp25uTu*W|O{t=IC?
zg7qcg0zr6Fpuv(LHhjK6#V9Q~vxn|oqe!L&@e?A<m7FT7D{}4_X^1FJ*W7laLt`wh
zC!}@PkF4_4AGSol+ej+HcU!=)0q8Nq4%(rVD478-apAh(5@Esg?NsZ6lyhr}Gd4JL
zy(tBe6C+iD1sH*9L(&|J9VMkfs3413;nv56+o?hb;rAR0+x5~afz_UoQRs(<e~VB$
zh*@10?l6IwOd7iI^2+V`VVT)9oB_vaapn4;E1rjuO#KMfFnG6;P(e#c=rI&F0}TG^
z4fRAjbGwS>QkhVAQYcPuxot*7yX?3FI^(FKA;#)PspC2^o!BYd^)HC>)CmnmNeW31
z#}QGVl8=vPs96ZC$QW7}($*xy$3xBWe6;*Bq)JH@V(i6&Dx821H8O=8RFz+_(kZPJ
z?s(<RiV5~F*RPDu18BA$Ex%aV>GpjBXgtkOAAy7lw<mD(uw+V6?ilM^xY4+G6pf^c
z!<go18@S=<nKRdoWR@UDl!M1jH?Xej6KiBR7r9#;y|x$IG!NT;L{&=xM=Pp~2ybvu
zO3F_6M?V+kn%Y=~v($IfKAq)K({OQT*&>k%UvSN<ejS7{*8~#*iPTYUHaG$1c9Dz*
zGD^_F=w7WBCR|ikpZEL|oM0-Jz@(YC7UR5kaP`AERm;<Lci_Q>w#laGLcvpRil_UZ
zxPDIa5D=+Y)F-p+B@@im(Bs0r&y30S7vnL+dX!%pCbN-LQ^4tt*DM2oW@sq58Suk3
z6Y}A`li#AAeU(Hj=DlQY{x^+zKg)f;o>MjA)Fh=3u)&y_RKQu!kO?3Z8&;fdxnneB
z8n(!669{$gJLKB!>&wiLS;+X53H;E!w}te5zpKHtCj6TPq%{ilPPwj@p^y=&OM47j
zMLan83T&uySwieSw>ez+3v_uat~uZu-6JDRB=kT{3WuI0w4JP|@znROkfcJQzxbXP
zXqj|EW5D+k5im*(eB6nkbso}`a1r9F?@70e`k=n7qgO>z5!$uepxFzY;QQ&NHuK_?
zt;)-r{gpdhey2e=#g(17t%`hrcJwaKVh%K!&=57n{mNgr50D0W(cqH-0{}O$iu?x{
zxp5?IW`_Pr`|p5Xt*Y{@a^EmG>$qE=pES2j|5y!ScK7+(`r4X)!zI`K$kEb-{Er`L
zNaJ80PRg_L`+@zpy<x7mKJnDL+<%+PHyC@9Q|k_w60SZ){AbILVrKt-Q-se4h`{<I
zNSeK_>&F(#`4S&JURXWfXRp~0?0a@)UC&F+WsYv}-vh6&+4t9$ee9oqz!%CD8(O!w
z2$0g&Uc>=2e}b@INjwiWdeuoUN-8~^KLC|Zq-fi!RDE|v*lPEdEcP0&<)*_W`<Q2F
zK@ilPRa^BNTr0rRWsCV)v=d1o(3{8sLwfTo-+GY0MipKa-+*!Otpo1Dr=bSODu*pu
rFNt?raoP1dJd1>lJuaud;kRQyPipG*Hzl`SVn+*$o65_=PDcL&f6)6O

literal 0
HcmV?d00001

diff --git a/.doctrees/about/index.doctree b/.doctrees/about/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..f729d2cde2beebbf6be8711f303f973e72a2df5a
GIT binary patch
literal 3320
zcmai0TW=$`74~H%OO|Y{oopIl*A0RqO@hYO_VQ2!XplB-+SH9u6hYq>18Rm633JFP
zFKhW>f%YNA3gBDV$NsT4`3K2==*QvANLCi80l~w=!*l)4<?p?JeK+2#|NO*MjLX#(
zDGI3yYJwjWDt9H7HhdM{|3`Qc-t$9EZjEv}r@`-mBWGL`n#%Aj_|wR4xiDAF`y)eh
z%a!b&vmWcSqi5kgHuOEqt)N?1zW3|7ayCQ;+&}n&uA<_W{AEZM^eX~=U_0}}tTT`u
z`hH{A%s&uZ(lB##;*W{7n$MkuKhgd}UYKa>Q7(uvAmY0xNtwO|g{wA)ImvG{T|^$|
zk1ScntH37ghu8b;!hbkrs-#oxq@;R!tMtwEPoIA()NT6@#+hj{YD%JOQ!Wd-x+;sn
z9<wof!VcXWo(^(VR?5j@D;OzGY!7->ZZ+KwRCD(h{9$h+RzJe;9sJ(K?|u9p`_Yy<
z_=<38UVAw6lXR6cB9~OCW%y?xe#~VqTtOFH)NJfy;s+&J6|}M}Tpu7@q^dKC;4X!l
zhqdoBUKB|D+MhJ9JHkM^a!xfZ!u6g%uJde^6nEe~27Wf?IQA3vGj_o~Ku8}TCR3!0
zZTsaXjr*6knp73lfo!`P-(R+|Y~>)PR|%~x+}O`|&f6-s<w4Z1*hkO8%%7w%5oejw
zCI*T92B9T{{r38R{cgrSU9;b>*&o*Iv+F(f$2GeKNVdEgLDUz!-mpLaU+YJ1PP+*B
zUTp<*H!0o_np{Vf#|e)J;$wE0Ky^Qe+a7jR!ane?TTP@9Bqlbiv|4IvQkC?qvP7ip
zH9U1PUDUVlPjjbZ4Q5@%jay2pJU=khS}vDpB4=G!SZF0J1<ZZ_uS7T?t!pcNR7Zu{
zpRviz50?}bMXb_D#IPW!T}#Rehmlsw!c+it75x4*_)#(;gWel395+X&MNA^Vlf^2I
zw=ks_k#`sT`2u33vUKb6r9X+?3dqe?3#H5O(w`cXO_iZKV-GYC_&8Chpfx4C<(6e_
zCxO1^LDO1heynMww9N#4P7%%nf6R#Jjy-=yR~%8a#5&V5^?a|ObGL-L(_5k?q6iL6
zNeRjjDa&beDcZ*+VwO>vs{&lX51n0H{_@grd_TE=nNh8k&QSG*KotQ7@GZB&Pg3N!
zr1WYj!C&m$?1*6h_NPhc#(?(ellvDtFa5*l0Dxy0;v?Wt@FxQO$dM&QxC@*Q&{_Oz
zn&wIu8QKtF12wUU&3r#saseL^4qmiqV10jzYE{{W<PSKoc2c;q$~rtEsx^TFidqs8
z1x{*8T@5ky8&YMcD^#{BBbpx4RizBgwqMo|$!+$W8=mkw$qH`BTtr72E05YbLCuZf
zj6r48MF*iL4O`Z>UjIh?f)Yb@U6@oxX$A5<LhnSL((!uVpT%MWY~&Is6ujg%eE#rV
ze_GKhLy4dP%e)TQI8x>Ww#T~gZ!&8u^Wk&~vbn16(BkBk(o2-nqsUns0M5+P66vu>
z#{l%pcb@(N@jTUwd`EnwY{dV#6_3g#5wAB?UvX*|wGMQHDYK=3vY{n&lu!{z%)nm&
zkY(0YX6Ha=s)&%ae`xNcC990_B@_4oybpwmxF4I6pvwN{P|^m0daXXEV<uEi(p2*o
zn0l}WiBH9bDAysx>l;@=MciQOan<wy%r;QCTmS=2Q#g-Z!npYk8!sbvg(9gE;|6jQ
z02$1|0_A%L3mByqIv(ZVy2)rsxX7?;#im<Zeb#JCz^YiJhPL*HR6Ykx-YoBo-2|tG
zDz69fOMia<&Y<Chmu~I{srjJVfn97K9MxnqOZ1YWnPKlagB!r2#m<HSfa#x_{{*W{
zkz*gRm+{NQW$XvLtXwAbZ3kyPzyH&;y1fy<(zb?)^o`jW>2VtQ@97}z_7W$|f8!{J
z2WJ~gN-paij(y#Za=-ty)Aw|v@4Hv(dxfM|X>|Vlo}+0OjC<qv3}*j%rFq}F823f{
z(D&o4LJVJE4U5F|%_>$JKa2!>S0Vc^0;|@By^a4R8ykPJT?h5+lboZPVxk~1QDVDm
z`BFl+DTxoUM%|QiB|7xftgr77JMQ5<iNjWF{pkFje2gt14Snd*uB_$;yC0fyjifko
zM!6CK5QGny#Ydk+gn|AQoc}WX1&sxZJz5l2af4b)3cH?sk38zgHMmF3{)zHiMqD2k
YacndlHZDZ|xGoVR92Sxn!OgG!2Z_BVSpWb4

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/index.doctree b/.doctrees/autoapi/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..8ced3bef6d38b315e44419dd732e8ac228362276
GIT binary patch
literal 4132
zcmb_fZHr__6`r^5>7JRMm(A{?WG7AtF&oTuXF<@F7zm4^JDWj55K)$!>U*oZDpU8x
zs=7NpA0n%mumkmN%P-<zQ2Z3gKggFr{*dT%>fYOZyIsVf1I^qzb?Vf4ea@+;oj-m1
za4r0~15*f@&4w&VvOM9&yIzvVWy&+_pZhm|??3T(U0<_Hlb1T?-mQTn7E&deXa2Ew
z`^0XgGQ;}(w&Agrc{V>TI-)DKAN#wa?>bgm#aml$@ZK+ee)0jIaL9vvVg%pThkQl~
ztZskmD~-AZ1Hbj+$n~SS!eY~PYdz21mXaCw=jE8W9cHbT<I=)DvS(8!hIHSK6*C4z
ze0MUIR$slOEEbC~i!U^v5Krm0Et}Go7YE{Hu_x|`O>yjgJQ8`zN4m^1u1A--z8F1v
z|EDUyZ2mz(GpZCDvC`(OkRzETd^XeuHoS7SQI27BA4ajMW`$@w=qDcvX-*0@<tK5T
zStc`cLPAcb9M>_+lM`|wbgW!>(zt_B6i$Bg+T`BjXc!!OSKM~*{%_)Z9RZj>s?0wZ
zufWMr|HdTGZI)Xe&2?$p(XBM8B7EDe-)o7w-xkHhcMv4;DuRMQ-SbSmhMagEzc=uE
z6XY$j-=D3!4sMc(cN>#?h$ajV!Eu8nC;L)-PyB!sc0J8auAcA&hoLjH7`C)nGKkN$
zSiPRaor+`e76P}`q(SgNFV52<FSDd&ca{5M4Y`xYR`Yi9*L#h*s~17`Be#t-*D_@0
zxx3jIrqT)JS4blNnP+ZKX0a+0xI+bA@iB9~l+6-eSm7W36Msl|R)oq#aZFV{_Sdc}
zWRjqeuic;)Femh~G94qO;Hx#a6YAFFnY!}N+z#+yp%(e$aqW*TbpAC;ZpnG>pf>&V
z63Ja~jgYp&N#}pn__2~xB~-6TsFpk3`2)>E(qBp|PClyO>$LXE$veW@!u)77nqs#c
z4-t?N_!BF5RfGZU8uoWTit4%mGATCZs&!KxoRv)d?8TY78alG+kFe=!n_UbmJOv>B
zp#sF)Zm?udD;b|Jodh_%u&Cbq|2P1Zu)pldmL`8`YqA{Kb{O4OReY<0To;;`T6%Z-
zje--(_RsL)^H5t|%k-2lnv&?ax4&t)E9_~??cYqE1I-I|8t7?^5r3r+{|%+k;4WcG
zrR{$Q9lMks<+#QCaG?MJ8f)-QJYeEy5Pl{e(WSquYmEei7St93>?`EyT&V?khK)s+
zSV$+4I}o*5g*m_)`q6&~bkk$STkfXSEHf&!nyAqE6aZWG*pAIDQx$g9JON5dx^La>
z$EBtY9?kppUPGz+w(A*gt<0tvNsg$cg&AX_(E>RD)^&f$R0*h?4=LTev_)zCxi~m?
z{kl2k+GCLbFpOnsfyJ&ymx8K5%9(fTBky)A6(Zbw0fwD=S!#vV0MHfD3_!b3r9-%P
z<y*~evH8pmX!ZeaMzcw-Qy}HO0eBY?V2CCv1A#AB6s~xnL<D0|Gl$^5`anIHoVz{E
zi(K1C@h5<%vkkW+n3^v;?ugGMyl9y%O~cf2orI6eDX!bUWIBTv!2ujhLFpr8F(<7_
zKbgU25k{{(0hf3E(oRl)cxoiR?<{XdT<cs%f@O&UhGW!%ZL#(4pz?f6N+(G1?s#Qq
zMTGc=+pmN!6ljj#S>9MV=?=*NB%TrQkAguM6&1^-CFUf!JHdJbea*e3d7SGcLLWrh
z;2LVB=dO!UY62VK4w*FBAi8d!8cDPuxeXdG21%I~(L5jF)j*I)MV&E40uKTuR0D;6
zK`IO{lRTQGlud_xR^*09?UNxzVjDe?M%s|C2T>vo8!IxT)^fY46V%)uj4`N;P;?MF
zsM(^X_q*Hlg-Z<A)b9}(l9nLeyXf5WB?hP?DmJ7|HboXHnM&*5KYYXO7ra<Zab(OM
zQN|3qQ(d@kHBb(+85UPo<k>LSQ<T#!;tUFaGqOBIcr3!v1O4>%x4#d6?(0dsBAzG<
z@!xI4qjDML>jl?C$?YWE0W=s1n<`w^w``0OLVx#XJ$H<ROrvIMSlkeqf->aV9hxgl
z(6dO$sZjVKdAC$fdB0~;hAR7QiTTyQQ7?scwTx6Avudnz$FabW9wYEovBAs1g*bgt
z7Px}mLt$8|F#wYlC^DNM2Wn8bK`3F$@d}$vDR))Qs!F8y%Y{eEU<yl6zE{vdaCE?H
zcPj?hMZ{AkRfLxtfo{<HsD67PSJk8fv}?D?vnTWpqUo2Q1htDy#ONqd8#b!?YIp-3
zK`CUG1=b>KFiaAZyBkZ?>SY8nq)p}v>Q@-f?=utHJ-x3!@#0}k+9tk=c2|5Yz7k)G
zhvEzIdFlGJW3LKmPc#45*W$}(;)`dcc$Jo4p}##4_rd&Bd@lZ8x<Na(bmZfA*4NhR
zTL!U2vXJJ`ENC9y0nHJzP&{Ssh)TR6R;)n)B}Y2W)jWTKOKz?Zn**-jhzM-`=I*e%
zv2+EnSzAKxTdT6_9gNpNoEQ%WijCWibEOa&Sb~PL{h+2lCv$gU?%BHe)PH~~!gCSz
oh*zsY;#3#U0ceT5J12+O@7AL}a&1aa-<^aovW63y0x&rL3l=#`egFUf

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/args/index.doctree b/.doctrees/autoapi/lmflow/args/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..7fa292a9da8e33c4a9147a9e3399876eabb1186a
GIT binary patch
literal 583437
zcmdpf3A`Ltm3{(QACLqRA`q4enB)asULZh#Kmvh;MUsGQf`FZVuj{?;PQUI)FWCTP
zlSLa)e2gf9>$oEBqvJU4<KV`Oqqw0XBks$%4=$kpbM8|2R@JT3_f~c1;osknSg-5e
zbM8IM_no>`b?b)d_wF@i>J<E6dv0UASgP(E%ohsPTETBjv}YA+!>uvD+MIaR#PS;_
zuAk^{&#vcpG-|E-us_kB0xgD%rAndhS0}EXXwSjs%~GW?DEiNA_`}Uotr~PMPA|?V
z&b)e}zc{-+y;*8jeC<?wpUT)srM6=*UmtBuU^8en?Gk?{st=!h=>#9mjPX3^ZcY^U
zZ_myJgDvjeo*~A#r9G`us`?XKT0{AEU%uI_mxfwRm>bSz?^2<G<DNNO$u}DCBOFh!
z=EwNatU_yC{V|ju-d6WVuuZEyvzZ^oQxnCx#YM$^i-#8XE*{oCZ)LGI=C7=`s#U+f
za!0McZRMuTXTyZ#|G@woEBU}y=3C8Le!R4jPkALyd1a|u@OKW@8!(^4&)=&xRGiYD
z&VJL3#O+r+0>)ol3>G>H{#ye7Erb7-<6OWT^%9#we|y?^zEq!R&mXH5S{46PPq%OT
z0RHO={m0^rEzOC1t9VRtg__uw;^AOE@WpJfc7C*;A1`VHYPT;bmKt7TxLz7>dIf(3
z%-!$``DR{Nb&a<fqN%dJKFUI=;pM9ZZv&if_|4$Qfxc!j-}HuS)n>j_^_oTB1Cxjj
zBegpF1qSR_Sa%c;-Mv=BFL>C#4!ug%(Lo+I3!i0!65$FaJ%h&zH)%f6jp%`O#lwmN
z5D2rRC%^}%^DHb*Z-4c?rdJvpuhpAidw;uM&sRL|(8d5*WVq5QfIW0S40t0Yzf$l<
z>a{VixoaFw4|o@hV}3$k{nx&_e@&}Y_j4e&0jI^udcN8KSB$}rgFMZ}dC<7|f(;ut
z<<7p~qTE^QFTHsErrf5@xy|b@yzsm&=M40r!!HpY3OsHQDxLTL<=j*ZY;li)eT*^i
zDQ@+%F4%JRdFSNL+O+=Si#J~EK{5iGl}4+O8hNi$YBasth?YN)V<V+ePh=-^Lm%P3
zLE+m=uDCEJyxAJB_+jY~Sbn1qFuR@~Zo&w7TtPaU#u}vS{|7Ts{=LfGzX6RfoSivb
z8w;^6<olw<i@ouBZP*8*DtJS?ylQJ~d>1f^w|KG48?p@_AH-3mF_a8E?b$oWDr*|!
zdF1z5<8{B`*SGr<h(~1^g94Wpz*zT@cqD0kp8TQRw_cGg<MM@}pOm>?Vs9=q({;Q#
zgT{kk2vgegn)#ttC11}C*DATjxIe71QzyEd#@e^%U3K}j$6bHzs_W(dvl-v1)U{vO
zHF95b{PAl}L;7#D#>VpXUF(4DfnVDTMfnA}GpL=OXz#E6m4ms4i>3?v;D4s%Qopm*
z#Xl>x(NVvSM;D%V;l@qpZP}Pxf6+OYZr-@%5{cG*eX{bJ9#ti5IlDchIa;r^#$#s2
zst#he;^>1}&}B!d&@4`z--m(r5SaZP;MsS}SnA|NO~4Cv0a#-W)e5_e1Kqn`+hII0
z9V@7KAB}s@Tr<9FO>L;WW~<5W-f3DzHE3$FZtGRd$o&w!`;PlnWa`{bs4lhfwAW`=
zA{{xiQfgv-ph+)&OQhri)A%M-a(*{CIib8{V{Y#?zH6dAV|>>NFv*0kSEJ~R$HAN5
zfa*K3)@jdxnFEkbv=2@!iYMT@0~?J&nKx1VNqZ*zg-kQi-W!#+_kq9to#U0#aH%<Q
zJ3GB2U$0Kw0X6HcHM}m0TtWc4?OMY~aAFIS-h_hFZ%(xJD%t2hh7FHF!QWd<IDaZw
z?kkW73yU+9L%yOr#7{ITS{@6?zd-?savsHQloznWm!RS2LE)Dpi5FCz9TJf%T~wY6
z7nGK|=o$Pw4Cu2kphv<3>Q5eF``Ad@@Uq2_lI(=a#IH*g?Ujxac$$4j_Aeg~vlrR5
z56>>9N+nhKL~OyX8BNy#{ly=5PO0*79c&zARq^+F?vzi5e)j28{B!Xy#izFvpJ_iQ
z4r+W_==4344CNJU&Y^r@zv;H50}4{2{Cu=N%Zl>Rcxaax<z-xX*4)T5G0A8s@=&Ow
zXOkj5mIlGiYk@FX@!w2IvBauhG!<To_QiFah3y$!ww49A3F8+JOP>Wdcd%ws7Q7Pr
zrOAQ=^ehlF+Jg!35;om%0^o0M3Gim@!4u#*(b)p6%e%PptO@V|CZ=-&aCKczfK5Rl
zCzinBoJeeDN|W=uC_S1?-#}mSWI8x;P+ER{OZYZRem&N~SxNczB=k#@UrY4-N}l8%
zOu4Tx3xrb+e{)N@KVT1@a*vA6HYxW-t~_hXP2bC^9xHQoRw+j}`ZUySa0Qp#s67oY
z6DMGi8>)x5LF>0f&CxzkyXy5SwdyFp9L-0sFWv;bAB^_a8x?mKuSGSlgNAbP;I-_M
zTa7PtYalNd)g!-(C3ck62eQPnf3OfitcA=xOWa@(`5XKl)b-^v!AE*b(EtexA?4G9
z9&89lL7jY;EJ8aJ>c#ttufb`+0*5B1o7XcZt~RZ46d#ZSy!=hmrcA+w2wZ?3EA95{
zTkyUM{t~`h(AT$t-BN(1Om?vry_wKz;4(MfF9>cqz)gtZT0P&a)d&0fE-N*<D=>QJ
zA~wKi1Gq^+4{>v=(k#Is^Ob1V?6S{>yD`lc+>Z*@-TV617xHkY1{P@5a(--ezOU~h
zAFiMa%yySV1%&R*)LBEk`QXE?wo$P8#_jn^3+=Bdjd(HMq=T8fOTfiE0>n)sb>E{V
zZesNHomU<4;fCOFXuc8EhBjT%eNld-xxRwy?!>%Jjl5<JZ+yX`e5>wzbr9p_xg4<N
znJ}9oEE8@6(`x2~$hh!=xHdErEMr9On?cjy4w<-72a@5}?W-I_Y+q#%avrWMHA~z5
zxS(LP9^SuWOHkNQ&aDAM2znbQlOzDABMgK>*27Y2(3Jl<216RoS1P-_;bN`U2-yVh
ze#NuGgCWLaQrw?-vqUWct8Wkg-JfcuQZ99t^NgKS$&jm^QpRVZ19Mn7Ssj$FVMq>F
z!{KwL68OYaa-zlVZAfms*qsyBsa3dl1Pqw|3i74Q`pImYRN=k~`lYFG578^!$jp$E
zyS$WQ;9*=3EZ@kc9WHC}H@yT=<&i9i<WmMXP+Us+N9@7N+82w?R3(PpV2!bSEmxkk
zto;xZlkB>%I(jy+lv3Wwhm|U4!Zk)XXTs;`oC&w3<wSGlTj&CjGhsPe&ODBdGOj&-
zql4j-a^{E7FLTaBX0``&=BsSl;he$WbaT>ChB@;T_TV}5pQ5u(&U~3GPt2M2X^}D~
zY^M$t4iBY;8HK2;nZCWlW}%Z5%+lj)4~BU&#h~oXO?xvPW}FuAW9-r-obIvHCgeCZ
zgS04q9EbMd__2RXxdtJ}lg_ofeX$Z`P<}{)khRb+4TKz~({6lndk{sAVKWb-2>#}V
zA{Sr}jw1b{Gi6Rmg-f{dzY`Q0ikRhDfg(9)bN2#8)&zMUT{8%SNc5};LrzI8D~cd@
zplvvU%!_K)0OaMQU+n<oHjqI@d=h}%3;ogn$e}ub#OAgK@#AJT@i2biZ*KVUHtfOi
z;|9^0vZln3>$&n|{8$&MLZX(ktV6PjQI*I?Bc^#)V91A^P2LL(*%IV^@(N@aO_JN0
zaAk8><wvpRf6#^;YYs^smxek|l96DCI^P2sRGuWE&M%=~8q`^;qYg34J%~S#GZTdI
z2Y++JpTA)bjz8ZNohiFY{CSKk|2x8;1K>V~ZklHW{>&AFvNw4z@Mm3s4~F&F&=Xtn
z^CrAmn@L_2Urs?AaeO%-u3y8IQ%Mio;mS!MgK|a^u55&UX>jFm9ajvq+k<#=JeztL
zPw+QAY3S&~Rh1WD4~{3tiq4chC7!I{%9HWrxi&Y;c<3x_+cj3TNjWmjX9YHlI-8I7
zPP$bR^_WiLeaLWWlGx0I6zVaZ#BSEoqqy-(v>(TfgA=7|@bM}#2<+hFZjeERdlGy+
z0R7UyN9Zw~<Vo&96nQzbKo~{vH#Zb{5BA_Fa=YkEnN_05OS$r76gfP&KgsTe)+2Y!
zC4O{jrjX;)k*Pf^kmD21ChG-qsH!4<KR2Az@$)9EP*p{|eP?-5boeRSNK_SZc^WMI
zob<3AEc_T`Q1P1t3x9xqX<#8#RTyTs2l3!ZHuW$b;BRhtFmrl<2TzF3lszRLe3vUv
z#)Cp+l_YMf4(lW~Ps2!5*)z%%HXVXSOT5FVKbeR2+C%q#o+Acj@4jAO(;2}Pis>Pw
z&?S$dfeD>X&nPj9PwUZ69G~VJhM)n|S)`lo0P1v*L3tzzpf*FlGyrvke%WK1-X27#
zQ`y|Z2!+4tsYAygUiaj%2d`L87M&@BN`yL*D^Er!KY~yOV|5Xt_K-G(QT2$SpB42}
z&Dnu;e5AYbQZKg}pH&Qjl&OUYrPRBurXJxc%G6@r)+kQB25rT0YJsVT22}TxzP1CZ
zSAz^Huabc3AE93wpbEW%YMkF5M5<S@$%l~&e{(~s4`UCGRId=7DT_*^x|1tUMylwo
zB9pa<Cv0VvlvRwvmM=t1^Q^#@e|I){FR*1pP^}mrYYqc}v55&?&gz2HD8Bp#?Zxrs
zK;u9(fcYKia65qc7095{AqilffqrQKX0cugnJ2gh5$5M?_F;s<-`o&p?u-Coej++k
zCY1>DL#{j#VLH4^gJRaf)1ya#c6)Wb$S!qtVR)QAoLhzlrp!`T7lsZN5~Epl7J5ir
zT^J;4Ikgd6CSF9^<<$BP9!bin3!q<`oC;lCn5MS}GioiHdpM);H{GUm4C2*A9(!=k
zUL!i&g!oBZd16MjkG6kLMe?&t7sbWR$dsQITzrkQDQQootMe5XL#W1I1d3&Myom`h
z6c?wL8byu!(O!a!jZ!t-cz|@c9d5h^WKdzAgd1;zera$c#Kq<b?m^_Zm(4zm9Qd0X
za{M#);K*^0=uDYZa`D|<c`|YgBqE35^)!hcZk-f*d?_;RX9aqE-r1bJKo50)Hh70G
z`UYSaK@yvpP(<CIP3&eZJ&Gp3L;G<wIXF?ehA4j^gTM|^egiV7@J~XNzd^q=h!VO#
zn>@)qh$_Ei76_vX{^o`%`_Bwe<!7QZWmbtQKjzAlQDt%X{%lFU)T&-Hl~7{QQsKvQ
zakz;^>?GX@Su6%+Z@6CI$3?+4dk~($e3lSmXlX)`=XYI46id!WTXHNpl$wVIC!5Kb
zu!EEHKnCT$Bsh5i^h*OL{rX}7InO<aC>xm(!ia*u=?O<iE?(i3um`VG&J>*~%SuE!
zohwg9lqE@s0<jxM70cENS+0vr^jT4v>~c0B?VohlH75jFAKDloTrY%9nXu#d6jGw-
z@Mg3JM~D4F3N<+RN79jYaPUTuK?Q9R9J~kmrGbNkbvOu5Y!BxD18mme{Kwzi^8ZuV
zgXjNiMQ6&CGXG!Al_%%_X0%I}x1m_r?y^*cCtr&gt|Ok5uTZZsf5q8wws}vtgI%V?
z#uv%$q8m+>U&P*{4$rT=0)8}YELDD?p|xJOX?sg)8z~9nz^2Y>VCo|iYI~{RyAyeR
zV}<D6Qn;}bvA5Lhvs+d6kqz<5jD>CO7v<ptW`$EDYyo{v7~{CV)ZCy4+h1x;%KfF3
zm~Ot~z4`=tf2sCD+)AneI|4UzdHA+QKmM~^Zf@P%4PA}{!-28h9o~#3ajRbVhDxyi
zb+Dm}kP+F?#bqPhlL7`)RtS@z`W+b1yNa3Us^yHu`ip8m+!aoOq9phNj3Q2gnmaTG
zydns*^qaU`$}FB7(n$<B4E@qD;6j}Nl^J`nF|;Aj*=7dMXH&+FRe5M@7R6I;EZD*x
zoCUXt&NeJ~J|9CO3tHYeyOoJa_MRkNu1-O!X;&Jo8PBzlhupz)ELmQRI|dDcleyz)
zLu(ypb?#6gSzZp^qPr8$9tkjBu}A2~W&HC$XsRfGAaE1>F&8!pfsX@uXThg;VCVaQ
zMPk1zF8MSL(Ry72x#SZ;55^^D!jJ}S;W6|@qZzoD9((|yJz(0n-s^{N0gdtbXeqkf
zW>fdjr;oG&b5w$z`lrDsvErYWz&-ru;F%nLawXq{E!?zy_S@$D^$`%+=KriCes2pu
zCMCbI1z-EvR~vosk(BU{${oK5jB7;D_C7GO74Rt!rlh?e>SP<~vA;w6^^1@!{|qHA
zD*q@D%)GV%yf&phn~ga#;POwfDTjIUl+|$m8b~z{cNPwpk98L6N*?CZNyBW<*S@8(
zz2xs0hRq2l+H>V$E)SdP)#{qXJ3v+F0|5(iJj5FM$GhOmVek!tQelFbLL(eV$?}}p
z5qt8PE9fbI7pB5e8@@^g^Pgz<p%viU2Jiv3@HZ9OGr0DN;$Pa+nI9*LGupF8T9pd3
z6xZLEYu1Ky@Vzcf=)bVvZiSC1uwPX(9NQ1ji2O@a8^GuPBfP_Wy)W}M?ze~^SyS=Z
z!CENx7(QhwPHsZ1ri|#_&o{cCR)EiWMQ15K_E0_r1hr>^CD`{K%4_gX78m;8)!GXF
zp|m^NGpjXuI7j}xf`8KEj`qIdr2Kn-Xd$N82pY|;*5q%p9ZF(fHu9TI0HBFGaEjj(
zrjqz3_AyU~Y~{y6TKg#Ng8|84mbe!`jbY^{Kt&jJ`^!Jnert!mSHM83{B!+$7=c8-
zhksOgel<owB1`m_p90C{Kf*s01YHpbVGwk7YpDE;Hlp&Cc%I`<Xz*$n+X4wjxGB$q
z^X1v_k8v0RVhqEW3ths-Os??p<0bO*A|22JCuBwsHHH{^TmqdqdXzmlEH8$CLenWP
zfnNXzmg4`-C@;fCsHxn~f5+onC(1|5-y^_qHiY<P*b>|c9vFZ_%c*NnAO*NWL3tgH
z>OTDEbo}Qv_|F-^(Rw(#qkJa&_bm4B2KMhp_V3y3-*fQa9;NKz;zJ6FrHAR80a+7O
zeoY~3j=moZlQ~Bp4ur5AJvW-8@5S>xN9DLxdVVtKoFP38Bj{0j;#|z>sm2gX&v!s4
z=k!EPGClEFrY9ap((_zM&v(JknbK1VNSB_!#L=?!{9im`>B){JJ3W_#_gMA%`5w}6
zouHN2aDhPW%bGP?qdVBLtc6l~E-SC{)oQH?8y1H?liOZ(<!MJX;ZtyHyk%^}%hnBe
zr?K`>7{XWX-~)YYyr7A<Y~4Dz0%VsJ3uRfTl5BQgIL-BI7A{LG?%Gp5zKF9BkCc%4
z5h39V_V&~r6VWBq^8VOmQ5n_iYoXB#cX`<{lQ-WxP{i=$d4MPBtC0Rcm>FNJsu077
zLRH9gRKgK1>T6}c)g4A+Rfrfy(j%n9Ihm18jUk3~e}&|CMmp3akq(a~(&2Fg>7EDC
zGzET6M7sVg)%U=LOE+)6!VBH*j=)Fo1?lLTLb%-^lLQxL+UJzZFAxKl2-~?HpDx5;
zUWe=6F_gt4PBc=1R*UEfCu+Z4#k(yK!jf<kB;hRhNtS)>nac|B`O00(IOvoQ(9b0R
z$PIx=Q!17Zhto+2avhvwSELI92pq>HG5E?r1UILy4Zv%y241fqb+@OE$HD6^?6RoL
zRa;H)^`sJ;2`*H2RsO^g=tWxi8rra=SQbEP@L_MjqP@Tv{$}`4ql&@D2+Y(PKFSO?
z*rm*8pr1gHb`*fyB0p{ump9lpf-0N9%|&74)R#9>fXwc&m&8iP>;ON)%N_h0=i4|9
zRzy9HN34iqM`?>FhB#eVmN<6(hV^H{_hL7mZ4Hjk)yk{ENOT0!G6UpnO;j8te~Uil
zAgMJ}iS>KobLp~FK>6!H2m{LJM}cymW%)vW2tJMCkGO#4^nJofA08WtB~zg*jN)z>
zSOC`vusEOAazBWOL9)O_p~VLH7t>lO;OAQ4Hdn}u%7u776*d~&)+1~@C^7oY_^3t}
z!^fX9f8fHY#X66Mj~l^CiTJ3chQvs0AHm4y14jN*gnDX>lmcYJrNGFma4HNVuf`*W
zk?d%)F>)nB(FK=o*<cNkM|FTm6qXqwXUSoSLM?RN(gYI<fK=^Na&5=c#0sv}6-zx3
z!tn93C_eTve5|#q1s8a{Es*M5OWq7!E#dKAS_=h_LW>PNzM0lS!6VlK@Tgl+IS^0l
z4h-iTQ8^l%)+2O$jsYE6;G4lw4J-zZV=&n?#pAz2zeI3UX(6!@n@6zmGQh@~Fm-Bd
zlmaqf<A38!7&d+nj~F(xqshj`V>Ye7=wfSJ482Di2@e#M855^*De;2w|ImrNVB~F7
zD*aLTSGp%$k6B(+|2`1Hu<*(#7S32!$=4e$I5>OXaE6D6MV5)vp{pehK8Myq;h@lB
zgM&+GEffxNE%-E)u%L1ro=Sy*250pM1LvE9+>C!}P%-@b6$Dor{5uZ}l!$*k4J7tq
z(+Kum3E1}=5!R`(PYRI973L|cYv-*v3x<7pJYv|#j-CbBxB1L2Vc+I6yMcWuC60YN
z!E6fqpp8nQ-FPZ%9k@LZ!mw|v3H!!|T(IxufkfvT@HXgbiG8o3wNTh6wAf(Z+h{Em
z_Hiu&`_O{Qad=91TsW_Y`p@93J%)V{dS>iXgNk9_H89aM*!M-~M`0g~MPeT|jbPta
zz`lksacb<70y1FV_i+{s`<}!jhJEa4vaxRkTzqa`f4MdKg?2)SkRlhA8UJQTIdKI1
zGrE#jf3lTIrKg3TrJKTqmSz3-$3O_fKtGCsv+!DYEWgtQ1^3%8oa5nvkrm)<=xT|A
zi)bwr3JNVYD7c)~LZKkn0w}0#sGNr<Q{kY&VLigZ1I^)XMnW~L7!v*tLM#muZUz%2
zBB4wNiG|oWf`vX{;U7eJr^Z4lKqgt3udM380?vbBp^rxl3)#_RW8uISL>_A>)E`fY
z-alcIG9%%1Atnxl*Ma2(5Q;`Bm0pbJvetsT10f6t%TXMhxvYvn=mLXx2O^z|!Iwi<
zOBj5B)<VIc&|(9F@1nI(Fvzt43<f4tuETS>>%!SZRD%Y0^#}!{_mO#snn6&FDh7fr
zm}wdi{151t2!bLNBnD#J2nLn`19u1`r^Y}jAOi;e2&cg?@W*(>FpwQhItE_UH4J2u
zvSXkS6UV^6q91uNIGs0Ai8N(SIFZ67mt`&Zv<MXO?nNz*fpu35+@Bm5Sqkn4T`e*2
z2wDq;fkKN72CksBP#DOyFbotXRIbBwsW8ysu04c-Jo?NSs74jTz^5R-(qQ28!9vLx
z$Oj`a5ZgvDum%|TClT7IF;EJ~fPqDv2E)J-9x)7LN0W|$2!R)Nr4D40vSXkS6UV?C
zz;Y4;c_WobH{rRgb>NEwA*>E;#xW2faNMO1d{rRQxemM=x>{o38)+>R1_~`U82BDq
z3x$DP3&TKRLghL<r@PL&4m7xH4`CpWJ~IZYQN=KD7tAyb27VR#C1W5TjKn}}8^ORP
zVBodF$f+?<3dn$gKf`G-4E#ABF$`o!lZ}DLTzt+3dkCcN2cw|O2sn*Pi38wXa{~b6
zZB!CX!&6!DZ%QD9;oq(({>@m{7%e!#-+_TdXZYJ6x>~~DQM48c{)84A_&bi)Lct%`
z0`Mm-s2qo<bjMkvpTSu@Lci$!U>2Tc@Kb||f!|*s=hJ}SrC^{$@Z)J9u@9R@ux}S&
z-!oaUPYTF@eHEMq!@e;*V%W!yCLQ~(xZs@gtjj+N`%qAJ?Bi16*mnz<O%#9ASR0i>
zx8kX+*mqMPgkj%{<Jh;WHd=Bi{azhNbjH3{L03!c`$t*}g?&Pc4fcJ2)<R()*TS%m
zTTnR;Pw9@c#y*3y_7L{5&@*G78dMDXu7`=H!M<-mzhvxVW0Ba0O(WR%V!*za2opmz
zrF%_Y3dn$c|BJI=*!L?uV%W!yCL8-!TztWW=bpFT8vUsI!CX{k{F@==#1U}DyZ`}Z
zE0s#K@MKmDoE8XS7<h9O17|I3)W(aYybB5*9LRJ=!2_YIB?>O1wNNN1wAi5FiL@39
z1-TYLL1jbbJUpp8&l(2}4(kyPM(+oUFf}8g8deMm{{{$<1_`eK6D1;{Ob3aD*f@fP
zHv<+<6=qJ2g;GEUEF8yqFf6<Vj~EuRqshj?V_+F!i!~Bbbs!4LjDyp-lsF3Bj!uN1
zA(eO=l}0ZU{*~@Ybsyx`KnTOY+oKpb1D}bhx**{FfkbBnd^L2nM8J2@S||h*T5J&T
z!?YF(0l5}d2XYH4$Kk0|_-AldkMJ*g4a~yRjDBiRG4#6;CYlEQ9*2I3=*QDQVjnh*
zVBhV4eK!jer^Y@hKqgn1qpY5V`7O?ZVc+lYh+!W)dNyF6{i84)uy1o$un(lfv9E9c
z5cX}>9<ALMPi3wDW(7hR_T6d3KBvcE4htkYW8cBh)e`%Tp|w!hC$!jL-zl^f3j4Sg
ziG5%}g?)HRcU(BHh^Jr-&f0U>w>jaNXf>!9_U#Q5O@n<`Lq8JxHV4l{W77!s-3iz?
zU6?pE_DKO5u&;@;VA$8fBZhtKXtJ?y<(3OByWkwSk8WN2QSS*$VVN;-mXZ|5!8SUS
z*MO>>N~ybqzoolU)qyVygfJ|;CyIrA%c`~QwNZGY!=)B{eIV5t5ATPrmU#GXS__4T
zLW>O^{tK;z!b7eF@KCp+av+{gg^32I^#~KA?+FLtYOV>@z+$-gQkZNST>KvNOT<N$
z7P2nH<`HbX2e9!rVd~V_C<Vx53-gy%UHE@E6NZgX;Ss||b~NeO2vz=ujhim9Um>Kh
zQ3}hBjY?7+8|NILurX+-QfnTb&RQq#8wg?8`06+|!WCY@uQZ)j3Ks`bow4yS=xT|L
zD`_nhHVQ2^*tnL~LSZA<!mv?UQ8^G#>kbU(8&NMBoVJIsF$hC5HmZTeuyH0#HVroB
zp<gmK2IG;~h|MF|_-eq$*}~MRu~7=hfQ>tGCJY;Q;Ss||b~M@8I0#mBDjScax0gX#
znUQg(7?kPa`4#9*UNi>HR9f95d@Ws-s%E?^5W;Zrbx~ZLjcbRlCF7d|q0X@Qdgy8i
zi|?bgP_QVp*udf^Xe|^haxDOhnh}-z@N6nnG`OrssJJlxMjeZJGbpNY#X#{6m}?qP
zd=mO4f?|*o5)-j?1QTBenD}yG=+u}f1;|7T^Ose<_-C95!^FSf5yM1wG})NA2H<D&
z3vw5qb?(Lumu}jaJNu%I8?7&^2Wfna%FT?G^9_SF;pPGJ1Kc!rQAxM}TV|~-=LJF-
zhQ2Y1p$D>el6T}9!$luI6qFmO`@YL1_>w?@GoCJnu9kRuJgtSoQ=!EMPuJ60C_Lp_
z08bMPs@#e#b+?8ymZ&%luIv%E9vR2hSiYEXR*f-+v-`lj)8MQR{StB3m_ibBv1bHx
z-w2qypKw5G%#{K%VD5D|DTcY%;}OGLb~M?TdonzYni6pb2*iy^&5XDE8$`x0-zU)5
zyh=56P)Yd@A||9uQ`M{Y1VR|rzBP)q^YMA_^ho=*K)f^3z8Sh&BJGE0EfmrUEjCE|
z8Cna4v|I}yt<jvyrPwYN&Klg<Bb@ajI2#XXGs>zF#!&W7m~|SI{W0`QL|H@XNQ}jP
z5sZB+VC>z(_^B~g3XlmIW=X5M_OCc4hOvLcBZjf;s2dm?9*@r2XbtgsH2Ez9auVVz
zZ{ZI*v5z9{yFgB0KRjX+U`J^uz`oPig$=`(z(*zh%KG|fYYaZ2*0B5PO1Mu1|J4^H
zc|>>qc*2~hJO=)uj~I0IBP&b61v;EqeKKG$(kV<U<SW~#&*7^oM+TZ0_q+=dFa1|l
z&J2W^n8xzkZiW$sb|-!=D&f=|^`!#w3p<R&YL8;WNO}Y4Wt@{4veg)3kbM|*vW9H=
zc{kJ~@S16rUsb{1BY6BSz~dv}XN+DW9`_ed4|bBjX9IlZvg&(zkAF!ioMm1MKhWwm
zi@t{+b>vM7{&20HZ`SIAeV4!q{v&>3F<+`SyoSHsujeaXZ5+Sl0SEBaO86XV)2}x`
zPQK|4=d0e350d?Y*Q|NaailZ~-&utrj3LYfA4q*R4oyXBQ&?DHeK^%41W{Zk_Jw#D
zd~;3Uo8oNzXcXALgtp36N{#quIs2gzd?qE-C|7FwWBNCq9%7cu;io`zwK{w#v|04l
zc#USgFOUDC59>aBGqwTnEgi}l)#@I0@@gYserV9FgA@HiACu?78o}_kargjd(;I=7
zJN(jU5#XPVH(x33W}j;r^v)acw$!Trfb^iZqg1J&kA*aEq+T2I8p9AY!^OcqzDfeq
zng&4;n#KX~X|Tpn{FzQEXWZ@LjPjk>ntvnpec+Ef;7|A(&Fpp`3+2*C$=AOn3tw<;
z*5xqRvC{i`C%q3~YpxH~bDheynnT4GS+z8M_W4no<eq2Y(3FSX@8Y2#P2Uf0dJhW=
zGOf$+M(I3~unUhQx>USPb;01fg=s0@4*wTm@|`>l@C&t&#U0X^wSx5J84G|4!0MGo
zc<y1YWAgRM<ilKr@6$o5*S41megSh<rM(a$(-$FHrfc~weS{nza_pPEg46Y%l==S?
zXPdaD89_-sF+T*h`4YFygSuuS8UNc!mq<p`!IQE4u%;)LfV-2Uyw@I(i8h%-P#*o2
zi<N_f`X_MbFQ8{@$U3303y%Qngu<Y~gc6lcD5-YyLL*m$@}v%--{_oL0L}SU1u7O+
zqCjQCz7`E-OwE&0`gRoks>(G{!WaoJoI&%V!;xN&pHQ~n@4%GV^v__J+0YrvLECIv
z<fKbv6Y8+ZrhIZXb;;<`oO?i%^3Q;a&4S$cC^%<1bB=XxU>ELB>)gN)NXm^{L@=^a
zuv6X0>%}_M^-yF2@!-XQMj;|unzcTlj{#(&LTO~guS4a@(*oPa>pU~h?4u1#dNlN#
zvi<X1Y_BuTzk*@TX6Cofjmw>MiQGUPHo0+4a&Gj<^wP4Epjr8^<YLnxcRmiz8)43~
z&K>N+{cW8)7$Qlz^YUhGn_n&M_Ur8Ot23)~aAF{?kJoZpQ)o^gkd06JI{LG6#BDB)
z&=cuXV5OUxFRT;kUMF25kx++CA~jPdQeZN7A4vO7j&kM2x4Kw1NT*MO3*XFKXq`^j
zg?rsPoiI?6(&_f$Jghq83Z;7IOkxZKS;Cg5N_Br2)}(fMP?5o6lEz4kY#M!q+Ky_a
z2G7+NI1d$DqoZ(rGy<RA9vrR(YZeU<l62WKI~bgFv-XR!{byZluV>J|fnh$$%x|4R
z|K+4hWDx4G$)N4Y8PrSLOXE(KdgaL{U91~q(dWR8Pe4zHiyiF3J#U>w7%WLyv>R%#
z+(^FB%mq*e=`>WURbW97|Idd-*iyAn8ipl2SR;U(fh4Nql?^T*<UQ;Qe+g#u`iJtv
z+aPg<c44vsDvA>guTmPS=j*$cd$^KVYnscr1LDJxJ~o|F_Mf^C0yWL@!RNs^&oCE+
zGie%5Uh3w5C~b3{w253oEjGE7>#^bGq<XGL*~@dWSCAcF0M{G_Jsq+GyKrw>X9osA
zQg*zUUw+j6?WG2+yLTp<#<)KW%fiEAeGFC*SU-_1{o;;Bz7JBNI$UHc>H}UgKPrwO
zHHK?rW2L57Dl9i=$ei$yq%)&GE8DMgvAxbQUk1ahVdf8K2C=ZV$w`;U3)Eqg7o*7>
z(^a!elkPEn%2!vr*e}SFN5N4qWR9|~#jp$at#zJYI3(rCxeVtaJ@By=yq9+++zGk5
zZ;L<dHyQx?NE=lzKa77Oi{MgnX?wmYuJ2dkEkjr|Ge`V%CR}<&{H1Jmt&7d{`2Gq^
zXa_S}IKKA_O+xC8{HpZd;iO;0JnFTH`Qw~TD94+kr<&7iUFH<T;eUZaU(F0^9f#P3
z`@lL5(QQd__(mwnhl{y_-3<}m46qrCeRgfYu+@W468m*_voha;v!zDgc)kI5H~^{f
z_BInP?|fj3F({3O^Oedj5Bin1%gY9A%mglL$Qz>O0N+O&wRB(TH|3a*xHv`+_OF4}
zKFE9%4t56S_fDHKbqfAZ+@XowyZoG!QW4>(%qGI`Oh)xd@Q*bA-IG76@89qm6*t;G
z7T<93Mv$A2f&ah8{BNC`*oB9Tb#7vmCFN!rOM7_0BZtp>jKMNkA;<F0;cBB$YHahy
z^5f&;N~G$I`D1t`UuxhbMW0`-wMK*I2Fjrukqgc0rNfg>tJM30i{16q`z9FWH_Y_m
z)FZDj_I?hekM;G8$T`$vlXK_WTwz!mTN<#dGL_XAxY!|x_HTj7_lKSiR}R>Pd&xT5
z(XmO<ei?=vJPvcR?bS5P1=iu3Rvw<c;g8I)r)f%6c2@*epTrX};mDpPRY_N<ACw(e
zx!6&U-tWL92bn1yqW26ZZ6bP6i%s-y>=eCLrj*9&k~C%M=ex{3h}iFfnYS=ATjMi!
z;hwRMSafDm#I8oOLCEG#7=r6hfy&x)53_1e)4-aLhYgJpxvwrt8an)=vRBQ;UV3DH
z4<@sXnam+FuXoZWA``XPMCRg-=90#+XsIl8m&=%gh<XCd^Kxb$>xjZG+@jVIg-%F{
zsFV3p1$@^uSMjT(>`6mFIQA?k{8$=mjd|jPxI4@`m{D$CRG2hn{DiXHLoSxnW9$1c
zt+zAdg=1@P7{64^j+L&DI_VOTg*t2^%ew?GCJ39!L3M7ZM*KyW5eISgB$(zAW*Y0b
z!Y<sJ)^UXnNQ$eAn)Ox{*Q4P{*akeH17L*@u%e0BpRiE3qgaCFHHB7<Vy#svc*Xp7
zeAZO<_Mnukl#=e@%vBAU8foI#@5-V-b+M=(Z$E_T{*W0p5^u>*ls@I8OT-)Mu!*-*
zGZ<6aDU%Fk(^&_>kZmjBAWDA(2AvK)9V$TV!hK*JrRcV#DDBVVQiNEE!1&yZ&#Zg+
zD?F4v+{6dj@hpFey+2&j%?>(CYihqJ+bngljUI15fq@^%%wrvICphU6@rF8V;_V>P
zIPg?=asy?JjV>b$BIT!GiZhuhtnm!HaO+t|3K~2qQZ5+J7YcAm+ZoJ)+i9K-Xrg<7
zW-MeP0kyBLL&jA2M`g%i7ene{^)r}dj@i;Wtgdm=CBh1I*o4(-W*BoYBZOWyA<C*Z
zxy&>O&!2-yU&2gk9iG^Qd%-$9(Q!%PdA46`z>6X*L=D#ka0VWUfZ`6C@f*Q#ZO~}?
z@H13(c$S&l2y^>rqGeEqepA+agNya_nEVAy?RCt6;g}?sUmkYSCZZCx*hJ;=+GT>1
z^&l8i^;K5<gv-2w82lxe@MFw`)-i}(xYey=5FM2igEy2$tMIf<Zm5)R<c3=Km?%D$
z&7M;jDeYv-Bpcw<a-~M21rKbqc-*wv!{#f;@~~<s`>cTA+zxLf72qwSd?gPLG5`e+
zY%NvCEAV&^yo&~}q+ww>*cbkv2~+1A1|_4|cux7@yDomvqxpZqNZ(?vaERugJ82Wq
zj9P4>`Lf>FUmCJk;*~F+cCk*76TbpS{E0cjIw!CT_nmc4pu1I0(A{pr_Zig7YTC<#
z+kzK{7A;;3w_a;_AC^C)Q*Din?<&-Wy~T^A#mP3@Y%r=r51ulmLA;|;++DmDUN<e4
zp{*PWAKm7uQ{jcSK{+Pz0?iTEPxKcT!{;WBg8!Dlf6L&%<yTL%XTtN~^^)J1xO$?$
z{P-+*jrJ726o8M%2JbNT>sySlVp6??o^OJHp;pshKG9y0+-@)~Hu5gInuqPX=%$wu
z@g`e;IrPT!S}>-#_h#y$V)b-vn0PT{`33S*10ncE_~yO99p%vL@sWGoe86}2X)m5l
zT_d9fj@0lj3?!Rg4=+<&glvMYFb<n7LS95`p|%L&T8dwf1m8mb>ZY1Vg;WuW&)Ic<
zqyYDTOVx??UgNtUVBrU3m5K7w*$}OGzL~F&`ppThrM(aQBA-~9xT8ILeAgNXJ=kJv
zqWG8gEclDPT{^+u-o_Nd8*vWixse<@xwU}M5_a`<fvd06TzxmKrAw}UfYy@A)zk15
zWNjwzB6TGBy1W~@DPOnut__tZLeA!TGI%=~?ghA(XF^UwaCA&)J`|qEHW~H>(j;UE
zeGa-p9*09{Z4g3hHN$_C)<T7l&|))}AJJN<5aL>(>W+pCp3q@9lc##K<3xLLm<3Jj
zX5NrOMN@1;iq$Z0>kTOwewCL22A2B{3HGFDN4NAazOXMh5yFn)O)G$|AmUFJ4uD8Y
z_q9bSKyJzq_Mp`+LW^-yY!{&;@rZ50!;aE!!o$joE^NXBck=4_9Is~B)AzQ;(a}V4
z6p@*^GMXC`7%z)X0lSK_DArtY^{Ln@YfW@gAcWOKv$|IkZ3-ki*F@(+SIe5{g|rq5
z<+zs4HBpgNVab!-tce`Va~(N$lF)>fFrk@iqL<NHC_;0X%YC$#RIZK^+*?T<w!Z!-
zpmeuNB5t5COzY!_ltdl#*|jA4ENKxkf<6gdElZ;RqP0*FB(&I+L{HFKs0iX(3?&hs
z&`~&(r+QQp9Udu(q7-DVhg2xV>Y-II=`{7wv_(Wc6r(y>3}LTGF*FN{p%aAZQx`*0
zK!##y5zdGeLx<uKD~8z7WEVrn@j8gzhqbMPmJoH25|_CI(l^54*Jhzbq2s}jqWaNW
zs>0_)Y>>6^IW7>w3ZFUMD}2rlBs&*AXF^xY!silN3x#D|OXtF8h*Y7I5F<OW7!+9`
z=w{*LV4m&d*h$(FT5JlRTWBp5Z8^;49$HH(S4Rt<H;_7PecjE%ha1S6PCMqaYvJ>8
z(jsI8eH6M{7Cv91wNMcxwAd6r-=?)t5yZ6^3LiY7qi`lq^{DVUI8yj%BxbI8R20Q(
zo&lI|nwsaY&@b3WIy|0<woY)e+`$f!a%T>fJA=aDsmmQHAVaxx;GweIS%^oh++jzP
zUGAI|-7PHqS(LEdJ^jwuF+|Z56`Hy9iSJ!zD1cUgX+;4PZ?DRrmDnt68FWk_gq1-H
zx>p9B5lD6}gHD64mSxZuS__42TubLN=qgf$Ej4!Wo|uDq#>ughBqp@jltC}1wNNDH
zFqby1C6%kAWzcI$9k#yiW*Ni{WKFIe^Vziw`Uq(eGJ-w`T`kL?&(T__2ohRs%Al{&
zTBr!(S`1|np3qS^lc#!A1|1QmKiJF(cNA9}eI|TsUByzY7+MOmPE!p15&9(-LvgZ`
zB@uRultc@#B<dH&PhAp80U1i7xrfP;XdWK1l87Bmc1d(>@GQM`E%aQX77C<gE`qd2
z`VH03GO(hkezb<F>{*WIvz9$e0wJvIIi!1K&zeB8bJ?>Rx>}Y!=h9jzB;#5-mpzw}
zDr~8+i)D|4dCKJ2NzxKpY|5UUv=)lA9OiN>ttFMKqh-$osl(RS-7I^!fvm~2V?NJW
zh9BjCeM)k*YGv0l_FPWC)l+xxCv8K<(!<czvH<!dt%Zsup~a>EdX&~e#S+&71yIzI
z3YoEqj?9_7*P~L1zfToBBRi;A&AcF_0xMPnc`)5HHPCONUt$fUQJyS*uzjTXIRuNJ
zBZawB7e7)!hT>=T;j;MY!y{Jwu%pQ?epYGwVq2F#ONsJFlb5*w3T@DBD1jD(Eky|w
zYN}H0C~T3n2s$DV!ipfTdqvO*fn?_*=ve4#Sp=O)YoXALYw27BT}Z01CB`lmK@R3A
zkYgu_OK7nvg6gyuintu+as#a;m8+vg&|RbsTVHpx2;v5^rqPc1?7H^z5NQ!Ig5C~Y
zElZz|(psnp5?X9ZpGRmdR0MG?hP59&p`&mnPxYwaImEp7gD(*!Ec>W1ij_Qvz=YG3
zJimZ`i6u{n;$+c-Jt9Sqhegk!!sMxo9w{J0(X;mvvgnzHN37^!N0VLjoEW^|&A+r@
z``piRqV5Tc%v}0Jzp7xUeingIMfDSHtqP#Suu;|m=-@yID}a`EuK*ehBs&*CM?+W3
z0_Zea3x#Q1OXmXUd87(kQtV;@<Y1l`kz*(6OK7nvfGV^WioP7?vYXbD%GJ>V=yp<v
zt*^UT0C59Z(`v_jb}fM3N?L@Bpf^ER%L3?wv=%CYgch3u=+m?oDuTEcLji;*bQI3y
zsU8(TheZmY2%(vmeN-gHYM=!$=QK6YkDy;-4HP9hSq5RJNEx&g%b<nA@TtonDIh}`
z^bAgjJ^K7tJYr=KJDTh==-A*RG}bFW)Qzt|TIS+M`xcI&?m5s4>K?73DtVrR=d+eP
z^8+EQ<QeE*$<rT5b}o6I3tcTsp4GG#3dOjV&Lz(VQiUxIcCqAfFwf=W*h#_?T5L+5
z5n2mHSPpY((pplvI$HAFMC!2hbvH{MZXjzS?U>K5CC?j3i;xlYI_PRy@;ppyp(03V
zu_<{zMr)xWh-)#FJa|G!;Y^<DQOOg%$)y;~T<xeJid8#vV7h6lohP7QVzr~un=Ev&
zeWcJCz(Qx9Fn8)gM+(SL=sbmUVTH~g@rV^V>}awJol|fJ^vFldY}b6K4T0ifGnYQ5
zub3HXpt*~K8pzZ`6+`=D!>q;7et{5H46TY4Lu#+Q{>aWA=6!&U3<NsYK!-zD%Npod
zS_=hhT#KOwihk*B9jU{X7-K~cSCeXQ7(R4sxb5uQOX@Ayd5G=4U_&NXkrp9|@pA6l
zaA{@U8>UQap@>mvv6;(GS_>6HTniE-v!BXHJfUY~CQtQ<7^8c`7=X^qkSdyD3^@zt
zorWP_2mKNm(gc2zDzRgPDpw&@_6Y~1rb;Ow166(iC&j4pgLuTKk{wMpRh~G8yHMwX
zFB;m?Bz0>fEHX1iMn7d}pvNzwivh79{i3Z^B7RxKfplYdpGM2Q@;)C3VbpkXj2eTl
zA=)wHe+TlMnep4u)sh*1L2IFKjcYM5W8|Y?e<oF=CPt<u6)*CETcd49$Ie)Wd-}ON
zA$ib|VdM`79LbCepey*+h8dUAS}0}|T5Oo{1X>FfL0k)@rnI0+Nj#N`8%^Ho5jRHH
z$s_1y=0!EW7%zSn5+w~UUJkZO<i#lJNlwI05l%cAIq`EMTT^qQ6d=JgoM%@1rPXjk
zj1$N4h;br2nru!SV7Ef7*T}t2PYGmZK<$gCk&R++1Ir2Gi+u?74m_9jQt+jL5XN`w
zVtki+Z`juclAW3FAE2ux^SzzcLg5zI(wX`GiBvHO%;#X9FOXv=DN1MwQ<Qlhv~SW{
zD2j5J%a3R+sazdh7ymt}!`9c`To>mC3d65Hj>tY}9rM|BUHpKfD3l>1XbyC>EOI<r
z3l%{^i%pR;NNb@Yh-)#di{l9$g)@1oM<vbz)1v?hPeG~xid8l@Kysw1Y%T&5B~~_q
z)?`70Z6gKEIxJ{z5{a3*ppgPH6f`BA2P<gGc*F`Cb~M=q&B^fPvPKgw^2cgTKZmb-
zSr<10RB;oPnz_)~KQ>UY5E<GC?MAS!D0&PX)YZk!*e+`U^pZdbD}Xk{3Lw|D@QFa6
zbKUa_=xSN_youIAK^fO#sC%Mo;qN7N=p^K{4_bz`a6WWvi0$ax>1XFF4VT66M@f^A
zZ1@H4+;BZ*UJU;Zt%YJkp~Z#`e@1JeLWpYtHWUU_If-ZVoXq5_9?_u}rNbD`nJ@ZP
zIK`-OH%vPXHO_f1L5&9RlQfC_A~d-HY4SQ@{?s%n1!SPfB{(NWlS}c4(Ih*XY?@rv
ztl=jb3r>53Qtw7;@-kE7J|SW8z0uc#Ed?otnyS=09b06j!&3txj1Hd{qr=qe0T%|6
zoq2H+bhYHg7tvZMP~%z*yck&tuaGKI(_;FSa0m0;K#rZ{FQFyOU*`3IyJ#&Ge>u$M
zb+ndLu8yt;ypz;n>+5c=2XF&f(`v_jc3ls6gtQ15L7#=LkjCK%GFL!fr?pTKB(&I6
zKtG_hP!Ytn7}f*ugpR_QJk_HDDEe-sPCe$*M}<+W^jQrPPE-2KT0)dQA&Qel5B7)@
zJ<r3U=VW2>)J2aJkfG=~0%yaD9uJRL(Zh}=yXaYo&w3Q#12NW10!I-`0#aD!a%Yy3
z6t8to0uzc_N3~Oh&MA00YoT*uAcPe<7sd)5*X#c01p=Mxo3o*-Wqorwt%ZUtuEkK_
zM6dftNFBDM*Ue|(_|UDPw4-mGulsKyO+vEX4cxgltalHsg<?IS#fJ6XKx?5wh-)!j
z_v0BoCo}n~N3{2x$aQ}JL1xZVQ5566V_~*wIPbI2FL5PYL7gN#Y#t%J3z6_v31g=w
zJSiXp;XRI1VTAV`JYs~$j-C~ShhKW>8sVX^?hqcA6eqmjp+oueZ7d2{J4JYZ5dN0#
z3O{IM`DEK~0wIj>F3U`KGnR%^K0GS&<F9)`S4+ZMKx?7UifgeTydy~+lR<cV=++?G
z(YMZocLr$^lJHLB&b1-DEwmPj@Prl{!n=yrLWL05Vj?^|lZx<6zS@I?$DqYbcxq%Z
z!uth)M;gMr0jxw59vhD&JZv5zyvvaA{#RgEYQmEOB!Y%B%j$8qSL0L|;oXNvjPTge
zWE0-8SjSw0cC=<Z>b37cT4u_dt%-_r-h04`g7dV7iuB%x=d)hry*m)XNbjo5r1$AS
zpz~GU$Dykw>HR0Ig+eT@#e($yo76EGq{oLIPECCrkw=|6`qr8Bo+eE~lHMP=b8Sek
zZy6QXWZfjR*pS{qv=%CaxE2%X;Tb(AGx@4Vq!)b&Sb>n4^;8(eSnusH-88Is8rVLO
z^)#@Pw1@2@w09NK-aCc4Q`4Rlkb(9t#<?)sy9AFI?XjcDroCf^O3j8f;~haPYNDXb
zWH*gViSyhxFq_~x-bPW}7@o>Xaiu^Aqqw0M#qC|HHpOFBVR$mX%lqO$o-@PkhOU+j
z*QT{l@Wizk7>-R;S@*T13Y}P-9$#WwQoXUn2X2j-9UVLM1c%pnKSElA<hBoTw}ykq
zeBbVKv=-`uPH3^=wy)7zsNmsRfZM=$Dpc@SJrOhcs7JguKeBqs@WxDODsp0!R)CqN
zp|n3jzeGyoxF*>Qn?~4d2-$2z7&$eYNdXzyY;M0~vw3*L*o+-bHk%!HO)Fn*=A0f)
zqCQ)!#ARl-K3!Iv-<E+P1;6PnRdOxI23c8dNg#x=Trtjajb`0}<<<o9oLO!)bhTu;
zb7?IUP;o6LmcxlE>t05xNY8SpB^Ar@fm<VJN5@WIRvkIP<+a=<X%Uj?#<^>4nC>Q8
z3&nIoiw)Dgg4RL>5Z8iNaom6^WbllhkePhdBesiv!B)c{Gv}!|igBI?^G(BfAB291
zoTp=*WIpT=VZI_V-;u)LshLj-$iRGG#L2MhxG&)mV?K5?+01u*q1GCzzzQZ3p!N0K
zGU9qp7nqs-_6<vm^WT%`T=*F>E!<Wm+z*BSrTfBn$}I2S{ZAl-F<>>ufUXa${9ho@
znF4<cT`eha+VU{igy&A8G1p?Cz~}=j2aq~ab0AlfiUj%4t%0?(Zyh&nI+nBuY4ri<
z3cj`Bz;(11iUWlf8xGt=YoR!hYk|y+ZrX$=Qc<AEQ$3=<MUe}>5D=M3P>n7|g5QCZ
zO+$j?V5vkB3<I8|LF^Kt!79?=e+$4#O@mT^#L;kmS>46E4X49s@OC_6G{}x7n+8|l
z4ZK|4ngACQ1jt2YroI_cPMq{!55^OuCtE4Hdn2CAdcpTVAcWCfGe&og%bM>G<T<n5
z!_d`|?LJ9sp|FZ;F|b`^S@Tg+MQXB3zpTjz9?nI57?Im{9UVLMWQS$VUy#-yiS8%d
ztu{pWXIcwIbV7>_(ak!V${UNh;Iih)+S&l>))O(4kLJ#vGG+OV6W34l7pE6zu;f1i
z|5yyA?@{pI68LW!{I~q-3AUcz%-2W#=0tlzZmd>lRs6<Ej#w+<_r^qfdfgu>6(-sT
zMp#Z@j+x(7?8Nx(KA3A7ej5OzC-R%bHpy(*IKpgAWVY7|L#JjoDIf#0orM!&%(elK
z7_+gX$!50Wm`n4O@uK|$D%8@Z5|^3b`gB=wmb(%RDVH|2mWt`F#s*oL?utMNW4hg0
znXVehb7s0SbhTu<>u4<$({U}GneJtziWuM&YPf;lM0xkYQ>L_M0=(d2`b0VQ9zuGi
z;{&%w(T<MY>-bd;RQGn$93(xwh5OZp>i&t=LQ$R2VncPGrL|De!?gg_q4iWM;?H_2
zX7W;xxGuV=sbPwl?^FcE`0f&zZyLV)3G_?kI~@-t=V6Bk=j}$$yG$57HRnkI89486
zI2p!yQ;w0G$BrhO^VTre@)AhaKh~FagPktz!g4dS;C#bi1ulj5<#`U+SkR%di%Q0W
zv1L|XJTMT#c=3ifFS_0GJ35f(%!^B*t0ga<OlzT#jcYOSV)&Nd*`$i}yqNlyA0N0i
z;&ybb>pSTM(jX)=Zsne}Va7VGg<?jb#fBMgptVr(!?oaQ=ayR4*SHZ+=;@frQ$6Cw
zBTd(V5uh`Zqzb1PNzQ|bry<F=K)*ziG-97*OY9k8%Nvj_4-hU$&6ZL?2DbbcoE2ls
zkKqx!LS#pi&6dYA0xHyN<F!_IZUO27GxOuVVQKN}#IK@r;b%y_a9hQQj|u-v_l0kR
zS>6KtN+5(W;;k`8oIX;i<y{{3{81p$nG?ScT`f8B_p}xY)3_D`CyJ>m3-2`$2Dz|>
zQ?sJblj<%YAG|f%c69IF#||FnaKY%279nZzQ0NNIwV}m9S_?&sLW>P8o<?h-LWpaD
z++^lc0fWD%Vnvgmdc=y+dw?M%GSi_NU5pOD1bLW-4!44(66r9Ec#;RPON0k+MIL-q
zKtXCAlmaA>hO^7+9^ej~4&%X{c*J;+9Zfb5p4!N7_jCEt(YimHZ~BECy4(7)QLL{I
z;`;hvq?$4b&diGo%u?h0csH0?@T0kpO3GJa)2vsHcLqWjPqt$`>H4V5n*xE(H2ENO
zwWP`S(po5t<5~<f8GTgd<D`z%EScd^89wxI8tdbT+&=8+Tj#g_zC)UXq{(k`=i1QZ
z&uA?aO$seGH2Ftb3l&0K3(%x6pvp-+qvvEMU-gJ6kFo(gGhwQTiV^1RF!MBodFTp)
zFwGDsi4(g<h_j8v`7+^z)Wj(TWFXGtaaxQxPrxHaoa|__iE}N!<XO*;<!a;2(pYIX
zqt1N6nmUgoR$R@ZGn41SB)M_=JRgiL=rgI8O4iNTI4gmk7YJbldQWBoeNiCLnLu9%
zT`dW;LTjPWk880Y(A}gCow%IdD$GEjeCXDo+}XE|&zjyxT7=}!3GQ1P{(LK~h2l@4
z#fCpWNNb@Yh-<;sS7twzk$6JS$V{H<5q&O+&}RZTnF&-yRE$73z}(Xi=r^HX;uU8S
z2ucdY&JhZ|2PyPy;fT}}Dg|Vq&|l%i7=`{Cj~IorqsgYwHR$q@O1@Fd<(o~v+AP(o
z*8EA`u{6ldOrG<NLgUOib1+~|V;7aAv$17X+MFH;VYGQ)X4*U?5a>*s3!$qeZT8bz
zDA40tENF8zsbex|lMme*j63?)`MUEG(j+8rUdWwm!<$327K%5878~9ir?pTa#I=~N
zJMoO3lbL+gBjSwSwlo2gnKe~F#aMF+Ogs&1PC&mz)-(d3q)hA?q0IY`GWQZLNKKhi
zKnBWu7tV@N=0kYID3cvcHf5d#U$d$X!$NJYS;L2X;5*x6`JK6nUma~0t-13g;@Xo^
zkeNgerKQK&^fMR?@H6Bp-B;1-zlr#f?hoITv%Ky3sXz#$)dw@v>emB-&b0bp(AAPw
ze?V)YFp+Dqpw-`!Iwpfw`OvMQy0dTD)f9$Zf7ynK=B*5qMmXX~R^1o6f^ThD^>A7X
z#i~M!&0Gd(EmQ<?ErzR6Jdui2O`hrzsU8!#8YO|s%&uy*F?RhJ<V+fNy$~#y$gVUX
zN}5Fx5t@AvY4+nHjZ@RC6d-XnoPSn#MMrUZjAn~?#AucsO*YN0F4S`HXpo<44CgDQ
z>Zmod9!)T7L}q3(of{Jv=g*gbT?K!}nk&k@5nE-w61_eU!YK1CnJM$0K%g^a-UVGP
zDf11q77FmV77NOJnA9;Dl*xx4PFZ~%kvpIred~NB`X$mNBxgRtoomCH-=?)toGG-}
zaOO{FEmR0`Ev73`Jfr7iCSUc4G!KuEW)zUjJgI^y#*;f?(rI{d-(v}$i~*mdM(h=#
z#<w6f?iQv`O^s4O25LMCXT+%Sxp>5=ksVDoHJ-qDal8cgK4HDqni%^DVhl;lOpE(P
z#KlSRRIsQZ#YkgCiEFV<R!Tf25W*<&oiR#u+)!XkAkUc*p9ftn8SyGw3k7Ffi-8d%
z8w!+36*>_)-O)_{nk65&HO6*y>{OSBW^X60L2}_O+^sfT_z$!eiVKAn8!mhct%V96
zt_8RdjHeP2f7KH)laG4DgNH_VFpNNE7E~b=W5E+(%4t~e5$KnAwHQG<$${7>!h!EZ
z4qPqFo|*%tfD9b?J)95Yz$fsCaUeUIYz{n;Ik?)~=179eJCR@@GBXR#2@lcqZsk+x
zV!pN-ZLQMpPa+Pa8^dc`maD0M2!t>md~a4B>^m-u3E@GJpEH;NT`hU=AX*CrWn4>V
z9$ZSQm;@f=1GmQ5j*fNW!40G}NFF?cyVZsVFQT<jJSeo-@ZeTj3l%(EOOyxk*Hk=c
z@==d?@Guh(DohKA@ZKp?rcS~C6@QQ1GY7W$0m@MWjWOb{0YK6);w@mbL`IB)o}@(V
z6rsfTA|?J-pjc{3lmaA{hSSVy#q@rh5TnG`;Sr-mb~M?P$iDdtUwt^)`iijAH-AxB
zW*TIlHsq4x*MT1Z6AA)U?G*KW2v29d1bkm0gi+r=#i-Bqy^zlZ0-dh`KMh?ix$kSV
z77D(&76bQ1-wXLZsUtP{Wq7QW4?Ucg`Zyx1q8)wfyec|n6%}$M13t~2Yr}wZXe|^2
z3N1Dacqpxf3L&n=v?_{c^qkD(s~)kS>6^a{f&%^v(VmK;80|d-vrR*LXMpJwX-`3&
zBs^>$A-sP=!h2X4J2l}+0T~GI1vnK(c$ec5BRqCA*@Ooh2;>X7QZ;u1d=<r-^r$_=
zb%B|Q4_=W~!!up|)xeyB{K9P&0bYYmvJ&7}AcPU%$6^FH=K^?vrdG{Yt{ke>D)6We
zBkC}YnLitKQy|)z3SSIeEvfJov=$1-xE2d4e2~<kQ;^eoXa*{DFwpzS!IPvVw1k~~
z9SZ@x{<%(D2mB<hg(5A7xjag1N$2dy);QlGb=W%lqqybZY}|5CY<kLf;7PS<1UHa1
zmv&5P{yGXzWSdgq&!k1j2>L(h3W*$!ptV5+t<@rE*6|dak*h{Ri_KgX(psnp;#wf^
zqp$Pe2_1+td8$WMPkh@5jepFQkBXyM<x_+ir>T6_fb|nApD@qK>Ib_-s-KTx^-~r`
zPhI^;0U4^F=i_u(^>Yy(vFe8%O?LHj(pcJOkB=eNz@tJlS3&dQ664DNB`~e1gyQX0
zHM9+zWvzxr10k#$`gE)s+ApYv;F7_m7`iTy>|6}(gszsw(938o6uNOOhGHmkZ}>h^
zg)KdHaSL7t^E^b3og^}$B}`=IQs|?!7OE6-n9C!ymQ=2e?h^QQQirXtyIBfx16dPn
z$9#6(SMv9yMaT&HHFUKsh4wmu!ZNZH5?X9Zp?S0xDuTEcEQKP)GoH{<IFqM(R0<st
zxeXE{A#*vTVkuS*4Z*C_ltU+g2NKJnIN8aP2)ji}qEBN<R1n5bT@pzF8A_t_aZ0Qt
z+Jr}}Bw|OCT@tMpuMEIiMy^pR_||Je)HCA|nVAcsxiNw9dT1EzD(azFb5#!c*eYu|
zln;cka_Es*IkdOie#_efdCo;p9lBZ;LATIaC|u)OOhr(5zvX*K6}H3}FM?9<x6B7_
zjkX;f>&lDoB@III;zQiC;mXRqMDTH13sqSOEjGOPC0Yv=KU|A}7x9Fij+s2wBVLR?
zE*`}qGcBr6iqYa#FzGb3_-p8wNQ*JdldOonBCPlbvf_(`=~J_!6p(=xXPzioaTXpi
zR%Az$&5Fm5=L-dRgUXE(soNyFz|4HOZ&+HK437YF3Nj40RZO@Tn`C9e!vZ0U2_MbM
zgsTF1&P+H6T`igL3|b3?WL!&UCfq`*m;@%|1GmQ4j*gwVtX>)|HeEh9ULkEll3|fM
z*M<yt(^@Dp6k2S^@OD}Y6+&DKkYQj#m6Ui+Ps&XG>JcACm&`+0WF|uuO))Y&2xgpy
z3?G7iiDVeYJjsOECBlS{A`>1WjGmearGN}f_!*oIyFUCZ9x*0lN0ZHjtNdCcCtPfO
zZMc+JE!X5_=D~eJ!s0aeIC>U-hGYvhRSEZ9;d|-2@H5$#4~l;)5W-mSYcUo~|FFRS
z2BMt_@n_K0k`SM!wNPlrwHOF7`rKUK>M*p0EuESXGdwKdV4%av!IRV_w1lb4yh1)e
zYoVyiVJ>TEE$N&cS?@fT)M4xFZmy7X16lKF$CP${NOFWU2^m5|&=oQ{975)$fF`Yl
z3L&Azruw;w)<T64*8+en45*+O&!nn-Oup(-?Q=+Eja=s+bLpc77b|_f2B@2+^m!{7
zDzWql@tmxDut%iw`5IO}Ul+Kay7G|%Bz=bS%j(^Lf5q9b%ID*F#3~<lG})EUn$l>s
zR`+v5rF<hd)EXJ_>kaD)h}!4VAUAU*G~XyRUJ*S8HWn3;v5QK_Z(z%;>jPg6gs{r!
z+g+@TeiDdwu8e*FT`eo4|D&}~NXNBUR7TTIqAYDoma$dy43&|Cfes=EPtuyu5~ekC
zWweymLeZMTTuz|1q;qzpGCGsgVe9N}Rz}=FVdU1w5qW*2V@f+$Mq5dfkRfyxbhWIE
zs<aj=goGBG%IG>;3l&0Ki>WfgGddJ!@>P#2qv-aoCO$G(NGhOW71E=CzG*6?H-Nzs
zD<mWD$w~=(Mk=LmW2N+;f(KGpN>V_EO6kKmD^@9e1dmvy#EvGrQaV1D8?QB*@LEZ!
znalMrC#$3}zgesmmQS>)y>&tYGZ#m5L*n|E8(u7F?-yz#pTv&*s;Z-}fCu!H46kNb
zt{Z(h5W=dXC!jhiPH7)d@EgOh<OW}9@tc0VQff3OUNy1&#)<1E`is-sdx`NEkHEhd
z!{XLa@ZS>nZyEfz9RAx~ycR?FdnkQT*`TD{BMgXcC=xcQIs4V{V%h%)Odw76KLz~~
zvp?cdk@xNBP2SMZ!Qv52cmgru3&MS=F+mC-F<}Nzm<eC^ad{oQd|sY=vIK;Ac*FpK
z9d!c`l1T<_aSVU?{fGtOGzfPR*MHN8;bRF6(*zC+CUAH#0d|yj0-u-#<2j5+QgKk`
zOmPOg<Vi}H_DuMQL%rlTCUEQA$3aB9|FTjee5vpCvsrkhs;4b(1B@M}{W%!x7Qrqy
zKJB8a`~+wV53|4gL+!Wr{sxikvTjzN>e*g(<!MJX{hiG<-ZD1KW$OmK(-@94N^lc<
zJU{HO@q#AavUTeKCr}KQ8{9n#+fD*s=-6hpoABw8PGLYKi#XKR8VZMw9_$j(X2Ogi
z?T!yLG1Pn#uPJ}7n`~d*BNEBOKdQ-I97r==M)sGV0?Fk+!at$>lSBF$oJbhbGun{K
zC!!XxH<;VtP&Jx-htbT!w4V+C7)SGT@@RU4_-tNgTvuL-;rbv<(i+!`Gs??w(r7^e
z>g@Q|iSp6%_Xs#YiQs&!u%r>3`-`W`q-!5>X1?J!)_7~icde-nmDg+y=l#|oPiBgS
z+PhgVjkrr)?6(gh<rTjp^_TBP89X%RgjH*bUKvpo&aVFAkK22V@0w`O7~i!5<`9Nj
zuuk!Du*jVtqcs%nsi&Fd$I$DV_d~z-OnhZ;*SJ5?-WwIL_x1dp<CW5IsX1{6JG~=c
zuTI=i{B!Xy#ix0|KhuuR2*>hR@i3OSKSVd71*zP}4(s#|HwjBXsib^(&jx>_RQ0{Q
z$3Z%5N3Yeu&5)Z#-z(&sdETVp55p&8o3;92-z9K@|2SN$HuI%w!)y54{d&IQ;aBqT
zC0{Qn#~L6f-}Hv_Rd2`#$$r6W);#DqQW|a5aR_7hj?tjRooD0F6yAJ5SW|Lla&+j9
z_$hi$Iih?sQ~@%cIoI}=Z`XcnKg4X3t3pMatJR^lg|9xW@!+Y>zC8Ym-l_ZJb-#gM
zQxdLXjcRodJ9)JcFf24^*1<h~p^wS)hKv62w(%Oga^#Ib%N;&Gk<tK)E;UQA0rqa(
z>a#lNoj2m)BFupFjklvzsi60SG;gF{8}k~&5Dvq|!9Jew<-7C=91x#?bg=1^GVmiV
z1}@);<rL43U!aXY%d(^Vpgxq~*<#=lcC3_r-AUO4sEq4CodA>%YkEY@q~+MN&sEX|
z_dE-Sro8fF7q0|h_e=Eu519YSq%6N%#SH7iV<x$pf<!Lgruwir`xK9}P|n`TLmojN
za<xOUvR06!JTCzd0T{i~2+uIgXiTv_nPQlc)mjr0y0#r&IVfOes^k_zWRfDpnv;^F
zK#tLpU{L)h<$>v^K%{F`H*HWY34e`Fm<qiiIc$?~zLPSMgs8(N33n$)daqp|lWj7G
zpj`S~7b^!z^&9l(V(3cML6{&~Qn5Zf2CS2c#SJDEMjgaXD6w|)Vj)+95~B{`-{_oP
z1%D*pszBAj$_%JZnn;sSiqt$QrEf>kuc~|#Wrzi}&5I64Iy!zrng2`|^Xs(sd-T9M
z<^k)By3k3P$SBldlTrEPjOvorr9Jn6Cgq^Ki_L=k_yc<9Ma(-iMr!%N`f!h0=Ld^|
zr2M!=1SBi`I#r3hI;=x|4kaZJ5El6wm579C*7|@(29S{oCA^UjbtX>@Y#*=f%zU$t
zHZbYb&~M88FL5!yo*#ci4_wPUV4WX#Iw=$RfjVsR<C^6B=#lNEX(vIma^Qn5HVyLU
zPw2h-nfI*ohxOqex6U6HB}w`7@@8$DUxiP**VzSEXLjk}$3R>kuj{g=(VRda8=rJ{
z^k?OZkGc3lPo}5Q9Uo!tuui5gI4Kj!ggR`JshK*N0-L!zK^k~+lq)wr?qb;>rJg|_
zev|poI;B`2?sw~yVzH8xQnwH1VXYxoDAhY>6JsLC6t>({s{6yR#<a_W>I@c#G-hI?
z)95SIc2sfCJ6J|ys}RN3=qQwSBd}mUI9v->DjFap=`v_`FgWRH?H6VKKe?D+&!Q=N
zfd~G;JYbzgGuQCi*J@p<EV2l7*ksZ6<Sgo?@uhVqOTBXCK`zz}GHELM@j&S6aPh<X
zaNk>J5{s9lOxg`~SZ*W_4+{i{1}TMGD8f=8{+|yEtEFn81Ye8Abpyy7NTfPm>EMz=
z-ow7|mtZ)re<(k^4U%VQ7bYBlqd3v<Dy5-%zP@X@hiiznrn#g$AU+)FX45I<g4HfA
z(6eb8dSMmwLO7fF<9cEpBl_~h1}AkQpHPcUKIM9Bd1=v}>rw8x)Wu#whD=A_T*Q21
z%{8nK_osD+uoy_nkQejIk-ERV)PUvd&V<t#_lHX(rD3sd2CE9JpU9Ygai=2R2kB59
zF0%FX0k4@K6-SUB!?m$7*bTN+SZ>aeIpHBme@1^+=5M%|U(b$N=z$vZKsY;yRqq>}
zl!@Fx9X7c!n#?s_wY#+G9@D3sb&revf?S!6p1PZP$~sqAAMRc2TwxKBlq=^lyoVGC
zwx5H$Nmuu6@rV6J1Hd2Yqw3{{@lRwETxKq9hwqT{Tv>^?17R`E9QD(gaOoKFmonRX
zUCgG(_`YblhnVHUF|Ibz!&ykhkzdsSKJGLC5%;LqChm`OwxJwvik@mt|KT#HAQtyS
zi++h&)H)VfAMOR~SY)0{ip4iVi9TG+73^-9@MeI~Sp2hV1%|O+t6K2u?3QJ|1!qf*
zzHxYV3U86H>j@@Y-WkCbV^Ai+$9^ljJm^>2E-xdnF%!6~A#aPCgM1%t)Y5~Y-;{5D
z>Eatb;OC-ye#+bv4tN68|Kg-f1UKri3GO?SL46YZBF%pH<d5q6H@rs0i}v`*zNdm=
zZEv6jX*m!5J{!6^TyU^HJXWmJlEqh2T9&cGhev61_$bF1tcn$KEae<7Itr!6Hg7CH
zJ}xdus@|ACh8Ojv23}e8`PEu$G~vO4g=Y2A=Sink)*bC)dOhprqxF|E>xZ+BywEt=
zNu9_$)MAr&=i6LpSXx_Ju&Xka+0S#aLlE%`(B@||n_CkR>%;wI9r4V!NfCb;h8%vm
z!}h|O<pb+*O)C!%;PB^W*dsNiD!W+%i&4hMV$?@D#C5CsL78&Y#guvkKL@Qj%&h4U
z!7V3sB7#wiO$2Z36v0-ul(E|-Y0A{My39R@+Jn%}H#0k1M=k5aePbQ9%%4e7yBaMQ
ze3(R_v$p!f3>%a+_!u^R!NeGq;SJM8WQKoKCVPvE$@J(v1g-W4W;KWCe7}=A5uK>T
zCOQ{)w3m!Li<ZhnpK%#;5LJuNKA&Xvv5qR%huhRTs+b>=qUvP6axs?QnXCBKQTEs&
zART+;6Mig>wZ=SgLfkxN9n5GK-+5NVR{VrA-v785Pmiy|(0<=#_6x^X@bztc01zgl
z)oH(SQYNAcb=X9gcL`op5Jr>1)w!WE&y>?ZxpQTCINIhJW*h6+Vtu$htz(P%ASt#k
zifjbw@n;p#N^B!tSi9R%EWwJKLaj!z)~Xb|VtzY5eA=qwA~!UH(y~%Yy1f(I*ii&s
z?003<!(EK3$DD_DJp_6?JnuH(q)fyd>adBqQ!`johGQlf%BX8y<{6;iVzlTQW>M=%
zWqr68tRt0qE-6y`^SCl07AEki>&^Ixx`)4(ns9>>pIgVX{Gs*!a_G_L*+FM%Q0*6G
zo-19<qsQD)XrUJ{3q^1aMk-ZC9&|1_DHAb=I&5O@AksPjy2%ZcId;2@Fo>4tqAhkX
zTUbX6>%;A59WBh}NzrnFzM&rpYQfz$PX{)+S)4>Q=EF!(g*WJvC7P~SW3~HSY^jIV
z5;W%oGpBWEz12yX2rblM6I!R4;mpB~5PsQ&s1W_I%S?k1U5YmSAhW4;h_XK157r^d
ze3ulWXKVW|L4>l~$6D<RaRg6JK(PnS_`PDdUT8FZ_!+7^Jj*O@guBrGNg^;qzbOMg
z=3+oSE|;MVzrt)7j!SZB=1C`YA~I2nO=KRQ)_f3<sro82{@!I?K^*p@4S&OIXdQ>F
z54XE@95P=e#o-NxT`puij+Az?C6o;tE`)``Mhl+vW-+;GvxhrMtsKk4s-oOhY6S%7
zc6eo}0IwnCD|vX{0qA&OYpFV3fhU3BH8gk|jlIm&7yh3KTaEi$h)6b`Q?58*EkuiL
zHLgeWa`eU==p8K_$%yux)QN~jEjAH-S#KO5E0SJ`SI#)j#X3P=9F3kBWS+3j3)Y8w
z&pI!d$5md?{cei9Z=~KGEN+u84z?F9UJSQqYk04gKhRTcjg9Xr)P}vqi>1}cHr#YD
z{pZ8<8^oKg;_dBX8QRLBU`Q1Y-+bC$@M39(jRRmAYex=VWURI7khh^NPGT?ed-*23
z`#98U`ml{a=yNKurh~fJ$esdf2K-Iy&U*^LT9_K7fRNDZ>U9h~{!VhZt%}Yj8J(6J
zpo|AXnDN(w@ynro4bpF%`_e$P>AimPTTwScR~U%Rc2BROwNTqVaV_eNaq+?>d^-xu
z8Lj+%BdH_w1~-0KwE*{oOVx?;?t}H$w#!+!hr{>p@g3mDAZHv5^dWNa<hBq(i_P|M
zpP{u-dwV&|<twz7bk2@!5BEJ%hpn?eioVbAmEE|rlkr_{g&}t5=$O*Z+rvFWnuH9Y
zKS5W><ZuYBm1Hti|IF?_>!{L)#0#Ot1}_e#wNN3%wLrNY+c+1`=v7-LU-f3AkoMsn
zC{5dZ0^h%!16z9l(5PUFZ6dM_CT+cm2;V(~?IW@lJRmoRGHk`t4)dMd77lyixPE{;
ztnf-<Ef!j1!t@Yi12AM+%XHXq2qlRPl%xRppnTYORvYMEf-_><!d;3-Y!@hYly(y(
zR#0?d6Q$L|EqHHXEXQuCbg>rF#b&RC!qRY0q;jYhs(?{tEfi|2YN0ANNxW>btcA*f
z5LOGF6|05RrBr_e(bfcrwnnQPUJ?j&u7IwEu9g+howODT(YO|i3g~`Phb<{~vjXBn
z4=1ZWjtD7s^sO@~{xfM3k`zD0oohpi|4wV6>MEhdh7=#8wNN3%wU|f|&*(Xs$ya-h
z6hjDOCPfuYF;etl(rHNXPtcDf#n7-MDPpe(DV~L-I4Vq^niQpg45T>kbV-W);}Ii8
zb~M?fxLLgen5*Qgqpkd?pMzzOGTeuQd%=8fB!_i__4;Oa@gmDYTV@J9%1TL`S^L2=
zf?2IZD8fA&duAovrGXGexaY+PH+VeRj&Dy5<T>;0Nzm1jZ_lH(P+-Zm82C1Fcl8CN
z3Z3Rc`^0rxa9~>4#&Pn^8Q3^(Htf0w^<k^xPZa;so&|sL-6&bb4&cJp8~DJj5xk>g
zUBBYmA`L>4?KRx9He`D<t%V|6p~Z%5UrB4B;)iPivW<S-98c)!n8{N;;@T6_VWXLX
zRfNSTcrn-_4F!J)`cV{Yg^-e&QCNhT&qHQDN;oPtGfM#(nE6XMQFgidWjtcc%#J3T
znb(RUXFGfE(*3ISIGTwK(layfe524f)BXVc&e!veT@<taNJNNqaQGgQ<)!TR10jrA
zFNiU#>!s{dfk0<!{T+0*q}J(YgkdZ^cM|5g76Y|L7xU+nI#P2hSCfid`OvK)yQ6QN
zFJ+G-O+uP|1#|`H+OX^Cv=)k8g%%rj-Arqt*p+KBUCQE_RP<`{RgdU(Nt|9y*kq<t
zHO?5Feiza=4V_*CHlyg&2!oP7v2%nzFF^YIo`9j$^eF{MbPZ>n)oSnUI59?_ci<7D
zPj)of^tqulhL^H_br>El<y%h$d;L0G&z?-PXtI*bWO@iCJI<+Z0P_n@rFtrAeG_)b
zdeQn|AcRrtr7>!aye^I-w7Y!$Kp@$fW8VW^Ejjj6v=#~=xfTP*MlN6fgH(~4V$)x~
zI+*84a_l4*3N2wSG=EL)H?$Ut3mxY2H(E<7S4Xcr_ghcpl&!D3dApn&C=BfSI3jlj
zI_9(MTE#J>MaT$R23;+ysZ(ezR0IhvHr3QQv=%CYxE8}&1)k7RIFqM(R7?%Tiz$Me
z%r%t?tXNI;8c57EHPtpS9aU3NG$>0elo2VdF2&O7egPJ#ODicLLuqvr&X1K=H{%g2
zt=Q3ImsabpfzKO^Whkmv(M6R>eCBd$fmv$2ka{f`TNF~}KB|Oz9X8EcLfsb#VI|a+
z-7BHq9Y}UAq22*qEla3>p|wz`$hCAXp*~Nl(CKUs-2reg&v(hOlLRQV*pyH|r?pT7
z=rET*(OOcuI$A=_Jd?^PTVHpxgyIIWCgzU$>{>!CAuU2i&|>InSwbC8YoQ`YXt61w
z*3(+32;y1{B@~{}Q8<&QdQ?I!jh9eHJ~9_iDx_k?)2kpQ(-coW7>z2P%+x2#ChQ$4
zo36yN>0SX4smmrQAVb-79nOrEP1oZQE1THSWS326R4U@(!|<1<99E-FpbICng3P7U
z!X&xz;%Nd*E{dn5UaEZh2W*_Re7YwP!pf(7_sXZY1(Kc1r#C}a%kt?%v=#~xxt7l5
z(`QH(wv-xsUNQX&l!JM`L5`gyKcU5@eEK1+g(5$Px%`gSlFHT5@@eW>R8HCYx|`(_
zH;^?ccg$zk^63cDB4h+Dg07b3(+XM(6+uFaP5E>xt%ZspuEkJ3;RzjuGkK~<<x~9o
z$Abx+WG<joP{j(UJ0U646i_)Z8&yCh(Vr}#uz#e4%3}$2w*ZOMC6pA9p@iCwb7Ljc
z4m@Hd6g!&i5^4?ZB?wQaj*T~S>@yQCuW=00#Zy#t=CWyCTw=Unx(f^}3Z{5_RVuv#
zn`JGP+JO*ODvicUCD-R*-WUjUu8<ypu9g+jduS~b=y5HE3Mu*=%)gR4Y-uxAB4v0E
zh7a8uj63?)`8k-!Nt2Mg`3>&eaOGuw&+(_U7K%5878~Atiq=Af5Z7XQ4hGNYIhn~<
zJtEE{<LeGFR5DYh3aA)m-Ubs-Lz#;<(3BYmKuMO^GeVZ5NS1d97o;XjDIfz`uEJR{
zvOFG-7+JET$tKHFb)s~Bkz*yzmND6x>2iOA$T(j<4{R*>($GN><|b^Hl`zi@gfPM!
zixFn(r(3QLBs(+b70}g^Ik(YTDA?m#49ppMx@8xsA~kWQUt4f6&z<DhNrn?z!VG7A
zy5)Xa3&n5_b9o1?C6%kAYYQJHb=dm4n`;Z)K-R?DF`r%679JxlLPpS6pev+tID*VI
z(v!3nDuRR-n;Pjiv=%CYxE90O0-n%OIFqM(R3aS}FOlL*WUi4^JjH6HvtZ_FYNP`<
z(lwHS^kk8QT_Z)(7#2wzg%eU2Nm4+DB544p#fqdAc*Ke%b~M>V(n*y{etVvMg9AUO
z;czeJ7`i|T$<16G?H3UjFN`*TT}5FOX{?H(v$0LqqUg*(2rG&jv7#vbCRr~CL_1eS
z7eiOes>r9cP(a7E7^<S^dP0NLVM~?WTu*Q?&~4=4NmdhD!mMVli|(bhP^{)Kmp9W|
z(m6Y_N!I&F9k$NyW?jS$WX-l6Q`&j)=s!r4kRkL%=xSLQeV5ikg^<u<QyBf6)<T64
z*J4^c!ZSJ)XYy5#Dx)Lfl~I_6%vF(!rdU<97G|BMDw=yXT@^(bPgX?OEm9FRup&BL
z7(aDIBn4!sh?e4%SVgo9k61;-jwZVzT8-NR=fu-VE(KACEkkv&nd_l_!_wmA&^j=x
zD2KvrRW)=5HpyBIofZgT)zGdkRzuGZL_1eQo1v>^HI$>ZQ0T_BSX4t}qz+qx>}ECO
zV4xex!IM-bwAfTbchg!ZDs!021GJWO&W=<=?;>^BI=h?I5I2xD!*)z*=W6Klq)EsS
z`ZwrmSq*)I)<T7l&|*^!{gBo|g%H<bs)q244#k;#)uU?2uw`h7gv`Z|3Z_^wbP`ND
zO))h49J&|^^PQ}PuverO+J&{yDZ=!rYauBhLoMXtj94wS7>`&j#EvGr7FxTVZ@$v-
zE3h9LY>>9gue+2)9rj=`NY7ju%{K~-7f2_ArA2{c?4s(VHP|w1owPa-!s?`##Ofq<
zD-FIOceU6BKFplveMQd?Bs-T$=RjA>GU*Ch3k7>zi=j-4d@XO3RAEb;u_}paNwpP=
zgL$qc$4)Yw&=PjFdCBMwS_{Q+4s&@mttFMKqqoxjk<?-9>uwfF+(6dE+cBT_TCqwb
zu}xL+8PXnP{ConsT2@G3p|w!)BedAe<$JUiDt@>YsF2Wl3W4!w9f31>sYm6JVaFX4
z6Pc?d6-}`!={T5snks4UbLlF{NP4nZ!p@Om=_Ob!9WNY_x>%9|G89XP;KW$5v<Qz_
zvBZuhyI4AlFAmDxSNV>0&d(j4WOG~1pd)kPba3ME1&)QbVP6Hd7v&StQKj$+*gk6!
zb!;Gn6;UrGiYWM4rAraDF_7$BM6HLemPORXv=#~(xfW9qg`eeAcFvP3Y>72qM4^^c
zMU;biTIASC5)@i&im01uEffhl%;lA|mQ=2e7E!M!b=dm4n?)2ikTpGb%xBl<k^Ysm
z2pK{D3|%eDs4vo5s0b2TY|5x_(OReo;#v&PBjE`hg)@1oM}^cemz5gqlfUchqpdN&
z+HAnPPWX(9iS*3nl!~iZIdu$JAWb>-4D_R3cOpno7F8%BQdGSZi>ejEDXEJpDIh~p
zwctEiR2_&%tf*o~lU-Du*?^C;n_Ta*47HlBsghJ=E~ySmmK<M@8UVwKiYmFEBKkpW
zp0%JlIuODNs+V`KpgKK}>|9Wt3SBJ=s?D?(3M09e&IQ$#qzYS-?P5XYV4fN|c9IT-
z7Mp_VdRhxbhYoXjIjtp?tD^<gYe*fozV2p0#SLUl(;f5KwV?ViX%RAlJ^)=U3#xyk
zwNMcxwAd6>U!}EB5yZ6^3MxFIqi`lq^{AjaI$lsE@{+lpQehRVr<Q;L($rH=K|iXV
zN~S<rPN9TIIrVZZr<Mtqq%NnVfDGl-oaf1MYAznJa*7>wqnt{PF*Jwxkm0fLec0r{
z6kA(Q=O1_Rb!02An++pAG?>6)!35Y*+6jDO7L2D0A2D3EA>Yh5{AO@*;Ptafzz5qs
zt;&NhI3KG0IT-B~Dec(kv|_6BH=F0hC9;w(^tt3!SDtoM)8E-#<1J%@T()k&JFVb1
zhI5V5Xs$61-;!VB1x>tV>(+@iD||+P`$iRftHBYvN3A}Hye`to;TsC5uQik$Y>}OF
zR|T3_Dfvn)C4UZ1m*3!oFL{oQz=y5}^Yzijp!`B&_(wI_ivnq;^0B}C6i6=r5&j9q
zt{l?OFxVEAO+yMTVxE9nz-?gexIfikwssiIEX<hM@Q-mYPbUwiH-KNpX_<jsStJJJ
zE1;7#kQZl^+1Fc9lR$B%_11~<(P#|(J%Y`zL~LFOzhev=vAMr^dJrt{+Tf3rs=k-^
zcy>p#&1*IAb(Ln(_X+?Ayivg)uGRC+T79tZ5;(zs9IjQH@B!|I*YLOd^?b#vjpK*i
z;9xvo&yV>{zuo{j`KC9VuX;m1NQTedH)|er94U>q>Np700iff-zOV3?{7#(5zN8w&
z^54~%RdnAXd?L{~3=ujtfJck##OHP=(499bciy}}r8@uIZh!f9?YH)M!iM~Y$Qln8
zl>74dFZ!|W!{SZ@)_SB<$MemiR~rc&jK&9_fNuu-#O;W?0_pb-Oqn_b|5u#e-d9%>
zpiJmTHTBoHOuc+3C=|1QHO~HC=naAVU@#6)&)(Y=kP>-Q_-#(g9zbPW2kHbp4{LhF
z2aQ#}oS4d#(mpABD?@(7Wo|(X-G^5E5VIl~bmey|%d<Y*^2yYM&~uyW!@PA5_f{w-
zD>5#s00dsAm}HI2G3g=R6#S8VtJ3tg=PNDWD>b|$`6pLOyZyqEhOpJ66=h3r1R>0R
zR7U=~i;?wceJ$Ggzxe!ZqxA<)%0#rH4x4DbJULpk8Cja}sxB({{td5Fe*FFo4-9H|
z+y3BUr6BI_M}Pd5`NKNySs(60>$qorzefxa5F`})Bc(=8d=f}x0VKdstyTe4#{cty
z4NKKRX*kb7c}LL)vaJgc_IOXmi-YFiKRaJBf;B|_{xQ|kZ-z4}{^y?$(WMpXwCyys
zegN$~7kWq0TJ86Qm-aYo&9{r4)QJp0EjAgjshdWY*6XfNW%yMtmI$K!LA3fxW_8+A
z8oIMS+*8(3&b*rx<=3jz<r}8u3Z;5Sq-WI!lpCzPEY;!B*IIp-2Ppu>pb{8Vg={Cb
zhPR_ssbB%wfYyG&4EPIFcQqdAKl4fDhYMZ&pcB9w&=s4QE3C_v7da^t;g33O!atXc
z&t0*-v}n)uDCg8%>=h)*8__cr<{9fGVSTtat&@aBKvI(Il4TIT<YBurcgzuac?9%O
zt9r?_?e%Xj<@Gwg;Z;gQ^?ZHTa&wBzH;hL5(|At#;7%7G=n3*>bi*CY4b};AzmqbN
zAgIG8L0*)cAl)#$v}do?DAzpXVyz%O-hzI42lI<{dayp+m)7aQ{GXH_xn>;}5OO1>
zO6O|?9lc7`@fKXQVrRUp#jy66kbcMqSQTV{BqTwf(kXqV|D^2t1sA*OA^%o1?dO<j
ztwa8sPRc~cqYj&p-<%xsDXc23)@7;6@IQ62MG(+$L#zLYS=~CISs(5x>wsq7O$z8+
z;0m<_;RwK;HmZHR$mP(l;u<O@@w?9ef84K+SNxr&<}ME?qQ7(0&y5UBIx+UUa>L9`
z5M6053f_U9*c*CBFA9hyoCBSdiTFnyHt|2^691X)FHO6LwJOIgb+KiTKkr1Z9mTw6
zoj<G(_q28Xu;@t2pHGbCcM4Ky<i{Yy8&EC5TU&kXlDX-R!gaEC>5Nwq(4#cg8uO~H
zu^~8%t4NX;{GCDuU*BYF%msg_H9Cr{0SnI8v`WL<@X}lK80<T51Xi<~11!npuff%b
zybNj9TUEGt=Zo3RBCJPo<>snu|Ja;lgc!~!FK%%0qE0^#p(D>=j*QR`7*eW^ydZXw
zlQNNhsKX}x-un09JZYN0m*J{V8Fukrknr!uIN8eL#5&<wA0AuQ3D4p&Dd9hb$$^lY
z6Z!6*?iyfcOn6O_xKavM9^6v?K;x`rfF+z&{=Ct}pL*)Q7hQTib18jO<DD)kshMRG
z3@oko_cUbX%6nW~8Kma>(2sY!r6%jceQ%wbEMAgQ^IW)t=;z0fDlM0<{c3&)uH;ld
zyo+ll#o0k4>0j*^Wxn^im`~4y526Jh;+YV-F+(m)f80r(ph47PLxZb3EliuuC+(BA
zyE5i~xXdny#Sft+zr-wQO%tpSx4v~OGLI$2;yZ>awc%~>Y|n0ZtcR`#gNG3KB0<By
zrsY?MeXrtIN8!$Qegqy%fO{?5{Ay`8yUgNuS(1u9SR0VV9y|xdhSA`+&!OzY@vy2=
z_BIFm+{oyqOJlz)U;Wa>S9;)o7~S<#<}R87?n|bC$u*(0&a-!b%1wWDaZ`{|A3-1e
z#Vw^+AMRJ{lwz@vlu~sTvpHC(;Cm8VVm#}Giw?XOi+)%)g$IBdYS98Np&&V0!;o$b
z@8V*<?w@nvCFa~X&^Qbk0|{r90}kDsa*^U+&<6)W?@rHuT-gb>Ggw4Uv!^<gu~xVk
zD~Q&Qp{0&yma--Q)`wfyI$D`WlA`qrJ;riPAJzuo0=(06C9+*|l$wL`kYsJrWmFSo
z@$+3QuEY4p(dg$`gy-7iwJnR;%h_a<q9%Nm%Y=gv`vlJR3bznreYl0JLyTEfg&5uM
zrpRMX>XAJ4H09;T!Q*lGgc5%u5+2N`L3uJ<X~Cn&UZVz&KIIFAlIv5E?B=+9F0xgJ
zKkEYloNlcF*6yg6nm)hHy>*~(tG2k?7_XF?IanVahG+irqyE+bZ!0_j2xvX*H}J{M
z9IRF4cGT*y?$yAPg;E2n_S|T_)*6S*+S%kk`(w47xZKOZJsJ-8*umckxmV?D+}er0
zt=oKmJXe8dONZdPVH--xR|}|}eN+t!{?>349{SwMAM5Og2SciR)^yq<DAz^ev^cvx
zw^?d}9qRrFJd6Mp<HV~bf`?}Jk6B%2&fJJQG#$4_bkOK|3$|WV#(}klz&3jg^l7r9
z<1N1w+6BWqjEn#|)I~&?XsYhZIBZXoMGt!s5t(w(2PX!vrskbp8nUo|V<>Fi1C2={
zM>yY*LE6|p*4wciw6$Q)R5`qb=JVPJAMNc3pOb(Xa)1XK0Yw|%6B)-R^C-c)`^SJw
zj%;oO(X=uBHq)49aA?%z(TQ?E97N?9=SAUF6FO~+OK}_ppsD)fn9=c()EV&lr~V&%
zUjimcQRUCToGma23<3i~4OjQT(98^XGYmI^awv|13|Ld$mEBb}-Bm?Z^*|3OBXVlv
zfvutfA|jW9crK!X2OjISo~#P1EbFzxvWl#t0{-8NBO@|1>SaVm6{X+z_xr%4E93nl
z-h1(Wzl_YtL|!-Y<|4tHqvc?{m>Y#lZ1%7VTvN+U0{C2d<hkD_qZ1hJ#&R*pm4fkl
zZUQI`IIhX`T~L7Qjd14yzp6h&?nr1&)u9cn9Kh4VeG)?4Ny1yIut}ky8sj6z@*<cU
zg53-Sb~_gK*sxb8;3?9M?32dz<tl7wnBZ4Lh2h#{t1sF;#pi{{P~?_}v;V(?<n?3R
zK}u}y=91rGQhqg-lp8`CUA%N@4q}Ny3GVF;PvLF7`c$ztl3Th|&NuQL-ZjYi{h(*f
z;r4_two{cV+Wf#Z#L|+}ehS;WN`tabVmrbfub&wxErsn=d%-{Zz(4))&$9Dorssjm
zYm|fL%y}~dm4{(J)bt_Z)`j8qSSN5zsW$i(TF`iSvK7FlzGeC*viO;Uy3yDUKRLhU
zE|>Hje&J*I`1A>L>_8>5d+g;HSiXBKdajE7X=LmNDpotkP6>@L*Z-_@FKKvPXxX&O
zu0rj*J_nk@qJ+)9>v0-|+IP)S(qGcBiG+yc;sTJ)#o7cslr>Y?v<sx-Jop0^-kFN$
zOB%$cx6+Scx;VJ!tz_THE%pM%<|Pf+(<mJZ^_?_IrclRU(r_<{VH<1Tmoy}U-Ck_S
zZK?D6pkOSnq|Rs8moz*?dW6iNA3#%>#?cHqO47)zqcolNOB#jBAc102ES{oKs0`vL
zxRWFHMt=N3e@4tJQr+3JJH0ziX6Eg@Dw7i1c~5{bTW{xO`*rz_-kr_}ck~+eM8~#Q
z%2zR9qu6%epP_y^NjM%dY&mQTwv-EKNxa2C8j!3VjhuLk!GcgA%tSU|^EVkBfCFN$
z3mwEG_9_N;l=V#pOi6U$O$MO^V$aS;KD-ytg<Lp7YYD{m4wmMg-kK@kw%r3F5{_q!
zg9PH%*b5FK66i#5X<P!abCm>oDW1~y0ZmH@bbM%pNuWnNCxOlmEjvq~wb0a30*%rr
zR8`|B=@O_#LfFz`2PKe$d#)n;P7;_vi4vIk?dk8PQ78g)7|Sg*N~Tc93GEXkhHb1n
zDS^0yqUBldM@#~x&gZTB)ke8~aq(p%AHGL=hRmdILsLsB^m7`8$|QkeGnOZ46e^QA
z3P_=-A5~Sx6ZER=6}cB&5;2M~OCpt72}!gD#+*eGZTAX0NyK_3B@s4?Nuozl5*;HP
zpIH(~10Is79|y!F(EuJXNyLtNEs2H_8^fb-`6PAGAh|S@u<X4)WZ0{YQYf}abToKa
zC?rDzH6f41^L%B}QK1nglYX6$N!dS5d}e6cStp$iO)YiOdK!f)c^t){lj8gOt0aa_
zTGK9TMI6TZ73uo=frEoCB|A^@n?SKqN^hr8D1LJo%Z)TjwqVCD{Xa@#*ao|kN{Ktj
znr~C5wDZUKzDc@-OrftqQ%i~T6B>m|A%S8ek$z31P$|SwKq3hTs#1)<=%v^zR$bCY
zd&a47FiCJ`fus^DA&?G-foBm&eXk@1l7aW6O2U>gmGo;=Nkbw8nN^ZB;GvS1;;5KP
z+7pkMN@7R7R!K*xOC$9T5eK0Z`6yo^k4(nBwUYVGCI-251h`trC371!O^?J=eHGJT
zp%JE-o=7Mr^;tf6Gz@NMGKClAQ1fT`P7N(P3#OBysij~#hen|aAV)C>rr3MyN+g6W
zg?4aZ$-zArlYJ+-PM}0XZC+S<GmS!Vox@nJqfs)2I=--UD~Vwn>rTog?jUOlPMyy?
z_YbnyH61uUK5@Wk>%by^+MwyO$n+rTB{HkN2u&>&)DLJBDysyFje>fVMxnBbqkw{H
z?^9K+@npSPdquQMqG}&IGBoQdm3awWwF<_bMOV!}lhjpaf|Sw<8^@&86DX|?7J<kt
zt)u}DX|)&!#-!D5c*LX?JL<KxS}n>!p1)*?zX{9!*4zqm3CdvGTUZ%ih-J`K2Z5i3
zt}-@J6ZH^0(^pol42>{Z^;G9%)k&dcXIXU|G_{mfYiSg!Dsq%`Sv5>T*ivl=WtD?_
zHj;fO2~eQe$g0a|6p8>H#&Ru<k}1@2S#=YMVH@jC$|~+4YZ6YK&wRwZ@Yu+X`$><G
z8FU{swUkibrctO25-2uf`5BEuWe`U(JhFs8=vCM&QeBcudne?QiQ&v@NhMT5EiH$k
zXHiSfLp%9JndH|Trj4{G#S*rSiKVAdEDee<WEM-(fQMLG^eQQqcETejme^6R#nM51
zQ#602A*FmGk1zF8XAUZp-qJ{aU!p+~9RSW0iby|G384e=1YaSve`tgWp=UZLgpLU<
zI}4%J(9}{0olc`rwT+{s3!!sK2wQUOpb&C!Po3;LNn--VMhLxuMxkiTVJy=$N~Tc9
zh0yy*4BJ?DQV4MeS<`Fke0CK=_mUnVGw4&$)KUn2jYgp|NTApVp&!vGR0eSrgAl?W
z^eXHXsV)hj#R(y#(U@8Ps60yOpM7D(S@h2{(2meQI>|}dgH2+x=NXhe1H$Q<WsfxA
zA$u0Cm9l3$JYuqk9rarF9IkCpcho-1N$nG}>n(s1Z_PERpnmYJP(g{)l@wZrXZcE@
zeL^El3O(OBDKr#Xc9ueiLQ_jAbTW-XRX2{3E`?r0LfFz{2c?jMdn#n#Ng@*{Hd5$9
z8igV<hp}8iqhtzoTnf#Q7`CzQq!i)~vL@Kn`Rpo%K0$hf%%HoWsihS9B8@_2kU+7K
zLf@rPs0`vL1}TI;=vCM&QeBclxr7vovzl2Csbos%p&Sf5iynFk+7WsvL3dIVVY`?p
zdLBj5p2Gc^MUgb%A&M5fT8g4RJYu4V9raoi4JBS5?(jeo{p5DSvbQiYywThsjh2Fk
zg)}lWP~vDWJkM7g?GYMb;%GL0AvN+TVf7pmydK2mIi!O_%g*9x1vIr3N5|19RK4RU
zCUJysf>(Zi1qoqGm`QPjD48DHc5qLT>^n(q0>wrgHE0xy+8oBRnMTPJ>bN+1D~Vwn
z>rRRz?jUQ*O`XrK;^?EKN5~Ai9hzE-qx)zSDuV=yjX3%SjY4G*M=^*a{6Vk6UXkjO
zI7&XWo#ZyNJW}bDkVm`1z_ZAsC!if6j|`M21roN538dLufk4_#gdnp(k_J2k(pImL
z0%<lLF@eO6dM%KSh<|LsK_4AJE*r%Sd&{Hkk`|Nu+7^R*g*r-}uSk1$Jk3`e?GhSc
z;;654;%G3m>@1G<gQk|^Xf=&O)jN)oE{<MCLf8^!2gQ+tdtOKOouoE_Vk3^mX%vdu
z9LBPVM#&WFxH!6s#ITKZC&dwWkTvC|&*zmZ!v77faM=ETE9n|Cm2QTnmMZB}GzyhU
z0>wrpeVIm~Qi-F0O4=ssNL82d481OU#jZ<oX=!3DDM3tTwWRVZp_X=pac5CWzlL^%
zT1rx&6ie7JCYJh8EbSx$kXbBA10G`OIUE#w4EuRJVq%FM^;#^g3ctJKppy0_RZ?it
zTOMhj05Yr^EjlaIMcRpK;=Kfa_Z37tghrSk+Abl8)c1o1V()e{Z&n)!4La+fy`ib4
z4my-Zp(+|jG3cQ9?*0=<3|n$c2qBN%{k-ee1vWLd&KtgmNtckM_&OfksJ=449<WZM
zP^2hOY)J79GzyhM90f=*vEdtk(dVRBthz*oyC%p`RWvgTszge#;I=T}EG&34v?Ewh
zD|nLrutkjiwnO^cPPjZX{Ye8J^!HgD4WqyN@QBeLJL)z49i+a*<-mXRgSE<}Hv#tP
zwvsCY-^b9xpJAryXDSN(p@_W<UGxRimTv_7x6lZqz?~8lnE8guFGI`DT=*z7wdBI5
zXcVfDaTEg=#y*$3#o5ua7WH#xDr6{`ZkRZ@XGgN{BxebfC}){BZtqE>P@Ls3mK8Kg
zrclS%1lEukwz2NyngDl@HI=5$XV)7h=aL>FGw5t+3ez~6L1xucp;4#|5-2vR=Rz8V
z${>znxM6}nWD-55NOei{Bp<BRsK=~&RPPe1=bKP+v#6fy!Bd3l(Me9q9&8emJv*W7
z`L?L~nPrbOAh|O-zpP$}{7D=RlRclpBPM&;QLkms;jxcYIcOjH!P=NzZvm9}nwDW5
z;9KBZp@I^pD=G9HJj-`A;9o-{ObYFmkV3Aj0Y487IxC=`K~qZw^aPDU6*Z1xP(bn3
zfWML$wzSyE)d1f0=(N@Q5qnW?YHXcX19pBbm2xC6?g&jSd2wGFh2lklV#A9E(<oF5
zaTL>P0RE!SNv~LSi5QcQS;lLenH^O+CD`#nm?~M=@oaDx!H$XICn*x!#VB$&q{y#{
zDVv!hr2!9$tl^j#Mb`0%Q6xL+HASAFJxv?`hNc5u9!@Tm6BypqNq*hYK%-ZJ{{@Yb
zO%<hHjV*jB_43dNqtrbUl$v>U;eDZHXKsBrG_~Z`+i4W4Fme<Fx5ic%K0`w2l;-pr
z-R!Fi4(|CT*>{oy1xobV(7d|v6B>o$K!>sXnnuYK>iFuy(<Fv%tnb?ezP5$0RBplN
z2?KcOmcwLPO5zT(rsmZ7JnImLrwq0~hbl2-`fLkLEhSWrMxoM2px8*LK^lchAC3YN
z3Y@3vFaA}pzg{uwl6+d8xS2w6o>@w%luAgc%`jcENU1Zyb%c~62~f%^v=Nh4d!nqm
zR7~N_vPv58kX2<IACpxTJYuqn9rapP9oP2r%?_gKP*PO2MevqX#J8sng6fUnc_FBX
zhH5&$3D5VHQ<sEBn4B8uoSd2oEj!Dpw?k7)IdwCQLKR7lk}jw2At7ujHnCEXT~0Z;
z=RvaXBp(VC8#(m@8inFRhp{|LqhtzoTuwbnV%WyIlX8kX$eOBC=d<h0nyuGSRff!<
z`Owr-QSC~jP#GjpY!uZ1jY4G*M={*2!5{Q0>=mgliK_h)qN+XTnI)A<tAwQ51XCu9
zq&gKGM@TAy0;Qlr3o$`8fP(5`F^Mw^Drvw&P>tZ|n4l`+5ffDGsMmt(=&?p^vYu}R
z7q*)D;X-Sqly8<d1rBQJAW}`4&3j9!o!Z$=KIL)|I9%wccCD0%dOe=(E25@CBTPgM
zCPWnDdM^2*d}@B?^R1yhXX*52Xlf~)uBTC`YRFO2rPG}xge{$RP&)CBTi58+z|O{}
z_#M_KzCk*KWYY(EWTW|Ge)QoX8iit0fnvj^zob#9{NX6brp$k8I^qxdbo7c;m)NvF
z!KQ6$lbK3YP9><c1*6YGrSs1vskB`clst;fV>~*DJi1W?BQuXm10FoO2M&$#Xbz7U
zkFuj)^XTegxmg$<5As^@9Y{3gee_A&-VC~fp~IqbMLryT82DMxr?H8etViIPzTA08
zXoPX+!3pkkeMI7Ap+RTzJPDdwlIPhp3RV0#ih(@ik4TJ?7@66V!(>|9=Uulh%IUFP
z+n5X-=<_ntBP4xp=CQS*&ueKEiarI34Sn83qfi;dQII~F|J01cAM_dN6{#-Kr{Ue?
zhH7W#PL)sz?py&w&%&J#LObFz)K~%~abnvTaUP7sxk`i~GjU1-9>n=Dj*AiJBY4D!
zlO6S%I1k?t1oixQxf%>Ng2INT18LIF62|O$6XwE%!Q{2(A2HH=M@8avH7);7<bw=n
z^l^X7Z54kAjWD7d@+Qi;=S7P}v{x0X%xwuxEs1hR8iguz9L0ht_arfTL6p4f*5x-n
zwvI%34CxV)C|C2?+7RXGGzyhL0>y?X&!thQ4B{vTqQoCE5v3_oT_VcllZ5etWF|?~
z#RN${0W~BGNp1#r5hR%?eUc!tU5p@ykRYEFwJbA1N&~W@MrWGU!}jmNF)@O?4v!c?
zvZG!T<f>NCY~>x_UrleZ4=s9=;sVW7k`6xxo)mPbov4WL<M_Mp72{o@5k`c^B#2Nw
z0SFs$VYegeXSAM~*ImCHT6Vs4ydRocGUE4W6sn$a6ayp1J`DSF5+XApGL%dgjt=g5
zitIZ{T>>Q<YV#9-TfC0SL{j`XjAdIIB~z&5x2$&~F>GVq$y?UkLD7<|_anBEB6U8y
z-m)GdJwj&Cq0khjaWsR>i|i-UD3tgSC^q8fH8cv9K^(<!%Nl>stFTw3x+H#*FI7;)
zW0pNCjS{lwb|@)XWX~pW6d`*wij#r|Tf_v<F(`QM6h$qw;E@JA1kXEgG)(YZi$_fG
zu%liJp2PV!S!&gMy-}N}w>q{W5VPwoeikMSCMD1v;9I#O5I<cBp^xBMzC!4>&<GPk
zCnbcC>(cq>Lxaxx=U!-PseisnqfjM{qZss0eChlrB!*5#PFo;6md<(CtxIicY@K%(
z{)u!6$%;?$;M%a_>^zm&q^=SuHmtZkjY6dmM*&t$ylV)5(dVRBthz*t$u)s^jWLs>
zN~Z)l-VDReLXL+*3<z>e)JjQ_*e*tqCm}`NBHW*uBBcQjiaZU+#3=G~JYp2dj(Saz
zYv8Nr8-qr(&?*B>wt|WJc%kL;(AFyQp)I3zZ@OI6#%z)_1Ms#W%{Gk`b&lb=zSKDq
z8e!CVYJxg5uOVC*T6X5p7Bscw&?{&ZssM5n1Bb@e5N1e-%oLh^4Z*=Zx08J*xlW)&
zxz4<X@EIC~x&m|<%Y!sZrclS%5Pm>n*v7h(YY5yy))bsNpIz4weouOY%%CTrDNN&N
z2AS6oo~Kc$3=$|dYH92BRK>Iyi(w4`;q@x)6{#+XrF|0*>X`}2ESXeFB_z}9VDMQa
z(+UU#A(`4xpA=5mIwqV>Md7qwgd($Wk_J44(@Sw=OgNp4M@%@eqh1TAL-P504Q{8D
ztL0WcKd>xiZDAs4m1@OhGt>K#dMRSmTPH1ym>XDTcq(sto5(rr449R%zbcVbfGFtG
zGkR~|@`;1<LL*Ehorxmp(ZjabV%ZyKUOzKXnlrtn=xb>Y{O_f3t7R|vXCL^dAO2Ye
z|7<E<gem(AXa$*SGT5cLza%mDqYgCBdDX$hoIe|!k!8+bj`Lw)0gV4|)7!;6H4~4Z
znCsKY>%2(!!SAt3a3)rQb40*0R|08(tORrUB$)~HNLSXt5Gyy}U|AWs5sz3IU`L%O
z1MMjWeX&2fcyZ;xmqiml+DP0Oo-;CDXf{m)_((W_kA?$aM_C8(z<lV>?mU}HgL3AS
z=CZ4ucGG5hUQlf{%0YAHyqST@!(d{1<!O`St@3HLVqrX<M7aSroLsq@(~1_<u*f~(
z_U&W!i}hH$zN$P5_|c9JRDPxXZF*PAHoMr1d@pa+y48E(_M@R(KkIn^n&r9G#h^I?
z8<EEH&3a)Z7|MmG<oegFfkJ|2aM6ZGP8GVZLYVYISFsU9--$JfR!340A>QP0G^_M^
zAS>RyK9-JO3Xxa|eKnTQ$93Ort%Yn*mi6c#)nNZKv}U@t9H{&mELWa^e?_LI?9%Vx
z2bdwIE=3*y{|cgjzro#9WL3BRH>F$iu{11ze;K>=Z2N9?hZ`z5Ky&?9zDd;o&qE{Y
z`d^w`>Bm7Mk|_S{_==gze)8|J0{&_&;P=D%n9arlK2SPZPQ&S?#}%4Eb0{}dpBkzS
zSBBQdi^BSF%FGj|YA*uWp}=9n!t_vRc-It*rAOqcmEF+_&yysvsqEMz1{R%c1Et?g
zZ&{z3nVwspIsirzEz01d(!=1F#b9G{INDO5by_qd>-o!|-Sj+sugz3Fn3>)R0od~k
z!G-nl@<_Qgb2a<<f<mJ@b9L#jrN5P)Wq~R^XCQ0r&5ujFvzdGrh70{j6$5OrMsKj4
z@Wte?7R!&E5RAgpu(?8x*YS8LbCb<-bu8B^1-W9ORp2KTgOOUJ(5f{C`_{q_{Eu*X
z(<+p!&0I5p#lpgPu2#qOLO7@w8U>(jSV06kg;s8)P|XboU^yt}TD2TB94(JcHn0<j
z(F|I-(OLukYrIe$n=Fh4xrtga7!NDdKUJSqE*>lVFB89gM(Lqjhp|?ntGsE)Bq3g<
z{cZXb;lq5r&?@DIa_|^ZUjhG(acl&z`ML=NFN4ZXs*Q#LM(0CJAU1=2d=^*Us$1V$
zw;obP<d15^uW=c1<r=^d`T9B>^qDNdD>v%!YynTq``D6U(FyDZ1mhTp2{rF-4M#{M
zRq_tW<U#45jIEU;Ctb!C&iXuhvBA7Z=33=0<$2bId%nHSfc&gnsoJn;ox!6ONyTxQ
z7gd=5sj0|LGN)n=%S<sCEr1Tq!8@5IgIu|p+jC^H*{V&z2J+&bDcOp)Wi-MRW`9&p
zey59*^=vJmpWnvEZ=0<jbOICEiWoN8dUpG4^>eax;kg|Icx?0aDvpnBzF|<)3ZHQC
zQkeI{7>|#!cv$B>Yr|t`o%by6XNWFB1&QSTXt|l^<I0O!025%iRvU+6jQ<w^36`tH
z@(A2jgTj142^N2XhN!(c-jdPbpgH+ZDvY<Q(JxLw%eWbSQThKp7ia7FUqpX@oB2Ch
zt<@tSxQL6R)?)i}Cpa+!5XEK&oZ3kzOYe2osB--Ox_BbY@&LX51oOIema{fIQr20{
zqT4RZFH%!CKT;Se1^HsRky`1Q=m3?$RAspljDVD!!YKe6P#FxxAT!*|T~Hn$N8JcD
zzY!G81%F4?T=hrB&-|l`!#2Z^T-xFT?N&{$jA1AiKx-%hwgj-)2~4CvV%VgAzI}b}
zi0`FGyN;uRv)sjB;Up<xWR|hW(9zT+5NpGuX+23;2DF<bQ&I-;E1f*!;51<_uZq}1
z7rFMdog3I#F6i>TnHw(;HwulZW#%cegP}Jvp2pu)3|{6E1Dzo%7>1Kr7_2AA*-l_$
zf*^*?1bJQi3DOD2OMiA7Mun#A;;nFcY{0mTvbb1J57vgq(t3KZ__v!LdHxg`+jEpQ
zEA;ABuGS~vVzmK_4yd3ot#f?i63ho!;${EQW)k!%Oc^WvPs*=Xy7*O3`3ZFGo0w~@
zQ~q5}U?Sxa!zSfVYoGENyehrcVXMmVx4QTuOz0|l{X@*_)(Oqp@JLxFG>dM#guYCB
zYjDP_4)7wEmwuJkpqRwp16J_sL8Cq%Tv%>R<!aU3Kzi0MjCD)~G4Xd5h6i23pyz)b
zBk@HR3G4j-ffJa>f5foK{|T4;_x8VZ?F)>m0{34oz6|Hj`53iFS=6lO4{O6CZ9RWj
zcC?#658(H_d4X;g;9=9C30D;5YO9Z3GPi;;xK7qCo$)FHT9hXyCvw%viDCE^SCJ$y
z1Q!Y$d|8t%Ef<5~$+0nH4OnMBf3iHX0WZBpi^0BAMqv@VwVX{d`PcAjL|%rp8k1GH
zc;{=_trDz9ad5NhdU0Y*GD8f%s3^`Ff#l9|Nz%kXKF0zXn||O(2_5@%!$K!8G5rw3
zX8PUw&k?+tclB*CYYY6JSY`fa>9@+6oJ$zPncu>M*&Uivx4LlQSG!VTZFrJd&wQ4c
z?PmU$aF$^0&5MceKJT^4?K$%`OX89#T!L_$`-_aDl9RB_uPUBvT;i!y@C6vsAr?~l
z?#7)ib5nCm4;(DL_RrL170NSQLK#lY3o(wTx=l^ihR5D|YO;K3H#JWVzbkILjvZ9-
z6@_@`F(^zR*Wz7VGdpTu2p45n`-^hlgp2$1nQ#$$u*_#dWPy%cp}xooPVgb3*zn<@
zX)Dxb_sRUv+FUtv+GTWME?$hDyqbB^S|GAE-22wK$RgG*7w=G?drc#V@F4`gTF?y6
zpA4!aL2f*#j=}Bm!YDkJ05@GW1l95;cD==Kx3p7zut*?PA3X8Ky3u566ZIc^l{G%O
zAX}T0ePOKkGNg&Wt61IX5-UCN-+<w|m4%C@fa}^*KyO_rz4MRTK^3OYxr8a4QkP(i
z?sc0|tPPK=^^{_{&~8dKSkC6dZSa=&Dp@OBeGr@HG5cWwHQX&IubN<TOpd^GYvx{6
zDl~$ZzGAI;ZY(l(L*_u6UsV8p>Jk8bo?MDC__4)2S(jF}Jv<_Z*;NeXtlzmfE6mn6
zp{IW1maVJ}_pWuevWT?H*4OGe7Vbz*TRaYTuC~omb5icw9!-XfIz@SWelg?K{N?EM
z+0Z)eIi;i8FLGIoUXI4oiW=}PE&~oz><S$1PHeQ+8iTdr9<oj`=2ewqbh}Gq&pW9{
z^3>CmXRm~%1AJ78KM@H}Xw+b#YGiy89x%@}Yw!$Ip;#=tJ{8IClFR2JCmZmejpboE
zT_5|v@A~C^>$O$h_{U(^FVC%S`}r08(eR#b_UB-tmKT?Mu$@V~D+PbShgl0%CY|&j
z`qszZ7YOGTszpS{x2)vhs$g=Y1rL3$=Z|&b^Yzsi)^*ymC`ZKdv~u}r>M6_V7bl!7
zCraWg#wzuXjhUn~`7*$Rq{n{1<GDLcZ@IkBx_i5m0he7RzY?gr{VLCH=d8QEm`Fs~
zY2rIPs>?x#x-9(P9F9hRk#SDQ9vWZKton9Acl);U`l5Ma%I@uC{#^^w)!pskOWG6&
z*}+A|`l0pj{hs}k^QcX9U!15@vS*3+imIsI<<+OTyb@|-q|49t$(kwqxLv%OYDORM
z>f|i_G0b@XNTv*F{3@hzBTX(6G&x!h#*4X8xS~c*#7BDNo&bCdJ@UM7ld%U3cVoF2
z<VwMKJvR|Fo3J^c$?oU`@bH=1XuvPx&yX7sa05oJ1UEhaqEAAIJ4kpd74{?)RAcOi
zyyoSGVDmzO-G+q^6i-gT)AM>)WuG)^FIPcfPVn2eh2h#{t1sF;Mc;+UP~;AWv;V(?
z<mF=AL245IyGwqDNjV)$$_;Wy(e&b_OLGv*5y&dIr#C!>xAW>##o9=2=~6l0$a8qt
zAm@JtJ!=vhhQyw@Qbn5|xMWyba@tL>`^smh=F#N|m=gKX+ISJxwFdP4eeL&2<_ay?
z9Wp%G3Si%6<O6H%+YIVJu}z$E&dZH1>6<vWj#^{eHY3{t=LVpU_;#=8xhfy#<0)-#
zm0IpPpB)-u;rJkKhKuYU&HQSfT|&#I?UUrY$reFV=!ebb(ta9+dS44iNq;rZAtXd(
zV;p|ey;z%o2ef7?n|9H6!&Q=>Rf8)6zCEs44@SzP<q_B(w*Yn+!d-*OaoD`^x9R!t
zU+l_ihHaC>T~a@WDdgauQ^>xPdv66w^u;RXSM!`jqfjqaaTrTLqhtzo{M9^762msu
zzOUx_%6iEENU+-r5V_x2Y=KUl&#tfLc?ancGJ~#xrZA188FZASmRUz>+U)~03Y9?u
z#im%?MWav|#8Gfp$LY0dp!FPo(5tXlq`I>Od%EA2*0i0y_~gU3uzMQnjmoLSKJC8)
z^IPxJ<{PxxR_*VD*W`OTsBQ1lQSy^7=fURK!C&HTb}l47h&pGM2nJ-`a_GC1Iw>!J
zKuh9<7Se!Z@@O=zzI5<g92$Eo$A9Aydn*S!%KB0ZrY<`0Qi~J0<|rWDwDPc>xm?R6
zn2xX$Swb-BaNe?M`{WQ<IT@%##M2g|kQQ7#nOZ0bwI!bDE1{kjNhcEVw$4eY9mwvH
z5^5ndwUkgxX%wm|a+Gulbs!00ORF7}P!8@np6okGfC42-fMyBxN*aYCK!>s9X_QQ%
zjuYYpiD4V-PD&{5plB)A`w^2+sq=YGf4xxb7lp@0cf5sk37JAyLQ_i(^<El<N+E$_
zGnU(E6e@){3TP<dKvjqF7rhR9#i~mp>VQ-c6`?w_lv25skWzmE8p$H1z6HLsl2Rrr
zl!6M4#01rCD5(A>xGl4wk_J2k)nhn7CaC@kkC>ohN4*wQ$H2Dp34Bo)T(QGjU$QJd
z9c-(mbR=&XwN0esCI$6Z@U&1+v6I!LeHPF06;IEEM%en(T?z5Dtz6Vzhui7JyUAkL
zl;-bCZ8Jvok5o?!psA&LT1=x*^^l_&)KmP;0n11XoysEbV@g|r;xN`<n9^4z92|5M
z*?E%u1d5G*dO3|kk)Oj@UQ4573wCTlVT{DE4R$B>6L*j`C#Ozn{vBm3F9TL<yqA+6
zAv5TW(9}{sy^}_vGDx7<$fpm|C{zY<6kLK@i{J59g&2R(3$a(Ex}=^4DD@N}H?we3
zNtF;ze*_B2BAmVoZnF|jF$R>H3ERgs(_N^U{!dU^X3ZoGcxa|aaBNI7{RbW~&BTs+
zt(i`OyU}^RiqmWr8m)W)&q!raPD7T;$$;mrpLQ@BP2R+N8hkHQl(C7L!hgaueYMn6
zp%JE~KG{JnHLpbVkJM6IK~qaDwTMQciX%s{&{BJm7`F79Sc>w{QVtF}jO;wggaXAz
zOPxrgP)z7BmRHdz*@7L@QUwyjHrSoiQrtn-44pcqUA5FDq({gMx(J$DYN=~z6e@!R
zij9_fKaE0V5Jxd+Df~e%#9oo=l9n1Iw3LaN%!*2-Rzgwz4-iQfMfCtU&q`4l*-+{#
zv=Y-*pF~~tzk=Q}>ndr$Lsvb7BV@Yj-|>j)Dt6RsUA4Ly)ZrQPDcFCSMNO@+R8t{}
zw}x7vnMx|BC&9ZyIcX;<z4Uwh-B&L?5gK87>9YyFl=<1GS!Jqwq+EIqI%O%B7SbqG
z{o^PGxfFZH@a`mpErBM~lG|s=^o1t}_pBuQPEwsfu@Os0(<l_xIgI5r8YNSx<2UZs
zkr=kI?xa}a4zi};)cHKGAGYq4t1h?hrbwrdNi+#fE%nlsGzyhO0>wrzy^BVnl8B>#
zUJ6~PYBBz%*J7`DbxAVqO-Lq1L}sC+(kda8egnlii%|M3xXel@Y2+v65w?xVqtBu|
z`ftHzndOl*;31E`hvQ=M==*rY<PkgSggk1WSLhD$0GIrx2RVrLUnKf@IDlV-17Js4
z2hgc65-mFkK3dV5tOku^#g>d_+b<C<(ZyV{T}-;Ue}QOwlTD`KB2SKY^vgrfKsfXw
zV|C+Ti%PUnbW0*xyb+K3*huV|k*CA%F)8@~O3CbB9@>%Yx8;|IOkIjRCHE)nGlUJQ
zx+$Wn!0(yT9jh-6nYz=RVqd|5nM=9qM52`cFEp|)<?@R|h$QMbLzQ0~!oSCA^aEI<
z{{Vl-EH>6?{fk2f#=kfOZ&|1|ag!B%RAz9kcyV7?3drvdofHe7QvK5d;rEAZ>}y{m
zdIj3#xwC|Bs@<nLGuDylbhP_6(dl4=dSROu_sG5QbD^E{w~5#cd2Tvwgb?ZQOGRJ8
zFr&w*;(-n1mx@?}<ArA;Un<)B*kYm1-uFvHUOr84ce0w(ErBR`G8zkA?CYKVDvnO;
zY&SNOT?#$3-ApFCA|Ac{2}~@X&19?DoA|`L@A&70B8w?W-iz!ci!7of&mBxyk1??t
z4X;k=+sIyQFqvFT`4GmMPgX+%l@lK(bHZ{%*#|=-j0u0*xkZ)FgqFE`qA5!*s(b>P
zLd~_gNP3V)p)Qg*iYl*S+rduLu|<_1kP!OZi<Gdm$9dcqRRl$`4P_4Q`8C;hQc?*N
zn?;qUX%y;u*kLSNZs2Teu7MU~F@!q4sInc2VH<1TMU`Z*+h0t@lx^yK<|D@XV)r!9
zAn6g(>-#}dn8wizGH)nbO`}j5Bv5RM#mi_Es#tIoyaN@xXNf=PS2|vi>W;#m-rI`Z
zG%bc=SnMWK36;>`*Tb~8*5FKV^ACSk!DVte#dIk%ZJ<55oPuq!ciL96pQ6gSQIsH~
zRB!tUZnPv;BBTLXU7~Tc+Ddjgj*Bg)T!BYyIfWf%U5#K0qXS#Xj%mPKST+Xv$tv7$
zEapdQ<C7EcAxsys6nXO5kWNMfZ^5*vjoGAVx&b^cL{pnaN;rK0&-J|<_5RQZ6HdQK
z2&e2%Lf;dbc2-Uwg{GFu>3$l8Dux`zM5nQDsC}Enu%*+){V2z$cXT>+aL}V<=Sk)h
zC{gA!?<9MYMxmI`VJy$mDA|G?W5T|1YFgO_yOY|9J1APo^?t;(Q|gqS!#-;0u#;?n
zbP1V4dqY!8@pLGSLZy&Eu@O%v&?r<2aTE|w!hxz3<1czC_KH=P)YE`9-I*1XN~?r|
zdN)kVEDEXwZX*>`8xE8z3fsq2)Gts)T_?&!W)&q3c&MmLaBNIPZN?*}qS#TdRn+05
z;+_sHH8omJ>S;NBi)S!m*;_Sj6El}oOYZ^C3bhnFSxwFN;yJ!5>0O}_rjmY@P)V7e
z*|{^c>@1RQg{GDw>0TOzDta8nAd+IsOJ5@)bh>(hmX{pd^E0yVB)<t18;SH=8inFF
zhq3&LM#&WF_+r8A32IK+#=4Udi95)ec2no`qJDVW?${*0<PBa{SAkcTz?&0Yp5WP&
zbP<_SOQ5NxWLiO^P$?x)Y$Vef8ih(JjslXYjYCzP#xwOg?G?)|NvNDbLPbf+ES*%k
zC8X2aVY+6KPQ&0XQaZ&5QA#Fk7n4lCLdkTkC>xn2lQiHVnWk_|Ofqf4BPN;HQLiP_
z@y!A(I~gj1lX^NVMLo3x@fK1$x3`?sRPO-S3r*F&otn1q#M6D{)!RZNOkO?SIeB$U
zXxUj_-2_c7<<%!>6skaSlyrIZMH0f6auZ8X+3(&txaUV?-$^bMC^quyF&c&9LWi;Z
zo<_+O>bSgmp2V<?btmN&caSwDr_N{R6{y{+RGA@DXlH0@DW~?MQK%FWC^mBHa2kb5
zA&z2Nfx=((I_wpzF3G3;4f3fiHJO!^%B+NvdNa(LEK2HJ@Eob6+S8yERcIn6svbvC
zH7zD`W>F;#c!;Vd4v&ed79KHC#g2L{st&?UgX869!AV5*r-&$N)LS&oSC*0@=`G+&
zA(GU2N)%m#Kl_TJt3xA96#XF~iumm#?8ZRs;RN%WiEa!HI%}c#K~qaD^idjxs%{*`
zpoREIm4ELeF>DDkA%;Be$MLRP7v0p@IuqniNSBZV`6C|OsKzos-|}l3g(66SVndKm
z(<oF5aTF6l;xGD~^omuN2y(H3Aj2|bCPbA<2|~OKMx2EZcdn%nqAGck0I^Ao0RMml
zc!h9!W&)H3JP7bW91bJE6?nu5kRA1!0FOcP%NJm2aU*`;9@ZD8;4ynAHcWe*#facd
zjf>itP4eVR!Q+A_+cZ)|{S-Xcmo!fdjWE*ubAmLtg3r^4JKxcQXXeavLVM1f`5I_y
z$(bb@g(`R)#lV?tpvt=&Nr=pp$xy7{<)=Tv!aHtVg;N7NYn8(T4Kt)SNVa@C4{J1e
z%xrlxjY6@dK(S%VduS9Yc{mENB{)x|3jS4}h+Z-35>uKUW-wPJGi|DLO3>ypF#IgE
z`6FmYUPrd6fRaP8eT+l@j2wEL2uEfPl?FUG^l=;;<IpGYh;b-8>NSV1VL^wQ9yH){
zobW1=pzg+?t5U8f4fehHa|ff*BztaE582b$L`~T_c&0CR&I*k%?tIRdJ9i50IdkXs
z(A1JU_oh*(^2brqxpNf>(F5+}9k;H_se#>c)d~mld<E$ak~~l4akU}O*U=~xc?uL8
z@*Jm8sO;e=K%VG5H5KvC`c(9aQkU3sU&D2#sV14ZQ>9aaJD&x1$ikg(2d9zTX{>*e
zIk9z&IiEx3eBPHir2!A-ybVXjnDcf#V$8{odd-~t>XTn{=EE-&cK+%!j^NFnTt+3$
zCi(LL@Ob3vGto#*+b`j{z6|<>&<JDDt!4v<MP5$9ZnxU6*#022=ggqrg{GDa`Y4S;
z)j*D7Vo?0dpz`jMBt&)wb^4g0e)Y*aZe5#G1G{5Ce2pO(ALRe$dFJxC*w*J$=|^(v
zd}s=hw&B!WX%vc61&R%)4$vr6K5-P_RI@)-Lg9(}boGj4msqy#HEN;^nR!+vU4m!t
zh53|)XHNyUkv!X`xJstQ_A#d2YBn(KePVt?9%lOjo;2XWv?Dk+#<WE|Vob}9dd;*)
z;}ZVJWV2P9C~tywUi|3wgcH$5J{D}yZYBV4p54)GHA%A<fx89Gn%k&p`+7Xpmu06y
zBaCGiBv{t<#^_r^gU&SjW@u_jv)9uoR9WOG2AYlE7`>Ck$jq}IH%58at;=(2Y@Kh6
zeuH!g$*d3Z;My?jLo^D-tOCV`S$|2RP$|SwOgBdH7ky58#i~nG+HYI*%uK4XD#4^J
z7=0Edo!>}dQgbPkM2gL0M7jWpbfXAHW+IgaJcx7;92z6i93C+uWk<ax(j&b{^uUyr
z-nemZ0^KfYF-e|>fr|xsCeK&Ic?6#3OPq&<Mi_A}Oc1Bz<>|{pd(OOh5;V2s&9i9~
zs_1bP18>Imu8ooqnQ1fo!_~aw)|EIluzmf_@>n(B9CP}B=w+l^Nb20o<7-2m*U~5y
zbqW+4>b!|Yp|Xgh0CnnKR8tdw)~BXd)Vjo<dl~pMQJu`(sd6j9ohxAES-A5-Xh&X!
zCM%)jO>7$D&4tLDt3(ho^QJW5!J7}`uo!PXf=7%u*-@`~^O$(3^M%Hk+ilXc2aSvf
z-VC~^joBoP{t@F3e}-AwrjeSk|0i-qhC2F2Kg$;m{vkBNxO9;(m(FcQt4Op{<ZA}E
zgr=5Ux+9H3RY8uD&ZT>j5Ix{h-f`=yoEliyH_shII)r4=)jYB`EP6VPLb0eov0>43
zX%s4dI10?q_--fsArpU^BGo1SG%e+tYm=EoRaX-v`UDh`EF`)a+(wdUn>r|&6x+v`
zbP+P?lcKO?W>RTDme=Uavw8#FdvI)wNw32r#-!}1*G#&at=;B}lM@qD`Qe~dSn0Hm
zoA#>SgmrHQeX+r0l082LUKZ?WXrP$$<9MF$Rp?!z5yqU06U;ehbi7t@c}Vxmp+)EG
z(EFjOC3k+0MxkmSM=@}x7^?E{&q<KX>?v?EJ*dk&ADzW|M`CXuOAYT<D+Z5nc-rV$
z(jz2&KEs1+L!W&uDzi!XB~WbWb7vZbN+FH{^vT?(5(fXSPf4#hb%{Mq?`=)iB{OrX
zbV@Mi?J)E#%()t(Kr*MH21?e%wlUURjI4R52t#Jplm<Ll^Hn%5#+qyKh_NO+`iEo9
zRb6FG*a$H9^wvxP;#4{SVw2$o_wnqMdkZ38GT8uLZpWIdw1;0N@H}7EEQdxIYcBO?
z%{PP=omq1eG__>SX&Qwpc^t)tHQz^q=w#*e?lB+M<ej&!#_8b&);z+6H}55VLh|OP
zczA7i^J_E;b=fIUY<TlWGzyhO90hoD4tJnpPW(lmmR_;yA#<*pF*B!1sRVNtVdz<y
z^BHL8#+<8WBy(ch7;`Q~<{TAa$jqG5fCqCfoRrMD9Ud{}WJkSb&LcSy6^ez4440c}
zcRb_9y?JxHq{ZaLW<R)C-tkPFucqfRJk6Ii_X&+K*6jCZ&7shuGix3SO)XjTWEzDk
zcpSxsHD5!5^n^8e=dCMoYIq&5HCv=dNY1SD;M#EJr8Ek~nF7UzGv7|5P$|SwlGmE}
zcYR8F#i>h-X}Z&ys4`~4RN0gu%p8n73t@f=+L45rteBE7v1yDi`;jmA6hX+$m(qX-
zUw#XR#jZ5JjYo_x+0j25U%FjrI`XCeb!I&J#>>qA!hrKzo5}MPU;awug$!%-2BPI%
z&HoIIFupv%pD+IuT6E^ir=Y1NU(VhbtxVCulZBh3*zo1{BuG#Al6T&^4yT9L^(u1(
z=@ZiN%b_WR*M>3I&?ppR3KSd0Je5YF7?Y!zt}^kLOq^+oRqr^n-9@JAWr8ywf$5ut
zGwa|nC(cyYnAkMNnFk<e{#+EH%$z9=$l@BEaaQ*<uf$<7&b$hb7-zDhUUTN5TpqdK
z&D<~Lfm+S3_m$;~BF2&zl^+0~3X+VRs`&APc!uxw<PD(_#*YX2^W!H%i_ZLbH#D{6
z$1l<-RB7WVHvITq5~L^m$U7gMt$Igd_byYz>v%o+1nCiy9e>M%Yr~FzrBNt$6eu?A
zIR64_{#cA9c|D2f`jqsFQ<u1LiQ&mvy%w1-B~>CNXz@B2bQW4%4*n-;F;e&BRV22G
zG2%hUi0>1w&&-I@fCnQUk0WA?cmf_VMr23-K#b^i0cp>O{@0FLxQ&;M>%gZ;M%3?1
zo`+}nGU97PBa9J`@MpwoXwjJwE6~)E5ih1uC`RNcHjMaY5=1AW7wo<y@4R)TO%Jc@
z731xsPe>lQg@@OM6F)<vP@E`GY&h{j8ih(Cj$*oE#9#Dj=@qM<uwuK5MU_kmR$L2%
z&cceng?6^AsIC*SRg4vnKvsN>aD8T0lm<Ll@$Wby#)?~9C|Qvm{R6S$K~B%_*t6n6
zoxDy2W6A5pmw-=`tay<2w9YPghA%5F3XL#UJldZX_X{mLv*Nzc)RGksr%|Ya#!+lo
z@gx$YC#=XjZ(V6q!|QmRSR_3{a^iX(TpLbo&?poq3KSbo+)SfTDa28d*NOOdeM)-8
zsV9thP@89TR3ar9aXT1v7Dl`U+SxMVLE+Om*eb?|M<XL{FI=CQ5v2hSM*JL(h%w^l
z@rYd~vZG!z;t_lyV4^k=R9lm7uk<;<YVlN>_U6THqg^#o;}0>^{61s+Y_*E|W04Os
zoY5O?mUkJy9~xooc%nZ${yMbi%#M#iQ%iPynns~Y8%MEW$1SI##V+dMp0FeDymi%0
z53lPrV~+F*$&b51QwXmOKMv9;6h8_S8-83(qfkl2QB2p2_)8{+G{ve*47s%JN@`f;
z%!{e2j|r0eHq6K@B)J~^MUrH^0?PGNY#8Io6OkvsD@sFVo|Fb;VU5l<t2>Mr;Gh^!
zUWiAGC)rW2dGaVuJ{JV#u~I8v3`Po5P8X0Xt+-M*@6DInC%bIo%(sD~1!tOCDBgSr
zp6Gj3c}-}9@#f3@dGqGbqBC#a2u&?{^Bx+7DtR2mhBxmgL3+ZQyz|kStal`K_c1lR
zvs~WN{s`$0k~4q8Lu<pCzoSto&J-v%ocRomLM0GK0nP;Hsp*J+)u*FZjJm{^`?TfD
zNToA#rpl!RXI=$k&%&9zZA#%xQw5Z~iH&2t`EumVH;X`I=1pn9gEtStfid1Z7>^il
zvZGG$X8X)SckI6a@^EqG4`nJH1^Lez880-OGl5<~Dr>+%<=Ail$Atr6M_C8(z<lV>
z?mTl!gUjAH^ZJ>A(wx#<Ms<7Oe=LPp?d%2r>;wPw!#~T;o0*;mSHX>P(44^!Qa%hO
zrdOO9Tu^8fPs3MZubr$0jbj^QlX$TVFCSu~$(5@)jog>bOz)%p=V06BF@9(L)!4Q2
zD4<8XJy7|T_P6O>2-9rUIP%)|RqIyo)e0_b4dwb-xBJ&D&#f*7&5=C3RxRJG7e<1i
zTzE>Zf6W?j37SEz&cmi^&pL>aUVADw6>uchD0)>$R&&Ig9B!Y7F%~*2L}DW2l_)YE
z*Il=@9w9bN7-jU2YOoiC)=UCppz>$1TzLlm6-jH^rQg91Fg;9NitNOFEu!$H3nH+p
zSC^;sYCg`Y1@JFpubyq+tL{)i1;=Nu^U5`eI)5fKvaa)`xs`q#G$M%$BX)enOl3d$
z_gMXYCD!k?!js1OJy1GYPQvNEjw>{S=1^{^J~dPut_-bDvfKJ_&dd`hYm9w|01C^}
zL!s&2Qf-zVktbI^iFSCZY@h6c2Keb1V06X}lzub4Wqk^kb?Z|Hz<8o%89Y^b7`*dw
zurWCtZ5b(<a~r{Eu{JVOdOGp|zyr{3dLI1NnyLpg(_0|`dv`~0VST(jQf|#$&3=aB
zTAjJN^w-kgO3$)zm7be6jSBno<I?VILca>bhTf!V0ybEqH@H=JV{%xF=@M1WM{A8-
zgBAN6`vgL{IyNY4L)vL-UjDm4kwI?nIlY5t;fS;jZjiyAUc7W^uHLAP1kGkp%neWF
zs*@A-DR6jh=~CG{@*LhZF#jv)S(AvQs92?n))nEuGE~s^)fWaLRe+pJa`e$gsVd-E
z=pvUf&*2f1G3=<<Dqz`(@M*TmLaSy$|4VdoNw)Lm{q0RQyG-~yU6g_Fm&6)Hrvgd%
zh)1QIWY>9pj9s@6kyw4$;79nYLTes`Z|YKn@OMNMQN;}rRTbr!l<rtDy{S80VtO2y
zndwz05=_4hG;(HoM3PL8$CBys7>m*`QRQ&T*nkCgA^aU>Ybw*f?}T8qTn%!C9GCdX
z_~a&=zz?}rDaaKItpYy{mIZ4KEE0pcweSP~<4CRADwM0uTr=1hGz#N6T$Za<3*&wD
zLZdJdz-J$uV5iW^jTEZ6;Q%Z{p>Nf4&~UUoHrc>#1hV{ujP#G(MU}fn;c>|a(GsBR
z2%Qb97_5Q{)|+;WUkX*O(*87k=y+`ezMR=8R5#>?a_~W=zLS8<i)A<}4^P5(sdKFc
zga86IlvAhnajjN)t8S`K{Vv&6|C1Wol*`B}*8qmd#Pe}*lh7Kn@J1avd~-Qmi^N{;
zeWerF4G6|D5EF3j)^Nlnu#&ZV+c--)!|r6|rt4gK9_qCQ`siKEM`XrS?oxhbZMa{f
zsw0+Vl`B;n7K?@j&ZV(7DmgVRxd;-lSewW<OOvCc;{l}NaIH2jJ=f+}<*Iu#xvCYN
zD_VE9!;jxw09_i6^(6U&>crPFbz)=NPAri=31Aiu!CFacd9E0Y7AD7AxubF`rE}6w
zQyu(KrVd`%wuAfECR_DMxTRBW@Y<pGuYd4B^>hDn*<0<8$u^H|zD{rT*yi^x*DlE(
z%hdf%(e7801#uu+5l1J=7s6+n>WyHe+yu6f-5e>^$|FH;v{9SLt=n^WbmgIYF5h#s
z4*z%FU>|D(7aOt-UIi`BT~G>u<r>IHkWe{nlY{Dtjf@p^VGLS2E%A31(?7d7MX#k7
zqhJ0At>d+HDU|-@Icl`BTg#GXKDPOK9OYx1Z&(ih;6UO3&KuOCIOk$6B&=wyBI+^r
z)21G;bdG}@ONN#z2s^oqDXi45#|Z2IO{p~k)M8qtW^H)XtSdE(^zV73Biuh-4B&g&
zbyi;bPOKJ&q426dLG{bkVab^ZwR)jd!ZcPtSDG+2xX#LrmZ9Z9FgP}tD>n1|kArx^
zGE8@nW`*%aP$*7;94Hi_n6b93QK1P6tymsu_07C%?iRDQ!2hwO)2Os~k-{C45sUmy
z#o}O>Sm+bt5)8)*77m*9FX^$Xqyu{159Ra|U7Q|HfX(Rf<J=|yYr~^tJpou`aRP`7
zn8+k3w}Oc}TwJv9YGEkHUOwD+ETpWsSCOlY@-?l&u>fvi6j}k^wJ1%Fje#Z}g{6OS
zz0qpa#`!f(8Lm~D<<=Co6m5V;28|W}MH;${Yvgar$0II2))V_p=;l06?1*aC-n5K&
zD)@Vs72x?!fTB_(j!mUr+YJ{>4|d16^30_!o(a?aGK|9|EDqMnoVDSxvrcyw?{?{4
zYXtS6RmRKcd_90mX~@&8S<l+TH!Wf5SK4e2_C^24LNXAo=G&|DWgleZOg|~V-{j(V
zeF|KGuD_nSK6>GpG6n8-0u)mKacri*iLO(?=x^z{4p~#qe!ykyVUk~op8h=Zv~`lR
zHat4kNzNjwlALa*U4k+(q+Nn`<8Hw(VEDFo3uIR?VM<F*doL`%`rL9qM6JO@pRqG6
z!-Zdchx_i!=!N$|Wj3}3ulmo|CDZS{>uW+>>P6Oew|G5QfWS!lnJO>0#1pi;{n0eD
zd<OmbFu9rA--oN-k$e7I!R5B7UC{@jLj5^wNV}@pf$S9d0{Ml|6wqxR16WF<P!B?J
z6vLBrY@o`!2a*u_dQjw+<)&TqTb7l5;ifVBy#+EVXu&;9bf@+A2gD*Y+_Z#+XXc*v
z^E-AsBQ>yF9>7;l2CO`O8tD%5KFles<5rKJucJ|@$LIu#P4<*&6e@c-3QS4#o=O({
zvpyBQqST#RD$~0sE1zD?%y(N<7A5Ys`~@n9_1%^VW-@=Y;u`Q(_{@2-PcxBfC!d?c
zMzOmw??Zz9n<!sUH?n<jT^f+9Fww|aJ#T&s4v0NB_hCF@H=Nl~)*H@@N;>e^*%5Nx
zRIDWr(A<{Oc`&jsD;cH1sA+F5+%|5H-HVR!;pf4*;$CO`Y&8+@$CG?H@xIUq<HVa1
zoVe{7;`ZA*T$~rrxWx*bnI8WwG;O*qOw!}Gps6K2{s)ah6*!J!pvU+p)*mM^GV`P3
zC)Ra>ba2q$$<C8(CQzbU=wf;CG3#QD6}EW;m5iMV_F@_(Td+^BRRe7#%Sa5{VBd$@
zXaPRXCBzW_1YRaJ#KRD+)_Ol;YAAI|&pD|u-VB7$wGk07C0#<M(DBd|W^y!zjtW!g
zD9!P&rco$CBv5R|GD4$JDa285)r>n(m16uwFU4N5>XIf}n$B!yU8M3Up^H8ZwIz!#
zx)S_F>Y_O7No|A;W7_B@)JFdzid<%GBn^0IqZ@HhOdEXwkC--MN4?fY2df*~F0v@)
z+`-VQw<21gnM#VG&ww|D7}8EulkZ;q-B$;FDm21$(5)TRLH`<>cGf{(g{GD|=-+7+
zs;+Sq3mx<;5<{mXr*#4k9pvDkzmT0LNlc(bNzAN+=3PQ1BdLQN#<GY;$rkLG4%&;v
zunl%6br5%uHMgctX=fdDEa?(5g^q%zmOAJR8imq90>wrLokyckDa27sItYK!OR-n1
zx}<}4Gw2}2KV}J}QYj&UJ_cnaiv)TjxQdiO8rMnngKc8!=T=ld9~Y%9v-*(+Jk-y1
zI3A{c-it>}{jj56tDi&Vatl84lA>uM%R-TN*cp;Wv+J#ZUKBBw)Ij%uPlX1GoT^mN
zC-4kk6?AuKgsGs9CRETi;c5`9?Yk@neI>N)EP@_@rj{b;`!otw)HsSk1jX)^{DOqA
zrNs{3D{*koAIZLx{3TFqB+#tQR3?%V$YCrCX_QQ%=4&)+PVG)&*v7h(5{Ns<npRWi
z^DLM9Vn>qhAk*hCXlf~cPN7jK`6Ezl<j+|&3Y9(_1>_GnPvr{!Rj<EZG3t`^S(3tH
zX7QtvC?S6CfP#`m{9FXiBE?UH<fQz;Rx$bWQItO)fnK2NGs_=oz(f9Aiz8z4=bd=O
z<PSUQwfs3mJ&^CDe$pNeQD(ixPoHipseSGQhYIbZpQ)zaNAUz->2rH%gh`)I5Yh*p
zfOnBTUkEKbOP|j|Q%mXdEgFR?XdJ~PePE<oiTf!DVM~We>4PYlZj3m%=YPn)liVdx
zY^2ZMXcUUO9LBQXja0r{jKvV@xb%4miD4V-PD&r{AZuDpozL4wH!C^Z4m*VO4e9@t
z(9}`~9Z#cBxg=0*bkHkl6sl}+6wpC&PpZm{r|6a0D|%fLMW$zkG(s|KB9&bUP4pqC
z0a-NBB>0NdL^}CNNra7JlIRmCi9Rd}QD#Xb4R}bRX&exfMAzUElSJ&O*OF)`w&Tu8
z7Y))677fO1dyAvQHb9g@1C_|*Lbrf_g+fZ4t|sHHc$TkBx;Zq$WYWEzlS%&)T6UI6
zpM<8CGU=-{3RUkoih(<0o6mnpLfBGg2W66jdwxUqouoE_Vk49Ok4B-W&0#ECT}tJ<
z#aIlXj?1JMlNh$K?xalO4zecN)cHKOANTh5JIJF0NuQ8@-yfP<%A=!b6e@=Vij6#a
zIgLV<3yuQvNP19JV*E+3#9q<rk}TTOuvQf3HnTQTDV5MhH^9WtqK#_cFj5;OXio|x
zY#9?s_o6`hfG9zk1(G!2A&{=XQ89sZB_1(>#EyC`kPcQmuU&4KrJOq$TJ;u13p7(n
zIdmg<Q^+ChL^b_x!ry&`(Dk7aCWP+qoDlkWXxUi^-33i8h0p^u3RT%Sih&(tLg+gr
zge^ICPzX7==Re85lO!fkY=qGN(kK*(IgI5w8YNSx<3i{~Z=&XuZLB*fgt&vOsWo*z
z^AYo_FB|EwjPwYZLHj^cOBr++jY4IRK(QIii8Kn8K^z5SP&{Ms2fYe=MXF08$n>m`
zq8_vSQHhk0KktEwm_`1S!BwRE(P&NzA8Zp7KKG;Wd9Nr2nT3xu;30fA<9L|xc_SV%
z;lqx4Eqo3wRGSwBjXb`C*J)KK?O7qsuDAGkQN&nM{#*w>74j!?suDo&!!vvZ(7Qt;
zOaMLDIRW&M(6X}tx(%9I3ZT!@C{#`3DCq*|>m-CNDRxi*Ik@LxvhO5)2^1Rv^xrfJ
zMPCkM`7@1@Db#T~n{yd8r)*>0Ndd$iWKFB7^LcJReh<NAC1@|wCu9!o0ZlC>&`KJG
z${~SbBY}>lQK%f^C~PH&dr(zk{7J9GUeW532r@l2q*IVt2B~yP$e?$?bj%`yM!;F5
z42sa56hhc4CWIbDA@ojB6fz4TX~07WZNd>TA#@QQF(Jf`dM$*GgjeWK)LVIYG-Gmv
zy(KMWaVPTR^noe&$l}JmCDC?Ci%DH{Ex1_dqU8B%O3vVEzT)WZp%EsIzS%i(^x@F5
zvpD)7G_@2*pQKT!YR6I1#nG2Y2wTGJpg3}H&yUHzlf))aY{b!j(I^zLIgI5GG)ksW
z$Hmd#NetUqcTya22U*i?>U>_%e^I##??rySi$2=@a;nsjX|xM8wbV!Z(<oFL2^1TB
zbOen;r4dH~eWW>2)noisug6|7>ykv;%W$(SK}lwvq%td^limWeCW}rw4?ITdq$K@G
zv4l-yV(FVGmaY*KII~!i20X-43x~zT(j*=+vBZvgEtb{<{A&|Iei&Yvw*fv72d~F;
z(MzjR^pe58w`AJEXf&ysroq=jH5r?ziTYMN(^ojXIW)qA)AtDB<hTRv1EFPS;q-oJ
zYAKxVqEV<a$Wctf3E!iq{QNl*!j?>v!YT9H`W)QzZ)D#|#uF$u!s$O~6pHa2#_~9g
zk}1@2;q(_0!#38P6i(bh)+C%dpSL<><tm4Vi(YaCRb<En+5wtcN~V2i6e@uPij8DC
zh(@6jh@*gHV(wG50spSoV6QlJNi6Ma5KAU9GV3LkPzk+sIZT!;dg&~18mX6z#3v;a
zwvI`r@1bP6QcT^<l1UoykW5t^8Iw#kJYtfG9rao=t%UD*HOeDSDk<`rU_%aZlitE;
zuC$ZXLRWwjg%*-$si}7r{^%=zE(?t?`STDVe;jwAy*IS%EPvhwO)cfmZ8Qp1(m0Aq
z{zP}7eVT-@rNpHC$-E2A!9D*<_MK!efnp<n{+&jl*vnxozoJnxg*q;O{zzij#=4X8
zhdaocR#WG5Uw>g_WU^5hnQ~b;dhwN1ts(Ph8)#~&g7%<Ms5}xVHY(@<8imRujshx3
z_oAxE__JP-y`t77F|<p2A~WkBl~D=(a|z6rEc)k_;3rc5NV1cX2OGpB&qFAA-YBMP
zX2~NBcu1ZS4u(mdG9EF>!;X3_d0tkojs}gOIs(f?h1N)kEfc|~P#XDSun|8p=dxn7
zCS}Da1<P9`?U4$QR8N~B9zs2(HdmAQQf%cbtKJYAVY2EGLRLAhD!ntb>@2I^22Cwx
z)dy)5sz!1YldOuaDt(NEu%+6htjfHq<lvsKkbNiVP@vezs_)Y%6dgK@<rg$crclRa
z)&Gzfwz2M{tl|!`Cg{}p?DSrv7hOe_7&3k4K~qaLwF`|xrH??dQB(ai3Y9(_1=LjJ
zy+-&~z5aT|s7oShl|e+M5tLb7sl-aCt0|Z?S=80b!GWZ@N+m-nyU<chc0Gc!>mo6o
zGs`Y%z(aNo<0zTz8o?tbyVy~$W!K60Zcn(z3dZp}j<rU96#fy*jqsgP7tyskMRbu!
z-jZwg6!S^JwFx321XoI9CATidHokJ}!q5nlTaP8=7JuV2_H7{cX#n%5M6L-9I;*U!
zp{b?HdOwXq)lQCLP+9R$iQGYA*ivsoW_i3qn|Iy1tf$7-`K`)dCtX6)?^k$mqw3B4
zb%`I-C=~q)6dU^eFB*kPA&vs{oA|l}{-V!GuUK`7f>#(Qm@Zgm0#@mjAmAprAqxS|
zyE=t{Q;MUcU$hgW-^Y-CCq;NN)2}q(LBC6IjEsKwz#~S#?5NlDd#LuUV<)mr+fk+2
z^`_Ry$Co)PMc%J-DEL$mY2;Kzn}_2WzO;F8XoS(`ZxghseYn<+H%|#II`ig<(A1JQ
z&!SPNqQ_ATyqWkgbwGk-CQgBq>GgxW^VZckHN3M{I6MgPCej}yYhJ=bYr~rFpiwB+
z6eu>V`2iY*N+6Dc7p*I_N2Ktt`gHV)QI{Cgw5>)jYG!^^NtEEn17Xlv`0*>yj=VsP
z)I7<C*eb?`zeP4&DO{hK4W$7OHvAcmh_T_rc*NL{9UZ%1i!H9M9LN5BJp1<v?B6G{
zf1kwu{Zjllj_eonMG`Hpbm}XWM<qXk?nH_!Q?5Xh=DoSm^l^3rMLu<PNRg%%iYcGQ
z6MdQT_n{HSlu!CI<(xND-6CIFKMR^#GUX053RUPhiVagPB|&<^l)UrSRX8=gPRqMT
zll~x?@<<+98>T#sMxmHepx7|wIvRyaAdV7W-o?M_)6pwNU1G|84A+*1N@pfal}ic2
zd>pDs7Q(y)d`1$csRByg#Ktk+d=h!{e?&>k%$w4H2XDR`2gZ2wJ$S@;lO6S%H;<So
zT$ry07vx*D4MDZ(#F=SdcaEC&=F4s4_LA$ecY|}q<z)P9#gq5oNxnS!(a;Fv$)^)M
z$v$qw={#EG%<HgU3hg;xPksTKTC(K7(I`}r<0uA}jD5xDKS+qo9LZ2JT}$$gTUX!o
zz^+{3ay9uk(jO#8{+S0hnm*<=*adH)a+(xb0>y?SUqYi$>BCWgBez2LsmX|c*C(S_
zoVvu4rtka4YLS^CRW2nM@-`TE7KS_$LO?QPyzWVM#D+0;d>YyD4iSLN>?jR*u;ZCH
zD8`Pj!Xw6x?5Gp$*gm7s9s4hU{99aEcoCI`1Jhg9r)H+-jEonW&6z+iVwE*upt3$3
zKp`9eJIXqM2k;7ccb+w+!DVlpdHu{lX-;V_Bey;9KbFFWQucy>_JM!;;h$ya%}mb=
zs_^W8(44`$e-DF+>HSWTkH$5QZH!G$zzM+dY#h0AHK$$lE}NO&L;KG`L&F$@vp#A(
zT6q)@qg@@S{7U=V^bSU&jK=5pskYOr)~((P-{&%v>t|i<U$Z>7x)?M^;K{zRe6wB{
z35IgvDY^bNYrp|$2Ji9kAgKzo2_m9bn2JrMyEN9QP|}@FR&B(a9B!X=v2V(643U_S
z_$vyD$91P|tw)6o6Gj>RqZ;hHLu)1(F;Mw4Sgt$+|B58B?9%Vx2bc_|E=5-DFGLjZ
zH*~}pQq`YZQu;F=XU_uom$5(3w(n1ON__=KX0GMRDT!KsJ~Xnf<)yimejGF+i9*hf
zub8RqC;uL+(|^S}-4vcQ*6D%L(Q@KVUw=X{TCN7ULXJ;!!<@@aHp|tqT&oo1iiK8z
zpH~b<YV0|V!Q5K-f&X!&R&5o^a9z+0Ho{XL<GEVBRjySF<9+o)qX7GPf<_bU6k55F
zLNzxWfMxg=XseckhNI=N$p-eL9M+&gnZN(oomBbz4dHcJi=%ZwFTN17(wfrlY!W|*
z0sE>7*qe4##f{hS%2nE*rVoM5?S=6?e5rIpZYT%aW%^En2WX0AI4TcMw$P&uhyuiH
zD5p*pw=~|WTk4Db3^P>!lN#86xD2dv4Pc0|Z7~PN_ETsLiFl(9&0bL~szU5_#g99I
z-GE>m12F;TZVg9ppvw2Xt$hW~usd0~=~<Vahw-0<KKcvu5t%TRyOdv98}8Tk+<}GR
zO4WwN;voy1OJi+RZfb?&Ozlg#kcRMh@$y_T7%fbWw{l12R`!L{SbNIebYA!x*Ian1
z+YQqyNV`o#!e~N-^m!PhEip)%BNj}*h`(gw{K;A?n0e<+SWz!Iv04}&hgul~^=2JE
zc3h|yb9K;ytc2DJtrC_z^>d|JtBMMk8!bc2fnacKFjs8md08C96PD#*jziHbj5mTp
zaSEnsp$Mi~Th^%1gxXmwkF<EvR2lz|(Vv~Bd)^O~0xL45z#L40<<N9;n9P~VerTT8
zcGgCvO(j{|2cpG2h<)3ESNP?*kx~u!ri{Yt;BxEs#7}|jxqQ#jI{e>xgMF+`uHFbn
z%1z+g+(fMyEYDp~3V^d4m~6naIc$STgpEvOkoa_isLqMMD?h&6#g95I%tKecl({lW
z3rm5-=)p=&uq=7zW1FwXK|Z$mhUM@NjuZaxyg_aK;I&R27Sw|Mw4s)j*}?!klcA*w
zLd9iFVLh9V5g20;pp#i69oB|N&6;#rq)%Wt(L^t4_t+uHN1>p#`i_MHCoVj5wNaMG
z<AY;C6(1Q6@Dj8%IW`8uW)!aJgl=p#YU5m+mO*ni%dIJFDcZnf;39IM5)nR;JLw|N
zBY#uwy~@SCdX6tZ4`0Sd9+5uU(?byZ!~-_(b^;VRia0hoy0&W`hM{!FxC+H>E}jW9
zy$|DX3yXtwrn5FYcGj8B;(dbXBFNB)W?!GYPnJ39>j7L*;j68b&89KSO}D>9vYUQV
zj{d5PqxGDB5&HT|%-7Mno|5xFbOIDPk2p3tf1+#7w|6ikd55eiXaCY=>|s`Kjh_Bb
z=4tD!W^H(Mth1U$v|U!`@yX)hpfHhd28G5*2~zqDamBxGxK<lKukWmw27z;7ae$p$
zoNVB-LU^69Ja^nwcm;7%VZ0d_lX_tcQ|f`l-<5~Ayap1*R`KiUybU_}dFJG3I&T{t
zrtCxf&#DXCIdwrKI3n95_?ixQ88X|~j2i7ym(hkPy)Als2{cX9*P0|@ZFq#NQ<_D!
zT}sbXCnxgav$M_M{7Kv=-qiCN_a*n8J~=TA${K_QKNWIXo(gKa5x|?8t2x-<#DvKh
z2$L$Pb9|BW@;-RU1Z+NUje|_7!oS!G7zmq<<Kq);=EC8|d9b6YqcRcAKdJy7?Ghk;
zMl8e_9m!%8oe|rFGXfnX`w{!A>cDAE9T4*Yac$;9try{e3G{;FDu2Gt#h>Af+YY0-
zjz!a&L|7Z10oF5)Wlg&ocVw$p&)4-S#;yg<>*IgNS=1~{)W?A=A{4Q5aD{PV%wg?h
z28#bup1Rn@Q+n!ekIuS~IV+m_3&Ygs9hcCFzpGBX*{KsE{Sn(H{r9qU9Hgw(xvJmq
zcj<SSW;>wgu4A6FPBYeqd)qqAScKZ8+41NVTrdx##VR@fYuh9{U}bKgT7yeXXuGU^
zx^0j#XX&;<TXFcTi^ueI+Yz1iY38(Oy2Wa$5cILXs*ru%sRJU_5Z5Nv4sdrLB%Na`
zHI|>aj3rFJMd-;NGf!HlA8W(CZ=HTDV(roo7Hx}VelrX{<^bu$o}kN@ixF{gD%`|b
zhj|Ir5ALerub?sTFn~2=IBbG+KuH1_i4WjK3gS9g>cVV@t#$0u7Fi}Oafb5!pIm&e
zC-P3{{->DxqlvsjI1|;dWtWV<sm{!PE99rOq8CYxeXvRDL$bRbk~?E-HMT`AV+)h~
zCFtetp{av-W^H(6tdpEYvt5$Ex=^o=PvuA84sRZp_r`J6POG?yYeIHDtZl;`VR?Qe
zrPnql%~ydt7?8r;G5x0edytEN^+etoU3?&OaWs+Rmwlon$N#LlaI8}oMEWAKP5Q3T
zt_JP>3z_TQP>twJml1_Yy9;{s4CYbm%FNpEC|D;gi(I>;J+oQZ7~~s46Yg9_ig6-a
z@kDiv2A6rPNhD$0uUl-b!^l*P{!zKJ?&3~8M|VY!jx&#X=I9%ox*&2Ck!^DHU@b@8
zyb5`mt+8_M+gwH#=IUbf>@@SNb*{2DJQ~)y$|Be<SI-eSmuGj8waN-N!_Mnl3p!!H
z{lZ(q`|SsJo#AFP$cAu{&6u>O*Q(|!yR{A%=4|AT%B^>~xK+>D-O#glFwaJFR%u!s
zp2Uy+RdwKVP8|>#i?}u!dsGLU3%NTR&1jbg-;VXUv|xHJ47OGs(;q3=CK6TYhy`J!
zHG>bu4AQy!BRqR?1$!|$4BcDqlrhTn@nMkJ9!{y<G4TJ&u<34U3x2E(PZknHfAEDR
zSeq$z_MBPpIAIl^*nsyxj=+ucd^lMIDUb@L7%(M{w_})Wln<}Ti8k2RclsG?PaMkO
zb4O)(0Itx2<v@6hqdW#{g5`Pu>ly**_zToRxn{B$z`Y2(2&l5_fl_UvhIcl_O-=-?
z*T%~uQ~XXTd(Z^`*a$8tG>ZI>eFNpeU~qY^%&sG*awQ}&JP8(6#AO6@zG1$&T|#?C
zW@6fDDp8(yO3!Q5gGfs-SN_U!C0ZonDl1idZT&XLdDm+PL~dIq^yev;Ys{tv(wrO8
ze{PMGH+Od#N;vKIKv(VxO<~Hb$AoVCXG&8jRsS^oRpHt{Q@EC@aM2Y{tCB3le74YS
zw8sP43~N&*r%%JXZg`JkD2K9mxq7l!%Xc^@_nK2rJ&oP@2&<^2KHj+<f}AAnPf#8{
z#l^$A2H6{(d?K3@^z);-_;ed@Ou5=OR>M2TWq9Fy-v`Hb78{#2&9XM!AJz%PeAq65
zUX95Eivraqy!H>D1|J^}#@VZv#j~R|`Kb6>;3j?^h!K8Xj(y1s7in2~g{fs+?;5a0
zl8I&CKt<w0mq_S|wl4;wWszvho&A9N&8d+Z&ebl%2@|Uy$8v>RVzD-S{ML!ZeAX_p
zKFY#9R-epI1QWH!RDQBmhNojdX4R_Vu`;b%p?_ZbidX0OxjF4zR<vU2#kMxUjoW&y
zf#&c^3L5u`#sl!qaQyFO@(D{rg7!uK$5Jk~KD*py05WIY8mU0v<`QT<ZwD~eAF{~X
z`U~x1>5>;7Sh@9H7q^D<XBqnR({A&Jwc&BFo<A&hYW~pew1@xMvzzMS<<f@m8TZ9Y
zm*N8r@Mr{X;2ECE!NUmksbXy;w{)qD2YC*69Sr{`F#IO*Jhj+tR;s|cvMVUWOG{2$
zJqHA=$8N_}ruY4yA}`>y9gu~V*jWR+9{1Jv*BMS8<VItgA63u!cChr#k8`CRc0(t!
zpYAW1Y<xeRJWFNb-^ffHs90^G`*Uc7Ir=DYbS1LoGTgmm_xii7-%ed}k^9Z(LsLMv
z*@d|)jY91==P0Jv6~JPQwz@JvLg*X3rWZ7V(PC`^He$_GHthoY1n0pYT2u94rjmGT
zEOV#z>k4o$AnqUgF-!^G@%A%X?$w+J8)o73)WFW^2PW+oG-c!TlSq${TVIdkp|yEo
z!CD%HdZ~{<u^G!SjY1_5M*#&U9H=HF{-Q58c*Uwa`&g%UB?_5&f2&HQ#QxT&pm<pC
zZ{-_X*%sF;z)|vD1;%~C)3VyhmlR-&*yh)xkW!x(<qPV@a@Zs*Su15fJX#X_`=tT7
zk24xGvA=&oNDHPt{x{%g*h>m-#3S~S0(O*jk3XZ34(#ziRPFhUzP`th1G#XBz<iMx
z!5U5;REE9TPv3>gA8Cj%;HSZ*f&uk2)uj6uJfZCxpCt=^GBm<i@c1MP!jl@JR7Q(f
zxRY_#kVYQg2<@46+mkH#AT+gP!G~xRs)%tE6AR)%m3MzhLS$z_M9IX0yyMYXs&^yC
zf~kSs@_>~N6u9L(sAMB4@Od8BX!aZ>74)p5G^cMzqfitmP;4l035`N!4@UtCMDM9&
z!9VL$(JM+_;=d&+<;={1Dvc5xcsC3>3kMzt?x#6WuXmCMu~m!*k4GN7N4P#S4@v_b
zJos815#zyg@QCpsJL)wL9#fWAdin<S$Y)3lSrK8xgVU}BO%UGPxTD!>@;dQ+aJXPe
za~m~Dn|P`(U)Dk+j4w~|<;%-Ld(M2h8Jb%1<+U^lRq;4VI$z#ILg+;0cGZ}fFL}qU
zD{*RIon8ZbKj{sUF7M-EwV}&z(<l^O3KSc<{27fxB@agdx{Q3~3;(K5M6Vcii7f{V
zb#c&Gl+3)T(ka246&QXN-h3X~(Y$G{fRaJ6eT+d*K?WTc;mFLO(trnpF1l7S=uUXV
z7?d6Lnn6!wH!e(pce}7WB8^7d;&^lEE`;eMp&kJK7lcYQRIGX+p6|=5`-esttDfP@
zs>g)(oLO}>G__>a(`gi{FmjZ1Ry~)5=mD$pj$2pk)WABi>L$_~B&%-ZVYOk^t7sI8
zRRxL-tG=5?p^}HA#90;ps!v3(7<GwN4=}N6`|@O_R+Uo;YTXxnkcC?BgLX8v617nB
zD;kOM>lw(e10pb)`BfV5;Mect0NEAl5AcZbD?92nzaFN)A;;;$bpJF))hv5+Xym;)
zhQ-?d#u)S4m61~wPd+ZPKn5}Tl#=CY?QcROj3-~6;7RQjBzAW#pA9WKGvzbT)RHOt
z-Wjb;(ZQ32o1++*GVxBKok@_)Tq$ret=aO<TUX-L@NTtY(B(zzhmsy4-M$K%LU3*P
z@&p=%;!A;I!<T2$C=_3E6yQtdJ~bur@0l3W6sImR<{l=-)GM2rBUKL*9Qhbb+bkTp
z5xhlnWTf&*R>Wp8R(v(G;x9$v$;^t<fGnudS!Q+r@){fxW5u`P5o1Mm)N58;quzRz
zXD@1k7rr@><Ek`rOagdwr0I2A2CDopxLQ!9sfC)Jx8aGtSC$_NjWEVs=g*k;h8CR}
z^Hb2&k}<zVqfq6Jqu4O!k4TW7FedMObS~>1iQVN)4X@Mn<x`|TNWOfMht`HKx0s>w
zn!H6WP;B^eTN;H*AdV8hzQn)k)6pwNU1H09O>Ai>b!O63$&?_?8)5KSNOKiLfhJ8;
z4V2u8tz+D|4!QFt5sJ**DGhjV=P5Wc#+@(2BgUQVsMp*%6#Wj38*dJ#@n+PxH)qB_
zO=IB80=QT3W&CW#l_PkPFIVP6BaABx39fYffLAND=ggFKXllunm(nOyq2nk9ri^{S
z>+K|jPERk;y-nV6>k6D6Sho*&-A(#~<jFgEU~PEv^E3)|T`5p(c=DSx3Y9(_1+Ock
zAMnDz>yyzdPF-TkJxxp*D{f|%RLPWJ$#Y=fSy=K(Xh*YTyz)tg#FjCJEFeRkD?*T&
zA*BHihMe;*$&hpLh%qEP>NP_irhnDK>9R6)%~iAP&5bXL7)#!%+#S3sh%s`iV#Pi1
z3}04U92#M)IF?{V^{x(hR)xR8!*1<$MQGWXAD2T@OMYBKqfiBoqoniWsU$=X_|eKe
z@7i_?cr6C}55FC^xU!xc1W9TFE=p<_hfla(tUW1OrBNtSvl<j1%f&QGrkKyhH|S_1
zc{7P&8}$2d*AQ+aYQZ-V0(dl<!(>`*=MJ)F+ti8ew03Yi=?*e|Zh@vSnWO1ren9#&
zGzyhI0>wrXJxHTa>BCXtYX|sOy%2lFs7s2-xYDjukXa6?EK10sonX{i<j`-S9W93<
zd?&RKHj8PYG1Njk3+HFnLehYT7Wz95iLDlF@ouSw*io;w&>`v*RZdE1S=vHDXxCc@
zEznFQRnSYopF#y`Cn^oJ3;yn_ffj{Em<F04G*I{!Gdm5mUufA`1MLe<Ej7^LGzwMF
zIEqOFMR(1fL_*loVF!23I=JT?vhO5!36v;znWayOMxnUNVJsVIluV(HOP|X~4BJ?D
zQu=TQS<`CjeC8wOtC2Q};bzh!WCq;`O)b^WJv0iHK?22QEceqWR0eSrP(SgE!5{Q0
z>=mgl3839gOG1i!%<@NNQbPW02V>46e;$K&wEWSCPKqCF6caxaD1NpVj?XN9qyZ1{
z^A{Wt6F+~&BeoF4j(RPAj*NeY(MbXwm?nW@*1d&L;v<c`K1Fs-zxX|&97>$7M9~g-
zmaiyU7#d-ssM$GDw0CIPSrp}<sii1dMWax4j-#ZDqGL!1TcYfsC~|PmTC(pXr3n-p
zQ8Y}WP?Y8{mO71+Db#U_djpAK8|zMrBJLn-s!g5GbNlhjt1d5&xt{b1nM2n>Q%hNN
zCyheokU+7KMfcJuR1R?zkVVpisuJT*dL{OXR+og)UM67_CpWV^QaP27M}08zEb{0P
zXh+MV1o26cgiT{2sfi+KYY~LZB1szX5J`W;VKI^PG#)XL#EyC`k`5`>^6cAO@XhiR
z#S_^$pT0sQ?RpEN1z{J(8$BYD=tb9sl1MvIO~GyOcV98IAT+|n&{RSUnO<RSw?wog
zH0`X0c7>*vYG^r)LX|g;Vo*b~@ioHQEt4S<!<Hfwa)`rZx@F?vpwr3DlUycHZ1m8%
zGz!IK4r8g%DA|G?yFYdziD4V;PU<1<AZvzAozipIhhQ9DBJ^(3C1eU+3r#KM&@D6y
zl|ll=Mh<;~Mxj!OqktR|4pgNWf6+^^SFF0EhRn-EA(xqTkjkZm4w?;P&Z2`Jf_AhH
z(%4QaAZ!#<KvSrI<_gDWRzT8#hXVQ^91v4LPvQ|%K<ud33g}3DYkgs4BpAmpd%z3I
zopey@GEvOBw;Ey}$=14S(nNFK8)~A&>1sO8!?S#q(N>`mri?C0C?ihad@U<Vk>*$I
z?Hn3()<Zi&Q%gOxFO5RgH;!VVhYlt&Y$>vndWd)3y6mRL)|n#DBwa#M<mo)P(G)U2
zMzWqpp(s+I*idAZMxj!OqnIcXf6?cpSFE~3k;Vmsc$qU3rOK!TQ9ci>kcB9(1&`50
znJ9shF0pBhE-yj4+(HB)GhIpp9&~vJ4vW#{op{9Pk{$g+(B*JYn8-JSLSv-V1-jI%
zcZMz_29pn&d<o;tUvI#2ML%89<yS;j$e>1d`CGo?;DOKxqsz+^beVl2;YXoqXX5-G
zG_@qo$7mF)>Twhc;`}{{(F@{qaM0}cMJsqTO-PCpC{c<tKW(x-jY3hJ!&q`OO15Cf
z77_+Y4BKFLav^~`$eMXmr?j(5I+b(@nL;N+Q<%xo6f&!%b7&MQg#?O?N-EJPR0?qv
zlS;y0GN~j}tX^o9q_c}zC8=H}RMLG=8ndXRY4DhZO49o#RT4IhsiezMC4F90>C7rg
z8jyq<opDyLxcCqbi>ahr@QA4-cGPQ?bPRk=CtsUv)hAn^lg^(Es_>y5H>s5RYGM<F
zw_e)OY&9vGJ_imLqRHGwDW@;sslE#hpAC&L<@A<>a>~3>>ARt2XW{fsXlf~(9-&dF
zLda1J!YQ_q_je?OEtMt~92iQb`!5ded7kV$$$0`L%6VqlwDtR`Oe9}U;V_n6X_QQ%
zjz5nwKw{X&x|6brJ1APj^?t<OZIU{lXC3Zv@8u-Y9c20(2Td&{(^?vZl1u`{Mluc4
zC{+4z6p&2dJXL@3uX_FUicy#3lJSkyMlLdICY4SJ&GacKEm<_v72q~nGnt7`3MXtI
z6HaeI;q)0%>M{!_X~07`-GF0b!s$jlV#0|X^;$TsY}M-dx|>qkC*{GFL1xoi8qE=A
zl3M7~;6tH>#7S!M{R{r)D}Fv18e!t+9SQNX@C@<o-F2hmwSv<=?Qev(o#oGi(9}}?
zJVc{VMUA5v<WJ%qnZG1KbTV>!lZnqeG96s>4B2~<!vu<r1e$w2m5-zZau~~wG)ks$
z$0g97B!+FYJ1K#<gRJQ_byjb+(&>e$$B+&o6KFLwwG=_8(<qb(5-2tz=v*3wN+6B`
zB8a(9Rbu?RUWvWp)Fm0Ti|Jup&O2t^qjD&rd+vsMl12Ay20zidN06KpI@lm4bl!nN
z=N?heG7BAPz(eS~2M5E1&UJXigbq9UhY&g&x*&8mbVlfanWWJ982C^ZIvcdBjF01Q
zzC!1&&<GPc?@0)q%&R<K4lO&&ocp1vrOf#rjY5?zj$)8Gu~nX*lMuGF*TGdD2lqTh
z_MK!Xfnp<Zwzz@HL{j27jAdIIB~z&55@$CO!#38PlsMc$*3_9gpWVLGIYjz{%%4M{
zsind>nMR=$jzF<dIIp2msQlq5pm3t!>BK+l71%3EFSNkfFk==tDuWUN=XR(jSp?1|
z@RNnW*)Sso4mOAhocEx>xl@$0%mPOm@DMofz`-zqb1fb*fy0h^EpSdO6pQ6nxmGQV
z=bI>DTeS^AHD4||-Umzlwpd#nZy~e`VLB;}?g0M_aYQs!*CQXn^L+)=ZJ`k+m~Kc2
zrX9kSp;ODv);fHNpUYePJ|Eh5mQD9UQ%l+OO&W!&g&f5on`Yy9HGTRM62q2D6S|4R
zWV$^j92Au_>zMAwCH!?-=bs=uPfD7$4vOi@zmgD{{geIfq0v1&chNoPY3|wfMrtZq
zxTj*n*t^jv=~)_kF2{V2az7e{QXj%&He)%QMoAxwX~xAK?K+9Xu+6AWs#oqHYh9FD
zT(?{&YpIRTi=;nD^|~IKTB_FujY8#*K(QIiW*UXcAC7|R6}+eD5dW;xp;wf;q+X3*
z_-ii`%!*cJRYK9;3^h25qP+!tM=M%Fkw}$`Mq(=W22{DX2x7>ra-{(emHRmyAXB-Y
z$0Me4*-@`m?$L#MeS9iEQYy6a_~EMYLd$U{{XuDWFAWIZ%612%(WIvRA$VJ8T4NJ6
zeSeH+`l{LQhenv1{ZK;9W`8{X*P&@=A^R9KwG^^X(<oF~<R~UujqS7F@&lBoZD}<r
zWEo5*A?x6voyg9U94Jtta@nk*_NGxN4s;mHDjFqQuwxqP7!t!a*qzi++(FjtoI0gv
zuWd{Q4v%`hj`RqbLFYhIOBFRvqfi+nP;6AxCK`pxAdUhmiu+F$V*Eib#9oo=l8)*(
z-R3k=l378ioJuIDZ$kmjqM&XBuh9z1$bV8hVe^=F`VeZT?+QN1tevC*5AF1692(P3
zpTQ%ho!C*YwbRRjYGHUh$QOf=S~19%tD`{!AC%8Gg68CS%kAFinlu?jgLx~eC297P
z;_5pP4k516TC3^(Z`jCJXMHO)!gSUh9n@KmhNhi$)+5l=QfEC$qfmvDqgd#yXGsiO
z>g}Y?a&XXgH&JOqvZO$<(OFAq6pAGs#<HA7$rkLG&Ke>yY=hlNoy8qw&EBa~+Ibt|
z*`!Oz6j}>SErr%7jY6f6K(P^8EgFSNA&vq<EAc)f{6#OtUa{(u(pqIwS}9az7F{aA
z5~AyCP${#BuIs>owCGCXMXA5gQcQo{f%@wkqK0SIU($ew{`we>lIgE|@QCRzcGPSA
zbxbR$HfxQ1z0fF(j|Xs@w!F!4@hWwbq6xxVYwc*Znp9X{2ZsxVWp1N%)xYAYzPjqG
zp%JF5K9<l`nIB*KS!mf=Ry_nwEoIejX%wnVaukECirr@X6A58Uv<X#}{Whb6d-^^|
z<p{}$0>wsD?M$OkjOZ|yeQA_Tp^iVkb})%y8|zMrD()a_qW)@_tZnAA(?`_KB;7%#
z&*{+AQc$g@QK<A0C^mwsN~2Ke!%^@S{z~okAO2OZzg{uwl9)1XQZ#arSwg9FN=T?L
zLZ)SrP}hRnXbEK|J}I2AeM~rg429E|MdoJ~PSSvfaJmD>#)Q+Ic*KMgJL<J?I#~*6
zc*3Yr9-eFkMYVe6wrsW9cG=1V<t?OkqXuHy0PrOUgb-A8TP3W%f=zsd)dQgsR){~H
z5LVf54E`uI?X0Z62Td)N)nhaYRWCV;L0QFL!2Np?!<KFn(#qon+zt+!eKVCNBsmHc
z8*Q~cjY5&5!&q`OO15Cf7OVzI4BKFLQd@BcS@U-4ly+XQI+b(@nL;N+Q%iky4vj*k
zkU+7~S0x&SN+FJ7TCl=j^iu2<t1fA*71V;2kp#_JOQlyrYuyKRGK<!l2KUigi{?bB
zxzJ8bbA1{$*XKq3&aAnl0T0dfAsi#qT({s6(_HMR*P82$qAB=!d__pP+H4ivU$Aw2
zn)I?j^j2VdS%D;1wmt{Z5kkz$gwkVQz?Qyx?6aW}rpG>)&|{ewy1pA)c9vt`gr=5q
z>=7D;Dz6;HAje`0UB4qCYzaD{#<DMTIk@L}vhO653KSbLw)KaoOeCM$cNojAG)ksW
z#}~Q=NDSLpcT$XT2U!z*>U?%u=sJmX2bn&{K~qa1ww6YrgqT3F5n{tM3Y9(_CBD#w
zf7R=+SB$zO!VWWuFdL>aYcZ8w2`%<1s5Du$*cISKT8mlHqLgH4E+)x7hm!0wq7Y`5
zWYU0#B)b8J$|Tv1c*G<bJL<J0I~ZU3zj1tg!fnYb?fWW&(yX_*ny)M+71XD}nL<IS
z^OQ{b7yQ{*CVetA!er8y5;Dp4>nPs{4LYl%2cfB@I(mpkp=usSF{q>XV%0B63|q=f
z$Rm%%D&BSLa-14l=k<#%Z=q6-q|N7faHFcr{NC^FXcUSz1&R%AE}>DV6yhkR^$Yw(
zpOap(>JoMCW}?op4w?B;rBZ?)?}i~~;m6~^|1>|Uq9;iZ+r&ulOGtwE2)Ab@L21B)
z1Ye8eVI+7C9x)PRN4+M&6OjO6>zD77G^wvSXp7@ble-Y6le~F8_+RiQ(NGa-6VLZ0
z(pqSQ5$V?xM4EZM;j+-OGox;Xrk0F)Esa8zLylr#)Yy8%O(aBSLe0M3;NYH5k$oqb
zPoPAZ&%7(_%QOnbd=6vz9*vSI)baI(pOYB2vF_x019y-$A*aq~r}c(sNq3Ox^9(eF
zIUG$NvvBJBFjdIN<tl+<Bb;`oQK<CcDDm|M{HtDny<*fQ(PZ3L-=3AsDoW*4LPb@;
z2U%3qY6u3cq6i9<vI>pFWYyPER*j3mWR_LZfQPJl6%LTes<n8;WEDH=wXAw6i~d-B
zGCvVa)EZOy$yRy1ys3aMlyX~cI5O=v4}s(@trnBUlj3RvL_vrvvZ<0+6WGF6UX?>5
zOkRB_A+J2%FaCzmwzI(61Whdk)-;Vml}(Of5Lk)#i@%Qqv8CO_nuX8%#T{I9H`#lV
zB?U^9CCwu1^E3)2vK+?pO&TRrxZ@)0CnSb#v^y!XxPz<-JatyjUNLyYAr4#b|3rF(
z%%G>Bsin@EeJfSbNS!56Y;@N4GzyhL90hb1_n)f8_=8@Hy&}~msdb=feTrf!v)EF(
zl@MD+@In@`btpuH7F#4GO1Xt*Vsh&{D7QvMa5Bp+X~08norXhXa_e+FVseWe^;&Kn
z-YjgC`|cgz&$&EJU+Jd31=ZG3Ye@wafMbP%ik_>c=@_2kE15<@BTO>=Fd><;KaqT4
zXxdpZwV<h`V!DDxp$Z^JF({_^O`aJN!<Ir5g3041kAs75Cp%BFoj|eCOP`@pD81w`
zmIrB+Y{8B_k^BP^!#3ER)Jxn!)-0SlrJe8S{ho9QnL<xMQ%k}0JdHx7kU+5!Ok3YZ
zRZWYrnC|EyzFvyGV$~(Zlrt%&2o0Gfl1it9M0y<zJBvhG0WqK@Qk3(gCc<_xP4q+5
zMC*n7GixGgz(W(g6vxCg(aCtkG!Z-MxhCSTA!@?@_*Thn%_&tA>8AZPQPf&e6Xn6N
zQWHhaRhp=Pr}%23^Fkv`6a6%yiPT*>*f=L&of~Dzq453JL)z}pMrhet5>=t8r6k%+
zqfji#Q4Er3D;TLZkhhW$wj`NQL<}X<yBi(cb2Hg@k^uxtG}Pt=q<d%-ip?CxazBld
zDb(==q;HcLwz2M{AmR?PCfU^a%twr+Wx(nU>c1mBLT1pfp{b=FdWJ@!GDx7<=%Kl{
zQx(%<EVzIalh6pSS7EP6y^wM!!YXDtq>?Eiht7s!XOTnuK?qXhP^4p04q>~P9QrBB
zp>u@$Gs_`qz(Woli(_JP=r}xLa)=%ETn^>)^;)w9yRFKte12eAx+0ngTBTZX+066-
zX-h;A)82Y$VZ_|PGQ&pp>1`tCNX-%ZtCC0SAQt+>ER^<m{mejVDQK9zu=vei0I^yC
z{|fZ7E<btx?0jgwxN=r#gvp~{pgel?ur0P&_Qn}BHfMTE(btG^lfhp=t2PF^H0zg|
zE`p*yG*5dq=)|-?6a0~7+Ft_g=xHB|qnP*8d!-Hn0uU<<zrez<RzxmyVUPyM!Z4Rl
zn3+fpRn}l?y$6TPs={@6#Hs>2>O@s&&obzX{Q)0b(0ogD21FYTOmA7Anwg$6GG1sl
zO#}FFIDlKj0kEU219)IQ^k;XTQ>8&Ug-UbTb&ftc_4zYBFQ~#BJA&p6PTGgT#Pt4W
z7Dijgj+e)(LF3rQ*yIG90W08aAh~iir<E?KWcq&&np(z+7we^VSyg!yFr%FvsQgO%
z+w_iRt4w0e?^BJZSFKyUS1Y)%HI(aTo$X(<Jh!?SG)LeyZe#goy)Y6C<-${P{cF~U
z3ocxT;c-&+?Sb2)(X~=@Uyn747DZA4A>QP0G>7y+$;vqoMhwaKxe$p}&d0EFK0aS_
z#@1Sx24yvm{!tC~;n14tVsfDJXRus(2L2V9h_Xw+gCE*-={fjG2DkEs+u=M28@SsX
zQ`MpW##6N{i8|7fT}(de4jHhanXA3(M55Y%1_o(e?Mrhj{aE+ViK5iA<11z=`^mq@
z3iV@HsP7e?G#2WC($O+H(+3_`Xa>!p+)#aLs5V>~THk6Es?E__V<KoY2iG<VAakl?
zVL@2WH3Y<Ro@l6D({FP}G<<3mV0ysGe(4d}zsgI{CeNNFbl0?w>=NrpERRaR;Z=HW
zed+)hbF@r@H%kwL2N!{j$>C^AeP(KLpI*=iMvJxo*W8&0%25;zJolYI2v-RCcmYBZ
z2rmZ-M<59amt5fxL4>eu_RV{{n|-^>9yvG^M2^TJ2rmkvoN@^P5fDKU5V-_I6h1^m
zKtM!9Kt)szzpCz@>7JS1n(moh-^U-tWZ!gEcXfBwZ+d2WdcjwH`R$-wV>tX!TU&Pe
z8iUaQTaa+plnc2`uGY7X{dsk|QtVro{de}m>_@CivmZC2T8MRfHanTk{nxOY(c@J3
zzygcZ0;7ay>MN8nuxMYWJ6Ci%(j7e7>mqMQt<;ez6}xji^@`I`fn3p%=E=efGfCCR
zIIyRvrJL89_k!BMKcYvJ2Cj4mQDf4SDIMiXDdSYDPFKf@wH>fqLV0ahDbq1!iW?-!
zb9mK2{ee)kDlsZVaLneFstEso2*El@f?iyz*UI%8YzhxUr%SUKtL?UQP3+7CN!#~@
z9o^ZTVH1Ho?z-3Qndm&eiB2cW;=U}A$1dmdTvtMljRH9~40A-*Q^MxbN3C0+J<Zds
z%c<l;J8PjR#A}l^m-A>V)T$k~B15wv#myUm1=>qV6TV2OhQE_1s^Kr6aDf)n9uInx
z@)qvdK-QfEGGWDJf%b>A6>53SqAicnR$_H^aJlU{(uQeYCtaWoceXJXN6zO7(`vot
zS(vs`B5lo@#!<#-zP(R6h#XMwK~b31{sFbn%{l`YD%djkJAQuhr#&)u2rDMBV?1qz
z8c^H{z@OHgGDhQhD#s`4$BjnHi;1YqT8u=sbfw3s^))t4^_IHo1*bZTUlYxO->YzG
z4yWY0`Wn+r6QBZm5Q_5SpgNG;<%kObeg;}^4EUVv8RO4~j&YZB^jAs^U21d}bFny9
z@U;<8LC%ODArU#EGqjfiO;^}ZlUVU^6}UOy-<x8^W26Tfew^mZHiux-*s{%`c*K@%
z*iqb755^%aSoK&W=%hDSOck7BPw)n}Zjf|%O$DZ!t6p&H&UTqfT1bpk<3p7eq0C2A
zUEpa!RGI>EvUcNnNiVurcuE*h4H^cc&9_P;?|9>vw?%#ZK~aCld$p(88mOvJ)DWo7
zqODLtkz3IL6|_`|?iZ6LOp|OFsL+b>9d9+qTDa$WvhHO56IT3Pt-s@aCvAlSDvP!}
zNLz{3)j^<onzUis*R2F9?jV2YR_o!e>WPKM$oV{I*LfEF^bY9{a{RmnMGg7sW7-Ne
zeuNbhej0rf6)z3i0{n#DlMxt=s}VR+FSQAmb}-CsdRmghDMU-lq49BOX%=)t1TAUO
zo(oG@Isi+9h5;<4L|4RyC09WLSlS<3#$f3HJYulKj{e`kQhH6hVac=5I#^OvhL@jC
z08a;DNfwY(^+Y@`DJ&iDDPgcQqIIxzlBd}kmU^M6AuN4?wnAl4ZY3I)&L>T%iOaY6
z9=ZAy8<s5Ga}`;4GC>F{Ca`oXZG}p97Hzqgwi2tWgRt~CX~VRyTM0|tLB?q~ay}29
zGk13jT6&Xo2swaWhoXkG^Z{*!8bHE|2`vr1nF^Q&Z2?+h?vpVXt*bFOQ9tzyF3HTI
z$0a$SLb!AcG(HY4&47L|#U)ui7niVf0GCD}E*&enA~r6$3KHPb-q<pROH1&G;SxLQ
zZ(Lf8d$Lz>$vIbZDw-WTbV%)6T#CM77dDX?m^RVr3=`95!P)K&J532WT|b8>CPk*B
zJtYj8#<dQaR(hJPk*NoY8Y0tL+6t9Nxs_;SI)^l2nnqi=zGUH^%gDNuSwdJbL8cpN
zD^$j_Xv<x+l~`RJM5c#G8>W5TN@U^=GET*j^LfzR*%rj~D(Me${JaE34T<SJ+6py(
zgcTEF`VVb|8b90$t}lW2WCX^a)d-xZm)ZnO`rC1u#G)rAIi5nqbOf|M4l!-FA(EJM
zsnSIzEFM6nafnQx7M&3rnOp@4kZA!njUm&Xc*Ky29rZUd?Zt3uMY@*BYCggo$xE{%
zZ^i}mCx)esLJGsAbR_s#kW%P$IZ2Pgvyx)c;hqwPNt?C~lN?X8H72E@s39g*X)9Fh
z<5r?E=?v0@Y3ghtCRw;=Jy~}$%?T?em~;(og-UZ4ZMluM6056&m~<a$!?dqkiAmf+
z#tArbJ`b5Qd#=rmxED!}kR#}MC~62wZ_!q$5hSdbfYQHcE7S<$RsfW^|6~-#Kd4bS
zQBSoAmA2RA!OmbJ(sPm=P$5n_7@8i3lP2F1$w{HK=Ry*e4M5VS2uX*EZio#@u7U)R
zG!NUwAZb1xF-T%Z{S8S=Yo)UH6u48#z>8Yx9>@P0s&0t*mX>ynL?x}2#K5#!6SZM%
zIt08fu&GHQIc=BWxk=&aAWsQ{r>#QplwkAMl&9MopN@l~hWOM=TcIK%x1zzP(B`kF
zkVZ_CY6zfgH-A;1WU_G41!V2X1SqVSAk?>LD^vosXv<Bsl~~;!M5wz-8>YS8N`&GL
zGEU5qvwHCC&bby#P0x`IAqUW}pr|21y-r)929U5~LQsFBtxyAqTfv)i=sp>V@pm;6
zC+ep*0o0DV?AWPKOnQuxgDQkk`+)=EVAK}3Mq*TxH0Z(<Y6!s8RtQrEh%Si@Q?7yp
zFx82zV=y%fj~Gm`qyC1eLvqD#r{cg@lT&ipDOGkVsV-+#E@O*Tdq-kbguTRowN0df
zFly}&eIigRvbdb;pTbf}p=)1H34^YV5Ol?UaQs+Lvo&&k28tRYR}XE43ZdMJ2Dt(s
zG^&#(OjB?Oxa>Y?q~ex^d)AS4C$punVuD*=r>#)g(xNTb(pF-1br831CvBMabt`d;
zJIFYVN6u%f4;np9x`P}(PeM^c*m{Y!LJc2b#RRtArmaxJhg$*I0_Vy2i@&PzH&Gw8
z3AN^FQ7bA%>7h#wtq^oA23N#^uJPZEgs#Yh=;9aZ3gA}<;@1+<O|kLIRgeI`reix9
ze$BumhF|QczwvADOsP<CGWa#46l@Tht~otxEmyMUMIx8RM&J4&0|pMj|FXlD)%w!V
zaoL@rvqp)RC0$=f?{6?cryXHxdfxXs-rnGN!C2ZNa$YaRGm~=FB2Nk9svQWfvfP69
z)1GE)t~v~g8gf;dwn9ZqZbipc{w-)rqzThp8|Ess66=LL3-_E!)}2g@!iou3eU-LC
zB}R+3TtQoj)zv|++CbVc?dw)@6?c$v&W@bVL#M$4hRvP^kCQGThtMNX)R3*7r>#&!
zNLVppt2byX)DYrUfUSfBWgN!8sBt(^U$u#^rfd01my+~MC5KgrsrG>8$6>0`w?#6Q
zHUYXEh2;YrwF7e0o}xQqbCj#V&F%jFG}^~tTWlNSsO|8GaTGgh1xNYoqsAn3huDn4
z-Pnz6MDy(@XL}8p<28UC#ce>Vwx6826xQIfz3IwIb-THy`%g|$p-)>hu#jfI0p(^*
z)H1a5?QPJ~9Vicieo>>2(f%|?1_}jsj-lco+M~9R3%nS<zh{YY<xJ$t_&ZQ`d)jm?
zX}ANWu1Y@i{uEk)Zi67-q*G<kuSL|zXoF*2joOm{^4K<gXqS};h4v*-$T+mSTTr4U
zcMD29cDJC!<3PNgiSc?L_#FqbHeOFpvJ>8du5=A{t*oVr>E6Hsy>`y}9(?GGj2xY+
z`al?2FB;{B-{|YZF@9~~3&WdGB_EysbLF7i5U}c>-iCYL-RLP{Gq<yKSA;+IH0x$D
zxo_1yP!tA;3FAFMTcKhcw-S9t_#4uMnvw0^I<~qZ6!Qr-`ka)vaL-@Jx|29Bte9L8
zeneZLaNeRVqrT_Qf&TGs&=yTs2d@aXByE`Xb<!)saA!BCz(6LBoX@<)d@0Igg6&Fr
zg!KAcC<@cqKZ5j|-0e$Sp+=CfV$zl)X)9E);8uVYf+Gh1ppsdlo@x&cG`7_yKz~Nl
z<ANMcAzb)2OnPHnU_`<<tzHEVbKf~`0$vS&%5$#)v1EWLI*}=^6X8e86efGwxeDC;
z?e9aQYryldRqPt@D|p1N0ohU9ODM(^E!fL$X`wgetV$Q^>{a6wd_<v=%lH@5wWGs_
zN51DmwWk9>h0P?!q)qi&!?<)UI9=e9zKq1C>+#g23+dN*N*FfHYaKTIz|(AvO?N<1
zLu`7Own9ZhZbg$z16u_@Lz*y6r3n|()l6#Po<EXxCzGDA;!k?|o$dZXTcMJkMO%j4
zPK|eiwrILKm;^T=ZJ73TE3t_?$REhndIYd3az2lmmd<4Am2_sU&1(8=(l4a%XF^d!
zcv?hTp~jK0VggTx&{n9B!L0y1sa}*38UL(C<V3yJCWPA2BJ1f<N)E9QN?i_tAP!2E
z!EsS2rB8=0QlW+bQq4o8x>AIu*hu9nNPtx5VCxuCor^~dsn}5~kgE9@Q<l};#VWE9
z&6m~Jdky%S*8p}Dw*jqMR^M^o!w)n5sM9v;(wL@T0P^gY)F)__GRAA)g8Hq{BWlbs
zT5P&2P{?8h0_|BWsNdvSVtlzE*@F7xo;Dp*8ZM~os^nWx-+)&5lBKpwWx#(vqCQ5;
z>AL!~C-~>FW%{5lD-jCn*FhoUpzbcGqa}Aa9gp4RbUY5k=>-_4Z-C!%7;EFS`bDQ5
zkIKTQe7rBX!Ka)$YVZq8y$*Q5iha(hqnfGY$~EuHJvK#T-+nA=<i<?@98nPr+T(qO
zXdyP{-(F*~?f6R1=sV26(&JV|Y?*&J%)cb}CUz~9GU)<r`&C@&zx39GKHr7+juFAP
zQz{O`Vc7ycGNbQFj8b)lFR^EW`^5BC&y+pii6<nzu-?&A!UEqBp$lvFfI45k^j~A?
zH;P~E>C=Ti@-D=lP}J~p`cT>m#m3x<<{B&Ty#H~e33XoC-G#8)`&5J>b{B$Iyg5e&
zuCO9IcF@jqEMB-<OS*%+va0ib^^YF?vh+E$73#`LSTVV>T2EV{Mh~|FS61jfITi6|
zbt)$6rS>G%#{XzypBm2e30DrIP{N%GjW$lWET{7Y>ARtvdljVxmxdbe+$$$670Ad(
zU=o`qT(8Z@CX3Oo0ylyB`_5=F`WM(DcIEUm9<eJYb`<x*gC&y|EJn{>;ndQ08F2fk
z42UWd=e{9aOE~NO5xbK^Ek8$2vOkHQc00;XI)?Y(|KKTM`R<rdzDuyZz=xi0>)iJN
z6gA9!Bk%Ob8owLJNX)Hha$jhBfh|ZQv9q7;_5y1DvvAQ)WbF-H6u5_~UryeGwn8N<
zi?$p<TcMIw)9wx~Cx3>tVcOeCmy^St?GMFjJp!;3IjiCQcpKWOk`5sUPzj2{Og5pN
zGiWQ+01{SAXy*dj3N?VZ6`&pFJ{gJe_gJ{2>!&v1&XzjdVOd8HI<jgZ(0K)da~#mQ
z9o$5N4o`6|;^6rK#5o2L=T(tcVk3^Lz|EZgd1Zt+Kf|^##CaT#7~-&_{zjZV3QoF$
zU&n^4{aWBTM9mQJUG482#X3ISiJ@m~KwX%Aeh<DC^b<H)PQKUhoTQiiFMCQD1}$$T
z2EFg;w#K0Mpr|1R4gNmm1~No*D<&8;o;0E+B&$V$Bp77jqN!x<4O|q+VtNePg|<Rv
zF^je=p{-C^tZ8=#G3YSThG}oN5`%cv{oz`zM*xE&XZ5hzowH}po^P@4(@Xk<978!M
zYDhw-&{n81B&?W_&^p=*HHNqqAR*U-G9Kff)Oehzx7q|lQ#6^d)1Qs>R3t}Kh>Bi-
zSRIFoHh{lqDhi}N7mu)D0FRa<9=#~?O>8`J6(qo;pJJmJ9zBFd43F4Rf8)`@LM2tr
zWF7nlb0t+<Ted``IgyAI)SnoW#)qmb;+602=WoH!0+loc<n;U<o|hDte&Z=&xReRu
zlH4E3@;&}{Jk8dy^cEB~gr$#ZD^%p;Ry42_c>HkmU6h~Iq-BLl%!=`Q{OS^dg?qLk
z>rN&)Va0#BuLq@_X)9EcvuMknw3S$09R#HVNgJkp-AYj64l+)?k@I=Pv`nd*D>^n0
zANG)5AxBXLiW;I)owh=aB4Nb@mCmHCP@{-j0aTJcl+hUfs7B*NJ=P{znyQ7RP@2;N
zlN?zgV0spUa2#N|9(+avlO_SW(1e8p(3C-FdQN17*wExENB~Vg#)dIyx(|;SG_j-p
zhNj(Drz^enGVF!ql&f~P&t^vPlBzl}Dvj}L3oj`BFF00ElK)&eC4Y^lBn6~rJS7Z}
zas(j30}${MIIF5ZYw8~;dDGKu4M?v;QA0rbfVM)#J8ne>NcgCS^z+akP<}Q|n&I~|
zVn0w~;hu?P-O1D@toXZH4@lE!D^zN;Xv=)sO02F90@6OD4b#4EB_MGJ8K>LG`8;^e
zY@1KRrAdd7186xEHDsd_ZG{>@!iotSokm-s1`xM`H`LI5GB)7vY79=)Pi?{?e0DYX
z5PPRD|LBQG4y6zgJq1xW4iQ}i?xKmvpX^*1!g2u^${`FrEuw#H7;+UPfT15?n-~n;
zjYkZI*irw(kdy1l)?g>@OggYDq-I3;?yp6|kfOQ;Fyzw~hM`}8V?7vB&y^GLmv~B2
z82Y)VguzfL1Vag)F?!Y0ZH+@OK~X~-dXKh3g*a|SgF~TbjQ&F!F-?*oAWCv`&B8?!
z?xscxnbCw56C|2KTcI+VMO$XkR$_H`5Q!F&HcWfFl}N-LWSnXvXY~-cxn^@~?KsjS
z<On(jiW)Lej<!OLAYsLXiB{8As1d}i026Wl$ykhkP-AhTp8Bssk(yidP$UOa2#Owq
zuowr5E(3Q(L6KT97mBc40E$WoMURW{92<&U1qq<&d)OuhMYrP-gCcg+-%zxmRIcTE
zb0^2Tr{>%IOv6x}e&vb5Xk1WV7>#}g?iFYhJX=o4C-9`CaP+9Bguzig1V^!NrM>8B
zwnn4pp{OAmy+vE0A|1D)L8HK}w11H%OjBhDjN;!)vvAMYd#Eu&CN*Kj1dAroR;Z+A
z(Uuvsl~`RJ#G>6u8>W5TN-W|IGETOU^LfIwGJF`bE0yj__pVOuGIvF;W^-5VGo*va
z0d)ivHKe2rZG{?8!iotcokUxq1{Ajfl%#d1jL~?W8lw~SW1FC9TZ5dZ$0j-0LfG^m
z1kE_uv>qHrV^c68y5NK*1K?CgaC%6D@7UnvDo6lMH)E?9IBmcq22Sj#zrpDM`BAqN
zdl6pw;G5l@wmqZnv=NpPBh@5Qd6=($3cVosiYzK8^21mnDQx}3Q^H{D6auy^Z_7RB
zX|{%~UqMkr*m|9|LIq52MF(5{+j4&+O_(O&Fl@!XEob4L;XkCt2$>s&6%*LnjJ87M
zMvJy=M_Y;2)j`;rL)tLy>sG=RcaU+Cj-1Z}=UF`4eHiHua`+q!MGetvIc<d+KEjF#
zTCJq5P{W5?0ki_=$@q)Es_{2bAGHajX6ZnR%1nBylA|g_Rrf-4j6+oyfcI#sBGaIY
zSg0m|Sf?Oj-6w)}Y{YUEBtWd|u!#(@uE!&WSnR035o>>C1mI4zu?cH|1z`mYB?hdm
znm0oCH29C93j|+LCFS(K9}6VKs~>qv7+#&xI=p(?(`=1bPeM^cyn2bYLIq83B^s~Z
zCQX>8+!o@Mg?k46h#DhgeiT+r@M<I43Y8x%+A^886056&c(oI0!?dqkiC5e~#;H4U
zK3jb@_dwDe<nY-KiW=h8F|-wG_y{W|c;(PmsNut{AYQ3ECgHDY{7uwHZNjV02)qjB
zJUv~>K^3B_yC68mp{sS^JesbkB<R8wY6-yA83<Q*i;x`~u3QBP;Oa_jA%m-{@QA?`
zJL+$^I-sYL?#emES_=24?yVR2Hvci66pviLA}l4wt4XBt@cPvcpcmZrE2^jjtb4FT
zQo#DYr-T94IjsY%$34x~fb|FzH3Y2ZX)9FN<W`~q>kZO`Y0_;WU|G25L$dB<h7?vz
z0BhvE)JP<^q_k+u7POUET^$6hsiY0lzHTL8aR(VE@5uRVwSKh^=?-%EEQX?nfOQ0I
zg#s*L#RRZYv=wUja4W&}EBsZBzlr*&O@O6czoOEMp0MPo3K7<A5MkmF)|ucvny|<W
z>LM1Z2_V)vh*-Caz!w{_Tm=ab>l@fahFF*35ko9?)Zd7;)Efb?Z$V~dxdgXTV1FmO
zyMw+jDCsy}ZzM5TZQe|8n5n)8ZWm0|tdzv5JMiSBIQ3mm3B#%LLpU|=FnFx3R7@9+
zTTv<%Y+loQ$Wv|2Qx8B<L!SB-ZH0=G+=_;$hG0{bh5uJb8>UG%1XSEitep%j9P~a}
zc`_vmD<(uW_{Y>>B#FwRE#qk`@j821spu%3y$xx@w6j}DRNO(v**bDc4>_t*cPwrS
zE+jodj-WlDs3BAxOk1I#N?0+0s$*#@)Cl5M0IInEWFW>rsDU_9Pqm4wbl;olY?7Gt
zNF|3=2&ry@5EBQfP6Nl$NYyM0x-^9v0yK3#($oeK^kUPLs~`bQ^<nE6O|8cxMpNvl
zziH}leZQybRVP((U?B@Wl2Nl`t9_&H3!0crOj_HUSqXF3&Co%DyUcXR$$twLO-f<k
z@suzMySRlEcAux(n!<hvMGYzJN!kh(R=E`e3VVUHVVam*NnsWadWWn$nL~vY6AJs7
zwnF7li?)ouj~em@ZP9ghfWjt{HcUIal@!JuWSrY0r}WVM(}k)HhV4eWgmn8ZP}C5^
z_M@#(Lr7RL!LU!$R;YNvtpJ7z2g*>4e^EnmqP}VqhwW~@@ZitKeR?XB!z@H)*FrFe
zLuISMjWm^+<w%#&P+x%2E=ESXUIeSyjOHpxz-V8=wlYS$0FM}>v7`QGw1vItHK}r?
z)LX8l3Qn=7mbC-2$W1y!`V-UFc#X<1YF!6@7O15uASd$;cwSPj`nIQpan+?Eu99yS
zT0ZQ4x2M?}tL}uNhFJ9oZG{Sx+=?dC20m)`ENQ|t(T1psS&6l@W#OJT$-0vnQCKm7
zst;%@R7SLD%h3C&@ovx-O;-nR7ET~-nD%ump^7`mI8jH==fU&m%(YnAnnOB-bo$Ov
z)DWo_(^jYfB&?Vq)uFT%DpqhSfK<$VG6v)CY79=)Pi+D!?be*(Or)nLIi5oFbOp@%
zIP`Q9_>86}P1<u&2@3~M=~6_ct3(KjjY_V91gLZ_HjJUtIy_>i#E$wKm3GD6j>AZi
zUv`y>6T{IcRacmbt^}6~DpJpsQ}Jp%At?@B?kQn7bVciM=yp%DH4fbhMGbN20on=`
z;<yz}jtt<?Q=|#g6xl)?vT)C<WZlVZCajp?(0jBMDw|of<v+BQSX~{&p|KB8bIP=@
zTZu#5LB?q|ay}24GkdPh>wi0u9wA52G$?AwLwnFxs1YQrnDEd6v=wRuaVvPIkNHnV
zVf=#{g%kBun}A6BQC}td=t)Qpr4R{S3R5u-39STY(Ilj%Iv0VkQ~-glKm@u>1ccZK
z<SIyjKwrcbF$6jjj~D{6qy9#q`PjcP9yp&7b=%I<n;3vbDyqW#^9}H);2-5gIq|-U
ze@}`(mv~AT{#@HS{Mq1Xw#J_up{ODL{E)UnMKo?D8h;)mO_-*{7UGYEdtM~#P9`s5
z#RPxeqODNL%c3p+qOHX0>LC6M{|PmxO#8Z(_`@A!oKz#{Go){;Ey<^n9wA526ewzl
zKfBOYs1YQrnBdP6+6py-xD}2+zCFqD4{8)n)KhK3pUJu#Y%=%g=|_&F5dB;XGcgYR
zIN&Rqew0M#!VeY-z|XY^KkG#}hz&okf&}n$DmIA0&uMtX;D;UcH~j2TDE0K<*M4IF
zQ1s`1Ro#gpXsoZQx?KzUz_$W~0w+rt`Wl{-6oxMHlrR{&sdX@Py{Fk4hOU94hA?y|
zZH0<&+)6YIJxH1`O^_{wAq)3BN7kK8Wx|RH482ZUp;DPeTmD8{iPhCX7#jE>HK$Db
zx|J}*9b}whBj@v=IW}Ks*@pB7IesQVQ9~5knYKcWA7RA=h4!SaP~(SN0Te>-$q0-;
zs}VR+FSQAT^dI>3r87MY$&nO-q4QxT#(|*}_=|?2K(cdz2nz;)=q3cB3q(kW4MeVj
z1VFS38^wTVH6Ae_Vn_WAM7vhgtDHDA6nR5TDo%_-qf}kt1);BiOWhk{%9(NsUWg|o
zg`kF~gh9}4t%IPeJk8b+^erfA2tl{fR;b9vtwcl6y`%}##MnXzvT)DSWZlUmCajo1
z&`Y!xDv4RN<!#zZtga41(1)ZA)4pyc1aSu$r`O2&Z2674i9e-+3^{-{gQA8MG>x`G
z4Ip8~go5VNR;U5Qt>E)P{%_Rb?`jNA)K6`~AnoliC8OvWNDibB1DyktBn|@|3(le$
zNKJ_@0AZ;B0NsWFv`$Rb*Z|}zNB}@(Y!L&XlkkWE5IgF30Aky4#aIrC1R$w6IRL4;
z!T@wGxYPw8<xB}cU&a%X0?^r>5(YqbwGMzT^E6un(AS};ApqS-TcLs)w-OCNcabJc
zlVS@2$ih93lXWNam#|_2K+n@wsQhKomN#fCvAQ}4K<|?_O#8Z(0K^?+oLD30vn2pc
zc!&xz<N(?TiW&mY6xs?kfP@tj0GdTxp#~7Q5(XgrU5&ws`su$3KuSi@1CSg@Apkl9
zCP^Fs`ZPEz0)Uiixd4Qv0swRu0??Uas>TK&S3v>*%43Td0IkF$20-kn-vOvpua)cY
zf!<uj?zOQ<0FsK61CXjK3_xE5m%0F?oGAh5OL#(306N`M!T{)A0)VQuip};%>pjiZ
z0Q6NTY6w8r&{n9R#;xc82wJKvcHc&tFinbK075IV-Vn2J&qHM0$^0d(m;lhPXe(6y
zvS`aIw3S$09R#3vNE@bo-AVxB4l+)xk@I=<w9Z)-r&=%6sy5%88~ZR7Y{-E$5{epv
z&?MRlHIRfA6A+p~TcHLLw*nyKbEAyO_;)oXC+fHVCJ-qZMh`@CV1<C_6qqz|fap+g
zRs;|!)pCIdO9g=FUIe1k#MF%qM6QAaK-7gTVnF2J5d$K2)Zaif4<5Wpbyu8|>~Ls$
z)br7<+Qcw4LTU=5(5c`|fkN^;Iq5!+e@+TQCwod52t7!EkmalL7kHYjLFhavY6wE#
zqODLtja$)y5Pq>u`uQf(glTdNgAiJY1wt0?xsR+nnZJY;6A*fmwnF7Ei?+N#TZz@x
zK@fVAv|-xUtpp+NAmfA@IiIaQTr&I-D#Vb(XD}2s#GlP*E7b52R!s0`JK730e7F_B
zAK!;d@K-hdChDU$LC@B@$D+NQqemV&ghI$uhv^Xqc@6+q(a0kcoC`cyCICDSBJivh
zQ#3a4xC#;g&k5Kb20SS|V!*?W`Wtw5ucnK++S(M`FWU}!W=27esyZ?FjPYv=!_O*k
ztbiZ?xiT@Y!Bdg~P|Z`q0O+w00L9-4d!47+nt;xNqJ{+Yb=nFQ*SHl80R?x$zLvCM
znifL{lwc=p3kTgzR-Q~^!iot4JwjWdl9)wXo~5nC>+HZDjIWY5Ogp=k48$E|oL?iS
zwDlg010SU#4LO8925%a2&_=WsY6uA{CLA=Gwn7adZUs0fv<D;pMGeJ?`l?MV)S+V`
zUmDW$kQ`1S9x6ez<M7ZD=mVOE{5j8MA}kkRqQ{VlP7>~q%|xz(1Wfc<Y!hRmWATVF
z5j*N{CYoPy%B4yzRV&#6QRG|lp5DYXG*VF&MxiqJQ=pJ?qU4|o{yixN6+I=4gPsa;
zklZuT@{ZV<o@Q$d`T`U+#GtRzR;aMXt!OZ4u>X$O6{HE%#2BI=W+m1_kcE40C+kjT
zFk!_6f*zo)P#MgkEl<%_Vs&+JA?QWYhG}275`wscj1z0*d>%e+b-LKK!sc$+hon=;
zLG%F>HN>Eik5S=_{HBbsVuC?i&{n8H#H|1Zc`lT(82_fm;za$`CJ@rTD=#yUo`U3f
z3Q<rF+8l?1_Jl5=DM(3oE&^ep00KRQ2((f-J~jfm3KAgDk=P)HK%d4VhCuA7zY%D5
z)p5G)=x6)L<seX*7<`6c#dI&k=fQ!3dH6YUiWTrrNpUCZDPg$tEP*?g_q|T@G+X1&
zS}1CWJLl0>sPM(D=y1n>-|JG+glP&4<4)}RUKZ}zK-QhiRKki0?);FpLg9`@TOOmW
z#OmrG?mS1@FzxGB;tqF^ar%s$&jW2<gMXiN2RVG+gQA9@Gx%pzAR|FXSTO;e@w63c
z_;4$D-%EW99)DHiZ=ybG6XI;CUF*SIqo*4=d_r`Sf#$}coB7~(nr?VvbHN7B4}i_H
z2sVyzcWkh66(j(gL$EE2M}jyMj~Ljnqy7e)d6jxGRZRCfc5oB<fRC#-G2o1ln!*b^
zUEoaj0Uz%?2|C^Q=cJ&s!c)RP=LG_EEZ22bdz!65rwT<4LFX*m3KhP%6&>jK*L5x?
zO_(OXFzBF_SnE0#?zx_<JDI736%)|8leR*EjzwD@q^-p2>LBPmP1-Q+>sEpecaU)s
zjhxRzr*)M}Hmf@CkS-yI&|6T{5OzMMtx!WqSTTW}(T`Ih)1WPQbBjAr#$mLt#^FSL
z)h6K4KIr4+6g~9Fp%j9i<<R6f&@&7CPeYGPh%WeGnE?2_fZ&r7ZjTK<u7U*Mvp=?n
zfzJVW#K4Cg^*8t|Wbe#%rEBS`Q%jZ8wN`&9ETlg%3XRvO3^UOQ;OF(-e;C026L6#{
zDQD-2P(YoX>FhSA^-a%CftR+nTi!Q4KLW9JBnU^1?%9*Gr{JLbXatm=lt126!YJvL
z)=|<)o@Q%G>V=|)l=KDK3Y9v!m1s&jpERNRwgJnw-Cd>LbgtN!KY8MS0gd7CLv3x@
z>C3l}k}TYF6<K#OEeI<nlyoa?g-UT2ZMm1W6056&l=L`h!?dqkNlDy6#+f&AKJyl1
ze3_m%0gE?DkB}qibtq~GNgvQws1YQrn6zc+6I8@BXbV76aKxZ_H3}!{sWvf^_M2hh
zq@;%>Iif<abPTjT4lK=pexPAVllWX}!omS+dIhQJSkV!&smWE4fSUHkhB0bdf=7&+
z*ikE}srkr4ckBd-9=veF?bL9b-WXI~+t(PHDWt2_J|7q47a=Yz^BQoN*8p}Dw*mJd
zW=!TIC)+vWj6Ot^q1j;!a8zFL#n$0Yu~x}B)jo_;FMx{1Ojy@kHFrrN*Hd&VODa9}
zUO1)NF`c!fBUkKD?%K`hYfMr8*{LlWpgLAhWyvZ3GFbIjcY6MH<+sLWP1Lf;I%1S8
z-I#maUfb22HMIpD(^z$<Et=V}SC>=Gq^ddC#k-u&I14(wQ#z(CS|r}P$7?d)Rg(I$
z&@XDdF$&`4Kp{UWl7SEH)mJnhdoE87hZe*5>+&oyo_qs&@-5YEW7M8uH$b74{vTyy
zzvOAtG35082cS9sPY}ML>{jVd@P{T<`dFz_z8kGTw?RdkbgHb;MG-YJf^MCY+7tR&
z+w^f=Rw5MFQ_x7`xSk!BpN63Wy(oe@J3hNFzoYwmAYQ+L@j5L$sg2juvy0uy*Vt|;
zT<TR9bSx;ZT~Jz)UvT37xuR36t8w5&ekUC|aJV>Gd5!V=fi9E8-`F}*XZE*lJ@WUX
z9zIMq30FZe_>+J^|7@F{eS=5gVdb?uLwo#T7(A7I0lf2L&{1FEFX;=B!`P)prZ4+<
z-&MyGP)=TTGzOyqwiN5EDHn2?T&-^%`}68_rP#MF`|s?B*^l@H__(2K71rn3>|{2X
z-^6Z1Z&L993oKF#+%CLPU!nBr%%ieSN7@VH-K9!L4Sr!EwWC&n3n=)&K}R)H$(3uJ
zZj`VqB<JjFq6TgrYVK9F@^52nmCCPl2SQ`glqntXsq&0dt-`}4E7o=t>%HZ*U8PLN
zlqqgq$#ZzsK>c5#W>un>MI_7Sm8uBe4j2v&O^{IcsK7&TrQVeG0U1rEP1hhJuc+4?
z$gaV!NQ6)CRI~>&t?N>^H=;A`XrD5cUw!d=!js;_^qne4z-D-clHL6yz;LN#yr+cC
zhJnxld0(EF4+>j8IkSVOS(oO>`{CO{QK*N>)z3WI3U&R=t;oPB_zgBl^*=eYlr-Us
zUsg*cR!`1gBI03dQr^NnN0W6Y8A(_%St{wGtxy-b7Hz4}R$_H^@WS_W(uQeYCtWHD
zceXLAM$YF!^K8DYemUt6a{PP)iozuJkDrBZDjB#?VT+q-E7bTAR!l<0J+u{S{BSEk
zV(2{?f$?WG0w?OF_ONmTzaSBj%Je8LM^Xr-$3VM{QJOO}qv_v4Id_Sq$wdiZolN0<
z?(zs0gy@%l4KIE;w+$Q)_%Kd%fEKrXcMjC4i<-pBf~&yI+x{LDD+?n%Oz>kZUtM?~
z8^xAK{((nqd4wIsU0q--(Sp^5ebbqYQ*bKjnv=?uirsKo-kmFipREaxjt;NEz(u>O
z*S7k>q&Jipkv4CpH%v=of9}yzvr=-}ZiFW%1*g%T5(ZAg32?G}5~st{Yz<CZLs3I;
z>ZGku(U4owfs_A9oV`dB>WuVV?X~tElbTB{+;cctcQWw_EB?fHs+agqRT9Vvv=u7x
zS+u1<TZz@xL2x>mv|-xUtpq3TAb&tt>)|e_i8a2+`D}ZS=@QZ-<Ou46qK4>n9c_gg
zLBfg&I^98Ap+*q50_YSRG57~H3McBRHo=pAfxfdzUea@v99SWa8U`+i!%@G6a#0-B
zEDgFyg*pOAH5`#@gy@#oNaZR>fK=~b`xsKai$@Hp*inBY)q-AF-pdtJ)lAwRq&lNP
zN>!g2pT_vLg`sKqQyw(=&y}b&5>H8rN<%#*43)-)P-&yX%C%gnm@XXGT_~k%;zm*6
z(s6;8K^G_)bSqD{H7sooMGawTI&FoDf82@&mWHynDoxpyG@_<1D_jzGj2~TAfyu%}
z2a&ZW)19zl0!*K!tx)OCqAlIDl~~;!On)`fhG}oN5}3GyjMH%BtR6h4)8;L$i%5r%
z1L!MI)R331q^(c`NLVr9rCVq#)BxgEfR~v2WF*Gl)kvJEpW1{=+v=AGe0fL@NpdiS
zAZY+JJPsuN63RtEl0WUaXoMvLXfzhlXprcF*l6S`NPtFfVyhS${Rxj48nL7PMx%YJ
z>0Y=Q3tw<Ar*jpjE7j}NvZe5A>d9A|^oA1S(&o+dhQVpzFFbH+R!UCX!FX~~g!+#d
zdSdZ&LI|PcZ92=<rcKG}k+<nKhN6Z5wJmLh3W(f_2A~4_`OGFwm?qSOPc*CWWZ|BD
z$-0xdPgpU5rz2@ARPM8A%Zap=SX~{2ry^;?w69wUPuxMqNjY*pkD1n+UX$YYLTq0D
zYLK2GN7A`a)R3aSL0h3llCWYzQ8&<5sFB3207d!zD5El-pho3HJ=Z3jYO?y&Bv0wV
zN)EITSp6GvN*u6y0$djbR?X6)i&&^5fLId{u|5)cE;eGh3KAgJtJpq<Sg+v`Lo9aG
z--xwW!729CvZ=BIOC`0np_k+%&}w${`jg*4Vx$@u)E7pn55dLa`NQDZa<YDmCnZIw
ze|bvSD%8XfLM2#*8uv6+K@y@yK~Y19+M2dP#Y1jI15u$>s2xcorl~cAQAt*zEL^mR
ztUa0jgcTEzI)t`Dr9X?de2%sftGk0Bl_hPM_I4{liaW?S9Y@Y;tDS<*Cfz{}pD#jD
zLwf3?tx&^9STUiet7$9L@ZnZ~o_sq6;jd~GPSi(jLZ<EYD^P(vq(>z=ltQTVcgP)a
zQ0ZZCSrjS-)1He+ST=x26A_XAA@W#kL~<1*K%^J3T?~<aher&N*inBY((ZWG#CzKo
zkAh#!2-8%jRGt`(Mya~OSoA))RbY{NrbMC-@PwpD^fyllL!xa$NF-O4EFb<J@k^?D
zBoGaOqJ}`UIc<dsaomaqhyo9PZ%>*qO^_iRVpd|^lCyBn?quD`Y$mLjz|j7*6)Kxq
zwB;z;O02F9!q5uRhG}275{9^gj1z6-d>%4)_PpKMeRQ+;<WDC(LXMzQp{OAbeTB9{
zjUZvggonOKTcJh}w}Q_wG5^UZjDJw0aH5`S6B6<F&w?)*tGP^%Lvk>MaOf|PH{#&X
z{ot-B98%Mri$Yi~fI{0K3cV}xS8Nn=6(m5R=deu-g?^1k429THf1}W1Bn)_?ueV%_
zNkqFuFBpYvBnG4L8kJ!_dIx+h_()SgPSE%8yrh`)wx@()($v;r(x6{a)gv+K6R4CS
zCT&bxp&}o*5{*g!Lz*y6oh`&93-`<=>rN&)VZ{WK_NJ{+NzS4z%V;aHx;ltS$CEZp
z`?{5w#2sXud?V-ckU6tw+dR5-GU*X=1g(OihM06NZG{>^!iotdeT}w4jUa9XF^Tz4
zMq&Je8if<}RGTnqnttUdoQ?FfB*#>Umi`ElIu0$}173@wB~9{kp$UrzplK>X)1O2(
ziw#Y#f&|d?G&YSv(=YLeK@&UbZ)n=1H(gyRsL6H#$zTHkMSEgU8tqdSMx-~vvjUNP
zr^;#h7M_t5j^6N;FgV&N1V{0|xcZT&n$Ic!1_vY^{S%5B($T1AC@+w~n_JP)QSgha
zTaq?RQ)UQ{EWfy_KAdFXpqXUl4IC7BkYE2Q+ydGPmEA1b@+sO1mED?mcHoPvN0T;8
zJG+&9#H((cbt9*=^%qxbq)W&lREDC4fb>P$3N?g;6%#<ZkhVe%A#Mc#DfGov{EHfj
z6ZKV_2&u_V0%{u4vymK6AvXFw#OXL}bO-n=ij90(&!r+P7@(q^kcwUtxh6Ihxe5|c
z(c{=CMnzBH5u+k@)ZbJzcZE|++v3m;(W^(OHZcbc;ab7~^eQ+}01!V%lF#e-r=;Zb
zil>B;&zunXBzU0cAD(V&^!Y0kHAJ5w&r)t6!!oy`L7&hAMVpXDOw(bAeUdy-WZ|M6
z$l4pY$OwMs(N?I;Wzm+Uv=u6IHSO-;;?Uuw4b$FkCH(NJ8z<DrSv`1m=NyY$XhqT?
z<N(S;QA6}Om9|0+AYsJ>eZEXvp#~7Q0_elsCnGWbu14ZS{nRG#*-F1O#PW`wc;p}o
z5zlWSJjWrPTfk9K#KY5^i#k{$fI4#!b$%yuOKj9}6(m5Nhp;sabsokehC1x1zfotm
zLaAC!;nxjpd1t3+-tlxNhMtj%sxbMy2tF0$qns!w-AnlQr1<lKr-b3pZXx`U4-Z+s
zlJ>5r*&2TS3`Gs$=M&lr70kF54g3UNNgMOOl%GwLVF-QfJ~6MZ`dGN9gRDE5w}cgw
zyJDTR6)JC8v}G^aO02F9g3m#u4b#4ECHQa$87I`p`8;wO?CjoK%5~YikCr9fLJp%Y
zC~Am5t7t3KFcMZw@aIdk6>1o9D}X<W6J<QczpC*#QJ=L5g0|5^kjz7R5|X1TL_+@y
z0XPl`-2mQ-A|WO1xj2Ny0ywlA;?S=}9*B)Yu7U(ObRRZ};n4kf#BhimwE~Blk0x}7
z*a*|zxSni8^G|F%={4ZzUIW-s+y=Dj6C1M+tl<urxm8Z^3Am=8*_f(=T+`wKgtPzD
z#zaCjLpt9!+%5gw#v9N_YP>PpY5Sc(p}^fZD*B;4YJ2^_R(!8|mKa$sM6!(kxsAck
z`5Wu6x8dhDbXD?g*Zw-z8M+NZJkhZ-<d2W2mC+|RbhT<v_|Iee^buZGA{60&2Zf9y
zy!*)wv?Rhiv+90w1Ah<1?}Zq@UxwdtFl*zt`pJ#$gP+_uu-NTXoMJ}#_73^YjroB-
zlNSJu$=)|NnkrR3y0I7)@xd}dbH1`C*ph(Bz~D9d>c&FQ5$CHL2SYjQuWqn;@Nq-i
zEa*M&vm1-C>(HT8T)+bEXE#`ZxxyX3&u&bSUrxeLZsgp6-;w4YR;qg6+Aug%&d`jo
z<z_^`>s6pvtGN%hOQ~i%KCp2T)Xek)8*Wv^#xX}jEKTx(jeXQ_l%@)~YR$h)(ln>}
zwv^t@D8qJ!Yz1<zY#Wa}(I4FyuTdFZKDh{co=;Cr0XahY$Pr@r0gekjB`nz;6Iw<Y
zc?fv<ILjrJYdx*jODI=CQHZ4`7eu$yRwyLkR%8VAKE5#oo2x{?2S^+0G_=|kk()7o
z14|?y2w#)(!a@G!jpMxCDUVYs|14Q~@(HG<4hpO>y-J#h?H{`}rbzejEj<B_1}<0J
z^8s0R1NSt)v<V<(<SZL>oV)9&$-v*}*PJ=@xq)2O7!5^Xdih7sLT~h_pT(F=TcK7p
zgcXyv>_l6kCNZ}Hy^7wGaS?x3FE0}HQhRu|u~XAbrAa$_@RegK1it6Nz%~Y7M!EcR
zbBBTF+@%fTGNPeLOYSlUst926F~}t6i@B!7WRp*2xC-1c@9$NkPi1st^Vl*+4<4~)
z4t5lGm4FdI3qF;x6h6&X%cTn`gvV5_YfT)g+0~3{f~FEf&BkGkVW#;!I9xDI_<T8a
zPsh`e-WfX8Q^JsDc?fCP+s{~=>47+?-{bkKo<3`sIUkA|!ps%46)FyLD;k&yyiC4<
zG!Z{nTYX7P&DFf({+X;+BalTSJJxEmmB&eMkV*6r-mNA{^m*C}l|+RVlO*~EZG{><
z+zO)6Y$b={uj)ih)JJW~p*uFup`nnZPoZ)^g;MBha6p_CI{JCD6dI0#ZU#jSfegAF
zGw52;C9yN8s~|xJ-4<KNGU#@A#4;#5YDETZKE}`;;zohH+(tH{`Hh0vUIXTM4PZxc
z8_=p71-p2I8E$kL-z%7|&G{M)fh=!-yI^aoe&Ttw_T4c!2s%p*M@Gv7M+OS{SJ21+
zkM^jQ!9Zf#-?PMsBa`foLAR$(#~X%s40KiU-7)wSTH&iqbl*xkt3_1I=%#_LV(m!~
zd2FIS@XJbs0{;>yWE}Y2n+9mfy=j2Q?o9(c4#e^d#`1mOcO1{!SgzhQI5>3Ez+GMr
z#)uPV%8z+#@0FAH7A_GLa>J^G$;Ni_-a=E`eEmRv*86Hy#|Ia29VtiFw*%(=Q`PM2
z-dBQ-IQJQDfpXUO8L(rsA5-_*pxeE>4G#7>x)@f!VYOmrheBF}=eO}ZPjskqU*}=0
z6!&_Q&s$25KZ@rzztS?gv-4BdB7a}0_^!-7-{R;mp<G;#H)&WCZ<9#j-P>W^vtBsF
zcW-BEvwJ(xA>sQwhG)x8_!HR)sw=U5?yaaCjwYqL?YPJDPpFycJs!6z;vUbb3GeYN
zF4b%0dM(wJt29{#(eRD${*G_s7>ZjurKtcQ$u{u>5q)oGl&XuZE%?&y#NT*xQ$16T
zg-Liq^9h+ZO0HXYN*IWo7Fzm<{ZQ;oPqPk6$hGq6P}Fb{WC3l3x{l>mWZv@b=mc(9
ze2O%o&N*MqI(ecRv+@ZaiWOYI?&nyz=XkR2B+&{h{v`|jtI&Dc3bkZm(Uvu|l~`RJ
zd?@y8(uQeYCw(Y3+}X|3QsBZOaz2lk);X(EFIH`yiTyU|6>=0^2}NNl`$v)f?$P&X
zE7T|wR!kPY?x(F#qljAp9+N(l(HQ@zM&m?1)}G6o#x&E^rf?WujEoxzZJU&rV=HtG
zv<bA{_!@{^X7T$+uR%HY{t<mq*-(?9yYhv_v4YBH)=xv$*;I6f_G-!G-jS=oy}I%D
zrqR8l53y-%<?ACnVk=+lDDJ`$Ba9Z@J36}JtWH<pdsn3{2W}v_&y=NBq-&XMs+v1F
z@Fo#8XnYiPkm=e|XJic$!_^c66Jh3>@Pfx&1`6cl-wewoMX~Xo5{6=DhEUA%F4PX5
zK5GEm7K$1I*gV<_6;HVp9f0BeA?e+vqzTic90o9}yHILQ<`wtH1hpE0oE+J)18v^V
zJ(2VVnT(I;-D;AIi?kIg84D{W$@o;-3N?7R6}$^IM;R*kt2z-A^--I$@FFUp8iYoD
za+ZTEl$^(aQ{p7&E1{fWayE#TZq7!nft-CN=In8z>tg3@S3!cDeHXTv<?J8e5zE=^
z=>KibE|fAW`!{EMHE3VXR!oF*_ER`+_!?%@Q79G2tn#!NZ|-pMuSyy|4Ev;~gyrmY
zNptq^J$=?W`*%>(FlWC@TcN@(w-TMRKOs%1$<l7oBzDf`6*tZYksWKhGyT@T^=Bsk
za3j<9mQWOW)+BB3NL!)Owy<K7ws)tkP~(SNfwUdmnI8WTD|735>i<>dmcd7#xn%=F
znfp<gMR79s@!-s)nOj!d&D^Lpkh#}k=Kh(O>#;MptH2G#{#j~twOz#)v&>z?BbK?@
zQGaLdgS$$pic>9@id84o;}qG;ouy*T8x@O<FR}Gz6X)p7o9PWF=JTK<L}G4MN@nBp
z@#Lge(Ca)UEE``C%Eo;EQLL2NgPfOp3a!)eH=w9tI=-2<LWNUqMU#$$8@k>@+K8Qx
zEjM&kFQR$X{Zn17N8lnlx^IuF)E)6$kx3H!HR%yDAwR?W)+8bSfwn>=WMRdmEq|x2
zP$P(2!Gz5GCub!7L7kC_da6wcnS2JWvq`AZXJI+MLRt7r;E6a{c*_@!vv9Kj>84&(
z6-d1oVCp?vbW-fp>ncc)dS_uXS?ZmQM=bTSqyA34i(vp%Go?zd*b_6|nm-pHO(jmT
zqf}ktw0ZzIS)^6<OgV86#1oRH()~OoES0WLoJv3EDYQ<dM?+D=RGOu&P=SzJF-WDW
zNE>Pjv%0iRkV<*gjRSOa-&&^9KGGv(D!q{Rtw}1qnzllvQenj;m427DLX9A9MUzVL
z59*9e)KhIrrS!dFB_!$7ryNtE^tm^*K2G|42Fe*;#HxYNO_^9ckTTa}%3LZsBX-Jk
z6(mTRZ(-9|%KS4Pv6RV<`a5MVfsXDfmr`seikL}rzRfkNXDo5r9HUlUTnG3bs2lt{
zFLC<Mm8o+mo{}_m4)m0;)OlIr)VaB*&^mQ)0!0l|=k~M}D)4bD2B~u{X`??<C$GA3
zK#uNP%hY)|=@Bw@9>V+9Bz2xZTcJ{?uws%r3$zt#1aT{x)QNvkXJn$DYE$Z*LEg~y
zg(ZFRl%pz?JU@c`5GQ$F2!1n6p8jCyrcYE5NS~Ks`urqm`g9c}NT1hZ<5>E<0gqVv
zWJj$?pUp=Yx<kBF=)R6aHlq1Ug?D-l_`cTwb`-Y(t$L|&&fy0hzTY7SF554)<j4a)
zbLf7{j!GT6<nY4}Ty{Xm^jxt69vQ0_oDL<y&**FTwjXR-K9Jw--zS_%sAgBm@ps&N
z3OYdzGX8sWhVRck?kQnu|0+!TKK^(TPk@?2lq1?AVdO<mp)Tzk-k{disy$&akL}aP
zH(7~Le0vxQ8OJyG<vFw@f*Z5yzC4G&2V&M$7_%OQ-?4vVBfqPFobE$#udY5vrI`F8
zcycyB>LoW^jm9GuuGmq3&-U$?)N3XA!qW6iAziI@z&!U)@QKPlJAFkGPwKG&b!;NH
z@2%kN!T;(+H+n00W}uKiI*^El_AK5C?(i(J5ONKMkoZw;Z%>;(eta+j2ID_L_=c=|
z+vQL22bem#D*2+?wrGV%wLoXeNVY7ZCPr^6>1xuZNQQ0FM>1K7P$ZiKg{&hPT5=;9
z9=nkYj{}kH8jNI<;rFJIOnnXbP@Gu3u#a#iT_~*W$Ye{Us?(9K^wfKuVy)WI4c}Ss
z*gscvYIV3JEn@)eN4ejsbfKQEmE^D9ev~cNDo02~w95^!{yC<GRPc)TUgLG>l9K0=
zY&+g-JRQ`VzSrnhMZDK|H_YE8dspt&<J3~+T-hl=Kc%4Wswozsr_b=O%=ltxN3T=M
zmb!q;<}?l4olO)6!nte?AMDYuG)`=yc=`;@f=pwJCTH_^!#8jKcR6mZA;*m&BVFMs
zVPW?k47)GyHekSvGx|>Jo1Puo7$oZ7^vw}BXubrcU`Xl|wkK3IrusU+X~*iLnXG3h
zntcO&VjRu#ER0P<)_DZVx!0Y6e9*i<#WiPRno(n*GXkOU9*hIu6ulT5@>~UO4Dw$H
zil>l<@#&?*yqB=)?5gx-JYtx~j^bh-n}6bkAAe;60nxoGb+ZH6g63DHZ+i{+i`M{l
z6t@Abx+>kiS}x>jECOQeYjz>Jxf<|fkw8?lzw+Ez>Y9#G*bQd!(QNq3JdL{@<xh8p
z7o3AVB`j>+8@%9bjFR<sL#q0djG~))n)IY-#Amwd_@d|#^c}CrfX28kr<zHX)0H$#
zD5uhQai14$edmp3=fQVeQbjnk$!YNL3ezoHG>dFeV_>l_9h^fuW06Vu8Sp=9QxG@Q
zPvfU=XK;FBVsE`r%far`kjA|xv(2dc)?p|UX+!71_N0yc7Eo<{PLP!d<%CZlDj4Sk
z_gWn-xrB?yf^ZvS<lqbk^m{R&4}i0qvToB<jQ$jhbglw-mic3-2<Ria>%~khS1Rg4
zdI2lXLV7PAv5?M=S`iUtbs8J-g|(eCrD6^797gAMuuwF0d7t~I4&_hN;7>D_V-n1c
z%1+9T;a9X61P)%RUTsd#?$sF0W8%(iemI_1@pC8pIh_q?9IR*@L`{aquxhDZ$-rZG
zGaH-MD(Pai0GAx4Vyavz^}tiFRRpNU&|0aMF7zEdpfSol$LWH9*0BWx8{;zd3VhX|
zmJ+&_HO9EV;Ed^8))-NBYBhL@r`iV_$(AzpTCPy-gew?yR##tR*wJvo;#9rh&=|&d
z*63T7-Dp{3q<HAIQ~^yRVU5PA*3!LY_{Zo<sRV#sak_JB`Wgdg^)*KGD&S0A2`7va
z|EyOE@UOA8QU+d<&8)<~LLhshu}L>5D3)rD^7FZku~jEiE7p5cYhZW(-oA4i<6tlJ
zTsf5~^|EU(P&JAF;uIYAOKNqlmgQ^@%>YZH04~R{AI4Oia;Z{F6`WNLIC4;9R5o4k
z{yDTU-dO|r5!zL&SKX6_HimXNE9yO<ZydbuTLk~X85Jkp3%`tnj>tIJ7_48eh});Y
zhb2l~a8_SqWWCnC^X@xC)a`3*?f+xSsZ>gpR5o4gDnQtV0RU~w)L@*7-AtsGp-f9(
zV{+gx0Tuc0z|?At<9frTxAy-bP`I%%wgE<6H482(K%4p+V+-kGPd(k^fXBPw--DcD
zUt>bWfhWPbQe93N#tfLrF~<&rY1IwQ1drsp+@X*i))<$D%V*eE*Zrw6C<g;=Y*)Rv
zobm=EIF|hY@1!`DV!D7`w-Nh=4Rh>$`BS;-)ZtdpnqC}A8#`;ta6Jw$U(1qglw@kD
zRk>;oF1}G0>}XR>uP9(cL@kHAlLRK*7-&otrUjdEaQawrTB<u&fW|@W*c^|*NY$yZ
zS?Lr}TNiZs=t_DubU6R#!2=uPS#Byg#U2=lg<Nm0)_3T}(;DN-P8ptF7hfI0#*D#V
zR-~csI4K%iW^1)_b-}Dz@N>1ZycRA!i=CxP56q>_@Jv<#I5SmqdZBx2&>bV-_noKC
zo(KLMSLx0K)Z;Hw{SK~rTw28;yV9w2=A2qLvjL%CR<_n#0AnL-=@l@Gx@y_JgGV$b
z!<g(%d6Uy?K&o2C5wfnaadmC6mR^&}=6bU5&5bOK-yww(4*fCJ-ZV_Klj<<5tKg^!
z?BD#KDR?C!EnGV}lLdCkK=<+a23}?^gwiYPWiT-gYFEqE!2di9tb@^0%D~(9Fa|_i
z*aQPbwZ0;7TCNwzZlRRslW_<OMlfW0S2$fTgV#b0vW`;&8=Ga|td%LJH=Qe_(p_B`
zaapa$qh^f5Rj1I+quIK~V5hjMZ&`gsV+4;TFnwVp=8CXw5}ci?K`wA0^}x6p50!$E
zRf9kaI=Y;0jN^?>e7_=&v7bR(xA%`>7)-m^@5Av&emp(@c}$<DpxA2=h&@n@!S%<H
zD1L<EZWOEVlq!gNV<eWx@0zkC!ZfU7rPr~N>sYCEtOP*R6i7780HnjcvLhSAcyqj*
zkR2+v_t^$7>8C&j-OJ0dZu)Bs%~c^0_nnIj@h((?uj3p%#7(krX^Az@p31Ae7+b-r
zy%E+OtDVQIi~Nfy{(#~rye2*t#f>O#LGcNS!J|NQqgaXJEfnvf=p7BBg5n(%e@AiB
z7!a#a{0+rFQPjtRI2pwUC_X~5b{vS$qxcZTfQ>+$hT@AT{)1x3#vsl>aW;y<<3Wr-
zaTba%qZqykh%qSEq4)}l(Gx&yjN*J07onKEDTr-R><U{Q<@ZD}cPkLPp_n-d#2gem
zYz<;36djX6OhvKP|A5#A#irYUn22Ir2Z#wMMoa-Q7R66dJci=*Z9$xc;^gf>d;!I-
z+k@B>#q6md=A-yCiuX|5umgw<D7vPB$f4+%4q_^b-=cU0#U(R9d<(@<JAybC#l|~<
z*bK!JD4s@f?o1Hpqu6(65C@_d&<SEFiXWkP0L7YFAWlOuZ#IZMP`r!ceH1s(0dX6O
z>|7ANC|2wOq6fwC^FX9gY&jpqWE9V!cpk;icLnh)6pz9}Q~pU5ox6kB1;xsRAj&A5
zMIiDhrYr`r1Bw?>{2s-ndxN+f#l=fNT!P~Kr64Xsan3#<8YuSJ7sRJfEZh&oQWU%G
z4`MGAa}EHpD~g>y1!6XesRx4C5yj32ftZV8-(?^UM6u{F5c{Cm{csQqQOrF8#BL~N
z9tmO&iXA=;VkZ<GM}e4%Vyn-9*apR>M}wG%V%#wxCZPBkil<Qg^s^uyLvi}CAkIQ@
z^5;N&0Y&XN5NlD)J|4t;6n{qX9*P^5gV=!L>Jva*kK%+BhzyF)o(N(&iXWkP0L6d0
zKn!s}{0qe=DE`(B;-4tq?g8;v6mMif{29eNIS_wG@y$Glt5AGxC5X#VTv!0H9>te?
zL0o|1OGOatQ0!d-u^)<$Pz)@C_yLN0QPfTXu@=Sb3W)hA{*2;16gN~sY(UXf1Cc|~
zQ3o*<#cxr(g5r`@Aijm-sMR2jMX~W35SyWR0>#rP&Rq-Qd=y8X4B{9RV^0CG35tK6
z3gQzKcb*2~9u&8J9>g6eZu|m>TToncI*1!keCrGlSEKm)7eRaz#aGV+@ii3ZeF?;c
zD9$<y#FtTg;cO6JLb3K75T8d;Jr~3p6tmWWn1|vm6z`(A{>vb4Msd=4AXcIH>jfY_
zKyk~3AZ|yI|0;+QiXAQju@j2lqj&?w<rjmv2F2(4K%`J?z8=IR6dQjH#AYZ)eI3L`
zD27}DVkC-BPz=5l#62kPLvhD9K-`Vumdilgj^c)Ig4lrK>TiL#9>o)vgLoRn!&iWK
z9L4=tf_MnUJy(Ia55*l<gSZ>TE!Tj!9Yy}zAWA5@t_6`pal&;VGAKTKJ&5Hfj=TZH
zF(|%|;)f_QH-gBbSpFRlD^MJB6Nuwc9C0&<qfs2Z0mR`b_PYhdK`0jA3SwUrdwdtf
zA{0w*1F=7fZ=<*o#pT}vaSe*=ZU=D_im7*i*b&7mDE@%rg*!pKjN;kvgZK@Kr|tsr
z42s8o0OIE;9=sdGqbTmZ2gFZM-1S2cKSFWak3f7M#ZC8u_%4cTe+=R~D6Y5<#J5pg
zdOwKEQC$20h)YnM{}T`wp*ZJ35DgS({1n95C{B3@#OWyN4}&-vMd=X`H59o=K@?GB
z9s`j@vHWKsR-icMaS+F&NIe0f3&ofxL5xQ+?B^gxqZsfMh@mL{jp9Ej{{9OP|3dMX
zr$PJ;#UFnO;%yYa{}qTgQ2h285U-%XPub@2Q@45i<Y68^iI~T4z~u28F?rm|K95`7
z=W%c2Jnns*#~r@d4q$oQP9%@pjpXt5<UGE<oX1zB^7sl>9-sEk<I~`Id~_|3kGAFU
zA&)#h1d_*_r0kw@9&a$P8zp&MR>}{48JTaa%Iu6H$7XUs{)nMKmi!mIG;`16xT^lb
z&jW}B#_wHGDtMOx?9@#IKYC}6^l2RFT4jYc@#%-F>`JW<u9-Fq>Joo}We&J-%#`YQ
zaR+M(nNp#^aU9k!hVLVOzZj7EtiB`RA_!Il;c^OSx(eKn2B!nRj;pS%!ZnlobC_cn
zd+Vi(_XXz0&%wZ3#q0xjaDiLxz$FE^@a{-=tbps`mGGCMvu6j~>|Ila$~awc2<Q*X
zF5b=E#CqiN>(;OEr8{@!&A+}KSb2rpxs&n}p#-_oG}K%B7=jlC;K^f_9Jvf0Ov6hz
zTqZhn&`@PL3K4z_ti`1KL+81bVe8}rhXTfL34d}OC@u`7L$~;h^b+?G8-bpqn*Lon
zV0UW=%r|ww;D&I((EW})@-P@c${L*T#L%J5n?EG7`Fk$KSN}WMHE&1Ev6M)MbmLLh
zZal)&je~?6^K+quJA{TC4I#LOb{d;_Ydr28WwBzk_cOb^9-S*@3iU3$obSOa>rq{}
N3?)`<_~Lfw{{eNGiG2V7

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/datasets/dataset/index.doctree b/.doctrees/autoapi/lmflow/datasets/dataset/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..72b03c4c2b070374968bb669aa4d7fde2f37613e
GIT binary patch
literal 104845
zcmeHw3z!^LnP?`NM`rR4B!D4FhgT*;GJymLOduE`2!;d}P;mj9p6;6IuJm+|{m26h
zqPuvBmWyl}5qz?Xh@yg`qPy37-K(g$y1JsGD_#Y0{oK8~y1FY0KJNdY_o+H{`c!ph
z#&wtPV^UpJ=luWq-{(L7IdyZ-9dl-N&BFhJMfJ&Isj`1KS144f1+U%;<`$~?=7d*i
zv<|n{y|#6*H5l~Qa{KDlW-afvf?4oJzE~<3YF?#vT`O3KuQy8N`Y>zXSNHObQnezQ
z7ki4m#lGuWgT?-!r%`H@J>^kwOnG9gT-`TZ$Tf0xuTke0Eo=fUy07x~BP#JP-qNC2
z_D<#ir$(!|B<RnIK@{f%y$tu=L3g=S@mjl^qq$%p*J#vAqs<148OJ!MRH);4`ts#m
zy$(0wx~GzxpjUGX%}M!VG?(8~^TzO<X3*EjjpL(MaZzzaaendm;+*2j;H4Xj)d_E7
zty!sfwT=6#wLKdz-E|QRl>Y+=);ChbHs+d*YHqT$k&<E~Ck4OQSgI7f{lm37jBVA<
z+0D`7te}V7QwT|ETwD!fDXsy^oe2M*4F9i%|JUKTfHbud8O>nOJ((-jTEXgxYN1*7
zwr1jZ5UiqaY>0cI*t@&Y$~B9p7B|Sj?=G$a5(Dr0fg`!`T5htaAR63rRk2jhTz;TY
ztX4945V}mk8w0-7Gr0^Qe&#|d3z>bTMlmzt!Hb1@W~^Gvlq#i0DOWDtP^yft&y=gV
z0$vQ{Duv8Mu2Pz8mU9ibli?K1jMb_WnL=r7%tO)0Amh9WbXITXi<w+Kb8&Ng92;De
z%X=B@<|S8NwtJY$-tzwpm@?r9;=hne9wmM*RuGU^+lniT>p_&}a-IRV&ZoLw><J#Y
zs_12|XFRx`v7}VVmz!W8G<Glsgvgx(f)fjns~sRT&2cgh5Hd67tOu4&PO98mPu{M1
zD5rbzK?4wBEaNa}hqZBHq3V=VgT>`wCg9zcq{eQQ8X$^`h~deC)}kbUA>+mk2HY<O
z%NgItMz3^uwc4Sbye|W?q5>=r-uZh@y%>6VPSP_^k)D3Ao7D-;YXD(atXY$ptX1<K
z7}P>$^gyQ4oR~ZSmM^nr4JWnr9f}{osntj0X&&_NpD2&iCqak7IZf8Qx>wulwNTP0
zIgU)?7Xe%5aY^THX)(VM3~ZN9f%u>0+AT@0Tim4qy~6A0>qO~}0u*KiiyOI7uvWEf
zzFN-KC%wEPknK!)H+er;eC^dYo__GgO$YhEWW<LhxL|BboY2Uob4JcZmsM|0On{DW
z1Fr<)9xP{}1o-P=<#8)mqWsFjxb*<w#RKqPxjH`X)$r=V?N@ETa>rHKt6p*Wjw|uT
z&K<8HKlt?}yRW=z`|b<m!z(Yi?1~)&13X~p2}5Ay><@Yy<F#sYGD2w#OhA0g3E40g
zn(QkT8pYPm0hG0m0OcP9=042DsU07c4*po}0K=rwYT<wp(K)s1KI4NPbnwwOiecY0
zGI?O6I_i&HPsyD*H^!|G)92f+zm^bn5af0KB=}9R$=Us2LKXPI>4`h#jOr_w8eT0|
zR(dObk7->m#XzyYKZw^jh-U~^!R*Nct)O@Ezy=s(%TvQ2B=1wepYKBG7i-*tUKkr#
zr&e&XQ$*H+z!)1$<VuxR@n3^J_|-Tt2}WoRA`r}j-`@Voaw%VGv~DGj_vLDp)*%Rt
z4vf&q$`aRtDLXJi0;1M#0;!R!jeCt&u*%LhVg|kgG3fNE|IRL;?GxzoioMcEKB2O#
ztxxfLU`hf(33l~$Y-wW^^7Cu(;#UCX;~{+$kxU!?!#=hZSNO{St|Kh_AifL;eGw4)
zI~}3HcnpIV3hmMCUc6>%uSo@cl7kG+@32Rzolo=r{Bw{iI2BtSEYN^U-SoHOJ7nHt
ziarO6|I%JF{Vi>{Ae6KCoT_&I4ru4Dc|Oj2{Vx178~*I}=V0@b;}l9CFbd4{>y^J5
zPz<RSxu;UxE|?JS`!9!HsCN0M;ia2``5P$kPIYXYbQ5oSnLaac_R*wJv$KR!@Mt1l
z!}uE}<vKG}KVg-6BhsI$)Laft(WW;s_7qp9*QP^l=pCm`cS5^PwJ9!}Ff7J8LyB|@
zA&Tase7u)Tk=}<bh@0|X#~M3$5C1h3c&Cc=NzzSwMWW!2xFSh^h0GM8xLupXYbKSF
z{wiIMVdeTR@`U*-tw%+}zQ_31iH3cr4I|<-?8nfqQw@vtSNecwNX?!gl+o1;?<G^S
z=dlG<v#+zpE^7Ao6nLj<HgC35Oz2Z^M_kR$*8B{a;M!G;ykb(XEh&2ntJOvz2ve(m
zlrjQPQK{4Md50n3hBicxQ>m@cu2Yp-6Bh!?!OoC6ts_*?)d}yZ)V2#ChJU-T1y!e0
zSYsD;I*|hJRGqFP-L&^w6r6oDEc~;-^aYrTK`ti#rJx%VJ}p>HfuR0_@b>$|<Or52
zX)9Q3sEpIZQZAfVxF1^I8*Z(>RJ^(PS_JbK@FLFyzMJICB!1d@RF1FSgzmDn9PDcV
z`eJKMBCBXgZ1!EiR<)Q8i{E0l?%&PC!t4~)@gRK2YWjDI76gUWFiBx6hJ48{Hq}ym
zXYpOw4P+~oZl)?@v3ZLrVOji3KFJL(!sH^$`40?S!dIDKxdWFV*n<otbzo_eu1ewk
z@zP$elEF1DSflc4Y&D|zKHe?MKYU34J(wy%$TN)oC#gn0talnqXjrTjSxTN{<b6CA
z>LWqk;tb(dIB{ZgiR=ZFbK0Dum#|~{9tt6i_oAyd{!>Ck)6b>8ODG<1yyD!_KSR4t
z-O}nfx5TJA)ww={=3s)(|2CnC?v(IeGN<$#Y(bsUH(6tw9YuoyNk9K_3Ov>+1?P0A
z!OJ`8L0Z(sb992lix7?sz4V7AK^Cya?gWv}gu4&E_>bL#+w{Lm2xmtFJSVdc@%@FJ
zr~h$7$NnvF)8s7uj~L!o{fpu+73aW+11n7uF<Gr4U8X4D$f?5*`%83I7Js>D7paWM
zTu`kvU|pq7t<Lfwa$j<LcOq`1|J*5B5dXOrP#O-IWVdpT-a4$zz^Ym>#1?i2OO*^=
zUx-V_tZToFpw)IL$WluRYeL;n4-j3q#MPpO_#|5|L19dzS<@EV`N|I3Y_hOKUd8ew
ztf5g><&TXseViBZFG!~B7bG^?|6m^AxB*s-3GiS6;zCwj$Zu^0i3z2@7kgac*F?vJ
zck6(6vx0syme4@_2EIx?6d$}q@C~9LGlFvwL5kg6>^&Z_TjLN5j#3t8_m;eUSaZ_~
z7V*n$u97cSYYOe#ESzYIQJE|iw0d0i5|mNFI#{XDB7{(!C0Hf@?l97n3R1)U61+i_
z8%nf*7Oh|aDMCXreT^^ZWlL%#zaR9_rL$shFn6q3E)&906b7=5YCa2b8|XD%3qTgV
z9+vaS1NkOmIEfh2a+!6&_X+$-a_6`xO8A|pHo~2L*zF_OkL->I-E!b)!w2cQNp}J2
zp~7Qj%Z~n|&{5C_^dKvb{^#%~krefJrMiLEiyaF3Dph{Dkl)@wmyQnw^VviGdkMV3
zhF2FaEvi)cJ^mi{gz%t%59_a5hj55{7^%3WW9jU4HqIZ%ft{``$Yh+%Tx_yT>#JqA
zc77Sa>$Z8&f0c4Cyf{Y8o4-$eBi>#y>k$8fv=={y9@QXAcjklspTKkfPvMWvc>Pdx
zLX6j`%~Agl+~VW&e}vDe#fn|?dsg&eG+cfQp8HT|YDR<^t0*F%Rf~vL<tmkG1FHG7
zLek*+Yt}ZPz9d_%ln<<3pCLas;T`-j2K8w2QE4JK?q&Hikl4L2MI#;JyI26}6`=Y?
zy*)h43(6E12Gu!OSfqHSt_*iUa3xtYzuU%|ple?%91Zyc*kpy@3;z%Yax@t_7`W_Y
zvyyK#BjfKnPk=s80ll&f=*cMP6)I?d9{l3$4`Rl*-U`G&1|GK=;PPqUpX3KpRaF(;
zwi*IXzZ+Yu@S{@czYPSL(`SmjBPyNQJ4$>a-tqg;h<Zo=190ho5dLVI?|&F?lLhaO
zz*Vu=e;98dF#n_C5&nFA%l|lk6mop00#TlZXVmfepMXn8VkoYTcaYT8`Cmk=3cwow
zZ}G~*t1pSGFT>R#|L@4(uaLiAC4c{({QVmF`#AocBA2ut6U11o=H`v!;&Hh++NmRw
zOUhH_gKi2bRw^HKgXo0#pp9W4^lE%geUKcHRPI1D@4)>SaOi=4xoii$YcZ%<OHA&d
zlZ||b6)mfcybGEF5sp_8h%xsM*r5!<<A^0@_MSm;vG<>}L$TR=3I$4rI0Yr+@KGwO
zY+%+jSY;GLvsIS(M67a?FaQJiWL6o0DOMSOa;uEjA*;L*tnxm%-H}!19i+3$|Aq(?
ztNb&(B37AP&9qfMU0VrvGtIFR1+A~nCkmnNFb^^X6NPy_x}K#n(tV;6Vx%{RjdVXg
zXA=c|SW@AS6Mb}GsSSu|mU`aWQl;L2M21&SVy=g*xLM8hDbUo>T%T=+Vl&qaii^3v
z$PUG3t|=5S*BTim3-Ng>n{8m*G}vsEO0&(D_(W{>Y-p4wl15;P&BmYHX5)3pW^V?Y
z-3_-pve~?YbT<1YM3~s@*WeYg+2m@bZT1GVAhg!SDY4^Fve9WQt(3Y<RWSWlCJmgK
zsqA!1bVBU(mav_E3qGfIS{ag5@cTsj4s5gm3C%|Lt*z&)wM60cBUZeu_W5CG>S&)I
zwL`JlX9mT^K0j`UVzbW_3fN~sLdiOOmdZ98*fkBd870wdn<YLG+k6`|N@JT5m|~mp
zC%4Ub9kR_^z&76ww>z@UyaTSzdeSLAp-;0M^Ln{$o{v|=Hj}HFw#`Gl%wv$Os=!9<
zxI_jlEVeCL=+rEaF>D5LVI&K?qUs|Y9sDuO9&hK{gfr)5c2_IjdcK)>BvWtH;O5#Q
z-TplW30_DF4`E~YXJ`*>(;Ba@WAlE^h;6#&+!HaUJd#;UD6w|i`pmf^YfDnaBN_2Z
zX6?3ZuyqDQe9&IDA|-YAWMHzYyG|*F4WUN5!?4;!gr_c_hwB9R8Yjzxt`U%kSJ?{Y
z^vEx0Y!^VxuE%LX!+@!}pb?15Am#+cZ^?_x+Th`|o)N_Z)-#q8@epNV<r19ZpsXfb
zH>E<%gpaRh2rN_#$dMcFY9zWO%jLMrpCwDC%9#4an8?c@`^p_j^F3CY2VIruvPB7-
ztnfYfC*s{lIIyk}Wo_gJD-hr2-5C@Y?|!Qtip{$-C_eNcGLOfz;s>KSqr((i{@8lN
z^1$9mlKpNguIj>wy3>xXMT<SIJ}xI{l>m0|WH&7Cur7_2^7<B7x?ZpBb_K@_8|_y3
zZvjNZ^-um@xEaZA>SE!-u#Z`ZU^SzZVW8PdgyFAF!cOLfRzwlRByS(J0uH*WQKA0~
zHW5PaGP<u@fw+tgw#%0)<hUF~ue;HWaxJ4jq*y&;g^9&AW4UIzk=>gsH$8ojP@>YD
zoUGOw**csk(0Em|ipTPhC7Y+!vFp*4DyiIc<>~8L_y_Dm0sfIX#f}m&dVeygL0?nC
zKGuK_Sp1e_{w;!w-S%LW)#oUR$3x;&pa_PGc___<&F`~9v6(Oi#YJXbV25HeVGN2!
zy*UtL(a}UL6?rD9xZVm`WwUCK{QH4|vcNWKkO-7Cc4rXqNo02tQ28S3Sgo4LpnO&p
z>KG$)l^x1gp^JzOWmYD^SVrwo93<uk>R1+N@i^Cn&|N7tlXmDo36HV8k)cEH$jJ1N
zNd&=t(APON#7ERMq=YAj?4z)8>K14UMo!bZZj2Z>Ao@d6>`pr@n`fo4IwH+nKCv=U
zXZDaF4_-=~AGG6aqt62d`t;Jd9j6fgDdBy}4$np-3NImz?5_vMO@PZRWzi91#UkBl
z0zY96LEPhYAmtDmL?<58QL^$!0Nk9+k@!TCImZ*dOujgZz?5VT{^ZFVybdLEwgORB
z!R^>&&Y*HI&hN>7fuC4KPj!PBAoD}q@CcLt$67)WS${AvR`SXPDkCy>W3vP}TZlYX
z=701F9)iSLhG%mFmvg}rmPNfKb*F!>Elvpy!0*GiVL1f<p#`>J%_J6~6OgXH0Cx5c
z4T*{6&`@TC@66gR5<Rpocz7TK|6Krexo|QM-NOZDkscYBsd=wzV(nk1Q6(n`k}Y1?
zEflW~4-5<qv8ay#fEZ?d2G1wN$o0Gvlzh<%1jcsEB;iqYHIJM-Rw2g^;^r{cBTqRE
zXD+0j06Nu-))4`UW}85~Y!Yi8b7eY)%d2%b@eqG!#0bdy)QlNg?DP)6X_E!*b==*?
z|D%=<4#XpeE|Sj8qJjbe^@wmXqkJg~CpJweq7%H*nIoyVK|KW^(hd9TG{6fqz~a4L
zU<d5fR8cg%8JCYZEs^{VQ~4c&e?V^4laGA7r%nsvjXavIL42EPzy+WXC(#Z;9Z7}}
z=F6U1oF5!}Ar;n$)N`j6|AJK2OF|gxS;Vlaz8K%8m?5H6O{Huo<Ys0@oWT^Fk%sYn
zB<UK<QC^#5l)|E32t$1ZVS=41FUQ^~->?bgo1N7s$R^0@)dY%vBK#xffRpYP=*U<f
zsyzNW6^VHY672a#>vLS3;?reG|C^wVst0y@GKIPU8if@K@2MTxsjMi}ML^}fq(lE=
z_(xEAg~hUS>07P8^YqQ{Q0zbaQliv5LipJS%G3$*qUz*dt>6|>p&dtr&wNyur0nhK
zo8PVve^?p+eM!bI6z#>poWF#Y;kkC${3wcsO(;L@BA_T5ihxPcP8}k1wU7@rdt8%_
zz<h>!ZIgoN4?#P^bFnWZp_tE9<FkZVy05`|sZ`?<pv$L8hq`L?ix7m1jB-cn_V+??
zokSktxrqESBO5qAF;69KKTU$05XYAQLH>mh#6A{99I*+dp|dz5*-hfOg-?$|0_hac
z4eI6_qF5dvnuIg_7{oI=5nkG_%axrbOh6Yxqp<D9d#MEUQlQiT=`fXmt_#7n4}jTZ
zgy$llOQUSz(rJ1mMROzxRzfs)0V%c+QrJhKh$c3n)N>Y1BsdjKdy|z+i(&6RMlr`Z
z=*%(LgiiKsE6^AY&z8M!hbF%hb6fk+e}nJIsW30VD85Si{3CT7wO<BS*iNM8JJ{j)
zX)!lvk>V>sJyPxRraK_d6cgagVB7|2Hb2OSFC0Kk#=15y%QfIofiq7naC)}i38^gg
zlFA2tZ{eFz{#hxS7>g1~+-B8}LAQN{Zxo%7%7rXsNwq2txsK(wt>fLIKU0x{HEI8L
zXc}3mHt*v5kR6I`nTJ7fS?2klb||)_J%xh73|R-1MhzdhnRvcV$4X2LjGFo)v|tI-
z5Jl+CX-tV;B#rrc7;=)DG@s-z$>zK7LsOZ_l*K7(k;MOCyYcYdP--&^M&(vUaFDEY
zyYYAjT&;D3&+63rg$2=fpVMNsXK@&$kocE)MT*tPRoY@TVrq`SF7eH5BAU!Xu5D)_
z`Sez=UQ2Et#XQ+y9TyD<>nL86YPJNQ#;!Iy+QtQ<6Ji?+VcQs5Z=*=M7?u-6$BB*P
zDro9xBu}$Lu^CAQ#bqqp>`-h*l0pF^86u=)Bfg=sQP(|n@*A?22G&i3wOqulC078m
zt(3?`Y~@@Sa7tS_3Qdi+Qm6;Fl=x=IQWn5c4lrV;wv@bs4lLyX90akHH{cbql;mor
zE#+3m(kw<|u>O(EVzLuoh!<5q-Nox~heN%099b@kbcmI;dVG01@|(t&k=LaP9%fR(
znW|4M94+Z>q7!0COJPfzG|IeJ^q1Ixz8ji48qkl}q1X&4gW_U9KW&F%GoTcTA<D!D
zsqChKQPW^Ij|tgLu4M3u67t!Qe5?2z8Z34Ol0-6MKi>cYPia4&fu=_L8Bq>yN3li7
zj+VfV28{Tr?I`bntDNrjoFdNWaWKS={u-}{9VJ&YZAVYTa{`hX&y(7Po8<haPGik(
z+Dp_-9yL#xN&1eojmz*+r%~n-(Fw7Ql}>HrDWbo`w($gL>S!C!utTxgMh3;jHonLX
z#bz5R6oYNV2P)Ha?Xtv_21ZSTZJf_-Bau(DL6n$94B~tkZAybU22G6yk*Wc=g7{j<
z3Rb`hE@Wg*Z3TG;9azDGI09k?Z^A2L1<BP+Tfwtg6bXfD@ZHWt>B;r&tfAU<$8qF*
zy|<`!yaU-xgGl{dX(jJulE4|6DuO&LIw4lF4q`P8%6BwZK8*}!C6pLOl<?00`IJid
z5HvL^p-yeCjQDy;8S9{oZ)Vg>t&F?_&O_a3I)x;k$B__a`~qGPWh7TKt&E$p*|Jy3
zW(U``*F^e>U05Y(n*(E=cC3N@0z0-osc+X4V5koE@!wMYo?^V`tV^YS-xZw@_1g=A
z^h&516$?hQ2k9Qr*wfW{euh=y8v2_Cbr$&<P9?J@OXMP&{4JnEN=@#WAEU{FXH1D@
z(p|DXq|<vrr~iSmB(+ZS4!B?#b-I^IyeX@_5(h*SdKF#~g(g=st<dLX^F=Sehs+|g
z@G<O(TqBWgZ))daqy2ZR=0`{SWGba+ZGxrI7sYbA;gPfhTYyPw+?&b{Y!;mmJ8+;=
zJ7DN<8tlLexLY?6%xn@Qo)MEU3?xZu60SkL5ob#(%UdQUbL((?Jn|qhAwzKh48>`T
zXGTMz7WWTdfNE3El*UL+c7{yV@l)>@{b2ca6AoIfdGIyz1}K=Qa8KGyili^3iBeS;
zNp&g)!x9x9uZA&=i!mvi6V_dkE7bu#_=q+fXj`YBA1RjLP<k8^9Fog0FNn|9CukwD
z-k%EVn{=$FqSoT|fS7Bo1t%b@wYaTasR=uUo6AyKpYaYrk@}2x;uY~3<Z7mU#u;O^
z>O{6s$|rLbZHq-jw;g+kA<@+!qM3ya93vULiam})ixm4sbM7vbJ5snGu?n|iy8mI(
z3Gott9`+Ke+YEq3y4s%Pg?bzIingO~E`NoOP~MOO{IbB!WOJ<hkROAlFcg=4$ltL;
zv8Bft6qoeakL^%w=`jXHqhB}DkJ2u>5t%-ebbQ_lTeaPZb|O<S3AQ4WY7uk~cnrd`
zghRR;abT$)`Kq|{X$#vd|0wBS-z=YaAM5G`wm{6c?bBQVO+f;6u3>34m-uk)^aeXD
z+a65{t0U5++sC@Yj<bzE+YR*L^|~xe#eYgkzS0iQMk5L@A&u;>`%gf1rn+I(U`(eW
zL>7f#(ruU{V~JiQGX4-~OUlUj0Pw^Z87oeU2gFD5fLNd!jf4LjjKf3BVzrHfhc2jj
zxrUd?@m-Vrdv*u_?xLB*orbjoB9KhdL`vSRR>@QCkQ)Eq0b{v^wS=*(f+Ap^hoKu{
zv|UnH<2TH&@E~xnC<CG&<fB8OaWb7Jj||*Mc59x1vHuAp89c6iJ-s82jfLCSkXyKQ
zl)jEv_}>Y8yaU<O;f0rNzx1U$vZoK9P8-vw?D;0;`6m52d{#pAd{92J0@>6C$fjX8
zARP8}0|y6KbfInZ8geHhx5i1DNU3;75~)zPMBWL4aThcOp@{C18f+&h!542F(Aj=M
zl;x5WHdd2oF>K~!X6%jaP-APTLye`O4)G^DNKW7B(RCwd3tUNRe=IGv-wM?Ja4fZz
zY3cDcs&br8f=y$nbFf*CYaQGX3?R3!QPBQe612tC`7XfuGX!VYDLikqT!kzdIYLO;
zml_<TJc(m)h!>}G^+Lh}Q;m%UGeeC9BUxkkhpAp0s}^hkVKo?=Q2zuu*{Q0}{v$|1
znl;x1;;AH*Rn_lpK-}*V;u_RXk%v>0A`Xk#n?V@JaETx&)|K7gq}>OD_w>#8Dbn(9
zNya3k<?S$_e<cIbBkGW2@|CSL*o4v`KFYxqm9ps&$@UKpP?W7S*yNj}i9*(H6h5v_
z^oP{)B@1macW6L@0PhlT{BHm_?q!282Z?xxM30Cz7d-{d<ZaE|Com1%CvbJ{pO|X!
z_rTy+L(6tsPQMP6VOb<>Liu9nzS0C)$t;rG3ExExILJL*NhY3`Y8>|o2zI0tDMm9%
zO}o8`BE{$pK9o#~vB?8rDegMRMHsH+(6381^!Ew~_HigfN2F*BVQ-=g9m9^dBpW(5
zQ9{_?!jy+A7dvCcpA!jd2mQIqy+A!!h1qfSmZaUEB%vKYYvpEaB+gp#k5wT*04Hj&
z3>?{kia%)Fx?PzYWZ;Wwc&ZAXd7>V3QmW8U8)gvGaGnW~CPU~qs7(@ri;P1`JRIh{
zRX-cdSG#5}XeT5@UjW57rS9IHh>>ic?c1SIc%LoaQ<uW*qOP2g@Bql|TS<rdKHDB5
z0$g+1FDZeKgy7lTDcfI*a8Lq`(Y)eNPD{!qj%M<PaztN9G9uwmKLF%=l#tI(0mU9+
z6H0MAlN5V|OuB<W(PeyQPcitXBoyPP82lp{hJK0x-b*FC9|F2OP3fW?@S&Vp!0f1U
zN3y(Ykq+0+LMFrr&qau1&M9E{#5|R_Elq-(5XT3BAd8@7y9h@SM{GiA=q!#%c9S@+
zmnRRv`CQ^8iGk>Lc{q^DWLU#z6Ab#ASq(o1qZ-Y9o{@xbTq;i^G}EOL@1>H;4+DAD
zlMZ#0s>q;CDtCt9I!PtMbCF8Z`2-9Z(<3R5*CxSA$m2(V6t5tpu#ZBKM{Gi==PZv%
za4L`XCdzyZE$Z#Wa@Y|&2BIW*6=SD%>&dh6RT^Kt51JfAJ3&4c-9GT^qlq}N<{^Gd
z`oA#={dvfDE0YR1s{zp0Lq1T}x>p_};T}sjv!pN${2=!s|0Bw3h`DrkG-&BE+Fj__
z5?EQG_%OquX}<mt^u8P4Rb~_SCn66Er0yk>scSvnOQrRn0G_{_bf{}Ruh8Sf<*`cS
zzU1})4&h)IQRbHrcDw};#To2?snrMJs}yTsQVM5rs<{HTI#|TRE51xdV+XD%?k5ES
z>{_VAHAu5}I6zSp7n|JW(1a;2YCr<T{W$b471urvcIF75^zr(UDLHfGH!uv>Iv9x%
zDOLDfyWz7s7`~Ro$Dt3N)k9lqCy!pt>tL4Q6{&+ES2JA)vprANzOu?%TKlSK`lTCD
z8+}^4`pGbij%#C14fiNYWMYeBv{O#b4t=YJ4kgJKYvvIeuwq$hoQ8qBs)2O+Zp~St
zM!He4o}q&9Y$D9-8Q7sBLjn?!{>Q-WX+=Ye!*u2LIQl@2`6Z%LQ-zIV36TLat>FAP
zo`H!ln0Z94ct~awFnrVNC-i4AsfX*WW3rZ*TuWG@^U&Z#d{$y|HF>?m5|jOQbZsRR
z4GQ1A#N_pMShn&B3QM=02}zT##N?fJoNe^E+dv;auVQlp{!^;N2kh`{G@|el(#ZaL
zq{PIaY|~I;!Vljaju;tprHRBcQfYEB&^KkJ$=9H%vC<@J;&{P{C?5|a9<4ih3__%}
zjQL<(qbCpz`qAX~5t#!0$SQKGmf?C_ADy{9gN<Uc-e<FEmr_MS&k)b3#U!bxLuWX4
z-*#$rid#qAlexYwm3h_@-KWQ!cBHJ5J^m&L!>@tD=8SHdNykD!B|-M;34uVofe5Y6
zBF?`eINOJ*Ogga1e>gN@lMa+QTgZY4DcORrL+?*P6PhisBgG~i|ChbagY_wybZi4w
zxK27GLL}^Zrrq$_q=Oq0J`SCU?I#^C1r%xV{mbx*OghNbOiwyCHLA%=@7wOcWSw?A
zx#$<zvCvWrk^2M~wgZ#@mIk8=)K(_}y3jfX#Ct?1L}q>jMnnsUlkUJY^fwJYpXYIg
zIhD+=ULqH9^<zMXl&=0TO1){jmEa-w@S^%vvL4OQ<Q(B1e*|Wz$XGJHj{SZi2&U;(
zjAMTfHZnW*tN%(Z5XEVg{6q)+A_#8(f>WzTNNJ{0uotNjB_*JuW=*RT#YeKKRu5{a
z7bP8H-BV04A@2vv$M`k=tJIF-&$fU?eHy;a|1%Zz@MrwrHh|T~*AWa44h(G1Ode<y
zs})i!z)Oq8S5YX*v39_#G&0pOs2wQP`8QG$L|6I(|JA2DGnp0eMVM=-21P39v^ABs
zaa!TOA)1QfLH>?PFolU$AQ9$;fQTZ)Hb6M+xB}plX<*3;z~hvJ$CZw6OE2*^iS$|t
zD}M&2=u^<x7*_sQFZ+!x^Za&Na{nC2{f$_1D}@<TQuq7gBxsA3#eWB!|B>RX6*B%8
zFZG?h)E1b9U8$mSzW{pvHckc<5eOH1k_{0d^Q6@%L;A79Qs1RX#v}yfSs2hFXqjL#
ziA^XC^m#~_Y|082lI?jyHa!FmH&G^&*yLH#L?NrS)OS-V0z4<+{HHGUy)f0_e+`3w
zp#%4ox?OBS`C{k3(gayajgVxee?Bh!vnqVG*7ucZ2C}QGOTe(BrL4rFZKU?x-b9gO
zbOB#WCdb(1D<L`dv&!JrYTtvY20t4IPhr?cq6{9<qOpU$i86Q$H~xx@M6YLfNk?&Q
z+qB-fopjHGvd%h@vD)297^I{CcXcC=-a=H(v`HjQ4)A``o86#Fj_>M`7`U>WVoiB&
zZ=)D)6qAoT=m^UK;smu_8cfLo{uFxu3uuxc3-~VfI?tX>Nfz*1zzo+cphO5WI=hy3
zwHrT61M;*3ABfJ9_G!SM0FE>b_*1+hX+U!I`;Z32@AW1+tjQ$}NInAG=>f5r2Bg@<
zr2#1znFb`krP1g=$`)-prU83ICq!tLfzWi82Q+m!&51xlCvzfDVi!pS&H_rLOayL#
zrm=}Yf{iH=hz}wwe7lyRJ)O<CV$KB0a3?rkE>2;0nBNikGO?N5{*n}^bA>E8WCD4p
zzw1~;_#A=ol$gPnK=-a@kfTD=le>mo%^;_FS~Hjf9H|)`#VcY4$<<7o!QqMAWFixI
zV!L`P=&<8ekfRYR@)<{{xMAPSqh~&b4yc<$piwJGuBE`<$fS<5Ob=QdE#N-U39*11
zA{KB_2uo(JwSu6;hql}<pqg~edT7g=peYyz_llZ#*<snLMJOyiB2Z`7y#9+78y%vk
zX{1S4QS*o$XIn+h!v^~B5}MeGnn&&MY&4?q64J>2dZePpB&5^eRp;@eP*4ZW9#vu%
z@u)SRX-bd!&(PH9QHcfM-gJS2UeuFrK*!Wzd^LMgL$S<Y`)V0=TPPDK^_H|xvyQe!
zXsRCJY9Onb`3+)hURn#T=O4b+cKF~rgz>Ab1InJANQ?M=L58gc`sz<Gvs>R*CL$}_
zZg3nl3h&#*d+I`Zyiy{6WWILSnJ8MzNRRq{1IJxPXM_Ojjw>t}Y>xrLL%{%1jvmLN
z#Dp=s+nA)rJe5MbBMELI3Ooxb@*+YJ`(TvR0ck_2Xs5i)7E#JCC2;g~(eK6fxA7#j
z<J;fzWHkCl7`&HCGS5anWGQ_vh9W$^4L)#`%$q`h?aXA7iSS$`)3nKriKXe0l*!wY
zU?pVo9Hd1+Xkj0QB9o*IrJu7*5}H$)w6~GzEGowf9dKj$Ec6QP@;Ne}wQDmD{W<CT
z+Ma$OHoY-^h~JV4|0@X-xHtQ2Y!u_o{;)Q|eJ2rlm{)p&NPw>UcrTUipN~v`ob;%B
zC0-?!U>DraLP*#Zh}i*z9d8HHB%^07L$S95SG4zqk^**3R6Giy+E)qH>~IwAC2hXs
z(1vL*YC!_+Jx|wO`%qX>>a$GQ`CNEP@-U|XJ6!WH5+O3L-sPHy;SoC@h(7xE>F5^&
zjx-N*F<y~847r-=Jj_|+UL(uCqnIckb85SMjM8z(d6`v8XMp8~ZQ>|_Yv9CodXQ2m
zLo-il!dJ$$d(wyGfF-JJ(>2a9V{@J8gh<xqk&I2!+Eqh`#aY4pf>+OHYvicQx_Y49
z0)DuHLXl$hgyT6tFJ+VTYXsJ&jE@@gCpzGmKuLSP@r{5JOeb8<hCXaX-@1vALDA2-
zel?eru2(BWy2lR1cFuKzu{>ypVms%WLNRbV{9*P_Sz$P`JD3AoVp>EV8gMML*J!a1
zz<0)Q-K46x@O<6@#D0ZuP>kvLOqMu-S2Zl=^pnIel76}x#*i}o^b9mLrk^yk#?w&C
zcp6GUH=2&R9G&_fGpTS+N43vJiAX&@a{m>B%3Z2%byfZ=d8l7tqnOBj6&?V=KQc_}
zkfZe$hAfgkV}F6xNwkB|5f$^**SYb~Z%RM2bh*vXgh#AocCJA#ErizQEK8agHbZ)|
z(_D#Ruay|Y>nZX{iD3!W^BmwPF)V4*=g@}5uqY1+V%S&UFn%t?-9CI4!~QfRr&AKc
zUJgTcjbSB1B+@9k#;{z?xx(p;Y#+mp1CBI?E#ei4Vae4@$FSQJ6UV=*oG6yvoFJCv
z06LComl?5U>v-DktD_8Wa$943CQZ(3fm0oawg*KgMBvIDhqk5;r(xk}g~25XcA7o8
z#5Cf`_XAZ@dh++6{+RY?8M(<l`brD>QO{mRQ+EU7-1K_(zY_A&Zc?Fm_D8T$jAuVZ
z^|G8}X+8W2#?IP$`lC(FxE7ipN;P~u5k(ZwzYn?i6=;3rJpWUqH@oXoJpX=)f$lKE
z@rvU43D#e7fTMVR(&h^eZJ6grc}d{;n>dV*LL0P(?UHxQ^M9DV&K>5Ic>X2FxO;wy
z5HZBJxO#qW2KYd9Mz;6-gMcIT{Oj<Fcz$v<)1H3>J>wW0KJ3*dYw!)cL|%WRi`Q2=
z@7UuniC}B;_KQMqaT^o4FU`w~fGZt)`E8;TB3!q2?Bz`zPJ@>}Ms<@a`^+9*Vj1!9
z=Kw`gdiXKa6O)IJaFKiW<6_W`diGn<lx=0)nqJR-qmYpVo_#+yit+5bE|SVi_W^qu
zmW}3qKC`b_$`{E&<F#s;?Hta*w>i@W0GD(Evu&KPwS(Tds&~|fGz+an+)|>B!zf{I
zhSo<e>i7%Nn{(9Bk{IaD$`W-Dtl#7SM~ON}n?nw5Sk!^CoFMAB9f#3`HVLAR8hf2P
z>nVvkz6L;C^9>Rq;u!96jXJoA;senc**@y{G2lp}j(@@{5_OQPnT|TnMh{(p^CGhN
zRl-Cu#|D>}Lv-D7#6i4`=nh;7rei}3Wec>?a1#`|D~-yWV~;?r@jD^oM5ONRIMy(A
zI1RCemSh)njwwo&I7T9k=YSq5BaOAd4`ZYe;vkPRwD3rv9E~#WMbq_b#;56xGS0P%
zSl9!w3e|J4QB0JPA#-;z*tCJe@u6<o#sUlV23)tox21BuG||WvkNqHW>yM!Ik@MJj
z(wno#en?`V8~QN6lso~!dba}{#bc8;FL7wYJT}Tj0+0Pc9L8DDCV|J^%wFe?Zc04%
zVF2Rlu_Z#pynf2nW23s?vG=Z7<Ue&nsdKcw>wZ5VOI`N^ctu<{xteL$y&0Vyp8Au>
zYY)13ZS1vUhfPuh3_qxH!D0=A%bv?e!7YlsA#KTLLsF@0vSU@tr$r}3l0MzBqc(MT
z1FW!#l3-FgLRO927O#9?AlGg-I(@|}PXbQ1^kS%%!9<Q}h+7t4NKZQ(7J_6>j!5((
zA<JLGKvRY+y~o9bEK-blw6ZjeXf#myG+OLWF;SS_KxG9CL^YGTiiI5?3sjE9MlpfP
z3KB4&{-=#mj*&gI4Nv;TyWEZ_ccdXXBhe_8pyY3nGp9l8BNvpsi1cQsv$A;cC5eIV
z!os{!f)axD1_wAwP(s?Qb7;eY5|n`iLCIq{jAhU!K~S=ky*{NuNezIw1|<?9;!3{l
z8kBI=<^$0g**+FI1US-I<PCU5Vi9sR)3M0LdTwvBIOL>uaR~O<aSSp<@ER+Xz3gPJ
zQ5?Z2WPOHxIvsyx&daRz>($CyV?LpqB0={?(;w27{AEZgbxn2*O5P<pA(HfDBq%w_
z1$h=&Y1@Jq>J8Z|+NS1y@rp2-@{SznCk2kC*u?rRg^xi~u$X#f>IjF{HKJ6ee%=nn
zR+-A6xK!DG!w$t(naZGO6ig7|{m2SgHR973;r)QHM;BZNe(GjRMV*I)Y`gl0N&ot{
zF^)EApGi7<>5E)g_)jU9^Il*JS*+vegQkvhd7K@JO)eP}7r9((hhmdU1|_Xro@0fq
zGI#pq@=U^>1ag^x(pOr?Ypt_qP(ItM%<;O;;{pm-QZ{pTD7NDj5{#v4hvFbLKTyZg
zOI;x!5dSGhb<hs)C(v0=lbECG&B@7Xt&y!y<Z6vqHLDG;-a3S(QcuIX5DC)3CNwDa
z*CS^;n0)LsOibtV`5p-(%#%`yT4YjsH`vaUlhS*EO~yPr4Pp3%bfJo1bTaxRM(S^8
zUducgRW`+Eu7r3^`)+ol3MVgPY?Z;FLL8RV{zWNo4_oC;l{643u`T&~i1<USDIlKa
zgDvWL6RhrEy(Llnd=eD-rrpnCqnLT~wL?Q92qT}bgea9BBYsuU%U~3n0m|YB%JDU*
zDDY}J-S?Hs<;<v;L9r{SpSk3prER-$;A$fZM+$ss+ra`27z+fn&tn?ifMa)+;8k+t
znMB-RdxXCajlw5a;5{`R5^r+Qm??>ZXHfW_Bt7bTgu9WC+_cgk67T0j;B?m)5_F&K
z21Y0-MIhqV^D53v#W{{@%GPph^H=Dkw2n<E@eh%X{m|5wfkfIXDH76#(%#N5B}GE`
z^=-m0D`LKqA|Y+Q;n0SqNKoAqq)48|q5lms*S)@zn>(Ian38pyZNLuK6p2KLOl<$d
zHATW54j+il$o46cmjaG7Me;JdA}JDbHPb1Q)9bm3$#SwJ$tmrVB-m%iX_8e8x5@~d
zZrE54Ys_AOTta%}j7`SWNH1-|tpxum)xHYkS1qWnU5@FIJ)#pLMZb)sN0ysWW^P1F
zi1gA|*&}Vb)K%f*fGatKTLkpPiI3MpQ^yJBVLKGt1d~B=nPA>yhhm#5Fen<u=75|~
zni-b1@R_9I!&b<uZ%a18L?~&u={`u<ql=_9VL)l18zU9<P~zht>0eKLK&gbPRVF!v
z?|yp)E{gPrq~jA-I$E)RoTB5BhWnu%icK0B6c=gyxgCm48X1(d(%5sn%`B-bojz&o
z0-VBOjtf6sD81S`UTeC8QQ9TlvB3_-mhMO(HCyda9Hd4|cg#Lx6W2Aol$v&l9Zrfg
z2OVTW^I?BIlIAe^%4tY*C|Lv&i<lD~61hmCV+?FT%0$N`Fv^(d;Nb{Qb%-JX9)z`U
z>L5Q4x9gV}a~GLaYMbmh=}HpIWh(M8=nQ^<UCn=pxt`n0Qu<zFl|I!%sY#FLaR9Go
zO^r#9BYE)V;UqZntjC+NQB2n3#Y00ClI;MmF0&n#=ESI1(+=In<OgM9l;X|_jtews
zI6Y1YDU{3TA^kGXuH`V4Cy=|)qRfX9(URpXJ^+ovISag}M%?lCBb2k4H5+8-e$rzo
zXMx<{HnN-ikaGI@5I8-W3dt$YSs)aboW<r8OyP*8YAeU}tt8_T`HC*2;Wr2k?SxSB
z6{HQNyPa1`zJl=TJq~b`d<AK9k3$=luRujjkgxbHnw~d78=9{;A}6Z!t<+`NPE%Hn
z)Qyc|42(YnwE*a#_+BhxZCD1JpW8<&m9}Rp-ah^TNB+4-{JvwZRDm5{#KzOE3ex<m
zc~&BaZIi5HP6!<tWI|?YX%w`{X2rb*25g~DOC@$XDG}qCHOoMwur<SbsjOKaYU1&v
zN8OsGIHhrG2no9wvb-$9cCm41rDHBPb7`APOLu9K(Fsd87wLNmp|71BilrlMC{LU%
z9pU3~4saApN7^iNXu~WWT8spit_O!MEuDQRZ09|X$#_C@N;0!I!8lwqvl1baQ(HE>
z9Rc`fj6!pEee~@!vv&ZFG&B1ayds%day8SL*{zKlnD#6zO~X3%WVw`V>-}l%a<t*z
zJI>bb4)-`nQ7_%#jbz5k)m#J8wx<nmf;?`qIW|^?yQ9@=8E#%xYkE%E-Nj)L+~`Cf
zNOSTauvN9#x^-~O_&y*yA<}s)NT>e2q;oqNEOT>OUU<2=L-wkcXX+QPa(XCtBq1IZ
zV@RAg{u^lOm^Xgh4#k$KV^Cc3M^D+I*m5omibla6BA=9c(abS>Dk=GMD|B^^ktkco
z;H1qS|Aa6}7jJ96E&;t4tg<PQHDAY|xa8|r*rC|+bqU6@&JM*vcC>um9I$Cz@5r^3
zlPz{gDKd6+n7WTs<t>!({}d4kJ>h?}$^4qkN*{ZX4^I4+1H9ZmK-=eYE`_F!CgqiO
zC^nPAptzWnvK@*|EEp8cq;#Wmi?b;7p`_zYR@f?o6B!f=CapcWfp92+JxM_9w^|2m
zwI>XUi#@r|4#j3q5{%`eb|?;lquG;g42QVR(H~M!{*N723X4Jq+L1-+bP4f0_CeXK
z$P>`i(Te=o4#j3g7!(&P@~j<-O&Ayy&59&g(C%MtbJ3}72*Dt&3F!fx5-lMnp!A8>
z@mfs?gW_UBPPap`nUDlyIo}S&L25J;61#-B%MK@n`JjVLXg=((N0tyxldox5LOhli
zB1dMO=EX#bW@It(K8V9oE+#gBWyZzC=$w}?E50BO@n|*mv6wo#pG{}LkbYv0+(c09
zB^%?FYU<Ok#5{z$oQQ>4Q1*+N3QR)hDuoPAM)A7tHR@uOCPho>^!~&uwW?)PmmXK(
z&~9c;VQA5M>m#}JcxMtE`O;$x8^x>)j1LWoX+2pKAXC)!nSwW#YnB`5!K_2BNKV$O
zxSyc}RqgO8;0CN57U=eKxQ4t7uf1;8y#j7C<UGRV0TOAnT^7i@U_%OeJH#ZkCICg&
z3Ka;hgTA(eX?O$6uth6n5#XbVc*GU~9)d>UMF6~~rj+9CS!hW!gEIL6(xbizaFl_&
zJO(sAliKmOA!PJ`J)|9c!4q+CSqwP06Q1$`bk>O+>61xDDi#LTAkDu`Xl^HsvM@l}
zP(Imtt1N*M-oDQPj<PU7+Puf14O<vMgV8Q8J{vy&M~%v^RXFt9q0JFo0-cx9W#3Ly
zwgfsC8^sv9$3VqNrkk0pj#fa0Cz78-$FHFhYj8yRXU<LS9*ijcFV2*PzM=|SY1$9i
zWf9IaSY{&Eso#KH5&_8G(hO<LTVPBVXy_^ibuf`EFoSv$GzuG3yqC(Lu0;vYkREk|
znqpP;{16g$$!4pn2;0SwZb`{dZbMVIm*(^GB;yn2a}X)Ki%{535XF3wHk2#Q=96%8
zC4r-VFQ(w2zD~q#tGc1sV{sU;No&k6fzr%ZZH}TdwIp2N--@jey6E4F#Cxf9W(Yaz
zQ%YZ~+KCi8b0~y_T|}79AZ!<%aVjN1GUIeZT4b&=X`4%pxjV_|gvOkT^u3eN*G>*a
zV@Mmy6K9Pfe54v<Z=(bpM9YqIz>NhQ=(rOE94F(@Wx!!ai&Y}<DcY1)B76zP;aZ6x
z5h827XS-G+@M$z3h(7xEl?Xoo9BC!OGk8TR5y;g{S0bEKfJt3eC|II{{f67sA?ST~
zT!pYw?{BbFDU|YFeI!#Z)f>hwV+-{b+<Js>ORMl3;Erl|btQK!DEO7=gb3NzNI}5@
z4ReB(zK&;jVbIr-$9=2|J|yv5j^(H|wy{`Oj~78x$JF~8I}}^$ok4L)y>GNbu_aL%
z6peb_C^3|D(UJiAP||U`6}CDFPqu(Za7dedKaX%o7f%P4>UK-TJ+!(!&(MGP^++<;
zB>2;i%w5iNq9JWJr*b8dkyP%#gW{%4<<@~-##C<D*6>8G$Uy5zN7J}l(eM6(nUuC^
z+{}ew&t<?A-Lb$8M*A0~2E5K$1Joq#Hbne2tf?_c+rDmFJpsddD8^fopvZHyhp|yi
z>SA<gh?!WDv_NI0Ioa*R-OwE(#Q*TLtZ<^ZVFbAvB|gt4*mDZ{9O;!NM2;{Lq1n+q
z1tgJ|P+m6j9Ogs@2lQ|vp0E_cgU~3PLcn{fSB|%0p`7dmD1r}=9`zJL58)@*1@=n{
z{1-z2bvvutK?KOAJ_yHU>T_-<%;P9`)`=YHcan@$OoJ{&x_*n$)lM2^8bsPqUfB7k
z<Xs8>?stHrOoK?9dmY-aX%HHL1k<4HIP^C|8x+X!`gd3|*ffY6K|Xt!l7j37C%8|8
zBtj%Yy~=eO#N7}dh|b9N)1Z?8M>-8U8L!ARh+NI|G-yjbS1C0PWZ}a|`8|myLF?O1
zg4CWno&v2^dmCh-!`@uE1czrR71zee(0sKmw<G#ZX&1Hue^isJtGVMuXhd{E#O&pf
ziO_MeBOv(clcF{+hbBn#)vm>8wVQIZR|!NDmrq{~O&zC7qjo5^sS<<YGF6(iL$OVj
z7!-Y~l%&@2bynD_bxbx@A~>X-D&0glqzkZhs+54(Z?_KETByXJxD+bA-wwrAsFYwV
zpRhx55FD*g$y)39IXkQr#Y%Lbx<gV0E;PTr6cDg^Z+ty6zctyjX_((0&F8n80XNTX
zC5n;R?U^v-l(XA^fu_dUt?oYf+*VA;RfMB6+m~Y~wS`GS+nMbtmwPolpBIjJq}^tS
zv(7+Wj8YUMs*+q^mrmf<3=<yQ=oUkjC3Tkj$2f>zi9t+BXg6@8duYc|q4U~ZW!KP-
z^MQ|E*QEBL-5S7=hIS|56$$Og)l7$WFDj1l4cJ*-9y9H}Op}@L8pUd1T`SnqF3O7m
z(s97IDW>CuvEMN<-SEID_PJD#t-#ifBf>4B6A}?te{UngKN5&0jtF-_Q^$xfYlmWs
z2pJTYh;YIV#pc)<l=KndYpk$UqdJWd;f;htx_!4sgb9d!mvz9_h>$^Xi3s0ohhmEe
z6O83!b|?;l;~o)y#tth*L`Vm!`!QAELJ{F5fI#Ai&}7f1AtGGOBf>bdZVnD5wvpg)
z6O2A(aQIKq)EFGbIu#x#u4yxjXvkQ_XzL6nDs4l?P4<xS;^z1`l$TtT%Zr+75k97A
zDt+az7=}EU(oKyjglZ^R!@>PR3~ow7$-a}^LrIPb3nl;5HI(H1;iK18uze`G3UH*M
z<Z8V7(j53hu8u?~sUs6&h0hM2ScRVofc&m?1U>n3u3m3>Y9Ia=5!v-(02{;r$W___
z?1WyUxCDzPs)c6R+ki&}d6h=JMKl=k;g86N-`S(J*R_i4uES&KiUY+J#rbrA7>Dg#
zt0uq)i{}Q%(0F%4fo?UZ^9va5iO>x~V6P1GSNH`W5rQ;Ffdn!XuM9zNz1pnhVTa`U
z;AkjJt<=jo-1v}%JrLtH4~TuRH5l|Xs*PN^wR2W5z~6x*On1&|;>fy!h52T!21CxW
zzIF%m`JHm9;<a`Mb7A`m)|J&;!IDBX-)xl1^<g+n4~bZS&kw&WS8l>bBM!IdXBC6q
zYO^ufY_xV47wr!E$GwVI%QdPs=&2tD4>f9V$`kyUSF2VVfT}lE+TRMgHo{S{6oo9n
z-2*QSu%FFZ8GbEjRP))<Tz(Het+$F_3zm*S2Nn2Wg>w7WU_sr>H{h_i?0%g6wr+)!
zkQ%kpWHw)&m>`_p9ju^tys}5`WcQUC#VnCd7zRLM4NzRkjrlc?d?KUl?e&0=vx9+R
zuH4A7pFP16Z+{8s*T^-Rb$Pd^;EguNp}&Ruay8r%1aH*5+yva|2S(&Q91P-DssOdJ
zUL{{Gz+0`Lzu6euaQ23J39gUVA7;H;ty;?#VRKR0s{!p1)jXWCyNJ?U>7|GC(h63F
zo`kyaeNeezA?;V<z*`@$KL|Ar7ULK|>gq)xQ5nY63Ko=emGNe7+yjai;P-5=(h82M
zdHHIskS%zVATt1_L<rjpnm7i-1d5ajToj7E!NOd#QO(L*!R!(U+JZuJVlpd)5eQ3e
zz_u%|R>_r-b&JRyBId|_d26!nH4D}3{s|PN#oqqOYTbh}F4>YuNxqTYTdJ3+urA6L
zO7+|*oJS_G?BhxT3O64HgXNY)<Dk<8?6oX@^r7AYjbBc_5&_HH9@&lryuscI!18&u
z+&=MhXIHQUH4P-EG7f@RE=`mgtzC=P1`8*>$t>sxd_STBg<Ak*KHeG40q&?2!Lh|g
zW3oQ7aU<NW4^JL|*>`2QS{nylT83{D1mI0LIR_3_1Kn!{bK(AmlTSMX=((^qmJfB0
zcVzdc@b1xOp~&v>YQrV3F-Bm3DcD$SOq2mwe<L>vT2yEhTRZ0lD?uhFvO+n<0O05z
zl#oNg;`#wNt8IU_SQ;;u;eU|d?s64He?IvNO_qIz=GNdS<*PK=ay6f0q+OXW<^UgH
zAJsQtB_Sbv!)8x&5<n~j>>8y8FprvnLm)lXe5o=9GQjM@QZy9x=4j}((geyboY6y-
z5vrTO7=g&ZIhO^{-~)g`(eoMrV_6>F+LQGra;0)MS16#x1uoqhEWw`94&x^;sWm$k
z%<(FFTf3X1!CY!hKz%_Hv1S5v13nX+tHI|{iQFs!q(HJ7U}&Ka*yM(GJXosTt>h*E
z6x@d1#>9_aG^Pb|zYlNH>p_1Ty49`tr!^b?+>U=PM$fwg|D1&nHp8E0(2uR~i=f1H
zkamYin?vONL*!lfI?x($ETltli{}>mgI<ch@Bzghx?{4q8eS}}0T+0#@U`6C1wExY
zIONu?;D`K&06Z%#>E_NB;#I`7_&SQ_orpD}>7#r1OZcaTf4+}@eu#gX-SFoI{4=`;
z{`BFW+we~d{~QldioXW`yuKg)yb=G*9e_Uz@Xs6Y&u#c;{yg}z1poXQ{yB_)zJ-6j
zi+_ga!=KIg=R^4CVf=IQV)!$Jf5w);pFQ|z)=}`M2mjRY&p!NfHGF&6zYhQGUk-l`
z;-7QiGuZw%{PP_C`5pej&)xd?Ib0vVG34Vnh<rSR*T+M8ecVCo;|^ON3$T1F(DE@^
z?PD_A$C-(bGnZztK-EDa6$|nu&}0-VkQ90cO)Y<qil2Iiz6Zt(barC2S{7<eUOhT=
zQ@q`;_1Mp$l159l{=sY38ZEGa%ffxJClI!PBgj{qXsjVd$XCl{s$LL}^j*O29|m2x
zrF8`uwaHu!%p+(=9rPFh4uX6xtRJX@4dl0@a(ce_B((R^3b+S=0Q*D1OVmqXvopj-
zWU70;8k(65ZX3mLtAwN+9=HnJIxW0!WO8+Q3Z^4hI{;ft_c!1>KlvPNHqDe8YwMXh
z9F1Fq-Kk>UG(Kg}p1%1m6`q`IJ&CHO1ncqfC7o=+jzMT)jYE4_DAW~nlMj;z-IwgX
z=rXkG7(woQs}iW9vOj?sK%vnB?AHGl{NmJSRzq^*7==DyD<Oa*;w$JE`&^Fqmx;&w
zAFkt_!^hjR<BBUT1L;!&ayIt2DFgnD!+`(b#qa2olZxLGXpu<#{GR~>x$wPBVT!_w
zZ#wYe{0qUILY)gJIW1y@*tHHj$RW&Xf~IFiA=KRiPb%KlOjBBoCItH^sX|CQ?nesq
z#>CWH?@GPdT$uY_zWs{bm+XGA(BF5ZCt<99vg*{>=W@t96A$@f*CBKL^%p`5WsqEU
z?nx*xTVDh*>YfY<Uo$C%rK3J1)y5K2E$>P-E-3yEq#Cnx_bTLbq1vEwmd0_4k!PVZ
z^|(TVEN{pmzc%rZZz4m6_+SB?<pEiWa&{CMsDuYW@4@v0!@~j*{*yulCw4n@;jo*^
zzYURF;Uh#^^ctU73%EP&<{H_1<&Kx_xMGLNzr4qV*b1#X5-+B-9rYpk^+BXXqCn*R
z(89?R7)A6xvxxq;R|dKKLSkM#>dFfq6lz|%1739OO_>yT+>eyxA0(#U_g$%Hl;roC
zCD|pQ{)3M%f18*W&%5%%D9QiLi$?5J^SD!zTKMJA!pXuLdGT|zUVH`eG1D;7JR|XF
zPj?+{yFm39N)+&q4x{~pw*r?X=EbG1yf8}MFU*qntOGAP_CrVBpZSTYH||P3qd$8t
zJ@sNIknR)ATN01<b*`grr?<aW^wvdII*PoHZ1n!_#8i8SE7goL^1FmGG6fUNzfL^l
zPmm$|Z-xkir#RcrC?A7131*ZCBJqsUe=lI95R}d0cg^bh&nuA6B<9Xjggf#1Z!r-!
z3FNM>E`@*!+5X%A!hG^~iJ9?hS7w;}Z%54dJ#@e?SerC(J_cGi1<poUo*g60E|K#e
zd@OlpVrFb|Wrk6f?ac6krJ}xtuV<2V)p_DJDJ_^+s^rVf0+uq2V>QY^0ZVXLwF;))
Inxn)2AJ$==UjP6A

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/datasets/index.doctree b/.doctrees/autoapi/lmflow/datasets/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..f6d53011fa1f3f18e08df4c76c99b5154b86c815
GIT binary patch
literal 95210
zcmeHw37A|}m2f&qcha5CN(e$os6-&0PSPO(6q<kpk_d)G1`vaSMRnEdu6n7iDr(6B
z;*UFVo_m3ezuV{xGcLcQGq{f^j*R|}f-5tMIs?C>qod<8`m2L8^Ph9>{+3(!RdqU!
z|K$5p{p!`b_pJAvbI(2ZzB~Hwnlq<w4*m<4HfqIE<;YmBP^eT3UZWi>EL8KYX|K|3
zzrMZkh3(thBf(%jceqh))$?9Em;+Dbi=}d*?p4}1wSy)2c(YV)jPdph8eYCxs#aw4
zVt;X<xZtMtNO3UeZ<d;6Px~oYTArRPR}YUBa?M=BYc|^WC_Fj;I`0Ug5PjgKZHCoA
zEeAL>+r{O<U{)esoEHpmT=xg_%cY9f-rt(Y1w*-JvtF8LHGxK?X<n(&K>Qcv%eh7a
zZo+kcB{$8k78Y7H^<^TLKUnuB@tIbzpqZP(pW4Nx#nr{(;_1bC#Wlgct;OoJx3%7?
zRJ{7u!`1r1tyk^a1B8h`fM8=QLu_lV)vV@frLBzgt%CHerAom&GFEQ@O>6hgZA}#C
z1pRbRV;JLQ@eH7$xDGh=WcYs;{NDioHzGa2l6r}#7zyUra;17ZIAyw8XqCN-GGSf^
ztJp7F!ao!T_BY$PR&i5tiz0G=aV@YCWHShy&rQ{HwW5Yt@Q2qGOO4FcN1MfJC6fnb
z$`rgw;C~~N%Mg<@yO@kKhfB?3X4-=X3ysWVwVo+eO3hNPT)MSXncAEwS91lt7|K-&
zndw}mRBM%UO}LX0oXSkrtJ9f6X>!s-7094~yb5&IXyuEUTqAQuYibG`?8)W540iL3
zYoED)jOo^@{|cCLB?snT$~25IKbI(Yl+`80HO0-KI)g%HAg!mdF<k5q-hExs%iPRm
za5EQ4sgf_ZVB8zx-~tG-I|mG>k)5j_C7$L4n}>+aTsWISWVM<ut<Cgw-9x=Rgg-O^
z5iT+TgLPP+BFocYk{T(l0y6;5wiP$_>)ZfUTuKHd3!@V?0SuHdT0roiBvuH10jFN;
z@POW-oxRrsTX6ua3f}Tt&b<VFc~06htZC06*vIO$;SGSXtJkf|)auo|2ZpndnK+uM
zw5DrE!NO(MtrM)aKEv<>IQ7OvGS7p-Bh%&aMh(U=IGtMEYk2iTUK=&NCUE2$zZS%@
zNNBonH_OD0VCdS`1haJ9zpOHDDR#ZoT^rE)zlc3fgKj;U#)ZJHTD4UvXo|>9_8f2>
z)qJz==}OwcoEKLIYpmYFH2C|2vyz_Xqm%iV_RHG)gKbHVPPfX<QWp4?D`&&aykMA{
zjLA}&d70w9#Xt3eh10ns1+UgDw)ggd(&TCs1Aq&jDpx1k$AW=ksZfBB;aISkcOY*p
z1ofGK;ah0$of8bvfNoSP<)iTI5JVozp`qmO;SxU+*nbx|c0;AX954=VISk&Z=GEId
z-8o|W*PCL4@vKHgFw{_bk`t~qbeB6?>}x2bg{T4+EnILp4>c&nSry3dO8-8SPX9iw
z>EC7mJ||e#%uRr|tY`Dpa<)<P@=-lZ3?1gv^TDzk58Srx_S?4KF8<Pp`x?;7!3EEo
z(O_r%{By>4pe{CA)6?KvE`jI=#1gFHp((_HW7^N{V7c}!3-p=+<(?t<$L;*k(3K*x
z02haDhyReNPWTX}vPcB)Pn8aNl?=u;5SMs$?tv&`4FTXIb}$%dPSvZeT7<zE=YrKN
zr{q3Zj`HDB0SC|!D%a0};XehL{z6!>4yFRaa@oOi^bRoUo~Ra%Mi2!&=hdr+Bljs#
z$59v4M1>QZ$7@H&fsNzB%Vu_jqR27M>`QLGk-WkP=;v?^;xvc>vPZyWYmkFYDY|1m
z2Dn@<fvj?6tsj3KOl-vX;_#KX_s_y#DPoU{ryyV^w1T;{qwQdzc61BS*Y-jt#D6k&
zvf4ib{)7#X{%9}&B!W|H7vBLhD7>7`l`3uj9B3CTfE&%DHE?Y65Js>FzI#V%<x;-X
zY~M*gAI{a$Tn+-8neU333xQiaI!?hg7<2^E%+;s7CWQP6vcR$IuR#LV%)&6<+Xs9)
zmH1R#RUD9Zj_&DlP&QatT`{3SO@_GE--9@>r4Co&B?!^Rd?a%vKDXLmPVoJc;ZJBN
zDP_K}M?0}9=%EM2mK!)_#Wiv2sVf&}OYw2elk0iQR6MbA-Cv4buJ(uFPp4jo=utFR
z$z2BX(I&QGUJDZaF&MWWhI<W8)(kt$0xMI@xJP{W51{pT!>x7h6z?d$5W##$IW6Rr
z>FnjqUwL}%mc`AZr?t5G!9A@C&68xTsohbszK2W9P1X=?`gh3|6mXsgXgm_q=p^4>
z{D<OW$Pn=2S~oLp0WOGh<i&p!NCs!1Z!R^m6k)OeGj}LgDx-@X**G-B4EsNe=efuG
ziTQn`E0-vIs{bL<=dSt?Zk6}pFM_U2-27#cTO)SM^Lx_oCGaKcRudisH=DU}<i?H6
z*ZONUOoF{L2!;Ch&(ym2zXe~o(>)~tJIx9IJwBc6g#E4XyPFGs6rQx%fg~6FeT3lb
zg8iNFxPKXbTn-=m{ax66L)gP65f0<-czN9Dcq0e-w}2lQhgHGck&PPP@wU$QWLu4<
z0-f#?*n&CTukyw=b|a<S>3)e&u&RECc$VP+{%5XT#u1=avV>QAi10IKkGcoJUYS?p
zgFU&|uw(P@i|N~cG2D!EWd8eNp4KtZe3@pt{Xc_dVUIRADLs$!X~s>9S3`GV@oLMi
zh8$~d%FA3{tu$eBAe~$E2j%0g0NLppXXQ8M)BWAD1^KugfYg|2VvPyE&L}hWJ<xzz
z2FwhD^ApD9$i~b_sghx#bvQG=vF$#-ja<1#08f|C3Fbq7H98N++0YXFCEF;$f<m)Z
z*YiTbsflPa`;K!dYK<#5xE!G+K9_|}6Z5O9g6I!QR&~kwxY%<fSOjQpfg~IO3l<}6
znt0H6V~)f%)~{iYtNpU<m_&LBh;&XcNVta)Pjnvc&gCT%T09vGkqkz|Lc}2%{TEr#
zUqR^0bV0cUYc`qy(C!k-iSZZ<Q2llijNl|K`*En`9mWL}m;j2)Y_5_oR_hwj4(8OD
z9cD1Q2P;Aiu>?gccv}xa;saLhVBo-77)&Nhg|`1Pz)*9eXczx5j5Oo4vXFiro*)YZ
z>ukX9Hl)yi>n!JD&ZmO`KDQ<R8uYWdY;ho1IN2(fNjMCJp)92KvcPe$Jj@p(W-n{Q
z+8_SFBx)Hnh`5i4vn7&CKryHAE6pS2V9GP&=FV2^_7m8T>P`goiXS+q9bpj-ki{kk
zVK#{Dz!#viU;(g-vIG7@_>~k@|6ZwXVe5j&f(4bTxLhJ`Z(+%sW5F>0QG8zxPjC_&
z@}Z@bs<<bf!=4ZxEXd-zTl*N2>2Y=mE6wMq-Pvb9hlH&c>9`S+JFEl77~;fKhTZV`
zlK{bV9V7lh?Org+p^p?`T`~)alg79Jk6%RkYNuG0%3Pe;U#)nX#oE<)rqQg!%?(Ai
z+BgXd6&p5ZMzJyc89fLYoT&zn$3exGlF`UjD%B<|XBsQy!S?6wcybdmo#UAe#EK1<
zY|iYU=V9bPA{V+I&&WqI8!ov7)_WJ=qWUT^Xp|}cS0O1OBF)pVzBC07`ZB^jz?JIc
z-$R|4zB}kU9LcTwPhgYPK7zMp0)&6~DSP}!5)k=^OIs8EEpSVSF2-qbZ(#FgE9SAW
zEe4~B{fK`%K=5AxA2!bwgV~F9%;a(+M@W8GT+DR9<JE}AYm)HbE+__%yL3EmI<5-L
z^-SoHbb7iDQTc}Z5RzR|7=Zy0GvJyF9h5W)?bPll{*GJu8}wD8T6zICS?!+(e<&1L
zDTi4NZF4I!Hv7f}3UH-B@9hA(76pB!4%+{XF2jUT_#7(`e<A!Fo>hop$O!Z=gh%99
zI-N&`vG$FqH1Y;KrdSo&+bN!ut>_xjx|R}Wu2Ig%{f_`+jCjQko|5Od+ESV<<;{#a
ziz>7vIXW^*f3Nn7fNOX*>+g<ZsG%`VVtcuj4-R4jcd=aRWffL3b4?@K3;HgWN_nxS
zly9&C4*IH5rMwH9i1FY=-Mv;IF4V#5S*b$BS=x}FkKR>K#l9%b`k)mi&POJ5ux_0_
zlq<J9lSqiATeVuX-pn>&eX)5<tBOTn6w}<N({2WAy0Qw7wVgda27kayoeO`|$l?Vp
zu|WY5HDV6M1-yYESbkSDe-XhYCd#b3`gw-pGjtRa?AW*MP_BocBPNC$G7iPXkp0jO
z#b(Gj6vL3sgV2jO7jLiVXT`|>umaa5ZLHMY3e8lSFlv?vN|)B>-=ObQ)+Yt8pETd9
zHx9hM8?ckAonUItwnKS4bP>@TZfjD|a-JQE11%rWX<2Lp>tgJM?kdH(%ntp5FpZr}
z4I{lLQ`1i~3~p0QNsS7M`bIUU@LW3;Hc34jnu4h_#@>949t0rvMRBZThh_7x3|3F9
zStw>*aom|ZD%pdV3g;Kvakla2_89&Qu!*1Gkoc<Lz1j}X#v=wVC6DZn2dA=vFLcM4
zU||(7S;}dVljKRNm|4i%wFtJS{1XQISp6+(p;i8AJq)q<{FA~bl7IRqz>DRd{J(%p
z|33IICVKvd@ir~#V6a&n@IQ(-5Safl`4fJ9S=;}(_$ib_x)h?OA7E?BA^D$#OJOq^
zGe)%P+aBrz`0TQW9+c`mU{yWc69Tcap!jm2^nTtF=1haZ&}7Lg7nqKy2#%`&6dcjd
z+5}PXL$;)RooX2jT`d$(S{D75BH+<91o&Uar&s%r;2)Ow18b)650yn7z~#NsQ91J*
z9nFl3im{zC3&hfxV?!DEhl?Rl)Wd4az$~)z<xJhXrB#B0#!RzH)$CMy#&vvbXlQhl
z$9@C=BALw@td+<3_2TtfGHE@**okQ-$ihLdR0CU~Qg|#{<2{Oe)L3R0>jcoLW{i#q
zD4K18_)IF!%#|4pS63TQeP6BQ${9%kJ<rToOpCqVQP|;7FdoO1eesW3J}Azn4FJ@6
zysBWO8BD)nmw<XG3pHKSn(744bCyU~(qNtf5Sb6PWCq~H24MNz0Eh!>Q*;z#o`k$y
zY{Fu0KQkEdiK);4^ylEytNq`A{({~bB~*A%pBy9`c{E%9rbfUO(Bc1O9l~&=Y$HsZ
zp;lHKj!)Xf401&4g;Ps@Kn>N^sYpGQk&2k%P-VkZ>4t_KH#4ew>%49RFDU7*v7&M;
z4VBWOUI{|j1TDi;fUp8=#NL_Z!X`{^c2WP_6!lL@)c!BvkMK@G&D{dcjP+rI$A3ac
zVv&Xf|Gf{Wl}JW@hanX+NxnJ_|HXj1idb(BD7@FzfZ7X!xq>?MPlG>_fYlbuF0}6q
z^O?u*c(Z2z;hPqvPKEF@Q`%w0;ll~>V#CQF)o_apq1`hgXF=7KUH10)9d9zpAJgbR
zl7@ac(5?dEw4h~*frd?(Je>y`szBU8+cZjZwU7^u_qcIm1m-i$YsU=`^G6sR;know
z({M}-jk}3i=Fq@<T@8)>z?auihvv{2lp%;v8SRd;t?v)Pby9hR=c4i_$BIGVlkls;
z?Ne!Rlj`_PV8|zkA$D3cyN^v651rK!*&e5kPZiVSkU}~IbR(wuhA379h;hmp{u$IW
zIuZU(8XgJ-{Rd*7si1hTD+T>M;M6y%!>$zci4a`7P`OP;crFUMA}SW4opC=Y)f^r$
zxnadhs^&Goia}_Z!ct-r#yw}%M20ifv^V*&8O8iFMKM-F{O@BEHrcZ+KVvxjRn_|*
zX!0feO0#%7gohafF%=dueDN0L^Ji*nq<;yB!nQk7)CIy;NI5s>kzyW#=dP;eSXIUr
zIRVa$ge%GTydWcY(?g})x`c<+qHx<R)d=&ibE$?lm7=N#Utfp!v#9S-43RYF%2~(5
zi0`xrf0OKl$`lHa75$l(@d?!G)riv~T3z-RSEy!PM5{nk0B<ht#IFb5VTWQ{+~H7M
z7I$7{hhj_8Gbk9m(CVKuV)%obNf)bgyq+W~C}nB<1JG-l%pgk%VsbE8&UIf?Yn3Yu
zWbKDO?SDB>9=->lhLVQ`7`D&hB!VoX$-~4tSL}$FoQV#EAyR~DC1-<jteQ1v{+EF&
z|1SXt|0{U)alHDfy!skk9rGWd&%dS5zoXBu)8{wn^HKane6-py8r-7rs0S;wiRZ}N
zY|t4=Hb}0r>L*%g29piG&K5|{JY($7S5c14{>aCaB0s{VEQDmb7)R^#W!VW?pHkTR
zgjTQ^k}gB{d$QxyX6XMwQ%5uO3p*5>8RAe}XqmUbG#MtRttOR00W%b0q!J@O(Um=l
zicSqv{|nG-m-c9>ut!40W37=?NMVcC0h2<us06m?VNPDxwn%g!)LaqiY>T!5YRndG
z$1Ac$bakR_(Yc((;y461n-tk06cU)|t>@@yX$U<&A2VX4D{*Wv8x?t6DfwP}DzW0k
z(N;ZOc0#tQ61G)oBc&T<f2mE>0ch%IqI^3Pn~CC3TujsvI}}@_#Gu4PO8A2=G;`$X
ziNR2$6cv<e>R$%Ec4?VTNmBG!3l&jEVX5v03__Nw0+wnQC%0=$B|7NAQr!zUFiZ6(
zctw_qu1>V28o{!CH}fQlDhqt=UDUge(N2S2XKeFXFIb{(PS);6@i9Zho>A84lU%Yw
zFsA=@v_5|+J0a@>C%W`DPy3SWFSYgg0yK5BK7VhAVzWLRii`F6ksXT7`Y<Rl)(3y+
z%KAhFrCFa3L9bm}pJ8ErNCRU{4pTf~XBI7t*qH{{nU8S7y0$Z-1EJ1}Jm-jLHK4}q
z%;|VVc80D_w4FJJTMZ}}g0uXb(nMz_P0{qOdX9yL&ECM!nC9nP;3GFb=A+6CosW<7
z8U$^VosbziqGwh8&G2KF+PzX}w?SvDmYehmZNCCwhrH?$(DpHoZ`ax`I_N>$4*(8K
z+i$=t(ssH!(b_(m&6d4NHj8_P6SSBgpa|*E#8d?Dq-PZ$l-R@s`oRJThB;aJT`72-
z%UX!4D?eM2osfdx28weRq=RH8JnbIXnEsA`d0`+Y(SmF7wT22g76d3$?w5cM-Hm1c
zZq#yFZm6XYrOf!CRk=qc`p_7@4aV?6E|RXtu;@T2h9Xn*SOe_j*0^NidyybYCjJ>-
z(Ez5a6Fq>>&*qC>{vge`vvA_<bgt<%mPZpt`&hKUo`?0yXrFu=xfwwRE20m|b-|dQ
z)VTjF*AF4yuEzbRWhXT5U(oBgkLhm~#{Fr+jm9u6-UKK-BPQSzz>+Razz=bp#F>CN
zAq#78dNT5$F(Je70x%2@aGAv#2E9so>~b6?jZA5hic2#zo5b^zG1|a7)D~=ztb1_8
zZxaTTEY(fhOsV7(Qk&vtS9<OO$dXIdF9e!?C27*CtT$ehD>VQ;IGGmCDrvCu9pQlU
zCXoaiuQ}!gc@}hv77FXZL0g2Fg!PL}tdGZ{@hO0q>!Pt>AYU|YcUm*&YfeI!gb_*W
zBgOzl<|DS^75NCdI?+C2Gaf3Q#WSv*Ttnijx9p~8k1#5`8bP$OP<cL1hj1qyGGI(~
zf<f6_xCiZyQsyhH%IuhR*(Ez6ukePjS2&{s0X&j5wp1=L+wez>tw;;SBZ5M8LlO8q
ziCdhFv2LS$4m1T)%$4@|?cilQ6k7&|LvhLA+-iqn%iwS*2LI+G|A=?-`zYD(ij{#C
zyl(hXZKPyiQtYIpWqEXTd@uK%BquYTaS*I&yL9wLrkwBiy`K6vcl@T_hW9=z^L*Pj
zv-dFgrtGjXROo%U^z(i@EZY_{2CFC5bhi!f5j)N{{(L!xKcd`_XM4m~rOe;4!?W>-
z!Ar>_`{VvQp}0)nZx>@yXCW|_RW8bT#K*%5y+}O#G+=C(@$k@4LOiUwE)fQwB*I{c
zYBc(N1H@cAxtu#jzoVDey<8Js7$P=(iDSYM0Nh1mi5sQrM`ie!ris$LQ-R9l45My`
z^r-iZK+7uL5@`7e=!?j<uzG%Dlg9x(XZjis!u5(q5Ig6bk;0;5HX9xvx{WrQt;dLe
z{dnfO8?N5HncacyZB00B%xl2y=h7|Q`Nkf{EBv?qpm#L84Ia3D=T*<zo!vIJjWuS!
z^53^>zi&5x9~)zP(eU}()lXI++dBZ+KIR5Qz`kkdc6I=5sL*-Tor%I4r)i?JVl0hT
z=sUXJ1d1^VjX^1*n><E3DN69U+lEZBuaAmcNx~*-`Yl1sg3VmKi5(hZ4RvUUG}Iw}
z<!54<h3b8`39-PH;`W|yx&3C~_GO9O)+VE;+qf!lIurK1;FyEGK4R3t9Vr01{ag+0
zXQx41PMF^UIA2FN!$!wN6Xh!8!sr|nZChP%JCjM0fJ3s_CWaR>9)ud(SPC=LSPGIi
zhCjrR&dRz4lr_fKg!(7sRHypBGye+YAj^##3Q<qPS$+7u4VYUY=Ee*^O&>O;MI2UJ
z&jn>5a!C<1>&owMH|~SMd;E@jHEnrO8k(fG{2>tZKZqbRq7FGGU)!XDO&AZ7qa2La
zwatmhwu25(w4DIhWQv+-Y@LtFCx#RIqDJ}KyJEl}NjUx_bRfKJ5SAdRcS!X}wT0?w
zXr^x)FMWa;BYlEvO8>sDh`$$ze;+Z~E`1MsXIoRT36qOm`dSkdCAUawBX1uw;Gp+#
zrG)rUS9H8vLa-yHX)&5fX4>sdG%ZGN@H7W*X<Cd;c2E;dJ^F4JDhyYO^ndA!^!G>z
zb~-fD5h)f!*qdmiW7x4F9qHIa3t_()31LV4uf^Q^hIY^WsECz|QbgI}sEFm*NFEj8
z|3HWQZa5Ht=iz7r9e!bP>rQQM0PjAeLm}|U2K|(VQiX-uFoT$d;}}SrVnV+WW0DZ;
zi6tpDIox^cel{4^yXFrV8@!_rz`&17;B8AKNPcF*7HAYcGXd}E^DeupYiB0B5A=2;
zb!cuA>?aivBbR?uqxXssJi9^V+XE2}YG7<MuUU+9(n?97nZBW-=!P^D$$;Yhz`pB=
zeRd9L_6VCWj@yN#*&`Iv*#yPZ@e{iV`?fS36MI|_6B*_nSG?Di^8N+zrNQ`OoKK)_
zx#sb(c1MZ)bs@NR5ppF)crHquu+5s|lkls;?Hy@wlj?XMFyw8-5IZfJI${&XLuYkF
zw#TXCX0>q|ju(;};fJETec?QvIKvv=T0P<$W;OgX7}aR*^T9NX6I%HI@yygpyw{ah
zeh}F6aq7@CshSSjwDOxFxOPc$tweY(S{b*!nj;hUlhVf@rNK(-<A;D1KOk1vsnGNh
zn=tM<>mxFp>7%`g7OA2|UGIPx-+7I(Q>RtruR=h@<E!^V?@Kttz}a;S{1&i8oLuve
z-xdE)A2zko&QE?i*s0hkA>@O>+VIMg6z=hKGfxV$zz=dC>K_f(n3&6SM<bRlquqs)
zk+8DD@ZpBR82RRpFz)B$v)XLp!c^?xfz<iXC>%)Py{<<6$3f=Xs6%tqi#j~PToJ28
z?kibe9m2sbYi<D$cCrPKB^msDoDQ$=z}FD_&m(H=z_o#Uzhr=Y6l!q|@@&ijiZ*bu
z$*4mUK5%gaq!_p#gWlD^wbSu1j(0Wb6ICO}<;a9r0U55<FJczT2mar5BA-{kh_xg^
zhe<qt$%myv@{a*UR{e4>UQzW6U7hIamkaZ>_LbGv(voYP+1ol;9Vk>hC)Grch5PKe
zB4$&#M_Cw?SQ29#(r`iOH5-hCG%wp&M6~k~SY?bp0n*eBqSFx&p9wWGjfi!>GQzW|
ztf*h$XCiz|LZb4Vhrr$HyF-IzVZI9c8~s7i{HW|SuCB(ha3}^dqm2Aik#UI#n0r8@
zWJqNXFkCZ_EjE8kNHzS_N|Uvu<VPeDQ-*rRT!@61lzg8a@3Ew0aIvX&CUk2_Nk23N
zHE=H}ImHgkRxZI{nYJ<|Yaq7Kj#VD<?}oE^WnMn<Waq7RoNXl~n_~DQ=23huAik<0
zaj_kqjYkY#N*>uCkCc?e47OP)DG_IdM%_VtWr@NvQd#mr7&Kj0mgGS+v6a?Q6DJBx
zWRZB-I!5bE9)cL>!`ybjsEQuvGU7**-7_-fIbu~geU#z)8=vyDuu;O4=K?;>(yB(-
zk;fUmc%<v;&v}mBcbp2H?bea-WUh~^!Tj1(oYk_$kAgD15*mXdMC&nB<eOrW@mA{1
zwC-WiYLkp_N(7P(L})b@aejpZ6m61$O<v~EgikV1=WHPhBBW&pz7D<Dp$SG+`sPX8
zi4-QCPco|faS@y=RtKwR>|C~o0f_4)Lm@<A*Y`V-&nFqekO(?VA$Fc*d>2q;lZ@}-
z6-_ee>O@a6Mw?al0{g^$l)RIkCliAb8y-p;A@KqUhVP>k-<5%wv&0sCIR?PLkxENC
z@<T8hMgW|4A7xB`v*6=Jo@PvNDc-d!<RY&9m%xWEUHh3JhgjDxWhgwltZr4JN3)~7
z81#o=c7Dx8GP{oaLaR1~O&)jTPr*j<j{Lw!^#V|wMyWRluoo8KrY?Bv#yAyqY6W?b
z8c;0Dj42Al+xSm4W#2p3gW_|cx*1tdCi=j?SRST-*J?xYc1OUn{!H&A1X$Lh>Cbck
ztp9uyVR-w{(9TTlXtP+YP%VHc9hNU`U@XVl0C>x4raB3=0i}j`g+q#H%3Q!du*q4-
zj9F$)B9fdSSd+JLrE5CBLOhKKQ)ieaL@SVp@IpW$f=d7*ZYLG^p2Y%5Uf`WHNJJdz
z__Xp8*Q6TPS{V6DFh%>Jacmgbxl;0XyR0{{jMMLy-M<2MPbRWkE4(-^b-lNyL0hgI
z{u*#T%y2de75_gk<$Zk`X6mK9PXIq(oumVr3WSR}X~RRvJQ+2}kZ$a;l=mHJXp#!@
z8zAUyM3D335t}d`nDdTs1fs1_A=?56DB9!^o7_oFG`3nxdH<>_2K=Xl^Iy7@_iNn{
zzpt+ksQ;3KU{ql)Yr`aCm%nPW7lp~pkP@c<MJ!$Ab@qBS?~l78WdKQGFzje)X%MuI
z%%a<yXqt>pAn(91O_Q<7&D2CA{wEsodOh!wrM4+!Xo!hXyl(*#&tTXo(TGR1SoC0T
zq7jb)$6lhutXOzPPmyiMMBa{Gx))(tZ=I-UZBrTnX|7}6Aj)V1w2lqeR4OV(*6%{<
z&DC)48&Vj!ikxB1WNvSxDLksl(;dbL&->vNwNv6t%lrKTdOsJMq{#c7!ygy<GR5o}
z%=?u<7_NCgg%IX+_MOt#iGH5=6Nv{wh$)iJiN8AlN0#_|Azo49kFI_j5`XwYQKyqN
zY>7X5D`~H1lVakJVUv{jV_;O`kG?A-@kZdTzS?gNQOCsJ-Lex>lG8v*ddvL9bvVnZ
zKjKq->Q7-8N&USRIMHS5?|#(txbu@R^+(vmrT*}T$Qs_h)6jmto=YV@_oqC3aJpK4
z!tOAmBX@1e?cw%^ltO($>Ox#8MN(L>1`+<8MEJOvy+@&YSF<NjVF}57TV2hb;CZ)Z
z?`MD`GkgDvS7i3+>O`Bpv!`=4CzH22>6|&}pyzdtm52qsM-FOecooIQILkMWpfRgQ
z*GfT$m(761TPQmri?=;u@s@_L<VYGNKB<pgSudf+jV)`9%vxv)hQPf><{UdLTm1)v
zWqN#l;D^eGPGhJUtm&>s<_bH`wi=mTG5irlF^M%Y*W2OQc*NkP<dOaHNR3RKlFot`
zT_g@O!7&)`K^0~Z4>}Kc+NB45J~WN>pkx7r_gt)@7xkRm(Gm4?xyF0WnDUpI-0S&T
zt5WIXNUMT*6*h`rJPU3XZ{9Ta=-`q^>}w;3v|Trm7Rft%Vm1qy>pQ`T-MseRRBYuN
z2HpjY!W#zgo*pqL4@#7eS_IiQj;eJp^=R%GaNH;KU<j}os)rSW?-D?Gs2HHiF$Fwo
zzk0&{G;aD5epOEOku<o;WY9&(k*^R(?8K;K&^~IzxM=6R$_6pczk}eI>7d_=4Q#(k
z!#lBo?dL>|xq%Jub)}gXqa1$1_|wh4%B9Oqz;<DB%|v)Eni;oEjjLtcPf8~<X|R$y
zc?t4j4K(dksio;8wPE~o)=A<y(@A?9m9AoX{8o~aoy9y~g<hdkzDDJ%_FaIa|2~R^
z3e{$$rt_I;gdl1<ufj$N-t2d48(TRQd$?CBl6IP7AMbTF_IIJs3)G|Ol|&s>ihXP^
z4k2MzQf>zjcCsDlW*^&Yi5fd_ZS?L|9I%f>&7%<4at?5`(MxS^a%jUxFOGr~qxUj%
z^x8>znC)e|lMi;0t~f4vmj{3j*Sw2Di00MTxaM6%#4ZRisqdU_{yV^t<z2pxSCn_5
zs}r4f8JqH&S^iQyr~J#4lkzWGM?L3Z)@q#rR$DHSoBy8+C##FI>x?4QMOqWF(q-IJ
zj{9dI0^OFGBiS)0{V%c;(xhiaGBjzcQDZtR&IyJKUL&8a)8Uj2{Rq0n;y47IBWt0=
zO>Fhyt7of%dB8GlXY;ZZHtO{-Kh?p+1f_>_iH89#m_)do?0lw`0_$c#4#hmB`uRd%
z`kt@xX}cYY?Ud>iv|MV3VmqapLy47n_!ZIntzaC59?XN?Ep5_@CY-43HQW4E)IE#c
zG^aW`B9}LW1g!RB28lllDJOY;u5M)FQ%?%RNb2ePU^2Q)J+*-R*wm9@;zSDSWRZf>
z(2b^|o`nwmpSc-xPDRZ~)cy)*b<}cDFULj+QTtjvzF~~}F{(0-)>6b|kj&ZoX+|f}
z4nhY^4C{{z6Jp*}KIa{&DAuw%H=~r^3a#TaEZxMf@2B2OucwbW<hd5V<`f3W;}qLy
z@hf5dCI>iL{7P-^acINiSJa0T@$2)DjF&(gHv6=b&*Rq@^2bHkcU<Dv#{h_H{HhS5
zsH5T<zlvcl2r&iOIez^W;K<_FU*i?UuXJ^y<JYHXCQiI1+9{6R<PgV-4ttJgPmaZs
zuh|)4>`5_C3fmI<n=&ygPO=4dj$!Q**$FAwEj@>|aUITr$6g)d0;M?PJ-5O%;<*Qb
zD_wf-r$EzK&mAjG;lbBf(2siZThO#E;*y(PPrg^`OWcxi*sJiZSXW@91W$gp?p1}v
zy7lPmV>|2U*;mFj6QeNxM>Px!sW_r}`sbq*Z-&-0=jrRzo88rEp8j@)f$1Q^a@0IM
zVSSSW9L>{Hn;RV3a8Hl=lETy9ie&7DHYq&)W&CmBAdicue>(ti_4EoMGQ_WN_4L9F
z2trIjcJ}li0UVj9|0rIOr>CnE?di9pXPktig1vgJ4)3XR^70#OUS8|0XAi$Tf?b?<
zUmAK!*p$e9Wlp{lQcQC&J9_npWhbOeuk6{Y$8|UhUi}o^E$ZTn_vi}Ch)4e%aHLC*
z{!1JeafOr-DGG0XY699(PyR|YU7zRDnq5ymywVmvhCKlH<b&8K!ISUXqXsM6#p`AG
z3|;v3%;91wU!-He>(w&f_nU+FWp?ZEuj~b8$GG649(or;y{A5uSvV_|w6rMW9@MaP
z(0b;gjB}|s(=dhgMT;{2Kw)4yr*IoB${?)QI>6DQ3~F<lLmM7tpf0C~GVVq)2B1xf
zC}SRfTsY_B5@qB8h-;Lg5F*ELuWOVcOq3wR6lCWp<5s|tMH#o@6-60zb)ut;bI?N<
z;IxM<zBAA%#yHCsW5}+0jxfl($nL;(V3A7beb<W(?1jk*-Bm`U9a2Yg2s_3WuaTXQ
z3ca)E*dngOS%@u+q`BmCLV>BmF%nt)KJcT<$l}8|7UCj{kOW0sVT3~x<!Ds#PBc}o
z<Z_zbsNxG!J5uBaK8KAGqKXX7%_Xtj1`4N#y6NcqmzWL2*b3iPgX_DgDAzpneJHKR
zq4msp=$}(>=I{=CTg^j%Kw)4S_HY}`Llf42?*K>h(A4Hphc?_pqh6%&(0`6(d<NR2
z@X(*;j|<0kTs-udC%b!Sg%Fw7Pq}(%G28_qrXV|e=!*bH=An1s6?tg7I?*0_6Z$wj
zcge{)XKc<Hd+FIP)6Ada1BEJBY!Ty`7YeL};ZQe}$#@oU8g*63%p9<e{`mJ~C!{bB
z_Uw=2I$Q*+K%&ru6__B+@o&Zs)t}rV@$2LfX0`rg8ql(p<c10n;+QfELCUgS>>x5K
z+amRi&wMEKBJs&pKxmio$s3?)Y<!}WS%fAl!iYwrk_XW|@8{YuyHUw|IMn!o7WQ&H
z*YQqlln|Ayrsx0#(rrL;it3?bEHWsc71l(%qs+;Jsi@Q<kT0TS9)Q*}7lC|*dNU1L
zn9o`S@+F0V>A1pev<QT-zTW|k7J*Qkk2<vB5eVu)iU{N(B;#IalOh6n2Y+1X=5dKY
z=B;v%KommcP`>RNfrz0k2r&iOIRZHiaAXn4YP_Ndgsx6>1hS!#JLDF9oSqbYU=KY<
z9;1ZEWU1_BYq@4|9Al5o8UB`X{F1piv%zmvD;r{y1@jpaY!fs4qHM=`!1A6$k#l4x
zq&bg8LXk6Fkmuo)v43}o*^obC>^>fpk4Ph_?kGZ^A#scgN35?g*aJ<$LYl>-;{saW
zxK>Phzz)S$Ov<6S6vj^4q1cK^ITV9~6u3FZbw{kgb-O*Qab1&yWI7%PshY-0$6f@2
z{N)svQUB)4DOSdf!8jcqU=IqF5nq*dzRAi|D;@VR#9XxVU3Mro?c`8gwDZGuC^qfn
zP<qhL&sl-%@|{)f{A&_Y3hhk6@9$ViwwB3rD8`#Z`Ysk~3Rg-)e`bebI};%VEptw>
zxg!@^KA^t^Wq`RwL6G>Ws9Iu&_W*R3GnD3}Mypn<)|=VJbgtgKrB!WujrK9DioFk>
zg(#2_o6@Y<ACDZ!5a(}aVNyCQ=6V!R#7{^SYLN-)CeWZRC#2_tm|`cSESwRO(Iq;9
z(TV6|7^QFFdK^Cy)lM_YTnn+A@w)0}6%ItkI4gqzhCI=z^Mgv?c3JgJ*ECQmxtI7m
zi1<$46cF#`y(vRi=ju6lNtaibZ;RA#DHIQ+K~e0qeHJ!Km^0rvIw}J)dTS&^tn75}
z>xy0mBiamb7T*hwk3p$`S2y{7xKu7@CcF%)T|s}lq<9-_$BO?0u_(+G_|SfV#Rf1Q
z4j6BxjClf1!_@*=C5=`pY4DA`w?L!taT9n?5A>5w4jMBpN$`DCzG>>w+}JxG<tR)m
z`=am;Lf}l-7gBVeZ{bBKs73MV2zAKB`CUm)pxR|?McbRx&?X1*50H;<BtF`ip`}Nt
z4dcCCURrvD<mEfS(b6N-X40VzPmkbmPmvz^9+G|yv`LX35#~;06^=`K<YB<cwS+<;
zM3dZqb4`y3ha(6v1=%@0@?F4@rANMpSCk&1s}r3b+1SWU*UD}QlC{YR66~Yr6v<kS
zOJy969Bgie#bd8PmyjAcZ+mQ7WPmjhRzZAK!+g$@Op9rbTE|tI-*ByG&T-$rM^YoJ
z;!);q!$^h<ut)h%#w4k)BItlCMZzj83D#-l3TWz>^f=QF#Wt<vP+X>!+wD+n(@GA-
z;MhFS6XIFIVix|a7`e*|Tz6;brj-b#hh4Rol8{WrbdaiPf^_UdNslGeznS!aN(g;y
z#c3nFWZNqUg=Ak8D~ncETCp!MbX-y$hwM;nI?189=;TZ6P;5HMq4c1WcUyt$Qk_+u
zd>siXg-)j6_xr3QThjv^ic6~FemfLfsv`w0pSMGC&>bVyG55UfV#ML48gJjQ!|5W$
z!H7(0MC^}8QXFx9a~4t@S_XmQlK3QtLN1cz_$B!JE|VO6rzRvhL}((?9I`k-1Ym8P
zvZ*(=L6iaU_OE&5(J|3+=Cu^ZWh&~l<_x~`T+e;Td7j(DO8b@pQTnvgv<teC(i0wY
z<^lnWdDGa0$4s7-d8UrzU5_;8(7*k*DiLHjFBe#ctrJooPahqvP|5@Ryh?dgTGJC=
z-8dc_GayWwQPw+WIW9M#;T*Y3NTGyAKlRHUy^+Y!et`sr5qj>_nc9(;JaKU`Gzup!
z@Sd(g$%Z79xadbEx{!JdB`#1J!dT9yFG^t#guq!9RwOPEic8|++%AM75bdh1qN|jK
zE}66#KpsvK5ABT5k`~m4@!c*fEonirdWr)aEonh*wmY=pNedjPDUufRko2|Ch9xa#
z<UEvncy}TfVCN|}Fn3|21OwxbLU90kDzVdwtPM|tixc;#RB30X;vE(*R}`;e#CI>|
zN)_0@MaG_WRZxCFPqvaCcFeR+Iw3SO$b`()iYVwfo0aq!7_cQKFSQ`*1E~bZt=apb
zQP`T{y{@d;FskGKq8?3a*2VdXpAR8nS3{nhMc6Jj?kU{~SD3kOn=4EAtu)k0OSc&L
z`wil+ogJE`qc%(?&X$hk_$~)Hnx&&Q_d2xUmJTgO3QIR+S~@!^53>^=MQ|o8*>OqF
zE<MdXJ*yC+^z3PKJ261KU`QtD!m&KQR=sn2_H4kBrDr$b6{TnC>O`k!&u!Mhv}a*`
z8WyW-<&xV*`jMnWZMe6dQ?~oVJ&rI`OSgLCnaOfB*M!{dIb+))nOkg4PL|>BM73Im
zo7dG_o>Pi<Sr~*cGSMHD`M4Css9R{$8aSqXcgRji8`pp~nr|}N(F?KMtr^MTRq-A2
zM~yVopnO#DL%X9GalND=b;aB@(9|(`T(Cp2CF(d7mqcCN4#k#i;ZO_?_LF`R_u|vX
z`~k(xmsk<d=fSCxbsSC)DdamyRHl+!ldvhc{uV3I)^r_*;*zd=j~$9FU6+EEkJ_O)
zXpoVvn+KLnj6S+ns`6Poq%IP6jLc-e9?JUf2LxdB8s=qXoDq&QE(`dBf;jnI5%`#$
zK-=qXzQs^;SpoR59g5Aca40T@<=^d4Y-+)w7=~p&x;TqnVZSR@E?sSNR$ZGF2BrtA
zvIy`B=g+KGB?ZejT4}ag6%NJ4syx*W#b#Af(6ZYO#X)@xt1=%$BQf&W7o{uL+F^BJ
zR~Vr^*_B>b6U%m@Y__BbO&x=jTkTM6wuD1*u_brfq1cpxLosYgx)ts>T7h?MNeGD^
z49VRjq!i&v3Vz>bCE037I20E{@+mtMn;}U-%NOlX9CXJpB#Enu-?qc)!iX>;QyLNb
z<B`?Gxars|tR_B5)FVgcq4DdA3e9jX!hb5n8C|X`_Me`xt{9yKixtMxk`Rv;Rj<M1
z(S<w;0(1JwxoXEado{s`R#d(1TFgY~D~njK1=YWptH3mLu2RV0^c1ffUb7+BX}V~s
zT;EDyLGqklH*WeWWClrF&YJ>h(X#8AT!kD-gQHk{JPR8otPM<!j>?HXEep`3b#tcR
zP3BtV=EX1rQEQU5dKI@dl%TjBUfbJ*wZj72fezP@i4i5(t%g^?orgk3xIn<d)t^9K
zP@ELN<Zd3io_E29>Uw*`B(y33wbmLJ2-gwcSj3EZ0;{oQD{UEIcPbh2Wq_T~D7*}S
z_jH9!wr8P5&9hJ^FQy*NWq^}nq$?u9*x%IXxG{u`>1{)!Lo9hB4lc_9J9?3+AfUHS
z6s6N?D3wbCXCu#j;<=qQ+R^~EVREv|R$Bxm*^WEF(Uu0N%~KrO@TCDX7@acXU&RzZ
z8I^rcM$*?q8&p;>fnSS^@)7?V+Pz=_e+RgbocykY@Fw1!SURxtlrMt*0X9l7bpIQM
zIOV#z$?9nhRC*%y>T-OODOrOV>7P3{y?ZdC>>qJ0W{oxU22|Kl)0tqGWw<zE35Spo
zT^<3*pE3+-!c$;O7aQnm2K6JUbb%Yx4?v@^LB)Gr8Pv_F;qRv&O@rFSy6P80NZ2)-
zud5<#7eo5gE(Iz^V%P1}$bBpgebRhxK@NYLIBaK#W<IG6lZvzXBuT!P;FzID!D0A1
zF}LH;4Go)RVZd>tCjKE9G{d^h(Z<ZYHKwZDnaan^e-dBJcRk|0uExw(l;{(T)5d#V
zyBIU4rJ{z98HDXJW}J!%kj*&VP!?GXnQoh_5wj@`b#la<gZv$Vrr{K7I2O}J47Fi0
zaUL-wM>b;YZM1-cM%lYa_IAbN0S7wn6amL5lCA;{J6gOJK}^v^Kyh4Z5pDu;xNa3w
z2+`{A#jdpoVj3+7F{$rdi_iodSuH{fuc#J*u1<6<!nOiT>aucxIi2Wtc2XsR*;mhX
z2y4v#MoN`JDepDLGv!jF8M||AiP=I}i|}n_1zrWx&<(9Qa2!htULiXnrMfgyQn1*-
zoUk(2@M0bq@r{)60^WrnNq$$f{E<W>bydLIps8cR{m<=CYzcP`#U<hXNjnr<(v(9n
zxHlg)hVecAWsUT^V&#{u;B{M{swRM;(L>JtA(E4+q}Ihd#thS1>F9^nm){iAfB5l8
z3O7#uXCZ~VN@PMqV?I8CtB{N&a9;{UcbUMQzcwL(8@4zieJk_MCeqR5?WO2#U(VIA
zWAZk$3#_;blcGBngxTo)pc(`x0ddK;QCAB+HG4TCzKl1GP0e;rRp`fG;OSQoa_hGg
ziW}0PC^EBWVxxqF#l+|+H?)+hK(oLyu{+7>u$>{~g+ww|x=!3Ef>Mn#U*r(%nFMoY
z^t?DkW*CXkoM@2-lJq5%jg9;cVu^7>yHd%7Pl7ImM&ZN(-qZCj*@}fSu~(o9?w}sc
z#6dsFQ;Y@vO=;6}LjX-X8&Y_YZACaP6P_Kt2v4BgTPKRravDnI<mXD{YYCdBnEX&1
zCJVd#v}`NM??MMS+T@4YobS+vPkzt{q?r8dLDJVj8&t?n<nzgoFoI&va9k!oZvmWK
zCqD`yic@cJo%{$lBnUAD*?IEw0l<+>em;m-H2I;c6FvFa+Q?N(&7)a(-BJFa)70n8
zq^XbIQ_mA0INnsli-!t_a^(`7oS{`($4-HU^|r!pm^YOjc$jOnkgRF(9H&2jBRe6r
zx-l~SIW>_2!q1!vb$B>5EgIIlmed+I6}A5;5luZodJLL6PLTf94#hS>;!s>BNPTCR
z&e4>dwL*qNF(*hVs~b<Ug4bPTx(O1Y(Zd93Dc}`0&(;Z23YKrN((J-=m!hPL>`-h)
zNhxT#(hkKozvfVkq9ki|<FoCsx+qLyD41?Y*SXNF_FFV~%@GoQJTj||vud+2t6eE(
zwT2mwpVKN7BXion29mp+)4l-26?>wi=|segR!+lpgroD>8!>?TEH{lE=d)*D?bY#M
zUO3f}b(<j<oq-w{tpG+=Ah|xST)``?I;ks%KDW&x5wGA)VY+gQD60upuUnUaLLJCX
zVj0fFQQY5OJ6fpbGwap~w`6^Wp$6UmPw3t?pcAOD+3dbsTmw2G2jTY3k<>Y$`*T#0
zuRuAO{}*^g0UceP=zwm#I4QPUXH9*~ByFF|z{#7%YGGqL*qRjEC3M<zn72Kl;}Jc6
z=^XK$lF*INyTsqsaQHmeX`v9Nt#ph7KPx+-II#I!8wWlv5ltNjejA!P#(_VvL$Sqy
z9EwXE_#bvCHb>5(^b`j!%h<wD-Hy&~95@VkrHTVnuzb`?vo#LnP+a1`3+zyAabOBs
z_Sm5~sE>Obc%2<q7jYmX)bw7u&V}N@Z_wa1ZCm*9NE{ev)n*|MJVV5RNya@sBvjZ&
zLc+fS>bndHZv)B2hJ=Z(MMQ_|I*<_!6q^`hJ;<$E$3StrJy5)&H8lkVBztmsSv4(#
z#%`J_kNFC#GV5BU2at!6xR>*$$0>lk2fB9+AO$KsfIQ|JKnnSYc)=Wkodd`}LxrLM
z@_l$k0VG|`L;z_b6B32b5104>M6bz<3!H-fd^y)>v^~8K|H~Nneq0i8za)UJx+P#Y
z;<F|^XoF?7T>fBg%G-kH0eO{XqfJ8?5t6^r%iH;H8(^_)<4t(#Tydzlx;V^8#CUA)
z4n5sHQrr<NW%2Ho0^3f2bi%nB)HVNUO#YNBhI*c-^$c2*Y*d4RMzvMX!`{cu!AdAL
ztu)Fx+{}=L>C#jk=1c9{;Vh5-X0@3sxA)EohQu>)PU+q`Eu^F`Sdwqm>p)_b_q9J5
z7I(^}ir3yBEQBp7SS!|O2g_mF)oPZ?jWIYY54l!=R|sF9E4Sc%5U+2uDO)g5Z8dAH
zW_y2e>Hc7F%By(wT(eq-o(6$vs5^tRo8Zf$dbQdFRK3a4k#^9x6^?{uC}aWdL3m(@
ze{I#v@NIFkn$J#v$ne)j+h5r?XHKwU68fmX>nOBe?hF<;ynGX8iP<AK(QV%eM;tZl
zrCK&$ot`FP?hjV8J6_qNJK4jfW-&`D36uaZ-T-x!ZVcBwdI3h+JLCZ`=LSQ?T)CO$
zU;BgQ-jNcpu$gPN8tQI;!JBAJL4QjQ=jyoA2cD>VxoNmF2(rj~NDtyyssO{XUL{{G
zz*FsDu+^O0a>15H39e5!f6RLIdbOS{!k(eBR|obZs(Cn2cPV4L)=NM2(hk;yehGEq
zKLO(`Si<^MIEdD#n?Hma2g{HKP`pMFSX2g@+QH&-t}@kvp#>Z-!1uXcr5&79_wv<x
zAzScjpf&)eM2sDP(KiWX0!K;(p$)}>U`ejktY+1%U~UO?ZE*qS{#mJyz*xEgyQ#c-
zC09n#Eu}l8=_q}5tJd&Zg=+T5G-}f_??|oM@Sr+NwInUcH?xOIjS|z>rP)HMk(+>n
z#w3;t#5jR5Hw?sJ6(tQi=yWlEEQ_ynXtcqOtfF^7z&f@^yNp0K*joWazNnr%EWhsU
z3zp+x1I?*Sfg+Ym)1_v6-?9zClA2e`!VrPCLsTFG0ziiG&O{Dy$8ixnsn~4R8sl5H
z!tKUb?I=vRD`VCA6bz@6@kv4eo`fTD;KVc-e(hi(+~0E6Ip+aCm((Zoq3-dH>i%re
zJ(??2*@IqvtmHK(2?kh$t;Oba8GsEoa}zL%3e94B@4{dW=;U-(4o*n`oXLY4ax7TZ
zI0}cb9my6;Q^hj;2mPI2uA=G>(>rFe{3SDY1}ACnoXM7}`5b5MntT!7?U4uZvGE3?
zBo@LOY7Vw)0AdMX*DN(bc+3nO1MR8iOO;8`0d5yoprL5ACPI&urcrm{FdjA-p|l9>
z5vU9tU|E0>d=xM!dR`M?oScWJ4raaST&bMR6$)r{K}vT9%duxP#rQHyX3&lW^SsKT
zHk=9>EM(RMhA(I$7Dd3=fY<)!>hK<2(wpUg6lhixY%TNwd)LsE2P=%b6?m};-G<&K
z<(C1prv<vd0B^GE5&vQIrJsk7;seF|{nvr<^k0o1m!Zpj8h(5RpZGL<w1PowiSPR)
z4?=B^6(2Z8&5lu%WAr$@0%#q$7}oD_aYu157+_RLZ&2)K`{Igcz=Oqg-~@L_H!B=p
z&|hjyS7B6uNAh>BLvKxPb-r-35YF9Sd=VjB{4~S%E<_%&_0hfidHg8j$3Nl6kMQHb
z0DRnpA78+aFXP8$3*h7F`0*k9_)GlQybwON;YSHSD){jU{P;9}JZ%s@cHzedhvDO6
z`0>ld@bMe`c<oa7xEnv_ErXAR`0@Sa@bNGBF?bSuEWwY@;m5=H@%t;`<2Cq!@2d6j
zUA8{HB*@2?2>E!Du8$|}(%x4e_s05IMCD_Vm5-@sA5+sl&NzIWk+g!v`Y<EGFV2@>
zu%hOGGO#;nO~rFe!}L4sIgEQ6N2e#MWjS)`(UqZ_^65dN$3X#=8m%z;2gg`%w!zGu
z9PW$%0>KH``h2y8_8208e6?I=BM7371();tuZL0a()KlAw`#dMSU?y64H&8ja0GO3
zN#ke(%$&F#)j{<#74KOU@Bjb-Qp(^68YM8x88QQz>LIU=mL!84J~5~&AuGp*t^?1_
zO6HrHTmyau>yWD-g-xMHn((5Yd=9pWW=hQsjZ6cMwJpNFQaK-*I&RP&zvB;dcxn<g
zjl(GMZQFX(%_dQ>*AH4)qs@LE?eqon*~^*dU%7wJGtqEkoVfSRS~!XWxri7*vCt59
z>zCl0Na=Hnprmn%#vd?;5TlX$O8zDOuIN3Mn%<YW(mPMk+rRsoYn}<(r-j{|_P1#S
zU*|yZ@4Wi`acXhA-Blby{rne03!!{((}beQV#+}lPum3^6KX<Wkh4NW2u~ZZ_ZuRq
z7L4@F1Vpn3;g^bcQKqG=MhoJ3)KsaYJ@=!O`ID)+_i<P5%@xYL;MqH`*?;B!r_1sC
zN8Ph9aeVR$)5PBu$=^&(^4DES7US1{E40vv6uNV7N`<-TG6+KtW++0<bg3*o^`W@-
z%hX)^xhvO%qWF&>*SMW~r^Y@PuIZ;>oT?dl9vw4}D@Ty!2}SaX^=Td3QfL7Y<RHvT
zOAuY;%Gn7NpcWnk1GjG;8XJ>{h_4JPII;E6g+t=c&meNEeT2vgSCbQFfpDku#fTiZ
zcK7wWuh|{vU*6@yZ0*F6p45vE+MfDQ^11|hktz_m2wFG=BC#s^?sygbZ?6n`c|&Sh
zT<<E2Sg+iJEPD2)T#I||M`?01HTQ0D<zB2Nzb9UkT>|Rg`S|kHsbvwk$|6>i|63N#
z#Hr@HQp@5`dXmMj;>X3!OJ-%F`Cw{lKjlhor$F@yEebg7K<)3m75InLviO#(EMl#|
z6Y-jNr-Lkd_Cqu8&z?xly<fR<FV>&^XZPGooItuyFjqY}Z5(<Mv~Y?;JB-_XeSO-%
zby1d{LhqRk-?yb^+n6ic;*@0H{JxZ0avY|ZSEZ);N}}0c2jfE|IXliNp95`D%qbB>
z>N%x<0brytl+WY$_5T;v$X06U)JZy)ItnorkJZS51daS}4-7NP*Q6H4D_w;V=YjY2
zB8=Zc7yRzj?E6z!_QmS*f&^W5iJgDvqshNcEsVc%6-KNs+lAo;D`XLiSka`F)kX3)
g)ekHxRr2Ll0qYf}u;gT@fOR&!Bn1<1t%<S!2iCsT`v3p{

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/datasets/multi_modal_dataset/index.doctree b/.doctrees/autoapi/lmflow/datasets/multi_modal_dataset/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..c22f5c804c98f864ec41e935739d55113bd0ef4b
GIT binary patch
literal 45539
zcmdU23y>T~dDiRhbf>p$%V1e(ge-Kz(h1q%z*&|p%MV<N&MyJRmc5?aox9o5?#_B=
zSGofmY>XemLxIanOsWV8RWSrZFc6&Z2n7`36+;39DkcRu##N9|1d>XHs*phC`};B7
zGdt5eyDOa(SNSwM(~tlE`+xP{-GARZ^qyxwv55T_t_kXHxi&Xpmr6Ci<OJ<-dC4!f
zs!pxheyqLYzV<!s@o=<Z9}oOiqv*85MJQ2p%au~YskLu!hpVZ0vs?)#ME#M#DK^V~
zEvoJgxx?<r?d@@QG#qM{n-xd<6mG3lXDa^jiIUy41E(2GR9ls1xlr{>cBLR*wy8R5
zFFWka5m@;uJ<#UR57%wL+HAY)!qGy6oVzp}7C23X%PQrX)1GQg+u@kqY&Oc%ttMze
zxGyc20>X2oSh0fuFY!E7v#b1Rd8t)bKc?;C(S|caWm@4#)1IYIZFh~k(Ov1D?=E#W
zh4yaOuR6OMty;}#>^|-{j_&^bL)U-+@*e;g?B;;&wp&f#u9tUn677~G+MPn8-Q`-z
znVV<?!1E;sm$atcMd1*8rxBZvgS#1cxm!Tdt@wW%{@;%OcMy&sQ=`l<8xNP&?Q)|X
zUWk!sRh&z$JmVC;oR{8}r?5Lb)oj}>_d<7<BG8ok5)1>zdK9B(&o=D3t054+>abf5
ztQ+Q=u3xi?U|6f<%wX&S%eL46S%-*ytV6^;)>WLRCOH3W=(o0DBmj8@A4LwNok>W-
zX`j2v-HD-CE^%X6jTHTAOt##iaO0LOR=wdD9k5r)nx406t!jN9d|++aA|W@H;qalK
z8-nR<pu*9)YGpF0+a$%y>kTJx8poVA!8#>j6fmy<tSh+eE8$%$UxXtEgeWkv+@O9d
zLKX89ilpOCUYrqZ4ddOlO?$dku^WY=UnvB2r>Jp3H#T0z%7<%@-16o<_q_R{d*r_i
z*Sl1AVdInw<K#sbP3|Qz4qDZ!-I(78=>&g=8-&`0KAX@!x5IVXuL5WgSC%Kn@ZW(}
z(DbW}&l#HwE*~2c57vy|h=~E|4gU3jU#Zwlzj2M<xVcrw_?N+QQGNc^v79-(TCYH+
zsG%S#XOW=oYPM?S+N=;MQs@a7#o&yF!_C=-->P?v8<_!!L8Zq*U5+Zp%cZ8<J~&1^
z`)-WmJHaZCO1A9A6<Pzg>J1PM(|&0_70{&(|9I+$AySo{b==D(IN_a>_4!GE+MB#H
zW5BGvIke<NM&*5X9$~s+9D06b-Kmu@D1|u~9}U{DaEd{Yc6p>yZaNLSqV?^5UMPoi
zGZ1JiW$~qKWdSq9kP4U7=iA|MeSQ}x+IDoSMPvB|FvI6yy)!j(I1JK2rMAO8JsZ_F
z{2!`Zwac}(`^|6!znb%PXwjtvOt=ESow<6YTr4-+53|q5?MAJA0yb=Zk{iUFE(`SX
z{3Nqu?J3r4({9W<&31U{0Ib5)-e=%@oE~*|a4`nwX-Mh@cUX=2)B2b@nd3%-OP2Hn
zSW;@|lT^(#&{W~$DEQ~-@DoxNg=$c;dA><iN}Fpgq)_>_j!HHQ3RF(#DEwy+)H9FC
zM?m5~1BpKxC-Jy}=8Z@h5%EoyD4$~N){4@QpOrv{cP%&!oSD8TA@DXWUbIMPM&_QH
z=%7@jx0$N38AO`zIqp8wT|s&q(S}KMA|2^It7|-O2kPZldhU<i=iGmua-R>c%A>Qo
zjp7-gyu5Q6;bQuWz0(K8EG|%C-la5zOpAGE(_=nG=B?wMXX~<^oP~OE+ct!~Rf=?-
zHzgo<H>L4m(kzqaev82QV9z+4Nh&muxzGoprr;!I7Ydibtqj_jFvD&<g3=EsB_upY
z^T@4=Uz_C~ubo8pLDYU*vbJ8xy~8a~H}|5TobZ4;m<}lTRG7h*vvd8Gn5a`LV`xPL
z6{2aAVa_&sO{>PjGJ!O2+8Z^#@9b%OF?I9a5!GN|ya-02A+=o>m)yPEy@yc2luT=8
zAq?cj7pKk2-1{Wpx4aQp(Oe7DWHB`_u%HXPIfSJ!i8^MF^B_&I(<;H972v1^JE%(2
zKkC%VZ*m%ia@C%73hehdTB+8XJFV#JZ9A=riHX};6|EOEU9}4{D9w8*)GKg1RbNqk
zRAjM`88NPwhiRdOFrL+P9gB5sm22z=TB|iJf5vLkbTfujM|a2er1Yg=jdv+~<H%Wd
zjdfnL>aC^*g*}7G0sGC%iMiuLlGuVv!uap;zGilEYUq6w0|-_eWOgXq_)rqC9-N;0
zB8Nj8<0L2L^LGplz}2yi$Z<Qu8S)6@$Rj(LxR1Na2bx_VkdmIycS%uD9Ct&w(SVEQ
z^n5Q;EosIkpVRYg#s}G6d6v`jW7O-%>A5_Q(_=!P6ygF+)>5<}O|raiGMZtW&Xzo-
z>o{$Is9<kS(7#g+?gV{9R5qDRA_iu2-d8z9nNCm_LqpB%Lo<t)m_6WlcgiYDQ!E4J
zR1W@!_hk-N5BlJqNcbW}uA7N`njp0**8yC^LIp%+j@*V{urx<*PtgUpokDPu^1ltl
z7xI5j1f7n&##q~_?10mqtz8B?OY--iUO$rmnmm%9jyUntX_ooTj9QTS^RitPZGfn_
zK0sz!q3O>7s=+1yd{H@%<iD81HK^ntX3b=q0V(;sJ2N)@Nq!iKnx_3b+%M+s<<kOP
zn@M^F5$~lag5WejmsAjQ#1`B^*wPg_jThe}*eQKZn$Jx*)`dT^l*QXouODS`bslBW
zod{=u;%G3+#hrzS({^aXL<M_O8Gk@ExXSQF<vc2*!Xe638Q~@ISxnxE6X(|reOFhr
z>?aeXFzY2~&(ky~X5Qe3dM5p;P@nJ4rJ#=O8C-WB+mo$0O}<ohK_N^k&+pJEapn0E
zlioPCM-$0^F!EtO&H49e!z7hin)64f*N^7Ba!}0)3Y-CI^BG1qnhUErR-`s%yhR%-
zD&L#-{1w&U+Vkt8vT2;82lN#VShn`OI<7qjiUDm1Pf}}2w=P{8CsbG6>iwyjlP(gK
z^Ajqst}~ZIoLWSivdbky)!v8M>@$v^H;0#Lp=s}ll+qc|X)PElh!c-dkII;1&{k3E
z*h@p4ngYazIQ1HAX~A|g>i|NQ*kTbyt2^GS5#Qg_h*QjZUsQv|thWNBiFlANb3r=3
zM{YR4whL_039rf6<zS8TU1^BK=FnR^23+Wm?WohR<l!^y6j_EHo}D|XXb;J1`ceqW
z*s|DcHDWt7Z_dDNoZ{ZtvRSCxO?T1?nhh;XG%4|=xcCIE@}cDZ#w^=e8px>RqHqoC
z*RIw4rk&U|8eVkU-mMg}p0u_zDsJDm)7mTH#HKLpWSg|2BG&eO`w(4(4IGDB9pbwv
z6qg*0#yp}%gfnkT+r1JlwxqX*C)G$R(MVkQ!o|lS5ZZ=H=u|JDDjPlOUSH@*hLptO
zkXPo#ov0(mq1(Wr*yqD(g`K*ztyJ#L#SX#?a`0BP2{K<{g>x8zk{X2#SAc1D!R25;
zg{ufgMtk-<wo^yq_znY(qYTF+f|;9$gv*opU7_**E@&(#G&l~+35IO&-)8`SJp-S%
zu2vnadQ{5TkE;#TWd9-~I9p0;)P>&sx{l_fT5McMz6(j(uGA|@u8+S|RQzir6~o1~
zP67BVC)Z!%AI8OJGZGObby~V=b%_1qNeKm>RM+3?(sjMF>u2<?y(WH1EuSUeKQwmY
z-HXpz63!9yKBxDx6iSgJ6Rp39Q1yn8eN*WD(%7@ERrJjneA|cGktWn$Z0{KBi0x9N
zEtln)axuO&&3#VacE)|<jo!}zPtpzYzLti6Y=E=<7b*RpB1AWOA7|}*Pvf6xP_`2*
zvUMx7`QTg~C-rDlS8|W`Mv?jKWhVE~h~QdOZB2XM)FIB`)t8!kc@|A{u#sR^(Muj?
zeTBIf6SIn5V$6EJup+4zJBy3Wl@#5tG{Tz2-roO$Qfijy{f>UpZcMo^(%gN|Ouvr}
zaonos;E5iY-dsF<F<T!l(2*3lR>G3tCrrr+@+dwAB0zvTU4-x~ZCClRvU8jcNVLN>
z^08poimu<#EJl}gvlyv&(wQ8%=0Tq0C^#d7Q7V_(-k;Er$=$c^vxI<0yr3h%9ScPk
z?+7ZfsRhp8z*4lsF|y9sZHuEt@!h}Su-IVhKF7Gj9S)bzv?>*5<#^{~*n&|cO44lN
zIri>B?CE81RWSmrNR;KrHnsu4JLxCerWH9~MLsR*AmO;S@q5RW@O6VpD~bnU!+p`6
z;D@A0OG$V5POit`t&WH)nz?%<Au!qs{!4SV?5b^t@1fRsc<4o0;Aovcp6EO-5%~f`
zkD-mE0g1`EVhY`&c^^PMCGu%|tWU7de2+EXZpa1|TMn~v<7SPI?RJbUk=NvpQ+e9H
zMn`z4{KJAGaA8i(He&BHqKEWl44ra1Y!}91yFRj|#uGz=6^v!L?nd)p17N0uMRH#o
zP7y_95|rE5@66h}mDI+K%%+#h#ili$XNr?CQMIBYeVAb9T6xK6h#gWa5IpVpR9kX~
z96PT7@a0f&B|Gkww09wQPIyx^W%d3@`|AA*@d6iDt)Q}|KR5Eqxc6Z+<~@l&sq?S*
z;RD9%r%@w}%P)s-a}=-QU}pzHbTizaZ3wq?r*pdLP@s31j^}{_Hyz&Hc=YbUpBNRq
z`|w6s3#K=`w^6yJL-=z)`xItQC-q@_sO>!<D<+J`<)H2h)D$xw-a~lINmXj-L8_@5
zkUXW#BgGEJx#vF`v3Aldit{SodB7_if$3yNR=kb$lTC2--!*?1KPw8=TJz=OYWaE>
zKkRfOTq!=v-|J99n8Y9|w5I0Ecd{I{MExNK)I`LL`!RPA8`35Zp*F;p;{HN$y22m_
z_DoV31kXuQ7`&-Yg|QdI8dDgopi&t7=zI2tntTdlGHQe=j4RJHg~4!6R~QQP4u$b5
zVAYSpcmwL?RTvC1sW7NwLSbA98TdUxT~M_@g&`a0OJR7M37s1KvFMLaf3`$VFU8Xd
zZ!7zI0sFg+{e2nxyPf?Vr@w^QDG=<#d@uEkUnyZP@kw(AUORwSF`<Jruz0VGx7I1)
z4}!Wv!sBI?kUvaC1{LzRM~yHczk1<?{DLg6AV#8eIaknk$oX4=SU+<90o3a*=kYO-
zqE2-aqJA|*{XRkCzC~R&Fo39kn$TjR{uz4OLVwuP=@#{=LZRZ+3Wf0<gULAp@UCBi
z$A4fDAs4_5EZ?IMRw8(*`}^TY)D3s~qu)xnzbkk~4tZiqp3j2+UDOB@?$?3=pMn=A
zozrm1IA58|^!jBGSHz%F+FRf$#O5qzAt{=!Clufvdg2?PQ9pX(*Cat*ClJxd3Kfw(
z2T}}^&>*jc26;vdWZxP@HXzw4b#^h#Whg$i;GV+jVW~dW&=XT1?CEr?k2f})S%gY)
z{3_Z`bJ8@))F7Iq6Rv@k%Eg^+jz?JU<cGitI1)gqW&FlvQ+~}6EAO#zL%6=P9l3~-
z_DLzFag6=I9>)byBTOkBOe&?#T~HDVHMWv88r5N66H7($kBGl&#T9b<BDlDYh-+Nq
zvcM#&nqr?)jf%CfOIY#vTG-j#+lh~!KsB~(vnEkcj$Frkh1u)-$LuBsD&pbmqGenk
ziDW3Lj209R7hfj9z?15D#@KPV*za7!Em4(?p6C~8Wy&>{tV>%}Uq*~A`{O@Um&eUr
zVqJX3M#`wdF}u=o;@x9CrB$!{jb<T0?w;mdEuWHtu_;bVhbpc+^+;{%IGc-mF2X<h
zI!}stEU=Azd6tb$Q5)&zR^iZLZa0IH{g~T7M!oLlHcfp=%eyl_T5yXJM))9ja*JTC
zzKyVKpkE{WS40Em!+)Bdm=R`Ar`rhMR45>A4Uz&?Pof`x?I6}y1888gyIN~(yqs!L
zS7=3~W{SO23hFz8izE}o%}mZ%{<oq=m<`2#&A4N~xXhr4n;7{rW=_OQ7)3lOD1R1p
z(6cdo0abf2hR@|F-k~H@@Lymhxiw3h=$w$Q9To5$+VSfkMnBqd)kq%g7~@E(7PsX<
zq*EW#;@{s29_d>j$_At;#4V**@-R$Fwh`7$AHIy9m_B4rCqo~`fzhoznFg_FBex6~
zF%X-y;;T|D{(5nQz$+sJ*i*j<?4vc7ld!OINIO%KOkY6LE9`IP`!M9TUs}1%O6Q80
z%)rOw<Lj>sO)c`-ef@Q}389qC9!!0XdynYRl6*U9MbI{1VWn^nfdNcYtH%L}_g?&o
zv5|B*#TMJ^qDGimxDk>Wi#r(;F$lnoWfZ)(MO~#Ch@56D1>RVEYJKO(h=GaC{zgPG
zZu4U1E%u3a-pHZ~le>(V80UReRPDjkm09t)QPy@X(f}l8uK6}Op6DNg2l~MvTQw3t
zb{OlFQC?VM?cEM6(`~VWt-}^yg-W??u{+GS6;MxNn!<X0knb7@H<|QL$86(9Sl4UN
z5-Ehft*dN6+JpG03G0f@b39K}TF~!r2pMKyzd}#UzOtv&ZJw@;B073PiGB>z09hOc
zHA&OyNbr;%T1F2U*ki>DY_rB?*B#n_?bU_D*B`uks&L)ZRaf7#(`q<4$Wtre6cu8B
zK`i*Ua8~Tcf@vksB_}tztk^$AjW8>AbJB`sq!SVLE90xu_rHjmAHZ^@z!<Y!2T$TK
zrkT89-hPV3)4#yr$)5`UCCg2ohSA>r4y5bB-u;FZPuRPN>r-IY;<92iwKocb^ZLhN
z1F8;)!B!rCHHxDcvY0EfuA?3+g)vI`qL96)e`F^ZvYvx^TD0Vs5vjex+{;~TUYa=S
zCPKBMmt%S_>pB!vglp#)h3k|Vink@9=pM1<z{O3UK9`dYw1~{7aqL+snl){FSI5gq
zgp;oowzFnzO^0ua`A4>8?!`2+qL<jn8iVimoBQfx{5S|bj-R=Bhku%)%NN)`{UsD-
zx__!*>+ny%k4ioJr__`5Pw83ur{_j$Iqp#<<KQ>LGyO-@8^|-24M;B`K3;h|(=UM0
zxM%uBdSad_dpg~o>CKv-pPuchA1^i6Cibp9Cj#NU2T|%>yGTiHF=;g$b}hi&5>=-W
zOx)0L4mXhE)F~a7iBor4t{rf;o3y5VzhYfxU1L{*Xz_aC{MR2C71+3!%YXfT)Clum
zUzhY>b7Z>`-cOE_%AEWx>UIFXHU;*WUwbHu$V3WbfB_a}^dtszerqIx(YgI^RPDhv
z{4Z8K?%ZD315wgk2+i9y`M>9k_U*y1Lsjr&o<pr$&ApfowTfQy9BSQT?!|PdRrC^b
zvb!o9?v`~P&+cAPN!kG;%({CSJL%T#aBFuEdOqNKIUDZyEZisroThPutRs~0m(zsc
z+zf4=%J5NHJ}%grquwZj&l(ACM5xFpkc;74=3Yz;FM7$t@b{a0F)_U8CC2c}$Y5qM
zHUFsS_#R{5`Z)K*xV)R5ne$OLG(9k94^)22h^aBLgXksCtoifiUW{`qhJ3Ch70;M^
z$-yzP8Qc)_t)#q(Uy4V5VD7D-IUGZ^N1b9W-Z6(u@mm&n4)?F{2h!(o3bu|p+_zDw
z=Q$kpB<FDSEaz}rySwKdD-Ey1eD1qwBXd5dpLBnHJQZBtM4Ds!aZy0YuWje|71*_T
zy8jT@BTS6#$5FW`{mU(bHUE=)|1hfq#oa5%Os<oz0kO&-od~}LR2mgEam0BvTrpko
zi$@D=i=e&_@_ykSPg;jv+{#d+n-ORY&pKXiy6ip%c54EQ1Wu@{BY5%1c|<mEc|)YQ
z!yz(?lhzwt9CQ#DENC5|1m6HG5<jU^>t_y38^SXJ7R3!#U)*pTxZ!-%G))$0kH~q{
zI!`G;RbmW;JZQLHu}mq@c&ZPzzf~mZ7!8p*DORRH8|-v$yW%Fc<`YcfY7oI~1zcat
zSgc}CI~9JVBV7=J-5O2>cY(zJ<1QK9w2KY&v8I>Hrlp-wFBJLq_d-7Jaz&0iK#ofp
z1=07ILQ_LdP(iWr_MA9MfOD5u_#Ey9pGY|rfXs~ynYb&HP$u5>3V*7?@y|AM#6ho7
zRfuGl<mly<@hTeAor+u$P4t#GSA+ybh{Ro8(s9TwR?<54A1r(XK=NQ=NKY(S$evDj
zu<!<x!N$!zeQdNk8R>EdUD&R6LF=T3kmPpUg^LkBK@z5IK8s6z&KThbqDEMZFi6G-
zyCroN;h}EDly3WrsKWuGgDFt|zlsii!$<^U9N{adYBKnVz{r<a@zW9=e7=7qf5f5A
z6CM1OxffHgMD&s;Sn^wQFQ%!l=p_~`=^kNYay3|VaLo#ne$c&wo`NjhVedHrvjUJN
zdD<uhoTk}6OAaA81<}EajRZF$w3|^NS9EZ{xfc_|i(c|D{PpHuObjo2i7|Xn(LvkT
zci)WLXLRs(HZ*bLnrAHYQ2EV9OpVb&LFGKr!4u|QOwqv}IOcJ4FF81-m+0UV=HB{<
z4suj`)G6lT9nrxQzh!}=gHOYdrAG%9Y#q_T_n}hH(Lw4-MhEFxMh6j5itaY;*qK0+
zjpSC305i__qk&wZ!3W~8yFPD;7`xfG%1x{0I26Swl~{blu2_`Vl!9WjPL0kU;~60h
zo)gw$fsRBZ@eF}RdEd_(#YR`Hdf4$8M&SFRCMF0QF&NQ4zB3|{#=IpJqB(P5I^!aZ
zKkthH=D`3zMa`s>mS7k?(nwV_o_UqVGd!f`O=_>glN3VquFxA*u|hFTRWFoC46V%T
zCQ~F#3m)DL3>Tv^B;p3G*V7R&+|UaSQNZvXAaE8#Anx5HM^y_LQWcJKw&hC%42ev?
z%fUNZz>uma-f>Sw!0-~_oi|`8K@kDNcjXKiN}JVx!0-@&<N?DedSU@X_H?=fhSy((
zwWI?+%s29HT;n%xZXp-Pv2x&)=mXAQ_I&@$Re28Cck6ZFjfGbC9$1DvPd|=%Qm>1d
z#X##py<7|Pexg!boQ=~6>d(Z_awaYAhbcxqv2fv|(0vOR@Vr0jJUw)nYg1$j2Kv!d
z&!J9uHni^CpsdKFU0}<V894>E1)d%d_UY3D1zX4T;0{#kd3r!S$>{+-i|GNQKUokh
zC;p=_2X~@Kt_i@o$t?Ihd%rQ~{nMgxsj0P`_iLk^_j)^8hWjs2UCuAV;-Ub>6<<m;
z92d-d`RqTBHcWn(He9hd-ZQAzb3S{<{m+LI6Pg6zf@mh|k6{#uP$q3b<z@DID*s7=
ze=K|bPHs3H%bByQbzJOVVNbS&lt;pZSXgX!b~RfyWNeOZkLs<G8i5xC1afHOBarZP
zBG?~~!+t6pkpBQ+c^wc*Mlljk4wl{iR<*w)S3=uvHZhjwK3A%`ZY@qk5GE|TA_P}L
z-SaIZb}}dO-LGF239nXAanITG-LS-#ORKMq8ewwu5!lR_|Ij_4PWoj$S8#Y+)K{8`
z^i09wEvO36@`NofHuqu*_3>Vmf)%r+4PkdGsA1V}>|Yn^m_}T<eb3U0L3tuw7b9%;
zd@12#rLfDwIsxI!dyr4f`jFKYSL`SpI-G?Sphj+IfzgpCG?+y_;V*>U9Wy<%qD+cR
zEhIdPk6?IK;dQ2Gk(rQ|8WOp8WCPN~#7DZzg$X>X>01anmf+y6^u#85?5W?09#fPj
zA^*-Tg+kG;RPtw2>&7}(lpgue$(Zp?NcaXe6lcknkW(mV;>IdR-WD29U9()4ll!~W
z3#FjmD|kw>MBMAhnUtm-HNx!1Q&7&aO%9_GpbV7w8h`R*Y&9<N5huU3Hwuf7MF6B4
z6VtiiCsEbpv`U@#DNLjxgCERMykjvXg+2?cD0g(@cEz6QdQzd#p(o!3lJuh|zlVA~
z=t;%BG7sB@xsmEzs(A`y`6vJwNT13Er0}HbQ$sekUl0YDGW|DtV#<^~oeX6f2S#@_
zG>af7t40j6v$6dqLg0TQ1lUu*2<X|^6awKIo;oeMNe5YIal;ScBtK!nOI(L-yW8<;
z$L;J6`!RQ;yOR6GOf}&mYt`(yyEoj*b^ET8$i}vd<!;-B{_Ki;i9|oLH&L8jbCs36
z4ZtNJSgJv=#H4J(VZPI^?d}ZEW`}?gx1<fv*!@4-K#uD6J-F9ssEMnfEA4}e!ZBIK
zDdA@e1M4mhR~K6iTsDlHynd@ublOwlO8JK61)d6*i(7_*Hj)(kMSkzzOoLId)DDMV
zPnmv$XlpSh*{vq_<+rEYHB;fJSW-kfWBdT!92lW$9Y0nyd>?mR)3t_k?Qrq#Hf}y<
zAka;XM^Ru*{A@KU__c~I9h|m{N9ilB|M_^heg+NTo@+;Y{cyM{z@3J*R<$rkC!*UA
zhpTaCY`I=2`qe7X4A9g@{>G^|>`j5r4iwlO%RmD?i3$}vz`cWbu@Vcrl(o3x9CJX(
zCE=KBBg3)yITWsQ=E|TSF5GSf>g`a;nQqOZ`f9pa4|9H$XgGEiZ$>c?MTd|f_}C$X
zf>XmHEZyZ4j<%XJyI#GE-P3q}{A0msH2g-vwQHq{(*W%WRNS5xuHiJ-S{Y)kw8Kq_
zFNr3+SFc7mWi@YC!60j&AODc39IhoC0uagti7LRU9j>a_wb_<E>ww~PTjmm{)((;4
zqv$tEg_2WuY9(}2W`rGv7|!52N>HR+lB|FmCs*68rjMQW@>RG5H(%l^>{hj2h?o(C
zWiQH*<VFq0+0o1z_J*-JjlOzS51dxXFU(bmmDW0Qbw6;B>PgjPTv9~Z>T*z~+}mgi
z7t;mybcGNRP>x6;fxxW<G662%;@2&MYeccaOc|k{HgwGfdO|0G(_m7G8@<t52}8c3
zVIRj(@t+45hwDhv=$4LIFk+=#EjQbT)@~11*PVI+^5H-};kE&$mGovB3lRkpitt5l
zvsn)&ckjmQV4^;c3p#2Oeq$E0bPknd0H7q2Rv{n-*=yrwVyC`q+bdoPdaiEF6cf$U
z8`b>FWb>r%h-HsDjft|;oMC-H73_AK)e1TrZQ9e2qEgdsA6y=80#8;8k#I%`h+E-L
zglmKOTGO5@xaC>5g8#w4%PKyx{z{e@oFDrMK&Z2{MBxRT2(krfHx*r6<XWU#378au
z%8Z0aD}J<9$92rB0Uh!sgZ^9%oB;Ru#d2*1JRtPKdQueF7n~@@lL%FOn+xMIri{Qc
z)oG^$37!W8uH)e9*zlYpN*ygYRl8g%U^4}2T#VAg;W}!W^l;!*X1F#x5iWIV$J)pv
z94_bD1mX)$M7FUqWTRQ&n=KgMtOHQstR@sK+Q8ur(&OR!*ju`1mAyu5Gtm#a0Tky`
zsQ3uI<j>>YO31i3N`D?&j6ZLuKbKOz%U9E%W%OY&{<OkTs>LprBh$g^onUoNu<|EZ
zStL{6LSLytn9IF5Vlpi55AHZ*vl#{5EwHeABO5HuT{u(@U^v?klY{XVYKzBXmq{ax
zc>zJ{-pRpy2SH6>dSv$5mA358UXN}lWS1{`boH1=SCe^kIgv+~7kPC3h)36xc$694
zqs-|ZWdZgmORz^d2tCSC=ux&8kFvpdbY{?_Gld@Q5AtaLkVi`a9xWAkG|%@a4&ad|
z>yfA13Rmge&gj3YScc5d0Aft|8&Ybr93N=?4KD|^0qLkt`;|z5Skbc+FQd|<u@*;V
zudLGgSo^R~jb<C#=A2|(;tL!EsNAC8BDDx>T*NvkXJlCZkptrWW8jPj+BZRc;09nQ
zBJg1VZBHGJgA-N<^SCUeDqlxlfjiW(f8gBL)u4#cgHysuK3D}#uqZU1`CZNAo+P%(
zjAJIeSVX}$dLKd!qfa#?%=%KC*}@HrxOn;Mn{K)u_-f7;DFvr}@HXPyCkO;}_)IYI
ztzg~{;)~o))Pek^SJDATiyd&ZU=ss&f)TdbvaM;D(4!XhajDf(G@^wAqEbYlf!a_+
z{A$lc{Bm9*E|EkWdBgsjrmma1mNY%1-5nY|Gs#G4*Ns%nWD&7)mHAy^|C64ve=aZf
zl7+l4qK1aAq`|y~cw#TcTsO=I8hb&Vf!a_M{B6$^{7qg8O8)VFNE8&TbGJsyObUW^
roN#^Q2}*s$^x}%>H8a91%C%ypRe}TN&ywGU^OA5@g<r>K{1g8NLGjQ(

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/index.doctree b/.doctrees/autoapi/lmflow/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..1473c10d7ec261596ad2f9f616d851f7d5764613
GIT binary patch
literal 8960
zcmcIqTWl>=8SZs=@7?>kiS`tGJYtK`78DaKhocB$^hiZTg9aRC&#b-IOlQxGb2)86
z5`Ex=O;+LqW8#Z%niw@C8WZE~!NeC0QKJ|{(L_l^gXohHjWK@zT65d8+uefFq|=$T
z{`Iea{p-K~{q*GL{{Ds$`6K2-pIh$UhHjd!XR<I6b<;EAE_0*g@#NaiBo8NRq7mr3
zp%(`ROT-8u3~o7Qz})2iM9dL*WI5qRx_&BTMr3(zzMN0;8lSp9S>p{c8Cj9TN>ifd
zblZ-%n-By5ChlQ-v~xcGNTO({`8sxt622%JT23M#7qyhKZ870kE=#uKmM)rl6a`i*
zj&LG6!MJ6HMD3K}=wXOKyidA%SH0EE*e^a>y0H_mHo?SVD$+YNmGF7K#Ao<fKF*iL
z9h=<ivdtiNT^4Nadcn@-`|r9F$H@MGz;II$wyDRFr~B5XI_Re5nrv?)2yv9FFO9`5
zJ|ZUNScxl&Vf^(Vou30=&g1tQ{I25nS~><e5m=J=H8J7qR)CA>dS>jfTUJtridi+V
zo-w$#9VL3qujA{*Nw)db;1{1(7hT-cLN*|pAC6lJQK;_dOy5v?J-~RI;*G+nA4)qB
zK&-r|!1?vyLLDSDAXZ-2_hQ!^WW)&W!84+Ol||<WCh;Y4>LAwlb8lO`wgRLFolsp;
zawXXo%av(UkMxj5Wss!`2-wA(GQjI90G=ON-64DUmd{9*OCaYeC?ao%xz>X^X<7-a
zx|7t*()`&9n#(K*%NUrLv0TH6P1d#?C11S7P2T35+uY$U_nD}7^*xjM5l=3SLZft?
z2I>y|>o{I3>4_S*OcU0jCuUP*`9R%`yDb(lGr2S(n$j4Ap6l!bwh1Fsh&74HJ$zY;
zM-$@mL1=ZKOo{naBGz{U-S=6L=w%h5^&jXLGYC4%QHW+($hDp+SY<R-H^N0Uq}rd0
z>@cmlt*}4RvO#M~Lxt9S_T_1fC~3`A(snFFdMkGHKr=i?3w>q`YR{o&WJ1D=h5PP(
z=*EX1ddtJv--J&51)MnDPJHD5VehuydgImwlD{zScJ*NYA`BqzM4U-o5DfoDX*v;$
zr7sP~>~|d3H1T7(5er=1(NaOCF)jPhv}QB&E0fjrsr6`xTGR==*dJgUxvMEK^=0xD
zR?OM8Oj3>}vGIFg<F|2B-_I=S5oU|bV20r1Wd!6rTb{XpAfe*{TxjO63p0*2q*8-x
zaJB{<s@{R_@X|u@wk>~u%WK(N4^&yZazQzgjq1e32kw(xUV|r_@tJE{Zb#dL2QA~h
z>6PHkE^o@QaLu|?Ldai8)oP(iPEsh>-wszUxJ^l-81wfLZ20@@ICR3QCgerj-k+<c
zrwk-&I4Hb*!Zn0R^xD-eH{t&hQ}~MZeS}Qogj!7Fo9+3IWmr-2sGQ!_1G170uw1#=
zBJ2@$49Z{yhLsRU@JlR8#A^<@1*x;oBQ|-}boJ6GnDHD$`3$cWJbtds<HEjX7Bvf5
zNRofgm+ghokr8-&LiIaZIgGco=2-wfgFSy+QJ+6lz7Ml;YTqS$1^XUzui=My8U#HB
zf_~Lcz}n#yitBROFPWxXol0YcCjBYHF5Z0n#Y$cNnk}{OAYM~geMU@|b}1CZzLOwi
zNPAG&w#NT?w0hX@I)bx@Xo!6`)*W9->}i18GxV6nW80pi<tI{QD4&jQ{8iTj`y#fl
zsDd0Tb91;Lkv6n|MQMY53%wpsA?%yg=Krfs1QJbuSXH)dss5$g4L?n)7Q=yl4&8w-
zJRC?&ki$)o7U2@V1?X>9byl1%1qCaP=XR819|-=wfz?k`SC>yG8hc%5EA(~BYU_T$
zLKa+R(u{r$YqLn}D>6taNbXB5OMzo4;nd4UuNpW~)HWn=T(}$Ar`};Jw|fX}+$c}F
z1}AbGY=1qK9lm^!8kY}Ed`0$UKasDH(d;=8wUOD4qXLw^<g*sS+aoX&=c|5rWzBL|
zlv_(v(rbr!FJ8`0OXQ^q>WG*?abhsGrI=%mX0*^ko*u<PIYShK!2#Jr8K}rbRP~_b
zzKEQ}RBBR`F+Ed!7wrB`&b))zy(XqX%Q^%~q878Xll0{ByD!ABzX(+Jm+-*F*v}B&
zCHt2-dU-eJaW^BPA?d9UYd=k(%EAvGwIKM#9OJ<R*NI?7ZJqWxiqIE^5IV0I>$+Yf
zlhZ!luHLY5!^YcJ^Kdc0I%F=ShoyTMUCC|$Z^cn2#*~>nSk%Sn?m!x7kI|AP`#&IB
zvhtLq5J%Sh*fpYffXvZb^Q~_!ZGHNPt^L8RPnNf~$MKaZnK<NpZydUbJ&)<CsjD0i
zBAc<fU>hC&c&@2mB>#QbO&12yeiAU!5TeWmt|myLAx?;SQ3vdgs6{OebEP8}lPbRD
zHBoQJjw7W`inM5INbWTxD6oIZ>64^iNl^Nui2{<;<`+bxsuY~ZJ^ze8r6Z+$MBcTb
zY-vfnj<9ir_I>K^GIW{16$hsfcTELI6I2zdo;6Jc?0bPoOo1OV8?Zk{pHj`r->$c=
zihw;a<$BrcTsFL}(zu?ONhh=KML?v-3v-}(*UQE-IKo8xp>!3&VkbR1;!jdmWey>A
zAFs?6*gpm~adjyaJJ16h6tWFB{`OCSIMW+-aclNdrLq3#R$6%b*X8+R*z`})Z66*9
z+kXT;`*}Q$w%>a>Drvu0<CguG5;gXNG(R*fe=NsvpuM*LLBzZ?oS4)`4-z9zmg+5+
zEw*c*T$zaZ>{ZiUgL}d8m8CB|o!69#hiSPRWp*U5mYK+7mvRA;zt7V-m+XQ@b=*S5
z><|#jw%G5$tNl(q`fuP48kTv%yYa?r_MJ39JhtDHPtoU<#C~r!RWTc@AR@+`G8^_K
zy!OjmUN$@+t<4NYhT!EI3j2Pd>O1uK1U<e>4=sOt0B=28m(P}b8uDq%CzH=MJ%=cC
z!BDXPKTGjtZZzPu((Xb&w~`qT%&n@91CS4tKoK++h}Zaqyk*03j9GJM0MEG6HK`!z
zfgGNL<n0}Py*L#Gx*IyuUu%Bgbr7?w_ReI4Xv0x8<rGGim}8=QRLmK1K+Pj9ZxzjC
zBdTR2>Zx5s|7$UCXte|BH%wGzKB_x0%HdgjB5Gb7`EitN^LdnKIt*<fJ%ZBU1FnFD
zk>2(3F&%hb1X7^^dx;p`L={Mppkdz~05sFjIB@VaOU{kjPc)0j|5<UO4Fm|KSZVlC
zF-t8rRMNG*wik3!OqvVP4)Rd~?o#6}1ga%9!W<??v|TIWTGsr+PU(WQ8@4TG0_J-`
zq&e&|10%;olj~0YIVl#|o(29Pri;T2Xi`j?tQB{#-5iy&sO|^6Oz&c(fy*)&9fs($
zTyRTcE^G~XAJK@T_WI51GF&*@KdG@G@B)qNuIZqv4*|g05J`x6#q$!DNr@#9%axf5
z3i~DKhnQ1%3mUT3XZt5Ai^T$+0a6!ou*i|w;;f^)ofujKj+5;iL$gDi44B~sriRW4
zYM|JL`g7Z&24!j^Dgs9=GZO`bJad%CYQ>NkgONnC5O;ko7e+8v4rI9%A)J>ZQWfa-
ziy=Q`vFT~ZY5}!?Qll3#t-Ml7i4kd+t<X}!I<J{lNQHGAB&BjHQxYiL3=R__Tu>1*
zq$UN^vgVLDLvPpwxid9^a5sd_#FbB@os)l}Pqavy2FXFf&LRC>6hZG=SQT?V^EIlB
zLO%`3FqDveNIPgJr4U!~DDuOt%}or48~%R8J$J(kI?$!l1Stssq!zI*t|yYoJoDGD
zx#1@8b1rBb74$Sxpnp?_o+3CB*&P;aSS)JGJzxqpdDL~VSp#(jXpt=J*Tphqva99F
znI9m{V|rpC+;<~=PvcgHJNSkCPEfTX5i=p0ACdOS7+M{IQ77fM`qUiH&{f|}EE^mJ
zdIN=TrElP}WFc1Xpk2|Eop;$mOU(e<ZJ^h3+mM0OE}S4kfksya%<7WtI-ahSk^1I4
z2}B01ITISZ4+=O#vAHKs8-Uu;7>e4ChHRNEuB3V~?earbowl-OJu%MQ%hZ1tb!APU
zzK}!&vletCLJNmV*s}M&2%;ca5&Dym8gv`Uj*An0qZH-KVPI?LA5?FrW<J2DXi&Yc
z*-t{p?H}R6d;E(1Rf>?lOphPZ_z&=iMT6jF5KR9QAXoHwuP1?f5)e-G98F0)GyDSD
zB(=2go|*ovxl)7#7jh4pc}q0(Lxe%eqvVq>j)NwZo=l*wB%6rHL$C0Ql0Ysbb+Y6k
zQz@DM%fLqlMzNSJi<vz5ELsT43xXMHgse#hr#M_5Ar+9L5X){0jlw?AQ<cFSy3q&L
z$aa;OPW0i2V+bN@2cNFuN@vjRg{?Et-iJWL4WiEHKqheeaJf^rr{j;~Vjf981ieM>
q2MYiJ2qCI!!`$Tc4NvF8Qrf1|jx(gYL;vK_G|4Kb2290w8~+6<q0?3X

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/auto_model/index.doctree b/.doctrees/autoapi/lmflow/models/auto_model/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..a0a9bbb58634b84c209ef5fac9ea876619bd627c
GIT binary patch
literal 9093
zcmcgy-ESmE5x39Ywb%Cf<1Xjq<Xo~Qy7S&~wv!WrV9WuEgapSK6rw1Egt^(7_RfrF
zXD0J;_7WWmpoFv1k%%$~BtFDv2ttSe2`C8>0*NO;;(-Sqkn+Gk0OB3-tM2KU*<J72
z7b#fsc)F*$yQ;eCSJhp9s`9nZ-!me9_*@iNwzs#Yo2KWREK2#L=^II#d2#wedj5;)
zjkLjQp}rgWNocT?kAT9kY}X8#mp-2ISt5^ZH(JZaPeja!ZQsj>TNSHnO+219tQxPx
zcI>j!Cw_<9-gf=nwKi7ABzhA2ns`rX42Fz9#P*1nA(T(2lFw?OgR3~T7I;m|#jwVB
zHRFDZkGr<V(ygSa^C>-!L%W&8keZ|%v(1PknlM~FiqMJoil?{b+oYKUMWd-3J0aU9
znuJfpdW$}#)||Cu%~*F@W7Zk|@Ve!<*?O3G9t+oZ{cvaf6IVY9QPm$X7_Cdj)=6wV
zu-BzB)}=B;uT;jm?U`(EEsP-CU004JO>2Z#L|;jlQq9)g5XM@D?(V_wS^Tcx_dE#(
z-GsIfr@_Yq-40V;$5M&QK2RUH4g79NvO18$s&2)po>=d;Rtx#Jth=CP>|YI=qPIdl
zuu81*vmb%PZ9TRP-E}+l7K`hKABHd;;)uERxD&87X#<N#rOxaCOng{wf@DG;s1oA&
zvUSF~2&GOc-UZhahTraT0;|H8mY3^6=o<{yVb+_Sx|g(r4z{trysS7KN+bEhI-iP~
zgPG#By|%j%1v<IgWDv55h1Xe1ylyETGoGix>$G&z!pr7W2cLK#vt;2)i$%LolyP)L
zk#S#><U(O{!V2eOy_vXrs2RSiMFBHPyL^~IjthD||JWl>-go24OE=VCA?p*xa=i7<
z_|nFu`!+Vo^`fNR*2B(aI2VkTFJ_<!;1Ph7eNOp8iCBZ+-B30+g&(r(t7O4bQwqT<
zy_A_=YP=e^LO%)mR7Ow<3v_Q$pOYA}Yn!o^UYR0u`6^WS9Cqn>Wf6yMNNIvc%M%cA
zH+{3y$0+z53;o@`eu4D_HwO&)km82Fe<SE@_)TZyTCTPF<{+3^%V{oOdrYXd0T7%C
zm}lBvOWOk+m)PL<zumdju1&ah%tGBQO=`WFIZ`d>B`sX(_k&#ta4Y!Z$AV7Et3hWK
zQm3qp&ZO?&VZYu$Xc~xDyb37+I4M7;A1Pv_h}|@zt=nE|{h3doMaY`+G2(zvqs{gL
z*EZ}py(vEL>Y<nJ1GYOG68MKjUf@$_L&VE;ODq-ZA+S8<?>c-Ng_pmENcn&C?kl5E
z@vCr&MXOru+^glCV*|BMTSU>~pNJMT&o61%u<fD&FN5%BSnwCh>N6kAz2Fd^l#4D|
z|2g79!`c5SxOwTYnO+1DU}Nx1e4MZ;MIHhHp9S^J-YNMhvd;`R>#py$BnIleCjSgZ
zf2ntLnaKK_rO`4^flx)CF9?_}_$yL2T_ru`t}OqdS*9>cmZ0S^Y)Al|1Ck><sz2a+
zG0-ehQNFJ~w<YJ^(UFm|ISqDZ;s-I+S<OcXmTrf!to0bn(Nel<1Rz96;;T74)f>V&
z>$W#A!8?C6XYESzN<3$I-UuJZ>7hUP6GqR{7cH`pB*#hEmHn51R2~sV_?)2DJ<pGk
z9(J80_@&Qo-V-A#Z`4<WTq~C^);HDc$UcxVZPas;`pV_YNPWf8W06-ROcguzXs-8M
zhMZG5R|)$0XkBeS-ik@go=Lj5n~(0My(aH8X~>d8%MWw1%<!E@B>6-6Lc&T;!%7hS
zq^@vwqvMA<uOU@EJrI7C6hkNq>7L9NJt&iI@M&!7Dl$~DJU>p{2rY|tS9nv>bcRS%
z6VmjE;yg?v=qR|zOV(2G_}n2pP7;rj!AV+RFmKNe;*FmyC4|>)wre24OZl97)pXCW
z{IIkEhk$xls9dHa9#AG4c>2RR6mi-%Q=yDJ39$YS2ywpM%eCC90v(+zpb)@_GcJ^o
z@+ndY;y+GX-AM?qW+C5tOZkgc<&)cq>k75W^{2GhH#8jU2oT>Aea|4#5WNK%@%A9`
zCXZLAVe+4(rihuj?HUqU`&8F>dl=|^3`_72A_bA>=EnJYW1uS(C@Dd*&0@<pfrl3l
zAF`G&k#-~FZ!;OKGjboCEp&9ldBgd1sgIw_r)og^%+-)WMHlNAvOgDgkdd{b^AgCM
z&y<JW--Cxzf?GkwX(75CSAw;_^-5usmmPqG|L7UijfIYn&x62_*1H8T{A<3F0K=uE
z>2P!@HS=1oQU!OVo&q}ll}~dZyg6UTcjp^;90+9@T=eq-<31svjBY-lS{AxCE^bQj
zyfkE^<v!<OGoPb8eySKR1ssv!OF>o>E?8Quk9d)ny9#y89Fy*;Ot*Xg_^8((%cf|^
zl2gT>BP4OoxFjDD<W2>D@-3?fd(VYm5DFx5`~au0xC>qR=moW}cq;^0q3Ol#+d=!u
zNBw>Q&PtA!B}da`j>gID)H<?JEcM=@OF6gWbDyne%93jp_^eV^bx)^4<q%~}UdT{E
z>dQG-=KI)%Ts!ny@Vqsg=Sv6ioayJ|_y#BI(oa<alhTlC7VAAWe7#{hlIz`~j)s%-
z)9u8KZH+E~DR@cGL=&#LPr>THga%7azdt+Q!53wy?@(mtC-Z1j3M~hR`hvs0Q2!@z
zDnq^VbG$k)<Ix3P&MW9a(0-M^QflS=njT-qRyn^BpUPQQ8A!j9I<J9%pE`ii#(6r-
zm&5%MH2D{rPlS8t4ZI#vgEZ0WG_sgL0aEt{H?rH5EO>Q^$*b9&hx1KH#wU>Mh-(k$
zThtVJV7cx2tMW2^pHFzcdYx6>tMVpmpU-3;ReJ#xSq2m3M01|6`V<|_MC(N?j&Vkl
z_I5Q>4+IrSLzXaf3t2u2mo%d?siB(5PCQ+TalT7C{VRHWj~=hk<NNvB5Ae3{{7^i9
zB%Uvd=a0qnCGq?TJ&Bi+0IMWRp8$MLrjxmjfWKK$LVd!me|EIi75-uBzH>Sb#f4rP
z)}|^3t%^Vxt?IaKknc!n)e?hCOMPpN$*>`^2~=0QBVTzB@=}s@k>3`Fx)-_Pl&YaH
z(85){EG1Q99A4e@$_SrQG|WUZA=yUxtdXG9;l)~BmYPvLvOL8nGfP1|Zh;C}qQJHt
z3SF7FL3~(u6BKe(l;>4HiGw6gx2!qb!?hSnP9kQaVd^Mw#d<ry4c&C;`!QI>VPY@k
zqwBZ;lq?V`?tow_YbK$K)^W=7snkVZ5ei@CC$=#GPUfuCeUl%jk{^m8+TOMwwsDU*
z8==q?pd=t_C=sSwl0D32R8@(zT|2fkk$XZ6tdtGN>J?jjCS-vh#+u8nGbr*HpR#l}
zZ&vsM+q0oRToELZBC7C;$(l(E^UcyNCS40-UZ%IvQ^Q6WjD#WnY!BMfn1_QL`6I6-
z@%HNdtKxKYXZJ^qg`pp6mhPD@L&dC0tm5Fr=cLX{vsA<^DL>QmrDqD~-|!$lE9WhE
zP^0hc{?Id+&yx(Wy2yfxTu~%C?&@AE(OV1}r-SJ+lxF!!6x00B)J#TK<XDC(e_Ol?
zhug*>5E`*fWePat&5{dgMHfGY^Cq%Nl(@9q7@=6vVZ&!bPsbhL7Vc|^M3_14e$f>~
zEHQlz#Vk<G<CfWvm{uMsti*`5>vm*Qf`_>@(~hXD0zon^CzO-G;btIA1OS!yLa>^Q
zOxw0`e5QKsBE4Y|l+5Vv2zx`bnb`7a)UfmBl~KMR@&K3NX3~Q7w{ib{b$*4<1}xC%
zQXKw?LLYcSg&}~ogLiTYeuovuLA0^Hj_znJ=-|lct@&XKzI2Kx1p}beV%Em?#Mq8W
z^sk=1_dU?(Y`AUoOiw+<^yk#{a9f>Bc87&)HjB5#8i0azD{i}3tcD9!c#$afO!70Z
z$+ng|XD&c?kG0R|qmCErdzxjpEEm79-*LK@P5DfO(pRiKk-)1XDC(qW%ckb~sLx8(
zo-r&WO$Ihz`Wv(?REW_#C}Hi3{{#@_{R{^7VLiTKd)u&q3>QuiP((?yhs<u1?Yh1$
zosnwvJ1I=2-DD;_xC0I>hN{9oKV^VwM`NgG;GAsYpAbZ){w80@{_6m?>TXLk+vj7<
zyH2%!J}J=z?h8vqc(<w0i~=6jDumrEfGJp344{S8m{i%nk@6E=y>zY;-I#4VZ%{>s
z{&4{06Vxf+8_pl$<IW%GaRIT>S)s?9)O;O}gx82xqy+T82ZDNEQ0@yV#Mfo&rwMRj
zy@@(oH7l(s%vS~9qFk|=M^_a}@jn6)B4m1#!u9*cC`OAAH?FS1k<$}r)-}PFxYiZd
zMGjpzI+RU{d^q99%eEz?J&ux*41Lgm>>&tIbW*M59<qDU3tMbAQ7!Hwy{)G+r>=F8
z)zqqDrW0NBBN~Kp3g14}J6HAv;TQmC_zA%mVjrp)E(`%%jyOB<K-T{PcI4^wK{zW;
r32=MtdIbBF3+6c2)jgee3lp2RJ;O~*GU*nf(v(S1oS~M?h}ZrHxr*X}

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/base_model/index.doctree b/.doctrees/autoapi/lmflow/models/base_model/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..387097620ab1db6addc3480b7c22bdb440a3f7e9
GIT binary patch
literal 8682
zcmcgy+ix6K8TW11Ufb~{ZA?R&c1+dA39+}0Kq|%xrI8R)h+9=vxwTAYXO8#GdUj?y
zm)J{{svxzB;Z!Q^^dCS76%bI6`T!Mb`%Iq^;(>s8;T?%*e&0EBW@gvkbxI47Y|ote
z&Ue1^-7k5m_U7%?G5Jp{#vyn8{SC`;{J>#xCT5(#PP@!cvNy8x-^*@hYoZZZdvTCP
zHp|2qAZ+e>PQ?7|xlAk&c;b5TM!tVKW_IERezBd`c%4r_m#y)Js3mUVvC1cLkJsJt
zg1wC{P6v{f6*E&lXS4@frk-Z|M9T=smor6YJ+weolJRBHFpF#ONm0+K-xgDz>$7Y-
zZCPU0N|MNJr3r3Lx14mHm~J#}dsZA{5zjT>>Z+$1Ck@M&mSyioY=>Y{F`Za#`jqiS
zeumHUd-)_kE1qfcpv#(3>iaBe?gi0q^J~|>f~)F3ATVw!!kSi^1Xk!a6*HQO8O@Rz
zP1kqW{zeqza`#=GNLzeN)Z|))mx|5&eq4sHf_D$#=Ry3O!_RrT7<d!8@;Yl`Dzw}v
z69$e-J@!Ros2Ifk3S@mK1g~!=nU(U-^7Zom+x$Lo8S-mDC{{bNLS7+NEI$rG)8WiB
zB)>P5yjPC;lFJv!A5+X#$gQDQC0Z`?v-~2MHKS>g)J)q!w@<lwO`KU>HNq&c8KmwQ
zt)Ahh-LMDo7^|zAs*x~?KAiJp+!{`lXzX{r%{a8kjAp`!#VoqPGNN@`)0opd2U_Qp
z@s-xIq!+~Wm0UxmiL8|GC8O*+XA~*zbp<XNwkWl0DY06qXGNwRcxD_jyCS{g6m3ev
zi=}73@xmiFU$}Hr|CM)rzC2E}|2H<WdFheOEi$P%?RKrGcNs<lWfUv9vx7s~sC>@E
za^=m$#rs{<6SMe7`g)Djb$0fv%nMnh)YeG2m5}z_aN#`-%ZL-pcdW=U_N<<f1cn_k
zD`AG^8()6>ijl^y-`1Y3Hh5Cb*M_Jk?I=jYLo6p33Jvx?qC94>WzTgIo?V?Kb$bmW
zcokCqk=Cf=1X@AxtttX?=&iu%9U>HTPDa7rp>-Jp$PFE_<x83x@rBK>w;8lLo7W4j
z8(YKN&qe`r`TDbxwQF#W^C9yc@Y>vmo34-{9{q1|YpG3pZo(qVt03in$PKYk(2|v=
z_`z^fgd3LhiHWe6iF(*u$E`C~_1eVl_aLu#5weE*J5k3i;jl9C(49q+3^AQ{bS>A<
z_=jQ|Z%HqNW11ui#2nt)e(1Tjn`F1-=RGU(vje#E-lhr&qarf=TW?eP^=w-nl~_@m
zC7Jll@nV!-;&+Ik|3~j|bsQ{y2PUz?>!sx0sY;Fw<vxjsqRl@LZ3xeAY1=4q(T2AH
z_)8r48?E)Z4VG5$*eBy>R>a3gRI=pXRhA4_NX?QLbC&$;Bd}yNqkjRuy>%ROe+nQt
zxZ#+@6nW7MnG{6vLqOjeL@7>HVuIlXep`8KYtT4f!S3%5?ydsy?{Jfjc?p0z&Vnxp
zNF{w`(rM|mt-6$FaD-(RSn?FJh}I)gg+;_H<afrEz)#?JW1Y_&S}#Q0`TY3U*quum
z*+b`2u|tydDX8_vfLh;KQd+0*h~O*M3rA|KNjGA+{?HWl{GGX#zBIK$bDpV=i77<z
z!)XphlLh)>#xBzQB#rv&mroazu3dnSi!T}%F6_crw&U|fFj|z`E#D6kYoIg|m%hFA
zK!Sv1(>NzDdhYT?V@n^0#0%Ntrcr<x=PqAHHZQBG6;V?jwiKON2%@-DLAt_KiCSXZ
z(At2f5{Wy2q~FTL_+B<>1D!T)In$xzt6Hs9Syv#rC3K!s;NL6`l4^4rYJ<S7c!hG5
zDk@!qB1=C#G;-)$jDRR__fm1_QANa>n1htpk(|rp#R;NDa#_CjMP_}Q7f0NtA#XEy
zl}eg0B5~-vehjHIM5>~5hRzvI;%g&FJSj<}BCX4i*my2#%nI8V2ei6SYNc4kzUrQu
z+*3d<%M(DX86a=Ak|3%Ak6+O8!&51zw?|%Hj0cCc@K@@VKgS=bN>X@awdSAzrpP|^
zvFfEf`GaFl_79)@UiIYCs}Ixvw-oqKB%z(3;g9x~Vn%xp#l6n2s~;yW6uhFxlH|V-
zLfLjWN~t^V;`1S!&AIg_d`nxM_p9gBl>^fZ(y!%t{~UeZhew^4^GKcd5qUHkT%7Kv
zUgDb6+A#}-#49xU05P-kV<PuV=diyKcU3CF8!p?kQQ>7`Q9qiNZ}UENJ+4xsv|8u7
ztJKIs@QsJIb<kh}yz6A03?vp=JpVIMJvA{vwUTYgi_Vjh&uH_Yc?C7k#4LFPB=G3K
z^;Z$1p6BfRwvHOSE@pO8&y%*H&Yv}tz$Ok-o^V^Py@E<Xu9jd#+t-Pss=sm?oPC;J
zWzt)Ca9brD6-!sFTjOPhdk@9Vc+)6KGOJ~8piI85l%J!skwa8;{UKsYMc4W73*%Wn
z8WDeALv$Xd&D2)}>in<@>PrKt`y<~P|Kss>T4w_|MrDH)Ay;d#&KCBKF2fEgWf=b}
zzs>(9CB*+;UWF1wR8BISo`GD5SEB+l!RtO5!9S2Q0PeGt(zTCUhs?S@XOWA>h|P*N
z-zSW<^i5Caci?A_@`csJyn6fl`)~h*)|gMOIoN4|uu=U%T_>O!SO_AQ<)veU*jc4Z
zmv*S{Uiu0OMQciZ^Q^Himl5^67trYelbD7Nk?nxacj;AT7u9z^SXa$w=%gR$#|6E-
zt~xviVm|+<-<JWAJBGLbTJ!_Grr`(^ofoqtp@y0C;!ovH2KST&M=sy;j$%<cx#9=}
z>bwW8KN<QYd>T<yp@J#VuJJ8BdsxQ-m{P2_;J4H>+^jxg_$1@dS(D0vNjpU|zu(VP
zRkWB~@QdQ^B(nV2lNE%Cj$<1gWHq^{B`7Ss?CO}9)iBJ#Yf2hEE*9)GqPd7!OhM-L
zikgOqnViuW|18toM9y<|BI&{%G^U=hycBZ-ofe9EkS1Z8WZQfZ?VUEmghQrvcyZ9}
zL=UtZqVY2q1wjH*5#shUG2TR9Ns&OIdKUn*`D+?^csoJShbAlZ6%pudadHO;P&Kj2
z@-1<KCQRtjoBKOK)J1!0A;w%IM5kJ|_G6$rqgI&5sM#Nzdv3x_naATAI4R$dx6!x7
ze8j>aN=%R4U|{5gnB|sNyw=1r+jqe~ly7OQfoh`WuvXdzz6ENIQMVCznbpNg1A?#_
zU54m$eQ?WUKGF+PXVFNLo%Ki8Wz4>}|D(yGD2PmsF0#ikps5q72<Ku^@w|eiCb4AV
z?7)`+6rB&qh8Gmxk_Nr|-u@2*o5d2{0a_Pxu*j3MixZyZw^OUlz;Q~KCNK#UcSXz&
zBFA(XbzX4{4Vtz^9R|09)C3%H9jyvTJr=0DVwOu{0$C8sElgg_LK(qWx!|IA5&0Gx
zHrrxRu1GZ}>6c4k%u**XF=GML5}KC*CW+OZQcLW_yy3<!)gZuSI&Mteeq1D{a#|Y+
z3~nBmiQ#`#%?_=}!OR^ONga&`SLg|cV3I?1J!C`J9Ef}lBh%vb>bO{z$*ITCM&i)^
zF50Tsmd=TVkcB4oT4A4<KY=DRu0v=$XeXl(_wXbM<IQFh%kf6oOL*XK1W_Bdbc!G)
z0f00U)`fTyh+_up>kmHsIq-8K+OY@F(@Kf{A&nkpYmv(CvS`C)$&Ne&uAs@2u7|@K
z=zhV9<P2d(oP|zy&B8c~3*<e91F;nM{KVQfx!dL*exSco)I-a}e2j@qVm_b3s$($f
zF8QurO)s#moV90dj*`@d;FZ0B%aVoIy^G1nfgC)`X>#rc4xl}O?fN^=f!r^gBu5da
ztpPB%OS<a?mNG^f3GZf5nQn_Yu;3mj;0(>`198d*)UL_UCihH?$jReMs+aR2H+0q8
zQQqu8Ofvrljh4lX@+L4}Xd<Gc3)@IASE3fU)SG1x1<gv}XrVO@4U%tW;$+_{CERiu
z*mjB+YV^~b8sO8ksGiq4Z^FhquhXw4!er+n{d$#Ne}G@9Xb`N7s5I)9&<7IoKtdtD
zuF^UoKzDHq<C}V3#b}?e=a6L=e5;7AI+S809V0?!w-9$a)04oZQqd{hvqYgul=175
zEE&dS5hF{pRGg@4l)U2!%pX+1!@boCISvh{m|0yRZI`Riz-|kp%RbP%2NnzHMju#1
zpDHn(?8A>35G5Hb`qUt<{0o9G+?pMv<g^g+&}jA`eh3=Tb^1!a{s!dma`qHV6y+BE
qmj`GTv{D^dc(=agxnfx=)|~6xUh0q{x5*dHI^=+JU!{t%jsE}xz<DkJ

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/decoder_model/index.doctree b/.doctrees/autoapi/lmflow/models/decoder_model/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..7d883664ee6e810bfd5a144bae6ee047609f6332
GIT binary patch
literal 9995
zcmds7-ESOM75C@bUOWCs8$*?(Od3ko!QQ1U56HNoO{xkhjghJfD5RRs&fN8#@$Aen
zAGTLjB#>woR~PZnX$236M;_s!s(3&MRH;<(4<LkE5Kxf%02ONU%J1AeU%O7$wi<ZA
zlI^*7&b{}1pL5TdZ%@8Da$!XNh=nNNwzt<bEX(sP7Nug^^39~nyf}R&UHMLWBW;M8
z(AbUqBs5tnMgU=Q+qFXGrO%~ep1@<<jhfl|sfd}e?R&*?KFMo*>bbPRXT)S|$1bb9
z5~tm6$MtucU7U_hEN0>%)a5XxWmq-w4BI1gh6BHtDr{?k0o>x0FNqnwAcl{NT1LaB
zm~d^6rJG6H5OpJtL%W^Cpq*$Px2=finlfD@iZF@iNzdr2r)et*%0b&Ow?oz;m_$s)
z#umLw`2s)9=lB^u&d-WxTioxmR+xAm3tPK>xZQf{+LItS{|5x4mO`v$B(ZM<c1y9O
zrC8GHV@b>QEVkDSBhYvM)v=__N5rI@t1wmZn|})Q@N?kcdHg<z-wXI%A=<#6(3TW6
z#6)1&VJgn!w8Uj!)P~7JJfNV~h5_-~W}F%c|14iC8L`Rl2fHE58A!+23XOnQ5Ef57
zuK6BYBe7^v((M{yPxCuk%$cT6)cjBjLLUpet6I#$uI(8yv$S?kGqiw(wr@3+x-8$N
z%`$ZY36CjRDiVx4vXv;l#Lx0o2xdB`95jDPSrMNU?>@zh9Y&ODZP)O&HKMnvY0rSn
zkVUZ~*^V?)3P{M?(jx9BuBDm27aO*x8D8DrVWHu=8jRHp?SM?}j;zAoUy%@4&|B5)
zSnC>Km>c=^Vn?kFRp<<E$Myn0BnD?@fRin)lX%j!BO0|Jkt8y<Se@+|-N0p0A%OpT
zI!oQyJicz!Z>t;2&-4P@G+f0~E#v8WuG4pVU1D<m_?Q%qN?PV@+x5+D{acA2_j%}*
zj_+%aYFEe~KI!|7m3rF<F@fpklbJ?bHbTrRWJ`T-9Cc+*<<S?EbO*<$dkGS7;rxf1
zvXhxI0ZVT#nKWhk-M-)mvp#+99Qh`bAznz2<R#sp2OHMToy#d33Zsa@IWI*+kP<U{
zU3WbS3<|^30UQPkcUVep=es$M8P2DG>um0J^60ymPlzcgOZa~oUzf{q$<-eoovWlp
zuPJcp`|iWxBVsW&+KFp~y6L-m6fm>m?hlJk6B1r5UjO<F58Zg-;T!pXN$XF_<An2H
zi>~XBJhZ++@i>b{mk@3dz{GNvFCigmR$ixKsWQ?*dw-VmbR9oWsNCinWjgixmzj&W
zrTi-boe`4`?bvYZh-gs^4{L;$whMn9`<fXtggdxh?eWW3;35zjl<+8}8aOQD?u@9#
zTcMu>2N+L@81&dZrc|b}WY@N0o?fk!(!B*iya`$VCfBRO#9G<l8`TYnhPH3@4j>9V
z$3uVjz`V>PDTleulG6KJUJrWfe%o1pzRGxQV@U4OtU_LT{<`FG1KK<nFwX+h^*!W!
z6@0|wAN!D8j#I9UyvA@VyYqi%20B;8D=n?`)}5^siJin%j0L?^)Pmj`Sdp@79!N6y
z5JLQ5m@5=DPz+y|ij^a$rzvg>w5W@eC3VI|M@GaHCgNTI&oxeXh*^x0W4N|y$LUS^
zde;cObRP*~Z(ZelhvfV4fxUH^o}`=dz}N`4Se%N+Q6((>$v=^$|8M=s)lo40eOS&i
zua&ZVzbZ>MG&CX0bnXLJk!aFZ|Dct_WJrR34}gEcnSUR!*3w=cyE`6fS$zDB3a&V-
zM5eq-9X0)aU2(-($yOebF~d3jXK?SG!*=>3fZ!1Cypxz9ubbl927&w<(7znqQk<%U
z5zY0zE#=*fLA(70R{wl(brp!G{8c*U4FKlR8LUAJEAcBYxki3RtWIUuMcSpvS+fL#
zV*HQ<V*xP^mr3mkt}XD>MWS(Fz7+KKu>_5r!{NmWnRc85O3eY~KH3eqM&T6kT0By9
zvsy#?77e$qA^v3LIPRwJW?aEPyBv>*3A)r(13*?d89|u^Z}iAUi6KtHzWn@b73;=D
zIKgODyLfRM{&Xu^DMHwS+`;fXKQ;#HGveWIZk&&i(yVJ2B+(Zxt!f+jiOAlNa;|Fy
zh<4%9B~%h*NoQG1Di>Tz^3_5v1ziR9YOe}~VpLNBaG9RUO@2JENq?>`Mt9RelW=}S
zOHMm<d{xPF&CaVInPE7uDDdwW2T3hD4=qBtSG+<^j0!q=9So`dnIWoBTz`iHl{CIl
z9DO_A))2E0_ZsqiiG(;oI7wd1abG0QX?lGKO*4|FL4x`-i{a^p^Y$SKPZPol+i5!O
z&WOG>1ksZcQ7RL488RZ*My;zNf`tJMFBDuUS+TCVW>T&xAeZC`KvwsWf*Y|PR)I$^
z78-f6qLJ?nC0>mB2Q~7qipoFXpHwR;LNa1~Pz)1ftNF2NQ0mF8Lr(S%p8R3;<kI2a
z$rj#G;D3;aI`86N?nuRS?nxB)oWE6Hj$JHxMUN%Qi(|vWw(|itIbgpTx8|@kX*(ZP
z&#5VAmNO_Hmgn6+{Jb%|K5$;fBX!;}c{Chc+}BOq*w(4Mrxyx|AvAd(;nev(h=Z#-
z=zGLHmHXw6&3368o{EM1Q8zr3_o?e)moCc0^ThbmR8~h^j)qm3QSQbP+e)3^0?Q)(
zcjhr7+}DZKRos>HDmmi31W1`spqdV4Peq;l0@8jo2KsLxqLyX)&M7)83njcJraOu2
zN+VHN)b-dmiIvDDB&?i%2^S6dqJ$(2zlXiDJmowveVK+bLoPf!6+fv(P`VYx#zv)B
z`^aSj<^Of9xZw4>{x;sKE=7|7k?=70d1WIU6_spp#U!o|s3`weVPmVuK7)UhFLpY#
zFZD^5#RG?<SV4#nAk-gX8_p<+<j~urV_}<NiD<tD_1z3ZEFFF`+^^gWoqK_h68<th
zk$O6xAS91MH)XN7>d7GgAhd$SF)U3W_`I4=@Xu!4f+H`lOxHf{m8ouf9L)<<drVg3
z4|jRj(o4Ra(ZJ8%F5#%&9;(!@^i%(0u6cK_?&<shhz%=!=9d}NISWB#vfMaCh|brD
z=+X=1w3q&hf?`7{dgg%oaw}2G3mHGfPGSmvOcpYnpV3g}In}Y}uc>w%RM+$K$N7AE
zO_i<o#a#9>A1?tSb4yVHwBY&qTn<M&(RmSzW2yqmcybd~AaBXpttgXkNuyX$j<z@g
zDMQv3%wNeK6HbZnYRJtG*Q|2e2H(hgvTHa14cs+L6*<)q-abENXrK4MiR0?dFRc|#
z&VJWaRn(%X%U8v{acFpvD{nWtY!;#-FPplPF)lrBdUZt9a~Nh}n2?^2ig_~$sb#De
z4dS_cMm37XbjEJ<$CpqHQw{G<D8q<_^4eDoH$go;k0+w$CvlL(=_X%5ZEB05`76^*
z3@j9KQQqwas9Md2z8?cs#M`}8jJ8l0Qz%gCunmBEHcUbnqZ1Sasp(8_kuJU`PNKsS
z<3`6S(>KKlYQdukq3?D4u#58AJarQTl;UOYBEnXu)eLhPbu%J;*N(X^Zx<i~CuIw=
z9{8r13mG*8b(ifhFmg=Px#1SWNwLKCZ14|P%OuKyCdH)1+Q}C7o2TL))lac6GrE|W
zfgnsq#1MYA2X5)iL%KssEoPFqv-X8GnLV88f2p%D^h2GaK<+X$u4{xU5(}}QcwX6M
zQtpz9vjcAiws3BdCC{sUOC0jmXZl|ZEEbDI1GFyUV38}EGbhm8-Ad3J1;;6U8^fm(
z;+_ze5n8&%sFsXlr~$tzYB2B)k|c1%wsKWK1~gBFAibOtV@Szx2}Em2FO(6Cl@m77
znb1RD7uzhz8L8$Z{c<XZSYr7)nn-|Jqy{fCnCeQYC1$Me*pW?lBJ8DGc0}z35Rzdz
zl^Y2RZVtpm@ItDd0IkWu^p1@rlfIl-rY9VNRu$D8ARF4vg2-p_O-V7lIx3cAvg|Tc
zyg0PKi?Z;w#S3CSV1Z7xV%R4dY`_Vn_yF1t+{q}!CwUwP(R!<e>8Kg>V(xoQKiq;X
zogzqy03h|4p-Bzfquv<{)*k%a=fThUuwxEvPc!BApUJm}*&3v>+bnF_EbhoN;0jti
z?z%W^28B9Uk!+7li?h(lu3i{tK|ogj*cXdY4;}qIo!eX7#V_=Cg35iVn2YfBL#*1c
z5g2uk9LugOEXzlWSh4o3$#F$CA$VnP;Id>PR&S%}voF8Qk)PXSZeSnU<D0hEfevJT
z;UqbVC}|IX*<I3I*Ef_g($_NEDO9H0W)>{C2MjnvrF~zVG6A)%GgSIr9p4b4DM{bV
z+!RaMS3uBJx1+q-z8GiT4t;earj<8=`9c#BBW>75jIJ5g@1@=>0V!x!3`YyCvFIC{
z8>u+iH%m#soW^dQVnFQ*`h*AIQ#7fb8_w@w<Idakhf&_xray1e@K^Ygh#7*Fag_pw
zgx;5s`w|LK_8iUA2IvNFpfgp=ZdJL@&p&f2EB6~kjLicoKGlg3C)1mVJI<ry*h?j(
z6DZPVJ!Xj+e@&t#ON8<|C9jH!I8oIoN#hChPgKAI-D-v$M-Hc$S)Czmm$T5oZX3PX
zKG1syCJX49KCqemRJqg1zWosc!Z?LRpBmgNdxIbhw`TeYIW0sy)WltgAA&}7ow}0E
zzXCbDm_7{?#jTI7TF@+Lr8=<iZhh0U#k5qcS=%$+#3DuBB41Re)C6G|2O%!Y3t{s=
DzX({3

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/encoder_decoder_model/index.doctree b/.doctrees/autoapi/lmflow/models/encoder_decoder_model/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..09875256350b7839064bbff8ecd67617750d98d0
GIT binary patch
literal 10231
zcmds7`)egv9nbS7xyk0S+l`j()>*f8lWlUZZ3QK!Yr9*Eu-yn!Y^z(Qb7#)YnaSLl
zaUOY478GpRlq3Gospt>lkN&7Xl;RH}N}(402SlV*)GAcaB5i;5^F3!?_a=LDW48zf
zlFT{tJ?DI1=lh&*j=%hm3q$fp%tZmWz1@akS)OmPC>4{IZzdh)#pz4w^0(3(>5`ZV
zjh)C(LX)Lp2oNT>T`Oc>`b;Wj2|TvlsFAIoh?p7MzE>>g<GjWvo=KPZlo*fg*kzSh
z;=J2wyZ%n2gVV7I^GrO1y2aFgIi1pStR8!U?GjppEqXpx*w+FBxW*}85L0?V93K_6
zjHV4S=Gq=hH<Fei>P8%gb}Na&3!;70wj!c)!gP%&!X%!@J)@(ZCaokW2Q9<g3R#<A
z5-|}QoAfH>bNmdS;dk*-epVPw?sr%-OgxW;%^g47YCeALF)$$i2Lz+0Lab>dv2O%+
zQ?aP2Skyd}MNQkY*lr_?K<C|8N0JsF6612N!fYi6{t3{<&w-=o@p~_RFW`5XXat)=
zTT-?p#sb3*Q*j?oO<eYQZ4fz$iwb&m5MW;0h*Kls_wm(|K^y#Thyk*mg4B%7&<J>i
z0P*M}n(why5|<VwosJRiX?|OaIn&h1njdOG=wo4LMT=S3u{|SZme$(S3@u=x?OP3{
zPK&o`vrNrE!oy0&iUi}9Y$l2?@w0pdf||@J2hCqlw~ddBw;pH4HX};4mTP!h8qwR(
zv?oAj$f8)0Y)2X?3nb=kY7zGn*V0Viiw)b;46p8Qv(Ru|4VT&s?SQW2Em?)VzbqlJ
zU~fgUW36L=VQ%EtiygHVRHNOy9or53kQkg@51edi?ZlHNAfizd5=kOslhxU-(Ft4@
z6$1Fbr?b?JjpOS^{ZQRld}1%KO~X|@)iR#0<vM+<*Ci&`kB>=Vz~s%GZn?g>rGF#w
z<1S0S-1dF#A?*s;$;W(uX}R7qLQG(~@mQu2myHnf3fWTMmj+##Q#ts8lJ4O6bT2>x
zE}TN&tL$V~nSiC0mrR;4{Z5w?#kD?j?i|@BlfjQji{vGpU=Md#J9jRptS^ir2IssG
z^+8HZ?RMO?C@{!fPX;g;EZk-(nVoOvIA%DX2Cmb&*~z{1d_Ey2q%7eNr2kzk$0b*P
zFgRC9i@vA8rR}>DhYyMQ*k~oL5$dMz>QTVVin%|^UyVt4F@OE3=kCAp+ygiA|B}`p
zm&b|lBlo&%58l7FPX0RcN|)e<;mE{d7GEK7X;faPVxcn9!GrFo=;=CszNliMM^z|v
zjRK^4{Y%V+8&l>Me$R+WrM7KYc{sQzhQT#LOWT3@j(yDx8N45?u=dF1E3g>w6iSv9
ziS?YBF?&kX;?2-cf&+4(_zi099#cq@Sh8bVF;B17Ne5qtTwa6df1Rt@QBtmK@U`j&
zL_^EB_6{HlJV!%+=fJ!SKq<sIWJzg^o>~j`*8G;U_Us}2*VYG!Y#N6wy!7mK$@wLy
z{7k?+3!>0>5iC})6i<FcN9Qu1aBYM{hFdw4|2w<erw&0nE${gUx4!2H6(u2JB-l$u
zE!bNH_fl4kEJ<wdLGJJNGp3>jcEB2@;{Ia?zA17PTHHbKlR6{ALqlQ$6Y*XE8#qe%
zh-r)wt+=*n$LUS^ddCR8bRVJS-kORY2h1#BRrl6pz>{vsgJUDyWN|7!`(J8@G+6&c
zdhow7R9A-~n|E+07kRA|`8!pSv%WqM;jD8fn2v~+w)#7*>?dl{hqnRvXPo)B9y?Ip
z#D{N>ty>fyexrgb&I*w!bGbp|{8toLoaJogfQ;$S@jrokZyvSN9{>bXd+VLV7}@g_
z`62}JD?tCEcS~`);-@s%_coQ^FnX`|&$0Svy{oH0JmoLbF|PtJ4+Fp&ILQ*fGRrpd
z<g_}K<*Bqw5gccc3<VQ?BAGdOPh_#$734?o07XP|V7?Ue*0BVQ;Kt#_3K`#=14@k^
z<sK~cu92fgKp-w3VwANdX?Zkcc>`FR<qu($x)SsXby<!+B*rMWtU8md&NGCn5Z>sK
zjS5Jdgk5REJ%`Y=bP)zLTG1|E+=5NsjFt;uIVU$VJkO7fo~n{~;Opz>V+2-f+6Bpy
z3zt^3_55JOk_dp;v;suCaOo229I_C#D8`lfF2((FA-saTf(W%&g-|i9DWAGXPvs^*
z>e-|_a}dKjY45#pen?BsICOmVCgkd$*JiTY<GiH6zgrw6RqZ@f4RM9y6)JgDcFL=F
zNIdQtKv9aMd@$gW=2r_;4l{^LVj5boiY!DTB~B2IlJ|1l6(!^}zS4)rDM@25frkb!
z0|hqD8%Ll&N$4xwC+X~4qy2gxv`<R3sif6m2*F$%b-Idx=6dwMP=BR@#k%U6ak-{|
zT#zRKS=~o|V8nh{1s=Xw=<CIbzP{a;crolB)Yrc#D*uRoQt6~X>89(wLK&l*oFA(O
zrPh3K#L3>lli#bJT>6Y#-IO;K`0pj6&Rh7G+hQ@9+Zn|@=dab5BNq!^(PN47{765~
z@4Smm4&3gHTQgXiw4C>==hT#`%Ndjp%k%E;f8GdQA2=`Lkvi{~Jn9cF?(8IPZ0l5t
z)C+~g5SqLXkL>&o#KDjqv`peuB?sQN*$!3qQ!$r6>V{|XE_FRBBcxbA54l%#6}m@*
zRLN1g$5Pu$o!<c4BH?gmF(Mk)h~-rplu0W&<GcV!8JnOp0Hsewo$Lb=3^YBua~@I4
zath}(os~HpUK5k;#C7FmQCQUV*f)u#2umcaoPGf*jeJo;5{BQwURlU<9+<vJLm6xr
z_MZy2)b1$Fi(=<erNI2)Wdp5@HLb`uyB$OeZ;7qUZ4e-0C+0q@?g~d0C|g{)8rOSN
zmH(@_Yb(b-gTJ3IcG|Qr^?nw`Cyz$4f)MXPsM`@QIm0B9BlA_q!Z!V4+HMW%%0WXc
z?Vp40S8`D24j`n2zerD{p3cVz$wBC*fEd}M3<d6mR**RQ1rr3HcOVG<>5N-2=w<eG
z?ZakYb=%`;xS%9uvLd>;&AXP?^2Lk>e)h01q?(9SsbB7<{`p+<ZeOj|`92WqSFX+T
zBGhIJL1eOAKSGGklSFiB1#;R;dqvK1Nhx|}fVv_sQOipvKfz970(MN6Oq`$6P{upe
zvFER<mLgQw^Yh2qe0o*Y*7n6r_A(zY03tI>Q2{jP`T1N9M?2AZ5sPCgRZ72d6SqJn
z+*yK^$+x6Y%qc@#9D&dw>ulz)WM2%YM0nLFiA2U$Zo9<S^FH&cw1tGaY)Y?Y&1p3a
ztDm1XP%rxc2BWH+Aa5@kx80_ys<=hNny-jE;?VFSSLQ^zY<QxpARE@>F;XQry*ec7
zISjKfOi1sC#jKfx)DG8+=J!lKqnhAiGGjl!1z12iPBr)2p>!)23VdHR+yp&=+`ovL
zpTt2DryG0@)v!&5#<mPUF|bg5MoqdCpiVX&`hE;l;eU5iG2BE&PN6^{#TEeS*)R!R
zj82eKq=q`ZMHu<EIEgM%j69H4rf-T9^tk~VDf({P4?8IA%~D4+Ks!P9Um|REM$Isn
zQAZ}yckGz!G6w+}I4N6@_4hZ#OvtFIsk>~OfsrGk&JDL1j*A7hYlDACNRucB8W-ah
zYbBf5Z<fl6RNlqD%;;cd3W6{h5kvUd9=N4TT_PnHQ%T%j{oJaIChqEfsk1QjL!G0l
z?=m#wYlJES4Kb&9UfE?_?vjeLJ#Tupa6X`$J*)ODamZKS)%~Jpv6v?spmh-ki(J`Y
zI)MiDW`g!EI8I^R2tErDr$Y2BLJMDUP^lZo&_@LuqJ|6KM$iO~*jBCzh=^vX#-x{1
zVgw-?Qbn}K^g<cISUF)Mya_$@rm@YOoRMly(l4iih$WV<qY(zEd1_K4qN%QwT4Kif
zwjJ4&D`79)vLot=fRGH!iTsk_!p(q~2v$i|A)qxGnBKP0(x$IM7U>Cxpxs853dn|b
zvmo+md_PkRuMUd^8AQ7b#WW7>@1P)kZT^Co4OpO4l^XXGO+w&=PC)=|2kvw!#K(CY
z2hm!yiRq{j?8V&o8h*HmyL6f$B?5rdV}?dM?jH63Sg?BUr#}OJ&W3HXXM36{xBpbW
zJ+7@mD!awPhRx!(JOien$>WZT!=_LJ#4VCNok?*PI@!^S%UKYRRZaHAe6)wJ5_WZN
zZ*mvE(BCntB&K2}!WSH|Y9&Wt)G0ZZon_oCA8l^M+OsA{>TE*r>b`-?l7(2kg-+GJ
z{0d2aT9lcAeQ1ww+FlzvklBTkWGJGf)dOaCNOxV|ppFb7M#cy#(`hjaH+T;iaE99c
zzBp|HYD;J6F}OOumqKHez6ZJ~7P2p%psQ|MS+jjH%Dip*qDV|CYl7<wO@z<1aW`Ug
z{HS9f^=1J`L9=2QT4;?$-w)kL#mTN&N&@6Gc54>{YM9XHNdTXqN%g$s{1$iId4v8i
ziX&U}=QSGs5`PjgMX=J(lEaYD`x0_rLc!~vqj}l@9qo0r(`s3^mD~LMldiHUuwMAw
z+`;1WqX>R8y$QeLJT!{ERDe2$`ft`JmzeSINwj2rQKnZig-pbWszyl~PoQ(793JRa
zGh{e&IK|BB3~9TZg$8z7=r4DH-qABzKu>jnP35P`olbV`4<8W5DQ@)X-o3InaKbQa
zrk{|}g2zL--G%reXn5C&E7|-@ki+xoS8zp<1ybq-&4N~{0}JccH9cKSOU0VDJ=0Ar
WQshmtMRf{M;D&J!{IX0C8~*_Y47lI`

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/hf_decoder_model/index.doctree b/.doctrees/autoapi/lmflow/models/hf_decoder_model/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..47a10dc74db122610bb3511c6efe97a3d04b61e6
GIT binary patch
literal 101597
zcmeHw3zQs3b*MhOEA8rUS;Dp?yCvIbwH~cL7XGX)fkAdGY{|y>qacZK?@aH`v}R|<
z(>;<F;};RYw(REX0pnm|ApC4%0s%sZ2@rTMoDfJ1JTMpnIVQvg41^Gpm-k3`_tvMo
zs;B2pb<gVacyYpNrl+dzw{G3K_g3}M{<o}{HG3BR7cBAWrE>Mis9h{pYemOv1@nuw
zLUYQg`mOs~8(!7Aqct21HtfS*t=T9ztzZ_kD3r>TV#BGn4zz;B*xWByyiw7Ap63+&
za;+M6FZGuOO7jl1hD(D%zhCw%j`CA*c4ca!Qae041<+xT(nP-K6yS3s&z`ifCv@z)
z(K&(`b#4FJ7RPy@ZUcsXtF$Z_%!eeE<^}_Tl089RrCfDddz#~RFl76Fqdea9fdfeK
z+;Y)F>gE+Hw&%e^_}pK$r}(G&#b!PFGHw?RHk=7;(+uYM_9Xt)DlI9kDlIInF3l~i
z3HEO))ux<Hjb^p#G&UWsH4bjNcK204w)_VW^fqzCHrY+TX4lJ`IAb<(#%zi+W>dLZ
zbdHQRJRtGxYvwe^OS6K0_Dta}XME`#AgQz#_;@b-?|k^*didW4Bo8>$C^MplgTA_5
zZnT2u0>Ea)xy<S!nBaWgaATKtrGY(u%Wjq~C~b^5vZr)5@Eznj2x77)8+N^<puP9^
zK|60Ol|2jo-?o?xtb$#sI7RE~t9Ekz*o_Kf9WIp%CEn~Xj8b<Rmfff|t3}Ii*i~<$
z)|he{p2e7G+128ZUpwej%ddcTLbRjSF5fEIp5-}*oQ7SorX0UiE5i3Cj1157`Es@F
z=kqyB_RZl)2%ucG{F3bhNN1?&0d@e7KQC7&S)&P1v5kJSir@W`!x0&^ZiJ6=TmjJX
z?Sl?bTn!Pi8<T7zdP5QwMn)c<dw@){>f}8@!gnU8IYvb<U#L|l%9A<47r@J6c=<J}
zQnLYPfLm`k40WeC6tX2Gt-(h^ipzEd_{6%f_DF)~_&)sWjq-W9^1md-ROXbHl(2{}
z`Ch=y5ubc^N3(!)bw_DUDF=pSzMQi#Tc6LDgHnI+cGO=v8>IfiT$Q>O{)GBguGX8r
z<%9iZGXk`)*}!=tf7!JYCOOO}Fu`DsVC5K^vb}>Iw1Amau&dU%V@<%^0v*OqTxYe3
z29vH_op8Y37aXaXr#6m}O!lOg$>KzrJk1Oyx0PPAUavKL&nlKD7+-+@s=2Ugb!2p8
z<JMQfP}-ICISW2j9sIXaYuNBF+RXKK(XRVW1HN*53#Pbsm>IhWfNJTanXqaTk<3P|
zt80hR#)9nvIdT9)DEV-FOV~45lF+7F)tYZjJHAz(s#hHJNt|LR-(UAa?^8$L#wmX4
z=mR;6eXBb1Gg8hzBkk~;pq@RNIr?E>0f_NHgJZ;pNQd^P*P&&JI`sH;YzeHOgkU;z
zWvVx7hsxj@;!0er9)i^q1OXEc6k@_Iz}$rChl@+%k{2z)e6itEUCiom$vk8?%C)9v
z?YZ`PSO#(O%N4I;S0|hHq?q;q0#HLs!cZ$ulnXX0a*na^wx;9ZH_%z{u+xCXp|bqu
z5f(|`O)vcwiu4D;$*oOk4nJ7<RcqIRqfje2;Qbb@@oB5toT^WQn`*6HE7pFx4aX1Q
zyuusrywn7PN2V%cULEuZ_7&incuwPx)4~<#h{RE>L}$SYv_P&va!0XLJ_v@ch&*_<
zXIQB{irDo!b9SK%e-CfY7Q?fdm%Y@t$D0+q0lrox@70}x;<<KnxcgZ9VCnw7w{N-Q
z_N{lwe;LJZkH7^}$8*OpcHu>1m*Y0WYfeqsjp-e*g#dRvSSdDxFu_KZpIgB)<y#)e
z*SByBhv2_TZF16S3=PQ*TYtQnV9j7K;7@{wUT@<X?hio~D=AqzAG#ba7yVM}njzGi
zKLaj)0tEG#RE=&jQ3iOKIsoog$7{vuHbeo>xsBT4w&!evi#w`>E|Kqa=~#VwtTyhB
z-O5>JT~2SHMtQRxx9(@G9fp10!n#vMbDKW``%VRo;F7FkOR3GPl)<dq6=l%U(_%Ue
za70?sRQhq}sRY|ehFmbGKHUli>eCy6<d&oEK9Scy1Y!LEHsBq7(O>{b1hdl$a%uL(
zEr?lQzbV+FwMzdQ%!6<KbRCS$T*M<-0N<S>^-8%=_FH$epAXxOYU>zmNT<iZ#e^Mj
z_gDj1tm!ef61Da)fIe&l9KRKuZzNJS>Aw%*ffGL2ubB;e{~k<|m8F45oZnN$>2wqc
z$_=&$?jqQv!wvz5`VRJ_#E5;KgoaN5%x}jgW;ivvOjHhcmAeMu8p5K9_&Ol;)ovp_
z4voN^bRH?_Lz~rt5Dd`xacKR=@gaj1iunXrsXEDRqaB|OABNt481Jn%D&1W=hF~6n
zhH_#00JGf@ep)QcmLZ?mQxO2dv4&tQ5vK|lIbxli_aYGy6ILPo9_o?%GE@u}_%hjP
z!XoS5682yyJQuX&!I+kGieZ)BS9(8^0gI<H%yFJl+qNaGpQR5<oP#l367^8k(9o_h
zsDi$<{ww&?o^MC{*N@1d#H9EW{rhmH<xQ@S?lhZ1nZ}Syq#WZbI%1`7=--DsPxoL~
zxLU<{(QQT4n@20-KS<cY)iX6TUl@NR2X-GqO~sJ(%3!gMMzk)v?+Vp9X#xzFe%XCh
zbl<^5WUw&0bD*WV+_@jb2gWO&J0F_6gZO6%{_JrVVD~Oj-O(8BI0KLWN|rF~aQSf`
z0E}ac8GEK0p~yf=lPII$h&!%le-5MY^~`-Mel%Os-1|82j;mVm+%)rkrE&Af-aam{
zEv-pXH+B7xq}~>h`sLK5(wu}v)qNvlNoU<fG+<p7CZzLt@uk&0lrXS+B|J=8O5H|6
zYn2<yYN{+9E@uwxha$wHJW#ZthN&T2RiyZ8ic1->nl9PRTutjmH3+9veUF=)l|kF{
zNZO0%q+MMXggw~mItnO_N`uzL=h8Do10HNT&@tX3wl2evt>Dgbmu%4nW1m<{7r7DY
zT(x2I(+8ob_ANJVix%TAd9Mr!I0!LoSw^{cLDS84P^>8?Juz1&6P*zEIH!x}E(O%J
zT_mteLiQ(3&*5MJV7w7R9}JaX5n{&n6zsdUV-gDMUW6mAa<_&9Gr4XCxxzxgD2tKh
zZo{U0*^;b?5PWesPBMb?5kZdKd>p+qVpn%UELg6D77mr2!<gc21xw_|Jh(okT0@aU
zH^s#Lo!oA%$6PV&*t{+|Q^?^$uX3@)_z`X`*$N6a$$eQIea?@_4!9ej1v3GVA_lIs
zf+6G$Y$+k<t?zDw0kMl@OKZ^2w~3{JVE#n2QeiCQSPbR;S|JZ<P?-IE8Af5<2|HN!
zL(~#6+>99V%*<xcfQ|T-ty35kV3LOrEm{(DPU5hKahzy4xsXJJ<C;0ly_UYB?b~r$
z#ZKFOEesUQ1Aeg0wtFXjWjdpNuhus5MD?*?UbQAaE|!ls@({|gV4?U?eqRPH1bJTA
zXi2pupUHMO62cQ%h!$>{+zSM&fJ|mo#9Eon61wY=uuF6&$Qouc22Z}bF-+-R(MHkD
z0XDipA9lAW&w?X5ro*6`P9-Uiygf<;g~{aYdmtl}vx?54a>3cYezM+#JP?HL^LC-&
zR3KX9*BaZevMU}m8sWL=k&%NCPMP#JFms6pcL`HYy9yZ{7O_|KGuXQC@^dk$KW43G
z99+L6XI;+P!{mYpD+Gtftgwl-e#Z{*znLLlDd$yW%swdPECsBr3w$gU^lVESbodkv
z^6}Upx+@dRKHQ2s9PUT4%PJSatDZyL*_0;yVcz_`As}YLw>HP!i{X(J=J8mLz8Hc`
zwD#_c;CFW?{Gpbf1iW{tc*plfUAIi%?1I&eh}BJ<uoCM|J6521r!h^!G7~85a2k~|
zRLm#~e(6Nbg>WixHzh<!MTY0d*u>;|Ig9%P=+xxdN}yog))+D#5iyE3{|m|6V0N`#
zoBtWRta5(}{}{xh+-G_7XIOLhC-6_G!7H`Ib7)^UH{4&SO&4iR<;09VjV9@$xqR`H
ztCc=U-Cx7dZL?I+H3(gqx3nxYGc_n3LQAHKWA=851bm7BuOR^kvuo}AT!md$xdJZa
zK+08Ce5QzVzBIiOX7nHeJxYTH=W)VrR{Z=SyV7)YG{6PpRRMVqaz6fT%^DWFv9&R$
zVy7?q3Hv2V+Kcs7BT34GVG~KH@+6}4YLZe?x;vpO7}C}7PpF0+M0qm}Wf*>L(;*>C
z0cE2Kr4N@%Nf!SSQB<U1T^fhgU0@uAo&*L?W3;R%iEDs2kVMI^5m|T8WKq*_8oC+^
z>RuWOB`5(!6I4HQV<b`HTSU_XG_(w&;z;x)suhVoSCY(@#9B%^+nz@h|1nK5CB+Y!
zOYyTb6iSK$iY`UwbxP{``L~FszoDUJkRnH-Cn@$b7g<P6H&G*!lG!85YV$Nr1;w-P
zbIE9nY#;jN5)l6qaqO2gEJ_PFte#jiUv3+cxHD%YWDkCfa9-3$x&ah_2B9n1AB{f)
zeCHxLB)>-RR?+Y%9&vanc|<o4&YXhvzMMyI0)q(L$%U`r<+8Q?AHfX*&!gQncK0d1
z*YS<EJPD9GMq47x+M+G*WG)<r9^49ibf@5twkvUO!^bRl(ST2-0oTI^2+#GypYZFQ
zE%%W8DHcfC3ha3|{Kf+*?h*JX-3HE<VWsl*-7mD^IKY?{=9lz%p>}p*RIOCMHP@*R
zRQ7YXeX4w<T!qZaZrq*k#xI>ah2?kND0t0Sku!QD|MyltdmuqIPvELdmfbPsEk!A_
zt?5bPlO#%x-GZa@DC4jxUliniJBFO@(3_>rZh|qW9gD7pTH34tUFozL#H%2F$O}bc
zX)~^KEOlm1Iw&G$&Dhp|8)zX)2eu_EgE=`h8bJu{!VW^CbSf^j+mY>DY8!N^5u+$#
zy94`oO!u3KZJJT83tZiL5MsLvy3T~yQqXQ7Hz7@L++c=vo6DtWhCLL?DyC+g2T?@z
zt{D*3JF<)FgEJ&5bJCL~)p<$9_VVZh&d3)FhQX{1%an{Wv|>`3=6@p><MX7L)On%e
zdx?(kU@$aMb}B{g3q*kivD8C#Bl}s2cgWQ)?AaD^x?VaMp$9pnJc<bTyEFv2-^SLf
z+;8GP+!&%246%sSVD2qIT6bh5igb;P3|a6WTQ%X-fH*T1R*<uX7Lt;)>b76XSq<3o
zdRR4XIeuYuXlP_a3p|aCSYud#Z$Vu<9)`$S5c_6fmF?^Jp@DUy4B{2o3NoTFAgf%4
z!z<#H7p6!c`(WYGKq%E5wRZ9W;807=T32dLPC`p8o`;bDd>9cCr0jH@I#j_z;WeZ-
z;5b*4h0HzH7TtG-;dSf?hh5>gTwn9x#8#~e=hs3Ou=d=YP3UorGYtn3;n<Dj7&KS^
z<IBV#M(*Kg2OB(Dm9Xsf^jM=jl#xy;suf%vlUQbH3GW6@ghDV^1F={KA@LsoAwkKV
zf{Z(>KV0#WBc^@WpyQGjn}F^dY`x0;Ptae=^?<qR{wmV(?a<|au`a0U5&slK#Z-+f
z&m~#e#+k7)$UZ<cxmTxQPH0Vc0J&#E&u$!g4vx;P0CwRVA`Q0tuR2qI!I<j)8~h{v
z_}~>;=J?FaZ?3TJf2imzYePr0nhVs2NOx!j{$a<M_>MHPk%~LU#G=&#yZB6<0ki}Z
zwSEBz?IJ!Ps5Q&y!7&|2i>sg0#kxvuJvX`CQ%Bz+4{yC=RPk<ZiQ(MYoy4hl6Sn5K
zqP{^K;0$;oncp_`uAC8Vkd*DpINDQ3A5>NkV1!D4DGmMM#JmES;6Tq56BD~|8PjT^
zOiWaZq=~s+U;R3+d`Yutk=B>BaMcOq-i-|5JDbED!#!yjCTGi?2E0<{!;66M-(_8B
z{W_x)=~87nH;7F0JJS#pYQRoF@f`+KDW(~_aH@<o02z^_0T+z0h-l11N%m)wv;d-p
z+$R!0su~dJKZ50qM?625hGS9%KFye=8%%tb$zWaueEBmrpnEg?BMne67&>@T1C%F`
z$iEeXYh?5go{7<0ktha<PlsP4+<ue>w@@8_3mEb=^h{xpunXs*u{t8#lhpBI8HA20
zq>;NktS|H6+DrF4Njc*`gL<}yn->n~dPU1uxT5o+69$r$(>*Re%cP*YfKz>Jz)T8y
zZVWC>fLN6go{55<)-D#Qok>4Ms<|x<R-tOX7+A4|v4W;Tv6R?_bI({ck>Ol5=`KIl
z{obE2zn4br$Jm8$wW%s4^pCy~jeZ$)`5J!hUJ~zqg+q{m6Iq;5nhp)yj2$g@CqNX`
z>0(*n4M&N?NRHTB?1Sl%sR){v{#_w&xES7G#iCL1i+W;5CB$tRP!N4lTd-Bn9_TfD
zqY`3EF9wV=x_dF;%jHPJB1{37JQR!UBr{z30!60~vhcRw*k~ry^I02r$l^xP*&7uf
zMtAa{hp=>7IoNy&Qclk3K{O?gkyCQmH7WsnZ8!*vs_zH4{%Iwl6KqLMv!F*SYMLiT
zVZZ75L+vN9=;8N>(1$`ClPd7Y(w4VFS8839<%Oak>weZ;FUWdM2Vo03wPfgN?QNw5
zlYWXw{6uyVKgw~|Q{%~nSpS2DLZyBM6s??~4|QJB$G=1reT{~dp_qW9(AmTr$TD>?
z!Lu|K6vv+7a?$jIEGGEhG%PAn&SCY$nry`cr_IxyBz-k?DgG>muBnR&EE*ogBMvVm
zkLc#^^I-8-s|MPwX?V_nU*qH{(aNw`Ss2IUl_AQ9i4e=m?){=!_yUq&YG77P?ec>Q
z;fY8WYX!9P`tmeSN6guU0eCS5UOA)R62Q-QL^lDjOfcqUz7Jl<<xDAbEmE<Rb9@Zx
zTIII$`@~gg1kdT@aOlZb^bN&9nF!ss(v71q`T0Q^v(gQoKWEh&lfa9pdZX=_Z4mhm
zck!Z*;LHw~4{yLL7SuyC`+>>7E9??XR=U>XtQ*B$f$Y92Jho#A1qtsJqMAfnd&pR8
zQEz1aF%P(Uq9@?`RhXO7hZdPS_lbmO-%*|gYI;sY!qi6*iua^JQD*G!!cJkv&V5#e
z{00clafgs?q=l~=POzQ)H=ev2Jpq|5?yCK>f}&)imM9o8L>o{R4wDwxU!ZgsziaWR
zw%?OI#P_F?hS>D~0d$IQ`tez&O@9%T?;$pzzUe=slOUzv)VV>VY@d!{)!9Dc^c5ii
z#HMpdAlzPxRi)!~5#k07+!B6?(MJPUcJfZhlFq9r;;Tfxk&6#<E^1nKa$}y)reeR)
zu4fn-dPoVMWzw!m;LAU7zGR7c&YiErr9~+82jO+rpD4%LA*v~?XNO-S-99}HZlSF&
z0YjESPik+6x}~TicHulUR!3xek~-$1I3<)mgmLGg_TU;62qc-kcqDUJ%fkfmpA!R}
z7o}mGjKptcJkzxjpJmcY7uYk(2Gnhy5_gx`q>x8S`f+y*u8~$EJQJ-<ie?IAl75Qx
z@#Sf-3ia_Iuwox$1x<ybkJyEC&sZOk;anf-E=r&iE$Z8t<)9H0fll;LQUp53Vf5Fa
zonV2^?(WsR8`x#GVKr3vJ>viC((qq8R`&?4fa^$bPpLUpAC|GSVqSVvtQzS?xKlO!
zNjj=<VQU#~vxfV3?dr5uariBr;*x5p8HXF4aZ_Vg*buC{XASf0dSx2@-f4Cb>kA#I
zSlW2q#KxDxA#1#57mK5=tQOTr$`BjWS|VmJ+;-1Puh4i;h<ij<k0kX-4)5MhN-W89
zT07#>7fSjm(v0_|(hMb~d>d%#yV>NVQ_6Rq8vK*oI91SzB=Cte5(sT$1K9sD#(vra
zDYg;2aEZ}|PH`Pk$PXDnQEVf2dAmUuVH?rNq_B-O7(KF$G#w%dBzLdJH3{@XAj32X
z6d}a21Fz~vzPJ}dMhqk!I`QIOi~;`M0R3hR_Z|o}xw8i0&%O9*Hhv1iPp^Yd$K2Pm
zzi(iF-^l*HiT!;u`+Fb$Mc?lN_`v)ck<5}89Q2UJdOnzPIBe2G5^Gz!Bo>F)a|-Ka
z0&jM;{Frr96)Io%L9tlQx)u&T;im%!+ApKP<P~qqT0zA;G-4ma?+3)q60VlM8jxu7
zS|a&1G9+h#SXGx(Hw=cUtCis(OgUcz3%{1Sf=g52B7IY8g`2LWtyYQ=u&nNt5vxQ~
zEwT@<5&>Y+bc=5h!J9)|Q>U<U&=tZIv#@5EU)PkcH>(Ao7dfl(<k@2qK75K`zkq}t
z%&zO_-f`;ez%Hv?4%+<~Oe7ge3p(5Yavcf8{VWX8b};;-I6*@w8atOHvOFOcJ{KA)
zwAv&ZY9v{Wce{JQhmFp}w8y&{l&mrB1Ar6grU@6DB>E&51(fH*2$8H_87jmp6(+r!
zhH?%3+$Jlb5h-YS6AgvZ2o5Dd+I?tv6iOZ<VPqCJ2h*dxbxG_CC9zMz|Dz=SZg)u}
z@F+GT{(=ONnME@(BXkb&l>|Z^vkxd7`Z|fe0f%(ogDe){=1%XRGDs$$M6tjypa%0=
zTQO26n14ajM<q~y#^GyrlT0wr9->xVk~JJwPprw7U_O(^nc~k1=nC@D_#^8DMJPyq
zjTX!GG(0N7%;BZv5#78k!JOdpTn1}bXz|w?`fCYxBW_!-BsrnZi&RT#Z?^RcE>qod
zB-_^CCh;dN2Zxftqq*P%E5>P(1d*9V3`*A3Cy_(Rw)J)reFF{|`83C9D3l+NNK3L$
zb3YA*T1f>I6aVbJG?ebNB<NIp-hGsWky)uu*u47!iNAqX8Sv=qB#cv^M?WO-&&H#e
zZTdG7L}nJ9DBHAP0qLt4%x8U#NVb4y_6mhwzHuupDwPrz?h|&OOt2zW?<2x99qW`L
zWZUYLmVwimu}<kyU^Z2!B*R3qN-17nC1O}@HA=667{YSlMuWdc*C>r#4{u?EclTLr
zG)(A-w|2=_@5$^$Bdl3U(<O2wuO?MXoqwPuVXXWvpyf)@6KFZUHB9@{peR#qzm1(b
z)G)nyWQ2vm_<?&2;oz(8V9U<SmRNcPUq<IS@aj2ycgGN9x{vPxeYWjRNTv<z*|d$F
zF;Bpn4H`>p0VR@l@v~OQ6#p4h`n<j3Pfn*2h$u8^LZ^5c6FyVptDVghnkXez7J&ZU
z#s<_=`U_QI%20;>FjD<D#-Qub-I(gjawUXnl-O^Vhs5W!gr6d;-jxQca6-Ng81nm!
zAv6_A@(R0f9@1=Al2^#~Dube{<;ksk`g|IW$qJLtFf#NC6MU9Q9p3<ad5rTVOWo6x
zF}Sn{Mcos^Gf~72l})B~Pd`qBTd0n21cp4r7(&ycs3UgaJTz8EWP6f2o)UFWix%q2
zL^DDd$wBB8H<I`)lUBYN*wfDj)QzNJ-P3t7xHJJmD-oWFRwflU8Pz>KFAY|qKHdkc
z*veQzQ=#Z1cH!JJ)<<MG*GIaGlG{g%>KaTGk=sYFrdx7bIpBE@jDEA2z~H8JEh`V}
zoP;RC@6qy9Ov8WajM+jf66&7zz67p)f<h+9N_wo)s5~LBUt)eoPI<4m^m&W$E(UDR
z-L-XBJMh}-99()-DXv`yx!`?UHgCyU7eDXPtvL%mZq8X-H(zoIe%^XOf*7~`LWvI}
zA>7Ue&6kT_E_`Kq9$#6WV{r5M>IwYy4mQR<9c+$&UAyi8AaMKoaH#cT@LuDpS8ELK
z+h-v*aO)JuOrR!k!{4?Z$c>JUZpf|I!K*_+_vrr*<o0dOZO(1M|2uGC!yU>kPt4Uw
za3{58d+!wr>r=5^r@b4?Vd$RN{r>=~V@~!wq0rLf3%^J6=#EtLNO651fH^(QOc_}v
zws#bS`(HwZ>1#ZHgtYd)FfHfc_8!#+!Sy}IeM+>0%mG-2#aj5ltPs3<KD0ydlKXU)
z=&7Uc+W<p8pxYvSn8*}?S&CaDWxx3PG`b$H*}n&helJsW<24(*@VP-Rz{-{er_&(=
zD9W0RT^a^m#F~wZUy3#Rei%JkvuQd+J)Vrj$U3&;aw)@Sfeh1nya*vy^7KeI@<lzK
zjE+e<bmF_$<2?Z=@_M{);wM&*$3C6(dOUsu#EGlN<M4W}$GceI&CCc@cn4t@wMLIw
z(&1RKoy18L+ljByiuiMonYz!>XT4#u-A}?nnC3kSUawYc$4lkfND&C@H4=<!_KTjJ
z?@~iIZxKZYS-vnAx>84lVwD}g24aCg9qY93AvlLdm#SP?#n9E6LsoWdq+wB23mlei
z5!4W!0c)~Vc3es0O!4Q61pdfUj*gWbFQws8JmT<D@`!HUR@sp-!L-Vb6^TW=e10Wl
zOk-7Gw~@rF@=jIE(Nh)J6qlk&-u<;C`ea?CfTCvulWPd>qoGha#}u@@i-tng5O637
z(vH^<{1FKwv$%V!AxM-&a_!FNN&LzDct;+^X2e%W5Sdvt6Ei~B)#S|e4@mS4I7FK#
zC3Br%m`+~iS~xcyGuIKaZJFz5U{TMQx$a-wC37v%>Xf<WKeykm@hEIbek2?#@HY%I
z*KD+OnQMbCk-vC4P}|vGRA+{ox&AQFvP|>@T8?k#dN>V=GFW;(cIuG1eruGyg%jPF
zvc}Xkk1+CRD9bJ4{U)bi_IeA6eG=xvFW2tkh^MIA52Z859g_R3cVY(n%2a|8$(bvl
zQ#?6?&r~JsY_4J%?2m&UZf67P$(bcFp-LLso<|d5e+&nG;}n|+GLMb0O%gTB6T~8M
zO8hOtYbp(1Vc6$SfH5v(3{8iUs=+Rtmo)p8un)3-j6u;t;=#$yWWON|$7Ck^UPgwV
z$;M}y)bdY(FL!aiWXWW|F9w$up~x8`JQG#yF_Zm?G`NN8_(@>M#~4FsS`>A}E}VzP
z>WFMlQpZywll`qUjFVdVb;dJYEAd$-t$YmF^A$FrZY&Km+5Z%SOG{E{CBie&%FLPU
z*-Lb8kXVKK_$gq;e==5(?bRD)unXs&u|6Whxjxcel&Cmb)Qb!d6H#&Wa=L|ul}z?W
zVf1w<k(s-}V{;n*OXqAOc1m{6_F76h72jMGmU(f%x@b)_YE$gaIhIh?lGw>PQorN~
zCvO6rydyr(JX87Y9Q7k+g_4u0GJ7u5;~x0!6-UpHBQg;ZM2;Px=n3M~iRpz3Cr!C%
z$iaOSgL>s2=36PwQLwvmEq#u8S1Juxa?GED*>)v#h7A(=@|1u7|5iA{1>V~antEz6
zfyfMtSq}N(p&m809f*?deOVf{3m3M}gU;Q;bk2BT!!BG+=~YVc`%$xBXaGf7*s#kE
zgDzrW!xbXM!uDAhJzCgkIz+lxhA(8A^thyZ10cgR-5Vjql5YRA8~GyLD?>_>4xRY!
z>E1s86nVP$J@|>Gd)cRxo$h^MxjNxA@Z>-b8Q+W3-M|up>^bSXCmd>+2U76Oig3l4
z=j12r%`t1dR;yUsE%sLSoQ1EXKUAqq#h=To2RATF8OsF^h9G1pL3t7d5dIv*s_qwb
z!(f;Z{$w}^Q_d$~P17>MFPZ@&1&vxFRvw$cM4z4_Rewp0PJJyzIrS|pe+ykxWRU-w
zHP<u9XLgjfAX7W3zCdX$g{eJ@#QEdwWIfIC)H4pr*U0=k4TZ{%2`DC+zIjV2)g&{0
z0*c1PKAcpXec_wC!+X^Ej}afwBw?%ZzEnw44kqh)?iGMkT=B?ShZM}dkR;fI*(PtD
zxPpd4CHhm)avcqYN~;Pe+Pf|KnHiQ7Pkf71<p2#W!`m%568ciDs$8rNBgw$5OkOjR
ztjm1aR8kl{h=@Ey6G>G}G@vUi2qt#rH8d2;t_Ub5cIC}96iPh=l<aopJtS;Z$S2UQ
zypu^tU-?P9l7iWvCJ83(ihyEbSH46;q3lWuTArk#7${B}yYdVTErVU*r1oT2vekFZ
zUPi9ulz9FFxB%8en{i$1yOz-KC~L*xrQ{Laysf?~!CgLaWfW(`E&zzO2^}jbB7EB_
zDLw|CV8%*{=Yf!@I{=jJiY%r$s}t_+^%PIQp8rpTn+p0v*HaA3n*b)As?%U^jE~<1
zFa>vgR>FG#%sWI&=uT28RnxE5QalNCTp@Y_9mltpVs9E0Ww`z(?9?G#|6K^(MFj~g
z@{N%^dFUT^wnfYKEF0N9Rrf8pkAX$+6+yrgXu<d|!ufDlH*3UY!bZKN1l1T#3=CVi
zQ-+2jI*@~wWzX{9BJWDk8g~E`6jEb595;;o^Jn6;n=%-`&YK)gR)#>$gtAX#Z*IXI
zN6xC%eYj&)+!GFC`y~g6tjQfeBjvZQzX`s3wPpkE2e&7DXrYiWK5fBE?(Js9=OF=h
zS%P%i3?M;M>*i~(-ObqBZCv@<Rqb^Km}}|$N`r(`aIs4jzN2Wt%xi!mu->Gv43t9h
z!mB1LXtcLFxE{_KQKZjfBT8(ynQ9h_!kOEkQ+#h2vnlH34&MW_GR3;k*@wq}(}X2A
zIlp;?MJvs{X%r|-4*Wf+#N7s3q*P&G7p_H}Js_;fQj!CxcrI5w?X^CouGVvgEN9YW
zT@nK>+Gw}7B({OjVuwK7qiJLy3pyTVjL{1^@R=H*r8%u682<q{{a(&#t;}e#zADKH
z5q})1(|?a)pl=l8Mn;r&AY_x$j;m%ytdfeE8ZM&#>uIPDC({pr`Cnzsr@5s}ChWrH
zMT=OOOeo@{$@J2Z5z0GISZ?kaB+Z+JDx9QA)7}7POj3O5-_medPM2RW2J6!WpJkdZ
zPXjN1#s<{upW^dEEZgzNk-{uKO~-&%05M$<vdMJWmPK}Qa%AfqQFcxm%EGDe3@~;z
z^z7!HC{qEua9J2n1r%S>R5&H7`>sgCIH}#+8P9a>#%Gzd`$xc@OWA<Bb{kgr?Tx`T
z(rJWeqSKkH`^sss3ia{Fz={dR3YrQ<AF&JPp0Pe6!?`}vU6djSbd4sM1)>oXMG&~}
z?p6h$RQLTIjD7=JnVGBZds`a*%W%P4gjOWnQxj}cFR8(uhX?~ekRS{h@Fr=?-Ga?>
z=v2d|uLt``DxN6r+Rs5gk21BScVufw0;H+$(kim)pG$*pIEVfjko^opmR40|4q+Eg
zA3Y%y_Xb(=hyfI34q=z~8FUeI2$eC#9QqlI9?c<|4pB%XV|TIu=eQIS{R@y`T1XTj
z#A-Eu-Hm)vNF-yak`A5t?uA4Pmdiq-h4_gT60uJwyO3ySJ|9NJdMG8jG+il?0<`C1
zqVtr&!g3;ZMFHHwFv7!mi6ue{l#bGHXwM=uaXtu5bv*QGZ&)U@HXH<1Z}*(p5YX1@
zgq|bogkso>t;I~${#+K~k-GNhV(3aava!_sg{-+=`;+>;ATOoSrLt$&arj#HjJ!R<
zreRT;GY(6)3u<t{fHm2&i_<jD6n~lt{OOuqyo-iM@rc7q$s@XXTXr#NmYultVmw8s
znwX9W#t7xM1mkL8cE$wbqtKN~FlwewrWjYpQj7}zZAnJ3VBip(E!5%E$TEJ4R8Dn1
zDp|(e{Yc0sBSMaEn(^^8ILdv)m$6fFnlX4>$u7xLK_esKRhEz;3%3st@ajF|2&ukP
z;;Qn43f;IZIG!|%H8@3*>&GauPD!qxq|yPA<oY3WqE=u1{P*r5)aoCwKJ=a{f1H`e
zc?Y4-J_Mws`sSXlyH#42Fj%C?Q+jBFp<XKkEr<=I`)9FfFC$C{9S_?FduDciD0syQ
z?7Wt#=4b&5ZWhCiCM4Ql?q+?sut^|FGKwq6Q(Qr`Y}Cm;qJD7}VO}VNSsIAEc6_FK
z?KESeV8rf4D1c3D#7yzRJu$$v6h#ai;hDq>dpx6_PlKC`c<f$`9NEt}LKCB?B<sVu
zNOL}ld2r4rspKgU$bMBC-bpia2jiL^$i`=xG;=A+VVaGYNi*LT15683XePol(ag+|
z?1$4}73$<N<i-0KFK99pon(DD|BQ8#@to@<-6smfant%;gY{ZO;?TY87K)0Z*}LZ>
z>0g9CGZ@bPW*Q+#xAf~mHGV_G**{Ok6J-T?4hrZeOgCxYt4pL>IFMahMN8hC6}n*|
zCq_7v&OmH`&9EgqC|5r0W_>t?^sG><JLAfa4B#j;iS_wMgFa#=p+=^dNh|c3M3W+d
z+H$8TgTcoosJ#s&X&Tgy5Mq(V^C7ZG6C$G8a_=q)(K~gIYF`65^5e1B;wKi>W}i-W
zRC`yRofw^w(DshBp>3?ej3D<M-~Lb=Ej&y-W+mQ(nzKB6s$PK--+bL}*i+t^b)Bfd
z-G`WszEu2tJ^S?loU`3it2#=o+=6=|@a^LJ1mR7;T=7PWwr}V8N35mYB`b_?a6Bl?
zy`}7zcHtW{_#KV0!?;W1B*@KN`wx*Rb4XKW7<0G7LD<X~Zi~6U{P;3cgblRNy$JJ<
zVI{WpO3rHdC^@G^tGgcLy*(r_buj-hbcLv$$;q^1G!*J&nt)<*GVOjE3U$CpK+$-&
zB$2-fcgDu<Lw%S;Lk%XRI;7UqeW>r{vNGY~pOfg5N6rKkT_cij2>xpt3Z)S#X!#Zm
zh0+KPB|+Np8-kxEVPqC}Z#M)dN@AZho3r;RF{b}R;!oax+mT1H88Nuh)T=z<GeXzZ
z<Qsz5kmwt5NcW4#v!UD&=^a!ZilqTP5FXfIZGG6<wty$B+puupvMp9`vCPEGpY8`p
z<yjINUQwmXXhu<~y^Eo%;bO9jhC-PR0mZ}&-a<p6OoxD?ar3OsGT_Wlur$jP`Y7Bk
zlNhQ>lWJBaLd|;Tg2TmX!tgsta>#W@K+)%N^15>`4TVyk6tp}*L!p$1LrKtvKAf+L
z_}@>$Fcg0<w>Iv!n11+>lXLu5z~S^JaNT~91~0{w{2WO!sUZT2$zk}%X(&`ZL<(BI
zOGBYFL_pCF!*ijDXir>Io~2=BI10~E(50+uoWUxUtyrc1Ig}^iQo@-HU15dMta8_4
zmE|-%YDMGlQu2sy-d3!Vv^~|{bu!lLMm)(7;h2gc+CuV3Rf|;XP;U<KMlNlANl12p
zuO#s&-8&8?fk)f2#6A?BzLdl%Gf%Ujm31vhltHqKJxS7QAOm{hU}eMDJTM)r+@vXU
z;_AE3X{(PyA5V$)bu99V@NO&e+7Fw6j745|0pF=2ul4|&tn^yjg_8DiuVD<8zg+A#
z!4lJ#O626F671xCSYu3IERn8L6f1lysebBoQ|l})K!WZQU4fwETlV!x8XRSQ?fuxP
zL)q7-qVgxWtROKM%r1wxp4}}m>}`P1Fm?9>ly$|nhU{*IFnSyd9LH{Uz<FTlO)T(D
zor0gUlnWft35NR{hS;X5`^5>@7|>`+pV%amC0tqF^znsMf)%w~pNCHIH+|r<OjQA!
zQIS8xM%1eUh*zaS-BX8OqiOZs7(#S%P*e*b9wyZS>kZ{5;}*v4Bm94s2LCV{x&^uU
zEaN6kuu@mR`fv$!UdeD3Q?j8<YM)|o^f!dSWdIo`VwOo5JLI)I0l1hs+Jyj#?cVaS
zU>5sN>|}^o)3P&kWv5v!X3Zk#6rVNtEYqyH5am3`M$~7`y6z@VDy2!&NFRq|sM0ix
z`Ggpn%%@x$;Y<9}b&4ptGz}%;%-V(=zld?1CS93XtPhux@yue9<ui-!qtqF&+4!dh
zvtHC0;L6-B&8}4Jj3Vjpfj%=>u~SMT1nE&YLY)#ScJ_V&?yF)OPBO#GoQfM9zuBlN
zS4dG^a@JwE^vSUtyHH}i;j-r`r|MgUTGhA9RgYZ)#WJz-CM>&JWDg6q23!nPuT_ie
z(yFEpIcD%5u*@Cq0;pQI#w_J%(UAqh5F5X-(R6ObJFpy!AAq$EM;BpL*t^Z(eG1Ge
z@^Wm@5Tp3G@U}CIr#oJFNtccqzro8KHFm2i;MRyUQleF^`W{1RSiH-l=;f?x4X(l}
z)EG1PJz3FJVH?n|lK-&@p5R|;E6v_NG81@qVfQTP85eTz0yaR&Ot{$v8(Xv2b)U4-
z;F~m!%>1iTSs11MZ4Axb9n8`v)V~dvp1<K$tvg!7rT$?0q$!tV$w|Z@;`E!-a5`L2
zFGDuJfw5U(>6;8RG@WJab<!|a55tzDpP{A%%HA2HyEGUq@ymY2$u}Hvhhgi!R;VrK
z8rXX&*R^pI-nKL-csvrrgZ588fUQ@#h&($%QZ`;=dgf%LGI7?Cu1=^jI}%mKGf<hC
zHfo)f;&U-!xDUgFPzTPVb%3KOm0%sI1gx#11z(H-HqrvbmTAF@yRGKt24BG`&I~g#
zUlEC!?4XwPGHb$(Hr+?r#4<zw3dRANS&FO8`W!OoBV28CSyQ;$mmukpt4)(4N_u4!
zP*%bomy+IrGtEnSBZOGt;&rAay)r&43DK$VUedb;aO5SuXX7VU(#t-b?2_J_>J0}9
zS{;bB!xq5H!&SLa*u96FD_)$gyqChF=ThHaQlk%J1+M4#@KWZfiAwD-tH_1#yfha+
zjPB$Oa+Y1MSEloYlI`bx2R1FX?}RVU!G%9(IaPZc*4QF!fg0uUrtcIZW`<3fbu@VY
z4i*ga@H6v6VaNku3GHcyVT9Gra>3sRKf+pBYZOYOH#$|X);J*RUU$t=Z2W496VkTR
zjiQC}V$ftYLZlli!%E@{!$H`DyrHd<_-&`EeE2%6*O(bAljy8v-RB8Pr5wWV5zBXn
z+)7=RybHP-R+8<bp-{y@0*XmtQ<;WBm3jy$8u!lWGJQE$+uzaCwO_>j+ewVobXcm4
zepd{#CYld3k?YfvEIQ(R)z>zatXTcZm27-{26Q!&Nt(Yr!LrI(koa39lZQx|kkr1N
zqp#OOB)^*KQ5p(0BLoza8Sxkmg_;opiZ&xsy_)LFBy3d)Qt5VYucrDdCZQAxoPya;
zlLV7ha{`J<)!Z*>C{$rq3R>o#MXkgpwCGiH=*-CZ*T=p@6K)v|D?`;BC$uN4Gh*&|
zCE1_VRCmb;x>3ZS5t>0%ZQOIAtKk}b2@Qp^NCJw9MY@WHLRlmMCA&r1OTs=$7U@PN
zp%fM=1+yC@!Q?C!P)sb+Q5p(mky6m|dK!v>;%FAB%ggc}q@iW7Nu1Q4Y|>1>Ebrqq
zk(6Ee2y`{HE1##KP<BN?F|jLOp`lRfA)sWpE8iz!pCr5TJtm<Pb|nR~|AQo$v?~IN
ziCtN+iu6G!yOM&IGifLWij&5!tf!%6uq&L@p6p7t69O-wNuk7ZIdlc<q0P9iCj|D;
z@F;7=;icpe-MsCDKvLLAJ0Wmc&ylx;^7r@|*$T;|?4}6gg{`bngAy0FzAa8ZB5;(%
zpDa)3P!f0)J0tsg5=3Se^>k`HkweJ^{O=~wH{g(KvYdXHhC*4zL|T&12!5J|Lb(P4
ziU}=WqM>xBB|)c_#-rv+9`H#LUS<Vr2ifC%G~wM(Nm|I2K7}6qiiR>ndN8z_^d~au
zK_BKy6xp9a!pJOp4yH$Wo7mqRiNAq$Nti1!`#VO0$jqXdx4$_Vr<t8oB)o@2-+)6#
z_IE!GC6lf$RI_;sMJ@>|vp{rkGpf7EE8~A7=`fHC?eQr`+md{}6L+-jg4lsDG6UQ3
zm|H|c+cCFM2r*|o=Jxv_1L~L?oe7pl-ZrM7u>JVk4OmvPS%k@9n?WCcTXQ`xSCH@Q
z#W(wghv32&DR*fV#cCfVl~A2b>H)W#5amZiS3vpr9&r1MG&ss~^iN}_4hP)+aAZVe
zn%P0Lc-~Z$JB+d-X}nolH0G~W))_u@s@z%rTZS}sqeo;1w8Lx*l(EI{+L4{M-=QEg
zA$R}XRB{w2bDo4w@sm0DOs%`<YzWZ^D%Ha;L%sb58&N-*b4Di}kZWq^2GLylWehDn
zm=&8#t~iB**qEHcIj4)@B%)p0MfeV^(b-7i8`gR5LmtkBuH9~oSI*k7KAi8gtds&k
zCabS;S)D?ch!3Y>n!Kv{JjO2ls%CtaNr?}j04z45uEd5f9ejQaFe!GarU=hOO?$jV
z{Q5Mwg(~@S<j8Ku5t<l9C0QTNMPrp@oKI58Q{ocwMjGBp&8#x6>6(epGHK?mC<m8~
zm`O8_#Q@X76b2OGnP_I_OT>RK4OXE}+Q^Ih882uu6rE&!IRA`wlJT7DB;7};lV+y%
zZ<%SOaT9gY=u4%jlirJ@KZfht%)JxyuhIxX`hb5aR3qU?VsNQ^aix5*jo6kHlO1b$
z{0;tPD{Z$EaKZEB__%JT8#{IgYn~WMuJGaaXeECt6?YVeXc9&BU8bc8$ElQ~KkK@a
zA@MV58lnD+G^mHOtAu#}6T_R<U}bi(KAcv1rYKGb<IlGZ;3%_;^?AaekC<Jkr732Y
zqt7my6mbecZq;P*{&6{laQ@lmrw}59SkZIS^b~^JCQCwe>bsvpxEyfgrx3Q|Cw2;f
zeLC5v5ccQu32_nq4xk<mBfLcHZ>DNR2kzW1P2|~WU#F30Pex+_3<G3(K9lel20)Cf
z$!GRx!Um&*2Z`qk&NZM)8cL(Kk)^yDG*)%Xbi-$O+Tez85H|6Sw4F9MB!V3$_%sOP
z$*^8YZ7hA!S<BCd@dt5ZNccTsXeH!D>azqT=xSI`^a>gZnJnT^OcGyr(@@9&Jcpuj
zuWLm<<7!W534V{nSdEsXs<-ZndDh7OTN#`5Nk>-Ka~;uFD3z?(S%M-LgLY%{i9Sp4
zX;LO6wSUIZ*OPY1X9>PUL!oAbfMPNuo}{5rGeSVoW<;v91kaMNRV7HJ+r6D7c!o(R
zg#xEw_P{yR+C=8v1r(DCyVGeXR0&rKTGr7}3=~JJu;a%E<PE&Bvjm%ISQ+Z;IH5gR
zom1&7!7iFXR2AKo(A98_zJ-QDStJ3)#3DH~6iTrKl<XGiFbVr4StO51D1}8z!R$AY
z1d|p?KryjM57AI4i<E+vM`<VqilbSiE@ufIqoHN6Nu1P_Hi>TDR?nKWvC-;T*LAID
z6&t#QlAw6m>XRgkvd^ZT%2vMtNVVC!<g(SDlK7LQCmc!wk78w0zal|oW>HUNQ;8f(
zF3wqMkq!;DA_yoZE5g|{6l!88(vn=Znxmo25G`Xgl<u@7=u~?RX|#L4nuM2G!P-G)
z-7s<q(Pm63(n>O5C$30qYlaA5$6Bq3l(t%}-LMJCSgTb9fm5|wM&UGBvb8Nez3r7-
zM=%tBo!IMv>20svDxHIu(X9m=b1wYveE8pb_}_*DEfl<XS*9EyYfSr2<jcKq$%%Ux
zequ3g_UUBDv~Nj_g@(myJ%qEboq=$6h)2(n>>Ik!X&zA>898_u-nQdy5E0Es)T?Wr
zx9{?Eedh?=)3e?lhwE<ag1>%8&bqwlc!j)Ip3Hl7y8!p`gsrUgJ9dD!%{u@pxH6b8
zJ1M2K?~K_{qgvk&#5%?NcgoNsJSx>CLt+mD-7vKt3?Z>#_3ap})^77To8~;D9Z2ui
zELv1P6+%o32^+@PlhA%zmE_j+8X*)yiw+*|`1n^zW{}aqzhQzldIjj;(@-eyL_jg|
zw0=rMp_EoY(e~b5qs+a>$7ijjJWzElNfkrvig{Ld=T}VRF=If*XE<NG`$x}mWAH_;
zQrZm2b`tbFlC$KyqAchNMo80zt|vja((ouAad;_tL^p5SIVYK`6X$rwYv;RI^o~Af
zL`9p=c|T}QMxS#ph>r3(yR9tJ-P~m2ZiXI7yQg_OdYTUi8<yGAlmle-H0#I>=4sxB
zpO~k~KAmh&^O`9K-cgvZ)EajFuv4Ba`Cbp+<qOhzm+i27_9#zp9}{$fd2+e~bXt2u
z>GUN16gfKgkT%KC(YZ4mg!yMj+lmS12&_A9gQUavkKD#54S3>AQT)wS3B2uSej?1<
z%+ITUOBv12hj5ztz#DgP_N-aKV%>_gf+QJBlXI@#c(k3Gb~AGn&CIKXR5P0yIY3r3
z^Os09Gc#YnPt44)Pbb^VTnFz#&DWg?KYzIFm!K>&e;b;d9!$;7bfzX5e$VFSjO1~X
zP0q^X#?tI``Z+Q^e=qb)s)z0}8k(N(go7~C^Xi^WPtt%V&h(rgo2(sp+|m3**teOV
zCxBBK%}?Jt+Wd5qwKPHNji_ojL$5|N^erLb%w|XqAoWali@Vo~u14ya8Crv%m>FW9
zPPQ4^?%9W&9*oY`bVdiE^=x9cFr15JxTG0B<tLhz3a>rgE^dP!!N6C}iig0ocuyb4
z1tMAy)!|<vyK)gosptC0t>GZduH4yXS1w3FLsHOYF7#*5^-W~>7zksZBf72)QA&zQ
zkztstp(}MvPr6@Xi@lRI*NZS$B#KK=rtO2_oot$5<%-buJR+@_owQpyt|m$P1`UNu
z(hDeQ(&tBMXc^My90}cgsqr5JDfYA3C8VbFY(&aw_J|T0et@Qe;@MlcWHjCA!^tfH
z@h_1Uyq|_e#YZ@-o>-GDhV&U4XNo_MCGe+f4C%`>Jc>sgUP>O(&D&x~NoMoJc~8q@
zeg<15I{Hcxu5G^3M&NQrU+EXnmD;ND)k=CsD-tkn_k-?4KWLMXe4-zuzVK>Tz8E=r
za~u6qWb+QuT%B_+waBVgNxd)P7!va9h>+uZU&P8ZILfnv%du03`yxI7_aBrSBO}p;
z5akBEf4J6|hBr<yf5u|RXe{<}!P-HmT7HGoz^fr*OQQDPV}8E`ydTMjcNm5Pp<~2H
zgSYNUXDFfHHO}<mTBB%zVQu0o4?TQIAG5FtM-zv&xAO&CB|Yp;0V7(((xF`oF;5J`
zp{h4RxLc5$+)aylGZ&^3j<^J36Lg~dnAkNBuSa#efc2qIHEu9wfLyy$^oS<St~4}<
zXB^*vR9t01xpKyl_2HCvw&2Ju<%}ca)-dOmcG1G+-Sn5z-6Vx!ulCh^#1Za&#NLkG
zu^l+agg1FG4V&cw_(?{%egGbysXGapmCAA2H=)Q1oRv5CE3cg~yqe*6V}SM0NX)v*
zs~HfU$pQFPgt(-7lWih}zAp`)p+dhIx${QG9hxvjp;;f!QCbR$LNh5ODfB6EO~c32
z@J?#!M;O<1EyZV<w6ukCc$AHp>6(Tw#Q@X76xTE$JQFR=d`-jOrNJuH$+sXco@BhB
z$xw8X_2K+8)=9>5u9I{h<sdmT%;y>4CJvIL3qw`cpxysQ9~VRE?A-y9zEp4xyvrxx
zC;2YK(g063eMfl~%;RrV4fk>3_sB3TJWtn2ny24JWZ+fl*r|i-wM|~5V8MU5S_|ja
zDO#R$Thpl)<Qte%-_WpG4xMV)^p`W7pNc2SYVvjz&|2sX!Z|}(O}KEB%Nf>?6iD0N
zrBx)vi_#z*&ZGws+Y1@CwA7WE#QJaw=~<yzcgB^o4d5s<iS;?lppTeIsF5jV(gXTT
zqDc`~GRWPKj7}eyD;Y{ahd16?93pa5Is&4L5Mq(fhavJw6C%o2<^EI>qIc>ZX8#?)
zk%!rj;wKhnXP;&w%&sF7cc26X@1CO@ExnVVpubSDJ+I}cW4QM~Pxo~p0k01UV4t!Q
zumJH{6Aw-VOJI-LtT-F-O~1I|Yq3ZL;v;`zue}#PMRmrdq0*|-LhdP`Xs;PnB|BWY
zJUE*t1U9lwthcc=u~FO(!Jh!pqIy}j9OW(B?vVhLVsPgwaOY54xM0AmH5&!UG~|NQ
zp}4i`!P_OvwJKbTQJZWyz<Q{p4*LC?Z&zB^%nF8N8>a|gn;31F9V{+18x3GUUW~OT
zSSX)V%2lVeCzucKv-HbV2u23WinT)1FT-^h6Ai}FVk;Q9*{(DlFMKm}Fu)gy)}GRm
zJ;9(@d}<9CX%LDy9nZI?>hNVjqgI2|FuYp2e54i3-UN5ga}@Fb_aHPF5?`B*3Vd7S
z*9!S@yKoSHh4u8w;Is)Cpjz`C<?-FYA`ik0)#g<G2;4e3)w(-ajIY(K=L@wdw&#bT
zR`DlJ#bHlaW;)Mw703Wcq5~=~d$16$53M!)e8oBB03+uFLnXTse(evIIY-LCKi~G7
zo@~?~^cS7+<|K@_80wR&z&~iwaO^2~GKiJP4iba-v6rakVOg#fp;apwZ2A)$FWCs?
z$*o|u{$t*0G~ivUB?y{SoCfe7Q7yo={YyB{l~MZHD6L>k?3dUO?q9<M3>NcoBOGM!
z)%p*y&cRZo0kqC50gEa?Q!7|hv8$6!8%m3S9{4`TskVaU5c`413*4YmcdA7IQ)Y}E
zfO$CqWCBOZMX3s<fnc%S^lSO(Q81?ringfOoT}$TWdz2u2W6PrjjCNi(Jf(5n3|*X
zqepemX%=hwBU7kKOPwS2nul+%je0UIDfszAWv>iTDjW;n$jaW#8e%z5P7;`K3xOC9
zrUjdE(0mcli^Sw7$`v3EEX+#$1VB8e!Dc0YUsw@DzMx?r22tVQFcW77%W$TF=2R#B
z5@`Qa*>CM$x;|K3ck0or*O8co_+%X3Je$W!5u8!-{kk`{X%js5M(fj1U{M{dH6~#$
zt-zKH0caV%dA1eIhvys5-@Fa@xwtV=hz*ZVqT!z_hsWA$RM~@0W3=q}6AT8Jf=wlV
zssg|UeR~{cQPD58u9+XK0iB$JcLIya8NNH77u&av1xvl@s&603m!Rsbf|X~jps!Lx
z)nDjM*)Y*=Yr?GdfKkiYcm5Sh)e5#??HYKWE*A5{Wjt(B0+$&Jq4&XN9Y8Dw?D(te
zxfwVH+EXh)wIJw#unVW5q41jHv1a8d)ZI$W=996H86!}csd0EOujfw#1|`Sw0mg~~
zv^tn~rtEShZx@SbaT(UjaAdT@o>Q6N*6dg?*Qp+A?P-n&^SL#F=?j`zu0oU)TIYQT
zO2YetQE!$3QlMEr7+M&k=uDs;4^GpbR_!SO1&?90iSWw+8q*?sJ`W%A&%^FRxX3<;
ze{P%&e_n!r=HoitkALpRA6oFI84O}Cf4a`b=KYSbKF3)5W3275W@#-fENnnnBrh)w
z1_K=Za1AK+3x_*A%X4|S^2+5d=r4P)CbsUzJ=U`G!~IO3TzbKUM@&oia5Qg0tPxEY
z7khUD{@IFu-id$Si+_#|z@OLRpJjvaXC?l5W+D9f3I6%m68Q5;{IhyF{8@{C_OF0H
zHvYjIv0c0|+r{%%E}qA7u?m)*sBp0$&BcN>7i&XYtQ~PN^WkD<q!}zywVm;QQK1ZT
z1_cN*;ZM-0$#z`O>J#1$#1FG$YP?nnClG6TdhB7?dQcm2P(o#$PSeJR^{L^bwOJ7#
zOZ);30+`%Ft%+t4R^vjgQsIgWEC0L~iRbr$*1fiM1DFqZ12`BW&_fSq3<nrmUhVOM
g@G(>R1?6g?(k$X6n?!vdLeHro>?ry^&GFIy3$f%OhX4Qo

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/hf_encoder_decoder_model/index.doctree b/.doctrees/autoapi/lmflow/models/hf_encoder_decoder_model/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..eb843889fe3756bebc293a720a860a1d846b28f1
GIT binary patch
literal 63023
zcmeHw3zQsJd8YN48O`V|S;Da_(|&3+o*5a(F}5NpI2Ojnmax6THnv4m(_J&wrGB}p
zTGD{MIIs}Q6a;W*6BclQ1ww!m0$JXK!(p?W95yV5z`}tI!4O`74QJ!<2%MPx{`;tV
zZ*}$Usvf;IXMJE)U3KsM-~avZ|KD5n(8%v!`qOjRfARXT?bVvcXPs)b*{Zr>H(ph3
zRXPo~8Fe4+?);tZgWbt^JaCSMtxiyJyYV?_QSoZ^YT!1zhr97QY97_<;jHLC7P^(F
z)@ml*y%BHJ8#~;c^v2_ns20^-?NNMry)j>J9h+?cItKCP%WkuR>!4h9`M>N*mqx$<
zD-XHH3A5b&UeV<=jJ6$M9Cf{o@pw5Q(OVIZ3d#<~E9<qU+dbHsbK(gnih|l)Cjv)^
z11oCPkf<H2)SWQIO<a#Ood&;JRqeFZk2$AuByi`cO(!0UoCSK+_11gayfxkh-U@Gf
zT;1ih8t$&3(`>rIu4Ap>$gW!t+z2+ve}G`PizBwn=|nB3UE9SOwTm-qS3gGWsx_<b
z@!22*sTba|yff#W6OXWa8rQi%yr+XS?;`N^68!%%{C@}jzmv!Wr-B+IZZclkc4|R4
zCJA-w?r)ZI3M`)C9cFU2_eKv!U8mz+;ms&c9rP}QJRtXRh|gIFoVKTtara}(&N=6h
zR|`w{pHpJ8DOH?$-K~~xx^X|(p97>=rDI;L;_+t3FiP7EN>0$~G^?d3aGK$KD`>bu
zSYpgAInC-s)H>oeYcEDSA@bSM%~8p7!cyoSbpxkfYPgZts^WJCBjbL#Tx-^%a(TKW
z`%cqH1fbR|MV=D@q&v|GfgRxS`?cl*Ycvnln~6G2`W<;LM`X5i2$ynP0k9M~M_f?c
zOb~H`1!l0}ghYjrQO5lc$aI=+IRp}syRgVHs)prCt2tj=m==5iUL}TC)GF0m4sZtC
zcHlD9-ReZbmV~r`k3<yLoI3c#y0G>{Lg+>j{taij-ERH=l42@zR!d5FN*Ld>xSQg}
z=k#?GWUKqU?cOvT%_?cLu-4z=(}p(^zmfD;T7%SI*s9dE<Pqsxt=aBGr3n6=Sp>AN
z6Oi4=CpYhhk;9(gn&F->!%Q@s@JNUju&j#HEX}#4dDs?om^yKtH4F_#S8L9@m<KAZ
z)XcLRM<tVk#bvTCQzqYI4wL)Ju+wh0f+#FiYx9gR;J@xJO070ATKTvgO*l%YzGJ$C
ztENl;)?0ytf5~U=aH>u_as&M2{uZvdb&NSX1wi+7(oK|F^Gas3rJGvE$j8EWL5>{2
z1W7&_za;z_rYQ2MrDiKCExJ*u)@avVie21lBHv#vh2C#H^oN|{Z$0#e=@R?ZbmcWs
z&aR1eJQLKjTPKfx3@iXK9vE<P@j>a(eZ_TXW2O#0{ya7X)=5GzXRb``pmnr{5Rzu%
zR`V!kPY41A2Zfk-DzHtMA5AWqQ=XcH`DDXQU08K8Wgc~cTB{S54&Hhjra>}(Ipfux
z=0e9=5M~b$K#e>JL#;Mnt2m^{(~O1pcifPEL1)8bZh*##vi#d4OiAA_F8wW<^v4nW
zwi?F503LqZMHeAZXjNQ9#nsZ>VyW3_v=<S+mM*$T%>DK@96!K$aX8n1s)@&sH|o2?
zHuMPV4@4%R8yt1JGy@%%IEtC*e9S<r<qRZO9vkG1c;b2$*|RmpR^ygp*Q@ihi$eQH
zcyl%xUdW>E4Usd~sXGB;t$I0ZyA>^R9Tb4CWbNY(_uc)HXFT|ls~(j9GK$}*z{Q=@
zv69&R>(}hwL(7n`(`Y!s;yx@b5SGVV#o`f$Hmg1E#v8R?Ww60q@~xS`fA!YFf*VXs
z$c5cVx|v|ncsv>{AX0DlaF5oLP|12h)~-UAW3_7Jb#Iv<{rMQU`C$m`<5EQi$wnLC
z9(@2>-OjbDi#><}&lN%ISkHa7Af}a8Mi&)BUBA1%xVtsy?|u<yS!qwc1=TEXzVAi%
zF&0l^9k{0LHp%^#k7IqRp%XuAsW7Az$Lck>d#A1q?tM>~*(gV-t4!>N{Y?zZPKIQ>
zyuH|sN85`tU_sZ_SE<DNA3%8D!GgVSj2e%ERJfyVeDyHv=q_>})W3lxT-W<)JceJ<
zVjE6q1>qB~#&7p{yI!l*qVB`&@i8Z8c28izy0{x5C|1n*a~1Gxi@Vv3)jh}nM_4ep
zQ8&)Hh87}cPU720Gn_H!aLY2t<?morTfI>w?!VK;?e-NJ$rZ~Ae=QdGSXt0e|BHIs
z#7up@hK64O%&(^mIWeUHhRxw_^S1+TPFQv#z6^xEIB2BD(FpFR|48vl@^M|HazNt;
z(fR}FA>%EYyG4lAT;RUcNn6XGq4#^!z4b=k!`=x3^KLYh6WY6&uUGgPxiLqkj#76;
zB8kSDz*r*hl}wJ7`FZRHkzo^FBl(@^k$(><21}&nS)pQb_U}r1FcdC_mb@vYCH-<?
z-aEZNCNePTYQvl+)4KZVta;aakHkH`iKfz<ZhSURnd%v@iHVz&+zf@}nT-&ghmoUJ
z!|#^^3sd=!`FnDxNm9<T{vtE1p~jFCtsLVECSu<IG=ER_cLH%and`Dyl6QK|?5$?o
zzn2)rFFdsX`PBOl5}f}>r0bL{-x_bp#8FMa{#z5f$g-qK?<a$&U;oWasySw0e>qy}
z^Qr$Hx{-TI^;e;}KTdxp@aLevnz}DZT&H=a%rIw~-uyR!kkn+!?&(fWIbi8>WwxHY
z6U~=@fl>H;>HkN%%r|BFk8<FBXU%wK2uHHDXfNg69~ZcL+p~01pRFX}_bb9bSD0`+
zUonOIuVoDCui%6RW^v(~`j3~M$o-=k1N+a$&8&&r4>DTo+|cH3ZKCoybFf-dh^Z=;
zn!c05mZPgxdhVu)(=vCjKfnUs(hV&neVS3oUDMXM=f0Bmy5&XZ^3_QXHkZ!=rCI6x
z21IyfU28PNQi#Iy>%_up5;YYbvgfRE?JOLb8$~W(k-}9SOSF|pp!FsSXz{j=o|MBH
z%85vf8#Pq9vLd(7)_JTcm2|1{D9gCW*6hf+bd-Y2-Chaw+C)5A<~td$1_d)n;4swU
zwS*~K@vz^<`c5dce+`Ye&A%!cn92Q0$Q`o@qclaPe>FAbQ=nvxLh!lCIN1oUA_O^h
zt7!E8h~3l=v3RqV7&%&Vk5M7J8?Tp_WrSQ_E6^k}NO5TuDi?#h%3MjqoE2eIs?obv
z?J~Y33t~1M!*}`jq{-lXQU1c;i5ARNpb!r3bmIx)5f;Fx^qXt)cvP%;**qPO@Wre*
z8n2q~)a#6u9E*u^)T)$G)P<Meb2<s_l~`}H2dX7ucqL)T>r7Wd6K3cqo82g<!88B~
zXYx4tIZDGGqjA)5a>7xB(@Z<bL!gyaCvwuXioAgT3JesFfgdb2;J=c7GTqUCH(N8j
z41XdXYqsR&I(d7BCxTAIYs5qOdm~y1^1`Ii`esYslkI3Ef+uP!U0TTbR|r;tOy+FF
zoI0Gx^mh<p&&dobOOs60SWx8mol6Vd?5@<G2Cn9`FzG)-yBB{Vla5K~uUpS5o$pn(
zucUsy_aKU3)1|6=v{rHV?pSDdQ2Ii(qU=;EZXJoRs1@wJ(W!@MROy6Kt3kHw*5$op
zwaCLh7q^WZrd!>6DCnSJyViDTCkjJCK$zmS{HduUNG~mfJDGzfSHGUAztcnoj8!-^
z3y80}ch4nM1le8M!GyD8-*jmYYY+29MjFZV-KC^SX~({OST(RIWUI7X<u>0VHA$+F
zHY4$|bkNI6atgtf8svkiLCl~kUUsaT4t@MTr!L!k0<Xs~=^$867JtH<zas&}CX1^&
zbN(}NODe?am<T_cppZGw_&4CWzaM{WlU@cLH|uotS>fef|9y1;j)w@x=l8==%zwQ&
zLN7NlJ)=!K*e2r!^&0jgX@ST){+}?j#L9T=l~AKo!poJ+R=LWb!-E=%)XHoc)5z{=
zjN+5xgk~uJY!ytrtl4WQ|3qE3`TvN2a-{hDAMxhjXU+Zpjeip3*lO$nNc<k>hX17A
zbgj`;+I;G+vj4R!_{=S*TGRghUt#DT2O#JgM^_f|ZA?5^3%j_;!|LK#wpSv7D+PGF
z6>z+))yvOq)McA5;GzswYqI@;nmAX<DM7+iHx%euJ7`3}^G>H8m5)00j%%WUIFfJB
zD2FJ!MfZ1Fv>%a8?9)1SX2g=QD6~m?ojIFjNqIbJA_>**C`xa#Qfig%esqP8z5xFu
zYS>4VciN#$;&G1-30Vp#Gdh%&G=)pD_zy)<)eh^rG_1h_<0vdiV2F5CxRE8UQQklj
zC4VZi9<-BXGs8u6%_*o?*`e43C7>9B8e!p;Bue~JG`+zNZ3t0uB$gzqEt#=kmdw_t
zT1q;5?kkEvV5it7#Xrq2#ZTFx*rX_+m{MfXu%vE;|57yl4?DCWq{xw2k`y=fhVEG+
zT%$LW6q+{`q2IF;Y7^eK^9%1Mb|^OC2`EK`w_>Grb{|4`9El|fZ*#^PD@%Io%yv?K
z>Dx+z7ucz_N$@;$&FMs^>`-hH6i^HauB0WVq>cYj6kThFHH7>)3QLmTrrzanmiX2h
z%_N1XyNcA??WEd-cObv;%62F=;Rz_F@NA(<!wzc*;c*m}B)k!pw-E9&gS#myqh(WC
zoc>BX6*i8&jLXGXo~)!~i3G%dC>`k9VcB$m!&(w+R>@pW7I&6UCG4R~h4XvuINSL1
zt_=Q+@^q5qko>9OeasHe#v=}|Adl?L<6moFrBo}k>{zlI$4?{|*sAEeSU_UU)|-*=
z@6}eVrzhF|H9+i}Y*!feB-`J@7QvKx_UpLx8~9_Si2VEMHmidNxbjB*kZuq>KT00a
z&sTQ+qw-NIkA4-{^fo->d366cF6F|Dvu9E}#__;!J2>;UyOeBXFw01p2li$&_N2UM
zy5A7VF7a>{83o!fI(vx!`yy^1*v&8|u+YrmM=AC6)UFF{+oW;8Fh8=~D%8e`pAs-$
zReI)-4A4;jc?xwi6_Ri3#Xn;VTdv1cKV$2~@)0Uo*mg6bOo7LIH)Cq+1=lsU{bFcN
zKp#EqQDo`atD%e}O5}I8#>=PmjIl++SM^cHLAvBAk)97O2$4>jBJDqjCc_J;d*6*{
z`DAz=#u!qDm!BjV<|jQ{X#JfwwU;NvNXS1BN)JINyE2RJ6cO?&xGP0XY5S}~)01a9
z_Sc%&kH-`9HMd^nbB{_1im3%#!P#RiGbz1l(z7Qwe49*`ClSkpc1sa(cM$^oJE`?H
z|KHI++}g;@5-d_1*6*%hm&w$W%1KX6O_cE80jgTzsFgTqnjE`f4NF7{>Q+wVO_u^B
zr$RbvQ*xuq?8L;>l#wEynkwz4Q%fZr2BH&W(<M~TSORqK#r&kz#j_0J_0$R*G%z4L
z?tv43;%F>YVo~HQ(K&1EXP+(Y=L6s*V5@Yo*I8IVOFD6ckpMo)T?$h6%W@l?Qo<S7
z4y&7mtSyC$48!NKb8k+a<8oUo#4*=a6Ni2i7O?g_e9Gu?i@S(Z-8ic#Ifmx?e|&Bk
zr}7=n!?D4$Rf*0BoSSM?#vc2Irdsh$DTxg&Eiu>Pm_2GX28eYg2#NnFgoORZ8Z!P%
zNYnT<nO)m>X(9c6kn8Uv{;|19*s6bwnsOcA3zZL97gBY_KMhe~s>%;tJ1pBcGq%RV
z51@?u4~T}oQ%XPYv=8KVb8zVOibK?ebI9tx{rgoHVrr8?@dNxL<Hh*J!`S%Av2U)h
z{&5|hjXmgyR{uUnD+C~Yt5+XI|J<_ih2nk?ohYu+4np{JPanuG0Zpx+1qr;L4+ynp
zMSYxYbd3tw(+7B0sjZh}xBJ#ZZ<Z%s-#n|uI$uoTY{si8#q=Mi)*M%|NF=)+#T}_J
zJ*M~UlcEihvYUq-?OPAMNt->u2%Y|K7ok51daef(zKNcLT$g5;)P>8~t`?eMl3HXL
z<_>fA>pSyhS<zafFKglJ2IQ|8g^0PI&#VTH(DS^WOj^q_bQ;7f%^q$5;s46mYuB&y
z`jIYGrhfzFnzt1pDA9oZK=HgBRB0MOT{u;_HGmkAr2$t=u?$=)LdlM2va|qMDjpMw
zhq?yD=0kYSbcStj5spa}xQ;Q)beMEc*Cjh&GzGa4eEAJFp#KHU0mb7<fG0ITyQ4&Y
zC<Ql{(<69!oZgm9F-Uy+JXN^4MQ}^h@i}0~9Ak)`7EK+g3+JJo?V3X(wr8p1GiA~<
zrI5M8<w<jzM`$m@?=0m^KZbhtrZ<1L2oI%#zMQeoR8YD%l!D$2PQ8>3IFy1umV#>+
zs+g4tULFPA)GHRLomr2RYQC=sR*7mp7p!<MV}+dxO*N?t=U#5rB!+X<w0Eh~e6JQ0
zUXDq8p1Sa*wr#5-Emdw;qyHwl9HgIYN3)&uTR8-oMv>d7yhRyV+26Gwrfcgv8@LpK
z(-RPd?UfAjyb0dFkmPj)!|p}EJJgvBzQg!=eE&)*m6z)G#<SVXjlA@lApQn7oYJ_#
zOzssMX;yBh(O*!ek<2srgEt(@eK;hANnhW_=`SCH{CO-TTPbT=Ne}x5Qj>nJO`7K<
zgRqS2eTc~4t>ye<ZFkM2|6lYceYWy$Ne0Z?-s*n}L-cIJ5_@?!qAN7y97=5PBQ522
zr&)=3CPrJL$9=YdU1o(HFKe6I=*ZwMp)T8e4muZsS6hMPArQN$`JN>uOM<V<vcC^X
zZD`Jr-)S#t#TR5+nzRqI+CWSfCrcFWJ;knH!pZQHCHab7yz;>6U3Y_(s@~P_tZOXo
z>32mfjJgqS+L@C3pcwxurYzV+Z!OIBssee6L=GT*vME_{iex--O6E}#{}Gsr-y(`$
zMOSUQij%8X<LZR}8us`1*x%Q(zprC|U(fzNN`Hy4D{#S7Ot@vaUw;V>`8N5Uw#l7K
zbjtjliSWqJUXn+a_;k%fAnAE=uur~Njx;H<%G0InqZX#O{0()o!kI%J#i>L^4p*#_
zowWSQ35Z<x47q&N>r$P}=)vlv2%arUN%@~wB#_wr;lNJx87y8U87U(~bwhc-cO-*k
z%^5jovJ6bV2bE7Ep!51kQmhh;t?px@F?(iiZ8sia^^p8l^nBb(k9EWIhZ*}#dFe?s
z$$qg-|Ib--bDMtQ%{6~xN7uG_<ntW<lRBfVn`@r1!?JB0;jm2qS&%iuZLawjJI*%#
z{3L@va=9r3JF(YBsn6Id)8tKUu&NJ-SCB{c=KhyZoYi;HWEk4%4>>N(4mlEU(>Lf)
z*!Kh-4?$Q%1|2gHS&pDXIX@A0#M$p(nU!=*?xa^y=s}5tpyXvjAv1##eS^;Vhs4{8
z*&8$IsBK9&w&cx*wkAp?_gGa@w?d$stlZ~;>vf_ha6N!ks&vPq?Csx?aqlkeUaZ%;
zg~jN%6pFVMK~a{!4^XG1{OvF3kY9>SH!mrR3NgK6JmDXekWXz{n9n^84JFC6K)V;8
zucK+AC}uXJ3btBn&LQZzTmJ|#gs(U=CbmW)9*as>4V?<)?vFreT-L>&yI*V2%1EIi
zoq1QO$IFYLo@j6#$UKw-e=YY!T{!;z{VLhCmKG(F=QzoRXwPhx{vAa)CX=9VWn`F1
zP`amwuy(#^S^6gU@@CGLVY2j}NWnGd(UjAYrAdO9Ckfgo4=-$|&r_w2pDTh}qK+*v
z<g<(+c3L!bq%NF?cD8HkNNmqi$7z*W$7Y?RQZmt-4#t+2EZ>vPy-ymI!#MpoGd2I?
zB8-z-`9sDtQ!DA-P+ECE*z-L$;83ag(Qy;5U6LX-Pw?_+Wmf)OsASe7rH@Z5f>olA
z0a$S^dJb~$nm$q&&b{3FNDSxtXz!w>=E<YJkjco-UXhxoFu=CxC4a4@=4%-Jxg-&m
znjaj)jNQTBS)c6ROMWZ<-&lnIGJ3gzI`xTOu9YEVwJf5TknNv4gjc60kF;ne*@eb~
zg5I#h+}m??GKLyH-_WUY-KO4w<XoQhNO7!Oh-2EULWPNYnbO(=jIG)9NbLGORv5O6
zV3^p{BG79v^z5pu*%Wo*6xj_#^YO%lyK_L%Y>K)Z%F#vG6e(u`n>vcol}*{{5alkp
zw2-OQQ&R4F8_398?otS`WY{YQkuP2}B$soN4wHD1A|B-_(yaXOM}Z>G4}XlVSbmsY
zo$37WL4K&><YkAKR<GmOEIBuPkD!Wumt=S8`AzHqz8A^f>C&w@a70&!5BH{vOEddK
zQ`QRS2+)XKCch7h*X)=X)KQ7Hr~`_h%FX-@gsulNrrXGwJ^g$#2vgk?Sgsq{Q@o=J
zUt!5htqA^%l<HP70&OpRK(R_RH8P0!E){@DM=O3Qf`5?Unw2iL=EuH+t|0AFBv51~
zB3`yzLc*1T@-sV>D_OEg`#2WG2%?%OpyVl+j!f8uXU)_LC<f=&XL45^e-b+^x8i+m
zQJ@j~G%FhVQb|88#9x7yoEGxw*qQNuat4LH44q!aPoLW`-+<2F2B+R`&dPX{LHagk
zgU!GRvdY87X)F3xE($27Mm&&c!~+_WcH5!YG@<}4H`$@s3g{e4hO}3bkI^W((+Z=X
zxc$q}O&t#F6*mX7M0uN%xHpl+y&C?$6@P!QBr<rEvWVkW5JR)*WLSjBAwFqft-S28
z#-Z4Xehv<qK}Zg#YIe{Wl&V=h!F{kXITWs0z1vQottRtMbcJ6r{HC>L^$|NPTg{5Y
zS`ur9t64p6$Jxf8zs}&#z?#)J?C@+n;_wRc$lkoCX4M<sB(DhN-;)dV?L`;ym@Wy-
z`UPqkUL@^lpR)2rw-S9~J;k2(7hJ-oAIbK#n^s%n6Pu^yP%?P50>Q85s!CQ6L$ioM
z8Fntp<WRP!z1oVt)zb<nc_N(OwnMQ+1(~#DM>x0Hp$wLGu4wIrb|`~s$<V2^-PNry
zhE}SjML3y~$hNzOtoY~9svJCejTOe(&!e|l@gI&yDc|&GRuDt8=yds}zqF#CgF~s6
za<&Kj;!>g5Cv#d@|D2}LZGP`Ge)?bF(r>q`$jkKIc%pFc*?96<M8rdGJozW^KF7ur
znM9J?PWo)05t*%?O(!S7`@a_X5QNUXJ7M)rCsViK!=CsYY>ACVr}M>!Am!IwWyU{9
zHlY;hqB1ng*4Sbb-8d<wAm29;gO;a+#)6hpyD4Q$5fo)vXA^bmQ`UL?)D%mO@z)Vh
zwuio+iq$=TBLOQ}&^DFO#n(*f15RhnK9!mNm}g+kmi?(x5$eN|cJWx>x<ch+<1v+w
z`3)`DysD5u#158ObV{Gqrh9tUwZ9J||D~0mUI+b~Vgs7x<2AZ4W!A&IsZ{^w6m&D^
zn^JwbwSl1KDkAsFL*lb3<B`JZo+4N!2Kjn0<Sxb#I~7{Ni@I<g+S#rZyol{P7!<RR
zd1m+6JWzyVayP?qMuxeYf$j~Zj$QC2;(YmnzO76<_bN6JXm^yzA5Fovi%{%lAb5Ec
zvCl4=ynAfkUIe#99e*DTc`IXxofb_UsSD>}ZgnKKXQ|^^vB%~!MHnZw^3#lGrdHCu
zp|tW3z@AUC0Zk{V>7Y$3|275JE=i%41TT+PW^JU&wa4Zki(r-L;~#<*KV+=1Q=#c2
zb>ZC0t&hZTu8;OE+I|4?sF&nGOza1ssAf<>T-#&wdl-GaU>L%*J2c-Ux4|0wgp)lr
zLKMkw<*T-@HMP;sPw|4R^QlvxU~Ib8iBPD_+E~K_5B=><Yzp-U?W>uTL$Zlh?B1^8
z^Em{kXSh^=G$uOB**f{HcyLW29%y09W01#H=nPp{U)i!H6LE=ApQj4H=M=#&v5`Ll
za{C!_cA086LR~mLc7@kM2x7sm98feHp)ObC=pt-{l&ye`ya}T#8?n<N_RPqYkKAi=
zO7_ewf{eU-W)wnf;{iTHY^Ou)nUSkeNr%~K@Sd610!6-O=5=(%_RO%WGrni0>e5%?
z&Xzqh9GfNYnR%w5iup0y<`^u3^nEnUZ<vST>Etf^Zk}-T@~844?}kM6sKfLLIXA?-
zGZ}>G<@@0Gj14h-J4+8K0%5bb$*5*e%&OyneJ-C$P$<05<rC;?+mxFs$$o@2U*a*s
zuiDYIl^nms;pZ>Oe%B7mR&M05On;#-m~*gZxRUHscARbe`9%hQ29{*kth1ScwW81A
z737h<c~40;!@&LpS>9w@)|SkTb=<x(iW{lz8&g)|b$J)6yDn+_#^t~XZX{2+Z=V%?
z>xL!)#Vqb-m-}wDL$Q_n3ea+o9g3~o$Dw3MJ6-PctT2Wa_tMIJnUcsZ_dQ_6-&&~Y
z%cGP<JYoegG>cA#MVPvpU1WO9ihd3b+1-;?WXf<%r~iC&-&d;ePX>?T73|ekrJMd(
zufn~j$n;jsAww3KJ_+9EDKd$`Y(SAogn~UqruShr^Pj~W2{)Ou$izk~R%FW2MFo?e
zv+A90HhPii1EA%zq9<rMwMC{Uil8WqOy8hReexQys-)ncU#hx9If=8dyo7g}@+1j+
zg&@Z{NkwS*iU8d>6P>Oc^`1+VL$ltQg{7wo2|*N=et}Nu!V=xnS1<kjR;sY{Vd&gX
z*??xwbc8WOVlI9u1>dsXglDdnQVK3hNdzZPu5@#T2qY$%&lFyl7r`sZ1bqZdxCA|I
ziyz{fmI<ORoNsnkYndQo^^=^{ftHwz&ulU2rXn1ZY5V<*S!UXv?hU1a9|K>W#RfE0
zFlXAnoPwK66$xG*Ra{~*sa*uOL>)g4hBO#M?6hd=NL@G&bE_k<Jxd+Wiel31i!e@V
z<!czvOs%AQLuut-fIW|}0f*AcKS{y0OHve*2wono9J-kFu_9O{`uIt(;=_y;b}BS|
zq%NF$x%H74&h^pWMN5>ENBuGLsCL9eqMV|aL78H$nDjx6{&kwNPTqm-9~9xg47$E6
zw8Fda7Rm$|ubbH?D?jvI6*g_M*Uhk8XSFmRv>NOsK5X+;zk1NX^3rYzWJmzT3r@>7
zm{w~SuU2sSOUUdgrmFUW(~g{^$gm@KdliaoE<!_MYyS-dY(UQfp$K*1G}{eF3q^=c
zzszCLnypcnpXKNxY>kw*fUSKRqbpmp(;<pda-}4TIj5v3^=ybDZ&6Ai#FGDC97Mh-
zO3Br$q{AeBa8c?$pva3-hv|wHrP$S(E=t{AYtFj??I%2Wb*USd^rp`Qp(U55E>4D;
z<g2`%n$Aw<I#N}r@uZt{EZQCAI*vk+deCCJznrzGAQ^<I<CigY7`3RU^+8<VZ4{&A
zmPo3enbj!$N#+taPS7Y^k9rtgZGJFSbomdg`4a0<Z?~gstGB#`!_Qxj`hXpltscc;
znXW=#b?0EsaP_Fq+Htn==QA1n8CZ|{svVw<M;u;39@(4s)T1(t?DVI7pPv0%OXX|&
z9xqVn_hfz>VEmAo->1+uN9I>^eKPrb(Ev(%-)#A11P(2sBLnlllk$+kg4=Wh_Nl(~
zF$4Au!sLiS{tn7{uia>K-nx<Mx!<pXjtSB8l;?ggDuSY1hFnOU`s98;t@5bJ`=aDw
zf>I}R@$JH}<PfhL?Wlxv9&A;jNdb4zf`0&#&r9_E#R#vR;sQBq4kR4bD5W*T$)!@v
zz1BRl@%lS|4@pJ5<b^X%JGZc}keo$&c{e)QX5CZ*`EQ}P*RU@371!-2om(=5It3#u
zYd%zjg~=-J8(_-qIryk$OsEUzWB)0LSg)<(i1okDS$`HKl#dqSm`vRTj0`h%NB8t7
z*v=O%3GhwurOEm71wGN8Gok$I6kId<PC0N{+aP#(Qg=&CC_h#Nw?rME07Kr$7-FYI
zQ%CB;d1z<5rjEq+EOk6963U+{!Z@jwpI|&QwUX`)rIqw{qK~ivhe{}aB?Z?mNs&+{
zczLvP=!Ek3i(r-L<KKf7-(jq<Q=#c2b>ZC0t&hZTu8;OET9S)A>aHA!i6j?=M1!(C
zx5ER8u<7d<eVbqyV!S&v-!$y$gr{vXwb9N`aXR5#>eMHayISUYWkOj5X$awxP=_~3
z-u)?F+dD9ge5D+kB7xk)7Ec~uS%@dvyzvi^&onwiHrCS$+pPr1uw+0h6=Gap1mDDl
zeh6guF=XvB*KCNoaQf^9q4{)T$<7>5G#jEWm*wapY>1SxfDL^gqbnP-(;*VYa#<x4
ztfwSV9D<CziDHEi%gp_J5cwieESI^G4wLx7iQ-3qB2N^*imq6qm|dOeMDaZhH&}4X
z^;Y1NkGZu4FA7gy#`t+l${6=zw&c|DxxHf|twC$JRI+$eZ$s%@vK}di@^*+<4^d19
zku!b#mShm7q`!nyG}6b*1=4*#XC_mS-=$B8ZSyn6JN=om$?W`62BZ6?%oH6xDYG|%
zWkaUS{)#Lz0;l-PEw^Oi*^RU0E^}v7nSEw^Y3og*{SxwdvyeCB+`EaguaH?XfD2d-
zFsys}Cb58hu<i-EV(y7uooV-UYv>$3i6yM-m*kKLmnFNPXE2JYHJqBFt1^A#r_2&<
zdvA2=p_{LmwUXn)lTZAI@<G24>Lj&cC?E9GWDw?qo=W?mD+-VxJgiaHT7mxTzFE^s
zZlj>Zt`%7uHrrgQ^?1x0bhYKxq`&+TH>XvsxmnHHk|`}gnNfFIpK2*5j-LC9w8`N~
zyOiU4!3ympPM@!>USo%1s~-s{Gdh$JrmvDZ@k`OP-wy3M3}t**Hr^a|+U-^lmBWS;
zME7@EksEf|cg&{I7H*OwVTN9MN+^dE3%|DXu$>AU&tAYKW9WwUwWUYwux#}<4r@uQ
z8LlSzay!m8{=77UKLcx$ud~Cm@rc7K$Rm66o|<Ha*F61^*7@1d2%9qc##RdRp4jUD
zU@90gw)!x{l_R#|GnkB~`W&Sp{Cnf4r{EZ#6mwN({G`9>EPqW=)OGTAP>%02Ru$D9
zmR5v3uxuF!`LCiY2syPUP`*|KM_GmZ3U%s}yLuPCQdbM)clVHF!Y{WHwSq+|CX(nx
zh0$qNlh4@%R9Z~Umh3c$`cxIA!tV0i$a7J*6a%r(BC6UX{~TvVzy6=O%E*^Zns1r9
zO4395rhK%BSimSCj)`H|GqZa56XGTh;8KCk&k6}gln$RnCtHYPmJSb$lDhqv^|2@1
zcIIY)j06hwP)4)%Tw8G0&uGSoiV1WbM7heu9bkPp<^7$zS?%N8`Vr?=prwxEJ%jYu
zrn^}RvpHix(av~xUf*_zn=OInt|DxfrI;%igUwP5-P7|Nc2=rH?}2d=*$y`1P_>vF
zQ-ICIO-fN@g^=Losl{xwh)b%swT)8fyNci`Gs_1ih&zWEckF~|3eEa(j@qT5a?4x_
zSqgnt6bu)O@J?!JhjGo+Qo1*kmaZW=+|NcFN=sjx0&Ewis2CEwJX$(*#qg~~uu61t
zE%D+p#tS<cnohDloPW7>lJT7Dq`i+ScyP!3{Tx%DsCZBaGbrV+s)h$v66s$;A5tO&
zWp`-4f7r6&<3)rZL$A*X)yO)B@^7GQ_=7?`(Pop)B%tpw-L!`>11f?d5F5}+dG}uw
zK{&Cba|zp@F>G^N66?b$v|EMd-5FQDl>?4uNvzK|a`X|FL>gJZk~W%_WG6)w59Mk|
z<}pu6@o;)e{^FrRh;pI_E?PE-0it>+*Qb&YllsBc!yAAjPtWhCE0&&TS7$msf9HZ5
zl^f3Sa@}n%MBd4(9Nw~|M7@dCl2i5@%<-}-hU?5mGFnXERxac)M5)InrbozGDSSaP
z2-C<*p^-}~6=n^1`ZHLUW@oTcEbg1jQbhFRvJQb^L*}x6hiq?foiHtNnbo>H4^2J!
ztpk^m<GfRdd0>L;8vfzp5^S>MR4Q&ea^~tTKYd!_Mv5P$$PF4Hd%Hiq93)Kk+z;Db
zY4vLph2*R^wK}ytOD4u0MC7XzB2S5<e+<Lt-6bMX;n}VO*X4EeQclA<`VRs}?&v>6
zSIp6~t26EBpHHTjd<Np=`T6HA+0Q3PE!oYlPthpSE;pkln7rd|t<gq0`P5xy4WAb(
zE%m{)hMYeBzbAt*AHQeGK0a%})9>Ri>F*DfO#8ZcMMIB^{|xvwq>F!&tZuN2Psv+)
z_{(xo)a&5)knj5|A>)A#{$k<akq}pDboreYo)ec^nWw2z)#Y1l;YdHn{ab|muM*@>
ziR0b?%<?*JiHdOCx94@-lK;|Bi$>@>fFpO@lXS%#H@iC1j{61Vu;!c!rui!0W^(eJ
z_rWDQZv(L<JMhgJv}QT+jXmw8$z(iH7IMASLUKCuy~!ZVnIBrRGtV0E^gHt%{av3a
z<Gv1E(b413uLaMBbm%Tw<6wuLA$RH2cjlz5*RdZW-?&@Id!S?gHO;XLmWlJ1QaLRx
zl#)h?eiezo?;=$1Pf$H2Zv15!KCc^>s0cS+%j?D^w}*A(Zvc+mjsGXQVs4yWoeVc_
z&MB!G)fcIvx6uwXB%7@ij8y7Q7<OHK4F3`I^#3>^;9Us;>}ps7Ruewk(<T|ON2t-M
zyEBzmGjf|z*k$%f_{c{k?a>arA$I5CE^T+1@V0qtIDzB<Z<*CoXp`Qa_;Nm}%&=G{
zobl#o_&4r?8U7m5fIk88mta%^D0<BD9<$zjS@bB29(>*diYYu+Xm~8QRgmM+u+<4F
z*!(#ipBn{EGpw@%>}4FPUI<)>;=%4@JQB4cr{2BgoOnXEajW>*A-b2v>nfc9Z)Cyt
zHa^zDc#XVMuQlE7!FW~ZMiJhO8g}E2cvoyEs@20;yhn;yT<ylAcRKYBw!}T!<;#(H
zlppHp9`x29jK{^Po>qX7#;XoC&qYq7jUTInRx1Lk?tJZdH(s`@8?WXllmYh$8cc|v
zouH0iYok_$MGW*b?0R2|H_c;!Cbr{gw;ztzhS=ZH>@>>9>Dz(bhvRkl6hN(AuCy8r
zrV<C^ZTyZ~ciEltv0CJnneKuNKoT8Dnc0msfy*`k*4?8n7`Z&2@SJ*t<Rbey5^r>m
z*TBEXi8`UWJ5qJ$Itv(Y-7zO<f`4ccxK0Ci#vzD`OYW5LV`sFva-dZ=9`8i+GuO_9
zHC$g{J}kRI&<e^PcB|Cg0K6wuD-pDt^IRKcgpJaTx2K+@hVZ`wgN)blahZZL9Axhc
z%m=B?@dly+S{Hg?Q5`gO<F$3CxzKTNc!R%Tc)8o`#+$L}rWI77`EBS7z|<IHM`4Wf
zAQK#^Ri!F;qwzYY6Sd0fR=m6hMO$0#G}`4v8Npa~10R8QgC_PUW0>{q4pVcIzPi;8
z-A=VtKHeZz+Tb2<w?en9_hed9iONT7VGY}ZXsmLz7E+oDga|Chq>;ej)_|B0X2C2D
z&DV-%<@s72#K8e<r7Hjl-5{#9nve~RR)xq{2hOqN=Pk?Pjbv#eSq<%P)S~Wz4Ljm>
zZMR*9eegFX5ix7%&YT0>$tdFUyeMjiyLauv?Qpie7<sMcY%5rREp4He3<0z(M{Wb+
zi6EX;xIc5*l~;qG>w@`8YIwS%hQC}653_YhWskVQY|V}484S3BU0&3v1K2pyd9b2t
z<aKXZ6>o=5Hp+={CIpnjHfhL-ctg0@jF6r8aIU9LXM4Kw%6f}be+}DXTNayaAI9EW
zwdWRxfhvw*?e>c203QgR+Z(vdScuj~I&FYh2kfF+1pae3a01%Xs^Ik}&;j8WHj$$U
zJ9DXKwFc>Kz2#JyA*^K12rAQ<bE~l6MPT5$ZUh)xDrj}2>^7WQz3f!0<Z&VShvSVj
zGD$pi>+{^3orqVs&7<9eow;}w_a-o3Xkx9&H*u9C2j`tIVL@*;0x4)#1V@W8s_s1b
z@pzMQx9Kzh3b!%ZeDZ_d=3T3@`(t#IUr+iUrAhZg^yknr{JEX}tfF~&g#Nsr9{ef(
zbmDR96)m>e*u38f*5?Fke}c8e-sg)jv9JL#Y3}jH<57-&G6#4gybP@liR?*cVmaN#
zBef87WA|a&)O5wuXg+6C?@BopBQ_;Wy+=5jcM;ZvrcaZ<zmxu4MStExf8IfVemROi
z=ZxXcL-gkr^oPEQ%f9#P(~HY}da=1rFS7CJMK?a3&td0~eA-Rw({59ra`8Up>U~Pt
z_>|J|Dc1LCEzpVA>YB}Xy|z+=eUPxge}0D?m~6*|s^8)5AZl1bW3E+CjD<BlH+3^<
zJ#LIRE}^ncn~d=>7X?ulo@GmVEb#=v0NiV()gkAJnYYra*SXSSmLI!8+<z1r_KNNu
za20JQfYX7_g|H3|F!8d+?bXR`rr@h<%}TvfC4*WZ-JYP>hrebA@d4YtF#CT2)gJTB

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/hf_model_mixin/index.doctree b/.doctrees/autoapi/lmflow/models/hf_model_mixin/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..02cf080000664c9822725f6b781dcdd5ea574d18
GIT binary patch
literal 81018
zcmdsg3z!^Nb*4wsjAllUkt|Et9?34*78=RY41Qpnu_arUCEG|N2}uSU*|ci9Yo@x?
z)75lUOB!s0AsA3n3oL5KAFz&1ygW8I0m3eUWx+9lbueN1fMp3}6CMt)EP>DYLd+vJ
zd(M5-y|=3Rc2)JL2hb;#s;ln3=iGD8`OmqJy7#``2Zqn;ISc;;Ltf1;S5J<brBc-`
zSza^fFS*6Wj8*lUr<<Ez-@Lau5)9PM6Q0|s7p-P+7NjWJ<w~h;RhxG;gXNgqFIT)#
zmVdEl75%bX4YS+5cAvfYuI7k65cK+GzhX(Rg0+>Isfv4IbOxwHA@)>(+!bcZC(G3)
z=7VgD4qGRYpEjxQZ&H5yY9?^!H|>?dKq0J-y)fuwwb&Ods+6l%b6;c93<gc#ua_qq
zKGYs-yRcmHux5*k71Q(JA>8*?%^7;vUux9Ek4dw5tZq$VnntkLH>dGc(;l*i?WOiw
zd!fB9c;%Sw&RAphMzw0y$4<EQV`F<Kc0-l<f54zOMj0D38@_AS%41Y0V^k<(s!+zt
z)sl5`wC+I#FW9r7F=?L_^pa<iT&c+I3!y^xdXVA8@NWbB+X(+QVdX$R^)ji_NU*48
zmg~*n5+K*8Sl1fbCJ8R4$+l`!*?s%`rrEGBwYLgc?6WTbiQ7x4S_!>a!5;*JJ49QM
zwqK(>Vhs!>?x518C7Lbki40(4Y^TZXEg-ReBC!>|IoYU~^+M6D6ug>M%pkl)Bz>^r
z&fD+4?B2UCznA~tL~{EDx!~BDYy0ubt{%S*)x~Sf%$W7r?Vt=G++YneXfT*j>2)($
zDg7!y<yAAheGvXs-05knj(4~0F6_AN@cxPYH|^S6nAmaPz^(gkjy*qo$AMk=aPR(u
zI|_$)9K3ng;R58iZSSr_A`ds=-Y7|?p#wpmKV5elwV3=7FBU`VWNe&%$a11w^6loH
zL2QpIw8t@Ma+fPvJJXN~c%@tbqSvHbnoVSMVck8E_@WmLEuIJJHUKDPPS?$vEoBU@
z9<R-gyOYlNQQ8{Dbvip6rHQs5y_3j!1fXDP&8n8l)#<`X04BNQtMZl18*p)@>|1rS
zA{A*r#B8*Wve6Xw`JUGH33x+T3l`L7n?YY~b}Ll3Y02<}Vt*Lg?cIRSEp0pKg9-w8
zHiOOSVY3O&40Fwx=v^KQ7Q-)pwg!%EA#xEcf#252TBTep`^~qI*C)(+wRsAfYj&I>
zUpqGx^g~kOJevCmIp3^LTYfXRI7L$u40$K`jV@!yo*t0*17Ig>>^{+q56I1EwQLG(
zuhaHA`>z0Av7EPKHeK6chPOh(KLCPnix?)0I=FjiLv+~wYDz+S_D=yfZvt+9U*%?`
zHG9FK(4dmwvW?*;Wf5llF3v@8aPEmvi++TYu%BgCd`=Qil1=|YRtD8BN9+%#Fz27?
zS}HVW`{S~WItw9EO>d_5r|rk>zuad(5nQG<w#2HUR?e5{Jrgy(XJl#Xbr(#}Sq)-h
z*7uXN*7yH9w?1bB(A707w{)PTuS5<5oJh6}j7zduAEr+-y}ff;e8lLSTd}>cce5r~
zsVF0IBaF`3gXu_5A}Q8QWp~caZQ7`n>D>nT@|fOedebAdokzQKb`serUXMJJWv*9T
zw7zL<CF1*>8}K%r`*XHY;vITG=L9K6MdWBW1>TmC)7K`48JeM;XnF!|3{sOb$uy6i
zX_8TMeK)o$HP>Mcuae<D#Pq;km)~&T*`{Gz8SeWbUmnA~Mr*jbde5f`KR~3UnlL`g
zX2PGq9Mpv0!7``tYtGv!@lH+n^Q4&eCQQktGGTq9H$uD%-Qw1Xf<@T(Zo&W)L`7vt
zJrxy4uv!{YFe+|!n)dO{DE5>f_t9u>IZ-fha;7rw)l3{=?61`=&#E7{NLT1pke83%
z+)0Kw1k1|Moyo_zB2U?f7uI4B%3{NE;Yg5XFlzBmHUP*BKl~l`kDUGms+#E=5JsY$
z-Y^GYVHJd62f|34jFYzP@Yt`kk6<-mgh?vqjCK?+e{s?vlU?8(2M6h>&;*X&PT-I&
zj2&5(npHPQ2iWj!(>H2$_jnlwmkrbKe6w0I>m}oaIcxZ?QLJ00Zy9FQ*m2`dqv658
z1%+OVlVHO&j{18f*LV&e5OOv`D}dxX8581V+^;g0r31M$^kTQ{77}dZDdnU|uJOjs
z3ynIz3DjBY{Fch(C2zMsRO|fSZVzdc)+W@-+uZ^A@_4&VT5qRRv>Wa%!Pamv#krY?
zMRjlZESq~fhB>HvyNPAiAu?7NL&hDHcuV&d+%PxhzGe;z60rQ*h<IA-6xI_1_zQ)2
z?#?V8;gPrx;`1z6JVeyab|OYS8{m-{rnD|E>Xe;J5(;*nfnXxZ2|GqYYS|Avzk;03
zzhf){jDzcy{0tKj%rOY|#2lIMzTvc)DGMR$deQG`T<;LTAxuRXJKd@e1Lz(_q%}e8
zxoGqM!I~{O-j-{LbwgMDi7*Fo#lyhTXavmD&7bw5+k*L&V2q8~kCdwh9V^z(rEN+%
zB&<xPNo-T7g)#p{;9TPqSU4N79N!eYGE7YRjk+@P$_MQc*~!EKNs1G}IETq6f9>92
zofao~)3iw?8tVI@*b`kvumrf>3NavIBUpx9kywKKR;CFQb#?v`%N=&U9Trbobse<o
zS-}9QTcl#ncQGmTYg`zD&7X(mB(vF%Y*HTkvGjKA{aY*cf>qL#-SM(@q6ooMGZ^Bx
z1+!YT-MZ8w?PjoWii6KNY{2<+FeOhrrwlXPkf&T~5}8Pj7Hq}oUl<WA6^MWV=j%{6
zf&egk4q|BrgD4!BP=TcY>da2iN7-rGj|aVUj>PT@`llL|3K1ZsKUnbHVgcsI(DENA
z&)x(xH{_{EhP?d)N_8D=X=wbvV~s`{eJN@5B88Cj;=^J$&?@udIHrgyBhZh~A@4;c
z(>J5FWHU_86;M#H7$iX^nVc)}Cs7yqch%iW=kHGii>ofbUCtkGrBjHff~D*w|Gg4Y
zu<ChXqM@qGpYe2964LW5-^T#^6jt#YOjH~Q*)T;u2z0&%r3KrRflM}SY@`Rn97Fbc
zXmq{_Y^jhw;(SYb79`Jk5J+^sCnszJZwyA-q!Q@z3RJMx!@ABhK->8jxU>hnLt#<G
z3@&U;I?qa+JGbF`igL-d6<k?iQHhx6EP(gULbzzSVmOz;)c|Bv8>3}cSD7}=3Sn5`
zWC#;X#UN}SWTFt}oDa9oFkIAHUI<SBj12s*&smR&kdku|eUA5UYC4zj=Mi+e5bW##
zrX)kd&I{o-r7R>o;ss>IBj+0A>K%A_30_X)WjwsQ7Vb_t*OBWsa@|g@*OTiF<hlc|
zX9zdO72F`xH|aqm>tsWxh><W`bLQdVF_8Wo!Nmu{q6l29kHW>D;Cl)e2~`zX@!!L8
zI>w3wmb!%%SleW*5KP6e;!((y87q(y#|pgXSb_HutXL0N@m_e`5mxX5@?pgfkXwQk
zKg2tN736NdvEnApEJTJ-BmYcCr}&|rK05PA(z9sJkmNb0qAdQtIEo~{#`hFSw3QZ^
zGSr5-4Dm+-bKSxetamb|2&Q6~@^i?P8B>rF#}vHhn1c5aOt}~^<(KfdBTV51<inKf
zpmiyxY{NT(DdcXxG382mtk)Q!GZ!#~qZ{{tv6~Fz$DM-4H)IY+pbKqP3b3fq8|OUY
zkI@NyzN0=(mhXF8jFMaM>#)#q+?aIT3jFq9wR3f<JiTq7TeaY~b-Y}(#*O_oT(ECe
zj(C3kE@PXqak|#nxCLfotb$oAS`}FK>AQ7W3cTKJRy@cJtI~1)kFD2-aicOVg)tTu
zdQ-XRW2uzpX5@5>v3c_`SXVjiZ6d=*IQ%q3dh2Gj>iTBn`+?x{BiCK*!;0o{V<Tyt
zjoY^v*OBxv&H$^oVcu}u2$L8aw{M4qHGn)Lu)2ngG6;~lJpwZFC>bqF8Y>8T3_idG
zccPHVNFnObZqRe086ALij$)Q!2g%E0sL>%~bx`Z&G}FP590}Ss0Mwqz-sy>uTykue
zp?8|Vx!gMq)Cnc#F4TuY>i4(dD}#TC^96cvHqj|umpx;_V&_U3zOj%7Lcj1kWO$+7
z(GlXX&T@TUn6AEIQDm$Z2b&9IkWP}^peTu(ll(}u@O!i^9Q0J<cJ(l38Ft<c|B+$8
zb3$OKHL69w5kr%nan1wW3DSSwhIB1X`ol75$An+JPu~hRp3o9;X5e)z;Gh+B9+9b>
z&s<aj{L?yy3yIm7@tQ!FMF``oT0Z636nU{O9S{xI)RVCiWhh5SoUFaak3Gs$<)XUa
z9tMl0HT4kLz%t!o=S#q2bPm&bEJ=cjil~FQa#d95eJ6@w*m;0Zaoz?0g^ddzy~&=H
z8i3j|@EILV!gWnN5K{g*l4v;!sIAMtQno%Kv(-Ye`+BuhJ_SX@bdG5@qmo9m&(TxS
zX*Q#xX!aruSz75ZeJSMlLM?5%iDhkdz5_KE3kTxbO37r_%Zq@UOnOP{KPP&rM0iPh
zIS>YcTrs65LZUC#N>r<vyGixIo30Fj*o6f_7(&X<rDq#w0PY0gS853dJ#HL@Z^tad
z4x<Ynl$ERG>rTo4EW+TE)1yBG$zytw;MJzgM#V22H!BTGB@c6VjatpE`vng^!t-C<
zaPccrGEQ2SxtfEvJ)cMG5TqOTkZ78R>PhX~N0ak9^Pi&D_iJmdZQ<8LR)|2;V9#ki
z6&?04DrvCiJ$foS>|s<C>=_gyrlPguZge4}k<|91S`Km}5TTM8fIdXqCynHu)bg#3
zM;H}V3+~|x*K>~~ov-Ps=(He1U7prc(P;stlF-aitawICBV{|YZw=*XdP*l}ORMOT
z#&pxD#?n4**q{?9Gpaj4SYgc60`Je)Q?a0kSa8AsKLbFF=&7WrizC;imyD2bRcF5h
zv|gpBc05|A_FjZmRc&Wk*-}TY>$HP96z$YNb|{jA*E;rG)PKLe{<;u-4`fa0TJF+Q
z(SZu1lEw;->8a>6hfz^nOQx{)UM+38{ksv?&XT542fDOalR?T4YHO->LySrqta-1V
ziVkZs)a9dkDk(HZ#Tqi)%B`-K{t~wKgq~Ustf5t%1FX5M%?L3L42wuMjL=6^{8^_h
zh0CQ)Dj~0L>gA=wrvC$3Q)1H(^;C4&#Hggfrhn5@(P<i^qF~b!sX?eGND7EBkrdkX
zFV;l~`7w%+$s7+Y0&bv*(g<&@mT@iQFe+&x-_3d|Iz`A(muvJ?bc#T!BsDZKNMSg;
zwKUQ;^jV;?%7|djz?ZVEf|S-7WiYFw+L~%LgHcK2jH-GnI%kxjF8AoE=rn^-QJfKN
zEjCLN?Hm2Ro>mT5L|IT<S=O}FVQ(*)cBG)@sI9Y^yO5sE60AL{uZymA9)_%dK?<}k
z!U1|ti2e|k^$|TSolBy$=0HIGd_p!!MhiBFLc&|Y`RDbV>!kF#1S$2=Ibkj*{!`HV
zx}Kg+Qj}gUNu|jYoHqkQgyjMbJmSK2I)Mm3VFr;52cCwpW*DT5Pdhr@T9)!=9PQ?N
zCfOgy^&$PZAVg*=wk&0WOb*~;_YBKokrH2)g7<t`3f@PSrECE5`{8lRWho<exF1k(
z#f>KH$2V?-UpT4@G$!FKU5zk`3CU`Rqg0<^9l&CiSyJDDel>OxUNG2_nTGucAL?*c
zVYzhS(?#uBlo(naVMy(4rQ7UdC6p1JeULFzQURsccp>BylOajlVZql4@R11J4vX%y
z<6w)zUOQ=ONoVcory3if)lvAmbr6^y_msjd`!IbUlMT%A3XPcP^y|aHa@Dr#W%w$&
zXpM5<nd4FgeqGJ#z}HWy?JKvET>DGa+LK-JW|SQ|;Ulf#_3QCF>!Jl)R+fy(S)<yR
zsm;RnXvX^Wyk6QgluICWGcf6C_FYq*?k=p7XdTvI*F!7i`j(D<NCInnaAlG)qFRr-
z59X-6Cdvpm@q~q0FG`(ll+TjL+f|0D#+#}AXC!=jg)BN!Kfkb1KokskDxnaLRv|{R
zd~h)p7DY_v1_(i30oKVuDUil2M^bf=?hXq~S_GsGgHJ%#SWh{5zVDQtimrpisHExa
zJ*cOm3xg>YTu4R-eMRHp8+rCVO`L%4kFDuSXs|AY0>LWTbhr&jjsdx3v4P++78{sQ
z|I|xXiL0Pr&3qE_vG5?W{43xV@@DoS4JJiqq|f|r0JJGG$%BsN`;dkgz%5*DI`(1T
zK!_>o*M0PBny+9j$O`$d;vHGdO78NmW+e`84!**>L9DgP3!qn~2T(z*sV;IQ)sfb=
zv<FdnMOXrm{u4Gj1yVVsu*V-Wt>BHSde=-~`opj&0;an_(Yj@uFHme#5^u773QlA8
z`7~4_kA3zIw6srI@Z3f*dBjF{fsKBTi7>Z~@&b5s=CIMVSQBES>+p`)D7l+&8{Mf4
zYUef6tJ9mQLa$>>T^1=0ms+b)&y^r=W~`CK!d|b!G!jB|V6T^lMG<?wrHl4@zGAPc
zV3JK%FdH-3tx%6VCc7W<wQ#8s@pG%i91*L%1*~?A2{O0U@&Y=r+9|9IvD#_8BUVf9
z=G$s_MW({@+U-ly+iiqm$CewLwNG%guSFf8jy9HDSnqvIQ+PvnV7>Q*MG@=W1Inj%
zUBnb<x~|KA3=7QMb@>mF73iyTJIP&_hxJr+T^B|rP1og5^;C3S7fJ=YF40+Id?Q1z
z>RaW32(`D*x-JO~)~)bptpc}E^GF6<!EOw2Pk<ou0PdF|UkkvE*$jtV%oBm!Jpj40
ztSxdwE-!#vqS~0LIx|1S>JZ5N5#ABVC3o`;xz~rwmGgq_73sk?q}4IZN^7bU0QRiG
z5MZT*0$LZ~`?(2>o@dI%#^nw~p;b)ir9;+LTGq6KLPL->C9<y9Q_&$Sqml+$$MjTm
z$V#aIvPS1X@rB%Usx7t!)`aHj7O<{SfK`}6GNcMdW03j_P`NyidI&_<0#YR_;b4lX
zBVc*}!1Py4-*ba0FQ5Z3b+94?OpoCm0aJ1}-(b3(f0L9KOE1rWrQzpK9S*IF@1_!v
z^bH_j7OIML0!ZJ8ujd9x?+c3}G3srgVBNCJ)v9H3!6qA~;4@~JcS9BO80JHeuZ3X>
zp>w;$#1Xr^4eat>CcfNu$qVSfF8?dmgV^PN!#iS^<Ziz0@_OGrW>w3tfnAex&Uj@A
zaza|22RTwv@e!cE!4{{z$N`a1*yjIY8o?X1gZ|@FVNt|3Z||aQs$NsnGub!=qcP+B
z9MmC?asB}EwJ=Ue?A$srb;LSv2kU&22{5;H@&b5cs$oy6F39s(5n`Rc#yeu2<ZiyL
za|e95ShNcz+(9v~eZDxo<CLg%Y@th(!s0f%W=UwHN>X8^=VFSv@u25~MG-4K(nTvh
zSMi{-P?GIbuo|<|A*e+jJG~6@wXjn~^xRT0eZ*3afTgZvLd<QcynqfY^%krOvD91f
zj#w(Wn{P|q1*7kD!i;Wnlh;}^jL$|Wc5Jaju>#{(dlYqmde&HSVY!n`Q+Pvn;9p-A
z7DX)gRb8}Pb+|7ooMh`2?8dD34yZ{U>%ANDwXoh;6L1U0JP`}V9o3I8Vdl1AUI1^{
z9KQA)SRG=)r}2(hFu9v=3%>Ey4YTUAU8?e$?^WrYtxT?C+g+-b7B}2K!KSB%tELrJ
z`xw&>-nboD?FYi5h}D{)aLTMo)}g;oh2>GRQd!du{e1$m0{JwPDqql3(GC4EDrtuP
zzM-e08~USE5{CZpg&ZoWcG-65FQK`*HNbG55*{g9oQ$x7*BHV+3R1{}u+Krh76_}_
z3WrzB5P??{;Pt(%6>`HXFMwO1+Kj2XGUqJi@VXlB2)vTJ`G(gWH8>ccLv+p1m651*
z46aL*!s6Jv1tiU|RY@wKbre&~O;=`fSQLR)I2o-oXuUoxFEg}W3t3Y_>x7<)4q6$N
zG|+lPPeli<lu81$;tRRyRGVxItqIN5EofDHp0W-nBdcIFhO8Sv2zik8b&#(GvMQFs
zffdt7fOQIhbt7wo+`!5U=m1#%5UWDs)(7#904uqhZ(zLvHcIu&$KfcuQr@UF!<t(`
ztYcUml*@{v>Yrhw(}-10DM0nZOe=V!szG6<HMft2MG;Vi!zoozWs6vlopf{6PlN?#
zM%7P2)|9CFB|Q}#sxm5RQ1$QiRCK6HsU)oe!Z#hjYC?l`3s%)d(vl7*qpILEhN|xe
z8RS9Lo@FgjRkjoks+c?is>c9S{|jq@+@Q(};P$6_##GVjxmXhds^{Sy0abE0-=KO+
zp#aB#!N;@~9Q?#4$VN70!B^PD*LF+rMcB6VLjnmzJ4V?cq0p4OcyBVcZZ~eL!v5Sx
zU@z~gH|5r6thzUPpl%(;y}hl{VJ@33M!OWdP0wdLV=K$GgH;I(@~;LH-w6+JIVg<`
zA{}^u(Xc4u0cv0sP5BeHEc+3w8@6W28QFhE{^WtXLg}jvA&6T-?i1aO?&I{EAZyIs
zB=1uEay=EDyJ1w)xSNumiq73ID$4c;x@~`nOym12>aqx#AJ_6JcMOzooxcNGCBAiz
zZ~Uqwo_0?f^1&NGKG~|cP8wJ>4efR%REjE^%hger<+81cZ%Ck$oXL3xc2bb{C5W3I
zY~V@{(5qW;a%9z$_imjy1Artxz7MymZi{Eqr}KY&W;CHa3%T^+k4Q_U`V9AD+KOq1
zT^SX1*fn{F&d=(p=wKy7UB05HqJtGmB|%9pN#pb+MftXt2iaUQK}I?@I#8ChSn!{)
ze#1`Opu4tS%}|DulOl`cC|$4|i_$kirSe4Seal-$>4{+t3)h2-+Erm>Q!^6I*Ffl(
zGTq7@&hrAa;d~zrYm+`oz5we>V)+a4j>PihZoXsrn}iyWg)}+i_zW}i5rQ4Z@Vg@g
zW{Ka)ijz@Nak?Lk8hsB%isq4b?70Yhu7T#Aqp17JuqYCBd%s;#_r8$(%u)ACA#2L0
z`%XO-oqJ?d(nQ^ko{G*rGAiAPy6@5QncIcVdDMN9G;oHfn@TaQr)5$1?gT2b3`D3N
zi@Fm`?;&s@;hrsYWtapU3Gt7TMgYY7&P*FA)N(~Lf1Fxe=wMkilOU*W4K=DOkfKPP
z9LWfFW0A}lR6kE7^EZ&M#Vk(DP4Fn@f)-H>^T5#v#)AmvGNv}UBN$#l-U#L&v7RJ?
z`5xYp2!`CvcLcLXG`+mWFlQihL!JRinQq7NjNhU@Vr)x)xFD>>%Y6JFJMB-6e%2&2
z!h@hzuLaaS&vcfnoC^4v=AeEV7DZyA<Iz~?&NC?;78)uiO?6}7@|+=EEU6u)S^-&8
zhENyjspvu|MkP%M^&&kLT?oafbSH$mQOl<sK4cn&nDY>7J859mqEcu?#t>>Lp`s4i
z|3>Z}Kc#Iptzaf;9o50O<S~9<Peo@cj7l0)d4ryc&Qur`#Z+?b9^cgRklPFvW!gPn
z8L59eZKgB={D78u?Z_;nqAEl3$n3}TRCLOap)Q}*Q_(2{rIOGyiGCaZ6)laljosl6
z@(GQtf>jzLPNN+^*49?59n8Gceae&fp!ubqif#w)3{cU(LK`9Iw1ZJezl;1DJ*^zO
z$Ws>7R+hC*bkMQK$$ItFC8fy}>(wU&k=^R7UfiO`r}P|>`>cZNSfBN05EbX?vu=Z?
zY|&?J71i^u>xL8+iI2e_haT*+EQC+&!OFXd7hwqnE-o>Ma18p%SOd<fgtZn1EVF<E
z&JFD07<FqJwsE)WTLusCjorP$t>GVOfq*e?Oeklt^G-}>w7z>^lm6YRr6qbqd$l$s
z>po6U^If)QL3+2coUniMK(J)8;uen;$f=g{^zRGyPVBC@Cz#;iXajoSJlj7z+@0Hb
z6&6NU54OxAm_C||>3f0cJ0YiT%X2jAS7PaOH*w6O)Dw=LpDI_$@%41aesOXrn~NR@
z;8bZPO`+^tGm=nlU)2KybbVq_j{P-MF0*n3<CD9#7%uxgcnFU@HY=(UdWP31+J@;F
zM`1JXLdBf4D&A3}Ts4mF0KFudw%e^AvR>V=VEgi&71Q(Z>|~<BM+ZlW?#zr^t;`y*
zOI8W9)h*b(y+oU>T%F!zoG4c+M%DFU=Wxrjd`y$v8Q(JKxklMHJlkzlV0&@Pr~xrr
zs<zTpCq_V-7J&uH0t?z4*~Ap5O)8;2_hnHZ=e0r|UJL5*8lpkK<zSIX(NYepJ-1cM
zaRYCiKydi>OdP%yQ&Y}RZ;3$;3MG_OkLhS=cfmJ*=hbOVaABJ|N;CeCM&u67Jmi=C
zidBH)46HiZjwH=sscdb$eFudFN+thkq-3?99Kl)VgS17F9cWG1I|}VhDlN*V^BANI
z=K?NCWAvQei`oN(MELQJs`2LE6r>TB7L{b{Fi!S=uLE;?gfx!69sdZ$MO_JIq1d4%
z;hYp%5N;YNsCa4?$zd(wIzbJ8#9qN41d9){#aawK!{0(SKgM!feg+lMS$hb${ZW{M
z42u39L{>gL;zTJI&aCt(crp6`pT$Ev-Re|%dJqCCIo28-jQk<mJL0@fE~$@}#1)(W
z6!qy@nV~*1?!OVL_8gFc>by;+xP&-X$}%UbJQK8sGSOy;jx;HvI}b~AV<^5NDlX1U
z{Dt7i$ih(=`P~nqSO+=V1yPdbV-_mC)S8b1|0NN)-o7Nw$1Gg^pq+NZp5k2X=J3Rm
z8=&f$PdwplF(i0UKFbEzmeNI&q}zl>B|q8-Yp^%nh0|~LAXJc>HY*A2OE1z5^wcKa
z(c#k`Y9_YyU$a)I0JWwwZAbBr%(Ria-=LW``CH^}&a}yNJDzFVE@EoA+_Vdb7%D!y
zK+=7w`z#m95u(*=ffKXP<gL9xra7}lSQK#&?*cphEn2nzwvhYG>xcdTvO;CmLEPjK
z{fG2abc48zN}3_Z59+DthUpj;Wkg>$nbgs${ZDE6ltYmDo&5R)X<*e^X@`fY6jdOW
ztCKj=_~ryE$(dp!jR~eV-y@9*93_u53a(=#jWv)(o{`3%LcaDRjS6q=MjGYB@sY-N
zf#0qJQN&QdfPAMA@dEOC@Wp3y58jV=#DkN&`S##16&_Fe<UOYYzdp}W8HsGi{(D<;
zZAe$9fCszk7Rn?pd0CxUBKO`c-U{cv9$I^jocFr0DB`>y`t5SwSBKnZcHUP&)|AeB
zm!68wc{3_$ocAF;6`k{DRJ!B5EiIq2<C<URZITAg;Jm36Gdl0F1S+x&gvXCL?*!B9
zmh*1)d5esi$-Y}K9rNAif++I%?gt@X3*X%W3AyXOq{SqmRE@ajeh57G`9QKG&&><S
z>$yLKbtbD^Ka6+8bCbK@0MAYR8eAJNLSC*qR?+W<=T0EovFF}ZVEcipwV|ZidT#vk
zexF;lk~-j^nTZV3+0GVkg$MsUQ$VgiDokbSet#}3ig@rxe!D#Qw?ghSd+@)5tSLSC
zKkKRJJUFA0#)JP#Petd!8I|sM@MWuX;iK%m=GTJ{0H4tjX>IqLN-?7c|2k14)c|D~
zXgqj==_#M@h#fXltd6^SPFiIySI1qEApc~^-~tQwrA~rO<}0d60_fzZMlc<VYW@<`
zBTrOw7YMI)RFlM7tEeU+Z9J-Z1frVHF#G{fjE_giE1#~}f846q%O%V3+!?<93CMuM
z_h3zkZyUvi=esjTd8Sse@T7UNq{L$jeAWc{3lJ`_l3L&8Iod~|o@JI3Ceg*3hy6Pn
zb(qSBH7qL6<AYm`1FlCF_rQEREJCTd6}rYt8h{^#lX2>9%_?|Bvw}-w!eQJwFIPz5
zHMu14UPS`V5oroB*a0irDrV6#Diw3aG_09Pt5kyO@K$xJC9CRL(u?REEFFUoV!b?j
ziF_C%PgWtEAeV6755jp1<czuv8MY9tq}5=UMQGHSnqqten94lG$k}2OgC9>hG01U%
z2bZctY(Fvh7r-m+oO~SbNauvy&3EVIWn$Qjf+DQaEWq|3uy(IHowJX!JJ0B1oO;Jy
zl__oAD8Y6;mMySkRr)Az95&)8?6~dl{)zoJ?b=&7e8+)ZiKCHgb)>m9>k|pP{W7S`
z9F0wWF)WJoY(5$7+1xT$Qehpuu77y<gRp{`J3miD)|8!}pXsUSIzNm`n$FL{b9C*l
z{oH|3=}zZowU$rWPt0%UXC?3%^<vu250#>d^>TGYKL#Bj`^w>4RE-qPZwn&KT$J{T
zoHEXqpGBSulz54@6}0tyF%e9vS@)auRCJopsHD;S+w@d)-F`+zk>tu4Xj2G|Z;_Xf
zQb>4O%ayDcne;q?X0BZzili0QUU!OS&!ygJGQ~#c6P!=C`mvi@3~MJu9m(BU!F#Mb
z``2LbdAhUjg!XFDoz+JMyjOc^>Pp2sx1WR%{VC?f5<9o{h4(f`?Dep$?qc`{tL<#D
z#Li*M<#+*kqvDUC1jvy6pW_{gipkx4N5uz|!nbmjt!&R37T?@iSR9h?I4(Y=uiadQ
zfLrh*8Q4d<XjLru1kA0E8<VbE*<yqTAt#2;>$KeQ*ttzA0oq>x&6=a&`N^;-5<Gu8
z8azAQliYlT*WU?6kU6~md&rtHy#A@4iY~}tRMG?)&*`b?f(%AQ2{O*u(!Qb9It`Yg
zIaAo7<2P6Qu>|<c6kby)rU@ma@cJ86jk+FQU!iRUZJaem1d}Sf-l3<W(|krHjppyy
zQ_*QYqoQbjLIB%hEo(wbfz+m!tK52?Kr@$m-$hz6LwHR^H3#awI?<0Nbs2{uDY>_d
zJr`2HUn_O3Uca75B$ZyD)>F~xHKUS7uivAmqSI?eMbYa;7_Lwe#(RqNrI6!CwX}0<
zFC~*pVLwEgCxgOL8O(vg4ulaNk)ESSHS!WZ+LWFMd463hPpw{lHKF2i10)XHP`{F<
z^09txLR{Fbe(mPAT|m)oNbcAQ=3^b(&w#7V)3JRH<lmxW+a_%1J=-nms}=9sej2*A
zpJlOK?yfB_AaB=p^%~x_U4wU|YfJ9tyKDQ3r0&)6N@WJtrFGW7-P>9JR^_4NPVQS%
z)H7{=jj%RdYZPXz8Mi)LX!zwy`88z5d)%0+xTe3wAZdJ<3WoVJ*aF3@RIEzDE5F7X
z$0_o}p6*5MS>z5pO;UlKW1w<#)aSi4EQ<7bpUB<k?OcXL79;A8cHP*^aaSmU%zfS+
zkTqqW_mG~7E<$2d(nLt6o{BC)VpJ4guIod#=mO}n2o3hMe9F#ze*3&GY2XYo7L{U}
zFhlC|ZX;Aw&A+I<K{10-T5MR8jf0IgjcB6>wQZ!0&`#3|rTNVBy?QD-yI@q(*u_Wn
zRCIR1s3=nIB~F3LGv0MzZ-pd3r=^}-+Zml)orfn#3uRDjDuy{w><jdsG)bw?Yng?s
zv?hgUh97C0L95W;BjQTc6@FGvMW@h=N*aakeSt0n(5@X|R1}3?2r*16btdnG6knjF
zoLiX*m0YU43K+^XE=dKDOO?}PiuE58VytfUAII8uw1{biU4v$FCsNcP)`@%}sC%AH
z<gL(ZEjp2Df??j1yewk{<K4+8pgTFjY&Lgyk{6J-J87ac$V}-Z-jVJkxts6qWNiIl
z&GmdZW7I8Lo|ki?w1f47GWCwTkQYSzXL9MFC>Y_6s2lIcChp}d%?I{ecNY1>*7E36
z3rM{i+I)`sb0@;0NPq6L(f-^5R<D-R|LWk`X*l%}>?5JnZXuX{WrrgEWb6{W$FQpo
z!pehP52KA<gj)(d9xRmss2S*+oTHY0Wrtj;I9`1g+QVn9oExur0eSK2FHiymuRe}<
zbiFg)%{yN4L*?1GhPd20^loP!uXNNq#;fhR(noM=2%R}yg<)(ncAFK?+G1?pd<?cQ
zo%WLPOjjfiuGeZ42^jk_(@L&XYVRV`7n)xTiy|=gWE958(i2^68F(wx$ZL{C#F8rU
za6eDVWgs4i`u-qfGV{RE(~uSFl%|9HGd&eue}Yj-)1O#)uFeH%`xA_cBEL*iYpb=i
zbN47Zm|9y29A)ZKWRUcjwzg>`oo18!ZF(xYP3|+)WsjbUZdip;Q8u~nC6L8Uj{Oq$
zc88u?j&1HK6KOhi(rDecsO>7y($7(rfr-*Y3pQ`&b;4Uwp}M|8x<;x(Rsayi==vzo
zb6WhTpm(31p6<AHN-vkB(qxMDnG+nq{6w5u-<KS53f^N8=U;>F<cT=n2QAks;?#1O
z9C7Nh#3RlpA>w?Bd7<19Cog~p2dY-I|HAImC;<|2K8|-J;v{$T9dRBiS$c6ABdI4&
z4SG!Rre<(U>O7-*`W;7|SLw@d5qBm=n-}Pd;=t7|nE<rE)dE_ki1UAkMUjZ}sc6Kh
z3m(;(8(qo~6|qHB$wmys15wGJgq&p#Gye&)rVKOxT~9?9W-=;i!py<*a)+6WisJDy
zg_&!$wC5wtJQq016lP|S^i|s0YQs!MMTud0wsGr(JE5C5>8a=<m<)9}sHdU}GZ~dE
zVdhaiwH#q4WkU6Mvc9QfZtOe5?-Scq`_RdmV!=}amUSz5Qb)Gh;C*uVB)E@-PhSF=
z<O!eN3{Bc1d}`-xcmTCk9_MT0bK@b@QxHOZnIUXq2qo_|GImoKQ5e3>HybglB{+?y
zVwtcR1wMu2XWT{|hT+34Ml$CUcKnB0otM=PIzr(d_+Nxton|@Zokeax7Y_E>xBu|2
zaf59!QkmUiynOH838Um%UUj2y*yeG|Fsp_;1*h-eF+L~ESpyD^D!EX(vbP1^m9UzF
z>fR)9j1HcnR5Yu21P>PnEXFcm`IZogF*$31a13|Sv+BpmLax!l!*;_Xg|jHjD0_wj
zo3<FHL3RMK-Ac)-lVhx=8n6|;71<YM)Hq~W<AX<ThTUEolaSkMSa*%szF+gk$Hu0~
z(@{7;Xbi`tw!#z%ZUaYtjFGh7*kvzy(KReEf@QHk(n#pL_r3FXNJf&w_|FkVgM(!T
z5|5O5QC7in|CBwWFoQpfaMv1r33N5;<)-st$b+K;k73?n2P;U9s_EIJsdj5$AluiM
z6#F_;`>Z?(9B1BVg_}LwXXWXX`>b#b%-ak62jG?V0>6cKq!&o;=DQcTbK3F?Gv>)c
z#i~yGcFviGYtr}nRC*ov@>Zzjv01{%?B{Yd5x2PLV_{D}hen#CiOqiviz5BA???M*
z`6f0KO6^vpG^_*^5y>V;MuOW|Wb{)IPoBtV?fESuqiD18NNAmg@p$C(eZa71SlhLY
ze6FHrF&SV;27Xvv^UcW$_~B};;TzPl@BlD87SDsw0Ro0&up&DQfbwP+3VSJtqqV}e
zii3UxbsS+ip^j(h$eot}h0Hk3*<v{TtCTp+TP!b5A3(JtqjfLCJA%{XZoYB)Ml>{{
zHn4QRo)f08&IZ#VxsFkKD8fBK;0Q=BQ_^sZjyx3_?=ZdPEv<$;nTF7&!lDR7KR-7R
zolt7GAo~0m(ut-_Mrgro453YsP9B7QD;iMpVv|Uta)4gjn(a6~KM&YdWDVCApD$wg
zjQ!706a<KQH#NlKG)aXPJ`}1>(imq8+@0v@0p{N&m_I|n`$3?P8F)Ec6nH0A^`t~z
z-WJ>*)t*B8d6LfnvlMy%3hxN=lDqjv-ra~xlV%Y-c8M;|$%($#W<y_vWXA}+Dxnrh
zD7-S3jw5lx8(|0kmnk!E{SMIhZ^NPxjVIQC0;-3M<%^h;itQF2Ul0Q}wQ-X%S#TV~
z<gbE^@?i2=!!0p6p?Ntj!`_k;c+~ks`DD2|8dH!MHba{MzI~0gV_R@OPXcEq7kYMR
zh{>81LIsDQZka>iiSv-vr7SD3dWN8Q9Z<*&#hfh$#h0dqVlH`Zj9H=hO4Je(1iu*X
z2o#gMxqxDonTUtv8+h%<<-5|swQ45jOnS-r-Azj_!?^}lw>me571$Y8fZXM+z$f59
z#<(<tA^1=iHqzb-=hng(w!YUSHi`W3S9r=he<k)Q!=pg$rL=-1<Q$fr?1+6`u$KCY
zt>jB!Z>v2;@3zjM2||_G-e)}BC`~sCd%}{t0!?R1d!rGW3niKhqBD519D_bONZ+)#
z1n0oEyH$^Fq>U@+r(p!&Yr-yUL9g%nW~I63tYDC*u}bi>frEiP!SW*Ar@O$)+7~S4
zPhjJ1tGO@ehm+`i+_btGtSq@j{C?RRovIU=mEfC>mz$M_<%K792YqhChc0z<pFOlM
z7?`%IbOSFaX`p2Kmgk!@HTbcl?z*_x-<m3)Yz94J&0q;-p#XG`K>|499;YA*jd}%s
zE%V(X@$mT8YuaxO&Ypq-V5HuX9=|16=E1-?9u9sIR_D$%-x4hM{Cc@ofU$GjINgJy
zhUpWS?jTPJxWjsZ=qXeKD6tH9@H~026!zS9>pqOW9k)P`3xYx0tb{*%gO%3FvR{CD
z`3;XJ>J55J)?{NE%3BUAWN|}eNKv=U8F(@P4N<fZMv*@<08xNb&t2RaycrBM{Hd+i
zZ1u`;zgB%&fNkPo&9M#h7qBfF$R1fO`k={F=29uWq?Bf`F7hT)gmWJlU$C5(D|q0!
z*QzffnS&Kr2hcjtX4{fCgJl)7I^8g*Es%K0Y6c6eYBN|>w~B7PR47?B2+)8`nFy92
z9t{#Hm$)j}eZg|G;kyO#C|FPiMO#*C%+v~@GJ;^q1G0yF6_z7|V28*PqUPB8;!(}B
z8nC7O$r)6o71l{u1&?Rxz*rB<NwlQs7mk-<YkIeiWfe+g4@ZcgB6wOhJX*|+1PpE|
zRK^3dAoD-)ZW&8fm@31r{Y^lZHFyU^V3x=)yH#i#EUg4hzNBuR2!HPB309(|vB6!?
z{+Y7hoLI3jSYESgg>e1?E3*`zOu{4`F^b?^K8*p7z0umNZ@bk|w>}NFv>H<q2Ecp@
zz6G1`gY7kget5oh!;7v2c`mO{6(hyt6H)vNdGTPjCaUZ)t3Fz`{3${MpkU1QXDUE!
z07lZlib}rS+|wVd1D%{HgvJ?Gpy1U&>zk*772YhIU~sZvm#6It`~&@6RB=)DmwGcM
z7}~2FVAUQ7Y8ClSe-<h(oCQEdyRK-%E(t|wK5B0uWg<ezeXLOf63c-*zwCqjDGZzf
z?Qx5+Rv&bL;lkMn6kcO8lB_&~x(nMfP-9#~zz9?Zx9$fEo&^qU%kqK7>LR2%R<LHw
za;0FFN{G0G>y=nCwzy|irYM@73Km+`<IR1I$)KO2379WvV!2u>gKfa55v-!O0G&W@
zRsvI?Sv~+Qlu@##5XXbFm8Vs621vnUC~YeI(T8AKBF`7&LwY~r97YHGGQ2DRAL;xW
zCQzM^;oV2^vKL?6ikCxpYUcsCG=c%l@6Xl<6<We6lKT|Nd5YvZMRJ^K*z3X35CY&Z
zuY*~)K3bj7C)mC0H1%*_(Cb1k%-vtmTlT;&Hs6BtUH=o<VWWzRxEqGR4>`BDQ?`4N
zcfvM0e&_3W`8HnmVRUf_FOTBo19<s4UY^6t1Vk;)LA;o7c)2r$m#6XaL%eJqfXn50
zIf0jZ@bU}1Jdc;V2jOxbUOtDHr|>eg1TN>`<qo{O3NMf1<pX%xuoNzv@$!DWJcgG|
z%iuDKm$Bt=xe_lw!^<!6g5oDftUGx2s)J{<I(U+sgD1T?c>0Nhr=vKyajb(I%R0Ez
zj)Ob#IQY4wgP&76`0)|>u*$(jp$;w@b#Q%ygX<q09M5-fyx+lbBL~No9PF()*qdnt
z%VgUn+O(`#rbmt-UeG59(L5bhTloo12g(e#J~Qc7LO38v&xt$?Qx7O54sfa@)7eV-
z;BD)E6R>l2v@G@p!X0oRMYn+<53xnjg-un6SwbAMcqe;)8f@tP=0N~M*mebA6|Bkw
z`$vK!pvTL-Sr4$2KaT6OIGE!Fw^hNd00pZAvk2hhsKCPRQOozDbM!B$^in{r_Rr<W
zkIv%`J3oaS+6b|i#ViYhMQF^;pm*27gZrVrQs{^fXEpcSi+1<V$ilF5D;#1XMWOeA
z>HP%WY<fLLl%X+;%3pgUfb%hf0JQ-go}O<4k%wU#lMoObGmyr$MnlM`0U;4eDHKpg
zWe6cIU&~ofH`q7qEP))Ugt&kUaq%m59Nc&7zMIi+5!t;~l4naX0{H8tc>PTP-&J}t
zPNfWMM{-KC<8p*FH)odS#;nqGE`c0ekZ(vM$UU!bC&`w^##X|%d@pLgD|5|nOj~nq
zj7~q+JZ_2amsBWn<jnh^4;q{u2DluU2tY0gACuT(Nx53AG)frDOk-FvSi%^bg{x@X
Ijmgpf5BMAgWdHyG

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/hf_text_regression_model/index.doctree b/.doctrees/autoapi/lmflow/models/hf_text_regression_model/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..316f5651b4ff7853d8f4012b67309a2399d611bd
GIT binary patch
literal 73060
zcmdsg3zQtybs+l8jArx~NCt$U`D!$t8QFjh9)oNU2#Ao7`PhIgH#OZgQ(fxmZn~-^
z4K^FhE*QnN;0=wlWc}e|631{(?1Y>UV+T8|V|&+Ooy8~K_{8?XF<FB#j(2_H__Oie
z`#$yFtFFHNs(K`GtaH@UT~+Vh@4NTD`|f*n$G`)Nj$3pb{)<)z4X;u=H0hMfwR+hN
z+R?Idz0{g>YhnBT_LldzZ*PxA!%gR4P;WI$ZaX>-o+x>hYPspw+WXqkDttVwRD(&;
zekgEDVWnP+n|lM^pf|LyJ?afd17Rhsy4tPioa)?6wSI7N4xmFP-b^uc4~501JKJ=F
z0QxDiJ8kR%I#_x=+=cem?e1!G8U`B<U>vr+wb5`fCed3G4GPM3M@y@fn%mypns%ZQ
zCk&gF=~f6FK@Kdblmn!8s8n@=04~D!ftoYNzbz}b8tTWiQ`+BjXYiR;G!#0sc&qKL
z_BMJeyi>d--lnL$)vM3BTbr#~&24TySa0s%diAayzy|pT5Dd0*#I`!Eu<kS}TREe)
zaz<_K#;C28TG>4`*$jZxQ+F<IO?$^h1MHf{buJL^G$7489r$`C{68E1Z-)O{kWAoI
zv%-iQjg~f?O0ylIgj!Yitp!VgMH9Tigyr+z;O?;Pw7l1Q6N*#2y;DIR-g2&pN|V;e
z3(?3WN>WTmHW-(L_@T#3IPZCnPFLCy0Tk6X{<t>|TwKPuxF&R_TUDo7EY+*Upy8G>
zFmx$<K3enc8{hlJ+uwWc?ecFM!Cj-kMXjTu8B=dOf9fJstDrSE=QQWHfvNzTqYc6s
zz!00%ZnvYg+OHz8!5qRXM&MtyK0E6+M@Hm89Y{VVbZ$5r3}>75R-=P^Xdi}vw;5Tx
z44NFQltZt*a|F3r18(jIfz_po^ip=MgPZjZ&{|E`%kv$G0-j5n^@APPnSsNRliFk$
z43IP1bQ+%4FuG)_F+WwG_NQ*)3@lt^H(Vz9<J)d|H)HrHjPex?w^pvyW{Zcw`sgs;
z-Y*O()1hi5bem39>)iXO7<7Xip|%=u4|gAMU{n~A(c;E@I~r`vPXG(ru5N=+5+4Uq
z{ueOET?cD42&96oY)9wkv65}@Y}kIzsnpuu6VVX-3g;W(<(42m(Q^3h9%@u8rApYo
zhuuEtG;8g{(J)Ajn>Tw;6fJ{B%zw0ZGr*zKoOQ!?WN9IV$eGD_5S+*{GZi}*fm|Md
zA-2IARN{U>7q{D0WGL7By-nV;U{|r94`DM(#@OIqc<_S&@V>;D76FR%3q62t^q#i_
zM5BKgu(`WePyYcP0yEUTr)VkKx;D6IAg}^Y&m_By)@x?BP_5TyxlMJF<D?9&Z%MY+
zABu(#%~hv@hJz8@vIay}Zu5Z4H0eFiR{FYkGdG}6Ho_oBMhS_lB4$7;MxZYd;>3QF
zaNR@OMKC2yNc=liDt{e>sk&ZuHw6eE{I<V3ZoyDk16s2up*7vY9`9yvFLnd|SnK8}
zjk9ylO>@a!QQ{t5iC*@~9T@*!8wba`FlrtdxvEwPD^9g?n;R6ILh3_>N-YSTTFIT{
z*4&jH3-{=ozejWVPw9aDH-Z%U^ue-^yG+Eq-!XrWb~g%9F?M|EEV1>qsjZdk{e6j1
z^qwONkW20%3C@2P>N?JgHbm=Faa7*ezbPIUX%X<K_jqq#?BBqo+I<Yh?$~=u_pE*i
z+Sxl-?=QV)yf5tbz8Li-pQG$2Q^qg)H~$@sXNId}*L16<EUvV%awdl62)w1=Lg_Jw
z>F>s`c0SYJ$$@uuo6$rL%dsJAD`n2_5x9Gs(v(p5Sd#E9itxKK6HfCL+_--wV^DXE
zMl`^a3tQB^zoawwC(}CqWt!*q$I>3vX5;<<Jm3%FQ-BTlYt1?NoIfymQ@E)lg>u<(
z-onu}E9tM%SrYzwHFz7y02H>@YavW?0&b2rL><?a!2cSs)-|2xH2Uvn;}*<+p9m;T
zN`u#HCTFBA1_PK2!9+EBi<nuBR%!(vvs<Pxwqy>?jU0wXib!J|7cGSpTW8(?XK$<U
zPBExJwkT{hjTD)jTA`_D6Fv4QX6!JrB(rqhK(yI%?kELk{k#(BKgZ%p8^NQ|a-d)W
z0!)Tlv=T98Qy%u)nBW;|?Z1lsZuDP{yJylr8Ki$)G|cEt5bM8!k8<ZH8Klr$yqMCv
z(P$Z>$#GnUz4t=zIP5PCz36x?XL6w89xOpb){a)ouSE#Cyn0iUQZFUODXE+pZpGv$
z47tEE`3Ff!Xk96{8Fw^y43@)Rm!yG9P#F&Y1z;(&8IbS<2DPIRWD`t<A){-~&Z9w&
zSlfFh8sKwUZ!lUm)2db(D>?j;VpuN~A@6~<;u&`BZkThktLibt?FT5#O<d-{=NE+<
zF!8To;xBQdW6wzJb}J`Yb|>AAA{qVdD35%WmYvW^QYmr+{>9KyGz6Sr=>dNm{$z@y
z|E|?1c%t=iG*qk0udC$c37#4{9IX&H<?prdgdi`7A6i|j%WLvE><QroZWw}XYad1u
z|530?noi+a^BGP5yU<&7MQSWr7-e#Xd4#?B>@R?$t*QTgz}55&qy9_Uwdj#lIz~Zr
z-D(kXFTs?g1b<BK!t0kbXW8OMFkT2-HMbaqO(%3`=cfwrt#EN+bFJQ-bE=!i3*`V3
z!L^yn?8Uq5H5Y!n2P!4^;?1*-7Tj$GZqX@~+$!YD!g}-K9ZofXhsF>R{2AL1iKN+J
z3$w&%(^oSkcWSkI=p<H8qI36NbS5r}OcgdW5p3QzUbu)o55p5O(J-r@D#VWzHgDSo
zQv&8IHps!LEMp0W1Y3oOol@+2cwgEOixw56uYhl=let7Erdy8|9c(8f9e+PI+2|v9
zJy=Nwyrx%hc%ywWIA$SEX-)e-)&*_t$ArJIeqK?0XN*Fs-}8S7`1rqqkKR5mg_=ds
zyHw?FQ#pZq&xziLdSJQ&?)xj@gTPd{yG~3&Q`a%g!({fH3)$dm1y-80803Smq2U#e
zF0SD*&N{y5*x?&@#X=g4(~8W4X_Swsp5&m>FDYGmf|O9SsMa}pK7~y-`hNg_m`^+*
z9{!D1trWI8Y{sG~i35C7pg-RObfXjW(>iFs3cq4a?P49iMgs8<!R>D5ffMLIu0uIl
zM71eKn$sBPUqooNA)=ZnW_*XlQ=g6M*$f^n#7hPK81!PK^C{qLNdXtQDl?UmxsCvl
zl(s+suHAbOB>pYvJ(=0{{~RuMaL`Z@Gs4kHF%9e|kOdq4hZq$9G58Y;cN6LV_;@e{
zwbE!I^)GQ%v46=D(-WCage3CU9N*9B_;%6wwKSBkKo=dX6e=&E*r@zzG!#nZ1r$T&
zmq4)5O}E)S#l<(0kafxHO6<P^bgPtHrxG(LnYDESkjbR28CblNB$Tu&918fZwXxZ(
zH+W#CyMjd*OPz&pio<(Jz|o?5r_H<xn{4zmQ0J3CY^Z~U^-7IZuxLJMnT%H@W$J<=
z@HP@QPU2^rRy8ahaH=iWBo?N>twy8X42uC25`?$5>R2(uCMJh;9xQ-8T4e^@jG$4|
z01d~FNOH9+irzL!FKKu_1Wh6Oyczz)hO&#=JVHa6f!iI1N9c`!GOt5fin9z!7XP6r
z`ZNvet|YA9V&f<*fY@O6uI5E)LL1}{NYdm_Mch|u;wWeMC1`3Xu5Z#%C~*lWhPVdU
zJWvuPekqz>q@m>y7)N3O1hzhP>Y66G)rlu1wH?<L-Ty|@P08{X_Oe{Dl#DDWSqdno
zEZO8;Qa8YVDVk2Eq2-VzM`8hFIl%G_LUyLn8k6)a8q<8;c{D8)=e9yq7;8oxvlL@?
z35fqte7l5(Md<>EwEzrSCNnW<JX$;!GYP*c%x|JGr?_-O3YP|XGDz}C{#5X0X?PT;
zIJ{h(vS}2ZGzXDErHF9{CVF{J0DeMJfJFfxgE$i+q0UT#f4mmFADx`|?_$g=E5^L1
zFncO;;tIWvocN8<i08!pv*4?LHhh?wlPz!svfDTh^alOc<3k9^ALrNc``vASQeH=}
zTDtob;LgqXBwNz;-vD1_WWl*I>cv$~(Pu7op!j&I5SJO43qGlp(~~I`Be#rq>jKJ(
zc^YG`a;Q?9ymH4b%!=;9AKe>OU_G!o18K-$5~hGy9y)nF|9cA`S5TlZBv99e(Sr%K
z_Tq`WUKVMThuc%gN7F-+m;R#&j4u~mkdtc~>W?K*H|MT))f5jyAIZs*F<(=vYE0e!
z7}}|+TUx0itUtgq5HD5a)f7x4p`s!;={z67S7=TJl||V>CTc^pcwA2m6CypYi_%gW
zmlWcck?mZFTTCGmy=Y?lDz?{>-+9FLB}Tc~(nf60L%Y0U%RsD!poI8>X>Gur+eJFP
zhPJ0h*bkL(Mor;#??jW$j}|~SFXfcY&-##!J+0}Y>F#Fnc`TL1EeJT~J}fu{<4(DR
zqg=tWdVm^1`Z&<BKf}a+I2xI$xYaU`6I8lFgaJ@^&u(kE8R-h+mL19WJu>)+<INH6
zl8U9aWFf%cj8AX$&%!@k5NP#cEE3xTZp1>ju`#s{Fg7+)fPc(~79bYpON)h|Rq_f>
zfYm5&wSc6I7aC6JjTf3QS;kFI;{`V?O^%F=jTt%Bv9SVQh^#jYHMZ&q%Z@BLbMYB`
zVe*Vg2JsSHu7ovW=#W+W)^NeN5PEI_SGi#Uxq!QrV86g*;WFMqPzdXVGrZR9EIf(6
z9C`xqaYjXua+&KkAb;(a3w-gDuRH|oSq%0WhSxN33Dc=^T=vuhs7}S-iUsU>9`>fR
z*y+y0o~W|%I99*Pf7~q$<D%(I1xsIHmsnyQN+}F1l^72dVGqNcrdrXJ35n&FmWXv=
zlNy9W28dNA2#Nn72nnjXHDvs&QPa4GOJ6+NlSzMfbN&4-_(O#nFjoC5@lmehn?dC-
zV@*)i75_9u#ZXmt=)$~g<ILC)<?lcZZVzeb>fgIrr%@o+wct=maR{4m4w2^Dzf?6r
zroN3q@wdYt8JI@5<r(7(GrqaP`d8`btnENYJavu5QxFe!UGDiEXm2eWUnuU~&<Mj!
zZKH{g*Yu8P3216P1`@cFcL-|DN-1C&&ee;vqSLU<gRrhrThB~??t6E9l-ptWGN}bO
zk0x;L9zNnwd>EhRxZ;R}tV$WKNR8<jdY3PVHb}~@%GulZ?)Zr2JreXknT7s1=y^Rb
z;WN;(m+8_5CN|+Rrqx0ln5Y(M19P+K{knR-v{AIuXv?1P^#SC60~x~YeCis--(_K#
z94-ITf>+vjfPJ?9bF2xiUnh1WU8+p?2b5`kB@02Z2228qFIiBfX#h6iR9R~PG9pa_
zUO&e2r-=w9J)TL^0?01%m`L2zH6Su?g5^x+M^_D+c8<0g!eEX-qol#aYq<>O8-Oo^
ztV9115P~#7(Qq8#Ne$4hD3PC?fJ+k~j2^<XF?#D$#USzNa#!K@<}A3y>i9-r$a#z*
zG%cDsViV3oYjs4nr>WxwGFhEaNULyp)O7O@+RN}eO*xadK|MRu(l=(|p|nxgGxnJt
z9k1n5(DQ&(dsv6L6ttdzOAA#fD8jQ*&~=?+k=mJdOR44^S+I&#^L${%?Ti&P6`E>d
z6V5$r)kKDK)ufwTjMiRFCf;IU!Y{xke5y?qh+=$nhU)!nXmT3<Wa)F7^i3RsOrywZ
zaDME-oN?p{Y^K<F`ssy6vkpn^0A{<>^MzV#t}zdr%?qcWE(J+G!zBv%Jqcn7h+XVv
zN!FjB_*Jq;aPe|5>bds8@RII5VI5}sus?KkOCgc4>UNL_<o;7u4hm58>D9}yHRf_N
z$Cs^}Nr_L}Eb>murgxF|yJb(KBC`<}b&KD@r1&y<8BcwT7hvlQJHP|B%)(v-U`%lF
z_03k-@@(HI&s`<~N+nPi)ENFIIfh656}TE?SA_d~+!ad%?tt0A$F)VG=-P#+TX+Pc
z9MGD*qOq|KJS1?GEQ^0Dj{PVmGp)dyEXsKangUmBssVpVL!qhx1(d4J`z{5r!$aI4
zNeg3|C9;@HKry&?niy$<i5&{fxf(B?iTUA%sAoIH{f#8Xdg8=bih2c@q1H4ys|b5y
zU|V6x)~_(5j5V>Z17wp^LJI}y?y*b*ENWt3<XUA6g4cDi9BDeguJc)`{%N<AEZ#=S
zf+X~<j5Ai{X}f4BR1JfGV${eiMFT47<3AKddudoXDr7hc3n0<V6*3`B3&pnvmyR*E
z$qJb}X;@U<2Zyx)49Zs_^ZPXB6qi1n!X-KP5E&EsQ)$>IX?PT;IJ{h(vT5YMA6D)3
zQkWEfe$-20qEl0YI~A3sm&qv9JIZ9918ZqrCd2DvSf$K!AUIJe({1BZqynrD@H(0(
z{fLto{n0IW_=8?$vjg15*M;wgP=T(pIr(}KEVBhO+|w#o&BWO$u)&0QL}}%>Nv+hq
zD!Mkg)O!`+`Yq8Ca6J`fEi&^WR<-;yQ?89_*CKuGQ!Xc^E{Fak3yN|%^i^yWFNgXU
z>X7e)%qNetS(YLhx?INZ@s+{Jqp)nqe3HLSLs2quyLK%avqT$E7S7&SHCe7T7x#^1
zU&nn&sH6nJtCmrLV?tqMMr8yVb)@_f9d)-<I8d~l@NQ7PLDr#}Q8}@jAf?am{(u?}
zXD6^S<KiTFG9ieKx~+ove+q-z935Q`0W+gKG0i@Y(MlttrA+W>D9#?5quTNUa&!sj
zsG)JkHo1IVCQ*pw@->VMGk1sAa^>#c0esoX`I0BO>?h#TA{4rW@Ve@fR#ej^eJm_#
zm%B>O-<t)u*zR8o3^~LYLerwDBR1hYv{pxCdzw0qtAsGrn8XSGk<Kh8RAHo<z+}pJ
z)L8I}lDAVc${)?bIBEYLWIQvq60hab%IkqW53mkR`=>2%$Yr~jN7_2?lL@%AB!yNY
zJR7Y{OB)Mh(rzhz{8ARIVtu>;Sn*ek6*Lu^K4KHjJ!^eLhI4(Sn`jwjw5Y#^5G0uc
zNIDRaQO0N`Lq_>(=>3NxDuVFqOf5=~k%JLx4PEQPfjzaj>@5DR`2W)^{Fg!2PlQ%L
z@gJaX23en$ImmKRCWTnp*Y!}~72lFU>v-W{1#-5nu;A3@3su*_Lt?SU0gkq)K;iyu
zfo({3s;mpvJSdFvGKT6r#=!GXiRBh9b)htD;+mO>>)=2V7fuqyJ}$47^-Ee8TOhcL
z|1pD+5`~$iFrSEVS7h}`QlI4T=AEP@R@2sXA}-^iv|CCuRt}qXiXLfNoVFLVbQl_g
zW{|ZZcfU6HC%N%jK}Sj8)GQK+ZR5Lv{TmtkX^CmJ5u0#{(S}Y79Z|@G7Em<Xh)ot-
zG!eEDjZ6mHcoX!lY$HvFSZ$LtS(ytrBK8GbWZMHoXI^Ae$g#B*^1NQ0D2el=WF|q<
zWHLgWCpE}d%FGj`{7Z4+XBKevx8pYtzg-r8yBxk9_II$)E7<3i?DH!2`CIIBCw?L_
z+R0CB*=zw$etJixHsiwKZKVZVC*0B3I$?}Z-wTDGh`WM=02{3kj$vxT@k~$*U4$c>
zrV7*bdKC_M5(hQm*46{n>Rj@=jFPt$CbkuBs6plN-m77=+uoUK-3j->p;1kz7R<mI
zQEoHf@3a61TLkBe5GSv5Z*3uWx8uRbtoHC4cIRE-p!YaID9&v%7aWIU#HAJ0t|&wH
zeo!Sn<TVXbrX|PQ<F44SsDYC)mK?V)00S8<qvnw#aDV}_iQO_clInMG<{%za)c;<L
za^~g3`=KcnN}72Cu{v}gYh<E-QdfNv#2T4@I9xu~J}Fq^nj-Fx^Aq<t$J1tQ^z$?n
zYHd_Nu~{4a?=%$Zq!a<g;Nnsoww!&P3!wa_;^VhT*!uJ#(<&zilWQsS@0obap-Qgs
zW#IJBNNUMdXaU9CF`j-p%D@O^c6@53Cj%|V(@-q5##pTsd-%oVxnoEF=`^$)>y?~T
zGfvR;&DzFnNOw$W24<OjR9Y8xK@Ev1npA4x?9I>=JcEsCxsrxLnHB-X#<aYXhC(Tc
zfRf*|cqD9H)JMy-l$m%kn3fEjzLTVuG%W&(jcNH14TUl-8EE+!4aGuhvY3`9XlOZ1
z3nz5}OiR9%{x8u4Q9}A6Gz9}<3_7{AFHE!isSMWhG(5^+ad^2nWz(o*rN1+bh9k)M
zX~s>FGgNzDqic`UY(nkwjw{2K;+#cGTmN5@srjuJ1CuK-`Z3oH)4A-Go@~)x4f&zv
zWCTJv7Y-$bK?7_)rvX_{0?ExG0cYV@#AlE&MOw5^!+r~iz2BQfR_A_RKtrJ{X{xy2
z7mNFS8ZDR8P!@=m>u4ywX-QG4ZT%$&G#)r4cDc2&Uj+M5V;cjDNv|KFrGlLf@!TfP
zP#pats?*Yos8qTpy^2cV-cdz$2*}D)MfCyDhh9}wTGB|CQJvm{fX+Iq8u0!$!FzDs
zbRE^`9yk>UQelNz7hZG9jwqHpDhl-(R$RqSK!~fV>>H?f;KSBK0n{hX-`@&!+%H-J
z9Y?mt>Qh-zl*@mAh>g-~tZw`cJ7$9Cz{I%>xXcrX>>?gM4~Hk<tH59b%UmNYfD{<J
zH9^3Y_yh{x6CMBoi+%i53SOY~F3+t&Z@5N?$xt}^U!;d2(dku!LIBH(RoEQk0w{6d
z1om6N!x&J+{2H7Nfu~?-o$<v*vO8E|4C+l*Af_eCSWOwGe8vlPama-18&C%#>oTd!
zNV#Iq4e*=4N4Mb2z_AQEOxV){XB*(rrabG*fXPX-1tfU1@Xo8R-o@Cfu_m=&x!cYt
zu0XbQFR(~iFC8@(O#fkQEY^aB*8ojm{YhWzD1+ohPQ$3PgDMh9FuQHknD}49F~c`i
zcV>vcoM{w_il=9x5tSQ8m!z%A9sslQ3~NHK%6|PfO;|Eg`^_Uvd75u$QJ~n<90Zm4
zmW38+E5+D^Yf<;82o+CTDMrQn99O*UTER->jg9rNo-j5S!g8h!*3}8%;u)hbHuV{p
zzI0g$`E?c<$O5rnGRBw%Vt7qoEu%TD4e|E^r+>ydZImspGQE-<5Xs9*oi?m6F)%ZA
zdJK=U7Nuec*``!%$HItJQn66oDe5<8p*|i=?*rza1uc7-WNk2E6D}`W#M)p&5vL8N
z>&C{Y=s;t+ePoa}ZdT}U(gscE12AIJvinzM;j$bqmoo;N!v(M98ZQ41czFry(5$CS
zjtk+olb4mk+>*e+YPcX|o8fYP9@)vkk*~3$?A9!l#Y5q@fw48lSeg^sP{1Z!7S=-n
z#g{e|UK7>A_h(_8)b9Hj&rI#cYq_-h{lK0NvJOq{wyYNZR01w7N#Ozzo{dhYl?01X
zmUc_&<I`ENiuLhMV8v666*Lu^K4KHjJ!^eLhI4(Sn`jkr7#cmr95#)ZsEEUPcZQ0%
z5PJUvI%R%*^O;)kmA6{><t+S{>4N_)v?8rq_<T`*#AeDk^AKqOND{;;1O7-{uPaLY
zQ)Zy%1oREE>rCAys%M8pC(@wCzm-ldS!s??T9jJ2b`Qws*Gw(x8QCV10Et<Trxd;?
zWWhHcL-<DGwa_v{c!Nziee{6P!W(4CuPjEYHiodtzgjdAV+fTo!x*|7dRJqJrbASf
z%G{l-z&uh_r8__xnX5__ax6Lh;a;o|Ri!eoDrqtq(Yva2AK=QXO5cOuSXC+ecC4#P
zuPYYg?AXGcKXYYYb)_0geOH!l(mIQ4OXEUO5x^T~$Sl`dNZVpuQ<mf)$Wo6@Ov{(4
ztTTwaVtV-?B)*NZ&NV!>re04fP&baikcl-6B@#>byt$-DKB!3fk1_nrHJXQ^DP<K?
z>!<Ewjm#R&d_@kA(U?<34i9rAvK2Y}DGiS*a^Ubx+o2~D2u|5F>L_wZQL3ZA64zxR
zOCM$FYjFzuj<vWN$RN*J+}}gHUTbm2SeL7DXQr>lX+(6c$Ki1}4WTTrjrF)6kgBU|
zqP`yY5uoEmMaPj{kNZ^?6lH+@b8M8p9v3~NRwd#X8;(&Hr%04xDJhP#Av4tf(ww%J
zc0reu_G@%tJbWZ+5WXeD>dxy{fdsm(?wo`-ZBPD|xq$W1i1I3l)t!eyN!PL_^jxBN
zgn7qV2Pwa$2uLdn*_vR2u{svPnZ%k(N;f?mU(-6eAkmQ-bR-6`<R(JLb10(S3#;xL
zUP%JGdo10!JOXkXXUNfXXsHQo!bMGj(3fCPJ616$<`nrgkt!?Y5ol2eR6l$juj%8A
zW{)-z{{*mSFK3UjZep1#yDb5imZV6PA-p4+D*Ip-tm0|@W59}g7%ONhG=0P-oO?9G
zwP`*woa-aqM4RTL3XfPICZ_r55_`>U&wzxOaGM{8-v34U_=QN7eKrgKr6>P$Y}Cb*
z{|2PWp3lS+ZG=7!^7$W3`m}rNkpvSiy2n#$R`~ra_{L-C2|)G*hOG4%!X}(PdO&Dn
z2wCzq3n<zc!X{s_Xd=cCDr1H*^heOU8bdT4B2^~CYME9#QmL{Ht8=8v6ml#j@fW>V
zAyQ>>$|Gqq8PPjcHUYTuRM{kcW2rLs?O3PEZY;80$44SxwyV#48J2)4y!uX>J={Sl
zPajQ*UDZ%JI9^~|tKr)|*jXJoa}7MLguiAKUV<sl-`R_pO<pH%vs3PZeXx@oUzn|O
zS;-CCde`i``heHf?Sp?}hB6KN?d1t$(VO;ObSA!oa;mV|nFhq1Qn-2Bc;TY>z4Ac=
zUY#{nh@UEK-nLEjglqJ&krvZl0UD@BR92<-*CZO5Mv&Zqh483hW+sRnpdM___81-;
zr=NpdgOSi&la!{MUUc42mcJBtnwAgDl-f+e&q!^)=g2CTP!%Il`Z@^`)THn}rOM)2
zGa=Y3bvjS7irg{WV)C}yruhLH3bnODK(X1T`2Y=t+FBu?7`$7Z%HNdj7fawlZQT7K
ziH1JM&9q~aLCLid;!!Rr8x}rI;!X~K0ma-{mA=#Ft27i!9Wv1JO&SWN4jf8~sFVBK
zUL;}UmNf^{kFZTK?3Gs2*x&YZ5_xj-R#yfk#=_7V(+8REBsYKh9t$RW()YG)ByqQ3
zj~P5#*iy5B&K=Wtilahq1^I!$G@Qe^9nnlm5CdKVV;B+2r1_22iEgC+bX-?X_boKL
zsHLShK~o^a6v~u@xoAo=u2;}dDAOUJ*qFf^XegBF5Ks(Ha&mVWaOS62n&VU2XndF@
zG1OHjlaon>n#&HAxJ+%h{a%t5(&-2&rskwOox?N~N^>&M@_RHCN^>}r6jfM?gH@CL
zV<Zeq*^Ae^55xQ1LwJ{*8@3}3rw?hH_KP%l83yAQNODQ_5KwH+gm{jILaoGSpyeNE
zD3p2#D8`u(Tw)?ImQ<6MX;?W<gy1NcQr6YY+DqjtmRPcu3MR-Cfqo6V01snWWqAaI
zT>9f1>9ldSmWD@pG!8Epr)(N^6iaN6L%Dbw{VHr|j?)Vh+8GeG35qmJgl#nrZ6!IR
zt41a-)JORHdcde7lt>SMx0A?|@g0X^!Jv1LFka&fa!KU#F(@%V2uKjQ`ICQskg7lF
z^MenNxLdF%5%8u%d7y77HX+!fG!)7&rE)Ai1p5pPg$lt06dPKep`rAq#bR9l4GAN+
z-X0yp_X3H$h2mN;=*J|C*EoZgtRn-B-g;`mptU57*EoaDCXvslK}%qDL<=p>BSGZm
z4}+3xij!*R(rxw?B<>dMvGNQz92F>~excaJvse!&4&X_NWRHH??>XHnW-%$FtyQMi
z{wfkWYJayuoSvul_uqpQde#1Ru87G3;88nrI;(&`2P<@M7gGy}s_9dC#^tFz?C6qs
zV+MUPPqs!X+xWYrTI%LOFWCDc&~u+?3G^J<YT!S}f}*S;dK4RVsUiBDs-lJJTt1_N
zr6+wgn^g(#VQ2G<23uflJ8a?^JLzZVK(%io&CAZFiI=|<E8mHead0RU%ts3KY2Ph{
z<M>i*c>oUPuJDB&EYX>=?Jsfg%uWj$ZRz8=)sENG^Ci)-F>%{CH6nQo3@oj8ZZ~;0
zlVHVxFV8@u<jd6Xn!ZBc-BczDhW{G$_-WRmxjlG&s!}jQfs!tF)u8%T0-x@yHHeR>
zg+x4TY9TjUicMl`{k+2ehgtBCE55!A%=`gkCQYwaHHl3)|7nqF6<;W_rx+BoiWd%2
z$vFt~48X(&+2d1yrwxm)&w*AEOF}D;H}#w5uNWx9&?q@j@LH~c@>Ssa0PD~kC};FG
za#9a%9#txMRsvI+WHE#gLz^Kqo<-ym|7?vEC2z_?Nj#YT8!-G_#&DW)Z7^XIE+gx~
zgkt4`iEg5mPvZD{(qfd0@=5f=R5248K&=G%ze4X1K@(I`2rL$&1i74r|1x}VgeIl!
z)?bNlMUDS4^)N~{#UZ|@-QFJcPKSxjnehK?_`ezcZ-G6T18_K?KL{U4yLiiG(?ZmO
zqlj7tjy`8_loqonNR^SDEZ#JzTfjF0{U@L#^WJTR9Lvyurxz<kL8^?tB~2zHdKaXA
z5^&`OsegpuSV1cLcB~6h%Z;YnfVaQFYzkk^QiPKv)x>JyHgn(9S8*zZW8a0UU#EH>
z6|di$DhLAP>-Z+cm&wcc3y;SOZq1ply2Uaq%`_|1t<WtiMke1fB#)U&UXM4GtG5sJ
zgxhrH88Q+RFcYweZq%&bJcS}sx+wY~<&ypiR9FuKOc#@>uJbSBuGnyV0gSd$*LmRB
z@L#adDCx0Q1<}}8<c!}1Emk?>--@gMD`rJnwJf=c{|#sgoHNTW(hGO~iH1VuR0R~9
z-2YE$C{*rWKry&?T8}}%6}a>LKt0<j?hl?o>9p>EGcBR^#2{A=bTJ^C+}CQMAg0sO
zX)=QfQP=i$u2se$7_+x>X~W<YsyUb9hKPPhiENA%8Oe`x7z3>a%!M=*YQP96HUnk{
z4TTyo0*WzUGQBnHMiREJCYe;fkGE!B&%|T86&rclaJorSOO`PSPTQ0*-9ba4%9t`}
z&Al`f3#~EAn0mZ5>mxL@9ED7r)CDk03;outKch*dN{&7QO)Y)#vosXSv<N6R3jen>
z6iP`1l>DaUA4%B9!nAysi6?_;$-wFVM^a0g76HY^v@BmwMJi<ZT?SfCqM=x5O%~I#
znTD3bv~W@vz_jG6k$gK%5M}=_f~H_Vj6v74Msha|k1|*sUM^1AH0r34Oquc-HIf(d
z#giT(KfK&Gecqc=^U*U1S21Ok<YR7w)f0nU(*Tcaij5PvgQSZzO#+HJxlAuCyO)MS
zIkXJ4e1wKVnI;Y;MMc&mwMi?1`4kB+w@!3|>_Z7`Mk3~mMRiBY7(Dvb9p|Q(Ly(!O
z^hys!PDiE3U%(_bPo>A#LFm0IJ?MOytok@VJDHsoATK}~{%Mg^0q4+J0pfLAFF>h_
zB)04zZxZ_l$P@Nj`<L(=OJcKc$2y5!ZPbGhUWHV0gFq!+<y+bN&SD24yxXc&n1dH-
z)XQ7i(apj)v6ou~lio}bj|?a_-C1}SQ2?)uFY?!3brC#;Yv0-I12ik_)2TtJ4%Eo{
z7PeV|y)zL`YEUOB@GJZFZ4t>H<1B^q)L9CggIP{e801YfGrV|%8TVS9qcD<alnDGO
zUxDzb)gd!yCeBg#x7b&(eD?QYtQzMiY$C-YGhjWQQsK&}Bzc)K@hO-anfP}dA=MF@
zV`LeRbgKP)l5J$B=?zRPtTKm}(om>qNI<cP?Do)5C=C}-jG0EyNR=&M^p$F#CNa((
zPxZt+S9tXvChvsVpr%utskZT$NVShMDCQRO&C&vEwpX4rqFs{9eTEbVd0S~u<WFTd
z?j~hGj(|HkuI79oeX*%cL!m~1fMPQO9-*O7BS1hgcz2q{-0u8NnL{KGrFfFWE4QBa
zu{q__X`)NtmG>1Ad9rq%LrGy!$4i(0js%jML;bvTIh93xj6so|@zUj&N$f3HWEH{u
zD-DIx?o?XRXYPZqGlQ#y!J|g1fMPRJPoSanrX@wGlB4bn5=L&FIy$57%_Q;`DrLc-
z%SafnaR%K;BA<^z9cF2U1d^LWN69SRMq+QlqC{|;Zv5`)7m67nSwtLW16>N{Ow0(W
zY#McBs8gmwM?X7uUY_hhdU{OJ(~%zgIm}n{q{p5F@%KuP^_a!V{Megv5Zaj_`#vPd
zekms3xf5ivgS-i{uOU}hg6wbb8%vO}Z^t@8c1TY<q*f>%v1HlIf+fpHy!%d^ebSyq
z_L-1)_Z=h$19%Vl6wjRrmbAi3HDI-yMgH|#_n^}(@2Z#G>Q(Yh?ANk}@2Z5}Ral{W
zR8npKlrZjAf$FP?Mu`xT3hoeI=2YAFVt6do_EIv{w#8bYa+;i>g_=n(K9P!?Go|2C
zC^AxTw;xl*I~Iw$6G`ThLBk2q6jZ_{O>qtlg_`^bC^nOy^Jys5_z+OcG)0dEa<0^U
zCF*vN80!&OrozA@lc>9t$va_ms6-uSs!f21OK2jhT*IK4iMs2KfRz`6&5$^}zHP>B
zy0wxkyot11oi~(u@<|zzBWQ-xYBPck(NL%nB%s)gpu1@()Cdw#41Qk15`F#nVlE|%
zIro1iG0#1c`bbK&(}bIz3Veb@o}5r}C@BnjYaeXa)&xIGqL!PlePNetnITmi=`-)=
zNSc%4c;K2%$FYC=+Gxg}=EuEraHj3EumxrVoxQ13aBeJZ8g)#;Q$nGmKLvkV@0lq*
zCrO`<E804y<KKpuHqUhY7oZcprsEl+TRAnqq)!HPPSan4Y5I3X;GTP$E<4COO<#AC
zoTeX--`F&reLL3E^s58sfZKPuN?fOMi7>MH>QoSyzGu}ZGm6R;I2~mSj<*f8_jQO6
z$0(0j%7OBi0P-Iylo<zMEzq{Z(YPx%cYGC8(U@tSkcP29*r-%k0uQq5X1#8F8P<&0
zcSXgeG3d<O{I)^U<hHM#MtF)fG9%UjCJRB2S#!>hW5K1aG0o53Oe0N|GryC=HzKd4
zs0s``0OCIsgFG4*71DE9HV#J%>C+gLuju;@8gq(E?@QrQ&!X>pX?PT;IJ{h(vT4-e
z&r=56(f8*kr@J#|bJAV4Lc7CNp9##*<Eo#8cD-CRw|vqIpVA%eP6zucIM}m=UZgtM
zQLh~zM?5Nz3Y&FnZWD?e>oxkAF#85-JU>fnscyeuP{s$C`~>Luf}*4Ih_K#=dyKGG
zMY*BdjBSm6Nn!ZSEEvioKE8pCx||T!92*0_Fg6yyEEs;lfn4?Gd?BpEi>?m11?H0r
zPOS`A_Pe#pZEkZU^J{<OQ$(!K^*UE#hHsX&!%Dab<3+whq8sk|^rO4{|ATzt4m{!c
zewvBR;uxZThDManLqo2;<M3ZVZC+tb=yx1$v8G?TsSGWYW{sRog)H54|7SqN5VY(?
zw|00BHsN%4H_yl_?IbK@)k~aJ+u<diu~E1{9ztQ9LN__G)15z9Z^Fy1o3pJsw-yHa
zF*cxyX?mh*w79M7N|d|}(SUS%iihJ_cqrdhJj&STZO~tBi`Vq{k>-@9R6hq!oy9sd
zPp`6kwe9u<Tr)?RPy%_J8^W_Wi)vk`Sfs|L-BN75J_}Z{S^fpE;vI|?G!>d^ViV3i
zn&FyiBEz|A(oM8;rBIt+w6GQ86eP6X)QS_1742N9pF;0{3r+aBQZ(t}6eQ`q<RtY-
zoq}{XNF(zpND4We|LhuB)QcBl2Z@X<BvGb#dLM)IQ9zb2kbexnv6(mfwh%LK6Pe_=
z5)|Bw2{7$F@@6AJ!9c0%1VP)?`|vM?mj2^00Z+sPuy1(@coq)w>f~m$8sdXi)txBS
zYa!a&HVZWoA9<^w-4aKGVf-}WZS+=f0@>8|L|w9@-bK+lJSi~2lL8ao%mi*;nNXWo
z_#Gf#96TmlO!i140E!7bmS}h^rkq(c7}Q(M5~Lf(qZ8nzo3)_Ig3%)E>0x_n+He?J
zG!WK9r`q0mTr?t|am(<tg>)~9R+U;!SRDzAqOaZ23V8)yE9$m)N6P{?4B;HspdGC(
z*GqiOeWuA+Ty95$?{umyH)!840_140-U?yKy}jF8y*nBfF><{LJq^Q3tP49-<{I#0
zd9z**0abUVa;P0G+S-nma}<gIcRxHZB7U}-Rrs|ste1+@PH8{hg~;^YXx$8SP^*Wo
zcKM!YWdI>#tu<FX1ZQ&1weN{m;lWakVyQmI+%I&skzaAEF1u1Z2vq__rn^7}KoSj5
znc0OEaMoA785XPV0T&p#I2!SsYW#B`TI(LF0RLdSQ!9`U4MYQFce*tT{jGxPgc`n~
z75a6ZIk+;6Z%B2K7{reqAy$M1s(KloYDdGZaAxAdiJ$`CPcd&6-DVR`8uK6kuDVU&
zJ)&BImmjU>JlA>|V7;`XO^G{+F8r5ZKt`*0zX}K0`V{j<qH(kaX#lMYJYZ23Xlh3*
zt4?jU<;=Rk@iP2g?AF@R@es#D5L7I?4YyVXFcrqwK^TlPKqhdcQkJUV4MwY+R#-2p
zOVQ#ADB8+$Ypzj@l@S=rF2KR-ZnNf8QFN=>6{hAWeRZi3xUF)%cxVn)X^o48J$QDT
zYRR;u6c!Iuf(pzyurEB5jGavuV>u)T2@JRuKuiF`f{i$Md?nC}#1v;LRUi&5&IbGj
zKmxakQ$&yr_ErXwFK;>r<DWYhMQd@Sf#%d^K@qE!xk}jHwPtg)s^K<Z%~GshA~7rQ
z%CrNx<DiI6^un+aOl{o?mxIa1JZ!Y9P1c*UFqYQilMDfP5;oewOM+qSwWDQled6pl
zoDck5)to6My2mT3`*URXSU`a)yWee2R@`ug!2naR)eGmU0BksPrePG7L$AGaS+ogs
zat=<!6N57*pvYbV+&&zw3Fd2|bExQ5X1yx>$Cr0i>!|uGf;k5U+N~`Z)d4W-c=ntB
zELQ6!N3eEN$#d`}!)}w0H{dd3A++A#Y5<5;fL&M#f&bhL90u*FmtZ9rbU@gJb!aGp
z)^y^r${gx$weIl2xRe<qP?@=DIGHX8=K%xHh4+mej@Fmpsr^NF&Z$(3PPvS`7#Y@U
zv1gQc;8th2H9H(Facc+KyIa%IGHy*^_<|-@YJ4_X<Zm%!db1Xg0?i7+&_W+@5_JG}
zJX&X5tvPc53NAx$Gx3i>G^S;CeF!h|@1y?b(dm8`KUUx=n8Wz-1^oCte*78U_!IbO
zMZ?%CoNutcdAq}`&0+TZVfO6dmUlWh7S<s+&5OL@Xi&U_P5J<DAbA|;MX@KA?k*ar
z1mGLn_h8Vtbt8}4;pESy(iKA<1~K)vaWq#Wu85|O&fk9lKVHU<8!^Jzhab-k!pHOY
zacCGmZpV+>O85xy15dAKZ*2B)ce9VXpM5Ok^s&&>$2BA$*O+`vQTdp{YDFt`EoOXO
zS*qZ+RuByEnqNU<C7<I0)UWX8z#M?q&rR2>@xWk@o{+d0KRs;pI4q&EM(d3J!S6J~
zHdvAM$-cxLaQt9OOZ66-JMglldbP?G6g>LSc5(fF(5AcE*MeDqb1lKZfPMuq5;(xv
jq8gW%$CsHhFR#=})m9k?&n()%kun;6VITPvVDkS1`Sc`k

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/index.doctree b/.doctrees/autoapi/lmflow/models/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..a5051e4b824802c77a588e05ee1ce965f385bbea
GIT binary patch
literal 3949
zcmai1U2hz>6_qXRYIh~AB*zX4*h(7|N#xejwm}|>)(B7(=!=^z+6H+G2HY9Wk`ZS(
zAvxBnK!D^Sg$3{-fLS0w{!o8b56PL?4|!t&3(efQykETc9P-=Ve`m)#;m;oHit%!B
zNwTbz8P(1XGMUx|EscBao_^=Pap!iZ$hDTWN~yCu&`22<nWClp(%G?BH(cn;X8pdV
zso}Ehma`t~v;8mKIUCxZ;YQGvPxeR@xscbF1-uLGydw1WKBo&Gz~<-gU5uz-5%@Qb
zo!Vj21<ZDBzln2d_XIDgo7OX8N5mM#XSG3oKA&BlX&?K3Du~t);<;CnB0deWx>~->
zNP4Ac?saOrZ^+z#a_pErW{2zv+hu3=#e~U%PE=i%R86j>x|;m$v%ey#_8$V&lNi{9
z)JBquPh#4WnD&I18C_f|jf~!Valf9i9oq}Cl~TnjvR@(t_CBigYy3XI?<4%4`fN}T
z#X}C~cCR8_Ir|9qYeE0?tMw||hf#j9E@u6yainIyWf#lzrtCcwksZWdSZ?M?E3l(~
z)w5`?*}r$!M9rST;-6!|qD8Vb6CZ%D@263nJwr_f2w{lMlSL)#GP_%r9c+$FjiRgF
z^_G&c$M*5h?8kaZr}o1)>w0Mt_Azdw`;XhIJ0qIj(S7=6yBSR}6Y7ro@87IW%c#C>
z*|Rq<V|lV|#V1==yv^5dwyq3yN(#Q<<w~`-dosX*MA5mTS_4$>6l2St@X}C~la%Te
zFIy_`GuF41`fWjNarBb=6h7AU>pN*|DOg;QH0AD6q4toMsi-rWa}fug%~-~A#@JU(
zu##0wF~eT61^YL(gMutFS{deE+ymy3D!?-kaxUb|-P%6mSq3z}wMUKFPB19zVn!9s
z+>0GM3gE3}DQ=)U0x2vJ$ItBN>@3Xc2($YS;O^Rh+v8?DFRm4-Dyke=MK8bp@9Q%U
z)S;aTJJL&#AuaZ8NE;;dwtNf?KDYR>G<MFOx5RQ08d5vjncN^G(&B4tb$11ofYj?*
zuBOy+5qoAI8bwMih_8H7DH-}Xa8b`lLqsfZqZ6uKdw=)rIIWdux}@W^!&Vac%?>m*
zhL>|t_i4w)xssRw;H2$;K}3xn>Bzh<Yg5(6P1*6(4(Ak{MT}IytYJpLkcJc$UJjI$
z2BD(G7S7(AIC~I7NZ|Jh3Zv$=RswHF*xkf)rT>ahdcHMooPChPOeqar`S{8n`7;7q
zD_P`H74DTCYfx8}0N(w{L&M<XD4~J|Npj5%OIp5$zh*&0;8T03XeE_N1bs<S&ii)6
zi0EE>_LMF-s%VI*b*t2~y^PN4Ioyq}i7HV=XedewNQT(5l=_r>e7r=>;vhoH*<o$+
zi$7dwj^`)q9}}vSR0$(xCeY?+1LT$(XOCm$SFH4c6=%;jzHFFa|Fh$0G$26;pk4Z8
z{lmsed*Ty7<7tNa2qff>RHU5OWKL1;4Da_b`t}_Kwow_lphDZgjXy|CZ9kP|jvP@A
zp0(Y;`gZIAK3R(FeU4r`%Icy@x_(4egCKx$NpQNG+T$Pz)8My$DXKKBGnp(3-%TfU
zQAsf6>dMd}sYyI0#(q6cGOo!?_(Ym0_uDza%pD>bjmZc@2ct)gTGE21eb+z1iKZ$H
zCM~_S4EucmaKk>uSD@ypA2zg2IY$=?UU1_+JNd|tD_Vh_av)?0u7Q(7F9|H~=feJi
z8B^&$P9_lR%j(83Szbyt$2dLmnjrvaCWaQ+9)s-|Kz{MT?><32$0|=Z%zMew{D-Z1
zOfK>LdPUVGrzQ_RfCdw03IS(BLuMGE%yYoNo}nR&q?^nxfkaonLvHPfz9|h^B#h6Q
zzz@y4C#3KDLtPL|+1E8jTBA_!g!lNG2$_<|NA@gb1o41{<FKL1p@dj{RabD~?+t-Z
z8VmsWK*Gx$J<x!{1CJ7b{S7r<_}&$gWML5eIseL`Wk6IJ#`gvhFiH)4JW8Q;mC%B4
zk>G~o4-p~i)8>wXUKM!+v|GDN%a{JXU<VOOFuUkPoWeO~!z6~NIR!M~q<xB@&{+oe
z9>6(R*k3E`gEeWqK|l;^dH2%yc#6vh<ojM2pU>G(>__&R{ZQMXmztXjVfu6aPq*y#
zZOz{I&p(3Dp5qYQkCAkU-X6(2u4K<Uf^5kPw+RiVbFaWEUybkW>@-xp*U}`8*FUgN
zYkRcX6(P=}l%uzNqKG|8{P4GWu^_9GsE0D4i;9^PU84onha199d$?wC*m@0%9<131
zh%4jIGLJTGHD7R40W_t!D4_p9T&WZSTZkOMO#6Rn<gZcT@7+H!X1GlHV>!A5ofRE+
j2(>0Y=)|Ep4w~Z<nsM&?V&qTKU{g2(z2b6M&o2KD#?co}

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/interfaces/index.doctree b/.doctrees/autoapi/lmflow/models/interfaces/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..bdc0c84ee789d15082001f13d81b48e3fcc7624b
GIT binary patch
literal 3409
zcma)9TWcFf6n0`umSoG9+z2G@g_6XfvDAk?lr{uPAunlFdZBM+8SRcVgJ*ZPbBSd_
zAtj{5FmIXukN&zo_nV#FmE@AP0g>IaXU?3<cfND{vh&B&;YRh(?dy{9Vlg6FR>+KM
z?|PX`ZB7g0-};BY`Ct99>nn1mrBx~QZUY)A<04bE@IQOE7t{?GdQ_j^)-*L-7VUA?
zVO_TUvp;5i*D>4(y7I~07Wqubvr&%QVT2cksu@Y)NjQTeTR+l8@a11LU;F51w<Pe&
zcy{ReNgD>+bluwfiQ5vqpnhVf#0`isichV<T_MU%p6L+Lb}ES05aPL0kUT#1GFvWR
zrX)R8bQW~1+csn#K6$p!?yy~Umu<2m_tTikoQ{<(3aZ9uQk{;!{^lt{ZT^FUdK?`a
z6Kf<X`8XzH9Fs9#CIgQdU5u1QaPK|4YNu?&bt>6PE@PV6`v``8h}?aI-#z>u;P)^D
zhTJG#g>&q-O2U<QAK)r0=ocTnvkdP2sB-d74c46)Pb~YGoh;*@u=kK<wi6+=#K1u#
zaDyN0G@7n!eQQlr7Zn!2j2Vp<$vfi<4uk?ciR$bTGS!0%eE?SGrL;x%&tz?&dNMVN
zu7dJE63Onk`~R<CYdN2|(SM#YwjfhMW4ZnBp156Jq{3!&#zj>*2ITf~vdCy@n16N!
za3y8cZvgjP$f>_@UB<Hv?Rnvb_32j7D{MZcie~=Vh8tAfsAVC}p*ujAFA);5`6D~3
zvN%9izeCHcrQYq=i90z{q%5iOWEE}U{I~1X03K*-!VUDY{G`F^C25V^U@Q1gw^mcx
zrLkl7xS@xW*gGf7-tn}x{%*1%O_wxxZLXjQ{k9(aYM?E-ut)BOQKZm<gnA^Ul5<6M
zB$tknhKQK-TBoXk4fkDfd#P1{x{{Wtb{k0~EZ5W27+%c5(1$JM&XfdS0@tql6A=~w
zX*s(qttqYX6ShBb{W%2*5hInyOGZF>hU6t)c9fI`t^!{R@2-r!+lekDxc3wagZkAf
zfwv)Qo&@p>U*Sq80Os7gYcrTBq@gPx-?(8I1pr*Km`RoUH*QaZfXW2z7zT`n!8=hx
z1+65>88<9x2pIR(2Wm#1xLrj{sZ1j1bBc6cbpu93``U4bbit8DLyXmpQpa^NI<<3L
zw|7QVfh<BpQIbQ_N6AteLJI!z0y#@)k;)8O-u11SoqTqpIiByXe@v)SQYDNOnE?0!
z1H_gZ@AhNnSFCg@E8ZPl`f|wx`_t`3qe};xqj%RoTsr9vLI41sX2_2~K;8`nDdv{U
zDbk(c{VE8-T~{=fDoel~fDNt*<6z>tsVruQ5$WJrg9g@hdx5Z$rN~|7z}hggd6~3$
zL{=+7)o}$8!Qg%+sk$MQekm$7wV6y7c|g+vU6fMOWc6ePk<=v5x#khC`$@)iU_}H{
zd%4~833_f9!Dw_w)pRgAtksf+G2N~3giAD4Rb$eMV>8ry2b7081>+#ZEAB8f8(>q+
zfkMG^Zu~a~58Pf!%LFX~-Ymjq0vHC#l)(4U7w$GQrqrL0#}MmLd2X02MpDhuPB(*Q
z<p4AjLvxhJpd3BOPwsv4De}3eX6YsKL9#S|zcG)_B>}IeRE;<_v+548!I+s`;Ih6U
zQ?yVPCPU910g!ppc4iwuqRW7g3wNN;3quwO<8vnP19-QD47lIbIYF2GV$sqXiMn3B
z$Ja#2l*D1=j#5V84|cZLY{+tzLYzLeC9VkTVKrrH4!}r)gcmblpeBWDfh8~uE~)W6
zU{^>IJ25P7ryh{O_{z|}=kS10YH-KR6k4YV%?TF?7TG|$mDh*$DhjNMStPUzw@HiV
zVF7f#NG0f9AQ8J1M{k(K5H+WuChSXxa0<vWSZ_h+7{P(c-L*AoTnQnDwORYp4Sb4i
z67uaJjL*mH6?@5Eu!VK~AT{SD+zi{=s|)tx!m<y;^Gl4m$5>jr(UX?Zn-!{^pzLuw
zHk;vvols-``8Q#z{8dlyt!8p!7h13<)(ux7Q8_tGIe-<SgyT^XI=|72*;tiCJ>&=w
z%BNDaD63R&ToSh0{WXjI=GN-awKe--aAiVE-n?w9{(|iVyp&?A1NbntQYi%LhZr!H
zw!f<7Z;{^D{(JNo{xm?hVwu-~RlKmJr#10TE3S~SQ*Tf};yi$2kOc;<$tLUwb~^eS
D`xibw

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/interfaces/tunable/index.doctree b/.doctrees/autoapi/lmflow/models/interfaces/tunable/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..0b99c4ec197e8e39887d682bbb034ebd33781a48
GIT binary patch
literal 8013
zcmc&(+lwSu8K2pi>7M&#H#<q#Y|=xBJ4<%DlZ}Fm;~KLd!n$J!!CNl1)m5kaR8MtP
zrRvnqG+{#&5_a$i(Ukvy5D`R(LLO8k27EJ!kK%(-{0|g-_V=Arr>eSVXL>>+4s=hS
zQ{VZ{cfZx^Ge3UjkqPxDSCWW(!9mk@-5_*XlF3;&bkaTxMD})e`McSjY*Ws~_I?tk
zvBNSs0SJeCz8kY3dnuF41TH*3Y3Aqa33G%O2E}nc!)v_$QntzG<c#oy&nk~(!|!+e
zaKG8d?LguMg2f%%VM$Y@f!zi$oyF1V=h*?#bcz#iWSZ_;WP?hP@ijST6>RY-S<AWE
zmD9c#uxvMN+j8C(BKF#;fE0-9DbG!a!@A?!NrFxMo(b%}{x$2SQMqW_&R)zq1e40T
zu)Fjq<173EU*r$*DSlDD+~Q%Mwc<1gSlrqV<Gt3`Zhi%<n|~lMX=%b*b}B+U@>*Ir
zEiIf@DV$a=9BjGlpcyA%^5Gk2(l(!vGit9QV=aFE2$<p<kmaNJJciFDd@d7%kWuU@
zwl?K-WP5QYH*sa^voDNSkG!Nox5g{7wkt9_<&X2NQZBpvVMrhPorALMZfr-qLb_af
zURyy!DM8!9`si-1Zy<eI3tJ<#k7b!?xW+H?D-hbOp+!+qcf$UV@$(sZVPm5a#UX4m
zVQ!;6Xas3L8bDo*jSWN5i7=Wz-1BVGKAtE!chL8@lL*ESAB$p^u=qC1h}K<0V@~q|
zXkF0GS$fy1*^u??xy35iSufY6pq$eSij4S{23G=GQC7Ap>~`wgvE_uml|;;`Xl<Oq
zO)Gf0`r@}<ee%w$SMQkriq}`l>ty)9^_lIfPj2s!t0if_Z^wgca3UC^T+f3d0#38?
zIFoCYl?4F|hq9aV_{;5ee*UY>k65hjwjsDJ$cAovh_Ma3kqA3*?bvPX+k=J(8&1q@
z!5VhZ`0_K?8)@PNT@&|aM(@d`J11+R8;5ChMDP@7VaNW*)XFT5?0c@@*^PNJy0@T+
zA3)bXG<G#ktrY~{sUo1T-wxfu5kf)dR2=Rf*;ko@g3>*Xe92HFzqlO@w!?OB`<96j
zJL5u7Gl#r(>qRBrO@zzEhy^a>Z5<$hS4fdNp9H_rUfuVQ&}_egnE%eWWll|2y5$GQ
zyCp)iqEMcR2AQlygDvo$v1%kIp}!BU{TAu$*w`m);1j_slghSFpJg)C0d%r&dqKuO
zly$7cAVRR3B1+@}*6bkiJ;xK-UG;e1j)UwFp?$Eelf-GVH)7vlTZj8yb)~T5E)&Q}
zr&X_tBfmz@{XfKy8<SALdvKd|UMuzbURAH`Sly5UDKGsKd5Q4+ijJM6Ogiu`0DplC
ze`!1;_tVl-p8j-V(YpNPh)R+Crz(=;&DDtHwOl0s{um-TS<pX++<rDrxIYCDg5dF(
z<TORv47C?j@qIww9Yrb6SF%RK4}-3b==P}VzKPS{8$DeG;xBWHu6Z4RCL=>GNLeL)
zWw~mbLa(}&*L;L!9$50?w8-lx6csBsRYlmi9->4KL}DtRBm0Gld!L(}m^i(bQD^k7
z7AF)*_rt84BW8WNGPXu36zN$$Ws<|D3MmaQ7+ca_{y3qASBu&T{drwGA*Ye?j}}Li
zd6wykm3U|tL>douMRk)JbmKH3R}`!r1fj4;x+{70g`GzQYL)HACB@^VYgZaO1~+Om
z6qMVI0;F;2+BFpZsv%pKGdj?gI<pE56dV+EYFw36CMO%l4DhQ&;*B5~c5gYkpN$%S
zuS-WR^yvDkIU5@`%}VY9y_Yojw~LFEy*vneK}OfILPtwYnCW6s$Zw2IAcmV$K$hIT
zUJ!AwzGG7^K*d|A=M_0}iKtUjSL>k!Zn%5x1n%Y(cO$IzUZ4}yD5Kgpwo_90qcIfD
z5`~()Sweq2Wp5ovnLJl1ns0k--$6s4$rba{vIB>QaYY^D%1Np=OzOXE>G~lV+YzHj
z$2rf<lt_wVhyMj3ulJKt>b1hk=-GP~5Grt@|A$aAIZvWNeS@LKaIqz8c}c@RGG4@M
za<-HDz7n6lf8G+IW1+RBfccTydlQ9}+AYC|wr>(`x+Q%Oj_^D!RrW6|+R>5shzK`%
znd%>S#>S{(yHV7xcH3#9r?^#kd_9M#xa%MCyZFW2=YL~ju3Wn}BK~^=(fbn}&Mhy$
ztb+Q_BdCWX!6yIX_;R-QA#j|O7As1wmZ-hI;@rtq>S3iy<$vQJ@xN<F`iHC=Rwz@V
zlvLs60`x+%Iw?RAycsPJ{0F%J5SdCX-Tb5l*}NBU7JKL-9o95K>r=*BhL3;cJQ({J
zR${e6sdE3wF!x_K7W3)NQ+g9;CMHfAshENZ^OHgmxh}7cQDX0XV!HH0!+RMkC^c<r
z^Ubs9P%M|VyjAMm4=QCHF+_Dsy$5Ki!fbUNgj;%U2b&B+^K;p3Z|QdUP%h>V&3X+G
zd0<Eipp_srdj^g$(R~RMf~E_H6my<4ip?vp-2Y1+<%$k+#T7{0`S{R0YM7hw=!EJL
zrCf=2lkb=@?iMb<;1|rB!S9yH2fwViYHZ1N^zJELcq+feT=aO8Er!ATio8$6c98h0
z!m==~>0$_{hru%fg_ob*n2_@ZhPha#Wbu=7*-2v>HCx4CdC~0X0kfRV1&;~)8jYgW
z$h;G)knW<N|Ay_S7{QuqMb^SpM5)Mj`3kz=F2m5+7RaAixEK{-fYgt$Sct<=fK;Ts
zgG^4g(1vRgEZn;Xfcbox#y(a{l#OUuO;3@L-j!!NK!B=}RkrWSB^u&m%xN8T!nlua
zemTL2IKr?`O-B=;x}bNM&uCtnSo@yfma1#O25!m^<deu<xfru3jD_X1+YEv{Bj>s8
z7t0yB#tuBl59Mu|7@!$B<Fa<z1-@mPCD51#c$wYD&KwlsFk*)2^8!fAVgbqyGH*GT
ziq6*Ow^R;)X!xPU;y8>gjuD*CFzc=nsYvf~Ma#T`WkzAi<i(LEBPe<n`Sh~JThd@o
zKQ#Pc<gi>N9$<9|hlqSN{$29zpqtuV28mM{bp|h6<hhtRVeDEiqmdP^p_%=ztib_0
zC{-X4&o!oiQe>IhPOIFKXHXTP{KW9uDvS|=RU69v0vny|E_w|DQD#osFSnwErEX|p
zObw`2%zDCvS=E!uN*rO`_7YDU>x$)i2@OiYNKR$lI0+nX5zHirN4lqh)#PAS$MeA)
zUQVpjFI<9QIQ2cy4PkSk@&&w=Dwa1U<r;Y!EC>BHhxPX{g}S+VNiIh$vS>5~|HKF%
zlw;x-!P-GPIfZ<H3lSyTtroVEW;75y44Prwg)f~aNJRi3Ey4OwkAOO6vA^}$6Q6}V
zm*b8zf}VCt^q(>4;kGuJ>>i7o9upmP2SP!Mi@uM`<}en37pVd4th@-D>|2F%77XM)
z*r8la27#~-Ebeu=j}Pp3nuZIRTud;`7S<~%ygGrP&Z)ItT7Kx*xo9ss9Isp)#9i%g
zkg^gXPVZr?e5l^-sTX#67&wIWgpL<<U;}wvI7@*dN!ue}UY~5&4{hy?^p<TegUR&U
z%!LOJKmli%=^e`R4xsidhN+@&;VmRZTt)R-{vHvw>UVTBJCsu_xJ_^C<gAV+a9>y=
zvZV*#5E#kRq)^$-8i;~r2?SbLjZ5$5?qu@p&@L6=Y8%)(#e(Le^y&}bb=uUwH+xUP
z$9qrER|{#gcZI$lr{yNTQaMMkD#Oy-Lj`@PAP*H3^6dug69SA~b};*{<=u^m`Q}wl
zIUd|8vaLz2c)OP%V`g_LeZNDvbZI(mhB;|lQO0j6veZCAH8nkIl~cJ?wJ62o62`SU
z;el_xLxID<X<=4($lBE|EU@3k+n*uO`$jek=(!=VIdiMRban_oazHFHc=Y*ET=^3u
zVT3g&OsQr-#=`{8hx(ytWY_xjeE)6e;f?G$xG35$I6HKmz$*2HMRXh5UMRMeVJ&!p
W<EJhea+hM!yh{N%k5#%kYyKAv(VcSu

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/regression_model/index.doctree b/.doctrees/autoapi/lmflow/models/regression_model/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..8791b623ee64d638d34f9f7c0d58a7321dcd3330
GIT binary patch
literal 8992
zcmcgy+ix6K8TWO4jW20qh|;uUqBc&jciR9G7%P-U0z`<Zs&Y@6%+4I|nf2_<bS|-%
zDpf&h6~n1i+UY-l5Go*`AoT$%)b^P^Bg6v*@xnV2&-}h~=FH5l<K0*dM6x||<~!&6
z&Ue4$g~``Xog0z=#9|zB*Wa&Oj^hUoi!(9p1a{hCev-YGUHE=>D_avYk+l~GX=JlZ
zi~z#suIEI|&z{M|0)Z#47uWOsQ!%p>H}H$?e3IAr)HB%{pAnOZn|Q49Nj&IvcD!J(
z-T`(T60ugq;u!l)xsuUNY@B$S?Gr^q;a|)Y&9%@1g-OPj#f({8iI0m~ev@r6;kiD`
zw$r91X00TN+-91919anY*NN$NQ?_TtF&6PW>02H3H0`8e`O>uP-H7cFOe&@lt3{tO
zzR1t;dHw(&=V!(B1`j%{5v9J*qQ+hj?KZx0?F+cJ{sRKzh9azCrAc6gZbLDqp%~NX
zXH3KO9kySOVqEdTt7B=CkBCXRR^hH<JbwsR;;Z1}Is80~pY!;+K-U9@B3E8&O-zKA
z8)f1VAWS{>S!0M`#KQ`3V+b6tZ6}$P@{jY4k|W#vLGT@Noq?FFR%C^|Lbf>lMFvJ$
zo>AV~kbI~seOMU;k!t{uJ+9cVkXc8vfoQwT&+>IJa$3_UshhHcPLC<@NpWU%)d-`&
zhL$qNXm$-h?Sx$j&{$p7)D47D^a0P4ar0=R#LRxj+l)htjBYxFQLyMH%ZS!(O=C{;
z9B7?WrdrzCl3ox~S8^SernFMNmyELSo>8RMHx#&J*rL?DrNnBco)wvP;F)pA?20sp
zDdB{K7faW_`P^f-p1X8Q|CM)rwnQg7|4S>{y!6=S7MWh0b~;wny$rL0ii(xoLBjpi
zE1xs5TzNBb{a&~BwORaos&Ho4$h*zXeu;S@iwx;hjD%YWY0^y>e%G*!II(=kiX3Cl
z>KaL4*b%c5W>~)Q`6sRzY3%wf?F}dU@#Q?85w)Zh1!;JQ{p5V1&E7F(GmR~Ku9NWW
z>MW_=D-g%akob?ab`2A5<$!Nj51>%c44m#ELP6(v6zm;Zm(hdV)g88cPE#YUZid~>
zpxNHM(Z_pZYp|QD_rYJjaa}Td4Q_QlWWEFLoBMF{6~e@q?;SQR*(uLW$n#au@IU5;
zyVyr#RvPeoM;kC4yQEu;h22cl!tMt6ld)<XAi=x?$-a$Hc%+vUHQXL<E)(m=iBK7`
z2in)MTtDOQiz&P%-4IS~oahj9cxU^e=h|+P-IkyCtjNy}5E{ChD#Q$r`tXq5O=(uy
zwnUg%QHv#+s2@-4(&zjh5&Qq{Yp#w$0B^yxR(P$H>swX1vLj_f0;CA`Ped>}=XbPi
zkT7Y(n*jV35dKzMPj1VlH9dZB{MCy1@CTJF`A?N4N2{x5$@4i&{(TIV3}*B%!M8Vt
zG52Qxf}=e8Br!qWI78+JQTzzdxBHJ2rz#P~@B+W3yuQ_Mx-Vh(5Bhgkf%td1NoZaG
zppM<(3&LATUzzG!I!&xD<ryQLGK*94#J0!)2BaU0h<(T>jVpnlz<<X&M>@1#h`9Z!
z(UFn6mohSv_N8KnB<Uki>w2GB-(3=1qcDq5E<V%eY>hP)wOxNmVtnEFPNyeDuG}Th
zl1Ib@GK9k^6~(3n`eMc|ijE|WdMYcU?*?lZ;S=L^<Ko3#c-U5ap$Kq`@(|1SgT(49
z-NdEuY@JJx{A?QMB?Hc1UN^QhQlxswPdAMM#5jNXGHMH1?X8GO<&jIlnuVB(s}&be
zXI0{!7&Wx6;Hf;s?K`B`2Z+(VtlvV~E!uLXP3WsSuN7U_L%E%_pHbl7DiBGfIt`^l
zGN5>ca+#_;T_U3{cxXtZ(RUdNSKjo6f`B`fFl%BCa^FB{AW0D?h%(7>`Q8)Z_066i
zaI+bCv;J!xN;-xnBkfm*kUdRgE1IVX=h5W8GJxEZl3XeuI}9n4=b{p>h-0x&+Y4n^
z3Rmo_?wOQ(3dm)N0mPaCvUV#8qAKv{MJ-W0l`?yC;N`_=a9BHkt#0`X{E;dq1xYq-
z_6uWz>{g>yFD2&phG6y&V}7TKxr_xz*~J?Q{HK!8_RsN0`&copy^7*q`#05(V;2iv
z(PK&SU+F;EBDs@-x8KI+L$;iA>(BU>Hrwx3@zj)K(+twDCBA<R#`pf6@Z~&G_>Re=
z!QkR_C-oB7q&AdUC?sB?$p?tE?Vk|2XWEDTkhrf>mELsOo{jo46N~!Mw0xWQsOzvg
zjgoYoGoPo{8$xzGqy>eR7<Rc%*3Lj^Q6KQX5bYCVeY7hHmmFz7DY=c76xwZ2_Dsx@
zUqDfS{$X!@B5HXZz(3G2h1bONPU?BmLKOa4GYM?sBjp$$$hDVHeaO`kjA;7?@m2ME
zPJ_Qs)2mFF3;%AbWTj&7ih*mq%#J_tgoV!Frco4uy#b(!ud0VxCqQID%)6nigQE$P
zZ>|`_BgC<aG4MYW*0p|Y9Q<9qx&0`eOI_3z@v-3~RtUP(M^JB|UE_ZmUQ)K}IAc(1
zTG5(nq1fKSzQMKPL8VsYf8!tUze|zxf85nADUn5;Cj<K#=m0T)P-;N%x<^Rx=W^b_
zVV4?l?Zehfv#!rs<f0K|v!c~`k54VV;nVpY_}M#o-)aF-z5PSIxBsfvvwPPVZ8t&K
zpx&siXV3^O1d+?~(hwoGSLxEF&Fj0Deu;wLn$qk%f9=VnL@n=Sw7WP-Ou=8tc1HVq
z^eXe2>boCosHQyB)erRJf?nQG9jyZ~pMTWv%Yew8P+R~l`hi~4aC8#ki&>ITgHXne
zr*emed&<Hnmv4DTv8WtsfdZ*PJ{!=V42=n&1{8m&w8~@G_?Di}Yybh~IU7;Y0#z$;
z>>AOq$<HCY<H|cqo5jGR*Bw>WwV3hnb#Z?ZS$^!vy2eDeRgRbBjAt@Ix#MM5N5rg#
zVGdqX()LlYV5bpHkj!FUG_P0G+(=C4Ovlh{nI=qf3bqr;(BYuf_I1lkF<a3|si*~M
z5~fME%@@(6YRM4?C6#z_(CJ10w-e%JE((GKq$0HLXJWL09+@J65`|p=%;v9Y<l*fE
zMLn89(N{#iH^s>v9DrJmRhDmy6Ev5@tis&i38D^KZ3{7GKOs8ya>Nnis55GXd5nf0
zvAO3a+>{jst^rK>hJ2i`E#@N@22o;q>?Q*v$HXkRyyA6IEVF$V{6kTh#u{i+OggNY
zws778HTtNpigTIO!O9E-VKcf6(dYW$mdSi1H>BEPCQWuWKD{Axg$H^+nk<Tf$mHmt
zdkjOY8j*@LLM$qtS5BFfr)1)6-<Q54+V7D)FQ{`%8uac5dO!4S7E5#oXkE;~B2Uh6
zPI#8zO05<H$0=DG!~9g-7co1C9MfUcV+I-;#%+rljBf`i5jf&HS{0D;EKo<tESJO>
zvM&^Vm_V6@GJ>&k!9~9%@+~x(w#A}ck!nuTFPFlYrA}aC)&!^}G-3l}E!CY;OYFqF
z>Bg>7)<x5CW14s1A~}^)+DKq<^SDe5U!*z-(3%{~+;Ndo(jaw(o`3`sE$TKvHguW;
zk<Ve!T)bW#70WU;_83}U9NOPOi}>2od9e_((4@XG>=Sc1P>vRT2yF-LWEA3~JW0ZM
zv(dnETo1bm5Bz!%wO~u92vQONNHbv_h$n$Krm?>9@S~puKNq4MyYF~fDUW|d9}ly&
zNM(0fRCigjBXPhLG<eeSfNTaGJ6MsNpG=Fh(8-Qj7-w;Tyz_A&mg26TSo<b-TinAB
z^ml^#ewmn$F_TKnXH!^p3`X52-_@(>1-6y5_N>iO5Ze&EvNv#9vJks>F%>$H<9a!R
z&)vWQv?s7#e+N2{`-PL_DB`r)2j+H2cfG(;#z@2S-3%(zX)*^E+yw=kq2+!cPT7Fk
zH5pp|o{9fhKm&bA^>Y4S1E8zkj`C&)Vx0LmX$UW-l{bO;LK6`wUD!r~$r?5CrQR%q
zC}>s!M+>cSXykq?6DNCCDbbh9IBlnRq0t3R^Z`Cai|To;{W@&C{VM%xAY``J>DSBj
z`a}Fm#SFp9=t==YLLW%T0||v#yGrYH06L;u7+ux!8cO?oJ&!Fr_gh7z)nOH5`xwzO
zyN$Tho*KtlDlMJReO8pYL>a#!$&w*m7EiJ;OvQ<+M#(##z$`@tJltEYkmJyBika0F
z(ssED4eT^A!tDXQzi+XCp6LOb(OBh4CwtCE42Y5p7JaJ!to#dtFx;9Qq~x>^@z5Oi
zAbtoM(RJ!dzWy5I@M889OcZ4vl~>R#Xr)3}c(<PAxnfx=)|~6xUh0q{x5yXGI^=+J
KU!{t``hNl1w*G<u

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/text_regression_model/index.doctree b/.doctrees/autoapi/lmflow/models/text_regression_model/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..1742607889e40c577ed86cacc6d7d32c6929f56a
GIT binary patch
literal 18043
zcmc&+Yiu0XbtWagC6S_Rv2`iMBiRmD6uFF)mrBcWVkd4AS+uLTwd+J44R?o}8O`pj
zXJ#cat`nuG9kIOu@)#FIk@VdpZtO%Y5TK0_)DG$&ZUVIZ(-v(}rOA)}YJapqQJ_HE
z@7%}Co!RAZm$Fj<TJGGrk8{pF?|aW)8-Ahn&;b9(rlOYPHm_E#dc7Ic?I>Ym^`O>n
z*v&ZkVsh-$$+Jm?jfd8iC}@W@J7ELpQFC0s9@@?1Lc%7gd+hpARgNEx>{{#w&2+dk
z?2I_07m|uI&W2+*_HFGIyUTB^`N5TH1FK__*uEN@p}ij3QG`iM-ki{Ij2?OnZ!x&H
zLC+-u`jM6eSmVT*X5(gxoU@yaNSs#Kknc9_WTm}ou?Z`VLwB_ugBFDQZnqv0o})G2
ziXt@P`*720h;L){b}M^WwQ83_dyV?E*=TI7)2qapa%P>q&LL;FbC_96PSCKI!gjN1
zhf7z2@Y2%5=N<w9)E@vCEeXJutacn&Eq6(fXi1Q0sVj+=+-BXrS`8!M`L?sW+N;h0
z8|G~qvBf?(cLFcx2q=0N{@;WDbNGLZa0Hn`mt$68LoLe<6LuV{wtf3vqnCYR^P=}+
zFP)u{l{m55&T(fkBhZTTHf#g-dK|lEt%p|2(GXyB?>9IVjE2>UjH8?sLDSzjYSgXR
zGUC7hGg)y}aLf!lp6ma^YhOx7{P5bBi(@3i0GRiQ-4ei9TM~?5KjR#B7C`bbg#(9v
zv=%gSGUN=i*&|1cRv6U4*>>Gn-7uQ%Mr#B6Z5%nG;4SSV;KMqPMyp!`#m28T{N<=+
zk+hDrAV_w2*-i-76$PV&c@JRSBV;@i(*3G|jXo$@o{Qg1_M8!_);pm{3NDK7oM2O&
zNB76pYTLI$vljSf)Us>ZZg0n~L%cuR|M4fDIq~c>C!bY+Ij&D<%dzk+7rNz>Czel>
zxJK<p!wNUfKnlUfY(}bA=&h>uI$_h=lL;E+mEs2`@Q;+;`BW30BmFor@!_T$yO!^M
z!j25f=+eb-n^9~vYqpdD(z%7{C0)naNW2~f?N$fLNmGM^{XX|&3`4HC_1H<yP7u3&
z4qNb9?DFRoM{T#E+5}JP6A%uoL4BhGP{6r646byvb5ldgtjm!56*TOL<<`b>u<9*e
z>_*4aJ?OA%mw{(4ew<5l1sZs7%Wl@)=DK+mW<i6JJ^mIgtt{oJ@5Xj$`P!P!ccd8K
z)&=M!6Y1Bs7HJp-jtJY;+DO<)Yhw}AOKjbS(6;><_Wc#u(k=BS8v!|>_!4%aZxffm
z8==7s%WWpk_t_|(;*Az`-EIPg?ZLBswdK1tH%_ke*H^5tnQX$MY%B}&)T09d?YXhc
z&3>}NSB<T3-HsFX-W{?AS4@8jFW~=2A)Ot>#(WC`Ipd6E`~59_zwIsej5d~R-H%}F
zXr4c&VP%_418$(>%UJM_l*~yDkjdcjz7-mbdOu?_ME_GqbZh2Rh<;Kc`X7BDT28*N
zgBoAkj$~g!Cn(LWXJSL7nG^VUAn>oE_x0kG?4V{`3_obD3kzTs1@(E1{-xq*y_55l
zW70C$&{5fL&;^z)gD>-pE#*+^P13idStc-x^n+6mpv3){g2{vDX*?J-W9afo`A8k@
z=@xtM9vm3xeV5^pcqh{l9HgHBvsMet+MByP5<$<yin9m0s8^%H)d*~JFK+a)ThK5$
zr&_~K`qcw$2ti(_lSRgAlHQn+i>M`Thq)l_9bIs$a7WyPN{kkadHHAl5|r(FbS$-i
zQ+!gZ*$iT<5EZhMA3J?l3@2yVnB(M`JF{S%RzScRga^B9q+N`;GiMN|@IZ2g4GX=T
zZM~UpWQuBvkeExeg>2AJ+=Or0ByM4nyp~{tSCXO(dFwP})}!ThsZgSzVk;?0-UZS9
z7t)0|+uZ@SgSQ}Pg-D#jMiqS{a=5z}d!rEQt#gL+wG@<Y;ZB9^0Y@w%jNm}AeFP(?
zJ%7&Geud*FOK=?LI2IA?>XP*2u6SSA2K+GsU%);_Yi|wq=Ssjnz`>@Fv|+=s@?FH)
z+6K**$U-U#f*sR2^zjqfc(IKGg3u(5W|3GOooATdl<n+|Qqr)&W+yj)L(t$4@uvs7
z;C)SW|9#%wdl7$>?qtJUC8Wqta{>}NK?{y)9E3XlgY$}e@hwB(do(@In{;DMHW+jw
z@P>fsbpg>gb%=&YGpTjNL$=i4moDYKg4Z3QFNr1A$_(OGqF+L+v3aRD1--n#(Ba-S
zpHhduW-$MIIn0044Q5H7f5W@>s+XhBh|t}NLh3PFZ(vv1c9hactk)-v%Db~W8f`yz
zO-kLFsZ`++qPYoo!~0z#&#c#}YT0dCRDRjDuhbA*Cu~Z6H4#2LIR@P>aHWt$1qTb1
zvxMIg^^&$jE)hfBdg3L3HjRs%zas<=5xsQ;gcM|C@gC({j5Hqdl#s22O^~)j9E22X
zJ}zS;0;<Hhscfe+!p7FxzR&fGSbxHdgBlT&ysewO?Rf-Tyfy1ausu%16`8L)0OMoy
z$OD4Zj@5+Qrir<rV#Ue4|N9@XkWyVX(r7qu=!$vZY9SqRQHYC!)Rde^la4b5Zgsw$
z>ZVh-w4U>-n#lVQtwFiz8CLBNbPdpt6adXT9Tn$$+lSlULztsXVQZYF$Isql7*`%b
zZ)zd5^H0uA=LeiEoPX{cV+sKf{Bnn97Cb{6ROT{Kf0bRO{`X2c!B@&SsIOR1ZdG4n
zGWF-n^_|1rtdSlz>5$)<hkTq(thu&d7kVmlA*6;uri;JsA~osA4u9ZL<(8z@Z9;3x
z5O}^H1iW9S-m~7P>5ten>=qJk68=Iyj{V6z<^FA-2L$u;69)dBqXq+Jm!Bs^M%1o3
zh84*aKH;%ow5-@!FhcvXOXocch8@?c6BF~?dzha$mW@YR6gOCYl~4I-5Fr6d&)K}Z
zzc5~1Mx3=baLiKAb*J>I`X}t&I76he=2CVQr%na`PART&c6*_di8Nb7+bs5>0vUEp
z0;$??=mmc>x5Q+A2?hU16o&})HDJ6KY43!3>h2-F+!sfDU2w!#@uy7Jka_q6>MFSA
zZZOgB@FB!J8T~XsCG%%eeQKv<6U3NdJD(sEm)8l0ExpgJ<ozC?``sQW^lC;S8X_o^
znf9U+NksjvYzPtcOMDgYxA8|ABzC;3;B*%n!C~H4^zm>+$xD_VXT^JI>zA;3MJM`S
z(7R`fd6J!egh5a&dQQ@#HhrRUx@gQc2fEx86N1@zoa#cM=7Np!RHG<v(OR<YAK0CT
zA#auo+wc7kK`hoLzK3u98d}u8bTIhAu3Mtu+2U;m+jKv$rAD*@04FhWJK@z>lZFU-
zmT!T^nzRMoSo6Iq#wqCNp7e7i&!(^sO5aN^&M|87HCVMXnt5kGG*ryG#L1UuR%+W3
zv(AH3$N7Zb|7y&_G<KBQTWiUf;3I`~dty*puNMYA(77#&JG;Hgp!!%pP^GMS9He-H
zlcF38jWubApkB|cNdy<HSw7@9NSHJ0^=o+CPG%AsBD|$-d|K~Q+3ZI#<QM62`?D19
z6JiDB1uHMqc|<9hH}8Uh_gQR3*&PFQE`SRKsaq}yb{e{TM;UZ)g(BedZgb5>vax2H
zYwc#usN};cN*8xr59)RbYoynBkWGJRhm%zEz$E$PdT}Sbj{*U;8Pa9+p3vIZWN|)9
zW_zF4x_I9p@ewXtv3Dk8H%eu!;(Zx2dcTQ}uIX8s=X@!hghyoe!DKz9LBfVs0^Err
zSSp^Qzt|#o{#rU;xA4e|F-1oPO(xI3gP~xX0dlX!5ro;o=Ev=hBZI;FlorCbj39qq
zx;7hZbtVe`l7`HBV(Fd`{6Xm=y&wpWi4!#goR;s+cBtgKJV5BOdxxPQ^6a<eVQ%g;
zduItNcC^cxP*GsCv;#S?hlC2tVDCt1eHHZJ$A}fLiq75=K5|U#y_4p+>Aj2IQqIhK
zH+>r#!pD2~E7tWep+ojQlX$1pNP<mEXx#_Ccu{T)q+%J}dZa5&^wR`IHi6R9d5cpK
zIi2xTOBJ(q@`8gm3&hwcGJ5=ugLjA?`8kIE+zb}QRX-$ant}Q{shStXO_)u#SH4uw
z)94}7bWz%Asu`#@)sJSP^&&ftamp#utLH&{&brbB$gIe%DF3IB@I(664$Zuxq?o*N
zZ0EKBXEshyhavdrP7Of`V7F`)j+abc$rD1JtL97wn=Kiyv}I|ZOdZ`1_R`f+-WluB
zu-uzY!VSv<5TW1G?MKnBk%`Hl9d+%IJFDo{173cyIm7JdQlPq8r?T^%pT*EJ_fKhN
zdA@V5hwk1_K}qO`6kTZcV^-%&-PCy*XLZ-@86q9J{XHPePIUV-7+0>_OSxMrdUm2u
z1S`krx|VJi9)Mau1z?1?-o-BHM;pd@smth|Cg><Hv>W8rl0HkXC~w~}<KkbDt1F|D
zm(M~@A`PxS2)I9!!u^&648M-)dkq*AC^BGpe7kr_8dODjg?m?nIwDkk-kN#_p!2>-
zA4llp27Me&zukpzo8I00^B(^BcK$iXKP&w67=02uyaiV8;Z(=JDK_r?JGF5s7`JTE
zPNq}Aezn>$PWU8HbW0yzL{ybusvi&yT@BlKI*l@8T1zGrZ(wh3*=W6<PQs1WlSQL7
z*CQev(!4e6YiOl#CNGv%_>aiP4EdWW+<n`%f51?n`B(6Cm3zwum(^1EmaX!y<*WQ7
z_u92e5|R`dV98Z_i6H;Ge3h~YQm&Ho5$@cE=LDwdsj$q^&aiFAWpvp$T*i93KGsiN
zh3}tUT;{P%E+X3&0)ESxZs#MpUCUo>#`%7BZJ9?%*bJPz&U)hjw&=%{UfK{u&Ffhm
zr-I}2x1680Hc8aI!|~}NcRyyw=WsW_Ak#eE-JA^U4mamjkYOio&ik-E<!+9q{ghL4
zXDRTVF3gkQgYQZ)>l)hXF3j<tfu`1d7x*x^auXTB8hn~{_p;sOB~(!8sW;Qtm`so!
z=~<A7o7Js=3&7@KITF}B(-Ek9*E*Ui-g8<TQyNv~+(tgk?UknbYW+Z_4$Uvpp!Cq(
zdqH3Ev^05!xTdpUQ2m-=)oOMtHYm#0%i3_Xo%<(rR(4Rj*w<uAx1ZQ{RH0I?su85!
z4?)tcGO<-pV`NbQ>H37MhRmZA0k`uDwLCh_enG;u-3l^~uJ}f|lW&Pz`~qO=)h*5d
z;vPtAJM^VntfX1tkyqC;&s5?TzXTA4Tl_M8<8Cqk_GY-nd0>irEiq7c$4W7f_r3GE
zK<?LnJ4N7?6aoHirwE*+#ForvWeBh-TyMd(zBFNml(MX!Rj&+{H^ZwrQBq6Vs#x}b
z8iQ=wigQ|3GcRHRlt3@WIOAB%B1X}Sh(@i}Q^j-#1iOWc%y|pt!CRZ@v@F_LV7JGi
z)r@?ehBr~W%A;6b7Cju}LcgD!9bglxk6p)On=l?^leKn8Mb~CpV!c<jh*E1dCX<P%
zIG?7XY+gdXh8rxX^{At`?B^}NjY?#7LyC<A?HHj%vf@mk7;D`|`82;pjEA~~V(r*!
zv{0tCCk%oZpyDpi)r1W$p&(3vK=(v1p~HlHY==Ic_R)PuD%+;FxIT4*9azHzs8+DG
z=Id-96}6*Y*SxwG;5N>6Hi_FdZVOebyv#SkRI{SR_H8Qsjm#@<?3nxp0?@!pa)2zU
zU157eyA_17>D!lW5ONora4bK49A?w@RTuO_*+e^1orc+P-Ck|4W4=kMYNOIZ%xhZ>
zw2WgTYBnK5@bO5&#6em>1&FxB*^bv1Pc8CWqKEP?O*;&O&~&V3-N#i4FaU7F1yVL8
zXs*pN%x6j1;li836yAS96xpPhH-kZqK9qk^7|iw)4&b`T0f~HG$Gp$Cn(J+A-3G<!
zuI?`U_yD^tv}?H8W!7yfv&1q~y1v3jAckwW3=4|5b;Sy}>^ezRQD)Y}cH!PF?%AUX
z+f11egyjt`_%OsRGn#OUw{SM6-Orj@k=?EbChD!xYd@+EgUCkeIvdHkq!ycWxs9&8
zV=lAqMpTarL?o1>N=P7Zdx1=Z9J@I90@uiH=9-He#q`?)GxQCMpsJf+RYP3{W~*b%
z_uy9=uvOym>>!&aNdxDg{L2CNH&73GZvPycY}qXwp~_QMLIzc;EpR(vC!t_J>BMm>
zT3%X0b5w0@Aa~xZ2H`qn=^*vw0HCLdt8dtz7~3(1_QiYN@h;GFGF+<_rl*!{`nRj;
zA+{E=>?J#_x^}$A*MKTma^i-M#l}&!1}Wn8#AEC*c(P%p!kHo<YkM}?{%E5aTUSlT
zU3Yx^2mcOH?OVe3MkvXS&8ORt>Ij6ojX#S=(+_HvB<<mvgPW8!I)C6o2`Y0EV)P|c
zNpJGsy5K)xA=SVpxF@I~Rsau3y>NgOMbus`baNZTyZC7WA&m5U6PFUOOk>rqLxMK|
zfn%ezbdw#dq1PqTMtv*pQr7V^8~j%%uCr<RyBgqCe@$q!O}5)^UZ&rzU}Hj?KzzZ8
zI014Y8!-yYsbH1!%`|`lXT?yoxZsFuc=YoY>_D!SZe;Ri%(j+3P_Zuk2nM>3Qlt1@
z@xD*y_6_=2qI0+f`uKBtd<`FMHcox{sSz0r-g}ex+~mEGO*=yEGy$ptPvdgnh`d6t
zbiVpcg6tUcbZT>zfyIwpM6jF5b#flhjHGV&kg7FATtJXH7dcqGB9331=XC)doi}=P
z9_dkb+PjxND1h@Qz-zO8I=^uW>_b_xFo&RuXd$(t`Uz6&EhG!P75v#)#g821y53%B
zOnZ;#x{a$<`Ai3L^TR@faRPZiSe#3~fr*CFs|9UROt6P2AN9c;U<26R(FbMw7qLsv
yCFdc)xc(1S08WU&vtq$i_vM=Rq|Kb|_PEWO->wszual<2T~DZIsmVmny!wBxM%V5D

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/vision2seq_model/index.doctree b/.doctrees/autoapi/lmflow/models/vision2seq_model/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..f56c09808bfd9c3393b6dd21cbc2510d7f254add
GIT binary patch
literal 79434
zcmeHwdz2hkd8b}8ni)MumgR>myJSb!j2VqA`4Jeukd470OSXk=46<p}bk|IEsbB7@
zmNa031AfV_fU(=lfk1c|6R-`*LK4h+*O>4KK4e(}B-sNYz+!vXS$xPkOO}M3c)$Cq
zdv86yzE#s2@drM~t)A+8@Av+F-@SF~*6jliEnBpB5&17%<G0*u<H)34sWh4u$M1$i
zm1en9cN#(W!S49)cHhw*3y0hGVZYgFmz{372wIfgYOT_C8r@sE;cC)6sMh>R-ha98
zl!I!s5p{P5+(CEwE!{D9I2;J7LCsNq3Qw=qXKKyElXajDgB+^*(08Zr+*+bnx}+y`
zTz0*4gmBce{heLL^I*#c4uh^c8V;8tk=&)>AeUrsxU5!fINiOSX*(RTgP>iV?gXF#
zBJt8{#V67&FV}3}hl}uipkde9x1ma>CI6VV%Lm)e3~AE|mk0JN`Kjx!aW}dv-P7Eq
z?xyhO?QXN~Y;SiO4X3^RaI<}I`_+413u23ZfI)vdV{E(K37U4Rx}7OzJ5$W|xMH?f
z8x`lsWZMT3Pv5<yGwm)42k13LyG-%!86c9o8T5D-{J#bM-wOYa6Y)Sp?J5;&EL_&I
ztL<)hHW2L8oQtd;h6%T@hTD3ya|ibZUAyDH!rdk{ve!Kw^zN=;4kkTht+)^_zk+L?
zI=}|`H#abhwTo$&jb&#_Nk|i8T*{ie6QHjls;{+yJ>99<?NYf}EBP&_oI`cXX!~&O
z%{RS$$2;D>^Bv;fE+Kb~Bp23?rR}G7oIiCjaTvc-uiNdpOTigHvEc?D)(|w4%Fo?!
zRQam};_Ja!8iD_===ediz6%Dukp<%$om=;i=o}fj%Bi)Swnc-&3S2v|TJ7ecD&U4?
zTYg|SDt5bK9k%DJplOxcjvY9b-LQ6DeudTXtBqL^%>!}D+);<a!C<!C?6eYwAmFha
z%#_ngL(t`LwGz19-6O;>4KU0>uzgebS+59D26&4)0D-dUW@RqHDDb(o-8`JQPSKBq
zTWXhKKr4H;ZMR&dV_2GM%}q6@y{QA4HnA?o!*4RP=cNa3rsfy};8@vm8Wpfl=?DOs
z%E4QYEeFCMme;C*)3$30Y4^iC@K4E<pev!^W7$vuz(%DCm$c@(;b3cS8)&ras5nRr
z`w{Sr--lF?2_xYk=mv1Q8%~fIGDYYvWKYtsZdV&!_wjH!{1wc#Ad@a7e1t3DZ|6v>
zRxMY9?mhJ9!*;vTJsJ*!@fpPS&U)bxG^SzM-Af4uc6-(dy5TE}x(CIv4?yl(aQxc6
z7)<dH1nCBMP+IsQ)xu7um53=HbT_$Q0gxx-+)uh;W+ffo0}by3g71x?0<VAEf$-J+
zLB0)r7)Yw*Z&a0JJFZa4&0Hnl&7l%p(MLgVclA>49nc6MH+!UT8Nu}~)G%O*Dzu)7
z4;h}QWG$=KY|OI!VaJ2G0=*B!d#jDY;Uo3hl;5&RDLd3^JHFFC<WOIKEA$m50;meI
zsAO5zfwrij)JwdaBK(iQD7@;63Js*nSp$g`iY@DkW+Q;);4?tn5GJliR`*VV%H7=X
zQ22Q3qaKu{wczQKF;6$E5C=&Phl(M*A`_FXF;*N@8$IgB@%dnrR#PhS@v?GJxC}~?
zWHm=p-D>hn$*)3{9(3B;ROB<6QjL|HjhX7~lohn?hCkD6*PXUMd3mkc+6k6}UnZ<-
z-JW$yPNNK!QoBT^0HtQT)UpHjlD*A_GhyAzx~2V8vm3J=(jDO5sin4P9xj2i(mB{A
zueEEwGae=HHB=(I(P##CY~mH}+<);|0Th8#)>dklt(Q(%7YhwQ$qR8XWkpS_t(RU3
zGXpwT+Yk=0d?Gz>SL8>Lo+E9tv6TEBF1Ca-!#6p|?${tYZiR~vcjHm*y`FU0=n?X2
zP>M(tvl(v$8h9&6EB6)NWvtUh<R3Z-qrSfmd>_gkrZJfEl9`&A<^s&0(IZ<5iF&q^
z@=iw#q|H;t!WH0(+W?3tE8!}_8Pz-eTdPuZ`3@lR4O98z!XbrJj#A4DdI&T`1Y*(*
z5k7he(nbVXM+G6ZaovFeu~voIu41kSe*MBSib@t45$h`3Rxh@rL8v<ru%eitjxor_
z5DqE+lz#UBYNl{;BN=D+GfRZ<4-J#`RQrT_tJ5e4odmipo)Yxon<V{k57Mn9>5r+T
zy*I#LB0**0$mdWZ-i`2c7LYIjz4xh9PR>GhCgBGKJAA&lA<XU|74cP+N4+kQlLR(V
zk@EzO!&+INO`T3C$Sk=!Q!VSW2=`0sG=Wqr9{^Z=8kmfiOx}mn)TjA}ERJv-zL)4=
zqjwjj;@t=T2vvq>^vIbk40Jo4n%zhEBGyFO<{t=OG%=*%GOB>*sVtC3Mxy{}!m@}R
zG}~&3F1|!?1m7g#e?<w0i<`;l_$BGG(c^Ssk)+z7OP2~1E+a*TpvV4@Bv&lKWF<*w
z#_rUD(jmLnadh(G34W*5YPN%t59>?8t(_)WB0Wkn>@XQsaMcHQI3=?}N^^wuX|06<
z(tnf+*@7wrHU4Jk+L`wL1xyk-a)zbP!>J6w&k2a)rsh;eph-jpHSsQ-)`mE(-gady
z^uexHuktp{yc>~D!pdkHY3#S)s==V{M(Ap2Y!|138JklvH})Z%Rsmx(7W!c9b#PN7
zHd9G6?@G0qu%~`e8t}cixR?R&GdJLeaVnSrITg)-19X-qMB{%+nI6Nb6)+)VqL2wq
zI)$gyAw^V6B=NwsmRIEPCzRwg9o)f^RC2-vP{ZyqlA95Ss?K3DMec@c#MhGDD7#wf
z?6oLQR#e3T=T<Ulh9wlg$Ne(6e+Rv)m9BRP7>yQl-6u#Y`v$SbMo;q2Y67VjZzpu3
zYd>&r5GvwsI6|g5u!0LWJN3onaF8$Bx?l3XM|3h64$X9GH98Ms#79a&vrLpo7JR-$
zuiXXf#Pq6cMi_e)8DJBex&nh2l0WH!W;Bb1c>q~CQVI{Iy)jp=URE!+;U4Cc6)ol`
zi`~~Y+gCK<9spTDuxqb#;65cS_YX|SUi##C8pdP`7L#OnSjci}2Ux7TaJ!C^R0^LL
zVlMASMRj<75zbg_eN^N}@1Mx{<ZkhX@Emgi6$XD1V^Ck37;_)%eG9?+KQKnTn$`+^
zaTmdTLcKBIy$t>Edjm(6djsxM?w8!BspGg`&Lu6Y`k6<q2ah5~fyv&6aLI(aN`u&H
zBC?gKKe*CHCoPR^bY}1Fifv9I9W%>yGuxa9V-&H?S<tVjZA_$0w}}Lw_ymizC-*qW
zjcj7W=egg49c9e#neQEok!r(r+EuuzTXrTz9_#a{G9O-q+E<MrRr-^4>FF)vy0&)(
z7WFo6-b`*&mmS#1QL(1ytVXBanu9xe*5=JZF0>8f5=iX^CVk`<p}s=*dLojv4ma2Y
zSgFwJyLT>te|S@w-kmS8?H~+)tuZ~6;#)<s?*UuUwij5Azq;rslPNrP^(A`YqV6V^
zS-lHD7=he8VR=)^weW=0LCMmz=PT{J8qqRC{nRef#TuxKjd|~cVZGbo16g(Di<Mq9
z6b-^{Fh_U>f>0MI^B$$Pl9RiB3Abh5rbsnJdZ{BNNR&Db2**kt@B1K>_XGIQRq`VO
zIIOy%Imr7lX?HjD^L|2q3X!$ubj#k|^?oV_fe2R%AJ{hJ&2oi<qMw6ihKeQwJxdsp
z0|*Dy(RGYv_TeZT;w2rwfA9SP@DVPDvX9=v_dY`Yq^sQO-;L%rwh;?bcB3i2t`?WK
zv0WfX!<GC;@%JdS;N_ejHCoeXiff`B8Hv#Iogg5aYH0a=J~%E-kQ{M%pv?zst~*=}
z+!9}B=@ZCYSr}wgX6B79EA`Qmu<|r<I0h?vTZt!s&V5q2RIFUggC|c#Ls2|AcR}&w
z>})*Iw3H4clJx|Td=aE70wh0xekMSo8%7`q=^aCma{&OJ<SH(VAYuSv?pz4sJxpYJ
ziF|y5eEbvncqIDvN%(fu`$PKqDf;<o`uQmRe2jiRPCkiX#~{+*sIRv5Db`<CIM#J`
z<{j;;<KrdcePme_@A1|$;#ZPJ*^`hw5kC|SMI-*a1&#Qz?1+zPDm~^U^NE;W3UU>R
z`7O}TB<ADB5m8V2$D;l`i24Dp=E6}g2AD_GUra=!QGW^fMx&m7JKj-$ty`^BoCe%q
zew)*$kbh0#kXLw_cf7CA21^F}^~5V!a@U&5NWYQ$jWGK>a{In$C>rS(E@-50$&PeY
zN9kcMIZuT79uTTXm^Y!HNtkP95iw3W$71|Ki1BN<f(ysE7(f`iKxV&-$VFrPZt{)B
zIQ@3KWBdl#1zaLesK5qqwpuVYK95+3Z8&bT0*eK2C>-bshUOjaCnSa(<GbUurGjuY
z(M~M?r2HbW;G^8XgnP^b3qBMLMX}&j;Ielriy7e(&Uoh1fxg1TZ@Hw5x^SxzugFH2
zq`K2_M)IG)nfHUjir~ywNfa8aMWh&>qYbvqk`N5c76YbN0Wf`lTe~nYi2;P?r2^9+
zi%EUOg3UMnj%b0R(htcuic0j`@kXTs>hfi2rroTUU<)MNw1Jz8mHq(K4TS*;<89tp
zwGA76Oe%ry98N{C)36Z)py7=qw*51AQQ=|pz_yp7p(wUp9LKgbG2tky`i8LdCfYu|
z)FyEIDsf%PxN-oSwV|7RhM+5U$4G}`zSV7+aVPKVusEsSJWtW#!Xf#E*+`DMTso*r
zQ4*m3B6vy>P~QoPH34<p`31hevY5C@sJ|GX{^#8J3q!paKm=l5sDA^|4TbvM<Qs*0
z`t5i_{eH1e@0g(cwS`eWgROZ3{8+}oM%aE*MrQ%*nZHZ4tAP3Yv_^Fx8j2$NWxqCv
z{<erQhAn#5LJ|ECbj^+EZR1V}L{HJ-!Xf&OY>3Npv~)0+QY659Kj@|in12-dnSgnw
z%L^PoFSn3M2)_&<{1#w{1WsWH7XuW9@F$2yD1?8Je4`Lfza4J~pVk+Jk4?S)rou>$
zb2o25-;a+U$Mp$#J~h4rb^=a6fK8Jy&1nUtvR!r^?HN9OmEv@qo=63~n#qjt9q!n|
zFLkucweR6u(NHuaTprH|7sYD)Jj$SgjV;84;D1F@<W3*YLf72s<Hxj<F4?izC%tG8
zOZ{kC{5Qs;$x{uhaVpq8X-*~2KIzkNY6bR5GbZ%<THRlkI;TrB;fC~}=6<ZrTGz&#
ziYQM0AmjD|Ty$*P`wr*|v1ww?D{v~9IXRU)=DZH4R=}K$i9+Ty>69qwQXp{Q3OZ!Y
z+9PYE*L6~uL|yj`AVQJ4?g&`eq^|3gzeIUAm0jN9A}6c7%c08qCWtk!Do+d`vZ?Mw
zd?OjMCVM~853TavN50W2kA6GeRo<KVBF_w*+5`t9lzgWK&o=aDF?Mg^;x3h`dDnE~
zsl$%3XG`In1~`GIQfhVrKXnE6#MIsb(ldUQxc(@Lfw}6jPew!0dhE)0J+>|>5oI%0
zi)6LZa#+?_u=twxla%-Ah{@dL)_;bsxy!95j5{U3Jw=}j2kzUmkuKfg(lK4ikihhh
zfo6(e`t#7w1k*EIUqCss9FZzu62Y$o1b-NqnHPeK0g58{zY}dx1b>-)qX<sF9d`t8
zJG1aO9vnZTPBo7=f~PVyZv?*}by%2;(>XSr-Z?>@`5^P{gU+1Ko@~!R^VE?dG*ACk
zqT9*Ku}N1hG+!SLMbZ4KIGUf1OGjCbO`u0)7Y6(qq{n{+C|55@Atxiwazpqz&^0%N
zpG`Z(vQ*NZDI%@QXhpJA%1=_n*A^G?^5P=u)nocJG?ArJ{N&dnOQnZOdX|zhB(l^R
zP-l@WRfB#;St^}RgDjQWF`1>Vf-E)4-3eki`M`?$Z0^(z6*wf!fzwnSj#yT;0*9tn
zET=y0RLH5E7F48G+krI%3yxJI4QD#^>@vQGaXv%H0C+p+SfoXLnNW32*y<8Jf4CcD
zJrc<Zho8b}WHZ$UeU6wt94-$j5f3I0hj6lx)~E!>f7X>pn8Vw5wP*b)1&pjQI9;Xe
zPF@X%|6T7i{APQ6%DNs@0m^F5DB7D8ofO;`o#Dh~JJ5Pd9mfV?U`h_2hJ32AHb7aU
zzXPc7zoZ)fLmoAHUDBF8$+-tKc|ZIK&j${NBT_+%TKIGE@*&ST3GYJ21yYcZ|2T3W
zb^hr*I*(j;FDU$zRN<(rd!LXtB3+n`vVogy1bdI{Ld<#}rQ*F8{t-{Eg{SRvAch;@
z<mBWYa_+M7AM-x%A+^Jh=pa-tFGx-N1tFaIghDKcRqwkF*!&7(^AUw=!sF0GX#RU1
z>X9Ax0qMV{q>b%Bx-f2y?LY)1c3@KnyxURIPDZ7~H}4Jb5Kd%5^r2yn&1f5DH(>0S
zKrr`%#eyx)X&E~kkvJuP5>jL)f%Cd5dM{wD84}Z<wd5O}{?Kp7d-`*OZy(a{vh`=`
zQ+%s)f}wd&drnFWIYy2cO<krMO|%urk@Aa#ka4hTHiYDw&}@l@q7xc8*H_=}3~wBO
zli5k6=%v1LU7N4S>yUZbACm9OB4WAcC>KH3+;f!QHtv)t;8XOna0NWFVLRzC>BYO`
zKT*6}pp_!Uy9519iuV-P6@@$UTzo>XWYInYFnlI3k-4;_7VVqf2v8%+by-7_s=Zo?
z4$J7#O1f^q=?X|j^I()W;izoyR>|?bkV8aj5+^EX6~`2>sUhZA=}e-Jyr_X_N2#Ed
zi3b;GB&{ClQ3oH*Ba>Q*90alMgw6mKXeDwUVLzuJ0sss3mG=5z9@!#$)j+WKQ^8RC
zs`V!6!gQf$>3AWh)SE<o?_%oHtbc6k%_s6`LbWoyh{O96#wKcIwBCFs5B10nO(6Ys
zO4`^CqzmKL*bYQMVuzgdCNZB9)W@jad;u8CU2h7uc)fYMVZA9pRe+nW!QS=eqG3^Q
zE+*e-y-B|v_j(gn6Dr4Gsl9l;nP6!C^=4v7q~1)l6>vM6{6&K4sbJMU)teilp=iDN
z?qgbSo*NO%U2jf6*WC5ym~p2>y_uqyUyFJ(=@#krrsO|SZ;paiiqxAoLO-K=Gs&Yt
zy_x8gtT*2cKz{--VN`FPbu~auX1zJ0tw<WRmEq@b9t1q2IWfwicLDd)5%;iQxnjE3
zEFUb<^P!a$%kcbUl}40fP;M$`*`<_gJqHokQ>d?0&s-t(+ym<AKu;{l$&!puMrR96
zq>I$3m#pR&V)?L}M_RR<yBD~>p9+i@?PxjoNFM5uE8YjBKTb)buB(>SqzmI#FGi5E
znh5xN22_-#L()aD=rJngz6MO?F69JU-1k0YSjq{!5?G;YuXic;cfc##Iq^gCjh1rs
z+wm^tt|L!$2TeF60gggxkM-q*gx40{^QJK~?-Fi}HXI}(vfid2q8P5y+6d^3U6r8t
z&tSbim259XL(!7$Bk_{0*og=!qb^**vo(7km#*gYVn_0xD0W^1WfduQP8~5Tc47t>
zh0eI4K*=HpUaIK*oLd+yo4gaAdC-M&!$J&D6m@nI>FIv0^T;=fI`rG|MxEV!r)hui
z=GBGqhO;qWq`B!RJk2gdsy14kpalD{;GJo)8~4^F7JDOFt9k_9c1B-10?n>zv9rhb
zv-Zlp`dgwNbfc<aTR6)T4$Oq(%u99KKbRPwZ3B=)w``y86-)ASHSDQNVM=zTvic>u
z4s6I=&7Ow0T_>25H}B=><k1oO%$U-8y4kGB7O+v7C_55Z9XGY5IQj|R7e|hTk2J40
zh66d^26d@YEyL^MGTN}6!!&c2=s~C2ctmrA&Hd-zGGW18?viiUTQxW}p1etbd=d5}
zgrOaKv7NQ;QtO5W7;gXd9cVP6zjTkjB9FYPmPzeTUSW59yLR=SYp5391V@%$3#R0+
zvC|)|TON5jJNzs9Tc)<F0V4X8+OCR*qN(lUvD9|c6okuf!lafGT2Z|JCQLHJgqLyJ
z3$@Pd?M}R#c3X8g>Lw~b<%$fA>BygRFL=Ebx`O+ETs`Y5<K=NJoXTfmh{Q{i_}k++
z6_dBe{SHnAdwU$GqG@n#JjXFbMlZPF+h{g~$%azD_oF<ib5rfj5#Irg$hQ;FSB_{0
zj3-H&z9NpE?G1$(eN9|QO{h<?(ML*iSmRMta`at(oQlblj!)oJuqPdJ$ntfZih*Zo
z&r_lcCf~zp6?mGG398TQRnJR(IQA{T21K_8Mcee^E>Y82x6vYyBYhewtyD)4z@%T4
zvGXFX3G7+f7Z~>@*z_x$3WiObiU~FiufR$WG#WS+4VzBR(mhkUd<#zmQlp6;%8OVw
zqTH#zk_)9$XcmI0lYl+&MH4-5M<qe^%&F);bDMC|#kVQ?x&)_!`AiO3cH>kq)L~Ro
z+<-xkjIlSPGz={tE`^sTz`L%LH+V5R^RTuKS7=ho5rzROF6thfib+y>7fuCBN;zb?
zAE$!32d845l>PvxRUj!b7IZ7C?q-~nwu<>?0^oY5qe0eJ#JKoVhQ}9hxv>=VdFX0b
z9)As|g24%=Vgild#i?Na!l@j~6!aX*o$49KB?bM2S<^(%|AI<_VhpEZl7g16L<<Vc
zXL87LB2ESK8Ain*1)YV`Ftxl3TJKi|Vyx@fgl3=|QFsX|E$SVdib)3AjZ?ugP!3t%
zj8nn9gHtijK(jcl0vU+0pj%n>HscI*PP9nNSEX4dVvEtelhA6tyGTay%Q7k5j;jJo
zO1D8*!=&^coC*dooQerpeh8<6xe%wKC8ZNmRG&E|TB$Zxs{7BPJgF{{D;4SF3YEr>
zGV_@z_RmovP@LgZOtR3oaVnUT<dEg>aVnUTFe(OF=w~PmQ=|7Y3o+Jpd_uEOj<6eC
zg;o!kMmZIeEVLe{f@Pr`vYd%iF_6VP3!RJ8Dv*Vkpt_Y+Z!^w97bb3YX58cKosrhW
zh7u_$ab4!5*W-%8a?+L1)i5W$38#X=4X0uPoeoX~b0|(l%Sor>cF0Um%Sb;ZEgeSL
zQhg>@T4H1hr6WJha{Gn1EZ(m$e;>-chgptMNrCW`<wfa`A3zBf)>aD5{;7`L--gE^
z7vG|&?lY()2C6e!IDHbQf;oMv8>BCsK7&)~?FL4&{4GwUw=5~n7GF4h4y9r0U;V5X
zm>Kj;g4PQ;An|2XTC|$vR7~oHRjbje1gjTv$g&Zqf_Vq0VqPzd<FpFY3ycNb%Br^+
z*9()e9Zm(7PgiL}2;7fdm5_Wnt_-Y3xD>h?)(F?)R4|C)R7`4wx8hVVC*oAJ8llhk
z^I9l(s=MSW75bI@YRsA@dcFgd1O*;W#bg@$08Rz-nH;iw6sLmu45MO@mOhKpFtvO?
z(-LD{2P`x#<%q&>pwgn=!Ks*}rN6<cU}-6bEYIRpFz?`0%+u0KIIRL{iLsztS@kyK
zv~;%E9?0^M+&tTR+B(R(iaaF#lnH3X8nnK^63}w!YM6jd!Kq+i!l{@5<5rvs<}RFy
zmVi#mP<Q6+G`Ry!>iJbDZ>n45N<T@Gg{EE?FjJZ+c@HWGiZ7gsN%q-~Q^6c2hb&c`
z3g$43ib3`{jM6YQd_S`fV_nB3H2dTT!S|ukqTa!&m}H-i;8d{elS7uz;8Za0;8e`B
z&yzT<0@;VLpj%n>HskEGNqI1XKEKd&N8@nRL*yOhiiE`P;WA^X=iAWLF!lTooC<~{
zoQer5{v%EW^A%1-OFb(TMQ5g)Wu%d^D7IuRmdaJz=Sne*Orf;$@61dl3fzD)j{*m$
zVv<&V3#WoPLk?MX;#4qaU{nm!%BxWtrpE4PT4Ai~fP<!$9O3mAR9e(KI2Dt$;^I`W
zw30)XLpT-8J2(~dv~mYdt3X;|Ea+BNz0EkSY)N^VPeDCo!~AHXr${p?zsPj+VO$qj
zy7>TfHB2{;;Z!h$;Z#h}@{2eX%!fFY`KOz2qCBWRu~6yeFPNE36!;^Qc@$7M6_a%H
zB2EQ!h8(i|3a5fO1EXS)ZiYwuQCsvg-7wa5AVSkkj_^7Ql@|35PQ@hM?7*pD=_ZFP
zm*G?}@8DF-)6E{7R)KWGSkSGkLW^;_8RJPK_DHx{##x`%StOnGA7$Q|!IgsLooVQ5
zn0Er43I;WtiV1AK3#WoP6Q?r&ywgQ_P@Q9;^3MItOePBa6v{k`ES!o--uW|}3g!$s
zWceyi1#<>Q#USr|2c=<Z?0)7Q#=4G3Xx_;YUOz*nMZJSlG08jsic`VzP7Yay)?pQc
zi7e)MX9G^FK;B`3>Q+|0%{cFzp`9*bmTp!keMFYguF2%G9hV$SF0X*DhRNj;oC<~~
zoQerH?#8KLc*3d7Ke^n3@}PRaLM4|sGc%be@K%(0G~_rHljQPFoC@X)Ib``goC@X)
zjEX^W`F)gzsj>T+To~&*?4ZddM|k}yDlO_AoQg?u`3g=2OD;KN`74|X<{g}hd2;y?
zPOCt2VJzrYR=v$QxtyJPEEU-!Eq4aMnN;i!f~TS52`ODgu1Wn(=A2*R3c+&D|AnrG
zIcM+$w0^)~hEp+t%=I`G%#k=1O{*&u-7~GH!BA`y``IWDs$=BKIh0JH`!;6+M_?us
z1x}&Nqo~5EnB<(Ra4MKH<dEe?oC@X)j7plZ$q|6cO}Yw7!_?UQ%sGs89gWbOlOw!-
z2bC7}4o<}+=Y%*FEa&8q<%2jC%sV(0E$1v1zzoke(C-ppKZR2&kZ~9Tx{Xy|GtM|;
zrl%XG<(yNqJB!qVHIWMXGOi#j1N|51YM6ok8%_lS9ZtmrK!1o+!Q6^d(J~ObTTfrg
zmpqj%zGL9$C~eh2@?{>nF`-cA`QOw$IZj{8q3PlEXhmV5X%jFy38#XAQ4U$Q;8YA`
z(ZFbco=7T!kpCs)?L3@X0YGA+76K%bPKh^zB+n>y7lkV;j$baZ#TH+EnZxD%I+iu1
zAQ+&By($&K^{8?S2O+0ZC<?CymJI7Z7v&z!@SKWC{dWkbf<bW(S?<88U_r&Gq?r5s
zplvsNIFGmPwEaoLLwyzRcmU<8uy+(=YaY&I61C(k^u0vhjb1Wdy@fm2eslIY@Zw<Q
z^ds#h*NG$TuYe;JiyUeH1n9%${XNkU&*Cuqvw8}VJjVXxp!qBL*^dza&%wdX`kNL_
zj<FX56g|fN+eBh|jQwAcZ}b>@`t5ihWADIA$jgoouViQEd9xR3z;WQEI=l#ewiTS;
zp9AjSQuu)To(#_S==<=Fo+QSiLzXA3n%$V~*t2jDdb0v2;Il^~CoKJNp9w2vDQdz}
zPeq-urtP5YmV9_?$t62d-|VruCu5?zF=#0(pMM~5l=+_VTrUZJJ{pQ1ll}+<*q!P*
z(e^P{9G{bE=hrGHVAJ;#Ym<rCk1$hJu1F;<IuTP6x|Z;7pcL%oQK=KtU(!zcIlL#P
z`Wls3U-(2Cf_xdjN?A`XF6(;e+80^l$6zH5m(~KLi(i&vom*V2iGGQdy6l{kEwdR2
znEIQP@~YxezNTMN>L)uV#Y_GUa_JYPKnIEobaUSXQn8ssjD{j3vVp*0^^%n4NO5Tb
zDvf6{LB9*9f=$pl6>aLgc!v-QzDZfSIF$lZW=7n!j=mRC3siFqDor{irpzh%YT-)o
zi?XZN-ibcFIG6Gyiu0d9LMc+5KM6)QDb5W_4pFvWG7rI%1^px70zc(-N8y5A3{bS7
z|1QxIE$F{TzR`l7emmX;{nfKh!)e32f%;Ro?<!ok3qIysp5Js4M7<bnOs$~B+o_bd
zO+no)-(3~G_9*_$vwZuKUbQQ4O3Ln%QrIi~$o)y9w<bl*k#%L#gRWNcd4u|1r)*Bn
zIY+3abnvj<p7qDc+n3_)dG_+vvV2H%6fiF@NVI<uMf+USgcqWr=rrL|v1!6hn}E6B
zSXR1a!m`|m+AQac9vb6)w6aC7xk#ag6k<y^6rKXT3<d`y>&u{1!rS2OLDJ1eaVpp{
zD5qj_N9qim3bqW&sc3pkIU$ECGkJR|rKOa82g+69h0qk5g;qJY0ekVu9Xj)6sxqm}
z$g(Ie=YGX}Mf2@(1;*omCbt36Z3sl~WBPPhnb30=SFT9y-+*c#74$l)86*EL<5V#J
z=2T4l`&OI^=HHx(=HGoiCgj~HZ-u?QPsfD3lbO;)$?r!6LGKH3DmoyhpPl#!P6czA
z9I|{Kr-C^Qqmpu8D1LV0mr)v~hVSRTAjY~5P$q$H;vGLgrA57ihoH$!{{@^1Hq*}m
zC;tzpf_Vq0qOB(|$KY5T_mY8=&@!XIssdv{x3cPOsSB`aX=pM&Kggep;>(j<3GvG^
zE1iz31KS65Ds(lhx+ZWc7|3ucCLu9}Q^DMbQ#p27>1vcW)m?J!6Y5h|x{8_7M9BwG
zK~UV`R7|o`1E+#HOb%IY!>M2n!>AZ!rF&5triSllR${E{c!g%A93l7#R9e(KI2Dtu
z^f{agmX&hI@-$8b^A1kMJS%+*r&S;;F&1<ytKMdum5$F768{rd2$q+A3SABJ(l2l-
z7|d`gCXl)GWVGbK9EnrW@=~EEBu+$mP#q&z{+Y)y|D(VWn8`$e<0$hes&FbMIp?==
zDws3mkmX983g!%qib2kKBTB>6*!|2ojCCE2(43Payy~d5sCRHGCOPNrI2A1C<dEek
zP6hK0PDRT($P*Ip$0-%aIE(?^#;UIwXPoWX4>?R(`t#p%({*f74tolMPl~W6a)R{7
zxSp^?^f+`iOhixNR50w}R7`O6FL5fEcQGmt?YwXi{ZC&3Ep*o6-=oZ^j*~AV6@PN#
z`^;V@ntTam9>p9^#UvvwIt48pFt5lV%W9ko<`s;JK}I?arD1C9er6=bx{g+8M#>Rq
z7oyUl-odGuWTe;NRIrScLzaCw70f$06)hw6cueL1POU&zVod0ER{hO5E3tVH+$+z&
zDkt6m%@dFKQzn@?Tx=}KbfBwYlDQkFg24%=Vgij1;Z!h(;Z(HQPG9d0{UOSm>K3_j
zPM=oc9%iOAQSy_hASl9cDkk~o8Jr5{FgaxTTbv5!FpP>p{&^0iVQTn(<{!qoj!S6%
z$q|Atqtc?@!Ks+!pH-)#)d`k=a>%j~r-FG0r=sN_^xn`oPOCunVJzrYR=v$Q`;2C7
z($Orn<eYQFt{3uECYZ}{$uL`83SA+=XrR1C?C9cK!35F7b;-fCI0smYV;mGxqDiO3
zw%wHdW7ohAQ1KjBn1p+p*ps_)k`1Lk8mK;0A2}YW$fHKrA0k(I(*2e{ahRgFh1>3Y
zX7;XM;jG>CUCom1#IELNVY5(?UCmukpUJLfxe-!qUjD7#VkGw{e+u>}KhHNkLa08M
zc<6A<?MB{v!!?JTcDo7>Zd(V0W)4_0oklsRHXBggTGNhYPuCps6<$MO1+HV!r_U_g
zw`{B2Y*ebGsa>;p|7LrVZU2MqeIs&s1%DondAmCtt_i9^&4FjfXPmavC_C!Y<0FzY
z0WDHhGKM}A8#UWs$p)~;Nb8LrVWl$-obH;x(2|w7fZ4zUthf)lo9Ncm$AF)Yav{KQ
z_b!(x;|>x~8S~z&T)XKVVA)-;^>%7R-HmCD9bj`3Woz;Qt2tx&uHAC1G5PSwgw^QO
zOJ&z?G@P10VYyCq)(s}C!_`XQj*m}`Sn%KL$)0BRM4`{!fj=<dR3=BFt#T1n`7*66
zqP}x$hnToja~iWiXu>+D*<vnq4kvm&IGJchyspBW4G7m;fd$VDQ#a)P2F^D*qHmVy
zFE@K^{~!_FKGdvMtd`RzPoGyCv+}u13qqx3S1ROU&^+igU@Lf~3eUEVr0m!R6Tr6X
z45^`?bA(D*ZTL<*pd*4d=s0X_e{uv{62(TJsoAr9a2{u%NIRg~q^4xgg<9mfP^;=&
zJ1o0VAu%4n!<1INQwypsfC9=4u))IIdq{E0aDaNGh*SQTjEuj`6{KAlnFq&bAUM83
zgVlRPp~nKFXRL(5F$xhhJks=Y<FR3ckpdiBt&gvu?z1t~dym6~C=r}+Ac-2niYa?~
zrA2=h8`cPR2`?0OFDQgeh*T)ldWu7R798rwl#LXJQZnSG-X;dM_h=>6wR|HsPg1KY
zH8uPZFuM$T#&<-=+Iydn0Ykbl`EZrIM|L49`5{w@nrOm1jseW)hWNB#+Sh-^qCi@q
zjIH?FmYDtYC|bv(NXddCh_Ld1TWnCH7$Tg|7#h!rA(264qDAC@8f?g^w_Mw=YDf2s
z$mA>}84~GQ8E#iD2`(e8H+qo-OUBiwGU6)6{Z|R;KZY?kQ?1}3y`z}(L6{flDMSC}
zJUWW1!v6?Dy_pJydxaAEqzlsnt~w?3iR#iq|G{HbSk0rTN<)G-9V0#?I9MAepg8~A
zVutMtCcG;)C~kc|ogthQ$Ap9&uZH2G3c(Nwy1Dq~ND%R@BSD-Y+?^F9_#C+NA*u@;
z2`-wi8v-WGYYd4Ff0RdW5juPx1bv(eiu;Fx4x|gy4X#cF9f&&9(BV4M=?wu~qp3|=
zY;QHB62<C9q=OQ+|23v@uZX+<b!-sa4BXubWg4csyT=*A?W4H+pMZUyp=09iKAz7G
z;M_TSNLT*HJUWS7`A<QJ7pV}q4i#4>U6|assuWiys!DU^cjui`6ICw2d>Y;Ulrq<k
zg5WFzfs|laeVRTmDHIGFVuRp@<iS8FrvyWsA>6`BF#H)<c!Z9L2gAPpTaYRrqUKl(
zEyHU^9xX@V^#xGNBvlJ;6(zh#7p6&E`AT>Z<+Jd@yByVLo!_UkPGiQ%M3pTD;ZzjH
z$okROQs%a$eNND6H((L9Y|~Xp-x{mIK?>tj)-H>Dp_8ki4GVET`N|e8mGjfAbJ#K}
z?>#x9|Cg4~h6GLFFY<!4vYp4u>cYksLDE|c{Qq@*5?Bb)xACM7+Ie&kdHR0>1vK+{
zI_bi6X6#Tz&BUR|X?C4yyqn_f#1EB4{x>XDD(sD$H(Sx-x-~sV9IG`~X_l?cn}wFq
zHcWFs;Q<iepU#da{f^!1feD2zzT*1LUQ#^^_aC5}xYtzlIGaxcuk1LRN69yOoDKcf
zkK=4`yKo6EBxgG*hZ5?Y#=<47xo$X6uGzldb<`y{?_toK_k~D+Cn5ppx1s{9C;V)R
z-+v6(z<sw)&DloQ<DEv}cj?VY!jJfgF4gm&w!--u<F|ClD&mN{(Ot;|q&J$@s+Jvd
zFAh&<C+%#bH}?E(hpIl@RM`pZN9+m+eHM&H2=SKe;9Y-Gwq!?A0!uNvOBK3Hu;t5e
z&~J9yaOcmR2-nj?erq<lD^!Arm~F!`q1|_M$HIZ28Q8V%?nU8<Xya7i&km7uak#qN
zX~X@)pv1@88?F>rYE^h|#NKeocLH)(#qWlr5Lum|TJtAo+Ek?#xF>w0UF$e<1uPt7
zud3|sb=T|-hof7ZFw$@ZPCD`fyWWC7R<xVV09bWqsz<uv;_cmV1!JKEbPqy<5&mbV
zU4y??1<i74+Abd?zxrMGJ>j|;7@*M%9Od#o;VR!L2jni@5pvI|dr!F9hg8%mm7DcC
zwe#L^BfH|%9D1d6xEi=6>Z>3IP~shk^U@0|+fJ+54q#F6kOPWb5{|fbE&6jH9CeOV
zL4ScAbbQfhARMST)16ruZ#A6u(*XTJi?(Ce;mR-=qU;c12!CwV6z=ae$=$GSINS+l
zw!LzjUxn|d=|7g>&NH0-3ilr?HKz@_C#;qO@MivcP&vu~9i<y?iv1EB!utgTV7Qu%
zo9um>{zI&DxR%J^gQ8qeQ4M73hO44eFhL&pdx_KNh9|V0a<g42Rh*Usx9b|^Dpl+t
z#N`Z#2^y(Zge$m%a7bh)XqM!qa7h&$ZB+%Zy%aekD3)HRLTI<)W;!w58hRx<?^<4J
z`A(<OEFGy6S6b^FX*GQZ4uh3Fsh5<45(!VZtwY8tRjNL@j}NZHxm+%S1Ojd)2;)Oo
z(1?TPt9Y~0Otl8WLEvv7-+&0cDG5>OG{|TbF!_qMeK`7a_u_DrL>gan2Jf#|gYKTS
zTj9o((~|f1i7+e4m1!GzCqWUO>;^&0pW411F8h<Mxxj@RyUq42#L|hRC1n6wmI7F(
z25bOx48iqnTh2Wn^trk{Lyn&3w>9bT<cb{rY%x59txa6^pwphLI>8L30Z_2r4eB)@
zHXPW~5Jhkgyt{iS+yp*ZFGax_Q7HK>;vq-Fwf-F39z9ZW;pTab+(7S!%W6&H`YZjq
z4S{xR2cp^sMV&zZW`CAy&9co^y9sW=13zFs7H^<sszT^}u+st(tARVVKZ3!)QShE-
zx!RZkAK<vKjsS(<nT|DM3n8_p&4O_m1tV~o`m|Gl2%ZBDT*nE3#))NUb+F{r?P{$A
zx3UT1Qm#kI$OMOdr#8dT>}a^uX&mbA?M#P53{4<>!4t_%Z-|Yc6u^mK4kRe>n^9m2
zJSza8g)u7548iemop!Zh*MSsVhS6rCKL!byR_OKR<Rbe%<~>7_?AOW1DC9|RHTn25
z`SU6G=!C<hZ7|oOBeK><Y0IOu)zOZ-84?K{0FvUxaNlE)v8!x3VRx0~;oKKTnO3B?
zaG>f#w(Q;mdB?kG6Uo;!wJsB>7B+AY&fH5WTOLX2-cIsCj+FBpQa5<y<BR0uDe|#&
z8GH<pk3HmLANhDc`S=j|c#(X(L_YQpz{dgd@eKL+7Ww$tAbfn1e5`?qmA9UJSVQo!
zg?x+-!^Z~l@gn(niF}ar)9I=39yzShBZoP9WP_7OHb8a4RjS8R8>}i<A#R8f!7l6y
z0XET&8C$)=+JOr|gw&^-wJ3;a)Ag~7QR`uC#9={|c3P*64>_tGbOCWrjE}{Cfg%B5
zx7_RyXoQShZq{ndnIZcxzk*+X5IpbB?mmE!mfZ$00$=pujBP>?oMW{==L6t~t4Sx3
zd#O&kp#fM71Yo}kS?mN@Ao^mvp<q=Y^LIJnnI)d2RHE4xJi9x*(W^iYw5%H7iGFFg
zj6?+#vRCfgcMXWE)LR4>obGPw%+rL0jovgkiBh`V4K%`aQ7#hdpjKHCU8XPqH%`m7
YP6bjxbCw{$NQI<Jo>|C{o$1N{5Bb);dH?_b

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/vision_encoder/clip_encoder/index.doctree b/.doctrees/autoapi/lmflow/models/vision_encoder/clip_encoder/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..36aa1d931087dcca414a33a2c1ecad1e609ac90d
GIT binary patch
literal 33970
zcmd5_eT-byb@$r)z4q?f*bcBZ83%&L?tTOZ12M+I*a3>o;u0VUI2qo~yxlj~^ToV3
zwiiMK3AkQ*MJh40AtWhHDpd_p)uu=dfk40hph{6Af>NnX+n}aZ)wE69NKL=e_IK{b
z`<}NmZx&B1dG>wWd+s^sp7T5B-uK?Uj|@Eh>yNDy|M-Mmvx=3ovqqs%sTNGT$%hNo
ze4}huoaR%_J&!gYZf5yN-8f@c8}+=|<m*r(Zxu_0x>;#{w8_V*xKk|Iv#S1(ZRVX~
zwc=K{2CPAA=%dZ7HNpp+qEj+`xA>k?d9hSIGh0UMAac5Bqi)Wu<Z)h~&6kQbebOW%
zpwWMqd6t;#r0U~MDal~X0N+m2+RR6CuAtTiKB(lI=l!K(#ca+u77RXWI8ME|&~Ts^
z(!+*g!6uo9@+HHzaS_J@6{9SVh6|0Fcd}sQPu0yuD%0RYj<H0yn%0ChWo@)}SR1UJ
zd}+?Amd&|(qf#;Jb7!jcQ**~}zZuHVf54zUCmEYF8cx-y73ZY7=A^pjg6f*{t81=U
zDVS$x>o%ml?AZFog0+qhh-*Ie%bu~`3F)j|(C+2<|8D%>jsJT{QYfol6!K+xf6XY?
zoBS%Y-6)yYXWH*M{vKIpru{0`;Jni`8rIKPGoH%kt;^6c)|j+Q&#)%-1wM2{b&W8V
z?cTYv0MI%t)h}roEoG;sskaWv;?^{@I4ra{=@<)*l2Om)tEHS>GxHrNv|p6xlP5p?
z&{YpVwC`d4w@J<2;Wfw4ULfN*aMjfZ4w7-%jdIzjFCT)nK*xN$0tWcntnYS{Z}y$#
zpoyS6ZyUwGBgcRC#67~1@2Z|L>!YLEbqB)bmHmzIL1(F6ZPX%~B+wXwo;y_YFshs>
z796X2Y?L%xfo4yk^QziVRx^AbfqVS~2*MVsh2;pN;B!O0dM0vRpdNWqY?Tq<k+D=a
zYL>4ezvn=0`9O7n9k@TF$;`pF9cVUG_0at%MK5Fl0~>2*rBJLa<<0`3{0-mzwlu7J
zf2dS+%(_wXwP<}_`RPQ6-KOWSzuMYgfk{FHzP`5H<b$>48ECC(`XP)u?^n?K&jQd}
z`Z^zkB7n9{er-yeZ6Zh#iLz0wG_9}jA)GnOH3ZBJ#1bFFxp}r$D&~t$^Sroy#;8}C
zD|`ffEzz-~h{lIePsE|-ylC4o>Px26<nK#uA_9Y+L15}7_8jX&XFUy9-EIwfHt@9H
z2F#ZINS*zZwbS}TU^UVCMXHw6@l@eS6#NA=_~{TfsO(2anN=E{vcBDBL&@x~fSYrx
zQF;u80B5Z!@qWV0Cc-`hETZ&cm?Yok!|_b1T3M3VY=oVsfZF$mYx@iFk+bE}0lQ{U
zdN^FGo3>d$Z3<)h80zZyc~Zm`k7X~5GIZMEzoaq{;(ruWRL<l^;e-@m<mN_2WT3VL
zp6%Rs#Wg9mrElFjzg4wH2jT&l0Gn`Y2sS2RGbche(<+g&?zQeCGKla#G8fL#_U(%c
z@>Wi>&#$Jqz0g2jpHn%dL!mqCKgs16m!>m&_nrcBE!lf?ST{8h6;oO8Jew-$>Zr}G
zg3`MT+?vMYLB^~ng6G-R2*4k417lQ2odo4*ZN%lwNu}nKIid8Er0A2=nUi}@DqWx4
zqhj}GHOu_Iw~-_1H_z#|*oVmQY4Ek3Z;49ag=+Q@*F)m6>#X(s>VcZwEjqllH@dOf
z`kFuHvOMawA9k%*tXHkyo40<ScaT!W$WBguK|^MYCsU|nKPS|N9@>%Qv%bEf3V!_a
zkXggA9tDz&h^nkXHAG-#^Qv-ET%c-*t?ZcOtjqAq&IvND?IPG*vNQ;??AA8uA<ioB
zXiYC1YT0ZY;s1_{t88yvQQ!ER4WIxUq*72OqNy)<usygOj;SMyZtU}-YT#l*T%~}P
z5LYK|7vuiSk!r=kfX0okk?UTJ{R3iAOm@WKfj@~L`6$M|>edi3_F}L!t8r+xF<W4=
zV$j9_2V+xyThwc_dotN#B_p%$_NOuSG`j?gz?_CpR5h)!j`w2*E;^e)L$-0cle3GM
zMsgbUzzo`KjDQJYTSXF#0tHE$_@kLAJ=tds@V)JiAasV+NfpmL{?YC1IP=KzF~~cE
zgi<ibHxa*L#3RlFqr8Bluzw)BQ|#|t^1_xbg)Ob)BSPx1$Z=t<t%&!(ZX59;43h|w
z|6yu*HP-(pf%Pqdb(%3Kn;1x!ikRB-b=#$_JKgPbX+|yyfAz_5b=kM;q~6K&a$&8t
zsIr|$Z+M-)V|@}P@V-hj43+Fssxrm4<4;n+2^zsxS!q&KV%zXX;|L$tIjpOf%|f*#
zW3`4U3a484t7iS)z1q;}(39m_YNI8-@xE$Qmi>~gt0Y@fezpc>Vy&Cfr(Sb6CpO1+
z<8~|E$@W@Du>{?<Q;A<vZj_v2j^@sDu6f}UCb|M_XXC_roJ;bHd=rhQi{=@c_;2zF
zeV8*Uc`MKztAzoYh|!ZF`*ITG?e^!*F{KXz#X^&<1JAx}lf+mEiy^(<QxW@1IIsx4
zcw7K`YVuJ+0L%g5sYq~UiVv!3AL~^<AZK!{K|Z|LD3yex$@WJvCdiXYDd4^;uAReV
zlej8zWor!n@1fy0`#yomP6>Wc{#y#G1P)5{&&ycuGxk-qOJpCBO3>u2ona+V0-2Sm
zKE#oO@v!V_s*F8JN*;oK#B3S+G@S~h^q*I%GxEvB3LmOe_2Ia_JR|2<SNKMCQ=e}}
z2_>TK7MiG3^)+3NC{cU1={QImq{;rZQWrEQvJxd`x_pHF1*q|(p;^vMR`h!Y$w=EV
z7g<ObQ+&xO2>Twm)G#S6A<O>Oca854Ne~!pOlN3fU^;^*dB$?CR5c1W+-#I=Q;%NP
z@d<&mMx|19jL<_uzVE(+mpgdea3Hf=Xl3`I>C8bFm&N?jfs9)uv-{8?j7Tvm&0^4_
z{KtdJ7sy6^$m~y{@XUths(hidFQY*hM;?(^LPUat58ihs3hysdl_^HeC(X|V;r6St
z(r>!W376f`SYSW!>%NmDOC>LzRPp5gw#!ItQp0`>uGoLzXZ0K`My8+mWs1!EE5;+&
zJ?`FGg}V{lXQTK@;7+AkG2Bg{qV@oOg+C{1W!=&UgBiv~e??J%1wwuFK8!ec(G&Q`
zi2^5v{U4!f$0^efMWvuX{p8s(f_T><fIRUSpV$y|!l3^oGCqc?Yr^<s7>rbR1-fI@
zD<p117XKwrqTT5wx?D)auJPMf-wnDs9C@wmPj2NVkci}YDzB8S;9=xou_9h%@S)s*
z0At;vI`dlnKw_&j3|KUf<mFBqrG^<qJ|uCCTCG}layFhJI}bFf^vZ-t{EiUSDJJ>y
z@}%$_QxFhU?k)|nol5L}=)|9B#&~w#r}<T~oWKLMu5!q~L8I|Gq{m@6ePX{MI+&6u
z=HQSu@Dl{HWn4iv_#pimWDijxYKeVZUZ>-8P4)?WJ(LXYK!-d}CB@S*_9zZD<f))<
zxevd1<|YHLx*W*3^UW9><*nIDWme7y-7g&qE;FP``Yicu$fhiB0{fC=g4T1ICH6@o
zBB!4xLXG>HdMwmb5Nz6;dwv8ol2FIjlYzPCCs8kX?iq<IMiugrubg`p9ftYmh)C!x
zno<@yz-8zu(qU3ALFMV8C%vFT<r_*7X{XL)s(MG!78#sb$Yk5jO~Wr8WZW01khLC3
zO=i~Rr5Vg$wWxXsvU>hH;v+@5BlBk?>_t+Yv@~Bq4<cVus@3g<Y>ah~di<_SN_5;V
zBzwQ(^F!_f>n!_*|4xf!^-o-aaY-BdV~~gh4VA6FhN@yvxYbR!Ycy<YL~`^N0qeVo
zZS%fbFkBEE{W(>cVzTMB1pGl_Bkc%?ae@qiUXN1y{&7M(!G_pZDSJ#W_9D#iOG3jk
zpU3zxix^ik7#`!1=<`*mM9yHe5q)IX5dmnT-wi-KlD>1FKmELlOjLnb@=a6hf=Rrf
zx&$5(jZ8^zVd#koO+b5Tuus**#E1(`!I=s66dZ+JNwPHQXg3`_?H>ISj#k(&i{H<P
z-)F?{XT|Tc;`ej(OM>}k5Cl@K!3@HO`~wRyA-h)ad@MaYM|kQQrZ>lsjKcLH^o@e+
zm<pbg-$><rlL1}8_Cc4FfbA<`q&;!?U15ihDlFddJpm$)e=TH7<M<yzJ(YRJIDSme
zXaJy!App1%0PsGg<J17433LGfHIhjHzyow706-k|8UXIJ%@U@PbC_W>>ucrvH>dY~
zKRaE!{l*}*sK=it(~t>Tu&n3oPbllro!^DCpL0nGXW!E+XWtQa_JEq={oLa{;^&V-
ztTcZ9CDd!}=fVEaE>3kqF1`mY{wbx|)Gn?G=w40{<G)Vg2p4~ej)aSgqh7oCU8;K*
zP4TRMt(^SUuAN*l)U}(B2L+T92*Huq8)P8T(Zi)ZSASDkkv4)ZT>V=vDdFnVy>j)-
zTDZDaeY~@K+((@Kn-DCGv;Pz7wRZM!pJ;cdS|N9zhP%J6w42)9H38kzDcrq(K)d?@
z9SL_AN4<9UJLo|fJ*PHH^kPzKFU{bs5UX0iOv};KG-f?sy7uyo9<8i;0vPadKI#?F
zlhXc6o}XWa4qK!A{Zf~d@bfuXWZ;R<dc|tXab|F$dK^PN*mn%oo{S1E4Woxo!|}-B
zaUMaAZBSJj<hY64OHM?`3<vx@tWbkT@*9tBbVmqJ=AiF)s9sEsCz?Pao(xKdikly}
zmt+*V!+mrlz(gGN8cgn#U(V)=Wn;;-*9cE;NslK1cDja?y#aDrDdcpfWK@<KkT+MW
zSLr<)d_f!wIGY01HRgn_da&atE7L~OMS4+lNeS557lxgw1W{Cav<*<BABo`Up6Fk3
zEBIvDe+SQG^r<J+6D}*#<AYKqK9oI&s!5MWg4noKV&f-8rQp-#i(+~#MwXVvAXw}@
zPs%T*m+}kgr3|Je{`Z?b)kGdu#@I$rg2&Eq@CbTKeEjg3kHn8t&|{kT@fzy2jvpa!
z)gk1aiOff%$Ua1ova*rXQA88a01@ndHPM_9MZQOpiYW3H9f>F+j(QzMKB$HpIXO^U
zqcGCN05r%?*Ky=sL3%Qf<QA)SJfJP)stw1EjUVH|YC3>~FMBBTU&_L?fdsKO+sX67
ze|JfV2(mvMLAHp95RgG4Y9Ma3EERYird27xZrm+T-pRqFL_IOKY(&+LW6N;cl_IDc
zqtBk8Ze+k8vaxvB^|+6~?!UrN(!lOD&}D1b4coZJ-AovFgX~9vcRzslN6MB{1Ft5a
z{i;*oohMlZ@ZLd30(iwyuYvct!k{%m+{5V;cg0B8P&=U|k{N3(t`514$r(TQ+(Q-%
z{j^3?00oznK+@~NNUBq2ntlwXt${lM#Pr{Z&ALC~HrjF4{UEA#jEtw-t`tGW82$7F
z8Q&3s78#$h&<9?nqSZRAuhHCqdZWQ(I|7M0C?gFd{u=7FhD6s&H5Nu*!qBE4nZvja
zAZUO%S#)X;)CAIk;L9YP0Ksq2kpMw))N2sDvw)?DmUAkSbTcE0`2c**<t$@Pw2L{6
z06Sd+;-(NepiDOS>eqhnLkKNEC*ML~2@WZ>@8!tvC}Y#M($))Ce-)|}%w7{;xT5xw
z-dXg9!{ov0WvVya_5?6r56=j^o%bf)mk}bQL@!SKn@b}OXFJNcW7z2l4o1f1K?90M
zLy!3g8omw{r9s27p(Hd6^@zemMwLPnxYq;x{#12SYVgwpv_Z#afSxA=-<}q)c)g3n
z5%7089SQgoN4<u>+Y0n4+8V*$`_pHDf{m`>Za|T_ATW0jJ?YO9+k&~Hp-LUV9JSXm
z%w6ZUCt&VS80HSpeF<|)c^<rd*sa?&yv4B56L{MePV=O}#KW4$bOhGk3)Q57wFgkI
zb&{ugK*Jdo4#C+Wz}bG)Evey56G#td=Sd6!XDf6h;7lC#8qSUt%+tmETH)D^UErDE
zqiaALl=6tlwYNWU`=YWeZ5VAK+LuF>IzTjPuVX~}g4>=zw8LRUdyei)MAO>yAng^m
za@UX+!$(gbZF>aLr0K<Dn#Xkn)1HBD(qP)3p<ZiD(|w>pjY@|=?Jz*?S=BMAK}{3T
zw%d7*_Rl1T0JR^|kpMMu)N@e73IdD8r8P>@x&SrFNB5wn>0FRBZFo3Enl>4#)B&hb
zdmV#XzuTSwwWDEB8>IUR)Rgu-P`kpd+%>4h@bS|DYD(yMQ1iHsfZG4TmD7OQ4bW)<
zs3~?ds8Q(<s2v5U^{I|Y4QiS|dQiKI<Pe~CHysI36Gy!UwG&pcP%tZ4>gmC?O4mM+
zK3(%M(>1(}1&Cb`jH@U^({|EU;&wVzsRN9o_BzHm=C&s=?v^mdmFT|2xPS^hpnJll
z&^6G-Fw+yDn+j)de&dNpJCF4U+L=&L8npWo>a|9@KyPTsqbecDy9JQ9q`D|I<Y@xh
zlsnJkzD~jj$a{&71muaMo<m-xQO4q0PTn%@HA?2XfIJ^F-9ui0*ad;SH<h7jJ828@
zz8k940pwA89Yfx?-1Y?I-4=$tH|V~CJiiJ($om(ULf4QN!^}?y<oP9!hdht<2;_Ye
z3Q7ZcBO{5B=jTa79#siJ-fe)q*Hss#hCEF`n{wxnmmy&U<n5v(0eRx6*N|7lf~r_f
z(99JpHN5DC4@noW@e;i_mJ@3@;zJZ;tv*EgSo-v@9WJ{D$!FSW1xp1(Q911WgcZOX
zTKbrFs45%wsp(A3u$|l~b6GBmgaT5V)0u?#xo?=SR!rYZ>*C8Fbx|)&8vB~e742}V
zUrKDRu!m{~V5R;?CqW$U=)vhVE-3-Ax5Enq-`HSf<_oS+HC?;S{}TO(%ZGeZGVn%E
z=q6t3@m=xMcEXi4E*K=f@pv3nJASX{R#7SVTGDvXDr1C+D*=hZmp!qp^kNm(O00>H
zbH&INTLgkdu6vR_l3ucpua#tj5+z~;^53EN$W%d({m-Olzqw}Yr~mlmx6-r!m9=93
z;+T(51pJEr$Ob>BiT%t(<%hBNysmmJy{vyEWNmA$!0#ltlC)N!YUL=_+i^cE>Zjl&
z6an*-C!R##Pj2gkSUAYHeZlI>0!kKw*;V{!GnD0Rm0uLaVW8I+UO0{G&vCc&ML8{L
z__8PSPm*O$T66n9+waA((WL0hl6q0K(ta;qlH68`UX(1XK`(|DW)C_*%T?I-B@2|K
zyjfI5dJ>5IdfjzBxvy4b_e|o(<o1$GLbjJmQ|(lWyvrMtiuGiGd3AVBt*ulxe&p%N
z6d8G51&>QJ@?1oZww`QjFO^n9QhrSuQKR2U+>T-FdlmM?4r8;vH9Gelsn(V=i}h+b
z<5*_q`0>N{9L~@}`k4hR-Z?w^L95y*6*3EErdGw8{FYgm)yS8os^<!gL|5=5Iq=(E
zrSb%1tE-w=cTnu`SuSH><zJI~YS*p|-ek?2*l8E6FK00RuPx)X@64`Un)$>sl1nu8
zIGVB(c7F`MjQS~J%;QfjteH=mXrkDp7RA?HmN;>c@JZn3(7K#9_&gnnX#;Vz7Sjf;
zA;c$vUKln}0=3?H%xjfjb_ICe6+j%N6~Mptz9#^mK>Wro#P0I`SYc9aUaW7y0n|oC
zhJPok@NR<GV#p(#wGQfSrDsHfV9&^f)n??!u=N?*Vp*SHWdPk~2isUjxY;1b-D0qe
zeVT(KZ@&+@yHs1#e5+G8Dt1Y%&`ICDFJa$Bxl8qcgT?hr&138Ms4im`aM~aj?BnD4
zMxFM+&bd2YZ`4=h&ewccZG(W#z&FF+<nG=87t&?{Utioq3qaehukt~BFo^+ubKaW3
zR-{WNcC!{csNkevU?FJ7DA%w_=~%s5b-*e%*ErkceRJ3pRkA?4hMYoyQFYp=mvFX;
z_PwB8wCOIE9)FT=Sp)&>9ANq`pXZxszj2xyJc}u;GWJ#-x3Oh)4VyCyNU=d`N?yUj
zX#_HM?o83Ka$=7Gh=G<=1+{bZJl}{JGFr^KWS%ym$n|{GGO+NqIvwDf&9g=54_g>C
zFn=Zs4e)`2xzJbwy>VLRm$ughUDGJz$_P3lZ;~*?pPY6O21T1q@R5eIICI^M*ezs7
z@Mg}e*D*0?8I?lG#NOM3#40ut;S*Bla&L4&Wk67A@|~eOArkEG;g)<{()D=IwRZ$>
zgevn%k^xp{TToF+?D)CK&8X$hE$gu_Jin-J=BxOWwqVj$!)S)~5TECR>LEQeQY>gw
zz@{hTv==F6*z_g79{aCg#c?1$K0y|Ha!~bgaiIu1tXHtF1g=boE5gjF_q|Is+iVo7
zIqXb|Qj^$uw`$Xjr&m*03D(gg_onTBKr2@$+O(fFOov|X!KESXByhNm5XQ#tAaYGy
zSdA){TP&6!95zeePDf}2`=5#_LhM@wY6W!pSlu{-u9Bz6`uJw@G*}KcUA18SW$X!j
z`{ZsuUNdVs_=gGqMBD=d*r~Y&YX|S-6#Qb#accH~xj9_6XKTw?@24_btuMiswoysJ
z07~W@vyAR>#D=V9ZRXwYxf=Q$uP^39<mrk>{z^?AZflUqo-*sRMblXnZ2$`9ET>#T
zV<Xrr7G5Ow*B<6OVUuM{AS>r|1yK8Jtnf*Dx#AdSb5?Q5D&arux1V;LZSsv`z4)A3
zF#bFSjNbb3SbjQhDAn%FTiEt0kM5KH1}zH}qVy^38omP4sAAm@2l|&VumbC;=8Kg@
z*nq-?Ed&&HV<A*bem6}k&ImW^7hnXFDKD4>c<?efuuN>Myu!ETQR-CA#2(M39QI8k
z#6>Ti=bMQ#O$y_4`9+ClD|~}lIo+h4)Q2USz<ptf*t@0(-*B*VI&If1>}E5Vf@L`X
zTF}5U`-J0sOW-PP^&>8W+M;_xAHrc?nu-t6MR}ZMFH`7!k$yJPRvjbs^D_OsNIx&o
zjo-#kgO5-xvGWaR$a*WH&Wb3%BFbVd+g)^*2x24NgS5MV+F;4sT_C)o{y{e$>)^`=
ziZ)_u^E@RRvtyto*DUYX+xsHTB~GnFl1<ufSFkCz-w_)ti7nm4^7xFF&u6q$F{7o7
z8LeH)Xl+wQs~a*}{gBbpXN;CcW3-f!Sfq*3q7;l4tzh(>G^6jR8GRne=<`8FpT;oy
zbcWH}<c!`nXY>vxqjxkJy|~2a#U@4%^%*_%XY{O#(X%i{V-H4SlLp`9_XW{Ao3QP-
z4Atl{d4(WPmy>4Tze0X0uEKZ93)PbAUgF}`&_%cONPyyqZdFv;5}=RhRd<@eqHSSX
z>JHKe0CK+CAkajt#)i_g@F+|adBM;Tb^R&W&Ew5G0WNr^1ek)A+O+Zz;o%k+mjjo_
f+{?lx#)_4EsZjtNR+p$2Hr_z!P$(-ukevO0AbPw_

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/models/vision_encoder/index.doctree b/.doctrees/autoapi/lmflow/models/vision_encoder/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..0ebf1b999ce4913f57b6539eca936dcf99abb583
GIT binary patch
literal 7944
zcmcgxTZklA8J=7Bbobo%G8kuPGa<&zjyu!31|s9SAi)LMWTG*Mgb9`Es?&X{r*5e_
zwbSFe!8|1Fz(IwSZxZql1Dc?ZJ_}wDAqpBnM8tQWd^h<0=Tuc!chB@pL>=hqKDYn;
z_uKh)<@^7-I-&mLQXF!}@9mhj?FTlCQ(3bEEAf~w(l^p;ucr6YO<9l3!#GGHi=}b`
zJ1p+FcEtSj#Z)fRcHy}3PIi7aW|nXQzc|h-yvk=^OgDL5R)iy5R$7u*T(9j0hdUl_
z2avl?jI##wEsUd-&f(b1^Q=esoU!}0RO4F>O<*Tdz9Q>JK@Ojm)r^WgIpaD$OZSqN
zDI2B`k<&^9s3&Tt9Xlq<W-Zr@V@%?A#Wy|ut7a!*f6y|mgNU_hPa<c9*`cMBFYz@#
z&#&-lzAj&A^1x%wDDiz3H4lU6p!xJOw?T6L1q9=!MyzQjA}~XzshQH$Olb}>rRn%K
z>+M7_2)ue{DrxZvSy6K(zG{~9hd~g31U!5cpU3cd9G`1M9QYGCilj|B6Piwx%1_|B
z#ATn~7%LL_sNS|cb{DViiPTK^r}%cC7km6FIL_y^TJ`l}IiHY?uO%%FKGw2d9hk}l
z42WOU+|>}xQ|TlGNEo;Dc77cks{w^NlsNFhAo1-{rcFSSfh8hVR@5^X%Gc!9`Gq&Y
z|2_HSxlUTH6Xu4Ybs&8`HRZhHTW(^rw&N;sa;S98GkZ1*1yAoxLXXW*86>p6;|8tt
zSXQ}X+pynbxsaVMCTf1-wOGXL^v;BAD07Vi-#x<K2HZejZHTqc^L54223Y)6*wTO;
z<Wk1P?Ze0nLs+v}RzW)d?4W3%9IOB#8*yJ3%v^zoW;1mrScF4nbA_14Xj=Vt;wwK@
zC|YTzuV=4*G(CHzq-QJ2GM9zfN?bECtiUzmkXfVpcB+xisQq$z{~Px{bN}9r`}ucD
z^nMrgX4lDs|DBK7z46)Io1~v{;(2CtbPG-ga+0fA;DBS^DJ`dRr8F`?{GhvkqJh6w
zg7{%%UQGmtEZQs&46EI_zOl7+aA-!I_*$cphk?p4xXdW)vMM@Jkc1<=rI<4d&VNM4
z!#TKzj!pK{AYT6jynYwr{Auo}&hQ50U2*^~lmSo#Yz6kw2%^AqItmU)=2g%nqdW0P
z9f8j5z`A(ppp19Jqurp@-F<14(Hl3%x$T|N1Giq<-`LN@vu`M2?UR7^uW#&c?PoID
zzotaAiAXdbGT(Omj?qJ0EMv9z-^D7|iCNbXEHd5FRs63iZCI5TM#)M0rv3NRO&cLm
zp)aSxqf}PIqiwJ=Wy9|4Z%~WB47<JZCRqg&5Z_a7Ac>-%p6U20|A(B#NF0SosHO?D
zoWq#)Lf5ezkv>q%(0=mJbqG_3#Iw>LS;JP9w4{6LIsyA(B9&P>adM7zg3OhF`7?5t
z|Fd7YGYN)%06SZiSH^jSk`O;A3$Y(O@^F#oJ7j}@;OpI~$%zT10Cdmq>Et-!(V6$L
z@wY(ecSDB%$D#3QE?nQc*1d>(&w1B)B7Oseesvb4Utkm5)#;Gr4Ee+q$s=g|5q7^b
z3{qY$x$zA*@H^V~j)d^Har&*{(*rh8?|JTS9GX-;)WV3xEV|2-ioJ=mdBpsI3M+k-
zWxZVl=zjDcm>d)QDfW&D{>xuM-fni-##4bW;9ZLxV05AI7v1MbcT`S)&aAWhY;i(m
zBo6_popUPAV1iza;ZYGt<rhXh+UB*5O~>ERE@Euze(gW-1Os_|iJ&Z%pO7;s8b@;)
ziUf<aV#E$=B9TPpTw0#sbH>`cE*}wG6iBL7sm#jjQ(lqlW20JLN}RE)FY0d<f;&mn
zn{p0RZ9|n68ghZKQhryB2f{@6O90iqjejt;?l%eUTK8@NUg^ae^kPER6>Y<WofN{(
z6WzMsDegF#gc^~c@vqU%XQ3{~pngH2PPvH35R6?1rD;hxA3r_0E7At7ECq=2@#W#;
zTAUM_@1$VHFxO5wvMNUx`a*wiEXU-eKT^MRf2^7E4!*{@z(FN=TOWE$9qRrNU%5|^
zwLIx9__Cr{OEd<kpiB`&W!6k>ZRNU6Kl`-&n(jE&SINm>l$Kv=w0^G9dbf;LRcGS)
zRr#=g-Jix@*ZmomhXY5>JAGf@?+-`rF}yp^A^DJ(xWX~0-D4E$g&|b*m`w6bLj5F@
zlov}C&RvHcQb#J4OZiX3^esLhj<bpa$^i3}?1quGImWko$~8Z_PypcAY4>};xs+NG
z8UGw6LtA~HMO{w`0ev8JxS&rdDrtBCB%f&44W_oTn&q4PLs`-JB(KU^J8@lQX!`yJ
ziUEsQO40K}HTQaoejCkdzjEoMU`2tyD+pWNI#>hqU#6kT#;AZ$8gJ>qTvGNkxJzJ{
zr<5#naD08338Y+`I;jUay8a)gJ<HL%R{>1Uf<G!v?><aJ<tfW!KiJltw_`c$2l>y%
ze0p0~ZpU&yTg=BR*pZPK7n_#+AfL<k5hS`VW<nqZ&?*~b>;mUhhA~2>Tt>D<`Au;L
z%suM>7Hb>%U055FNg-+JE72yunYUoKp(AM6ZfoZZ8g#>!759v9?2g<yt=*Lhi0Ek!
zwvq<&$-Ifnuggm!GX2;!sl8={QP4qlshhYJ5eO88s)v>BQ7x?mdcr5=qLoC{A~uSK
z@O(a_8^f}eu^b)om7>Mnj+9#3s0qJrx(WJid6<*cAQ53A(mlR};;X~Z>{Z?x1KUJv
zS(sjkvTQC20s&NE$h}lfHc|O$6v#szU_&DtCXtKL0$CF^nQ0Yv_P)H(1^_=0tTg>V
zE>P1Qojaq~4k8Z)<YJ8WZiuFnN3HxApw{#ZbD5ek4jsXbyv2>1vIAMqdr!_sEDR!H
zxa=+iBd278n{F|z$Q9Oez(0ZHF3vYqWW{E!q=V}gseGkA0N^sy!%Q8*vKSFV_&Gkf
zrQ4aL*s`98_V(wtm1nv#xM;8_3L=A>zU`umh%P>8gVT~rn&%}b6$K@g>q9F;Ai8a`
z=0y!xsjk8yKYe9zVd$`2CK{l1F$arWRo5)Irr$|mao{*bi77N!<;93uL1Y`~#G)sH
zYpCD8C#x{LHXJZG;@G(=z`-t3#bxxT<P>~8lAOc~jY1j0ST*6GZi##oMckfTQZvdF
zNc#P$FlLDz7^uInYZ>i@AZAASq|y>gs8$@M7Jy~gPE1WZ5Rzdzn;QuXZXU$M@I=bu
zuzexhX5`&xxk|rq393A0mB^6+&4$S5&_0H!^bmdN6%~_RhSHKl`#m&1o>_ieE`}^L
zs2c+Nv=qb8QVOB%z@3ajevAtd#=FfXrsJLPNbtbl38D^c=`!t92w<lnm<RC)Rd%v)
z`>{_y0e&t<ZEFZT&Gf;4G6xT{HA!U;ShV9X(N=fB6*Rf<TwGR1MGPxaUBa4NhfaD%
zVVng4==qX{9LweS$QNeM;7*6T_&|SWsIE@se2m7bFkVhz)iD@#QH}M`a0AQKjeWUp
zairK51h4H4TvjZ^=>zmWkJVe8>ey#)0PTI?>-cTxK;{=NkfT6jeQ2BGk?y*INv%6V
zjGPfv22TMtEcgf*aE4aWvAk?y*MY&%$Z`!-V&riZ)+^b&A9U4iYj1Wer<s43UfE<#
zdlQ&1G!b#qfo%x1hp9oOUf5QE6f{fVEfKyP&l%*$<%NM+N-NYfptXwu_1Ngi4%=sG
zQvcrU?!v~qPvRfH$9vtU5j4ABrhhkRzKMT{tkZrKU+LvfojB(Aj@9O4wGlbuBebO8
z%=4S*(^j(_E_eO;J67>>!EYAPHjk`$po<YO(+8BMyhk|c>}LiY5cOz*60pZ_DFpc|
z3Og0sRo15J#Y8TYl}nMkfM%hNePDr}As>?O(|j$@km9RZD5KZHgT%nrOGA^z?)t#C
zdVZ^dbYTEK!a*eH(dqIqtZW5uO7L(N9^RnF2!7~8xzGg29D#QBscilYh~u^NIan+5
qOV~Zc9z#L(g@vOVn4T-9m5$A2?^@954n5#DZ1TmK+tRt_&c6Y)527;w

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/adabelief/index.doctree b/.doctrees/autoapi/lmflow/optim/adabelief/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..f694905001cdecd0e62ca8a8ec0d5af677437894
GIT binary patch
literal 23759
zcmdU1Ymgk(b=FGz-qo&FNGpgW7!?Ms!oH+nFl$M+zy^arPz10Yu!r86-tBJnbdS4x
zq+J{;c9p?exJ4Y({IH4RkV++0HvbZw1UnWE#Bp9#sXTCzRNgUFajKGXC8^2}yOQtR
zN8j$5p5B?+RVb@Uv;DaDo^$Rw=ey@V`h05Og*P7C!T+(zux{Dj+PqP%dVbXmV>Vp%
zD~+1zMe+0Tu}{ZO#xrarFfN9EBdC}$+kp}l%XX`Q>BS$5*#s4jY&V>j^@l>U64}0&
zRJR7KL2Kw^@r*UX1|mCh&D1M4?bcRY|Khw~kL()QFqTc%HdkV*feQT(nQO#J+j39E
zg2%zS0gj^Bnqnj6L<DOW8<d=%WBsn}nen;CvcX1;C<^T5Mg)mT!d-SXBx!~!t`UZK
zi0c8*sEMoLYNM`imW|5!z+9m+4K@@RtMn?iCar_kxOK?dWgTV@Em(fdTnHMTX9f!w
z{owq<z4zS-G1VV17%m9L7K}#Z8+Cg@bi@Mhh=rsh7HqF-uFVG_M7#Fv&c?E}gAMR!
zsa_NvY+VOYtRv{R>+$~v{6C8S$4D&nOki_iW>|mSu!ERgTl1?8*L-iOjRM$VQDCmk
zdu#Aq6dMieCTmVhe$KiU{cDX0!_k`GqaLuKJ7f=Y1yAdHsYtYSM)aa+%V;YVk)}>R
zEs9&S=$~QUKYJo$x#1c?x#GL!ux?hm=&F8Rp6z+;BcD9}<R?!&ss6@9?tV><nO8!`
zOUG|sI!Oi*Hfl8^SU(M=qeI!WbW=FPeCl<~rc!rhh@Ew=dq?r_Ox3uZ`_$;@-L<-F
zl5dBl<Y~!pSN*__tlE6(KEG<O*k-k~68JSLQY+bB>4Ei!{GehfmmFv&Bdu?Q4MwYh
z->A3rHhKOKbkU_MhEe6BU5%{x>?rlVhu%LA1^ddFHZ%KF13!~)fN*HpudcT+3O;uQ
z{>7H(9J<Jda;uC0dW_Y;s9ULu?AE3F`jWrwEIqEf`eeQ@%_k*JKmHi+<{5bTc-{1>
zwzpbd1KOlpzNNR8mHviYJ2C^qO*LtKNqXZ*!bz<C@ul|u2sGi`v7Pnxm<`s~=OA%x
zroo3g{>#wZi@@EszRU(8BS0}`@9G2&V+=`DzGm27Y`x5ea2KuDF+A-e`fLpM=33pg
zD|QrL;;%0nffsME5hz9ATKge|4Wker2;y^`P-FzFW)!nFP^s0&LE;M-0RG>=;cOqe
z_60b_v^A(T_Ci`?X4_sT)jV$<wq6HvQahicYDH?I3eTb7vmp5MO<pL21#5tv+6I`J
z-fAnlGu}*hMtc)fo$;9LjKAojGm5)-13mWiX5IQ36au)lZ;AC2g2o5|padJGSDKq-
z`%{BK$@RTefz(E`J62Kq@#fm;LTqHM<}QVGgCf4+dSHfTaKYr({c+S)BQClMgOg@i
zhfjtILDG+8G(!B3q7@k|CF6EM&@hRC39(A)4&RGt^b$}u&03x(T6Yfg?by+|;t@wU
z`;r=*r9H6m`6e6hEWjEJf!Y{|*-Z(WmS#Af&}f=J1RU!GT{0n8ieDLZ?O^?g;9G+o
z8cZhWO*ym?tVWF>6M?^$a5+N*SuLC`xxuN$`NhTKv!!J-GQv|ci}Sb5mhf}iZO3Ly
zW<5N0+?+dp%WUbQX|GyQxoTF7^;3(pCDg`1SUq*8;f7}VHNpbdJoQk}z|$3b&8*5g
zvS2`ShLv?$plYs~2%L=w#Uj5Pu2$uXdSH7|xnddKs#$jZRsQ%`GGb10Aq~&-Bco{w
z!cIJT^7;tj(^BauSH;oOv!#=&QxGJ=)R#(0k<!uArx82zS<5sV5QBj>t}}_LBwe3$
zuxKkaxU#;IvR7Q`M$R@hl7%MLcQI~uT<318a?m02X}?p>rs92P6a}0yDrMc|yiZhm
zFa6=u6K>vnVcv)_M2}(eM{^i8;RA*r$T!n@B}v=&TE69kU|u9Y!^U8ZbHEl(n(Zb|
zct7*|ObC&c=kWr0Mz}o9!la^_f;h*CHeq;}7#7?P6P?W%4GI|D%NgbVW+EDMZOrdd
z-T%sur6(qiD4ZdYE-e|&(}5Dgj&=yO6E2mTm6j<ujo4Du)AsfGLUpk|uhoA(Cp5zF
zr)c=g+S@at($l<>^DO?TQGpFBpcIljzmR^p(+!lj(4}VltJEO(#5~7vzJ%8;o-aB7
zb-d$_JH_pX`pvHsQu;x<llx97%DB>mzg|d~_4%zm_B&MNpd;yGif4O#6(wb>1AM4Z
z1m&G3`X|Li;rA;>!;Q)p47Xus8DLk1je6Y=qH>7IaP*0WPwO=s1Zsbj=1WH)#YkA7
z$-S)uU7k({c(FuQai?W{yHM6b&HfXrHkO@N@h35_HfsE@Vk+Op>lXWwdX!ZDI89|d
zJ-RE=vJbU{J9ie^?suDMZ>~bY(l)4YB#BG%G`9<Hm2mu8%XF|<rXqhlfT|s8?N~9D
zBCSa(8LbWQ2uw*N@3c&}7E|j%PlAbU(9=XVXv|aAc)Ey^G5b)9`arR$Me4h^bM<|^
zm`aiQB$bT%`WBT~xYDw?#Z-Dwn;^aoYTKQO?enxYBugmCl6x)RQ^oQX>FU#+>+1Pp
zDn+`IR5H36;FA<3mAun3y;MxC2b~Efwn1n6v>)jSSf1)8n+hsnTb^sNUo959NR?mf
zT$SG`rc$IzNhPDoewu+Q;lz!W=*?nUJt$JJunmfwYMC?VX>lx7N(q^Hre*z)Vp)r{
z_ivqR??=T{inJ%GWVE-5BaZFb?$|^yu?_kgPcLxfsSYdHd<8_QExFe!dsne|MJg+y
zYKLLUiDD{6Dw9+)Dl2rj_ZQRZ!Qli8+n}%mZPuXjbceZo^Lmw1W8N#R(T^6(U!>88
zJJ;xHF_j{XN-CXev{6i}2aO6Awn3x&a~BWu^tq?Gq|#>V3$4rNiX|`7<uje@@)wJ#
z6zNh@>0FmzDW=tfE(Hrc>9Uhb?3x-D66`W9P?VF!C){FfjV~H}h9=Gjoz|65%u>ZM
z7}{UIV4D{!I13ZANp)E^yo!}sV%co#nf52uw&8VUvEvxFIpTq%CMa!JW9I@GOtwO;
z-=>9*KOilnc@}F%x<8x~e9ny{A2`N>1)-RY(mEyfK{29aw_@3#-1@XGt2J$FkPWXi
zT$iuD3Hqa0bFWbU(LT#%{_JUNxbmmE81Z%+_2*%+oD3dsp<BL#n5-;U#D-03ZAx_9
zjHR~=7jHLk;%2Fo>}q2%t76FmLHsm~{e&6J-%kgG=<0Fdvlh;d$eO7M)W0?vt@XuZ
zHRV`$1gu|HHJ$&YM(I4#G`nsS=2EQucM~hw_5B&^tDA3{J3l}hMJtslV@Yp`JO7J%
z#k=Agsa<jFo7QFPHLhvvTbqjr)mzx)=6c@;y;GkTZAVaf+GeH$I#m8`*)dqi)K>5-
zTiP@pr44^<CD0~LvYk~*Oa>cMRRi|pt(L+F3qdfWu(Z5hiYya5Ty)-ycINzLzRg)J
z(H53X`;4=tb7mvB`+=V>omf1P%-DY%Sk(j5fqvnY*5SO>=Jc*_T2EJJ^~BHb)H0_6
zu}O6B$dOV#@GB;c%vGsBy+*CRj$<69BS+MjSy)DJ2~tPFWLR)MB)cMi1Bs;VNSfAc
zuN$_`BEjzaTBEqqOq*RWZDPqeLTupv)MlmUTS>&(JzeD*N?Fdk!J5J$37yVT>KV(6
zF!{cv6Q+teAEGW35SCX%o69|g<}%~_99nfgg`a{QI=S(6XVMgR{W;7)-k;(LXFJto
ze8#ya{i4l=>7z+|d6w)vf+kwlz2%PShf$U9o9s*WP3jG=5;a=(iQBBXpD(1%`s&%8
z3mP45s&dc~q&pJui9#Zs2skY}RqK<;r#G8);846*B6p2*bod>>ch)4^9x%$Bp6WOZ
zVuup{%JZ1im`B|#10>C8OMr9+UBC~d%s6j>cjvG1leONzqY>${^Y`?YVld|)=<1hH
z&-q9Gilw%Yb6xJUvGdQW4sn$Vjc`2HxkG9JF7Q1R>#+N%p?@Lzx&Z|YSvAgxP+}xO
zqvp6z9+h)m0UvA#u^~S#=X{NB`CfMV-t*_g0b|4np06$^)Z;mE5ORZ!%a`hY3MFK~
z7bb-!JzqUj<)|g17n%_b$s#tt8Qsq9RN}fEt$I6xBM|*$)1qdVaK5-HoHw<<Wh`%5
zZOsAl?~)}6Aa5zIz5Xp}Z^~SnV$W_+{I8OxIEtUxDvEzUX|HP(e;ZXn-oYQ`L97j4
z|4%WM3wYf!R!G2>R5}@v1_z|k=EhcqfGwzCM$x<qO)s_p*g5LD@_>Df2JB{^$VY3<
zWeZyW8CMNWXS+k|pM);*(7MG-6jW1%Ca69Er2TW*|2;#sYC!p3S5WP|3i8ewlI81k
zbvs>sBf0urTx~et<iD5s?`!<`Tm1L;`0wlVOM+biNWEDuQ+Elao5s=e$?|vltCD1<
zsJe3Va5z~v&w<X5qAw+MCdD*lK1y%fp33Pe@cnSo6o<@Pw+fkV(q7k)X`yO|koiP0
zl_JQLR62poOT|=*f_OnC2QukJT6I~EOHXmL^GF%v=BVq+L*|r*OlAA|P^r0VfyxIV
zcQ1kM%cz$Jm5Em<h@^r|5P2&g@*&yZJwv2wK>2l6VVgv=0wP~088}3~L023i`PDlO
zk?%=u!gdLhvztboskXNqZDw|cbKvrCNXsJ9%oNp-`M1&zlp41XWxka(#Ub-#Gi0W=
zemh$%`1hp2t}*lbsM;ZB{;-%z5oSs%onYqZU>4u9{ahGk3M#<NmSZdQW(%O1qp&Lv
zn)@|qrad7aI5meY!1<Suy%*q|ha&QTGvgl$JgIOKc%B4!zA5{?XYf=FD6i`QJU>Ws
zaPYj7t~hw|t9Kkc%Vn$zg!p{JESG1Fb%38WGqU_DR$tqEr>sBDn@_jBW%Rr%-P#Q2
zSbm6-S$JjR>2hjOJo``sY8l!{`()P9yNW>{O`76>dIsk9RC<kdfV0dQ%ARRYzsQ_t
zPrV>K5%o3ijf7w>;1+Fwe*#sDcDb@q+cUIMbkM1FT*>);tf8ehfOGoh%465G#;%M>
z<b#)Hu?4&yf#AJ>*Ndo^2VPlYl<38-&S!V#7<~npUfw(dU3>&hlil`=X{v!jOdAxw
zo?Y7%(co{8Y#h*jldd?R@vC<l(2mo$lTmj_cH^dDAa8Wb5QcD{^CtGwv#^B|33#XQ
zO5SN_|5K@OrLQdj)gLBJaiID@Gf?e>$?-6>-N%<%bdYB}a$3FeE<UM!Ms#1kZ)@IK
zeESV_MlZho1G4&@tqQ3)<<yx;r_g<qH-7-${B5a`p1oN$psXV|LMia(ks;;Hqjbf+
znP0ur-n>Ys2fB0S8#Z-jYIDoJyq9xUagjJAFi)NHJu^2N=@-PrtWMl$r@kJYy-m@_
zbxBj)sqbob>hUzEIYCJ4`<d*=bx731MO6ru+-q@eOL*uyAbB^cg5^#=#=5JRO3{ft
zNhNzCk5s3G>gP9FqKAuV^>89ju#ol7_UmOb@;c6YCky%cm0BXN9J{#kusoq*nVWw;
zDr?qSQ28eya4)F*v#6Jc%EIv!DDTO#*BW8n1x!99RiBH=>C<!z517G<AJoE<Q4)Cw
zO7aXHKkC5GKh2Mx308TIuW(7(TG#coSw1UyQ*htjGBDUAC&BK4RF@K|ka=fp+4U>u
z%ls?P)PX_fN>@h?1;ceI#nCJm$@dKR7>;|D)ceRXqEcX*^d%`KkRd<bg%gjaJ6gg&
z!>i&gvLATj4FDvZrvypqf1*1^7*YwW*q%*U1jGSS{lTj|`Z>aImLyZAjJ!IvH0PG$
zYzv%s&NU?|>c_X_85VBeIt$7|wxcO+G%t5!G|pG3v^YU{FYM;maEs$SBWzTwAl30w
zL;7&x{@h+vg0*?8dA-~XuZe-*2S&fa87*`O=T$8mRS~kahhz%aepOeY?s%yKDqqf4
zd9i~kU&vO`&g$HVEyiRAzVL*fTtF@?BRU##d=CV>J_V@QlIBs~<Rq#(P@@1<fmXJ|
z<rYLijQT_HDrTHNqAMPw@~iELQM1gX$5zt5JQAdiTNg?oQ<r3qJMlf{(P1UP7+rAz
z_*Jh0?4xxVc@RPiz$V3+^W<a@4$8^&gz8mE9T38+qatG=4HV5ef{>6$H#=jUR4F0n
z`1m$wB&j3Kgh0FGk}Y(0iMW>stH`m~-eHq;lGZHS7X^kFx_n$M;~eZNCVe8E<UoWB
zG;Vx$2OCvo%qnghBv&7ss5AmfR#{F`Q;w@AB1I(|#^j97Mj;>P6eY9d2{w_GkJ2Ai
zk!bA^N|q2NkjDo7MpSP^@i}V}Da+*7Ncdm@H&q%6BcoQw%~;_35m?2zwHC9!1!PJS
zEYLR-=TTr(-Zlakcf09B07`dBZ}Bm}b8PPl8bEp-Gxhiq+f6AyarU>oMv(^+u1$nU
zH(5veMSf&BL{kUF6H}%iEMG*z!7@+x0x>{IR*<<N&oO*3#}hxe<^>ZSxs#1r29iX`
z+W|IZuG#26EEG3F{dAyeE;m-u-UOw4L;s;fU>Y?%8G#}yCJ96Q*&ceUjPv41c!(2d
zY@`vb%)Muhe?D<2`?72XfghBSbExW?$eTGxtm4ZGHYxf%)ye>GC1!`4-ZVAgd<YK5
zCPcfM2UYt}_C-@=wufYZ)rA(i$mJQjcDqQ3)G&|}Nqo1m6RBC*RoGVc1In{Rc?3a*
z@_(LVgK)MLd?kV&v8&1y@D<1e<qj+BM{FlPNI`lQoam+{d0~v`SpEQ6n9Km-7d1G^
zpKvqhacA<V9-57+Uq&V|JlcbFe12$_(>1x3RFL_@4sA+cg0{+4JERB?f=Dihl#{^W
z#vx1yPQ;;ISdA=JUa?&WheW5-bOj<f$j?JJBo{(!Rj7OnIc%XSaeKCpP4TaUToXys
zELeXHi30E2bCgZg&3ZZcn1JL%8pb-T9lVoMu&b>os)tJp3wRvP*VnQ8;LZENDtu`_
zmE;VdWErs_)D!UpYG!@zhQ*uFpA*4KrKx#(qMN@_H4nEn$Yjss(-+%}RyYkn!Gaal
zTo4;UZZ~)l&yG3F4#OsE<-|D?0c0|&4Yns-_abAhY$0ElOF6q@q{BsqxR{Oe+$tjM
z4$)CprE;wx#};C4(Y1$>b_@JK@xtHG%e)JbndLlkes1tgz&yRLgn<oMk6*F971)5p
zg}nq6VPm<em|Y{=b$xtwK^#$@-t#d`rnZa(U}3Zl4lEO^ksEA3!dc^d*+k+;w~U;{
zgt%bu5}T4)hheMkia@gsw#)P`P`+O_EYJk*3roa-R-2AoAYCe@b>wz41*TwG5r7r}
zKY~iaake+}l)io9kI~vnazlwLDS0r8579$$J>wjrLG>E?nZmf}Owi9hev6+58=<oN
z5G>^%=A}1y$qil#yA?<1IW>Spc_$I-4$9BTS|%dNy6DMdWL3i`G6jbie&b6NHatgd
zi5R+HrQgGJk~p(YJ5$6P4dMKIGX9O6LkFjwJL!iG<vDa{&!ICX4xLGHXcN_;O;(3?
zBplkYaA^I~q4iCN=FASwsT*u}+Aeud?M7ZsF>Ipu#1n!aRZgfR{e*mmKZPCFmXXyr
zQ*>X`!=&^`rr3z0%3Io-X&)m=5XHcS{mpI3HyFwRjupQ_(1&po$rxQ25zGZ+^w1sh
z`SZ}`Q}KgvbbLPt$biL!ux%m;6;6cfSb(Uhr^)cG@+7A|hdS`E2gC#cd{{##u;IK!
zOXpu}_g>FhUL}3EJ<!$^RASb0(D^OYC=7lE<oLS_6HT6Ob>M>!K6pRGRY|r1(jv?D
z*?Y<4ULh9rG+$cZS<qm6lMzF82iMA2@|Zj8nC(^EMwKV~BYekv4}+S-f3f2<|9{T#
Bc^d!#

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/adabound/index.doctree b/.doctrees/autoapi/lmflow/optim/adabound/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..b510d0ba2aedb3f218f88386fde10d71985049a4
GIT binary patch
literal 22071
zcmd5^eT*H~Rrh-L?U(Ic@5Ww-I(ElOQ?Hf%vSZqKYa86O39jvRaZ6g2Brmh?&F;*2
z-kaB%dG>D8h#(~)n_i`*nV>ch8Y(1;3Kaf9P^1Ke)S{|FNR+Aq1PCOABK{~vNaYU@
zzjHt4&b*Ji@2xixmi%_+&b{~CbM866d*<FV=b6$gbC2%f|JY>QaJ~A*f?ciF!>SXf
zY_u9ynl-1Mq%WtZeldMIon>Q@eLD`DQN>Bw9@MC~UQmskdir?ECa8Mi1@VGxKN35Y
z#0%?LbGPIUyCaXMv+fuxC0-IZxkv19P+JSa+Y4bM@oE@`y&5*_)sz~b{?Nyr4H`(d
zYR{!&gu@LR<4IC?ij9>s2HZh5EXTdVh61ndq$|x;n~mE^5_zl51pE?zgI+ZzUPdZ`
z9mgodb*XOG#MNlE+0ZwucI9T|tWljN8%gYSdX&18?ooHYdyhNl9%CO{bi<mn7&Ys4
zCtAE6MmHC)UV9i!sy`TDyeI~?Xg8D4Zg`7A42xV0i%JZOUcKsUEJQIlJAP$vbJg9$
zO1vy5LLpuE-QdJM0jb@K|M%hlN&G)WTtP6A$9b7$Lk-)DQg*BsR-1wI0jsn0**)U<
zeCNmR@Jf=}P4|9xUbDU89*1z<eL`pSq#slTY~&#+UOu@q`d-c=`nn`UDEhLy@<gOb
z;ul4AcMj4S<<dEr*sIOJj>?rVD8~(_(ubgiczt&8u}?mA`st_6JgxqwMDDsK$Et6`
z1TUXHyL_IMA#T=ccC>jBGY)yO8EK%fgN5AVluhOC%HX<aR&(R{cd2TBkgREZ{Bo@k
zIAqkZr3x)OSPvsFacc|8+Dfh6NYxr+!^wISHXCiiB%2<AsQZ-kD4N{%s)?Ik87JY^
zA>5mozfkGQb~>Hw;8Xby2w_&k>So(OG0wp#yxmsLL5U2cW0Nt!ioG7$4L8@2on3Bh
zE{CiB@{^e?&v!JG1@-*mlaFy3&%%E9H=KIatFM<g03-QsAJ{>+YUV})FL5F}$o1#G
zCe3VAjUv@{_8Z;p3>d*j#`ZQgQ#Rb#oCm9^lZO<N_*XDxzX~wzYNKoz`~m4wc5e?D
zmtv8j>NVS|r|$2t5!@x44J<f=M4s)#y|d8>yo#5kFYw2=?Wmq^u`x`BK&!5+1RF&i
zj_v6Rr<2&xx|5{rxOv*R;eQ#3{!SVEl>tcg6&S*dJFF+`mHcEmT?J0F^0Ir({d0gJ
z_45U4W||>t@DeJ14g^2nVtI1ucT4PeFW51&Z8go#_w(#@*B`~sW0IY}?t>k(khZ|{
z^V<pLSyTcfb?=D{5$>f}l_5qCwbxp^WQTJQZ3SU{UEr(TYD-nLezLW7z7iYTs0GV$
z!zK@Gv=KS66WwySdOm@+Y8iy6uw>{VYlp^GUO~P{ItVoWar7eHmuxXLtmKo|xcgu|
z6xI`3R0Isou(tAS*8L*`d-n7`@mSpbLs<(xq=V4Xg%&OC&F2}80k>El*(rsD)+|Q@
z%d2;GfKF`>4Fw-4H&*Ee?O{XkrrW&_S^*~LNjdi5IwsAi;0^z_8sRLhJhgbv3Zi8T
z4zitC7p$d)rKQtz)~b`(@rBu?g>!Q#I(P2esX1%StJ^`jSQ%B<Z8#Hc9XO3xmz{Rz
zmmZw6Zadz(o0O|g#op8pm*y;tg1h9))@m3A7{$YO5Id)`^=y)JW7q3pVz>DH?99)f
zzc+#BvTU8?0zY|i&N{EShkFKnUbeC-*2#+(;j{8rg&9^7fUPy?Ql<}?j5A4!zH+M{
z8?cm8;!1aNrKOW1{;+}DX{&Ymm#N87pUCHpOIeiiv;A>Y@b^(I_kRDUM57PTAO2Rr
zb$A*&3=cp^3~yZY>Y(1T0JSrnmm|eUx5t{h)r?#x4|<mEgVN0dJ~&}^fX2b4%<l_J
ze36?c4cv@zZdzH<PDghy#80*v?kEjcjCGW#Y&XovFwAK_Om0;UT&2LnJ5^5fU-^G-
zZ<)CXmCyLd^FS}buvVcid&Kk-%#>Rvr%-VQKAvc&_2`SniD3ivb}RXUkUhg6(@Iv<
zs~1J1=XoRlbNHhGo=p|848SKu;Qw;I*4`je)<Kt=gRfK1-0(VCg8v#Gw;8%*;kWRN
z+vy>$G9<u8vtCJ>ZSXd*tcc@E)BY19Z8p$om);*!lcT;M?SBM!nIO6V*>{XY{73P)
zjX{ZIl1f34CBDW8c3xGq<y#oASUlA%{=JcU(RI;&y?CY2uDyi#f=ez*afhGYD!nzY
zQuOlOH}n2Hc`sjv=zf3^wJKf@)7+D+e{AHo2U!F3r!r*<8+cPo!YFUL0~az)x{%YP
z(Y+=uW&>fnCJj@QzBK8Gkw`DxVSm7@^X;`9qL#R~qS%smT7ym+sS<dt+07s+-?D?I
zQ)H9f6*n7=FiOfXc1n^@H$&RU+oEOnzC8EZ5jL;Y1{9iZ_q8}jvqcyx@oE{b;!g8=
z$;hiwOD~{lbJZWgA7z!Si?UodQ@IO|+Ymu&i=^`2JeB=<2UD?DEYuu6Zl-;vmG<^i
zD2B8PQ+T+eYusrTM^#P5tuD}<-ZFD)T38}9#aw<9Xrkw`^Ahr`naa1&McZ7;HAGVB
zWexeFnTlx*kyP$v4f#zY^=@m3{}y=H+e7}p(=F{CsXd*YE+DV-*(tsc7)`SeLBC;S
z)d;bYN-qHLT{9IE0QAAj-<qlPpfm-T8C+Tro}D|x__a2*e=}1$gU1CU*d9~_o4Wui
zhdRAjcfynsuc!b;-f0#G22BgRQT_f0?DvQ!lV&O=G?7$#L6c)<DkjB{RPF>#7LC;3
zaWuIf!|BT%?L*L)jjS5cL{jMmO+I0!VnUNXcyY{BdQh4Inr!2A#%5|eKor61E+EQu
z`=(qcJQ+kag(m!3V+k`eXYx#5K+_(v<W(~j6P8FSy<o}L%v4NjA*tL6mVDDl{T;`W
zKjgF1hiB4<pucBi)rcjMN-tRQ&t@tnEa`)nADXH3pfp7+;XC8X*h=C~o7?}IsqFwu
z1gpD%B}ck!=63>={rS3z*<zvQ`N+_Y-Icr0v`2`!&rHPxF_KCz5c7bUib-1}l{<l$
z4;!h!;}CO+&rTl@(}$qTMplgwBdPQPF*P$46U6kv%Ts15Jt$29Vg{C!#o<bu*_N5g
z4xmM_xC_uS)A=~C6S9n|s)|QlpjrK@nN^e1^6O~YBclAinTiQfB$Zwe<qb0xlX6HZ
z1w<L5^Hw3d_8ocgP|NYJjI?)*CxT4hoAKY}Q_}~W^daT9jGP+bL{jMmPJV2rVuF)C
zco{Ak*V7&}rT`};el)MlP2Oo8Ib^1`16UEP?g&<TX~ce_h9dy4Oh*gl?8p<hIJ@QB
z{m)|a4~HG?r|j4Y5^>EGHhIf)ZdVZ8oU%!ES+?sHw{QfrUBqF2i}83_1f0d4qjnKI
zjQvQjn)<hBb5g~@xqn0_6sKtt@<SHqlUf}988qVYItWb0;bY3i=@=YQaEPle#^JGH
z8TjVD$@f9sVK%zf3<7@MD~OMmldwXvqad9(dD-)bwd19_8V&3m$?}*8Q@{Y<N4GqH
zEQ?sEh-jbOxs?#wtgEBNmOf}BfO^@=Vi|GJRdHo>HxF+1KI23S*D2bYuAUSt2R8pk
zWXs$-c%wyO-3$lWss0GKud5w*|1s*6qRVI4ySEu!ZYWQ*45f%|v+mco50~^mg+5yM
zgbK*4H6<Sn>APsBcf30VL~i8*BHcIKH{CzwGwuG__AH9?Q!(#{Fz+ODQ<w|Yrzk@T
zxu*KprC@MkuC-vdV+Q?aBo+5qr{ny*5cA`^yW9AlzMNRD9V3#;sauUGY=p5>rD!mU
z=!I(Da;!%pyXqmf&AR3!Zdi@2+X$7mKC}s=@hTP%tG2?m@v9D^P!V2c-LMm9+)WaM
zRxK{t(S~<xA&k}+?bUek^tq)o3rpu7Jo~`-3XUqpiO($b@9u8H$SK}SEu15ec^^iM
zaPJ!85%K|MY(N+jI>wcT{V%ojpjDSXd^m%%)z&WtBz)21(-dUr{|a?+)Tgmvo@dy_
zI}=lGa2?nqq;(LZgjmPiak90YY}K;2eMK#rZ*&z+n@q?yn~|6Le?UTjOn$dTCc|=a
zb?>dbX^d~}J+HLyDk?3edo>6vH_QB(J1>!+zM<q~(dxFGF}@B>ZLTg<!rJ21W*Af!
zRZt;)6Oj8SB&MzQbl)9j;C-;QX>N-DvPH=L@1a1!F*~8=wd<(W#P(2A%l`W<M77el
z%AT*!d}e=uHo2LlarcUZ$1!Q&;q|){qOSjfh9$t0n&iNfe@9U(e0PP2z@sNl5I$8L
ze3n$TRyVDBv)0%|RK0cLgp!M~j*tjQ{U<2LhH(92VAg-oIQ97Cke6WmI|{+|m&g%%
zU=_XrVpL;mwsWAP(o-w@J_~`kC(U&iRHtb%B3Fr4tr-4KVhBp>q~GE%=gQcq?ukNB
z{W$l~KQx?q4Z5CIw;6ieHQa30{{{5rKZ74bkfn@XoX)!9YwT&f;-0~56o=mRBY?tD
zc&~mcn^Kb3((LTE>cO?Stp_k};kNn@VZ8o_@KfaWXGp1pr^?+E|1#BEL_7Z@{1MYp
zqq*7mrPRNwDv}N59M#j%(Mc&U$Y}-DdXy7&bd8AW4wR-8VR}|Lc4L`Jwz}Z*n-%|I
zjDwB9P3NDj_>a&nk6O*&*TZ@7`5rvBdZ;cZRPns{5^0O=mk-tb6lzHKJI*Ri)<adM
z>QPTbFLsgyUWs^1It#ty$|((6`cQXb%I_<eK_W)Pw^J_9p4!%+Yut%~HQ&FDHRt=i
zV~akT?`@Wk_LMb$(OL=dB`A+GFBm!7pRcdrwOFW4-j{l()TRjheVP1P))hDTvry&d
za^S;CeCT{Y#UR=IgTmfX?g892d8fzzOh&MiJ8cX*_$->5USf*t$cv<UNBxbS8~KYc
z%KW>1yKJ)e(a6E)6ZJUn6AfiktKSu~C|2=MOW}2+6!I?>h1apSz*F1nSblpz(CKy4
zj6)&dh)1#hTF-(a0DF}fIO?~H=<Q!I4rePy!Dc$I8hU_jtA-60cCu=G8|^x*8bxE3
zD+fE$aj>(e_`6MRIXMeqR!F9N-(!c%iRwV%LdWG~SO8*iE%C1qQ~yXmSLx@U=;vB?
zwSucH|2qHuDF6Lg{(FP}evJQqoPJ3)-UiHHh!dyL57yti4c1eCcZ~7Vd{mVH-%Ur(
z1u}p6X;Kl|hXgUP?;~zB5d1f!woBgIe`Q^9D1Uz|l<&`vv;(5IKW_;&Mvel`s#>11
zd(F<|h>0B|y6;0%47ryb@?&NyrX6xgrLcFML+%;Yiq!X;Y3*R=S`49Rn|Z|VxQvZ(
z=@b|G&W+lZu?~9iwtMvlGT7tF-x-_rz}m3+htTDnVDk-#sRK3(t5cXfSr}Y97QY`@
z{3B^h9bNnUuJe07>O^Z{REsU!5+8e7a<7?xvttqeGRwgX>#edjC)=oe`T3_+T<wi6
zWBgCZmUwL*WB5)I@#P+UA}LNV^yZfIP{tU(I4aaB<E#E!-oJVLQnb#Kv&75e5(t*9
zOP0K}5*OczMvg-tO~5seQQ%lOBX9!Cu2=DdJ50P3C=xH^HiPJrV+v(N7{V8F1-j#!
zt^%HXkrvix`w^D^8LBO|L(W59w{Xj4HLm55<6eqMZ&*FF3nEbL>`Gr+4xjHw4w=$i
zfDjlLLU)j%)%`^+H)<l})*Z~KNuryOs6X8Um9W@EYqS5O+>7=u?ooXM1pD5jQp1vZ
zRQLeCb+3;EA>aBHw(QhR`xeG3+_Z1g6?fD4)o!?H=Bbn%s1rIP2l5lU;zkVjGryNH
z@HZI){A#BR@bC5I-hyU;O^P$hZ2Ll=H>gM0zLseOZ1piUA!%z)0we&?-Ojq_RZggR
z4)yc4N*(2>SyvTWb4#^R85GVR_t=nRsk`lE>7?8_b|{JLdK~Ztri?w>b!24_>7+^t
zK7|j`D|^_us^e5~+a#_A*hHlnQL?IXmYQn6DiJBF*r?pTK{l=_N@m5=Ypq3GN>$`!
zd6d41)9<}(IBX`3W|FSBlgMwf?jY?H|G*G8RayoTyVgMFlYLPbCKxK#n2nSTEF!Cl
z7y=!z-$aFRdE1Nv+#R5kV@h{LPjPzu5}RH_2l&q4$rZoA4p71`>`Rw7XbDDgmWdeo
zrW(k=#J96!bahmeI5G!Ona<M6{L6E&0ZOugOb@lf_D2r?zAtcYIS}MtHtyQ^7Es=n
z*p##3L4MeNYR0;>RCQLH>*#NSQo~S+9rWwiHI$5DA}S6sL*w)6kX9KdOd%!W!Etyq
zS(|@wo*%Z~Q+!x<q9}~YE|$;0LE@}o8Y<4=*`$zpu9p(;1-Z3ap0sq~Pmf~%U_$h(
z$Dmr@Q+&|Um>nb@pmnhe5d}OS$brDFuQ%;=2NI_shrLM5#_o!oN*Gm<&w<hnfea-8
zTVca6wKc>PKq6jMsRF_VCMZ)$Sr@Ur2w%WQm^f{qjXv5-q~;U6z=O3$NH;+xle~nh
zId{OaqDJgAt6>>=OHgzWse5q5TF$rRT2jFWo?h(H*RbfTT=il~WCcd#C`Xi$z~J_S
znHVDxpJ795WVP~|7l1irI-8*@5J7en?q(s+0eY)q%J(6I9i~d$UKzmG<rGp9I7rFj
zLi=k-Ds=7ONjA}N8fDmrIDRDlkgcf!ZO7QjDA?WVv<k)XLSqw0!*zbni`+w0lMetj
z%kbMUJqb_O;xy*(TRIDQPDE>!mhP!UcmH11J<Qf7mA#2r3C~H^I1NC-qMOtL815J{
z$-s(u2CY$c3_4jWXU3T^AhTj@v4inuJ+U{+E^>SXl;<NwIy~f~N!fn>39UF>j3KDI
za-Y<eaqehK(H^U~ND@%N#0z_al(`7;0q#xY1>53z0C{ph2?JZup0MK8*PsIu7p4g)
z;^t~gHLpgxixelq7%5NR%@itATXm|i;7yFcbsR+eu*2~8?3-nre0xC|84C$<G4n65
zDVcK+x*DtrG}~f>PW=|;`(vX5O<=yzMC`eHu#E&M(<lcN*PAH}3YwJwXyKg0_aq!=
z(}hwBmf^+dZ7sW@Q(;OTi0UI$D6VJyf2Ae$U+8BFyM_J){rm&n{vCdrY>evieR|42
z$ZK!$np?aUc5qHmIdy=vJm=w&56jOR+g=*76gcNQoa1{j7T@#*a(g}r)R8XIkV^K0
z_bM8hdyx-|C$ZttM?9d){~8^Si~-Fz@%mINgZ&s2W%9?bUgk=HKnmyy)|munmX!X*
z3L~rGMzA8hf=J`SI~jtK<_S&_JwXcsVABFt$)fCaZFf7>9=b;Y>3U?>{#u89R<y`_
z2V6pJQun_Kx!J6gtzAR>0Emrz;?j+k%PWtN=}{)iEAPsA6Es6r8c^Gr>oH6c_2^R=
zhfjxWJ{`jOv~kLJetmjP_vtmh$qwW(fNSLe<VzDS3M4N|2<lZmq5k<2!Y5t|%&4s*
ziDIGZp_am|_E@3Xn4-#inlALu=RE}mA8zeSK7reWg{XqWh0r4SC`hvvz+hpT@P|eo
zlI1VMa(oLSyRtWaF1-P8MUWd{TB<1#p7k9(Z#fXuZ;wZQu8w61J@e2XtdT-cnc{VQ
zu65{jnzFuh=ry#+>yYwQmE)V~%^+UFdH%c7hi}}t4*paoQ>=tifsUKChSb$z)7jdb
j7Y*B&6?0hJ=hZ7gvq~_$PAe&1z~Mf~m6kX}Jjwq7JPkW2

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/adadelta/index.doctree b/.doctrees/autoapi/lmflow/optim/adadelta/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..ff8f7b323a10188703d4663bf899843425e51127
GIT binary patch
literal 10182
zcmcgy+ix6K8MkA5?R9)_<ARd3TZJ}jVtbP`t-v^zN~`ja#zBA(qAip0&RL%s&(2Ke
z65EX;T8Xx?dMHSpN?e4vJn&MGcqj-7iBb?n;t^hW;ei)k`3sQvedo-X%i3N$5n#!)
zbMD{$`_A{xt0V9J=KO&C6Vp+^ZEt7Uv@FlJSd@w}%Wo%L=Ednd>B4L2owO<Hp}8IT
zN!Vto7{H7+w_Ph_Uiwlhrf7O>yU}vCel%k3*!I1AIUnIQKKfGH<aIF;+p)_^L*kU%
z?YRE-vLD2D7hsqcb7M251(-klCALF=95(IsR6$q^Ou!STd{)$r905KgY8l+?V%W7k
zmaZoorWiNlIJ7sC82A!?L$(zWE~9PNj3RX6eZ(`n>TS$Qf}*iuwzopop*e{djm=FO
zO8GQD%_sO7KE%(8FSfYfWvwvrJQlXL{cx-G{PpL6r2YYbQA+{VGLzUh1G}Zj(2`_m
zX)?5I&tf~vVFb(`zIGzn-~(br_LWGesLLM#Cj3!Q?Hv9e!~c2wFA!EBOlV77nqoLG
z?JyN*yS|mU>@$snsV~l`@uh=@d2KyT&4gdzO9k5N{9zE6pHwnaFnvaMh|y;<@=E5;
z7wwFT*z0LULbaE9C{A=VQG88J=Zm1un5523Y;Gj385(WhHKKsE?}1Rma=w^(@hi74
z-MM}Fj{Z&QxEsZB!ulUDc;(WSl~odkDCu_1aCZ$H2R+4nrl8P+<<f8}W=kytSXb3*
zY8-!0XWBGAuJvi8cVZ?yby15qLq7@jk&Co;6vVtoT8?4Kwr#~cy*5td^FY2WaLCuv
zaFo=RHu!3J1F|6-zO}m#P{279`rG^ZrP(1BIB-cFhQr(p&48B{iYqI@?ux(RtlZQj
zY^)xr3d?$Y?dFS;#!YDGM8G`D_BM?j7>e??Pu$10nscMB9kbAMOZ)L3WQsJVp-2l&
z`tjkK1S=pR6DNY*RMdjqC7_zJve6)l{}3$uKFsBzS}1D354J27=Z;~;Qn(SC-ZgD6
z<v$anXvMn$T*?p~FHWM(b^_OK+i`kV4sV;Gm+px=n4#>^p?*M&VUD!k>AE~8Hp5L8
zr{dxMY?G?}UD)jZmr7q71XbUKBFytz!K`=7%wmTUoVfB1Kg<6Bb4UApo0j$C5G{Bc
z6W_vt-|Er4%(-hreZ--s`CpHDP=E0E0L>dm?eq;yg0VP!CoxPGDTPM_f;MJ%dbbp(
zO6IZQ`rf9pd}gndTUdRwcXfG^sPA;$l_)SN`WXvC7P0UalY;X#tko{)MQN@Jkc(i#
z)E>Fql{sgEe;m7Hp+oM-`gqal>(X^Mp7FgHW-U?%VqWa)&&A7`8ypxYOW5HmP4K|6
zI@9?I3DOMYXSqjyju&s#A{aurCb6JxRHG^Nv|)P(2OkSZD{UqID8ZVAAOm6;k?elY
zPqulAMvTZt*cK;YB@$iI5H`stbfd+F8(zJ%d~vZ6a{ubZ<xfAk*kD0)^%7gU_|!sf
z?Wg6Arsw&w*%SAQ%P*{+i{ZIf8s{a_=hqe+s~Q4?Cy=R?Mn0u+er*jgn+%ob#fUOz
z1s7Lyp5%Du*r>fqzCjE&v|!+^*yOpMO{$_G2Dj5*4RM~OC8r%azMoDyBbeaSXcoWV
ztf)m#(T|LZr5w#cj^G~^#c<=G&{^Ly_=uQ4*jQ^s4jV48dnrHepa`-lPJ;1EP+@tn
zm?StPm1Vo)0IS%%&<C5k#HJUO!X+Q>qaFVUaK{K-1?w1{auk@|K48vCFr{QO1QM=|
zJ4K1~=Z+!sxyUQA%l9hpLl-gVj$+w>A~C6((MzDp#Cb#^YF)w0Z}ugQ81(i_>RSr`
zH}TWYqE~tQx?1#_T;#lgA8qJFO&I`%w<F3!C)`krCA}74|CF|q5`T~gI^V(YJ`Ky@
z{t3pUg3L-@5!L(vF<<P5WfBiw)hFYvIQUn64;F*|etNx6OHMo4xwr$^p1dTN<YEX0
zP#-7Xiv$1E|3GM0$4uPVxMjKts~#>Mh>{@i!`O&$*NI<Fe7cJ7k>U7b8K%OWY9Fd^
zwa4b+m^R)Aad8NPId=43VE2!{*!AV(-?22=aDIuOT$B!B+Q0gr@*#%zNlV7H?39nn
zrwo&q(g>+WfziMFpY@mCvyNtx0^z^Fq<ZdG59AU?HBv+LUf?u;;=Z+Q4oi>8rBC!f
zr5~5FQz~2<kr9N(C~FlsUG9I{eeg*EabJ8o&LZ*gF7(_s$aWaHS3nakM(Xs}N%BrR
z`-1_9m`XoVcH@@Kw%f=-rea#Z8m8Cg73+6Y+JNqu`Q6V@x5nt;<Qj5(m~30A^D^Me
zFSGpDWQJ}K6UrbfYhIAUS(5sKWE=8_Fx9CTCvT0bEYhFV%e1Ivw^#mwj_P<#jCB&%
zm0nSuKW@Z+o5)8Y{Rgt|4cwV!Z!wKvyF%nVs{$&(c!`?wBACZkZFM&&c@0IkCeLqu
z;^Jpbq&im`c}@aurp>cFtV~1Nuds0W2IXbx?WR&qj=WvAvZRME_sA^&UT*mo?(G2n
zBfZkOO`B34cwRhm1aKu#?(~3Cjc}X%y`$3}&Mx-o7eSPSrJU7p-o(28iH*HdVuSyd
zf53kyIn6&j8bz?4BUK^!{vh~HH18LX(|pPnP;!*!dl_Nw_b%sLAT?QzViq(P8qI@a
zb$y3wbXm-O3zxZb{U^%((&~*)L(0!8t#dBoWGxn%Bs))*`ox@ss_i0s8nUbFzQ?i)
zn&XwH4bwDjcvYjYyYFZ8(WmwZj`Iq3a&F_%=c15B;J?VXlGJ|)e&*FO91~?oB|O!@
zbN_g=8h)ZuAyL)1<n02TKj$#qb0GQ=mLAhJ{;mHh{aj;qO2st}6YDfms!?F{&;DoK
zhhtPA6h{46QReR@b*j=)P8b!TsdF7u%O3Cmx%fP$R7LyHOk6dwc}(XfCbdc`O)HtM
zlAkL8+vn%r1AsDx)pSg}IvaRYHK2nYc&tqtC0FWjdYc@=!<Iv$xD5-Xb#JE57KX%h
z1-Rw{rPa#*?IGB$pVJ=Fzd2pJ9+OzKkxMI!4YXoc`Q6O?fihUVo;%A|JC6YyF$(W4
zQ`XMo)RcGfa@+HlRB;Yg-t+bAl<r<qRfs(?kqzqhEM{abJIW_bd%o_|^JphJFJf_w
zw3-U8TA;YRaAp=hvzb*S%ko*Gj1C?8QYK48ReAnN;_g*?T%gAqJr?t~CA{r9%ktTh
z&x`VTNj@*j=M(fKSV{;?6E6D@5YsAt%4vklWLD(VLk&$T4E0H^B9B?nrO8)yS$Rpu
zJ4>dnKC^uPL(0?pP8m{nerfJer#;*{<P~YYC?1SM(~Df0HZ+3J-$Zg*RisB^WE0)=
z+JG3>bC`u@0)p{DG1X2&s&^ZCb$dehs4BM@%VY!v{n@-i-w9=0XrbuyWxA9{d4MQt
zei8>soUZd}6i_x9D%UdafQDt_avq!A0Hu|aq3_246>;-UDh699+b9qy_}{{W@vNDI
zE?Scm(o)5oMiJJ&E#^Ad0I5<|>b@%`>6-yB=^8s7KkTCVGZmp^9iWKYr7sa8Y;{`o
zFqg?5W803oA(MAN11Dt*vI_IMm<X9nFu3d%10hd{ac&|Rku^ueEZea`KjdtasOTNB
z*haF6{idk)L^V9@%gio%>R?2h5i$gy?SWbb^KfAyO%(Mc?ks(3NnQlbR0j<fhJI*p
z)3aQL+Ix*a#T`LRE1H*f8Iil7Jl`|YvxRdTx+A94z6A_=^_l8G&tfq{I6&$m4idSt
ztTE}D-ezKMGEkf@U?=cxhj<`lZ9lXOi%~rj$Iv$f>!Jo#>)^fvirAJG1>9$*sNiB0
zUE&0;R!9IQ-N49&5rmZ;Hfk54XQGC)E~aIV6m#N!(G^52v3vvN7tESLZP1UHQC=yf
zq>T)d9odv5!(N7EM^p<3A{msUT1lXA6F??{=~Rh&NKH1)=-4ih!^aZy^oB!F<e_9f
zctg8cVEIXW1d}(f4T@QL-E|o%OB~YQMb+~9%y}^tu)v^FCiJr{35IfT0BHy8q!i*~
zJdT5CrPV@rv>farO!k)ja1*+8ie^dxFw=-x7wm~;)rkd5k6pY1dQOF%cF*?IQ*8gZ
z-X3aelE`l15@)lxBhP>-Xz{q~;;=djV$dR4q8}4yA(LGrSI!)Ptgx^rW};m$Hg^ne
zZ*mv^kl$e{tEFN>=J1RxiFX%w`XZ-ixPIHrNPD);k>hQH@k-x7Wl2KJ-ol3md-78v
z`F&7k2KFF5e%tmskb%rD%#oppl8v5ec9&$=^-ZOW^r_EQ3X$n<Fbf*I3kWzvIe1T;
zLf~v}84Oiu*T6?lWO2czyJ9x`YzeaJc9b>S6GP0qMV}0bF=b7lzK}#Xb{o18<EslQ
zGE2Fc1yGQz7@tBxYLNOQJC18fxtDJ7vKzZ~@&*;r>H8y0AEi$9-gIt2$DQZtF+nj^
zogSZ~=CgPtqE2&VFhHL$$=Q2y=AN7d&-y6!(*~$6uHx&6T9y|nvb(*>l-1m?YA1@X
zjUqVT^e#E-4{0xTpB&aTOI#BOGQK9klC?OQ8Ij2zhhj&EV#`EKmSsy~JBdQSa($qH
z>LC-N=P9z5dr0nOFJ!U1f$w`NQy=W<%xBjt)9U(Exzk)_e|Up1PNCbUdiTmk;ErM7
z+I~Xj3%(CkXcvM2mc!4EK9lvo1CG3&egVpgY!1{OT#q21>cHI2Rk}~+-BQF(+FsjD
UEE4HWvP$C?nc~c9saSaVKL#)#+5i9m

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/adagrad/index.doctree b/.doctrees/autoapi/lmflow/optim/adagrad/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..65a1e9fa79f7b11bf3facf8d3da7de371eba0ed7
GIT binary patch
literal 10143
zcmcgy+ix6K8MkBa+Uxk<#04d8x0Kc!VtbtsP-PrTrB(Zo#z7TAh_+0|J7@RIcy?ww
zm)LF;(Mq(9)k8t*MB*aE<$;%i#6v+mQ3|3+Ji-evJn+IRe*qG|@0_`=*J~%Lu;ke}
z_wW9F=lkaM%Ds4LNdAeLz~iQ~x2hY4;~Ff8#kk?NqAqj7_`UeT8}XgEA!@$98@Q3*
zVzC&)j21U-!)H$XN-U;ndT83gYO;PTV6D(}ooqR;@G2jBC2sJVsDx%{v%-)#V|UxO
zySwUoq1go(ddJs|nC4^p=vUbu!EwZ_H)91~)zbk@81p$%(=zb+h^Qt|Z;4UcbXdF<
zZR=t}4@2MFjzVBd*o~M*K$wiRY&{6jiT8@5ch%dt5qWuITW{_7tW9$wF&63_8jAT0
zKgTEec|O7y#fweuc3IPp9EbVMUDw}fzHt3HpqTyuz@VuBYwA(x>YmwDL}*GPG!r5;
zO~+t+t9}5)9=UcZ+U7%|BKr#TQ>5jO0ulZgX!bb%pTPeT{x1+xpo?!xR2pK`(@j4X
z=ew>E+3ZX8zN8oD)bL8*QC{5&V?E**`AUxS7Jmek<)@Xn<oupZJH*&C32h}~7xH$(
zLhSXlqMzDJKN2H4n#jGWrt@XcW?a%{Hq^HxTlck=YiohWTK7SwQ8{1CzWnvum+suY
zd?)>l>9`yDaiaCfb9epHmGuo0gdpm6b$@>o>;^f-LZX^bfz`rrEanO=4M>+2YH9+1
zPaFF4(4&coRFf*b^Af?SiE7yK-N-vcEK=7okn%n$IgTZ}rV;Y^+5{2L0r7UgA2*eS
z<0Q7Q!8eK<knz}djr~J_0?rZN-96MV?G1^4-z7EJ46-#IFDw+7*1i38ciUROnNYC4
z(N7IllhMtaFG~V9pqi5&a}3kzXnQaa#XZ;Vch8h(W40MGU$+aJ@gJuOlt2*Y3iPw1
z6$q*?p%SOO{a94J{T1LCv!bOSQvU?J`Vmazk;*5kzzy~*7N0rZa>ei)G`p*tPRxHP
z#?T7)J@}9jI$fMbo9%hF*)qfUt{mRgeJ4H;H84RLp`$Fn7{?T8uj4IwOsM-E7Q(g-
zVwcqBAHYWcpS1YeFbH}N!oR?)IiKzo`NWQ-H8JEpzR3Rs14sLOkCqK$4lQ^W6W_ss
z-|f-1#FM9{`Itk`@xPz&puylD0GhXs+v(ew1OsvOPGXd-Pz(<U1Wn9t_iic97A#`j
zcAbv0c6zUX8(4j_cXe@+sO@#_^}y39)EW1D7BK%7lXCM-tWBNH%hE#UAm_1zo_gS7
zSK^Eb{t4`oL<*Vf8sIl)u1i;4f5vq}n6f~bh6Qn`KNBx&et2l8C}BryGr<E_YRzOT
zBuKN6pVc1uIaz#B4PgJ^jl@!FJsVO->!#B;*0^w-zLuhm0-Q;}F(gJ2vL5#HWR9n4
zL<>wrXkp}+Le47*bPaL=-C((H`&ZXi*VZmA*O?bwy~I}5o?5Q&GPA=&&0sBk|LWR>
z%+AlqjdjOyL%k=?6_;Pycszu^Uav1n#FsXg>l-N`h)Ez;>-B6(eQ9$OVVVq$7eqyw
zvz(C|8Cx=RGo;jB1;-$U>#1zuE#Kt1o=wWqAclA2UL~=fr6uPqI)0E^S`|#Nsx*sV
zwAR(4FVc^UgQY0VLzLhd6~%DFpkO(@Rq%>9+t*m9cpNdB<Mv8+SihjLAx?w)E6`zi
zte7GwB#mXe<owFGyfgrpn#83Sk+IYpAL*Sf_ZVQu30MW`IGu4Eklg`5&PyPrbTdQ{
zwuvi5f$SIWL+&%NSE85gS=^7VU(h4Pw;@Gn(mbt~Or?nnh(px6oTJ|vNF*`r9G2F1
z6dm5i&mhBI=J1<p(HnA+^%j0odnd+IW1uj%L~-zx?WfX7uQ}kqpbe$m_Y+0y`xri?
zX9?s#$Cy-;ImtDmo9iL$3thiR<l(F7$#}~T{_Vhn#jtyrVjs|wb5?RLE&!$@FUJK5
zsYr8}9+<T9L;o`PP-t3PkL*yprP~oJ10Ws>BF}UEPz!L^315v|x`rQ+3HfsouH3h#
z4mG{fo|u!h%$WCM<p@So1m$~<-oFN-H;|eC#L{Tn`VD?Ei8_LF{~3JBM;JaNG70OF
zQ$8-9Qju;e#VBdzIQ@6<X@BcI?RY*ZAU**;O_x1wKW0r9r=&QRJ9AVQPCc+L&SU8b
z+4tGOrwn3Wa!QGP!)qy8c*}9QJouCcV4FJpf!KDEMdH(4XuheDMbR>EfhJs#)bekU
zAfB@h2MUllm5QW%$1RiXwvdmE#Z3CD=}wE6Y~pd51iF5v_rgU@T_}UTsWv41FxNC<
z>s5f4-E#Tw$S&O=9uy%~CO)TzwIa0#sW&7OVZUQBLCzePTx3Gaw`@^O?zQ~GG~nY^
zG2V`BTRKa1{)85~EutO;{2$7`w{V-5z4<hP?FvzIQN>z-@e(!VH8BgjTI%Lda3zXt
z4W8Wz#gkvwk@sA$XNd~<o)%A1vNtF(Yre+(n@Te|vUkb+f}p<IBdGj`nf-h8B-`?z
zrt_`av=!xl7sT1VfE13s({pS&ylwFJkI#5m``BYp=ui-cVn)My8|wxqH4X|%4gP!n
zA^(HqCI9Fs#K2OH{DkD^8Ss;+Iw%IG`IIA|Bq+^y5^_A~4bHkq;=de(%xhezH~NO>
z>D8{wLhc&4!aX)<+A8f8m+x^YlzdTXlC_3YQ}!p)X?>y4C*~zYEgLCPpIzN@9hTh3
zET=eaR2`;dM>TR&`awdQ^wb_vv0lSY)@?in+zgVS``6i4lJXD3WxQU5W3mXT)S7aP
zd3d~0jx$k;kf^3UA#3O8{56B&z9Y_$vGjyq?jM6s8RX@XQ%YWLlvtM{r5ZU#{~mnS
z1Nb@xLT;!}6iWV4Ql~5(#S~E<e_GcuwdmCP$i-JMr7YS<=H5zq%oF-6F`1~N(zAl8
zDtN9OtV5pbeSjxZRt-zXtF?_sS^nGTfe-4?C^<)q)7#Vt9;O@;+2vO#m3uq3b}%BQ
zi=Z_wP!g@^uO5cI`W5XU{gu_l>j?=(8`-on-yl`$GOrtn^;f2<*Apk1Y3m8#BF5mY
zWpdhjlA7`gUTiz=iYmRqraNx>I-Pc}s5--em`n!K_8ev;E;z_0%{Xq_m(HV|=)8c1
zA(Ch+R%(LY@>ZEx_QYP6ku1t)fih_*Fpx4y&Z*4t&k=90(&HjMHtDgPy{+Kwz*?2h
zrhKl+=Oy{PET5mJC&5xcV1{rxgn*b)@lZx1+!m7(W;#?)Nrj;SiBx1QbGkJ6Mp`mn
zku}v7Jt-`+Z2SI&=R;?VsB676^r*oeZ5Og)GhY^GLSJ_RTjmEf&v!dWB&%X`B}CfL
zj;{@giF6J#(2PJXJ}jnNkx!*;Eh}74rah{FEyfcGL8X2!E4{aU8RZ$M?R<^y+(8!Z
ziK-igUKGY#d<K=04nt8|rWVjJblkl|z3ZWda@u#@5TGIo-iyU>6Xh8N0!8vWm@ttv
zBi}}AiUL$BhSMm5)OW>v8yg@e$_m|g#T0!E;6z<*ukHF>RC%TYl%qYAXuI@@L4d8!
zsUBuC*`w{6A=hM{4rt({WI<9t-V&2OllcRi-C`i*DKWuy<QkG@Ma;216ZAuJHVX3I
ziov#{4)&X-8WWZ2urJfQ=&6AbEk?)?e5M0xY0SaxgY;0;qOiU4)QSx2&zA=^=KHR%
zaosU&hPrx{Kt;qaW)#f}yHw;Zu~_UG>Dj`%4b2hLYTq1&boKf2K+j?^OE^I40uB<{
zvV<{Z>rN-qI}8-3+tew1v>_hyS<Ceeq_(MKiDT&dfGtslqP1~*0YywB6$RW*rm481
z<z3<wE=tJqMO{zJgb{?59VY4&zN4dFvn6I^j}&v_e%|E;EHYdTr54PZMGepmm{wdV
zrKE)fk{Otk2g6>PVFpy{1|kWRW2urr;U<Ah0K=&=@{pQjn$|XLAcqej7U&I!prAup
zd+>&KGr;oG__`%)UK<v3^6qLgRFOEOzl-YR_1Psc?J-ZIk|p%BB?*R#uLo%d?4%Up
zQ#=g4V7=Kycd+X1BRqCiUB3feI!iMp0GO$TtPA#pvckl?l_%D&fS%KSyVbKj_2k<>
znQjlY)k$P`n7?YWur1GkDQNPrYvZsQieJzoS%M!Ii;&5#mMLe3KvF?C5VOI)6Y6^!
zN4dkMl1Ge>Wl%zk#iUHpX-SUmE-vVK%1*Q0mY$GyvBi<dZGrJh-#}$aLd@R5hXn`n
z8zK4OPhtiRAU$r&blQ-C#4gN}p$MYwo@r*6WY>0erHu3y&rS@H>25Ov8oUn(I78j{
zK%7O$tnX+H)n;46*Gyz_!KJ%mF8Lk_vTC=LH9HU^%(+Eh1Br2EO`yJzM7VVmx)I`|
z3aTbcxtRk{kgO2jI6-QV*(5uT3rVq;?&Pu?yS1|h)ynD9B1|8nPW9ffZa~Ma7w9ob
z@l=f-&r|bRJR(t}xiS=>uao5L13B|R&VpZkjQVK<R1!Dv^+PpDa^wlz-pt5G=2cS{
ziZ6)*xZe0Kx#^E+FLjF?P3x1mAP{7HQ-URHZ!*szGdC8+jTXg~k(esVmc(`n1%2iE
zKmpZ5CM2Dw$Xe_nxtG0=#qKt~-YHEz)6<#Fu9c?M(o^M5^QHaa4g4^MZlCSlD;a@1
zhJkCj5t%ReK2)4-2m)9RKRfnJ(*GVf@@D)Zloh!ds6DtIKt9!hnVl<jpU%3ah@Cc_
amK_-+(jBr&6DTjbaL70a9$H4itN#Nx01d_f

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/adam/index.doctree b/.doctrees/autoapi/lmflow/optim/adam/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..fc9882aee9e2088f0891753d484ea2647d695f5d
GIT binary patch
literal 10086
zcmcgy+ix6K8MkBa+G{($wQ&n++O0~PHL-V{(4y8jmP)Jgkj6m>Aw*jy<DIj6W;{DH
zol9&tifAR;#_FLUWh!wI;_|>tLE@nxo+t%TBp%^~7an-wmA?Rq-*?VjckQ(kRao-u
zocnkGzVm(adgZ;B&X34HF&B8;boQFMVK}bAf>=x%ZaeBRCyejJ%WuTD;)baC`flJx
zew)Q&1T)&)v<;s*@hh>IrRkw*2hC*tM8Mji={nhRUg1?f@k-p_HBkx8&}M}pv0(Q)
zw!7POz0m9d480fAWXv7^D%&F<4x933tRSm;I^YOnz94E^1~?xR)db!xF>ad<i?^a}
zT}<g==$qS72s{a|G1CYLhl#eW2LU?qUUBrEdYd#NFK=w??H!+WXig+1LcL2vF`wgS
z`3!%EkMVQjT8q0q*776AVSa1Z^><n?UVQ<`r9S{LXeq#2dK9|4XSNjjEs1`M&~KTJ
z!S<Sd0IVLqaw^*9BcdYv3S?6h<&OXh{wS#R82%r}|9Si`6GkA4Z%RBGV%*bBKNg?r
zxkhBOFVqJqE*?~4tAj(lx)sKH#4qsG9MvuUFbK-0m3-t3o=-c(#B&KrB{P@ucEUOA
z^{gVB+DktY7do1#ys4)16;NeTQe{5Ww<BBkwYF<(fydf+L7;ItU(CP!^_v%O-Mn-w
z{f+6kYx!~F`X|iP^@~rgZxHi?sMpi|{Y`Kgv=mE;B0}byg<(j4p``)ivieNX-?I>#
zsi{<ND*clZVX28~*!A7W8z2*D<^+g%mo%Kjl3mjXd3<Gx$mW1-JK&6)O2KhbTG-$l
z#SO?#Y`ez(0HA<#%y)MO`lV?hkw0`v4Msxm`ntyp3q@ny+h2FLt@Z0fzWT<1=r^f<
z^ZLt@t_^76jK>_qbh_Fe3_)?L^C#RY<<Nv}hRoOP!XEs`iK2|tsd26)KRa4WpxF`<
zamw3|Mb+D11v)V+8Umu~Pr#NR!TcSmVWJ9rV4Gs`$T3zahAW_{J>7I-{!1}|R=DrM
zEsW98Vj69>=h<f44CC8!cvttG_(0S^Y-M8(wdZ0I<I)PpTk?!h_q!~N#nRzyk!tz_
z*xdhzdR`d?1>b{UFY#*5nD>f|VMh{|IB<ub<9~wLqJ6$c%Z71-7QBmz@8H1i_Gwb$
zkW+(u%%Nxb-%ohZaPSWR&0ELq^leOn@*lmE7$*x8!=nK~6SF)0TZ%IUQ&+cLr>iWP
z-Y?MxR$uR5U7RFpdp&zS@N|k;COw}8%)h~;(0mhXQ&;k`G`%^<c_5&t9<|t&I9Gyy
z3cDn3KXXV!yy4tc>3ZwWxlRZ(6(~cnBnJ93@v;_2M@EVgcC`8uJaCZKT(&}jG!OY{
z_Q}u5;)QAeBL_Dl9+j4{A@#CuI)j13^3e)fiW>^BC85KJ7)P`@=&{Mp&eDh$m<X}L
z$S*~UPZ2~7vgf^ErEdF|*P3f<7gy@rEYyR`jkV@8D|I}cd1iT~&b;9AMYg*3^m1mw
z=j6`1<G7*T7o&<xFKs*)!uzh*&r5{QZ?4ogQb-V9K$h0)*_8VE%}vBwGOS$^6=k||
z?rmhe$neduQhOC_zZk8j0)e-DlNb6nDT{y@-HrP-!+M^UoVDoqVY+BlFu|(QEPlaS
zSBpMRKQfY)^0Nr}fzMME!|i~=-SkGlD`IibFsDcyHk4!aN_Na)kzYehgYm0SS$Uv1
zP3KGM%67@Qm9cnf2o^PoML!D4)gJD5E%z8OCkadi<0S2V9C*DU;4Mn<q*OBm2DXXV
zzd-8NUCMhV?n>mceT#e2#R)p3ST>@_Oe&}K6R9+D2~mbxm-F*GLkT5Dok1ymN73SK
z{0y_=W&XaY7QG=CS#RMdHEv=uH3SNCOB4rB*?uaR^qK?y3))c1`(c7;eILUE`jtTb
zbBsy#Sdg3}y15=Ay3qBDL>|4Io{YEr;NK2CSd6-Z6#IadoVAj35y_j5yw(<^q#{jW
zdSKGZ5B<yVL!oCKJ+edXhHgi!41l;Vh&<2rLoL9KC44n<=>mN~rsK~=xN=9Dde8Je
zdSXs44jS^qcv_LuGF(+F$Ms)BaUIIpe`0C0ZT$v6nRp$+#{UdI<s%Fa$WOw@<dlz#
zr_2<+SBhG>Gsp11!_WR(|Jld0P=WLbu&|c--NQIHhS3x``JUsnbn2e<aS=;T$hFT5
zKV=x#l2b}ttH=l=#VBdzI9(ck+CA_|0dY@!I>{nriC_gxjjW57c@8w;lBC9eizM%?
zH5e>F#8fJgG8i{Zw%bNNF&1;_tEM|`Ub2tJr3dJ)ncf3Gt*P5%Fqkky&JGhzBeq@z
zY}sX&|BejQHDW*!TxIKXQdp}}SCBwMJ`jdH7E|Qeah*jPvwWEr)#L`tKTIP&UKNv_
z$hM`|ROe4=q1z_XQJDXs?0XCMX4#uhBiNoKQl3-67GS(cO?f%Y;;y#3FBE)=qFIAy
zH$k!XMIEWj^?H_rfD>x-B%dnNisoy~Z(gJPD!pA-YRQq4OI8+K)2n^*%72(y!R5O<
zfd4dIY2BnvDW|(69ytcM0w}loKq*JR4gUV|X%A~3dkl*(3c^y%Y*=q&-SFhbK_R)p
zf6qVUe~_H!A03Y(SkIBVkbJ)%d?%U@3)X2qWe_OwN%NhAF!y@Dvo4UDJQ0M<Yb@6r
zgK@mP_4QcDT>}@qvxiS;()zf3%S$2S=aiONYd9$tgG6$zrwV;yQ9{wSkq!0P<t^7?
z$z{!Qiqpo`VM>NnBe$_1B=ks6?GqI1HSA>F#AC?iA&Jg^ooywl{Q!K*>qR(bijYb@
zDhHy2<Bf7Kib{h-HFX$SJ4ffQ84PzFmVS(-Cv=1V7=FqyH<+AKa)aZ<x)dqZ$T9l&
z@U!m02`UhBBYvWI^N*4`W$7p;jq)JXx{9eqZ+D1Xd>K>9qJ3mGu9VO`i6av=QbDC*
z1rt^9XE{Iv{_K4KCev39OUJ9VjYnDXJLrLD>e48=OpDXo=`lP^IV9B0SSU?<JGOSv
zEv7S|8q1VQEBdboV3&SHYo-6PdU!n{acCo(R^}U|id^PzBeD0&F!g)pBqwb>4lKk3
zythnCTTf6^-p`9|$6ZwgHCS=SO<!lz?p0MuI1n?*VA@{5jKn1e*`ztgP5aV$v=f~d
zurNd_O$AOZP+DF$6APZ$$TE^ec`Q&S4Go4;CP_k-dHq@9>ScOdpvNXXR<gHMyd79g
z`E1GOntWcA&r9<8IeHQ-1q9{@mjMLCoQjV!8sQq56eiQ5dP*t`4N0dWcbU_r!8g(p
z@T!bwR&^?oux#%`O2`AdkExr!H0!9T9&H7($}wLN_lLgj1hz~SYM$?Qkw{jR<4TAO
zq8(ou5mV_LW}q2?P<&L(wj-bF(ppx9o=JODHCjw2vVns8LRLZV_%hxzP|W!nUBiPc
z@Do)x3cV<dxA+_iBwdCIv&=T2Vd%JyhkDOL>15h>-4LK69^Q+^XbUA71p)>1JD4z)
zG$Y?e>okR`RN1Ccgsbn0#SS(=dXyEqZ;R7Z(|4ktw%2j}9x6Mt0m{oB3a>r-QXs%q
zXH^fgne5Sa&5&y{aR)STQnDba3U7%SpUISh&2BId@|2k3I?@kGvmzGQo(cLPPa6e!
zZ^dBSQ5XBoQeBB^aoCqZ8bAjyqRj{yg3ok7EsZ(2gpdY`S`>CxpI(&#{zK(KjrqRo
zYg~5>o1so#B~TIji#bK}!Y&oLODxXyjr46{-Gt7FS+#EtL%RB*@<88WF;6%^>H-cD
z*|LOj+SZ+Jq<0x8PM4}v`1V5F=d-r!8=AqWE{S8P-oGWPP_Yi~F`$TPq@sX($}AOE
zw7g54!j%b$zo_SFnJ|K|vcp83!gqAkYqrFk?2%$l+|RqbfJKI@q11v|^Qil|0n>^r
zrIfUhQ!)dSQe)UlGt7W$*gzzKaw1g{DBKK?31Bi+avoBXOw&514dn1)!xFvW5EOJM
zfe+r$ZU$IBjgM5a=9N*H47$SF3{@l!>F=Q;d3FB0nDv;aQJE6@*_H%DS=WQK19nmh
z@gNUFFIaE2&>b|reT2bI)AhU1r86{B0)UxX$a-K;C@W0NTYY@(NzilF@3i~2r=EQK
zC(`YqwmOOI4lZXV3p?@*n1U7$do~WMq4)(YlI8VDaSk%s(=z4E5J-v%2Vy?hcS3zn
z<7SuJ_=o(CQwc2=Gcs$ZC270cxToh?Jk567dP3TBZI0Y-8;n=_1}aMuV)hO`8aR-j
z`pEBh5;JfB>2cep(}4^mc43hWMG$TGO*4BWySA$<Wu#AXc4CN3Z<`s=;C(>A8Opr}
z;tT?0eMe)cFxwhFN+OF3F5MOj$!A57RlB3C*?|~i&JFssM@%Yf0`-L?!l9ecjSyc;
zP!U<m%>saeWQF)t2~vY(C)sgaKZ?C{JD1(qt&=sVI8NUQVfqAhs`rL<4LWYUNRJtc
zp=$K_5;dR4BN8>5D}w;~)JV=gkTVbDEO^vMsh>7LC2<2^HdK>bN1oH|PmpY6{xo%-
z_%bMf(~WPFgZ_y2QuoC1v_6UJ071q#C0Mc!Co>*0fn!muXi+Q~iPJ^dlGvU`5ns7J
zP(byN2}$QEvKD(t?qx4zvA2zHa!OP0@9WHF*GkiB>8Wz3#nS%p27VYrx6kzNm5jh0
z!@#xOh|Cv!A1cl^1OY6EpPhIv>AwSxycvH5%8D!v)E-<9AfM{M%+8g%r?YM;V$-J6
Wwj+Z?x=U7R${<smSS=L;H~$BKUF{tJ

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/adamax/index.doctree b/.doctrees/autoapi/lmflow/optim/adamax/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..a2117ae2c16bc6e74a7d40cda76fc026d779b5b8
GIT binary patch
literal 11854
zcmcgyTWlOx8Mb5Z`Vz;tBrfFAZdF3oG`=J%s5OpJNoj#lS0IE+FHFX}$38Qjote%h
zwp&%SsA*&MP>?d^RtjyA`T{}-0YV5-Dg?Z@&qys2Lhw?F7akCT?>}d*yX&#z7A$#o
z=G^}K|DW@p|B1@;7iN3ppBVF9ZrGc%db4ReP3A{pu<10y7PEusS~UG+^mtSkL!Q3j
zJE7NLk?6sQ1~;sx$L#3gNQ~0(z_9$;c>aLT8iC>1>2zM<RX*@=ROdsY5*UHSa$VxE
z)mpWjjakPHj26()Tl!{1!!f%5^K6sQ=sxI5q_C^HI&cXhJ|Tv*6nWk!sxi*XqTe!X
z7A=P>x)|1jz%y3D0JIWqeMZwKItCh+?)zxOd&SmU>TR$Yy4lBy-dOk8Dvb%nK%lQt
zSH#EoaX!LN@IHQ0e0GjIEjH(cw#~e`4aZxb`^>#}gTUkm2>Nphu{k{q9NjhM6bt4g
z3+5CH<_x>ZHfKE_blrIIK)AwtL`Amcn5X#4Zvq|sW^n8l{GY=AY5Y$UO<;>>NJ{FW
z-_;E-631IkGql)8Yu)%Rj;Y?6Zk@ck97KA^Z{;%?!OQ$cu$3QFnv#inJZTUE7h=Xr
zp-yGrvHY;sor-yCEqzCN2sH6}Ne$;`z?nhGnejkh2`$~z8jhv;E^Ax|lltX&G5*l~
zkDYz|v2%|nzY&4EF9Ro5-gD6|oV{&fk(9s>TP@w&T7s;>O)(W4CA5Dw*Byz8+?NKL
zi>5R(jK4dj1q}};HdJXJ7Awt=s0M4E6S^I&B0C)bBd?Q>gP5{mGy@)8946M;VBI>T
z<0P7}o3`c_cp$$3d5smPxz&Lv@a*%PjgEHdXh`+DP8ou`(APZO<++JscER0Ra8}HP
zONxQD#V#f=tNNBMJtVnXhfR*S%x)U?nzjkYkY8}-x)w~NHDDP5^K>h>5`Qr^oT`Ez
zWp?xO&UOQPml%ly?p7qK?$!*5i&)-05RbnPA-w>%xTDR9D(HgGip0&kdaDS5f(EyA
z!;bim!~nj6Ef+zfkARDV_-31~Wi*T+x+=RjbkB~q#SoaUJkJikUJPP{^w80=gc9i9
z8Ve$EY-ce^1O7IA^8d$hFZP0$&qLv-cr_Ez^LY`m9obB>xW-TN-@&EPI$x)0Wpbej
z&tc%R0Qj{w3yZ^c;*56zdYr$xM?mH1Zv&gBcU$Qx41!bGc_q<Lo+m=Y1AzucueL8K
zj^><T&2sED<>B;p-ELy`rS{qRL1JjLWi9xwPWzfc*JD2OE;Fe%U&7oZ%seEWa)xrY
zRnU`&n{SGPFX2CoRpO078mvkp<=DM4G}kUTb^!O~E2l6eI@(k9GADa`dh!~!vmp~c
z2%hFxIzyr~4*i*J)1STVC#pW&9s-a!6?@FOw9lGhciUP_mz!!~Z;`_>-d^;Ge(X&<
zqdIxuQR>lr1DmlR^a^{+dlY7M^7t+POwIDn&(F@!pF2}qVS(<Sug}lkai)gH9d}Hh
zsWI0-f0oV6-+rdH!HhK?Xie78x6aQ`r`~={!qRQq3G}u-t~mGL;w=Fp_CoEnWW?#E
zGquG8H*89vWec@*NbU5}5_V~FTRbHy%C%*3T}-8vav-Hst(A)qqPLc)3f{6s?rvM8
zs2rkqBWkx9^Ddfl+$8X2R%%u-z^u|Jeyh2lCVhl{<Stlh)+96w@lkOMrwiIDCnpMC
z5l6Z??u3l)of%~hr=Ys+8SCO8)L{k&ErE(dgoEU;{4NB+A|(%&P%<PbX(uBI$~(nq
z)7b^pK|)pGIY<z8L$OtY;-o}TYCFR=!ZL7r$g%um@6rxZGf=9S!pXy+!x(H$39?7A
zo{UrLU|fFS6m}?TUMAzOmNHKC+MU|_tm4fx_$l-IMVVhwlb)25%%|~_xI;0RI1EL(
zDe|2MEH6=4dd(1jl@^qC&`mkbZ=t)x?qbYeL7z062`Ne9o8w|X7&u;@$-U<j$au>D
z|DqJI=yf_d_A{Dt+>GJkOkmjZsGQT9igcL?!1yZz`p0shu)S41v;yt2ZiTFffH>@j
zuIqS#=Hv7eJQ6x|DBmWR^7}kqS?EpTZF0ihvm_@w9ey{NR^+ghR`r$9`gSR;rIP&<
zriLr#&+(J0*A8NQryR=b=<d*;SdKB2Kjfi|<fB|dTCp*s@ZaUI-)M)uy9gCZ?|}%1
z()ip>a((Dch?DObO;ZQ<ZI6?fx<{#fpd3n>)M6+Fsa51oB%vt&%4j-Q4s9QDQb@c9
zavCjep1KKZBtImfvDlcAda<0;GIx5<-o^LDawuiui=h<6*E^q(g}02BwQ?x?5Ss$N
zFR|@ql6Zd$&fCx^C}?TG$0v3c(ouhz9L908bMFVwpfbyp#JX&-jRw-okr+!}HQjFT
zLRi}^KSare<k){gQ|SV~TkaGITnsguk@*O)OEVb!mlRo^A`#>fRzW&rhB+gh9CE2h
z4kH3bVwkpQ$Y3C^UCf|}YMiv-uO(Y}UKN9@p=HVKj)Fg|1x|yQM<=@1WZTn7mB`j?
z7~yssG4rI_Jp;$H^eOWI=|;Mtj)u9-fZ|%6r-yzq|8X68?1fsIfWq#f!Q%wG%rP5}
zGH>=iO1RV8C1s)<so8jDPGetf(^&qUG<;289{~PRGSmDCElMf=DRI*-$mLM^X&WlV
z<4m3ZaQ9r7`D3h6wl&NtOFo%p{u=Yj^I6-ud=~!=e~tfEN}9jEJBbiIN1jT`{Rrev
zJTE(a(RfN0QU068{~%_}evk6ztz;(WG|gv$kAp4KwEA?d-tB-EAN5--;7$`~`qMiP
zn$52l6%Y6c#{8tRK64%zB<c}swE5v&n>d_btuC>QkL9MFA2*U8m-ygfW9E+k7BenE
zqfMjCN3pW`7#<}@#`rw;mvkv9mK(5Zcp{HiMWQSbDxNX&9fk8nt^;-we`UD;Ck1%j
zr;LALYMBvq-Btf{&y&orU}!!#bm7ev3@OTc$1G(b-McrxFAgg`EvjJ7wsV0XL#`tb
zyagm>imz^d1Fz;c@hED-3uu8$|1R}XtTA7tw^Mz1d{1_XBtIaEI`~XvzJvkdXct85
z)0BeD2ZI}6<bS61WH2ya!RsEGM+?17GqVK}WzQlWovP}w^Hi4I9_HelzIh&$i2;N{
znbtR#=u>7b^53>Iqb@4p5^X1W9Zi~N)XmJc7>PTR?+J{E<C~uj8nd0GEg45E5j=RS
z04Yed519k^<)J%vhp}VXUv5)5#1HuabNA{9Zz^$ww9dPBfRhrNhK0LAkDXt3Y}VEP
z^($a1xT=rLXXudE&Y6k&_4vDo)1{=pJJ#;+M)H_;aeJNp^<sJj_wka{#D}Nf895s%
z4%h7heN5fs!R|^ql(LgS45e@~=qJ%7M5&LAqHE=__HjN?C=_iye+Sal*E^Cs#c0@l
zQireaq8I;TUw<%mp;Ga?`}hp04j+F%NZpT*{}twS_;_V+iElrYL#)%MAA?JNAQm??
zaeqENS>XD8y7_S;;0=0QpvRl^xGR181m3pIyXEs9`TV4Oeo8(+EuR<ZNdaaD-@CKY
z<)Tt@#wn#Bz8)#YUP-`Pz(Eo4HoZv#<lBA;7^6562ak*ZF{TcBX?#K=GcLACx@t+7
zLRZNZAnx0;W$Sz~DQ%jO2eTPn)jzRx^{(Yn9T57ISC_6BwL&{vZdwz?&xj*|r`x_I
zFS0b(bJlPHs%oMt0q(b~=wgo;PR6h%K0^qO_lnU*=uyR$meyB|BrU4WN({!a8RcXX
zX?@tLCl9Yplp}nBlEHp@iWOBS4BRk?miZX2_tzL|r)16>A5EM`uy<~`xcEQlIZgmn
zak$@%MDH9*3KR;IpIFC$;rKK3EPNfJb2rs&Q7_Kk&xy%ZEP%^9mTSH$4pGIL9k#U1
zRmW?gGGWw5*_4Z7q88P``B>_>YGD?WE!u_=a7|t*feb*3C&ZOb%VNZ1GO=y3%M6S>
zAcnb)BzXK;5fg0F0RM2<3jM6L(qt>)8rB=7x(KQ&!MaRup=AhyXfPs%@H1?1OJg>2
zDzpO<Lt(HwbNh_UR-7nyYRvN-Pvg4Xv>58Ys)Q;M8DdQFJhw_kt`dopZ9Q#Em^b!g
z+c&D#&2UI&pD1>;O%~%s1GLWPV38%S-Va&2y%y?g3>>E%(E(Hzio+gjI9?NXuvB3K
zFjU93EUGZGRU~D=5u=%?0+KGHR1~0PP2vFZKe&DhTP_`IX`@KfM`eS7`*qLOQD?9$
z#$=0BbCQ16<oYaZIvPp@FlroiD~`{!{7k7O4cu%PzCjl>SW9afK2>FbkQmE>#7JOp
zBOt~HChF=KS`!b`Rt*cpp<r){-T(xJ2z0#++0bfDi2NW*Fw@VAy<$RU<Sd5j2M+CT
zp`zyA@zY||Wv)hLHn7izWEcvRTxdIRC!-MW<3Zs13v+X5_GjHKoa60T$6JFf9i@>H
z0gTiF)`EBfSsB6HnN#z(fuEz^YNKs=YRQ)WV6r^SRwtERN2bYO!K#D-S1`wemIYu#
zDB6G($#StlaS}S&(o*A035c({x5c=>We57E#&Iui(cOH6nog8{L}EnV9cl69(N&}!
zv%4hCavFNf+LH~A8^#6%uj~z6mMp~RbrivE%R*LJof^A=ZD^0vFzi+6K<pPL$x-;>
zO4~4_MY?M_x-v#8HeHXPGOZQXgavN_1I`!{rs60zwfefoP$6Y$D7ht%3n^U{6LHZj
zbk$l_-fUa+G5azVt%^bAO<=yzL>#CL*hYZ*KPrKedNTo}pjiP5aiKN1aw9)3CJU{U
zYLv}bZ8iO%@-C`?#qa@YRPS~39~4gCp~ncFYlrCZcl!J*9-$bbvGNc`MYMACwj8-F
zM`2rVGquwKsDN2SrCv2o-DWq7?fKrtG^Qp26cw*N0&jGcHVjR8ojB>$pVZZ0<3yP8
zC5e`-8<FW-nU6N<{BF|u-lScgNxQ&M9Lj5%r27!c(NquwFH{RTlVqG?aK42sK(<0F
zTPvtIEet)<)|ieSDhwM+pvsjd3+p2;ctHd!Kia-l+=IXjSJ!Yta%6~tsF|{$5)eKj
z?!bk({TgI*CHfr97RhHg0Lb5mrYeA`$15}+Oq->$9W?BQ6*ft^*T_E&qawyZXa*p}
JYMG6g{V(aTG9~~3

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/adamp/index.doctree b/.doctrees/autoapi/lmflow/optim/adamp/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..41e0e8394651b94463d894a78250916bfcfc0c26
GIT binary patch
literal 26361
zcmd5_4U8nmRldF3|GV40y`S~r^Vxm!&+Xy;yR&2aHg^XXJAw1XYhzG?96NKpGrim0
zH#0qS_uTGc3<hlHHYrSq+mT~11QdQ08)E`OAxL(9L<q=05C}ymi39=xA%XxBijY8j
zud2GMyQimiW_EqSlJ~l+tLpvKtM^`2y?XQH!1MpQyo3M8#)F3K)YoRsN~P{qtRP~;
z6|dZ^S@ke_E;{j((G$@W8}ZH6z-#(tD`GoPqij3Xif`4U$09aH)kCKm%*ysdfmIG2
zubwow2kb$6=&@+Z9$^Ea6IQL%C$_&@TdsPmvtA=~YCyxRH6p5w%KblNtq~T5b)Jp{
zZi5XIn1qo%!A436>h>-+C^24S{Z*%KMT^ZPlZ~2T=sQcz5R?*SyPQfuR1B4?W)Pqd
z&jWR{CZ2{X&4zkeGRv2JYnkda*-&V%(5J{Aw~yF+>>KP|_EGlGobA=DIlo!2TmIat
z=U<+C_XF<&dGS9$FqjjF&6&;6GaJsFV89$_z#M15oKvq@YqNd;vTnS%v$<sNU<14?
z#W}%L`zDZK9|ON`#{akC|Ks@o1d#-;_zovziuE^4$B)?cHLudFT5mTBn9jaSe4Hux
zW)Cigk=e9ww`UZ!i}sD+sl8hWi5lg7aRD2;S8|mP=%jj=gNMG(3(kqY%(m1JXkzp^
zQQe*fUxqnf_J!tBvugUKvR5qy4XfM*i~4zew(rp&e*EMUkDq!X{x>3U4=QlX-0pE*
zIC<y784`e?S*w};`Z)|Vn8_yPbcEo~ranh(BK1}R$=OMC+bI4!kGb>E=;(!7qiWIo
z4GdmlnAH`}cS5^18&9l(d<RmNM%ZAu;(N_T3lC}T4uQj6GI$tGR-H;{M;Awlr*-i3
zGKS5IMPxG}PId4|x&zWAOI~HY1ySI+%lB4W%DM8P8By3|1WLqQ@y&*vYRGP1Xsj=I
zOYXuY!N@bMGh>#2Id|z%&e|!=(>)EVUUBLxr8Ve^bZ_6?d2jJJ4ON}c^37_h8~Y_W
zmj*cGNX?{QZa<SS|2Z~nXJb8LgN^kWP!(BebwSMiC5+C`Lrb=ug=`QsL4`%k*i3mv
zFd9_1W;*rAeuWL;EnIKFfb1gJY&YJmwMNw`J7IK{e_l2HdbGhtz;vOBiWNT_Mg^{}
zqeTuRH2oDTjM(9J!^5ZZ$Dp3C$uz#$2R1$rA)mAd)#yB*9v!PKhiMp|vya+efPSNX
zK1|JYLqZLnMa2&R;Lqk}l{DY>06Vg&Qkc}&QH1<;nvnLQ5)<;MB;+r<AVkle4N&{c
zW-NIMm7r+a_r&^1n?*3LV4#EA%lR(Zp_Ixss$P9XXiYOeZ7OJeDc?F>iH)q)stZBG
zB>!o+;ah>_U$MAUz8`I4n+8_F1Spi1Q)0%(JzXU2JmEizUZe|>SVoOm9)Ey)0mi*v
zJtR9J)XgMoDNjb-y{>P^j?M=jmfAg-wBRW1gM7~B<g+u^WiWs?g}Gxl3oSgw=f2_8
z3!In}IxUcKVp8}?KV}E(hyUB^K9Eu$qc5evfy);*{fz(m369ki8IW2qZB+dQ1OBWT
z8fT69+4=dC)5ek&n!(wr`PtLcC^~)m^oeQ1Y6Pm@No!{Qo@ry%a#rlHRI$qDy81Xj
zZB(pkXr@YL(dlZX<ePA_bH#NlfW`2xEEr3kR|O95GOK}gBC$^6oDj2K_d;r&cGB6Y
z_no;pgyXSb9Oo=Pes0=06B7>C3gWw9Bvp*#=gz^?<bM4m8xRUv3C#Iq_!IUfOcZ^k
zj2i1RVj0Af>f~asldQU7eXCJ^GP)P2$q|>pr{_{Ufnx929YqCqH`TIlcfUt8dOQ8Y
z9eFOY`ysNh=YnHc9+9^nFNvU24-}Z2n2N%0ims&uP?6s|#dbr|W}qiHNVb<S;QZz9
z8MB?G<C2Ds5l%;*2<b(%@hRMJ6GVpzQGw+!^}iW{z6Qbl96>&hEXZ)x!RjZ)?w9^w
z&#+{I2_a9YNK-&7Rp<yIBRj;fl1?e*2PIQ+68@TKr{w1&+F@aR_11auVZnKZe{`lQ
zD^DL3jh^9++z;WOSc9{P7DXpm;Qm~?*3POQPYHTd6ueA5^Ql%e1KgMJxn*ig5`G=u
z_{==W1%?>dXx7VNvqhoxEyTd_q+tJn7B=f^v~utFsmT#nfOhZ4Tf&Gos_Rd*K-~A>
za|?k|cS$H2Mh<YBBFK3$hAkIQ(7xiUBJrzQ=vm)Hx#?ngpgeV{)H6=D1jYS#S`0eN
zPC4te+i#>jbaGEV2G)HA807`8gJJH7*WcDs+X1hk<|h(i@)}rjg`S_Dx_xI8NjjU7
zq~V=9DQ10Mt0WCllddG`IxUb+sKc_}sq;Osls+Bc-bqZBd{Yv1Tnm+y*Rt8HhNUZJ
zwP|I^WCwy~qv82sDZti7_#;h^cH1_{(B7J+UU|BD>#Wg2`&)f0PSI3W4Gr*W39aHy
z(RyA>t5!<SqG@x<9l}5HtYmG3<v~4^1Nhvc5u~(8D7U4d>`6P6F=^RCMd4$5*r)Qa
zHy=WQ(sm4Ct1DbIgu75JrjkD^;;!h4)7cUqO)-3*fdWy(S8O$&(nI+ox@Z|bX*ML3
zPRzzf^iXtWLqge#+4zJOdONe>egm|t4IcNK^h&!w@y1NAHsj-bbh1vrb~bjw=-0KR
zYPG6_(#f2EQx8Qq=ewZgZ}d<)5Sol;8Q|;Dc(TbiHQoM65A75_XC}htx<rt=4Z7sI
zHcRLtg|a7IHzqb)sA%rnr8D$enfxE{qeB%lu7{#iF%n8AD(0vjicVT2l&z?kIW6>S
zu3~NnI$gP?T`>BBmQ<~Zkx)8OG4Itw(W#g&XtDHAIuM$Sis_q=XB?i?Gz;`ldQdHb
z#BES5*qz94!WC(i0bVU8PQEDunVvwMt8x`hJJcgD>Y?cLh=kIK9{CMD6rD&&C|l7Z
zpV30U=6dA!_~>-us&v8VztWPb)guy0Cwk=X^iXtqqzhWUrH9gi&}8(;uKC%M@nqxA
zYO4LO9#RjwM3A`+y5vyd=6#Vm8IEhlM5+Qs>2>`*J0}Owv_qwGs~(C@sYocDD3v?(
zP;~Mkp=?E|yi*JPnk$v_d~~`{DqS$Tq$O3WR3wy6luAtxMW<A{pyhErln#WZQ>C(@
zhtz{o5oB(IQkg71`YY5ban(etBn3*V{ED7bovZSTXxgDxc})*Rr&T1BPPEGFdMG*p
zkx(*PrJv4L1@Bt-;>AbB$1iJP_pDO{n69_s-{eEng-Ypy%iqvas#PfxN+&Ai+j=NE
zmC^+*g9BQd+JVGmRLTH9hL5M4d{Z)VP!FvKwIWFENv(9!2(hBr{C7%ps8C7{JMoG$
zTfWi%6!rjd%+Y$Hj@=*;a7+=7R~&1#jHu&?jmM8AvtG6{hcBCj2&Q)$kCa5@SWpmD
zi@;v&NjjCty+Rw5aVVSpX*!rVNh6TPR~$<!Ubv^wh{xd|Xc%XX5gVnmZ^W`8+Bq9a
z#|CBin*9df3$+K?@N%<S<;T1N_-HBg%EUMdzIlU}J%fljUaG1QVyB6bM@2jW5WI_C
zc}Q3inot%|JgMU<!Lcb@#edDe!$ffNf{}zE;(V)Y%b05(c<lVB<<CAyG1&BUNtg{A
zA-l3=YL)(QPFn3m39=L20kA(F@4CB>Ql}L5J;`p`1i2I{@6Vx<RcurC%bQ0$x{sib
z{1#9~Eh|Y$$27V&+NoXdqOit|R9K_^y8VXzSw7PC=Qbx%9Gi-9KZtQBp6lXSs6NFF
zQd~6Ee^4?82jxl%wl50MeNy&sA1yk+PqQzL+2MAWJ#is4Y%@TplvOtxzSr;qt3n}O
zmwphz^N9a4?(<gbMrd2cdwG<UvFQ0V<3hdcTaawj9DM+R_w}$@n^{6Z#YjXsxHM{q
zA;PWZ=1hOhxiahdD|6;jFn4Nx{;t{idrqCaI}!H7?a)Yx+_O2#5pX=tFEx}Xq7^Ya
z2=YrEv);EI6vT$(9b4x{ijo&{lw>DQZVcUzQtcxykq@DoWPPKlt_qk!YPppiMn_|T
z?l-Tf%;zf|6T-EA(|02GRiXp>^^<9`l)BHKNT;Rv`m7XV{BaoW)E$ba6DClNzcgMW
z48bayIweGQ*V2OW&P5!PFN{7MbIl;bxR9`61p)lcrP;Dqn=4nnE2inp@xVh0{7?NS
zMHzQz5JHvde}MNNOOkw3Jp8kcc}QG)lWGfXb0fI%C1U0hX2PhVGDVwYf>K2D-BOv%
z(0%!qDIxsZ90~5rD2NS=5~y~Zj9IfO%Tpl#IR~=+6jGA+FT^Y*d4zSRn*Ap5`s=)Y
zdqC9nR|zdKJk&%Ck5;(dFRCUqqF?9$%IC68QYMGr6{%(JkWmo&v16p$%N8zLRg9%|
zqu#7F)^VM{ICd;%zP65#R$#Y(aL0~7b59`~#kLQ%(J01GGUg-%wx@t(cb?qGdrD;+
zVYF7YM4s!^31g}t98?4>2NY^sHfmOAdlf_|-`1|?Y^KjB8?UU;7Z%rj8_<ZQTY7x%
zLaL1I%XBxjCZmAe?@d>_Lk0Z`JiCed9TZJc6XI>e>r`A{NAX){bB(gXtz9@0Jl*Wc
zN`j2@J(<Ac{?aMzRRn3C?9DJP7R_9#Qk_4LbXVjHO}RgTzTGGBM-#a#17iy?dpeC-
z-x|jz8;ThhX7(s#AsOnm)cCAuan&AVmpQAmW0Q6gvaJ3f(8i=Ib==6EgE~pC9SeAI
z3RSWq-}WL+CIF?QWg$Djr)zE!7OxW(ZY}G?X;3WIiSE5X*S!yavIGD9D3MD;z8-Kd
zP`#f)JNJJ6iOHyJmJ_)TMDDv$fgLJ9bm|1%CrK|CPl5fj=u|8h-9@zQkayJ41B8j{
zAolaKMr2AZuY_vLuY%>}FZWl01RKIincx3%e~n&ww0Qcx?#+k`u2?kHz4&n~E}jur
z_cqub`7wT<Kn=M94w6dabuTW9>rqbxFR;Q8>j|+-aVH5WX)9?R<YHzkcHyp4)m*py
z?ybTt?X809amzMgEREhZu?T-GMKG42jx54o<QiqQj@BX&ysj<6A13g)MYua}5q788
z7TbeNnX(4|kaV?WYfyky+8Veffx@0i@8;mYwa9AC!GEEtZrh~L9Q<d;jatk>0W;oO
zb1)&bY-;uu8v}(|i!t~t=<UTA7(=bbAR}+FDTr;BLbBBm+zs{ic{$J!?A{DP+(A5z
zJ2wRGCjr~NhbVfPo=(%#E6LMu;%UQumH+)M{`WQh_qX}qPw~IMLx0ICy#)&UED{Ic
zhM|R<I5h`(1e|caD{ubx_IkUe<1MT2sRCrzrj2!ep>W!urATQRmtn4jtN$>@v1?cV
zkzAuL6g$D|TCp!C@VH_>lUMBfTNxVb_o7cq&DWETx2)z1a7?RtcCcuKIhIqsYyPLS
z_-i%)lYAhuBe&^7;Kz8Q9KP#fHIrSqh(cPptfv+(ztB6rFKidS$sL63?UJMzIOnb=
z{BG}9zqVaik0e^5fZMr`pdeMEr2Wm_>G)FbbYyKzv52+!2?ew1t@RTQ$E&ft89}BE
z?159djLaFMLa@bMxBy=D;x7CL+U3`dMd~Z|8m@1La;x)j2F}ApIr=?24{-;*IuE16
zvGcH-p1AYCpT5)1!(0g)nsfq;t$(f8kZk^GmU*hHE&B;~r15G@8%S+anKp2-5!nlt
zWI_RzLx6h^vCELz3EQ3OC$@{3GNmGJ!N{cDfvkP#Y8T8%y5dIrJmfB85O3>@qCyiX
zl8a>(IBxvPqh~xRiWU+yicGxLMBas_T`!JK@kUt_*-s-^ptn^yRJM3gh(Fvr;)~mb
z_>dH~0^Bp7fPV6>C~$kHV0pVJNQLre#U=Ft*tf*^|5Wd||M+&{uEK=*{y@gT%6<a{
zsz43Ei@g)^(d{B4@vE}zPl|zh&fWS{?^wUuJJvn9DFr(2t#wo4?TnO_DfUkkW-b27
z5iqG2|KuxZm-kOHW+Zk@;_VEDWUEJV9v;b2InX_OByk75dL-W>iuiWMzta=<Nchut
z+9R1l{+~u?uE^20I~UZ)mVJ=@91{d91jv|lcG0WHX^66iZT$k62>lW-N)3&UY@1Ut
zm~_RRiu>|T#hx^#!fIs_p0s|?B^f{pQBG>?OC;}#93#P@YmedvGzH3?M99wSq39xH
zC6r9;YCjEKOq6(06rIt->LGGfpwNj<r42!O%y^h{C+l){xLz!fi?>!0i#(mPT<nWA
zutKRt1NVW@UNmqS?eZE}Otx6{?h$K3K51Ld67Pe~-6e;*Q0JzDO>cR~@|O{w7Z|1?
za<>^W#FS^~F$jcAMLNkktE>dhmsrG#yZx9$Jd%*e?*smx!vFIvXQA#NrHa^N=lvC)
z0y$V1m9u)mIB&>Dgdrlmn{Qc^`@}?U6CRg{m|v@EnDq+2V3k!DrBDgMY$`sB9{F6F
za{n|9aa2K!r&_v5(f@&N=ugLKT>uMy8jUf0UrRx5iOsrq8wid`*p|K&?JsmgdouR-
zgYf4$;aXWq2Tf5E!K?Oqh_p{CXo}eTWCu`EK~vNuC6p65<cJUGw=q`g(?F{0Sg9B-
z87uWZ-M%|9LPiV3Dk7eoS?=kg1;ioIp95FnRehPBxL3uWw!^E+B9pcbDGK8FdMyPq
zFIG0HgP(GJEg|6R2?6}6R|0qzO&OM;2w>ylhCveffCNl3{dfEc;iEFSry1uwB!!sI
z6pFN*s@p01Oq^<chHLB@lc!_1lFo~gwYT)YOKbyiHw(GT+t0crMY}!C4u-y2531Zk
z(5BN0vZRU>?E@ilRaT>mJJ@Jk$Ex79N%Zuwv2xR=1n{LK-TR)nM5KLZ!!qUrx$P&g
z8!M94FZ-N76>QSJhf-JtNxUi>^qOI#8AgltI1&u6SV-m0^MT=|Lh3%G6!nX^dgXZ`
zP(@ttTEzP1kkeS8K(}Hpqr#|sZTeNb?WG$ZltP}q;?~ErZ2vMkKq5{nReY80rS$oT
zj4Z8@4Ml3?u>fh@8#t%p0hIx|IwDFendZB+irC;1PY?+*07*8Gxz`uj9^c~WxT@9_
z3yj>!Mr{+RyX5Nto3Pd#@DE{%%|Ml6V`Hhgg8s%RF*9YfM!%L>L&*pRqHGZ{gr8Ff
zw@L_n_bBf*uCX-3<(YeC_$8SevL8#9?|Xg;k&~6Gg?#h~AoLyFpJC&I=c!%>crOt<
zn){OL!aXvIaH28Mufid2eM9y`t})w3G(hSC8!W2w{JeXsW__h;u2|qW-AvkvpE+O$
ze5>sF735~6?A`!FdG#0BAST%|?gD`$P9+uvT=yBH%)up9#CGD^5pt#>-nEnnBN)pI
z9L!U{jx??4W}KIBG3OpsQq%~nX2mNZk1&e%A#pY?NR-koxs;TVkIe}jN?wP)N);!d
zMDZX*VmTB~5=^)~ASOU4l*pP0sgc!6%T5)<;RggJ=?OrPt(b?lBONJvt6<1?<M$vi
zRO0nwACf`R0eaO!iewwoUqedo2lgFjV-2gJZbcA%NRHlsv;%jV6zp((8vw<@Y-1e(
zvUPqxBVvcBCPx4@ONf=n@Ps^lvelTmb^cE9bIe~Z=enm7)%|U8_n5XOiR@+EadND1
znZtl8n6txL6%%&^nVm6<c%J%Ub`&yMD<zW?6I|S5kZEf-*uG%B9-3<<8~J*xl(RR&
zPfQ>`bj0@XjCUoO^6n}wyQvI#rK(ppC2Nl&co_I#;Kh6cmpKbj`!cfcZ}4B$5VuN%
z8rXpJcx9))3>lDmVLvH~pt+Q*=F~`bk%3rDM*8iG%MnDTwq#W>gV%w9Z6RIy20H{#
z&%9i+kfOa>!tZ~O#>L2AWfSrjJRqypWueVBu+>(-LcgKGhJ`l4^o1lMP|v~K2=S8y
zl!Tqj%><BwWQ9<)xDbIG0i?&-{!A%da^uD5Z8>?N^D_ED4pbkaLh(H1{u>$7Z_=L$
zMC`d^^yeSx_3!bg$wsIy5ALJi^x(BOc+Cx73y~wosGK@Ln(Q;Ur!Xi}leJvbNK#Ut
zNi1q?KJg120nBc6mD~jrT1PrZ{c-Ybc-DlOeU78W6HN14gZvhxI}P{FrMOX-f`VKM
z8gl8p&!zJ~mv+`%+Ie$nh3C==u*vqOWsq}aFS6we6A4a<5>hpBJ;DBT3C#~)3Yo1f
z;TJM8RS)J0liDMhY9ldJ-qZd}|F9Z<7(oXd%J(I|z-&Vimc1saIoL>~U9Uo5AR4g8
zL-)$^=P+_lM-O9C<5DEl1EeE>91DPnKFk#FPKvoG?sn?sYO_M3ze3szS5cs{rFJ8a
I1h;4Y4@GZK6951J

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/adamw_schedule_free/index.doctree b/.doctrees/autoapi/lmflow/optim/adamw_schedule_free/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..5a7dd8f3455d002493813390a903d1d54406c86e
GIT binary patch
literal 17928
zcmds9Yiu3Ibr!|<6-A1&L_3uguaMZhjQJ94*)lCEk&UEi71M5Nv_)fkwY+!8-PPXR
zn|+94Tq6l!J7PRY(|YwIag8QGfj-g#MGFL}UDPcC6lv0b1>FAy+5j!k{^?%>{mwl0
zaqs2wl48<_04;ZBXXZTT%$YfJ?w5z}e)#EQ;vb)i+P350UbLD`KWMTj<zvmDk+hf}
zr*EZmUrAq0>wG-4wxb{k8!Y9=(4t{GUNdBV`cld#sd?;p(W2@<8nH&~1b)`t9=1p9
z(U;P?J<f+?C-zw36F0oprWb542JP5s0S&9QZAJ~7H4~4Sn;~N<^+ngA7uapWsgEA7
zrxNdxwgoKX)Sl+!W=57h$VU`4>wL&_e3q^!8y2rwaU41uNenI!)q_qmA__+vo)tx?
z#QU&swdC7aGim3A4Xbf8WSi6`;iIv&MW0f8%06dL*bmx+_IdvDk{z_zQkeKY3zxQo
z@aEEIpZyF-*MERuv?LK*vXVHk+Rl<>&XQoxQV-@VIewGfUJN4;`O)jglMVYA9~N~5
z9!vJy4}mEA0(ki`{C^n#XYqfI2n3HpN03$LLv70mQ(kWcB!Q>)7Z0D6O&9hT<j8uQ
zT8aIry^!-|-To-JZ=aB=mg~wXUBO46QT!K*Fq4;+m@(Eh$zwT|wI`(nnnZF{Hn-=&
zvoXQ5Q?a#?cvfgO0?&-vtZ@ML4T<*r)Qi9V%Eeb-S$S3eO$pp{IXE8tbTwf0;$y2<
zNO7X1)w05!t56BBmd~guj_I>l_?+_TLdgUJ$`kVwHT=7V`S@$UqNd|#FdJ(%U0R@`
zAsXl@I@~#k(!6m#5^se;((d9a*@RK>{D3SULziu*8Qbag8gbbNmv2IlfmWLRRK764
zZxjb03$+n6ce)S-o`Yeq-BmA)8qJ@cUB+R6tgX;$+l7w&msi_6tHFl5dZUZI#+ChM
z(PCGht2bU00;pr|Pqdlebo?#zHmp$*=JQ7lQ_FnRb7B@+UICc>j+&{{UAU(Coc+$;
za~4)e;Lne@cTzsm-dO;*QdTrtB(&dy)Zd2X-gCb55pV%!HRVF_`!}d5mOSd&vK&9P
z|AddC6z{aL{0$N&`~=GEcH46rPMqEmpSP{hPj~q^L@VuKKMNHfLvyjbr0W7?Y=v7a
zPWj5g)j*h=Z(^bOA22i52Ox^On1D0(NG{>KMG3P#r9`59%RX=a0M?ks`37~X5-WB1
zIvTzSfWOr-<CV)|5A*x>G*mOazvYO3P(*xh_0ovHs1X0h10Y__j$Kgq#(pe#4UJ&*
z_8y53k!4Nc7Jy+6T5om^$<G#+0K*IXEos56&Z*W!?;D-Hi;ejB?UuJ1wJq`y$J!x_
zSa^#GoqQR6^@<BtVM)qS&YdeuyCuac<)IM%HH@NMschY@a2lqb75<CyOyI{b&yh4&
zGrX%lle&BVz_DZf3m$%idnW54P&x&DT<p-t{+zCn2sRtbFJJ4jm_}XL2gC9Ab;afm
zXE*GkTUMY+d1lA>5cUP#4iPOklk~-m9Jsr25|;e>KkTBUPK#kHnm4>~ZFzBddF9f)
zvB6?1TB|QFJ~?mT@#K?p^9E~2YZuwV^5uDBn>kx{Y&KcL+F4sh*$P`p+l*prgR&SX
zYvzS!JJ@F7+R7rnZ3ZE;8uprPMdaAm;~;F<^Tw>HKA5e<GiT@KvbA$c5M%j%5L+D^
z7{2oPD<6yDK&=|HLL9SK=Z!0xcsPtu`&A=rV$5E>3g24nd1m;qH1N5iUCY!aV{yhv
zIaXmY<pYM+UA*OkT<;j9Y^M0YcG@|K-Cv|G=Uf6`oP&De>HUCPLj(5&wXz>|KPfvs
zMSsLbMyUE}s5-n2$uVplX}6=di?F>H_FcX-O?|Y?Y2L_?=;i6x`3Y#^0&I^!ho2<u
z1n)(;<UW*Xyiq~pxS+9<z#g7WA3l#8??b>C5g@T2Be?q^9#$ZJS|Bdw9D|4EIoKl>
zM0EFm<!sC*qnr>K_C?HTR}VWPHRzZWGua@sQ%I$TGw>s2zg(Ywr&1hz!0(<z-;f;V
z_*1nUmi6mvveO%)llvzA=v9J`XU0&H?*4Z1>v1pC6NO%L%)d(`is{&ECb;k5bJz4&
zsQ&@JiP<|X6oeQTwBf<SJ{Gw=u%;p7EeHHwC15@fbhGR)smnQ6!Np$C@x`G_K^aHn
zdY~cdVh;8ns=;EaZ(4~Lo3|`4VPzEfeNobG2VrbR*cHcLNCG<G*(KZdqatRxQ&MdZ
zD};HvdpYYR)nU<06RS!&t^Zm{Yo%=e8C{bN_g(zSG;9w+{#P}Wzr*J)4N{U+P<~v5
zGErQ^HEHF_oWkSBE4BBZI$`fGLW$B5h;TgfK6*)R5MMQM;ytJ7Ts2KqGxC0P?Ne%V
z)ljOWrl6FhHZ1%sO_3_)G(Axb?GSR3NF0HjCd;m2FJVm-n`j!#l{u-;SCd*LzR&hA
zzL%?^REbYPDT!}jS(AmgoEEPd$|1xifgg$3N*;V~u@#zV`X-8=e}UB3t4XaA+gJJ*
z+gsI8s>G(C^dq+KR6{w0*d*{H5ZkcWX!VlTK<1#}Emy&Jt7)l{%lG=1%MYueRLMm_
zDamCJUWisW@t)K3K{cd9$VMV?1hTnLnYrTbx0j@*I@)PUyXteQ|FfFvDk=VZ|5E(0
z8cLNE6_m0RvHRDw4bpo~&xygK_B$j3M<B)7K94JUNp@8A)l_HYobr!XQ(h%W16}*{
zPFAX+R7p}nDM|97eU5H4z5BMy`S7V~3=c7nCE|zDfPOmhkKmXpc1+qanb~$2MeOFq
z9_BUj<IlO>=QD7cWoVz`ezzR9-9UUl<x~3AwETu$+VkxfG^3b^4p7XQGRhI{8=^+g
z9Q~YT>fQpXSxCnIKJCC>B6$?Cm)qH#GwzDm3?O8UfIRl9DX-BP38FB#Wl#<Y@evhx
zvESEcZuSTt+e|!9oD4|#H8Tzx#7R06eqYqRfpCwg&6^Q!PZDR(%i}KK_&60sWFkBH
zYRJRo!nRqmux{sJh)15baCczU$l`C}98z6(c#bYTIA3Jp4LPrEM7LG@g4y~~htk^L
z$~Jco^)syfZQb78rJ?9P#SA~&6OjVgH#@+VPZ;a=clQs^xnIQ?Rp%rHbts11+`H&k
z9d+9+MBVJaw%@n^M#$3s+r0>bNNt42ggnncp2XFvV^C^OVJW&KK<$51u>)r(xpe=k
zN06y@Eiz)88Kg}{lx%IWD8{L`9~hAeq=$wR89rl8)-<-A*hX_yh)W)%42!iF0s|Lv
z)NwbCxh;$rqLt+v355y_%kUTij`WI;0Ha1Wc7nucpqZOQaV=u8z@+Fr8<`PV)b{iL
zHW9gT|5GI)`+ToMv0}8UDBB12lI;&(C`rW#*%uNAfi03lY%Inhg2VMW!}6QPBE!KG
z-ZvPS?|BgG2BSrpxL?~~TaGV86>J)!myjU3QLjYA8ZbCrfry=3%-dNsUV!>!>R?26
zka*3S<wXHOXnTk@G$|lN{S6pBW^CKcHzMm6jSN+2+-wI9qF$MZ|6kCOOZvS-N&0+Q
z(&FYxt8~*1p^I}DE*R}FXfRy)f*5xUKWVjhaEZXUa6wO)$~F>9fci0zj4Ey?l+TF&
zo=D2Aj50hI4;xZ+muazmTIpqd-*`+{w9aU;*bbVADL!%V&SE~K<-nr0pL%w>mw=&O
znAE!It`_R}h{Rqljc!!@G(?}YIAkH$j{m6Cj=K9LjO2b9kBX3ximd!wHk4QgR^SM~
zUc}>gUB(ahyxdS0vU1jXjQTPFm6@^Ukfn<vhw#JrOqZ~wyb6mk`9I1x%aQH!&He>Q
zi@1H=4e;u=@hI~&Mh(2Iguasd=iZ{X(?fV{i%+~qB#ldfel>M>(1t(I8(naouq=3K
zE0{k&p;3jG=Dvd0K3PfweUW<R186nPT~P`P)RoU8Cr79MsV)q;p8$P)6h5lBGvq!>
zMG?;{mi=HsUetsq<HK6x9w+V(P3r0ed8urdPpFT&JdGC0@r<%YQ+}Z9bUPY};K2!q
zu`I~l!efwuh#M(OqgKkpU0aYg<{7;Z-jk8Jd<Y{W0N=Bf8CCfAGBd*pg!GtIDK|9h
z5bWMjxC0rQU&@9OhUU`I7@8iV{wxj6{oONHGPJj)$+7LSG@k~04r6IT^sBNoN~GG<
zd}I%VyKT)Sn2KLeGJ~uRXKVBT2e&ou5)t+_dMwl9O?q6+-d6Co>;AlWJ|>=*#Pf0S
zd_p`g(~}I}Pr)WUO*eGl58snXIE>#xaPQe7oEF$OJY1v>+1h&GGj&O*!v~STq-_`y
zh1@j!KDc{Crs3P!P{K4^?lcV(MNEYiD_=3_{$4NNGC+cZAXWQQSve=?-3*6bTdT^L
z^!LyeDEAYS{$Vwgs`EnyrF4EcM55C~$wE%i2i33+aS|v|DEkY=O=sB^RM!apTyUp+
zw9>ni`*}Dh#UsBSdhxR~?mfInF$a3v_8i$R+x|9Ka2VVEA^KI>b~z=rY471hifP{E
zMP7zwe@7`tPs?6B-TV0SEW}2?6&aQx?{MmUyWG|jh-?$yU*lXwCx)wRNKs2*#<R1k
zr-1*s>WSUsILE!lO~-fWN~1hC&M)4m6W?)urlXCCOZ~0FxhB7LE!+}WVbN+$nG4)Z
z#aRcXjm!(81K;xIR*h?h{w~f*kR&2cImE#U6UP-^&A>l8I0+f<jr#Ju3`Qtgq@<I=
zKk1ZF_7|?Ol@O}s2<a(atjf7Kdk`+V4^wM-=(h$&UO-VWv6eHaz!u3r_Z(W4cN(ry
zsN!3iBlp0u{<-BoQiT2Vy~Ec0J&K`nxm*L5d#{ik+Fg{(Rmf!xoiRBdFHov<IX{~{
z+mmIQm_5Gc_<imm_+_*2lR)k{f!y9Uot&;qG`iF?>LTgcd+neu59M8m1*?5P`MGkJ
zC;I48K#ZWs3D7rUG-n>{14;og>XO?l_eX_Y`w<?-y#}NXd>E(EQi6JZpKyUbLeQ(n
zVcq(-fva@uf0y2bTQA;@!mY0sr8)%Bv93xA)V=odfsfdi{82{0A7=!Jx5E-}hK?=N
zSw&6&pOOK;>^KA$KUHq4{$%J=5qz4J>8`X^b|#Y`b0Z(NZeP(k#tY&Qe!-%w<OLmL
zSrA#tECaP~UPd?UL23AeKaZr{y-h`yscg^lGuS)(ktZAr6Gzlrh#AOC<>4493toEt
z7_aFztchYmv<~peMiNpUw3%f+Pv{z%{mjSIAraEQrzuZb<a2L^BD~o|2HvmIwc03K
z`1nYW#O)+b*X=3ft!*)6G>iMgC^Qj{iejtPMn>C-FbHCx3jgDF$_JK^I44n{aQ00!
zsHtKSdMKTw$Sh@2(^o`izs^r@VgTf0utN16ev&fGaV&4%Cf5u3castFm)pqQ6{q<T
zhB_x}n8)aRKQbxGW{O+@kO4@lgGwx3=My0lSA;!=01Pswc+IwOomdrz`82!jfPY8~
zNus=VxXCt>EsQsb2pdIs&;m&(E!2!d5DiAe5PpsiZkfzS2%e7O_;?a;E?iy^@$3i7
zA59j9L1-e>*Yp??u}271gu3~Z<auF~VKGX|&v$(37{dJtCLy1c<K{T%-Vc^PbadvY
zhz4j~WP?Q%c}n?7&+@ktYm0&7bfw@pehGlz7qUhWHj%4CDX0KL3FYg21QT!*VS8}I
zX=+tKSbmZc&&<4vAIA*=+~r5G#mtluj1?6KWw8+6B@H+wYJ{2#$2+TPM=WUuCeqMQ
zbqd*;LBz~rPoX6Z+!1#ohc3HgEVJoE<iLUuh2^N8B$#j$ASS}usmum~)~IIYrsIJ)
zWX{jf8-O6$O}N*{Ji%y9i2MY8G6ABJ#p?rnT3q1p7&7c^XnzZNqtBk2<&$mJHYu4D
z^AjN_;Dij`HnbhM)1=@J*m2yBR+pAg9WAzZ5R>s2gK!IT=`6Jr2%x12e;wk9A&xQB
zFFd^b82C9EZZ<lGr<#2DNA&QRwic=EO@yBv7H<j|Fa=9?-0}cy9EpLLMIz;TjGu>2
zwvb4rCTB)~%3j*#r=lG{wr-m?^0z$7;Y#sK3rH_a`GmNIZ>nqfcW_H3zm;!#0k#y9
zwdWf)(lr_oyqs^~vS1-v-$WugPRfCu_%#V-26mx6LBsJkp##b;oF+pNB^w>hoEGV>
z7g%yK(oaC#OrbKZ4c5dA-T?+SLl*8XKMUvCx@j`x0edDQ@?>#=-W@)ze&hnW>TODE
zw#x^Ze~W%ff{#gSg6RuQ#K}IQ#?U)?bw=pTG?0R3#W1wklwoT@cATFs)zb9=QH{|y
zvjSx_)9+HC`6yM&_qzKYEwtaK$25gmCh74PRQxkM5<X6CMKpkZ+CsG66)kr~D;z6a
zpn4hrS(aC@;TlmleY!S_Sw`cPY_-)(D}H(+!jhccA$N6QB=c~GbnYfxcEZfQD$o+S
zhT?*cxW434ILW1OlS{|FE*%HEbU5JB;X%Sr7WGYV;3U$o<#GtF$Qm*$x}9Wov4-Y?
zsD*yEHt<sqrKV>(Dznz(rDo$Asu<~XX?&~@VVq*VpY0q=eZdk9qt^%$GHF;3k)i8B
zIiLnudq<y9^>0B+ucx2G1jcO*m;z`*1f7)tGuv0HK9N-mwL9VX4KHbuns1S%!u}QJ
NSy?jjK9Y^a{{|@5D!Tvx

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/adan/index.doctree b/.doctrees/autoapi/lmflow/optim/adan/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..6e8c622cc3b9e0dc973c06c8085de38b9ebf4077
GIT binary patch
literal 50960
zcmeHQd5|1edDlvNYgf1BLzXo{*jgFwt`6g4Er~J)R~#eRiCvC@WX8QSz1!XD=^l6Y
zXtg#58(&)J7!tQ#7{@UV=7tzT4vrW@AOXq&Qjj<S5>g4d2uI3P2vmhs^1Y+G-<x^!
z?VCem{=iipO?SWde&74P-}}AS@0h)7;NgKUtf2qJYs0E#d#9$1a@q6CW|+joWxrIb
zm|m1Tl<c}YxiguFM*?FZ^lL%MOyU)gqGZ`_IWWECmLy(7l1H{1PVxLhp;?M--^;RF
z1J<B5bW1W}jl=_y9l56PC|>VY=3IYa%CAOt1<WwKgd~Qv{WqDXh!LtJ_a)4-!Kwjv
zM2WRN9w}yqTdU$hZtsz}-?cq6IZ~T7;!z`t0(-U=K|M*mR@vo{)M2RP8es@G;roDR
zRM@xSa;=(Q%o?Q=fjLLg)Z(GYm?w`CYpu1#8ne!~R#{u)>!&ThVonD&&ohJR1wS}3
z{hGtqK;_bZz`$^t88&UyBHyUm)2#Z_wEEMe`qQ>oHcw3jA=K)E>sHogtrhVAy(hSh
zdDPklwXn8>w|2n)7sCHL;s0HvM&OjdruCSJ`>Tc>B=NZwzg%<8D+`StkI!MxCL14F
zgGZvos97(wCiAKuu`U3ITB{j9@&K<(Z^T1abC=R!ZqBcH<e;ppnA=!cM$@_wXyWBV
zEV;ECyfRF^vMw@aYpxL#OTJqSt7fSM2ldnR@w($j@7R0i9sBM~|4j(o^*LO8YnOvM
zv-gsj10?uitx_?9#X}Hf@MXN2H$>q4l<*kXFI*L&#&YwS9)<s|0@{p@UR$ZUCTX}~
z!6;N0BR?ovg_A~L8(viK=L#g%RNA-(>PzB`8i@y^`M|GL8+=Kc?+|#pMMn=qmIb>U
zS;=*y#J3*!_5_5_PpPsTm4yOcD;7Y+WY#Y)HW&(at_u8xhWk{1NV`yG8383R<^!W@
z2^r&wnd;(<KkLlgO1xJ%(9k}n$o)gN9;betfTlWDHNCR!%@<FBT8KsM)GR8E(2#3K
zW?;BN8P?;xHT4tJByUPjHg8JM{HZ1J%Iabg4^|f^p*o2vss-`t6A+fiKqH$r!*~$t
z0|J)B+m?~81o{h-v|`wx<sXWN;3`_ILf^59K*y`$+B{Wt?UEfOr|IJbBk+=?cm$ly
zgh(m+@i086Qk@*37?Baon^6*PR)>YQ<wrr7d(nok>jMuy48-1S4d#J)SPYEW)VU-A
z4_RBS&w;j)a^6O=VbLHN9)yH%0l>G`wItqaTLbZhEl9#<-PC#Io);_EoH=Rbj`PZW
zy@kr*E?WW*+`pWg?uA4kIL%9n`-!S0(A9!dZAd*=Uu1lipuq*#_vV>;8uhKH47qQu
z&n+g3M^07TnXqb*q113SFhesqX;PxT4)Ugb9JmVlgxutOgkq#!y?BfF`o#WGD20zl
zvYry@rbn)+d33ZEGTGT23wN?eJ7@Q;SfL+m=&hYiSq^H(I^gLIbv$i0c(Gn<9V1nN
zURWrm`;B<tepN?o6;m_g6Qpk~h9KsyS&U3C^n;=S!aG5_ib}Y<0Nq6y{sA#78Yk!B
z+tk$5E}E%Qb0fQzJk0%Oip@JkN4}nDtJh4sUFTAWNrydeTE>0#hP}D|Msrm<k#&R$
z->SM;MKb7d3&(jJj`KjkzYvMzG!*b1VsU&I@0Q<@S{J{%E8r|z?@UW|juW<%x0cQE
zMuoL`Pwl)eYZt0<?}YWi<@;&pwKR^+1vl%et&fXcv136#{odI6wDqj@nIqO`;|m+R
zS5+lRn>{1Ft)iilUedXl`T=}?0p217_meE5kmVAa>T3E5MQQ_GvU5AhL3?HAMxGfH
zc#;h|X6FWGThmss%&3>4Z*ks2i)kKe84j-v^p?(HW>yQW)y^ACgQzJRYeTBG+v?R_
zfUa+;Pm>t3mb#(XV5-2BkCKG+y}td1`r&1w*Q}v_3i7qnP+zKOsEu_boat#heLr<W
zOYM|o5xtKyo3zs}lN_v_zK3VlX{YaEwsqQ0zfFr-zMV2Syf%3|#nZvSmUqX^)76h{
zo&5?!ofU9XHI{Xz6{*GHtTPG;oYf>1kP1ey!hp@0g4^}u*pazxyqS+=rJKlDmJA}1
zvFzHzbbhOFweP`TEy_l;4fpeL?RhI@n&AN?<Jz;c9CTcJE|@fx4g;3aba_&4kc}&o
zcsre0n4s^q1>0+!o!iy?1y`D!6Kv!2b1UM0n9^;W79;Z%YseGm#$g^ass+-N?0v+d
ztV?x+gL|g-?Ag1!Fl$CecyMCR)D@TSF2LtL`gz4J7<W~}gL}=%J#c@)wCAm;ST;+>
zBFVVBP%%yw$w;~A`9X!<^88{o@J}7Q#&AP(cVW&COrvDY6lQ(jEgUS|6x7Ug=(ZwW
zOD#1#&yS3{iRF0Tu>(6Im@=Cw?4%Cdd1!axKw2}H6NLAynL?JNu=CI%7^H#xPk<T?
zu&IQc4X?^#oB1#E5-UrXi;4RRtZ&&&zU74+sVhXLRNS|a)VDO}H6+Uxhro+3ef1*}
zsj)e)WSI_SaFmZ*fR8W-#XJU6`ecqeoyLa=fenovL|O^eqr6sT2BYi^<JCZ<NqC*5
zfblrNr{1O4(qxldixN_c5n7A-DzK?9^*orfu#B0*#7t)6Fe!aG^J<8B8>o5oRnUZq
zI@g8?d%@LjOkWaTr1ryUo;LK%PSHlPdIB%9i~=iI*oY()>zyto+zhjbEMLx#hf&AJ
zeO@E$A7Z|JEBu3X)G}XwgJoKxnVbjVpHzI~;ndNr=1w9$T<HcW(aG1`@ZTc^HXY_U
z?}Wz<?Uh^p0eD86+j`0b;y1qvGqy1CE;_mIV2TXia^O#+z;U17=&?T`S++PF7iQ{h
zkFISAtQ(+GodWW!9Q3O=D74!-17<FYCk?k|$_B(6!&<fK2T?JEWsB&xnom~ImZ0Bv
zzAW0+d@Q0J$#a@@G^)rB(8O7d*;QWS@1iwE<NGg=wKnTK1OH?sYZA<#;3!{$#|<pv
z(c~y!5m7cY_OWRNRgd!84y;1i{LOkhmJcd3uM0sP$;Jq(D6WF1X%*@ByiOP3b;4Tc
zd5~2zg1d1PEP@<Gir@ep4W%{WS9zUgaI`k!#Ej@doYvJ3;8fun6BDJilyByhy%Dc0
z7QpNE18CqVSO7VS6u?y=q-n+I_q;wnj?_l*m;qe~-ui~Iwkmq#QmV9;b$9dnp2q8o
z#qVDI_&tK7VDaN9y77AoM`|N}%z!S$Z(MpYQpIkFr%0ogUFY$72Cp9$uTSX5>p2_+
zix)?c;x#~D1=CvbtGrI%z|q=>88e~_G25hidsT*xCrm3jlwRl6{1IMFEObAlp@Wyd
zP1$I)O6FQ>4e!i-GpPmeEyrHjkBuGTzN$PPBm=({B+C|uK|6=VAxx9E<Mk+r^Xv5k
z6oSHEk2#7otnVii2CO#O5)Yq+*&MTnc?EZ%*y6O7xN5@IWAZ%=eAYnPf54FF#s;!G
z#-6iF@@@v}S#j?InJaq~{I(4Ys(<xCN2Dz{`W0wpbsViT?mdj7U?R{$U5?=>8l;h=
zrk_kNrA>fc<Rr0itTu#(RkSUk(a8kc`si9Sc@r#VZ}~+!9YNpD9)LClWNVx#g26xA
z`$@K(owUt`5^MlX;<f45qT!V+>0NfY2~M)_C*6~CUXktO2^)7f!X`^dXqOXb5o{9{
zK}c)Ap9DZOmrXe3u5eyXMFMs_!Uj%w=}zKNGOGhyi(ucWys;@B<hzNiXPy0|+`)Kw
zuI9RQdV_(F79+nz{6^-BpQZQihYd{hZk~)7c7XVCE1P8l1Fs;LbeB#x;Z|Z3&%(<;
zbJ&EHZ^_y7N&_~U&J?nxExJvQ?|!A5K<(SjVCs6Z8<u>#l@gUqR}S%<f@Ht9j;vND
zg7L1_0IYm1`=g{#wnDaT8RG;~K2~RnJngBsuKhHWQLl%RET>!0>ejWNfqeLK*^;o!
zjIC?aNW<FpvQ=ba&Emcl;!b>y&C-(eWGffhrb^O3#T^4PML7%3?8YqTuQ{f*RW(~H
zx>U4wdoyeAr7IK*Mp%GJ`>G$BWmu=6n@$TeWLGJ0^`u!iVummc>YpqehE333RI3!O
za_9ZPj;soq;4fTbkmm)?Y}iy&c$H~ZA$8e=&Eun16jj5S>1iW4WuKh#gZXJ=Hk{tK
zZ_njZdoI27vdgk|Q0|2_0<*tU=S<jMN_5VI$!^%hCCt4Qw5k9tdHIBSfAoVobh%xL
zp`<!m3+%-CPsl?Q>w6^c7Kc<om$5r!4$vLH73nYltRigLzMU|kWWt{La$$C{;MFSC
zMc5l!*uFjOSJ5=AdH~fwuwq4sY&w;v3|B&G!9L!|E^BuQ&e>0TpQ&OoG@}rfmds*t
zVpn0JafyTyu3|=(Uxu}~9qI&9If=Y5muf)T9-YepIh|AFeS~vRxCgt4M60NEd_l}P
zDW)3})1|(bZ{}RuN&MGzu6x3H6O`=S10TqO4PO!2krhRS?sAxAJ1yEZKux9w$+Pt0
z18FEPJRoaxEPrnM!3<C(Y3Qi=3do9eD>BLW3d75_GdcG{`w~Z53g0{cN#uAp-3lzd
zSm+F*;`M}|GX2zJs@$-KG1Uc7UAiDY;k+GebsmKenFWu-9X^breW&v~BwY#eIZx0>
zF<Y67W%fOgI8Vakcw>W+6T8T^S8-T%IW+D96v&2E&buI=rqf74ze}vh3rO2qK5!)K
zsS|vA9c#_?qeZ@F-MIsdh=*WANw=*#uP2vuy->XN{7H5y2L>0OpMG7F-kxM94VL0D
z{xH2>4=MOaG0YOJ_5Ac+nvRr2@Iaj+=q=e&%Vp5WXcOiG3*N!Ew+Dd<;3RB9hCzD-
zi&v43{=KA${t3l&EPvxdVZal*VlN9H93sf9PIXVZmrKVJoKh(x<Sw)&*?TN2ipt)V
z^|H5;8{4#X%UZ)<%@n1cQNdV8+oI?|wO-_QHR#17;D~ng;&YI%PA{aWrE;;M-o!?=
zxDp!s+j%@74(+K$T0qJTeYJ4zB31bo`M8^W{2lptQ}*p1__pNSOF!?UpZC+x2k7T%
z`ng0tiFBQTR(%l+wyQeLc}tV#O!9Z!jt7|)ss`N80N9HryU3gDkHIrt8mYXH6-Aru
zq3$$UMX6`B$&S$WKwYM8tGPi9ZT0)$fp*&Jxr4xHxo>B2N*n3w%&5lJc?eqP54r1=
zty7#7dFc&i0I!G@a&VDti!1P{5IT8XAob=a7dOXxiE}PZKsB0kNqmEXrkK6DUj_DG
z$a8}ItDq|BSfq>ZKbHbR%kMv_S^WLy#%12^(t%MLjdrE(FrfpLx+CNpl{)&Z8&W6h
zZe38)8l+QhXbt%MygYcNT74|5!11gC^jo_XIGYT{x-bIUuQ*LXq#HJaviVEpo>0wG
z8S<MbHCcECT=q@$==)$7ugX5Q1#G798rb)CjLj3?M2|?_!jI#qAEjrd(5J=lHkcw+
zmCUmj%l6z{lO(VUk#09_^7vx5k9F@%0e*JCh}hLW>1uy|Wg8)!E)N3slLkU2`GJsM
zegL7pEr8vr9Lo)|ui_4h>?Z7j$-eW2UbZQAccJ9k)#AMC&l+yNn|7ukdOjT}m&G@6
z)xJ(Xy{meZ>t+oaNpsztaH@df+y)<VQxkWDJA8ZRZCO#Y555@$Rhq?6o|)oCDxc+6
z-jfxrniptkaIdgZoYOe^5VV9}uwG~mKOd~;gUbg+(x2BnPCU6Z_mu8u<&;Y=bCx}d
zvJ9PldKj_--}{8gjJU}P{CjbfaVVl;7a?DQ=O{W$@PCM-V5_MdMe^N-Mww<V&KINd
z6?u|Ym|Fh=YKk}t5Le`#z2Jy^y@#&sNwa4C6sG0n_|K^#b&OU9W+eNL35I_QHQd1>
z#ZVN^855;k!1^wV)wYvlxb8TMovk3wx<#nK|DqMpaF#qc)M#yT2GK?G{0pD!{Y@5l
zNkPt~XS-E{5S4%o(9|ha_*E_+TZXW%6s^$NkX2JYcHt-(DLIOcd|Za3VDiCHx+Wjj
zqNa#p?}dE4S`{g^eB6Q>j<yJf(t>=PK(X3(Qit-Pbe38^-iTH}!&$N#XtXvtgXkh{
z`8XeY&rvDJaGD_%k^DxkBah>*z_tfI3RyLE<bya0rXw6hM@K$^qhLD1QM#rh&!MJ>
z@$ZF>d_ff{wT}D)YB<_77)lE|@&gpBZ6|f8BT8qfb);_??SwH}GZa}CP#s|ml{1Jg
z($<k}>4Zm<84;C~tQPX5Y9igs1!WuV6HHLHKvqpbnZ!{r&vO(VLAeq~!32e)bWKnW
zqo#;N=!KwMtBRCbPz=;?R8Sa73xeXKSZzD0LqSnGOD!n(pcT+?maGq`pfHBY8AKOp
z3(EFX9GcFcsw8Dx%9V;r-Q8SPp2R(a$;#u9RZ~`;#!)a?;V3$?@(hlG$qGm5nyh>k
zHASRDFJ$F8RixCi@*UK0w52eV7G&iG6sv6~bto%JXQ^dnV5IGF3PX`q0+kiUP&tF>
zB5hgOimtk-BxNMak;;j1Czp~PxHm8<*#=oPrDQLTf=LNS(UFo@;3$}saFnh|$*-WM
zh!p6BlpIz?N-ZTOYB(w-45bAr2~n)Joz$U}D4nI2k_XTVXgEt&22@HIL*)#ji?pSr
z?_w$lb-%m7Zj!0xi-oD>cjFbr&YwLASzF!}tePnvBR@Hjt_-V`j{ZW$-c;rIRjzQ)
z;B~+h?h~{QT5C(s;V78GaTFbe`v#7JDI7<U6mEbnn={X4N1L+MB>pt_<o8f?k&G>g
z8pmmQiR9mCbmZNpXcJ;C=t2{sluTD=n^JP0M9F<Zvsycf4nZ(Vu7<3do$gi~1*0TK
z(V^taa1@M^9HkRV9z@Z5PRT21bXuS!b3wN#nQp9UO3D2aCHD)IG*CyP&FU6fX)TrX
zaTJV_97Ttccj71*B{@nblzc0S-g8R6g+`|ZN-`I8fs$v>%6ql2K73t$+LWk`5Ay!#
zW2mQ5&OS}OqQ%+I;wTtrIf@Qvzlx(^oaHE;aP}Wj^qzC}?`d>e;4E{&8Q?6|$=6de
z>+c&M<eVK?-FDaC2U#_{{*5>a##xS{!`baP3dUKE(g|nxqv$>7>@<x|3!G&x=mKZY
zuhULxPpoPvaF#Mds<=&_<`n)l)bXh7-bkIKCA%h$f>D^G=ukMsQ7{T~lujsoABx^{
z3g1nm(*lK=3%Wqzk!<6(imy$7T=e~@|4=5ro7Po}iGPfvU`*sFI!ydC90g+{N9lx#
ze}kg;oQYqj(P@E+%mrOw;zq@}0Tn0ZA7kVjo8_CiV*U^6U6iLkp`Os<>Bty5w7_`E
zQFM5EHjaYvl%sUQ(_JWf&v|+wm{UK=sS(aD@N_gk%A%qt@_Vm`Q5T}tUrVd2MaE+|
z3PwhbqC-X-N5RO*Q92>x9VmLw$+$qH(}HL+7j%J)>CbAivs5ZRW<R6Jl`^}|Me+&M
zn<y_Iqczv!<%e+;jF%imhnJtiQ7~R|6p5ED{S4!mQS7!^+2(IBeu;*r1yZv5cY%~U
zG*0xXsCt$%A596%V@;msG=2efKT6{tP+w`$`2TPejK&;AhsNXM=!gQNF-MVTEHBJb
zC1_lAk`v{A-G<uIHmzm2me-)RfK@FG0GMCeqPI?_h965(9J)M{KcwK?2WNfMzo3v?
z((nrkE@aaF1qEx6{WKCm;y<52QSax!gmjkbG<CyoA>9nWKj1+L<OD8Es<rp)CTRib
zA$2)I{I`(c7ctqN0F{3YiT%FG9Po!7`kyD?=%Ii5?ZrMbzpwF+w=I8UUhnUtb$Vj{
zz2wAv!=GK_e@{tuqU;w3=JO-=(&XUoLN@Or|6!@@kbM4@vAghMADF)j?EzXkF#r0j
zD0*Q2C@6q*U|#j73;D=Z^(gP99?S~v__(~n#$Ozlw|y^NMucCMkk=Q}wVH+pJ>@(f
zQc8ziyM0{#DU@aC81tQwRdW&P_i+?#jLA`S#+aYMQLr&4N0G*uogA0{8fr@0L(Ddh
z%YQ``DfOn!e?<*P_aQQr7B+4E7mC%klR6xSE1jj@v^l&6eW}4{%~0e~o<?hvGl(uy
zZrc297I;ZPx_VsxBHRPm_Oq8jR!#Z11V_RA%u#gY<CQoHCLbK7Yw~duYKj>4mbOQ>
zDIYheBBhoO3pE^-4~Ei$e4IqF+ICWh@}YE=T0Y*4RzSm9vKpZB!5AuM5M87#A7|#c
z{QGcMU^?;?WYyG>Kf+Nk9pNZCI`VlO1=A6Z(ls4<9yLXbe=l_8YpO`8b>u%#!_lU}
zP+HKDpQ2c8JE=n*Q94VlBV%jZ9)>U!Sr$+oVGNZsh%VCBku!W;{$kuGn4nw)Sv3Xa
zavTK{6po@JDA(X9n4oZ!t_jLf)D)2jy%3a}RFP5(ii;YK3JODMK~Qc-vD$W0hk~MX
zmRe9AL@S`-ELk5=L17G)Gl(wI7L=YGm;Vs%8BA8*2U#^`<&!uHCMz68M^?UwqhPYa
zQMx88-$YFj>Cg*Vd0rJMwXFOQH5_dz45bBG`5B7Uwv#%P6{WM(va)Vn+v605BC7-{
zD~zFX2GK>@vU0|b%TM9nz@+42$f_wN2XGWjN;ry+l>9P|f=LNS>6(-rM@<na&<iO!
zs*03aN~);gsFX027Nq12C|27}>QG9Q&QeRsZ=n^?aF(nLsFW~<${9o#X-i2f$K^kQ
zR}ecc{~^d)zmcRZdpkWY|3$nGn8JOY)<J9k((^b9rf?iZN8$b%N5K@1qeu$Z;&J)^
zM$yIKwj^rp9+&?SjZO<Y@t6y`(1bcYE`Q#7bO?e`ax-Mr>~tq^6pWG_MTe4?;wTs;
zIZ7v#{3R5<=ahUEjZO=cWG?6yB|A7S@1l-GdnSukT1zEw$5Aj!augj(p2kryN^+D=
zDEST)z2}sCghr<YN-`I8fs$wZxcsM4PotduB=w3GXa536!8prNbU6D>90lVnN9lyK
z-$&7V&e?yb(P@FR%mrtFv*{Kr)xP@1Q}KL9UE_m%Ki#?wZFl`+kX5tmzW_(UILlFV
zIJ+B1!8prNI^pag6usx1ouScbfwRm7UEpl5kIT=Zjz?wpb<{~(vRlPbFbZ=N9SXkz
zN5Lq}Q97aU+fekLQ#hv4X@SDb1zn(Uo5$t<0QDcr#E;OrYBBLoaTJV+97TtTe~F`D
zOynq?F!7&I^qw>E+cY{YFp;^S3rsvC$K?k%w%zHkgshsK{w5p+<0(hc;pt0o6pW`F
zr4ycBf};1Fr+a90THq;jK^J)1{&D%Es0&fiyopv<OEl+j6pV}<MTd+Lj)IYqqjW;X
z`%(0slkrV7IxUcqxu6SV?8<TZ51`&edHEh%b1hze97n-;$x(E8`R6za#!HSO@v^1k
z@_&nBx6R5nkIVlR4NVKAWcBX?DPN@H@;^h}kJ9+3)K^+G9^HhFC@>mx6df9$gQH+H
z<|q=4<%QW6kIP?-+R`?yWt&<)E`JeNRlmxq5#zS#t&^$YxV+-f<rzIL|28=5qdqR5
zThee`egQJId|aL&rN`yT_w=~@CdE<ehSTy#;k5iBWa{X&d|E*Jr{!-0`modT0r^Hx
z%hPY&I4z&^C0@(^Ec|52_af7a!i0L8G@A5LL3p$ic0264C8=32vqr5g))@Pv-t=Jc
zxwYB6*a_=Eyxy(AZ>}#)`qjv;Od4gwgI@*M0R@n<`NJvtaEd-Oth&-y38k;BZ%R_L
z9t{0jP%;y1cf2VI3@>zz2#(JdtARfsm|>XQnM}k3ksldua@~q}G)-fc;j%_*(-*HP
z)!=Wldr^^>btE23@3^*SCP(7o(2OG6n-9VKvR|r2wi`~(1=MlnBp$rkaBF6mJe2IZ
zn*t8{wWwN)k|WmIBk{<*>6w86f0!3a8Yvr*8Ae8>3Ky#b-;cmlbIv}M#C_9AyqcL%
z1l$vlV3c3h0vE2vBfnIfHA*MQ(=f3fj5o|d0iGY3!tK-XcnE)J-m6uLr{?^ilAMm$
zgi!#0_r2s-Dm2hX;w|is>6-LTalwwPV#zha5UK%4JVSP4EHJBn5EWhXqzR5(8IM|q
zn_UjX>&;U(_%AY|T9_sphzH8%Y;7LOTeDyU9{3MZ1g24eJ0lQ?l1VB<?6W=aR?+lI
zei>3F@klM2o4kB7wBh^t^24GT1b$Go46p2(0r;MnT8e<P%;!QW1GJPR-dgvht_bI8
zXlC&mR&H)Xn)`hDL0#r}9jODbF0{Z!F4QTB$6dpluNm_uc)SePD@`wnHwGsBK0&!y
zHmjyr1{j-qdJvk}98?oLVwY1YScCBzqZawa{8qfu2BM9ZYn5s-BO^GL-mswo2cF@Q
z(5<C+D9uUq^IO%>td;%ZsS2UeI`~suerOiOoRlS{sCd#2ZAR9$#j+h5vo5I!w{j?L
zB+%f-pfVve3)<qK&~cuuIA^<1IcV#f$u|HA&49K_(<7yoA>^wAV*x_NF0bnYRe+WT
z%<<-dh^}3+qvY_qo$;EgSuH~QFrj^xsKds{omm5HCygRL$BLqAI5RyBx5KIGVr2Q=
zlpoAPTRMxRqy|9BVq{hzJQ0Lr817GAxaSh^=bB)yR98H?lNWzcT0AscgHZN_8BEz`
zG)FN&3Z|{7;sV%6WXwV<Do0jw-Eh1WI9VxXjWer2F{~1XEXC`>MK3Z=6)k(-a^ZjA
zZ@=pk>W_sL0~*?GHE7i#IBFxkW|u|RFB#mmTT7M!_CVlSdjl_17eek6wJJcY0qY_=
z0{=4^SOWI=CEJ?=4sc!AK!hT!&DJHeD}=kQZ?MMLPsIo*Q<*i((1I7i0?RZbz<}S0
z@{JQkvtrn8(I}UR;!>;Elah%Jho(En)NCnUWqK!*BemIhn5hXgUtpr`m2GGnQ86-t
zc@w%J(no+%z^n*_7Ro4_b417E4bolDr~nk)hSKJ;i$Nl$WqN;z++^P;oO?)DdKdXv
z4?nEptRWwpR+3BdaR+&D3O;J_2+0*KR%zKR-xAHUMAI+Pv`aN>J9I3xK<Ek&SR?Tu
zGe7GCtO5QfMz%phYdds)2eSS(?e5|MJA_U+ISu`g^J9Rww$cu@KkZ;)l8;zw9b#r~
zB32VKowJG6FDD;|$j1c(@Ueq@JVHJmBOl)<A1{y(^20+8`C%f5jHw(l=BmZxB0H(S
z#!EJEoH!hO$nFpkO4G5xh<8{z2p6!qGV8k;;c3!MbvLurBT|VYDJsphK`I~mlORe!
z9nPvRi$8%59VD{k*N6Z@?^*I)mk|$o>Y=Oo{fB@h_a(my&An;_ARNG(5E#w?Go_Sn
fug-2$lCHMBl3Od2MlnzLIa(%S&UKFs7pMLoriqiB

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/dummy/index.doctree b/.doctrees/autoapi/lmflow/optim/dummy/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..144b1910fe2e5fa4e55a7a9a00ba8a8946945ed0
GIT binary patch
literal 19815
zcmd5^TaX>cS=LH>*J@WQTN}%=?2&Iei+9h;A{6#o7IBm^q>R_9U`&XtjAqZ7-I>vx
znd6x`k~S_UhQf)q(g7!#JWwG4LShWTv6D)wNK7gY6!#Q_;sGAOhETlZiAt&<RQZAL
z@1CCNIhWo!8<C2tbatkvyZ=l7x9<MCpB?+elaKD<|JYpIcHH1<-D)<2uxZCBn{0-S
zPRkCG^yBH`ucXhUH8vGlSK_b}HSCn_L5+sv`pw7=(hDh@rRs_6$MvlJL~J(_Hw^OT
z&X_aqOk7B7&J-I<+{Cw)PwbH2+VI0G^{}0|t$MT5YHg)d8<j^Nx33Zwed|1z3f#uq
z7BERtXP!-&IqJ?nHlAU;#zuWNu+z29y2YlgB#GShP6A4avVCqdCMqTxz7@wP#Pe8S
zwZzk8v(qkK)~&{+$ljnj9X63zoAfDl=A5I>jB|^#&pF1<tvF%JUWqzEU`H!g!syb<
z&p-Mw$diA7V7wv_Td_JxXtmuH!GIOcfECVw6*p+wSL;y>vTl83Z)e@v!^U`-!a2cI
z=Qfbx90$Md!2i4O{{;Ro5=r1n<Z?o4Y_x5;QOa&^h0Ts{KVbA>Iy)jhF8BH7jISlB
z)p71|mJ4dvoLj+DjMEf`z}k$gwxgiO4!(oN(s+!<)qS5G)y1$b^!bpDCAxl_V5I2U
z8b};ra>hC4EP*|f5+jbwL?djKgx48kM~@#j+ELiBF~Ut_eai?st@aj%!Z?0hBBiY(
z@B^H4@%mt-*wodQzZ$nKQj^JcWXE=N*-i=9HHl+}^M2sEUnp0h9S3Crn>d?EBUgll
z;=N#0qdKEVKrf2woMCfZo(?9~ddIgSvl04c+_oFa*zF`*qr5&l_~g5vzW14@Pdy|5
za$27%z%lE3>%;1)dsk1B;=~XxE802(1%VK<h0Kh>VAPe*DVtZ`OpshQJ}0N~kIU)w
z^h1HcmCxXcND^x#j+GeA(2k8DOkgHAGZ`j3*40(UvMDy6Y(`<H-NQ|?eh^N7NLEjx
z$rZPmIO!wPBuUSMyC1;-epCv}PGhBX@E)}TqG3I3ZuKAvJoiQ6m7a2LzR9NaY4S4?
z8+OlXduugZ_f{_o^*`NfP3rv1nTt<y-K)X2&9v>H3HF*-VP6%%>?f-CCY{EF?<RI+
z`ARp=*E2mD<B-!ri@rHni(oT3Hf(QuD`n&Dtz}S^+Ny~qc76ln@hUvaKvTuWK@)6E
z%8Z>1ehNoRbz7Djq|Ue41m2RZHmu1$g3b2h-M-rP-G-Z_*ZAivRurV$u)tfZ!g}|2
zmM}+KtK7w<YaB>oMVoe#vYQ4C4_D={!lnI4RQQn*u<<Jp>jh`L7@e=Eqhk-`FbxAa
zga3dtpnkqY&2&RT4PHUTmjLjKQgSmrDkS-?Aq>I-t5~}r?!Q&y28)j*?)@2Y-yecF
zJ<C214!pP%n_fUAn47^pu~D*VDS{L*^~0!rt-DKhSTVtd9|oJkc3R!4^g*=#)$Z16
zCFi}4NnoBuMd{GM7uf#-zao;eWJIJEWx)~kGL2rcxGi^!8lOD(C=Z>Cvtf|H3dS;U
z=qb;K*!!uGJ$tH;GTgLxDsRD2dJ6=!-X);b;=S=0<`lk;-670yjVr(527MyGMV%Qa
zMSco?S@5@qjbhT!8#|CypQSHm>>^H1I#DU;XB?{<#f`0a)p#daY`Et0Nf<SpdJxq4
z7aPU(b5w_(E-V>-v}$0o0H^$*ak74L$ym1&3+*58w9%M<SrQd4phoRv{r)A?yZ`?C
z7nh7Hw!7&hX47t1TcutWb2l`{5w-$2d#h^(!%m$)eMf@mV%0doS$X2jl5tuRi1-OX
z->Q*UF;1L0gD{*=dKTE2u)Kv}nE5d0Y|NP@`ck|X8!@D;;i>4v?dqg#L)gfbv|Gcx
zO=@z~Bk-!uNJWuThfL4B3!?fj=YY6S9)VE8WeZ+m)<}~eIa5TieM29QBJuD0t)S@H
z98kabzsB}MmX;A$bCB!+VZb@d-%FlZQt*Ba1yh`YZuGMdVL-s(eQXC5CkaJ?-z4=v
z7`5j$s2$>{<vQTIm<TE?e(uLYZ*sK}qMKu@Vk-2wvWDkf^06h;VRnxXP4{6lq(sTr
zLWo|`ju0COde!0!VkrI)f4FSl$k|WO+nu11bb5@z$f|^orvmghcL3e)1^o>bwCCVW
zn)^X&|7|T0FTv+QdM^m{URI&pUCs7pBl9(hT^21Zuu0io;lp>dD5@S^4R}SMUBtYZ
z)hvap1vDNE(_G5ka2w?a*m*+@cgfX%6-MuMAk|s-UQ>x5IiHW?c?JI;Xz};Hh0i@K
z&IVFIaeDc5i_Wzvu*dgSjTuzldH)5Riur=~J$j|dgpBj`;I9ueg~M|e*vJKipqsS>
zRbXZj+mWC0UhT*Ik_ML>sNyEw!i(qwBJo-Lfk^xT{z$oIb6qip5d5;v?DZolH}qH-
zfcx~c>QH|xn)Zlie>3y29*RypGbmN;%;S0}Iy;j=DcPAZ9-Rrc^=e1<wczB27P>m7
zL#aUqXJ~m}=c7}Ou(Xmm1f!qTlBzZ60t%wB`P^fLZG_UWHL4dLc~FWgo(dj6uLaCT
z!d`dv>(u0^Hw1ORtp!qrIxMNV0bjvaLZE#J|0PlUxghIhEkp_%H>{4In3pZTW0y%}
zH^iNGJB$)D#^Pe~R41f;f^E1E?<JM;AIickR-UwQ0&8X4Gsas>(zCaM_`lW?uNB!p
zN7ITH=%4gZbQUOsQpE!Oryh#V0%cH27U+gPvqZt+KJI#qS5}nef3(Q#-XIA$LmQ;;
z_ZzJK(aipRx`;ySfCZF(>=D+ykh>#VAiHOefOCD=vm~`<WJwv=Wi93v*i$KLe?SjK
zC(G3b;hkC_KOTe7KS;cw#k|5Glr3@<J6FX5J*7uWH)@#$`VHu59$OCZ!ynK?`6RmN
znQml`cnE9y5j~U&w0uFOWky|HlZsU?EUd#P^{~IzP2tXdSP)(wbydJCMEXYtt_TXR
zQ1-hsYVDlfQei>wm-Y1O<^ivuX+<ymr+O$lFPuTCq66R7L($1i2BqYMhnxp|M+;pY
z@gK=N;B`JaL(Bt)VDy_>QnitC2BnIznI6?f13F_f1T8n|p;Qo>lCjZ5`*-Uh?ZUJO
zGKVxR`szrn!HmN-X%=JV3#E9)x-Oqxd`2ut^scnS5{mpRN^H37+E*G#(n{H!d^D|~
z;gn{3JLLnYt3pRjv1uIl$w5LM6xNU2X6mItF5ks={+VV&XNl`ay#imcU|(>-JBLPm
z2M#$_SV>LUG_8zd8ym@C<y~|(p6y&aZ}4?1XPixLbbOz${0iXHW)e1takTCB1~2;{
zHp_WwQH>D$C1Rw=3pxS>AE#Ho8=P-hG{pX$GCvj^t2xEi@7-^=kPEhI<k<&e@s1~k
z)boU<i^5M}!D}L`uY{G&F1aj%Xm(qbZk^X;*x&`!6=lRNu)}>3QNXslz?Mx_&H3A%
z(-}M;edxAf6<H`5-S?qgu?E>MXJK2(S#Vx=-f;enkEQePg9sCgw_-exVmyhfx}7Vk
zFZ05vzLl{9%lgIWR@|!3SYEMq;ru_A!E}!G+r?5{mWu6dWo)MyHX}FOz@>diVJM2#
zk4(j<7&a<Obxfqbv=qr^-#sUHPvg~Tv2$Y7M0$=~w?98%xAo%U>a>A>vYruc7`f?+
zjU>$Tq>L~!nl8`zv?$@w5LgA9MSX+kM8)-KKa%kI%<kfnagnqt3@jhd+0qQ}{bUCJ
zamO}%8)+`s&^EBNWfSO5Y}YAuOZk^6BeF)ZS6pJkeuW7#ax0Y;WB1Z9W9NOIY75~!
zj^X=#yh6~X*mN;0%3z35^L`h#q-tlD$9q93V;>lPoUZE_2}XK<q~dh%&^YC`aZ)x_
ztkhqkZ#6MOF^(rHN~aje!btuDbt#NwF0z9}TfyWXyD-^}z5ILhu^W5&uV}Yhd$}`>
zs_DEP(n&mrbS|*Hi|U?b_fQlrygMrZ9jX#fef}Bm)B6+rDT}mXZ6Ce<!U-a-bFYO3
zcoVd!qFEM*LDu^YFyntYGqm-+SE+YlF{p{)pPuEPE1D3?U*k~97UPNGtm<$UwQLBM
zhXR!wfl+FvsNKzDWiBfQU`7}K;U2oR{7z~ow7h4_=<cpQ%Og=O^tl~<>_(MOqTOy)
zIU_{X*E=!L#3G@uOKR3n59bgm@G})AC5Iw62Fd3|q`;;N_bPaa9ee3;mGSsFE$_Od
z-g}s8D;UoAcy6WZZiMyHuDDcU!UAW4-oP9gD9@5&O^rhF8^f?54>Iopk{3CWWpV3f
znWFgAM9`&Yg`)Vx3ZeM=Ccjg*6<_DJqB3G+UPKI|hoT@xVb5nOKv58*CQ4ZTm~<T1
z@05o32@HttphmVhNT)J#R5Gjl(@?oxWn6k|G)u0<iQTR(8nr%2a3T>Csf<q8M9Sr{
zK~?J1M_SIilnj$|3hx&HO>#~~k?&T@*sUCshL3|15i%3cTE0&imPm_S3j;gLdGtQ2
zHW?Lwg?SY(g?jw4Ry}Ior_hu4Y5erf!+Rbd!0a#Qo$yS;`>@dYdnyJu4Z@deayd~$
zU$<-56vw+bBcw&h-wOi&F2}9kv`xFC@Hc2$an1H$^iXtbwi%QvYqsCjL(wh2Wl&0M
zwv$qN1p9i^H;M`cKi|~ipxVdMBEn9IOa-W_n}U3aL3r1r-VgXtl}%!mv8}@Oh4FGw
z(*>mh*Q+ex-l>P8TfiMcf==n7RG_7_fICV?M=GCqDJ11>dRV(y!WD!L_F`STLx0gn
zC{zKZ>fS?ebLv`cA6xbgR4E(ffoE2@%**7wO3ScJoKq9tL0N*MX<4=$TcFgl9Q!gz
z6w9&RF9J*N3H+2Swf7#BWGk>-W4!ZJZxdtVJ;^_@EH#=5*YKg#doRk_4EJjst=b};
z8dBF_?}O3*VrB?nAupmqnL+BR>z(K9Xd{eTv0)kF_<~$w;t9A0xu^UPX|>9QkX_aa
znL65JD&!jC1HeDdS_1#xwVU!p?CFQ1O@4G5t8JObk;`4ugqBvk7FUgj3|WyMHjHeW
z@{+L3g+(x|d|_$a_6;j&8j+1d6oDv(Gs9Y1`5)J=T6wlMUEm{?Pc16A^WkB*qpkux
z2nKx!jlmr4Du5-a%vK+-KRX2dSQ?nhK`FD4-|8YznQ+dD31^Ox;M8kUPQ4eXsR)V+
z=Torm#jc*qR=?V`^Rt|T<(8^QD(YTxmR~}<-A+3TT7+*Ds}g50p2Vv@GtX`odOr)r
zmRzbj(flhdw^W+Rl4KLj9H9q)D2368!(dmTS_5LvZ|*Qugtju-qb5Q>v>-fXPDLv9
zYZag<b1G_5m?`J~V@~xwU{rNZB~i)dRFCbjW8<0RUkXPdCpl$~)v}!{F@fsdLd*`$
zdQ%gm5!hzr=VrZWdgTXk)b}7<7MCetKPm|2<E$)R7Kfa-*-Z9PzR#mZwtX1qmF9v_
zmdSe56Tyq^B*CtQ*y4Q`O@Zi^h-}W8?GBZR>{ft`GjbbK%bD!FMA@s)f&t!f`kAAj
z+v(?E{xpxLZSN5Oy@CJU$bS#>-v$2r7WyTrRtPvnwDb_b=EO#B&Lbq(W|vaPPljX_
zKApghChn;q1?xVFiF1zyUuw>2c>%~W-+^6j@=HSOyz_oni1hToPps<li6?H<&>7q8
z;t*$v-IzpH5c@p-FmZ*%CN6al7l(``I3(z&kL+R7vX0%vYlr9=VY7`+MAwp-`6VPX
zvP4`)!X`86z>O>ObZrU0LS-Z3yVFe^IDdlDljD3XkBx_&q}@r<HD?Y-xHoNFFv9b^
z@zSJlA+cI*9PZv9g<%3zu`zfxWg{y%H7-z~<bzA7FrB@2A|G!DXk(o&HleTBTz`cf
z+CT?50ck76*VqBN)CJdOm{%!6!{PJU7?+E*ad`y4Z6iikM@5P4+jRd%Y+i8_$K>bp
zKn5UX4YHdp*4RvB^Hgi!zHEb$d)c&OAuT(59b@zMRTuojQOr(Ul#Vs+_0A^xo28@g
zbY}qiwXGIPrZ5l<n}{L&+yLA%u^t-I%>u|4=p-A<_bu~Wgj>oVO*@Lh$aD~9`8IAz
z87EYcroiR|&y`-rcrRGJ>H5;ug=a!**sSQcz(KaYrTn3*F*`^!K<Z)#Eb{qvA_sgc
z*z8!FHaJcRAbWB74!a?;8)4MMnO3?#0buC1mo+vHCEGyC1vuh1r6?e)VwTQfn?(`Z
zi|h{M!{ag(GZ#iMmKV4<<s4yGmO7l{C0xvTe4ZDzW4qG~O<a3|qJy}aAdGEOZONsi
zfn0kxcIi+8`ZAktOe;|!B*St-DhU*B2E@cLn&QYEq$aCoZn!>(!?iXG^aLQdT!rtA
z<1{FGYhuXv<CY=}m3Vz*gw6AG8{fu(a|hDj!len19z4Nj+jhHHEGPPK{X`qm4%|s8
z*iBB7wByy46%@zy_7)cGf_fNjLYEFxO^yI+n%Hc|@FW<HNt7?&b@Hv?=WMjm=<1$I
zitg{0-9v3H64^^Q<?PzY28RJtu;L^wAHb$?P99ps@BWx%#~_m}Ggr=hwLd#Ayv+{A
zTR~!7H65ID_UWv1ii?kM+CF77{4|4^9cQ?PoTB1%gXxD2D`V}khJ$p01_oZ}8@S9_
zh}xHMqscbE>q%@j2s5w^=?NQdumKs!?7|^36me(0tD4&)*~QUqp^S7R)1?$5(^|Ki
z(BLg#;Mlm3Vw)XC_-<V?ZCrxko4A#VEG|a=8k^7Vm4dAL8^W4xvwe1OnQnGslfs%n
zeIbc(=`M65!G$k$%>|d6c_0PJN?>S_QGp$Lvg7Pfsg!bwcrkk0$Y1CJ54t7_)hDP>
zJlDLRgpPYB>1PH0&|9LPd+4=>pAMU%x;zY^o2Gc}ZC-Pm*Md(yPUX}A?jbmhyn*p-
zL7>>b$ud3iGykV^k1E|K?uClscGGL*qCZD{i9mi--fn=I8DZvJ<Y@5|{5&~}CvSPQ
zF2I*%JM4feTb$YhxMxJTK5#&kkO`6X1Y6Y-l6zhXS!}K2UZGOe8@meg+Eb-!QxYog
z=}@VEc!MZOq1%VM`^vt+9mBvi!VZ})_&!`t;X@EG<nXf-XS4E;V?>@y-vwnwast#I
rqaH&(1;E_Sm5TT0#azVpyFtV6G)bg4$toe`1Ex5$S`=w?*6aTVLTVyu

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/index.doctree b/.doctrees/autoapi/lmflow/optim/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..39608258eb580ff4c8ca8b2e9e779df19f5c6264
GIT binary patch
literal 4106
zcmb7HTW{pH751(@;~C$*m!RmblWdbs9CSUChddPRB0y1~58Y;pHbLJ4fg^F|Fs4M6
zr0j8!6lu`}Yk+T2UjyVNK>kqwPWw@mG@7Z6ZVe199UUH?%Xhv*e%1TuzYjO6KX+(y
z64_$Jk|a|JH>DdSDlSr<+49@+y}y-TmZxs0*`-m1j(O=epb?XhiRM}PY3UBUx)sul
z>h;@($5yDUT~0mf)Apz3DGgoE3M+Z%liQQ&OsdO~%B@ICuLrfQPx!*`zxMU(GWgTa
z8FsfN9lK%Fx=Wj`Uwb)mTT*1aoD@^$c9^wVObd(Xd^DRPG2Zj-STbWE#B(oWX?PkW
zMZSEQviMx{nb#@Ywq<kwsiZ@CM0@m@HtEFuZcHlWV_jq!*W*j2&&R+2=(h-|`3DEh
zI5;+Dg;gvU;}G<?3VJNEgfB+gAfVS@+%2ZG;d+&<lc$hEdIJH_6XfVE{GQ_XE`IO%
zV2}?jst``yR?dVj-Gfvmh2+0_b0x*@eh{CnNU1-uB`fIXbheCbLa!r(v>RZstm9E5
za62Ct(_qeo+P$_SstX2-zYf_77TK+Fcn5rPKMd;h3>g`~g&{zv(p(i;^503>KxI^H
zHSYrJe<X|^xu4!%jMaQj+}k&*vV={!6nqvcSbkGZ+*`M5df#6~G2zm(j^fX5vpCnR
zYw(TR^wN&8C$}oGMMv$;Th;Q8*63EP%gDsUJ@}#-%F<bLtCv|v>E`W|B2BM4S~r`L
zES<iBk~STsoA=95-)`=1E^VeR)LJE;-k6uF*Tv`j6?R5pS9X4Oqd_xI{?9aHMYn!8
zgTW|Fgf0MY&hwR;i#CPa?TIXwMZ#x71`?tvB{U<V&qz{6Icc(VK@0kvzM#L-SM+!K
z2Yo{?xf`Txk?`D7`Qi>pnPCuMdVs>`QccTi*C&xA7&g~#zuu%346-7fa?O+S#fICd
zW{Xjoyn^lyX4aAuJ+FL1Cl#OUfM5QI!LUL??y#nivrEnLoa>Tx`PHxg!zwR<8F_2M
z?U*GIvWA`4q%~K2Ti%CyfHWMpG<HhQo56P$#v*Smu8vFA@PA`-v%{O1B&#uY18PSl
zJ#!DNW|@)9=RL|b;LJ@Jl|8F0lOegaPL=Dz?{~)?#D(^QIBLi4ULy(P*9{D}R%COW
zD~{WP!c42oaxl5;|Cz}Gcxg`>eO1`Ju;qjfCvG_B7^%!E4KxfBh5>F_n&V|xD`nv-
z;I=5;opI@Qg9{P%J%_?h{aR>=w|&$$@}t~;g)2P|jH}Y!o54(`Ebn}L>Gu5r2Ed9I
zGo{n=r8_Vft$Bn2<c~!L1|J0pm%Nfhm%>ui94D|}y`Vl{Ox&L4xzaY0{DLE$cij##
z*}nGNF<%H|(K1_@Mycm|37-~oY<F<UbcQTKLvxlwGDOK@?nCnai3~Z5cowS!TBREn
zc6RoQGb8Z)c=cn%wN^SJmL(D(2MiEfY)f|-GT*V%tE`mnWbMnE3Hp~i2u7C<G)o_^
zepp-Sj(h+Bo+0E%BB0XkOP0+GHs?rpg7>><e0N{-Sm`7JO8_?5#vgPht{<yxh8U3!
zku+$azB}-A87)Qbt^n5dF+cLC#Urv>2?U_1GbX*k!%9+hgD?G3lp9_oDq5r-O-Fo@
zE5oDi$_gT}jV^={0k4NqA`F{i0=8amH+_Ph+e0u0ol!L%jPBQJQA42aHUET74A)g-
z^2}RHQ14w(8|oBZ!I{Ut*#Mht4irj}3R`}3^sYO|c^;ufz>is6C2Y@2rVPIOzHmPw
zYjg9<@fcz=%C9V`Y^3xY?ex%VRt`WjvOGn3EXpx}{OsxbA0VFxdKRyl_mZXgcN+8P
zT;}n5&h<!eJFE5p8;r@O5}OSzo1%phKLG~r1b|GVwlmuRB9nWBT)QK4m07lkNX$v%
z2k>r5<#E4fQid-3OM#X)NYwr6J-kLz#VpJwcM=nWKPY@?He|U<Ay%IkIkxbZnrfQV
z9DvaSiO6QaKurqwJWGJ|*VIJnu`3k|o#?M2=OrM6F@<Suuiyc3Zn4M17+U8MPnnPr
zu3w&XE3c313ly*_XMxbJ-6qd2{B_0+0+pb5fkfOA1bV|pmgzYMHQ^?345xrBi%S{k
z92417xx2R_4Hq1UVQtpF^gW;AZUp(Z7l!9k`jLL1@9De34ZYM}<#5yALw>xb@2?Ab
z;-7!OID3v`dOvv5GJ11qXeT8-Z^vUZyl~qwxI~mMF;zZ$d}m{$9@JaSki-SxJNmG2
z`(0F2e(uKt*z$qG_aO1j-{^&ebR|I#X#$GUsgf<wD%A&T!d82@VsY5)S}nS_Vjm=~
ztv|^;T(?z!!9fMo6yrJo{9$m#N=lRvF<>%n|E89Iiwu8V{t0b{yDl0Qw^IXTg$-MV
fS`qKI;tCwQ^<fEMoO@jCB%VZTzVXL|VmkT{4>MOa

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/lamb/index.doctree b/.doctrees/autoapi/lmflow/optim/lamb/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..9004d181ef9d055445ad7e5b967bf698442a56d8
GIT binary patch
literal 22973
zcmd5^eT*H~RrlKaZF_gUaW)~2lQC`L=Vrg`rU~BIF1FJY*U2UdQ7I+q^St+FcV;~A
z&0}Vsy-O4kP?CC+t5BK=YKqd9w2+cS2=NCh0TMz*K`2y+qJp3x;4fMb;y?Nih~K#%
z^YK3J&fDDtEcxxsoqO-O=iGCC_sqR#&XYqg{?1SD;{VuW*tDI-+Pqn-HT;?t#%#3a
zS6g+f5ydaW$9^GxA}+D9z`Ppzt)ObfY!_-&ZO5wxRwI5aW)oCBa=dU}wjT+tYUKEh
zq`5t058ES;#U*=;4Mk4mS(!)dfLCAf{HybRGji(lo>{NNR2g*#AF<YG5Z$Ug9gC3-
zH%*KqitPhztegzo-ou9Fc$e9r=QON%xm7XQxEV!(Q)xxulla=>)I#E6r0SVrh(cTs
zHO#uW8m+aO=}pC~UJk4ks?%a4k-18bVtdj)Wbd<Yw)fbF*@FwVU$+*5R-<7B3s?Q%
z^1}TO+y~~=9}F;D5CdB<Taj-zodv=E0%w1L*k5oOHEV4?2*K4Y7k9TR_AWNW%Q9jU
z618sy7xocI>jU_I8~z`~|6{}vgc3NMj}jYfnobb2AF2DbmS^2#bV{7vD4xxAKC*|G
zqu6ZOci3|&tIPH+5U9OZ=tnxiQ>uWCoR=cylR1;#%b7!8XNA~AUuI8Uh%`z0jHqtU
zLMo$NDpQeJX?bQ)uKHd%Y+BVm1T@I&v#Ccv_4x589zXGf`Wq9u2UBwF(hg7R((yZ&
zPSNCtt$N)I*3V#;AxkzR4G}s&pLq=J&)k*4ao&7BIF5h!Lv6;#FVvf!MTQ$1RA89i
zsvkI!U7uGLHPqf*s?iu5j#dM|)$9-;+3X0!+oyC#(d4RAi|qK~I0>`?fnLU(`AR=F
z)89-7pU!qb2vYHD>m38dIQInp)sAuwI%Eu8n~VWI%+<hb+L?x|wA5T*@+<DrB`(KP
z9Y!%v<!3HE%H>*ueeP>ojhfR~Ew2G0vRxf5*p-^05zmRN!1OYG*e^>H8YHT5YCpf$
z+kRm4d?ajlb3JCm&Gk936I)q`AgO*C)AFkT#-6svhQSX|D`vNDf>$vX394E*okncG
z!A5Wwtv9h?>><)@FYc|irsq_hD89xYUp0e9yurpGY=J&qk)Dm>ImhRCnbU~OVAYCZ
zHq(7txB<TeG=4t}_u>E~_#za0#vV>5=Edy9SUrVHGw_0a*#0H}jr#c<HPg)oHFzEs
zp98^PYO^G{RNF)Brp<6*Mq4Fi<GWcldh3W{<59`RcluyMFPshV`pjkmc?y*PExmhU
zgM?o(R#J%1LG6|HF4^@Nh&4RFu_|!VY`2#hT3>2!ovp;i*6Q9;*fhyg8f^wvXa!d+
zu9BZYTeXBiRDk?+$f*Njs+BxjBo}oW|2TS)?nbf@YF6RN2l$e2oc9|MEerx?W>`mg
zGVAUQ1G{z=pLi_H?!lx5AJP=GbG}VGi}@+TAs`gX8#}^rx5Ta9a2lNs%dy_(pYvBT
zW0Nk+E;a}+x6@M~6rZ3c<<Nm^7qx<%m-<y2QHfTIdN^x%!IA-|)r^eO#^U_q;_+Fd
zVnt?ny0kcdau!7=Po6wBYgo-Nt#{m-TfA%5xN148c2ureRdYRkxHxNo*?O~l#q?TM
zT7Dc4%$iwWGAh3Bf!ckh7h1DM&8j#iI?t9KOID*v&WG7(_>tM>iL(<wbLs;Te2gXI
zC>QV1Gqc7i#W$QNXzr4cR56a8IRlT9JMS}WNWgHaD`%6LPo$g3Q1q2q$Jl_O)DKr_
zCl}i~$s-LLxEi<Hp?iUv9CC?#){K<xC@<I@M+J8;)w1t!KPnpCLw~rx&b4(K+KTmC
zNDS{l<kYM89)Rg|=OUw+==Lb(s*;fD=HHgsUZ~j|0D==_`-wi6F2Bz$)Ojv0X}B2U
zT(q+w+=lKxgd1%#%uyPq7~?4QzS-~s&G4rA@VE_GaB@5cZ#WrIzx@B&x)S3O3Z8J0
z<$zvBfe%7Kc8N(PI4QSJNv`4y{4>!m)t=95r-cnPI&I=}Lh=lMbSqR|o1PVop5cw$
z=kP}XH(t8iK~9Lk{pD<}-Cm&Vfi6=HUZ<Y9nRU~D_hmfpFl))et9Zt(@*r0j5@55{
zs79?0FdJA>#Br6<{(UWNHqh*p-tSS9L#`n0egb!iAbP;opKFP@pTy%11|@DuDmg(8
z@%2Qo^F>8lz9j*9#nY6<?`f&$U6Rx<7q0`UYc2sj=T=Kn++nA;K4-<L=Dl?L?W|8u
zUdc-k-QQwF?TS}mnhVMLhgxn6$QlrTAW<g2i8rjs53;5^a5~YX(-}<~-L2DNHsE(^
z(l9mYOOtNU5-GwR-rt=D-{#8T=@9o!6kGBx)u5wVssvsuX3L9&N0MigT^F{RO+Sdr
zA$BmLpKbZHp|(Mb>}^@@QwLYSGHXz1y3?=X9F_7gG{mbVyo$S&*Rxt)wOV=_O<NUr
z1b>uOvL4Fvpq|Qgc-(;qQd=aI+p|>mWt~aITD~yl@G(8@6YaD&pF%ODotVN-H@Mpr
z?m;!hCBIIYyP{`Kx0VEGis^e30Fq8$_f_L5J(X{vi;n4&%Z8*<WZC$<o{DbSkW{v_
zY<x{iz1Ong{s6qEJ38*a(=BcHsI8c-t{Gq9vy*rDwbs~&px@H6szp^vrHIYHtEZx~
z`9660J3W;GN|QsDA^t9^j7{F9=Jv08YA5hGX9$~PiC}dHu;hjwug~28WnZ?gVm4ox
z@;tCdx6o^K^1mUEf)F#Qr=o)xNu>zH9M)6OX^W(?9f(=bQh(ne<_-*}FSoQ0L0{0a
zs)ZOyr3l3Ql%9$XV*234(o-p*G&zVFSX35=tJKUwJ(Vp$i(qjlprywKTsLUho2je#
z%$21aGd;&T-{l&b76g_r>Z#~}MN%mOEMM1C(dme!vK?S~Q%n7Q2bSOEv(ty~(ubh`
zTFa^ySR|Dq!16D8Dmq~4gO~r*Qz@V{Ibhj+JV6$^ObzXS^mMiWErQ7%fR^dbE&Oh%
zGDy`FoQRv0y&DF%?4VqSrUh~2Ha!&`jz}s+aOA^!Dms;rRB|{nNT(n|Zk;=4;$bSs
zk7;Rd8ASw{zBkRz@~P<qMEa0&S<9&wL?o3WAX3*;(Sb-GygaU_Qb1#JATq>{(Uhsl
zyHrOu^whQhC4$u*fD&wpcb@!p1CwD<QSqp5Qzl>0GpTb-zJ#U)0p+*#RCItMsT2W}
zxAatW>LICY9Z>#KOZ)u>ly~{m^Z_V+Ncjg^PPKp{sT2W}AL^;-0HqIJhKIDvX#tJt
z2T%^`scivJ1gl#Dlp>7~D~i2;r%Y!G<>agrw>Yxp`}|K~3lOIqormez3=#pw6vueQ
zv94ATRUETPby+qWRXca~vRNo#cBAp8vWOH5JHuuXxr-f1rxv?cXnRrxuG!z9V~OK5
z0omb-L&;PW?g=#F!8eEs#))Ih#_8l6p=$_c&Iiu1VHvGvzs+|-?O`^$((*ih$Sa7C
zmm|MQa-)cww|Ut!2$18YX*C+yNs{AX5s82S-bJ@O3M>gqsEW{?%wd%fSjkR9z824!
zh+bYYl3+ueZdGj=Qq3caou9LU`3EV)nyxO16~jhYu56icq#tfmRy%R3*s=Zy*k4h5
z?(U=1DTQ^<uv<47TxKYrX&Xu&*Gl&5o5wf0pGF_;8$dZ!PBkSP!syy)m+pCYi(=f!
zL^0ZL*>Br_#An)mXLA<Cp{bbngP3=cxh|50>Qks7#XeL02c=+eOrC1NP6htCPs-ly
z!`;sAv%<@c>&9N=`o#;8VVfaBq^yR~4E&}aS~UvmLU=7gIgN8>RJ9GMocxU6Sn&hn
zUdw74_ge@OMO4_*_#*)#OL1kK``E*eQ0u$zGU`r)V)-wP+fjsAtAzzKSaYt-`@!mh
zSqT@8pIp2X(OD<&I+5u6jlIT_(Yoi`*hgsbB>!xoOfjqo*+Fn$=A8Ayj<KPfwU6(R
zTrxa;p^c}!CACed`zuuYkV}GqhRv{n@l01;#6u9>DHx;Btf2eNE2eJi8-*F+OF%1d
zV)wU*4}k4!S+-K_Z$F&QD)miyrG_TpTpW5xVRd3TOf^9+WMd3Ydah0xG2Qj7$~?4;
z6Z56<hn3U}S|pbeF{~nJzg3y9`t^nCdc_K&z;Bq&LUWxX=NtkB5$&J*C-ocO%pkBT
zM{0<ZQko_2QgME}usBJ8?^11nLbpILe@1exz&yrNhBItRd87b!&o&rNEO@HDlZ<Nq
za~n7A>nKob5W7R^V*&Xpf%VlU73=rg2GZNyQZ4;Al4h!<#7og7;Z{t>Kl1v$2~pSI
zqhX2pqb6ehw1(h*D{Vqk{>=hZ-pMz~tX2G5l6vmjGQ?OOI&y@Tg{p;5uxdtS-DtGx
z&2@ZrU>rH3=0RIW$Q`4b9^SPp)O^m64}P0MZ8TEjlY%)8jowoDvb#ur=X~a9z$kS_
zkMhXo<TEtHG5=`yI@x~Wd`xpXV6i;A9|2*d<<dEHmojB6cQ(PJdEGGc-2E^~LU?Dn
zs;NSMglhB*=PS8CkM7(j@uLY{m9e46ldiZsa{@aFKb{#3Q#-t+LZM;zCB1d?RkSaq
z=g_oc8>4$1*dux>x@`<erN}l$MNdWdQYxrmEzNI=Q*n0MdfR&5G-RxFm#I1d<Id|^
z&+l7uAtFWI+}IRZM^5v*zsIHH>UfWv1;^q&&OMKzy7%HIuaiGP@FZT{xWDdRpnA`s
zo%;#?h>3{lyw!X*cJEj92ud?AZYKaD^HR{zr)5+tC?V?T0iv09pnUH9OqIlo&R7B*
z?XRTr!z=e?jDwBfwUvK+<$j57dEj>TzTwY_?@93n)9}^hgesmBpUZ8qee$8YKY$wY
zbu>&WO*VX0rs`2oL@%_W2p+rGF}M>p&Q(z&u6(!H4%Qx}1Cv4^d#DG<iW8UY7p|G*
zpeVYg=QgE0tiDZ3Az-!rIOiRyFKW8;>vmsw`iDtZ98d3V$J5MacQHu)tE9ueA@$p6
zS`bqINl!%wsgg<&Nc|5z6&<7sDuC3E^8tFY4OHziVecubj&W4ws@@$=(*bqB={LaY
zR^W7cqytWc2`C(;3T-%gH(>BhDcUXLsOmr&R$n;kz5?>@F&g_<>FPst^|j>c*KxJs
z{s#YjmH&R7|NbWbeU1OVPQS$1dw_Sx#F<VZ#5>#r@zk-<HZf0~Cv|Zq&q8Pt^;E6Y
zk33HgcgBzWXwntOy;Ix9y+@J``^LS8(6k`#RrFMJxF@L;!M&!QiVpV#m2S94PqLGj
zx2^0ZQde)X%Y?nBxVM+%o>chm$d?YO1NoM~>{gI3M!PmgQkj9mKC04&eW!qZr=)nd
zjD4zuZD8N)G<uGGZ_pLTK7RFojeWO=k<~06>xq4JE3*9>-lu!HdenZ~FVfR|6jcw&
zI|A$Uvfr@eSGPIGj|~c9()HsvDFptBoMtsk+i?BfNxI??cov%TboNzxUv{K?D<%Jp
zU+43-?`n`3hqJ1-VUqiloxe}U&=oaxaRh&ZrWkUOQ{nIHspw9HC6(NH?I2B^VoKbk
z9Q~J`))r1{#Sn_Z<+JprbjbY<mrh<gx?`-nfOp93v1}K;cu$d5c$i$zyW?y+pbnh<
z1F*RjoV{hV180TVNt9&*{<*;(qYMn$SN5}z?^|-3yTWbuboGM|Spg2D>Y-s8B13^8
z&s4=>6_JoBc4k<y%Bix=fsO2%+IzmiLUIOpFUJ2N*%IUb<Blx{_fIoL?6I@KDo^PZ
zE{#izUNX)i<p4KsBeIdnAh0Y-K49WJ4F#5gLvG76%tj4QkPV?BO3@;m&mBFCk8?XJ
zxqp$RIG$2eW|l2V`9IeW|JehG`yhfdXbkrMAw#+2(Bb?JFsxYEk-k#if3hFmlezym
z7=MT}uGN+7W<NC%vg(aNWFeW&eiH9J1*l{;`>9DrDG%UHNbQ5|;2wX7LG``IuZAV}
z_&>GD!HU|HQxKt+>b&&irrfY_J@Pwp;!W~dj8%A3&(Rh4rufxPcvJboWLE~7O1Z75
z0!p+T$j`v@+ob$e_zMXGFC`4{tF1D?bH&K5zmx$sDMD(J-9>zFEHm1vM>rytDQrxY
z-h%+KJs%E|&!pX!>{BXt;2cM6esqe7Qhj&HJ=oFv9)Zr>!9p^=-Zqz{9JFWI!6+~r
zp~u&UGS0SEkqJno92|;}l*NlL?qcJrj#a~Li?|tJ6V+Bg2@T6h`oevxM5HlfqjFaj
znHmpJ!b6^VaRn#DNbplbuCGs02AME9hGoNkD{8i)c-fvr{;X9CsR{YV@3^TE*hfe^
zgUnic1K*D@RGeI|#cW^!8N0*~C<g2@DvZn9R^Z`oKLt2ZN<?~!AgAZq^a?sa#u+P9
ze2wj=WRN(|E3eTCio{<NA@UM7k+F{-1BU48kSMWa?z-|-oPL*iE)=i<O0t1WC%DY^
z1s4DI+_SD&5aez)ZkzZ{UEU6{1J;@Y`62a9D@;p=@E+J&MSl~Nb_~1Hs1aC*w}%`s
z5mk$rq47BlNUMw>51$gcA!@4?t<3${9FGaRIsdS11%V%wZ7h<Wg#?SkG*pC(u}LBG
zOfN&cmzW)Hd(zg0`(2nBn-Kk`V^FPc&Oc~t%%+G3XkBPSL>|vUwcj%vt1WZYg2X8f
zZ#R;5vg-n?>IXF>A)>T(AVY~Fm)S5(Yz3ijkcd-Ls(|RW3CfODPK(%XMA;!h08VMk
zi84a4yug7T1`T8>LN}AVgsVCCpOT_xXtip78M(4hG=;QqhzKfYTXHR_BKd$5I+Upa
zeU)oYNXZYuh#cjJG7=cvJ}?vFutB5<f!4@s<rT*RbI6D}LsuYzj9ff;1DSWwTMbjb
z7uiiQRpR#I03d^+9z6?bz-(xL9SQ3mm^#WPnpQK703!a7^{)wS$Jogz*o`W_0L9@v
zqBU&4!DEqPc0JYP13=9(qCqe{5l<0jHRo<yyc6=A2v(|X-BU^0{q3rIn5{`Fdl?~p
zjuowN8i0ZYJF0sy+%aS?gB9`Yi=*r?bh2Jfj1vYfB6VbbrVTa~t~Vlct!yJPl}E{`
zVx&w(4z!r<<LMpBGQYz$gm$Ls9?G6yHKk||S8XJ)sbb=Vy+O)cgs6QPIVCrEj#v>v
zC179!+T&N9#tL*m;=(imMcAseRded3yGXSqjFED=UXGzM^@>%41+QZSwuNkf8|-@c
zdFJJ^g#?6N8QE+JaWV7P*a4Z37P{)K2sGPZd#uJ4$_dLx1)9Kop@}#Ob6^`0(ri+W
zL9RCkFeqqN1fWI84+32X$JunQlwvx0F?w4`ZYZ#j63n9d2o;L!lKUDhrLWS@0lewC
z6ZG>6-F_KAEjC7V`SBU$sO7acc+Cx73-5VHsGK@LD!5Y!zZsU_;dg{0C27}ACF`kL
zJdsj1#G)HtBWK~}G4v(kLIzcWIr!ByGW!f47Eh$cgJgNkuS>_QE*-<VbVA_L34%*I
zW-jfhx%5uw(mP*^?ayi+m&JZ$fD{WO<ReN5U{pOJ^lS-P1uuoB)+<Q$nyY%StuU!Q
zma8_VsPdksbNyqj2%;F4e|>vj@(C7ez+csG5x8MxL_$FiY66A8Dm-#tmcM}cdOCg>
wri{p6fC6+Pg!T%82@2+l_a?<$0rxtMs@JNK!mko$A+8EQEisM!5JZLjFX$8?qyPW_

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/lars/index.doctree b/.doctrees/autoapi/lmflow/optim/lars/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..5516d2f1795254279a1929eb46b3457d63d585a7
GIT binary patch
literal 23708
zcmeHPeQX@Zbtff$o+OHtW!ecXdllQ#$s&)G;@T=LDUlVsa3xcY?6`4DXSuxF<L+wj
z_Po2tj|xH)AhN||kS6sSfs>|onx+lv7A@KaNSYvakfLadqCij}D1smf`d1P3kEAGy
zpnnAYy_wmsyW{YVQVkdgVV`zqXXed&^X4~iX5PGy4!-i4#XkPW_6H5iuCJGjN~P{q
z%phVT6|dZ^ne{MwIXd>S=vq`{qrR~gcul`-MywAx%9ia`e6t=sAF(mY9@=hDlI4d3
zvmDx9Jt=MtT0_?G^HI?nWrLv|x@P7Rn{aDY*IO%jjnJ-@T*D6{%8a}N&zS4fh)!9q
zN1|mz4Fl~6BWr?<>Pf?`-E2s<cYzJKcHN8?n#%?&7-8t!%gqpc5?{ORN<cgemt7+W
zkcj8Ox=|BPBb8<&eOWfjSA4Tdd75lEG*;+SWbL=6ti9G<)^6(%yEJQgHFMT)*6XG}
zyXN^<W-mVZIGBt7K?8$X(Xd&g8G1&;o)zrRa`tD5{aL$SG1p6e0Iu%7u&cRj^|3*o
zmJyqfsC5sxunt37_u~Ki@&8f$e~dVSP<)&7QDg%R!}cTgu9{bAy5=KVyTsX@;@eF7
zCu?XSjEtsr+?q*QU9j$kK&?GOKhh2!j}zGNc_~ufnUm?e>^anRPKZs^Wo*d{ktQjh
z71^z6NM(deWjr*Nn-G~^_FO$^nB^V>G{Ez-@x>2cJ#p>o$!qby5s|x;l4DD^xKrm(
zJUD-vIzMREYKFgY7QGBvvPn4*q4TB8XJ~)sO$W#M;d8Elzl-NCKV2w1wjP@GN}xUc
zYah_;y7tt@GoD|zv^6`lG|Hg`WdmbjMXUN=O$#kkYZwjFkB8e}Yu!@EM%hrf;(N_T
zn{;VF4@34nihcw|*6d1XMHdPr?K-4=1s&?e%CntvXDay3Yy|`_%U)%pt)Xb=Zr@vL
zOXrA0Bc)@JQDBR);u{SsQ;-$s8yoZ9vNONLWqP`8#FQxg?9w8aYY~HYZv))|x$En|
zk8D*(yR0hip<&k!P2X@cby#1J<7<GZM(N=C;?{!;fWTYAb~QF4Hq_Xd0XvbI#SfC|
zYv`8G12wlCm~06A0Ma6M&o*EdVFsbBHN&n))}OIqyoDPL%qF{uG~0uBbG_l(Wjl;+
z@Xu?8UynA~D1<GbspB-jM(~}3ceKE1goeLjh7p_W+%0@`zY3^+J4W?|en{{YDE6c^
zly=N3*^V){6fX6^%hn<5J3u?C=hIY7)f-gcQ^@#95d7ViktFAFYmhyBQy?+9^YW<x
zzLyoi*4h;dU{MO-UwaUMTCAIp(u>=P^5e(^P};gCHbBT1VS<L>ZRD=DR>^M9Ah+gv
z^%a4zM(aqcp!8B}>1-x8x?XeVgN8u{*hs@S1JhqMxiWtQW#gF<qQXp)HaWFxjMzZP
zCP`a?+Fw8|(uzsuSH<MM|4D91Xy?5;rkqe(Y?G`lJ?VAlj{d&B?mHgyy>l=r!J9M=
zy)L!rb$8~^Pyq17tjP{@#4d8Ypc%3aBaUqi962*6(>7@d^|1jM&+P^X;r|$Y(E}S6
zW7za_#_w0DMMautYQePT`turWVI$PeXmh2x6VqA^_GCS5)^h37TE(bgP61af>1(FF
zVugCeEE^jwpXz3Sxx-tX*Oooc1%by6H!!EQ&~FAI3@Ec&1^R+mt{j?RHi9(iggG;J
z>R2+f?dO~s^|}`tEhaxZ`CF&&4Png8Ye%`jkDi^@PREjfwFX_D*ODyS(X(e^y7E=R
zBpVcvo@&pzL>CfyCsGx4W#&ZIuZbCRlAh8^F0@pV#~#+d7PXE{=Lsq@<q-Moz=}sx
zY;HRRWN`LSF6+4SK2hiq`iCzJxc(l1{=yaz62oGPRuAz43&4D!O&Lou6P-S#TrDSL
zIvLVMwg-wg1IXY6**>DrrOV%QlYX9yB?T9woQqZ#gbC8w9&y5LnmIzv6m1-#+P7<-
zuV~%@-aJ04Oeno;V=0vp^<TWh##=Jngvuv;WqG1?Ffc@@OrPjxf*ifIlX4j+Vf~46
zsXqOlvZGjky?vB?TJ#0OKdM<QuV0@Kg<j-^oKNDP7~I)-D-a3^IG@R8+U3fX%Os=6
zlz}&>Vm|IV4G8B8_}n(mB?o_kZ+w^@<QhcsYc%WSu-S%g{qynY#8XQ98%o-&ztJwd
zuTzmJN04@Y9dC&ow!pD}QW9~V#pgB#C6GxfIYAEcDMzsLxtO+Gf&q%f*ObK{D5>YI
zmelAMYmL<Imk6J;*(E8y8lWYmjTtGw1hC%B8vJCqyb96%8(P$=cU^RISF--OlG`q1
z4a}cNl*z+#+Xy{BJ97KaB${+4qe&yXR9ejXy>?9+q9Q$M(j7`7-EfD^0lUsO*fIb+
zK*meVm3T?@=ctk<Ay?ICx}m;mxJ@(9B)ctWHX5EE>H&66!WWtzZRBmzoO^$k_tXMb
zPt*zsy1k?2#d#{`acCsY88e(Fq)eYvGObkBGbq|zcK#NM7!OXi1^Qf4Q~7&T(FP0>
zek7G{;OBWY6&3tQDmMc^o|5`j@Z<ahyr<g^&VSM??McM@1v|x_E4+8|mSi{G*^Qt#
zm8>d9rxf%sYDkr)GthiiP31OxZc`eGG(9NIm(^6dP?{rIUUmav;_=3x)1mg)YDy<t
z`Pm*+1e>=2R3<yG+d84jXp%K%Gn0}s`_F1-RkqGQplFx4@_jWG6|P7s-KhJ2sHvz_
zL{hmKT=`!m^|u{Y{)hKY54h5Upbzd=nHEZ1kyN_DmG`KrsBonRUgp$Px=@;KapkO<
z(hhJ%uz3q`<y~9u`*ebqz1iF`y5tj6fcXtI*D8Q{8b!MV7*kC}1u&9IHvki;si<^C
zQn?v`xvr%Cwgb$^c<=N8Fg*zRH6^P`fRR+X0hm8mQ&9m-54?OsO{EK^=@wxAQB7$F
z03+DE1pqVIdH1*zx{SnGV`NDaQdIeOHLEIA`5}sSi7G!;Q&FLcq|yzl{6bAdr67_@
z4pjzddtS({eKSUUOyxK}pqj~dj46Ul&-*ZY(VCw2VS14A14>SnkRqvc11YD}R8)}C
z122!OsdS++IY=4gCpqzOlW(bxJfo(z16UEP-U3*;XPd*PPWUp>Rwm}W!<SSFHMJB}
zh-0H@mxyy!O+|$`l1ev-v#F+{(jQ4Bhd4d%|9)OceaD!y(+!x<^4{sej_E<rUsbZI
z1RY7G8_@ZVnu-c^df?@|YARhQO%8OFTf9F~)7k;#2u5!K<Q!-}&hCUVyOAx%75<ze
z$<NiysgUHqQM5}W85&g0?2ZaaB$aNE<e-|0N+~3jn?aHzO6qSrlDr4a>B)TQLD274
zvZ_Q9Nu?VkSx{3^AxRIsEUKw=p)@%p8RR}C@nDm0>2Pz@)OLU+g4G>iNjHVyW{u-3
zTc<-YJvpn#E1VGcQPszB_JCu$cJ~nML<?Vhasgho&9yQd@*}oCe$<V6*~%RrZs#bQ
zJ>ht_F8nBic8_y7z2hjvu0+l%or}b7oz~arWbGyDfNXQcsd_33=OhYo2QxU-<H$T>
z1@h&9V<+wb<lRcyko4uW-sDFx)({)1HeHuHDG1^PJ@m>XH}Wcblc&81_fejjW}}Av
zD#`JX@U1}uAEH<8Bbm6_l!bd*=CoA^tZ1bUk#mn4aQvUw5?5Rt&z3ED8-Y9J+rMr4
zrAu@>fu5GcbYa1xU6#xo+J2=)S*>IV$By+zz<NDCw{iZ8Dy16{lkA>t8kcFx-?TI(
zZy^<}H@5fwb^a1{v?5{-mQzj1I`}%@Lb>#?t&@-MX2!?Y`j+*k^%n1G>)YG2C{F!F
zzaK=ulgw4VS(Kj~Y3bSk<^Qr24E#J&Eoko0Ew-?9ksOY-Wg~>Aedxo#nH=oBD!E3k
zdM+-;cy+_o7R->Yg9J;uUMiGI8XV@#dEr7*M8BTeDfvu?e^2sd9g;3`qr6W{cz(fa
zG;=zCviCN}=v?ZJ+f3_z+|s(dI_8vWfDd!lD=ju^)V=yf&1=Fbx1WPp)3+mbpB091
zVVs>c{B?V^<oPSJjY@U)#KUu^N^|c$b?Q_JujESW{E*r><&YrR2z+ckch#(*E@;Xm
zE2RCFVXIZ>DJr&l9%R+<&b%6qvVGFAUte~;@|8?yeYlQupl5JX!)TCqJa<DP?TUR4
z$cvROa?q6z!c5H$@T~&R8MC4p<+A6~#RDz$3Y06p5fX$(y<+&4pwx0N#0~@07<#-;
z^9AAdBxw|?o{OtYR9)2wEa-KqK&o6bLTi4$py6+E*<7*fS8>T?rEXTP@ia<b(XWQb
zPh8WE!0mLAd_u3%Jty7uuW1ixMU)B+JT#E&nsz+SeQa7=TuiE1%~r8kYa%}-W%B2;
zQ(ZAdQSVGlnG==OO^aqP;(kWj>cX0ZL8Hxc#+DjO)A8d<`nZ}tE}}I{8Uz|=<NPcY
zDkcqbdCdmCdE|)*A79)jsgRrPw6<!Zv+Zyr_41wO)rjL6jXUucpoZ{-3T<TGC}|q0
zf^3)+;SzN@d}d&6RJ7#{?Xee1+C#Bg@cTWb0+m{j{=XpW2k1O1W&oPJ6$}`6Bhi}J
z>-3kJN1BVjUuxdlrg_5Z2RQZ(_4Ngcn@4$1(q9m6iwS<J6<LSr9uY7rgQ@KGr^x4(
zk4JGiE<571zzl(x+8jQ)u0}UF00e+u7lUNk)Xep=+pL(C0>4D(8XKlRvt|dTcJXq2
zzpjK~iqVDtP>)#;;PbD+6b-Nfz+Ai`X!4;M6kPiX(A%z^M{&ZCN1wZNQ4Yb>I(dgQ
zc=8<}*>quf5IVNCn{d`5$xhwVP_S{7RGGzOON>$NLMM;6>g14gI?LE(M|EB6%1+@6
zf%w`B-CsTd7s-~(zs}DvAe{fApP%9f1`HZhqz2u6VJ_1cY9GT4Qx!I%G<8N9ZgpTf
zvB`85%ht4^a=|&u;2@pLqvK{`Ej!4bTObp+a5=6WBcOr{A2E|Fc4#%1OJ%P%E3J}?
zbajP>>MYC$@iepSxnVF%H(WD+ash_>;h0f#GzU4&*jSZsX^)@prN^CfX%C)+(9Yl$
zhTtf}#2#~=;a3VK9kU;}6`Ut)&GFXSGB)V5Eo?bILZ`>8NA}=OBp!=!XN6OVf25^p
z3ag)%l4P$tOC;#j8Yk|$UEaMH{q!`?zcnGM`h(OgF*vD+=vt-e?mUtfA#FI-1(k>L
zMKbG7ervjx3$cJ3KvRbgYYpEkn}|ffO}Tor*4V)PTJ7-RxI>hAlIim$Q0{*EyiV;*
z*TqsWk3rdYbluLGqdlaOu45v_LL)SFy?9J3wqI}Jm+>$STV4hC?(XfJAy-e7zl0Y{
zG+yyKzXHm!;>(>LXFik0rgE*~tvja8oYfXQM2O&*&zvBee{VK_+z0u@RN4QO<hbR%
zlcMuGsM+}_eiWCk<gKPZPO9SD7LVi5;Ci-QgS_dyLHV0A*$?@x`LqNUvoc%S_O}v(
zJ>Qx6CW@+N!@RBb39tcW`&!q9oa;cd?7nHcY?A#x-^BgOdXRx-Hjs2}x)&$0dQyH&
zrSM~=6tX5kZoBIz;Hho9E4%e78quvk1>Xq+O!2!dKj>Oe1Z>|Y2Bw^L5ji(9bAseQ
zwjx^1hn!!0H@n7cf`)Q_q;}p!P8-&?iTp)0U2GyducEEa=kSvsD1U$yxo^xx<h(}t
zhKKR<I{(DP1w@t~pf5$vAL28+t*w#8V|?2uvyc2Zg!d(>1Q>IFgmQU2QC-(}zC^7^
zD~Mq!Kd}_$zN-Mf*6nc_BEh)_T(Mznq4OvSjz+KiT4eUU?#+m>FxXA4d-3B~oIHbF
z1_ZX)WP9bu_<aI7<konQWZGZ%;<Pv))kO3F<RLAg_$Yjk1~8$VoH63v#+#Tcj$mEX
z=w^<X+G38N@;5y<9N?`eyNJu;n`a0%<UQy7lrSKOiT!x-k^;zi^wTYx0^XNY#ixKL
zTBm@$*_L*g3EIE6+*?A-BhG3(ttIa%JI^I;=y_K75Q?J7-JDvksHvz<EhUxQd140f
z+Y!F0rnQ69MA3x&7|+fW9jCt!a_MYsq^5=1roRq)@&AhLqZ~H5(s#!6v_Wl{J`Xl`
zg6Y4HaxIuHMp}&Jdy+=AWB3z5rqi;UJ7ReD^zXi>Ouy>+HSC3IB1(iN&-}zzJMkUw
zh_dc7r^DwB*@}1~*t)-IwvW%KE~5QkN!$N0rlb5E<ebhVvFFeED?D0TFkg@(dR{xH
z$wzEIBbtP7nsl2Iwm7z=uosLI0ux(X*lv;*6Hf&k1lZEg{c&(8I*(>47E+2ac4U)M
z{@?0_|Ll5u0V4P&3ZwfjXDGL=vCrQEhGQ0PNnI)L-|dC>r0)@T%=s72xKdZL3s@>5
zWVJO0k!hG&z>;{s-UXG+0+xzol=8()uiV0Q>(9`rp4Y8$v*fyUX`9W0*h-5bA|9N1
z>8Y$V=RP#k85zOP{q!_SPe+ocqj=hMiv0H&|9yb}9_PQ){C9?asaLkJrt*!+PA61I
z+L^J#gOUSz&pvNSa|R|82JT20;7>bcfXC~R=3vSI+b>)`5{on4uZ%y@=C}wWWW)hg
z2oPIi5+MLkT5Zuf9fyUS;fOs$=V36X<z8s~F&kF0t@ho*R^tX1;w)}G=8|ZV*f2ea
zQ~rA3^7%o>e$5Jw=tMNh!4P3^+~`6dE5v!|{J5`4-1M`ta?__+m3k7Ha&Md>B2=;w
zX=x(}=LE&F<WV(o&Op+sAX3}2bR|Ar-m)RD88(_>v|#N=^s^NcQ73tH0lZWQ>_dbJ
zLR_>xzUPH#D)uecBi28Qcx$2w<fC*284B{X>AQH_M-Dg??UKI23FlMnKou3xX=Wz*
z2HQt*KCwrouhR^Qux(=jLQXaiONei61*mFDq?oQrn_YoUp$K&zeFkiRk}M!2MJ}+t
zzR9omyC%*%5n78CECZnc<m(`tFxPF!54%cD93P8JgKV&3E;m<D-<Uk?K@Q(EYDgJH
zN0d!shT3P>AuSzeFgPK{dWMZQ!|Keb8Sb-mSN@}J`o8Du7A8p-Arheg;0^9fY`>6s
zrj|ioOT-Sfd}*n|c?QFbjfr~GHpHdx%717n%*KfaXkB1IL@tj5w$C-{D@|j?gv7~D
zY!~9OvfF%<4*d}Vih>S;48;IlU_%&URk&tBB6cNK1$bYL(fO91Cb3=c0z(W2>^#!`
z9WDE)_!v)sXMpMB$e2?0^AxV;+<r=u8iCoYcsk;hA!!^z{k*``vn9Eflo8*-4r~gA
zfx7gH9Z+0LFd|zy91ju<xV>N|Kuh?DL-u_#n_jhDFo$5Llk@~4h>FG!EfDJnwN=pN
zdl0x5T_s-O6*9rS@>~<4*eqy&4dDi#96!p&8fHVs_z(vn#2;cdHlXcjI}HkUXS|D!
z<e=2pz@}xLyOKogcFM^cfSfu`LC`%R57TBgX6~PR5b_-JtL2vJDJ8A`NL)RJtwAb#
z1ukc{8CE$BK*6jP)?5tSQACi#DB`g*N7y0gWKB;7XTpGth_%VagN=G<tm_s|RQRb%
zglM>kLl?2VJc5QULulN9>tPyELx*32Aw_$rY~ir9jE)!M4N~SJMD8ny2)fC`|MHl=
z5(YM*J#cHNSD^zE7Y-0m1kL4^Y<7)w7vavtV5Bg;S0boPZP~0~1aF`PmTAIilHCqF
z&$yzS2p{R{2vkgni=My1CS<5#=&D;4Xtv3AoAp%+_e(xJgeDli&_ryH*ccljBB)Y?
zNUk>%XcROn<X&3PG6eU%7O?}lRB}e*$*8TGywGU{#STXHVM-LwMW@IoRr;C0K8Z6%
zKN`Ir!cUWpQeM8LLg9yb?oFO^ljnj-aG28XM`ncMJMA3H*jcT+cBk9U{SYNMz+@ZU
zAR~dF_lYx(fjGDy>}hI|b(S}W2eISMo7@-GnFBHB0s5i6IKGMK&<f6>6<(9=%jy~D
zd>_15#54$Lh!ld8IG+%4HigCjPlaCAaJHAIE>Aqzl9=Qk&DA&>Q{{yY<m$)F;D_Y>
zb$e@F@(ZSBfL+;Z5|m*cL<mP0N`aG3%(=toW%|qL9cLMxaG^iA9$m%&hHn+10Gbd$
zXNfQl0>^?4EDvjOYG&8DZJ~0<hw4B=5W>L@?bBgU63t{~x#jnCLGgE_6I-4xw3vuk
zP)s@BCY4heHG^^n-Hnq}9*lMHvCEe)f!#QEHo#W|&%Q84D)lBcq3xjP-Jrn^Br`!Q
i4&FO^l4L#z_t<srJb^*KLb#2iR7_S9*NHit2>&0XBzL#~

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/nadam/index.doctree b/.doctrees/autoapi/lmflow/optim/nadam/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..18acbef1d525a8e551bd5a9121c4311c2bcec87b
GIT binary patch
literal 12268
zcmcgy*>4?J8Mk9!@9~zzElJ$os!MK&w?qN8#)(vl%0sA&*aU6LWPI<j&y45JOlOJh
zrlGAw+gLqq5uGB+A{HTpL?2oEK&cRm1P{=PMM6AO2=Q3)2O#nL&YAVz>#^geM9TG?
z?K|K8o%5aV$;!)b&v(gxV$65BVQ<dqjfU+sm>-FOhEorl%nqWLqLWWWSEHI3^z;qi
z3B5XtL>Fe%xnVUtW=D@iVw9!_hUL%2>-&9H4-Cgnm-7nm<Nc3BH9jaRfe~0NHzX#l
z=Bnjv%sFmgH0Nx+p*JI%jmf>AVw;3T$2m_&3b#I22PQ$pC&ZwZqRx9nUySjx=(P-+
zMa$ueE{60V@Qjr(0Hs7(kJ0dnivGH#`#w7HU9t71`Wk42Zq``Q>+2p{r8%MK5A-z}
ziuf2m#z**Z-ovNG$LG1zWb<BV+svEaaJ=>Tk3Dcd$V>hJL4RH$Hm`?)qr1kuV!*s)
zz`SI@ykR%k=A7q)tXnP|2v>NQsK~w?=M+=<tssNn27cX+|99g53H(1vB!MfQAqlC8
zURO7~NE~fCjnHBrs&-(yc%K@a?J&msmV-zS`6)h|QM=4<0Z;j`Qj$!{<4K3;zc1#h
z6zEjej)ezU_bSdQSo)6C(ALE0^J+Su0bd3rU&aG{CA4%;t2>tFyR3d4Eb5i>#rVa~
zUO9dB%9*Rl--xz*FxyV(H(IWXr_U}fkplQ(v#EPq=OJh?Q%uF?2*sbv4M$=k*U~_8
z(TIkJ@NfBE*v`;UVm+1CS+T+lioRgYb3(U`QDmU~VB>XiaR5s;j7Gqt3q!;>8;o0r
zY@9?7cGJ`xfQRw`$X%>Bjjc9BfoG5BY_#=D4?}9-amgS&g}&zLF3&9#rxx9<MQ6oa
zyd?QoU20eVIXQCv(nZPI8jNwoWp=}`*R)MIggmtO?;Tnqr+&)_n5SDgF#Mab!BphV
zQD!dR+1XrR<PsZkz}<>OpSv{+sv?&63B=rQK{{WDE8Nl4L?38^pNd3vHxCsdGSJ+n
zZrBljRrI43Y`KUMJ+!wNMw@NAmQgo?=$ahf&^<fa7K32AayOm5yBNR(>64>nxk;dV
zYb=Pw%{z%l+UzUv!T%Rqz0eIdz6>Rw;(eKPUd~I0?Z{yg!%KXc{|Ziw;CzXem5GEF
zynu<%VZ#?&tSXMtiTmAU(_{RtJ#JKv{x-0AW;aM*#3VR^ogs-{@-`7790)Wpd$kp$
zIFj>wRm-v0lvmSR)w+Szms(fnCyBw$rnTt1IvrvLT#xz8yUe84d>(6)z;aRg;SA;M
zWS}PzHs2KoTf%<`P~vky8lFmG<Jbc-6j$$a>;NvySMFd+wDqUzWlnZ?b>%f|XB#Gb
z5H8KJbcIA|9QrfYqCa~(KlJ(Vc8EFRcI72&(l)Dx-Qgs0vdl^gCy5+>@nND%^x_cO
z9>dA$j?#$c8#s3bp;tIWUX)nX$iX-LnX2WTTbNr|I5SgSVS(<St1ZmkGgHOSJ@=fP
zsWR6;cbd&E+&xp>V8$8`v<9o|Tjv&Ls!g2U>>z9=Jpgj{Wa{<DBm%l^JAvME!WCy8
zUb;O%)LyKfkgPdzex|yVAdGVf)NZkwPN|+ae;&s)c_^L|73JtMK`*6ZOWBeVuVCeZ
zh3Kv(x`VF_$o(xqirON&H=<S}GCx8~j+wN5nc12ZOfdUs7C&V!szo29Kk^hTRc#Wg
zhB&D>hPwqFm6Q7fuZTk(+;~Do$FYpAN79Wt9U5z57#cDQ`<7dZgM@+PuWT2>VUdo9
zOXwJsbhHu?N9P@4wdw4F=l~(AupA)xyCK*tL2yzcDD|J=G+`OII^@{>>G#AkQo~Tn
zneLw7nXYHBNhRAZB@VJ-t(BRDiBmYgsCAj#pD$&n=(gMS`8mb0XYr>jG!*sVX|?Dn
zxyXD5e-e)=1`@}jC^toZ@POqd8cd%V;@>BL(lR<Jtoc<8w^?G0`FAlU4QWEkm-yzm
zI4lN^muGVKxnyU2WgGvbbYs!&v~%o7wB(o>?~AK~Vav;NPM0duwI&<Jt!&fZmTwAs
zT-8G>&@Ss%$chMvLw@MGju&V??nJ?(p+nd7ZE`Zd&f}Fu=OiR2m)<=~a<V-zbdqUB
zPD^Q3t&G;UOKB~Y>_4zHTrq!)KdE}{AjWsfcX<QDZTb_-G2Z1j`CUfx(J&#c*qKrI
z-}1fRY~6cz5h|2!fCvZEINnKeJs3@hliwLlQwR2KkCRxsN2%RWzDt?Z;#~?-tH={e
zLQ&kxXgX8A+dkx^khlSI8Z92OItgneKP92D*qM=fp`6q*cly!2i|=#gyOfD9-lZVE
z?uCRbd}Xw(mG80-v8l~(fY=UqItzCa-e_@RLT77tM)Z^AM3)Kh@x2T1h4Nj>1Q_p9
z5TJ~6XA|P$Rz}k|%6Hp`AQck(669VMA=`^6Y-kkSwe;kMCJv4=;yyv1|Crf+tcQzN
zPX?5uylk+II^GQ;F_wI4x?Sgm^T}@cKT0Ac`JH2$N(=cN(&9+XW2VuF%twJ-ny2DF
zqZ7i<NdkF%)nO~+g*hvO71HEL`{N`Pi6J_GBT<F-iejEj^u@U;{(5pb<9%XaHMA^w
zY*qUYX@OHG-qGd#b=mg}GGekfn?|^uC0<Uedl7IvO--5dNv}wCl?BKh>J-yzJk9Tj
zyFRSrg<`RqrWkPwtn>Kc<Ux8qF&<;yC1s-=+4gvT&fomKMNRqF(!<BCdjZS8oy<4?
zMWE=NW=b6Ch)8bRf46K~yo%QN4|mU5n*RihvO|1M9r9U8^FLTuo|xRuB_{bV`0M<a
zQkMLUod|<SIWmn>o`)e%;%eD_ndVczj~*>({@-GD>^B=@o+9;M&@`U~K2i}((`qNH
zwGLNmm8@y9fIAJOW$xT*%3K~+RLs#N*z##*QRV`6Pt+jRQu713K5-}yttK&ylPIR0
zpEr`9m$=qqXXasV=}|_LpwXg3=3{_tUcpbv6+gZ~A5B3?k=%q6-IIB|DiURhQ1Skk
zA1K`Yas#l3xRv34Hr?>LZ-5I}TBi0LZ#*-5US$3hx|sL(9eDE^rWED9W0JX$s^60<
z7n4#)iwc;t<(zBJP-}DTM?kR5&(_S}16%VC_$lhaAJGFR{U;iw=wZG|Uk_oO`Ia0K
zaW+R374X@}{4>VI;dV4@Cn<ZJckVaAxNibT<=oA`V&NVcM}Yo9cw_()Mb9D_-F9oS
zt5i1K8n)u67}En1q9376zQ&k7HD!(|-?p7u^`;C5XgkU0Xwp5a9@4hONIaOdCom(9
zSAIHa%yyE#WFA4H{othnWRKOcVjjGg$&lC$#!h8_X=&vM;bEA&HwQSqqXQ%u@7nQA
zN=)h&UUfZoZrQO}N59vrHdDb<H8NlEl~{ig@xKc{_vI9n^m9jB{oF_%(+=*geY{po
zd7)}TlA8GLw4IUjYKp^kyS^V_=^iiJua)mocG!=1DIE5DNpuNOs*zFjyYjvEanx5R
z6zx3!bk)(ve@}9!s2!abwfXoCdhy=+cvYG(C)K{QZ_iL_^X<=r(Ea%Kcd)M2w<}{y
zeENv;>C$N1eECth;um94Lksuk%M$>JF&AZA=F80uLchNkKb!P5Kwn$w*JJqFHm}Iv
z$K~%;`TGU=`-J>`l71<$?BH{Y$mDe=7mbp;N+|{L16DEON&=3g1Wcs_$glkpFh=ns
z4jLH&VoY7w(zt}ga$FCU3{{ixgrSmGTRf3wU~7CSsXd#O<ym;Hq53qI0`FA4)&{>v
zW&foUMlsjUHk;P0@fmSA@O0a^<U^z8dd?bNtX0igCBXZm6<z2OL&+T0Kr@8cc()j>
zhaOeLX=(l2NYbO~;KV>2kx}e7kyh`mdh!a|K<UV5C^he=cTv&jgn=6d(J~)HF~%B0
zZJT_7K%;@%2TsdP7iAj5p5p{S6<74lNOaGm{zIWaNrrVy7>b*rXQ6eFZqHOxN29nw
zzaS=80RWGyEZ2Qa9He?e8*j3ktB%)18Of-Rx;7W}UQK$e^#SUb>R}d>J=%s5a7{k?
zfeh>vFNllLmc@w2<THcCE;BIlfEeOBo+RRCMNF_w1N_6)D)h77N`tM0Yk)UOMJ!bP
z1Go(806stvbw<PxeufQhY0O4qgw8!;Fbq~_@1B)8hvUUTjd`BqX<WA(7DItvAEAnL
zgcwsi&q1k3D3O?M8EFAweiDWzMitx)hh+8f;y}w{F-|l<>wFFtS+Z8*przYup}xkz
zaZ2<YK;5S}<gvQrHSkVOl~CA*3JsS<AB=1jsTFX<Xe6qD)X6ASacEhWIDjk;9s|Rs
zOP5$W9@6qr*<qj{!n1W0hAfLQ*(24Qq@Q)UJ_{R;hMEq{8b^_l<1;P4Qff&Z?@Weo
z&~q7JX$`}t$~h1cW7(e=2@Gxo#Q4BOJ?%ql;%VBdVSzYQR8G+sHbLbGJ^w>C1g!y)
z52HXfZC>aW6Edx4F_f!tXnzysPY;Zr5Th=0HL8h%ebyzzP)O!N+krb7g}9jqf$J~M
z&!gL)bGLABx91#h4YqWIW=aGwQwvxV;t6Ca3v*}hTsRAUj(V&07Vy-QfxjyO53|)t
zW!ITEXRu&Z?g3XY&x59g%?43r11pj>eFI_|I@#1x<4g&NOA)rkxW8ow`liNFO<_?*
zMTDAL)apcHM84!{@l)P4<PNi!J<W3Jdd%ACI>&o&9fDW(1};k$V)i;J6}M%Xv8)}A
z-M}`q$Eh3kDs&+B3zOt1{BWgZn$aZPwH#d;BNYR$M^KsO3TwcEw}1g>3|UWc1jku@
zU1KO`vosV$lgEXWu8E1bq#3$uttxM}Eqa)JnTm?Vfbu3VUuYt(Q3h-yK#e99tVz9@
z08-Gb0A<q98a#lK9~YB_UP>6sZa`a28&rx%70Q_2Po3(!W}cv6dI$ZC&|P+rer~1a
zar}g0kmky(7Zo|n+1qmFww#4S!EMw}08p~DgmTKhIRBcx^R^~!m(qxugin+t`v|+y
zH99W5La<bdt2Zfx!m)`k<L4z>ve-u^VP#U;q&vHLlz!-hXVM8U6bJKKCh0zi+Cdcr
z!3)(x&Lo+q7@Y4R3y{6g%H|4+f(ug*w{)hn2Mf~%lU*gG$pU=D1uuwT<wsg!#UlvJ
zaCLPjBu9oQh$1%&Dgogm;`ZMc_rC<$JRN-kW{VUu9026+LsQj;smCjH52xKy*@g|f
aZiNj}?ltmHL#ShN5Sp<OVzo>#&;1XoOsd`h

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/novograd/index.doctree b/.doctrees/autoapi/lmflow/optim/novograd/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..2ef1ec691628ab077d308b178a6f06183f508b96
GIT binary patch
literal 12870
zcmcgy*^ebv8SkF%dwZGQVH#y-;F1C6PUvL@(LmcXiDDpOpovS23s_X&d%926)U7IO
zneGOOm<UYABbZPmu5o!#6LCSo_$2N=n3$+fKKS6fk-xyi-*--(TJBQa(}I)C?K<nX
zf8ROZ`JSo0{NB5V<UcVR`rLANSIl<X_1Y{<#CY3l#U18G$xF$(&m~VMO)(LeJE0c`
zEtZHO%xH1TX$Q<r9#6y!O^+-mTuIlDg{&1>o|`S_HD2dqk0(t&A!?BoIjk@wPB@({
z$J<%){K)EHkE`BxV73!lfcc{zVY`ILplQ!03d6c@0-q@1^J2ot2;d{4p5nbBMjgv#
z$ws_sib*qy0&6plKrhiZVzon}Wvu0xVTewA*IcuszQ)_JpEowm)|G&5(VSR}MdmgQ
zC482j<Wu|<AK|CPhgZ4RVXHyxx-3}T@q#O>4?pq%h}3_8V7RIfTQ%dzGkt4Sv0+uR
zVO6tX)pFZxcO?ix*)5lk#hZLc)MQ_Qg^IiUR#3uk1J};r{~h>$7XQx?RbWhDNm`m>
z)Hkgl5vMy|J9gN68~wR2PO0(b{=>Y!5hZ5K@8Zij*&F;8Fqa=!I+F{1PIrj0`&0Hx
z;V$OwREpT^eTs!@FLNMG1e!R0NloWV;Lf<@&Rk?}#*P^nEzdDRpS7-oQKNFcn0xGF
zPhEKWsk@)ne-i@tXbw)a-v$A%UASj$os=PrI~_CFy99}YpJFjJP}sprVK@=<g_Z%T
z%Vsq*iGLfgrUzk7latz>YTXl4<(UxmXgly?zlU99wPRrBb@Fl?OLnYw#FNXD#6B17
zy8;P$S{n}1+QJ5(C~iPLWYcT!^&kp7M*?rBr(e1qGJ(EJCg3<scpRS>7K(e;{Jk}A
z(_VW*v#_z=*A!Ot_@yTvlRR$1LZ^J@wk>zt*oC7gZu_nyY^x<V=2#I6OsB9P|3+#^
z;~I-3x1?_mwj_7~iJ3U&?<Jz{?=6GWgcY3zar~PQ+1KDM2U?-1gFg7OM4UN<7fTRE
zXnM!A+=Ra(#?XrPe1wz{0xyoE&31jqYFSZoO%Csvft&1$2?#^^qk(ZijAM@U-pPi9
z6Pdv_ixP25wb-O-e+54K|7Fsbhr!jCVF-)7o(t>cqOjONhLcoY;-~p<;qGXkFVV6p
z3DJTVF!2ij_+ppkrNLc0>H~nD<bOLPplbB@fz5LV?euv}g0mRBlNcqBlprF4Knt_C
zy0;W33huGtc<#3Hd}g<nx3T((?$yOfVq&-BtcAWwr=M{@U?B^xGO0M9!de}I9+U1m
zM>#)Cm^vaCyVBrH_)lV&^yrWUvWh4=`-lwPjr%<}f?ErfgIE+j{h4~%3&TT0MGYHl
z(u5BJt38{okSNVTe^$Em=WyqZdI(2|&?L@jAJu3|J8f8Q|HH?*gN?Ry{wUy@9zll0
zC{D7yk)M3?42>9}g=1S32c?r}M`PHen9vEA8cuNW{L1+YON~tync>Ce`IY;Y8u+>I
zzH>_r=7$$Au;ugjE;V+Twap`=&06N(#q&!II>8#|Rk-VIP=E0O(+Sy916$I7YM#rY
z#jM20bX_kpyUxwx?oY0ti4ghM8fPWb&R$w-tZNEzyn#-xHL@v<vzIR6%qNf9i=w97
zVXg)1nM^Z2W!zAE6=I7RZfM2ASH8&uU7M6uNeu5K-4<qlfR>!J34E1J+ciwE>okks
zWv{74@1Z~PR4w&z0eXn|syK!l3mv`nErr*_Vt;3^2^ln;Q}=iVuHOl?DUL&Nmf-><
zU@=X2NG{8EDS(!#`D6t(6Ox*4Qu2^KIL6!F0qBksx(e5Ef^ra=oeDG;B%0DP7!D$i
zg*!=s^{xMd&15E{)HcJVhzDH+VQ5O^LrO$s?nXC*OA{Ay3R3HGUHD=pbH%XRtLa}*
z{CggMsv=5RH=b3CJ|`F1&*6`Dv|=oEK#Fi%6vvJ^f!1sK%<=v%?I%s9pAOq!#&C~`
zrigzBW75LrrLc)@o)0e;c|noK!xuGVeC2?DUI|zXd%X;Mjh39WQ@FT<SgyRK7Zhuh
zJg|nR8ac>6R)d6rZke$Y8COjwW@Yf=hA{ShFNll~*RAN&u}9f}eeylOFJhHPbREX^
z#rn`<T<8rS{UlnG(=@4RE2s6Jm9$pM^&eOoZ`!ZpPo`S~g!pDPlwV`GM|)B!rcnM+
zgfdl(nVPh6XHMaNs$u`N8}`8>R4Ba#BAm#geLu;KU{n()zjK-vj~&?>7qIk@QoFqx
zN|n@7C?%=Y<he#ul(ur3?yiP*1UV@r-U2zzl#gQlgf&&1qG>F5=A>S(Cbi0$K6rTX
zeY_e<mH1L9CGicP*JR-<r)9eu$`Qn-fWHM|yRqMixu5W6$`du6-Q79S&r}m#CBV-d
zUVtxDL#YyA3Z*2#n!GV+;?h=5)7Pt^9YK%^iMK$I)1^yJKM9VdGc<KsJE!mUYWk{V
z_Uhqf_S<SGRWeJVT#wBDQVs10GE+z#NoI#xB;MJ<$;mS4JZ5A^Z8UN2m#3u9QnWs4
z_ukMDmerdQrL3=7Y^Q~Xr9{l?Ps4OuymX>GC>KnL!z`g^s8nFs|HT9kH<)a-6Z_M^
zFUwH!U(mts-z0=0%Ibuiv%_ANXAHcc;86(2-$YE(6$<%DylIs)q@tdtB>5Zqs?F<S
zd@FVwdDT$xCymH!5&I}l^M>qu4vA3Nn@=O$?jd%bR<~T>c!8QSyO&*fTPopGIEgF1
zHF=iG6L-Gf#QW7+BTF>n=-%S#<JzP2#AZFof+v(=ay-GL^9x~OZbU)mr_(gSSF<zY
zt%uo`e@oA|Z>Fv2A#YKf=!-}J?5QrW<-|gh|KQ+kwfzq4QFZYus6#OsZQqG?)%ocC
zLOz=RlE1-!C1uHfJ%})flq0V!<#{9INnEYUe$aeM?b1sY&7V%$an!_{eHW?!dkiCF
zQHaD6GmPfBMzeo{L?^x)9Tst~jSSpf)e~2W`zGadoW@-1k9{7PXq8Ac)_!-PPux)4
zttqih(_^+<oHtdRr{hPtGY=vkP8q1t=+Y<qN$hMtg`bMFO_~pRG}}sw=O!G=pDE&1
zlPF7s%J=u;K<V;c7=Uf0tsHkN19;t6c!s4_CeZiNY94x#{V%xCqWkZ|n>R6~EboDD
z3+1PpLwa^`LhEZ;4GY#?aP~QZJ<k3ikS&w*O?wER_Aq|Bv;jSE+jSbHcw&#y*NgNs
zE{8;(rxRsOd_J)!F-6?e7un`HdO#>T{hMI)57VB~>DyEIJR}ooqvNzP-#{yU9@!{)
z(M(;avhwb5mcGo{-y^KY5c1^Xoc(=j%IsmW?Rv|q3;~YO_4MbA?p{{!t@~ms9n|f4
z%t+%{m`$2>J>940(M|*p-YY^1U7a{qApx1$O5LLlP)B;1P!4h>jF^8tj`7R^j*<5J
zpB>_Y#H!_>79e03H$0d1^@pPhm<q6}k^7K8kwy$m12(V6|NSxBO8UQ>yZzr(5z{_y
zuy?#!&Sj$RMv|(1de+X#`Dcd1b-ToWVCf<AC~sCnsk#%SP)c`#Q4*adN;Ptd{!<O>
z2sZ+SLfPhvFPVLP|KBBd%F(g^q8{JhM=#!X-#?wXQ>py@eSVHukI(-N$UTbBzkRf?
z&sSEbef?|+xn3WCGhFoVQn5oLkLKg`2G{T7?N<QXUMEU^NM9G}>qptwkMXr{|3v=2
zDt~_}e}5)_Uz5MD(=P>_0ls*!@)e?#zS~q%kUn3Q<FO>*p^Sjb83FR^s07SXTuFmT
zPJozI7rZQ9;r$>jt<ysd9kwu3@t}{_^L*PTU)RNq%d*&S+0`|VEZcuTQDhI45#`~f
zTSj%@U@Ok*Ao-HGF$zpKbmU{R;RoI}N*YuhWG%uQw3A#O5|erkYoi%MczjsQwBmrO
zERC$Da!U88+Db8=#%q+3&S!O@TY<c|woz*IQF@07vm30advWB)QL@2jQMR+qP%kMT
zo6u<E&VqAw$45cW@xb#Upo)wAZX$+PQ7oiTpzOpIOqfiYap0geO?Pgp)1*<{yI&9s
zTi5^(?5xmzO-xgzsvCEV-7PQZpyp^MM4_aQ(!mbZzlPZAr0QV~lRd_c6>&p8`GO2U
zN*APcjT>SrVDe4LVOJR#c}z@l6Yo!HvnJ-*t_A+#LKcU4Z>`NX<8ACWLlsX{Ifi|i
z*+I_)1kqwd4B=<F;FiH$WK`%>Bqri$Yx&+~nW;Ec9yC}Gc!9x9x9u=geAWq7Br(LS
z;(1}0n%o6Np<N?gTiDlNZem95o8zEYpDGV@Ef#Y`1GFyWV38w>X{H_1-Hy#|298sP
z=ol)B#SH;#c|jY+6jVhAFjTp^A?h%-EhJ>X5v#3L0ST8GD)TY&E^!RGA3XcU9iOhg
zba<rYGqS@%#Z2I$8UsDEvPY^pNk8xMLl(C^14TcWHHV5cFJwk>rPPuZ9u=+7qL)$Z
zWwfo3DlkDvishI#5*XYRhzWs-D(`^Sq|=No%K>pH(Osl306}RHm4HAtv|Ae@KaSGt
zta*7@%*%Y7!%)w}q5U1y$~`i7R?PU!H>iLP_SupQLlL46Z3pgT6yhB`iu`bGbrs#=
ziob_@zPsWD+pwh*G*cpgnMTAq5Kkm)otVFT$N77}&zWGW)wMnK<lEn=w};u9q_S67
zuwt=jOTvIFSmja20k8>_VZn-IVd%Ix4V~;5nQ>+Wq~$34VlLcsBXif_D6ny;$R<G@
zH;TCuF(sdEjr8628j_CrGp^xyEi+~9=@!S!ehY$E_69CX7Gm}ll)LWB0(M!Ip1Og3
zXph&j+%4!p>K7KsQH1ek*EFj`y6bqRGDa#ezmh;@I-9Hw3*G|;oH3+J#R(j6%_{~&
z?W1F$)Sf&pq;yTpr^WKnRcA|ivwbnb+^bZmF2<ENf%!rcamlh^8xd-KskBk*%{-8T
zW<@B#ht}Y=oBX&~DD_gBQFddut*k+%n^g6V>0{KXzMJ-A6jDD<KT~wyouHqOQ1gTM
ziNyrXl@~KAoR_or<;;CK3rB<7sGl}KP1`zZa_ed0Hh<FZ&i1Znaa9LV)VPNTzR5K@
zH2j(NQi-uqT`7md6Jf?LNwj1Im`vTuY_v`HdizoOq0^pCr@>fE7qv{%J&i(B6$HTx
z)kDri&r=L8_K*e0UT9@!6ZPe#sW*0YX0s<s(<U@jxzj>vf5e3#N?_$Dy7x*)5SZcW
zT3$?!3{enOkPcJ=!bij%yFcxJ39@-M`4G$&iDx(f$UlUpDu9{CD|H{wx}~xmx7?N!
Yw@JCT$v;h^w9P|k1|Y<0nW0$uFRCtos{jB1

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/optimizers/index.doctree b/.doctrees/autoapi/lmflow/optim/optimizers/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..f91a1b42fa5adc85f3a218eef4842e425deb8eb1
GIT binary patch
literal 3094
zcmaJ@TWcIQ6n1RyuGe?x)<WXMrlfIbyxWI9lr)r<LSNi0rL=tugJwp%64s2Sy4dSN
zpnV9$pr=OtZ~Zy_G@2P-f@7HV=*-d4xqRQz`K$Nu+WJEM=eBgkcsU=DJTGNVwRbDI
z%xpnR<KO#dfBQH7zzr2S)zYerdba?LjB$}GTKd=CZ3cD2g&sBg2byMv%d*?fdaTa|
zul)fVx}M=i&`Y1(C!&}Nc{-An;YIxAe^RA=*n%C)ztVYd<z~gVK6=@&2z)Z0?YLpm
z1;LhFzwv$ImIW`VpV*YRRbq_dsWpf>#JI$B9pV{ef@lpP-g_k};@e7YtBXZSvSUT3
zLC3m*A+zwwvn}?3t+R)0iS4@I#!MD;tZZ3QH9nQ<c>L3=9}#H#4-V>abZktlk)-0|
zn2K?oiuhZnV$92&&PPfkwns;cHf0O0SIaIXGUk{)Ml5U}nfnaS6Fg7xJPUClHHz2K
z9Ju9*aOK_OLgrS`Z}x7gfqN7+4sTLm{fY6!vM<=-Meq~$2q{MGhNuXcDN?aZj=HVy
zh1jcOtfL#nO5gub-V_#wzl&LohRKaV20y-KyX-mAw9@)hdosvmaRt+?=N|0u?^Q}>
z6y?tM(zCtN7S$Q*vbVqA`t(tq=pWAcS*JHU<%aV@ywDX1>{+Q4)l{8OAAFs(J~lqD
z*_3v{9k7=z<`3&^#otH{+Sa#K`H(-PEqyp?6(=plB-Yshh(6+ROBnXt-8G7oT9Cl#
z1lgLw|45jgk%kC=wBS}-9hzfl!+E#dW@c5OW73gty_G}~x)n{0;pGfHXQyi-Qzc77
z%iO#E??hPOwd+ECX-#E~pRlco8_p<*MT}G)i=2SQh7=VRYf4H3SLu|`y}LE`ZY{cy
z;M`*<tTs!l1XdeHW=SA)_zG8gfhT9)-JZfsDGj~!@txZU%?&t7=2NK(|ITe{u&PQL
zcxf1X5G7R5063kTa>J6gso=b3L(^p@Ze7tzDw7C$LXplzx5|j<mOZya=Nwry#8}-b
z^;|Efsh#1t%~PUEWDy#Qk^+(;%8}6!Qt*$L$XP<mOy<z?ZfMQ)@asd(@&54Y$Al^+
zRl-P_3ySv853Xj$yRDe{OICWd74LSheYs|W{o^*H(F+IKtq-q$xVF=6hX4RP&5$2a
zlV?Maa%RblBHcO2w1^wX-BC1?Do=7+0X8@$^tg%ZXR@3kMx=x1EgD$gZH8W(T!`Ev
z2i7)nP(SJLh^*EE4k)UUh+uH5mefcOrN0nWn%Z0@AUU+Q>AaGfCYN{C5Xnq(!ZnY0
z-AZz<Nh(4hjhBO#C*a&Vg3(|`O*$CeXw;G^7j*-7_rn_wK`W_=Ny}g@N4?i@RijRE
zd34L|1hN4(<qRkkyx_*a+J5RbD_SL>2n9va0CC8Akfa2@2VS^OnK6}qF&;y#N7b2O
zvK&b@1D);#&DsHICWaO$kEwebtqz}j`4#fHsixUA^FeZ9{`1y6m`eg)kEt4QYNquW
z=mukE3W39hhNPfSZkRt>ak~Iyk#x-L0!U(eJ$KvstTbevFg{}f55T)DWWfEpE(lom
zhXtiI5_P9u$7Ld9Mw-;_W(=2BhQdd(A<K0NvHRFo{=6P|>M5q_0p~!E%y>Bk2AZaD
zJ9G(c(`#zH2-p>pL?(t|<k$l;MM`tf_Y58|N)67qmqF_|p#|Y0!DtqmZte9>Gh6|y
zVj3ITxm%*;Nf>I}N^B+IE|7=;frA?+F+|O%sRx=JI0a-G3~XrUc^F~dpw(TG#xVzC
zSetf>e(0x|kRTrfVSHzlrP>XHz?@a^@Qr8mT(kW!g`^mUUSbO9M+dq_-p+&FO~+o6
z_Bw0tEX+3=qlbSNE|dS7>v6fgE^TVv#$^=Ler;qNSPFr`*C+{OZ}nmVE=!^wW+R}c
zNTuk2Rjb~;ChW9_S1b<OQ|nE)uh>V^R>p&c_paM&zToaeE6St|jRqH0CWS!B5Cd+a
z!4Hl6J<|Ht{{mWJo&rHJlWHI-9@urHE8?|IT!Ueamzl760N*SOwySw)noWNRV}VUa
F{{!+v#UlU!

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/radam/index.doctree b/.doctrees/autoapi/lmflow/optim/radam/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..b23d4300ed3426f8fc68b4b9d98be2f2c746dab0
GIT binary patch
literal 18078
zcmcg!Yiu3IbrwZlzC@9F(oSf`E4z-aD)Ew1q(q`6C9<3Vu4Kv(T&GCvu9o*Mxx3oC
zd$SKd9JomV+Yy_o<3wu|ZV@zn0BPZ*K!N^f8q{);q6Lcn=o>*0AU_HeZP5q%0+Ip+
z`kk5C-P!wCUebyM=<@FD%$&!ZbLPxBbMEIWZ`}RF9{!I_g)PhWHflzr;rR_SOxS3{
zug6W(i;}y^iJwniN~&xuFxEpq4(evY_Mk`IvfV~tddUk3o22fM?S?fuek3&Ok?nig
zaI0bsTO%(dRcnk@B0F+T<rO>RHdkDKz2>(fyIBj2hS5x@H#!eJYi<x0UHiO}2;7ES
z1~7>d>mVD`Gt{j?HY_n-VneR&naNVTY_M@7iUNB%jzB3<HfT3OqGF`(8exb=e6M&$
zQ+yk3#I5{c*{I(R%oXYrvysSHrB{hHWzAR<)<f2yHOrovxBR9#AH<$#2J`EFaC`px
z)1Lr&=^r2%&I`omjX3g+mOU>RFwYq<&lxapdku4=7K9+{1K0M(%hn!N;cW`%1XHbt
zL56h<{CX7sAH)CS_<w>(0#^c?6H;YEEyE5H_Wq{dh+Xq>tqarH`^3w+F7K@2r6@6C
z>!dZ8Q@dn+06ew!2_eZ>`9RvhMy^V(@&%pFpXK6Vt}B9bVlJa6H3XU%eOYw3PJu6@
zoG%9=V>xz>K(G6*9=6Q-4p=nA`?CYj|IBAkzx3HNFQtDI0(T<^$Bg^Eu8XJ7E?y)7
z2;*kc2sSTcp}|ZxE!7c%UsGNu?4a_bgXE$TeP|s2ZeD@zjE_InY`G@sZ>aGWjSH6j
zqCrZj8Sbhd*pbz&rCL^LpG^wZ7#ogO13zwcFq9N|1nk}+w?{E#-EKrya&4R#?190z
zv4(ysFWX79GQp?R3COH0`;E;GM1kjE;IDVIb4x_((RIieEQ+xj7%fW~$WAV{HW&S6
zXYrQc>cvhqsqr_LZ#~ahTZPh2v`nvId#m~e42L?~Pxe1sx=tgm9hrgQD$`hBm5Np2
zkds`keyz7!!5DCC*xuG=!iHO$bD%0QRog+#{Z*{aufSyXR7o}rnqbcorfp}*66_$<
zw`tg3Vts>+;3?W{VK*5h*lZu3&5f38*X<~|!(Xo(ftPHtF)&@2rS7``8$}0hyOSjj
zBr<|kGfLQly_SdT_M5QP?}<)d8vq;MfRImH!};pGp{|bElf$$Ocdc3LJ1}-M&+9a-
zY)NRqYv}kz0Q^#$R>{5Gs<7(LEX4E<2IM6Eol0VFAxcSnUXu9NJ0P)~aa-WStJ^W_
zWpsiG>OB)1B0HC04+d*(^j>M7k{wknv*!BVs<5X<yFxWE`d0gBwG$iLXu6AG%OE#v
zv=x}48LXLHUVj#2)2$Dz!cLK+ock_D>I0}v(j6fD$1#iaRI&}UWcQwWnmY^HRo{zf
zI}wIxnsu~itL{87uxC&I3y%%mIg*XwC>?<0*4iYuKM!X(gq_8%$Q~8;xyr4AW_w+H
zofBpDq2TK%_(^wX4;zA;-05YIxu2vrdT7HljN+i+<{BKUDs3&z@Ra5TiyGWsBhoHu
z3$=xX)2Fm$Gcv+U)rH!HQ)s$y;lhbin%N5Tey7d3h4ZJhb<<w8BE4bOjm`Y!!ij9t
znc`R&p65q=&sR1KcIML;AC2H#ENaI&e~w>1rCm(1hWiBBThy{H+VRVm;Z1VieVSE-
zoy}MKO15?xZ!!*uxm3^tS}H*JmQQl6ZIYs`U<2z(yCykL(U2L3z^jUoN`LABJLBl!
z?4w@RN#~<t(Bt%nJL_ED4nf|qO$(0U2t*!zdhCIzJltg$NvY`iE~jfb1Jup8t+IU(
zlQ}3k2g&vm2AseAxv(1->9|!w#~7!hoe1e1^l%lNXd6UF2~mOND9yhef<Xy_hd6><
z2~0RMu8k9p!tPJ~_|zlQ4k4i#3MzOyN<uru0`C!PLfWjibF|QL8a|X5moNTTN_ol#
zyiTEhU9gSePuV6>6oy|AgI?u>oG;=}YCGA%4ih6-;Cxx_wbu<&d8V&91#i*JTu-`*
zuk%&B?$Ap~!q@SRtJM*{+Qh(C?A4>V!x{}NroiznhyCqR*leKH$-QsUkQqmSc76&^
z86$ewoj)rD;yj1f9Ry0lBcT)+S>e`Pkn_taY<Zx8=@f5s62D&xz33I>uCs92bMIMN
z<bvZXL2;j&4rum@T`xM));-m0Cdc4Su<pCSs9o^-SmwTX{lijf``|Tf^ubJ+{1%Qu
zksqk4J8&tJq)Un<jqWXzVm9D+O42Y5*^wkYPzt0U>TuMyJszY|jA4cQ1u0qbDVLz*
zrBKOutr)Qz>1&1?n?*9&;V^Es{2<ap#2%v0#6E??wrH1pOr<{eG0V5s5-W75)2rbW
zRf}e*!n<X(il>~`E2XrSO6esGjhCGf{7JQv^$?aD<xmdeb%#Ze(juXJP=zv~x^gLL
z#m1b%7s_FuX@|Z25(<>=#}alrq1~2n5ZzKL`S+Z-wQ}Oh_L2ZYv3!331ClRa_g&-V
zawva-DLR%<ZW|IxKikGv%Au5P8xqR9**1Q&6nd|1!}({>o(El=f1yW;$fUuN-R>D*
z<EvBj*h{r>2aJBZl+;qIDxvhF^M76rrA+5{K+E^aq4Xg%1<O+5C&5(N<WsJ0|5y&~
z3|<!$VS8I5NWBlX<bfWC*KP)7LhYLpTWri}9vCd!=u2ht|G<wvP0Um|lrj?|q4Z;7
zX3L?JNsEN?ZcNO4DfIW;#GC{=J90~R!04w+Ni8)o5=uWN=9A@6%1q1-Xfey7^dU3_
z6Em=oY8<}hY8IA5*@bBlB<{+z^fL%aO*p}}PR_NS9s2Qzzyo)9U&dYtcfT_%1gj`A
z<CKb9v(5E7(&Gu6O26ucSGNj>knPe}>W(|3i>yZ2CHspEH1=k@kvMC#QKxAZ>-Wh)
zf1Or9#aA59b6z-SFo;Kjk=@4GJ7MD#z(Jk`Y58KBhYicrigl0s1lBMcU5Q<nhdTuD
zaXs?u#5fB7-s5erBAdfo^KOLL1!Cl^2&w_W57Q%0S7h<Ex`-ny$2P&Ss+H#^7M?JW
zg<sUN^cc>0bxU3jkcmt)sNJB;0``k$a7*kQ7IMpSq#~`q+a|46b}(lrb_T%uTIzi|
z-=<0Fe!?_+cpKytRQ|9Hm7>p7wcgr38|QorbF`bnf<4Y9Ma{Q4e~NLr_t!1uwxy)p
zthcRu);oNqtv}kHL=ip^>wX05PCPFQ%2I#IX;A{7`u~n(3<5^E6ucK1D(43hvNhW+
z!lg1+#reS^&SMPG_7&T+DY-7pN`8cw24RdL5LW1;Ba_=y^ng9Sgd_Ok_>(3A!gV7u
z8>ciR@u3P@jQ?w95K`{G75Hm*!)$2Fo7yv*&yr^7wQ;EdxGsZ;YGKw2pEWh)@+mjw
z|GN_gI6tJ`V%^>k9{e{Rv1VgzJf}^eNRpl}>Dk@TmMQni9gJ4AZvT)lLg^9%-F;vp
zE>AQOi&S$sY&2zP%C5XYa11m*fayXdBTl!6&LFhXe)C3}Z+d;2(s#~L!ehqiKxdi_
z)Jo9Jje0u>RU&6F^Ed#MnFc~)l#mb$NJGQ|mV$8pv$*E}&eyzXy4v)_`Pbr*uIe2<
z6={|cBI9Lpp>Fvhgc_N;TQ@u{vP|tcBe1c(nwr1DpK-AT27k@euA8_PfIQ#$O(=<G
zcn$3dbJO=4w`${76d_qQKW_vZ_FBykR_Bf7aQ^g#h4ZzA51sk&xlEQG>;+v}i6@Cy
zO7t6oGsSPN>2y;8NgJeSb3JE^yGJ|~xAVItUBP;(4eRoix=y`k93mBpJk9pPl;p;t
z3kS%5>7-rji}9F$`;IyE<-RH5?ok}riPIoLAm_JKx=Q6c+lXethg?#fe}fFC`@Jw^
z&VMOQ?7FblH&b?K<Qgt!^sVB0RlHoQ`_1|5_GWB&*W<boUbo|Uo&%<<TFO5mm(Gt=
z0NWVupX-}D#1F-A9|N5)5}THtG5jggsO-NV*(GV2jvZ*9K#{C3wPEiV3M%Y)Do)#Q
zG1Z&C`{%;-OKk}BR+C(~-ozM1xQK>+y_`qDbjJJl2GnMsX^2>-Qu$TwGcE0UA5awg
zOha<1hl#_TzGmU#FEZdW$BvQTQ8#frmOCC^+-z;)+Lw0hSh@zKeX`y29bl*Kp6z>D
zF|+wLdMo(Jjq5Bd)vm6GI16-?TF~{-j6z(CFm=6pLaTO3?~B_bO*69m2JVSG)~icR
zohvC`8cDgD`W?<WKrd=r)mc>9*kN_HD#uV>=x{={--OyP<$tj;7xcd;9xA(7QU0pV
z=P<MLd3=;yH;~!<A7oQ;&;1-M$SW#d6^=4TsF;RUUlhYq`3P(dvRjV(k1~MWyz|n4
z-v3}|nKM)rG?~2mkG=;vufbHSv4bwW`Lbfor7PU?7SoQy0_+f%(V_r$4-@35b%Y7N
z2ZBYI!1;Ax>-+{jigNIEw8+Q+KMp#-Mg1PG;N!RXD<*H2u%ZBdA#vWqdv>G~&FTs6
z4=b^Pb5Qc{%e973eG>zVic)qv()kAAkx!87ThVG(<=I;3bbAOy-b!^o0=n1;!X*4+
zs&k$m`F$7l+4JYbHGG_yJwN?AnKsXfJH=aULcUC&52A+*e}q}5DbG*a(tb1(!Go2G
z;O2@D)mbPSS7T|e<$3kp7(!SWv(=vgob71<X^wYo_zrPQ>MkzA2j-<E-!r?~y&(Zi
z#9hQgZYw^GwYpDs?vvS6+|HeEw{sIJrd`Zk=lixyHwsNSsj2PGo^x_Oli{#q)7QjM
zpxjSb`g7$_%EHnTN+Do8L`#<vB_48$ez6?ZF2cqFg`%FTN71e}-sapX?t<M1b=deW
za`EG}@uHfe#>Kw7ZO>8au<bWN=x%KL?_pfKZ5PUx+Vlxw)49@i*z)r*#m~y6?rO``
zNbgTPV+JdJ&<r(06IXYOg(?I>)~HOg{yOIZ_Z^8vx*_ylb9mm_G4OHV|4xqocE`XM
zt^}*RP%c~?ms-83UD4!M1TJv9H!w}QfCGm!<dF;n+)bAUE&>x*ggwy;lMvRFGa-HD
zx>PNm3&(Sasj8_>IsM<=3H|B`@e)|@HyDiND?K4zy$=MZB;1#|a@zlCC$wj4e;I`T
z11G#xR@5jO4H3NRZHMS!r$o_+y?@yU6y=mgLlmLB2VW_5knZE*Y;1TJhqDwdc{sbV
z&51bm)>0FZDrb><W!{=|5xjFW;ObnWZ^!7{<?P!>@omfb82`M&KcC>ASNZ3Y{PW}V
zNvqPst12R+?hZ6a`kQHTtCRwHIkV_fa{_7^0rMFF{M&8`;6)VDEzAjEQzE^Wd7gQw
zI(>!6m8|VB(p&Rn5ZenyP?eESTeU8x1-WxPBR40CcFpuapS~@W^mfcYC_FaqW}(8h
zx7KBayw)jpBnk{KbooVM9bt)86v>D}-b#dO2RFI4hmEIw%myA~qGy0j*5iQ6g!QZ_
zcp_~P#ldV;`rN31K1gN6yrg+0;QVQzPVPCnM;T^OUpDN=Q7ev;C2I<mXR9U(e)*+T
zJT%DMM@F-SnzVg^??*rt380OH4a}q1O`t$2>D%ZqE+6B-#nXODw^E@oy+z9PHFjtP
z6L?4uDb07-ek!s?C`R9)Jrp%_lOc+ITPT9%5u6ZH&4?CL7K7^R2utd`!~kRfk{lo_
zbeGsfVDdYmuDND{k$c&=W#Im(e5|m8=7tUaA$Sspd26L%F2}2wZ;~qZsMHbjnyBFh
zAFvR0lZYYwY!BSh5jpp%AQCy`I9i!IKgToF4;5ePW)S#+ZXv$!ny6YHCRCBUW>bRa
z$}AN=OTuQ`-n32OxKK1UDdx>_NJl?Ze9<<T9UvMYb)f|oxx74UziW7_v9W4`<8&=x
zFN)#WVI=STpn=jHD*FT&s=i)g!%(snB;vslyOD|la_y5;IHu=KY%g;DxVVb|8-=Op
z07%0pd4ml-4Lk$2XiG>rQWq}f+=I%RTA>*?d>v(EXgYw3N+ecvbtIRPIxYp<p-nfa
zF_+%3Ln<c*Ari}xR7s$46Cfr8CgSQoq(*krS8Nx=q0D@mz5xh|)p!mR1udAZfhFIE
zUk$)giN|XLs7Rno7_Ny5ISbO?M6Ku32adDJmf6bFDnuWuW?PVU;7&@x9!x{RXbx+w
zO+@BAp5RH?QR>MNKu;Y>RV+`$tLMzt++z!8!OzKHrQS9@wdB)(Fr6N1Ymms^#svf0
zj8-@dn1XpLYPwL|F%<Sei+BO^D4T^$HuX$7q2MAnA`73k*nx1<i;NB3Lb;YpC0hx8
zhydk*37g>8u623u`VMZI<kzru*RLCrwX<~#g(Y<?ywEprnX?eRZ=)=Gi~sh5NJ|Mb
zum$Py>$bN78IX425E+UvUT*7VH%WF;fF_iYej(v@0+DGhn+<62CNQu}6L)0TQFwaB
zZ5<g?+tu+a5oB?(@^{!l`8yAgRd+>Lvn@7gdTaDc3v5(a6R0mF5rGmLx)GspovL%W
z+#Cc_kgN!X7MXJ-k;#s;LxonlVZxg++e-E@0ui9xH>9tSJ8<4daU>BU7?J5*CGYS}
zn1J3<0(GZnL;PwB%O>I1BgA!=w=fHN0y4`)DP3${AwQzgZ%KfL5$Y+vSDla14u6S0
z4pOvrl0MGU<2jm$jZt47@u8oi;Jvqa&n?~yp2{(5rwLHHc@aVSVR>mcPd~N?7%z6o
zNZTQNf#eQh{>L<z$O8|h1)*^F2{Y?5M~fFd@_R)5UX(+jLWe?$4jn8VI(RzAiPvM;
zb~bTPYwB315E7MbkLj;L<VhuoC2r09w0GO{*OBSv!k`fc7PUv@qCS!)#Nqb-Eif(1
z7|_;!7ybT<SgCjr9JvlAILCmp!76e?8^nDHUV>iv_|2O)ZqhCze&GaGPyArWwHad9
zU+~HKV15B0X8Tp8=PSM+b?d?>ft-pKvh8U<A@6Dn86@5cBhg&O4_6er9%*aLdXE*l
zjipfeOos~dW19}51V;C0`&{x3{3Pr=bw8$k1x^XdoLy)rbP*2F$W_^X7gFiG2o7A^
z8@`f&Nt3t=2D>bmTL8{BBDI+>ioy0}FS#YyXM1%wZjj|!rOgAEN3dhc&4rW`dDH&~
D9Kt@w

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/sgd_schedule_free/index.doctree b/.doctrees/autoapi/lmflow/optim/sgd_schedule_free/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..e1af0d62f257fd638765257ae3a09f30f6f9f360
GIT binary patch
literal 17307
zcmds9Ym6nwRi5|b&g{(2W_O&hUb|PpJ9m|tJN9COJ>DI&cH)R^Jhqa>Lg1vg=ia)v
zyQgpWO+R*KuuNi+-RzK3aM(5ypa}enAIXm>5{MWHgCd~__y<aU@dt?@5hNZ0gn$6w
zsj7b6N8g$8#0XZ}neOVUI?p<F>YO_Bxyd)*cy2`e<BLJZwA{U#(P+3{g9S04X?XRh
z&D=1)A76ewel4!@S>M<VyvVP!n2(@G-L#yB&)oP^F<+qWq2&ZMHGVo^_0aO%bhtTb
zPMOo6imT=<pA4<gVYyHIA*bDRyxp4D39WW5Xf^bpZn8$?FumzB7Sl)!oVdyM2&19>
z@5B=4sg3~*!`NKnvwBLBInJjPC0l&Lv0N5!McW3i7-8sJ+ffJ(5XIwGBOvOg>y8lw
zXvFiRYqaIlOe5-KFWW|a$7fCI6Y=TLXwj$GTr|&{^X7x*xOsuUvTk~9w(du+%l!3S
z&)-@9_zTa2?Bo{^4Av!L>qZoMM#oy0j9C|qS?^)Yy5%<5Ud<0c;zKvaqHS}8Pl~o2
zcO~P^he48g5q$h0{y&2MEBJqz$ODIbOAuA%6CK0yV}7aa5%-_gh6si~B0H`P>1R%D
zg|QKtA2Qc6j%=9^f$!#=RIp4#PA3g~`Z>jUp#aO-yAm+wx-PjZ=Q0LTNuWs>o3gvP
z3VzK9ew_}D?Z`2Fz3w@B&|&o>ux>*1=cix!nO7fs?bR!<CBI_=_hJT)w||(j-+1it
zjjN<ELDX&={{AME0gUC#YFT5^)N-F=zLa~@L4I*DUaH{VCqMe(PpT#OJXRnY)Y}@p
zYWSQi%E|5tl)laKsj%gHQKyHWWCNzb?ISXK219nOMrg)2D#TwG{M~^Fy+mDxDSB>#
zpUqD|7HQjS?DrrFJjZ=+x2IhgF<Lc!hs?t07%kuEn7M)c#ztp<!`rquZgsI%yIQ_>
zYF+&|Z@nT|U&UIV?=ZJvxh;JUwkHqmV<!hKk@K`;h0Hgc92oN}YF#dN;TUJ@^R2<_
z6V^uH&Br?XF`w$}uYo@?%Nr*W({Dn~zXAI^aBcG`P!Ho7^VOr6%@|u6jcXg08=GI_
z(|8N_JJ{aF2@gJpcedAYthyD(_r&L2!*}BYJ_}JwyEl9*;xp(hHjsEr01OSk#lo1^
zjwW_tQhp8F$a`)|Zj3?zZ(^Y>n^T!w-^|OE4U`QD@V<G${1)sj&GQBgD-$LS_!2sP
z5deR=Yh^1}#GK?y1I*EKzb_fm-^n9ASfLW6Ur|W^`4NyVXT<?1yE}~h&!ZCz+u)h_
z1X<4*8$Ot2p?9-;N`5Y9;x)%}ThdA!-HWS%(YLxs=R5J)y|%LvbPV$QW;#9#n17oI
zZTvLGCi^K^g)JvTIdhtf#No&{Dc^(euV5DCBBi@>iA%8fg78eV=R7xr2@a%rTIN0N
zsnqQUMn^^tFL-zf_StlVK<PAevDT%FhjXW<0$68kul%Vl3#nCwP0%cNh>Lak=$4}3
zY~^TC{?-VefREqn=Fo<+KwtF0g8Ldqe!)k7ql<znZFB8lRdf7nkJTPs)!OiG-7soj
zd$jh{s<z9lmKo{|RyX#q;VVpMJL>2`$T|VN`{+i^vYX@U9dDQU*RHH;P0wdW-MnTR
z0XgK=(DUo&s<xu54|+Rt^p(q((_L^;AZ@s=7aCoM4u0k5u6{6t1GJ&72!^k0u4-2k
zBH`{qtv9r^i?*`439ne3Q<nLpG~1aXT~E~_Wo61RIah9T<fB@mn|R75xzROA(G2m?
z-MD*^+CN1@&f5e&zs8b<m3Z=Y1s&`;>Scb&ent*@ntq7`h*0G-P-XZHl4CgT(J3W4
z;=!&yJY-W!C>q)?qjx(+pr4Ok<#W)sHP{t_3O_}d3BHT>f-_L0?^X$Yvx2^E;(EN4
zq27xf4ncl~ke8Ux5X@mn`z1)95lD+w#o%c<7EX0J(R}uQdA+5}O{#o~XC6m7Z^I@?
zEg6x*Bx|B~i=xnR8NQz!mubzflro-=y1lF94ar-MUu7FwQHy>_4!SD_*<Zx3#Ln|6
zY1ky)_LuXY#vDIc3G|qu{Z*PythRp2Zhr-zdser?`?v8;EY2mN2gEh61BV;tFVEu9
zYYAjLWq^Ob6fht4dinKD8gkxNaB*C-ToDV&>D)LBdjd{8X8`}K93a+l(}<i<zil`X
zD+1^53!+ZP^FuwraW(u*<Wa=sfQ;6k<Z;TpiE>L<BE$LorqoZ8Gf9Vp%&Z}!^?RkX
zmP+-nFf`h>-^Q<0u?C3npUa{AF+TU`j*_8*@~3$y<Jg81qQrAX&)8V0mVUPz(r|%E
z1WrI;3q`-7pUCF(T@nh5jTx!u%SkO;kPl$!Aw_q&97>t!6qJJKM$<)yr;L`LD2H+k
zu}R=3BDR8i*I#V8E}7W!jTxygm6KW~wvQiPY@aTNQYJP9<uGD%%Ap)XY!dhhh;34M
z;r*mFnr?7-%H(pVoR%`VyncAO+%JbxCKm;zAQxD_bU}&djGnj3Ass_D5`hzt%~H?N
zsGpn`yLu%HtEWAq`R#I=%VhYwhnL~E%b}FXP(dlkaP&$-8J;p)zEck681j?APe6Vv
zhdjsjlj^h@o6wxTXO#baIpt+i{Qlvk_=9pNWl~g73Q~M{$eBK&cWA$i4|C%u?aN5S
zkEH>J8N@$;_(f>x<ZJ8cDGIMR?g=0G^W+Dfw|funaKmJtmJ&L*Ew)=nvNq<6$)j$#
zb+d3<7?u&C0C*CQUeRSpI~bC4L=F_AtVV3#2BK-~-TXcB->W2zJnr%sl<~#BBAiR)
zAdz3i87JlyIzl7hj)ZYB_RptOjNN=E3B{OGe5M&WjyO(A_!T|$>cmG11->KN?jo`;
zTC;A1+f&5X3o;4<9Dkf%MKC%Ih}30FF?ZaO46K@2<oVLG22#NrTAE-H;VG55ev$Im
z*3UElmR!{)QkJTJ&S-tCOJ&VpPmcx<A7)zfx03$$O`3|5<IDV9Uqo_XU+w~13_MlM
zuMW?{*q^{0WrypWI^?q}_AQJnPp};15-jE)nD3Z>C}e5=(ICPgQWN<QA<wgrCvmkb
zI!65|6G53@>i<#24xG6&>Au~QSEyVMw9sT4>5>*itriPH1U+3(3sgqb*Q`Ku8EdeH
zwrhnZx}!m4>`4`Bm7AW1n-wYoipZyd`F!*;TrHybux4luGkljG@eyFu%-X&eX?1k7
zqadsVEEJgJBdJ<yKx&EY{J%^>X4dzVbj%C=0x9_fMP9D&A0*dr-7LsN^Vw%23%LxE
zLa5b3A34D4WzBFKT8$xOf#+=o);kV_y3J@iCf--JS<7;TpuDCgMhW?07>$ZV_!*N^
zW;?WQGiSf1-GuU_%Af_N7defJ;RGH*=r{;KHz*50;}Jq{Fm2alt`-=#X=W%weW&AD
zNOYtE{(nJBCh5D!lJte5q{W4TcHzPjqS@y!Uer3iS7*4r12OJvZq)AVBU7whyqGMP
z(moPPfcjM+8I+`;i$TShNXjgXGCGePH;ZIHN*nE_(#z_Q5PesgMr*Us^cqMb{n*h4
ziggedb4iKKOjf1+1TafBBehrB8@V<<C9#&<p&jHuP0*)A4Duz@if@)`Mb-W+X0ktr
zA0-(A6>R^VbSh!{uOOy#Cy&QiRi=^#=CF&xRqk1zSuX=nkr@N0EZhV*vggImb_rR^
zD!2KPzoUGy4ACB6?CU^Sq-U%41|IFF@uSGkXV3z#>L=+dxqJ3?diqQH@vQj7dy*<#
zit-Dw{bBUr%l%LV#|gunM|K74<r_4k@W|}v@pwp9(nKGjk=X=^dS-4X1(2)CzmW@~
z+t*b0K<s}Zq^IGJimM>@ztXEn?d0EGZ%y8^gV*B1I^z-^u7xZl&1>>*&;g%UACvbb
z^iWP_kak*hy`(MaM>7#TZ22L!0eLQX9Q+r#1!Y;(&Un1cw6rPDBpctcjmqOkG%5n%
zyS6IR3h#bqWm18VUa@-149f>56c_tC7DqBH=hCT!VR_<Y3`?I`-#f!{u6wmgX7;x%
z8Kym!<uq7x9LsVI;|5z6CCtRMJkS@}Ufc2nOv4!^E6C`0wk4V1=(ffFci0g7r-`!v
zpr@DV=|9t_AK>Z0{x9+Kzs1iHbU3g_#m_PEbDVyXq5C1&fK9rR0e|<{Cg7-k1;KmQ
z7T}D)y6)i4tIw`&c`j4;LArbl2~6683GtGd0uNjroUA>PDcDJ;5~kqEZc{Ly$5dFY
zVy?OOd$pKx#vOD)YT_@Y?-@CFQXKjnnM$*iuVW}sK1`PKemRt~h@gT}hzL%Q=n|sj
zOGeRK<*<$s-IXX5{et{qvgqom#PVx`JH>EG|3UU8TEMhWm6zB@FWx(2-s2?-{qJwf
zGemoAc@yLx$CiH^;|AMuxfl~u-s2^T<=x{YJ_$>1Dy8Ua$@7uAAAO1WIJmb1&CujE
zyyP4%kLd&=JyajAbEcZ#Rahk3!C>uxHl%T<r-A=pr1;}{Ij*iXE!U!3Ycc|yUA|DU
z+gX0LtB;O**zH`5lHb1Ww?wf{uu)NF0@n)>yP%u>tAc3HHJr;E+I21YF5(`z$uFZ8
zA`rqv_`#`Y_(y>dpWzCiE2Cd9K0%G@0&@SP^F?JjxUEz`sFER+EbV+##>Ia+3K#8v
zpx!cocMXjEJG|mZJ<BT@gK}(<{IkD@Ud5Ay{ry}MZzUXg7aTi0x9oq+!+!SN!%p~n
z5>sVznV&4%8S}P}Y+X(+n~=*424iuK;#ZOJxj5^&zAQ_K8Spj3@4`{=OIP1BK<-Bb
za;2-)9--N#p3x9V&)_|SSn#(Hs~SQqm^=iOzb+0LJH(J2VgyAlfI}ljYvykNv@Cv4
zwB!(@A(_3h-^A{gi1Y-5YcGQIBL~+Kv=rB08WQ?XLIla~k=X4b_sT*9(v5EdSLw#L
z=t;Qo;^`#Z_#!gh%S;71lDt(Z1?uW@F_<IHAz?~DloBAGj!VE<3QMTNbVdMQl*zj^
z+<_~As_-!Rq$N=gd|HvUh_qLB*8|}4eKu{?yqXk$u8E-f8fqffl6=XUC>CWYpkZaD
zJ#fe6aX|R+sHPljCei{@bCsXP8QBdS;Z*1dH@6V~mj$GgA#S8Q@r@B)N&2t`UL&G&
zlrPjHpUPA9v>0_hX^};#d`1O;P?5Vt<*TB^w&{yBWdkKsFVp49Al>x%loy4aD2%tv
zMHCyg7z$0r{ZYI$kXQ;rquoIn(wy&kAy9?yu^02vb<{dZ6euaZgANt-8u<?1PEm%H
z3S8+cvaDa?XPTG**X>!Z`5r$-MZXBK>wDyOp=4?yKw)VIg_R;QA7HBUvV}Q}V)cPe
z`7&Kx_XZh&qz0&}&n-UhGjUPWVMwN+;DuLA1D8tGE9!vQo(2Bl-hUKit&<J59knpu
z0u^i_fQBBZDQKf*7J{fVB8Kp@TyRTgE)wY!X5+I_*j#&ZP2{B?EPm9P?|Z(Eq+P>d
zsF0l^RFSggi<0NLSti9SF~89DrE3cN8(4sRLC%}ukc@t?_@QesKTR}1>jD!jqAXI(
zPdSF$ii{Qm$LV6g7>cv``+QdS{00g!s8$eQsM2?fPhs&jk$wk9tVW^=NUtwY-H@I&
z@iAojagiH|6FpT%Fjh1mEya9zmNem_Xc1~IobI%#6R@b^>8K(?(`gjcc>&Y&BZZdK
zaZ}g|EV}HBx%7q=kn;*c6qeJ;BEf>22QdNSqO#ZnTBEw@P0ImsC~01%CjddMl5nk2
zT!7gc5cwRc*dZ$UdSjF?iE9`RLxGeD?Qf&h?}gJVe4)cSI@SMSeIk7XoKR}nfwlv8
zS`_^KW*ByYjrDaj2er;Vax8Am^IKR;=cuPZ06lg1>JU!|am=87?U6?x2R|45X1!~A
zYRRU*luVChYmmzBAdzgbuqj}`6s((J+X1jy)OBGMiK^HcegQh!MxBgWoGAgSIOTw!
z4))#9*walE=QvcL6Qckg)nPH87k9CBbs75}?wVwGvvtSAaYC~8Lfyn&{W=6M*BiJj
zScu*`sP#n@9N38he`N*^pgmsQa+}ZrWf#tnp$MYwu5MPFbl33=xfrQ3e<y~@w6_`l
zAV9be3`~a7$pd~4j<d0&Gn8&RI&$k|ae>}FzNG%{0Cd%9N^5q&$C-PZD*XA3v?f@-
z&_qP-kpYI@$qO(-Z<c@*G%JLm#eodR3bNz;Ore!70*GeJ)=Xch1eN|=0J=|8qkOK~
zcWIOTMf$Ns>5~Qe@eB0&Dt<(KmimfFKmEl4(fdI3JP^GQO1MbvGy#g_uHvvWrS9wW
z92C=nw5#b(n{2E2n*af}<M<xAcmGCnN$++dDZPWcPMDdS0xeN8CNAoTOGq}QiEK(E
z*%a2bDeP-gP{5|3LBvnx^-XZ#6e?!rb_lM>7BVYIKgsHR3#|pw3jJ(vqYl5&^=wyT
z+IzOpZ8m`_W;#=tA3KB}##ry?y5~|~utme@)xC&J8umjJSUOM+r~&rg>E~4YeMsp}
x{0S^z+`E7&fF=acSqU(;eTC+^v{|U#oaNS?s6lGpB1?s{E6lU9WHR|x`ycF^XZ!#F

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/sgdp/index.doctree b/.doctrees/autoapi/lmflow/optim/sgdp/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..18906de22f4a6c1d0a70cac297a61d6271494612
GIT binary patch
literal 28091
zcmdU2du$xXc^4(V^&myb^h1`ta%_vT`I79~E-fjRt=Lsm(?%R5MPv80yjya2wRd~n
z-4n?gZW1(h%y{BQyhh*#ft#jnnx<}(K7b}ct+qjfG);k~Df&hW1O-ww`J->7MH=+?
z&CEV;Z@71)71V&K+u51<9`nub`*!!6nfDF6^4a<h{vR9jo2J!Ro7HQzhFdfIkPX+|
zYO8KEg7D?=*!#n$!YMZ5>8rlm@~TG2cA!Spw49n}G{PrBwvVa@mgCRL_Cvl=4J@}2
zH#Y~&L38MdaLOEE1A!GdM(P&Z@6?wacXifn23CF6U#T@is*Jk*?=;qkiJ~e`hk~=g
zrVc)W(A>{PDly~cE;cCnU1a@^)iA=v){@Rf^&s%9rB(nrNv>U1%_kX#s*djaD8%o9
zhF%xHhHI^6a<QaWFM7r@)oHPzKwqI-p*dzAGWVF*o4d@z?2$Rsts8S*tI;sLxmDM@
zIQOFuKM3KXf54zWCm5U4TY;-LtvMn792b9%#GkVoHDhho^C8s@=XbW2%pGiim!;Gu
zG-}=mDa<3#)?4uZE%^T^{y#<%K`EZa<(OjqP2KWBc5U6QwH)JKtzF~n8gXx?{gyen
z7=(Juyxp8hL|rs*fI`jP!akA#9*+vx&;wGXd@v`IYdLc0>x@vF=u7V?3y~%*pBB~4
zX=r7bYh^spms*bQRjRI2@ta1q3kCJ_`fU91AA9onQ%|0FD*87ha*rhB*oAE#)P>`B
zE}W#n_gnS4?yaB3C_|TQQcgtJ{A}trtUq;CfyCMAbMq+vd+edJkB*L>t2Z5kCa$kh
zfu=hvu4e^ieKwj=1Nr9Cj7HdCu;RI`W}5<O&JIDnU21n2O;)X1V20;MNudoW^diR0
zjqGDH`%QK5<LM3vL6+RwdK;tQbC>6?wv}_xp~+C#WCZY`uXuXXOf_Uv3(fTfcgbG3
z!1Z{tZBooq`RNOfbG=Sse(q@+jhfY1sjLAa(p?=b*;O<`Lyi>~p6;ajFh4G*P(M)(
zllk+B&hrOzo^!%>HrGQo*j%51IH8e-2-50n7?zI#7(32AHVAouS|Ph}6TAxHB&cd#
zw;G}Ot855Y!Fm%8V;7NTyK!x-H65#J1>v*&_NwkR!VNY8WefBvM0z%i`y8LcMNT8o
zy%i$}*<|ss@Cp1X(D+JB-1B|V;43ifNpmn6m{-yRV{}w5jlj$1Ve?A>H0tL?YNi?u
zYVbTNJ_mv?<Ytm|)#d=ZZZjO1R8~pE_=~g{ooyr%<8dj*H@Xl*t(*<W`pjku`2Z>b
zS~~Z{`U$^6cv7g(Lha>zm+Yz(#A=S)SP?j>=jTfetuN$Trz^3MwYsz5H+71XhMS(@
z8{Q>@o8(WRZREnBDnNe1auR^(kta_VNvBTykD?bDZp4mI;f2Q@<}P1*z-<KN7zE5r
zvbOSg)a|SLcI+rW@Nmrb!MFuyX&lx$o3qYxamt_%2!(rNM>y_I@ma4~jrM@$Sm)`V
ziC0p*$&h6S>qnH^9w`ut@1r{v-$G~?w7g7|`UT=>io8YLpVl02K||202ihrZes+HT
z__S6><ktvV_0)}Nt)|!E7$8)p)M)z2^>Jfn{+?-V)v#8~pi(oc`g(G69$gqtpr=Y^
zL2I>E@pOdPx#EW5!*{rs7PKYTb->z#y5k$iVkb4m#nT%NH=x$(0G*w9&&jt02t5|G
zqg?ApPfu$nBXuCQf<Z56aTV?8>C*@_c}zda1_U}MCUhn?h*;^dYDHfu_s06P$SUzG
z>EwK_lPmzSzSS^4J?(SU<d99|)3Ym@Qc-YhkD`LTn`)W2+iw+(?xlZt7|)G(0!9qC
zE;NQmBZ~E-$01<sM7!^ageba|NVOD`DGGb1*lt+g3~+-JWP6D|*Dk-#`0uO~7Zg&A
za4GU4h-jiXw6KFs*c>J{1&_nj`)2HU3ic*AdwgmcuwTc*gHKA;zxDrd(#3O5*nKQq
zS}=Na1)2!E*&)W7z^9TQsZ7O5M028DVpJbdjuz`{v`>{6#egvUqw;52Bm0nO^bBug
zKZk!Jn8zb}9ukEL?3dEDb~;`(P3UJL!DpyvKKqKMh5d2dZkz2=giqlfpR5PD9gzZ?
ztwuFywP9S}LPQ+D653x>(q?_lcI|zhnjEqPY5VQCiWSiT&AzH6V!sQw+XR$wCaGi;
zIl#S-5a(wj+VYtSfGqANBL1C{dNzPbqImJ}kOcA)=QBaLB*nvfdPcOCt!g%=H!r6n
zc#2_Oh3fta9OVtK#4wlA^>>xhme4hje}8ODZWGUAf$OEGZr`ccl1`;8X?Ul~idmoA
zZb^gGq$^9hT1ljgba>jg8vLbL3T6j*921EpuM!J7s-#NbwXC<CpmIrfT1Hk(cEE2n
zo30mBe7x@nezN7#>$eT^y|<*LPr_lf!&acsfr2QEi!_zRq2Z`rByUoXh<rvVveH;j
zp=oQ${sIg!nw+cy`aGhh@?~_<1`HB@B$YDo^Msm;3VtM&t-z0~q}~aB?0<&r$$JF*
zJ9J4e8KM^ey<*QrK04WWvdnOnQS^pVRORfHiXK7_iP01T%}3Q#4&Zj1(MY7}!f1X`
zO{Ij<+?tVPKALTP^C{5$O*NGhd094x6CvU@z{$a)=eHs}8Hp=KBBn|bMEROpR8^Go
z6*Mi0D1W1-qCyl&rHrBflbVXkI3$&=Aj<zJsb6tK`7b^?T_8#qiaxkY6;vn@MN%n)
zC^xC8s1T(KS?1MLN*GNUL^-Xd(gUIh5w`(RCfeWq6yeD(REv;=|4uOE-D=@f81fjJ
zmc$T4O+|$vl1dp2@zqpR<{+tT1w)=zQorID@_s%#U0_HTihfNgsuDval`<Iex|)g#
zL%NXVi)tz*j3$F21N;SSG}+`;GTr`2O|1tk5u$DbmR#N8opBMM>`B*+gv}NvGJjJo
zvkGGV9!*O^%)h9qs31mCDFZRzS5r}0i=?s@h#Bu!`R*$YF?+yHS1(<<Q1op|QI!xQ
zsg!}3d(>1^5YvS$52~q@Fq#a+;BiO3X3sqMyi-l32WSx@ZUeO7m3;oIQxU2R@M@88
z@+!fRx>`ULj#y|~5=Wj?Q&HiFq*4Y)Hq=y9CLyV81xG%nq<+P5<fD9ax(HReQ1ovr
zMOEU6q*4Y)zNDt2!jUdy`MR1)38Ts2$gcU><I!Z}zmuu<4{Az1V2KcO8?fZ6!mrgu
za55a#jD$=I5=p<SmQ)2Q|AD3@L1l12<+yDXs7NYhK;@vCipoADm92ovtxD=w98_)s
zJ6#1QT`2lKrKn0!kyOfn%A%Tz3RJp~<#9EY5=K)NRBSb+9zaEixecH)S)A)qh$>Ol
z7**ne1XX@UEvgDt-ixLsQRO8y6&0#TDrHdRH8mBLfk-MDROzQ=BB8tXFU{g+qQ}oE
zY4?mNf=t&RnP2Ba(?y)pg_i$VDX9`tB$YCd^4DrADoE)<mT#%4l(3i#qzv%9vuL`>
ztHehBQ%$W0SP`Q31S@44Vc`TmZCe%kWL}9=d2oqDCH@)v1NbtDkNE8=WcXGkR@YEc
z|0T;<tzwBy$i||d6}?e4GpPle<;ADJe_mG+Y14lD8WJpH!G}_-7TTBSt7(+QZT=2@
zb)Oi_u`6<J5>?nI(1>TEVu=hAYC<+j`CiE5#>$>-7C9S~soUmd{(;sUWW&oX$Kg3l
zg7|19aI2&@N<Y2K%br2DI4@1A5o140dOR%h@1W1S=#r;I$EkQ#k;j_K#SjXcGL!u0
z`TKP&<yp|;%zGqnR86_Qh%d9Te#-D>AEEU|^y`A~8Yb4L$d;+ElXsA=Qi(}sjDhS}
zcLdCzi@xgGr>Ij}<21=`+=RIlD{sxQlEt+t^D~<#``h=Tk9?4oLFL3!()s-MIkZbY
z#TTXYZ=_QC%{R=;=I`;5Hh+I}5k*dr825u1chb2kf0OD{-aIV?qWbevF~|Z-tYEvc
z;B8&@ZXPa5O-m~;&DS-Z`1<K{foAHywqzI$t?9W<*EedE<9y*UEEGekuy)p6ZD@gM
zXz%7Z!P=th)wOets%OBiQFHWRnQA<<go2tD8?t|4cGL_4WEaoP>E4=kY1Z{t=JX|h
z?!^53U9<D|oH%}WZ0px_;v;2oZccQpu^Hp*_bRlQ14*jLa!;iYo!t%*BDZn>*42@c
zWIZQIcJAZ`(O#w6hinobHZ{q1&W@(~D$oj{a=SjXQiFm1H?Np@&kvWDgrk1TvqJj?
z5(N1ATw1OK?(^wZX*<3yYsWAEWO`W*i}J^X2TTk=`mPa%Ow3H33f45$)0XnKMdW}j
zjJ_k%nnr$cAy&f*7Ol0GW~**}uIjj#blsXegA5BSam)NCZ5azikv5#^e}MNNnUcIp
zbolF~bx2x!ooWlTxdGbvG%0flQ(-ibG9{Z-LIOnd-AWlx(eZrClo9?xP6V5wAaXE?
zL3Pq(q?%1to+A0x9LdgeD6zb+k+KrYBd*IL`;E}`WnRBCA?o^fi7hca)I<!A(zxwU
zB~55Vf1w1GpU*Z)xg5SevYuHljRgXSjvOI$uNqiZT+^1;wMMJnT*pc}?Z}Zx`N}%N
zTEX4FgR-w!qL)q4{#W8Nafwnf$6>HNEy=OxDSW)EQZ=z`#BeGUxK16@rrMW3@D)2g
zEi9>Obt5p{8rJCCTwHH9(_fffu@k8_^0)R6frZGjWwd86q{`TMrn{*%x#GpXFJ0-5
zWPyv|vYQnygQ7`lGE5FZ&#AbuRATZCxkg#TRz5QlJ=M&~T7g{CG7;0`vC;{=rts4`
z*_%;ZWR|(o1U65^-4#VbQ}%n&xBWi+QLIyu3qcAvdpgZo-x}v8yT%|Y%)FkEh47;I
zmGJr_N?w)Oo@cqLvtyHv5VETN5ZK0qyRUR3dmK|V{n%Kbivy^VHTjN<NHZC+Wu6Gx
z1Wnc46fAy1OgL+MLfi{^#S@}^Gnln+!Jq7~A47?JDCAE9_H9(Jfp+%o{1y`{kXX*%
zJ{a0FxX%u@vpIE)R#K)P6Hh?(W9U*mCfajoSW<7)(Jb+ibPyp<)_F|H$0`A8`KMpG
z;Msl#Ot2w5A@eoQ_6O;buZ&7xH{2Pqj{_c-8gBG+UsOCJmZ@*BJ@RIBy&pB?ldm6F
z8f&;wSyYdDB6_|N1bC7VFDLFK10|%AUO_&_Y{ehkS#k7r!|T=?+|<b%B)x6f7wk*3
zc11kFuSoNYtf!PG_>EknEZQhN0nzK)6Z}$4k9&f<^PXULN^OxJ$do0>`sKK*Eqj3i
zuF_t>zMn|pJxT9w;BS@8D&4?eqN(atNue9~^U{sl+(3ac-c&a*F7P!q_lliB!dIIU
z_ynZw#R>Ecw>g1~u|+OGda8uQb_Z}bVC|D~kYUulIe@5xXbP8i0QPgBZC^|BF#2^J
z{rYhH>v{azuwUSRU*vyZ;(tHF|Gv!shV+-b(VGC&_hM%ptf@D!th8=nqZ}((?;?6X
z)(NlMI^8mCPZcn_Hm$8sYsiG2*66XLvW!)oS0c2(17p~AXn$|6Q5S$t^tuN0xtJaY
z^pkl&pJ-QTgzH7O5_o?k?s&`aUch4--r2#T5$0S@^^W+4l7A)Q>wF-xBe!Yb{{(N8
z<9A&=y<{I$L?Jz>tfwASp6Z?7_3h#}UT%<GB^eik=iDlUS9<6ArS0N+Bu0e-ZRc)6
zf>cRj?XUMP$ESOjBkNy^O{^_GD40!eYJ70DeE!Ni5Ns!fx81ZSROV}9!d_dLa2qt$
zOPKHtw97p!7GYNuEnM4y;r5{6B!Y%%Ir2RR4N(WZ1`XdP>3Goa9s0$C2L9^@9W=~T
z@Zyrb^x_r2B3g)F@^yTyOS;)|d~ipar{=VVE!%3-8dm0GpPnVTNlvma4v!k5P$Bci
zYj>)j$RB3P5@;A0$%X-0ztGj&uD-Y{?yS$i<}!}()^ZXRx<;{EW)`@yWkr;5-b}fg
zLopjgu3hOGuSV0ZABGO_Mp@U`Pa{`=*GV~`T3jSUaYyfr&utguL(<p^c+cDd`|))m
z!TH`LcxbyMNM-A1%_VgM+_$CiS9|BZvR%9<nZW#gKqj!t-h%{5K{5bO_AbQgb_o&3
zQ(5)L#b7-bX1&}y*Dv(Wbx+|*0mi+l;fZ{Gk@7LcaY@2gTU_!z_}yOOlFvi6xws_b
zL!y90zP?CUY>!0FAQJihRwI$9gI*(%uai7H68SUw#Ul~^>jxc)+=~6|nq@+eLml2!
zQ2Q;%9uu4kBqI3O6Y$ic+laQ*%YIDj7sNzbm$*pa=i5+h`klfSf{Aa%UGZSzY(ALS
zljc--tISSa?e}w=<6tvvE@~7*#Mg;9eWNOIl##^$f_<=DCi!(tO+}UbDyd}BQu}G>
zBB8`ZBGF+rtsat61q)@clwSVH&xrrcwUf1ulFx?)cJZbH@Cbk0;znPLya_{X$oox*
z*$eWX#VF;FSIo2sa(Bl}wPWpBVC_H3$y4~$mrf$R`BB4LM&g~X>6+M;RFkQrd|Nq<
z<d)lNPqK2A6JvW!X>957hH1p9jt_zVM`cU!|H-y*o%S!Mir5opycNEUzrQdlXY_)0
zMw360V2IsIJ;R_4yL4>G#WN0(%WF8A-l*XYo~#<86ma0rW-_lvNBKOOvVSd2aWtVA
zO|x`SBEQ*<{OSCwb5KDIjUoOUDavg*R}XB1z>x^s(pMt;dN;DiWB+yteu)d-85;;j
zQpr%%MChtB22qn#G8Cy-FF_@h3`I>+M)_~4QQ5{Es8_*M*EvuTTQUdgJ*s!?qzIV}
z5J5z*(bQ+btB9sPN56P9#eZ!_G^HL&DM6GsuarP;=9tad;9s-ehy}PD3&4N%DgfUf
zTBaQ&0<bZ$av;uZz^;Mv6pQE<(n967)q1pdF9BkH4{U5Wopd{8o{Y{5nc;{%!;cg(
z;=T*c8)@r(mx$GPfQ17%I?uWIm=SZD9Sl6Z;XB+v&}*g@?EWo|85sz$<FXT;-@!(s
zIz|nbEs~~>?W?vtI>e+BA6c>|DiKGQuwj|efzw&`<ISr$Ze`iy+Nt4%?7L{!ML*6$
zWrJ=jXtsiI(Hz5``6~vFBjH<E;-W@iA7Ec<?54llbKL-}B3pMYWPNkkrC+c>OI<Fa
z!l=A#c@D1j(mD=028HfoEywe0Vi_G^J82_T{4Co`2f!fTv9d-U6r26;^KlkQ6JMlw
znxv1e4v7*&o>x&>MaFN1Z*&YXKuI=`M~E!4J)XgLkaUbo1{ArIjhZ@ko0OLWY`?K)
zL4Qa;Z23tkUS%w`R?y!*I%)tJF{pto`Z`KRFc4LPgdzT{2DDW{KD$eYIAC!~D_EYn
zXNE6*xjuWdVtAhGRge!^a||4+g496I!m1cHCUl<aWq|h*vctJMxi0MQVVbdhqThsv
zsP*;P8@a}8oMeF2`6g84@O|g^I(lQJrLP#!I4w8YiGzCB0nezqUJW~p(@_;5Lx;mG
zvO!F-WvuXlMyy(73RuLmk9NebBt>i|787BcPh_!HVq=72d4Yvl>NT+0IJz0*CEUz;
z{1g{8eWO)#E7%_&MdLUM0t+81>6Y9|s@Ou*@-5nL6@67|mQM$sKoH60P&7#};r2in
zABj$4qhDB!tX5gJ90-R~a3<*&h+r3fp0JK>wb5G*L%tg)8eyo!<@r8927M-X3~c>x
z!usntq~hW6qikQ(XeNstNIsnO(S)^wcbXLJnkdH~#r|w_9f_|EzU(4oS5ZyQ0BTl{
z0gvGc_^|*+bLN)$JE6~g-f}h9J(VQg-x_s~X{(dTUc|~J%LtY^4M4%18Ppw2+!35u
zfLX*(Tp4DEVUzVrJUKDJ#p(pP`T7PM_tzVNzE&}@>%2od&xbfB3FjV!Y!Bb0wIX+E
zeHJU&l1*DHj$75GY7ZkN82n)1#e9R7xe8JHBF>T7;O9b##UcU*Hefw&)oLum1|%*_
z5K#E7rCc?uPPU8f^~GeQ^P4V)Fq!(2QNs*g2M4Bs6F@fDRfzNSixmThjyM&ZU`2?F
zk$;x$muFDHR-I*mW*d09*0@CHK(S$gCYZjkM5N(am>U6()}eDnxZUgrQ?RT6K#OG#
z$nz&0XA_xHTDr!I(c5x-LEpycP$^U&qC)X|%KlSwrGG+y_9Gw9-ba7FLYIGtKP@&w
zb$RL@olC`QZ}6HMycY69j!-#ufa4!dV)enG+y%95`9*xx!pYd5M$Qw*NcnKO;j<L(
z+`k)riM;v#=uip7)x?>3nzO|ZPvC0-`O-#vFEUkZ$_ce8@yDjbAe+AJ+4SwtrnhM}
zy=}AU>CUF7zZTn@wn46yy*U3xxJYP9ln~TJ^@RG<B{V;HDQvdBgoB<kRS)J0<Ju#c
zY9kR<-qS>;e|QZq2!R1t<@=I%;I;vTRkuYj2Oo*!L>w3lOauOS=mA;&GDhy{@Ew@c
vSn3FPfOYt=V?i*+!%Xq+xR{&bZmUssS~W8L6+$m8Facysbfbs_YhwNnfEafU

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/sophia/index.doctree b/.doctrees/autoapi/lmflow/optim/sophia/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..56ffd4b485d9fcec95ec004bbd9cfee1eeddfacb
GIT binary patch
literal 16883
zcmc&*dyE~|S@*u~KD`fny-C@)9jmmyo9y1*yb9jfMkZ2dapDw03!+KKbMKwqnep72
zVIJOHs)$s?!Ap)xOD2D$C@Kgf9!`1u7fNZ0f<P*$2nw`3g+RRjgg^L`-*+DKxcBbt
z-K`Cle0S!|Ip=$S=R4;+-%De!|J#)l{2!Z(0?YPx8%DeB`E4^w*+koKB^}d?)7R3K
zPp2=Ybv79qJCUD+Ei+{&(4%G9ZaXx+^o5koQuo+)qlO$m9+|D!_PuPlHD=YU@fXs%
zHOa<eJ9f>&CwAKHY`OkU!w+J+(};YaW2Dp_y+?n@+$A*nce<4d>}r7lT;kMPV3T@=
zyfwmV66b4d)U`b`y_Pf$Hf6+dXg8A>v=VJ2b~_?E##^otMQFtHm}hjv(?mN7@|UL3
zx)GXN)F)x%v9V2`Qftmywq~pctP$%h`+;@K@0jah;(2DczT<~C)}ManDG;ds0KsTo
zAhvEKv2O(Sx?sUNX92zu3)XF~ZSFS02z0&o>TuGuPOvfFR$!jstMxw6VVwiV-jDwu
z!2k33zd|&DEuqaRsk70*u)~xscl>tZnonr`_|CpVd|m7J(W+gGQzNk+vet5fuUYQ}
zTdirKDY>ZURRbG;QZkkcbt!+B^21zL1oOmP#)0$@XyWx{(cM}FXC^pj=3}FoxJIbA
zd{>VGvvmX}jq?6%{`nvI)Wa`+>X8@K-;}^Tn}cJ`yDr+zhacO#NJ<bTosJRiU52c|
zO|~SB658J=d`{Uy;Y|n4Wm9TS;ooyyeZEgUY-(4u=URqKeYNLezhxOwY`3)M%$D!9
z*Zi<;hT5}ad-jjxY0D3_rwwmAF}6+Z88`%2`)Fvc#i3z)wzu6-mO0i9T&m(EtHs-)
zp9DPuAnPB8c#bHY2@Kh>+p(2iog#sF5XcS4+*kT_m<<#r_|f77<cpeqd#?vk;5icd
zJ3Z~(eUU-+9Wn_gWNe2<U=;?kjm=<h({DPP*QHpri+wDsAwOQe{ygVz9kxCbm|hzk
z*LUG?ic_v1aY`kzao3K`&~OWLS#K(nlnA7`N&dpYCJ9%<kz&KaUdn30-Wn)O&7$Wa
zMt=d4`Z;*q1I?M$KpNa(%FZ3?2ve*<)VTvaOs%)rINsvD0PE2R0cX>AH+KWqZrO2q
zn}6OhLNDEClaPRLY6mPMYyv&F^G~mFD6tW4n=w3Pg{Zi5cmw|a|Hdg?9Rf37huSY$
zwOmH87iDA~$Y&D9Yt~uoUARk{=d(1dN-#9wGwAp#0RBvuoym~E8e<<5N^?&f%+kG@
zJ!jru7MXXjk}BpsFPZm^BQUR8B>Uj-D~E~UWpskOJ9s8GN<KA3gaD!0=)KiFB|B4C
zG&I-uwuSFDx(&9C(bv027dx@Z-Hy8%1qQ`|6G3Q3W_Z)&rtnFORm(D1g+(k!IS;K2
z6`>THWUxZ`Phl1rNM(zAMPxCD1qz{z_N4E{aLkc#bxW+LJ(Id~-_VH@g9{$PhO?NB
z;3&<*CK_EfF<AUni{Q?&V6zXZwV=+uj%IuPVzHG&98f75D_|)DvJ-3+(R^<lL~F||
zebFNu;cJ|PrD*&miCdkPvre?Cx#6Xa#>U3R!>d};jE(40eWP*VLyxa&_;X=pRSPYj
zIy|<j?U?qq73*!YW$d8|10Gz}I>s)KEH<^K@4MP1?J2{J%vG&r1OQDzN%_l4wgAp?
zBn{8=W20+3!5;a<#rMYujyARPoQ3BvuWA<+QHcMbpqpCOMLU1_G6G`016g8Y!qw+0
zaU~OQ#=49_Vy?oH$%ZtgWq8Ubx!N^J*`ctZowVEBoR83uWrx5QtwouL3gVq9ba1Ar
zm-UeIJz~%k^oMUaxKf^iQX*s!9K%MAb|q@_22Z`%Zy{4;^#7Ps*33Zl3(xCp8hW$_
zU%-K~lY|53Fn=!v1!YRES5Pv^Dd{F-Ttz7dMlDW!2(lA|tiW`FU>t^ISb^jzjwH7e
z6Ool`W2;tR`p*B81!V>#lsLn$h#T#gVckLnP6(lqKhV2bUh24nSWk@0wc}?i8P0~h
zUcLXU;3>nOs>P<PEk7*=y}}1Mui}sL!E7>fQG#^mbH%U2Zm2Y!9&^m^(2U%K`suav
zCO-F=tVI2n@r_&D0#^ZIoF5>zj{UI6<)KRoGM;k4zf%dA4f(w+`)wMs>`1uSW7!@*
z&?u-|O?YVqQN8A1|F{|~40p>&+*rS9xQSUtft`+$An?OjkFZOPe=PCo@MNF-(btNY
z<x76G6RQyB={}20Ke3MSeu`T8mJ|CgmBdyG_a8AdX*yrRpG?ILkmP??L-_-I?olBr
zN(tqUicn6LmR3buS8Gn=@NlI9f3q9<;ZhVx-32L5sHmWy=tl5S5yzi%l9sDUs<O%Z
zFmyn%tyDv)5}Sll65A++KZ+pnl2i0u)v%5sB!R+R5Yl{iz}8P#GsRAd$Z}&&>_@AK
ztrFkUgNyHz)ljO$C!v(YH-b=A5yqc$dfaMA#}J-C;4TPn=s}Kimzxre))K?#oQzx5
zWK=Qi(_Ad&V79m;9C(pl>c0`_r*adjtCC9Ikkj;LHBD8j_6Da3@$zD}dIyQuij$sg
z#?Yoh0Z%#XFIK{4LqU1<q-F9B4Ow;sXy^NjUjKj%#g{9AI3Mr!?gT~dC?u3p7&J=T
zFG0=c)v_i}PM4VYm=pN*O4vpD7f<EHabSKhCpTK9BQ*hupVrYiu)Srs%13n8JH_KT
z5M{jqzI_b{b;pX8i}(>3{U4RY4#Ma`LWTdShEipNN1&xPQW+8qpfn}dIL0^Y$_C_H
zt|p7s(2n6A1*vzzJ>GxN`2r`VUfv>1<((C+#ipF#wQ7Q^oX!e{B0PBuYM{)I9pGM`
zsD|=3rs#<(rGJr72JtT+sfJReDiX@^{mUmRVISmQoWBL-`N6;Q_w-6x0d-z~Oy}~$
zTr}l1uF|<2fy+rHrIpS_LK(!l{A4whD(7+pT0U0|WdMyiigWowHMC<m7eVT=oXa4C
z*n2wIKHEBN%=PTl1+O?3;#=jHv1DT{>pi-|#!TG7pmgs|+uUg(@0zkX^{5+O%PJi$
z9G3GeTEzpp$bLorZW-XJ0}RM+r_N1amnGz^w<(<XGKrvwusA5mnc+ObqY&IWz%>e-
zf~0JU&exET!wryfLXg#D4$gXq8?IGj6I+Sv@&hLUd`d@lnm9)X`S0+ySCIDOt$8;>
z>;iG}tT_1uf*+z+p3=$=L6O_0j6mTuMzE}I<@u|P?=^7UV^hoWsW`rAS@IqVzluQ_
z@kW*#zb@7eMAzeTWFaQ}yDoLLvM7tK91Vc=^J;-`{*@-Bn=ni4eTN`dK;_@NP$_%1
zx^?I9ysh)kn4>%3D*0uVy0zXehJen0VO;K_`{ic$3%ObAE7m*KSGlCE-#eT{5e*8t
zFGB9b^Qx0`>Q8x3x+X&X|5`EzM|Zgv+`1=O%&8|~!rE5ocQ!REj)Q1(eSO=Gt)$s#
z`JMHr?ZiSJ?!tw20ewAFE#sb}Q_z&@QDHdNS(*8m<ia#@n2!u!;c)Rg2c`k4I#}mC
zsyfq!MOj~av@Qwl)VU91$P=8VvC9q-GN>lpU)BVEDWy}o?uLxX^0{+b5c(|>H^SOl
zb5HY<POyi&3fj4IDn_jABgg}&RiGACT!JX?pdS?UC{2~hyL#N5qq9L?{ZU<y%s4_C
z&D8b!idOHJ4wAVQtz*WP-$wGWs#l{pV_MEc=C})`avlc`%B4v!>}(d=*y-YIb&jV@
z13FQ$-%PQel0vyL*Q#fUlLsE}*PWlh%+5>rsYp@D?9bI~D(*_w5Xavt;x)!m<_MKD
z6U7guXu0qK1|#2c+|3N&ky8@}hE_T5vbxEG_!kBq<a`M}v>46w;mz-&OIh9r-b^Wl
z?}G&@Y*A`zSqJ-vE;)8Rq03XCnO{e!J6{LF&Y$3?tP0;ii;PzI66O3k^?McLoWJ0o
zSeB4vWi9-v)cGcQu%&*e)>r7lL@|cJZA$0Em{!Cv&R=2Zfc&G0zD2m?6DVJvM`@Ic
zsmnEASa)}5Ca+aFzX&AQIKnr6v&#7;dgW=-;=AXsiEF4>$vs~^&Z_1$aYJsO&B%}H
zeE~gWSQcfS<~(1useUvQ!Gp_+abZHl&X2*~xHZdFTrF6~x``pY;^RrsCXZ;r$J>TC
zuwQ(z!h1H#bAH>NalSksREOVUD7(ekawp{&yPI?O@`qB_l&X{$+#|WhceAOuYuxB|
zjl&Y<13f_5AIU9JpIPs*Q^_o~i<kY~N{(NTTloqYavZnvpBUHcR=Q=V9LxR6v6Rv6
zbuAn43}2Npf^3fGTGRweUj}zA&aV(zBV+jaReBnwr(esS?%-+Pd5iykk^lZW|NRaA
z`<wjtx9FGr+&yps4@I#V3~c!QE@`QoyA7J@+cx)Qj!(<Qb@0%<bj|m0QK}hTdi>!>
zHu_rrsDLVVh~g!;^0VOeT?&BSlTF2~{LyYJKhqaw>8oBXhDtNVp0YzJGC~ywr(Sun
z>^&#!(F~g-M@Q=z3giblQ@c_PrRq#gLMfegjgmAKQQ{@1=-Fym$2jW}D3lGgxRWZo
zT^R;GB)C%&*tpccezF>FW;ZVBF{j6>rl;zhye)7TL_#;Kp;SpoLMcm#%q)`-f5|C&
zxf<3nBqUHcmV^cw)Dx)pvEh5Xrp~BEihHU4D`}2Z&q}%omLF#&y@PSRE2;2~YW<uo
zqSm`=K8jUyNlN_KtEQUZxU1&7L<e6rf1jTCs>z@3#;SR+3Kg~#>Rh#w0-5D4Z=LwE
z|EC!Nf0hxzpN>nwB5eX><ewA3=ETWzwz)v|UDm;=Pnz0e;8VqIb6oAtr>$ETRn_4d
zKMY=rL}{X#&EB_iu?ND4ScLh?i88)&uwC@YS-GOM$`(;@<wY*Pk)Y!sY8!X$MY-Zw
zjQh}Tdi4aGQhm%eUK0q;8e+4pB&1?UJu7{jQ7xhbl1<2s6)HU!s92JhZsMep4$Rvq
z=lM9@|BJHIKUVXTI7s63nl*>=m~9hfjr^h`UTFV>vyV=I!kOvN_hX=nY{G8JhSpJ%
zBT%4($PILulCMeV;_W1z^;0<|eZ?~J8Fp$56X3G9S!lk^PEs)`4&U@$RHk)M*)toV
zqA@_7A3w8;FcqREYB61tPVpjr2j{Fhzk&=h07(vzMTpnfOlb1EjIMdp1S5yplx5)F
zqkJ7>3+Ap3{^7VRiSpL5w%JU!G2blJ6;WXp<~5BDS|%ZgmPy1Aezpg0={Q04seB6w
zj3nM#`|dTK9(kbrQ8&ZT4|U7%+OCPx%o?GJ%n6$lJTJ^L#%D>{*{&~LQ#f-q?Ad3<
zyg3eP^aJG&U4z*?(EzQBEU?JsRURi@Squn{Q&MUeMZfHHh~mMpjjR4tqy{ilVtS3$
zU}js$0)ZoTTd4vPJ+oAQqUTL)7zrm_I>wPI9Zb?LpN7x!1{-=9dMFY=%N%dvYEIJ6
zn}Wzp+P;oz6!e-$-IpJkdT}Jzk`^wr+L2ASaxs_Qwj-)H03i~~ab+YhxET<Gx=0j6
zP~|eTMt0M;Y!}3#mUfAr00cEJ)CaPm+1e2KG^(Ss*Q-Npfu{gn6NNSww7-Ljv}fke
zv)RB5bShKh$LmB4YWxCdJ8&nXVE3z2b~HzgU=IheUc(Q!VM}MICr1E1_1Nq{JTWiA
zF@v=aY&-^j&W2m9uIZ^IpZ<etdYG+2DtiOjE!&K@I1IRgbt~?;05*yGAXpKvxtw5U
zp_3gwGtP_vSsJm==A%6?Hg<Ij<sUAUfTSq5Ms_1*GyJ-+F7FH9MxHajKCHVws!|1O
z&!Q+5_(1T&-oRzfLiE0YO45B^W6lf5r5o6X_V_K^+ky^Azi^5iMU*tVy4f8RvI3SN
zjFIZVZ=_I}PSb3|g7<)dWtzy)GTOu#H*^yfhOUlkc=EVF?>1YI_3Y49cT0G)eKulx
zH>sYSO$cuS^MxkjG|Yx=#3;?B5<;#w3qT5*6~oa&YjDe&{5U&RYNhlhZ^mp}*$d^^
zs5l+n$Ei^~*PVZ&<@6uuX8|jsGfO{zL$BY$Pr@dtFF!$}`gY!XpZDD7y|9@+N9{BL
zimfi9+_ffeP~;h(?gaD2Y)w@QD2mY|EWGJ$3K%{HuOo96qpE}pffHe7UFK-<0wta=
z<_U0zj$fUt^g|~m4xOkt6!AF}5hm<pQQtTRPNJ|=EQjEVXd$<v`UzGSTgVo8EA+F|
zL|J&L>ta`9)_bzlZBjwyGo32Uj};<}Q`q~N?z!X_EYWa!Ek7ZbhV>AIdM=a$YJjzO
z{7KpV8l-e9{XrNoQr~a|(1Zv&D*$G`uhcx9HFLF_w!M~{v`Njk$x}_C1k1-F%{`eo
HRBrq)J36kT

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/optim/yogi/index.doctree b/.doctrees/autoapi/lmflow/optim/yogi/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..6a77bbc11ba021b59e55af229012012bf3e42342
GIT binary patch
literal 17001
zcmc&*dyE~|S@(M15AVZqHYpn?W75ReitpXEaol)oyK;@H29r%2mGVkDp1pT=XJ+Tl
zOy}X<rHTk5Nxkh+Nt2003l&8|0+CjN{?V2qH6Wydc>GmB2=NyJ2}*&&UkF;n?>pzr
z%$fU`z1OyaCEuMnbI$qR-}%lt-}ln^n;(1s5&n-YM1f^{dkv%2^8A(=rEIF@H<PyM
z#pxUAnO{g>Olxd9G<G9D37clhj-W@=vfWl_dg%)(o2Txv?M4kbeljwfvF&^LaBJL}
zuqIzfYt}RykL}nsl}~KRZEv~$Zo>~^yWQCLx9yZVqwm;9%soP)SC>~(f!Rc00FOAe
zme{nOBW{hd35o9p8*^>XOgEBEgUuLm9NL>n3_6LfQM(lp4U<jRh$1xNdE7JF;%Tat
z1jWmy(cB5mE$Wl7$=KMYPpP$Ftypu`L)NHul0Ca>`E7GGOgzsFS9kqzXZ6RQ`2dK^
z{s6&fRUo!%B(ZM<_NpL%m6N|p<geOZ%iL>(5vY3j#?fTcI>N?zo5D81QR@*<VVwfE
z9>xF1@c%UapCO9Cl+fmM)Yw>H*kQ`v)%IJ7Yra?O<~Vy$e5-eVvL-g-)JUu+ta?G}
zhV?KQYRwA$C<MHiHL%I+lBHaj%f-8t9Ok+v*e2#O`qDz6iOW|-ck3*;GR3*F7#o|3
zYlM2!cl9VRn?o>YjQ3}Y&;QgX&b|1F^Dkz9Qv&yF0gm0g$3k5{cVYc92|r5OZ6n;j
z3Mqq^Y*`v2biSc{hW0COIw&riPh$rEei~{sGxKyia7{AYNTUYrS+Xem=gd&M=5G6;
z9b4^2W?SQ(6G?@dW)tyt=qEuJE6I8%!RR6RJcS{<b}P2h8#BaO53JpR^!!Xy4%1|1
zf}c?*AYZcSxAwab1)ig!zuVQ$-4B^X&mq%rB*u1V1eP+8)z*Xkb$`=YzsWgzxyw8n
z)PD8m^PI0WSn6D0dM(@A*7x8n)TvGnI8`R0N!O0e&~TMGtS?Ao8Y8G_VM)K<-;!Yc
z920gl*iYF+uwMr`si`^(;_9zKEWZL5+1Co$1n7bPO4%cac&-%71$Av3wwGF8Vv~4_
z_X8{)qXe4G;@#W}T)Sz<>23ab*9g7zfK7wh!VmTI`D_Z`xu;GyIE>f`x6L?Z%e}?I
zP5E>1%kP8<-xvW0--Kc>TN8y~-c$u+_GK<f;0^1f^+#}RG|y`^tV$X*;4|p>X#o7C
z4oi}Ywl&V)J-A0$9<Xyk+Fz@r_1Bw>wC5#h-yVXrYK9$v{#OrU(#z-schi3+Hby=z
z#X1X?+UUL2IVHPaalo4Ed)vZy8l5)O!swfwqt#Aqdav!SM}a|c)Kn0fks01Hxte|q
zW3vSgtipm&pj?C}Mz+qYO>)U6{AVzW3`X*0vSN*1c!mcA+I8QHY1t62XPI@i=Tdhb
z7&&reaKU5Qc249YI7*Aq*hYuO4i?2sL~yHE{Mae(i)-8yXtviK&YbD*5=zmGf}0Fz
zj<7Mrz};~KdG>kwqDM9Yzc>j?G3+lAMm1VN+R<6f4c9e<v_`C5(bgJk=gw-IW^6=P
zYHN*4XVG%$(xo$JH8Y4x-Kd}K*|BZ7y3s@gj>snVm8s5|^|gy<wO!NRwqm_yHjRB{
zf;0IFw7~H+JkO7fjyRm1|Jlor#)vT1wbPt^r>~yXE@yNhgo0+SYk3##^wq0~Cwb_;
z%*KW5Ep*{pF4>%WIgi9#%G$+7v`oG5R7`TCW0JBbVI#X~r=2-Z(~uR1z^i7G*+v!v
zJ2U9u%u+Aw3Fm!c(0l0*kJh<19)~t!%@!QP4v0efY}W$^dc1pK$%yFnsi0~z2h%IM
zt+83CQ5_C|17ybtKIblfFD=exDsEO#G0myyBte7<z2gQaJ_OAvLQ~*4MYA7<URZ(N
zagH9h5fg!pYh!<-u=>USSL?}*M<{TPrHUbKHDTXE0gecvkvq^kSzhY6jF?M|E41VF
zO185RuUqF|6P#uEQ?<gBwdGS{(5rlq^J)Ca94}k!bW(x?&S%w5N8K<}cX})cc#9_H
zR@6(qoiE^Xm(@xNeiPrg(VgJ>K>Q05uNfy@zG`GW1CFNx_U~81W+Op2^L~$ptT+O+
z^J93+`OwEJ{b?l-=cD-CMWA#`5=x1Y<9s;~<a|DZEjK=Jtm11y;@?$5FGmzbs4XJ=
zBHEVzx)do(P&@{wjhww@H_L&y^|l&jQ&90au<mbwQK#MwAm)L1{a=;T4!~=8@TFXt
z`~aIKtedLkj$Fw#>58IBQ%9?`n2q?|nlwQ}hSH=5DuE0_9k%4Q$Iq}7cR5CZMMjl)
zDfH)bB~0>MTSnr>`Yppv%rcSez9<O-KaBMVhZFI~6Q9n?4rm2^Or^buPOF#Q3Kw*|
zJ1XJ?)yht2oOjDem2U;D*D7hP)X^&#nru3g_><Wr>!T{qRztZDpS%2l)D;QkaTUs(
z8VY8ll^Y8RU#NzCz7zK0A`~dy3lVk)vAsk%if$Q|{J9|RRyA=|D@ll<5Z^c8J__;m
zUNc^<hVm^;(Iq~)Xh<l7EE=y@L#bLcB$T^ZG=8HJdcQ@(`3`6=j%b|k(kq?cWM?aP
zx?+5hOQ#&oS6br`jQ(aNsg*ueLK(#7|GXMXmCX-9%imT*89-@Do@E^CLa{K&x5C{1
zsT$gOd@dQn;l4zWdJlZb4<Eky?d4nU>oF|zH9fx-%orK1TK_AR^#@?ofWBv;8cLP#
zkx&NlJtwQ7RH>1KayPzbwG#R}?t7j9IzvUPLooX3N>VF*kAyOa@A=7UC{?~^2wKc)
zC<7=>+4t}hs?7Mt#am%+Q8lzX@I8Xmd*FK>=yRyn%i+waeKTUqjRnoDn&ztT<~D{7
z=wd!!4W-J(NGOB2nBS^~Ql%{t%H6n_Z&X5m$6d@<xpamIZ-!v>UsjS@>0%_5L0rt=
zS3{|CF+<Suy=o`}C{4-5jI3oAho{2K{<j*+9k>=j;vKn`K?Wfg0~>H#r)|BSpElwZ
zhoXG5{W9Wk95i$vS0f@52|JW0aLYD#o5<Qp*+TZH8(z~Y9e*5l!LCO8KctKM*Qh%!
z1(`9}7TB%SxkY<~>^j)`I-L!yjCG(Z&IJorIOj2lUtb~X28Vtro1wF1TnHmssC*sG
zCgdHk^)}y*SQBh&D{)<Z%qoD-=&|1<zR{i4+q~^nTs!mDq8lN0iTHR@ob`jB7wMJX
zt>(x6O>z09oEHg()vV$ob?yBIvV+#O{0bijOifFsEb+`8O0#I>2^TlT!eJrpMUGUI
z^)m5FIXt!U6G3)nH~`icvyHFw0h*N3rk2?whajh*^1%*N%Kofoy>)ojfb*l6qchkl
zx#L1p)WiViIgBedp1pDd4wT#g>zmfw)*o<5TYq>siQ?Q+$o&N5PCT!=E292%0Z+L_
z)c=xX3{I*GE!Y|4I(_EpShI`>SAM3a1)(4Kk=deK%A40(1`={^nOfV7Ex#3MTYji*
z_#PDSmbsT@>+oann<7uA25DZoE1-(39Z8aPQr^~0vjv`vr&B%#uA*=$tsF4chr&1W
z%1;bIq=H-Afm>M*Rc#`h4ypHwLj*xS%WR~fOw~zd;et93Dpi?3*iYs^VZ|{nYFAeS
z0|(2<akPV|(QJy>)u+xqb+K+pxz?pz>yT^R%j8;@?t#ly(;n44|Ko;q57XQrx_ST1
zJq4P{<{r;ltiG=gEIqzzB%z%;DG3j<d`-xbWwM;gSyaqVA617rm`GR3B89f&*qa{~
z_?}p33vHK6<TS&jRH=`S9IUo#79ZKbf%y8&bD7C#wEV7XnZUM@WS49<ntprrLWrUE
zldF^(P|sHe{RXZlBFfoQ{^ja0hKtb>xN#0Hlc0PnIP>)m&Q#mYSE;w~aVNo?uLzz`
zvzdZug;dFt!u`GPE{T>@F0;kXi4^nwvkr=!&!Qn)<=E<we9jm*^tgh>-|j%DKZ^_H
z{wv~aq1;5%po`DL5X_(O{`~>b)ZZesge+-@kY%M|IKNX2A@P5C04Q&jhbXHe&joFl
zvc!=uv2yAZt$9ro$<!@vb6@k4cCe4ELG9G3Ocs@W1mA((_W?PoNJlJRbpJQuS*+TU
zF>6>I?kLsBS)&N|@hCQf+8M3Zodm)Ykc7KOuF*>G4h<TgmJ^OL4rYt6^CTc;>Mi3=
zXI*Jy$2k^F7wN*Gc?IEuDKmYXNbNQzfQblDyc9b8$avX-mQ}0f{5)oIUcygB=7~)0
zSjng2vC=u5P`skzF{fh6MPAN3=>Fbu-G@|sPHLtu`FlamlQ|AUXCIuyP@p`>{`h(|
zl&bx)gi_kMj*;jxqQpx<(TA&H-NF7<pwQnTl<ntJ5FzZR&bd?84mJ2H208D7#i<+B
z9(wT*x|CJBZ~klP#%Yi}D4Qk|6c*NdD_S7bwG~|knPMyI)PbYZz)xB8&!a`|KKYiz
z`4Q^3hjGpY{)vgSDkd!Llc{qN&1^PDs&<C5I@R6hIq>&EDQ0NkQ|Mo2j=JA`_lL|7
zQDi&P3{7M`YNarej9sSkBXM`2&V{mlQeu($p#Jg(3Hh$*0`Pyi!2e5K_&dL=G_e=1
zh1<M{C|aM97QL=r)8wNTTgZ+KO_MSZ3}hIf0sA6^Qm$rrEqp<Sr6*e94kBLU&=oK_
zUsEB@6c95rQ=1C<e{~r89VKYG2p0S@219&*tU&Gxl&;?cf-@5C$y^2PUmk|`T=wq)
z;lIrZuhbPaD4`*OSN(kuS%?yp5PRPo0E!Zn&=5r_|3{I^J&Z;E0!R%Vi)3iYSmc9;
z1c1(afIFzkTtsG^W$xAFd5j1>I*-j{$%XR|z?V(pXpL80IR8kmJbhDr_x!r3DZ?p?
z=Vy=eS#w<!${esc`7wK6LJxUl7Ui85JU?s8`q4}TFEZm8dE4St^8&3ibo3=-HF*?K
zCQ@B-NJS)PvZK37&d8j%vN!){Fu?gf{d|Xhen3C}l0SVHPY2Gw^51{szyHpEzsG<7
zga7^~{Sqz;0X(n0ivYGD4li>aA-zi0=4GF-Wt8dmMwY5fj<z#*ABpk>UuxFntO`;b
zDAf0<0?5qg?ombAHT$U8MDxH6Wrh81IWL~H&axA6Xm}A)(NWQ<Bevg0)q*IV9FLJ;
z?xr`6u$ion*}`i=)QqtCW)f0qrk)pK&Sfp4Ad^i=-9XLh5|wW9BGRppbEkz`rH@j2
zT9j`|*@U0OK@z7M)&lB(woMdS@=QOxv}h5Djdp;_pxMy(W1xzi{9ejNR#9;zP@vQH
z9dwwHuSw|Q?HC;;Q*kGK#c}dy*zqk)fb4lwX}-;lQCTbYGx{E_%qU!%k5FM5p!Se&
zgCb0|B3evYg{SZ0G(hKh-5>*y<N#S1xxwZ_lc(Cb<}DM9Jj!M)18I5kb(}4kdp7uo
zH2x$iTE|=FX0na>=BaFoO3g5@X|&NY4M8+bB8Kp@J#b6M!JbbAX*eHG;;s5qb$<H(
zQ2C>7hM^zo78YOEM5*Wmp^Affwjg+}%reerN!iJcFC9}j>l1K0^J3lthivpi<qsW$
z*&@*Zt&1$M$mR7i$6UkPPK<369H$$PqbNUS_l0KD4_l~)pt3uFp_11PHUSgc!X*PZ
zVz)9?z&*k|Rr2UX6FZ9g4CF&%m#XK=2*&aT8&(>628x3=aCSspxSI30EN==TGimuc
z>Uz+$h_W^u_v-3Mt|d+6TH29KX{ne?Z`l!*ae@$u<z!|gFt|Ao6TxMQd}wHm?51zo
zE{H>I?=n3B2&#v;RicUjv$Y`dS=4DmRO0o<2wVo;FS#a4mMmz08wGdIES_fbff*F1
zzC<6Y76WKIa3`Z+4`$~qXpS1eKF+JXh97RjmhPvX90Byyal{4j#Jrrz4C;@qT>wAl
z!>wk=^wd&J|9Cb%%+?^4-NEIFZN^(123!HoYuYY=O{3BZR>Z4Er`Sp8WLwXT69z7h
zPGzmh0b7jrz1Y~(EmX+3R3($5_#5?FDVyWjth!8Ry^Tw>BD+<0{iY#Vd$MUES-%Ov
z3wr~XISbKy2Q{$=yxLuynhQ5@0PXRcwzmZxkbdDfIf^LR?C55<Nq12eB#e=&(sxp*
zOncL8!GiaJfn}na@qpcr_|e$WO%za~6sUzldtRG<n=Q%8dg!XVCA`@I8#TRKRNc;|
zgg1ftLKCr|+pvuo1;bP$$@OLlNI|n=I9m8}9Qu(T#{q-dO1Edc8MAHWFLZ@NrSj-L
zNsZ#U=FG#!omu)>!piQ<)6WFGj^Za_)6|y-0aR_zdmr$g2fP<n)Kk<>6QFGEG75Jm
zWZGfZ(QIBKb~#^9v&9p|@evl?^foQfFVS4$Tx={WhC{qgm|0gjTD<6sUq|xWRfl$i
z4(%2bc1+bRPVF(&sEVZz91tz!Lb85>t!fMDJ#U3Bwl`6hUg~<HqcQJ2UFtTSLFF?Y
zFU^nDAdFMk_Whl6$uC%r;ozEnLhcJ|AIckDC;~)|b$0T)Y<~k1c_n=g#)?~1m_4K(
oK|ck++|QMoXY*#RVzai_bdwgT^fq}VoFc;&ORq(d2F}6%51rcar2qf`

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/auto_pipeline/index.doctree b/.doctrees/autoapi/lmflow/pipeline/auto_pipeline/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..68f325f55b74cf81710c9b48c18876eb045745c1
GIT binary patch
literal 15937
zcmdU0Ym6mFb>4a1dG5^4dUwDYuX7jM?A`UwjI$07nk;K9dlQ`X4jbDb;|INcZ}&`h
zP2cY8e(cNyFBnId9Z^w;XnsH<0*VwO1rQLN02UA;MR^M${_q3gpoo7VKLSYkmwczH
zt6%rtzB994M_Fm_bXT9MI_K1R)u}o!j(zfh3x~u%J{NQ>+udH)4a4;eGl=<w;kBZ+
z>4xzq;?p0Fug48O>FZm87x^tS=7#{$vTVojO*ein=CcGI+D@>X^dAq*R%m-}+T9wn
zYS#F3al@MAW1$^7X6_R|;k4HsZ)>?@cTC52&1F3bJx#sE)DitgpES1#p+Vpui6zFh
zjt>07*qY~)T1t>L!fOc~YkZVui`Sy2&ZqP+^zCL8f_|cR#5Mw=ZM@~^L4Za)kGXnV
zK1~==Cwpn?txextCzyzjhx!J6imf?o(VDUDutuyC{FxQYYnv;6<hrK6vgP@kE010N
zO%SdA0Ks5IBDO;0>m7SVvSmdvLbRuBS+QNi++OwrP<ZF1;izdH;$xyM$6CpC>sLUL
zbrM{>3;*xI|5NyXny3SVd|S}e;G-Sg_G5le+cP4^{Ed3`s_?rd+)_1|Ra*;VJ+kh#
zmNIs%S$BfzSkp<Yk-p*U9V>@4Kliv9M!s9u>x%R`v81hsc1w4hoqAIbOr!3(bvq2|
z#O!4$Bnt=Xng|Tgd#7A2iJsn<ZA8rl>x6X{jGs_c33|p`Ub~<-))-$rd9vQ|y_N|v
zGV0Bpx*N4SJ6P}f$&(80N*IYB=D89y`y<6Cx7*HY(9ualCpx|vnEo|0CS2DPjtS1w
zz;#;cb*7j{R0E%SC{qQYdkcl8j9sn0vx{`$sstBon-lVSB-ERcqx)LRbF`piwsKN1
zNOVU9JU{a6Qy=)%>mN9GUHuglKc7v<om;4StLM(Go+sH1qIO&NcP>B|A=!K(F{!Xv
z!k8tWV?Lh))<AvH5Uxz&-<3zNTzu@&wTs&2hpt?CbnUkaZEF`VKJtWi@u?>tf9Q#+
zDP<bR$|EJro8+}{!}p?27hlN`Ls*@CGJOJFwrnG`;!9H`Ssw?Ve;-Tvhe~(`ty*q?
zr}G1lTWEU5PESNfd~d7ghcPl=-F>VF6*~N})y~eU*JP_#%WSQmuQXiCW%vtMpB3zG
zz|PKeOxLj84Q(4{KZhB=^g2$b7I)mSL(|ut+&tE|61_Z9#xl;d^Sk}E6SiJp&WAfY
zF|T!YmcXLe%)16!obO<DUxjn*YY2G_^uv<H{N8=|tQf%rb!+Rk8(aUv$MF{KbYQkd
z2n{}scXPYr*eyGZZ-~!Zy6?uja2GqP((4S20$>7nR)z15*Tj6G?r)f3%$N7K?82OU
z10MX1HY1mY!2MUCTMJe#Thv$bi)!{=7+O8@Q2!1OMdN&xx>c<tb@(y>zlaIHqO>Pb
z=2V*oD7Tz+(fauwC#pvOFM!RT1Dh`tv1#;2kH5ED`||d3&l;cal&SqU6(amSw`3lv
z?|-9aVQ(5bbSP2t<EfhGQOk6l-A=F~((DzxG^~Fftm*7H#;97kRN1W`<`tOTg?@Y2
zVe7}%Yu3N5SwG?T?@<v8B$JH&Uuh$o7aS@mq-e`?zFMfElk9D@RzfY=EIsZ`GugD9
z{3ex?T@=IgH|dEXNT#c%CQ;F9pg)5iBSh#A#78L(iLr&j+I|esH_B4TkLRMMy5qSU
zGTzb44(ny~eyQ9$4`jXH(rB6&0jR>!4~md2!>_Ov)YX=$)S7IdXsjuWm2C6U(6%DF
z9)CF5P$s&azNbc>y_ATBb)uIx=joyO5n8hkr9A`+Z-(9Ul_42mwXGMd>og2P=G-v1
z*4^EIU$3ay`mjQoKTSdJLx`)dBq@QZsV6P7F8p~NDGGbT#a=4(>iz+DiXT-|CEVe~
z_r`%DS``wr&q-aqnW$Ke@%bWR`*DYlq;4wD$w6K)58KQ6J^VV1T6T4RPvT2io&`RW
zpFii}{xOyCUOKTi5BGOsS^C?)eje@z=ywngcc$9I6(}VYyycFqhu%?m$i6EWQ*><9
zwm--AQ|cid+qaX>RkoNYM##xtm6P{%Z2VoXuj((njss>k`@c)<>MvzPH^l*;%GmYG
zeX}cbSIR-DoE{Ai$IQMi*jkMSd~;%F`j1z3&+Icj24-zE_HxJU(><W`9xUgPbM7Z$
z4pJ!@Vp%FSbINIUR?m}P_Q>Q8D?IYtWs&f#KkT_75(<H|P7A!NJzMPuhKtHk&1Uk@
z_oqFChpq#q<&tUNTb8yKAn|~-5noL=1xcc)-mu-e+#XlwZcp!j7hFi==7^ET7%7oa
zc;)W&9VJU<>5E1#TSr#6kofOTUyv6N85YrrP9!Lmu1+D4O}(?)rmD3Y1ePu3)pTKc
zIu<?+%$JZ(6^QV|gqH|I#CsuSl87>UKaIG^o=yiAEB+R&_#r+i=qeG#o+VJ}|GHU{
zq3Wk2_C|GrP?dO2&^-N7?DRvC-<Hc<UbD@umS;fL=G3F6yDiJ}b1O0+ye#Bi-l|FQ
z462VdaMFSlkZr_*F*&ac+s0lhqb_-o*$MVLfDk4Er#Rq9%%_MiNG;(|s+i>BwIrKl
zy~f96vdF6OiS@{F1n=bhQ(EY?G@MkBX?RVvy^I`|XwAR~x2Fg<c`oo4NWgpORc!Al
zXDC*_C2h>!`A5V4i6mJKr?v$d$>akXa*ID|=%HSwGRd{EFJd4*4o--48~YNy3Q5Vo
zyWWyKM%v}$uBRSnRr8X}x$g3r<fD3@2Sh?%kOIxQo@!HYG!o4hm|+N07Vi>?Pb92T
zmNMD-9Lz?tcN1X`ruH$PLD-LxJWz(B+ETJ7fm+e-H`x1fZM;veM;Jz?<sk>zxWu9i
z0?X{n_c3}t<STl$2HV7U=HREw?EPyxh%kG1M@{xTniS3Kd-Pq}x;~_42)&k$)zjdy
z7x0~3$4^g$5+m1*5VUblXru1BUWil0!bvqBeg->{#1Ey;3&&G*vX=8iLv`hB@RZ^F
zi%OjN&{kZIlh}VzmqqrU_#>ReF@YS6mez^fRv2|*IP{Km>UZR(9vi7zKR(pyo_d73
zsHxe{fq_~m-iN+pm6Ndl!4LxoTo4@2I7W2m#ja@AdIXE?Kd8qd>t0#*dpQbjsb?ZF
z?(LZ}thzkYKcJKJOf1GD`y_q}qRT#o7WkM?)1MmqBLbqC*dNPwdVV=(pHb~4r*k)$
z@=pXQoDTae9@TOVU`Ec(r~&rtX4qE=tH&T7>@Vr*GCh4QeR>5?yX>#T?_Z1GuZ!Pr
zh~ICD-&g7P7P*7B<+t1QhD;?PV6KcO5h?Fy6MwU|znJ`xCjV#PiCRQ{?wRv<g*a4R
zt)CJEoVswfem=Dp!bGjsQ;_<p3m23rTTlT`rg!TlP2Q9GyQNM=A5p3PG+%y0SKhg*
z_&L$;W;&B_Bn>!{TkHN_N+;~?ZT95;dSG2}e+Eq=-CuFFvvv9S06g~O^?Km(hS^4<
zvA1o^u<Ejn&!N-4Z6nR3Y$H7@+eptP+t`3@)X?15Ha2qC2L4~9oW7P&qYQOX74v=*
ztVk~`-3vZ^CFO(l?$owa-a$%4wh~54FoyqI3}1CsN#%TSFDXSO**0A`XkJ9d{O`gM
ze|kf%ak}29wC>9Vmmobt+YBwwfZhM~y?0&C&l(q<fMT^;q`@12$Md5#)D1X|cA94?
zy<<8Wt}+B?>$KBL9#bg!<xKMD%qA{QYy_ueILivI<n1L_A_?D*LzxyFll^j0=ub;x
zxhJ>svr?y`ELXY}nr|O&<tOP(!mT`5+Mn>5Jo{CcU2M!u_0Xs^C~9?!XgBt@LFvW_
z8<eBMtJC#JFxd1R8}~eOn~z7*6~76&kYa(!<Sr+jEI-lV!bFlEhh!Rza(Y562f;7!
z(kK}NPgF9H54qi2mPP8a$lioMRmTe|nm8`O=LI-hz@OCOO=a372xE6teZ}EazC%XC
z12jY5%NMMH&s{sBgotzac9TR*re>fA4W*ADA=i>ouy$|?e*5Mr$b^H8RSzn90DzKG
zYevob!KrE9JT+s&_zEg0kv*SA=UxZ5_`*W&ln<Q|fS?9X83iv^Qox72Zmswrby;M;
zD+nkG@@UFd%p+g2ss5mHs!HLNa}D0DS7_kGnZ!j&ez6Y$n^ZlA&!pU;#|-DsSL6IK
zeC|@Xgg<|ZZ?O0UMXyOOsa8V0X7jyLJztfA%lY=nqiS$|dpmMMTcg?q%1TQIfme74
zv3Z^&bQf8-TY1B>BmDM!WK|50dT$IeWV<#7KZ3*F^TvRQ9f8Q3{0MAF_6DU}1tlN<
zQJGNTi+z!9oTh7_`SkvSVDOKVNCNV7uuO{@Kt&J5bx#udwIskQ=7~hE<rhSu$hG{E
zo<y!iJRLOGB7)T<^<G4#w3|>z5^6(LNkMY5(u08EF-5=#JqZHD(?JQiAIXw#T=<+k
z=1GqyU_O&d81<>HPATx|6mljgj-gFc`6RiX4>TLrc~y(Fq@tuH9ha8UH=M($c?Q%t
zb<H{=BMK3tqPV9RSmswFEoZXM@?)W|yMZI}#Tsf(Hc$m3%bCVPB)XmW(jh*jU`zwA
z5m7nBXIqg^#Zp>Y8a1O@WQi1?NRDez9X3zJRHEE#-51AM2JVo*S9c=R5vdTB*Ssj~
zL}9#U&Efv_hKaH#;r#G|fW-_#z1_jh>}lWguw$d+zU`O~t>E^zM1joPCIF_A*T{G9
zc9>4$sN9OaB2@S?Ke~<qki|E1%{TaADnLV>jkdk+`E6VxpAAsw)IptzD9;Hn)S_%L
z9g_-n0&U9<EluobKn7+?IwY6%*7%HXc051S9P^q9Mh^2SOLx-OF+Oi@+u$FL6QV!?
zjqx$VY(^UxZ<emGQ^^3vHT5=HCb1ALlZYYwY!}?pOc&>^bSTCrqi}uc?MveP@{Zz1
z&GdcG*DT#N91{gwH9{3fn|x04JU7ai7zLN*OJ7Puuy;cJ_^cc^!$I}FqxhlJnI9n<
zAa#KS7CEB+=&+-^8<D<Yg5z|4Jd8p?ew%N$Jm1g^lgcMB4HcfP@fuWb9mllbh;1lQ
zz%lGBsgjm8@nM_@W9NbDDJ>O7Fjh3!xVquHI_{CK;cSOMgqYLnXHA{Jj0{ghofM#s
zpv1rnOfBD2NJ%TyuGxW2ks8L*3_GAQ9T1XWIj)oh3O55{0-R#X6B<ZO0;a9o4v0gA
z-2y#f5>#W+ITO}~Ml-PF)2OISUoQ>udEw9<6SrF}NPio}50{Ue;<FvIqftc!^b<8?
zzzNkS9Y{NHC#B$Tw!*Lztgft}Iau!OAb;U5d;SJ=={P|O1OTapW*f^BVmT(zzI4xh
zzXpEJ`s=OI@YIqG|5i0T?K4PZH%))pHp6u>2TZ|=6}BBrHi=8Q&>~SBG{H|mCfi!7
zoGAgxIod8i670C4zO7mIhUMTt<ad;=_QrfBKq*zIJs&}<12F0~@h)FA$3qFAWbKKT
zh3Kb+g_rsUE(;c-_a;i7c12yPC^k*Zz%Hc6YuWBPWFWB%N6AnGQL_YQw@G##PnXI_
zb)uUwM5f&|4QTKVFtAKilkD>2EkJE*CW={*moZQhE2>3r@cE<$7P9KBOKY~vM@;t`
z)tmAOX-%NMkVKrZ+t7^=1z}X9BIITsNI|kf7+Oe;LDi+#V}7*IN?U)?jM3K97plpj
z;#Gi;Q=@!tuw&$6=jrDK+{t1)^m8BlC0nAO8Tv4ZpNLOVFL6Rom9V1UuIRHX;CBTq
zBF&TZl?KH1=kus1t0g5i$_J?0!mLW=eCl_VlSRpD0OuUvp!n)#8cSZE7*&_55nK_b
z)>VlnU33;`iaSW+s)x8}!{}&D90)Q>Br{4hGm0EV7##7#c}Wz!IgIKe=_tW3*+K?K
z!O2C)w~&^IR!DBUiGsxf=&@2`3O!i>n^d!kk&YI|hwt#i7?$98d936MoHxv2%Ztd^
z!BgViyaTa;fWRw{Kb*9G0;~3s_;IK<HU}^gkdXk=EGJB@Vxf6DZ59GKZM!WeGDrb7
R$c9ZBWVjO>C-dja{|g9d><0h<

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/base_aligner/index.doctree b/.doctrees/autoapi/lmflow/pipeline/base_aligner/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..872ffa0a663fc7d70fca2ab719de1a250e3e82cb
GIT binary patch
literal 12504
zcmc&)TZ|k>74_OX`?U9A#~UnrlUbbDW8<AooQGgzISG+K$YhCvMM7Tfo|)dM?(ONG
z^uxQW;3z>#$SQ?MNFzQFd;mfL1PP#gK!lQf@sWg2ga8p9LPAJ<ASAvyx2mhVdS-U^
z;Sel&w!8XP)qUSvx9(G=uf2QI9`TP)hHYkhYgMz+@cf1qCVZsf*W;GuMafIasi%`G
zNrjIF=4$B2LETFD9zfKY?KT3-OCC-51c67k8&=c#!=Y7=Y~RbKvl1I(!;dBvHp)wp
z9l4hFi63)Y%dWp#ZQE_jwLPm^GegTTUAyU7K|=E|ZSbPCM)+(2{Y)b99cr7vE=t%m
zA2l*^*gig#QnADbX|H4{u9<wyjH1A<#Sy3{YWLZVkSH6jyJi?-5YHvgY{{pQM%>O{
zYG!>Uu$BoX=EITMq)!Q(WJlS4c8u*~v;3h2=C`bcAof7e!m1yvEPVRjPk?0g2MC4>
z60rp{j(oFiFG!{=D5fmvOj)qKhP74=LJ)Y}g}rf&?cpUcrtwv>oLvus*c^Cx9RF{`
z|C9KCiiiV$0$Y$&;e&0{4ibJt%WuT4bw|0cNceFHHs2SD4J}298MBYD`J5L^>^g89
z@*IVL%w}M=nT9ey@o|X$VhlDdmd$b)*MtDd{&JZ<+@};oD#gsU!BedO8h1&NNHok%
znL}8gXS3`K*gK-I7kCfX{Z_#=ScxB<n=7{izivTS8s*x0*^684bqKXQH>XhT3nTHv
zKKF&S&5`1xYb|#%Y@1|+BkjNnt>Chi5UxuK#}wyr;5sf%H@CehHNc1ON%dM7*G&GN
zGpfFQMv<ytlHh`2lR_V-BC{5|W?<BP*9hBIU6b=xsyQg&`P3s1K5_GvC(d3`e+8|N
z=iBj>|D9zmp1pbT92sL6w_0YfejcU<)#Wp(yM#NcYM&E6t-To_zUbuc9K%0S;=83^
z8ymZOQ~fTlGP};oc32ipkd~?{KUnHqIpyUjABviRAGf>MOimYC>+VnnBbc&kHzJl?
z7$X&X0c`&=B>SS$qOF9eE%1=O0EK{>-&pTL6nO3n{MD{;5jDtJ^_X&}!iL|y*j``s
zYtG^&jmzb8{ajR4gFJuf5y9dLT<QL{<uz=tX{^DqYv}O1u67S4#bMWutiW`&#n~@Y
zV;tA8N^*mIYjcBy`xcn;z3ugc54G3l!H2}sy*&xw4M^~Hgu6|>9v=d|a9asKz5OVZ
zAUB{{Ez|ZA_6{G$TeRMWi`qw6@Nv9bYi-xA+fnkI_`GTcUb2A@u)Zh*$yQMv-fn$S
zgo0#AY!{h9(~1&a*{*nn=lD5d@dxZNE)0O_ufc3)*ibIZ*K}E0o63WPMp5Xmh(ff^
zPia~|0n&t50r(?q_!DIrsh#GQQr(T`nc+K~mXrG*o!rgUPm%j%O76dRKyH5~{t!HQ
zc`L?!4?u8mo3F$N$%`h)x4`D-0sUO(lKik11In)NHKn&UJ5BLf%>H8MY#oR_#tho#
zDFCY23cer!<@n|4si~4XeJIWIXq7Rnk|v{Bgzl3bOx`Q9!16u57s2m_Du?MB&qVCp
zHn3;U_N9!B!8x1F5GY*>m92KD?DkUEp%AVb!Is~mInHuLW{$SkH<dlL6BklQXf^!P
zoOTZ%#ObL!&7p`gL0^mzCmJ(~gW~B&!@6=Bel0vxK7D!xKC&5}%0kklSitl=KQcRx
zR{ZS4=Z;56M;6N`1<@zZpDCYH8zS{WzPMP<K*}f2pT`M6oW^E&NqW^>kVYn!jI0cK
zxt12Q_&{0d2A=Xo?2bi>HqQrElTM3pnl$C8L)+^*t`u9HT2i}k9+lvq&o&ZDbOcI-
z@GW_T<1C%T)DaeW`fT55p(xn`DyQ<PZ1Y~ni3%Twyyuao3pDrv!b)&kyceRgqUy;$
zRE-L%I!P)d3w;v^=c`+wJ3{D6Tt{fH&Cz_J51I!Bnsm5pSx9kQ8)sBa1d|;ao+-Fg
zvTU9{rzGZNkkeuZAZz$-<WP|x=)eP~l^pSu%j(s>#Pb2aTO)rgsr&){2(=OdBpWt5
zg)m6As<xFc`Oa@`v9s5`^Belkxxe4c7G9R%-xY{D-^U;2NBM~IB$9j1EBeR1r!!vB
zV~+CgXhBiB=%v`s>-gMdzbUtVgKu%oc~jqKNIEmcAoZH>_xJw$z0+&Ilt*&E9rCC@
zI6u;g-N-hmJY-}FiC1Xy2BN9+Z9?~`)9risA?<8<*|t{esHY`-Qau``S7!xv-RcZQ
zskO>=Z!~1G9QG{@p-_c6b|Y~TpqXX#?6-vZ;NA|*rF?UKIQI!&qf~?<7*svsW8@W(
z$fKfHJW=qWG<Rq3sVKsR_{ehXy23u>{$obu*NKaiMZ71*p2dkljOJj3+XKW^S#dc6
z?p~x<k?v+5y)M%c&DtgVDlE@!@1STJE|#-{T2Gg+ifZBjsH#erq>V6?L()YwlQ=<~
z(@cW>BD1kucfA*TQ!V1$MC((fbB3SZKFS(`XFCuqhTIB!d+WoYQ^gwnl2T2Z^h2L>
z4)gk-`!=+5ANwPFkNrug0{iohsHIdEXE+h+k3vUCDE*QJf>%{Mf<K<}2<|-Bn4KPL
zQYr0pTJG&GrdY7qUfxwd1nBr*U&Q}2N^{=7;-gaoV*Tom>P&)SW5&u<8fSN#y*o?f
zr9NrYnN?pg>}6T@r;1nQRC(K{`!aG3<~NYdZ@b=8?XhK2%<z@EC_y=&1+I$ksq1kb
z)W-PDJr=FNO<L}ma>I<^nxZr1z(Q`{AZKp5X@c*(sIPLU$0{m#_WYLX)-!!|tKxhU
z`#9giPtWu-&Fh}d77><q3*z~Abo`F>qbN-@3Ry@=Ob}m+nSu5JjxT-7srg2RNH0rq
z`sD>*$JBnVOl90@?(~uXFXm3esl<BVUHC)+LxJYF|7`cTzjF1s@7I;A2LavRK}B&e
zC*|$klk%(Gkdmh(y)s^BFV@uaupa7K0!YzNH?6T1%85HnvH11c*}bPj(IO`0IBEWs
zW89U0eF2z@YEH$O!J~66ehRkg9K{GSsbln&Qaoptp8iEY*NIQOa3<tEJ;s+3rwm{|
zw*>)}Q&c$BbF*9E{GX(KL~iDM7?0cJG%YkoGxG(MXDv#+lBO!ssdr|bX{Fg2-A73W
zGE-4(cE;#coLlvG&!3mqqDVkIUp-E!;dyzhu)+7IAJzLbAkuU*%z!36UyUg^T8Z|9
zV~<dUlIMf_Ar+A;reQLT^Sdi=$pHB=aS^n;kXY@RPo~%F`V~l3j9qmQJT7tV7bH~J
zdrS{2x8-Flvld2yS&w#7$r@3Uw4`_*N}*@%4%0W>q<D2($E0E=?L=6j>-3veuV}x1
z38{)YY4z$<wv!07w<FLN5`j`G5P2?0-*UhFscd?$@VL!j-vfvLPp~)oQqwosKP=c@
zl#`xEE4HV>{s960eg^y1-Qyl!J?_Q0P3Mzhmeu{ePq6>W?n!xmH>4C@f2Y*Cf_)El
z{h)&Vv>q#^W%ds5IgVZ7{RXgTH{ty?%<CQAQ?H-``k^i)yF>Wxh`=XOF~jBTE`+NE
zb{@i=2_oiQ`kADkU(?T2_B4&B4d<ZvJtTgw5x<AU?~M4pmVPOp*d)|$u2fnQqAE;%
zDM;@}iYb^N;P=2m67UCl5(J2+-4Y-!#L`?LCxA~PkfJ3kOGi*HPCL2Or?SdQ@ad#%
z?gE)zmwvcm%a^UNbE=bWURA^9YdE0J3(ZE~ot16Vzdx^Q&wVoDijaoZw9Vae){Do^
z@M{q7z0ehRTLyZmn&@tly?CVvS6*&%VGkcuFjfPvG0`}{C+cxPoqk5v*SBAd$o@V)
zk|t+pY@DV}K+&nV9Ef6K1ML!@Gu;?naO#qR5BYJ_j-zCWO`_SLX`vZSTr%JV$2=>H
z%vKw12;+h8M?e+D*0qEWETGdtqCkb?6#$H-uW{hw?Euw5saKG`q89opKe&tqaGz&s
z!_V;p)H{j(F=K7n4_auLm_RCIx6wr<+Rj2ObySX6u0@S#p|NU5%n&zAAOkz46Vev1
zB|6u&{U9=2>#_w#?&V|5bhFnIpSIR)@DCS@aj1YwywtF2aTDuJP`d_o!(d&@Y++=S
zk~52lA^dC)+%hZ=#coQ0`Dh$1&wq4YRF97pKN?mL_<_Mpui-+)pa7r^g=Ib|d9JNe
z600P9w&P3363&r9<oy$J-5dus`&jWq$7DW5G(hV@1{S%ZukL_rdd=8uTHrVp@Asm|
zl^+VMx{sPo!=lb0Y(s66OMD21w~Qh=IAS-HDxe5HL0tewKE(H;9*+w^bOjojGJ>&U
zz{VYW;F)MFSmKjnM5sAQKObs`R^0Fn^#1{B3hgvLI%M^kLQCqAaoG-Sx)H-#M#B!N
zkOo3hEQghmz~J_Sm=I?cd7BKaNx_U|+XZpx51pYWY=SOBx^aeVXtf4JK8|MK?DfI`
zpB7gou7!372JLU5(dXW&lYF9WwGHb2fqkMk5;&o;sts)i?qn4FLoABg;o`yqhQn%m
z9W^|!>IY5O(qV!W2msQEtQN!*K^!9(pTF^@TfxtXV7cD0JdNbbpHRzFb%s=S#R{sn
z6)lTB;0hL4)N--eD7rLYMWWYjgwH}JTSjJ_83F0-_XeK|*S*MGGnn0EXy#+k-$Cm2
zNcjE`or00^cnqr!!Kg#xUA`KwUpG_M&ej<&Nb3;1v^Q{Bun@CX(80GMTB=2#bm|5+
zpgn%w_LiXosb4rqjshLB9bk5gbl3GwX^hn5yplj=S~aTy3tk5X%tGtb20vT})QVxD
zamzK({7oKLU_G6-Z$nqzW$Dc}_&&?KOwH1KM0yjLFEkM+Y8$o@q2G}jvV`7D11V@$
z1V;<4X;5qQm4qKGjMCMQ7{+SL*$cIwQRg+lhiOngSDb%Rfc*#kEa33voS~n;(d%FE
z6Z27m6{m1&-WJds0&+t@AyJs4aasUfFz3)_IF#P0DW9);Wb-P~xh%@6kctND5TQ7E
zj#3B%UMDSm531f8q*8<#yCl#O{UPEyO<WW@RM>H-@Z(Sp=unOr^8>n;3AzuUOHKws
z@IsD|Gf{Ao!TJbUfEa~VwrXg#Er4FrF_=M*7QjZ;u41Kwh4m2^f+&HNAMRW${er*@
zS6BCAa%6~tXp(ZF5)eKj?(jY7_)Cz@Gs*ohTU-Xh0YLsCG*xbxdA!2#cs4AQZQS<i
WZrmW{Zjyf*Ymie;{g*69RsRd(G+adh

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/base_pipeline/index.doctree b/.doctrees/autoapi/lmflow/pipeline/base_pipeline/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..46a585542b3cc1b0e508bc35b24bdc679086b2a4
GIT binary patch
literal 8011
zcmcgx?TaKw8Q;C#o!OmzzvMOvmrLe|$!*N-OmZ5Kbva{BL^y8^A^4WWzMk%ysh;iW
z?sQk*?hr0SA>kHFh^F%o2#O#=6!JkuV!&?(@uT>`DE<cue)jiNb#?d5W@qon>B7uT
zRX_FA^ZwSa&%N`^!&B-{t|k%pf`gXrx<Tl&B$M-Q=%fP{i0tj`@$Y6gvrV}W+xtnF
z#tzHm6d)Y#`EJaD?Bz_Z5V-LCq?MnqC(IFE7!=3(9Ix^E%h@JhkaNNlKC3*E=lns}
z5BFP<7ct)pSgT_v%rdKt4&vPG3+#XhItl!ZOjBHoY|tk%zAhK6f+0R5YdIgga+Yq(
zcGHe68@3R!*GUEVCw^x<HzD5Yj&COkHt~Bdum}3ryqiYlqGLOIG3ydcD(k}T(W8v7
z^0R!2KgehJIr(CnhXdA*(;#4Rdq0f#+F!l)WiW02fxx7#32WP_2<^ygYq7MAAlf5h
zX?uao4q9;n4j;ODD(&zoIj8n2qSaF84}&9q9#VV+pGWbzfY0N^9Rw76il<FE8`)l*
z$wvpFoBHhY&B<DkpVDw!lfigxS7dg|pWs`iICl9%5IS_a03F%A*p7IGXu13>^#7bO
z2Ce^%@!ed_K>Cc9u|{ei%Pi4wnV;hqA+C8ti=v|Lgo6=l=X3Jx`SZ;v4jl&NyUot9
z8Ki?~2rV_wpEndugwgcjp68Oz@kGgmgMq)DL^k==d=#^U#Wz?+wC)-jbD9@H>!S9?
z(w)}KhOA%74OO|zM!7Bp<=noY$ab%5a3!!+WnXK;?xemQTTbX(NyMCr)=o0DSp_fG
zUi#*%Pu+a=(oOSU@%l=6o%H{QuCjgUsqGzduOuA|?09$?z5`2?8+iysglSbCXL7x=
zvcUal;CjA+zpAk|8ed_4#Nwuk8co4%K~{9bLs)Iv%|zINYsYSL-ySwa*mPoM3)ZxQ
z<`<v6(o7RC=$U9YH-1?z-vwC{y*NyxBci9+3QP9yP%ra1vhTTqXIC3!c5guyKY+G>
zXe{d_#a0k}yNZA!eJ6B>M+gO-GjX_oWM3r*3P>Mw<O_xx`K9e>xE*%-+t;gtZ|+PC
zL#-<8<?AmgDQ_ZTE=4SGA$IEk0lY$yymA+=F}ke#o?x-<SMc-S<!-uEr7SCb^@HPm
z6_Hv|D^EqkOxB{|7Q~UUYG5ZlybneG7OCynfGBI=84)d$PuxMq$xsT=v4QOc8UIk$
zu@b`w;cJG-kc(KegUI(BPh_{$<9$00vO@&>;kM2hC+2X3!r`{^p=?*(C+xV#L?%Cd
z`)XE!<=4o)|L=fubsCa?4=%L9Yo$Wpt16Tos}#~4WvPE6OA(%5(XmNdqyz5)@E5r7
zm&QwSpDn%SnY$C0HsmKqREp$3RgoNTr$!{N<s$j_9f)MIpnndz{p=*+{uDq6h{t1+
zvlMkR)L>A>_W^xt9Hl%{Nh3`^40<}c+v7g_CQg5E{B#wFe}`Li&FcU(`5AIS4lC&^
zYgXG7eATVI_9H9}V96`gBFj%GF;=gs%CLDQ3<TnIV(OqH`-O`8Pft%x-M*Glb@VS4
zClpEd!K_<jW_^3*Y>m<>^0j=r8e*H9IyHO2<ofpUJBu|VWp0IsywaVLv#0=$mP?d^
zR_KY9c<35L8jmzaqsc)xP*ZYM9k7ES6!ut`B`>|W^N2tdv)#O)_`GoWVspn_fRYV$
z=616HX<oQ|8J&P?%r@kl4!EWAtU?t98wIf%S0$s#>87y<{3?-nV@O8*Tu$$2<6hA3
z(UG%#y1r`H#+Xg_lKVpcWextV;v!`@55RDc-?gmJ*-}epx>?ls4^2)ahLw}RN<LpN
zuK&13WK%9e!&@lu6(Mq&=u+}l>yhMb_<C&uUki$_F{b)2(h2I8H{_`4lw|(s1TyD|
zOiklF-Elm5ZyiUTyr)u4-|*PJgQ-F$SItk$4jdlF6>XeUVp7FnDuBl<T|*?3+hcU=
zIOMsR5=v3(@V_AL^?y80x>i{kLi^7FLPbwZ0w9r0Hb^cga4^^yEwyAVuW9&4#)EiG
z&UaJaSF+RhH!Kl47P?)EnjfjXH&IEc-4cvw`v%dbyVM8Z1TWB1l>mh`J39Oxk>4gS
z^Zor4^OEgmQN`LFr-dftdg1N$9HQc?f5`9R=W>7ljS0MR{o@hw-y4YjpXji*y?TCG
z1@)iDP>)7|P5#G|i`o8%z%eN|R+L<=QTu<zxyfbfVWmvvf8!tVziU_ehb&m3bcte8
zg_g6>3(0Cya3Xj!Z6Np$asePJm0G&?NvpDk7jPDPXe%AoGBN8eV=ZIHzj7XoeT*ux
zTBcOFe|VJpuNjNEdn1<q)TybdNmCY6HDRPuC?eP8r4y9cf1j8x{m}4U#tKSHo7#Ny
z3_23cWi9WO`uBlKSw{>}?Na}KTB;aZT?gTo9^%0!gV6k3G22_ZKR%R8`9rf_2Sgqi
zk^*Qo2+f{>BTRH(!i1o4!y(0-=Z#|X@+<fMl1I6!gIsY1l6F2zG>@9*COn!@Vxpca
z(Qfh`GtJ#HQ_igpB3H3OtTp$XFeN^6>x`~DmB(T{dc3g~17Utq-Ya4|NPJaSS(x1P
zFp<*(;W>eF%g?S($%cVpE|w|T`m|he(wN50Rxw;&GCO+6Ea!8PW4^vl<7hQ7@5U;a
zyXfP;Zu==Fv8G0mwJ;S?DzaU^iXOPfFgUga(k2!zrbd`04I(TS<1iE;70K=(lhbW<
z;F<&r_wE6pkuTHO$7-4K5Dl#9DRR-f@^lvnP)oAP_AR+gBYey}t%GhD575Z3B$yOO
znEa_xX#!Md^$znH4Nntm-xJ(YH4WInP5FU*EV(O}Vitw5uzYrdL6E0pgWG<woRjP9
zz=Qly)TW66nv-)b>!dy4TcI%mjcS0G*#qn>KoJfjW{5s7fV3<YpxPkAmJ6xqZhdY`
zW$y<^A6hJq!`R}O#`z2*?;4Sc#4cC0%qv*t6c!9D#-5C!=v(B?D;jS}gE{@+=!3Dt
za*cR^)g>Gv^3^PO*|&pUYWEl<PSw*XylRp6#LNj}*K!$+t#A#E?ssJkuGdA80*QF8
zF$EMME7Wyb<(52!dI+T_CfQbDj1a8a@X*u6fsICX7aazHC^IMRms?T7Qa7|Pvj)@}
z#ylZj)KpI@D{+K%!%IAEtgDvmB{Vn%BRQ3I<0NpnB`}j966sC`R+EESUC#$|ct^28
zzi<gA<J9dyH-ycF$`|oYs#so~mh0qcupG449M(U;0P5P>1-TNj$fBtf{1fwhP>!iz
z1ZxNF<P`FLE<}`Ux7*lGTG3GOFldEw559DUAQb_Cv;-SKJp$^O$Ntu%Pkt8iT#38R
z7<$?%(LZL;!)<Lc**zAwJSMv84upa>7Xu%cEnq4DFH*DEd3g>tIj{=nEEvc;uS2<(
z3<F^wSlsJzA0OE7EX@`&xs+hIEv#2kcy$6n-J{leY5AdJ=b}C5aJ+wUpm^<Xkg^gX
zPVZsDe5l^+sn>OR7&wIWgpL<<VFP(wI8A{fNjqa;-hgb^4{hy?^rmeugUJj!%!LOJ
zK>=r&=pD*44xsidhJm7Q;T<GJTt)SI{uU9o>UVWCJCrjlxIu61<h+h1a9>y=@}vjf
z5E#+ZoKV@#I*5X02?SbLjZ5$6Zf5fI$S#%MY8%+P#e(Ld^!g9rb=uUwH~XJ~kN2OX
zuQn29{~~=oLCZ~irE-B_Rc@s>hzj~pK^`h7WZLtzPY5t|*})jRmNzvf=9~9A<#ceT
z$g?J?;w4{#WSrfi#QhV(rHj&8Gs{UkiZXs(k)>u4s-5XmgPh9cszoUtmocf;2@ibh
z9SR%<P7AZTL)NZ#VS$4VUImSS-aEEgKrf7dEtp#srqd(%kpp6p!K2TN<I0~P2_vjI
zVM;*@84m+EAL@sqkzMOo^8L4=hc~k4;i71~;Ox+K0;|*)7SU~Fd$HJ7hPCJgj-R?@
S$UTZh4VMCN9;<X)*7`5&Se(`X

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/base_tuner/index.doctree b/.doctrees/autoapi/lmflow/pipeline/base_tuner/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..cbdd7b530a952c69922d4bcb693057986126e1b9
GIT binary patch
literal 11914
zcmcgyTZ|-C8Q!^c&voaru!A@|(8CJbyY6(ua*3IpEaE2murd<dD9WX_y1Tkh)l^qC
zb=jE-E`rG-6ORc<86ONhs4*xih~Ys90iNaAs1Ks?MIY4!@0agCRi`d}sht^-O}3}%
z)PMf-pZ|XT^FLPj`WGt)#6Ldkc9?E$Rn$hqvKyKk@bQLS_uHD~1<wS>9}g}CWj^7k
zo38CUbuHir08wYU*>E%~crf5I1n%jkTZ!h6xmw-RZ7ZJ63arS+9t_HCf)_m9Gquzw
z{z0?dH0{kwNAGB+ZfTX8>S~JTTbdKlEKC|buWb=Exv-amkk(>H1!Z2q=J|vYZ-tHU
zVzh&GK1%BZ>wZn;ld9)Add>H+bK2>M-f(GGV|7z?T@2#6V5#l!X}sZgl9!rV-*B`h
z!T5a4Q(N>YV6$wIO|v6xge~#=s?2U{RmZomo$99TY*asa_s6ka`3D5L)sV2N>U*}@
z(W@ats*)kqgdtVkYG_*($Hn%pJ2&jt*a2P;V=0b?%x2ePJ8T*JJBt4|;{P%HKTcZ%
zcN|@8sLV$@s_q2*hPK`CP3?3kM;LrL1X{^~Va0VXP<{4cwvupRom~fBLyi*=joNb5
z4oi{CZ~7PnexBsLT2f1{UlXh^+07DtxJRl*s0<6c29NXsWPCIfMo5M_kTFE{8Meeu
zg0<t4dO`PC-EL<Lf))7U@^Y!;*mVtB(kRuoOP1g6Y(t!-<z-1~E=))t*15;64JL|D
zY_-i*x1*BjjdvW))trl3K(ww)8Y7ygK<iX!u!+UZ$pJogHqvNeR13*_!l?T08AVEd
zAp{o;n-#h^=czT{R2`*mn~K}f>M1#2Ni9bOJfC~uzDI7k^vJ18@~_zI!^v`7|9{!k
z>Zw~+*T@K6zui`y?K3bkC@x=!TqK-OCG|Pr^QkulThBVS)06lY>-Oa2T?2Kyw8`{3
zE9q_tZjaDZq~|MiV~w~t!HZtYvHeaDgUQK4U(G#oe;iXb^@hiSbCaZ8Pl4fIg-oB8
z`g0}mr4W2RjeuN!&2DV>5DGd+9DB28Tto?SQGKS|DXHP7S3BFQcFkD55OcV+HrO#$
zVyH709uUke!*xz~G^?RoEoBRCJVgj!-9L6IxiM4sG)Fa4XxZzL=}pAc1c~vzHQ4yz
zrUiw3xU(JbVrP2=I}fz9S0|Rg0hzsqpf=FM@gnvK7ZvdGZevY=*iUoXs%{1BZ9axK
zZ@U9$G(t4+DZFc29aFFCUht&&ys0`?umfkmy&49L{J0ITw!JD`ey}bU^His$c>({>
zZp10P!%q=u-&e12ZV0S?1?I89iireYNlQ>0C<PJ~#hKqB&JdoLX<DAxXu?YX{2><n
zv9y559utc=xp)7v!1p#O+3LILtqxW=xz$Ibt^RorTg_+S3*g7|S7O?C0R$&D7!w~Q
zPZ=PS0*k)_=qI~T^1~?~Uovg06?$Z~+xVWq>@RiCP6M%rm_o}u20$4(!50LWguW#0
zRAt(h9*QzJ!ZHagQIZ+Q-<;q-d$-8%N@s1$gKu?ZhSD=0i`ck*=)i&9OBwlqaVnl6
zNV*2<TIo{P-6f?(7fu<$lviU%St^HFp>E}*sK<BfG%{&witZ>=J-|nC66#G}C}zyi
z7sb_)?0deGJ;}t>mQTQkxhG2}PHey%w%p@!$e9(>Rm-wHwd;_?Pu;(E)I*Z7S~?~+
zd+f}~(wbZj=@fFn)lv*nI(Fs^GJ0{YTHuAygC^orVqwIaib)S~r6Lm_DoKsNQ-VbA
zLXx#~KC~Hh+kw%dDT@XzpVn-t&hiuzS%dLl2>!WvA)z|gLUj<HLtf!vO6Mwh7^U1h
zCo;%=<iaF7d@NqL-|?W#ry%JSq}_r9evl{<oEGny2rPH>XwHr%#E!bRlS&725(MLG
zS0HtqNDb*6r*#IC_*4#whXjdqP-|;QU`!on%@pfryEHpi>`=MldFeR?F(-zc7b}2R
z#qJ<O@@yv!Jaj@z5KoDmUdr7(AF_Kj@kik<zmGpcorLhna+PiYjFOGYWy6<b<+rY|
zvemot8|jr3?>@)|o)5vlBM3FVhd<J5@^R@wLhczaraum!h<Qbi3CX`AgrWe^PpOU9
z@VUosBX0d1-~5{KW_q1s=)@$0<ZH6t@AKDtyWe^dkHYo#$fJC4er?+~Jzb$9j}j{+
zUZKf5h@Hl_h}=b^*XQu7QYXEOy0%$IMJ(X6@=;N(I?Jf*l}<&JJj;Ccs1l~bZcc#*
z#U{+q8-Wpk$T(kTzaY9thr8$wrJHcWxJPgr1s@c$py&agB%gpp9TmCkd4U(B+?>55
zV+Skp@uqK@!aBnBCl$}G6Avk4ct?yqffIokO~8n@&k|3=+R3%x?Rk0?X>IJs>tX7V
zvh|R8WtQZyrzwKEtEKq()z{I>!k9SJ$@<WR&_Wmr8xfL}LEJ=~OBn=vHMX$Z_PrK+
zQz9{LCj3;ZT;L~mPjZUDQ(XjR18tf8?8-+!qXHax31~{2(uX=@4fFEPbvvnZ9s4zV
zhy6yV0{iWrc1vk1PHG~+FG5F1D0wLY!OMCa!5@uy1ZSRT%-#+#k&N~>De-gXBMR73
zKfjtz_S3XqpQZg%QfuD35~5K9VR=<VdG<ibFlJ*UiBtQ{(v5ZUP<JVHrqwqTy-6i|
zs$G?jm&$!@K$P^8LQDzy1+yE-+3(m5c%^5dAe+TY^^n3bJ`MULuOnAu+?N{T%RMN|
zaEh9FveZyLI3h1f+l{Bwv*vqd$-vm>OQJiMb9JX|Jc~7q=kU`vsf;qV$73YIif%=u
z{&t$SnJoQL(vZn03IarY$wv9q2e`55EusIL@zVWt-^h#c-@w#7$0Rf7C=Yr@fM@fd
zu}C6)=+1m1_b)*c+JCZt+F#rc?MX!Lm+~3Iz?n`&`bf0^kgV<wEGK1ZW_!vp_+oru
z?PFc3a6th+N<I==dy<bYfM!v)C>slSG_Ju<MwyL8j3A9TLSHGrF_!4*m-KU;_{393
zA>PNFz8n}O0P>|A!pq00CYVl0ZiUx-l<<p$#JB;EyW}k)TBex^0_k3|GA>8)IrM4W
zsa8}zHU3Dnk0Bis6;$I-^eT>O>37Rs32*cebuC*y&dA}Fur$8Ir=ySZeI5`|j^)Nc
zvz9H#BphL)_2Az;lv%=*_+CguBuG(whyvXH>NjCzxQ+b%-h;vMT?GTe{XhHvs|AHw
zmnmKTLut2qds3KDYp&<0b#E`VnQ=KOmGB)Ni5MPn+F<Dg2b5!$x+Z0_M8gA$o^xGP
zkD>MQk}c_jqk7DIypjkscOcMYg5YQ<fH36^UlMnFOFX$>sMuww>qEi&73w~bOMh;t
zJ1f|nl}kD=lx&Vd-JJsby$p5A{%OB|KeT7vVE0ygLR}xJzCWRECW#nAb^AxRguI^U
zHVy9VC%Rq6ynfLwG#VMu=Cb7V#;`jOf@UJw!Wj1#!z6;e$1vkv#46)i+RI<)={b7(
zYy9*#Jna~N7r*}yzyB1!{}R9d7Qg?YUka!LV$EPpN=0E=SIFH#R144M9%2J8#2ffQ
zyaDmF-wlW>kI0258{o4DDrl#P<1tEvQ5TN<RFYvCpK`8saEFqhEweS*0ktA)AS*Q-
zR8~l(9b_GHk-VO&p7ln;%7}3EXek=(yy6Zjc9LI>C}_E+xWiD;2h&1lO4vbF@Nl(Z
z2Img&NeQDh@ajX1Y>3a)eTVw9l(-vfT8@NWS$sUoOwfupPyJh>x2)-i@=*iV-=9%U
zAKgo`D9wwu?{$1HSZA}iKyGPZg(xcH1xEqR_0)C;SIbk5ZF?XU<+ZJV4^?q@9g;wW
z!VLgSMz6kO;_V<+zNo{CzM}f|5<k=g0^HkZso^L2LF#rx?~k(8w4FAtz-N#k=pA&x
zh?XH2s20N!&D5xg$W=CVk16670o%Yz(S)cyXPr+wTE})g#ndipVB|2LWU3jz7Wllj
zrGtOCK=NG)RN#e%R`XlHH$xZj)YSmIn%c(51f?t*ZHDO6EpSWGEEG2>|Kk(B*IfC?
zil`SJ$$nHc$FUuSsaC^;f<XbWH<S|jY{>HzmV&?%@TIOVT_}t{!0`A?h&Q1@&OVa;
z&^4LQ(H@|6E(429(GzvhRIQe;wlr{@inYV&O5|5LTHSUUN<*U#4=h9NZtJ`Vqidop
z3XbRvsR}52&d?pNk__=-RA6!0gN`mGRz@&Z4CuH)bu1Ov#_N1mj0iO+=_f-SSMwXT
zg5E7a&7ncUMo(jUrqGhQr(D!soo+yYOKIpX6;QB|h{`c(Brv#XY{o^x9^M*4Ya%eE
zshijw`uY~=35%dZjBXYo8^YFr$fwZ08NZ$z;`5?3Zfdy9X3+jNS~u>VJH}@^T1TON
z4%jF9+CUQ;WIE7x&`w6dKfpY%<E~b#7<Mb2ZPah9itV&uONR+k5CBNU)7lV^2XTyJ
zeC5WQZv#JPoMycXJ&h#jZ<6Tg1Wzivp*a;@^O|A}xPmJ4+9np8z#TlSNOUWW^CjqH
zTZxS`-avHAxx?q&ZOc=)6sEVBiT}{wQM$<w__T{YFHd>chgG{^)K%g=d{s=lu12g~
zsxw@O)gk!M-oRzSLd@Pk57v%of)$;uksH{7_SkjZYC;Dhzi@~g1-cu%!1Olhu4${G
zF;Z*lMgWy**R%#KcpDTj4ec;H{BRvm8;XWj98*ENE_qx*^?cND3tcsvp*P#%Bbs%Q
zT3q>f=uKe0&_o<3b=Zc7ZZ&GM5qdKZqM%tG94)k_K@F{!0)8knO2vLL3~bH#h1!Xz
zUl!nFG#EaYjn^r-zDhq;97&9m^z#b6zKkEAPY|p)f>O(^fZh?1I|2$x!ZM8$0`xen
zp;4?D-JwXIFZ&ylqR3htVP!Z)W2}oX96U+sLv<9m!qW4o?3_RvMU=4%f-KPmATChE
z#gIYO41=m224#K*WrjXKnAS3}?}O-j3WFec5sr{Ek#He{(<5X7Via20uAxmb1A2AW
zU<^Hx0h^Gk3QUJG_z@QzFMyRF?#2~;L12cftJ^*~GDJbNz?e`82p<u5>})jt3}kaT
xxEE%N>mN7($lrygh6~0XFEcz94+~|R(yhAbH%Pf#<ew%R<dh@-71m-Z{{<&)pKAaB

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/dpo_aligner/index.doctree b/.doctrees/autoapi/lmflow/pipeline/dpo_aligner/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..e97cf3a39bd0e5260be1ef22e746a6a52119fea1
GIT binary patch
literal 29332
zcmc&-4U8ODefQbEk3HLWXU8!ij>$M*-Xp#{Y|@4vX9wqt4=3l42GV@MdTw{_X2*Lw
z+nHJ0*CZwVAmr!+DzcFXDFP&l1Zh<jkcwy_4N#Ckswz>bQq`g=YLN;RAEFhlN~Qk(
z?_=J3v$J!1dyZqtdoyp|d;j<U{=e@hho4=Kw();#GH&_x=EYgJR%?beFYd6hT3Btb
zdd;NsOlSIcI!|=UtQ5Hy;;<c6y$;)k996&Gs6}40^PUcypzKM#5zor<qp?>_>R~f2
z?hpGT{^)x;WxvFRlX}wdj8E)fV|BR^UYKpwTVA8y^k!?Vu;MoAD@`xzP%)I+`G9wk
znp2$l{T<QTk(P_LB^`e+D^*ft{2gpW;;_he5~R*zd&y<vZjwaxrFH_`3D+I<T1<G2
zRvT^{BN5NTO?Op1jn&#M{j%g%&qdxc<!Q6g#9g6J9e>h4=<oKg@pt%#*h6!Exa!SC
z?I!S<yAVd_=I*)oZXm4wpn>t6XxN<FPC~a;pA#gRQzV%)NHSM%*1U_eQ4IWEcY1q!
z$=}9?d742~L2v(B;Nu?#1+T~d8}a`L{!bHbAWu~17?s)1mRpZH?E2NP)^2!jc8d7H
zt`oUtiZb~li%G|A`#1YDnh=Zr>p*UQmyjqe5>qOHjou*%&Y6Bdze~QNu2X`jqAs^5
zegv9$a8hLVkAhNToKjPXyVP#DQKcF-DsjuJZh~k#d44wa=x;oB?1{&YKcW712;3PB
zjs>q6%jb{XGQU7V61P`Z-DvG3WB^oU2c!~1OJ<GF9k$nas{nUf^`^%0@6P+qoD$kL
zKCbj_IGa@pREdowD^b{Pb&-v9dlW?7Bu&Rq<U+lc_?^?^M8+n_cn+Kjl~imb$wmba
zn-!3ySqf`wT@6J$cSPZZu5@mkNG)=UlweZamB?-RMnQIRzO^<VE(P;vH4&YK0@aw+
z`A(jFlvBG5{oUR2nzed!rE(F*!~}3+O8^u%M;rCTi`<4$v;PUHWCr^?TF*Y+ThCxG
zcq`fV)>?;+wAN-ofR1OH7vjSw!N})fUVG{&8v(AceH}L4x7F*w5mBL4x8Cgdf5Jxb
zmaMhlly*=f*e<+#7h8>bwVrgI=ASRPQM0qoN???*TpRchHin$s@^=<Fw8V{8yrjc!
z>W{757(55Ne|e3-=^@bjSt!l{e?+tKS(A-kPgW4Wp79U)-+-B<dj61#74eP=d>9%3
z002Ld(Nq~SC<DI%(1ZT>`UF(m{B^YEsf}v-ePn`h>0J}sNfxXFuL=a~$i1Acl3i&S
zRHqR(SA?Z>v+`d<>9g6=W+qm;xZ0SHTQ0>^W39-Gz39Bh#pk^!t6bNk+`#K5>#)RC
zUfoQR4xZXSj#^|8k~+Ww?_0XN-Q;0gEX>9MW~8Jd73>?@w#}5P-U?1_fD;ZT(-OQH
zQ;?baGBVREPVpc1AE9dCT#ag8N!2=jJkP)S71dz2L@pBsp+Z5ESF6<A#EreA{PI&D
zb>wqAd#6a89v{CabZaWzaho-#8aB^+QR37SCkdUz_nf9z_2L-L+*uBzRX3Rx^!F4<
z=>Y@#y*!YMR6aI12BvJ$&8j-yZ#T^Up#5I%Z6Vl7>fX93TI@%=^TVGykLI%B(P`;u
zKdxHF?!Bad1;ak2Y7ZVHiKmXs0k*HZq1w#`52dP+7txme7dCdX!Mivod#hpUXZ<gm
z-ZQA8T>lQ#|C;}z|JRHDOKi5t59VM?`uhs|!yKh0KNh@`lLlNkkSiC{1}7-1Y1XvC
zRaRE@od_Q|%b-mqxW5b*W#J-oPsQLegVUl-J-sHoWpI;mV8FuKy02%*&|9n&8KyNE
zp4v1Sim42r9=wf{tGF3(>(UnWt}pA^gW0@_|2NN#2d~e|YV@XqVPptKC>NLv=WY0^
zK+p$$G~G5<T25W9cH&lIfg30PS>VPe@8!LC=Z>(Mpf{5`^R9Hw^lzi(rtB9}n_T)g
z(-PdJUxg;kD(kmVhs_##7>UvAL7$W@bX$({v{SA(9TBw`4N6RRHf@xw5efqxIY_aM
z?Zik}_t*gS)h6gmCC2cSo3tZqB<mwNNKi~riNY|Mcj6>E%42~_tscqO*lpI6wMy0Z
zs^{jNr7&zbC!D+8M(iDRsxFd6-ZL#4P;(oPppN$-$70w-hGu)U(u%_BiQ~6SJ2NL$
zoRP*K((R#t$=ggLd^9|zM)mw*c1@8n7jEsq2AQ-R{VbJ=TyMs9foL-bba^A#9%>Ws
zXz_Qe18Heg@Dr-{VDO`~dM<QVL+G}#62~lyR-V`G%OUUBUZe_THby9j_Ky+djha40
z_4R1_KHhX1S6KDXJ#W-8qGzyX<uXZ5DkMU%)ApEcnXucE7U4W3U8rO^P)bWEDKb*D
z$Q&-DHydhpvKBeGR*>`{{%{G{&&fz_XpvQIyP2V9HZ-rAfF}+5_y(X`-Jsu)0iBux
zj3iPw&fwmVlLB=Npm%Hl^g=h#Qzp>B!JD$QYzM4?0Fy?x()u0{-!!)GE(C1%OZ5<o
z2tV~G%di+2vBpJsBlyS*4gJ-Z>s5QW#edNpQXvod95mv7v?m)b1xJxs%x~>PvtgNS
zGG0sQ9t^JJP=c%RCnee*E}6;#bMlpKvW}@7iiJ8s6ZCO{z(@G=j9nor7e%X%nXT#}
zq#rDXasr>b2qvXeLb=_7k}s7XD~7U|2StjdgfggDe!dWLFR=`2D5m=^!7{zl$e<dl
z8%!oYi)uC{lLGlmAd^@34g>N8bhnvZ38|f^1@&(WNn9wX-{Axa`VrKBDuz-ds1iy)
zg8IW^C`E!Qp;&^tlj1EwzitC6K59DtvJm#bqAI{_+J62wXN?`H_ruG6sC-~Yp=sNg
z%81MMrd~O0Vd^0XBg;FUP*OW|!tjna6#`~MVYhd@fr=aqHX%d5QV1mI13OHD)th|&
z*l_s6JZM*F$u})McNIdVV0hVWH<HSEx6$@&G}#q#yVVM#q!MGwFZs=ONHf*z@T9>V
zCia)W4qan(IS^YokMoX(>~C<~GOn_e7EqWQDH{7qA@+rG`v{7*m-@0ntHn@?Y><S~
zk9Is(45dhZBoxaAjcYm!0{h*jh`TdQ%?}l}U|>r$nA2Pr{ON~spDt{7q2@^_wkLXA
z$=uN6h7tL*VkkvYunAngS`4K~3IvoKHt(cfy@APh3t{xhXE`%!Mn?k9rd^I`w{2zm
zQQdw-;|GPP7K(;csvA*58so#O$nL)7U;eWg$}xQIil(K5*@S`EzH?yrDWTkH;v%>u
z(aDO)R57gkvKVdbOGFcFE}Np(cRy>FF*+qITc($#zAILZr3u>9O&8-=)M304Md3L5
zk&9c4p%lr5gko`Q`|(>8PV}g;xvLn?0NfJb2jW&gh1hFX5tP&`6zpKaQ;t;O6+=}#
z0J(yK;vVB$V$d=2FiP5t&FF86`q!8<wZ|!Q3}SaiP0N)+TAC7T&3bLrbL<AWX**fv
zf@g>3W_xT6atV(-3PM%I!c1Dzb-wOhsA6?yhfS)-ircLER#?5!$h+wSZm5WH^|)xM
zA=W*jQ(dohf?@F4m`J49^LMii7vmsWtb&&@V9Vz>u%;5@a~(ELqo-J?gXN+2f+02{
z*XH;y^3J?J!p4@{jRqf86u`$TNmwNrq$!OTdD>H04#ZQ%fb9O9(f4UuPZ9h(jT|2m
zqsE~159B~HA3hXPidlJ@Eit525VP#3%UK4TTa$)+uvtAV?w{#f+-Zc37W2@W$3wvO
z)3PM6EwNoojj(#I!ut>gvg{upz(NG6mQFoZs}#H%EmCluTAi;9p*0E@>WR;nEi8u(
zEMmp<4b5w`G4L9B^W%?=JNS15lM<_~<jA~p4=e?rvp6a~Kw@LBne!p?NGiOV{ppGE
z@%wmHv4a)88c@w~W2}~-2E_BI`MP(uy_m**=bh`Eo1Fy#|0Eu-6H6LRKECZz;i`=M
zrK?c$`^y9I`&#h(ODGAy_eK0k<Ds|Ya^y%JH;&|TLZGU3;SyDNfZg-*lOGe@dilxc
zEP25P(Y~Ays*RaXhO7O~K=eBd`hAPjuebRogYpHcU8oop5hOQE#n2N<Ht?jG`u%|#
zb%TabVv`sssAI|yvM$`P9)ghOzr-*jV&$n|_2;Uo?YH%zRDGxvmgQsK%BuOgjcZnv
z=_jiu-T!Qcbipq)e{KTxevVge{>C+B40=eyY3V*#fV^t{;g_HMBW-o?V%r@g7>wUN
z40-7}SzrbSru9C8rk7AMJ0_jUAB<A%LK>-vz*hU=>`Vy>@Xu@l{g@M|Smyu3E+S;U
z7>c1LWSI<4q-k9`)r>EAm=3>AQFfO~v1xdGAR12FaQPbbRcmhxX2B00KJ35<(K0-E
zyrngoSZ}RiZL)LtuxeFd9wDaybvqgv7c6?VW52f$N}4W`hx6uIkv!&ok~pEm82?Z<
z@}fdzF?WYBO04S423&Fxb?^|{z(x_2@paU}!}Q9hw$1luI3sqLth3Q(s2(R&@{CwE
zzs`2ckLrCda>&psPBTq5LzSlTQB4F7oq_}rh3H%StD5hZ>rEt@gRL|P{spzMV|w|$
zQ|5D-4px{HtbxGNl782H7cVIZ{tfVzKbEm&@Ixbw&FQw45e<8d3KpKna>0mrl$)aX
z2eqbg8@U85NO7CZFzK56KK{tU^$9u<^Ue{D?vay6orQF0p3hp(J82f@$jOsfwTP9A
zWh|PLlfBvozmLS#uwp?!BRe0+6tay>;m8ibE7Ap9nVuj(2ZOZ~9*=+*5DNX&B=aR*
z@$!5MEN>FH9Q+cHga5<N#;aR$+Pxi}E+sO|+@~S?6lRj<(&sj4(*$Y_Ch=3yCb?`S
zr%n6KHqnl;jpS5_lRX7LuFLVy_P|i~D@4&u(wM$jh-u-_<X@s_kz2JzORgsP3tq@Z
zMXYKe^~}K}^AU&`DH@M|7#xr9T`D|wo3<qfmG1Ar#!l1-{bX>2el$2jy<-v0&EPD2
zr#W(!BjUe@EK2QUXxM9ig0<*ilS3*m0Mq;-72T$+V-FG(5<@CM2h9wg#gDDW&mkop
zMF~DmnFvzwyhx|#r#itWR5}HN#_-4jEa-V;Au|ul`6*<xNnv)Rf9-aR;>Gh$x|Tlc
zNSE+k?c|-M%yOUMSz#0)L8Gz?BUW}sY%in71=;ZdA^D$DNCdB^+QriRa=*<r8h13)
zINO``w_v}FI*M$MAr#io<*%ZgUi6Z;;|<Xn#_Pn6H?d%!^B#-#`^S|!AXu)?o##nN
z1F@sX37Z`j%6~{w2>q>rmIRAH?S%FaPFLHF&CtevN+;dMjiBvVmbdw3W;K|O9lt3w
zRB*j<{AT(n9zq4C{|;QDix50Br#6~ZNjL>DBU-^G{B|X+inH+)H2LxvHFl*AVoll_
zB25NI4jt=ecG<ayfkEGPP8c|D>K;cqY3i(q$%-I8AnhMz&e*?Gkl;rp^>XcB(RRi_
zcFa@-ij%fuk{z>BDz0|VEO;+j>t_g$3O#k`>1_Js;%Plt;=fh?TjReT|6S(4EA&f<
z89gX2&zGPFwZPX#QoJ*eUcRAEy-cBOSuKr-D@Q9|AYKctG%{+X{8K3lij`YX%Fm@$
zaizTT@+swhLn$?3^L0|UvP&mF0|W-q$#0=tu1*>hS1L()Gb(u}RPwWub_1)VszC8^
z0G0e_LW8U1_vwkNB!7CfRq{47bRS47-`J;Cnt-;fn0DtOM>qeAm@ahF&a2h)7g8=1
zTeqN|KToUT>iO2or=G7d)YGDHzJ}_Cc4_F3fx{pgIyKruLv6Y%C8g{cC4DQD^rw=7
z11qVjKyh;bB|Soza3w9%6IW9H^lB^VTapN8R;b<L18M0keQK!%Y0GN5CsQ31vTR4T
zw}H}9PcxadqMo8W#?avw6!k<}6<5@EK-_*C0gvk8F}+@@+}8l*{b_wdpX{a;4p1(l
zD8SpR67!dyoh^n^v?@_T>1VLDSq!CURic1`E_rr|8GSHWXKST77o~?;$U4aJ*c^6U
zroPBkhF!IVINz}9mUbD|H-QF&7}k%XTyE%-Q50oaDSyVYz5|wZUb0|d%c?5af@S?2
zp~5Ze=jn-CR{r#ATh`k=teM}EVeL2kF#&DauI{#L>o%(|5z}S(W98LW^&3(y6kBc6
zRo={y!Ha2C+^U}0s#X2_w7yMS)xSg0KCSA@#ZZc@s)W*yRsCr(lp?DtpyXIp`mhD7
zn!~QkWL3>*)e_5mqpBO)WmLZc5)5Kg_mA~3s<uign@ZU;HkFPO`YXwRfo-a)Kq;T?
z7yBCJxtTEGHuWeyahuAYUTvFtSB0-JLT}`^ybAZ7UPQ++^cR6T*w5~VH4MYf_g7#x
z2i%qo?2S2<ii!23Sk!pVYXY}O&Z1iZExRJ8B%iT>9_N137Wppl(Nya#SmaY_Roo&!
z0711TU#IxkwSkM6RI#Tghk2%)`8?wTu##^Yk4IAAx#qiQ-Ti|oS~R_3M~QN3_1AeJ
zyX~cvOF8Y&d;|_gipC=z9FK5tJZw{tA0l_nMCS1OGMV((8G&-1=;vE@-QF(Cz5vn<
zV%eWVx!iC$SL>vKXHIeR%Ux^0AAtEkDFtI-^RFr>H2)*Q66THY|2g5xZT}bOiQ9hu
z^lIDw+bRv5S*HT(L9Bki0~HpaE!+D^t2%JKdq|P(vGVX7upF6GZ9Tp#g+%dq3)cMG
zX;s{sKbW=V+hv1$&M(<!Q|`OTsr54146}1>6E)vj=$3X_i?4$;gIJ57klu)84fb4L
zhM&q#Ty5G3TgJ1yL>`2N_@?C8z!pMPprpt43Nk<}un?mqWg*7siCYN%^lDp(TlpyU
zAeP~1pYb3Euw{#Iy=bS5ljQD}qfX>ug+z_^KY5K@D={p=i_W)v(^layxN(VMyhCYK
z+$ubrwF=i1wOtxG>u?(*tg{!@86WcEy<5|UY&zb1BZ_W1-kalv?09cfawsR-vp%8y
z>AU92nZYqQeW@_mZHiV78r|Q4i=C(uS{WRn>fi|3Myk)yHAjpulN*{cY`W0eeD|YU
z*yVn%0<i{hKOaK5T=ydlhH^XmvQ6v`NFRo~xmvP&V0WV`7}VYTG2z1l($CTpcQ^d$
zQn;JmVnstUa@~wtcvy&mT$OGwdgN<nzL+BLr4#}FG$;c6l*_Ko6>Jhl8TVwQ8yj$1
zi@aV#eWLA?;_yw^JT*3ZAUBT9)O0EP3+i%{8FlT)3|6hpaIwZsC|(Ms@WPZXh&zO@
z;tmG4h1fByso?4DDE?8lAA>&4xWV@m&=Hd>IG$YGjxwBJn?a*<dK(*8dAu54+hC+W
z#3riki0)pgq&KbXRw?4f6&#@`HhJKpnZ0y33%@yLIpP$n;X;9T(+N0wzc?ES+exdP
zbQb+d9LT@o;T9EsoDg2%FtF{@U2Wm0{#|s&CYp+^Zx=gkXbzVFh$hhH^>fHDE??VG
z18;k1H!<DbLSM0V`@?MCGAh6x3eQM>n(d*xZ0KNui*z_Q?i857Zr6GX*N5<3L@}y5
zC{nzJNBfH63hj`s@ZA<b1CV3^dAG?T8GL@4Ps2O!fsotTxbNcRANe}W_IelVpdU7K
zv}2t*T=SONE2wXR)_2lf5vbR5SCLWzBdQ)DL+z_KL9L3{#CCDoO2bO+WO?RIGkgQ~
zHTK7f7e!%I@o_GB0}=)S0B+cu%_arSjar6zEgg0!^CeS7@H40$n-KNtHmK6q*dH>5
z*%aXbsf&G(sKF1$-_vlLD{Xhh1I20k`*z$p#IA_EYKRc7=Fz<w07Ey~EV2=(-7@x>
zgCg~s5(R7}pP(cAD>{j7$Ch>+xr9w#l~fo(Se{Ua=0-Sal`5R%DO}8n`#Px=d+l0S
z!2$lrHH8}|u&Ws(C%Pn;k}8gKs>gLY`wDg8AlI1gUjZVrm7_{Ypm4i^40bT%Oj+#g
zMD{&0TV=W40CKpp=Kwtc2(Ag>yFYP^Dr&2N<-2e>6Idl)PY<!Z{J4gOhr0=UNdGGC
z!nk+p2%BhmtqSzRgMMPG8-U>2krt#KZ6~E*SNTcOis$F%kQ~pp)-Z_MoDHKD=+c#x
zlQ#f4E7(p7_9Xn?0<Sf5;~Q=PJtv~&YNmQh(beCis;4-QME0B)&DOnSnZtl7nDdj>
z27r}tr2@2wUkWqE4nZbYE2(m(2*@)x*V$CO)=b=s6~DgX;~pCy^1G8RX6Ufp{J<=+
z+dT%Mt}qVJLZ`LrO41&x`Z(JH+s`ARZ=f<KA&%QRhpTSZ`4w0E-YRJZ)*(G%wccEY
z3`o1Mj|@fJUdm*vuafLiZ<8w{T|0HI1Cd!>!rcLJvW6D;9<CQzXIEB{>s-ad-6q)j
zRm1IC{PL-%*<N{_7G$-tEUei&+u=3O)AdzsOjr}BFC-B=^KdBv<W3xf$mM1)8U@Kp
zU}&*t7#m>8j<bDMDjlG}lTq7p`a&19&|O%_K1zw=xg7irx!14L&u-e^RHC1k=;v$n
z^Huur75ubWiAwQ(eRQ!FFSpLitn>WqJTF$N9;UBUA?~YKz$pbI@*Kymz0K)9jD^(a
zDhG=@uwppo&eIf7+&YfB#2WUU>J|(HR@74eq-ZAHO2M1SPk-iTRqzvr`4$Vli!7i;
z$N?=<4ru9PKuaS7T4EK@60CsM1O&8ZAfSn?fF`;E8u$!oU^JjXjerI{0*WsKia*<I
zk155Ru6uA%jc~G{wMZdjrSb^|nkl4VJQY&Ex`aEdtgQPpiD~YVm93<p@|yNp^~3K(
zNe5Qs%4}Wo3miX8X*Fz<xr3L*jVTR?9E1xVee@2Q{tP(&{?7d{3pla^1_tsKLlX!f
zU9>b2uf?z;D!FJg_G{kNgn<PFY|W*er_gl*HmgSOdVbeksVa#>&(%j}3`9Z9Gl<5|
zqC{b=G%Vfq4(x&F7Z45Kb^rZmfUmk@4mMrfF?aefN#uK}3F@SPpn^BGV5w{O`I~R2
zGddi8Mu!9c7V|SV;CFEl#}eG(IfsnXo1M0%OB(?JQ6?qImg>-id*7xB_jtdA+pY*V
z`rxVi7vH}4R<hc3C*SEC3>t2dbe<tA@X2u<9@60!*$Hv$3f1p~k;6SYKSN=5`~d_}
zwC9*EWWk{#2rJ-(an1?tnst`Es@sk|d^-|5ap=T2a7Pn)<rRa5b1h7Gnjz0`njs(S
zmmx!nAv@o`c=wt9xZ$yVX>_D?ncI|IX)<?AyXwtqqg^9MxI(cE&S5|RCu1Gb%=Xgk
F{{tEfxG4Yt

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/dpov2_aligner/index.doctree b/.doctrees/autoapi/lmflow/pipeline/dpov2_aligner/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..fe14d923e9b5686f40d7304e241a332ae83fccd4
GIT binary patch
literal 74023
zcmeHw3%DFdb*5zL-m81{viuO(vh30Np%F+UTiBmL$TEs-Ln2Fl;3u{+?mN@>_O#~3
zOphKISZBeugxU}Az=Dlqdp9g0HlG*RU6O2ALxR7B5Xi#;e_<iu#VpBY7rq3HKQB1h
zbL!Du)l+?@y65Uj`F8bvn%mP;Rp-<>=RbAoRCV=TV;^3zVBrG%7q0BKy;}2N$*ESG
zt*YDYhfAuhO0VHIgZ@MP4ZqQUPrn#WcANv<R<BcW`{4p8QSoZ^YR7H%_x8gTSUjlJ
zyCqS7qU%<ITB{jV_r|<&Z(?u1=uL)WK`p4e%1`0idSj;EI#6oY+HSqpbW7ECYyXzA
zQ?Jc7-A*4XLcPT|xCfD;Va4z6bB@Q`4)7QBy;b34IqHzNC>$64*b^>BYW+REX(ya=
zf}m5I?gh|)?C+vlwTpe5sMMWq7jDA!Skr0nt0mQ5JN`25RQ7e;87$KaCjw^{f9iWH
zy*1u4?<{YTw>G?Ov)5|4n>)QG^l0;eR%hSlt9M@ootA%q!R}_x*k-2}w48QrGZ)Kd
zDVEKeST@(1Rrg@2(}f<Mz3Zgjw6`D}WA_x%a;bagK##n2AjNax|M~ELJ^bH*y#oPt
zYOGJiaB<tIb^77?jaIc+ci&u~wGy7k%Wa~IdE<M6zSHww?`?|3vBx_bgzhcnwiO%4
z$?`@xwIfys)BkGime4V@xPuFrx9Fty6d6E$Z0E(jjUclnOlBts&UCNtbjp=hz1(fP
zl?>ur%*ux+-+s%T7rp1sE$@+k`$+DZm|R#thU(vX(Is2Ap~-Z6jfT@Xv>ogLq#Ldl
zum;qWl%M<ID&<=ldar|V{S^GGw`OPE4qjdF&bR=hin|+JLw!eQmidM5)RctfSh59S
zbCcnCFxzSM+5_@LColoh&M539P~||a8hHI(Qz+ji$af#8uqBOVn6WDjyhCjOUD9-`
zddSG=qE738@rN;l&R`wr+ax$1XSU<CJ*8s!#;xr`TU*oq);qXx3)@KGlz6G_cihed
zT?BM5YrD;AtvOph2+l@r`qG?DOBGJkYk}Kw>Pnm5qXKo~oR5BtxF^yP2R?-H7M|2T
z)DOqohc-cn`>yJ8Q1Fj|<{km3n~K(O9C`>Iu^+xZgD>pEAcfT$POaJZo(d=6TX3ii
z1J)vBBU}pK-Gl9Vtx^m6huO~uoKCZU1T=PNEB9tYN0)F3lx1T=e-EP<IGtHH=!fUo
zsFOK{kHL5_-wt8dLXi1~0WGV&@mQ4~R#oYyssnZQkhj+RBXFA7&VyKu(jQj19||4?
zf)6C%DFO=V^oI~#<2`3dh-Uu?aB~lE^8uZkVmf=_J0etFZ}x%15S&sihV;Xdnef(Q
zMvG(oaY@O0S>XCLDcmVQe?_$p9jry~Nehg>M5!*qxc9p%Zv90NmD)qs`!C)z-WT_H
zUkWdw@t4+JG}Yq<$iGhl^6yoV(-sYY^iKy_2~7Uj7L$KBIwt+|fi9(6iGQF)cS69D
zF7$rDE)4YPg~G)cP4poM24&Wu^h}bu@N^~SDb!odSst}G$)TwVweLvQRttrb2OIUR
z-L``hp(X8(+jTqpUFO$+9qP*X@^%)3#Vl*{3rEJ`iMt{y$6ic9TVn1I#i~TOJh4;E
zQUtoA?~$+ft7t21GGZAk`%YAYv9JmZ=i11Ru<2Z?^Y-51-G<$O7+7iM*p7NzE;fb1
zURkmnZp3)@ifgagzhwuXS>pHIZll#X^mb>))qmlB)0LG)Z0YOYi@AbN8NK@NV=&6q
zh74e38+Rue%P`Pg5e+`72!@SrYFX2ewsQ_zO$=WCr_drX8?ZWDrqYiGE&s{LXia8r
zHE#KzWO~s@Jk1Bl@H8=W{YR*mXXttzJ#=Y3MN}R0@yY-t4E+3m%LJpvknEo7S7M(h
zy`0Q8AnQ14{Z~{GizNNW@zO4s^dIHKM~o-^=UFr9%0(_Ab`|94-oWtXKg`+6fVW#B
zWSE`LFyxtQnJ{?sF`|^j<u6c;9G9!j+$jis6$_^hM0N;$Wrz~fg3F1M@*(tM8lfqD
zKCW1OiQA<*-}CXFDtg+w5vyl_zywzRK~$!qBMF|LaN?t4bpuehCr9@t#wuM7qD?WV
zrk6_woc>*`zq(WLcf(Cnfa$;6P+A>d{F|YGe+!lZX9QuT5<~fS!tG>m86~OX$WpmQ
zgq0Z65n<&@Ncur4xUjR;3?SXr<+xuR4&0B!OcxC%72Rqtv2T!)^{1m6EX+I;SSsn!
zOlp{^=dZ`RkQ{?#W_Y2<uoY{~0#EDEx$F(L*&8jHtuQX~*$cwOuzWI@K*r?V3jC$q
zg%u4a=ykN^51C>^56bd<tSGL=;4*?-f$_$|vuQ6RwKJ?(tgHtj(Hn!Ma4GP+3FbJA
zk#ITk#Zq7FyO!3|OxC{-+g;-yiJE7cJqt9uAe>~qOLWaYibeSdBE=E2`9#!?naw50
zCg*Vpwmyu#$I{pfPgQc}`)lrj3M9Sz;YxW~cA6Eh)lqaZEa!|FNtq$Mpv;Z2OE2ue
zIulf?Rr^dR%18!)@jsJfol7KkRsLO2g1G@$!veYV!zq*wBtT(FPEW;#<09wjJrj=c
zl%_WxE}7}o>r8^2{!}?=RmzakM)*I&?%e~KR(3ZohP>U2yshP`2JOEezp~g4bJtAk
z74Apu<%i9FgLhdr=dDe48RN_%Pe(1TI)RhyrN{{T-+_k036KO!3j5#1ugqT5@6FaG
zzKC`toM^V><qCOw6VLh|373f<<@Z%kLUgYi6<XPB$$PRKwuJP$ZV-UG>L0-_{wpCW
z=?z5}^Z7ylz1XoGDG9Ga2B(-arAB*6|9!xo?%#|4VdY+!nwn$|(!XCVJIvP!pE9gi
z>}wV*;#MooOdr#W{=-1se*`{;yT4yTDLxc3Oq|`D_8(Vx_x}-p=RPmB%Q0gshA{Rd
zwB|nr9|U8<@fjHVtisqa^=a5?vrmh;N&?(}87gs~=Km2~`d@|*y_;WwJK&wZihsua
zKgL2x$^RO^kJtC~{jbaW33qii0Phtn$=sFyr*NrTXlewaI4IeGw4B%{(+<ji0l9iJ
zsMr7Rc(o0$eh^)~2v<k^AF|K?!9IV)KL3V&{w@3bpZIx<9LXhW`k_!{<3s@hD>4>=
zh$>4`V^ojv`znt`7}eZ*8J6X`TP2tHrL<dH!QNQFvBFU73b+m4Ia=!8SP&agOl6g9
zqEZ3IZyf6_xY)pgHjMgKvJuj0Hd}#{*gF$$xoz9I0j%S1Ev#qqU%!20VH+zCV>@hG
zfMoR6LR6%%e*1RFk+EoUwH(=FM{;w*k*L8bZ%Y}U!-WMIR>D=>h?8hUca!151O4R8
z$1h`*H9nH3lKFZhc_XiMT||z#(gL{BV>=A&j|dt=&Md=#fp>KnNMtg_fRL8neE$%>
zGK>O)e1RpL!mJ86Ff?4)6Hu?g_9A5l@|2xbpBUJRpvC82QK!@nU0~Ots3|rF!f9NY
z#lDJYrs9?)V$@qPbRH*F7%prMLgzQJ3Zxq051T`s7Gb~LYgU5ZfWuh0Rq_B=G3n0?
zA>AG%{j^HjzXQI>kul9Re32yL*Wu?h-@rTQKc-SSQw%f)7tZ?_t)h##WpzkZiG}eO
zB%f-gLS3Kc^NeU~35z(ieIOKZwgm;ZJ=JDv6@A|c%)ykMB`}slmVb@^4d5}E0P~+T
zNl-HpJ@in!qI=hmq6pUbeMZIq82pKpyWVni<jBxfHW}5GR;b*G74R}rPfZn(-TW!%
z_}{A>r`Y`1Vv@=qKobMzD2!fEu`&8nX)2V_3o4q?pQRdoG9XDad%99A_wz|!RclcV
z-~T>HDqh_hG<=$TX478>TxK$TF2^}A{mmKUJre|$sclFU=r?19{3@vs5(rl^Dfu=)
zxS6Ix0YOl)0fI|Yp@1N$Xh2w}0l^e)r=h{3W~_k&B-gniVZlRYTy&Yn^?2LDMCJe&
zud|=lm>{u2uP#B6yd5j$5mG5+pZ_%zk`*jIMN^?*A*k5E;`1~W3KoKj1`8_4U^9rk
zLD5%9K63+s;x{t_o@L6-fB-JRTnMnK6rR!O)q`3YA11*>IiEVgSC~7otjvdC<b*J5
zaKh!!TiWd&9K7RgE0`-1O9?So18KLbS!{&a^$ck@q$JaBcrDXzc#UM0Re}NtWe?m=
zO}iDnXt!|qreA@q3}keoEe**8)HM7qpBXA)VX@2W4(^O#&(MTuLYWawjOo##7Xc^S
zl9Q76xD2BGRoFVu@D^>^RhT<9!JKaLb~`0jLL132N`rB16k4>m7V5>(o@oa?ERcXl
z=R4^6He4Td3|n#S`^x7ze9h=zfHj~xsN@iq$DQc-_zsp|EF_e7%~-6J#8dae0PosE
zS0y#kSco1OgLLPRwGy>nDJIv~GoP|MGKD;Rtwc>Ru#BI+#PBwh4>v%yWIV?Hnh>C*
zm@qLvLq6%Ms41qN^C3q!U+yh@O`3eoJ}+i<z~}9NURW<RFoWl-4Hc*-++HH17pjuK
zS1dwc@q{-#AMAOjkRiw}-y(!;K<ZN;zr-r(WMsJdQYl}Fco+7Btwh|3S8OGMUFBVg
zU{*Z`OPW{e0c38dzA5R7(;;5cNsbKHWUy<1?1#_<IAoKhW7zhEok%s00NV$nrWkCC
z^9kEE8f+7?HbXaNcL2Iap(lBu`)R0`0^OtrBz&<-0=`9n?|nj+x#24tkb2Dl-!Eft
z7<`|_D+XV7b-dyGmMANj8@WTaVQSVCQ5zYwO7hZx*YBXtIbM~5F;xFr*o0K+2vGfY
z)D(m2>*o`yt5bw)Mrq9H07}0JoyddIm!Vz?N)@?FfMV$cKwl34eO^c~H$Y_rQkyvd
zx?)@cbR}LffU>LO4bYux&N4SfFU>Hzssu-dsh(XmAoT(ew?L|1HU?`Ei=-||6tjfG
z=S591u$JZ%tY>Io)ue33Ys~BbUeAGk<iYC>sF#9QUHTHTSRnz~5<qsHkYjGh$_7S&
z>@C<AHm2T+R}8Z3>Ucx;%}lq&4VlmiEBjun3Uk&AvqE)bjK0eGWfS0Na}N8~EqOEL
zs*4+LJACg<>1|hC!N2dutuz}8QGA%WX3&8n0m>K=z{Z}aw)oyHiHp(|yuikSyWgp2
zC=SgxyECm$L)Iw+wRWBD%9(Dp>V?Y+H*|Wg*z2Rs(8Dn%5t$4Z-(rW<5!Nq_cZBGn
z88yWm(nSEMhag`glU3o*|KHS#SgOz#zc0=s3ek|3rypYtMcU=xV(lJ|RKne^u3d74
z<NZ(-G-$K%_Ys;3wZb8&*sO5;7EOg(;Sf|b$;o4dn8=1Y2QJ|#cN^^l$=FJ)ng2xg
zMoXY7hmH(gBhSc>?a~O#L&4a|_KbX%Sw?14(j!^5hsHf4cQGpZo{=}^_0T5hpji1p
zAmO9ji52y2Qc)y?zQwy>1EIg6sZbCSRBRyhbD9bTAwlItg3yu)Dwb9aOj~&PK8#86
z!o%P;llg$qVqiVFXkuYXa{!@Iwot_coaI^*5<H2!F@QFbY9f1GWMZ-!gSOICC;$m6
zHUN4XO@#uGprQfj91$O<1&Um@gQL#Sda?B1L2{fMBttpKHL~2q^qpa3;X<7QSZvA^
zk(*GlOl?7eLcbYn<-Md<NKCwo$;b*5Ax(wCgrH)BiHB(_6ea|f`NqUkB*!NhCVrdg
zI|C-TQ0D*>>n*39CN!LxURPp+DiR~*&q!U7Ncj^cIV+_6HBE&=ilAbHl>bFjp^ze|
zXh>O%Q-3ba!6AbG7|Zb=N!qz>o|DPt?Eeo_PX@!zAnDgGp+ZCQ5dlsGo>W;CJp<8d
z!{L>I_E>hf@G@z~a21QWND>Ygwgw-7xB#oH@iX-9Qj&;G@8ImzT9chktUO?{7;{Ln
zSNu!dvA2>m@zMPm2R5mdA+ZIGXdB+)q4!?9-RcD8E*u{lytCKBGmzQFwJX)W=W<y#
zmHeN;hCX$poFP*+Z#zj&8BCqSNv1Ft#p6i5-^B#A*f<w+Igm52!9*@k&b$TkPRW^z
z%%Dh?ho=p)S=@l9(5d_DE&{vVC+3v7)911QnSs<bDt45N^PIA2JITG+Czd-uj8`mo
z&aRGk?tFK-3`bkQeoeQ`W=wdpE<R+GH=F(|8M5gqq({!LPp33o#EwOYZ}upaYCt>W
zvCbO`o3<CC*ud>{OE+NZ97GcpVIHWcC$V3XLRNaK)Zb$V_5`|`xmkI8G-`?^x-U*9
zy4PC@gK@8CxH45O>Q6;2MzT^O{7c;Z&qq9G&IErBs#<1(pQWi#ag?B96GuHqQ=#H0
zK}CzBEFPj@tGa_hmPO5213x6W&K-eS@Q^D!dx2>^LwLqTt9x&?SH>*w(*`Q$N&|!V
zAPOO9<Hx{sTsQ)$*;SvzJ=?<R#Cx-&{Qj_Nv3gfcQvOBlQ>JX>Fa~nzb~*5xOh2j|
z>Z)3bJ%jTKBm0fK-n<1(=c$89DQz$?hfy)#17Xxhp-Xwfs7pa2DPdIVNF)QR3v9YF
z7;0S%8hBV}Aa|%G8^{}KU4#8*q1Lr{#X>E1b-Y8ZYv2^5{VvQff)+d=Q|q`@xl*1t
z*t#M^u$4-D<Z$bUsZAI8t6XF^qtGiYmAtXgbsF$$0*FC{fOtf`wa^VZuzz4ZY%zdy
ztk!QV)Ll5d+AG7mAe?$|Xlr4n-f{vcabfAgjRly%&BES;ZpEp?i6HW)i}0r&JRLLR
zbYc60*phISQ}4Pg;Im0{CnCNxQhP2V9HK<*m22p`<|gWMqo$buyEGZ~ee$1)MwtAH
zde|vm3}kIhq75yKTA}w}swsIp*3G|+1e!S{y%VZhhNSn=RH%?tP_YR~AEBvGA*rCE
z$#UrON=iUFv|cRzPm>(0(LknnWhe)^0+vrPeP;+pxll6(qX!rj9T^wcxlI!wR-3Cz
zV5I#VtLV>36_L5`uk)_hY(4lMO@#u8pkf1%|4mb&03xVp09gd178m5;tc?8@OYxsb
z%DD}lQORY@zhJt_V9XiBe8v<t-jS(cL1jGznbDTLlq~mpn#vnkSgpJU4%3Dr&7Gky
zm(Wx!Oy(-JFH2;IBLNuSh>c|@P4oI>&xYr=d57n~6RjZvGn4ySp_iAvPTr3DemmWF
z>QtaxpsM9~?9)^zcO$6SxSNAC70Px570um@u}C+~;fP;jNq&H&p4;UJI=P(AJxoIx
zoK6N|{}$O@(&-2)HcscWG!@F}Wa!HmX(|?mqdA>LFl&;?W7n~fe1oQx!`1L^&Vj2r
zBXyc*ax}}-(z3Jq&A8V;qI*p_nIAw^ODFRZO@(qYf{KlkS+q3wU@NGM;$%)IsUH_7
zvkF*B<`u~>HG{A>liekqjG$uUWG<trP);U8Uv|+{C?_MR=uQUmFA{f)@H&Rutu&<^
zPKGlu2To?SIbme-FiT`n*~|Dw+|>@<Rm!t8p{k{4xr?Skc@{y%#<Sc{Q=tq*P|-X~
zrkz5MleBaDl@WFdeVnN$L+p}4(!WpkmUJtEij7<O3QdJ_D;fIoEt-mj)o5;Ij3sa-
z<ixiaZ!gl+a<~@W)m*N{rqaL)q9LQZ4Cdz1?00A`YxFGKqGePFksEA7InOnR_$SbK
za(Gu+G%gz&ggkG{hWc7t`ig>i;V|ZWVBzo%Fq1qBhZ{iMDGP_ha)w+wENV-Ksofb|
zJiHVZ52r;M492;^lYi86(A=GTEv?`bx|J3zr4?FDw-B@n>=3#(ysfE-N5$h(8~?qG
zU-2-y4Oq%4@bx8j`aL`iTbMos57y$#VQQ^rv<Ov_Ze^aR#cu13)`-LB-Q-zdTS0=C
zh?;N~7<<2bqXF++QTG_ES+}mx?zAdyw+m8(em8rK_958US6H`B_LnTfxdc*g15#b$
z{pC8+u0qCQ{>1TRR}VSoOuqL<8b;nl;WAL?i^6x{6<ZW$SI2u%`06sOQdY|F2Hke6
z35%0FAC`Oj-Hr?^!e+`NF9$p3c8jfE(C!7W|L$r);cYiI0cp2Ef_!7)idrS$SMYW3
z-o5<Xd_ZLxUlGLCe%S&ot5<6V;;jz)=DXEqs?ws<evTn`2<>xj7GZl)Q*06T8t`V?
zGv?by5e4f}VrBGPy+K?oe|J)>qMjCVPl$R_O40XNEDuC|$-Mq~6slTISRSRRP!kqG
z#b(0tBu#}1Jp~m_dS^0&B!q3)s<jyFC{y1rkSwWjOQr=*P9|6U^m`_)-6kz$;cW`@
ze@-&*Q(EIx3<655i+ztIky}3b*TsZ*Y|QPaB=6*`&8UQTM@o3NqJ)LZDZgc*1gqCs
ztfZ+>&`DMGVs!jUPu7w&a_gFtIo4RXh-BVE4a2c;IY}b7d`8BCzJ9+kveJ!;Zr@Jw
zZXq5kD|KipxvVtdkXj^(6GaI;M1sSf7w%P*@II1v8ztzfHdFcFs8sYc37MMb!?IpM
z)qy?*5ifeVmydATSvRZ1mc%&ZWM-Urcww1ZO4_-8Gme~|rsY7TdY^=<mXXsJXe!hg
zEU4HFWnZJIP{0yYw8)8JDkU$Kz}t68%Brq21xWdw-d{1@WXSV!0pzk#n@R%-L&NMt
zOBhZj=23=$CmE$qte|3y+)5ZBN?imT4IsoErLG~FC%qo0Vi3?Gn1(8jbpuHvw|p3t
zT!D*GJZ5)&8Oghacx;Rm_g0A~Gz{iBZ>iREwcKP%#=x9sjoBW^v$kP8$&+W@3zARC
zv!<mhWRmrKi;fItSg(QnTSrVv06c>kR+|^R$_Dag;(FL?_ISpAykeO+c6GcnaW_eO
zfR{N`%Li)kc!9n-E$>qryK>~@(s&p-JJ%lC%Z`i0WuA`hDT9=p$V-as9cJO=y~INp
zajK426>*gIvd_XWCS+_iykHz}JwQ6LOsCc!jhbQ!yz2ohS_1F#5oIJ~qD7#mnc5Tu
zwcrWfH^LMoWfXmnMe&J<|IA6dC!p%Uz-oSv=I3ZCROl(F*o2;6qNz}!r=X%qZz+p@
zRMRvJ@5%V*Ig*Fm<NFBl&!3vqWR8D+L^4kf@SKW4Km(geUm}U*me44hNsS^h$3T-S
zsrc4HM0yNFPFcB8>J?P^7^kd{a&>DC<CH;{Y;+f*SmZ9`_t;&mr@K$hRL_N~mhR#b
znhNDE1Qi>1v6H4kxeGx>a~C-#OE;1nsH)C1V;#X+TxU{}*;&*`=1C-QDh6kfm@FM2
zN#vGK{>hS2JZ496l;qt)JUZ@4N5Rceub@iD+EJX&5ub?DOui!d*ab_`tPYd6W6$v^
zS{{_=_ykn7^c<h3sZgFnP_gkG&(c&V&mpL2o<j^>DOs+80rYv2x~lg~USpI2^f{)X
z44E!2hFlhFQ)wX6WeC%?OjoX0Z{f*QERb6fBe?HZnH~~H1e)FV=_K={f8taO?t8#g
z&n1cEme44sY7~*#g<nc?Zy_QZQ&sX^1|yxfeAgypnkY55V9tES9T>=b-2-81p3K+H
zpo5gm*YFv?%z<sr)VaZI*!7SNJ1nLxFc1x9!_@tTAH3Ofn~75=3fl{Cq69mFqV_J>
zeCWA_daF6>b_!khojtc%adCTHw>J%^`oPA*=IoR{5Eou28Z69oS`Aj8Pwh*En`%%Q
z&I;)|g#!k{+u)oF8PA`9oW=3M3<);XRjB(IZv@R$L`~4lTOnl`ZFeTOCW?54EW?=t
z3IR~aynRu!CAKemmfIJlo#joM9tB=`(lo>?mNaEo$31D9JU*aSJy=F_h8>M8OEn)!
z(_uV}oHRW$j6%_ryt6;eMm{-YIF;sx9&Bd54f3rU3u#*)qy5VA(43?Wq^lTSFh=et
z1pcJO>S17}Z0ZwHQ!Jah4<M~&Q*SsH!V}XQEn_m5Dc1iSbt-d;_482Ga^~|@nhG`Z
z5manuKF`xssF{zTqRo7Ah24Kk@}Q0sndUzugxx<hsmUC6ze+Ms#+00jP1wD3)iIfv
zj1qPmMPv@U&mp<D5Ro2slbI84lzIi#f{dAy&vA8Y4&!*9t8*DGFe<h3Ca7xZV0O_|
zC<h~`*f^M5X)2V15mYn>lVSVNK9aI(6q(#g{_R7vOgH*yN*Z$pG2cygmfUbGsMtIv
z`VdWpdQ3D!U!I_;SeT6Vm}q>Aj+i6FH{zJ*X_{t^hedgZ=fD%q^>gT7p}R_XmoGzA
zOYib6nhNDz1Qi?a@*+)zG8I8Z^Ddd5L;pERJGXxs;W_kwVCu=>U@}Ph<Wr~+gmf^1
zij9Lgo2EiJm<)Z{NK>(}8qL8Bc@BLmO)ZCO;a$z;T5Ku}B-;(MI4#+}(l}E!1-7i9
zvU~DYQZ~6UG{))V+CO<Cur%P~%&~ui<em&{ITeG55-X;6k|c5qDF2G7Q9R}cKOP`?
zw-ArjgCCF5R46dpD&a|z#EGH=^W^Ipl6MOw*mw+O&BNd;j?<dQu*6-;TW#VF46J#4
z9Oj96);zukI!Igd7|Nx2&0|=#!8MP4koW#gk*om!My`3>GQM!Zf@}G^Vc6n?c=)9N
z@Axjj+6%rRg}oFd<?w{5o83YUPV;v01cE}fHZ$XP@CqK7bpm&`=63NUhOS$OQy18A
z3|&ldR_eVfE|u_;zPA?m-#V<L@PfgC7L@7+4lJ*9ry9<|a--ISys6s_3j0glQsF?W
z(+&6!&O!BuF1$v)2`3&vN%pg(5Le}Q)YX^<@1aPb{(p&dTq%)w*7Rf2xu}j&UE^tj
zCH|eL36^+_)_8si6f&>zNVdcp&)t@5JQ7zjW3Kzm;jf4~^;EgWa~fW;H6C_#yw`Yc
zE)RD!g`)i?_X0oEe?vdK+P+e0C=(;E@^prh7-^Y@ZHzYt$@)na>xQx+*K;gN#xUIk
zMm0z0TWpA$VhcKl0l(UU&bCn&bdL4H&CZDV%nLV{L)C#<fq8HIbu<+!^C76%WIoC?
z6)Nl%RJ6=Tu7#U-l02yHKU0QvgoT@iNloU-+ubDdWMIsx7$$E6`=uTtiR6~hC<`}6
z5t-)>pCq}r5RpC(lDTeflzIhKKE{QcH*<BLFbg-&(gLGWt6$<>(Vwz6JDBHaDwKl}
zRBRl~3p5qV!3ZjvgUPUP^HY+tY808=O8$kLpD^8I$g5@$^YYWEv6;-P3Mw{v)ipE~
zDzBQMFB@no7AB+RRpWIUV&Udenr4pdD(`SEPh?YRAiHV^spcrVdJ`#|+&~&7yLvsa
zG!RFb2WX$<o*aER6~h3X$gUnFN#qt#{_LtzJm&0bNb+tW9;@u?!!#8N(zZ(YZIZ-^
zq6Bk<^amvG7D}-37)o~4;49`WyIRsWT@OnPq@-8l9t@;c&x2t$PkQw`poEn4s>L)}
z=2tJw*1f?5>tRT1T_8diaB+hPR+}RhWCM9KNk2nbup<^;!Yh_ZVpqpIlXUyg)CPMw
zW8OW(*Ue=*iE%P=PU(PUk4MRv*d~@Pu;WE$mu!p2DW(;tQ(<hTJ;PH`Q!K@F^q*CV
z>AZ;f%qgY<R2>M7%rn3(G!<$FD5%)X0N+AWp~ikeMVkTUN-<qe@{l_?7$L>9$D}56
z1Uo}APeuoviXnm>NHGN@k=zm*CB<YEkvV`mOmc4_B0Yd2Q%u|_^$Mzdj47tGxVlf6
z6w|+>1xBTq{tfSnO^WG1&{QY~BdFLom_MYcP!2{=(Hu;M6w_allvSgcrxeqlG2LWH
zF=Y_*kIBxGDJDV1CdKqinhKR-%Fvgo)l{5e)0gZirq|LmbEKGfhjV!%n@R&ICPPRy
zM=7RDN!jEE(kLmWHvmfmag=$0-bHdxjy{}<VSrAgm~JIW<Q7o=6q8Xr<`h$d<lRC%
zRw<@CX)4qxV5@{9B#9G63FZjtQIdBHCD?cjCB<a$74w#2dPBAx44<c)k1mM2F_32(
zgOMsvp6LsqiIhB(?Sxq-n=Z-G(ZP(<QOGz=h+#K(#z{7iH{<k8lnTo@eG9Kx#))0c
zMaC)JcF|(LaS{XrYy3Tn^q4@eBz9koRq9T++jmv4`!7HP{tu!Kycl(WUFGe-G_tjp
zo1IsZ#}|0xY!&OA_w<Y2I)M7Q@Lzcac{l+ZBs;J(vcI=q^xq3jgqQDj8?DZvw>vYg
zYA%H$>k4dzERabr*kpAvS)gQYKau!iy-{9{<G&Ah(>qo44=eY=FPqyfHd<A;UUoXO
zaKNZw48N3i<NY0-*&gl|g`J?d+0;<%3b+k#22u(-_1bLH?U>45d(HkWJNVB^g^oK@
zw&P-h*o$-%*p}>UxaGEO=i<A>wiebi6|CRBv9PV`b}Qv>ZMNKPI~90kQB<U`e*1Rr
z(u;5!zp%eJWA-W-o4P5>YOEZ4jb14+VW7rJq(^OM)MvjpBBs{(uS86;T-^QOW$ss1
zeKu`iGM_lCSS*z@$PO0$hk?5P2z=;3z8Ur_`?tUcx`q4TXEX2GNwllYW-573`#I)y
z{SaQCL|&hwd3`LES8ZF}Q+8H+Vqix=M^Bm9bwE_rru9>CRjC{GR;-W9NPUp0p*KO*
zfjNUY_qvOwLZwCo6`Rz^tuz%1=z@wS!!y(UEtl!T(#eToYIR6eiDkZz<Wn2qCc*nw
zTeyk5?|&cMO9-bI&){p``u<k|1uW4Lz<}TeeUiBCQ(b$&7kD27M=;Bjn{bs>qKr#W
z4<ppB=&n+{TI3Q&#n-HUz2#75GW*G>uC#jPPON|jNCl8af0T1<lMj59ra~FLpkib6
zPtsH<qZd>(qd!YE`egKwX7+TYSngjSc~z|?QwEVHpV{<(kEu9=>2o>Gf$49~7@V0P
zuuN@3qCmeHE93=IAtVsK%cPVzWzp>Ye?n8CfFP*Y0O3`d3Izl~MFYY*4G5;tI1LRJ
zHDe8|d@U86=7xj?51DbX6j)8ZDcZtB<^UJ3v!8>QAhANPE<usJ9V_Jxq*BN}Z)QTW
zg2fJ+3Iz*6#ReAF&{QZ`2r3#Zs0fG6)$;~L4#{V3AW-~fM!;UC+zcZFmtZag*i;%=
zvoK6O=PM8CZvg*mPAkS-4Wt!+26JWWw4yie|1nMvk&?(Mvg>>L{@3MwoXg&e^TV}#
z@(rQv&&7P)nC&Z$l5E-)&$5O+&}W`E7n}g{SJS<~g~Ze4bY8&RLMdh;k%M|90jS|Q
zD0PO2ST1T0+WDWs2H`d&X4I1CHa|SJZUOs;iMloZ=MtoK#AMpo^mNn|)8~Vr#QC!5
zaNTf`n+-l@dcfd+2l``e@Z8dw&HoA1lX>*CEQ_WVku~?RVgeKof{LCJay5d&X49o?
zKqhT<e<c#glaYNJlgQX7mIZ$iuUHnGUFA(8Gn1c#B=R+Sz%U2c-kA&AIvXS7I~nj8
zF#ZcP3y$$*={P)kRoISH{Rr@WC2ESn`=R-S_xUOCPDtGh@tE-ei2n@wln3Ig&#-`a
zQY#YXSUCanhj8o@^34r%*?`n_4w!GkeleJr@QT5lT^(<j|5}tSngfJy&xP=agOS0l
zB)SZ^z7ixaaIF-KA^leTU0JXi0n$68rWm9@GM|txq(C|rvl-1X&jV=Q4&BLv<~yL?
z&;eV~g#<FzO91&J0OYp_spbZ<Y+wW+AHXg#ARojl24r@1yn(!1&3VlMmUraBvdYBB
zaMm++21GxA2Ej*ay=)BX4+@)+DjxyV_eD)HP=9PbL49rts5N<;aUHWgfa}B1qdd6&
z9jG@H*Sba|w6SUe+PKv60U_Jm(3TBIJ?9vszl6PF(EdZbV$fz+#~a$$OD)gT>Mk7E
zA9!<s@2f__cf`raAYV1mIg!3p*WCTTLi^xIH&lq>{@;WhNi!G$?$1X}F}QzxKH***
z0{2+_X3WRD4`BW;pj&w`|1+pJ6!Qk%NZ@1r1n@r&!2e4j<=nuR4M>gW0RGZ5CGeNw
z6$3uII^Mv)l3`=`qm5>3iO<a1njh;U<9kB%!GP*;jQu9pBE+{C%;$qP<|rUv7d6Ge
z{1}*<wom3H*d)pPOWHO&ed+!~e3z4<)p<+Ion=;fiHp?i7Giz}+`?HPpgeBjR}nE^
z!({U_VQNlJXjZs_@NA0p!QHBlftxs6=r6aMkPVP-Vw|5cXUfw#*iRN5Oyd>v5$tL%
ze1z_^)O8;8m}0l8OyQF~NZ+m6iaOAaI>4^-c3_&VFeD8rTnWkeUftaU2WJMjEwj(8
z3i~8~Dkwj#hgAB8y?yAWH|4GImT|Z<jGn1(IxKqI!nNFsZEDxrZoSrYH&xrM{aZFM
zgLgZiG_f;Q7A(nvC04L33p!qc>LP{eNy*h+RU6@Wx7F)Z+`hLlJPnQ{ZFcKUP-`{I
z?M`d9<H9y!$o+?7K`U_T{ap*fDOtv?!q*=5a$&fl((B-%re)FAo^YAGgT3nS375cb
zRebkbw;!&mwko}#R_~VJ1z9N2Djbh+lT+`(e&2`seD75_-s%PIUeMp;t=tn%&bqL*
z)d^Z1XbI=maCWKFXv3GKomMLVR^6G}!G5@Kb3a_lSttYDeNbRZeC>7W@NId}s+6al
z%0B$H+xPAdPnm%RnytW9ZXXVp!y&0bv)3peghO5%{lno3IIXqTE>~I&wl5c&TEp+a
zX{_u{`9Ll3%4{brbOR`f3b@gl-B<=E$hJB`x$f?FL69efQ=U_gzK(^f+=DfcU*H72
zt}HYbj#b_1-Ym4Y0^TCl1o=UUj_Wkw&Lk+J;$mlzzgiQdRfd=CweZ;3emL0+W;VTX
z6YRz9hiB<OmfcRL)hT;Uvswom0|S7*RbZppN-lGyl`+;zKU|ymCDDZcX@Fn2g0~y<
zAZwqc|B$F0o{W6}tLu6oqB``cA1<#u&Dowa>w?6q@cksW*$+?cz{$0pYPssdy9TR3
z1~-WB3C96?Gtf<tNUbVO!5a@(IK7}%j&Fr0)xgk}S9^_iIWk5NEV}{k6mvUGr;ZJ-
zWOtaEqxR!l?XKIaw#o+^Xi6u$2ivW#TUKi_E2#wK{k3k58|%t)wbpf}VTu!RIUzv;
zfLjKg=>jYm;-JyxqF5P&Vch*5u4Z8|ob>B<7%JT+wpIm|FYP!7Kvn$fu7%+$L>gF5
za~2G-UTf5X{_d04hb!7{8(u=k4jqQ0f`P~~yfY2k$IA$c@U>nLw7Xk3Z-(35Qu|Qg
zwVI_?XBMz@I+kP%K*@68!YRgpy?(d^?r%E(!izzkD>^fkMDutjZvFz<Jm%`qWcRt9
zQq2u!7!7a*o4ue>2V#?fGYu%J23~*Hl5j29WTPB`GwJ}}S>8Vqp4>gu4B*&fuQuz!
z-gXb{cQL+pvL7z%HXH!jJ9~iYE(q#W_MLw%*IN}wh<0tobMTFnuxXxw5>&}V2(|b1
z+CX9ja2M19P#$*!N5Fbom0EKKY(V&hQ_xZ1t1b0nwFcU5z2$H)E@sXMOr|mIRsq3>
zfCJBky`4wG(<@MFUm4F6hQ|S_RrI)w>s8n?`e8Wmd4_wlBjF;qxxc@sHytkF-UQ$a
zmRM_6Yk-ZQ%y*13yIBQHfn^2YXrYa&JA-~aJVm?PbQ(YkZbNG`(U);_rd4)-0&nu`
zqW>E>#J(3lejPWM--aJ$T%NxjKMvs!`{AP(PGT+G%gTlhR__R_bA**Y!pa`$dFx<c
zVGY6%xy_pl$Hj(MIR<!R$;U*tMdPj<?!vKJ7Y674VGJsM3gpH9fW>m)h2RP~^|o_1
z9poC>^l_;7@4}BG`0+A+{1QK|hB(;24nKB`!N*njaTGru#E);{$G7q0oN@R#4?k-7
z(Zr9BE`g6n@#Bt3_;B%~4X^F=d-#E`iS_X{vOXT3?&IO=KA!XA<2gV+K8fYylUhD*
zo?uTK__(U<<Epcdi%&i-PWhO)_A#;TW2(!?RG5zwU>_&QKE}L0#>BmFxr!d9&*haG
zU=R%gbj<Idla%GS;iz|bIdDUOw#Ia;9s!aSJuPuFDm|&SI4P;JN~dV;!+6&T`ry4z
zPqrm~fq(;Cc%{`t2MwcprB%m+(xDF!Nlfe%_a6c)zPo=txTChy0mlY*-30`4f{`cJ
gZZC~)GY~E1N3J7iXVLFYRdKKpei>t_-gN2z0)d%$&;S4c

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/evaluator/index.doctree b/.doctrees/autoapi/lmflow/pipeline/evaluator/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..26a76def692c2735391296a46af5c17df6ab13b0
GIT binary patch
literal 41337
zcmeHQ36LCDdDgY3q}5?t*hs=H8KafFJBu8Q@md!6A||pnwh>tIHSV41-R@RT_qe-9
ztF^7zI2@MK1*8~U<q!yvKnij}Vq6e{F>wL`f<s71Fcc}uBmqLqQ5C8}2uZ&G9o?^c
zdU|JibvP+h_H6g-cl`hV{_Fkkx_{v9?t50Sf6<1ZX<F{Fie9g~Ufl@V(Qw_XwHk&S
zw%^v?{krz?_INbn>kEO`@@qyrT7euj({k#*;kIvYM`M&dw49(K$`1ubEwns0E^ZE(
zgXYlf?QwG?8VIe>F_cfyc}`=_@fIpgt7$lvYgCM*y3^7_&u>#9lv#7gI7Yw}Wqm`N
zLps>h0aw^IH$@}W7!h-IG$`PoiPlgn?U~lB9*yc@=v%X`5O@=wtF3xKI1SYtJqVD9
z=K)u5@TcK=tC_sa>a`=jF-Lh?(NL(*)2Ft%!Q5)DGq;(m&F#_6Q>NE2ru>%c8vfLR
z=O3B6{@^Qtto#E2gDDQ!l->$Gy=hHxa!g5bOvU7wvfR3Htl|g2?ga-{wPwu~(Ev+R
z$jK>fo)2uy9iZL?`2RxuzYG8GCd@z_-(vWTM{An8<+r094X@sEjLWs|JrQX<&t&&3
z=HN`&)?4O9=43*G8S@1oH2QD^{i4tNdec;ZJ^aCyD^?sb4eeUl<r*Wr7Fdm@W6c3h
zEyU-h@70VT(7ZW~Xlc3gnxG2UOv8V==IHKxOP@CgPQ#%RN3FV1*Jc;-@w!)DW7LrD
zA7mdbHwbmNW@u=|^F!pC^SuUZiKC}A(YL6qq7iT$q?(6Peyaw))I$9TgUmM!t!^=9
z*NLc_j{p35COpA32c?RHKtt(Jw;1Gc2F*7domvk)-gXLw39YV&dhewvx7G>>89<nj
zXF}86x_tNijtV^x*iWElrHSTA4U+s7E{|J&)@%Q_dQGTC2<0odUgAQjcU2N3!~4zc
z<^<Htu#{y;@yoe2Fb58Qa}}3cE*d7-K3wp$h7p=x9gJV|4UnixEa`YU<ZgU7I9v}k
zdgCI^*fkbA=-*JAvm9eWl!Y8I{%0i&pEI#<ge~6<G%M5=tk7ga6Igc}q!)-~sm>I;
z8{<qHh{op)XkowwJ<<xBtxyYLR<wERsNrgsD-$Z(t3r|^4rU%E98Hp1LPu%Y3T6FA
zAyN_|)U7oj`jfMISTl`4K<JV};*BJYCwLj`JLo1_6DVdh8KXbbQtkwiB_G9BKpImK
zYNMxmp@Epz<Uq=*>nZ9bRr9<(R!k~s!3Vu~-LgMC^g3|?k&M6<>2=G8N8$O4TptUH
z=uSXA0OcCM$f5}JUPC9xLn>5K*&b@6)^0Yec{2oC;rS{{r~b7#*PySf3rL=gm7rz?
z+9AJX#0u|c?9PnT&)EI{IlCv8Vy1sfF{)UK5qQvEBQ=IF2coSzc1T}>duDE{(OiTN
ztnJt#e2d~d96q#jM=;ynd_^P28qRb89}c6H;ihi{hJVy(lU2J-!YHiUdRVo!(yGZp
z%0`(G4P7mU0c;?$C3#N>byx3%qLG2ivoq7WiFxoFLw&a8=zg{4In|(P)D-`CDQA8S
z%O7pL?eIN&kKc3garu|wdSB99WS*jd$n?cOH@%MrE<vl&(EY{z7@5G|kG6<GB@Apu
z`P_~+DQ{I^o*oKbGm3xlSYUMYRjw6Ux&sr>?L_=Z_~gtB;08SQs)f>~k!GfPLCuk9
zFr4?jR<nc9G(H92I(_!SFp4Z#_0Vh|7$qiq7YP3j^zCm-K3b|9l?rZAD<B+Zz4~GY
zpn!9=?=5tsvjH{@@N$a8&RaA+-CUgZX6@-a6MF9JI<Bk4nfBj#8<Wy;j62sg4Yv-j
zqIwKNU<HooveK2y(f*Lb{3u7M)%?5=z>yf3b|Qj*-CYDR#AN_RtD1}LXt23B3B22e
zIwB^5e*r!AS;Pandc@Hn(1fCDN4t6)rneC@p&|{8`r77KqanP7i%n>x)dWJc7VpL}
z403B$*nW_GUeJBFeF7up#c4j+E(~};vn@_D|F1p6+6o~8M%a!n=%Fi_g7^#~6sJ=$
z99Ri5KMBFuVh$#~_oUi;M%NyoZlW>C4>2a8dY+(SMO{V(9!JKHqrp#RBr%bNOS38_
zk%>E(R5Ih{mS|(Pf`l>cHHdDg*g<SJu<%!vIv-&w7$_NHgLz!5v@OkH9>d#XnR>hP
zEr?*Rqug5=Fhu)kTOnR?$8<oybl*h0Z|s!<z8<@$xZ;nY4<B2qrbm$pDz|%0(Hhe0
zZA8t$7LOqJ!`UjMa}?8rafLh2?T4OK7HyP%YqqqS$#l#HHS=a<l=dE^fn7}COCpTA
z4AH4cB3eha#BP;vS;%WBa9K7W#D!R%b~OS{umpi*q>l8Yqi^nqohZI(ABsybN9H_$
zRLO`~Zvn_b8kxceh!QKKjq{)ahJala+3p^-Tq-1~KrNz@E21?Rf_FMxWZ=f=OEs_%
z)kV-DC8sZrjSvqnu#lP(uA7jtGZAgu9b3%}tWw=|y^sx5Rq4O@)_oUHKw?_k#R#x#
z|Ae+r0s+Sm-t4p%XVG@;-;anc3#xC42Dlway7YB%2gRty2=ThqrdMi`OYoFb@~fFj
z(rPAJxzNtahJAvHY_+L*mD?qY%P6#v5c@u!{hqiX#!g$oPVg`|tuQM=VRc1KKEEql
z$nknqlqJFMy>YX70kQFDEqZ?v!v_XHw4UH$bY|}<A41~u?gD&97(Q8K6#ZV{;@FQa
z0qQV;%HbTQcDjT3M+G2mW*}0a9w8RPcPz{_D1^U|1=E(|_L$i@!^O3!HH{`Uan4)8
z$xHH9&621i^Oi_WO8WClg+z*0x}E&`MNW;s!k->Ay8Stx{nISF{dxS6S}z)*;3K0A
z!^r-df{!r5O7;YS?^5+_<|vOL?XTl=hr|gCzJ+hg#9+*jlQ7-T$jbp-vdIZTs}UwH
ziO~O)K`61Wc^W4=uhDWst4h;<)tE2w3N}AMp3<Ka{kPhk>L@x(8P6WIj0K9_wxbR5
zv5MIw6Z4gcuq-v?qrri6*mqXByB~C)`ok<F3Rv~F{TN`3M<eFH5ZLDvl~u4g_awxy
zeH3DT24*-RV(n;@^aaKt2&<+?L(w3IrENYdt%^Aq4bQb4hsh;xf3zBUH6kJT7tgY^
zH)6oUQj=^1+ai&a&m?RGj7RB}IqtEBr<pODu+7qLJeh8YF29Nbw$oZXw#=G<S$wa`
z8Pf{ozCp|;-zoGFf;XZP#ll^g;ZO4~V|ji_>9)+LWF__is-Nbtw?un-2U!8|U<San
z2_83}UV30_zY28}sRo6c)G@1lGs+bYTTdv%R`cJ?XU*?1RxrQc7q;Z&VkpW?|90>Q
zbxM)*K>6kD6y@J9Xauj}G>z*#(W1GXOPakNY%M#z_hH;ZrkZr;sv`=LBr%Mmwt;Ey
z!?-N4iCW0&NR_ZV(u5s}Mll)h)VZ?EiYmsn1MG8-5XU7u%=!(|&xpq;qe@a@MBdRC
znI56sTkVJG59cW&R}4iaoObwf^xU33i951qk2b9xq*Ve}%6WW^>&eQfhJQ4yLPUp$
zBjcDOX7cZ9VUj96OTuccjE?Tv6Z;0N`fK1R1V*_02B=z8#F|JOZ`MvcOex(`)9`6h
zO^#<UwT8CTf82XRSfdW>M8R6m&1VscOrD*|R7ISxp|DJJYp*>~q8Ef2z+xhJwy9Rf
z(yhiB2eI4(Q2-UdysQw#e0TKE`r_LsIN$y${*ZVp5i*<7R4=Mfg6%({+??}Y1VMO=
zR|o-MV;5Reqxvb)mV9RC9lNFL=R`cemVvHVi#<)bIaCBB7tMq(6k-bxDpBbd_uuOh
zi!wzM48PY8hOued1vJ0Qh+j;R?@;Z$zo-cBui`F#G$}$|{9)FL{So|$jnExEE1iC*
z8Y>(wB=*PECfBE$6gfV%L=H$#?oeps>NL+6(@+S{IuyzZ52ck7p537AFsso1s*1$9
z-PE^aw(c1cmfDqpsu(>XGz1M78v9=rh&lx2C1tu)LSL?Mg4xUZ!7LV*Js|cajM&AL
zP=tkw@c!vpSg1?$geBawr$6)A1A%)po#SZ{+T009o*Z?{24U(x7QdyRj<JPAy_WS!
z+8j|@DT$f@`EOzsrX}k6-5nLll|@++-n%OUSurmOsUnDak*b%L>vV}h+2RRy5A=gw
zELm8KYv0F+UQ7)|vZx5}r=BH?x|T~;agjfS-V(-+EZ6ViwP<Kc<DL&t5$@10F)Fm%
zN$oF0kzb?NrDqB4x9}F^I8u)4%-dsYdeRoI*pH$siq@{kxg#u8i3hfVVEbX(O3kv^
zO$>6P%gA`M$euQyjYY)TPnm{^SH@$qrTePyq(CLEznz|T$<ozsDWhd>yvFUjl{D2O
zU^5){A1hz&FOwYbph~Xm@DGJ@GH!nomDx|=CpUyI;^TiCSHj{FFK^A-8rVc~#HkAJ
zMQ6pjP@$(1p&^M7@xBDaj(FcwfS$+u?7e{6z8F6#Z0(m4&p&HlN?$3YXuph}et>fJ
zW$aU=%uJ{q@}{<Zxy(wyC;`5ggJ-`e=mvUSfowg}iz?bj%_bE{DNO5=aUSm)iJ3DS
zy%Y;B?4dOjyu!$xt+}v=>6OilsPC>f$+u4+kmY*vaZDyp@|83vqIKe<eBXo|H8H;$
zXWHO;GEL^Any5XD(L>C0@*r~vYBTdG1YE?BmLuBxWBDy9-xv3w5Gv_LmM!F2PbEjv
zcca&Yq-S#{X7M7*)OFINM{(a3SHi^os?#Oz=O*HwA#c9SCvbMi{7#@yM&_?UxjdQA
zQd|l?<<AKIRgl{k3JR7Md|5#$!ME2D7Pr#RdiuGIe#YXb4R|_XZ)Crl*zaccdlvgW
zoBf_czl7K+5KvmrWjG~;v(k%jCJiiGFx6;Qj!*`mlMqTZXCja-%CKC4yeqDR3FNh>
zOCZlq1X3YszAz>*b_k;b@XH9}n^7)L7*$G3L8QDHLA(}%=nBe}7DQRWG6eB`ga#AD
z_tVoD{jjGqE{Kj-)19iXyGQ!>f_oCgq=DrNqFNWN^hx`{pP-KBe(<Q0GZDm3QikOU
z;>Y7km>^zv1_V(-E?*E67&`><F@Rr25Wj+Qxq_$wBL$K2W(4s%2;xTt<w^^ptY8^}
z_<cfy3E~gvi3uWmI^%+f&Cr-mWV`qJ9**`Th)Dy>7euwLPC*<RNCZ*InF!(tWmv8t
z4#br(K|F8<1W`dQUl0=*I|OkBz%L_+J5erI5EWpgAX46pARd4qt`d|hEr_y$WeDPB
zga#ADX?kLU$ezx$AiflX);Vjwe{0yWpd}u_C@~>k`LYFYP}b8ad^b|(aw{lvC4%>A
z(etw7(jm|u$A-TcSHc7j%eYTZC}?BiiVF1eg)V`wL+E}1pqCLk2j%jFPEuG39OcUh
z+zbTndO@+$0w*hw{nl3yh>rz-g-~Du_j-C_0>_@tw7^Zz!i6|;DlK2qz_JBzRIRI1
z_})Ps&4o|RnF!!J1t-XkEG2-C#Fa1syzz7ipfVOz$dE6D35*>=*arAzgzzzx%M(JC
z(ozs9Z$=PrgdjdFC|6n#Wd*YTN(kZ?2n{BPPtg+-MD}#11@RSmXFK%di}C{qFfVdR
zFQJ)^W|l3Lmv>Y-E>>nt%RMvVJz%p4v_|Y2zE(`^MM;e$qfH(4$yv*sFNtt}OK_KD
zj#9$?&A1XK+=sxbe}M?OoH&Sj#*BfSUX$Nn+41z6{UJd@Ns9PAA;3>#*tyWB2~fCR
z|Hmkb`MedhD30$lsy170EyTH3SVN~QgNjy8O9=3kw7q_?XiO2UY<6yTSW88=+PrD|
zA(hHqb^vWFY{Y&TRdj4f5L<vmE2(V|=L>P^#+)3{cn!9gWZM!gSuP<<r4V9uO|81t
zna?W4TXfm9SZYu0k`S={ytHo|3+!s?O}to1pKR4-lWow{Sn<5o-UDE>)89RCO<D=7
zILUoTVxP*77?Y*BGMnArj(InA_d||Ao|c)<t%|#wJw6Cqx3T?!_Ft!VJS3oY7~Ova
z>Sc_sjxNbFx;cta*xqP+0pR1iX{WV1NqPwV{tclwN}FX_fz-Tdju4ZigWOD|XGpno
zyhvCwn|wDtF`LYu&a_Rwq>2-KY9>w*>dz4OG#KS|FWdTFDqzIUKsx))sAKm}us~;2
z(Ww${9f!n)CLQubiB+2ay<*05ja<t+C}9#KSn-<Zv$vstmK{2JOI!)Fu{URJ>{Oqq
zFg-n<wzV6xb+AlnlbW9gtJXR{Boz2S41C{4_I)Ut)vsCoN8%~e<$GB6G+JwwKFz_X
z<1>&+r6qWLsyrT_SUw&bv+OAPFlC?DDw)`K!bVL?uz9*XHeY7g!1dV_yFMOPOR@t$
zlbt2e$}1!YcuE@nPGQ5*O0U!J`4$z~YKwMp{EOwX6F`;5){unC_V`QE<bM`7iH*T?
zIyM$oY5Sy+ZXO#4TTMD+vKrucobaxeM<<OkbNe4ERA~jiB##zn?=a~(@_$-Px)Q+>
ziY?!cC-8OH@rOXAGIo6DP`(|P3RfEM&57|AYQ4i^-wfV~1Phe5*s_AM7JHiTVHW#x
zdSVuvJ)LQbeW}5B4bzc={n_k^UIx#y_GMe`TVzfA%m#j30|v%Rj+iIt1V-PgU9n5A
z)mpw@Tii9F;e5wgFEFm4Qv!1h`><?IDwQN5Q7nhhmCH8nH^h}N<9<ulxZl9mF`Ptv
z8AciFKfoGcU(zOD`2b@g-VzGfG4OpG_c;{p!?@3~>}lhEs^bF>l*jIuIk@Yfcj8b;
zmq@@{i(5IDX;9?|X!3YOC_uE*ODN!7#jO;D0z@k*S7XTmH47^*D-`gdLWorxtA+yX
zeV|ve?fG<t0)CJ6cY3H^7z*G-Ep`MH-!_l^H+3jBA=;^)qG2S|pVVXWHId=JE2MKF
z5?^Fk+uu%cdA{HGOmQp!p2y|F@e{55SZ!rA(dH5}H7UXF2gU6U;BzVjwzS8`!C6+%
ztyAq!UZ)L-3=+5aJwa?_xXAbvcH01o!p!v2ZO<!irKsCPE2(Z<Lk3qO#a|MPE-Y@V
zL|1VT`s}LW>>bYe4Kaho_WQIr=ZfbW`zrbFc>-aFdwzhiKkgV9?{A~Q_2{a+Fk9j!
zNe5k-gp*))2GwqXi+)hhG1o;`4-<L84PsaOwcK;11EcB4e4U@yFS%=}Yzg1pTF7^*
zp@Bk<H)h=qm=BARfcdQ*yCv+mcO;G5zpkW3>Ujxr8aDR`1+O0n?MKiCfODu&e1wAi
zLAB+(#h!U~&cLCTODC-!q&3c|!xB;-a9q0F6N#g``RRx9Yy|k>>d6SHN-B9r4%i=3
zppkosQX8PJUJ}Bkp%a=lwaa-Vu2+0b?QqU~&2xF9zWU+`-WLy|P<CfCrKzKb?s|v(
z0OeSKS#<8}S%v9+m}jRbB2K-K;*kF_5rW^!K$CW5GD2XBy_W<<FMBU{q#(=VBnf5@
zITM&4?FZ&qaBCpnhZ*^baa49xQV~w^;!aXF_ERUlj<u5B*k9qh?7K-!uC!!!8Lw5U
z$YbJ;?_lGc+0Vc4hk9~D@s}7%=?%q{R*I1g(C8^nqtq7t4Pvre>JiBt9q9=xJez@`
zxN8ObQrq3xH{L$F#AZQt%jZuRHpPuAd^Hynpl7}!nsL4=E~4zzC00D%17KpOF4>Jf
zcIw`V+ASg^7RVNj=Shi)ox0L&%3#_l*{OTcNbfs!6M&cxcXTQA#a>70vPeAAJ+kyP
z-c<mR2L-RDCl(ZBPiHzPc(5vcwJL&URXV1f9t?+v`TGtPUe!yeu(-`-2N3ra*PjV1
z_7Ea`L2-prX)Tp4(P6iN2~<6rRwF&Gc0C+d!dQ7R8&2Go4kxDZ6xJklFs2d#>?<_r
z$AI)58gx;#NR?!kN!ToVIy4yDJMyd3aHl@yo5qLA!~VwQ!+tWN1Me-5!Miv)dx;Kw
zsJNA)=zwUY_vpam#ch>{4sa0CN?UbwbJ?sodq;F2M;uQ}bYNTzjy=D!gqq5a86<Fb
z#0*}Ijwmyw_s=Mo7c*E&LCQ$No?cPvj4v$0M|q9lpWOID(X|EgWCK}2S!@4O!k<k6
z{x>}_YtNp}w6#Bw!y$FUXks~>QSaN<zoMkAk6T=}rJpXur^x>0P6ci&s!m$@;!KH+
zf8OXxu<>Wbl`tEB?-{f4lQAHD+xXom+J}w5kYzuaHvZc3uwS`+*iVLyzr8#Lzr@Me
zi;X{0+)9y+7p?Se;}?tDDq-U}2*pNQnP%f`R-C=V#^;FRX|eGyDl~|RKFzoA3Dg}H
zehl4E#=^fF<?<|iv2v95z1U@m(do4A_rkhw6r5Aqy2}d6TKA6<?##M>jGmZvXHRF^
zx+~+}X4C20obT<=oTG_l+wbcWgG&&n4R#OfebmX-T#LRw)qphNnbgE)e@XC`WQ|he
z@~7fTn9aT~YqPId7Mdq>;Qs9x(!P!QKcQ$JM*SNsd)la<>NPzp))qNnh12H$#lh`m
zP0v_yD@AL1L@T|l>DgA?O3|7g(MoEi$Z1&9Qz?X49eh6r*7WQFuvu>IfkBGB2gTXQ
zWB-%k7v5Az<U$N)7|K1Z47szol_I}Tw9>m@=oYtC!Y||?^x0L#**pBg9F>1s{KDtu
z`GsVd^Buti%nnEJv(R{D9KpAvzw#WxjER+=-+5gC?sWBVC$9Z@0L3SpN*+-uE0Fd)
z-K|Sc+x$LZ$6UQXpeN?)v8U&dtLHkW+|@%9C)3p<h?nI9E#>Mlc=<jpg%2q;arHh^
zsAzhef_XBogt>YTJa=8aug8$~?dm;^qFLjp+V4#8{L3u+>2>vfR35t@a&UWb^;WMd
zbSR5lJ<&=puHL5NR*GCb(aPy|^)4ucSY?&xfU9>tfX#Aqp`*l!nzkG&PCc;y$#C_4
zzL3a;7+lRz?#b1AO>rwluAXS6cUP}g+*S!!kAu)>R~2XPaP@Lj{+V<2$fV`FdI^{v
zuHNI&cx7C@htOY3xOxOnp{s`sovz*kaP^)L8ay|ur>+U!FFZ&i&|G64XK)-fa0T?d
z=2%Az$1**yu6c7>Xc~Ntu;w8~gayOmAdC3k+2vM~I1-N(vazZuFd&SVCcXhUKP*aO
zxoC2C(zP?!U$0%EP41nV(SlZ^dXII*-s5+OXPchyU8q|3YK}YJ9ACia3H$iHICYFJ
zB*yX`x9%<M-aR2BT_>rw1l!*!h3%WVV0(B7=aMX#SYoWrXcoV!&22Si(PijF>N73S
z8%@UuCNzPo%w|u9rWSff3|F(<QH$Pq){1r=)weTElbxWcpjUv~Jj3InLiTdeUc6Q+
zm5I?NBk1Jb;~K4m(%4s&Z0N_vUp%4N+9cA6#8t<@t>dP4k+yetMcbK5+&KzNTwKPc
zJI*5KME#i6Xf-rZh}Ks%aT<X(D?Zb5Y{{92H=Ht(JahkVq>1v({qJ0oEYIBk{<362
zdnv-8xss$w`0pd7`0qrD|8kb3r&sPCyve{#;%+cKdZjk!czUQUm{!fi-GMkdA;6sh
zf>ZPLPi7VTsEF96)}{~37Nnpi2??&B>jzhL**LDVw?B=-P^N{~PhC<9JqdEj(g}WF
z>j%GBbG`-0eT5;Hj?QN^t*tD+r6Qaj-Hkm(>y$;^M8YR~Xhm6kOGOlw9xp16PTrj~
zKLJF&-bo`t5gS9^-Q!AEX^6{)Og68VA(OM{iG@tq)5!>#6nCFkt9uj*R9#$|E+m3J
z1MH5)yc2h{7=axz0_>?Q0*445ad$`pOSAzYkd|Y_*OTBl3hd4zE|#A(rmdlE@M#x5
z?Y^CkC^ARQt>!w8Ad5Pjr}pf)xi318PvlL?Tb3scd0ja0m2HSj6`3iyG^De&)k<rt
zvV$j8?xMlKYxy<o`<sZ)4smvS;ILEDs!iXUhw2R4$J^u4K<I_K(>|~w8kKpBI$m3Z
z-O6YTd-r`P*{Y~(CR!&`*kK?u(J=N@&^0YV8z+-^wN_|3L1hkCfZ}@UIvnEH((xWa
zd^B4$=(WOTD{Rl08)l-BdBYWFnBWCIh5;+zXtrl?Fqr3s02LRc9&1M{r`pk44g%dn
zegqjt#cRuV@V1_A)}GaCN9ZeV#(g~6JckNg92}w~KNzhKaCbMnn(8sSp|$;BG=}R+
zt!B06H5z0~P}NqRf>qZng)hTq911khk|^Nl0WQ}@!a7(YI;zAmjv653s%X^Iajc1W
z9f&p=$1Kn<)WcSgqz=@L+15OYk1goF3;H1k&hKg<Wdt2jGYA=i&vHSns^Q}9bh->D
z8fk@dlP{ZO_mOW)f2<n5@A*|zck5U(4%!o_xT-wbz-g}3GQetSN82-BGF8}jK<sck
z9;-KjL6+W@{*WmgZ6qAPb%6;IIl!qMt#@>HzNOC_pg7$Hzshji(OJGx^Za_XZZyF&
zXvSiMmB;3SB350p0<Kjb(_5ieO_HKjxZyooUvD*<)tDJUSe9Uc4}Dj6sHZlt6vpP%
z`$<wWFj{r5daOaLw9z=$#32<`wIt(`8qTV*0xQIg^{A^_w*uO~2Sn&3Xe13uA%VcH
z12O@`f}cwWt`XU)a~2%vHuTCCdO{<C;j{aH4YnJwjxJy8>kH^AsvMHIGTKCvCY)Sw
zf5Qsf2RH7D#+pX63i;rB0vRV#3cJ&tgd#fE48vwHJvD{opwe6nO<Y9m`SXyab0{YR
z06D9;PZ-@3qC19>K6&AbUjllL`E#{Q^^}rS{~}pE4GoB8j~IT%GQv6522{b688#d=
zHWKQykfM5Mwhs(P+rg8KYAl>F0@a{N9C9Ms7%bwl=wnsWnl~N%5B^=_(2c^;I(8aR
zRU8NOAoh$Wrvp{l4FjCC+iNB+$;KIDK9@I8nUN5=kF=U-VhqrQRtWlYHE;sl<JGXg
z5<DRE!e&wwL2EXX&1w+u!VTrZxP~bsundMtIF>(LMCY1@5u%NAYRGk@YBY2USJ>3+
zq;Ub>gV83c8M|k}tIizPW+$T6hI_P)GqR&$u1z4m;6w~>a6LBo4x^-b1KJ6Evk5?f
zvqC6Z)PX~kNRLOGQ>k=mI7>!tbMXru258k;`Vb}Z=W+XDGObheb0e&yJwrcx=yex<
zTG0sQW%qHB!C<*hu$(7YE}Y7=gVL!2n8JOyv3ih$8|!>?AiI}qUu<8cF^vYS07kg|
zAo+g3O?B}-v}>e+h5t;DnJxp%CPQxf^m7mW(A8}=UF~Mmg-$kI_+-;<GB({dW7BmO
zHeGjN(|P(fowsk(A>TF~0&deu!#15nY}4_FHXV;>(*a609k686UTvH9a@({5+ol!T
zHm%gMX(g9UQ*|~?<=GU;u_@Biiq@;#&FH<pW<f@%!_Wsjg_M}g$2(R{;rY<rkcP&r
z=fncTvYwqujB}5qY8;WRvO=3v^|R45X_s@db%`%<2%utXIGPV~1nXP#9EUS5Ed9{c
zBK>XPi8r)wg1TtxK9nZ-E`Wsa24hW|N?sc$GbUeaxizO%C!v}r{vNH9F%Wu_PdHZo
E7fa?2Pyhe`

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/finetuner/index.doctree b/.doctrees/autoapi/lmflow/pipeline/finetuner/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..b2d6079e2b6d5ede1a9ecbfc2b55105ccf180a2c
GIT binary patch
literal 29804
zcmd5_eUKbSb=OJvE$L2@?Zft2_6p-@b?EMlEMxmD%W(`iAWNu%%BEuQYHoM#W=6X^
z>zP^I9U!nt0A11)P+k&31wxX7gpdzNLinmAB$cA5d{vxOfIk8z<)jiIRa8QyLP90v
z_xfYHXLfdOb`Qy@N_#Us-LKz!{l5B5Pxq4}@3o>G{2!YRTaMdYU$APmreCwe4jZre
z)po;fMxCcSN1p0j>Xg|;V6BCIJE+<nwgWY)j$5w<cC+*P4x6Itky{TJWc#tuu12ok
zjGH?n&Zsl?`cB!IU?Y(m)otSuyP@7#srzdSEw^RY-KM>;f}g0}w1W;cLYv`p_BtVx
zSM_Zjf$3<=0$x$a+0P~_F(J+_HY)Ky!-lDs&YAYI#U`yN3f$#(1iFdNU2ZKTn#QVi
zD-2PH>yf6_5Le^1c1zzZTh$AJy+U=`Y%H=?=~2g-c4nPD&P~oP=OBA@(eWGhV$g1a
zn#DCgxUl%}=~sYQ^#=%sivqDls~!1P%Uu-gSXAs-jM=g1Hf#3!LJ)$wn@{a*FFQNf
z2ro0(DLC!i0&1K?;NGqHe;fWE#{VNk8JH8eoSrfpZdq>7VTT%itzEb8E%hD=RuXmQ
zdslHr&qN)o?Y!8T*DN^W+zduz3@0!y)@oq290OU#&N+7J6>7Xooa?2~ZM5p{3Xm*C
zc-#v7svU+Uf2Bl>bepRs$r8GZpEgTYsctn_+t#X0$lCSNiXWixT)Sy4*X`51$Zdv^
z)vVg3H8*lftL{a+S*lr)7245)5Vx5hGNhzngQY(zq)M>V>N&(T0w<k=&QS=-xEcvQ
z24hvfk<e>rgv}m0q<Co8O3NFiX1meafLxXi9a0R*tt0S5KWD?`-bk^D^+tUuY*}PM
z##@0M+QCJ;L%5z%I7*y%1J~Wc?r6iYPZhAq2eis^`!JIz((I}hXBX-0c~PCSZJO)W
zzQ|f`*R7yZ_3M?eWmgT!-Ac`dd40C;b+3KnOD?_f*d_IsQ~ZSPjyYGzL@XV9$<hfj
zDPg<Ou!4<~uo=)qHY3eFEb4;sxWo1vcNI{cG}3z}@vrW$uG+!mq_WE+>1vWI6KphE
z4g7Yii(%veAcgflxjBv|Yi=!aI;SQ{WZn%1z7u2g9wil94Tk~XCrtq4kCy$~Mi-*M
zb64Q6b(M1$MkX|)$^8l&c6_O|vE(m%OXmg4N+)`n%>_~8<oVZeu9jib_q6P04YsYa
z4)<bU!0s3TtQxDax*OSnRX5-{e<<~5RN&Cjy7MQ!bq8L9<HdHiHacvywJ{GuJGSX$
zh{J!3A^IF#Y)>=AMnN6yOotsV==nMbL#Ro^a+@9JFWDIGqKy^|$}U2I?Z&;m4%bn2
zqt0dic+Co$olQ8OjV0l~a$^V>ri~@;gF0t;Uy&88+EIt~c8qfPxLSV(9{*~p^Qj^5
z_>)kB8D~@v-6zeVvwO0i#)!PcbMO)b=Xa@D-gr@ikE7ymp~FuoNtb#TOa2ymC_0^W
zzEiNHeDpsCY<_(!NFPBZn2O$z*f80n4&o^g__L^eISrCsXPCxP-EXc6yJw|k{_SY}
z)^uyLlJmGzp>Cc;Mdfea%<ZzqFA<YkDt=3q%4m*YO#+sTH)G$G<37h8l##8}{rEnX
zAn*2}9Xm`R>aD}nKYXV*AGhEryZ|B+q(r1wK<d27xkO;#6Ads|P>=4(a*NJW3SIUx
z@^o=XIyw2urW?6d-F<@$N8?8_S~UVmSPMRH3S}ty4h2-I1xwzS2l8La0eY_n)3@mZ
z6^vg?BIf*N^1j?#;n`B`lQd&;yj@&LJG{@T&e%Nz!xCQZ^9qvJCV`~M;|!aKyVU;8
zTZ;uYE1E7lf3dZ9^Fq%5-pUa>IOi*-<MN(DyW+mf`I_^r^Yt^%KeAhU#}JxCQl%^F
zhr0EL{D<drZX{e0FEb5^Hgd|Wsc9*p4R7Q<P2<8Hl2@n8Vm`@pMdv-;61yv{#ph#|
zo!?TA%aEddX#j>WeMDmF9E@kX(>%*jp@0P02+Vm@CIvID-A3O+ZhHvn$5U{7iyGMH
z{WjH}^#~aX3D`}!bZKj!#(wKTdWd=wgY>Hs#XP4+V~fB+4pKgNxO8uYQ;<un1FtS^
zS8sS}$L>9x0oa>GVU{EJemtXg4_|ILE>AEAGcApn&L(Mh9w(SAQw7q#oE&M6FP}zE
z19_X$gMKr@oFr7z)K#uW`@5jZRLOiclW+gexCQs^w*aLDWoNet_mat`(GYVQ%qrPY
zIVmc;O_^2d9bX?Q5_~5zPX@wL9e1!{<Tblz?=;t%q9>KmMe;6c2MIk@vm){%d|Jk<
zQks<Ve1<~b*xN9x5y`8f2j_)5^9HBPb_36O<W4vOY%k%%XDj?ZF$0r<rdJ|}v))ZH
zU_OS2ForwW1g9rOjCV6t6`oTuKqGiOhLh<w**GC6up1}ny%9Us3o&+$k)XKf+G|zh
z?K*5)T~?62bNs;I|5idslSVa>dtpUP1H;}~RwR7Uz^!#SD-0)2M)qZCv;`luu6x&@
z23K<=lfjJ+n<SoKeu|uUa+b_S<&@NUmW_y6r!&gNSK9SD=bPw%vJ&~#3X-R!-p}%~
zw__sAOLaBE?MA{)r2P(I%wIsa+<#I^fsb`n*oor2Be0erS%FhdPLoW&ye#6SVM?W@
zsgx6T&j%nj22SufyVs&yE+yuD)1Ma^?M*h;^ws5*DxMdqmrb@uK2-PnQA3g!#+9a<
zzA97o2om*&d~XDE)Y&8wuam4&)=v5Hf$W^Op9ous<BL+6k8KIYuBV-Vl`9n8<r_M0
z20bTDbJ?3U%Gd)jR%OVZU9Vuq5+3CVaCy0+a`EzhPSlv2yMR>6YIuYjG_r=%T%xRI
z(~qpw%N};@u@kpONH8vy4s+TMpFCPR!Rtd8kV!=nc&QXuDIGp}5|KK$vNK8(w9R?1
z8XPq$hS3(V40&clC1rAPrGflX3P{rJv7xn2ItK7QKuu;n>fUtb>Bv=C+NVYJcf}oX
zOM7d3**k<HH7YF65lFL}S4C6w`*BB^rr4Xub8i7Zd43@S+dE8bjk%LV&M48i{OK0B
zoFrWC?}dwW1>Lwn$_{X;p;t!@8|llsi+5j!=Y8p5QXC7h@cuD)6ZE78EhK7(aJUpJ
zRJh?a@aU?HcM?W>&8e^Da*7Q#yXEj7sL8DN75vGIvXvZuNmTz^UfugL{=~93LDre;
zHK)k?YR*HNIP2<~fd3-!{N-kr*m>W;<1S5-6nqEIwn@QM(mZ95kyQh<q>D8{)4K*H
zXg8W}lb{i9Ut@XP9~()(g=Bq{h;!J`oRGvBy3!OJ$)$h|`Q0*gn3~Lbw{ilKaxhG0
zUiBkxbXRxhc9lE&qOWAskTIR|a^!bhc_G0suMBKr;odnHbB8W9&S&#+K90v-5-a)h
zFrLAHC-h`gI6}oCb*uY(Jios@4H5nI$)o)0?1o0W9=R1-@<36fu%ozzA?2~p-;hGj
zdfnF7dpVvc_Fl0V`_Irxc8skt>WZ;{28~3F?LC4^?@|0D+?@9qN)Ro-j{b~#ucu1X
zllQnNr|Y+Oyz{C&6+YhzhA^th!)I?9mq|vHo%oFhyYt7ecpzLV#ktFLl(&rM6~3Nj
zfvRFTf*qU}A`HtkCYu`)`|gwkIXXkSFhDQP=N0na`w6m`+`Xs>ng&w8m_jNUK^4oG
zz7H_+45lFniHzw-&@Mk?irE4tZQ^a8$e8l+M%J_#oe%&@oAQ|LHjF+EL0;5H_0<{+
zs{c;O9ZamXWL(jt;~&W$)qKqR9pq`$Ko`$3ejx=WIjp9ECLNJEM(P`s7SA!Jk?n1>
zOvWBbGIoMZVogxJCXAPkzNBr!t9<_02;#&OdSx$LcExRHl|_tQ-lS2ajruqHV$;7;
z?OE@i>5q^pVpohsYQe#4-;a@-o73^)++1m?beiU852zPPC6Pc_m@MHRrTg#-T&#|Q
zyB6=9N`d`E+YPXYC-U)<(!yUXp5jm!CMW0S;%JV8f0<bGwEYeUEG?d*E|Sq1?<byc
z4%`z0A*|zS7;C<GU!XhWJ-oU_(@(1|cyrbd@w$%gOEG77ec`q<TAZ>surRNds7|i{
z)j#3sCuqV8`Z(}x2_!kYHdU#DMHdZ2BH3d_2YP@(mOcgRB{5Y>^g5O7HLf@sMgd-D
zn+RrAg{EguXZ8;ZC#j53zV(=<+Oytn{2}XMELz}Y=CmT$KzI`bNGR3<0FDTN&@Y~C
zpp}DWevr*%>yjAQnVyfe_Rmp&JtsYBBUz@}0xLpNXug&UgW1w@X^(NN|4~;|4AN9n
zcwavh#-48xWG->$=MVhd1YL|5H4)>LKg7Fq6B^<jyccg6e_|i^cww8?4_9M*CX~dR
zG`rlJ=u+0$w?z$TPv%s*f_BBFN-ED|Xp~kSDl-A)bJkFvmw>a6a)90u{E0bxeJ}5z
zgspc4Em9X#Smld|R2srYD2;cIfl-&D93Nyz750NQ)|9=YAIf57IR<9`G+O2bz!YN{
z%0f-V_~Z|%p)53{S<157uXirM;4IIar8Nh-`_59GUe%yKz0)@z?Pn}JsYD;<!;thq
zR5p+j9S6sMnFCBJ(T%-BA0>!+RkXN$K80B!MIwxfDAG#@8)rqqpv^UP&-6oGtW0-;
z>3_|cp3e<KnW%{vr^3oaLz|^cFP`IRyHqfl9eQRd1QJ;ye9cmW<U<I5ItllkemJH?
z{@a{cNs*_rfkgf;@a0<^U{d5K;#geChf$)3d~zh&Z~hQU&p`bZ5$NkOhe6@f<Eh4N
zwjbPLal9K0xgIV1u<g`D@UXBr65F%H@h+7YO$lV7nN2wv;b0c0W=dJYnSKoM?9PDR
z*AEYsfZoH|mlRMc8%RKxz^OYqz<~txT&mwf0!sZA5zqtO!=j`!>yZ{syC1A#(YzO|
zSmvzAr^2w5)I@NvuxJv)g=pqC`9GZ!p|JjcaJ!Yxd{-w^w#dkN&uQpK0sZTA%Tu`d
zq<>BHpmKO>GV8oL4v2F<fRYK-+&W?ufz!`pDDu9OQY*yqQB<5%B&ouByN40?Z(~IQ
zeE=cf_?OD*<Zqn0G1e`_pGIPJ!a{)t3%Y|&a-l++vs*7zFw+gh-?C@Em-aqQsvz>%
zW=*XtpUu^kvRCMh^dH1PIR8)MRnXOVC&I3cBS({^Msl2{334Tp#<A-y$tQq*;kP8>
z^$34a@^@$_UzC`OrxPilcgU9~RF|+Z)*hXP<o=%I95~kf@+9w*6t&t&0Z_V_v?_A7
z(L^kv$t6c#MDKOJNo2I?>RWVmF}_;E)uy-3e>eE=8~E=V`R|+f?<M*r!i=RsT#D5q
zlx-;DJ(ObLui^XG!``3YYQ<fM;cr_Fjl4yM7VaF4wa}=jm2jAz_e`4=RKguG5Uzx0
zua**Cqm@weGg}8Wid{PRJ&4{QI(QINXX&89XQhHvHKl@Qp@RRrT@_RSwxNQ@i3+ZQ
zchVJCL4NgotKiAl-wmjT$BOvFWGCBJ#&i-SLn9vso26e&*VbxzhAM4WEngJ_;cEGs
ztEHCLYqd<VH(NI~o?W{6a*!~HZmyzTmTsmwt`wB&rxf%xP|$}Y2M6|%3cxlL^vy&H
zSJ1c66<1Jx^?WPn{rE^Rs#G1jdZC3?z)}C6@{S@Jn(AfSDmvW-use4BIU08nyLMIA
zO8P!29cplsGnj(Dvl9d1O8N*yt%JZOp4JSX@v#w#up=KdAM{v7`_UMlP^KhO-fsXz
z*HWqvpb*mjVKl}3We2?>6(3wwTJ2^vYIiLj9=cCWhjFDD@|oPe*-)!{mD{JNiR#)b
zLf{v3J1IgS(si7v8gZiBG`Go6)_i%WmT=0Y#2KS^Ri+;|XcL<j=WH9R5$&?E?*j3I
z*w}wVyDS^qr7_CLQiGI{eFS6w9?ATHjjRHo^e{Pkg(f1q#*~pArYmk_`PK7nWbfg7
zbW~`ie)&i-3yW3Qj$bR95YX_qZCP*Eu*#8t!EWMv>{^8)b1Xi=Ck!I@(6H7kbsOL3
zJMu$?MB0E&Y9PwSm_=G=4`QH9olP3Lg4Xwj7znq%k3j~XF^%ktiqIjg^Owzr2f7>b
zYRQGFy=oz4oApQ&e@BdWrj^fIz`BU0dHFnL;l3oC!i#zJ<RbNb>5<GJExQf~>wcig
zIyE?14-G<A(o&@Dytj^|qqzV#Lt3tiwKtMYhIV8ddW~b3p}!U^8pO~)fp%Gj-qcHJ
z<JpXMOd8el5!!pjOh7#byFVibJJar)t9LgZqJ4_6lnzD^zdp8maqNJW(D6lQ%XPnc
zfp0wq)8aPTjgl}grSoYqJzuh#HLStISEO=fp4^!X3-ZWJm8@{mD%D)RGamM15?e&j
zr{#AIh=LIuJt@gmUMjg5^guVV%KYUi%;gl&q!Ba)ng&2F4JAxyeG0q1Z+>W1O_S4G
zmC$qirsHvp`=xl?uSoX(LjYeid#_ND+50E9dU+wgvs99+hB8UJ3^+!GAI?g9ehMgx
zDcz^(ibos#>iLc~UR4cjE3zxqb{P2$?!J92-K49a@9^XPBD1@k4!0eT+?xZuELQvn
zSfguW^Ed?2aezXv{WC+8*&KLES##@X?RiG(fRdo3-WCjA{wfB-gO?~Byxg2mBS*QM
z^y~+^n`HR*UbQG}$pE0q{+AfzzQdO1(6o=R<yl@mNu2o_iYC;_SS<Y%#F#~zi0=<h
z#P<dvB57Srn@*uxSCEn+6;~xlxlMciJUz*dS~SvKQOg&>vq7Sk+s3n_mRt>0Va(w|
z#B|3s5&YYqOZmu*Yl^&BQvj54ND9E#?&UtBoM#MA(iL|v{Ob93FL%%vR(<=CA}_)h
z2=cZa#w$5~JZcml++*H)v=jwaGo(mUO=T-?v(l>L1Nr;T1#Me?nvV_Y>hv+_eUw_t
zbRWY4N0C+(H`=H?ijms(Z1y!V5bhAx(hlLl9{_1tUSovA&2SXP1DI2Jr-^RF`1b7%
zTr};&9oW2j(jAP+%tQv>iATUZzSp?FZE)P5+CJ{LY$23#l)h@FByQ~8hF1rjc~vy&
z9~hkUUl8Mv_$s(ZGWlcqz2topEPBDGKBI1v?<1~1M8iJi&{;*VxY3k-F1JtGTwuk*
zr@EDkRvkbRd35ZF({8o=AgYAe$~StV?bCkcJbUmNf;#I7tdhn&?agVHCKE4NB71<p
z=1!QJtdZWEsHTWbm1?U0K9}m;FUFriQ&^cI<M~{EFL~oBd-*INWV{!F+@|yS|DNC1
z7x1{x!4oL-Ie7WiyCSy?^L<q!w<#U$a2J&wqG{~9LbNlSpR~nLfp5KyRZQ1UWQS<N
zIjUG}UlP6Uplc1vbykjeX3%9OlWsgKzf}aOv^Txje_lH<vVu8PwALI3^1VH_sn})$
z<PE^}MY1JueHIx;CDP&p8t<RFipt(Mj520^{VHElnnfDJd;7uAlOM$1O^w82M$+v6
zHhX@P{6vwjVLoSW*>8UOns|UDo%#etFg*Uw6|vRQ+#LT{RytzJ=)p&c9p`L(ESmVo
z&6xLn10VV6aPo_@2e!m<EBvHIO?*@~5kr+v6CaeNo<fAWLY>$4H5T$4m~&{9{*a8y
zOtI^2W5|~nACkR=w1fjpuD8B6!4El*%4cLh)FS4muu0}IQcB>-&jbjIBCD`(NnqA3
zNeam=>uG$S><8aCYxyKtcZstupFU%KHZ>94&L1G-voIQ<)4Z3YL?4^u--SzkDXuH2
zuLky*U^weMQ;59}5o^TcWL$Q#WTdZur610z4+cJ1fN6$^z7-sJzu-XbmxV9r!2~4`
zJ)dhq_;f$y#DefsK=6|VXfgzWnh2T-3j(o02ts}nW1&9D&MONbCKu|Hk>4uUZ7kHk
z1n754hUHDV$l}O_`pU<t9L5z%^?wsM6;1VPg!tU}>8;3@seTn#C_0iOvUM(IXj0{3
zcG4Bk#qg^iL@p+YOxoH?3e+Y!xfIB~Et4Dc@Fm*QF#-Ex0{GRS1n|9T<q~a80Gmb#
zhJAA54I(kSlE)&cN2RLTLN|9mUTl7=!Iqs9>OiV_by(B<irPOB?`BjVIR{*I^|nj+
zUoJ;D;i<Pzh|jZfj<Rd{>>DW}Y<y9{YsXa_uOrU08i}wyW4&`~2b)xN>>6&{6gLjB
zscJi*)3Pe@*;sp2i8vz*JGzREGI1=}emYHypYyeXO(U@3a1F<QJVLY8Fy7OEjr#4V
z)s8x6oM{}Pv1;QD2;asGH!!R=4qs}taJa_q!1p7diY+zPJ8WnXhj$1RXrrnNs4yvS
z+d&<7dub0;I&X`fVyl*qvjZytfW0Seqxdr0ODB)v0Hew}<<xKn$P_lXa9cRQho5^C
z0@SQ1vFkRSgcMe2`+o6`24tWo*??Z!@Pa*o&1c?q`=Sj-?qrjWg;~119bx<Jbr<}@
zCYbF|myXoz<@PGzP0^l2be0R?+ExQ46Bvl9O~ep>ZWG+9V2f~{PJF?}uI*@L{_c6c
zbL>sYhZQ>r{Gj6CJ#!r!TtEOo8#a|?(}L#)lo1Z4!w#mNq(FG5p>}Laz|}aY);A>|
zq#Cn*L_-KhIbac`EIVwkI)Dotr(Jt@;)FPMO<-5C!AhlO)44Y2hE5+l!$zTUE7%|x
z9C2$(6tDsA6dkrv(M4=0_Toi)1jkZUVqpYhc>$J&+5t9uA%N4ogo`<izAkEocDv?R
zZ~zr*?ZXK%J~resTXHF>BK6{8^BC;82UwMw8`2p_AVgw0rj!H<w+F<8*gsY*UWC-h
zYLykY4&rdk+6-Nx6C88JwGw+a16mD3z8gpBVW`CIsUfzXG)=Brg7i0VcFpO1huKui
zZt4Bwi5MIU)Pl4FcTx&=y%R;PaA|Q7#o<D00}Hd73x2Q)UAm5Has*Hl>$`BA8g!4&
z-LhNrx7~gm{G19_swwbPqQU=&0#ES}iR=YCSa9uVh4%qdu;@gMIy#%cVJOfdemK)O
zI|!L<RAS|Xf{P9C<xZ`eY+ty6-HX;Mj=SpA@gMR#Oh>A8*dCtTs>tltWhA|Hva5o1
z7?x!1!K#B@o2nRip>N<aXCZ1|z>!m%{6Ipn|GqE-n~)yA>NZy(1JW)WAVU$hms8c;
z1`dxyUzSitIw<f$2O`s0#$vND+5iTQjRS!;*>zRax=^uk77=#Qt>G+0ZZ9sg{qpcb
z$ZCB>ShG#G%Whtz0|(i-uqIGnNFsJO#(qtZJFz=8mz(`S3X&DU(887@CrEaj9Y~bY
zE>^r4&{pCbI!=jBFGTe*Diqgc?-04xo9Sl``%HO<>E}P_=iBsi13j3*Pn%6pE50`}
z9fio-ZSppoy#6Myi=^WrdP)Fs?92&Dgv!mebs|)5*AXB2aw7J(YO{NsTNuJAcP>*{
z_cXy0X@OyNm<`ecgsF30ph*YJa5VYu?tGJNzNxuKTLJRT3qAVW&7;rZJo@OyqmObt
zTH^1~CjcHTMD=K)sz)o8JX*=*(c3?d-VS<{Wb!Bp)n<E5+2!2ai{qq(YXyfz37IHW
zPYjz`LVCqZA?uB0oE?{_dTpvOu04^cHlcdukPal^!`lQ=2e#w7G%Wc9?j8oT>bJ=}
z!naaZ7lH<nf*&7yP?kT9aeiCpCtwV)F(k|iq$-5`iw<Kunke2K7jqHb?KZ3Rc8yeK
Tm8{-mjZD3?iWJGTmlysYk(<-x

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/index.doctree b/.doctrees/autoapi/lmflow/pipeline/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..d8e429f24ff475a0ffeaf827a7eafb405d73937c
GIT binary patch
literal 4032
zcmb7HO>ZQ(5!J3VBaOZ#dp7}g<-{=T)y9^_NOB00g@GVIE}NYoiIJPp;B<4QXtA4Z
zk}ZvbBuEYpRse5i;EMtOiu{ZGs5~~idp=fUBLW6$7K_EI;;UEWKYIWE&(TKx=Z;KC
zqL^K>EGtyTjdwelN~@d~)_>)n|I`1?pSyu(H%3)D<=$;TBPAg-%?tmTcZWgU3TZBz
z_1lK0R;Z#|PCe?=_GkW_2Cip?m3(2!9m#wm)y-uoN-jmgeNctw)~9?HTzFIAOCP=J
zmkd5wPbY4WbWYQz>o>lR+*WOIRE?S2W!7pjt}J2<F>Q*>1pl{F$&7&z-@St6@oOim
z%K2c-(ksm;L8o%tmQBNyrz3ho`}CAH>CC+vlFIo|S4F|~@J8vY;qPAk7J;?@;Gh{s
z$A+x3ij`s*Lm#%G4@Hsj*`+p!>7ADkt1)f3UM*WlSj;8;2r<yR$kcoIy^r5B{GNw+
zkP<EGD9+th$%OXq2f4~B$$$0Z)m*#Bapq!mg8C!tSw$bx#XQIny@OQJUPQ(GSdUwQ
z+x=5DjuuTk$Hz;e=Hy`U*D-C;B3m0xa3Ey)lc-J~Aul^{VE_!Oyi`?@y_uQ~9Fa<`
z=8HJ{p0v>uclMp9vYgnF`*59hkSEJ0>i^|O?kDROjG5sHlVTcx6Hhznuic>5zh{H<
z@7%yv%jTc2)6Pou|9AuS>!04U;MqF+{F=#%S+yAJTIC4{Vb$}saf)KXK?vzm8h^IV
z4`I1xR$TMs{)pF}mu3@tPrz%>OXtbGo_(;+Gb~t*`Mq_zHS;c@=3<@VwUqbz!QH+n
zQdwnuB4lI@8dF9SBKn*p6;zTYOV>1`ztCUl@7(R=Y?krTl7IOC^vTMa?LoFvsmA`!
z^+{wIn!}wtXiRs4oubOeIHb(K+;F>fFEFZ*x6s{1kC{_fDEv?9tS<On5WpWm;Y%#x
zj+z?3xY4XEx%O<42E+Q_FB1o-i+3j6u9-6xYe{rP+E8k+6?|x@z9pf#v2%LScBzZD
zFg@>#zR}@WOV+LB%>{Y`?Jq~V7*^NX=p*;G)vPcQ7*<76YBkl|#Maic$}$;K-00L9
zufl#0++kYj&^VK>P42gn*e>0U;ns>`irc_R*WM>u6_$f=UH{K0VPK`ZAM{ma%gXu@
z9gW;z%0V@u!ogyLjG^ILmX{dpX{9V&1=MEVJs5ho7hOoO?-dkwn_;CTMh7_IB(&Zz
z3s-sp9=G0|PGF``mM=`cb_e0w0eB^|iPE`$?G6q4V40v#gv*S9!6#9|C9fsPjj)uo
z?H~I!3!3}D$n9%hDs2<VuQ}5B(CreF-LU6Q_)H*+mf6a*N<G)h__&&4yTco%3uF-*
znzI~|0gf!?A*A4+D3G%RgrzcQc{ix+<l+|>M&SGD@?^rbRyrY8WD-yZ3=mstaXE^a
zU$D}vt$25~GP7cW{_75-(YXWd(x=N4D=Xb`2mrt{g#1VZ<lTW}#k68mj&x@jKScSv
zw=_?c&Ju71V1sSKtz_i-sVXLj5$O<Fiw5eu!$4ihT;v`KVC^8Q@-pf0h^*EE0VwK%
z$zbrPmekb{j(#pG4X-kl%<_PyV?HYtI_=`h8X~Dpu7wd1uSZEH3>(W3NaN*pTPLWw
zeFS4r8FkUY=s}~FwB+d?hc9emxULJ67r|PF^WFo&;hf?S%y|-u4X`PuK%o@5u>RHY
zGj~|>GC_&NRu+QU50bj;hq`b-Bx_6a%i$1Wb6MV6QpKgxQ<T%ApjkTr&BXE?$769E
zJCI+z|MAa}&qF;)SIh^=-24Zvc~mY7c)jBKQgAz|_W&CV$>tK94J;d@gtE{FcH9{N
znI~Okb^#=&3<$Y%$L6-MY?hFi5-vLg@NP*JaKCSIhAR7_LP;AW>aBVlhlx}vi+#zR
zrNrP54n7tevRtPStFNjOTZE@b-BTJ4Km&pV%@-JGNZ~ZF1gL*SP2>T)Qn6Tx;n{HI
z0U0!_4CQ+Z4~TP%Jszddx=MJ?giP>g3Zz?mebPL8fK@q(gm&jPd2#L0c-&5;64Wk`
zh)bS8ZP>&zJ>{S#T-#6J6p&@n&_U<u#(~P+>5?=)NFau_NjK;RKE+!G^6emu-{<r#
zeM4Up9xQ_(wYMeQ43CCy@968hirx+1-=L|zz-_u8J?R*|z1w&Fl3sL8vTa^?ni$;m
z{cCiU&rct0Y&7k9t8JCIr+fOOatDj3sQo-h1+WzYh3`=kioey130agxJ){XJ%EwA}
zK&w^XUJ-WMgC&cDcGr5*>5_erxVGUw^JvvpGlT02s42yZ1NcMZN|ls2Ld1a1wEde#
z{uMI(rT-(!3~y7A4PKK5$ch_w47DWQ>%=uU_L?gbz&H)K*v$foHhdGV3Dx-We>JW-
An*aa+

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/inferencer/index.doctree b/.doctrees/autoapi/lmflow/pipeline/inferencer/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..54357cdb33768e59cb9b0fb06105d485b5696b14
GIT binary patch
literal 112250
zcmd^o36xw%b*N=YBh6^HC3%sT7G5+Oq>&}dV2@=Hui$SaZz4E&Y4!BHneL~a?xq(l
zfQ^^$Mex|0PfWu40kgyjn1jLKu!g_~&Pjk^2lI{D4E&ft0J9{J<llR%>Q&WyuW#3T
zt(hVJ`y994yjN9szg2bXR@JQw-+0tf^Nzy*f)&krsa!ocT9}xq)+W4WCs;gDE4HV-
zYOC{x&c@q1H+P1E!A9Xgv(|1Dy-sixlqi<Um5GK|?cCT2mSgc&xzZe!^%ph0Vyj%M
zhSf_8O9Q1vH+F_ggTca9xmEFuPr-?m>B&m%z-YZ(_bTP8H(IVvdeC;!Yjm&{R9djd
zJBZk@W$)|=TnFj}sNL$6Rt1Ckkd)H=U_er^J6M2lI=kEBg<z=AYBkE^?G{jul+G_t
zG?A)B#Y&;sgqQHVuv(ZFPm3qob^XV9p}4Q%O=6jLu&7m-!cU#jiqg8$($Yz#`K6PC
z>$a3?)83XwyIS=cTMpD3`?frP*JVJi`UenfZV`xWDYRR)LcP32Fl38j$QI2Ic%1MK
zjy9S=-YHic-5xI;6)dE821f<6OQ!-krS-tS)8PLz;QtNq|3)MYxYH;TLBqj<dZFCt
z1kaeRO|&cC`8h@y!Fo|>GgGECu)EbMv`fz{ZPr}aT{;DL4RRa=(G;c{g?h<AH@I+5
z$;(|H3Oz@h&o#@_^-6gX@XWQ~d%aOBdd+68Hkkt~yh=_nrCgmF6?(GvNVFui2$26H
zAwq$Cp({I))}5u3OPfF-ixtg8?V@6B+A^ck!eHI{^@<d4A~$|0S8Y$%4}lWo)~{Do
zuw?{(0OzXacz2|N!GqJ4v1Yx12D`Z4@S0v@zt=%rcPktv&PxE-CBn?Lovu_bf}snw
zeULS-wO(m<72Vl|2C`QaC$_C1Jzm)=jJGR=M!s09<ePP`Xy`~Um0m#QgOxX2_tLF5
zzjWKp>R+Px#Tr~to+*1A+qP|N2O4m*Jw07$9NGy63Q8ZWmF^WB*r@Tj6Ra}+$^+@P
zQ#^AB{%f}D^;)ClP2?vEtwPgl<y(j99)79Rrl!2c(2#PF3!@E4h7AS-t*J(>U61h+
zeK9C#B_*R5LzM&NiB_p|#Sm)KYk{|~0`a|0DNHX(84bM1Yyf+P@!G_pL`3H|Y6lWO
zEJRlpuTy%CLMM2^SpCpgZQLK*o6ODJj_%%UG`Y&oy*Ciwhr#(St$Wo8a6$Ql&=D9Y
z276`#t`hg6O1b4V3Kauj>93`Au1!X|qb>C_-7OXR10p0ix_+n=4Ac*82G(^vv-?1y
zeF~)h3Ftk$I>lfB*a2Rx6AZg{Y8@D4V4dkgx!NgxC0GQ1p}Kd1`G`ZX1pf98LYGi1
zw>tOG=L3aCwR0GHl|y5qXW|D4;0O<mQRmXxO~6`(#+29U1Y1+90Qr@NpnsoLe{#h<
z;QL3w+SZl^v^YOviqq>V5|kVEQ{RVv3fp-At8p^K3Lk`m_XEHWDSMLU9okn9a~D9@
zm44|6h)4eefX#aWo4>HJ8Sai=a8783!ez4NE{bUL{<wk@Y|qk+u+~qjM*QPo#4U|<
zK!DOl>7RffY0Q{3;217_xwlFBYoHB|(V<nA{?Rlx|1_xQZf~V;m!2qnXLsql!FtAS
zB!Q>^b7Cj{Dq@CZD)i1|GOuo6E&e$u7BU!rIX=4CjK4$x@6&AjOR1UeZdRJ81jkb&
zjfXGv9<VUz&bt>17T}nn15*zW>?@)4JyC@V)))huT%}f>5@VA>v=g}tYTq8MZ5Ars
zRLUcmTcMyDF5W_;Dvh5te=De|PUNke_+VQ@(3YH=g+otf^t<R{ISZ6_8~z=-U;lQg
z7NwJu4A-Ow;lt3^e|cDgP&ghuT0=RB9{#lSveM1i3=C<FW@fM#Iy-SxTe?l*8~hN5
zqk8g&&*fIRP$|DePTcf2Ik7uXZk2LV<^5hY2mLPv8k(a+L)XHHABO)e_@Ju{4PE0+
zj^(Z|G_EV}havXXt)p9?z4g3p`?uXVTxzxI&9N<83XOy1{iC(U)Rw|{bBivsacF2y
ztyVD`i{fT~=kOE0{BA|I6jNaTJ)CJ6tj5v)JA~8dXDF@OxG#!WiZ1hlu&>jsW4ldV
zEUD|<(NA;HF7yj%Iv5jzS*QkUbff<(w5=p-bn8YxPjcw)b-Q(=Ux#{Gy3sS3Zqy(W
z?S5vv(gb72p`!m8Vu;<9;=6QR>62JPbfuq^l{xQ^<ihaN|AYYEwJQzIjav0y&9uEe
zZF(3~X=StpL!o{fp%bRfD=37S$N30_DUDfJu+#)Dw6~k89(}aU!R`tNtniP*((8Og
z2PXi*Ni1c$dy3+(yH3?vs<<FB9ELoz+nJDl{VKIO@KnH1$N!jMR0_X$ZRn^HyCT*F
zCe}#uhSwZ<AW<KAHfoCSk!uq1GPL^~RtMeO<yL5eGvsdW@;Oj1i@Q9FahJ*T&YkDn
zOuVr@C%#MPIj_bV!gG$u${fR$=iDHGclVs<_c6FqzOAn|w40iUFoku7bL6H9sd+yW
zZtBv|gG5HJTp-gfn&g-+ITFoAUC}F%W5N|Jw^1;B(5qO{yZNBQ&<1A`-F(oSpk5Xq
zw2|>a5klwA`P@lNv7HaTOXqyvhc$%rxm{M~IIWz|%LVZ6&Syt-(6!fQT>nG2ha<H5
z^WcS=7zHb%@RUdEuBm?$X^?eNYQ%<Eq#ChB(dL~92C44F@skkILlQ-I5%JCIp%h;Z
z0!MQ0*}YsecI3a6#%MMd{b_ia6f5%IoKV^fuNdJXBQ_$L150rlF}=92BSztRf599N
zN5e%Jt6{=LgO|0dl={$A%4EFJo!!#;wh`JKFlj<-!jREmSc5`F&&U`u(#_1^D1ZA|
zNeiJy$Y?!=j39-fQLg79ApoOA`F3ma++loGor1WH;NHweTb>Cl1vd5-ZINw6qAh9G
zCtehGy9o?oryL38p<mD4Ah~$MMq=V{NiSmTKC-Zcko5A#)u^MQvp+TgIittFHq_?q
zL7FQ`eBCXlTaSMu)XUQ2k1{=e616km?I+k7&8Dv)&OnG@u`(!w_4qMe|6juzqH}qk
ztnAXcTq>~W+W+?mbR+wAg5$fDn&~cPHQW&3w2My)ZeU+_kwbgfmkG=6-d?OP8%}7r
zdtY{XLTS_Qn|+yf+Wu)!ExG_QI2raix;L|4ZV-82$f{_6h8+phpRL$MSz)=0Yt<HH
zT7<n??7i;W9!3<qu49@sB0ku+JrvfUzHJ0h8damT9$I3jvJEt$_kc8-U`K*M4sXm2
zm#dgN25D|gmfOZokk-_#oHv6U;oKleWC5gA#<Q6)ePlU)$v4Z8IoN79thC5=NkqsV
zQo43XKcV`Vx`~;GDqR!v%ouki9QRtjjF#vRL-{0n>tG3>zZp`@2!&u7Vnj(6^tY8d
zBDLE80k*i#|4!I234Rj@{-|J(D2kGk*rB*}Hopy_B%{9=(HFQcM%dla{xyU4F@!dz
zH%@zyno}u5hNdBo0U<cSc_sF+G*gn;i3*aBvczg~EhM+n;K^|HYZq{q7|g47qQh)|
zHC9>YuY`X%AFt%NTofO2C?1?J-LACCc}%>;L7!j~{0V$I44%MWEZU0eL~yK;LASr`
z9VkNjXD3*p9`l82u~cgqJnZE(FwaP({+yN<9-+x3-9iE|)F@AM{C@)tFa{Qg_t{ar
zg_h}_!+#Vik!ObFVi05}7($JMoF>SUwo|Ku0hz{BdLmdT5|v5=!Q#nwr9#pd@Q3oP
zT9F3M;E$i6cXvVZ6ulLCye7$YY{Z-;|J$g;B3X44Ncvm&C*?F?E*t4^QTTdyeqys%
z3ofYU)LOXa0`|wlB83I$5tF1S@iO<z&qHIuB47}ufBAdyPclvO?`mzc$a_8<EUMPj
z<8t+Svq%g)94wU|)!(b2ge0*U7FtoQsduUzwuJB?>#_yiicr#TN_MGUOD_3m>#zK8
zKx@GT$?1B|omGb1HNL;{9|N@P{&(1a!gvQMt%-T5qEqspG>aa=Zs-nFv91G)VgKg<
z*Z&3la9z&J!$zpf8EudIzcKLjZ^Q3m4%3ZF-7&r{Y%H;7@#n*Le*yd?VkCQ$I7VQJ
zE65TH)oQH;2@%##ieTOK8(NS|oUc_Yhc;}=Z75F{ro249!f_tpZjCD-ks5Qrjjm{3
zPWAurq=|VCOH45DK)knj7gky4za9Reu6s3MjeS)eExa8=eclBj=nD+cAMF9U9tZsq
z6SV(s_)BSQ(1pIgWr6q~g3qz;&?XQDOdRVqE0H0lrrKW690IG3^sQI}GJ3*_p>}SE
zSBkWm>(@~2llwJ|QmkKl2~-mOntv%g`j^3vrRn~EgE!DaU4j1$_|L~e2+7|i-sAIK
z9sf%8KGHvp0!LnkC8>Y%uYpHLP8dCsYJlyL7GQe4-5dE6h}D1KkB2}1i9eI!(-b@%
z_Dl3zreB|a_t9^Key8ylQJH~WW3$X=b)C{UWNP-?E2xxKazQiJ3mm^tD1?%y9T{-!
z1y+ZRkezLd_5y41yXXZJRoc4G3@i8H^%F3d9ZN^jk}X}M6tnadP${*gBP3<%_^d1)
zpCgvO4J>^Xyza}=RRc;hY_BWFo?R{dTEvzteGfj7rPI?~Tl!f#nJG6+a{}1;#>X!l
z^U3-(_-z`qVVJjgl{}<#HaF$anaam<xT%j`$dI)EHt>wfl?U6d-*Fm_UdD18h`bFu
zH|2JOBMBP5jOD^2xeYsas$tGrHA&LOIW4tFDS^|_pdG1X`xJI)fbh(frD)hO(!rNQ
zNaT37gX5X0;qu`SRI;0L8ZIZ`YYvyK98BpjlhZQa)B_u*p>hH?gQg~(>86Epp0dKM
zI;5-8XMYX%6|+x`3R?Z{XUX)}L!+@lri{6~52~h!xxAZ-`xX&{q2(pe5cvos7;iKM
zpU+OgXZojLMZ{qw(2-aKsENJTG(Dc3rf+7U$?kcg<9&ZCXn`5>N>J<B84KJC#3hdf
zG)l3tzywrEJr+PnYAk@yYAk@yk+Hyb7z<3o>#k#gVPkXW`4_>i*ydO+T-c08y~p8u
zUhShF#nM;FQwH`5HvpQmNH${3NS6=TqfBEF)z}FZIkun|tkS1dpXVQs?TZ2Ch&###
zf+t20w7V>~oT;_Y2Ip}M)v$JpoeuT*6;uexLBLe(&J?@Fe3Ufo7)wJfC<f-cQ&0IU
z&<Yd^*Vkfs+~9~C@v;08$s>qOgxeKoI9Ys$1Q`xLtc;@+x#5>Y8nWeUx)d}YnH!GK
zfmrw4@F=?79hb@5F`Ku8!4MqMP?-?1d>v?#(F)ktO`nYrl=y-h&}Gf&(#FtLYJnhJ
zaWr3P&t6K!rb#Tl&M)FWLQRleAr_I{=hy9ke4~*O-OrDV<i>KlaG-mk+P0JvQT)-N
z9Q=oiq_B5c99I$UY|1sf7q`m|IP9TSgWYfi_!myIpjKhi)5u8JXH)Z+h<*P31Hfh>
zbWI>4ohkv+WB1Ik>Rw~A)|d_t;_v{N`5!r0+%7Qzizm1#tX56d5!3{T9a@F*u)%%W
z(3#+RS5agTgX{|6oEccUv!E`wp@s!<wGDPQng}Kp1zSuWD843@;=W2K?qT=`?c3P=
z?c+PQSR*=$yRo!T{pWxs+#wnS3!s@79=2kxqXlb|wOt5yZP(A*KHnR`j(214!O{XN
zL{iwM)(eB#(zj^0zAe4f6%|>kY6?G*hQhFW`XeCoW5j%(BM)NhLS9%!$cy#T{%yJn
ziue|S;@=McsL@Aole5=J!`0B^2qW?DF+nb~K*|!Y?ok4)CvlGQIMg$Bp68)7o^L{>
zsPW)Cvmu_dhVfhkoc#th=-&tbgq%IToBvkI*1dpMsiz}Y@#IOPL2QK4_#ZVeiW$oG
zEFx9dXRE9!Te!&PH%nP)EEfZ_kA|AuSQVX{VJuih$cGoHVJs-pBx7lhjHK;99a!wt
zcMjV&bZi19NtV>L9M}^xOWu%1IKH&1(?~9MmxS*!S=FT=f>WtM+p5+vgB7KkY#FVA
zJ0rNU8t6;gLY!P|>zqu2R}^Hdt!cY54Q-)yT?R~lJ~5r=g<)M-Maaq7x=^&jy7($D
zlWq*AaM*htgg<+k{eG+>M)f^rhkik~egRau3IC+UMV|B<1OzpoRFk{Xm7(8fPi&JD
zeYT8{BH(lzh=M!VO$L<l6t{4KD@O<~gzha<;-pAKIUhT43e)F?qcI8X0Y|kh=E9Q`
zA!kyiPbD4VXjB;uVRY=(hbV7*C!rW417z^9oyYM`v;`4hF$!v1c^7Ld!)~b?Zx0)x
zn8S4t&7YOPU9egU!QiM7ufgt6CZ=g5W1+l11N@ysc@)7BYxsW+^<*^9S^+ReYj?|a
zeJqRE7<p&cf$F|RvJP05ZCzhApp4OWD{{4OFx1eb#r=OkGHA)~8~8*^e)N=e$&Zxy
z2&9BvsLd_2J)Yxa!p2dn1gUSUjEyf6Z1X25dtsZg;@U!gDm6q2*mlv57Wz+NBV?i1
z&zgmvXjo`OnsBTh1V`~~ywX1!cB)a0+3Ax&MHV|<w783%#@Mc`6>CJS^?I<@A4xW5
zwpP`E5@rT#Jq3v&Ydsa8$Xe-XuB~;4ju2+)IJwgd3?<S)`nFLsB9&mC&jR*J^E69p
z%RCngFo~E>Yq{IQM#wU6nl;N@lg!j)qtsZ&jB*PQki{skhI)yer^#$(k61Wjk2irm
zZk5c-Y>%peKJ2lCbdWuk@rmq_p61*hE43nQC~Xv~`)1G{HIV-8(QGR|l)V*&ExJoM
z_E(W+tOedKwLu9pQ%`wI*a+EUVNUGPATQY-HI^}ZJOl(}vBy7yddc=^V59623rFm+
z0QUG&$-K<=s2Wfr%+NnRfOL>Oei)y~9_eY$?GX;Wht<I*90)jr_NamMZ;xhMaeMp{
z3R~KvQBwDhUy<6Ngqg`6zZf<`_Bb&o_GploY>yhtm_2?D2*_fOPeQ$9do-|7_K1Ze
z_Ba9d_<6~^%=V}nP$JA=kH1Gc$R7V4pU58RX|C<@++wXdS)R&bi<{ggl>$KDb~vEg
ziQC}n!O#X(DQ$mi@ngT^(__L$$o{5g&Hh#y<5S6iWZTmi#%ymHAfLtdwm`i^+fz(d
zc83Kcb~gofw?Z;4v)!o%`mnppkOs25=i(FD9X-vp-Ca~{c!icnS2<uFh}Xa^=}s1v
zgFVHi+607IpY2qBGzO|~V?00B<gi@u9aG8Z^6w^vBr*5o5|M+Vvc9d%;Z?Er)H+DQ
z7j4^nrG_a5$<(*r5H>=#?Sr~ndBo0?5z2NBZo0a|cAoJf$wI}7@Ndn4{UPqc>TL$F
z8O9B$3Zy=0W;V|oQ?TGkL;3GKl+PyRG0QD^63V+RD9Kw6U(Z9~w&Y1DR+jdu(l)y3
zp`bK84o5DHuMK+E(|masi?NxOXJ$-_ZO`F-7PKGEx__c426XvF6As9R6rG}d;{|LL
z*VkH5kRBW}<W0i9Op<qM1+m9ECQWKHgWUFCO=v$_JeKU7V0*K(W^$6*uSPv_mzb<Z
zHrB<y5Qxvx#Xbu461&(0w<mks;1mw^;n==@A7uY007firCakZR8Pt#y-71#ww07FJ
z18zH4s7~ZCv9Hj|m0Os#hz2Ektni|(wAHi;{ZL|-?7>a_K|NOaywgC__ccv7!-}k0
ze3jXxc8b#`$ZBkk4M|s!8@@z#MPrnGsa}ihO>x>-B-vsiHK3fM{7X~uyEIgoX?Has
z<Tp?m95I`xbB4l^MP1?NSX$%1bch=xcauULkRODay_V#STmh^i=<V)Ta3pDD;vkEj
z7A&&T)*4r~TlIEJNLA=#Zg^a&??x?J`Ate!GExR0%GH)ES@GaziW1%@hY7i|l$pN@
z&&1z|W-O9uDBg#EYwnJu;jXD24IuqYVlb~A*CV<!!~#XvjKa{oIzJ7sLVwT%MxC2Z
z6S0cmt+OVgI8YOXC*`XdnJcKk+F8v=n#lD)yK9<=LPe&D+}SHj=4u$a0R^fHQI6Us
zU#_qZh?z;!@_GYM6ce!~KG8&sp5}TYHda=<QZr1z9P?a45PeU)z8*FuHra~(wa`!Z
zoOf-?O&1R4tKI=P8EBtZg>6#c{BjrMww`s?rW|C~Lk=PxIyRP@1XqGz&KiX;aM9*;
zy_JUq_com2(eTD{@CY9_ytG^e_j@y(InwZ^Yx}*tIC*5UQkWXcjl)eVQ0THkrRi<T
z6-x!U%uifb7}lSLO#u0+dTT@Cgm5sVQ#l~xjdm9S=u+RooqNJYXaWdV7}*oRFVbr3
z{{x(s{g<^%bRiK(U+meZ@c}YL<zJff?+lridgl37sLG8QA`^=@Q}NlISARS^bq@-3
zUFOwa<e_l$Y6-<<Ui}CUg_~DPD6>7U{xOTOIUF_T)&2v}H|w2<vo){&0m(bXyjn2T
z?mtbM9Mu1WuC82~xFxaWc4_5Mqp!7XFI~cURW_=$7^*r(l}_ZLaM~}SxM=?{4~5fy
z3B}^&eCRX<1LKh+`lh)!#zHnFo=S@eN~Q?Vvq&~6G&KdA_prpWhh_*UaLn1N@R*Mp
z4Mwa}9D^>Wg-x)4gL$?1X<_47Wu2dbx+V+6g*v!-xm=}d=Jl4?U_Ok!lr_?`R-QXp
zh<L2(WC2dl%<nH$+MZ1!oRBFF<jXhV=IPdp+cmt*o{ku}&7{094|IID5o3}AT4EGd
z^jhyU-EU#(W?j#{P}R}(e1M0-xgH6{#q~VIL*ZPHgkrg#g)}bj=6vK=&B-sb&@;Oq
z2`7^S`T|KPg#${#>?c`rSre5|iqL|JB6jmOzvQ9R;B)Lm7wK+N&@yi++y6Q!&g~{G
z^I;;V%#NP5lC0n%y*f%yZ{H%w?8~>T?>Xm8axyEUwG_$m51RgAo__9xp3|YKqq}()
z4~26#5{iqvxrB$pX^n(pxtsYg&+q1J=$q!^AG44%yBdO$$<bUxvPt1+Qn0zs63aRo
z3B|?H+`>cQ98C&Z0v?Kk%2<wOFboPQ+cVy1b$K%nHG|U;B=_ZXPVEuJN^&`?Vs#YV
z312k*ALi-joX%fCRY#}u86FDfbR-lPr}K9_6i$OA6wB!>z`$WQuOmKcK7N;lo!RXO
zFq!<$<0PIGekTQ|f5%eG`W*?y#qSI*V}~A`-$_BsaXb_Ut+D*hLW(mh+mnB3JvoDi
zmcic$Qv32Zr|?1XBzLpatfHv3UuvRv@<ek^W(=x2I+^G5P&g+ep}07i7x7Ryb&*gk
zCzE1iQ(+-zb}`vUHYJix3J;Tl&3CcHvK~f4aq%!69t!7SQqb~F9*Tp?SRQ6*L%3X}
z49|R}&F+Id+zh@(5S`N3@Wo?kSP3g&vy+BpZe7<ae#sfm8p&At)g2In&y;=z(n?If
zG8P+E+SS_2)jDxQ?~{(FTipYZ#XIHt1vt_8c6Kw}>g21Uho)hwRjwcogj1T<qm)*@
zAF|44dMh)T>Xn$EAZr4~H^!2w{PEbUVSmba7nsNKsm(CPp?w;|e@la*FHh+E4OR+I
z=<^SlkUt?$WTR8;_Oz>o@rt(#59PDY;<G+L3R~>F9d+~`ETmIQmxu&0tC3MUJ?|P1
zR>5KQd}M@9`4cZ|l!`>T+_V%Qscw`tGm1#Qe$V8!i+hsTn>+#)$y^!?+m&Cf1Gpkz
zla_#wTh5IYBkHD(2s#!jMGxM>cjhX7ckhL{VaCB*uLadzK@HkR1g*9>B2Q&h=9K(V
z>*#0%8@scOk|hrXLM&Vk1zK-mmZXXbO^1>i=jWusIZT^+9k6c)v5%+GNZ`XNg5$i9
zj08Ru(o%wACuOb|hro3+Ik2oGtzBZ&O48fdchKB~#Q8!RhAG{>k*Khbx59UsboULw
zmOl|}u?~$}5pf<$A7byd^bbaG;H4=w7GZbS*zhnTnHPWR&2W*>dN^tyF77f-z&wEH
zE~179Zm3cxWPr0OA<vRjY30a>;g`E;&Qh}wan(0Hp8qanh}g-wShZ$sf_P6VCd&iS
z-s!*+!=L>Rpzp0zB`M)Ei%cGJK){g6n?eMfWr8>-$z=P;NbhbDyHE4KB#S@Ip`iqE
zK8o@Z;ykqnB8HT|iM6`!7Bn><yCNy&Z#Xd7kn+92v9D5<BPHdZI3Qq1`OP5$&QeC4
zg_QX!M&2v>{lyN5?T7Ik6va_quUu5)#I{#Mm8Zi|EAMpViEV0-t5R5JBo+6h6)942
z!_hD$9=^X9`B#fnTs6H=bl4K<or-%lpeRytx8oC~;?mPxr{Z1+OHN*4Iv>siGi2mG
z%WX?!6jI-bxrd`I4#Rd`yvPRfZgD;mO}~YlTTH{v!&C&4YrVRagS#o`RpE>Ty25o6
z<mp19t(si)Ot(!;r(GGvPK_$!AM~(d7idR!U+8$|*5zR%q~rI)4Aa_O3HP_o37ce3
zW97mwlDLT8#EKbOpKwU2ifVQhLrPO;#O9%Dij3InskogHYozO@%Hw@$cB~Kdk99N$
z5j$)f{iG7W|LAKihBss<=ru=xAls@FG--K42_z`;S`+kv>;&D{KS5TWR$M%>&p?Xx
zPSf$Z>~wspe>#S=M@<k)^ea$dzS2Z|D?1U7_D{qzt1C#LBcTA05_zks`DJ!$e%3!V
zv$*8GWF^;?v1x^ipsM3)*a<uoZZ%9oaaj%9z(e7dHY60QXG^u@ejW?koF1jhY3^gm
z{TvdHJ^pu*rwgaAWvOKs#3ZL(7Q_lX6mCH*h1S$~C=Oa<Er|74a=)2}mSGu8kZMnL
zOnr;&nh--IzIbdIEMcrVJIi3karL9Ik$UoKn8r1>8g>b|<}9mW?*gGEu7<@Y32IU7
zWMfe*ig<i&?0%TDUnaYp#I><fw=g3HTjcClHIQ}o`!G`J!A#o!YkZ>FFFhTJ*>87+
zGbm8^*|8KXq&s+$cG7++B;d;-0rZrWfQR79dxcA1N5)?ZN0PKFaL@aGPwR#2HMQKy
zw}}l$rtCz85xMJqX8e_j&DxjOHVTuiJWb%~MC`e}&h`f&Z*xNKbe-)_B9$UtCFl4N
z9v72RK3CpX|D6yLP4zzrCO%Ww*&emZ=7dd@<GtSr_*#cjtx41m8g;{<y31ERxL)1^
zN2i046L7S`C+AwanL=BMjA1*5V4gV~pXx=w;v7jsH)80gvxKvQvm8`S5ePYnirYl2
zoQQBCVVaa&1_+9O(DK}oovd^FC(F3EA}*T9M<B#_qbb;vor0_SryzFBe4LKNB0x>-
zy{4&>ou*P2n!5M!=nZ2-$^`qJl_90S5{OG4Ick(*Bgb!n2X`JhmIlQA9|%bf8wskq
z%@^O}sK>bL<AX2&d|dVti3310IclA}h&Q~=SC1*!+j`#uaLr!QxUF|}*S3t?c<;md
zV&O3P_VV{d07N>U(wD^vZv*v1n<4eY9>M`bN}mEQyc?M_Zr3!YGluA_ne>ODp5Egg
z%tru}s_q@ESIWh53yy<}-L{L8fZKLWY7h@~o9+XE4>FDG>an~!eGALK8JmRB7Nqj%
zYZuSI#+w|=U0-NiSKdFiZQItZqg$W7^}KESx7|2gYPIUku`OE)jf3U=qqWA=mcn>*
zi!LKpEx^?KYD2q)FH8K?xVAf|P0Aj$D4W?-xF^zJs8ZqnH&#kcg}d%Sx}dgIXiRzL
zkd^Y;b{&~}YDMGbb+wVc_DBMwe%yowIe9b)H#77yUssFAWtm5P&H2^1y%Pb;BO|G#
zdvS^zh;7Nr!J&0$=bA+5vZR?v6EG#gz*og+Four5Ph}J`{o=P!DVkD<@61l1d$$et
zZ>AJ}0$jk;)S#VG*!ShP#~x#2#Cjxk*C1ljJxNP@-t#icngj@9>7xbhbx%)2W7wZO
z3@l#{HM#x}rDSBVU=<-HUa&?QA`15BLa^33f>V%Cn8J&tVYKDJl5A>q66_?C>Ru2;
z$4y!;OCu4L&vX$9!_ITWcbN?8lfd_#)Zh^|s2d{~IT;jU;$l$eW|gzDr|b=AbE>AH
zFf=C|UhX4?^BghE39AShIhzxTHOZX*tKTO5+B7^@1``mwZG*vgnGEJLz}b7KLEB)u
zUluKAE8PodmAWs26|W*PRT!~xF`8@_LQCwkRo0Y!EDdF$v3wSo{Sjg|&k4g=u!@k6
zv$3E^lZ@s6o7<!xOCua*RgaKdY^%a|nXKw>Km=c-25qY{mfm7>3FXV#GFk(F9>I;*
zC}~@WlZ$O-yboF$SH{|!wxf@=`OeZ7TG!`+>HkSgXUm%-JFFt)<ZN9iT47y$6(ctp
z2ZR4igFzlKnVXE$?jF$*^EPQbZvUG&a;E?Z4?*T8t9XivSIkIm@^%oS>oLI^AzI9N
zq`%vwcSVQT>{Ta;>n_&)(ibvWzKOL*qX<M+4>h?WDB64VFh1%=TZ>#g{Q@)_5lf7h
z*Vgqy*18-sZLbX*A&ET-0oGXwH3w%pa@>e%;&e{lNUhP1g`;-^Mb6=9iam>nbRFuc
zM20TQbP1Ctb(;~^M*NJp^`l_WS4s9J#;siv7gPhv#BFEgbVv2wNEqcV+=EY)yFgD_
zm;OoZkH8((mkaYM_P$FjQCBS6z96G*3*dd*^-_}*ITtsVh5fgoA_$wd%W8Xmhtw4%
zYP$<?wCA^kjgUP*Hf#3$j6@kp4=TBwY|9$um@WS)P?N=$KMwVJ*|IGLWye@EV#kkx
z9lu$!HM1S729$6!*zs49D6-?P;S<?0J<YWpU)3zu8XlYwpNG@*#(QkPNmLU!9v%{a
ziu#VSovF>)!l7?_E?CVC(_e7-eXG_emPYr$LHM<Xwf81+{(Y-f&)4<4t)pjc%x&Ij
zRhLfUZRQ<UXJ5bLG~E3&mfIj0vSH_@+>Y>o+h)B`groGsGPw;qcf!V696&%oZ$MNr
z;^I$9rByoTl=<_^NG00~vFWP_j~nntGJk#)LLyi7IM|lyGJ|E74AitL@>aXgMXR~V
zhNJsTz{$#-xu}0lkb<}N5UoyXRu;7Kt_MUx#SHbQ=EBJ=7uYnK6QC-{z~v13XYx?E
zoOKDsC6oF*9txL)CZSloTWVUb;9&eXZo81?>lG|6)*}31!<(F_O&7}5PS4ZjZE()V
z$H^=Sx}xp5B({i|awl~O-q;<GNq{m_c}3eAYTr(mnyG`FYb=*p?cFEPbkW|Mc_^Ir
zN+>Sc`$`@Pr@a!2rM)Q+a()X7-4wT#DbnjUgtP}an~HibNhpP)reO9*S#sH4OG0tk
zbn$r}3b*MZ1ub9ap*SdxwdsO6$oacGqzoG_1etyLmSf_Br=*^6Ar(^{q5Mnx>wn`(
z<NV{Vp{ip)xOfdW=I}XBBB8i=lC?Y(P9Y?eemu!A3w=&J$?1Snblsor98xg*B9>g%
zlSn8op5!VX3g<~u&~g(G#X)gwPXhB8WpMNHS({shhm^sS2r^T8625pW&o(~VhYLlP
z(tN8SPHgKK7jM{OH45gqj~`#fGAOeZ**IkyKOP3mqQ+$6qlSZbBL(aHIL~G&V0>C)
z<Y3gEBqm3gG|sUI)AwK!nkB;YVc<ru2vcOLs<ywK;Xp`y8{Ff-`ya>|b7B<A%z~~%
zBFcWR(aMSGVs3c%m3wl{+GGnZb@CvCkpdVS@m^kx*c8Ugm2#`>HFF2baGR;T2!;-o
zH11%*vq$Qi>$zrO8j^;qQ=90LUq~E<Hp4r8XCJWkL7#OtF$nO5tBv0SzJ5y91ir=(
zvK&5wCs{s{21S*=`*p0;C4Kjul)fvZOc}1`9GA|Tyd2Mgk$OPss)k6Z23-)dDOanv
z%C%~tLY4ODn;moYLZbjTbb1X^2#Dmi8np^4K&}C|JlCeHUb8uxdtP~}1V!j(p(b1!
zxz7`MvSmDvX1tAULso9X%eAXzycl{D!Ku^^L@L2;(RHt|uk1BOb1y7;Rr8X`+zz;S
z(%bwj&^*$*T)CNRH@%6`-1woKyxR!xpTu)HX>{0~K1-7s67gE!VtZQdPc*FM5RljK
znx$H0Vv`9Ty2adZwWhyrgp<ms)9jt9J3KPdO_H`0<#^c2v;~?~W%=1kuZ@1DfFXu$
zn?^7G%T($q6KQ`2m7<BX_%2hz#gD)mpP~lsgbTZ$Qwu4kvKwR!t+k=sQ%<vQ!$`kO
zsX(YMDHY2sQK;ci<gIo}>(by7I;9^28;*yXT)&LC8wo~OMR2bBAPB1%$*suj9}8w%
z$*qrr`Dh1_XhgH&AZExpBpJ_&2*pVjlvoDLsLR1M7o-uqN?tmTWNs%f;k!(>^iLq(
zbErYvmR3YOzS7~?d#%NLA~-l%6vB3~sE_~GRVKx{|H>q2DSK(89QvxKK+VQU&3LUa
zd=*v^n(6GTP??i_)hoHKe$O0q^YQbt<xnKar@(pH=2!40u|U?{X@sKu$Q>jP+mGP8
zOn&62!1|X{gMIsvw?=T_HAeQ#2;0SvRAy7$%JOFysJ7L=O2f&}R(}Sf{0kB#&w0aE
zv5Jtsv#p}S2wUZ=7^#&w6!`%SMR>$yY9&sIxRquc=orVoJ_)V=J&x}vwX*koZjn66
zSqVA(Tl4?hY51>Za^I4gTsq|nK;OQT;JM~tmL^A6g;R5w-I((zI9&v0O@#@+-E2*J
z)mGCy51DXIm}}U_s9$VB>jP56-K83*|0Wd=jkAz{1@inQsd3U-$a;g@GgA<zWSy#x
z)`!K%+1`~^<Io;|4TLO&nkmNlSVhpz8<#Q8M|S-Sv5UpZH|!Cs{KTP(v`19z6!!Q_
zXkFVQPlrr(Rv{Xd%Q7RW&SM}7*HmYX5N5CJnm?}>{a1@~m{kl<5n@x{`yA%00Y{Mq
zb`3sJ78pIvbr#s=6d3f5Ncz{a)2Dx7RMP7^@9SJi1cpVU>0Z_LbY6UhOfb07JqOV!
zO6$6$Tol*45BoEj&b1HZZ4Tg^(z%+EO11}Jm$MKar_^d;Kpx~Y6+$9|7yyH?^SK5D
z^OU1ie`+6kci4gm)7!W@W^g5;VWn^lyULajjU0A#OuVuzW-YNUz+#F}@rNe!Ei9Sr
zI_AA35~mcc5Aaa9rD+MpWoh~$9tyX@D4|$<OBg=5FR^`@#l;N!rAm3RamuvB_5~7~
zt<DZ|wz~$C0A+eIMeB{!zMZ1AI>}un>FN@rpi)`uoz~o+b>Z+ECm%eD{u5Wdu%taD
zaCOPk`u8s2Bot1CB@`DGUUIzcRwH8KR9Hf>RCqr041$eu&82UeohP%9O?jtMSb~yC
zPfq}ZqC-&^VY;yS*(|YaM<m(ol2>sl4~5ICNTD*<@=zR9#>%Uhw^i94JZbeP@K7@3
zR0tANx)r{7ET<x=|Gfw1-WzZ0;B8O~g9YJmNuRML?~7V1KWZM_%JLwyi3l7r*~rTP
zqo|FTcqrE_(<PGFF*_UVPU@0oAO_0Gi4q#ySfa!@IO8md67K<i^-7dbLqU}xaSDfe
zEHz@+z&w~wPRP!pXKut9*AfSEla1OmT$(9zA#$}zV{sQ!A(TZrmev~k5oe8sAe9OP
z>{@~(JRqyWYHE0ESFmpVdc>><tL%`u0Sn#GUDppy)QY+F>lKo08G#x!|2b&hb#|*z
z5qSr@PIjH$Dk)^0-9Coe;^8^V-@+%F-O|%s&u(8(Z+H_h&&yX~e-3PQf?Ifw$V~U*
z)HB^UB7M(#UlMOq9B4e2ljzZrt<BNWWPZXc!m)4-alG;9^DetY{7q-YsauE26S2Kg
zILNGe2L$Jg`7^2T%+8yimS&}N(rMoO`$(k}qe6t2V&42OAtahN=VJ3_m~%!|PZ@sF
zC+&(>p1_T49p}?Y=vnjWNz#oO*A%HJ*0!nbSwPE%tU<eR4-&4yQM1NK>`a!C><IZ8
zP!+l)m+8)S9tt<zkx*R5;TQ2xxUs8*VsUPBDhDC-Ha;;**@{;2JuEfmxGdEaDkVOd
z29j5iW<>l7w=Y4kpC69n!j}vRoI>r}L4o7MRX{QSoMaPln>pSf2W5rS;CN{LMoZuz
zD*@IvS^`5C+qjE|!r6v|;$j;e9tvk05{hLTcH}5!!uWewTrz96jZ-GYeJ6=6h2jd{
z^ryI8#t%t~yWB3Tv^V-%tMHdtR<kPnx5Nu4_x_JO6i$UD6c-i#ArFO9VF|@j;S|xV
z-?ET1D{S^?*3%@L6ndJ1&BvU`4XxN{mW1LG&C2mmxM)@iT1I&&4l3gk&Dz03$q>yF
zB&KvLeDPQ`D{1sQJK>w*<QPc8Tp&4=qcM+#a-M=2N0v}d5k%N4l#?_gQDL1CSCZl(
zo*c$U{#nkHdWLu)hl6HJv;_et?67qOChWN>xMK)*6hKaYIFbo>uB57?`;G>y)~44p
z!#!spIW<{zhQd9!Li4WS9)*ew_dMG*+@mC-rq;G~^-gAZ6KWKNd;S!kDBMF&a~<xv
zw$O$QDQ^n)M>Iic#JWH1dC3&>f!#Bu+Qi0A@VwN)o-Szg9r1ZbmxhN!!-!@-WSMGn
zN95RpG?h?sPBfvyIuYi$E?BD}Uf1v1UOyy_PHC(?JaCMO+#fbVA)7G_*;p}=Tj!dM
za$IahOJ-p5_l1xVNf{gL$oX?n)iE&g4IT<Na+XkBM$X^op>V?>3B?*Y+Y_ynlfPfF
zxR||4s<4KQQ>Mw^&q-{y)i}u6?ifr0)OwkWT6~Jyx1$!b73Mo(9XqJ!Vl!)?s-re<
z;GuBZETOn)^Eo^ePMakZOPf;$#JZ?^%2u@3elAOmsZOcvAtgSU4CZ3ej1&eV*x#4I
z%+&Er`&cosZeWVo;1no1#6#f>L_%>fkh^&(oPkIvmVu-^p6Ol|dS-?0=Xj<!l7vzy
zY6@n5m?f7Dmq;is;gZkrP`GeO3R?b-hvJ|(R=9*Yp6PKOQihO;AhR#ua$J{z`w@x6
z{+g$l>w$j(RULca1t+onJm+8}6c-0`3=f6V6A8s~FexIjr?ZeVJD2Q{*c>1f&6#3*
zixh0$$r8)D774}0wLG7P!nu|dw7iIi;-E5?YngYBni#{A9!!;ZC>h*}ATgy|;fu#2
zu}O38?0JQEvpmRbB7Ni)-T@dzZN$Vw&68waVG=uLCosJ^If~mYG?N^i*67Eg)9-~D
zV3z3g{lLIp(dn)MA{C<El8U%^oO%ocyYH8?cj%6;F^+c@Nv`HGP0#J)>eJk{UXf1K
z**3aOnfVcGPdkf;oejl|Z(o3Hct}>Aq4@Tfpn2E$wn9b5x36@KZ!1ZvXrb-1ddIiF
zi#q9HI^*~7iQ?PzG}rO%OUu<sui;gT+4k#&5$&<m5$zCg-!bg3giU2kRy1yR$(Fu0
z2ItnZ=fF=Drl$*GY<m-&i*+z53-YY9#7)NbJusMWBZZ_4+<2o+^mo$El#<$0ILDCo
zZ^A|>q<u{+r2Uy}3jlvWJeD&QD<(c$p~Y1vvr`UkVzUgYI!<g(<)LsB8wtf_VzZft
z!i`cT6l-FWG6;91LyOO0sWC@je{iA2^GP#ob7!Yjg8g<UW>Toem@*bWNbTFP__6E;
zE$z<|{t}}(<>ah)+RYVNL9uk_1tBhphA-x!a3&<7xR}t(c_^F-NhlT<=h%c^$5LY|
z*dLq;y^1s=g$W7v_hmvuv^Y#-ha{6&NS~E4$iK81e1v5-YX%=8Vx3~}pW~r$W+0)s
zn88<hD4ZEcD3%$_htZheTRg*@zG*IgpM{)RWeG~AWuhlYHYpTQFrY6*oT+7^MW<vQ
z?97L%j)R>wJQPkvB@`DGJ&lLLsi=fvsc6b&qH|d2nbowPWumPlp%jXmg4tKH<g&3`
z3B@Irdm|5pi{+-EWgid4L2;~DF0)K@h=-IRo-4@g%eO4Ir@o0D9Tf>XI`Nlw(XZ#p
z;-*ZmhN_NU<n25Z&WlJWE?(p>c_^GRNGO&UNwF66X%=#350ZT?=wXsg3jdLU&5yCf
zvi?Itaq%BN=Am%@BLywL=Ak&KjO9PNtOYGNmF=oC_zpp0U%umnu3@qyH?rI=t^7&!
zwRS8g^E7jg<pike=vbb~L*X2YgyP~@&f}qQS|Xujbu3q~kmta$JeOpX!m*@av&Rz4
zIu;4V#j!MbD4b(SLCbAC6bF^D9LxNzXBn3Q!Lv5B*YJ=sI2S=?O6S5Ck1aGOMZ#xi
zW%bnL#Yepgki3?v5ss~;{sN-<S=Le?1j+SUOEuOa)avQ_9t6bKORvF&!l&ih19(<`
zz4YX(yarw#0^4`w!U)|-Qh=L9CcKLB{5?u({ro(upQZ~l*EU~(D1TN~1(av@#+YxW
z!BL$v{0LU+a%0S6bYl!LjBa+YPw>dq#(h{<9u>F1w5x^jinj}yi+2UVjY8!~xcCJ3
zQZ;i^Ue#+9TH;U|+*3uNTp<d&XvPTbrh`-}z$Gw^+Ws=!#FZ=5;qhR38t#X#0aQ4r
z2Vu3~1{kkp0HdH8fq@{9L|^Umf-`<#3)paZ)Ek8zDex9roUXSviW`pLZ{h~wOp_35
z299yl%|OoEI2^f*zy<NZjd71<0o|okuI9W#v6L&rfz8yuJywYWg74+&_B0;y;{-6m
z2iKcs@*^WXHDFI4(wE~j>w+3M>t4c?U#<ITB43kE;rMkbQ;@gQ`~oUPZ>Pa`=0U36
zI|w{ZL7zLl>qTh#KcyD!qwv=zTaofe-9Bqqv@B<%WDjE^u1H;3gP5S1gA@AbfIeYO
z9;T*`!9KqaQ7RJpYD3d`N*X%D)3L5czORF-T!IwJN}pT5i|Pnj@$%INWefQ}CFE<}
zh*KIFar9<3t5bWjB*TG=XDr}J=95?kFQ+kT8Y8nmN+Sk!sl*NvfPJY1zRP4vH=;<-
zr52B{DZMa)l9MSRCN8G5vp@MNYwEWdZBV5&)P@Fi6LQ}p?(^(142tRqi8&h-Njb@&
zE*cq$^w@jS_SQ+J0=<ZAHpxgLU$Ku7Yi*jHuSmmAWf*r6BW=ULcbN=hFN)@NYSA`~
z1vs!#3MBr}c)ue8j+d&uSOuZDDE(P!Wuvf6Ur|%^KpKid#V#QC{tBw5=qad<;JCA5
zla!Ja`}&cQ$QTlt=Q^AtZ!LzS3FF5u8ZY5>(RlMKYCQdIbOryAhR;gl|DHH(YdpTo
zr13?R*;lB=h{od>T<OvR`a|o-QxR}Z8jnz1H2(4<CNzcq5o}TO;P+{G5ZcHD^7~(j
z-#pU{=TCKnjGS$R<eFq7H^EH6N`%z&AmsZInFB@cr8y=JGRa=L6$8WR9#1}Hy{-5>
zQ{}wmM5q*<m*Bfh7Bq<xT}v&du%L4y;CQ@cpMg+ZEa<|%C8;du2(+Xv=CU+2hZZx1
z{Jog?%QMNa7^)*=;%qS_M`1C19eq=en6vzd<}5sJ`(cU*WMFKFJDn7#QTpnvUDqM$
zUxzxVm@v^Ck4-d({VC&J5Z;wViJEKm-&zRmG(u1@t%g**q`R{&uC-fmRw1QS>7b2x
zDS|(-V3i8_!0o)^KT@}z_hvoEf`mGr0ZPO-Ej8OzT5rG9ynA&j-WjorX%yruNQe1o
z!3IaF6sFGQH4$%1LquqWRixn0i2`2J4J)KNf@<EPj3@wc>TU-(h80qsI~?jrD@1Kh
zVTJo_E96O$H)g32wb~XiBWH?#73gr?grgCnEe0=j-Grl-2NWST^}RRY{1|W)n{fUQ
zKG7x|dYbD^IM0U}omZI7hpQR=oi24@>b*Eo_<e85IVRd7toUF*5INxcoa=X-hB5iE
z+=jw9R4Ei&8+LBW?SOcHG2bjt<(u_F(HqN!rE(i~?u1)kuq7CZEmoC`Y0S~ju*XhL
zhkOr3D%oy@UDic-oB^v;dU-k4e}LwRp&jY|Ex5hLI@)Swlyl0x#y@C3dJ2ob<IbNX
z)U3m-j+R0!or)KDtQWWMm?3FJ>tfjG1}-|O!P!|rJu`XzF~-{DbySUaY#sGqL3~-(
zQ7-}k^;$=b3s^0ub~yqvN=|%L^%m?dekWB9G%dcW+Etb~{JUDnQ8kd&@4g6$_i*#$
z_4q`7m!6J<-?cjdQyUEm9$%@fE-KN|Y2ZqNqw9w{!NOvt&}?=*vkiX&(JqAqltTjO
zDJuaF%?Er=7Pyo~)r~lEy{4OW1&iQ{*9Klmf`0TH&_u9(PpwwDT&;mOa>GOdtct7a
z8tUb`S1DJ$(XeZ2!103{nGV8kG13tb-Z1`GxciR*7B>3B{u9Q#;8}KiVuNNL7HylG
zaubDCK|X5q=tW{@u}yatL}r?vLdsTq%-cbb<`mnhn|MW}Qi^T@;ic#%ZV4e#H*p7c
z6Hl6aUB`$}^%!g!&5E~%u!KU|R1Nz-2e|$(;K#BI<?DX{AHc~ZA#C+)%WXwaI7VM<
zwEm(8TJzB|7r;*<TEFg&mer@7N%Oy+3Mqyh_71oqU6nrjRt8=4D-2i6SDMbxvUIZh
zcs>bLx%jq{k=(}jaVl;TF&G+Q0u7OmK!Wi`Q}9@J3Lfd7f)x?pkw8ac5uhgaUeol8
z>@@u}3r*d-F%)7fH!aEBW@TvQ|0PhDJZ#c9MLSvl5L9v=HkAha=c8>SBspfH=es)o
zmFhkA?cYWdr{tzxci=eZrBGf5r{d!r^R$h17ZsXbb1XMjKQsnuPGfs@54u+vI_SnC
zS;&}LoOYVsXXtC|g4JEyDt%h@kN#$?4^tO(1sbFQU`qr*#1ogkECwnA^+THhZ+oNy
z94UPYxbRG9gYyZ7?5s_g=Rv(-QNx>T9jbesU_Jt%4Q$@QdZk<}w>pRE^MOL6+BsbM
zcIk=IcStOy?^@I#9uG;e+=*gAu5mpdR@iA)$Vo0hR9vU7?6|yIZj}p_@=G8UtOen=
zy>@@HHz$TZ2g<FIJl-1y8SQBpGBw505l$FaMq4qMf3*Y~j&0^0t;aUO$|yWi6lk)0
z2lgkC25o7YqfFH1qW~Ik3$eiEfZ!ySAnmB4Xm@ucCZi9dWHh4_au<6Cc!#bkGzxG4
zh}Q@&hB04h6c40g#H0vMx3E%|)Wse1jyg&vkx|c0%SO8}JMls#b1Q%G;###;C_~a9
zDo&^txrX=Rb{P(c$CSoOtpNYxq7ybF(iYY5l$yT;{%U%y{sCZz3k?y7NY_Pxv}ec+
zs~*jc1z_fXqKmxQh}O#^iF_;B36`++5QpP&2L1{^>syI|%WZhFh?O0fR0Q3r#?!Fq
zPAa{V>v6A@23Q)<3XJc}$pUL*W}e`y(aZvB(O!>Ro2;$MsdW3SQTjjxCEmHnH9N!v
zZ3-q6Y@Xn@^a53wXQ`^GdN>VLVXyHT<nlwrWu6VjdH~fClHsLktOt-Z4-*{QP+%Q9
z97glbA`N3hN0DSJ%Pf#drjk$sEG2OO|LrtvSE(P55hLu>4}6!&P+o^3d_=JJKEp2s
z$9MC)O5wT}&}#Lo2wJ?BNMk{4T#RLV7MUvSvsKoV4V-Q>oTV%@nAan>=R;L)pooJK
z!(gb6kdU*%kW`Zlrj6-&a_2@vci~*u=fGvBZZERtdTatFN!HZ09M}{yjT}xRA7xpm
zlVohm!grZ0tAn!6QHypLw2tAnQct#w*23pRkYlw_C9NY)F7|a!Cdn%bGS=3#U6qEm
z(8At`T;D}p=b2$x7}XJya<(v%uCOq^j<G~d-PoPvYkAz{5;cy*draWWM3DeVf3ftq
zy&Q~OqE<5zHTRm4CF<Kj9Ii{$8X=0mz6av3{lw|dr}1fSm<ih2)7f011)=`&cFTis
z{TAnHqXZ`swAZwVxJ>V$y>Sb!eG;MnKhYXQ2qShm?z>q_aa@FaN7x84{yq$R&q`n`
zSZzx{veF2bV26120-%4Lud$4Uy8avpa1M2e7#BsmJ_+^I3Sl&+pe`bo>w!L&MQn@+
z*nMbqZ-w_=BVaBoeX0Rvhqh0V>tutWZrTwju~zmCq=Odw9>FJC?4zfwYh|RqM_{dN
zXE@^MkzHZyK`uPv^lW!C3_<m6ic?`z8P7JQjScdAQajmJ9Y?zLtZm!Xf}6EuV{Gh|
zHgCMqJ><`%)+wd2I~7N}e=2N*?Ed{|_trwq4dmMTI-a#V`EtZJCntg~JVO|5GW*!|
zpLtLf_~de&;jugv?i30M<=qR6aOk|88eqYbR)zIE6z)g?3B_7No8#jQ&t|DH1GrWs
z{eeL<lI~vte53DTi*6@KNtN>{a=2cCG2VxuC)h%oVOyeuD%t+tq%g9S`0rq@pm~<T
z^mm-$3t48fxL-@eI+=mTL*dLoLUA#JCJ%)(0|~`4gA~UZ-pN8X1)54_(;a7MD&lP<
zn-q#D7|@p@p4ijTTkMT)=#YI~^tJYK?_-(Fs_MIm9ZstHQ637XsuGHes(zk_!l|l+
zV(~J?fn<-ekh7~QY!4F)m(99YEKin;uy8pDE4?s&Xji;W>4{)5{3UX|X#A~w=GRF!
zDO5EDn}5y{%O;pgC@$v){U;BFJ2xl=Ez8bi2P_UMW1Snc0LPzdG$H=brnZiU_39-!
z**XK3)HlHY8*l6c$2QybdacpQH>V4Y){EP<7Uatv#>vucP!@VdL1;?<!WWNaXCw_~
zXJ-<7yghl*X@8yMiLAynHj(`ybn#gxvKN8Kx=v(`34}^+J+T|g@pRVr<0SXTfLGT^
zu9@c1<M<!x=6jUVs(3Z4il!2xLs1zG_an-?WmOn4#SaNOf(z|l8Wh!js&TB;Wug5F
zz0gkUTFIxIB%Toz3&sx(g?sXIkqufjM@?}!kKMl$X4WY^@IBSxIW)DT1^H{dR=ZJc
zj*W~EQKR%yY|A1rkaC@MUl}aGlb1`_S~hGiCtfPo!8w-O$4}I9adyi&t+m1)Ib3Jh
zw-(nUUtwh<7UR1;m4?YB^;@7)bc-v#%e0>H02<Uws6~4{qubd!($2em)&_oW1SNYU
z5HWDIih-E8tYWP1DKOOqI+WBnzdsGmp;P)W^6)<5Ay1^SIze>=-@EsQ)Xxqmf3k~Y
zb%5a58vciR>c>~ouv0D9eUS*Xm+SCdCiVUZiso+w%dCZP_SBE>N5Jt?l?&Vm#YJ_~
zAKoENE`3F<*iWaSC{*kRk$XQU?(r-z6r1V@jyo$hNhwLOUjVNA2%cylJEkO8-tB~f
zh)?fU44ixS3GUhrw)XP8lwSTAs1)_`_%4$fd<+G-oLaQKyg2<sDp>R=kj)WrJl@g_
z5Q>W#jNx|ZY$FTh+WTojo5s#GREDPU5OVZ<;waA+!!)Rlkb<*mkmQm~W97(*F}i}0
zRgzXOH49MzYl!0CgtUwG$Hl5OBmTTM4QrJDzQKVh#{T(FAS17%I!Re{SE_XH9=o|@
zK((|RT5CFBV2I^mje)aR2<s%VJa=RyI=)3k7w=rpOi*LnNOBE~_*L6d*w%P6`WbxE
zfGBT>^bJ`=`Wj-W-6i9@Oup!o$nIBCixFRhvjJt>(Sx|&9|6Z3o%BTr#l;snAGIKD
z#ksupGM`9;f9PdCg}nM0@rozdP-m(m`0K3BBom>|d>te2g+`-Ihrz-A0g%EvA1B5s
z@?JiOr01ayDw$*6iwX>=wZ9q3d--P&hil%8Mu?VS{|=U6dH!Avt6M;~3d$%#>`J|l
zW>_$+vIZ976J-t1)0}4w6k$I=-aF_OXFh8n1l4!ez>2V`VSGwn-4Dws7dEKhXg_cY
z$isA-wplx_?w=SoLTdH3*y{cQ1TFe|YfaI3Z>=-v_t0sGKQzkQLuewXX`{@p&zuES
zIopjagKlyt9*YhqF=K9`!&=_S(P52gEIPafD9;idei78`9v#*Qb&C#%CF0THub~4z
z4%!zEIOAals)4MYtbqhl{!<g5$dl32k?>@;TaAXKq%T(cAXy5!A9iqSNWg6&0rZrW
z0Q0Z|O+c_hEXZvx!saa8rPm>KLYma4obgGn?_g+ssI;!MRJbJcCRdD@ebjJiM{uG@
z)Y_~LKiC}RP&FW94Gw%DAhOh`Dy7#-8h}a!-h2bz(VT+_2AZ{YqX@oiQ?MEiw5c{L
zbYOiRj+&opcyPW1oShOZY}HzYO6Q8Bf+1DLn}9#tNZ-6*d9mGqJGo&IOSH8+SgPLO
zVGf<$!D84!)+$%0nw?<PM6K9vl`GBB$p*1<0@ix|xKL@s!q*!*Vm1<vt9O@H><$Ly
zA!xM*w1gW%yk@H~U57uGG-|aLpz2MQ4|aliTVNqoppXZ+`=G#({IlJtz+cN+wPJp}
zP~3-KVH4m7gJUM4fhsJz8L#gNmNnrj0XX_Je-JNi=-d-5hmmu+o-fv>X@e#-wNAY8
zDjvPb<37$jEz$xR07+KB6FTU{(uPONd=+m$T=;Os(ZNuuP-*4mp9_Ok-oY~PuT^NZ
zoBHj-32(eT1??@zZLGjQD1j$0!J9!meaJ&%5I<Ux$iubhwFxNI2?pD($<60(h9ikO
z!AbVVyw_;d8u?P8I#KZ&z<Wfs2y4(Q#9YQ`Wg)fF2~LiDi8SGV5)3a`F51;NsM;sl
zA0m~5l}H0<U9$u%ssK%$U|FS5ooW}RJm5HVv%%3`6;?_cUa{7g$WM56uQ~x>@MMwQ
z!2lTDB#;RlDNiU>C=CS53++}duU`d6mqF2%O|+-$`A`{wvGf8~W4uPSP{9UQ&>K>7
zl)ipdZ+h*CTK?cPs?ti>W?XA}a0HgFNm^2D<@cAH<rXRHiu^>miC0koAri|)%1FTA
zmI5(NFblFcXmpt@mY*zFfH?3hYw-zyG`$91BLT9()+RvYOB#g(;h$H`3s#|}f#y`F
zKoKkDY1jb0YvqPudEKk$!9F~&&!W&od@~MPI`e20!SSV5tKJ;jvISl@N9%`LrCN2g
z)|diYT7xAC0VtVodD9@C7Kmdpyx)ArS=)i1%Nvu$Nb~qcH-DyT9(RYK%I<>;CCXlF
zl3;)<*ivduR{+>xt1u2$G|?(`u2>wL3_3ZT4~;V<Kpe0I84@d-hhUHF!F;JaRjR=M
zpuY<$HB|kjw342eOX>H(3cFre&sS>2f@JN<#S-iXE`s=ky#berg;085yAB|h19q))
z3-~YGz+upyS`mCR=z#PK$DpHVw#OsI;L;D!UEHEe#<+l-5vUAYjWGchd<ZZoc^>Xi
zUsHrq`|{p&p<KxqCMMA1f|TwFR$<F1@upXq6yEG`FyE{0@9b`m2aAO_0rLe-ELSIB
zS1{<ENZ2F2Sp`UeX0^c4LK_p_B>M5-80&2nHbc{EXl*k5V*s7$1ifE`FU9k)|9b3P
zUxPm{gGE;V5dI9zhoAZQ^C13w7=K=YMec^5b})$baeF>>BGkfRs(qMh9;RA{sRm>`
ztcQ+<06=HCqcj){$oxaqCzKXOBUC%W9$0mMa4vciy6Mh6xUjQeBMhy?Lh%CC1;dIq
zVqV%S(0(~mL1+&m?ufRJ-M{}R{(J|2UIG(F|K<3zVj=vj#-9oNDdW$dErOr7<Ij`$
z^H2D*WikA0$Dc#^a|`}_4u8IgKg%H7(mw`&UWh+8;LjaH@N*CTT(SawF2|p5!4XjY
zckpNTvG8*({`@Kayaj*0hd)2UpYh}1X9|CwIv#%h1%KXqBK+KsKX{LlkM}J3coG<$
za7Kq)(D^q$=Kj;(03VYld`zzJaiHwuK)M|)GwqKwVi}yPAchB^CE^XbSyfJ`p7{oS
z1HA=%ogS}MLX)PKt0OPN(t}oug9<8DI>u@rddx<v1O9JKv@Q7shAhwt6l-mC)G$6M
z)+!ZYS}+z_1P8Fv`!|4j+}XJX99q550M`hX(F7Y70K>k(dc7okO=h^HTrF1GF!!$1
SrqFW_O<;E=J-Zm;kN#hPkQPt?

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/iterative_dpo_aligner/index.doctree b/.doctrees/autoapi/lmflow/pipeline/iterative_dpo_aligner/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..6ec03ce4e8d61a401552c58bcb0718746eb96f62
GIT binary patch
literal 52740
zcmdsA36LCDdDcpMYqgSm2-~vcm2FugSvxYa%VlpA*|LQqB71F&ZHy1by)(Vr-I|#m
zx_hJ*U<wlKSkMp$w-YeL2dR)0I8;JX;Nq~2p@2#eLJ>kG#Dz&>0s%s~C<+otsQmvs
z`h4?mPtRyqg{m~$-Sgi2kN1D?KVDDw+lC(bpTAqh{`u?OmSHv*%UZ40v}?K>_@gzu
z+HUAgFL*fE`S#$xpyZD^+Hu!zJ5@dKS3!%aVb*Jo-VE*u{7G!?nRT};`j5DJ)idp8
z*xeX1hK-SXf|4=j4|%3n*JEG#+v|<_x_!LdGFy7xZ0cpx(;dw-kLi_K%dTj3bD^m_
z0S*BD2XECEk+Hnp?hm*G!z~T?^#Ws~KUN6^G}idTLateV5UB^V?K#aK*F4WL=h`0V
z1La?1)?Aczq*~Wp7cRo_P*ZF0qtRNs6`jm!)uWC+k8Rrih^H;!SAntK*lbJ~=NoH`
zEq-;6VK?+WPP++`?Ky5cNB7)t;~`Lj`~wWSdpKizw6<q!Eprc7(;lg&JxVq0F`G4g
zvFx}Y>DD8w+jGV$e~4X+shrP-aREqXYy+)c1pl|g{~hpuCrSy5a!e*$$scTKrW5$n
z4ZGH^>%TFTZ?^p1y!~{(=Em@>7iewcQe!$&)2y)-Cc{|EZ7?#T4f29Neju7AW*?gp
zm&Dw|hzGdh`G{Ia$&mq^r~SOSu^V(a%5=EF)8^WB&8bxFdc|$&)h^UI$lCiG?!M#A
zue|Tgd+(Ef10;7-M9%k?!7OH8dBw~=G%2^;XlTxf{a`1cWq*@c1^`{<*w=x-F?Ln~
zEhND^F%JLg_QHbhjE_r<4yBt3gB$aQy#>c^w=&AZm0<)_+ofhlq04cz<{80}aa3j#
zlz9~9%a#_hl%>Z8xHCQgu2XY%?L>xA;B$>*AJ1H8OB=#Q*<}nCA8o<WT1Kp+f9*``
z#Ed;>&D_fsIJGYqSjxQh{(J9c`Yr*&Ct7;5W;Pcpi?Hy-NnFz-2|1%9b++2nV`CeC
zED$%$`3WMlJ=qy;u%0j;{nf1#fj`_jF%1d`^!W0F^YA!K<Og6O?ub->7$k*NGVm`i
zus#Oh(6D<$Gn;|&ls^Jz-ia1Cw>8L)zZTB*#a7*{nqKfA`}(-%G=r1=7)%*o*>Zaz
ze-!#KzY)wbx}N4N=w9IO?q<5#V)1_P9V@j+92tN)c?7_<$rz5N`jPll>m8>H=k#G?
zi}86_P;s1xup2eM*x^0U@LfRg_fu#VL5Xw-OAy^`{6kJcH2Wukn+KK}=^fAr7NE`}
z`GdI11>l)MKoeTer-$^<jxD%Tb-TI1S5qyGks9>AH{Cnl$RAs5)Mwn5hEdyS%h6rk
zIi@p1dIR*8PVR2z03wzX!H*_`gVa?KDxehOFqQ~xLWfEC?2+q52ql(}@H{js>x_X_
ztKxRmc^N?Z;IOUrVGqW_2C$l&Q&!U{+A;1l?!sZf5yysE9+>UjtMbK0MY8Uni|+N+
zG9Es7^GyeMG%S6sE?WvmXC==|g*>5jrPeD!t37&CImZ1-#*B9+&r6*_?%x^uFoh!W
z`b<UF$bmj8nektL%4l&<`h=upy$&rFQ<_cwnF{72Z)&|hv~5-RS~5Pmv@f-8V>8(q
zN1;14o{qals|x+{dq?9L<5}amS>tp5)%jeTN^r#DviUg$0MS2LuVMO1xH@(%zA8m4
zrd%Us0)--gx3k(fbIjpcvv`=#=UGQM@gCitbw3-X^PD4>;9><iS{iVE*5MrEf&UYT
z?>t`G1zKg*z`s=aJ8PG!X)JZMhM<8pjIBUA@N%&w-P#G4)4ncD=p#oLyF_?7bTNdN
zTYn=<m#40`n;xVEE{FdnKXX0udY?n~Sbac+^SEP2+0g4f6ZT+U?;K#MEEkxiGM_|_
zKkP#C0EwM{mB>U&W^;;%x%u+_o!v&WiG)8Uh{*X@`GZi3$|l*E>`dY}6&I>Pn%8y`
zr6ie(;KIuCBGwcuTUdgT)huJkza-!KBxfye^Jp@E7*369GE4qi;C~uIGRBy{4*6s$
z0y|HnLkW|%{u9UDY<)Kzp3VGuF!QVYF(z$FRO@@#lsg)!o`}u=3&&BjIf`s@9!GKX
zrP%vrC-(d^WA%k&rhdE%wZ_0-FApnPvufB*Y>t*PV9elTCRMI*V|DWt$)I)zoy}Un
zloRuP04!_W>Xe4LW}@}XdI5xF%LUX0K|g^%j#`4m5z3y)RMsCBxuWr`Kg3f&W7r>^
zZ`bQglbrr|#j~pwC=uW?@hrRc0A#4_YSavQ`*&3D7Cz-*1>X^~$7X+m&3=`yH~IC5
z!|viD%i)YIX_3*vm3WvjSkpW$Ev3k$t*c-le*`qal4<K|JY{whKX2O8ydZhfA8Fe1
za8h2L=2euF{)G5ao^ON}LOeHYw7zM}YqA}Vg!Ek9^I%yEPNIn47OIk~rVz2%dTD(N
zM)MCUqL-->o8>83K69-wux|DcD$$Fm^&)VYbO$BtrP#GQ2H->?L<x*muUDW-;m$}g
z@T7brUO(V0u+0H?_Y|6fqS-;X49br+6g7NBTLt-8KR_uWI~OyK4keme?&iC89fkG_
z?oPJGqqA7g0HigWwg<ZpiR}x2?_K*Y@}M+2Gqr<F@Q(evr}nY-0DY(jLyA2!6*igL
zv41~A_RI-wlE9A^-8UfIVvauU)A(31`1%7=(hI>+G>DNJBpK)V1IL4O0A<Z%m(3QE
zr%Dt_hjTaYbVo>ztv=_s=d2&ZOzG4+)cZmPl%C4^8E|9$96pwIo(ft1H!h1ZAIq1k
zy=w_BN8o#F6g~(pg%8bg33hS@vpK9LH*}~9)J@puiG?*k$_tiX2jqsp<pE)+l(<4J
z$)}KTC>DreY*}HBH>TS0Xe5cp_@xM#e@2>sKhVqq^G~tMX6s4#!(91U;mXo#H>+Md
zvm6b~NFLxQBK`Cdq+40ipNNyT7T`>Jj7|&Bmq;R(3tx9ydANYqhvQVv6Mhoc;*^8o
zGKR$&5fE<jsV<Q+{+Z-6o=(Lp9z4hi_XDi$FiN6mPZ4MP2DpV~&YRWbW&s3%v26f|
zNT9~C+4@HqJzc@IJ`NW%8cdi-6pN-S56QLraa6%(>jR96^#uG0CwE8A6@VinIb~yE
zUulK0E71h}3aQ83T+&04V*Y}2{K+`S9c+GL4N2wmFhoX~!srE+JVt*WO@%UgK_y}I
z=f{mc9WHe;d%9Dk_uV9~oebal9%w2m^<@p8Cf~K`mw?NzOrPtq52k-*v1n9<go*f2
z5)8?U(Y)MD&I^f$8<?`JJa~8mO@+dPpppj<b(#u=2SFu)hf*95R7%hZ8~F!_=HsoT
z425x$pOvmL^Co5nU96*A*L}duuKbyp3MZ4v?h+~K%hB9?h@2Y|D~~ZH<-*Ee(o`s{
z2r7B7@(fLd!iu1hzzP+Buq35u!1Ohe&%$`1`0W}2UtyEm1p>GR`+$JyuIdXa5Nynh
zAaS7hCW69G$!Q^>@G{d@E-0)XEF4A&DtVx=fu=%1K~Py9C|pSLd3r!$EAZG=0L3-f
z2Plln+8MWUC`)V%6IQ#0_wFVN`Tpu%yQHP#VPx8eNj3Web~ax<gk3gU*D>LfX~{4z
z=t)xYG$Qvpk{pR7POp&e@^tdgZM5s2S-~5SaU+0Nci<GtJFI;DehAJ%>}*ZQnvCU0
z%d%ziZp;(p-FL>*SX>udCJ}e({R8W@(1~xESQo&dH3c8Z64=FX1vWo0-!_J=ORy1A
zvUc$6c>F+M?UdJ%Y-~5=d{EA9*ph9rTbIJ2Om(?ZO1)J2Z;om|Q!>1hw_&UARxc}p
z*s1WIa;W+r%(XqB8m1gE6C^hP9!S~uk{bYiyn@Bb)mFR^YJCs~<XeaFmJy;JOHoWZ
z=8`pn)0MB>`Di#4D==?`0`qd4w!c*|YjJg}%|7BhWA=}LP`S;X+dH%TPeMOo`Rcoa
z5;OOE0vFgTg^R6#i$4;Y1&wBLLEStl_XXtu<i6loXx+)pLF+fs2EK{{?Z>0P$D?b)
zqif;lq;-IOUdKMKXP*bz=d0M~A^b$aVw-r(*88xDcT-r)>KC%FFOKYxm!5$gE66Eu
z{U2x%9M`di5v;#2EJjXv53qhI9E!nu+e*NCiwf3}uGOfH7|)>kMUbfost1O^^tnIl
zfa=(6NKj*^6sWfWsQ*>ywJ@mV0CJ9#VT#!7r$K!V%ELf?E*>#Zv!j&;^)2zTT)%LA
zs5q|U?DP!nWO-46?#o~@1iF)LBY^M4CLK%jUHR>DI1~f;#VY~uDHXsIN>^h%;yr`$
zogh{bj30r1G{%#2B4LjGQZT<5V1B94Zef_q0eXP>9E!wXUd1B@b9S`SFuyffRqPkg
z4;Ke|lA)fVp01B7!2TvQ5bny;ts}U<MOcwEf*$<&VmK6o`;L`>`-LjFr&O;7e8hbQ
z@Gb~e1n`eQKN|4qd6AIEo+-%h0LZt6h6_Vp4j^Z{fM<USg<_EZ3p`?wXGbdy`PXG?
zsQp6ykzP=ry7R1O;42Ek3haLYErnxW(IJBW7llPhgXjVN&xb=X`0rW?_-|LkKRpNP
z6(ABIvjTh;WGk`)ya@g16+kgNauLAJsYPHHECSC8Jr`aC<N$Kc3oHUZL%G-@@N+z3
zivT-X=|$jHd|(2*=MrkAp3$!rpa;L7WT<EOPo~EMke8CBxyj)W`{~vZ@Hb$i{)G7x
z;ZO|t(<=ddrQc7fN{#x6`wZ$wK(HdH-wFL_)Tie~LLPgjAP*0{u*QUj3qxKG&;#VJ
zMWGnv58x4lJUd!x$p1FW=1Mzz!8+VAYuGiYy<bz@)iXAF#`g@D21`(IpFG><nVGr)
zHhb<8oQZGTvN$Lo<W27pcdIAX0)I$w#X|vc7Oetz3d53i(S!THJsgUy0+)fUJ_^NI
zxz*xdUO*(aUzfZ&vG6dZ<c{D(oE^C@n7kONsTIn`*NCK=$lbv@bOlx9xjpNxG!^QW
z6G0`<KD$p-q4wDYmBc>#_FT5f6kIa0x1S7R_t>`YV<dNlch@<YytkCF-4<Ae*?qW=
zswVI*X)yzL#hCv%$-G5PAg7{G&;?7tK1TjIl6+y^Ey+Ru)XIMt13!BoxF;spH%Pg1
zs5Kej=74LmLtJypo2U(DBFRNT-N||*)aK*6_KH<=*|#>&RV7W_Z>U42i0Mqo^cqk_
zk&tQC$WkGb@LeKO@;7&6y0jA?V-mju;4%oIUMq~Ka0n#_kghUmBVtozm<JN7ZO(I0
zLKZ)ri$^SeVn-_-Ki$nAqX)li!QXjTO#Va>yV<9R>NUk9st%0x9AkZ_!@#9bcJ>pW
znJL$Ec2Dsi66gN$V5SO1|4uSlo21){kR@y+&NHE)G7@}62ce}W>-c>X2JHpI?NiQJ
z4u@jV;^paR@k_rZQ^5nDWC(kvLN?JZ5n4?q+sTPaUykPIaH!O-!_<S&HD{Q5J57a(
zas-t;VQP)0LWQY<N<x=w0Lol}*%*|47isYrNjdJ<<3A2t@;5S!N}*86VRMrVFUTN<
zD<J8h;(RNEm}?o8q@_-Gwf3oOcw=S+X~v3gqB(e!oC8w$Kjd=cNx1%;rb1b=ppwUu
zKS5KWELl)VSTc3bJgfL;4V>x{>HNzipK<%?%I+!Rh2Zc7Ho;xMforiJaOldnE8&nC
zA%sKrn`ml&L{1F}hwn3O<*GmYf~G>jK~Tv9htW|gh#-F>ET}9G9L^*8JbmD>3Ha>l
zUVpB|KEPqQfB3bZoDmWQGfYXjP;diHg+hU#k_QEM(o`s?AgC-43hE@Erw<AYHo;vu
z1+K+DprDjDD^vM{5z$$Cf$%(9`yV1_fV6s_DJ7THKSoobtX@#bWA#tbR4A(#R1#Ld
zWM#5LJhWuLNc+!_92d6#B{?V*8GVY)cNZYwN-YF}Je@MtPDR$g?A6XoU!8qqZn;x!
zL|y-k#K_b?cR~zRr2hFbOdY*t+_63*E27g~2$rppUJf<VOGN-!xJD`mC|V;O9g{WE
zF+5^5Qg*b`HPQ;a76e{!fX`+Zg+Q}8ufro6t9qZxq__4^Chf>u&y~`55puKNnuWDU
zcIK9i`ASX4=QlcBPvygh4)XIG@mX#Bxp1(t9ro)~!`wY(x4l-|tJF+~6=Ib|&GS0*
zCtWJlBwFV$1bgdKX>)5h6f13BnJ#VG%d9vNcRUf_jK87?52^rWhz%rlEzTmfUKQ%1
z>x$$R&^2d8@(@jh3SI=2Ji*JYG!-g%5mXYvi+uYHQ{a-o;Tq1CcDfuTnwfc0n!<ri
zm;4k8bLN<3bP?uo_2-FLaO3cC017z6%NUhpJ@iz^`tKwsg^cwd;Oyopd%lmRLcv8)
z$pe=U(^M$92r3D<EEzHP9_v3%a$Fb^OLATa1b@Tky9*F-rS<^?>r;+d<rc=HHqtl5
zFGb7H3#86Ti~lF4fn3?`4`?cs#S1EVEdIY~DwM?wDhZ44V#{iLoQhrwTW;|!t6^ZM
ztD+=VKq2eQ(<zgQDuT3S&qS3^a3BI&but=Bl1WC-1%oV-jP8Tkqmxl8sFwNYWy*ZC
zW3G}-Nv{MiyjftXa7ro%D4LSqin?I8?c9b(EG1<}r$tJd*fcuLQc^W<J*T9PB0|(-
z^M{j%sbv&d<KwxZ&jP&W%7v$Qhp8#PFZ7o1@{Ok~LshdYJtG5ChbEDyY=ZgqDMPiw
zp;(4`bvi?RE7RPH=}`m}iG=m~KCq+;RF=qJ-w|r6>-_a?&^2fN`d*p}6`lwxdBT$?
zXev~ABB&(76Zt~QLMY>=pG{|b8#DsT$4OZW2SmNGS18W^7&D_T;(R_=d4i%?vilZB
zWqFg`Z<2FFhW=mY?B+>!U!ti{*b-FoVCyF|6$)E|N&;I;ri{HOyMt>fcTyN3OLATa
z1gn7AuF?sv)INZqugPwS)H!MK7c&jyN_MZLsZbU#sN}KugESS&;suq2#dneH-cC|3
zY`Mjg-CNk)bg@Kp1r)N*Je@Mht|Hz$eUjY{fv-B*jU>q=yE-hvMUveIVfJz)yB*kE
zGTBx4%O<;5L$bRdyh!0>R}N4#*?j_a!IIr4@rWh6>}aKv-PdE%Cw`z(YuS9ezE3-^
zv%T%Srn%}l!TqFCRu*O}UXpbqCancui-miyiXjhX-q96RbH#6BlD|yoygz%EN*0;-
zs=tha_)mci_9^52L^u@7c=x3<-X~9c^{SHMuHImj{P}bInXN3C+?051Xfz353pLku
z2K*K1nll6bHcf?!Sp=0lG0Tr=DpbrOs3c+*>dy8K0Xo$s()q7QJ_|=U6!Ah~`7hW6
zcM<q-E#`?~V!7#;8I@#iy4<%OUpPT6MP%f@6}sj`!DTcR3I&2n9u(}SsZb~oRF($?
zH<NsxJ}9_>O>h?|;9Beh3YOtTT^>0Jq~W)iUUH?Zze7`@3|~;mWB3ozR4Bt2R1$_y
zzx1ku1E4!aF5p8XuZ4}DCSS-2JjSNE3-jl??1TCDwKnrbQsJbbKhHFf%h10;Q=tr9
zP|0KH|3*`x3|&x37<w1AnV*uB3!826+RV#rZo05eu7EyR=PCP@_c`kd$0BDz*POA)
zB{UVvQU#SfmU;zEg|bvZrKqJIA}OzcrC!J8rVC5u3Mgc$c{*jPh>9F~*{g`9PWO~#
zsxa#MVI)JQe)w0AW)!I(F2cOg^+R=?LskxV<rOJgGu#I?!@m}}BEY)X8$9FyMQer+
zq5Q07_y>5zYKH7+rE7)@__ich@9?t?R`=>t1@Z14Du_$+*mFJc-8uAtZ&(iR_h7GE
zR#;`C%ILyNsVY^iYT!$$JcE}2r>6H$30?3(<d0;Wu&=0N#os05U$``fvSyg4X%t0&
z1T3;o6~qsRL$M0tHGsFo;|&j?z`d(bRpu*qH?$Zc;TQ*T2&qGU5^3qVP_(W~e4l}?
zIRlEX(p0E`LQu&QP`p4>p#lm)C80O@MlV#K>UPZo&O*DPH$9=-><umX2Z-k5$D|B}
zLzDch6v_jCz)Ycw_=W2`Pbd;A20qQGB#VKY;I=-+YNWyeJZ(GGRDweBO*A8COj7Fz
znH)|+*PIBMqNz{_5mfRZq)bzx5F)4~5Yk0*cr8h}uw55V4)?LS>B3~W0t%UIo=%yN
zJL`1dg_;xWg_@4;wVkFLJRF2i*p>I*@WrZw>m?pB+Ob~*g!-wO!X-H`v?3Xb>5W@i
z@-bNRDB5N#tMcW4*Wi_k;H#V?MpEZIkwlrC=Sf&ii{w1-1zpiOPo6kUW<XaJ6E~X=
zT?6y?C&GCa&WGdxMf0JLpl(<`^ie!w`4BrgUGgFI9Vw?*K1A@?b3W7-NFr#Ia4F0m
z%<t?m?;$vm;fcf()iheBzkq95e_rnJTsRcVd#+38Jw=j}WcXHOgw(^_$@J!1p#)v0
zH{XP=IfJ+F(^RP7O;E`by#0)(LIrPvN+Nj6pAvM~&dNVPWSpbxDdUVg=&qhKn4gtG
z!NeeN-PJ9~T;F-ZzF2zmb!KqMh%~N|C}hdRiVC1Ez-Y@E9qv6fGka>H!?I#R)vCt%
z5z(PS$;(~?r=mm3J*cC@NTN)1_-QboBGKU;pxJ!UA;oT<=#c22jSjDa=<u^*{VW_E
z$^nW-hb`0%iw@t2M=UyIM=Kp2-Yue{@a|f8f)7SPeJYXn@)USpB2P2ca~$~IP7-$)
z{AC_gw0p{wkI;`EBcBNo58iy!vAqZC=0a1ai{r28z^050JANN|+;@Vh^(h>DARLN?
zgRe@5gSVav<%vq(6Y(b)o$9FZiBQn4qs9+H*PKz~$7m{4)F`Ori5fpcQ=y_pL1pEm
z#;=kx#2xqP6gB=WGlgV4N0y7ZzPpSX-^ZvVv#P${9Q6}Y`J_NU=Dg+ksphH;)G|fB
z=SNV<1BOYO3Izs1B>{#m>fci&<-#Uhy#9SYFw|8}!xhj6>pW#QM;#<}Oj_!-OdGjw
z<GY2XLM_{ZN*+tSm!?8ls-RNTQXP`=3Rr5B%}p2UC|5usOU=_MlYJ}VqD1z+Eq+@>
z=h#oNJCmbl6zP5N)J$9s9tssz-lv$>xOGM4j#v(^Q1!Cs;Od76cSuy!iFYJMCh>kd
z_{<`S_w%4mI`Qt9EXZtI{SaZbVA&M=Rghx;u80*1r`U3UqAB*bPzx-@{uexADK<M=
z=@k2xN@d=xLvGm5#QL{-xc4m1PS5G|_M~(Tb5XA`LBltwRqL+uA-?MpJ!M{-x*Dx<
zKZRNCQ#Si@I26lf52v%)O=>Ar`RI7mlzpB4xn-yhDPO!pB;5E$N;oop8HTRuXs+{O
z`Z#~e>1x&~nL3P%2VKK_sYFSA3vcU;cZfu}u(&8&i;Ggodn@c<*<H#u<t}&vqH0GH
zF`scHKLq<O;z$mI{%J>&LZ@^e7cD77)@d9Dr}19`h=rYo9H6MvxECd6PD8^Z<}}z*
zKb%Hqo5ZkY7f0f4_$Rh!o9rasFFI7MYp$z$y=XHOzzzjqM@0pAW&rrvk~S%SJ^WI>
zUDv0pcGF`wJ~I2lbw++Q75i!jywP{(J@`IWW8BzmOmG3&VtY<|E)ub1?DMyCPd42$
zTYBAW>eHf(dQ6{&+V3=bA+7F!>=(hHNKbSq%MRtRLnT%$4m2ah#hMrwtEm;!A9n4w
zQ`G}ww||!BXic}SdHDRmmSeML2?qBCC4b1XJ*^%bSq1B$XrtHQw2e{^_><MP11}2j
zDq^fze?neCDT7&m6rP9g!OK_Oz~5N2t8LG$yXAR@DYzE+!?$bowjSN-<qz}6F$A;5
z`dNQ$L5B*m=GhL6GzL#}&|Ob!wBTf|W7{6Es?VE?fj_V(@Yiw{;HC1-qtIYnoVJ}h
zoUQZhYGqEV9>s5A;eU^R#ykwrv^_m``9Xi3t5>~dyHQ!3x1C1tpg#$(Ry13cs@>ps
z!p{1e`4zpcvn!S3re{=`<$@SMNpwI{W)~(Ly=6P_vU~lQ4vJjuj~g1?T`o?C{Ehme
z3HtLiukA)xhidv<djZCqJgzxS&>ysLbgcnb#$Y0<ItqjQnN56;0KBWgu0gB7A8UK_
z(^pNqCLEui{Ia4uj_p(oty!yst$_i6Y*i0zTRZ|I8fA!$68KwE-=v1Ho&q@flYHE4
z@AH#iq&oW>PzJC%*8mmOL8ic8SJ#>gZEZmZjo09OwcZT;GaY!zf>VQcFSPV#4ak^G
zvBLnxc@PscV%DT77{mUg*7od5bje?Bf}yRewHvKUXpEp(cEJRoJ53FOfmwe&yTZ&I
zXFs~sa`iSmqGhpxrnEs{Y}xS3i+E3FB~`C-%ydm|tn1;`4z4x_uN)4!9FZUaz)gTK
zF2I5z4hCH(nxRJo;Q;@e@Cb;wy2DVZH*vHYO!-<zI}Q`hPmc`v8xd(>In4zy#Jbrq
zz2L?TJN(I(-h#)1iWhmHFcWxXP6OT%6#hAe=e69KJ$vA?TW+2347*vjodv+s+1QdX
z04*z?-hk=xU^+(O`t<h8t^j>bI`h@k@OUK}{t`JnK8OWP_NeZZP2HPkG+-&%V|a}^
z5F7KfIY3d(GlC<d{uZ#wMkNGiC_u$+p@p3EH@GL7p0-#q%mo8>m<+Jr!Mcs6KjAhs
z0NNYdfNB>MbtXIKr<J;0)r4xdR1FRI!CZx*5VXux2)&QCTR>tGxbsX8^v{=plVCk|
z6`nu~HXzo8GjO4B+jFUAW&>@vZfhKjgKRMZlWELBzTkQ%fCEF<J)m)R6<Qsw=nc)R
zSF~CUpWe#2-iRaPI_&E8dA^#R^w;RkW5H~D&L8Eg3BVUD(QLx|L!h<z6%ezVjldLG
zmIn(hj8W6)U0BEcGZI&uS_4SIWf*NfJQ>Etw8pNF;6;91vfhR+_dfiX0Iz9{;m4=(
z<7xa@#4p<L(e}r%mv^GY#^(J_vOXtS`;)9KydQoWI2JY_xXgXVm_N*e?%3VF#xT1j
z=K^SGYy*e5FZ9FG-T6bN3(h)t5QB`L0C{5zbG(Dn5ks(roErN%o9mHRWYa>|Z@qvY
z-^GvDLS$myfgg{++v}}I@#97OcnLpt4#7tmKV1A+#E;M7$Mg6x0<T@SCh!B_3~%9^
z<1Ktgl7;V3vhV@b7Cyk*!l!px_;fD|e}`w`@ANFZJC)s!YvG*;7T&2~VQIv|(usw!
zt%b2~+g}&Q0-K$6RTB_}(+e}kui%m=+i^39U*YXwQ332U=InY1FV^&|)Wxv%SYpI6
zNtJavBQZXBFUJdDEjl|rmiPw32Uw7+b{m&l@YPkjUgu^Galy#-;`+m2llKR=z~aJd
xGGL`H;E59q*X2t5^4jn+gTz|1S*^Eg2*CwhMaS`dJrF>EvLK>>0JJ?<{(pAmCzSvI

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/raft_aligner/index.doctree b/.doctrees/autoapi/lmflow/pipeline/raft_aligner/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..24aae3a182992e4627da135c37e9ed373c29c886
GIT binary patch
literal 53167
zcmdsg4UimHeWzB^?rK+`D_gS7N_d2Dv|6;gwk+H7V#x>$#+D^~j>~r0*rVQ=-tB44
z&WvYzBrT5lNC>t<bB1_Gs02uH2_|vEVZwz#av=`I-MQ{^1VVz7D!{%F?r?yR3rQs*
z$KCz@@1x&)-92x5dh}6+sx;gE`n~u6{&*k#`t8H-|Fu6^#r}nxTMe&LJ2UH)%e8vh
zZFRzR<$9^T?AC(LlbyZq=sen)3dfqx=~lhnEV-R<6>5~cO10c{Yn=x>;U=mcRI066
z(SEe$mV!#X7B%;Vy%BHp!OoO777hoMpz3Om!X4G+#cKWZY@^a}tCgBN+jJI#qEoFb
z)!b%>+M(InyWKN{PmkI^+u`_*G#p?Tbi9dhtQe8wtqDg2701K11gmqrz2Jo7P7pLJ
z3+({Z6SZq9<rYykTB<s&77B4aTyvKB)w*)Kp>7tO(#fW~NOjubXy7c-qmH-PoAfq#
zJG?dC&hY*@ufFWgHQO~1G<Uk*JUMsAu{VHZ`3DHL<~U+=PCKYOjmjKn%A91%oX(WF
zO0DdknQgW};LDG$ZZCMN!eLgX@s+dOyAlL>yTHTU_<t4t@4^4QL>&0jtT2+M!nF;j
z((Ht<T&|bfRrl2et4PA#yxNSlmN#-d=s0cfRo;x^#c}WD;5g(t1_3!sO{d{$D2Ipd
z_T0h|F8KmuexX%aZd5CaK(r9xVWU|uxvf^AzE~iBR%%NHrYWSZvs_uW58RE=BB1>m
zE<%pB)0LS-<za89w-4f2C#hw0j+W}nhADZ&;pDDek`A|ASXe34+RKd<C_!P@E=hv5
z4#y90?rbe|M=BgUvs|5TH5@Y9b&aOma+{~z4&i!S;wW%l4_w!CQ&+aSMHYnPN0faq
zd)#gmDRz~zvx^Mm1h3B6wwY=1mcUtPSDj|DRIe6W4Y#CeM=yn5%j$<)-hAI9*F5^j
z{zv6sM)AW6TzK;Q+2j2FYvykv6K=Jam!0OyVHhYBKHM&xES%V^_P7&HXm>>rZ@R_}
z<M>ytFD<#v@p0)2hvU@*W5&XfV5wPeHxfJ}zY9gI_Q}w7XmYwz4!q9ManhRif`{*c
znBFH9qnAXq4(``GpeV3VFRvsJ1)gh~_0x%R7C6XXr8K!&ViVpl-&mQiFZlB(ILit*
z*&Ng?uXOmtn;B!L;6pbw+*%nfrFaG&T|*+gK3~sLVx!ec;5MDA)}{9c!jMKd8XaXs
zf7IQG;HVjD;p)apCmd<4%z)gEtNV50^B+P`zmLGy)nA1pAP^3x6HaH3HyvaF)N0wO
z)H>cDhoiU)RvK_FYY2;QJ?`BzaMq<t(0QCaKJ7GXowIQEEAu>L^orl`Q!De#@pq0h
zaDmfYa)VB|JA){hp??+u_W~RE(IN2rcVHOXy%8nG-_gbBb`=H5i$cdY5jv=!-==0Z
z*-?XMQ1Rmc_(`ez!V)7}=wX>0=%n|9jDYOu|0b~c^<F(ag-UQ9-Fphxl85RbodJRG
zM(xMrU548<XIZG$YfIb{I&uAf7h3;xytQ7*d&nyin75;%3_@>Zo?78&WQC4QpJJsV
znWDbN(U-{hqJU<JsiUtGxux&}(S4*r{?$XPR_RLA-JS_P0;)d~wO}Y*1{G<>RHR!z
z=>3%UD0PECsdaOHP3gw8Sn0h(VjRAa!rxsO4TzL*eEharC8!|bdy7a1wTB{KJY5OA
z!cygwTPwhIuwhI~B!-X4j)irE=KWNzz$PQ3|0eK$z~CtR_?U^9_gm)uRCl`%k3^A5
zXN^4&YBg0f{5=UXe9fR;@-X<U?9gwMmeR0pd$?YMtwNxGI@05`%zMiFZ@mMc-(vFU
zZZlB~^!`E*Z2mjYE_-bAzTv&#y?EUFAK{cWLg{ppnVt7Am9#AiFMgeI#fVg_O!qU&
z2THFe2aSX|f|dUy$%n-&ze<<cLY80Qz`I7R@W!}(pV0VrqNjyVL8G?ByU^6?DS@lE
zGa83(X44TxN?kBvfRU1%^1wK4qXC_3aaWanCf0X%#RC)le^0e1eL{%xhH!_KHWd=$
z6!aF=tptgS>Q@Ddw#Y~lMs^`YW$zww-(O@DSc&xz`h{of-d{ZW`*);t>`$UFE!z8U
zOsU-^;A_#%=gh$XO(CWxeN@bk=7AjD5aI!!qOB$1Z$8H6Rtm4H*8&V>T2h}9<x1NB
z461a)w6mEKfPaiyums>rpfoG}ZjYSMoDdpmVH||PXgDXvQd5;$f#>kn$)UX&Q*~o@
zNJCo~=BjWlrXrK0e;TE2q9?@`rU#v%-83v#4yGu&umK&bN;8`@DW{qCPOGz)qUljf
z3F;NAZGxH#*8|%b%)l@-!i|I#8|1KiWB4XAQ@=nxPWn5dj+s1nL7uC^F-BF4B>&}9
zl}A;{28G~(sGrn;8m=P*Id<!)_wI=8?}k`-i8jY^s^Xq5VK%uFZkCrtr&jXnO-%~D
z6q1IUa{PBykt<)zIx&iATC}N@JB%AzgeRYSOB{2~3T68KC8)uy9#g<zNGBX8wqS6J
z`D}BP9gc|ctoK4V%*U(VNVsmXU9B?KarooKpk69srjxAw1y=ShjG|eosz$iIjBw-g
zhr1y5%jlK`RhpGx;x2Kwk)3_it;0!{-O0G8NTvWk#gnPEWhZdrREp8QUq?sbC^*3e
z{C<OOnWE_TwfYR7ggzUN*6Q+dlPsR$6F6tX4dS7^pFj;kUMs4!xmK5DvL5wB@LFyV
zz&mx$5{cUctEBUjp*@$|^CyU~Yqg}KFvX->fcLYx47k1Q`JdhQ`P+em>9eQ&Nv$kA
zqTx3MJ#?!@40T%j3M4^s$tK0!rnsBt&!?wPVn$}EwU;?Ka*&&uZaKADJ#b=6HsSsU
zZrUAS!g9W_hw)&~;eCahSbbOpCTB6PJYR^a6!siGjFg@^-tE#RlmmLd6qA&X=Er$o
znr4SX1?hrur8@bQSSMyY42Mp4;>m&kA!;(|6L>w!#}ixWcAw<cpNIf4x4XN&;O|0_
zluFnmFbFr?qn4uUKSbbCy|X`q=l(2yY@s3r=WBJGjiHvn&?r;*e7*-hhY6pXyWt}u
zL=qq9+g7GOwDz;?VoJ1H!4j4hmJj67=17@&Vi0w%cRa2-=9;yNM`21Mip&*jSR^WA
zxkGEO`1B1cpTeP9QXjuYO(y-nz#m&0F7@%Ny!xNA>i+Zi6Y1bsWbKS1|Lc~A7-4AD
z6#@T5-Lpl?G{y5@!sCQR2?~CIX9H5OL3dIqL?oYs4zo~^vUyEzQr4sCfTRpF4;yLX
zMC2yx4NUw8M%oh&O>-UGPFIS8y;cgsp?XrA_E3{ae>Wq*)PS|*=p`I}qd*<Bg0jk;
z!!k37l#!Ni(cE=14sm|@Q@|!t>($c{cj!{#e5W1f19+U!Rl%P-@C+{8u#gcR3MGf+
zt%CQE9iGiLIJ`c2WUn4xvD~f(l_G6gpp26HPu#+RvOMTZWXh9%((C%K#}mok^O<sg
z0dUe&ZiQ4L<^DM|;wiU(4=(+C@nb|e{{fUBd44nf8Sx*aN(9M&h?mp#yE^^}Ssu%#
zcY`M()nwVUzko}V7gJueXcFG|S_iv5<_pn;XFTOwz;n4XWtJ+7iRlwuli{95MHial
zzBi^wo|;<SpA(YztR%w!CF-3I`w|&wH*_`RJ{UvJOsTTXw0;16*!-ua6vj;JQ)p+O
zX~jSS<2k;H$Y)xa?9*mjyV44UpxIWI$Xx|(AaU4IL8>qJa!~#E2o7QBt!YYpmeH6h
zv`*&96xlXW`8feo4%qxpU|vUxyuW$L{~YVe6tM2O$#TTJB*6#0+2<wW2zPg8I(zq;
z>>UfovCpSk=3Y%DWWs4-*8qF0<#qfC?G6wv6RF&t(p5&Wa9k@=-u`d;V$-*&_N4zU
z`olGg*cD+B+uLy8hatV`X_Z1wPZ#D3$7tYoL@rSk_$<Thcme-tiV%zJe8&^wIA6so
zG~I{Wl_s_f1$C@`I{1qXQLNY5@$u>DC?90q--59q2Fim2pl;C-1JTrW=CPqA1Emc|
zbAv;gv{-L0Yn!=<B+%YsAl}rV(Wz7A{JFE<!ZIM;YioQ|pU3o+7DwF`?7J%))oF22
z{^M!D7>$P)E40Q~ucd1t>Dr~SN=5AMSk|;Md?2f^a|l{Q8tjO~_{o5}$%GnJ9BBnj
ztd;8srd5Ttr;l<b*7EQvosr#xY^K_i{(Af&52EdeU}gGf!|EjbG3tmL*DN|5<{iSu
z*klCl2-N2i!|mzz#09&(>tp2srU_oxu~6JmPEl=+6(Py3U+IO$Y~j6B?AYJG)Fl;p
znkotp^+RD41Lr{IJY&9H@CT`PE-z}r<z*M~8r6hEyoy2b*WynU>mJJN*ZSdV<kPs3
z_~SarjRr_jV@r=3u%6Trb`kBeYn3pb2hb>PJXEH4Waq4AJlBA;_p%QCz4#Mx_VR8K
zLMmJL3QDC;#jvugh%g$$h8vB4qlQt!P!8k~sl+~CV@26x{ZJMe%YHEXooH!Ghbhm}
zjD?zT`PfCO84HOt%~(!OPv?lx8MvGhozpC5TPplC3$oONB~2u#&_*o3(T`w!VNoAt
zGBLvumF2Ri17QD0Scj%X?XV?6QZMb5l>R*%!!DC$5ms3&>$+Srmn7tDt!VplKeR=*
zbsgCLMaFhJFErbtCR|dPZHt7PW?QdH%|_ER(KOq@q!-U-)6Btq2n&c$*!`d%j!C8e
zK4X@t^i-Bh>2CmEzQZ~+m3}Nr&82#1B}(Mu!zNt22t}YEcv;kWTdEi&K3$$F+$Q_M
zEmFsWV94cY*@vsACY*<v)sfhqrj9ps8SG>l=uDXjk2Y{hQ_lEfsAqDv^iV%MlnQzy
zW1p#@RF+FY=fSD#Sckb3^zInkObSZyvMA`*q*$bOrae-s>Gp$Fq?)e=D;5|l>{Mu;
zlA3VtWmZjMI9E-3lmDw`Nhr(zA#Quw%s;F7l^#=d{x?<c`_TJ~bj#*??WBK$Ly(hu
za%}6pDawznZ(xZTS8E+EMd0*{5QXhUEx8*VZ)HWZjigoBws!}H0R!x0xLM6P6%w<<
zdXhD(r7nf&ZAk2DxHdio-mTTF0@X1$IK+9}dxL{M55cBOeSIa{{|p%gpYqmgDy#WD
ztJO^TnbxmA9d*JaaSRJc7ic9V+^$JO=c_i6PJ;-sp-CZ{SY7!fXvn;}!k1Op0?Sv>
zPAssPTeOWSqQ}It3)Lokv`tLMpvs>T3<Tqn%PzL}VdS<Y*?}}p)6Ix&OJk9?ntPZ0
zfL4V5h6riX)pzLXRCIM3S7-e*?01FzzJ>ih!hU~>{XR;+M3}b6h$*w&!!m$n9;PV>
zHieqgF!yIR%m~uJcB##9q?qOEk;p8yipnb2((|qpY#FU`Rn!Ty$~!NhRc_O*QZY5%
zC>5Rwqx@HBVjiR13DVPzQe(HYNva&P$va__KN{2~Wd{S;<N>0C+2nO}#cYyYU2K~?
zqVGq^X_nVzHcK7Yz?Nz5ib*lfJHToYGtIinK95tCLGAPQs1s(NKY0P|^K#uj4eq9!
zs6sViqOS)Pc}#Q(?b1!uWV*Ccsvon`pM;g(Dwvo%a>@<{u+p~@Da=aWMpw*A+116i
z(u21he?v~=+?Uxn0b^jBT*Eq{4Y6kW`U@mop1!gw$^t(i<SxZ#4q-Az$WGJ=v%q_x
zL>(kD(YR*pp!_EW`ybYycG-FR8&Rj+R!q$Nr_n`X^VNq?2xa~#nqm~Qik40JdmF_@
zyH*O?i4Dg?hvc9bSBfj2wZaaE8p(Zeze`OdXg>>q&sl+FArK2dlCcLh?1_BctC~ry
zzN{$se2P8DDRR6}<B!d{CnYUTcc2Qxgads7ROfM^-$lE02g*!E`c0}9^PBfT-tQB<
z&h0m42hspdVPzU|tQnPlb1hvlzsasHw%@$9$Tl`0Fuuht22JcQahp?n`|zaLvF+8%
zm=nnSHt}o<$bsGJfs{`9TM+xOG3KOO1AR?g5`}b@2hF;R7aj~h9A1*rk#s<*&y-8r
z3DN3CZMwsZj{1tI6XvKNfYyFlcg0t-8I|rt3HSZH-e#D!VNcBMuG$04--<g$(6tdR
zss7!z*>w(0ZS#53t$me;gIBTYv90Wp+)^+fGamwbtwd39G(QEm4o<;FL)}uSNT~oy
zVx@|jQhsWRd8jcRT|6rJXE0l;?=ntN-3t|)uGC{wngyrFDur<(R$U5)<%v~i(Jnn!
zrFst$yTV;P2$<SS=h*0zX%8T@Z5IlW8rt+d%y-}IRa%8b%CNABtq?dT#Zei|GZdT}
zZ66d~vf%J7hLu{Q9k53#M6k_;B1}XFtDT?DmD+zmP$}cD=~s2N1dF2Oj)%aX_ecCW
zFH`tWq5G^;_!1Q{h2QA4f+1egloFP-n<go8j}ab()22HA04VYS+vn(tB_8bRVkaI)
zinLj^NHb>GN189?xIXj9%>XuVvT?7`-Bio1;#FWbB4!`oerv>NwGLvukZA;n)O=*4
z0gJ^D3K3?hvX$qB#!G!MZ6#w4@|CC)mV>n7ImjE&6~&C4DKqSEiFck7H`D8YFUbdr
zHQ$S<>N}fx2~GRRX1>L$n`CLJR$n>T@H$&kX?-bc2u%mXT1%>Ea>B$s_OM$HwlhBk
zR}4;pz807yA^sFJ=tW9-uFFrvHG>nOP30$Ph&?1RsC5WM!Lj@l+%`A`v79zZMM?!w
z5-U~IxcRAZ@=#;OT|MWZkxU{LPVt}@DiyxM$Rzb3+hxK=;hIQ-zX(T`CkcKxSe%{&
z#{!p3e<!;lolJaNNbJ5Oe0k5r_XacZ#c3;eO;Y5(BKIiZ6_bK{%5AO`*nymsAxF7w
z?)FMmf61z<x+>}E>nTwAi=rhaO3%j*lwU{pS<_dEib!8^3S)PRpn-}=U!|O6I$}2J
zoxc7dDH2<6c$ThM`pT{@cKUiNPr%q1X|N==k9jj@%m&DTQ`tRfbO^)H64=z#^pcde
zGC@s!rflLbg=R~=$d#u4dDIC@Q{Np=Q@1ARNJ}_JMYWVX(|Ck%6x}OIz8@jccUJiw
zH0>il`(LcOnN`M8uw41s=veLya}ApIO@ZkHbpMwcN>ikKp+c!StCV7W>9a}-<3vFH
z5@eVspw58N=>av>&}3LOXO)r!B!erC*!BNZsC3WZdh;FhzA2w4lEHMUGV9LLLH5gy
zVD90z+hP+=;^|%%YvrIf3bF(Dqs;Lpt1|1#rU%=93#xAvEkX7F!^Eas85J)RoA009
zj5k_a^W%c5`NEOHEqorscwv`5TOiqjmva|f_I|Egp5=$O$M#zfvp%K&6nBcSf71_P
z`fHLnpWQ!(#-J<tUR4UICZ*R|XzY)^mv!$+5^bxZI8n*ViT8mMZZ}S7ukl~el`x5@
zrG~=uE&bpbO<VpvkUY(hG-F~+b@g|}s0pX5yKAMKOnX<1Sh0`+indFeniMlM(YB9M
z6Rvuh5hHEAok216aOo}=$2!)8(UB)}n}w+pI76r8&GPTekbceHf*<lqZu7+6dG^}8
z6sBlH$$p>=6ECnE|Coehmbp&Cmf}|hO4?5Y>F$evap`GqVs=3BQ=H-)_IW>1os}Jf
zv*J~Kfa9!0*0e!yq-}!ue2iu0c@{gZ!j+<!v}k|^>c7T7*=<XlcR~2k3%B9wz4KaM
zakTBatx5X*ko4!eNm}ziSN4#Tq`hZ7rJdjEhyKVH{Q?mFD~7P0Uz*8N6HcR@Et)SP
zw)}nuD4H*#CZEaBL_>_4a5Kw{7+EIyA~Ne<<6R9oa$zc&I4>jj%f_>h+$Abv<o>Z<
zl!%eLOakQy&E#6|k^3xA<RkYvx?&@Dc6G5w?l%`pRoAH%=@83&%T5QfFFPqv1CP`v
zbi4`q)G_&bt(No?Mw#*nZ-iXBN34uXMYlwqurc`4@i92gBf%Hi*mSS4B}FU$y1rFK
zmBa7y8-?*h5wO0;+V`VrA0yR!GE`4Y`lj&WLXA`}Gv-e<txF%FDohh2)SE$ho)PLh
z(Jp<2YN($anVPfI3gP64^l7-$Bj}ua;hZlG$qw>*vkww=Y##DMbj7?GySmuk>~+O*
zrB!m8<zmahC-K~T9_?Uuk7j}!*sE=e<KNY@ZI0EH-YrqC{Kcn*+DNIJSxiO`_uEk?
z%){XWOXh|=<+04EIjuaqhx_9QSl=G*IW+CV!+jw`^@N8@;l+jWaOMh8ObgS!oWeHY
z<vtGT^LV+xN4s<{mrz6L>2?~f*+e|)?LH1~_Xz+Q$lJ*dq~7<NyN$1x-flfzF>lAN
zF1EM3y@=1g6&IYK<jF-B9Gl<B=k#vQ?(|Zj4($8>p$j!)hc73S&93gl33v_9Idt{x
z{(WqbvREk}x_Tex6H89D*mP<q>8{=muw$>dT)`=IE<S_i9lGXvRqC9Pr3Vh8lNvs4
zBRJW@-snRpTXM1ao4tQg<VWTmdgM`U4d~GEdd<bXzJ*$pHC>#>S*DLMxK&XkkP%NB
z4zQv2sY&bhbVf)I-|ew-^sj`P4xGvsqE48je<tqezxLxS7sl(@POQ9JmgXZy_8r8o
zN7Fun*nU>sjJxYKyWs;5Bm5>FW1L}>DXqFIKOM&hr(-NKc_ABvjA8avT8Ow)sQdX*
zU*tSBj^FO`4%Qhv6x-q3914!j9I4dUC%4tXuHhQEIY}A2R&+gXg-ECOFFH68yLif}
zwq5fLFno!o-DuRCL9vBz@CFaJ>-1r7wtxLy)GNMP#V9v}r9N>RduJg{`Nn7m33JYA
zLK_uNC2`TcBJkf?3AAqa`!GY_JU6|I;GVHVu?bE<c?><sC9-hnYDp2U6fK{#LwOR9
zjn@NvONav>NJ1O*n3*Odd|6orRY{dzq{w^TPM+=Un6IEI!b=t#`<5MwO;iF(7E%4c
z4kd@EIPif)b!l2clP0o>L`5mNlqX7n8#dUo2CJ>DL(_}`e3>1JO@IQ5A;8twTrY{E
zOGV9;9Zn7ra`5LygnIUA6=76SQW=R*JyAp-u@h~x!9&?a_$E6Pn+OFIQ-lW&O5#@I
zQc?4u9Zn7ra_|F*aBFg?kS4$_@oJL9<O4<UDLcV7@inuH@11riHt`84hWLge>ETMz
z^7D2mIfTc74<x)x)5dgZ;+sfRl=4e?qKN)YJJB`)ek{8HKWm3#6QF=%2ypcyk4oa`
zQc?3?>~M03kb@scgqxxl^Vk47O?a_gfEdweC5pTk?c~{n^|kE6`dd2`o3I2FLs)BR
z)<BAj-zbXyj~!MHfpHWD64=({3`v^MVy`yP5*Y*+D^;X!+n769nLyKw$;xg!6r1n_
zl<dO0#ttio@Hh$s32!1Xm6ayD4SFr9oMxet-0gNcZD#j|?2^0B4#g%p0VTWS7VWTd
zNRFd0kmN4wGUJ#gzpcr-Qi^Ghlq}z7r`#sXN3zTEJ$5KISqdoGW%<i?SUF_LQRq{a
z_Ueg+sFb<d3$+mS3S&JdePtkhJxU>+SdW^-_&d*f)aN0+^!2E2ljd?+>Xp68NUlyj
zgIThjVyXZUC3%47l>b6v@3jBdiJ~e0C9N#fKfWa5pkAcR?|G}T>T`9p-&K5%>CdPU
zKRB1d;J;{)6Can@p-mnaW@g0RcD|dPe=CZ<kGrYLUKFbqli$!Wrji9MMwue!Z~G$V
zyHtD9|2F=xJ&D7Gzk2(V`e$(IiM|Fx$pqh;FVR<^<8`}RAKxX9k6JOf<^p=0-{O2`
z;&-o2X6$_CN&m|kdKG6fe=$zp`FJC4$ENK2>J&n3x#&OjB3bOKljf6mZT4EA$akU~
zpewdE%dRf=+Uy-Ei+F4*HQ(y&kwI2x8Qg)_XaBA%MQJO6M`;N)n*RT>S)0AQ8xv&0
z)V+f8H^-o=x^FW5NXCsgw?&;WgL@9X+&CHQt3RGn5&5gJdou6JY$;-4-z&34H0@($
z_5oJ?e67qrlAn&#gVV7Frk*yc7&<C#0#{1;pUjW!6P$}#)`>o1hhiHu2q?ywq01!a
zC+tvij21ZX?g>b2(KSu423~Z1&Q6tWRQHAK68fqgicLZSiYcM=Mc2Q!L&+f_4tyX9
zox4TXAKA&XiRxdni)#I5TfFjZq7qQDi0TSElpLbszy}i5`CfG0ZztL&z*#iSIFI=n
zI~1D$1r$SoT^C)C+u`I8AqRhML>R4@9ug~#Q|18yDNCzj$eQv*1+*nQ(KZp5vWu{7
zhhh_<fMSZU+oJ2w*x}?5AqPK@2+!xD>j&)w+r;;Q?BaXc4#g%u0mTqsmwe<ib|^W7
z$AJ$dyz{;2`ei%OHUa+k>;inj4#g%w0mTqt*G1Rw+2Q06AqPK@2+!T3>&O;sDq$1W
zYBbFl?=H1Nu?b5+F@$AZbbW;#Rt|x26b2I5`B`**wVhO(@NUR1yxZ(hY{C;zvJ3BC
zJFFbS<0uRyyz{W=+OX4UlUy~s<lbtBVw0SJl3jA49aav>aTErU+yz;5{ivOCn=C()
zU6!A+L$S$HK*=u4KeEHhAxn-zpR%-9Pb|8oOu$~KMb}=dm3+cKec4r^omh6g1ycff
zmR-LM`K2$rcApKG3$H)XW8qb-nZ%EKe-4X(ZxE9oND40a((8|`DjQ!46icr^q(<j%
z>DB)p76GHT4vb}0`%>%H30o2tM_w$oZbGB<rB-^PFK5`6TH|%Q+nii#UC-c}OC)g=
z#ZoKrx7WI(SZe(zR@L6CSZe);n5fUo5hkwzm01rTRS2<_pRe~KSuC|mbIH5ZdM{Aq
zORe|O6<ca$R~LJ!^+xs`yL{`avzgab8IXY&QFn7(#o8R51cl@N_Z9RFbgA!z;+3|_
zQcW(Xh&we0^)Mu&o1m%K8CMjSqfVG!y%=9n++oA|T&~MK79rC2y4=s8X&>uyZ)4TX
zb-CzNQ*xr_459G|ZZ5i4lK7SUxc~CtxSN|UlLRE6g9@`yk@CCwN%>SBQo6^2)RRzC
z)b&E;UlV%L!YwmB*-|(rk}VG`%9Ct;9qrPSE#VbpW_3vd$>a{U7m-H!Ld@xDly(r?
z&O6;E&I??|+xPHA%Wf@T=a)MC`ySFa^{A!P>$?K3j7O%=iiP;-5Vmgu<Chf1=XHPG
zs;xG!8pBZTulo@pm!9O^w{vLKs_ALwMyIC>^M%*K`yZoI0*r%FXK@PS^t6Z#^YOD(
zr>EIDj<>PT5s6An4y;>zw@4mb;EEFvTC)W@L62{KqfZ<0-M*}s0{)$@RI79bWQ~qw
zU36VK!WO;oki=!ndkxr0k)j-m5cLr014CSoW%Gr*+$9_^-@+mC2BCrqB31v#lqlsC
z)+zJ&%UyORqE``)#@A{Y@>tYLJl$e>3cr~60Nl>L^kFl4hq7LQMwB;uBTDtDOb^%H
zJs)k{)Xq%&HagtKI`sGI$WJ6RNt$EI6NSft7#>y=^rtvM1fLWf{vvror%H~+X3{Gk
z?M!*Q@G{44{J#qS_u&6t{Kt19D8)YxieAeovIE!7E_^9sf}L;Lp^C(}t(<Rbb-w+U
z)y~e1jSXgBnr)@&^bQ*Vtbwm<xqpI$$KS2Fyqky7HtQ@<^{$`0*5yfxDm4C~4C2&u
z`a6)#{aiLleM<2KrYm@Ca3SRvP!Hi885lQd#pj?WW%Y0COQ@Qpz6(h`(oIrYqHxQ(
z;HV`1+;pa>e{Vn3M`_3RK>B+a>6zUWHR0UK?52ppXEH$1j#s57?F>yc|4B`_!DU8_
zY&6Y(&QDKg4fm;`K6bS9-pt*lxVF!eV7Y6%DiczgpZfx}?bpven}O|`_52-p_#au5
za}_tA&wzj?_x~LskXi18Gnc!)iKdVgYByy-Ow8F)y4-65i_O{bl=^?7_a8>-zugwa
z{GA*v$Z>{BQNqbxyXe~tB_yawdlyy;wf1sjrCcu+cI}e>$6ANG72q^_N%r|Wg%HiN
z92*+yMZcK9lLIwLh$)iZXAr*}IPwX+E9r_&;IXT7F@a|y6VD6;1tERjSTi%Tk)U9>
zRCQXdw6AyWiwKyB2w+!v3HV2h$CBI(H)AN(uDUa&dM&^xw}q3FMD{8l6|_epp`~PY
z+?(_^a01yB+LgLwr@WiO9elcRrcr6Qbk5UE(^(8=SQ_CrL9zrR>dne}vsAAr>p3!s
zmN3_7Fjw2AzrvALz1=L~4f1{ArFb#B)~fOo?qCIMoYq0Q9u5a}oTA@3x+)x(b=)#;
z+XQbY+*E2eo8U`P^mRPkAWPWURma11nA8mLUGY{YoPag)4-zak89U4P?%F+c9#yL&
z54j3Q_>i=7+}nIS91~|R)SDgLAW6kXQ*f9oZq_&J^`MB4mMvD!bi$#zPPm?<Pz2nQ
zs4y;W+s!KOHU{+)&h#stq^EdY?U``vB08wyq${oX@o-}c-z$aX7the_M(6Qx6JBJi
zG>WDAGEH){(A6X_!Iw>0N%3?g@Nl#=^b}+Ol4uYWU`mahtyRT`)WFEq;kf7E94>J?
z98S1rD&Sw>1nrh89WJ{I?IrZL3Fim0@3Ny`*I7o%7z9yri5S9<jpd7OtyC|A%bjqn
z9W2fqoM}~Xy~BK1beqk3v*=;qUxkW60YF<R=!BcaOTVg@Vb)70+!=ck>%#Y8eBmbE
zZ?g3c^Fgd}xP@qd*0nsas0x}o;l`>{TWUK?E;wGs{c5+?2`_27r8*8#EV~W2RtA^~
z<Map&ZxLjIBbBmL1#cwW<g|l2P8yR%;pz$$ZDSe!ycj7X7|RMUzu`7(PL)KrnUyd#
zC+VxAM$2uN>%}w6q)J=dGmU!7#iz7YOQt0y9IjSrRq)vb>Wj|gbrz~bh`@4G8VL+;
z1Bk)*e`?I)P<^AQR$Q!9LEM9#a64TANXu=~cY+{c>a7fsuWvf1quZlH;RIQlIKdLy
zzg!7A$F}SVH#OWw5%%H2J`vRbWCN8fV4kT+MiE}_1wo@VKR1Ws)@)-X@bGoNdUFZ3
zw2f*q1W*&_!Q=d7*j^`Ghw_=Lu08;MZfY);V%<}T>i#EW_cXR7l|AV;XDe>7$Y8(~
z%z43b6~M*<X8~4J4!q9Mb>U9v<Z>}G&WM0wt3eua79X-*sRh`->s6LKOg?(h-?h~`
zss09bd|gqTU-vi;O;g9&;e%{AyNt7TXUTJb51mKHtPosgEJW>-?FK+>0(L<q0ROog
zI1BBmmvD>^bU^rpt>h^1`QKQz$};ILeRYHx<67p7pfbw~_y}?<SOEr}>jr?at%O=9
zi|(>hsTQ4bnP$d--s9l}^-L1S*M1kdH#-}yacieK$J-0xI_^zizR<)<tz3a^1jWE<
z;&p4LHxob#nias&q7R(<*Mc7pw;H82XBnVSjNTTbn-OxRWmY~)h5Q-|MigLQO+V9!
zk^Ua~`Fr~LKK)!z53a>eI~=1{!AgVm&D))2ZO*d#XIWhwE4T}Rg>{Gkc@rIU$kC5t
zfHxeUFuEy<x-#5_!<807bLVlIY<LFX#XRd;8Fn!xMVNXgIGQtrE1~IAp!Y99_TlfK
zpP!?jC+O#s!}$3${roHatQx`3>*?oq`gwwWI`nfXzOd<EK|k~nP@g^$>eJWDeEPbX
zPp8EDbc(!BCn@`M(y~wQu>16myHDFHeA;&5)11FgYXa?Xqpt0Y{~Jpc*bE5}GT|lU
z)MPy_XuX8jgWG_0EHBinkpZ!)m&OXC+G9qKV-hNBwAJVz@u?Yf;BB_W`w~x(Ai(98
z>TPn1h{mOQmA?HA8W8zMUnk0+gx39R=PtMpd~6hs2zuCp#c+U;=QWDgN5xF(*H>z#
YYP(EEwnTn%yiCDC_)*Gz+6%M)A9E@sE&u=k

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/rm_inferencer/index.doctree b/.doctrees/autoapi/lmflow/pipeline/rm_inferencer/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..a255bfb44d57ed227081f232a01cc19600306219
GIT binary patch
literal 61288
zcmdsg3zQsJb)ZI?zdkdP?ZNR_b_*kGG>k^JFj+jvvW-~Bj%4{0+ZYScs_Cwo>Qety
zU9C^R<hLePQhZ&IZ1ybSw~%beat?umg8_TtELk>$jT1Ny8%~z9kYztC$+E{MA!K3i
zeV?lLs$So&svc={R_Cav>h*i?{_nf@zE{=1IP#tof3t!83#K}4r`9~TXjQAtR@Lr=
z!FaV*={D@9A3hcC{iX2na4wketh1d~*Q?lJumO5hoLar=+0F1o7;MAtey!eF6ywJ_
zcEzu?n$d7)#2IzQPK0yLL@?sl{JO1t73{1xR_d*@i|tz5uGgCOqSq+bnkzO;U$MOq
z2g11FhwO8RP`~a^gdF41wgnUTp|d@hC`SZ2n}Sh6$FX1-p@qk~%T_RH`My_M?)pGK
z(z~fv?I3MqmAcjGKqFj_G_3}|8n1TS@y)VTIpx_a*rywe`PM3a6*^PSth2?r*4gCj
z436)2S`B-@*KIa!Z~xhrcWVF9d+!3G<sU$>v!5fj-|G4;t6kgA*|J}<WxvLj{k3M*
zKDX#~fWqsJZ0s&O8-fwmrm&WC-MJnpa&`e1Z-D>1;r|}^zZa<k26;6`(_ApzwrXA&
z>~6HG-Mani5-pYBWxU@$x|=h4%nz-u^Kxfj%#LHub-;9IGuN$HIi_VpFnMPz4krH@
ztx3ojX1tR#n9peSq!bZAc^u;1odw|8IOEy0Z!LH0mRGK{>g7(`uH<0fFzX*ozv0n0
z-Te5Q4m>XZh6wJy7+g@lh~i(m`Ie=Fs4|^yqhWdH4}m@acY_&W)?hV@%GY7AUAZd*
z^|~=`nS_7!*6OP5O-@SN8%cH(iZv08`m0{6+fFeJEqx4Fnv<jBFyw5l>O0|)Nn~OZ
zn0N~0)RLOfPotFy9#bbkzq8z`o=+hPcy97qXH)IWRiSwphfIKTu~t2+?I;6-S1q;A
zFSVB4rIVb2rGun#Eb?B5PQHQhc@8XiOWST%Yt7a2IdCZI)c>U5)KY?D^_p*cR$ZCX
z`Io}7MmZi~Y*ioVZB^hW7;eGF_W3XvZJ*x<1c$cjR*>(%4TAd@;8=TFWH1U8g4YX!
zm*?<zAuL2VtYOufq4Tj|4DS5%ZCI2xAsWGExVO)>>$OVF51(XTpS8SZcp;bo(QuE|
zcLfQ?p)d3M;V}l!x4c!`4}+Iw5hpYLe-1wXQXBn|A>jFYz*uIS(O8u4QAKI@6bH)e
zDQBniMR1Hb&pUA#B|jYScIf!)0Pr0N>l8tQbnJbA&N_dd5fF|31;FO5{bu@S&<Pw$
z@0o&O^jaaTs6b#1daop>46affXsO<6u5#~aC2gPzqfaJBtDS<0bB+2^r)^<WHQx5@
zj_sYXnF75L#>&O&4a}j(C~LurB^UKXt61WZib<GD1R~MGN36YLhegOE+(UF9DV4ip
zXu}3ouX>v~!iR;|or*><6sAFI?oVh<uPDWN%sGzJz=Enwb8*e<zyZ^m>XaqQ!2`HV
zK4^n;soo1gbN!wuUY(r0r&;rBR=xJ~cBkY!b}9a@#Jo!BY|VE{tF<$Bvjm=nMcSQ3
z?x}4FK`eBxem_@eC6r64doxI3z|~aF@q~_;^K1J3TyLikJQA&CMh5m<HI0MfMfA5M
zBEiEK1rC?f`y~|jb*R~xqRa$W8BvSZRrmFg6_^6nIp_EKud43DOoqKJE?QWfr`5&M
zt-!eK>!tH~=UL~uW6r+~UO_K<W|GlfFZvp2Fj8E*+y@wkv}KUBsU9o#rqX}PqzF-m
zmC)^?tk`1c9>dFQ>!Evu1Mj&c2CquGhm)~?IN8@h%z>IsCudMJ>`8&Dvr{Bl6TuWq
zxXTzBv5Z;?0fQ|naIV>qX~hUBGrv1Ab8ig<4!ZBe-m@;EgAqw^ElF9tvLx~EFQ>f-
ziOcB;fg-i$qIe;05ptrZhls+xn^8a#>p)N;++y$fl5vH*Xqxy(rXa$dH+5C=cy|Ok
zxTDw$sDsc!Stni2AIJz}+~TT696dyZ4ofwJWaxnFHW4~Z!CDBlgVNzv(}&!tBQ+}3
z9t-?G0oJNv(wxRn;)Q4g3ni`wloq8o?w1~F1&`4VWLl8@1~-XxZLZcV@f?9Je(cSf
z3TV|21#w}U8-ih|x1}o%m<evfZ_1q-6qEd}r`5}3zJ{v~OBz{MEZ<?>NtW!Kk>EPo
zbxY~=>lO=Zcq1jOxnMKkyAR@6hDfj#@nR7VyVr6=O=aDUIPa|cV_mY>fMhoW6O6h9
zY3>Fb!B<_$ju_3EK4^|3njFV*oV_1<SN1|LxI!tioT=GoD-a8Y!IZo#Tg{5o@)Vi$
zQ&3EAWx9D|nJZ(5t^`715(d?(A!CZN{G%VfF^M`?z1SDHF9Oq<8-SuG@Fff;kvou0
zLk(BY>w{5_Pv|@wjPU%}84bo)y7fBaAcsF$_FI)Q)JM?#pJi=tg|wZu#@!INe?)Qb
zWJY`qNdJdI`b_FunA9uWk7VZ`PJ5V>ET@wzVN5cv_;Wnc8m?Ntm84RX7u*M6qF@X-
z!72>ygLupIMZIsf_VKFgg<!1Nl9$_L^FCfex)5v;U&{OK&_j^di8@U+Te3~|!<i6X
z$M$`2SK$RD@$aOe2yZAC`$B%u{XZ~UaL5?QmF9TH0*;aNCEyVY#fh|obpIYO(_Q<V
z`+t?T;Dbi2=Ri;GdKq%7&XU9%Zx-*x_B&<DwNOIo8@jy~nt|U+$6A!$8M?>ek}_g`
z{uCsKtDU{f%c7f}Vk&Jln=Ri;?2ZHnjvu_ihf3{IX%A!go<j?zgRDOoBb23~La<bd
zx|H@DIs^#^bN@5a6l3>y77c*yJ4EYDw3_g_6oU$eN^%{5t9X)MN=%}!%fZmuFd3@2
zzkow#T?DU28_57$cl}*H=yU{*`Gae_%kCy<lEPp6rR=>CK+YtczBzXazIUhLqyMsH
z!2K!}_a13s)}-;neK0K{rn_iN#R`_j6!dZj(=#k+H*Ba8*K4q)qy!Paei<24WJ_G1
zBb-%yZ_I{oG!=_dFwiQq)k&dzGM-7Ah}sm}%5$WIf}v*GRz8nIX5G)gAGY3I85w)K
z+pPHAl!q8vk~qLs4Ep&#pxbHC|4{|){vzB(R@TE8e2WC)2Jm$+XTS+`pH-n;BbE|$
zj|rE|X^cuZB3haWs39>ken8@>S6(4x66-vV1frb;_YlmYmHjE;Y{vmNxoRu5ioVSN
zQIfKe0LwO1L}uOZ!|ciY)cs><OmR?C5j}cQNzvWu)5wBZ_md2Y`-ku+67HUis~xut
zU1Sr{SgD0dODuqwY|>XA-KmmfUI9bF>Hn#UaSxr}O+)!6Op#)yP<jC+i_*V>hC(U5
zfTAh=wW`u51B_m3PY;Uu{!d9<Rc+Bqf*W91ToOzxJ`Fy%>fZ_Aa;ZM&;~J>`O?l%t
zlMPHICXtqqYLDeoBjrNcgUy(g$sW$qP$+v4P_o#=6Eqac9t0H49<EpI!4N_BGKs$9
zV($M2iE&}8=!-#a!+46xI)}653|#}m7?)`ZS9pj9w?~0Us|^PRe64yBY@DB$YLAyO
zuRllfIv8rD<B(@?$gKNmM!de<jPl$`f|9o}xUZ1lNW<dr3h^$>px~z(-MU{Z<Iys#
zg7BakZXv2>x%($zc>=Q2p}4z8S^F+(o#-O~$Xq9iu}am6>TFrX(uw;jxO8uY4?Q=!
z4O*b$gYlCy>fVl>5R!WbZ^!GmhVE-*d!ht%Eik2tJ=sQ!`&zh^p$limoD-cZ2wwdf
zNYEj<k9PQy)rn>JzAR79En@Hc#p%M6Le*!2F)@jv#y?Kzl@o=c!7YVhIq#Q*xZ}e>
zATL|aXGd8<a1tj_a(mrrWZOX;g*lSzu;vZr(VWB%;*~Hi4i`)(ZlDGRaWy}2!*>vw
z2EZ<29%^~0%nO6pfmWbUsDc9a$FnPVs1N)9NC*)Ut5{<$?zY=PLM-mP)GAeuUA9{i
z6Y26*rG(}qcG(a*5Sx9MEs1V#$ENc3r8;jXg2|PdU9WQA7U#{vb-{@{_O+6mOFNE6
zrZVpPrNfN2jwY3+*#AG8i%pMU?^*Zf@DEoLWLJbm;#}aP2SIl8^KtS&KVMoZ-K&)j
zOT0e5I9Y;!xZ4Hi(fDBxNNV|kj*@4e?$$iJTJl?PO2~r0a2ABkx;QyGKObdOZ2DwI
zP#6GC3$;rIA`!4MK!#8&FD!@)R$5-86ipAn)PH>0hK=VcY)jx-m{!xQBY4IV3Ljdh
zGMrOvC^{1y&nk)nVh~FK94?2#mj-p44mIjH+VMR&@uebY>I$}KE*1SkF2#MEEADyt
zgZ8bQ>vOS9T?j}McF$sOuKIhx5;}ZBumDzVhS~+3vyUw}m@#WR7wk;WuVefC%>;I|
z8+$MI=2#(;+%{vqP%sm|C3fov@=IM(QDkUL;qT?4Fj}7GfXv@v%%?f>5uBaN3x{xd
z(IP$_4?z((85GxpKQj6V-k90z<l$=Mak!DV=TwkeHISmmTl(|>>oJy>{shLUI#2UZ
z7|)ksP||p?O`VA5tYSR#z}bJtCUk!T{zRO;x|jc!%GSF>tWw`eU`3NBj0UmcM&o`|
z!6;=Yw-gbn#J<?zn6el1P!<`>%YoTHfRR+JicyeaEI5SAhZd<~EGSZwv2^F>qXo`m
zyMBGw&kQWiw06$vHgtsn6O$$N><9Kl6scY^q8kIPcEYNzfI&&C!nQ(I1&;=~Q*1)r
zs%D8`MXDy*C)U8l1a35u!nP2nEVlK^LV}kR6dWDXc6%P$BI{ZNrXOTXr+J}R7Y^Za
z%4}UIT5etRkh*YVkiy~W<zz7X6&%8&`aapA`#?PVZWwY5Z~LF;bdPcfGM|(uN%*5p
z4ebF*IBhk%Ygei5wJJ6|FIX)tpD#7LjrMtX#HF-rmy|Wxhf5Uj>ws7~#IrN{md|Mv
zf1LG-u;F&F&H;{t+=Ht+^LnMZWT8TP4sp?V0;3c?P6;PNZYwasVgx_b&xj<2(f^Ey
z5=yWm6Rx^B?}o@dUXf8c#>Vwt{JwX=#VuhuE8Fh-yQ8UCsp1aU?z&VZzu<bKm=rFX
z>wz)Kse0hsfEs>oEJHnz7Xn#H@Z&H}lmvSnm{+5}UlkQY9EOX%a%k)h(E4`>o&qn^
z6+`O5BY8|rP9Tk2cT?h+*hHkx`dKme5tROykfPV&)gR&2qtVst;p&3>82dcVKHtDT
zPq5E7vd=PpB4NszE~evaaHi|NL;y3mDIYFk$_Xq6Ht19=V=(K#LcQQ-o$4MN_qT-h
zNa5=%;anc${$?~4Gw#Ee(zth<jXTMAvx&!ur%e0>ps9$7{}9IYGx3yEq@m;BgrOe>
zL;r^0Z(&216G&MXF!bqBY3SSWiWxe)T5m)D*|=CaxQ!nwV&gFa1Dm&6I5SxG9I#$k
zw%Ri`>Ur$2HY@gx(NxT+@4A#mT{0WB!f3NS$2h0#c^8mV#GVhsxPJDm%0iklj!T&G
zU0}*L2(A`3WjVnBru-O^#MbQNc*RVaU9Gz**IN~<UiPf!sdX~t7=gh}S)B`yE9<eh
zhl;=#X?W#?BwN|-^o6EKK@YG#x1*_;Dc^G$n6g5$*_30PQ>J_h$SGpVZ-a4tO<6%q
znlg?{nDRYf$_>HQ!lo=IkP<GiDt{PBVy67B@QRr-yIOZshWGry_Gkwl1Yajpju9B#
zl-0S?ru<n{1Yycb&v;e-ywDUW=mAXmGtpGcl#gBprmT=`Hsu)Slqr7-$SGpVFTl9I
zrmP?)O&P}}O!+97@}~t?3!Ac>KuWlPDSsbHVy65zc*RVaU9Gn%-yP*rb9;-~GCW)1
zw5mm`VG;O2?f6lAbw@NCe^4rVMvBFjokeG*jL#y$6XUD!N|Fw4-?MKP@4gG)!S-O(
zd@r7f;7?J(k%MR&u_4%D)g?TOAU4MqN<M@&9XvZw7JcA7Z{>EKJ>9h2YSl}(l^*iC
z_Cl!(r?<}3>y4$7Xuiv;cc2}fEQR+~m8)=?j=!y~8owzn*;gKI+*{grNInJD3pKHw
zqE-fjiHJ?IdaJa=j%|Ew&C-qGXe#E|?gLZ&11JwgyNusqVu<+s|EIWwDE?3-{O}qO
zF6Pw^A&;<mrFg|%EZXZLd2pw#uUX_N(`#TTh%d|I3=1?A>iD*RlI8gJ5)FkqzAd0=
zoRe=hE2M**;nC4bu~$743^r*jx1%Jj>LI#Zk3$TGUe;6LtW*x`GkaR-Zl*@L^-f=U
zRf6@cqo={hVCkP>Q1qvT=7g(^yiFh1scFZU=qxQOx{qblCS^q0R-L1p#kSr;L!oR-
zK*?fT0S$$+EdfR2o_xT7<am5CIWnKV<6`c=kHlD2m|Pap7lT4Z@LNpYIgEfabq$PQ
zPsV`7w6<K+d#p4Is!J@ZFOae#E#>o!bD375Kc%5imLi~Jv6R1}p-`40plFsdjL~7Q
z)r5Z;^YL#<*oAeTgDIrg-(%v*q1ZV%JvmN=rQ|ca0?KFM7=09a4&5e|7OrBRUQI*!
z_b^53QOsxXc>b)LgO(d<DE+ktPHc?Un(S@Q$|=`j%rzug{3fRCr%9OjZkrVg9`q@n
zvFb31MkKryrrU0}Jipw57xMb2yDfaHG~4981!tACuYrS_F(*ols5$+as&Uy>QXDtL
zRNqfhP5PRnFf?Pt`$ifH<!c0#EWV~rL!q=rKq>BP-a^7&4`1{1OguS!O%6`~CP^*n
zYXp=mzUCt|6w24+pyiV^lnh$)N|+(?n<K1nAx%!)#rF0E8rq%kbxzm9NnHciGM~NH
zVsbavQo|)b=uR=4{*q=B_4<StU}#3S^xtVHlv@%|vbd$cqoGg=CZH5|OA`}RP_Q0u
zX#_Azp6@1?nH-$HfuxpnO9Dz3w{$ZNg>p+dXt|w+l0j?oxTSk(Xa(F7Cv^?nQsRkq
zR})ds$J1A6f+!&^!%%P_n$gJz(#4LKyp0`}Ps5`e7Kc}eQ&|SN{}sY-^+EJ>(HY(X
zz+b=T+3@PgW=DDcD(IWxGnS-Fdws{nybDR*6?Rg6F({P6y_0E&9#*L+>h-p_BWbg4
zn&s<$5IElJFz;W+D5v%>e*#-|MfNX01I*~Rf2r)ai0#W@m-$>&M8Mo$e1X1tP8eSt
zKO~ZWu;gOrgyjU}?&c`p<ue@;{xhVOZEgNJUa_rBcD3GHoA<;!XanzU7CJf@!yj~W
zGx$n8*V=4tvh59+wB3d6&4-(?6?$Bpj_g&OT_}0hdA9Q@w>?i3+_n@^wYDcY851aD
zXtgPJqTfSLvo?ny|64Q_+m(C}Oi0_6d~OXk=N9;b@$18B!RX6j$8qB%B{niJ{xOIt
zW1=%nL!lBK0VPYKa~%zZ3RwjdE#?|*_pL`*GT5ZCx!pw4ss;qP_EQE!uTVg<0F>lP
z$hfSsgr2x8iyg;*;L56*-0hmP?c-)kZ%+AHVaA%$Y*yx!4pVE3MfeaYLeli^XFSWa
z&9Y2Gp-fLe$zpnK8VY540*c1P14V@~*e2yKlT;Kos*95dyoKvP7U{f~q>HRe2q^m2
z_nW0=54~C0n)xsdh0?Sfw0wewLTMU@Vo;IkWCW#9^&APWuuh~w7IL1PB;6Ng@ErO8
zl?Xk3B%J5vRIU9EDI<~}--4kTgP0%CP$>NtP_o$ShRtL^M(MYJqIuF>8`e8W*s7Z3
z3MvNJu-*n3<q8gRaC(8Hmh_4OidG8jv0=SLL!k<RIcT|?hLS;R@>FXdp`jJ1)^bwU
zz%3Q4*0yMZC?TDKq2NF?qm$*?p4HmNX?Rq%mcuK=sVsw1)mlR`e#zb;(Bp|jr6ggl
z9mY7PYKPy3JfKMJ@IxT6eziljFd~bGFEtkrV-nKU!v~>y_}8KY0-9B<dMGC-T0MLQ
z>1EZ!&*BxU9<r<Tt{(nuS>@e;k4D^H<jG?Nfr0CXA5|v12nEDGX%u@<ffpH%!UGP+
zSKzIYK2#6cjnZ;L{Pv6XFlDe=!CfW<T3ejzex3Apxn6tld^8m+4n7PPq7?^EuYGZ_
z4;MulKr1b*)AIxWGh$Be=LfzELo)`#Kct~hfv|v*B@iClQaBJ6P_#99u*%j&uIBv|
zNo(Q2doY>{C8SpYCAk82E~_k|iBiq`Hdj{7<QA05TcKwNv@)Ft-L$sYpWQ~vi%d9g
zWemzxx;sKcq3ldR$zo@(r=d`(gMgyhS>Iw;4^!$pF6RCz663<w(iek52C>TIt?$=o
zP>@0fk!4USz&9+Y?||qtI#9SfM}T0ctqo%wq14mfb8&>n!4#6vyO@L$7I|DoG(*P~
z<bRh$-lZ&(LoqODghgEnkWZ693aht(vvw@vK_R^AbLfPE{U4LqXJApL9|-#f4TW-A
zMsb@f+TW$2^cQy~T3)1~^ryw3R7tDbw3P~`3hUIRvAUO#$Y)Tg3=G;$!nn*CbSsH`
zF$SgF($A1U3UlZZxuxSI_8C}|SRc&JuQn)@EGsz1!Qxz>VeP(T#kn5G;?$ymxjGjk
zovO}#1>*W5)wv76_kPtmtu!Iab2s!T&#5)0<mtYLp+5IjkpdU4&&dgj*5^Ke6tnu=
z2l0y4=h)SH*XJH7n_reO@KL`b#cFlNH+KwNxBKf}RIYvXDY-YjfUkK5{p|SgTkrUj
zRJEqRP+GS9io@Rg#2Q#-k5%vl*SdbK&fkwx=9ebLf6(s06Nay=`OZCf0MS_Wdr1$<
zWxU+;`#67lhV+uTs)3)5ref8=N0ZgSU%gD^aT(PJpO09UyGHmYFf?P0@ITW~sL)wJ
z$x^%cUo;ddbQVyw+Qne?vL1zv!6uEZ<|ibrh2yWm&?}Tp{)nkjj*<wMRo2)@99sD@
zgQCA<WT#dmOnO#fq?$@-1wgtec*8a-$0BP!*Tc|^hIKOyg)%GwC5vI*PD7y#OF+@M
zm_F^)!&uTCV*_}A#Hp|Wq(K%^>U)_8b0{_E(;6uCj&wX=(&VjLH@RFVT4Qc^Np6$c
ze44Q$kuaHmP5xJCD3mq}C|R`mT{IL*n*|h2n>Rtk#rc<h?2LUEbMZqY<ia}3pcGQm
z-)6GOp{ASxYoMl6i4fJKqmywTsi108%-266`ATZ%7a0dKY3J8zD3o>zC|R`g+cXqP
zI|Y=Y+W8_0c^$O#e>2(S&`!>PLfV;SP-?TpkZ)+4C3mh-&Xu8JdHoa>ZWI>Y043d7
zz$g{Pm^V^hK_X9<<2V!pgI+yAs;g{Z+)1KVn6Cq2SIE^H#bK^nK1|Y_K^$3(NvT^J
z%;=KUE%%u!AKqfGvWB^a855AIVSWPgjv_V8F36x?4YOx)Mb<O-=OQkB_V7`tZGKWD
znqbPs&K}AMiq<yYhMZuv&9~zft8KEY^{#E+S7z@BR*S;}9zc9;v7#ovsW`@B;A-aY
zr081vvKd<#;U&ts9!`$SnEYs>h#AkQuj+l}@q;(u+oP9Cd#q&uU{(A*hZaf)t9GXX
z1<}=Vr_EmX9Q7*gIdn*rHevI6Tn<u}fR85JVWw@|k0%Btq8KWSLwM9L3hEE#boo0}
zKNP`ZrO)GFuG*8^(@CMqh&ug699Mran$%Pr%vG8-;HMqbya=V)LOLdNN;^xoPRBRD
zL2{qWYQD}?C{tGRT^b6N)d(nAvYHoZD3mz}C|XuCSedd%wP~<PW2+gPz9^yiVCWT!
z88-nXxsGpgS!E6XL}~VUuB@8L&1j|B$dZM1YPyi4Ci`+yGNeWAVXVkxQMb}iD2ozM
zvRKs5&`>Cg5>PZQP9%72$f%O1h+a=(P&hRjAmx16BqVdn*&vZ8a~uxEz@SYKZzyrY
zn@A9a`NN<TGG`-u%qieINZd29NB034OiLf2az~GM2&Zv5J%s%GGy|z;`hFLNrV1zK
zg84Hv6w1{IC|O+1muV=Js}WE%SCi`@<i8|gt7>wo9zuSBi6;@yXOL$WPQOS}OFo1w
zIGyDo<W1YD0L7)AZ_S}K+i55nv?k9($k)@*3Ot0&NiF1-vJ6TUlMNY?R!rW^R%kWz
zH7r5NHTn?Amg44vxP*lkp(Vg5<=xC{^n)bw<RZkO7#Nf~$a<0lQkX-79Aq`J$h=Pb
zB=#9tltopQQmsKrFIlPf#$MITME%QLqm407)o9-b2}qF|Z3xWoSEEgrS7dQ^E*m-N
zs_b#7%D!LZ(}k<Da)P2&*^eRZtSbBac*Ux+>}tKMvX7O^?N-Mxx4jmehwSj5jw`pi
zJ{(~j_|)lr#R{}N5gNEwdyb-(=g^$=rTEA1Y_GG(zN$4>1!FpkKWCrr+HivOaNUBR
z+*_&H-hGT^c>MEztK;{my=pbaSppeZZY2A}HIy%rel1t!@e9#ZtTK8MOh&6b9$iZ|
zF*5a2W@|rx;+HkHs02tp>y0sEzZr4ERMI8$w7-C%sg>KDm;4VJ3YC`#C|RQAzo(&4
z5vG8m<t2xrUrZ`wA)NE+GeCn*8_RBT2Nky$4pjz4uuvE{3bf=ZEpmxv4Oc{!;A>1}
z^;}KmQ(Rg|h2sXdF?cwB%f*r4041}Vm>d#yO>;PYH;FtMYjP-A!tqB)Agb(g<@<w#
z<3<*l!*Q3yJ_C!g>bH2}%b=u}EF8bNSNz&L+A&AwF%hZA{BvNxMI!TG13C1I%*jYZ
z2I;TJO=LP=KMC>rvtp?#9Iwj>ipJ|7M2@g{{X=-g;&pbl-tqd8mAd8ocC%cElET0N
z`W;0AbQObvBl0`c+2)X_a%+VVw-mog=>!X>AL_R2_Hitwh?gwENn-QMmA0spMj@Wo
z7>5a;C7n>N*!k&bDi%9elCksEz34N<!F1nbOsh^D5v(=F1pirtW$u{vt1vWUO#3Yw
z3Ki1|C|P3K@6%AIm{vg1V%ka6E~aAgj7vH|P1X!ITyast%mIRwe>6$S9Hd-HB2TWs
z9Eu@GNiyiAB#go_%33qX9HRU*iF^hI=^+Z4F>;O62dM0c!*AtnPXB=kyHLz-a93K)
zK0|q2gWyM4e`QPeF_L*IY5Ag44yRBe^$4Jp@?d7$Xp-2IqUTTyEK1tOn@JdjIW&N6
z7#Xx6*OZ|JMZE7Mk<Y-OOh*IXM?;}pd}c9xjD&I7h{3F<&ymPy5Q9Na6Hj7&l>||k
zKgFNKGHR*WgZ~#2_YCYwWcy|)e>*6YtZrYVvIgIM$x_)KujSR^Yjf@zW1Y%f{{Ysp
zBDw4ImHl&9Js*(C>dVZ@s!B#WJFP%=`Y91~7S2xP1VyvcIi#6or+e{=WvA?Fy|dG!
zl~$we;XV@P9s{SRcNR%cH53NUPw&&F;|Z$YD$7=$q%M>?-DQ?-)$Qi0@4$OSSR#Q5
zYP=JKO(Lz(51Vh5voWs8r8dzT=Wnk8nXl20g&m5fVhQU?GGRTyh{{DTLw`=+)0r_(
zBf4rC*mSb1WJmgom?MuyWaQ3hAA+H&c+k8{vqD3mGFky8OGfL{P^fT6K+*WgBf%a=
zxRcT7TS?f3m-zvr(O+ib(X$&ehvGcQx|oW6kNX)EU1zW9ZPF&aol10<YMW}0wf0|=
z+$O2~2uE1|VIi~DK21ZRv{pdLqP1V5p-@^YplDjlk5{An<h(XSs7c-Z3lf9%(A}?@
zgk%okzegfZ#;6>MfkBB7{vSvXh51uFgg3Is99(a^idyk9ut)a?8B9wbpmN7pgy9Y(
z%b?W8sbLA$iZDe=k_SkBsnq1EyA0sR7XhQxGHbSlyGZ0oVRI-3TSz*Khe#Ng4TH?q
z;E>2?U{Dr4RSq#4H1v|Cnmt}IsRzvFTr);Hm22J#zOYEH`6O_@-#%#~Y?sOAZgaA!
z5t7a}S0LMbwQ$^pvrRcc(QNbmNHxngKY&*(+hkX3k!|+YMbyB<{At}L{!xA&SsyeP
zY;2zogOLiJ#WVew_rHk<_;^GByDCb+3qyd<&ZJreQ*g4ZTetUBT20uY^*bTcG4v+#
ztCI579<c7cC-D8K&ZINzY~cj5JnU-qdFDCiV6c-X!u#5_wq36^?R{QjU-T2|9?*OP
zOonho$3@w3(Ni9E_W+>;)J+Q1jmZN>njakPw7Oo!4xNSIN_O%K_Mzbiw(;QTDrWrQ
z<KbK|;<tRO9v;~cOv*lX6>hso<xsG#()B#xOj*ozEZ8DjkgD)lFy66!9}Ze}!eD!~
zRpD>jT=5t~t6?zuh*j^}o$#rUzhOTZ<;OL{W6spEU}Dv7+MeaNJeX+$e)8Lf*L*hG
zaI@KKweXCqy;3_D21EPdEIvm8GVbOn=rAd6yIvjcw)(9~dD*I*!f!jF^Y&oJ3QW*!
z`L@#hWU#ekSA2Xk*twOK*9e~sw!tH9wKi4<8cYF>1+%=xuG_4od=?IfmYJpk830KP
zK-FapTRgkn^89k$K4SwTHwKfARgZ2*g6;OX8t~7z{4Nw(c&CwIq-rmBS7E+wXDzRZ
z2UTHS+iF0|1U}SnBQb~{`)%|xoLp^Hp;s78bp4fmui6Js`iH@_`j=(f^IBfnv6|Jo
z?E&u*)e1buFvWSU%re4e0qskClbFK&9++RSjn5n7AV*)Te~}m*Od|~)V3Y$assl}7
zu(fVASG(4#4IHn+{YJYP23L4?rR7!2Rl99Bs{p3P7&{84w*q7WM`~563eIS-&FVsZ
zBW?;d)<Ds=R=bUMIZ{SoENg&_#P*t29Yr_ATB1XcaZ|fvcdM=Pxdy7zw0#Z^_}K8i
ziFhQ_l8RqGQ|r{YvQFV^f34*@5+bl1lSTptw*`plfLSn$gGslFZsnC)9f$+FpTR49
zudVGdtF)UqTNOmU*|W}qsQB%Xp<p{&8fXsIsX+T1H9x#}dQY&eZMVx{A2!%090Ug-
zTd-x>0^HFkf~y_hZ+DjV?}z5jV*9-Bw3>@8Zxw9mD(uM+fSzUFZh&}v5XU&Q@7sOT
zEx^xh-by7gJ+{Qt-zcXCv$arVPubpL&GuIq3~&Ye9lucrU=zNz3|3V2o$$zbuoHB$
zQI3o=B7ncQBD@excg{C`>s;BXtvYr15BfV?Z=vdM=`<`bw9{R%>JBjK3U<$L%k@^p
z60F@>aV&f<hV5~C11>Wb!st`oHh|a$*!eXd_|M(I1<;;W1>7^}fba`D&{05zAknSX
zK;4Dk4dcc*%$yOZOk>%uf(4%k3>@3`0mfAo=yj@WH>_H{Y*nl1aT(Uzac1<x9lMU_
zX26?W2sYWxGvTrBaxl)l379WvVy#)Nfo=F@-||*%Se`&{wgXb2Sw1*gn4@a1pdSx*
zXsu1F0id86W?PAFM$ws8S^F3^^6NSGIb32thaX#Dm2@ZY<7@cwJbpZlUwjHay1@jF
z^3S)~+<e>xHs%8Be}VPA&~<jf!ontmCGwy%5sY&5qcy-8aYk8v>w4(u?1F{xV6^Va
z<t`YhbzpH0pTwZzzznY5Y@r>N3ois$h^cdkqdA4RBAPBP_3n%K@nig`LnPv!#*ep+
zz{k7s;|2WqCVuQ1g^wHY;}m|h@#9nY@frO134RQZ!3X}1qKm(?=;AxuU3{mzi*G)1
z@y$psK40VF^EWP@=X3GApNqRAF7BSV7`wU{+jfJks@^k6Y^~J5dQg-g9o~ZOP4?p=
zS6g^LPz11(#&WA3nG)-IWuh_aJ)zAwA)&HCJGA*>rSkj`{LWR$xx_aRDuDB?w7TdZ
zVO_4Y>UFNt5Ce=I7VS@g{yh;s2#%y}dEk;jBRgO_9AM;wwdT!HGgJS~wPvN>t)jWD
TqUKLlahVVv6{Dc;^5Xvkml3NH

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/rm_tuner/index.doctree b/.doctrees/autoapi/lmflow/pipeline/rm_tuner/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..564a3ae66742673e810e6fa58ab6007f65c8a2ad
GIT binary patch
literal 22665
zcmd5^dyE~|S@+tzcXzMXyS5u|Qm-9fH%ffH$=*#|D>RGkCQa%la=lHWkfwQ!XYZZe
znen|dH}kN)Nz;dDQZG5`gG@jzDx?S@B!X1Y{!xn5By9mfi%=;QqVPuxYWaiqk5UkU
zQsDQU$Gq;nb02XkEcx!tne+JG-#Op;&iT;93m<rB8~?}V{I+R1muh;W;kXUMkJwbh
zt%ohc38LqtqaTi*j;d_h(=YmN=+%vgZ9|Q^X*C<3;Y5!`Y?i7AR@1M^_LII*4=mS-
zo0}76$((#Fs+!YmBCvv{k^020X|~pz?!{W$Y8y?<F=}2*3qr^6B5H#s;|~~@2$q7H
z&qM;wQd<X9L1gY>(^`y%xr3D?^s8)~#)wwKHJz39An>fUFaXws<qoUi6GoHurtbSF
z#Pfusx5U#_BWx!xYkK{>XRK45kWB{q27QXmIdj3BF%OtK%!BN~Wz%gL%U<XJndOVF
zcYgWZXHNrF^#=g@%L1@vJq%pEZ7mC8EGuFxD`G5LPQ$oV^L!w8?V0W2nz@Zl@Uj#w
z1)0r5K*l@_+FghL*W>>Y{69*Ffi|AS(W$cWwr+V5tF+uk*ff5z(vuDBkf^fMvxZq(
z4I(`>Z#0(@{8!CuL1lBN5TQf{cB=wbz9XSO=k<K1NHP^Y-XSO|denR3M*t85PKoN~
zO`y~ir_}C1UkjVMr`6r2=C_Ud5JVg2_1W%6AA0JRr=L3ZwE7#-aPLWmV;gS_zgKQK
zzH)-Z!w*|6-P=3`0RUmyyi{B$M=kX^VtZ0=8W7K_+e{h%n(oGi;g!ou#U}FAq$EwV
zQn2B<VY`D~q|1|_=8(jkLX(SDBQT>gWul)0`ke=hTqP9!L^##K!|4vlvaGp{%?>~T
z=MK-k*ip`n5$QwMCett^`i7^s%~V77)|K|=io0g7oD&4Boam(zHBsTzxkovPtI*b&
zw&64^XG6OLyOHko*zmn7zD+i*!0_~Dsyp*DQk6;qgeXy?&-PX$m;(+L+uq)cSgF0a
z1T-TfZ9RzEzXf*v225v96=WqK2fG%r8wRsz5qt}^Xz7*{nJ=+Pyak(W_?aC90o#dp
z<5Ih6)vX}9%s*e$Jtx{?)8K_LOhul7O`#gM=g}%3D$u<RBZ%0cKCI*F{VAC7pQBoz
z83TP^fI7^ZrG#xSq}gWlBt7xxdGnz8Raiah=M&Vdh*#9$W2pEX2K;zV|D@wrR=dxj
z3+A^59H<!lEkN^Zzn(sVN-!C{dt&2cmm)YxAZVfXdcI4xFJ&StP1o5F_D|2tdjqY{
z<y)sKvFS^#=8E6eX=XIl_6*<fE*M;P9!Fc{i5}%9I{`T}_0W}{&Xr29O<0uCmz-9_
zUai2fPTnaeFwzvn@3BzXd&jnIOUqSnm8S9GRqeUB1qWd_q~_k7)byHom=BweP&e?F
zsczmd%{q3ho5M6U1uZ*CKJb16hNN*8bKd3yG~Fzh?{Tca(wo+khF=Lxqmr*t;dZ1Z
zjIW`3lOD^?zE?+bY04+sw}9BM(=Dn#p2-4c{z~?}+FSS8yJFAOL5kj<FJCqBfuB|*
zv(s0mW5J0RRk!wgNknPUFwb`9!ASg|{k~X|yUi%8=I{6Sg7z<Q2KCmO*b$mvOnW%H
zj&_4PIrGcr%jPSq=Bw;zPj}UUC~4Xo>A1S67dagJK2DO1@8V@?Gn1G=W%ShIgrJ7|
zvO{7H_h9xaJr3r?>@#BUo_>rS&1?6$JmJoz^}9+kly8!fq016B^PpVkOtU$@wAE<B
zf<-7Cf?7VC6KrqiXhL|4CG}Txqx6<Hkk9@E)n2d(G@3)ONrg~i_VZx(=f_d%RPf^&
z2~CkhBsOAFA^B0<J>0#0HwU&5FB2YL8nfR0<(<A=>(=p4yLotfv0K&Dy4jvU1-nGG
zfF1mL%8A*6K3I??uq6S6#I?sU1wAeBdB09hwZN~>!CPUiUb)kC0xVs8B|;tL3B$hw
zsM88qOlk7-e;c>pe*PMOR8tnV-|8$AU?}-m?O^fAZj`H~s^!RNL(gEt(E%prnP@`_
zT&alL*f_#iohy7=gU!+x&9@M(2|_QUp=!BA{(~>qcvXsuQD7&f%5Uyzi4{8vY9x#q
z-(JL+D%%MtmoTB^!?Rrk4WEAS_sqg9BcAql>UY8ZpDY*mgNxhPG{-52kiCss2(PJ#
zkbpVg2h1q~Q@}Vyz4rs}s$Sr+H>Khk7cApq9W%3t&8bICcj~6=r8v<~7-?atR(ppv
zu~hVX2G|hlLw&0eaeAcOH(Ayj@=yy|OH|(echH+_JHm({NyN%T7%X8C3(c;hSxK&1
z&6nAPSecq7Hnko$o1A!J{IV9fbq!Gn((;#i*|S(H^U|am!S>(8>4RKX_k;2ODH+ch
zKEoMa7v^Gc(owf}3Y1lM%FiVzlLN1cnbLSe5A-~ia!qgFhmP1JD8ZNY_WktAMJ4_2
zxJx3Kxy2?OS3S<E;w2HN*<v&Dqk7+i8WKG}t~BSks!Y|Ro@hMZ2m)B8Xp2z%52Yj0
z=qXn|lDV^gg5KEi{G_5%<t(ni^!2hEuz5qw?uxCS{TiUiTI;I)vs4+oI}c$M;%GEA
zghl+DD#S?nxQ;kEkHgFNMNwmM@jTW|8~#yl?Z{HjaUs(k#|`vc+@BqL<ivFWB9tqY
zBb*LLPTf>F!Rtdk5Q)XQd8HCpsT?_V3bS`^f#;QSB(~@xi6oD$$EH3u=%TNbh_kVZ
zvd4HzI{9#}ldMBzV;7_R9Krq|HCeD}__UkP&tS85I24VZh==1EdLUf0cc4hIh!yw>
zX_E_TBv}1gJd7Ys7OSc~hwt`o{Pg!6T_E0<hPZQOB?0aHkihplec-DQe20tSldhu^
zKFHu+E>je=Z5fD&HZAOvq<r?Rqse$I;zG>g?$ceDldgrlZ4=I#W*Hi{)G0VRDr3hZ
z)jpxuD~0r8V@{`(ewmsq*k8h*q6vCdO1~%?eStT!{|J9#$()XD0msXJx!@xf8>#9E
z8h=OKr!+PzmiE{2xkK9|9^b~dD}u*t){u2UrdthQlrK)u`tj&!{pd=eHNow1ETMa1
zqw2P?BM@X}N^ERV$mTpfC3x)IQN$iL=5|WiG&NbUC-4WREF%-+<QCL8;w2esf8kJt
zk|4&)PCU{v*nuT_O8G89HM<O$VsYQT7}JOz6UdJhLq3ns9WpGbRKvGCl}f^WDmtmx
zWW3vp$1BnqG2Td&8l(|>b1Q5HmPR}1G&L2r7Oyb2d_whyq!$Zzr-inU;EN*l8#QnL
zU5wFv-kt!|F>i08k(jsJZ^NT~8-B7r@+3+y13yK7O7`2S5{+cvF3Rco*~or}D$h;H
z4}cyH)#Pz3`z}0Y=}<M}y%u)-9XiquRw{9HG(TmGO<C=<nw^~!!c5^F**Yb=UQU+G
zBm}!O{2=Cw3UvGZ)UjAIUN-=cDUd#p11akg1`EDEi9U*aLrMxV!Pm=ZR~&rB@&l_t
zvBM~WubkNkzz&2am`|ZsKGnMxoG1DhO&O^^?b%VL0@ZgVaj<wxq~Z!kE?<;ms1=?4
z$@KDI#6acm<vPuBHEpd)KV+biAOWUb1C@Cwd)ucp4d0Tb;WR6+TSl`XOjI%tlNJa$
z1N?JpIw!tRZh&mrF@w8T*}d4+l~YBD8UMym6!{CPy<q<t{So3ol!>9pWi~wYS+Hzz
zF_}3oE>>15XK5XGhqu9VxBQBTJ=Dq-{G)&&HrU0s7TmO~;u%kbmS;37fs0*G9e=Uw
z#e1!l%ZrQg1dw-s7nX)tKEM6}s9SWzhlmXtUu5!e^6S%)2gE#R-St`(IR=*fd`O`0
z^MT&erq!z66ePOW^|AX$@0FN9yuR?)U0R$mHj&iT$W*6|MfFd(c<c)_5LcqDMc3&r
zh$L=T<tk}N>}aI~lRYvhzenJ;bRkH!!`dk`>TGt@xMIl<JZzh%0qj;4y4`(!^OK>p
z<HtffeuO_{FH*@7HkGAU7=?Di{uin(^y>)J?Vm-5P%|E!uwV_+0Y*08U8MwL=X-um
zjKIuz7P?}ySf<(nDgsg%fJ81*#Fhpu@e#*{=eojTl%@%Whlas0_H|XDc>pbooH#j(
z{nWeQ7c~+5DrWKbvn>8c&SLuq_@gE-?D2t3+b~p(&6-dW`-kZvcV&i@HU70v4QNl-
zRrMXi(A<{FbCQ!gt2|UTlJYEq8MpBc?Vb1&Q}&u(PC^M=?+S@XJ(PoLAf+K_gwogx
zDTq20<@hKpRnU*tIKj*s2D4aMZUnK{(Q*i7p(cVq11k%$w41Uxi;E*U=>h26%|Ull
zp1qwQbkm?-)u2YbLpje4!xE)SALZQ0>JpWWq)SMdwtty-nAN4jeJ4UnEC;BW$l!11
z;2cQ12);qI>(<e@t`Hfud4k>Nhruq^uo_tKInIJ&YNXr|H4$7JSi^|X-8AgRMIOq_
z%`m&qN4u#9!dhZZ+f6C5AE65Qz}&Zop_o#DZ*pQ~6@bb{Qh=L5m%rp4W))yMo}erF
zNR=ck|6y+2Viw9d7L7NEK=13y1_e)#uL*2F8wR#m9B%<Z{)-c&7?+edQWHVLfyI%?
z-c1~TK}AV(0y)rfr<$GiU=gX7Pu+ww{~6-h84o=)k(H}rn#kGC0vhF)3RE_dfW8%!
zx|(-5l7QZt8@Ct%DWEjoAOgC#lPyX*yM0PT^IgM06^rI^km4>*iefBMqDf5z^#&GA
zBDfIE;wIn41e%D1=ou~N{sFK3CQ09+CL++%CopIKNz(iEXz~?$?VpWkzamCZp*$64
z!x|swWm=hYbw~_cIDEL$_T0LG180rO+GfQGTkTDppr{-^tQc8XM{pGIJq5P-1xGWo
z+a#OB{v_ZisrXLl%<IU0xQ+4!sz)o;d`MWGVyornG!<;unr;Ir5Ve8Z)M%6_n+K7n
zcK<wxPjXm!uxmx7X7D$^X0B_Lvw%#_4J7pY6t@vUvAa*xQN;V>Vjdn4NfQWY-BbxY
z-SH{rq{+pgW;t5Bi6n=U4|t&=(<^W~p-CHbCs$nuWy(a#gbF*e**-3C6_IxFlBn=+
z7OHU79%57d({Wc^n@(fb>hr0s3ijOpBSJ|}lG&`j^9sNx`_1fWPVjz}BFeXf17C>A
z)-CO#a98tlXbNrHMv*^}lc=>rryhhIIW%J{YJ5B;V}Gr1Y&O>J+`Rn?HNjMu4{d)a
z&EOu{GT$m3#AfKCWApQaC`?AlkW-x+;8Yz}(F7IZ9W1%DpJ%VH(}bYkFT^#yhmqb?
z5*cvuu{FL=m^o<n`oQA{;OLMgf|K>^sW;>gCWqVjHWh7!d<vxh9%pdYLk}Q~y`62=
zcgUoXmxLH)^l(xdKF+IbdO+qAW_s~^!hv#0xLD;ghrpnOrbWJgU<?lRG2$GvY8JKq
zz1t|&Evk|b;4PyQ;95?AY)-@2Ni{s4l2L9h9;GOwL5zaAOv!TiLHcA<4d?XX<J5rS
zB^mSf!Z9gjYF$Tifp$S}qLYkGL1u8+Zo6Kf`8fX^JQ2EdV4o)(oT6SAYys635Z}Ed
zzH~@DanRlPy(?2qp`0sCaDTWE_d>;d5KZBWGnpzqxM~)UQp8m`%G=RH9LgQ1*@;3;
zyd)T1C?4w!J`c%40m2nv;b=)dA1_9zh<MMT>Cj0&pD7-%D9J~RHxi`=X=Fc)^-DUN
zr_0pv=a9|A_T`rfV%ncd?Vt@JnV6rz*pcA#0=(3~2|gmThv)cw6&#W|K3xtIVyBcA
zAs-cqKkVwRz*djaLEv<P{%IJiUzgJi$V_oI(EfYn1~!Jm=5+-Ni;6C`=apU-RY|n<
zuM34OJ+Yw)gvv*`2eAH?YzbIjg;*%#Af{gSpLP^g?XRcGSUNL8W(%c@5(s}V41~5!
zmiQ-XBv${VfPbGJ`TcU=n#Zsn2pzfw<1FPzWYQusACLo~HY5bW;~WUA?N$~Sc_M|J
z-cZ5>e7-o(WFUhe^N(}K{$2_mL9l9e8}g36F!qCw50neJO~<{;ubExO+*eG)I@UBh
zCW|IvIcZS=;r|6M^6@q*8!6tl3d#E^?=Ty0yDCEunct#Rq5H=~#P$z`mW;v?6oW+J
z4rL(glq3aY-+BqWHx2`Dth98(zKWJbQ(>}6srVc<5!CMO9;iubW1SfEnqW|7WBr)e
z<(B#~xUS?xcqZKqC!L)ZV*3n{MubGB<RIyV@!4T0rv&((0f?5$9=Q(`cqeaCyxn|t
z4<aam==nVng!(Y(#DZ`?06aGUPANg4CIY8{1%b#Q1fjS|Do{(Z^KV>sva7qiVUdAa
zGV((NYTu3Cze`Fxc%WAKIJM}spM--}p6&|9XI}s}hK|oB!11`;T0c_AQ+sNTp>WD_
zN1oa%iP&rQ+@Ad)qBr(SfYp94J^cYaJrqB^4^Lb6!~FLV{`)BZeT@G;&VM!fC5w?d
zU&rIK@_bzum^|Ml1}bN|5CeI9GMlf>5%^M!z#qp5@TXA`;HRQH5MXnNWZ^`1d<GFK
zS9$50`c$c_!`g)>zj4eup-a^~p)R{yQa8aZd95YPOOT*u#6$GHI<I4_9l{fHDZ=e>
zy>&0XD$l%$UB%-XSPN4K9J>@e9L}%Gn+R~&uNj@$#>%RW(ZFj+GsrPETMs?Dh);{J
z-J4M*;`%){1t&o#F>&AE9=fQHUm&=SoI)IIZ{P;12PqEV$0={D<c2{z45C$Y4)-{1
z7$5{c<boHNSi=wWRvR}w?ettX0I0~gycDsqWn3U7K%hel=TV_7Uqi2nw_S9|mag)n
zuSn?q7~8v!4sepuNEKgZyXZ<o?96DFD7ue(sAiG>Xtfb3;2SzVx>^t=M$@1z9$&j?
z1*XQ+3V{Yjk`3g=JgaQRGk8p+X<RTs$nC6b>WF{H*9o@AxMYETI8qt<N$EtxSPM7M
z-z?>*gMO&t8F~vP(_lp1AY=$W%K^1CTtnf~ontsT90u!4Z(HI=lMiG+YKG^zo@VM!
zqY3eX001|fT4r;C=BZvLcrOt<nER6J!hQ`($7V&p2@I<Bf$WD|W44=c@IfdOB%;le
zi0x|X&PJ$j7@#<v3g3>aC)t}kqmKMrtzpp3Y8ZxYHe6*TDBL<uV1pu7Lx}=TSkKa(
zPg+vMw&Soj0#Dd)(_&!+VR?ZCIrNYVOdZbg5-#S%{iLYv8)3uMa03)-?Zz!-u5Vz+
zCuzy0q>iALh1_DC$U<LQ!}4h{0z@Q~lS)aTa5F&0hgyg|07#9jrmb5|Acrd)=jjQ9
z;GQY2mAD@Oy*0q{ow%bDtP-zh#@HUxG)N9Ew=yC9E!>}WcJ~oB+cw(C9VLVeE~aZk
z+5tN$1-r%!g0{c1yo_SM*51VaoKthX4d~K7s>uOBP3)1iz@7l?m_qr|^*0{}J!if3
zdaipaNxHv5bx$)R64~>HSF?;@osR)iuxtjcCI*|vom|i&eplcWI|!L<X|Zxb!Nnyj
z^4#wh+wE^6$NiFKS{r5){~^EQbhB5)W_Z|8lX1h#2qz|CM6K!8bxGQTbrZS$b+Aw9
z8>q}lh}!3I1L77$!^Op>xV}`Hfh|anTeqBb$bhs9d&yAv;aaYm)gsxY6nn0WbO-7A
z2qM#3GaAs~O+a88xRq{;?W?2KdCkCmcsP^Yz$L5vCeq7nkGxeCvf5l1)@+OIFq{i?
zmnoYP)&%McNyPDaWc5St#O()MZuS5uNLBzt3tNt8D_wcY_GU`ytOGAbZ|m_3-RMUb
zqoVpG6^iGo{S9)ge@;I$IE`RW)6d_~&o}AkYxLn&{Df?pTJgj8<T`k}E#78}*WcoG
zF=04NU#UY}b#{V6%<}SvB*ZMYeBx`yPQ>0;U8;gRQGK}N=rYYePECo`8`1;D)dgw@
z6cD85DFG&3YR19j_lWRY9Qf54Hk}LO2ia^2AlVdP3fZo-{Berx!nK6LJ%UQ2gv^Yp
zCrF<zA>H7mkk{54ZX?Z9y((82*PhN)n^vRpp7v(?hsW@O2>QP--<SLXmklFWcSAB%
z@Q)PjhOj^^;EN~kl;zKZSI<QEL!oid52gW<;X{VSfUzyi6z`0Sxft%WoO&~Ckl1gK
Q<tjJGTuaL)=I6El1J5L`1^@s6

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.doctree b/.doctrees/autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..4ba3e10f237b53237988534a240ebb1487aaa0b5
GIT binary patch
literal 47038
zcmdU2eUKc*b=SvzOS+R}P>~4Xm2J2c=<Yx!WqcMgAS@FE2L=HGA(pw_xtkg7?kqF2
zk`6;csNf)b;!t7iisQs_65AEWIDa@*F=b-F*hQsMu2lRHn{v4v7qN-sIEmw;5|^E%
z@_XGqpF7(-GrPx@Ri(YzneO+x-+TRjzwYUsd1UkhfBD%B?4LVhw=}(Zc3!E~nr2P4
zJ$IsJR@)7=>3C0i`+mcF*ekn}mU6~6+g4Te+zlvE)%1GJQk&jM&z+{?j$XItMg4JG
ztvb5d4619R+L$(e(kp9|?x>?Xbv1Iy-B)j{)Xg*VExo1I^`<({h1FW-=>wIT;wUZ4
ztg5zcS{@03jFDsNS!!IKs*icRCu1!I?RGqEhdWsbI;w4U$3*{@+!1Ppx71!%+$qIz
zEPc7{z%FEm&3etIK95)Hif!X0jz^nHgC9-Q+O6=*vQj-|sVh{b?T$OjDqZrl8Ev<=
zRl88ztnG1E7c{e>E?DhmQ?(Y(nAWL<n{GY=WB7m2Kzo5VY(Z%|rqa?ExVaX5b1g7)
zEu@%hL2uU7v-6gX-d=QcQ+rw4;Eu9$5%Y7;Xs<`_w7szI8}a{Q{GY}DebiGJ%hFl@
z%I-)@(JjxtqG8tBb@gqf?4ENk=XK_?SJB3n98YO$muYjM$(FQ>;23QSkIOKyw)-dC
zscS=*v3S`PJ0;=(60hY>;1ZRt7N7=@&kph8+J4w_g4uGrqb#@UidCta^@`n6t2xX%
z!pghb?>+wRD;|FLfrtIS9;v$}ROg<3jY4GciYpfnQi$2@MnkdI4k2b>X?K@M4;XRt
zk;|UDBl4{ROT-8D)+zj}o2#p;H8tgr+tFk(5qy*Gn6qk`?N)~YDPxSoyg98qfhuS8
znxlC~r^uj981xi;X8KXmE9xTx?uiPZ{IzV>);b!Bc5b%JGacty%A+BaT4fT+NLjU%
zmKLe#UccB{TQrxA#rwGdO9wL#nt5L1(Eax^o0l=-x3<(~O>eGN&LSB_)f^sx8sD?y
zb=^@dr5=&5{h=6rW4ujXIP(6u`^ZC%VeND`wbnd$thF`=LwIU5C6Pb=2oC))5_H#5
z>yDwf$ey12mV$ZH1L%>6hN3q;?J0L0-<-7;0B|!k$=!nQ>e*IZuj-EXIJ<mCv6|jF
zcM|U7*)A7Cxf7_zaK&3<x*f$@RUOa0Jg@s%68RADWthq1=tlVH0~mC>w6W0jABehM
z?dm@A@srvf?TbjZB<BgLCi6E{cs~lh4+?)LG0Fvr@xh{p(%stEGAfcc{~TKLo?eoE
z6NQkNx=V6LDEoOpZ*)*c>6N5N_xwmAF4fKED$lY?a?I3F`~GC@Xd!p<Y@@ztw-jP|
z6D>=%RqM3Mg6Tfg^#Sl+1`J`7V<buO2}I(opcB-KDaaB8C;(Flo{t|ER7<3j;Cm2M
z#?Fl!Hbmp9ds-m<0CQs|sKJ`B9Z_>@B5JxZBJCdS1c?DiN5s5#L+!wUG@RBde)HW~
zg5FyxJAvp{)$75a51aM6;+WQ5x})6+q7OLmQCaot2}m^J-^+zC(+6GS3fQ?%-1TKV
z7H^pL-uU-&_hIio7$99wv*f|rp29*H{(Zl(?%L4W!m;|JeoKtIXbjMVYL~mWXIDb>
zHtr6_P8xMCYoF?k-Nqd(F1nA?0KT=SqcCh#Q7=0ZYtL%WY0ocdUvQ6RhtS@ppwa)D
zqHD^KCsZ`v#jF?uSavR&3&T9^XJ&tDN8JF%8g23?gIQyV4znSxag;0X6Ksvgn3(R7
zK&=F@3w83|0CJ5RGROn!imcLIUJ`hX`82^lPeWPb(zK$HNwqPG0>&7XLhpdB5g;-4
z;dByVgXM-0UL&U@*h;`A!Pc3Z*+jl{*laqOq}V(~>~cHKhgkasY93p?NE`i8IzfFS
ztbI19!LarMG-=*XLcQi;vDJq$8@PpOrh8D#i^_Vl#EEwHm56;s_;hS3BGMtT8nMA0
z!NOVR^q;0*({!a`>sbA9+E#4g%byrg5@s_sR+QG^Xrahod(%eUOS5M$--CIogwFlf
zz*T9^ExTLLkGU3T1vSv!M(t%27WO?hNsEQJ@h>EIxAFZzJd1>1MkH)-Ct0r(T{ixi
zigIxB%@j8Ce*-z`&77cS@-|M8^j^*TdAH`duZygHoYvJdRjiPD?u>s}QJPf^>yn|b
zdIdDiDgAlffeH^^JNuFzRvRHeuX)Ts5tPJWGq!C?SeaWV%yGuI(NUH{u#yQIdF~Wh
z3e#CE*v2Q>?wFXtYR|c&d{U~7xf3hxdYxI5>z}GPX0?Jv3`!l(v2*XiyqTR1i&5MD
ziA=nQyB(46UEx6%0b5xFRCx}{Zc7q-Detl`PTQ00GJ#&16Os{HOHTGu%<qk>Ajlnu
zCD;_-cq@Hnu@n8?H0Sv8={a}2Y5IrL{^>cs&U4P)DlYopcc6smo*fjLX`22yzZ^*-
zeYWa2NOj&h>f(2Vsr*bTx>ziS8sCC6_s*Uir&MM>FCoEZ&OOMi1EpE=q-*>mdJ>0`
zvhlshxh$mC&%84%0+nXdbd<!M3-`c@gKu=O`ngz|Wj2{Tw7+zamB-k@vM*-Ki>07Q
zY4*?|P->RZcljeH%v|?TyhhM73J_dYB>vrvB|n|xC={Y4gv7~{yYY;dWF^K5RoQKj
zdf942JZs;}D;*EivEh4Rd)fG5)XsUJMx^vSRl<Hh95f^~?Hm7vRvG_|kKRa>(x?B4
z^vSq~RB`ypH8*-+k0y@ex-o$dNfQMcr8IFG72_e_buBBlfY=w%kSWrECTw9c>l(MC
z5(gUN^*A(2_=tD;O*jL{xS0No8JAEY(qzo?^K|?k&)DamPhiM}2<AJeB%7=lm*FsO
zm~ub)5lbbf)TaKRYL&{F)8BmWq#E2zM5?@FfP&|P_ImN8`IIn`kBhyN=43<^B{3s5
zEM6dad{{&Zh9mS#34QU*R5o&i|MP(;mUS*d);Wn&zL(uMIhNA2S4^v+S~l7Kn1zLU
zwRX&R-F1qMyR9q{Z*b2G5_YTItUB$EEWdFvNVSWRRQt`K6=^BQ-Pr0(wf{g>b{k*8
zAGCQj+U73#{1vGZ!^Y)q37u=vg5@PaKX+v`Pw2T<rEQ2#>!Ry1oLB{SM%ZYMWu9za
zV)s($7N5SNSK~KOKtIFVA}m876JYhlH*iR9lgk^aSe6o1!$_*`Wj&KcWK5@L->@qk
z+4mV1F~oc`WM7_kSpwd@IY_`+*pnN{0ly`3HHk>%>Wkp4&x`QFi0I7KT|tZl35M?r
z@H1K5?}Fh^1_?G!^Np*a-Z((LJ48osrlYq9N7vx!oN+DtypDYyW}nxy&v&rTBlJn_
ziO_MDT?atNPhzU2*4Gbe-Ze<5VQuXjW=4gWPKdb?z7Y^33WZ7PCb|}3&waql4S^^I
zFK<Y~%O$t8Xd<B0PjtbVo5^Q_=^NibbOgxM1xii^8q=V}co`x(vb?~OyHPbGmYkAn
zkzonfl7b}{%1jUe%QdO)ZWv;0LbY&BzY{~`>I-2<4uk%-v{O=W7Fy!L@le1s9*UUm
z!*VSH3XV_8wG;}DktGwG<-#%gS*ebUK_IqRL>J4ro<6)CM~rAgQkg=p1i{<G_+=)N
z_YJ9Cq;d3BRK@U*$I;b699<o;{`cfsWN{?4WFf3Sk!z8~5!ZrSj>)*83(?%(zq^J)
z>}GotI}Uy6q%MrQ4pMg*2Fg68j*z-tBOp6`R-K8|3Doro9aT>dx;KE(y+gRGFro7W
z^g-z6sAr7O&C?MhbnIx@gbpj>Y>&nIQM%g)N$Esu`zCecfuK%W_YQbS(7K>l2paFC
ztAinShXYZJ*p-vSZu=cg@VXPrrdi)#vK=cAg*zU|&Q0Ntp=w47w<6af%dT8Y3WXb?
z{zh=Zk?M$oV|2RTld5qwg+MxofgYDOUJ9i`OB_nmDcJ|*T4Yd~qc0zmYboUH%?Rs=
zuRkMIQP|f^OCPSz3Gltf^~f=+E!O)o@v*;`T16T}Uq;o8eC#`NEwUgITC(u5AIY`I
zf{1G&KE~`9rewMhP09VcYbbndUvEC<qgFa23*)YXksX4GiZHV6qxl$F0GK`}BLN9c
zRt6`#M!2jnC-VjL!O3Q+Z;X?b>4<SMb~J2Gw%kz8R_ZEtM{Dax%kCT`EejjmH#3`x
z%IYL%Z-=V{Ig6GIlfkuAU@-jb>Od6ZXY)yZwk=qA)0<8na&}uFI5#<4Le-4q?0&fx
zS*qn)Qpnk6^fZFkO{t3FcuiUg;dKu4JSuIw6jg<mEY$1?xfU6!=IF~u<XQ^3I=Slc
zlvGDnU%!aSc<i=YAB$z&KsnLA*Pzb3SoXJ4vq;0}MO4j5&R&vhk%f`al7*bTBG)1d
zBd#T7v5YQ6^Kt*~8VWhv-<zC8fGeG!g`wBM&#r=%itw|kv3&e2ikUt|BN+*bHV=xn
zC>&RqqWJ>)plExkdyJyJfsPnOV@JcLXgWC?KlX35oONevCkM^cV(spmu<htD1>edN
z-;VQE_)qY*lnNnzdmEJ-41Ie`Ad1nqD+fm3ZVCkFrf)Z(YDW5Yk6eo^8FMYY>6<21
zkrmQuUtA2*8A&We-*TAeA!*yCxGS_|p>OY%Ymwn@j=ub!TuUKWC+XWKq&l+tdXV&u
z8z?6q=rySG(zh>3%_0q>=TS8yefz3hi!6+UmMrw`yK*hEFydNL=o?*#=Hvd|H5B^x
z=4|vWj*RInE{wnq7Iy%~D#GGM#`Ceb6a@9j9EnPhxhp~Dt`u%8Oy+z6eUQ1`)JsO@
zE~Fzy=Gf7&$(&JFmeo4l3B|90u$#J~*Vb>{?Y=>>xz2|7P3Lx|$?RN*dox@p_*`0z
zkkI`K6&wtqyCM+92;E|m&@tXYjVdQLprmh@J`#w_P2;XZ)r>Umm|Tl2DRV8oY230@
zMOJ7ZAdTY<&j}U1ygM>?#pRcasYMnvd_-y$DY8Gr-5V!}>ARTTFV`YNcA+IpU;aR@
zMHWO{3n4qRUpO%7LUdsE@2;UB`<1<sJtd>1Lwgv49niiX)+z$+&!b+RElp|3)W>#`
zmB98zVEepqWMORg1@wXK-=uCbY=4Q47`C&cVPm^0Zkbk2EA_Uzez1P`AYnb$=Drbq
zTS8nXnEwYw5KbdYBnpS=;y*={`9adhhW398L@^w{rcWH-G!ai^@s>$<tN$Nbk`c$Z
z%eBZ-FxS!>$1jqq7y^#-hUWx^Uf$i92gDCbO(F&Ht56lb%>v>#%C*QqTxiJx;>YD$
zWD&%*V0glK;gF=WeIR&>qlSXu3%wCMp25<=JM2yec>e&yya;$d1Ow**??m$Saht>>
zaQhnI_76oGER5T}06!(i2WqAtnmtK9WVr3o5yNeEG;G{nb?~I&D*Np#J$hsvk^96T
zkvq}qzOj3zLtrO*e+r%x=-p8~OctM}QiGX?e=-oo@cT#ye$%GT?DOzv197?M;h#g*
zjQIU!xfWUW<yzA4JHC7MTT&Ip@td?1x}Cxs9#0F=W~Ba_sF!!AW>2YVtAZ!>&gPhn
z|5R!lDUiR)oty>8$0lV#E~SD(OBNtMPp(B4N?Z$oJZi;okkaMoAno6ELqYT_dLw!=
zucc#p7>ga4{xO)U2&P{J=jFllj+E-7Itfak`VpY|Cxi<Nqq;Ak4^+Q}`p8iIRyty+
z&W?tS>V|G3anO^=tV&H~f7fqHKG2o+&W87m?K{(CcEbBATqxi@twxwUbSgL)gjWMm
z4B>A|BK&l)^Ob(AlV|JegMsASV1F7_GlKo&axJn9&9$U}ee!onAC&5d!t87HyQCZj
z`g3XHrEn~?#NjyocS%pnwaDhxIr{RVTuU)$Cw?mab*YZ5&K~6M32vaA1fkcs9zCMe
z?L0qC`iayo(m48eRK;+P$5HxCmJL&~%qFFELQ57}H!as9izBWDv`!c>Ov-d6nw0x@
z*iaZ<hPx~&XqHau!XWG*b@#(sMM&KNxGoQ=O9N7$*O9CQue%AnZdo|8Ft76k^ug<X
zmAc7z-Hmj_cpW<$Hm_sbVfnLSfAO}%#@gLCt7Cg}5+XZEodW*}QkPO8OdnM$HyBP=
z2}ChYcUzLv75+7o69~>t=~}3ok<vXX*CNZ#TuTb2OZ=MY38{*#AV0u;6gkZEF=^YS
zKrFPxfjFJeeMYWDM(A?%<$1Z5Lat73hy6RLj;y{OB%$L5%E=FU4eGqxVSgkwi!_XW
zfT|h!+|T4%WML$<WZ`oYTV%O2Ltnt>l3_#YqxrahcMXNkW!w%Mhp}`v7e-(Qo4W_b
zD#GUW!*O}oTnd8vWR657$lPrpbN3227AAAPfIi6FVd^C#bJx=mBXjI%*krESQ0!CU
z$4%?EI(OF~Z@WcW+&6*S5);-*-A=$;g1W_uhNSHzl^6_ZyE_oYNZauwX)C@E*9;`*
zW^D$lW@K#-$+gI!oNGy8ZOMhW_eynSh4K#C*2e>aav10%(#A_cN@&T#+n$nZkpXUw
zzI;xurI@o53vquV)sfZNgXC@8Kskv(uW_AwA?_8aU8HgJGOA`IZm-I<$l^$7$wJ&V
zZI$KE41Gykh$H>cq};#5hQi$Ddf!@&;#WFn3!|@tvmJw(ig30|;IllOEe4%FTO%O}
zwsstB?M~sk!feeK&<9()n)=7s+95h(Y>gcao2@ncw;)uO6{o6I)VkWhOM1%t220%<
zs-u}TybNM_ki0Fu?R`_Ycc#lN^Pl%E(z6uzm+&?$TWyLjmP+rHQfut%zS7*G()Bvt
zGsQPCMc*so?n;-<iX!iHyBxB^yAVIoL=@kyoa@CCw+EsaJG_s*qG~65e+0d)8$MI5
zx^6qs7xm0$Gy?B?@Y0_^UR2hWh^(MId(|4fMXthLs};U)46n#$4@~e)<$n!LV+Z|7
z<D~A!X6Gv`9-u0_jRyX7U|9Okz8;k-k={%aTC!l+6LKvwE+n*AkPtkmojt*j%{n^Y
zpOJWm;-5%0XJtqobh!946hDFnCEw+2{2ESnJVQR_%&Y)?iu$|TSY=v_I{pM!+Y%Y$
zJU)$>Rmb@CARxupFd$zf*<ELH(oVd<+Y@K!=@I#LxfU6#5?ZoE<jZm`vWOH~Vz$c>
zk*`WM4<RCd%zP71ycwK1fQZ}{1Ddp$obXHdQ5l{H<MM)Sh2!!(RLz(vFOh4J#ih`a
zB`&X&YmvpJ&{8xmk4QBSAug|DzR3}n+yI5*GD{`*mm2N5qgU#9TNS-JlQRl@1*Kqb
zm{DnD?>16-2!B<krQ%hxO*_gOMgnS;wz$JChLLWskB@j83T>)3Q=AlymH4LW2DLxU
zLdH<0k5~1cK6c@^=`+dD^BptECB#ONndFCHu{<-$Zj4(@D&57RT`uqQ5|dGS{q?N^
z%u*+#Wz7r7N9mY&y<15s@wY*j*bJ>CW@z(%g6~~091ee$c{rfy^vZNLF;LL%kBgd^
zBTTw8HN2U(snctN_-GDUI)5!&8HMm7x>%u#?x>CEK>pdaPRY@98}lh;Ro!1Yqn2uB
zbJi*0WwEDK3m47Ot!u|jtE!do&TiYZ?9!U*yp0MPZQChnO0!m1Ap`$|pZiSVMXpu6
z5RaYaGltS-VzyQ?tx|Bx2g;JA+AXtbt9D7Z7ZeLMO|w)to2&DsBf4cfQ_v^mN2R{4
zq_3bUv#4M@dc7W0V?Bvg;KXpj*Em1LksC%D!4xZIlsneND1NNml`8$^M?Tc2(IdYq
z{9>LEGrmBjd6ZQVWuM1a7GYCi@I*pJyz?&{B!Bt~;B4RaAAgP83a=Ij&4?$zk;juk
zvZ%q0Ut?~J6Oky!QC|E?=y|HbJ>Q*GP?cxGDikbFGcCqn<4=$^zjK#b!X8w!RJBya
z;lq_Zzrily1)3};BnD$R{4s7d9l3^5fVol<PtaB$Xe%_frzp+X?FL<<LvdMr6Ilfb
zNvZ=XbVr-&e4<1tXr7S7VAP(UivQzz!+0Uu<-6Cd%YNjpqlOTgUM1&tohv7!Cz3Z-
z5eh%%q43p+T{?0_dngX{Z8#+Dp?ycEbJ#Z+%^Hlm2{n@oFiC$zMl)66{>VI<$%QnU
z$^WvKc2g4gD=HTvMvRw?SIq{kUPmCq-Mx1&VSZJ$ZM9ZfUMn@*jn-Pttd{of^_?Rv
z!%c`5&cY7%a`&tjXYLo1y+ht4n*3m|?unOx`ZQ5`%VCVKs;9jH^boZ3`wefWBX+-m
z9S!?_!+lk=UPoqLw;K!teP5JlY~Objru|+s3cpXJ@z{+4#4a<8ACtlNa0nl#km{Nu
zb1nTJ3q&!7|1hQ_E0~!CI6BH2%GwfN6OCSoEu2Iy(EU?hKWuw5P?)whFI_z|P*s+H
z;^R)N=dNU~WYm#SqRSB}wi(J~JHrcI{Fq6fU56d%m|foj^Awp~d#IOZb{#~qKb_9-
zI%t2abj+V0Mm~KP8o*(q;Omur0Yx+GQ)B^lC->8I#4;;88g^!_R4P@aUawTvEwvsS
z<h@(JwSDK-MPE{R1(d&y+5BSZCM+?YprybQ>|X7)&9##!gWIy<xwJey;fwPVad;xk
zW-p2;^n)l)3UX!Q7Xnc%6F-v7#23kXm6D>f)F{XAWKH-=f=&k~!cKoXXld>g{t~KY
zyle17xfa=71ED3$U4x&>waD%o2rV)2J1;HZcyy&LQd8;3M0;vRhU-zPo$sf>OiiIh
z>M^va!(S)-;MjN~()f#{jhEgJ=UUPXlvu^zFV#`pKupbg8|WHo<BJ(6anF89s-m!c
zir=$Oji3EN{Opg!&$6`b8O)RE9f&Qt7TI|38l{^7E0K=;np8z`Cy<uG#VCK+0vP(n
zlaaRnuC(o0oDkorkgnyU{c4FXC`cy_JWIzbM17DtY4|#WsW_OYSL&EFq`-JHvKr-1
z$Jt1CY)GWOFG+0{ZCpNVD4zPBM~gCI!%Nb}OT&t5Nx_Ch>U%}1qOg66r@mD4q^G{0
zN!y;mJegA8#C93GkZy<W?u2;i+a*;|+zEwKU#b(*lhUQqwr6lc)*K^l<)<Wyp{y=k
z+WWqCFcV5&UI=^AvApnkB+??w3pc_Ud6pN#xtYJdkl{sZez!Vr&_99&hUdi004eS=
zyqU;f5fir)UjMLH*7hQj-iZGf<NqxF?>ouv^oyliUw}V1jK_y~tfpk?cA0Fz78t5@
z#1<IX(I6HW;*IIjgFDdg=sIJ(lIX!`wXWE9+S+p~=)eO(2iQ^34m`UN?b(wIGIs_K
z_H5VHIcyie_5jDm?q1;){}Sya@Gpgq(x$ZC+E(5{Hs`(|x+q`P4!ZmJLdRT7Z>h9-
zX|9bM{PtX}Wu87TS5q8C+)ywr^xxMhD$V<q=2@kAR;d!Nq(s7owPtgqHJfB>o$i=z
zwyi3jv9sSj&#{!Ijm31@hJhBXV(G*75TNd;V>(LRJG#N0^2?|-d~MUNj*aegwQX6j
zSVhQMa<}?tsJGsdJ7KGiqc>M=&)tDrCA=pq7PDr}bI0zar=f?tvfMGV?X=pCx1`N1
zxs$7EQ?(Qa*1;F_T*U^DMhjoISf=Tqsn{5D)^j&5fY<US&=#>%C@>|ywyiq8ZF9^j
zewI)@MOST4d%wGL1p=C;qef0Y?ryWO`5w1*Dra$zx8XhRPGi+aZ&k$IARD4~^D|=a
zj9sDIuoV`;=ms>23KY`p#8yjfnHIgA@3aa-ZgQtIr5=18b$6&|b=c3rf|KnR8g)l&
z>T-J(@}|!yRulF^3EYQn;LIc(QB|ok)IPllYw<_fp_J!Nww;x^tLCul-E%LDU#zH>
zWm*-!*98yq9HXXI@qXJGZu5wgQ6|N6_av?)L>Pa8Vd+kDx$JRztPOtc3*#3OmEG;s
z2SlB%!9;cR$#b{WmF8+2dw5{+8oqB*o1XhR3lHeCYIsgx3(xw58l4$-3`24S-GoK-
znjZ?<m^-bs9TRWP@lUy%bOhSA8gABAf?$MU*$KSgSGAhB&kr#(><kNY@_u-#WvlI)
zSvlLFP};7ZZQ=JCm1s>CB~_<#TDNr`tTUCGZY#@m>WFCNxIaiR;I^VO*y__{BMw5h
ziDH!%y^hYE^xR!^1SK}MFVUz}n<T9Umv6C@Gr`xR8{Hi=(!|4V5d96^@owHe>rS`u
zY*dU772^}PETLp8oms|)rV0%T_X5pvTK3|?0#4iWtu_3tra5m~s~Ah?Q%TkUl&oO0
z72M;%9TPY|ck$&{!k*LCN;M&#&V=GG@x@cZqL4kMTJyT<tS}u&1q+(fs6*MLqby?-
z)f~+`I^pg?Og1XP;0!uYv0D@&=iKe~TGPQhFExEttK&c7cZ9ZBdG1!bp<tjr(8j2?
zVW`)!@BC|}ZemHBn|4oC!wypX$j0L14OV6*MD0`U7F0~5b&l@9{yYtwL-d$cy}5!I
z5P4xIB?`N}oG6CJ4<UBzrcz@AVT2_kgiK>utziVOp#_?%V$Z63eifxoRn&%}*DFe`
zMj4m2dIw1+kK1Z}g=e#K?q;=l+FNQbyV!$G*+iv;xLt47bc_wB!XJ;o;${aLg~)P{
zXdwf8pD2&JJ7Z_*=P>Lvq^$&B#weNA*!gif$&bs%qXfObP9Ix=p2j47JVPH()5n8!
z;S4_7?j+T6)>=$9uXm2sImgPMV`XuBY%hR?2?TT=#Ll%b-u%d|_9R6*7~ruFckZZe
z1GK%z=|=vmP%rL^kNAKL(nT%R4)JErP+O^)22lXx+w}2$`nVko#W+qM&(O#7^l=!k
zhBj`Xk3%E$l|JabxCXs9*PwT+8T9TogI>jA(5qSudak=c&wV%OX|M)8E!Ln%h#B;V
zF@qjIVbJ3#4Ej-{K|gvlXj3ZNu58c^G=pxi8N_uB;=*lrTXY04pKPn@7)Rt(_=%sP
zMCX^|&X1nq<&a7+Vj9b4Js4E1=y{2gLFvhu#7SQ(tF$vFADCr19x~ARNm=3wNCVPv
z)ofFe1wzy9I3C<!1LKFq`6m&ek9oHtrC?(>k`rRn#wg+ngKQi-y(Ku!2ErD-S*^Eg
VH1Jj_cVf#Q07K+fev5DZ{{gg%oFxDN

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/utils/dpov2_trainer/index.doctree b/.doctrees/autoapi/lmflow/pipeline/utils/dpov2_trainer/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..c54b2d2e5e1172fedb592f68d9e47d2d5feeaa61
GIT binary patch
literal 89960
zcmeHw36va1nWm1ub<4JFqr)X%>b6_mk{1|gEP-v=Gay;Eg$%;7T*|J>?yOXGWl@<W
zsj)pAi#^toU<McWf^8mV#=|kJVK6g*&9UslatthQV9h!EFzW>a`-Zo3fX%Xh99bDr
zmGxC*c1a%Q$?v&Um6Z|Sf5rd*@n=NFYx>{1Y|-LH?7wi8Uv)~A;}ce~Sn-Os-w20_
zUZFl`SAxbpjj`7@Zf}f+!!_%e@6~GsyAdv;I|@#zT&&rZ#-T>IlHDGZ%Kn6GKj_<q
zpyXAW&7FQ{z!^N$7<Goj{-6|;ZRMx%g7Vx<**i8-EmiGusbWuv&Wcs<=w<n!X3?cu
zgEgntef#a>j4ZQ8cQgd`167M?4I0jxa5&#&&RH4`NS^kFeGIv=w?1u!BUTX9O4Id#
zWWjh}S}OXC&%r|3@_o8UpZhD;ocJ_UtXJbN(^la~&7NWR)WgBRnq@yVoK?<xXSuV%
zS?X*Itx3n5vnOly3h_00%&Q%l+;i<UB!K82BG8`{giTuYz_Y5QNg<-iNJNudM3b6`
zCQFr~eSD(k6Q7%QFR4#Ei^6_>O%b-xgL5A7<!mO2UO@j}ME`H0|Hl}QB&J%4b2b|G
zRjpF35nejy73*dD1vwKvg_nrCx0>AS4D1aWR^7SO*%}LIud|8N;4BlC7n{!N=t4Mh
zMXVrh7wfc3QtQ;?6++~qN2?>(i~y#go#J+9oa8se<+nPprt4*^mM?haykE5oDMZ=F
z?+;hseBjpYx8Hi%?a{vt26tTy7uHU?K}>Cb_SEIfl>GYKoK>6ONftsf4%f+{Ph)CA
z`MD9UQNHC#2Ks<qK0^PMz1dm2HZl^8%>LHfq^S*u1Ho*~t5@4Z$GpKH2{xr%hp5T1
zQZaBEyGNKPD<sMzq%AM9k8U=uba0c}0rO4MUU9ySP@;2b%{$h1ox3wOU=o`Qlhd(g
zYgW}!8iqTks`FFcv^#ZJ2rzfK!9bZ1_wGD=Gne)#jqK%ByHYGwX7k6%-KYRN76cG!
zbFf?r?3z_pdUoC;$J&6PrV)>``#X;_@*<qVa7lH(5e`)6w-WOWTXi~22ltW^-%ZZ9
z;}{JGh)?p2jc|KvpV*-A#Tw61C}}t!3J2+1Fkhumwv-VQE~D@E@oKqLC<TqX_|M0z
zTBUI!942K6e`XkQ!XawG<3wXGM-8mntQ|DMOU%{G9mG2+Qam$`VfSLv#an4Wt#bxq
z6~9$gvE5NkOqcgK8=WtZ3uXPhnKeUtWewg$H@<-Y?`|2vGWm!?Ko`*U&bJJKF!?8l
znmfDo^m@99oKEMS!anBN8We|#!4lm))7oWtmf~b{Wv?<TJgC(=Hj32xaBFMzrf~T9
zTzSf`S}Z#os@81Zt{t_x8Qnr{qp)@}clt5OTB2h`sX)tBnHVrGMyM~D#56-Bh|+^s
z$#hA&jpp~JQMqd=1y${;^YCZ*6r$Z#%@&-5)nql-w^&W5+{U@dImo)9(5rNFDhAnQ
zmnB7CCm#_Wu4Msr*Y(#Oy==coccZXsN4LnB=IY-^GX=dUYP#D=m<tS?Q6G2cggO5~
z|32DzY=<vt1}s$qW^qw%5l6>&Md0C!d)BuIjQ2%7yDw*k$mS&L!sRLzaa43)(X`d1
z1a#E-VE35l-pEzdd0aHZqVq{L$hie-XCCLAN1R8U&+c_T7j82NQyRZ)NS&f!l`xK2
zka2I|Lek<DzoxpD*zHA*E}HW&1}J2?b*3&JwcNezqgmi`cMI?Z#xC~`-c4skGb$-g
zN;u37MKJe|4Em!WCdY5*{#xUhJCW4!Q%OO~y(H;YWwz<|(+%zbyNkG^P^I`~cZ@E#
z2CU``JdRe%B@n7uoB*L})wO(vo4d-Z1eDMFVz93Z+pfo<>vN1eKA}$<+-h`Yys^;r
zNV5eGUFQ%<6OlXVHf7PK<pVzD{WK#BpDU*jqoqntWT_@o^s#iws5VDen2^)-MPVP+
zjN0eLY>u*${gU@fR0Ij?HLd0o%^{cz<<l8{D=R^<dP!7va{9vyO&n6BFSDCtB|h0y
zNs<yr!(~MKR?4S1QQ-<kBhT&mcWn-%nX>y+*6(`vC(Z7;#-BwRUla~=PFvV?pJ2C&
za1x0oCiAz=ev-)?Vq^*$hgk33$a|_2dEpsKLEvb~K31TDU?W@=eau^xg5%W`{d6;2
zHt~w4K;wB~q`pZti%M+Nv{Y<x(I{b#hLgKuNsD}iTw*WgexF$6PJ+sJB%MY$!em6V
zJ*rIVv(IoqPVk&Z!+tSya|XhpnR>a*<tXq+@_|>#(~O=uhe!FfJ84?Uug14A+J4N0
zyHO~dY~b&u?zrxkbKMt&mob+j>-G}Ca@3u(v6W>O)<#7}(O0wrtCdSR{dJ#59fgA=
z2|fdMpU=K>t5LsKyse^gcp@CEc+tm|(dDh8=y4)kE`N-^uc136dw%n#RTVF~7Tw2s
zVtBqC1mw&bCm4%Al%k4!rDQRimvp~Ny@fAOQ<dB(S8tBIrGZl+H<>;e!p_qD5z(Z(
z{89ID<=TNq`WG!yNggHJw98Yu8!J>!KS;Ikia+Dk=Iok3aed7elZE27?4I#l;MEGw
zM5Qu8->F=CD7PcG*Q?m$IT}Z7<gtFPtFpShRX&)d!??$3JFup5GiA>T2xZ&E_HmlO
z`hK3NbSk$;l#UK=p&>Wtm5N)&b6e;M!>r9d6l&~7W(Qt%o?SRnW`Aq;(Y*X*%>-K*
zRD6v}DE?GV<DRoleO^(U<xeX7Cb!J?S-G}skw6rVwdT9%rmN&VF<WJ8+Ah-|FVe6+
zVpsB|;uIMRV{z;DZQI6k^b0S!9kt4J;#2%Y*8{5M7c3@OF6XJ7AJiBFM{K)FtTF)*
zda06Uw2{dxRLLv;1fz(4iixoaC82>(Tx+xO+^$j~I7qnTxz^tf9f~{Ue3f2hcN9!)
zRr{F;*sWI*U2zZGkn*KUN#)`yi^NSu;)4edaaKE8sX|JD+OVf3q}irb?kV|!ZslBV
zHr3zd-4k5#c8wvj)=GyS+4dYGn0nk_uM#z4MDXhq41@m~4HR*?T+90f$7TbjHnGQ=
zn=V>=_Azcp#n@qg$4K)vyIS)KHkmHvT&2Lz(=g1>+4QWWpfV5(LSXN=E51iMQNJ+}
z@u4byR5+Nkj_1oZ)qxyECS+xb^yv;FWUd+*+$iF1mm8}TpJ}*MW=gXPD6^&*wZsN2
zO9g8AqRszRZk_hLa(rvC<TFFa7ioBVb+QrmGl_G5txle3)}}{ZRzN@3YczQC%*LCy
zTw|4ex~D+g(W5skT-6aqCL*31UQJ!@@x1DF^?<q|S#wrm0_QCMh!kN06Osl}LZ0gD
zRymkg+?;9#jQUD_F3%p{af=hqVEY(PMOj9>ipP4ZQt<-r8kBG!UUu;E3j%uRGDZH7
z70I^j9M4_O?<bc@&zY!@H<fGNlH0O#Clwlaro1l7QR3|I7M2+(GZ8iM7JaGEi*r%3
zOrPRToR&`XiD<a^Sfh1@=gzPu>s<zq3ZZ&te6whDpb5tlr499I_h(8RJh@r_?M?8+
zblUwVqQw1Y`qMqbPvGNel@Gl%)TLhZhAwms((mpN{Q-2z)V`fAvWm6b!q`(Oa$I(;
zOwLxZi~}E~+(V`bS+WBf<fqe(IFylBilZP4?x)iYk$Gx@Ih}5tPB&;aays3}m~J$)
z4N+=q`gX;yhnGvI-eKTf%VvhT+ghiO(w#Q}c=3zX^76dgp;YMldH+#XA-~0$#h<~b
z7A~%|XBH2zChOh%=^tJ;IkTA>RcWFX)Z6l@#ZwVHeTtz!*#)}V4*dxg+MT6u(R@AZ
zkU9NV2;%zmbEnivFz9|jMVUyaCaIt16E9hhTNiFX{U|be43e+bvDN2GqR!M@N8w&X
zy=coFFg#yz!Iu4$W=aKpnFUqhlqD2Y7NUns>)pSj{wYT^(Z??#Lr@<Q<zea}*7|Kt
z)4EL&+tV233-oiF)}?Kw;ALc~p4#hKxdAUuyJ4&nlh0<INMY(62-nFL+$V_H_*rAS
zEwmwKRDWFFWwlQCp(Gh=uEWS)E`v-%88;o&nx=#Ds%(O*mB_=zJDMi6Ly`8C5I9`y
zwR@)*u_lpYG~(_C2x7t=t=}tE_~KZ_=l8KxKjKP!iQTRRG1+638LM6n^1O_uGf591
z>eZ@O3-Ug#Qw&~N_t+v_ULDS>yz8c*TBbhTyq68r_3pdrA95e}(Lc?B{fA0f6#2@8
z(0Ci4>1YvBcjD}+hbXy5GE%O@g1Qw73L5w^YDygGt|~bSp1>%`RY?>RxBGUCg2JFg
z(Zb+!GO2`1do;o6;cygd<Uz=VYG-OR?CJqAd}??2fKa}PT0R3Ag5zJJ=q8pN4Zng>
zkcp+><uQzcOiZ987)D!cif!Qu2$$Iw+7M5s`k#V4^h*81eK1Ywk^0ZZD4EpX8oDlm
z7=}8RtBTb}-vTG2ruATF5*AE6!;2uR&@`mOzJiRa8BRMSIeRh6wbVsho-DJR6v5|a
zjDk!<q8wIvf!U3VvC6F+bx&yo3A&b5Q!`F!*0epB6#-FCPWsqgQC=dx#7^(uWA0GS
z{c37zn3DV{MnMikqL^6ty%+^qg+$R(l79ZYTL`XwS|xvtCHbEqx~kt)sf5Hy?Mgn(
zwbL?zG*G9BpuYlHg|0*j+QgMShEb3!NntckU=#zR(Ok(=S{xCLHU2p^kblHT861gV
zvj>jkrjxgtBgrETH{l{tDOX~Ft)vH3>VwV{#ymnz4SmxFjDmcVL@^mMmtYiRW)em7
zO}l!r=0>O{AxR#Ty(`6w%OU-ly;~D0wSRkFl7<gP_O$q*Li_cQb{A<_pwP3_vzrwk
zxlt}pYj0>#fLeGTZZ3d|$WX4tW_=iv6E61k5(!%=cMEH>-c96rJOW!huJG(Z5Cfi#
zs!g|J6lB$smpt{*%qa)uvayJx=J!~5Z@`F0=;yZlGQITPiBY;+jFI&I45Or#-l~?h
zc99g@uE)~+D@;L;r1=Gm(j#ep6QiV+W*=Lq7>Q7PiRJbqjMXE#{Q{%(NN!JIl(cdi
zik5LkvWqXo@*5t2k-bOqI~$|)NPZV#l(h02*ivH)Vk7BAUt`(*JB*iFc4DO&ZEIV6
zNQoa;V3h8%)3Z`o#7L1w-H6FT!j>o|GmipBK{JmOyu1>l81SOaJmPUCmz~8IV&i!=
z#>_AW5e)0vR?|V_V#cK@R(<o)G1^k*VC|z})5pKWe7^(pjpoG-YHB!@_y9&hgGQp5
zOeH>!QIOq86m2T8Nqh3LpRHS0*Vzk+w%0$pHH&5cUm(4zt)wc9bR!`3jN<cL(J3aG
zDRlPtPzJDKB2i5I)-N#%@>?l*c^ab_*q`RNVDWWi5GD#4{FPv|2mb20Y~O{l@kAR&
z>H<)Dpv<10gU81!w5}o_)^ar-P#0jKAZK<SH8pf*+c64qW)j83nY|FBAY+m!nlsZk
z4JBwnM2>`-$GIK#r({JpKx#9)GJu`hi(M-;Y$AIbas!=`L@{|*>|+#UM=5xDEk;3h
zBv2AeqV-|n-4Mf6`_HB_u(otA<Fi6Rw(iJanw?@)J_tF5mLXA0!qvkV1%;~=ynGR(
zAj^;_=HcpF7%M}#5+vwaRxQmqz21a3I|{$3XVh!kn?$iG>8ChA{Tzz{1*rc`O$`Io
zZ!ijSEE2`UNe&Ic3<Q~vM9~7&Qrb5rgxH=^^WS1AJ_{nNMvy9R<|wJd$~q#Z^_5i6
z+oTZjvmv9<+ej1>Z*wI^LEa_>FV|xf19Q>64cmwu4Z1%5B{q;-FjfX{BN*+0w^^6G
zbvnt*tk7<acx|~FGkpv*jeLtwO$~j^9T)}q7KviwTkgRq$XFywR^RdfL_Wp7<$kW0
z6cg_hB7PJy3Vn-2G4U;5!zjqNq~PUy7{$O`G~cqg=~(DfJiva6Q8IWH!D0`*%I2;w
zUPy8<t6Ew`E~f2zEQ2M(Fz6#E^Z!VEhE8TRMnO(SqL?_DO&A4PjYQF%3_bT25==;7
z*l)2Ew?X9WPKKeRPGH7~m{bW&3K72qG76oHL@{wP2QUh9GAVeO!6*jiqB)t7_@yY(
zxKl61#u8xM3~om--2=C?l}bbE%UoEsj$Yl;e20rcfluop62uDqYAlO4Vp$;fbAp;0
zx}SGr6y$y+ii!Jq5ThV_k|>({IlF`KxlEIGf~;<Bpg5>cLE17qsCLNI-s)kYMiT-5
zEo1^ZABkd8#rpw9LB^7TmnSg_G8TbiP{sQ-#4uHQuc~;0Y~4x0d3B1Rcg6_JVvsZ?
zib+`7j8RZnO2Nw{Mls+;3rp~!(+-T4AruKlbuFuwW?aR4iT;+bR^LXE2)q}pdz@OX
zb{~Z){bC%YUWUblqSQWWY8a)87zH^ViDKeJYZwKYmqgK`)N?W^CQ}T1J)~SUvQ$;V
z9>lQM2ql|{{~eGC7{errNeue{MnR^Of|rkD6l6*QB`JonQLQ8{pNANkmEOl;QYR~k
zDo9?6{cT7%T$bEXI7{_e=i?9}vurp@Z$y*4V*6hp-3FpDT5~+GEOSoKQ336~A`TK~
zLcGk{Xou{LYLXX2jzT7SrJCnrl<tnB<?R5;Teo*%lpZ<3eHf)XFG)_2zk?z+t-}z*
z)U<k)$cxdTr+6@tPm!Ix3UUg=qeL-D<nO{LD3MRW%iAytvJ8o0p2)u+V`WI>1qr&A
zRZBBY<aa1<=G9gnb$CNw_uPHB*<+NrD_7!C^BF7?l(qjgH8srIzk*SamyswYY2#xU
z1=*5B(X#eSQz)Wiu706OCrT>!B;-N0tyKBD85yaA<`cpMOqBCG$QX2562&B^U$GqK
zGsw_V@UkAGAVU)<2H|WBVwf62ufmxiTX%FYoTV7HS3yppWk?j0aCQSmLE$U~FL{iD
zEJLE0hqF11l_8u75_B!AmS!BzcJX&Y%GbolFGcV6cKGg5Z6xkE3Tx`6IIi7^g@fYS
z>!_(=Tx(zy<ZdL2iR*kHMnOg;QM9<WJ+(qQMz^7;D<y(`98#;AQ>q9SL#2*VA5Btd
zazWxtka#$?OM&w39-|~rQy=d!N^+gzR|`fl83Kz|z_^ZvfY3`q#KmuAi(Sb|h@IJ>
zF??@~JlQbLhYXn*xrxfQVU+GFGpb+g#3;xE%K>Y`%N~r<otFgjYu!$86T~nzvtA{9
zLO6OV0296x@hpIx!h}zvm?V6!!zd`>OTo*VF$%H_iDI7cy%%F;NcaQ^x|UT-Gfw#S
z6yoE2GMwVmJ@FfE0z`SAawYcF4`UIbwC`io)G+P)B1S<@N1~WGoo`_jWLgqMOZ%p>
z>!oA<*T?!)a=)KIl2yw}mHP>p)FJVICTTP|@B1|*9EL=JlHlOkI<fc+5c*Q0(~mly
zWTi-Pa3xB1GD|TfJ-rckat5*<a%doKqla5#7zGWWMAJ!rg7OC#1vvwWV&V+0#wgu+
zNifjX4CJK{!_@A2m4OK1=&=cAASqny9OM*cAQHtS1GyEWpbR7hFDEbxvJ8o0o`Jj_
zV`a!d1PQv9RZBC@KyEsD=l_ZDr*DQFiQ%GzM7<PeBp<~hLmA13sHtH_@(4yjPDrAd
zIH9j%6l7`=MaxKDxWHo?JEkT5ykjMz{tyzb+FPo;MB=25l;2HKY;t1qD@ZwvlmaEe
z@1G@(01+o(UKbsVsfcUQs*{p{binP6_>!wrXF^U4#AlRfoR3k^xag>)&8m_Tk116Y
zwha=H*}e^E=#BD{v&3DHD+A@}VFe}+DZJnTOcv6%L@`MoW-tm$9#ZfUU=#yhwB%vA
zdTeeK7xjyAWPBaQ&X7z9#&xZ$#&4WVSX~bMjUIa(lxGZ^ob%hg#<3BiM&j&pS1jLm
zWBH=Y<L%VcF!Oj2qoB+~qL^eJ4`UQ$rxHcWJYKK>1$N9j;wMzG4!#WOR&6s?;t@lo
z_P<|DQf+eX@dHRa^uH-ko?IwOLWZ+k1Wz$6l{ooph?Q9#I$Up5lAO=1SdB6>6P1{Z
zw)GeVjW(h96e<~mSpTY2lANXNfDD+ZL|@pHtoYpvMln(RD=<n=6+a8Heh(CXD`db#
z@g^!cfl<1v#ON{V+c64?;?mWd@bUmg>CQ`nUui93ehgxmnpLlgm_j&uMhfR4DRQE(
zK~7;2Q=*v6L%xSm&^#msFF(a7$TB2~HV=XOIsX}BWteve5_B!AmS#Nf*f?h$&zJ4W
zY~W1g=+$7|=L{opi)ezNUWzl()oWnpfMx^Bsj1;?U=v0`eng^}_>pmpf~-ZNXtRNl
zSkfJn`<9on?SvGlx=)qYXL}jjl+cff<n}?jq0f*gda{!INbE31K~|80mkLHfRv=Ie
zqRDL#!&J|`iY9_=-Os>il44A~4RQ)CL!y{OllNm36irg_@(@NrmLXBhqseD6R)%OI
zNYJ&cTAFb*xsb(%YR#Lg2I+#znzn{fY)SYf4l$2mQJ@g>chuA{#5{pfkaLkJCXVtS
zF$yvviL%fTGq@H7Y}Fu6L5NvO#E^bWB)1;Y4IPO@F$pna7zJ5D3SRyIqaZ5~C<Y<s
zYKUQ~=U#;vLALIEV2DXErt*+eXc-d4B*e^N6cl1o@Nz3gL6#v=%tOoxjFln82oiKH
ztCp57!~}G@ik&6IXbrUxqyG|znD=2(pb+z3YHAo_K8jJ0bCD<}j`9(Vf=o!FEHuP?
z15%(G#3=|dUlsZ>k=)~uZs<rPib;t16-Gf;kb;**XP_`|z>7hMSqU*r_1vovBZQ+n
z9~fd%jHyw`DU_x}F$pox#V9Dmq~K*2MltXY<{@Su#>x<41f#l^RZBAtG3Qn7nSc(~
zsmzpSr}+EKx>v=PDQ%*-qFsw4$`Q;tiYT+x)G(sVV-(~|B#Mb&{1c3VEJmVe5oH;B
z|C1DU$HJI+NmMM)cS0Ie3rH1HGQNxItwKd6!uv2J9J&sPViHt7g;9_pq~PVt7zG)E
zKrsj^-+>sWitbfV5oGJ`1_qTB!|ER(r_eGaib+uU9Y#SxB?T`l&di){NEGv+vL0h)
z2r7aEUCXMa83&b%OMc$26ucts$SK<V->KZR=asw1mbL9oqY#tyQ|uX^jm3c?%{FRk
z7-_D=D9FJ`6ccB8Jw`zeMxyAEMiG7INW*SY7V}I)8dS4L7il<5>P0-a2o;$K?<gc3
z`V)y_5@}wKQIH{|;N>kC1sQ@sF^DwxK@3wx_bSo|vUMK>BTb57^*4}HXc-d4B+`5d
zqo7EWf|qY&6l56^#XQnHj<GUC8bN}tW!2J*Bh7|l$+xD<cD`8is$M;iVP;2ef30o{
zU#IUZTGi4Z3NCHe<H+(`%sz@Nzow>!k!9IgFm*r<MWUEE$8#|XG8>7aMV3D9&G`}7
z?FW>IA7eRQ3b9r7pNfdGg{Vn&L`e!MUky2hjz*%GIGUGY6y#`9@M2*U1B=leO><y6
z>%u<%CAJU`V`XqOg3%s0n)7FE+Fx!3<Z0?PI<h;@Uf$mHHrH5XKhfQsnba)uImti9
z;<yWo138}8Q&U67^EQlv9FIgXaXjzGD9DT?N_NNd35b2F9nV8tH7O2YOCjZNKu)3K
zktimP=Z6>tIi3`}{2ZegSd8X)`r|Q5o=1L*$J%c&S_ao6SnYx9xv4;FOzPx$=Io$W
zD)>`5Wq(c5OEBs|sq9Z&O%7_0=T)z(2h^Hg4#?__hq}+eSGB-J-m2|-EW9(;!Q2Ho
zs#Vm~&{1v1D9BMs6ca}^iBXW*NfgadUAb^wi8MB0CKvBhidAt1BtNsSi=k3a-FGBu
zIC)FdjgWY_=062WVWB7qsu)Xr8%@$cR7xzbgoI?)RSH^qqs-);J9j|t(knB0=gvI~
zMlo>;_hS^~6og?Wm`v-Voxg+_nXMt?qn*U@l&t74K*CKFZK8s2VwCPGFxsi}BaDJ#
zjFgxOFTcPj-FZner<Ma7o`M*r=G3dQo)C_n48gKqiimyI*)aP+(v&DBN$&+11tq;H
zc-f9o40zF!Ubs`|g%~SC$}1SvwX9m2aak`nXZc6u;_B8XM{yoMs+abbDUG6}Rl62v
zNe3|NC}G`CO$`&)8H|FQi9|7RCILo4CL>X_gmqb#hedZn8dMWF6{Wq`2^E<L?_H2^
zIANA3CJ&4L9HSsZNWse|FbXmRfnpF<{xifdRdlbSiXdC}H!!NC7*;=ooI=ZxC?-+m
z=NJV=l@z@E2BRR$kSJPI=~7A=ItRw%3~@yeq3c<-HRHIlj&^_Mt9HdI2lG?8nX+dE
zxgEJ}VusxkO_W=k(L}u#N0ZH%VH8a^Qd7fdGKo=;6Okw;4si!YL8c;6v}n@LWh~^^
z{)9;W8cXszh_0$URUDBxsb{;pxpq>NYEuZh1X+dtMWUGamtz<O`Ii*D+<{RHj7IY>
zi^YJXGA(_IjpH7SlEJSC7JJ}V&X3<M#hwjS>T~&HUhRloqhc-F{=6f%J;}$M+p$^X
zaljq1U_OKegZ$5*Q&U6#^EVg;`5%d5;(xw`QIIJ~6wUweE#DoUPq%D3|1Lz%?0d3p
zI{!A;OA61ELd5?UG73G9L^1I^OU{K^0P;L3cv+2649rFMJYtwK&$C_eJex2|2G1i{
z%;b4Y8inW1)yqLCU#6F$=c`uDqDM!x7PHLWCx4>uT<X4a32A4&dw~73!CC5TbjF=c
zyEl+4SV@=0Ufa06G3xX?1AOS5$G&W);@<`I|3&ow7W#keP=m){v6rk7j_2v{sCwD<
zC(Cmb^p8zeOI4fUPx31GWU=ZUy(}-^)7}XCYxYd3*l^!SaNz)bDSCy5^GQbAidNe8
zI-B;^r(?d|m(m>;qh{1Sk3PCN`eUn<xfilN9(6Bb{|vYnvzr)_yG2}QpYLqAW6^a6
z`{TuQuapCc=0<iWiH1aSDSgyMGU_y<3*mWJTE0zweyTb@<xRU&hl5&qV%PQ8?H8}4
zKP((Snbe@XQ8BC=S_~_mfraP5{+#<FeD@&hV~{kmE?lWYD?o=@fOO9?>U=O<Qk`#v
z1J(Jh#6v?9CJD#6mn8BE>Z5hVj4E_Rjaqg#OYOozQgJX}wHx752EbD>`*^inDwN11
z`On9!TBUKqdBl0t`7GC#^SL&T7%lIVTG~W3F#)Qk(Q2aX{(F@B_ewd`r=5vNg59iq
z(5}#PSKDted;2a~Nw}}>(>jo<muDx>)8?6S*`inC<wk|&PehhKA)fzemd`BX=w;G2
za#lb+Pt&vAu|{}J+SU_^R5YA{y+OlbjBh0YMGQL|#VB%rL=5PhkGhX5*TS<pbW0{S
z>OQU9>Ar;tQ&?VyMsbePkb;lsXfZ%{kPXEl`lFeiKCyd>z=@?dH@o6aYcq||L&^x1
ztBM$5qeffmA87U`M5L1j<Kg|(v@Lzw5hZVncPJz9V;BXcPZGr>efk_mLGe(akQZrP
zmBoG#L88?Ww~jVlT{n|blGSx;V@!CyI_5fz8pV0h(1~JDY-+8nyKf+=7{?$He0Ui8
z3AK}9C|s@**61|y{zL{U){@bt1ggy>-#1G2BnB#{Q|x9PT_O||br22BIMDJCD<Z>i
z(^cug!QDZfxKA^tcCt^uWuIQy{PZIFbi%!Y|Gkp`y^8<c#s6N-|6arXGJ1Z<#y87G
zq}!4%YayQckCckd0{!#lsr|Epy<o4rOzW)O7oYholAZ9y+O4q%UdQg>xn7c+8%^VZ
z*EYN29{A!`4=g9}X6|=Ov$G_+p!>amni{&_=U^1%ekF>D`@IUIAonX!*vvz%d9bUh
z()6*ZW?zZEH$hRS*!P~Jy0{6|h?6~TOmUmXT|u(Q;&ErGT?db=4VB2xvin>7?8W4g
zR!LQ4_Onq33-Ggl%vj-m_SNhY_p|)dGwo-0v23{P`Tpow1~*HyQ^zY(4*dMoJ~qN#
zsDnN5G73IXXQNe*CMNUwg5&x0z@Y5SR%wDK-g<ovduR~7kzU?cvt7Fo*hO$BO@Hhf
z?TNIzcN*WCs^6KKZR-ot$LTS8ZMIaI%CTceJbLOOU#PF}aENM?Y~{#!P8plZqmi-P
z)}1-^BpIdcc#=rkq(n2*wr~{=MWT*!<@iDzvECs^Tr^<x#5`5RYBamz5$jSK#P6cC
zBbo>u7$NPSR(i|xu=^AVg%_KNM=Dz@!k&@PG}3$GZ?Oq{q^Y;W<N%en|ALyfB?nUx
z?&2wBZuV)6f|3J?Vv-#EH;jUk1Bs$Zbxqqy6TG!QW=ps;mhkr>RceNz6?L8<+lY%g
zywIkV0~0V!AHVU*5(Qu?^sIOP2iKxL^5G+XAu7{6>0V-kP(Hv>T+QI4AkftbCz<?N
ziIE{Pz7D^}N?EjCPy15v|FmF5UzU<=_A4<8GJA<)V)h#_3Nm|%ax%<*0#fyioBbt3
zUTdutn!ONJ56pfnPCvS&KS_4KF1dAN_?>=^Eq@=>3lza#F0+<pzR8w<7^5J|mnbHd
zU%@EI@+Hd2u>99Rs-AJnzm01#g~Jn~>Vf578ryCrpMOrL_L22>z9F{%_e1SK`@fG%
z&dB~B!YIi8C5nmte-@)4`<Eyu!~Xv}r0V3^|65<2U&Q}A5f0O%Lh-b;;XE1+(YHd`
z@_q3%G|GIw&efS>APAZDU?6OW4Y_?%ndJ3X>J1`$Z@nB_`9DD&Kr8<h7mJaV_icb#
z6tZ%OVq)cMF$%JBiK1C~s#hvp2(dHUvY;VTzMdnBQY9`aB)t=I3oFGE#bn{i9*lw(
zuB71QCX8ZWHQK_JE=zJ9jFw>mi(s_}&SpzgqiUPZcHgQzAlgR$Ci)utnb%^zQTgC@
zYHC<MxErG&KO<30{LFhW3bGi9qWPI~I>}o&_;@9XTpqD5OqkNglKx9bU1m>|fSJiH
zeNbrAM9yD;oIo!lQS`ZY@*ytY#3;ySQt<L4jDl=Npd^?<>yco;fEcElU*MJyt!5-h
z*L@SrDN>BW!HqDdKoXTGCNbzNjDliN3SKV2C<eS}F$iup+>Ws_L?FSau4UEQ^arA#
z(06h+g}ekaj6%;f)YLHa9Ka~ZjYt#|mpFq_kfBHvE%c<^6mlG*t13@bp-#CeMDs8K
z*HDVNSqfpl1+oj{k3=!?F!x~;<Y7|q@)3+;U^tqG>98^6(-<X#ZxJkJ@+~Hf+~?CO
zDRmuo`{PMxQMkNl`-ObX4(hdv-?%4UalEq^&tnrJdq3}2c`o(?DA?>HPvB%q<Q^lE
z+I(a3+hqO$QV-utAy5)Tl#q|cIq>fwU}kY8VD?UQ$!}FTV-w7BP*{-)G6^f2F$(ey
z*%UMh0e_zqbQR>pz_yK6M&E!@das~71pL>jAhEuME!;b#xK1B(VxS<CIKoz`eTO-n
z^=^9vImJ79!n3rMe$Dkg$;<p=a@&^q-Axa$vn=zwha}!%nP0rPNUre<&rT$Ebga>`
z@b6L*?Vl1sYz-!b9lR6hhO5Mj`f`C|=LjIj7R2RdXlIC(o3?_2lk;YBVqrkesuXk0
zs(Fqc2H0h~eKto2N={qTrE)2t9VZh}W$PsL9EYv<K~q%+9eXqp?WUO^$=oYj(r%jJ
za8;3ZbXACMk5<gFjb?GiB1%Z3bkmtFH|0x#J*V7x;Bi`zaYb#`pBhP6G&Ggl(<&k+
zQYLOGRv}0xa&}QXz$Q8SRj)#uv?g*db*w6_k{jjf$or*N+2dmoQtLy%bXZaAu4qlI
zx2y#02-j>N%_Jpv&E}L!>dO>UBSsx`<#E~$tf}07d#=jfnOm=sexgl?d_1TFgjNyG
zS4#z}T%OPQ9$N}ZTc~N*G}j=#-g4GvI^H(1omLytKhK%Sz1+@~W-A``%KB;YC&gp=
zIR?+xEpD3_Y5tlQkFs7o!(R<Po;#*>!rnD>v{a-eobCP5NG;8YooYlP?T(`1-0W6@
z1`fSRf*#=!U4G7fW!)+(BYB=tPW&li8<ADk;joW3W~w6~?C0hmjePkn9<85AGg{qG
zvAe~l#%(lUAEvK-u#UvzP#Jx))Z3hC9QXl^#?#|ax|mj-pB{(W5NZ|A@5MNj!|b%7
zvvebhrLCGgZRr073_Xq@-$)Zdn)%*#3LjtN;X^Nbw&aKISK^4nnuv(gc{0QD?Z1kf
zupspiM{z$v|1{U<Z>074`$Q#dYUGMsP_LGGS`aw2>me7|72jj;S{Y>-x9yM1%#Si2
z+EiYo?@{IwT^XAg*-yEHxLvl-(J6?5RiXf{rYAgKu`Np8?c)}u0b~jyHQ_0w=teG-
zlxQ-ym$_DSQPLQJ$kc_C*G!3IzAEMhqJ5k*Fedye?K3GIwafDpBl>?LpKw2|2rV3n
zBo-wT@rBqLe#wO2aa&-kc9neMd8DeJ2~~YnL2b(+`bl-CHAWhXV$ErRNE24dCBI8!
zg-xIId@{kOxd|HkB-TV&t+7vHhR%GFr~%++A7%H&`L;Ge9;6rR&(XUil!QB6zj-s8
zC=_gRTgBY;e2!inRh_4Iq2xAijz%rKM+{`5aPxVK7Wvbi<)A+fJ&*3DMiG6I<UL(p
zr4j8Ik18V3#w=YXqw-xMVpm_%cJb#Cy4c3}eD;ZNW8|Nn>1~Yr*h%Z6Ho!tdbmGFo
zHZ}fX>P?NwR4jChWBA9({f_dY$+##{i8Fu?`ax|xcVmTWIS2PpEhpOPx5ul@ioK7A
z=s*enG9P&m&K@e{)M$g@7JhNdcy0?jAZH8PUMMe($q=Q!LL^+5Oj2a}oqmqP*FI**
z9m7|uJ%W3iUGY7F*HOJs-6I&zoEkYvF==~|`Z+ZJMSn}B@&i?%<+oTmPLrk7+vkeZ
zv~?k;nhogwL@uv5%p2+ZsK)qsQtgUa6fN!(3(ypklAP;n1$gVo9{R<%n8DX%XYe*b
ztG-w(dArD)FbaAcCQ(cthy6#4f}XZW6issHv-4iWK*j=3c+A;)Ho-?AX{uSIT7r@8
zYyxcs#$P09FZns}=OOX%QH(%I5YSR`kV^XU4G580J{%>}v))AUBtPqY9MWze9-~Kz
zzrrZUxH~GLkJ+`Ng+)0O3o>g#z$`=!CZqpMNIYEjlc?<EWvAz36y(VyiU}{<FiIw6
zuWk(&isE-dyv&Mkhg^se=z%8rsqsF@h{0IYGjX`+Rm2RvfhsEF06&o~4@`>kBXJi|
zs8=t=@xF=$gjQ9SscGBFq2ze~#~1~T5{Y6mN<xf+Mu|kx;{7?DQg9*ggdjF3TPP#x
zy^to=JW>VvCQ_yW^sXe$CCB=QApLLvNtAqlTy_ix<=iDQ%I5djB0q}}k5G4(DmPZ}
zyoD~(>^qUK<RJZZjDjpuqL^6Z_c02x$Rz0vh&f;+yy$BzyPsja)Ur!<s{evfy30=Y
zk<h8882-yQqih8Re2HSR&i7du1+DW<!OJK{L53qyv~|ACVJ6>b*Zdxv&T}zhhBd!}
z5M9^mzy<Npq*2?N--L3*DSkKG1==F04wabX^;<FVZR@vQOqqU`^;?HXqaD_7bt(fz
ztGO;RV5EI**LCEAcN4K4*LIn_JTvMbYbmtOxaMzkJIX%sQYioQOqW8hjqQO7x$NB8
zg;hbXTtF3660qO`=zKd*r>3TtSo>VlMoknc6Ml(3-`&vjrK)(oq1hF$c<%c*Q1N`A
z$x`Zy=l#?)b;a{NywNGFczz~3gMTe()hD9KGoY_v6qGYb6q9`OF^q!J28p80fO=W+
zd=iqT<^`!HLJO#PK9Quo<XO<~An|Y(Bv2A2i!Bw;73ZVuHM4xOS3DENlRTN)3~4tI
zk5R>Q5~CpF?x=*;isuf9ky#51sCY^>n2i1#An|Yll&I|Fif109AWxQpmpP1*N!jh?
z$2bYN72;)9d^=>Of=!}&k|*bPLq-h7qCWS6lXDR>^aiS^5@&cx0sd50JU@&Dgesm7
zQd7f<=ch0V8YL3NWW;|Nqo7eDQM7p9lZxl}AWf=yqzd#4t9X7VNps1u{{KMw;Q*2-
zX)B(;#fYeCm&jLgkY08HN-#|LGO@^WF$%KCB<V#ZPFc5#zQ%rg3&u+=yJV;O9E^fS
zv*1eik<h8882&HDWFcWo6qAbQ%P|V7c&6aR#wZ57q^)@R7%@Y|Q?RS+T6K6J9-1_2
zt9T}q8%}Y>b5n;xVN_sBE_TLrw-r0@q?uNhV&}U^nH`Fq3B{kN&dD~FHV+S*Xe%GJ
zmpJ!PiSsTw8Kg`AZ(n_R;`UXx7~P`<@wJ>3QggNXEG<N@#2Z^nm1;c@+g<t7+#K74
zY5PI6?m0YZ{m1V0K{!TK+g7)--bhmVfNV)p%27=?+IY~z1-cItLh1#&5m|D9?yTVg
z-ALzItI}U*=E<wlf6G4csx<%9i>h>I-@%@Z%0sE5;};<Z{o=K$`jK(`K&tOI8Tdhy
z0sbi~1CK1GbVMFV6*Ca7qWUo%dbyQt9igS&euFy$W@yn*e8Y$QX$ysGIyaNq-iWi_
zSuPf?^4jY;s%A%>%fkyqy?kqQnBZ1^nBZ30SAX=ft$cvlHCkIOkK~MQhi4BGaV-=q
zRVY}3YM<eN@6~GsyWxz7XA*kFry~#9PH$SgK1=&je0E@H*dKU-Rc`EF6tdMD{2seV
zU+awH#o@|Ay+&uC1bNxl-f($zg&h>y*c%S{b`a3^U%wHqp+)fG{KT0W7jBV`r+k@J
zuG2w`(a!R4z^e!3bsBq}ReQtXS-Zlv8+$eCX_$^?B<DjrcIeBpn&$;XDs50d-Ut^@
z((#jm1a?5`5xQYSey!Ka^le4p6=)TB;RyTHZ#ZuX*UnG}6)&)r%Xft<eA<n}&U!vh
zJLKjXcZDlyRbZ(~2SW07{yufJUR<%uHe31c=Z}>FC(o^x*dQd?fW7^fUsztVt6nY0
zmu=eQ)7ZTv9C7HZ`sizaxW+zSBKZZhsoIaP_80Bx`YiRglAWQ+PN1fKZEKFM43i=X
zHe-g-$9;63j-m02bQc}qS`TKnK5r{+TW*9K^dIwft>)G84*CBwSsEDtdk7SeT??5j
zz4Y^58sWy4Us}3wpLHRf8L?9I8`BWA-k|@`(l}hrcp$6uX~$Q!OguHh6=m8TUbklH
zOdWcELL*#aR~q3NHM`)|ius~lwJSw}VJCO)4F_n1&Jdd<ky0@-1!o{!Y1IQSA72WW
z(5bZHiXv^p$Ty9V1j{ebQQmfqj?rNquHsj?nKSLjm#V&9FM9dobIg=h(;269hEQH@
z$*rUi<d2s8k}%d)`C`eprpt^GN#$TPNNB(<CuV#a7JS4}rz_-b`I!=(w@b&8hU?fT
z0`ct{AC-26^;RU6FRNL{NLAwN?#1C6Hqyv)DzjvW<<eX!Xk5E`OSrOXS7~pYJfny)
zvz%R-rlW21Y*2*fI6**%AWcrvWq+bNPX{(tCcN4#jis~Lotyx=Gf(exAoT>Kjv>0f
z^`dRhCV8%`&9FCFsok?HarYNT-Ln%_naR@G=@TV8nBf@Y3MQRku1sLVfi+E|s2Dhn
z-9zC<vdOu8b8t2h@_v<B$cb>ZKTmu5j^~}ytW&1{$$tCFw27e+F8Ar>u|fWobsE(^
z3F-{~U3{f;oalW7LbMwT4!yd7PIsc?li8poDRU80>m&6lL98U|f>J>87jEDLS&vsJ
zRc6Qrq+eLe9ED$>Zn>>A$847l-4TPak2@nWnYn3t4}l-d69tZKv*U5kD$rd=^7fon
zD(9_Yk$GHB^%~YQ^TWPfo)O;cM7Y$h9Bt5v$>EUjCNzA>5=)h0iN;2d7biAzyIDg-
zk!1zsXsHi+Cjj%~;acr#g$|GAm#Mdz<`;HkbE(L$53-Bm^Qikv7H*$pf0k3kbcflW
zPq9CrWPkn_`{Bp*rydToR_sJz9y)lt6THm{e*X!6FTEpeGX)mjA%)A!=?x+S^1$gR
z2DG12eR(qmN8v8)FZmRx8+Wk;<1l$4nNIaZL71{WMyaz?khzM{%E)wC*mu9r{``df
zdFc}R!w${n$A9vZXkB(Joy(4`bJ?3~_=|X4wg$pwYb0DYL2%gwq8_eL?UYMy1s#4Y
z!YD~oTw(4gx=*M`y~4&2ze;vJH|>?1rq3^)*>bUY_psLEaD>Vmt=0Oc$WjX$<R{K*
z?Mwbbp`Dy%!K*V@NO6=Mek@FoV)o!w^7=hwMt3y!(ZHwoevk_x3-Za<1z^)>YL}Na
fFLQHUR;m=r^&%S{v&_TM3oFR2OAjbkY)t%r`Mx{E

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/utils/index.doctree b/.doctrees/autoapi/lmflow/pipeline/utils/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..3e53dab839bc222b2fca14dbc7b8fed52caf9e5f
GIT binary patch
literal 3819
zcma)9TW=(_63%WilbJ~}Nfs6zX*XPifFQ^WAS4j1RzgQQFAIZ!6>rG$bbH1o?sj9l
zlgR@TJh0kG781he@9<mrDSU2s_gvW9O3}8<<#JW|tFLN)>HPoi;a2s}9h!oRd@-VF
znyZu<?|P|9%8cdK|LNcT)&J~IU0>5nqe`7H@3x?ka3NF8a{q&O!=P@3G^6_bu3?E4
zDsPYT4)5~aAN(oryN(rBvXw1&TV^w<E=PqZm=rl1RmOZc0R5eZY!MuJn%ZwZde|)}
z{IQ-NyFO_H;M=ZS`#o_x!MdN6Q|boPS}mrfMZ6)tZIPM~%Wfj6F%aUrlhZ7I_0qCf
zE~Yd&*K8JaO1Eq2JZyP>$nWrdewT0a6ZhSitBj3xndeN8FO@zYfBDt_5M=Wo95mzT
z*qD}9(L#)48pcf;##I`|B2U?3qzxi_^})?@%C}snlC7jLW|zN)NcbDb+MD=$3xDt6
z@7)j;5~D>G%c<KbsL<Y>WGXEs`~3AMXW!k8`)5zy=G}?)wB+ycvt`&5{wk8o_abJN
zkABn$+~B)%8tt0+zOSr_>bHc&FJdC2Mf%jpf&(G%_oF(0A35s5g+36fvO<-4`tLMt
z;n7rLHCsjDb5h9fxHn(moz_!5ai2W<Oj@Xmk4Q=_EwoCQF-phx_K*C;efT1Wt)>9=
z>aowCeJo>{(pSXL86)w2D#bisx%d8y8@P}%BO;$MlwQJCY<%!+8wHzL@^9Clb4t@E
zp8ELNcj@e@86G|N<ff0z?Tb8-Wy)qk#)9Hgp7I&z{6{W%&I{&x8C|5Tu-rem0Y;%k
zMTX#?xl~hs?Ydl~DY*069n{8JK`$?}DYBXR2U~7XQKeD2yn^lk6uhkCfc%I2q(b}v
z74QvMZLLh)VGZ)LOHGS{X-`)<A)Nnuy>d{ytTo{VW?8DVVU!JN%`d@L@S$eErV5wF
zPWiotzRwzZJBuvOTGLNy1GHhW=9cZ1S_dDlN4*+i%l!O(_c9(bH<E^zASkxEMzvMU
z-?7S48Pi?sRM{`XeK*`NDRt;8r0pX6jU;vx*E7sok<ZaMkK0Z-(<-+N?BlxMQ&|Ej
zZ7c1n(iWxl6Mi^x{W(MRQmZryz7Hl~#?q|7Voxh&;VOW(@b1RgyS?ZF!M*2D7}U#B
zORNs?+9Y(QunSi@0V-GC-I~Ent}I*GeC!TFhX&NhVy1NFAG@JJgD40{GxU7}gLk5Y
zNmfb7rLded-4XZI2kLe?ar>GTN?Rh?1w%S-x&f!MU3T0tTL@&)Qd^ovspC2+o0fB2
zH@u`eM;4)>8O<Q+<H-^hLJIzg962Lk8I?lIyMAeBXP=%Kf$zKPn}lhtG~qN)C14L2
zAhyJMcNjCjVx?1AK{MRg*)YNXa>HnJ=|FSz?)t{YNp}<i0PqY)ek20&?m*IfUeY;3
zx>Kxgf(qPAnk7o71iS&*;F{2`C$5{Qe1;g24v{u!;9WNi%te+WcT)gs2WgoVq{Sn$
zS_uT8sB<cV!NW>YRYQ3CrKm8hOchyV0Zm72Q7FSe?aPxDL=sCbgb@+1ha?q-PGtzB
z_HwtW6V%*3f-$I!s_0<!pjIOdb-Gu=7cMbOSB1&)U@gUa?}6;_PH_nqJr2bN*yMAd
zP>M`g|JBhuZdkB_phRN-3Bl|K$&|wPP#5k$+}gr?HXcK4M#YuoDjz95M>*XNnw104
zB$j1(9*gJbL4Nku2OlG!Lp@72%m>NR{M(KB(8B{>&zT+xW@ptMV1qHYnZ#v%OQ$HI
zH1vd?I{_dw(pF|005L^C$hA8%SGlDN!o{4UJ9B_{N2-APeUnjC*-s@(+8|LcRqMDU
zQYAEYC3li=3V-l?vDlF1Dup<GUKY3_Ozl;FsW|}62om&iV4x<2TY)9e1~$}07O*Q7
zjg=Ut>T?gspmC)r-z#{)8MC<Kb^@()!ZIo(!S9Pex|P?*^)CcqRn8)zUAt|TUxex0
z^&*v^c7a69x&pO<ae?YN12tiuJ%&?2mPHEzoud~ADtEWmq;Xz{7}jR(q8s=Wqdw%j
zK^VVJ`5*jue$5}1t{<fKs(_nen*QUOUtgE}jqv?D8r(gMmfh${%jnIZ+xAX=uWgu3
z^TNz-F!=h%=qf+%+}PTxTlY@WHZin5;`d8;u!@Sx&x1q&TOm;R9wnjp8@-s2RY}xC
znt-Bgs$>haO7-OpVXNI=v)FHLtq$E<vkwy2Hq17+H*M8B7+pY33C3^W4~;8PQsM~_
z13J_0e{1=l$nbCeTa+1ox}jk4W6J<p@xqp&*2H_QxB|ysJw5@9^MH#%8c4L}n=mGn
G)6qZLn&RC6

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.doctree b/.doctrees/autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..142c90661a3df71d2171021d1fe74c7bdb8847f2
GIT binary patch
literal 12414
zcmd5?U2Gjk6}Dsh`Zso*CgCS_afM26M6Yj~{!qCgY8BB^LfldTt=hDk^}TcN?0WC+
zvOlp+8&m|XYC0+{*-%<kwSdHr3gQWg7bu7aB!m!x7hVyMNJvFOh&R47v$H?fzBj&c
zNQso|-PxHlXU?4Soik_7vtyUSTZiO7F&FyWayORDX4CbWEKI~i(`&>X=0?fI<k**!
z3rST>2IhL`#X*B5VhBAN+;W-$bCV|%F-zSe%L$j#@#7(DM3(1f!}%Dm@bM>;DxVZ%
zkrg?t@JZC1PRsGumwd}-j^(l?MXbX*Ua)C|W{Vk3-#dTYFdeJyCPWByMnB6oh`~XJ
zKbNQ#ReTftM+u)7lSZ~+J|ZgVI#<Lfag(gXtEQMTqbRUe;|TI0d5l=ikk&rlaLh17
zBc8`xv!k9Snz5h1teTCrfVHSkEXE_VO`j4z$B*(Eemx)I3u3L#y$-7fvFoy+zU~EU
z^-rI<A41W8z+hNcjMdFJ@=V{VE3wtJ*y>Vj^&VoYTW*tWECnG}e8cH|@hTq@W3sKV
z6>69Ghp<9^Bc%Lc{J#nRZ^r*)v|0!(u;d!6V$?USAQ5+Vyk_jMdlm=VMRA83H|XY9
zR-(j=`A2yz7vBoM0h`39m9FN>c0f0Xsr&NnllpU{)RgK3DBh<wL{T(*N`e@m{X40;
z^IIU(2`SS9k+~W>W?(ct#|VAa7(l$EvcEWR_OYjLyYTe!3;J(D(>;_=C!V>6>au*>
zoy#Xk<-)krF@w#M&>u)!97$mVR=8C7oQV0tn*nK*js4~+{Byi^n+5dr2x~EzVS}9k
z^f>ngZHYVK)RZ>!v2F^fj!ue7)DFDZ-zHxInQ=&eK;chd$hy^xcyfA*<n2P<YuHjx
zE79eQqd?&CA_0QDRj;|ZBcmgMx4z?tF*3F7W3aYK0GHVgOrIA9icc>4o6Fv+y?jmy
zck#qP_PC@*oji9|3cd<-nDLq0wA{9_0q`nL^}tn{N^kMFV?``5ox&XajnqmjimxQM
z(C_rNP+*eeRqXRO6H)OuYY<4niV#K{^Couo+W_^R)-5VnF%UNq_v`^{6F5yG(=jbL
ztsjeVyhWQn9OwwKC8qJtHhjlwSW)tV{Jd@kZnA|P-dtAbIIwpU6BsLfQnDiFkIbOW
zqD0(rCF&r7=Qa4J_Zo0c4`a_R!5WY7O0Ei*imJeRZYS;eMZUm)1CS;<FVnDo`%eR2
zLdO>|;n%y&KMjbqlex^KNBN(3IZ=Q1x4_Nw;O1*(ZmPZ66Lw~$kL--GgCax^GC5?t
zu9*|oHJe*u$Xj}5{_hkw-q(ZM0+9Tts6AycsPeaWfg%6ux3w9D@ShiPz#bVM8rnS+
z@L%$G_^(#@uf-<@v)^9JC)NJHG3a-@4f@^Umh`s>7_@yDvP^CIAG@{bzwh0q?VB)f
zzrEM^fw}rGA~oy&oxTd&M<CL>lNhCFCqYn;&9%_G)y<qZT!_gQ9nWp6sLJeijZKU`
z*FCz}Nlb2ZoaN9r=^!}a2P|a4c_#hDQy8n`;j=PE%vmm-pG<wO=xR+*8MKNi&`Qrn
zS!~@Wyq~;3cIAmPRCsVi6k0NWVK+%%<bi9K@x5(kBP1L1P@ab~tdU3Bo(_ON&Yz&A
zAZ{;^xwfA4`0YD|?%dEz5Hl2mQ-D%ErbBL~Kc?}nU4E}#Q9!ICYWpkFbgqbo4yrYc
z=gN!?;u{~#{1dHxP)A5(;Pn>IR?i+wqvhQ7ow^F7yPTT+Bho_3dK?j>GMda)`t3|X
zcT(wH`7!$ka!<-0uw?T(9PV;uKf|~yab?E`b7gYHWJ%ZDZ|z_{6hPYFlLEtT?5iHj
z0}9z(gqIxtG~nq(3ay45*l%g~WKe3qMUT6SOYJw*{Cf;d?Y~Gez4x8CL>Qb-*MBwe
zWcKU3pa*A`+HY@q-Tuseb_c@t88q(XzwBpr=v`E2`%!eTAERDaAiP!qp6#d6-0iz2
zTbXysU012GkLMzn(Z_SmA@70g+2SeBjd0@6{MWYj+~g1KQ<}mTb~5?mN@;z_QrYv_
z2<gQ>0G5`zz;kygze<RF0P;EF-ZaTj1$wHJ>R{==V^?}0<;DD8VPQ%X$XbjdgS|cH
zM){0c`eKAgA)8Sglrqpd%RnG2Qy;P`CB!LHtCQe-Oe_rsT{;>WbXdOGC$p{FX|t<h
z8tbVc43mtC{lvFC{mb`K=A)$c_6eeU)V@0-FE!&RG-F6i%GGx*-Cm}y$_Z(4<!nBj
z(b<{J31U<6I6>4e$KFaW_QZ9CwBvb;tv8T2OvIdiG)%X_y`ZpXmrL7Gl13*<?lP1z
zhlAzTk@^P_tC>h46;6IIK>J+xQj~;p__OQiA%O+g3XoGGrbt@I5#bi4oV^m2G%dv6
z5o0RL!z*H<6+4cUsG5Juh`fe@EFr<kJF@M0B%@?&-i>&>i+Ed5TMuoxonEC6qV%8C
zf`-DS-OH26UQkQc<TOy_OD2C<RVVDxrWu*tt4xz-_KP4S#vuurH?sx3O8qImyIxJ*
zV{D0W*VB))y1Awjsas+u{ixsP(IZ`6n01<SJ>91J5ha>0WKjh8OSWi<x22-e=$c~j
zmEo#gqlJB@$C-U`4<LJ$vIhN8*FFyR%l@Qlf2`0ZdS+uJ;MyNA_8k-}Q%o9kSdLvk
z#q7nftsD?l?FaGQ{uF*L52tTuq@+zP#H;p0XwtH=e?s3ChV~Rm&e$(9YCD3e{W!kc
zPvEByV=0hsXABlTCoSA`T`w}bu4}~yQ%8sFakzvSrEoWs9(jhuAvqpcW%v{z#oLGs
zFKY*gr+i(1>bov6>?hrEo&84|a@77k{zw;<`Ep#bQ7WR~qK6l=Nf!%~{_Qf8u5X|8
zucb+|BVOuIj`WpLzS@8GowyjT74w7&9J4C}djnKT`TTwxB(E0e`parrjq-g097o}X
z07!;Ym6mH6pNpK8V~5`<xop=d+r$^G?AO%ql40%acXKYc`Q7hhkn+2BfJZyTPuY~#
z(E<-j!siuxlRD8%_7_z<JwKn=&*=6pm%0GS{*Zb~mui0rkGm93!HH@DgE~?BW#Xzf
zf}dCDX^Eb`kv(0+)0Umc-%IlMoAUQn`TH&T`x^cJkDT4TbgHi&+lz0zk~G3z9a%-O
ztsw9{_FE5NN3W9KI+c-<ervJIZ>3qM)N^&Gpgxxk-m{z9&IjDoLFs1b(!0ZOPwZGI
zVJYA}<zOUdy^gSD@M;jZNA2x1y?5aCC|KGVymBtLf!9e48W_B2CJkQnoPrmQ71A0h
zjbA(yEJ9HCqT8MTOcUq}U^v341b+D>pg`<p;#z{(EA%8mOg>!=5X+sOm{W(V>`;J+
zAT5B>pBDAm0H6Bge2`$t$yWIZU7}XgrCK#vs#Q~^S~aS0tH}ztnyPSP*}MZ*M{VP3
zL;-}fYSes{L))T<&#K<|E#goVm~QCEaL(`puZ>$CRrNL&Ayjmd(?eoP_hC)E#!wtS
zEM^;VKs9wnRzWwTTT~^Tm`D#%C^DSSss>wuJY+UevhjJ-iBVpr&0AEwIP&8tS>bai
zXJ|82$C(k1_jtj%k%f`j@$oVpcwPinaR}c?#Bd$O6^aD|^RA)8RQehR4&L_D`IxHm
z=qt|1FNuRK5I`Kv3e7Kw{Zz;3#vNm$<pmwoXv~HvfAdk=)uHOh5TuT(7UnS7Vys&c
zH*`%PW=aR7MS?40CSbl7M25r8GYE2@nBu0By^e`_wqZej2*~45cN!C8O|}}hG2JXx
zh)_)q=rXf|mPu?xgVAD$Kg)%*4AfJ3RJnsge;l=Hch_WA;rjANg9U*X7~FK54nrMZ
zg;+(xLCh(c7pRO$Dv4O=`qD+h{sYWY%qqG$4|??V<quth#Q|Cav@YZjkt6RH_dBNB
zj?FfM#3{?O5A~+vx_~vjplLK2)hA&Zsz+QA6_{oV$rDJ#YHC$La%7fnK8?Id>_eUg
ziL$uk8<{dfu(H8I5;ky6T&AwzbVOaGn$zy*O@7GYrf1+D6ul0hg2xM)Q5-3?q!AhC
zt<X}+I%hPkkSgJ@kd(`DZ6q+b87wA5Zb2QOpfzbXqh&d=*isy!CrpCdRywI+Z-`nG
zTRx4#-0b!Au$U)HgXZ9hmqYtIs4Y5k;AS!FGv9!HsA4->49dHFXghc(qY&@sQRIiq
z^*WlvC4V#Gp1b4)ZP?Ob>M0pOPa|R-Y)>Swf0<vq>DD_S&)J~W=ps)oIr)!h@|5l&
zm0e@OlEtEyoC8o$=TXPOWRtj=h84*=ya}-Yo$MHyab^ohk9}L>K)C5f=7zzoHb-?Z
zhyISz{dOW|LR2e8#us8(bqGOSC*Rep;dl)*741TUBeZB><CVQZ%2I?Fy@u+-Em__y
zYnD?O*n;+W4a;po2U1)(NI((Bt6kl!4(YDrnaUWcoOmsP%5+v)6BfJ)4md+`)|NQj
zK(94}!4EqI^3H^~lI!`jlp4C~v=o|ci4o?Wr?O-*q0j{83r$4k%Yto0sJWz)FIkd2
z52m155&R9brb$J}7ZP!>)JiEq*$ir}?1k?9sKy!H$Ei_0SMBeT3x1P+o}z1%C+KGu
z2L*eYeqN^!ui_^blQc@^&8T2nj@y!Bwq*Y;*%!x)8|f<%#D)I}ls8t=WUTf9+bgY3
zWPVvYVN^1QaNfxa6u=xB1ud1-9Mxq62xy5@ep0bX*ZPu8d4nJmurd{H(~;1oBco06
zm5dT&vA?K)QX>0NvZvf5q@-F1W^_L#`eF;&hHQntc2-dtT<UtLt1;_6S?V^aXO)x=
zmgvJ{1W^L}Kio|#{Q{Q_5NvocK^*)eN=6(g3)BL>c>Gk_ei3{1T=EDE8u2!u0h$p)
lht-4`7M7Z)vu3G=)0W$C;wGtmn~LJ6ngrJ=vZ+9F>A%XGe-Z!y

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.doctree b/.doctrees/autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..4d1a43cadca23abe3c54795deca677ad7ea5449c
GIT binary patch
literal 9704
zcmd5?-ESmE5%<~Nwb$Ri<W3QH4vx?z>m=*VML|i-=^z{&2q)%4A&_v#+<2#Vd&aXf
z!+e}=MJGTar!2GrQYH_85by#IfW!+@kPyempTHxE04X2@yo1Dt_*M7x%<S50pM6ol
zl4oanx~r?JtA15oHD9Q__Qtsp`6uS1fZN{As%crCZ?Py9la}8~y3C8ySJRa*rZ>{M
zsD<Ws<R@W^rD6m<THJQ6ka_9VRLoKL*mk4UZ2UyTTCweW`EXw0RX%Ywt@E0w#CGhm
z(kt<Z+iko4_G(}U%(XqXs)%)2*AI7%$ZRv?y6big+iNpS2^NSJNR2<sc8JGgP<Sa-
z>#7DONX98&6g4AXHXjq!Y_%I=oOnt%l1)=gnQ<K2n@J2QkyOTPD<TO@v|KZa(1`nr
zXLi-yq?H84!=~BV3R#=_Bw`{qJM=2$^ZX2-;b-|6UlLxE`(4%y6VGE|bK4KMnomD-
z8KTiYU@&Sb#+qgl`(|J_mGGKccugt1=5XOPZO>vmt6_vi-@87VZ1NFNk!__tQQO7e
ziAD1JAnSMG{~Z1w!2b#@7%~fOxy-s44@^5u#S>lMN?i7y<s)yWcw7x_9C;+KZp5jX
z@P~M#kmClw7rVu$mFgC{x1bxu)KkSiN?khL*OVy-C_bfjMo~0}%7Yl7jl8J3^M@hX
zNh#Td*xXE9Gc;PhYeWHS9YM(BvcFik^63}P-+1xS8~Qh;={{OaCqDo7>do5u$JQ>8
z`b9~%YlgcQp-2$8IGq6rjB>T~Iu(nhCj$cMH~*O_{JDOo!@{X4ZQ7N=Zka~bL^bY&
zeiH1FBVo$~q<lmnPh!ZnZN)rYpCY+>kn0w<&ezIvm?@M9yig`UNVn-*yL%V~pJSoF
zy{BD*BiYryLuvpjvlE&DFAWsyYr*cCzv--9QvzJRa8wgoRRb?xyCOwj2O7)-%(HB-
zW9$I0%Cmjst(#5n^Mq^1EHvHHH2hncNmdnWX<?4v9d3?*A<3>74R%ve4R#w4OUlZq
zMZ5EDZ0|P#<wH$bRIzB_Y$`6^8N#OUnMA2;+Fr_kC?@a}?*{OnW5k)5#xvUqT)Sn*
z>C5u<wi$Y9Pt>rd3J{Ou*~BD<OMjGZNETu<?65c$j~`PtB;0%*?&<#kX4X$&|6YT6
zp61m;EnX|D1sl4xwEwU2CH`xmGtv1P4Le8|XuvDz_!Ug})d4)n0we8R4w>`}|IKYq
zbTIopaC7r8N?$@JK-h3fVw`X?h4aS(ZS-yrk`$*(kiG2sUPs}vIcQH7Mqe8oUG5}m
zJ6(4z3QUR|CxehhEWFO7PJI?*wa2_7T}Z)lU#Mm3_+y|oi$-V_Q=pZFQMva$AONUc
zPCOZ+Mhb{di&9IjSkB24BO_(q8V(!810n!tJ|7|3Sb*kyY(R5{1y%e7{uv?z&t4+)
zwp-n!kM8H(xuI7fq`xLqtFLH(&GgI6SvvjC^<#Nc?T(%Ekm((IWKH9xe#ZEh`=9H>
z&0IX4J3m_cF&r%ogPUExQopj2`Py&j>BZ%@hPu+`?`X2lr$`;?uyk6COGi}b_Vc->
z?ibzG`40~Fq|PVh&J4GT+?n#%%ihpwVchLIME-Mri~nMS|5CjB$SzDSnyl)7;lTDW
zC2~B@$EDb?C!_V1BD#^yWwa^KP~)FIf6;>A$()4tP&zYbgKlrnpE>Jl{yTJO&P_>X
z`0f*zaG%+92d@T>&H3<c(1UNwY<W1nLHFjY?nn5a`}s2G!Tq|HPQFeB9h@rl!s_6`
zN>J^rpn1@b)!MnIDco3J;{jfb+{Xc4Z2lQ};$FVwdod0Uxi8z(UReFFh((!g*w5<A
zj+@p;9_>~>LVC1&z|yM5m&49ReQAPfgfj(>e&SMgR;dF<Rh)a|l)=g!gtL7)gVOS{
zw84lNM-puBIh{^MbM(fDY~*g@B<xFK>B9=4tvp1@u9Q=wG>^`s@rpQqWDL_G$}vV3
z8~?X_8~5Azx|qhQ8VJuMyJD8Omxl=X+;@!XGXdu>ME8vIr<}ahl6OE$Mnp}@U_b!Q
zpQ)?zNm^(HoBzq_?9b*Tv8i~RB<hE;_pf2>iMvYKit9GpZXx%Xig|r&m|lzfVQJ?M
zQ+7J;>9f|8hEnY4$muyG5JAVbQYojB1BE?0bE5;oRzfPE(0K#vl{kasBt(^pDH0eC
zoXE)bpN2&>J7e-&qN2`=yecN!iR((Cs`;mk*l!ugJrKIwl5IC}vX!kxH{$J0lJt_=
zeJI2qvW=9AFe4SArC{pzMM`8JQcKq4_|o8#$v4(jtTJwyu{pTP>;&#S2|{85l8}dR
zXN?}E3YDKdzoAm0Ju%_=`gTq?H&lA2CuXvj`gsvOvgJj2r+LrUZMq*(qWK~g$3Vlh
zM@#%&CMpfK*%FTjv(9fpO<bb5r@pd$XApbLaO+@zb$$;9`yEK#`D3XKiYb{_F|za5
z@~dM&U<NhE7+IjwRZ_yCUbF7}9XvY!z~ykDd_E^7P33H|>HNFALKUVDiP^XE!TX}4
zGl}<34VMG>$PjTa9{}u=(oQMU*kwq|xi*RfCD%pas^o0g5w_no>=L(UoV_zS^LW#*
z0z>mBeJQNB0#>`2dw*Ch7#+kI=UEI=u<9)1*0~Rtei@%b3jp$d`l&h(P$!znsjGIn
zznMA@;wRVd0n|JTSv^O+Wfbo`jN99kQ3)qCfdg?8Ag{6?{aEQ-#FWnW=<;5=e4j4w
z%kQ4TUC(*Hyk3&m56J5W<@K_>eu%CIU|Qj+#JqApIc%myc$~dj*64{>2L#E8NDH!c
zenD3Q8@eRckR`E(Dv33yp4gyzB9LC!6PZ3>>9I$sMegBkOd*9dZ<N=DpO5kyBY#-j
z9fzhDxh9oOjUe<pNR+7>V<q-u<j2=X#FXyCEIcL<Ge05bT1iM1Oe3#X&gd3Zw-l3^
zvq52ZF|X9NL+Okyq{^Q&-2|09ZReuuCvlL(=?0%i0=dIbi!@_ILU@3GWl?N)13XNJ
zz8`~C1S>nKIMGB#Td`nZ-Ys;P${v%@#nUXs5mW=Ew}>TP5hvRq;Q28tHNPxosdnro
zU1O*1hg~Gb=OR=Z1JuvDRFRHA>Wpe(E|V?BwjFarS70$yHXtj>ZitzX1%4PCF1yYk
z$WbxHO*en6h()$zLw+%0`bc-Gh>FEFlMbevqqIDgLqM0AU9{A&5iLfGA^vO+(lSt&
z_Ni!s;3$dPjVBs1`a0YH(qLiehXyx2%Vj8;R*6*vW@28+yhNoUsib0Q;LQLD=S^6g
zm{W8M9`xw5{VxUviv?N(v@YThkt<WQv##lN60^e~aSDh>QR5ePg{<X=mSHg}4PhE8
z{ceaV4780P6B4m4tqKSt=O}q>6is3j@hr|1NjETZWrSd5gN?Im=$XhzZ-{x>BGsIB
zzi0{~mRP=lR5E%kplsqt4BrD3Bc+zKV&l3U*-BaG4a<(G)`*2<Tux{sfx*pSF%e=g
zX>lMrn{_kVwkzw};xyf15+tmpRpMv_Y8JM98XrSotJLHA39(3)2F*d1oJ0G&DCj)1
z@PL>LSYS|<M}D+Gi$P^AfVP8oG751IkK-U(Yc|mwtp>X>_q|m=?7)^zQBTPLdKxk7
zVtZnlL1#hZ+#`=co^xTlH9($P3i9`B@)Q=6%5Jf6)n;*9&H*TB^0@0_vKo^6up(Im
zofJ#Z$*z$bXTAWGuSr9CVj<e~Vsppfc89z8hyIRJi69j-5$eIQ@wo(69YIic$!GOw
zxPHsbM7z}Dh%s8&K4ou^vJ@dkZ=tl?lb@8x??f^f=s|n@762PMkm15f0*WZv9O!0u
zNq1e}RK`djZ)~MdneHaDV8OfKfHRcCdg4?Iy|xU7Xwx-tP$$HdTrXyyuRvGbwnDR>
z7-Qab`Upf!Dl~!lLKAU9uwff9s@ha!lOKRAf+=WL46g$$x9F3P8>u+i*Gfkx*$is!
z{DDf3^hF7}Pf(+}uRE{9#+{$gWrhOH8eM)ym!Hz*$MoVyxFn)RqvZL5K2MS3dU8xp
z_V3BQ@Xz<rTOx>>$pzG-tJ%3#`+&VwtqZw#)~*;IjYM$a>C5Eq9~=WMb@&_Cl^8^}
z#3{e1*rdvcWK*VG<Uv(t4P<O3gTzG4mbFgGVHVX<<rE<s)k5H+`ze8!TgW10E3~w`
ziH|P&y52p|nD?&rb*t%FC8d*n^x-GMIEB@p8l;uIfujcWwfuzO30@JkHWw-aWq=2s
zxRkZOiXD0>{RGSy;WhvP`Vc{L)r2_$_BBuE%~JWMZLj4f7Abp&K0KPT2&6MSQ$ghF
Fe*r2g;X42T

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/utils/peft_trainer/index.doctree b/.doctrees/autoapi/lmflow/pipeline/utils/peft_trainer/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..a1d870b5c594252d3d4aad0161353752bb15e11a
GIT binary patch
literal 28064
zcmeHQdyE~|S@+uBr|n&@6K@`l?dzD}YiA#cMMT&*PN->G5$uLMN+E5=v-i&K%y{md
zo0++`mnyB>lGF>Q&?q-T6@RouZJ@Mi5|n_N2ug*3KWL@45)ueV+J=9M#G``#gDQpJ
zcOElm?&Iv-yMClnE&1*|&N<)XeBXDz^P8D-E(|~WY;6<&$EM<j>($TC+tq44tU7Uv
zja9=+v*y&3)-$ccA8S3)Dzov(J{O10sN%HPCX}eSUQmskdh0_iHc7=3FNo)5{n6N|
zBwkogtGmPQh&%eBR@ohA!-<y!j&_UP9MqPB@Z5aEYdC>dcjiT74QDyAlE}uDs719=
zdF#W@dE%;2ohMs@_mPGTPLq~9&Bm=%G<ORdkrJI^TZwM#RCCE@6LykB-cmDxCP?xv
zUNt5uM=OCH$GC{&;ksQDM`P7yBRg5LD`z5SnaVWTXkxF>t(H6G?s2!fH@REfz3lXY
z8`hkKs9A?t3+KY<%)+mnybt25KVUFk5R5I@%_OuN-hxojf>O`|SI~m4parj9b<WR6
zF~r<=|K{eByNM0+YnrA-C%F3|mU{sDJc$3d;Qt}~KTIM*OOeL~E3>T)+lyN4NG+^3
z1Lv+%flgwFMd7&uW!;feNy~1!x4Ltgc22qb&<E(}aderz64?z`W1n63u&{>Gau}5!
zfO(Wa&k15-9W%Y}OG^Nudqp=0LUvD85MRgLz3wcOIi~pIoQ+n(TFy1xVYcVMfl?z1
zD-P_VT3T8y)tj}(Dmt=s;DF-ISVr&%IuFE4y_sU;=WD@Y+_1?B#~P6nJJDIEMZBI;
zJW8Iof!A%qbu-`FsV=b56VjHsTg_zWnW8Gi6-DNKS`_CBo8s2DGqIPNfgM?uFtFl=
zQ_*^IJwx2e%d?#y{J@8Af8xVOpHP3ftdD2p*yESmn--5AT|7o^7&mJ*J6b&s_kxMC
z85uMYj^?%7EjF#4Sx`Yf)IBtTe`LARlGW)$_F1pKa>5P*3Lz5{C&DN~3=LLGv3=Gl
zJ@D@P9u~G<vg_4SJxof=%^(n$RBRk>rzW*L&PI}zC~P)5I)K74>^bPOcg9fVoL5cU
z*8LM?e4j>#d<uR3w6d@Dx>sx9H}nQ*a99eft6dr05{2iw-WaB+)mg{AN3p}+v)EW&
z443@H)0*l_$JUQh^IG}ir$5LwT}EWx-f-$wbeMG>(OzSU-FImysV*H2yu^v@KqKM)
zh4k5N8gH%4Z@=E#ZxPcuzie}3wZ%pnt8>sv%h87b>W#0VJ7330*fTh?5lD^b)?&By
zJ3zINNKoAxM)j8a4K|9iWVL~KwuN|M+i>ojZv<Y&OIjEC?Q?ciZ>?cOSX~sOOR>Qp
z0djSb5816#oM>W4D^Aj4^Zj)@4>@1Oz<pIh&HY2@j91}XGww*%$FJ&r?DXs>>P#BM
z-ohY8?ffZKGj%LgcnJkx0Ku1)_ef_>y=fii7DV^B|5K2Vnf)(;n?D9OU(Rz=?#&*1
zB<F(pz*^Yn^VKrfe@*eh9=;r8gx~*_BH~{+v}u!c_g&=fdA2gI_ivzrJi+Xi*_7SC
zU*G5b-DtzqX{p=0|EPPhe-QPGJF@$R`=<NdQ||ZJ9R(e+O=vQ(tLl&bG*?v4CHXbI
z1I%U(uk`PrF5`~rPtsvA-}JW$;yvAy&Gw%=%=B3)jYR%?CFkzmZg$r`vQRMkh?M!=
zeamcCsf-lnLeC?dAXDjGh;5~GsD%j>tn7DC`eM6Bu-#gMR|>-Vib%Wcc441F?T@$D
z)(g3hxfYSRfPyN0hIWu$X8dKdPFu~^^h<K$My*Vs6*<RDlW@ZfVCp2F)|F0#^#u7-
zI)&=Ep7r%7`|4{<3jE(nYjBpXM<Ab9v2C4UBtI7&iIHp}lV-o5B|W7wAKpt|y};b`
zaQ_2;ZicGKA}6Vv*j9igowFLs4=3r46?>TfC(S4acI47FitSW>ikG+RNbK!4;DjB0
z^w_}!(~QN^AujKs<Fln>s(nnXFdJMfrA11IjvvQtp0_*0hD8FD_2J8DkEBwhyo$E8
zRFe&r#DFwIM_D7Uw>6T_6xq<ZR=YF!U!^L0d?K%#x3X!4KuDMHUlhf^loI0hb0h2r
zxwp_2z$(I46hMVp_`w3>iIS&K*-YZ^q>OY6b7i&-T|I{hIcJIOAP%|a`FSqERuX>8
zAmKQduuVMw1*(C5b!S-4hB7YyZ5=Mhh)coZ7?J4B-S>NO$F9-ly=Oh=Tm=A3i%qFR
z%dS`4Fw(kZJwK=UfJ)78wM2#=7XZ8f!l9;DZE?kDxe0=y|G(}0i*C!}s{b5H@Q@0C
z4{B+#2~rN`AeeRMr$lT-PCMK;l|Q*7Y>ejLT!DiAgq4I93#b)^(l`0Fr!d*!SF>Wo
z+ZTy9fh^q!mw28|`PiS@c16VG4kebgDeW!qvVlb|meMQ~KxsV&E>?4*cwV9lr_(?%
zDvN7gJhNF4RztU&<+l3@wCZMAYKdABmU(SAHZ-C)w-L>Uj$(Q0KD1#PS2bJ6<*Bo%
zXU<dCv<aukQ+Y=sSM4qrjk`!id2qT3eMu@cg*Pg%P(><#jnu}V<D&mL7!UhpD+Wen
zRa~cXrV)Deq<pwkE)Wa#p|s{CZdk=2c~5^8=i6zw@}s7bP#Ky31#p~C0m}ZLXxCV|
zy~VOMTLLO=2eTz`8F`iX?o(miIV^13e@m~tv%Rv4_nkMgu=(q<7Zi_eHtGK-DER+{
zpRR#ihRR3MMz}TX=R?&l9>@vG2Io!`ctGd10c)naWYTF>7Liuhs}5O0vQ1*Jp;W`-
zVMc3SmBHw3X)f4IIqf!<v7vhB=p9p)J${HkCdX4_TTK-AcyYguKdQ%p?@>>3-u<M<
z4GeIySk_0sOHKBi-})DDyCdh2T>m!iZIJ8jy5WiL$#Xv{3y4kup03U_37#KZg6Fsd
z7#^m%JyBnW4Zwm)ZlGmDcPsKZ$_W3NkuVzyJNx?;s<OwIbb%jxb^d%zGp?=VgNmLw
z$w>a4nI!U~WxE+9)>%7fI(hO)Fq(}<7$sH=xIKBi8PapOH5xeoM&~LU%6sLIIV&_5
zXE)Tqj%SgOG=>Xz#VP!ilU@;|JDuGBkrsX~d#b~3(qeXJ=+y>3(w9tvGNwC{pa&q9
zNP_$k9Qvd9$ptxo8wviVzn$(<0_yLeqyMI#Nq#Gzi?Yr*`^lES6D8R`6YJ%}1heX?
z(SF$9gG7O+M*c1wsvsiNtS3W{{Zd#zl!!+qOUR4FbCKCn*hsvZ_nS^sa+1pY1fJ=Z
zDv@I+PAMBK@hmZ@qC~mwI8{8oqeqJvPch#t*;TuNuq{R!M5WUp^Zg5-xr~G|7kXRD
z-hEKgVFW#?xEWgeajG@<eM)7Q?g^#8W6P#ZriYXY)xZ<UlogpG-LXz?<WCdpS-&gU
z^UhnAP$ohwQkM2yT|Qdye+F!_QRMUd>4JX?o$^VKeqImf#3OPfesyF{c!G{64U_8f
zoOrOb#<t6w>U<g{WRf1Ig{JDEx~9reOGGbrk^}}NV#(*>Mtmrf`JY72@<O(Cu+og0
zEEqK-C;fRxJ`4r;$LQw>{oGAIx28w4I9l`P`0qUbUEsgB@!#9|?@{_CFVzI#0FM-4
zFla}lu=i4M$m6{IqB{ek$@z4Z%cG?k>!+*TV0n(d_%CTA!=(jr-jANplOg{#om3Cd
z=CQU)`7GD}TQt#;K}pnx=$Tws<$V&hc`&8KQ{g+COa8ZY1`A;?az0t8Y!)@6lytY;
zw(S2a`0(GrPtG|M{}Pkw!oNPr@E;~3V!GR{_cLxTtiw&AiA5J~u4vA|#qFM1WUO^$
z7AGNk;mpDv5orY}R;Cr4>Qf?h;Kyz0!JQZ;-Y3-y`!c5o$BkGC5DR+4%&N$jFBEv9
zUdAwjD4%cZVhu$3`R*V}F>)HZ@)aT>%xx}*uB_LRHV1gAjm^CC7lU2Cj5bX3D9uK4
zu*++x*Ej6Kee8RzPk|{FQh3ZFEQET;6u9)%iOT<>R1+re>ErG`AC1YL>hB?37Bapg
zIZy+9zN6Z*f)q<Sp6C5?q<!B7V@1<GrAC?d{p!Zxt@}8p(+8&2mPd#@5zy9M1);5c
zeKE~ygS9dVMdVvQPq<b{0Vvr4kR|3ET_LPtUQo@f<ymGRTejxM{=9*di-A>dr&h*C
z=m!x&U2THJcI4RPyp-E;9A1Lo$s-}d*$u*R#97Av4Mz5jIL>ZVHN_t_ipzQbG%ud#
zdPL~G3)}h~@U2cW#%~{-@wq`5?>!wLIsLcLT{<$+rAM!5mcXv>%;2TxXyXD6;u4j}
z81Bd&{twn!ICtR5fk+>I4SH6nM7P9@NoP`s#ErdJ?nEOVM1cLVbegW|hUpzFHF;AF
zsMwzGV$-zN=d-tv<m0oq1|9L)8$TL!_Qs>wC78XP4C~@OeG6|zUV=dVuE7Ge&c%k~
z^(%U7WxP6wXKkI?7rg>X_e9Q!vtNu>_=gkmY$?X0d_0-&tkfmfe2B`<mIyzH!k}QK
z6QcBN>Bx~Ycu8_4?mC`q*9B3-O8#ON<<?+W`pj~m(dEgs5k9&+(muL8|C6E3Zo}Tu
z{#lpJ<Q{fxrDjxLh?A@jKAWnq+Z@(7ME(J)@&`}XrVk1l&3YwicDz(Fw5Z7AD5L$N
zkv1D@<iUE1=Pyu|J-#4a2#NShBM~DKA*tk)Ios)Bz1>nsoj&Q7h&|qa>#LDWVc#;!
zqUY=SoJb&;>ATMyUg!7KixB;>!{3^Y@r8xA>{arwO5>e{>^*2Tk;WWpkN*vD-kwes
zFs}Tdr3=`WfBimhKc9cy-givw<~_r_6=lEjEbEdV8N0+N;tvJSU%{UgIq%^qTZT=M
z(tiz2biirSRV0-npxv~YiV3unRC2Czq|?fK4Mx2jrJp9UJ~?QVL$@N`P5d{Y#SA*=
zbd!FV8OTZYgYUkaL}-0OPIAXUSwJ^O!KXbupHQxno@X8N9%IKCo#h1AjbFrB9x+of
zIg6xH#91n4Dkf)<RC3PJB?fnK7MX<ga+m%Z$@-;flx1Lt>5rX(T;_3Z7Wu@nfE{hf
zWyVwnEWO}W9n9E#<cM+rI?QzZ8KaK<ucMKU5dI1-V(@&H3z<hgMnn>>XsOeT-XEFi
znciU$^!n71xp)VJ&?RlZvJk@lF7xr+<fS{3kn!DtBwXO5Furmw7kR&dKYrlvH+&O4
zWkeG4^Ctx;99TzyPF&#;*wV+rWTCel_5zr6O3Pl|4oXW-EI)o$Zz(9dTc=cJj{jmb
zM?J*oF!=|-@juC$m}*=OsQKT~d{L-bu_B>ne;u@6K`~0_gF-PwBMOSyOh+7w;YaTT
zD8^}o75Az`F)0_91d2&p`$>mlQbClzQYgtTX70HMhP7cR=DM^I4#m`d3dLY3$Yps(
zDCYK5eSJeQbEpb!7s-zAF;g*R$C65s?D(XaiYYsmRC0Rjl1W}U)WtH&q7PktO(*-4
zX`t-*QEnD}WXD4H=2VVgIQi7;FmL}Vpe~;<c8PHU@KM2Y5m$NIOvU6Xl1dR*`K+0W
z$yFqkoU8PMQt;OYI^oHFn#hKbmyL1^>?Zv%Gmw+Kzzw1gClOlTkds_d)a4(H9b<Hs
zZ*$!g8du&jQ!zP<q*BCLerTp*au!J?=PX@f@RdVdwv3u`u7Mq<KbZz{nYX!F^x-l>
z_ZxDV!BCe&MjabbmxHK^oGNE83TBb-U8&P72;O0)XF^>By*_nhF5ZE<bji@)5vWTx
zCFzd3WPEp^F1uihgS@}-D0*sL)FtC<9n>W)*onH-fMHxK!xrqx{PO6%9Cg80l@30w
zt$OtF2!E+b;Vy9r>km}uqmytMhbR^`@zqF`9{hWC<iu*CxABHDqxOwPtp_R{yJ<k~
z6<PB#VK?W|d{OL1u_Cb>td-H*cPPb`Ux%ok8x*_wJu)-?8qKHah+{YW=xSp(?*$SN
z1LC^`Z_wO5@EbJ5z=jc<7nLOazb^5Pd2eP^%AnjgJ<q)Gi_({rCFY0Ug70E}A#H?1
zG*A8%qUj5w`C6*Jz9E{|QMDjM^G!1qQz|W~6iKE3*-XWhN=qua)Tm3^dF3FQ|1ip;
zk5KP4h~@{}Eb_i=oRSIM_a3|2UX1y3-u_j9Xr{+Zvm2v`J5aTttL!&ZF}aGQQp8o}
z%v4OSBB|tDr5{M+l0r21808q)P5O~%p!w!q+#vdJ5~1}CIms1;Xe!2zF*=LIbyEnU
zdE89J<Sdd(5oh_RnTp9-B$b@AbQxE!97OZ9QI>%nrazema+yzYv*^QRgzh)wGJ`=h
zuNrl1glJyoLiP#K{I!{$38E47`qYuRcn3t&B}2cm5KXraRn+XHJ35oG+=0$~4(Z1r
z=*$n$Nk()goggV-hA&nE^aM$#`j&QHO$<BM`*;%A#B(xd?25@0`NBpupeAJbj$Z!@
zo1t+94Q!?(@mdERT?%L*&x{_aG;l+G(`=NW_W}=7sQ^1u0r=6N0_Y1oWCE}$Kw#Ju
zB3%Rnp9RRhB-AbbEel^*)K-f-5){?GN-35E&6+K{$JD+ab84%LIlk4!Tm$Pv(8s`h
zFAygsR%GXp4*FZfYZZLDjg2z$5n1n-bbk<cRxSV+@DC|0?DepM-513EAj1h3?F?G?
zZ(<Yt^=g!HsyJ<uyhCiV(!^5!b*yJ3c27CQwyP^*&k;5zmj}VdE7P<)3ExR&IpT`|
zR<Ta|`)QS=I9>gVjfBml(M(#W+$pSVzT)8Hb-ol7PO6}dx8rLKta-jI3d01fVv)r2
zEjF}(#nlB1w9L^N6qt~w%_zXx4qCsEHXoq7_&DVywrd#;U`0|#yL^%DpuJtNzlC+4
z78}GG?UOOq^lD(22fi&vjHdR8D^B3hMj5en&P!a2uZIaSKuK1RTTnnD5x`ok_!Ky2
z9Vl`$n{aKc4kb^A*|c-sgZ}W1R5Q-54p*I}<_g-Iq!r?6W{Y+myM`;{=!l9#!VrI6
z9ooYCP*|#x;8-@^OqS>Foa3uL-ju&-0lf_)%f%)O0XCh10nmbTSTuf0=v-@Mn77hm
zd)w}`HQ}FxBeF@+ZpMSEeN+BMTV=MBWQd_C7b*()z8^aRyS~!2SFqq8UwU&h_TXXH
zV6%uYs#;ZtHYNZW+Qj7)8^H_U%UIkJ8u6;i6tH;ZBrRcVWtZ4ytfPsQ<gjarl^P=y
z%P(+D8}A%agH!wpH*@O!>{26kn$^(4{v;^16D#<Kv192qxs_C~<d+wFv~C&NvZ`K8
z`(r>5$>peW5;)v;2oqz~9P#NAtVR~Imc0PNVYiwYIsy^wf<kNNqHn0RD!P0dHY-X`
z?;m2*d||J^!CK8OtiOi61y1ff#3ma~!=eob;GfvL1w3J2hX$-2ypvO~>)j-2#ET2q
zX{Qm-H&(IWP<=j(R^Us!sU&9rC9TA%p?ea(EsN8byXCf@gFYvt<w{%gbR}#4XI1lX
zTboSwj1$d!PO{8tAQUXPNi6`eaV#+nFXB6HjIq72$(ofqXDWbPzjux8j92T4ecp1t
z6*s_t*zeW=3stw+cD|^ZC0A9uh{e{jMb@k!#1a@nwR<ZrRza)aVTSNGXql@JrO#j=
z5v-8}?)W}VG7PN2dculVUxp3HxUh=?MciC!E9TY6c7xCs&PW?CooT^jYD-QP9=r+;
zT*tw>YixG~rOsFm_KFBBY)(ZH7wlbR({g($*lMsWqS+eT;?&R522N~DL=(6#ED>uu
zd9;xVc4DF3AGqC2gDF^6f<Oza!CG__$JwsjRa#4tUq)-o=?U$7LAybr_$XZz$7TNj
z4YvE}=VP?M%tz?w20ER=Pm_&PS-uVwZ79V{ukn&=ycE9UK0w#00W9Ht44bly$PdA?
z1^3$5mOGY?xN3OCc26;e<<><?C!V3U#KNRo)xzLNtB5oAG-r!1Xw8?w^=X-9pVor&
zX{|V))^71>Z5N-u#P;b+ZJ*wx_UTP)pPugd^z_fCOwy-Jv&nYo*2wj<0~>gV!4f(X
zS19PHa-t*jE95EsDy+A*ge`+|MXznUn3f*T6&qJnc}u%;?PL6ik`|)D?)J9i9Sq<I
ziIuQP5AiTsVkP(hrUN6v$UJ&NUVjFC_+;xrxHlFGM4*73#IR^VFpZ44%iGe++$gts
Y^-9pJk`t^@9Gj?8@Ro5-ya_)4|LPP|r~m)}

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/utils/raft_trainer/index.doctree b/.doctrees/autoapi/lmflow/pipeline/utils/raft_trainer/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..42fa60b43d6f00a20bd095a2b7c69762d04b184b
GIT binary patch
literal 481422
zcmeFa3%p!abw8d!o;QII9^nxJCm`frxVhmKk&pl}KzIgBf&^lmJDGcKW+roIF7wC(
zQNfB39iLo8#J8eV@X^{9Z57&|Rcp1j*8Z(6wLYr0)>5=$)z+WC|62QT_T%hXv(Gtm
z=W0toHJm%=?7h}rkMCOhedpZ!2L@)(!vAY8X-pPN)dM5>v9W4xENo1*7mU>kt%<PO
zoVssn=&q@kP7Ssf*7N%twN||lPPJ#j7lmS}GFA_(Q@f_xN8;zrQl&8>%Fl0vg=VQ%
zjfxlN7Uvb`@0uDcE^N<jmYS7Nd(}RzGBIAM?H`#eO@@_HH5}oU)$`-cT(h2sU+Pm>
z8Vb+Z9v(ohyvp1?#d)7MnFmgrQ^lq2g}JC{#lze4M2ohx=OEpwEv?=8_CUVbte1AT
znji^m`QfFp2DWm3p^|Sj;753#Tg^}Krv+oJN%hC>d|_`r9LH~3?fK389(*-bTvA+K
zTwFZ4czAI|dvr~)HW99=x2n*rHT!Gzy=$(!{z_=R{0}hLSi>1xlW#R^`N`56E}%71
zKx>$Q*60FSQ>u=I2S(}*Xyz%`9@g4joYkJoe$yn)l~6ntnpIp0@;n{>cLw}#75r}q
zn+Q^>msrCF+jA!KrTSF+tcluKs}f!ucqpko#6Mo`@ojP5mgZExRXn@6T1jV1@f6TO
zaS=B<Wi?01AKC*Om3Ej-EHi!*Du)_3a(VL_^Bvoa44`^0;hz_WL3#_A^p0xgceg6}
zdah8b<QkJ<A%P(0u<zSPy=ccv&VA`i&U>l+?-Y`|L6K|UHr>{+?z{`vtw%Fzv?eC<
z^@Epyd4P!9%Y@T|JB?_sr`k)kzj7b~!&P_O0GKfV|5s|{nzeeNnA=ki^UbiHtJQPi
zZLNGI*DU6%xpQ;po@0Dg2^&yCygC<O<)DgczFFEA<_eRoTz+CQm*1B!RgfaSzx;|T
zH{NvJ_S~k8*IoDgjhn9BrvG^3)*Ci&y<*$8=xx`RS8vO0+qn6P>o;C|#n#;QH|B1<
zVcYhI&Bor+WNy3!&Xe1nZx)JS1M92Q_UsAk__TfN#;dnny=8N5<JQgFa$7cDe?{~w
z)V6K=#_d-`zus`;_N%YI`d0W|ez$GYRaabo({;u#8`Yl@+UDBA7T&tBJ+HZ^UTaOp
z><`g&K3Hc$<6Hnm_Ls(*#i?rt&>pK`k9)!7YtqTO!Gu=9t$GEB?YnDZ2c3)_Ua#$U
zzL<-k6E6d8TL`$F-&4;|7PW%y{p%(VuB+`XuN(C+!eD)OC>rso=aSJEF+B|etS+7m
zt7E0=p4<U|dY!X<GsBrQp!t;&cw)Yyu~dA!z{t}*+Bl`a^WB}n69NNkcl)r(gH!E!
zlLuFWE~Y{~j-XAv6O8I12&Wx^w>=M}2eEFdeNNISHwD=UmYsmqV5<0Ndp`WtJU9uN
z$>GRLdlCE{9+<3@3Z>@Mz3lb=e7!n#2rTO0Iv%3@GKls9D8S<W)D}jxnXm5&n^Wy8
zdt+8C9y|zH!HmR#YiEO!?g!6bR-C7Ff4{E#u%rG^$@djk6h8w26YF^c7NZo71zraq
zz6J=s-hvvDWy*lxh3N9)S3L>Q>>mPdUIpB|+T>=iGkfh^+{Mk%mJD#rvI=D0DmiMu
zczOwP?0A<XSNxg4i{m4_Fj>?v;upG|n0`K3e8dADeo84Zf`{TKbZ;*o4he|A!xtYf
zK2iMREyX`+uS?>vR-4gsXVyu7Y&+?Xb>Y)i4o+G=4x}dh^xu8`^oQB~w0s88rL-(D
z0cg>UNa3ob$aAPv)5~9}3S3TG>=$Xvealy)p0TIAtbH=Yv~uI}wOE98O<i3jFHbIS
z?xN;S&U_P;OXJL!dN?!L{2BM?@+KxSBl5D}bXCwUfquOQmB-?5`7(U;3&G_LoOs44
zT;9*BF`?hi10muIa&&7_IQX{6%VK>+FRzZ3a=G%>-V?E%_94_9ciN>fo-|K=yU=iP
zMS4$tu#3t%dFltCTpCZkz{6AHZJv4eyr0R*bWi*(nR|W`OK|smlPK&F4a=|R#Iw5R
z7g#mj-IJ5+xqB93o)a6jN#v%n(k^%1J^e&&_k9XA%iZ@xY05EPH8*}*SV9&z{$Ur@
zc5>riK)Ez->@$kex$QIW&i~BhWx6x|mdu?O9TvIscST_zcm6gfp4FX~19g9L&$_cV
zdo<F~HZK{q-T6{fasqeWh~H<-pEq<-Z6|lW7Rvo^bmxnhyi9k--*i3D4uHw?E-b;L
z^K(UEAOAX^6VK|-6Reu<(V3H@TazMrR?)fR%!pTX?pT_R&K-+oC2G6#8&R{uojVfM
zqVs*i60$_+*LP8ECwG23l$$Ph?%ej7kIt`Q@-p2Se@o`hAH@>fo$nQeecbtOPCTnS
zKf$W$?#`TC&)xZ^DDREWWqP?(yqL>@w@(Kd+hzX=wZ~od=r~)NuYOOcIg77;yNljB
z`Rdb9E{(6Q_2>{g8a?w)`VA&1(@F8SWKKH!@W@I3UKI9m(!b@zvpVTgpzcqu=T5pI
z@*cjlz{@fDr!Hq)pQSq6>DHoRxYNz!BsE975MO4@anJ7}`%aFw5z3`;v>^{i6Ahht
z=UT&rV>%c9rVE`mM@(q9VF?}qhecr@YaimovpQFSRnt8Ja&mNQQY5r00>)SCcsW|U
zn9J!@1dM;5nvCs?_n`Jf1dNl>9P^M+bCw8rcNe{Na?JanTpGvpiGYqq&%Be~#RO$K
zDgKtsNgu`%+(}<13i~+eOE~eYPWt<-n(j`@$xXMD8jI)ry~il#;iSg*sma(*`gf>3
z-AN5Hnv*^$)SSgh|F(<XIyve0pxm@NsoCh6chaviL77g9za?|hpJ55^q<<v}`#9-e
zaN=2=bp9N_G?kO<xszTW<(1}IRxi&ni@2Qgx?WaqJLTCRI^mQ@nyhFJxf(ypn9dG$
z(NQOdyb#KzamaH$(pjs)Gw+ONFd>=Fh`;H=r_C4B+N-eycgE91VIOBal@rhEjJL3A
zx<^b-j&4ngv{pHzxs2S)8O<WJGn(I~s@`_Sm!h_Kggnx$UUSBm3zcSxjxXt=qfX9v
z2+B>DGg=Lvd1u_ugk(A+{+7%cAHovc8JnW8k2BuJiDz}jM_4u8ospC4xifB#Lf1Cf
zmIA2{Zt?JRmFRny8(!I)ifsq{OVki|z(pFzn)iK0=rD`-{Y4k0bn?C@p<Ej88}SH*
zdTVFi^}fI)WV#;ymdy2@!V=u|J|_zMxZY<t@vN@*e_1u%U5}IVay|N^SHkLi*p}4!
zqxKx!`!ofc9btWT3BDe;#nC=a+sqbJYSle_k5%4Yr8owqNA1%3N9~0NCMxS1lX=|#
zx?r*%Hp2S8kZs?1F_e|N2VcZC*D02b&7^s`-_rVZ72kb{tr&p1#75C*CnB*Ka{eZ<
zB~>_i^mnwIqr85uX=bznVJBqyqNoI8VFfG$zcJb$z;^L=+@4XqwYU?T0XrPEYWlFb
z<@~j-O^(H!<h%Vc+_1M5mV|8Q`-x#sAAbl+jUXR1N)wZnFn~?J!8q(^Y}Lcng&O|U
z3<I_U8Mb=``RZ5@?#oxekZO~`c&#4XcyK#zf*cNNlg-jZ>Gp7p{rsbMzVCqm>aRB7
zqlvKI7~#MiN_NBsyR!NBV5%K>atmL1C+H1yRWmE?s4LjQ9c=@YyZB@scUyHN*dU`q
zq$X{z5UG7nhe++i<Q3Z?@-DP?5hAsUwGjDRkw2tgcKdRso6f%1DMY>&%1u{@)LYxX
z5b5w#+&)%*36qc+BJsCmA@ZGAf``ZhqA-Q(*lb*&-r~eNhRF8B44}obep6LJ<998=
zcQ#y<Ae(wpg8yh@!K-*FBcwc$0IP>ibGgnGM6&_;g21oN_1P&yd55Fo@)_`>D@7=8
zcYLj93FS5LLHQi~3K)mPK#PLqZSb?59z<tKsN6vNiAW4EDievplIz)_4Z)^bwF%p+
z8yxkPwPU|4o&N~g>o_Q;w-VciiNaG+36?0F0W6JJfy3{liFpkOuW-&lduz1ou=BA3
z!C<Kx@IAa9M^X&+ZmX_wV1?Fr9u;6;_#8MpC4NW*?ifE3-{cx4I1U7J3vGjT+iPPs
z1}L`koYjNpwBWHK@@$sk-1g}nT_R}%Uf(LaIB6bb7lZ9Z(5%&v5HptAM<9o6!!-Nb
z*cNVhNBIP-cX|2PsCs5ZtH6k6wHLB>T8%6(!%umr_L7=v!*E2z8N%BOunoKc3y^d-
z-p}g9d;1veSdD$9aDM>~)R}58k&n52wNR|pHPv-HYXNt<%T4EJ=eXTBJkEoGGisn@
zX>5v#P0KI9d&}!>{<&BbZk89p7Yvi&{2LI?RC@qL1iS0uU>|d{e|w(TP+xqaJ(q8l
zFV1T(7;jZ7OpKiV02~)m$iaCIh(%AZ-(CgV_}Q=OGvw`P<ZT5v2C$DM_$R;;vx~*d
zE(#pf{Dp|s-ol$KtCQm$s>v972Klb#Ib-=|-fpEhZlJseDr(OMNw5<K%0>K>8IAsT
zwYHj{y>qBNzgm-zN6Md9^CK?~wHJ$*^6#bag=k(Q`e;eDCV!LPVNFP{5jLBEol}Rf
ziAM-gNoW?^#L|Isc_B9J$_S9nEzyZVILr2H8hFb`0;6Ww87v>8{nqZdmz}|C`9%F&
zFL0C~<O8}2gkA%5ItSRn@)_{Hd?x(U9W?KWs$zaJ)Y@G>8-9{fC?CT294R|CUNQDy
zRGSkW%g=@P<qP2-f-!-J4#qY>LCM(kgB`Zq4R(sP7}&i63URP2Z-K}14e*cI)@|?$
zz}9yB&%E+Y_z_YnzmWfq&##&)?~uP+I5q?x`AYnf;aK@rc=VKrhFr3Oq{vmSAXlHm
ze<tvs&*MMU=&1%zhsu-e-`m)~b@p$A{o7>!w(#F+LYQl*ecFj(CiBMdtXM<uy6-RG
zd<)dpBf$BGs450HXWM}DkNBPg4lMzNI{zM3nHhB)-1QfAu=#G(QLM#K=Wn4<V$?xO
z5_Rxdq7FV=sB<=;&fme$nNUYokPda0&zGokGCncXVNWxSI+w`}fzj!5iJ>Ol&i1VM
z;y56nzo4=f6xkzCIX9|`fy!zdRL;lu98?_5R!FibswXp&I9Td0l3<(NNTOJaA;~Zl
zN{l2(Ng@e8OC-T(3rXOB&hiNSoC!%}1?iAv61iph%We3?kc2(WG?H8;ktDX|KQW?Y
z+hm^=S%|X>`in5H7M3Or#T@lb)K7X<R24&*H8#TBi|;wY5Y1Oe^OmT-%t+&4t-nZv
zZFeJ$Vl9R=cR``VNQ0Cl(%`d18ho~pW(^?CZ^6%*kVaM@4aj$lhxe2|i`+7#`5Znm
zq+w70gOCP}v_5C*e+*!JQy7fYcLu<CGOCIJ#<@0Nd;{Mn0F0-i`Z5ECgSGww2DaS|
z7>czRV0;Y<B?b(nBmo1TC1BvQ1sLZ7V0;~Z&IA~;0%<_W0OO<u5-^tI69WwPG~<BL
zanJv>q$VAyWk(vKYH-HEz764C01EC==jWWLDuy)Y+emXBz86T-x%nzJd45!1W~6bj
z)?cK-w!4u=u@*y`v!PI8q(Mp&Y4BMh4L(~)b3P!=YWO)5(#Q(ZAx#arWvR&|J~5<W
zPcw})8{;PyB!(EDZJ}{8nX$!n!bE@jJBNf(Nt?)!zT6#E#Xx1P4J!BGdk!kDb}KY_
zAgU-cn#6g63H~K)A~#>H)|zkygmF<!d(M^`T*bn4pf8<(D>!iRt+wc4b|9BDz?0(r
zQIhxe?EO>r+Me=<u*mZA`{6%qp7~hD9?W>>CR^1)vlZiU_J)Y`295NmyO5rYll~)}
zbomkZOHK{7JGAdhB$4uC@VZL_mkAjwnPJQIz-DGZ`r;pR5y$(m-12f<$L0P;3^Z(a
z#89loBF3FeQaJ6OVA>BU$%ui^GGgGf6*1O=WL^P3XNnlI0*MZ$W;!oe;GNky3uVNZ
zi%%?Ku&0@h7;D7|C`lJ{B{<b1YyVargVNt<a0=*2_rwg*;H0Q377Z@4qru7eo<{@O
zR)q+|QQ^!Hz`;j<(E%IkMhC@O3>}VxLW%Q9q$JS+pCvlrvxN>90XiHHKW9P*SwT8<
zxCOao`Q)wm#L$5~%``e}jGzB8E%{`eO!j=zRcBv2sSRON(k3z>NG+<0A;`rxg4~Ah
zMLy|hw?dP<qKY!3iG!zyA;?Bq(=n^$hMgST)zcW+U8=6)*`!x?^(K;2GP`;m6a@)=
zO6R6SqWCVF%AY|Mu^dihih_zyruY$>3Y95xDsYX2Q=Ce%_?;fE%=Dyd;^<Ije?tRW
zusbv;)?%Td422Sh2BaiI13t^pfX`NFxEMl11%A#H8e|0$229O#OB}zC+_KQ{&-lbb
z1AF@Q2o1)mGt(Cu3^F|m4OX3f4GpsfOmEE)8h$D4Q!LfK)D8{5!uNV;Fx#yV<j5{O
zC0Kpx;K>LL$MVnsw^cS^*dZJ#H)@Gu!>J^zWNcUtMLlD~YMKfa8w3@f*l;mTg^CTF
z3d9DtJe6wk!wkW}q0asW2ee^#a8Rtpg2T_CP~zZ#lw@$gXBiyu*$NJqLU8y6{G2H`
z$O_U0hX&}C2Zts;vEaa-W;!@*Fi${A92im@vXpg{!9Jy`zoQMWwLCOKKsXdt#rg`H
z?11nZe9r@d-DZUt4@H$^?k70-F@nL7tK_Er`p9H6Nf+T!l1VZQ{5BNz3<G~iQ=!6u
zpyCq-{*0zVg#k_l!hkJFrB3`LLjZ87u)hHSZPpzC6l<{n@Jc9@H~=6e836EE1^|4v
z0>CB+0I!0dGX(%yLAn4iYmp29v+;=q0QNNF0l++UYgz(;Ng{Kvz&;tRzX9MR(4X$1
z83Mqvs45l!HroN<1bi<7fF(nP7=uwInFD}>A1eSf;HLMnRwYat04^e#Bm=-&DC!vi
zHq%t703fLN1b~}pDpUaAR9FBoC8^YjpXlag4nlMt9&)I#zX1Sk)*S#8Yq0>Z6bdB{
z07yv&0DP7K0H3V@uo(itG4OMy03a(!7XV(1+_K@J*WnWj0PJa|1Hd}%;J3u+pY&k4
zthv8&CSQMX{{zB)r1~@9{(GaU817$d<Nm|=p5wmJV1*LDAJvc<^c~C?pucEUqmWOI
z_@5&=BoY7rKv7S`{~MYLh4_Ms58{7^ra~b;rviv?h*2pL-)De(hw}Oh_h_naxL2&j
z;Qk#@C^6h4B?<TVEa4uXEx5lH;Qn3kb0)Z#6-a+K197+6fRh$WxL=M>4DQ*}OvC*}
z+977aAUk~|fa8I>AIt#f0{Ni5R(;k)f5(U5aCoi0K{z%gtgH)es>1rWoiJyqP77pj
z;8>4bt68cvMsBQ!{P?l4>#@ji(8NOrN2=8k_#4vKUBQL`*USwEIN`Yt4=hx5nUkTc
z9;kU^eNSrw4~B}0;*lpgI1#XthuYT#H^700wQ9a%GKL?G7+;QD4nJ)Y#S|lqY7W1I
z6H4^j80WlMtTHSPpQw^}z@np|kWz=@!@=dHLUU)MS(n0bkl3Y&G;4dqDje6?v7zDv
zw(gH>A-5a&)KsxuR`GoJUDeD~mn&6EdaF0(L9Dwg;m)1AfInS3w5BzV@ifX-Gt5nd
z&3dT-N{$qyDmQ9o8i1J&E{H2Ys$fQAASF;MaDnpDs|OBu=L>uFimxj*nx^_?oe)9U
z*CX5c8Jfrfmewka#@3+5?JYP+b0;@i_WKAjkH=+3>f%3FU}7ffl3kad*gclNZY#=m
zkefIDJ2V`?VN;X!S|MySa<Gc5)NJJ7m`z$T%x<=a)kZBM$^SxyXfLA{42dNJM)Gjp
z&h@7^!vmn&RSarYT{0Z3kCuwE^{MNE=#ya8C6~a-u8{c*Lc|nDBdJ(<0U)U!E4@}H
zthWl8X&ae`AU*0dUFX8ib0T^yx4QunnSLFYb7Mr>ofIZGF2~Ud{kWX=A`O=1x?bZq
zrL`A#;VQvwWt=Nm?tC;eCpaey4tQ&Z^$Pf)w&Z!o+aZQ9%y>3#JR3JUW(X0A4F}K0
z4M>an7&ipo%Of8nN7kIUeXa0dBVN;)KyZhh^ohhRqQ?$i#3Lqi>^MDbo5WmJwZt{?
zmqN7LNknVUuEzH)9Ka&W%T4%CtY^DU5}LhE+Zb_(BvPh&wt|XJ&-MYD3e~d}R1QK7
z(MrM%nG_^6^Hy1%IQwi_`|JeOY9jm~**JYDwT)*+isK6Tfrj((5J@;LaEhH^#@<I~
zN+#i^y$C95OooNhc&RXTsJ(FV;5x7_=!3waj&Lmg!FBuq$Su%?Ej|b)@-Q$BCgP*K
z$I14Qsg0mwF1)-WGQ>MHDSnBjLK$L$w){O!g)&4=r5HDxC04IbYt+Ax<T{x~`7tP_
z4i}G`MvP)g^_yc2<-37wY;mgzDL(dc*b!#_oFEnE&JNku(kOYaO@?*s`RX{`VeRhG
zB;oe#S{yKs!Xgo1`m}B(NrdzU(Yl50d8KNl1Y7B}SfKXSL6?O8p(4S#BuzZpcRb&!
zG;`3GY=vec;oRHS<YcYh%rz$9jMv**wPx7hr;@JG+iseIZkcBHqmCB6yj+F<0DLsy
zKN3dU7o{HL;D}r{Fn$i!CXs^wp#*X*DG<_-H$zd+*mw&~g@PnO#mAub&{QZ$5>yPi
z4I5zt{(vD57_dTV@=*bOkcrDT4r$3MQF=K@wneupJrs7uyJYMn2E!IJ2@9quFifD`
z`$&n9a3QEHg&HCs>wtXkrm3ue*D+uh(3hYsAEl{ySn4@?TS&<JG)-&0-J<R|#~a!M
zoVyywN3{`I!{ON2squj^5DVP}?DH6X1QYj;FbYcmw0>0{_cdBhlokIq6!k>H@6uE#
zBotJ9knlffDwKT*Dh3jspF%M_5;>AQ6N0*ES<3+D97!cZJwn88>?B6SS-`bt@8ATo
zfh52QDn7k~GifT60VZh61vC}P067(h;n=y{^GO<>2GKqoj=6_Z`my&2kv(iqiJUQ|
zrcG2N^QQ#<yo+oqX&QoxPwd-EQ=wvCg0>u_sZgdNsQAaeduUoIVjpM0)Us}B-Z}nS
zmQbo`K;HF@&*^!@Iisd1&Y8ceIQK4ECRChz8x-}7b049pP%tB?_{6zS(Nri~5>$*h
zcWM_Ubc|^ybSkLDuP>9l>Nb=pqwh>QF@pU$SFexYze6^L1SCPlCszH4ra~D}g0}pM
zra~DJr{WQ-7BA^V^6~=Uv=^a@vu%PH8LASv<e6kkNy`vad_vU)G!-gTC1}g@X)2Uu
z2rB-e>PDJYicrN_Ftx11jCZKI)m*UB3F{zIc`Q-s9?ur*<YYKAe^k+|M2n4zW+4>y
zjAr|3Dijb2Dn6ikH%)~yH$laSW>@q?QJG@i17wSIW1Oj&cON$bAN4#!HikqtLB%KL
zeUhd^nOcIje4eI4nHs0!ih1b$TEg@Hku*{(e-0y)IJt4EV5fwa**i7He?T%$&g6C!
z&f&0dPb2X&l0<6RFe<$f&2S`|VU6!aM^Sl<hiJU!Nl&Dyq_WP=d9n&7t4O-3wbGe%
zZ&bxs;K6Ly>|L6wE+!k(D^*=hQ|YcM>ntJnl=jUul^%g&NK@(F78fYuAho8I21&!$
z#Cp{c;BaN8h-60~K{oSRvZZ9W6jXdV0uRwtsE$B_wtSGLLRp5O;@=T?jHZ>MBfwcO
zwX9p3cSqnRZBe-~C6I6}dG|iR!ibU#L)tG2u>Xpd6_x#e35t4V|Nlr+q0mQA@kuZL
zjiy4`n4n^0|C@TCppLo!`Ye@eX!t9#&AN>y>K|ldFL6-*FK$XcN<MlinLklV7F2xt
z5CKhvGS~!d8KJ3kZ;MB`Tu;*QHIQC~OD-G}NXc-S!2MrHwv-H)f{IVL9HXgF;W9y6
z>NFL~G6WU>aCs+9D@C~EESOr>EzLVzZey#G#ZD=;hiO-`4N~2M<$$6iqow|n3YQPk
zvZBJ}{ZQ01Tz-J2LZOeK;)A5WPg9|6Oi(ex<z-p*(=k>qkTq*@@AD+{y0ImSdx})z
z81@;Le%-?(Pm;`&9m@n%e$Zno?oRU0dQ8PVYBKj|GU`(vAtdJG5y#L}D39P&9Pv2i
zLjIK`E2-@}5jVYY0k_$#CEMrY0zL}eL{sUmP_MC-7tmCCWSlvgO82%njMH8aG(pnv
zwYpv<i`=KoG=fYP69m6|$d;1HqM+iFEZ$F3p_0V}ZFw(Eg|ZAm#Xnj69hz2(WRbIA
zYFW25?_}}T*fvI)H|%upP^Wgich4D@Smb2l82epikAF@JkIEiD2Sq)z$FI{=C^!;S
ze4zFFG!@F|1QjEDe13LaCCVRvPO`7tU%&FlXIvU~=Z}kzNu3QPpmI`HDvp3NWDa}r
zlOdePPiVRRAjwE-T_xZpaVF#xvpW-7PqvQCgq&2|3G_9YsrZ1x4w?!D3cM{2li{<~
z3Pi_A8mX-z<)pY%1#U&(PBQMJXde~aOH=8t0<V<!%`_D%#t4b|wB>y?mF{hEnUgi0
z_`4(xUvuhJ%FBggCPQS(n;>F;iEJsE@(L<GDevFYRH&3UL0kTXrb1bUpyHqM{wGZ<
zMas)rFtw~(ns>^3T|phml;R|&?n&;#h=5FQwO<q_FFBS>NT|ei5ft@IY*)}!D9{m9
zd_d=HnhIrFf{KyYu1l|%jtT7?tW!&nHjpgqmX#=(<zy1)J{P$(>P}=|Kr&8-L{7y4
z@T;9ErnF=GG%`(!Ig**wQcTECZ^Z4+KpJEVJ;d#`_Uta23gtjf({V3odOb~r0s}$C
z2Mpd$Q|aCohk@FIZjX>OeC@7R83-4S8JoxqBmuO3mTW1Rfe0!-8OT>@DpUrNpe^5~
zsZf?7sQ70fPt&wgWFVXcQ_H%gd1oNEPTv`hJp7qE(gzednULr|sf=X7aj7$sIZ)Iy
zBUwgMp+HDb@d2Ud&{QZ>6I6_h<kE~sZFWpc=CX>li27WTaoye$<t2hn;z)U(OR?_6
z<QkH5GE#CX4%|PE?<V1=nBE`{H`T<ogCv?-Tpj54MtttkrV7~-5Ak^=8n@F_D8J~a
zq?NjooJY-S3cHtNAhp6gc<7Dt+*#t=$X0nM&kQSM@{j-pKSr}fX<Jb7Ngf`jsZhy7
zg0_5#rsB~SBY9Y?AAKz2qWPnWjDJtlPLWLThMQW~jo&+&$agukS(_u$j$^hCXSV+A
zKE|<vq$<VP<truMpV0E9#zLNkqMn(@>}6zrLuDR<icjXTgr-8-si0zH9v5d&V8^UO
zE!k2!SV{7&+h(G~Ly=01zbk;Fm}=d*$6Atk5`Pm=*_4%vBg0wDgC{>UU5k@1AZevm
zhe@tCDsksCA=v;QmH2pDgQh}x8&`Y+mE1+r`n6JtJ4?BrY=DnS%zXfE#lJT*6(7a_
z4o#(}ihqKn^&6o0uaXV$QM`{zzD-l<t`e_l)Te1GR1_Cb@6(q5OH=9I76-1_gP4ns
zr}7bBv+C6#CKryGk&^w81Ub<eWJ}3GOhLt`A96lTh3bbSXv;>L3S}9BiqQ|@Mj@_o
zvn}KXnpTQ_2WP?5vTkWazr&|c`@{)2$hefl<B@aG8U64l%)K4L>$fWW{NeZ&9p(OB
zNHz+5xGyAAti>)QI~ximzK{$l$qUKwSzbtn&(?)xH^9lvtKsL47m^JYr=p`|8{7Fx
zu%Ofk@<F3C0e1lfyIaj*Jj^#+^>8(=wXDEhCwMUuJ486hSH}W8<|e4rYLfxn?GxO1
za69%2hXXlAJjQ<h(L3MwKmhft+pY#S;tg1`V>ctTRafN|q+96bjK}5QS6spF*x3rw
zeYq&v+Fe}OUbMSXE9}j&Gqts|ymwp=J1g1dreJXAC|vHu1<mgW8{HLv*hYEM9$geR
zIvkvZ62yzW&Kg=b5WxR!FNOiHrCH-D7-qt6<v}G-N2?JsSwl~HjNcyB7zwT(5302$
zT*Oq{R~ie);Nr1hRPD1G4JKNRX0SU9Ci9I(I5xrvt>0dnyo5dN92GVJT!K|7odO5@
zYxTXs{!+6TG=VW!yCl(I_d)herL?=AuOA!<ZmG3`LcSU_;L@}T^38}HthMSvZGV+F
zvJt=`=HT>WLB0_e5B^xGRrlb3f&4+v4KmVz>v)3w`Gew35VJf+i;18in!ARz7Z1cn
z{je76XJU`;>YxlLA+Baah3pUWdzt*3!aGKSo3VM~EOxjR1Y{3ala&f^?_HyruNFXS
z;~+Q(!AP!*+0k0730I2c#!B_k0Ip!;Rl#3j0nFxLfY*qO1!pxUCUcB-SU(G|)<bC&
zTJ<_y9v7&OOQu1L3>e1?2ZN)RYuCheGQ;s!m0eGUoS@mkf57=CSOc4dVlZB>;U9?7
z*v7CD@+<7%d6cmz4~!gT#;>20HOj@t8m@_M=cqi>-ryJQ5*vgTA})#>MN5^Y2ligo
z#m%bSX|(ZXFyR~XV`HFl^hJ<beSZlqkQ*=<1;KC^jUR2v+YozuHCzgXSIMC~L1~>t
ziOmr@l@81Z13S!nOHGc)J4at8K5eiAq_IjcOt=TIHBoIaC(joO;bap`f%_Y`9cJkx
zpq>5s`q<zQczoC_*1)UkVUTY%YZGvD9<Iwc7}Udw+CGT^Ti{v%u%J9p9}h~+RTvWD
z60qCI?Y2<^7X++_i%m+mvy1sypaB17RAuwfM)O?HU?=qwgc{&|Jl`k+><tGD4oX6Z
zffQ4tSZh@v=z!8f4taT~2fhNJ4DwA;O)$DAtnzCjMnx;20bC$lQJd9!3|r_cCdaJp
z5*-0Lt4vHm4R{8p7Ps1kfEXA9vG{foJ@<ptGfETrYKsY$sTZLsg!X|8G>c%NI)@<T
zFb{4MPSy&=A#{miejlhO7=@xa_N%5jt`*Vd%0Wg8zGFbB(FlXBTQ)b)JptBr4+{qQ
z6^(IqDpOS_;C23L7hLcc3sNjfNo5IiKmj8I5U4=2y|8K#!xEV_hC=B~B0fSe8phvH
zWs*Yx<toz#lWSymI6NIz5hR$hRRzIbh(6FzGy#6)Bg12W4+%(S(BCe9#@i@6h89&A
z^@sJV@o~-ozy_0~LORw82<5Vb>D`lgZl<Lwgc7h7#<8%|1Eq=9L{JH<dmx(cFNRf!
zQvl=GSpGS}=vX8vpwR&P1Poyzn6(_QJH<*T@_R%4FoZw{4hp~2!+HyxrpjnBSFN`g
zN<}yT=1~t3{IDjB`Ky~P4E!9rv_0Se!fNmuz|TshcDXQiW1xZ+gEJO;n<Yq^xW_Om
z$fyvVB~|8_2$1**fUt1$5ym|P3?_95=k+FQpoo0caK9dQxOFfYLBg`9v=2p$Nf7{n
zWRE8!+ff8Ex=uhE!;k>(--PTzq#KZ1*8v+EpdLuj@$!H|t=beBIf@X{v{p60W-FB5
z4uV{>t+eNcNyr-NCBR5HCmhlSoS$L*LZdNAVdDzLS`CyiRvI4<S<;Y)L>>URfXP3=
zXQ5E5a|fseMYy7>4oRc9{19NU(E<R&tSo;J-V|!BDiCSF&7qJsjtkubB=h%;;b5#@
zn}oauk#u){cd1fpmf+K{Sr{4_ur_rWYQMsHKdMXfhA0_m+8!e42)~6Ug57F3bv@?B
zO*H5{Xx=0Ynn#zYhaeno+If&1ONszhAhWGAr~rpx*T4o?Z3qb`S`xdyH5kP8qgn<S
z*3q`{of8!{M%k3A0+j|lVwmO56S)-j$3*@KsSOw!Y8F06^N?AY%&wyPn3l#uFr^B3
zA4KP!5M#q#gYpB=4E93+M=_M|&qH1bRr0if*&&N<U_t!$AxQCMTM?3md8P;qM7@d;
zUN$E`1flZmWJFs}1lrHIR>m@pU}Mn4du~}{;;zH5yJgx%X+_uEVq2jdj;!9uFrjrD
z$MO?5V=h_cNen|6Mh#5{JXhYFtB_zPqytSh$dOk!>*la`m4m!(qBRNCA**$`zZrA`
z*fEOzmC=y_6&D;=>>8fK6Ca<ebuEE0)@>YBbh%Nn>wSZIX-^URlMwO&mLR7Q0TX%y
zGRKJ=44fncBUN73+|Mzo$!`Azj9}KvE0$Rn+|C?U;FWSd{i_;ws$_1)dX8{y81$3I
zMr7|9uft<$NCQIE7`Fi%i8-bNGQw>N6@WkP49E;LS~z(FUKT<Ca;^*uu|O~ADJqF(
zEn%z!Qih<(F$JDO5xr5|76bhkOcBA2M9*NX2Ed4&coqua&BVym{fEt(Jq!Pjjq@#R
zF9Dy!tE?fv;$vg#CNX}S;UEl89BX})18|?v$0_cGI(axuJ5I43_>KZ9L!)UJqu}Ec
z5sH9oOzYT0zs4n)+#p1(XX6oKs(oR54s>pGEwwjfQjw^4S$l!3Mw=skuQd(^Mzf?@
zYF5G=xB?~x1>=sorLn;*Ii9z~O3gGHAGT<i*KAp1ZRJPs>*Zx+42Ia?a&9(Urd4ON
zTz(Wjjm9*Wj&jsf18Wh7)(fr5Co*w(*l?iLWwQqoT@X%879l-uXs*{DP(3Nh;Q~1T
zs8#pXM67)Uzt)6z*pPnHtIb;H;dmTXemB;|$33@0oo|Of*_h{m0u`-2;tJJhaCyfy
z04>jgL;68WCpau?U)3v_YBu&G%f?KPT0F{FVaI`uN>kPPXj(--w<_}TbfnHa{e;-q
zw7`9h0?q2eX!Kh0IbZbNv?>Np0J}c9czAJz>+<>za1=4QvTN+oNYUJ;0@IT-aevh=
zOqSThfX5g;5eE}A(t_Q{gko3Ns@a*5OpUxI%Uprwc8Vukmn3*bGe$Q7az4cjoc4tu
zqUUh1#UdPRNfPCUR1w7Iw=gQ@x50m+VEDBE6bHhJ9B<%*Mbk6AF|AV@%<Bis@6ly2
z<SheU8d8q=qKIbXFSIPugvetYQ#ulXmNEo_UeQK5d8AeQVyjx(NJMIcyfDB#97e>n
z%3?hJwJrzGd}A^ZM<U;dI5P7Mn<3hswR~eI*z;D_%JRQyZLo8U_GkYGSuNPzHN9(&
zW)`opG}<H6z0a!F%XE<!X1a@T9F@-`4|wysSO*X3Uju5cRIBLWOx&H+dD2y*5fiN8
zZ$}u@vy$6I+MKOc)nkd+6<!w!Gr{DqM$=}W^v)=ty$D?MUZzD_;aWh$BHWuv5_&*G
z0B{7f9gCq;fukhR-McW}g>zV&V3Z86h_}bF(4nHibv~Sjqk1q{)zhhKIsOS%_iEbO
zZ_1j7EHmJpoj^dmJV4J9tB__V-?VJg#E&CQ`Aa)rtZPN9?4Pa5Xd@B<4>@3{VfM6R
zj2Zst8Sj3Ph#}#5iXo$aZ8JohyB6<W4CekMYa|`-zV*Mup1~e3&Dc`S<c>Yj?C+WO
zyTn?aA`SQmHkK(@t!UMqY*puF!N>!%;5j&zr`1L?cjPF}PDv)Bo>Qkg<wi2Gb_}2B
zz~P-dtiVK^)GwE!IX%9^Y8@%g<3z2jEfIQag19Ub?+fIrAgxB4;YBb`DAx~3CY@Jx
zY|X^jDdgHDfBBwhz^uN4Niu)Dj=0(fnncDjHD$5BR!XbYmNSWZQ67;4$Xg2hw06@D
zq~#G<grg)$LeC=*SY3I<M-Q7d%W!bWBltK)uU2~@eql^yOM&>BIkxO5S}Le5PD!!S
zi;u}AS`IUyCZ}M&X*ySUjrD1Dmr&U>uk7mAs;(|8x|p9^V;j4s)rL7mOb#tFn)A0U
zNHK%!Nu8Za!dB-y=G4&JltqL_43RV<XU>pIvwT{dWqZw|WIc!>hV`W;r}t-Sf|>Wm
z8jvRM{X`;$MBW?8)Ewur8KU7?%X@bNJm1UoMCQFaUI1|YETB&Ra@qfdfC})oI+1B<
zPW@HOhfMH0#<6}|LdVkB(r&HVC#`C|;0}3VaCd5#HR2i}n*XJ-97A%ygiVsD<U_0S
zsZ6|IBvym5RpaJr)@ZG|mBoDL&IfU&y2b+Cs%gUa|Cg!Vck{u}M9b)VpJaK<aFMo@
zUE0LLX#Krb&C-+1u6Hs^UIdGxa61An<ku>t^()tZsH9MGDCQfZ7h094XW|o9^g^e;
z2?>8MHx$rmr^Y-j4{u$q%_M`9)N((2OM^G7M;MzQq}mo~J0i5IHfLh@LJ2hX){1UP
zxJpWGv6D$p)a@-m#`9MUs%c-)y1iJ0CrKm;z1xd<m8;uZfZ=m}L}}VKhIELi`daJ0
z_kR;_2a(-(T=<%3CkQiLtE)EF54m%~kkTCMqNaII&GH^IbLmmHZNEi?ws_HPL(W(p
zvv@eT9G2JbgoX9SEO%d6Ka`2}Ik-`Zl6IO7+e3+15*@Zknr4U1mLUx+T8FI+z_OQV
zi0rVnfAuT?Obvn`u1P2d&1n9>(zofNsXG^sTh)3wCGwKAb3yZ;s&nyInRsu@nBRFL
zY#&2}N+fJSk<J(;5_`cDKhD%bYfwuhun2cvl7yZ}puf8kiS>|3MDr&}wmYIfsDg6i
zqP4XI1Fy9y&)&#ob`ITJ9Vhbr+2)vHi=!(;xIoFp*o?7V&QrPek46z6PGu9*S)dLL
zMt*Mjj0qzrYU}w;I;>(qu;bObe&Yj;#H{5hdU?nfgA{8KTa4c{vGV9d9En&Nab(6y
zn<3hFv{+dM-+6|$l8lu*=8J{$N`e2o-J-bcS&^vC(3+pEPGZ$jps0Zvo&_ZnOtfMV
zZn?CvYb%h^*j<6*8(`dO`+%`K(PKNey#<jUqcSa9xwCBJ#-J9hGMnZt+<n@<U8;I+
z-F8K?i&AS`g)_Yh%?GZve83E6r*=6lKnArHn_#0d?t<UB(_w`gxos9XI@c0iSmcp~
zHgno8#(J8_7A9gyL^j2cv0Tk&h<0`@vem)aH?v04k?p<U{Li{@>=cGyPZx$<Z%c2L
zrTNjFmf4zadz5|1hV;`T&BHge3h%Wl^m0t(0v&V56=0agK_1Vdzj1hnKl-a~tgT`a
zO``0i0Z{T%rWRU*S~nJpaP%Qb=-pTZEJt?I{%dO&zmcC@0ZT8T^CPzZL!7|j$k>T9
zM(m&7Ge^A68*#<sjhUMn#pR{3sx+tknB|nFm!GBu=MJ)!ejlj#?EX?C`U#6DZAT(R
zBd3lK9cw_E5dG(g7!n~`F=T{jn<3ibwGe#(EdEi}NHRpX-|*{czZi^q%b8{|-?J2C
zn(0yYxj)h<k2FnDtMDnSLN60UE|>|PXB;f#=i~}@?Z%K662b04Fpuvr254S2*u^5;
zqDT@t?4l7mV0XuI2*2_KAUH_M|5&7c9En69hXk{h(|#0^3&Jl*Bd-Xxk9C#S4EVT{
z%@I7)BbLft^@z&sxGL!qOSbbP5akQ4$`w|Xw4Di9Mn0HNSc8Hw{?g#`nnWB4T#h(0
z2L)}0XqV7%`6b{IOIa&vT>k$5CQ}Er|I_~o?bXcx8q3&Bk2+C$RO~!7=~>4M&$2(z
zNNlx8czGD|#XM|bbn=>3rOh>I)=@CI18jOTtCwa$3no~Ey8%f;4<_h2u3!>CFp<Yp
z*CCor*V%FXJ2$TPXU4fAN}&%zivrMMotS2Dw_7%By2-}Ox|Hse`SMMbaJ|A}nYMWm
z*sz_B{zk0%Y2w;cB8Eg<ix?UhS|^W3B=kaisTSW}4qm#CHI|NVpZd*rCx}|xrlA#@
z7Cvfe*7T0~;@C-P<8#?5lv?q}tm0mliZq#}o@F0r*%_8Jnl#HRyB>Ne8*^%x(=j!Y
zU&z$VqwHfgwVLV1%;N`gN{;8?p)p$FCo{1%dqk_s=7aRcC@kLzuJ|a^CariaEMpPw
zy(9@eEThA_!t(bZEURT5(|1y-J!GwD>c@m;KBcsuCzMXh>7;7gL=$E!`Jt-4Ihhm>
zB()a17L*pIP4mp}aE;xmU5IIpbodIYpZi>GfRb<R^Ex_L0>A~>3+9uG{-rhM7Fo>E
zUMms{Y>^|eh&41#VzD9-Ln5(I3>gENHbXSNXbHtD0AKzeLoqU;*l`!&&9hj*jwA1S
zQt>on0ZgzR_pCPL#F42D&EKDAxtE#UEgiLHTB6fGmRK*eDlfCD^nwuNgF(o7Bia!G
zeo;cBO*dyAKj~UxrF~1?)=Z3_?sI^)>cQEZl2{a_Q@;g3w3dmAma3LcVG)j9Bnds8
zLb&7U6kX)aa4ep7IPaU_&u$C6-e49{_n4I*h@G)#93#2o&N;JYT|F^btHT{uO?Br^
zW8J{&;H<OQN}aRL3RbTUHdf#sF1U-S21{c3A(D`5z$Fsmh;X*1w}sbOz>ZBgQy{t<
zE?{-2n^|y$1e~OGqkVB2oUz_$?JiW{>ar0<JGZ}{pPUTG2p%SCaPvZVK-?O{sNr=7
z1vs9A-(ye*7JyB585DBUs=yKKSTAOgK(+=MWC?E^VrB7gdDxZ+|Br6Mt7CJ&;<6&(
zpRM7P`vT#=co`GBb8RHJ5_SuQ2l9Be9$b-79fLbEMlTP;$!(KxUC`(-P{UgJ9K2wa
z^8#%|uC<Tlr3G-^${x56guUg-2b_O>xjKde&xW}X?(TrQ1S)Xf7R+KqB!C<z0w+{g
zSK(3qa8-$rC_9iJ2sXmX_z3$-%9`D|g>|so*WeZq?Hf1?AMK*l$nhe0X$TVL`nkCX
z*HJ-}K?qei;y$mabJ7hs7Z9&YfMw#*OIZW^HW*YFmf*S#XaFu4XUFR!nNni{`3B`o
zgkWL;+f^$rs^BeUH<rMzQc!qf67C_vdmB`pn{T|S0T*9j6DBM9LMU1R=j_8Cu2La%
zHUpV8?$%;OHS+)xC@x+_u-dp<C&Zg%poY;a^A)%v2+@o)8nqjQH3k;aKnp>$L+fI<
z!oa7zT6Sg=R1-gmbi~;Tra-v;t6ndS!ChBKo?mSeaS>^;ROPqKIU6qZ2TX~p=kQ{&
zR<(*=&d=5w4F<!JJtO>{A$Fn_USP-=z<DWX(f%S_KmyF8akH~VMu!Hh|KsVtdD7i+
zPEe9Dro&qm2M#3)I$BP84}kRDEEKqMQoI$i1<k_SJ>bq#*S!o2@r`{MLwkRJu{1(b
zF0kn2_L=1audSE|d;Z|6M@@9?g^_Hedgn41(GrkH5~=fp`1SJg!%U4hlox#G&ii^C
zz%=HNj(+I4o$T46PPUS`R^iiDm2_Gn`ioI}7wsBQ2B3Vx=9A1qqTJ|SkmS3WBxxqz
zi7fL|uCd75SdsF(;6L)n>K*q%&q3%~#+vDi>-~)C2-AIw={$V&Frr%+@vNFgdy#rg
zkD3O{AJROky)@=d5?qd8^rf3=X!TF%r`}g22q^s~^Jak3PjrUTjCuGp`>Q4aY>X}e
z6R#I(DYz8o!YGB;s`;r^4GoV1qa#5E!vi#GXq0-y@cdKEb|+)_7>84M^C(Z|A$l%S
z){w2jO@$+!zto7Hn5jkQTAd~3v2gH^v$Ld;rgd_x)@71D7!C3q9q)2!6`!4{O-IR&
ztL(dJHPb9F)>w{M9Z;-l6>Z4GqA~qt^W>M4yc!aWCmAHsHvCR>QJ#}v5l2qae!P#J
z*=gv?TCw|oM02bdcbrp05-yg;h9WdWEhI8j5$keLh>e+kh{d|zk7-PY8K^8D+1I0+
zpGUn8++!zexVJ}P5tm2(vY%rsJsDZ);kb(JkIxFj^5zpeT8H%L<D6q!npYpPES$Dj
z;njCbufEyk)mex5puEN)($ZbfOfct*p+MQ*xol)mEB=sGoNk{;ETi^aoMDJjV}fqO
zkhjaoR<2@;UiGVz++&v-3m?uThZ8Xqat?xOB<NOl)zlN2I6N%S7XsO=_~G^&0EF)X
zF1k$+X_*ifapCs8EY9|8K)Bm*a8Zp}9+)Y{nC)8lwZvjZ+V>*JL!|691N&Yg3lUiL
zZKhD@ZJ+DHDsog_BaO9CD8vWjCI55TpIY&sS;f6jDB^svJ};ovN;5T}W)LJ_3qpRy
zbf{Ib@Km#9$yhrUfx(!Z09({*r&*jYCNU&`F}W?z#N6y5f)hVTxDWjM8_a@exoD7p
zMK~mMCefb9XFjIdvsX{y9kOp#ohuyheZZqXrp=#Yk-J%u&LokFb`A+n&Ah^ciZ*|a
zMbu<COh3$r1r`%wRiC<PFJHM*yJj3#=v7-2lLyCYg<$1Msb=yUo_qtTn?Uys;?j%q
znz^%PmCu2HuxZ=SQWSrpqxu2pdyL2-b8#!)sV{{S=T`2TDxMDiI|Kf=3jQ~=YpOjT
zno=)?jVWBFw{YI9S?$e@{Jsz_VQE!iLOd9(OpI4*`$r~AlQ4N&4M+GD81?*klh6Bw
z^^vXc2Mf;deZC3R;aTKEQ|*hBl^xLuPZj5FX-?&#mBrOY6Zy5Q<s*SrvyFr0W3=De
zi_O~DGQ_jB56b&cc<?izr1`Dt2^}Ckaj^VcsIPn>{1fUjHFb0ryoLreqpFI-#Z#{R
z?bhz{!wXas+OyALhB8=w#P~Ik<}+Idm_t!zu1jjmk3kiAog=iB55bSX$K6mAXn)F}
zvLT|fK?9EmXexi^qVh)u6`!^|OjGIJ7SJCEPbm#ss7I~N=wK@9eVlcNqBdw|)6Jh2
z!p-PSSnM4(r`l)Z;s9JLvW7dt8tFu9m=mqhooK2(7fP1KrrIZH?qs?|dmelOs4`Xj
z1Zs4>ZCTtHRC~)c5Jetcez3d+J}%z?|Cpw_4eNNKydB?;E{1<@!ly4lx$+CytM(#|
zohi#x?wKm@kl!OmM`IB_TK3yuYqtqafW6%cZ%lVBc7i0V=oYL<RUj{8R!w7#Y9BUv
zaH_p9f}AO{Z53qgTiBEdeEK#%RimdGJRK@evVU)5|JK>R4fb!7{oBHSk)I>rf!UM*
zTX8vb7MQ)1Un83gA`N&wKUvukwgn+8tn!Tvd{wsrr$0w!b2zm(JN%ch1F1$ct4IW+
z{~1-qVD!T_jPj*}oIM}(SriDxI<NaCk#s&3^+eJWX(|+w3MxKrSw&N!kd#vaB(;}b
z;)@Jm>QGmI!SpB%y|@FoK~yms1JNIY$)y3&Ye95w5as?M!4to;!1Kd^eLoSFo*F!5
z1=8zMfah*(0t3$iJ~8lQPcsdk&o55qs`-g9Q+oQ`g!n0FX3k9KM-|1B(%V7K0!5?G
zRBCevzU?^ko(M<}L{%{${oSlU`r4?z#3|_^DC!BM57AU8AQe=6fb@ej6$(f>6$g;w
z3te@lH|oi{Gb?qdtG_^cj1HvIfZbTCc#UCc1Ei1!OaB7Oxv@0z5(%gHsRgIM3vk*J
z@=pz?vVsh7`dw@VgVXQf6N6LsG}CZ;xmrG)rN95Ygn$}R%?zmvwW{Kv`b*S3PffKC
z6|DYB*oIVXhWzxuqN*6IK9&_$m#j2l!(@%jPZvQ^Pgq?+Q=wp0Q1OA)vuP?6ta2(2
zSj86^VAY|n{=({UI;<)qb_1*8HU_Le1-nZFtT%xC+`y`NiiB4D*n-x_09t=0Y(6!#
z$_k|anXz%ur<*^H?O@Ql2cH<UvZtAb)+@DHolM#4x`g1W(aj96#?%z=TpNA=mx0U$
zz#5+_h`kfP=uPH&M^qJq*iYFID`vwvn?AVo#;C%?sq1T@s3*q0lcqvptf1o4mJicZ
zD2(M)0Apjz3Gj`sJkvkB=B^Hf^%rE1*FjeI3pdOveq%7ZA0&|mX1@aE+%RkSiv(Ky
z+ydH90ni>4dPohlvVsgi`!CoM2DCrGCkC|aX{Lepn%qPPyEG5ttW;rd#+j>25`wKo
zJ2T)eiB%ZSYG<7mfo|+`1-yshSG@t=UkURQBm92?X4qdpUatGGEg`orE1qM%{0nGK
z8ed)x5_S7>ObybXv54i*e**se-$J0N{aIF!!Jp5=)-ZoQAD@^%v!|K%=PMgU*c2U(
z<;KP)`<ggubTj+$BBQdn4_^Z^mx+`9sq)|J@QdE~?^RJ%EOGi%Fgs%kI_ccWuBg7m
zdDAUW6v+F`Lhq%iP<fM};*&QWq^VGO6Q|<no8k*ybqUiZhr0R;w?^8eTc;ap6~8g4
z-2{?I1GR65a&D+K981D1es00+p90KYA@q<MW@QB#VD|U0B@AYNAD<Y^vZtAb*-OV8
z8SypgG&B;-46Ji_EpbHs5-OXI_3>{Mlzv&*fmEY8mX>Jt<j<q37?ge?E0lgSsxL8=
zJ_$uVq4X)53I(NtiVu|jH%*0tQclGIrT8KPlseSaUnn)Ep_%ErF;p=c!_d!y2-0Ba
zsi$|sQ0^ZRJn=gVJih?o`FSDv)Zi&AkX~m-L9fBR=VB8WcwUH43_RJ>OoQi*&W$FS
zvenBI_B@>&WQNxcJ81ZrnlbdZ1vDb?wnHHWz&GNzy~$;-i>hJ({ADoD{(AIEy)Wuk
zg4@Lv=P?)G42??T;w32OPLn#QMEX1yw0!={;PY1t;imR^SwRM$zZ2WUeEu$cVm{BF
zX4>bk!p)bkJQ_~(gnj0je0_aFUpFXc_VFXE>f*W7ThI=8E@geJeEO}zR-~#kBvo&Y
zs$xF<6)?X{{nJOH`VuEqAAq7j-lu>1Ntz0kR0%3RN!90RDpXR%sW_4<e4)d!8JW8d
z>pRrdUkE-?hhWo~-O#HzjzRAmK^AGC_lHo<4ZW7fNXW$k7UX^fAoo6@i`0-SE0BI@
zD$^^?I_x<Tau3HR2D$8M#vzyOEX$#a8)37rkZVxR4!Kr!amYOhBrhS?{8&Nm$@oog
z(x?-nsu<*cEi2@%j_OMcxkFIY6LK%6sZfwBsQ5tc)if0fayb<T<l+lmb*7hgg<OZa
zW)N~s`*cID;y4Dm$AB!-K<)&T>j1eXM-p<efCaf<1IRs2=pr@b$_g?-?klk^EX%zc
zpBUt_r<sP_&A4f{Cy1ML6QE8wGvF>ZYm4LU+t3JjifewW;PxHDPNbSMWVmmQs$y{a
zWLCKSXjETfxP1hQdcy6eX(|-l3MxKu`$d`x1-G1v18(s}2Do*otG{q-Yy~t-*^Rb}
z;TYQ9528qewogMjH`<#1BEc5Fx4`yE0NV$ICQ^f~tU!9989BZ3+j(b5u$_-j3~brc
zOoQ#^&D!3uTDm>sqLrk}uOq6NVRoTbRUBha0ci`2)jm`J8{qri<g+J7RWX45R#t#L
zH>xi&z^;a(o&b9(O@#tjLB$7PucxU{0L!U302W{9sx!T^Yd-5xSAPNaIDH5~8L=B(
z6}K^TJs#wc23>E1a&C0hJVin)er!SOw*XpC6go%^t+Ij)(0VVngXOV@@QFbydzxuz
zy~;Smwy#462Ib5UYn{f6csf`%W@6rjcED3v>thAA?-8~lRh=Q1eMeLkgWB(9h1!ot
z^(BVdk3vyTsQnX~3I(--iVxKOB~68bT293QwfG_f)H>AFU#K-kYfay9!>!^t2DcA_
zEYiU3PoSI|ZY__Ikc$N@$o(Eb?n6QssUcTZApOvcn_k)Mz?l+q7vU3wT=q26kb4<Z
zZYHR`AmPLqqnH_D#hnojXgv)iETA=ZvBY!mU2n42AgYQ%>c3`%)U{E4i6Qk|DC!BR
zn`kN&qzWoNkop3e3I(a0iUU&dg|0f&3%h2o4t4bxQkUx4EB666lqybRP`VsskOoRy
zP|gjd!b2p4;)fQ5{wqM}3L*Q{5GpIk0HJMc1It=pi%$$f+0#rz=rz_Qewn~D>0TC#
zc4k<O-AL#F*!Q6|@Z2@_xyoMOFKkAtJVW;So~SAYwNGb-+Q*{$5<~6Bp{OU+ewL;}
zL9L+T1GQhJsZdbMsW_k(Uu1w<hr0R;wZ_&X+e6%-t9Xuq?%P2cX+Za9P|gjyF|U#E
zi$yH>eH!5RokAa};a655ebJ1aUK#9>t0ep`!6yd4>}jUqcSn~yqmyN)#+hx`B`Z5Z
za*`RES9Ga9KFDw;Xi5Nem%<9R2eC|VGUVq(RWaCpCM#@T6xEj)w%0;YPuSi}Q=wp6
zQ1OB7n`kN&Y;!6O*v1#S>P#>1njt&X)nC{i(lcb**xewmS`Y*AQ$aRqK>Q$-bAxyn
zkCKp%6<Cn|3_$uyp{vx8E-T0Y>2JWcv5fkS_{1QcJ<T+vUyH|e?}m-xxnjOLR>?Ti
zc4<Nsw~1#4?xW&0#<BXh(I9wI9sgdz^M{1pNVR9ks^1?~#o+lDV2b_q_mg#hkIBgG
z;)>~*i@zJ%lg7nA1LfQ<9#@3)YAj=U^)JAy-z!9#+N)&+QoSaGdnIarkIi9T{Ukmy
zuVzm(?bSErDz!a(!a81@P->R)6_`UG%hXxkl+d$X3}p81W8JmJJ^Vk=IJk$q%P22@
zM%a-wgA88&qo^w8<-Y{eG}h@SK7eQLpqXWuype~T&4!{t-)EuzF*Fq_3lmg)vappj
z6)FqkR3HnB9l(QcGNfS+h4mMf109x~#_fh?#dr*we+WjH2AZz`QM#eo?LiWrv780Z
zzXW)GO87!*c$O7Nzf1woFTy4<c;1Ci44&E3OvCf_xkhPEHQ#L2@ji)4Yogl71kM|i
zf^&>~W@tV--Xw8vgE1H11Vsp7c9l@DybnL@P2ad4RfVv8{p{H`E+;+T<DRI(#GT{c
zf})<Vd_PTvf@ML)2bSMUQ=wp)Q*m^T@rABC({;Ocj$QidFD##8z;fJK+@P%Zj)C$7
zNF)s?e;&%YLD}U!5|Xi$3CUpefXOwXm(-XnE69M!|A;MOnEXwAVwlXHW*U<>wHjeA
zKN%j#gv1vnL}E!aGyX16HO156|3t;}bXa|-u=eM|E~Hw`G&9lpAJ0TpF|3^z!`j3-
z@WLSzOibp;9C$7i^#t4FX(|-h3MxKedj?H~0$WbSl>_6O3~1}pSbx#>SRHL6Gj?OF
z;x>k{PlM^D!Pu)oer}9az9NwozqOEc9wO_Hh1sV@R#|~`K{I4}4MvP&I~cO&@rfZT
zdzxuv-HbXdw8rwe{6w}B>ymDn(g|lq*u`dTag;p(VizcDeyfo7cKoC_S?#{4Du%R+
zvLo$lq8byYwf8_#Po#a2ra~dDpyGqHAE2pFNXx0XkQU$Q+B2QBYhLTpSbve$*fwRD
zvm0#{$1$|M4P=o9ZNCWR+-Ph1j6_`g-a_0(h`0@*i`0lKE69Mj-@&#p#Qi6HVu;J0
zW*Tv?WV)SfWd_`&XRB*;Gvn<dqp~>M{uf#RPil=%725t%*oaiInK~xQYkwY9#nASs
z>}Y$$Stgj6%#nHRLMZBqw##WM6xs?ZK4?2cQ=!n7Q*og!zR7^LE{*jUZH;3>bz^pe
ztztL^wm$*uO9Qsog9zPVYj}*rTKwF?+M^I_e=3YWHP*@sq$8Sv(<{3j!=^B-4e^O#
zEqj`2ti5KemfKU$kL4OLty-$?$uxYBbV;g3J2U1kiB%ZK-8(@30(WDdE9`w4e$|^y
z_x7kNhP}sS$KKaRH73q<Ujs!wvG?sX6$*O=6(8(<gr-7aFQ?+dUVNi#&ve+XnXXG?
z{l#8mS*mT+Zv0hD$MClW;z)zPe+lK>_#5*aiNRRJ!r<c&gZB%Kq{d)bK?V%|J~oG8
z@IT`d!(jF_(-?eNqnU4JyYn*X&MHPRGv3Y>)x=TuSEz5E-HMMCw$3`+^iDH%OqAFD
zQkWILy5jni5L^4}sm5L%E)KVUDpq6u`7>xm8vi^I1nKrqA#>@K_@U*MC!tsVLWnQ5
zSIP=9c;z|R2<DaN;uG^q_B7L8`TRny+N{?qnKGiJchL%(nLTrUR8ibFuL3y>-;6#}
z-gyna?M>!$WmFZ*d{*?|JB`~-rAXZVshExV=SFBp8viUnIk$gCa+e;8pIRQe0zGt-
z5MgWg{c~r{o`wHczIEQLSwnYCy>x1@IJY>DY4TM3W96=?;_2|eGvI%#;D1BArm)Ie
z;B&2}46b@RwuiauOYn)gDtnq~SKVA_)$3uknH#UvvMohPIz+D%&g`^{&D!Fg`zEvo
z9#ze6mHSQ!gOO^^5M197RmI#lh;>|MpCi`D_$2xdN3|vHw7eUNdUjeqPE(;eErN<q
zTmFQmLUmd=73j2p@RZi@R~>W{c351R>Teok49y#6>_%I~aSUx=1F}eiw%>ztZnQOh
zMM5rqZ$WN=ko#Jpi_}SxtU$V;sZ6ioqF-WL7}EY<d}2t;o@N?puY)VX8)3d)DCTh2
zMOe+YEc>#A>rG<RGh^;jN0o8lJ$5((UPl4t$II}$-Xu>)M^!QOJtKy`NtgYtiYiPD
zz^6k|PXJy^Q=tG@Q1JoaO*9n>fH@TaU@kwUJA9$*&U9Nnt17RV3r=|s{Db)Mz6DTj
zdD*3}{$lV7eIZKRtldDY*p3155g?E>K)eUaxq;Z>I}(ktkcGx)AQ~?bT1kz@vVsg~
z{4#71L*tj@6GLP6G}CB&15V2{TUA(}QVa`wCu^l@Gh6<fbf1iify|hEth?4YEWa6z
zgXhEUGRnK(BJ4<-L593|DyoVh^I0)uPP|C)gHer%q4{Ab>Iuz{(NrjC7F2wo`Li??
z3Ys|;*ML60$$-r+jrA9sjh!-1^LE3t;yeb=uLqf=f#)AUIX67JeMllSma`E0EJWxx
z3f-heXjy@DO*8O%<;$~HON5?{PYj{i(@Z1um6iPNu#&6hC&FyNoOCIRMmIAaFET2N
z!|(|pa{<G~rwV;f!Y_K07cYydV(7c3|AUIg!U$b7ZWmSz$6R<Rv?YxTpAY5SE^NqO
zdM<u$dF~qY++&0&Q+uwgAcN;#gH2(%)wTG<JeNJqwC7$^$X9dq5H`70VP|6_oA)MN
z7-7-Q?7>T76~?`IH(CQvredEfPaYFCBUPRuj~b1tVxD|n|2^55X17J;_GZO&%$s*Y
zd(wFGOQD?Gn`3&A9*spTk3J7QdY2GsYLAu`NcE<O#Z%ZE=F#`#6Z2^HG}9h^U5?))
zvpe5};j0FmDUfMuFX`A-jCy9TUh1ea?$?i^O>n<<6i}Z15n(%0^%*?-Ls3=Cv)A_D
zvyHKkxR~6&t=NwF_QTMkG`{_LDChQVhc2XdV<F4C*P?fSK!`Q9cgqT-ic@&^KVpNJ
zcYhO~n0K?MnfC50^J8N<c{I%6P-54f2%E*)*w9ov=@g(&H?vDG(`y@)%^7m+MQ<<G
zKV=_4&)NJ@Irq<n=}66IaPDWKs+e;>51o5rv6_-@FOY0ArpgWC-NDZ9pfRc<)sQ1c
zsGdamZm2M?kJX|d9e^LfeFmUtYq!~Q{Zn&Tx6TXA^Da~_A3tl>A*5H~MO}jG@2G<@
zD`%LC8vqr@F#voTw4Me4F97+u0njv6fqm_h^g(Q$<-sAgn&D_%&9FjP9w_z4!Wx~J
zI9QyL3nkjuZyXy7@`1F?pjiu=#V`<al|iM{Xa=?MVCU$|AWNt=#%uM7us*WAo-b9y
z`X+J8*yyg2fsOS&jdk21!A4hkR;x|+Zn8|c{3sYs#OEnA1Jcn?rr6YoU275RZ2oAu
zj{qkiROGO=8$B>;ukt{feKUyY4h1z9?-R<p7z5a4f*DZO8}&B-Akxa<X!GD?I2ynW
z%-u9Gv~D1P|J&I0!=O;fL;nq`%vXa_6&C4L3nA({-?ivSaCH^Ugi#tV)xkyza98PY
zu)kEPh`Qi>Dp<Uk4<^Ei-EjUK)Y~lP;X5QcGGP55hr|c8#`6_C<+o`+wZENG9{rI;
z&*U&zevjVj=ca3Q<cb5TyygH^%K=Ppo5$@<det1XKCSRRtFV{%AVuaqbD0@y)zQ54
z*z@nr#QWj9hSm|BlTKfLH>g$k{RZ^=GLuO8-!yHDg=-3*-}>uoype?+4m@f?d9Nj5
z)3kc5&POa_UZ#i~GE-c!i$cMfYSX7Pwds;w5j>}XB%;xzbw?rX5;X01F$vSk_@QzK
zR))I+Nu>OcD#Ki8j|ZJMTV)C!W!_+w(L(h0Q?;~Y`O1~SWW81h8x1&$areQX+M1X=
zI94kJD_2VNB){Pa1W4>D;Hu%BhHnEVw9G-E$_t>=LCa69dX?#+?Dv?3#N5(6-uIgd
z_2i{oflRYJXN~Oc&c!Fz-DOWR-rb$7P0mPnSErl3yQ|mMrMs(tDlKlY@uSL7PXT51
zXuj#Bs4CX!*xujn?gbHb-Xk-ChV1U114R>ecZa<T#k#vLLCs)y*RTwCcUN&7>+T*0
za!AwNy%oxJ=<XT}IlH_17xC`yb__SiL;YyjUAw!#y)~Su?ZbXalogNHYZIyZywGnN
z3;0x6BPcaT2Cl|F4)mQxuQ(_*f@-Z9jMrM#vEk^;(Ji$qbma!IbhQ?Q^*VIH;8W;Q
zLkt<Rd(y*6IM#x9Nr*&3{Iq6aw^_tYf7IiL)}2NUSr+!ZT`5KB-Q0Ox=n^08HIBcP
ziT$9fE)@dXS}_B!C8#K+x*2VJTrg>m68t=}x8JBy0`O|)Ej&kcjS_Ijf#C%9og9gh
zt!<1jZI-(Y8vIBW2$G3shWcGS5=&pk5Z-UOjR`Sb#t=SY750J*r05<)pn2(Wr}K|9
z@otSF*qoD)ppPNkim31&CJ}lJ;gi4NwhSY#))2#!mI6%6>#?@~$s*=usmNinA%?!T
z>8F|6w2K&G=mCJKaB0hg^o=N7?nKA<I#U3xD{TaVmEq1r646H>m;=7kgN`-=!OA@3
zQAQhqU}eUI#vuH63wgJ0W$UvV>jt__7OV@l>j`U}CC8Afmd1JTbq~~A!DWFYC(|qS
zgVN;iIaJrvN_=#b+l9&;gwoJ(a-?))=EgGq2LWwn>%*BFEY7DDYp$?}(e^9mrI160
zHK*<Bs!%Ouwi`y;edJs(Mt!Vf6lvx(=VE2J4oM=qb1}W2;Xy}pE>=c!lxZ5Os{vPu
zhw3C-VyJG?cc@NwZDl{s6kzwEx>o_Sd}Yc#_{4_l*wai8)m_cerYECyNf(k@q%#lL
zooH1^kJ&A;zLvNd`(453yFhn68o7I0R23V!YxH;I?hhjBiAU~!7m6kxx%-HBq1ecs
zOHlnCxic2x+D79Zyi-ia2JapKQKT8X`v)lJ9=wYgtQ@~H))`ui$A|D57*ZYtDqcf)
zTetRZ2(MRzZ9f#uOOT${9N{MxF**#39056GIf85Lip{<mW31#X$^`bI+WtkUO<%jR
zV-cLnl5*r_hL^8!bImw5Wvy=+@EWVyad>*6%}^JA`h;b5GNxx0%(8RMuxk2kml3Q}
ztioP?h!ou;STrv^9@97{6Ytgtmd!Z{A9~;J03yiIOd@pO?o+?X<{fn_twF47EghJ~
z*JFimvxs>aD{`1@5Ua0k3Ny6{7Ol!<jInhxJ%Kf?t8I&g^pGgP9z^%Jis^t>m)6y0
zWw=X`MD(sUbHgPbbhNHEE3?R>j5dzN%4iA0G<DDK2d)x#&m~);d*1f#p3AgVW`3s7
zyLZnY1!j5o{KNRfy65a^rn~1i<Zz-K{uzUf+Ra*SEZkQrgjsfwT%Pb$BPRoyyXwzz
zRy)YrIUMj-IyaUa=PW4u?j7E$9{lG(y*=u||DUKT)`Ne2tOtLD(Hb6q%nc`V*%v%3
zHFle5FJK^4{H0p)4-x6a-S@9U(Zt>NzhQ+;-sf>8Ihfbq0Qd4w#r`wt+5b^`_DxLC
zyHjd!O>FeRrR4sOKCIODoVbnOJqDpzkBvcm8l;$J4C18oI*mbea6CB@aeCKA#K$9E
zk8$@2VJglgMMaO3MP&tPBg1*vWVWU7e0*Y&fj!N1WVjWMx@*QShRv16bJZ{$3&*kq
zi5rpzi7vckju)qOZ><1W{Yu<+(5)CN1R&M@Lj_;gfg$uLhFld@#bU^tVlm_}@C&v-
zx|69Z?9w*(xvK3iY7H9>-_1JQ0H#_I1F%gXw=@7Yi9Y;7+@<$Kdy&TCl!pOGEF00v
zjdZiGIIO)1VeJZ`_tdZ^D@Y4#_o5&etR2E925anTreW<yIM#%vysavnL<O6s`4!$-
z!0d{oVCLi?GlXsL(6X_76L#%2*9GFCF>*Qo1@quWp<bG7mTJ{?!42%c`O3~lv%V|X
z5a0q<XNtVcO-Mprr`HPL-XpA28kgDaO>_ssJEE!>z}+7Mxb<0@FCgF8#C>AN3dPq(
zqCH=HErk>Pt)%gZs0oRa?cal<;NqWxLpY-?w6pp4TwLFv?_v8KO@%uDPEheV|L$*S
zD%AORf{G!%qlEM#ljUN3t8A>;N4KmU|4_;I`y@|#p59(q568!96R_)fs{9!2V{gxg
zAK0$Q@^%1q_P&{~?+Kez4wCIf@O5J{ER@Deg{edBg_8%@K?kf=38x?&EP%h*w%Doi
zC{&HxgC7K>{ti>2=>g=<1}-O`&ISf-v625{jEdR8I>qpQaQfg5=JUe!UDmuvD<Uo0
z_)STD;Cw2HAzL^Pih5f02{aYTss$Avt3H#aLRqz-Vp#Qnk~J4{F>cU@!#CQ<;`2xj
zQrj@2lFDo^;5zb=-gPAJ<YkG1iivM`NXwgjhbEz0X)2T@Bxp;Krb1Z)r{Ykxy-i^s
zNyAsyFF4Y|D^4bZvQ5bG@n9e8cmvs5(mI51Ej>byej6g|*q~X*J7_8^po-YRDM!JN
z0MH3=;!&CkWgUXbIeJ^T#>DAWww3%LO>4c~qV_S3*5qWZ4qf_*e7$*FtJVw~Q|!c+
zHF`r$E$h~1_5_W<;!`L#^yNz3`rC;-&FE>lvvPOWQSlz@Iz9-YHci*@+n`W)*RfMV
zEjx_Q>E4WZH}QV#CO#xmqtx9*S%J*tO+)BDFY*g)H0vh*7e29WB72(YZsLn_kmO<K
zu##`q>bd=;W>NHddeUWlLDDW`S8g(QAIlys#&kRq%DP|^Y=O-0uE2>fT3*<pBN?o|
zB>J9bJ!NAg?X>RXC9?Z`suIy-E}#<8M4ingQB|z7`IcB`b6t7?37HsO&Eq>%C%!e3
z$|H<#rKqf5l~jhJrX=oko&iNY)AI{yDpXP?sQ9E1SI|_buq&t-l3XConhP<Wo2w6$
zJZ~m>(8Fk=PNpK2DskJ!RF$A}$>m|9xX!#Y2Uy9d^s{sM5>nb^YPz2@Yj!EzLneo4
zDwOpKDn8cx08NFmUO~mM-V~k750V_{>Q7|5={uJXa~=6e@6#mjWcNW(F)hK}x%?tc
zg|dVMZTUNz3S|kLibv=2pGg|Ny6$D?lCy3?j*kcXSjX(ORO~0ML->|Y=W+>6h3Z@;
zz=;(!70NmU75~oV*)**bolDMwsb$^Th`{1gDAu`jq-w=k?Zud<LLQIj+Bc@|Q|GU#
zILo@*elh<#Qc8O365*!eB~{G73b^uLfO93b0WO+~U_oWJ{q2jJv!30%JMMnDYDcVJ
zJ^)Qj(=V@sgxvjdpMHq!qn~5+(YrP;-d}$U_SY8+H97n1`gyM#Zm#F~!5i2u2xLDN
zPubvS;$XK8&Z>c#&{B0zz~@5ov(`ja9Jv7#NEvHU+3{W8o`TaN^3>Fau*okIMWM;l
zw2<`mKq2u$QpuKBNcv&u3U<aRoQM%evPli59WZ%6S|b}U`2aq#0TcE#(*q{gvHm;1
zW1x|rn5=~LEOY7`(+rRp)H4r=taG*qhB2avau^vZ6h~UkQl&A1em1g=Hyo~?j@?Ai
zK002V?4`S2Dd+iP;Q`WQ%<f#G0gO*YRWX?VSZn|z-EtsTwVQ#IoThD`<|{Hs275v$
zDQm;~{}8pyHJXt+5_cw@Dl&f^xC9vUNnC$KHi%4I1r?ue&97)GRJSHUTNXbzbrR00
zIE=!c4LOOV;cEcB?ACDBJI4ew;T36^%aQt<hMf}2Q%q2Gr(KHgSlaa%*jSph>oU-&
zJMD6zrA)Jon+D^68&9h~hB?M3g&#Q6DxXEbvH}UwrrNttgY3Y@vZ#IwKC!6Io@P3#
zZ^2BJTL1>QN-aN@Ev_fMV%0@Ib7WuZsui*oi0!dpb2&8oI@WAgP(RjHlML(bmnyK=
zNzh6Z*sD=hEU<sPM}gf{tr6IlcaxU`yU8+2I<JeG)YF(wQco4uUk!{V?r^&#+h3$N
zpwel^Zmd_F$FRNxl1PK~zX#>qSnqZriT1|)y{idvy#F}j{a#^OPQ2H;(L0WXo%-U~
zD|S)<LqhB@;~`Xb@~@BrFG{%?@5q1^#(4Ov=nvs}TcZk0VNLF6Fjj-BQStcL5_Xv;
z!fG=Z&E+a#HJ2M5hQ2!7#sUixqZ$VDWBJJ@oIws3xQxN2E%nmw7W_F7F$N^6A)Zs7
zuOA!=YH(0(eLoyHJ2J4XwVU4rf(O#VmcM$vG!{am;7D420621Yu+|Fp=Ya>PzEmxT
zg=WyGP2fE-jV5pbUEKV>QmxL<<isPimbs;)rRJXzvX-GSzWteyqeN1;@1^rs8nG{0
z#LTXeo(5u<T%4JK9J1}tCv{evMuTq95=+6d1xFZq*Q)zQCYFzN8IV>x&FsR+)Fhkp
zh3#ooJe`TbxBeO%v5Gaen9TgmvYTjp>JJc0{#b}-`o^c`UzlQiO0p%!r@rVrJ|&@1
zx~nOP?wRZ<z%0*XPsJye$+D-J&SbA;y@7CFzLM*~@KoaKFEKjl#53oxxp;e6{g|e+
z-5yTFgf}eB^&O8f*XS&b(UiF1c%!`Qd0+`W+Rt`DR29o}zZ}bR&-4=l<J(-Zkfh{b
z*_d-*tP_<YF@IFj*b+6tmDiFppVvT9z!`IB$GvvpMKl#^Xh%@-8Kx@JRH!*gLB)_`
ze7p^FDzI5D)A+1h=K@OJcaS{l*q&&Xt~2RWornWWwPwEJVGAZa>y0tTm&9t_D;-qa
zg<=z`F3&T%g3~?5`~H6EQ4HcwrPAYjn1sE?c>j=W5IM#xsF*<DUb^yUG!+U6613%O
zG!+U6I2Fe@fj!3iU6O`w#*`!Xa*UU=?(2hMjMwEz{q6Z1$2P`8y1U!2*p7Ak*MMTu
zbo&RM*Qwj@NGD{U|J2w3wVqbRJN#e9B>h}rI?fKi&j7EiAZ=QF8a9;;@ScuOEG=eF
zGo2P+A0?)kIA$N;eSX4uTL<~fsqtP%rGt8krY%wLlqE+Spi>MAALF1evqI-Pm8`4>
z>*!Ic{Jf|tmMVWGmMUMEQY4Pt`2w@5PD1Uz!PHEe#nvbDM<uMSQ414i(p#XYXC^&L
zQ=tN=pyHEBS7|C#02Ne>O#0L=DquPsbOhbbX|0m`og}ZmfjG9Ix-;oinb;j%#XiEH
zA{#_zVuFfKCiWhh3S~$M+VU})3S~%~io+=EOzd%zhOYtiG85yho0%BdeM;b(Ung5j
zT8E(GvxxZnG!<$Qae}ryLsOxwLs0QwM10sqR4gl}SVYWOFtw~(o3qb*4K&$rdM?)I
zb%ju)&r41wc34SFmQf(Pl9Zm_YO1#r16h)KDx6&bjQYtJ=UhZ5$R*kSc7lw4gezfl
zcY_q;v2M`$U}$N&L7PE<?%7XgvM4)3#<_it2E@BUU%{@>T7dx0u8=nO*>3Aodzjy~
z@_0^=$JA#}ry2(~*7r2{7$;6Rp3sRaZN}<_mg`LGl;evnV&)K@?x)(6BXY>59A5}i
zj?*D7g(J2wI-@7>7fo&b4Vua}(KIj><Iw*jU{+jD`0>g`*Yqq-a1RG()x^0{XAP|z
zz(fD99>=@1_rdftTtT%Tr^%z~=+Q=486VN>9);7J$Kl#lygD_HwkPH{q7#8ecMZ#R
z>uer9Unqn)L(WLSP2G4^>nJ-0b!5C&-=D9;xrL*0$~#|&J7QZfwGF0?*BBh+OX9;Z
zpxO*3VAk6;FYmrCtJ{5DJLaYrTc1SMPjT4?|LaQkHz?7MS=K(%EU%%#U6zBIXmC<z
z)0JKJ&ixYQWrr*xUMPT^F%&p#IG`B*>*KYh)C0<&B!dg2hiR2Ol!&3P00nS(#n6y`
zbeJtc5<>KY|Nj<I<fW{Y^uhm+|7II_WQEI#c4*4^14}ukNgm@sL8<DcF@V&necY<n
z%LtJdW`swYNUPP`jommVN9@L4Y1RH!CU#Hlg!=--G(t50OJmVNa=*k18(Nj$$;A8G
zRPje6Pq)ZxS<f?>S}=Qr;D*G9D8=|Xg2Ts|_0fj<Lj<Q4Y!H)`;ebLC(XTyVAauJ2
z9qrl!R%V|^8SUBwRz^#irfIj@bJnF;4J6qTV*H`+Y9QHJNIUIuF)+)gJ=WtBoAzK&
zzZTOTZ~#^_H*?b-7WvH69=BpA(5iGW`n&s@AZHHWdFearoJQ62nInmu&UY$Xx(SS-
zN3$F+h^k@$@$a!&jx!?`NeM8{tpz~|O4<v8^rBKIi;W+ZH1<YKNIZ?P2a0-5V;rQZ
zP@~U+iqA;;Jv0?+d{Iy_rZGCF-W>+lIu}s#evsr*@3AKuS?f$X)v((AOtofYBuBKl
z{5pq4=iq3;K}DB<Dij+Wb$MRD21h?LJu2+|;OG<H^3?`MU6P&Q!O;l11kSq$M-}6-
z!O`nL5NQTSe*oo(!BLy9j)SAthw;JDf5&umi?FM%gQKs~has`<86FlrI?j)lp#NMn
zOQvyVe-p2ua}D3|&_61}m+u~oSr=2mT?ZF!KomLb(}1WgjLwLu4~YI4&Eu!aJoEw4
z_J2m!1v!TQ;1wI9+z#WD%z(teCT_ifZ9`^f+<(|J;N4$Vb*y8Xs@oau5lJ`f+5Y#G
z!kH$)^hGH4;AT1zFwetDk0Z_UnrEMGIh*O(JsR9R+ah8nB^^8(Idcqd#u!dBxVa$_
zLt=1KF=PyG+6<9!pbu_7gNSe{Yb8Cn`GMaMOODL#3@?T?E8Jn}!?eO<9Oxy3%2H~8
zaMY^S%kGdDW_O(i2qi*#^Pg&fu%3zc%mak9M!W_HUzw=|#sHzs4e4Xi0O3#3!ER#e
zq&2S%5VA7dDM=#w03mbalRfBY1B9&12_9v%0YX+rOChFdfbjQ#tHj4yNw&lQ;ex|F
zkF$~qv&<|_p?BXd_(for4-kF{pV$B)dz$G1!X1Trm~VzToLZYG-5%C+ux~ooC=^4y
z&n45*R@Wz7B29CWdD!qmy8gjxYl5Eu0{h-`jb=F6SO<H<o6eh|PND0PsNd}~HCFRI
zu!<gy6n;CZij5R50uwJ!z(qXKAyvnBXer~`92z7)wZ;OEj#n$?;CiLn{@)S9iN^(h
z21Px`1sATTBwVJ(1qBtKalzwhD%7~3pkj;*9u5Y`g&5xx$lfU_4w01g<TcS~AET0L
zVDAiI$X+i;4(xFOn8>R0t;XoELoIgt{e)RS=eVf~#Rm3Vj?&-Zr?d1uK3yQiJrbqb
z5F3g52arsfk*J-ZH}^<XH#n1nQpTN?bhF~)Q;QCRZ0H{aBy<~}+OY=K8f}APZuuFY
zQ4kC8@Q)^pI8?%Xy^05!)FH?PQCNdxOG3Fg3lAR&^Nm`S9Thqjj^|qyI2fduZwC8o
z^}UT?e^{xoVG8}!5Ipe&j|_o>MWFVu(tyh$@O>d)4JPvqT*L)j!P%p55-Bh;u)PkK
zt%dbf4fp|8c{PjmT5C@+7>!tNz!I`rJ!o`2zGM)-8nvZ5iY;DKs}IOGQBy~UkcE6y
zu{zV_@!urDedo>xEzxQHnzDs`X1{E)E@T;!29)5nit1KHopD7Q(?ja4Cw6uM4QfPb
z?vIuljTnwbDvz?QK&$kXnOd?ynv+(eC!5Ofq^KpE(HpJOH)mq=sF-c5cxbcRIj-o#
zpGPA4>=8@>`b<;b<s(2Lao<I<CHgL>`}SQVn5ON!d={AHeV5PS6YIOMr<v}%Tw~jR
zmV<aN&DdMX+&?)w-WujXgXY;J(wibUu-#0(P|om8VbRhAOk7KJ0^gHSRjhk)W~_UW
z?r2?Cwf#-TPNtHvDR=s%xQ?Y?UjsR%Nxu%?&?)_jYhET?#$l!oGs0kXG+x%tSC^fM
zAoq2l$8ITB`x}ll;Ws&d>W~e+<JYN~WVUlOScpDWkUHBr$*Rai1-<K~WjjcnWjkv<
zvmHy`3Y6@2_(kX;G~OkG^XZGw=K+Pp5n8e({A`17gqFT2;n|c#_tC|xfmt4*ufZo4
zq1n?+N9bL+R4)ohxk9DEh8Oq3_LM!16w$iT#0@NBdUt55eQm;^O|g?XfETFR2eHNL
z0yrJ9+JOBz6JfnEA`i37;kyymLxj#G>;r4z`IrrMfZ2BJMDE;JJvbb|Kk~r+hxLol
z&)>QJ^k#SfF1}yI<+JLN;b1+x0j5!cq0GsAAzT+k-vp~JxkS_hQ?B!P5iKPz2;Y*%
z=yk@=UaOD^CFI^=q(_ays}USTq4z<{0RVClJ(h)BhFOSl&Em<FGGxj`f25G=<q<X4
z2xX%5$;G%aI(uGnLt)J^2X52WVvXs}L1v`1_#a9t_mff~hx*>gbnbP)+k0s$)VQsn
z;xlghJ2Vw)+*VLA#%+g__&Jwve1JDuB}y-!A={#d^+Y4V$=FFXWc&m(3ky5$hPjK_
zoz(~6M}Pq?dox_>4K%Ss3>m-HLB(AtHe~GbUSloj5+`0ej+&UuKGb0Msp-+0pp!}+
zKVX8gymY68gV{SZ?q^?O=A@PmDKF(z93na#0-i=<2}wd%U?Q_-RC*_#;Yd8g8t<o*
zypxkpLJ2+~b}mhY0x@4DTuPGobx{IOZy>vwy-QQVHj;M_CHUfuIQzi`7X2O47}SR}
z$b`r}yrEhU8{W7Tup!OxMgx@Ko?W7oF*(#>Oqo$liq9-vhCRm@37G9V?6Ff{X&iN7
z0_|*hxnct5d+K?Z34)Y9*bS$+!m;yg&IglynCyY+A;_Y5N<X5V@&M*Jtu$0Z@CwrK
zb*I&b!mmJay<8LpaTzOtUxJAtt^iKZpBWYBBCu6=G0_CjD`Hc2k<=P+x<4Gmk#adM
z>td$huqWKv>tfSMi-i(g7dwxa*Rbp%OQq%*&`2`9$dHvR!&CP4UW*ws_*?cRZzI4K
z@O87sGF7`_oiP`CgYQ;bW;dg`KENuam#LnURKrp{$qOnGJ)T5F(aON9uuXr&+C<A#
zTN%j8a5>Pjd&rXAlm{JcWgwIH8$8NrD+5^>4N|6QcI~^sRpPNb$(9(qyVZB>P6iY?
zTw)5n``F!o1G9YW?w9z)#_rhDOpo2k*|lieHpO@yzi|Zy>Qc_GMYJ;y&TSRbB2kT<
z*89PzwhZhrT*vkcu^&cmkUs|-f-SXb=v){m=V&4_5<KEB6gV7rDb;yOG`zGls)~(^
z?TXFLT^3)Z$Cf#IH=73~bF`^1%LLDkn&ujWBZr$-K~czH%)=|)_X%7?Q=vMtf{IU1
zZ8J@U>Zu7T#&A=DWrDYml=WDhtQU<76;t)1Z(_PJ!<&aPO_<gBc8>X+&ZwAHyUNqi
zg~b~ZUl^5@4zB&8Jo}}jw8>^3<Q)6h>}zN$l+6k%J~sOxO@*>qLB){ZaZYUI(u|LV
zcKBLJ`Vo>l-98f8bqBhsEd2vaqX{fMfpR`eHlLh-5mbE6Zv84vg*v-6L0i5}Q}Hk~
z|Fc`4rfH=(yOlRI6+HPAiVaD-B7T2InOqBJRQ}^0SyEibMwS+XO45uhExWAK$WlD_
zlA}qkg)<TB@e!n5*dIGmm`v9Zq}QkwB~hIt0rozl;1Sfu)tE<E=O;13-Gd=qK^Rtp
z2JEtirOjcf3D+CKPcTGcU1ivS1;>>N4ms5aaM>|D<zekHEcg{ePS%LiuN6-Uz0UyM
z=}THnkkrvzAE6pWb(|)O+9Oog#}nOP5cl(}V=92RyFsQXe>oJz_Wa<s^s$<I%&9mP
z?V;I_sdkw&WUNKW@4{5_E1>+&hobg24NKr<tqae}XfmIsF8sB?RpKtZWJ`46TfSX*
z>CtJs@VkLo-i0sV6YIjWr<v};Ux`*49?0XBt7(?oMBVp`6Rxw->1OV_uh(lEROgt<
z?s=<Osx(IM_mRutAAXz%TRbCoPgzT77wZ*DUo?MIu5|~fv@<j%>YN{ls$!k<1F_Ee
zD$m9-o~_PIlA_isj{#LyN<;ri$>L2>0}^+^Uk61!yWsDlsZiY+LB*##^AVZ~m1PPl
zMt6ohQmMlsUSt`R`u($K!Tu)rzn4x80#yGo$+8aGiTXGs8>xCap8`hhm_l}TxkP>P
zUmW!p>o{KRprQ*w4FtrJW|t=#>o^jOsQl~nXeIr|%kO*31%)Xlt6dW5Z#UuueX`m>
zXm=MvaUAPHG(ZArx)4Wg?9_!Yql4@|oYaZsc-P?o=2|UbDqXt{zjIYsnS}Hdl2MrB
z4nP_rS>;#y;!ZezN<4L-)h;{%1doY_X;hemE;Sng0zGspMuMA*VO7I;7Co`I5na<B
z9)MfdaYJGe`V(-qJKWU1r&$~h;0Ul1oFWgGxx??f@jx)x%?O(wS>V9A$}{~~*XA-$
z1*z*1)||`i@u^*_EMjC&L(WwqhitBLqQ5d#k=K?{XQ<RWNH0O7J6W_CjBc8e^-F<5
z;$&U2C6e{m`6lbqhtekNF9c?JvVJo@v1FY+&2+MUO|CeZYruZxLJ|H0n}_>3-)7U!
zp0wLF4mRthd}Up5Qx(>9?1T}JRy7Z6-r=|5;PO(TxwFx%^EK|f+%s!-g)&t=D)vIT
z&7`n>X)LC<CraKcQB^E?e`73pKg+*?j{N;OqO>j;xJJ;F3<p{BPVTv^pcIDqOG)5W
z5!;DV_E$hr&y;<Nrb4Ccf{IVd{vMhNm9h&eM#{dzL)MOEsf+buS{VC1k~=*yO_ZIR
z<Wi;MALEMj5%HgqjUeX{1Qm0_%6(V!*Jvt~xg==IcWEk=xo|2D<<CKf)Rg`oB#qQc
z=VY=d+f;*l(e}LOn~9f+j^u*ajsiLyU|f^Wu_TGq@?lhZCm#2r?Lm@v5Ak^IoO&)z
zC6$$0E9WmKNu*Z-uFvnC65NZncaXe$D8bB>$x#R%B+LT3bku-?HVW}FKmdEJ!(Avg
z3gJpA`rGF=rtV^Sl)K-pxQ_L^_kb+Y^t)dPN_H;_k0*Ds?`=%o*{sL=-*3dOP)VS5
z*Z%jhSKSDcZ<vbX92`u+6>4Ll%(2`rRmycAX|lR)=zX;*Wbh55==Al~-VGEI_thj@
zqObOT-@clZw@klH*Xq7D@)Kwm6L4@r`IGp>`fBWHru%AJa`kYsR)=dyYRdGo^w%~e
zoM3U$&)jDlch#Dp-!>fJ6slNG2y;jEQUSI+t1iqq9ID@R_wJ5$k&YbB`BFLDUkk61
zW@hFwiF$coj;dn4ybs2Dc{lp^^8Cah`gD!O+6j%*)fxPksC9`ugWrXso}Iz}q^VHZ
zy`bXLQ<<}gOkSzHNl=;2&fqeVJ3R*UyEAw+a0ZI>5%F1MBgoF6pknqy+?~Pa(NrjN
zNzj%nX)2Vta4H_1!R;iC)Jo6L8FZ?_-5D&B%#(?BM*&%#!F?o&)bdH+8FY%r-5Go}
z$-9SmygcqLG?i3VYIO$RPm)Nlgw&lurxM(q!9OH<_fUeFfRdd-9wf{Hx^$eK!Ef;>
zX$13yQBovAy%Bfj;9HsK@IO>y^L4tt)X4~c14TXK_4jEi6aWh<KEdP}nhIswf{G!x
z1QYc0E~hdDU8jj+arz1R!+;@s$bhs(E`T1`;)xxza+ht+7hg;BjQ&;@dKM{LvQcL;
zDR^zlc^*xLvQR<A$3m~9sZbUws2CP{l-=>+BKF85c94vuHrNb##CDgi+<8QWWS+!m
zPQ@W0E04IHB#~M^>GKGuc-(o!>qy=`#N(Amyp5(pnQ})Z%w^qoZEXKTB$3o=5Oh)v
zEISq9&Ltivx%W_n3CA8#XBN=qL*x=Zg<=ECu2GEs4lD=y$|&dXjeBreaUL67{sQDy
zX$F`76%_3rTy_sS$pPlmx-}s_$oxSZ!~LQd{($W3Rv5aJTLy~r*ecLdaVh6YFv-*5
ze`mn|R>A*<c1_`o!Yz12tExZ_XPG9@ebn@@E99u@;rPTxP1(~-kDA_)gGC=<t^xa7
za&W;+Vee!OR%mA#HoZLQu&IlI%;ToVcW9T$sf=?8k8^%2Gpr78RlapHXjES-mtm;$
zgs3Vu)cI&^s59L`&aP_vi#;n1$8~Dl4L*wX82B6m!b$@^m!Su~5dHdz_9Bh1sdfid
zOY}Ki`*_4@GjaHN6yfJMq1M#!BP&P?KRd9=Y|`cyd}8p!o@N?;uEwo$xe1ubtK{$!
zjx4})NkZVUNN0we<yM7!mS#i8b19Zs-%G%W{jQ*<E(}T9K?bO)Mpga)?0pHGoJEy5
z2{|Vr2Z3+~z63%hVd#WQBoZKiTmnIi5D9UnGyP4bCq3OmAA~_vluO;w1&N4)g2!K3
zVPWxJ*K2pZ&qWtkS$17;-Cf01SK0lq=l);4s`~2q>U&*Zb<Yg=-~43!y8EkoXTA68
z)vGErH4mpuO%JZBAo_@zxwpms)!M$&rULGD3Gm#D9ydW#7)L%^qwl1lP+Ow~6#q@j
z@1<b{Y+B|hn1O~Koj3#38<0DuTDTjf0#uhmOs$hROt-K0;BdPKg>=e2lz|l??%}J@
z&g~vj15-MN6$IjGm+&xi3A=^91iJ*;fiwlCqO^}lf56B@m+;T*30(p{O}9(91ouz$
zkCdCjFT~drE1md-yzBTb_#HckQ(V3FqT}nsEg<}I`@UrB^QLo)aDkPx?D_y#U($QH
ze^f5ycS7-`-k7!`*TVd-lCIE&d@$ugR<@-NvF<fp%WQ1LeU6d6(|&+WE#i}+Y}vU~
z98M+!7DH1z`|CEFT9w%XY-HlD?cZ?_VE+nK=wB&PjtwVeO}9u{Bsbx6O%#$c1Ts&)
zf<FQ^>I+57+2ORD!8z>X>t9Smp?tl7V)**mIQk@6;*+AOpN1CT?Ku*r5%<(UpGGNf
z?;5+C=6I(@a8;%W2U^5%#MH6OP{Yl%jrBYNUq4nJiz{%o-g?{yzP?@@f~qWj+1}oX
z+4Xp+T-rNfq@$b`&F%6P<|&u|3>c&kmwyY0)9v!z)>Qg^BOT=;A?^G>$ejOw3Bwue
z{ACByh?^?V-ub_i5sc3NUiO5}AD^b%`Ckm0b8T`v2@e1|@%?AzoD(|0ckKGlaC9m?
zzcirS#zG{}93P7}p(+(Hw}0XZ&JbXkqfcqAoWCjC^>L7Rm(o=qPP)PX;3KI3V4Vm6
zJjfLO#aO1-9KN-|N`(lnNZ=a@Ub&~2UxB8%BY`hqBQp}{mizpda9W<?X!*F$XJ{yt
z`xH>}xX+pAQ7SkR;66DLrWtJOK3!s4<`{uOHI2|`Hohl^ocN>$*<mz=RF%j<(9|=*
zv6hBHNlid8q_&@7cO-4rw~EG7Xy^e_<VXcc(Wg-=2y)r4sSSdT)H4Szo9>Q*6ym8E
z=wTSiAu-U)L7MIu$mvgH2xKJo9R#E!ppUQ!=n-LWf+HZ=L1+YY4I>vLpgrsfBOrX5
z?g;3G{d?-gvCbo(jX5J94Q$5|&k0&*sYqvQy#@DjtK*Hb6XXlDZlpCbzEig7HqsX5
z%4$p`U13o3@l;T=+74=*6bYx%m3ZcT2}-%+nRi3e-0{piv5^_i?3cuFj9fvT;S$y4
zA6M1g*?>=nlky3Ul}|YHH5v-#7zGr=G0sXzhG>(2P^$Mh4J*J6auj?#nwE`n$!xJT
z%3M;@6Hr2K`d5n1U($3^A<fVHB{$=IGBl(lC!m;;W2<k{cFe*LilW6dtN_Vz6oMq@
z(<l|dxNOzb1~3Qd0gP_1-GPfjI2E|u3Q7_ZxI7P}=nh;AUm*jQLoyIf2Pz+Dfy(QJ
z)&~bFvV+h-<q}371}a<F69y{yG~I#98R!r?NzR;<Gekjv9S0|$<9OHO2HgMDKUA-c
zYK2c5qb+cyM6Ft^Z0CpK;0ZW8YqiRb9pa7&wnXE5YEIImysHZ51o!Wb#~b}{VsJlO
z)6!sF1P6Pu!*(<vWq7e%Z$RR-JQ~aN(7JW}nqFs42-e);ZYyc*#3yC#hDmFeD?}<K
zU15my7pV~GQ_pGc3O8wlZX2>F0;$27Kukm6|CoajGI#mf<?c?T$g+M`GJI9S_}qcu
zb<h-yn7RAUec#3e4TUQ07f^gk``=DOq3WRp6eIdvni2$;K>O152T4TqEK<I`D1cDd
zQ5^SkQTZ_Ovn1~1c2EJuR0#KW&~MOCD22#D%M&ydN+CEDhp1<<fvMHU|CodkENc!X
zs60LggMLpU-z{M;ynu|ryCm#GX((aBw&MF<5{9p^U%Y6aeT?N!K8@%=*k-6t7AEAd
zr)QDWlG-7l_{^*?rJ+#svmCU%oQ6Vahk)Wg3#`zv0%n061yjnpwmB>Ju7SY~{@DmR
zjunxne!5-i074?(fEW-WRT*B0Pb!+YnWmhY(Z2?odPWm>&`>CQBB1y<!gtY7D2)+N
zjA)|E=Ma8~#7@_ZTp>mm&LMo53(|+hUnGej4UB-|6Kgz1L!s0q2Q5FKp-}3=p}1lV
z>m0)WAYlXxy@M^IPBFN*jLyC=I10#QkhNuW83`hoKjB+Oo$PUM8C^r-PVQ6`P<%$*
z$uyK8EwzrGIGY3!E`;EtC!9in{djZXKkM@ND@fcugkZX8G7!t*igwdzQBo67d;+m?
z8VVJN<)GzO8j1%k{(;zAXjlP(7$?+}vaZtJf!J~G{9Tf^YoGc#z}17)YWI&SP<x2x
z3>Bz-0GfIRYM-Q`P@Y#n@v&)Np`lP3C7>9A+7ep^nOUD%>`g5d@C1p7t|qx+G7cta
zcKBVFtlaMT=Opr^>Elow@r&i2e@}u4=1;hLcCyFq8TNlsaP%Ue_^5J3L!ngJNsBxA
zR-mC!!!idgr_oT_)8bGld%g8S5{9ovbu%Bs8Du&LG7QLJ=t?BD<a9|u@d*R!G!!Zf
z$U(~uG!#lZ1Qh=;;7v5FfG~igU`konHt#TCnPX9o`x~=7c|YSl>E*3&m3x0bO)=%(
z-vdp-z58g&M`$RNrU)p84BPCG`D+?lfD}0rU6A5Dxr!=8XE1N(x^+@s>`@8n37Q~^
zec$CmGUApwd?8oDk{=bkAJgzCrQq;_IOWqQRfXbO4xZX7l%;x62VVwv*Pkd<Q}rkP
zkbn)TKUs27R{e=sf|r#i%Th?DYfk=x)tnT?Iu8u&^l3G{+Kso-#jxO?>==#WYO_8b
z;Q=nz6~pacREl>b(38DCM7NvzQR$y%h#xMR0^+wHr1p|+J9^)J_Hi$LN4*2EOgg|W
zH5*_{bpw6TI>>fo1;)Cy8+#)7cfn9pjPSL-s8WM{;n7&JIn3^1D@N?xno^_=u3<Nz
z@$c9@YtdLezNTDjvAfwC&0=*Bu6Z7cwhwIOd(j4V6tHuI!okBe&7~;5x&?c0BY3S9
z143+PF5U=TDZ<ri8Z6-bI@ki&DAZcbu~xIcRIaaUhXBHIAOV;*Kz?ffZ?vaesYIM>
z5vx)`6tMJkJ>$@5`MS8iKDopij`_wL;BXspDT2GH;{rdU0a&cBA8=l)bMPQhv%|%8
zb|*G2A4K(5mB|P0yN-rocT#<<STDjUgm4vJ2`*k?q5`dhq89sCepr8@$2Ic>q>_o7
z2^|@<IjZ1QcS8}Ym|-{TG3cDlOoZdL7F^8DP7Q(kqlegacV)QzPQzy{G$=Ns#&E4w
zDZ!n15l5mF?O~VfF|FDKIuzI0C49hoK}ZA1V6{1|A|os+weSrF9IXnxUg+VLo9pK0
zQmM?fm0`N4R)@R3#Vy~BsKu`S1~nZnm*S`t1DrvSk@iN9^$)$l$Xm6c!%wK{9l1yQ
zZriRIp~GqNhtgCHqbV9fB1Cw%#{P+ugG?Fk2~wB`XEoP?jpti7*t~vmHX1`2I#|>8
zGp(lP{(j4R%gYoQJiJVC78~i_h|J;AmYCSZ7MFE|%FI3zk#FfD=}}HYA~R3A5sjKs
zE`%Nh2;kHb-ppLPQR~YVQFi3SinQOw`dWe2a-U_M@+4&V$WJsP#9?&FefHaqhhhG|
z12*P%-q+_P!)!LPP7~OxE#c=FQ@2`8y~Y%ShhyrzE`=@Lw9HQe961=~iik(+;%yz#
zG*5<3T8}h?+t`W+Iznfm9=GK{2_rnoq*=JjLj}U0>IkJH=)jRe(lBjHQz53rcd-?F
zYF&Oa7wW>$B?`N`Z4k@6uLh2myq7vYnG53wxM@x>zLKlsWCYO!nWV`dzC+mizhw6Q
zdLWP5hQ)lu)5=F+8}1`WAR~9HHt1K5M>s~fIPn@D#^;{~Z0*(*fE(ozBZ@jD5G&pn
z#8Goa2bzH9Uu;fYO%uP!oCnEaWERn=;lI@C%**IA44i=!joOj{YeFWGTV#H((Rx%z
z=$XM)3OU09c?MS}=R!!>oP=LyaAhMz4+1T?`Wzc<vr!0Sa5Zs3ca<CTv5P$_CA>5b
zQxlW(dB%l{kqfleY9<HUaL&=()=Um^+huaTm~+lQ$@Bf-xxUMA74p%R*sJC4cG{4x
zTEaHX*8(kLBF)46B#EMRbB)yvtpdW8GT3OiOv{i$Da10oE*C<=G9(DiSgX-y<3p>e
zW*EM}hQtUGNgIZ{p4GYtcIMfOpfX$U$s_T^Y<-c*=suJYt+Sfh!Zuv4v@U373w6O|
zwl0^cy-7&x^_t<&Tmeds<m9tr;$P-rqMl*+GUMZ?Jaql^&ghi5ciN32`CEBNPW0w0
zjNWhN(;IBVdE~7($XBj6belV^sQfsO=>ImWjkcDu^+>grGI7|-m5{Nn4aN<YU4~VT
zYHM_Cyi^;ER<4u+A>ZNL29EtVa4h$=vl1;axp@M%e$u=U7bwUzt<Y`Ii!YJepqH>G
z+y;$L)4dJ)oPOBz$xk=zWH0ooIrl;<s2y*J9#MVu?kqM3hx;3_wuHyB^s&{9hIn-6
zt;LTx857U0KnH(vZd>kA;HCcRUn%3T76hNYFqvxu^XjB4Ov!#fwSoENlgkJs)H?Qd
zu?14RV^j_FsCN=xX<sQ`oskfjd!z6Rps8m`$;C7jswP4}F-l5!y_l%<wu{HM)35>x
zN;nGH*2`HIbPDHAW1(h}=O=^2C9_%VS1B$2LCL8}Q$<yIjB!ki>XjLrBp`TFP+m(z
zq2wf>_{-@w8diXuI0{{m(_w@Kb4f01Kfch=te+L%|AgiRCEfqyFWryPP$=mNDE`v@
zJPj*Ax*UZrNVl^aDgKcrj1tx3{-XK`4TTbwfZ{KzU(>JxM8#3)f~XG3j1XMI(7iiC
z3_|yhY6vddLIn+Et?y!J>gfnq(oiV53Ml?^T~EUbkSj+aNUlDOQuTtX6U|i7*VNVv
zF4s%(wD{9qG^p@R6%9r}N=VV*<sd}&p;db9D=P;NcB7jvAN)R>39l0Db(Rkfdt7!W
zI|!YtRT*)3WL1ql;am-$rhBe-3EVMNjq7~22DM!Mm0E4AlUZBdt3_P!JD$HaUA<1n
zOzr>|A~GO#|ELW0>qxVfYj$^2(iPg5r~cq(ckfIH%ssn%J2drlY7f#-D5oZ%_&c>v
z(69oW8b`s$KWT?(-pRSst=ZkTX{xB8@|zqJpV{4$G!#ls0*b$!en!IzkP}Ct3v!zB
z+1<QLgG0+%(A0AjE~lYT(iKqrrF%3DD?qv&g)T_9v)SDynlNf0p6V~E7tv5CQ3)vi
zqIww(D?n5ng)WF{3TJn9nhTU%tNwDmfrdiKRY38V>zim;0dnOi1j*H>QEGPQioT|H
zc4yoy<%~()^E`!lYM%FM@OvTiypMo5-Sa$mJS=B?#?4YL64G<Or`X)@HNrnT=YIOu
zv3NtYbr?3y47IAbhbL;p^=sk=+pdB;T_SdpDLZ<$=lJNQV{nc*e1dYoMqC*x;0yii
zZ1G0igv~Xu_a~0neVuGKO;hhx2LDq8P0c1gM-C%v7eum<{6ogVzZ6Y@g_E@K`YnKv
zd*M~0B^F-)+jrqrYDDN{{zpuMaWem7_Jospe46gb{004R`CPx+de_N3KJV4h7U+(r
z@XM_Z`PI?R<<iC0t8$2?zE^``-xpJXYObmE|A@gMwD32nskL_Gpm~;|{)4fmsr>-e
zP2!V6{tyy*GJ$&_G^L8JoeO+Ry&I)GpNlzD>-mfulx;<EJ3WPT%IW<YgdO7a&Hxs=
zonA^wrN1+7P___HySm>nJN;XsicVMOljf8igj&Wc7@24px3MR*jQBL&mT?Q@HH#(t
zX!}m=<2iY&Q(OpiY$Z>1)8$^(?c@8#vg=L`boM6AX~wV0E?o^W?@~3&XwnttD}SFd
zs;k<Qh<IBSC0Ps9Z0XEAu!MROjB?h`iW|2kB;?LW-Uv-|Tj^J0Bh$n5;D&G%xZsBA
zXYgk!-_-5V_l8q&H%G=NSNEqh6ly9dp!iQE|AK}UFqPye_{<8moGKTUPotDWbq$QE
zb*RSmc5Z{=cB=~Ylv^DJxrDgYr=XqNt!5Z2>0FO8mJeN~IqhP9&s^+?(4b%!D?12v
zvA<>{ql^6wdqNkBPt)yUFXXzgGlmOV(5`giUC#*et_pU?j&=VuHNt`(J4dT?TVlT^
zjcNJ^<rxmRtZRO9NzxVi$r<~CUXJR8cr!tdu`VQQk=u;;E6^6S+|}*WTA9Ps15}f)
z+rTKSQwC-M@F~Q=JRjP*4NO|V(zYByp`JD?Gxi0B?<eFKY*u6kp=RY$Mktz<7qcfc
zEBG|sW@T%CxzT9F{d<b_DkRey{ly_TX9tH}CkEz%yapx{i;hjr8iGRfOf1x`G@YrB
zL$ljTOGLe*Y|@BOK&e0GZlhcYk)fn3v`Mp4Hfe!I6=G=yt?8eQpf&X~MB~PHil`eB
zOmc_d*F)3XA^5f4jZ$%)i)mA9agQ;~R<^pkZLh*VWqXUjl@QzeK4|B*y_8BzE8Ocx
zPTC&NV)l3tV6&-gusxO?$nj(b`@Iu0$Hw<g%z-n*hHIt0v7UV&sX%}+@@Ymq+T+i#
zC$z`-G~M?2vi{NH_|6y%@UDJ#tQ_9k2WNxDomk`Z@>=5z96Gke7iCh{WIm_S5IGca
z<6&PEkJSc;`@~sl=4FOXgMDxo0BNqXUr=`KyFxpqa&=&=|2F9gjrH7=u|6R`;Yf;=
zc!8rSkbm6X8wTC@35<F2SrPox1k2op{KwEVw;}%*Y-GA6kse8BinJeCww~uy6z+3g
zLPh&zwtglw?GEm^7tO)@HY*KBgyVi_*Ko%n=E7b5!1&HTDZW219QPBtg}b{HKdnO9
z?*V1*9~GgOgcEvUIH9IPGaT6Z+2FA@us<WPgZK3;@g?S!E^9NjQO?muSdisL+%b;A
zJ{9A96AVU3jMD@GxMQ4bKOiF<GhoXgB^}?)W%13U!k`AnH?jlif=qR2AKy$c!ZE&i
zBYVR52A`%ozPSLWmi=`og~#(_I|*xYuc7makp$gwG;^3ui#n;-8JsM(-bbSg7AW<-
zGCuDE>2+z%=iNzH7>e|yLXjSYK6eL3QF7{6Bgn9S1*-I~6hWU%u*e;1d>oqQcGn;F
zZj`d7E{095H8mI4E$ww1QH6BMh`s|@5n@E2f_82rno?Y8JI%#)6Y;dk>|rMJUVzfk
zlu&jMYBGP#$mE%S_QW&)>}k49=7oGBs77ZeHZynTA19#-cE?7uFGUST^vR45=Ir(I
zAU^@ZX+@XV9Y`~p{y~|Y1GaX}jxI^MLOZ%&%8nkIO%S4LP2IHljk!dw{lEfnd*ZV$
z!6UaNJqDWQwxp}Dku_-xocNp-j{E7|!hNBxX+kiDPn#=VkouK;R%&%+I31UBcKR&g
z57SVnC42$Je+mB@8dks(K1aa}VBPZ>Z3&+%n@^*Z+jfo5sdd}t;+kzh-F{nPo$}iY
zKqMi4`yJ5E?YGmGQaWyPam_+K?YZ~E#WkTa!IPc+ES)pO-hQ(45k@F_?vJu3^xXI~
z-Jbgj9`GlZXY|8eAN|P%8lAZDi}Gf4F$T*(q+_4HmZnoh4^D@^f^03VZ=1K2P5!2(
zO&w%!znXM~u6{|%)$4N{Ogl}+l<9{F7P%e$lh8D`qyGo*Mk(9tV%XH$-nB*oiqv|y
zAyy=$4DlC$F(HO{!Ar9YaT}$UR{2<OqS7XM2{X}O5-Q|0(fTHx6X6bn#7f{1_ldcZ
z_x;4z!bLgisQJ-?9HBvN)wn$rxRjmh&?5$j6Os)y%NThFiKal_ByFJS1qit}&`7kz
z)`itzK1|nU5j7%CE|k&=oynfT^bP&tnd}MuAwEsFKip#8=G4#cYwE-g<}DL<BG9S-
z+jcwLo8s)bcX+J7F;*OmH%42;9b?<ihHY!bot1d|b{J@_YO%1jUM^O4^hN66^&zlZ
z%^f>-M4O^ZU{_pUG*}yj>%QXt+Rl*}HqTw-?l09rUnF`;I;z4+Bmk|k5pK9@!nY@E
z=z}|~;F>6QrPM|k;&u3P&1kubw|%YYi*^<VcgNL|hW|^e@alGW6K@WyHefH_Xk2gb
zJ$g_&)M}P1jl%htz9cTz2Zzt%?KN6}B~`f1OY1R{BLGBeC@)c58y$B#+cKC1`Yb4s
zysT9fQ#zxb@Ia@OqyfwLRhg)l2|FVPW(T3gOOviJwCJ@%i)~w(PW`X;p!R7>(6ROi
zwJ-v;vlOE<bul%#+TXxFPj)L6Onsqrs+K51?&xkeGzFLU$$47E?2L^P7(6K`ucV<+
zS2_zQKG#$K5e<d9v{yhe5-`W-76uot<6`J}vMY@z?;x?&eRHns*cGUtE7{-X5~}-3
z_D4tp$Sc`%K>19kP#ogOx&h_uB&=YOnBcl67WY-~Pm)CB7fW9<qWd)M{aMFQd`9%|
zXeiW(<|^h8{w!unHA)s-?i={$8iyQA_k`^pgGZ9cdkEXXAnSg`<46#}g94#+k3DWJ
z-Av-{!JZKxE!{#x3DVNp7&>cex`PA~ECvCmdxCHqrK?HoeFR~K8Ez=A?ih-XxZg-a
zX)kUoWV${vu-9wyzMF<Zg-pV``q1(xG?ezVIP8agkom_*7`~eO;zj%Hg9f+>$BW0!
zy%QWv2a0PZL&#Fi96{)#B()<HM+Fq0HM#H8P^dMz9JKs94aI{N|CPVr(69nl{y3qg
zlyz-0_r}%v-nd;5aloX;n4g{F$bZOye*5;+%uJ>|+RxCIraG3U!10TkO&xFr6|s<t
zyabwh&e>McP$&Z=p!gV|4Kx%=#RL>%&bH29F1aRaXOU>?T9qqE>~OMnx=WhwK=E>t
z0MY^HfKuudip!l?@p_$v5iE)h;`JO1x`jmELktcES@HT_5=1b6!sB%(d))E*{Uq)l
z?D2}%KT1QPY!0WzA%s-C{y7pzuoyat*PViJN7COVvG);#PrUvwU54UTs%JWe;^W9>
zZX=y?ca&-g3FO(XR7aE8`zV!<AWop6v=@Ze^z2L;3KboRVdz84B{Y=wv^Xr4Jw3aM
zgyE}G-JG6r2AP4oPgLu}(Q8O*$?2JhH+`mOH_=e2=~<50`OP#GN;?D;|LNI%G^~K>
z8ArjCvaW63)3g24tI9mKH|v?_$=Bo@OMRiHR)0>DNu96!5oqc;wfZ6rg)$xjijS&4
zMnj?0LqIX6R{O$|J!jq9*S==&EBE#e&mNwGhtDjZz~2-X|CNNS3pv+p384ffe13qk
zG1tpUi;{!QGhP<#_yrUni?W!8LRpj?w5%YZyk-7AP!$0GNHZzG8K!0C2^-|7V~<Kx
zjwJz(1eg^Lz6)mMFrqNUWnbne?@Lm(ZxqGn&=gY^=1genX<;s-p->h^K=H9K12h!M
z!U!m)g@J;bObf$)Q(UZ*ki#tugA!z6swkTr7A6OqCrDz+!6=~kSeUodP$&zNgO>Yg
zC>|<f7?vJYEG9>t{*_XgkJ4}h3=Jo`3x?*rq{t1l50b`2ZJAZp);_~LSM)8JlYCa3
z`6kU7${>9entB?ff1sgI21!8iF-T9-P$=yZPz-}~N}xcH+r5stmzApf9f|HVSg~Jm
zf%>p|>6O7@kAPxkh}^57N77IzwaG!taWoW4Z8#K%sMq-qW6iI<fW#zNBk~{<<o<Fn
zdkaY->CJK|&JG$%mlS6=4JAl%?EK6)3By+<x|yHhVlX`)Ih)F1!S5icCAC99@riZb
zMMI%togB10OhcixLqIWN9j+K+#%*iKr)XFK@eN18l(MdE-uao6lA>v~IBV@a06)9K
zBbX5#CBcmRtb&>E&}^WBnQuW;&tT@CX(*I&5m0<Q>@R32l==uLMlf@<5BD=87qwEY
zm_2Jd6$0qGk}J)lKn1y?eE_SJ9NmlJ2a^Pl3lce?9MdTjM|5G96f2ROM8XOdi3zTI
zVsS4%o=*~yUo7rr!OJ^_;xmX#G!$wOa}{$4-<n0#Nf5y@4`1+ijt#e{Zy|B_5w(vH
z_R>(=3&Cqi@ZB^N$}0$t`OxxbG?ezVI26hbCO%EV@YSephMb&1W-vmAoH_jLcS&l=
zkW)bM2|53ThC+p$IcRx?hC*qFfZ`u=&b*3p{~@`1j)Ey=UE91v&UtdPCXZ)E3<1qt
zw@yZo>`^(-!)Ssi5gi0gA?Ppyt~tBmRQp{G2}^!d@Yd4ss9=@D3*wYdqtwX?uKl9t
zovdyATScTh^c}Gd>*Lc@G3`8(X}Y%L3Oap|2#R~oM$Ir626QY<ojw<M3MUG<*gdre
zSs3>KxwkF553o?Erw*{V6l$wO4zL&n7P?Oo%Gfe353Vro0df+NKCq&fo#=2GV48Uv
zpmt!zw))I{_PMACSIQ2<X%&$P0`87wThOCcgIyClz^A+eQLP>gTvi;9wzjG<|MZgb
zF4&$wP=E_+<K*~@Qgn5zSZ$V@;}P6iTZ-XY4A>VRmwI-NM+2$4@qxZ*&u|%lmExfy
zTql9|+7{JixAnN$!W%a5Qrq~NVg(L$hmG_RJM|oz#?`1=i};CS4BlA1c1^hym->3}
z65Q5kR2z!m<N)zXBdWpo^*!ZAT<AG%Yk4dZ7kF%XMYG<DUwL}Z)*8^wuF!?fpj$Wx
zzb_iEwW4Yqm)M=V!*IBK0wpR&19-k1T>M*Z4sd+fBb*gbzK(GZn%1N316%7w;6Z(0
z$Hr(+Ec@!I$H}|=(q8N<DmJ6I2rQ|_d!l-)%Ge9Yu>VyV|A2lt1|iy0ZVn?;n1Jxu
zfqG*g0>j-JjjM3bNTa6X-P3689IO<9XBskiP)Y5YxLyaAVPBkLc;{6D<HBG)hEw$W
z6By=MOanV2IL9CwY&Dv-(Vpa`^~vccz~%M8=F@vNgGb?etb5^9zGM7!PxEotTjq1*
z!Gkm_K2K_C;<Nx4I3z;r;3lhsY|F;ZQ+oh<hw;Rqwsr!-vzYp8z0v6OsDUC8nkR>j
z*2$e+LrI|8*lDvx3{i>29R69u;r@<rSm!_XHJCK>mN~41#Gl+=@`;W}`RsGu5_vOt
zp>DGzXZf9sXvegUzSa>-56Lu$8VFh$<|n38B05)CCasHq><G)PlVOxK>}UcP>gHS0
zs4Qti>*(hlk$JR_Z_uF8XnV<J<_>D2r3aHPlY7{XZ2#nq=UQ)1CV9Wc)Aea69MBP+
zg`7TAO(|><WzUfGqrZ-2f#SH(C7Ao&t~t_e3+IB!#PK;%?|@3Zui_4spSrYahvV2K
z2<6tX2%uIiR@SZE*b~9O%_@>>4ntf6Q69uW5adC?=ngvZ*r!5tMHyl?5feh#g`GiE
z0+TfsM~TCd#P9tQVHTx;(}$o0oOS!R>&0<=)wA<Pc^C8#ElTki9PPxSJBULuz!SwP
zg`QM=Yw$30$7LJV@m?7~#<%1!D;AE*Bc~X}qs4164vdECwNV+)u|uIF9z{a{3o$tB
zVn77a@c`1!qrMc=vssuTb6`iIN0uoA$xv}?kOYTiY2p3x8#V~_*b2B*TcbPUIt$bB
zh%FZJ;v}F6$59n}lG;S*6&BlwUI$i5UaT4r-Cr>b0X*~#d}4|@TD~rhc40gV@aiyy
z*uwzh*<g)AbT&+cb^>p-e%M>^S@3lgKxb2d#&E4wDMe%D!QIF{qXkY$1T-lzn6;Ek
zYTD9g^vQjs)&fp0V{jWW1m?wL;!@}lN+uY=UXdyeaNNg-4Zt`}jY-CI*%xqz&30G}
z?;MV4od85(h%x9gpk>cItYl&qLjxHD!-G%m7)dyPfx9azmAF)h;JidOD=XIFhpPA%
z0x)ybGpVtY-g?A(sWF&R28pf3?l2U@!w8|ytp}Zp*x_nva@jd@g&v*|1Mq4DZ?2BJ
zft5-y{bM#T+8vMYsntt~--X#JOu%3i7ArM2mTPiuEhkxxW~(&Lr`BMb3P5{Y0MuZD
zeJ=2Lv{sLit>r2^JC5s9BaXSHJq`zUsWynyKmKEF2;Yw)d1RiL%psf|y~EAsSYzY*
z^@7VVFiJ37po>PKTw5Pk*MnSOrn$Zz55@JkIvB6t6@&4G>0n8=!JZ8C3&YLP%36f$
zMk(4Z=~5UVOhz(?TLBP_=6d$t`r=r*pZ&Z(zBV3&GxBOxgm4u5@i<@(4T{@r1*kYS
zR^Z<iz=o~o9al?ZHLzF=#8_M}v*~kVJ!cg^cz5Kjnv3Ln2YN^D(Z1VO&`zDBwNz=4
zIAhS(*si%uX&t)mF57qP;VLoQ>K%Ev#>I(qr6cc*g}iHam5S<5i`GuDqBV2MT*w#m
z$f$9Hij<zZ;T|!f@Px{zZg__)TG9|Y!$Qa_Gr_39%*4z-Jui_W?pr#8dz3Sfh|dFD
z(OhW=y*L*_j{*dUC>26$^%L}Lgvc0m<Xx<<6?nAc$t+Ag5vd&ciAIFjyR%pB<(c@`
z|F1?2sP%>2BUlr9#S*$XEP8!oZ>d+u_{HiP0!<46uVKKj#$hmr2mM-qG(oG|62ztl
zlA5F1Q3Ee)pl|7%l%DNce>5rWWh3mdq^7NG`1=|HcXdS2;nbW-gJP3O+KS{T>*p~;
zQARzWRpx1BT(Au{E+mkVyHy)BNE;B2;ru6lnVJq6{wwA2<vBZq=by|OUsq;(s7ki`
z?hYS#*I?(CIr`JI<+GMfng(mN=Et)oAV<Ve%6(_PtbzWL1)A17v6jH_b$IL)86h6~
z@mvTAkF5|gJhqJxJwP;%eJUG>pF}EYkNu$Lz@8(T4(fQ-WMZu;*MC~FH1)bi<!+?%
z&Xd_3t&`tbop@;|gNzz_Xiuh*&|uJruFG<;sZpJH+1?B<n`)sUvH0ZyIp&1WnL7$v
zUp7Ic<|qDVBUAfdAnRx@YyJn@@S#Ei(ftn&FPHzhG-u|mZPv|Y+za`DBt~3I3u>G`
z#gdw-<Fhe=DAn8Aqko3gl;*jxy$qhi_NLGXvAr+Og^-*rNraX;a-#&JXl>Dq@o6}H
zLMXH`Ug@U}U;sVdRBIEJJX}e}(k6`M5Nlel+E~IiCt@3#J6bG<+;NPhZ8IT}6P$CN
zf_(GT<eFUu;Q66!Nld{4tp`oT-(YEqZ3|f1Kd}YyL2LSEt0}EB!WJ-i4qK2yBg7WG
zHy1*}79<EQGHroIXbL%_`emW66_QnpAP*&1L?8#Hb@tI*NT#@E1dBFe^a#}K)9Gxa
z--4o`?bDb4z(y4q(vyQo(j@q(B|&q@<Wz_6S)F+eB8CADBD*lEJ!;8nnxU!UTT@e?
z?g+L0)5~>}#lzoqgy=D;H9%O0iBhYExscDn<;DF0A#|buy(#1~7BC6)Nvn|r2WK#o
z@eNcwT1U0u0NZdgLjutQZ8S}e;9%kxa@mxxMTafHvT#fGGkMFe-CBC(&f8q3Ia{LN
z0ULO~U<>CbooV`UilvdJeVdJy0MdF`o663xn$nsgoEw9O)@(M))EkqPVvd`=057pT
zl22_jU(`UpxFZVog~cYVvlJ&oqME^67|9uq<mpT!V6YLQN2L}pY{m%<c1i~f4?PD*
z7z~8x8ro{A_o_VVky!Jym^@#B@}$S8X3eq9iP(nL8*QqGdgHR@uL!Ry(K3OlvNo@Z
z5|>#$H5eadOp`7&6L60u9@Chv((>dPB{kB9X&dpKYus3XqXB-e1(+7IFvSd8$LuUc
zM2H>xXfA|=9a9JycFabI)<?~box}A0E~JvSW1G99Wnfym(32YNkLF=wGTL8=u^UoN
ztARG!u?^=G&0TG@BX=F6ecMxz*zqjm+B~@>T&eOccbycJP)&W=mK&9>muzK$;p@(w
zmTOtdlE?UoJu#y5<olW_`<bQmrVU+U*9A(8xkxQ0vU)+S!~bh_NDm5O8yOZ3+n7Qw
z#5T?!$TJBPwoxHu*hU*6dh}|x@myvlo<=Giwo%)^IC0CfV~{3Fb6hTbG(|Yt;=8F?
z2a0`dMu}yT4P7Sc%=EHOt?NFkYcCyR_&5_gG@^?JjEYq`02b=sYN(yw5o#xB1+kvi
zRD*33LTYE?z|Lc;xe}RU?3kFiddjUdXKqpeH?oZ}bM6i}*n0WIayK>_`Bz%3AD)l<
z&3T3I-aC?RC$5~l`e3OZlL<maEQ1!s+2a6rhPUeFy(2NSVVjbQtnFbVjA=M3W~?Px
z>8D8-GEKT1JIuy}!M;t^9$_1yeRr==8((EH@_pKO6Re6mg3@NA61>|!+*ZO{8KZ;}
zI!n=Y1T-GLp0zuCcwc!p_{dK9BB|{>Tk2p372;Cay3&`@s@p8Yvt_BY>dnwb(<(-L
zTB~f1_-WOPm{#449cE*~VBZU>B1&wr&1<pEh@nAOxHYId;n{FvVn!YT*ivbZpkZn%
z^C8wMQ)QknRVHt(ztofP4wQ0y!a^}yBuY6x$poQH4x=)y95(HK%5f1>j)$?sY)mZW
zz&2btvVn}8aU>jZaVGp>t>12`7B^VgtyvpwzK1TBM}Kr1ZN7&QBcJclZSGXY6bm`~
zb};tR+|Ud#w$V0CrgwyxHKFgn3=ghoZ0y-SU~XCzhsv;{AU4AeD5mv#3{%`$Yc=uP
zov;Unpa1Z}WM4|uXpN2*>*LWN->f!NYgJ3ozGnco!t{$D;S`7_WTAGpntW&1WyNv1
z;!x<Z{-M%a6Q+3*cR3=|7fQo_@2%lX%36$nG1J9gqb@pPd}SOa-t_F^9Oo2-STpp>
z?8QaP78wik2W7}!GWI~HYG|hJAgg;?U&S<$;e?jslJ4r5h8Ioteug$mVi#yCs&%!t
zBN7iIbX@~t<4f9#B<8<_sp>-HCe3)wh+!M9-Xsux!4ZeV|Mf7>ni0b`nnBtMHaH2J
z4qLgBol!Fw<90*X^b4D8$HpOsidL?a;vnDQ;~c1e2~gi4%%ph=)Jp&fXe7}RVMcM^
z)=ruiFPnqy<mp8+7Bn06PRtw|-#amfm!|HG_45NpD$v8o2p~H$%AR(!r)u(4gQwSz
zjN#v_@oydfHt=r~|F+m)W~=st2MjL`8+kQ44SpWfa6z(H>_9q_T_v;<j)y!7{#y<I
zt%3j6?$|pq7d8;q%W-2bOvy&(LvItCDz#m`Yop%EC~W%MQy43c#TD4cUV#0Kl}4dn
z9BTFpOHjBBK5XT?+{O7rYxho^kgs6@!0jELb4hb=5eOPS4n%^K=rPkGz{@14cVwCN
z-NaE0XZWwNF;ZD9zGrc5blm9?+ZoGk;>fXjyL}Nj%mlzx;<NG@H-S|2z|9;~y(7<q
zenw7#Kc;Ng&xWrzz#k@^b?`ILb3@V<ZWb$nij9oIO~{g@iREt85o7a~Wa`ac^JKhP
ztW!CF_y<MZ+Y^+y!kE;Nhd$m3O;bnz^3#%-!_oKC-A_ZI4(k+9d=BgUC=G==j!{4{
zP83{XYY<l)$C-NkHe1EiFOi5C!)`w8Q!dp;;ed#}BOe1*n3xM+G{?u{y(3#eX)w!?
z+`-Ne3cE1pb1qrAk7<5_M4mh(mP2tc$SQaEF$p4=KjCNSI@#ks$oTgp?jG#%x-@s;
zPRi(zr=(^IA$4i)p(K!CF?4ciu2T?TmAPe}p)W=hNbEfXVLECLQ)f2N`C|qY;&S^p
z<CrrIs=HC@08p2sFb)9KZYN0J4`?4Sd>P3p9VO}z&u$(td<kGg&0|e|ieoxmOqtqa
zIu~nk4u&i4Lpc?qsY5wm0Xh+KDCf1nF!!OH7_G=7IG0)orVrdKvDv{Zg$)DmeI+;<
z8GeYotTvn7b5RY>(q!=i9Pm-a3H~+3db!quBPsa_G;kDUb8r~XjA8K~JQR9fB2R5B
z!ZcqysWCa6lAR(_u8zcmO>qb#9AN`TRF-S?f_3ELEQWwKCA!s8Wpm+a4o6j*3p>{N
zW2<rY&}1>2n3pn$I4^}mIj52XDw?oq&Nf3VZoWgzUv@D;-y|ru4$PX^%tUIO48vn&
zc?Qu|Udh{HhhEj-$RQA_^$&UMIAX+1D6uI-7939)R+{I~>>fpUwtQj^e@9cB4_cBp
z)n<QfC0y`9s!%!xRC4)*)v1>{F-%aM%mPCjK~lzgVrqnm`629-PE75e=C>r$^^+p=
zu{>lZ>bINW^)1AU=8Kkvz&4x~ntm;niu6C=0Y}S1V4L@Qw9!(j*hU*pla#sm4Pcc!
zb0N_ZBmb4YnG5L#LNgcp4}w#|%*8VHgqaI`n(oZS`Tb+{cnpr|f|EhVTFp+97@Kk?
zF-*uE=QA!e`|E`R*@_#Zmsa8Mo9)dSoOfN=8dn>&dSAp!buh(|^&>ZrX4+$s*}XIf
z_Sec7^nswelnm)jy25104XI?v5j3%gteF%!n77aw1NU43X9VrW^nMn5SyIeDC_2wd
zkjtGGIUSmMrbRBLp-^cN0mUaR@^Tsql@<|D48HB3#a_p}f1&=a=8wlnlze9-sRW4$
zF6aV*3YVS_UvDCbAm``;iqD;#Z>FJ8I+BBy`)DYXj&LXr;qS`^o*GgQksyL)j!-%f
zwHaxUSpYs7%mzB%=B!m@0X72JepB647Qkg8jV!>yw5@l@#2g_?INzNd;h5lpr1ZDQ
zKd+?OPe>xj0V$yP49H*8P^bZ!gO+(E%F%kz;t;+yAeWOMbj7C9LMbC0J|H=Je8-o_
z3b_W!)MkYa)srq(K;=#lDV$RYqN6~&LJ~x$fiT?(B0HFqxuV0f(N3q0ZeY{7)k5>z
zrj14}<>5DEiQo*~0TyTtK#B(rfQ8x)xRQjWi^@<hQp1cA-XGG(PBDh>icu2-TDGOJ
z2$zX8;M~@zQig+7OR)IC&U}PA5_aq|zSSu3$PU(<_A_8=e&<D&nwxX~IUKvJIPfJ_
zV|p$tQbr6S&hf99ETWa+wyDb=J$iEJMkdbl1#KYCNt(^>1_-%lvl1;an|+J#Y*uP#
z=xp}2fGnTQ-pHPCHj7WwJ)6D2v?+~Frn9GaG@XT2<PO*82W?t#f^MwVXT>ypcB|H8
zGv5=FJ!5@;u?1I-_tf`lWZff%htwx?=;vDZzbokqr@D8grn<-GU<0CV&U;s<o9h!V
zJ_qKjXjW~&9Ntp0Lw;0T_+Wy1?uqb2(A0Ax{8<_b6;lZ)J`=WY&`_vJj(}oJgjXnC
z%;dHs3bTs^{)xmYI1J-pf<o>mT!M2K3p_(2PcFi8C=Lc$#R4;9DiRLnPk6C_lRfTY
zf#oFb9_;a&_Z&?_q4YXa2<FU9*@F{ESi#aT!F8gwK8E02k_d8g<rMz)(tggkUJE?8
z(oiU)CZPDxvXh1qBz$Y4I7Wg9mU;L@(J2IXmB>vb?j9r145-Qc5O)%01D!k0{Llx`
z*qXtGyHP4X<O(Z{{LoR#L$!&P9A?<|Q}>gc)KSirF6h8dy%#V_1sHBW^+^(WQVlp1
z2ZJm>^%W9CFn_}Rl#@MfKlP6!?jG#%@>4&dp-`qOQwXNxP=4yyB&=X*nBao^ln{%L
zZC^A*g+8R8athz=rw*r~P<|>0E$e6~LBhBE)bmLY!7>l`Q%)hc{nSMy?j9r1*X@Y3
zsmpgvZQAs(?EGT`HcVh$^x)p%kF#2=)8@cHfCFf(*@mKd<PP{LByU;)zPt0Lsnmc>
zoUR}cPiIc=VyU*b2*lfFPHz-h%m@yXhROvfyR*`!u=8%zN}LYXs&MIA6Kn9~uRJ>%
zl_60&9FOCTWz`yFS1atkEcje4^$f!w*w3*m8mmDvm0kWE!PQyKSfy0wNo2I3_7#?%
zn%UHO5^Wg}rxOY{S{-CZK3WkTgNhl|#&qyQc@>*_rGWN+8a>;3H&cOY1>KXEZM_>H
z<j%HAv_!V`A>VAP)M%L<Gda<I*V6|9S)OhE5PQOGD?Uwkw)Jv(Za=%Ly&uL_GrqRj
zKU%C7cg6KiQm-${nR=!0=r|911l2njG2Gv7)(Ra?HId_;en}0auZhth)yK2~xst12
zO1i@2>LaP->P$h;aMs>8b6!?~T{tjaUENNtgY4D)49R@Aw^S6QRvG^qcoyO<f6m70
z<#T{RpP5*op|^K}7QA$w>$EoPE<<o}ald<@?PcbFe+G)*to8^qqMrx&to*F4A3bSv
zG#pppV#H`q9gbAs*QxHR*IHxXJa_R6R>jzc8)$KCi?u{pN(^hPl>2`KmTu+N<51m8
ze3%jVTSed`t;y}TE5J)gw1h0a;Oix%yd)>`dWoX}S?(oPvnTWt_%z*KVq3A*tPK_$
z&He_QY|!kcpSU=upP(`6*i)QC_uo5M92}1O<0`v!wX`wXA}(g!zOz=V?1(l+TkEaZ
z=}(TJJC+`$-D_&toenC}rP|buNmuAozMS$YNBfYCB$z98E3I~nORzl0wj~Q*3D;|^
zH0@`!lAN)=RgBo0V4r)<?ow!)N~yW`^zNjgP-z1J#U}+aMnj>}1_Fw~$psq2Ij0?o
zy;<xfZO-#r5(nL$=gWK|n4lcetGJka7<ea%JeflhP<)Ew-b+KFlpzN#AEcpB%D|zx
zgpEf6sKN7D5{9p^ySZ7Cqi(uRavdVa@Oqr2medXbW$Auu1L4GsP0~2Qlj7;W(NI=E
z7e$@3Ec9CfUqV2XgO>lIp-|c(plr}-nZ-PpB!_=c6wMl@e9g#dHboQ58m+Oh8bk^h
zFI?S%oJ(Ubp4_osC)AX(&J%NzB8rPI0+IuZi(M{mYUek`WS~tj;GW?q8d5WylYuiK
zGn`{Vu<jX7yFf(Far*Q*4oy}%Tli%*2{=V4a@$$XAE#$I!$sIr3rRjuFUaymF!!m~
z#_DDG99i?72+A>|0g2!MT!>PxnB|$U3A7Y9%JsO^Q>^d8gBcs_vZt}QiY1(4KWwAi
zY!&g2d*M;>>~4Gdl@l?|dTg}DnwfI!Q6$L0I!_X%b#j)~iJ5TF2O2XzqL7n8#<b%(
zUEz*0A2vzZ!(N{uehrmF5Hfjr#47<p?mXft34bwS{=V-FQR;f=4Do7!%V&sn_JlJ;
ze46eV;(2&B66DwWp@ey?lf>a^IVXe$;Ev~iJB{vou~>T}BvIJj!)mo~ajn#<#C@=x
zpDkDDr!Hky9N;cRy<(1=4Kv%gp8-~mZ0lP!p#CUPKi8BeS3dF9q$||qr&4pjv%A0p
zL2|=5{K?&~c-5H3%*IzGSHvg9jdvv^xVS;aj_-h`o)f@_X(&|8FQE9$Up_@cp`tng
z#o*jxOLI97(`Trp-c#KDCW(>mN^(sHQxJn@hF?YbncmGq;Moqy8F#)7p_r+SNi0JC
zCCPD;yq|GQ%{i<)p)sRO#`Kgv3n)JNyqJbU>9c@h=yR8fkXMpe1?zMN#a$}^qm&_X
z7k8aZB2U&%a43$nfK}XeHVGn_KjFn)PWHHqyRIN{_h64#aaT-3q4YXa2xhgFvcL@z
zR<JZoa6#r*h{eYc+)5Hbre2)FcNcfPg@!^IwH&m(hlUa)e5<(Y5fVhO%)^VjoI-FH
zcYU73-D3or?%BiPnGJOA5dPe!QK|^p6)qb^$mXcRll;=*midwz=|3Vlr_+@y)-u5b
zjr1P^W`3zRmQWB?cP?g4ZPM)!Jv(L2vD`T}g>ov#_F9OoLUL@2N3wEk*8EPU)QnO=
z1L<@I?J1Tdxk&_!Z8K=4OXE7^!A9A>V{v{RRKdk#?5wvUoIej|+p*ooe0H08vYpRq
zc2HL^HNe#1IRf~cy&$R&9zDzSL&m_vL{nhk?Qop5JRxw~3uo-J56(T|@VI35ub&#n
ze`Gtm89NI-oO8xebD4<&y&utzogm1MzlFWzSj-q9gaTYN7e(wyGi_V=o-$lb0Nco0
z)v@B>ZrC?2YA1Kbtjuu;zR-7^7ov-6a906#*N9@ctpJWAglfqI9$Zqu>PCiRge5be
za2#TBXIyCv6r!_-<H6n0c&%0Ey{u_i7uc{2YY}iN4i3g+Y$LnuMAVu%&d=*L576WJ
z7Mt0OZi~?(OYm+@SQ7{EcT7p~7eO26ieI%P=_!`Nn3>iEn#5)5VSb_nS~q7{-I%jc
zOYdZ^mBDr#jAx^i?W35P)oyI~H>SoXVyE=Mc+0Z52|RNCrUuxRc?eC8#`_V&><Yw;
z=8SeU9=74s&>YZ?#zPKl;2gL|(~E5jVG!pXmyAhm$0U1cD#oK6M%^5X);W{wb22Cy
zR`X;-Z9rUW>4`b6R%^#kX4^(N;=J~BDHXlJ0!?e2SodJ~I>w?+2Ca$OSo|^5_$JrH
zpJ<vW?j%UAR+@O?4`0o|s@(bPq*7DZyDhCT6?UF<fl?>uaF?ic@E)rJFSTP(QSHu5
z`_`v!cv~Ltss?~f3aPY-jeeSG>|KZtjk;!|u?^R95{Q0@3XTkJqv<x<@ug_j`#f;d
zjxWVFT2R;SCaA6aHl>X{1NuVEz(x$Y#0fon%FSUFHM1&SenK)F^NfMB`^;d=&dwEp
zq$(w!l>{HprAwOG`6(0PcaTwZ2yj{_RC3F2j*vxWhmtsR+Nmh~SuP5Nk$xHp19MWy
zClzg!vM%%q0;z(@*!US^>%Sp%@3DbxxF|><`q)76ecJ<$Ha4)0rstDXM19EafF*K?
zmMEg0G0XFUAvu{1T_WEA$nquf<Jl7~k>k^JFOi?C?*Qmzh5Q9MSIBj^9WRg{ulL61
zYjQcfFRJ5(di`)a9$OyYbZ)WIa92Yw)O$(nh4H1b7Z-vY^>J$s!(6M~=O$g@YWFNq
zO7lvhqi~u?QzT5Fv6`(dQJdEhX+MDRO08WFBnahRyS@sV`lsYX3EehqWG-dOU5hMP
zX(F%x0@%xsN;KDoqu=Zn`UkkQPRPhuJX~bHDAmsXT2XgfICT@8&pzd%@1&tn<)Q+L
zQOG$PN1Y^2d{Q($NJ9%K<>W|Y`#EO{p6j7cqf`psHHN1)t9-DYRW^*4JE^QNP9>E$
zgE&Hx%HM`|?!ryew91^aacZZIayq3vYZe4NXA3oHn^K-QzqSYSh!_z=Ff2NUbmTbf
zhJmbTTrWgtjYA>jP_b2M_F;+;t!zX13#~c7a%8Anhg4-;iAQ0lc2*jiWg7=ujb?2W
z5})Ea6{+>lE~M1pdx{)<R;n<GK<6@nd|wG<lA^qy1BBd#kP<Bs<sIr9<w^Ytjq>JJ
zWRy3LJz<oGPtzUcZ7DZki|1gm0!J|7<(~c5#r0Yz;odnp!#yVg9mjn~I=cnB;F?L?
z+2hzleW0_A9Hbe)s?o9%B-^F9Z$;7-#(gVNabMW3X;-(#YB1khm9YwJZ@k)+AnA$)
zb?cC8^_PPQTJ=2@F!B#E#j3B1DO2m8SL*)Rt%Ghytx!)n>Vtt5A&z<w+PNKdhW<;B
zeN<Z_(yn_2^Av{)HEQd+Z@FTaZw&`e%8pBE4uks^Rx5(XCaiCuDT{jJ!mdIe%XhJz
zo1?G?7Th`>w<1@O%WMI;1P531JIzbcP_Z!#=U1~fjTjDWDOP$!yT)4R0p1so_yH_1
zyLh&sqYO}8a8f!cbK<EFNPR^)VnV1XAxx6Hz5yWQcGnUu;jT~gb=Oj-Bqz+kz5V$o
zZv|v|Qsyr9gzg%jo(=ALPpw`lO{cp~BhVFho$j{1yH2;!-L>OaHCi4KDlc`!94)!h
zG=G|Oh3@)!U2@m1Zo}R6*Apb4E$;fufKhID?PALGx@(7)x!tuwJ>{<N2Udi*>t8~<
zY<KOzEX!S|UroE~=P^(50ii~1-Sz+RaM!ykwViOXneJbs-aW%H%&S8@bsvujAQq@q
zE923w8lJjVV^v23Y`^dT?y6)lfpF$))p2TXd{&{WEd$LQHK!HkTGkv+n6!HKPUs%1
zv1#P2kU$puFo?L)yT5~9%7EI4nNHT~JIM~5%9QATloCx+xUmXo$Q^D-v_!aZk#D#m
z$AokxCMVj58>a%YJlxpGo-o|Nr|Awi&S?~{iTg48)k#X=)SM{+1-0YA;zHF|FJI^B
zFO}=sNjNYCU<cu_b0+#CxRJ6M*Q<~Xfs9CHBW4QV$Lxin9@V`Zg8El#u)G8W+oj~d
zmZU3;Hnyaqjng|L6bb1y*IZWUT>{J0sW^B*2T!ILs~0`0MZy{FE5(^|LXImgA*<<Q
zXbOJFXZzS58Va@3LO}7^HF7fzh1zK$pqM;VoaRDlzZL4OBnG-s&Xtr2-wJgHmyr+m
z9w2ci_uL35rY5-e-26EWh0=r^w0x0<LTLhr;t=&L>oA<hNEp7d?&e`Q9Cg$E`HWy6
z?f4l<EvX%1Z27Ei{*H!1t#0P96AP-8H}Iil!+xsne}XFprfrvt56ZqCLc<DJLFI&+
zQr5N2xq?~*#)7Rjst^vR{H<_9hCy6pHjs|R{#p@#0!{qQ3)!@5HS7yp1OKhvfyY{2
z3tvKvT7agW&gD!R3T09R6rTZb2@QoZDFTYY%Urqs0TQ;(!CY>ogIxbrC?0b@Lz<c#
zoW72vmK>e}ijS#z9SwytH92UxlZN7<HHN8aQ?Bt|8d`v{;iLu`8=pq0at&8>J+=9z
z<@!pBHotHul@z|Iq|$|8azm0zUj-q$lS=yBMrM+X>Jo)+I(@W-&HOJC{<UrTXygjE
zUySvK3p=9Dtx;n;x}rfiJgXj8YQ++}m9*7hd7goxdTo?lB#5b_xYWCLpwNSzU@{1M
zVz2P-QY$*Wu%xpgFH9{>^SGt?W}0T113T%xmZlKbx_!#()--s!;bl9lMDa1h1rs=H
zd^w;zrcFooAkr6+E@QI&JAo_6c9OC>zXk}ovpNzjk=2QPvpP~$L$f+dYci{|KYPNg
z4n9qHR_E+~9Hvn8+DTI9`8kt13DAynI>#nmwN2?b&o-W)w3kCr`$mm~V?k<NO4+PQ
zy26yra4KaJsnm!&H-(s*(eotcWHu-G=FZ7%f~KBf^A;Kk6*dbfK4J3?8VVHv2`EO`
z+><{iQzJ3ZZFsKWIXow`o6E?DdpDA}lcBJH;**oPorXebLJnH)rJ+!oz@d;inYYZ`
z2X^klzgO<<1!wrDBrsp0cQZM|X)ygcIb+H($i7HYOR9;0;xl7<jD|wZm~zna0~!jY
zCIX5vV<M9?|3SkFm@jb@OeyQy=1k6<3TXJP#HZ#<F4s1-^Q8mz`I2(&?&*?3H#J>a
z1xyH;E*(6UHC@ue3OQRk$cb!vvNX&B>Z65rI44V)7v`{uz{}XxEUJUh5OxD2Z4~+%
zIi5XX2#ZhC9m1Z0a_%I2Jt-u7?KpIOKF7OVwAaTkdxNFQ!R98G^>@axv#~Q7!-`KT
ze!UPx)}{FM+@vdvUw5bC*W+_jglJFQD&s&xLGFn5Dro8%(N<|FR2(Cq_{3z_(@-b}
zC7^T~(Y}eqKzBD&6w$tc%gBd&?<R34LlptVC!+l`8VaQeIcWJb4TaJK4uy<pc_8~W
z5{9p=yBWxG)Xflw3}kbRt$!h@CAC99@d;#~p`lQLYz|sxUQIawA6ona*(EfrfIyZL
zYD!txHfJEa4Hyfy+E`^=j2EXWUn)G0Z3E#F27E{QQ<&%{Un!@$mZp<hP+Sd70XHB0
zIE98n>4$*gFS!@eumU8<Q3#TpPovZn#1)-P?G$8bDyq-8BUz!Entp5s)eV_`R6u_2
z)fy2I%Bjb)6q4y_$8I+5I9ph!;Aw~KAavSs6C(+y9XGQloOa;TbWc0Zf&IJnxPJ&X
z4|g*6$XySaK<#(}vQVc$Si)mrf+x?AoB*>YNz^Q1@|CiL?<DO;uKg2lPrAbCMKv|O
z2;1)H>UL`Fg1MBd80*eWD11}K;Vr<S5aaMyZ2UN{gw^FNErhw0o1mLE6IEs=-YTRP
zY$jv}p=RQ7MjV=nzh_TqCh%#x&BVq14Y*7!F7?CZn}fT@U>icSBYSaH?p#F*e8+}k
zw~Hn<>&GlbU!(|TEAd0+Yd2kb*$dCTU|av$=fCjm7hSsLf=jmcpL4-weXwQ^b>+?S
zXe>_MVf(T#g8i7=Sv|JPE-iDwmj=rDo3d2T2t}28X0D;<vQ$4$x<X5Jb;?p*&^49_
zCYYhD)w_7d3}w|jl0SMawqi?9otz#}q3j%J>KV!&OhchUSpmgAls$%q6%fjD6nvsm
zmUR_}7R}_`F(kLfb7Z(Ax-`{R3fA$9lJ6NbHz@hOfK%WzRk@gkLdjP^G30xo(_Tn=
zGk#SxZ>O;d&;X8P7c^k8W%68-o}auUDb~JGR5ocUDanrcOZK%i6iTuJiXquHm4mm@
z&;lgKkqDBUPovZ{)a4tD?Yi?V2j@z@u~)S6uZKyl=;@GLiPl*NCg_Zf2e<;6IZ*e_
zKYvXkPhPCSp*R?1-Td<?2_l$3;iVN$_PFmI`5}qB2YY<9l({8w(1}Ywr*=AYq?v(8
z2dVCPkwQE*FM1l{v5<Mu>_*nS$Qet@36gPAiz6US&y=oaGo^nQQDg8-Np>J3SX1)t
zH<+$q<l;=}2=;_CC48FxnNq!mX;u~DOzli54Zh=<(x(1&r@e_|=kpynoCXgaWt&m?
z0xz2YX?rP+oXj#uk1{}T{-#2OQ$XHbS_D2Z=?dpQ&D7k-p46p?6d{ih7<Vx&x;Q~F
z_Zsg7(A0B6vz>-Qg+&5Np0KD)LkkFtI1;7>(H*68UFC(GKc?g?6?O%*`pi@8Pt(6r
zdiq+LPAa5*701nIvT_>@g_50sV#tmx?RqB-D?oA_g&@iKG)e`ut^scZv>N^COT+k&
zO2VHa`Jrnd70PyV9r@2uam;Lxj)yrCL=@#>%5+9iX<w7E^rR3^MNtJ{L`W3%18C>Y
zE<3%FJBmsZkdC67EQ;D76eu`~k{yIbQNLm2VifgT_JmOsK23KNbyj>$vC@Lo>P{A=
zPRW^MmM}Yxm+qB4^|FWlQn6WV#7%7*1S~|rb>^*RxzZ@GUkm5(SJb&3<y}>_LbW>{
zhfB`4j5XmTzhY&3GTl}$j>7J+@*sOr<fk`6rClA4_*lbWufgL%&w-*lmWJ+&V4bhn
zbY`RHNt|UEroK=X@4#l)Vy~r1R~UQkNyT2TnRec}g1eJUtg-~z{xp$=dPkDy=9h}^
z$0p3m9S5#~reI&p?1uY7>QiYbRE3v-;!~}59u0-6@Dfmr#epN+h=2>xk#{*btD%;x
zdKrnU?pY}hzZF)a@ppc+;NFRB&_M~TmvRyNFn^RJh%BHIP)tQ~A4u{F8VaRIIcRx3
z4TVx94#lAqc8K>@5{9n=bTh=`sGHF*St^)gXg)$xOKOLJ;xlFW3=M^vvgDxU>ogQf
zI|LN}iP7KFumUDV90gO#y0)2tyjZ0>3eW&kZ%k>AHUfzxerWHX7HAEndosRL-su^d
zd}^lqbB?lURNUTaW{ZkF$mFts;$t9}&`>C~5l{^8w8Z9kraR&b7>cQ@NJMm9$mM!E
zSSL9GFiJ%d?sbyqk;s#>;!qr>*;*$#hXfJKpYU}OCwtuMB-=>bJ=iniqsl`x6iSty
zw77kJlZHb1dI80Ume<ly+SB4tDBIWHM#AvbsBZdt&LGnZkiI^Lq5A+yE$Qn86dzyz
zaT*Hc>vPcZ1sV#a9RiBKum4*bR)DYPD40^#wawetAK4}sz<rN-!PwGN8|myFzbNnY
z?=%-Em;WPZnsOs<m;W0Y3T0aa6d&6%{~9u|ptMIoF<d@+kc<mQEFX3l30c>HTs}VB
zhaCh6*}JAmgOh{JCzHgI21h{gF*s+_P$+|wgO(T5P&`z|FgSB~u1=0T`B4qGei~kY
z(c#1f86BTSsg)O3y3bg7S>jGoy8JeKQ%lmkieygkD4K;}f)X;<0Y)hk=Z=)$Od?N)
zf*gv2L6()dj|36SpKvSVWKW;e{uzCmY(Grm?!g|fNcpd5DAaJy6vAvwwQHpO4GAPz
z3<6H~^wd3N`DYS)4?*}?4Xw1xWiO_7QSt!2v@2O0ao2MxY*Q<d3n9Z1vJyFKPu5DL
zT9}c`kH)4PiE4VyaSvN_TqKe;V4tqkc1aD-!4uDi_sVTk+xCaEe=ciPBheqbN~%~b
zi5nvu5sw=ig`TZ&NfkR=o}F0O7_67an$aG1&>>t=QI8vAwQ8fhvl8RI3vfrpHRTd}
zvB=sF*Gf^dChx11J;MEzc=OeGRII~EiN(h5<4W;ZTrI&>ls&y;MY#683|D2<@YKU{
z6`lv=b(#XF0cg3a#jdVgTj+V2I9w5KLas%P*3Q8S>w!^HgG;gM<q|uM5$@Me_>`+7
z@nAD`wPnxd`mP3#Q`yOi?|W|WMl(RmK%?}+Df2wXr712-gS^;cuh!*ZR+rh82W)7I
zvm6;7c$VY#$&`}{o@~0aEfFA@-0E>18!Sr%eK1&VKS*sU+;-E<efH7U@1ouTv4T6$
z7p<#_<=S;?H}*vE?^4vbNGt=hTie))pL@!UVW!538V)q#%1}Y?cAyYlFvPSS^s^Xk
zA9$7Mq9NCd2X?>)rCOymT5TXCc8OeZa4;T&ty=8*S9XEi03I}2fZJO46zl8^QK{-|
zcokc4eQU8<W&?N}#_K3xQv#^XXcTUCjmo2A{0>;&Q-6OYuJ-p2AS^y$3O!n$7}#);
z*zn}y!{(?0wct^0C}PXn5!^t>@Z&XC>|5YUJwj}wUSN`>y9D7LTmG((uZ0H1X4Dw2
zwJJa#T!4qG<fVvRu!BhLDngfaW?^8nmS_kziW>%PksL6E^>72pK6ybdSf5%nu!;#{
z)qoJ;X1GUBNR{;r+rZdMfB{rNmW08Q28>ZBUMfWA!qtB9wME89q*j=yXcgGR)~du(
zR*R!?2`NCh{rp#80^^VX2ui?_ih9LWzWfZGk+*89DeU;jJ=%8@ztNQ{TUVz>Pp3*p
zg?^$H>m7Nw&X3hz{NOFY)Gsn$*X-Zvdnl7SKQRi@l+G1aLhIsOs|zm^%HX02o!J*@
zeWlR(mQHn#GE5Smc~YshPPXMj=uv<GhEgH4cB9spjSy*iN8ZKyT7l&tuV&VFBVYoS
z_b2eRj7qqAt0-mSu|Hs?1=U;QNuI{@25>Fg%`bC!8L16RQw~V$_7zsQUV6%KK|Nh$
zdNQrwcIc*kUV?7wC#~Bzc7*PcnqPI<D-95h|I(EAmDn$~+_Ki?y&d6xYOsITP@t)r
zKOiFkNzGjGh1SI*9Z@ll`D~aV6to^~q)2O$1adbpQ!<8%NbBs=2PNhQ+i>$k0vWkm
zwLz103c@kH@7;b8(j-Sm?#Kw`XM0C4e0fVSQTrLMYvTGI@FLp~^{7x$D$zU{JZhah
zX>~%2P|PqGWQTi9Ar#^sf0+v*;T{!2hI_OTqIFGkk2f-%ehjIk-D7L|v<g&XnxLzx
z-=b?x6(qGMu^2Zo?k<3)_ELzRq+=6W&0=61E)iOan#DjVx-7=ubTnU?nz>5;d5UFb
zPV0oo#QH3A^rD%MK1%`2oX2dO1WNN`rOi*Wnv%LBa~lkv!}6ri2(dip=0ZqVo&=#q
z=KR*1NAu)NShA_-sT=NL6QZYXc!$=>wp_>wJDngm(=dxRx99<-nZ;Y!!0km5&}Q+D
zXX_XNqj_)FaRY&|d;$o}$~Oo}Vzz<(n#wmQQENE|WnLS3gr1n!TQML*DbZs}Gq2c&
zOP1CK&Ag&EaPvyHx$~eT@WUYPM0Bh)Zns%&v`k?-9b?9We{)&fY}KocjXj$qmD_?m
zUbP8#_^{j+OZu=O1{+{l+8xej;~8sA2GyF7*e<g>rg7&-0~(HwwOGP0xeG|oeMMV`
z;~otR=D`Il%(F-FSZ#1Ps<uXV#&yV~mEi+)88=ja$<@_(fMlV^`p1p2nPQnIiIiqT
zeWCQ{zC8M)rw3loROlX$f&Q`QD2pZ6p0oHXAwQBw$cd471C#JW`HTd%;aWitc+E(l
zCcVuAj%FmVjW*yX>1wD)0ju0sLrJtm9_<~lt%~M_I9FdT3x@6_{5L?B?<D*Yd%~TB
z_%z)+2{#L4*vUr1=jGf;C_r|+kMMqpdoyk{)i%PcjZ1xEJ*1;ugA5ekH3(L}+JI%F
z(YW3y@GI!*{d~{jrOD#T7CiJpcmU&whq(zC=0Zuq@`>cLG7U4wsp(m+W9)tpTE~+O
zm_NQRwVe=_#HN;qf+xmqjAQ(e5KWD;!i9QM=^o546=M%38B1359RN*JwM*{u%wuRM
zRC%U=;!~b^5)Fka&Jj=yK4$N|%c$ARZlL&mE{UU_Hpo@TnGHIq+~+J5tr^#P=t8#t
za8@QSK`2wd3u%Ou0ZHc&N7-i=(zP@cO6LU>ADzF2hC=DQfMV!;_%5WoNJN4ayMtXw
zdtI_}*H%7AB2U(=aVYI;D?dSk2<A_CZKabv?vlQ5khpuW$E&vT2^tEe%1&C`yO4fN
zL!pMHfZ{VO|C@%=o)(8f*}IVDTqgq=*H(XDjq2tuB+eky3;38*AC5*OwdAG%Vd#7|
z1r%r~)TV$OM*B1x3Z)$aim@qxD~32?%GQz#X;=Xp0yqk$lyz<M_VvUrB%E8gf?LO4
zBSE<17Zr95(p;c+A@xI3&oK9D8VY4w1Qef<@M;<gr9A?Q;quXgWJI*qLdWeSWL*bx
z1+U>g?2Ra!9C2z6Ha|cTOBx&j#mC_MISqv}I5}wfA`QhuWekHeXN?-4FkhD+)o^=^
zh8JLTII%%S$EQ(h7m_PWVC+IVHgD3y;n8#Tp}6}D$=%>#m<yYrY30uWacZU9^c>>Q
zIhQ1+c7KO)d63Lixwm&HWK-Kaeg%1_knJ6Zz9MUTN0KL#yE_ilH=@Xl;M%>c0AYXF
z!67oe_pyZeZ$uUhJR7;WW8(X4_k!rl-r1qP=BrL)_KH)X=N#OS!Ir9Ed5Wz{$=Y_d
zuL8@sp)MMBg*4(m{*hUh<fYN#xLoauD#e{K9D){I!1{sOZr(=Qk<o*jNpRJwj_YP}
z-K!6($#+8ucr_XjYh`d)+(5zFrt!5uBto_I_O+lt*~L{0b(Y9vVUJ>y*3t1+N2c4=
z&5pK(gF(kKhq;p9%8uEXWgBXJyTyB$`W_>20rfSuTU-I#E%fxKR>Cgo^F>`XE6T5f
zX5&I>lA~^bmg@DVpfRb96v;k{$WiZ!h54yUPLDdv^VwFWnzfAr)@|9&r=*7R4|WFy
z6WMg^pg>j0+ChOxY1IrYMsB}@;#=A3H_56uX=*3tVaCRYLo&(aZ$ntP$b3b!BwH;@
zV%A!DZHBnY>cYzyF}S?u%BUi;(IM@SUR>~gW{NJwPU$M5k3Y9X55ioZlLAfJdn{3z
z(%wI9VWk^dV%uJ=(^p%addZ(*g7QCtHsTr_8u_52{o6Z2{frKsi8pOtrTgy*r-dA4
zNp;8lY=|{cZnTKC5-@DThb0L_ugS-ud@;f?Yq0)*t;zqQC0TRSde-EB)9Qp4p{Vs_
zkR6q=DTG2QW1q~0kf@AR2pMj}Mu^rmt@!i<OsD@EsifV;G*!imgz7XwSM!74<`Gt6
zF&<#t{S|VTRyWOJU>hzGT8dgd7)sG)F@kFHG1z2#h`V0X>)b!Q3S+Z8k#9F@@+FZ;
zh}fnk|Iq8HsViBNzshP#tAemR44%XCq|gYlJSXKsNLZc(q3oLc6mkzjy|LT`I{`KM
zoar*AVm0~f8?z?=++4_|xMi3{8zFi?X=d?H*}z?nBB0IURMq6?9wQJZPi;+pf?-xo
zz7VxGi&f^em`CV|d3}gU>E$RTdQ55N729yh(%PVzSJVb>Ug<Vkxh)297kHR{ttKDa
zsC1IrFuv^~IM(L0t>v-Pd*EmdOjYXnPB^WBXC$NH;x&-gEDsIEb*Q(G8u8VwxH=d|
za014z=5W*tg^`eZgo1eX<<5Ap$TFVL-x@eoqXhRO?}|#}aH>OjFd8eC*qJvykf|&}
zHfVQTjqnhf@(>)mfd&5TkQ;Wk1s?SvPpwGyYOwZD-QO!xt@ZM*;U>IYi{Q+T+8#Vu
z2kQButGLR$gUgv~J+&&F?oqE7#|vx+fJ6dNF2&cf^F)q2e#3@B&uQzC6|z`AT6YOM
zh+=CCj;mM~ZI19*iYEI6*Z|cG#i1Gl%Lv5FDWuikcK9SS_Ng1*_MB#|HbuMD8lxkI
z>shD^A+hjEFj_}{Y;{DBUEzKhbo7VEcuH6q7aMVM3@66r&zNyhgVD_G9s+7}oVFCv
zfaGp<tWYhX@<Qyy)z?VoMgwRfMIr%m{>SlA+g~nO&CSzzuq!dX*$FdDokb^dK>`L^
z12(_F!BNM_g$9;wwk}XH4vvv1Cd(w+ZW}n&MsCx|1PIw>juQa2{-N|o?p~pMzdYaN
zKJB}S({&S;pM6}%l9Rti{)}lAq(#{=SNj=VYpVA_OZCi|<U(DBi4H5dY90N#)e)`3
zB1&P<QHPgH?wbmt^-m+Z$W$eXYF1(OS35#>hJKoXjhEk8`Y0Pq4<Yq5qwWO-5le&E
z<^gOoV(86_QSZ7rYHU?{n(<4h(gF>dCZB(|SQ@NI&sZJN{1J-8pgSf9DWnVw?Rk>U
zoL8C?9U38FofCu#LbE;X6~R$I&H5PAtREpeXhb#5!Zt3=x*@T8lWaRl%c;V?iLt3h
z2uA|5pz{;RnuvNW)i9NQ|1{I2VM-s9tGx3Dt5aHxVgN8q90MSQV8{SCD;Gjy03--q
zAWeI)W0lH5Fy3!$e4LGqqmffIQ?;>yZTQ&81~PKSkx&P6Cj4Qm>u+CS+U)mxj7DwJ
zgLYnvvnFX@Vg>k+dtahNOSq)3`tD1V6)mCr5^n@#`M$)P*c0wc#HZ=rmw18N;N4%T
z)y6v6pLlxC{fQRnj`u14)atO81=VoVU@5L|go&!-uEUG)M|LCyKQb0-knrBZK5Zrd
zY-u$5Yt_oQb_3$hTCJikKh%yM(%wn#+)*!b8>}4w>9{44DE)gOY{tI`gp|G8ak0e$
zX|Ge?D|7u0P)NOm$FwZDc0ax?=?Zs0ej~N}@z0)X65;OC+(ov=lL*n)sE6)RZKY`R
zuas0il(5pZ|AO52@Gvy>tSSFX8VXfYE};0-lz)?kLKO%LC<YG?BZe|(YWmJ(vW?>U
zlO(2k6<V$ubP{?{vFksejLo6yp%g>}uzoi9r48_hMF8R^w+9f4sr^fATj^5s1=>qe
z&yDXC%V)oeGCw3uGoY!bqA#PNP>L>~_$c~H8VaT80*ayNIW|w6NJ0))-0;m4>rpm2
z6qGZd3kuq+QboG!W*)o!b%O`E8cPj!|EN^?3Q`E9D!&BT<#j!7OhchmSwQho<pvFf
zQe^?fP~`<0`?<kz>{yt^UeYE>H<LI7D>VlbWOiS}#pJ`lw~@$`JA4Ebb04^S!|D5J
zD3mhfpyeYp6iOL56qm3sfu$i$*ndsJ@D+A9H=J_RO^f6+h<&u<A4qCR?GVGuXaCpJ
zG!$z8R}NG0D;f%=9RiB6|BKvkI``FNv=FfOi=$vlS=Tmm+Yk{LO~n<r5lud|)AVp?
z>KQB*Xeg9X5m0=Fz-crTN^Jxb!^4ChE^!Hoh^`B{g8mK;m$=X+EBE0NJ4xip(-Sxp
zhiSGp=8lmdg8363TRPd}K3w7^5_b>wcpWbBW*Q2m%1&C`zWzQM3N<VR6rW-F5Dle0
zEe?gUj}7=F3By;Ty6NjVgG?_#`uZG(?mHy4q^}oHe0=>s(@-d1pM#cP&`>Dt5K#Pm
z{XREz+t+gxOeyQy=I!f=!zEY(c8x11;y6M=)IvLcQIpC;Xf9AL|3GN!>GIdmP$=6X
zp!nF9lW8cF_6R73%g>P?Igf;_>p(6aAMV4>M%m=ZkK|x;ktCKhI0A}~!Ku+uD1(!O
zmRHhHJXFRoIPy+gG4kX`HQfG)h8Hlw;KT+Q9iK+2!zEmcFvj5$YxAuzIsAE!9u#Lk
zKyo&C2<E^gDCP4$Ks)!b5-u4`?XePv>8CZ=i-PXsB^1)B<0baOLO{sz5`P2C8@ZgF
z4)K|Z1sX<sCsGTN@|cMg+G}_VAwPXwEKb|UO?-n@MBEA3u@y}4b@Xu)*Ib&E=dD*f
zfrHI!V|{R)AKXNU*A)(It-}fEab1FE2Rh(YoyiX(ON<9STv4m<ZbUn4P?wCw$xx8Z
z{vYlWl&4CFgy-^!WBlN2S+Vb1gT^e-IUqquP{S~l+3Nk%RA?RltJSgTiS*H^owvZS
z!6Kd6$PTSL8qZ}3o{Di2&7~5}<R`7$Uv-4#(S;rl;V27b6OnCx^&=x5Wh(fDKonGP
zlFD6{z9yjDMWQ9jUH;j(+(nM4&~levK$e%gtYuGF?t)L#UG8#ze--v8#if3J4^=1C
zE}L>zyO@wWE_PXE_SXyBE*iCZqYpPrY+}iEel2J3h;bbDBD0wsh4$BKP@Dw<>{5Bi
z=}A{u9`diL@{s-0L~z?_E<w)~ui<Ya#+1PL3925G&x)XJ2^P64E4D&Y|5!leiZ8`R
zCI$1^jusS>pwA#;7W-97hJR4#kA<UO;h6XY_phR%P{F-`VuamfPJe=i6%cZB6tc~?
zbDq!T<kKh>a=R3JYQw!l^{B}j;_hfq;hc*0P6dI4M0+2CcJ64;HZL;RJ3Jfh^u6)_
z${g(rgtD}a^=>+^2>WX45xc(zZg(DHn|07a4n)FMMr@ZI+yuRz?<|8?xNnHWi)%5+
zfjw<-K|9>^J{*s--EMF}&Dc2Xw1bUzagVm)4K||5R|dFqiAy{dW7XG7Y;3e3=dUdN
zH)Fi{3Et8oIBeQitS-EwIR+P_x#FaA#GBe4i+8}fd6Ju2Cr^aan?1nwpzg;^*FGbl
zOkPm;bAXV$hFPK|g1X=N26b}iga&nUUn_&UdF%;;I((Y$pl(Y)>`m=24%S(r?pVDx
z7&jVmqm$U~oV>A}3xSS9yyM(-aq|Qv%0M*kqus<d(K@5O1D!ocLz3~UiZxb(oVyhJ
ztw_4U*l(ZNFsMcssttt`%iUy&_{g}vWDb87e=@^A{TDDD@}r{Ylmw03!QY9{G<Wc~
z9vhj#-y&0Rf<hw#oS(cRWu<+isM!)u&5Jl<KC$2q8VVH)3MghQ$h1|GH48r|igwem
z0%AdqLbla+#)4dIK8;ebpi9xGHWplI1XWJU<PHfH>Zy?MV312lNcd)G=MD)ojE{^A
zk7`RqIzZfKHaP7=g$A_^5Z`cFJlLu?%Gbn|ab`<#8@34j*m15!84ju#alN^vTE2`I
z9>OUIu#M!JGB_U=CbB)<4VcD7jq-Kyr`7{sz;Uz1DttS{AnuGK8Q2u$Xm~cEq{igC
z$T4Z93X|RT`!Zp?Qwd{|(qaz-gxnF7L`y_a%fT6Ck9C&+6t~k!-3pDMz5vMbwAdHf
z6Gl+@G~E%@1tR$h`=9zDJvODgz%0-mM@>JrI_w>-mEwxr09LIQF0PeY6*v*PT!s5e
z8yh2as=`aFz<}+1d&yRuE%rs5tK&O%WbFrI1<T&o-K~v|J3YcPO}1|tV+XDjD+Vlp
zlJ9$7@?<6W7F*;?+miZT4e);wqg1N3IeW~t1MG>UD-81vNric@d$vde`O<4fX3O2k
ziC2vXZ8p9tnJYdimi|}5hum@IFQ6$nFSGjFee>O{8>#r5tZoxfe5y9~r=d_w@B)g#
zxy5{ZB?FQ3Fui3f^`7GH(IiH?N6J;rk%AbsU>pH*_B??sCF4xWc0kSu^`~4*jCs-O
zv_;cx8SH*x#>}P7b9hUs&hn$;`1vHqN%GD?et7Bg<unwk3Q0im(dQBkh0<pM#n9&!
z%70{X+fhR?i#5|yTGx<R1?w~i6Qsy>m*Cu0$ZsH#Ck+vY;$RTgRw)g<iv$tOpYRGU
zCwuzj5Sr1a$@W1KcMtY>RUv<ZhC=CerVtkDDy>oUB@$M!G)!=vXswSSc!DH?tdnyJ
z-(9x-V;Tx&)C3eCTK=1c5+wY6!Ch!FpK}uxZ3N34p>!f@Gg5Nz12~k#-D3or?%BiP
znGJOAc>D81wNPF=gaUU-;%Qc9^3%ONRa82<e^lo6`80#5GUJn=sb|1?E)9hm3j&JI
zSlCKKp~ixMVmNrdUljS6j#BMCt3xDgT@7-DrXB2A-HGBc=X9k0avpR+f2U$c+#5)K
zlXTvKjPVLg@1mhldMlv#=<WMxD3sm`D2CpK?}+<2iAcB-vlDUl2ATQ~yJY2#MZZQO
zPnvoTrF|^=I0+({KjE>clRa)v_)`*h5B7M)qQ9k~P^#>t#T{Dp+)PCz-J<1S8cKUw
z913Och&zUa;j2;I+!4pcV0r-`bLwO0&LpWN18-sIeAX>5p`lRgmN|^}RWuYzI|LMS
zbrg?95J8)*B_lMffYnisf+=NP+lW}zr%`Hk)RiALHm6)X2?=t?_?wMjw8_ouN!IEt
z$rZN=I6*$|W<V(wSG#@QJ4ozFL&%{xSY*z{)lA_55>_y$OmLkr%*O(Kh9rWtKu)o^
z{nOWJC{)-ep!m@8_cRpB0%Z!{wmd%~VfYHYoBqiu26y`HcO>#2qr)rYS#S#(j0B0n
z929B{9zw$M6~|WCn1$A?S*-7hAu>0?P4*Z(jwGU6WAFtulrD|I3uq`o!neoZ6(kH_
zp-<Bo+)X0ituZ)GLkSXtIVjW^yp@FID~@R%gLjiebZZR$2@R!7WAI}%lpx{TWAO7N
z3}2y7`WSRi{l7;d?=c2_LVi{*b?(XF3Sm=ZQUsxjL8i78(%1#9Wz$)W(_VKGq(VAX
z1o?MBPe>8uuYm9FBFI!xjx2vPc0pT+rwbnsVYPPO6WnfF_*l6#uEXw?QP|!RX}h2q
zPvC~~h~M-)5VacYa-{)Y8332tyri`g?~B4+F*Vo|Q!kfdR_a);j>Ln_Xt332)<&VO
z5w5H%*Xr^{c3h}{TI?fesM=I)-mPXJW$uTWEAW#Gt*O$RC027Y^Q{k2rmEul0R|KI
z!ps$Pd*d@ta!4YSE}mG`5k}7sX1(%NHcPVyvc3uCaHdwX1@fR)w`-UGZ98S=KKoz|
z6uT%%7S*vkCL{Q_6{=tPkDwoNZ+Zc)9%B`E#zjV%C3kvPqFu2HMY<TOHrR$PR$_}l
z3JFZDK6g(U3UZ;ywpAS~4(?_f`*F(_{2x>6A^1X6>GCe3i))Q0@2(NW*N!obW&k9-
z7F<|XXWt+!K%zCu?qeI^cRLIeqO*tN!QK2OH|%ju1MXiN#JBi$XzZpoRwP^;9E`_c
z3o=}b#ye>gdbAOqEc5}J*h-(|!xh6ZK(Dh+Yt<UyR*UL!vsJGmmaGws-O4!dcvls6
zrm-XD8KW<1G`BaJ^&QqtQAosQ-2KKli7!LRGb0$zEdU|0Z}C<-E=6`}RJqCSYb(NK
zbc0yDSRZG1#TDUBIymc~!uT6;*rjOCa9l+Yyuz6=Tv6EI<<9LdyFXUIHq~pZlsaQU
zqFOlktXo76aJA<zf<2lme$@v03$1Z#8lD9j9&)VDPc&ca=3=WGF9XD2^ZQSX`z~#i
z$bqZh9)BboyPL67`o7EM1Ve#MSwAb=GMtCpq&z%g_{E4H&7Y^i=!iP)X!cuX>@#B@
z_AeohW{7s>BQoT4&X9X_L$(ll%xjn!&)OO#_L3Sb$vui)H%QWsbjcXGhSK_+3`$1X
zJlRkiAGcTvV-B;`+NDF;&Or{qqgWLkvv7S;=F1xBH(8)*)f2G+!`Cr1Z8B(0)UKjk
z$yEQfToZqyX(GQ<PrEGYi9dWb0}FHKvy)0qYyQO28dG8C$rwQD<QyIoXdOIab>O9T
z3@WPKnd#sHE+NekUTB!uq>xIR80%F`W8a7P(5P$18ryI!CxMJ;HX28U!$wb>tHK9$
zroR>-c?L%=({=>ibz;euyWi7z`#np_rkc#*(VSEgyKd-7t1-=IVVM|2hh?%6q4}>_
zrqxW&k71{D(6A;C!(ohjzpuiI-{xURqN&F)3VwwY&>Yk>729yCX-;XHikxz3>fL_v
z_aw*OMBHr!$*aBaO-8|f#_O8$ANo3K(%z#cB60xClT#C|lT}tHv<Sr@V37GB=!>+z
zQV50I9C%VLgv8B(3Lztuw-KUsO&gVKm`*Q8D(O+#nm(<9F+NSumBrYSM_7r)=w;l!
z2)RqEn`SYv4VMTlMa^QM6uHHq+h|wL<D6xqhc(k~U&J=Eh4F->&0A9u=7>K1nRp++
zbbVt_R|_vTMw_Fid0{)tQS|jxYP-si0%EBg9wtS-adB{1q-APiUu4D=)Ryg<D^Qmv
zCHbs$;`%%~q0gV!F}1kXW6-_n*~K@>i)R~JN+fU0Ba&p89?RtMhJ3>m+i)4uLrxo}
zDEF2J9Br6l8*Ru<(pL9B1*~#!b(d&~6w2woTixYqg4`2qGO%}I=GZtapul~W!?n`h
zSkJ1DROY~P$L9c9zSaHn><PEJ<I{9+b-!o`ZUk*YO0s{fR<1VtYhxXq6tp?#NkM7o
z9dCF)Bi&~&PrG>SZFe3Vw12u!X<1x9snPw9Anh(48T5G46)qocPVFT>HV2i6y}8?a
z61U4flc1h^`}fbGX{zeaz5RRU>nUE7>$w7oPkGrA8Va?nE}$5@61ucqb`^=0ZjW-+
zYj&_*_6WczB{+Av=JQD8Nx^X_j&e<FyX-k6h+zJNZ&7fv$6c?tjl|u9Jzn*CLo^gh
zuQP>^RV&maVFgRW1Q%4VC&c1o2wq1LLDmX6h3~GJy_1GQ)e7aH<-IhNAmLlvWj{!Q
z2$p&Hc3Gzo+?7IKAaVB?fo8Z!?i=Aw!fc>($GHXJC+H=!C+w<GY72rZteD!Wq$BjR
z%^ZQ2yD&-NohnRv47d?enDh$}v%4_K739g<q=-g8U7oa=&E5V^ghb#X$x9lx&4cyW
zbD_!s*XQ9ZHCAF!u0rVp6d>S9Yug7dt;hV<fYQb6Y^MP@IjIFFe)4zuoP<>e4Aw@W
zlmTiuU?CQc?cK?$Cx#-)@Cf7op4wPk#{!K;#40pmC~Shm*`Q=(PqAK#j%8os(^#?5
zC`1<wiTa=-6k)J(48*c8is8lw_*5PptHf9s(r7_aqZkQfcx^=wFF61x4R<ZoFcetu
z7OCnfnj>1?bH*FYpwL`}SeoLU4C@zLOx3!aZ*`eH?6it1h6ko)zxRhL9X0K-xt?v#
z^a87M*ueO`u(dES^a87iZF1?{Ye8Hqyb|p)tEKo<s9j!gYhLg^%eV;I%HnmkwxG{a
zRa5&XwYK0S3$)iLVfZ>$@N6<jb<)$8=QEA!#ZKw9f}_2~1*v_%m(^(4l7|&Zithr3
z`$dR5%{wg<k8L=;G>^0tAM$9m2OKTMhiz7Rw9zv0*hU-O?MB@BJe}Rx*aIb{&4G<r
zMVa<n$D!Yfl;0}!SpQt3Y@X~~+I#hdlH2up<fg0Lg-mqUB711HTc49OLCS;LBF4mQ
z>_gJAFYrq9Rk!CNK`bG@5itivU+2$s4JK=@=e$TwOw|2FjAw5^$lmIXZ8)DvAbL>+
z%BJN3M^krfqp9vBWwieSu*#j$mS~BL_72~SwhTZ+Guqz*WO+vW+w2K5+W0ix8SV4R
z)pD~Rmeu;}@mQ^sWA9JPnanl-cbvyQ+UTyg*o4#V%gv2&wOXwbZHmq<RvPX*jrKDd
zO2c4%tE}8lKz#aeHAi2r<4*rA=?YWS1F01CflewAMKc*qyx40V^E6j<Dd|i;D+*`+
z5k(<68~;5}l*(ed)6Yw2C{(6KK=H{s9YI5(GBpB<k$%o~n9~Llww|QOmt@8hyn@no
zeSlG}!$os&`a+UgGQTIF_?*hIjfO&<%8`SXAsUK@))=R9%tqrQN1XVi##@ty7H}>H
zCpFv0tNa$5a*LbBevM<-G-tY!sz#}#n9Gn&ZBpzIJ*#9Ibay^Xp`6Nx{Qwvdk`KEZ
z<mk?aS<XbJ!i@ce2GZ#)*Z_<7ekdG9yDZqmsTY9nD;FzpdjQ*X1Ku6>`;4*8cF{m1
zt_&5R>c3GKZtd*Hq*8ym1Y7hr!*|?Pvpc5_&F(@AUO!;zt~p~qI188@3(MWeXaGKF
z0m#-*EqlbU!g=vrNtGIHHpbbbM9;MhGWNe;;9})kwjB;RN8wgwRdHXmPF5qW>w}ve
zhKj8U6fQwpaR7~hy)UcDl6OO`DsGNqJEaC;vzOX-R;-RkO?Is;+YH6`aA7u4<1gD&
zxd*bGaX;UK)n6*t2cRe1Nd-5kwqkJ$t-zH17V+=<FvmFtH=Z#BAr;A$rFYM88PbsK
zH6<pk^R>}vdDn20ZQa^a9*au|34V9aDBG2iZCet?MOjNci5#37?qzb$%6OfSU)6B9
zu|YjQX{!yf$E+b{j^Q?y=l@_e_8OlIqNC_Gg-A%z?ay){B#Lemgcg~D)tgIm#juu!
zE#>5AOi)3!?q@YSmY_A$s0Ft9M4B)?(GoK&_9!x+PPBae4{B5;>VHOuWNGSfxFt_>
z&=?leWtgR&9HCiO$9o#Uqb-151C3#c18o-b1X^!2`g6F&lSs~D-a~8rypB*w-TqDE
zGJiliEQyX|%6fiB7%}GP#s4%yZJ2XqN2si9>x(rx(0DF)du_?Z12oOUP1`JB&^oVl
zM8|=t;DEU|t$Q1H(uO5j&mm^O4gt*US*&e9wIn~b;YN=HqGvtPP<{gqB|XlxtS7el
ziboqQ$&YQcuyT?zrS|~Ga%V~<S|U?=oo}X8PEtZMr5^%hd8YKk><Kfa_%z*_(iaWZ
z<08b@e9ANk3;dmAOV7-iElmOMG-JB$RS<BedhM+gtGimoUGYYBofJC)0#c~#Y|%bA
z^<fb1w#xjuS%gWX*iDTatJ$f|c!)$FcAz@qY-E>}XMq|jSK*L~YFz4Pl^a=aHH(ei
zo*+v_csXwN`y(k@_)#RgXOnB67S7t~r74j}i|zPDO>4gnW2H-J=`SZ;VOsi?c3OJd
zZcv>6Rg-G0ba_(M1m$ntHFlWdKa4WCn~4ohF-FDvM&f#&d|xV)`a<c@4-&!Wt~>ue
zH1$mU{|60)O8g5bK8gSTp`lPITmi-4UytH4XK(u&>ZNa_Mx-9F<VydC*HG^V2!V`z
zxVM_bolNEnDCV^)?qb~&X(*H?<e+6U4TaJK4#gpA`(VE<Bn)3!ck^IBj=CA}l1q0v
z##Wi6medXb#b>E!4-JJ{>d8UN%`_BBI|LMCsfQ~@#sTbi(69m)dN>NElyz-WYc|?K
z&);)fZCw6$XcE>bLr1uix#D%{O_I;bnLJ1nOI-<cKQ#4pCZC|8P$on`@iB*AqM=Yq
zBA|5YOdcmO&_zB)&g45>Mn2s8DTzC2E(8=GXYyMb3Z)4-Xz6(q<<vcB@o**wlQ4W`
z-A!l08DyFn(wXEKTPKjzk{lIKe4NRdG!)92<e=pe8j6RV@OLIx(XaxX2`AK)vaW63
z&g3XQzviAKng6$+q|9RNq)Xu+ls{?F^iuxhYG~@|Pi~~4P*y}h@v(@v(@-cq5l{?&
zvfRyS--*h5NNjWsm?GbDFPE1OCqG8wPFfEE#mBdNo`yncMGjiNO+%ryf<tln7Q8e~
zjiM(>7{Rih6@E9HQw;8n&;LmxPwp$sWRSJ-dEY;#{ChBe!YA!c_PDoQ9z^2q!5+WV
z01c(RmReUPW~2saC|w$Hn`tQRX>pA>yn;)OxGf|MUybT!G{;rS^z&pim&3D_Noq-Z
zETH&Ab9-ngR5X`^mYZoPly(Rx{?XhWG^~JVj-y~oS=To2Xl{kRP~(*y;5$|O+0u@T
z=Jc=B*m#hpnu_V}ho+t}-6v=$lr0fZd?JxA(NHLD5m1bnE+FCjIEjHS`zea!zQbkY
z!@Zx9xRZ86K=Fy=eoI54G$98qJ-1OYga<7iUgcmChOex<=~Xy`OnXCml^kR11d>{k
zqXLSLS2>f0LV1-Ov|K_%@vsyAUgatpR)AOGgql*;waweBth5*Ry*$ezvz7EN_See8
zv}g`c9_DIj>gi!_q@hrDML_Yfk+;)OD9sU2I`uH`Au-VPV2V7<y<A2<-1``bJ844%
z6dw=sc^V3(2{~x_HVuW+1P;Z+!#qjC@RfBpJq$<Pv^}JU$uYKmLsCm>hk)YaVdmdX
zMh29J$wA9OG!#lZ1QdS{vzmq#;9)olrj&JU^Y$=Di^4E>a+FxyoQ(~oZxNrAOF5k;
znQ|!`p{b`!xrl~B84>}-$0%M&L!nefKrviOhl%J4iH)uVQ{-NTxx9Qhc>{?%X+Q)N
zANTSm8VaQqIcT|uhC*othvITCRwDWXB#dBLcaVs7iou<T{xpd^nQ6^rkd=u38VMqp
zKjDdJCwtt9=zk({_h64#BKoH^l=fQcm5BZ=4W&yXuIJ5UV9=fx*NC$c(Fc<-d^M_@
z0UcK<)76s!T@KH70!b}tj|CK;fbL8h3Kh`hpyd)83Z)$aihn?N6%8vOpyMc*Qr5N2
zJD@vO*4GD?7cA7<%ka+pQbl<!nh#WzcQrKijPh=zp-|REK=Fx9Zl|G8`XiupGs=4p
ziH)uqQxxUh%jM<6$&Zn^lU7AQ@rm+2PeY-!A_pzsrlC+;!J)W(juquSNx}%0bq7(N
zQw;7X?>|Z8$uK69K~|Kv?;Vui59UvJl;>oRJIXtV#NC5EUQym^8cKUD^@{RNq@i?a
z#BHXbw5P>2;;bld3kkzlqq-U8ag{RtKN;oa@N8w0TGAd1C_Yi%9vTW2<>jE|W*Q2m
z9RiAfly?UWD<I0_D40^#waq)qJ0{snmpgGcS2UHeUGiClbq~^nQ(@ix(9|=m`veVz
zGA9CxPcZT&8VaQ_0!lZ-y2nXubXAz5u<kotUOt@sDTzC2Mg$a}u<o}s6iO>{(9*M)
z3Mo8jaXA?)tUH*55iIKt!aAoI++kfGi98v%a3~H2Sz+Bq5=1b6!oxZzd)#5&B_!@1
z?C}cguA-q(w#!M2yNYy#hSH@GcP$O2JuMD}vNxFCLc;LXsBVUJTnwheC&RiNp6x!8
zTGECIC_Z7`Lo^gBtjj^mCut~@b_gi`Vcl0~SOH-jN5Pb`u5I37U6Q=xhhyZ9?UFnl
zYcB&l^Gg-w{eb2J73F;&ntDcg|3O2ctc!r+6Px@G4TaJl0i~N!-qN>F(S@!VQxxUx
z2MB?@d^ow9#GSM%0*X(RcOngi(uy3kY^I@5TEU^Xe2x|6Z6RR<%esRo&nX6Xls8Bs
zPlhp>46>rUt4R>S{0Wcpoa}K&c{h`|d$7lA#N9zdX|JVT$-8&aP`WhY9;TtRr^Pkm
ztmNINNEp5v)y*i6tCZ>g$tW*}XZt&nTGAd1C_Yi%4{0b=l$V2+U(!%0?GRA>qr4e+
zQa*SjAj;z?m{QiYjfnDm8l}#QcbzacwdcjJ(9ercpSb2eGhU&cIy3%raHdMgnen|K
zOZS=aj?>=cx$#Gm$fwVae<eFL^9$lYEC`nVUv<|4<yLW?ZL_=f?t0g28ygqfv1}gR
zHMVSmd1S%73^u_Jz(5^BK&zGRt|V(Ep)1*5AlN3Rwv&635X3V;3L(&PLIV^CX-m`7
zJet?ZgEWLDoYOQkfdhF!LYk5$q~AYx=H8jBtI@rZcEOy(Z>@A^{>S|PKmW}A=O672
ziq`h^tjr<RjZYTiBPF`^5LXe>Rf(udj#~}OCRHJ)GU4eQ?l&A`_a8>$JLBnWd^n2>
z6YUz}s3D7j<wmipvi!ytA&{zxp6W3PwHj}yLyA~9Oy{&oU7Y3VLSOfjU;RXQYZaRJ
z(p1}0<U;d%X{^?AS}<1frGD>wyv;4Vhf4EUL=DM>=A~QwM7cK_JQ^u``1;cr%0dCf
zduX5rx_A|zI$hZXAC;b>k?rYhR@Ai@BQCn*I15FC)vQMudmBA&*-WmCfCi!{Urdjt
zbMY*z8gKtTn%-&TL}$*J(M6oN#eiMj8MiCb=kXM}xS%{7?V@B5&Hz=$AEF?Z)tmGa
z&86rn!dgH{QMW#wOJvJQ5NKbqN!Mq_=_P`~EYtd^bb?VYFFN+9w2yYFr&dBXr^`u`
zBo*pq71hH8^C(zM1^+w)>OU+pL39g*U8i3i1wKMGMu5=Z3JGZ-xcCQ1{wTj0cmOT}
z(HJF@E=rAOF%;&GNIGfcO6ifbQ8ZEgIP|#4Sv`%?!TVsU(F6_6qHmETXp1B$DV_ll
zf<PB#x{0=-5%IfF(C7Mx{><I-#e5qbpXKY>E`z~4Pw2cF^YNZL6Bauy{ra|B;Ch5E
z_c3W^81~Ez>g3YS+_0EfJWIrkr)i)`h4A^@j!;ELBex@5ABrG;JAzA)at(o(AngPv
zUqf&s&2d*VF7?+C41^nWj9`cs?snnrVT9t;BL5I^QDI!@hgGUY&f3_V>!l@KILf4T
zr7bPRVE$3ew|+X^#Y8v3Q+MOjwx5wM((&WiXs7d(WTH=cMlghlKI3VsA4?wh5k4Oi
zts<il6a7vog1Cvg1Sux!B}lJy_rV?WYVZ3=xgTR(>P__C!$HsB7<Ro)bV^G4if3R|
zB@MJl`HiQGK*~yZnX<l1j^L_8X$`4by>UF^O{6q0F>HFmB#+42*ovk_BHzxyl+({6
zCIc@9npg<<xvu`8h#fc9VYiTEe)R3?wvMVE+Aw38r>S20+=mf9rWS1!kq0KETUuc;
zZs=MmXVE2fv$iRsrmfVH)QDr<5Q-q~SRE;;S1@@8ORu|<V?97NGM|ZF?^sW7Rz)!`
zP4(f%NAWO%bG*Sp;`D09sa_u?Z@}8v($Py)ijSB?{k*{`EIeUli1=#Avspq@g1k6p
z&qw|2dr6(H9J8J&KPVZ)9*+n^8215BQ@t>`aT7kD0HunIM(pU$Py}&1atTrb6fZ&g
zQJ3uK5NX$Sj7z;8&DV@FOr+M&uJ%6>hFwSdKTPyJ#^}=zn56xzjm?u@4xc>A3DlSz
z{Ivi3K{lOyWz+f%C%#-4TYe31Hu;(6F#-5=%(Kt?uw<godd4|~iT=pbR6pX}L<ygd
ziB^%(h>89p6hYiXU4j%7^%A64P02(*LMHHS#--jwKX+KFA)LpD-#LU!iah-eT_71~
zoiWSPMIePHyiB1Nv3=<p8^#(!wVLI4R*TfCjp5RhC3#lX##Tiw5_y%u)Ys3mULSnR
z+q1jcg6(cU7dQ}@Ajc6-=S`&1IMp-eA^hK2o~HV7;r@^C`K~|ogalrHT+$R#>iVOK
zq(+=_A{0T~DLYb9*B@Sj^ja-B<&To}oWw-0cgou)NO$2jC!%XNUb`WTm>sY3F_O!5
zOfGuumAndTW6M%6e<@yL^7r#9dqTg*@(hr8bwe8q4ZOqhb1ZAZ(&YG;UbUs<X72V3
zPY5^jfTyW`ptzeMe7<f*Pe@=l^Hftrscxo<q(<D#*Fq7*-Han8)y;Sb(yO`TW<E~l
zbO#f?-py>C04;`#X?3@*dwDU8cpdk0Gs)v8OdfizmD~$!W6Mx4cgejlx%;`7eZoE1
zFN^f45;ex8x!2Qv4n~{`2=cbt@hzPWE0SB8dZ#Yv)$X2ldYbA7h`SZS=j&GVgamdg
z3!5TJbt_dQHR4uQg(8T%6-P>{Tk#U4S8vI!e1gpAPXXM@<rAmE@GcY7yX#~w2qSFA
z$=pKHIiD$}UWX+o!`j#q)sKSYWY{S9Ihl%>^#f<1)EK6gr=NK8L7c!QD4~w?aRM8=
zrR0G|JYyCju*rFv>PL@zAj0SCf%Jp~_CR}^B1-i@RU|dyfj$w6Ant)2DXAXFOORgo
zB@a{~8yjY#*9SJ2PJn7NErkD>kiK2-^GFy$JKpD3lFY+QGJ5@$ybo(*OH)4rlJ{XF
zV0#~Zo7;|ZcF9=|&GrCqra$g!BW2=Vwq(kbfqj(^R{B%z*21MrBZXo<VVIPhIXoW8
zmB$L>$$TQRbg396?K8HtA+hg6V&-sdiK)5wmx2g15@hjgjfZeWw_X-kbg*_J2|Tsf
zo57ws=Bi4+HLWmSu{w|io609EhH?ZTiyktsfY{6!eR`EX<(yA>d^(gVFnA4vMFvd<
zOAM9?(mkX-_`vQeUzpj$<ZTm^iVE?jI@k@WOGwsBaexW5$+TG0k(pRb8)gMGXFAcF
zby2KPHdE}L_*f!kBz6??>0GIQAkv>58_DK(MGNVIkwpS+v|LJO%_t?d$L#FnVpK|}
zrwm~5K2WjF3)e8}OF$*nv9(l*Q-b#@3>)KiQR)aG{!A}uSDEh5%$2^g`hC0YcUg>$
z<g-bm=zEp?Jm1DbJ!|~xx`DR4_%Ly0QFXLGa|#H}oC=^S%IbD}y#|1WdIf$4<5Qfj
z*p>1Rd#{wYW>;~<$jm60)Yf`PIpA93A$U!nUtDI7b|?&e_xXG@EeE#X;!GxyIe{Bh
zou2U%=H#B9St?mf4aN6ALF1C{^*@>agF|WL1kKqHr2h$;8VW8>(Hv8kiRzOy53}_L
zC?)shdy*zq%W07{a4-BP-hprfs_U-rh2KXkGcSAvePUjieR{jS@cGOMm-4YxdAJ$h
zI~daUN@$z*xFb?${W$S3mM+C`{=|%r8iSFG3Z--&hgR4D6O@o&ets^=&U(;MC!C5}
zzA`eL%_nx?<6vYspU>i%%Zg<Ki6<*|?4l*ZKxEaK5ggoM?yWy_g6!<OCDg)8E8n^9
zs)&JSQ)u(PIp=i6eDfn!zIj{grNFsYynED(#;y@TM`ZmZ&AU*1f3MnwkfHZ$SBwW8
z-a~u+{b&jgpq2y}6ey<-N-M+L^jLHUlsT3l<qaOtW6>Q!<yaJ1HgEyiQ(6j?2(Ghe
zj3o18u$oHd^OzB>>G+~FUN9<|%U}|$cti>IVw5UY;|Jm&x5p@m^uD2`U3(arV^KB1
zzaY``dMr9k2*JzG^;mS8U}FhzRQ(Mtj-X@R%A@Kw>FQjnJ*pnUwt7FQbGKUU;8=nj
zRbQmXqC2V{f|nI~EIRGrSd^pcv#aJn;RAhsaXs&8dMX+?#BLL(j=!vj>Oz`_;d`OR
z@Z!ouG|zVm_nU8Kh$jWh?iVhnm*_dw6(ib$rUB=em>!GHvv4dyyyTc3i%xeq7G;iU
zqY%lK-CgQ37|S*LM^}z}w7BJw4pn-{Mq_nj{xJ(i0_RW@l|xo%VSEs&yyYxzF#V!6
z1Ais5anr11hA>&ta{f-NYOiC%qb`fJ9w&sP@6>Xu^*9_$5Rdba9*fT7gy7{#J(d7k
zqj;P)il~Lf*?+iN@+Cd42K<W6Xbb$x46z2_u0$E2X)9I;mgu9)&<lEobVKw!n!<f3
zCf8vvLj*4Ir;G0uJwBbgvhg(}sUVH2HbVSXE*npFchv02k!>YbyBE(%-5Gc5X84A^
zOXIO?8wd)%tF!@Z!z1}gWOBELj*@p(<n^rohLB5e)rL?SQ*$F5LQx3Le?v%FON(uw
zzo<cQ_2$nbwE44x$GxyCeHoR)<%MiMo{X$8jqFI29Sd0ziDSFK{-TuMVdT=+8pRc&
z;Kgg7df;}R8ISD~ly_h?E~sDwZ$jG#QT!CM&zPi>E23Q&$BQU8Q9|hid+FJaiE}!~
zEIE;N9_`L3@84$1Gz!nq<<c#6biSt}b@G$Vg!a4mJ2=A5-c+_T0Z85H>FV=}O3>EJ
zXmdm;b-2AL`i==O77{W&^$Tq3t4B?6F`sc!^}B>1-4r)Hs(QFC-b;6F=Sz{!E&L_Z
zqon?WoFb_I&3JVqg0}aqpE_lVy7}UB<Z7iT7-N#0hn9IZ&z24G0ybOh?KXPI?lZD0
zl}2G9awphXi*|{y8#pL}R4ToAkWj23`rV$w%N?W0>}sT(OVD0k#7yPO*(B=9l*+{%
ztS&naY*dB@hr@<Ok;G0|ha#ysO3+}B&rBJ~XqP>S4zfxr=OmM9dsi^B!x-O{M};Je
zsa-uIQb5p%7liR_o=PU=BUcXHz)KHBtEx^7U1gs=t6wU^Xo;xniJ9rCCvoDuR$#U}
zhNtTD_9p4R<<b2R=@b=DQ@xYo=@f*IIVm-Bp^6N}b9SxB^BrL)dYd7>k$8D96hYi;
zIt2BqDlPR}Mcij4o3!ToSeWoV^M1PbZjbr>bniV<*H4Ecxk`r02_i3<T7M%K3VEDd
z&OtUX`echpAYF#W)ql`+liv>`0>@4MJ<<GKMzdb;q%;WD#^zHm8z~Kf$>t_oHcDBL
z>l^TnE%S_iKI7ft335Dy<mx5meBCoLs&DA9Ck3JE-hG5OJ&pCF#5J2R`gn&bLK^W7
z-S-4tdPp&qLr`t+P(_^L9Xe284O62dCAO$`m49(4qN=2%*uR$$y^c%CR!@*E{nFO)
zUr4Tv=Z928_+h=b#R?2haboB$B?UdrBRy3?yWF@!h<}ENG^9?>_H+_RQ3*3s)Lu1Y
zk^1!#ryUN*uRTd)aWd;&Pl)8#SQ~r%v`FM*cWli4{MuzA=xTYu<%w~ob3|fnoWEV`
zNQD|2Nu!@4DJGc|^)uS$)^iD^j)v=1QkroO$Erpz(uK7`jz`xHeI-QuJc#tl#vKn)
zXS-*_(C#pKiPkEiqv@ZgG1|#`)FYQPowc#YRf|N{bfycort8~CiHvO4-^V<no;E3w
zk+qQ)Znsx?Qne7@_qMUMkU7X|h?R?3QO%Ms#&DW{JYO#Iy$W7A)2<vzyA@`-lrN4`
zna?3H@vqwu&zc67Y4J#@T);|hSGts9D=fSBE?R8y-f4>@j)mB8#z>Uddb}7<>|hTg
z;c~QywV{U|1rwXpJOv25m<D>Re1_mssHgjdt3XeOQbx(npC%o8g6WXn&Yz|u<cFh-
zZJENSoYf=;lD&T2BW~5n^m+A3j9jm(XHtjX^K_`U8SYGo3g%4SKXD2x(Wqy7mdmsf
zuk|=Qdjhu?8)sSTg&g<z1yb#=G8yT`DY-}1#x@5n64^a66M8HF9?3njHj=BGpq%Z_
zdmG5v7G&}5v~LB?*%mv_jb;l*K&+iDxRgGzY(e(v?amf_FA78_(`*+mmd)o2&14Lo
z88%~3Mcs7P;IF*A&`BuT|2K+*n8b>S6i+;-@1y4dY-T^w){dDdrr9piASDK_#JK`V
zET$QwJxM9`kaj$;imZ!JVjQYzkLL14BZi$vQ+md!$L2nWw*<v>f*y<k!AN>H!-jsA
z8$09K<t#4|3Gq3TO1k5HLX>C6$Lxgp^YPGT4AMpAQlVVp6RMKEI7^{BUl888>U(#3
z9fU%*D2wqFrz@7l_?@aO#zzmg5#cgcQd5=(WP-O<-u!mCZN<>YU%8`lu_MaRNs*h;
zG+@HY)p{(tgcXh@NWw~1k42Yyz_BPYT&OW+Td379nze0Qso$WbRDQD-D!oOEy`ij^
zYuOm9Q@=Lr#TI1kdEYlN>jH*FP0Z@^`bs}F@0K1Cdaiuuiu``9$hAD(Ym=wueEMhq
zKB~u}Q*@3ch@$_q9*a)VITl6HS>8gh%!%)6F*mHZjb~1Li;YbP1+@jx0tM|?ra))v
zW(NVTXA?G`xBV3NNWaEHgZ+PWRrw9=AZS(jH72e=iR{xprJJv{w}x>nK~%X<k42}-
z9E+mL-IDlix%;Iz(?imtY?+pZhLyUBWW#rXB22A81bB*;cx~<=$D;Z*|6KO-^jLI$
zEd(#y^jLJtU}NzcY|pi8Ls}d`2fLNI>^AADM+#!ZLA2vqEw@_j;O3<yE>2x1d=x&p
z4(XsCi!N<3gsZqsk42{)9E+M+$`pf5d&ocbXvqV5Tn%KF+9ar>ENfff!01?I73eJf
zB6a7SZmFRdM&&zqK%ddGugeSl0-6R~6?{{VMdwsFmLMkZA9^f0wc%Kl8K%!G@!Hc2
zJt<kqt6D;2T?iHQH*v-4OMYYJf5qybwZv<q0UL{tYxZR2_WZRj;AmJrjmMV0;_<&?
zb*Yy20OAQW*RR%N(W$a8FaC4=Iz1MhS#m5v%yN?+OMPB^6w2$xuV6TCJ$~yH_Mcme
zL0vBdRjpP|`nIIh=|DT_hwyl2<D%tQf=v4R^jLJ0eh6MZrpKaF6OJYLq<@bdR|Av2
zO@cbgvbF`D^q17x(X}Tb_4LQ=sy1TQ_xZ)0`=8QFK{xk5j-~<U{x9pX=$s7462!?o
ztH+|#9*#ws`$KG${I?dftOKDY`^G1;pRln}9iP_Wgb?#vT4uGY42~s;!<l)XE<VWU
z98L&cj@DxdpfZZX>9F6p2+I?Hy5`oe$Jc<<v6*ei=>%z1bu!EEZH#g<YaraKCZCBt
zL=UcHH)<u@unC5srJ*hN^&mU+SunpbnB22q3zef)@>@;39rQmB<`P|X9_&QyQ#NuQ
zYz)HlKMz**W>TC9Q*H-Wwq@xpb%oM!=TKp*#r$0(5+Y9G8?cz`^yk8MUhD{BHNAN)
z+t+8)3-FaVj1tGMR8LM^e;nn{82tl7?DUlL1E&k$N4vJ?i@1b$IFGC<mRf}iz3AV3
zxr94YhYb`w<4e-{*73pcON-R&Jw~Hu<s9h{BvC4S+Z=J&#r;j{c)zD(btaOnktEKu
zbBG$2ol|v~Oi$-rhfY^;g3Q^1Oy?(w%G;Wv^4LIw<9aAB6Sdv1d^YP@%E=k$ghB-;
zDDUO(fFX3PLO~X{mY)U9dl44ZXx__LK&+kj@>Tl8@?O}dw>$4;OKdcbgvVkGsg5zK
z8<`#%ZzlKU+>p61KIogye_7_!X@AMcnR&X@n=V=Ntx!1Er<pKH-``wA`A@FxLOoRH
z5vn@jkDacVKYzI@cce$*#CE0XtLf?d3HF=1=2Ct_zNGWF%ggT^0zzk8ypE>9Ljzv*
z<d>|G%0U|qk`MXr_n_VW!zH}uer^1%TNrktX@i8<_KEJ|xhf>}8|29ihnC3Uke}lF
zheR&nRUy%H7{^9JqLUzd|B%Sn=tNkgmVNYdV&2f`Wtx|s=h|I2G}^zB`NarDMFu5#
z4{@s+if~^-*A`^4g}&NOBgZ2-!z>|U;tvk65Gj$*?KFx=9#0u#ltIpF^G1dZWVRVy
z_%RX3;|ROxNG^Y5$mYgK+6SWO7=53Y8Yw0?*Q57pNYv@jEqs}e&0py0LiJxYf*-=m
zf}i+=OC&J2IL*X{O~?+Sn<2iDI{853{MNR2IR<)_bnPsTW#VF>0x*P*fdpAR26`)K
z3?xiPOmeCu>Q@Rq0AlSJ=pcP!F%bLocE>>5WAS7%W{$-Q@e-<6<jfeVhT;sI{XVjp
zXlQfTXh=s$({a)AdVHc9sPod<Qx0^owykh2b)ItP>if892whSgPN;zC9;Yi7Fiq`%
zxs9PPmUA9+8dk~L65j4aO-sA1T@%@_TV>bCewkAvEzSF;=NX4czo1HXu)6KmR=(i*
z0w@X|^6-*AenOetxMe_zTeRllAHldRR7Eg%KqQSsFfXE=e*{y<9z-m2LSUw<H+-gc
zV5Yf?t5@Bq<}{qSj9iNH!CT@xP<&*)Q=OjT58?hOOo=-F50yr;hP`Ej*|hNvT;u-@
zr`F&2Wn+-1)HY)D*Dj+Iw5>4Z(+y0if-F9z&W8_DXGafHqWHELHz5hNQT-|PC=hE;
zsSD^6n^M`QR!pfXF*P_?Mcz`nML=!DIcT@*6{@BA*&PnYISd@{Fu*=F%D^vC>%Ur>
z)-1bz;wl8(2)9dSh1ogrB!0q?Pv@uQNTL|Hiq;RQuGGTROq)TPUN#*kk8FSH3~QM^
z#jg@~I;>(xR8|#nr^70KY6X{9@Jl1a)4CfVK$-{t5fc7sI-hNIm^cfYz{+{0btHSK
zk!2gYIAmiVR;3F=toBkK$AK!FrdVC#86$~5%f#<gt2a?D79pb;?`x|yQ+z{rR#diH
zGfbmI3l_6t&BiSw_L-cKB9mVdd+C2kOD3~&3onUabztk4&gpDT&Dv^pjv6_m7{}QK
z^hCF(Va*gDE8vgrVm@Eu=YcBL)YTQM+a@6fygTqf7ylC{RPk3&DW8Z9<2Ww;YF1Jo
zwdRbVgIvC3NM9ecdQ6<%#i}c|n+_9I4qCl9*px2B=!h><g{{^?`x_%`uy10tT^(aO
z3pRj>H=v@M?2DOnw76J`WsRK%1Ub#>O2xCzpY7IcV>b%5gRfHA6c4pq?MY*}Jc|B$
zQGA^WwBw1Q5g)@hofwFOLCg?+=^UgLGjfT15>HjE&T?sF)oH8DG(Im<e~cN$V!jwl
zVfQX;6d`*eHGw+Svuv45y|lAlD%N7pFP<(k2^gZ)YxnEYAX+a{fABQ6`iKV;f=WR|
zS@2Y`db07{Xc<!uB%Z|I(~Mljnp;HKsA4jfGzvy82{d$p%2ulbMmPdCA(3=as6wj4
z>W!BXlDl77)6!71o+S1wV~#RHu<Q%mwSpiXhrH3vEcOjka~gg3tAc5illj>0F;b;I
zV>gb98!@>h(~?9fwlmGnkW*i=WZH}mXNeI`<#b^rFu0ju#)Mfgi^JnR{IM9#pL9En
zHJ?5KiHZFHRQtl%P;W^Ld3Q0s%lUKDRBJX_8Z;+23PsGOQTD&GrEj^_TQCYSoR>3T
zpSbM@NM_PE!*S3~Mqw>Tl}ZJ3aP?|@ZAJ^@c<Yjj=8L1SrFrxuBLGjvO2!z5r-b2{
zf$vuxw`MKm*;^b*c)F)=-0qJT-P7qUQdz_h(X>$-VL0FlR;Nm1Ss?2y#fM=<$x^Da
zX@<2JIyr{BVYqQR48+U=X~-d~&m7N{;=5z1^k^!J|DnHaRHUq8%{0g2FtlsRuxb;6
zn#=yS|HL`XM4XFuaRM)^;t34AZEuh=6CqmfC>MaD7u1!~CCJ})1BakJ`9wN50v+If
zVGcP8vpnp1EImfLOUEafF}5*h1eF;ZHn3<cje~-efzt4Yta%AMwIgPrl6;nyOH>|!
zQ9YY_rV*zid$u<_WF2ATc2>5Qhpid5H-Y&=6Vth58n%J^fKa#1z(fVTnGK?#StU4H
z^pQkuVEA!sj`A&Sn6j_Y+lcc=2OX|VvhSzUm-grW%&9clt|f2^U66AUfmaB;L|`@j
zuo9qbby6!Tg-?Ch?GCXvhuHIn*t57tX(|0m9b&RPBh_hj@Ot@T4oJ0ouTeU~nSsT0
zXSJtIOw*Nvv<mxM>WjV7+|(wfU_`e>Y3fXy%=ZwjL}rF2{Y*Drf@Y2+a3g^O1b#^1
z1p+Il1Dr@;AA!Fj@En036F6Z8z#0O33EV_rUMIlO1TqBj1b$E8j|BF20URPQyBlCW
zfqMyjn!pK&iZW{m<d*=H2)s_<e+ZZnfZYUsN8k?xb}a?Cp1_*~-XgGm8NdYuzCz#|
z1U4Q6u$jQk1a2cR_gH`h1a2j87lEUW16V}hP6GE5Sa>`@gup!n?kBKhIl!?5?j!IZ
zfmaE<MqsQTph)1?1l}NU%>cko0>35jzXZxF0IntQdjfwXaLx$;8wh-az{d&ntOS@t
z;9~?T1ZJ-Sm`~tJ0z(8oL*QWozb5bofqAO|jwW#X8i2bAJaZzzzY<vgZh#92+)3bG
z0$pnXW)b*2D$-{Di9q@kfE<C<rvj`caMEc2rxDmZ2yh937fuKGF9Oe<0q|o2?_UQH
zBQWzUfY}7hvjKJ!7(E9dOW^D00(_gm-17kz5J(fq5qN{Zn*=_$0pJM&2Q~s&1Y#Ee
zBnd>26`DDYz@Cc$_7nKS#Q=XMkh%n5j6nA`fIb3o0wV;zL*U;D+;b_w{RFx%1L!00
zRRZ56aPj2;mlL=y4zQQNF$sWv0=Faq?jZ1`5rD4|IC&Ickid;8fCB{1*#WSDz=ABm
zVgi3d;1L2BjsaXmU|AkuIe~Ku02>J0dJVu`1YRZZ8iA+@a1w!;C4kuk-rNoF7J<qK
z0q!L5o_zpk5?F8}z+wVFB=7=(9~=bu5rG#!4Dd36FIxa#CqOqmv75j$RN9x-D$Y<+
zM~0F<GW0GZL+?5=v?q|EJ%h5<BhO&Wr}iY$m<P!z!_V5^P@p27vz@p6jr|;4I%d4F
z;e6Ja6xpLkdcJg??o@j06j)iKIZFSCIg6zVB8Yk3zW6T)`VmGY@?{E=5Md|s*{nS|
zA)24Qo_~J}y!}m;Z3qbp@gjl?%q?{15FzY`ckeaF@uFi)eC<T~j+b|fp&fO34#5>L
z;JpauHU!*EV9tI*&EMVaf**VuK3`g5s(Xc=btx)?EX+KM7TR@EJCEFtu-e!?FIM~c
z+qPW<ex(HzP4q@((`7UpeV0g(ugqg>!AmhAeGk8goyMw4aHpS3_b)|QxDr8}YdVQA
zaa_$5&+WkbqnvSir0mM5jD-P7$q`UfeYiq=J+u&i6;z1Rgb=51KWE$4_io)t(HoQ9
z0ZE>|Vzjs5^g^tuA^d1m&EH+-dmad@L0xDON`o@5LR=ENTm)OzO_dlwJ1Xd4Th(Vw
zt*1aWc?-^ae`wy1YZdSBmw6vqt?+(DUEZ5gd@i4+`f&AUeP|h;9aM(Gcrw$`LMY8u
zQzcE3^v9LqBpJ@tp&7q2DC5HQ)-s&kvhuvVMp2$!b(N<DhBF?T_nocc{f9E|m%c~g
z{kh5Dy%L)Dn_I>E8#3>|Ln!WUes1~gWbppE(7b=PRlIMUCK>ylGZo%9VCQ$jC(3Vx
z=KbGV#d|{LeHj*1(zp*#2Jb%$&HGPV#rwa=yf50U@czZg;C;$x1kQS|r=g%M#QsyO
zct80FN$1BuAn0HL(ptLKRvTfrz|SuX&HDnhXi0E*q0IZzA?j7nZd370Yh$8QLW}LB
zpknili8dYK9}^vBV!I@?*tRw$wk<Wp=33e$g_7COjHiP#UN4l)$Y!>GSaC7w$#5|T
zLi4`ARlL6}^S&^n@cvR=-rqs~=Kjz!{8UgG`udw+)i4}ahLdDC&xB_DAA&Mo&v5p%
zNpo}GE=756XgiE1tLH<@@CQL<=&L-3+WaRgSB8_MJiiUi_-jEKuctf@$;xy5b&B#l
zcsMCf?}LK&puurBS_BFZeU<0Y8p`9!aFUc~AT;C02W7mT@;s$co+s-nPYYHB=Y{6|
zoL2F^q+Rm5?FWSNu@%v~=s1{mMf47uvSXoT_<^7@^fjCn?fzcZmEj~A&b6T#-xZYc
zdWLg>tUPZWQj}-o;iNpbhL+(cg38cWc`mG>Jgy8UNqHU(&G;ih8Ly{2hO9gfeMC{7
zWL@QHf!F<JXx_ipD&B9Dd0%^*!u!6-;Qi;Jd4I80ygx4U?ztM`vAVp!gJztm4~cOL
z8mhe&REEAz<LMes!<C^;hhfcea!^Iq1k2v&cD`omf8FU^B3n*Uy5T5g{&giOO_pbr
K(uR0>IQsvo<X7SV

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.doctree b/.doctrees/autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..bd607e81b6dec91113af3fe87908c156073ecbeb
GIT binary patch
literal 25674
zcmc&-Yiu0Xbr!`pC6SVB6KNzlv>lr(M<ioaMPpW0qRI)}%BJktt{qFtXt+Denc?ir
zb{-;GTc}gSj+xvhDLe++27Lx7P}G0=rwtG^RvPzLfdXyNx6!80AnuPAX!-!$7X8k>
z^WNFJ!(ECJpylkHd(S!doO922=FZI7mq*|DXk#D$$7Z~?ZrGcPYNKI04b2PLM8l~E
zEzS1B*Ta*a32%foHtDJxo)frrEoA%9qpllP!`1BYYRIOkyKh+DVl;l-)9SwA*vW8x
zR3FpFuZA^!l8yR?Z)urR>;bFQw49B_w$aus!`2oBF}I~OR9|hoPF?dn#|?=B2#h?V
zZIWb_$GsAYIgGVcNbHCD5jLqLQ`CpqSTwU0HbVTvm0(q6Q>yQ~#%kb0Bcy?0qv6qP
z$Lp5rd1%D(sI9id(L^I?r(agp`nsz%sZYSheRYjah5C$sOh2d}*N62Jtg)m!Ep5pS
zY+G}eHXL_->G3B%0X4*bAfUG-1Y1%A-%;Dfl2FuAtf(cfsHLK!mJGY0Z7#YVrh50q
z{lTifkB#!SjMjxg^bcTa`aRI=efWPr{?Ft8Ntz}!<r;i)H8#>#4L4+qEvFG!+K<dt
zG#FbD0~RXw*T+`;P!04C=?kf*R`k1J5B-2}#MF}x#|>=id}=A~BeS`t$o!yqUg%y>
zRJ+cd1RztL6W#Sw(BcHw;$dH14J_4F>W-y&ZLPiog^uw4?C_Q6Uwr7si>GhIe?ww-
zDP_lkz3`0Xht4dYCAac|R!eoa&cRinW;PoI11zz{%;}IF$$V3whWrvfIE8<fv$m$W
zQ&aIm9W8c?d~TAB`D?Bdv^%s%@nIag-JxnHFl57M_<DG8inM7%o9i%`6T3(+XU`CL
zAxnUw)vD9j>JSui4!h1qM>`L2v~o&^Od<fOYp&YXGXvTA<@VOHvuZA16AGL=TWz^4
zitgvGUEvC^VU-_jYj(r1*OW~}pDd4$4ay_7>9}S1nyXqF;`%Q|i*8H^6s8OA*Saq_
z1QjkM+uz;_*;spP0V)W!Yz!hx{0dC@6@=rii<OOGs)&^#JJU~;41sJkzNH#=sK3d^
z@y*|A1LcNE9(DlVwavC=)D1s;ji267T|3-nldzfyZ`F~AO<)kmk8p)E_EmRH^FvlK
zLPjR%A>$VSBDXD6TpWU3-oO%@)yGnEzmYY!*41QW*VpwE`nM2QiO!d4n9RmB;EU+^
z1u*<lVNFK~N{kCVm>$!=U6GMo{B20{YA;HkMJI%z?v&UF#kvp>jR_j)-7J!1cV)uv
zoaNYSBBrXvb<)7-YsJyoPHb|sWi5Mcl`^=AwySxXdtKw6^l6NZvF-|ofmGyN8l=Wq
zLZLOv8E6($pcSP!32sWDd;CI_Ohq9i`JOnHd30#szN}w$4+X>@xNXiPBe)2M;WbYd
zyrx^)qraeECNjX|44FMzXQxk>L9(vIva|aLWk0DQ05u*(W_-c1EY)}1=L}ze64^Tf
ze+<4_JSIXuGylET7c|D$YCZ%74+yU@jaTx5>7UPkuXSJV%t<g!s*=iZDy5R5>@UU=
zGJ9_Y5D@&UF?;hla&x*Qnq?=XvZsJ-ekxfSWl3;N|E=C|Y(C4a-+dV-h^&7t3$A7z
z<ErDU{tf*t{pO1PP4;NTbV#lPTGe|8nI%h^pw4`X>n?|yye%8x(%>71+xVi4<^bR{
z1F|B=PIHA0tHGyvQ7|76LCse<neO%{DS_4zJNayYqvm53=mAQjh|!&15lzj-GQ#hb
zL8tj(S=Y?vusMnj<{0(D>;RY<bTCh%xd@xdE<uW!nI;Kf5?V<BbLI(t2|jnhv3*=;
zcp@mwvX1r?K)*@i@lA)aHJ{B9Xl?|c-$+Jq0DT`sT8zU^uS==izQLFWOv05SdpNqZ
zsTua1NSZ6|A)MTCZk*dh$V?^LN7%<ku*cPT;ZB!N({x7h3~bl<ft%Z-imz=b;PNXi
z-j%k@XlEzhCeugRe8p%L8?JKy)S&;8n5WF3HFf|qSU`ryMPr9ZQhv?Bzvr&7a&B#Y
zkLVsV|0yBQt^a;le;=FV^DRu){4RAB5ELsY74v@+I^|+ckeEWo38LOhy!X0^$L`2%
zT3k1@jXJh5LpBp1Dym)Av45Cash1PeHDP>dwxkG$^(yYZVOJ2NjYi0|lL14lF7wd-
zf{KM|(x7I(gIV&JfgMLEC}dNlCR~eROErIa%*LY2PyH=6DlRJZF*eZ*EQ_mB@SjqA
zr><b*fnvp5yzNz7^zzoU8_D($q~8<5=5T}WMCRk>Kgi8r7eS}GB8lvSV#+Z&{X}uf
zgmN{JcZ|@6Z*i8<#ku(r5Mtxd1iwBvAEvL|akAfSXF=?AZnJUQi4Uja<^{1gv&{}h
zC*$u&&?B0jmvox3owzOTN0f-4r};i&TewYA{F_Ksap;Vu*k8~z{|eODC22M@SL61c
zLo}@xZV*OyCykOvSMzTndmbrj=09fIdS$9{c%9+?rrNgStA#rW?DXZc_xaeLT%Mcf
zT9`j~YVIuWkClR5SzO^R&m~>v=Fgo&j?BaMY`jj=AoXeT&I~waQ9#hjcsm=Ki^DdK
zQX+bRNIo}WLmOc+=`fo#<d{k9Wjh4<0DVOadOl&t7whq0)qFRr<I0c{h3?Az(;44L
zL@15z=6^yS^S|)X8*WNw^FNu{m@g13fiaaYbH30+ym6d2C-5N=FM^vA@uo2_@8n%~
zopPIe{UZtKpd3k3@rq&9d>VrUl9(UBp*e?-{Cq!%7GT2t^k>X`fI1N;b6&L5@vEVE
zGHx$`#BuoVv(%GcewZJ^VO}e>K?2|lzO$~`#!qSPT&-G+Ta<^)<aQon;mU4MTudYk
z^HK55NGUPY^^S^@8B!Ff%-jn2eWE8;K&E557~d}N%ZHzuxdU&%nULb4<!*$Qt7wYt
z%HHq0s_iu$x23t>;(6RZdL}kg6l3?XJCj)44(z%gbOi9B<s^zOXQJqzC%nsI3L9#7
zM$tc|A;-)=#ve$!mX)+isQzau6Vt=#Zn0W$(Ubk3D2bevDkZhj_obY(mtomm6Ok7s
zcP02*%>@;iMflwdx|e1&je7o;1#-spUIbtDSl!qliItgVq{U5$Qb;QCn>~+`Y+f1w
zGRw6*l%{j&2&G@=Zb5l!g-|MDD37MchLdR8EAwHQrTw`m=+f{)(7hWL`n|}L;PRb8
zw<{D8p(q!R3B&;^Z|o=+pG*Q~J|Q<B0ekZ_&FmZ<{U9BEG&%Yhj<(J7{PRco=LP=x
zDF6I8|NI1fl02EjnFq~XNSvQ<i=FUWfEthQ3Tj014U8G%387BJ_;J`ILX4za8i^jK
zvzeTC0K9lCA;t0H-XdO{dDg~rK$lk?$2v9V`R=K?9Ds-ieWH68PvexYt!+iwy7_1D
zk_33VV8+#ie;H<&e-9EJ;XDG3Phn_9&{&sqk%5NbQUV$yq>~K#ZAp1_g9mdzhNYY8
zo$w&o^aTVvP|(jxg_ojg#3gU+my!hGrHqQcBImM8#{Nw?m%fZW4A;+0<aebk`nEOa
za;r`Khf>*h%hZ1<=dw$t{;r%$U#7mdcYw=y!hc95=-U&j2r+=8<nx8Bl%>gGFOV2{
z#travB!{soIs6?XGW?Rd;0%T$=;x!&BT2M*Bx9O;<XmJf7;&jGmnAtDnF|Uoxce*g
zm*P8}$m09p=GyyQ{(gBbALl;hnSAPS9hv+EEV#;>BAHD7Y3MZC7wMVFCn;B)$J5|K
z9)B<L_>V`6zwbOgCNMxAZ_>PY9=}dUJdfu`w>yu&+)_6cOTz<C`YlN1FYQVypNcqe
zHb0f6)tS)$42%;c^x2;2n*S_y*qxmI<%AT^>1)ND{?Mg1ZON<FWyAJ&lB~ax5Z-aJ
z{&N^wF<JkroQo`?3N9td`Y>jifrj6fvdH3Lb}Of6uFkpi1<N~7&mTx-mx5=+B@dqE
zsrt9&Tx8(6gSq^joJ(J(9$9hl21vTF{#DAOqOGHciL<wr1)l+|y+c&*wW@_Y)swL4
zQCYx|`q31IR!qW<%elz>DB@Bj342h^Mdn9>3ngI%zo8S^fIPUl_C5)#$&;`QWR>S(
zsn2!fVJ~7?^^=FKK&@5tuq;x>sThqfq+&IsVmG2y-ghb%6Br;B(`afu6>HKFPsRAr
z?M}t&$hPskh|&ry-)LK!;xunTO7`5Yq-33<4xE}DEu-0)qTPVeq7<!cM2aXsO&xb9
zReLcZ#Z$G#T}#!zoDkk|s>U$1VygDbaxSvCE4cJd)!vk{s0iPk+p7JiYCBNRO{wfs
zh>f^ZN!9*H&P4{XJDAHmaxQ(Dx(K=NN_kYZ_5T41a(7q<$kc>_b_o5w7WGauwTUrV
zypg)m2!>Y7)Mn*eWNs93sgkLEP|ii>MuJO8rbZ{S@py1^?R}>9VR@!jf?(wtTk3lq
z8QXQNu6{DM$DrP-8Cw}}#+e(@DP(So$lNxgMc;Sk784jCb5m%FJafB7M?7=mN4Gn3
zQ(O%XvfBz?6+i(6?-pcjPwz_BmKSs2jBUC|uQOZQfT5ynt=K!w)V8S8?qq5~LW*Z<
zXNs8`zjr6DzVo16{;bsJ6LLGw&tApQiuu{k%ely+t>99YpXG0b{)UuARm`R*7;~82
z|E*9V_zp3kmvu+x+y49x?r^;=)ru6dZwm9~ov8c{*PqC_$RIo7QU$WVBj+M>BEf|q
zo9i!Km~<k$Fb6l+-b3~&Ib;_?SUFs$e%ArlZK$XpxE>j=4%fxV8ACPEC_wcYK(!lL
zt#7D~2@C+$$7p&Ss*lqVhiZOwyP^7dMX9Tnr6{$Ny9CoM&DWg<o_2m>S1>x3Z(wv@
zjw#ie8h_8uTb_H|@che<8g>6Ne@N`Sy><2K$+?AdbNRL?gl3<PW)m^F_%43hkOqt&
zg0Zs!V}M8A9!f}YBwyz5M;+zQJX3iWXkBwHOeg!^$ldap3d!U54-+aor9kf`^0)oy
z6-I@>+?Brah8M#kZ6|t48d5zyl}xTIuV6!V=QG%sXb9f+#UCu}JnobyFW00@qz~;y
zT&m!XE$1S`oruemppZO(cxRc53V$rgaz?0k<}T99QepZ|a|D~dZX&$|i8}0cIrfdA
z%NgNAsqj*J2`*&{qJ^E|@#|6^{VRyGxz!5#9jWmBDTrsV>B4+l%A#-m^q(h}`p>Dv
ze@<m=`xjE#E2yX1n;mj4vi088ONRkOnTdQ~%A$WG5SQKaQSnFyvgV<y8QCW$WC&Qr
z2>H-n&SiE`F8NH}yaJ7PJiwhpocw^S4f4+-{utQ~&s^b;tV~imi2VS+-=$DRNP2%-
zAUnPyu=AG}E(nUH`G9y|TIBFv7RRT{5}FF}@e!$Fv(w(uBPS#M$H#{uQAGfGN-Dh6
z_XL*`1r_3>B4yFHe)^A(rRph<kDgTa3hJp8A3r1KB8!jRjgXIzUy`!u-w1uj$5JDd
zN3O3*Wv^g_s)6W-i4m&VQ1sw?i}7WIx%Ybz56Jf*k}Iq7-G_7v9lH;wFuQ(sAHD_U
zRJ~b{Ugt&o4y?!XD)Dq`C#mdRw%)}u46+y*r0?B{n1FP5VoY3>m%UQ=*EDy&HSrD|
z@vRAdv>RI!-My1k5xo}}zs)V3K~!oa|0X$qC-8g81l~<1z>oT!0Ix&TQ7?qeh`J|9
zy$)QHM%9bsQ*&|I4xF0DdPhNHOtYgnAqpF%da3DW<BHM?ae?Ip{({<q+giw1qSg|I
zRclvw$T6H*{3y|!-;l_K%D6hUeu~}6UoJt>KRkb{pfL6t$~cPZ)T2HsWmw_GeQYZ3
zqc!j~pm`6m>3ZN&?Q12ebbT;x5f!i5L{!lP<-d<4^}L%dulUkHf!3c;tpL@a;|&Ei
z<^+B_@WT~-2Bl2bG*q#sT7*>B6uQzpUv0JV<$&urK14-DoXwC8Eg{bpB2Zqnjt*1N
z*TA*#?GP2Kq3YRm76og*$c{Ea0A)b5O!I5(5LFnrgO;*M@1mnH>ooR_j5g{rx2SHp
z2U5pGi)LxOMcF`=M1>a&!8E`r8W7c=Tww=Yt?jrzo;km+L6Q5}l&)II*HLyv+ccm*
z)FTPJxYH;bZD^~(8t6?^Sy(E)3%Z)xLdzr!QP*fPB%fhJTZ(3*HWAg0VUvO1T=>WW
zFY$9ce^Sw0*KrkHwHua(vfE=MY8_Q#XN1l(R7N?Kkew);DUdMdCvk&hTF^~ph({mK
zpC}AwhiMKT6s1E&7B3)r$WraKKwZ<IajLbnAC=(Q9j;b)+=kN7s0=aKQ2Fo`HipI6
zL?t6=#Aw8>fJ#NvRN7QYo7jF7O~M9N&}u7*GeWVv0S{nnE}l6jf-}4&dCw|sYI|DH
za1<0eMX$ps+v|9mk{!vtr0y%%4bKqHI-@iUkE&T?B9WBi@gl*3JBZ16DD)$07{F_y
zZc5XzFgesopQR%(K{-{vDp7R{)EY4P0hG*#sl?ZdL+l8@k`S*n!24T<A3kw-o=vy4
zHty3lHLOn*x`*Vb;@pO}Lv~se?0vfLx4q@1B{X}B?Jex;+KZ06hP8AT_2dGeC+_mL
zU>={>ch%Yp_kZ{d^f~P|>jm=El9GQQCQl)bTy|Y^7Y)sCavlf;OS<2(z-$twXR(TS
zne7R70zTPNlEs-!AS#-(%?^88wvTr>bYo4o@E`s=LIrF?cF;q`abNk#0IS-AqVC||
z#aG30unjI$d!nx6va1fmi}eOAa}{FrI*NyH^B*4Iza|icfo*t?Q#b4;d?1PoM=4Nv
z!D^wK(IVfq991kvD*L}4!ev^kS_3P13liuW3NvrByXxq*u4pI_Z7F!QoFXol`bhL6
z2k=#^DWcgn8`kXW^z#91LPQfRUw9(QP#IVoKI*1Zxo7S-M<5hD%Ll?BmN)3f0&axt
zXs(sEaCtMRHIpw?@S1*$0Nux_Q5@II_Xu?Vn?4Rw^_@xj_&R-jjXwUJPW&4_0yasb
zco8f5VFNyHn~&M%{kM5vT;kqCXNe#x`kqC_^RcKTT8sgmHJ8sONF2j1et5tGT!*hw
zn$bcOiYkna#C3v^Zjq$=IUy$1?&V_g%6+^V8ZSu4tJ#=T?Z%{H4CYPxpqIH!dfCgQ
z$B9jP+}NbYR7`q|#iTo6Cfy-3DI4Vpcfby1J(lb65Gs2MK!sjK3k9aQpD<Rog;o=9
zg@?CR@k0Q)u6Gt1liriLZj&)rPU&clJ}}1hLqv?binO9L0C|MXx)V@P1frsfv<3Hq
zBLTg~FGTIH!@948pF{}2?gfGge942?3x-Mb%rzfKnz`d1FzmV&G-#o$QLLM4P=JqO
LA>}W@>f-+c)O0od

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/utils/rm_trainer/index.doctree b/.doctrees/autoapi/lmflow/pipeline/utils/rm_trainer/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..c43206867cb54d6b3fb8629142416d5bbd086414
GIT binary patch
literal 20525
zcmc&+Ym6P&UH97iw!Q0hY?H<hZW<Y1x8B{vDyVE72O24YI9^QDlD6JtGJEgbof*%)
zbD4*|s}L00CiON)ZAvB|8VKY86cp47l}b=V(})jz;S)%}2Lx0gK7fR3s-__X#P5F|
zGiT=BJNMz2CEuMna~}Wy|NNil{Lgu2;-wE>*unp?`LJy}t<5E~UT^tzD@@pQ-LJ(>
zs}&_LCMSO_c{-`GnZUdf`f*UR61D?1YPRFm1FMxhp0GKp9ywmPB->AgRxNV;R@&U2
zuqW-Q$CIi(!zLmp@+|EWt9Z?Z=U-WBJ8jEzTGo>2ENB{0VB$SUs4bd~f5O@%m<DL^
zY$D*EY@2{GO6&zTW29)=yV#_}W|fW8n8|9qZn9Z3iUManjz9&%dY4lV3B#$HXNDmP
z@jTHoo8oD@9=EfXb+dLUuo_e+W>b;5L7x(P-ac&av5(lh?4#_7W!rCB%R$@%Qp;ET
z;L`FV=N|&<>JI=6mjz(UW*qru+gTRGSyseZ=EPafh_mdp>el8`5CX|#=XS>H_6|0|
z%QPAb^4mWSr0lyv%X{$uIR2l&|C59uXcRacttuODn@*6hYSXXBp7pW*)5A`Ps*C+6
za&k3F%-H_0y_nHu)jkI5+q(s;WxTRa6|kuXCH*-g9LnD1G^4Md6*LxonLRP3p=nCa
zit6?$&}^F1Y+q!q$DSD&HQzJBwpAN~eB-=6+qd?qr|x_DsnbuZzX=WZ3)yhYzquT+
za^Fv`oFT>u<7U$gw$6ejKw5T4ia5k)N&B3z1?|lM0rG-boyEULtt)0w|Adf4D!5=Z
zqC%k((1~IflJ92NWV8|Zal3=2q!Fe-@gWI5jV4!|dSoZ(W{Jiv(D)Lj(N`?fkKwfr
zKBIR)`f1&-Z*>K9SKwdi`eA~^r?U-kn}Pl@Hv+S5YYo}Yue7&T{B?Kb$s8doXZi`y
zQm+2lC)YUTs}T7;ZL3vxS{ueDl#o8w&)ul8)Fe-NPGkk9rwwAiA_a4A4!$H4&)0g3
zCln5cneA+EC2X?2wFt^2mad;@hJG7U{2NfxJw=;M0&{54gmE4pTz4k0Zq%{~W0Tl_
z$focXZM9+Pb`d0OH{Pwyw&&EGD7nf%LnK<sHfDEgMX0_(Y!aJBb#6(LRX${71{+oc
zc!!&Iu2sGYdvY7J$+<C1!%L8~L-u4gyD#apYxSHVn#LFHqxN4w=~6#0P_wcrr3PO@
z#pf~LmkXjTJriXv`V4y5{+j^@DhGcX&^!ldzL<xm+8aFfi>Uy1XwCryzN1Yh^e-zI
z*o7NCgpl=D)d2R7q<k-QfTPLyf9qV3m(!~K<^baT!?Gz##oPW9U0&T?V>@;XF0%G_
z?6>UiuG(+2(|v`m6IbHL+bMY0ih_4dCwV#NLh#&!pp}%p{~T2I{^R;(&pnQD%jQ<e
zImRkSLW*B`rO!h~;SBVy#Ky_vNZ>4EJ{{C<6oF$0HOHvp`K=A%*_cJER7dM4i>>ua
zY-Y3Rt%PlpBB$wgV1-t2+2Shv3A9yy=o)v_GLW;7!&J_?UL^f?f`1mh$N(etW=mY@
znTO&Q56(g%<A<15l4^9fPMpG{mH|C#*GyY*Fc!cz7gGJfqraXi+Ww4vkx+rBtaWqK
z&D-gZbaR$%LqW|RA$PatH`@pZji!YNsTNkP%cf^w9_uIRE8;NE4?~KTEbmmsX`ya-
z@l++SqBv+7ejHK3<ESghZK=+sgX-qrt0Q@^$^~{m$K`CKK9m=%`gpbo*bo+PEj334
z;tgQ$3N2vmDr=nPYn<q7CpjIfc4qt@xsjt444?Z4#MQL=aEQ$pK+7zj`)11f-K1q8
z<L7>j&t2YZ<`ju?ZwvPCTWB|m{kylX{VU>2B6=&VUl+L~chP-=QwTG$TN6TU7U;XK
zvodX}RH$0-AWX;}{w1~GzNq`EY+R<xsTtf+_Z0!ib$X=k-}7#I&v<^ga%^B2-IoPe
zLkMak6&<O^1`tqLQwSzkduvMXtzE>U(0pr3^{qYLMP$!h&$vYqeV=Adi0G6APGgGy
zqpv*;*<hmiUZ08T<(f|K98EN<eMM8@dAmjO9l>&k*xh|Cpf)GeN}rQ*)+Rx{3zU?C
z`UBZ`5JCOE0PlJQ^}`r_@Zrj_$pp2$9-LTdihI{9`myda9N)Y~anGZ$o8RjGN|)NY
z#c>})1@}`_3yKfks^*-zPocQzv(7Y9-!k)BbA=^-D>;lMe(U^sz8YP5&~HUpZA-n^
zj`GZh-5{pvDf&95gdbLg-Fn)B`>=Nbq@}zc)YFH}FQHF{Sc=19Dmz;+qScd?s?(~7
zMWene>n9U7yqUK3HFgapxy-(UjboFib9I{*5$EWO5jxnrfcu);q`9kzNahDij|^*&
zygBV7(bk9BJF+L(-F+3BQ0mYl{cC0u{FQXNy1HF#H!xX*tIok;dkH$e{KnsN3ye8d
zcfU;i9(JEkyXV|+1l+KL&2Sux2zK~}{e8wa(%!oRKTW_3$fs${en7ug0`z_kG%ZCV
zim?LE!Df**Ju~TCifK@ckTxi<%n3_eV-(3!>=?mjHdHo`Z<msSjkS`Z{&L?E*!~&*
zlxgPty#A?Z^sRv#u@5w3FLDgpAv98I;1%@ShSG4qDSGVW6n2NU<#O4vuGA19CTw0k
z8fL3z`+-K-emfSlMy(c%PZ~mjgnf4hv0sUHPCapd7ocj63PJymMd$@5W(LRoENXC7
zj=fM!aKdJZE3qn$4c`0`J)0CDCH7lvLM(FIlWe*XdmiTsG5)L(`8A>>De$*=*>hNa
z=cR(z7m9l3B(0UW+XU)SA%_sn=jfF$9@Ewb=kA&?41-gYx_w_zKy_yyEmD9YhpJe*
z1TJFbq=>lOIdErpk-G=t@XZ5vH@$Mv)!$qGqS&0>W>YO+J<h4(MX{5#&GyKT>U{w<
zB#L2LX};yFGF6XyqVYm2ieQ5Tn|@xBOPM%{=4i&veFnX;pOP-<&<1UA=eFFCxz7TO
zym79&U)0LjDFq@pGh3Wd5sC^>?tjSiLQ5_yxG(4(3}B3;o*$rQrtX%&7`^?Hs{5OO
z!F>fk{Y}(|(yq8@9E;c8*L4iifT(wXbSCYhs~vK`PSAX#92#kRx<K>CIyCnTFcHF;
zaKEj$o8ayE7x%k8K44iuR?7(eXeny$U!kjBhGi8^hc+w^^G12tC&iLr7vS|TfL^_3
z;Jc%P-?&lWAIL9c3Sp6Rl$(mVd_`UbSQU#i>^?g>cE8H8gXUl0nx8xmzl|;FC}&XF
z*!>C$JY@s_dg;Jy%<uFzUZ5t2T{$kY8JreR8q$2I39d%efbuOH^o{aC;X*Y`xEcl)
zmt!jr09=W<-S&gX2$9nf{c`M6dJ=a(Uen>q%yx0PtE(3zn01gw?ms|Kba$tVod1(1
zmN(mzISZ`2domedMLX^uM<e0MxZek??lt`6_3IB&0-xj^`ZMYNh$_)Y?z^I#o}WwH
z_wa`ON4f0Z;hY=;o18#R;hebti`sb`)iZ$9+!58mDDH?mPcZ*8{p_Qk@6pdf`m`TU
z+wL9w_W=Jr$bS#<-yh?@chWC`a}(T#pUx$SRI1!l#J0&Gq42P264&cL6p~rUgi??^
z&y|kZYWa~_$Z}+-FP^z4LYCD^<pjs<#Mx7oGpUl|yMQZ|v`Xc~*|XSM;QhjU39mWR
z8Si5rbUy)OrvW-L)M!UhQ$S-CrF`&|buu}g?<BuC#Kx}Vl?_e%I5j!!((q-wsCljO
z9szsUeJ6#PJ0ul2B)7t)I$Vn7P%c#$40M;ueU~s4%gqG_$%uZk4>&{kR9(QimAaO&
z?XGJxz&dp8H8dJp*V0Hz*V41pwVd9>m0AQ?0i*o@ReCDfs-0#BvEv^%k-584Nw*x3
zRNM+1e$cdna7pGhKAB44l&r3a(D}k_YI~2z2I$pJ(`ekss3*ZY2fBfw4f^TApn16+
zY~SLO=%dVu)h0HVQd2{_@-0OKuUJ+PS#TnyM##n6tUF$qK-xOwI^lzuJtAqDZlU^=
zDMsnJRR3W~jZqh5Ma=q;T3g^-_E?GSyV#hwR#-f<{W=z9KQ>&$C4Jl^Z<Z8h%GUP}
zYwy#0%Wu1s$F{l}V|(;ol;y}5X8P;}Nqz-nff<L15n0+5pWNdx5Vti({Jj*OZsBif
zjQBd5mWfy1K+46(-{OsKSd93c(ZT=rjRODX#fX19I(C08==>F!3FUzI3JCwTe3Vzv
zMaMFMj0oi@uj`}qi3I<ve6+6@N9!N{2_OcE<j9WZ_;8R$9pPUW5x#{k3A-z&l=He(
zvuK2?T;R?C?r>GUfY}(>RpB;HM1~)jNWB$Sa03gf^$Rj)JklHNYDXA(95!!TG72>0
zXkn!4Kxys#WXo0UNnzCTYuX}+ypM3?LF8$A;z1;T8a0UIwyx_EdtozHt;;JXuM`6r
zH|2wQ9tA&;BJeXQ0{m%I1o-~&4Ol$CzazDHxJY=E+~vI6k?nK~k?8MsbV>c@k*oVZ
zy7IfB0r00WYkC0ePCz6wZQecLH4BaMzKXlR1ztP*MtVX<yfj{!SDH8IGUEMYIv#gu
zs<$ZOjniSf`7J|5yj?KRT}AgX8cIYhU6>&A5<k}mq#>f-E+E}P6T>m?Zb~x1I!wt0
zpg*uFQ7eo@TNU-vvx<7@nK)C6daJMrkD_>>sP}<l)Z0B`X=6`Wow8jbr8G!9{51)L
z(#l#q{LwTX9%Vc&QvM?vL-1UWWxrmaYccdU3xMWT!C*1;H_%6!ldrK!E{6Vdv>Q5x
z=92byuk&C%F$SZg(e;PGB{V^0dW`C;=$h*Pj-=D|Mb{7O5hZO4cV_`Conz^Pj&*Z}
z%XvkWG{Syh_{MopFmLV@hKgnTo4eZ(W&gXJ1_fk+U5AOX|1Ir`8>PkD9cAyB%v<t2
zYNb*3`<$8c4Rm)gZ<J+p^Db|6!=mgvM+g4^nwII0fx*9dQTDx~V|R*U=MEBOpDiDy
zEXtOn3=?HPQa;)UQMLdgA43huhl4cgh_btg@GW$K*fH`3a>_ZcUp2c&eu|8=GoU-%
zqwhf^2X>FR2PPuz5Hl$~B-X|K>=A8O%E9l9xN#qa`!7;!jvR5T4wMGZPkjG~`+35Z
zN8F5_c*M=0Mh#fG-5Vz2Rt{z<2E8NhucipRoFc%VMnyo6xHD?9d68_GX2f9=Q(kGJ
zK2=m=96p^8x15mMQ#>t+6z{Cts(nUXEwiZZTUq4ytt<x3Md`I*r9=(YchkCa>|G+B
z<yIMYwe&V?=>;|RDY>WO@!e*eg7Y_Uid0-sGZA4E+Dp#uV6%Kb33aSGUSq;_jLp^J
zfUdtW(#vl4s1kAM4V#vEg1Gf(A-%?@5%65fI&Ne5CCWz$(|u+(DNZydtM)wZFxar*
z#__Zpywovv7@5sBZa3H+_<lqO_8QJ+!p4?yiGl!uQfV)t!mNCa0}pR|DJzsN(4nu$
z4E++@-#`c0VYjs6t86b_EYymd#wP9G;Vy=`5Qj3`xU!{5*A<26>aZxWJd2kYSDeT;
z_|7`ez(}%zyiaD8?FlTNTjN=mEf8`io3%~k;mFqswqR{KpdYsS<1j0os9Wpt2Kt+$
zJutd<1N~ZN6D2d4h?+&n5PUrQ%dlEV|D|k9HWNpU#g8uXT-qb~kA@Wleqh*UtM1{>
zBQO9RIK}l4^MdADFB80%gdHt>DRklf9fXk0iGDK}RO=)84~52TAK?&!P&P>9@$<`j
zJ+rkDn;RA=PU*}$aRC&&Bd}_IP&eupT|t6j=pLa}HVFZ2AY~X7aq5Z{kP<vcC##LD
zi0wpnGLrM-X4^=a5rpLh4i0++EhK27n|WTs*_>uSD{6;UT=xx}vqr6bxT(VrEkkd~
zxuk~VDkpR(M+|)#btk0jZh(k{a!Lsa1a1$I36VZ4@;ksavYOFwJRpY)lMc}n2El1=
z$^pULP;Yfi`EJ~lmA;-EV+$l{;2azwx552Q+#zy)-w8I?w%P{d!-9O)IKgm(OB>t{
z*hwhZopuzp!<FS_6o*UgEo?oumi%A?vUHGYasW`%h^!{2C*sG;t@h&adp`nt&IOHH
zp?fOHy8j8)J;c@|mc4|4%(0>d9|Njj*^ZhX2AjdjcSsSxYiF7r1y44OR5()vWIoC^
z+ZS%NB6HKQoekT=fAH@(ou5zG9-h8t$o#dd*w@I?*>H%uW=hf?t=TxJP{YIvc>|R>
z2~qnJZV=k$_l@yO#H1S72KV?ir_}%tNWHM16h#=X7pgf;;$6=-g)q|1UzZZFOmp3;
zLxQ&efo<VNmTh*hhFX^l3wOeJ22Qk)#>FgMWef6dGw`a{5ZY{;?Xp^z=>{=2Ewl;5
z7o3RXZU?e~1r~nK3+I~!00qv9U{%01$oeBa&i3a@DG{C*qqj!-LdSpUIx<wBqC)Xp
zb>AdI`}_3s1RdA9NIxbeuRTFOZ_tNV@e{KdYQ<B->25RLZkxB+=JmIEUF3b;O<$=)
zTq|$}R~SvoQwLeDQ8B^qOlp0Vsl^RsAxv~~m3+gaQ|L?V<&Ue&3gBW9r1n_>CS4xD
z!Q|(a_^|_i{D!CPxs=xDUZWpMI&o>w#-%+SmzLaIT9R|gyL8F>jM-kDJ2}1f;^H1*
zJV8TILP|*06BDDCkmT@EaC37V_krcA9w-#1wP$kGX4I&>r~SG9VNZf6fp$1p>`Q)u
zL5GU0`7tR#SWjG7;enOFG_cfD56bcvF_X_Gk3!sWqymZqd=rAh#ek`%%oXoWi#e0-
Zc3L$ru9Hw~kY1aulQNfjPpmX9{Xg`{G7JC!

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/pipeline/vllm_inferencer/index.doctree b/.doctrees/autoapi/lmflow/pipeline/vllm_inferencer/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..5b6a14ef087bd840ea8a2ae1fcb964a4e2c39e73
GIT binary patch
literal 92863
zcmeHw36vy9b)e?x?&+ROqmf1<&7qWpG(F8skI><jnqkmwT{99o2MJMAU6tKgsp_ht
zDr*iv_Tq;?Bhs+8N=ASd*b8hS%)Mwa76>rU2gbW#V7amJvNpyTgU$MS?EW7|MnqQR
zugI*P9`WAJdsCei6%qf(zvIV||90fDWv47V1^>nCJFTGBI67UfRvXQ#-|5Axs?ADw
z!EZ#pM|(S8(|fQt8IQNiM>@@JyW;oaQ=mpAsMV`&ztOwD7q7$WQLWyY7VXD6ekH0k
z8%gtEBp40G?(a<o<MBvTi|W4iQ+!T+VXodhGTo}R{Cch7Pam$=7fQ9poDTphe!GWF
zq2<au{iBG|;06!%IM$=BGISC3f(`L_DH&9-A|4fEIS{Wzh`j^d*>XHlj-qyLwi^Kr
zaLg-e)eeqttWqy`I#3ADBaQL`e_B=Tw$zu|a^-=xKZkX?@mN%z$Dew^`d~}2CfFLR
z2)4!d?+Ka<{+@QX(eT@Qjx^g3?73zC^)PVx2N3M+;fU=iccW&xRolZ!vqzF<Pl_~q
zYK^LYbh_PvF>b$kd3QEAB_3gA8i6_OgL7a^!MQ-m^Wpym@P7yV--+V`618h=RFm<_
zR=L*h#XA<7)o$H?p*M&(@rAtR?m-oU(F0Mh+zp-|>{diM5Nrq12dlZ1DKR-!7Q_=*
zDW)(v*klw5PD77ZaWeBB%bBA^1duEHc=ccx&}|i?+o@4`wp%Z^OO<B5)M@#Z93))H
z>c^+P?BJ_j@ZhU1eo+4HA-G!=xVUjbg<$3dm(J`((dl#-7Rv2o`#>syzVRlZ;-Ej%
z+RwdsgZ8Zi<2My^%>?|bH|OX5_QZtL!jW_}AztJ0Xf)q$c3XX9L#-bJqUNOODrj<~
zR*iz*%@at*29WUq;8at}#vr-YI=ENw0Bz80vwEx#QNVLWyLqIqoS7|D5^IxjFf!%&
zcDWU34da=a*0Gu9Y&dg>6VTf`KtZN?y?uvX#wa}rn!KjvH>$P9eCa5d7QOQ;3U@BK
zI99Jke!E=PdJX=)(63RBNl)q4f5_G?uow)xczNqsFCJ|j+YN*6`MQBY+P?#s_ck!H
znR*$I!Wh8<_Tt^SY+nyt6E<8Z*BZUxsdx;&MaNp;rdA*#@oM<)A8plZm0Hw$nEiaD
z+-~%a$K${tZo3>jO1uhcGe6Kfz#vBD_PihU;tSo_$_&9<zzZzB0l0Y?(Ec${oK3-~
zV(DW#OZ`l?AnzUxwgsOCGl~8DJ8VYr5gWV_D!u^#zA2@tBAAe7egM!d!519?(dZus
zY+g60r`JLyFfQ3W#VgT*^}xr%0BcZtF5P8(hGtB?db2UlEoC{a3sq=+DBW7G6ptTW
zsLym-WsJ5~wc38iZy)xV2)!KIO4s!=X51CZMgUWmZa!5ioIZ|X0{RlcN#X_xPds*w
z2!n)~NWLdR6`lqmr7l<5>W=Wi4TtNK77T?`L27PGNljMV65JcyhuwgC)w(&+3U;y8
z!v-aZ^uaYSBJ`U#Ch_Y%wJ5lC4no{A8uN*X8~l39Z+py|cu`P}yjHt;xCWuOSN1wl
zxlt{*tKO0FF)wO*m9}4we6QT_uDbdfuiIfg+<nU}`&Ivo(c^x9q1irmM|sX)?7q3j
z@g;qP(+<t=lZCvx3=6^+0L_Pt3}hb<r4f51U65(j;+H3GSuw|C{@^GetEGwy-YOA_
z??0jNIOE<ap@sLLs^Uy+Q@l9~OBFMOFHJPt7Lrc}?-?91gm*E%W?PXYXb7Ix!+@{?
z?c5`Q;F;hP!6y#{&&HPza^{)iMnkj&?%iU2&@G2AVT3X~JS)?Uv2qmBHOLttA`ebJ
z>>|sUpAQe<qnoP_Z|1<W_2SB5Egl!eP4Vy@DPQ%l7-6u@CZD=H7UVezo`v!Dq1>#X
zISuhpcmroy4%>E9VrxtXG1#J+^=)?S#zpHpZL{8owBt5wor#BLsXjnTf}5p!FC%Dv
zOZ7X@u85_Y9%QLfqdm#?>9-hpO#6go`RvmlV+(Gd-X$7SVKj~zrXt~g;=q5-?9+Er
z<jIz;lWd>9nVS_S!ak|kD=U;8Y^kzZ+H97Jy|P*j121jI#+D~`Y;zVo&7zHCb(VFu
z`8GF;HUfQ6+LaZ1ZG?73EZV@>D|6f@*|sfXgfeX#mgx+jB_3n13$O*ZZ9mJQQ2+9C
zY1=MOk!PT73wI(-IC(A;jVa-jrk|6CroM`Y=z5SM5|hO2dg&}tIL-Kz8sERzQbOUB
zt*W+86OKTIa1?8SV}nGHmh}+yfn2}BO|4t0Tu`!<Rn}wwbRr0|Fhn9~{eHGK<XzKj
zM35El$mqPUTm{>Qb1@tfL!G1rOAMV2C{4?$=%B>Ev7j}IDHB-Nh%Xk2_{mzs<9Xjf
zi(5P0)4Rs%mWDjHzmUsc3F)2w6(!8_uftzTof<3|M%}itGAc9i7{amz0ag_kYH@W@
zF3tuc@t~1ROCAlbsF?Q7gauYF60ZiFcSClPArr4f+*nSFeK%GB4ABkWj{R;4-<ovK
zIJ^}&d`djdMwl92_yktvVZ0=XLi2-3KXx=%A(|Y=RoMF=^xmI^Uc6CTMmt>dk5pi}
zyce&RkEL>>5;WTyUj_*(E?vllgvk;Y%FduQUs!~OX0>XMkwtUEU>3qprIF_}QC1=R
zHK@T10xWO?WqR=h(g)T`V9nlKw~0r^@<{NBc!aNg1f%h)xo*AAXvpDDl%i&(1Z!Yu
z2tL8eUI&XZtW;G)+#W~VwsB?y{~yDzEX9CJU5xJ)Zc*GRh~0jh4_S66=ksdFnC_V5
z*`}4%a#T)_QY;>Ye*+!GV?YVEeiZ%?zcPu@zc-q@`TF|tc&yQskLzUdZoW)-JYFMy
zl;1Z%4Kch<QfYmoDa&L%><Qs@{3rs$)jN(ud{mH1+C`DFC}t7EH{rmp$x3B;lRPmE
zR&u}`6U^hlpe+kZk2aWL4{$Wi{bcwStt@_9R?m|lw0^w=%X6I>IRgALy&uc3YR|Ks
zEuCE+ih`xuwNMPi*DBVMOunVMLL+IhlNnpI*Xx;7mm7^{R8DP(i7&oy@A(ld$j^8?
z7{Pb!+vV+L^+6wD{{bxi%y>x^Z^yoUka}f)V3X8FWdk>^;JPG*Uw13KFHNW7Wu9~g
z@T5A~mg>ZeTH<9#dg-_&+=5NEga}>_J<<tA)BbPZjn*dMm`T{)oekfib67G@?tMvb
z3o7K(E>3`2okjQ_z#x1IJ_h?Q3;rL_@y||@3s;t1LGk3u0r-9#@%>#IU*XXD@dYV4
zjY%BtU|8^BLqNR-JBhTo;x#8-nG#*%VN!yrjwNnEa!~oCD;BC?#8qMrN*nKl>Pf1I
zUZnW)GLkRxvPQqI+>1@Ngm=Rq7V4ayD0{2hs6^d93$biQ;s8$y^uhqpt$xrC=%B-`
z@J-s*EL-qu5=b}+KW7;OKEUuk9m?6ll~@chA7iqpB|@f6G1Wv7<BcSqdKO<_AmYJ5
zvQ;Gf8YqyltAhg0wv^yrrrKPsV(ue>KuOyt0-hVzgtml_0FLR&YB+DBpdlh=0HY7Z
zG^ckW3ATi{F(~0(@F!vJ4o4ryHN#}tc+yr%p;n?8@F9ernQ9`3`96+uUB@^>=AWgZ
z9EG3z#8C*nfZ`(be?mi{gkC@~gnp|o^y#o7OYG@Jitc|#;;KuFF8JXS&`f3L`URf`
zpIh|53E*;xKBwa%i2g-+qc)oktWR|!HKDIu@nvLXVSQKzO(6(!(T9yR6iOci6c>Fs
zmxe;=gMeb_gWTuMoPZ^6&Qghi?G*JdAu%qj6$3HItr&Y4ugwUXG?tv8g%rc3QGC`y
zw;t6>xEmfP^E@1cuMiKiDd+*TH(SE~ATs>47WAIrgvZrrz2rvG;Axy1i-d>5t1scP
z3L5c*NBE~0Eq|#mTK+1OK<WacfM7KII#xnR;Wv0WKEJLPeoK}kSS^>a9mp~PRd_BV
z{0>x;4w#c;GDtQ)#xJ}YCYLZVO;)_qbL?68y_8IOu=Z`@Si~W2p<w&XnCL_c3SW^D
z{2&Ra;!6v|3f?IP621!i=NY&uca${*A50-=ioV;D)G>?F7&@&)!*F`9hjz;8*)|44
zG6!;Md}A=*4#Xq?wglxFEf4s3hRXw~K&DXBVyuswYw-*M)^7+J!Sp~$%>H3F&rH-B
zf#0sdzQT&1%$XOtS7m|UA;usx5hkB?Zt0_co;iAUUSPpEfdonJE$5z#J0&YV9LA}7
zv&XzfccFC*_KbVyo+~{wS%+f@pxy^ac8JaBW`O-(9FpRWVA$nFS0s4;?N&9@^d{39
zl{)1!t(;|QJ7IES3374Q6zyeNp^{3=Nr9Bi9GsTZXUi;G73w*W@V_G|xf#)GD)IRO
zDL#{7j(o@GldhP8T>+uh(z&Ym{4Ad7)VBLFg?OJY`vVx0voGULjJdPFhIYc85t}z-
zohPps9uJ$QJl++c4Syu43smj*c+}2*nXi`}NZB%@6p^nVPXydd+J`?zXS{p_K7N8v
zEAZ*>lc#@xr{m#I+2{XZpZ~}{|A~G6GyD7({KSE2$#o|0i;!GDkj$MINw=>qtlJ4D
zL#wtn{kLfL*}wsz*;+-V*jw@U%sfFZ#onBB#T0w*(kk|OnToAQZP#mscb{Hwgdr8t
z>&u{>gI;SKk!p=CQ);~z)cOoT+rnxsI~YQ(Z^1z^wcd|UOs(0|$yVz-YvxMOA}RL`
zg_S#n&(JE~m)Ec;dKD#vyYRm1O5Nv#97!=S!@ykXKAUvK)cwUvtL~GT>YgUPUFj9}
zeM)~n46KOK|1z|5Q2IXZNY%&oDb;^5sD4S%xv=WX4kXVDc=X@Ep)u8eGd?lZXHO?v
z^<UyQJFunU0l!hIRToL`uPdzgIx<5m{Tj13Tw*qZ_TNUq;2Lk%RqFnJAv==wL#X>x
zNmoqWuU=YpzaUfH4Jz4nU7_Bm>+gn9718xSfOZbLZt_JcIyO!z`qiN5za=PKSkYw%
zlG_Co{W~}=rs#i$PfXF-)5%u!{fSwh+&P4v!)kiatir6kc;ucSGPH)TPP&>DdsD*G
zAYmz!d+}vzD@u8`6Sd2gD9>(`RkD$m?)zsUV^Tz%7M+)mQgZ}y(HY@UsWvmCBX*$t
zG(nP?n(IIz^lU{up3qIEm$VXP{5Fwj=ZbYK7~QIKYPn!w<^<aur%2jcdR@XyNcMzD
zk=Jfd72bXJ^zVRmMeOMe@XEoSYJ3*PG=AQVN$_tb;ii*e0CQ62_Bt@PKM_(^*xbqv
zq%K*_?I^c6wlvA@IA&&R@4zQ!YuVGuwzc<m%7=Y*(NhU7I)Fu}&f*x}TMI{5cC3cB
z!kg{=E}2QLr_7X-NsdyD%)pK`$p|mENe&Wt%p~8GHp!dpM3nY9>t|)GA5H*pn_><q
zm$A0sx(LSF((4i$>phuPE`#uP%dG~{XSplDfFhRrDQM?lxw9ldn(-Ip9ALjqzX@!*
zFQlljO_v=MwdtQhYA~Drd-%j`I(s_lHXW}ydG6VCJ66NnbbG%`X4AhV<WP!{lTH6#
zs*z~|Nt=%Fa@+K;CGeO{zqQY%+liRi^o*aCP5<Wvz|c0`g6ksKbW5+#37eilS-VYF
zgXpvAUjhacvFY2#+-!OVo`Y?=y<xvi$2+9HBBZFWO_v==Wy~9={{m8j+4O&dPt2yX
zr^T@8CNimg7;;TkZet=xAdZ5W2PD{f?6t`Nu1f~Mo{A2j18cMrmthV&U18pak?V+4
z-kUatJ=pYXCa%w!<4k7CXi-luy&R-PUmJGvVsoiRIlLIc%k9PPP2e#vb|;vL6MtOk
zl?hP0XUlP1$%4O;>`LE{BQpN!V&!!o9)MeptvVLm46plKIfi7OuE=%l$UI!3Z?qAF
z){7L8-%b*l+%f!CXbM8=vYq@X8Va?YTtIQzPW~|(3bmbFKryzHQ)hG-O_MFKInNY*
zj>J=UoyIQHK}Rbnd~@#~|9!^Xl+B~g6mbe>M*(KySez;P4F<)O`XxA1^dpi1q|pC>
zW9%aIKck^gLNB1W2>qH>rf*BJj1qbQ#SnV>d`^bg(~Xp|*-qkGSnz4^xkdkMz%tjF
zB2LFe5dD*Nrs!&tFQh)~W5jZD0=LmnD18u6T=e1PG!#l71QbIb2A<5xP>F%<lvcDz
zj0<bUKn!v#MxF6GWnd_uzzJGNF<csjFNHnA`r)B_6N{UK&`XyF2_FII_RNODt1q*0
z7c_FtY{<hy2uY->8B}?A2+LCogm*&X;T|Y<Iy|&9eRwD<mrWiOdT&aewWC7kB}awm
zp0$HPAI0AI<{+2*4E|dRxGAk}$ACU!k#j>0GmZg$8rtPP2829=V?fzs1BT+wzf7X<
z25usGsmUeQmxq3^{*!_tKhKALzAnZfb2TPcb>8Tszt<Rj_}gGDk}WCZIUXTzWF2n5
z0m5GaZn@9;NTvwBHikRvb5p0h0B@jx7kRLgm7T%+ip(64mz;ut4MqC~t!tZYr&F*~
zOCD;V-1H4T;j_N4!~PwE6XS_d=&P<A>l&=JI7b3jC0#K+Y{I;E={!rkeE`YrUPs~F
z=XF-VaGbpkb20qX>IKkFu92n_H}IY>Qwn=>n;^DIxuqs3&WNCC;e5F4Kw1SeW^g*;
zvlj=%QsOVdCzcXtPen7zOx_pagwO4u$#76NYL-$|b=kaPii?-*75v4Vs-1<y&^mr7
zt7o`7QM{GS@hU8D_dX9#Gfv+jgzq*yTh6bHG`AVYo3nt*g~03z%BI{cgjUL&>4|b}
z@3=kbirJJl=z)=x+>?_QY#3%PrBD}>W-4)=iH0Orx>%8@o{Y<uQzmDV0W@_?DZY}1
zLd_-x6qniL!!#6XHYuPi^=$I3B*uEElq<QoXtPP?O?XVZCHxJ>Z&QaH1jID-`k1U)
z#;OI1y;0wcvR_A&yCmB_+@{*-==X$e%(6sk3Y4UZt#y?WefGq#(F)Zy5n!<HQ_e(1
z*Z)Zz&X%yBoJ$!-?J)h~bf!PB$5RUVzIbXs46jH$^#f>^8Bb-qBe^eOr{jS7_an4{
z8Mh)&*!`#lPgu(iq%SnpQp8`xzTuS<GWZ(7Cl-9Mr;{CgU6Z)HVWO>pcbBHPmjE3$
zygFEdJG_(L_yT`&RVllRp)(!Km8w48FWkm!f&KdQ4cA}GzwgJd{Pt3~rZj1XhmYZV
zXZl+Hu*-vcnc;n&qMN=Qc{AQ@vsw3E<lWis`n$ZgU-$9mnfX>X`Q`d@y#qfbn2M0^
z))ro6TCU$GqSdVHobHFt4>TLlgTw?b6XNgJ)iwyGp-oLa0N1e0cXndrcnh?UVE7w`
z*%@p8Xt3vYwsjmg=HJ+8bt7?AYN6vC+d<eDX<B-A(iICKyQvWJ;3g0R#@>XD|DXMw
z{Va@~G^6E9i;zlmWQ2JmtRopIi*L%eUz%_&cYu5mG<6J+uco0;0kVMN5+L73L!km>
z0mY!4e9N<&m~smDFT>~TMR8`99E~0`<_ey3;!`LddpVPg9HBHPzv;8}5n4m(T?~q;
zHteEn%2n7@WK8%IsU*pF#f;aH%pj);zrxXUS*7pMP$)GKP+ZjHKhjVrH4#t@H91qb
z#<Ww;Qi#m@irOD0G1LVkm*{6gE+pjdXMD{e<eYemAmq=NF$N2g=xE0#@7wA2QVj-G
zQ~dg$B)>@Q_)|tjr#1WU(NHMu5KvsS<L_xGly(RxhIXvP$TLe7_>YQ?qpPWCtgzg3
zFolGCIbf7)|49x?d*o=zqZu3uq(3$!_lh^+HN)k|cq2)?>_t+R;YpGCA`)=Ctl595
z_%t@z66P4)i%B3ZqZ^M$YYn`kR8Q-y#3_s%75}0Jb}I=J-&!+QhC`>N!{s`3Vq78(
z?|bRCTFrJ;>MX$PO<vJ$Mt+CihPq!L`y!aHvohyjY+R|7FTB$}nXSoj8)a(vHFEf*
z8S<g2V=Q}whC-Pk0ma1(Jw!vH%#eUm+zh>$gncs1(4&k!IecvnN`HVHEvcvi${y$;
zS<KI{IiH}Rybyji^br%4!sz4}%OBHF9Hi!eK9&)d6_HvezA1hCIt}gK^q2<Q7(UcR
zurV)6CvpblZ@gEt^o_{k*z8bxjY;gwNEa*W{5?$_YOmUlpsAxZTE2z~8A8e$2`Dbs
z=u{dCC07E9VT~?y;|6DgWxZnH0+8auc_ikAZP`E!3PoGn0a@U?3rBu|92aS|1Qc^Z
z`f91=%U-QTO4rg*DCx>EmfLA4lyq?@77<yA*;s8Phe#NXav*Ng*572u;S9lT)7X(g
zN)Cm66**i|JOmUMr*@o%LOHb@V|g<Tg%S?|#c*nzW5Rr=#pJ)xunM>|j)KW$UEG{g
zTT^n@K6qup=D)U5&870wm6TO|h9(o`+dc+O9evy9XegBR5l~z#>DOo|lz0g!hHuL?
z;r{^%TNjgDiLN0g{C~;VlOxj3LFv)8WUS&qX&0k(It_&~N;$@I9u37|EQV1U5NTgT
zLn~mG_)r(YEUl7@n8N57ELuJ_B~1%HDg$yOJs^rn*FjS-AcoS7Y71%b&Jg)k!Mlfs
zM;R;*uMnkN8ue}Sv8<y$3%etd3FgeO3&pl*3H#L8(<RAO+%7RNg@UD5aITv{>wQwV
zmffczf1E^~oDy*;76M%=W-PfD<~5w&OyX3SpvBNyJO=Qs2cU7w?$-wU9&)e_3~<pK
zZ6}vSW0r73{AOcsMteHRJoM~aw-mO0Ter@DwTB{GxBdtSp1F0)a=qi(St$2!&gw$e
z`%JMU0#>*G^%wdJpB}l*Z_hQ`3u+sU*Tpx|dr{!y+iVWks_-%!ww~^l8&$bjJMG<5
ztJgiebWLk1E=crha}$TyeQcuOkXP$?u*B;{O}3v0ns|p2qPN+`okRTHKa(Jt_naV{
zG<pA8O5TI<czslhU_TPPXUT_m##f9K{yK9w(uS?k2dVY`7pdOO*i1Jas1kpUwYP+b
z5}f1N6mPUpOY1z1diY&vn7En^eRP$`_T9IxT<uSX!gtv;fnBy;fT1S`&5pwOzao$T
z8jQyib2YzS<?nY>@4uDrD;tkT*zejZi~314?R#}D+u4sV5$3!5fzT6Lks`*~>!^@7
z&AiQ68{SKSfA}FgWMVRa*=F^ahULK@z>cb^DW*<SQ{IecZkwC%;NLC#udg?cq<-<>
zweNl<nwXfH5{ve%`L%pAo!(9qX${J)R{dD15|pD-<ikO)a^$ndxXiiBYl`i1tdDyF
zxEI=Qqn@%>9=7siD=bE{P{S>3yYLNHVs(ls^)w<UXRP*RZU?Y@SD{y)14eoCegjts
zVLP7zjitC5IBlWHp>I*!gH436p|I2Jw&7h}89gV~i5Vx>`Hk{yN&71GmUe@D8bvUd
zH8C;4l@E*yP^#9b;ZU-L8s4~UR4Ubg@|yI-mlL8xiVZ_!62x*Pkk#mNJn1s2GYZ=v
zn~ixLaH<yQPT#Jyt8%H$i@Dmo5dNUe3r3axVVS<ak+dM;6<C`q-Y<b3UdB5F{bMh!
zhgTr`ZG9l?OofxSF<q>%xi^IkZORW~ZH@&Z#Vvv2s)mW!$gkd`x0Pa`#FMV1uwKZ6
zbz+5o8Bh-y0qIc&*gK~=HsLg<*>s6&f^52oK?yH~KZ)nsWjAetEGe_bg*BYfJ6NrC
zAb!7+{2j#DvTy87&_<U(nifKk-vEu$f{bN)M>K^rL4FNT==H3_@TKr4q0pEhf|Q2j
zyJGyiQZQ-cg}5RtF0NrsgV!gbXIOb5vB~&V;r7uyxFtgQD?p47L(BA*mb4?*gc6%@
z`q3oUgc3<^6UzCiDM$A?oa?m7^O+g2ZL&?)#hB7E3N32;Y97{vLgv53xMCV_EGs1Q
zuLTDD57wb6^XtqgMG9NGSV`HBQW!XC077=rfENwLWr_Tdy{NG+TW?aE9BZOc4*{uv
z#z;-p4>gU#CY(FY8ifqCX;fGwNYu!lBS5q%)p>)#+w{w|8K|239?bLeuup2;InXHW
zX|Sx2*1aB>w1sszq}I)(FrXz%_#K4oqIH)QV!zb3f~}R7-JEA^iIzPKq`iTWmZpWK
zW!Qvs!&%Fafi^9>Yier1+(@IT^SsEWQ0r5WY}#Y31FDj-DR3bV!KGf*7_Cijgk^>F
z;t?RL&pI4ZFJ7C%fR+WJ7YNx!FD@y-L#Y*o+A5uREYG+Sop=L~_176$X*y^+flW9k
zoOJ@3XVZ!6rlxYw>N95aljh9hwEmuLYLHb8G$DIW+DG$HS}Mbb7?Dk7z_LQh@T)-2
z_puI5W!Oe0M5Le%sHH^e^C|pj5fQ3_c)6&``V=E2c75dvpKs^EClT>C0<pfqh(+VC
zi8wal^mP_-WC<5>y2-SNYlH1K<FtFqA?lmNCOn}z;F_>-qw4+D(BxG7$`6E}4v%u$
zDl)U-Sz1{<cDuO#Ky2zszAO5#JJpm|nx2~x_?B=rHcIla;Z6<a!9RiAD_e*)mQvCu
zb6FmgW9OB<IxMn!%{h;1vdcRHIaA;B%awq&hTZQAek1ZK%|=wNH99J3iYqD$t%$9p
zU?#QFY(pZl)ofI8$)(wiAh8OGO-uuWZHGk?I-e?zM<3e7An@kik!H7E^=5qydpy_N
znD6Xj9Vb|&ZU&lez;X_xhy9KR3qq_%6FrW5KJ-|s7VzV@U1%st!qys5hoLpe-d^g}
zs*sUwG$FGLr^}Hj=}hoWKp^?Y)HvLuahXe-ky$$UQ8&gHXZp|#KOxK>p3#nIxiPso
zgHzu2yunH=&AUrgWpQ@ovN&48t_POqd}s`-t^El!eYN%2Ns}>2k&}plBJ|~X2%Wgr
z$AHS0F)9ZtZT^^}jN%zAcRM+(>M&~KQuYW#Ps_AjpBlQkdb`8X;7e{^;N%}P@CQ=h
zX{RbyrV(dMIm9(*M>8}5m9V5KMn*h|usm)PmLrbBve*VKTO8(6{4m8~a<DjXRHZyz
zI8Yu~eN7-5DFmDZ0<mTSanm5r?r!uooYF*b<M`VWj+6EEv?f{<Znx<s+Cl-EdXEjC
zrj!<WW0S28O+@64A#9Gw``4j&6?xO+SjU58xzj>kymX?jwtW&tlKW~~g&ez+=C=ls
z;@kKRCbi*7j>#m%;LD_*1zh<}o}a=ec9|4=I@y;=U8^n&80I#qE5vTdR;}gZ0p)4D
zYfg3Do;GGU?P>MXu-8e&Ur~J(yiAJimw{ejb~M8;U4j|S{k!;l1ttmN-r^bCRi4D%
zQ77OIrwKJyX_NIL<<-6pQl@8POly+sBG9iST``S$3ZhWsBG6xD<jZ}9(+MIv_l4#a
zFiUlF(%&VN%6)Uv-#}9!tht$aM)G*sjOM=nFAaq{10$fgoPk-pff69{xR`)qY{$=e
z#}G@~EaJ^c=a8e(*ZOmvs>wO3Lc6TD0P5+@3FMhTPJYut=_9lrFZ(Mle|N1=iM4o>
zT}6e48SGuHRh2?(lqt(}4ao{}WG`l<a8i<&&`>BP5l~!|q(noZj{FEH1{KFy<^g>V
zr?KVK!Ah*68G~1l80c)zCD<HHA$eV}G0A=`;~^4x@>BqaVj<89AfHB`M@bNc>BFEb
z9(nACCf-fr?m!;X1~{me*+8d`v(b27D(!0xnrtSd?|y!fy6-DjrsVf%0#Qe}o`t54
z{`QMB6v~tcC@!Yt&uA!=DG^W%Q)0bIIm4@3>nbY$gv3l2l3XUn3bl|*{Vn6A8R3$`
z&Iz^%!hXW;EjsfwGWwz<dm}V;l<W&=D3oLiC@zwH2@QpkYyriP>|Cd*uOnd>mfayv
zQ(wi{lS5c@P`X5pmW<g26yvCF#=S)gG!*KfZjQ0MiiYAKHF-``AE%)eI8DumTFAz@
zH0nDeY?-Am`5EEgB&k?j+gQw7$Tj?Dz^Km{+Jn)LlgN|N7KdU9MvEQBRj&6BNt_B3
zv=~~8#{he{`z>;?4h(S78|{p+MPrumjPP062~qnYVTEqrA>l8;<h{rt;hzD;GY<*t
zc}4j~hb@^%_a75}3bN*ZDpFOTg8j#YgL5A2O$O(JFq{wnFM$6$;Q!A1d&nb^c9$I#
zO}lT}DAVp|;uA}|v!|1tcE7DuwO!gXObY(GyeW7q2E(S~4{n6rFd2heghX+>d~CW3
zr&85>9$e;|u;r(Vs}Y^33@5KUGjPNd515vtN>J+5Ug@&}*zI)b(q{8F+36$`Xjwli
z|8Nlq^&(~YcO_l1EdTpJWsNNV@BUmfPb3_S^yX5n3S6I1HutK))zGvre6g<z+(|>B
z=5hjx%Uo`jhC)T<0*W!0%b7pV$Y12_MX66jjz)J$xiT9$@hKEnwU}h&m_~B)yM{=2
z!a4x>?|lr4xtO+1OZCrqX?twT>=_FIBCBVVDULi&a)eCk^f<0AD`@Ycp-{Rapt$JD
zhiNF3t_UcGuAFKmwPn)LUs2Kbvm{Q1CB7eIA))>hV_^=V=5$&Fp<c?Y_4i30kTU)~
zMj)pJuz#eXP%<u{xXAeE=~MthW}E~RL&mK+;S8&9t*az`Gl^MYNw-2RB<!aHg1Odp
zIl&e|*iY*ZhHYkhtx-(|rKwUy@Vz9#NwL11k-|x=Z=#`4VlAM!i1k4l3T3(l6ho|6
zz;uJtuYXGe`%O_XBq0|TTLz_&n9ecY<gm${0E-}|8!huQo19LldQwRBB1PLblC&iy
z^btk{CkcImhC)fGfZ`&dPts5*2^CO^O6bQ)$R|NUpJu$tA)z_Q{AF^mq<<4oTvk;6
zf`&q^sN@*Sk7+0lB4ezmEW1={8$2l`JLL>AA}Fwu!iTsB2IYjU)}BR=nOdzq1DZO{
z@ORKqD8nM4xEPkpXegAF2q=bO$+cShizMvAre%oL+G`kla@dv}lrEE_C2fm<;$mBx
zG!)9V<QU6?G!zG^F>K3#yRY9sLn~lo_)rVk7?(zUtF@MF>yoe5K1ousxV9Bvt$i0@
z)MpIstF_ON$dikO9Ev4zQfygNxxUYlI29&nF|-N=`&I_PqTCq#Pu=MJ9dfV^3~<pK
zZMD{-F-y2wdqH~P)4q0QU%6GN_pRLSfGkasmD^J`5i7S@>qc@FcZbtp`d4({2N~g=
zBCP~!T5LsEc2IOhcQcOsCvYe$JPV)LiY|LP=_|VIvSxJyjJTV6DOYsu7!13j8&~Z8
zPVx&OU^Vrw2HZP)AKSdnK5z#+9G_AJ><N+}_rpItR(04f&DG#C>{dAnW<03AQ%vkX
z0F0Fvd&HaUaHToQ`dJy|i$OvbX?b)y>546neh}o}SRTE35s4vaVJuc&l=QEt%E%0_
zsZ%8yokb@`Fg_+akrYh6E0W!ij5GIQ<+aeX&wJY!EAOVEP*WKJ#bqi}rJ+zkseoco
zZ-=6@1LZC8Mf&W<VG_T>@xl;iH`+E9+5@mhNaRWX&!IRF=y4LpbDTg=k;oS#Q0j2Y
z$4C%`=~Mi0i&cK?@#E)7+#Sf1O2*oud}C-RE{5j&G!)9vSSe<I^Ws0!P$;7*pty`>
z^i0ZUl3597y*{0UQCMz^8NSD!3{PSIg110?YvqUC0#1^naFAdJ3-}8pj3uU2UIN`j
zBEL8UI!MBJjuWU(B43O^srAHHk{}Axr}%oJRl)4W`;8>-4&-ri5l<`$lqZLUVx~*T
z#Y!H$nhkVLThDS{3El;wcVp7A(K1uB%`4WK_2l#-U94sqzfV(xS}^$xG<BS1e3^zq
zIZOe?#bN#h4TW<40*W!qV8K_$mX}oc_frzG&hcE^7mJ5~KW4niv0B0jP)Jf;8ug{!
zEwlKOm_*D-Z`hNF3iZAu;!aTAB1y!_%~TRG1G}swqP=>567ho&WIbO*vY^Msl8CZ{
zqDjOnapWwCcojaeBqDn{=}E+Tvs_ho{J`7dI^BBIS;{1$9fP5hi1*t2ofIR~Qi0y?
zefctq!X=rxLpwO>nl@$ptX$N+AQ+33JG?vTiscSJ0%C9E4mb4CfT3yn;*3g@=38&q
z=v~RdnuThJtpu;!xxP9yb@al&L_?vxuz=#?g<~2D<%I>5qF(szB;=Feh2P3}lfw&h
z0=PI-T#pk+0|E;aeXuIX?DN7FeJ~ckcjsD9;M(snm;U_|!>D3e2R4`f45MLcmSRsf
ze1k-uoC|R%4g~r>2}5UYu6S~32=p@&`C<f0#oKGnqFh2@`V^11t@2}^OM4{l4&+Hm
zsU6Dn&`?}#{azXhW$UdJv!`2cqM=ZJOF(fM%Rw5-;ITO9^&APKu-q<<UcZt=-a&93
z2=p)s<2g>CCrIQMhd@t~FrMQCdWJ;47=cn**3Xe33e%@}mep$K>^A#5B<>F6aq<j5
zToNe%IxG~|Ao!Ex5DIJwEs^X=%qL~cY3=zWg?e8;=_L?K70D-UJ)6oWWnh<;PqJ6<
z&nJBZ*l>r4{R-!kWCum_NtfctSw879d}8?|_OuxJq->egE=sv)%#2OVM&f?*5%$vR
zp069E@W;59?&@R!*CYdAPeljN!E4jEanTB<<(4k-%HOPuU*a)%tx>z?cY62tCc`&D
z7x6Xw{e@=x*d66L-`ut6O|rqjO&;EWdrx-u;>!nAna+ZeN_(+U!7Hfb$ikksUroM<
z>MdGX{I;x~`BmlY?eEFe+2!_pXL^6L>esb5z~gNvmTKZ%&057|>tWkc-SRFqx##XQ
z{m$Qu@7sHR<R685(04HU?%21>+gtTJ6}Ts5zSL=zD{xs#QpMY`Zy)z6>}4ceyi|he
zWgts>4lQ(vh^whasYC`95hFb6K7W;yK+V~}?>IP^z+>6Dr@@H4L+9Wi@tT%tw!sw@
zWsW95ZK=1(@I8P*_!NAYx;g@P>x855f%v~2eg@-b!QaS)ZqAZGiFbAdh41SI;QMjJ
z_jhT0-=B%Ekqvzfzh44%;Up;r&nTJc*9+aS5>1U<;)H4{Nv;<uB0oVAnOyRH9GW`j
zW1pm<Q1e;=#YKysrlC+;ET9-PJX;uWi-qJg?LSyeHBmJGLlRHjQRd2dQNUeKW*h(|
zXKBLEGUlf2Ahkx$DQJcO`cO<eC=$9o2E~;69gfb&Cide=TPgKgiDJMHNd}NY|9y_J
z%Z)GpN<*Q9UO;gX`iU*3ze}-<5_$o}5c;jU(5L-<me|vc6y3Lyxa!i9OYmv%xkZ07
zV43TZ3{J;I5dDkthEp~jSfA=dYC>PR;>(pJUr2q}%ZQblj@vijAE2R7`XHdV=)=os
zD3m@3D26_qqw9kupv_W=f$bFa8zjbswPGL!xfLU1yiOSy$|rDw7E%nCMtzHX2NR2%
zgw;!z8VMf;==Qva!mBUuaR)SV&U*x-;ny+fMM&ZfRrdV4Uid9pj#2vK81?cu20sna
z`dxx7)>MW5D$jGSE_d*z%9+-&ndWRbbI6?j9^&Q(PGydXirO6v?@39qN`c3l$^L?O
z%9(oj5$uDX`&?&&)&PAp1;~`g$>2Tl^476lJlZ<88^&NJ9x%Fj2T<fg7D+Y~ZM5yr
zRhyMw@U-EHo`rVt82lC;Yx%u+1p;7d<R5L-Yn58mJI;PSQf@bT$Af2rPXwQ2915O|
z(<9=M7U$B(04F3MHIc;nj|#40{ofRHfR`4kSBLJu!LPUcw#V}nQBaQX^~{HBFq`(u
zaOkbvsFvGR?@0NW7d5>~+b>7(Qssts)z#N{-448VY@*hH7eCk7ou$bvbCG*hrthm_
z3^KQ2@>1uBKKf^j(VwU{{rzviE%%%LBvS;Rz5vV_A`$gc&6Ou-7rn`%i6yTbL8fcA
z8~Ko)+F!JnX&X|hoRuO2H+X|{R;Sy01!FK7>Zyfr)z-vD*xM-Sn!u`B=FF6?T<@bB
zO}b*9<%?iSm(H`qyR*_Mc3#+>kQzdt6IuqNb9O@9`!GLrKD3i}d1q&IVmqHi5<bLE
zK)ZFP^+izd6@ubcr{!{ux9mXL7Biv{+3E2_B}_-bGt@IUK9*vCAwIDbJ9{dcp=Nrq
z2pQ`AiMF`x(#(<`Z^4g(W)+OV)x#Tup)J6T`j}ulCOi2m@ajgjdWYS?UE+^MbZ|S{
zuzF7|3U0*LX^P~qVK(CNq<1+V(n^%cxKl`-6q+GS#%)Pg%w&8S1lh<rE{Am`7VKoL
zOqdyy13YtV>DAcBHZv^QRNN*OPBz7S(}f~ZD;b$Bk*s@dBY1iOMxcVH@M{pHvfm>}
zu=!C>fN`CbFf+tss4Za~Far9TSsDALw%3x#lS|JWiiJSMZc9-F{dE$j!UQdb*5Wb1
zzV`nva<C2zFb%m*9aCe(dbveome6&dmt}mF>xQ{XGTT5$_$i#dZ3z|Leg67pAYl=I
z{aI+|;IB2yC|Am#H+UHR%j92%QS2Aowz_eB?)k`VetWLjUVz2bIlt}0D--yH#ESx7
z&S9z^Th@RHOVz_nZ^w_Oy?bi)I=kLOYbkc5d$qZVL+p@<C^+QRIv%vb$<Lkbu8%9Z
zhZ2B%4f7D6_Tb{;Jtqh!Z8rO*ls1IlU8~(jvd+X_x~?_)n6%Emnd;2+Ub?2Boc%dK
z0uCVt?8WyREyU3}r!n2oM@ES(k$3#tq3~TcO<<R8B+zHbdGuFd{9h1AfP}7xcgf3S
zgj`ua-BNazRu)g(R&JLelJeUfIb^*^S^2fwsOd<WJDd4??I*AiPZFSq+o?f5_ybs!
zotk1QF*W7QcsE>stzUta{Puoa&Y$q$-!1#EuQ!jRe(@k};a8%GiK!`}MXb5V_Ur9L
zk=CHxYSoXGDv($xMgBso4*3Jtc(&QB@A8_h2yW#mbJ+I;5MDwq0(6bF^01XBTVXM(
zHhTj)ycgw|SM}%0aPWvJ<}@PbH_EecVNaC}s_%f316bbDE6)LQym`Ojx66p10qy#7
zz2h@*ZNKh=Nh!^@x(<Dd+8)&Qz`8WMZFtE`M$d_bVVfja=Qm0#*VcAZ806C^a-iAp
zCnhGCc!hg35s*T9sa6(WPRIl)53K*ma7OQJJn2mQdL{8In9LC-j+$al<&ydrL+_xI
z7eXVnNx`U6A1u>n^Q4A{S71k6Z@vv3p2s>2U#cU&&eRvFyy;?v!Br^?X!ClI4+SA3
zBA~oA4Y~-fRpxFh#TtnlT}fejdmc;^Yxo@?)UAwA^vD85TWrE<OS9t=)dbmb4ucXJ
zV!O+3jRaXzmdGZmtF;bn!rJ#cCbESA)C4xlL%UE2Y>N%Z6j&@PB(Q%D6smIy83H>d
zh#;jQ`K}oMh7?SikwQcfmWzm<VdaIyCgWFy+Y@<kON8>fK#aFAV$fsJgc6%@`Z)_F
zlH4Yg^HWof9&b2rXOrhMGho|fo2(0RzR@xYP3E7;!x|~`A7flGWgg24$^8EU2K-mn
zVaiCGfrS*db*W+{WnW5RKue9#0EFzK0WTVg%M$q^dr@QkUY@Zg8udLO^>-PmY1(KS
zg-tkjoHYs=YSXB&NL;6pJx5GuQ>ya@gSY9IYco)_?R40xY2B*rrn1oN5T0fwG)jj?
zSXM~u{t}pU3hQu4tve@$0X-C<bqLu->n<zAeyME*TPrQQD9_juE&D4V?QTX|niiUt
zVH3^`XDvep+O+Jhsi^_e8I7jSQyH5=txrL+X^*uIsLE;$M@=to&qHwOjSetco8Ab^
z3hBjP16glk9S*4%wG;-lEC{_o$S!(uNdX>8ttix1>BKAZj4RQJ9{^d8FtXBg&~yTu
za85Yu1TxR26W2{m<(`OVOu|!><kR|lwy8l@HPD20CT5y4JduaeQW+j+L^hQH%L*yO
z4}qRN)}g5k+sGV-6tn@glt_Ijg&!><LRAni7gbrGVx+{buUz5tnLPL;BL25PtY;aq
zX#6!1$0nS<&LWO1;UZ2qnKq~0TO6XkNo=wu#JuBx-ALg^)%!Q0$(V>nA=5E1X*zbh
zA9dS}Bx9O<SM>i$9{PuyIodxG@(^slxi4M%Cy*{>tE1%NF-w1P>|n=K9bTC2HRn94
z$u92*q&a=hFINKA8uEw>aH!0yG#gPFUI4A~oVbv&(2Cds1|~l%%{Jr?Tg^rlwqAMs
zrfkSXV&)cXJFHF6u_U}b8dot6?P3skbMHvATd#Vvu&pb>9<Gjt{yV!^#|c)cmsn3X
zU=;;2y0B9W)?rwWCVCwAeCV<4D&W^<?AB0{goO**I}EMKsQCblxK@QEYNH9sV8~=6
zQPRoVoq#~{kEyXGBWh%p4t~^)@x_@wG{a8_vxjH2BU)}uF3#YTw>@vL5=-;$QdL=;
zHRqTflD6PlhWQ`B`mKV-)*P?C$a(Cf$(W?bNyI=AdRrbsC$9CUK;^R;m9-78k2%U{
zK07(A>M&}v<EaciEu4FPYUt*A?+!<UFS&VvlYh{_Uy=e(i?~=JMx5CSan0G$3{5~K
zENP075l<p4|6~)EBaXtd*aj_I9BxkW!xV?f!Q#MCmGW@mKzU&GHG#Mzg@BVlAl6JE
zZW`p--HpD6Q<^Al9RJsZ<7E9%i@edIaJx-6(N?d})Q_;?)0EO8Z)~#6p^1pRF@((#
zdH)1@SCKb8j&(dpmOCEgCdU)CgZ@o0=-fN#6>_+3a{t7#LA3Zb-!3IL8p%PKln~o}
zNBNFd+g{&y0kV9r@4NAd?e%3(i?P?&L?*q}LySNk(;-KYeVxXKlL34r8321KIspAT
z4K;vxJ*@C|>;7&u@i5Ej^w>@`9FzQsO?<>pJ75#w&ii|~>1HC>60G4F2(}~kY@M@{
z!QNP$SlivIwfuUm;qN|NuP^LQG9ztZ+xgHPLXwBuWYuY`s<w!<0jU(^6&mE_>EoR`
zN8(YwnWGo%icgQ)<wgfKxxlWG63l<*VYb%kJ=mL!N1|p_uJ>*}C7zIV{3?9y;z*ap
z>ndHC*)^h)=<7hdMwZ~ndI#cFunjH3Sy?aM04I(3w&uAuBWe}41m0b)ci~c!N5u|-
zc$81?dIy5_2jX$DnWfo=p74mX4@Y|zTJU9cyV;BYRe!E_v==Yi1ILp%3MGL1092R|
zU%Tx(d|MkeE2Y_T<pKN^X5(*+Pn&}d8qLVpiXV>Gc3{I0zMK6h?$PT#9IwMUWvf(a
zF5pzR16^(5C1NhrDIJ0R*d=xj35EfXM1y*{182;j0B(qFHQQ0C?jQDnkjvwVpj=PB
zj>H@MqcxykRF1kGS!pC5srs|sd1$^4&sYKdpaz^`T7Z&qU_`~o!61HYUr`D65wdAy
zFCOnkbGt9!4L6GP;;rV7CBNNnwo3tQE3Er%pgp2mfgQ=~InA|RMp!Sscw6e1R2Sio
zK?UP=ykCWbY`xX|A=NlO6~_Ql*9m|`br@4GURy6W=DX#2A1Gdh@5}v0FW%UONpiai
z8>Cu(qY7Z~7Mla{C}`#!3==3)t4dJ_M&otmZqzKPqIh`?1Z{1#yU;2n!U%+A1u!r7
z+pv=hJ6z97n3yB?RZ*+scdO0P(FGKxQ{fC;6Shz3Et!;5qSE17rxr0`U0;H?7_x0$
z36^70NkHM&z+gI{7EI%y)3q>O91MGP8(z^9Z(;|bAgkiHnO5?MrwS}z-7X&iR$=F$
ziOb>*sA(WMjd>8ndTjwt=j}grN4&1(x8ROzvB{K4B9_eJ)-P0wcyka%t<KDzJy6`4
zZXJU&5{>C*dmeP@46MlzfSSn$xL&*p%6DII(WOAob?v!Is(UO^-CroX$E_zQvJd#}
z>6#zSF&JP9_5{&F9e|BT<yp|8Y83QtUKMWxnOrC(%9#v+-;V?54o~eIgLKo;Qc#-@
z>hK?T)YqF05cf5mg)%7GE4rZ79U#<3_MLw%)ti+v<b05{+hBtp-~){3`Ub3IB!t!v
zbXx#o9bgyLBG{A1&A@Sxo@NCMG{}Ik3#Xx>=yYdO)oKeUyLi7As4OCe#t1}aVb-sL
z1|I_q0v`?m9FNbaK&=N#{zAD{FO{oRw73lG4cIf<;f`OQ<JRnWyuxoB?j7jP#;dqB
z0rdq*tU*2vbOW|1vZDoTa2o(AkgNy{E%X6<4bhIrrx~RU*vH9=p|`o@3+~Q_v=mk!
z!$ST%8GaVs?5FW#4J4Star|gOFcL1{$20iD@4!bl9>-RAVwAZK*6uiKbDY&b&gvfT
z2IqofVI6{-+#8I?qvB+Q^Z~($*fo}%7}=ZnUg_@Qky-~Fbnju@;d{Bqv*VBrT`8R}
z<R}o+U>`?wJz|Y$hUo6YU&D{L;>RjDtr4!pk3U!mAD_pMjU(`}89(mFk6HY9H-0>Y
zAAg4*KgEyAD0~F?@g@BD8h+dbmxYD*;>QZO$1YrjAN%p+cKkSF0zS^hk1w1GA78-_
zd{1eJ?=cPWWdk9;j3C5o;@Ay%As(D&_Yj46dM(7$Z6WT9XZz;6@mgI|81L4?@m%i6
zfuFnt4Wg{ag+wnwEoP;lSqrnxdZK)+@bpw+QhVI!aa=-WjZQQA2jA9?&{Ca|?o0dv
z!3CJ>O0$c`7(98US+8>?fT&^Y8d3fzNajPm+rjj-%55-WAj2KdHV!bc+D7r}q?jqp
e>RO{x?^aPk=g|;MVD^>om4$c%PEdAdr~e-ehdQ_b

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/tokenization/hf_decoder_model/index.doctree b/.doctrees/autoapi/lmflow/tokenization/hf_decoder_model/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..6d2291ef36e02a4f788fe1a72d4c312802fcbf58
GIT binary patch
literal 30901
zcmd^IeUKc*b=OI{Pn|wapi6=ftOA&mrMru%5S&BE!4R+`bPxv&HiBhtckX89c4s#;
zv(lXb2UiT{><&<2_yZHlj$IWeflA7MP)VGWNldvC|7p9*m{i$L*-iya1>2R16XTT0
z?{)Xg^vuro&5lkdRasTqo9XH9_tEdY?tZU79vXb=FITqke{8qkwrqE0#;DibR^9YN
zHd1fZI!)6J!Y9K+9|<1}r`V`xoc3ECuV#j93v$#fyHWQ{H@qigyC{2LH~bk{e%Lo_
zf!%WB;?|%wWDVaFPFbUDFtCG$seEFWH=2u$*6Ep`wQRcfhYTt@V=Y$eW)1jw)h2!$
zA(cn*?YEmNgjc~*4~GK(p|$~RgV35_qt%!^Ya1JqRL!&P4cj%t`ObpD#*85F?1fGM
zZV>g`?7B~s4%Zrn?;{b<gRapOPb2kCJ9$|!YRjIvNO?MJI53vzQ)unB_FLoD#nv|K
z09%-~T1|7->$tAz&7N*~%d@wfxCtzX{s2LLRv<QObb^-Awr2&CW>qH5W|=f=yLEGA
z#`8hwCC9gR7OX97kf$m97ENJY3c{>|;OS-fe+B+uiT{U)Nbt$CIcZaDd)u(RkR3rA
zbQ<Qhl|oI$4vXy5g>qX%^Fe5Iti#rH!lQZX610T1Lx^7@KYOACHg<i|MqD-~Q%SPn
zQRDT3>!L=ZC*6bqHPSJW-8upej&Kg{35<nK!|<xLR-@{-%~~JK+|Ki}JtyzF{~ZtB
zf7OH0-;luFnt)^e*_VmAcU(Ppl*GyJG@FLEdJIAWCbLOt0H9bi%IA<xC~s9TA+5F-
zjp1LTwX|e<^aR1IMsJa_4rU8U*&Ag;!IIbNw7WP-7GfBD?vu?UD014a2Ud7|j5zCp
zv&(3_RwOR#3B6Lm2h|G5UM;lhtGS47^IE5KKMaxzb(aBcqp(B9l4rCnr67CPTzhq{
zwcyO%CpcI+T10bZMCN1no#gzUg0hddO}B2lOVt&aAQi%~GlLK{=y1ahOwVX2Ra?(W
z?Hdy4go)<;X>ZMgW#K5Yt?kv24YgONL4IhehKQQsQ)tI$V3B)jEgJ%%Fqa{_s((Wn
zqFbZVO~ZCW>v=Yew_vr6E^ZrP!*<}^TxmD#njM6X^3SIY&kfhmjH`3P#ue>|*a(X7
z?jxM%U<1QjGJ}vE?xU%>)p!~`$T_kW#|O|JPeHFHt)ZlupHiFI?AcV*o=;i_te0R|
zsh%gOn67Q9z+=ex2>^UNqsVex5_Jsg0NrnWy&xbx`fI@E5n%K2G&WPc(PKBqikBG3
zg6f?VNL2ij5oC7L##Agc`_oai*8fC(X#!o*=%1=hoz}`J>$w8T{NK8Eh?Uv;TU9fi
zZ3A1j6jx5`i`I+QKh9hK#N?PsKd_{)vg0~0)a%Dt_4={ObG>{*y_|humejB}ifY*F
zXRcw+6@aU2+(?@NOHYJUGJU3xLKD%!_s+z&(<mauAQ)D}M()Kda%`V6ys0!=?vfa-
z7}-v!j?(vKORJgK=t{FO=eG@-9gVa-(>J|SCYPprQ8pS;p5()=1m)B;!-&S=nN&F{
zCo0BJmz+Yxqf^ZQdGw}^%co(!(D6y8q{P<Ssgu0$c}c+$vSY+0I2scWp4(#k!RL8B
z&9U`C>uw?iBVVPOP3g|A${qY#)d+L;B^t#pG+MP~44|gI`^X<xZnSH`Avw0CnW<S_
zlq^+0zsJC&tKnAyjGwD5x3OA50)M-Tk}7p0FnlvW24#LbHdbr7r%cc1v)!uvqz9mu
z$gfp2v>A~k2fa@EeX7)`I2u+vf5`Q`R2|XCI;!L0EMk51tGC77U({p;d(|jA`qawF
zxE)U(;=G%(k`DIWXAU9LxjKJAnnQk!Nj4JUk@T+4bFpIOsbZlH*7+Q7o^;oovQ^Q5
zIn&3!fO2K}*g|vsnV48o{>>V|=4d<(Je|*S7NHHs;^sf3v;uur4Ps)jBNH9DN{ATh
za{rl1@IkHfX<1m;T~INMYn^8W9%neTb-u}~>D}UKaC&Oc_d1^xX!W6s3$dnj*|CC3
z*;Q^`?Y#YLz|89F|50lSeVyn+Q4yJ+Q@);kXfefobDd`FH8>vGT266)qtLvwo9gpH
zgMT7Kw3i0cXSF%>GEW+O5tt}7_`hV~A{zW7fk#gbW>=RQ90)O-EL%Dm+szPJJ>^pE
z7NjXVS0gcQmk&#9i*sJzu;q-|1F>OI$6MVaw1SpRU&*>=&fQ#WvIg)DY7$`pFVRz?
zSi?KDQdi6x{vPVk$w4n`XrNq~VayGMtl>H=I3G;ot(<W^2RHOtwFr6*BGc$o2{E|2
zMHbdIOf=efufU_XX?%NR8daw1rZtnSpCXA7{8)x${YZ&OzMr}PLGnlYCRxwX#3t$U
z!M`DANxk!BT^!CsygN$M=bS)do=d}dASbulB{+8?gL4<<!hjWy3}wbHS%m0Of*u=<
zX2*V*cr#Lo8fON15j4(>-6wbevT{Sq4G>(3T^e2KNxOf!uXac1&-qeZf;%?$1Erbt
z6t$;gBONsz^5H^2w2WON<2h5dTM_YsLJ`zMeTD^6AykD78AaW~wj-|J9k!&1(JuN@
z^=*WygN~Pq%SQnn>Hv6fiD#u{0u<4XmIhdZ?21ChGaA(t%$+pk594OZoffklz<wHD
zH;xwDNm%kg3x7|Alu{z?{3q4B-+3*rp11k|wE7k{%4y9I>by=_MRyo6CqZ*y8$R|N
zY)!EdLQ~*4Lbb1l-dnxUV;3j^+f%lAnwCj~Y<KinHQbuj@{|@@FZxMguqYliP3tWX
z>-P(-IKVmxl(p+2=Z(@U!)7=W#1WOjqWuz+;rttD<hBB9EWn<SjS-U&e8U2gbTp6+
z$#|RfA{!KOG;4^BEOr_V&PV}&4CAy~6){$_5HIqyM-b)XsYy1%?LUaS2Skel$M4fC
z8jm;rIB$IPOJr(_vyrNu5hO>|SqHKtlMi7^1WLA(XD~~pj7B;~Q4t#kCwM^8xsG1B
z*r@MrYg(-CSYyL(D|*}&B~Obm^cowNAEWmP<dEd~ai-mFD@u#<QB4F7?#locEnFiK
zzbjc48B9rHIaBF;8?~`(G0l`Qm`eYqvhOgD-dsB014?PLKjpltq%mc!12<|@&Kv6I
zf(A|+l7hJtJ98h2SZ_Pyya`;KpW$bH(|0niiYvmUodstLk|N@rd+EEdcPh;ZE^b^w
zu4#65;JY)99}O<j&gI~;7X>3+Hp7Gwx%nn6K`yJ$&YdVEOqg>S9-YhalcshGDKJKd
z=+BVzcFIJMoWmlWo*xOF=_oy8oi2r*eSmUu>*UPhu^>+rdlXgBuRU_EC#;^KpLf&G
zpU}?@@zafXT62Dz|K7xZZ|1-6;lJ<YzrRDjXUlFxe&wk{CG19@uIn?k8c)Q?a;I9g
zj_A@Hk1-vo@J|?peaM`v*luuy28(8++B8<G4HJP$OD4CCy6{9)?Rt{LSN!;P9fkeC
zJ5mX-valusOL1!Dy2_QFy<`Pf9;!?q<6zH$?Ro*iutB!#^Kn(&b{)*vE+xPsEm&T}
z<)yfiGqzvd_+S{hmyaZ9%~#XJ!l{Ohm6^)G`NQW{5`+N6lYkga6~#SS6yrjCpiMJ}
zDj!)XWKmyXx?I3{Zt_MbNlEwd9WDFWfQyNwvNV;f$hWA-e&-eZ;bXN6Ifn_0HpON;
zT<lc>2Ie9h@RWf5(K?{p-JsvdfR3k5SQJX3e}w}Bl^AG60Xnc<C;n{UbT`nqRG`kc
z@fKNsww>bB0!|#Yl-=Z{{Fu;tp%%2tWp(L|7^ueU;&>mHTD>N4;49^5$)>$%*V4=5
zFkDwwzM%yX{@d^D0XA7LlJf%)r*H3M=8z4=rDt|83#NS?e_$5AsWNSM#?*5DO{Ezt
zRsnEx95thwlrZ%-wM^9#bT#LI^FtN49-?})9?H*9MHkzotV$?bP(v&>J!JC^J(O{L
z?t&tPQ$k4z=Qea>y=0SrOSrhCh1^RtonJ+<#MRp^njA{sl37EoeMzPuegh<PUv8hD
zC!c$=IU_0U{*Z9_6Iw27W%Xm6CxvA7X+0F3tV$?FWc71;C^}h{P|C{c-)SLlvaG(q
z+a=@g)mc$N{07MCMAtwxPfo|xT#<~X6BGITmX@<x`Fw@*p^$vOu7{$NPYI=neEw7q
zMJJyUN?G~bF{m41Y_fa~1DU>TsUUuX<TE`h>Ls6XE>%90#6&)4wVc(;=V8u=Lh^Z?
z9*RyrC6prad0Y=gC!Z2ZT0SxB6zuCBhi;>92^a6tLM|_#1f}o3{sX*S`mmpZ_zjTH
zOY~FpJYn6PDH6$RS9-$ik7${#71;+lYYK_%<9aAMk(E%2i0rd^C_0gqP|AwzUuz+6
zvdBKi+ocbY?SsW%(Gu#c&tiBICgSOMtBs9fK`IQz2Cj>w7Eg(szM=)p23p<I;+Lt&
zey0!WUey9ALLGK3+AiNDtjzR>rB^K?EPhYO`(G_&3Ro^0okmbSWi&cwnpA}PI_-AL
z3#vYLYzDvIY0;Ks?oR!Q%7IcQsYoIAD@~wgSBY$js|i0AUg5~%XnYm3Pc%fX_e;dl
zJQIOs*H_&meR(eS51o}EO-awiF9DnEm`1e$X-C99!aa{W&Su9pzKnLtk8LDP*A?4%
zp1T$(V%bM*gCIq*4SJTb4VLV!I7C}x)ClZB@aV6Q^^9U1Q3cu<2Li@oQ%pMUD&idP
zM{9{V$DQ=V;~f0yT*o<X<lfWN29z=ZkFraC&ouJ!RyX?QiqULik--tZ;;V|iu>M@M
zG=^SgZWOy_uJ@MsNfgK!RpQN6?r?EL#NB~sxc*|xYhoMC%x#`|yN7+eX8m@t&Bo+`
z5PgoD3_rN%NCi8^I!%|mu>q%HEFknE(iE_DnN)3gIF#Z@1?yn=VF0$#5Fr`l?f7Oj
z8K#}fKuenp&&C9a8*x=U&~jNe(2}z<UPfGUz9hZkhvRC`IP{W(;n!eIVvpzde-Wfj
z&KXk9?vaRa{xR+3PJg2vNpa?+kwR@FX=e~mplI=qQhMweFO>G!UrEbk3RnoqfYdO5
zG%g#B&y<ve%;(FK`R4-TB7XcY=%MH)N)k$uiPFF5q3A~O5=x4L7t6JMxkgzqwR_U9
zFO=~8U$wZZ!zwi>wXWx1558|V!8dpl_c7-Zd@M4&rg4x)^xvoSdF%Cdl?M>_U%$J{
z)sh%YtR+cG(po>)wwBfyyvbQu$el1gtn0ZPot8)_MYQB1JrtdmNGK^Sxp_U-_tiSd
zCe%2byKdVm)9E8x+DdEH`Y80RSyRB2h7)j?$DMVL@Z@i4iPBErC6u&@ejpO{fd>@U
z&Fi7)gs%@;PU@lPgik=p5s=H*lRz5lDwH}}97+pGF*HgFO4_0pFgWP~DpBcay2b>L
ztCIsRGwESe_je4X(us)$eN4-AEkPeg(G3|+?M%&Yawy8l0FpAP2~B@io~F-i6it&+
zca@%`<q)UK2`YPiPRRXYd2+w7QRMFJo_pqy+A9~x%=?s(_>J-;er=;j+?Ae6=a83|
z1Egh>6SCeYPu360lU2k7ITvDhq<T|xCbW~t68DjP`A^43bOS2YDE2kTF6loV0XD_$
znDn3K@ME+6rw@Vu`To-cxi0_d5(@RbHk=?u{!@C6{HGU0MlLyprpu9f862rL%Jg)k
zPFeBEX=S&UOXp6>t?h|E9xE)Dv}zFhGP(QdZJ_X4SrQ9C`Du~O)LFIIkhcHe!Ajd}
z)l3}LTCXguR@_dry^3Q3D+dp9QvySw%_Fb`sCj_$HRlhe#}T&>VhMf{<#(#PB(n<_
zX$Ck$uJ5Gb$~&-4z!wheMS9{6Z2okv9oWB%okhCSj>>qmH)PB98M?Obn!xthb!dy+
zYUiP=JB_&skGv6J6M=O3h%=pY^UU0!wCC-z;JNg7{{d;<#`wGMkE`MaXD;jSUhg0T
zX&=|imKef6iUFRnCp-uDlqdYPUl?KJW_~K|NlrZ6)JExfT>2aOha@3ssn2R#N;`l1
z6p9uf_s9w9GrUlmh)r;Z|9yEf|3+Y(_O;~i!1#(Diq6*}p%n2Oy{w0#^R-AQDMz2)
zwait&ygU3gEv}_U>Ux~}_Qm`=Z{j}O;ewAvCPIokJWuq`FL(IPQC*j#H3p+7if*Nd
zmh9I<(P@c<QbbD*>7nSfL_$ev$$8}tzgA0IX{|a7?(l1bSQKfnTeU=K-Qf~STKMwa
z;rHmF=!CBiT9)-tbiyZ~Y=Ap_Rf|JuAvsI#a6xw3q82bX=>jTI>1nKH^pW;#cMqew
zzhfwsPE3YV&uW>jCFp4s-H_qbeD3_F^c+YIJ5$zhUoMEq57Fam!lf6=bLsP(ONIKG
zm-J9{sxF}vF<W2NL(!?agp%UyH6>@Ng7@76wbJz^t@S-Eg{sc=<!vpES>MCE?{Jyv
zV|XVTrj%_dQmD&0o-@ZS?;PJbrW?eSZmyEf@ljw?+^R|Ecn&`{+d2L?rUd!U@dUXp
z=lCfU>f1R^kRs<eJx9)Qtl`u<7_&=`=iqQ%K?Q|8<^3EyDBVB(6iKYdy9$e=>Jz^X
zv>cTsF|9ZOvf@yY*3B(()#ee%0m2DD=-U+@p(S16w-$7TM=ev{6<!Cn!WC}P6L*F4
zr;TuhCw7bN7RQps$7x~zhP+TM`b45B&MY+4gTu48IpDOxq}nO#Xmr8dv<&<3gDs}5
z#c6ph-Zb9?j%Dl-xMJ7nb>Xm0xJ|jQ6F0`C4S&i2MpxxoN7#jYwF$h%SchFjq-6<r
zUWlvm27^|B`=O3+!I2-d9F086b-Z>+U<cT)TF0Y{_^R=>d*e}xxPA}E<cY(^a1Y@G
zUDU@f7F_f=59_!p>ix7!CfQBGhFYDV-3h{ZYd0=dS~78^9^ZC^7cBlX{QwKiaplqu
zuhj~GDh{h#3E99bb_)s=Xlu(dGK|UBj@Q83PC7@9uJWU=I7{v^ws#Q~xY&QABtOb_
z(p86uE>%}(Av!LJ+J&8nb{lKOc~HtnRr^JX*)S<g<yTMJfmMwz5(Fe!K;A4k&&EBI
z@Azq$r*Q4z@vUskGO#;Hz7DbpbHxV#0-V?AC#i#VbD^^YaJ%S=D!Kv;^_oT#DWhnJ
znn}bEezps4RZSO%2GWskY_t<BPQP=SpT2i-`eW7fyp~tBaGyoP#Qw@5LKR1ju-$^^
zN-cxDmXICDe92Vdd<M$Mc8Piu9HP<}r$1y0vpqxuB+9qIq6WWh1<O+0rH-*=g5z`u
z;#OQq$u974bEH?t1(I|f8o<zPhVyI)inxeF2Eh@#9*F`D5!^-hz~DfoD2Z*wd4|}Q
zg_vPA7Dh0ZC*W8M({l}6Ei=z{^OShuRg%>9%}%{l#jP;NwFg&-wXiWjEy<;%hAm&V
zZ_^$*)P>`6e7a%}gh(uhBPD^tje{5;XYh&RZ6Gx=TXoUKE(*Gja+00^1b5GHt;BJd
zsI87B-+{Y3(NyB~_yC*Wo4LffQjq>8E;l=|=SsG#ZMLg);~Df5kyGG=+vM7icHmA*
z!G75ag0??5JBwt0ro9?ixTmw_EkT#|QBIBka#jNw5TeU)%=YvZzxM0k=PqxtmZ_dn
zlIq_URZnZSNo1EzZ^kx*MGga|VAcwn4FDU(Jw4DOe%s#&I{=w%R%7Li36O{Ktg$`*
zsvF=SEeoe#HRuGa5LbQTwx5uV^G$VCxvTC`Y_>}_+2J_Cnju+xpk`s`TMZ2_^bK6*
zEJW^Q++n!JZ$RbOno2XU2I*<lY<CedAnn3lG8BGiA(PE+lI%8GhEPVjgLFBB$TSyB
zY%d8`fq`Y>nz%K#uZCR9RTG!hHL6$uL>3pV^eCH<x2-}}8;io4t+8#Udy4KdWh26x
zKz$*JI3d!8ZUnf@kuIO(ax(#>AXxznEo?dVB9a|vdsC@&z!OhKZHw^>-6TR6qayn-
zC5q=M=M;6gKK<NFCkWq7KQGdug#SQ4E`3<WPlt_CDSo~r-LlHdt?@EzJpUTci*+#v
z=_^%;>yM5i<Ub_ODowUwW={t^8u!9cmy3H)ee{^&qvT_LmFg1f3ARU<Dq-M6m|DjK
znsmt#N0T3?%MU8$C&lvf{v0|F(4pONeB+}->pUD<2jbA6)}cXfhwW5ll5=S%?hzDS
zCO9Wj$n-?{M02Pqq%S-bGTdCi?W3uz7iJRU+@q;%qY+eI)8175=taCBgiY9&txJAE
zHx9#CYjw!%q0hu>$OZ%mqJsW;_y(E&B--!c@O@BnY!QXIfVB9KYXLB}kE!GxaWWUp
d9kyF*bn2uEOJvQ)Xq1DF9DvZX^P>-E{vUpOcNhQw

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/tokenization/hf_text_regression_model/index.doctree b/.doctrees/autoapi/lmflow/tokenization/hf_text_regression_model/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..a616af053ff3aeddfa51d1d55ab1f0cf5454416d
GIT binary patch
literal 60876
zcmeHQ36LE}d6q2gsdH^h9J0(4w$SR>mGK2$3xlzd4Yt<C7$UIb=h^pWcV_f1&&-on
zwgjR8$AT$C?BVbMHkXNU$|M0QKoX9mLP9E#3ZxPUq>4}}3Y!3-D8l7H{=d6trh8`I
z|7PaBJyPH*ujb8k_kZ-)fB*ge-7}{K-uZ<etYH6n8=RJ5Hs&X_YPHd<>W=RXSDTe~
zU2nMleg4ij`X~KyZ^YJ)InB0R(S2_Pl&BbHt!nEH|FG|^$KtM8b0$UoAxE#cX0s7i
zHwKJBW9YCyZj5*XuIbiv>5;dmR-dgkk4?JGBYMNUL&K_*#%$Tu=iRcc&)K@;z^!r}
zxU2ct4m7j+W_Sv%<gEP`pUX4Y(tvx{H^#h?awx8`${Q3?PJ63sW<&R<+cTOss=2Oh
z&a_?74r*hSS#?nIp-N429JmSB0}ZXtuZFAbR`g>=s~oZQSuE4`hFon9ANj@xV~erY
zILBCJZ1t?YMzgN(wcCw`Ztp$Tw2$n){=j}vN$?La=<MZ;?bX_DQ)`)fxzhH^O558>
zX?x8^RiB@<9T523>&|G;7%RL1c2DXD-bKc9KtN*~==ePNe>?oY1ODHM!h*hRlZiO)
zt!`<i?RytPpR{ZG<)vI*=v~CCP2{R&3{Jbg);4w<6OrzwjdP)gj5Qpl5zaOSH@wj+
zqrPL<*{s|WfPfaS<h{XL)H><`8NklFf)_XTfJTRzMmM_JOuMGp<w~<wc3OHRhjLf5
z^4`Wn2k(6T$vZDR8T{)bxf>#K-kpnwm8s`nGPMuk=Ctc|%|3nwfCrTAZ59Rt2H>Rh
z*!RYyUu95+I?SIv3jfubb91_lR{-2{@GBV91Mx}%lOx`sJ7+iBt%OF=q6~q)bE<n7
zsvI+`uHj!diW+W!hL1oeHUrpn8w^qduag@<`!>_89#3U-mEAm+`d|QuR<aJrHUf4^
zo3pi+AyxFQnQ9%MYR*_ww{Z=Y_T?E#lf2Ruw;f`79|vQwwxu_!W@D~A52i|{w7)-;
zf({<4nXYbYHHn1rUNI^MIa_`-BH!P6M1lokta@j(j{DwV>-Yqy!`Eelh28Qn^y`CQ
z!8?v-Zx93q^Xhw3#SN<u(<8Q0*UW}*{E;^Vzue<3m?~ExSKb=<t<Se=X2o><yV>Jo
zn%(eEL-!t^;<hmFMCJ`cRW|+j(~P{U*>k$<dlwae0<$vr!%Vcq*5$gD&|mL_ak$wS
zj9~Ci83uaCuEhSn&)91GH5g=U=N(v$>U*qk4-|YG5PW-V;0wPem{7V9-C}$#BO#jo
z)4<JLz|C7#ZpJ&a=iLwv(8!?XGzU?|$Rs=vaNte%mjJf}j|9ydKMUsMhzrRA{6g+}
zbrg>qk7pSCKc)I59Q?+o<PmJGTDf9H_CaiX*7%(9`Dx<|-V1UKJxO3h;&K^&KaLN-
zAIp6};}?dXwF#6bM&J*#j==BtJ_4=nK$q(5V6+1*S_m-+=|XRYkqQn&=R)3U^h$hi
zwxMrLC_Niz&f6rpOQl+~F~_|hEk21=q4sU@+HxUpWWHXTa#|X$&W2mCUefKOIy36G
zLfybqJ;dC%h-GD2qXqtc?5^<sQHoJ$ODtJJua|JVNA|ZHY#HeAF~8Z9?u6Fesv=(4
zIw+&(ZE0Z*#>N<c=f=>Uuytlfb8WoNxCKQ4k6LPGQK!8NFH3W(jdCEt!G}RSbdYCi
z&B_t<$#AKqSI56|*S)2y&5FBIewcojFP+^c`?qE~u3C&Y2+NelQQdafdb<qP2p8Xa
zc~<nAS|xkgYc*JF)|`@Dqg2&g&Cy+_B*-wDYz@$vXwVP~12h}8<0WXhgugDtKONhI
zXCO@|IxX4$rd_h@C9^TBLp-3O?_or$&BhM5)X?>630D4%QeAV7um?4524tJ$fYyU9
z6N?2){XJf2H4!XFtT!=xR;Z0&MG~~}mN;WM{EOLe`j6o;r;`z@4W35nPzbR@JDEnu
zjmYgdxKMOs4v6&_S`&;ZZT2?An2a0|>+#U$q*<v<zKHb^)}yKyqnPnvEfRA_J_+@T
zxFh)%g>l)@OfI%hl42XtmA5{~GzUF28bW5L)H(QAo<gE2E|}DUAQlRRPRie64Yo|M
z9ubwP2^OotO2K+B=cC8Pg7tOQOlKgd<Y;NYTd^MCY~>i$$<bACICmY#uSksJ!j$48
z_i5}oK5~Zx^h&7sH9qd>SY#WrzwFXeomLRTXA~Q<m*iSOghDT;vHCw8jGe|RK9+lh
zMmLVtXM--qSpA8poMo*3H|L|*vAP?mXD3HXQ*5l>kr+V<_pacOO0?5;G}jiBm-u+U
z0fE5B`?CV*Cq$48z$>wQj|SjCm+tRm0B(VLMds}Ng$zJa3(LtOyqxK<<Lr%(<!++K
z5L$#f*5I@Eeo;Bg5WR}?(W^yxJ!_`>?9Iv1(iF1@_lBb|(@HL<b;#6aS4;8c<-lXy
zf_wnc!Y#=5Okzsb<U;~!#@6C1fJeGOr;{~#0_qj9CO74>CfPKyoGi;jOrISs3qH<g
zSw4?7xMjItRL){q-o^Ro)w28(Yo@zp;p7&=vP7d+4&Nk0C^=0E-B?#!ax}ChbVGgS
zDsDY~iiqIW<6N2%$!z>gz^E6q@slpN=wvolt}13WuFGXMx~O6~*^3`BMRv3o_&A@v
zI14l^?8SeH%318ilbnx^_QKm01V3fvrbj$EN;l`K5=Gy4pa30xWH~y8{Rqz2Qpd_!
zJm`GEb#`4Zgt4?jvy=#)=d?tL`dv-pp@pO!H+<8mZf^L-60AzL?nV+!*&;62LmO1L
zcCvM^gL*|=q$_gSx}@~W$-G_5bl1_m;bVEM(_;b`S~aY}-JzF>%2eN=E-*~`QqD(b
z^L9yK-g4!30!vr0c4YKU#S}q%T+7Aioy^?>bO#q(dwP-K>6&QEYX!SD*wrk)erdUj
z8_thlmvO`STn2SAFDY5l$4G=`v!ow}HmK3s$&!8?>J_o1H{`RVT|0Y8Hl=mSK)s(S
zv!iLn$N5a_-(U@HTHhxsQ=Ju<)(1HsolWbdy=1gEdqt~9#`PO9RdhA3)}0vD=c@RZ
z+^WcG<LGj*okt|gJC}>^=&??*41+44bpURr?clLaq?DHZYwP!*fOQZ{!88i{cBBm|
zLPfCoI%3Zo33h5YezbW<xw(Vy+{gCfdv|ct46%1-!vVIhzjRfz;lkeJa0gG~e$?;J
zSgj5LRWGun$PetGx6Tf0uzftw0hT7!oeCZI@u-Ip1|8TY1Dnmg7mMxC<0kCc;1c99
zN;?bgo>fjJNMi9oA#AT+4aaJdrvY%|^?H0#c1$=G;<jz&SWK`V8Yd98N1GMJQ#yEn
zBsiI440ySYQHZ@OoAcI+I{M16+foCe-Wrf*0yd#BHobMoHQT4mek(h)706pJ#&)+@
zmxs+WRBQuOtnfyd>@mTu7hzF8y#~sP*nCykPCA>z$R_7;7+deg-YYw?=ba^;hd65L
z$0~46$@ew{mu0O{F`Bm2W8DrH;BN9@JNG4c><Nz8IJpk|z;PsK!>syDN776WW6!F@
zCE*H+OpA3PlwdXmjyi#wd~Xz$1$)`yT$j4Z-WwF#*NxA41AIHWG3X7?wre$}P)>go
zB8U~($7Z7?`5e1<7i^kmccWs++hxexR^IafhD-4$geMpX)-n)4L<CJzc6DO2r?|*L
zb3rgJ5*as&jPw1(t8uehTuQOI-?|4H@`gYYY=6J?cKpfUBmZtRC-~v2)80^{8C<Rp
zZcp&j9;dyv;$iT63`z*`oUqV_Ml-k<l*5*g9_*-d!Sea1QN)Xdssh6*(iFu^s&xSh
zb`>ml#TlJaag$njS2S%_txJGKb-Eb0rlfnGbi{}`=Hu2&<j0F_Hvbf8%PyiiKaN}1
z!*lBZ{ON8Yp9q^`1LeH-jCB*-3Y1~}F+S&ZQPx3Z8$Ta5kZK#P!|>d?75)&s2z!{q
zOBE^xyey`vL$T9MUBqO<)IA23xT&++aA_TdKdRJsz#TAJcj7;T)(I?xl&q8dK3?DD
zTW<*N$E?_SFdiSrlFW)(Z-mQ?x{>Tw&_G_hW!;Bd{Q&>?_~(cC=bhozyWr}y^=|g}
zJ?!uO?C%5Y?}O~`L-==r?8_^{T^v01IKHz7Gt=8{$6jsC<8e_~q_A9t{n}G(*Y+OR
z5M6868)cmDoT<|Fuq%5{iPZx4?+$$m_8`Lzm`(LsxvtHZYdRbVFvRVaR^>^!vRRF8
zu@ifaS%u1uYwsz!aDuqPw{wSgOD`(zu+2Hc-LbPYaRp;#85qN%0X`xEK=)t_SA|V6
zW4JwL43nFBg=Hi*`)&%G?AbgfxdU(ZOy<bo$<DeC$02cpzU0n$_bxx|aL9z~m5_Cr
z?TvA^!%A!ofQuXkez^`#PkMt56IL1nEYhuL#(F;4CEm&gtbwJ&>S(a`VwEk{Zup0}
zk7qNjBNr_^;?QOTMP_E@RKNpVMWkQag>)-P`Xw>a;VKV~h2tR|Y4NtQ6cQ~-L|@f~
z=&>ZxSIR`K3*c8U(Ri!z+z6)=?ywg#^Jeidl6RIQEo)gmG{=4C@Z>9-E|qhsoCkIm
z77uHivt~s-01GP!=}as1AZ$P1V$A?I@xYk14{p-^t$Ow{)9RvgCkng8+Rmt0yWpQt
zX=f%4HdEoas@t`4v7j&2n~`qchU7Y?TS>WopR;(S%wh+`-cM6G0FM*e6Oa{DZc?bE
zL-Zpwm0Q!O2#g9UDn@Ze?1)kJFw*0vN!oIE%NVs>s1}8`Q!vWN<i_YHq1jv*<?{Cd
zqkIyJ9Edc4@`GfoOaM>$R;1K_CzVQq^BYV9Ru*u6pQb{AQ&7nQ&R@_}C~yiY3OHAR
zbJhu)>{+D6HG|X;FAPmaB{wpMfSFv#<ihs>nVV8Q*))7^jF${hntTvxc`vDD605tJ
zPBLNjMKl!(tAa`vtX@Y`p|C2b6vgUclJX*B^|h>D)RjPnj_d<g#}dn{G@P!LO9dEJ
zZ$|ihBdJ+ZtdmR+neh2`nhJ$aK_v@5-%C@W@F}Pi#plOJ%8QK8$63GRz$X{JPxw^V
z`kn9@mXh%q-Hh=0HBz%AKL3&FArn6TgQh~^Q&7o*&mYlLD0~VkDn22W!qt~tE3Lw3
zkrr1DEzDexRC3St|I7L%2R^y*eZc3rbl@ostQ%rg0=y>fN6Ow#Dw~AtHmC}5p)Bs{
zMKl!(vVuw$$X-cPp&%=$6ou@oNXo@cY9Ywp!1^TzWOJyvP6}#uS20Ap#=?+Mvjwj$
zbCoQvxAH{+YT+uv=}D5Xx3ZZG-Q0;)wpcl&^GG6Dq=R?rnGN=eODP&PBx1#ZVBzmb
zyoX52nD3v}+BLU)RI9ahRaD57w_B~I?Uo&QpP+ktyNNH<WZ8uWWDOKDNm(j+&#D7!
zW*NoCbTp5@!cT}iFgDJ^W#bkrHEPS#&$bTxG#ZW^!pjw1`y8ph!sD-CbnVZ9oA~-r
z=0OdXeVTVtQajdCrTm@&{gs~bi<nHL{C*87t=uU;q!gt5@LHt&ys>!VXD9Akl=m@g
z2SBYrT=kLx3>pX$CMw#+&IQ<Z!QZ%~?q1+oz+Xb=@hsr4@QP&t+0{~K0j~-)*i-ti
z;F}YA&HKG0WIjq;1PQ<2*5uyV!<1K?Yp27;uql?@+YwLhMeqF(S-t2+G-EwCY^3LO
zUJ5_w!_W?M2CrXQ6bMgSNn><K%c)5WIq4?t90^i$kuG0I>XLNAFNUfao$&oM70L-0
zRI)hXH`7!o4^dE2obX(3z${5yo`UlwFj%55_tJ5ObxjUGoXf9HLNe2;-yo7~7ceSn
zlC7@^jJJ|XCBb<Y=Pk<$@jWyZ3Y>yU7H~dFQ=z~qs3_pfk-+#ANm=gYT+l3*!1x5~
zmmJ9C!uJ803!K3C7O7<tt6yh2$rR0ail#zgRZz)-)t}Q;D69%9MX@?MLb)1?jMYJ4
zCcc=;(2;$>>H;J%CP~ebV(nsj$b`>*G!+V;f=U*AzJjJg;Zsm4iqAtN<weHlLDnxh
zY$+GMPxuV85Nxxe@KhDOVlFK#6-1!an~@`WiqtHL&l5}!necfJO@+dzpppfj57AU8
zd<rUx9$AutdGTEn7*CLti{rCUH29;eUvl7+3*QHPE=~gDtE93?$bN;XCKF`8O;e#D
zE2w0F><?)w6l4XJqL5uNx-e6IA;><>`XvWsbEx>)q@ZMoUr;##fxa*_(IHzgPE)xX
znn>&w6zPf_vRq12$pDuJWm&?sgossyzoYr#YMSCl;&OG*O>l7wS(_}CtWN<;auOmX
z*7O-!3Ay486;d6A4Sm71Lm9Y<d&)8oYWP5;9a8kMRB4CvVZoE0c8Hivq#f2krQB%;
zq!gqb@H$94gkJ|3PvjeRK)zue6zwVB5HwIc-|#-@W1eq#7_V5qfn6<izTxH|WS@9z
zYEKD>e!aQ1*NnvbJBzop!Rh66{^(PrIn0%!_(a$gOHu5MrznV56^q%6C=tzPUku6g
zoWw}s__r&G@hwu9q$B=ysG2c}@f1yka>NCdEROikX)2UwC8#KlcrK@6bj^ae;su?G
zLEt9tRAlJOEWR0L^hFZmpE&@O6kA`D7?Y$@Nr84T4P;s+?xU$t;1pD{fb$hJ6$+e!
ziUQ6YNsL1zW!cwUge1m6)-O3$iCp+TAaj9}7^g@rlUO~$bdm|H_s~=*tO_bwu=)^9
zg~F<!QWUFCkdzl0s~=_kk^`$;_&#8D0g@PBB{fUp^D9gbneh2-nhJ$aK_v@5e@IiI
z@F}Pi#pjB(3o^|YO9(&B`XvWGx$u3$XPBPoPZHyNQnMsJOHehVBYGiCg~F$xk_DeH
zp{Y>#6jW4vVmvlm65}S4a&h}vC?Y(~`XvWGx$u3!=i($Wj*`kIA!{?$WHPC5rm0Yn
z6;!f7_FkF_1zAC*C}clGQeI@pKEnDX2V`@o__L&-WD-MA$&$qQ2bv0%#K<AbztL1O
zz$I@I<9js40!a)mZXs)vrBWh^krHbvp2Qehw;+y4!Fa<e;3gh#kVkH65+g+)OO?d<
zJBVbYCov)>6G@DZL8aVD45Sn!G4MJ_Vr=Ogs!n7zc0yL;PoQp3S&g89;#rNCLZ|br
z##MO5vKs7ask0ilW%_zkPZ^Gd_?lFY*^YzHg!d(9lU$$I>MeMW>K^tPm9l2f!CTnG
z+kx@N<kkEe)mPec?K-?^6AH3?3A~%lo+`b%VKy7LxVF}CW}9{$KASyxqpja;YjBdR
zdb9X;HhTpdU1GoHxQEqu!Qn6YL)<#Bsk!h@H^E%nHQ|%fsbwawl4{;8DlVry%OTQY
z=gPAj44Yzkmfi6@OU4hniJVKu(oz1!3d#4Je@T%<$-g}LTLaa!^D4uE^z5N}OhtVL
zFN=$AL_Kwi>?v}EassMmpVCD}V4hV{k2UPTcL-;&F+u^#GGZjG%h$h~;qBK!Ytk{}
z_Y@b?=NxA_?C>Z}g<6OSDp?j{pP;Ewp0%K&Xz(1$taGI%cSh2cB6xj~<W+V6b7dB2
z^0`m+{1xls9BVqR$1Fa+lnhIQ{!P-X*=@cooq%_icAqd)M@dK~93|16sJEUXdy6y%
z-(@Py6t4I=O@$gIf=ZTAGPs^v{*Y^FK}8uQ*K|{TuHG?{BdTlPO4jU|q-=#pRd*J0
zkF2wSQ}jLLW1k&f;<<+uitI2!MK#eU0;sP%A*pVPrb2-)hb-69R45b8siXkp{B8(D
zZ&{M`H6#y(L6VJ)f<RF%Y6gR&R*;2K0}6!qp}u_-Jzmv~N<IfyO1&9b&^t(_lL8%u
zs(o>%rc+lxg;z*v^U0=E62XA$@HEoWy~VY357SbndFBzC3N_RPl`LlK37QHu)CCn~
zo(aDHkar!+b)TG83bz;a*5^qH<uRAbhbzokZrkz~3^O_0JKiycY)h6(iMa2<&?5?0
z>&4@~Pc2M$6^#3SH%0d{WK=UPHm|4XW2xf4HRzi3xNpQ{BJTStsFXYIi<E-6FJ1?6
z-yPW^c8M79ZioTD8rtb81{^d{JO;ddLl6Ug9$v8+FuPjn81QS=IZQ+)%fVZLdJ6vb
z^_$?mMt#R1>f1%uAd2+6aBr=7Y*H+Z*(VPsjah#3!Pe(&-EknWjo%p4YrTIO?!_=3
zaBv1=OP!!{g?ld#n_?5lL_FMkF5Ulu<JPTM<P#hS3HBV|P2v4_0p9Q+VidDYu1wTt
z9Rj=+vZu%uNg1l<UXk3wDlK||_s-&C9^)LV(S!7_o1CVpP)iCyCCifH-82<yNg=2x
zOA0zz+F{oiT`4jIkCD8}o?foS0Zl&l3gE-6i&X<lLZ9m~i!(0;c+;T2yaK#mCVPw2
z$d{N3GcCNoNmHRliJ+2YlsrjOp+<?IqKuMd72y39DO=%DwEzL$pK(}Z>97qODSw9y
z@CqubiB1pjZl$SE;L9P)Zkh@OK2D_%0p80<9twkGfdaf-Y}KM>FgR)jStvDfAOgH;
zCRmNtCf%syb0WZdm{d9`&}*S;U)-tXAK+~l*V64wOPS`GH_=q6p)ROoF<XB?Q=x{s
zprXt(g~Og59JRvjMLygINeSgKm&=DM%vo;t?qP<R9PS<Om_oKCOQl4B_wsN)57+0#
z1H4~enC>bV;QdmH?q$fRW?BSzQ}nS^0p1&+YtjR}5tE4k@1;;FcYqfu1p!{X4g$Qf
z9gWFAWc(fJ6A;;bAvDl2vO8}0VE|#*wORwhRgOZ1?dpUNqC3H^P5yaLW@0vbd8$NK
z<TsgN2}>s9KJnf93qY2qMNRnd=!iE|Y1VN^sr=1=E!(!0T6VJn@i)C%nmJx-wCk<o
z)n=u%ZCen#BFk_tfz*bv6)T(>)lzgu>pCbc@h4b*X?v!#5y0i(DR^bR)I~vjPpy>Y
zE%eU=`h5TN4!mNqb9S}VvGePL-9!BemtP^apk)iZCTROTM##P2B$^yw#u4$FHUnS1
z<{wg(_FY%St#Z?@>ae*QUJnT$(1mZuYBl~fY$)4y^m6nqq2;vq`6W2i`V$P_7dFKv
z+pFWj@UcX12R0<-No3!q!)AL9d8crwguMUrw*h@j>EmjkSdD0_`*74}9irV;vXjW=
z*=<lY_wwuztE39C2-}%Y6c=-zbF79^(xcsPrKwQsD?ug8`szJ26>5DYs3_|zI&$1$
z5k*&uY`{lJUS&rj*Gh>dpL;p<DC=U?$swW7^{DzMvM{O-FG4M8&|hBB?yr!&MQY^l
znF=#SA-+XZp+<?Il4X=UMN^?hiJ+p4l4TX`ewvg`24b#JwE)rXpL1Ab>9CDwQVtCn
z?G{v26P+IIK9{CKfiH(FduS>Y_&Ak5M7yU*9twkGfuh}9Y}KM>FgR)jS*Xk=I2p}W
z-KgYqBHDc`sdQ4H*Fn|3xKrzt^$znIRf$pxfJm{s(HwrXxJ-7xWZE1oLDWTCif~j(
zkh0U`NZfmhi~H7oiF;;pEt(=~r&1th+M`IsM~jR2fqsd&UR_$Jh?iCZM2p{!#QJP;
zu|88=tSlxd5#Qby&fwvayLf#2Tcq5wQOxBI6pU|wJ(ns)d^<%SOBLU~5xOQlz8x`{
zh;P3DD&>xEBc&j|jn_eZ`>ep&Nx{N+Nc(DtmtF>SGKI7+2}0VgKJS*BjoR@N#JEfN
z>q7iPN4E>sj>hHsvnbC_#-|+Uz6K;*BWglKYCQ-W#^}k|wFt_SWjJ#{;T)ilJH8#T
zCF0w!%^2Sf`m1<+`y$|$$G0!WD;D2oS4$n=KFpBiHrdu+{NOQ<pu*cNBn+3v%|;3K
z0=H{=@b(FTbb9CTPT#vBU%(q-PBPPT1pLVi0x#pBWYWJ_&av(T!VClhM~&3v+W7tQ
zuqih2-V%>>Kev0oGFJs0kt{_cBX~<lxaVkiiVR9L{4f7s1COb<-(l|CN$r#CvN}}F
zy)H9Zr9}^OpDr%uTR6vR3?)6x{cf5H6+;nJvc!WvOjDs^D1wR-z@tOPJ%+jejO11J
zaB_LzG|#!i+#hFMtQt=e`dp7$7C6$T?lkBxuQ2yF$=)J0@~=#VnZn#p(p0EXBB*2;
zB|oF7P@_apQAWwK3UiN~ops%iYg8>jn0o*?MNcBbTRXOrLXlx^K}9vu>0$2OG!+Vb
zIb^wvrb2;_Q|UvP`x=sm!lrkD!rWYJ)uLuFIBEr1sMIibnh8#ZvDKh(@;SJ;>dnZR
zJwhs-6le~r_Qjn_3vuE!u8P&ERM!~kn~ID0hJJ|{4|OKRODjRfNFOXN*1L;~l?A_v
z80pKysV!Wp6_1gAnv`1}QMu-lf-%xh=2E4Ik*4TlsbZvup=;7(q!E*e80pneDR+z%
zDFrc7ybfZd=XYPHC%;{I3k3Z3LnE1jrl<Ee?UG$DnT=W9)*BTab_hbywAyU!a7zsx
z_SI^x3-K(?Il>;WteZ1AI(@^<FPnl|t+rb>t4_&)+9g|uMsV|R$u;y~4>2?m6>HA2
zN9;yv*DfHnYgeh+cA*)*B8FjovLj31Id%VnGbOz$>-(|M1ly1Y;d=(n#+<whc}R#6
z1WZ<eP*Nj5jWwb!OynVntBapt=`9vghHcNAy`exk(X15bs<YLR<IJ~0tKSo>EL@y<
z-KGL@=722`XTCjSoH>Aa5H3^okYykB`M@oYGjGQ$7H4Kx{fIM1?06e^G<X8uw&3DR
zk$q+#u<L_IxC=jcwBxXE>^zK_5M$KXVyxxcU)kjF99cEv#y)Qk4;W9#0p<x~b^^`D
zMD%Kk3HE9VUpLN!wvnu;m+h4nz;2AERT58UP;L&4muB0p=)SSXJDa_z4&KU)@5+E}
z`|Q0L{z-q_8*rPhR`ai0;f)4m^eX&pqYhSj>nm*=Op{v{ZB2v2ChpYW&C&j}Hw<SM
zT(dFf_}&<7|K~5snuUQ1I<5NN;A^y6TX(`i>kSHvGi_{`_D1IPhHh(a(}tGtEwQ@e
z!dtH4#~QoYbb(cU)|~gfm3w_}4QHVYbdNxRQSq~F*WlMWw^=FAXq6-Q)bWjbyfbH^
zfkxBSrQ3IV>m0q}!cm3tJiZ0nzuQ|6M#XHwK5cfw!GWf>@H=`<XK&FiA2VH}T&Zbr
z&;cK>6BWW6Yi+&NwB2$|KdOTw&+tYKtrq?q@W%9c6ZGe5Zrce84R`}peWpDJ?X5qi
z*$vPil(2QJ4tGYNBPu!ygZy#(4YNZNzun_|BW-tf;)N5A3D@VS56il3H|??kMzE&a
zpnGJs;sVmS&dX6N1FRL_+ZuZkYr>j=!Rf8%?IvrVqdtgL_BNsn4k*e171cl{AGd%s
z=Gxkv4jQk*?=$p<@114q71$wPuIeqo43IIIrUzkQ&Vrbr5wjXV!5H+`Yi-E9M7O*%
zOaR)tDjYy5hhPN7vKuDg&~9io?79u?PWaBo=vK?o+tp@yzK&4ZsL!{Wuqj`z$xu>p
z%STPe<Y3)Uu9}WEQ$s-nmqWoIfdRJ`gmGY4@IyR+8d0n~Yr-qseJ}%?@d}7Iy3IzV
z-oVzX(B*4v?HF_w|9Ra?ZwyBoAg3`0K&+W{)AbK*+~KWn>8&!G^+97yVQcZuj0U{p
zpzxk)xNggt+PfESJCm*BuF-5vHtjhWOPjDHV*pA*MvcF*-S>v!{>1hRE(U$Bw`VJ{
z=J8I{{PTk5@#W44*(16=Y3lAQqXDL1ui@5fKy1X-W?&RmUBka_*xL%2te3;V843Ug
zBM>2{y^YTChO5n&4Rg+@!T*5Y)is>=y|qqVgMoH?8%DJQiaLw^=0D4|W<?XK-C8j;
z;0HRMk2laVQz6to(ry8X^}wBLx}bk<22KNdniX)v00Y7<oQa0QY0t!pnRUc&t*P<B
zxSAOwfJ}Wxhoc?tap1tvbr)!Csz9kDWxcMMwK62b@J&UG>oIH@?XaWQX1O&x?XA)q
zNB!ycj5o}!2@GFAqS>gLFg7652ru}DqhW}fF<=Ui<$|GwHmdrp19sdyQ@PvF>Ocx^
zLu<3)k3lr1Rd#;}Z}RJL>r$M6FTy{!!fUClTky{qoR_lJ<DW@<up9ogy%DVC9&fR>
zdA-xD&S_TuG%E{VfZqla3u_Q2(0%B*arVPGz!)$FSvdSTP|(;0^Wwg6Rt}~+Z@_e5
z=JfAI&+HF@yzo#~2h%b*Qpl-s1!q%3u8~daY~=bv{Bt?}c?|zNfq#CDe^x;N&B8C4
zTll4O3qJ>9;pajud_kOrFOakFB%y^T4K3VrVd0((3s)8vt~}b_IvKl6o9ikji~#Ho
z&@jJ)CMqb$yG_2s%Ry%X*6TCPS~wh7(X(SW!_p&4iz5M5R_RQoeVA=**9Y6NDc+WN
z0@FPh&`Pt7<`L#rIJ#8hu!MPh=qhplKETjh{F}fSv@{#c3ZTjX{Bwe#9aV0x32!q9
auQ3~yTDyuvW)9IiT18VYtRi0|P5vKaWZqu@

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/tokenization/index.doctree b/.doctrees/autoapi/lmflow/tokenization/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..e2aab628542c6d378c03a602a12fe142a4ce8d0b
GIT binary patch
literal 3441
zcmai1TZ<e;6y9WaW@l#a8xzE2qd}r8x;r91NDKiHe2E!}if^H*>8_b7wz{jWOLjI8
z&_E;$^_KFF`0IT1tFG>zi_rvzbe}$T>Ri6_owL7n{+#TrRsY<cE*URorzFb?nNjUs
zFO#XwX<__p|H<$Exj%M&Mb5RfDy817K_g{cWQrF4iFdm}-Eg5#>+_qMriRO+J<d9;
z%Qm0*W7c;a!;PQ|pWHQ(PlP-_HS&xW{28f^26bp}{7h%TjW;X3^wGO+N#Kj|?7;Pt
z)_Jz>y0y<Ew;^~z{m71q8xUg@A6tW1Lu~6j(;<${R1mEp#B--0d3@?+ww%9=NqVN}
zB<NVTX~;Bu@@$XYWZUc(TW3e^@sP=!4wWqms)pxMoejT#^c@0h{)2;h7#$lDYa}W8
zFs5L*n1UfMGCDg|8gbozc-4;Cn(I`uh2+J&viA`OyMv5<h~HiOKF05p5EIg(cooU9
z+b9WF-rdY)W(EEFgE!^fy%)cpyy+wBj*KUkeacSe0gu>iq?K(&q|8gb-w52`Cp(T-
zwXf=WV@Xt35f;CRDU24$e?}D?2pNA6)!AprPY*8iflHZ}(iYjjQ?!P1$<!#ih`qO@
zkKJ^4-ddsMB#+z||965-l8mNkGnF7CRKz0tKm5qud&@b4PDm7;DylV_Yvnw3+q_7H
z&FF-SSZ_9F8Fjs!%raUU<{w@G#7SAT7aC$J<k(-jF5_8-M!s}AwSiX9D{MZdie~=d
znj2Jot7RcBpgRB`%%K$u@-sWC3Nb)o9;3~cs^|9V+@G8)QkGPCvIx^~{)gp)0#dX!
z;Rbq+Aku(*MOs5X*a|+>?bv{0ZtR%dZ`kByvEMvx4Zo3*NYhl!4eJZ=Lwhbqyck|f
zL+ms6j!~r0q9VDZRI(~(MME7U4G}TbwN90J8}7T}c2lbY$t5kNZ8wrgZLX)OF}#?9
z!4FzWo+w!uiiU9AUx=_kM$744X-#R3AF;iW>rW}@iWsQ?#ePOWj)vqVUbd8!2Cf2F
zGw-epz1xZ|B)Ins3WNI9DuK5hR5}SX8NR}mPJqdUcZU<0DWstbA78nhFe(7EWHym1
z_pjWp249s4`Zx?A4TCqMgbG?ol5=iY(oi+-s}I!NJ#yQMmQtBS(5DpXyy^ywi1xMP
z4(N;{i-s7h8>NoxWOQt&xNi5Hr~+AphN2{gq>qxNG=voV;{|e-&;m^iE${l)OisQ!
z(HzgWmOmy`DX9`hicA3VfB|Aljdy!7^9xowl@;%fR=%v5V1K#YXmsvCbM)5ohn16V
zKLh~aX@>j=1mxY0AjQ;@DMh+7yk7+kxN8dhq%zP{3E1G8Fg-@Do62H>7?BR1HE3X6
zw;Nb1nTy<24y^5DHZPMFkH~5z;DDkkhzJJvDoNE1q4aZ6sj1CmGRp&+_UWvYnkI`U
zD~O~fdCE19c->1fuE|)0Kx!{Hn?6C$Z6g?s&WM`@AXBR)4OO}u;R%;$sscTtFu_`e
zdT)W~+E*~>L1;F>rkDbSg6G`$NBbYU-IA6d=Ly)dNUMZx2g#Vg_s|#aU1m(F?+u3#
z>(lbWFj<^RHAOpJ51N$&&`b=?Q67VG^dLXE`}vp1=dPNhE9Qe_ZvLajJUW*Iyq-~Y
z%Bh)DcYqCs%;W->^$i)Lg|aXYdhQ5-%#*e=+W-<>283L?eSJ|FGD{erGJzk!yCG!2
z{kF~ty6jhrmexqrwdy^-CPJnpP9b-cG6H{4_}Fa7a+N}yKC>mR2<u}tTWSu#AcBM!
z6JVewg~PxS7y&D4JP+6vlEh97E8dw0WH7WcwC@ExV3Zo%aXp3BSweHdMS?{)%+Si~
zgL+j3R>dR|+NE2k#nZ4Vx?ZFb^e&Ky)r+GyOk#+dQcx4ts{=R%WErf#Kn-SZ*eKoM
zk~A)l5X0J}edz{1#oh_|W)Q~bWA>80V9(hz>-s@zE=sr=mb;gi?D?f-cf#`v47dB(
zcDm7%meHHttDT$demg9i;e{1aWApH@!c_UYp4l7Cyu=dqj6JY!XAu&WlRGH~utJn@
zJW4|6H+nG}i;}2^905Z4Sc(>9mFhby!dAP#WU=4eS{*uEvJVDVCM-4AS8df_u(N=d
zQmlgjAEs6+g+Tof1IE(kx3&B=()-f?h#tcq5$IN|{Ti@}7q;}YB;IPp6*9K!y$ML1
V22c#bhKEs#i!fX9M}Zxm{sT-YO~e2I

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/common/index.doctree b/.doctrees/autoapi/lmflow/utils/common/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..cccd2d6fa60dd5daf63a4b00d6be1aafa9331833
GIT binary patch
literal 47565
zcmdr#X^<ROc{=viYIRxiv64Kpv9(%OyVy8j7F)I~Y;0t&Pw=6Y$GtPX+uiEv9(VU>
zwSW^lWeW>8wi6Goa2SXUm>4PnS4@n#LI{B$6oe2cpo+_oR6%}_qzIM5kmUQ`(fwXe
z_nYq7S;-YuX?EuId+$5mcfId@-F<KGyZ`PptH{4_gI_mH=h&oHtvYU1_nYBB)vYvY
zx)U@XYwo$Pc~5gJ9Q3qB-)(pmy&0~89u>p1tDf#O4>!XR>>il5Kgq`T`+6lXT_+lD
z^csCe|KaACF&Op+W?<{WrErT~o44J?NjgWxt<_*Cc8A_Qx9G<Zjr2|rG%0p{bq#O{
zn#QJZupA+8tPT4Z&NE?;Z8~~$rZK05Ls}4c=3FBHT9LN3X4OYJ`YX2P`_KsAdmXJt
zzYSCy_4vn}RypG7^Vp{m_6OPmUTPW}j8S9Q*lw&fc7!)h8E#FV@*0k#dsB<9cVz0?
z>#hO<`5!>gpQ4COX^p_u>gE(>!4zS^6lKAb=~VS&lb#QB?L4@qF=wm_dr6zXJjz$&
zY@ox~1sppU{+|c`cf<cZNE5KdGYKVQVNYE%y=FLCbE^$oe|;&9@8MbW>fW@AM&C@(
z)EdSsjlD6!GsaF}tFewsO1#t?c|$m~KV~>t*KJZ0TSb_0KV>1EQOl$X5kQXbquq@O
z;LiZz&&EKTYuK7suDEvDuj`cptm+~C!;Q1Iz3qZ~-ge<V{BIM%T_1xB=eoYY(-&Mc
zeF=(!->B6zZ)qQ_IPf&w##9q@U{bi;3^xhC%0RQMP{Tv;&vq9UbZ=;gYf^8r8xx(u
zurFBf+(x~HRj94~z{rAp9DpH<W;HOH2ZxY#4zTVBtd7g&p`FYM6Wk$AfVN}KtuD17
z3V5#d+{Ko5Vs20YQilwJ(a;t=t!@Yd!^!FT(zH8gP2WWsSh^%d5hm%ieRs_g?v8<G
z4%c<3YB~$$V_+V{3HKJ7kgrz1Z3eoh*}_c56HIsdDD-BmH-Db3H=y+dqi{`qsTuaw
zm-Yg2O<gn+$m0*fBE1icW2Vl9eLxpju4cHat<7qJyTHyh%>+aKiLf7j1xs~s9cvMI
zxDI~n$Lh9OF@xrV<np5CIn5K{ATXcWpA3s14nPlLr<*ebN}zcQde98d%3da-!0!Re
zd~y``;A-IIyFuW$8GZ2*y<1!&J(JD27LOS_j4y*p!+9RXVXAe(0S`mRhXCLs2@PY;
zo15e|Ku3*l=LDoi{{&!jyxmOqLnkl_*)xSbXnC68_kchXde0}P47UnKuw=W=0<~~j
zQnsrw`mW??u~RsBtY%O9bq%ANfx4&rx_4A3Vsj^q<!)w{808pcDUi^(>rJ#WM~qYq
z!CWjzh+LM!YYtx3a7Y;CQ=_mg6j~y|vNo??wMvw&Y!yZvz-?L^q7ej*jUYKUBqS#*
zo-pn(4&gN54uxq}La{D<O`1nE${f+#{vD`5KUdR^=w;u~ZMzJnz%S2xZmnF^0*zRS
zu@{g3Y3a=l1fz%iz?&#tYx)6xJA8Q0(9lhK(D0m+R)PjF-6hw9?>6)(x$~^eF9n7c
zlq#B2n$t@SU$2%lzeK>5033;u{mG%Bil=LV4q->#)T;t`*E1JPN3-LmiPGfc<RS8j
zI2<4Y0WKX{iZCxNnt@Sr^u^M=soT{Oi9t)6Q!Q2Xd97gwrK6hN&=I(&*W9Ce45Ewy
zZW-iw-aJMa@CLII5PGp6P^6dW#S$`${Ek9m1_v0R3AmV!A81ZR2Vk&GXb)+BwJL%M
z+=4*rupU8qPQws-3<7aqThM_3st7ul0TxJD^6#;ZA`Kp@SQmg?Ei+o-b3Bkl?48Mm
zOp6`f7ljRK0Z)%l)MFrSDb2>3F^dW@OLc>OKRkPQb`J|}-pg?g53eXnbZtMxp;@=0
zmBLNnws5DKoH!7&ZjVe`+8$!e_;~v;$a*sofNV1wMMB2&Vw_`DU|jxC$M}NrMdNR0
zj4y>3rbRAoC`KE+LJ>`h;l)VAx{<I2R%0}loEDoZECu!#{Zi~Vxu@itR3rvs9;<;P
zNZ4b|u)!*?jl&@Fu?|wivQ_DtuGx9+h`Tsr?8r|f7?v%8jA19q?UL0$UU)&&>kUDo
zbv0eM0!HyzWPvjLZMrqg=uFKZ_t3MkR|8uJK6>^L`<rG9ukSJ@N$dAX<Ttf`r)Quq
z=-RVN(aC4_{*tUlw$Npnz0bk8PR!mb)68BLsi(l+Jwx~b_O3L>MI@FGDTosm0tXc{
zdf&hi)aX6U2Inz)AEwx48@($IEIY8|)~*{<#b)ls1dFmI?i8509~WMNl`wM)k#SK)
z_UX1xH6o7ta6^z1;oZJsfRpxZa7|?2wyRJUbZZ?BAu0CGW4nHsL#19>Ms6#N>%_>t
zD$U3#={*IOZY^O7Sh}HjvncQ#FN=IvOFRa0yKw}yX{*>^71NPq4CH>_&ZZr>iEMPm
zon_$-EH`RhnJczxuTL=PWo6YaF1!fpHSR2U(<0-NnL<c6X_?*I+OwNi0A$jV%^`DX
zq;_T|rh+<Em6DP#R=W%4P-#?_8FOG<CuS_|Fc8m+<A6v^Ijv;AJ!^M5OE{8g!|-x3
z8+I>_pf>DwHdw`RZo}R}@yL$L_U&I*YK)t&<xUe@u4aNYFDuLSCxuzR5|-<#$c%`2
z$#i2S_MrBOr7u&$t=DxtZ_)^T7WqVt(6ESsV0S*RT<3gt=QA*eN(-{=&R1bvCw3>T
zLSa&(?-Ybf&k-hosTqti(*m3EYB5XmeH=k8&9iK<ipSj2{1rtb)6%p#WyfhAgP`k|
zWE!>7V{k^n<<%~AKREnjDe~-o$@4#Jl0=6IL|;lf?6bzxx(Zb>Rxfn0`mh&p0gm{D
zvn*>5G$#-G2Ir%rxcHn;YTQlhZE(KiF(00<DUSIzTt}*LN(WpgfLev<u&<>(HreN|
zm1Qi!6oFF22ao*DjYg0ozwLn1B+s<B5fr(EsLzLkGN?8Q&u8TkW2RH0C-!N|x<`F5
zFVzzXNV1ZmRbda*qqWxXV7<i%UMc$~6kx$2o>Wi9>m$%0lF|&)6~Yx#tHn#&jNb6H
zwA~r^u;~}$MYthiY3g<~TnD)Ag=01XG+d83kP}JrTdKd1#L=3;c}J}`MbncN+X*YS
zDjXz~B*?I?#je!&bJAlp=c0Mi(HuZDDUJg;dpq>*%0e&PA`~kfHTA^`6caYX4g6~v
z&JPXO6AM~MO<-nKf>hqIy-bD7Ppg!Jay>ljtTqWN1SbUIZ+S_CDIa3XW9^3?M6IC|
z5V+9{hma>wsRC7ua-~ez$BI*oFNVFe3dQIP2j&~LP54IP50wMAQikdxROT;|w&PF|
zLt5i*h}+eO+YY+Gu;y3dPg3uJ1w~}7E7U6F=NeA?2ql?MXY5Flj3+Q-^eC(c51f-!
zveF&v379DC2TqXc9qR-5lSqm9+i~~Os>BmvzvJ?+BfNPpEx<Yv4zr8=_a^AU$n&F4
z8yuIn@qRcH!t?bY0OQj<fg~Patm3wfk=U74vo1rzE?1wqjqNG5wRN2w<*99;^#(vq
zw$x+RH9}kX={9Ib1qEd)(f%|kf|w}HyIxHTrm@hWbZKd~-<Yd=ZUxTTcTbf3BWAtK
z>I|k!bFOPcms`AsK2ZXT>v~Y*K#DD}SYo<#eI2WPHTzIpT>;%^Tu0wi+Pg2k{DPR-
zG3$DflysBJOk4WJ*wB3mW)cH0*(r`$hv2$33m@%m@8_ebh-#J^bCw28d>PuFM>CA`
zA6tJj0+?!~EgP;|HTY2A&J1%a?qEwc6I+7yjx`<13USp0ixS>*N@avfLgdmTEJX}Y
z&%iJmOZ+Q_17%`i1t-6$WhnRXDOX}~`Zwj0gsYuqGW4{5fkQ^Ee}O;5Tc1X_jyTmZ
zd~9fmn)q<_G)D!##h`!J26Vj@^luZOqv!~Vps}h=;K%Ax1hgaoUEiZppK$eJE6{Z~
zgwt#N5>SXVqz5elg%j0}usCyDNH50pmXx67LOt;F%sEl_6_u7+?*#Z#$Uwo6VpFVO
zHs{TXTww}MT&N}muL30%qt<r7Cb|Ezegl)M*E^0UlI0C@c=t29Qa^@2U{-z(e<G%x
zm2Hs;E5~FW_tEZrHTln&uU9Jhs-$T@VS@FNh+c-Y-mHew50_gw$HbLE8HO$q6Eeh8
zQ$sl;4GNP_21Sz3wczx!#FJc$x#%k)XGy0u3B%%)N2_!aD20X7g;@&;X8{(!OG&8J
zp2c9oH$^7TtwZ^GAjSLP>Pxw>!nc^iPbdM0tKC-9^FAChY89aF6G|X?sDm3NOo!Y>
zBv{j7F>d12$&E3gpHt$1dqh&$Stf;@GSRTRYSinl7nFUtK_s}l;o{v(Boux@<V6?E
z(3zRRAE7T{g@(o65d);gm;v8aGC*mO{t1SH!OUZkey)b1vPcX{9*gv!YA7moVNfKC
z)Pq4b<zI@&Bo|{o_VlWpb;mY|;!xNqtpaQc8KnYT-l?QiskRKt`(d48?~-nTCe%<q
z22-^7FlK@Z&@!!tl0#xXE7H<~`!TL@^hYcqSF2%tHAzu>`$7rrf_)iJ*+E7d$I{!$
zzL9Mo(+tk0mf@TSDOX~KXljP20=&a8G^got)lgKXhe63>dhSs}Q3(=*BAK3zN$yj&
z8!2{(g!M5~A5kLGxg8>43frNF($?t^myXqkgplJYCE}K9ohXzP<|N|APbook%$v^Q
z##Gk8kp;~fuRbiW<||6PbFjwBBf{TSLs5xvDlKo1XnDIp%TLu%+Ani1T7IR5(w>$S
zVd}x`nLx?!l`wKj6TAHqZ_ebmO;R`=s+#N!l;K+e>vl$;D!5U)d<G?t%RgTYMdk7f
z&~lL)ib^yX6xrnysbE^06p<^`usU%06a{&eMNv~^Hu4M#&#b|1Y-SnvRCruMlaufh
zQb%NG^Fx{|Nlj_e;}MVC!RL$lt*xnd`(?Vcgw?JpnJ5x3-TSjLyZyTW83;V`*wbT5
zyp>ABpvXI`bbESS4Mk;73((S3Ls2OWg_34Z6Oqf4N*EolbB8&{)FnudTt2NtT<M-O
znUm0gzgL3jm^Yn8E~%_Zk6gZ|#5)IT@)#{Ve7q9@jasdC$?}DSX}h`#(GBN3-u@O$
zpMJa@<B~k<wVsDT^myC4AHG@-z=wQ}bOKu7pc*siMxXT%c0x#2NZaxI@uu}K{0U2K
zP8;`N@uzsMT>?hEfb)=ZZR=6^%6H?GL1RYrn1Zm2_th~c*~J;-ZBI`>rH3D^%SVl(
zg5aS7KF$H3P;)MDVl^8H=R<?x5V^6N`nx!#$hiVfJ0aHv-xgnDLz8D+LpN|Q8kG@=
zO)&*Ki%?)~$KIpXnfQm&0&XcLAn{C=+crW4)%bYJvD^4~i5@)<mEhn0=tldfV~HJ5
z!%gPn<MeQYjDcMJM9F1m%w#aTp_BHR;CJVrLvG7I2XEs)CqB(SXQDK>6x}l~ULbJ9
zOeKMaskQ{TTMZrVnfFR?8#{I{MK{a)^d^1zhI|Pg=4b#;^c0>A(ARWKJPFKRcsMwe
zI+fvBFf~CA!f7RYm8@lf5Ik864uYjgx28@S_3wjafM;xoXedTNd8G1q^~yp5cz_Dv
zW$*_CPkgS5g>B-of3^)r!*nV3rXoBB68$<lA*doL=7A$ZT`W*yCzU+jPJ|G5CLk-V
zfHjM~DMUno>QB7R0$Z8ih^^vw)q?VUuewvL`F9t=I<j4R0Ck5jOHHMXvs3=!5XxWm
zlDsw^f=hA{fnr?@f4H9rD++VaEmJ_{+}bC?70eL3{07x!kond_*vDXmD4%NnF+qO>
z1|<a@+r){~TonZUmB7`9$b{Bm_!Dt;y<}Ip93@&~!FeVDUrh)TcZ5sD-Kq(&TSWYg
zU0Oy?#m(vSy?F2YiH~^k-Y3vY;LoV)3ECvfoxC_HWMe*Cgcp$#O~67t4<nOjC5eE+
z5|jvs(6v)9x1dD0+-XXrkB_(Wb=~rFW3n1;OhQgmru0srQt|;~ERR1dvIZ7YtnU-{
z%N_*VI#R4j;PiLMgtB6d$RPxm$z*d(%C8d`<Wen!oJX}TU+$`M{FgsbOmXjk%yT8h
zkpfKt!`Hyb0t$pf=o;l#AY4|eK<XhEvIT(99v#;tkpv|n)&vfrfwbz`XBerR9?yP0
z3^@&d(k;{(@Rb{Jwj;212yJ0MHGa_+Ao?Bi|Dq!N=kfmqI4Fw$t-T`5-KkkO60A>R
z{LB9lVX|H=45cK<4*%Tr*;;2xSQ<s{4=hQ>)tD>S72=BEV*dbE=o%uJX>N`@MYAS7
zPq*zO`Q+5<oru_|b%=mai<sa%UK3%Z7G%MBAR$-ffFj5~4!I)75GMO5L<MC3)i8T3
z`)WEigH%CxzQ^XfbE^y>xx$2_yLQ3D&q$p!Jj)GshEAhaUxEi)mUivpmQLA+iVMJb
z7fkCbo|u`sgfVg?Cvc!0D_AK%&!BLcWJa)3ez*^TJ}^)&=eM4OoQw5gz}9*SzkLM1
zeK`8|5%_k(`Y8E)ntXnYeEtRbe1?2}96!-2tw7oR>oT7ClBeFj&@*I~TWEhI3(>U4
z!_d?Yeaapwv4<=LK3$qUm5(*S)BWNqdweDbFPkTl*v;zrFTkRUL5^%h3RTCy5KToy
z_)0jTkgDS|pI5~8(V5p{L;Up!#Ioz*Q?UIJNONQlZtDK(vBi^WdpEa&p{`J>#K`Ie
z&MWOw%&-@g3{#%W{s@Mu0xH>zvGU~~kU?_!@}?G<XN-_Y92R?#4jAK0jWJoj?VPON
zPz>c$t@L{PfkBmSD9;TTlsqSN+tg50r&<h(#7A<PLNuziOeb-kGp2+r?mkb#ndU11
znWVTW<&zRGJ3JBR{PNjifK%mD>=ntU)ELX>wMxz^$-0{GAy+-?9cn15Q&$Ehk9-<x
zC@T46P&$>*w<;lvYkd;r^C(#t**)f1$S%m|8ClUlnwYK=I&&$N+F~L7fRepR+TKf;
zkxNKFp@yOoQU)cDkiMXXq7qUDMH12;jJYWLTCFI(81wP#O4uEXDFsuw0_$sJX$r_`
z0WSZml2T<c34@ZS0_)dmD5?sq0<^3dQabk>5|h6IYoi)g2NhV9(2k5to<Y`OsALdJ
ztx}@p6)tJ`2E=LUB@HnyEhP=l!Jxt=4G4*sG~jn$(r{+hzOJQ~;Y!%`{|$^}i8ysF
zLpPq9$#zWLm@2ko2b5AO+A*<u;cB4$ayAlnd|O&Fcw<)E#H3h@z(^CDVia#Jf+9~(
z-hzYD)04N|Pf8M6_h972KPo_o(v@)Dh|k<-n{{4<KrY`BKRh4G0pOt`SQUUz)MsO$
zGgf8z_zD4RPCnM36dj1+WZv1y7&g0dav?GxbEA(-6f__a&@HNppl=euCjzA7T!J$M
zhuoMkfgK1V<e~?T5ofXR?f`f-7kP0)R0fff{vDI^zvqvS^NHd|%n@SZwSr_Y3yh>Q
z+imGXSy$=&R@JrbjyotMqU`vo4ug{Cy4WUWX0q*eqJZH#&`6t1C?7u!Mm*wyq0kcR
z*h2}paww571H60z;bk5=Nr_f=#GCaZ5qa0f^4sqr)Tj{=^4l{J1Jpbd^4rKWlk!Ze
zFt|NK<55xAt1>#&_B0uVvrgH*q)BIbCq=U4GldqEN#t{c8S;TBwsj<tHvo4(MY$_g
z8pOGZZSaI0xbr%e!hcM_QZGKMDnJ;SQYe%Y?qBxu(S-q~pP4z8K+I1Q#8{HK8H_vx
z*GpF57jxqmi3!x)U`vQl(Wr!HxA8m*OqB{0oxk1)>-!@DO^ukKqd0_$gc5`$=qO52
znvUK!KA!3~GgpT7Z|gHNR8$6*p`t0BP({UNf$dw@$(p5RBU|51FevFXu&pB%y&0Ih
zo=hk^4TXwMBw(qLW-5v>@~G(L%U!)<8hUAhmKF^q3pYL<X(;(Etl*W23DnGBD~M2u
zhVCR9%44OL&`^MrL{zCgk%Ha=t2<4AsqqpN6o*iOP=XK@6lEw)K_4F<ZxegsVlOA|
z>^$a76bfuepqi%LseM4XGtz8#cOn@zCK<w;0xfLoNVngD>p~`!b$dkK0dUz)Hpg!0
z!2||sVPYDOkn?D~ck+dRQ+z5w$6S3+5w1omeJd>L<783Qt01T}4xw_8TcwHU5YMh2
zBGi3gD6QuR1}11#2JX(=6*~wuPX}T48(|140;C(-rWT?aCze-;e#iX(S`q&9q{3I2
zR6~&$pfA@vjYb=-C~?5coaojxWuq(abK}RF4YjRp5;k#+MxiC9@~4HU6l$38F}Ob@
zf|gRlWQjF2oowxXMP|+|!sQ4VwLA)8BjW#q98d%i#v$L$F@%XQ%0dAVo`u<C5mwW|
zN|$(ojhFdIZ0};}(%HiulrF`{k-87N9V=Mr5>J?Nnq)@gDP4LEU`tDvUW?yI=@R*N
zs!Nylsovz0r+%sU>tU8##B{k5Wi)7I{;tmnmiZvHO5{PF(P>#XE0b=3H5J1h*+vwq
zOu9Cjib$xFtV~kBX^2@>b*I?S-WdT~c8yXB-cpUyS=sBt0#~WlXuZ^(^pKw%n9>2_
z+C*ziX`rN3d4ODpp~-3wF|{bC0a<a8O$N!uMeBr8I!3c}lG{hG^GJ{XjEM+4C*nbh
zPo7eu_o|_&PQe+JJg49vQ$tamX)-7h@8Z3G>mhg|UH5_#vbeG*p;Y73WL;!?p_E2S
zyzGEPoE6fTXHd&&YKmT$;lMSjJuFmy@bi^neTU@-KS|+zHY?c;^C|pT;nU^20L}E%
z<rtTi)8#k9pu(ri2#KFA<9B|#yi@6LTaK6=2pX?};cVZdK4R|5JIt8d>&jfQx!ty2
z<<&(=5f7Q~2I{x4kt=@4JYEDvz8~C;gVGO~!{hV@>L`Fp99ZIA@3_kwAE&;GT!zz-
zlw)NoGs?HK2armO_MaslvlTdF-KEAh<9Kyb%P8iJrK66FUaBDDjfGZ&B~}l>pk!(s
z+r)S>+j1w)Q;&l1T}CF9(`UjqTeb1NGXYojO9_eR=cx!UGifI6{yN!o+k_f5B0}D1
zF=Bw4XTqi%d3FiqnRK2y+alz0o`3lQo=E5lT9qb{$;%*#DIsE^6=Wj#AfaA9K*hF>
zL~se%c|T=m-UHO960p>(&2qH}BU9{z1JtX#wQ4*r@7{#5tNH699EkF&Zv|F=im+PE
zF+o3Y2wk1r`hg2d^+P>GNJ^nfU7Q0kmXtz=)GiYv<W&Qh{S}O1UCOKej7`TAJZgS!
zr|ICI;GhgYx()LTou*VblfiAXsY2)tFC-v|jaZ_7SBN4Zhk7q?_rGB<aJKbv2g^3+
zLnG?VH2O0^T<?pJQOllS&Jq9rBKWHpQ4m)g@*g>dFmc6oFCeae2(!oHs-}bGP<i~w
z#Zh8=7jvl7uo{JL|BR6%8LEfdv4Z7L`Hq0oBr_sU4)sRBmgZ1z!fzypO1_=y9O~;c
zUn83*fm-mzw9Cz&UYd=plyf^tsna8{lwzDAD{Z0F=|VIW5zI%EsZ+(Pjag8v=n-4i
zr3lQjlcp&cznqe$k0>csrl}r+p@oyC?;wM^oHTv9b0VIi_~c2NKBtDFN}4h#d6K4I
zRzp!GO&OF<lcwKMLKb(1Cn0J2EwV1MrBF&EC0_P8BF?EvQ!0&l2DKzjQ}nv?r0M^z
z4C^~gn!c36`)pRSq-hF2Ryb+;Hb65yX&U3wk~E!zL4}j12#F_6@jFkNj%NGWmW=76
z5F%D#EILB)TI!7Ha$lRyeBE-Vi5=nv8&ob_6b>;@iZ5=SWFvv*mAx6~$|5N8Jn0oU
zC}V$jUrP=V9t;}gGNX-OO9}}SsB%I=xcw%2CZ}8k9ZmcHAVoS?XdzgB>P{GxOp#)n
z7&m6y=|qb3&p__(AQN_!B3(+rmAz0x*m;T+;brEkq|IF?DbjZkYSf4bdun_^AZ37>
zXTqKud3G!1nUo^EwMD+WDBCNL9eujc;xXxaiV!YmN3pFV>3bh=@`IF<d9$NmNWfAr
zG|P@6j7*6WvZDvOuwXnf+LgItxAM&*bVoVS_XC5!P8h7_lb{edgsw?$g}`N`3ZWh%
zB$-f$Zq9)iOERJJX_r|Ma-xsJ?3Xcyfj8|&t?B=;>3G6M%}<sSeF+C;_{>wD6Fq&S
zthH*2gq-M;z}+n{7&yBEInndT%xYu=aeXR6MlE}S{YLyZ=71uID-IdXF@%XLu6qG-
z{SeF^i>sOrmJ{W%A{R%=ZFVsy`eq=ya85Kvj-;MG*^U(~C(2`fPLs@tJULMtu%$WC
z8h#@=QS$9n=R_y#p6LYTIn8l&FHah@WY^|xcX6^2n6^Jzacec#ktS&_H}iRxgc97}
z?rRG=tqf)7t=EfBRZD+aJKH^_UyPmCaaav8K#;Y&P+Iekqp65&Jqt!idRE=ANEY|S
za$}5FGn#DK$;%YHq~ztW$YFw)oY`x=o_H;5KbwHl6#b6Lc}7W&GAHy=7z(V+liYh=
z4MmmQV^H!W_x?c*MU~uRP&!TSeNPEl+|!(d<laA%b&>TY$3n`cQ^Ywnx%XrWl=ML@
z$-NZ0U*Tld<A7{>GAqWVC7E>(3@V(=LP$KBh2MEHYecj$V=eisXThES5p+k32RhZ7
zzcR9(p@5rn<_!4+ojb#`H*sfJ!f(VGl5gE`hOvN!8)|N~Ve5M<t`q1^;KOUK&xO}H
zop-p&FTo>IqD#Bs(w@WkP9bB+7&V5eTOhW4yBH^o8JC1xs1w{voFL4PP7t1i%V%eo
zB_9dE6A;!45Z0&;VZuJ&ZFm)U0q;b3THtApZ)@bOF7R;S1rPjV^Pc8d*c-TkW;YM6
z3Ws<fy$U}YNYUzWq|)#_;6j<rH3N5!vljBA@tJVI*Mq=x7W^i>PsOcJ2R{!lVgcq=
zn_=HuG`j&W0_U#}3j5qfP;Ugy8DqmtIJlrg072fu4nH6jru%_bt2bxhX=AP%u*cgp
z!_`yGa2-VfU&eO?It;O&4bO&O>jSp};Yj5OUWMn1KOAnJhY1{b3AxbxV7T7bD*-%E
zxqJ*?&C`4^9D({qvtGu<Ad5W{j?xz0)=5iw(F}|-xyu^J07z^AJ}Z|r40}3x?YXTV
z)q#;~!XZPmgEITs8*b8%nZUn53mSgh+FRA<8VfMr$fD*sz(45W=~@k124Nv8Iue8U
znGSHPtUDFA3cZ@)U?Z5{d+}c1gzwwsi)G#OT(4|sPSu9j#DD+*ZIuAzmhxPfrI*ao
z40j~1B&M(qgQ|riblw;TK6<-+Au%}Ih%|uI`3A7a2AZ1TdRucA8W5QQ$E)yrjqWtV
zEuLO+y(&CtxDGM{U`)c;K2WoHAQL!Z@+TcZm>7;|jleC(P2n071Z{m49+O>;gb^4^
z8X!#7Jx8-~={ArSBIdaIaZ}yb8&$V_tcIerQ9o99eI1Hq<B>#4Dna?E>6?M;;auga
z>1%T~62h?T=Sl(!Hw?u1pcX{qp!<5(tvqkqKpfbTZTJl<adnSqrS9NtRao+Mp0)@}
zMSmV#9d078pR{#n0R+)DYi7{AZsYE7q^{S?@S;>5^s_=G5nJXoz#Ww$Ji`crx<5TN
z1<n3seJL;?=61aW(50=|lOO;+%kb_dSe^iuV*uLsp7*MYfS)7Yd?hhGw#3t)&!@)>
zHj3;K-J3M^V4lDLQ!r%&H5-5p2HG5GQ8h4{2M59(Ad|Inq?{1}WxtLRaw6R5FFApB
ztZbMIh7JEgetT?uuM@mzqo#R5`R)d2wGWKiLVnYqW!tT2jI}!|2E1jg0?SAB4Y*8L
z2&0cQ>HuN{unWup_)pEi36LJQVmk961I#XLMneJbUP*K_Ybd+6t5Id_A;t(qrZ%Tn
zL4%h714Gy0?V#b-3iLWs)@zz+m*E%yEiS=&6V8lp*Yb6Ho?5dL;ac4}+MH?3g#*-@
zfck<YnoiXO-3ZF`Ww=CcHUUx~SpgVYm;;_SfOb6GEVYv7dto!oHXr@ygCAy<bnnMT
z`hCp00Nv>%ehkAy$gDy9D8XLI+JPT?@WT1<(Fg}|RIpSBGQf3^aVN-_6QutM(ia}q
zw+pZ0gy0S@fw!>rv3L1$AJF=4flDGE%iUerYx>}bn-5|P@f6NQ^G-e7$wDN8m>T;i
zni~*nMAJGKvEGj#Z^93JxgdEFqJ_`ov+$XH7T!N$;r$gB9tv7`XlP+XWnskC2-l0!
zN%*zCVuFk#!-0pi1&t8zM;AtHq5VM3L7HoGt{n+H>3UkCG3q@i%{a)Rl0loL`N5xf
zK@+UQ*5q953OIBykrlUrW(>S%#kFlJc;Kn~53u&fU?m@D-UR9nueSxm!8T<SU}Pz!
g=5<js5z=+0Q?VOWREh<(YeQ8u=FINVka6<=09Uceng9R*

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/constants/index.doctree b/.doctrees/autoapi/lmflow/utils/constants/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..1dfa16691930e96c27defb8bb57b140b04a5912e
GIT binary patch
literal 75458
zcmd6w3zQsJdFS<-kw$M>wq#*sOEyOG%9`;rwq@DSp6QXt>gkbtMq_!6DD`xgrn=PA
z-E>z=8jN|w!M1R*!N=_-ga9FI91_9-1KvQq-oTzXEbA-;b2u!U1e4>KM}iOQ!*0N^
ztoQrwty{0Yb?a8$X?J{%TX$~N{l4!${=e_`^S*oFvE|E_FT?-rO|6Deshu4zmdmwz
zS#NdhHRXD#y`a~u&ZC_x-qX3Sv)3MK7SFWm?Pf{u*vsIKl2NIan|iHtykl>~+pS8q
zHSDxs-O@`|rCxKJ8w19mvHEytuQ6l~SQV?P`!3nrstYHp^)tiNMyXzFS;d;w>R=0K
zuxds>i;+a$_udYTaj;Q@h^&sW#U9e!E*LBAK_~LOy{cNN>79Igu4u0<T2`|%*S4U?
z*u#~TatnL2x>PN;T2Kk)fm(5ame!Qp4eyV+V(C;<KZ*CW?bTLs9xru_O~ysWdgJxR
zN@J%zGh);i^pR$}R@0j!XX?#UBe}_O=qCFI0&I=Yz($H~t6pqWM(99{kbxL+2V$gB
zE9+;6n=R<t#fMk4=Zt0c0IBnhA{}7k66lJt8^-N2`2TYFe-HeB1$GNYrdc7K*=w(A
z6f4b+eepuQ+^*_3?~0fJdndhNf5dfTFmH8=ZDXIY-|KnaxERJ4=4}Y3qd4CzHVj{w
z_JyPMg@t;py11*|(#yNtP7c%Q-X^%gnP+hIW;)2!RWY*D7}>4HPUA`#wlyq565Hw$
zMB!iH#(;g%?%lf@&3Z|P*)Q*!TijJ^FEkcmns)8p&4Lr&L*oNKM_O~yLD@rR7pezZ
zjUq~EO{1x|^yX>3gR$mWj84q!Al7wMfS!UjvkH4{##_NeE87B9-f)%VhYOeR33@vj
zwoPQUZMKSY?P{^9mFiWk)zC}68I4<btH}NK=C>TX_v-uZz2-jlFX{2yJa@L<3&k9`
z?xq6=QC_X~!a}jRcq>Q<mZ!bVfgYe`*mt>OZ}I)5LAL{_I`+wx%a*N$|7NnsX0)k%
z?rv=?Gn1Ll&S+!V>CwW(k(r69ynDmUlr}v&RmhHMcz<qU8gAp?ek3!&ZquL@F&}X&
z&-F}ZA~((4UL$HfI+dTz7N#>8XG}0Mw-Q%Qz@q-)ec`y=pvQAl8Hv#h40moSf5;hg
z|BwfLaBkuZsy3N9f@3$4pPtF&N3+`aL^d}z&8h>#P?*YPwR~nWOGX<m%|J~qThMOL
zW(qUfZCU6@_Slh$0vTYmde>CpaPay>ekNO(&E)WEZtBq3L;*_^hw`vew2Ay!_89ql
zGIJ;kYvFJ<?^TXv$1_KBGuniA5$_FFAIZ#&-cDNvFKW;~2$I+gUCHKQjiN8Chdh;=
zx=YJV6*3yMI0UPEa%${oE=$Jb$VC1S-Zhz>oGRR{O=rfl+H5X22~mz`3t2jV`HAV<
zweehr+*!!Z94+Lv(W$YlmMs*f3fgT)$Kl=>E7|<4Hk&CxTv`Eq&1S>*j~$tsy++IA
zVCXc?&20ZL5LU56tRZ{Qns3(IjV_qsHU%<7H6iKNK$A0-vSoA*uSJ}H0O0n1;O8TZ
z*m0=z8Thw;1Gs6LtCttKfv#-U&u}jcV8-umW4wulV`~Q*iwEj+=7AGDsJJ}ab&#L`
zhlQ<fJ@FPY1AAe&vc93$%9Yx@b{2L`zF_UQ_BH~x1XfoomfkE@eXfnCos_v#*o%&r
zMZXfAMPWxpdS<U^EOzX{#^Qb$)sF7pfZ=@p1uTfCVDA~ZVX+5cd|=1XvELxGCG5bq
z9PeK!R%#vNukF?Fm$le{eZ@-jZ?A*D^|Os?rBtyx50T4fip^T*9PG^&5750`^tKu{
zCyNKjX1J3lj;vyHUbi~-b!nCZN%voZjqz)i<_|B2;eHbIx6K&zX7Ne?Eb5UnfzyRs
z#}{Dhh;}}Q%|ug&4L%Duei|G;!4&2Q-c?)-0I1W8jGsw55{LgOh~|?J&7TIN*&7|6
zeQ!{Xa{CuC9^`Czu=sNpt^M}?1w*;<WfqFL9iA#XxY$Q<A=fJp9_9|TK%N=$p0w;W
zo{dL`c>&lEtqhkQ#{ckx!MqIG$+N-up7Ff#{k-uH_H81aOO608>DOHK<`yzi0pXK6
z{|t!51KKyQ$2lYNH#cHYjsWI5>Rb^5W|o+V&J_-D9Cza0q(DK-RiET@lSm9mw&x+B
z`Pv5&mVO0i9$$#sX77>giBI;YoONUDOfUQSn1v7}`!=*oBl}w;WG~zM{z$$=Mky$H
ztV<^O2eAc}{0XPAh`3CH$EkBw$$yZTiI+Tex8#z)%T*h1OH7`wM~XbM1y9zqy>S$l
z?o&8}RJuDN<MWC2X=lb&#QLR}DU1^9SD{@Rv0fD+*69BBN1jiTaS6&3>ypXyyV!!t
z^K(vP8F_w&I#-qFkBOOhc~W;vF3;ez_24UAzE>ZG5{+!dlWFkDd1TYH4#RTYu)>uq
zc}^bb&?nhV*hX<%y$)=MR#B8>cR;%|k`+CDj_z}RWV@P-O;EO2=O1(NLdT8tK5Ri{
zyUJ-S8gQ0Xms97evb~v@iI**PC+<y(jkNA2yTB{SoEM@ag%sD2C*JaF;WjK`4QH2T
z++A@&`&3+X7KKXUJsva5Q7S$K?b4|DhKR%)+wJ}c`ED{iK_O#ZG9llGEvS%ZoyIak
zo~F)Kh5UYECSJ(Y-I5Er>lty}R7dLBc_tOto+s<BXT)(v)Po3%{0W@-;4|X5AbeW?
zyt8gpwEmfxg%G9nr=eXMt;akgj_rMaB>yQgN<qnET{6jk16xqZf6{3zBl$n1&Q&G<
zJTVh5dFpP-CBNV%QSy8WUh~9%qu%I_l$uOqp4h9erTehVSHedf9GSmCrUO2W4`5Tp
zUHNjbAzElr8s7x%(rEnVh<q;9_5KL_*HmNvx3^5J^AEjP61X|vjgv(M{!6E^Xx!Q6
z{O8oUs=%*~ndYE7ac@#=&Vw&`g`a)JzvRVj!IN_EB`<C>AwIY*Zaone?NOXe-%DQH
zW<I6fh4(8ebvkAWqm){Nc4?F<ddZ95-~PyRii}Gz|6-kgyu}L|<=Mg(RGvAfv5Y(?
zsB=|$-cQWL=U?hh+?y2gEV>C${28%G>4=;0gnFXaF2j=jBb-$#*^9;D`gHrp&Qeg>
zMgL*U+(zm4^UyAhZm*BXurZzPk9a>$1}7+9tV<@|XRrko??;`+GUEL(b*?JjZxb`|
z;-&7ET)gwHlE@xP^zqF|{l>QBN%-Vz<2WqjS8#@@kax$%?bGtFoRy)X<)6n)b(EF|
zR;APOjS*Uo?{|OX{4+8>K{;bxGC6OD(Q@ScznsQ0a(;<ASC#V~@GkGJ$8rvCv_c=W
z7Rp)JGG5NY<|>ZEaz2cc?US?6u}{vE*iLavJrOh2QF6Wu+NDX&A~}ou-5)t;$@m22
zjCKBT7YhKl))j0)v-53EV;OzDi8@!6^CB@5pPi{Yac@#=t*6~2$$2LVzxxoWrHK1^
z(k(2#ldw!bf-^^Dx;-K?pEy6}%(F_a{cz00MTzr|p<Nnr4oBqLsNVKRk{=|a5|kv?
zC6nY+*n&#3<206$<RjF%swAH!X5uAD-7UE!54&pNd=)8Dj+}dWQoLj7jE5!oQ=BO(
z!L6KVeDZtQnNt<{y%aM^QS$p0v`Ztue-j};UQhcYy&sbi2}%#^l1XpFfGfQpIE`hb
z_X2gUD!rZHUEbZ2OV7)Ofj9HKulq&HFXDcl1ifq+(I~l-uoN>miB2|*aN?8WDBi8O
zRee*;#6`(*656GaV^lVb>TQ1{c{3T6pd_)*Kic9&j2l!9ThMHHqtjSMBiB>ss*<b|
zGx6Dwx)b*%#RfI-xhdh-Y(=sWHspyk@VO~rYX!Gq**=Q1OJ$3nn-aS9>Gv^bQK+QT
zM`C6<O1~e2c4_o0_}rAZ+x-#p`^oSGg^YE{g!}}yphAA1(^y8x4^ro<LjDRd6E9@y
zZpnpwx4ZL<{w{o^mZF>R#Cxn)uEK(S5oeJK_I1%A`lR|pXWCVy`a;Y+MoIN$XqQH+
zS4Zr81s(2>NS`MI6BH@dB@^k&L06=I=QNfP>36AfRgrE5@AB@JT%=dK%Ar3=5h+M|
znJ2!hl9~vM>IR$uDyl*1#wVbg@S@^g_4=4GjuOx)v`ZtPoe=_ZI@ceuTulZcC>E^q
z53YE5;nwshwxAhp*l8@YdiPQ1s$$WJnfQ!G-HCgXVrzPfyVCu?+7c-l|4lsM98@tE
z7STDJ4JxAbe#bs3J?u=FN-laZW(K09^cb{DBc;nCq!j34e*|?u8H%8wur8UP{s>!8
zLA~2)EF-A<sB=|8{TVS6FDUA6$pv-DQ-Jr&M)YToqE!{XlP9h5r7;{9+27&}QIT!&
z{T36A*r&9=cV<#WY2S^Rnkc3H5Za|t+TJLog~z);zvyCqn~X<LZdjL0Zm(bqD!1pH
z#xipI26e6~x1rT?n?34o$>rv~(GdLQF!Bpvk%Hsg%afe<MgyloGUH*{?T2A=WVe-L
z+$X#dyi;+vxG!dsqJ(!7v`Ztrs5csTJ?&4X+e1bqC_Sw6kF<EP;@0snwxH5`gVR_>
zdb_D}Rq5SB%*1Cp>Q3C76kA8{1LwhM_5Em{q(*!#crx@paK5CD!jgO!&LGW%J0czV
zg!vw4##J)neKAuQCCrDRT^eCVec(L0zx|QtA{m#UJh3jBJU@ahs60<Qjb-H7qRv(2
z`DtP%UY^w5lFRc>cfaUOuK#B<CA8_kohQsAOYI^o$#3GUQAuv&_rj;gZ#h$}qQ`H<
z%vqEk{~p?<(PPBVQ4dFSwm)KgmJCWzj98aUj6cN|RE+=CX)GhgXQ*>kF}_O7#EX%-
zTXHc@@|5EKQmu>>-J5wr%q^w)um~?(<BITn-@0@I@=5S=yj5|px*KeW7Dki=N1$CA
z3GRz{lF039e+0Ob3`tOcSmz&V@sh>OY6e?S0lv;@EE;F*D*Otlxt%&!72uu3OuPW8
zJ8^GPY*s6-LgIfCI#Q1DO?l!pUMtsO;kI#>Y4*G{K6Iaq&p69NC6l&d<~vHp?}2t{
zWPDSEj3qkWA5quI00l*jb;(5iD7K)Yt~!lnL|viIRYm;?VkTbH)ZLPc`faYV<bJ_D
zQrL3MdE#!o)**yN{#BfHD)KkVb;Bq0uQ^LbMd;7Otbr(@e+$~B5&A6=LYM4(f5iS3
zGDtzOV_h<_zlbfU*#EQBSVrtmQRk{+|3Ab`yx6I`B^UcqS7}j)Fht5bssT^XGre#V
z7VVB9SF{&K1?Q9LMR>d7wt73*5G|G{nO+X<(#X^~04jVE)^KEZ`y<S)WLSd2#5(_Q
zi<dHPt#7~<RG6Ec#-g!iTk8$fxvDS^5i{|^r0&GMNwKxQ&sB^7!Ci-%M2a<u5uSkG
z(a&(gLO+G8foA6`lXT1{{sm`2spRNN%nFGT{~2hPM*O46b9C1N>C=0%IvJ`U5U?&8
z5YAx>3WPbQu?!G2>Rc5FA0lSrfk53YIS|fp6(~I7NQwqR3p^B@?oS|L2t17|fI{F3
zKLmvR@e$xJoE4-(fTv=XL=*wO2JO-i;LQ>Q5cj`7`++|r;}n!X)+Lkw-(m|Y|1UU=
zW#s<^b*?J^e<o()<xky7$zOca<8V9=oDLR!lf4R$AnU+^pYWpM!*KUQ;o-0^@EumT
zt6Hzk)1%Ic;e)3hfY!emZtcIxc&nkIn|H&FC+M^p?<6NBdhrELOf9klMnkoG=y9WH
zYc1GvPD6E%k`$a~xoMIdM!IXXUbEojGUxc$ZFX0^dzO>AcWrQuiDwUvwlptyTab{h
zh2xQi*(pCUhcg9FP#tW+ab|Efr2XdZpFHvZ82jEzZ5KUgE8>vVE7G2)6*zp=*D>e(
z)n)c7ILo*DpiVrFZ6jXMS`|39)@nBcXYt0IhK58Y2f7gpcw#Ufe9KM=hC@`hM<9&i
zeOvT(Z&nNL9C6P~vDZQ1`{CSF;@aMTk&si$$lrl8*PK;t9zly2nK{=mnf&Wv@|W2|
zq$i<nn0dUFW}lc5@!~GK=J|22!MJJAYtU(QymNwhH_?%=)#_F;bjGs1V$U8j&Vf0G
z^IGw!S>GHj-^FGFN?!c;3gfqzpXr3TVlL7^+wcz=%=1Zb&`+(=u9d8ISA@%Nab4fy
zb8W}CZgjhTpWn6lcKD0w*pBGmhlNhe2jFr{2e*+C_C^f*EtliiN^D^Z-b%E$`VOQ$
zUD3~!;3)2ny@?gIVy$G<o4#p`!xbJ4$WH9LM57F9MVv4VCtbtcm2!tHeBZVPIBtF+
z95Y?sUScpu$<V=x({PG5On%2+i%T7jAcSMUgQp<cgU+dg#`E?7JzLNiwAY+$SF1!0
z)c;z|s+Tl4J{q&Y^Q7(pIO~wqdbeS0cVKMvfamLBTo2)&@R|=<jO)o_gaf8AoyZRy
zn*Ap2GBd~akY1ND`|YJiwyi1`tzx*B&Y6tn^I*tc4I@F0Wi-E!e-iQg|E|^d)6=}q
z*{f@HR@}&{_tSGc&)Mspi|p?$aEH^qmV48tTAkIg`_K~lYw4B+S*vpnyO?!`iX}zo
zfU0!q(9B@R;9Ofc6<O|JxLq=daVZX<6Q@aY90F(Hpo-dS9`@DQiO%~Y>CwE?-+l=a
z;>vYF%mlYNd(GqE&wML9h?3u{aNXJ1;c@6Uwdc%|FHUngI-waaQI|a$ZixxAhGEvD
z!gO*TCrry1rkCQRkbkK#Bq8!IhNdt^5Br@*WUs5lUOosWx)O_%z8o(y>FZXp7g746
zUf={f+#q-phSjOk7iW^5^o>(*G7~OPerEE*>B>yMBx8@THvbBW=2zh%pc3<0sDT{w
z*Z9w%`QPy-^kjaW)?@jBj`=sNK9qEB0(|~oyptpy^EoI6X(BpcmZceHAZeO0{}H1)
zjnTe{r88Lifm`|^l+KwyB9A{NkN-p-|Cv1gggm~4k3C8UcSs%>D3uCgj$mM!hy2U=
zVXTg78jwEwu{SB;12DCIsisN+wrhku|G6cc0{$Idrzt?XyB<wG>RRVM{V}ER&k0ch
z|DO?tB*g!xps6JOpB8%&;otEhga2p6UPSn(Ub^s))&4aV+}(0s9WtF~FPus0J^pcV
z^6~G5(}n*Bp%HndDhdA46T?528UC>x!v8IR|8r2Sgnwor9sXCYXZT-*C4zra>Noxm
z@?IFDCo0J7{iD8g5ouK!@q#l#djs{IFkyZHrh@to*9d`nCJgn9@H&OMpo1R9_qp~I
zFwO}|0pkaSAqg>lBQ%x7_#I*|B8)p;WH5eA>_voe>ZJ?gSnXd$(irDVQtvU2%aM<9
zFPtunpAU@^V;ns(jANN$9Lph$X8_~dpjrvz%s@Jfe*mK;yZA@2L@-WD{l++Xe<S!C
z&`bILMx<qBoX4CY+Z&|+iL;8>B1nbw$6X@?>De&SKZ)0UZ%#yyp@;P^x;7QC&IwEb
z>t7RwB*gkN&{PuZ-w}HeVcqc}gY_51UPM@@Ub?W3)v2(~nWWxh9hW2@>s~lrSpPUQ
zN{n^%#ITNKhIK55u$~31e*&tNu+9uHMF-Ps^m_pG#hINO7}hVw62Uqt^&9J>f!|Bz
zyka8ts*K>r9A~qI>CDW;R9@p4SB881>B7C?`wcLQepRO8`?anS!uN?V->=8(l<)EV
z^bntQ%_<<C6Pp6!XN4gN5nq6&l87&hy@(L+c#%PTQ|v{Ac<QAK@mTF&OhL2=UV!0D
zQtuIutCEj+FPtvK4@0BGh(}Kh@mOYv$8req6M*=upjrv>%s@KC{|QD*cJ80T5<xsE
z^&9c*_4g$qea!3caXyrR&t8y@cLd<vyX0@d{-v|V*b+&_{V%&l2=^z$-2X3lopN8g
zyB_|ZbFC}jpA(`2{$CJ=B*g#o&{PutFN?j1@b7q$!T+mbFCzR?FJ1V@>QwmWOj7Uh
zkBgI!e=nRa{67Va65}5|G5lkh;UCK({7(Y@pN48B{4)bg+rcy(x5K|-Bg6lVSR(i*
zrGDf8kpDf=$Ro>TUW!*b8e8>^SGlvtd$ZfW6K2$}&{Wi)bd3<|-x;R<6key)m+PyC
ze9bkjfP7AT3dnB=LlPpt3QZ-E|4y+N5%L``GRU{ZUPQ>JUb>Kv)&2z)M2%p_&zYp&
zBOg~LANgK5UC6&38YM<PdSb}OGDAL=L&(1qkbehMD<Pj5NQeBt#AwNm{~0V1<dagr
zkx$-o4Zj+^G;g_btt#*8$vct}rhBrF|DLmQ*u<wI`u}u|5TYLq6aDY-IwiWGgC53z
z;@VTdI439tjQ?5~k`UvsKvPMKuiF&FVQ@A~5yl-aG8o?>_9DVK_0olLtWJe-&Ls67
z<G398827^I!uYqLQDTgvCx&q>GmK+7gz=+*@$W#j62_SUrsiNejoZ233G0GlynrQw
zaZ>6x#xvu&smzQPI50iwd#=|<d=E=(S{djU$Jy@*)lUf(F8KlZdDjS``rTowSMWNe
zdR#9(v@g0w70}L!Oablh6NV&2`vcHa679bu_98;N<3$GTkBhyC&`!N{p&hGJp`9~H
zy+=E)Nj}=WaJtYw2aOV=9X&C$W0|2H%OSMi4QMYzwG!Hyfplp99!5)c=+9$`pq-TZ
zjrOtbk1yt?@`p5vcYM24>U~$ydF{cx5wflf_}#G`px;1U@>k%0;jAdOAcA{ENyh(g
z*9c+!@i6261+P=aOLWx3{NUywHNwG3efFIboC4;z3qulOek(MU#Qf!AFCxr4USu$T
zt=NkQ^VCZh=CL{z<~ftpd(7jq<YV3orwj8xg+_@nkDeIjvCJ@!<q+nN1Lj|bY9-7w
z15DM)cJQZQ7AfYdSR$AwrGFKe55BGu0<kjYh1Qn_^D!=ZvXy^tNSI-gN`2S$J+2Xg
z`C=IJ_v3XR=EWWLF#o7)S7A5L2~Gj?|3Mg%5c9teO(ik^Ik6WJ<{d9GnE$faiwN`7
zOBd#`Iu+(Qlhk|6<Fe#q-V3J-^Y4I0i7}6!80N9eFpuRB=8J&&cS5xi=9z(XnEwSv
zOLp_GV2NO!l=_YN1DuaGYLl5GQgDwt$UefRGSat48tw_m&)X7|UMi4pbd3;@>tT>@
z#_JU1k=^s~z00+xfNxGny(a_qlTQXOoGvm9L8HWEKu?SeSY~9va)=B%kYOEEE0KX2
zNJoZAjFylgk0n9|QtCGueq$m(J(J0gX0`E&Y;J5?nhH^ez;Wy;6JaaYXisP`@2nFx
z&8cWma*Yrg%!g^9<8?{{Zr3~#v|URIB;bV8dlFzD`6Ted=_0{fp;2NIpeIHGEHe^d
zIYfeaAi)W!Rw4m2z?Mj|r-FZg(Gn7T0!xGhq||Q`90+`3wJ?>-YWd7$c3NqV5Mfi9
z2-_nK_k;@n$yqaO%2QF{i>?tug-VzTe~#CkJwjCXJTm+@*O~$uI3e|(4A@UT8N6`1
z$nbe+l$Z?YiID-zj0{)~k)Z-)coM3W$iNJ+Ws-~xKgVbZ8GeB!LIzUmHyQ9NSTg{G
zT(+Ryp3M|ywA-@qwr}>>k%<EN=9Cm71Yf}t8dqioel;n^Lr<Ww<vhBa*lI`xjSa35
z0*zB)Xl%mk6dLjU^yqPkYgU0CoY;ad5M6tf^B#-Liz-(MLlV9~bR{%}fsy$Mv|Ghq
zL@y9IUS#@`6MK08e$1-tht`~*D5YNT1tJ(dZ^Ez|*~Au85G{faWO62{_lU<;$w#~w
zP8Z_WK%>NnM^6m#SZ0XFatQIK0P$;~S_$#Y0Dm6hKFRZ27%f5kd$B|iPfGnp{4IA)
z6%GgSJ^{ayS(we_q~Lya#HV<o%_?L4bt2n60s2Rs#lz-470`dzH9~-14+H)8@Hz#$
zsD~aI{?s+7fO1Y$L6l#8eFBs}EeuJB@-IPCNt8b;_98;L<3$GL-xYfip`3a_l*8zG
z6Nc5PP|lg8-lH7XBOm2nI9(`z3>qayIeKC!$1+1XmP0771Ij-L)k-L52AG_KDK!q|
zTedQkpNA!aa#HFy%CE>x9U7Y`NMU<t1h%OOWke4$J3YbiJ{Whuj#A-xk86bBxE03n
z6?mQEnDxd}@=b1I1@v+v2%vZ6o?++5@}+)B_zq!ALiUbBQ%Uw76MGS{*YP65-jiZ4
zBKA@*z+N(Z-jrdzf6)X*FW5(Orm6Sv#ns1$uNO`id@qGYiQ$W$7<{qJ;EUxDd|LqD
zH$b%#e3^lC@ck%8OZL$p!xDinDfJt^BNK=6Q-!QHksr$*lS1%i5eW7hQU>w*ptYXh
z_>0bxU^AKu$6s)b5FDQg<M=<~b&BI)pFD;Cm1{x)%A7a?P+qnF>Z_8xxbYofP(mty
z3z|w&`9-l85tSV;GF1MV*o%nD)C*9Vj-NMacp(**ITO`;EaNKVW7!L*3(KE@Mv1YE
zo*0&~%&?5*5SGsXmY;xXB`h-oOvJ(T7`G?hcRs`Nl~^KJCZ&F3`MQb8%%Lot_;EO!
zPr&BAQEU!dQwH>wt^j-V>GUj2j9+M}D4lVQ5K7-0rt~OYr<CS&%v1feYexaqoM1wz
zz9tc>%fg_9s6GKrB~jfJdl8}9@gjrjcZ<D<P))rMRO9%0lZF@kD=3Hv!48@;QN2es
zE=4}7y>PlveGnQYMm2h3sKzowHI_rDz86q^D^x3?ni)uk>MvrnWC#5vED=<bQom9C
z8)Mn=%+cJ8<|CriMmp*PNgR91xZcV&+LLYdbIv+p)0_&}Uw4fVV81I2?7zY56xiIZ
zdFubZYe@mwoNxlj9@=xTvQV7YZ%y*T$IHUVgxr1!no4r}Rk0Tlw;eAs+}?0Okg>r@
zm*@*{+c${b^kEFC;LVw`-oqQ0B_G~iI9>4m8Z=4_Z}i0Ajb#RJEQjFzE`ay5P^|=S
zW`L<Wm_+0D*;!Z~6yD=lBJd`qe#84<kc&q$Go!cDUAGj%qmI&uw5km99Rkxm!MP?>
zw&b@%?{SR~oZlbD`Ek5XaW3efM}-B~o&wZ4K?R_`9-$6rWm+X85!c@-3{J?xw?k7&
zuG?ZSBCb1LWVrsjVlSfoIQ0Tt500ZZiFiXQv~wn|_h`o@%165wP8Zr|p;2PAqbG)T
zEHkuYIfVB60qw`2S_$pUKsvNPhtZOK`M0n{&`wJIM*B@3DK+@U0DK6LyqY4#_5BgA
zqwq~CLw#G6-JTHrC(eRl^Bvr)NfP}>t`S1?2g5}FC%jIH9@RUK1iy5RDd3zF5eMfj
zIJntLg!6UVf|v>pmk{Txp{XR!cZj`+aPD}K!TBDs7ZJ{>7Z>Mj9KA`z8&ct%GjYAg
zIj&DW&b@HDaQ*@`N{n;##Bh#fhI1^3aQ-0R{6(l%!Z|a*<ehA%uE8u)oS(oF!8s}Q
z8|OFp7@5tFC4l+9c$m8el~H~^FU%gjG`<*8T#!tXAaA=y2*@7}gZvC$ry%F|%_G2r
zt|<kKbK(hLd@b3P>$P$si2sf-G$F)0&{PuQkBhyCAntgPf%q51UPKV5UI60$QS>Gd
zFQ-B{XVQ9)a9o*ugnQw1A-n;N5+fWvF@$58Asovggg*=jZ$h;a!kK||2>%I2OLpWh
zVTmA|l=_YEY;F>M#PBXHH&w`Jup>W|ozW(z#*XH)<XgN)Ch~`*cIP)m06pHkGTtwh
z@!{^;AN%whmy!S2$>sL;h|o!f09{62_e46lJt)#t)L-EmA=LlvF!fjAbxM7iE_>0P
z?;2Lf0i5W1j}z=NA17WoT{!tQC@e8f&=bQ6mKjd49Ky+O15RF5#R)Tz4ksCmmf++z
zED@ZLQonIBF`1p5D%`D2XU4PIY%VtmACwr+7P9oW0rC^mw`=3Mj2u#KiO{8-3uUal
zQPz)(mZc=hJut8Sw4aJ7cezFgQ632s<!-!Ai6Yxuk17?{xB^u;5e9aayY~oYVQea&
zP4ed6+l3JeH=L)TspN+9A+Z<HhSTvPv*G;rVlSc%C-nk}O=57p>Bd(61sQw_EBM3)
zXNr5DIB;e26NeX0SK=r@qr`~=J+Z`rWtKRw97-IIK;oEzYNdTBGmtKEd>x}DiQ^ks
zB8h{P`kgp#E@Wqp7V_HY)L2%_779}Z?Y5)ikUFGxpCb{eBFd<8a=0+sa!=C1kDO)0
zW<6Cp_(#_WNe7RG)4>n$I!y=B{qu<MFRnR-1i*<YkN^hmx%%4cjwkrq;OYy5oC=PV
z5b`UbsU+kt5PK0p-ti&>`OCy!M3ARm0P@ZNdeeqiQh}T^RlNr}u1P+~y>Plf{ya2F
z4080uK#pYwax8~H{uqG#AD~(Z<jeq*bTFaDeR|^_m_-Wmw_u5YoRs<v@+rZFEt{X!
zW-|qNomeZt^Mu)~6x>H6;GV>hGVrfV=B+15pdC_f5N(pLEx*k*LXyCT!bzZo*J%<+
z)@P3}_q#3>5(6jBKw_A=XV28+5iNgoQk%KGkj;!u@40f<p4`-ECZ`oL`NQ}Z`}a^`
z>gW-#7JrqBU2=b-YR{F!!@~*kOh?#BVV?N_G?mOVe<1cE$}^4^nLP7Zu@_OEp<W=*
z^kzQ1MTL%1r7g}v>V4Wm74g%S7fx5&szRg0X$w8Ew1s7swy+#ZTOWe7RfB4!w8adh
zOIts}Xvy=BA7hE6EmG=t+RFL)K6Yek_8KjdgD*U4>2lVaC32R~i*mxcLb^jedM9+n
zj-X&uWvl<^ELUSE$yOf;XRBrCgl4M*9rmcQIVMP{qbxW<{;Nt=yM>(;rmCILR5DeK
zh`oqXmE%PwRox=?B1%=%%WIIT&{3*X#aT$bPgN)(eyZ}q=}J|vfG!iKD)hus6_#15
z!g458eFRd~uN>{e`g;z3>x{j+ZVVE^Uvgh(udy3m8MqAoza0MG1OH!fyn}|m3Ag(U
zq)Sz8jFzOT(^w*@ij<ZjRe4YL>`e>xa=WVUFV$<7Ub9*qGOpMqc4?RIlGjUPt#OgD
zo~|<j$7|bjp4q*|L3<lbLi^iRrP@OK@BuM&ptCV!cMOv|hP^wAzR@IpR{H#`XwL;%
z?jL@8uvKq2OM1t+(mvm67Hh3)(W=yITBBK?Z|bd92j<Nlu<BN^+Bv+;Ud!&$%kXC#
zX3tn|Z!EQ&O&AN!v6Z*ivzlt9rg!r8nwD-^mD+r(V{a+fOEmtIO)`Y#jy*VAthV)5
z=g|&bHTGb=Z8h3fCvR-Z+e7nuO>Y*hdJ`-Sm5Y`R1GCV8Kh`ztbqhk(Pgc%$?BydJ
zdmRly1HY%>hPBS0?PeAJ+F;d7+FY@83a_?0#%JyGPJ)42-O_#457`@9ddaG_7qqh{
z>&=DEL-xj&)vPqMQhi~8OnKhEh}P&;oz!S&Dwd&@s>N0dx&fY?25!arrrxMGEv>4b
z)?px5*lUeq)%|n8-lCta!1!53tKDKZ4cG%^eXczZ_BNg=Hfu0`a7R-wE<nu?Ohie?
z&R~3%8jO~v*Gly=+|{v%+SbYaH|}p$p#1vaMNMxu>rKrl*2-1A31g3;mMmDkbj*EL
z28fl8y)$$rWWxMB2+rO}?Rqh=)~^p<2sO4hV;^AEwG0@dD)gyiZ>Sb)^KJMp7>sxs
z{$8QiI`-D4UaB|CT3K(vCzrrYg^ctdsNf`Y6Go&`W=p{sv^N&pR$cR|>=hMQXdBAy
zg@)!XMi^L9Q2{A8YsD%~-6m2)mO0M8SJi0g?Q&f^yMRk+vwpTwZ|R!9C0QjUOFLa@
zRp?^fq?IeJ;#?Iw;zYTci39|<9y-$kSrEm6(GAXR+Q~{4ItN7Ah9z*)(wjt;dJV0W
zVanGvi)Ub}=%0s|+gng+uyShiun?=2g^Jag+`Px$*w7mq=tBp6mdJpu$C|k!#Ezn{
zFElKx(K;|P0@bbI#-e4^Ys2;CJZR|xypsd~cWRct0Mlc^bgY5;{g+>L9gOG3=E+jX
zJl1&T-^k2^Y>T*LPwCC!if)}GJ^%$HhP6-yXG2zT4pdaOjLzXT_D)!n3z{oVw*y+M
zfh**ky}7kmvx;XmqcU$);eS}atEzQe`s-T@@bP5rZEa9>3kGT{`J4W!RqLgqGqgKP
zMiJtHiKlvlQ6@tOtxvTZ;9?_0XH_g1e~N)~uzKpHO6?@90f!6cAt>PIqe8cl?=ga}
zA{FB*0wY*53v+rI6ubx#7`kqOj|)n0*C|b3C|0Ukv0O&tlBl<!W#n*6ub!kdJ7=%d
zYo|N;_ME+j(gfrSE3r~5S3nz<1|Q3w*CAiPy4eCj!OF4#v|yvGpG1z^=LKqO#RYH$
z)nM(U`^O-{v`p$(V<j!`H6O&J`dj#LDP(K&5`1_U{`n4gXxl@0ueI197U|vR$eriN
zUFX`yZb&4=03^YK#*jTo!*!W&41{+I2i?rd(wjX{X+iq!JcPT3SHQ9JJYyA0t+4mN
z$cz&tEc0@VaugqO`0z`7coiQWTn7&}KAgWE9(Lfv$MNBDd>GyU57*+uQ~2;@e87|A
zO*~oN#KVV8Je=6XvnowI%hJT-%uGD)%)}FKOgs_C#1k$|JORVRU${5%7w}E|MQam(
z@!G^6J~i=&QBC~aJQIIc&%_^bGVw>AO#Iyt6Mt94#IL2B__cKtzvF1)cOXstN{oqL
znKALx)Fyt~+Qbj#nfReS6F*O5;^%Ele1p`)H&9J{h26wg;7xqd!o(LfOxzQixMytJ
z8~jQmbHAZf0lnhlf%QOZ5UA`vy2$)B^gdX6pv;B2des##x%K=|rF-{Kz~T_|N*bLP
zuuncM-vQEH5Vqx9fz1ivzEp1`%wZ=`s#mL2M6l~vJ?hjy3M>A-odTe%QEUPPK?5zw
nG}NI>{Xq3Px0(oUU8Pp4w#xw9`aEiFZ5b1j!*Kdq&G7#RzL_P8

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/conversation_template/base/index.doctree b/.doctrees/autoapi/lmflow/utils/conversation_template/base/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..b7af236dff7314315bd284ac6d062e7307155f9c
GIT binary patch
literal 199913
zcmeIb34k0`nK+(FCX-}x69`~9I)RW$NG2d68ZIG(I|+th(2(u)Ojpuf>FI9zNHRhc
zS3x4BqBI}^0v@Y~cmw)(y{~0mbyZl_YY$!BpK>T(sHptE?;Ta|RlWZDRaMVS+}$WW
z)z$UhcOUP4@4dQa>do^e^-RKl-rRb%P%Q5q%;odtO5Ulryy^MMaC5{dH(EEfmcOR8
zuQlM!sO9$5E6v)l)AA<46T^jKDPMESt!*uD4nE!}mg<9|{j|C>+$dJce)GcALSJFp
zw$?ylhBviQY?K`3CvRD4WJjs8XOQ<XTq*B%YV{m;kZm|4)l#nE3=ZY$P7Aw$9(p%B
zdl5DJhS#<@6@Aql;M`~x=6f@;K9PkfUZ0?CgV$RsmYvpy=1|V-&ovsg;!v{zJV72z
zDdy`)@3i4ku3m?ma6Pr08{t>e^UZ4TWhghitLE&$XPVx$Ms6qm)GEv^EGo<@98s83
zIMQ=g7b+vp>RPj0c518lRBF3cU%dWYV1)bw2-a6~#8&5;jY_UsT+Lavnz3rNkyWda
zRjZ5Tyt8+(RtH*-x@dB9s4&T!%I+zQ=Q1fA4YU;&180wc|CYjk%izD|NGGtWR%Fx-
zc)itJvDWfdk5ux_l5=XtR%YHHZ?Vd@L7{I$qm^qGjxVeVn6;sB6o{lSlWSw3PV?jq
zum7w-SWHhAYPW<2K#yl}_VXTdZK+2DP+Dj3#|tZgkJA|+=QVOe%~GzG9j=tJ^{O)*
zgQdOfd2ikomtS?_zN?<UPyTHoxQzigZ}+$;$C?vgxaKret$K50Bv%_f1GEMB>@5^7
z0qks0`MKrISH5L|54s_r-4Fjtm7P1C8eXkif63;TXE(#2i_hA;F1WK{WA@TZ)~&tZ
ztc$aoH(s=E!==hI(#EHTI~U42!|QA8tW}!T2-DHGOaqa`l*n{wvZt7D6j~ScqY%m<
zgk2!#iqx!5R-$z93bh0DbVHT=s1eaAwaOmj4^z?RqHTb-8Q{BeJ8QXWL22lny{0<4
zrZVKN+0OZ%InC0}2Kn=6Y`=oBd;lE8tg2Ja7t1@dd%<(5K+Z}HM2dA<sn~F8xsuX*
z;Z9+$eH^1!V5;}DH&t**3`=ivb+qO6RYz9=6Izbyzfc%=fi!OiU*6Wty*{879Bs=x
zIgyKPL1e@ZMsme+tMCDD8hmSvRv}7GL4>@S@ZH&4Eft50jn?(-=RLVvxwRkUIl6|s
z#u$Ojn+~m5_-SomkQ=$$PN&iG2IGmH`J3Ay2u+-yxu^$Zb2C`l!a`pl@|#tWJ8i{=
zvc0KrWZ{e8m9d{2u^A<2Z15&{@CE?*=8$QN97jf#4nP+bzGDf9Mt?tGa~)vw1|6G$
z_UL(=d^^-#tG&_d?E>R}yF|i!Y1d(J+rL{vC_E$VeU6TdV)uVgl|Z-Mfx>+jCjVPX
zLcYltKA{?~I|cG#d;2YXrtsOq=Qb2R@4e90^g{HaN*(|v_;lC=pH{g~>oAz0dkC;i
zSm9H)R`_^&D|D9vTuRK+;sF*-L;^$H9Ws*k8BJW%Y~Pc~Sep}qN64Vv3s3`aXtvPH
za2;<?MPTLbMc9lDg-p4)m7%-mbx?wKwtgA3OJeKi$Fp^U4+n<zyK5N(wHVCqse+-+
z2YtX!6eWwq?o05}E*QI~bKoiCvAc(LqsNJGA_%cxfTO7q&VYXCbr8B?jL`leuNMa*
zEf{!%k1xX0JHnpIJ46{tWlELuPCnqtg%Qs~>+Rvz>O<a)y(6VH^=b~MW~NtbPTi^P
zc36V<H_%p&U$0;zy8vZvb|WXp{-L{KCIG4EhrYzjg+CT1hU3%Lia8SD82s;ks^>l^
zG^KE)$^KOm5k8E?-H-V#7zzhL6uHDtE7$~6+uW`2io#ax2FBP*H{+H5o^*m~m|e(9
z<S+j(pbQ7do5g(B+R8|^Qiic0+$uPkkz%<xk}GAJ<zgdK*^$X*<iyy@%$`DVxR4pn
zl`}&Q{!rQFlvlz<9p-MzJ6C2ZHTY60)>*?w1@v;TzyI7yZ3JewaGZX=Q>r?(3>#Er
z8iiaVQ>|5Y7hxihb&@ORbG3YCPi_>t&J5Qan2pN-u(Qryn`zbo0*v!Vs*TaE_IN47
zy{lbb3>@fchijF2N+~uJq3J#V-VB`Y@8_Y@k#ZMv3$FfsAkpN5oV{=#W$rLhA94ai
z_Hk_(u^5Au%l+I$a>rL{`=lt+)RYBn86!yHJm*<-KRnNwA3;n>Z2s)KDU)NfOKkqY
zgxEf&wM%TKO-`2Byx2Cei4u-(bsWbr249L(0Pb(O+N-k(>|VS~=5XMj$Yg#c#7G5S
zq#aX>UvS`UGa2vFF>vEtKNpTo-$9&Si7aWGr-$P-q=OR}RhcF%7YN%6E=Y@HA|#6-
zC}7A5-6yz!WB9-GeXrz)9ecm$H!=CU^Sab?*uPzheB=IYhL3>a*DfO!!OpL}6#Ag#
z($24Cp<R*z+ljV*Eg<*+@n`E9r}U8--cwDH)>|BzU5PFD$m~MVm}0;5XXkO??fuzH
z(x#2ld8O|t@Kmo0v8R3bL^=-uPxYGEBB3Ro%u^}T%33CG@2>O_fXQbm)5=|f5%ym9
zAuqZ2nxo^RxUl<44Y6}!_d*|(4YhM&4?(*mF3f&@S>x<@mw4LcW$$6^()}3Ti|5B4
z#unU<y;C%%WFY<6T^#sdnIHQ>h&>0qANy`>k&J~O+wA8>q4@)QM;2;m@?@8e0hF*W
zn=-}sWphI)Db8#vKGilcw{vDa&<ABm?VQ<MXqUv9ono7hg~>iZva&yOhoX%#@t!IN
zTDLJPTa1$9-t3PtOy?O6{IATLy?|8H191F!qCl9k3mIUzNN9;C%gQeC{fstsY46Ll
zW+qRzvC9y`UTiD!k!NJHHDnYYwvAK{J0JFP=!23~J0G?a+9mN}C)@fko#+F^gI&hB
zrRQUKFP;Y*#TMLyZ4!+s5lC~{$btWrc`z@;o&!D~+aFsbW8uO6#`iC#nuxt4GqpDP
zvu$GvFYMhuhCJrp?O+pzij#YQ)E_%1_YvrWvaohe?z7M?iIcNlM{1<~0P%JAGp6aj
z4)4YDb>G4k+}FKdG^QjWecgLG@V36L!@^?h<|deV{!i8o?Q*aL!NHA%9hXSEf{9bC
zZKgA4(TT~bca0<Tu#Hdc^=*8SNzVn^im6Y-7AePMQ=kt@!P=Snd}x=%)YsbPGZYIB
z5Q~3~8=W>j!+WYsXdA)t*)b?ZZt;H-jVXahi~j=$p4Q?|>L7m|98FCy_Mo4GB~7B}
z-~PGqswVf_tSF|tCigD5X<EhPe#7vzI{I_(g$LaC<5S@JVC|B!P{;i`+zzi`^7ml{
z%a@c}%lIxP>?^!>2^S6#YnSG(XM3D7pzaM=f>Y;SVWAhfA5<PHs4#8KUZoQ8!L>{!
zzXe;%lmV0m_0<+_*D~oFv-)rm4(!A6)`%S-1I2QNZ!ECgq_sT${sC>5mO>0)@c^4i
zdSL-kbY~Q9Gns?GWa~xP64z+fw4H!*3kc>_Y)1-v6gMW~?m4*`u`tyeuyt8dcJ>Vd
zq2A_;$heNfn+Y_mf(3UBId3-N%2rXa@7h)rO|{)yvEN1RO@8-G1D1dWO!8(hibLeO
zH{+u`*OV*@(7ey@$BgE5M3dt<9eeMD-o5S6^A;%E!gd#(J;ShFwB^l}m)Tr-xKODn
za_W%cXggR=$a1Bv+wO6M&6lvbrkHOr_9#ISOw9dMm<Z0Bz`40^fG3#afQ_BNp_bQ=
ze1c6ouy<A8NaXd2eL98DdQ<u4n?j#AeMhrYV!Y(=`?HP8a27Ucp~LtryLTPz4q|tM
z#}K!-AZ|x;p@Sa03BQ8xVDg{E<Uh<kk3A=`+kfRG%kJc|%7A2C+c3a)PxR(<ja-;Y
zv5U$5DRksb15U7wOzwZ<SEf1Y_i|+w-%q;Vn^vyK%Q^D)D!z|xzc)+#D8J8#Cj@zQ
z|Dn0%io7SE!=4ac-Dxzy!L{}yiJoAU^q*qoXfn^~UXO&G6Isxd8Q^On!Hrsa8StMe
z)=$x{)O{mhsRxDu_pQo3&#)Fv`p>ycn{wrHrNQ=7D+=X3f9q+-G+=}0n#?jr@3J#i
zW=><zgB`%WR#;EGCgVSnS$4)55Kx(4UMQ^~@IkYtavhdRr@DFGmkPQkBi%Aw1)VGm
zb)qL0UeBIZIInOI#wLqg1W&EU*8OikZ`9|5W2Sg)bI5(0%2)f4AzZV4gMjAaeI$&|
z+kH1+<h}<!I>!?OJ?~fOai=3F9z+Z(6QK0s4k$f<C_PA{Bw~mGrN38D8mI7pgqy=d
zfL|m$^gtsX9^9`Z!QY4w{B5`cj{ZCNr_X%^A3{j(qx?QzU)ORUllMcO|5#AlDewsQ
z{O)(*v5=J){C$aIv%*j4!kDZ~476#Rpiz1T-KfKvF$}(#F)jn{yu#-W?0)XwQFKcA
zYzbCzE1pCs5Nx$(KSzGSJ>jAy3&BbT*?$Raryg-&3A~@*tG5CI_QWB;ZQ|35+#3Gj
zq<~`zun27kzx=l#_7y8I&aGIHS(CY#?~dNOjIH8dwlc#nvc)`J*V(FgeixSUFWc6i
zfq$zqo1r=G3TVJifnm9E02cZ4U$+#Tvi?)1h)eq!2);Xjcf3+O#Rff@dZE%R<#A8J
zcK-ABLDnIzW6c+L>~J9O9d_XN#vVriXzZzA*K7{~?|o0PRKhMGxWkzq4JVI#2>4HM
zmImu@2=)~~m+WTol}>F?JPUafY-FgzAE3TIusj2@Dpc}pzeC9pyBjhRK)zUmw5>GC
zx&*`^F$LViUGpFOz6LXw!s!jgT&Xm=G6SP1AO}$yssIfL9JV>+Gfsu$Delyq%#NCK
zWfRhr(kN_z05G}XTBTn1nbPq0X4C;$NA9MO%-!vji^w^KDR2Y2$ZTH*Fgd}?wlk>P
zmjPV=4m?(K*v_6Z^bRQIh8s=T4uY5p9682y*bU;8S=ZZ9skTeC*dEDpwxg)FGpe=^
zW?&*K<Lu4BVJprWt~lGbZ?BF5$9Zp&<^$OM>As$d?Co@gkmVeW?b~^OXYBw}%P<MB
zZ7E2GXyr>0h%7<c;Y&`C$yr~_OJaNhagDsB<iLq1nLILS#2Fa^FiZuDJ2In{W(E>S
z#xubX;H6Y(uJEjP)I5#`oW20ORP=~7<|dF)$jcPV^@fvUNbwCbKvq~dtnUt%$esZm
zZ)<+Vas%X8%WTJT;#T;!Z9DAmDd!oEVFSqIs(6Znur%PYGzQ5;-2Efm2JQvQz=1d!
zNDnjhX0=*jCOYag;F%RGgmjo=x~vSR@oWVytOU9nE5T6#D)4oixJ0bDJ6g+!E>c#i
zKf>KH@*htCXLyjI?tslr?64oBb4M3`aOj~dxXBU4*<b$BbZ)f03tg4zqWjzo;^>y^
zjQBQ?7P{DkZc}PWq+9+&e3!}94V9})?sAM=Wg0gbxX48A5t{e+kBwU#6PL(Up&Rb7
zomC}wg^AoCm_6kJ#b_7K8-vKDx*Ko0OxgTL`oI^$9$v*Kg&u;v@x2%_mErC?VkMS$
z@Wk?M@COn~t-gWx)X8Uk#wV0wySL!eJT+Yjnd)14hma?;S%4xW(&`MJw>oA{Do6Hb
z%631x=50)_Kf30&l^OX3;+_1?L#zD8HmJgvZ3?3v3eiUnA>0S>X-*fC$+PbufG8s3
zqX!HyGvPq5lG88`FYtjG7@NXj$2+u5=FNE!4Br--pOdU=_j51?c{C0O*ugk>Tp667
z=6;Kbm*&_vv3IT!*yNvB6ZfCtj~tPD7j`cUhM40%j?l%t(pjm&C~EU0cC6H7K#X(Q
z*@Ia;lZg#q6lL(H%d}961#KE<9s@D_F4QB<*5BaMi(EtlE!OK9RJsZyN7}3}(bdr3
z&scvnXhovfgiEwN2(ORpf6MDxh1&+)e?-&+jrb}j!B^<Zv6T0J*-VPT-vO7+Fr^<T
zEWG@j{}XMxalH_&V(Ulr>5j<#E!dOqT4!scf9ZVYLZtL31t1<O-5ne1j&=pY3n327
z3r%G+21<7wu~TLuRP6lQI(9Eih24oT_Kr8#VNFRNPxU1U;cV>uRCrI9<6UcX78s^v
zByQyN{3}u++KbaZ)<{kh6GRfRsVtLInwL@`Ge@6zQu+-mW&79#;mQw>CqS$V$13ma
z^ushx{@v1K-RqEQUP(Da&XOe?0xBe5ABUBAu%-JtXwhj|lAKUK#l)Il|1TVUGFd{P
z^Y%?B@rvMY4+&lm4;|dryF!hvf&oHg!Jrr2tu-nZ8S(?E&?!(eveD^>P>75%BTJBM
z=V?y$&B^_m63qFCQXe{PG>>T#&vkHT<mck<w~^&y{Pe<PuoQMY5)z=Uj8nJcWVDP1
zQ{aoz^!uTvR;q~bget-mx6&Jp$V_E2T%t3>pC4Dcc`_AJQ`tnj(p?xUJ8d}=wB?@}
zp)^mFVH-C2dWR<N^WXt@5k94l*|2dW4xW@?j!d#PQQKhLJb#*=0FzwsqxnmGDmvQO
zk)0><lxq4B)YFn=`k{cf;}X$!Hqf>NT6PHEuA&jx<me7f+G+$g<{Ck2tovns;Lj%q
z=_WS@O>jQ(jqnv#(!53Sir&^F#iTfeqyhI<72Zq*9+T&QdnfrjqyuntJS%)F#=bd@
zXk9TudkGVfK0%B3)R`!np~?j9Il$11S%>a}DgjHa>Tyz)EuRi_dRGWv(q83+E8@u~
zT-~oIxQZnje!TJyZ!1`dFY!cx<JEC+^nJ=YVCX1gC{2@M2iSzmfEJr#2Pigf2XvDg
zfKOtY4Nd3u@U@dexE7o6^(EATHw*~Xp!XHfWCXvmWu`RgyEp{7`bMtXf!R|qkrC~!
zFt8AN(c;Cp!eQ8f7oxDbeYrVO9fbq+Gm95Xd6Lg?i2{Bf2C>wMllt{Fd>=&d2Wx}`
z7heXcM50ynd@bmOSfR;SFXM|0`7-_sV_-$-8J3Gj2P54KvO<x*EHo^R&dWmi1P99u
zU2PrUw~%#w^ctQ7`40Z1OquJNGT_Eo`u9!0E4H9(3&ez(pjmPZovf~edjeoF0N*~4
ztv1WJ4kq-bA+KkRR6Mu}9L7(5giI?fy`E}xb?JX$lSS@-z#l;W5EcD4OIUwP0&y2Z
z7wy)tawrA=B!1yIAu@?dnnKHz>6Dz*il2JIJzx6mzoCMyT-27}`;gdcEAT}sP!1f{
z-&Fx?;DdKiumz7e19Mxzrk?-{d@1*A^<}YD!T&H4fA?wVBElw37@!A-`g%~}tLCzz
zUZX7Xn<f@0NhJL5fm#fZsCYf~s9GF{O?<V`>0K#7^{iCrJ%t2fL9dnat)roIRz7zb
z^x(^`KUjPyiBfL`EWDJ4dlLK{k!E}z=4dFLb;gQ^6&gxh9!{5wHYEpx8v+0J(J;Hj
zzZ+;MUE<$uG?cjfLxaVou9AKJ_kervr6I=V9^d7F^;&fe!UHsv&fIIC3F?WB*qmSp
zpX0xB-dorvH!-iK4{(mGrvFFg(Q*Hg#cF!KjGis1e-zj*7Sx;exrjA2uNTYg@oCI_
z-vY}4MMk&<9Q8kmpbRlQy3D@qE=%l4Y^jnR$Xr;LrPxw6!%zP7eO2z`fQ|bE67mFI
zJ&9LO`d3fE)qeLE?DLoG^J(_^EB5(o_W2w9M0}L}Pb?wq!v3fAV%*!UwD+v!X|EiC
zr_Fr*p}Zma?VSOt!INMAvA~z~;ji7ve<%B0vHW*yIRCXAlyegF9K)R79_j-$<+o&x
zu>hJ{Cb&y!D3pKaPz(v~6iD2ZWa%UlM7!kHoey>uoPH2ZZW)wBiERw_oJZnL=KlhU
zF6ygg;On_s;m~C?6e<|Spk<hbl1$Wk$>2&7L~>Chl(81|wIuG{67@|qlrD+-9vVs_
zQTJjwMp4=ikuVa=nuAFrY~A9_1J*B+$Xhtq&Tj8pB#6ZPN$vJ@_L#HdpOCm)u*Yg_
z_G=nSXDxMgTFlP1r=Q9Jx<$)e8cJtc3<?#txFbmzi8X3`Ebc@Sc?&TZ7!)d4pGks9
z%pV4&dk)ELahH;~Td+qT!>Gg>$D@XHb^^wuheyYwa?oHNk_Ko;hNORne9?M?w=yKH
z1LvtBsXwfcqf%@T8kKH^QR#C4pVwm&j!I<*DMqDlKw8+S^ai|Qqf&M?(WBA}M{@OD
z-5QdnsKg)|t+eCOUfv6?(a|e0?*yjHN(_4@7>T|cf9lRi^lrZ^HWJN-N1}5!R<VSV
zD{UQuxi)dgM<`U9hnx@iJ;xq$egv9Y4mrO_L!m+?hhm5Xy-2E(oPLXhp?XKPpwe;7
z$-yLw1u+=(6B2oHBqE^L<VC-xp->}{7_{`vr1ai`7L%|yz^fD$Vb3LDSPEOb-yPrn
zF5ki4%iyGtcFdUE$*_Aj*P0HSKp3#fuLnuKki1_3O~IUXJv!Z=yf|Iq{%JH6N{<8-
zn?me`G!)8da41j!;QR}GGX9|Y<n-d2fE3CwrO!f-Ie!XNERsLn4*noX{!|7oQuz~)
zmNI$5r$fn87Lunsg);)3OE#r1J4lf{U5kXU<mo!RV#yP`n&{+dQ#M<z;C59wS0S5)
zz56LvwD40`U^&Zq$>;SobW&$ibG7aUmJ0``E^+Z@YtKkG6uKK2_FW)9)v!}6d%4~3
ziX~ePEJ&K<9mJL@z|nyGz@gAN7v5w+MB7r7J|epxK`e5kklYDyCr3v|WIFH7ekDLY
z;#Y0~@{;(KFGD-ZuY{y89E`WP0|6kLkpFRj`EL_qPV9eV2c-Y$<JM=YC;kwrWzOf{
z@QOJfb~Vw?XMGm7S66FY^E+p!@jD!o)GlYfLer`<@a!p*)0r-xlKv*R5xAOXgyKlK
zr*JjD_Pb)PrU1J5LAd3w-$~|HqP>Cx<EN6=95}DHTvs>Ev)N@fgpICS3zhW_nMFBS
z%dESbWxe@;5VhP^_quZt{Y7q+Jrm(|pJ#GfLfOp@H^4jMaGwIDN#bx%1&&Y-H&B1+
zYmaD;cGS%lfZM+mDxTQQ$_}Kt=_$Kdxk0$uOOZzAW;f#%bF=JfqTTGpSzPYeH7}cD
z)rLSMwUeC}P}JVXPV=FOH5~GL;9*D5P`fj8R`k1K9(GrEJglMDc>u`w+cA9Vejh>8
z1l$Dp)N28u*gn<3kO|0S4YdWSTxORVpdE3kd0;~lm--=SN4ZpA{iR1;Wb&xOF@*Eh
zUEol62!%=PP-O>69qLz+LgrBa0k4=tWmnyBsJeGo)gM!Mv6xRDx=T_Zj(OJ)U1kgU
z9`y<M7oPxjm6U)IEbf!Iq}jZ38l2%d0&5WAAj>2Nm?wBqx?UGybSQqe^9~i|CG&il
zo3=rutLKyd0%WBI1*`ecXF`qiT%DW`MR?TyEIqdrYknX1;jwU21|uE+1IcI3);XmU
zJrL(Wasxf;ERFjl(#zwxq6$GIKfD7fF?J&vP#OOa!w5FVk^sL*Ecq#L$vT$EgUFG!
zB9Jhs@*r}&ADYf6Ls<BKC<(A#V^|nawoV^*HXOBEUz73Q$(0)#9z5&pwfsx~JXT|x
zctV-+KRqPJphV-%H+NLHTV_M<dDvJadjgo!^#!4>L)u!nkMDWvtB&HEJ@knb-~&5G
zWgYZE<t&QCwBp~T&@NbTD(XQq@oc%h)lzY|*no{wxY=q?u2ydCFMOu(*}~_z&VSyU
zCP_ggZWC+Os~`#F73Dbad3mM_KEFmV0OlEkEw<~=cS===1N=P}>@Z}0M6rVhf7csu
zcyO+k&+N(JQOB9#nv-k5nbPG92olbthBNQ_i{*k-E5d`r&Y+ym=yI=to$nGf$S9}_
zF%-sq>KhvM?uXQEXXK7kp-tp7965k63b@7Il`4fI<a&L&U8(0b95|O7MPxmXAEW~Q
zrUaP6Ut>F!<~JXdphKg(fI#l9I>-9yK~K?j5N}5X;5HsS7wE?wq<+NOmHIZnD;8q+
zgTvF3jzifQB9Wvz&Z3C>eGsNG8M!j+UT6v!+0-{bMMI(Ln;c5ug#C@C;q`+LlQ2}f
z)cj(nh4?Yo4~me*wwMYA>1Af~irpWQ$XkfPz##Hmw#P|;iMeHfX}!XrBkRm-3!ft?
zv0z>MahNgh?PQdbw@Ul7v5Z6K@La|mO~|Z$0W<~s)78toieV`Yg<8cRpxDrI5)Fm2
zat;L&GUQ(1OYyck5K1qq@z0~ZgA4&m$<i>v74LrwR4bBK-vnB2omaEGjlbh(3vf`*
z(I~?fFNj=n?E*fDxG38Z@pYUx3-dB2+ih#W*pZmQkR3=3*JWt69kGl=uxXeIUa@Hy
zc9nD%gJlF=*p9e%2b;x+vuMFr2Jwm*$aG#5KhxoGQ|CFy3;b8C*adn2&ie8UOlH$b
zg_eMv9K4k+Hv|)T@CuYTFJ_xA8!0A#JxEwJd8_>5#!w?YCnvoq!lQ;T#u)HAA0A80
z-VDZ}&J}olQa(IQlwSkK_HMreQ`!|PNi^W16;|Q+e=b};GLmmJ_1AZ(1ycQ4JilbV
ze{Mwu#%Q^Q_QFhW@`@F5yoHwmeV-xe^Li@Lb1MEGn=Eoa0e{#s6!wNveDP{F{4%UQ
zm7ay3{2D<2W(VlyDD<zZ&~+E3CPTV5Jjfe;l%cHaoVk&A_(-$VD8kA8Wd~or$yeC$
zgZ$Z+X|Y<yAXBaALMQ%)Triw0|AOR@iu`yb%a1cAhO!xRvOHz3?jm%~lNld}VqlOV
zaSiloJ_#@}w+t{7CCf(EnUm$EBqbKCvo#H|0KuTh6Oe(;^Mmf7aLnfXE5J6Af4v&i
zGfDon4j69Azhp>|r7<N*6frrnlIzXj9bY5tHgPs4J0P<$Vu`^vq=%(qe}h*n6=PQu
zor+yt$kns#g=~dNDeu%$=494O3<A{(A&@#bn;+`Xym)kO=qVYGBliP$wuf{Av67Fv
z-xW)QTA-I&@{w#+m!a1Q@EV7=7p>WS1ZYRxM;Z8(#C_b3M#f*!>Z!xGh;*eNS=1Kk
z=$fS#IFzaoU}A?NJ4otKK86%Bhw^c}Vh)8}b;F^wk5<Y6RhE-T3fh<0KkpOp1)l(R
zm6QNhn(o5j^sIFw)yAl=4~gpRFSH$+>aaq-k5Lq>*ghxRADf2P#-3fRxPKo+7kkwC
z<4~he9!kx3A-ve5&c}Rs%-P-wHaq?m_h)=0%p=zruK+jDqmH88?;<EZf)49L5X`2|
z(<P@1AyzOm;4}VVhf!^gF9Cj$`0{OFm34fPWq0JSEW6{iEW6`%XeRJh2o;aOZL6~T
z^TK6!-v^O}_vIl;2BS1@fnRtJb)}TvPr~M0Ta-u{D9Xv9?n8=VReV1Y`bbiIKNH%;
zF1|DM|GXOZK`4It9nUaDj!|9&pO@u#eEvAWg1_eS`xZfijMlms)v>rr{hb>1V^w~4
z0JqrXcPSK6e*Z|i^7}?@KU17RkREY^+39nVr2vB(BB6vJpqYEJ&K3DF*hhCgsY$US
z$LsvASY-ZacZNZRUdKNkc}M7q5t6GpQU#btBGs#bRo0P;1w&rQ`T(?(BcX6M5Z*y4
z2SZ0us7J>`9|b4z8X?KV<009B^iTQ_r1Kt*FCm?5JoII}V&frpm2^14Y^@8!iSu{F
zu660`4_wOW30(@uAhj<!Oi&^VBdqGhwz*>dpr3luxs>y7LSv-B^?(%1jeO7Vin);o
z!fr$_&$ti4WT0Gmh%azf$|3{qr_~?Z)@gp@Lo{Xc?!#=>b1BRvFN6P<Z)1z}MWN=G
z&=e5-h6+cPVniM$zXnOz<oQ&#?tae%Pn2djJl&ksaBnQlJi(CEMPhE`{v2SdH5h|1
zv>HsWuQ5TQ8Z1s?KL&JXNV-*w;?-Li80(^eNbK3-q$2<$DwWmQDW>xcJeh!qbcpU(
z=sEoW<wz)c6qF-LC^{Ed*ESS|C-x6DA{z}p4?ysFOjx5a_^1n;M_H#C?#FNs1zctJ
z6r6IVT#<KSVJyBqv+Ou|CqLM~xd30_|INCyqgmn)!*wkK^T(N730BqSM>Dx9tObTQ
zqlk{Ni$*2WC^(sXt^sSQ8~*az3B(|<s>`gcN*xP)0yw=%v;<C<low@i`*p93+!}Cq
zEBCx<o5bc~;pqJD0dB=OxVcT}9nji=jrg0a(YIcw!aNRby*~s0Hr9$ei*N!~My^sF
z><|92`G40_S_-1HcU&!DoptUg^r|Yw=fQz9As_d*u_#vF8})gh{8i8xC??;<bWjvu
zS?w_W6v*PHIGFh(<}UzRH!`%^#}Cje^(|!Bgj3Q!PGJ+}?Mq1fRTfZgKn>xlflc;U
zG!buKLZ#|3OQ61e>9f%LFf`#)Jv8aNxY|l<k}1$wjjz89XvH312gtFc>YJTdA(pL4
zUo2_T1==}}{{i62^Y|a)70ct<RX6f@9T_!!K<yF~$oV>wg7$TdpZf$n?i0YSk`mCZ
zy2ja;Hfj)=&P}y}@D$r2NC;SH@DabVF*nj78*_|ZX`}ZbZp%^)tQOSG3N_N*0J)$J
z;ZZiGCzfI};Upg(GtzHEC>a0JM#e|NG#HAtnb1IwR@#`1p!j5UL?ME>HP`{67+VSr
z2#tT{?jqUd8KVHdNSgjExF72@UDh@blB{juwXAL6HTp<pmGieD;Qbk3jAd<O6M$9g
zi@}i5s<v?jHs|VMtZn>FsC(U92druvuRzWaZ!}corj1bx&@Ogu1I18iTibX9B!V2H
z+z1HK&(!ewjo2gG6#Cb@weT82gLKup7(-#+r~Y3x>c=YgdK2IlJNJ@85sB*)>1rF7
zXeEU>wT)2B4Q3+OB}@ZCv*7yK0s+<BmrY1@<0GUl#hRA9-|vbA=O-aJcVUs2q1W+`
zNZyel)d&ezae5dv#6Fz>^+>RKAMnaLSh0A>%NvhEJ2|Hm$q3rjH;yKekIpeY37+Ep
zLYN@LWM`4e4y2RP$0D5vMo%K8tibUUUa^4@yGlB)VCL3^0>=e>)~hRi#rhEUumm{{
zp;1PBhtk~=$Fv1{_M*q7SiWS6-xc#EzYO~ly^PU~633A~L{nBzmN*WFrhs6)636j0
zJgUUO;pqmYMtyzgBzB)1!jRNOVr~>EaTtV=yu@(<Nw<oTmOggee;u#HA@X}x;#g<R
z8T9R<#9`pc1car9ZTx~c9tAi@;?ZBg$0mtK72sX_coZ&j9A-v08i9TZ5onT76k`NZ
zs~p$!DIQjTsIxhr47}653diG@8}-bdVxz!*5{HH4GrOG8Jry`91a>ku@y+k`!T$Ad
zvUF(_iyo|~p%ymkqji`+$JfEbGTD{*NDWREf!6gpOtk094agGf_1sRU?iW1{9E5^7
zf>&9aojPLrHOS$rKn~Fw$Ux^JNdF~r16y5A!>hLf6yF*LMOg}YGd5~d3fZ}0#TD>B
zUhiJ9BC{rwFAnqC$VFit0F@~>a&Wjf)UtNxN=-Oa21;6R`e-RP>|}~%rb4oiqUaaa
zRP1Ug!)9!{k4dN)Ge?KZkR5RER4PB7=@Hmu-#=-^B)M835YvZZiRlhL^?x5Uf>GVV
zLSN7k-cu(#+6OfByGkMG51^IrWgWT?s$}${c%_c<F9Ec#grK#Lgb3{cjy6I;?E)V}
zmYgo^o)jhi-Rv^p!eeo`;7=X=5g7gmV>r#VFC*8uF=7)gBU-FVi3`R0ZVM<%i3^+D
zZP7%OxX@X2m@HCDT+cx79yAd%e%&bd_8qM2Jk3v0?wx{-+W5F-ez{i`I`S#%y~nWL
zEP4w07_#7dbS#<!J6}Jk2L!MPT6Z$P1qO_z9T|TM4B~`1$Vr1(--DQ~Vwlkq3@n1T
z;ndig1n0nE7H|Ta;B5}JXd{dQwJU~E{6#Ygni5g>l^KMb3K*-o$0493_S|ED9F8>C
z&*|yJ3sE+e>8T`2=S1iEjZr|B&u_dEuh{$syXwaLhK`InsKVeTD3HYrl7jYy-`D#D
zT<;UWu96bat-|k1E{4QhdW2->#-40D->ZQNO~=G9`pyn@H^wF3dq8e#0%le6y*JcI
z_kU!`7vWLXrO#W4lJA{9JZ7Ks!9FLT=KE<M3)AdSteUThAg$<o7sBFIypTc!Ve4xh
z@QG3KHQ_V<QyDH&ZJxFW5R6P)ybT&zPg}^EFG7+vU%ZwzU%W>57}R{%&&N3bcDOH6
z!01%CIuVg@L)$Mz|I!b~Ec*Tdn-fLfXG7iV25VRJeFpkSQuLjANP?m-%Asx1cl`ns
z39^jxBp^mVS7QyH5KQ=M-qLm?FjIBcx*$VA-Y5T1<X1rcSmj>F19Gu*FNqc(+^jzy
ze0!S1;?LU=d+Ui7aD#g9X^G;$Tw$)4c>^N3`|4cg?QT_O1F1H#sxlY)U9nia5@WGe
z;_ZG}ChvnZ1zxf&vmKfOg7L~SH5wjOmf`T)r`~#cBg!&YFeKZXa?mJJmN7{qd0pm4
zl5o|#jAdO$#9&sJ@vK>cQCrkyOiUU7Y{olSb1c#^nDd(e<w$<B4p?BF->?*lFB5nW
z+R3s^D2McSs>$7t3r(m-N;2zLVjMY7C<&NVG!?Qr<x_Sba~<7UI**#afrPPD0pG+c
zHfm;9Ne7+GJBw8TIx?!tE2Cza4U!bJA2t8LC*X%Z0qiO%0o@ukpSQMB-tE-tIecyU
zW=9^*&QIHi`iEsigmX%*gQp2%>M``e#$In;FsWw}{=-apKtCS6t0%<#4y39kX;x$E
zzl0j;;eZ@dBRnb-4&n#uuozSS%7@3|`09xnQy;WYXO&KMtTDBTAZ<+j8>E;IlFSMc
z46QRA(26m%Hla2C!N^78&7mhiFcNy6ghtk(M-HtKk{nv&wH#XGHR_)-v|i1|ad4j)
zS`W%?%VNXq(fe8hg~hPN_!#(4*>3DjgED@t=+s+i+}Fdqr@Sp#sL%2N0egFtrT-c<
zxsQzc!ikI0^faiW74mjxg;cNeP+?BEKjz*v)!%}ha@7w%9Vq?FLy*I^7(?kV^Sffs
z<wSHY+O`#CX*&1D`qA`jWaXx3jdD`3>-TCL2T(`T?|`O&gU!+Odub@tQC%ELFn8nc
z@m9zU_<bD-NYm{Z@V1?NWdznHwA@aspcGivyS;^NvXq+>(LN6{MiZ03$KT6&$CxZP
zY;R^a7*G&#gO`Aytlc2<wBj5RXeTppbHwq&)+iEk_XE&0<Sb7_XSqq}OycBQc0i`C
zGkjmCzgWs;FADr3lEw1xFX0u-zu8sN^p&~1F09JE!e5UyCQkp7R8HT3TWZh0z}RWj
z{eK^|hP!`blfWnaQ0SLbj}$TByM9+J2AtfT81R(eYwQ^Cb7*QA1O7xqp<)1sVjBbc
z59P^^ncou@12_>eV}MEE6A%Lij4{AuDds>BkP!(4kAPYv2?QsAlsg0hqg~2qfbB!k
z;AD&j|02{haWs$}NcGbtE~aAWXmB3V#-hRbc*UXtyPD`|uyseXJd6_=V-gHDrV0j8
z)KbTSgG_x!LqQgVBtn6yMGy&c_@HtSZ;D8;&F_juf)__3!Geud6jZLX)nBp!gFA-2
z&A6cE_a8eZl%c6*Ot^-ILd66Q#Wp6qi3BuZF(C%4-bqqG&SeQGy4bJwhl5uu!@&>J
zP^jTx3|c-zLrEm|-VLxZL80WINEo&fKWZk-dEiK~0lONUM$6-15)A@5gY<Bq;$a*F
zcC>Sj5I1a_q4>{9zL0wKQ)mi?t?SY0z8;;faR2u-6iSZ-6dPLl4x{|31ud``Kdd(h
zUiHykkk<qxY|D%Zn<1Q;^R+<hBKg|oz@H@f+6s_WhkVTxJY~9u%|q$hi!oh$g%DTb
zbWL`UB3(NRNn`2S*?7g$HFh=8>Du2^hG4dKH!D4jNyfG*RmNt-E_Kp&NSkh>dE3iD
zS|V?2(<n&Yw&JthN!_;iU9r^d%x<S{CBOgJsapY>TBdHVq@hqToI|lq-L5AAsR1<B
z0*bU#x-nRF2T1`LrUVo{<eF2r_tQ|QkQ;-RPtZ^ji9H;Gze2*Wm3TK(H_o7#sX#|N
zkDI#v7s(e=kNyLiTBdHlqM=ZFB%s)&Zj%nD{HX;khSUwgt3J95@|u9uZG|y)GY2(u
z_7*5!BzxNgEJ~8S9Sh>>kiE4Dp)!HP9zqG+nV7(B7D7v$z{w6$Bygu9bu5894X;=N
z$F3$ifqOaMB|j#)+xk?wTNts_Y1_PLkI{_nQV@^G*rLw|$=YT3Sa*`Om-=0?WNlry
zleHaw|FM&`A!urutTkvTRMh5BY?HOWB>~xn?(g7A6Lr=lZL$`FRkx57kby}+u}Rk6
zO+%rAYz$gHN<&E`_HeTHITD7g#JicSaR$W<{vGW+ZnE}oBwt89`W`g3OxB*Dp-_4x
zpx7j9f25&Mdc>g^k~RE6_0e6B*90VMOO44|B$S!+v_R(~dD_Lmog{hMQP1m`r$qy&
zOwq7$C`DU`DcS}hs>CUp>>x#ob}|yiQnVN16-&|B)kLRgm%&=dVJFK)o*77!RB6j0
zFG$l9ZjA7eN}Za`(tC@hW$QsCA}!OO4N|g=_(*qBvWxt#SW0$LBqi$|s#HobYMtBt
z4rAwFS3pzC9IQ-3p`0j(V#vWbr2&_(CSlkH<4F3&!6Yhranxfv_l^>tdTy>%cZ7&-
zBG}tW#*oqY7A|5vt20-<-bX{BqOpKtL(79S6iSge6o|%xfq~}WFRFcZL1GgSju#ul
zvF5(ap*YaENGLuRIFlq4e-GMqm@?6wr3}N^FcgL_!Z3Wk5K-bVEIUXMhMz{FSQ!2l
zUa>IDt|mGRU)}(pUmuf+l8aMC;t*b`gYn!*htYUEW04<^Baa6rKNFwoPDq~Scf~^T
zOCupU<!YoO{QhG{=)<6?WrSWqL!lfihhm7(%FIQ;`4^FZ5{GF8R-#xPgH;!k6p-;$
zK+)r=xuo?98VVIpW6)Bdp(GM}c$Q>03By+6-JB)i42l`%JKA~NS&|mX7gCRIgr;EF
zx*nPHvUk!@C_NHTZ1S=X(@-cq;!q5;B>02sqq`ul2}sM1HKt|ZKxWR&0+owoW|M(A
zNiwrXpk0T|ED|<la)xa~$=ORWIqMaIN}Qa@4pJm%k0V_yIeP-HSaQa$COSFWRClUy
zSaAd1lG(ipl5<m~Wj;o!6S5gfSJ6~#$`O7lraTy=0#ot#?j&J7epf6B+Y(8_7WqrU
z<v9dv<$2z5mWD6%;lxhH4uYnZ$=I<p6v~@&D28NgeuQpi$?Ee-fVLqzGIwHtnJ77n
z!MbxvO30uppy)x>oSbc@p-@3J1}#H0ltgL}PR^<%3|q-}Gdbf7iW%%X+JD^S>`f$J
zNIiN3G__35?x3MidL*FOBxmoZp-_6np%{`g{6Y2IU69uVBxgq%ld~YGnbWdB*&=D#
zpCI{2l9qh~*x4a1Q$nar$gou?A=`op*<aEoWU_-43E5AOB9@T-6t7r9#;zthA-lZh
zj8t|z*;20F$bzW5n3koREeqk5Iysvg=`fn2{TXB<s#lT6gGB8)e5yN%+OvLFEK%DQ
zNz~#Nu;v{};pS2<bQUzV%*c+Sp-?WALosAz;R4o55{7LkjwE3z3RoQVn6AB}gku-5
zE+iR4#^Q6ii1nP#JW+Bv4TVx90mUX3@1UVjio~Iq3Rw7yYM)(@*aQS)^Gr$DiJ3!k
zpmdRt{5$YCNkZ})fT<lqa>QlIz>IA}fq5GS=HCnNmw2W`c90@4--C3q!2C|UVu6`m
zO?Y6gSDoQvu9SsWVAZ;NzE--x9KtJoV2*Sc4a}be*~q{gemn@wpTeiQ6PO?JyJCTP
zI1-q<G=uU_KAhMyDE|OWE#vj~XegAM<xmXq+Ay0DNaFvI022pq1I$D-C^1;~Cz28}
z+6pLov^CG5^dCiOE}6E)pye<cN+LA~&!8Mf!myQmH)l{dgJLED9qm8v49YnqUr0TA
z2{g5wLD@t@p|Ums#U^XpPD7#eh(j^Vpx_Ux@9u)U4w$4Z9I?zvTcC20r0t{Ngpwp}
zuLGubO4`E6$)pY2hLW~nOxiv!9Ae_6O?HqXX}cZiVoBQ_c*T-7c69(HZIx;j-ecOm
zB3HMPHilO>lQz*|G->-7$R?V!@yCNy;Q@TAJ4xF|{H|EiRvfFO?JGW<*h$+Lps8ij
z_6QAy^0FLC+@$SCBtYBH9hq_IR?-%Ob-yDiA%m@eVw1G>9!+U3nY6{A<sceLA~gsn
zZ5a}Vt>n9zv~dQ-OaeOEf83<)43aOT9<70<mPy;iG!!an6HsiDwpY+lC_Um(3`raQ
zp!)7E$m;-1+IYk=CvAbsMUu7;fD=lRw7m+LYM-=m<Ydx@Z9_?05tFtL3Wu0DX_FnK
zNZM{jx>(Y73tq9Lja^N2()Kd>((!CLH{x_LWxFudx+D#&)Ct?{P<PRE?S7DoNY_FS
z2g%xp@R{x;YxntGv1F|hN!DijYn1AZTAWo$pYwZ;ou2(YG__36zClBw{3(ZGNYADK
zrOG;_ACMqy!*V1cV^9*!o5f(yFG<|V&?%tUq+oxcp-`bS1}(EPl-66&Vi0vW1$!O|
z!&cVaOu;yVVutjNRvkA5dlAVOlJ_rwrj{w#xil0?j|3E(6l^mMg>o4jiXjEVA5@Rr
z1$j+C3bw?!@<<O^<^(KIxJUx_ci<e7Bw%}ir5zHmP=J)_7q$$gUlmNh-Y48t;`B>)
zkRtthGg8IUuN(1-rC;o7qSLR-ve{|{79SPM#YQ$eusme~HsUl2l{~y8EnW7dA(cA&
znyYmOul~bCYrs|CBIM21o{?wihVBNb*u5Y{)$miCApRb|E0&5iF%{EZIyzZk-1hy0
z`hmsw!%@kg^`>qQ7~i5Z+w5Bc0wTWUZeUpw-|}^|y+`3tqrzvsHZQFSNnH3A&%7Q)
zBP04s$tHTAcM9<)_CB%$();vr!!y0o<|jxrb3H%BE9QFG)kM3VtxkEkl6SJh%8MDh
z=YZ0^(Z+yUYB%(Vq0@mzWtUSfzS60!$uw%Yavk=_kHA|h2QR5Pn``h+EGNHNKCu$s
z-(0Wfb~^Pn8GNVd)^m!(jjb?fS((8x+BWzzT8F5;CZmnrGN)(oNcNQ(w$DrXk@dk^
z`j4zwv1<<;uwGxDS#?I{Qs>GhpzLhj+^m+It?cn)xdCwe%WZOGsQJ<b212DrHU1X(
z)#rq@NM%pqSD*E}Vt#dR#IN3RVDer>r-$ZDnj~Kj=p7#eRncC1E%Xe(y_AD;C*b}e
zi>avQQZI*|4^1QKzWG&m$IwuybYDQRN%v2rp-}0*fTD4*qjhq|zFmsUe?4C3h5`TA
zk~rHALnBiTol!_MA8{rVxo)8rI$}FU5-$*+&7ioNIUGehLX!ndmrbMr1UCZy?<Dz8
z(l*S|)Td3%`ro9XQ2H;R*y#UjX(*Kb3n-fYvlo7cBMxV@X?_JGgff?QGl@)M#qMQb
z5-IOnO!C?)m7`~?g8aQC@?@ovLoqQ3W48k1AreM%1~D*W&7iN5$R}e^XfpP@B#6ZP
zNj@2C)Ss2U{;X7F`y`3G1$(Ss?fNVYg|hdyS{f3<j3rchaG(f*r=&1h*RxF#!jUBI
zHbT%{yBW&yX`$%B(IVpL4OAKEvjRjkJzNeOU@VZinhh4jfPnjVR|^Dv{AsD_@@o)8
zFQ9oqC6Vi(sbvzmg@!@}Q31s!h&nVB%BTbsjd%S4hdFBvks)Mrdq@madB=L|Msmkj
z=Q6V4-VG$~Wc5Hm(QVGWdiypS3Z)4#Xn8LUh0+8L#USc%73u*JhOMl-S%u=L>pq5b
zN-=Efn<TZQb_gh+R)@$vr%RUvSAjSBAr0kW=pv_b77GY}4>w`vE(R@+(@-ew;7|%P
zyuM<&1oJ(un_6-VDc*@DmIZ{h<yjJ@H#_g_$Tdri?CxBt>9jN=y#@7VwOXmcp5qbN
z-gITN(s1gn{Yd*CuvbaDE_jtq*1_c{G3&&pjBv84bA*}7nn-DiZ-GP)IflwYNbQ^t
zO)cZgF*Fp)s|YAI9`ZyQ3Z*v!iWXnyg}K}|%J7j>z|^%QBC6zLg_#rsv@=bzG7r$U
zkjRs{9fxA@&7m;tkRTHCCwUk)vd5g1j*_@ru*Yir@p>8xrOHNH%)$O<8VY5W0*Z}U
zzKe#^nHGaWh28f@NEo&n)y-hf8KnDs8+U5M(XWxzlEGd$I-6ksT^b4%>;)7XTK=1c
zLTQJ9@<#}=c0vBPG^{@OIc8+%DCkmFwaq%RPnX3a9%NiiK$BOjkO3SogFrLqSTcH3
zBI<{x5WTgC*T?Hs5|;cLz&nD5M~y}~yhNO`Y2^MK(oB|fiHg0cU^Q=+qA5#bROAdE
zXu52!h+_@O9#t=5#r*(OqPfQxa!Ko>KXdHgKoUR(Hx9+XsHv>rsc`8E5=de$2{>cT
zqAH1fG8TpN$5)Xs5_2d;{uo1t-bx~$j6qYNQm$yw-6V*_{9#bKr$6TW@#7@!7VHU~
z4Q!qz|3YdgHnIDkX(&|eZtE_>w(xHxjKs1YA6xhpiM)ld4gNG_3zL>oDL`WWjEgNC
zOyX|Ao{%k=EpExUKsg~T6q_su=al9`5(9Heru=3CRwND>S38@QESc9O24qCmCH@YU
z93@$ocnL_N!@5M{x+%Fj(fq1f14+>}ihFU5;_t;$DafwB+Z!aDd=1M$A()W7cfDAF
zmMJ(jCsQv}nx%YZ$jR(%!nxArhU4U6nHk<FmNV5{u~uK1*;8y3GL3?h*#Tc0at#|i
zQ!MXf_nURtsxOvjWO8Nb0UoY9uq-g|-wvJ@FZdGA5B8svYveM{-rNYRQCri$efxH|
zUMcrqC1$=BdUqe_C@Wb)He``7D?GDWs|=N#kzAuVtWFJ2-o9N9dhjwB@DxLR1nYCW
zDaHKrVIue#a5_ujzh&^>@@*||YR%bE%(w986I*DnJF$yJE@A^3QGh@GV714G3ReNp
zSCT+qKm$D~3KXiheL{`_kT7@JyB{E;a8uz(w(jpLRPZYVe$63d=}@UMyerFQy;YVz
zwnSZtnl0v+tjRooW&aYcaqtDNisc=Z6PLhUHc1UPx3a73iP^kUA6~Kzei5o!#~=GM
z@b4=2iC0T%m6C%!q0U1ucn|*}<c3pj;7^6(5_W$d|Jl$uSFgkJ8ZfsuJ)G~9N|nW;
zmwp~%R4cx83<#CVxa&vP+!12qkFL3sF%o{^tYkO%H^xr*zOo;_GnTR+w~B9qt#A(*
z3-=^z`9q2V-riOL!|JZ;#;Z{`ZiS|x=C{BfU(Iuz>n3r9MFnCCJscK~+CqwEANNnL
zH#LCD-L3l;<(9Yjs{Sw%YImS3qN7{YRz2MP;Tnjm6efKfNNm4pLG=mf1#W>Jg$=zH
z`Sf9E4Sf2F0!S=!4C2GlHpl{6t3j4%3QUf0jphK?Ovb_#<WGenTq#6+ISvuN7kC{~
z@C8Oe;1CU};eHv&6I8P{oNCh5xHkrE7^m*HfOA8r&AqIR5^8%<v7}Y;FM%uk2|R5r
zNuLSVArm+tUEK$Kv8lSNDw@z+j!z)(m<<<tFA53Xhu22>9Imbb&m<!LdL;hWjLWnJ
z+>G7x*umOFB-6u<7p-fbqL?#}OCpmu1TvwOOhJn&8FyQX!oYGTQ7P6IJ{_G1ts4+6
z4#FBO=1rAIJ5#z-nMcl(H{98X6l}r@^YDeSkev%Sp5(V+(f*CN%Kt;t0;lRNTD&*|
zIn*#Du8^D#jb_Ttk?Lr^GMrhwSTc-!hFcwgS_`1+L+zDEpE>^^AQ8pyF2FBfD5hK|
z>&X%7R1VWJj8py4fn=yq$v5_R{gAIg_6<8T@~!LD?o9sdUm>4|USx(Z_f-TqV50$V
za&@%j5u1pHfp)%$D34cc6A`<b=uJc$vwWD-y^TZXq&hc3pp$yD&=(|K{?3i@-yp=6
zYslS0vf7hwYtS^GKivBL??G(a=UcSt0Eq$j0nljoLHKagj72xTd%4z+`du;S@Xp8<
zpc@Y~?lO-wpzn4Vr_Db%`OGr$oh$^u8JdFM)$29p%@6m`P^bdFfMPSv`ym<%HO(ua
zXjR6}8)#TCOSk#w^CZryuE#3Hb;dnW{ovC~<a#i)&=EcEs6^=oES4+Zz@X^nFjkv?
zenRq}r0qXBnl}3XYZ?lr{{o7Q{`V}Sa#3=*g@B^zf4a>-3rJ*C9f+maDe56}0i%e&
zGS`=vk;s!1{v3*-9?~1y{Btr1BRPYTZ~lqFpbJRklQAf?_-+dcA~AoGFTOLf$6VIj
zMdEJ39;?N7qcjvM-nG@zkPu!^f;dowU>*bCPU3DO1l_fpq1>AmiXI#-B97ibm4S6M
z?egZIuhKlAmRo%hnns+mIf;CfhC&5V0mUYW{+NbB8I^#d1<?eXe|}41pvpT|<V^1P
zp5`*L;ogja#3_h?qT8H#J@BD46iO3f(6W?<LTLhrVoCz=P!47K@FWt3t*pDb9+;!9
z`xw$G#jvgQB(-FMDxhe~P1|h#*-As97MsSPrAR}ev_n9#Uk|*8hLvDFFh@a`vZ`&?
zk>oJ3(<j*DNbDl*u*c^{noKH;c+k``jJ$(}LOBls#l{hSfQCY;hk%kcjC_W~Ko#~l
zg^~Y{%gBa%-zIS<9fg2m6Gr}nhC*pV3|gL|p-`H@p;&~GKantOW!=p%!co^93>ijZ
z*w(`3iIWTg#U_j#OGBZ;NDNw@PeY-!LqM?)BQK$0B?u!N1zpOjwpoXfWZOz!P7_3l
z=rU*u0Yh`LavzDPQOd7Dkzywej|xy6ULsD}G>U91F%?0^dRxi0Bzsi7h!t^CZYy~W
zm$V)b%+r3ikpz%41{{jPRY$g!yq5%$m`iE4mBe7tr%3FRu_(Nh{$UbEVh*L4I*!4h
zACkx?V^C-*{o^Ev#QaIVl-|f5^K{vBB<>dM3Dq0T^RKg4kQoA%*$G-~veZR16e@Oe
zC<b>CwuR*+jKs1YA6qz`MBYN!1_p&};UW@5V*ZSaEo4dDE!Y#X1+&GCj0=>j(n7Jx
za&UpF*j8f7Zzf<{$q{RnV~jUDupJ@Sa0caa7xPw<0QJaLk_Oa?lWZlq9k}0ND~Z46
zMs6QDib6fQW#pZ>HgLD7V?*W~-7?blq;eFSSdk_>NV+2JV@N1F*81ak#a5)TtBGEb
zcDaywR=z-?drQ;ScVlUqk5}q7YI`D-N|MC^)zS>sRz|8Y+s)37j<w=STXe<_x4{Fb
z#Ud%~%;|YyT;$r$=nugt?(3whinY}2VZST3)a*TxrDm6p6>m&S#uj%3CR%Bu-s!TH
z<Xvzh;MtFTLSruz`*&z+IW_w<4TVbE1QeUJ?Kv6>m9_~e8vhPu8X8r4&TB&k6Hf8w
zA5SH#iBp&qDc)?rC?YcRka;PIJek9AC<X=@-d-AL;z=aH#N0B#OteDI$U1Xkb{<KI
z1?%*biJV>KimNwJd1aia`#!f^?GWy9T9)e2yrE{=a?muQyyg`Jbs7rgZv+$@fAbm|
z3gvGE6wTjsUfS7(WvMrjIIEfyYbviZ?um@<%}nGmmZfsOcEKEy&09Z2a@VEitv|*{
zjX7`qWf~r3O&ne#PT4ey%v+l>DsA3+MkpaP#E)=*{V_>?Vr5PdVE>&<(PmEi84`Ii
zN^mF!28HHOrmh^51SI(!ijh6$AbL27y9Ik}v{adQFzM$6%sVWl2JgXG&pZJU&=8q`
z_!5NtBoh!X1~GJ)fQSys<V?geD`KKk67Rt&iLZ!k-Z&+p?tfpgj=gv)BZoVgVL3jn
z!?_i3egzw&W{Uaxpd1OsX%b|p=aVX_x*Oc>Cr>PR9}=@pGzDTxQ-ke{Vk@P4rQzOg
z<(}7n2^`1)F9m~lNl6;iTLF$&#KF;R0t~?J<=BYrk-Z2rtm{<B_pDeU8M|UdrtM4s
zJLBM7c-vL3(SVJavWX&#!Tx^eO1}My^<{V~7VBrTd<st=8D<l44KBrqMEm>u8AA+P
z?+5g~_vcy?(t!v*`g64n`m@xo#RGS$OpHDNTV+kd@zGe~yMv3b360<=jD>~1Ko9Vq
zx~06mH$)$J3wFdeN#Bp^dL`@8eNd%=91a6HZGR7tx-o>5p4f#9$bEoQhqxeapqS)g
z1xL|>lV;QDu>jk9;$Z8W`v;MK?_m6+$y1IdU~M?hX*nrJ6EHcI85~{DH`!QO0t*|^
zb$+%6#-#LyCqPY%Zz}s@99FwR?LW(i*VP{HB~tqjqu4&h*{P}h0)vy1x@7z<P@wOH
zkfHTKs5|0dqwdGY5TV33R-1s1$K%l98^TABqyLX_lqOs;1lER2!P*d*<hUWwZIm-Q
znDMw4xLwPA5C+IjTl1@d54{gbf0^JK_^poNecC3c(X5sI*zSK1gz&sox@OZn6@mRw
zY}6*OAMcO1Sj^=~6rT45Z?I&^5`<Ryk13EL#Oj`BS<qwW-a_sqH2OnF;rFI4tB=J6
zC9FM!B3TWsJB5*hqNq#rG5r(>=$tsH`Bw4?#Oo}E7cEi6N?03C3$5`=xM1|Jw1A^n
z32QT8(MDJaYFi8|c|f-knv`AK5M>@M%Zy`nILK}wCHCPU0dlNp^kwaE5aKisInt3t
z>73|%n#UUfSzbGR6JD{}DZ85J+UYCUyU4|HEm?MeVE3x08&g$JqnM?xpRSDdIpDCN
zX`a)L@9Y(?2R6R6{GezDzUwjk6u6%Eg4orpNB1PL3ZL)tyJ9;050S#>VeLp_sD{qE
zn5{j`qlh-=ZFm4&f9P&N=%;<`Vpl6a2~C^U$Nh){3m#yN^aBf&1QzGx^%>jHkL~i^
zN0Za>t>kp*enlN_1y!QRysb$gCSYF7Je6dy3iHfZKt*I`>>a?+Br{{rLc0z#V<v-^
zb7RYmb7N70qO)WFfN}F~p+S(&BwMf}J4hOI77WUub1+`9pu?^vI_PYxIU|+bPPUY*
z!wa#TDomU;Dz)y#olU9YjuE-kq30fBw|ewhWA_GlB}jEXqAK0#QHL0@mLadruR$<a
z1$x(|sJGnjibcJzMWS9#3uI%#N>liJK@zB<qcmZ5t<R3wVfIXD3O-3MfSHT*FQuVS
zMS20{5EvB5LkPW|jMN;s3bZ6kL!pZF0*c1TF?eaql_Z90n2uGFPr18h#3V@bQlV=}
z0?1MwhhnH?Mt0ZSL;^|7r8K*1VzB7FB=*Tz6t0FpK*C7Op%m5d7z}!tL_QgVLe=oc
zNDztnle`*kWRJNT{y2%d1$#ms*}S{v52>Npc;soTsklw<0dMOr!nSY-2_vzr$Hx|y
zlgL{L+u%<_wr~mwA~Ao)#TG6iakpSk$QH~N_wsRpvNJ6deN0a-73JxS-awUfFbRMq
zVKD=8D!2s5=O$vQ=(RL&sKsrsfu@#O=uI>fDhm}*Y<%85G!!Ze6;QM+wDWSyE-V%O
z7>Tp0IkEE6&bTL9D*7QNa((8LGzZStMCM@AC^C^|8e?h;AT5U$Hy<{sXA=HH5f2z#
zeUSJEqo5y-jVuFDiDdR&CbN(`Hjgo$CJ7*&K8IppRAh|t90??`#-|x$#9+~a6Ue-Z
z@?>pU6wa#_lQ0r<C`DcsgFz>g$S2dGP+ql`1d*6O$@3~Bd(0Wn77}+0_Jj=CJjQTR
zL$Qgsbs7p4Z>??NH6)D0vK}8>xS2%WLf8g>8nT6VksuQDXIyOIArf~B_JnM~Y;j*6
z7bxFJ3&kegQC8BM5}gUC7@FUrXf8RJi-rLqk)q+TFf2+^H2gKlqC?TJP2EG*4ObfL
zhDNfYrNggb>F_u)%agctC_6}6I-Gf;EFI3mD^@yWR})=2+?vbhv-PSoT+EfS@JhZK
ztheu8;qa1Fg+l{ssmq3U89F`Ci-rpfILpG1@wZ?Qv=p?hO9jKlepjqu_>KSN1;bN)
zcEm0io&rrRr)n;sp-=-H0mUYD+(JX4Qbz$ro2s!K)yEmk7fB2gCsrv3^NvZ7=E3|v
zk^nNh;!q69XJjyc0|_KCm(mR8W3cEh68mH<3J>P*Ct)P!P>KOe3<iCkL_QgVLIap@
zk{}ZEC;0%z$R6|1;>RTJ7VHTPV9bO0r&B|*@yO57P$-Yw)?I{cVfOQ>+#s>6$Hx{H
zlgL{L+u%<_wy>H6k(fW@Vhd|Y+%4D>vIVony>wilWYa>?2lHfwgr_rl169(-g5gPE
ztP@l)+(+|<Dj1GJQ_C#$1{w;Lg$gJ(KJPXf3gz<z6fFzwTy*O~!SFs3XH|0!go5FF
znaE=l3^`vDnS)KENWsuF#!ON$j5D-&bZkVJa%k}_CaRDZHV-X+MiM~!dJe@93L`^{
zXGkE46+O++A_j|Qo<yZO$ygLlrw%7!B<4_xbSegeR*}djV^AoaI+X;Gm_NzWDI<H#
ziOxn6cMJA}4B0%i*p?cKO_<$9L!rWKTXzw*g;5elVp)%mExeIL-a^<0e;Trd+er|K
z`7<uI@F5a+3-*L;!EAA#9v3KIOAEy&%Teltru=3C>VyNvf|{wyV6GAdWJIcjN5EJp
zNtN(%kVJ<np|R8<YlP;~r-7tsh433#Av{V<xFoI+$_|oN2&cS2RtS6Xid6{N)d5x^
ztW>K9YK4%Y*6j)*r#H#2v~k>47%~mP2v{r$yRnVG1r@?0LEE}iAw1mgid6`|^}oDA
zc#_YK*cHOn(A08n<}4ZtH8&%m*d&b`X(&|ED4=L_Gh<L8%#j!-PN`B3<F}avX&%Nm
zNdm~cibFA^osnVuYe^u9xs+xYAA?1=lGrC>QF!R`9uh`k4y72n#9+`TN#v6;C^U5W
zDhVPnf07SfjO;NFD!xzRZo!_=(8WBA|5<7%HXivI8Vco++q#ReElfR`$_*0BdVFl*
za1wb7VH^Bu$QG87AQJOuTx{V~5_b#sglxfVaTkmWlr3qY=)-ujLdeq@y@4ueV}<Z|
zFxCmG5H@MvP!+-}p{ZpS`dS(am4yl@Ha_nyG!)9`2`E|?IxZE$cab=&nsXpj2=8Jd
zk5M7yd`)ByHjN?`Lem&?tSf|H9~%*-99lfgL>2PF=Ap%Zk_3>xo<lK&!pP9#7bK9x
zik@a@5raijUPz@n$ygLlr{<F|5_2d;Iu(OKOG)IDF({NyokW62%%9}xl#xB=MCUvb
zcMJA}4B0%ixGXgkn=m^}L!rWKTXzw*g)2!IiDf-Lws0+pyoIn0{xoC@H<2I`^JiRa
z;k_j87VHVxg4yC87#ApCObf*(%TX$Xru^mrtq}4mgSkQ&kP)d69t>llBo)FRgCycr
z2swgrDule)v%G}@c}p*31pEad-@*#vLNVczxI!p9NLnF$78$_a7yKu@ViiJmHPIEq
zE%4^rdb8$ai}`G+SZ{Q1|KP=`_78^8N?jRzQK-X#%8<Jzb6L4qDQ|_hF0IV?$M<hz
zr-b@n#rc%l>wf2kI+GjvBKLz*-65w?Iclskujl(+u~OY5ky73B+SA4`)sIIym^}?V
z$Xp#TK~Sv?xu?K`l2UdZP<OnKQtWcsGH7Z!QSo9L3N=w7px9(k=h0B83`#)JxHi|w
z+0IGC%SkjQAc?q)3(AItBP8zRk$D1&ejw}BQtf)KR*p)&iiSd|LkwE>(@-dN;7|;r
z4ky&Nk}wj>Iz>Wl6oYw!;yx02GGlDZpin~nNfJb2{v=PRjqEX}x?d-8w_uN+gp%{y
z-2dneRPGq(w~qk??cOvHnb0#uB5guXc_~i2W&L4Wc$(ywN=>Xe*A%w!Bx6u$Tw}I{
zsV}0UJSl7r#o#PLwy=N%k(fWpZNbPMvn?zmakpTPt!mGm1#(#An)Gu524u$?mu!Y}
zNb_JUAR#gsn+1BAWH5FfNTI`EERx8|f!J|2q(ql#K7xa=Il{7ygD~|CsTZ!#?Q$|W
zD#<hoP6pCWXSk6m)-$;b4pB0d9r&VDXD7T}bTD&S-PzGBWp-3*nL@6dFBQu>m6w9a
z#G#vA25IlhNPSjq5eB9HJG34{W?U+o0y91|O*s~R(c;A!@dhGzFX_-|rraE<j^-=F
znZ=7GG2}BGJAjY}2(jM@CD9Ut=bypYqP@-GNc&%Lq#ejM>G=36)JZlzz8bIC_?TUF
zV|;8y!7mQcUANvyywz&zFubOt-f~pYx{m`3-2FZQZ}16VS4jydu_+90^o2qB+5r9r
z-onxQ;3&<-;AD@1|CYjk%izD|+gjc<r`)I&oq7w-+H|jnF1!sUi-mWOZmv{HnE^%*
zSX0Jwu)e(It+jk~kN_MS&NEt%C`^&s<Gv9v*GV35->Tg67LlFfNu%ye^=%8!-+J0H
z4QFp-O=cM*dD$5&GpFU9`f#>h+zDqW<%XR#8UK;YvNO&AhOz@o7kWbIf!V(cB&;sG
zb#2B(_k|jT>a^520^w11sHb)P{MB=J`S92*!4o)3@HUl&1GaLMZpXesz^c#sSePaT
z2HbZ8M(%syLpP~=;mBC`{qTV<=2rL_T#Shzbymi`2VwDb)1-O?*nPbNb`KzS57OA(
z+a9~|Pc;}1%!x$6zDOeRc3^{bB2nmbA3?Stw?!7guCHskkIDO?DcC14;N2m3-7)$v
zJv&!->T5D<s-tUgYHGX5)VKTAIZZsSObk375^IoDdq-LHTDV)f9`{-7iL2DXCNvbF
ze+~iF?R%hbpEtQW+Vc9UqpNTpP1ASG4(<XT{Q>%*hGL3jwZYG{7lWgpR&#bVMypQC
zW0Q02gkxuKwNxA~Hd_1H&wFyUa%+F#GlkC<KF7`F^PZV5#OOAW1U!Y}L!MGT2ph;(
z$g>7d34XwvzEec6A-;JdJIvmyUe7gPzBcU?wG4jne)>Ws4-svPttT|8faq<|fM_Vn
zNsib@xkn704)`x+?EJo~m6;KzUeE1Bb7B+P=M;w<TkDP5$_#vL+XjC|>kVgQO=e>i
zIh!ltkJ~b*XEs#I4(@@el%T2qBkO~;^dDKVVi#!D&U(66Yb`JkD#MQPx4?(L0CZms
z#v!|NWBBkB{H|EYd^$3*e#?Q$d+rPMS^eW<peovH)7F-6hP=>cnJK=LlU3(JQwRq7
zOt1OXTbI*NsF_{?#U?}BK|`T3Gyz59;JB<%+)LuD4Gm^Me8^Wua>a7Xoj+X-818o<
zaI=5{XNc}}8ZECg?$!tBW1hgH$s)JGM6O$?g^uV$F_kDi^A+y^Ih#Sz&EY885t=Mu
zx@;mHQE(&hK6jA(Cuw^dN7F|C-%mrK^j|=+(f?1-P$>NuP&EBN*c5R%ql1J@jAs{m
z;c4K1B#}{dK+A+WDR#_hU{!g)Vv?768u;Hx<jHwX4#mVEjNMAU`2-0gIfEFOMAjXH
zL6g={K|2|Pra)q%@MjJQA~Am$l<u*|yi{-riMs`RtY#!mprKIq-d0OPLO6p2ai9o+
zPxZid>z-|j0&OC3w-JKw+Rae5r-h;iM~jG~H&A6@9Ze6H!v-)pST%x!1u-Bn1=2qL
zv<##2YY;?VMe~5#=JHBtYMDfOG!!a`3Me)~^zAeh%BTbsEr|999BvyrLpJvT5(8D<
zu_9-3$M-%iBOC60j>Mf@I3l3vHfLTq@(mgar3o=;`2h`u(gY60lmwuuDL(4wBn(?w
zA2l<S{`WF4iJ}}wUH37hQ;K0*y{D2{1I1AR<<oP5x`Nkpx^zi!6?mhAXebXu7de%)
zFt~C59&W-USqxe-G!zS3{<uV;g-c9KRq`(ZMT0b~zGFfZc?;^zYPC{>_tB5QOC_#s
zRvJ)I+K+_(0f731>QYvDVjW425>rh6%u?rYGL<!vq7&Z&V$Y+ArGm*?XlfZuE~BAP
zE<`}Fafib+6iP`16fKy{3v;<`NC}U|nj|8s<YUE@6r-^#O|mkN#`crQlZhOMVsONv
zn0qS;A~AoG$6O<O%o*r?B<>dMu^McAl7>R5vXK^ZWdAA+g)&P4#l|cjrJ;1D#h_4O
zpZ#MJhOI_*GqQ6A={}!~>@ghOpGj)TyiY)}iR?2^Bl8C;vd5t1a2g7w9RiAdWM4+Z
zN)XvO3c8e4ZL^N-(`99d2N_or(Bu^>WB|v@AbP)qCWsQzY0wm+w-)jGcyUU?l3xRO
z8)$e`w!q;f;*?FJ$YNVlbxK=oI|c`Hl}PredJ!w`C)ho&&cqhDr1hbmIrd*e5<mtw
z4#nWABfIC{L;^|7rMSE26SC-?B=*Tz6wV(%Ou|Uap%nRJ3?2FsiF`5!g%)*xn*@=V
zKk*iIchnzq{`gZ8cMJA}O2FnR@!zC|Vw2fTKAp@EsMy`sU4(669tk6{tjEU|jwO+|
z5VpaehHT;aB#6ZP85dhPm&DzIJt12#TioVxfijd9icOY-b4p^fhAF=p{}q(p>bQ%%
zltp<+&8PT!(14oAa?9hPa%;WZlCQI5D=z;Q1S0oskPBQr<B}=4h2zAS1V(prJdFzi
zSBd&Fq}<;HlI4;~UeH%}bKJXLtU=2Z9N5uOFI1W(*xBG@b~bah9PHzOeGZKZ-hgH3
z)m*VwUzyocY!ukuj~(!}Ay=`%Gi-MU><MYsotl3ewtcXN>kcdv%=@>4r+EQN{xH~o
zPOg#5ID2y=)snNOfBW|BZoN|Oze-GpE%fd_&{0-KOm7L9saseYNW+`Fov)OI;wD}O
z1E1SM&4NWa-jrhglonG^&Dl<c^+UxxtS&9)U&szw*a?6@5fA}T0u`)#HU7lc&Gx8J
z-%mn)F%9)aQK&#xpO9TZT9_m4{Serzzi9P0sMK!~_yIk)sBC;}324@C*l)xZpu$-R
zOV(tbzp{S`S1kB~SH<#<%85(hE}Ni+n_Jmc_QWi1G+43?evwOx@W=iP{JV;M;?<H`
zrQ~2wsM^pA-ot+gxdFR6@TWp?3A?|K|7>UsyHj9?6j)W89?o}4rOIN_OFs`Us<B==
z287Dd_pTpZb4Q4cKf2~l#z^>uvy$E5-xxdL`^tX!&REKR+$z2aw!%GNEZmc<<qs*E
z{ts<+F1&k2RfpfBI(&(%wx$X>&UKTx!lD8(g&q!zM{OZRvyZzW*P9yHG54zOTa;Vg
z;;Z_@NT}U`u859qRa&_Zf?7+<yh`EHf5zdGdkb_aR)_xy8TCWXsIMrH#LBLg3=H1U
zHrRqeuLfbFEiggCHQEDQGa(CGkUtfQaHSCObQ~gl%l|V{@D!sU@PY>2aK9W-&DwCP
zN%!L37_?!W`kn=x8$xaVg|#UJtn5Y2l4ixf1nzJC86X>FGZjgb&xGfY2^^5_`UZTl
zsmiNrnowMhPk?01h6}wHg#_=zYa@LQSJ!}J5)t1s8Hhg`no^rDFuA*Vsjv5}O;|GS
z?BfFS+NWsdDT(l!f_QNlw3aD&F)ibcE6^Bsv4WSn$mzmFc=aM)9ELSs%%KMHwJ=jU
z2AMI=lQ-HjaeFLe=K_w~{1$9Qf8S&X>|?aub_77h-fkPh?=ZkGV5rwy%%6CIo)fuS
zK#)1y>WRUsA4rre74Eoiviov)<a)IulP?qdc4ZbX_oM{Sv60yI+Q)aw=zEs<w$~eg
zYQAsR!z;FLmt7^@_R2QN9LdK_X*LeM47S$~SMpBRHw#_Vjm<&=xzrnkz96ZLvD>5D
zn}TpZP?xs@P4oG~#}NMaAj-YRw`|kKlUV0z-Q#z~oWvq@5+8&mQPLlIHy&u*6^Y$L
z>T0HO+VJx!pIIipll9<-pee9cFWZ<m{5(uUq3ZYoip^~AV>A?MwpT#WijAE&&9KD#
z?NYBqhP=>Xw@sk;KO=G02J8pGhM%7>k?R?#g^uV^NF~Zxe*PGPGCmuAW}iuA*CcH-
zpsA()FQTDP`Y)i^=>KvW3Z?%7il+bRHvF7QBBSa+tWtA|vdAeWd6~=18%gBJd4CSY
zP!{P8ZTQ(n!br}b<QslsFz8AW`D6?Vt-*UW2_i9nlCQxtvd3KAyoto!f<0Di@b00Z
zQ1;$dOG85V5DDTy5rVm__C*qR8zJbajTy?f(n8ULqeaBg8>lj{j;3AS@be3r2h^&o
z$DyfZ5dAX^g$kkqicJum`4TcApo~gD(Sm4#4L?Vc7^w1&6*-eTzQX|_kdY1dP9$+B
z7k&yTy3LuF1)oVnp)?@|E$eA0lqPT}rX(P=;b$ue!&cVaTo%kx*L@7>lw#Oclcbi^
z4gp15b=qdb&ueKY)SA;6w7i9eLTQJ9V!tf-E*e&XWx*T;UCOGqSx1t?#D<?>Un8+|
zw8K`PPtasiVdUe`)G~~Gg@!^o4*|u-5k5jgq0~b_NgGCfN@Ac2dz`|^j{qUdF!Dzd
zchXS^C^lha`dMVCr!*l3EemNVlqPT}7GdOA5{9j;yBS6}>bip=!$=I<I)kK^)D8i~
zCX8H6L!rV*3|d}6L!q=oK(P-a1sYa@Fv3yLrL1b3br?yuyW}-AL6nI0K~r!>nv<2=
zNyK^z`8AlVeKQS@3Q!zgB2L*fitH{i6+y;&cgg)EdsMxM6>(DTF1eRWS`P^3X}`~t
z1duZZ9E!nJM|PKdlLV5OOKEnO#9+~nN$iucD7>QnX%a?a4yBknj=`X*XD3c7+Uihf
zMg0O2L}LCVUr}#lk9oRm8Hu}v{)Fm{=K0r?Q$w-I@z>H&sMy_B)L~nADG4L7tjEU|
z91?j8VH^Bu$QJ4(h{XIE7h8BeiMs`RLbhPGxSPiX%Dd7+vB`39$)4C<V#;qOV0Xze
zYn656n;qDIkZU;d2w}PE#k`{=pdhlN<PIo+C)rW*MG%7BjuL+bj@(7E)Pj)c&XGm9
zaPS>s2?M0l(VZi0Uu}T06pPkm2T2#LeILnYFPr!QUa>`M>}sMHt!0G<WZ_I^rWsvd
zy|%fFtJi!iQ!iuN6QNr&K%84GEoE(GqzWfsl^bw$b*x2M@-TwV?{=IQ#zn5Pj{Xn~
z^nOcvvskOqp7y(9tI>{%tVX+htaxKuRkpYzFwsgI7*Ch2B=3S70ncWxB@<t2a6Ao~
zTF%}cPD7#Q>;x2>%x)PCh05#%6pep}vV)qUO3!(1NOHni;EPFo5@$InvcQu~A~TPw
z*OSPTX%2^CV9@*sGnK{pTS<V4xn+Qv=tw;y>&&@Zg`~uSb$V7sPQY@-)f=e1GS1~q
z;c74uD_7r0^M>*_`#IM7l#qGl>g_ZX%HIemHvZ;b8Vcoa1QgBRjLXW^PmwsQniFfb
zuQTq6jP4;O@)#>uIbXYA4#}pr|DEKnOHFNmpOG4KYWqnV9%W4&ULsD}G>S}Zn=&eG
zYCFMON@t!!B`=AUIYoe-1{g(hB6EN}nna$A5*&&lz=oz-R*@hQ^C$T<i;+F%AbJ*w
zy9Ik}v{adzFzM$6Oiiq`7|{nqKJy$!KuTne;t3enB$=bgfi&#qD53*2Iaxuyif}MI
z-iyv!9EGzMPl<7van?fJLcd}i+jy3dW1!5i9JJQ)wWSrq+ezaz2?kJAQcqPI!@Yv?
z<y?!Am=Vzwh$+ElV|_=;fP1Ck-frcd*MA8d-2uA-;Y0*E#8YntIKD0pj&2iR0Cumz
zMn~fC_aby@*Qt>2S+PPgcEyTJ+v$ORJURj9)kboS2A(e>=2H||4EFa!SCw5(x%f(_
z#`?OX=4`IP`*EE7W_h3W<A1|*N?;2*90bweQjAEnzrUX`#ISpRK;OQ9t|cKIh~Ps)
zSKZ_9%~HD-58SCTG5P@PyEl!|?};V8JGl7nghuezw}pkiKo9Vqx+%TAi$uq$94~Mj
zs_Sj6NB2RM26BuH<h1=gK<X1Ar1WerWI*l%oI1n>l@k<`JnrBqT5!^AIz1L(`*0j=
zeRE%i{QC;yA5ETexB+X!c}~koIoyEB=@te@*Yiy_R+hlR26UaDt${Hqz2ON^6XiW;
z0jnR6!)jNk{eLs!b+yNPiPU~MitWdootoM&FgPixOUB;<swS_~k)icLs5|0dqwdGY
z5TV33R-1s11##%`4PgaxbRIPAbjE{X2&@g4g0&$q$#FxV+bE}eFynbElQu1sYq<}?
z0NH7Se-$5PL3I7GNcyW#kZ>#r_^poNecC3c(X5sI*zSK1gm88oA;`dfCN^pl*pFv(
zo{%7IV=<Q}QFt;KMCeR0Ux#?@|6>Yd2(h~7Sr+uzxwnuz2@MI+QTV;7Gl#LbpoFy(
zP$XNRb*C_LP!x4(KBk`n0hQvQ=3B{$h*y!}MQfsBC9Dmnh1Pf_Trm1KTfkAQgtfWE
zqK&W;)V3H_GN@Y#P0FsRpvgfVEh~{@b-2hofRxyWi^wEjc!4vt!$pWQLF7nB5~XvZ
z^O+zI0J6Ni`XFAh@+!OPMtN07CY+;*0<nD8G)aMYt-JOjW`3y27kvW0<P*TIk`ln)
zZ{LMOP0n7o{*uiv&u)f47oW9xU1k7IOdP4?n<Xb>S=WWP9ly}>(Lo(psL0DmQcjt2
z-`E=+N4{0L=gkTC2WNW>Ou#GYe+V*Fb28n}#d;<Ecm1wdt@bRe)oQP#|2{sz&oBv-
zA9(7={a#E*YQ!oYn-CiRsU#Q4Hct=*2u3D|9)U*I6GZYY^ax45g&wcvTj=o`N0M&=
z4HW+GEcEaH0{2B0lx*#Z?0|GFy?Hob)O1|ZVdqNMcsO1$*T}9W+Owcb0nf5wWA@TZ
z)~&tZtc$aoH(s=E!=)*G&H3?sO?&XEy-%hSfh?<<2!Hf^(1b4eqd~tb=8x7!{Lu;c
z0QX0o84`G?m-zjr_E07iC%{7?Bg`HuKrrH=mP4c19tt5z4~5s#L*aGEL#;&*bv)cp
z<)LH;$vo5uqRZCCmGO#sD0bBi4;3VD-dvuytb*xOJl(b)9CIgs!kJe2(=w>k1ZUnA
zW^sFAGsGD+#UCh~<}KqsXH~ONEY-1p|J>eHavFHmP_FKPe}*IOFoTp$23eCqwTWNw
z7-UUyg%BoF6imR{^jt#CPQAW*rCA$xT7{L~!3{X~x(+Kpi<NS=TC41=!3o)|eQ?|!
zY<O+tO0A0~dHwPkCl6nnID_Bg%^7aiYQU|m=xc*FOWrBLv~_EPH@)sO8gNK<z2(i%
zSB5!3J8F!RaCURw%W|bAyvy$<aZIPzS7|n?%|>fOVeSTR#!jc~)N+kV4SJdZG0cIZ
zgh#6IWoE5XX#lEl0`%UN*R#6i&EzO#0q!n%pkI7#)=KbgcB3+!9m);w!e3z);Z5E_
zJD`Iyyhl*EeZ4muPV8=!n<Lr1J7B85b-gzSwgnffS*VzdFs0bwE#h~alEdy~@nrDq
za0$)^1u_7VXyD(NRdd*z3ro&!2N*fo>o4R=jjZ@O)tm3^Edu`<xkj@d+?|?thMGH}
zzd3tywKDJzo~Svw5x6r01TpL&F^FHW4BX-sBY3Li&1g1uta{O^dJ(RV(0|N2wOXZ?
zEx_sIC8q|wM^uL!EpIO8xzfv2)=SGfGW1KR3->-SNpBAC*TE_i!$Gz_LjNJu*qetm
zfY#Luz@ie+)beJRa%DJ$eWwE)&%^i0PPyeRsKGJnwLBbJU3JQN08?a~?gO*f0b~M4
zig~FDg+6aiuGy%-iP-X%HyIBJpPg@xRI|P^0%O??nBH+}<y;9xH<#UEYL3zmZdL0}
zGhfN>9YIx^=j^Rk>P}W|$+Tp+k=<Ra7rC;|&E|{s+)xP#5m-)>Mgj&m3s2w%vtSkn
zoz51IWp@-yKpc30g?I%(;H+V0l};IZ%Y(>g)^fN!m4Cgc$D5Cq2AWge35r-MjuacM
z_4Af_bE-}?3-$p=+?R(L!)D<f^cq<-3h%H&qfxD|S-l!=*9WVkjY6e7SgGv<TRH@v
zWC*~MzQ<{K)8YQAr6;@q_&KMxV>r}3-U+%tPIix}C#vi&r#4u08ao&aa0ROijgb-n
zo6*P(ffeN&h1NyWy(2*<N3y<g`UGU_Rn(CE-n{x~xsltOEfjYaO7I`(Z*QrBsy~ap
z95gH54thPjG&FcwXtq=t&I#5YIb48Z;V=lE+Z%A1u@Ii#)vN-DIe=ZG*Z}@>H?SYH
zr!tJkfBSynAaoS<=1}Od;t1+)sgmP{(94_=sLaTalLreP1q=#~(*PKU48v2qvd&1Z
zSjy(|dGxp-rR%-<*faX!x>MT0z1e<mic{X*+Rz;GrgLus<_nrwEa!`08;xuuSKH}8
z$O65Y4@iM#HNeqAA9-g7`f=|d?QR(=rR+BJw!{C@ht4$5?oY#;{CdEBHHO`*@#Dii
z@NqwW6k!zO?!b>d_(KETK+itF(3^ox8>3YQfH&UH8t!L}_Ok~2-KU@hyuWoZ#2D5$
zM9|X;GrT@di60IMQ^lEtN5g}K#Sk4&WBHC2$fg$S5W8E~WAgGV=vPeW_7)cLxC~=p
zM7nSW$M|DN0AlRol!5yO{J04}dZxlhAAbDfH2C-?e*Aeld`zAJAGi0z$KCjG3mnDi
z-i{wn;m5D>BR>m1iumzU{CFHcZki1rZ^MsQ!?~F5>+qvI7e4Cv@ooHg3_o5y4?bRp
zACKb4_weJ{`S3C6AozG6e%y~Af5DF_3*h5z2gApm__6K~__z>1p2Uw|;m0Kl;o~y=
zcmzMbiyxOC3Lo3>W7A>qu?0UaJ{&$?iXZ1b4?Zr!kI9SRV;X+^2|s#{fREqd$FunH
z(IesGA^iBjQSk8*{P?@0;p0C1xH|(M@4=7X;m5Q1F}f5OKLf((XdQB$tT+=tD&OFg
z?$sACYs{32V4%5D1|nmvu?j;~c_(`f2ZW>y!~t<@P)<I_ZWoLurkuw!T$!M@Pnt06
zxe0RONzg(~t{9yW&2;NPea@TehyGqPjFvZT%UPQ?T(IFh^jw%Ft-q>A$;MEP|AK0~
z$h`u&$L6%y@aa`Opvu^%Tf~ny;Ky_Dq`q1rIAq(qSn!g?DoB?2%7eX)Or@5o7HSX*
zGQNYgW^{Ku4Ww{%Y$+TOmI87lDuvhK-G~(2!=ZH&0bqCTEAivf<KW{Y{Al3ED1Llt
z1$;b=AMaQNAMe7CV+Y}5Ier{*B77{ykAq(TABW?|tW)4)K7PFYMeuPKe&8E<Tzq4X
zd++J+K?*sx;*0Ti>~?{u*2EUoDPd6|T~SfJ3-8jR`b`^K2^~BB4B85I=i-ZH+)v^M
z$;R%gwcv4gI}HSIS!@Ap3JU-^5*5J5@NV}6fZe(HjvaR&?6G5~-@CXs(8axjF0R0H
zafP0XQ>`ve#kx49;o_7>)0?ekfh;eXJzRub5)&J6Tl@|t`|>%S0I7HQbC4E6);Ka$
zDfx*dd-UMYP5<c`T8}d%RMzMqt$!AsT9Abw67Ea<0t06liVRnpIM9Gm>Tso0;t3v%
zh^MU;_iutE&HW@$eNlh?+SVpWudBHlB<k`VGLYE~(CT~Y*-hxU0bvtfPX*zz?#Iep
zW(Y=AyWp3yb84nphVcs^;N;;F5=a=;`XMw`AAxe-5L-E~4=Jad2a%H-q(WYFf;o78
zbt=Z_zZ$OnU2LwsD;3vXlALRYUKS`3S0QGUz8<yxJ<yTQ$ENl(w$x7XM4B~q-KI?&
z!IqSXE|G7&x~D^uh54O}=>@GmZ3z|bzwPe<BYqg05#Ni;2=_5)Vc^1RRW95L2~8(!
zaTK}397w$gGZGm#0?t%mP7@srW|nLQrHF2Nuzxd5TH*9xrr5~j>hLQJjB~Y7SpN!H
z=#Vp<gO@46Z_DbLdIjIaR|~w+&M|}bqib$c;RV9@eQaU;)>arjQY(8e*l_MfKl6WW
zn{4J@3vrxz4h-1KjM1swe(HV#1|EFA9I%+`GU}*Zj-=*83w0zF%k{6nE(QDA8mGPS
zu8vLaN?UTJ7i(v{*Q>^Bqb;YN4O3vdGHiA%d4iClk3dT<jLo+5Y}qzhn3*=ZdqWB>
z`5*7|oY=w`vK5B(dF~u&A&1&&gmD1r@HMg7w$GMrMjgJPjSl}`+=h9w+hYskZMMQN
z>Tnui96&n!U~IO1%$99N9roJjus+QRWez&)-hPS9J&RCs#8)Oa;4RL4P|ed(CGTv&
zyP5suG-d`mG!}r`CZFGmEsby3O2eqccG6h?Mx00^XU@8iNIemo_Mh3(-YJp#y_!gw
zCJ}M!(#7esG3Uax3*u(az0krcd+x-AKdD?0Zy6?q{y#V`y7_vj-r+y2IWBUS#Fom@
zwo)<L<)7MkN%1aRQYxv3a$@Y0x}PA~KRq`4*4VPo7-XJH!M-kdvdd$Wdzme{oou*g
zvNEe>qb&zeDpHTlwyG`Lj9M~fvN9nSlWqTlT_5*&Z;CC98*GJP^mu86aRBM?y|LN$
zceZRZ>Tqux9sXY)zPS#6DYh^^Zz~L=4yO^u0i?tK9-D1Huw|Q3ho`pDVLcK5-#R4x
zV{B<WV=E1#65B~*J$q|CnTH;GPP1An4#R@&Oc4g?MW~=>_|m=%t`A(3S+Z}5UoITz
z^ze_axntjtuDNqi+2?%7g>i>U2SE#JsKiz%1WO-u{zOZNz=L3&);3tppsdqE{y5%2
q{z_uCjPK3#Z?i?5Ghq&+)Xd|e(Ve({vp<iEoyEFf?6WyE`2Pcam7Wa%

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/conversation_template/chatglm/index.doctree b/.doctrees/autoapi/lmflow/utils/conversation_template/chatglm/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..81dd3d4b102719b8a0fe21aa94818d8c879d0d51
GIT binary patch
literal 6705
zcmc&(ZEqXL5tb$KRWFp}SSe~NRZt|6n~<y%qcBv#fMKK!U|T5)qbO1*=i}Xyd(^%4
zZci^T475NKnC&7h;PZ#J2#OTwxBh^B%ZL7gB7dmQ?A{%Zq-0wS+8Q9mot>SXot=4R
zcKoZ#vp2szr~b)S9CF7$?wGdi2R4gSS+@f#@t7~tr|H#Sr;pPe*@(=eI7lLkrE(4&
z7I$1bVt)E@Dp!eJIBwj@*4JWY2`BK2<-EeHy!LUr!yB?99O1Ikl>DCS_1)m8qh(lu
zf5@WPBmstCUg(;Fbu4a*f$OCt1wt0@vt!ckTx&l~^;xQ+3C%^y*JZ;fj>;EgH9O(H
zTy!0urTa<Gluc8J$mt~lmLW?lICe~@u34@b#~|W&#Wy|ut8ORZc+fMggNXHsCy_N_
z4rnUnEq<9V^DBITZ%N+efycU0;`=P>9tF`s_wKzP!5sM?C>VFOVqG&4ff+hoZLqE~
zSoc(ebu)u?9p7fhohZg}U%NA(^!S{tC|c<T^yTpH;CTFX*!R2myn)YK_*|u<!dj7|
z4!9#1L(_>;`K}k(iOYVteU8hM*Yu+HIo9y%zDUi4zs=iYtL^jGa5sEOJLlND)^dVu
z-WuPU@|n#EQsxSfcuU`dmS~=715$vl?WShu@4%XMWz98V_7c~O3@dPrIAqp247;fK
z<=TUvKf3n#(YufHzbS3^;dncF^1t$u-D}r(_sGrS#PiJP<R;t(rk0x-NFeMwrRh|z
zmqrFmF&WygHu3NF2e<D3`0l;;jQj83`^nu~_up?e^Ej?dugF}nA**5#1xYw#OoEUa
zOnpvE*RkZtu?0`>G|9L=jC+9l3v!QnQ4Ev>d{h=dfYuA_lNm*!=Ry=5&CnGFQb3)$
zqycO)2ay@_(n9%*-SA{L=nZ!t&e(E$@8yH3Gqd{U!w2drb^r^@A@gm=9~j4gtn%(3
z{r7gy{h;PLf<>lV+LnKj#qIix*6BEWzdSp90afZGay~psWi>o$!-6R*BO6`hU*HLU
z2GE}w`mzdp0EJU|<3GjU6#0{6c&6j0{O_`ckvIvFOBYBdxr8x04qeA`MEXQcADNM#
zj${KDuR-&CxlY!xP-UoeUu`bTXuw1&zmdBrT>2xj*8dfc?p(l?Jwxzr^6J<Vo|Qd;
zoq3yd<xlw*|0e*P<otn_J@5Ky!RKK79X9;^6k24}QJ#Qau<2#~^-FH_eDz1r=Cc<`
z`VE)>aA!-Bi-e*nY62Y40e62|q`XwZ_H8%t2O6!-=@7E9`r-8IGLvi^d+u%=np8g4
z!-&N!I%LY5KEc{NsXkD-W~}l=?Pca=$rLp!Qs@*-$ja)fB0)c=MrhqrCCv8iz!w1I
zScB1~oS_#^Hr%){H&^zvvr&WeLBTL=6)TiVYw)7Zlo!2R-Bpc&&B(6ugBd*B-f{eG
zjS}Z5$F6=;FqkOKN(arVvpKnl*Tif&M7d^_W{lWDK`fGJ;x%vz-0P<UGL%|H?-VMd
z^S24F$hXgt1M^!x*P5{h>=pNUTFJO0mvE>yJY1<LS4d+8!)iQ{e1;W>8dmWIUm5O_
z+{?os6ylY;Y`|USWJ8^M>d?a*#7fVhO?isV#v1PzB2U-2P8w@<>$K5}G`)M4rt)H`
zNIZ1dk%i}EDqHza!}Kj4M5T*+QPD+NGf!Oa8QR0+bG&}=GJ^!iPL;8WM9%*~7XM`W
z9NIi%_!+jrp<oBEJ{T#LO)?cSHlDYWg@mkTNt%BpD?07+s;u`D*Hz}!`!@{{SOzLQ
z!kVuX?K6}HiaKT^-L8{vTiONS71wB}@)Xr+${X<NBhY$zW0KfM`pCIBzBPSh!q^>M
zFfH1qFsDzMl`z8*1j!mKp{khSZ)vEUr#$w9wti`jWX%uqpQ|~!t&8B1T+Sx*@j5uN
z<HZG2%MWr|&PS4HznBRDY)nUV#BOG)9C~MtC|mlfiY?M{X33_u^PMpDob&=LA3Mtq
z-^-g3ZS^iE`9&k6oi{Stc_V`r>Fhl1j~RUybT(3+g4g}o`KV}a@OR`ZA~O9LeHCi4
zgi$a+Le|ZViU<U%@H=y|ne#9k!vv<~7v!pyMASMliblzDPSFh$S<fy5U8MD*E!2;c
zv)ZV_e`dM~8VNae%W9B_FcIlKZ=p^dFf>6-0gS)^{>NfrdLaf&Q4k2IieNrY<%KRv
zX03vOeGkCU%!WziVzfeWM$H$RMXY@;H~J9Z2ZEKzPvi<Us{F(=j{8C6q4Hjh(dY>A
zl=rCp6+_fzO<^um6ywMd+{jx$*eP3(^@8^0a>T+Q5{AnT84Niuo7{AZVMVU9V+ZyV
z$a`_lRFM^%^^yU$Tcx_69$}EnOb?U>F2Z7T7}C%2VJ(CCfL3yO*+@jceWR_=>B{7!
z!J;UL3~u_i%h0i@lBxhM+0r&INvS9)soa{HnG!Ml7sdIimOIuVUwvhAVrsEmqcgzk
zVh$6zsuWytO@EM>0|tvzs+dPZQC^Ig6-2gSGkTC>8|pmm%PQij4-ke$96NUffb1$2
zsm7Qj=YjV~yonbYg)_piir@e=Bi}^byDwXcqTHNre@qHvme_%Tk`-KQc$Wn+Gs-KK
zmsr9$bYe$4Ys;{mn3^9rNTy{i4-y32G7b|1qg2E}=t{<B^c`2V)#N7q!X_w;RaByW
zfHWIdzJ$&zu1XK@T#)M&Y499W;~d`a;l+4w?Jc<)ve2MsB;wOj2E%(kgttR?3JUox
zE<_mbcDo?Qo$y5P!0!an0I_t5IF$n6EPAd~sk5;C#`mwoo~u#cni5ZxvG_N0@su{m
zWe-@?ahT|<Jpct=E<6{THSmZ)6scBBU2efAJ);QD;sEHRk%x@rT72RQ^Vs0dfV=p>
ze;4T~lFH>64Ju*$GC@?wFw{jg)<eS$EHg9h79bb;;No?>!OF^nSbcy7(@6ENRZ}{H
zff2kXupGY+AINZFgMcDVdQ)tUN51O@rVd8x8Xu%^8L!7|MDPhT;0&+sk-TJq>%d^>
z0JsJ!SVCN-^?KGThp)PQjb<ac!2Cn%KFhjB6NE235gF7$YzVa3=q;~0&+AYMo+XfS
zfaNyznjfcfV}eShg(5>*zZg(ggId{OuMtuI-WmQ4F+Th&eJ#^hgTB6?uRq}{kqzQi
zS(m!uihHCuM~VwM_;sR_0KCQa&;_ez6<m(_v#;g7A_M2?7Cq`1X*zvEDgD*?q8M7t
zUwBDRX~u6VwbbK7m20YiPUK42qtqE!&|1+659iht0*;(d8?#IyZ&y@!pw~lhcY^hm
zDPqChm|$z<yDCW=6Y`M*B9S7ZFHOtJW{`vdYgUjDXd&a_z3$@ranZ=Gwc8o}DemFZ
w^rr|>6qX2f+;t4E)EgG)HbGu0$jY&n9N%&in;dyS|C`aY34k-K(q(7o-%ota%K!iX

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/conversation_template/chatml/index.doctree b/.doctrees/autoapi/lmflow/utils/conversation_template/chatml/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..34f63ce2ad2df25f35a24163701786f0fb5a57a6
GIT binary patch
literal 6664
zcmc&(ZEqY&5q50vt6$c3E&-JI@ChB~!0qnk6v1JvNKq7ol*ETnI3bj~$?Qz;_N?cP
znI69=iUbmO%0lZv%H#u}69S2E`~be)2Y$hc--E<c-7~YZ<2bPo0g-mSU0q#WU0wB5
zb>$be=l}ZJocbp_amXG2sAt-?AJ{BTWz!C<#ACilpQYD+nLbIkWh*id;~<GFmdZJB
zSln^#i23P<saz#?;ka=xTi=M8C7i%7mh&2~^TvniHgCzAaD>ZBQ}UYY4c*|dr)5}y
zf54*HBmstCUg(;F^(<}*&rL}PWGp^lN2J_&mVTV-lhi{KdW)2=%a&0blrPA7cD_Bi
z=sG@2_mY7r+olkaGe`vNLUvei?3m8nuv{~aLB#KxZ+iMy(@w(aU|?GN5gQUuA{)XS
z(NxMi{0d*@SNQ_pl%qZmJl2mA-)B+(Fo^d1_wN4y#>oFb!MLv#>zj!P%+Tp;bM=+E
z`Y$wBUz@A%_%=K0MKMnM`rY|tz~^L5(MsmmH^aY<)A2W8-EZLYCO&WBbB#_4V?~ZS
z-?m%~O(#m_buX|JmwkWh>^CRh(QCSAU&8BqA~h5KZQh-lY>&T=i{VS!FQ<;RmJ?+A
z*7VAhzg(UnWqtsOxAY}wiRP&mAO+~oZfbV^Hf-5cwp<hDAaTveumabJLuQ@Btc!|Y
zu08ze<LgfzzxyQro6>ggPq&jt|05UKxqf43ms~7PJkN}dZ^CC_X?Zz=2Si-2G@Z)z
z(#U`%W`p`t8~<+q@YaL-_lyU3?*HiCtp|77?L2^MRsPHiTe2=jQILcuEJ@JOfSu22
z<|dXLI=0~H-8Na)hh_J1c|q<juX6vAfS;8G5ReT5`}l;S&~qUQ4o}b(_E9Xoa7hcO
zWR4;;<fVo3gPrhrCm4)(9#t&4wR`3;>Q$EBeDqM=!ZyHQIb^==_#@*8I91;GC;yd=
z^V@H@j$o1LmUiTyWue-ssGCmX_0OlrE09SYLC%NAsjP>`U05w;Wgw%=`x9>ekAU@4
zV_nu^2w-k1zxyT8Hbtr=5uWMzDgT>nU?h%1q|F6VNiJc`jzZV59Faa%(}!l{rxV%2
zU2BXyS6-7%tW()2-BWuDGa50G%I|$?7f~4Wd!(oTC-&UEfXjN0P`%9SQy+L<_5pV4
zJ<?@A<D2{+z-p57J6iU#i>C#jg7G)l@V6Cg$ZDWG*SuoWEBv2l-00=%PoT{wuafjD
zFagd^mn0VnGgBmg9MA#xuqsksD&hE+8~7uQ&1N-tY^;7%U0r69ts~Ffi9?f$#AX<=
zm_-LnInzg2o9EAmDz8jco~g0Sy#A<AvucD+(T1$7ekyYDOUi-HeO0S$-41*K_>DCP
zU6v>4g_DgpF3io9{p@tSAbn69j61~&rP3O_s8{i#vlUqN7;ubCD(_TqaBJJ~w=_JQ
zz3RI5AECfZ{Z%?*R)WpRMZ6YH);yG0R%ynF9hAHxiDs((69BIt21rI~4ZT9BOwL~(
zye7YOb|A|y_dH9c_r6=);)`m+ZMlR4b>Y`aLAgTuD&SS)nPfArLDaa8FL=m!hvZ%v
zf2R<yoMZz|GACQ=)RiNT-y>Ff>%dPdjxbG#<Gn)Qi}h`izFOHPZSyKU@13TnyjUs%
z4;*%A;h~tyPX5y{eTxTC>C#?RR#B476V<zh_UriUFB?3%Ai%LxWvL=-^S_h1KdK%>
zTW5+m<1RQ9(%?Y{8>O;MmO_5UyLGl|ko7Dj^Uq~XXIWmC&0*rY%9?urwjlz`KnX|i
z^0}gYf<i!1r);F#4bp8>-#=X9It^87qH0XJ0v>dPQ)jm&iG8e(oQva|)gu$EZtLo3
z(KdxyJ!Mv~j3*Ez8?c0`S;oJnq4J&b*bln;6*-X&KgfTs=H#xfawl>*o6N`S;K+^_
z7fc;L$Z0trNuvE?CIsLxozM~cnW=KLJ#j?Y(N|S$kxr9IHno-SgsGQg6u|e?S+@CZ
z-e%~kCpXD2IuzZ!L($DT6s$;K=V)u3&}KoWALS=_x1XMXimnBJTfQbD(~r?Eq0UJd
z1tVl&-L<HRK%m^dJ15&Y53?~$U|4=Zu3AY%JrSemfGp<}-3gJ+>?+V4S}(di!$`TR
zje`3F(@oGK$U$4ygG7XhNcVULMd^s4yI~3d1P1Uv77No0F<6R%KtNT5@=+=;^idvb
z6%6dV4~BL&Od=Pf6^b(Is?aPV?NhligaAJftVDh)SE&EwC!TRM3?dIz_G*k4Mu=y+
zM_sKLqONEPbD5$ThmPPz-qXQO*@CRevnQ7$76y?pTz0@<$a&f3rdteaa-AJHu%AG>
zi*u%$tl4akjIiA*)%f(#f?Q^LptNui7Nf(EevS`o8O#T+lEceZB8J`fx(b`F&Q2OE
zih{`Crf<6py_h<w3fPhzZS#_pnv#;r&B{zg#P|z}@>MN&szbi|>g+^iv0S4w!0TcT
z6S=DPTX9W)l$aw1i&LJMN1sq$jF=TfwqY}RbYdIoDecKRqG<>ShD97ZcLhM~DpjS%
zlqBbY_Q<)37aD~#!mx_q05T)rM4`JUJBp&*oNj+g3S*Yofq@zoTx)oP1u-+qE0vd6
z!Z>hZM>}iBu$`DX8#qX&Wg`y~1l%$X69b`C#6jpv#%2s1S2fJ!W%`9pPzkH3MCkx&
zHm-aL4OLu~9^Snm*D2EAIVit5yx+s?@c!Ccay4Y3K~F}+r=<*rH+u+ghwc;<@~d2k
zFy86+L5_RjvEYH<3!)KX=@M}&1;AOfQmHg&VfW2<Zor<a(a@@hC(2a(H*@inG{|N5
zS=4iw7^*!01$~abEH-Q5)qp5cotCEDgim@#5uC*V&?O@enaH*H*caxJ!JQFz@qzy?
z(%U1I%Q3o7!uWZDsE%Q%i)yTgh8tLBX4*{vF7(00>v)5el?kzWAN`|=YD=qLa|Qzw
zcu!zC{t!Np;lc(1MVt&OY>r31>jtI{MrsG|r*Ii>z-&bDF*M)|Pwa`jWPxknV0i7j
z21-{#T&4AT)&hsGx<id-6S=_r18OVFrbZKlFFX<X(?M(q^w8+xu3F0LPzs(UkZ^$I
zHZ_r-q;g}1N`-|YL)x$yPy>T{)L?HAQUBf^{}nMl{tJCA(^reWKBKQc;46_W;#E19
z+T4nJqBtjt3mNzgqLTnTzIM^%s%Pa}j`=5_$Gb)T&66#<&oPp8`jis-SLTaSXfc1%
zB?F}yzp2zxZw^(gsoFV_D`k&TXIw!WMJGI*TT=))az1U$GKIWdQQ?8!0DapT*4HY;
zg1a@t*2;HPk~U`KBL_qxMMPh!%F1StgaK<-kPv7g<Kcns;`(vX$gYjs8T}dV;p6nj
u2vL-k2zK0c46oE17U(uZUMk4Sv6dX)aub^zc|`wB(Y6VIGpy1@XYbz?eZiFg

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/conversation_template/deepseek/index.doctree b/.doctrees/autoapi/lmflow/utils/conversation_template/deepseek/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..9a6f5698cdca8254d86bbfd54193f84b0e61b062
GIT binary patch
literal 6724
zcmc&(ZEqY`6?Sazt6$c38W)kcxI(3ln%yNOLai~8qGBZE#8eO=1ZgL;Gk5pSdUj@*
zx#Jh2NFdQF3s(gxlMjF@1QOr)0sYnw{RN0W#B=VQnVsFWlPnbwY1cbv&OP_sbI*Cs
zOMX>)_U>2b)IZsX0`7RHEz`C=-)2!FtF~{&1Lle3X>$G7$>U^4)<W|n^5f8AiJSw6
z#U0lUnU{Q!$W>w&jvKYo^_7TO!tuR)IWO@tuY8c~@R}?MN4Tu;O1|X|x~_lH(lRXH
zJ7!^Ik^n=nLExH#wQR<Mh_R!D#6Zg8eRfLf&9?ZXM4zS{n9yA$d|lRz{IGmMmeVsH
z%0<`lSaKM5Oj$RD2%SzWU>mZ<f@4Q??uzA_Q3N7>mppTze^u=`7(H}M>nLPh;)!KN
zm_2%x@CLuim-#imz_+B+=Kg@S!`Sm!*go;YqxRi<--ki6KTt4gYsK1TEPOL?+S+7o
zWwQ3UCTnLVYdfCJPFrDw^S*X4A9whiEGb&y67>D>ui<?Bby)c8_`HG7oA_L(v%*}V
zqfWRZ7X#A?6Zx%yZ^thC&h{+#C*RhqnzJn8<wKE}F@K9UM}|A(ui<+5lJ?M%i>+k@
zS-&&7IOR8+W2Dp@An}gA3N6vR&<dmg-P~=>&Tqh;Rb|gLVRmBI3=PY7jVNH&45nRF
z{BrHVPaoZU{OIk++24e=`{8Ii`OE*AqwL<iwYx{I7R7^s8J^vS=fKu-GldGoUaRmr
zk?Vyg1GX3s@K@{j_r3l7_kX;<e_-6-zxV##JNNhN^(>f6lPgkhtjV(Ig?=1NS(AXI
z0$a~$>ME9;IJV%)L7lAY!MaDdKtFSt7llDVz|V>T2-Z5jeKw^i^jrx2lPS7FLW-$#
zm(+kyW-l}YURWp}>;`AMey6|taLSh3dozSnYijxJhY!?U>;M{;1LoO|*E3FmTgB~v
z{$Jca^Ms1)2o{=dVPF1f8oKLKYA2%z{_DjN47^fDk@LY>BFn*96E;j(5#H!3{{mn5
zGvNNh=$B;}1X!HNcmHD)PLN1R$iQ^Gg#SZU@FdOxB-I5{N-p7<od&LBIU;$YUZ0qu
zmkebMcdwCi#+)asSgA5qa;SC}X4qpQk>CANE~C)tk4RttUsO7{gll_-5Z>hFkvBXm
zdIP)gM(Nt0@-6-aaGd1)ftEeb1!%!1VEi36{QU%Gq?J;ZhhDJhRsOG+-01o0524M+
zFOu{dFahQ+mLwMmOA{0aIG_XW?xaY0rGW6;uJ83UW}A~?WMlQi$<;+BSvwuLyHQ|L
zDOn9d7P0V{DR=r1YqQ+?KxLef%41cSnbju~)U;ZmQ`8|Vt+4Vu{hU&vaZlAU+wb|F
z04zrukT&HMJ$JJH#-+KrqMu!i9HbBGhJGVop;TIf7quq5=;aEqas+@zj+H-{!o=+z
z$J^FGF-u)`{Y&G)SdCUVYFehv$wl-Q({&N0oK<>bL=K8$5r<>F!8xF>8wn&TwTikH
zDyg$hg_q=wS%59O=-C#HykIZC)AOpw9l3;KHR0t-Nx4EgD==2iV@arAf~bBOU+|ay
zF3G*x|8_23xy%M!W=_`B$tRB8e}`D9JKB`zSZ$>7VJ`A~jjN=wR<}wUy-3r$7ilUl
z7s|zBhn-kxFB93wej27{aX&0v-HXaEN}gHzx@BlDk7ntG(5OL#V<*a5c}C}-k;y-t
zJczc>2!r}fa476S69_9MvQD-_=0?9eUQx($nymTfvZS*wFUx8-c3ovpy?@;hzGa};
zBf$Aw(LP3rpr|7@((M-MwxxXlZgG<yRkEUrO}PV_KtisUw<d{wq>r45<6DzQCZOHX
zmD8ea3Ul(5X(`hmLXfP$5~`Z%|CSz=_Y|K!zp4Aqp{#g*_H#8OH+4Ncl*{SM?0Fp=
z>G7hRsp0t<E#o6ev|q%806-=~I$}FDRffM)N0beHRrwalD7~bwwzHk^>Nz<EfIf1T
z9ln<hCz`5FC;9ngqnS-Mnpv9;%cf0tw(R8wB2yYI=%l241>OF|X(^v}@Eh_K5t?3v
zQ4397g3#|FFY9SXN%#T<`N5p5XFSZtV+`Z+OLEnULz<-+`Q&6dqv#2Wtftq2!P0s@
zDe8vGU2PQPKQY}HgM|#hW!aBK5R2rHH&C$l7^Who08HQk9>^kL4gx$Zg}yJKD#H0R
zk(b&ip0x@F_B{eaJ$;Nr7f&k`X*7+Yw}`e+<VF_)JYTQ^`H5Vi*_9U$jMJ_k4p4)y
zMi_hqXzT|x5sM(|s-`fPDT;C82ySGvAncSbNQXm*ayevy9}2@|#|(y?mvwHs`D010
zvr`B56G(nh##E9en|0zIwp*nRfEqH$W##~s8ZN?ObQsdl@n9{3c|cckdRdD_xA{&}
z;nTJ8mj(+%KQy@M*)GEbr%b8>x@1G!ydb5dq$F}{;?0DJ{y!<wSGC-c4%zB!<1Z!_
z%QZR!ye{G}k*g}g71#87vDss=IAx1@%oye6kXe3c8#bfn6x+}w>QI&uQC+|=EaKRi
zD*$F!sa7>cBsmYnNA8UWfss2S466tZU^Dbgl)s0vp(x7D>Gnq?%pPOgH&D5PYYiQm
zA2Fl2QhA9bjAJKqw6iu0+lgorf`g=5R<a;Lz%Ao25l~7+9E7f<Y)02{)tF6g(l2a+
z`dCFJN(e}^apg-G#^S2<<H04lPLT%BK~c`({R8xl_txH&s{soPYA6w(mNFQ6`vBe!
z-6<&KSGW*CwA*fj9JPWo!F{jghdsp772;G1fHNP%Qo+uG<{RI<1$(ZBU28%-QAXmw
zk%_0oK`wj5!j{8CSM32PXmc@eu~`kh1fodIWvX%uJ~=S*;LHzzxf*%MP_9L1o-j`h
z?)12e5BztLI+8>#N0?R#;}<caI)b4tt7rXaxV~kkrrokQ1}qjXUdJ1(tW1d2N0>GZ
z)ks^-sZ$sj!h3wn@w)JV6c;uKD5AJC!R8FecU|Aq!AN7|qXaH9=r9`*d<G3TL;F3H
zS1fQH84R-k*FYIdh^w?-Psiu*Rky3rY$z9)cTD4GS=DHQ@P#KLhdPK2ftedM^=cHo
z4yE8(0!as0ZqxAjaUwUys8n1iGNg6$2O4nDY#ZzqBI@5e{l6o|`+uXaW%{bo*QfOL
zCw#@SM!YKT(s*2P4;AN7aUl!8PIMB0mTV6rv2t3(WtczRH1Fj(I7_&gSw~3J$rDQI
zug>S?&|=o;#T}&?zpd0#9f&H}R1F==m7+(fGp=CBq7xp@ttkW?8J{+0kwV_CsPMo+
z2b110)>kHoId^T0t(NVoByEhzM-B)@f{4B{DJy-0Bn()y{Fp!s84qo{i|fZlBfD1K
zOX*K>4<98zLWrWYM6lzoBY36WFh{pB@={J#j<w`?mK)pT$UXY+jk-+$oMM$OJX`+;
DcIMVp

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/conversation_template/gemma/index.doctree b/.doctrees/autoapi/lmflow/utils/conversation_template/gemma/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..74de859725e2b047ead4c76d2f8ba001e1f03dc0
GIT binary patch
literal 14508
zcmd5@TWlOx8IEIneMx*vNT5pHZh??Zh`nur7Mu`~rUk0hZdD>EZRupZJKi(LvoqVd
z#J1W<AZV%EqjJdvN>N2Cgequ-ctNUoXhA&i$O8}X!XvyOA@RUXh|Bk%GiNUA-Syh3
zQY?9P=FB<&`Tzg?xAUKWzBqjKm5=U`e_|?VG230AH|urRt6M=VM(bWJYFcg>Ux|-@
zDZU(6#F%fc1zzOWtXS+pj~cU`x^KDh7h*9<-9y_6=9BRwfmI7_&rOH3VOC-zUx+Jg
zObmy1=vcW|qT)1H9B*x2!KiueCCd*?0$_wzv*nnfHQ%tBO*1AK05NpIS|{9gF!HHb
z(NSudz&ecCj2JUgGTERgCDbj6A;)&Dcqv*o#kd)UzP%iU;0keJ(5?qW@kq@vg8+@V
zA9l^Ax*M%Wt?XghtgZUi3iXM^NN6_bRm`T?5w@S*$OhT0SgA6vX;uBmbuGWT=J~7D
z`xZY5R_Grf7*rKvRWk}bvt?Hmv#OF=)sD=nN@i7Uw{ES^`vHi(>HOYkne7q7vMtAZ
zwJhvr5XWu-Z{LUiqxe6D|Kmg`c;(xYyowlVnYJH`Tbo`za;#J3zL!l@)sVTqN3haT
z7@HCMAe+nhwZv}1s<3@ZBr}<s)(v9(Y_?ufRt^`M5+MM@vuXttM6)aXga9qr8P%Pg
z1Sdx&C#OSmIdV+jsCkYNw5-|=m^vi;i|I!ndg8XrPuzZ4|Hd@k{n>P4?fS^a!fhW}
zI8EXeL{Kt+;|v4_j24Fz!vJNQ&%KVtOzz15BZ}&L(>VSduhFo4y8F~Ui;HKC3->HO
zaNpSr_l%EgB_3`cn8@OoD1{B*i&|SaO4efp{N5qEM=@m0u7@l>KTaHW!QoXbqNl}X
zJ0-{gcsLJ$Y}v9`-{^?wpzp19d@)QawRH?=8-r~!8@}0Mxq;%`LTh8eTjmQF6;I2j
z`_-a(HQ>y}M<vfIFaY~omRq;ohOrLQl%M6UYoA3g?1*EBmTx*a@a&~T8%qkMIMc*e
zx@#gVjzm-JZEeJ&)Y_N>8)7SOplCV1hxPm}taVo{79|i1QyPmqc4tImxIco>G;KF#
zuZs~pg&Qrn!a>4F?8CFQ-g4}k9mdbd*K4Nl#+z8ujRj@<cH*qWD27QF6EDfBL(^|q
zVJs5kzrCkSmR?$v@4(Z%gO=p{0M_R!^zAS!Wy^jwzwB1mrK9z|!e-eoVR#A7*J)U<
zwWa~jqvLa!@M~=fp2RZRacnc`5%!xtC+dy<Ca`%L*gRXrrqUff@tIT^i^g*Ys$U$K
znSd8H7UIFTe-dQ@Ue;5vzb9HhS;Qu%{r}91K~aM%?6p1=|F6B)C{=v+MqY#Y;J}_e
z{VOqhlfA`$zQleZ?(Ez0wh>HH`TwAXe`#04zvTJfTaKZI{16zIXyTvy)x<wtyC(9Z
zn77wLYpuszy@^Pav3IAhKw}XgbnhgFC<=)YRAXIj^j>L4P8`aGNM*-!8!Fl{+nrY(
zqc66P&UX@H>rH1NXqmM08g2PjVELCUY1kjbSRI)@Dnqsm<-*p*)Ny@VYZB!X72|-F
zY)jH8w@08oaxU2bB^sVSrxu>qNmk@Or_U&xo9PIN!VCoGfz+1B{bN^?%^qft5E+P`
zb6~EU+q?ZkodRfPXzaz!G7_de?77_b_Cm6|)j_eP`zH}8FFsdx6|(A>mw#C*(XL?V
zuvf!)s)*Q5e4CxQoo?*diJR*Nqm##D*K6uFKgfSbf=#K5!{UZdb;CIGiu|=y5jv?(
zUyhN#Dp$McQVVuP2gPl^@uwKK8{c?W|Gu#+2S|gjr<?4cwiIRZ?@Qhl0#@0UH)olt
z)y7pPO^6C|mcOD^lwm7>nQr@wT=`3C@@ouU`EMnd?(0ujB2Z4I>rD;3CV#OHcyOGF
zh3F2i-FNaYb^y#5(b&mX@+Uj=&Wknw96Ioas22!?hs?PH{sfxa{bZWT%baAcsfULo
z_te8fPA$qTSsC^|#Ew5uHsP??(w=$DgL{>fu-Q&(pQn;&Ltctcrz51tya6c9x0&Yt
zQhTKUNdcsB#Hl37P_f;z+PL<e<~zQN{b?c3mm^0p011mBB&4_I)+m)RNpFn6MqV|H
z{6a!mXBo)m%B+X%O8IR{!Rnkg8y3g=Hd=c5b}%Aa*pH{n)M;TWVjrlOLue<_6BC53
z-2KbvLcXIQ-ux87J;Luy!AmJP0x8%d#w5*c1oMwkSLKN`lQJ~#PvLY%bCl3jIF1tZ
z?a*87hMqW>%Pd~9t+g7?9bz%1Zw=F}G0)Gf)OIO4%FpN|$!&%b+@SANIg<DoZ`WhV
zpWLnydd7`*Dim`v3&-ctLs}0U5P(Io7$-&{GlWx+V!BF{l4KBjOAM<74=ah$mB?`<
zGu8a#M(EWHBm>E6yd~S7MhZ%{X59$4j}UINYSkeKx6`9^GnDF+0)PVoGC%#ZlE6Nx
zNY>!AJ<3QXmse4{^P##Kn(b63sWSd70ErQBLT1bObMz=hC;#kvbLv=PQ;fKtzMa&~
zb1GlDDfTBX_45pRB;*BYrzzLdZMq*pqWJ<V3}NNsO(OBugjL#zCMI`x=98Z!!WO%R
zcIAp(>oRO=zXk`!^zDeKdfQii8}KgriweIZ*VZds)y8#7%9-i9UMSD(a-t?~f8_N0
zLL46~l#fXc96NKed|LL0hTzx)o7#nP+NFH#%ozm9(r6#n`jMHZpXepiYnAf>3RZsV
zfwDH)xXXb2tPMypKoSFMaeLUo-=HB!I8EOxDNr=cKURZ&lujp=bT(S%r_rQ&e!U#f
zuTwNST6X?vyyy4gvc0qEK+#f;BK|r}r94bud!64?|E_6*3jaLb^M`Tifm-5dI-qtD
z12wa+mk%KhboL<`f?IsZ4=_mi5dJuBxsOYc;!9|OH&~;e5?`lIG!x%Y?R5Wi%%9Nh
zZSG+f68tLll<t9F#%+-=6_%#Q#o}Ei4!0M|>3+EntF??F^9Xj+vwNM~xR`46NCGye
zjst_m)n{3L`jc<t?CfPtEZ;{nD*5c^AgwD|IU*+7h$z~W{<iC5fYCDz*k!x^Zj9@_
zU58giaHGznR4QBP=r6>z(GLNm&APN}H$%E;DSO#R>Q6^GbcRm-|C37NwK!eVaw&Mk
zF?=R@mE+~UOG@UuJME?wGOrFt{MkOlUzS{S2R7dU&A8~CH9vH;wY_tGe+QtI6F4di
zPL@xcSjCbzg5yg2c-}^fgB=EHZ`tu(=0ET5E$cfm*Plp1Ntt-B+_oH$(E%NVgZ-B8
z*eGAgZP=#T$V+UCjs((i`7y(COxklMjX%5=j;X`KK0^$g(2T%cM)t)N*G>*p40v0^
zhR0~g5&kfKBxQ12NBdPn#_&0!hG?PmwCW+7`SERLcDK&#<!3Gi?fE1^F<^N}4Z0xF
z<)6ooHko2n3zOm@H}Ws{o=7=Hw;AHk62PvGov-5cmQX9<+0%H3NU4>dr=;Xrvq?W@
z^S#h}z9&1)mvBnWw<}KdGf3RfjGWLmsIH8*bxO4H2-%jKkp~I&BYdku@OL175p!Kl
zVn-|1In>E$*%IoUk}E+w$UZ_H730Wg=Pis&qMcs*lFruct12`S)g2L6IZg*{`EB-l
zu}&YS<?0qkbv>dYK*`2h1JI_vC`8G|x}^M1acj-LM)BQmi4$L^yWi2>H`2Rr;%<{)
zk=IyWugdGU<n`O~`T|`EncNmw`krK4RYWFlFme>=P_iclNrs~s#!3SIkP`66lmL0R
zTLPv*l*Z*+!ijUmgZuLC(xAJp6DURmlBk>o-992{FSL+*>33wu`CqJSck5r-$&T}1
zDJU5y9EE4dM>G82^j<~u?eu5fVPdE^K0DZEcEINyFw>Ik&bBv0aEtBz8+f(jZ6+~T
zZ!_s$Z!_t>%{(53dHe^OugN@W09v7o7O8vQQ5nYtth6$Yr|3=^M|t;l*p>_xF{M(u
zX$lf?SW-Q$UzK$x60eS_f@-j9OQ{7;Z8ESGc3M}p&&d*Rf}ifG=X5>%oZN4~H|Qz*
z-tcV2Wl-s<R4$Z>7k96D`=8dxvy<Xop>MhYO7p2ozvX)k6sD*e`QZ?WKqo%GM~v$}
zRvnKK(LNw1YmrYC_(oc9zhAegx_dF2Bx>;?!c6*Q!HO>v*>%)`e%5p%RHW+wT9mvf
zY(-(b#HLU!*|1PcFQa-qAUACVq1kMqj&h&xc_C0mqJKRW167o5DikO#TSbTQ<T3Ic
zJWWszo@)5%Eu!=1#eo$7K$g?WH9sRJ=vxOjY8vY+p5H|E=VXA2`4%>$P5Oc&0H`CX
z#d0j!VyxLAGh{v#WMHObKvGe&B=-AO%kx9Su`XF)<X$n(OecLD7Bkkm4gMkB8U?!3
zuo$je%TWXHCTXuvHNJprnN75eVIgW35kvUdF1TgLEg+@C#aI-s%zboDo?P5md}&y|
z@A(EZ-MV9;^uI)?;!s0ODW2z`3`;1nm~DI027<3cDaE9Mo8h2G-&lOnHdss(4UoEk
zfkloiiJEXsw-K2QN&(Oz(q4R%A`bdi&GYL<-J*JDOhcbAEQu0yX$1!~;D}w<qJRUK
zNh;PdvL>+?XF)iJi<&JX6-F>tHrO~&@m&-3UrS<2wn#Cj)z6w*ffdy~1EpK&HI3R}
zFR+aKNGT<?(70p=Hf=!x%c$D{)zVWgLt#0hl>`d6AH)PWzfd`LNKMksSg{=thi^3w
z(;X&3ftb`v947)=9ZS9sAMaqP)Z_U9F(Y>(j)kf;2I+62YI$+`n3!x?Ed%;tK|gDf
zVJH`GLE3>kDTR0s3&U2hP_3dlm~U+$KjO}NegnF6h<Zu{&@=sPhbq>r*4)usKMa0O
z`YW|I@YIrl|9}QhM@l5JtCl}+Tj7eF1E!$L!lr}C#!yTLEs`buqhb~^*)&q+ObJNN
z1UAKVu;GU0y20!QbMO!O9io!FSnLn*y+dd`9zm-EFzTRuR*#0`)y#yovo(foTMY}Z
z^bK5=EX3$le5kM~KY^0pH6><X6Vl_=Y<C4Rkl2L-WGI4YxviVsB-wR5lgjc5F)~IF
zndY)phX!u|17@N2ds7^$q1UQmp$gqGaF|UNS7JSrd~O9<byk!$+Z2PAdx<_`5~IqR
zKz$*JI9;=$8zH{hpo(%SH#0yAk`=<;z?RqPlcmeCI8bP%l!a^tw3YOMa+35#6uOU4
zqq?u~`^fR$OP9x}8sibVOdvw%<8-;3UYx}x5@R$<p3TwcRdU>>9J49=Z_2*dtlUCx
z2_R||Pvc98QgYU-eZba|?$fC^)~**HKLv2v@iP>%yh*TBE^<g$J|eCpOxYQQCRHO!
zG-X+WJmi%p?VM8eoKpFmcCvEU&MB_q6n8~pA}@)OHxu|QKsieAOSO=}(f!mS<XcEf
zWGf`Mxr`5t3SHmT)|mDlD|8#vvr0$@3h?1O{4j<kIMfa+c?0JSb6E2tGIsEksCRWB
zHV_ba<&kqq`xUI(Q}KgPZDciIBp@RJq*+avTE#;1zO-2i<UZT2IT7-=C_5n=HeM&g
Lo!B@PM9%*Yx&A!6

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/conversation_template/hymba/index.doctree b/.doctrees/autoapi/lmflow/utils/conversation_template/hymba/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..60b58ee594011c26a7f965b0c06f493e6f554e4f
GIT binary patch
literal 28551
zcmeHQeXJeFRrj;s=k>?49mh%f?0D^x=GyUlFLnb&FOG@x;W{{e35k=Yaop>D_wIYU
z>wE9s?(VhyNK8}IguF0HLAgO~XrL7W8bLt_gwht8R!9g@A%uuPstU9rd<moqwe%18
zhu@i*-JRXLci!InY&Sw!^7ZcQ%$zgloH^&rnVI9~MqYi^TgU&f$)IIB%}X<8t=9Bv
zR?uPNHLu!kSk18WN@xEkI+r_THsPBW1F!8@tqxm<8dck=*L<tlIon}dsCwwsgBjU=
zEU>Df<2B>v_J}=dkDcw5?Flv#I$_;XKCyDWu~_#m&WJv$Uh{(G2PSn;39Uw}Zid#3
zz1&zZJJbid7(Q)XBHa3D`Fux=W3*)g>#$>QV-uBlnDz!XDn~cZhU-q#>ddznOja_(
z(03NvA*e!B*x=Lx8u3`QZUzAg@jTKr8{%oa)@~&)3ug7aZ!J=tHX94gCHmB{C+#VF
zlfBE{VDDy&v$of;X8m@vY5B7kJ^%ddL#K{|6wyCGFqjpH&6@4dGh5EAAl57=)@&}Z
zW;wBDoo3CtG~)*t?9C_Fw-@YnY=oC7v=?l#_h4}LUeNaa_<IZf-ip8bX`rB$@9^Q3
z*>KBr{0_Ul;nmu8>n@`(+t{pVGF`ZZJvtwD%(i`-J)O{N-o6=Ju{R2lOk`?nRKQBd
z5`J-6*_kSmLVzBR2@Z%J&7R{Y1c+frMRofiC^^n4xivHw+I7>fRK0p7Xj#=hh&s&c
zv#n1*`OKlq&m6uS{o5gM4=3Q*#cLxUbBFGjJ3`_Wv>Of6Up@+f0ioGWsTeTXW|Yqz
zwoQ4ffDq~FyrYCaCq8oO-eZ;151e}Bp<|~XD3zkAJCdy~C2)d`hD*NJZgtU-bjBE{
z+$WL8(d4323+>Ly64A8@x}FE?yhvD96MdzFr_>HegDrTq<t{{l=LX-q*j3Jz9?hfN
zCKFId=8|u=Y^5Q4Y_7FD=PkH%=hB2Ujuf3jGwFs$&ppjqP=*%R)Uuj2r@2(Q1pTB!
zee}vhjrcxRcS6fI>k2sgbvZ@1q*3i8)AaMbrzzACN1v^4EqB;xYk3;v>R77wBCh=^
z*#65<<vpi38wC-dXFKf22Gg(|*hcE1VLHu@{o8B|Z{czac5(w@#5UsHy40#WRVVDc
z$Uk2+{bpx{O@PNjG4^B0*f^ST`_q}{phMGNvceAg@c~7HtDVol9Q_d0&B-C~=~Yb6
zo%U$L{#RA@TRmAwe165=ZT|_hGxhTe)J(^7YVa~D{yYGFDx(ji!xLGTRe(;}e_are
z9{m+y^TKL9eG-+R!g}|_hDlF$VC*qK2elWoU9#&H)os+h=919IW_I4x(E41qwOWZy
zTx!(kf|f}h;dslp0?WT(aasB(+C~=hX>Ljqlv6I18TlKTQt5rrC`#x{x=XS3)i?lS
z_esZ0s+RaY7Akl9(7JW1T=iBB1RvJlos3&>6t+TY9?3{eF9*tg%05Hgz^W_VT(g;W
z_;8+Kw<{6$Y!A29<XhjTxZS6v*BzN{E2=MTrkj2*4`fs$tLz>EdDm!|qduNbBWAyp
zelPc)?aWDRUoUd>V|;t{k=WtaBHWmLtw0N7{J99Q`(YAnT2k#~H=XE(abk|$kHixp
z&v7o>zqPtCb|2^5?mg*Zdu+d{8dkT8cExS0{Z;!d`)l*|*V(<rtzu6K&<wto1~O;b
zl3{ayiqkG-BYBys&JtA{Dc5K*N27q5blb#5ZYkY)dMsuv-ID_N8ttX~JnyDA{|QUj
zW(iI|8dy#DM1lTbIHiW@-Cx#rx-)ql|4*KwbPwcJRhH522r9UvR12enc~oqHyC20_
zBN@jR5(}vmY0M+pCXIQVJjIvp29$k>6-giz!cNvzo|wz86M8&ckvHd6kATKT=JKm?
z3vMoN1WGfJ%2{pEm5Miv23VwEZOKmcR8B^jd%}j(EF6v4-?<_>Ar+xe#w>#r>)0?t
zY28r)T7PY!FO|SSL?~?gsZf|k4U?j#)|0%o*=%|tkC7=WDt7qHk@tsKuFn~_a=P4l
z^q_Ghf`%{!oV7V4u43GJ^eB`+?{_B~5o?Ns^Zy?6F&<Nlx#&v;J!C|}gr}sF|H^cd
zp7?C&VkbMJ-0xA7DVM<OoZZx<a{o;<`llEim%aBv_ORX+6hnBEB5+Yi(;i{_3)_Zh
zY<;vyXs~I$ZVHvzMu@^RJZO#{+f3MUUhwzSiZ(sw67@Uf=8k!Sk2y17_g{!SeHi?k
zSHXOoFc)Z#6TsCl-qj0ZcAXMyy5LwBtB6E**ktrrF`HG}^A+Z;rb%c~9<7WIRm3to
zC>%URSQqV`T8EQN@kyZM+?nhs#QaEXi~B9mnX7Juj6tFfD-od(4nSNy9iw2QG6G<~
z70p0<l#MU8>vc{{0l!oUy=n!KB2w9J@v;{Xci^Q-HNx#5iNm{v@IyAfE$4$W4-hgY
z-QPvZvK=pd+39;tM7QUR*fA|IQ5Mi-CT|sS-1AR+Uj1CGX~tylikf&NBWQNw9iO`L
zNZ5%B{1gEeGvdB0Fi2%cPiKad)_ldB<tNdHb`c&0uQJZ^BHHPl<rO6mC7flU9KO!3
z2pPy1fXB?`a2Mo|_^VUqRG&gP6kw<NrzG(P^&Q-ALhzYsHA#;kMxSgcmG>Kk_nT&V
z`yCA{w7nX9i6;t34d0LH-PueInw_+N5z)imEp(0h4;WwCsVckgDrM~LoKDN3ciw6`
zKW+JQM(CT(VA1m%*lU}4#J5iSh+kT@(@~9s26nXqbBWZzLrxH$x!<XVXM)f_XyE7U
zS^TpcV0&)Pc(_Gdb!MGDo;B_^=DntM&<HW*(KP<Z``|6(DhCdnM<jG9*l$c9H6F7*
z)&|PfnbYl7-8#doJIxSq;>WYnS#ftmI-<d|>v9Gfnd{tdiD7<4N?AnJ%oKIY`0nn<
zkH!c~llG~&D=x!(xWe7ejW7lHAug6*2N@N$%YNy-$quQfBR@4pUq$`4K2VSQ<DL!q
zy%lb=6h~|#?k?h-Qj+lNcVi-pS&-%z8)|ktlV7AJQ|=r1hqGh{X9?lZYB#H4yUT?h
znu}1tQv&*HtAK8GgZ`=t+Wo#dIeR$Ef2jrHuEXbE-m(~=`#Y!-&x}>)xEN_{YNboH
zldaw`A<TEQIIHVJHJl>?Me$aNdmO!_HZxX7fp4tPrl7OvRMWdD@THPX5^mRoJ(_a=
z4sfyt)%`LGyXcw{oW!-u2367?`EL?2rrclSP~0!zpO_gp>(!~4Bx(MQM@=G$ObQbE
zk8S|Sp6he%*w&+Id%=B6MYD(a@6bc}2D<2?pp<_JrHK6R(?ik8zl4&K{~g?NWbBq8
zbmCveLO)Ekl^K7B7MWgx?S2=eNL;UO!4_Z!65iYM_;n@{j-fLO@+Y*&yE=XeD0u`T
zN-8j((83s;Kpf1~CXlN|elP-UfP16R=aLr0!1Uozt{iy|#`1Ge!JTPwFF>9`_U<)3
z6y1F9DWw@Ed_fE1y}|_H*kT)I=&ZtoKhok}gb8WYo)6{E*9s+V91B>SbOV(Kg{|pL
z(c%MB&UltDIpnfJ==ZcC9Tl}BNsnF=Bl@p;8t5E^chR(<5&eN4icTp@C`F9u@UYe)
z&`l}{C1pfQ355k|b1Xw<&h6A<pz>RF30Jk8gDbx6KnOf4LcIf8+_f&agp!_f&qhKw
z^sF+o@6<!lNkSiEIj)DIlLP@Jht<Qhx>rVWS_`8fujNT8HDoK`3}og6?r9~X)k=L#
zE>AmLt#}A1*pAp1FG0N)j>Ltjur?dIJC>Nc6*_-h3z!Xg-RsZG)MU!-V{|WTffN}X
zj&V9oetue6-3wmIC~H=LenktDc0(4;c0H_IFzanAJxFXCv|BCD4=VwVt%e_Kdvs2F
zg~H&URmYwd1yv>s+nbwZIE$|+D2lbZ(HY@QB4*0B#N@oG9k*6q{{T%3nw&q?L(wT3
z38jeAd|MAir)VUUl*!qerE*WhlU=j@vlbDR^L<&KHLTg*$>UZ2n(e=}$ZOp_0VPLW
zXDsvhh|V_{m_CDB=3MgR`>DIMxECN#G5fEFq7&uZvGlP2dMLVSDWMdZmdEu_Rv$}_
zKxNhUX)TPR5><uhr^0`0F#m!;X`Nq0ofe^JQ#)F%fs~5wXE4$@CfLK)KdXoG61wOL
zwn<~(hthsY4@D;(63Xl9ShPm|(|TBMWJk2RWfv%<xvWZCVaq-q?LrA7<E9w2{=k9A
zfYW1Qy?;+XARUXoh^DaKDT_BMHo_uU(Q5+lEj>Kl0#(2ph*Cuwv6~uj&zuT5eRM)W
zY<1xk!Bp;weI5Z(xS_clWa4CwZ{ZkJGf?-JZs}t?Fh@&`N6i(MeOpUHRd)LF4H8fT
zEmpoIq#?a%%=ZtzuN{Ea017BMgc{+ytqPUak6slILBhG(B-*LPelQYceas)x!WdYN
z*5G6IAwzd+kspje8?a5UNYHUDh=J+Dp<Frn$@ejz(&Ao#Jeke3{NTF1W++8`l?!?(
zI{V&JU1aCNC$%sJ=JmCi3!l~^Ux3*;`ZO~aenSglVESB>x$tEz?ghw`nG5+d?$53X
zl)qUklp>xG#ht!MHagGQac7lX_G@%@l}-0Nzm|LCEdQ8m;<$AkbA5?#`y*#S*}Vsk
z?lJtNk4im&5*%2i;~o4QhWjCU`T^MCp5UK!7mxCfcKuw({Yf-svwaP&yq~Tns>fsZ
zfKQBY;P+qLhwvCJ%mjbb1F>(Ml8508+wv_Vu)TJ@W-M67QX4KW4%AvToOq)G+(e)>
z3Bf_*q7&LiXj{f2Uc=}(9qQm#121pm%4A$@HqiqrBXx#{M|t!^sV(-fq7O5r`_0fa
ztV<>mWLR^hbLY;vf!8cOBV%Pd+2`KtC@Vh0$XMFz#8vu@E#E_8fd&p}skbQD`g7tS
zF*Zx-F<Bx_X8K4Dhf&&&!w4N`lUv9m-t_qpmyXTU#gt3fV1>2?|3nhOhGM9AN`M+#
zpm*qjey|%TMl~w=m;lLAr?)bqV@}aA+I<+wdWjy6zJa0(b+3B9!o%e1TI+B3spn%W
zPHo?uarj_qpAa~_(38_#^bYMqDUWKS@C<*d9IDW6#=f)o5}j6}k0k?tp5edrw9ohI
z7WG6Tk6x&Z{t#q3j830yXCE)WEI#KnM#>eeDlyx7^l-we*S)>6my)nI>Lh;bDhxDQ
zvY&tFxzA?E_|9{$b3)>aASEvlZ=9TXKUl&$Cn^7UM!re1q6~yZSwvb<Ns+%}J>@Se
z?W$;e0HSd>T0+Q=;GbB?O+oZLZe=+IE@dibxudp-ICEKQrwO{nwCFS)p5%we3X=UI
zX&omR&T6LWRFFcVp-S%FXG&RIRj81t7~U|>-=^9Q2~A07E}l^cdafUWx*w(5;?(*9
z(DW%zQ%to|g1LgP99VMAVowq5f#EDffMyV9Kb~n8;hBquoI%ki+rg7Eo}i!c#Jb})
z45q~glW5T#fpVM4_^F8)zg9iDPbN)>WRDks@>sgb!@P-dTV<GpF`9_tB~hNQP~)EG
zBUKk%ogv6N<O(?H8B&fpri!~NbW+p>Iti=$Y`Bx%nc;a1uLt7=+%)rleW0<=V(g#i
zG}bf!F6vzfKQ-xM%aEAuG{Y_{68WWpkUA7Y(2JLX6p5G6a*C|rfR)^cP93%Si-C|j
z978CK6scatD)+vt>wbk$5;v1l^n152ROv$Qcj6ZOZo(bh8$e2>Mk>V{?n6%P-D|+@
zs>0WWKfADOG~12VGLqvMd-q0U)7BBw4j{h=kbx%aPC7jQpM+o+p2yV-L*%6WXyQ(h
zNSW%zcH{|(=_-X1<J7a|BzgbIA|Sd2l5SL%ISs^xpX}v5nfAxsK2E28alwyg{&BCP
zSAHW>eQ$cxBL5mLUp2kxaZ6M@EpkMyuubw~^u7%>q?!-nN|Q}5DvRn-PXsTp!Vqf=
z5q|%B+&kd<PhP5$7l_h>%&0+EQYN}RsHKd&H=JJg-2ei(2k7TU`nipMZi=7Yho=>H
zivRB7zq|SG&HQ%{|262BmTbxZ_|0-TfWhvRi+f@Br4vpWWh~mnHOzi)kq?}2DSfpo
zxh`MyYF9R27HNsHyHQ!dUaOnU_k&5jjceJx0XVrg;wLQ~YB~g0-eSHd?usj^2jQEl
z*Sy?&>AO&e+3_dF{?WLTybD6*elB$7;PXQW;qYPS$7K=(yW%o;fehmMae=suJt~4S
zL|_?|;ZR=?aTfke^H8~2{2;XE-ICPcQ+Mb@mo+O=E>Q<s<uY7Dlq0QlU9MX`PS|qY
z@*F*J-NK))gl<WcBbyX@A=bZ&36h_TNDX%xq*UT6q`^qKE!#(;+@;ezktg*ZXFZ*c
zvYt-!i=!C6j9f%*3S>&n8b{<C<xVi4Tu27m+qH``q1p%84t`4%30Nr?D?EcGr1=t=
zP;s6uL^i*Z>sTqOW7Y85CNFx3ZK<|>%ClODv#M^2N<?;5HZH9c(#>w8Jghu_?4r-(
z0W~D_dxFmUC40hb)N6;WcG#J>Cz1AU$wJ0desmHqkP!=Ufkq1n@izLN7XnqR94~d)
z&@A%r2^469`aCL><ZIin<83qDtEWt@^c8pOUuN4E(E)O6SW5AWY%^uFg*Q~WL|0~!
z#%~J_0Xi+*EapyBfUc%QiG`RI1|C!>{8!<5954((k_}{jvU#@2xA@t`x^=+<A=k5#
zZQ`7xd>vuitV<5)ha?~EASoTGSqtqY^tXkQ2vQP9^lO<7luUpTRf`5g_&H5btHSTb
zk}tp}+Tr5#ozpxU!LIbjisk#BU$GH_tXoL%I!dTkkx_6`&|K+dg!j^6yE9)hUATXY
zNy)Z|eiIy`*1OUlGL6|*8bbg=*&tD!Cz#t@H=78xF40|XN(`|cnTOeRK4PVQ4G}0x
zXbCWssCJ%>VlFKr=>jO?)FM$pQiUy)`VQf(sEDmceg_;dg^yc_g%O111=!QHd|Ya#
z4kvjD7jxo%Qq&5pcFn6G5gux7MUqD^uqtXxE+thQUUdS8&iSIRO3eu<V=7%r7g&x(
zlLQlP6NS^@yoqb6kQ!O7vgp(?IArAANlyTR5CET*$Qpv)YGC<Bq-X}K#Oui+wv8Xj
zs9VVKXG8iMPS`oM^;Wi}Wwk1llM?e2o<49wKF}7V9k|n^U^m)f*b3%mXHgu?w3czv
zy*cCgOPEX7Q%#NlYR1W#DgU3<n!e@sJ3!AZ{$e%LJ(VQge<13fR6B|6dCQ-1tZ<RT
zfGU`^!$uvzCXkvCvxui~9cQ~ClZ{F|IpYDyOf)NOYp~o5%}W*AS+eW+3;7+U1cx2A
zi60LZfzJShx=uMRTtQ6Kl%(BVwGqs(g7IR$fy$hOI9z-liD_4OqHCVLTB?B+NRL;g
zSYE6bwv(a=+6$R#PJ?8(?wOQ#l@KFk1d(YhST)SxWnf@iNaDG|uCJoj`3eHxPQ8MZ
z+@x{Ar5D*YnN%CHT3-~}Y=v#GninWlH5(V&1k)Fih^2*txe+2aFQo?Ma<dIcL9#+9
zTIh1br$~>p?Wt0_md%UN+hY7eIVLFYHL8zMp?EI4zetAnW%}8KHHJGuKVPArH|b}E
zKD<Cbqx7=@KW#Qa?fDXg(rNP^R(R_b-g1SvTH!6=`|qVX1OSHg2%>7EGM`IVs4&jH
zcqAvt?*uU6ofpZ4*)V~=M8sk^$}0)~g)p~|3bZNLqd=Sda-PVAC)jamuZtgIa`(_`
zbpjTE=Rc^(*sLzjH~SzLBtz!FMFIsJSDz*=i=7~{6VRpxRG2K9iPqYEw@W<X9W>NF
z!d_GEHndRv0zS?zdb&4Z^d=jLt=2GK;j*z0AA5BE!TApf*|s_-_bGk^(SDK;>RO}U
z0`gOjF74j9x6{v7RE#3hT@llVBqh0$g_RF!-wg?`49_=OwhtX<#G0WnnXhc82@h`X
zn+MZb9uP&kd2oPAyLjLpKx>)3z7O(KXD;ny@j1$aMJ_EMTv}kX*=AMcawV`CiHwCu
z2vUm@vaC@(p>5O>G6%dAD!Q?NeCesGJ2HiF?TJ*ii3lq1X?v=FKA$=;dDmzAl3(DZ
z!tPhSHd%GJgh-QHhkk@6g!?&mpDcd`O3D2OMtyQ5c)s%}Y#Gio!YW3)FEGDly?Ea6
zcoSX#dWKk1d<O=JfW#BJ=l4Wt7W)>OYDQ?H)j|{=iv?u^laqDifkz*G7~_ln&jUCf
z;y*n&`2qb@B^}|CkY05%{A}N3_;|l$aGyns92y)@XdrjVv^vRBm<|sxaNv&_-2NLL
z)^y}Tkb2_&Bj50vDV>JPP9ZwO2@NyASGWa8yf0MQf>kxK!-H?P21ek~VFN#mMQG`&
zL3`)9&#3Sck$k;xUc6S67ekRq4nH`5{NdPm98lPza6|6MXu0td_s5%tu(vpO0lON+
zM%FYIu#z~BFHQUh7VTuVwy~C?`5M!C&HW_w`%>TZ`@^F2TQBLC(*4teQ?I9jlA*Vr
z-;>e4-S=q!s_1AV1=efk&h{xQcil@TA`E^aVo*6*Qy&S{zS}p|{-Y?>a%JSu;4<=m
zUf-2U{rb*hXrcPfJxy?=@7&)9_adXa241X5`sAEmQy+;0-_keL_7$aCt^^;>m*D@m
z*jQ!V(>E{fF3O8s30?y))+BvbR)Wv;O|>VBQY}}4KbSAU(eVHgsX%Z??`)rYB=R{O
zWV@u(pdmj+N(&F+hARFhoiL4^(mp-Ug{uLmr_*_%Z)OBVnUO2Pg_v=Y&J?iiaVS__
l8nTUXF^|G+1Uu?&WFM`2OBA@pnJ`4FWnh;=Gwp?${{>7x>@olV

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/conversation_template/index.doctree b/.doctrees/autoapi/lmflow/utils/conversation_template/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..0536936b2ca5200876f49888c134d6a8f5cb2e30
GIT binary patch
literal 171730
zcmeHw3!EHPnQsP?NiyU`0wEFzo$$yc<Uzti7?MB;;h6~`2_VGS>6xxfccrJh>Fyy3
z5RHm1*{!PxEwWxj@PP=fitDnXh%URbK3FbSQC8*pU{Mi8QCIJ->+1c!^Qbzf>h#yA
zsyZ|I?cV%;>8Y-+bH49<kN@|bXYHTx?yhlN<M5w1y<RO8%X^mP^7(Ql@6?;#q<m#?
zWY{S;ns+o8y}7xs+3QWN<#yLABeg-N>5YRY1`EYfzUGvh+ne4re7sRC)t8C#6YI`k
zqgW~X#S0S(-GzzUo4tj}-h@W6QF4@@yalD<ou$g|WxS5TO8Ew-R?lGt*@iP*E#(?c
z6H7zk@mrle$dy%@w>3HM-PIg$+GrMz^d@KhrWFqNx<!jNdgDvQveVo+GLZ9na*al<
zI55%xNnpzlFXrpm%87%eT)hrA;d(+jH_Wdl<wvT)mx0{iu9~wGpBeEcHgZGwQ?oF=
zFt;$Ja9rW=!tvgY<%P<yv%EG^E<3g5yDPO_%Qy621kIO!fWi85&e-zYNTZUg7MF7Y
zEoTB+ZWPe+Vma^ZSyrn<Gv{4?*vLR(oHv2pQzXrmP&fgaRhSR*JQ4n%4F4Cx|3%nD
zkW#J48rJKLujY!irnh*wk{>BK=VmOF<SpV4FSU5K(7mzI%#9RIEi4VBv#~G_bWk{g
z8(m;EGvp1gXUoU{CsOA&F<ZMOR1P(+<?`k==32HH89?={;g1)Vfb=E-3zNabE5p^w
zNI4%B;5g82Ww23mR5NZP+`?S%w7BZjHL}q=Ge!jixw^xRqWSjbM(^wxMF$JH#!zWk
zDgD}*rH4yO;d5dX&O1)E?l`-Y;%CJuKI9A!=Mt7J>>VCZ%f?~5SZ+AAvL>=PY?n$o
zjqNjHG#Rj4X>c41xqi1(R*Ie;gXmtBW*p<V(Ww^p)_kgtH>Fq}ERE!yoy8Kzox(sN
zU)WhF6x>3oP%cypwL+tCLt&5OO&ZSa$vf3Xp?P^1pmDCsLJVN{P^mJ|Jm7T~iupW5
zqyyemQK^4pQh8)}z^OU;=H=tO9u|P=m2zn>Jlg|-Er30QgRO<*84UCQEL;f@LBkGj
zy72s^yKA{>)u}aeDuiPBD|9#kKy&;myq<c1UpWcDp}PW(u`c96L=GRCD0EyK#4Z*w
z0UecF0Xn`X9y&G^bUcp5mKlxQz(^@q%MMmb*?QF(jG|=AFf*P#@6CAq>uy@H@1~Xe
z<liQ?_`}d*LC(AWV8_b7Gf(SVg?3*b0pO_ZT?4@r^ySSKDG?+a%aos+-jT|;EQml$
zWL{T*jH(Czt?%Eo_1f%K__JZ{*7d=ijaOv1Y+AqW(zP41Td%l${l+cYGwUu{yY=D?
z{ijE6_QT^!T^FujziG?*^_OcmFJ9l@zgD|*$+i6#=yxvNxOM&JjT`z`YR_%hu(p4#
z;g-$^dn&k<z37U~(1t5EVB<Dja%u1j-njCb^&9mX^*dVWYcE}qJyYZ8y7ikbxfZHd
z+PrD=`Yr3X`nR_9^vH~GLRbqT4o~*F8$-3qNHt=Zn1oIQvyExBkPno07xM@|J!q)^
z0EYT^;0s@pxk;M@022yX!F6f{m=X_E@_UVp9$u^LHvTXHK`UAY+BO-|-P{nMQ$Z={
z_4ifx_EiSlz8$)CW>&RN)|Tl-*6esalYTGcm{Y1wIbSRfW%oedtJ2%hNqSP*6H7%1
zw7HT(sPG#BbklSun*rGU-_~FUX)a^jJFL35>2+84E(H-aoiL>T7Igiqa41k8^17k<
zkhwPtS?DU^<#4W8Zn}p7B6t(wMq_UkdQOL9?cNda-Puzu6$gur=B@1K-MJbDuF0Sy
zp55B@aJ)%Sfb~lt#Y4J{Ty4l{G`$tEmCC^Pm(W2O6VSc93nctA@TS>??m%%rQxz8k
zQ5iO6zXhG?r-kF)IUsCE-?6G6V@XP3Sms4|@P|P2M<EY>IrP0%AmeoBy1hWxk`~>F
z?*l{614BR18R%`zp?7`A&02?;7>?e$lEC#}l{|XY(Pn|8!S5sk?grF5jt|WCG8_b2
zHz0rocOw>KopMtrvDfW_-`nGYds!Rpw}OV(VfEI~;Ld`_-J|g1X!zLZ&c@>FtlgZD
z2%7PjI=*`mv?b*G?4GI}+H=tTU8se-zq<}E?GV5{pA%0D0yna1^v;4BAq3zEa&&7F
z;J_B&2_tEwmE%WBnLK@S=ZS=U`5x39_vIraJSon5uh4Me_*Bk&cN?v>a^44_Tw3Qn
z!@_x^A{cvby^G0Ai_|XOi|4J6VF?zjT^NeC_SV}u@wDFh6sxAaw{mhF_tq;kM^gHC
z*1oEgF?s67&JqZF>8q$Q?xoYwL=|onAN{q^Yf2yeMH^kU^3gv)xwJmI($Ys&0gSze
zzRV=1dnn$E=b_UM^F8z@qOe7l>%Pc|r}fZzK;52P$2~NNr;*_nD<6%NGI?nbPa}^f
zB@*`33s8w7o<@i$zIq`(mp&Ojzm3*f`RYa}m)2Kp;%QU_WACl2n9Ou<#e1qcX#2+O
zcL$c>@$?)~*v5U%;>6Q>Yn4^gKAv)NbZZi1zro;^J`QT-MS2mFp9W)D`m>3u4|`}6
zwZ%PjnqIx)o_7kBrVORGw^36o_q-R%rFBo6A*oRN$KE&J!33uJCf<wZo1elG+&BM1
z6t?lrTRHKxzWD^JroC@+atF~j^%)5(=h2JMzNtT(sQR#P{us3-d{eJp@y(ZnN>lpg
zi*3}@$~S)v<wnCdL+u}X-~0g+nC_c+FP?9XKiv1t|04?9_~v&w@wC2q6i~M(*Kyyx
z%I_vd=LW4DC|b<qqgxLKnXs>(jf&*HIy1_a;;()9VER<|tTw7^<*ye)xwQUzrbV~Z
zAcL{@*_BLay3gW0RUfpKW3GEOmf-PpxhQPoL`yjFv_89&RntD6a&mNQ66CtU*imTa
z-pXM^WlY{0j2(p@PDUW?r+<zb<9<3lL_l#;Pv|veP`$N{u39<iolq{VliG|Og#|G7
z9(oItnC_u?FP?`!fF-zxzF8Eu@z6JN;%PnfGpw5S9?HobY!40A#@IVhP{zVTgNKt5
z2z%)Bs4>+;0|JVNeoyE%rH6i}jjmdG=u1#;)I3xXz}S1}x0u9q55;@&JoNWif_vyU
zL}42beTEZH>!Cg4?ebnuuHzn>^Al1Az=3-Gm1#*l$uDg3*&PR)WY~+Bfry0{&uP_4
z#hp*X$J1xUOWG*5l{@!AxwP(lu0>YdQW9hD*S$=1x?kfxRV%b7U~aqtOK`tFSroSM
z>jj*6TEBihtEPSA<>ctrB*={~_nk~zlWOJ6S_zW}U)H(G!@jy7HN_+DRE-bCLEj>@
znKItKv5kgWIj9Ha(mLpA77nU6f9$>TMkX)aEAd`Due=9KaIf4g3fp*PgA-5dl^<u-
zwD(F*uH#-A%+Bb`z^&X;FJkh`V0K1-Hc|Cq&-^-Si+koYy?VtppA{-i>6*{9QBy0|
z{2r7`>zX#RGokj6y>C9n1g85Y-izm(uVM-An_m%yZG7`dPCTt|j+<bY33GBC_sy&y
z)O79oxoiS2^2?gM^$iD?VAy|80`Us}J*Gtq6-Pb=pH80yFKDCFR*rlcluPT#eHJmc
zl_196qvtWn=^l;uRHe{9fI095Sb}@>Tv6D@qvvqqX+3%~tEPRl<>ctrB*=k-Xshq6
zvT|g-h{=D0XsbV)sQR$C-hkQ?(N?csane0Pr75Frqm7zcIq6%VTv{i!iMFBkkG*eJ
znZR`4#C!34vxz0RZw`yXHooa{;%R;JL#&$izRAgT+&6={o6sIdEAI)FF?nY&cN2Oz
z8G*2Wei=0;MutNK6bJpP&}&KueX@<NS~=*mP%f>5+RWXA1u*s=`UI1h?xA=uo`?Pz
zOK=bUf+%d`p^tImX+89JteW;7%E@)yL$CDH)9{vKD-Q}6GWqDHPLl|G>9O6um(B<?
zq`2vE_*nX6cMeoT*+(lkJq608b<@)=lHG^|#@<hlVj|Q16z{3ppe-A-+_SL+_tPUq
zVM-6!Y}oPYPUFPW`spRCn)ZIm$<eJ5SuXw1cVKfY?6+_IkT)JT^*7=0BWURt;OWQ0
z;_!}Ej&I16D&--5ctGR`kIzEsN5ZAm-F)8UJ;SBGdNqgVVNI&moVruH!C`x-KMiH&
zc^RK#$4LY%YlllI$A2hCPUX~H&Eexe@azrZ_>YO}#1T~@M*H9WE!FOy2}?1@X0?eZ
zMK~GBy~QuVP9Ql0Sh!zqy=`;kLE)2yhp`$s(?h9dG<KMuV>;HO@Hdig?;Jd)Vx4l*
z*j91SSgZRNRcc!+m6AMn;&ZYlA%QRI-+L3C3L%e`aPQ|X`X#+m{}7q*{&&;K60(l(
z>UH$S*;{_CAE-={vWkbMf~J`Gx`Rb392=iPYvvj3Q6}b;`1rJM_^Gy-ew8V+HBh#S
zkN*ti(#FTvTEs`A6wqr&X-|06Rq+G|_emx;Jx1cacro&2EWu;s-;2T&tYfn=Mt+eK
zPa7kD&#GyydYxPlCc+Mm&q0iIKhN2UaXiEj=~n{O5__^JZ*u6JqeD3C-t#8<?mgFJ
z*n(rl%QILaeHMHiR6|)-D=%LR<<ffjnuK01g)#QdK8Hz9cXqs|Du#9sOo&%u3GVDi
zi^7yjr1#C_#M3(a23AdbXXoVT)&$N@KL<In2=uL=3#Y~I2Mws?R^g^84R+secv?*p
z-5cQn_a=M_PzsV?CDU{7f!pB}*gr=#$bXev3;9u`SXVgt#mtjPerIfg!&h@dP6qtG
z0f%Gx=YvLn2-4wOg`L=^!*BCTuylAeFttq2v0CS8+WD&8bvQN)&X4m}>jzh5dW+=@
zKLpC+%+*DS&U(@gUv+~f@k9rX^%@W707sAX!vnvj;V;>Gaj2YYjMUTvP`&U8T$pFE
zGs0LAJYE=2!<9!07bbWoS^yVE_}O;P28#WZuO?&i<{beoSPK0%#*#M`Ib`Q(vG3aX
zUz&ltpTv6Sx*zkaXI5|?Siv}NGHYk3k?zOwQJ!&z1r)IR4}N`Sb|)dboYzT6r5&43
zv|`hnr5tT|L($ni2uD6Qz3K8Yn=20%Dm6t*ZTfiHv16xY#o4?zCl=!gwQ!U&oLW}Q
zH<_4}APE5AJ{xWXmrVe5?g!xs1|x7xGf1cj2Wf(w;E**qe_B7q&FdEDs<AVayb1i!
zv_iKxY3E3(#Kg&2=*hwXpIJCU4*-Opw}c9OdlQb1#vg(wk;4a(Lw;K5d0-VE!LN`@
zFl(5?tYMIYk-a>z+FxNkL3I+Gf`(&)*vrqR8lTTKa^beh<LCYa6?qdu9_;uzcibfS
z%Irt|Ual<V2Sp$7CYCGma+<uolpi#Bz?&j|l;4kpCq(n={zKEt6?soShczL+y3=R?
z<TVdq6F)3O1<hpeBaZP&hNbR(*v>23o&6OXSvzvP7LB#;{lK~&KYHDVlzSFOnlYSp
zA65$;L}Wb?j7ZJP^o*z1{XFpDJ`Nx4QS=1ADu$wKM+V%#S6boD$KN@0#yQw+g<q`^
zOx>rE!~aEdC;-&J;WrfyM+qBI4rXi&_={lUu~3L(qx)lQ@J}KQ{u$n8N6P&it_t1m
zD|iEmxxes#!mn>{y06NgLa?_M9QZhRgu|ZuYj{kz=9bxz0(7#1xIpKQM_%869~1B+
ziyz(o)kL^D;7($nli6nv`#geuPGO%@@$(=7(gl!-`mZvO1B{to7J4#}<h9R!AiGOE
zgp(F`%)?_cr<=~ATdrKLG}xOOlmO?g+_vh(1{|H-ms!Z#x^T^s%qsRgL}xfG8V1e#
zGX5i(g=^M8($8|m*)lK(nbD1)AQf)Z6qMJclJ3gLd=ajKN^S{NqGOiVwL5Z(xBC_>
zGS@}&)TD!+9q!?U8a_FOYwG~lMhl()(kE@kz8L6iXoZH(Hy|mF&SqN)<o}U2T*W|S
z16N~!$k=u>A_x3M5V-<{k{~h?QxF-yN<_x%5F)PwtGym>w?yRLG)dOx5t!-9?yS^?
zVb02_WqR9#*fJvw`cqzjaTg`5WSP!PQ-{1S3n9p;(mvIi7OoFLNl!_>hm~@p4?i84
z`wv2tb-3w`lYl($SH+-YJwVA`xFut(HyH-;xGDsMj`NQ4p|(0w9&C(6@VTqc2gyDK
zB!A<RGC_yeRgHq=udv8m_Z9d9jJ;K5tOXJuHkn(dREjPd_nlBhco<4TF{e`S4;~iD
z*KAb=fevPngwzo%^}uf4r}9tbbuls0dbo<b**V^Lh<tHpaZo?Jn|CsU!?|7Ir!c1?
zrKNY~RDO&hC29vj1@3c9YS<^VOZ6B#EZg%=G&yJ3uLAJ#%t-Hrec{hU#<&8s5W=VR
zAcAiL!!7a2CS#NpNHYqAp>&NW`xvA5k9pd-7FgqH;{|xd(nfZbG;L&f(}8~MHSn6L
zV!e?oH#&{8mnX$p14pUBc9yZyDB@m=j?59)SR`=yZTKWHl@lY&xW=!F!S1E)VAuV%
zhJ9v$E%}w20M`AZ3Ht=X7NDpl!rn+zp%9i+u|?QhNkS^3sR--NgJQvq*Fiy8-VPlT
zRDO&I8&!Ewfx8))j+4462-|B!Sd;O$hgqWoNUVJTcno3frGT|tg_^*8lJ(AI1u3xh
zGr$AK+DGw<VJ*8FYplI`C#FdFroj%Uw3j7?+9*S*k#?r3&M446gNDL^)>I<E*{=(W
zk_MsYaIw<cr~RrJ&i03KcGeZ_Enm6PHb2i5V+B@-`!9a=u@U$CP}CA}e??QF5SLT2
zMchA;gj9qZZN!Zstyw)(lqP!sf{Gq%Z}vMgH!Gc)`7{+O*2ZYda+*pa&5z$$fme|R
z-nN>gVX1WS<{|YJL7Yy~P7D`P5Fx`$mW&~`owswA5QBQHI+L#_MMIj>>!2t`L@jMw
z?VHkSMFB2Ng)${U#inDtho(ZA5~l)hzX_WT{-7qrsnr!!;C=<LC}Bpl(3lY!f>is2
z$P}n#K6EFlSf8ZrhkWR30z?z%L$U%H{sN&`_dV|i)_6Yj5xipg5W5=deCXQB0Ip=`
zFJJC(O4Oe;B{K4qIx9N5MYYky=y7x-o*1<#6rlAN@!8JwK_Bz0V!08#_*ftMOPL$}
zZ@>E3xzRsCQOn%u`!p2_);Ja0+~^gOkY$)WD811b(&{>bvK}&^2`YL(Gv`LrX)07e
zi_w<jX)1{{9}aM*k~D0Uz5z@}8{|!w8gUuvQ9)(Jm>RWn_PD9hrKDs?OIi;_EmNaw
zXeyK?2`V<Jkwa6VEQwPwq(=CIiqomp6;$B<E!b1S)M&9WH8Mx4_SsR3XeAS*x1yr;
z1ZfK-NRJ96O`IUf3Q{CUcK~ZVL2BX^OOV*rSSLtZ>wD`BXSmbDMVBPajl%4tPK#zl
zYm8<@525$)j41kikPQ75KGvCJ=%apBEE&4C!^zO&e)X}Fq0d55%Vg*&nhFJEoQiES
z^c|9r8jMDJpeTm4enB>Z3}J$bO)~TcnhF)dVzi}a3grnFZ82y*oD9t&Y1k^g!^ser
zQ4k!h#);ZFd)#E`EK)M0^jAVr%Vg+6nhIq}f{INtw3((tVF;&UNQUqS6{l0HE2zNz
z5cpTZWawmLG8BnV?em~$jFKr(71gV!K-WSFbiV+F#3_)hAVmtaA6Vll(3|m!r9kXz
ztW%&(FfKIcWVs$XvjFVEr0I{!O6t^Sie6hZ?YRs6gr`0Fvq8%9UVNl8DbJmLRV?Mn
zhEtyL18^V#U-cCl$2KoNc*w6Wc8>EAC~BGGJW5lc@QG6~MLO2fK*;|{(y)zmk+of1
zph;&$IQIeaSw*^bI-YP*u2gq~l5N7nk4Vvwq3(xV$$HXco)Y;DO@#_|f{IOBx~5XT
zWYHD~bwYrF?I3;CK~pO%sK9*})SWQa%{Rt64UpQ$IUST_j5`6u7>aROh;eraK_rfG
zvVs&bZaJ{VW87(Y#bO-08tWLhr2*Jl?{qK!;-q0N#7*i*H$75eG}v8?-p7Mo<nh3r
zF2$!h6YtjhRk3*IgyUVx`Ht)S>SKq$tD&f6_$$y<sPM<B7{Z^j<~R`i4J4t&F;JnE
zG!Vv+)}ND&AQ4_r(W8}lzT+;M3KgwlwB>%9N+QjNyZE0bY1k^g!(DtXqaX}gB|hz(
zJ#H8O>!f5zOL`iLVi2@g3i1P*3S~)xicQ}03z`aLNt}wIi;q93IGtKuK?UylP<O(#
z=Okm=6OK;pGoMI=lF84MX_4f|f#l}`A&$h!kE|d?@^c)p#*?4p@rors>}sr&pZ>Z7
z=O5%6l@4w1TAMWO@%c!d@Jv>!il#befc!<OqdXYI+OzQY&LlZ2{i;}!vn!nB%=H(%
z$_>NoD#KM+5m;`-S@?RfPbPM<vkr<{COcQrR4Ab0R1C?^k&$*QOYsItg0}H5((yOY
zOq%?}klt>xF=WURRP>N#PJZ4-Q=vjujJDiCQ%PhA;pFE&l7_AFJDmJ*83n=6D*0&#
z0&$a{FOZTUE$MSm)H3<`Cz=XnNrH+^^7CDq3S~*0iXr*IA5{EKt*)Q~cPZ4JF!`Bh
zOn!n$)jsV}qLoZ|{vG%UjrQz<glD-BLE?l*R*)j$nLJ%4JUw{D5*~In)(Ov+nloIv
z!O6nu$&D<?zcZ5>iAQ@v+@wx^rbjA_ra<#SBXYDS{CJQEorq6$CJ{QpuZkr?*M}3K
zxU(B)`1Qrkcus?&mKo0lG!+V+I2A+2!&@2%c@s&)HqJ$ooOH7rocjRxtm0fd9mk&C
zD3hWg1D(s2tfx%o*^QfNDpa5oRBQs>pVL$*Q{q%ivm5w}>Y%9=7F6IK3w0-qbmne;
z7-!nYy9koWsP_Qy7K(b;L)1G?h$C^-lNF?hdY=IvcsKu1ykb$0U5$CvV`o?7N?AB-
zpw_XoToOmU5I5<gUZlck)O!Z)L(Fl69}fWe>%xkp5v1Ag>Q}|0-tKVJ>rnUqUwksL
zgW>m~sAVwx6-|XgFiyn~3=LiJKoNf;2_}w=2AWB`|1qRDYlaThx>b>}Oi<BdnYsHv
zpQb{^vKVbyPE$!_3E}SlYLbSn@;lu9=Q0X{q*Yq<v~R{{5Qy9T-$qJ?w4`gGC<aH1
zxsM`Eg|Z|;#U?A-O;e#PiBmCj|M7<uNl{c?fxg|ZfG#_l6otc(MN$+lB9o$bqjL4d
zjk_T!`kFAm#7U8?K*qyB9BBUHTw?n0r4IvZJSn;#uUJyVt`4E3s8X$V>VUfrBt?vy
zP9{a7!e~<Td32XZQp6t*V(}O7sm=@qe9o_m;o-i6loUPVlZl-aJq1NAlcMj@R44%B
zRN^K@KPL&=#>L2RK?jqf7}6VeBxO}(C=*olP-aewrqNWWP!^*t$I(<0Swc7|T1?Wg
zRepz)A}%96DyXa&H^8<7fw)Q0C8T6XOS%w>S|&wT(^M!+5>#xGqC8E7vLsH$kQCt$
zDt@O{S5Sfb2td+7N{V<Gij@@cA~Gp@3o2JniuOTL^cjJoiIXB(L5ig4c3_PsMR(v8
zON!XlSSLkS!j9K!Dc5k=>n}PpJF-6Mq7aRj)CtklP<7FC=s|Q8o(_c`4w9jd;WM2{
zh92;%V#&~3!^zN8e}PE7QH!%A<nw;TvD2SNp{Qm0^EH|Z1y!7iA^kZVTB<Ap`8G*H
z4K%Sfw=yb8$9-Z*=oONEGB61$HYv`3&{U|v6r(MZXHs@=(H4WY!zs=&Bn?}2cR0o2
zG73VT)woYP*N&UwoJmTCl>Q1RYMJ7!qp47qB&gV=I9JkCDBR#w3@Hx&prUeWbp;i;
z9|ZqOnBpujF8k2qQ2PWY6pLhfQ$fY*>CIaqz4@?!gT(2LtRO{t^Cn=8r#JiYilsN~
zYOK?nP1$U<0xLg?<zge7?Ol{S!5MZMg-RaYXplI&(O5~H-Avc2>s=%^5~{E9@TO|d
z$P*AlcY{>tFGEhHPg}$~{Po>_RV>vxz+Rg+lfAtF*B^s0omvS_LRVk#a+qjD%R}Pz
z&gDmt#9TK@$sPTdhInSj*TrvssYt*-&)pOtx(!;G42aw>Kv4^z(HRs#<C&-QwP!(9
zAt*Wky0~3vJuwu?3P>pG=B{Wu`2RU<HhW*xKjRg{A$B#^ICQmB#<wwM2bI@Gbp(wP
zpZjlMC^aTMYpArhQQ75`i#Iy8zD%Q*E7xJ;>#$R+FWXdew$^gRvXkE`pIDL^cItI_
z(QCagvjNay+l9r!#y038EXiPxXgmC2M>O?iw4PFCbp{XCU6Ns&OOzj39jv7P$l}Gj
zcIRqC^+lPbYcgA$>qmfPXWQ11YRTEg9xuXb{3V<^UT&BDFb!B|85%0#)%aTgvcD2O
zB27F6$iCuN#X!~rw|f9?`7e;U^UxGMk5vB4zWjKg#)4C0u<qB#UiFIaujQ}uRNh8A
z*{@D&Nx2i~=7?E(R;la8>ncZIlQS8M%ylQgpGeYeKC9tak_h>#2tmar>F%YeP)WC-
z(gQX4rqtGTxm=?!@M&K#5Pu)Zvu$@UGTqUhg``6RX92_E*O0rj;b!DLN1B-gA)fAA
z9***u>;4IR8f<GkeDn1&j|NKa60WAgNwSD^f#61<f;S+^mZM!{+pdG6mM&1DsZjk+
zLB+-eZltMDE+D9A@;k~Dk+`5uLml|EGP327%p|t$@r+DT`@PkqvTf2ny0$6&-%IjO
zq86uOk`ab?g~rE88p&nE$c%!Fo*?;8CZo^*=`$pW#3D*HKx(w2CB79cQ8fG_$-aez
ztlk~)Ynlq>_$_TUq=#`wQ8;-B>4B%OFoM>#UD3l)B>Ogc&<Co`R2HPAq6bQg=%W`<
z)u4|Z5RvsYvWo%zhRn^ZV<9>Z9P9O<+RdMq#xB1G0rh-Z22`@J3W{0=)QvP1DxeA~
zHUahZG!@FP1Qjiy_5>2<qBVqwklU3>4pe=Q-u#fu)np^Uw~*|Uqxyo1?sn!SFt^cE
zC}W7xmV0O_lreBB25pDuhdxHquvK@5=Z84=I?|BP6vM-wBpXZ`2&V$Q>LdMk#8zOD
zBOdjjcC5QrORR#c0H(f0685?((Vp+KSY)mnqjf(biP*Gmve#WKmtf3DZBj6|W|%1Y
zElJaZA%mT{ky0ajL#{O9=#BJd)kmt;N)7gn4#O^*>qjaLr`}|5l>d#|cHI<ItF#U}
z^Tfy%>tf)5Rr|;@ffbR;6yE}cPCuGTP)K_{0*YG3o#SaL6x0YRHn4drO@)FQK}Cx@
zGs04C8GU%~DiG>wk`Yz&vBFQfPTV;rb(uSHSCagbX&<Lz!0b?j&XOb&izrnG#V8?j
z%33Gcw~&xk2j$H)70RHEZ87Ho@1&_v&MBzaIOqFlD(%~1u&Hn!@DNGE)~Gt12XGnb
z;BSMgHbVUp*<dnK3dH*+L@0l>utm`S22JIMP(@@QQ$&41#ilJkq^VE_BB=bU+7>d}
z|B|LP4t~~J)IP*>7IZDE24@}OC&`&79&KDrL5D3~EF(Ey1`%h{Y%+{fDjE+(A&hIG
zubWSPNm}x2KyNlpkLs>+dWj@uQ^<SGFyv{)EM|XMf6fTL!Z0h#$;5hq=%g6r#8X&K
zTyq*DwQ^m}9Z!xi1!w@ONwT=2ve%L#RBb0#k`RzeI+A%hHwe88X-*U_C)+@VK~BXW
zs|jq9RT0#+B$32|5_CpEO2cFWl1V9?a^6JJNGzmuDQ65zdMC+$G8r8X6Z484y^kc3
zSVW9UN3F=5az0G5Zy}-37GLwo`xjDEv5EEnL{p(+eM{U3d&GB18j00CY98@Rl79<*
z8?ZIx5&ub&NGzh!@`xjkA(I}e<6=Z6v)`RCDpZ!HrDBsC;ZnR6sPnmQH2oQi<&eF`
zb<n0oQ|3jHK`SDQB42{l5D6DW_Jc0kFN!p--;zrs%`b5@G%UJ6(gW?hELLpc@Y=+n
zgI4IqT}Zu!m;D~wNBhM>xJ<#RIhlH)GE&NC2As^$2rQp1HykGqtKINMv7D*qinaQZ
z%<f{NkZBa0%ue{)kW1;{nPPc}-5-JXKl`_HWvBrjt~;<MG4J0Fo)#|w601g*^<0>1
z<TB2l9BdbJ`g(Tk*x}YI<(`|w;Mr{N-b0;bC1rSh$SVtzy(6l%%0S5(hWBF+sy&3m
zcI=SdCA<u}N;^Y!1dGMI!;ATqFqRA_CTLhYq2}x?=9~EQ3buSUk5$ZdksGiG2mbg5
z(B)HwtAOZklIR&U(bJ<uVZ^*!=rJG?mQHI30^lk9v~WCIe)vvk$+dz#SndrH9Vk@>
zcV*eQxhm1;7O1OWv&H;^zRb!cJqx(O!56$LmUmWGEP%Ugm>X_xV^`S~S?tO!*bcwQ
z^-TC<PX_+o#6IzAL9J49uqHHls0HugKcw7%^Oo_aLU94RzmNZHD4eU;p?3-1)}n?>
zoKmSWU)0jWqm1grTMj})r7iAyVgKDBF}|?>eN2$>3l}B3!M`za!uKUT@STa2{kTnh
z6JmvXAXvC3Mav&j?C-9Y_7}Ex)imw^(`Z6TF!DR#k8k8TE_9Q&!kPjtg&q!TM=c>W
zvrnMYWN$i5^x!6tVx`Q-41(369DTuHj|x4LEyB(n)pF<G)Dz}ar3x11(;E*P9JI1s
z94T=n#!ZSq|2B?5-M_)7xx2jwWco0ZDRABsaIWwx8Wu#A6KuE#4wr)opa(Ja!!>#l
zTr<-U9wdJ%6yZu~&sXEL$H$JlpfUf*8WW(X!JynP1?|Toy!|9Rxt|D%pvXRELFGZc
z$o;H{5}e1QF-k||Ujp!a84I^uL8-<A)kEG3tZ_QJhxGayQ_Wp9I$`RZoj}zw3(oe=
z4(Z&d*Fy1}uCDw)Bx?KJ(Dr|0GNzS(JJ!yPAB#lP(#5?UJ#3$1ku&BbQOG?$1=_q6
zrqH}N$|<%2%dW%@b8P|ENr{+xzfZxMDTH1nDjy-SSEigZ<PCQwvcxyTYw|lc7OT^N
z-r0T$mKfZ}vVn<%mEmCsy$U{gbLY>`K-x73xh~|h1A8;&k>Tpzd}S~*f4&qO`3!eE
zAbA;(tPixdAAOkq5@ayK=U<Nu1Y*S`?MywLLbJ*h7H!8Ys?ejzRT$-cUJoRAkgmgy
zl6>=4wKfx${R%n&P>Jjk$o(Jz3s^{;xJ7Jva-l7J%hN8rVq2cr6>fQY_xN#L<M7`B
zcM^O(;7(?rJ?!%c_Bn-pPQ_1Xm-}Jl3gK(6`(?N%xhgD^b|Qkx3HzYGG9QdPoa9af
z9oLGmPy(2D?}OUB%d)(G)RFy8=OsN`LNJkfkJA&fHU5^8(cjm^R*1+=PjX%<-HxV-
zzI-?Y_}_!bb_;q<%g7dMd+2_@Duxh`Kw_xvT)OR06D~ueUVRTn{5?wgWSp`EN#L$>
zClLF4eOVa~_n@Za?u4Q^)Buhw-Ph|f?@D@rrb3Om2`V<@!;jEZsPSPzMVroSzq5x4
zHr=kIFOxj030<uDy!N~&oqPBPrgc3Qk_!{KX!XFOwo4D5V(s+pjEe3N2WwZ-&q)=K
zZF`BcY2yOFr>RgbAgI{5z@*nuC`vBv5L7f5NVhBLSdtml24dNEy1A6uz*Gda%=6UC
zNdC#W6i&r3moh%ID`^!;Be{%H?MjLvqfI3L$z&8-^!EmmL}C%8TJ&d>ka=#jMzU`q
zA*)4y`)Mjv>}zSOAw9f<BykAo!Q3hS0Li|M9(0&DQ~6|CDte%_h(3A&RSnjWwZprT
zo~C6$<&<BAq7ig9=akRWRH%R|sMrM5muV`LUkNH&KuxeK=?^3as=i}I(NysFZ(L0_
z0-SMd;*3O4(cR9x8hIW~g))X1ZCOH7p^Sl3F(m|exQH^K*hkW^Rd<J1BXjO`q#>ax
zhKFq>8%(CGf{M1-wZ&@W0h$W6)HOz1sx%eKKm-;0)yQw8X(d>V%vsR2tQwqkJeeal
z90gl2iA}C;wi>;MZYC8>?tr3}!Q{g<6$(ED6&o!4YnlpW9)e2RVDcrB16A3h7fk++
ztI0-y-zM27u|iO>2_`?GsZhocqb<LssZhqisaOP)!{<`L*H+yf4kny?9cIX262rre
zCmT!}h@fH<Oirb#P{AZdTh5`WPzEBX*awq~X<7+_31>mqvTAVF!6ez%mK}71C>6aP
zibB}XpjPf>5ku+nYcP&prs+{(iqlIZDVsv>X|Q%v-NRy-9$5*G_+$77Vb_XBN=G$|
zSYapqu9aK4uJx#39u2&gYy&xdz^NGUI<jl!<0O&9f=an-C5DuqARCZOO5rv7&yX|{
z3n|^Wats;0Nb;XdMxizOza~i}7E!7-`bG(vN71^Eqw*FD358}Q%_Fl%rKVz&vo4^i
zP+`0!ZiGGJG?GSQb&r}yTuAb7p>G4WhCE_3Ng}a`M$031lI&YZDC7}lziW&NmA9m&
zVw3jZawD;8#gOof!6Dwqty4~C-YVXUvrI0GG4E6fIFIa9xe4Zz6Yf;`0LZ`nP8ENR
zj@+Fxk7_`4N6I6x$ZMaN^u|$`$c~hjC$XdI#QHT^fm|u3Yrg&Zwa;Qx*^%9!!z;Fa
zja`lPk=<K_;<NJg4INp(c1g<hYd$xr7qab+v`;ov9KJ0*U|nUn3M1g`T<louvb2?H
z?9d%NQClpf($3SK5#~j%{*3+*bm*QXK`PdIv}gRP*m|@_!5g&oXs<h1qA{&2o8Ok0
zD5dp(C&^M$bis{4WH0;ejJ=lZ$57OAsP^AzD%9wlpkkA?bstY=pj6f-sA%##h8bwo
z=DDm5*-JRXn?v%GIDJW%;T;7`MYLw_Jue~oC-WCh#ULZYn??g;>>~*#7MOu%(v@^Z
z>6!Dgjbvjiq^HMCa<rA3uU<eEmvP|k1@6DvDdMBFg0)1;h8lS*K+%Z(npYd#NK>Iu
zM^Ldrom*)t6zT{n8tSxP1=@iXtnVdxRxKyi*j{_ylltAAOzSaLuyWCMz$20kWj{^|
z*rkTDKg*gLb13^MnjYm)oL(YH*%XQlWg9XpZ76$kC@(Ywk#L}WnQVPxb50j%f6Vo0
zGfe$Ql7BK(a4H5Fg@#h5%%gIn#3D*Hlwy>SIhbZh_AMl2W2(yVgTYA0VEEx=s`noB
z`^@7IK@%e55MP7HpKu)Fd{9LDafoOqO%6ybv}#gxWa3d6nfNEsi@@+78JSSG*)Lwt
z-mR699iR;CG~^xDI-HIHr)02BYo?g5FOz+vI7Nbl^$OA~Rj&fu^7l4B25s3OO0u1h
zJv)m|DR16|-V2+#C$Qg@eUfsoChYU(Ma%F4J>_?=XOn!VUR}0Oy%n^e7^em95I#NE
z&Ep@w?-p_~+-=Ue7^vj;s$AT+c(GK$;>DSkg9z-<g^S=NV7W#EPcaeWY>L*F_4Ghh
z@}*#`F2l>hSUp?iQ+OK6ARE4GaD7K~-_z5>gyTL0`-tUKst~RFN#&k*k!3^LG=ldW
z-AB}_r>Ir)2SA|RunenG9#VxFeGv9Fo4WLW9!sUSa+Tf!e?XP93*CWj;5~JFUuz6S
z<Wx=~_&czX{k%f3jlmM`90v>0jvXg^*U++-zX$w%FvOpp9)`Tny$_$}e4*_Eh&-w2
zL2P0Tw#(0>bt%27-D3f_pN_+=j}=b<%MUYA(ydd@LBJwh`n1B7a}ZEr`xq77?l;?b
zT>_8m)qP;9Mn-ta0pLeErDu2oY{a-b@>_A_><aV$I%~Xc{&+8u`Tqk*^J!M0`vLgl
zH)56nv!opve+!J~l@KqqF$hyf9&Akg)EF9+{KhI1wBwI)+TlCFmq4KZ0VTt4Rtvkh
z;sjWP>%iIxP;J}^=pxEtAm}%HEg0L4&E|2l-5~%2_$ZI9Uw|T6{0dvF0j>t&QreP?
zQU&FnH*vE#&Q`2=^1lc2UwQ(sPR`BHb|k|6Vl3o`dv}jY^HjgL#-cAzyznG4c*Q46
zupr3Ff6R6a!B)qdg+Yz&YYPEMsPBjn;n$|D*S{#16qLaBHPFZfP`F+2m>K0<I*#e5
zKtI>S;mr4sr+}@kj4fK9ihp1c-V)l_l^}tQKhJ`S;vZP#T#F*YKhV%(_{UeE_P{^r
zc5FvK3es+A|L&zQ+bNx@gQt%ifR@BQeMI&Ngs*+OohZb7tLzrZHtB+BKi_&U@XF^~
z@53uL-^#AWdcO54_M&!i91NWB4Tl(ZWXAQfq%*Ejep1i5E{WFJ>#*Kyp0^m^VJu#k
zY<#2p%xDEZ)EIsWpysonY&E^np(EC0=_7tsY_jxufJbezbWW=lF-}8=XiU`}<}pJX
zAvZh#Kp(mrwDTFCx!5zDPeD;C=P*vIe3=!}k3mopRgBI0;BM&0R(b79$=mT_@^<K8
zq9$BH<0C`7274HT(KGW<lffs<<7Yu5BI9QdgFF+CpH0iyjGvjDT8^MCGLE1{n-v{H
zdmaMdUkfV&7mbdg*{tr76(o%|r=S4X>Yjyo#i9+n8tZ8Dx|%awxxvZ8Xa&6A%fS~)
zz?;%K5^(yH1{@<xsbkK4#%lGj)0dGive?Gn(y-0XqnBj-%kA5h=+yqut90j49c|=V
z#<~{224UcQu(l3`yH$QwEZluR9PV;j1Und#G{wy`l28>DrHQXs`_hOVUpGTh1R{&m
zsPZ%wYA#<;v6;)S(Nw6pd_hH%<U!b#a|_9#8lPj$+^64_v)`mk^SYwD$TpCZd7O%2
zx-+sX=YEn%VnL<cl@mispCucROiJO2^DmP$5(_Ea#CZ%EJx}tVOh%!J^Os2yiA9uZ
z;@l`9^ThcdN%k!y6hde7uAC|Jso-iu#Ri>^rKwQp%q3>Pjj%`bk~9*ld(=E)70JJa
zCJZtPdBkNTiNqorEsuBu$-aezLLOoEyV9soxiKvjy?altG35!4UO-iLFdl%j!QxEF
zjQ~H8;f=)_)4OQdP)p*PP}DLny`QE+<)wm(4f1`Ora~c~prYla?bm5`V2$aMB+shl
z#L86L^PY5#=@U%r`Y<Tz5nQy1Ji?|>WLVA6@6=XK9)#}3s|QDk>AM?0XUYmevAMf3
zenH~ogi|qu#7K8zCP^f*wWsWE#E{Y{WCO^s*iuU2<mwEPMq(kQORi$b=n|6uWHJgR
zS67oH5{oERa%Gf|Iqh*t_AMk7@@R8+V^3--HbM98G!-i7T6;v3q>)(NqvjD0ko;Td
z+kmYhk9dS6kyu2d<q=;c*|(5T$Ro^t_s^q3<;Q8M*rYwmDttr2GY0dB<~KZ=XCus0
zi9tIeQ;DyI4pPFY#O@PqrV?9BjL6x<CC1rAW8<O|ir<F`#dE{}6ZV%Pi}GzI6lDcT
zClu$Q7}$j333$aO6xr2SPbglU%jdK8sxw&3m9p^Gz#6;&t|QZlmnWT0G%%EUGV%R}
zN)Pp^#90QO<wT<Kw_qCaY_PEoO(UM+SH-3gUjPr%R?YnPnnv8{OC$C);$=|Oa=7O8
zG!<&NMo_WI99^0Ul{pG3+HlQ5=-1y&a+o-&O5d;FW74I$Uw<3f1~RkaR1CRiq+fpz
zNhGnLQuga(Na<5#1CmK8+^_#4Nh7h4()D6u$mm&;|70=>^<rKiNhB6gs$PsyLgxO(
z?@0D7Boyk!nEUnPPojdW4HX-7o=H=o(77dUggqid(nzfCQS*p1Nd7G}VZhdqM_fRX
zNGzh!@`$TR_AMk7@(8ov<wu1|EiDzjUr$b&@B~LMpeoxqjd&5j4`g^_F^zZ|EgNbY
z(SxFvdFeeg6)G<kRBVv%V>A^C`2-a$FKs^+*MVuo$4Q=5%Q-}*5kJec9%CAji#Cx*
z*c6IPBO3ah2YVXvrGuly^xcgYnX*DqZ0>ISfouZ__?(I%Bu2U$JttEcP-1IO+1-dC
zrFmonl1V9?TrDAKBo<P-<SK@Y&L{a#CZkYtwUH!|SVXCkE2D(WY0m)3zJ-KB9&PS!
zR8v#23A%5jsZc?;C2oX0!Xs%UR`;lR#Jwc{7Wy_|Yse!$PLfD0qS5k*CrI`!Boy)p
zv)?^4DpbCgmWoZ<qf8?j5}q-bM(j1tw3%ii%#(;gD<YGK%b|Oea1!y4po{kBc^M}~
z<P@TLe$>#g=mg>mFoC#23@u?#DbWO?tRU$G;_Or81mZDx#U>Eh)gd;4SgBSItqDZN
zP^Tvlc`K9bWgEr4h#`Lvf`FafuzB0~TQGrm8rWEeCJ>kSRj~=gAAtw`H=aPe$d^X!
z3B(JasO3P-R+<VmP$Q_=<ctF}6)I;GRJ4JbgD`>EAURB&PNnbBS53M!_vmjW+d$@2
zoQfgqjP&U5B#9&zRLUNG3@JTGHXxal!u^-ekTentDP8|1hK!yf`A;UJQ2*sSB#Fc#
zO4WZcO32)^_yx(ng@i)=7juvP52>lxpmWbcDuk0SENzJ!VUL(Y(nzfCQS*o;B>xur
zHehSWBl<`ZiA6M89<h;R-$Ftmk1+e)>qmu(o0f{+qbDa2d4i)CP?c?*KwJaB02$s`
zOd#G$%Z8djyakF{=B0PiRH(dEP_aS257AVpyi`!p^3qY6KzxMcS+$%)WCHP1OzSZw
z5V>d*d4x@&$ONLH-+8bn5PxuRl$gG|@jO#j2#U?!jejNEKmtCeVhD+m?#5xgRBn>k
z+EaEnVn}H=*??qH3MW@5ku(wuDP3|ELq=zl{3nx9D7m_bB#~G|sgf(Bgv@ErbtL;1
z5(;^=xw}zFO~oeY-au2Kf^JLP2z$gWB#p%C9yO1+i{#%z-v(?AdBpuBiNqorEsyvt
z$-aezLLOoEyDyIlm2ah`Vw3hL6NrX{=MbGh<TDZG3B;fkkqN|w&^<~xf%rSnMeGSg
z&SsnmL_RIFsEJy6a3bV5{5<*}!35$WG3bLmCHO_2<}Kdnal*c4Z(%-I>xuCHWca@j
z{x90jDiQ~1%L<ZCAWmB(ClIIO6`Me0S7SYaxEWreTOX-8*&=_rP)BwjUYvCIVThU3
z6NhJqD(tNcxP6(c%Ed}~8@%O7J*SkNJnA1?Dh|EW-ZwlwRGHik7`Y$J<n@BGJ2aDb
zl3x{@$$JS*S)0i_wsq4Oqx!(cQS50DLFVS*tX9!zt^a*EJSZE=u7k#{@i~e;o7M+K
zEk`Xj&{U{V3qi#ufx3>SLM2dwiYB$`M#;8MAW9?~DmKL$h)9({6u7Qzq`04CpB(iT
zRP=LRZ<dzVb+htnj(5;hD2s^Emb+;xltpkV25pB^>IX?0iPfDhr8a89JY4ZO$v>Ge
zwv<sQrGAPekyu2X@Ce;jA!Xo;JG!<jw)z8-eG3WcX(&1B&4EZSpbE%1>U|b4X!rht
z$QYj?B57lM%6oNCG`$uMF@7->FV)7x!oPG5F%FoD1S_*c%pv(F)y=6GWE66UQ%Dkt
zMU=`Rj1n?CL?6k%g@kPFd&O)prMa%bNXMXGc9L;nXE>iU_sN15MEYbWfb%BolU)aT
zXx}G`q_(m@c8X1dqKi3S0*%ZUkZA0Ssjq0gtUtHQ$zbOs(<nF@NJO2%My6QL<T9mV
zy^*Qx#5c7%Lr$%}EOS-e**Q|m?5xx>g<LsbDwc<oH;hSa?qrogB3vYGT6Ib2n)+|#
z`WeV#rzi=sI7n}q+6@$9zhz3YC3?{RY5A5ZsoA8R<aeM=vSWJQiC3(X%&t1oNp6iB
zI9?#S`PWa<<BwiQyeDe{ds}JKQ5EZshm!8yegod;H-KFwZNS^uFb4Pg!ZOQY3vZ%R
zZq$lSy$M0ry$`DJE;YF;yuEa5rBcfDvi5*4W!iV!7d5>#?VnmkkPG$cSqtIih0<o+
z`y=Bsz3xNGJ#Q{qIVKwJUQmV<mbY@-suLT|o<?70A#3QuHA^z9@=kp)TQ3g5;Y+ze
zr!V6_l3BQB4Ub&Cuuxa1I&k$zK(A_t-nHo#Jsv6)nzyC85=f8oK0T+~gXTNe{gh7+
zhYI?~bpgCQtcvhN3l&Ppw0brW(ARz1ra^&T_w&Gq`#5~)?sH=|+`b7u&>24oKeuRx
zIw0eI8fo#Qn>31myC1aS?rG%if6?4M(VDx_&n%b>%y~m#1(CeruRsiO@`ggU`wB`1
z1s)^~?CaZ`?yK@Aec%=B4wU!_A@R1+{^SdCb*J8!>8tMT!@;Q?CO6-~tP(wQqIg{C
zYmGlu_st-e_U2jCTKJuW40jsV#0~2x6B`QA86l#&Z}+-gT>!s_RrfZ%?&{v9IGCmx
zJ|+ae1u;#5YN&3QqGGM*b1aneCf1ytjlET;>9G+ycDS{(r&=lw78}h2?C0INTDf_^
zodw*uN8!iO_`&o!8#y#L6S?~(PXhW6Lv5mnDLBFcKUE8K3Gu*t!&8K>?R>9Cc96ZR
zy`F2p_-%@TY8m*3okpROhmf|}0uN1gASzmBU@sKrq#W$S+_Q#Cd;Rx3wtth>lFYDE
zujhu)r`WLeg~h?fHh5?Jk_>#{Yq{!s>kVhPFLOl|<(VttkJ~e=GaD;q2RF)8N>J4Q
zk=4OU`j0GLybJ7VsGjbnU9$`gmBGjOTL9-53+s|n4Le^9IA7;i#bV|JjG5Z_`kjZS
z>^UIngZrn(U{#dYMz1a3Ho4swnJLthLsi#8QH%&`p4QbTwGLN-j(5>isL@_Q#U@4D
zOH-jzG(km^;Ha!x{0ow2wRfV0jCqsQkm3FuqBr{;P9Gg|8cnY~?{VIqfFXv5k-6@l
zF|F%PYGEdNCroXZp8bmViCo2~=pHeT#zK=5Op-;URR%W#<NO4v0<vunaW-vS;PW&U
z$^`@!8yEN*O@(p+K}B<cqf9}E3pz;5#030oFFZ5ye@JFj8;E7w@dtuc`~9{_W#)n4
zUy=Ni!<?LoNk$mK6=eJqNh7(87@4F#9z#a6mgxC<2pA}{ZYiU~Awy9_G@m4qSVW9U
zM<rxlJa{I_zJ(Q84Nt72sZfq@tA{H|5{Hl;%##)l$-a#qbeK0&si&o)2TF_RqZd%s
zU>#XsBYP3xc(9fP`wpUiU<#_;{An3d<<}sfzKfOtwG-xlKvBz_^8GXw3V8$-n}GUP
zG!@FP1Qjiy_5>1c8Ae0!_XUyzRo}6qXe#*o99NT#0KY-9PcA1BRCKp9FDLmSO@%Us
z7;X6_O@%TBPQ{cEptmW}?mtNyw(8!{6Y3C*XJnFwJI=k1G$b^|@UUZ+k{JXgR6*sN
zM=3J}U8^N9!Bv1sC(~4Z2vy`%;X-%EeHL!Qcv+0LoK91*Xv@E<ZQ)82gP8nF(4w_8
zt#L<(TI9{Dk5sFb8ocy=7~WBF{Ya$&vr7lCp}$cZs%u%5iFG`gC&r!p;imRMWdbWA
zRVTg$jlGU;EEQ6&hN6}sr9e}mP()C%LBtzqDwL53Dq2XH5teexz!L7C-AXc|YCcwE
zN!LHS#iTBC|Lks(e=?)vR19z&3ce4LBod1#Rq!=R$efKnPO@(yA*;T}Q#2LIpp9)Y
zhxqT%R4C^ZRBW8{C7Me6wis+GjNHE?Y1kT7heJG<kq-W3h>wAIGnOR|@q&s?h@VGO
zp+bC&wk)BkPzEBX*oXK&npT1k&sos5tQwqkh@T|qjd-+iH3c2Ec(IJ+co~H8Ep&q@
z6<q;EA&hIGubWR@Nm}x2KrctrqtXUWFOj5d3Pl#;n&z&wmAD7taIZI!;!*7)RsxW0
zx4k+@dlOf--t#jj0e6sXAVVCdVgT&OZu|R4B8df+V7GljDg6!EfMik%r;~qA(nu_%
zbm?RaOL~svKbed|>%M<Pl1MC~*z3OAT9G-O{CARl3kiiLi_HV(6PJ_O0+r|qZQCSz
zN7GcOINlOB!X9xVNh7hkN6jN%OY(1_Zv(c5JYqdbBC&`@%OkEO*|(5T$Ro^tH#90#
zM$%HT$$N0PN$lz{Wjv$5_|jV*cTt$KjE{{H@};CfYa%N(*T5WaoRyk<NgrFR`EJlj
zuvn8`HtiV~TFK2GD`GY?vhSmR0xljrU(8qIh#E9puD0ZB0Mv~iAMF<_*D?hMHh<I$
zm61|DGvH)~Msl?rZ255VusjWKz?$`Hu2`!t$?PsR3T&ImPWal8Yv14*wh;ujkBrov
zntvO1h_Hw24lF6m`?rIq`8<{UVOh_GxkfJI?8yyROHN<UjvYJPdZpZRlNc|X?cIB*
zv#j)_J`{3R*EuH8hIiNwzQ`8V*y3f-A^KRTSTJ4h9bU}8w#m#?L){6mF1474RjI}N
zS?n-{JP-myKnB1FRI=_~{E08Q?NX_Jo}_vXP4(<3RcKbX&|T29utZv;5qPb?z;zND
z^?iarXwMz07@u1JmURPcO=62x;dF%seVLU@dKPfQf-iVgEbpwWSO9m~2sYf@#;&p}
zvbaBC!FKpXu0p~edou9vCiaO}3u={;gEgUPLoIj@{~_fDYyrWa3dIHN{yzS*p>VEV
zhs{*rRV`|`#3_|3^F=K^Jjkfdy5%4=RQAeuy|Dl8kQiUs|2`&2_=Ss--QeGtIN|$}
z9{A2g%6{A?z6r6yJrFG1lcMDhDVF}-mNsWNA;dR_DQFJQansgJA;+a|Qdd}2pr+8n
zVfCmbq-yqY7;+y{43ML$`$^@NH~*%dFcWHZU@M}cN0d_Tqwqy~=1q!}{xy!2+`qwE
z*nuqlGf+^kazQ<*P!fy5UY3EuJHduqaOmZrO!Nh&NVrCSfNQ2?;S2JoLJ_W%MvOlV
z*iyD11)xT$`bca+7ZjzI>|=ocQqXEvhPRr87xxoE878S$n50O?+y_Hte#*)e0#S}f
z%aU%zzXb4iQX(PDLLqQMI`sAWYEz9@wKQS4oS%Tnm<4BhXNMH;(`%u9PFJ^pvl2D_
zXl(pS3k$eC&<raR){HxIPT*eq6wACU5nsmyN};Vx;fq-r2d=<k+*1_3)DsQ1C*o@k
z^2KRb^Thx)h_8hsBrwQ~afZCnQi%0f%uWX$t9}W#58znb2XK(K=I#elvA5=i_*(+}
z1p@WP7xODl({m#C4WP)JIL@lj15K176>xlPm%9TVxmB&m0Df|dtjyx&o{NAw77|-z
zyZP1{eXkSWBYO|Dns4%bKVGp-zU(UL9$B`v=6K$7O0&u6O4yD+Sjjt`+*ou`%8f;W
zrPP~=o{(*eu`{I88;Ef4P=~h;P4wl%dk+5hAhvzR_h-{4m0mZ-7NJl3RWWE-9@!#v
z+o2|1k<N|9k>*eP!ZOL7oE`ou6vZ(Di_JmL(^RPOUO~lXy!U0A3N_v<sAyA<?Kjx4
zbo*5q-9%nI+hS)<VE2C@dDgn=lcAfQuMFpk<)%A!n}hz1X<biqEzCp@KWe*-^U+T-
zDx<SGD6@j<Taax#4vJd3z%rT&<pP3=jSH-zsZcH;sAw*bZgbE^k{Q(oV$D6Ln;W^z
zq%!lo^8m>|Ir7h`80JRChc*XQNgByzlxlNO3>n=*@}Epbp#^%klOz(0DAfWzqlC<J
zo%fUMTS&-if!?QSDwN~5wAGLvo*+paLV7UIuzibU-$oC5remh^Vp=MCptOiSdI41p
z){(Wtn}ZHpnK+>S5yWR1P-oCosDLV{*aXyhG!@FP1QjiyCfFQw8p(mG?^sbZ75ps$
zo<L1D0=$r9pWGZIsOWBIUOBv(ra~D*jJE8csZhqishARi(B_~rNyAp%9bP%ixz~|~
zgr*oCb}QLn(m(_iZDDGQ%|Un4RH$XCG1~GWnhIqgf{Ojh;ZM=D6098NEa+NR4bD2A
z%n_S|f}N1WUeh+)g1$yKlUm>Vk5JSyn0%Y2Lg9y?VuOW0p{Y>jA*i$rChz{_J>%Fv
z9Fs5|;TuZk{+(n+)%xf~mEUrG*+_Eg>12vQNm5X;i7InxDwI{kXv-p+O8d50M3vW)
zG;Gz~;i$r8q~i@4Rbsf>m1Kj-s3NG?M3pQ}g^DUM+A>U2p$tS&v5zV@(X<jo70!aL
zW!2!Uqe`;<C{4OSl!|VHq7Xkc%#|BZ#7YbKH5kzS08NjISDaoVN!b*N>_;(8lpO5+
zC|@MSquNERxRZWA%41yFdRQ=z3qDJ>fgDNTR1AO}*^lx9NhGnLQtn5IA*J7u4M--X
z@ap^VXHaQCVj-m)VU8iAIVAtdWE5I`e+o$=v4~QwzBfw9Jf7A^vTq@w(EOx%jP{b$
zRBV#gt7$4!9B+vmVUNg@G!m<O)I4Gj$-jlZ4cHp;h_{m@5{qcGJmS41`xX)kd4$>T
z9vBrWkEEqyllLh5QA`=n80<$mah<Z}e5(VS6><$n9#SmV%9!_{1TBc{LHPvCnJ3(X
z@-5JV^&S*|^^V+ka<WBpqI*r2<5I(i#qtLnritt|DOkMlQC5(2ncFY0<?LM+zr-uH
z%#B@*^)k2XgdJqz>}O^lom}sBMQ7H#`MjlG__jOJTG?!IDz@~Jb(P^NoTgQ7z`@wD
zmT}3$4BEdxaz>aJxgI<EL(uJ;bS9OF##+BN-mi+SUt1YjzxKL=B^%Q^wE1m`iBeiO
zd6Fz8Wf$BCWH!%lXzaCT$3judq1`1k6>1PqP_fDF`e-UtZYQW{@;io|_!KpIE^9-M
z6V3xSko+W0a?<62mzcC>?qlyD`6sg+PQ@UjBO~%u*727~f{6uYpqcc9KBM%^+1p#l
z##l&C&#TBmS#G|10aaYa@x9sH62@Z1>%FvWr~$gWIoJ9ak@-0MkJD5r)Dcu{Q0Fn4
z3WYj?iiSF)vf}k=l4sR&VvPW{=RK+4eU)iF#)?-i+75U`vJvlJlLB_B5$|8Hrp6rc
z9(NX%`jFVc=_QhsO`*t$w<)#KM!XZe3w0jZ{KWR0F47(gOhxh|bEI8K@=wMJPQ?&u
zLnAKdlOz(0DAkCIQ9|Zux|L+#LP9pSs*F;YtaJ=UDVA7t>w_MkdAK5IN@Td=Pte~^
zI9$;Hg;)<)M0;&=+=6=j;j(ZiFgk#-5(hB41lSq}Fx36`i`TPVYZ=)w$_&b$Y#m>Z
zS~0u@HBOPBSM^rXT2*IL4@o!^8?#>&g~q&LUU1fg*vIJ(8txU8dtT2bIQIj#6vA-|
zvfroP3V3`^93I^vpaI-9u@KvuemP>>9+mQaix*45E?%5zIcC7lSb%}H;asDEhm(k*
z6-5`zdU~L$$}Xo|ywRz#x;E9Etu@$n<K(x>`>Y=S8y<23``6(Vi3ZnVM58@DJxm~m
zt^d9H-v5g%8`7o`yr1Z*2LyOi)T+e;2UH<OAAn8!rjGlkW2x^}uD(w~A$Yyq>_T^7
z2Y652*54W^5gwJZ3Hs1nKgMcwKcKdO>^MVnTK*m|^|cUFdgd2$Aoo7rI^+e76AY6)
zD`6h3IO#U+9t*gAJ`T73AsFYP{JzcPM>kJ78G)7IGN<*VoQ%Nq^if7fxAV<5UY5Ya
zdUczhs*y1nz2ONk6Xn%vfvEo&N7Sw```@$1>t>Jl5}EyKRNL>kI5o4MWdJE@OUB=V
zRvrCXof+C5gt;RRHs*e63>8X#W0eWoaY~$a_>Qm!C3+$hjd4T>E5o&5?FdYB+!5$9
z$`K*Vc|OdvO)F(L#~=)l?M??6z(@J{E@xxY--U{VgH-^x+J^Uuo1Ml;t?b8k|9hZ>
ztKuj@2KLQZs6}8ul?{YKg0!8*T%JVXv0@ORGsS!z;<f*e8IU2w>d3P&sIh%*Avg(j
z4-qK*+SIYcSW-~J+J&eRmsuknMrKA?m+oWwDNxXx;;`m>$$I4LX2ut7iAuO&Wq4a?
zi&w%0YyXf19mPvnnL&#(!b{NFVtB~~x|h&R+0GqN=FxIe@?f1z@-Q?d_Q@mxb8LEP
zvvx8GaiECo>BvUul4yUR$d`dxKGXUYykax0?5YzptvWN|9PL;9{6_hXYqAC6Rr1=~
zpZSR>-|`#qoZkR;m9znGL$*3_BFg#eE?K+v;tl<$XSc5J-?U-v*7ccQIB{{fk{>BK
znf6n?@Mh&R+CQ}{LM1fo%bKH{(dFJBnbGfcA5!jlM;fZc?F;V1>Qh#4wioMht)30g
z@KsQxnuzJ(*6V&A_;4SG58WJY?1tMn!3XwAJ_$bqI{w_Rin-NQIBWa&N;BMF-~;^p
z6GOWLp8n`pWSSL?HIr=ODd|Ac=~-@1_2t}DmX9=wrTTK-?-i#f%Kr57!9uPvR2pu2
z;%G_Mc=HHRz++^D=qDhFI3q-bZkL^?ijoe7hZya*H{DlJm-rJZ?Lkx;PlK*Pfd8ow
zIw&pKIuu!f^k*HA1Uin-!||l1!<J4x8+c=@P?q2o!$x*B*3hyBJ+yyAa$s3$2`mbY
z)UYx=#5Bu{Mh{fhg9<wYDr^0!7^qwm0hM+500)&ws{=^c;#ZRzQcN5rg_M<XA!WD}
zAPTnJj3@z%5ky%9g_0l&5>pTbze+^G>ky(`gNU*Q?u(99GDMLTBtw+_$T35dH{%sU
z6m~V%h;sgg>(_7EvVQ&L$uY$yS&2|djVebPsvOm1<-Nk9q(P)eR_^qxVz_c`1Xu3D
z2Sl<GZFhhy5BL?OMivuKNs;A@ILMNBoNC>1b_Ljitv6#!z+?nl-UWq{U<(pcum!(L
zY{Bagwp@$YatGW`g)OoIX-e_3m2V=)3|qd1R}5R&)mUT8>WkO+_peP3EGsOM6`e$C
zRGAj8Z&V@Wm%@;w9i)JipZQfWNZA&FlvnTp4k_VA2YB+Iel4l-#KcfiJXsM3PllY~
z;amVF*lshJ1RO@d<i}7b378-;1x)a(1QWatfyp)mlb7IrDlm~1NF#~|CX3IJU~(#6
zF)(3QV-6;lT-$%aK?W0@M0zj@*EgzQav|uiL;arf{i+z4Y>$A+1^9pflTf1rFxl+a
zk{V1*42=qy6!s1eC}4u^HiJpPVFXOhg+fWd1c@nNf?p+=;B^R0wj-FVhWn|&L{^Xt
zOx}bXv&>{aUNJCXS7Qw(YcJinb^Ydz8~Rr!&rZ&?$WEeUQiIA&Q=Ov<EPpACO4>w<
zEalyPRSa0N5ny=_KEQ#+)bIdYKH}Gu8eB|FB?XsQJs+4>b!ugGoEDpJ#+QJ}2)^6~
zg_7V45>xO6ze;?;>kz(V5npbH`>F6nRv>LDUb6Bma?J4Mn|Q_Wg<Xv`zO31>VQv4~
z6yRbrP8*_;8d;`CDjZc<`Gv40X$mP|<z>Gr1}nJ;to$4w;II;Db$}^<@T*CUDJG7R
zVoI#uPpOnsdq3E4GoS<<MgZkUP$&tYATb3{@T&wAybb|M4guvSa6c8G$O@zt#RHT@
zeG*U>;}ruGc6Dd~N{Sg}Yd{HTbO=z?3P%-C)`1E;)B!rruZjW6U<6Rs;sYw6D6I|v
z<x0Pr)PQ2*Xe<B)8*T=afWru&^g*E{fP%ynK*6sPQ1ChgD1!(ntKfbrK#>(B1C%!+
z$1Fp66J9YuVOL`fC>OI}A6RMjqANCMw_b6@hS+OW&b8<VwIY@pV&=52c~r6HE@5WU
zN>U^*O}{FJHBJO;?!*T;*0gSZ06O>k^`!<K6Ki7uIw%1%=mZ=_K<AxMC<*8wF$Hw+
zs{|dq4uOt?pz|)cp9*wj1=5`2Wij7Cjv44Yi&qSE*wt8r&Z<q9TpB=5Y@|8O0%<e~
zso`a6sJc<blvjimNh3&sDL?hAVwf@%!IYoj0~}LAZ4Q9sKm1Bk1Br>Bq(HJf&SX%v
zPz<Jmu+e513HXb^$cs=Y35*~y1xE0zgb}<B!N?H8$dBQEDj1O!NDGPwBfaNJ7+Hi@
z3`W@1Si{KbE3a9<F~#7D&AhTsA~ldq3)eTQNOA$_uR}edHGWkLNs199IUgV3ND^*z
z041CJT2e!aiJ_!Wa#|cHxqi1(4zL89ZN`#-!3dU|1BH@c2@+GV1iwlw!Rru~6cJ0#
zh5M<nL{^XtOZFkhEG2m(UNJ0TS7VJO7wL|fyc^VKk({)kl^R`+X<6~8!pwVwjY$(p
zk+9t1SH)mvR|IC>jSp~`Y1#S!Z$9i-ml|(OoQ(nAVEfH@6EGOTn|DB=BzS|w6uiN&
z5^wN2gg3hoZ*GJ8sqjWtAdM+r?(%iynBmPg@QUFLyBce}S$*xLE3#)MpLMp$U33zu
z(Pdh=zEK60p9@2hc8~&8Uh=DApfVf*m6!1Wk-LN%9bn1t{aR9EiHV`4Sh6fm?y^_u
z1!1GjND{CYL6U!gLP?MWi77~eUnP>@bqGm@5lMao_fsK>tUy{&JS17TN+L-wUNIzL
zS7VJNYpz?r>5^+<&^GBNLz^j~5RKGWGCfk^sDjB_P+^BULaY6%7?@NdU~(Qlz`-Qa
z>Htly@T*CUCMJ%OqDid1KsP$o!rofY7s8gC5hY+Tf+%N0p(Kcc#1urquM$!4I)o?{
zM3g?bp9)cA1<4TQX5^UVCj0P;Aqu-1YeYG3)8_SC)^GI@WlQW4+S4sML17}PF=a-y
z#!&^8JB2w(Ye<oy-0oM!K;`-fs5J2b4l2=R2dMHPzn;{nVqz&Ns+<}lM+x8r+iZrD
zfVT*o{0}IU1Wu5c0w?%Y!U<l7;N*IQlXt@XRB$3IkOmYlLHXavF@uw@;}wGwcGU?u
z30U!FRC9y7azoBi*kj&+6W{AiCSN?<!T$}<U6a2pgwq=rZEu3a3O$9ng(*DlqFrCU
zT0QWix3J1vz;R<K+r*9p;P5uSiG8Wu#18wkPlOtgns|B{dwQArba4J*t`MSsxI+Ih
z>U1HmyIvWo4LZ%j67Q%+Emy9?>mZ7ia<*El4Aq=^y}7U1>%nU>8@W>R@^M~|e8$Pc
z*AZ-Qmp5&2q*enVWkp>Zy(#ie368F7ZuBPAokpWr9;!FJBlDF(-j<y;Cck{s>%KZy
z8gc5)JDQ8$%m{W@MjF+TMss6f`bKZ^kW+SQxkjZ1HBE+X>`uLr8?M5aBWjgO16Xx-
z7WXv0uH{Yd2+l$l=<b3Cdc@a}S_!^QZBz!c1E5&^6;2X*(VMvwDkxVPj&l1}Z))8c
zY?McavwL>Jaed8Oy=m|=(PA|VrzQ?FTiEE$<#(Kt!|r5v7aN7_U@2FxLo<MqDB$0i
zQgf=6S|eL>Zg4=5hj~4PT*?1B!8_8~Qv~@na*dI?d}x9<A@2-~3_*R<cIRqkkRLoz
zb8^FQXEG>a(80zaf5kFLi=P?^Pc^;CBaNL)&t6(D!u4_bk6EWytJJcETsdEIAdl%r
zRtFnRZ#tK`Qp*HZOVc|(^h>A;clv6;tZBSn2Tz)0Jjl|==|6-Ddo!>NV0HBZh^Pc@
zYI;*kx$@8mcppeS58n^NvFBMeXRuPsXY)?gDd&Mqkx9B6JYpv_6C_g1OH(Lxd((0w
zjY>AS<sDW8Lz|i(8Lno1V+6sn8+<Ktu7s+a&h9WXN9_l<V4vnlzLMPo8^JevGn_ru
zO5Mq-C7G2BHnKMq>qTy?)3f<vJvUIoh6pYvN+$t_n*xoggIh3<gG#50$Fe(%C1@Nt
z|7^ShB6T?45xvqWV{Lg*`4P1ozPEyZy}Zji5<Lwpr#u9PSSk(|8_oV13%zMor<w)-
zaKJwYnZTytoq-(ij!xmtDKr|@df)QpaJ#;&y0=lNl$TX%L*PqC<CBa5crpMRZxY;J
zdh%&!fIO$wb`FNB$2&pwr^xE@^a(WCT~2LT(P`{tGyn>g7aGGQAU3&?8vrlLHww+m
zCwa$%O%7*$=kyzptphGQ%>&+y`rdLQw<lXD4i!r9AMAI0se-0Ir9PYkN4tImyt)p8
zn#I2Juh~*%FegNNJe(5({J<h~?r$JvCPFB^YorP!rU7@2VguyQVc-B*Pi3%J-U&7!
zaA76_MSWx-^jL8iZMRg(aYq=>zz9rcc)-bn2k!+A3XanN8b=SpQ@gUxaIRR&=JI(&
zT*md0STkz8?v!?NG&|rO?v!t6ZX6l#CUG<Y_XSHVmh(mMjYhVSs|`62slaZI1g5~U
z8UVCVN8Z_qIPT5V?v~-SRCXI`+v$JlMlj8@`xEgdzwUMa1;gzRF!;Z#@DBIQcnHlK
z@#8IcvdNq9<Gc98Kf{ODwGX)TCS&o&-YO%)D>zVi*8x`c04sNZl{w&U2K+cMQkV~+
zhLf05SXG$pb@K-KQK2w_9|jw|nrD?Clw|<)CKT&%R!s9&%w4XAdd1mp<7H5WQ$XKQ
zSi^WPG&$q9U^|d;7f%s#KZqY6!;fPp!N&>sQNfP}emsXC&*R7D$?$OvetZQ#p2m;A
z?SYTS@Z&>Az{f}N<6TqW<4*iIe=2;e$B(7c;A16z96KF8PQZ_^;K$SW@yHDL_#A$G
z@JRT006%V<2_NsqkC*V{75up27--*Qh)#R!kT_&Popmt0sJve0QME9^TYf1s@=U1+
zcAqO{AmG#*OEChJhcXz$0O>OTuj1A+d9GXRazS=Jgi26OX3{$hq0_S9(6_m69!jV=
z3u|XY({2@7pYta85p_J;TGN|&&DzZyFWq=C0s<x<m%mfVanLFTkfFKmwJ1Es7t7La
zhtqekMt3KE{26`>!h=$B!tw%1ZV-}W0RS=(ex&c7My68BR0}o0*Nl%^)?)5#rGXCK
z8e0eZ!#Y5TM0N0Hyc^MhyAMjo7e7|!?#7Sn=D^3<_;C|{?8lF<9S0xJ;K#!!z{ex_
z@!AZ0tiq4wC&I@W__6R5_*jA;CoF=GlknrO7Q@G<@MF!X@Uad*9$O9{q>5u3z38jP
zE*B{3p4f`|pJ7E|yP}GE1n-iH>YvnQc9i>|Xy7RQM_@M{t8+hwA5Y;2DaOtkwUBXV
zD-9IzXlw;M5>^08B&vWX@owi7fYrI5$B*0KEp~1dKkyxVF21AB#pCT=JpSIrW6ap`
za4sG{<l^x}E*=)*;$b2#?(}!@Bmft;>AJX$*ToHWE^fGUaVLz6J7rv4!tUY{cNf={
zy14Gt#id~`E**1mVTX$gKU|!0b#Y48#VHdPr&L@V>U41^)WuN>7e_Hhys2tx$C8by
zgGERXF<SxH<aaQGm(TI6M!mzIgIotv!Qp{Q$<G$qqeq2q`cF^RYMd;ovO+Vp`dMsj
zLTY_<xGwPv^oF5-F<2SFos7`+9ITW|JZpn)?!<NC{vD8cxyOT%T;5ZETXQqy%+*{C
zvTqTsk#H~VK*wBFhTaC$45!9JvIjj-CMZ8RxBNXY>tDt;>z{|rT8>8qiIxbwhkGLM
zPFTNr^A*rOWkf+l;5+EBi}BvYfOF8raKsu3Wu}orQ!X+IN~pO=TQRs3poBpN?@?sX
zgct`PN}{HtC>?s5^&<59GOVAGsla#<rZ$*ruzqS0{bpIuRu~$<L6%Ihk;&EJSLlA^
zYJ1^L1CW*sID@$nNJ@&0h4oCmlBpM9IKnrfp@T;4h5di2(hE%Tl-RmB(N-5-(j>=U
zy78hb{6z3QidqzH+;5?c^Dcy5QJKjZa_2hBk?VfKoXZTrh{Z1WrR<!W840{~q|E)*
zmvicR0{LAKTYl%+%I`2CKP``YZ*u9i1Sk>ATmBxjc6;pBUT51{31F=}_wJasa_$8%
z%CIZLMjWs&mBda`)e%T-Pi(1;*h<Z4Blje?k^k~|cSmep+-9o_iFfWQC?We@DRhy_
z`h-uXsw1%AkH?nUL$*>gTJZbL7X1GsHjJ2jCAKb}wAF>tf>Y=smGyPjf?td+wIA9_
z&1k{*nk`r#?t(#NI{n<SN#>kIn9zZjfWrV+MvO`GG*rnq8wJQkWl|b5^Bg!Bgjyz@
ze~PV)KiDe6Xu?*?xct6$+Gv>|*(XrPo)@?4a}1Q|MAxT(N|%yBnMRXv49mrF6xt>`
zAcZqyOJQZFq%cF3!gDZk(@yFC<ujs9v6azps|=Gf_RnZB!1+@PWu)%72~SO3Pmmhz
ziY>l^t@uoVrGHim@pZs;-yXZUZ?$c1n;u~Q(W>j(Sj!=le%u#ZZ1>oT&1faZq_C3z
z@`(3XY-K!Zs|+LJrBTKqWW&$J7TdG7Vl&$CoEA3x|D!ha0)HJ_8NakuhS7%8DB}>a
z;i>1x&A)q~gjN1+wBgsZuwgy_{_olk?2WCBlWnzOG-5k#^dF0J>*Tc1Nf(Y(OT|H0
z1Dh#Aue=EJml?iHDuc_d`Z5dlE%0X>dfPqx!v4GWy|Dj%%aqqgu8S?1^V>*<%}@tZ
z-@1GvlZ7aQV37{aYb$FHP(F?~(_fsaPNI58__x`*xFcZTqBN4nb$CO#B(5iq>qo^>
LORRHbVA=lz2lc$@

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/conversation_template/internlm/index.doctree b/.doctrees/autoapi/lmflow/utils/conversation_template/internlm/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..6c0ccdcaf966f74ca415db078790902299f69334
GIT binary patch
literal 6735
zcmc&(ZI9eY5#H?Xyza~FZZ6A-WRE?e<2Jg?3<(iBmQ|$a2ofY2&?!O)C2>4<&$MSf
zwz1p$LKF!k?v#PnffU~dK&KN(eB%f3?LP1e5PyiLx^0hVW;e-j0wRsqU9PUKuC98j
zy7J4?r{DYPg8C;LQNSJVq-EN+=i4kwWYzYqc*s1FyhyJ8DtVUd$y#V0M}8byERhS~
zu(;#eA@h<)iCiak;kZ#NonMKVB^=+&r}Gjo^U9-SkJn^LIKpLxPx3u?*mM2kmX=}p
z-VqBUlLQ!o4FlH{tmSxug`PW1NDQPbKVT=M-h7ikO7w2ZfeGD3!Z&5j$T!QEWI5eq
zM=raL$C6ImHD%otB6Pa3fNjVYOO74UzAKh%MiGd3FL~xrzg6ux7!SIpbr`Z9@x-zs
z%szcec!OW&tNaRI;yco5bAQO%VeEM<Y#;mKVf)_w@53P3A1D~LwPI~E7QPucZEdo)
zGFkgfleIIG;l*YrtuVrV-@3aPclm-WDO%wW^!f0wVSoH>SorJsy@TJo_+6#F!d#)F
zcDN^(1Jel;`K_UE$1eNs?mXuwZ|GUgc_#64M<iy<-{Z}(;X3>+91ma79y)fhjf^1c
zx5o#k{APQClzIas-quH<C7S12ffS&VyQSIrHQ2MN?71P#ZtR+&Vfn5R1<abmw9ATL
zZan<y)9cTk-guV%O=!6vjhB<Z_@6n-{`H&t2jpr|JRF+g=`DB;Y%RA_s6gzs3ZE0X
zSr{3x#bkiLS;xO0bROLKuygPJ4dcO``#-*S`@x-hJ&Wej^pw;cYqBi*p&th`<|HJk
zz}R!zx{4{sjxBg{w@&8uVBSNVpr857tKy&_;Acevglk>jKAlk%dM<_j@eEx-AqCZ$
zQ)&PxvmcrPFHDq=_k+`YzdP7}GHcA;gE<1KH9P;-lZWaw_J9tn0rPCf>l-J4uA%~u
z{;LXP-cWHJfW7G!R_33j@w+vvc`}aUKc64RKrOW!xfq-#vK*W?VaS9PL5|Mzb2!AG
z0Q%=*fGoo%K;%Sz=Rd~c1nHE73{A&N_}^p&BXJrawJwoTas^{{61a}#h~&BYd~AkZ
zGLkhofX2@`GoP$trpi}IN3AZ*u+KyyZ+$tpQ6TjPWU>D*INiN~bNdv5yv@sFclfmE
z4(!|`rE`D5clbX5=OpL%H0@<RKodRz<8QIx@1~F=Ets-&^om6<^MAeOLN8Z;1Z_Tk
zm84&T31D}=B)Lq8nxIC&1|4wsrbWt21)$$`eXp;P+nkOg8?&EG&n_~_+R4z}j{=h_
z%4!g@h=oT?xzl6J%~I?`m3YP~PZVQjR;El*)1rlTQHQLw(8|;GODcxOeO1!ze&D0t
z^hBf~X<N?Fb0-^YU07Hs`q}x=LHeM47&P)3N~H~WQESSJUat--M*wN0TKU6SRNUQj
zyj=|y^VDZozdRyL6l;aeruEu_T*f71wm_o9vrb=($U&7X;&9@Ua2D|ED+ThEnnw2*
zDz39T3oprQ^MG4+*7HpoJHkPJtY_7cdvXQ)YQoW#l5&l7R)DO=6NzY0f~Y|m54g-=
zpX6R1d^;Dfyk-ktvmk3~=TqAryicrjPuiAenQg3cCzpA)%2iTX>szIjUZv{2^Hh}=
z3zg%M!;UT7G!xm#UJcW;xE~hI?p1{tWzZ~#-88hLNAp~ZaA$)E$4->F^0dzXPBwo$
zy%BAnF%Auy;86I3`yi~8$U4~y=^Izy$+AM0(~Ql(kR_dhd0AF_vFj>(>h<e}@GS#n
zAA!yniuN(;1VtUQk#09hw;k;S@Qdp-R0)eJH{}nw4-$I4zBNhgQ@!O(9N(GVGJ)-$
zE}xcdQ<&4cOzWA!2!dn<mQW?l;5Rf>-cuZVep6pTN3!Di+3R{nZtBW-Bv;dq*?1Ei
z>Gq<Wsp0t<E#o6ev|hx707xby+G0C3RffScTa*obRQVFgIK`x&cC(f6=_Ppvpgwk%
zJ${h&DVplWPV)06M>A`3z#?grgXM{BuHMLuMoT&|DPO_G|NPXHH$M0^`GyEhFGAOa
z8ZbfV_mP`*<D(>efwKJWf~;pe%*HT=arp(gZp9(BRgAoevYJtJGeuU@<3NvTGjAI8
zLglVDs`8(hZj2s7hTyX7$0CSD(%}tMt$l_Di75aR7{CKrB+Oxe!Aj`+0;(dMPZD{d
zjS5<;U|`)tFx1mw9J&~-QKV60hQ1=&K9O5J2=IKt3gqW<joMgVJTy*vemF!Sz8;|m
z65x(Mq~=%zQI|D^xlB=vV@Gf!YYSnebVAxE>d4iQ1%4<DmmM(}a#7Z~>E^?d++-&X
z>?e@^qKv5|OE&AqeJr<5*8sYhK`t|gpww^>7NgCOevSug8O#H^lGDptEPBoNn+l(<
zOnx+282X{XP0w~2nmJ`s70@Lc+U5l*B_$=1J5yh#L=65(k-o0wj&;apUzz+cHCb-Z
z9^iEmhlyNO6|T9a*N@FUgT*OZETY{gFNVzWL))+!T~e_OHKjVTjEL$1hG7xM&RhX7
zyH540F(%1HAU=|BJPeH78DUsO(D7%Y1n$U&q9`|~(;t(9h{d*Vpn3(@2Civ-#Ejxh
z<t3Idj-1HR&e||+C!(eZHj-*t$$|s{w~Ea~Kq(b*5W1GK89m2U9XGj6Z&(C1vWiO7
z5Rhi$$XC#p#Zl?u-3xM)A`PB{!kokVhqyrA-*{K92P`n?R*Cqul)-Sv58&<4oq|HX
z$%P1_{dODVs1=+F?t3ji>?4*g5vNiBoOwr<igp$>-}&ZE*mFJXSySSPG8X@hOgtqH
za@j)`wj3sUY7Ia^n~R}~#cH@uAc|CnrYd*flS3m9&U^!Cv5|+2<VJMr3G>9@PM^E@
z!GD+OR+7lo2#qUY{5(cfM=;byHP%DJ^(`|s?T*FKYOw&kI^JMqWkSq8MBizoy4$K{
zox;Ef-s4-2*MkqFxUfY)5yjmpHfKn_>-we+M(QXZCUBWym)VHmQ)s{$?%*SN$pY7*
z!O#$J4b-uOxJv8Iw37~Bb$c4kMskUHN7R{?RgESHUw9&NsDs!LXur|TUUj86p%gqz
zAn5?hZR$fmOXSuBm5K{RhO}Njpe_fsxxro`qJG~S{1q`i_zOK&=~1J{XY}|Z9<i(u
zugbgBDOcPh#W_-3$ii<Eodn=oc7WzsIW6Kc%%8nDALKbWOSouXM@ZDkb4u!8S<K6!
z<?ON-ca>)RmQqXIK~%w}is)Fb6+KGraSfdoo$#=4O(EdO__Q&L6!LaOg$EA1X!=gD
zzA;72xoZ<_wQN--X=_40azH2&MD(R;S?L!fVZfT@$5b>R<KgP=;`njU$gY(SQu+&=
y!$--72vL-l2zH!x1h3Q!=IAy-UdhSIu~r<<a$}nuxljMoQMU<zQ>@a3XY1eoZrrK>

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/conversation_template/llama/index.doctree b/.doctrees/autoapi/lmflow/utils/conversation_template/llama/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..d012c252d328aa0ea3d325990fbc46232c44c788
GIT binary patch
literal 42119
zcmd@-3y|GbbzXb_{;X{hH*{n2LP)Zkcz1D9n#|&u8k4kOdlQlkP#p83eNVfRtoQUH
z>Djv>;WHt@OYcBvl<+GJ2^8o+nPJ*afhm{({Y{5XTPRRIrOXhZO+!kde1=kb&izPN
zdeVK8-e%V`^^D(2SNGoY|LE#S_jLmg-u1ps^dA}XT87zNp4RI1rc>9wfDG52TDze)
z{osLM^7X;p!2}s`wbP!{c58Y-HbIY?VcK<9Zw4m<vID#OrtMAh@k5?o^G&B24mSpj
zL1XAdFky_40pIj(UAjaj?8c()oStTL)STui-Ssq_pz7<5maX~vv~6n*Ex<WoivD?h
z8F4Ex^4@^aG1$@oYd<h{k&$Xhrm=+#a_Z(tzil@4V6MHOkx|X}U2~!B16Pm>Tg<wL
z6c5#G&GVoUjt81rgB=am+pXwiL93l~^+oK{CPTiqgqH$i%osPe85bH`jJ;%W#&8<?
zjN5KDb$8~p<DQ&3dh7;Zh4=#qdNT~M8LjO*TFaba%$lLhnn`8W3}x1g*{ti!)2;`^
zKI_Ql_JXmA4A3@-_e?CtML?W!G4S>h_<t$<za0KgBB8)5*QDf4kp7ltx&e7^!>PAz
z{rQ#LVk0wb$W-nT#^9VEXl>(4V=Cg;obfD>im{ccWTaC&MFSZ<6v>O~%AQygR|1&v
z5R(9#QOl$s5kP?*WZjK@z{z3C$(_Eo(6%+VT665G*V1bRnA%VKlbyG{;??`_e)Uy%
zi@yPayD<VsPOp!C%<jK-_5iAv*KRa4cjX``3^1DP;f?`}ZCbh<kX_PQ6&MjWXKNJx
z+0N3E?&8tWqlb<iy1F`l{jr;l9-6;CrKx(ujW<{4Z@lqnc<rjr?opG}T{wh+BobV&
zN64VR<T~wEhjh>r4T0<mif$N&oHpye5gZvsX*EGwCqW?`p@Q8EN}Ax6@&xD?7o7S^
zDxzCl=XC0Y0W{CfF+kf0_&IIK)mnx$kQ|<Et;{+L*6i&}K9vJ`EpnO-IC%SQl;;!R
zOt!W3X5DNqRhPld$xvR{I}{<<A=~tISF@$*jdyd)9%Lv5kySsEZPnntD4Jw*Yb79q
zt(7TYL!isf4#n{>$n#y``7=EN83ba%EeGV+%Q@r#LMhJB(9C9Fe3%TunZMG4u(|~?
zB3t2HUvAlE&Gdsa^zv!VZ3e3#(v?}}M~ehLG7Q6LxC-VdXkT-ebUz?GkaUl_T=djC
zy#u1w2709<eITC)!QS>5gOS)Dl*O)RiVo%bfU(#3EI4tT=YAZfk~I!^Gj#kN0Qi=K
zf%Cjb1fecK$Bi%K1f)iP6|lJvuz6D)n~7}n$gQC<CLD1A+fR&)oWMH;9%TOM0K>e%
zdjt^UC*1PK5(r5y;AgTf#LalZ_)re(|8JF`LhCm^CY!OfrEk-w{FZEd$~bF$dd_%^
zT$|hdCMibUJQs}d2T5c6fh+*EHiI!*=L6@sJ$^s0JwDO9JzAFnT$QAS?E@?|5hYA=
zDa%&E%%;3;hEt>&>EikTPMEFrVw5r7x$PkrB=k5$H!@-C2zIA8NNMSLE-SWvvx|7M
z4Ea_VSH_SJl{Msu0q2NCTZbv@ViA_M$s$pk1lsiy3X#TH>o@Q)U#PXNW59ciw$^Dn
zP2Bw_Q$2|69Gse(5(b`#xzVjlQM-;trkQgh=0=@+0h2V|ccJu{@v^WRfst(X-CV}T
z-tsp4jxIvWve^&7xH2|dF!)MzpDU*NAmv!xRI#m?seT+sFjM_qJ~&5YwBEvi_iCzt
zL#OF(stnwkn`$B!M?wxAb9rM5r${s2L@a)~04L4)NtCc0i&FqgcKsArh+gdai7pb(
zvg@D0xH5KKFczmV;9N2A?^4#q4IJBw8Ti)Cp@Dy!56)xY-(tXfHSlqOo*!H;16Ox?
z1KcFTnp}1#(vJ&&0j!C`O_T|}7nAsr^Q9HeitRL)*aD-LOd{$q`u5~#xf2;#ZrHP4
zOTz~z!!1|$boZ1_*HT^uW5vSqZFI#oLOJ&2K@-dJiB`Tyk5r7pT>Pm*xJa#7Cm%Y@
zpI>lS5T1vND%LAeX)?2LnM8ySi{#eJ!x0pP^WZM+rtsFCKK{vEB{5!U9LH&3@mrc^
z-By~ex+ZP?+o%fc30t+sM^INCmL5;d^QTjE-8%E?4URkSIQHl$^ToQDl`j=r@$-pN
zYsq3U+FGI3R%#Bh9xdj$H;Gt*Ri~5Don|IqiUiioNzy{AwSK^p;q_s(0Ty0=^=Ky2
zQh<FVG@9~>^Si0cvs0+70Q(^rS0=z7&VRF;%>p#{^<2P8v2k1lw%$RR7Z0)6Ci}j~
zFAGmB9*`mxc)R;a9KmwvhxlMsRKQ_)|M>s|-fM{cJe?+60pj4))WGeobw5L^fW5Ee
zFgflA*Tn4kTxr(7yiuqmeZdb<5tuKyINKL+N+g%?LmZbGdGoD*J=H}cvRuM1U|boO
z@RBuf34#;nif4F&@-XfhuuYbY+Lq8WjBN=$!}s{$Jf7ik2E13#a1lVy4^B-@<Qdfc
zAphnmM0USq3T~aE^u}G7br>4cQf|vl=`E+E)-mW{&0#O_w2){^S*3LcG$+&Tu%IW(
zwWTISqK#RuBGDc@Mr&m%kTv@7tlncjVh`zPj}q_Ws4$u4FH?!MVdDLYa0E@f_X0}O
z@!Tyl@s1bx40`Z5ABs81^Lcr|gxRdHN6xv6eI`phS&mivNaV1T2ucn5p)9Sl!Ul^F
zcHot&XF_$A-*#hVWuiU+eK4(8pk1-b4GXbEWt%ZTCUZM4A-)15BEjApiYYbKBU^!r
zDY!$V=#lM+Eq!iJ&tsJeG4;0Y!Fk86*M`$mEw~7@U=takG$#nQUXNW_WD-n@(0p4s
zPdb{zh$h2v7-#Q>-h)}_k!MKtMW;;tbPcKx12QHKt6H;WIIbkAZpAfNSs}_B_E(wW
zdbtbcpp*~BoArS5M~aYOXVwRjR50d59?m)iJ*fA9GC^QbKt_>KP}2dmknutjGRVs~
zjI(5bm1Y=&WO%V{+mx9M{%F;AYE>xXL9cO^w%rHCBeXT@hPd60xb0<92SvCOPwDd(
zEYhJ00A+&ce)6*tr~MowSxhIE03(ty8J=K|)cfn2uO+GEMK0Dq!bD^UI6>=NtS{gx
zRUP@f=}fWW#8ooXbj0Bf(LBW}omR;<eo>t7f*zbaFYGkdbVQrzhch8OPxpQBa=|K+
zxXf84oG5>KQO<o@J`y&Uo$6I4sIV*GRP%Zq@TO~Wx25*5)(T)Aj~ElyYos=!x|tRJ
zb&RT-)@(XHE#8&1ja+s7z$HG^5zbaFr#!g);J(TM+8>Mr>L+1&Y_<}1sa$^WAUrOn
zetVBFk;p^+R7g#TO9}&QE*W%RMR;vEiYD3I&#XkKbv*STeW!zD=3#BZA>$T;S1m8b
zz4MO|*>UT?LY%3cUff==-T+O4>FX7Penm<3gmpjQWW5!By2qOoiXM_EvVH_7JjkfD
zCz8meU2uB@aeGva8xJ-qxP4H<ZN1_MvLHQ<L<n}oktbmgizC*@;n4a7{KTd4DQJO+
za2EdzTA#*F2+4Ykwd3)9f%RF@o(K^agEl^eJ!y!r{sIn#7Bap}7-4Z6Irc&g%G_rw
z;XQQ@$F%@MrXk@CtN6t->FuFO5AkUwdwg+1`bJb+cEcJ%H^OQ%>lTE_vY<R=u$g2@
zMBJJnBJNc3-MYUN=1|?TNeUIKL94<z^{qQZR!DGIH7Tn>V@CQI>5YC+FkH%`j_nfA
zE)=E8L1KTnH4gjVz`0iTzJ&$emsRy9W{?w;mE13sOe=zQL-!4*4gvT`4xP7gp;Iy2
zG)FFo2Vz0F2twTh@n-~;zH?G<ns@5%Y{hrArnl(04R~#4`X*POccHdeug{Af`zr8)
zil;51As;n8|M-i{nt$B$-F+4KIdKC1ta$LI&1~hy7QS_(*?93p<+{q8)71A>e0X{R
zL(?x_am7i{s3mW*GIg+Wi+)ENq@W+4Z?|myI0a@leSi=ip5TGql7j^q3i-nfGCYsM
z1L0aCc$Kg%t3Y@-6HY~S@G^R<w40_QST+xm*IgSJC8rVhpKA=~&Q#<+A>uO9R~A4z
zoR8jl;w^QR-eyQQ$|K=)BBhdAB1wHPWN+FXB<X8*rdIFAA>-Dg@P`WHyf8OxwVO4+
z-H}lC%?cFYC<6WQE}&bTp#Mw;ZS}#K2!|vib^L235NkVJ&dMDbfz}6PC|&KBQ5wcW
zp1tfL-3LaD`Kl6Uc~wacD(vPid_{q#9I=-Ox}!i}Xu!8@%tf;ne~STbf1{TZXeeTl
zWZe1^V4IAA)*q+wHKsW6TT_{+cuxKvWX8DldlZWG2>b~}{w%etWJVH~->?`WG%{+4
z`2R};nYpT`r0wSn&5y}wW~l%60VR~r!4w^u$@QN@$)o?{YA7oG=TKt$e_mQlVvLTG
zz)<P?vA1OLQ%YoH9gr2<`W~<%%9c76n}I2%ypw<t+}_3IbzCS%-*E}~LrUZ=mAn{~
zGzMXGmtfqegi)SB6wK4jpqdi-atzu6$&JLHJCq<w^M^uNJNE1g^=F@i`@KrsbFe2@
zyn8?mMP=`qTAC2TBT5kGiV#@p3a>Hrosfj^r%K%O2qEq&(xH5|S19q|m?PrE2gowW
z9ZjDpZY;p$DtFP_34T|>{Chl*4zk`t)5U2NM8Boxfhs}x1`O?R%IP`eQ)(zGr_7<`
z38KGHLs1zOhY|~-qY;N0Yf~aa!sf;XRmp-Z??REYyyM#n2!V|9aPJZ&?#dLLLy6nm
zT|(*l?vgBfzZ!~46AIAs0yPwsCNL-|qVC5#cZrf?N*Fn1&EHU#pRzGHrNSJ;J?><b
zUa5d_8A@`M>cODEle1mnEvVCiSJ3^~lStBcU8phFNqk;b0w#S<=l$m?95QYdpze)I
zAbF^R?;@B@`k52y?w-k#uZSiU=xs`v__TUaYukSHlxDZ}IFaz2x7}(vu3z=w>k|GQ
zZ3jQ6LZ81qDAOKS1z9I^$D0fIQdQ&(72#%pb{1mfXHl^EsFGr(rv4EO%^7UYs-dV{
z42P1(U4C8-Mde~RlwO0)H<cL3!d|Cf^ED=;Jly-C5_hG&aVU9$P2Z3z94phv0<?^&
zp{O*0LCFzp_9|iImGy1F_1Icd53z=!9(OQ#EI5yLT%)8`sU6(5P6MUk0#rs|xn2$B
z9+;wIWrs(W0)FBaH58S0a3~MSv?wErriS%slA`WmgrN|ZvaD^n!^p6BsKHWcD<+`L
zS6m@t1|CKM<4!d}DiNK6p%5@)PIizzR}rwpX$0?#8lLKIh`}qxsXT+oGaHZ#m{rU$
zu!4y_QGrufxubFaov``_IZx_y7*?_5>QK(t^yG&bmlk-sl;Wj^#c)Yvf2d@ktUHCG
zQUt10X8Z?C8{$DDeF5>4N&=KYgF#7Q)Bt^WCvoX-l|V{!iNkriS@caM_T^ZVT!em7
z38OTJdRT-mphJDbr4x@#9ohm<eI*SVQ-UbX9|~pd^e27c?Gh#KIrJy-Bp`hycYn`N
z^5iNnP(xA0_e@bIZQ+;_Mrm2Ek1c3Q<Z}o+#h)f@!BK)J&7XC#g?p5^=U`957Sb*5
zt?L5i;a;KS$%L@>?!6cs(vlr(18Ne<v$IQ{!XhQ9#0&}GweL~3)={%~4}GwV<={DL
z7Fn?(En7SbGT>#4$`a$KKIo0Cu#r5cI~fy`_?o?39e){!;a*HVvq~Q`-;=?Kcs;zB
zK-3|{)qvN-53=g9ctyPRWh}}3GSd7NJo*YA{d0KqRXAF;zD9q)PJh2afB%L4ev|(G
zEB+!fQsp`=BIBjdab#p?B2gf!Ka~{l7Zc)@r<8!lLjt}N5<rj267UKjO5oB<>HV$Q
zI>Fy9ELVZQmM=HP>;EX_QdLOz`oA1+lG3_)>nFg}a8Y4R%kO^)QkP3vay_sa8i@~z
zSMMoTNf4f@oHTx0&+F}f9wJFg{3hY<dL!!X$CYFks<%&}NG_qYet{IT>d72R5-lT4
zc0sW~U1bW28xZ0sC(=V$gkVPq`!TSuaM?ZbT$J78u_(L8<3#M4gxK>_XwHm1a@qZr
z$+G)w&!#M@KbDYfl%102ghln;=9Wt4Uyk!J9n1Gl$8RSPh?{-Bg7{luj$|>Q9N8pM
z$5v@-FfJ<bO8dNWN<loeqQ{iU2naSTcfS&(icD7dC+siE-Ld~mI8S;jcRwKt!PlnT
zT?ELm>iy<5sCws<hIwT?_z6CU(BD1y{r@d~zn|-j;8ntoiWI+-a4H%Zry(-F3v%Kx
zS2#xt-Q#7t#S7cxi|bO|I`?N%OB5U5AMz@#l3n=(=}{P(^BM4;sG+Ey0dpvM)_y*t
zhN4>g;ZS0aC%RXDQ^xjIwf;pV&aogMeO13}ac_6rb1y3LNDl#F-1>Ve^0;Z{(2?wW
zws>aCtJZ&?LW!Hh(^a+pQzieEwEc*o8Gne8UT`-!s=PN)=|6{(NB_@LLs98JhZ58O
z-m2CwQzBDZu?sGe%F6p}z^KDtr9X6<RU)rk1!Pd7G&CyfPgJe{mJ&vJ1{JGXFUg?W
zl*pH3P@;I#Qi3SWpJK(EUD=cVtm8FG+;gxe*NOvCLs8j#rj{mz@J=O&b43X0i|ij!
z;+{tcao3&><<EPC5)Y0!B2IjOEQ8$9bd9UlzpCbes!roeFto!drw7sRsiCNXD2I|K
zi2hg&MP*bRN-T($s9GQ1s!A4QdB+w5y1nmJ-tqMVLLj3&+`CYTyK)hfLy6m5de!<B
zYA7mAC_u}#YA7mAU{KPMfJD{$4N4e!W!-Bbp&aEH>Tw^VbV>zm>vkozO6}lKVs&a6
zg@jEt6jha40b1@-Ls4l5hmyaL@J(u1B?<`{3UMjR+Lk+#tYtm=d(~vB!pOrgG-nw3
zs2YmOd2lFs9N}3t6qR~#D7}V}e^6o|3wxcy$mf}i@^J5QCGJW`!J*^{BR^C_QE5T}
zTKcxB!g&r_a)gmFC5*hXUdu4T7!-Ff$}m#EwkDO-Dmlub<Ow6!sG+FBNC8@|S3}9+
zC-R4pThy>hgb_w)T*|Vx<qjj|>d|dAK`Id$Fcbnt%*l!ZbiQ6LPNT*0JJs-10gAyZ
z#i=}ltVf{)OD;T5c?R-6=n_5yc^XR1->&42tQLg=PKi=;IiC%fpyTl%ebw*7N&=J%
z1`JAyr|u{<|AZ1qX)YBlH807czf)pgjz!7x+OI2Nl;%(mE5`+N=s%Unmt#<(y!L;U
zAWHM6Sb1$%{YhUf+r3?x45?DCR3W4<zgBvNk|#->Rzp#R?o4-)w1opo7^P*sKDKa^
z68SZ;g;yy-l;+R6*n+FXJqLSom72e9U7)<VS15Us9IQ9vrRFK=%?8xrk;&}JvSekD
zRKtaI@QU+@j*fD@H^SrP+~s<-B9MJ3_ah(*QE{H5OtQ1^{L0)!b=IFxLy^ELf20oS
zb?2vd^VFY<2}CJSTsOMcpZ_ItfYzV?6&}(0b9z*^e4EyB6{$ZLWsph=vg^;k7!vSL
zAp!KLECFj(fBvGx=GK*R+h-Kqm-g!F{Po9fI(lgS`oza&ix-!(4SwYbq1JO^Lkg$A
z52BTe%;Qxxg$k$tJ)DXbyzK|eixp1)2VQ4ItsM-A?r#4loF%OkY@$d7UJ61R;GQj{
zJKgU_2zL10$6-)zzbn2@j*vu|G#-mGX*@=EFKuA4A6(>jp*gcmI**GK6DS=_%9a)E
z1)a$q!a-hvN7O;mqmA}2#}fV}++Vn)qbayiZsZq2z&(4M3$jS!UtZXVcQy`#)Yr)M
zyda#4x}K|(uICV5XRark6Oj*kX*gZaJ}3pVjqpLpfOH=eA=u%AW?@iaAB2#E55i;N
zgYY=vgRTZ2bO4%r@<C#Paz5xah%HUhUW-T62hpRA_CYMiiT2Dc>3Rz0I1#*_Jq`=;
z@`M|f<Gh{gm{5@(a-1NXiaMI-B^}K}cwNqMBnCwO<^AC-J^Pmwgf_yzAl>QyB|@;n
zzq|zo74|O(N%$8$7XAf~6aM9S;9uSf%{}=SF@aE?)Z4(!9Ony&E%h&7#3Slo=+Q>|
z7nb8h+8L%AB^^z{947^E&mM<G_SBg-%=J9U)lX<o53c9m!l|h1xhCm)p1|vJj+4fT
z$OrvAoUUgdl!Dnt_#k9Jx(|vF?C?R~hCzjW5JD0@2#<vi!sCPwx(0mEcc8f^A0#Fa
zT2#yjO^gX2G>J#l2hpRo@IetOWQ^^$8151XUnhkxRp9EkxKt6F1;VAv;nL&@{Bndb
zYK$A(*qs1XZP-{fns&lCK=?jGQ*Gb0J)A$>@@fjU(`nh7uTR;yGZ1_SLV(P@H&G`p
zTO?dItKW$AoVHuj17jaKkM0fxsKCZP)t2ilxv&>da5rq+2wyz)H9I)6iHwRqdL2&N
zNP8dIQER)f87X|VmCZFrwuu({<=8ng%=c3C0{B?2Q=?p%UUcbRll6cMz6`$w8-2o!
z3_5MU)%JrqV{DF$Ea^@DSzS1R?IU&1*BY&04qm--93N1Hy*QTx(l-;p&Zrax{GQxN
z=rGDp+pY~~+i{!B1+8`xufkTDZzj7JVFK8NP?ws|knJ99#Mx{&s>_S8RZ?(<?0~Pt
znyqThX*7UL52hMtExN7Kmg;HKH>z~MLLdVm@d38x!KP@?u+7!!cWrI`ln#vCOhye2
zzJbe62gojc*#!Q<c8F~+Y8|NS3+*Ksz5}=7g1soAhpTH1Xc+-P)N~{U@iUviE%ur(
z?pR4i+Wz9ywNrF+(F@}jtGeqtZq?A5bz66V_lRoE2MuRDmu4BDvjk*s;!0u)YY~i+
z>|pcKowX<qV)TXa3yHyGC(_^nqYPk?4KxL0yR9{s+OP{1a2)q}-K;kQ@(fq6Ic~jL
z*IRnC4q!~m*g>$RMc6|XIAYd?D!?A8JG8d%RHG)c8MbSMj}NySt!k)@z*yP<pCHp+
z*vtwgHAY*gnxpiirk1C->rQpKfvU7qUv4>`UX@2uEvdnWjZM$QZ$QIb)w=0v3pNtM
zu^bXc0tUAYi1A>HQMP3fXbtaHT{LYV4!q7DJOUt|?oz9S9e-iAI*5F$tDOc>vC|`c
zWEcH{qOCWVKoM=TVfw+botKjxExlC*`_REYYfKZd1$U^cqEV0w4Bu~gvokZ$>`k{;
zU?;QYwBs&;EuD`&DFV>5D%?~6JMrkPsY{>xYrxMP?qV%5J+?&CUnZu<U0+dUPwMWp
zsr!o*2DpM5!*AFCY{b_Vz>4a=5gZvNdqF1~)zCOY0{EVtt7NCQ()6|Es$nh}Hv9+u
z?YA9N{cZHq(N+HK=riyw(CAarRokg)oV9yv27GR^27+hy23)2rgx)9HE!ci&2Vm!$
zKJcHpfmP5Rr)D-6K?k^B*o}_DYcC|anGMuk+tKRO5c;Vz0+nej=ykB*6~MsIb@;#~
zIll(IPFD4XX4+NQZyG%=Na+mOg)^fc_H=uZd9zirMQ@%8=GqHnn0XU0U(iIeSvSEp
z{3_eenCi_gKngU=2S*EYz!$gCkCWZ8R@}0SHp6U-;mII6(>iS*!bWyHVa;OTy#{~g
zal`bN;?J${;TY>?{J9D*%)n2ZjNmA`eJ+L$I&PJYS*86~X<w+Yx)`tGgb+9n81R{K
zz8!P)nQ{Kc-$l^TxENyXfiN12&`SnP4?=cu22+*~<6La3z<v>qA;UyWje`u$mm}7Q
zCfzxl?!!zs!?tiIO}g=_h2N*5UmLaXGanXy2E@WAh!#F!v~W$^!ZmLTSC%YXd9v^x
z(87B}3o|hbGqpC^F6%Dk=JuM&HoHZ8VJ+yQL_a1rxrOxuCjh$MSa9sn9B9|`5{+T+
zk=TqQ0xBJ}J2pQ=8`lrOcbuP`i(i3o4-T~Ew9!36Y^^!A&D0X&_|ReA{s74N-r#0%
v2Jrn>a4X>+SAhEN(2vHNw}#DBg}0i`n%%CW$t<DQ8?B?O=U$PeozwpZW>nJK

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/conversation_template/phi/index.doctree b/.doctrees/autoapi/lmflow/utils/conversation_template/phi/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..88c5ec96e55ba924b6b7e53de8f8c2a380681bc5
GIT binary patch
literal 6585
zcmc&(-H#kc5kKGEe%|*vK0x+`i%{aT$nIjWP;{IutSAU%dlMlGA>?G-JJY*8d$Tjc
zOwWDDvIG)|ywD0ro;(1G5J<cso_Wgye*xkj;#b`>v$K0Xdt)K6w0qmt)z#JYUFBa?
zKKuLE7W6;05r@L{Put9K{J`OHs_IT)CnN65^jUi2m+6yqUo|3j5(i0S^HePW!xpaR
zMBGn5Ow~G}OV^9r+4)+`ZRrMnaa>eHRn$IA_eDchq$@pM+EQ<MqrMlMv<(eA@Q-;E
zGa_I~J_<c1c^fRIL;(brAMjJ+Z0>U(r{)dSkbz^FiY?W!iYJOCRn1=3Rm-01^R%1v
zm})X9Be$1GNQ5M?<T^3Ewq|=Qjse8)iqA&oSKUd%$)d;XqlotjCs8%Y2DFum4RKYh
zifdv?>?)@tf)VdTiSP5Ma}q>HoqP9x05Rl$z+l`ljCEKd0~Wd+BdCrRROh*Z>OfE(
z*LV17JBsnt*Y7SSJ+YuF8dj3JQHc0Do+jRaT)%<OoA_ME=LS6!l8RjYxP7%8GB-+9
zdlWc{$G^8Xr@~ar9B9pnFRERcvP67Sv?h}3ir1kJv0@ByVo94hKsE17G^K51X9|?5
z9~AEx*&B-N!nYFxq^jEnUAzq`*0mHjCF>;~i!3|vtT^QMmk{W(##fsUfBN|5lgHnF
zlK)NVy7wp7spNlQ83#9S9UPKb#mQ*IqO;pD6^L2wWWfL~)-G+QYOA!eAckodzTCvW
z4}SQg?^qAsyZ^zxI}hG#HgkWi%wlJz*HBeCh=L?MCqxPpHHi32O1O?AC$1w!dbdeJ
z^&!+F=q<=C<z+TsBJi^^0Sa2Zz&Sg|DEM58f|GM_9p}h_o_nN$IK&1K3q|RmdjB9i
zI|zEigGW<x+&i4ldD>H_Za;dc#lH{NUk$nMxc<O8MMx^&*!}O`nCrXdxspfBD_tc%
z&-`+2ipzB3nt#67H4%O^=V~!LOI0;IYe7mWFGnj<)}Ns5KO%5n=+>%=_amyN>ej2o
z(-b+7&X1Vur{Zs_hLt=Ekq?)MAGLxtKMg(Cc4hihZ=bNpPsgeO#hO4kf6h{M9MYL4
z?dppqiw0bR_g66o9b<lvtn+`1Eq5<LJ)glDcSLof_0P&$=ND>@)cZ{Aihm*y6P@4D
zu@|YD4txs4-{8XE&Uk-T(d3EbC6`_m|DN-r7qdSGH=n$W(yxGofORn?wM<blMNY>9
zU10ZTNvbQQIKAfu{=h_EHtQ}9PCuGGT}D!k(~)-&hm1<TdKmGTN5@=S(nmO(C&!06
zZ%kO8s+(9|R?I-NB7<Jh1g)$lDiZAra{b1AU4rbr8~73dHa0<LN1cNgMmF5Ow6IXN
zvx_4H@q@%a+$c_HmNsEU?HMbYuWqWwh`Y#?>ibi1Z*SlA_e?~Xvy{5=s*XQZQk5Q-
zRaFaW85RHeGKTWTI_+4oi{e!#(Nx7g6UI$xk4&RaP#;0%Zr(eHin=~$(8{$rH~vJ+
zhlS3bSKjTb6+EN`E7q*3HR4o<SG}G}Cc_Fy4XgNqJq!<s?$zOU3-a0|wqX(rs-fRI
z^Vs1#gi5W-j+&9vgxhXG?)lu-iCe>Joi2G9m-jB>Qe7?;bjL0~vC-nCY9s$?G2a$J
zR8rQ<iX_T{c{1u)#){)PI}>zFIPE&A7E+Or#Xm^WAI-kOh-MO6hAm*|kb-Uu0!mer
z#DwgNc5}KQP}MBqiZ4{fq)|~-^?u@cT9oGgrX>U0LN!Kl<O>b^1ZiGFCuqdmE#ht0
zs2<jEla@LMQGKPY0Noe`r1=F&WFMO+=j6oh?2{>e?wi7A*<q5+zA`IEhGP&?HAq5N
zB*S0RQX5Wr?FTK>1&&qC5AvVuIk;sC*0Ea6HuLosFtX>x1=5Be<ggr%DA9c}mlAO=
z9n%v#nW*xxdhUs8!>FpbBAsN7Y-=yS30p5n9f;EtW7!vn`GBCM`(>hE3>I4XV4+2W
z1zx17c}EfF_*gO-Mq3Cv?u&CwF>DZTtJh@2{222M8g+zGFhG(u!-k3sBr52;3#ysp
za0kl-A{Cd^x}8Kc%CL&T#cB>Q0}NHqlz?f@RxzIGN7_goRM9_SUV`R6kI||cBr;57
z+7%n9HwPR;2__L2uz=n1Sh7)w#Yz+e60E|BPg8ZNgR<AKVBy{)AT+aO5_wpyk$=%J
zgm&R$pQ`OX2>5~ICGb<VMgt>18Cj?OAR3{1UXL++2+?MbXbcsD)KvrF9@h}-#FfIz
zM=`i5JCMz0x@tAzVGv2n<HsC=TvScLykc2VTm008{3LQ)oFi3K#o@hVfa}(&cBiHZ
zbU7OV(tsjtPLCn}Tp!Z1xQ}p3rmh-^?6=-&>8NyVdeh=j6hxL_zT<IBQL4l$0+!k^
zGA~i7Xez1No!OZoG5i<#_qw4w;UPbLZF*zou-c?I!0KWF5qY}2Tl1JdNZ5cw;*=g1
zG1ya=BW?$gV>z4}NL)kHpsuRIkNOC{kcjK#rhwqOPKBp60jWjAc_i9o6k3HbLa-X(
zB3MQ~LoM4?8ycd`oU}gyg)vW@z(N@ctWES*LCmf4No^&zw2s}_HO9JOIc`kD2|Og@
zvX(mu9Bvhli4mK$$ARcthGz9$Pmj9P4*kL<D0a11qFw+s2P$8|1Qe<=%e$A<7I_*h
z2Nkw}^^ef?-QT>f)<Ygz)C$5sZ7neL-yy6WypvO?uL>!{_@L7PIBth$QUrcGhz9Vb
zD}<>T0A?}!r0Sc8tvA1Q3-Vl#`t}TYf=tN2m6NA*K_+{|qqfUsU*CaH&=GRv;j#uA
z19*`hm(<lRY;t53&RIMFQ!cWQvD%E!e92BN;SPj{5A1iD+KW`J#uy+;>*oo)I)<Pw
z>$O>0USP9Kw7a&z+{1?AjlV(4T7)=#gdxya&yn?TI12+~SWjTPejhfF#f5DO6mim<
zL32lByI#PIGtw0ID22(4dfb5rpMe9x(LImV6&qMb7DpHFS*TPg;%ct9vY9n()$5yR
zHdagAKc;E0s+(v6_k|@QbGq;ii7^=U<a!dk1*TwG5~&8U+@aa;lT>X_L8+wBU{LE9
z3z|yMXc_1=0-E3Z!@t7Ehkv23Rr+es*XQ*02Ye-}LAW~G(sWv5k2U64V<GpxL2x2~
z{^}5us%loX<uU(!2Ygs$-aO4>ZX6>$r%x%Be`T>Kf|m2HE$L~_#BI%%ZZLG2rVHjo
zt(7fGzi|!Y5tH!nZUdpfk>eR*mLX*A8VU;>^)PpxLVaxpSYS7%&>H!zn$q?ZedK^h
zrts)1v$V1uBw>U#J4h&KA>*M3_n>|#8rik>ZU%n_J$#)07%qx}63z}?$FNFsVG-S?
fz$*n<8`g^J+g{?3ArI((7n%+Q;4D^|a<ly(-~pnz

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/conversation_template/qwen/index.doctree b/.doctrees/autoapi/lmflow/utils/conversation_template/qwen/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..9a718cd8d0de054e7dbd850c5025d6c021060fc8
GIT binary patch
literal 8759
zcmd5?O^h5z6<&LHW_M=)?W{p^Hf%t_6Oo-sB7)GyGBP2NknJ^zED}MIPH%V3RQL9D
zce<<hN09{qVmzS;Na-9B1PNEd0dAaPAP0T~TsXpk{~sYC<-ng1g6~!TO!w^Wc)URY
zOViu+y6V-dSMPoGUOj)Z^1>bO98>?~eCTt_-D&Ek>3Sv$V>xAdM$~1lh@X#de>%P#
zH)Ji)w?i)q3>M2V^f0*Pm;rO+$6`50-G$|Z&2;=^$P8h5Za$n>c$H5+7B_fJR)i%S
zR`?`OIo-D7Z8sAhhUadvAk>M0CRo>ZbitZWZeuQD0S*(F*ba#{`oL%6WEE9khs+}8
zi?XKWi{#_7nyzhKPB@m!;`L}lmvvo;z}koebV529x6F{%J83w27=noRimP{%w<$C7
z`;86V*bG>k`b2V4=pFhL^Lc)f&+wc1IA4~m#l0?T1(EBrptbD<o2~N~9)KdUKM*i%
zC4#l|NO-z$wGvgel&V_$s;Wh*YFVzycA7zmwcc{>Xtcq{WJS>micj{0zXNOIw?eb;
z!tXZxPT_YutrMyWEVa6ZobYukh-ItmnUTZpT^V)HWHTADI_dykT^F$)@%Ql6zOvT&
zE!Yx1omgYvoE9>Ite@>~ma>x*B~oeu;CMDsf5K5eaP=er?bey3JHG>ZoKkvR5c)>s
z=z(T<ju!gNxB-<;sQz-{(MO)T>+(}~U(Wu<G~LDibaLx0F^;vn?pZrcrWHort{&{3
zfwe%-@<a*=@Uv#&b1WANEe(1oJ94Xze-A!<|N7n9rTZ^@<owx7_wS`>4_tgmyL9p5
z`FcHb=*l3c)G}+bDmsA|`Fk`>Kr#va-=O-ZFl5^@1&`0wN#iawzKKosGBdiKOB5J<
ztjK_1ZNoEn_Z4(J@V58;P$37~I|gg30jTs&p!>WqP(HNg@2+_p_SzF=C9j+w(Yc!C
zac7=*RBb^6=rH3m*R<S@wgcEIPWs?~ebQ{(Cml<$Kz9nW@h_*oJ6D!8?)&do5B6W6
zl#*T^^><@g^><gHte6!+jW+FN?E04g@CW+7tU?Ds-&o#rDBO*4ToQw>ZsE}Sx|~Ey
z?D{x5$4MYLjW*ly9m}vpd{uqk)&n>0Vc&Px61=>D(@joclsc2*bv3ilgANn1eBa^B
zLV?kXIHCS`Aaw2s_UHw8<_TWyZ~Y6!t!D>rAMNq;e3`!v$R<8tq+!GMn+80Gj-SJX
zUl@RbGyuwuoa;<_lK*tXiH3{cgEUtl&9h}`8V8Ff%>(>2$J@iAM9!cO7cXUEN#iY;
zHNlHlvN`$RQjB<886$Ez3t;h&qQ#UQzQJD|0Vn=4Y&&zD;NL2GyghzoY;1I==dbZ=
z{M+mNJMsgE^Vj~u$&mjK9QTbu$9<!?4Z|%Tj%zPLf2r^Oc~sy1)0_8Q`!>uwY`-&{
zz+A%@k-G2jPM?8~BltMDlboOkCr0>=ZMD$5Jt&-9D#T+ej^}ogNJ<|(@=T0=VsLb^
zldSD@owd-{DZ!ia0~WGii>Z^~ag5EP;73(_*cZ8!Bk5TNF+ff830g%Rywdz9kEn+P
z_45~0wy|=b=L*D1VS)rF<Q{r{M%X6Vj!LjbjPPweAE88AgdsI^tWjy}fuLVbk*Xo$
zCq%^ZgJob^X;|(`a;lG-Hr;;s!@HC}6&9DKP-AieIsV?9g<`-t`l5vvl2s7}r5t@Q
z9wP`=VWS#CR|k~eX4eJ0A{(P(S+=L659n|9bNSZox7iIjjkT=8n3ZsHmSk0Nl4_UY
zLi=H`vLC?%W@<~~d(ys|vsadJ9F{RAYih*<%e9}Su5>+dLJq2`FYYTjzx~CXB5@O0
zr)ZYziTRa-#FWPi3EY;&whdhD#Bx4+)pXb3UQpPt>m@!E&}L!fomyhd;i$I}xF*4H
z%Z!zp3a2LN`Wu6bNOaT382c0Gp|Ay4Fwjse>!c?{(zs<RNB6Rt2G9JOtRxXJuga-*
z<Ty%~$^3Oqc!q`yivZ=CqFq7AuBiQPB-<5|Z8_O|*u$r(sbWb=n3Nsh3WktrbWIZ5
zA16!B*zx7TB@>V~lIUc@)P+8{$~3jGe+EW!5}Huih5d7CD$6OhU2iqH`RU0?*UMh#
zGV*GYg!bf2`Y~%SqDQ*CFz+<)dKoS2N1SNBkO=_*jC-`ir&CoyYYJb}B_0l0?aQ>V
zb7fS`qn^XT>zj_XhU2XLDTrJ?MH=>J3p9CP^s1oM{(Q0T=-8NI(C7pD==5*Qc<{;D
zuwTS?`z1WC51jwZSt;LHiZ<+*LCW-D|BJpSXnCL-`hv|K9WL0{z|DRYk0FAkP`Qs_
z-z*6B7NH<5e}5G0%QpxK_Fpk5K|%Ywc(q^0qrBknfdV}HKK-fMKcG%Dll{YlPVZM@
z`$rjl0N|G3s(+`R3gGOY;5FN?>!_swX&HkXg*5vYB-I)=+WsZI-Aix3%HMvCx1RkQ
z_57`R{!TrAubzKU&p*<02=w$16gi)q5;?lX{m3=_w32}wJ`D+95i9oDHu&kRBC)FO
z=ZSw_%2>@x8LO(4!SYyn<m$&BNyZazRW61r{&IR-3huM=2ERkzBm&(HQB$GviXV6#
zg!)N&qar+k%eZr6vYz!}CYlkH%8$r7BMPW^qUEKMnT(Q@OypF00-%Dln3sgwfjS{g
z+{b-XcOq01vLI1by-4_xh}Zc%?yWiuWe{B;DnSE>9t(xu_0gCPJWoJWL=Zc%JkrAb
zSt5dlc{kCao;IVvL2H%{cPhWoR~+!q$zyFWa6Q2a<g0R)id1gY)ppum(8V>}T!?Ci
zk6ZmNmA*o-I+;+I!xTl^wglI*;tys@2c*@YbvY9--wTB1uq_5f9+h>jJ9)Dr7uk*l
z{Ru?ZVb-Z4D<<2BI+$*bE&=HZ7JQlB1*L|KFc>X{<g;98OJgpAIXY)$EfVe3_phpu
z=;rcAjRk=hXk2$qhoNj!B~cMA$@xU*1uhlEC6>zrUj|s%J8(%km+<Y&kd3~%{9$0Q
zT%a|;>Ou|`IV#DVb#%8A=^X}*Q*e3|1wMH^V1^f%n#t%o8q-kuX<b&~No_=a(1>Mb
zrhv$9j?!PPPm)IwWFpRwy1tehBNVF$79zF4)sa!J%XvjnW=^}`C;1_ZOix2Hj9v@4
zr1e6k6-O#7F@&~dg;rv$^O|XeR0_dDQYj}hCxOGwU@;*ABjs^mI-7RW+LohgYw`rW
zVG^X=$}5qsfSZXepGMUcTa`4=9g&OVX|NpR`5e~Y#r5%pg;R3QXTC;v%J5G^DGayy
zKCB(GlT*mIb0Pe2t<?fKZ2G%`dv4PUI`E|>>Zt@k&%EABcL2;^z3tBTLZ5R%+ZbR^
zls@}+XY46fC6nD`LDOQQt>yp}w7BRxn5>5D33!nzz)Z<y*ko7BoikqmifUvbJ-HC>
zx<cR4xYgkfez4yOy5xxEOo(!o(4LIo)gcshT(y&?=6Hsls&?7nD5MzJ_{86!Wu-!l
z-b7icr)t-#Af3WM57y%umfMC6q_}X5fFg`G2D({YvR%j16KABl@n#H@>25F+9=r<)
zIKwS`Pc9kgwW%?b0~`&hG9j*#dNHk&!&aSkf@VEA&fG1kKg+2EP2j$;M1<`Yd_$n<
zMtA$F_Phw8U|9l(4Y1s#YV+k-9xG8PpHXCRYv&EBZ%{EC-6x5dyf^IgbnJeZ9*^TR
zwLeae4^i_h9+9k3Ulr<8JzVwfsh&O63#ad`L?;Hg<vEQyRyEBTGtA$6jea^md$WTU
zh3XIobbOT#^Vf)1l7UTRcSndONiu#$iKX%jl_RQrUByW@#Zi%*E!vh^?JO!T$<YT5
zBou;>tY4zmB8BW;QDKYS4HSG!U2hs7=DlmBZnbPy#pzgyKTd-{#PIE<L0;(>9LE4S
z!;1*MaQ5L&(19Uf%W=+5-j~v!$BsM`KLlq*UJJL!u7|MCWWpThO62LBtW0d$at$Xk
U$)r2<KO5>ML2-&&Nl@7Q4<5Og(f|Me

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/conversation_template/yi/index.doctree b/.doctrees/autoapi/lmflow/utils/conversation_template/yi/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..b933c11943268038ff94d1fe5ef135407dd2d792
GIT binary patch
literal 6577
zcmc&(TW=gm6?SaT)t9lI#DEenUZLzdA~QoEpp~&AqSYdyB;FN@5TNXKdb(?-dfYd<
zt8I^>NFcGR43rk6bRGb$&?3PTKY+J=;1?kN5Z|fl?&%rF$!vrm(u}80ojP^u+|S7`
zE6@M-wFUK0w&IXG{%OawZ9lMCoXWZ#Sc%7ckv>bW|0;cw?#V`Ep2R^CSuB+cz_7UE
z+7a{94^z2L=)!U1PIkT)GfOyuUmWKZUgfn9(>>mh72yb%mA2$H*Xz5%Nk`ML0{@sr
zu}K6B!MxBl1?!BQlnCJb@&k5CjLm!O<5a((8k*o&q<m90jN*ZONmjFW9m-|b@mYG9
z^i0__g@~M9A|MZX|B_?J^wyf?nsE#ueph_c)4%F=5>6I9(>jV+pKuac6Xt-nQr_ZM
z_$t53m-vpfx;*e$H%fe;MctDiI_loL|3e5N{{sf&u4b%jCL%CHr>n)(RbuMCP)yyC
z({+5Copz!aFMZ?gV$$OavZ7!mk!x-6uj6I>O-S_{_*}#1ZG5iND<P=JQE%Im%c1E+
zscd_Jow)4#yYp#F-q85%`OtawP^4zUzscJZF&*+ZpbWmEZE#{l8#zEW?@ZLB>|=Wh
zl<6K6?`Xkmisrd5Ck9AQw>3Jy0U6en3^#<?OI$NDtiUzmkXc_soXZMdZan<y<C{+&
zfBQ-PH>K;|pIj%!|G+HvZ{FHJAd`v{&oiUZZP*E<EVr{T0PpIQwo|!TS{abSwDVqS
z;@{7Hbkq2*@!-AtAKbh1;Js!u_tnZQdS-YHSrvmQNWwEhqzF-ih`*$S>o{`a*n+2b
zn<P{pLOp`sg4|MGW$`5fKQ9xYc-0H+(HTa;=Ta1$oPnzlNAB~&BMk&0a}b#!FCCO0
z?1!WMpf}uqG%d*8gSniiGkxmzqlZfVd+_|#komUb4~$a;r1F*T|Cg`KHC}Ta!6MTw
z-Nir4Jac`T&vfFQpP%oYh(C&VxfqU8Sq(>R2r6aea7Ft1GZg+O1nzS^TUH?fMAcM&
z=QZMKiUdf<J=5`1{&!i!N{m7z!zE%zu3*hhL)Wn!kv>)1CuZcQW7&XebtIfSW63(s
zs63M%s(XbQ4VXyft=BOJ6=eQ^ob&$*E_W|LH_zdW+q^na`SY^M*|};X-9F<x{GSNK
zMCbQ(>}7hU1D^u%x47_kGv1#SGkGF;#idvHzvsN@<?N5a%_py-^lKm?V4Y7%E>l!Y
zk<sx$2iW~tlJasXQ17~dKhV+Fob?qOrytFpE+fgtspsy;p-I(VJ&ah)qGP5E=_8!Y
zljB2`I3_Gl6-{PdRLnrL5`$jR1g)$fDiZBWa((N*DnNGM4SazB8|#p>EziIUBO7jA
zSXe0A+4<pt_(9?ywu%#qr43k7XU2-=tD34Y!Y)#!{K0h8+ud{gT^$hStE8^KuHR3U
zRHcVyRn>xAMzMdkjG?r#PCG{IpmG&SG*zt6260{1Bh#oO)I(6QoA(X8BHx-XXyw|R
z7k#4SgF<C5s_ypW3f|F%5i7>z8u6)ut6EPbl3@j;hE;sQ9ESTu_sZ~l1$ku=Td;@)
z*-)>YdFt>TLZvQcTh5AU!tP-~@5SuaiCxWWoo;y*oA=IRQ(i1pbjJ=mvC!S6vX%cd
zOyA-`RMOR}iX_T`c`oW0+KA)%+7fh2IPBP|5>k<j`9Da~AI&~NOJ)*ThHYS|h=NWG
z@=0Zrq=eLqMsvCtkku^P@-JjXCsAIN^?u^IN|O5irXd2$Kru#P<O>D+1UX(oCuqdm
zE#hrQs~*;Hla?w4QFW!P0G$}cqxt1XWFPA%=j8a#?2{>c?&+du**1ka`^u~q8ID0n
z)*uO0jtqZ8OJz6ZwI8%~3pkcFKgfTs=is)kSI2TS+sxOSz{s8#7f3BX$YD7iQKI`|
zCIrG>I;JOfGg0MH^~@7xORK85BAq0SY-=~a30p7890=1BW7*>e`D~!AdS#+tj1=1W
zNTH390xOc#oHK|sTrBDIqU;0B_W22>7&Gu2@{)*5KgL{xMjT-j43J^<n4uy9f#Uh@
zf^6nE%*HZ-K=}o^ZY2>7F^pnlv6@5l=t9;r6<~_9S&U`+kup&m_47|mH$h*Y$7fj$
z5)mdMJ>)IangfP$geedduz=OESeRaj#Yz+e0<6M?Pg8lJi;7pXVBp>(AT+aO61iBd
zk$2G;gm&R!pUSO12>5|uCGb<ZMuQ?h@r=`c5P2w^*JI2bLiE@kjh|wWx}qV>WeQ@P
zID#Ab5C%782ePTmp<Inv7(~Kw*)fA47iE*1Zn3P$O?K))egcUt&XFpzVzXW{z;)|X
zwo}6dy3F(dX+RMcqsI_`jt^-W%ttIGLzj(2^xN;WRZO}%y=ky03L=A>zU?v$P^!c#
z!j)`knU|<k6qQu&%<RmN82*dAdtK9=@Q|OrI=wM-SZ>f8V0AHvh+I|Mt+}Q@NX!9)
z#3?r{VxT84M$8H#+prlmkGO^gL5H#mFX|)oLL!cxn*u`XI#r#<1SA&`<&kC+FEk2c
zgkTlGL8y#;6NT)dY$=E`bJG3<6viyE0|PZEur|<E1u-+qCzX|0!Z>zfM;mL)u$`Dj
z6L?6*Wi59SINT~86C*Myj|0)Q49(~}u9|YmZTf{vQ0XeKM702FHdMZX$tP5$mv=A7
zP4YBY4$5l|>-W&`-QRdyu7@l%s0)ODT1sH(y+c?#cqgZjU*$rC@qV`paNG$;f(L#l
zhz9Vb%Y>;I0A?}sq|%#(?Q7q<1$nMVeQSn1K_=wi%E?o*Ad@{}QO9ATukJu7=yKt?
zxU7M$0A8eKC3U$2oAivrIg1Bi#6=b|mK*WN7v`zKodI|8f&DI1Z;{H?7~>;h{33x@
z#}L#-wbo0+4J<Pg?T*DU^su0K?Qf8>5+P0>VFEN(^J6s>&ceVL))QEc--iukabb%B
zMV$0z&>W9!*9}bVj5NPJN?|fykJ<3x5jfxso%2{;wt#hHFf{P4fhv_EuHt$#n^nVB
z-M)@yW4XlqW10iYx{fArUsxjYrUTy)7=lqhu4cfSU<#HckZ2IgZJPQ%N#)iQlnM$3
z2DN^%pm_uhm4RL(p#Hr#{40EX_!s(GrLP8keMVn@#8)C4gsXBb&7~FgSYeJ87BcUf
z1SbM$t`0D!s%9ly9`nz(zXwI$&66xf#xas}`jis+R~CyBXgP1$lAhv>-&Sm??m`u5
zs$5RwTG^u18`m%y(FqUl)({FDIi40~8A8^sps+x%hoS2f>ZKWAf!&xwYvi{oN?TL(
zkpm);!lN(G(#m#_gb~)PAfcdzjE4r?h5DgrWY^le8T=Xa@NxQMxG2g=I6HJ5!z%TK
jMRc12uM}WqSSyZixrt4NJfQzqXxbEjvsk5z&CY)SGf<x_

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/conversation_template/zephyr/index.doctree b/.doctrees/autoapi/lmflow/utils/conversation_template/zephyr/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..0d3bbcb78c41753e9abad14ed9bb00fc2d611fa7
GIT binary patch
literal 22686
zcmd5^U2Gi3eHTUY_@NI<vZY3)WnY}Cq8yQw)uw4^Nr`JaNFa+gTs5{O%ctetlDo^h
z-Sd8kWGn+NQoCj{u-#}~H!0k@C=8=@5CjN}#C@on28JK|5EOj~`p}0y1O-~8MPV3d
zi~jyIJG--ccf-A-EL(t%yR$R%|Cs;ReEpvtedYRNJNQ2~6|^neU7s@=4cBX!LCnS*
zUOj4=ZWv#WkN<xBOgzITd}A%}BEN3NYzJ!8ExXz9O*g(Av%OS3w41?P(ta#3>!Izr
zX>)7Ts#s%};~8s$jfQsEG?h>6c(b+K^w#D?A9c^YYWjge9n?az)ovQ0Irnw5ZEg55
zb%H)dE|}|tTt7{piv=Z>wgJ4u*xJt~YAK=CFsmf=EwYiO?V9mov}CYJBMg0eDGI?C
z;={1r2#DsfdeaC36ykZ*HCp0nyb-mtmnEaV>YK|{Ct_ovu|l6>Ys#9oc3Zbu!`2bD
zGH-b;bKZ|!*YxMtJb!im!Si1RGvps27|aXA=8Y)yjJ7>5*fr1DHQ$w8^MYOTw%ah*
z=llTF-hOUpv}EmIqr6OEzZe<o4p3)(8r=O1{vXBvWB7lZC<U{8o6|SLM%sq$$L#Kw
z*NB?tJ=O9f$4-hiv*nvumBlbNBI~o(Y{s%h>vjx`wM)olCRqDq0h_!x8!|30hjK-U
z9H7U0#UO|tjh-AJ1Zd38it5%0@N%5<a$jgHMNPx6)xBmdXq)vu*gC@NvwcrI_VnG)
zJbmgJ`8OtT4`txk`b`p%g}XnuaGFFch*~Yf-#81I0jJra#5_Rb=9JGd+poOUz=^z4
z-!X}QO>bqz^y%s10}p@Y@keSG9ytH-gZEx|U~*Eb@@TPTB99ZS60Z1O)b8LZ8Idt?
zyibOYqsf}x2(9?sB=Oh<k5@5_o)ny|6hZ0WqS^tOv?Z^x(G}5Q-&^bYVU*OWvkhpQ
zfPpere4}kC4cYw*?TrO*$yvA}m|8tuT9M{Ni?dgr;7p%^71-T2-G=S1)Yf63R5175
z05EcB$C`F%`bJaf-+C?4$cn%z&b09_dTS%>4#$)2Y;VM@(%zT_BVtoEQ#2rdjuHJ+
z810^#%qpN37Byy{AJCS@aD>!F%dp+p`W_p@Te#7NQyeCg*e<-A>+Pmpx5M}a{&~&t
z-FOqDy0IWEU|(K~jiVj+HSr>c9UA_M8OH2nzr)WB%Xi>&K2C#jZU|%a3bgK!Rmq0_
z6?N#%p2J6DeBC-?{S*e6`gxg}>BgHHyoie50>GCFYMul((t&INblUn^NkDq^H-XLb
zz~-BIY-W0+$G(<oWWk2^V*v_HGgI(gi3t0>j|7ge1+Pme)(46HPZcmy48e!0G~|_e
z#(JxS4ft=}P^C7&`mw6Z&hXHV9p%;8`djN=>+cq=pRg~K*TW*oB%YrLh5ThvA%Ce3
zfnK7akj_EyF44(<FRPRPwtby+jsmW3tfl?~mY#@2E%lZD0J@8Sp?4)VLQzSKKpO*V
zqxNzUId)KqnW|0CT@lfbQFLVuw7ybotyW?a>#gQO&^Bm7Hs1El!1S-0+{QnNwlY?I
zg2!+f%DMTBA!GeQX%hJp6_e;InX9C6uO{>!yDym)C0d@or?%c{kgV{@PszC3G13+s
zh5Znmhf{mPr<Fa8w{_9FL}VbIR=T-)9`Mv}bPK7inna&fd2Gz*w)ZLX-V4dJSBA-^
zY>>pLPX2vnpdeO8!Ol0h8V%Hk41{GL&*c&8$Imr0|I?EneYw1zy>#<_t2YsP!6!O@
zMFLI>8;97hUhIi=<{_Q8QjO@QL#3Rg^ClnkynD?V7#S-U{N&qcH;A9SzYjmzn+K%K
zH`Qf!(O`-_oxkMl%Z0JLOf_wp*_DP?E~*e2<T&4_f#kug^Ey42i)WqJ1n_N!w9Y^A
zZhDIVVTtfL0jH-19@P16iT>b36I;={zoH*?p6$}Ha~_4={HAlQOKnxMokvl@d5mg-
zKzK{VH8@YBxac#}wZ6<@Dn%M^Nv=xcEvL@&RkSKBe~8t9K-h&ttfM^hp1152QeqwH
zruizBL>pGEoPB8v?mZ6!rMZHUF1Peo39w#(H6C_4S%8?a-Kv<;mR|-uKFGv1w+^Tf
zBv^pej1jC_cdotBI>%o6QVVP>y@rvWTXmKz5M+7zk_fL#i{P|MD;K`4QFd2(%Oyv!
zpC;Mho=JzQ+u+WyU7%wYp&mz&?IC>m$^d`Qt%l@8+_{JPopwH-cF!eY8j`SsO>n9U
zBs*WAs=^~lHf3l&l=jmd&2d6g;5bgbZ-w6ZUg)t~l%>b3wz*cv_C(C4<YUcn>z3y$
zW3|;19xdL;1(Z`YA-qBP^>i%tqXWAUa|S6hNhq6BE0Q6&lbJbZ4mG$5!8QVT6thX<
z6xNNfTasVRVwGg6$a<HJiq#*h!p4`QW|MPMz@MyzUcH9ZLNXoi^0Mc#K*dY5YJ}V8
z2)83**dYn0=#~2#S_<S6fNcb_LFIW#-Ch$U%kHF6%19=^HzQ{3BMl=oic}^GY0fv%
z5gP+1_*$CtEqdj$qrSV|tk}EQWMi%;ANR`QS+StH$#y3n<@<irNXQG)N>i>U%Va(3
ziQolh7{b)Wn?&NtgjLd#CU$o)8|s`O!XD}!<5g$)U{_&TOB)@S)Eg69>CLQjH*n88
zjv42-lrnAXDvj(E7dFFny^wG4De}fnT{``l5W5Tu)nl9s$IhOpp62zTCfHuVRCl48
zR;eC4dln%xH{OS&l4M5e2XfrxfGNH}^reCxs!Gd^r>v8|DRhz#mDtc)Tns>*A5)WQ
zhrsKW3i1Z}`=Zfbr{K7z9*LHm(<qWmzj==6A5uiRT6pIx`0jiaKU+JUE;KDFG&$cR
zP{PxcoBRATHR!p1FylOq@6JX1XmCqBP8Zy+prPdVO>-v1gYM2G!?44dyn#l-nK)13
z(ed$<C;2K$;1<^CPsLfMN(9N-5asmzeC#|e%L{(u2&DK;s>%I?^9&yIjG3|0C3JTF
zcMR->FI3Y>bP4OVgeG$cebdc;x#)2v)#|aNTSX;@hV#R3WeeC}zJbuK*CnyeE`lf|
zw9J-UPqGvu_7sT7+m&+j_X+f&T_Ef+fBzM<)6d`GUJ>PpZ7Z?PEwuC}i2S!csL`8#
z2pG-Xxox`x(nUk5vyoJvwtHy5o$CKzDvjImzK_;Sm*U8o<JFlG<cNb5a~hZ3GDFL2
zzzaWkBf6VQ(2MilSkf*-tgXHA-%C0}Z`E||ubcitHS`TPSoZuD&ezO6?3)*SY-5^@
z3$n(ED$cP4#tNzOgLV*Jy5Fvcmx9ngQN_>Y%lNYq;7HCw^`SN$&oP?x@pAP{b<uOp
z6V=ee`ggC7cipuLnOX^sS7*;wf7g5}!Z?_hE=297d5HtF-4GDc$IFR#cT|7D#!~(W
zl}z6=uYJE>yZZ4EUrM{;Qg|nyG9BPC32oA2=uN)~MyavH%bzrcC+1WJKq=xX(x2@E
z>9jvS1EtAzjn6ieDf2|yos6rLl5C`|r0ne$XRsl+GroF~noK(v@rMt_t!aSRj@)_}
zbwrgz3lar9WuTwh0(83*w5Ni0-ccMP>k&Ubs|Dh`hc0>rkb*#`h90CRVq5C3pfvS}
zNdR-GM8k|R-`3)+t}v<Fr(%wmo`&E-NA9@7*0}St6LhA>UbgG`0~449XU8D8GZMaJ
z+IbP!7QLXeioy=Q<^(4{Lsf~&2j-6vGp3yj9E$Te{-lF(yWUg^CCQK9xNIUNGAqdV
z|04p~GrrZ*_CtYYTSc>n_`j!zat)t5IGD(P0;P=n|4a`>C;thQoc!O~%_j*)XJKF}
z^dnST(fEJWBBRQHD%j5JXqLrGoq{dE3?#h&g03;g%a7M3DI7zW6y!hDBJb$NOF-$y
zAoA`CjIj}&<PFXs4(6j~(6ko$!5B1*;6~xkaV>~}`NN^yIQE=K<>!Qg`x!0nCD>ES
z-<{J#(dm0nDJ=}b<601(ltB=oEB14TE-Qo3(&Aob5b~y?JCvul3nlLyOL&}o19c2a
zd(*>mVgZ$_UgOh=WL6>Ud)|>&L~SYQ@-=g!-_i3x7a+WXrX5DPdkp!e9*WK=Cs4{b
z(YN(bbV`*#$vM%<jKhMpT|7fU=ib+1ppJK6o^x=+_fs)OWw`fqE$-S7Jb{wexocAB
zhOQ}^{R=%5oh0-@%LjTWI!O>vx(xLQ&D<4AMn`Wz5Re>(R(IP3oPpe&z&&qdv{tDP
z<*I7Q)ryCJf(^9&=@is!<G6d6+mU2LXHtndqwx8p7BCy~I_IBr)MVP}gSszjfs~;R
zw=is%--c3V_hKxQHPNB~J*0(6yVlD_)C_A^jb>!#iNt<y)NXrzSPO6gC44IK=++hA
zhC8p)o)-nx{*?ANpGuahGGnN5H>13=9LD4=b2he?Vy&c_Xj;<Qtm&cXOiTi$jJbSH
z4@GBU5-8hsHm_<iPzU=aIh*TZjLLBDEiLX^eM_K}aW?Phq39%`4_e;WL(xfsfKtNQ
z{G%2|*|EMAz2w%Kw&808>UjfGMuW>p$A4(4)k;UATl454U4ZK0SpHWJ<sN+QSlLND
zOCNS(S49^ll%Zu&rA6ye4(nlEEK;<!8xe%&M_HA&(r#p2Zfb~7+Q|*j&b#iCK7$@J
zhjB_zkZy?P&=d|MXJjj4Uq!-_uNk~A>EY>SLjvAFoGR0Z9d02Ouxk`y&_O1#qk>ng
z+;PAE2(5cfJ6-o-#K9%s#KAr{P&YG<_OZLv#Y&CK=E{&6S{ACZ)0bDuKn)ZbUl!7k
zcN*On5F;%C+O-D(r3<4*`R1L%rEh3~49ukj&PUCnSGCv=#-idP^cz|j19NB_i_m?@
z&^ub>2V>AMc6}8IdS44-VE%9@H%@-KFTDL+i+c(76m|l-ujKw>+fd5HDj(>f=%T-#
z=Ax(zqhq=lV&Jj9Il6FIi+qW}?qW|1x^PSjVqpH<6kWJSi+c(76m+4x#(njsKzVe#
zP|8F?bo%ZRDP6Z<$N4ypC9*qvA9_kpDRn=#n1yQ{#}@BFKv(+MqByU}PcE)u5R#LN
z+JnZ~0iiz>XB*kCZB53^ak@CGo|3;4<8UU~b49$}dHS*El~d{{ITtTGkcYPNa*&*h
zuZZ(w`IGU^OE?<hyhJp=OiwS<(;uZz-^SCXbDjUj{Pz|9`^WtERsQ=Y^h?MnC)oKZ
zvgAN?9vRh|C=|#8pjrx&3kvzer<{P-QUd-gC4fH-O29r4C2`r#A;1R;SMjR@-euax
z<^TNoK2Df&>SOEkeH{rD_)++e>A?7pFuc7Ft<3auVEl(^SNzD`QD~of0ML1xzKgSe
z9W2PEyziyGbURtr*D3KX2z?ypm=oUJ+-4YdnA`7z52ekmJS9#nmZ!w&S)LN7=YpL)
z3Oo4&6!$zOUdB$!4y0!1O;hjtL&8e_Ct=I2<bTl<x03wnMi`k46*eWVWTjV$uy>K9
zN|T?e@<tIp9Ye%|>}g<EM-ByC!OXgyu};e*cC$R88}*+aiJz7E?`Fk0WOxY)vyY^C
z9oP&DUF8aftZ@0`=c+46bK+Si*sc8ZFp}|6UcR>Ptsv)w$ccv?kq}v)&h226vX0rn
zYedu!vAy-kr;K~GG{4?%St9c6vGL@}6;coGPjd+_`}|5;0|}qLMteos5)Z3*QP_^c
zc+r|dQlu3Vx%2q88(whB%nU-K)keCcUB2gqKowW^)?+p_kMv9e1zK-fMTN=aHS(Ky
z+e5eMC}$si#f`cb*@0zrfU8QTQv3qjL)i`8s8w61qtwXqv=_Tbb{k3HT9h*}Kv&bE
z#B7?pq_$><R*he30vUjmG)R(VEVA9c$@jgQ=2a7n+{q>_1AAo2>nPiAuG`=r?o35N
zRyx`+m!cK)x0ep<Q#LR3YZ@(-Okg1DCJ{sU*)F&x4o=g>HZ~E3%d@{d%kLoGmj75Y
zec$tImf<#<CNlk12-Q0BO-%`&E4_^JUSf8n@TJg&b8;N(^Ls_V{Omo)LAJgv|Dn*B
z?IRijFv<dpnmpUdo~GfhL^!1mj?;b6oybDTZt=~!=QnB%lQNqD3}q}_WEE)BGOk*J
zBX&cI0<Kc-r93RPtcdNztxjxCMy+-&6-F?Y7hs9P^l{joI-KGqT+C_ov!ZrjMh&lq
ztSqRt4>`cRz^th)xs=qg@o5J(?GdA|TEh+~ZypFqupE<00)^WRVglSm6i1pNHA%JF
zvfTu6NX>YNo&W^7U$|D{W-xkdV90kNT_=W0yq+6k`}qoM(?r4<3)0`R!}$EZV{C8R
zY}cS4CiJt;8HU7gZAd$CC#7JYvcj+(EX>cNIGAg1;HrZ==lLtprGr$HBY>J|qE1Ri
zW433Hp8QSlbFaT#FLX~OS@*v#yQj;cB(kffKWCfaGKT?EFmHveCV)*KPY$$*XYU(l
zM<A1}TB@8W0m;R_O|~!Ca6@CgX4xxN6aOK<Bb03?X1n>;axK|gegS(0*;aF{>D7&d
zwMXg}mZR$!c%g6LGG`%buOcVmCQo|GGnpo4U=z~g)ophfGLYDX17s+IXsJ-mZjtOZ
zJ%bYP5n^PFATq5bvjGj>00x$c6mOgCU>&tqYbLVIHEXz;P8JuV^a9(TBv^&4HkXAp
z+hoJ0dzF%wvT<QepuUhq+>^DT8zC|@QsOu+H~WDUBrAl&fi1_5JlSz}AXiF9`gt*W
zTTWl-QW0f9MfEW%6wfox2z1;TqMs)zxxyv-*@Ix+nWUfpA;<h7ej+wOt@tflN~g-(
zZSppoy#6Myi+RhZ=__@J^h2kS6tR+A+s=+56tCx<PW`cTzQ}bNz-h-XP|$Lf`VyCk
zMr2YW#Fm7qbylEBiH10uJjVdPuFWsaJ8uvXbQjN|=_}uja452JDDsNf9#s-KZ}uSl
zpKz4mmnb2FBkPGlP)kTlcqt^ewS;V<xvIAo3e(yXxoQ&<D(~q)u7CIrKa61s4i@`L
zzQB3I9M-*vj2%2BQdu=2HV_ba<+1yc^6MD2=i*18+Ssjukw{Op!x|k(tzxctS6a*k
aa+mGan^A)lV1;bhWP=QMV&g;@IrsmB!2}Ng

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/data_utils/index.doctree b/.doctrees/autoapi/lmflow/utils/data_utils/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..114375b876c4a6b7237140eb5f8002beca198928
GIT binary patch
literal 50455
zcmd6Q3y@q#d9GI4x8AZWTb67_c5IE7)ynu0o3*U?6$fEyA+`(zNg2(~oZT}=^J3<V
zw89oKp#TdGd5|$Nl@RhMk_sWg5c6^%I1W%iF8AJCaxdH@wz;WXszRm6CB;p-T)E+X
ze?LxlpE;-JoY`FoS9x`&Pj~-+|NZy(_y70fJT~&ghA*#S{{@@dO}A2Cnkkmc^+wrg
zcZ2bAqtvN6b-(*--F<KGp6}*^iB|D!yU}TtoNllNIZAG&T5dV@?#XVjiL(2ZYI{bM
zA8R`$ztX6O#oZBi)Ezt7&ASu9h+pxmj`k|JqFP(1HqOrQI?6@ASYVG`DuNQj$DAc%
zB{}0eyPV6>W)VF3U3Y6RQ3xC1t_wy5?{mR$wNiJwbDjBOFj@5dR%O29qs`RDb(M0P
z+A~(F7Tawk;(4TAtnsJua;K?&%oj_iTh0RI=>%hbagkni-OcVUcZ0jzUFTjMJbb`y
z)SLsYPQC854xDYYP9HdO^nNr{{sRWv2RLH~iXFdEY*r3%Q5<0HKcGZ$pi(b8OEaxD
z+IGzYYdiDqnqY*bX@cZ3c3*+ExO*VE>+t{e_<t|{-$(6&Xj&E4ntU+aELK|G;1#t-
zxl?uS$R+hba1GC~KPiJdI_Gzb9ru;)e%1In_Zo;8eK&y~DK55(O;=+yc=DL*<eIGp
zUet1xda2rh-sRfP8K+gO<`z12ratXiB>8@&zL+cKTE%*~QA2`Lo}R2Wie*YB0p%82
zNH6AI`_Sv=atoEJGo70+`p{4DemaMeIcQL;h+3(xdi$)?YV&f74mzjkJLQ~@EHhl?
zcKjGKC{!JyzKiP_7j-eQ+o@*`xmUZ>P@HkuL#%JcN{yN^%H5G**PcDpF(n5Bpq!gO
zm#cSb&2!My+@3vBIMzIzKhSx7dp?<|U}C9OoozRZgeT+8meY1xXPhqaIwyG)JZ}K6
z>p2Q37}+8dg0XwWU}a#jL;Y5QDy0_`jo(u|I}_LzHqy5E#raOP*eaA7)k3@Jlr&xL
zH|U00{$R_AH$C#I^N-wgUjED4`W8iwcVn<afH5f}VM4o8s})=4y1`cMR{{K(5PEnL
z|GnYJk)wy}3l1Q#<UHuKJ5~RU72iEvZ+3io|DbcW*eV|d##fWGNzfgM(hv=u2uA(I
zR-@DGkpY1_RH~X$K;tNKwo>-p?gNw5|Gx~G1Q7eL$S&?DVXcBU>lKixnQxTO#WK3C
z)i@jbVuV1lw~YH!lAYk^XPf6{8}r`mDVxA^2NM8u##ZXksS~UN^1$~EO{Y#MS6G6f
z(Rd9`t||jk0miBopWs-d?0#MhvfZ{ebk)dvHhJX10<m@lYn#yS(dN1R5NX%Z4Hk9g
zpF<m-fw4}6w_p^~0X}tu>r>g#E~YY6pjNEZyY62FWBBEtYXXMW5d*<`{C1X_)k>-2
zcOPf3F>31F3ovEpX1Oi2%tQcY=Vsabp*zPY`Nh_v<9CBw22lqF5}(CfVqsP060<km
zJbO^Xl|wYS7=px~#`xRej;fA)TJK0Fu`{U0$kzOWsOJ+@%+`BU;Nu|pF%W!8jxaIc
z!x7g9065WI?!TraWM}`Q;N}m&%^#WE<dfM8s^NG{fw0NAkfN}HhtEm&g3}j~I>&}D
zO4{Czp*3p+BwUw*eT59(l~j_=ljG(udH3u6alzY;I&6I&LWBD)eb9T?qFnlM?>_Io
z;J!HLekZsiX$YrjI%(;}AIaVpCL?1ovou}y_5p$M?A=1W#h}64NRR18wKvI$FLPjf
z_pxe{O@IIqLC)S9F}&@WJ^~ED6eKxQFii8DE*vrR>U)s;-BCRZF4u51S8dc6`FyGv
zHE{1l>9<Eq>zUl2cMC-37m!g*P5F>f>?x16D4jk%N%((M$s{Tf{%JVBwM?!j?w76@
zANe~1EhV%L-rI#vxL3ywX1y`Q1LoM?+rkozjhz78mPD04aPL0qo~LRsch;&|Eem<`
ztvwc!9NWhvvu|?I5}I7YQ!Q{GwLm^<p8o8coP5afb3UA1&3(-+G+N^KG<UD2bAF>$
za>YM0rB1mxGs&FX0(rQ3@@fkdU_gaO;QDgE_blf2%;C+^D->XEZ?*DPC+{p3Yt5?D
z7Orrijl^j|v(R=s3ky}M>)u8k9=8wYm}dtYaHNSOIoM7Mb#iC9Al>IiJr7^4*{HXj
z<K$SACtaZF(@9kKaN9v!EAXs4bqfDbb)uln>7JYv4rifKD=s>Pg=%pzPu@5vs1zc=
zmGE6s_CitKYyQp;<T*&!!~1OpHUrfmy$@N(??)NSFsC>gTJ?xxCC?us88KJLonUHs
zVjmM+-c^%^dsm?P=w#q4G^QzHw*w(Xb2Mm%|EuB1kHd;|hW|^fkCF{SlHvbvC^v}V
z-<xFkqs=5-T16ut14cAm_nu|)fZRtp>}cYqSJ`a+KT-*9>%Slh+sqR+88dkA8O}qZ
ztq<;gC;<vjacn#_ojb*XP4vL<FnnN$c!~$-@@VFjNmXeLe<{*%1SdAL$a8@lJQgC+
zJlkMgav=@fDCbJ6aqU`Dl*#%BnRsg`_b!jvfW6oetewtjbw#=JhLBytMKC4=A$b{L
za?y%!Um;cmgYEe!eJ=lyKKrTBzhQOhZ{YpUyInFd$(Uei8Yj0ei<6-lt*U6BsQHzg
zRLMv>={%;z7je$PhWY=^JRI#!z+Bk-4A;*Lmg#3g(?OF2HbF_YNnsXljc6isg3V-N
zY|?E5R?HPdIi(d+32uibLi$=ipSG$@It#RjI&7$zWPv&;H;4t=lVpKZ122*7DKSX^
z;MQx?R889SKAQo0H<jQ9=#(gI6SXuzCpja@259Gzq^Yxn7)uW2?EFTgK~^(xDPZEs
z%tBiYm~g_;9J(dRFcD_qFd_Ys6T*Kv!5kR;;x-vSxeH?jk+9BGWTPPR90^f?NLZ7m
z)?cz}K{^n57IoM(I|)SoGs>+fh^Pi$A`tmJlLSEIhIFp0))7=R8@_ypN^pGnoG5JT
z1&J@8<~%0j%bvdOw8WVdIThagFw(#uJG}WpW+AQy-nb#&^n@&u0iY)<#~&HuIB^`9
z!}FYAQH~QQV<1OOh<d8aL^%pUyVivew6&*ZZ3J9R`IZ?<--$YGgCGft_M+S%P_!>;
zKGWOKO9V-mGl>9_b|yx=v_7ERdLP&|oI>iarV<>5wu!>F4v;9cnX{FQLbnYVg_GDc
zaCwDTM<c@cu|uraW)`@;KL>N9)C{RN{+SFMy?HriQNWbp6$_c3U{TaqxH=pHWp7p4
z$VLInC$aK5)3)Bawb63MsvzkmPc7788!$<T^%%+xf>={Yh!xY)O9WYUCK5o__PFr5
zCV$G94Xxf!B{*807KLrXmuTg3juO$T&r%Ona|l6--(%I-VT59koLt(mYR`u?KoQ4J
z&KF$<^ghI*DJG*{6^U`ne%=RTa_fs_+RC4{(&asgVr+Y*Vl0>l_e^%x5}=W|mu~_k
zkwzB*Y4y#r2)G$L#;`>+hv}`4%~5S>DSFbC1$a^DNcsiAzYj~W1;OjV(u|CU_R$h^
zU+ZWaTc)r%F1R%koX_ve<tz0Z-@%i#-F06#NFD4qHokhQ0eoEp8@PtCskL{jAZ_j1
zL~r2KV%M<WX&IZLuZprw3yHJXB<ssn9PA8Z#c97bZ90~Fw6S~h$|R7IeU_X-b>76V
zi4LYZ=Y#d&dq1{pGUCBT;)QK^V!w?ITtb1pF{*Kww=S%h_3>Wx@tR<QwI$k+o&}`-
z3roAg+G7`%f^lM-^E^&8liA&AVRsv2mo_!m9BhxPR<I*m>wt-{x46V*SBC{GnMpC2
zG)Xa7kW|k+Of}1H<UdrGz1rILV5kmM&~}0Y1iagLBX7n(tRJ?sh7si^t@(F~kq!L8
z(5$46Cq??TeMmQZNxvpSIvfqyTT2^)HF*x@LZbTe$?-l!&-N1iIi09?JAO%c!M=Yw
zGC&iTaDh}n&JO-gwYF|0tjk#6am4pBshx%+e1o8|KaerOb|2FI*~&ttWNtZhU(mM}
z5{!QqL&^g`QA6zAheUgSE6WjG+K@x`Rn*X3-u{^S^;S~$U}p_GjVlE@SWyTS9X~NN
zE|99dlGoKc{DaH1-Ogt!&e;<7CU=9)^082?m)u56Q;2@1fcAFE9hbQR$NF~Cp$eSt
zK=w+x>s<{FwFx6J{bZDBjw(uW-pepq%-G`O2*yn}m?VV3CQ=-JF*mXXqhjBv`-1m2
zs&_OPU+7e;41zfE$%5Y~k@U#EzQEGng{`VAm4nvSsA1SgyF<Mps*msP--{0W4`D~y
znC5n(#O-#E6!Y4i=8L>I`wTIyg)GeF#}sR^&qXRQZIrH|BqOmFOH?73=vu6unZ9eW
zf3rfYp!0~6%xkgVU0wPa)?!~;7AMA9Y_w0*u!{l~`j>4vrsFMBkl|+p6X@uPU~-}2
zRLdNSlvNQRg7X3Fb<DsBOR}M-N(Pg!mzF=A2J1=6Q0yKtHQcVCS5fX=-WBu@AIqeu
zAs3NT48cuPQ#2cxn#!?>;bac~(TS?Evrq&SvqMtYyQTlf?Hl$$AWx-UA>*DX7EA?<
zJOvyCsA*CPW+L>{(5BNN!<aD{6^+Y^7<-guwC%UBSyHEFX2S4|&VKIAB<s7mtnb7>
z1P0nJcaPF^Sh8xJcL(L>y>$SccN?z|oyM#?EV!dvE^U_4Gpjfep=(!`H&bp-kH~XH
zQC+5qPAGb~kv`bk(^GPK74t?Zb0Z6X;b^=OOjemt?CtTWb}mOM!sTce{jI79iT<V(
zRBkkjOs6O^MMe0iv~30^3v;=u9j73@ZS=GMCTVXArywy)y36kC!nvvStetR*ta*4Z
zf~VgBE!&Yi8EwG%yAVW0x7^jQ`JW=tRg(|d4zm;Ud>V|I4?o@+og;}eIo@Z%XfTF(
z6g!sVeU5%Ik4^twZ|vvCUoQk>^@e=hB$M~^Q(PB<4dSKzy%jmcgs2^6+FWnQG?|ZT
zB6^r~(V<a3k^OfWq}avXeX-qnut<gX5^4($gc~^0#wSv21+hmay#EKnQo%Sx<h`}%
zL75ipR7{0ik8u{)Z;23uH?HSSjwuNJOwOPn=I?1FXfr|+GV<PbQ1mXxA6tOo+hHXv
zz;I7z-n&ZYRND?DSkqDzFW(ERiH#(9HxMs-lXwx)gBV_>bzU}^l4UsT-PA8f-t=52
z!b&8LYbmNhYesFtwfc+y4p~5qhjqphA_1H1R31hZJzl8@1qdo{HmJnA{=4l|7UHM~
z-(OHMeSc^*mtBnSPw$izXRMTUpF{Wky+^^3T5w?gpXvA$mCTO+6qw26`18h_h^+G*
zar|Fm{ceuJXm;-dXqxWf$LRbAwJu#H=l!3-YP|EV_~~)pzllQJdG|gF7QH{h9|Kyv
zk0T}Y*S)7G6H)R$!PDvaU0v@_@he!b@Uq7#zXx{lzauZ74tSqNeu=XJdo?%QvH?v1
zmL-SM)M{!?)xD>m2B#HK5<`9~n2vDj?TAU+rKP3!j~>gyq9^>L^HeDGkGxmvl;IPF
zV8|RJ467OC8R1L|Ou2Dfh94*KUZvH-d&K)mEs30+n2Om%I<eBccTlflqfrHeo4j{V
zpB!*>V!`}FnItE?oe!c=)Z3vn-Q*@?1I=(XZ|Aj8lqXn)-f_JV!z6NPi1-hslTSye
zB|{5Q6Fv@6;XaOcTB9U@MENy4;=4oglJr_}_vI|yg&xiW5X5s#5cbw+9u5`Zva^d@
z^KeMq?@vKR^KhuhFQq6VJRBNV{igVuhw~cL{#Fzbkk%Iz#vXLA%2f{|CgE=-{~wC#
zq~~Clr+|YWP@x1k_`|8G3f+6rW>`S6XaxrB1-tL1oog(x<9T%E4fued$<$P%<0lm4
zZ$wD~1$I^r<=U)Z)v3#nj3S1_MV!;?p%VsYn3BezP~L>0n`0V8T5oHNKpF#!&_F2|
z@!9~Q=xd<RD9^9Wgtm^E$YJj-0YMv4C<;N8HV_2;EL3oURcJ!c2<vG%?!_;qX88y`
zJL3X`h)x<1dPSe^kX-f6r}%w!7JfrmdJw|-NhTb7`!rahB3z!SVTrn6lu<Di^_I91
zM53{co;XciuT>PMsy*+at7BsDYRY;%iwr%XtZ!ruGoyo)HjuI&gQOnfk}~XNA7yPt
z=-E3{C@axPt*n+FF_hJoPbup=vhW)!>%$PvV@x>q_G!vWMYuduD=T$Dg0dQ@O?sLR
zu_IHwv8gFmQJkV0@1UgXkW*98k7SXXRM6jHA~O|~(gsq{H$pleVilTbI*f&JHKAia
zRCj+qLe1WNLOqE}8uh#?xg(@z+OjAvzmbK@P$}O8fqb0_#NH}RDX9pTVQQr$v2&%g
z7ttnm1iEWeFeWB;G(*ZTv3nhAR}(usQzCRN9Y#5ETb0oDkI>?@p=-quivc?QnlD1v
z(nXi;FdJ_VU3*`EZ{qD3$Dco^rz_~`i{aC=c)H;I1^fFP`}>#d@0Zx$FSEa2p}*w#
ztVW1AJl?bH5OY{lcex?vyTYo%6F;*#n!M>(R<PAgdFo3Z`kKwnHyYK|i$vdu-qTUk
zM1Yi$=;^Q$riF_=k?64b63!-6ibAhLVKvK$M91)AM51q%$L%t8fG~7s#M-P$WHI8c
z<1&kEw@OgWK4xvUHD3LC6ipegp0`u6#j6FCH1X=Dor*19EvOjs!}Yav2W<RiHU3h`
z@fWSM^>A*ch&3mZIb{7d)-~pEO3{}l66;JyLe@uE{pK3VW;UwW@lggQaVp90&0%?@
zQtC-coWE}sr?n}+%Xv$)jPkUdicKvAl{9MkoSlkIEd>=rEi-JP{2MD}y?-+)<zV|K
z|10a43_8ghzbu{H&QCF$_AE{@H<=lwaz>Mtg1%%GtyMvP$OMsUoo{r)W*|MAf(j~W
z6m+|ticLWU6+=OXY0|(Y*t^ijUn)7?V5L2<igGfUar0VmlWD1sw?Ctf+Ozj8^~Fpo
zF2YjZUC`3_r9Q<^&r;uB6w18RN0j7JA3cXlec?1M-?P-W2={3p^3v=Pa}fJdUy3u}
zV#zN>PNll#)T*m5{dwTLC`zvS(%(`R7Uj}khYBUc`adu=C1;h9X#af(>oc!{#pl&r
zzzoyRu(u}IvK)jNYzgr&RtZ`iT%qS9KBVr)%ZYm^STgRH$Hq!=={#yoTwaU>_H(h^
zYn%(yU_|jZ=!HZEBaYvj0fDEux%Ly8dP9Uqe-(wIv3yF?{ib9<i7Y2JpnUIR6`Ikf
zKH<@iMd+DETnA}wL6PXBi9by$`Af5_C#muFg)Cfz@ueo3{0wWdoog+=L`Ari?1I%6
z4oR@@VN}cr{m!YWxS@by5T|OJ%-nIx7MBI;7C(!AF^j}xxbyE=Gt6)&r46KTEy(3t
zTrNgD-Wu-ww+J<R$BA$!QAt*}aF=JFjWPPp%;Cx2$v{-sHLYofO--#eB%L^|dEL}h
z5=0ITkC7<)wq6=>`j(UpI+r{-{s~A<R~LH+3ZTm|ahrNaX?jI=$!hy1I*{ycR-p-W
z+T0YI+u1}_0n8l{YIe4T-Vv2#y;J)pZpz+s5)KC`s#<j<ORGZIJOcq9W&*TJK^p*6
zgv;FCO&V-cH}L^rFQSFz32EQO5XH`z2+PxCre9oI+iy`r?JWU6n4PV}ev3!4$Y0LE
zmV{Q}qz%Q`jO|x<s9Di0n=+9BrEPIeE5|M>W)>$WKDX1LtA)e=m`cU$fH1ArZ90Xl
zwMTGfy;hf0Xd_K&YB!Ulw9)=3dhJ)3vfB49<@&WnGT1+(;`$?5xDLnHd9eJujAgqT
zYvYTG@K)Lfi#EPUFz-u2MH^pK<U)!fVtkRJW*A>@MeS;Q+1nwOy`-}wmy1?q+3Q<q
zM%raB#SmL?cu&9Pi)AnAuF7_pjZd@e^?fkOm%aXlp4hS%d%D!iUbppJC6#vhE5qef
z%UuTB-^--Dd`#Xv#x5SabFNW$R&UK~e7!lPF^7N3HLu~Y5~hOqfP#7YBIXh^5#fk=
zqo8a@SnV>_z+!ka*1)bxmO6vqyt%9}F0*sfWkedycjR@4nN_2vt&OrqYV#->4F~ET
zKx9?qCs=lq{VhW4=ztT`EYU`z9qYAB(gYc4sx|iw-kQ4@d${>xYk2=eqe)-nYZ!+o
zhwhXt;7JiaW+fa9HG1!Xc#w+h@&sKRIIYy#7a_GnzQZ)wBt8B^k#y`Nae)UrgIhRL
ztacoeJgz(FG@Ff<Uuff-ZvI<34f^sVTlO#MT$v!;w{Pv)dm^_nL*XJsZm&cG<0H3<
zpPtC=k6=+UM{bFdjNH<*jND$<3$;BF+yI&}L|u_4f;;d{c>?tYF0YjFgMwFSzD?ZL
zHAj#3{xVwm0oBUYJ?QbrSy+^w#z(18LbUhr)D%9chJD@4&A7X@$i7s|cVJ7TWzk!l
zMqpW*RRm_~^dbdkWsVF1+U+4hjhH}8BPff5>YOw7(m3GZeU)0o@5URZOd5!NJ(Gk)
zVB#w%WU~U17;6_@{v0b}cSa^v6D0uZNlNp+lZENf!FU2}JfDJd&B34|obzOyqsD3u
z1~vAJys^efo@qpv#K&tw1+gnf#BD@ST~#17AR2X#WJs|8Yd3I^Ni}d~5cEG-o6KMo
zr41B}`gI8DWiBKm2)YgfMRq-VR(-xR6VoCTMKsccqISghgk&Y@wc>6n3wNPLJ_#}G
zMah0+Nz+Iw!X=klBT3j?Bke^rmzR|I&shIhLERHBFCk_>pHB-Zy&tvz0rd<wbSn{3
z`nfFfmq7mpDwF{9$GE{3;UtIdhANBOm~;7dr{v~}?cCkdxqGH__fF^To6h|-`?6LI
z-_FLjulD2Ss7lGn{p{WQvd5ZUFG=6Gl4K)jGA4*f6%T(aoo;24sTO?tE%a}lskI&N
z?_5>`CFjemMYZ=GS=t+p;SZy+kFmzuhl)0asR);keL!kfg+zKf1r==!Q<23KMZ_4U
z!H{7L|0ZfzW7ysf5w?=EK52GVC2aL6G$U=;N-@M@2KWrDy&WQKCFkg}9cH04VXOZH
zCVANE8}!7&R_y6ghple!y-zr8+$zfz#mfy`-4w$nT|upXvhU8l2M^O1{a%C5Z5`UX
zdeN)zq6hUkkck2*qgVeBR>Cy#PomMQxErm-gg)+-!n7ZT6)z)x6~mnozq-aQdj`~z
z7#1$$TH-s2%{xU@I_9qYhic5^Mw_UuajQ`jwav03ajUf~yBW8-Qhg56)Gg6IBYrih
zGD|z9C#g1FJ9wL};Y>b+nUuULAxH>S-Dsy$Kovc=qlj3Qpi<VUY&6D}Y+xi+wegUh
zdNWGBf2fKxWdg4raER?rd-k4CRgBhOgizJLVZg?RsuVvxp{nOmD08TaD9KP2J<Cwl
z_84&OiB<gx%>Rohlo+eBeWl$KhE!HCxFLNjG>aS3N9c(eQucJI4e3w4;mDDrhwI{7
zj}JO++>H3fitiq#YXNe3c1a&j&s%N+za{B5ZClODJ{h&!UOQq)bWt?!al{Sc#d^Kr
z_uSNc)A57X5xCFh_Oe#)Jv5y=7>3u`9m})1FiURlp+mU(nAN)@5LXi`<oG}YIlMHR
zw|^^A$h63wZSywKv&{k}u5^h6)}BL3%xpgev%T_{lYKHI5^us&T}~Fm=~dcgK;HX_
z2#*ljRY3L4vwirw0rmI2s4RxBivY6J+IWyv{PckAdr>Ge$Py(9vh*xKmdTp}huQ_o
zPeGyHhb)PZs$ZFW)jhZ|q&=GpKc{jGt7+x?Dkc9^IGV;p9xdAYkD`8wxa!XF;IgEu
zT=j9w?^AKEB@-b*t@{5HsoorU={~LU&#1#TQ`7p*IHL11%Bh>qf_1pUkzZ`)EH$f@
zQU%w6v)5;ft$O!@`@H*t`yyBS?*v+7XsH4pQD)=$ljsqWm<<sqKfb6G=Ko(p1lXV)
z?ywB3cVr5ITJ!Kx2X?k?O7GrERs?rn1xo{c)q>v@c|X7FB|nH+$`ySqfXsJ%>>>ze
z2dQC(QID86)j&R2``;}&BB~*KW9$#`ZiZk*Pl!wk>29TWmg^x?A>FION|>&E4YuM5
z=t?+q4@2MiROg`X537r_3f6N__n;`Kr!l3+>{M)~lv7avBeogi0t$-pE8)&tN$8V$
zW4N#A#fB1&*;frX<926(pj*rVV)}LUEm&}^ZJs6ui-wfM?~|~{KoG@mkA<5?SEROZ
zOlJIM!pBig7`ebcdkDnXdN;qj(4&&Xn`Q>T1|DaGPC!e0&oQO(EMx`J&Y8U_LUa=$
zDw&r7^BWaDOpRhUD*QG*G1r1U4eDAjv|52ndw)h;h_ieozGEOH=H!<R#iO#GUM&6$
zbw8hG$Xp69J|}cTc5P&ik?wxnPlc5*F#bjqjQ2cT#~t;@kKnlbalqBd=^T4xe{d7c
z$&<ooNd?tk4~b=l>aU<^N~r#}or(>rIh7cw9;OCsSp8iqje)V6lSz9t)c|WDAoFPZ
zaT$>sI<7JQvXyzO#Y_}X#FmY3woMcUmJg${a^m5-i}>o$NliSvt-PlYPpS#Ul$}a{
zEsc!0N5iguRBOxab}F_*;zFU)OpfliQ|aH96eI4KmBzptb#X?VYh^x#N~I9c87qw+
za{;}_%KTsgVpmZpbNQr|#K7{&bm2vR#Eyrgk6C$7A)W|W#8Y`@xm3)_r7qG)xG-kc
zaS>dF>#VMVn1#jr3X0(#U*RcO(?NWNzlF3@`U-OHC!GZ<8gUlB0cYWp!mbbOEXWGv
zjMD7SRL;T=s8!5a_*Z&j&H{T{31`9NN7pZshFD9Hkwa@EM10-6YM$L6_oJ`@YqrP+
zu%|&Az`jVj0^ZL34?2k0mGKo=r@E~5ltgDI1%ZgO6aCt1|G3k2sfaL7$OsCp=wixF
z^sFAnN;P5I6Def6Th_%CqGt=8M$qx{kP`FB{u<+X<rh<K4T;40VkviP#BgdXrd&Zp
zxc{1}<CV3(w-0+aq*+smVQ=-#j7N=*w<3z49xJj1g;HA)ca(>lh>~1Wp=Y_KLeCNV
z@YfVFNAhZ}sVs{6CGx6CYpOMsw@`7eEZQ=JQi*n?dUF7#TT^MEjzQK`&ZAuBH5KXx
zH*rnno9Gdem<<@viQ$?G<u6l>Y_IUDtf@SK>eFr+l&pv~m48~^n#$2WYbq(MhOxAA
zcTzwvMMbeIEULMD==QgC$tx_Mre5TV-m}8;S<%CC3`LYYT~P1ouo7lKz84MZWn5wT
zT3B72sjzw<Uq;cCVY_eHsn~)DoQi^*q!pI`Z6%@4Cs%QWMSwF~Vd0D?PirF+gP49@
zeR~${OFw4f&7NYr$LxI^J(Ak&G0Vy=-}<eg<qJ~HETZczR6)c<eh&cq385hayHK(M
zIeyK)P31yeOKoE=)OGa4TqyQ5s0YHZYXv-zxqj;{DHO#(>3f!g(sI>gbk6S<$xzZ4
zKeNNM^sNbyBP-#v5Wm1_nM=(^4${lzE`{G7R>FY#`%$32maVhUW)@m}na=UW81}e;
zTU|0PzZ?yzWybNtD4G(-kK3u(aGX<#!SU$wOT|iKU?9&J4Azlc2q?|+%XusFR&$yx
zAkq;n?*CpZjll)P$fRC?8BjnUwlY7MfFkQte{3Z&uza$uPxTegbhxH7t@Z6+Sb0w&
zo>bwE|6->y&@hg4!t+)Vm#!1iEP4IF%DXih6blvc%XdGrQ?V`I2`XvYvT>Wul%;4(
z3}TNk`=n{@E-R6NJ6q6MIY^3M8aZI)KFydjEkit&JC{qvoNektjd&6<6OW7FNlc}f
zAkJ9VVHDpzuEfn4D}%TaCn0pX4wKH6kdr~V978oke2MR4E_$mNyaW3ZvVvv!5<WGK
z`4Sy^V!i}>S_xkwS=&e&VmU?z9j%Q>UXJ<2umSG~8^E3hZGgTUqZ$xwt~JUXI<DTR
z`}p{L8&?Khhe3M%$*z2r(_Zbxt9>WC<aJECyW9=j5nyxp>-BhE-aQyx!JVA_I1gWK
z@249N`6IU4NeUv%49hY@SqkhCOhhQH(<rUA2Qu4@POF6blct01xD&hHuCfCv1)OPG
z#N`<6?)h#$81Zp#rP_U9O)x3*IA#3oP|JpbO{Gq&1!)vSU30+(nZhn>nhVD9)lDB4
zy0p8vIjT|O4OnO~A(p$r=o{$zrFK`|>J-SKGv{ue3ns#AD7*N9W;<=aSZm_P`c|Xi
zgH>mtveXTR4q*Q}XQ2SPr;%Y&{Oq)<__fh*lnV33(rJ2&@0)%+*tUQQa45&ok{=H?
zww;n+@6-xQ3yoH-`*^Sku2iL2C^c#|*6DM>E}r649hOp{OEU}X_zs!@N}@ou*v6>?
zBy4Co>?*0MbH;%n*9MbrvFaDZ&yirOvs8ina5|{dR;eRpXTGzD;+tp_KIDfSEvHyR
z$^<&1<WOUXKNc1(ICb2*fLz^RqT?^@zkPqZg6G}l%YxHtHChF?ST9!{e9C;3ScOv%
zZ03ln)iT0r=>}Iv-bAYKp2d(0Ht~8D53=-b^F^d^u!Y(Ht!ujwQ59|K1{<rz`eFx1
zks<Liey_y`_=3w?xZI&tE|i_7Q!j%|g$Z^P!)*c0ghVQ3sS56Bu&LPb8wHgVtgS%N
zHkLcJW+7BY2$m&ahoaM}7pv4$n^_7|bLxGS)NDJQa-*<RBURdhLlTX)Q_xE?Eh+hh
zGnICQE<{6Jg>t1`oUc+t1earSkYK=VKx5h%7W_yUv_@nr(8p@qU0})%_ATp5z3sGo
z?5{-MP;F&&`TADzZ20qmp<pYGG-wWez82bF!|{!yTlNN<nohHT@qshl^%B=aN}0!D
zm;w!o;7Zr`o9)>H2aw#JX`b`lMt!EyTEtkooN_V-kh9=BHFS@U?ife<{_AhN1@hd~
zS|~-TrxaEF4YGPdD^l6hPHU#(_zR2%px}V(*Qy{k;TPvIipsv*ePBGe8ai1kgo86|
zK!F{H!@-m-?Q?a%xKwZ}i*6PFLw|>>4O0CL?6$UoxUTJS9F0}Cw-u_5Qc;NZ>XKUo
zKj?Tq-q4jygvfom(*%i4;Lfl3=sXSs7oa_jQl-8C9T2#%jew%vnU7@SN15q5T{al;
z0VaSERHioXlre(OfdkiZ=zPuPCFD9?aB9U$wNNaV331U&j|W?+X6o^_Q(fR_b|F~j
z)X#M1I`ec#A)yI|FEp`IFIO-&{DP00s_=1R(wnVd3Yz5uXi*0)xg#78wi&7QVhyB_
zjM^5$AEN}OWtKihiTpY5b;+#$D*bt62!EF7&;MM9KmUXN1oYwp{&a!~Dn(ZnGSk7z
zU0`J{u>2QTUfiO$2Zn`J2(x(5#m#=4{m=%uBhh=-4~Di@nma^m+c3!8$7v!ldKugk
zey(U(8e7b%h*S3zXY*~uHL>ZDdG}tXKkI0P_?!5kpZ9J0L!Te<=<_5VU1#Ufb$1?}
z7WU{gu}4SR*da!bw*IrD0UmAa@o3|ZM~kx_E#7(*ul6XO?a_48qv>cT*r@9>lki5I
zvf}0zQs*fITQVQ-N<D?f8cT)#*XA45aEP$P?UBSV_k>a7grv#}Z8PeJ1#0<S;K}9D
zy2Kky0RXC{Mu#91_O{fhR=KLewvXK_(tizl_0H~t0ElL>1*n0hwSniHV2G_o^7=5D
dsrC9wy;SX#0cnjz(&))D!M{LCzP35@{{b}7@2CI(

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.doctree b/.doctrees/autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..60435a31b3ae9222089504e1d16dd855bc66ff55
GIT binary patch
literal 25636
zcmeHQeT*bWb>F?)@4MT(J)bu=>pO-J+jC?;f)gRKK0AB_$I;n~L>#bkcNzE2^lW!u
zPxrV#e7j(yK(KY|RKVbN1Vo5HB!Y;EkU&vF5)6S5q9jrjIY1zwgh<IhmXQeg5VE5D
zUR8B>^-OO~&)lAUf8?Z_nXay?_v+QJURAw%_4d~$pSrO%#{aRSVb`*~?Ny`I^8A(=
zCTynVH{*`!Mah%N>31cMBy~0$7?(mn4w`1d#*m|F*={Q^z2ri|<|%t*yWy%VKOLIQ
z$o9RgxHV}_S<@Gix;4utBRg_U<rTZl?X+G0(yFMV?HZxgFro<bQOD}0>-(KXPijK7
zp|Xkl%x&Uwq&gl=MBApi23is&)&iStWKFjYvMJf(4L0H0o|$aKn+BT$Yk|EPN01ar
z=AhjQNea_VFdiZizb8GTBYw@a;%@%1X*4eeW}EWFY&tTw=v87JwU(^I)=kzy>m=*0
zS$@Y{3u4bRgSAV3aB=OPd+&mH(l0O=t_jB0j5zX*uDvD%xW)yz)+4I5QUR{nUd!BG
z4MH^g)Y${^rZvVUd79EQqSLIGqnXw#An#Y=|IPS+EB>FRCPQk0%^O;06J5g&61Lv)
zTd`~2UK^oP+0Tl?D<hP(rZ%F)h^?QrR&r@>Sf|iu)*%7n9EHcy1U7qD>~XAy0#b`=
ziUbJid%fr#QJ=A=M2I|f=A9zDwG7$LaM>P<jLq0Jf=1JK8)4UM4k6|Vo}V2%fB!pP
z{m46RdnEl%h}@fVa_n6%Hg?uu{hIYN1i~=xbc|r<PJjslXNyv$U~sM~uM@VQJT)Ma
zq9S;54*%MIaLEW-_0V+N%QeexwM?%OMh3>)axE}JyA>Pma?Nn<O?$bvy1IIfe_l9E
zg&W<#q|r?J!)tVm@M8UYkA0vP`9afKeS`12_nBVk2Rtuwx<=D%i1EP-T(To8S2bQ8
z`BF_Y*~_gQ9J6dH+6w%*+t;C_?55G(L-zIzid?c=k(HdCqYn1a!54ucKgH~zQK?k$
zD{2L#dN=*nPAQ`Y1OHO#i%A;e{bkU$SqR$L3XHC$6l83@yR+_ZI_nR}j;x&-el)Mj
zvUfgkzIL7?;d}~(^AZb4NWp-g91Zve9L}GU-xtoG=ABoEYCGIDy_W55HMXICRdzS_
zm)#UC)2<zvf#E8otdH<HYmrUM#wEFW`Pkm-1&WpP%no#S5;oP{S%Gj9Q`N)N$sYwu
zejC%%o~nsWq2-u6685?)oi!4y2B@x%VS9=7G@Hg#wA01P;2`nF4&m9{?z(o<j*`p#
zHAb|T>;f`7a()<X#%439fX}nZ24^BNf-N(`tTQ^sxEB2oX5MQ;d!8LfM?Qt&zQ}GH
zUIi+UdP;?qxhE_LJ5O3CtuJGyBGOM#F&&Pmz<ZJLH^AfjdQ_5JET;490HRCQvlR*H
z+1~{>kAa)Vi`>-rW{<ry8}<DY#z^|AzZ#*s$|j#oEqyp;n6+LEeuQrNos_8cKT;nZ
z?`K}Y(T`Ld71c}K`uz%8>A!THoM|QNk5r}NOpK3>Ro5%l7p!NjFK$?0Vs9Ej50o^Y
zKz?nhgdcE~uvosDQM&5~YIo_!4Tin52sz7<|NW{X|G6t4`OdANtHVXAWk5^ch!F<=
z<Jn*vyaF6a84RvgD2Eb7@Sbadw;n~|oF$OZetVJq_~5qXV07-Kf_z(BwsESYKAc}1
zfRw#7$lFkE6b<srBWREcLb>KNit{EeSoG<E`e~)lC|T_@{qQ3lVp9pedT`z#3s)H0
z&d&+@SGb07-p;EjPE5UvW2{#sIelxSGSO#x50)F)kL{g_P0*$<fl&psh>hIsUU9JN
zm2Gg%^}Q{zxioqiZ=v)9y`|MmY<9cju7_RhV_}rT<~KBhhfKbt{Z*7r?GWerwmfH<
z&9*bMhmo3fiiy(3Lv5HtP0~D-Su`{zm817^3sCKizDG8^%&^jzp6!2}KbE+$PL?SO
zwZX&>EG*6+W+gaF$1x|ZW-A+RYut17Su|Tsh1eFt48iVB4G*Vlb=#{6OVkKftkYMv
z8!6balx9j())<?BiMHQ*M4Q}sdeaDPSQVo<P))4&-w1J~?JF^?tV;`tu#V(A>)v#}
z&F)J~O|{(K()-pmcFdsBHNt4U_U0~GbPf01eR0<{&k3tcN>hEhaG`cbZNv9W?D|ck
zB`eN~7qVtK<Irq0jizO;*EV74LHb>W8=?w7j=FJ_o0P+12E|Ii1xR*qGRvm6Z)=kM
z^azuJm_-*yE1M6Ef6ahiHZ<z&5c+!s1D~_Zj!+Bu2AMw>%mmKUsLJ^?J}`)!f1!Gp
zobP1S^Ko<n<7kY{ayj%0A+uZTX~%H>Gpntf*%@M1a5_UY1~K^`8YYi(CdmfXF=1<Q
zZJ4t2F1!C2ivB;+i%92?%r!+OJ4J46f`lfpJ0{>n8l}+-1|{2~ER$pQx&y^EL~(Xz
z<6hD`(>V)NWXU;#FWr1s#PA_eXqp#t4&y7G0@zZ)a9mVMLJZDvHUE)9{&ej2q~<ML
zCdzxRY@Jhhy;l?Kc*jTC2@YTqZkHzG*yuSjW8>>7ANZBiKBJ}0#=HH}U8f>TPDVPL
zDR2})ONV^iu<=?(w5AZfcL32#{Y1~IM2pfstZHyUFEfTn^Y$VuX3E?DfR+_?v7nk1
zgry=kRZa>sV!?`7q{!`<WW2ZAcC%=_w4PBdk+dfH5X5^PjP-)I0u=IPS-KKS57;WB
zeK9qK4+i>7nl<g_JRukHV_Few+w^WO1m|{qWrJ}KASZe%zkt_$gGeG-Qn{x<r5w2*
z*HgK_jEaP=q*726MU3#NusU=<r{!fY=sF&X<$GDDOOIrNPp#iYo`%NlXHefzxD|5Q
z2i#s)vY_O+tQYmgB+r*3RzBr${iar|TDU&T+g=H-|E8y+gR7)c1+G8TQ_;ayQrSPa
zP94<EaxWEJ4}hzo;40*@4{*IH^GYe%>XhMjzE~gyaBp&s<X34Wu0`@KTr8E4{8~K~
z9g-!LDoDOdPeq4hNoD_#e7}~LmkP=E@h&T_jw|%>J|Ovabz44gH(q;~Z^q?LyA0@a
zd3`C!)ubHY4{JL@+xTrRtx5oYTu(&@a7m>Kz(1&`q64_3QUG|pYCnh$u=<xBle`9H
z)&40hha<yY(iwW8@rkk?JeLAy{G7u17qpx^x*il%O2kuKCgqs?8!fGorBo!hKf+qh
zgte>)>z}m^s30t-3f!O9Q_;bF=q`CdOY55I5}b>Pv$63DN|zj)&@G+zt4mJlsSNIt
z9&P$?)|Q8rwk+$Z3?7S>;I*!&GPo@zBYuKx$VyA@*3zg5E9vE=ZomYc5r=qjI<GPX
zRvE2LZBrecIK|`yJMcngq4K-%W`ohDU@5vIL(3hCj6*G9HtzRZiXNmQOU@8e@4K}`
zs<aM%o3_Wjx)tM>=tnu8d%^PoElskOwvE`08V?z6Y!(~It_$OC*AJpb2shd2!PqD7
z@m;dRKcTk02n7|7l}$l6D+ea3m2L1EZZ<2K_N$VT&z4VX#h``nCs8!s?9n;pM$XUc
zspxc$q*7&&d`(Y9H%KIv!V2!X5`lAB)=P9#ae0u-_gh+?R7|Mm&8uxA`nu@4Dnfl;
z+W_s@lvIkw!}3Y#1w9oVK!#|`+@y9I*8xOODe3rLy>f$=M%4~~8{|<i3XV7@3AT%h
zsLG(Of{fd=E!84J4z~cU%+1Q>6VU7QRNjdy`pn951{y+jZ0M=zkRhqOUu}za<~gsY
z^^x8d4W4cU3&mbmkyd%S`LR7t-G$mKT5)Hyyx9bkr{oGS*0)x7w)P;3R-9mdT~9@)
zKO~hZh=0GHiVh``N@0STAjbwFy#C$4c$v%bceS+D-X3Zq5oCs5>wc7X%@DS(A*B3g
z+LmhdilkCSul%*1icYT#(UyPIQ>lQMf?k>AJ}~KclTSHD{#{RP1kECvx(}M=<pbPy
z$~4SEUzxP!C2w*mj7@20d7Z|20RpV3apv_@bQ(uesiJXi(o@mlM^YJG<E(0Fzt|e*
zRp4f*bD|-ne7ClxT8$&ARM9wZ)l<=FoFUrM(o?B`m|-+dpr<y1#t}^&N#j&0#BS_h
zS8q4SRscUn>28Kc_?z(^{3M;&mU+@)876Ml(>2(KZ1YkR*Xt8@H2v8yyrxyKsSLW>
zsoI=wZwMFFaD?06xS<aB0lSqr{|z@e#f6?M67QoPQrjbZ`Epq~&!P}_8^-N-_%9`F
zj{N#?_Yc?pi&qoblwdNkp5bl_))bp*$F9rW$prDaM&vh1u;eNH3{QIuHw<~Ia21?Q
zJy%bYW3%&Zatb^t+!ev+H|ddITBCcO++x)fp2v!@YJgy=+NVXc)9O-J<h<GBol_TX
zL}-J!HQw7=c`4HQJSt<;kO;pQ>HG;ja*s##+4EP#jr(0T?fL1?`80V&+`Qdohvm!k
zc>y_O6T>XiQO{4)(tK1C(F@Hef|IjA>(`~YAUW<sEj_MB^PYD89<{Mw>c2r%zj7z)
zeY(n}_osHgfd-@-N&%9(^G`||d$go+!mqS=wqNwTTCUk1`5cFq(S<*<IN7tkh|;4A
z%Qf<cZd^2Xgt0JexF+rZSnKrmbZzC%z6<MevGYBZvzO2*`MWe$_H=)&JI{kp=ll2=
z?122OtSUaD7qOW7q1wu_+lkUZTlOkn)IFC~T<VMLj6rQX<M`0<DE*I1c$`AP6gTX`
zRRx6X`@<^5-2Jw=#m>`lu}3`Q5;-uf>ou*B=YJv0-nEpvo-4H%{D>DS`sH4I7n+xi
z-pWN3t+<%4>#67#^O8!H#r$jZRCG%-Nu{uuFPfZ(^j*AL%Zmzzq09+Io`yCn{5<cn
zA<POwF2z}0WxhI+bsD~Gb9zWAkZwy0Y57QXwX)WZRErCw(izL!^;C2imQ<=>__Cgg
z4#Sd40mDao=29W)5;J?BGt7szjHnnG3c#Z}!~9lRN0mFn{DGEptuuk3QX-y%SZXMz
zwtuQ6F|u$tm7xc9sc6bwU;a+ZcLmW@@=f@zo{A2`RXgB0Es1NP1Iis;#%8pO3Y`)W
zqA9B43c9UWK$T2o-Myq)f0wE5#q#{A7T@c2uFy~ErKGbkz6?bxE=rd5RCIkIsZ>$u
z>v}4>!62y=6gpQCd#puzl+-<1$|@3u(%_>hsk?c<46*VMjUNfURSG%3iRHFxkm}#H
zWCoJJ#06n5R175SnDAZ5W26ai_a8I$XUt)~eRVphpXAU}N>tKYDqxDLBcOIHIsFoJ
zo`8_kecu41631YAMQ4Akb<cx(xwS55wa;4j4=5z8b<W4|)A=|)ihJHqASE-<Ilo7l
zh?4V3kxsuKOPo*Pi=8MjhxD1{79qQDP+4x4b3Toq6=kLv;?fF6F~m8~fTZ(T^rG`c
z`t>>b^`-3BpX1lA^JV_|75@1b{PU~)^Dp`5*XWbjxe5le=HP%K`(Qx3N|8*i)S&hg
zRT?iR;cLa-^|z!C_Q&4!S6Nlu-gQH-y{k%u9cllnk~P=O-_8hMu|2G$4F!7`MU>$@
z?iD$<sBPr|N=dCRt+xFRBv!6%b5{Gb?N?A}Xl+ZBQf*7WQ*B#qtn?g;+yL$SRaA0C
z+BdCWbnSa$Hr2k@(=V=l`LF%ZzWIWP9Tg``*@*}|Z)Kd~^i?fA7{Mz|SR2HU$*Zkf
zXVRF`D?F_73J<Hik{SJ1c-Yw9RGbH;#7Bv0-N)ZSp>?>bK#=C*YExamWU;EPWp*M8
z3@>zfJONDC{uZu$i&)i@ksl!_>De(hm*z2Bc#HudYn;tD<A5SzH?p|bhtm`h|C-In
zvm^uwU&x|<w*!9O)<SsFx03I3zOH0b!XYEsu#O^N=az|>*4&L84=sx58yTH0LU|qv
zd_SU<Mcdv^*!UV^hzb_y)b%1V%*n?%aPf46P7EoQHoe8E;d|NfHY)JoRjnjnW=ALv
zxfgdD+c-h&AQb6*h<Mdq#9;1Fbn_5ZEr}G<HF-*dM*vPEd4rNHAVYg^u)~<W$dB7K
zA2K1x18mMR;PWjXC)t9zZ9{$tR~*x+4$m~nCR^rayajUe6bO~#(4t<`=pbbl9nmzY
zF~pxR5t<%Ois{6T&BjrC<+UsPkoBhG%Z3>Qe$cQCujQHuygfy%;{25zl}<2uEt9;K
zgq`ep(^G}>77SE2FY4_teN*v8Phob9+5psrmYm=dcEmNjt=QN?TpoV%eE{*|*>!>0
z^n+HTWl~IDkf8wL8*B=Lw2iZ5NW^ZXP{48VJcY_?<VoxRj;-NrA9uQq42%#gPk>F!
z3=m+5Dm=<lILxW{^Q3NQ#x1{rz*)$341sEW1P)S5a+EaT$Zv->-FZM=jg}o!v}!a&
zaygw25)8P*XiSJBRuRDisFB$kZQDiT5Gj3;et`(Wb5WF!2Ez5Awidel5CZFGk7vi(
z0{1fz=Yc?f2O)ItJ$5Ub@0#5P#nr|5Y;u7iq;VH$2k$f}*bP<`b;I?wH6(|t-JQrn
z?0i4i!dSYVa&iWcvk{pc#J!GqyfCx7a`R994CFZ<w3|KEQ%YX_Eot=_wgy4=q8Y5(
zX4K|1pbFNksN;gzEJB%K6!G}gGwdWViQxU|;LI8z<BsgIW8sb$8QTrZ-m+Z$2Yx3g
zAX>r>^Q#gKc~{~xoGJ3F6AjmI8d9_;n--ihn&@~j-XLWzLgc=PsNB2!&lB*!IUv=*
zF3{sQZLbXsNWE~J6h#<s_GGg=gk9G+#9*X90&p<_WICH>3nO?399Sm85AU+;o5*#s
zVIm~5+du#*(zxiQ%WOgZDFk5EZ3}I-%MO~}L-fZ47-DabHo@=(65#+~V{Al-i%vm}
zIc^rf6p$5R3Ib|y!-VuWJ6=d7CkdX6+S=Izg?pyIOn~gulqi1Jom*%!ekFZ;mJVq@
zLm#K;aS0zWo29(mX@LH?0nfe5bMEq7*b%*g(y0Q(ggb*U;8Svon_r{p-4~z9X5@7G
z#h)4oF>NQ8Y3cGos!RBDOr(+Kuqr0btUEbd+*5&jo$z>74((|9=G>vxmP4ztm>p43
z#@l@a@u$Tk2w8{}QYL9WAz(FyMgUI*Dm$A9)L+PYq9-xS4gVA|`S7x{DOLV-yih;p
zg&;~W%CGOOOTNLx3{}_kV^U<81rbTm1tie<m~p4yDAS)r-#nVU1;Z9@Gf)8N{}4zO
k1T&3SNIsM$bFksusp-Zo0{0f_r@0m>WvRc!!g2Ng03v<os{jB1

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.doctree b/.doctrees/autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..065c7440ce462d1080e846020f6c63953ec8977f
GIT binary patch
literal 30168
zcmeHQeT*c>RlmL4@7>$IJ^NP1=d;J0V$Z>TU@JJu`fQ6BJ6M-HD~Z7-zN6hc)3e>Z
zJ>5fp+}$R&ln}?cg;IiX8xbHF5aJJPlSnonwiO&cL@Xc>j2#jnLUDp36oiRM2p>Y>
zy{hW&>Y1K5-LvbDKRD@Tr>m>#_ui{l?^V6&e)WN|hd&qX;s4lt&@nCd{EFUex?a-=
zVm8_I8d2MD!}y{2*azeL;yRo1^>cw2`3)mxd!R<cw4A1Ixbc~o&0+P>a)K4vej+d$
zq2;-0b92lbHz&@-b#sc1g;wYo$}M({({4H5xfRhz%h3Z<)58$(p^ue~PI#@>RTyJu
z(AVf4#(5-im@YmX3(m$nI`9$3<^r42QpU}FY+Uktij6v!Ys9Cbb)8KEVZOB<g&-%C
z%RZ|apbRD&Kz0Cy@O{kH+v3|~GwNg)>w06;H(FRHViTdhfwy9F-aKs1nwOdT%p=TO
zHNCd6>PN0?_^anUe{=PnCvOGeB!7UwU{w&dsz;%xcdS(*zEv*1)h<b`7Kv}wa+}8a
z6+Zy3FF&z2S~vHwF<zz&ix@HU3gFVb5~O_;{=W(SUk(3{A;%yz-{O4M*=R?%{FuG1
z?KLCExV~0>MA=(K+vVy_&GA!VtVibeo6DIfPnnm)fSEHwe=`L<kQA_~TO*gNX&}^T
zQ7w_$0sZ}u7#h)^-d7$7k0W`bsBSKSWGA^~4}|)9<mkTE@Ek4Z7>x=-9_97ffzx-r
z|C;;Wf9-wA-x%R;&*0bxUSq{syXLyJ<EVf^)Nbqk){US^AaAxPjS0-l73Fr!7L+Ru
zB$77)4^P9tmgk?-{bv282Y#-0N7QkQyMNRL3vhSn`3-aB4#N#Rf2p>zvT_V-YaQRf
zsjD@O2AC~BGZ|W24>s$8;k1@&sU6{kS!-zP=nX^T(*r8G+BqvUGdttf0VFXqW8L)1
zxiQ7Y!wuhyIz1zbhHe7JyW((9LX&e=Gc@B9(>TH|jPNEXk(Vg#ph>B8@Lsh8G`s6w
zbE}ZheZF_D@WvQU@!mGT+Z0G%-|+R0sWfEg);e2j-nzYZj~vR{@yfG#MK-<hp3}9{
znJSzXx^OzD45#^co<7FMsSZYLwqv+W%iYk<gPm0=JGrZrC3-R8SfSzTj)G)<g1e?g
zHX-?lGfVPo{VfStP);)2+u4fQcxP)F#26c@!NuWx1oYvTzzy}aLTnru2Okl$n|A0O
zVsQD`Ra>{**nEslz*V@_0r$TTNn<l`ZJh5oR>KP83;Z@rUN_zb-Pl?awtOhR%qF2d
z_oneFPD7~s8%7AuWoTXE*7D=vUf&2-^27)X;=?eh7unHFzx$dxMSmVv^~dO|1E?qu
znMce&1`mW-Ka0&MrNIUtg^E7|M18Ew(#Z8+;y(v~9yVVn2}F~>4b(gU)ckawntFfo
z*dyr-AK(~<u}K5E6t)T-4p^kr<DBL$r^`nDV!{!-?{y_un6O_>s4@Rjnyo7au&Suz
zf2-=4H%WE#(Gr&IKPX*EEt&cIswJ{VNA~O~Z;Z@8FrPI)bISZf_O9yIV~{9t-n^l#
z$}8Ne943}N%J99S4j(-o!J62MAa`j{UM_1;{$=L|#l9MFDLqMy7hus$3}I6~l1|Zq
z%f(@=%D_&AT}jtBneF)C>rPmg{jg7<df=vfkzF~EuS^T=lh}^$eTz0&rR<1(+W-~m
zXG-pbcEgyG_f$6}gGKU2GbZ*eT&gghd+W!P0mEW-!01Vbmczmpd?{kTT{bQ;q3xRl
z{tg!;_WO7@dH2-4tcJypgrm8UCWNkZKdfC~&(^;Z8^w)b3~?9m9~RVZb&G=?QZ~yq
z$8$Ht22}4Rya}!E>29r7VpHea&RWoctu6%ZV8<L7{#k>shkpXvCc%o+eDj`>%%)mt
zY(!5YKlwr#20|XDp(h!@O2ZyR+-3eG4-(aG_FRndrO}t3@^t59KPC0XJW^yORECir
z2*cRFo3`L29RyFbk}h_5EVA!XHXqp<4`4qD!3y?{Bs_7fR=3=m*x^(UbRFB#MkW{P
zQkay{u03oN0=T^aCfpm(;T<ioAnF)Kz8ZkN>v<4X!kq}4m{zlCxLOeE5a?X1-QK}a
ztnS<`!ZIQt17S%$6j7NoXO?P?z6C*;jvj=4>L*oeH`GpfuA!`y`9fJ~)!TpC#x(S1
zh3bYYVnhA-QA0~-Qjnxox|FYg7zKvb&>N<q@U!lD4)%Vl?gYTM7loZD%;Ht5fOA!k
z)s}B0B4O%aB*I;)-QqY_C$IwW#o&*5RD_Q_hW9bvoP_kvF?MZrM<)FK;!sVq`TW&%
zb{5U&I-3DCT?T8#31<6|ejX9v*SV;O{TOs*KMo%-^X;!-zlZHFr`_|(e=$t{J#30g
zp<4uLn5J*o#{PENTQRYdNUWfA5-|o5`R@dg2RV@#Tx%N;T5>E1(Psm87lWq%KS7sj
zZ<Gm|PBCFz({WSAjE&-eiD|P(s4^O1ty>Y5YKyW>HeW8<n_qP0mosd{jk}iy_5wCJ
zZ10CZ)B+(te`Z9Z3Es$_g+GbUWGh`JR5d=TC#c=at-FwneNgRUf3Azfl5~}3OfMlZ
zeYXbo<#4+n*LAqZC;i2IKB2rjxS)yjuAe2@$XY@Ne9N$plh|ye(<|^bY;xF6p;Lbg
zzxm+uE1Sk_q?Ur#6ws3cK+pApo=`#a(!Qe92$|`oQvS5C$i{d>g|2%@y42N|8mSQh
zpe$@WCStm|kgkN{!DRfY#A;cMd<5BiR*eVa>h9wp<kLV=v~FJye^Lb<xjPl=-3sw%
zN#gBy!0jF_l0zv_3M919vp`V*?<WEKNy~PjS+*IqJ9vo^_$2yWfVn$j6YB)n=Z&0E
zxcCK<ixOh9-w8crgHx>Wze%Iq1-E+!N2+|G3f!Bjz`Y7DPtquZRiG3ve?+4U=A}T<
zMscC6aPoN)qoi)hA8=IV5;#NZQ?_lyoni^QoNu5~UD_KhtGf-=(8+bM^g?n)&$XhS
zH;>a<4~f#KMVTx3HYszG-ETtE9y3+!aQ5vZogJm45~a+1TcA-W9hE3Khkj{)$qSJe
zsC18aIZD!`YKQ7w2HMMEUCR|fPS2PXyOy_*l#@16pcIH_R^5>*GyfJ6YiKFu;VK(|
zLSYp<lpiM<C?Tv;MtFlpp%lE3mtuz!(J0hpl_+H<>xXER!MqfxSGPlXki;mfR{scg
zF1J`2(w_+7<lRFV16xL#kCL3)lr)9ugAl?(x|{Gi@Z1MX)!eH7hQyJK+@Q?dXGmZ+
z;`OfTpTZ`G?Fxqd&q$Cm+`+R<mdl@WR5mJN9A#*vTX+7B#Kf&`OOKpTJF7d9k!O-!
z6hxhl=Z9JV508aE8F~0T<u-<JzNqq^R|QoaORp4GyMuT=Yxluxg}Hnm$uM^pXKRLU
zk)k10`rn{wkN+uNGwdB5dd(nF%1oRCGzv9wBuZ}0a7jP;_gyg*?0hmwe=kX$sxg&T
z4+WT&o#W*|8jNHa8DB?oK~7(ZlHUp!yR9FlQ7E@nftPpFD3rbkloD?19ulMMsPC5B
z5@hFXRhc<nMl;?=a!P83oPAG&LCN;v#Xjl+jq<0^MbAE5I;jel<mYJ=N;4$NmsDOz
z=kyyi)<1P~G}t2v67r+0YFcTJ^nJ>2SpAG4v2mSBn<suqDaj1ur)dFD(bK1(X-OCK
z=QIjsQY1>5N$@2ag;E@e@@l%EZ;;ffD)O4Ups$OOEF<IZl3b96MxvB)LHoysUb0l+
z<uHvx>5D)q;ew8l7-dI&w_K1QJ8zsw7gS;P-Ar;yYKBB9<AUBrqfjoW0xuekLTQFX
zDer>XG}aI<NRW^pWmVHkyP#{5A5e3@$gS0cI}$O}!P}xkg}o$Rsn6ZaLp?ythw@NA
z1x-tOsF+5fJd{K!V?BSJMxpddqU1c(MFj)VcYl<R;broDlB7x1n@WBu4_A3t^eHiP
zWrX@1$pC41BuW{V^aUD)QjiL~e2qq-6hxqua7nL_7-a{1w_K7SJ8!H=msDZ8&5o1H
z2uhj~rHo6ult!UkQUzY>G)f6x%DbfNXsjVzl3+AH%BrT7c1dqZoJl^`miQ#)5tG4=
zX|cGk#547~ote(tX*p4@=>#+_>6*^aD3s-qC}phZCXGVrmqf|Ard^KT-A7WVYEmU9
zwHxufEisa1Wc(n>1?j0IN*Pb}2#rGNO9fsYr%@<<5hx`*)iWeU*-_suPbJ9C+c45o
zRhWG*lAMy7AyLYBs;|;0l&7k|%QtBhN;4!%c~A9S8fyqoB}mARvZ`sNJ=NvPBi{J9
zcIJ=x^W9>do`2Y6Ds7VZAf+U8JC{t5%Nok<TntT1x}B?N6v~W9lrkpq%`^(7ED|N>
zc4ibw_uUDNVkPCL*IP*%)WNUhTm($zozL6F5S0<$dr87ct07U!_!E;xp%kG4FK1~K
zN)ZG~34d}wiBWc-cgvp$vh$XP^d}Xj)vu78lA0k=%J`GVXcWqyRN&=l8impfiIVdt
zW10Ddr&;AyW?Y}6(T4CSf{6S$tGZU&qg+++Mv!9fk~fEnJj+5))5OCR+{uLT6<QdS
zr};88E$L}qrco#ZBT>p2%m1ZOC^eEOJNGp6la%LHmEyJVG&4XA3{e^39VH1T?TJJw
z<7uv;Q7A>Iz{^cE3Z)1FrG%$BNn(^8=-u)(g6zBnB0WuoX|+LeN@|8gDdTA(8in#S
z6?pj&jY4ULM5*Fw9;DHR@HB#m{5Y$+MtGVsjoA0J;Xyr1!^i*N4PN>3AzZ@Shy0<u
zF?<HT&wG`#{GmwSfR~EhKz$zaZQ|+hq8x4o&yB+y#NBVMmUuPk2~yOmh$`9LVILm<
zB#_hN8;c(v|13#4`8>2hDKJ*~pJp;||00Psw3PC2mG?tZSY^D_Ka&iQwy1DOil2}B
z4vj*2sS3P|O;JmQ61)@)e9uc27m*-CkN9gaFRmsj-`#ofgCy3_Qrew)kzZF8zf^G>
z$v_EVb-hBllx!YGmXt8cojXP;Gsg`Yh4RsT4dP_tQWS0i5@+Z^&!C1igk=W&BP8)9
z2E5F203XAD9BpBd$A<?cMY?Fr@OsuV&S8R$n9V0&HQjBPxfde_<t$ZSL|oHEUX`GF
zQe()w46n;t&Dj19WJyqRJ|R25jB-=i5vd$9Y1l77BcAF8ayrBN;xU`Xv<{H!5>i>`
z6LGR}ncdQSmcOQJj<d;D<T(6|egQtMg<b<!QkdQ5Szh)4B-G@kB2&avl5gc0CgZXH
z9&<t*5!ot$&M)F6&)SGdP5A<=A+qHt3#<Wh#crPxY$x3%>2or+@r}*8cvBmA5b2P+
zxs^FU?axAIYyu?0(}3E~;U!Oap<cV*vPdMo%_dwg`8t;rFN=(_+iX_eOs*H8hGa2F
zE6uxJQkK-io)9lE!VnVkh~?=Qq_{wGJnxUp5;R=K9~-mbFl|8czxs}|0KEzc3F+)W
z_TK_y$&=thdFu8*C}qqV#I(pglK+K%sb;yD&L%MR4kWkX7lCH9*NtYdRKt8g+NQB3
z-U13V$ADy3rp(xDzoeqSZrNzQfwQD<qKUfwuRx{!GJFiqJoELmD?XPOVKeY8mB)eU
zHRZcj1M6jr`~ONI3)9osdtmQokH7~(y-Z0{K>av0Omv3*$Zn0ljAYKLoo5M9^1q#B
z<q*{s7a8;O)6~$Bp^zD9T0JugZ<NoB()}734q={!q8h+VcX7tn)x)!Obhp@=QKmH?
zuRKpZw}GKtS;oyx!*g@}ZgI0eXO{~2DX0J(b(Lm3-8no@@7XP$7K@E*0drJM;IiNC
zj9qVdb~lG-H@{9%m*!c(%YF=2s%ns`z=(YUOk`rj2CSNRm?+7)vMa06{{isg)o4a(
z&ua7~Xe3sn_HV;i`;+jIx8F}dNxA^FpTtUtWdDvR$L|ls_V2+TcI6<q-m?Z>gq76S
zup_<(wV#HsCFQ5AI+G5DS#{cf0g(1@f$G?w!*7q`x4%rk{S|!Mwm;85|C)b3&p-c$
zf4;y!U&K!&=T%r8!m$$rkB``CmdIBn%j`~9h(}8>E}i=uV{P{*q~WTxw);%l6<^z3
z(!I7Tk@r}x?n>0mOz-n4?2gxW1svqocbMsyli2UjpUmvy#FD=X6!iMK8r&B^V8sSE
zqqN81J_C&^8(c(63@(083~s4ef)Z4={g;5r{WSEkBa@qSFuci~nMq9UEPmrAmw($0
zlbiW-HZNi*>7F0*L&yWBl3TT8>kqexWW$izH|w@;9#4(}Tjs}ME%W2BmeU>UGM@L-
zT?he>1bc|1$a<*TC-#!uNWj@z`CN_aA|^dy%v@p@hraFx4$n^r?%LacRL$aqv9Zt#
z;fR?Nd)RbR$7sT31PWx1u(?L$<N0M;dRp0RQX)<*W0Nvg0tdS-q$jntd>&6~!eKK%
zhWR<NU+&nrc%&pgWzNF^EE@)#M8;E{z(o_&0f&0K14pyW_?{Qy%A#eRkJ-p790?;x
zz@HH}p~AGhjC=>K_G3g0Pe8-F5Ec6<JJ^B_T*&;P6klNb@$5J^YHQ~qLe_?pW#$4n
zW~>8;XSMM}x&XR5EJ_T=;3eYZrsVuJK#~pQfo!MPEI2L9HR2d&4G`pBHf`#VmqcEU
zu?6G21@ePj#}Ni7c%?Bm)-=|m4S<`&b8YZYA?Vl8+fXtE1JN*$8Kh6_llUtK7?EOA
zQP^63>oN}_UY5V98NToNnyI@@$AI(F#*tKrFS2?07*E#A81E%!N4oBGbz$EI)0EAL
zetTPAmcP-}m>obKK<ffi`t_LYcXW3n(l;QBF%RtSg){!xMLwLd<u~D|Ej(ZeVDQ|x
zQ*0cjXbWPcAQ7vXr~-sT=P-S_mKCwR5SE36Fj2dsrOF6`<pmaG0rw%dG<G=8OSqcj
z=x0Tpz=)cj2Fc8!)&V$o$b&S?YD=yq4M;X)1s1$Khka>HE5LKcfDuXML^4TW!p#CR
z0fZ1mMo`ciSxswM4loA?+%4ia0D+Y0m^D;`Y{bx86NY>S4ns^YPmHhy{=>NlZGrZ;
z;Y6g92d-vw9iyY+F-b5#8(d&;K2-;_9caf%!7eewuoJAUu0nCJ(%A}4IPcK&H()Lu
z!kU}_sHufU8`hg4&*5%#mf!RPZw7hJ`K?A*_gIp3e^t^wOj{jQcGK`zEF)}j3@`<&
zX4rNBYzorL!z|+Gj!m*7pp$Jaot!BHGLPvtI}mKSp?+R7tqs$G|DeC4n4~^tvpk)T
zCiD4RfZqzT^gf#7HFPQ3BMlRtGHAfSi}?mp<|2gJn{b-iHa}95pA#v~z&2=)*Rb3c
z=zz2f2hmUj(Rx=ktBtzrcsiZ~hJ>Lp0+qpdB+TF~pujZXIH_%Rr~$P$AyTPZjt1GW
z(c;1=U0@6H_)5@KrzNb}Hrr>oXYmL~HYuzLOkdDM2yk04H$phk4iBl~db0pTfo6r^
zfIw>?3liFKb}(0pKLYS#=&hAr;Ne|(o+MPCz(Vo8ZXZEcd?|iBfuYDx;m0AoJP02V
zo5H$0D2_)`^4i<H<~FYdo1iPP96NwK{m0?>wsE<`%~CyeZ;Owo{y1^Ia6)7NPCLGU
z3zv^!U*h+Q(IopntcQ^@^F~e<kHqs|Rd_mX8+Wue?s#onYT39Hi`afu%Q)Zr;W#nj
z1VI*}1dT~jPY75o!5P3yK`YzqaKd7)>cw4!X>IuZMK~Ycb}B*TR|j+bgJ1B&7-spQ
z?!M$5aLi!p8eW8k47?zm?%;q*z~F<&ow!++KLlg*;rQJ!ZQ*AjFaR+A0cffKOf6on
icqT38$_CN3h7&bWxi`>0P2<6b;F<vld^O*xulyg0<kAKJ

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.doctree b/.doctrees/autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..ed4806e1ebbebea564f806acabef8ea6cb12df26
GIT binary patch
literal 16719
zcmd5@X>24%6}H!&@v&=ry~&y+n@t?9%ql(z3J9D{AO}lei9rZK0EbR{rpDdvp6;fP
zU7I8lkb`w7C6UtkK~O~C7ZM<lkP!J10^%1!0tq2N0wfLzfrLN`r{D*Cud4bQ&$MUT
z4(v)h?yjz?_g=kMuijDb<F%*nylqJQ<1>M0n9kOc-fB5+iv=+sZ@JB=!<;aFDn9jz
z_`$frCw%>U;6{Fv#e4`onuck&eCEXW#(bK(ho&7Ysqte0Ylf!lq{EGxQ8&i!jT^=U
zuZ3o4v)m_slig|C?)fD-N88o|L({_$@G-~Ih8JoMbG5GanC8QDqwix|glENB9*HG=
zbx#L1VQkFu2`wex7~yqA^(r5=O^3y+(Ynqjfu3)!M<G^=R%gU)1+)@lO&}kj5zjS8
z@5ra|R^(+b>w0t3XKm^e@v%_fpii+eW1KLK7*`u3#!2ok8*Ya!`;p@?fBC%YZ!W*<
z?lV}T<PQ)GmL+1#dK9|4XD-V{To#MC+_hNCeHL-qbXsg{$qzvIHD?b+>&6hTiMAYn
zB<qZ8L8x&Z*8ElY{~G+i9{*1f(O5U%6r?uzsHdBL%ujdRR%ElA=PR+6-z<kORvK;8
zSHoD3j2n!_Y)w~<Yrr|<uoUM^9*-sseBw;x2&oOEXwBtYl$c<?+a!<Ve0twiA@Ib~
zTV;1+0qZ+1*7s<ruSd4-YfabI0*^H>VyQ<(e|~iBp8Icn@c!37nEZ_i+`BVy{GtE7
z1g+fo#+B0~j6u}t=>GPt5GgD<pHu1u+jJ@SIp(vuHw}wYR0?01#6MC4P9tD;dtv^g
z5%aef<~JFB&gpi<KD;oWY6-2Q2b&A?2Gd*O@!nH3yzTnub-$%1Fw9nqIa(0v*tR_e
zFD)&}VLtP0y~(s>(-0Ecc{4OJ9pp3sAW=cJZsgUDnc($s!*?TZ7l%m?j)DId$^LN+
zId8T?BR)Gx9CyI+O$e2n2=QK9H8;VB@)M9&UUysDJrN!8-Sa&^)M(@H9s}AYpiT7+
zU-yjMK>oyvx4q)7TPybozRsVnvb~qY*jw*gn_m-rT~m^QzYq=l3fWluFrHHBSQ9dV
z7LpJ3a_`zH!M_Gn-Vu*EEz{Z1wxF2v7{BWvF;2u~%r-;j>vj&1@vJz_=6Fp|7ia3`
z^ZnHgl&!#*AM&<iUiY>av9>YG>t^Ed=O84X#i7+#Rq;Bg#)%U1+xLH##IO`-wvKK(
zvGHX-hPQCrgViuX*zv=7XIq|aHq9`;AU<P@I`Ix<W_v}d{i<g<AIB8p1ddk)8lmoQ
zurTH;)dfvx*3aNfzC<+Y*<rBrY3%tq{>EyGF(+J4=Y@;)6%dl5r;L-vcX5Uh_$O&t
znMBcm$I<cA!1IYN-J}f4<OJRW=n3P;0|F{Xe+Sq+3Tz%LV$<l49{*Ij_th~{QD3Ri
z1&t+-qBV6E^-)$hkJd6R^-O{}f6qZ9CKc0j2>|1tN*RqQ)N>N^ue`(*RZ7G7(g2$1
z?`7;vHIebnyc)4ahlhp+*CfWbj2De>uNvRs)ee}1YLenhZRfuvcK*V}STs5Ay_Dx(
z`G&!Mx8|_yYRmt5&@KPR{cm~edcZ9clSI7$Ryh$BcKws-<{HEbtf&$OnOf?L)JVy;
z|6`X9hFyxoI!gjWZuuO4^&m86qG{btLxuO-%d;9tw^;AkLvH%1mk(fE74`DLN@m4C
z3v&tT9_wvldBNL5jZ-<!QFESi<y(^Uib*4cDPi5N1`n{^t+z<{`!p!5kBMoDXIHn)
z4C6(CE2l;&nDR~!!}fre*1r=UC08&;`~@eDiQet*<>1G2zVf{7IvdjE)VmRHVf207
zqw}5k#8$^%2|W0>h~jxZ3z&b72`l#_7@LGE)`ah#q0A@RX>>$SB0$AP6%HaACNYzW
zWu<|SlE}-<-6B{te~0T(&@YX@>}pTlBI|i2H^#|c>LO1t;e$Yo^^J6dK<OCHsioB7
z5y8p6rf4z3RS)1PB38lglF^A%^9|FPm!3<daM!8*IVgp2SB^{-@fzZzi0SSQGLhdq
zO<%OYL<lmB{6b9j4oO^}&`Zv@7OwRztFFUx;TQ3q^RERQiZ4^BGl?l1HU5T58dM%T
zb5%!YQuNzYCVELxgCEAyEMkucwD}ZKA>2#xUU0OnFJmg}EBJxDS-+-vPguW9rxzl3
zIYe%VPl&bXUI^v)_lW{ne@kcTjp#TbDzO}=`S(KbFC_>b69`fOr^67ZuuTMqb1XZ(
zACVB8%Oo!ak_RwK03?ft<eHO&YTYZA>o&*bOUBsC4i(R&qPFM5PTZ|Ityvm!!kWUL
zGW}U(^kF$@OboJ);7_7bcumAB1s&G0J|A$Fs%{zT3p8b4L&&-YpZnq7z_$azeFTxT
z9+;=iz*{_LSl?M4>$e^h)-@6N>w$Of1Fg91mf^lu9rt?<3ir#>SY@V91tpE5@Jtu`
zi7ue2yE!8+tWKOKh=a<S6&isyq3a=V8M=OczYgD$V1cI$@WZ8m`LMft3x1M@oUjyJ
zM9)n}q}+1aXOs?_1fG1!Kt5d#68gKXM|P-50VwPuepwKCp6iENfZRp+(a5FL-VQ1A
zr}9{37ExjZC26pM6~lCzq!6cMFD_4{T@q4rEg8LEEv2_qReT9Uqjl>c{7E%dA4a}V
z4&@1aE-WRjnc}2^66c|ma`5}*u%7FNwKv-&3YFM4$RK`Y2NGp!B&S+>oZ%IRrjVW*
zQPj&60O~UBPI<9&COh9m?mgx+$)lz_O{1{W_Db$h;5~_W-=ax_G^msoM)DeSnXTCR
z2MmRr+f2OueK$qYB*|9G`Y{FxCjp6ZXpNXpl6ivgKW+sS^Ao(TBL2pUye1>}MxBqh
zBik0|y@Wrhg>I8pmt2DvMcbpuGl*7c^o%J}octK;=VaWRlqMIJ{%7<m0@jr25vr^y
z9gUn8+hc8M+WLs#WHMdid1Q2xMrg<oC@7G*<Zg0RV#K-}lkqVuiO7vu$LUpAZuxh|
zU6i@_9X{r`$>Vg=yeQLeJN$_Hn7q%ThaxdZJIy$5(w6k2nFwCM!Vo@<6xg4r<-*Dd
zmq|HT)tn{kN0^PjQ$<u82jMF1FQ!}WELlGVx`{rNBGIsZk!#}*B&0|yxR6rr;}qHY
zHuE99LD<tvT9}UlrZx4Z!EQOD+X37tHwwKdB#{QEa0MoiACy03{U(pl|Jj3j6K#sV
z)@sB06L7WujGw)|rJK^J#MYg|@$}a`rD-;zZ+5~ulTOjoO|pidx~yURlwhM=qaN7Q
zF)$Ghj$1x-DdB%A0LoM-xEYxnR5TQ`c9@6V?$KK5I+Yp@_h9Hn`xx&NgNpvqRAK1^
zX+;WpbO0$`t(h?tR;S53D4Je2E1p*;?t)zVK`gyz%Au55dJ4+mmfjc3VO6m7Bnrjj
zI3Eb?X`uI#%$Y8Y-OECs%=#oIdcG+W%kPxaS?12aFgWLaSPrF(a|+7focm=ttO}fy
zC>#Li4$igudpVJ1Y<p>NwvCRKot{=1+Z2?+*><cPRt2_66b^uGGhG%?uMIbm_DS|z
zt|i-UH<VLZ#<%M+bU>H*&E-(a_@<x~_*N5kUqY99%V>IgIkXC#lSmu@=U&mrRPDvV
z*<F1TihF#?sK2kA`ZA8L4bD-g97-8S6_ktN=mX`@DsWUHQHi6246=R)+czK2=wTlJ
zA!i=%;Z7GW=N(k2;x6dj-l2*N*<GQEWek#`3hQ6M-}*Oxiq64*(1M`D5FH40YZ!kl
zf@B?{h8M;2qp>xLH>5+=edANzafms1(>Kxp5r?qG@R)cKdn{Eh@{mkWCGuch36Rz|
zA$QhQ^z=MEok*Xq#?y{<Qv6;cey<h3^Wyh9@%u{pC3Ief2*Ji6mm&@%eDE^f+1PJv
z;kAP;g3_&*QWW4vrTXlmq5wZgrxH<sE4rfqgWW5tcnRrODe!&A4|>p1h#$}$8i8fM
zvYMj1qQb~MX8nJw-QQr%daK<G*<EV)dl+=lYL_4-YL}i9wL9p(>52_s0iFDPOtUYY
zoJ>$%Ctq1hbn;d7By_TPItZPdo$Y)^M%vTBFw$?TkSqB#pTvjpspK*bZq8)WHjL9r
z;oYLBJX;i%XNzg<auG!|izwggZbx7|0bi~BZx@auG7c%!1}fQ#DgS&RAT8-L7Wn0%
zuRDRQQ<zord>02%AWQmcp&Oz`<LnThO!}}EUL%N+G0dl%kxvDKT3Q}>Bx#Xlf_z+s
z)lpA5n-)m6eUW--p<d(tRKpacIS5{t_Hn#w%%FN<gQ4_Kq+0RPqKc(Z?|7(dIPAM_
z2vm`u*^2q_GD<2W3Y3G|M2AWB8u>QfrYLnnC5QAC>66F#u{I`fkl)NTU*J<zvgt$}
zZ3{+B2bCGq0gC!O6sdKn5H!G4Cu9q=nP`y}QAxogAgKYW0&<lf!Ffg*Rhyk-Sja<s
z($JB8Rj)NZ%eG9cAIh<!AZx9)*m|@9aMM&5LPa^4m+2j}On?zhM#K<)rb9t;<{)oI
z$qPOah3&;REsBKG)y0n*^L^LX4Bctj4AqHsLKV3vKBIh+Y?hjsCFUo)zI07teE^%4
zPs@3CkG{J2p=&TdN;E*~0z)17F`u$^XCu-#kO~laz(Xjl<(K)a>H00L#i&3DV5rJ+
zmDjOR+sLb7Ma)(r3doyH(<MPIYvPBH!owAssN-p=Fk)dvgNeI#zJr7VrkN2fLd=Q#
zS(6vAsO4(7GKgMBQMcnFSCJnnq@;<Ix*3>sF936CEi<5EJ`kd?97{F{Hrx>q6Cm3p
zb{x7-scu@^v_TxoN$2PZKyWcp>`L6e!fY+D{4i>e!EpI{c9_qKtgFp%8`6OEcTiDu
z_tES5w8uP+%8al-n_|IG-{nEtfjeyqeuWW+Ua+#fjOJj;+YSws8N2=l_R?|cDG)$U
zEo2>30*2zIB=Z(u^SYa`p3{E2*)=`2WYfPknI79#Cz0J`{*uYUwtxXuuxx}K8^9)T
zi4(g>)bow=laR@dmTu0J0Mv-mWz8LaG}v}ReM>XU4a3HN$nPj!1&#R;acMzQw-zoS
zwVYjI&}_G<tED~JG;pn}3C7F)hE*0zh~As1Qrr=h+@d^Nsev6xkJ~hzHe^8Qg=3^B
zf@r;~o7o}RwOw6qMyi$Gj3F|eb=JZT-UbE+LoL=0Ki)*IO^u;K%+_!xkTfp1bb-&R
zs&UAw-Im&HhmSDl9MxR&aj8wPeIbd+g__tKA<9On!b`}_ERcd^g*b+w%W>C(^f*6O
zXr*kiXvS>q^o8o*sDK;Y$EZ<0H>?}zKyJ{_*C~zlHTroKz0Tt&;uF+Y<WH#vT=d=%
zJ$FPe_)6DNJ57Mwu%}TLSy$=#>_SVo_jfuym6IbF_1FQ9?)U<kFVE6kGPOOL+_8m)
zOqdzB3baJhR%GDC^+SssT#FoFi)=27Y_f<?<)uu}J%w^Sc?e=HWD6;iq@P^id<*RW
z(F&>TtfP{-(Dm}J#<VwFWO?|Au@ebY@ph~*Kh6a|jIqm)ch9B1;9!QTYq}9BGMs|A
zLv2GOz<ivzV|S?br@)&>;`d_P;{E^>0Qet3QYFAt;}x0@r_Dmx4x3KXj#?z#8>F8m
OTcnhg{*uPy(tiOzayYU8

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/flash_attention/index.doctree b/.doctrees/autoapi/lmflow/utils/flash_attention/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..0f63383119e57db22f7498ff5f9ce7d9efdb1a72
GIT binary patch
literal 3751
zcma)9-)|%}63%WilbJ~}Nfs6%?S?x9+-46Zb2{Avf)&3`x|cl$aff&~3AH@kp0O`(
zyRqGy$-_yD5GWCT!@Ti_@Yf+ex4V0O9E3z_rCnVvSCzl|s^)p;haZPq)t@^w1><sY
zMzTz*j2iEHnM%u?O6$M(kH7N&^QW$_$%Rp+PN{cW&`22<nWoZz;oUH(TQ1C5y?@ux
z)N&=;?X1JPZ1)R)%KENjxfOKflY1!gnNSyJ(Z);=!xCaG&JQ-iK5X0hj4pydzoY$)
zkN$QG0`IJ6$F85W;jnGjt)rN@9l<5_lX6Pj0LN)QEiIxC(Qfn1gvfSNIAb8hdq<Ky
zzV))QST3d{J=b&=bV|2t$vk}W?2tWR`|KgxW+(1HW2SOC)}@qGk1v!yAOG{?e<0-M
z9~?B}=-8N)R*`~_V^YRdQpOue8FQJ@#hEq;@Rg@`%PHG(ol3Tn*O+ql2L!}kMGjxX
z^9G(rcpit~kRQ#fkWSrBLAdtr^;~78pzr?i#w&5JMYYk5TC6*<o|NoOHd^Ll!d^kf
z*<OUz5*J5}zzzOgPNR7f@BXDVQGLO%_+iX$v`Bt4&fq|}jQ65CdkY!s!G%6RtnxyY
zGW-9eZQ&+WYBgO2>hC0!J#c@%{pGFaf8sv4)z+y{Do=hD4gOm{aqr*C%Dk|Dy}`&w
zw>6STs&24zyO4w+Ik~~ghqv+uNVZD;rWxw?xlBcw(HR#J@@&d7He-zam%3g~78xxp
z^H1-f=t)sA4=7?T)YM<QF5_7Svb%N%wV77XlVv`onr8m#mK#*uW|S0{&>es?mo*k}
z_!&E?@HYS+KLzoufz}<?kQ-fSQWR8svRbcU|0nA>MkUeKgd3P;;gF`<H>7p#2V231
znxLARSQ<NJPa2{fRSYz$FgIG#GHp+|nZ8NGH_e&bD|HCwTF=31z%5m?x7>?X6KMnq
z_mUJ^%{4WVsykL$B4YAuo$4;iaNZp^OiLXaLDJIxej|zW?|Oz>%jFz>@3?JbGp!^V
zQ086tb0SJ0rtO4XRobGoe!>nXu0N-sbYhi8@%1x;24YEGV6mr_vTzlETX=V8?A=~;
zA;G!lP#Dz9QVXmOaQ8`QCE+Vv=>*_hdUtOIGg4W)^6|<Yga!-XC5xHTxxaElg90cL
z5L#&41_mEQ2^F-GBp2MWq-ke3uijAiw~5==v{2e6f<B{2=Uq2oM6}C}JEjYcELvhq
z(<pUZC!^DHj^l<GL`!558k&+El0L31r6Hu?AD75kLS?ElXnEH!?QHb+$Z)(rT>qF*
zt(8s~k(odd00xLHwcZ`Z%&%DKR93t@+4!<yg8k%%(dg2FX6wWC4;wq(Q3wFQGYt6=
z2*|qwLFBw7bBc6lSl<OPxEC}{mCh2d2w;O_LSvn{ZmQ%AF(Ms2YtX>DZWu@{S&H0U
z4y+wyWnLsL9+A~bzyU=qi3kP{D@j!i;p&&7!q75P$s!MEI--k08Jeu_tRRxw<QX@h
z%7E9yB;$rmMF^z!a<{1y)Z9LTF{q5H=%9X3t0fJEx|hNm4lz_$g-K<wmf^nlKz+EU
zxCEyjhhhV4<QymzJm=PbeDuf-3tA*75%8x>tAy<b$&|qNP#5lHW^G~K8IK`0XT_yu
zN}ee_M>*XOnw104Of1cDJr>u|gM9SHU;c)C4)rYEFdrmK^M7j0qjE{W>p9hDoZ4A+
z2H0TCY%XwE-;yaxC<`5;=S~1fbm68l+W-<%1cY3>BXcP&StN|lnZN_^?g$ldzi)Da
zD*LiTNgE{Ug=!s_iBKtt-N~J#jKCjUd@MF(xk@2+pO*!W2t#+(ZE6lcYl4K!88A?j
z!o9!}XbKx@JP+6vio{9`WAM2LWYESkl<y@xV3b;%aX*FDc|vo-MS`CPfpjackLv*+
zSQWEKXxDC=%4cDmcD+a?s9hitQz=Jn*u)Y&r=TWGp2u(s$g=1xpmX%(K;`b<nlujB
z5X0K6U33GVVm60-HwfeVDf^CHv9H<l()EMXUKVgO%)Q@Tv#+m9_G);)Lc4o{v9B9F
zX&Jp4LfcNtp0w?=X<nGF4Te&G6}rmLbr;`hIw+>e=j^@G9jrp4a`GVM09J?+jz>wT
z{6;TkV^tFMkRw1SpDNL!tWte(L)dEf*DUs%Q>#t)*6f49wGCs={Y_i-7fdnWr4(~9
zz=y7tDj{%xhyiVB_wTj*d!+Xp|1Z=SerTau@zclvR`I}=p4P;Bt++zQUOh<xiSq!8
QK^7RaCYvxLl+&|c09X*q8vp<R

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.doctree b/.doctrees/autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..2bcb0c96a50ba7cb247dcd5870e9f922e0a63a67
GIT binary patch
literal 25552
zcmd^IYmgk*Rn|)T)@oNPOB<WD>@kpd1pA1R@LI_VBo)U-)~3p35`!gLy)!-A-RkKc
z`XTLNn~K3vY0WK%#O*2qxa>d?pllZbQYj2oQQqMXKcI|(O7eq*KM)`l^GHHffbZPf
z@9DYIGg{fIVykv``u4r|+;i@^-#Pbm-?Q%@d-@BFJ^Y8wc^%!bw^r0<({`Gg7qZEw
z(+Jv{?T1f=#~uqG4eM;mRWEr?;5M|7?Lm!(ZdgrMv%_;Co2BZ$VR<W2`w35L_=aOA
z&Gj*TT%R}>*7Yei<{Q4HWp1&zSnZbOTv`!*v@F%r71j4aAAPJ?mfBX8LTO07p|jBk
zv@IItP<?zL6ugahRB+;l`aGLb64v#7Y&_z4osC+Ct%d8shRUWfG}qV&d`OBUv(IRH
zB!!6v2JWE{-^XmVExt`QgHC#}p*AkKT8rugY{FMJ=~k%E>5KY|eyhGuKgv3*y3^KH
z-N3drclDCvURb^F%-s-A{09T{R>i<p)xdYuj<G5PxXJ~%S`gK0sQ_0EyQyuhxE|QP
z?eyMYL*K*3cv)s<#H8uBgH8Po$omKI`$qiUiQi+yGNk4joY6WP?Wl$uvbDC;3@q(L
zZHP%_KOq_~57Aa1U-v^b(0^E8PNli7--bEU_X{OXweVnEz^3jFY_6-GP^pD%Nu(g?
z@2AA<i2l@`5+U+5nWsc`eF?Ii<gz{Js~dr(x=O>b6tAN-1`zWoug?yieemJqk3RgC
zN8`UCk$X={jy?8T>(1KoA6+|1O6Udcw(4%5f-*tiY$1XaEY6k8?U2o9t`vwQ4+OtI
zjh~j|UQ*p={cD#$TzeqsSlUCr<2Lk_2Q=Gr+@;!kIs^-<l`B5CR9jhDIY#Z3j;qmv
zR+?G^aOS2^qO?`-LfzA>)>19O7hagQMq5j5Xo^@gsAVgc3|~+2$F74)3_oK<7NQk1
z#m4<j*9kgZvrGs$0Ua1n1t!ttlF{__@bomzx{X=C0L5}*Mees$GadX~wgbZB4X3$X
zI?#QtbE)*k7%lwnHsEavqE|OvwWDVmGPc&)UUN3gwe!*J)=mz-#8;xWr_P_PolVu_
zY)%o*rV3KHaSpejBWI8C$*TkJW;&YPH0(`f3z(c8?)vTyH`bO3%kVW<wKAmiXSg9-
zU=tB@VTw5)?TtA=TRvvCx3e9x@y_-#gd1vEaHdIq7FzO~uvtCvij9MF7?hCx<h7cT
z5H14s)m9BV)L&o|xbn9<a2NK`*w}ttYg--5Xc&HYh2O?9x5FK%%ywinhBM=A5*={k
z8?N&~_^P|9`7lMp>llaDkH8$?5RiI$1QYo*miq$xkyP<>dY`I&t~kQQj8;9J)hey0
zW{~<kr61Kl2Mb2je~p^SREZiqiHg69VSccHN|6&8+uuG!7xkAb5|R&p2cx-+(L9kK
zO}+Q<*mKFk?_))V!mmDk3S1-e1f)vr(^+ny3fT1<F-z>1Uf&T5XnQUmj{cnp&Tj9+
zYepgeJFAd+sH*GFS3tP`l4(nVaQdHS0m>X5*|VoQM(Ll@uj-#(*FVGV9|BYQ$%2;7
zjfFL@a;#aD_)3`xd^J0P@+ApGF&7~D2vojV6;%G^+Cjy<6Le*o5(5}$$(a}eSDr{#
zX@CE6C}`>5tw5K=`$-Yx@f%@m1spL?lakODe}R2}f8J7MG|x~&zLPH7t5pIN^WHw1
z(hF1GhjznY%FhgeDHVir!@-LAE-qNi>E8OW%$!kicFyF>k1)ib7Ths0-w`#gu(Zv$
z3;NgSo|wPLyUClXqU#M;C?Y54CIS<ArN`mPV7J%15*wv0Wr#QqOo@Tot)e*Ck<8A!
zW;ynz*t)94fj80me6e-55}Vp;TWekiyJW-xvDx-C_oBuf=ATE~IP`IrM*`9zv#C}R
zX;I@yQob+>D-jRV=qZYbB|#KP6ld-X4?5M}>DUxDOkz1*<;m{Kd@0f!{b(6bWErOM
zAgE)0GHJnwbO<(SC2@Rsc(bQtn~#)@d)Ra$^1<$jV;#q8b;GWS;MWj=vt!q`*U5#>
zGHgcCvpsASQQ_`@6>Xts>5k$V2xR(!n~mx|cn`!C^Cx01y3uTEw&M9JVyA1hd{`%O
zc)7p3)+5YQyQ{YD*xFL9<9G(effPhvyDHxY^%mixb7=$B^OXzQwsKLm0xi>|JH~V_
zZRG@hC-Bn{5@>BWj)h+DRxJ<X2s}+`s0|%GW^Wv;EuR`FdX+@A%7z0w2wTWQ7F^yG
zsi(kk9G%z4*jtCNA;Rb`4A(YYNncOaMcGQKv;9zyW#9xKGCM#V@ZCDU&c!Uu7todY
zB0jL@%zvPM7tOCG-ScI47|U)Co8odP3Ly!i^bD$)|C#hwKG;bbtQhGe(dakGf0qpM
z5FaE3#M&AHKbC=DduGbM_5W^FBx)>RBauv2Dx)M7p_zMxvJku|#i`CzTVUcM)ySLn
z<{gT>sb(W~STuX)JT+M~58#i?_~kWpzi2eU8<{is6War}C;>Ym2J=w1{()Tmcx4t!
zQx2Dia!)tcybZT|u~^4FzJ?BS^(5hT$TA0N;gKC1S&PSkZz=7QQrc{!(=FXKYO-i1
zq!Y7&M<L`#WLCvUEg@RV5Ixg}=%sF=r?W)!(w@oM9zieJ5Ji#n1(vUr@_$}Bl&t5F
zjW~!vMj90z6A{Z?z%oyfhfL`?fzdJ=`8cV5H5>FK2jwFW?^z72sJs<aA@$K>w=90Z
zS6SPCsA#>j@~u6>ZFXddxwPx~>6pzoj}@R;oKC@I=@dwL{$(yC^924Ri?xTMhjJ?K
z#_g^}7HN7!<-Q!1a?Sp2Ih6;?s6?tAQON<xO(miiGS{oC_$Q=e>ZRhQjb`Zv*zC|H
zJr#%p^4F-}A4h+|hFL|Ze<5YSlnFJWl2?#NVx1g$Bm+vX$f?K_WB^|NQBI{lFC`N{
zN?vq^lUJoQDo%Lxq$fM$g3iE5CZv-GrYfttifSC#CpDciaf<Z@_j5imolXZ&^swY|
zChV?6Tkgup_^6aH8*#c#*djGqGzUPFB`Fc9;z!)!*|lNw*IhDJS<GYPRu`4$q?9H(
zaV<5le6jP&Gs$l9f=<VAeZ|B31O6`r4!u^hL+<}tmiN3WWc9JKjatfY83jzHJH`H1
zY9`x3Jj40I)Iz;iDif(L-;bt%xjX=syJAf_6&V0URI02PS58H?Vj?Oz3-yK`qVH*?
z4i+0^7S-d@v1Jt{>x}iaRNW*8#<2H`X{;jPUzai=U9=IEyt0&AtB=X4$dqLOUOpkG
zB2yMYrGmBktdvI8NgqKkIrn5pV<m>2N2n?byozRgMarpEGoqz;D|$#B-EzzH&vGht
z-0pI8BhxehjC@B<MWz`Ml@nQBq$X*6RBGhkUF4|09TG#x&vI7ND%+vsokAqCB;wcK
z7|rH;iOo^5I5kMO%OxU<7%ZY`MT4{~ry|3ch)R`JaY9Z-rb!W%oI(1&-ZRkCCXMkn
z8AEhNI;O0q3}lERIs;pud&}nRp;%2u9?Fbdm5$t$&6%K5g4p?;UaEzTl-AHv%9Gn2
zVTqtS9#)Q=%LwaNqzqILR;BIhhvZab3SK(#<qpBKaw`2Nz7j7l%Bl3{r9{1oJCHw@
z(x|FdyX6pwDae}&sY5UT^!<&LQ>jA`QK{k(d|ggO<`4|P%eUlIWSS9CsqPT`ubkEp
z4uKd#ewMSER@ouAz0X5NVT1A!)A_D8F|?Q5N#XbAF{xW5a{>;dX+<aCO>!zST#2Yu
zaRS~Zry|pyh)T{07^TOzLWtdQQgJht<K0r)SuGgI*b6cPo3D59smVX&lH$z(Qa&%`
zREjqdl`42+%c;olW&mCukyELlF*&>$<F5e3>n^%Vb>s;-wIL8ku(}I~^M)R8EtDb6
zR8lwQG*gmt`#W-OWyta@npQ-XKax|CAxlK13bI_4Q<3ROL}hqn`70^y*BV*=l26S5
z$TEPGzaiyRiYyV8D#-F(ITabQ48Y68xYVvz(3nAx<%pcx5Xd4}9ST{hG-9{35&t(7
zda9r#Z++krFU0Uz{}?@f>hY@241d<*YIwiQ_Og#ze!6?x-uo0xrT4444S&}*|Hq`V
zA9{TbyUl+qhSP<e<=gzz(veGdu7XO*J}kfQPOZVuN@)!(r98QT{iIT1EyZSSWGS=F
zZ%Y}dAgoGT<A>!`WUw*tl<Y`pz0Olo#VS7|Wk6~#N+-TNCi1+Tip;}^s8r$QB{`L$
zCcYRG`I3}I)d}CNm`Ldql&`%nO9#GNYww@rREC;@E^F~^DUqsk(1CT#J*avOti}Eb
zsYla)9SmSCZjsU&T1wZ$T9gNcm!%Bsmd2itQ|Yg<g;xs70q|ZqmHrc7$@0BVPNhFD
zCF)fKKut=cs#;xF04!IhZRx-(s8bc6oL;dxPViLbcH?oV$O%LVJT4mACCa1`vbp%H
zqS_5T_tI0p{1e%i!cHjSq>DF1-WufA!3%{(Gc>=AH}5ki@ra)%NpM+y#5?|}ILw#P
zh@U<}-X6U99I|P8-yaDIkgzD9KY@*lL5BKO{vxA3&L&%dW$_o$1@URccN(;xp#w)(
zdD&&8LExq02*Ff*0OlB-k21eZhXsy`V+<JQ-_a#MR!_+&_%^2@4lrf5Iei36-99ha
zj=PIbG^K3Qd%bn>0y20I`EH8bMwuJTm(UrTfJAum2J_G8lAnagUfa&H$Q881CTu7E
zIvW=+i}WBnY$m!HU(cgP#G;o}nzNm_EUrgA5j{`yeVmmNo5WWlaY1tYC{}cEYq&#L
z=3k>XcD~Qi?E1AI%(@;2#^ndL%)bR|@hcxfed^{{GG)x@$F?|`l7B0Esb<)87RuAr
z4h}x?3s2G78(P!Dd92(^4W43YIAN$qxmL`tWvRc8N0q)xE2ZbLiMsi37^nG7eDpt-
z^!G_ud_6B<7ychvF8gLch+d}bTQA*Je<LATdUna&13%Fm!G~n@(Rrnk(T}5HtTpVO
z-O_+4q&$=DJUzRUf2fs~<EeY`OwY)qZ0e+sJH7|G70*Rv&41~Wy6^Hv`9nR$0|_*R
zc^1lQfSF=(I;Xb`&(;w%-4(X>XBIY}uRKp(x4}@ZEam3-@Z2o#7B>fS$W+2lNd<6J
zC{1~~cX*!OwOc$bl*4NYbF!M?ve)gDU3GYN&kfIR-d8DKoaUS1u4co<B_Q@etlAjF
z`gl&c9rf<Gr2h))<t}MD)NYp)jf6{TK8LU7NAZyd@ZUvA;)R+Yqe?`{{5?@l-!F&e
zAK(wWqn~B(azq#4PQ5}MaYxjA0beT$Fyng09Sq}onx6tm^S7Wo=2iOk41N1_^6fMD
zwqt&le}0aCex84Rfq#CHe}0KRX*k!xMQrrXw6P0b;&n;^a;=Wy4_0A5GVwQxkM~I_
zfDPp1{c+M2_wjBn`gm1R(?o7wm8vNSe?B36&7NKf4>?bdQnBzc_R8{__l}|`btS#N
zzR>$cNUR)s)1h`j?~7<OF!U0o7<%b@487IXN+AqzGtl=B(aANTZ`{G~==(k5iKFlT
z&^M00{M&AzZ))4woQSC;F#sF}i*gFbw`y@H0JkLhV32$z?Y6F;jPpM)^AyO-JO%P{
zvVUFX*<6dIh=9bzhsi+QHNJgfXUU-g36=A2O=Mk~B(Jo-#18weYI_zxdk@>~Y$BhI
z$SXbOJ3exSp5DWz<2qUsmjP5rA7QhNz@@C!N|K{`CN2@Vs@Y@|JwZb6`6NSk%jFTK
zCbEXUmlC0*Pt@7Cc&9L2*XNK&b5le9X@0U17fs68;;ZcrGHmX59mgj(qh)M`Y-AOw
zJ;e~{@%{x=n2s(3*TU5Sijh&?Yr2a#*^}&03mw=v=$9$J!VXZ*a64!#TZoypks)-}
zLvHB~@+G$^6S;@37Db6>X}m-v=8dy%gHqHWO2fU*W?;SOXrZND)F8;cY+6@w3^BSK
zWAoaU0r?>vO+Wz)UTKVtHMNak6Xa$oKPjb$MZcQbM#&T=qM;EpG(NFw(rm;=DMrPn
z0>8EV<I6m(cx(QqqPec)D!OVnEe%Ps$7!gDII_9uJ<hb3G2Tnaju!3|x-d^n0q|x;
zzum2G&EF_AW(SD}Xq~4=20mm5EY;o&)J>$b;j!Mm$nDN<a*;#XZ6cL2rOE{vO7^|Z
z#<56Sh@L_sMl)6g1W0Eo#}qR1#6@f`0<}2*8MHe}qKptMFEEg*$weM#>Tr&ia5bmd
zPm4O97Bn3N8K+R|Ao9mL$hVVi$+e_`v!w>|0U;udzLcinQD$i{5{+^qUL;s>GhoI;
zP*G&NfYwCSl$K$EIb=0opl=|8yjDEJ5E9p*w<e~1Ka$`p@#X0eHqT#Z5uq+<e;b)?
z&m6py&33eoLg{X?J{w$M$TZx6wqxwHDA>)q?{~bl)m0RGE1hj$N1l7f-NaftLN)mS
zP*d@>Hquf1JkgfcS$^Z2-v)Wkx~)c`dn!r0e^cB&maR%EdqHzo49#zG8i0aT-EUhU
zHiaBuSVcUo^dvh9oop+~;!GHba&+vlgWk67t6Pe0Z0Z(%p}(V)Ni1YD{PeC89pAlz
z#|P<YUd3`6Y9!jD4ISBV8kl&o-XLWzLe##1WZ66XUkUL44-mn?4z$N<7<LOf5aGfh
z0tzqKC{#1rq`Q`*ip59?`!9r08H!J01#e>nx`wR4JM2gUwJsoXsT!7oR7!-nn58Rh
zKKi!>&{eA?&}@h8)9j0s_@7M*G{N$PCL-i*U~TxwVNSV*x!%lUP|z$NCJ1?{aOjzE
zoE^%Q(qjQ$jNV$w1!ZBT|000u6I3X^*UdSsadU<~{)7UPAE%EAx*Wwvz^14!50leB
zBH*=mc+DMN3p=7asGK@LGPRS)&OIJwVn|PR7dOc#6N?<1U;JAD4`w^OLaxhW)R%aI
zF&d|6gI7!=(@*hX@z^|n&cZV{nJ;5+X43Z8ByY<kZ!BO3vRcOZK7h2)!U#eZqJ+RC
zt|tVXEuj^_OQDtR4J6giRXtoNOlsqaj4(dD?Nm&aUmeQz54+&{Ay)a3Vqei67-m3S
z!wCq;U<Hwa&w@%|@?mi&-WioYg}M1act4gc9uxrpF#jGjRS--NFIT)jDdx(CSX;vi
Ynxx#Dgiq6y*dC@CgkY=rwteOQ06A93ZU6uP

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.doctree b/.doctrees/autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..90c4b1214e18f3e01873a3b7a276f594dae2242e
GIT binary patch
literal 123188
zcmeHw3z!^LnP^@!$xL1(KmbFqB48#snF%4h0s%r2B<8^@K#*Wl-CZ-?HQimEuIkAQ
zijS+VWGxl25g#b(`e66-?&qrP#l6V-++DAzc!gc>E}|$Z?&|8Q>$<D1_y5nM&Z+9^
zGu5YR9_|8Pda6#HbN<Kq&wu{&pZ}cm>YjTR&6__D|MOS6^-{SuF=Q5tHK%B~Eq_VT
zDKx8A&1=22wedBr+gk&EU&B1=I?YDGYWedZN1;@%6dP8pbxX@%f!V!s#T{bh7rR!$
zD?7EIc&Vq<TUvZeYoOHU_jqNmVo9I;b1K!5igR>`)-h5s-BQl<Jb({%40(;R=hSkM
z<Q7&3wJkbmO(2?jmAt-1sq3wqK#SKZt@8VF0r91UelMf>fWN3xu34=E&AjRN19FXW
zzUcw4kT(m<MHl(7xB#fTkO<E`HM2^emK2+H@sc+SV-0Hr^ECa%o;ix2TBVhxwWVdH
zQ%eg=>-;0bC8uf)H=4DY)fhhNG{%Pa?Y{yT#Qy+-?l47c*lc=^SuYP$77h~@4o7%3
z9M8hxa;<1h3^iP!`?S3an)%W^zlWqrjYHe5^n9SRv>rHqI{Z5m{%wGN8<A#UTcb=U
z9q<>`&2ppVUsQF9&5CuAq0_GZHd=a<PT^AT0k35?OD`&I5)3_1It`i)WTg*U*&J<{
z^^!#3p?jbkoV9zRZZ*m%@@C~Mqg<_5Eb@X9XgDLr6)2QDxJ(-5nsJbb<xu}Y%k`|H
zakT7}49e&2+qVpDhX0G!_?FFEwww(=kd0(F;EPc*UBfhL)=>k6!KxLFf&KgTT)WW#
zi85Tvs=7v{JZ2eX&nP;UTif6nla^<oAQ&S~!zfv1(I_IaE@0>OJ6_3Z7+%S&8Ru=?
z)?f0xy1Q+7coZ;d=7$PSb=av}HM7jbU^riK^25b*&OdMSmdzs<Zr;3k%lYS?Z&~I!
zmU-dU^Up1A9XY=+QrKb@&ILfj^+~Vf)OcB^?nb3+s6Ki6zAc;2@4wO7P^lQrdeQVO
zK&R|R`AB(GvIy;rA)EzHt-uSp#z=X>Dh`n@jv8Z$RjmWl%C6LG{)%v$QGKw+UnaL?
zi}i1ig)3<1qOxFilxGy3U6B1U5STUor=t|ivLt-x)$iy|79<8-CNls#185-&lpZK_
zuH;KS{={ra4{yKOLqN#NOJ!DdmKDDjSlDcoTcyJ&HirVSxemGlG*z~pL7*3ENlPip
zmz36(20^(m;o1<y`cgWgLu226qh&N*>~>b0f(vWFk%m(>u7)AsI9RS!KpB~a>zNI&
zTpKm&PPyg{B~&|A6SQR`P%1EZHOgki9o~H2rV&7;xJiur^+~1x*7*CfL{#b%uH4Y2
zn*zpMm#i0bPYmdkX{lQX$IE4slF07uTz1`NKAA8Dwu&rFhxRU*H*Y75L8F$tt$&lT
zhYOIAhw;fMm`&HL7$yvMHK(?Tzw!ZfsJ(=1RYrs~kdlRl<GLaZ3tiW&*PRB)V<TVo
z8fIe>2GAoF7#WR$n&TOE)AfvMv*MLuKzDH<KYPms8wbfSO7Md5^r+b=lIn3N!k19J
z(1|1CEsSu6=U_z=jI+-hG|t&F2m|n7*C03(jC|R23GliF76pQVj>$2sgJ}uf4#*rN
z*gCZe5DWc>_TPQi?FOp^Iv3amKxUQXsbFpUO|T`fCQcurzPKF;s5T2FV}$fcBR^?0
ztYWi(eY#<dma(H6ZXNnNq+mr?)zKO<D*zV_DbNFz@@<i-DVC(6SWD!_29aGw3z|PL
z0176k9cQH9twJ?`m<MRMSlL-d4cYC|!N0WTIk9?9te|^lCssmJ-kD5OqI%@R=<2v(
z^w0!qKA49==RnPOkXXLZS8K*umSS2l1UVz!+bucGO3}z$#*t<jjOd1wM+FA+g8n@l
zs-QVN<Lu3w&ota7>Gh@;?$?A%e9BNV$Dslftm=YOE0z&EP>RJSbXbQJV+J_P51@}*
zjg8pFDehc#n`HKb(-op@@|IOIV9txv9hepE093Hdo)KfxX&Utk5DZnLL4jQuCMZT|
zW<oR)D4ju)oso*w5q%WFv|oiOC_opcMm3lsQH0!KtVg&AYA`41e?cTIM`vd_gp><-
z@`52_&j#QWqYDQpGW#&9W(^c#n1B5^{VJA6h~ezD4GaTYHxC+Hfj~A23>p_e!Ud3U
z;dwCV;Peig(FIpYMKmU$GEy$W%(EY+aHK_{alsvd1EpnDtP#o}%FD=D#OFwB{5A0s
zC>cBNk>$O#J1}-ugr$@l;>%>&m8$uWTuo;KgHNR1gA=Jkw?Y?*%%!>%f3T;!=>TM=
z^xwV<eCO5oCF%s<dG))Y8vxH{up^KL&9o~B)gBV{WJc|}Sk?Y*@D1%^{k;GNWZv?Q
z5-o^2o+;3AtATzPqTbNV+Iju19R^Dz;{o;j@N8T^yUcPeHk`WCgb5UAUz(F!d;nt9
z0v8~$xR{YfygUQ~%Qa{mg=v|^%LX=H+3gpFU2q<97ZEAUvke(n0bXc>MPR>3Sc`y$
zTPcs0JhV4SMX0E1lunMT$S)Dnk>o&wg0bF27X+xe+Nf6oj}h>Z*@FY+A!}&RxYisU
zwTcJ9q<0#_2gv9{U+91pW-enY489?wnlnjm8tz*QEY0svQ(F^N^xv(u!$i-W_C~yh
zq|xk=<f9ok0yjvSq7QujPBpYr0)CaX>(?9gh64%%JQ_H!sx_<iN$}bm>(}$)Rhfrs
zGJx|HH{W@P_4_8Om2GYvY!q0KsW&XwYK&Vg9C;t*I5PjwW*CK+@)4ge#T>^I{C@Ck
z^92pEuCgYSBoyn@$=QWof~RP9;)2==D9rPZ^UQp+Vm814tK{6eRgj#IQ!O<tBKiH}
z4j+2?*|)!Z%kBIxq4*~Poc}*&d;w+Kd7HOgj7u?Yvswi&#3itz0&5Qb8n(0sYg9wh
z=a#=pddmS%!V6*t`r*G|5vP_rGBysaJZ2ShBh6YN@=^7TW~VZTD{m1%jN^TNuQ%Fo
zn)SBEK<_(9c%{phS^`CmmWy7gwYML{(Ov<nwY1h=3V)h;dnNqgqBm8WOEtZ;ttLRO
z$ICm#$ymG>Hk_leA9`@fs=bWl81=T*C$~9yd)uuY8^O3(V|{9<V|AC@dYE}T=lRQE
zwWbI@m)r#Smn8~(cNWm%va`4X6H>4gBz)|Puvu{#c8z~ZN6Lh<wJ*i|WQ{X!83vXQ
z3BmpX@cjC{^~p`p0xe5kZA1y!j+LynFN8m^7TeMLO`P+fa$Eim>}si&NLysFr07)5
za;;^*1R(p1p=rFyIw;<Sh^4<2S5qeHm2#o%weBFF!IfBR9fL*3Nw%V?xke4DeR3OF
zS#BL5h=AjL)bd*X{!AqhwWxhQGI!k^QJ{P0Lpz^C+PSo*)EkUKBu$p1DJ^{4M7mBR
z6>i2y5JHvkAS;G2-&*@c1in2Ce<C_7SVx&che2iFKw4`bAvJZD*fyX_vh0l%zZX*c
z&T7MNK#8;8h|?vZp(M423)L1VIQs$#RsYD0qd^p!eTjsGjS&D$kDV;(u@L@}8oL*>
zcUEKe9{8S2iM<tas^l<GV)hMKCob7?E%ZK2NUVkI!;sy+1%F-!KL_j_7T=drfrJ?}
zH}%I`fKYTn$@&A+<OWoBG<5#jAIB0@fB3Ahs@<fne2l<X_Qizy<8!2@MEZjQRg#rf
zf7}|^anamNtv0Zd4%#9&%|K&x!w;~zsg1Kz(hZ2Dq!NC}grc-A*0SkBC46s+R#vEl
zpF`2KDq&|zl|cA1H#Nc2gtK8yfN9-nf*$Z>GfnUmE1Xah`~!iXg(f%=aOetX7n(qw
z^$aALPvkmh^%vosu>~=OU`X}RGQEWSu=?^(l%~N(#Q`56nW36dqkj>t10z;dgqNNy
zeG@>I{!=&uO1~j9ae-?ndeh|bjFD6p-H3JdM_eiB$56OZ7Vl&+piF55@4|e&7DcF|
z{SYA?BSY=0f)d1)atWaD(}*?NDL}LIAEjR-8Q>q0s+kQ7|NK}NN$Gc-5dR8vY+S(t
zEo;~8z2R!&^SFY!Jc2JBKhT_1iT8l0@S5oWYxlP$4e0tJWUs~%(i?SalD!(NrZZ#k
z?ApGs1w&=Pw-875hUxgaI*iuU8ATyx?7kPf4IR7J#A7ZE;qM6gSlS`{9wHE7O(*wi
z#Ss2SP&Dlj{t}%byaRhsf6tyzAVy=5Dg0f8!Qqh{(`2P86h2q+JY*fUA=>m0VhL)~
zznK+QjV;Iu<`Z}$fuF@l{smG~XHjVbsw9is^mkDFy4D$$_6{rS+LS6ouBwB^7@I)^
zV)VriQ4FZQI5FrDGTM^5_$SQUaS$Kq!s7qI6f&St7yk-H)2fR-snkV?J9AT8e3$Sy
ztSvB2ZarnUBHCi{!a!Smn-xx|Ext+MXQ3@l0vx&msw69|wm970S<wJXsU@(A4%*@7
zX#^Of4KBhKr`lk7pbZc`NegVpJeg>L3#VvXg%)@*6iurIcBIk*ge!Bi_n%An8t(m=
zCbychn~>gr6PBR1!WLFIVMiP$@U!UsWl~cjTY&;qk`?a#>a1+mPp2oHv&PN(Yee~l
zk+!!(Vy7v-9aBIyo~0?iG@rH0P%Jvp54K~9S-zB{nBwF7S~11%LvbhZv_xyb>Zg<<
zruiRaf)ei&lxc#VL#Fxf4@!_}{*{{3d{ND8O!F@|H!c!FO!E(Y45qg@)6b0@E#}6o
zM$M`WTr+6w9W-tjG%klN5QD~5P1xkH&$R}PgZr-@G=_$THj)x~C;@RUEVc$V>gF8y
z@dil#T~`enyR3?54nl-&10rg3PCinWY?sIt$8yE-0r;(Z4FFag=MS(`p*TJW`w~Xn
zTrE<RBASDcToSqhdngNI17r^rW{u#FsBomqwX&D<oZM|z!x_NMzJio^;t3s4;V$6F
zE?^1l?0^SRp%mtvW8rrLM+OZ*e*9(#GA0b9`)hy^&WKkvCvtAVtXQ{D$Rp@s!Tr^>
z91jB=C^~}%s{`mCG%CpZBBT@vPoN+jAZO02@_DK^Gl&&oHD%&ed3|7lNYblk@kkYb
zy<;36xaKf)0{C|W{KIYle?YF`7dis`LQjBSkgCHQNm~qr7gBGWCd^_Cqklv-MGd2+
z5*3nJ^rJwLrS+veOT-{Ne#oj*VHW)oil#M-F4ZxMWS-2fQ52(#i38;s!c`c!PEVkD
zrFO+~-5E|x76pdW)2whp!|5r4V-|+fDS$&)Ks%c_*qj?-`~qkCKT!O-cCx%vbkq*R
znWjGJP#aY)RFmww3RDOgBjGO{BoD(VsnRPVKqJ=yK-TESXq@dRMO5RQ+zAayOfSXk
znas*AnnH9G>SqrWO{;#i0yH`@Xm0AF^9hSV7a3ETaZ(FoA-RQ>J(kSiZpIRH2Dg<J
zPN);kCJ1Jsh$^I}&Vv~?5OXn8ph~jRDxy|c5lOajiXJW1LKRPO>@YE}LB|xw#p4)#
zr^fc-Tur5d%xOn7M&;azLPAx}aim8{lu3ekH<P9KNnST@;5Sbp912x)9~4ciYIbQ^
zIyfZAjG3D%<_`&LLB(uHIO~%d7|V92Y5o*TP)+j&Ryd)ic^$zs3r+Kvq^3lgh60^d
zO>=*^1IJCJQ>dJ{>ZtA^({rj^VkTF_80t8m60!EA_s0Rr9OoD{^j(x7s)kPKh=?Sv
z-(wP&ULQR@g$OD1(NCagT79Gy+}@c(b5k*WlkgZ6)A}i0Nm3hQsqR$L{GLE1{XHw3
zP$m5x!7vMzv>I^e3h1<|q(Y#QWdEFAkCtnoiU-|%@ot)%rVVb4@;DEhohpxIGLn)k
zU4U6KnJR6aqJ0(m;&LdOR$pAEW50x$GB;JkX2Mob5xvxTEVUhelG{$%VaO!tAeNw}
z$tG4fVV69Mppk_hut-gbOcM%JNmklP&|N`)l*SG{H6T?$wLmAEV;6qdg|6OR=<5al
zK0d;~j>d@O+(nC<8Y3BC8KX0fp<Ga%u~ec+61z7t$w_ZwykQDaP^gT1plDi^q37Hc
z4KT|)<~lidUrV?O8e=f(+*PeCNezq@bf;H7j3uaEc{MAXP_Nuhkj+A`JVa_rq*o}=
zY1J$55_&~iZ{w$BXzd-2S@)xQ<_M?`^rws;z8QXT0TBMs6+n1kOMr*z8sOnu;_Z^r
z8NxS4<2;EHLp9FvF*Hho`4p3_^qS}Irw|r}=6M#1rqw(z)-hJ%cr!O`mcJ$31r0Qi
zbYWU*WGvjB_W2JiLAB4<S>c4*=P`n47TRZNZ{pbv1v;(T=WVgQnXmS18V-^k<26zJ
zQ*=Ph2qgqQU3iBG0{AUrGPr<=^EY<gx;$g3#;BfQXf>vKPLR77r%w{gvoUKX6Qxa4
zh>Ak@Tm(ha>Ygjp>Yji(b5r?jB<v+Vm8qgyq~^s^a`P&?Cz(s_#}ZWQypR=6sC7;!
z7-pe$UPfw4Wb;s<O0vRpDRovB5=difI%kaw3H%IMI1eMO3<>;HOaVCtL=FiQv2XUn
zP%Ih}*f$b{?unqlP~rx9uK~z*+s;3X$5Vs_uG~*<Jur4UHN3jQCHe;Buq|DL2)>gE
zPP{=Q(}?)sPQZJE5+p?MMnGvOyv?oii726~8hTwgVFpee^7l*ODgzsh0XTt;UcjJp
zd&9<-uQ#g_v4E}69S#z~G?WO=kXi(Hl(e54H2_~$;Fp|RhD$X(I3iU#tL6{}h{}{g
zfE~owb&5C)#m(_hiQEy0CuE_8M@r(c8a$4P3}Yh07z7BiV&fvixX3UrGUP>uyvUFj
zATXwphc}804rHJQC&O2A0Co<7_t+DJ@bR=9obm*heiR4!5v_y9&K*0i+@0II`=;D=
zyRX@|`#=sj)0PITx_r-$>v9Kgx_b9EcFn~Q9QZogG)FDCVge7Io~SnljXl@pcJ8?T
zx*hwJxh~&#)y}=SD|heMwQJ9QWzO9<07eP29GJS6B;Y7wYIrn{3LrCdKp>Y9fy9e-
z@Rk?po`tPCqcGPfdWD7B`Fx;6V#D+NrO<;mL7Xi?+Fy<|kn@Mgd+0zWrh4tqVvTF<
zM}mrpcDf0)6P#K?co1bkC~`1xI7w<ip!>H$t?}qCL3AmuOAuCPG#^)>xtgGf2b)(d
zIO(ba=buXrcIYtZkLPp-Io3HwKRl;1(&`M-x5XG1$e|>I{`_mG2AD_tBo1`BXusll
z)?jYYhkl{#L;n0)t8GKZ&uC})XZWKsY`F^f5zYQA$!`A`{&2g*pTC#jZT~zb5ey{w
zED&A3IBbYV?8M(e6`e5cgQ96My&(=$nuwU*AYpo@KBnu_W4Z`K#AA>Z>3}bN5zO7L
zkMf1-QC<i^s4*A^$$&6<7WiJTkMA`Z;0uBGarp9N#223hzEypE%NgJcTLj|p<;jRI
zJ_~$buaEC*(&M{4WM#)-9+LrRh@=W?-mOo~JJM6r$7crl7)l}^feI-_Q1GBW1%H+y
z1tk>)^rK8cNs_29>r?Q>3@NCpD4-u@3aSzX-_xhy>Fg*NQ%Mo~C{X|@LW+K?Pr<)u
zNP$X<#@eI^QUnDjF3~muRzcB}MgZ@cV>N{ov5#^a)FcWv=~J*VI|{~CQp7$=6hMk-
zgWdWR?8uM;l@yJ)NfD$73U1M-;HC^IP)X4^{V2DAEJaOy3XWt)L0%<A?4v{hq=+_n
zlRgE<GNeEyMfo-<f)qi)2lXj<UxpN@Mznm}hz2Qwf=}sF@X72ba8y#nK1xyqDT0D0
z^eOmSh7_oz$Z3-zND&nLSf7F)W=Mfbikvnnf)qhPPoMUfJRgdt9Fu#gcROZ4mcM}f
z@K=HOI(@`XNsssv?s|_wJ^T%j7YPFWbM?_bJ3I8p6zJ1$68dA3*zDIwe@}Y!kK^3u
z3m-9L#ODEW+CPX^vGmC@)04HBU9cBJP3RL45xfiBZ`a3tGCl6A+ZLCoS2OG#?34>A
ziu)vpyGNh6yQYJ<IM-)81vFPIaeJlC^gs29`{VS)osnb-kci|!xH0>Jg5)cZDW!;}
z`+`1opCQ!2V$%s^6$c&TbzJFhk}E7jKJ;xIhxywi&OV`p?a!~b`^UeIMb_E@=wR^~
z?lr<InWWXe`9=WLkbr)F3ZO^Zfj%Px4RiSvIP}JIYB9^Q1bFDU5O(A|+}qN*kZ72H
zSK=v$Dacn8&?^wc6$L%?4wKLoCJON6N)^0!s60|Ggzr%y7a7Rcq6CphYwdppT%)#z
zty(pp?bkVkCBgc}8A(>XNh3r%Mo^H;N1sR${C|GGw&}GNio%skj{{f4a!h;|*rbN?
zZK$Gc<%RVD2IYGpDDg6Xks8Xs#6e*~&!B`>krV=ii^?g4e!mjYPC{=lhGHT_&+fxl
zytjj2+M_^NQ{wjkh^{1_@-daf4<wD?h?Dx|p~779BdJ3Ak0={L*|)2NHHoDERca_I
z>1R-qNdKGEP*l>-pv;i;->*b;4oUyJNE3IlN=*5fO8VDN8I=?#_Pw+emv{au1pRYL
zo+}0YF9{bC3HsO6P*j4>pd=CWZ>gcE1f4;dAwmBaC8Bdk(0@#txC=q2e9S=5RiSR3
z1^qx+&<{v<?}<yb?U_|jG@(6nx*Cc~&>55@f_}Cdib~KKlo=BAok~RKkf2{mnz#!=
zr+iE$=o^S`o;pG_PTZHtWx3FY6NQc+Rkngs#}^1+66yG+8j4Ev8I&ZV|5`N^mFP1l
zGbH-=C=s1QqJI}@;x0s=@-da@zhLT^=Qycf%!+Y=58j2We^SYIrK~?l7?DWU|5^=2
zCF=}I5?Oyj4MipE49X12`VW+d&LLU<E@|Q}WS#Oc3t8{hpkcBdSTw=AkoBd@v`w1D
zP&A=QbBY>@O4b>aB(lCi4MipE49X12`h`kF=a8(QOPaU~S*LvLQr6Yk{Zp#VidW86
z%5a2v5MB<iu+QNbI#9n4q1CZ_IQ-+KTkOZn*3p7fgpFe>`D4zk6-rJ+a(GX1qiM(?
zc5#kwH*<&SmYCtDSFX6j!ER&NN>_Gj!_-eLHtEvjmfr)uwQ{j#e*>@&@0>-NOq39L
zUkcqpA~<C~NCL`Fzoj+cFN9@-mfw%NN#IUztKfw%ANPAHqAhxlE_(J#z5bGsW~D-c
z@F@!YaMpVP`3oBWX-35C58w)M@_}wsSuXEH+epq4c^J9Bj&7R)j6aF5u-%yKprN~n
zV3!r{D@@L2WWXya19)|OqoH5`-i|duUG+c;z1JX0H@|b(hsylLz#DS&x&0=5B|C29
z_nNb*3|q`5z-7O<=J3ZAJb4q4(3pge%h*T$z6x?M65Sxv%9_K|cs{HN;klOQ!4^g;
zyZ?)^3phu1NV02K_4Y>EABWofmy*+O;%*zrV3TCF2m7jbOxh0tIo$bAd**=sDJji=
zwj-$!u@6TR#a^rszX_407Nu>9MMvyY1W8hCR6N30>80(7Q(LF7H8J6KLVf_j=Vapb
zHq)^kYkGx++3->9o{Cciz6zMykHL>>*WE{gN{ITn8H|Z1WC}YUna!x^oIzmvNB|^u
z6Q2DvV)~s<n6iy`F_`{9#`G)ywd9S>n?ie`g556R#oHhm7RMu-U>;X;UOA0=42njV
zhGerSvbOms$sXRjwP4p(gp4q{JQc9TM<FIZ(nt2$^vL#*v+ZN>l|BH<>`h?3Xt}Bz
zDp@=ail$^Sk2g;fA$h7k1t(`nL5c{;t@;#f&X9r>5t94#DY!B{1^r^^jp+c<uh0wR
zB+(U$`b6BCB@uCwBrpzwZRI3E#4GfPn8=cdIAIbP2l2LYk|3g`PsE++iQsXxB#I(5
zpO?RY81Ywu_(${+e;_^LB8Ijd_3$@9UL*+gKd+Der_-a~9(3AHMtmM1r~QK<>q&jG
zzLB1+APlvgn$Ra8B6t_L|3V-4pQgus`aI;zR%qKDOQ2{%ui^95P*h$+1|^Bta6k=3
z<uznbLRv-<Cq+c)0=R`KJgEiIQ4|uwv`vYqY#z(ejI$8fbslNrF1&`6k73`AOiu!H
ze<lOFUZ!jarPcjX!kR?uLlrd?mGm<xNu>WaH58TfGbl48{k{@WSz2aU`tKl3+=cX0
zK6WYn>g;w7=A^Q7JjQ{Mhb$3A^!(df1EuwFkR13{+c?StK{N7ue;UAd^m_|*+x*@i
zhH~TxyaC&VN85v+kdw)tz~oy%&5|J2Qm;LU2?)==E%=15f6%gD0e}4Gb#x=Pxwvlz
zN9`X&8S3J;Uj>f|c_=xxc?CL7ZTl_&X@4DL%)T3+9>=G*22U+`I%eNPe&0rZ?<K!)
zC%^9?zwg9fL}wbD&({)GN;fNIaWY>vS3IS4Hb2}(+SE7kk{fy9&^mFT9oqmkXiOQ3
z$+D={d|8&M!+++U1Np~Fbv|Gxav=XOsDwC>Z;3jP+b=*v6Ab4Y(YBO4fxiwaP2n((
z!7Su3{*&Jy*@<%oFJ|b+IBY_na8|^#3PX0)O6}Rhawyt`*Rqdf4?BhzvwO2+ai>>m
zlTQS9_AYSWsE_-Z>2Y6BjN8x3QvqXq6v*z-M|OJ_$d1SGctMe-5@g3EWN*?(_WJb5
z_R`DKV-S|V0Mh(baQTQn;?=AWFLp$nehCmSN*q6?kNE4;BhFn|aS|Q=2FQy9!T0y+
zqyO&o=m*{&EyvI&^>KeNJ?_)A59Dk5M17eMm1ti4Ej1L?yqG~rGB5tI8j5OO%%IHB
zy!dxYMB8R@JS1l3xBLB?G;tU6V#-JLL>H$q>;TWWv7QRL%7i9*GqSst?r;jCCi$z-
zJf|P0ZNNMqiY9b_o~?$W5_AS7iJ)JqhN2R524#i>{Td~rb4bwllP2y$&?z5N3Hprf
z18FK-LFxNELimzM$G=t$MJ4(SN)pk(OASRO`V7hpiT)oc5uHP#e?MvBE<~U5u}jfc
zXK(X;#>~ZM+4uQda4~fBeF}8je4o!iIUV08!jpWT_{x2stD~Nffi|D#E#UL~J`_pi
z^W+ui`#k><WQ+Pd{|TRn&yzgOw$Jls!os+VwzF`4W_+J?N-yZ4Gx0hSag}7wfuQ_?
zQ_1BkPGL-W8T2kVYkwEq$FU3UE`h5<;Q{=j&WKkvC+HP_@Cw25MJsm%Vr|)@d<BVg
z*@(zM0&{r;kyNlKl`N8bN@8vo37oMz4qml?&(0iN>!G}kBRCF7nJYL6MVTvU?C%VF
zig{3S+G_#kp2vC7DwWZm(BnKWsDyZ&b5W18;;wt<dRF8R6O7dXm=r$f7+hr^^r3#}
z$}X(xc{%7=&^hI6g*x51`do+ouX^_nn+S(uz85p`h;c54K7oTSco#xEpcG=|RN)LL
z+J&3nAlbuies-ToIQ+ZAB_o&VV|sB0n8qEo$F3d$&W3IoIi!#2!3;3%;+B!RKBkrQ
zn2Os(!U7=DfwR$DM()r@`E}`0=35iO2=i1xH@sotz52-Bmj$wMhhHw(#V-0mWaV3Y
z9@0nl6Imb|*TWX<qN#+l@?n*a=_C7S7RbhRvIV<nDnVAh%;#BsWS>co>_RZ)V+MyH
z8PFxqVo;d3T4jYP2Zi6H$9Mj&7;IT0;0jLy)06cvJpqcQ9KIr##k4On#bf{qBB_F!
z&HB^~rl*GQ6gm>Ou^{phsE|?w1y|}*uq!(X(%h|at3CyXv!j4r-4GsTILD{B*I`1R
z0xvrX*tG%i6ijiaz@7RO_~|K-H_Wtm>F8JJ338I?3J>TL@qrA9i0#b*#_X$11SAO}
zKCMs0!x<70+o1y@#BDipk|5$6`b0dQArY~CI>1<TYB@;|@l$;wew3aF9@=pvW@rw7
z15pwQ0{tb&Ynv2{plHg0nq#CL#$xEBe`<R4(}l5IppX03^thjy$Zbo+Dzd`gFG`Zn
zsHCJMAzN4L6S$8Mm}t54Wojs@<xU1A$#Q2!4Mnxw$)L>8a_1|Rh|Zzq&f7>6cd^__
z`KVsMltNnK{94m}nZf&&%vVm(-b2`tNa8=LhN6;q1|^BaKca@Bl6VGXh9v$mC8Bdk
z;vXeV+=awbKBklSBnxVBT3))7fGxgADMH8pLdkrk#Q&7ABay@}I6-Nps3e|2Nh0yb
zsiCMOo<W%*iGP6-(K#gXrvai-6ET5Iq>}iV+{}7~vLTd|UQSq(Ncs<|p{S&vK}jP0
zc{LQ3^fM?kB>kQe(K#gjb<)IL=zGe?RMJ09(dlnj@?0tCEy9IFg8m^j6qTSeC`knU
z6KW_bL1$2ANYMXEiRc^>^v{we?n2NhA2Sg2%%ju)QOR?qpnsomA(5c}N)1IN=nP5{
zLGL|LX*H<?ok5u)K|fiE=o}LC69Cb!>>0|(RDwQZ(dic}TR|!M7ZAQA68)E`p{PWk
zK}jO|htyD1qR*hrkm%b=MCXv`kB}zrLdR1+rV{<>JA>)XO13Ly{SOHv63O~~YA7mM
zXHb&J`k$zwsAQc%nITz!Sc&KylJ$p36L%r&l#f}+dbb8ma4P|G(gg3qMe}VX+m*8Z
zBw<7%S^sA>6qT$qC`n}fH)<#<S!YmYNY+=ZQQ2N|NY?uS(XLDy%EvBcU7fvc*U`Kn
zFe``@pXFUgCjdqrcO40I+jbrO8n)Z$>^eet(yk+X<-3m7cHTnLw*4pv+mC(=D5Tnc
z#4FI>esnzu8Qp$#13r=MN91X?w;vrS*UBE;(U7~%YB*W!IJ%79XE8+XvlxopXE79i
zpT&@jL@VVcBY(AA+W=lYG55BUI7;o9a68FZPzg~IW^_BrGMN#4r8=A_2Fcrl%2I3^
ziNQ$TG~zEJ7lDx2rO;LZDLphiiF8<Fr!NBIw<<AK4(o4$qM-BdIrZsz<iAp<v+E*f
zN4}AkQ)X||<uOc|WnKO%Afux$7wER>@)0PfqstMVq|5P@>+%({jJEF{H9>usAbTqH
z%`4DX-(Nuj$ccGh#V4Y^$<u7B@BPwHn1%MfR7ZPDIHXnHhgL2O?GPPEx3erCYy~|c
zZex-5d5(z;V<LmJ`Ey)k7#A7Hj!?SWG%qsbMTWe*#Z=^IyAvkZdMf3SFfB@`7A0JZ
z60+f~s!}}?zGMq)4fm_Mz>t`&tUN8|CYqf)c3!zVw|DnVx$Aacvv2o-9JEMV8f^ZX
zo!zZ~)AhR#?A)EZddIap_U+rfPq}M$Mo&%rue%raS*DP=J`MW`5*jbh1eFlur4Th<
zmWH$nJ$NycCJcdJ234gnU1IQ&O_xIpKrub1jJU#jP8pfC(WW_S4Z%fiu4hfu8-vDw
z17*Ylr9#GgH+yB5ouu*#DQ)&8P&DeNk=;lUQ+*N19@*6!x0A2KUfDDBF*P#4G;SN;
z{9qR!JPGx8u|B33q{me3i4E_l5$QN}yS!|=63PelQNB7o%6wOB7-60Y=!W;oR`iiA
zWr1v0du3mzkL;_nKsK(2b=)g^pFXm0&jQ)FPS$a+>?ib*{b+h*gH5qvp$(EjY9o7P
zAJxbAZ_?x2&0g7O^fCQ*dQ2mmUfVa;#bf|(M^c44{*690ze-OH-->x8PloESv_w7v
z6;g_z;DnR42kGOWXv#r4Kq>Anw8%%9f|4}84C+%bkRb&r_ABqwr{J>eC>T>o5&I~$
z0i*~iI;>B@p$sWdNzqtvp&pkjND&lx`V`bNq@bqg$zq3TKtWBSz}KhXj_fEHS4k23
zC`l2dh&K3uJ_Ya1kOGwyjkie=qzDQg)~Dd13@K1ak=WwPr3g|41&`}f@K|;f<W*9{
zK1xyqDT0C@=~M7*h7_ozDBmVUkRm8pbc*(vJP(RyG$yAxJK<D)3QkT>L9ll{_PXKd
zSCAt)Ni@S&eIhnzNkp6^35)|#l9L1x`}B#pGD{-j%p-yE6!*jy^@+GOLn31B?0^XN
zRTd{m5-srxeIh2ZBtj`pN7}^+k^~VgeIoA6k_e?ZO?95gNA!t!AVVTz4emgkgvBi<
ziI(`hJ`tbJk_e?Z38PyOA&b+K`b2yqOCsV7CV?^iDtBx!#YIc}LZ67AW=KS=<sA?q
zOm8_!5V7oe+BU}$D4Npd5I8BtNm$=<0YStW`a~F65}_0)VSo!FWO2G!pNI?66T$B~
zZ+DH!U!X7WSHb;*`iNhh9&vuUj`ra|6@Bze*`cp}IMD0#(SLP%^xLmxZx^8WJRnN#
zAB0fdr%%?~)036%%%e}}<Nnd~xUX)D+=*LU>XZwbCGL}Gm`C-A`<v+?E^aZZodTLG
zmN@xf*Jt#J`*wQbru*QY-{@2KD?(kOJu>}kl@^6+j|_v7WRJ|rYAC8bG7L&62>BEm
zv_4d63;Cdq*$d#Z=*TtBYkWl^(O=F|A}R+r%<`?|F9bxR=A1HEgYr?m=|wuKDlS;#
zjA=f&=YW#=O6p!h*pW!$538Z5B%VP@BJs8wib~=clo^uvS11vkLlQqhnm8QUm_Q~{
zN&HM6-1A;#LntY|kFX|@^#7?Eic0z!lqAysuo{X=`WciNlK!tK5uHQQ|9?mmcOm_h
zkEx`8u$%Bn7Ka#Ww^eNwkX&Hc#e5Ni@XwTuqU;I(OxTpDC(Jul+wx&hlJtZXYAC9n
zz@UUUsxV^6R8HUWF_ehTp`LImAljAXL;0AhCrr~3R=bouR|@)NgbRrT{aQ5?mC9#O
zk_h^(YA7l}XHaHH&>Kob=a8V+NE3IV@+luP5cF<OeBd`KqY47$&R>N_f18r$N<qJy
za3PVPe?SdICFl%F5<&l%8j4EL8I&0k^v^31okN2DY0|`92s-6sDnXyIBdnfPwt`ah
zpCNonB>KNpLs5x7gOWt_7p+s8Oe)c5P-aN<*C-L4L!!SL5besCp?pjw`qOuW)ix#D
zm9l;wVMHQXzfuiFCF=}I5?Q}N4MipE49X12dP#}s9Fp}SY2q%_Jmq5+vfiyh6C7E^
zESlh5m^8kU?Mhj{gD@hItiMwYMJ4MDN)lOrKn+DD>kP^a$@*U?5uHP_{^z8LyO4Ft
z$1Y`EoxSbYlb8+FvwZByuV9~e$74?fx^2gvd<)9y9D9QBq+?I;l^=VuT-x2<e(Fg9
zPCa=Va;G}=gjc|K>4%$dz+X_GZ23z@nze#gc4{q4mS_8}Wq|tp)A*?;3-F1YdP1IN
z`_z-`bHTY%wcLo)IBGVES)6|IVy)9p+HpyH3W~ocj=&>>#@L`SK4|1&Gwt>RPR$xL
z3TD$aE8DLyD=xf+?$;(c=HAIyYeAGyuDMu-wN}EzBTfn`A%{m;(ZeH-kKrG82FHFA
zBkjhZ>J*1b#Na0%CgE?@X-nqw4)48>3vHTD6TvqEPlB>bm6R!Mh>M_TbmAcEi^$&l
z^GWu|-ut*Hruj4xF&&dIy-^?2>oUM}oNzk4b#gvUL`=seOdWkp$I@fk7ia<|8zIAJ
z5q{-TPhW&Eyg?uF*Am1_^ZZps%Pr*G@@USfTMg558m+gscqD{Be>;Z_PXh40O2Gbn
zr)A2Lg+T)MV3D;pgWKo#mTQ%A&60sFLUqXD(H8;o0X4|wMQg-tR=nJ}S!r5f@c!{`
zvtD-^Ue2wW4ev<P@hrD>46*u<j8%AWl?h1DHI~U*Jgy~sX$DRoe-#vdR!O0<Yd-=-
zL4Cdr{!G<$kEx-&A3k?#I|k*$GL#-NYH_02o1p0_HMEaMp-tUX6p2(#wYq(Xjcc&T
z=5RKzi=+xt|4mKm?|Nex`!h&{PVsXnny~qLp065tZB_FzD52(4cDdu#urwNvqR?gI
zsk66DSxyStAuuaudCKx4z^LPtMWEX@Wmyg7bfzo_PnxpeE1$BQ7-LhlPhBjK$rGSd
zs;LXFfZK@SW=t}5xeDZu9@uy_K9Q*ld7AC1%MCIY^JZZztGUaSY3D8+nbeb)d@PAY
zXV3_&UU()GmLjASVNz54eV^5omHIB>ti}o|A?jx|I;%M$j<J0FjQ=Rcn4>}EDdsgX
z7>4FG`?73hHn4`~I}3^qNe&z5JQb{Sd=x~#NlCQQE;<H9yO{F4o@9?qdAgXKeNZ3M
z_ho=-7n8G3>0|oI3^46ta`uEireDhd(>O_m=?yam;YrBpkM%MAVR}p#f-a8fRY5Xz
zEAlMx?J<<5q{?8O4@Fb<s|Cflj>S`n5|I5K>-3R5B@1NZ`qqMC;Q7Eu!P#^5kv%&L
zWaGNmf@0wOz(;}Xetl&3q(?S|_ruah+$4x6nlHO*D5?n;gOYs0^%^xSjR_Y;A#8dp
z92;>1V0y>KJJn>V6wTY9Xu_HC18OL$hGI}c4K;;h<ImO5G@6Pck*cYt$FcEMHL0rR
z`%?1e`=%O-s`(g{Q1dAr8~>z+rO|j4g)SRUoxRPm5o5Q{vSVWs%$AOh4S{Z(W8+pR
zr{mZ_c#>lSU%6vrP29Aq-Lo+Yo{c;dOXb<%73h05Hmv8KjRAZjo(=Le+n$Z<uc(-A
zX@}?4uGxFT)!^B%idVqmvN2GAbk``?3=I#(#+LsQjbcObxTJP%tm#xQgdz{XipWZ`
zB4XBRj%Sipb!k0h%gq;`4hMyM+l&o_BO5LmG%gNSP~5s%u(laN7GuLDmq2(ksdtUv
zLrO@N*G_1Ac@@=85_q#WQYgIiYO4Zo5MDE{);hQ-vI9Y^wYLY5h$o{A<M8LeUmYwk
zY}IK~K6BD5EUeaCfLy#M<g0+G{TTd&LL&G&&S$CLJ%++3Bns_I5eN-X?<6z=qv{kG
zJ&hQBrxQkxcf@Fxbv*K*qmCC0Y}4@<Ksg;9kMJZNkFQL}ljcDAd>DDPjtV8n={CsH
zQJ)W#E)4{B;BwQo+-=6T`s6kzZ*RL57CQ^2A+zZ@qYbk-gqjXJ{H;NkUCc^Io%SvU
zjVpg!-aqB9Pgb{m$$Zg~>~~?6RLoB5fRcp3yCVpM6~KVK2dbuVLA~|KO+aTjDhL_4
z9kKx9-T^hKX7S->5A9z6FckF{H>?qFvTn8fg$Tx93h!`<LZw_Nd#z*SGdRX;tz-6K
z0BGNWKQF@{(#AP>@;lOtXr2!_SUZmay2yEz`hp^*OQig7hWu16N<&-*O1*^9oo)*A
z7h5$rC;(Q6VJNUa4szxfrO+fDg%>~A7%q-21*2+tC8r1j<h9w=J;X4M43wN5f#5y_
z<c0Y%V1G(V^Dp6~7rY7Z`;Y1Q_b-QL;9&f7#P~@$(9HnaON&T-;c?c)0ZmCp&?h5>
z!W~FCf+9SX2=eiTji8SOkchIXMb`or2Q5W=fAB6wm7fMx#E)}{MwJ*mLLUEfwHlae
z$*{M4A=E_D!@m6>8NHC0MG)Bg0@_P?6GmD?HM?8@MOFHXx5kffyjW}hCuzJ0z97|c
z`0`}L7oP>br|IMSy!7~{aF?B@kN6gXcp`V%E;SU@G@e08K8?Rt4NGGhPf?iKLRLsr
zdUx5Fnk-e17){<#lWHichGI}c4K;<k><wyY8cju!NYzx+<1Txznp9Qu-Iu)i{!|S`
z)qD&}sQHxcvWL~MG#Zbh&}HMPv$wg+Vg|k0ol5zq#oK~`y%x5&(E>Ihf*(dMK}y>+
z>q{`Gbu-PPw)Z6N4EQcItC(r|oo2XxV3oX~WxoOn`i7d0cCT9vX!tTSWWWY`6;#44
zLTZr9v$Q&lx7&<?yy+G2E>3dM<)CrSpuz9H#49k1<?6<M42|7pICY$}nH9)eGV7L+
zZ#ZVLV7gv8U$G3&3991dd6o3<#vwBP!$3dtfQ41@Gi4W(S8U{2jthMK2`kymm-_sr
z`HE8*%aI6IIkNWsS2zvBnlP*N3hXf0v9El2*{GWCm;nYT+kRjib()o;Q8LFZ#>j!Q
z1<&D;v#UQ+yynz4(a=_P6%n@gNB*O&ASUZ~&pnj6pv*<ogedi%t}>`QiBc0pX#tc}
z+0&?z=3(vB_ku+v>H9M8(*D}@>y3KDDOj#+6^;C)QEOJ~lSQXstY6Q0sLVr&f$E<J
z(A<1@N+46fEd0`=5-JS4J5=*Dq~w)Uw{pG2=O5uwhJ`8i(s_E=lS21boee0`AjK{C
zL|pmgN!OK6=IQIGqEF+vuy`yWZmP=8yU^yjxGuZ8hd9cqT^Fl(O#@_VTXgsys_FDb
z?8Uwz?TSj|&%Z8GC@kzsZv?_qNjabYGH=9-14zUh@y4h(VhQIN@1r6?XqVBTk`(@k
z7)(O`h{deY2qFWK8GF|rhTD}WE8_<z35UZ?5~F`?HXm$~>|uO)SYy~B!c$3{lN>~L
zk3O<@=^z_ccTw8fjOv({n`NW=`5lbWK$tsP)q<8btNL<arKVN=zoBRnt6Fp}ZdR|K
zf|7Q-`i(H`?P3BGYgd;#Il83*kXN8P0Dc__z8fbLkK+>=0LfDp17K(Ilqlc}KuQYO
z>T75@kdEL_2Lya4Ab>pS5^yj28#5TeZyLeH_?!u=@i~|j!QwJ3$LDafw=n=K@l~hT
zgc**;@C)k=`!$LUb;Kt!UK0k$1L2fQ4W}ro#^v)lwDSCG5E6Nm504HBcM1GkPzf1}
z?~IJa_P^tIx+tJvjTn~u&QK$jXqg}eso7QTgbf|_w?MZ|fBy{F*Ny%z_1eTGf@~Mc
zn}Fww&-e*-j<ooEC#c_FU_*hZYpY*g0aqAdjmPF6i;AXa#?3y?gLbFN_k4UJ%9lLN
zw(dF@3}%_e5$(}YTd}C5R#_)@sFtX!&@>~{prNZHJw4GR<4~kfSZ66G83?b7Nyfea
z5>ae-L%*0iK|_@Qx%kOTqM)G|3PV9d2M`EV-5o?ksO{UPKq!&gjzMUa)i$!AquLhe
zwyEtGLpdF_jqoJ3jjvp7<8wr9-wkSe4<vU~+p;%0lbE6RFlb!IXAm>=09Hw*tYZv~
zkh4FDAQ0~V-NX!i2x?OK2*b@DiW&L@6xELzB5mAN%uoy389A@gULYDl%n;^(7h@E7
z(^JF@6<AzHYR_9}?2LvCU6a{Z7_2~*dJlLl6PdC}Vn!ggQ=0LYal(lM2X_k3y_lh=
zSrZ3TC5ajOVWd#F11Sekgr^e0h?nl00VFbj-W!=2h}AgO`-9av(c6C;R1rVSB^p&?
z@Cf<n&ev?>Y<zmhpGoyMZ^TG@s215*LQ$0l<L$wGdP_<6uuqSj%@lU&bvUr;Mf#X-
zNRR1)*zhKvir&XiaMMnGWG_vRY>MEfoAnXDkszKZxT&m$q6)fTP?85-996^82)dvs
zgy)x0hcB8+3c*hw+;pd!ELF$#lQ-0R)KF9n#h`>5YKq{dkEo$(G!;c6RZ~q*aMKsm
zq^g?lGs&Cp>uM;f=3`Jo&8G}*dRh%jqwy#TT{fONd)qop%;-G3b3T7vtUDyIEW`5x
zR5L8LNl?-@9sCB^G2KiDZ8Aq7j=Uu=lxKu~hl#i_ruZ6+SWS{mZ%#$~3e3G=CjUJ%
zw_;s@@?7#=N1|~}IWUoVSW*zy)E@qngVS}k#N<pB0EMt=Oipf01VPPd9EFK{60~n;
zC%inW$uqqBptVk5C84$EgYPy7T6rL|cq3$Ix*(}g(?sH=`swPt2$XvJTt%FR5zJi#
zg*<K4lBfGYp3a1ls_6u(Iw@!hiwF@cO-InwcEBiU&=g071x<ZSHBm)MSkM$#d3^d6
z9=)axnmPz5(q)V5@rf*3kSE<`3*xuQV9B+}!v?0;b(3zlS7>PB2uId4xFd_VkUgdE
zW>!L`YS8?$9#6EIF&-%tnf$A^KqEYrjPuDq3z@P5NW>ZQVAL5Sf;gBrMkEN`@`phs
zDcmtJn1tLh*@jHLSBbJRWa>V`;c%14<1yzjn`OS8WDmPncyLtMzrs`7LZ%+lNA?ps
z$cB|&l(sfgJ*MSm*;E(%5aKmSM_XNx(q^k$z)4M8{V6CaR&P4m>Y{6LV_ob+h^3?5
zT7M7*-%%zlvDSK$cqv{1*Q{Z2n0oB{C6Y_xrTz_{$k<1ovKafq$jGuOjoy6mOG!cJ
zc&SAjH~~HQL<k^Hx&&kuFLh`{#7kuyF11f*t)&w>k)gW#XsMGs*G!@h=FW=yX&@o;
z_#Pe?5(Y~>FQ|l!!haDNh3!-EJ6#Ly%o#Br4+hnx7>{Fcnq93<xX@8^3v}Bw_ldxc
zZZtQKl|rrywN1eDMR)v!`bLVC`U_CKCo#rrt6pA#QgL;RmAVNLCW`l^_(b#;d75p#
zwJ&g;WpdWi4jo+;LLs$+S{|v6Xr<|kiE^1>ak)q|i+FjYP*^o7mvj+c7qbW_fJF4y
z!yx~2M?=3QKrUVzCW?uQq40@b;asziAP}mdBcdVH^1G)%B+;sM3?j3vmXY+1YFVJ$
zrj~6er=yk;o}`xXm8)fZj;Q5_K`mDxdFEoFKEt2|eKr00NngS$sdOcXh5B*?fpEX?
zCKl?8P?O5_7jE`YEY#mYQT<pb(#BoILOlZQjGR|#Ef9?$77Ft}!Wfk<7V2gm5|nlq
zA{iT_F;MB^fw4-ywi6liIPrm)PU%0poD&)Z#^~Pj78)osJ~3(00PXAnT}h&!mPHDM
zyNq(EM0hHp<L)5laGMuEB17e8BBs7ruw(r_Ci=+;s)!%Z5{)A<c!b<-7iu?hHt|pA
zDT!1%=(dm+2{)naDBvx^9CSk@d)PtOP5jdVeN11H0j6>1ws*L&ZbTo`d<K}tokP{(
z!n#-LV|rV9OjCquy-gqSy9wfn!n8i1hN21|VNjBXk9<rGOCx-QqR`oqm!N;957YXR
znk-eE&tE5RsQ;&iqG~7xCDc$;glYXi4Naq|C=#ifYI?%7=4}j*@?mYS93Fm?y!lqB
zp{SaVK?yaVGED1KH7t$BqbPLQc<StJYd<k&#OzK-rPY;?@f)71;3ScSX9)`0W~jf0
z`9U``RN@7o`z&`tYl=y#AMptYE03@AoD}jWag)7$f#)+YZTW35ZJFYrsSgC<SaAL6
zZFuuCIcq9DCJStN9=sLK4WU_W5nR&{m-YLBtnlVstL(3`y(v#S{|eggDpnFq|K|d<
z%%VGJ1cz^B99j1J6rfeqlBX|%JlzT<RbCoYc2a;A77-#iEdg2+fKk!_EshEc(0Woe
zmqtohfEHJRe3~B~x29gCehZ*T*GlfhC$d&To^;nrh(9HR0IkD3-YIo}R;)iM-8~g#
zNaYBob|}qCK-M2abIS|1iI!6SEK(?9ZmI&Z5S~h=`D};zhdvlUBL1OoMg2n}NQ60v
zM1mNbz7SNB!b22;NytN#Z9vvPC{b1#^nXV<9BvYM{N)^G{+_Rs>|uWo54}nlko9wY
zWPhTAY*@KPX=^jeV_I&OjdF4HYFLAGw8{kuZC3evfrFY>`6*{9t#Z+oxLJO3*evIo
zuiY;H77VKIXHo%Ow*3IuB!O4F0^Kp~Y$T7Y|8K!3GNzHIEXK6XGAND8e5FlELFd4$
zO9KKf3kV=jx&&kucy-l?2)r7}<anx`Ix!3pEFuGN_hD8m+bSe+40C7g{w5Fud9V&E
z`GnC`*9Vo5A@!NakZRwE-|6aV8$-mHIvUiHVoZ%eWp*_)A-<y)7U;HV;j4ii-DqJR
zRE5kIju8T$uioP))Fo0-)ia=0uVswaR;#=MrBdn`RMkR+iBi1>pNLW=Pg&?Bp;Y~q
z^t8K81*hi09X)P~a2LmV{>hL&ZGc<XHr~<#c9r@|YfH;$Cn55Eqik&slrHwqp{iz6
z(<@ipO{B|jf+P0;2plWE32p=O;7=qOj^a8UKp=Ehe25emC*;xML$tWKn~)|0@(~Od
zN*FFs%~t(h*J(BiaMkmme}dOAYi@;HQkR1Rmq#0L;&SWu)_~vR!QFM0*4}x3KhI+o
z;kAhzneVSCG#d>-CCBPI;4kATm2%B$9q^a9mgkjgqi)Mz1qU}X;u?hc#g^ZDgIQ_9
zElK=lJipgzdiAE)I#619!0#KiYF5L9tE`}=J~+C}f@S+^9bT3;9LEDxt&#FX%b!0C
zcScbZasYP>GW4_8W}^ac%RQ%%L#qP6x-ENXK15Bd9)UXGZaz!;a)-a%wF(}bGo71&
zW%z3A4u6FUyszg9PPGb@x=`C%nqpNfl9D@G_DVT&Cm&D(z*qsi=#V5VYgl!s;pHmU
zxCOji;P;ozO7Pm_ud*h}z(Tmlsp;}eJ$_Ho$~Q-$z7=@)6|fL;G%T|UDSglu1#&Di
z;#aN#!*W)w;1nTO%kOJ?BbzSR<d)(2)bPig)o3`4T*<5zD^>&8kEj+r5Piyasg@p6
zOUqvu`4XwZ{=d*${S~xcfdemnYWPE>uzwu32?(BB0v1((rk20FV%A2R=BNc6FT(o*
ztJd<5Z&(GVQOp&sx>YLz7`$2QfZq!}bp*%+j+BdB8cMzX3cQIeCzAXHWstSyML0b?
z7sw+pmL!xx4>aJ|f2?pNNg<MstuK=5uGK6$xrr)D({XS&nd4eHxg?R2f|naFyU@>3
zURUOdW!HqO1Aq{Q<zn7XpwBG>V&H0l8oe13q=se7jg%`u94Le}_yj;)tAQE>+6HSY
zLX$6Tm`8)xz4QH5*x5jGYNH^Cm2wqsP1}Fm27g7}s^{=3HRz&nhC1N13{&!Od0h_s
zg@1C%^Xl%l;bBO2hw77Xg;;IKX^cX5IuUac1R!V5v#QWM9yG@iNZ)kk=B>cb6^)TX
zq<Tyd)xVHekB1JT$c|Z!p|a(T5E!5hhD%<x0>Jt_GY`F}=#^S~m-y>ICabxia|Q(D
z$gP-gAKY>7WX&@ta;5TUsRI8%eiv076#Zpx)r5{lt}SqZQOA>a`kI3q4NS(`b%l}%
z_&~$c{svqoEQHcy%{qWs0oZwE5BN{jz%h^>r%<kqfDACbuo@MG+ssF@m8&Sb6}UQr
zbjC$QAA!hJ^Kj&<>rDa%CCkEV?M^H}uCbg|HOrM8+(dy&7h35Ke-+k@E$&*C5vtIR
z`3tSuc<Vqj?=PX+1iCLsV!2i<L*MXnp4k|+Ks$lltOBG!vOLhWP)E@kK~?Ur4yD%2
zDu9AysBI*8=|z27B<YJWkv<RDAHos&kML*T0{Gd3Kld+$pZoBq1p%7&UHJ1J%<?Yy
zY5ILw9`E!c;|QtY7%6>>lsraC9U~=<HB0MZtRVnkguNK9pXjCJ1j9n9CwkxB#lf)6
zM?k-)?84aIx&v*SlV1e3CcRx?5g)g~utCgAPf@h5LMjN_??);SZ5vH{`$7D91b;q^
zhY^1Sf1X(ZKR?8uUmpiQ^H#yn$Qt+=!=G>A&olV5`ZV}C34icvVjHh6w($Z>atEr7
zcdU>*SZq8O+QxIEZ9HMr#uG?wJhaZnL+xxlufWFh3~byDW#eus8>8!OjJ~%qpwq^H
zPa7A`Y+P8gak6gXWZp)D(ME%^=`WZ25NXxraG@d{<)Ni%3aUz;kBX0+g8iGMLZ7SV
zol4LxN#Y5S#2|NHsK!1Hl@wYXsvkzZhSvfec4D+H_5~~tP~-)tiE0{#`GQla(Cz^y
z#p0bT{jDI&uWwxoN~vx(K)r!<yU-62-~h<e3U|^49mbQ}<x1>=Kk51!sCR$>H%>tB
zf+3$UEV$gI<99KI*@eO<omw28P04O0w4yEn+j|F;P#SAJ%ywAlFQOM^_Uyj)+N*$G
z$%;c&W3~38X=Y!FD2Q7%SXc0%3tR?Y_*n(4Q)Np0tz8p;OSi<^FNG2@l;1B=eggCv
z$rOIi3{d{^t|_0$it;Z?lt&hN{syKg{_4P_;az}q#ic<q8BLa!YlTX)h@;mi4kZ0W
Ow7Qsqg3N5@hyFhXcs|<z

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/index.doctree b/.doctrees/autoapi/lmflow/utils/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..7fa53786e09cffec2091fc004621518ae3b63669
GIT binary patch
literal 3744
zcmai1-)|%}63%WilbJ~}N%jXq>;_oDLimvkNcV7p71HUXd$>L3bbxq6El;;+Y~r>X
z+nr1vkhlj<8)RO<`w!h;cAwkbJw0<-5vi4Sb-7$s{_3lmpL#D}j<>2mcW4U6<?@_l
znN%4y-VHL9mN}Ky|L#BdxBt<fx}hdlMwL3H-fcl6Wn5&MO8=dA<DhQ2Fz5CDT|-mL
zm29`O9_zE+@BAqnx}N1$(9S2fFY>uiSLc-tAM~KM^9@~w^EbZ!?4v*Zg5Y%P*|8fY
zt-Ea7^=mJuZbxuQ{j{7BHzL+*J}WJv3(;)z%mmMOQ$dV@5br%n^7uB$%3`&ck@P~-
zdC)1{t|g1`$+JWDi0!k-Y@409zfYLT=|q=OQa!m+`eO2rFaL&+ntyQ6Orm2GQd&g{
zK8Zn3s-P!aW^{S34FY=e`Q38Hwp_20b@CKb$lgK#><M!84xXoY-pBJn2nPAkyb9se
z?G%J-?;hkTD+T@h?X?uU`%!$hCT0Dp^`vAUv9ncdQ}!k@$o3)(R`+<+2;AuFau&^*
zcz3U@iRuf3#lOUCMT_K>aRdiKa=(b`>|<nP02hV;oyrSU%IyD>vW1&bsnxU#tUpK?
zd*t4D^<}K*bLyVlstS*CCFAw^|M;nU_Z4c=SR$?IXnyi4%}c5c3FVVmniqmt+S&E=
z)?FE~B<Ucja^=HYRp&rFOHdC$3(n}6{`6LTA;=|3QYBYz2|nv=d;8XHKm-*Xt=s+1
zOJO-wNmuoDKMG}dl@u;5)rCS{*4@ATgy1#wxLmZ(P`A%zD$0z`xrpV$W-McK#@P2v
zFv$w0nWb)!lVwH=%lz{@pbAoi6LH5o3!!HI#`PJ`GEm@+JE#q|f`Kga8PzoN&$ryD
zVmzayxQ6ZsOuDMnfc9_LNrmzeQ2#H`)mmw|!y4FUSDF+B)t+>#KJ5SJdbOZlX=}ob
z%&I&|!x<aWnn8lC;6qJoP35hOow8>Q9iK(Arme+w5+@BSH8!_9)eL@HkFgt2OW^Ed
z_rPi*O}y!(&}yNni3HxW$`TQiTkBN0F2i|u+&C?DXb(x-4E7sIYzl5*sI@=^y4`Wx
zzUEp1T;Lbi|CWdn*l7D-UzN5ft)H^PsT(dRh@M!bfrcT9kD|Ks0*gJZl!dE+*3!E>
z6Yut-3klA>fWoL=mRewSfV)jXg9%^ZN-qH8+Piymn32lT&c_$-AhcBgD_PE!&ixBF
zHX!^0#hHg@ZD8<Glu$t{Npi(4OPVH!^Xd(C*PFV1O$()MBIqSWI`6s>BcfgQ+%a8p
zWYH2^nntPTdKsOS3miATB0AI-G&ChSBtu+TN<&D&KQ588gvwN9(DH6r+WFa^&kV==
z<ModT)mrI<5t#{q956s^srBwKX1-&kS6T7yWaG<*3HDz%jz(7wG+Q69f7saRjzRzc
zo?*z3KtSFd2qG6HSx}@q!}=}?-`&?VRXR(+5`Ybk3GH|4`k>1>VnjN4)}Vp)-8fKX
zvJ$zw99TQZ%DhNgJR+-=pyD}6L@;<*NvdiHSHBV!hL)L1mU%$a5nX~MY0}+UK_s=w
zB?o<0cs)!qZpci8Kx!{{n>s<w?IRe2%BbiSMh|MWq#;Z9T6n`DhU$utsSMUK-1i=+
z4fhn6;LPJtY=Dhi0EL3*-1;w%-go1I770oO{3z2ZVf#TcBk(=cg?pV@TbR!#6Nt@u
zac!BB=SnY7P7i}-<p4C>28n?b^0TL({0aFS>v_6iK1f#P-)qc=wi)nxLG?MOc3zzU
zHkdG*3mi7IWQG#TLI)VQ696(#+RAJLNK6qBa^sH7wX|fJFuq^{55T)4RKWed$qB0L
zhY}@ikf{6BIxZ8TQWE=+J4qRVKWK2V*pTHah1h*j7C0h|+Eq8HIRLE(5-#V!Kurqw
z0!x7OH`I6@uqzaal^BNO3lGSkO=T$GYk0sYwK(Hp3ayKT=7fs`KMDfrR$d?1V?D4c
z=8@2D+%}b$VaRraNF}IUAQ7`GM{U@|61||HCd{YDa0<w>=p2}B(0v1yyL)TWICet}
zYx8!|4}6M=9rE2EjPIxHCHsy2%6=)`Fi7on0XM^R{PKqVdQ-9|;r%zXvu79{`_Yq@
z(VJ1W?Ud|U+a8<dg&EvnI`J>iRlXnG+1jcb^-j|yF_!+qz9`*67ZsJC2Pp@(LZI+H
zN<#5BdNCngNz_A{fTDb+L<_V^^}&X))gG={95$y`o9?aI2Z?JNCYgtuw(2h!R6tEB
z25jIDjVo0`;0h4~I@9i7Yx(cU@X!7$lo@`3p<wZw$pBgLz?Pxb#Cxr{0>@rGECGy*
SfQwNUNVMjgFea3<^Zx<WbH=0q

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/llava_conversation_lib/index.doctree b/.doctrees/autoapi/lmflow/utils/llava_conversation_lib/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..476792c38338bea0bac5347a809152b832949757
GIT binary patch
literal 64596
zcmd6Q3zQwjnI4knEot-u*<M*_WC0q<(u^d)fDkf5Mz$;=dkubKa@})p&-88G*WT_s
zqXFaKI06aUfZ-}(7Xu;jI(xiim-vv4v*)Z?+j;oO@y6LDi}5-vwsYc*P3(2--H^@Z
z|NpA$>gulBw;z`~(mCp>+g<hle?7kM|GT@Y`mVuqNB{2y^v_%4G^}#%)Ra*w)#@eF
zX?r83da+eCYi|4D_KuIY?`%(cqfKMMskfR%v+Z2~If_=fQfivD_HAu%9J9OSiZjK_
z4?AYjE!S&)acj^TvW9PKPg<kipj&n;X5f;ytx}z>)EB0hj7r6rHwwjiZQg7;28t<E
z$}?>&4y6ZfF;Agc$qJorGwX*M2H5Jht##gL!LOXP!W-gsI^hje$~CinqBUc9V}|QC
z%QGz(+5+poqFi#Yp2Nk8;W%&-(t|al%2FexRzv=nF^VUf<}BuEdBd(ThnL#c8f%la
z%6g-<!rJT^(^kD|PB&Y%n%SIQs5eheAHV4cG=TU620GKsuxX>^){RDanl;fhHPN)b
ziKff7l6h*X=|DX%KDN9yV_o14(z!rmvwpC)K)tN3(4yPm|E2JMJN(~)b%bVWmZ@fw
z-ax}BH{0HutMyW=VqQCuq@TQ<EcfmtnXRD{Zrf;CZ?<;J20CF~4BcR@WFt=woweeG
zH+D#N5RHqC$|>IKAn_1uawgH}SZg!@d*~p`Zta5h8=>}F>l!nyiqVAeSSdISvzVeO
z2PnU{_QM~zWAB}J?7LI^ZKJw(Np)WHH6IZB_g%UF01hRmRjnG$#e*<HppCta9QuH%
zsles7w=VFj0BxWG_D5sz4-s0JFBe-iqcFcGa%Qi821aYuDC`UV27rgpRvT_8N26kt
zYoT)iMGOS<dkWPmz$~0kQ4i|o_sV=FbJl28T!mx&@~&A0#Q^=WF~K>5;RfI!8uf<U
zxn{l9hzttkxnUSGDM!o*6j>;jT&sO-3<pUK2FXbnhjlUXyP-EA;6p(H$htH2(xTST
z70vpB_QD`?L9`6iZ4~&_m}?phD^SoovA?mnzdmE{|A;y^CJrRvf+@Ak!H;~Hntu{V
zY*oXomCCib!YN?Zpu(Gas!(+IaHZ^;O`{UfWIe`dW?Z$hEve?o&Qt?LN|t-e8;fmk
zsIj;k8mVmtc^5nD3Fz%d!T)rmDQ^hs56+?O?M>|#+Tg*la1}7swx05a;g`GE0B^nm
zjqz5(Z}U{6QZAO=_8GdoU^Hv(v(VRz`x&Jtbh_RMlpt5xK0#Vtqd8}~ZEs)tVL)8@
zo8UB;h9i$HgFZP22;685$*w*Z>}s=PcVVACY;Cr_0^S7WJcz~O`WFk_4;k+Rg%5-v
z&*L!R4ZA4aWPK;0qImPqf;D%8HTSC4Om;S}6AA3iKmvOu3H+8Y!0UkdjPe=g|Cx|(
z{Ys1Ty#bv6ZLsrItWH|Ln*iefJFaJZh+AI>0@hx!?1BrDW7hhj^{n;BC#)}JfZX2m
z0dhYJL+)q6R*xUE06F_2XfBSrpC(1!U*$)geJSXT>t2Cy&=p^a48VJ*>I`{&l>&CW
zgbs8^dnP*sIQuB}6yDlx^j48?<yJ_<*~c&o-8$*w_(Y(y-`=HTI^piUP%aDZ_9S+7
zqP{pZme$ePhpCYiUq|PHjUPX35qopkUF7lX>oJ+k>DkvY<vIPHy+C4AUJeIZ;A6SE
z_}XLu;rk*$k+(YOpeW1f0Quh6L{Au;k7EBZaIP2Nly(Im`WWwMYjbvp{$`h+>jcq%
z3FWdt^et&18mRSo0Mv8TR4P#MTsolsJ1oI~`VcRiz<1dXFy(oH`ZW^M9jHuQ&w+Y_
z3Ltp;kOH2NTL;QRxn+g{_!I0G20*yAhvzMzIRLv?c(>(&-CuR-hfc8jTPT+WcDoY8
z4(oazfbG9i1F2xcbLqgga=8z-AMnCS!1fYTo)>JJKznj^JqO!+RA4~J1AI!4_fmi^
zP)Y~BU+=k&Fc{y8UC6+=HsB4Z9sy7u!fZLi)$6+SS0_*&gK}9wc|~GS272Lq0OvK-
zP%1d_T(AS;hckw(w_yq93E#pCCqc{QOnF{#R!K~EPsr58*Cs>AdRzyb`8}zTIlo5-
z%VRy19tPy6u&)>(R|fz&zbAmkhk1YIagh&n>7Y(%d<@EELF3+(XoL!%2Y&KCY9$ql
zcrG0je;-RQD4yYklR)t_Q=S)!&ytw#P-N<Q4#f}X!NT5LWdI~AsDtJEUn5Onh`x+{
z$RN5wfhaRA0MsAy-p&Kmm%4OcC!oFp<+1?vs?<Q`)jkj4`aNnZ6|Q(L9bA{K@ZtIb
zFPsFf&okwD;kpjACs)^VxavJ;nl}+Fnwr<X=+Jx4Y|_K9yb`-j<2eH_I_#@3PtG{@
z@-7|J36O`NToyp4_MGR#+YWo0T1kZ>o(r~q{NTkP_GT=>Jm)T6I0-g(Fy(onSR^ss
zJttEaUz-d;tlo3-IUnggCoh=XbLQ6+hUh)mhZ@hxH3dB9S>D@uJm={y-PZ}I4@0>u
zKuzsAMYYcZ&-rm`D;2JIE*)Hd8%r>_evB7R0@pj3^1N{UHzcMzT$#F_!}UXc_^F$V
zl?<=Ck~+YC@HNvFhU}kXKQhQ(sIxAB+3)f$&x6_LyYyiv%>EF{Wx?##NuphCz0U*C
zeuo-M1udRS2ijj@2?pA4^TJ6$`z@wCFKCAcl7+iWUC%)qn%?J6luQO!y8H9X=ztoU
z+2iNUCO-_$H$&rbaI#5!zWe|zcVniUG4GBpz0?VoS3$WfuuMLyPS*H50OY0AOe!Gp
zT(J4$hcE`cN3jG0<ePZmB!Jw)l;;KH2T4qKFUr)#*M<kZ@tOWH%*TTX!JRXC132rr
z4YU7YI!PIF&xV`ByC^WVd!kaW&9T|dMtGJ)2}*w?TsoM^8$DI6>~|Ult_c}wG)>2B
z&YSco@*hB1F`53uG}T<1tSqlE!~%rS+3jpO0@h*-WbvgF{>1#4wE@G2`63MtcK^3O
zf75=1j}*Q04OS8L!@>ajoL_=W*aTC>-tEsBq?IilCuCY5vTj8&FyTKSrXTZJFTX-J
z=ie#_!#x`(z29u&TyD4CyxCo>m}6saE6>iFxL9Vw5mzQ0yfig7b~vyoXKd`y1Wodt
za4o}~IBA*<XTqpdCXndG%AE<T?11RSDaOX8P|fGD+v@MhOcUM3OnLj?GEmo5p&dI0
z)-d_A?ZsqyA>*_vV=~(WShZ=BxiPg&yqi?>W|JRwli$OU&SKbcRer!te&2^f_US&~
zrQYhyot@m||A2C@nVSsN7;)RDIe8ke;-qo=d1@xrP2#z9Zt_(u!QAA3;f3Rv9qWyg
z<o<*y@8~AIU+c^kY+!AS%_1pzd{C`Y=W&6yf#+W$0;N7#$&F3kgwp#7e<v2)VRgXu
z+t0D4PBGD5&i4({axxYOme6sGobMaVnMHP(U#`TqWBjrzC_BI|SH+HnWb=J52N`kQ
z(}`UULAfmKl4!oKQs47{Tc)XzRBpj@>A2-)EWx;C7cU&w@`77-Fy)=OWmo$6VxyO?
zkmR4)P_?`AkNx|SQj4jpos<va^TFK!4_G?q3*Qjf&k)h5^|L<?Cv`Uf_S2f&!2sF5
z4>H&u<^m#vus;yv@+N|HrqVpQhFV7c{<3N#NQ&XGA7k#Auz$@>v_^RXFy=z=>InA}
zIWPPCGpYI<Zv&q6OVDk=C1BE&8Yy;;HkF0dLk=uwgf+F^b!x;hG08&eaxKYP>m3P~
z)Tuh!%A)H)omeaf>n#Uh6M*PKXIzOnj#mm!8CJEst){Ywf3vD(awjbMb9PPa+<9`L
z<m{mCKnZI|f>Ep0U0T~8SY+w#yY;{}7nUOLpV&^#u>Ih!i336-tiFaD&HWR8mWk~L
z55ip{_>xIjJTgcn0$$`fzpwmy`PMU8LDkD9#DIa693uZUBt(Ucw``#uMveVvSOokz
z{0R>Hu-{jJV}Fws`l_#v#@A&qzF_q$YcbfI0e2D*8!}jf-X%$%$BJqt%PedDif@WG
z#P?RhAlMD}{A8538f~N=px?^UR<))-iK*!*)ihkq@NG!fyOU)@stZ#)ZPFV-)0y2P
zsIi;Le;;G=dNLXBpsOY<8m_?YZm^41D=6?mzy_jUDp+EWO87a~k$f<?P;bDRWVn|O
zW?r^abT6c2-))I)0B>16I%+p#kxlj+;13Nt>QDrFA>17Uh-;yi{g9gXbkSUkYMu&e
zmX0DGFRx>THjugY>*0@}VQ+{r4r@XC;^3v_JN@>;q%`{|itG5;*WL-2BRJ<Rb|YMC
z&DhgHJp)L2eiP9Tsq6jmx~|pf;%&rqWon}&g}gUZVONAQ?6y)f3;y7QpD?D+f_t~u
zq3N6KC`)-80{ckJm(7JD>?70m)`(=ms1+?`=yu!402ix^W#hXFZ27s9WD^wFnFWfL
zOKp1vSQ`lQu>wxG0*u>a<Jivv)+sW9U0-1Ix4khW3s@l!8``KV_Prs#DBgNjI3H`s
z8<}lYDg+s(f2`ovi`bYLkvvQ1?uMoKbXI0V+x{c6%x1>>(8yoGpD-i?E}+Oln*af@
zR-Cl_#o^ws-C)25HTx(0cb;agRh_yK4j*BN09QH8^8^5J4jU!_086g7-q1;u<*%By
zpTOemrf$?%7!dztNPLKqlPsII9|IY2_Z5MzR#r~`E|iPEqdQBN*=`4H<uvsZtey^3
z2E$u;6h!?q3H^gD7U$nFKRl-J^FPF!0Qxh~TfKUG@)a&U*%NJH*Dk0>z%JUdq^fND
z1X0GOty?D=&3e&<osUWrGm8^2mKuw&3&q6Nt%BoX^DycIRo?+4otW({)Ii}+uu9TC
z-jr{J_RAD<+kXUr@@~BO=!v%<zj0zRN$8{ov=jcOP!M)slfL{EQyq%FMe~b}v%d^x
z3SY+IZU3jhId4eF!NAwP09r|8w|{|6#$YINNx$<;ycmFZ2!r-2xNfh8Lrn0-BcY%B
zqG)(u5B|@EQ^Jndp4+{P)V>jI8{mKd-Ai*LbpI1Do6)^}8j|+ia8T=iFPs75z7MaW
zN3lITtQm&GC+U(Gj!*p_x~FYFAToy-{(2bMuV8Uvc>7b3OxSuT=s_$g1&FaD-53T>
zlWZ%BQGeKm-51{r#=Z>d;thj~qwQeq?f8@8;^6OEeK*@l49K!p7s+vPdN<oC=&ZMj
zUlhOBK@RS>96!^VT3wtI`A`z-b4=HTTVob)e2M!&K?8hoaUKG=;g<KmD>wQRtsD-G
zP|%wj`YT9C6!-6dF5|eTEE4qZ#4Ei4{SjXjf&Os+LBC$pq<X|lTO)`+1l7ob_#%|+
zh<Ivs0e8$9g8N|@!H0SKLAzyxyAY5G?)IC}{@d}e3lDeTVYi=}g49`invQ$uxR;Lm
z=y*9DufQW#?6rX9P4B*CNni<j^TU#aL}B?5G?;)TW|6RL<CWgP@<Cq|!Scw`!xF2i
zho!VN0?YfL8d+fZub^CKSfVKcmY6dH%Oe2GPxAK53`-#(2Q2?RR)S#pA278W50qMJ
zSibGu#}6GnksFwaCXR;4oiyi$<{%dn1?P*{?<`j4nIt^_0<ZN3o-g>K2%bm#56=s=
z4XX#Jv^E0JKZS~90qC!wTt|TNmKPAkydj7l1&BV++b}akg@7Co9Ul=8U4totD5aJf
zqDPJ&KXlWf!oJ)f-Jju>mg~+9(_vpw6sVU$dvl=r*(6ja@oI0NdWkQJpnAOjP~8}U
zs%RBGSf#xYux^2>WC80TDAy6Je(MXkV(t)Jj{{t{@^;J&S0Nw=Tt9#nA&>e&Oc7iu
z)eE?)_u0X>;n*7d=3G3#jjaTKX{kPOph$hzufUvN0ZL`90KK~I!434a@Pb`V?@ar;
zI{h@6_?Ok)AD5|<jF3ns?QaF2r}gxu^IgzcK{~RZ531_$lKD)ikosiKei=kW-ahfS
z%&0H!DYW`XmA_^FsIQNR?p*-Xp9alL_Nth3a<a^lus`i9XRM`GcGCVoV2S-I98^#Q
zr<vIN9kQk#%Hz<Fp(+;QsiQzz^SLh8q<AQ*v8KNdMQyCF-c?B3BCcWq3MF<G))0IA
ziJG{dAgy=vC!Kgc<PmV%`y!l1RA@ay@a<K)*;~AC(-o3^{~WhUFtj=nL$vO{%Xtsf
z=Nqy<|2gpFwui4uUIY!PUzG^6_^T2#->Z_t&Y}YghLXjVic|2PIKzvQMC*KmrKgzM
zP)I4O#s+YA-X9Wjq7miaVfQg!5V_>2`1{yVF`4Ey$*aC7;#pXUqk5hdj@#rfWiRkx
z!6=(qRcN$oMYk219JOq}c-sakY47?`)#Iq6*jv_!&bU~MMK;;1;18IT#6xY3RTAr=
zm=^gynZF#`2(Q<n_Oc$+u_}US48GlE?>PLu?CCX)vgJH!q0Q~_Qw-(Va#7vU9$wzd
z-Rg<UfWP2HuV|jmv7XkoYP!6?PQcrs1C!n>NMj>j;g>LU6MF?}Hs%=K2fZk6AXJ|V
zU)mu)@t?<G){w85g=YU74jQem=s3GF5O@m#g4Wd@?DAHOJtnVWE$Hr{hAFyxpj6i2
zo9L?t!S@Z#dc|~dQgWiVSIQ_Nt+^R_kc){D@@Z&gPRKly<b`|iT5oRfKjDiaKDNVr
zJbD~vI^7Di#AlV->!5&mE$emQ-$#6*x_H*!43{IpJ15{aTuVs6pN-cNcZ<NZXb5-!
zs~ce8XJU19Vqp6^C?@CEL>ZW=$wa}tfz)oe)pvup28{n|tnmrDLDN7X5A|oUDl)fE
zb`Pa(9yWGime;b|0?qT|SlbgckIJIK%4+?f25NowgX=X6rgwwV)`%PY7&KcJH#j!d
z(G7Bf6VK2vXUGfM;05pE{gv4Z3IRF1;8v^zdBHbfio77Dmf8#62p`^ajJe#7FwrC~
zWf%#?xqab?loWM`*FbY~cPR5ok9aL!?u|#h$`?f*u@?4-Yse$|ZZVBtJnoB5?HAtx
zMHBkP55;SV^NUQ2hW5jL(Tdd(q{85x$!qBsc>^W!iyw_OK9=|sH_*MYI?nS3dOX(n
zYz-6&9G;3*k-2@c2M*ftqYoVZIM((A%@Ycu^;(|Gt0iT4qT#(hhbI^5{fa(9k>*As
zlzq^0St695Lb(n%>eA<kAO$mrf)w~NrF{kOwah_^5RfBC8C)rXlp#z}kV2`Y4pI)+
zXJ=tDYtDQ1^p8)O+S~!kkdPA%Pc}nCi^r!tmkds};KkkqC!2gx6r41}!O4m;e$bih
zQ$3;nLFpb}T<U;i7ZgnxkQ|KH5*Ls#Et-IYDk_`!c&v)-?gX1N(``C4JS7EddbiIM
ztUf8_BUw+t(%1+AZ-m~+LcmQZ*MWcq_5}T6))4(Rfqq}l$3bTL6#{b5?^&z`-6P(M
zDWYFWEj9ffam+@6wyny^z=@_sDbC#2oSTG41EQil{3P~2oQ7vTfsB$T{skW&qI=ah
zL#f_mJ>iQYDxMEhad-(#a#nuEmz<iFpN66dS^3NHTH;ulY0<E9xX!ZSza6VBGeL(H
zX5wmAos?Xy>Ha=*^~Mee)6=!IH$vCH1$~o+u73&TI?#0he1fqtcZjj)fw4c$2S#Sb
z76JsM)qYI$Zfo@_!Pw)NBF3iFQZx4TKz2E~I?-3G!Xqg+ORoUKc^s-tKe`m!m-91a
zkp!_Fuk<EJ-sX!UX1+bl%=C~TIUQf^i%U(%S3uE(bbKUUOB@|DEgCuwJ#)DwRz+qG
z&i>4W8Lk30XomJRO1It}+`etww(WfIC7H<4jBS?Ki5xZ7zv`{lb^gn^>jH0^?uyqE
z_fC{+Nm7^l<F&+%BBlj*4Pp%_p@~<5(9GXqefE@VBmAcKk#cB7eB{N@23dUM=b>B&
zA4vo#yd&lddB@wqJ5KPX&+HwAfE?cOYgh%kpZPkb$U9PMslDSNa54$kcmxN4vT4DU
zNmh6SOHZNF-2QQZNr~RTyo4RjZeUm*3B#B9SP)&A<KE@_z9{mLcjfeuzwpJS_K^Pz
ziYD}s%U7#3q}D%{r<s=S9&&B0ilA#s^;#>thh&DQB;0PjJ8+_2GgCOosn}MDZTOu~
z6uLKwgFFzgCC))|ElC{YJL9#)IY_1j93*Q$2~50@!#`@etIz&%U6kPT4pN#LagZ+n
zkh3_*8Z=%92gwLfcu33`@{o6dhkTK9Pi7A(1PEGH!ItR$@if+fJmlS&A`eNarS_0-
z!&OP;dM)QYWTLgo${6B$bNk6*Ur^LlJ_c>gU8SE*g7OKx+MD~yM}1M`E$`0hEuZ$q
zrS_IjLD7WX@{93W;)srE>FzDR8LMImyd^VS1#r*|?fqo8-W}ck-rC7peW8}-p7O`B
zO%v-ff5iGYiOc+(cr9@*lWR%hGKa?Fkd&Y<_%L2JqJ${g5QJ#{j_b4Eyim(|de<q3
zS;Tez6KIDlu5%}JOb6HL6Qb~*m^b7-?*{Mr2=9W--ctz3;XMywHOPD3fGP5xlv--<
zdBizcZWL;;`EpMGnc;nz(45<Yj{23*J&*k$v^e*nfs7KOx8n8Qc+vO!qR5MSVJ|u`
zQ?FOJjVk%+=Pp!yd8yr~1w|9O(c9y-#JN$XMdL<UO=TmWiPaHwV5#mpnVL**#F*(R
z`M6sL4<0cpPMZ79&&IY(tmFCTtcTTcq<=8`OuUvjC(5-XaiU+2*Ah35m==66%NkG)
z9J~@lXZ{ZBvjYvi!t_Z`??UAOi@4DDKuctCp+ACh9b9OD4}}xO+#x6GffK!jcSB|;
zDg@+kqJM`KAt(C(Fhx$3QcLbc=S;T%d*xOe?&OXPeiXKlx9TNWHGOUBC$&;@ZZEn@
z%9`ZgZBVwE^~Pid@%krtN@DhPYgEjtPcc)?K-u7nB9HnIjKB54rT5FZsU25{s`o>D
zCv%O|KC{B>l(0I`uzFgM21jUN99k#~EnExbI?zI}S2-_uZ|Gn!w10>$9s;^p%eyTz
zT?hfObTPz$qi%4^mh#?=wIi;$1yjTolv--8c&E{5n6*-YF9gcR7jI3?7pnH$oUvJz
zH;GGqcTMoRX-Q;UEg^^`d|Hx8l@BE`5^|79*%w7*axP3Js}=KkoUVLiVgPf{h0vED
z{PXzioPNj}e8N|&jpF0J{O}K<Xx!%aD!cKoW86W7)URU<^FGnou3Q51{ok?%k7lpI
zKh0i)B%z-D8{iFsAF0svai6Julfsmt;jX85X>^3%OVEB<=>1Djt^>V?u_h>fbBy6p
znm-3LKg-8OW||iQ1h~~M=PxE=X#UT!mPGR}V2Ws-QcF$qN6@90tGHS<AH`=237O{H
z^!~;`Et)2L`KwlFlq=WKJ(`x?DRL0X*(?6wDaj~*%}0*t_#CYMiZ6;-{gE)MuMJzy
zI8fnreV$2`<j15S5&>KpJS8hPx;8HUwc`kX5Q;(tl+PC})8Ev;E?!IA$BUR2Yt$Pm
z*YK-9fltebJ8=HQEOI}-Jyt7z5Ny_fuL%|AjY`W@_aK6=1+^NDdebd9RroN@Z?x*J
z>98IDFGZ=FY-X2A0>NiPT6edMnoJ!)`a%(Bp&2@Tb{5h50FEBLqmTwi9L0|T23Z`%
zyP)|xIEo;2g_DTh2MCr6-v>Mb9r`~xHZwa2AwckiIySmHh&iksIS30=<RB=u)DGe%
zw_ccQ8l`f*P=xQO&(w21I=CVAtw6+p-0osSM5O*f!n#O4!Ka!ll2$(sJs5lxs6sx~
zV}(0?QRFio4f~92j}<gh`%EVnD<NF8nf0t9?T)a@DQKuHtn%B4HvXZU(1u`|J{IJH
zUvc#}nVpK59}Pku1&+C$_hn{|5dyMu%pYN0iDSNqDdHGPEj7nnU#vG4^M#-3=Z>Pz
z+)OgWEGJq}KX!%RHAV$|{wV#B6!APC3!*D?P{en9QA81s_l6=gQu|C1>O?qdKRqi*
zQzNYK9JEaqR(KWhp=Slwt%3;DHM)E-gq|}#4%q)T@1)Gw7Xrk9NgH2mSSPT5A*KlX
zlv;A^m&(PQAw_!Zqt5);XO?TR&lGU%(+`RL-OvF&3J-Spq6qs>_6GYJsY{A|YCk>p
zrKu6@Ujl8D1^d5__~?LrvP6%4ycG5OPXhKY<DHZl`$9lg?3b}d<o9h%5%ww73+$^z
z6uh5B8sr}v5F2dA)}Zsp1{VAZoboF`sjL;~)yD=779kES%$JL;no*eFGcj4NO|Y--
zP9(U^ffuR!5@efFm4tYUsuGwL)P3ja@vuCa&fC8o_CE!E7{`ApcG&-<FN*l@GXODV
ziR%M+ow0DZ){>8(@P#lA$q|cb?CCFa+E`;fRZ80;RCxvpC8kQT$3AM}JV{!`9{YGc
z6ajq(SnysrjlG_k1%-fE4jCvJuAzI+`c1T!IPhDTA`Ya~`9&dn`#~XldrTo9ug?_n
zBR-hKSO`Nl1%>>jFN!GSvtbJP0bbWp$lkz$;$Qeea#M)Lo<36u)>uy=(zXbNd=Cnx
zrV!L5C<N1jLNFbokk0~zd>>Bdq7WfK3>!6e;SXs$Qpnczf<oSeDWVWcEj5K4ActP5
z7*zwmvz9gnN%YO5po-k|u}YQJXY#lfdap<E$5p;4B9Gq*lgBl9osoxHXBmAQ^+n{S
z4vjT^rVgyHo;sv$5$f0ng;G-oY7*3eX+a&B4pGPN0Cijer*lz<5Ri>J=FwV;J{B-V
z)Iq7Grj8o~66cLVW&%m{!G@@!+%&RYC$rCF@*p2rVqD~iL_A*<k;&)6WO6@VXJn$Q
zxulfGeL=Y?MPpN+DFth-rxa;hgi`K-La8YQH3>?=w4fABhbZN9Kq+V8bS_E}0>scs
z7ma)attCqNCZ>o|D7Dm-ay6k6rkJOXM0O>98Uo^T)5nM~wC}esKjcG4^n4D|c*z$<
zr14akG+xH*j5K6@B~|>jUpzNeXe{Y7RbXB9R3UAPP{oT-C^c1}CP5XL7F2=h5LG+{
zRPh&ZIu}(40b;<Uql$?Qf-1IRil~B8OHCE<-WML9Lfiq-N26lEjZ5mAmqc6Yh1KMy
zj<r#VeU3Z!Lr?bTCgv@^D58+x3scC|c%4y5wBC|R-r<YMO(hzO`b;HQXFZij+agr5
z7Ye1O64WH91k-{_Fdd?j-vcVS98Tw=5+NWPm9)@WiaO>oMN~qmrKXbW0*K7=B4%&m
z(1X#{{GQw-vRW~@??K1|d?bm1kRu4W&lg1u@^qL%K8e>EgM=zA>Ekh9MsE7hnA2za
zzzXZ>L)sRhkGr8zYWhG;f<7=U=mXOs`gj`X;|!e6MIS<d7&Yl4k*}k*L?7S46wwEz
zmYO~eD_AU4tFAGZJ{U>#H4KfW+*GnbE40rc$q)G05<?>gq5L;r6cNhj!-Vpec%2c7
zw(62re(p=kO)DCs`b;ZWZ9T0>+ak2`T_}{AR#20m6-*0S!E}gLJ`c3=Jvg0<R)hdC
zcGA(xmJ0=~Okj#=g;GmRD@s(7;oWZPs3iZZUByB7|IO|-(1AU=ues6}MI`b}m_)9^
z>zb$}_+t0%zKq=Tp)se=^nn%D(}%PzLLYmeP-^->O@cl!E$9Q&A^La*=wlz8&P5+W
zKsNet(OQa1T9_jGpwv>+hZ2?S?ORmBHT5<s@rCv|D!HGJEip85L?!q7qKHubAWSG8
zUe`n=qUtg#`Lr)7H?3%l>NBliwe_?jZHv&#X(*JMR#20m6-*0S!E}gL{s3s@9ypzg
zR)hdCcG5*9Uqfq&R=$oYq7_OlHLbi&tbN712H2F=aA{lGwDU$1?L@0nm77>D)JyF%
zv%JiQmlz#6nC1JvC}Nf`gqh_fyv~?KUw26@Kl4T9rWTD=eWn(yx1L(0Z4qjD0Scw2
z7StrD1=E6BFdd?nF95Z?2&Z#Vix40NPdaM3_##0qTQEh`LaC*u7B(T?HLLJC2Nunz
ze}s~JLcFgeH=VGl;Q@JlCX%b64|{aabA>O8h~y8$MDiBA&WI#jYe^wTd?C3hL}O2%
zDFkb*rx0mdghHmFP-+T6O@cx&Ehq%jAqx3Jppd<AIv0fq0of?TL2D^GaWO>{LaAO*
zh-5Qw4O<Vh8)6ODtho+swXqGp33=&lZE<NLaA`YS+Ho81&|r;Oo2*p~7BrSGY0XHn
zlhy%m8+&E5yXBTE&Te`wgIfHEW_P7L14{vfHlBTovQM$MIKk`%%x+jA3swXySRS{-
zl{e(nTg@VDy0y!@&}|wu2N)3ID%@{zu4y_>`_A^HH|W+~qtZThfj1`dm?ikx!Wu90
z#*3|H6I!XjWu5R=i8HkQ$q8=+zS-=SYjaN9TUV+VS!HIM)VQU#H}pQ-&cyNeN%4m2
zEw|Bf+b66wC%n-)vt~98x87{S4_GQ;I<8S|z>k&9dL6fhG-u1F+TOBh_|Psh0a~SY
z5;BbOpRHyEeyw)v#lnnHJc(DGwspU^eij7OV55}4=`-GH$1J+FR<&?yw%)9^&v@gG
z+k}0bi}h-i2Ez$&6FXy8Ogd9oD7#jnSTP(2ssWmK0sq9Rrb%0~RLpr38gjWeW*M-J
z3;#Ljtus%Rq5WLLZ8`GnV9A_m&4Ik}1*2Jm_JbTv)2PCkQRs-GiIqY7$~9=Mf>|rp
zOOUJWjkes`-B<5+%8-7eda+<OoAqYFGHRuY*@U)7Q(;}Hw}!QOK*}IVX?vSPS3)A}
ze+^Ld#+h7cgDCw*^+Kqyw-)OFqt3CQi7HU1wzs-s)aF{Sxf8T_34Sj(Yi(~s(=67T
zr9#QX4gNq)nOYi_ve#yzn$RNUk{AlskT-6$V1omB%3EHBfwsETs=^0s#bAVnr4#Ut
zX0us?y@f%{8ahM69D84$YB*-AR4<&W;!s*^o@&$`vk)vvqofFX^OT)38?0*zrLtqp
zRIno4%3%Q#0Ng65j03PB#DUP&JX>M5T!G2~*KEWTC~?dtq0+3Ov=Vgr%BHaZUB!ML
zTjs4pq`}Ck&A~vdl&fX8ebd_Q-gpB(*bDeD0iQ)`uvK_w#sJ$96yEDC*KIiar>Ehx
zGu2pxtx{`K_2wL4=_1TY20+e&2sGN>2%O)2>7FZ~J;$4~#gKSBBgJ1P#N!Gl9I_|P
z=2Y2qXGsT8!L;R8E1+!DHD&-sCD&>n8}T;7n5-6jaQYP}I1L;jXT7z~V$C&96|C}{
zRe}F7eg`Ub9Qvzh_oD*e{^$(siX?YKDpcx4gE#GF*cS%ufsSW*gI1;{gxn`v4Nx%-
z*12UD+Mm(DSr|R_V!1X8V}SF*dL#;`H51B4TlRo1*nS2r8Xz))Ayb_(OYG~(j%AuI
z=(wl|xlR_$s!^^Kj8X|17kcT8w+<yE4?AXMma*AcZ-rT#Z=Yz*cq5EW0KPC1%e7J&
zu;CV5qX`>|(YRR$M#0E(foMSn+;<|6d+U|6H3K#+#M2;c*8eetWLl#0!+4UVC+*Yd
zZvPP;9vXnd19*6R5Dt&x;ZD48I~-cxD3)>;8zh^RJ4<EGQvS1)7q%4K3XX*Yg1bC`
zdu1{6eIH;AT0<1`ZGnu|R`9R~d><^_oi|u^z)`o)V4(6VP;YG}r#m2=Fa(llsdbQ<
zc@C{cGi_XdYCnsIuj1h+u+5?Ub37axhQkp&+=qt;@$jd3cmWSvM&NJ>9v1NMQ9S$-
z53l0kW212R1RnN{!Qm=AY{W0z;)l*{{6?;g-_W)3OMo_hDbU7G9@+RwBpbirVB;4a
zY}|<2#*M0N+<Vr>y=iUSrqIT14DBD|0r#o0aqk!#_m;77zX%)mldvzrrAfHD+{VR`
zHZGpDagCFWYoKggvS8zq1{-G@+c*>1#)*VBPBgS}9*~Xmglv4JZ{sU}8{fj&_?FJb
z2kJIHaJTUpn2pcKY`kT)@s`%c$kWEiwB@Z1qL#+d>S7tNj{_0L7CVFFEb_4d8$84E
z0sR5Wt26bAPXv_p!q7=S_oyOqRA{9_>lOLnXPa&tSo5N=EPe$79dI4RdJ7!`L>0w)
zrNRIS5zO#me*R%V&*}DifCd{z69|=W0?HKn9z;34(mze$U0JRbE3Fa`cYO}|eyoH}
Lj{5@^W>5Wpsc<KF

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/model/index.doctree b/.doctrees/autoapi/lmflow/utils/model/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..f6769116ceb5be62e811f15b47cf7b63ca0e62b1
GIT binary patch
literal 10686
zcmcgyTWlOx8Mb44?R6Yqa<SStDH}oBbsO6x>I0Z4kQNG38dFePXp=A<&z{{g<Jp;E
zF0rjDf=VRBaMYq^%2fyu4+y~<FGZ>dmp9%JPY9s`0)&7!9zfzEzW<!L?e6T_X;UnD
zcII6E^Pm6x_w#>x?3EuZ4#__;6L{QowwCoq!*Lrdh{Z(1HKI0i!uVFa_(FUwu8XR#
zZw7AU8!Q$>7-4YJZurcJuf$@ShKHsdEGP5F17?J#>txgU7_ad0D{-Ay#aL*DHY;?A
zgLb=VyPL}jMjJTTF^$IH(I?p!VNp8fdaQ7(csei%V?HOUT826w5tRhvH8E<N4vW{K
zbzMyAVd$IdQHY(=&PL2eK)V<>Y&{6ji1#r^Z>zV7M&#um>$<VwvnGv+#CWK;s4M0(
ze1T8#V|;`k7f-Bkx6M}k$Z?pzvg!I8E014%4BJb8fMBqq5L?ls(A7P2MG;^{5@1CV
zV8wJAY-`yMu&w(p?2Fd<kQkF~1<EO+^0#3d`~>LrcKn~j{{#46q%DCez9~0S7o(nT
z`ms0)@R7~lQ!635xL0*9mGtq-S{UmQe~2&TyIte=fu?-FVv_unXVV5Tc|NDMJk$A(
zrsN0#@w}p+f~fc16Cpt4KBtEBQ=ra-q|R)puSd4-YldrUfyazHAk?TFFJ>=){Q1+@
zo<DOf{f%k459iAX{cfM`>gk79&k`F1QM;}C+vjlHAgP#7gc9t(T<DI)T;WT@mODjh
zY7&38+iEd?axxX7vF>3B-&92<Z24~F^-zlBbR2ZNLo!Za%BI-}d3<4#DCdB38#o*{
zWry7uwgBMSA^_4F>uzJa2T|ZT;=7wY?NZSY_m@toLSg7FU-x)nqF7q>wpZPCYxOxr
zz}nefF<4ez=bpPPiCc$MPI=5}m`+REf@UbfJ9Q`U(z6=3&5-%JT>!+to(N7wEgt7W
z^R50u1A&(aiGAL7EGpji61EnzqJAI>e*;JKHK@hD0w*fi7IaoDYP)Hx7*>MDwsq5q
z`S-;*zQSz}7Gs3g7W?tdwmjQ3%rL$wyEk>;iFZU5lvk>!T+@pQ43PRcUXzQ2y5C}9
zEbi@lL{fCW0)_m45!nmFpyMlG^?6>&kLQ)*@vy!mrgON(kMo~Fml2#V(6m8kLK9xZ
zz|UjBm%3ytG1#dt-eu7R{>wcsG#LFIU~^+PNT0zVsDu8H#3<>U80HQeG%>o_4N@E_
zsK1)+IxVH!^lrXxVD@v}vx|d7b*pW!2A)nH&4lN(fcaOM<eE=lZfajHOLd&1+~Gdx
zsm1MTP3$l2ViK?t*C4Z11I%UhvB;6GQ=k;WyePC}jAb1h9vUk0R)6s&9AGc4nQVqc
zV-}qAL>K4ub0he({1SnI9V~!(3*_p|nKDzzwG>Zrf~=dtnX#e4$#PrFVP?2p|NV`Z
zYwNCSFQz7sr85$HV|9Ml-;hB{Et+*2q<oV`EroHt6EXg|&hL7E=@w69)}@CarG~3#
z6l_@Saz?w$i-|$|YKnq*<jt?3nE0D1Nb6(7h2+`Ki@6?za=T_dlW}XAKUnAA+uf>J
zPswBIFBzF(<F6L&mStex-c1+(5x>oUyvBbb-ce$TdiO|z^wwCV66qmpVtqu?0TiFi
z4tudcYU(bkl3dlL$}J5i2z8h#E20_FQdw)s<Uxv-romWQ7Ze_E(q36NB$)ok+PU1|
z-N0&DAKC*v7_LNO`@`$DUDk3L;D41FD(k_rp@l%s8p8mqLZd)Tn5TjswieOcZKSHr
z%#7qVs-tm$i6YxDz(mbllmTSzQP&CKVP}S_r#&}UKP41pcw8pVMXOBvBXe~-n<35B
zJwRzW)grqE5gk!)C4g&=7)hK=ocg-y)Re1P5&<s`^e8(*-~x(C6g(tGkv-`RJ1Iz*
zrXDRYkuV4&za#sRdiA6iWjrB=QZj~e3uyv|kBLJi3MTauO8fEy_*+KKvID4#{lIJq
zK9F2e93Tv&w=KUrVulV@wEjYH7py;J;N@u@#c2(Rs@zA{CagcxP-Owqjpu0oH-l4-
z<^-XsaGW6MyP@~*e&~rqh4A{S$u<ononkSQzG}K-aMv$Te>bM1$R`a-7Bt0RK}lW*
zxh0G=8?mHB!3Kfzt?7MT<WywHm5%ir>_|#9@>U>5EGCI0h^~-u>kRHhB?+kbZ84@o
zCteW~&B(ST+0^=zTId=Y!eNr*+p_Hj!Z_KQ4<p?EK(sur&KF1iyX2_kQBTREHk3lx
zJJAU2Lu$(@IDWi)%VdS>%G{4O^ic2KWfE>$?*<?-4ob*~(^{oZ$t1;Z$6Zo6v>h?-
zxasS3+PtJP20LOZ=}f=pFe2GrkPVt~+_WtnN04Z}fQ2FSPP{`~d_5sms@9S%zA=Sn
z{Tk53nI6ZYzF6B$v+d168LZ^2-(vYrMO?T3SZEWE_o4<yWVVKgJ<x132v=*5<9R93
zX4JfVkw%>NTQcb!Tv@Cwos%BD@=2|~78fX0b%{zT?aTGc6=Jr&cBpRs1M67-#AA1r
z_e=&#io~&K-TF@v%M$l?FOKg<2DVfKS`+BEs(1{*FwuIw7(yu@l#)v6uQo%TiH@Tn
zuj2F_AC)pdse#J?-vyv#3#BC@v#CjCR4|4!U6`hZu%|sg{m%}htr&LVZs*LpPE!`F
z7w{t$-C;>S!p9>AE7H4eImGa4iU{8FrN6w((q8Y<U+h{s^B|Dhrkq8A7Q?k{(OO~A
z@9whbX78fkDK2W=z*j2kVppM+IMjI-KPDD)UW}5s7w~#*AYMgpuo#yq6Og48t@{C`
zBNeEYkx52MTboVO=*%p_5Ejyi(98(L7OYnR2g3WM;$DX@WdyxcK>Qa25x2gL?j8<G
zj!~^rLY$-h)nH-e*9WfLk7U-*F)a^edr3BE@w-Dvg!;<a8@^K5&ff=aXAgwTJtA@s
zte49u-yyr56N&`ubZ9?Cm@Qb(<7ZGzn@_@(S!9PM44cEqk5YP7+Bs$Yw-v-vZLItz
zpCKf0f}%u^QaVeKu0)<>XRch1jtsJD#l05gX;<b@INUKC%mbUfk&b+$!-V^S)k_^~
z26Ri6&)s*v+kjr~eCMcid!p8pn53L<tA<zW1RkAs_#|53Mw4#h73%>SL@QZ!)lTm>
zV(UTth(o=8(jsM_i;ng&kmhNcD(?fVQ+VAYNeEElTBiU8y4HvbRNlGMy|&I_N$V*b
zlJ#DCdz#+fm%W|G+m7{!d_F3l@0ZUH$me77`9XRTI(^*WTrvgOZd^@go;K`La-*Gg
z)l!EWT>~;0NEhT_>-=n5J+LG%*q5a1fQ_cW%jyq$;EyQBQF?p0j_;p<%Bl|dDRDUT
zbtkZ;o1l5V+rl-2syY}8-4M5Y7ly=SI)*jy8G%%MSWFv{Pc;o%R>3fpwx~)5F_FX=
zDA$<F>MxqUjKLbX@OxUfBi#I_Zi1+|QRqcsyvAp6g+{4!Uk?%D;saq73qrl^;bXt=
zx*<?SjJg$z;T7E4DHKS<ZD7D;@)`Lyz7A0MNL3NkixBcfaj*#hjvKN<^G$JpDngv7
zt!*`3zm3bk=>U}p9`5SfRA&+Z)PibZHj^#drWtZAt;xVj$%LdxV@*u?%ya!vv)NVD
zi(J?zCb@2BpJQT<ZJD57h=dE`4mD^@j5XML)WUMpboEHrwt&m@Hd?AU2!qjP2tU&S
zwKV1+GeD7?s77IP>0L`Q=swojsWIPoeU0l*!)7S4s1T}%$Hk1Ic>&6pgc6J6T|HeO
ztXmK?F|FX{IHa?Wb#`=37PGVma9zMbB3oW?9k6w$73nPoic_XzA1ZgmA)gtp-_RP2
zuCTETm5Qv13WThQ+yW?KHd0nV?qHg(-n6_)>_Zv@E_&4Vw2T=+SlM7AIp{k&Zu8c}
zjBJr?PN$zYc>#+Wu7-;^jGD#Oup2O~I8$<o5o%Y>z*NjSqczNcsvfYB1k3SMNFZ=i
z*i3-XTRG<7nq-*PG;LW2Bj)K1i{QGCoOhfJL2Ka1_oKuq`@AqL=A=!x8SWT4xW8?N
z@x|E(#I(mejc&3bpN1qDDgZoiJ8&nV5clvf^n%ru6*LFS-gd}cXW8{zkfkFuQX+tn
zTFBZso>1ODGH>bRJ0Au;r~RhU1)f@R@b{<SDH}p8yTSZrlZ8#W22{Zc58E~ttKw=D
zQY6bUCd6^@WLwLGGuuECPV9);VA~1xEsdKkZsQ;PJ4zR<v6u=_nG$NBj3Ctk2z5w)
zt53~#4Lu?4af8DFGH~!p-autZLd@Pkk;#rM6qD6oi5l1e_qc}XG{FOjUN}gKB8b+z
zhM8^RUE9@_Fj6_!Mhuo|ud@aucpDgShJt_{am2u=4UM4)0Xc^TYSd&=*G(~(6s3Vz
z?WWRZJ7R=6SE(#aOek#v@dYO$F=RqELe$q#nSm?`n*&ndtPmyyT+^T;uxlt*=xC)N
zRW<`!Gy9+d3aT-~@NsHX?{%w1#<W3?Dat!k>G2vpen=0EIxgc8i7L&KX(cLHlk;}u
zoE<rSM~;P!Jwd$$5I4?eQ5jQ75|XJ6=&cDjn_1Y@oTA7qfPIW_lF9xR!BRQD(exf0
zu@Yg*&nYzNURk0k^Wic-DbvByS(1KTBn}k0PSW83DnOJ`1Z`9cDVKDdB6zWdBto`=
zOWW%x@ah<PxT`T6UF{fFO;?qW4tBtYo$$jLQh%fyR?-7Q4b^A35h*rUMN}%-U=c6_
zEb#cFN&79F(DnGE5M#swPzd0M0Gz89%rsy}^Zu+^GT(mFG3=;8%-*6>#K{IJ=|rC?
HM{xOn)A|*x

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/multimodal/index.doctree b/.doctrees/autoapi/lmflow/utils/multimodal/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..a70cf445e19964c4eb3fd55f466fa9157ce0c005
GIT binary patch
literal 12325
zcmc&)Yit}>6?S6p+UvKS#H6HY-J;O7i0u|3C?-k;RU@hpBO#!uxJ<@7clXYCc4nAI
zY_|kjM5`FCS|C#aMF{wVkSG!o2o(uY5E8!;ssIW20sKW&Do^o?@7z1j_3rwSq_yPP
znLFp+d*0`sGv~?5^KZU)K>mrTz~iQ~wWJ${;~Ff8#fafHqBe8F_<FqXRD3nAi<+-*
z25#gxSS$tr(cq?S_{@nPi^U{?ho&7YCG&>^)(B14$)@uPukzu?;ySO1N@#{QE4&hK
zvfEAD-CR;@v?Dt-+peM8F-^gYp%1eyLZvt0)3L&`>gm8KjQNbHX&DiGP*fA#SH+NR
zIxJp|)^#zehoNt-M<Hk?8V5}yAbN%ywjKl+#C^rl+v;w_h`jt^U2kmotVuAD7!LIo
zy^8r1KgGxS?R=1*79UyWZksLpk>fCbdDHbbmOpUm0tiiifMBq!5L?!x(A7P2S+QbS
zvSK-B#j@!bY-`C6K-(P`k4Ec!Kvd*dfsu;G{7%rq?*ixEg8wu4zX$&dL>E}&o06Kk
z81i(}kHwwfbY!#l%=eOkxI;lK_5$G5)iBm0{x-gt6Tiyu0E_vU(w<!W)9HYyU5Fg1
zB!Lq1TyZ3kC)RtfVy0S8?@1eNPrN^`!1-BlX+&~qI@H%ATlckwYiohW8V6w5kc1c0
zYae^!zN=5X{c8Ff({>-sw-e1<BI=d<-nnv)6eEb*ZQb8K57~pOVlFXQSjAG|bu4BI
zPa5bho7weI{6(IjhfHfkLFl%@6Q^mm>gwn0JS5F*P4`>D0>Rp@ZfLfxU(q$sXQ8i~
zjwV6s@_Kf@!C7O&b4@4IJU!%cl5Xf;n9Ne^g|4P_M+>(-Ru4i#z%Uyj;8PQ?bYh$6
za!pjjmhVR1E~$|#8HP*`DAEy3*))xi#}`LQiVmc>0Y!0B#oWt?3JW}1T!0+gx@&BA
zMRd@2H@m*5kV)>I1KMhEWqQlkJzkh7S}Wf6io0&DT<#{w`Exytbg3Kk`O9nbYl-aF
z65+2&>8~wFxzyog#y#d3rqj~4;O&Z-T)L^4q<S!Hn<4Yz`qRJz*YOP*2Ij=^ZbZbn
ztN3<rR{?h?@fSzE?O0U3?M3h~W<~Evviv5L;p>PDJ>86`f(vllu{g6YhaDqs(u6il
zC+07TVLXu$#A1*T5My{|Tb^w;%rL$tUvKKZ6YoI$+bb|1xaK}#PmBOc2CH~g?i9jy
zSs0762hjuR&Ax_!_4;|Si$}oe=V5k`YB%pwkm~bAsj{AuA@N=3r};18_-XMk(zJe}
zqzTUg@C(4=xent`4n1ku+GEpG{FQxf)F1sFVDk*H`Ft6hdT;bZlo?@9U)|d$^qis`
zgLK~T3%QeDlU_b^4lks5ibrk{8p<ntE8U0xH}M3w^+dg(IR7guPT8dE{CoTG1pn?Q
z)yxy{mx?xS4ILR6*uR<cpYfOZ&sX^`#C^Riwl}Gy!Eb=sz9!A~WUn(oK?1K8CD32@
zVY=2FSeIDutNXRyzuma?TK8bzexgo21@`Js#32m$N@h8Gott{w@m@2PIm=_$nml)P
zv0(PP;L%~v)<u#y<#=;qelIF=y|OORG?`;{OAGd8+}6W;2)&nmKZ1FOvG4nP+4r7w
zzcFTRJtP?f$&43I>;)+hKv4?)ZG#Nlq!}_iTklII@5An_2NWhZ7^JPo<uc_!(0KsF
zfgstgKQ)P^_XqkM;yF@_-aulA@|PItJUqDx=w>H7#EC*iH*dR6OXW#=C*lTXU+$b;
z1QNBaw!IQ~Iu&q6kj@3nzry6v?-Q7t=FMv|!^}|@wPu#4>uEuuJeX7$h=x(DlvF>m
z47^VXVCs@Q;LbnjIwA6)tPrqkJUioCKS}h4pYCS*MS=+*R0OOaXEP*9vj`YV*?~e9
zrFtIk$_0~ZfHa7X<15|rkov-W-E=5}#*w{OF=gSPC-YKmr9etjUKtQWsMqbTXi(xd
zNpG~kL}4n7{8GKG-j#qZMSwgDl#5Y;kg8mHZYECla%;-f&-DSwh5FY_MBRj17h|Ae
z5h^P;7ZZewOnT&VDT%XQ#8TGxae?@)KhwIWtUqPTOAR;<4HytLNl_;`-BZl{<)7BS
zvVFRfF+yZ08b)aMz0m(hAM|G>`cxulGbGHmiNat(EVH|XnaNTqMmDuLt*WLe*$pU8
zlL~2_d@2FXiS#iS>9M2b;81qE;)oM>x-e^orkt`S@YBzcl|?+JCJoC;);NAr+ZLlT
zfsyqLNsu+$?=1pdaY9bXS=yoJVZ*uuuX~YF$Gbz4l6&CP>84fOoPKeetPg8c&_-sK
zr~)3L3p=F$6+*q6{)=OU3gQ)$Z8lK&i^WuWtLaXIyM946_Nt0enIkPe%u|sNUNGoY
z%tlom;HD8<|AC4ZPFjTY`#X_V#*{O}dIghYEJDQ|${CAMGAz_n@X?`M(-BpLNz7js
z6;&tURWZ_xY+Guy+J96F-3Bp`BGAin>=_hp<ft;&xvma#fug7N2eQ)B%Cmskzo$o8
zdZ6N&v{D!qg}424la_5NHl)j?@h)cr=|x>-u^>p5(mSb73Qtyu#l$c;Bx_GrM32%1
z6`vh<QGHO@5yK8@y7=CJI@Dx3yr?RbJ7PR}nLf_|A|WxzfTkQb9ZTV8CE71wVTeeo
zRP?tAyTCaasFSdMxDajq2CIn&dS=Y~jnRk5#rp?p>vupbbxg{X>ee3%V@UrGT5DJS
z-1<v#m75Zcze<|d)8W;vzhg)1Rb2Ltym7Xaw2!%H-TG${-#&Gm15|pf?_}F_4ZPL>
z-djg->4R|+Y`b7w#l%#0VonB?Vo^-i#1dhY;!j0_b_qR&7l=B^Qyy@B+LxNX0p}%5
zJ#fJJk(^YHol|+yFQ=~@K)Zr~o?YWPWv?8bvR@ybva-u8rU-c`wJu=&V$AHK2M+@z
z)q`6U4!?^nx`)FYpIzbbIZRUF(0UZN*2i%vE4qde1h&icQ?+ygqMfXDHBR@>#Fl{{
zajLf;+7<5R;IzL)OUQ6%wQ##n{uBaSy1-!qoVATDt*Jp=9;dr$x_ctK`xNeWtgG_+
zX?gvOygn(fPs!_N=}PF_f<Sa(uhPzb<I&-3M>iUl&hAM<;cM<Ki%1KQ>rIMCXR@Va
zL^|0Sk@hY_9-?0Pu58mAj!#|iFU6<PTplX&ltR+3x8;<mV=kpcg`wP!U+POw-w5?E
zrXDy#J+$vhyZm?;e!UKU{M|sfyC2W-+2zOQG3mg5oOVk6INhgy{O&yscb7Xq33q-s
zmbf8zo-S~BcYcj1lJ0zm?xZ`HcQ?bG=Vt*irBcHzbwNQQ>6c1h&8Ha*UiB$*qN12D
zTj%G}F11D3!?l?Aa3MQOcj(#iwri(B<@ltxLfcolj#50#WG!y|tT>Kd9Vf7L>cK<n
zj@v@HSGBlRLN`Qr#l-<Jn!+#xj}c-ZKO!a@kxvbDTGlQ%o{p%tIWdx?Z|D-7$=d#!
zzD#iqv{rmnw<C0zrL`zgb)(RW!g!TWp+%v^&<dxAC@tWD8j;{%+8!RpeAh)25tRkD
zVsT^{tq}?Zs%~rmU^ID*d>c;_l(AC_9=%1@{;Zg7VgbiRXHh<UO-xWDq7${Xt)}a@
z(P1$epaafB-&32~8v`tLN{uj^$q{YS47rvz0Ai<PLedeqD#m^0xjs6a*cApw9u=cp
zx3kBJm|<Hc_=hH^C`f@SqGGW1sD<q&slS8zX|OKS+Zd@q5Di9H6Mm)xZfVRxMTU}O
zQH#Rn;{A)V&T)JBrN(^U^);?LhRx8*S0z+Y0ufV+=Y>@&a+O$|?s(I&g!L<!oS0PW
z<~XFYZ!f>-m@KA=254Qt!6I9JXqd2drxocfG@Qu_(@`{xieo-&xW1tojCzZ(4Yee$
zW<^AB#57V>K)Gg;K7nZYkT{CU5o(iB+tV^-1Y_j@TG5#A==iF#DyHO!RCAJkKI8=~
zGF%NGL;y97{u?)7T5+b-l18XqF#}U6>l8km2GmfOw9H}RaB3tlxN#5@psFD)4(}!s
zFs*5#4n}>Xb99GI@YzLLC5qWt&49?q&{+#nsmF^)#Eh)x*$m%$IJCcw&Y(-v_lQZ4
zc^b6~!9E+3VQ9_rpzXk&j6$5`Vdw=b%gY!Jmb~qdyUvp9w_r;r2vQ;dNG)V-h$oca
zfta^==H7RJpOb#G(Xl*@<jcP`T^?qulge%|f5~KFQ|<v*u*}1@jm>KKL<K97Ep{W~
zG<349WyYBifL2V>kR34{Y&)U8rE#;xZTv%jhv-9BEXD&gJ%-vRBUp6+MjeyS>QS@Z
zhMutYbc3V#*?{1cy@AV;g_ym8#zEAufSqhQPTar_w8w3jP7^wi_=Q<=6hXA!0cN&I
zcPS^5#z?)x8!=R-z0M3+@HQ~u3=LO1;zR>b8yZ8W7dj{nv_{Jg;%j0i>4k=_+D+xn
zcEliau22uM7*XB?<_k?k&CP^uglOHQHZG|*Ge8QO6(X3xmmAb^d^HxcrBNzU%3-Y5
z%pRy!jT)K(K1_q^zHYrhVf8C?`5{$uen6Km)8livM50EpvQ9@m(Gq$`LheW?BujVE
zI4yv0edo}rSWODQsn6ftHgztGt!W@d^KyUy9ABf{YkUZ6sS@f?`UMdObHa?DmuSi2
zpe+2#uQ?XwycXre7M+?bI#oqtqNrt(?g_NjsUQems1b4|DV$<(afB>DjzTNj>u4A*
zfgbM|%%E!}uv)sSTxqtnKH`EO#<21eoogj;5SZcW8g4|63{enYh;66@gpY_j{9rPE
z9kO{k{t(O-WotM9$UlIlstq%bR~jD6hNZHNnNGuw3{vhE`KM8XoO0s7)Ddv${{@X{
A`2YX_

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/position_interpolation/index.doctree b/.doctrees/autoapi/lmflow/utils/position_interpolation/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..7f407ebe4486cff187637fdb0bee1a793d0a63bd
GIT binary patch
literal 3531
zcmb7HTaO#J6<&K+T1k6%;-G-NX@bDbCbm`ykcS{O0t5y6;zn&Q=$k=MGn7ayXE-4_
zUM+$E4h#ogfNx>`M}9(nRiFEjoEb^m7Ih2=o;h=PcrM@h&e`X^Uw%8-ivQe^skkVY
z=QPhtl{4erAXiyku+sXU{Rcn#AN-jcYI<o@tuyA`7Bn(0WUg81pL%x~)UA-_yg9#X
zSZ0MPyW_ma`+WDQKjTB!v%*TY_Q|~`i@8*n=gCH;j0lH`C@s^Kk~EqM$8mb+Teb{t
zz22dhK6&4-D15b^pSU6Eg5%q+-^4U^J5rR)PwN?VBWkS{v)UpTA@*&Nn-J%2CaEzH
z;<;DSB0UZAx>~)=Xm+95Jm}PJ*V0A!<oOZ5$M^YtzRgb^op4pKiLT3%>B*(i7n8qz
z`Bx;M{SOYBNpfsLYpZA_CMiXen4-ylPtin_Ia{7<gE-%Qa=V`KE!T^(wPdF{@IN9h
z{w^~59)1t-dmq0KLX1e07BR9jw^LD}y?a=wyq4_G@4QZ(+=oef{5s0KKee9L{2?E&
z5;Wy+qYQj6A#Vkp<5u8C-_)~oi}4W`Z*7R0s>9+JDdou`{hv_>2SSBDPwM;;@;87B
zL*QH$m8#49KU27c@~X^gwvOfhlU{z$ef+yCx>1y=`^WDwASEp*(W+v^WK^=86so*n
zSEQmg;|VeTCr{nJC^K2-Y%XLhK1Ux63cAc$Wx0QH3*e_!Y(%umLaLd+c6~1L929Wv
z4w}=QU{Kb@jA@qpCtGe5JKm^LUO{&RCRu?jr2boe8dE+(3BE!DZq&&gHCY~CYFbrH
zd%6zTaQ^R`r2}|aXTpul3M;gQ`weLW{$MNk&@^fbgO#x}{<!6xaqPr#+lk{%vSQsu
zuf=V&ZNJuo?KL2xkvB849)3q){E>UpYFe5^m!#5ap_xep*0ah|nM%><#LCp+zFY1v
zt9779(vjhQD@mm228LNH$^~Y_Nk{f`t-#P|4cGrCl{H|~O^d#&ZB<)8<wsLDTrdzF
zwMu)u<P@}OX;I;2Pb+2NDp0rd?$*S+z2pMHy%$gzHLtancsl?RNFdno6|VFGjIO-9
zGl!W{S+@4^l{*MS2rwkexzdGy<qi#)ts>~?Fxm_Z-b)fDS(K1VVL53@9``i|8rGk>
zea$MREs^XQLppD}5vQ_y?YR@S6v(2bwl=L&&-HRPs~5QL@RI5hS%ijWw18xYl4UG}
z6#Nq<az<F0sT^9~4Qo3ee>ye-&-XV!5~j7%gwryYXbHdov1Qh~qm=nIE4^sNyVDzA
zZkXV|y2E62<v@G%{^o}pC*5%f0KhXG`H=|7y8}tfMNJnB>CW+f8#LkG&@59rC*Tyo
z2G@i+Gj;t;m2<?1bcnn~1Mj=TKy+j!a<>Jrc97RaMLIkpt5Fd9ucR^<Jc^Ro4WaZa
zQDs=2E3zyCnvU5L-OJLN(VUDB$t-y$u!BUr9+6xaI+G!g#>?HdPtbGw2*#i@(vAYi
zG-{;fQTJAO!X<|3xYe*SSj$oGJrE)4lwQHGC!yH@n{ojZN>K>wzdU~59agL&Xc4e!
znZ<<d2g!`W_s|#a58T?yd^VXtY|g7I%T;-<^aAa4H)uu&ph+w%P#%kN3?Lsr_~;Yl
z^H9&T8|H&#W&TgCd2}ufc)eiyTrfM2cYqBh+!hj-4K1Cah4L^j2JRGqEJ)XxT>!*X
z0U_7!*j$yCE(sS4F7X3+cccoq-!}zCmwi{Gr417GMto1NM5-{y;tqD2aSDI1C8lOW
zmSYNW`l7CIMOaMZ>}favg9;K+&Vhl36z&9;z=*h^CW?Svsc7oNupC}^Kn6oCNBdsE
z1J0Pm9d|QmT@Y4KAqiIKK)TWElV%A9R^>bq+O^we<+HG4x<R56^e&KyB~73=EU{EC
z7^n%0*a@5hvMlBdc3}+WK;`bvhBU365X0KMd+7&0#l{NxZV;yDGyZ}<=il>X?S?^W
zuPV41*1#9n{QK*gzZ;&PWB5JB-qTN>bd26^VclHikGo;o4lgX523whb6{gC+nwh@S
z&Qq*YOa6K74%Q(Nojk|{fEA*I<4F=aztv0GSeGO{<OmQdW=eJ_i>hzl5O&(b4U5C}
z)_Ca7hJ7%&wqfbHd(&3)1v?FRDZ^q2@L_6YN=noZF<>n1{-u%sjP$<rU!%wH=LfnK
ne=r!pDqYyo(}sAj6Gvq1HG371xCo#a<$*yPvI#pvJv;w5Td{O%

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.doctree b/.doctrees/autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..7a974734f02042e5cd2c6beab4e217ccb0e6fcea
GIT binary patch
literal 19420
zcmd5^du$!qS&w7;`WZXUZqh7?H+NgMx!(BNNwk#8n?&laSa$2i?XIBE-E3#a_l`X?
zo;x$0M{KJs6vU=+_pr3xStLLRifEB2@CQh!NWfBF63<fj6IG?CR0zR82=yQS;rE@#
zJnm!cTwgbdl<S!@=X~dUkMn)s`ObIFr$*oY`sy+MkIja@X}LQ~db8<zO(RU$c++dd
zEyInHw~~dQPF_oDY$DLNLoW^*M#7GvM#Hq6W?;C<D+!yS>XGGyOS1h~Xfz_rbJOPL
zs97<`UP)@^1RIU4$T4!C*wapH!|}G41PtE`Eo!V;Ze#?$=jimd<T!du*8<Nsw6LK&
zMpJ8f?v}Bu`Fhka6M_!lBR^v7(5MFj^YujFUh#Ec8YSi&o6u6yn8U0hDYC{!9LqJ5
zwRl};lX?^d)_NR)l0>;-s~HlN#u|<uhA71IsH?Zc(|9xXvzK+fu@x8_R3~O*k-kZv
z5_8r(XHJ{v&0%w%eQMeCTE=n^yRH!|Z+pSk@`r9d3-YNy7+|<82DYrnk*E9CvLNs>
zC-8E20xy>nc-eBB#?Dd@f((zY9*@_}V{DX{<v1yrY+eA`=40T;d+`5p{J)6*3q%4?
zF|at|H8$ewR*<j{w!CKS7*ACPAO!4r0jxd%SY~A{O7z&gWY#n4ubGd6$L5rn6q!_?
zR0VA88OdcXkY}=YDNKO%0l`TDOYcc7>Y5mTLsT~xLHBV^_mh#n9y@xVH9SWPeWP&?
zEEwVS*~wc!{>dw^ee&vS>Tg2beK_lmeeQnC*~*nCSFVu&hjFW=2fH^QcpyGIBh?iq
z`cm$5!sc>s8c0-BGV_!8_lyTAc0=PuF9O6LYONd1rsZx<PAdH|+TKoz^aQI!n}HYm
z9i%3eHU>uABOk`mWZP;+X0kd-q<2C3EpXRUlC<Ak$^m#O4}cWxy4T$87%0X$9C+Iu
z<y<Gz#O&H+0=iG%40PYjHDo`v;_t3_>-Ng4Jt$hecK9i{)C2a7S8s84)u4i=eZwUV
zYCBMy`4Rr){T-o_r!mKZ5!0RA(9B<vs^MKd@JljP@rB;10t%Gl%#QoJ39I<Kb?_rG
z^7@v<;#VP8zYMd~Q_ZjnC<mL7u#XJdS|soq2zE=i+{FBCHioxo*M|o&Ok-tJcsF)@
z$7)zna+iPJ)&n=$V-pZBp~?qi$=Nsna{HUC@j+qwZyHg;UOGB4<r?)DVX+@bO}aV+
z!FU@A=Zsm&<n8Ueycs=Zh6L*^bKd-Om?VPpS!z}$WYpkIRD1&+{(PH;lOCorru%ey
z&iv;Cc2qw6J&fkgevp0^m7tV+Lt-PO`4jj<K+r<%jdqaibj~VO9nakq7DjKMs7<tf
zwY_z|5}Vj*IV+*BlM6oX2S#WFw+$|3A4gl|CfwpyG8=N{ck0R?%NI%4jK)6+DALPJ
zosAMtXX4q|<z9X$l-(JYD@jGlJ~edgSYE1nYa$v093Fc%ZNZ0e5;F6VHks+=VVW<Q
zFB2FzGC430zzn>4wVN+vY6>s*K5|b2!*{@5O}OIQR%F7(6CO*=x3qv(LW@;5+7gco
z%5O0g80j?F#rIlYUWIZ?>?`2MAw3fX<Mrae%+D0xYrT~N`}NeImr!XSZbu1xxE|G8
zV8t!LfLojy>SaNCCBLMQVs{>lF^l>6Yid~bkI@99_3{~Zt_0c453xU%PRVYomYVtH
z{k;(TN4XgFp8lyLVtys>H`oodJGkp${)zdH`KN2<SJ}r0VA{*5Lwe@HFz;QI3z=p6
z1x|5r`E*z9k>fAb%q!{4d?=ft*6@TiOurpd3vT`GHQBh#cu_Mr0QRaF(jk2S`wj=w
zTb7F5m16@>z<%xk@ZbzcHP#zmyFXwrbp!l=-HgBeM7OHB=!rdw3U-BRftWD)ISXem
zpt#-ePi&+HIy3!+!V=?8R=33X&)(!wn<|zt5!Q~W3(!%XIRbx1L*y~6ZaO6I=@5S8
z2)vWF;Eupq3~5OzvHh0Eg-}x^#1awV0A}=9a#d@5aZt6~0feCzj&@m52u$T@Bx6&@
z*a%{{ogpDwEzi&wEwm6;jN+gWgEf0lrq;Ati`6d9$6~dm?`UZ((NM9J@cPxuPriS#
zx~_-D^(&Y0Z?W1mZd(oG`kLn&%19n#vwSeR>mnA@7A$60U%vL92mzjz>P60&i#Hal
z*VNz<@`97JQcbH=FW$HTYX*x}gR&bH>%2_x{wo!ultd{Z1gzZZoDEf#JBz0b$amX-
z6m<_9+D_Uhwf(=;<eW|2m#rZR^V<GT(davAcU<(p52BA?hoBh3^b{3Vp?PzZJu?7P
zr?4Iftc-3aj+f2)8k>SN*0Gr3D6tcSA?F`|FGMqnbebZ#=j^U@n&5P5CsNOF%Ynl&
z_Jw_r87E`}CgU`Q{g9Y1LE;ogf}(FN10f{GLgYQi@(&MqUdrjCbVS;3zRzLwDa7QM
z;1sD6t)0$Utuu^2XTos3lzwc;?VLnU2@?Gv{*?I_MdA1X(dY_qWWNu8l-a<-S+1iv
zBKEc3-=Ipe${7wHCD=X9nEf0+cW45M#EbYg5F{qjC14jU+CBn}TuFw9+dm%mpzt_f
zW<rES7CH2;9u0(4o?$(|XL#S~AMc;xcta7-@wrLDkL^gT%Do^zO@^+k;ovFj_!mk$
zW<y@5@F&#doGtss9)#ucea4&~7$Iw``V%i%m%mxwCA95^9y^hCTX$lk*gbnFjD6n=
zA}z$`OZ17@qy5Z1xMubj^P}qR3m<q&PG+t`H=d8s=M+Q!l41DOQVdIH;~%4Gyl%gZ
zKdE-;!Md-N_wogN?wIqEce0l+<$D?BZjnMszGXOlqrA7@YwvA;R*69jfK^jDPp%uU
zD&o6BO}%EgeYYI9GVS?4XnH_?jSQFS=rVrEUJCpgA;(qWBwjL%PL=o72e-r!`r_6>
z8nH)OnDdrKngq!c(!anf91K3cmx)K+Tq_8g$P}Ti*V~q{-9WlX!e-T@rn?QZpbPg~
zrP2P7+KO34LI-66gagP{K!yb%Tg}9N49MgnOQgS+iDP;E#Ihixo&61X-+al8><wu7
zgiVrThdnE#wG_A5SVeAAneV8Dhgo6c8?obXzew~ysYPCcXiW>{cX-(yY>x5LtQw8&
z+vNAp3!fij{FYn><lKJ*glWZ{GdP0CH8b5vVb!OhGQ*W>8h(U@YM4u|yCjzFuZp<~
z*I-Q6%-N|I+63Cn;+c<>12nS`<I4m@gcxVrI?r|WN?TXOMb*K=iXQ`vGN&rX{6bi9
z6Ya{wihH>oNfB1$oI_}_10{-BAQs5^b{>2q8kVj3s6K6=QBDWdUz3FBH^O263j~Gv
zsasgB4PZssBO*{u^P3RpXdYcZODQ%<G323$WzLI<YyT^dS0-Fe2K(Q0WqAjfM+xjl
zp%qjM`}0J6A*)?I<wP<^_17k)_7ebRU&2qxa&;b`!RODWpg8L<z@A-3ks8W<SC{Ou
z9!$?MBR8RP+BCzgLz}(<423qeKY&O3Y5WvP^(;!D7k`+(lESv1qo?1YpXd1})^qWm
z=k2ExdsQ`TQ>GU{{GTIGe52aFiN^yHodA7^T4n&0AY>MUqU1HP2QraPk9JET6Cv!s
z05moRYr=CO?7yT}9xlkgyIx&n*Fg)po_d^7#dVRev&W|8NA*638q!3BX{A}$Q)Q|i
zL8AUbBZ?3l6_)>-(%dMGBqvyZQCFdpJhk5bo~fk95vnAG%%L?>7Fp<`kiJ9WCKOUu
zFY`G6gFg3+K-BI<AALIo#r4s<A2@wP8w}kt_LfpkbXP|iVjb$}o4~mrbu=_`5OtI>
zOlc!(*QSl$jS2QGNqUgHKW(G{^remLzee}=SwiV==;;x9`rGvB@9?x||2_Zx2mbpt
z{`-&o_v`%kpXiqmI|8k-OL3mi_=M&duGY{SrqTSK>5tYC>W@ye4y{Q}b%RM;Aw?(W
zG8im%O1Bo7cDYIw2cun<QczsGJpRCGmnV8@mo7Bzu4gh7JM_!~u<u9DtfJliZcjHR
zDvd)B+BDAN&^VVQJNni*3cw*W&N?B-HBN(`xW?g6_gmw%HoO4Ir_I{Jkt!PO<(OU*
z3#auX)G`v4L#vjviY7aYRjG{gd$<J!!^YBHh3AsKXR75VqyQ@UImAM0Ck4e-ON~?u
z77og%XR~cXmch!xR?yH6FD>pc$U=lh@RSYx^J#x#o)?E+8fShJP0IolUE}1R;f;!0
zSJQbONQ4UeS2%XXEhhg)!5#IU;qhDj<MAaiq=UrEzf#^yS^QP@a`5=;UzPXONBmU`
zp}4rqr>B%K>>!Qox^kRkptYMUql%K{`yUjz(92A!Dc{{TW_Wej#t=;F$2R^q+U;)}
zQ`I94V(+MiQg92DMx777)?hDVDOr8nO9epbyP|Z+ctJ&&gkrEjxaXsz%2-zDi5pA)
zGzeo^986xmi4-#8q2kY_7|4X1Vn~OtFCIz}csNCXKlO_M&j=epRKh7q>?98Wd0-;0
zKbUl|qfSLF#z(WXIWh0{kB?q1BZpdP=M&GR^_k9_+<_zx$tB7_U}aXX4(Tgf)10N4
z<0JSf%5Xm6(v}HF3n?h>cwB;V$On_l6RIV42TZztD($M9&v8$2o-U&<xJwUpVY1|L
z!DY&#8-z;Hb1Hs=`;ir7oOq!Wr&2%aCYs)}AN4$MRP>{ALH<rG+CBn}TuFw9)ju9a
z|9JG)*_lt$;a+s({e$FQ*w;aj?jA%o^bQZ=`_bsY9)wvDI||g3bRT%Hcf^JPn=jLU
z9lpUO_y+F<q(ki`C;+9F75m8Ih*dZ5;5P^vzLfqVJ#p`VKMlgTW|GEcg(XaP6_6(+
z&-JTMRkgW*PZz~efA|m`=gE-2oq?^H*VG*ib$%^DU8NFZ-y?VLL!F<6s?)s>b#d=Q
zUEKRn7xz9G>0kzNi$urxhlSbTx*nGv6qkK@rzyP)!dzqzM}h8!;@%3)54=sB)f0C?
zj7CUyc9PX&Y*N)Rns|+gUPElA5eIbBgqGebF|A6({Ss_k?$_Xwj=A*Kjg5fk1T=9B
z_@k7`6{g7ntm4IyA4kcWIg5j@n+EQT;OT03Y3j)6iu9I`Be7F~=S3JQawc{XHnfbh
zxnc;EC9;JIlkznV9K4;NbOySsg1#c9;Z1gG0|1c8ZRCpYvJ-R<h#R-G9XQ%89PynA
zasPvlLv1a(T_gmkbE3p>3|^v%LvQj91@t5vqy^KMs+5)H7`F`&@;I9`b!4!~*HJcS
z>{y^5Qp)2nD;;eb>+vSKo1p{7bY2{A4ZVeu2{58z5Hd7A%LTPG!$oEb?TWF9INGSc
zzs{3D&KEywMi6*`X6kO!F>sMZg@%g!5H>4lo`W*Vp(Jd+?MoX7``<8;*^GdjjX||O
zU;NP4n4KgXAa$WBwOPVWIJ&zT>zf8BPMJE#abF92C@>mc(A1g+ot;NFbRowYt6&mu
zAlC&Hv6@O0kjpYd=U26?h#f~-4bp$(mX8QM-#nz|GrWLLU027E*fr$$Qxz`e#Qm(u
z4~@9#X*kV_S|@P;+zSnH*^6w+rKEwhL<={;AfE`Zw5Ao(jSh4kQjBs;O%hDFX&@6~
z8&jMkfYiup+J=REBD!$p3_YO}oSEf!%OKwz(3)WR6fU*_tHkT-5S!z9QP_9b#OX`N
zYT;=5&65||jBofFoma>FY;c0%>H#0pj<M6EU}w!J^23$oWfX@?{x0Ht?vfX5VlJJg
zntTALsYON$r#vHmh}ZDzk6(Tg^qdJc8g1aIBm@6m1s>B@Cz0JUf+fp{Hh3RU1<PjC
za?sfX&KF}A@tY;a**s*jrKOWIML?c=*kdQdT{psJi)n3|4*o-aN9dGt!lrrRrzSH$
z?;`mtO9a&%kM7{%q@8b=NFr-MXN&m;DsvK|_7?8T*n?<zI46udf211NgY<X}%iVws
zNWE~16h#=Xw^g%RB)g8Mi^)ir>}(|v89Hl%8N7=Tm<EpV@3GSj)Y{SvTs7cmxX6e!
zE{1oP&B?2XAgj)X&}MsV*l=&tWj$<MXcJ6dNFp}bEzFGwcZbmB035kF3<{DJ!9?Lk
zBIL`G9%rWtrIf+NivevTeW7a==;j|(AEQF?T(kdxjP>u+&op-1?Fst%Bl`J6`uSb@
z@MZkO@H;8<pSI}f8X?|pkGI+5_4jyPY`{H6UkM-%Z(qYD9~GHUmn8|eN06?i7F?NY
zT;&tO%qMqg?O&?^mdH&SQAfhD%%PE*H^eaMh&CT4KYzsY<@iwy`we)OJj=zVZC0DM
zVQtz7v1y~krm&AqVIiB`H=EqMn4QQ=C#Tm5+y}v@GR+84LP|*06O751h<cdjkmlAp
zF2yNSeYmYKtvykwHlcdu@J|)s!=40D0_||R9hUq8gANth@M7{^U_EgyfCEv2(7;lU
zJtNEC0w-TjUc_`qE+7;K<R*lKiw;vwSty=Li@A_aS#HCLn=~mlNv};dNtsK%CmhnH
F{|8#QwJrbv

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/utils/versioning/index.doctree b/.doctrees/autoapi/lmflow/utils/versioning/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..c99806125e87b57d960b1f1b1fa89b499ab54866
GIT binary patch
literal 36060
zcmc&-dypK(dDltz*6H*B3CJOF1_UcC-60udaR|y78!&>yHYT>w$;;gC+|7(;ch)nr
zx>Ffquz}J!6L7*1Dq!r8#8r-Q9#wg{QWX=&{=-$NN+oesZ2z5L%1Jqi6FVfvPV)P@
zr{~!-`<OeOs<b!T)BSy4e_#K8-96jg^Z3y7f9Gvs|Gb?}OVgXHGfKVQwCbwkd!u!$
z)^4ax*MH7G^o0MIU-8ClWyP`Dc1`uYEyz*RbhB=&P5+GVO;UDOH=UWF{D`C0T-|Dh
z#kC=ASQ|OxSF|y2$kkm_jotDtGaE~$wKBtHoKtNFW%cH=Peo8-@I&e<nUbIJDW6+8
z+)`kZ>ubBcv1(WeZJRe7*go$KntD_9=i7^lH?Fv@tuMA+)S2qIO|LstpOKoWI1UnV
zJ=9bh{A#q`Zbe@fmD*WbU7|c~Z^TuW>6WkU)TXov?MiK%w%>bbR<j!Ftle%lReN^D
zvd_-mcmG|esQ3c|omp<!tkQNZrKQhuSIjb3%to%5)thy7b;fp3w*#lPwimT6-VjTR
zIg<NWy9#yD-UiQIgZ~Hd|9bpCMD>DeY@OAn;tjSG-S)k!8dkk+s<%$(#lSnja~#Rb
zpbgKvzS7p-t{sUgKd&8ti?!{%^+e&{B@(>xJ0j<^IPXg&1rdeBJGeW!L?yR&WB_^p
zcAi~33_p!BKkagr#kQ%~)tY5iot9eb!DWLizqf1QBcC|>*e8xXCjR=Q?tw_1x3nR_
zo;!Ne+(~L0PP@@i?DMxH`tYl_FMup?F%!G(d%I)bs;GAoXcOc3XIjh4sy#j~&^46K
z7Br(VZ`fV7t#&KpE<*7LoY|u<M^R)&ue+LmYMk8Hg!|4SLY8O~o!W0qz{7C?#3_qb
z{d~qy*tyNNRx;^~h6oJOB4Z#CW!Y9*TCAXVY_4^F&RR6)9^nq0KA8iE8GirvM;4fe
zD?sW*OKsLcXw_9vMqKF8o`s62jhMQt+KL&Iro9~CXqa2zM;Lk~A46aQW|g<Kb>8=e
zTj!6UlD--z4)XOE5UiI#nz{J%hEX>#n(rO#ipzX-DO8}L=uKbyiZ_C9?)etFmu+N#
zw;kWr)t0H(bl1PYZm%eI(_iz(5D3mm1wDc{id>Ax{duOzRqSQe^}PedVPnAl4Y2pc
z1N5l@`1W}KVV^b}h3xrw$kbf7Q*55o_G{k&-I1KnQn8{qQGsWW@t2|S*%b1E&R)=U
z2c=Wmx4TqSZ2lUodAgINCy@zckuS*`Bu4VlH==?%axbMtdV6E!IBi<ZWzKv`x~<ny
z`jK?$cqVUbwPDUVErmuyqb*x?RQsIDn$E{iR&+rNjI1M*69W!KbkC{OptGi0j6+s1
zm<YSGf*y73u6C0RyBtmu`@C36*szQ}16#Jl+g3iL$p&<(#?G(=Gh-K8&ihj>CvVK5
zJ*=H3G3Y8|Vm1m~?>HvwEw!pJ_9G9YCe#Ol397C0u4Xl>!5FJ@h|;UNQ$<wID$8nB
zIj86*^`I4zLPNaMho_yhdaFe@&ZNsY*+LG>6B8Tgt152QR29cfNIyK?&_k0%NpwYZ
z-HM48H*Dp+B=wwWHYCYQ7}ILd>Naz3o|UA#wkb(%v`tqB11qVUyxFP3+d+43B)?bs
z>hHyX!8nQtmK-w(8Be7R`%F3^mmj?c!j3!Jl=4Rg*}NDfOfOUx4h2Jn7lj4h`_@eX
zkIHWf$;O8X%`~yv=k3WF6b%}TkA|?9jlC<{tDT1o#zQPz`4kh58nmy*#{)(U<+>jd
zXkXJ_)BbW^`?_~SUhgicDe=+<_Vf}mq8{4#0CNevzqN8G=50!j({j{ph>k+c<H&Aw
z`9_;cuwLJo4+<A`T2u@jzj2D|&u!ZsM&Vrhl<_nZlONfny5ibk1Yq35&FTSzb_fwE
zZRf*Knw=vtp!$U>as9&Sg&5l}oRPpZ;sL49o^42j^~)4k4%WR=4Y3yV70SyVJ@k*)
zFLj7&9$x<x<@&*E-cyAcd1a|lZ31L}hj|5%ol5l9$+18z5|lq4iQ?8zSpF83;8=bo
zDBKN}KhF&-i{<YzF`co@)fLyKKP*>eSdI;+x&l%xe*(>p1h^kqcZF%_{+!~-p}Q~E
znqx-BTFn0iLD%;8kKMOAL@^J$V_W;f?o3`^n^W6O0`Hs5EdcMh0Iw78cj<O-Se=5u
z0<8WxDBKNJf5h!5i`B!>-o3hw!0KZmbPA?5`G74blt69J2eT>Bf|#aBkbHUr%}HbW
z{S<nR>8XNx#DM)lDp9gip6?LyJYX-NTtA)iJMw^CROL-V_r1(FnXWlGk&eZK^2g&*
z3=ry?n^b~#&36ZdyTRXkxM5}C{z)dLbJxt(71yS}t~u%{5&rbEh(egBGjn)LAbctg
z;dH4URiq*N+Y~bn*-Io9h&7Eb2GK3k2fxxGTzOFa1C;9rs&6lVYTA(-{b)OH!uvAw
zNoK-J$-N=*>r{dx@x`EUH%NSe>o1GMzh`1PBay4y2qY$8l6d%6h>=9jTnxT9j=@Bc
zUWBL7_g@qzj=o($UqWsSdjB1Svkdfp+#x=B(A&DLKj>8opqH%YCUuSf&K#1#9VPdM
zyG!Ar0CztO3U`CMf9Cqj;_e`{cdu?EaCd(k8p+4&h1f`D%|+hb>m)vnz4uYrIQDh}
zdr50z;5$Y6N)A!)?hvFr@I8cb{q%c>3c#1D=q91>PUeve`Y1W>?Be!H<IiO(!F#=T
z1%<nT!);uDS@f+kF`auouCBN?{f$3A8i$AMSFAz|$_nHn@xcvHl}6>~D25!BmjRVB
z`(oJqd=T+69pj4~LYRlmFQHsN*bLUNWtQ{J<W+Z*5c=!PFBybVa&HLzb1K0R`dm=B
z8-#wA>o1GYzh+`OBb2M#2!uvIf&>wc{gm7hoUy#QxQu=bEvq1n#($%DaWwA9%8mi?
zM?qlA0P%+%B9#Zkw@|JhAm;uUDyiuvVekjcB^eA-a&H*iF&JX-`$6GuF!+yLe_0G3
zfcEayZ3G4rbHl{4jY0$_a^~VMF*i&U=|y-Nd+(-@MRUW1+!*rSL)l6WPv6xcK6%KS
zN4b8GS2Q<F)^n3^cak|IgF8x&`?|QL(&+RND#81{TY|#f0O2^-Ulw;CXJR_{eOz5}
zZPsCQDq9#;h!9zUTofior?QfLs7hn<d5U2)I+baOLGy(m;$=F;=Q@Ni51Ox_TtCn(
z8l6h3yGaOrmiZ-vP)hC%p?^drI6}Rka5o5D<NC`Y^qWjfXM}Qf-6FI&AJz&7Yy9%_
zrI!mEGwkQtl%a3Q)&?6l$g;*Z+89H~U<^|(tXIK`;Mmev;}DY5>w?FY!j-$xvfxDG
z^?VBt$xE*arscQ=tAcml&o=l>-)S{nY<URR24~Wv6~bGFl5H&Ilvqv49$F#%^Ppj2
zAv*25xf&+Thy{lo){`Z7B@H{+CV@>WSjfk=sZ3Y&<}_dIpSMHlQ0F!Mi7iSoWBFPO
zY|t3QwzurIAlkYyNq4G_j{RM3+fHm-6Wc6^FWD{+mX&tV(N-|Ab50xbF3)>d#o2dX
zL53(OHxve?;%$eGN3ax~Dff1eEo}P(`<~d&k!YSqh2%~d*M{O*1g}K|w|HZ$Mycu;
z*HKp9M+g^0W}Xb?NEfnrqhuzxag?NYGH))=Jnxd&9;kD=x>Cd5Mc>;gE~`qjrdf6@
z4xJhz?MD&2M5e2}(K&e=@~|@y+4Z{5+!5=f014y1w9(ub5gQp($ibkFU5D_9?~RjR
zu<->OCzG3RykTyMuf66C@eM87us6EYHcjR`u7AAhS~V<Pw+Y~{v9zbL!-l0s*~qr5
z$u<ttYZ3RW=qpCoEZ!3=-Zjn?-FuG2>b%NAoEQ;ARi;&}6~1D4u&%gDx|YFCAY%oB
zyb*YUZ3Z${=__j~@$XIR2;cm;=8ZHhaXBfHkMO-+Yu-d~Q+(fz96|M*Fw@SaCDKGb
zl0^C()pcnj>KavXf8Z*?cfnGv{+Q0Vf+}_*yDYhKs5{Qfo!n<d`o;Wb9DpfFwyqf0
z#L~P=vof)Enem=+D4ua$nC>M=j1O^=-A~218TXCj_~^`PFNC5Pt6kS#G;T$bsH}7W
zd_Zl*YvX@66fC7V<1XAc?#4%fjRCb~Z3Oe~VMwHnhz%9{{7fAie`7rgEgffe=Iy&l
za{Ql>wNpk`lW{-p2ybi<KfVeTa(;~L$?)T^p%CXs<72oqs`yB{vW65eU7h|68!BZY
zO~w*Wr|YMEL&Kk(=9?nJ>);@t$}%!E3|w~Um>B&D0fqDj?&kZnk`!n>3YEq`(Z^%-
z@j89{Qh4=AT&)?8v(G2k=cm}`r`hL|?DHx5+#ssDHMXiVz9*<Vw(4=6cUi)^LvCz*
zBiG!t<Jz(1{d3ccmSs-gHhq_3I_k!;+y5d)>%?w94Mj0_tEAa2XU9N5a5;CQq1?o9
zv86q5oW#<v!gv8N?}qLp!@HeuWFAsvoE~>)#D%5Uz?jDKY==j@&_uitXV-n$xgG2V
zWP_;4H`f*V#v5q*>6xQ(EftwErtyajwk~B9PnNWZ586uV#0+m>PFR7fNcB5AsBUFd
z-<(n%PHJ!*0_|aq^(+I^q0;G?(tA26UCAoFE3VWyh;L!Iw*h)wO}ODL=zR8JIE;Hy
zZBG{}PHxJ@fIJvQ@^NOkrJVH=iCy4K3)^E(hZyKfdM&xJ9E0uH269j;ww+8FABHjM
zXD7x{Bo+rdIS7v1ob09#P+g~tUtn5{tMDiE*k$3S(<DzZFIJM?9Sie{P{vXs&puAt
za-XXgR_jr2;qkbIIj!}XVlD5+?Tqt+<{D_ZJE2A1JU>^g<w2R2pnV2f676%4x`Mn0
z%5O$~{QW}R@n|P@IO-@CJ-^JhP_CxuCi-QF?Wu|O;N@=>*3{_iWw7?{koBw<&aZG2
zGjwm@Hqm5p6}kAkg^IlaE6YCrKt-mE9%}bSp^|QDhZ9ruCOcjzMyNw<JRzzTe2c2~
zW}#Mkj<cj-yLI)PVz$*}HSq+b-D+93TXk?YjQjDnMF$zN@!wD4weO35x+&yc!5)}b
zsmk>1fzMU&6_XWaMlkP?^!^uN-Z2Ru$mboAJ(+pOPcSO!dEP;q#Jq#9#k}J(`6%=d
zEyRvLhpC{Uz8Q?=Fox@I>LCR5KlQkIL`*%7(-oU~u&d3UdYp{;F}8iK+b1Qx?o%r}
z`M5X7++Zl_%uPSk!~tig?*mRwW3+ST@HD&pP`hQSlyI0$Mb4Z#vvISK2N2=>&cCN;
zET=+IY!-4udKQwm`7xM=<lT-a@=U0p^s^R;t%-TavEF0Em_0dII~+*~V?6z6PLNso
z)Cg`=2oT*yln}YaDRfKW_{~C5gsGeH+Y`lFipFn&mTtyx&lYPb8d(Hd663e4yLG#k
zoqr5c+1S|Ql|s|{AKo!FJrD0*mPPKgXeI-vV}}1pq2WeR<hT}zb5fJDZxpKN+c*79
z&ZMrv&!+sBs(~{x*L<hY?k-$obknF`FV<4jsHJszG_1>`vAX<cv6jwZ?y4>WqecBi
z;ovy8F{P2VvrtFht%<AINTgj;Xm}URNrJQ?HX4Y<`Dme%zP%IZ*+}=ix6tzb+>;`r
z2Mbm7?VNQaqQfD0566PKTxfR}uIUQA?P4vRiAd5u(@#M@U8tk)2=w_BM9QD?zTkyI
z!@CJVau_Ao^2M@RlJoH{Mt8{q@c>94I-W&0l74O^FJxSTtnul+WKew(ruS7?A)nqy
z_GG5_ccD<v(|gh+ruTF$ruUcU%=0qy`x`L7zXt_N{j?|q^gqA<KH4DvY4HuZV)J`;
zwb}FgJFp)inDK~nKBD6hyIc;@*O?V%XZpvB3>P(`Uyf48QP7yQvH>e6wu}{x(t4WJ
z|12adHmg6Jp4Ar~e-ezj3-d%1*~t!8_wYNT#OlQS{)*h7u<p5%oX01B{G&=Ge`v|x
z$8z0dVpPj(3u{?O=+h|L9XHd1#o$+@D_Nmrg|^2|7iASr)+cit)k&m9wK>^;ZBF!G
zn{N8FO!p*d6BnU-+J#!myQj#WO!xFAK-Y8kM4CkRMAxEw+FwLYnLg?;`l$a2FyHqx
zXd$5gK5B(3!<N6S(iQ8Y*wtqDQMbUG=>t`|eE{9lAuVP5q`gu@1A1Hk#d^PrFs;*r
z<X;Izu`XvO-Q`G+?+LmZ>Ai@qz7UEo{RmlNWulAeZ)u)H>O~-_SKt;oNk#T#Na_L#
z^-NNvNsts>3zEudt*IZ#W<XA#L8(#$Xd$3~a{4Y+f|1kr=!%gOyV`7WqH$>A$eC`L
zsc4^KLSX47#I(CQA#np25dTNeR79jpEFk{ZP!!{&8`GSWJ=7>5r0k7|jkb&zb7l_%
zUWt9{z(x|O7lDno!Yy((itNd-(GL)yp4o^r2{xi@!A7!1nxdi`K}G)(G>X2*NkV{V
zpviG<uuGzFneZ)C14c!2bj7HMU2QfMy%RW39>CKr7xi$YPuVdNjTX{IZrBLvK?G=>
zMo15Yq8J+;OS6&Wn5Te_BzGb{QbMt%M@JG1*MW~DQZE7@)!-I6A4T?L_~>2~>Y0y7
zli(w|7JMXarzt`@212?IWlD{Wgn<4D=|!pnBczw;iV+gK+H69)4Oo{QveYdt73~+7
z$t^unx+2%$4Wp<(4%&|he~AIpABLhBMctI9sN6HJ0;0;jAJNrcg(^zZmBiY0peu>g
zi$GW3gj?iv71@)at5;E|XSyOyg0ARV&{aXpO_A13Ag$j+$x@^x1c+AK`zY!yst6;k
zpV1W~Eq1lpq;(?TfY{+!-SSoLzR;}5(o}V6R^|py>-KJ6w}I5|P!waS<7t+Xodp&U
zl<Zc-P1l5iOLLRN#&zH(iPVe0O$Xr?IX6Z2WVmSpg?i>D(j>Tvt_3&An`(-dj)Rsa
zQK%Fx2?71n(gM|i(b5^ZVzk7rHk+2{IaA{Js%}ZC=s8nDU};JscqSxnAR(O#nu>^Y
ziD8o+ieiLxB27s2I4MyrHWtM1bJ-ga89f;aEloxe`__SsBvLN|89fcR$jK<OCqqUJ
z6zZ9bNRuEVx)x+4YosYEIsq!OP^1(U2?3&k_C7-T3#tL5qOZ{vqat>-*;GW+I@xhu
z-7-?qv`!|sG%?ZOSSEM_`RO}B`w`(U!B5`~MKOMwOY;*=+hny9G?jZlVyo9f6{Xoq
zV(mJxl|<@AV5>La7CBo*_GH-VZ&0Xbwjxb}t>{{?RYA*5k=7hY>u*uA6ln<oqSf|J
zS_dWsX<bEEjI`L*dXQFRj<=Ja%oLs|gzcF)<cL;Hid)m-U?JSP9=8siVJ~4xAK`}c
z(ju`H?WA`ZKQZh`+tp3yNbt&FI*3ROH3V5Qr)+06ZKH4mM?P&WT1Vs!J67Aqc>vmB
z?^4%RnvTg1C8)M+YZ))OcKpY1gwv2~xp?pVsV&~P$fMTrwT;H04S18awvAKBuql<x
zn)fC|3fmexk0ZeNsY;HI-N06j)p2IYW~WZo@e2B1q^+(_cqoB4Y_;81+x6$Qop_b|
zvf2#x4dV+Al2RR4X|(W?_wBZ2;jPYeD%Yy-4b0*d^4tX4pL!M<#)GeI+r+mWbYRvZ
zPIsWYI5g`SZ_g40G_i>`mVCk6;owNFX1h^cU9#*3-b6o%9eR4JTC*ArRMUZ|DW0O5
zDod%Z=&n{}M|GhZ&=eHF3z!|2Fk!1~+pejeQ{l+1-ngb<i*N9C$lI;1>hPb76X%>L
zb*Qc`wwEDqaz(M5Xbi|<t4aeYV+ce|rOJ?fdK2ENs!be-Moq>WYr9KFP8?y!j9r<$
zSygS@va6cXtedJ0-;=2|7wwn(JSJs`N%6h?sXHkV#sVPcO>()B4Wjgw$s4J{-Y%*G
zTAic8MJ9W_`VLcRF1P80-28yIt!mTvF0s{`W!I~9wWT)eP@^-)4g+pWI86;6(d(iq
z;Iy<!dhvTTO7gbi;5O{7ZZ}%hurb21EJ4RPiFOl*z(CASmcp7jML$YvIcmFZRq+~l
z<l3dKwk$`j#!IqRQp3J&-O*{2J7iVsx<juGM@0fFM+8U!xCvCo!C_<k7!b6YAX{}w
zH&Hoo%09Y+61;4ljtW6+NLn2s-)<`_;n!0G-fp%d+f?y}dkw9>p}YS5yRP>pTWYHc
ze5k-@jkiQfSyW&<fx^37bKRCRH#>`DXNJEZea5nvfu+5alNo@VRab2wJTA`L8%6q&
zgE!s;e@@y<wUl^DiNs$g#MAk2)MU@9_KdE&OH2o-U{-S*CX|i2$|6uycQyaisJ9<&
zvQZ7e8CD=TykyPW<(zN2%4${9mo*dr(S8R_i<<rf+ZkUCw#Hw;j`?V(eATpSO5oc4
zH4WR>YY05Y8@$Y1h}>t}EgTIp3F}<lh5tDXtfBQ-HNClnHW2W_9wG{-y_m|TH>mBJ
z7T)1b77a2oLX&AMs&ydvJS@;u6&vQgy*1=ITU8r)?|c=<ND|{BN*BD{B$*=asOA!9
zvo&v<+C1maw->!p&L)5_S|X0Z(}4}Q$`3MR?PfQOLd$YNw2*-V1&GJJJ&9CWq|A~b
zZ7KXROk`ST=_8cLuPerl)U6(&j|n>Ibc{ZJiX%XcH|e87HxA;X?Tt|>c90%*9jx3M
zE3?M(ud%$?OYk<jO9HhC?WBft6oPXYqjMC}hm)NQ`&!YldPBN{4%xpzgOcAOS^Sj4
zL2U|B`KexHsdhUzb0=9%X0ii8*@=wo6ikE89%E;v8FU(eL8k>6v~j?ojRgj+(KKic
zr$Gy$3|c5<&>{qb7AY9?^PfRK4;u7j-JmD$20h0z=sA}`lSqRmnFbBe3>v_-y&dtE
z&3wJ1rUM@oEcl<N5P^w&JgD&$o(~iOBs3N+GXx9EdTA;#%srNnI3~2RLVFVO(HGgS
z4`$h$mKEH=U;y-5v)V+S=y_|FY4S#kUVh}xApJSCf~Wisf+}!Y1V{&M&H*~O!jNSX
f$=kzZ*1)&x&6?S+6QGuf6UXb+5d@scgFN&9L4-h_

literal 0
HcmV?d00001

diff --git a/.doctrees/autoapi/lmflow/version/index.doctree b/.doctrees/autoapi/lmflow/version/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..d4c56079360a278e4881b95230baa866a857b369
GIT binary patch
literal 6597
zcmcIp&2Jn@75CSC+McnU#C#-Pyh6Ki#Li5zL})SINUN2&Bry^~fKWoGr@LmV$35Lm
zceOvXiUbl_WkhK~O6PzOAnyDD2o8JT1XpfIT#)!b_`Ry`o}QlAP9%sl<EdA#UcK-4
z>gA`Emw*1!g8C=hQNSJVv}@Y7=i4kwWZm|yc+5PJyhv{SEP0yj$wp|NM1CAvERhSq
zu(;#eA@h>QiEI(NaNMYyp07pB5{~cX$9aWUdF^qs#~ZRD9O1IUmTbA>f$N`ik69Se
zDZ&Ke%J<kQaWJ?4EYUnw0}~vGgm21*k+aN~Wi{pWP_DR+$CAUiZ_1`AMCkNm0SQT>
zWyg+)`<msNQ3N1<S3Gm9f7R_cm@fLJbriAz;l#2g%pq+hyv?ukHGYFH^Bwu1$Ne$u
zg|X+cuy^8zN4*CRzYVdnKVUHGX~ueHEPOL?dRp*aDtOQFY<AiWBZ&0Y{iV3i7i2}j
z3fj_&<X?dZ{B7vztN6Tw&rN)8kvPyu=qMrf<Vs*VVIsH3z8$;lTe~Gi%WHb0Q-bsA
zp-9Y_e~ovh5+3rmphv!{9b#(HjSL{0A5Ar^Y`8rGO3em}A8DOwisrep5Cf#*dm5d;
z2d&hVRyKs$k6kk~EZ;SvfLU*#pcRELHy-`)lRHm8dH-qlH=*mkGrdke{$E;j|IXe0
z12TIQkH=<sb`K_oI^}j6d~o${VLOqVg_QwuXG3PQiGPNXI*ws9n_1LU%DAb+G-Oo_
zLq87YR7F8qgCgHhk98b5acsep`%O}v2h|<HI)3I4mw9S|z>kUqC`I&r`)rO;@VOlN
zCv$L>RmkmMd8C0XVGctx;Dv+o>-)jkzTY41KPhN&_h2><x`n-aPaY{1?;$AH0_NF{
zH#AO>6pB>ee{-st-D<8QSZKNha{RYx6s;EcNTy-*hx5Y-p|5z8OTk$ptHD_ZvL>vU
zDM;nNhs}P2<allf$|@v8=1Sy^i*i<i`apPN(?R<FqpV>i&H~hmWx7|cV$Dti*RdRt
zJXhN%X6Pjo*?<{z5}MERvW`uaos&a#fiS}%6N&ut`DRj4{A=Xn|9kY_Uxe~r!r`}h
zb!wBBMVqj5^-9Wr!FTvyk<y6HujtrChM)tV0r3~O@Ry|sOKWtNk}tXRI{(LOUUV`0
z6L9nCWt4snBqW9NDajSeJPE2D1ayEsD3g@e3Yl=%^}V6aaAw(=ZJd5mK3znTjnlEa
z9|a~gG4&v15etu*@|YjvY*uC-smwfOd8TtPvw~ZKrsbEUXo6PS733xBf~whmsEYIM
z$G#_!#3G##w&ff=ce2se;=)4F&(4ou;s+&k)Xq;RmNwu;-I5o*UPn=lkoQoF<hzBO
zw7chcyE=+XEyk^jgLS6SC@`7!8Vhm-{pfsgrySg(9V2qkiitR!=_hB>rEWWsvlTw|
zxYW93eJ!uZb}5<U`Y*$$`aj6E`KlIXPp*Qk4*W}9CD-X@mB!S1CJB!!AT_Gu3ziz~
z6W!~hZ{*~av9@5W1=&#imzW=YK&aF%Z_62}r@S8K#9qy7op{xZ*6E7NczkdHk8Q1E
z&+~;@>i=c!=I!q7?l*7dMKm`c8aRr`f=L#4Gb7;FR1|ihD01;6Dd&WdrMWfmM``h@
z;t(PmLF`!~o?~<I!~E(G3s)asa&<6&wJ2Uava!knPM#y*vkNa71=zks!0oA;r)neN
zi`inD*|J=gJsxJg@iCoyl{d;Og{Jk`VJ8-zjfre$KMm8fxE~fQdRZ$--9uK~zhLN0
z5|v&@c%1>uu@mL=d1J!=M3M8mGOya%r=pHJN_%)M;pGSWC$dS7kM0Cd)!D{KR?~)r
ze=aM!{oqwuAH=S!f<xcmG=y&%=q)K@f39Gkq772eDH`#1mw4OJp$vWAp{1(!)JZCT
zv~=o!eT5O(zi5#&a(t&OGFf*|H-Ia)Da^8zX}dQ18wkl7G@&}S(cfvQQbBR;`5pZd
zoyeN!XFppRxTD+eiCjxJv-Ku0Qt_f3sqOh0EW;y8bYH}TK!r&rBw{aBRaU^}M3il9
zs{9IsVfrA>wsy0duysL^Mn#-D%N{?-h7O%<D8X{V<#ELv-DMqZ%Hnv%pI@ExK?Q$L
zUKOF~MXpLnMiBZ#v?qE{Q4vU|ZgPJ?HZvS%V;Mt5d{MTnIHWO%kq<rAGKe02$a<<&
zOm;T&QOzJ!PG{qF@O{&bF)+vyu&nyA2x5^O@;3VJA;Un!6o_Xm;72SH<~YD&HS~P}
zRuPn^iCpZV&DSg#xc3MM&2$-uE>`OlFEmJ@UBt|1a%%trs1>XLelFK(c;v-n<8<JM
zW4snx5e6Cop88`NNkt%aT|=156vQ}j1UIrV4sJ>hq;s7^xfZg(4~5~fV+KVo$tE}5
zd|8p3?9_q&P_*JGL#oJ%&HC{W*R|-aLeD+WW#$-=1`J^_5{CG5JZQ^c9?~GWu583&
z(D|UFvdoRyO@oD@9~#{BY?om&QzcfBfMi?iyg;R*s3dZyv{NE6+FU}swlv)-583G(
zvm2$ua)V@m*F_vEa#drz?wZ~(Hirxvr?kF=A)mYwGRqHb!)EmK#WgfxI+Rre*8s^B
z8gcB*6_6}j)Uq2>kX%CEMk5rD10#1vC{_U+^u3{H;uUZx+X|xGoUA_u1rdvF-#`No
ztPQ*+{fHUGlgdjhVH`V=qn)*F*iJ-)69|%WS<8Y10k;NWBIFv~0>EohG-KenYJ4WQ
z=@%}+Gl9Ap*oLUtF!?H`q51OuqTHlNgXiFF!r}d6ymB9I+?1_=1qMBR5ucV)7$ye+
zydAt#P{=QGA%bYX*8@1}24{l%Ue^zYh^1?UsTcsJA=nt^5$fH-g3dd)??RugaA1|l
z6J$#M-Hbe?e{$I)7Iqyb2I>x^f*u!R7ne2gq(T&_VNPA{z$eE>9-O%VX`?og8_}63
z%u|CqL+;`O|6QSnSt8dW47Y^wlNeDQK~Y!KS}zUPx6D+vI~K=O#)9E>yg|!Kg*bhL
z5!FP^l+~a(O#>5nk8e5N06viBg)K@HQQR+~Ib-r&*Ee-A(ggS@fy<2h%ti#CfdkGk
zMVQEI7O;*ChN*>X;AKD=S8=_W&adIC?m%a=iCkvhF-?+XU1t-7FFX;I&_QenjOb`K
zp{B^2U<#fkFcXK@*fckOn#ipgD76R*3~GaXLDLo*O9Q<|K>d4f^dA($(ZA_yjlLT6
z^)LGRC%$6YAY4^FX|k=bCkk_-uux0iCO8qm%k%(qv})RIWjTL-c5ska(yVM^79FA7
zCeNuf{{UGhZE9As_io%*obh{#E%lsJZI^n`#B#mpQA);jjDvK+gWMWIi6g_)$}B?2
z+Z7ZZIPPQiJA-<)1kAA;GiZ(MRz+!RhCXUQC=x{UwKA=A2SpfZ&GKVPTBvxKUbrwn
z42|kq`#6QafE}JC-$jU`K}4{_))Bl?UzlgN8SrWjR*tpmc$OR6<j6z%pNXbT2{_GF
Ix_#{a7YNIdyZ`_I

literal 0
HcmV?d00001

diff --git a/.doctrees/blogs/benchmark.doctree b/.doctrees/blogs/benchmark.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..5e308cff6c3d8acf410ae002bd76379b928583ab
GIT binary patch
literal 169635
zcmeHw3!EfZRd=$pGyBLUn*`WEHrZ6dBfG;)@6OE5W0#Py*<_PtA9?I1S<Hr^y1Qn&
zdaAoRRn;>)(U8bP5@|qTKtVv159A>zAE@}@6H!DFL<Lb%KO>?D0YnrPKm7jZR^4-}
zy1ROAPr8c$o8K?9Rn>KFo!33*+;h*p_rbx}zGdCH=g@zRt$wX!SI^|kVzKHLEq|r4
zvFH}+6{{Moyk=$B53k(6va_+tGnag~?iH++#yNPSP_mt(XH{43S!rBAuLri{=OXwG
zzEucpx60t9!O~D^!#yiIOPd;lfgLzj?5D=fM~@wH+$H0Bt6C^kOmAVoaiD4(s0VJv
z4D5n&<D%)*aphKxL!Mc&mRxVam~%bjmYP)^Jt43%jvhVcuh2K}iS;L~GlY8&TYl+E
z$cLet395sY(nXC;v&^Q_Kw~K4&hf^2$F5o{$Lo2svDpj)&(7BaaFw_?U>AMj*oJ~*
z`aT}wdaw$T#MQ=Py{2yRW?{jz=IEDtV?$uh(@!g<t)(4D&MjTuxOR8RtysIgdbMhK
zyNM;cLpI8PfSkWOL~3{5ap(QrdCA{grMObsUfNQ+tTa&1gJGLMl{xR3wNhLz=N&Vb
zjeR4=_@42JkhR+yiDwa0(9~{fs&x7B(hj1#bOnelT?tt{75_gS|6h&&cY%=7GfSf&
z0I!xtVi=qkYMD}#3+02xmc!M+bBpy*p@e|9H9b^lJVzf5;^R<+#<L?v<m6XRy#akA
zrgrX}VH$;!8I1bDvSS%XO>dt5R_E&`E{>twIL1z>;4Y(N`bOTessJd}@f)BV$ExDW
zx2pxq2uhZb_bhWED0yyuzGR^7gJ}7tTTkYUTdLxBw`!rCYH`HyEZ;J{Ldo)c0}l)|
z1m&VT{4L@ep5@qP9w;qAR)K32+zM%^5iGezh0qG&idMlc;-!*ZD230(%O%(KEu&s7
zS{@N-d4XvI^IFZZ3t@-z3s%+iY**n10=x)2TrT5s5uc;4RqX(sk_Z%EHfolKURp7U
z>|))stMlT=z{5xAhPYa#j!WR?qH7luAhz$<E#L6#Kp9L$Q}b3305BIR<5?*^zNMu4
z<#GLfN#<c=YQH>!etEkEoeT`&1)xHA+J$=6H0;PsKx4u_VEgbF6)PyYMW0lyREH)S
z7W<MB!^|z@%$(?bc{eZ)z_8=?Bq_D&_^wfd$3b61KO>!1kWs5fei>g~bQegM(S1kv
zCORNM+<H)}2jc6hKjw^^-6d<$@<z}sbw7!tXWd)3(RHmeaBIGeUKOJ3n0a(^>MvgO
znFF^THVQB+{z%lLsmE6>8)$llk6(mEsN(&CTV1rOHt~xbCiS%<zU{a*a_`)c&a#!H
z*E_Z<%|G3p=5J2b{9B><hsizIu;?DbBFBRCEv5HH6h*O@hk`>Xmf($|uqslPn&&RU
z8K4<BC(Epbj>YtsM+0}X4uy<q`W(kw0elpcu4<Glrv|q(=OD&2iuT+bdI;3Z^lNw^
zbe%%6oN*WVFetH0L0TAxgdM492BvQT0{)CTRsn8`4E&PiI3ou7o%j^>kY#d&a7gqP
zsV+Kpr4BE%ED#}7!yXj4uCrj{<M1O9@?eg1NTM5rUQwYdaFl-JEN}b3Th#~3-SvSj
z$v$9lG}+i9jl39Km@>L-c^->4H^;&}8Si$8HiwQ`2-piU+LgD8mKW>OLJ{Wq5iJiA
z7(#81FdG$gW{g~<zS7`^JKKnH+?q%Hh2C@2oi~o!3vd94)r!W><42F~GGGFr{6%5X
zWoLzdfKG3|^6q<f=5o7d{eXr}yQ_|qt5xTB34(T(wl(hP%V$><gMK{DAbjxL&8a_P
zENE8{KgY%#;V_XUE+s6-4~+qYkt>a@P2{aK222NIvZ2OiSg4{63nK<xjdO=rO7HX=
zTU^gZ+-y3t0{r4u28of$n;@`rB7q4q>UpJn&8Wzs%mrx@K6N)Nb{KFMTn~YK&8;F5
zfr*mlOhn0#JotLws3FL(#RvzNZr;q>4p})DHju2e>(%NAgkdx1p%cO$+QPR780W&&
z9-b3z#@>T5w46EEPBcr*h(-6oat%g#P8ihaL$JD0kX@-efjx?W32d{79Xwx*14hu4
zA9&#?40`8aypE6rM%D~U0%ymnz;-)};;H94)@aQ!m&G0Etr}2V13DfWY{Jy_KpFf=
z<c^Gc>t+?Hfl%!c6QLUlZb?@YLWUUImd~-vSvL2V&8lnva=dG=KFGf~t!Ff@XtGiq
zp^T0xYgKx)9F4DH6iV-CTp|rZQ%ftQcLDg__pg-RQ~H!&dT-;>Sl4fUmwI&E1big4
z*{c)!#8g6QFkviUJb`%2L?DJdTakJKx)XF85tVc_tG}qk@R!|*VOuJ8hlX#mkUSC{
z+>ka0h}B%*Cg<ZI&LYd@z=9NwB@8d(hAINc!Ni7Hw~bb%hU|y!$Lx9IhF*L^a;Z0s
zDNcSxpOasPZ+LRV$$!#S3<=Qc{r~0=qEa%clpH~z9;QJ;ZxN+U!ho2<U(5HA+*$G<
z6nJ^R<kp>H7+Y5{NQJ*3+bfbd)LxNZqS#tG6}qK>o#J$4Vxn;{wK-yx>waK3h%chY
z7!deAjJh9Cjw!-~8l>RdzzCpXRp3nT3%`MMRB;|zNiytWFyZ@o*Yt|gHaEW4$z)Y3
zQ!-s&n<&X~X==Z`?Rv`>12kbN%&H3qigCSNu^kiffN;$fjBe0BMHm=5OVX6%M{gy|
za;xRl9P12)5Ib+Z_2{nsvIF&@WyOLEyJo><sVump$wMF`xi-(hp<>iR%p;Va353Zh
z#N$EvT9Y#<?}&)g$!_mRBkR-xS~*T5TS(MFPaB~wCPhVq8<+$HYz`z_v@mX_%vCu2
zr4({d3R6k&DS^9yw9b7n%RaJ3)UP=g1Q%1sEeyh(DWXIg`if*&-U^l|GacoHg+W&%
zV%emJfRe0}b;hg!HMlBDeuo)dN^YBmU!jz?XDqq}Ghc^JE<-u*qc;O6A#@e_08oIC
z7ty3odTAPtixfDL!!39VeWu@+<j<oHWvF7Dp`0AzUup+NNDP{gBJ*rYVG-HLU?UYk
z0VY{g%PdlO3%!?S=q%eaX1lTpd%Ly?Mr!BTg|=>?nGu>+F^h{BF3{u)WKE)%<6!ic
zD7GSXEV}3kFrZMzFu59fKT6v=ixw$Rd}bon2HYw-5xT+wU+69+sL*K*jHReC=}e;`
z7rrP)W+UjXNNA}%GbTRTWynH*FNS_H?lR?cC-r{Pc(hqD7l1x}HilU=bb}iP7Yc-=
zP4hn(SH>obSoO~`P0Sl>GEFJ&^<69ORcVUb_?-0Tl%nK}JjH#&W<^7FSXP>JFpYUi
z^n^YgL;EO8#0JlF;)aznqHfj4e1T=$c0ky|IT~Y#6sl;QzL<>RZ#zJGQ}@gQf=U{i
zNAo3!@<`Q`!lETBs8``+;IVvUsD%50{YUf%jT0_GD%$rrkxDTrSKM8o_q0^SI5a$b
zQcTs1q2fV`B<40TQ(8pO?;>+WLW)URCH4K`Z~%tnUTH|7`TQm{G08uK1P;7#d~|mG
zWB<GiEql?FLVvyq{f>k1vPEz2IP#2;(Pc?H@~tU6LGb}neLCsAsLhfb!cqMSi7dv&
zC_MoN2lL3B)mna{s+SD~*P$7bmLJn|of`qNWEX?d%8~OBj+P%HDz}$k%B+!8lpn?;
z&14BGch0tYmIL2W4xl>>!qb?=*3Pi0G2ppNEzd>W7=kr#x#cG{rLszzFihBQFMqA2
z|0In|V-sc7F$kFpR?Md2rSez8KKP}UHp*XZ?SpEvXET!JF6ygLXd~@q{E@PIL+iTp
zF5Alk3?a=y6FKJza;Rj&@_I(lU>;Mh_exr#J4MD0MoO<DY<qb-Lu5(D5NfMV&$3Jg
zaf>@e#}s3wOUDF5WJ$*+q!Un4xMa>tk;I-Ta*i`{y5t;Th%CujU$gHuB_-jFB4UnF
z(j}t65SbF;=3_*N8%4wm86{mJevl!uBqHn}JRdGPNFmliUdLGJ((xLG$dZl?2QY-8
zsphDFl<$>D-ph#TlJahb=q)K7C!Nb*3|-I{T3k@l?l*g`&xc^G(WCIjdAHc=T6dQJ
zxy7mN?Cj_^qSDx{j85bwqtye&>Az&Y>2mtd86t(#8b15aTH5GL%HZV8>D;uWWBud|
zZWIk)=hk#-_$ouB(4gh0yQX4l>~wC5shAqWjiO@nK)2z>28PI#iphPabCXQP<UZUe
zDz0LjbY*b`Lu5<EctWGb6B@Ojanhw?njtc!VtUW%+ziWNdJk@tEN*9<bg8(7Au^?6
zFJv*!RO}tcjiRE=IO$R`&k&hXF#%bq-aWA=(Ys&7IO$UH07GO-#pFbyUrZ$W#TytW
zT`GQpA$m)NmS6r<i<$4<i*63R=yYz3{b>A=($fzyrMkTNAVZ|^M#Gc-QcD}Fc+#o8
zr*nI`uTAa6jiTXCxiwuH{+J<BXwdSc-BSUJu}@JkjvGbAml!8qD*lNfvZZ43bZ(EL
zViGrsih=dr#slXtM5a_wJfQ3z;(>Tnvz>9$mBo`8B2y}+VfPdj(-R3QCKxAOD)ul$
zrc_Koqf|UFF`0-5jxbKTRNTxE*-`;nC>u2~k+4w(#z~io7cfMoR7_5t&hc8!^^;S$
zQN8;I87EyT&M-u8smRQeZggwG=u~bW`%(2k@#T$7r7mAy&k!km(eR?LYiVOOFN%Jq
zBG#$NM8x_5ZcUel_c25Y4O(8bdn%xzW6J+b;YLZ~<BXFo6(3`WY^i`Ol;wsj;{NCJ
zjFT=Ef5#BnQZb#-sOf}8{Ws&JOT~XOM5a_sL!(q~V0tQ%8+hViw-L=n43Q}n6OcvB
z|4b+R&j{nBD~nwWku4R7^i{u@*qi7VH!x1RR2*Q4OsSyEpNjO6`HM&Trx_<*DxSv>
zy`>^EFS>!e=$K0C^1ZU$_cM*UT)B@SQn;ewLzi0G(D0$Dqo3DuYq~VNnjunX(DI>c
zqC)8>yiZ&|-@`cRQt>W^$d(FNyjUcFY)32-_+7?Hmx>QFM7C65B&tR~Q~MI5pTB0D
zbgB4DhRBo(@}Wwj;6vjY^>2)mE)`#Ah)k)NfGkw+p4gYjZJa;UZ6vUnA+n_cv89@K
z!gxA9@AP!WNmmwEF+`?RP=-cn6f!h%je0iYq)WwqhUhI7nfcI-DA1^wqtj|6$`6zm
zeLhpE%a^+uB84v+Ui4H;8>@LySZ*cTsp*7l7r8ZE8hnOGp+U=wc1;Ck-_&>-*|+!%
z&?^}yT`FGA5Sdaj1zD(!4`dO~_`IEQ(xu|943Q}nBnxG^A&Yoc`ZpLST`E4r5ZO|J
zaf6b@^h`n)pJAMIsrXZd$dn428C4@u%#6loYreub=~D3}hRBu*)IG$g7)wwww4vKr
zV1OYqrD7^kF))>^7}&u$>B?d|L-dx4%)IEP&}qiTqOm8+k6y=g>T>29hDhOzh9BM6
z(ne=Lx^Z|I)f0Q#FW|U~8w*J&&&6u78tVy=v#8=mQUClV(9VTn(c+0c(c%gD-R8v;
z^0|0TYw^VP@<~P{g{fMOeGP1#n#oPJXX^uuQd_nXAhT86D7OAc6KGd#mEUb<t9;%K
zTYrcV$(XIvd(+tZbBt13wh|z-Rop1HzPAaqE4Iq-HnUYe?}n}aixJ71ty8)2_S*VM
zMyWkp0jStYH;S#FX#(wvt@69gY?aTuVe7{kk*wLeFE`ent>0vn+Orjaimh~`*!t}z
z(5~1jzuU}K`MevpeuELon5|SQ*Pg9cY)nQ~O)g8+*hTf%>zfMhp3wx_6<g(Zo7pO#
zcf;1p7?G^mnrd5bVwBoy>)yC+6*o#-k2isK#a8*<X12=b-LUlrMkHglA~~KCV7ZJ^
zTecD)8v${n*t*yR+7(;ncbnNNpLfGnhY{&5Th|!RcLw_;RYta3znTu#SWnAlSB`9=
z^{+1uS1>$C|A^hjQ)s?Gu5k57X;X#PHE%DZ7IqxjB{;pEm-edVde_Ny+2i}g68+ec
z>}bOW>?hFWg6#LQsM@L9v%-bgYx<tl(#O<>Qtn@BLR3q+$3tG0C6&#}EuZTv_l_e`
zn^HdI2Vw)DaII(gB`uG3_F^&gCz%CO#m=^RmIL2a9QcQpp3!v~y5@T7@lC7PV~xzo
z?v^$>XHMdar=y56{zw_Ko4Mi8>UK0{IGSR6`9_A2eLo4)(8^hKOVBFosb9cI>5_2|
zLu5(DYU`=bFjl&B)EOd6I%2D*BlU_sQQGxmjGQhxuV9EQ$?3SB`W=jtE)j2Ih)jv-
zw4VC67$sdIew`t*Bx1Go)SqRnbm{msLu5$@ub)m|PyIDUOqY~L8KSqOq>nFXSP`wK
z9$Dq%DeI}*go=NRaZYkKo_;N)zI3y6y}B&EgdtK`tl_Mm*wRL4XFWKRDA1ir7U+(1
zYq~V-W{4CTv^;dTR1A!dbD5X8R&L<AjF2t~H!(z(B&?g^(p<V!1mqbVT>@qqB1-}W
z#`p3xg1A;B+{Xy%lCZ=OnUXLz!L!l>bgf8uH6x@;!mAh}OA-cYIVa0uJ(hFE7vsN+
zanhyY=NKYODu%}PphAGj7{qOYjt?_dx^(;|Lu5+_t-s^>ior1y48|uc{*tlMrQ^>T
zqPKKtx#5diKHl}XwGm${Ek}p371gT8eTzS)%bWjTh!oyv_|Jc9X`{3MTsOm7N|y@#
zC!E)<`@fJOQqb3Oo!t^JFh0f<h-*c{GZ`UW3G8HuEJ+xgi5rXcGf7)<ka5zb;yDbF
zB^3kX6I@s<u9YC}W`uM}IK>c|k}x*G(=za%@iIZ55z-~0#t>POFi1sR+#%MZs4L#7
zU(Pt`Qt=Q&WJ$%keJYs5rP8FgGCH~hyoDjMBw%Q493|>3he6yX1o0upN|%lgGDL6b
z$jou7#nF-5R1cJVKgU$+^5riWB84v+PV+M@ZFF{;oelOkxHVlG{)Hh@XwdSU-BK|y
zzDHR*x>jP?vZY(+zlkBTB%!k)p2j%o3gSwJ$dU?-_|>qDu9YCJVT5!^*vAl=k}x*T
z#%_SF6$vL9Azcz~V~8wCSU1Jv9J*8lIE;=i0X9QqN<f;^d<mnYOTZ5^M3w}=X=3ot
zQh?Lkn;31rk+IUH<Mj;DTRJjxn!FfVro!c7X{DaO&NS+B<^M253Rg5-=C8E0(b;9L
zo8e+ox>V?Ynj6wZ|4$eq1$`}V*)0JBWBXK!maY{Ek1|5KB>Xc&WJ$u{Ok!*>lN=ii
zp5Lv{U&jzxQUPxn3tMol1aUbdq$`L^86r~>(!J#*Bcw~hI74Jf!XRz^${k`owtkKG
zh+~YCE)~yZh%BjCw~td6bg4ASVsvx~$TLKi1i)otaU&A}mzmt(@_xojmyY`wqPKKp
z<}$U`Lcf`*)aA=hF+>VqG@Ry7wzSdNX%0@ulL+gllR1Q6=GJs+ct1m=(4gfxyQN}a
zY>E|BT&r&O2}Velgg<16EJ+xgPGo4OlNs7CFiyHu{5?ZtNd?m5v2iA@l_0*&2<ejW
zEr!UHgt2iRA7GprAEs@)pxgN62@H`X3F{`=ZPKNZL5|VUmBBL^B2xm=ri>0TI=TcL
zWQZ&Y7#bT(I?b`9(|kT-rAx=%4AEOUGIN@-h0f7PGxkKO=mSirE@ys-AyPP_;Wn3B
z+UV>yhp_Z{U;9;LKfz6f1SD6Ht()eJN9a<~{pKdHWYunS^;Wd#`I>0ig8XLlvIY5E
zf7!zJ@(Lr8!cHv@-Yr8n3=fY@PPFIfZ!$*hI6440xr9VqE1v#d6JS?7m0xY<seIlI
zPk)UO$&#m}q$!KX{*e)C$I^AvJp83g#n7)bfpx`D`ORjA%IDoM^z)2JmJEfGj_;Xn
zucTYICS!wSx5q??3Q*}<@pM}gU{^eqUv1{8eBKRDw=g0Z^K@)1#gtAnM(wn80C1|~
z<680b+9trRcq+fz%v1Ti8=mfEM6%>5Hs+YbMoR4!b(T?T$JF%%D7hNmD7MZufp*1K
z`Q2u=%IDp%^)w@rC0k)nXHuB@GDfH!Q`gNfOX*TE^p#CuT`^RCvzejtc{dDw2_upv
zLpQ<c@7b42v8cbmIJM&`3LXI|H8r|Zoc-k{;I23;zue4O`Mev>{yZbnTh6XAZtn~h
zEpcvL6p#+qZi(~9Th^Z@*<XGaUT8U+G+Nd?DY)BiS@SK!!-wqoI?j#9dD91@Q^9e%
zG!ByvmMk0uP6vkLj8b(xIDYGLne(G8u69ayq=(;^s@u~xRR{P+6Cp_nq>Lx73wd9b
zTxjk9^0|Ho*l{FkQ!ay6=iN50xbp6McII-sXZ^qo?85H6RV|b%rnfLUo~u>ocZD2!
zb8;<8>Uq<pcQh_ZX=$bOE-VIp_x&rS_mrP};W_8{rS~>2P5<x;9ku%=I$=9v$9aIE
zCHwV!M=oOh<&Gl<Qq}jL4_Czj^=95(w2V2S$s>kawTyyU#l1ZCK7PKv?Hf3fd%ol>
zLy<kta?Ak7*#|DEdd@gms{2Ce@tQb?9Y1*1oMRON@uf!|d}GC&ui8PqXg%`an+&Y_
zuX$F{F3_3mhGQ*Sj#0Gd<}44Kx40Ws+@j_9Ipcs)s?S@-yje5suya*iuVOkt2B-0w
zwu7&RuhcAWj((z6@On<__V*LXs_R`DsY~hd(1xt8GzZg^=G28)j1}q4oa;C)4q~1+
zd<Ym4^hb;(3len~?V<%yz|=v!C9CS&P%vxJbn56=kTmV~P>1$Z1BlLCgye*_96fr>
z2xTp-reCk1^V#=Vq&v1}#DtnNZm8q2Yt(QCso~faI|x7SR?%Y7ssNe5gKpvlFR)D~
z{`J`I6G4eI6>k(Ai_{n0pzb)(S-<WCzUT*eJag;l3ngf!aW~<35Oe`CAC#OZSyo`|
zncOpW&(2a1)cpOscNg7)pPP5xd9(=@?DoL~v*^>C{_f)LvFARow0E|4=J<V;;|KN~
z-t)X;<5q3{_5;&*p1S_9b@-+eQ+wxceeU)1lgDp9up6BtxUTNkp6!?LMcAO6aZA)e
zO~-d7S&~KWQq_oYvF?*}+`4BRJ$4APlCP5t0vtHKU<by|Yb4&Nv#`;d{DKL@=k1_W
z&*yA+cL=)MI%8I9j^*z@dEms6*_-Y-e8Y{oQc!WO-4*t(ouzG!Z+#zYMl6cn-PGT3
zBJ3uZ!O*J34qR_MB~_cG!Rqja>figCzL(y=8$-aO?%6AiqLg|+wUr!Kd77}>apcCx
zcS+`uD{tK1dv)$0hw=<ik>0y;t>nPk^j=Enu{NW`cuZT2<(u&8V<uoV;jBZ?R@h0o
zNhgvXfm5_Xkw+x)Hi6bJ-39kR=F6@Io?EQLOOufz>xHnx3hGq^6~Yr3x4U_}ie0`(
z50=b8xCQbmCQLC*e%vJ(OR!jm?^e(j`~}>FTdO;!7kQLW(JJP$5qQht*QyS_C7eZM
z9;0sr2Af2uG6%CMtoXIc%>DG^9Q~No=9+XW*#({;yFiRisn>0PQ|eG{YB=5Igv#2s
zxtIpt6qyH!urLqj#uZThqOX8an^FOl#x{m3R^ZtMY-RP-dsb-d5qokqp6t3u#N3;q
z#pg$%(3UXZE}bV0NVuKL&eDbQLE}Q9qqnNqQ}VL)2gS%&YcrDi)mC71IAYbSB1o`G
zZMF4dB8ag)ADzXoQQ(bWi`)zXodR{CSQIs4So8|QBLqzqL_GE=f_sEI^kUAq-t^H8
zDO^L}w9Wv~u27uK@sSAkN+x1EbT~vwG8}aCzU3{7P|x2ldGPo{Ov&IaUCDrK+jCoG
zu&J@ts@5wOg`LG&$M%DmtCz}xNPvEswF9ft@^eg#Hw@nebLQ3qX%G>}jf8<{7(z#J
za?v%aE(;amH=<WaObBo|Wc{P8pVvS3Hl;Ls(Uju<uXV-$FN6O_BmRGo`#`z@lKx1-
z{E{9_`Tg{s`qVEnSos6Elw6H=by7X>$nb%=0RAWPWGRB9AvXG*;pQ=Lt+pGr+SOc(
z1P}MDX(`Aawy5O!*Ac6vsqQQrlIO8Vo`1ni?hHeHP{Mph4@SyUg6Ych@bE3Le_<R#
z1|Go(5_%;Y*$S9c5sZk@YqNC?1Cl%KLcMCzVPCD%ys1YGOIh};JtfI+cP+`|ktF|1
zPbwl(lrZ1agOQSyU`mq8#Y@Ltirjf)BN^vlxn{xFo1U46_iqegWIkt~St$)QhWz>*
zT{aMNP+0O_LPd4{w!W6PwVVjIwS?hq_M%474B$p7J8{%qf>j_-iS#fGMElvWr(T5b
zirqNd<$I$oDcAHpsU?r84W${wO^9~vujQyd<n~B8B$dtiT=`tQrnM>ljw7;7C89p>
zZ21(@kB4Z_)9vLOm<w_?Wv#m!eh;;p-Cl|H7WOCAu%S_E<+ZfYxzuXgjn&h$qMx6h
z72)#fSu*{ydp+UCYWQpTWB8luHZSL|>E&X#mmgvXsbWnuYSlljiAqWzTB%fuU)xjp
zR>oPE%C|5?mQ-qNyY?YQOqY}oGDL4lNq25!kfLqZ+*Ll_Jkj>~R6Ev_ZPzwm@3?Md
z6lYQLIc@Prs$cvA<E_i)M;Ic7%^H3GZ(G{v-1i5@X1J1BTq^?pA2+2-z<)7B3ISRj
zyK550$GG?d)$#G_k0)K6j6ITerb%-UxNR?A%n(_UFoc!+@tr1co7ib$l(Etk#V|wk
zmJY2h`?(f9=zfc)t*W1m#=TL?bEUAynPy!s9bt$RE@||*n_Jpgt;bEN1sr3zRs_s(
zQ@R8c7$SuLtsd7s2{W+;9Jp2_ypR#nCE*7dB1;m6#-`)zXa{kd=y9)OtaRyk4MX&n
zj?8-8#-q*WkMaX0-(O)Wb@}oEhDhOyMjv}$OB<{9G364*rs6K)liZXp0Uu|G6auvR
zSob8%#LNV)l??uo5z-~$^9+$C3Fu>3qQhH?ppQ)?`q=+4R=RZjH$(K6j?DVlrYL)+
z4*88eQ4%(u&~4muIYXpyMx&Qq+S0~qy=;<&i*+IK{bFXgDP7S{GDHdiTD`1$5~gD$
z;9AMxRz^sdgkubmEeTUG5^$|ZC^161Bv=g5TM{zsV4GEXTIMg}j}&hnW=eH=^CJwA
z!W)e)_TrW{R_$WEPzYTtzEbFCxG7x%-pmjw1ZZ`!?nzK<od(7di=BRz5z-~$ml+~k
z5~gBOz_n7NPccHeBz%G)dP_oPUF`hC%#rM6O-~hvzRvXOa_FlJk-{O3j`ro2Had4S
zS{=D}s{IVc<|ih-anr&xz)2<y(6u7{q9(wO=?6KZa5|b%kY8<{QIOB|XB4)VH!vb8
z9M$UiYvJkC-u65lXN*FvXy$3^7<pe4U{^eqUv1{8eBKRDcQYbc^K>%h<l-M-jN0=w
zHCi*90K4L;{Ax2#<@0WMdKV+oTb{1bHg*P!mh!Ytn5Bcsr9A5?za5=cc%UU&mSlg&
zr95{;#zIXMs*;;izlKb>EbNPF)h$ECN22zSDxnhgJgQQ>w{AK%EgHcJhGk>XHjTPZ
z%P4#c%N7n}@k8CK`lzF#We}*rN;L7vgRf5(GE$|MDl0&dyFwyo9CE1^OO#@d7<p7W
z%8Gd^D6!_S>LEapwht^s#WB^$)0A_;tu9)gAJz`xCa}%~VS$3EnD-rfz7!Z0lqQ)A
z7M0SAYIpTvQCNxcaFje0UCXZm=>Uz^tKq^98|6&6MM-}FCH(|$J64iRRqF{xP?i2M
z>x?ZHnz(M!M`@EsZ&+Sc;D{o#Lfys;lLb1!4HTYXCY{z8gmR#Q|ESKhP;%`8_#Ana
zvvT}ym!paZev%EGUNE%d$n!h&@W!w_hIH?K;||$BsXnr~YNNdqetQ!(Nhkb_@Wj#8
znwGJT?(2#f?`isc((X0Q&dBg+IpT-Jz1i)^=lX7E#}V14vZ5IlTzr(dAbnmh3NF6O
zTkS!?MPu-GD*ct}8Qm%E`xf(EuPbgZ{|7@zNhLK_r#9o?TH5G5F4=fn-SWvzNIKL5
zMamPNlpLTW(Q4{QVw&?phDaerYe3K~G4V`{Ja$ES<k-OJnT(#UY<4n4mIO5odLx>e
zo+>#VWF&RTdJaSMmaOy;HCB1?^%h<2s95Ew@}iCD=}G0qy4!iukuDWmE6gNaww4(p
zg{>MHhWVB@I``zu?h+9wVqp=!nl*&isFt2H?l^JaI@!BBzobZhDSup-<QFkS3dveM
zzDuG{-HFAuCn;K2<P!CFV8t|6Vw}JL1MxRlGCX{j>6cKP6}Z(A<A&V9-0-lm^9Drk
zSkxj`d>nM0dL?h0STBm<2S6ORAeL&=%1<<DnOGoFaOp32Dp*XxyocBw)#!AOrPcVy
ztBMu0@+DdrkM&h^#xc`dcCfWU&S=JjK2|j=z}FOI@Hbhmhc&wK*Qoh9u|x&+w?5Vi
zV-yi#kswv#ZW?n|5$h^wad=3g7-1AJ_OVQi2*j8QOIhj5v={;ldZ|6Lx@-u9F;K$D
z3%}Y}6^y00XeJz}7-q3zSJ5VxW)_V+)`k&g(YNNYD&DKoikEO@aKR3isgV*bF~hJ5
ztIL-`DR~=UM$DkC0x)*6s|BZCgmgr6ca<PK<7O8F!V&S{APLuXjO**RLvM|U5e<F$
zL>>G<;|DR!#79HHiFI;`Z&uJ$y@C&6l{40xRH?X_c9{s5y%n&Ir%t>gVzJU*44Ty1
zwWpoLhgQ?Cn6-ElcZ9I(B!k`y<6<Q}ls&Wc2h|(j*3@KYGZ|G5zc319B|a3$p4x%_
za#rup>OC#@OwoI#c2aZXqLhwKU4(saU`5T)2{hiR(?k2=0pM9-DWk1x$Q}vfi8PBc
z_rlKHX?ydqn&#S#%14>8Ju@nwY-+MAqayJkqtdwdv09Yw`KXjsVxq64A7R3KtE3&f
zhpaX(wtT$%(#3VRt79^7sd({U+_-Mk_DzOJ;e|$^`i+)0ItQx52a$MPcI)uBaxJ@9
zgf5q#&mp}@e&4e=Z0!a_vE$-x-40u~GDHeHw1U_!8FK3M;5p}9Q@ge~G9WL7Enz5#
z5-ashL}8SYriH`ckYBK>2#_d_5zNA_ELIp|Nk$@63`|5eh<C(mQGOWVpNsXbu1hU>
zZY5`&pa4xmQR5W&B?3^2RH87=+JDXN+O@!U)Hr~(yCNff&@p}AF8KRtM;M^NsKild
z*Q5wUY3(yYtgsE$C_>i~qIbWJu+>BqX;$)e1jZhMw-SQZqB|djab^wMtQa-h$}wgq
zER*Qi--ZYk*;g%Wc0gwPAiiVPEF|)=ggZ>e<C|drLRIXrK$&*%a1PX9!-TNu93pyz
zvrPn2YeHfnO-&>eu!;_$q?j-8OIRFUi{j8_3aEXB{*6JYY8Q-yVM4zR6(ul4=|d<w
zid)<2LmbB#!-W=+0`Lw};YdLEFCt{g^&E^>653&fn6WB_p}QsLCPu(ZOhHiO{#>(W
zhF7GbM_g>_Q&kjRZ+dyFP`BF9Mru!6T_l;Gj{xEvk7N*zgrLoFLPG&yLkkB3m741<
zj2NDsk2EHay@(d9K%p}{d;^9Cfj>Ol9w9lLAhyhz6^u=cs|18`Gqi8{Dw@9uRR%+^
z@b@A$X!s?!CcEz<HYBppn}P3~V=lU0*ewIMR>P<!CjJwFN1G%a5*mDxdz9dc7(`)A
zM=uBo(q=3)%hRrV$-X0A47_?_A^arlLKU67SsD0s(=q2PXT(^r@}}fjU>01)5lDgG
z)XmFv+BgAYK<t$iZFnSyrUF%3Kq@pU#DOOB2u$r^)ei5FW+(zV25dQnks(cYVF9tX
zOxvviMr=eC8b|Dxh5m`bIZd{~w$YcQXvrdoA=XapDKy}jOR+r!8UwIma{MK(mG9WY
zo?tas+86?A6*E|Atf$?bRvMemRGj^O%`9Ny>_!@|lX(+MWOLN$<-)Z*fVCjQ(Kkd8
z_0I1a^B)U1Bn2|~D(?nKa-Dx;Ju@FcVtAt%e5iW{m#+g%`8o9G0RB)WF_oi9Ogaeq
zj{f2S%3t$1bd<f3VX42o)0W{<fAN3}KR9Z9cZL+b5%iDtm%-YCetm!OfS~_s)cBqa
zzIvkzKi^*&XzRj<`ilp2;h|9s2ObY$YHtMhiT*N7TX28fUpyeVb)!Znmd&AVjf0e-
zFz}aQpuAOq%Gb2frGO+#zCREp-(`R#Lc9mG#}?T3H#WMUjemPFzUagg{lIPACAwJ^
zj*R;fxLwp=JfKEzg42Kej>LK+&8PL3i`qtEqQ7`Rnm_yaAG7vGq=)*;N^OzektH4-
z$C}0dV$w13FXqLrW*?GeUio0Byt2$C$Mefev<o@%#i%|&j*4@9&FSKsWs0{YZ6BqS
zD7jCC%CvDT>Y~GJ|00T$1DQJ>qqJh(5vdwV8dIt#SVk6Ceb;b38?$Dj21Dd~FujI)
zaAd=(P4yj=Uu;R%=3S?lNVAoY?M3npslaf8KbhA=(;+rgF+pkPc4;Y@?3_a$kM{hu
z3kw!ft_AGlX?jIt9!WXc(w*?Bm<7{wt9F=p$4=%xXQ#(=G{msonhh33Ox_=RBobvU
z^!PQqfUV!rbeKEDw(Ak9;RnNN4wKXPVUc%RG7<@0CYMSuD6>{Z-H5T%BxZX>I=O&8
z802Y3_FbqVLH>?@6xb|U^Hw0X@<aZ53BMA0*dl$-RqvqS1i5<fbeBxYO+faKAw4fX
zLiul0<bWXsRI-OzKV;fs+4%&KQO;h;HY7Hlk9Wfyd`N?QzlL%jQ{ufB8@yhBTbmdz
zOvVRD=r2;Hk~7uO42_t)5xLOj{4~8?uNgjeOdc(`?gEsG*cIhy(HXH8B0&ef57xVQ
zL#pES7OgDsDedmq#`W$JvbMKc0S=46L(#?V6hiW#Y$0tYD)^A`3>w<tGzv-FWDYe_
zhubYD*pEu7_PI%HcQ4Yi$yEqNyXLr1f6SwhI*5r15RM`n6#4|1#sPT{L^zX$8U%E4
zm<hqFFj1b$sz_DElm$8$?Yj$e&~}<4i4t2f$u%`MakyG!&I}JnD=bhP7R@Ay4CZTx
z&cTQQ|55$rPHq49vVM4-7e|eX0mUoY%Y~U6n>x(Ew8x?siRBvd!f)s=*R-Yg&i>+o
zfb`$KFP&g-wB*<ND;;eu`NRI=0WJB@@7r*pH=6P{{gsfmraanTJfJD>`>qSDdL!X;
zp4^9ol=NWrytah5_7@LG_%q*|A!u*Z;Hv(zUt0~v`ilqDU;x2-C!Qtp+uPa)^)la`
z_Or@=-=IxToBX%xM!@?YzfZ;y+OpXw#~V@_>Ve|_oxS3JX-G9H<+6qVeO?=%F7p2p
zpDxVpVwQMxOfSB)znH9^UVM;ai{7ReJAS#h70tYz;=0@8yD7d$o3$`LBGvjXhDhOs
zMn&__wY1T>qIpJEG~?*Q3JM*Oqro9TbEx3NNugp33k%tmIeQM9Phf3R1sN4_glJgb
zOy>YTRilNNS@~T?u-$Ti&FV|iu*4rJEAU56&1fwFcrcpUlsJe?#0E)_YFz#<O-x5+
z5~BMsqmv>!t?KJ8g?DOlKvZ9!z^TL|kP>B-6#tCF$jDVGVQdvgjH_fhz{ph)q1Y}&
zq#ltE<>WQec$C%?iCWYlhBcPDscbqzDl9mlgm&G>CMnqWqhca48)gwHnHnzML&7|k
zHl&qz_@?33kZni4md<qZZ~&*so@3P_9R`c}FOgoPrd_Nw#Mv8kLZ*0y&;{jH!#8oZ
z97v@HVM-YVoT%|8JVYU5u}w8Atw>ab+DM06zGE(7X$(r8u}YAZ9EL4o!xP{bE|)C0
z`6Riq+;y~(NxK|8Ato|}VbyQ<HNRm-HG;0Vs9u$o@0_@e=V{yF5Tz02WX7mq?d^8n
zxYZ7##9{bM<XUkkm2V{;QyTZVE^?HKrzcD&Fpi>nI{B1DYuWSF=qaSAS$V5Xj!-C<
z4iU!LevVzRsk~Vzu&7d|nFekv%)yGRv9K-r>Ck`W-mP6x+qrJwL6p#bHEL8Cp`ox3
z$Lys1S=ylFTFk38`?@ru&+PO*N;YKQ9YMiH42hwq=#Ecg0kpKTk^&;|)BvsCJZpAj
z5&ie8*@@|vR1a2g{W_i}YpbPZ9G~ou+T-P``-=yHtMd~Ex;Nb0-(Re?xp%m~cz}Bs
zHoLOkFz|GL5!YtmLVxi91Gk0@WT)R7{=Kljcx&_T75&9S@sGo!-tg}&{l#0Gf4|UQ
zJYxKlfn;y^_u>BHt<Aqr^%oED??)aVA$M=Y_WAy@YxP0?mw6bF<sg6i1|H;JX(a2{
zI9JoWZ6>Wa<dP6ksfIy+o8h1w24XrNoBYgUOdXh_{D$m4#@cdtDYx%bqf2V-K+&xD
z#xZR9gW)?B$K+8&CWhKkb&yFlZ%9ED<QiL9GbU<p#K3=IkFkvEHd!qcj%{r^$UQc?
zZ7TFaqj*Cs>cnzeYysp4pm*oIXNwiHDm6fBXHnW^RZ;Kem_=Z?Q`B_f?SiOlL%KwO
z06xH#=|_cN^ghx8C^B-astYhbz=@%>8rQ1fXl$Pg!f1xu!5%&p;1Thx;NG`vV1rm}
z%^Dlw6TfH-CiiNBBxx5`Y+^+J=3Nhq+KX+O7<gvUrZf&dA>|NN1t~U;ubGV%q&%qP
z4XfcxBT;dj$fSh1o3LWZ$b%2qbO@D|f-~WZw0A6xwNbB!#0ioy;9e*dBz;grh@Gs6
z<pr!*u8Hc!h4AMTCIjeR?C6IAyhu1{OvhyLLDM&lTg)>kRg`?M*L>u9csX|)-k&I$
zORfAiDNR-nQ&V%uIw6@xWnRguqqq=68pp*RRn@2{FRFCwLU#++BG?VSV^uG75XFB=
zX`v_%_TI(nWbAY+_)8Uw7OAF?Rn?mJM3G`Bu_E@N3V#q%OY#=ovFgr4^HotJ^`%g>
zA}v>Q^VrSGm5NPOk`ApV4<%bd@^G1z9aFd`v69+2LfyN`LXg~{2;z8a2;Hak&QMsf
z6PaA&P5R$b_JtMIW_%khUL$QK?Mu9$h=d?5MyI?_i>QDJ(8q=gux4~c@(Q=hlHDh=
zC$yfMHc>8$I?x*U$yJN(Wl0HeRlgXTA<}b*$tNZY<wsF%SY0b`>HDPE=*DDXVr7d*
zVOofzzH!{m8Do;CKCB69A+e)0lBl8;+MNfjXxk7yoz`VkNBwTc6SL^%nIEvZoU0{+
ztZV?AZo#)S1Fqf|rbZMlhtJlM%G+?MZbqw%)WJa1s7U3FQ7W%1PaO-7WJ{bBNlb`u
zkqS(8)-tu1e5fWdF4d;AP3}(OAt-uM+t8sC*7~WO5p68jBoFOa{8|0Q0~ye96ylzZ
zv$~D5dZT5>`zu{-ExV_`ctFdl(6X~ZsqhpT1*N^w&{}^rLt8^1=r117(0iewXCwHG
zBlzAZ>8tyz9NJ3yGyTN_N_rEN^z3k_C&QiIsNM(qD{^hs`<?#c0o6MQ)kBr+*|A}m
z)?4e?us3S=nf{7<^_=rLj6iR5&K<v;ULV;REIRh9^^}8juy)6OePorx6*?aK^+QpO
zpE|Hc9sAYTHavXe86<F#U&A)^H(&*@?O@}?upFqgEna4P-u7goAX$HNaceVjJl^gn
zH7dWjsadVM#TQ33Nt|d@-X2$}B<Tzfi%oYNk!`l))2Rj6pj&YK8IIw?v1W3ALXitb
zx(>&6#ZMT+L;>>W%e<R(sg5GIunOxS_DsC#){{6m0106%HOGlz#@Nh+abtDfp*&=~
z9bZ=G1#Wcu!`{8qhFB{rHep0^k<yakVP%v`j4BgTu)*ww>Cnn=6d_^(uRSVS3eQx-
zQDgIt6+SVJA3ciQ8BM1ycK62zEQ3~zFCeiP15BWbd<dNhK+V$jt%>HuVQV!LCmuAN
z-xh6z8z<R_SQn|zS+-})cGWF+b<-_T@w?;5OIz*CrpEc0_o2M^Y#xd4h4_A*m)tfs
zVjLLTy!p0q{5y_+C-CpYW@$B2e-Cj)z2H8e=;zbzMPnlp`SW#i9(#ALuOf}UlH6}{
znn(<H@O&AXYuE1^&kK7OR_celtkOK9JY@;!#vh}LFIk@GF0@|;c2mF<3U-l5_6E7H
z>V8RQAHSfadAMt7zM@5%<ry@>TW#eJhQ}!GH*SLrZ>a~hdaz%}Gxh@sc*M93!sQom
z-R1?-0K7OsFV=8s2U=2^kj1B1sqWHOs`AZvH&vxxl~5_=ei|2&-5!O{6<f_#v9qjM
zdg?tZbn;gC8>XJ(zg>;rcHKjJ&XzW#n;i}XQ`%CxtTX_>cb*hUX-I5oP`Xe)Xk3D6
zSe%k_0vnlF*hoS(8l63r`nyt`d%IGct-$Z6BE_jjkU%lV7=8Hwj;O(|F9x>D#tbud
zGsi>`)>lL)))+^R9>dff3YV#_own(w!!J=|?h6vkJod#1#Gc+_*P7^6Y|C87{xD|Y
z-a1aUftM;+XqsjXG3O>uO`0<c0nHrIF4WjU8e8Jf<Pi>saqRnWnoSAEch)?V+k4Kk
zft@I*cG{+?G>VC6GYd@DV7p}Fb_)do(a|>KO&N1?z6v?o(;cmF5C8rmTItRMUFi<F
zkT<sIPWcxuR7NM#@bU**p2bw`Im3rgb|2KMG$#wyq29(}Tex?L*bpextWdFvm_f$$
z9Mu%$!q~U;DC_LCkF7mv=O1e`imyM>6<@y$zP>r)>ub1Arh82p4ojFH(}O9$A3}+J
z>J<#uSzHH)DXdN$2`sxc6`-J3muz%nDnNk5Y8DQ4!ge!p6WJ=U1CO~$DYPE6q~!L&
zuI2XAk=#DO%v@b=66Sq+Fj8(3Ovz2TEa{#~kt=F!M14qwHiKU&Ux5>;8$;L`ea=3!
zQW`2>iTm<Z_(OB3?@j7NTbF<>iNdpEC*kx91srjKt+7)2>1_uvrb6u!=Fb;WMyY?<
z#N^~Cb!u&&lIpNlaHRu(*Mwe2dCCn)-x{fbq^)_hE1!$kb{sjUo==yZ?B23XC6hiM
zV|yMc#6uWHBQE+Db3qQz*LsZY*;dbT;5)*B9vnIxhBKur?qAtix)MQtGzgGZHtea}
z%g?_|=G|mxNg55ElZ5YXX`}NZ=uJyD%H*(nBp21ho+xV;aI>o|1C+$v%MdBVXe|rw
zmYl)7>pJ&JD<0h`G8&AOE*TFqM3!W1GQGm-S=uU5lClv`!Uu|)|HZiJQu8wmktsDL
z6cd~Nk~uG_*`%H*a(<1G(<SFu86rz^*4ONNO-V_3qlox3MoE{5PccNMM7a4F5#mM>
z@nuFymxwPiM3zJhiEdJmvKXX3Qi#odt-o9vR9!mGWr!^4*l>VKhLOFD(lvaqL~<D;
zrc26G7^1hNbc{IV^x3IZ^r7kWd3i{9n{crwhlFpw-f`W^=yYx@FL@k)q}Y6jnWoF;
zgA9?v=1nv$GU`}!0n&d+{l&a~`72?R{G66HIy>#bslBIjdnG07r}pAT(eQk3O_zqd
z86t%SEnnR|72~IK`xF)9xKUIrGETZw_zaOP6_clPdlVIuxKUKRl5x_d;^hpHDHW6Z
zPUohWiphPrQB=I0anhyYtqhSV71I-ODyAnARQv|xq)Wwz7$Q?DCZJI>EQ^Ur+$dRm
zhH=uR;!hbOTPh$6Wuqo05;p29jFT=EUt)+%shFIKQ!zP}pkio8xADaQL-dvkEvNkK
z78Z7Yit$$ENk=K+F4eT<xzgCHnPy#CKaC+$xTN7zuWV^!HJ`fYbdL8kTR*i2H;RVq
zxHVlGu3?B28nk?B*Hn<cD%$~ljoXg97$;pSPBKKMR7{N}WHFYIh08eUQsFQ}rc_MM
zoX$<l4zhl71~*EhUdA}-Qt=Xo$drm{L}`kO>G1>=Kg~GlQt?KH$drnS>3F}Gm`?PI
zUtyedsrUdxWJ^Utqb8CX^-0D_mx_-wM5a_sBEA_{{bCX~s$cvg<D^T)=NY26RAlB;
zHy*{-f9B{kk4@DB#g~m&bQ^07GDHerG`#4#mNq(j(e+c)@gW`EC=v{AOjop*Geimr
zT28cEA_i#~p%fIuh`55zFiyHuOfp24RFEiQ@eJK4QQXQX=@M~_Au=Tb%vXbJaifSR
zF-p2bSPYRR5rfk(+iEB@J(CzjJjgidQgJ^+WJ<+e1OkeRz2gZgev)z0rQ)>=ku4Pn
zEn`%SC8+pC#z~io_b^1JR7@p;vZ-WH_WO*JE)~Da5WS@$GZ(t)PP<UA#(ik)iLw}9
zWIA;@^96=T;f#h8{ri?SR&$~gv50$$ZWIaU8r{YL-){MQ)3mad6YZ9W!D&R?s-sO$
zCpy|wx~5_qL!=0zFI21t_s2L?6*o#0V~monHsu&1TOwv+Ie{s<QA8YOlyr$W#1L5$
zF*uFPhKgsVrxNju$vEj!@qC8Jl!}Q&o_!*jXJ2NVbg5Wmh-|6oG!Ff7#z~ioS29GV
zR8W>F*15-Uqx!`=87EyT-p&xcr6MyYx;b*9V-q|QjXzSn`2(g@mp30}h!oyvxY3Wa
zw6U5Sor;a(rszhI@CY}iOTynWL<$L74yaoq2B%>#)FkQjzQl<7zZfT7D!$1OSyDk$
z)GCK4Zj>l4zOvhhW-CKvN<^oz=rE(CD~YQaB1<AfSf(a=5thX#daq}kbg8(GAu^?6
zFHE<hV((ajiXUK{bg8(DA+n_cQ%)*G-n%CeB72OJE)_0AWJ<*pvSlhTLyR0xH2o;!
zq)Wxi7^1gSWadWCPpol~gNLT4%KW{X>DA@XI~XE`LmJNXZ7prA=1e<S@ZaIabV>Lv
zhDaen%b9jd#Nc#d=3zQH^YB-UlP(pXWr!@PSU<HdKAfW)rA`0JDCrXMHHOHP2uvg@
z(<g2e5$9diZA7q<A+jVwj7gOn!k9GfOrOd)>59T&h)k)VF{w(jV@w*aV0#whq)WvN
zLu5(?rKeS(jP!JT>gfdIq)WxE43Q}nl-N~NAh8>#Vu5kerJ}?Ty`^G}&a^XFw1-Y>
z5pX(~+(Ty*tvP$~xiA~s%m2}`<}8`-?6`-{+oKhDYPDQ!#qo7=1>@9yv3E*h@uFA5
z`CGJ2xLC*@E;rU##`t=cP$<*nGR9|x%NW=6JsEjCrZ$vOeoGUgNv1YeHd`c%9Y@|5
zu~Smnyr5A&?`A=xY*UFS!v&51hq)j-R4*1Zo^ACk2QuS8b_*IWf2xe*WM@el4V?vz
zm$tOgd7u(oz8I~hi9J!)Y=WEBb)|b4B83>OAxpR9tg@i-2qUFS#?1_oB^i1P8Vih@
zE;TP;h)k(b%NM1NsV7RuevpyVCFcx7WJylP1&yy^lyr&sF^0&Lh)xR{-_0oL67dd(
z$dZWF7Bv12W2H;SZ!tucbnpVk^aYK7#fa&W@>z!HEh*`v3mQ~>F&tD}vdYJs<KNGR
zU~QawwhI~upC(<fE}PdeL<*ZVvWx%I!mZ9ud;PwtSO!AeC^LFFBcw~hr3{fmf|jf9
zmWYAzNzQ1AYem8&Bcw~hI74Jf!nzsGJ<z2h;25K$OTcp(B1-}W#-}(RC$1F<79*rf
zLY^TqC1GqYPn8bPwIbnuMo5>0`xqii64p&AHNvGL;I)j7E&;D*h%5;h8ry?vqe+=T
z8N@B_6dmtjtaRyk7ei!A$6nCE>NGgEcPdd#_`8giE*&3ch~Cnn<%Ivc)vf$9{H)x-
zPwpD{)s|<m9A3*p#VAB+d9I}W1*Tb-OMlN0DO}R<q5rp~ja7W;3@ts6s97&=6bawv
z#&k*e7DJ?vpyflmC1PNFjC-B9RwQhDdbg3x6Br^(5(Z~bo4`b@pBYP7jvV8pD~M+@
zM3z(xj8CxH5Z6i&hZrGU5)Lv%rX-9_aDNz}YemBI86jN~?q-NANf@N*Af-!~4vKrB
zMaD^&3ZEgeq+;DZj>+g!iQ$!ujxGT&XNW8b7#bVL6e-JL5Vr|Iyq&SqrQ@v((OWt)
zbE9gBV&rku10~;&GL^b~`3OU#@I}LM{&q_nt2j;+RkCc?iyKA4-*RKRBz%q`Qb^Ep
zoZS*JFgB&^HC-zbzR3vblJE_N$dZKh)8r~;03&V`5nG?pt@q!;5Lps|pe2?Bz_lXb
zYDP#`4o_o<Oi36U=c@Ptx>h7y#|Y_?a1BFbNy55G-UEX!6#;iKI=Td$WQa@&nBktV
z4ws4mm(kHBz+s3i2^bn1Ls=-Bq(R&!MxHNYtaRyk2}AUjj?7$UY(-*Z<YG^hgx|q*
z>T>3743WYa4X63DEp4pgH1|!%Mw{YBk?>pGm@Wyw&JZaiXgSSpiGZ=@M3uN!Bz%?;
z(k0>343Q-XsTtd^F*>>gJjxJR5`auO+iG#GWU%q-ZhihBLu5+A*ksIM;#!elFhaUg
zxSS!fBw^h&?{`6$ihvnLN0)#}hRBkDfw4X8F~zka;Z{aSmxN;sktGR3V~JHQNY*Bo
zwUihuT{<j==q(+Yxy<<T!^mgGA1N6>%#`Z#=0_MJg*O^b^TjP~batA9`>_6i)pY&7
z$wU_6XSg+88s5whDKuz#&2Fg}7~jKjow!zF_*F(omxNzth%8B1KQj}PgSb&de2P)h
zCE^ndktGqB-(;gLt`!MiWQ24{_yR*@O2XJ)6@byTBH`Sf-8%noe@94IH_dC`=u#2z
zl&%Tb#t_-cAk}k@F+#d(lw*i2Nr30XN=i=l!*lMzof5@i#!8otLk!VdIx_Q|P0I}<
z=h^gBaj4An>T+nFAyPP`;XjKlZLH!yXA&!wW|AwFUc{~G((nL7q|l({Kf9#@K9f~d
zTq`lWff3Rr;U^d(OA^-2@PLOd6#@U7(a|N~7Z@T-0tUwS#$<qNMZzC2Lb@b;lp!)D
zVQeDiJ8`W@c!UwsCE;%wB1;kmrxNk-R5Bj^FUCoiif=MRrc~^U*9uW7V?1Gd@vd&;
zn5_(vB^5)MA&+f`Gl<*7emKL7m98|dW{BR>u|^--87x|R*jlff4%Tk%;jgqT4@<6i
z>A3dr10B{Lo?1px)8T+l%k}GC&9mv`bpuDS&)GQOHlU+VaV1Yf4Nt+v8Mh8jf%oaP
z_^QPCA*jXy9KwmSbeC{M(<q&qi*r<6oFaU09mg-?p!qqpO(*pFe%+#vmrTd07woD$
zF&;<!TH!IC_>4uUxW>FVx6?7}zMaPz@%cJ%V~p%-!Sy_G5HJo0#ud))6{iD>L$z`C
z?k!czu+G$QqPO@4qfjeZRkwowtAvN37>5Y^#v>2D5$7x85bj4Fd=nny#8A%)Je&$_
z6=TN$7waA|EXE_8_gyv2Ah0U6fO$zLdE->ykdu}?O*R+JN}d(McT@<<!JT5$qGybD
z9QpN5ed}M)w+>+OE;Nr0boX(T`<&-igf0Xo>9Co17p)N^k29@}aF;A`u(a6e109l%
zkd`1&&Xmp;_M)6|v%7?YZgGHo9Nb4=^`ZWE0SC8QMhB^*X9+bAdte0zS1(%V0Ky35
zJ-0qz3XJ@+)K=2rB^&tp74%Wd8})JIby4)bl0A=pI(jco#>T1Itksf|x{p(L{gUZf
z^R@5<<*=a_?%T8H*!4Ko9!F=>G1^O}FF%Hc(dmo{!>YN3l0RZBm29vnBuKPZsXKvP
z!!zHldj)&|rUVBso4yr9txvVK?p9&5WbX?<F&=#)XPhKM<k#otEg$EF3(nx^?^u5&
z4)_7i(Haq7G9<mQCClPiVGl<N7jQbb=Q<Ym>9fNBeiXj4C39WdlHcrLOB#P5)D)-Q
zb8jVg=tq5R6npC#rs37AqW{pr_c-<5T1NkYH;T@^mK|C2CenL4XWnsK9Mw4w*JxJI
ztw?6>QrHW`d!q*t17Ht%qvrbP&(x*EgS1UgSTML*%W2-9fWL~qkN!>8L+t@5d&u_`
zNlNZP*K+^uPIBKV_lB6>Z`|RNBOwvA><2L%o;ts?w5>FxaCuP^F3C+Go)&Hb(Yaw+
zIHmWp`t-w1pH9ZB&CA-NZ5@6m;)KMzd0PkhT)d`vPPJ?^vuz!oLd*N2bE@CUT##dx
zwQlS1J=AIsHVSBLysZv*ZM#*epX!0q!9Qp2>xFaM%YViYQeH{D)!A9#Q!Q<D&OXLx
z9y?A(ew7>6CFjcwkwT7EBC=b8nzF&G<%l;7cbmyr&k$LXwf?vp@KLJajZ(EM7$sfV
zT*eT+B_chs5L<TqiB%M%**-_hj@y{Co-8}|uc=*o*qC=|ly73dV`IdBHIf1tmy3Z0
z1>;2*0|P|rj$2r;im}noVT??WWk8gJ5PaUmSm#8^tvf~Iu#xvHb3ugih;fSU(skGD
zu3d|7(Lcg)>@4rzrWSfN2Jo9SDB@ZP_!x_8H32t<Fodb#2)puiXuJFz`f~t(i2XVI
z8#%Av$CgruqejWkHRhW+@W_L&?*|WJ30?X9(&ZshNp7rXHUYcvM)BhX{pCjzlMZrm
zv&5q#M_$-pOgi@K(kQg)5F?OcQ8cE`4z{$>(RrQvGNljajAggp)C)%pTO<@PVsX8I
zGI5l_!yw&7+D@eZEc9;-=?#qj@2%T~1q{(~Ny7}0-lNO_2Gve95=YXaXw^(F5Sa>?
z0<sC{?P73aR+n?e!KnX}k$@4w*;f@YLZO$$sGf2#c}l)j=Pjf;@D@IS;Y$@Qn?-!y
zo{NyKx&hLCZq4%W6fXp}V+Le&Kx(Tk8WH3V4`bv+W`}^A4G+FsK;9Cm&HxD0Tj4G6
zv_Tojq12@7BcK@`#%HR20F*|Iz*{E1A$3s%w#X78AGBnOXO>?uYm~{6d_*IHUz9+>
zwlOt<EeQs66Hdt4F2>*Y%{j|i=5HXwu}lqAqMQscl`RRg?0Fb(hGRF3GENPh6+Gla
z?2XyeH*xQY?CBi1B$p944)sPsq=V{JbIPW_tqJ{P?Asg@L}Rw!kH&10cHzt3__)?#
zMX^L$I>6P9>>HRzDMnT+r0&+to;z^PImcaTQ>iy&sskHBB2Ger1r{PSI3{u}6z3Qs
zfk|miJE)84j+`+(d_6@&l<VQw+FC?<7#XW-xefyads4(fg7YvQE@`~yTh1Ik752y{
zgLR--L=M!g!d}CogqDAJ*hap1-gJhC+hB;MU{_(z%*fIrflP@aau2Y(70e8wy_gya
z7ZY08PF7L085+e361IN5P$IX2d^V<y1nmn*ivf8&A#+L3<m;HF0z&xB6j7#M0Y1VA
znm`8j5i?h8RFT;-a55An@tG9aPG}dR-p)P@Id)7ZRSV0Q2=Eqdk-HAvSj}~C+xn$2
z3th3`J#8YUjYl0GHWyvHD2c10ag`S@S!OUite60@kg+GGgGNNCg_4C7yKs+|NG(ah
zo5HigVJy&_?3$SPFyMw`O;Cynw*s^+^fBO%Y1q)vqK!x7+vEN`ofk2e-pr6NVn-}`
z-DB(uF^m0C?#}Hj_FaAIGSY<it$k(TKG{TIS2kU8N7(dj?Yp@)GdD!$KoZaQ3faHw
z6|$#(0FL*p)j@2S-qRj~)zOyu6{}b$sSB&Vr*!0Bd!-|dkA3f0M#Uc4W%VPPk1#_s
zjA$C~P~ix<%_wf6aSM&sEFN_s2nYuxupVM)B?nIADA+R@MGT_~frDWn&w@wZ5)l$Y
zq{#M&rYacd*c94WDw+{9Rw9U!0YQjSE}T~>J<3u1+u0{PuJM*V>$?SArNK{mW;glZ
zFJm`wIRtr91X<5_^l~Cuaxs6Uw_4wE1R-~Qd`?T3D|fx&)SU+rr&EfP6t=RApcw;p
z3x!W2(;^ZcdCRN^_FUaDYG%QnvkOR#p#G^y`IkFR93cNlg-UpE_~5al#`vD`iR+}^
z^s9L#=jX5lR?Da&b1G!^Y=%gYnMQ(Ze@h#k6I>O_b6j0PSnYWj+ECHiVZ*FoI7Oo{
zbBV?Q?i>wqV6^<274lj5RvQzwsC_ZLB8dp{LB=U81Y>q18B~SP!kV~nFc67|hC9@@
zXOU2aQF0M~qsT?1kC4a8V*-psh$z0W5Kgia4v7}$%_^#FJWLpejU#!EBssLA<WkWJ
z%_N6|jOeRqu4rL^P(h^{rNcsn0^Yy`gC!IgQ638BROG?vA^ne0N7cf3#y2rqMBc1m
zp)8Kt_JBTed!8sSLXi(12z@1TFz%D88b_KlOpR@akpORHd@S}j#j<P{yl;sMNUC-6
zp;E>dG&R-HfRv-u$H$`7hs1-z<Wq-5<at&ZdZP?VLXye)fz2H((-d=|m8$91V9cyL
z=Ny{O64~$2O2l$CG8dwHK-^|q#vN4~iDu(OfI%}x*QjcB>cV`qvDG=_kg)e*_ato>
zi4$0KbW&u}$>1OnRH6Qfeu_s?-o5D~)S;2A3Dfn-l#Q@EC98sVfIw(o>Q&0WaIQwA
z$uT=us8eP>tgWRMDWy@RluPrlQQ8#BGYt<HOxRU~Z$9v|kW0cxU?j<`5J6$ZBT-SR
zgAU|zV1HY(RzR(0xu{-J0e#?}!2{E;Ag3NC`hBFfQ4$3sh9|<3;)lfS1gf8fRYX}L
zQHtz|gZvNCLaB1(r@%f=f6?PG|3#T3!nMh=DY>Fv%5#?}QAd*vNW|L<mh!l{wa-cZ
z2rKw%9#nHq@?)4qdQsG<81oN>Y8<n<@@Lr%lO;u4i4ie%mu9!+x$4)i*Cv7{P@62T
z#6}?dx^Te%c<Wl)%sW5P1tsg6o+`$@yJw7pGp7u4zdO4&BQ8i91<8r#$BGrd-ZNHw
z{rk`UbP`LVI-=w3_b0VQuqIoiT@Ua#S>n-=BaikMla2#TF<g3p5y&w9Y5eH08r>{W
z(gs5nq;)W6s2Vq8yky%{)lbWNWFSD!-VZ#R@;5YmDA{=t1j!gi&Yi<4i$Df70vNOg
zHbydJC~Psl2`A=JlT3q}Mcb?Ut*M)b9xs?li}v_INZHa0N4ibF$YIcy2vR5gvWdUa
zTixu~E7H?1ox!3-e67`Q>0ojZ-+BsKq5}lqvP#ii$3=X<7mYPl8FOs&@omGyH=aT8
zk6A**mp52}X*-Ceg_<>$gx|o7Z#QPXxV0IntL=VLT63hSS*`K&A4W7uoG6Nmm#@p`
z;x)~3Z`o!$KApN0WAmF~dSHRDy3j-zPBl3Q>x~;73rP{oj2@t@gKr!?hG5-v!pU>2
zPV}lsZy>Tod6FHt7_scUrW6D<fB){?^Oz;B=W`gU?3S$@IK11bjH(7kqXw?sMUfU&
zLE!{PvrwpeCZ>a<;~vEtEz|KaF&=(dq+Fs!kU&Zt4Ozh&M7to<Ha5-7i?vU*RElP+
zK?5FQm8B?-#>j^Xq%bFbhLTShvS19gMD~&fgY}?>(G*UUgf&I>MrB@Z?iG(>p+&w)
zLCP~1#!6!**1$}um}`Pnnp<u{EM0TUHv#3wnm$KXU>eX!R{v3?^pYWB=<=q=w_+%6
zNv!4YFWmx$Nn1s)>W~ViaY;D97pIzz!hng&St#R*S$gU{D^zimc!d%iu~)9fE4%Jl
zp)YL)XMZZ>Y-vmBveE$9dY+uKi4JfmT__(ko<tRPUcEpyE5;$Y2uf;H%CCwGuhiBV
z!fh*X`(z~bmq(C5FUQz?{3Oy1sPUuJETxc<<i*6TGy;^Sh+)f1wqzJ(h9-i_9x)#V
zXe1h7WgK^zs&NcAkEjV|flQqzY+I}~m<?J+7Uh6rU&&iE8nm!Pl<HSR!HlTop+b%t
zl}370)*;d;SWt?@4At#W>SjbT@q3FXCHpsaCHqmx{*PN^U;c%d@sr1ke4yo7OhxV(
zz6<7^m>ex*g?L7gP_I|v07!`|Zqagl<l#gvD(WX<`8}ohV<HDowC1lhKt9ojk4Sy!
zOWR5;ALvRfG;8=Lkyw6_yJWiQm&Q!OyhjhF{Ql79OR#q_Sh=ILpEZ0)ZMp#Q6(}bv
zS2>4s<k3n{(X(S#u~nw~*4&ho++%1#$>-Bu%jZuc`Fx6*wz_;I%qR3<q<kcpu6!nk
zZ>f9Xs&uhFmTWHqE(~_z&rqZ;$^u0^7F+h(Qb{nehDM}J)&#4h@$X$r<If^#d|i*T
zNI>j3B4NI&2P35+!IU&q#2~|hQsny^8?9=+5-iuOmGTuAAv_zxHXU>JnU&H|`AXcE
zufiYn-tv2sJ+ZAHo|bEEXTo#X(+p>#Kc<Yco;;cyMkn*+6XDX^uD-*fUkj`f{S{4!
zb?j6!-~O3M`jV>Vk(zuiUeg?@$u^ZJ`h4E{IO##OFn1qwLFOOUxG?v%|MxZLi2shP
z(tHnAy@ruO>5BVTc9yP0>>Gv1vM+>AYJ2(47TS_B)>*`QqNR<FWfF}iVT+Hwr)RC|
z>DgGUAyxJX^ecW${H9dW;g790k0nX88A6J$iA1fb-Bu#g4Grl|bY|<Vt9-n9p+vNJ
zvW?31Wbx$C0SDzhrsTD}RqTBo<EzWw*Dyp1do>DeeypX9&Kfi@J|5e47T1b^_i|IZ
z1iYIeQV7t}k5&@2^y3vRAMd`pYvWO@Uo)}qh7_xMp!o7h{*W$TKF$y+e9;i>$6DH0
zO|Yp)*Zd<lrAxr)86t%MEy1>ukeOg(XLCdg<6}>hXgB7P;X~5AHCY%j+cU@zDV)&|
z?YfpWRuwI8*olKK;yZR4+?1|pFK37p0<=WiN<wC$)!xYA08^^Vn`;>&g*O_)eO5~w
zs|j}xpRb4gWa9_yJ&&8xCE!kmNFhK=xUD2)CfqJJwYZPz)#cC<L!@v>L&QN#8=XbG
zVR(3KW>5QxqE~ZUAqB|+%mCn|YZBMWbiKX_u%qdc1B}0o1{m_I%>xYiTz`PEz5FUh
zB!#0|I@ik6HR@bvuxO64HAqYclXHygC^n7e7^hb8j~(Y2{~{8C8j-~^kT2sjInvCl
z>=i~8dRXdcd9)hPw<}maPo=<qjkW+4lUK1pf=Wmln95r4)t+T23`pNS>~VvS()>SW
zh12WwJC6KaCyt|IhC}^*V(Tx=<jK9i+H$XbH~yO@oRZymq?-@O{@xvsV$VM`p`v9N
z{yyT6q@&p|$mjZoVaE~Krqc61XCAJk9vWpH{)f3Bt<oAZ4@bTut@a>|p|^S0uHD^6
zz|UX^$=Rfa>Eu10+R{d652_#Yal?AHVSD*nhDaet%YAiAP?vGe^B758vhHMv-jbE>
zCu6A_W0j9L4Xnyh>ZVPf?Mdq9$+U)xR{AVqHI5Z-gM)=e0k6VJe?ghm7c)(|B6yG?
zQbeE;YTn<{M#qkFYHqz)7V$h<)gY?>FlAy=Y(RVYioa_4K(tIj><UEez|y06s?0CZ
zIt1g^WlV%t_eW*(MQcv+Z70eBcgbH~VBo(OV5O2{RmBg+HP1H262FQ!uwo7~*jRGl
zMNNi)HG2`Y?9Zm>cXJE7cN^pPBwll@IeP2vGh%hq8A1>cv_R^Nag8AwNdi>tVi6>#
zfIR!z?O;6GwjmAV+O`lOQOv@fgy$~fnMU;aXU3Q&THh60hJfWOa<^GWEZqvHm{C9w
zuFZ|wz!vMG-YgOq$i~826c$mjfgl`vOQ7Tkw^b}a^a8O%NceO3ZnUVx!@k&lJwGZQ
z`uoHEvQe~zz1Br7Kc>~9PZ4mmI)rw$rn|6oA^M!yX?rzDtVlzl1u7((@YDjH$2uC?
zKqL9(unvOg#M%e!kAjkluu#IGYE=3T3nHe=5eh<sh*6=G6(>UWh3Xl$f>N#mJ|Ew~
z9;`zTL7Xg_VPI_QL?s{9=*LK6v7aN!Na&_-%S9>e)Sv_rn+EMjB4#@s>kMuF4JEi`
zXBAZ^^G0=-QN*tP5qZoJvE2%KKdq{vO~%z;SX|EwrlDGR&cyCl=t9sM)W>0KZLt)Q
zrA3pisQb0Bt`+YWN-nC0B)+7?B23la*zc#=kgaAuH4E4O35#fcDr%IX)y8vsXY^u8
zarp^ZiIuF7?^v1szxzva5|a*l_x*N%@c`?e(8fLVhKGOFU);5M`1k$AqqM290h{|{
zB=4-eW`!%()|2Oy<H7Jo1tj0lMu{S`uhZK*j{H+J&XZ~_dd@{H-n%!#8``5gwpdf2
z*B0KB`i#fFg10QQjU+SrPs&X_P_lcPHb$$-ZnKK3dLy(cZPfZJv>W@3M?z>FJK9NY
zOjZ-trRmXbZ-h6ejcR{|SMM_(`oepOHYTeH?`b+?klu*#b=pYxSB!7#GamY4{D3wl
ztBLVy&5>bmB>8*Vc=uP5pYAgr`jY&-HYTe{^66UR_1?(w-?g#suN()*`WsT|%khcY
zn5-tp@hF7h!Tai2l->yS8QLWDSE!R&;?Z%)dPA0&bQrRp=r11FKj<9nAJZ=FayI*h
z{ocP%vTwbyB6I!KRqau3y}x)M&GXK79p-VFr|gYnAL=hhwI%x#S>n;rKfJxam~@<6
z6y=vc$w{x2IW?`>+HPmnerG_;s-3`gduONBzDuXoFoS_9jBtXd4>M~t^M*+^o<VCn
zamK%<O^)#knYiF^CNP};QgdDrkD~1oc)tTVrHRR{a9R-)PgtQ`tYhAYCXK>{)zL(l
zctg$>9mhU0G?Pawmocf-Gz%>PE}EjQ&SSo|hBE;0nm8U;G%H%7HP|*z7yvYV3cJk(
zhBb$sR&jzAR&j?XwS-gHVqO>z!ea<{NFi|Pa?0vs{V;aF8#O8t??-xQQDQDneq#-D
zR-fxH&Dt*SOZ~-T^*O7@z{79%7k6zQo<H7)T!G|EJzv((6<nr`Nt;}O`XZ#(ajHtb
zxdy$@X=B($DnQ}`UL|&T>oD+rcAxP`$V|;lMQdc#12t83R2!q!WY>PGN-}8;LVLb8
zYW)>jrO$XIgw}DY>W8#3Sxs1aQ&o>ac(2q(wZFo9Q=jqB7v9fnW3rm?x}B<e3}XBZ
zZKV4v#y{#a9{OVZoHiz_iE#~6RgXcEU(v?9zmoiaea1syl3ONp1E|#`+4WS_W02zx
zZLIq%$I(9Hp)bd4v@uytj*o4s>M;oQxHbv>73w`%;?Xe`>12sXhlu?L`-=yrs$Pce
zG}~p~&t|k}Q9z=kyWbhD-XvLG*<Y2>9&Em;zj$D{{?f-q?yome`^ElpYxTL04|86r
zw{ssIzbtp-xvyp7rTg7h-f>uMeYxMjhIu&hW4Y>9mn-3+MU>kO_sLvxJsbo6JS^F&
z(y?+#HHS;LaOf`1D8b3QVwr;HF4}b72RBl*=djzkG3U7z@dj2i&5Qj$sXbchKxZyt
zVf36qr@e~39qT?$+#SIdoQ0fW78kK;FP2%z(%tVKgjK%gv)XEkU`NcUeVQRs^h#r$
z+Ml$v(RrQPR5YU&7ynhv|EnM?I+4ec%g9EJ8z5S{P<MjmXa&_Mwz>C;e#x%!gOivx
zz)8V;f6Vr=jcrh`sz(Q;i1{UM7;A|5AvG=5-pEG>usSMSfQ6l`Nt<YwYSJ~RCbo^l
zaXwXO+B}`bRj*>Lo$2E=VWVJsIIkx>(im#zse_{Z*a?bpy5-SkVmDaK$>5+D%*U8G
zxhEo@TESWz?1D&3{-a~3q3oDl5es%Ool>Nua<CP(<<VBRb6DHsdRP!7&uGS4F6<eA
zqwLU<X<(5Pw#RnpXdtm3C}&*799h0fc-N~+5u>ldf}kob9TJB_CO?D5u+FUMljNHT
zgjq#LVi$4Rnpwp$*WjmQ9W7TBy%u{IP)7yj*e(~dJ9K<ASX?p7#62s>n}r4N?y5-C
zSD_Is)S}I|C8s0@4ce`czQL>$6obwHQ(G~9(8N`>K*t;eq5cqK#4fz@0lueH=_)>c
zPRK}*CUDG~SRGc!8F9o4sLLFdkco*aNJ_9URH&;$ZU_(@04=s}z{!EwRo)dR2I9c+
zs$0P!f<&+@@RB&MWfJQf+%bE!?&^zJclEWXQ89`2^b_$;{K$iE<Zk@PgKy%F9P_~v
zdJP%h4SRJ*<eI~vBsJQR)3*NNfyik~+~oC!dAs_Hv^Mi*`ilpcw^=!d-mvaSe^J(E
z-4FB^53p|2s1e!L-f*tcUyQXm_e1@~BkZxtOZ0|$uj((-+RS@%fAIkG&Kos2H0urT
z-q&B8wR!hj{lx>kyCCK&5XAO|eSg|tw6)pyNPqDN`>hK1d&9hc>o3yU%v(R%*Ki5B
z$v2~nzMo^G=5SI*_us+s5l~ViBnok9l!J$IlthmnFXJSj6?t>QDx0wqzN!2meKsL8
zz421S&Tzy;Z78sA$4tHH(U<hMCv{B0i6qUVJdR6A(r6VvcblfUGfdNrAzP75wv6X+
zc(PZYN6luGG?~b?=4kfjj9YMw4Tjq|J0*H=GKT;YqwVN+3g;Hj&EXVdI$kAuHcd^1
z!}RFMOwKrHVykB4=AtM2L^JuiBR0*69**s4L6oD4m<n-5fX3J|1wq-HmV6Dy;>B=3
z9n~;0DG}u(vE0Il;WDMz6^1B8VJ%Zeugb(jm=y`B5l1k?m_<1yk;jRQhsfk0_ffR`
zf@dS^!sF@@aXvFrptK(hr9Y^o7JCArFqY1E#Nk9Xa%Cb9L5GeZW5t;w9A*ugqn#yD
z%N|L!IZGqc63i))JCUs7w1#AaK!En(iBO<S5Pddp1_81k#AUZCgdA5AVK^1n(x}KI
ze^v-vv8&}eQeOo|K=4_#qgahZ*Ze#VW|PP;KafcaL{_CvSz^dL(nLYLI1w(ErwS85
zC~z(y0WC7cC_x4m_@Y$0W~wFA#GXA!Jo)lF^id>ud~8}oTE3Pv&i;PbpQ)%(k;XYM
z*2!ZwQ@*o?WX-Yu;;bD7o$fClG<u1d!`|@D?Jv&SynA7P@c{46PdJX=aPP<ai?ufQ
z-qK$@h<gba+Z*n^zrR>(bMM3b#RJ^iD1(vSFz(a+MOd40pYJaoq;`q0tvB5JW`D8P
z=HAd$9|q0RjfSo^7U1@VcTedr&f2`&*<U=MdFRE30=?ngv-*p(Ht(L>Up&CO3zCDT
z-mveU{-UkTKBvEUfPGuy!^_?<?}7dzt<Ahw_7@K%3RfsmD2p%7j>W||@a$Mzjgh}o
zWAWbf<Dcno|5-gfIKjbFZ_|Svznoqh*cmK3gr#-6_jE9M2+IIf=tqaJ{CCR{56OcF
zIv&FEg-%C;Y#ScF5odgR*nAmPTsK(QBil(ifb=s>l20D|adB%ivbM0@PpXRVubY}p
zQr5hGzJ&^LZ0`D{h$e{>1)HTI)u?>l%^@DLO?8NeI;`eY4c|RnMS80kR>D=x1wK(7
ziLS7!GdwaIN1~$^h!Te3Y4ChJwqfRNY<;_oQYqT$8kKb7X!V<JJvq8J0YL{dtI|$n
z=E`)0J7x-ERi(zsy4cKI911L&z~q74Jpdq7FY2S3th$_N8YP01{uN(v%sh5s5XEhx
zft+y{@~7e?9%F1|&z?QTjo7Kc_DkXGl*~nY*y|h_*g4xl2^@g==pKOx6{{+GZQrj`
zz1x!5pMQF4nzk|kDcH?C`VQ3=R<JEQ^2caSRI-LG1Ihi7{+#8#De}JA@?Q0&Z*<c^
zF@v+?$fGS5r~C8ihmX3@Q+dQwv^R%K^^wqMR8nJ)c3Gb)?MycyioA`}x`W`big?G7
zuSVnzMv&Mhq|*_)@)T+bwvNF%pV_|cFp5LzGz;u0VpnT*Y$}3MS6@<__Oqhwimoa9
zT144pjG`805=@t}+6g&vQ*^2KW2cnrqU$V@Wh#`c3e6ua@6OvUHvcDUDa?o13gzh0
zW0?9WT8?iNU8q4|)Lb8X-GsZPEL7bk7&H{zlC3fwY?lJV438#xmi>p7G^V<i#=k_;
znCMO#5==>>L$J~1Y3EXhm5)?yH=Y#pdZX+GR!n)Ubm9f&Ch$UZ_BwU2&9EPz3<)o7
zDP2|?fNMWbZaE&FSG}`zp?uKTd=MvCIM@Qmtx6%sA1U_V(-r%-0*il**#85}{#ZQl
zFJU|&w)vrxNHA?D+&bn1qn(0|9y{c?O9%#h!&)?*dU)i?oM%=nRMjs?S1mRLI)3yh
z)@fkEtw1H<n46=}z);FgAw?8C<P3V59Xx6TTvSsl;OJYHn6QP1MpS3gykH;}gP5vO
z%tjXqh-awN;iGOHlcfsJXiuaO<90W1S8;5xVHQfZwTPI)t&U;~D4OQOk^`7y?23dQ
zN|osqJV!_%B7r-xgrRDVj!oq8ApoIyZc6lED&j{$I*yg8N*J1h4N>;-J{81MQ9FVm
zl*5ZC#zNsaO*djP*TM<lK*fQ55JK>=R)(wqRtJQimX#@Ubq`Sq3e_vbG<*tElz!c#
z?U&?t{R(0l%NzCW`|v2TFJwaURzP7%ghDRTuv2G+|NSU@WlI*ivLzG|eWQ~tIajQD
z#gL$66&7kPOb97C_ZIXM_%*{^gd;Pt=trCjB^KdWm}wWT4!zYfJ;39h>L{>w%yH3I
zNQYryjGfn%f}rN_-@SVthhEk5IrP=tx709wZ#pLmrep06t<_V7?vLK*y4UWaBZy$S
zWWZh}CtI{Ch`a(ig){1FD1*RCC3`fqF>wn5w6W}s*0cc<W=JeM3C~C>!j3p@O`i1x
zJLiXeYbSD#vnMfz=p!p)X^`b6IhJQGj2&;Wr(2#$2h=^FQmr3uLMc7|D&GWT8{L14
z(hAc6g;xH1q?M8nV#=nLx#Sh~5)oN~p@It;L&QS`>R6Hl<)jr%V>4=TWN@IOkmS>*
z&?9pCTA|17v@A|6+>(;t9DAgwdYiUPDFC9-<3A$7{A>gX!Q?pPnT-6~LHN2nBH#nS
z!Q5|f7f0<<kaH^jvY&Ik`Q6vSS60pF5G(YrZSBEh-QxXC_$Ir><+59}{f^|sDsL&~
ze5mPTt10{{4ac<mKO-hea_V`d8foezVcIhOs8bnLA<`&<5HgA}&c}@FjgD2Pu;akt
z-EE<iJT7Vb`+g%PkUwtXQBoiiQXpM?UW#c=uu6EJZ9;4{;jQU&F_C^N5~*Z}jz}d;
zTaoSw{a$++wsDY3Y#SwHOiqtBu_Gy`XS9)1n@>p5=)SI`@o!C^UridiUx*3d|3w0j
zl<EjT!YBc#e%QcHd2B>uCOYx7$wWuSQJUyjV>daKIgGhR^DxS#is>zkn$=OW9w5np
z5n8<O@-PhJRBzO+S=G^S$T8}`a>j<NH-}APBoH;ZA3>GDZN>v0Zc_wku{MpU%_cg|
z4S87_2uHc;(rOHaxUG~%8c#&<RrSMC;8_f4uzRlWuiU?~voRRB0nSuAa!zBje8(!{
zwoXpv+{OhsMv_K@vr${e8(ZWP2cxi+<Bg3NWd#@w`YVl#iY^s!I(`mIm8rtDxY8K9
z6N6Z+xsivqHHO@JfUMNY@zU1gjZM+%?5>BFHetYo3RGm0adV#Mx&ctN=Ik>ojdORe
zG|mfAn8kM&@WSTkw(dE&J3nv>v$Xb^e)U(%UqNu(xOfh2(6Z;)FAp`&N116*tygBx
z&_HzMp~eLWP3_t&f?EnLeYCbcd?Gg+o~5DbY{8+iAN_oO1i%lrcvcPT=w}^@P*;wu
zYi!1fcfK8LTtsJ6&Vsa{?#q`38-qnFU!O;N7c9XifQ5Ji2Mbj2WD{gju!tVQPo(i?
zp-L_dyH^^U>cQOT491<fzAX9UtmS#GH;eN~u?w{a_7kdwV5PA&WP7ZYLD9-e<MO6o
znwlsdoB=B@2-{USNa)LwKQsY1E)?t0!Ke~g<bbA?#`z8wx7V?$1UO#A{W`0<(s%-n
z=R{UvwrHVK74aEbr**tBL}k$+6CAONQX5J`jSHwN&8kO@bvAVE{9?UQn`M0jW5olx
z?4BCjDxL^UC+Vw4HQ%Zi-Pto0(xwZoGe{6xvvEkFC52#i(MGMf(ATZAC|ReW6$pv2
z+#vf2`rH-}gIetHkW@5!e)QTbrYb<(Ju8iEbcK)j7E(cO6|$k$iV*pEp1H)gN6u|r
zM4b(qQ=Nw*IyTDuSB_nHb>o7XRhx|&J@ig`l1Juamik5G$<le?v7f6g(}~3*yjy8J
ziQW_l;LX{<LdqE3Z>6yj&qtrWXD|49fj3uZYM!2`=Ghfcs)JN^!SZsp70iimz#8l>
z1r-N<cT-^I(Tj>fY30bq#^un-$}D$ICScaDk%m0nxDXqb1m>C963!cT=*;1j#(D>d
z{guWRBwf(a?yb|d4`9?2#C>=>>$o&16|B7+VJ02gYI$LQ1D6F0$-LKV_{0Uk4krPF
z|DhRp7}|r)1gdk;fygdgOojrxc{ja=(JyqD2Fju{t{3(QDpSc@MfBihU{JEG0N;30
z0dFnL(o~f*YZi-SbOqKIQOhK8-*V<cgZ6M^z^X1nO`cp?_5;Nnp{inq1*p9^`5SHM
zz4A~!-`E)1A9Q%=E{4xGdQ&hPnBKevdj?Ir2xLGP15sWNy_^e;|5n<lqu|!7!Km1>
zV+F@-Uef$3PCq7G*@Bk_8|x>=C#Ph@5#EIH1hzhSG0cVEW@@9+^09aVxBSBZBhLnE
zE>c)k>}s6^U49XM=Aq2GG5$DR@~2;u_USe)rhL5KxHv(76?xYbUX>rj=h`CsYWz?q
zq0C9}#t=EOBDK`mKo7&~o#jjLYWa!yQ@X!&U%7z+cKKoYv#=e1O7!QU5&U^6{kaQ)
zTKOdX`Dyy|M*6dr;J1kHZK8H*PY;2wX!+sN{SS*q9~KQhEE;=QH1x1&<YCdk!}ZdY
zhy_ACwv_HGJzF4AdN9P`RfLehph^e{3@DQyVn9P^fkBxfwR+?Hc)u0|pI@+H9Z225
zgzyQOZ~0!Rpz$Z+dr({0flA(WxHpJbpV;(}-`<pHag+S20J=EQK0Gmc4(#fan%jzg
zK{N#CB6<dS(et8dchG<^XTw42F?jwOi1Vc@x5H2(gAJ1lRiVw2!Vj{KNjyG}AB)a<
mUbKx5%*j0U)6GS4gMwKIkT8wW*etrd@6-`s-M@ky-2Vs3mkO8w

literal 0
HcmV?d00001

diff --git a/.doctrees/blogs/index.doctree b/.doctrees/blogs/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..6f3518277a9ff9161b7e4c69094d094f384b1d55
GIT binary patch
literal 3332
zcmZ`*OK)Vg5}r(^JL$YS$uNL4let1yTm~^ohsOf3KnSF}$tBW7=<d*w&pGb4iI08o
zW71iS1Pjna@(#1+7x0gG`~z6CyPtg?-DyIkR*uW%s&du$mFw%l4}b35Nq>H9YR2XK
zgp{RJB{jisma1?Sl{S1I-uxy!4^R9^lS`wVE@<#~piwX`N=;?>BKW<iZn-cg?fDHu
z3(J-4kFx<AvW+jo6E^Y#%dMbGTYm7dP%{&v0`~8HPUq2aC;m1x5e#bre>SiKKg#+5
z*}5OLah>~nf=e3CT~7QKu~zfkS==Yazs^e&W8EkOF$O|>4<xCYug%ici$zX~3r(j{
z$N3FQW^pU9F?)Qr#*X|46Q(LU(N0RLCzncJO#bufzlFN&{vkLsX@Z*M@oEz;OFBQP
z%D^779rlQ=yPRz|nO&r2zZ3X-@0`B#Zj+Cf+eMoNoIj1{+0i+B952fLf!ndykhy>2
z_d0%W;P)mXXK$s5p~{}(*m3K2gxaa9R!){nHE7Di?%=wru$nH{r``KIN9>{BUb-bx
zj4ElK5_|5)%~noj!78FJ!gqoA9WD#uN;>5tsj>eOzgdxaNo&i(*&3=u>I5LvbS6|D
zuKkelvIO3){cd}@7i>yb<y6x$oZazT30_7?aRuEiz;+Q%%*iJyACJ$5$jL{m3H0Ok
z`p1`=)HT(CtmgE?E|;Ahm(#uo5oT}WOk0b#c4WCGTR>4)lNJg~=LacWOBQ;;Qdy%3
z05O$yy(#P8<DjPpd+J}Znn)u^Ol?+cHPh5I^cYxWiD>fI>ZEkKn6fp$S2!KXm-STJ
z=_C#1{HCGSatRvygT5V1wUU+sXny!VA{=Vn_miP=wstn0v+=nf%_yiutkMCCl7O_9
zR5cdcS}BWA0hM|1Ym?x&n-DViy@0}2yL4J$wTtp)k-l*kp$uZ}uYy0E!i-dwE^WT>
zyRl6HJK21ybQNCsJp<s@8Tb*KpMk-Ljf4uCB-tgmEbCeb{Iv(#o^tMYG_94knV`=o
z(s|c!F(Ufqz#q^#M;0xy&U8uxKPYMLW^lK6Nwh>3p`j_MAQ_=#1&vpV@o|ZqW#E!3
zp%whd+3E57$A;tk(ap__YOQp}h%5zg3>e^Ug$;h(WPZuYAXy3iXk}-`1pCqNHAWW!
zbVncE+*mp3_u~ZsJj0M5feQt{D~OyqGNVX$iS=D@+uzr;P`b>}2ml+niEZTE4+|xy
zxFgcR%MJ}}==X@TDqD#BT@I}6maeL^9*@Xs5^z9KOCqAdagro9#L_QBwV|$5*}RHq
z+Nbkc8JaDhOb{t-_KX|e;B}mp+>l(vi?mU0bUXp)c5oR3W+c+V=x(c)b<NMe9KYbi
zP@RZLWwchJ-rMM2s8h2<TRezl18n3BC=|TnHhjAOhTp4coq-}~fwD*m+li8#Aoj=$
z|0=V#HXlqT5Sx?w$}%NSl%9c3529uY0GgSl70P2#j!npqUq5{h`P|dfV#R!vEX@C_
zGY{sHh}R3MPdK&H<OAJc!fYkrY-C9e3YD?RZ~7wuvdVg9_7})Z9T9Tv_sx~GWS%iT
zV*)>b_nuG@_d5n}iOs%pptM1v?x%IL%!DdP^RD=#f)T`n!Z&0?mQxDjOXO;}h@(q-
zFxnn~*98(Tr@%nl6b@sTFfLzF<5k42P^4jE96c@qAcL2v1bwd%0i)Ey$Abb|7a6Sx
z7a4}E*mP6W2kn>ytcq#V(60SDmCpc^SF0;y7s*Lb<>^U$;SX=_88n>m!sULmX+B^(
zu#1s{gH1NGM9(Oi83vvMTmx9N7}oFr;Qeiy{}?k&Szsct$MFl3*vP-Ks>)^KzU$y@
z;Md;nW)yri0}39;-^?vjiT{g_8;{~I@ioH@4i=GAT&CfT{o0R3Yp1=x(}ll3ZuNJI
zrQ19@e|Y1f%@7y|#=iv2e!exl-(Af6v^(gB@vTA*pJNb;!sh!4(@e21+06M-l-jEr
z5Qw9jzGh!uJN8=qRu~Rv?AZC;WmY6xy9EcgV%cE7k;LZL={4nDmNa@OI#^WYO7vJt
zsxPewd+pH;i=)nKI&^r$J{~dP4qfTNs;zbhvlv=)!c}~LwhJW$0Es)`W#0HC;tlfe
zkp=eC*_!z_e1=|wAs<}}W4Qsn8n1nuz9HW3#R>1*?NA6t&LY&e%J`DBO*hU+ek-xY
N2!|Qvd2sp3{{e5lCV&6{

literal 0
HcmV?d00001

diff --git a/.doctrees/environment.pickle b/.doctrees/environment.pickle
new file mode 100644
index 0000000000000000000000000000000000000000..b7481d5a0aa5d906479769814461575c6f30cf00
GIT binary patch
literal 4282851
zcmdRX34k0$xqt4-ec$BRTx655yE(WL&Ls&U<Rl5FVADI(yEC(MWsXgPAc_iP92Ir(
z21OLb3vU$l=|fRG|DL?3kLP{7hx$MiK@k0aUsZQcS6BCReW@%Rqs&%ySAF%>_x-Bs
ztLy8RjlO;4u;Ih-Ut@Zul+F|mcUpykOu1OdTZQUiW9kL9OfI$6^XQF(t6n~M!(e-3
zq**Er?rw}v6_eGnWu-FZ!NyFRMy;C3bs8nJn$Bf<1{-6I{nhebhz-6GG7j$=Y>Y}5
z^VVQvOu1GlSdeklp<=lo;yZU;m@6I{Oiyl%F?z7*!N#~wyM)SMb#Nb~sg#pgSbAK0
zdVG39dSZG~dNO1f1<5M-x>zeGEjw+pSm@344K^m(j5+Z!oB0d3Hlr(6wZ`gW9j;o1
zN~Tz-3|<4=x?(w8?KErEB9t=NnC-~QepspH^Je+TU}K8s1*RQr%<w!J$XJJx#S~P0
zLjFjl3IZxuu=dlCuvyA<x@Ef@8!4-j=_?F2mgUWADOaq5Ksz}}NZg4}43LJEtY!v)
z?$|;tUpj(3T8*(ev(Q&F`{3~i3n)y;6_aMp!sJ-BMuNvHgC?-egl1R;U@ubufsLzG
zdppkTfU*Z06APf;ior_-l_yH&VirndY68~NMX1A+18J*`tFP^VW#*w)b}}PbEFCFl
z`qF@>33iJ0YuBF&g*4x!Pj4g!tTM`durUmnD_c3stXPBTjg8SDN%IJX#<7smo38>J
zb3wDUvSm2cG<q{Rt6(CoX1}CdELLyme2Me!hL?b}ZWwH|^;y*_1PwC>I5T={*IjqS
zsFuw_C1+Nd&W|=0c;W`sp|5OJDn>GARw`Cy@MvRv-mFxuvcZrrI#Wnlhf!Tzf|Xit
zZ{{#-Afr0V6>LI~y;`m%tH5gt%N}fu?92nO#{y*fQ^iAt!CGUgbvT);r7UbSK$~H&
zF*Q?=AC6eP8Wc7$W%Zi1T-7KSa~8_NOr@YU1%N@ZfVhIzo3@bOTN_jHUA~yAf$mW}
z_-uN$m^6w#s6|kEwVYv37t2qLN*d(E31aI?74v2WRJ0KC<c(gCa-mqYjIz~hL77Pl
zP+~&sJKC7YGIDgF)FzjX^kJU@II13DJWo#pCUS@iSB0Q=yJ7&J=IdgK>jo4!xoYK0
zpew6_qDr$SI^J*}4mNJ+g4A8<BB&@;Ef#Z0Gnux!_8tM07P>Zcu3HEH?>c<?sa?>J
zx=QRZKJBVxs#XW^-VY6?0xwc#)wCgRfGG1;7yk`<lPu;-#R66in%xwK8KbA1NkQw|
z-I&yGSta9;SuQ}^LSam+)Ji1=EB7HZ$mvXE_7kIeq{Ngm8&t<Y2JI-HN9}^J0b-D0
zJOIqFN|{`-4@zM%z$=eUX9@z0f~cxSZ_b2*Zv`!Y+MuTb@NzwlC_%3^l{av}rqlC*
zq$lRhLJ5Uo)XF&^KemvmmNLoyBdAO}Y(UpkEL+ISOz{yQq7T{yych#r686DI8#8#w
zs8~q|LC;nK1@0K$m}EciEta8^V{In^X;9}-$NfCiwgTbDn>@he!c1Q$do;Rg_5hq3
zV-Mh8_Gmozu^m0oy}^(&erKjq1+n9sT4Q0}Je<ix;n=B~*d&cyrT{-yp%0kZ-IxL#
zqzz!G2U<9E5~wWp$T|pO2f#BHcS8qmr^Bk<+Kg7L(qLnFCzM;wRSf9SGQCIeJ;nfv
zRv9n?s;WW1Z-8)Nj0IX1P)YS*4H^<_o1Um)+F}@aa0^s&ruUWIWmK5J#yLG5AGJng
z&B4a#>(a*N-Td!?%W*_tLn`#Ltl5kN1;fb8x+Duk!&{jF?JI3oG8O2Ha=>u9m`h;}
zXwg=w)`L}Io%FWz$1=NNSYteU&RPeM!w5jR`ONUN-YM*HCI$7*^k#7AOVkSe1?&SG
z6WF^7v{vZB(|0t+vUn|%!U!)TS%lIDSJ|;XcxORuGD`@pf(-y{XyfISpc805o1CuZ
zb4C)navUdEGR~#b^}3HIF@n$++6WzK^M+5rM!2=%CaAyw2Fza3-~~dnG@yAwn*njn
zWK;le{E=}elR~%;v-xFh9a<KmI#WOpLg&U<1Z)_5RDr&m5dj+@S<bLd5QV`Xf++({
zyay}?u!ZhuOk*hzl`{x-?Cycyco>;8*FicnhY_s5t)>|P-58AF%_L6m+L-Qn*3@Ks
zW1?9o6k)&tZK0v~+smz$16G?oYNy&@G;GUT`C|D<hXo^IG6SjFdZCup8g_zhX08u-
ztETg9JGbxeKBW!Rp6rKQ)wUO%ZMLPMt(~(nU9FZXo4UHp^5M)tXR+MZRZ8`Ctv_`w
zbl7KXT(_|kqASncYnH6?*<I$j*T^~hpoCgaCv==$dq8}7Gq)d2pe}BOENk-9k;_!f
zmFHdx&9ZG9;))S$gXZ1_b+Fn}R@KbpDoX&_JupTX)gt`A1SV>=3Y`?#3~9@RW`dnb
zf5qsrAR`!B<zk7A?>j~{#%DlZ{53Q_w4X;YQw0pyk=VaBPIhYBnrCQXNC}2V7)QZm
zNp?YZUp%d|G606!C>%4Oy)(lOG!C`|LkQpzI`lDBvkc}B1dJZ&E&2y5HM21avm-z0
z+kutw482AQ43ok1*BWC@urOHvke*zt*$8rm?{N$mEc@=pL>B2a2N-iO;<0$v)7GZ4
zC!A*oT<wj7U5mj+|8W_&xsBT848fJIzOq?L4`i-$1D((V3^vXN3jv07KK<yz$%l;z
z=MSHKbU79Px56D}VUTCFfdR8@I+bb6V$Z-#tQxfvm=6|MWXw>VC%?cpUaKYnIb2c|
zfLssE22f9&4j66Fl#2y32L)h{0&iw8LVW1PE+UhJ=eOF-Kq~-zBmUsgOqR0w$l#zE
zz6=I?_Nr3KWvX^JbJTu@*0^C$pBxgbo8r%qHjtjp^D}c79~rjE+ZfeZMdOw;i4}yt
zu~4zhax#s!J1-YIStrMC`<a)@ZaY6Kw>dqywK10$i`FQo2~_x`3wlY<J2r_0HaQ}(
zlSGpngvUh!Jxr;FHm@_$8O<g-(5@K`wB%sOg@#&zf%;Zk7u;Y$i@>Sth|V^LqRovn
z{=#_9C15{q_L+7O4);muBB1Bq=m-=$rj$Ka*rMij=z(bvb`g+tfz2d<1+5!qyA+C!
zNuB<vT?DgDz#NAj35u{==cxt`4iJZt91~<=WMU?C<1DA^Nu~;&*-FaFWd_Qf&{cO8
zO8G83?fDzJtjcN7vJQ9T^SO>Basr8A&~S)(YC73@W{+$*ksfh%J4{5cUWF#I^E#CQ
z9KzI&zkd<FzXsmB*+G41U)g3EaJ&JGVCRuDm9upQ+M4P-!VZv`kAMT`Sq)8zqmAiK
z9OvBj8pExP`R>zZv*X<*GZf}HuUYGM=7%`#op<4-J9oD2?%uVv?d+|)y0`B<zkAD;
zeOveMKNlKuGZ&wTLtly$<C#0VcW=A2dmCigx%2#728T9CJ^`mShn*363|MD1Hg5CB
zc1Jsv$6)OR)XUocHG_?l_(N=SJd}s2A+-2hEsulKJpK>{teiDq<<R~{;lMBqlAXz4
z%AE>-?lDtF6mU7i^BDUj7^vCojz5MuJDcNSdz;N4dHONez?#NiG0WYa4k7}^_7sc#
z@DM;X4wx^N%f&-Dqq56nri<5_v}M7c+q1#ORBSuUn($0cPHH}MtUZ_+)i|$AfNmzX
zFaQk^ChKfjg>}4>oB44@i_@BU%|~VHU=1f(+y-D7rg`7OAPoI9OmsX?!5FciFB;sj
z$omurA~d#ljP^b?3rR5KusfRDBx^tPRC2_Nm%z$ETUZjDyRfO;s>Xt5nlfMxx{*8#
z1`bQx&M$w7nfBgve7&Sr;1l}Hbr!azWga?KsaSStR16cWpmMPv)}3I811R%#1ZM9Z
zZOq39d9xoz0uDuc2@9-6Y@yH!GKJovA%c9W^H>0x^H9R?H0O;6!DF5E3TT_o>p~GG
zqG;R!7-~@ZT<1lyT!gs-;F~XOp!kty)!{jr#__t5Sba{%OP8*>7M6h6+k$lnqyqYG
zi~>s+R)%J<$G8UL4kK)2acJ3dO<`mie;7=4n9)?LIC-3sG%_i}W6S~krYE8EhA!Us
zcxuw%MoGVQ1i+NyGra`Z1u)%3Zds_<3*13t1G6)My=M+ru*RY7mnpDi%&`R+oM3hg
zcIG78WW&|3G7Rq%t6-d)DTCXXR%3AzG}(GiJttWzaZ^KH!kEV1`D-^5p9a-$3BHIb
zZz?8&A+pv7To(Ib$_8r>&1Vb{xMU5F+q33?$tPElWaq&$UZ&((Il!3!tQ%I+KxheH
zN-~&XS2zegH%@R<Y!$~|bvxQvZYPSWz*0M%Jv?IDfwlz(n^gvLkS`2Dm7sm(thANG
zc}O7zb`=b8(4jXUr>p^7hhgg|qpd<m4G=K5`P$Cq%;+1#*TEFD`Pw$+r#7GT_t<Me
zo-)cX$%8Hg4V2xDInA_u$;#=vU=V7)%4Z5N5|;DLRhQY#mn<JxTHG|5(p9UJyLvK(
zu5<=B6$ZUU;4%-c<Jf7ZxoMpju-@Aw48p_}B5WV9z>0E&2EZojXv#1#ipD837};~|
z<{Z-xjMjrN;(*ZES}I%Z1PjZ4W>2q27r<~IfRWz&*gJHw6@Rv*j0O!XyFtqu&Ek*=
zEfW}?446No%?ZsItTvc%=U`!`$AW<!&4M{jDo({+&jLF6%wg!Bn<<>NVvm7#l9L<C
zL0H3*L6ZX5{?6+>TA3MG-Nbb(=yn+2$QyzXwu|6tV_Y+5W4V*jeUBR$&f<mP%s@xt
z%38A&Obm-{88CF&=s8!WVlTWkMzgj6)`gSAUM_|8Dliwy&|A)T-ed}CSbl@;5vPjS
z?ap)F;_|Ci9)MXoUr9$K%)poSw1Vq}PaPOP&T|9_svi1_5}Q|eGr@`;w9%>sB`pei
zZr}<Qv%?|90&fZ~Wau%t)|s>y3JKU!tFw@g0%XhhsUmc(e47AK!;CQu;M}0!C>C;j
zNW~`>j`Hq;^U=l>XFtl;Fm@j9Bd8;5Ex4R_w9)21?=7;O6Nz1zE4gWK0JPUoSp#F6
zT`<tV>NW0pI9yF)&-;4w&V$+Pfm3pGHTrchin3Qu8?GFIH7&NfJcm68X|VcM+|~^|
zpTu6F7@Cg_7@h%zSf#1#HS}8sUvos{!Kw_-g+LIa*_;RhV25XN)gpvO_Tn4Zt6>>&
zHJBS6xS-3~=w&+~VEQz7<bZYl&K~BmYl7t$XjV+aGa0RFrV1SsOh>o{3C!{57{#XP
z^rEnM$VKG5sN`Ub!)+g6VKO7dCWv`3Xb_sy*(1m3fvG(V;9x|GPuXmSEj7TqS?nFN
z91K`JFrcfh0h)I_Y}2aLdSErZ3SBYok3tb#FYp-J5i=vvd~FPW3H*1%;LR}HFoLKA
zSO~ZG+n7qAFfN!BP6qA$xE%Cv&?mzp1Iy-p#rKthGNm%GWQLh|Sv*s!SVQ>Q^G|5n
zj38f^ftJgrzmSE?*_n<ydwh+xV4SUI6XbFc)>tanJ1Jm$#f@JN_B*a6fEfmh?E-e$
z{G+FM7`ii<ZDT8j-CfwLD?&G)u%kHE+Ao^BC{vtAPG61lqJ+~WG$z3~Urctm@hPyK
zGXUL9^T|}SZaVC)4{!s^fiu`6z5&?bESlg>wUx}6IZsxnO}mdQ-bEwt;#f4c=LlaD
z0z=y2m<^sVYUAjNO~S3Vv)??fv$G5S58L-JO_}awV_Rc<W{5_DEfF-AVYwZxG`2s9
zW-uF+kQZll51Sbj<YsdEL0(6kbM0=N>`?UBXHd@^tVmazBRHyw^~OlvPAQJg+1J7`
ziDpqwKM!-13>#p8KbH>}TF`bivYA5G1oj7j4>n$5V%X~dd!-MzoM0{vi?LiWgRoFx
zFI56PfXPVqJI(tLkOrH0FbdChpV|;~M#qsQ+`FISQNO{GvRMLaud}`H3fLLpyh46u
z!Oms#)iiG|Hc+#XuL3KaxVv!3s=#V0UwgpL3!W!ym1+_8_?)8&tb|GIxB~9fU^fU-
znZX{odSh#fKwv(5%(^nfuDvG7dU%lYMD~_f8O70_{U_>XTpG%%bTSEygq;CcIN&L9
z{fr}JwvVnu19o)Sn{pELX6AXv#-5=0c*Jm@A8l|FI8VopnO6^|t+3?HGC0_T2NT!>
zfIA#5L9I=PA%LCDV3P!Vi2yyxeafoJQRg)9TN^9H*R~$OOz<aM<R%o}IIRfVPcuaZ
zw+>)G0=8}&6JKO@T(`DkqtV%M&6>f+5;s+I!ilqaCK*rx9BF`_)=hL6$3gxMtvFEL
zEb*PktZOpBa@oBejF9u)ECY}nmd=qk)`wOG;dN6)p4av|4!&O)lpPH%*nj5+Wm7n;
zWSb|Pjlv8H#y2+d?}jaW=q)$l&i|%sc`vXDA6`3{?LmaC+60x|bS+Mt+s?M%_rOA1
zp#vXTW}yT4gO#9jVKE?KKN`%g8v$8I9Vpji;`k8Kj=s>$!AY0KOtEP86kd;FLWj3Z
zu`Xx$6f?uICEWj><$KK@!~2n7-eaxA%<vH2jp3&MVD@arBdjMv$Kx*tpJF+^xgu=N
zOXaY~SPo1E<Z|F4yc-KM78u%*%{iOUTyZPmyud_kdTS@b>7kp*d4Y*4xYovowK<!F
zM9vFHG}&JDhgBwAp&V?q`wHco0(g>2ObL0%qiLVAWxB$jFPG!%Rh7<WGRQQ}9u8ps
z(U>c;<-UPrqs?*|7Q5MbhtiSG0&KIfc)|%~FF`$wFO=MO_7j#6_9**7f=&Sp^qbfl
zXqNp5r18zzVD?g^HKqa*Hi50>9J@T)cD5&TrISBou${2a1Pdd-ANmI@iY4oW<5BRe
zQ+PECm<)KHy^56zFJT1_ri(VBIURTeOpd|o;mHH>B|JhAv6Rp$ZGvHO)3q#iEu7m}
zz(P3jl)C`50r(jJVRL`1XTIH@UXL5CI27MJnBI^+HQNV#Lhv*=cLpYZK8MfGNS~g~
z^5<u!H#$i+HG^k0gJ<t{-mgxtae}9$JDlJ->2sanc}E)wdnW)(Isd5hqC0(o9c+w6
z+Yk+jLHAjDb9#%DvMasT32skc>;!kDcRIn{E<1Z}b)M~YLzlRred+zq>jN$ym%5?L
zb~^&NJbi_e@XGX6PVhzPtDOXfoAg>YyP58BQYPIZQ)$b|)9Z5C=Z4a5!c02r<X*Qs
zy%9v!pU$zrc{g?8D0@&$m)PHf=`#CUaapMD814|Mr3akgjV^(gxhY@nKD)_f;T7pu
zI$2)jKD^mYc#9jl)qQ(3U3ap)I^A%BZ*q#U&x)W8f?Wn2nPd#V**!j!ek)Y;E$O$i
zzi&&w9sfE<YH(VJIq(rb9s|}ap0-S%%AWCaDDKW<dJTJOpNE0@Po@GZAWmYtiMVGm
z)2CzV^gGgbu)lZOf8UwD!>Po(Rym=&+|awzcRLT?lYXxq%qF4LWK;0NNWDLOuao8j
zJ4QOe54s}zP!n4pcI1@)$c|x7%KOqE^;85`RPEz1jQYoIMn7Ty{bc%6?D?nD_d7WI
zO!~7<@PYK_?4WZ1h6%MX%RY$O25pY*1M&SrC#}1q%=R=;4G}rg-*?-?58TiX-OxSh
z_c<K>DE){N{Ezg1I>ATN|Lp{S>gM{{QRmsu)4y<nzfAwi3I6(4C-fUP^xO3BoQEHC
zAO4R^=zra3kGY}Wr~lw&`6Kjjb{F`6E*p=#p(orte{w@lx}iU(|KenRD*aa{__X`}
zne^YBhtIkspG*JUdH8&KBWoQ0aC7|AWhaZr)YBielV*oQ%xR3-5fE`Im>mg`>?r)<
z^~jFK@EFg71ct|Y9!$XSM9+gM5Ox@JH>=m<iXhvKb|_EvB+E{N$gSBb8$Pnry^$FZ
zIqJNb$=>kMCp!xw+1dDGKbV8zx%gufnup=}%?FK%_C{*~wrT*BV_~Ahp1|n<=p?4_
zw+^P+>?nOfV|*?TX4;`nSk-{>mm@#m#CCQ86jU8FkyLgegjp|~U4+Fih9CC44Faqa
zb_&Dx3Hx;^SZ8Djmbe+0Lg-fZY#D^I%i)K`RzQHKJsF>`be^|kY?TvRjj=UO>?DjK
ze~dyW#=4x?T8yo;V~xqK44BCawmr+T9usYFvYd*s)127p7(2s>or$rHPHYp#&T?XB
zLtw{nM(`XAE@HuRF*uq9&x0VZ^Z5|UcEb;Qz8M0mSZE7`+$~o2WGg)3^0?3mHOAo<
zDL?oFGhEQ&7_LK@S=c=nOexc+nmdL&3ft!7&0ge$VCe^Dsjx=`1`e$6+eeJtu7vGp
zvYl2xV4T8W>^U`i?np2s7w;)pxV4N&;oSB7wqkFaeaIiI+FhAS(gHib02|vCxPQTM
zv>lkqUJO61syi^Y(~0eZ02kJ7e7?tdz83;qAeXo)_F;<s_+z(~0~o&4^WZWFJ7Y}t
za(r=x=k1l=w^!kd7kS=Zjp1uN4_=I6!}A~wVYkQ0;ESy1Z9j%{o(Fjh7d#J&5O(vI
zAi^tn&<&Nn2`c#5scF?4sd<wPc;3KDoPR&iwm2H&+&5r=W)Cq6F9xkb<S<0CNASmf
za2<wU;(2gAgm>F-VGYhL12zLe_bIs8;9|04(4lgJCsX#N-pGv*IqD>PndgmnKNHrD
zN__hojBCyqTw`#HW_!<I8M*JKG~cnc^yWpAi(seD;i_>`GnunR2z#b@7#a52-8y=9
zNZnNY{QNZzrR>Xr@$60bV~gz-7=ESa!K)zb)Uz=ac4FC{p~nuys}AfT=mai1ID3$8
zipDtJ;@PACd*qveqBAOGZ}CJLQ|xU<H*b2fGe+I&NtZnek!&4*?6L9H7;bnTyavO#
z)ahXFbr`<Q^WgOueuL-18!<fSd2l;~-FEmUeDP+_+qYo&t)2&O!|>ZZ58i>{J3J5W
z#PB;k58ee~w}iXAk#|GHZ8vvgmiOR~-DSKN!}s8iUH|t%*d_FSj9@PXd*q&dKw%41
zsLH)a=>zb?y0#B`X?+MHu1WG?Z{#E1$bAsugX~A~F{AS_2=Pww<1DGo`zO4SPkJMt
zf=Kq$_+ux%AH$#VJoqexU8y|ajeHJMecqGw-y!UFQxD>cFMHm81;Y<{9()zTF6FO5
z#I@qH4@1nQ^mWYh4g9fr{3e85$$!fm`8Gs$JnW?V4(9oP_+w}ME{1>LdGJFByOe(9
zjXdIw{D(L4pAd1`{;~JXPau;0FZ{8~d(@lszwyOS@y7<x&oKOR&x2oJ_?Mmsk74-t
zo(F$`u-lCPh%f%n^Y#e{yTJJqL>yZ<`y@o%e1G;v{^HH^6hwFnwpsiuJaCCT?alWL
zCjA@!*h!zo@N=FAf5-6io(KQH@IO5dhT+U_IL`dAiyDEkkxpzB#zs4_F&IlYv9S<v
zH82iejmIBb=o299VrQZ^=_H7_+MbMArr?jAZ#snCd^5aBXJV>Zo}{xO>@qM1B5uC9
z-pD-6G9Q2JB2L1$tg;Ir<g&8R8(HLyEQUz74S#G3OT2HFLd0cb8Kzq9NxA~VCwm^O
z#BjUk!72=|h9Ab#8jPJ{$7;2n#w0#xVF%a2D1$SIGlVDnv=|;O9s}D^c${N*dVOQE
zeP`9R?Gk(K8=ua7+4KzG_1pB!^enud3#(RJ)7zjAoe9@$@OoNBxEGn;+87H18X6|J
zwBHWbRt}rE2w&;i(tV(N|JDQh2OI0{J-aSG0CrXJ9AV}<Hbd`9!?wGH<1%bc!*8yO
zuMITXeVIy`5^P8ptS%fBjcN_fWjIM!_>$tmqDmJ#GYce5m!z|I=<vbHs{j>Pc<zO(
zQ5a;GKiBvOie<YoJN2rt)Tt7jdBl@?O;#tEJ+Q^+lyinJCubZKibuHku8Fh%)CI>A
z!K83<t|jEO*<?p~V8_y_y<1qNY64Ik%By{p-P$s_`>NxYJZ`nX?pUSE4md<l@e#Ep
zKajzzXX|09Us3lkx#6cbUEMqPb)C&xr*^ot>Xh!Ul2AIoI?&Y6X+Bn6F$L%ZMdB2)
zBqSdml5x_o^FJOHxZ<)^<Y&~Cec9~AE6xIx{19*xbB;?EWP!6F&}3X;P2uPRC>8oF
zIDig2bv~}~m_puR*V#dLWQr01eL1iF&B<i|P67(;b8%oHX3BI)vEyqV!MH7xHAjCz
z_FJSX;-)REpnS<;`cz+%R_cYTCr-9yzHD4hPz7ktg_M)(Okb*2#IZg1<y=_@%Lul5
zTog~@%7Y@<;qivQ^9V+DG6zRo+3lDn4E(K@SBCw<LGWgYen|6ll_JQ+bgF!QV3lEd
zRe>`=aCQXdVt^PpAO(j9{7-Gv`zz8)YS&q%^br@&%z;A?Y}^hVaESYq4!7qiaobm_
zt`B3@--|YZ0Hc%D>lEY{F|QN0ceo~oqmx~CKOAi(!k}pDEO>A{*H;g>>2cz4+yKrn
z7!|YEG6r(FJgm3d{iai{zugN=&2-0bwr5`|@^hi>&ByF82w(-5Q00Dcu(8g^pWD%~
zK^Yq@oY(b@5q=o>>I8%7z<QPktQb7ARdivn*H;>^4BQrGr{JLS0Ebc$mJtz(u%MMU
z;pi5eP{n;xJfK-D!0CEeoU10&j(kD}EI53PxBP*Q*zp9b+wL!6$<86`9C>njR2v1s
zjla^D3hlOh!hs$ar720KJ5VzIb_nN%;nr&y;{Cd`A=(n3{(-CsYZiiIh_J%!2qm;{
z1#7VCb~KxORq-2<?D|ERe$;v?4h_F_Y#puimG5f6zI5E|UT3)N1PtIE#<okhZ`nH7
zpx6}$4_U5s&I&76ZtA>=Ff+lC?diVUJX2t+?MffX_qgKrn<{Q~ZE=K}$Q~w|BJpFW
zWec@SXm&T~Tmz1};vuQv?k2=QU=}L_oFk41n~9lOKAS1HZNkX!4&DAy$Q2c@C0awG
z8RnOU-8xz#c#;b$WOA_KZ(+Qw5bSOspx}sW^8{N{1#i`Zg#$fok0bGrjx=o6=kE)A
zvb3S$uaBFGoh(AnNk+iji_#WYSK!paF9<Gc0a&5U0VQuV`Q!6T3MYM5KJT{5xlI&U
zaPXhG0m4}^Sn_^B+s0ZSTvLNloH+~Xh10iAokAG~xVqMYJHm!`eiojPGr?trLw6$q
zYANudQ?aamC+|GBc5q}1Z^gip19LC|8>9Y7w%uj|mkuU5JlJsY>yL{j1=r1wI7L&~
zX7)s(X}UN2`=<X1*#zRC0bC83VmF3<i)=<GfSuf&wqD>=!(KKCZsK69!X=1e*y!!I
z;e}z-E_Mnpwg-mvG~VZsnRfxqdEkA#uUA{@W7oBe;l#2#L-Mz8o~Te90IyulRQ?{-
zC6$M>!$D~F*Rq+xJnRTJ7zI#jvnLO?##{EylU>}}0uAmKj}-gdzO}<AELXsyx$ZQp
z+Zt2RK277D5_`FgopDXzLjVvQS=<Ze?%}-<1)qFy_7Xc9^h)+BSi{SX9|g;Z75IT8
zi@5%TCky#B5|iAFNftRtvWGC;M0btAO)*_ak*z|M<s9KY$AxsbcFKx|r9w{lYVozR
z6P<k-Jii4?UhgDm=PU8AfOumRI~vat;VNWyLQtZy&a`5%nQ%%#!t7BbnXprP5+_0v
z&j^^PcrtbbWMn7*@wQ8onK=Q;vTs0UQ2zGfg(v;o@bpbv#*=<aX!=P)$LPuOqR=d}
z0xfz^*3(0?PW2Dxo?N>_bIlE#<$7|rhvuFPx!EAVClj7LXN2a#5(8VZCwG5n?(xlq
zVo&;i!oE9eTORl_9cMoblo|<p1;UeKd1wh!{qq-3uC<}L@aj7_4)je;Jy|ab%{tY;
z0_({&HZ<2{p37OH@Z^~omM8mpcmutPXEn-`ZE--h#%R`wds58^NR@pM>5jvNIJW-c
zNq%x@^0Am4*X=y1PimQ(%|tz^&kRkS0NOZ|c@p=ACLV=@k0<4Yp()2;O1L4}>{~LS
zX-8vPI3$5REbBbyg(jVWNqu9gCu2G^BXVBwq?{d^ax}BEJV~z!O*#rE4xW^|TBU?L
zPR$xF3{5$j%~w20cZMbfCkfmP@ucnxO+B8$ldYF}(zk`CAI&TcPtt5?(rK{rBicEh
zY?p;*8-)u-o|GFxQ-V)jT*LLGT@sph6q+rbl=*;^aBR`uo%19-KOkXt_-JUZ%szJ~
zMV>7CL$gfu?ml^Pj0w##iRB33YglNO@hl75y6_b7d{76^>M_AH$@F9x9$LU8VR6Kh
zr8_jsICqBSNq$9ma%aX=_9Q<mG&u}u&W4&N{kYKdvjQjBo~(OAvrhExVR>>)4b3r0
zn2LI`TppTbEH1X-*03k_tkBe>*=)v>bWd2)>>9-F46Z(5rPY(WvvqD;`Sz^Ilh10M
zZ;p30-vb~Ig>zd|nD3<^PB1(~R<$NF(YvAO$u=z_n|DpklWlA3Y*RszVRAmTb=L8S
z0lwhoDQ{a?Ch(dd?2ve&QwYce4s$%m7d#~-0}^Mq18%Z+;Tv}}{(De=!qR;}NVhSF
z9^$Y`YIRtW#u)CM!NcC<(6r#2tH93yc#@9}O+G7l5w^+IsIaWrE3g%f!d})>(xleO
z6ZS@_C&z-;ImTcPxB=GWXk_aQW1WR4PnIiNXBo|w<vi&RM5Jen>z?%IwN4Kcc-XXp
zMJ7*{7q`wbnyp!S(qGa#{YWqYJgMimPCW|uBs|H_X`Ot;5iji>trL&oySg6g%fd3i
z=Ly_BR8RVqq3Ngimc%`oE)L6-y$NwTD{#}=lYeRJ{4;~Lkv-X0w9Y;!WMR)kp{F&4
zS%Ldmp8T6y=bs(CFX$m~S{MOv78tnD>B)O-Xx`CmTJK4^FEr`+<{>Xn`hNuVuxtdM
z9lR;*$$LRqS=l=f6pOe_!?uq-gu22AHO4phg+1xl1*A`(+3dq+I`12!*>Wx1sdX1@
z-8H(DlPLRs94Op1I@5U$UbyY^d2poOIicrff6ftH_9Ix?NzMIBH_6!!C7xum=Y)Wp
zW{sF;Dfb-_dR)NGcZ!&AQgb=oO=1X1nib3m&%ry4Zv(+4aFv)af#!+(a=Z9)dRg3L
zanoHYrdt}ZLFwk(4Eey{kh6q~tGT#<i%YqDjdlPDv!&SUFvONzZ@|z9{J9Z-*m~+s
z7-9>jufh;p9laGpZ29xm7-DOiuf<RY{=6Q4*m~q3hS*Bun=!=J5Z{I&wlsJLhS*Bq
zyD+pHf9}SgcKo>qf7trky%_4ppAX>=TLrrhLu_^H;}~KqR-eMqMfmd>{27ZspTnPt
z`15c0!&ZO(9Ybt|=PMXG8GpWpKWt&<8yI4XF5kvbFaCTNe=fwIAK*_0e;&afw%qY!
z45jhsQT&;WKR?5tYw+in__GUteuF;?@#lZ=XD9yr9)J4q=l}4h4S)WGKUw_w3;tY&
zKTqQin`J+Xp(Xh9JpSbIXBf7?^D#6MLu_t4218@;**FXh!_Y(wu}S6>3=PL;(=pVI
zp;;JW^S!wkVpF-3Ff<O|F2WF-b1lITn@cUn(B=4cC5G7aXElb{?4|=lo%nVwhAa$i
zzz~~5oQ|PY_-rGFreWx83~j~Gc^EnsLl<CZ8-})GsDPo1FqFj54h%6fcNd13A-e}d
z%wD|&Lz6Mx0St}C(B&8!g`ul3Gzml3V2GJ|*J5ZSK1*VVnP$BhI)KkI7&;F_ISer?
zsfeLV@L3r{^D$Jz&^Z`7jG-MEdI^S>Vd$k8T8W{TV~E)Xuf)(&e0B?lR$!=(p&ksq
z21A=LbQ^})SolT^U5n4&gdsL?y%j@jNO}i`F2J|%#1I=H-i@Jk`0TwHnu($JVJL;6
z4`Rr~(1$U^`sa^fh;_rC#Lyaidq0Lw!O#O3GBET74COHNMGReqp$9Rv9YYUci1lL+
zV`wuz`zD0oka_w{be}TSJ~M9nMC(ir?iksa>wm~jlA=ExIEg;T|I+qKgYTTLF1gg;
zTT?s6x!;#^9`pS(NI%a<tLfGg*>JwW<YYr{UV}SEG-mm7ai1=O;I7|aK5kDz2LIzq
zH3g5;%izD|z{MC&Z*a%?HSnazPrss@AeaGhXYSx0XS(nlyp4koo}=j3q>3MIK-WX<
z+%XvXd<9oY_^Fr12=MILn8cm)gBuq9i6-z10}k2YYY^Zi64w9lt2%JsA7l$Yc#@E+
z9~}OH?>Yl~Y2hI>IB<U}ctnB^(XeGJbl8bb)%hnD;4!#PVqe2SZ`57jL<Lhq@s9Uf
zaOhR;O?vXWA1pu%8z#_6@Lb%Pv8ug{0PH$;;hi?5g!u$YlQRXu343~NV=`pmZtxDH
zS0)q``UVEaa^llzj>q7NmFwjmq2P;5Y5Y(Dcc=xflV>wmWEt>1UuK?!xI@7dRDXu|
z`zibuA>NT@H#5>B(+9v;vExkvlm;%U(re&DM)-9W!}&Z*4Zn&2hY|5(V>k}u=vz;m
z^4#S2tz0vPjllK9#&J_vYy^8ZV-}C$NL){R?J^#lz+!*8eI`qpV6oI2=CK%#%Jsyr
zKRchrI#}!%b53G09L4L2>=`_U<8nPQbuW)?VX?kG9>ejuo|t(PkKq_yPh9dg9^1`g
zR}3y-G}~G1E!_)Q49Dww;+2o`SU-!EU%QCCo5NzSzKh3j6t5=|ALg;;>|M`;Jcgrr
zJ#p`kdF&z<D@|C;sElQ?%Wvl~9M9{CfuHagj`Q`z#ZU4Wj_dVAqrHvg!?C@dSbaH<
z;rL!pOfK;lj`8)xCvM>}9Ovta2j0bFIM&w_3qQwWINsM2|MnD*rCIEz1xr{7IPTXI
zKdtf@j{Wt-o$u!{v^MIA>G$&(S_AdO-~W%t&?2ZO-n?`vqk>jJJ@H64kD+ByPi%TI
zkD+x?Pt-F!h899San>7n46THEBK-i5p`}nyy#B8|hSma5S-gyufEGhNap^W5LtCMq
zI8fp-v={1$pS_mH&}INC@8L1D8|sPQewoM6cBm)L{0EPr{Q#0bw49ZIHbgz~y6^KC
z+7b0c@uU^(-T5qb_bMJk%c7oGxAbK84y}uN;(uPYn#It<s3&rx*RU8`8IZEVV`ypA
z6OVn2$I!;8C!U^h3QLK0Mm=%keLRMi2*~WKJcgD>J@KmF@)%kl^~BGg=P|TE>WLqn
z*1@Qt6;e;U=UZ!83@wp*;)-?aSPZR^dg6Q6@EBSo^~BdFuV?Sj7O5xRw~NQn9;qj$
zeS*i(VyP!S`4EqxT~bec{<IA&AKE5>!Ef*wS|#<wh6AUucW9Z^6TdvdV`!Z~-CxaP
zXra^-XWYSKXr(~D&+`~sD)q!ap5`&MdFqJ|Za9O{MBAmFc-f^qhV~1zqj&Ha+A#IR
zmXGro+A+{pzQto`%hVG;d6dV{o~b9kF#1eJ6K$G$;@hv_F|=ktE1%*qv}o#y&-|Xp
z(5k5?{$t`smJcl;sNXyuLp!IQ7`KJT(AKFZPW|?|EG1fX^~6X2m&efhsVC01&SUS;
z0;(q-y_v_*3aTeQ@NOPMOQ@ds;YWE4t)Y5i{eSToT10??1?Mv=Xcg5Hb57$iw2bPB
zyUcF(4y_}Q{Ea+@782mI?E>}=t)zP5JNtMHEhXSH&0}aS)f0zb!((VM0nXmfV`w$i
z6MMhNV`w?m6QytP7+O#D#KGidRsvd3^~AC_^B7uD^~8ry+rr+BXR!^RV6hs0;t#*^
z0H-X_lVxK9d=bI%J_a${YcX@#h6i}?%U)nWK*Et|UcHeW<Y{MCAzrhaV}H7)%UuKL
zSjSx@PI2B8ayj1<xNpy&1ms@N-rUBFL!kh+e$OYuoqKn01HU;=0qhN2w>R1I+t>^i
zv(9!BY(D^35b!eK>AYC^#Tj8QP7Ql;TG$I-fq=qoxyXfYV5DvX1Hg7z37htRLu2^w
z@Pp^38k692E9e{F%pEe1FxQ2%VcRm<kGBTUzCmxM?0bY@IW*?-@3G(u`Q=!2;fscq
zUBt6c1bpEX9rQV>MZ*Kgj6**ONjROrKb$w3yG__J6}-kW-@S0@b6Bmy43it+*wz@#
z)ShjFzdIIXm*5|Kjh2f4_3-9vY(l=*yFN3Kzn==(u~GTngC8!JI<_+3>s_xM!QYRB
z?AY%7?`PQghr@epiN5zcES?=-W7G7%eysT#+p7O{x%nC!umAM|{u(8Nt=a$DQ#EYf
zzSq0<cqD@j-v54|r|Q@R_}?G3>5qWou~YEBKX)W&Z&a;PgRgp_L#w7O$}DNNQQ3*^
zU7<r5%~Akv><{;jE1QS-tz5kKJ2shy&+NjNB+w4N0Pq05D&c$(o(i0}85||rpO~va
zhl1bzZH&Pg3xF-V71CCE%=9;K#xgR)J~aWKlYpCCIMo1ur*M=4Z}~!KEPOVv8Ntsc
zz;poL)Zo06316lF$7tEjK(E%r*RZxgm~)%m20`!wbrJjl26w=p#<<Pyg+$I&b}u~6
zUIRZ4L)m>i{Q>^xGXCcZ{^u(8r<T1M{yN2C1IV^6c62v83+XK~3lxsINSqHu6gjDT
z@AkHR&LqrM0k7>LouVg17d?TWBJN}dE4_7orB1;(&xxq>9J^rb{qSyZryUS#|9TxJ
zaC#NN#1tPBLg9Dn6ppjGh{Bio3hxX$T<TTH2Xz?231tLB&LLcHnV-=q6Q`{aWzMyQ
zzz^=T6W;v%rl5+4bhy|;=7O!9@a-nAEWWS91y0i=WO0%x3-PS3myw_8FoG)t5sWy<
zw3QNk0@%yYA9NVPMT!W97KxIP&J}xk`I`<exI_}c%gGWi(efFS&=^2)p(cW-1rkq8
zb@mx(uXJYUaMB;mi7cJqW5`~f+H`orrK$+&OcJCcbSxb@MYG+l*2aXdd-EI>_X=Q>
z4j0R#1<)o40DLw(7vVddW9VLPw&`$#>wyvK@HJ-Xgu0iP13J9mYGed2zMd``tELV^
zxF8zAkWV<$L42=p3Oc;ta%==It3=p{$NM{lL;GH~UZTU+#AvpBot?;*(7JEc;RM&D
zBLp<t77#o0-3b$U`?i5s8n^54fGgn<Jj{}KU{@HtEWBHX1zdiovcT^}cv-klhXq{o
zk6>Y@UkdE9g_ncR>u`Yk3=tem^K+0Fp!jv2;xp03&yb3TYa!nHKcd3`?xjT3f4am#
zNhtqUI_2Z;O+@+580WS{=>m&a8c*nOgS$r&+(>9B2*P+?r+D11iYQ)G2ES_Kt^WA2
z8uln|kVTX~(F+El=5usP#?82hl10I@n?v4;FV`s^cmIgR!*wKY@#}Po-x*!JsCRzH
z$y@%pI_2XAXGHbI!N|Qh<z-=q4hy)e8o`3IuEhHR&uuI(6IbXkk&R|T)C0fb<t^XR
zDId3wBZS~skz5G!y)iE@WgT8{^E!eTu|=}WXx`#))G2;Rbn)WAXy0)2GSJXr05|AK
z^=H@Syv4s&r+7RO5K;U>yDN6CyzwPACnoex_vo;K#}Oh}aSV5-@J_bQX8SJ`IQfJQ
zCwLShf)hz7j1z}1ft4@mu)@x7wq7Ba@7E7wh2K{5;^f;pd<>5kiL?0V2%*z^i_y!-
zPjndRj%GyCjpuS3+~D)F@;e<?@Z?JbOy>D@)4X-*W#cb8Y~Ya_DjUt)oL)ADkJB*f
z@T^Y+8_s5lgAhR#lXVIo7hSm2OgvXqy+SxihYvi|6d{Baen<pgef6@`uEP?ZtcqYs
z>Q>najG%7yG#yUxR96HiQqvW$>UxE8fetHpq%49JXW`WWYo~h!*vrKp9WL-FTLc%*
z3YyCW-?0z6FzjXPY8|%lOkD(9&Z=i1Th1j}FGCp}hVWQl1VhfeL1akiU<Pzpz_WxA
zEOhuAHdmGRj<jbW5?FhM4r{BTSz8;;ns`}R;O;gZ?(m>xgy5D2b0^#j_hRV|9e(hr
z=dtm_I>jJte?W&HJZu`lkFz5tVjKG-Jn`WAW)*aHzYb4$<TZjPXR%D?DM&(J(cuP<
z+(vM-*pDC)&7#fsJsoD)?W5NA_;Q;W?{)spmK)AL)!_(F9Y=`gln{=hH>-ZH!&x$#
zGiPq?m6d?wXLK0AL)Q_aaTwrzwe8%%+u4jBuVEA8IqwJtoE;#C0moy4mxt*(JmBf_
z2p(42_;7i^t4HEzBG^n@(9B{Trtpw^1XHUOm~vb=c!hO}4qqdq`Eo3>R>HC!MR?iT
zsKXYX`;QRU8U=B=?kK#BU8uttUO|XptX+XI?)1XT)qWkW@G8V{;EKD}@N#vn4p*b1
zxpFL_7T6NnU|xqGyc7~4B^6M)4n@3zx=x2Nyq^-mm<p)4+Y&EVx9D(%_hcftatyT=
z>gz=3C|;fhb$G(tIuSgn^zF<Ai<hapbeP%^&6KkS(n?U^=*7#@M|4<P7R{1Mw{E+W
z@v`+f9k%caRfPJS%_uG@wD<XUJ_Ol3ti#08XeQPRUAWKjO=qw>8?TUlq{AQHC5sT!
zx|aL}y36sh_e&l2dZO7oHHy6;Kih(WAJ^gV;%E+?1wf86pMnKW{;tCbUiOPnWoxJf
zJ09}ta@+(Ba~-cAM)0?xr7lAp8+jR=t;68zXa=2w6HJwY>Xzx0jJGc%MB}Upau9|1
zY4VC<tqw!X1xD+MxU+}O8In)qpQBSgbG*^Ie6jELc$D%A;bI*oE}&{9#7&i#q04m`
z!ppW1Fu6czLBR}lMtH;W@|)7(7q0_H@GEMu>6AlgTL*QRz+1=>OiZ?gW*0A%{ZgH>
z@xo?A*%#U0k-q?LDX`BtY-;N+v~`1@d2o-CNw#g}pXz{bSG8SOh8uctv9!Mpd~>wz
zfosVf``Kqb+jj2UB_3pYwGP*KjWmMm(HGz+F9booMW;ySBBynGPi)~g(S$<Zt5Yc6
zca5m$Y`=Eew$<S7w7|p1b$Gx#wGli_*qpYK{iPyI1x1N}QKw+Mts7DB;?2$n<J#Z~
zs=I2{cEEz$;bv8o!nbso!Mni`%#6Q~e^ds)n&a)Deyme2UOJ8_ckEWj^PacV-|Cc#
zcbTI~wQDQN_|H0};uYzLQb+CUzVLt`-C+|o3_iSc9Z}+xeau&1+b;ZuuwS*4bPC5C
z-4TT^-tT;&Oab5Xb(q2H;}Oh^;XREY{goPJ)*9n;#X_I&S~|Nf&M$Z4J#Frw5U)>j
z2ZeZ>nY$%?C5G5FVZ7tZ9Teh)Tke(+Z@O|vhIs#!yCuXspxiAX-oxZ>3GpH%cT0#j
z5V>1Iy#L4D65_2p?v@a5y>YjMc-M?OJjAPC+(98;zv2!G@#+(IP>2_gxPwBx@xvVy
z;(Z(LmJqMFa94$R#e};h#ETx>Eg{~w;BE==+5>k>h}RFeTSB}EV7n#65Ib6rhvT_h
zLOhtx-4fy%aqgB7Piu20h<JLMJ1E41%-lgCo+;)I3h{I=cTk9jY`KF%Jn70E6yljv
z?w}A)d~yecc=(b#D8y5c+(988NaSt_@jM`RONi(3xLZOz4#(XR;(;{omJm;bakqqc
zAd9;t#4}OcEg>FD;tmh-j1YHFh^KV8gF-wU!yOdj;S}zm5YL8i2ZeY(gS#cfBN5y!
zA)ZCxP7Co!0C!M`JMi2=A?|i_2ZgxL%pDZs4lj35i2JJCK_TvGatDRDp~&45;zk~K
zONbk1+-V{1U~#vExRu1+65@sqcSeZ2DBKw#?pttYgt%3}oe|>NJa<Nj>*3rPA+AAl
zmxj3d%iR*<aw~UBh)bN@Eg>!}a<_!In#bJ|;vyP%ONi@M+$|w49&u-cxJ<*H5#ovn
zcSeXS58P=X&iuJsLY!lBw}d#E<<1vza>*SO;;fE4D8%^{cTk9v9PXeHXBXT-A=>TS
zK_ME?+(99lvfM!-nupv$AzE$RK_Qw_+(984G~7WUS{U3xAr9}{K_L#f+(97@g4{tN
z4ouuZAr2VaK_T|u+(99Bo!n6>c4)Q6sH)d%8~1iGp{E9Zb-;UE23-6(J~Dh>CzxA|
zahSq@Q(gIxWQ{4dJ3(-`XS?2*20M=>Gg$?<t!4boO9{WW73kQ47vXxz3roiQUj+H(
z;7%bfhH4W0QJ^OqbQ^Bwz&G9)D7;n%H{9r020l0ivw|YL0@obrjkd3hi%_1-*Tz<q
zr|tZM;A_Lpf8H_WKgZ3#>6r4L$#U^gXd{FJHF0jpi}vi6kT==wVJ}*_E2%*yZCy=Q
z&{Y`<($Q6s<<F7@cC7MRa2fZU*|xTGZRfhS_FZPVZT*?2w5?yee#0ukOc@*!5jN&8
zv*+U-YvF(EI>7&9$NDonu>3)(CA?l!y01N^{LPl|>SM}}ErGXWu3d|{@lkKd&GK`{
zTs}DaVP1F0z&hrZ7~sL@Tfn|{E5vF<^_X8jCOK_(_4HMiKeQFLXD@&R;7*CV8EcLo
zw&Nt*jS_Qt^czU2s&5VwR>1_geJZ0bx@!h^H>R);Lw3Q5h(ezQOEU2BMzb-|c8rGp
zS20=kNhrx|jD;_efipYqb2pJROX!sqXpHL16??$b>zH&V1%B=zG|495J{Vi5fp^=o
z1#Wgh!hB9qEaZ+r+VSAVs>u-y?l6tBz;|zYR%26Fx|p}R%C!P~+NTQ_uDbXn$N7Wa
zTf4XqtS(*={uN62x+SmhFS)dT3xDiE;<(8%`vee05l-SI+h5j10cQWgUS%gt0<mTP
z3XvTnvQNXG-Pvd0_h|NS?Dtvr`yBl400Wgpo`+v>r<MIDKAe!Al^q5oyj7t#@MZ_W
zfk(~X>>$1jEYXJ_d#GkUvR=3jqJ_V$)gVyqjsmL_)lsa3g6Q&Ub`;B5{?Lvhd%Q8t
zRODcKCVr>XEM>a*!nXY_T@SQ(0e2kQ7ylFfsi6DTtbpz@(7?|jHE^3H-o(+9H*Xl{
zACsV)4TB5%*c--@V=f;?PUgYwg=-k0-r(|33GHTqm5FK=*<+Gcvsv`9{GrXFJsW2C
zAFCNVzPrkuC&EzZUBPV(iNqQcmYL1+BrEl(ru1$;ru@x8;1yi@$6LmU&9=KquYm6x
z8Uc@8=J>Xhr2C;`l1>ww_a0OJA(!sB3u^^%$`4<A69D}Ep^@+SWxZ~3l7#=?F-fQ?
z;cp#Neq{+C^NeH)>_60U7H$Jn!I%{V{nXG1dg?MbGXaHmz-K9Vp^~_tJSK59#r-(T
z|H5>(Y~?JGW7jQn&3R9N-`JVE_H5a@)7ZRc_l4WH8Jl-@@87?5zX0v|lS!kQEw{Ol
zi<=a)AVjd;g4f@q(Xg?^IfH}FSUUr2A*paWk(C*h{!GbmzOKQiU4ysUWSg3uLL+PE
zZ#JZfEdTKwZ07;Rz=b@v_%?w{>%f(JeX8CD*$9S<r0Oj+L8z*FZh)#e^OA<Ui5w_u
z_*@#<*fl%`Rv-F&_y4P!$~{PvfvR$62+Fnjmh@U67Otq*ei~V-UXNu6Y~6Gx?}3G)
zJbp2KH*+NJwejxKRPdcNVW=wjBtgMG{v}1foydfuqTj;EzA%GtYYom6CD6I$(p|f*
z6b9UHYU=S}q6h;wyrKr&z*<P!e2B=3qBb9-k(HY57g_!nZor-ElUDbZ?hCf=+`jcf
zLGRB1S30BbQ#7flw%-*k+HYsGCX()-B(kHZ`^RZyW7qwh0NwAoboUlP^(&_6seUQs
z9|9YDd6eo|MI_BHCUT>w`2{qxv1@*YPxCvw_lYyBeVSU|LlcN<gKul4byf*U<2#9*
zC~AB=jqC}i@m&`P?fgbfjbBd_$Z^s*R6^4D5h5pw8XusMja}ojd~F=Y!(H812s;0e
zrq1uBi9}WBOIx*ayBd<#?;-M{sP(&PWMkL5--h3f@)DH(V@;($LX(K9(#1tHA9i^y
zB%S|&$cm!Qze^)~!s&dUr1LRT^^CZYkbek_xRyHSwUBf^jL3?j&Y%Am*Ey<uuh7mn
zXzF||QP9U@uFh*A>AZu;ilWX})5yl&&iz*V{%u=Cm0zo=@@r@UQSI7WTJ5Y7lE$wh
za-yj5%V}g!IE`PqXWMq6k-tV$<8_)qj*G@wB_xgCLgYkI<FBNVja}n@t9}2Ty%%lo
z7IglArq1uDi9}WBEv<ID8j{vON#sRQ>mQ?$ja}>ho*l-)-Ga`4r>XN_(*&Zb^Oik3
ztAwQSUl2J_)c8+nWKTqmi_^53)AUUCX^?*iO!g>^?~>PJCKEYP)c6D%+1NGiUv1yL
z=dwN9V3H%W^KMP8pGy;pYCCVa+U{0FYU^haxlz>onKZH|oaUipgo8~7#8JJVsri1I
zP>ze{-HJ$>PZPOO)VxI_8@uLb`_{2QU{aHRyQbpbLX(NA;+I7%!m@fudcU2>jH2G(
zKqDKw-j~XH-*vgMfAd9Kw_Lh&t8wAJty_g||6xrGJVcX^Dh9eDFd)@Q!o!0^#uV}J
zMH<=I@c>Skyl102VbP-n8lKWb!;>`WsG^~x6&mCkNmzKC$d)1&{y-xef(7cmpys_I
zetwIZU-^jb5FFHFPIS7<eYL!FhP`vT-mGH{<R5}rhj>uW$G-%$b|MpsKwH7co(Ko^
zmWUGQgbM;z&Oql6MzIXeE8yy`_w?lzno7NtC{M+mfKH)5iLR2Q=Y2%J6!pA^Mpo)j
zcC!3Ke}wP$VZ(+G!+&sdn;o8C<db2pST+rg4G9yk0w#4%^xZ@glIj?<rWGdq^^pK^
zBataZKwM8F8yg@T9L)3KfJ2~qq$GNY`?w|q?xTrC6#^%>f`F$glJ-AD<VR8a_tMD5
zuKmS6?ZdUP-b|lS1SctFa7iiNS^JqL1|FpeM->C7w8DT?AqfdTCUT{Sghyy(V@JYV
z9}>)RGHpoRz{DAPBf&VxKLjS9c(yyVCX(*Q5ZO`G{YV<w*mXbOCo#LxFb82?+i;_%
z_D`qDMYRK1sn7wix=8xpKx9Z!|7&SvW7q#epZ@Jnds$^RO%SFBy_yh6(nO;Qfz_=b
z;H!>=f@_I9DWc#S8rj%U;2!~!W-@IVsf_q3zBg(D;5M36Q~}U(1ZdVo(*0|Q>?rEK
zP9qz;?){4dwTfl*n&1=DZ976Q@Fh(Ue1Rq&RS>jXBnYUIgoX!*d?}*gej3@>(cm8s
z@@&bdYz<^8=-th068upU36If4ql$!<<AJX_5(<7t<Vg_)zowCm9R)LktO7y#^JnVK
z2IfHiA(#z_C&dG6A?bW3krhRqPot5IUFY0qzPsGtSuMlY8jP})FIFuh3EhC8|81K3
z-$Ij)YU3C8p8WckY9wKyo5+?T7S5%SjU5aASz)z^O9ClrO`xiYf`c@<sG^|dtk7E*
zN&f{RLyG$Er;&|a|8smT2Mwqc>=FoFz+IZkzmp~uRppCk(84MrY5wg*ZWJ~D78=<T
zQuDZ<B53|QnwtM6O(@4r^Q<D0<{u_<qp10ZXk=s8ynhV<+x$UE_0MUl{u!E7R8`+{
z4WL;QN%v0?*-_N}lQgoi>wbo>o9xN>d=|FP(lhs0K>i^x_r+7@L6wj+zLdy`qQ)1~
z$i}X5|K9JxnpuD)AOGI(rJ8!*M-z)`b8or#>#2&Q{XIl}6t%yTMmBcspX6&QsbZ}s
z2i-k_Kp5$7($xQrG{LCqzrEF(zfc(o0oM~bQbfQJ8rj$pFw2Jkc4nq5=>9%U-G7KC
z5>?%c_XmTkA!+?yA}@+szlTOPcCGt+eWrCYS4xZa{iB+i|1nK0s+w=v>wBsqY5x%-
zKZ@G_0gY_z+Fu&fSgPQB7`~I1XTGDwLqFqY>luAxApZ~;ed28cg*r)i7)fMI5f8&?
zWMjvJ-wI$&C{-?&iZ#&+I9(G68)$-2?GIX70YYUY1gs@;q=<kH8rj$pu*}zbD&~M?
zn0<X^s}H`%oH9^|f=!UrM8dT+0jVNkZ7X{)uu>8tt|4-!h={9bWMfCf3LheH?F2p(
zZxro|Nr(wEC1Bz<O-#IoCL&c#tZ#*hkZMV&s1tcpM8z#MvazGWZ#kg&df^)_X#>7a
z1DAvRM+3f~35f@2@==9EOUogsP7)sOCo-mphfmVT#*T+IzFq?MiVH~?G2qOQeUSvd
z4QS*n_+Gep#pp3jVEm3IC{<vb*{ZK-Svd(Izb101h>%~<$i|Kk{~e=L(E!8JGAc<E
zKIqsd3?Osn=*=@`LjEC`XSBRy<gbr}gK0#j6mc+_MmBaFEb#Rd3<RYN*c32m2q@U1
ziGpsLWK_F`Rjs-Pu{IJ0&Ly&>h=H?cWMjvG|4bp1TA|t>0D^;>ASlp8qY8qSX9|7Q
zkx<Z2<Vg_)X&TwsQ4n;d1in*$rqBu8sj2_B(}bd`|Hv~XtRj-;-$LX@QS-Oc$i}XD
zd7a5j8JU7{Dr_?g+W)4e_8+DRMpgR}>r9@?NC<d{$dMuf9;A_t9RdEkrg?U%JY<gW
zj3x@6q6tS81ugHIN)?ij@FbBdMI<~<BO5ys{PO~Pas>+rMg?vi3*Etrxq9=0rI7!H
zo)`G*BjI2%ktszSETEB%9S8m~p=72~FfWuruul^NduW1D?G0Lv2|{Hg1neYoq=<m+
zG_tWHz<<WvDptTH4MTy@3*4v)fa_^;Q3XKDGv?m9NculQWJpo}12nR+>)*eooarkR
z%a+lTF)Kz-t+&@Ii@VVu(!{~NH2J9Fpyir!P@N<^+(Tqc5f69M$i|Kbzhz(;r6OEF
z$`mqH!)RY6^a(%K#KI#q*{EWn<yw(YAKCly4~Q%&qTstUvLPs-daY}Ez2nZ%sQF{`
zdN=K|OhsI?6p%N4o}NuQ1@aGpO}a=}rIf26bvhG?d?><g93y)o_`X{t386DrJOuX{
zE5%wlDNZ!b)>P_7qA(R5--#O#5%rNYd^(XSMGbGDk(D}>wJiV8d%bg<oCV+@2^U+7
zdA?C8H0})WrsM3cmnIw4VI>L&LVYA0B#BHZ;^0~u+1PP#n%sU$Wvf~?GX*^8WET2r
zW}ju0%Ehde!~+eof6O;&!sCrJk*UJtg4TV1r%E*?)Z9koRS`9>p^=RpH7jJ)*sX1V
zIm8I)M82$vk}uIjq>2*pIFwqoBvgEX$eSW69-xtp9Tlg@;>+Z*_pl1+D1rOV3Usmi
zgeExtNRyN*I5tLfAuVbrVdOC)i;5We9gS@47&%$S$id!XIo}E$3+L<EAM+vq5KJyM
zM4%(EUJ@?m5SdfN#Y`I6*m2P=<06-{jnQz|+!t#iV;fCMs{O`k5y%LsnS_lkMD`T1
z(M=;8J2sZe*uYb%V3X!cRfD;*6Xq+2G*MBd$ww6x;*zLBog_RQBr>LmhXRdk?08rv
z%g&1hd4}>HO+eg56Obw(#OoO<m68x~Cy_HnM7*6wHg-fTmW7wgnFA)fmEk{E`#nuG
ze1|3+RWyhPDWWPQA>o@ut`w2*FpX^NNC;ThI+UrV4cizK#)*Gu0^vEDY*c{|wXP-9
zN5a7~M5YvR@Dz<~>^KNmp%~yx6BR25D_O9#b;K$Qa9DGa-ZY~f@(;l@BWi^rs9F*#
zRuFkpM8#4X+1OFBCaha^76bXhm25#=p^1%4X@XMiD#Qyql*&m6*+=A15g~hMWMfCj
zYFUWvqHGjEUZn|;n`m-U1;`l@GttnxN%**t$e<!VuBVX=!3T9Cy}1$3cV?;iW88?J
zy2Y%T6|3rcq!q@nf768F=V<Csh2cU0hEfeAuziNeh9cNL#mJrr+wcqhGU!ZkwaP=!
z|7fc8H$+(~?!t?Ae_B^Z((f;cJSpn;XEd@>XYwe^|H5s;hoeCH(9Em_ddBH=$Ug+e
zsUix3zcn+3$de)pCep~pj)J99<H=MEdpp(c6@)(i0!=8KN0X0gA1^L%D%45p70xCy
zrih1)G_tYdVTB|+4u;Hia3Z7m;TG>VCyJVo$k9Zk3JG!Qs#Yxt6&WILim2$Nk&PV{
z%O&xpY>Of6+hp(11jSov5>f@lx|X(}YX^C2C1K)CMAj5B@kSci*fDXEgb8+V%jnJI
z#FLy~*M!1XX@XIOf_Ph~Rb?aue3{6RA_BfdBO5yc+9aV3fFlv;8Stc@1cSe7!r;#|
z>8Qd${5nZwjU+5QL1ar23xA}MjU5ZJ9f>!Id5=&r^QD~c<F(5c>P;!yApa0dDI)Ah
zxlR%u77`g##KU|V+1T;0L~2PWI}17pBM1(AHQ}&}CLYz!LHvSUbd@ADTukIk5e?gD
zWMfCed`WbbYS}XL{@w8#G+}TZO)jc15bu??sEefkLqvuY^<Sltja~n;t=N+*Ci~%L
z`E{bL_yJ7-ypJXrRRBcTik`|y2zU>XBSi$<MI#$K0%TheKWcxdoWT#-;g@ZNRi^*Y
zguxGK(ouy$gsteWk%Wcs5!q72!gpw7W5>cNQV;5xH&jYFxLjPa;67lX3g`C)T#Q<z
zXEhFo{6k<hibwTHwUaRN51KL?iWqs0MmBbgtd^t*o8WNJGHF#R=#!*S%NvJ^<$k|M
zf_0h*=^Pq_h+E-=x=Hw0Lu619AMG@<vEyT<B*VUPu~ssw*5NAnsxYhYZA{6!c(Eop
zUPKd;>acQZ>qW_sib=@0g2<gBGA^Z&jU5>QYt@;GiJwyP2vQhTUabj>TWJzfg+<g_
zbwI5oOuUN7nj$7{qLGaq6SDQ$Z&@V+ydGOUFc0k)z1n<66Az!FNk<hA5!R=_MiLf2
zPGn0F3-{5;#*PJfEuw0HPaA2y`Zt;=_$5s)swjw9i)hwG(*Mth3@PgWQ5xCU^)L4Y
ze3cr#3$ROF!q_l<vEGbh3gjPx8An84;IEN{g^5JA6tOUlMmBaVoGe+9u=5M+5*5CO
zRkdK_0zRQ2AmThtM4U~Nk!rsnUI<pKmxPOrMCKH6aXO7`?6{Eo2*FdB&+}PM6A>Aj
zbW{-`UgePxA=F6reb!54N)ZW38rcveP>)PEk6iOpxYYbH9=V>stCp)~!26k*^Q%zk
zmfoug!gtdYqY6UNTxn4ksVjOXks(C@zJrlH5sqQ643|bHcka<!%z@_;LCyb9Q_bHZ
zN>%Ydwz!0$S}{rIUng>>sPnJV$V#2gms$Q7?r3&LqyWuI3GWfgUu^k1u&U!O?Qb+u
zsdfXKge5Lo^(2)1mB^zaO8!hE8#_wcTZyq^<Z;{pANXcY=s;Gr=>g<q$Ug)V8qsTs
za?K=cEF-d~h>bQH+1Rns*+PQozJ!l275LC_6&z%V-;TUY6CwL)vQq6r&Qk3{g6b#X
zWG|6PMV#!Sk&PWE%Uj^20xP7I>Ji^*-dAX%<7G4nsiH$%ELW<Pgozu7tSMsRIvUv!
zOi-sR%_$GJc&YhgobpV+5Z2|ZwSrX+zGwP*O$dILrVv#KE*3hdfI3Kk`!taeMZkT6
zkv$QnILl=zbh0y`sjv&Dztz<3uZSX5oZpDui&Cv5ZU3CenxeM<n?_dZUVg&zzi^YA
z_5kUXtvno_GHL}FyWyHyF;~kM#67tgOZ3dvsgQrj%~s`_N!XY~WKR(r<7s4L$3|O#
z1bgkl04}GO#ZS|nuZf9sXwp#~2gJRZ$Qnsl*hFMY5esL~$i|L^H35=i9zz2Bf(tX;
zye2HNG(oAtLUflyshot6J|c&T2uab%#*UDa0)&W@zZ{&55fJe<O+>tzCKy#jh^NY1
zRYpR<Adw?Q1iYR`Hg*IA*%GLI!-UTT48YDUenkLWt%<h8*EHeq6`GV(;UF%71;8P=
zW)e33oyeXdHvWx9Hg;^B940~dF0fT*7tyL#Nkqn9G?DQqnv7JDA@1!f)=R?0{}Gu}
z#KrGvWMjugXrIA*3d5{c%bA{<@6^YVrF!#^MUX#^`3LJWq?$?CIEl!fA~xpI$i|M1
zl>u!FEeekhlwt{XRRrr|w<a`p(1fJgX^300suh!vaS@R_MPzKHk&PW0%K`-05*07e
zM8#p6fK*W-y8TwEl!S;Hkuya^lxbvRM?}y#;zxvGmit5t<NcbjcrQ&#s<4Q(J%ei|
zVdLFI_7t)4P8!+Ru@SWH%?1u5Rm<m(7(G_iTqg`7Khng-_h}MR#YN=0w_Ga;6aSCM
znj$8?MI-w!fr<43CPpmNvpN4slhBZ1V!eF2^zTI06fyBP8rj$}5oA^H7G<W)d_XtR
zr3r~sApa0dLn5sTsZvrev5Lr<A|g(vk^NUdgxE=3t%-;$X#!I1B&di8=p!y8a;Aug
z{WP+%BVtj2k<OcwY(*T^1jEfV*{Fg+yonG|AE{?}1(7L59K4K1_5{a)*eQHk69=E5
z$z}*}5YQidl*p7K4n9mHdxGO&gJ41aN)rb^r^#jraIitP9sirilp+p(LL(bH4wi)W
zqxoVUeCO1}Rm!Q$^(GUOApa0pj-rRX=qgF0!+0WJifBmC$i|L_6``U#WM%r&RU>63
z%_BnRaE>M-Hqk_++Bt}RyVa^Cq2df8Z;Gfml}0uM71X1+&7-{hWG6Mh@=@OGmWjiL
zt$O+34TJ63ZDWQFOK;3x1b?m>%x({RF>i86jvbJ(F>jxB$SkMe_M(+L5PqO{Kof}-
znp#wmc(QPyH@qqm`btE86w#Nbk(JOl5p-4S2|elAe%#eReJgCU*UV}$<RJImntFd1
zO(3dz7Y}j=)In1C9YjVH75+9x_Cz?yy-=1yCwTA#Z@y)-&@T)t-_}&>H;A%SJi#q~
zET(mJB>jGk$djUezd|D`HRyk5`9mA@ac|E{0>lRTPZje<C0*<7^<5Qu7TD9di~2N8
zJgUuGG&!TIB%$FiM7|W!@FyDC*wL^cKy-Lm;t(LBN9Yn(uFwO*a>zdfbB|TR_};QM
z5(btKSyIHnA{yD)F|a5=YEF+}S>nCY1DYtfgeDu+?m%3ejHr)<gWW`?6mhVFMmBaF
z1lb40TD4SziD#xPbOtZiguzQ`qEUrGq<!G2j)a1j5P4EW!C@NN*io=Fp!MW4hv5`9
zIPXoeFJKE$_=qMHK1h>~Dip+zfGE^S!o&NCj49&by)?41;~~gGNP*)`a9-jY8~#fZ
z3I9oxj4Bc$Ere!mBn<qB$dV!kzE2|?I|hO*1hY^%WR;ETk&@_IZ|uo>cH(HrKLmDS
zq=n$EkA#B}M5YvR@XvoS94rgyN4dXEa57P@C9Acv2!_)%!LXhvcg1xC@nC~Wr6fdj
z5jj&t#3?kgvG)uo2MEvN3x<~A`J{^XTcbUipfG4MQXMD63p$GRl5lY~kvT<NTuCDv
zJ1*u0$S-ffrJST-Jibm73JscARG}bl8iZFx(*98*KZ@GFnMO8t?XM3IS*Flym90V&
zx&`;c4)}=}*wh@zB*kIl^O~^uEKOdjusAPjttqOG5?(${WK<C^pP-SA9WQGG@Pez5
zW-@8zEO3xmEE~P>pXF>t37yFQYNF-0G-;`#<(w$Aw639qm0uCrRK&{9X=G!^O3+*c
zeS9NK{O0HEm3niL8IV7Yxrnzu5)P&knNq~TBpTV+aj+tw&w#H>S=9{tbaBZln7OKW
zSZ1>(9L}eSNVTI756h@kOG3ptMBWrpv57`Dc2orIZsJF43)Uf{TI{!c9;r&2h{)5#
zql$>g-A!MWBs64+d?})#k482G4b-Dg&7)KNECn^c^3kc|e|TobcE{=sIX!c)CIIiD
zX+#x(ZG!C*R0#=kcM~~LgxtFr*%RUT%=`c`bOuabexRw+?-B*6xaK7~@o!NVNweQ3
zGNh>4Z_vm}oypf&{ul1x%%UKfg`A%n)~;uGK2MX4YO@v>!6WJ;;ow;!Q;IlvnnpHu
z9JB|?E#vg_p4AOulG>q(h}DpP2&N*U&j96`N!VCPWKR(r%V}g|$3{nx1P5}tyb%eI
zt26;}IZafmU55D42U_(clpG-PsECqFXk?E&N^T!MZ1^zzC!e`*eYAdYkUYJUfA6-*
zEt&v%B~41I068VfRP6LtNP@)6iCif{;-xgQ#~l(>B%B<Cgox#gPig|<V>B750z%wJ
zQ>>SSi;obQQ^dswX=G!^#nK@8`DXv#PU07unD{A8KB|}yPh%<6Ny5W_5gAj&!++Aq
z#*T-;5hLud?&MW^vycgpe+X>O=n=zPAqfd%iCifnVKj|w>_}J`)QXx%*}R>?S(-RF
zlO`I~PC-1a+qyau3Qi;Pq=<s`G_tXyU|EpZS{>I-YhuBo2}l(S;)yJkN=b<5A#$dO
z2!lp8c0>gC3q_-1!gr9xV-~k-V&M%m(WqiUp<i&TBcb4RM4l8;(4dix9R-2?g73hA
zus-#mCKkR(6ObwvqWcAZr6fdrp2(RZB0ftad*ULZ#{w^P70WCq(*hzM*F?l0XaYLE
zi11cQLd5?PIa5T$Z)s#>M?~O4cA=Ix>|0V5>tGEJ?1|2A7p&HsRLq0?Lolg`UdRrt
znS_nmMD`T1F@r`nc5KWGGQq0Fl2H=+i0ztSxR53m)jmQzzZ+f^N&B0L{3vSwd>Yvk
zSNo!uhyhLQS7>56zS{4XosyM^{3vQaPa_+<_5(Mu%v6dUtOIW$73>GtVJah&5(kI7
zH4*VHny6F}5xt2OUOlOgxP!=}B1+yyBO5zPRt9yeW~r1rVkFaM)xce!9DKb=7%#r7
ziH~p7grtfO@iQB$6_b$h4I+1n$oLwKZ0yKbA0$9Hj@y&7(1RWL5%cVZm#qqX<6Z#C
z^O_)emL@M%kchWGj-iecUY;g0s)(1r(8$J)m%w>Q6^`n{jSu)<sF}-IFz95i6T6bt
zYxL$JD<OX@^N_%rN!VCUWKR(rOK4<c$Hv-L?JNVAj@ehfQq2xU=t(Zu#K-}fv{ZW%
z@n#UYh7wjTA+o86mEAP5v18?oAW5czKw{upK&DWsn$q&dD>WhWa+=UoA+sg&l#NP7
zCFH!6$gLuBUP2=qJ8~8TAqN+h@#vn=qkK#g9v`7eMim~SkCT?QkudN<B1?)Gct4G7
z>=+1~^|GUzfdjU8fBmPLp!hGEs8m4_J?jmxo`jPBB=V?;k{{8?#*UJ%pti-t*jK6<
zc`IKmA2DjxOfGYsiT4nN&SSzUdM4{w$Ug)o>)DZGOREY>NEuD!QV}U5Xk=qYN?Q<8
zD&~M?@F}mjaCW99G)|*QN3{#-h=hhvBMA%ZiEJrip^HZL#KeLP1DN7g#f?`>6AL{w
z=^Q^S2sM(hU=Z0-#KP4yvaw?!@MMNzl!}!qOi40T!)RY6j2dsygv0A-0#b#8ct<&?
zvlA*M@3u6Ed@17LD2;3g9;kPWn|GA?g;i?)81E?0+-LSyyK|Yof)#Q}`J0*ue3+&Z
zRRk^(2B*MUNS)9_L{=0*_aGyCBHU7586<~J`$R{XeaS-}ah}pt@RLNTDqdL@w`Ei-
zCh7cfB6o^9{{xMz)Y1Ge%m2dNUp~2&0P~rGk+TYYK2v0Ahu(~3G2|bD8H;$uNU>fL
zE*21(Q^dtQ8rc&Z7vMW302g~Saj}yoBh?XsfQx3mBwTDKGN*`(3u$Cy$3^f+f-C1`
zFc^$X6+LeGmpHE11jZ4XkW_)8Fp>mROhU#0kvm0XRA^*lN5=BN9+#n_>Dy75!`!P0
zi+gAiQia93NDH#lUn>a{cN1At#KgO3WMjv~>Of53NioOhw8}~NqNUj<ZsR|qiHsl6
z<fMv>Goq0ZTsH|H-z74rh>vg6$i|M3Qv>ml32>Vd<`i~hr=Bf34Dt_wE!rK8l&D%t
zn0cP2+=e1%o~4nE9W(8Lm`N7lKv0r7d4%&HaKf-p^zyh?6DJ)*gA?(Bf^y9yY^)}-
zr-+S}G_oPspe}ngm(BPJ3N?R>%Vx7KfV(m2yjkwg6#BZ$eKq`Ev$&~~(L`b|O(&{j
z+ft#&3a*9(yCjhpMc7@-$esubXG;SG(U~%<X2q%+1v4)iE^pLS?rlVYDz2G{o;DQf
zBx(9JM8*^~U8j+iI+$Bn{uge!Y;~~gz?GNng&d~7u;DL#(%?(Lr_Kj6zd)0dYNHno
zT0-3<d^|v8P!S*Z)5wP4gF1jT2O!=PQS-+*0L|F4cTdZ0o4;uy@vk(Es3KAHEgMt`
z337iXa-s;iCm7iiVenZLAcnASGcW7X>tNa-{}9ZBMBkzj^^r8ZkjRvxhUe4BN}b9a
zmj8ttbmj)ht;edG!fM)H;7VtR*+r9zYQtV3*p;C*k#v7CksU?dZ=;coUH40ZB$fle
z3<a2Az?!@7D%}m5Ah?bu9#s&C$3mj3B%$FDkuOCwRB2>mN5hIh(Upx#GHs=5@KGH3
z%Zne-M8x}OB2q<!_<0MpYDuVg50N)TRNO@)8#^l6g2Z>oEaz(_xRwncWfjJW|Ih@)
z4{6d-1%!C0G_pn#7QRPhOA!m-p^-g7v0!ET((t`*D`|?WqNCR8*$l%W{}5P>3Rv*g
zNW#KDXi8IEDSwVeHg+s54QfTjQZ<u@YqJ4E#X3zibPf#~#KkLxI!Sm~Lu5>`Z)m5H
zjU5kNL9%OpY|(&y6FBe{vRVIPO>n%3CM?xqL%gO!t)PUID~MbwBIQyV+1QcN86ikq
z><k0St2KdgD@|6aKshUN9_ncc{`yHcc@>dKMV#D3BO5zT0(+FcvYE=jwikY=vzD)c
zyBAy(mkrU+XaeO^G-0U%CAvpxRY3_UA189Dh?M(iWMfCl>YyIQ6(mewll`S)rci}d
z4PmzO8%>b>k|rlrkcb;6gt|%i_!*HwMSMI;BO5zD&QzD7fp_rB7W6RS5ngl>G<}`k
zd}a#dAA<SJ*2u9ZxFz^%Dq&|LkzGaXjH8i_9Xo-mV)!|XQn{EfRfCSIpQj0wvuPqy
z?N*{!#pG&9sMtv4O%WBR)5yk-ioo476kov2g`6fFGBnw!!XbM1%u^o;2fak56mgKG
zk&PV(3xh@$l$&gkzEu+jZ=#7t6$YZ~`_|QwQ1C_~Pl_nGjYc+h6a??175n;d8%;*S
zS2dCFWtw<Yk)W`TCR9m6!<UGBDWc&EG_tXyA#iM9PAg<6{8<wUPtate3Weyg!BZaz
z2Y)0orHF&aXk<@J9I(Y783%3a_2w1}A^#A}EuwMYsgHz%`9!7^aWIEQ_Qb@2eLX&v
zDGMtSyEJidF-<nduRrkAN5a82B2$Vu*g_*4I}R2E8CSVtpV3>k4hkr^P7?)(Xp&Jy
zf#~wDWo;x3REaDpV&EW+Z0r~aJWWwC!3hugFo$(Y0SE8X#KC)L(ow}h^bWdEBMA$4
z5!q72!ksj-v11`{?g$^?EtbKReNh0y4>f`CJ(_4#fe<}+Y*t4?!FPx}DWc$;G_tXy
zAn^Qc#j;WY0ETbSvk(74lZq+;qR-#5nn=2Tj>wLp?w_HNja~PF7D2gIkgtAqYU+Ot
z<R60hL9|8S)<(iWJCP+t46LA$jU5Al`wMJV<=g|5yfwZ^6AV|-M5NjoMDH)i)sj$g
zDUmltRP3XX4M7F<s#f!=7r*yI%^%}cuUTN;53Fx_*=tY}hOejTL=}eOUE|<tNU(b?
zkrzeSy_%6d5w3gT61A*P@HJCN74ch~K6~TeG<EwqqC6F^8cR#m5ml1({23x&ihBMO
zjjYtYe4OQf;Vylh5{w2oQ)R(D54iVZB#VXK3@lA#d>?TBJ#eaX5$u0xl2UE-8~ukM
z0)P=xI|(DdA+o55kzdlt#*Ps@;2j`Ea6bXtT5u`I^dV#Jsd~tm1^I`-M3oMBN7YHf
z!*n8Jig=hpBO5y&Li!B$k$2qZESZ_Il|ugn>7qD!-KvR<3uux$o_$70?IetxM`TeE
zBWKgd#*UHY!F>il6a!wi3TCb<UJ@#6VxvftkSaDrH+@RAk}#1YvZjcM42^8;n7}3P
zz&<zdu+%#>@$e3sbX4&mEqS-Dk%Wb}64_G3!kcJhW5>eEa4du$r~H;CB)(1)k}4!b
z2d}CXlaTRMB6o_&_%e-b?8pe2pXSX<zkSkHSP%c3CMNz$6OSq;ROYANDoJSgGm$Sv
zG(15g8#@~22RA3lEy>BJ>CHKoLH;3_bF5UDjk3B(`fnpLq^SReG_tYle{rz>bHz$U
zc0#dV69ju{!cpxDPEi0svqBOQb`iN!M8d^1vLQ&IE@U?s!ug6QHGhl?;q$h0uk!;|
z%f;}9CJ2wx^r8yFlLZilS4D!~%|w0_;r9wg_C#0^UlS?}*8wMPycs$8ny+Abd|Fe}
zpCC$BacTTaHBARpPSX8Hi5x2G{=+n~QrGhVmj8uYBwrnl5d3Od25zZmpl>iTslx(5
zeg%B$ERp}5CMVUdK)h8-sGEe3|0Xi1h>xGp$cEs9I<z#0Cf+wu^T#+eEpU|x=2FYi
zX~F4wkemnkhrmu2kB_vhjReBkM3xjmID?Tr5eBH0t>oc+Ud?}g{6bAdZzf7rac~mX
zN>nQ*>HK^mcZxbchelTFcs8;8FWkU%Zc72$A3w<#Qx*(R0bjci2CEA2t}|GbXtGls
ztab!jGXecvXLNlf9Oa2jE8-|iBl|Ckqg*lBZ&bj^mw=;pY2xS(n(SUc95w4J;plBd
zrWJAYW*XVpakQajf0QcX=f_LM0yx&_v*3%b@+aKBtqGoQ&;+Imp7Yf^p^!>Si1`|k
zQ$@slg+?}Z#I&|;K$w+evF!KO^Q<OHo~8*&6(t1Q##1o~8Gj*ir-+O{(a6S*j8##6
zNvc#dO#F1TF#N1ML(hy^4*7>*f^#~l2WeJKLdOy!e~Rc>L?au54(gP-IpyZFEoy${
zDR*|u#9_l$y?pS7!S?L7F~f$XH)bz_Ki3Rqw}-u$H#sE74#?P;w|D#Atvk2x-fDF3
z+ji-$t-B8xySn%8-M)L9fW~Vz(Rd9_Evh}0xELH>6$yP;5&2O>-{my268a{Bu4+A@
z^RD*euKr2gwQ4c=!&|S>RC%2y3ssfR7djx3eM!x?5P48k^D7zI&?S{n{T+AptrY6t
z*$3a)DrHJmE>o}!IMrG)k~y<dX<sF@<@+@i`$?iu6<1Wm%Qi|C8<Z6OF(Pk@3jYX=
ztkkYQcue`5qY4Myzs{&Ku~N!qz@sTZ&<DOJKoUBuGk>H>Qngo^BlIdw?q!5MMr1${
zVZWo1l@P{94k|*L2zz`Ijy%--O}ph4$CSTWe%~?WZ<e28`OWk<&@$S`O`-$*&t?42
z75vXt?2j!!EFZ?(#HeAz=)L)|jK<97nv=`-=8A`$jmwG?5PIpe&LlOIV4!2g8skm)
z#%8h1KE9aVD0P`<LP3Oyf}MXeNG`h<WSKpU<qs8K`|%tNnfS(cSK)(!J+-POeyQWC
zA(2>P!ZNd2o+Q1?HKljpnDRGqe+ifVxnb5neZIhCoY;(pgHOOW42^)tE@M^-6edac
zx?_?~6Pt&QDZjFGkGTh!va5*SA8OlNEL5d0-o9&S#5{EwY@k4)9dHqj7b=PSj$;y6
zQ`~Q3`Cpi>p!KwN40s{xvqVo+U#7{`md;$5lJSo2OEl>y_Q;LJj_G0Fj(3Lx_;w+y
zgQTV}5E)T4K^~xywb^Pmmit-$<JsM}*72k*-3Pk&Z#`ffxN`5-{leh+1hArIWzs~U
zs%m$+LQ=I@6-ncd5&2Qn`0r?BW7qiD9b2zt+7^_(aHF2G=R<yi0i>$zC86!wt%aoL
zIYd?z^*obCHg-KvLJ4i(z5hV>?#;fIeX*vpx6wqQs_gKVEmlR+_!c5RiW={xk&RvB
z<4{cdH}BcERcP9WG__r&i9uD{u0sx~A8%Ge()2+hFN&Hj(8y9XeJmz!>tQ>w#jKhY
zOFV-79!*8xMU#T6qTSDIONw?XA*uOJA}5M!emf(3BAC|G?5eX>3%0~3G}-THYVw;z
z`30^nOY0sD*RfxiCOy@WbowxnCq<n;L?bIT-3M9z7tY|G;%L6dO!iv^(Ivujz>JQ`
z{R~YQsx7!Z)C%)gM$-CIM2-}-{v?fT>{_4W6O>`MAfeT_Z_?BL3dlbMHuzdq?Yotd
zP_UH9nIZ}n)5yk-f{6|asz*xVImt^kRlkoW3Dt)0I`XvP+FVu<Y9i@-50M>3eea}^
zja}a}9Vyx317|qZw9w>l($xEnG>NF{-Suk`s&}DAlK!tJvZbj1BQ&zH>)*BV4Ez9;
zVfeOj@6%NNhiKwZRetnnZ&yd&YPgrkk)q!3p^=SU?+cuE&z-lkana~4mh)!S2Z2X5
zA@F0GWK<!rA#%u<YbK-N5h8PnX!rq*Z0u;5;h>=xJ`HH3GD$xO#+{{S_l<%4Ltyts
z&-UC3$+{m&<VjKY!)Rn<*Zq7)_sOz_{xe;P`TBv=H6gHpCKuIy!1Z@#5BIYD=dYKH
zgtbK06p_$DBO5ysTw9=84C(`tngF<#CJt2qMB4&(b!5F?L*z(N?^n^to=|#6rxCt}
zf19S>Uqchean(EXfnf{yBKIg$oyd`*-fy9ija~1PoSr{#mV7$@f~L+Npov0N=gT9F
zKB$V^&hIC3qp0ss(#XcH?^%w%@%*^Gw(HaTW14#Z9Ze*vdUriig%0wbD#`l)HIXYt
z{r`eSHg^3l4A8%`AnL=xoU`@p{h5$|2<(0L#-a)ioT|xqm`3DI5f77TWMjv}JTD%4
z;nJ#AF2UVZ|M<T}6ARrmv8eV5(QE$LCwQtQ<KSE(XNovDi$*qf9Jo6_C@^eUK^Hm>
zYNDV(6N)MdT)(9)djz*qG6wpId?{ieO(PpS2Bv%a0^IxY4G4E?s{idYd8n%2^~2Os
z{cJDG7IAYH@D?IVikiQjMmBcMPj)nqAIA1+{hOLvf0!l<Rjo(w`C(n;w*C;29Yu{l
zNFy7&#-}zlo-gJ6I)6q}=TFh3p{jG&GjV83XJ5Ps(E5`^h7`5_IE`%VT6foZt7Q{D
zg9Ep!ss_6N>1*>V&e1ddmqPv_F#V&~dHwa0+y7!BYl=u%KqDJF66QMXKUFT4z|mTz
z3O5HW{|(-KnmE`)lZt9T;C>Y;v>y;_C8J;`kugOSY^RZp9R;%;6nH;*?duC})WpE`
zG?}PkAo}cww@xwwju6>WM8E)zZ0rbF;vgX1>#UU+Ucvdm@F7hw+)ERWDi~b%(V-ne
zNcCh~+(YD05f^vU$cEs8dS}0R=boP_r{?!x_@m};md~$?Qu8;<=Xal}`J3hQo9op4
z&GOeD)Ac#q=fIQ~*5|+!pU(l|DBs9)Nv2(J3&aXV`5G7o1rcmO8}ZEx#?P}Ba}2j2
zj^!E)cQ&xU*2C{ji_Qj?4vEOX*+8XPo?ZRs%EV$#@hxEaL&cZ<ZPSZ@?N!Y7F?KaN
zdiV{zRrRmIZ%i|cKLOdU8EoI(SYd*%DYKO6;!x~z`ei53l_{jG!-F>u?ry9LOPa6c
zsu=?m0<P-q{BAZY`xjnR_Cly<_9FPPvVVm~JBDYUhCjQr&%p1|?BCe$v+Vad_}wuq
z`*#+39)7QZ-v6KYaCv%GV^>$Yn76v%i|^p%v+Gc?+~2ix*M;D^!}*gcCM#W)Vy&FC
zy24mywdyRxHy5~2HXf4_o4poS;<5Ky^Ee1z5RVZClts7{qS&!CT6m1O`k18K>|w4v
zru@w@VThzV?!sCjS<QgUS)t>;X=vm-ep#<uoYV>3cuW#%O8EMy{I`!7HhdWV%LcjZ
zQ5N+vcllHI!u2MzFT^{?dxu8k)0UNZsV&@dygN#AfdXqv{+%p;X!n0yJ@vG{13E8L
zF<4j5dvbHY%;d}-Y02?nnyhW8Ha&JcL=%{5_amPy3#pDglX#HGk>X6^i!`#fpqfqT
z^DKYpO~PYoO05A9cnSC^;DRt(Na}u)CIwY>&y@}r+4Eyry^j+)P*m?9{y%l!9w0|i
ztsmYb1PO!$0s#_2APF!F0py`zLdeTJNy1a&Fxl<hoz3jbFpoSy5aglgC<;LkL=Z$#
z1W^!0QN)WEMG=LIqCA8P;>9Njg1qEc-PJqQ-Bs0dzN+cz{rSh_dfiiX&Ud~#b*k!A
zHFwXEEH6>L{o28S+bjijlP~%5?_Te-^Wa1#53CQU9+dCv`yPzD^g996h3L!iShLbl
z9qY+!hicoO7>&L3yO=3sGh!EFU{L)nU)vw@_t4E{A6Bunng?n^^!6OA+0?z=Ecx((
z8d_w9vd)c`USEfSLiM^lEl0fGG)D4%71W66{W8{U>fYC1^8K=<%WuGtpt^hy-PD{>
zmv&qSszLPgHCVH$`?>Yy-bXC`{VNO+s=ws|kD(#n!ylF5*OaC4{UE3g(eDpn&8F`6
zX7;AO)H`C9-E96g1_#yY`q^B2nfDD)6QaLg!<tRq-}KA82U@y*29r-9ao=AX-zk@P
z_XCw8vS1&q+0<Fk@)GZ<md-y31B5#6^{ck#RaAKez6#WZ==0;TW>fchv#2Y2$u)9z
z|4~cVUx494b-i4^J2cNn?U5O99;gzL0q0=Nrp|!oT(7>gdxxdtzlp&?b-aGA*Iw4W
z71V_2@0+n^hn2q-mvx`9^!HO39BuKp`jYPBpe96rKZZ4%y1!#C>Aq{WT^_$3lTRR-
zo2X6Zl$Ue21+^ghc`K~h)cs7jockV2e;<KCLLJ-krR<?eJoZxVVW2idzt6;)P2KOn
zrQCi?|2r5oRR7DDvlaf=@5!zKwIcE$i#3}%4~Um>FSq2tr5G$!4(RvfTVBGw1k{M=
z{|44<>i%z@$i!X3{jsI@@4-N!dS8AOKNtaenn%b!$l^5aZcrbh@9)BzP2KmcF5mvc
z((Ql6@SwU~{_;i8?dp@67eG~r9)BKdHg%83UAmoktX*0^fypP3*6YvU)R%3?gK7{x
zJr-*=bx+4%wq0oH>v<R&)R8TJt7>Ru>n_>O0Tm&7dluGg>fUbd$3`!sM)qUZS^B+-
z;X?Jhem^#9kGvmS22~<5U=V9Ibq3Hb*WO_1|LZVRsQ#BP*$<8TR+nn80Tm*8|0=B6
z)V)u+RQoGS-#>_9LiN4A<|23}R37&afa(xEe?Qi2>YfKL)4pv<f;TXDs3edd)fx%_
z<V&=#ftnHd@Cw$fh7ag<Dsdf)*Jh*g(b9WVUi8l|b)fR1e|}*Jl^6Z<E19Ug=>H9)
zOVDor?8Ds$cJ>mjbbz6_ju@(xT{Q<<vZ+U-Yh<g7wa2rLfVG`J3dGf*rJj6et_JBp
zFkMXquDI)hWlP)!jpPOU2RQESzKp54F8JAG@QuALc$pDGf|;K-k|!f<a<$1LaO@|%
z?@9(}!Y91%G=fXOeY+=b5C}|wjK3qc%ba?(RvKL5Uin_)-uU)Yj!=Ys0z=vleBz4W
z?=ZZG5xVhS%2bij8R%;%_QQVzYDvtU9x-)x+iw|7)<D58Xt^IA33+|%<Dhv}^7<A`
z9+)nvUY}><^?<38|3`rO68--+Mk`+}5nJEj{AuIt6(14eS(2;tg>T0nV#$(&K-86=
zB9*eYMjKe-nJTg5Ku}*IOJ-orrp}Umbg?8q$Zqwsk69IZoJuI1oNh^!)fht5aI(-U
zoP?~ExN<6}GLb7MVa=w_6>UUO-;nvFB~d<(!9yj=aaKgp-lF*^s4bBr7huh%&XEIj
z@ucGPv9IDdm29oVK8Kt?)2T#GbG~Oum+xRmQR%YGiY~I%^4{+qpyEX4d=qOnb>>Xf
zF(;d4)jau{Gbnvn^G}wPc@Be!N|^;#lnIzCbLAOOV<K0c!kX1^1>M&a`=Weph02TE
z`d1y!t@kKboN}ea-un$b%~L3_$9W=QZPHx3)z=<OK7rNOED}d!%#^vg8>lIfo4a7m
zO5E%;H+E@Ga9N&JdjCDbiY1n;cs~XTHF6zAVujK`nGf?p9f^E64r?}bK1>M=A{HP9
zbNu6=UaZKL3+x&B2sh5O<i>goASyR#EAf`b$~>uo`Vx6^2G(rqJed>+-MWJ#B>9RZ
zNxp=kLnX-^QUHm!RA$H*L1l>y`2yA~%8>SN@=tO%_2-mW=%|${^GcPK^|Di0m#;cK
zs@?wov82eaF?gsHnM<NbOG|xkh;9F0g6a}U@^i0QgEOACZ~M<^g-ND4F^4wn{$t65
zH$h;DJG-=nHDI&Eh<}5M6B+R@tXU~+{L_=~+@o~+x5O};?ambn((-!$d3H;eX-qzW
zrOVQ`1dz;-(eiXN71W!^o4v7SHM~I=$zqYsBN-|mU1X#3qJMr`j>?PvdHDw_FZ$<Y
zJE*+qf05AvpxZwyzM8oud!kwQCaE?Sn4}P1!Nu16+4#^r@DXe{tLJmqj#l+mA2uSo
zD69C8C$B|I|AMPAT%$A}F8E?1khS=5!51vyyE-1ewtu+5qjvMd1rs7q`@5&F&94qD
zZR`>GpL_UIbiqH_J@O~)*=wf|{Da*he^4sd@`Evbl=Ee$kS{ucKbah9!q1#??yK>E
zKa2KMujNCUOpG+M&Z$&`Er_oy{De7W;C}b#XYS9>-Jb{DpNHI^U%Nk#upf(WFpOf~
zU^vd=8w}A;FMOA&1I*t3bS1(qr=fYcz0&Jn8yXGLYcDd6nq8|^@@wks(|w`n_-ED~
z@H#IsI<ldj6a7gU>Yt1-7LoEfBY82UwZPaop2mL_VEg%rKu!57z$n%=PHF-8HlDn}
z)MP-{`0}1H>>J}wrN}<*;s>1&n-9Vu_K7Jf{SL%%qK<~<>+q8D($JUzYCufF_rsd?
z)e<9SA5XrWQ7PWd+hvFRo6UwiyqYNj#4E|er(!@*Jv?@3aTe5Da_&iB=S1hO!kX1M
z7oD=il*gk3Dlf>z!!ar^`sax=DlhuqXf&PpE1X?ByTaMk|3(|n4l`~d=6>qRDTGGJ
zP1q<=v)r|;)tcu<BkGBX<$6!PGZTxK2W>N_kgHZBNzna?z!i6=<UUK>?lqDZl>Z^e
zU00QBoB_q&c=&TNFcWS({K*I|LDT0rTpiiSh@Tio&SoQ?-E@cy`0W?iO^c(LJdhv~
z14d)Y(6lIptBkNinkKgal_5@(TYAlEV@>Chn{iFsGwy>;9yXD4&ciMJc_;{>^6)~M
z02<AYC&&0x*d%W;9}KER^lS!eR!V^OH<A|#2xpdg0<xK9|2#aI_s?eoR9@K6XK_?s
z%tqH5g=aqjJ1*4(Y)!7}C=#$wF^>Yv6m>jxEo+s4eZq)3VrKi8C$9~P`b}qpmAyUz
zyCo60;u5g0S>pB;BY8pjFLB&En}Gc)8JGzP*n>uJ37S5@;erz|1D3YPzm8=CUe3Q>
z#*p=AAhGfEB8Cq!UP$?uh*lY4hm?Q)4OE6WOaB>bR+^>%WF#+UDb6ADS&9!?R9^JY
zLoF&VLdJ#0_OpZ74|H~x+&#Qa(c^TpB}+vyQ696f4HKnnT&s!lJ**>Oh2yhNM4%%)
z`Lrj>)rr6rH&LEyiQ7p=@`ClNIPRUDD9=p>X2L{ywh>%{re|`v(w`_tAIHbMJXhX;
zfh*0G*I^K;=SnGLG@9=h`fJ-z%*uqUk!H+mK$VCy=2cj;(u{e9k-V6(IGN06EIxKo
zdC@<Qxv0E|9fuj)&t}YbcXq~%?36i0K4XSh7~13iJ2TFDiRl{IYUccl5sD%t{hueF
z_RP8c`;&kxZsr`tdbG(nwlR_y?BCLp*KTVjv&Vl>GB6Wn&I66$5;UE`;Yxqz96Ohf
z{m_(i;imI8oxN#b`PQsrY9)rTG=VP1kWx>eQsihHK?z|Qm2uL1x&+jaIG?^BYgU?1
z=Nrk3`HWM{d_LnN8kHCQ^GJ-!i-__^qnW_%pUtOBJ3F858D4?9PmpRCp<iHrwVp~{
z`&vz<R~r#X#HcGh`3_E{y0XC^B!XCrvcd0JqW2vmc|rd>IQqJ!sKH*Nen<T2WB}fF
zO1EGCAjRA#jnEO?{JoJp8Qo;kl`R$}G4dOPbY%qV*(N)Do0CzOuC(>wCVoqN>q4oo
zFY;Q}z8J1jiZTU5N1dWb5xLR)*pR=*kIo7i-xO(ZOak>GCM0`c%}NQ$Zk~KwQ~Y>e
zziYufeOk(t0cKIj=gToLs6Nj`(o)Ii5tAg}F9G!;`u_dgcf()+PY5H>yiXgcNjcln
zzh{CtDr-{YF?~4Z-==kPb;^2BL87B;ShLa~Kf{yP4swIEBYuqU5}^a%ii(ioYfLGd
z!TuEt6zX6<3?u`$O{T(^K=p`J_#)Qqu%iNBLPe<XTT3eZ4+ct7RN%JBRQNTh9+3*a
z#F|Z=3cE$3LcZARRM_xwB5c@lv0Y%;jL8FeF_jI`&8ZeyK&EM)Oo;zro@)>Z@g~-6
z>V%jWAw)LY#4JLIgDokMNrn>5SHE<N0lk4TDfS0-B$8qp)@<sem@Hwou22}%GUJ1m
z%=iEX4>ef456mLirLT~cGCfWP6(!Q+1gzQA>2Y9$9`1`+xr)R1QLfjj?Ah&#mMWjH
zq{_!IsHjv~1X3l=VwpA<f=Ux<^AW7s)M;~Qgf`i%0><zVj7&k_wPeq?F~q3sSpl+#
zV7N@8Z-KfKN%Re@+0;qYMIn(ch54f;fBqK(i^`uR?ch(-W|=yF0M#Z^=LxLY)TuKi
zLY+aU!afpFC{=RV^-jL8zgCTeo1K=}ErNDn@(C=0W`lH*?UX4p8dQ}?k&#%lsZ(UX
z2u0kBePyS&mR+B(^_S|k>=`ayA~boQB~9Lofkcfe&8I4f3nR6yGF6TQ)g@Bp-B`1!
zQ)OI)D%IS&=q5_Rk{Edm5-Kqc0po{jnM{LTP&pzEy0K<8G(d}EL~#v2&q3wI`8Pky
zL*+&P{6H9$7ya|<22@`3&uddqdC~vnM)d=JanZES78mU+6&LY?gi1`2(SKoMT0JIR
zJ6jbQ{m5t_h(eAZc=E%b$mo9)L99iQ(H|_)d%{Rw41?cs^gB{ywAE55O(jw?t|oeO
z)+3PE`axQ7^PObTO{U0bS~4&bij1aO!o9aA-<ePle<-agGFpV;>XS)ax?O;wL#4l>
z;6yn)Xp4>Jf_f1Xq+_vWr3C4Hp1igW!DQ^j`<x}pb6b?cZ(usvlp)kHOsEt%T27SJ
z6i{2}LoXiruYhV28ByXM9R?*x>@ZQ`>P(NbF5m5hu036E>G5kpD3w)6QX1B1eqK?m
znE{Uv87evc^Psjw$A1oMRtg25_2jjopzS3|lhpJW`tU<!Tje39olQ{q1%?haC^Ww$
zWJZx8W90?)&p>^NB>8Ww+0;q0r<x>u7)B`amL)}A$FQMN#4?I-Q)Pm@3TjIv$jex>
z!<8V)qPFSF>|)5iOdeS1bci6j8n-E+wnTzV!kSH;Agp>tm4!49$Ra`HL`#mW#NeTF
zWWH7M8CUtX98{P{lO<TQsncYJnkMpTS%fd=Tk_>x3@0jImfP_~Hd~&6&IUCn66Z{;
z+0==nS>-4WfNry-$}Jc&RH__nM-}yX(APj+iRAbS)@<tJ*jF7@ycc3~8?ptb*jMY1
zM3g5jY4UpvBPvZ6+tDO!vK&o*3u;Xy%>Q7`rcRitdcq8?6)O?Gj9PA&)o#P&6Ub^6
zMU=?qiOZL!#WG#C1eGSzWizbV)af!^PnR{h?zK*_NBt7DC1no9V4{YY=4Uj`Hd9>6
zgshfnb1<klkv18uSq*K_V|H;&&-Z*$d2v3;PXSPQ(LX;@LghvO{CpIZ7ya|oZB$<L
z|2?A<NWV<s@XnSgOq9wLywQ>!%x}mSL%Zsq!bZ7zl)BcnDo*%>5&gvh@W(v)&YVcb
zemL!xL=bCHobWYE^uA&wFUbETj($gq6MmHpz%ehNJZOZD;N}BH@?>;P4p==3$9|OM
z<z!$c6eqlB1ebvOZ=QT-LP7jNwW>H_{PA|F_E;uwooY)Fxv`JDGz-iL+LDAZpf<#0
zXnU+#DH$5&$#?D?a_0r}va?TEwQR|<<K`~TE}6Z2`NF01Llfy7rjgCr<17pj;zZn-
z8Si$bSw7mL3gKu_HzEm+!kSH;1UrV%T6N0uxgiFWEg3L~!9is}7r+3|CdvH+pkhS#
z_hHSZ?*82uFIzD?yK44|`EyrgSs*xJ@!XZlP;iYU5w603p%UT9cp}7ECsE=GP(>mo
zHet=8lrTx8;uC<;3+63xzsa^lT39_`NrL+^D5xZu5l@1UIg;1!19c*L{a&xx;gHXb
z4cb3cn-zS4_Z3U8z62tuEJBl$yMyUo9kNPt?%zSxh|c{B)~qz7|Id@}T&Bk!!4e}n
z!&Bs3r7!&P^vNsilD$2dd;&SX;t{N1j<2&!qQdT=dPFMhiZz=$71+y10V=SEgESxA
zU24gOMHnvBP_KCTNXv$>aS|aGfEp4BF&Aq#bwaRX`~V^9>;r(g?rx{xRC2XaCA1be
z%aRfsFl4BdP`rGkr9{*`i4=8EM<OXIShJ~<Vtj}cJ=JV?so0zE3(?^wOFG<$VM3+D
zp-9@$G)yAG^`Le{B3z3#n>rELt(8D@=qY7um0aYs<xxu>Jd8m?<$>aQw>B&^?UHEl
z5U3oH2EV|XO`Qhecp%CXg3E}_SK8%|?_iivd4R+NVVFdOw?OTPM0g!*HgzKG8j1%j
zL+I|$_S80(!`GU+EU9n+lTTnN@or#h2%095VLGTEkqrA{&8AL<aO%j?Z6071xx#6d
zgg6C5h8i1?)G=zFM2ZtZ9f_n^i8Y%#DcJk#fsEmddahVgWsetIGU6f(7AhkY@2_jK
zhLCL%9nJ?8B+}trtl8A*5S}o+N2C<-;Z937+>W6_Wdkx{1Wl93a2u!}kqozB&8AL<
z@IEBlXsG4a<!Vk(XkGcNB^RE?K%sI0*@p~RCQ;!@P(30QevdVqIu-T~O%viN&}IrA
zDfS+-%5HVBJ(Eu$r91{mDHTS_g}S3aJ&E+#25UBTda&!yfh<BG=dCp&)HvFb8b@K+
zPy+^%Tt*C(X>mBHBas$|V$G&b3wBFHNeiB%@x&s+jsZ({^kEQD*`av7RGVlDb_DH|
zNzwzVN+d}RYj)U@B$%&9NOFZGNj6~+B}<Z^oia&21FA|S$)~VpQzywjnn)5z<0Blo
z&ypkeVhB+=GRr6f379EU<cFZ9M2dVLYj)UC#E;LVo_fLk$islsU5ilUB}<C@9YZJ?
ziYUyKDe@OkQzAwF4{J7citMk6A{;~a3n1QkS-#j8ValE-sFyRBZ$4tP5$=ChZyGp}
zeIA|tb29(uRPP_(V)5+X?yR8%)-{S}H+){WMS&Tz*)nr>1r;VTXB^gS>dZNyC4{<%
zS{7-jLuJeY3@K`mSzt5|iGk8ISEkHdP-7xxj>VcyoiYcvq)hBrV>VdwrjEfy<&EO2
zA5f$T+bol(0;)|UPYG)_b@If|ROM2Yec_=}>UOHpuYcTVNt^32ps2L5nV^E!%7nQV
zRG3JZ&tuIFGr|mQ)%!=hao;w3*pe_0VL&B8m^f=?!u$eMm`IqPVa=va81^~5CCzIi
ztRg=S#7H*$jwNN@!l0s3W+9N$#M>*A=5<hIB57X5noXTFdxS_6ek&<Lk^@duFPW%$
z?{p@gz)Gpb4YU~hWK!%4sz@Zo6s+0QNfE9q3O$Gv;lwGHoH!9fh8j4KTM8lbWJ;_A
zH6&7EIo53IlwfaW2PPNk!J`N*F0!P>`4~D>TD%vSPb3p%YMcvdNu<WvShE^xpl=e1
zH;wqsNmO1uJi?z*K;=dM{3Q-lUi8nOl|ki2|NK!PR9^Idq0vhu{+mu~JNu@S{OPHh
zv!Rx)IDHki63v&2e7zB>4|)<C-0H#U+S%%Lr{5b51o5QHZ$0_WTmg=K-RZ4F5Nq+e
z)9aS#y=o*chQZ4m{f@luwAV>cYD%PJ+^44|vL1nC)(_Hxn-h%W$>^FqO3?asr(=_W
znee*P`;6ccaKG1+?@TC&f80U!x|4(9>XS*FL9M~iq0(RR=_%#xpnc~l3+hEokWRyz
zl@g>=JbCTyf0MBj?{jvEZxx$XS2Sl;FO?2Ws+TgIY+ic01jB?%0mY}M)D%!#=tD0a
zeQJPe5*cwZ_vkSA^b}hsDO{Zys;Z5wlkT?k`duKL%9lT+RIH)+wh-X;rm^yh=}u5z
zqWf>hnw0{=ZJxX~Ahi7hRP4(jDv~r$U?VJffvIN`8=lADp~eQq2@t}PrnNFno&^;q
z(&TBZ+0<z=MNN||t9|NqD(p@3=!d??pKM2zu}mIFYN<p~Y~v$D3D_%BWelh?kt*9`
z&8AM3X*#NePv;|nWsW6ZW??8%`La~rro;FWHd-dl(V*T$(j0{~n>uL@Rg<P%amu-h
zTVYwQ*Q(jLqvo<Dfd(<es02FIjzDn$$YdG-Z6K1V4{J7cG99KN6Dw@->srLLl`Yk4
zF1I4#=o(8lU4=nLWzz>N*%XI>Os6YABZzd`gf*Kwoo2VDQ%}BHtK`?z+5Od))Ox^@
zTK8k{QK{9}K5E6HA=B(W&>$ks?!}r-oo4K$TeT8Xd_(OuO9s7yAwy-5;u~s6uB85k
z+Do9WL~{HcYgR)J^k7aL?D1V9R9@`4^L;~9Ui8m5p;39!KR;1H<wgJeJPef={r`{A
ziI9J=xl3mcHn$IbtR}S8y7Uxim8>3#t~ssFFBh?nfaSE$cX4R3z?0V=T4?vHOg6Cj
z`Q^&`8dm$wu2+R$rX5TKF?N1g75(|vi>=}TOZ56$p!fXf5hJFbx9RLn(*_ps8$R>d
zKmA7zoW}kc_#^vY`yX)s#})j4T*d!U-3c04?n-zBf02o;-ZX7#V_)}K*IYT@#d+Ap
zV08-BF7eoFSH9TeY}j<sChgy@$+5RUJ-NW&PFJ_fKgC>;{`_F|pDVwI7wF{xwHSMX
zN!QV#T8UPD&Cp7Ew)4+R1$5K*q@4Hyk{X}ni#4axn`5Q-fxp!t5G{-dNNAvLrIe<l
zb$Pe?VrJDjL-MHxFfNhx;chDC47|cQIq(@aAO<dTe=c`_u5f=o=l)#n{#@h!T<iXP
z(fzr>{khTo`HK7VRrccyyvkZywAsLG?4P9r|7O1z54`UEzTy48$$l4&7<kM3FK@Ho
z)hzFP$Nhhf>L1rw)YV@abh;|_V$rE|tuIy9b}e2quTWY)^iNNzyV_MP)hpeOH%{cy
z;|-ZirCQs>huj5BD?Gr836X~g^9ixXvVR`n&HJBg+<$$|z&l$ozv@oR+CR7K8k&&V
zcfp4SU1)B8lvxBEb*SfN*Rob~^TS5aiMjb9Prifrjnd4`uOxz4i@EtFOZ5KU0=;(6
z%^uhJ+`P1rQBOF6kzKWVF}J4RG-GJakD&v*o;qSg&Mpj|*p4tbG<LX#frz*rQ_v<N
za>tlQq`w#uLqudgtjzoWg>nBZBJR`KCCVgSM9ld73q{7`m}jjcqibQS$T-`GLLxFA
z<H>6y<23EmVsOmvty;F3Q^-}TktIuaB5=i31fFh*+iD|uu{uAM<F2c>G+q?Oe(&+3
zWMC##1fFjMm!Rpn94<HkGGJ+Zao9Gi_-qpz?O(-k^%u{gBJj%?I_f-1iWZIOk^O2h
zXAZ@z%#bzGdj1AbCE}*yby%~$T4LB-V<azTEY2hI8H*2FR9^JY!!0T=?0=83{cOfO
zsIxQXp5YlYx)=|QpjVh(*7K!nU#t1@B_jYuAo{x}pZ0t?_5(@46*pgwVLjrhs=#e~
zBY8pmQJ%au8%buqe0MT16Xwf9jNlS*AH?BGf4<xyu8jH=3}0#9JP|`lJ#R`;qY?dF
zYBX|XqBcph=Som9;_SH`YgU>)ml(;5*^9HveD>ml7?l_O^Wclhiy-nBqu}rM&t}gB
zot-@+xtnrhIy8lTi8<GL3U!TaHHCiBh(#hceZiAYdkX!pMBs{>LVsk5+YgN71^d6p
zaqsMs^Z8_8CQPBv8o?!K`ZR|No<dDhk@z*|*i|jABaJ)_8v4>ax;2vrk~{T0Dus?l
z^j!yK49lpDlP1(HKn;l#>Ikpd&V-_PAMdSzfgQW}HwUu4^<sA@+8<)+@`FG?m3NV3
zmq%}zkX;_NRbs$_pu$83%)pwJQmXwtdF{HS?fXv$t9Zf^R90Q-R04&SE@5iyOgd7u
zdpc7WND`z-u^NMm8Yxy$vK>{V$i$c~mr<SyYER_QNm#R~b7(IWhnle_!kAB5GUnqL
zI8??&D=7^Ejm(dag1Qp<aRJt>h9Bs%M=YE8EQiXA#PMyTOiJ|6Gb!`_dE$!7i~jk7
z1C<x{Uu|qZTQ+Xn*=1v7(=}R{6v`o<!-lqcXu9^bS}{IjGyp^@{*))*nKVs*NV&yo
zXznD!6}Ms>!FmMpazAhg+}<|AT@e2bj(cZUjQb`7GhxLz#S-pGp1hW(+9a|4E5-?P
z>eX6la5k%IJ&~6tu5`{=;uf`r0>DBHYAG9>hXF<%5z#jyqd_dAx6W5l)FsaWRU}U7
zv*KOd*0PECh2w!*Ff!d939^2vsaKyHa(~s*{bdka<r-3&KN`^|&gt$~S}ajw5LBB;
zi2<xxUo8=A`#gDVkZAi#a$h}h*5_;e*)!Nx;z6epV$5|+N1Ns2H5f+JfDwHG1z=3b
zREa8Af%+1uas}3G>QtGgr%EAL?5pSc{HqTkvOH)>mIpADsAP$Lc>^Fz#8`<g_k$V}
z>2e>|Z0dBGs;5hTzNg12W~=!#oe)>vu;j{X7(`UA%#G|z0bB`MDiP%sP+cNXUc#D9
zohUK6i}avxWLs*#58BN{`!IQ6X+<T>B6Gq>7R&onlR>44wAmADHg(z@p^q<ofbvUN
z+1^TNkbS1b@jfc%^h6l6%926HW2jLXbfP(f;*6Kcv=r2zNTx+tv#FEm5Ivdv47df4
z&a>pvIT&D69vyGaBaPiMh0X$1CsJqw)@<q&I!aHWaK_kzNw-=u>1GT!Dw9q&XHvZR
zGNEn)Eg%x=My%P?33afZP=1W>b#k>jd-bd86y)!!K5ogQ$1u34JX&VXBc;_ceI5lB
zC(`F(tl8mApXPBnd*y4eT+4Rny8EN=Qf`~IORcwJ@(HBY7W7eCEz@UnP;nxC-ueHG
zJ~68@_W>gJ!-qQR9A-(KnV<)VJIWTTGRb0@HeI07MA{sHHJf_WiJ?uQl<SG3%^FME
zWHG3y6P5*SB#UL*oCYdQq|GT<v#HZ&x<1!&cSMGsVy{G2@0VCorh&mkrA)N`ktZ(V
zVHo|<WhP{;OqYv6g^6^z2x~TVx=hj2g`HQe&sCz^Ja=0X<t_{$Dp6*eC&Rv-GDYqL
zRV7m7cC6XdDRQ))BL3U9B~~dr=pI^m(S=pbvS+J``Mz@P&`6Yd!IC`BW6)8_6Mf?r
zh%zk^kZJWSXatc~Ph-udPOE+NwCZ!%BU!mxbRjkVbi0MrSSFvqLMmE13$SO%P?;oS
zKwXI>*&b_FLlU&~Q<RSKIv-SC6m=YJbRsAE=Otz4{qqVwR9^JYD=$%bVgKVs6^L&C
ztaNs-&X&$j55HI7?>YG|^kupW`SQ?neW9c4L2Q((W4vo>t18+7qp={WLHj&;ZFyo_
zYu5Btw2egIimRf%*b=vkjO4`_INy_RXJMCz?#Oxm6`%0yuis$30VP^QIyOd48M;wa
zm4pA+jSv#d{Hl>W8DW$9S`Xsb$CMsU253T^@<T>&3AlgZ$#*6Y#22sbyr!Nn^bA#~
z)=KN0=sN#j7|MPEAf8tGCx#VuI7Gj+7c*I8WE<rX^bb%?V*2z~tXV02dcl*|hAxwV
z6;FsA<}6;g{6w#0eU;0G&`i5`4hn4&7xrTE*7>wFQ#X!MF6D=Eg;3&<iJIr@DC*TG
zf;tk3Fu`kfI1Kp-5ePE*YS!H?b$UX+UTo>>g&>Z~lLXn<$C>*&V5sEvd7!pLug}4n
zm4^Q;PhLCx+g_4BIm!(`ta=kewwN0X7maRY>H(>a6c5&6@KEQ5#TFcq4VJi41+^w}
zrHnP3I#(t}xxz~!vW48n$h7fgOPbt(Aw#8!`lx_3ZDc~WN({LURG7$+Yp`ZhXULu!
zhWNEJq3PxkOOE^s!-mR{=(jlI(##gtN-TL0RGG+<2e4*SXNeR?yrWz1vRvdaV8j}`
z>E>+=87fEAcNA!GMA#}XN!|d}CGz7ntl8B0u~&3l@eM(KfE4{E;(?YFnZe`}SfQw|
zJkcl;GFN8FexSxgmh6Kyn>tG-MOhMCpn0k#O-{nlp+*v$*(PkS%#&51%0!+Vk2RY*
zPo&kax@h5}mIS!~gM~^E^~oVEj%ceE&I5HNGUFVq+0>b_M>K-;)CUJQW>v}R9hU6)
zCI$_a9X7K}*jV2gVpr@|P-7xTZpNBTog?EUj?^}m!+RpnSaRbj3=%3gW?Afs^ms-}
zoOm46l*oz4ux3-|#CVAl&boYem=y2owp+Pu$K(@8OVu@SwCSXWTPd+(TToRZE4IR#
z9e%9nE){$8eW5*%_gJ#x2n-WysIXy$Z>7YF!$4JuteA;4n>s6YkH(7NBLx?3azC%i
zKh@Z8Neu@BhDr_fJ1?|=5j0ie#~M&yB0sWNv%`-c#X7q{pR2JuV&Ni_%PskFDF#d@
z_z^T!;>RVRzC?aBux3-|$GfBPgFUWL&arE#`C_?VW9PWJHSDv=S+>;64%Q2`JljRh
z6(YN+KenXKJs4(G>SQevR0<06j>z4hAw*8yg*B_;6k72qDn@yw0V*$Qxp>J5Dlhuy
z#Z9QZ=>L4L<OP)%{qrhIR9^JoV^p%^SImy?Y{l%vxQf|Ot~sFxT7jrXt!rJYBH8h*
zBVeiG(_hrajrHUW3S~@cp}ZnP-KT<&NCdJLp9((A626)7@C~&4Q^6kLhdvd2xcXDU
z;X~7|U?I2s&EV6R0hZqkKGpu4!4Z|CUs7LY<RuZKxR=ahR2TQM?p<Jmx}dm<<|^wA
zDDxr4j`)h^vJt`}j0_sdi&>xr!p6a7%<^3KT6Xho(wq|eq<gUzPU}CL4AMJJDf3<r
zWb1R@tCV41W(1f3{L`MiHg0#WhOKo;^IaIqK2)Nl`A!Tf>UcOPJ_DvyG2aeqMoejM
z!<v;++FOj|MY!bLvJIE0yy&0jB&fUyCS#24XP2<{-EzbTd{`L=sPTSESt_&TV<y+*
zt|;?PZz!4E(rLql;M&zHncR$ZWRpz(#{*uAUWaa6O-Tf<xEoiKEOFbzNL~<rH&0$0
z^Nq<Fdmm_CGB6YF1I;mlOU#|KI9$WvOLrq@_vDV12YeobR+=$;F@V%FrW7z5lZR%^
zNU+KXE2KQ98&rijMV=n-!nQ`5_!~SEo5p8)oT2;TJ?y20Y_%_Z#p5zduYMYYQhCKg
z_UfT6y{a-ya`Y!b?TC*4IM%F`3w_j+*M>%erCU55CI~vLck+GxwQP^mU8<K0;S=f~
zF|~j-o;1RLfWbl?;Rm<oLewgW2;T!$BNE{|ShJ}UVMjrP3VX6L-@7rC)BVko0Ds2d
zpb}tuYXS^eBRT$0ph`r?KZiA&y5q+Q$M@zpI6ZCv$c3H1Td&>d--XEoc{<hk2ex*8
z&?1QdJA-Nw39uv9Z0ZEqRS+P{9;~fZYwW3<Y&LQ(INy>3$6=^YIgn}1fuK?HG2d)Z
zFCqz!!J18-1flTHND%RU&C>g4U~o{qzh7(bR~%-I?Dit45Yg>xv1U_u`xr6y*$0Zt
z5wCyI((7Nq(4cyK-_~Atjgg&xHK-5K=~rUSYMhSlABz1)zPLo?#dgeUql`uL&-W+H
z`{&z1sJ!T(Z{wlz!u~gH_Fs4J?EdRGY5%oUFNW9c|HekGdepgQwc39Dml5y9zSciI
z`LwrRcj<$YTOwR>+pjyb9)Wb&j~4>B9gXA#-{0lQYxfwE*?v7L8JG#%uZJ7KCEz}k
z!!-=HU$=E<YiYc%#*p>tB+e^O#qgmHfW72d+Ev}#eD!Sb6%n_HP68DnPS302{nx?0
zS1-x-Kv<KjI@v-ce7yfrOGjP+;-}njl>&cbS_?;Jn#M>TJrC4~=+SeqW_`6p06fc+
z*9O40Z?f((G;TA&wP+;SzJn=cv&H&N3>4}xKA@HNgC@x-_pP8-L<Zc9HJdsEc4#s{
zvCMkL((_MYU{F23Pb<%>7g>*k+7Lbd7}l)D<LDwmEHd~kg361SbG(uMqJN%+nD@_<
zK2%=x&y#gjUfBN+qs4;TKPz%MBIQL^k5j1SLgyhz^h42FJ?31qS}n2;V;$Kn7G`?#
zX)m%)N(8RBMb;`y+>SSr7kppJao5%0CbP&oGZ~l(i>&oVaEV1$jl(re7FqImzXn6r
zpMJz5>naQ%V!UX`i!4udcag<y@D&k@tSdl8h|}{XL;tls*LE{4j)5Z6y5G`^_krLk
zS6EWiZ^)Te*ozr;rgbl<4bhc9#F~{dt?wJji%f%4$vo5ejAj2k6q)zWBR?uH?B_E!
zDlcZ9E@S&yrq$coOl#-pc6MJS7v9a@Gv9_p%eAUiqP08g$R^R+)ss&<(K<R2xZ)D6
zqbzYd+(=%~{ZNj3XA`Z{lYyC#XstGaOC(ySa=3;;qO}d1spm(B`}r8KQl51#29G+=
zlEOt}4{0yN4ODKL<yjzQS!aW45U1ud4V~BaEbCpq<z8kI*~q-b(uH3Gu~Q~lvJ3Z)
zb)hE3`U<EE(T`uknw3(lFB-{<6od1~JjGaV**_0L=Kb@SkID=C`CN_4i#cbUvHdK?
zn%CJBYsYAcH5l5++H#-``IT!@tNdy+){#wq^&b!OwDYSeiNF<?Urn;aZ4V=P!S&rd
zdF^3YG8<X*l7X3!U(GRsOXOFxI9$UZzuJ<`(c!V4$554$t6mHnb#f)eipDO|%sl93
zR)g|sVmGJ&aXLQT&}VH=tG08^R^?QeS$gi%AaKf@O7`4DjpsDk)F(kLi0=D1)~u9G
zebh)^WD}e|=Gnw*%l>)DG4G#8cvN24&u3;-Ud%9W+GJDt&Sq0%rEDs<A(T)3nK=_E
z<50&)*Qi$c)Srx4BPOcnJo&WqsgY}wfGaMa+M4xflZS3$Brn)L!jtcCK6OAcFcb2r
z=|*q~n(oWt8V32)*6!pS8STq3T&0X^F@_E?S~PZ-=jGf6FQ;Pq`$~u-!G)j_#A$h6
z$a9yE7_r$1_djQEnl><h>k%XRj~rOQ{#nibt8VYM?fI2^*O9p}t_*LZrAOC+U@9{$
zDgG;pFkL$o=E&}?f*KLsTgIA|vaUf-zH`3c-Umx@Ec|7r3NU9&qxc343hF4H6f=r7
z8TNIcCPZIfgEg!1HJTxc43mcvR9?)vbBy#C{qsr4ynjCDqw=DEo~xnq!v1HBaz(d)
zwqH4+vl;eIQifd)CD?B>Ujl<wJ=k2ES|!+T7?DO~F0Xm=X(!kd3rWBgmtap|J=!GL
z<BjA6*~fbF+KDBZ4a@f>12ZAPey<T+f~H4uxQ0Q3y_GvdFAoj&H5jbY%$&u*AqI=a
z1bJpId&$^xIsG~fRDn1ppAzr3PAApNGuiI`T(RgBvg`8B`pA)R!_tixgBU84YANP7
z<mUqheK&@!k)3)Gs1VVq=VQ%EdDgj}yf*Z;eLXeS8?hPwQD>Fo?_vr8leIK{@5BJ1
zj^C-;@f);7cKhw1Iz+eMhBceI+jsNb-d(EZi%zzhAIukWm3(buWY|A%>HTLhT&Uic
zue1e*ebgkG1W$un5lQeQ)@<q|7#jQKN@>7-FsxLJcz^7m-Pj+)<bgbr>ixK}FKm(B
zzCEZ8(e0zKW>a_jXg>B8yC}0P{e3is1l8Yr>SMjSto|ra52B+F$C}kR8r=gBdk8#X
zMCHY@>O`aXB>Ly;4D<f^vJRCO{qyBGDlhDR#b`Ug-$Uu|>>kP_X%A(6w#vQ|)>AJy
z+1`p1e%$j4Y~ZQ~o@-&N9hFT+LqIH5KjX=#y`yqxB5=j+sN8Oe+igbjg8H{`+;zt^
z$?T~7E*Y2!J1W01f=kf!5f0Zd*imsG92q=mr9blLMF;*{7{vaRBrY|+jsZms9F0Td
z89NfbyuFsNUA}VSRP$9(IpQ?_a=ddpxbNb=+ZlFl##twQ&u`zNT^2rt$y;aP(&W%M
z!obI#d2(Vt3Dl72=smDzrF?BSPrh?|G44y=5jxZ=xqNh0wvwr3vm>({gM~Wgk2K+f
zXP-=nC7_B#LcAYqHg!TM;)7z<cCIBK&c;xo^1&!RsMm96f*KO(upVnxLkDy{BGxl}
zB0}XwQuiV@Kl=KM{&|*T-ak(&QF+lnUksq~!v0Sg+t1F#$9Hx;C*9#GhEB<Ti;ZFR
z7<5f)wSfB{BhHJ=_}89%+6%Z>6M-vk0r#>cZZ8_i3$Fi-<E~r48Jr8neGqa&2}&`D
zbd3KX<apL6kf!?4OfYk-kvtjUWR}S9O$KPf68T6YxCGqq_T&u$feDcDU%+v%Nrl5{
z7K7F2l2{g>hJi!nzkE%q=&pl{asqq`r~onJInmT<2Jt1{XYSckF|LaDVoMiZ1R|&`
zIFM#k#X~>bg&|X9f1VF&L-gmlShG?Fd$uRvxfSkCp4Zt29csBsEz3UY7CCUdlj&o#
zw7ng}gF0@fNK0Ga=>bb*pWg;5L-hGAShK17+`SQ5RNaVt*3#KeV?a=yE#HXrovmJ*
zJ_%|;^z!eqW;I?$*H&U}#%BstUSv}%J^ecuS;ny%mSP_}8)IL;u9hm<93x#MxJ)eD
z1(yj-9!OBBe3QS0!sbQU7&*R-2lXK`Z!FfV#Jo;>e&_69y->@?IDVm}<L6<BP#rIq
zu`tI=w#a^;1FA#x`z)+kjo;CTAtDZ6P@wW6;_wwRDlhuyI~%CH=>L~SD}1+qwi~s&
zv%68@8yY?J!NKt70ykkJM$LHFrdHcgpD`kxScQJdlQ&p}8Z2nM4U=u=6mr$76aMVO
z?TNq@w;gqxC2qGE$qTZ7jpMG{jxt%}d+^47|NA$|z)aXIdBg}VLDOGxxQ0PJ!B)pf
zgZ*_3RzL8I&5~C!aEQSodPBntQW>tjuY#x)co|fII3>Pl=r)7k62DtA+Ou1=h?;VS
zT~a%V$y+D2QlxK0N;U(|(_DMq1Jr})!QHTCrKDyTBYBZja0Zzt6@SBs?dvc4=Yhw(
ze;(&idC@<gpHX=+&wRqzewI{??QBvtI#{w9I^8>yu?X1KQb$PFnpWx5dLz1ud8+2g
zYtt$1a^85pAGNZ+X0YTIwT3gQ&n1FbW4kF;(VsMneAW`Z%UYoKJj?Z`pSQ_<K7+rT
zG4zVYY3vn^fj_eUwf_P4e_Y``sxh#N|8pY$=VboRsr;Yi>>tkW7nvA)fn#Z7pIp6G
z%9Zn7oQ7R&cQrrQ#r8qgIEB2^+m$c&I2$%yv}tK$iuNzvFPl;R(0klI@P~WrT6Evo
z4gB3+y1yF?{Ly6XALTX#{%n%=&&q+n%4q*8Qji#s(P~)?{E@nQZ=wtQmAa`yZg36Z
zk1Ex`AF2B*GXDm#Q@gfUT33pAVyY4sk2Co*9YW8{v7w$1P&QcY>j{^w2P{%!!z6<K
zQ<I@uD);AeE!)~y>dQC($QgKr50!x*uyHzYkNb14`}1G!&rjT+``w?Pxj#R5e;#yy
z9&&$v?fyLC{yggb{MP;Xo%{2+`}3sx^9T3m8TaQ|_Tvn^$~s%L*}!Y;pQQu;X1^B?
zyzc$J;r+hJeiw}xc+2}QZ?oUkY@_EL_y0Mne_Z4JUHzp&r>jzD*KsRd>r0ijU5l5@
zE0op`{nJzGu69*R^-8zXr3Bs^?U_oowrRkN6}K^^fILr1BzSBypU6LL**~9|%=<sX
zxPO*NEa_|_5v~<-gGDHt*rC#fY{E6ERW>o2b%b+9WD_Gjd2Ke))<SFj=1e9LxZ*Zv
z_P4}snvuLnnx}HyJG(h^d@?W-HfNR^!6j(Ah{H7uHfOe693AU53{^jDi@ljMFl^K-
z4=Gke-}`k#RL0X?%BhQ>0>pLw+IXL}HMGQU&5UvF&U7;f%2{@IqaMDP_XSJ;T@Avg
ztYVS<7b$3H`Y&XM<jX5TWr)7K9BbBBON6^iJ$dcgwC$U5BR!A1>2GLS{y9?vSmaAX
z^`{sN)S((FXArLT43K>M6Hp7HkAI9co4Sv;^L*UntjXuX8x#Mw^zy$jAgErB?6(Rp
z`zA<!{wJsh(a-<DnjI#7R$Qu?TD2R^doy_;cc%JTJDSzSNqd1>5WPGRYgXfBbTKD(
zmHF(C%8MDJ--xB6f1WX#_s<h_R9^JY7gngeuzzb~``Kc6k8~Hqm2f=%5H?QL<IFXw
z)na&~5$DA!W1T0T_G0+zMBs{B3}0!9+vP^`g6o%Z+&jA%{$VmO6BfhYH-byh^t&9c
zVX+vVy<%l3!+r)s)t`36V)!Wx8)B@G7sHimX0fZglyp4~DnOi$AB*={2N%O*UAse9
zAtE)n+tuvSs%@FPby_8b`{=C>_7om>VaOEOpId?25dFD1)~u9Xy~FvS3wdo{5O3#s
zU6Di|Vd?F|Fd(P{R!X9Wysb{6XM$P~z1)R0tMM|L1d1e*M-5b7OstP$!OYiR^v`D+
z^Zxn7j>?Pvc|wND3;V0a_Om2<ac7h0aJ{d0XdIeZFTn<?dZ4*BwaTI!Mx+<%$;F<0
z+FA6iiNF<?Mc-_R+f7FDg6ubP+;vlmas5whz3;D+ftip+|I!F9LDQdexQ0O%z17m$
z(T(a?Fj)P-Ez;<hFmQ;$B3ke31*r_z-d91?`~DqNfjA}qCEjfvOr&@A3=gHBkxY8e
zx?Lu{JCnE0q@|GGh}>oicrj#*oJ;Qt>O*wtIILMI%i774@7&?iXzqALGQEgtVspN<
z07HU0Y$GQRL5Hi;>A9dDL`NTsHLGznnhuI|lE)5IUd*unhlMm>f6+gma?JbZGdwCU
z`sbM$DlhE6$k=|CPWN>-ogNpg^v(=bSB5g`>#@<Q9&N5!tupFsjd(9omCt+fX=l{;
zCIVMnM*Txe+`exlFZljlj(cY_>gSSynUGOGV+5C==~Eo8VUSU8yOdQ^&u8V&q4B=W
zI=k6<OD1ofj!E&Nu~+!zlA%zQY1$yA)|-Kf5U1z=#QU#<sr4>HwuchZNOpa&r8_eq
zipshz*`1N^o=ENtnIz}f`-55$eLD?nR!Y65dh(siv3F=PK#^vDkSSx6W`6(!gE;X=
zPbg&1tCQ`ML2Za0KLKl2<8d@u6v-x!CaAoaeBU&hjYR)^HZt#@PyVR9=$|KTsJyWM
zbH?_wWP6fPvTbnc<kj@=63Vs31*OpJ`)zE{st28GSF3FMTSmkeY0Ni_<b~tDZY1v#
zhdFM*$Z^JdyWv9z@4i^bX-~JqX|7GJoc5#<Xu@f~_vF)FBE6FcTyaaJw=8jc-AG=L
z{Z)>8XO~EOuZQM;btsWo*&r;D_F_E)O8|dT6*Qgb$!lq<&96)bc>JA+tyV~f4s$VB
zedENT!?74RRJTMv1M3B;4A<UQK^!`~4^)AeOT9PVZ5>=0?d%yIT0=yZMI}ow7C;D<
z%OZJjMGhSTUJMx{FNpG>K17%HV$DkFaknS09k2$eRQz}y%^j~;4Sk+zVzV0h9EJpS
z*hUT=f(}=&gFXxDL3H$GShE^Oqj|ho2k_W|%8MEHNF)74|9r|Z@1M``sJ!T(XJ)9p
zu>ZG48H%4ypV--Sde6{8NP1Wylv=;c{0a<P^{{j8Yn58RXha~9v;57I*XAs3o&5Ay
z#CO_|1YB{c^$x5@n_O?Sk-Q-ONKaln!<eLmt>5oEEE$*ysr5`FxCBkRI9$Wv^kIjU
z^N-Vh4e(?PU@6Hy0fUJcHX2>wLkKB!Wui7osrL#{G2-;SEZ(hcjXv=i*!Tzps_y`P
z#L~MT20>IN-_oSeI281*<{N+?0<|N$cq7)Vl%cKj<h6ma?Wy=KQ6i|m0Qe226fnI@
z<NWIwDAaL&7|I0A*Z;l>sz;>4m$7D3r-Bp{6zj7mESc~-3==97;FzFZwfzRvj!1+@
zux2$xKvx}N)xxJ8R9+-+i;eUb{qt<bynmiNqVl4Dp7f*g!u~%SEjs*F+aaA@wT0jE
zav!$~Ez}O#2t{)B_;XEawNN{Vb!4;PIM9<%d!e=>5xC+OYRfEfTWlmRxW16%-q|yd
zdNME*7HSnExWq!O#NiqS3$-m-BDXv=*001+m9n<WF>HvjB6@w<4N>lbY`MH!ekrH`
zaXP*v-e(<Lh3)Lxoym6h=h(ZOh3vY#vp#YF@?%RM-UC9YT!TsRz7hG9Ow)&9TV#*k
z4XQ)*=v`Q|Qi64-C*Qe6*G}B;h1^D`qVoDbm_9b^t-oS;P)Ba$6}N!bg(b4jUjUUM
z`uusU+0=bL!5`}-_Ngt0eQ#?p&u;zaYa1g2e&U&SuAji<fxL_A`Wezvn+I(+;G+h~
z3>XjUL}b8Ntl88V;E(!pr8MAl*Yc(4_u>{>x_%yp2i5iZQ7<f!eLe?NhUoKIShK17
zd^>N{D>hQrSvtFl0YP<k<f#e~;?)}_Wl#&Emj|(CHC{#+{9?m^XNai0SVT39<|ffU
zUsag*&lhp1yy%}VzEOE${|saMS=H#|&Tg2D4IguR5B-OhQ4eAxRXx&Nqgrj5JYd9n
zu@=4GlTUlg<gbaq6}M&bf+cRx8_5f{Kg)6N?3T$6XF+))k(HFP(b24DAj9;7v!Lln
zPhLw??JCTqMj-wWbeomlXrCV%?lUo9ee*=&Xcq<#)iKeD+0#5TWPq=Qs2x23)PguM
zPmlLq2e(ef`=*BqE+cEH(=1(i3J9Wd^F)gLjp>oKRLGSfd*rnGL{KH7UsqzyN(t9;
zPhLB8+kR@j>yY>B<(^#4$@V+dYF_#P_r**rn{AYfFjS~RdH-1dhb@vBa6YIMkpbso
z&8E(PcMUN>v9`L?((kupSWx{wIo9v$1=ek#DnxJJf;Fr0Ho8C%3k*Jwpz<Q++Sf>b
z(Lc{Y%=_mlA1W{U=czg>FYLe1C}nl~XA7*s&MvURC(`aG(n9&x_8)?Rw0f|)Cbe2#
zjba_y<gMFy^4h$$t+|iBbohWo;EG#bO}E5tUn6<J^(h?p&MvQ(B?B|zM0&9iTw-~(
zki#_`PNYwXj`a$Lsz2?BwN(kjh8QcNC(>?+%6Pi_3W#M@0aSoE9p?>w*7m%5JJ)R0
zXTLvh>ABB=z$uqfQj~8*R>e)vX+HY>Sx^h2`!2(pm9nW%8_A1og44%5o9MOdpNAat
z{&|E)<%RuxW=7@346~)N{VbbW+1YGr4}ZJ5oG-J_EEgT`d&AkG{}#%uUS?(i%U^Z0
zbj@p(S-oh4pO~`#=E-+3vl=zKR;%RK)N7HifbVp65{R*{fagSi(wwpb>(nMM9c?5p
z*gw*fZ|B~PK6^bZ8GyH)!oG)Ctuk*)=*~1kM{u*tNS=(Y$)<Mei*F|<12Z9eJ;4Yr
z0rv_|zB8d9{%~N-@`cOiE?&5FZg$D+<;xc?4WAO6g(2+IQe?6lFr0|7S=@XZnh?U%
zLq@Vm8ZUKFF=C2T!J75e62qwE$+t5|#e08T6RD+h=N`8*JNM*OD`tnk4s;V!37B3b
z3fzc6LZv|SDoiU1gbb5-a6PCUkq6ge&7wRo38(Qy7(KgQD=qhvtx$UPsHOiO#-O13
zzj?Z#@qfr3$?Xq;N)g@u3$IyiSnu4p-aTl3rq8K0Q|HLp;X9VzeG3Fqx!{$O@5T{B
zX9xuEhRl;Qqt`(ViLQPXYgQWJFB{2=#UQ7e`C^dIG^o7jpJy<ryqM{DI*!VV{`q1G
zl^6a0*=RB6_Rki>TX%LbJTbZ$UX!aj*<2ytS9B_&V142_&}6L+wyt%p*1{`UM>eau
z<(|AYQg-N+p`Hj_ai<IwOWaCE@`CmSj(cZM8LmtQX2Lq{awE9JI_**p*DyF`*zQ<1
z=4UI0{P!_%{fS5{(Y}j8L<||C`lmUA2ZL57XpNMeeH&DXIAwn;-kEKUIq}){<be6^
zT{5S8Z8qP_BYJL4!HFbkf3$S*|AKfblQd~oXdEpiX*w6HY?L##KY(fy-TnmDtdyz!
z&Xez4?mjk14sVi=&|#+!LzBET;&)*3z}!t8@y+LmT8{XPXO&EV(V${P0*u6()erzp
zHASk*;|?k>=In165m@xkCo%K>dCq{!i~f18hsq24ml)g6QuPs?P1U7yD=#DtjfhaL
zp2vo<dKkLqwaV3djrcE;pl(k-?Ogq1iNF<?t6yk|+eeJ#1@Ax1aqn!deseM~6LR&N
zjNlS9y^+HO=jv^*E*b4k;LBBeslUX)m1gdrV-OKTMnm394Mi<4S#R3oD<z7BehMl@
zoV9;~H7m{9KQ@vVvlge5`K-l9FDftk=TR4x7t!N%WBb{x`SungMrZ@RHWfBGmhet_
zcT;X~)#}mhtqJEsnXj7vu5GQR%<-%vAVv1cCt}f9PhN{)TbG*pGK3=%fh%sxJj@ce
znMU%0^j#eH&Q6&hNCsxYlzFldT!N-2aJb+p(`2B>?}&_a=gd{ofIkO=R+=!+!T=%$
z4Dl||P-36)CQL72d9x+AM4Bo$fa(yZ%6hyT+Zt`+-MB~7je#v628&`->n2Or-Uwo;
z+|-ihg2v(MO|6J)W$WbKjO#%KiO#+jYgWpOKJUqE=kT^?!(#*`To5CBS&uSpY^qBh
z#?YV+@a8)}u>(Bg8YDaZAy6lx<9~rQt8qM<P>O_<M;TOJ%*_=e{YC$LvNG?V=LV>}
z=%43wsJyU$jIsSJA>XXC+gQ6t6Y`#NNqOLWz<JR4RgXW{xK<hYbk>ngM!v5npLRyR
zC=s~gGV%qMxXm?^7py;)<F3od4W=7!y)$xNz34qIRt@c`7L&nu)Rf+4cT!Tf)(9by
zg!dcClMyz_)mxWde<m5A32UBD8Nnss{)8vrnLrSqEbr{q<>m4frzgvvcVG*kS|mHV
z14G%Tra0UGCWaL;IxEVr1JeU9#|v2{r%1PgiV;(!o3Uo46zL{UUYi@645fGqj5==l
zGF4H>GfWYiwD>6u2&&`vlNLuo#|LbYg5=|%LPUo@#vP{H!06!0cbu?3Gg#_j&m89}
zec{PzyN}p8bz3G6q(D@s&LlcDXqM#Ptw7y~{@omFRvOUnaDKEg`-mUUI|o?chIE+(
zM_7{JFbokY2@a-`Kv*SlU?!*<kpo>=v#E1nmjDO6LKpV<wG;~+OB$@fK%vq=zEz_R
z@VtU&$S{csSx`G76HddLO`Qq52AJS|lW8c;3$0l$wWPu&7%EgMG`~94+yc{3A#9n%
zg$Ae|kqZ}N&8E(UU{r9{<qAqF++#_FyD?CxR3Jx%kYN%N?gF(VGT~0F+0>aZDG(L1
z`C`77XS2kaPSzVoP9?Gg{HrA;Uck_yQsO<-d_iHMyc~QURFcSyXR&6}X2#{4jo562
z`=7>cL5dVgIo<xig!AoI6yuqE0xJsg)R7I^Cevaps2-6PW3Xn^rbT0?02{o7G7=)@
zS+Zabh6go797qijo=vj%XMt)Fy?->;tj7E3F_1V$;#(G|yx8vLdsL{r=%4R@qVl4D
zzM+rGi~jjR3@R`BpMvcq`TleJXUAx>J9~^KRfDq>dfiOrLE1IgAXX1T*SuB-X;&HX
zU+kY>;mLPq|J*x=(7j@GcOr<jc*W>0OZ4tEk{9g1oul8z4K027;2)9!IOY|jCydY$
z-29!9JQ-b+{r}cSX`5dFC2_=&Px-{bJ4SE`xZm>RI}-}x_g%Lxl=}K2cj2dExcU=?
zI6T{%$y+DKQbbnN4u&R$jBkmrjwrL=3si@g4o$?GmC~UJp1k(R%;0=K-qX9RaC)7J
zQ|xw@xTh?$D}D7rr&tT0w5(uC*&LrO!$6@9$#<(VVv!GIbf!sMSPbe%<ibL%S(FPV
z;WM5LquI)6-OSmFTk#*ZWWa|oD5wmOZ`y<hf5;xm?HfU*h;Co!H9H)_xVuT(^k?wA
z@xo2#Z903?!1ArxwUn<~y7tQ;ips)aIj*-F+&a@FFW&&_NA&V_ShLdLzQ&X9T%g}Q
z1X#uC&1#O-f5Q~B3HFa*z)*+!Q7wXfoOu!_eg*1C<ivwmv#E1pd<Zmg6deo?TU=;I
zhY?IZfwWJ#CmNXhHI_+CcpGzBgUEz8ux3+d!h{fB-VWl>X>^bb2U(Kgz+}iEKX0c`
z16mxl&j2+f@?k%$+0^;q7T$==LA6W1+L8*VVu(<u31#7pE-q+x%1;7SBXVFB)@<q=
z*f+$1Y_?o+%DIYvu$nLSc~v-(vy+covf`r{MpRbJL6QfJrSgf&1)#D-o}7m@n>tUX
zg?Q3kDz0-XwQQ}FEr+U>BTV^@B~$LeP@*zrF2WS8tuj}>393uv%B@(lsdHs&h%4Fd
zT%kK#ajNA~k-flKaEg7k{%V9M&sh@X84MySQI183qBc}!$y1=NM3y{`HJdt1x|A$g
zpR4p_*XL{fK2fq|r<g0$Hb(fe<3)B0mUl7v1QsldP57cSS!T|5pw>j@Y>PEJ%$QTn
z4VKy0Zu*91qI}PWY;UDB$U=^bY&6mwW67NNU|>-r4aA&QCd<q@0@RwwoWrnYQ)kXW
z;h573$4JmwYe}7c3@s{kmOw$LmCZ7D98hf{ch+Fdrp}$iFz#e~bJgh8#49ZcbU6kY
zl|U=nN+8d0nMId^x)WJ+3D&HJMQCM}sLbN$K&ZT^o@j9t-2UUg@k^Y+Qf1@HT(1)(
zbm(@)uPu4{OUx-$p0?T!)>!8ITvP@898`};#h-f34u>Pb$!a8Yhl6igy8GWCp2Wk!
zSuMg+Ya=Dk{|nTV==pzQ%}PP&AD(>YjtQq~po=*$oc2+>1=mz2pTL6aI5Un!43&7Y
zH>fR<CwpPdrp}W|8lI>R5l^tB$O;S{>a6iTGl~csC3Y+W6(zD`G1hGA?3kiqhxTCc
zBbEgDFa{8nAgwm}C}v6=`4Ffpks}+iW>e?Lv=-p?<_iwnxT*Dr^3QKr66Na{N>rlE
zH;W`g<3ZRevE{3v!bG-w8EZCmw(J?lmdIwz6P6tL9fl2+Bg#9Tfurd-3*|-gZ$Kr9
zym$m_R>KQ)(JL0se5HiSM;Fbgyy&0T;G^=Qe}16^l^6Z<>p-Zy=zo*Z9R#<3cAe|w
zEj#*5`Gn}Ji0*y&dahOqe}wOlk3s7T^>B1eYju_DAl4DE5b?P$?r$9E$#-UzufJxr
zA`!Ua-b7qxiQ8f$dBORG9CzI%E|ZP2*4K>c$-qo_v9DqTm!N5h!=-!VOS{)#f@A!B
zgOPKU1O7@3T7Mjhn?{#o01*R5V;?EUYsRaLFvQnJY$;v}YD1i;FNybJ2cK#i+cZ8S
zR>r}q(H~p-^BxdE<tCyu0W>lce+EpFJbO2&7tynKVa@t#iEwzQC$AmGZNEactMGqw
zK*%q2{ex*`bFJX77%J33eF%vJ5xXQNyZ|akWWw`Uv#B#-XTb!`<&udP+p%B*lLt~X
zDhuRqEjB$aLVNIxL}I{rP%$C{#$wG5Ck7Nsthz5-$raazHgFbNGGHEtNP8I&ut{RT
z98fVL17=~(4krezFI5UX*(!UCE5v|xmJFz3h_sgh0h=TSltIOa3>d_kO`QQ_#4NxH
zq<iyyA@AQ{>HX_4G^pN}ZxX9!0N)<T^Vfh%5j}qu)@<sY56u5-Sg2epkNsa+djCNT
z5vuoT^MAl5i2)CQiV+!bKh|vO4A>(u_A5@V=6LUNy3f){FEPGtNrg8sXsA?>@0}|H
zLDV>TgZ?#8KO!Su!J18-5jzP+WCwG#?*7PO&5V!RtqAsG@(HX6X!8NLNcR0cpi)HN
zPsW-}-S>g{z#l}>@0Fco>HSq0BGe!tzqP0e0z)>*{y!d6i|GHQShK17e@0;J4}q1<
zb$9byJna8)3Qom6?}@~O3oKc29)=T@74j`o6)Uo_mdac?2UM2Gm9wyBQ|HRGSg!Oq
zPPtlkoSq0%zG=ynTQQWVOqosERa7vgX{XGSn?Y5HJh=&LHg%p%Q1PT(E<}j&lqE49
z$6%onL;g5Sd<bEN$xL_*)Q!l5N3mvyCliW=C=<5(gk6HUEt5|m!9<zB43n9#6{s7L
z37cchYM6i)v4|oVezu9qixaSh(YcA}pP!JM_s<IwP<hcmFC;<bh5a*(?Po<m?@74`
zs5f79yw5|2whh-{qgWk}UDH|>0cDN2FRCj}^W-~N1f+Y<`Mg9BYw@1*IhN?1Wh5_{
zzk#E#D+Mw*`_>m)eJL4$W8QQAq7gcRn_n=JC!=dpAk@0h>i%S4CKOuTX9Smk`(97J
zGoc{9&`R~5^Pe$X{YhR_S^Wt^hZvC?lcl9ob3)L*<@_9|3Na0O25VMIgP!u_wS_n)
z!zJF$Bj+jS&RrUI8Su_b-Z}%8=3YfTo$$0U#79-s)9naqLv;4Lyk>{PVBNWCe6Zpq
zSWq|H(xJzI5Go7mWQWS9XbOi0ZIXQZ9#Ao&Z;!y5l?L)*p1gJ-8)Oyn<9MuKfV!fx
z$P}_Erdx{vLXF*WMWu=W!X(N6{h(e%|2tT-sr!Fd;r}Lpaz*9UmL#|mLxoBLxuUXZ
zLq=<t#DvR1<%mqU6l*qhCQJ~B71*OO+3tR)du_RtFV;e<<DXhm;U^d@R4T|Hx>Qji
zYM8`@AA{Nvxo{8GZ0cMXFSyWGDb>r_nzJGLjL^R<iSSPh6DkpCnSroM-s$`Ys2Gt0
zf5n<jodf#`4s`Ppl<s=9RvKhAmOa^0xt4eTujp~=-k-F~7WQKD3FQCsBQC1w5M!py
zl8K<EM3zjznoXT0V+2dwoz%$D*fL8xEXL5F#)#=L3;(b^vfmeiDiQrY4{KKAcXS_L
zoCEMRDJn1a>-Yu)Dlhuy8(65k=%4S#qVl4DzL$>5i~dhB+Q4!9XXhZZI(rVXL*z1a
zD8KtMHe}U9&NZmj>BkL5j2By;*Lm{VT@>vS+GH=x+d3Y#vc87hHL4e!@F&ZDk_ck#
z^rI^J^T)f`n)<ONdiS(I?|H_|>F$@CrVT9Edc=s;n+E2yfBKIcIF0=?@JIH)_CMhM
zk1P29xQhRCBLC-P{?DoWAMevo9>rf|VyibzTiVz^SFe?F<$M=sVplm|cG&B=PM7=t
z&o=*CzS!ez*mTjRrHvUnfi*eynz#S9NtgJ0Z3o`pD>|4K+ktmT7-)XF?yuR$KoD(0
z-wp}1zn`xCW(f!7kDED8;E%O-=8OKv`6_|GK18QSzUEfb<=3&IYo*0ne^qk5H3Xr`
zAo~oxLbH8!nlXW>x2EX+wqDB@ssVEk)cxVQLP4pAGw=$h^uP;jJPrKC{rQ{w^AGpu
zCHLoL_vc^k&ui|_>+a8+?$6uq&pYnVW}jj|Tev@4u^(sPRo3RB%?4g$|12H&H~YPK
z;C1i!4e$3&_Pc1rz+2vbd7J&NW?9NR?*DUC|G37IuKv=X(^aV#i%zAB-62}rwRp+A
zLTUZbKRu=HYFD*XuXOX_rW%IcK+06AHUGnHPcSuswA_zYVs_=x#(Z|Y+p>Q?SD5#I
zjdA~Mb|2Z<*?qUj?4A+hCp6I?%(w<@?5Zbv*S1y@eTH>}lSfSS`+M@*iN3AR5Acb8
zcCA*)uc_A}6aDf;5NlE9u*4F*_Z!KJ#CASMzawQ1)novUDRU?rp(D6CXe6IFx+CY+
zi`}(6`_xHj$bU8&d`C^`ZFVOms+SodBoO|zkvth;<JG_p>?6i(DKKeHsaTV~VtiLJ
zKobf{?lgi+z<s+X-<e<=e=<3ar5}}inZ3(cspZ-A{(7zVh{*QR6Bx`sH$`2^?=Y~4
zQ(!~hf)-P-n^Z-uk~f)t11d(`WO@W^*2hbXrC)jS+Ld+N52wdh>*aE(Qezv=?n!Rd
zVc+`NSPozDA9)Fs#z-94n#lv%36%qK&2@kS8ndL>xdo^jkp?5MW>cqus-*1QmTo@;
z!+`4cy``PyfZGeDzCNcSkM@H=U5IWz5Nj57v&qaEAMr;mY_9Y}X-l`IuTRH-p!#~>
zI9~@$k%skZP#>bdPxYD|4paBe!ur6YX2G(uPg=V4;~<2}vNCD$+Bk^j(tuHtb3Y1d
zMs)54ShLb#KF^cy+|<2WfCBmEu$C)}zQ^<eb|HMbL<;vE3>WIymY-%2Mr2}4ljv{<
zs2`CI-^7|toetvym}%Z?`;#RTp2HxaGC}^zvziG3t0WRU1FA+O!BbeX!-@of=gori
z!p@(z3kN$g`2@0gDhUEsNhEj|s2Y(3+hNTPD-r}=I17?swj~LU!60dOBnVg~k>EX`
zYD5wofi;^t33d!j2a1Q=ik1Xei@`x9fc#X8dMeO8-PRAPMRdM{HJiHg_Y68e`nXkO
zhwo}jCR~YOLuJAd#C3vSQRr@7@rTLe#sB4?hD1_aiZz=$DaHp#(HvC~GW^t%3_rmz
zp^|~RXl$A#li<goW<(O)gEgBv3HAt(fNhgIxj@lHgbV+&<ibBOXsBE`oER9Q*2$Fk
z2dE&C5`V>-)ldShq!5)Le3J^5Cm%tye^YYY%Cgg4FL=dHf}WxM$;@Z$7A^ZT`A#od
z)K>ZK7G;LhK-GvWoa!|@9JVU=RRWRe@#@q9Yx>~&<YY^Kp8$eN+`N>(AlRI7DjA}g
z5&>3#x)KSn3~N>jLyJB6&TVJz9fzM{1#=<O&t_ZmBN#r^>FJob>4{*XM2`=HdJ^gJ
zA*|Wd=`mG}UA4PZak6ejbEUK<lE8e+k|E!~Afhrv{!VSnAQG}tPGr6gDoUitSFvVO
zr^ucfisS~%h3K`ZKUi|(2@D%5H)2=4apuXS_#LPrkrcndnoXS)Gt{V+D^5?oyOu4o
zTemEHtaXZ!;4%7AyXj>llTRSyl)r=CGI+?Q%4FFZ)Rst=EwE-&Crg)_Ebg6!iqltd
zs#W(sM%4`{`C?y$F-KZ5=G_=r)R3|ePe?TO%CtEIRGCPdgRo{(r_E$FZJHx9LX=)h
zqI6^MP>CXc(YR@VZgJE61DP79gGv&qu^MYOb!tpgQzN=L6=BF_mJInch7y$_2Fu^5
zp)yfE3F=BD%Ez&0HAF#o(Znts-+w{nQ{07n*Q!#fu&`N26xx3Kg(V+<hB<@E#~E=`
zTf`jS!D8R-zd@ady!$V&+2OG7wsRN({#0AYt*=_T^<@x3;x1e4X^+w<$+<6rnh~A*
zH>_DHsQlTJ@7$i+ZV|Ma^L1ptnR1!kVrLSQPhhblzc|%2ppwy<CeJW?fZ7qsup8Fw
zup>ivsmIAW8=US48J1g;VF`u{HP|C$XqqOI;r*a?L^90Bn$?g2O&3JE!KWQmKALWz
z@}hr!sSTAE{qsApsJ!T(-(W`NMgP-`?rpjKv+L)}JA3_ns&xHa^;AS?rvC~yrqyH8
zHM7-C^e-8W193U>i=KRE(x=#)=>MGvVl8f>|Cc3tKQfXRqu>V|{f^v3e<2xwV{W29
zZ-kEE=CelfWOPjmD_TFeKYA0C+7U;7gK$rJB<mSS3jHZjz`eC6-<ePle``*4JF*MI
z)wfREsyP5dhw2wa0j+9s&_2FD9aM{$9_@=YE2T$MJbCTH$z<rnyL{}zA=r*vzHD|O
z-^ZT%3C*adF^z1>4^F`lq5A(Qc}rx-kl;H~8I^fH>SBxJL{LW}8CG&<9?6m?|NS%n
z;WPdWT3cV6iTCoJ3K)V<+6POLF0yp^`5>6e%@QdEYs@uwc+gVG_veD@5`BL*)~pl=
z&h+H9fuQa0BkZAKh<NipNRZo^YBmw!HVhhSL{OX<QzJs$CEQy;ZHfH&8rE#;{FtQT
zhyPfjd=>X;ONu;+p+lv}e2W+ovQ>&AzXug2vgEf|v#GOWZxu@_PH$FI{<{6;cA;bx
zlLu07Do+%p1msW>u~uTrHlWHxrfi8dn>tgbs+qDrSLxA_<tR(C9F9RmCCg%qpi*(7
z21|T76x5o?mxHlpQ|F5^vbax@R?E3=hrR4o>klnk`Yg%PgW*FZi%n#SSSv9l2dYeD
z$_KG#Q)kNVs>tHa-jVa4O_mh-3<eC9BJu|&^5n_KGgCeb`V^=skr|)BnoXS<2dbFC
zUy#mLS)F6S$@);2qE)i@T5{!w7*tfQtgx7SG=|IU`97#Skv-qVnoXTO2PxUpQ_9wI
zl|J^IAWw0d2ltU+^LI-E{RKmdN}v;L2&A=L=FtCv>JvHiN37Z5%b`#rsN>M?pS4>p
z?aJg6SS{IdNNu~!p>d%4L=NqQHJdtzX6WLOKZP~Fq7aEd3oO|)7sH7fftFcDps3X{
zZ;k~OC-UZfShK0~<_INk*ptZYt0(Fy%-h$Cu&8dyq6&r@l|`#<Skw{&nNuau3?ior
zShK^LQ|hp;;?(t)oVpf6t%ID3!9eEJ=Rq@wocbKrZ0ej+uhRG!ZO&~@MMb2CEQ$0B
z3@R#-tX64i!)5mT4Ah;-p8v+0O`Sbcl(Ss6H(y|zC@RjpWyzV>F@UI?Szt4v1<jS2
z@+zn?ktr`@&1#r}R<?@DSYB;`%8RNsUV4Mdi~e~T7b-9M=QW9_yy%}-l%n#Y|1TI-
z{rQ!_V>(+IJTY1s9Q+ns=p10>70{YlJv?3OS``E@XB`2HYoGq23UG-h-<c}F*n;3n
zB8asp2rgNoS1^(n<j-^TJ5msQc`^XU6a-&tgpT0mB}VdObWN%nTNea>HyM}-1;O7o
zf=j^tEl<8Pp&-6)P*o88IEJeq=tM#AV;DNbh}_smN|=Ikg0>F$QBWIVGW0OktdtBr
z<jHHx|4c?qyr;KY!M>(CTT%bJ?Um33D!F?rCJ)SuRCmkYT@SfCXpOIuDEHkQREg;G
zceu}nLAmevfce4V*I*UrOiRCZfgmc&e5J&yA>V$~`88;m<l+NB?T9X(jx{R{>wP_W
z?XYfpRqwbU5!BVZS*DUrHScK{B-HVJNPIX~+a=N96i_)L4Nk<GO`Qh22H^@7vIZkU
z!;%RXW2jJ>Ab&qc8xcaLNo2SP)Q?Dp^RZ@AC&LawGAvv=FSJ~~%aQ|kVqj1?Am8@V
za=^7na{cX~T1401hBceI>n8_YpUoHZ?$=QBXR`CZN-obXenrmup0^~!vlu*7G8|2x
z8Wc9l<aioXlSqyyv1U^z$L=9=xZju<`ci6y7GpnWmobcC@(HX6<oEKmF(MnZPA0_m
zpn^m~jKZ3&U%t(V%|^KY88~fg_9M1aI-Ny^!-j`lzF4l;H2j!l$&aHkoRSC|vW+rJ
zjsn#rvgB~A+0?_vgizS%E;zYjw#FWtjRcKBOHvGAuuw^H7%-Oz+hj8If$9;-(1SIb
zIvMr|k)bDF?RM|;c<;zYAEv&_k`Py5&`=3+BtVF;eKILFfhrP7@foby)Jd^Vh!k$A
z^qO3)yMJi*V6Re_Bf;W+OK#kUAw=bdd^1CzVJOX%DRM8UDUl*S#F|Z=BKx+Wi1#tf
z2uWVCB*{w{MpTl>Z^t4e@eGw|@^?^IB2E5+HJdt3#s+EPZT3VsG5IRHrOBR5K7pl4
z7kyIktda?^JE$0u0J~z%Y6yVVoQRqgeq@5mi<J;RCPU>#|NNvCl^6Z<vH?_H^v{bv
zP<he+`;Cqn{hFPVI$N`|hg7o@_#|KW&7lq0Fjj|V*SuCmJ9Q)ei<A0_C*PTq`q-kK
zD-uDhMbXYCOY}ZtBrn+iDUN<eigv!248So(JKr%vM{x5FBY85qCM5{1i*}w&24+Ih
z&hL%j5^(?4lkZF@h%e<(743|;+AcAA8$*W}ksDLx#3(o^XsdSK0M#L;L$6`YO6kxm
zoCCUmWHMypT|I7v_khR}_QJL9lLiM`dVU6zw@!(rxmZ=_5%Rp+FduPI%CjG+9gzt8
zc+C!n!Mv9eh`@to!4v$GEPcHSL{nMQBl}wZRAi;KuLCAZUOyhxlj!xOShLdbU*ySm
zuDWNcnj6jQ<NUr$=zRYIrlL)?&v_U`)OkRDyN|+<sG$;1&H=S0^5iV6+0=P5NzD_@
zn{3~-q{yupI#i0tUlXNJBw(Y&j+;S6iR`!uYj(J@!`q3>R&vF)p#bufB|9F+&`FjZ
z0UITDJO(OCWXGdev%{4g>@BhaJJ@5jLm_r-_j$YNWm_hnK(c8TI|4RJ?AQual*o?F
zv1U_e$7J=)(wvwh+w(_Q(&I1;9%}F~%TJp|%DeS5K|P7==)#&!ogI6t*};nPdYy_>
z?2eWMIF|HSgW*G^hgtLxR?7Uyf{GIPaT?Za>in3Z;fH`O5<f1rB*-NgKvaS#>H-Ms
z+pI8CW=I3nl*o{av1U_eNSBTwtn{W<$*-x`oE`yVge~`2vgK|JEGk>%_kt-4nQT^R
zvdo;jK&^?)xf5$Pb>>XdF=t(&5Ez^h%KX)mGB02#Q7I#T`+`Q9u&pvzo(I(>a^+d9
z+0?mmh>k1eQni*XS4!PZwd%j`S*q7q<#U8K6Rxpa_Kaup2`qb-w>Vva&6XK77Sx=`
zpfOmpsWWIFHI%)DT&?C5vjsL=MJO}Rk}`8Jgs72bwh?85cFG)?1*%Hq$kAA{sdGfL
zTi;z8ELYt4q994tk|bpeA}UD~C60k9wUSW}hLE8$O9nw*i7Xkwn$@rbt<Vw`Ui>5x
zl^0bQ{74y<7ya|H1XN!1&ueT@dC@<wkV54}|0Sa%dB4JKTxToXc2!ik)#}j?>OFuB
zYW1LWEo)WWcE1tzMMcYfo_uEx!DEZtUPuJ77R7DPTcY=@k-Q-O(;R(WEs4QX5gx|d
z%qirmRi_%-wjcckD1RkVvJq3<Hj?!SWW0V@7Tnz0NS=(YNu^Hf;<m12U?vo|9bg2P
zfP1<p-<ePlU-6_WZd-xj>XS*lNVN<@hf4oRs$ANe543e{i$P6@Nzg*9St$vc=gDge
zTTDhvyqCuWQpi=(4EkZFj!hBEhcGy({+^-tw`7hFu_$EQ2<k-i{5tMA-J4Weml{kj
z@gChJ3PHvl36VwdS1ld;We`MV{hE|oH5Au(2D5x1E3J|jH#dNa5xsjI)~qzHuQ8Gr
z$vtP6d2-L=3Mw!9=hG4@AI$<#dC@=5kWqQj|Gh>zqQAJwbartwJ-WE5*YbsGhP}wX
zZl-sY*WV{Cg|G6AxE6}l>R9WV+G@4)wh@;_it~mi-<cFAcD1utB8atE?M!5y+T`OC
zjO4{g81Koqb2`$mc8*O3;F#6U`;5>L+<dQ*JQ-b+{H^tB$4Lfe!fI!Y5nKZ9tS8@@
zP!OLrsa885!*G>WI~QW;5F>KqU@806P7vDV&qqMjh-uP?v1X+-=|i5pHaM9KpLn-V
zUe3NE%}b<?%hhsoN(FY1RH~emuk|m__4Ke$Rfe{2zRmQrSr2^+!-vWUxzIt+h}H&6
zE2eLNS`*3gb?$84ifNjb25nz5?cWlVOs%xmDY7EmaL)AyOFldSBCA|RNjY2NL?A*S
zrc2ED9jHH%8Nb1rmBPp)o_y!{vtMfvz3UO7STg!LyI3-k$peWbHI~SqP_l?6{N-qQ
zmA*BoJCQeAV9ln^o9V52Q&!AEM_MxH-55+%=B#RK)EP2diaCdX+7r2R5Y}w!+&Q2%
zcLsACvIVEuSL+W2o?c7(bYn<S>2p$B=@YPBV$kWJ0YnC^#+pr?L5I*7#CDqo%QY2|
zF0&-kr!l~&MEYRci4=xF;?pNVJBWPxIM!_Hd^)r>p9;A(P9e)eJ9~<d<<r@GPbf0|
z$dXV$z!0MnD%(~<DKL;2^*zuKBBQ>8HJdu4x>_^JPZP51a)o;MhUDKY$@FIoEGn5!
zZ7Z3==1V;K6KDgGN6%r+rp}{-Tl1(^sTaF*wS1}Qm7j;g(r#b0TdD2B<P%t_oz_++
zMFEhwv@>W0kxM&b&8E&J!Yr4sX4#!f_EJzK+v9luhUhfkl1|5Ah*1MmduKTX1`?xY
zgN6_pbqv;Q>Wm`h;k8`Fed<`TpsQID>I@7pDxun+hle4M_*4Y#Ao6J~)@<s0I<WO@
zH<+ugb+22og<U8#ebJIhU%;TEGU?>D&U8WJB@SH;T0rE`l~}W>bLgnn92z<i@!yzt
z3eKQYj9$L_r6r+$j^RcnRCikmrA9*D_5LYn43S$u!J18-TZGfLZtom6TD174C7J$>
z0YxQKd-uJ4(`EMj3)G&-o_}J^rp_M1zIQg;ohuZw*+|XnwCnAbaZ{Om0?W7)+8T?7
zY?t}7H>f_5KYL-#YWRazc8bbSUPXY)i;68?a)Qc>{&@)#Dlhuy^?0bf=$}_pqVl5u
z*Nmz;{mNLUvz4*2BgOx+Vkn>eFgE1X5#KekRZZ-Nj7EVdgxl!Jccu_7wkGxqi6GXZ
zCiZGe^sY3L7h~XZj($gO0N$Gnz%ezkKQux|aP#{{@?>;PiuzjD#6Fh{%!HcQXN=$y
za6jeAcP13X7e}gUVz;@$E{WQb$y+B;Qbca-PnjCD)v%j^N)c0|{}?er44gMP6Lew8
zWZ=X*eDrb4m#v$*ihrUwG?5-^>Gp#`gq4MkvfJZJq5X^?6SB%rK*fE)45%8>`TKj#
z4u@g9chLS!z3ODyRU+>qQI;<@BNvH2XzA+@fPfNxJ)bnh3ARcMI2lx!$bb{DW~G(U
z3QxXsRjj*)uw&;8*|Jm26>1wpk>C?dE1MGdk71|~Q}TxVsq$diswxs>f)+}gxDZs6
z$cc|&&8E(YT|%62<3^!WtwuIhzH3Q|Z)2cPDRC4YFoq12cTc_r>PY0nH?U??=fl(x
zAH=6uT~=fV9rjXaca$A}v}DKsVh~Z;u>fa>+FY3{e*iTma^(rE+0?l*Ny8O2O?LW{
z-Mq2`lTTn_F~^1`zNIolMuW-{88Q-UR>Kf<FH!78@|+iyk1lvndC@;#I-~NUf4&cd
z%8UN_E*>f``d??XVD$H5mu)#>gm$^1%_2<}k>2jhIOQt^p;R%?=m?~W>d5Vy)oS;(
z*NFFG%ca|s?;v=iR@T?B8)5ZA<aw}<C4v~+eXWZA{829!z!zGg_mLLpJwJNHi0S8T
zI(yT!fdyNS7_oZOz<l;k|B(Zyv3~~s$o|*<2i*U01^*vc@qbR_|D4SKIhFsjoc%N4
z5&cCbwtCaFrH#XL^;#)c&UbM>c9rvGr;sl?U0(j*rFb=?D_`t!Hf*|R)6&Ml6k&hy
z5D<G$yx5>KSgLHyR&%}HQk%Wn;eBg5(7-YD2Es2(2iiC&zKyceTg%F-w~+F_pe{=t
z8s8UtBfE#)o^HNN63{Z!QZm|`Gw=%c?ZDR=u?B8+e{OSszUBVh;r`s|{(RT{`5yam
z23}<Xi#8j0js3H9;NR@`;(^z_-#5J9H`(u^5d&{||K)A=yPD0&@3{ZZQT^i@E4%tj
zgH9K#|7Gt-cCk;<tnFI7WL}}Pe(0Z`Qg^khTB=vNovwImJYr=k)!L>3kEz!(?SP!b
zADLo4;Umj@KKZa^|2!_5_kX%^|MfKk$85>`tGgWD{`u@c<$RXWjOEZg_dGMxdY*I5
zZ8gt5YlNqm=brZDwV<~3#(nH5<5o8&fmn-E#?4u$_-GX9y<>#F$nM_a=<Bxg4PG_y
zFdjLN-CS@lE`(niot6y5qo(vWyOT1Bsg?-uZ6r@d*f`hSVTqTkO`20GvPb4?VW1Z!
z12o~Jae)zBg3xn4`Obuc_=6|aNn;Vi)hDBPYjiD!j(YVWMdij}nhgQXM4>%%><85&
zZUH!0vp#BK2(9ttwbQFXkc#&_d#%!ce}|1Rr!c$HS9i;Z!$a(9rW7#2N>sQK1BFV3
zc_0;Zrb==2a!_9)H!j7RO`RKC59ad4$XUZrEvfJm3<D|^4gsm)S|)D`{uoq_==XcD
zW;K3CmjPlK!{-@PUc@AxIHB^Qf1dWD@}mDIjS@e%f3}R7zGX+3F$c9?#)NkiLMxfE
zUxos$dUUuJw_3@JVIA4z;@f-jok?9{S28mbL9E3}rppq&1B~RwfSAtF@6bx-gk&I&
zUCFF4LP(_6%Z%j72q&|W*^ms-gq2L)2rdD4#gp$$IEc?@RV$gRF<j;J6+?#@l^aJ8
za+dI1p<T>e4r)kDTrS0$l@garJo(NoW|Rpl)4^u#@)HaLs`KBCJ71Htg8C6%e-GBI
z#`S2<Cvrv}QBe7407vCT|2)A#<wgIajPUbw)*U;Wv+k|VS%W87;Vktl(2!CO3D?G!
zSt{$uDogd`J2NeLS*r4zQ!^8REAE?9U6!~VU?eZdKb_;Q%TJTZo=!*xW<vI~!U!&r
zJuTyK>CW2p$3iAR#-~r)EoaBps}vUyHel#VIaD3Ph!`^zAFSbJ$HCZ@3EJc<CaNzg
zpkl=NvlQ>s&gDdV2CUDpnuv8yw!7cyURy5ZrPm^^w{-HgAePFT2hzmQn5D^;bWWDc
zly~|*4{A#E`sc7_rL5_*M)D$S;jA;yS_)qOT5`5O-t6SEb7gKFd+2`-doy57u6u1L
z)p(pKWRtx=h5<?qJG$&$Vr0ZLDf&DL>PO_^!`zSfjMDb=rQKu)^WhmeSJ>(%J4bKM
z<bhO?>ge}ULr>Vz+(bF-ypxQ(-@=-e!p`fQF1mcJ?WCBjAO(B)(W^vbPxtjixG~d`
z8(qn8Lsy#KA_!!HmdX@4092M35vF6!4p)je<x+RQf+AT<ikybQLk%Am6lq#2Q{)s-
zSt3PF#F|Z=A`{h7#7%J{0i<Ecj*Bs5sO&h#aB^V*#5GbT#zml>L}HweHLD>8nw5&I
zmS;1lyx5N5SuH9r`seF&R9^JYw?j~Q(f{_|^o+`j{zn<52X6oD{Cw`P*ozF-#fKIM
z&tgMcJv3b#TkS<YZ8QkPCeM?ed<XX;N6oIW>!54uwa6Wp&A$p|+C;=!oS463iQZdA
z@?r?Q&e3mUhO6K2otg~5F)s}5&3XhDKYoA~+}z7Zo{X-^wt)w4Y~AmIWMC%j_s%tf
zOTc}sC*PS+5Wn3!g*_9_KDWqH>b{DTtFZ^uOO>p1Mm<-^*7|eB?7`WC4+@3bwHW9=
z(?n5lKL#B&_~_my2~HasimkrFVtd#D6(*)sYp`aelq&1VYd7lJzGB~{2_CyL$vESa
z<KPfEu4GEtY}Q?lfkGw6do*dbnj8`PBtl#YDoG^7C0Mhm6JlbPF`~#m*x{Zhl<V2t
zU^$yxm&>!S>lDJ@f%%CgD}Ib2LuG~TP$9&MSR*BB+yiP#q{iJ?v#C>K$K&SCn|;FK
zRoOYS7cV|`_MAm4Ls8<NmVEdJ1_zZ7x_c!dKFF3yB={?+9+3nuV9lmZf(b)VEnl&0
z{))LPS2o!YBEw!^vzsy|GI=2LrIJDSnw63avV{^UCV*-ZNiiO4_Wu`BEVd-YLJSrv
zDYl%F?>R``bejkEM)d6*tl9K^yKrT8<?Q)$m&{%?cSUx|^6c_uD_6-2h>e!MU5CL!
z^(|?@>n)T9d=*raNQyGnZ0e-gsk$~_&i3ZnlMMVbC6T?TFI%$V1`H1>8+6C}q42LT
zO`^edpngOeT!S^6It^4s`Hxt-{#O_VRM*d-EEWr;zCNcSPX!NxIuYId0M=~k?%rwD
zirEX7E?hc4JA1|amD#1Um&^?%eIstROZwi%@SwV$H0l+mNu&M^P(LCKUc;IlCNyBl
zz{*v#SIw1aaG)g(W-$2#a(_Av!lp?y*bmf?NP~T_W>crZyOu3qwQ$M8QyE%PBskTQ
z0w-ZuP$R)i$|MjmOLG4zP&cCckH?x#-~B7+ESP)T35!+k|EQ(=FTk*9tNSBnN$x)n
z)Q#x=bFgMpcmH-P*)E{U_jg$O{+k#ORNr?|hJMf{$?>;>iV+=uGuABXc#|{Q__M|B
zR^-%INuIIv{8JbbZS{Q6Cdu=UgNhM7|Crb8a5ygBHDG?mEkR=+u&A>qxhkRbe!H*R
zrT5!1`2_MhT`6~HQLi;na`slBmPBW7jx{R{^LIE|+F0lkKg!1h5o2G-4nG`qge4UY
z!_c4(^+PE^UNTIgz)Vm(A_cmzW>crY*dS8<Wp*k&7(M}ZELpGy1BA+g!%-Ft*(MPo
z3o1w?!f9BusS{ybkO)1FQ?8aBrzbQwTxv;$OE5^NR5%i)g0M~^LjzQiNQR5CW>Y7_
zE<rM|T9Leezb8b8do1a2HwFro4y1}9oqZA^?gEu065>v*+0+TKdk`@8ts3u*c6Q~f
zdu=3t{HrA?Uci8%lHz@6jELGO5#xDKQ6e#(#hOi>7@<f}?ax(uoStk?PdOARCfs7T
zs2I=W6IfKhks@fHM2NAVl0-s`!J18-5TP|iZ#BY%d6rC=gP}nU5Ad48GfbkuEKoZl
z1&+p=O`QUhgOf$Mkn47`?0YRvA<OQ*^*HPD-Qg8R)shos3?3>c=Au)E+ER%mgP^)Z
zk_=$YrcROxL6W#1Kv92B^*T#(T!X<vB?qbGNN=G;imO01iKMs!Yj${%g1;!>QlskB
zLixmlmZW$9gC#Lige{avaX+Xgkrek~&8AL@(8jg1j$Ln3apDb2PP~SpLgfU!aUC*H
zqQonpmPAUtgf*KwB|@83wbELrm_IYJ5Z&)qyEVr?Og@1%2fSG&tdqzv8B~!-hCQ)n
zQzt_xli-Qw`bw@`j_h8pvLwXu7%bEP0cR3n3nfx41=S>yViDGC>ZI5)n0$GVg))fq
zEXi;V1_zZ4q<c)VWfBR_0@WjuU<1}{>Ldv5T6UKP%j`Xcp$~0`=y0ng9d5>eq0#}~
zwT#**5#uIMQ6e#J#F|Z=7}JAes9bUUk8fsKZFs&`&Gxd3^h2MYzHpQK^;P~!%Ev8f
z@)!mal_pEzor+kCCDJ?!s!b%#!&tMalV)O&G~ND#msxh1zNZk$J-7XaUGBLRlTRS`
zB;A^hHBzF+=AfoTYP|FR88s#aQFEgRyJ6y<n&xYH_Vwa3Bbn%7mh_kjg06h2S9&(7
zaU2>(l!n$X-(tjOBi#S+PgIN7IJ!W!iCj4VYc};TvS)}Z)qEd&h@oC_tMLl;!N_+=
z*H|(oi(x~ZaE^tUBAY3NlG8w4iS#%HYc_Ry>>Q*=z3OChW%j-GU>0(TB`F#hB2-f7
zzKBy{mBd5eQqL%jlc;bps3DOG7h%n&PKD6Mr%Q$IdQZ-MtR+N;yDjN(7lsOz4)Dfj
z$Uun_cY<0HDRDd2Z0eK<m0Ng-@#lL&jCjG45zk|wP#FQ2TLkTs2=OeaB#{tLW6h>c
zh|m^CPl-JRm+Q$^+3JKn${!-d_}lE(Bx9L;0&5a@iz8~IM2s<@qC{eBk2RY*F+v-l
z)f#&rFhqwrmUNhf!9k4?@CGQiOd`S2pn60S9ECNTItfD6{oSQvtx_t4Xi&DK!5{_*
zl?HHi|B!7G5e7g7iA3nbnoXSup;WTFUa7Fp#Aka8rSOfPYb@Dt6^06x4R9(MGEkz#
z6`+<xN^HWKO`Q^Z1lMBhc1+dDxo^q3U-4mI@d!nU2P}DUKL!nz7qiiA2-!-B9QT2$
z63KBd)@<tJ2yJM(d3dc}Wb4o7x5vV9<26fuyn>-a<p;c>sWeog$V;HMM2h?!Yc_R?
zgqEW0<ycnV>s}rVGh?4`+O0Y!Gx-Eo9q>{#XrDxgJwYXjgxDQxHg!UTmZIJ40rHB&
zHnxlGl%N_W#qpM;Sc(Bd4HxiIG-{(nj76ZLL}Dz!noXS;p-jX7P}rJWjU^jZb~YBu
zG|sW4##tCNRBFJPhHRxojt!uyL~_)zW>Y7}ZoxD(*VB_7I$4i=2I^)@X556~LS+WL
zHVK+2KSFpTs3nmU*JI5NFH)3CWgRIVvn0i%7%qvCB50yaiibfhiKKW4Yc_RKgr<=S
zyMpXKVdIzKM1sXu-?B?PH)rw*q@8ev5wTLH#ygk~8$@cng*BTxHKqh75q8;-F@u*$
z*Gm44!w5lUS`wry8G^uf4wbgbG&ulNmq?T8ShK0qBvgLDHoR&q0gaU3Wb?h*B76A6
ziGIg6Ysr|?Fs!Ik&2n@Ptd-d^c}@W}Cz9txtl8Abvrj0-_>s9@WP3&|XXUk9ks#Bs
zq|C(_LR8AkgDImlRwl|tpuR++oR2k|I#CV^5v7ks8Agvnj(w^vyFOp*XWJs~1f?U*
zU6!P|6GMwiniVi<S{g1B=yp(dB7tthnoXTRq0RTKizB}%TxF$StkgUbWuCVr%(ECe
zRKnmjqLQI9L7oP6B@*OGtl88FvUf1T2-w_f(S=g3N218s+wGP^W0-sb%OUuDMrEo@
zlI=lli6j|?HJds~_6m~3iz43Da&-`yWyz4EF>t6s1mAacDH60(rpHmBsziDmjy0P)
zJwlK4G*Q!rkU>j=3}En32{H#wOB9yM6zKz%B~qjZYj*fignf;?<~_cloMo=Eq{tN*
zJc&{yVyR4#O`x(wihKrZHg$>|5S(PXz0Xi*OXXUA(0z|1%dSSUTI7DGr!J?h_gk{%
zJ`5=;Tb7~$CEjS6IQN2j6N&Rftl89w6S@T##W1qMdBu_{FJahFse)gpjf@YO9DfHj
zC6ePWShJ~<V@5ElxHGm0Dcyyt7gpA;ci*dz5M}ZmcI%uynS284oW*ENi8EOy&F-Mq
zMAGbvHJdtVR1rlt`7E_0$|4M#q$5hxOqm=DKuw9{n2R-=IypjBCdK+-*4f}~rbQ#j
zS(e1ufZ;+V23%zlG*Ko+9n_LYiVD_j>ZAxojI1aQX9t7za%5_`$&wg1V%SiL0p}o5
zGi7pI4{Ayz$F*3qsgonL$I8xZ%B2e1hc5~IA_RHVk{}Ob_)rM~@3E>(l}Yjts4bBs
zzrdPJoh0G96Yq`$|FTx|qC|uxn}6FbLwyGWh{}@xN8Oje$yHSAClIpl`ywzfAtZ$9
zAqz`bog|ZF$joGz83JKzd%Ew;bf>4g>FyzmMn%Q7)$kZm6y@o|hx$Z$E{LKiDhO^o
z+)xxj6h(RZ2SN7bU(4;Tu3PusKBua0-#mUl@AY=(oI3T@SLd9nTek|o?!>LGDotKS
znhT`K3pll*(?nXiMwhJFtCQ#kK@Nhn8WQA0A|JpI2EB64QA?E?tB{5Qsj&j5Hgsx8
zVTLRbAdBKt$)e2<`Xa-@k4p^su?vR})PbPGj9XPzlI%oU3nWP&PHpHUk>1rL$$mhU
z2@y^d2S?sx$dPyA;DK@k{jSEPrYc3=jx-iXk+<U1hE5S_6XYn}|Jaopw;EF8%Q#q|
z)IfiaS5-rm61O0Y1XAK=oZ8SSA#MJ^RuDP-#2*dG@CXhNC>hXufUvr$G<XPU9*_nP
z;?#yt18M6Aas&_gmOCFyW*l6YeXH5PWCoEBAn!aKJ9RE9s1jl-(ncU5CgIeEP6#RG
zB%73G$eUV539@Ar`B*pWoccP;kQ%4s@PWDsbjs;gRh1;CBCQ3IWIax8=p>PbtZemI
zcC<+LPmd;&&IPC`Lw*e5pn>uOJ!Ey%Ql&-$X(*5yF`U}asUc5#(WOAq!F+b4YIhJv
zZ27PuLB?<ZK?#DN^m41KN|URR<^pN*0i4><X(GQ}VbBfG^<(VAJO@L*Z^)24ao9i^
zf`7Z>sHRGeJCK$F$#FYQZRq4!S-rv|nk*DXtteU1ks&DvS>G{OBq<4n(m|7_3~BNY
z97<4{oQqCUJu9pd<_V<DK*BtZQyV&AR!D>yiR~M(==Sqb_PT}6la7)dE**3^>T70W
znq@>jfHBQBG+jKZtCD3g(qbT47U0x|P8K<NWj;Ai#x`uPPzOi08**eD4jQOS!6&bB
zEmdl?BMk*oV<S#&=+uz2691RAdkwiUio*uV4SZJOsHRGeJknAiIkGslp_5~^6l7SB
z5-U&r7PEiakSU+S0R?3W`d}!QbX8t1ITrA%w7CIkHjp;g<J5*u8)-m1yf0_vbFn<x
z3Xx{9d{<RR(D|7mUGBpH1f>glK<rvql_oz%nhT`Ky*Ra@(?r^lo#d_#QoLqJikES4
zKuNI<o2`<XsT6ntX&jIO&*IdEP66o%xhOq&Fh)*(N@mE!s)G$D-exv-Sw-Xn7`vQ}
zWrJKvl^83Kb^?jfj8hvrG2{`8L=5K=^<9R{*olJ%>L>6c7P*!xHTsZ-0;$o1QyV%p
zq>rJ>2fWffb;-6enWD9q%&5jPF>>Ib!-KrjkSA});RNLg`eP`+DyyV<E7EEpX%67j
zhE5vkyWH4lF&mGOsrSN2j4WdJr_GlQX>$t>DJX5w-{tyOS|!fSNV|c=`5aDd=){pW
z@vf{2C=@N{s(?ofdGZhr7${H9!X_wE9aUO9h%^&OiwAIOL#KuG{am$2b_AUnUpGrz
zrxN)9(pL2MbJbN;Nihj&C6E*oaB4#*g*5MxERe|~vg)5C{MoogvJ<iaMApHM(+$~i
zDh?p1&p^+6xYkvr$$F%@K$@(<sXgRqvL~BQC;VtKWJr?)4xk`uqSjTVNepQ&kS3Sm
z)P_zIX`L#2W8t<^WXzBoSL0BDas$0iRa8(V#0QWz0txXxoZ8R{AssNpU-Xj|h>nnQ
zry(Eiz<~ng!v<^`TB)B(hue`R0_kuYPHpIPkUnQ%a5<>(4?`+Efdd3e1@z|(tZpg|
z9!Hu7q`_l2wV~5M4in@vLW`Ya>EOb$Z<vik78CgZ#v%AHQK_Ixhy_R+frOZgQyV%V
zq>m$`1+q;8-CMiz0lstke48ON+Hvqey#)H>NSB(b6xoP07D$nEaB2@RisZBOG;AM=
zj2coTkAo*Til{YJDUwAR3#3RIr#5toEUaGo<FTIfQ--{_0fz~c7pGx!&2sfrGF*?e
z5J-mWaB4#*!-8rukk^!nWZe0F{ysx4{1^ublndyC)9u=+RJa#uAdm`o<J5*u1?k1S
z-EopxdZ?-Lo#e}glz0IL4U`h-7xRu<s?>NEX(*5yPvg{vP7Qf6PlkNkL=w-TSaDc+
z6~F3sv(d;3A|JqL1izR^s-#McW~7}!Vl?5@hE5FWy+|}izK@K>^K`+6{c6N=j!AYJ
z@}mz25Y%&^-;21`Ri#M}(p(@-x^QYkr->9=>^MwAi_6RQw;K}VtvG0)1VM)uM=e!q
z96%Zhq{e=n+R&*Xod7jRK13+ey%}=ZWQOjA?Q|cv7_#GL96C^Tpih8OE2<LYb4Xi(
z1i1;PHgtkW%XH{Ag3%%JKe|h)V<Pk+LvB2X0|v?s^fDc(jw&r4K$;1p#V>GbL#Ksw
z=97J@0zU{h9?Ltwrk?suv#fLykq;m%MW6Yk)KjI$1f;1zdW^@Z4V@m2#B{VUTw~{s
zQw=Gy9tRNAduSx4E_GFDvIc1`kR}nF+R$kt?IoC_$kBbTM+Y3CC1FU87!DRFIna9v
zR@G3Y#AQe$ft0uyr#5s-c#TzDropZ@q{Rnt&_HQnGFG9rRH^Ykq@h4+yc?%BbZW@&
zU-|qOJEDi2L6J)i?sIsJI}GV@I}RQwJ@D^e)taglxeaM7kRo5jsSTYX@^|&I1UVIM
zgv=Ec$p_Mz0u3%?@RA(LkhL5RvOHl(md9~GLCJ#uuHLWCDs3J^nhm7Qqd2u5w6TVN
zcwlU0s(0eJahHyz`uS}qzg@_07qOcyf7<kM<5E53m(;KSiw+pD@@G}#qo3807v*o%
zk{9K_@ENo6AJUQ+<v+;eV;cC64y4vkAT6+gV7cnbG1`7xNsA0`NNwXScDBbqtW0h8
z_?J|N*FS70DoZQUBV;;fPb5b+@lGc*R%Eo8Ocx^g5q5s4JrSPGk6krJ&Efpm_$}0c
z)V;UTdX}cfGV$STK0`KP7#rST6LA|cm=|;oDF`ugwdFTe<j-XCo@lO2dBLfR=G3-g
zF`pb5CC8FVh{XaytTe5e_~vqX3VN3rqIa>DyyylOa`fws@uW@ZSfStm_#MFjJh?e;
z7p9<lr4~AZn{U>V2cxU)5N35{=`pZPo9IW&nWD_&n}UIPb~Bm$B89fFPhM%E3fwnp
z!6o4SB$IE9D_Anq!+SH2JCSU>LB0Z){KF4%xZ3C}CgpHP`2!p}K#yFCc&%7+M2Co7
zC8gEAkF*PzAl->mYnvtd&K*qN)BEf69sf2zvezQJW+pbrim^@EbUH>>3tx~d4)>6B
zle{pJsPG)ItuTO19t0Jh!C?cXg6CA8lM4PdRT%OV(p(@z{=r+=E03yc<o9p-<9z{%
z6tlan3_XNfBEd0tn33QpB9BO%pd{D<^#{;8D|A?fG#p5W#W=N!!&t!NJspNdy68`u
z<NN_6OH~xV(M?P=@f+K55J4&9IRr=fjYzr13SqV(%?1*t9j7*Q!mRWs%t&l+G;L*u
zio=qdd7~j`_To^2a;B#~Zl<cz3T;M_mIG;%$EgjSHYfShhAxRCvwkC^>0&aMwxZcV
zi9Vk(q|c{uXhG?7VSVW1Qh9|&pF-9ENTVBYYD1^diT*UolXIH$nJ5`flCQqx0p%|Z
zDfBZOQcw!@*M~xmiYxTF4{1G+K0n5(4V^wUf>$)k4k?W~-Z>ohZL@^uH5^J%>UeH|
zg-dwKl~!j$UPjstB+UyrwH~CwzBCY%{(QuSm3NNmvGSsPK0$<)7v=NWPOQ8rpU+ie
z<wf}$a06+({b>2*rNh?7zI15veCfb;`Em?~ci}p^dq<@;to6EKCn*RqYPKO1bBujV
zzA<x*Y%0-pDtvDsaFu+f!bc5p%WKIC>SsCb^?uPXWy@$LUL-61<&@=P!QeZ!d9YHP
z%`EXE>?2wT31)ssOCF4{&V(r=u<y%;JAwfk@Ur1{Ew}{Sw=wy~I0FBPq&ejc^2jdt
z`DEPrMZzC&5Zk;Ilg+=!0R?o;r8VB^w;doNj(VwS(r=Nb0n?;k<J8)SiJtUJCf~T3
z#3|%WKHY@akvz}*j#=_NjmRTHCA8sJ)oysZK5BzcMw$d{@FQ?)u??={I{mx+3_cj|
zr-uhSGO05RTm3W~7SL92t=;N!os{@^GSV<$+plG6J^gs&dhzR|`d6HvOUJehn|J_;
z<Ye-x8(AF~OOL&~G%;@tj^>kNY&an$+a=G9T}Swl##TP!T}8F8N0C+n+xrrnTBWb=
z(vlZrK+ZS)F<_10Qh#4Fr^gyBN)0DX5PyQ;Ve(GuV>mpa{>JmD6hTI{UW(882+}wp
z2S3Er9s)k6Nx{I+@7!b9)OR7#1o|D%GuV8ZTB@e%d%lab6xi<H#;H|&&o{N?h410K
zGxa@Kc~L$eGhyXLV$CN3u=1jOK6!(c7v=Y84Gd`cWd1GF*!j1ko%3&3q|+nhc%wlp
zZ)M_+r19uGQBK_53DXKn8G*}c-`A0X5Mv6v2Z&Mk5+?8IIqSOShs|Dit@b@T5X5S%
z_C3Q8z0<Vh1^-Xx=+}D^&UK1277V~XQ=FG+p(D6?v6eg-U7gIX_5|EJf`J(@0e7Vq
zTmtSlGx^520{_vt+iKsNaJbrzPRzR9h(ibHkxR!b!*EAf@LcBmNu)u*l;{&UwMt6#
zF(&UhO4sQ${*Ar3qY^^t^4=YTbf*+$rQR=yStg^R`*E;9n_t}u-P!z}wX=~I?<;<a
zG!aOMAMpk~1XlGndtnjb+jzwCL-JdrR}CBeB_yJ*<0B;tD{ZdX=OGnU+W&c^wZQiO
z7f!9>5dO*JJsm>bm+>C!30{JH2@q9CviiGbo?#`CM`X}Y&!B#D4fPBWQBQ><$03ac
za^zT?+R!<&(vu^}N-$BGa-ktp&c~qyWlD#>J8{-l;Y%;lWFTL5;M9iBm*YJ7VrA(!
z`s8zxXfh#9!(L@bmbc*`f|6y6K3SxyDol9`(q14_-h@*dI#ZT=GNsbj9B+lcWJr`R
z;t+xo#UTDvYO2oTejaHokR_kRsSTYaYdl%Pj=rh@NH#4jknaGT3ke=Jq|EPda6u`f
zenJ9HS=_6w^5-{5%Ypp)6;ACT%pW@T6_d&y{F(kevy^uVkq;o{HQ<kXwN?I1L|P8y
z&*3<=9{j<+2NCa4_!u24FJ?L$K4~*^XJ==ZaDr0i)NaVmjkv~udN*~!jMKY0>SVXF
zn4mcaX&8`vXEL>i!o2HIG6Io8Y$Qkaw;`J|$MUfed2S|a*ve@nn83LibtgRkR#xh$
zw0aV0D6rKBacUKp6KC>`o0eJS;0B>Y!J;2JI!NNkj;|+Xn#|i?hl2^~47SxAMggm<
zhNx?i76bWm4Nh(7d|B!Mu#yuy2y(9>LGH%k0wqZ8iG8p7sf_p@(nKI5zJpU6IwO`j
z5EE&xgA&ggQsQYGFi=XU=UdhYEuw-dC!R#w2;{`yacV>7#PLo}B$5StbGE8skb@u1
z-#3dXO+-F`;ZN<UIZ$O)t{jQ97s!>RIJJiySKQ|u99-!!<VqI~B&aVzbH%-~Dp$@$
z+6&~$R-D>Hjw|l-3l6RvFyzX997ut2#l5mBSN0(71#+c`Q|rMM>|}_TOyT2ctbB-*
zDT}+Tk!*fnUu@8lBZM@qa;qUbzl>`ZC_8H>%AU2eTV71C+=4U_NXVO++PY7$eCY6T
z<HyneTsIx|U-Z=C1xlncPRd6IlVtlKvXi*H|I{OfJa`BRGH~KW-NM6}9pkF3kmEt5
z-9U0YfK#h@t6wnr#?8YluMMLktj)N~EM1;T<O4{TYrmkZT~jrtO+p$AWXS}aS`U_B
zW0Htz+=XG~oiPn7FUsd@gR%0We7?{fD=*6DpDAGFMfnT07QxZ-$%iH{AKuX2<(JzZ
zn%KLXlH(DxnaJ>9G-1WF2`e9EizTJhAc5=Q?xCI5vewroF|BSOmY840<m(=ZT?4T1
z*CuZd1hE=lo4nN!y#reEq7&@r=-2ye6OS!l{xcYeeYbr14=scQKmT1z9*nTg;`!QN
zoZK4>(10&a?$&}!!2LZY-xzn`pO3qJaq_=7T<w@IzBu_S4jrIZE}d9@b>$l$JU=`6
zGtw$xg7in6S|vexgvmE9jh;%-BYV6khYBpY+bmgLNaT%^WhE4sPJ%UhRi$i%#o&4#
z(lTJf&&H_@zu^bc+4yd<sKGHx+G5!7n{YTldwb<H3IE<+u9P|;dOgxEV6&fvQyY4-
z&$IEfqU9aq9PM8)Z2ue%5NP|K3hwyqM1)H1R3eNZ4FnP)g;N_k5t_(xIagTu0vlHG
zrT^q4i3)BGR{XmmD?W}x2g-`d7ij)o!N001MLvqO7D$l~<J5Xk1UvW_lLtJ_#L7Ff
zdaS%CpO1R6@}m5UwQ?(a@?yh;hE86b=rMUwwDuOsYUUxbAihB6Gh8Myp1^fA_pU-~
zTWb>IaV<KEQOjdYzA>W~-${(AKS1TGK*VZHVoV~1A`)D?GYj-4XvvGtFrLZRbD+R5
z_L#(I2?k={NsJS;5E2L<uO$ygIG9O{o?w6mOk#9t!6o26m&rHA9pEQ13OHQt?kXlR
zayWFLK4!gFTI(7oJSQ_okk$bcmK08{lCTUh`Fi>&|AzOP%=mX=hsomWkK=HFHvVb2
z#`l}l_$bmsAOSv%QyV$~=9I_(WVt{xZzUX~n4cK(;D<OkpgcGW#{)+l)#1wzkY)nO
z@O_-x(8(~h43vWfFB-DoIUEj97M!BT0=4;{LD~mw{-<ziLvQ|N<>n`6N0BiX`9g=@
z(_~$>gAd2wV-_8j6M4i)0Lq80dVEl7s}kfGq`^Rf9EDRGIzf)-F(T?eJLDiszad%9
z!+``POCOFbpqi`1=|&n4B+ho6+R%xEn>xMHkTGw@p#x=1Cyp`xokXR|8<AE6sj?TR
zHgu}geBJT|Lw5WZ4iG3iDxVKKC)VA^rOHbsl@Omn8VV%Dr*UdSC&Y4RGE(!^%Wn-C
z@@pJIP=-`Cxd1Z6k%Ot#Rw?sKq~$=${0yhogEH9H7~(YupD4r1JLe^_@}hh`Z;zE1
z<v*(R-oSnh6L0Kmn0YR*VWcIaEAB-l|L(nmR;t$9mS$2AVw7OBPE4;hF?rAFRnJK_
z?P=<%eWL?piS20Gk~h3)4+JstwxuA-vuiIF?QAqe@0=Rw{h6$AwpY(?n>cRVrQ{eI
zavR=|x`f<PkCA^J{^5bKm8o9-kNy0%lix1nw~P3#hul((=1&r_OUG7rm)6Eci`iH%
z*~;12TJ17hohzDKlbM9IckG>G-KEoO$hijNioBJ2h8vJdlG>+slZ%ylmi)AReCj!J
z>rOpSuJ1{`z^*T{>r3RieO&5g_V5b1UP|2ZYxKvd!wX91wGL-Ttk(Q!h8zynx+k08
z-P+l;C7s<<z9q8pLTe#AnvYwpezvnJM)HN?Sc<`QiB|Ipm&Th{-=&>tSU&HZ`sE*~
zT|QZsKQkPcI$BJo3-q{?k!(hCsUIelLCm(hyHr}KS}t`=3pL?VuV(U|F12xIKHM4z
zVl`f1f7uYdTeRdw8gVm6zd@_;e-jMEzAvzUrG=2-=g+m|!3b+-2EGTcz8nnDfEU;=
zXu&1mewN8M#vS;-$8=kTfAkMgUe(T^#0%`#5qU)Z3G~XPW#wt4>Tuxso_Yz=2;hir
zAx^C_qMOI$JyYnqZ|&YR65DM>3&U1A&35OG4(78XQM%oEJRMVxTkRy)A>ygh@*OyI
zpe<iH2v2Hx|C*^|{4Gf1fQ;CLQyV%Xmc;W`tZ0#U5xJz5@T9^%Ln>T>g9S>3%BMII
z6+Eh>a-e{;3dn&RPHpHMI4V!JzqrB@jUSB_i}`4d9E_UW>v#?IpN2&E4;(g7B2>=a
zmWbe2FO?1djx-I(hL7XahR%j%u|$<0@S(!b4XN-yIAEYusC=*^QNgoPDi3~wv<t|C
zAL7)8&V!k`JUJyfIuN6CE)Fj+?nh>s!7Df{piO^_Yfe~Q54E*lM4AF@?dNc6Jz5((
zu_Gp?_?QAK?;OHm<wf~?iUTVz%Kw1Yn2MH9CZ>*P?8MYUd#=Cia59BrEK`&xuealR
zj636Ly=u)(Z6gIChFdoA#9*|Y$v0*&>icm;DiFkK%uNj$qL<K;7qpLY^u6Y$f|)sa
ze=tY`W=`I#1(zV}yEt5jz|nd~_P1ZyAKlU2xhuN4t-r0Wy+69Sy>C-*M^Aspj&5n1
z`W77Ab`KR3DmUXm1G<vZ$<E+uyRnFW&200;8?(<L%>xG1n{aBCfO;d7Z`@SNY6e^X
zj%eSe9lh<Fqx7+}qmR&{20b1khMLT}Jct7dN{`BJVU8%r=>e>#LY4=R#sbOm3!K`}
z$#Q&8TSu8BQGzAoPA#HL{jnKQCJ}i=rT`^MWd)a$D9}18B$<FT6iAZsIJKdZWQm(2
z?r-bp?32RFsfM&zkAnqD3uT7JFTC(tDMVO<G!00E2u^M2L}0nbT9A)O7?L4|g9b_l
z-(17HW(pxLLz)L9#KkzZp%bEM(~j<)?Y(_%G}o+|U|em;i4Wk=fpVhqQKB<$)U2FB
zi}xWd1k&Q&IJKeEqMTLK1n{b{E5{DlE28c&WXJ6|kf7|Sl~vTNqe7C~kcI+D@>QJL
z&`GkK#SuH_1PxH0Fl5N%IE0`KIn#IiTC0KzJsv|^38crPIJKeEgC(8Lh^!`^i+^G^
z7Fj^#0~m|=CY|oJQiw1YX&R6Svv6udC&E!%I(M`IXSmu8*{~6Z4b($aK6G|w7&WS<
zP~se<eLzZ_iBlUoC5}-Cq@6pux7rh*%DiXZkQG@RJWy6t&d+eNqIT^RVx*BK0*R5t
zsXa7^A-h*UV%%UzjO%gm)P)%SwNr?39nwS~F|Ng_4V@T^T;?>RU2Q$m1lErY>2NO&
z6et}k#}_;Ogjbyu65Ne63`m0S;napsf_WX?ef@3So7$sWI@&uo_bCK;!H@vY;sAjX
zV71o(!>v9_<3Eiw3fTBh;?#!T_=}{0dhd?T_GouoSGzogz2c{4qlsoBAHZnB(*t<b
zNg+WK(l8(ij>M@Aodk5gvY)hlXM1mSTYFn?e{^Fz8C$ep*wfL=<}jsr&}Ybo9vnJQ
zci=Z)>0dd87F|dSfwVXmr#5t2EWKby?{=9I9o_xyy*t}FB|^N_kPruOxIhW9)+;{v
zR7;`4exz+cD(u0j4V?-TJ9li|+|esF`^|>U{y7{5&}Ki%tJ&=;C{275(hgt~--uHi
zdJ|9Y*xF52#zZ^1H@9CXHS~jq4gCNP31~wv_iAXVB1&`r0%;Afx$noR4ZXSNbab_C
zZ6}Lcwzqe?wE3j_%u>DyL_UC&Z<S}8E45J?emv47V8g%qf2`r>ZEoMvc7A7nw2A;y
zcVBPV^lOkfyPl`2EUhki4)@&Zqcna5X%w*WSL4)%-s_jhVD+^1Z`#IvfJA|qAq6hO
z!2<R8p0AI*Yo!q3Vx(z6B3y`58#)md3do{l_J<7lSR{~G@Bu>>ybp&7lm*HMW#yG?
z9#v8(@NT4CKnh%iQyV%3=2zj>-c43S(^Nnr!0m<vxDAI0lmN<C#x4X96;hi2t4OPW
z&Hp8w+R&T-Sh8bN=Z*`aojZElqQnDiC7(!i?bv*NXFHn#?&;{>Dv{xFLoz&u!v{)+
z)4ei6K=l-AJc_gtNR5YaYD1^Simvvq9lg7veQjIXqdPl0yT|}xOM7oSe@D>W(YGzS
zrL#?@$AbSc8!*fz@&ODO&i0}Qu!;&rW+CkaQe-+#ZRiwP+}qxNes6bl(~iyU(e~cn
z9lg<w=Wih-k$lERLsFcBLj~$HDu;(UUmbc@N}<7-NXvjU*nm?TIt@<r=|=6{JEJ?>
zdP!ap?Ir$XXS+m=tRXehIHaJ|s2n8dqz0;@3SE*&Yk_nb#HkIPE~~0h+}yKc=h|pn
zCz-B``g7%aL#|whg9*x&Hb`voucbniYmue`iE<52tp`!C+Xsm42l$7<Sb67n*jRZ{
zKL3gyD=*65sPzFOEuU<EFr%^C9~|%8{-Ek~S8F7fCfiFz24V$Ea>swcbusrYLMvKp
zKZHMNF;RSC{s$(12<(S2@qSdo3q-8Oeh7z?LJ?WA-H!!&<Fw>OKX`?s@3kL-#s+z8
z&xuK!=yrFG{iRn01MuW#wznQBOmV&|w9pagHfza)(G6y|iOyhP2JAM`p#_(Kdkd3q
zj4QzJ)0xHLYLiiHE0M;b1NATJqA8E?;JJ5a5@{GPMH$4YRZ^5VllR<ELZ|ciH~O4@
zJDP6Fj^wf#D^rvr>h;7PlYJwu!{GsKe|5ghtNqpLDGu{mq>Vr-T*I645ZFd$p#nmr
zxGyKCPIntN^!Jb;y1p+}Qm>Nwg(kS6c^#G3{tnVmU~7L1r&j6s-(d2d9bY3S@$dIb
zR8EjXv5L<A|1>enr2jvOg9Y0EH!6cxI4#Q6RG9I1q^&?^{0*lzbY{4Ci;-Aiw-gqd
zer86CBZ)j>1OTOliML>NR5-B|X(*5ri*RZ~=fp9pw}?h_*+Maz%p{A^sADJ3E<<{p
zi-QMBk8K8CL#nPi=-P_37|4^&IJKek<aKU5$y>R+LXrK36xoAA21*fiLo#?^iAH%%
zRdy7S#sb-KIZkcp>{#Z;jsn>f-044VG9<^1IAEaUFv&WiQC3lv8=pkl3gpHoaB4lc
zft_m>bJcwGgOwMvNDUtdPuY^qkB}|ut-O@!JYq=2hj8rxrK0*-h?k!d)w0`IY^wPn
z(l#LV9$;z@g&D26mC8qo_B4moxKn>_mc~vZ@&Tl=XMpn*m-;EKJOODWu$9N-)G7|<
zRnC$+PQv>Sc&oTD$a=aXXFb)B3+r)sKs&m+kwndo9^v&<L(Lkbi9jYqaB4$m!ZOio
zhGT^&d6yWshO_B}av(v%kP<N*Fi=Xs(-6;!s@%8?X)BN$7vt1=a08o0h%|)9BCLFf
zX~@j>kz8?KrQxJ-^>IT!eiYXjP(Gdt4p)x4DWU4aNCSZ+9Ajz^g@oiN8G(p2tKbMo
zKQL_V?<2tk&d9fGwzgbX)dAg!G#A+Rci_}24(N6!-?(}CMGl7WSni<4GsGy9S^1}M
zpg<jgx=h(CS8=VR%87p<?F4e-37lFFPGEzI2uj=^VdX;%O0)Wk<k+Sm_o&qJ3o|mF
zK;#1$M5#Mzc#+XrIVCI|kF*d-!{tovp%9jic0eHeYKQZ=z_7jhk!S+r(iU)B@~y1u
zgw8|S3vB&voLa>RZD;b03rvfhu(?L2cMz*gBGZ*PRG^Uw{$kL*mMSaWj5HL;iZ|la
zdawc;T105#J_#!yVrZJtnJg6D1Jkz*x%drSTR^#}ZW!wonB=M{Vd-l~`+yYu3RCMD
zmKx`A7E1NEXSE!`=r4wC{3j%az+iM1I2gIrQS~-|K$;0`>fhtkD&FR|TJpl%aH{<u
z@;37~WiyrDT7H`;&CxXd(kuWTN#p~_hT(*Ct)b$3mLkmr5^@n!>*;$|HqQApyWtQS
zEm-*|Pkfv&3ePoc_^n7pfu5)n?1>;1R(P-(X*G}sZ8)`xV>+A3H;yeUUBM$?gOKEe
zoZ9L^nLWfxlVrJwLkXIooM%QEw^}Q_xg2RakT)5e+R%A(yen@8M>FwaGMn+B&5ee%
z`6Lb`C~bPoXya08g*Bf*S`K8*$8c&xXUz$&tjP|L;B-aIhdB2e66dEloS?+%Gb4_B
zy%p~K2x&Z!JNMw!hR&ViT)9)&S0D>yJp9e8hO~JJ2N9Gu-Db2=s;n^Pd8FMy#{3JX
zHgv`u<H{J41jnr?2T>XTt^So+QnQlC2QUCS*NiPvbrq%@hqM{Ulw)yfLublzSEdxR
z*>u6fwOnXOmh*84L0wCi8Cje)R`}A3G#kj59XPe2^W`X4z7(unEFUXo<#o7M88YQ<
zIBcLy*=oiVQCEc{Z$TOi<j9+FYD4FUyHCkmBiSpgDA{kJ7$q1=0q9GHO!*=XAt+M}
zeTuWj3ST~tG#kj5&*Ib`qI@ajtavh(Cf{Jh^3I97hYk7iI~+m{;)}D!3SWMMG#kj5
zU*Xgq@_fnWa*i)Traxeo;7%d(0VKGF{zdSG*I41pM5Nh3z8sEI8#-T(b<J7IUzbEP
zu@QM+g$;&mIR%Fg)Vpjq%Uk67DqLBIG#bd2lW=N7=Zbso;<x5^(2y;096}9Zi&SIv
z<@OtpMg!S$DNb$ZY;m8}wKDN+!ivVL-a9*%KwM+UnS(f-pqw#$i{w&ol{xQ6nhs>n
zdvR(*XO8<gI?5q*u;n|3Z21-r8z@`gUlq8<U%Rd<L%xAD7s!yW;narC5cl_tHfqsw
zG<Q(t?}k+Q8xABWRSf6-ot0KO^B1JuK+gOLr#5uXta9yXY}1!qJ{z|R<kQJmGGB18
zX6diZhDVEtd;r5E!?)9FwN>8CM_LZ#%^aNC(0OyBD{mb0SA2%oL7lCJ)Y*(f3hHnS
z-%q>NTxCxi(s&?y&c>+?ojvYZa4eCCiWv$vhv(o<(U3cr<8Xp<$1n?askh3U4AOKU
zb9Up@{-0nDo7-|Q=aYua`2-H9P?+N}r}r_W=|JXu1gAE1=B#jySR_#_jOMLqG7(La
zWt<My{M3*&Kf+-IWsTu1u%phZuek?lHjpuQ;naFC2Kym__z;1wJIBh0_}SAkH_DRF
zp!&&T3+Zj>q~DmO;}eK{0O`1TLQUnfipp@5o$nCwtEkNbv7^;^q_seLzRKu%D14T%
z#s`jw%bc0RYppS4Lj;L0@FRr_jowDsw0En%>cCbbYXD@(N}O87fgQ)>8~4G2>oU+X
zkoFXbgFBZI6HUHWxEKc!G~$_k;NqyV%9jg~Rs;ERK2B}ueDQO`5$BYegE;RqB+k2W
zSV4(%eqEf7v+gQ`u0om*WYF7iYCRZ)O?gDh#A6~>9-K1OeZ%{i?Oi?nyQ2N%*4ftI
zE+t9dH)QCYxVC^Y6tS_ETsI{zx&vt-kb}45)G8dTvB_)QIncy8(7hwt*VDeKqpg!4
z7v1O1il+=&@edq2P*!w+{Z`G|Dm-}tX)=%}kK@#O@C55?gs<XfEMet^ui^)WVdX{n
z{J1x)yeR+oTF1@N^2xz{uWRhVeJ218?u!q{ibLrU$p@eLAS%6d?=7^hwa)I_Kng;P
zXl*iz1G`RP@{Jh`vUBC8wiU_tTLYs-$I)^Z2ZC6Qv->VIMDKhpdC?VmIr?7P2J0M<
z#Xz3bl}(JMttK+Qrib7ZByPSr7^DHG%e_$xE<x7494@azs2VrmoU>_LTmRP1u66FQ
z_LDe-?M^Aqnfn9|C!qHzfws-Cy;(%9p%Py|hBOlxUq6CVtHjq2X~~QD##yEx-*~6S
z%8T;3OUB9zr?SqY{vSGR-1u?yze~qbJ14T+h5U9AyV*e6f%fpm2HI7?KszE&2>&ms
zD`FbQy{FKs){3)#^#Ii$OX0fy%;XP&I6LjPK_FHm&Q2zUnq(<QXvvE%aTt^Lj1R%Y
z*+?)*1LEvzEw}_(S8}+*kF&FA6zU=aTeo?R^KkeoQMMb063};)K&N@qC_9p7LAJVj
zN{HQ#v=JC$x8c+(A+}vhUW7PKF#Qn6`!iNvl+V2}R$lm%KWaq-JF@O-Y-C*xh^z_A
z$`ve2J{IwM(#XcKl~%S^Y`sp4O~PGW%j6G%*m`>)h}DR#w;7`MRV{hZ7rw;NZ|oTM
z_rV|yh^@cXf=iI~*Bq|!W9yvF?d?5%?d{v$qwDzJp?bd(U0=oF<Q`oW?@<En%3pS8
z5w(g+gnbEVCosZ3k5j8e*nersiwMVQrXS&Wug1!Y^0`mO$_uY@jduOXByoRZBkW2*
zgdMU*Mq<)5TqmxBxpxp+(^}!RgA`;k_}aqc4}tKyI}pTbgx6t1^eioT(GdnX`d;Bx
z<E<i_A)d5}Y&TMH%)x&k7=R}?$L+!t@B2P2bVM|Lx0XB@-C!mL{yP|$0m;i3wBQnO
z{}+>Qj4SXz0C8eEJ2d3j%i<R}T$PmNejGYLk6bzl@`{_qhlpJ%rJMW|X&Ep{`4LX7
zlBC?j<m)=E=HKc|w$d=_wrW%gt>Ygu%w2J~K#2f4j++y~yPgU+UP2lR<i_*7ZNnd@
zw0dIUwno;mT5q)>6IK#=MD7G-!Ub@bfU3Q^Uhg<$1%T{07N=J697k)(i{S-lr~dGw
zhUe1As{9?-%&qNRU2X0o?F7LBG0$dWDc%;xa9Bb)3OaJk_EvJ`6c=?F(n25!FUF}=
zNXSWmC6OQ@&xWw_&g2s-FUscwGpxKQe-kcp*!a=%$r$^bLvM^dyl-Sc9%FwI*R|Zc
z2CZqWG4|)Rs3;Qa&ocSOB-Xymuznf{Vl~FtKQct`9xZv%5$@vXH)M?cR4@ShjIsZr
zg^u9n6I$|Mbb}dV&;LCtr~-R_2{6W<LrO-ZjJ8)5aL;7&8m>SGNdK&5$u{ayYYw+g
z#bIsNSR^p(afm_N3N+keF(cw#Pup1WQRW(?vA|p=;@|3?W9+(Ts;!<_0ESyJLnd5?
z1lRS8OeJ2IK*OyXOmM5cI^4P#SpgtBF2t#|Nh@6N`C9TK)8y>b&oo!F@*79uncH@C
zZS)v!eU#W@GTizw4oj%F0u8rpj>?r&T+|rSLLdpR#;H|E$Vq@Dksu)tj97Wm33!%=
zl^5mn933k!%KuNTJjxz!O=#?J>)6__G^H8W$8ep>y;IOS)*5I%szpMPT|La?8<bt~
zkyhX6zzF*sL;eh7!ow&(1cZ!?v<jj;#Wjp4g_<O%ulmE6`ZJjvSas#tfw7gTZG3I{
zR&pEOkh+B2Qjd{;9sVKx$6o%A{ru*(8a>rRB^d3WBx0A2t?VwH92+fWW4UB2XJ_kZ
zF_|v3#;bOdt!%f^IuI*Zt;tNn+B^2nvF_4o2vPeJh$`t*k<ugSD&Y+X;quCrD)Cbg
z;%qujQ<lY+Dp`!#N%EOdK9e4)633XGPRB-KRkG_5Rn843t0XZDxqOe6sS;g>D0E+S
znHX~Hw{pY#@|9Ap)H9shsg<N-rcR)j6X_*FFDKE<T6$SeFQ?GUY4mbBy_`ueXOoMS
zdX^~JK0fsvxpk+WC)f9+USQW3+4UuI-99e$GJAN1TrVYw>udDKsly9OZ)hFPj##bv
z(M-n5x9-X2cei$SZAoYMly8Y_ywF<6j*=s?S`iqrz8A?CietPhFC(VnU5^OA-1X>(
z-<P=Ss`n(1*Y}L)lj<VUOt^cGiSH*B_ui|8v51a4wB$vK6A;F(i&gdp1F+9T+Nc&f
z0^PipJQ!V_`5%^{%<hTBcgKc2J{SF1FgVX?&hb)PNc`p>sVa2-h!$W1@DDM0jbyLx
zvnqc3b=-ji*hWcApWTi_3F>LqLf%pHO&Kbm=O`_B8`3u5Q1q)fwYFKJCw+;@H*P)x
zw^hdz#5zRiPzdoj4jm{VK)c-Gcj<TxX)TZ<kK)u~DWc<S{U_V!*PLfx{D>JZ77%$v
z1_tE?Xy(eba*Df{i?k5Pg;`APp%4d}-EfGElC!Qo&J^8f*yiUT5p|tWS0-jkppzmY
zbyX;ECemOa1vcQ+D(>JECf~S-vC<VhvSgcV-W$slJ=l^ZR+>bPG!7-GhXCzr2(GX~
znIzI`AY}$|YD1@t`>`m4qnUUynay}G=6XZMT!#Y*${5gzZmx`RsjotqYmr6+33Clj
zZRmt?Kb;^uK%&tVF(1y{Ysi_qaX3La13F5_l{4;DR%r7*q}@Q;d<UmCblSM?wi_wz
zE0B|yJiN@ahKzX{2N9Gppv?ze8KcxzA<L6UlYwOUJ5FurWI4tazQHV+OpbC8rT2)<
ze=tj6nuvS=Lmbe7ZnbNwkmN|Dxj>RE#i<RQB<@Fx7Rm1a1rLYPW5|;(970fsvJEn4
z=E@UiWfiKNi?kR>m903np;N{EbjX60<7aS5&g6h0N%rHgfs$mSHc3P^6?*JJ+6tsc
z5vMkEdboQM$Dw5sNj_&tlACY{K}n+TNt~5csB$CHVjxvMiBo%sQpIsDn?#id45{)9
z974fU#aUT}D)%ET22$mxIJJj7RrrBqSB=qyPxiW{N6qrs@kBm=JXYVU2&#CTzV<4v
zO-n$kyo6I5I#rH!&+e>|Tsl^?qM6u;CHa&!hD3=3LzFGrIZFBXWnNXqqpU{S3#7?P
zoZ8T7;-05OqwHkfWF}dRMjc<WUuH;@i*X1+gO5H@q{^yaQe23%7f6)zacV;+iu)XG
z?X%I|XGoWK<8Xq~MgL8ZOO;j9T!pk6NSe3d)P_zP_d&AzL1?!b66LEnY@kGWy><Yy
ztErOXOGrzB<oF^^ZRq51e@*B-`O6V~9y4Ugqd1VDEYY99ch*;>%fm>MfpqyDPHpIP
zS>>A0vOFc1&&I7nA)1RN^92WC=Kj%a05prp2QUEAf3K|8Sf$K#q|rdiOu?xQoiZo7
zQpPba#pi1stU1S!HD}_Gg1Q;~H_NV-R*AC#X*ZBKr{L6vP8|1SH<n05#q>igPd+tr
z&?aq2n<Ne=C~fqUU6(4Wq!~n74J1t*r}qB@Y1o{VgEZF}lIB_*PK_pw$28qFNUMRQ
zIfzpmI%!t8hA5I07Dn?{G?|E|$vR61VeU30%=d5@K?$Qjr|YP$>S4Zvv=~U1Z{gH>
zkOliCp!kxIF9*lUgTEfB|EB@V+}LU#_ASjOzh+3$mvJovrRX`jgZXkDZ3>88cwRsn
z3S{Q9OzolYp}-oSCU@J4#^I}0{mE?jwSve8F#PH>B!XMbRkziQG#*HfCY)NuZ5_$v
z8}|u;>pD%lN0m>xchF`hG127vfj%5WQ2(L*wTYv)Dph)r1_P<mg;N_kRU&@g#yM5y
z;LO_%IrCN=R#49Ln7SHgwN>&QKw1tY&wiX*5AtBs8<8IIAc&O*r$=@Fwz|e)yx%nB
z=htz~0p;h(Abu*x3@e$?tw{5LB>XZ?twO>^e_`!=QsiF^Dez|;7*GnVgHpivh{!)8
zjRO+k5uDo42~guWvxSeDWj6DOd;keC$O`~YF`JDv4@iO;IJKdZz;B!VO@<73Jq`<~
z698?q;W#}&2?A#!Z38y{={U9FH^0xepE<+kAHiW!*XDOrOKJWT(l%i858>2?-u!br
zy8GLEyF0tqx*xRhaYG7x6o&_t0;fRI0GFyMH25&mJ|GRoaB4%R!K}{Cwyri04*bxN
z13$ol0p$QF6?E21A;9;M#sLX%Cr<4lLV$JdPT)C10z88Qqy7Y_s+U55r;x@03Gfe`
z+RzEGnDrC^RCLRZ-e~`h9i7t4lI4Fk8xkBt<O3KIoCygBo)uImaTL-@ASIUJ)P_!p
z89m!N1R*3IoM*^`ZX6U)kFW;91G!d8<8Mcr25kIoIJKcS{><|(Xz%t&2;OYi`ftQx
z0d0NIu+UL0rTO<FZ38y{C{AtY&A&*-$vqnUmmv>6g98Q1gEJsrz^i@=AwG>X5=e+o
z;napsh?%=O)<jQs&kuiX$b(<vuz>Oa<PRLxQkwr~NZWwTe;-b5=*>UtqV}F`yGTYz
zHY@FH>u;B`fg}E6mK7dG<O9eGK?6i*y%YkxhHEM4N4hWL)P_!g**(4OeeL}f0`y4?
zSZm0D)?gU09uf(ZiYXL05osOJ2du)W^`HRurL%Zx&A0l-%8QrQeCK+syeOZaD}$96
z<^PVIP7s5hj$-3S%O_{CUN(Uo1%W=PTW6gj+b@1~?Zbg0<r6`qo^=4%v)p4Nt!%AR
zg7#~%QEVK)hsk?Fy0V^&xO|Uey*?1cY8=OUogsSHYRQYfa1BS_>!c2C-p=Y`$1pXK
z#ZLUdf$AMQz8wtGfCEFnsRfrH>(@D4=>A{#g*dg|zdVYA*!FSapo52TKmpxH3A8%7
z?9U>Q7csv>8VZcDzrm?hV(hQ9<VB3*MAMIPykBGGMfuz-W95ZkDQefB#Mtv18)H`j
zVk|p0OA4?pkD~@J?mdLov{ryUffR%o<=9jdKI?cUe+UHF&Oi{W5nwwE(c7XWFFL{|
zj=oob)d&nMiE%xqDH{yHKF2hrwa^jVOlrx4(G4bj`L|$T2Ba?^)PhUEU1IW$aRvS-
zuuM#6hlZ@Y<Wawh!&S*yzJx;u=#fh&K_&xOe2CbUvMm#zP<;_;88AutJWj2Wq<ogi
zdk(O4dXImrFYTnKPp)<Q0zzWKqr@<iZ(|<D0Rv?M=u{@o1fPm3?D!qhS|B@q!&^7}
zvC499Fg(7En*LX__MbxJ5h)Y2{Xv@y)N227mDO)tCL*l{^5AfsTE$C@(~=iM3C=_P
zp+v*I)G^+CEq^3iJ)6-^@G)^#8*%tT$qL$jrxsbu^;LY<IY^^{Ts@Pi^-NGaLzPAv
z?H{RD*W^ICA3HwL%NnvFjRe;95HV#~Spt0x3}i#O=BfuvB8>;KVi2cR@nCT+dEvo0
zD+BXjbtf-hmFuwt>oWul6L<A#9GFmdwE>=GI~L@rH2f6OL?8`sz^PSeSR;_tod&R_
znZGpT!_RQ2K=}YVxxsc=bPc*-yI6CiM6LUfwgQRqW1QO1iQ&6y|FFNA`G?nVU_kkR
z^$%4`5EUA{j5HBQgBNgWL#F}v4+1FBajMhXRzp6VNaO<;i(&nPTQwD8tU}rfB*qGy
zS`T7i^AC}~@M%A+ymOWnD=*6DZy>PpqWtB!2@)GWT0VK1w6w7=lU4v;Cgp~c(sb8;
zTo-e9nY5m@UM20(BBGc-E;4z~;aTI>YFrlxVl`eRU2BNmHCpnb8yw{5d%a51nXYB;
zl6=={d@~rN0WUkgt_7DM>#ZEF@aI2$zK3}jhpz3}#4?TF;V=SvjuL3A1meXaHNJ`Y
z4bnVdJpC0;trAavt|c$x8E2M$JmVc1D=*6DZWt>s9LeR{^(XQ4tj5OE;{oyX@;z2Y
z3aBSMff~cOcMDq4S^@QVQV=4!wV5b<)N&?&2n5uQKoF}DP`4POw@FK0^n=%P^cx#c
z)4?DO2&hRdxCB`TIb2==wXWHp|LnZqM}{B7;j4ty5)LJx?<k!Lnct+2EFynlrp|7@
z2WcQMsJ;`YRtc(aXY!4ELj(HK@SDUalgaL{<3NFu0d(Sy3mL==jSat8Vsb0eQXnh7
zj8p5u3T&z(QXTGdu<}k9kChkY^8k&N7v&$Lg&!@Sq*~h>n`#{gNVWFKGZ&No7uCPq
z`v<LNtwd`ADaa(z8qeepfkbOXAc)mSw3-dkYtoVzz2HcWeq$4@j$n`mBwAav;1Y?}
zCJvWZqSd$<>-X_?8V9eEXC-kU0bK`d>SA9qQtSKeL8N)WcpAs4RpRLzwB$uR<J8iR
zXS^q4<wg124`bzpCwbK*o;Ei&o~{DK)BRR%c%Qrs`fH?#5jm@S$DmcM6;r>W#U<gU
z{+r1k0x|Vhfgn~RrvBUzz5mgY7d_!89DT2ts<E2H`@@nKf&tj)!;)vU&=K5xT1y^`
zZZPj!UibGPdVUFzu`D4ao21tZnS5hhf&bKy+lM9R;BZw^l{0bZ06lUEwCflT5T0L_
zY(Sa@OjAz5sa4XHbxhtfQPt@_{>{GhA{Iz%ep!+xh9QEhLWLv_7$_CmA+v)v6?|%H
zn=59829f3h=@IA68~y-gxi=W)RmzU_+1DBp;2I>NuIsav?5xyP7Xp;)tWFvqM4AmG
z!uxS*6+iJ_EqO7V;6&6PPSo&I8cCGDoATWb>plX8iEsKb4os+TIvqnz$8KMWL%J7f
zBanx8<J2lV<UGLgNbr#7KUjHZx`~w+<@2E!R$i3<1+8?MmQSXsk8kX-dXY4&PQ;3_
z=x8ySE=Vf?mi`0Pm)!dTtyrx=^deG_N#;DC$u}l*_8#DE2n4Q@&j9ZfL)_MB$qUAx
z#Bs0p@OjFX(M-IU%w}XdUJ?wxQ=11X#VM|Kmli@|2)$EF9*nTg7@+oY#kT|lG++q*
zCM~!G-20e(V;q5hk}{`Y6{GoBCXpS97Az~_$e=%kgV^S!ShIKo4k(~&F0GW)6cHC9
zj_Rn%@%2cHfJw=9IJHVraxIhh97NYWIh~fy#u8BqoTJ6>BZe3yrZ^y=Eq<(Li&rY5
zw)MS8TYznSH%@K%Z9Nbx#)s3%{g$JxUp8#(7jQtBw{@i=YFj^xv<2AKPvg{v-q!PC
znZh0`AGP)t^D&m3INE&G|1;~~D~LQI2Zpx!37-91sgT<C%}ASoZQq1b8+zN%&*igm
zt5Aq0M`A-(bTAzoay0%<!^ZE!Ap&ju)t-&7)<|vr9;8vg*6+fp#kRgqO6;G?y>4e`
zXIDpN(8^nxxYcVFM$^R$lEvYUOm4I&4Vd3<NQ1ZHpn=lB@g+)CR_9wY8+kDiegJ76
zkP!Qs+Cw2}Kh_TqvPHDBy>B*b@6RCtb)5%SCRj?2GdO_lZP!m}^_!4J0$cq?oLa>V
ze3HpGF0pU+<wkb2=p@I3#6XjT{{b98P-oyct_w;IyMhWkeu1<S$d3DQYCYJ2oy!$-
z!Q3Zd<;A>BjbUf~gH&^`wI`NOkR8LVG$>Zhd(w=avx$5F`KmWPAr(~UIRj}Wke*YS
z+Cw2^tpdQ}61mPYZ2i-bxB?@WV?&B6=LD~&!hutfwgNe@9;a4uQEQld;{w<T5C~o3
zSc=$b632#cI6>psCeN`AqMiy@5=di#T#4b-dT<3Bjzl=){t7Ga3};w*Q9hrS$I6TH
z`HLp3yeOZ)AjHax^5<yHR?+gwtKDN8`)YTs^lFz(RuqP#u_8S{guGFU45YK<jfCS@
zX%KrSuD`qYS6b&<Z+-93>J;K-&+SaUF+rHUH1l{b`A8sum3-ezK4gg9gIe;ULp;E-
z_j>cIli${UFM0S=s7M54CS+;vI8rhq*|q(*pz13eF0ZwI4gBhR%3wB62MUe}%oR9j
z?Fx$bk<B=OpiSeL$#A$WcV`x{YdALgq4DF!kE8#kE}2L!Vj`;vX)Q3yABj_|MERvm
zzHu4h{Aez3<zjh@#)nLFBv#n%phX|C$Rzjc!65>r#TGgDgV92*tUBf0g|rt)jB{~n
zLnp>^LW^`PZbf;vz}T^e%y^5uA>$y*TMdbF0EZBiC_8nCQoHgharPta2NGuwPOS%V
zuz9b@n|a*A$~)5pth^|nr+8R-QT}aOX^x$@AJ^EteT{G4K9no272S^~^HSRWBU}%3
z??JT6wbJ%`v^Xhpp}UxT_-Xs!0s+iFZU2iQc7M{67yaQ69Q($m?KA#~N+N*Fgrx0L
zNy#Q@`y?jc7^?cegZ4<<Pr^Z~r0p#@fS^s|_}amlpSk%mk+nOM^%Iab0z>QZIJHV>
zUC!him#iOC$zKyzoV?}9i%A`~%>7bgo=N6@5e^<GGkX0KGFN8QsJNPpU4XP5$e4bd
z+Rz!(Tux*es(>W^y@s557Y-mOXZkfbQ?u$SYu<sh9>|(2acVtSgH3@&3e95}R^FL1
zVC6;mJl(^}i}Js#mGamr{6J$<`1Q3?_-Mw;`eyOJ#C1XUE<|fxD~tb`7A-{<e;<=?
zOlsse`Tk5Gfca<fPZ?tO4=s7oC7$5eH#Uo3^fW4=05TJj#m^@tn`H5Gn0#ZX>Yv3=
zViT>7sk75@z$!WXsW^C`4O3(4o#Yg5&Md-es0P>dNF#y4bq!9f5?mupzHv#s$1rY)
zm|>EvCvY%8DR5TJ#LJZemG>{skz5RECXfV|;napsf)!QCDFYCYnvWSW<Z2v7P=;Kj
z#}Ehrst@@9vJOD{ybq_=gFe^<S0u<hJYnUX*#K5vl+QCcth^}y$6DEpouHrF*aZD`
zehIoXo&GSc!?|}DTDw|_`0upnClc}BFnQ1H#xu5QFVUFVH##tqO^l{3`Sbi&13`>z
zAXpIP*_}_k$$!ZZz2|G7_h+(%Yt@xw2gX*WwvjaimyV^jlH2eG9@-uw|2q6b`j5T*
zAN%=jC%;|DZx``f54oin-=8F6myWIMF1<cBTFl0B$yQFwR<;F4t7E#f)pf~MYci9t
z_Kv-Cth;m$NQ>LlT9uNGP$h05ze*3RO6<%@=RmkhOB=|VVm?{SW<1qcsb@IQsbilZ
z&@ZQ#74))_UQQqvEA=e-i|yl6&yibq>UnZ~PwEABeUV*XBG>KXQZKWISIG5J5|&=0
zKTaK9P`apfI6Gps=0`IbE8n^&o8R5q*|jB|-BZ3LvhhM|Av>CnTdklPu||#L3&pV%
zqsszfB!1{4d^@MNzHk4RR`(L+bLXUA{;#ylC+jXxZ)|MwUw7%scq!uSByEit1h_{W
zTIX63r%#JiBI5KgdC!Pb*TrmpQ$M4D0OmjSlQ+aJt0gaztu)8Ju~R=E2?l7uelQ=>
zf=f{K-#A={?z+p{anRcNk63qk8xA0#({O%4>c~hUuFlN+4enQwMgj+JU&5(X25w(u
z@}3ED-M5B8zQKK*n1_f53N;?X!2_j6pZ}2Gof_y@xQ`-@2U6x?oZ8SSQ(1_~VFgSS
zE%=uiYvvMpM8*kaO^t<^?yM<K6M0u&og|utv>!;E={U6>#K9)QB9Z2S3@h)<8?f@C
ze4g`R<wf~jS{;v;PZIf)8=J_l^P41!xDPs|O#ThHKIq<uXr*gq@|S9{Qlv*0G5H2&
z@>AQ21iyjNqGRspErB3b<M4?$8KSpOOJ4MeD>(XIxx7aH!A47yHjz*M3J!oj6%4>W
zdz0Rvg^u9n^;+_QqwDbv;9bGs>-!DhceM}_2!C5k9*nT|1hel^>W>ElG+-?Hm=;_D
z?njw?W1Ox3GK3k?!99uSZY!U$(vHOl)1Ea;A14!e<MdGpeI>_PpU&~9drXd0RZ~qA
zk3d=qOcW2psZ|oi*Emb+7!>@uurxYAxIn)HwD-76WTV+=B3pD2<1|BJoQ%T-m=igV
ze1H<eqs}Tz)*{UYvZNKKHguNEWh^Nav*cUR#O`R~3I{tZLv{?{@PM+zaUu+c9aVKz
zSrJ8=3uMJ5IJKd(Vi{uvk0a4+#)`(XY5K`%-1!;tzZufwgE(NI^yv1V%JPaT)Oom7
zTIERzX*ZB3@4=}JohLJsnPf3q%trTH`K-fpeA|#3-^5`7rAC_;HLB~Wvf=AUQ-N%_
z6{j|IHq5B<8V)l2uOS)!ih}}52F>hK_z9H@e@0pg<ia0uYD4G30<n05tmAOqnPkax
zW+Re?L_UBK$u=zq;Zj|d8}pDB1GzC9r#5tMEUX4=AQs>4vTe;4Lvn1wVFGm+9meD+
z*H>l7>yah{*>M(5EtVZRlR^Fi*M(c?&SqpjIdk6jojrseR)QWIBu&H@4C#@>fdZw+
zb}f2%)Yonau_ST?X*7@^DW>*Nn7(RqN20v`z>!mY+>iqwMZ)R2)2TY6RpT|f$JoQa
z=Bl&!Fw%G+7shaE6=!iZlW*Y43lH4NUe-aG9}@FSR$lx72M^S3)c6h>N11ZfRnB}L
zX+4lLcjDA~a0dGZTD-yL{tGKF-gY#4Vs^=3dFRed^zz+T<XD7@(xd965O?_VW?UUd
z<O4{L^`=(5tE@!1S8#MH0a^MYPOZYy8Z4=Q8nmF|+B~zMlME@+5)4JoHOqpe>MHa&
z0ckUk9>+7ahe8r`v}?P2<Uki0GT;Ivn!p^W#%3_J0$Mc#$`x0`SwGTtAQR5RsZ|_B
zH<NE%3M7(d3AS<?;K+d9MXWQ)fZl;a2buvHCe5xjR~d69(s&?a-i%Z0!5C};BN8C)
zyRh=%B|wX-vH|b>=R1Z>{T8lOpiJ%33T~b?R^r<?kY)op`Zb(dg`*8me`Zv|pfdUM
zcSC0U4TlAk8IB`PDot+h&tEI(v1%%G_zTijARYe1)E)}y&LXLu%hBEuwHCc#7PaOR
z`2fa3hKVk1@p6ULU^NG6GqB}n;?ydRU>cKeTza!u=7&dGvzb_BlGe20P=UGx!}(xF
zkaCSxhMbKw8px0{aB4jmf=x$6I>UVvRzAFRW>&@1lQ*5tS)2og#N3Z-3@9--n`SYr
zqDr*dgR~aN$RbXy!bqOHU>PMC$rt!y<pm@8_W)RVQ9l2g11m4ezd&n&GcBKdqjN%I
zztNdq{f&+^A$kw4E4g<CTANxQ-`u4|H?bc4yG-6Qx$=yiI&+BZ!w|2HTmBLVVl_5y
z`I8}ff2e_8y*F-Q{H|==a*XS49`WqRNH$ZojmdJ4Kg$$~u}ra0^_M4k{pA%_z7V4)
zcSVcV2sxCaXjMra=PD&mO|1IU6I}na5|jAoqw4S1xc)u)67z}}S>~A$W~Gw@Rbox9
zVr07@D_!+h?skumbG4J?Ph-`8>TX9aTSzicl9{5F&t(ZN1frTHP^-z`B{M^nKNXwA
zOnZ^+3NwRVX3@(WdYMNr3+QDLy)2=ZW%P0+xft#qbFT63G0xp$Ue)STB7pEdr5`|k
zZ&*I}i2CJ!Tf2M`z~0c<048Tv?)^%NYaN7fh_Q=%T%(n&71y?qf)Jx%JB<=?Z4;CC
zjBEAZ9>(jV(&0c5tMO5(Wr*H@mb^&fq8$A?em~^>QR(}F0odoG(syg2Be;2$mOL0;
z?bOWolgBRv12f>0$N$oTOThgZCa>WNbb$0<E;zG~o^%%@UoiC*_oXe#wcd{d*!Fm0
z6P};qU;?`4(mF_th&bw{47PuSG!8h}zDKudH9l+iZ`lPh0+GIs?ya5eQUkwa*uc*t
zF?4<Zt;F7v<Cq-xB*4+Y<?1Qz{4b=9z;^y8POWX0@R<LP$$JiF>&}L`4mR{(utTE4
z%9l`Sl+xFaBl3vs2-??A)#&RHS~rCU$07{`^5AHk+Cz>9T|NC056(B_K`#!EKzTsx
zrtn|~(m)^&I&o@4=fQl3Kj`Uf>*$u4@HRsxyak5{lnEPjJOZzs!i6^>O$2gbA5LxP
zTv+JfLT6`NS6g(g#D*^#vf=YMOrUIVoW=o570UHf`0!bzkw8BDCr+&gAF%U3V*ZNf
zq*!^8tK8+EtJHr`GJR9kOsYh~#|>%t7_KRxG+YC6N>Z&9@;!<)4M@I+nOcppX5CZJ
zdDZog6!sO!%)QjOb6+;gL1z*90CLdN(2ZNJoYKbAkro2mcnVIf;$S8+`34R~Bqrpu
z&Lb1fA@(5VbnW_xiH9?BctCqOY%t<dH-!fqkOl&Ia0*Us=sZ{;VKrhE3b7%_27PHm
zDkO1`K&b$mqHwRC!iGVljX*ZUacV<ngB%~SgM$UDD8+~C47qSE4iG38G~$C@IfV(=
zAT0zk;UG?J=uB8Bb)15gix!m0>AMZd@I4$RP%^*<Bp&rs`0yR1kw89t3#T@8KFoIF
zCAo#C4VmyH4h<+1PS!{pt(-bx^>?IsKo0y3r#5sB%yDu+o}XxX#Vln!lE?=zb~r_g
z1Z!0mEJfM}WWgew+R#}rUt)p1Zb|Y6U4}$B7l#Pc8^ES5U2CUAg{?>vfn3;(QyV%L
z99hEdWR9G{wpUIW_ZzZd4-OM38+5XSO8pc*6p=;(`EWT-ZRmVhBK4jjt4Ox*&5q=p
zJNDjWNQoP9us|tsCO%S#Dys3~lSnIpocIJzZRnghQsP7`m$Nd7s6El(_)6{<hP=2R
z2Mv@Lj!hOl-f;L-Q{~1_k(L6v@gtnt(7B=di{uEMe{%3*{Htb}<*PVYpuEuZ7rcrp
zCtgBY3FO4{IJKd3qDksVWNY4`d@Pa7M&rY=;()UGd&H0$tBHI7gN{x3P@>dSWyngT
zu|S3#hf^CmL*!Y?cs94s;V&*WB*ld|K%o9YV<bu|r}E%@q<ugh^y1VWVmwGB<IX(u
z-G)543I|AlJfM|RdGI!*eLx<(1*g`72iV09VzC5&UyhXrFP5nLTlJZ7vQ{E`MKV5`
ziAAqCNtz3}(~yF9;MxL8LD&LrN4@NZ7W=v0jx-L)zuRzX75>#&EE911uaVmSABOG!
z1P+S;+kcJfe;!8~2W<bxaB4$u|LLrskYV9S%&}x|*=uG4m&HUrfb0~uR!XXs()Sl2
zO#`<5T%6j_+kS2tCGzpH8w0i(GN2uY2ekjg*3r1sP2s^tq=7&loP$#vIu9l@tjJWS
z<o!ntTR)FO0owX4pwVfiMoOD!k!AtgJdINudYjL#>K|nPgJec(`%f9R{S7!YpluI(
zC$7{?;lTAs^MD+<4yQJB4$P=R%8~ltXGnk_<Dh^NU@g4!^O<f^Ev5DEMcM{z{kw5$
zLvQ_gQtL-YMvAc^$qBq{$blDdfIv9_TQBEUIfV(&A}s_m;c1-O(3v3l0eJ|xYTS6E
zrPM2kd;o(0ydQ8G0X8FT1Gat>PHpI|FZqErE`DIAAqV<!fCT3U%9T_6Ko8PFAQQT9
zYC~tj{Dd_aBfCXOJ5fm?;q8W0cq<MOC>3C<qg-pJaNz*bL?9RT<J5-E1wJoa-W;tU
zvEUX%7Tk=(0?Go!ys)U3(*B=A8V79un{aA9+8_H~LA;mXd&Xep#d`_90TNbTl+U-X
z!pe*CU$3=c1}&d#Le|^ZO~~d}@1hxDyYxx@=OtYKapym+Q>{J6p4TFs*l6fqOujK6
zfAYOF$-#s9<iKdr@fp|B@u)Nuh**t1#}<)75xJ|~Wd(Zkwd4im=P>ztepbmao<cXa
zru#a|PW`N4AfDPxH!J4FDZch}ErbNZr)tTA5!U%Kok8ro@7NoH0UEIH*ri%<3Ais}
z@{Mr^{+m-xOlOCNti0q^-;Tr8CZqW3_pLZ|pgv})JY~hALBuv+=@18yHUN{4{W!I@
zS)!lpVe<9#ME=b?FFtI=cSnb_BiSJ<V<n5uMN+pAgG|2gz8MDyw5^YGYip%0>Wt*)
zkmdlJ{U)4Rk7mc_Y$E65{sJp6XvYH}R$i3P(-*9~DF2^Y>64vvzIr&>&v*mLKUZBj
zc3^B}YU{*t<Ayh+wvpSVW1d|}JLD{qbI!`1&ytIL3D<4hX-_LwGx5}7o`^L6V)CBw
z)qk^FkECenVJJTX1gzwn6fGiU8l|0D@`Ckqn0);i>YEgu6%5dTr08@lxCB*C<#2iR
z^tw{jKWZ)<&5=Fot!R9-P|S|dNxZ@2kYjc68*n%)Nz$b_w4fcr@ls!GSs!t$lNvxT
zLYf5(pcmlODgm^g$$NICx-aHimd?f!bSV#6(38ts#XNbhA7vS*g9h&>hMD9~@5KQF
zrGewMP8AJ2E2eVcT}a!2TzCggZRlJ$HkOFxiZ&S-2cqOF4W1=Li~DkxgA(5`q{P>7
z_&_P)nA)$RM6LR%%=ik@L?AQ%8>iNT8Q3&cq`5o-Vdb3xA1g1)=gAsYUX=e&tz^+o
zvo|(&YF8fMxu2IJrQAmxj`Fnb-GWxKR?2-CDaa&ke2uZ#Gv)RSV%j4XuXB@@2ZC6Q
zbCZrSMDHjj?~Pu)&rM>q=jSGMmwH@J2XZH8D_`iF&*rRXAs$Ox<a_39X1BF3nu`_V
z!<7SkVyS1y*p-Y_w~?B~QqPiy?c-C=kz04_d2)SE>IHUvkzHRR*X`p{FSCbN$n{d<
zeP5$LP90uQiWwit<53sqNj+z1wW9E3ycP95*)qfOc^}j-|7ER6WP8%q#(L6?(u9S3
z14sPHDmmIB!5)#)x;t8mK`lpnxmF{K0b_>Ad%B9c4j8=;nYktqxJo{U%p5et?fqKv
zBF%U&$GzT1_;}15-5LzOzB5N()<Q@y^A;_6Fv8lIg73`HZ-N0DFmv=PEw}{SKWFk9
zjz9-U|EZ%zo3fdNl_^-g*<vieuYF{|N+ig6ev*fM8Hcnock}`dFQA7mT@1;|YebTW
zSN-iuh|fl!MOFehhJ0F=3w57EI>rNvNT#?uN=|x8W+hUrn1Bkq3Mrb2JYv`aC506r
zDXNfA=+T6%4v-#4;?&w^3D3Ng$$O6B>P{lM4~m;ViR6+l9O@(1AyPMmLp?Zjpd3mB
z$)O4m3Wd6mwE<G-T%6j_DRhDtg+^j~qXp~oXxhp|<1zAfJs~-yw;Gb^01hW8nKD6=
z$psRHR{N3l0@7*^PHpJ4@(gc-+5DbZKH-=ZzuAyYpTofeWm7yzHrYt1^XE4qD+8p^
zjX1R)6vED$h*=+=dtv28aD9^JZgm`3`X}p0<}LEYZrrlJ+24~a4%3&tEZ#ck`)5P?
z{t?$SQ2IveL0=yfRO&v0tO$_058>1*)aBH{QcqBq4<WJgg1UT81S>Df=W|(Dc~Smu
z+-S<i&z`Vc)7S}1d4-iY^Ipn6)*OLKW!*artyryD$OtJ2F>bPv6O#|CnY_m2gT`3F
zcWHl5Ac)mi+TUe}-nm-xg7;fF`gIH$eoOlc!9eUg7n##SNbqw+OCF4{&J<ehrTy0i
z12kYR@)|9;1l$Lid}G{!e|F=xwEyckTy1A3miFI@LkH-UOGnFTj0g#y>-)cqGz6H2
z+=5f9q#-vm`NmBT&1YMW?<)>xGf^8gXW~jMGMN(kBMuR0i?8-<akWM+Nhs1NVCz4G
zQyY5gFN!7$WVJzjH`y&aN|q`n)3JfHb2-|)iDm?tP2>^z5tINCPXf4BO69-|q+LJ`
zOvR}Uodb)j;3{~M;Pr+iI17gglmx915>(YnWx?r4!+<O}6{j|I793gS2<$izjgd90
zR;=LM4td0o3Mm{kP%5nVq=HY?R6YzLtpoBQfm0hg9~KbM>>cyH;=)G_dGKKzBv2ll
z=*I(RmDKhhLs|uF|EqCoLvR1Nr2X@;eV*~)2ZoLReH<Rp#$V;v_)>k;mcJ8e60qg(
zz^M(r<>!%>zapI;@of5M44eKb93ar9Kf$l*<qD~7{|}^1z_xz^r#AGqUqae`FdZul
zlh|L(crxIaNoFI0qlkO}BZ8Cs7~obdl?Tg^mH~ON7^gOL9?Z9Su-lUY-G&s{jza|M
z1iWVj)EcR+zYS>=u=U$<YPDMbL(|9?^7KE~Z`?b``hVAd@SgyrNkB25_9Vd@4N0&U
z2Mm-1zDa;oAC&^5NRxmR$m7&%QNXo<HCvtr0`Y;LCjmZVNPth{V1W|AHxRg0OXb0*
zkd^^?a05=Q2M@4sdBs~|{&E2;?|heol^5mncb!;yQT|G;30C_p_58-ZrJiYjR%1`6
zNH3?K#&sNb;?w%ndO7{17U9I(sJ}D$#=MR4eK|dMGAic;B39$&^ej>+B2l%wt3YqM
zmb{?*6ejQaHYk`^rzZ!4G~m_gS}nK)+^rlguMI#vanx8y;h#87i<V!eMWa%`zYvG5
zO(^mD^n4sb&|a{tJaV#qRixT}+X(UDMK96_U~JrhQ>(<rPAz#6n>c&)V-xSJSb0%C
zcePk~;Vk~9<?8L&m}qQl+}7E(C7s>Vw9(4Mheu-h-KRITWt!SX$!<C1yOwxU`xUYD
zDE-1E)3hZY8?p9e^ShhKYS^Y7<lN_$zU*i|ZZ&mwcF7xO{s&<I;#IYK@1RK3imLyv
zMK$55KF;JD7*(ftWQzH0Vw7$FDS4#t27*~>Msu>dM8$`E+YrHTYRQYv@^#(<^<MTh
zv5Rc=lZ;CUKNbwalbT0(Sqip~YGEVT`LLFJ;MmUIM2>E5>ap^~4~)bzaZ5sZ@{}O@
zerZlKJutc@S5>Nl_YtIMlPvKtE&0IlCSRhCkPRvN=*Os@u&)jV_QlO{R(OFH?vMUT
zE%XH2kJFM59Q_3ynL;to9d^@}Y??$%=U2ep!C=3Txb8~PErXS!72a>xLQWvRO-nv-
z<foF)-Qwxd0!cn34|91ia8Dytuau`?oYBHaV7yyPK6s3KtwAeKl0V1p@?$|^oEPOO
z7+<Z0k-+!^TJlbev5CJJIPh!|D=*6bs8+nU2M)2u4jhiyNH*-2v;H5~gEO=mwX*(u
zwc1(m<ZdSKne}^4PiRl`Ojvu;+I5l-dm<35N)wus372EN9ybK+u^NEYdn+%-<?>cu
zvj!;BS_jG_)UkJtRd(W<HPvV*u8CwPuG29)adA`l9$O=cu~ZM)x;k|jF#?}<3b(=0
z)pr|@YPGH?pLYlS^1r8DJ~@ZxrN-_FK528T7%N!vvA$=J1~i_opcSd*xK7hzif~*f
zGkH(PRsVgmme_ES#Fq^9icQH((^hi&Kt7h1yxB#809abuOg3+(twoYwT6{|~(?SnF
zpv5c6!v%)e_G`(D;C&u%8?T8B?bxxRZ#0+7=8IOMsoWT`4Eg_7k$fdUe&j79N8$s)
zw8!zyg|aOz<(46~v?Od>>}D#cjk8}1TtWOj99*v#3f{!mhJ892e9%dFHuou!{o}!a
zJA)2CM2QN&KWYf!hjk&;cqzkPAWoxnwdpAV_^n_7oX)(3%|Vrd-!R1QYg+QcKYWGb
z@8utKQC?8Z)24oFgzSS*lxO4~2?p<l5@B1)=Lf3DKct16===|A$p?=7(VN(z0?GXr
z-P)<rbKTqE@M%G$7^S9WezE~EqNUn>__W42VhKK56|M0~U0MU1ZHjD_`!TG%V8a-*
z-*b$o+YcxATYHN;GU<Jit6LjPvrlO*lG3uYG$l^78n*9=TJnM8JB8xAmR=-)&kqLR
zsSMz?6lE3RUM++~589z6A2`DEdSb~uNhqtGnS}M;V6dLwoMQz`4opRTR0}nMdR|N3
zi8|Kr2*1TsSggD#pN}uF@}hh`cErky^4DtleOf*l<)7&{Iq99f>pJ6w&Q%RfTSt=#
zIV<{RJ=jd^Rcp}ybuHov;@+wyFB<O4Ox`o#X^)WRt%)>scXl@Q#|pcf`mAEn`C9Dv
zfgnuY1ZUFe7BVleyM;^*uu_#E_**T^1m?fynBm8px{N)7063bH-0iEuz&Ej(phk*P
zFnh@mv*)$s1yBFQF|YRk>Bvoai;Rexc9FG8oU7$&lr3cbL_++S>2*lb6_-#tsyWU~
zXxT?h;KVJ@o(xhGl$JP(*kO{QFVm6_yd~x^Ptj`)CCOWDbzWt2FzC;1X09S%U95t5
zn-*fiU!1Ka??fCMoJDZw{t7EE%C8Y4biLY<Ya>maT`g6M2t9*EA(+;u3u;@e$|A-Z
zVI!y2jDiLuTJp|jz&4|uW3BefvFacGr){DK66dmHd7(V{d_z52l2)cxVs*V1+M>0t
zWAdJFQ#>Kl?G3X#J7ZmKO#(M5_}mc)Y$RJ^2gbCpBbQlgsqyA^Lm+R{k{7i9s+PR)
ze!L;E-A0sut#<ijXmes?yB%uQ@mW$a;}@H>g0;Hh(^?3MuJ|OA_w0)GA0Zx11}~c}
zvbK?&vSc;U#H^`${#5p83uXiXW9i7|MA-tK8p;+ZNy=PehDlmDOG{pKlj%&pj&5;f
zioGbBL$ZWT!&%x8IeG|NzDaV7r!n+XRUi7;2Zl4@Dg6Cs(aK9g=P;odJ7T<4^*5<Y
z=!y;c9nbI&=@I0;{Os|5e%r}!7xLRh>}DU@)=ZX)T!2|BV(pEQcQIC>wK79bcciYo
zCWw~bmqaS-NN7bz2f%L3{i_M3x0c8fUIuOp2iJZ<Mgdb8b~3Xd28V{BDZ24mE%^W$
z#tF7yU8HH_XfmB>Dh^vsW#A>gq&Q!oHHTt9uX%u$+d|6qc_lNX1)G39p(P(U>`T^1
zn)qmsbZO>dU7>$}Fwie*=0iR!Z;!jnuSONtzgG)B0sp(S<O7F){>hQ1^UJ}_75Q6&
zL4I!YDEU@^Pj+~MezO*60`$*m$vZ&@J6oGHBIo283szo~|97p7$sX!O8=G@Z-IC0Z
zIm3)|g7@)y@Q2o=R(koER$~jYJj&!7m|o7>GMY{|^~C5WlO)p0>ExuDbx0=_*h;gT
z2T7qVIZ>$M=_ZhpO)|*wTJnPQukv=N_Z07{&e%RHzb8qYT?a|%^F{hmT2mj(zY<Mt
ziP*>mO)EP)`r1}WG5>^M+F@mLx@=BMvSLn4ft%Ap&S)Rm(~?dWVru?%ycWKK=gYO^
z18<+Do!LDty|K)0+7`JY|D1SbXSUbTAYH*U$dYFA%~ngEluyC8%Wp|%^DZrLt`>r#
z1-5F*J6ix7Rzz6ho(?N7%70kP)7oJv-Po`+x!oRONxk~2dXR|Lq*hRRn^sc`s=S5C
zYXl|jH@36cip(52C5Xfb`c1zC+>L<%M<yWY%Punb6s0P0>63;~exeqX_5MJGF}?hO
z%ACqje{)-ZTVH#BU-c&{_Ym9c!{QIuI4ph=nRhr#Yu>@}mCEfLqd<|EUqM1lHy5hG
zc)$#d8N&-wKPD>pcQ?F7SYIc6C1(v*Uiiw7aYVhc8SQTBm2Ur?lxIAft9UlD^U|P5
z@0G6kH&#jgDWSh`ki1t?1WL$x6+%|k@txxPu{pDhI_DyZj}wS8@<Jfc3N5ZwC~wiT
z8KWcFgq5ZPv#MGhZC0zMhSzF_s8wvRSY4GCv#K2LQWb267Y>7S87mKV7`pEjr*vfO
zZ@8s+zKK8=fV-GVb|okmr+ACkvo4_TEzZ)C2P3TSlEo4R3FE<FFrL;dmMo~F#JCnn
zBJq3!llM$KgLPoSBa>8Vdvxph9h=*$JN$czaR%wqI;q1S&_YUdc#f25hsVl`^7#+}
zD=)gko3w&AEuX}(i?l{_x=w7eJ;x}o$ond3Kw}?9D^hDP_a!Z+h(7g2EqT#WpVyMN
zTdG5=rL>*lgx<C-{gNAeq#n&gYf-D29@2tIG}D7j-qZDYuQSjFYv!6J&n@3x3k0^B
zTfV()2;vL=Al7k#0DpbEXia2IQ+d&D!Yah`$(%BKarEpucnXDrrTLXl(5nkp@}Ad`
z(oH7TmuSfg_pwk*UIZH6r22v8Sx$+%4~6HfC2x@FIH#$rJkGgl?8>nN_RbsK!8H18
z^6Hot+EN)@DGk0|3vl6xwlR4PN2D8Hk6ueU5!p(>%3nd=laOr#Dr5epm3>?K<u@al
zU~pYZ_YbfQXb}VQmI4{`DG2X2M0i+BUUVsow~!a5b=zUy+K4}=uMP%#kDY-&poNm)
z@%yyo14p@eZDd^&eZkb!W#xygCc0zBR;wsE6DfoFN-(H5kylUoSjz}CqJ?giLC!>N
zsdg?Z@Bdp1QPCn_(2{qy2sRXmP{KVMR$i1JR4AFfu8HQYO>ODqP-eu+$lnqFF_`wB
z)to02S}n0knTnfxM60<(b3er78yH9yts{Y?Ej^UYCyT=)g{B>3N#hksYmd~O4xbZH
zHWsTC+>$L9tYA8hlx~uZy`t47g2FHIHmP^$n7@vMj&1u$1g&1=QJoL12nO~!&BM-8
z70jCrF>lh62V<^1ew)6IM3Q#57tI~PfIX$TJQ=RWhAmnk31_j1$v4nhOd;seywfpX
zkPHU83C*sV<e(vHaV>d4%{Oq=18@h_A@~HfQcy4_6Khp+zDEltLC$w-$p;SeOiE$m
z0!RkgoptQ2^u=HRCN6-aiWyp(;%7gvg_1z|vs&^_l(FfZNDsLS!pe*Cd2GeXi}HEC
zgq0WN^Yj!eFUsdLkXU(9{sUT*ceH#mrGI{7CyHnJ&E33KkG?}IRBO)oWi7^w9`gc|
zZ(z{$J|pC)xphdZ70hZ*7%wBmn&i`qwd4io7clv{&nq9>SG00X@L938V9-CNxj^h_
zS<`};96MVJQNigmwB!SCk%g>9{O89K!L-66-U@4_*%`X7Mv1^NE$jsLmubldj=lS8
z+IIzmw%;J<9a{JZe6Q4!cjAkU&?3U~c!8A{<v*(xuj~lFsj(6MX!1tE-ki94+c@_<
zFuz?7F3}3sium8uVt`=K*O`0+BmNY6qS98fQ_4DN7VY<e09KmHj#3&T8>6gKz1wdM
zvHP`_ykP$?IrjA)EKj5ecXdkej-OYDY)au{X%d5X#AS=}SGBMaynIPZK5%R&Z?>#l
zpJnZq&|Mx3x>Lxe30AH^OH(5AF{DV7JoP9o`M~i#VyiVW5|hB)8VtA-n}=9Y3bLEE
zkP*yn(~=J!*=_qq24rMYK_NTLic*jr(n3Zco6wRE99gm^nSA$}>Fio7@%H_}pi9>z
z)8f)2F7uM_)j~)h{4Oo|z!9F*nU0ObT4a202?pQE_9jv;=-#Y_jzITwTJnuTSI)<N
z9xS>p+1USRp(D`!iI#le=+5X$?j>75rkhUNyLRnH$wNFF48}8?M|hDfw8#N9cX(P0
zDS`BpTJnJ-JzXGustoj!`GI7=Gpc|-)dlB;q*#+2ZJw5V;5Z-AGn{mc-p&jL;EByS
zR+N%SY|uhRF!&TL`M{B#c-|gz(zqPvqQO8rsrhnVmV)giTG$9|cWKE7j_qN)l9Ka%
zTQImD-n=iVdfc~YVInYnla_qom`>c4Y*`aIS;F<x!QeWH;7Z;boUCH|DJ^USwl`?W
z2aoMVR&ID-Uc&aCps?M~%Tj#7U0T=(Y`?1|@5B~6zbxjb`RpcEUX*{U)-04gKRvav
z^V1XgK@bvQ#x1Br;zR3EYa03$t%epvd6CICa2neC#q-gDU{>?R^Xo{lCduLwEqOup
zg-qV_#k2OJ!pU36kp!c8$C=w3g8_L;^I*9&#h0C<1(2ZVnH(VR-5u&Zk3WrVAm6X<
zMK=%(c+>gm0ruv13Vu;T{4UXw7X;nK@vk?^b2~ERh=$QP{mQP%1?;PW0ec>q#;cUu
z;)VI!v@jE#eT$a76LW0*6!Dk$8LYf0zlKxN{V<6*|3WcFHj7t2t^7(btv!RacBMq6
zZ~eDcJBfDwf|h*XID4L(@IWvyPbaS<#X$`U%D>P;NuYecmOL0`?GQVOtVbw1zw&u8
z7>Fm+lTcY<3cSy0!6V>(hROTnfc4&Oh$ZOdoz13Iop)coNykwOE1PtT*VxnSJ!DU_
z^_V@)99x;qA;kx<(UMg4ndPrAT3O-MwQmu%KTM<cA0%6p97|O2P0Simml?6qVm6jb
zs#Q6|tSTRBcvYsw2C}0?wIUs675SJ;MX(WCM0nn5vGQP_s2kxY?o1Y(n?6PfZpIs|
zD1%|L28MiPPRmjp;w7X&lOf$ME&0H)oxCZ@_qCO#8r~KRx>K6t<<b;<-=c+&2sm%j
zk`El;shg99csiCOUs6j5e>xb1r!^-;c?!m#(!xk!e1n#J;22M`7jsGg-xCbLlj&9v
zbmge(A@0&bN1*#%E&0IFC1dJDGHw;vUIfM1067{)0{V$yfTm+=RxaCqz%Ex|^y6BX
z3CthUl6PW`jhG_h@~Kj+yeR)Ut*Kf*r#)JlRm{etnbDCb8E@p1q+RLBLPwDq;_c&8
zbLs7{)M5PT5`LRTZ%?Hr(c3YpS@d>Ps)gR>rq<Hi#MEMXJ0i82-X^3j=C@P$t()HF
zrdHD1?9{3Bwmh|)-e#s2(c59E*A8d5<@9Mn>Lh;KMsJf+lj&`K>S%gfkeW_!N2FHJ
z+qBd&dYhCwjsG~vZ(H~+NpBNVOXzJ%YM9<8r54iL)YPT)c0_6(y&ab7qPNMZBk65o
zs*~S__^pZF#-~#JX(zo+O0A=}DXC-l(+IsCk?Nzj$*GC-HZe7h-X^48p%t5$8c&~Q
zrrPLjY3f3Ho0~d=-e#m;Pj82%rqJ8bsUCV;keW?zb5j@4+q~2QdYhDr(A)f!MQ_tm
zJLv7O)Y<&sR?*wU)Es)7nrf%FVro6TO;2s*KW6A{aq4Ai(v;K&`ZOaIr?<mWGx?t!
zL2pf|8T7V3^*VZ+kUE^-cJW(3y-iM?LvNE(6X@-v)MfnN&ZD;(sipksEP6X4wVgjT
z)7#|K@$@z;bu7J2OdZE>1N?R_y&aZ1oj;vSZ%a}q@E_kmZ!1#0{NG;Xux{l4)=F<v
zQ)}pLc4|JoO-Lo^ZE`9`Z-=E`q5@Od!QQ0U)%?d3>1|?)o=rgn=vKqTT)O=pX>hu?
z7HKj1^%?n*E}JH*Y38_n#%O6K?Lh@APCC&5Juc(G82@RI?A0NDnoa(yn76EGG80da
zk`1|7N26WqQqtl6NxpGRO`9=poRy;8?9#DR7kP-W?nVE7%h}}Lw~>p*`WXG`t}W!J
zAJEIa^ztKm`3b$;M=$r&%g@QhV!etAT)m6@bPc_Hh+aNSFW1q_$LQtb^l}5ee3D-N
zgI+#OFE`Q4XUWB49g~>0_#$c=?R(_mk_3IAeTSCut;@+Df16(Jq?hl}%U$&H1A4ia
zUVcO`KcScV=;eNT`8mD(l3pI5m*3FKZ|UVBdU=>${y;B(Bo~YI5n9+^-%oyef?obk
zFaM8To}!nh>E&Pa@*KUqKrb)R%PaKq8oiAF0J$7aFGrAz#rhA``S}~kPhX&yTj=FW
z^zs#Yxs_gSqnB@xi^aMM6-eGoe%ehg7NbA?Y30M@rxWPqM0$zP%SrUImR{D=%PI77
z8oiuOFK5!r+2mrSDDA0|W1k^EEvJ_i^s<s(P9PVH^#S@9(_SP$&7hZA^fHHD=F!Un
zdRas-OXy`8y&Op{R*H5J(q`KZr&A!bGmwW_Q|SZcJ5@?|d{5Dx&Qo-I=@i|KIYsv|
zPSL$hEyj8JXLJ%JMduPybQGSVBUo#+G@Bj?U5FO5(Y(dZ4r3gr-2Nbup<F-6NG_fz
z%jpv({XxcZ`b2}uK}L7+MCneSXoxw;crTu42M|xRCx|E7CBzf$8{&y}67fWPi+G~F
zg+9@kbdYr$@kINNc%mIiJkcH{o@mz+PqdGTC)(M>6YX{4iFQBnMEjw5q8(B^(Vi)u
zXxF4qG=v^xeUm=Xu6mI5P5MOp>Os~?#S`tV;)!-!@kINtc%mIyJkcI4o@mz=PqdGV
zC)(M?6YcfliFSYSME!txq7FekQO_WrsEZI!)K`cn>NLa?^&aAhx)Je2-3Wc6q4yy3
zC*p}Z7V$(qjCi81Mm$lUBc7=95l_?$i6`og#1r*P;)yya@kBk9c%m*#JW<~zo~RQO
zPt=LgCz=!-WZq0XQMV?ZsDBer)X|A2>hZ)Ab$#NA`atnSeIR|J>BB+h48;?5hV+T%
z1_zl}6i?JWiYMwP#S?Xy;)!}r@kCvyc%r^kJW;1Ao~U;fPt?tdC+cs-6ZN<BiTdAz
z%<+mR>Vd@*b;aU|`egA$eKLKbdCEcNoW&FM(&CA_Yw<+=ws@irTs%=vE}p1M7f;l;
zizn*j#S``N^octCgUrp-C+hVNGB;13sM|lt{JnUhj$b^{2q2zlC=gFHCWt2*6X+Ao
ziw?5jAf9NH5KlB*RGvmlGltUH0TMM>epDFy;8<x6Pn@iT^y}0_b|gkdgycvT8dyr>
z<LovqpUoDd1@ehXHa%8aTp&wKGkYWaZwrxlr1DP_`}p5nGq#Fu!?}`8Vt1FO7m@??
z6oR$^vMbk^c;0A{qnF6@5y=F77t>vuMvlBqmrtMRE=@{g;~8>B+E{n$H4@^wOS6)N
zUaR_V-cy<rjSrJEH1k#_I<POw{+gtR^QB+QYG~dVjb*Z#%D)mQMh8lhvI8kAP7W$R
zFgASJ_Hm=7De>W0X2>d7MOObw_G9JWCs$_JNF}KVk+A37Sav$2{6xa^ApK>uba*_M
zqsmfWSTufIX{i&#T&{v*>K5`BGJ>hEke|9!U!~X56o%H0{Ei$!Y-aiB#vXdGTY-XG
z{-rHHM4nS$W1_dQD?>5$b@Kn;Q<@?`xOA*EgW<BthH~su0!xaXcWWauywR>`ioX1z
z)u01dlH1S$EV=P4yPR21V?VB3S(=cGCG%sYd1R;5(X@3&Q+i~Ow0I;&Ei29CzqVBW
zIJ}{|NO(7~K-iMD#!4r4?bzJj8Qrv_drQaG=%&uLzP|Rprj^M|6Bn-5xN59a{TOi*
z9peS09wQKxn)zS!@9Oc>xv0uMTPKr{R2`Mh4h@lWOg$}|Tdjrvv83R*9c0tuk>q|W
z-%SW#Xj)17L7~t@Dq1B-YPhuOF(S)I`cB+@eplD7rj_1*-V`g6>4<^RqU8imRvTfO
zB;_ZyZEoAx-r3Q<1zf;#Z$ly#ZBikv^yeMtcW(xl(Ojbps!S?kb!S^IVUxF*gwfI7
ze;1*OoT|Gv)=72OMoD!G0j8_%La_2<eC+0`B=skD(;vawtNH05WQ1vwmG{<7InyL7
zr@DLUVmZ?!DPP^Ubu(PIA6KX%S@o_RTRXsJ`>CdiB-Jf_J9@Ttw1JiTafPdr)UPHu
zb%T}rSxyxxs^KiGp}I>_-P*ll=Z>v}KVZxKJWE-bTm-2P30(bf<o${$E0c<#ZWC3B
zWBvUA)g={hIw_#*f@t5SZS9-S?`)55>1}TZBNp*5%&Au@_%zZ(G>bjar&vX+ROoSC
zbPIWsGmX;Se}Ms5mXY*rAPI_<W(jFDn~&yV#bFsdSyqZP)hSU#{=0~(Rx0)sx}!8d
z_6t|$3Dw9s%5><I3S7etDv~W=!Sy@Nr!1jPD&|zOhKfx0M(MV`d6EE=2J%)Sfw|iI
z?;=j6QjsUc^6}v)s$x=>;{dUuO)6xqZ2}#*fUC5kc9qybs5;p_xtQH;WoUxQI0-Im
zl}}l+N~vBa#bab)VKk8>IkC3}siFM9t7wx7*)UqLqJyzQkqlz(%7IJs>tm`~so0Y_
zNAnh6#R4wU4@z03RNRTIa^O;=gv`M~rb$wM8d=Cq@XK2xWIxepocJtoL6Wa^7Ua|_
z6}q;VrClq5>RW!KuhJ(K)Jj`2K|CV3Qc@UpRI02@ErL48x)ni{NkyDQ3%(pxhF@Q*
zXp;(AJ&>eP57<|we9Hk7RV1libNOg2L;7gFLWEE!6|*jp9VP1~qU@i+H96k9#$<*|
zD3pp~$Ad@Vu1!j-9dNPWnwQkCVZX)FxnVe~{9>w5Clzx#`@KkJW2_=g8ORh3oB>jT
z=T;?IuT=0lR&XMp&5_lbb!n+;g;G&x6k=Cc(b&*X-WrM($#f_EOK_2`KICvyEERng
z?Pz3DE}FHcc4*<T1i0`uK83rhmWn@}{>>nnMH-HhnI4kU(yw)({oAi~UGz!?Zz3Tj
z6DMHP*OY`-Bnuvi?Tx0b%n%|6KGmo8eg#OToRT0fO0zjDT8NWFHZnut`uk<W4xLhg
z>se{JBx!eoH@G;TW{F4|rNT~T|CA~2MwaH6hsY|W;#e>xJ1Lz8u2!vJN|i}P5XTTp
zko6nU^(2UZi}7>e)e8SVb8i}CSCXBF^=Qc9a2YN$oPF(?p5FTomTok5piuP+D4@2k
zDgce1>ACZ&-mR+l@V0&n&=`xBsKs*0kZf9@O`Eo9)0QmRQcyt!6;x0`nF=bXpn?i2
zsGx#2ZCaG6ecvf=KACy*-uyC8-kbN`ho0e&Y7XAb?|gakWai0}Co@6Rt20|>(SpJT
z1(ByxyQPj8k7*4PT)`r1v@k*7$zikABD1e1tw^0FAc#7V492wJl;VQxRpinccp&I3
z<@^jrNxa)`j^o|wK0P)(pGk|s1;HodcsL+ex!rG1;`rI;3TxR*CWnI-1=fptHUNUG
zXDJsS=Tz^8z$eXd(kTY06dSltp8j|+9qn1h=M;pM8Q7omX=K8pz&%=940>u3Kz%tV
zRRo%_AaHFsO6cY&dc2P2TH3*zHV=|`I2wrEJ<;pjC`>Vrg+k><0n&ipg_QU`dHSKc
zTEHW4m`4mch+k^=gg%nKSmvoP03VuiprLV27c0dt1jt-`Ykx4JeEot5Suu?h#WXm0
zAaJ?cO)Ya~3NIH^-~|8)>7<c?_J)&K#yG0<5^&NSWkteIGoeB7IrEpw{Q~@ASF(c*
zqF<#`#P^!g7j|4S2Ma6@W?tvh2yj8<sr&RAS)7_g6&06^d?qXiyl_u-@p00np@+tz
z2V!%$&|ILVB|V)%2l0za6?%M1z3{4|AGqK?bu%)6$`!dzV!o=Pj#Gu2i+m!0$TR61
z!Q~<TPngo|Qh@o2$dsWh5vBv#<)C@N1Gm47Oi*5&8!fg$W)SCujL){maXP1@l(5YT
zWy0WOe;GN*P|tLpq+oSRjf!p@*+zu=LhrG^Os7c%2l3}yX{&l;^EM_(KUQQuS>qbm
zApE5CUt*ExTS434v)k;V91=vGzDJ8<s;lsfD$0Y{uOrXG1)*m}EI<(&H+w@T<Ya?^
zYKx7^K?cER#Lo#)i>u3p%>tWYWCMfP6Le)s((E}ios9zZgoTm)Wt1U;pp$eOfA20W
z^}aek!^t5*)M>E`cb{JPq#XmX%~Qcpiw(!Z1))oZRgH&TT7=P4mwOb~BA>Pb0q|ZO
z)fUJAGFNCqBQJ*z?1@Y@dAjTVV0716i`oL9m0?!1zZ8oLUIGxgbTArBhw&tNdqT_M
z=7hFYDq6tF!r_XC2MXuuFswMR7#rVKXpoZO(!~bhr|;5z|1rh(l3nU%cdbx>tyEBd
zu_;=(AoRo}p+G>zA+xwoXNVw(cIRj@8;g~zTq9IaYjN!?To8JZrY7p&V>L18Y|;v%
zVP$4Ff!K3*(Sg7#5E!u|EjHk1vRDf^>-J|6fP%c^Yt!zeP4Sjy*DmD^yFhFD%Y&Rt
zYH7Sb=+g2V_31|dlZC$OFB6%9fI>t4(i{)PvV;p#`&D7)i=HtDjZz|}IG}Tx{J@@&
zNO4oT?;93KepFBitik<dgvXgn0usw&p{(>mS6q!I6d<uTX~L9pHc%;yJRo#_Oheju
z^1${0{j~#)z<}y66J=6B_;brNmeS>F$*6px^eMB#ZaKg`FI}&ecfI5m_gBjOHgeEG
z_+^pbK>L8xKIxv;6oc-x*SF>h7ZeD9ow?setOOu&dOy`Htrkbt?uH8p!eKCXzl{JH
zgf7apF%t{L&gNo*e24_N->L;TBLt|N6n5XG`Mi>7SlKY5H7JO>EX2f33i{qto4z9c
zGT-)QB>;(w@)P;=h+K8hGmh0Lz&y%SoxukJN((afG!}cnD%-!3_RFkZQm`ceiA$n^
zR1`rjx#ebfB@Qs@F>2rh0EKh4#6zr;5pRQo=*#x6F=bm+vx2fFU?l*F^HS$=Yuf8Q
zh<B4o^LQow!+gRhz(Fq-e&VqTKLNteDQ?XcWnomWcv;Ka;Dw^ARlA~Sto12Hta^pP
zf%dkv#D%&V0;^RsT%fmxCo7p#xWJQ8!!t^yy@5y$pY|%rgy4}^s~Q@_UQlB1C+&m7
zNlXVHn-409hCrxSv<VO5kEs{B<xccYm`ez0Xx`!=<+%n;kJBdQa8~p<-Ex5Y3dO@H
z+Ba(U@1oM&a^UPhaNhlvw!1Yfh&w}q=tyilKBS}s+jVzgKqQ!m`)!2KAogW3{oN<g
zyOk;lOD#rXh6=(?X(CBtEs<OZs37z^_uH(3gbM;+p#;I<luX~wBX==CURfLFupsU%
zJp)DS0}8$E$^E-ZDuv**%P+F9LGa7+*~(#6wU)JFh6=(?HT&cHwDayJ<hraut7WZd
z;DSKC*{p8wrkm|6=@sTemmvj8<@R|74x%q6YH*>1Ay0VRMT9zo6YhR17oIeDnfO59
zIEAN1?Y&AhhC!5(sUShj^>&{!p(wS3h-6V<5h15p;~rfsS4sXwb6wBKFOdbru8EPP
zxwl6f;S_G7lTeiQA-7@Vh+m1;`4U=%l7Q4H5!DoYs@h^%4=`{+-~|fiCzG~HLQ4A0
z?!=nNT`hwZ!bI+VOVI}#9>mMNuYPi0os+TFA1)k-E}NUg?pF>CVxM2pMj0@*!;u1}
zARwRMtePnVHG!Bd^Kqm=K@jE4>DeA_l_?Lxz*3lxeQw0%w$_=^k*kS@6(c8_69Khu
z-0M<u6!%#*CJ-j(n0Vy?omY)+RgC~-3~>YuWDt4L?NQYv!@R*sb_D<dxlf+=t7^i*
z72$*zblO+FJ{|)ECSR&-0~zZ2O6I3jt0s7P3_I%36mt_^6^J`$d1BRsf;++qwa`KI
zD|C23r5{uiT3%C)3*ydMQ(`rtr8TwCUD3Og`9K#pileWOL9~GmqA!cfR#OjuoV+y^
zUXeBDG3^GXTsnkuB7ns4$zT`{tJNC(K89i>F<mj%`j#mPl?xJc7sNC~X20-WMEZ<i
zAy1OV?N&v-mQDy>IY8%9b2#iipbLzebgYnEIeM1H@^H&^06}Hu<A{*}6t0sOCz2j0
zC_~xQ##~7W_OxACt>MAp=GYVq17erytk+oHI6zM*cj@qzn-9WB00Swteoh3CxTXoW
zX;~&O84|~O(iKWI-GJB6Fpr8v0jUj2RB_qJ<Q?&{RI~apv1qPHIBo$FAASF;PrU01
z@qgxVeHqpf-cdDWK|p{bB!+AO3yM+h9@nVVq?h(?>F#lIfX-F<*}-rc_h{R8^dO#2
z+THe7Wj;zZLg&ibr$7==IxV+RRoAXK<iXx8*CZx13QgrP*~JVF^x#sVCNv6#JQ+Lr
zhLfA+-@r<d`LZAd=Y_52{$!y`XRt!@4&aIdt|ZjxQtAqflKagO1^1krSU3dGCQKIG
z*O4m%C|pc`r!-LZ94v<k3}7|EeXSYf5_kfhgSZ^zj07?XK;V*am`2pdF=dU)yd&!{
zaw&uIMsP&{h4WH;9v#D>q*Oj^gK}R}5Syx(<Ad-k!fDbL=*e5<K%o&JVU+t?I*5)G
zAhO!l)38C+8z|(20Hw?32+MsPmB<5fmnkzx9GH<OWGP2+k{noZGEM{#0^f!Eni3El
z89?I2E3{xZ>@`PsDYdtJOA79GV2c9!S)Bsqp5b5%L*-yy&wierFK|wCT%m|gqJrT<
zy<AwDMFaw1f$Zl*(G>$4cz$VfBY3EXs321ikiEe=rXW-cf_2QtzF^%ku@{cx2vZtF
z!Obkp$B}@9TAk_A0=h5K%teQ&CJm-!P8wi?*bCheU7K*2h-*Q(y&?t;DheTM=Htl0
z2l1zAvhGo4``AsJV!%K$I6utCkqK=H_3pGYI2DB&&>-}z+~lX^s9CfOBLc%QLU0g#
zHQydePvqp2S2Z&OKbvDwAPFcf8y||n%9x|7F@jv&5=NFFAawO$)NHjW=|k*cO(}y|
zWyQ~6m8Ci=kOY*jWFO{Fx4OrRm@zg>0i=5gU?Au8!wUf-oBTJiSYy&VEcDtDUHeq6
z*_KRqBAgbA1d<m-s5{*v=B}V&gh0P89mX7d5PwSiJd51YDJcgAfv<>vpTRsK!ptZ|
zo^q48AnuH~%UaA$Gl(pooJ?#GdrB^vXOLN{)*KiG%0c3B2+V;|V7g><GGCas0fQ~A
zVGaxeU#6Cf_eaTF)p{&MoEUE8Iw@EXM)4H7Mnyz$>GA1m0!wFi6B>lRLR)V1rXa<x
zt4S>F8W}E#J2rOGc^Lgb+m$+p0uqG0Om}TD7kZWnDs43l3&Ji)Th$8~H;_^$wR9|X
z@Im~Ggn#cu@X`SmW^7SF&+_-UR%67Pv#C9jxu1J8+?KN37;*ikyyKQRSjbk~WPrz7
zxz8hB5a@t5okuj9%j?qIm4;g@Fkaggbw8&$B!>sVOL9|G%p>B(2ui_IK}DyZx|l#A
zKnuJeAah>L*Ax_C5_MQZ0um4?Vn2^i00=*w&+AodZ;y<wvg9PLgoDsaLTIs*<bpw^
zAv(ub+K~YiC`D;}mp|spSu+R^=zaz|h&{os;7ckCi9~#5omKp+-NBe1ZgcKHWs5`Y
zAgzM^oaWdR8|rp!MTJ`LCBb7RZVmg2gOo*cJ`m@-ESKL;!W_TggqgQU7TpqakCX*e
zu86zY+O%pEF=pDFej9s`7Z%hFD<f|a3TvV$AaXJ*N9`4?e3+Dr5I~mR-f$enA_S;0
z^xdLTdXs@f0fk#>$eB*@(ta}+@bxqVrybKOuA9rjLT055!U%hEgo{LjD(e%@7y#iL
z;?mJ%JRJ0CZGDi?E*ZBQs{3o1S}z5wl@|-tE{T+;y}`%{K?Ex;-2QS=Kw*LYy)_)f
zO|gMEqp7`8MJJgspmBV2Ws_b-T5rS)TUT$cHP(0HwS~>imG!IsccEaw^_B-QPc2L*
zgYtcekD3*NB7k;YT>d<44-;{RhCH1Z%crO8onJu6?_2<v9=?e@z##o8b(0IZg<~Pm
z_paB6%(~=76jUN#%6{|Fv(I+^ws_X-(=-g3uZ^Yi(^yX)j~;$XeWBe3=L0hl<9YR1
z7sJgiMo~3=Am08-EQ0BIFsS<RrG=e^?Z!^*dw<LZA$)Bu_D>ODftVL>H10?dVKvZO
z>y88s1K`KSzpkur?<}k@vTvx_Fhz9$ED-aO_-T7_W2+HX3BB^OrAHYKz(2pFZY6Lp
z`>+9szTS|se~MBNK$iEz;gMfoIhaOt#KHHmG-Zhk;!f^1_vjsr{AiZ1JP6Y#?0p|c
zDJqCMm;Pg%mdZw59}0-nnWesuwMdN)qK|<S?SlY$I<DgTI1-Q`<aGKeri=K>gD|}e
z^L?zxAqfpaPsAcr43`W0K=Ki|wE0p{kaR&s-lTrTiJV?@Vhs4@LUlfh@5{)=2l+4W
z)1fUo)wfqoY-zv9U_si#9^DN^N3^nEL+Yv?V9b)6^s3%lV_OC=pchZ7*S54}Q&2~e
zScHXCF{@mNBmoMND3~y;Bt>5b?^PiIf`n7Tg-8dQN|N-F*(=Ea28kDqAN1f;NwW53
zLnLbfz~HQ4uusl(G8&RkU9BBU$5jg)B%c;P3f^*JsZ?v9C#5Gv=?iAiAoZk>Du&-m
z0`*e4P|ePu{spnQQ%O$g#32x&KF8JZj!IH=Jjq)dg$Z)bsVm&%y>Zebt|YN^G_%n`
z_8IYX3$63hoasuEb(Gj!SHuRnubNvsm86!=$tE($ymGj&`zf8=$v3SkiPzCBZ>2p#
z6qBU$9qEJ0YJu}%&jRPs!w>Nb8>rALa1&ZgS8=zy;v|EpkhA{Mgvs%B<_HfC2>a|(
znQKkRA5K?yT9_U+0O=+1tH+hrLd?*Svj#{%^0>|~cwJ`>+*t%@aL!fMUb!|S*Hz|3
zIO?`UXf#mLv%G7};p<DTTdIM^TIsGUd@TLM;|g=&&ajz&<ad3aFK~Uegv?#>s-JUp
zUvXVguJ1ZR^gepaR*;JaZSZ2Dm+6S{+Gz#@=x1y<v#2hX3+qgpDVfN|1c}cVxUNoZ
zye-4Issh>Ps#ZdFT+^pLuBn6ianH~rd|S|h^k52HQ3u&^cGE9RAdp}>ySj51HEz}$
z9A}r2qya(V3CkM{lI9(kEHIFCtkku1^yh@S%v!bdTmVvDC~_qo0ZydL6yUgy&-h$N
z2iH5WI<De#epk^!bs~$JE#n%Vw4Jx0hV_mPHX7*sIJ#>=k{owe*9Eaa4!Q2dzL%ph
z?;K<zfsC^rFD$51&eBU)$^)t-C!LqTwR^$i+C}p2TJeqt?QXWzjP&T?^Sryqer(3H
zmlx%`OwQe~1p!b-K6!3Br!{3X<|IcRVW9fg1cHo^xZV*X+J^~oxQ0EpR{&C8qKHp1
zG{tVk*Aq>_2Wak?_dm~%>9slvH7Nnv^#aeYLW|`AQGF<P7NeviiZTKV4DZm}p%nZQ
z`T6>pX>LfEiR^kEow~e$_f||>#dPUV(&IzQGZDU8hp<5in2LAXDE&x0r(oRD9m`jY
zgsQ#kQf|PZap_VB$pV=`?rPTR=`G>%f6qcopN}rIOja5Y+BTSzih4~I*iPSk>e7n=
zvyclKE=wf?(aoaI?;W3oC7oj1r}qZ+LXw|xip2uOjU}3R7U@KwxSMQwFxsBd2^4x>
zD`|-z=xL|QbK+76!!oJB@r}$R!p8D4aF`{72B0!TI(JI@j^YRt5Pv<U_hGu-@X@DC
zPCm%5%|pvH+ADsFb<#Nj&e*r3Gxq4=`@r9Vs(w!1^DVAN5szEY0ZVt#G9q2}#{l)I
zSYyO7i}L=o?#6ez0v~}2d<H8J(f>@k<ETmbBkD;w+Rlb5F<+Q|<4YpJV@-g#y~4q!
z>V7sQ>iiaEKv44wVgV;V7DGdM;m=qB4jtt!kSCYl@^`JWr4W_v(oj&WGUE$S*%d0e
zR;aSPq^q+0w<rVB#J>o{KSz}^vtCjisOxD)n)_$fS9-sO{m7($a!FoqGg>RG0SKW=
zSWO2%M<$}K$|<Ody$DKK>UT>4&RGN%er#6lN<+PaYVz@#H7lEHsssB)QX@u?uJ}4s
z{Aq$=+EA<@RWs;j{Ji?mYDVWfv%1H<c}yvHyEN5K$6!2Pa9K^KuPp~Dl#y0an+TDW
z2A}Qxnn?Lt(DRQwZrFk>n0!XjdB48OqWEma10I)@oz!5VuP+E>q2wV4Wh?I-2PF?U
zP$G7E*HaA|R^KXV$o!rOV6qhF`N%c6XwN;cSq!1TW`(Bgu6eU&4NCgXuwY6k8F<|k
z9&?jkRd4s{w(YUJqjw%nsMDbXITo=1zF(qCid@E6!5XT44PB2i%L^a_>~FIL=ArM^
zIS|iy0OuR@R5!gH-W`mZv3f5+ChS&{myfmj)|f!MaAgCx4dFHp6S{myyvIz(TjRF`
z&l(1x*KmRSc=^Ebg0+7ThMa3mC)|aVlMgZ}`AO$l@G*po&-)mxkMro^OR9i7JM-V0
zK8HLUE}pt`;x~CUSQPR-0_AZ&$kl7hjS9XUb#EUDefMAqUGgayfzbXN)`itGA;Q94
zhdF0L;;y>hEOeq+74a3%tyagBX5*;&fFjow5w>&B#Ic?pl}Kb*;DCJ%7&7kJYuOk-
zKQ%5pQ$dBgdYrd6+<||7gg0stcA0?<ta^w03R=ng(Lyb*&c^dC8E45ZCYhg3Z`9(6
ztU#YP#cz5%7|oBc>!4O>9~n!^NCDqFf^V8G;)`hc=AE)ZRYz=R8&n-)ig(g^72JfU
z2%UEmq}Z4)IdAtLJxrrc?qqF#Wx51k2na(MWyulHi8S(G#}R4G_??^nFI-!W7jEus
ztZghcRtrwxv4A8{A8BKMr^a!?q3O@U7vgYNq5or4hGgsP*Z|oA83Zq_Zfq^YJ9PGt
z4wX^mZ>~1BXQ;luLxHDIAOpy}d~JD6#0=Mjop{4)*qBs*eZYekZeoJ8^Iq>t0x4^y
zrZXs+xuQB7x?C#fH(gdTAS<9834A#>>Z34aF9Y5xLhl~^S|nBGaeX^@j(P?_;NgZ`
zygWc2G)BXE7GPE~Zp_6?#*zqhF15z##VWHB(Pui!8#T&g0fmcnK3Qf_&C0-C{drZP
zMFD|J*(^0Hg9<vxWB~;V(b6jraSJk(%QXt-sn2Ma7o+fOCJksbvaX;m+nH4#6^zr8
z2z0KzHKjZ9YTBn8x%OIzDGrD%$oNS;om6NZLmJSyoIlz;Ys*v^OLJjB;#wY`%WRo-
zIelu*9rp{4L+mR<d(k?inF$3}6v@^f$WAmPqY8#=2?RDb@*h==l<CY!%8dsGsL7P;
z<7V3XJIT$I3*4^fdNcaKjMQ|RlhlOc8k_EpVO@q;n^~%K>G{)EK!<MPekYhQVu9If
zxz5z5;t4@HpVBeO840@Kij;-$SHVFtROd<uo|JE=duaY;aj3;7G#^29zMcFsw2c|@
zz|`1*$UhC(J}IZ`?m+m;uzE!fz}c~a;b2T*f0Z@(&Rc;g9eFnC{`!03xOuMHx<*jJ
zT8pFu&+S%H^zDoq)j7hl?#1R>EFZWoVrXbaZpABG7%ovBaJeFW%ckfVsZ>~KfjD5Y
zEPg5vGWj}SKE~)Y{(wevL@Gu}=Prbg{JTt(<n0MvWpzNeS&1|`ozddmW#Yd#LrNjL
zOe_k;8{EJZ`z04<qFtsPs>@bZm9@egDOv@v&_u%BWn#aiS&0z{7E!!~Qvo%%4R;Ht
z0t}xeQ?mU~0Y9hdj*t~#JlhZ3Mj8j(&nt+?nYtrn1sIw%v$+IXo>Fab?wlXtHD?V|
zjWF^8)-CNX1?Qbq5S^?1POTagafI4*g-@W5F=Wk@zC_8VOH`>2-L^2Hhs3jUpKsKw
zRS{RH*4t*a_y_IOmE!ZS;{M#KSjZcy>)*Po(pXt5xm$M?U?7j<!)k_AcR+lQ=4y9;
z=F!9NR{xIknyO#3^E>Z;;t9GZPa$}c!)0pvev^*k_>#koFXS}b>^uc?|FbML8e{1M
zl?09^Q~7o#-6_Gb<-RoM`;$AWxOZ2<l|+-Qm6Pm#6Q!_9SxM&@RLYZ@>pC)gYOeF0
zS)ntbK7@@z)v`8>>@Ok<T37BGRL+xd%^IpN;hO*0tmc)S3hojMNA<NFXZEv*5d>X)
zD<7Qqr~N%U(ijX1!D-6_2?oiXk64faqF<)=R3{ISv3MZOo#2B-=5xq44zR%pQ2{Zi
zLH^v_T=-g}X*l0Zf4T>}7kYrFd(%NM()0;>L|U&?`0nj|XWIRuIsh$>yxdZF+sRvN
zg^L9nj||^hyxL!+C!!f4&>XMt$&=r)Dy}MJn}0Yo*uNF|2}g+-PC)uKk=7Z%C9WwH
z$%JUK;c|ejwX6w8iA*v;bGzH@A55E+B}=oKI3hI{qE`H`0#T#`R5uUXtrp!h2~R^y
z*5fE#0w~l<nGHvY3@$)Y=KPd|H^Rl_W1f}s6Aocs-I9(6W<URmLQ=n|Be}G6ka6my
z<&J*ga&zi8XEmv`Re5fp^)IIV?D14!bLI+oXaV<KH2KF<h1)Y%p_m=i<W6c!`*d&V
zl+I<06Y(tSe2mD-9}BlbZA3^4d1rn>`kKgIlb(3`TJvT}0q>h4M^bDi^oFz>KR?d1
zk2MY<<l7xI=_%{4BfbtH;QJaxvFBqCmcw1aj)4sDyg^QecqM#346SYL3Jg;kz)jzf
zL7SiIG8MTALGeIJZ|k=YhLe})!`g}~RbZXQzWfBCKC9_F7IPXNtffH(+M0)AmvlZ2
zOM~MoMsZHx7Y1&UJ~<NYb=yPzKOQ|irpjOxMh~B9hA|c~9f|9^({{I|ee}uTej+oT
zeTQgNO|@`<rFDa~xlr(N<cfv%T3TFP*}NrJFgt>iH~vx}g9up(C<Px!mOx;$LD!Fp
zF(=7yyDx7U8xcHM>mW!4Vo}BgDBeyXWDDXWeKo0CdG=MPI-G#=mU1-WZu5a}sl^&v
zp<>yAItV5*=>XB4^q15Z;TKy$GQb9`&6z5SMM6M!Lr#ve&3!|37}YGQ%&xYAFlZ~2
zJO{}IIBuz7Nlh(GB)@_x^7RV4cnzZobb#qL`k!;sXpNRfKEQRg#GeS#MXpbFjtjzw
zIFDkhHSP5t#B=CFH1pHFNs|XOzSn4TpaXim&gKnQu}OoWl{QxaC6Y9N*$w{aK?Tsy
z7}QvgF|2_V59G$@^S<-k6c6B59)b$0wy8WerWb_bjX`bpr#*4+H7#QED<5Z=yp1NV
z!O|cDx@=op)O4Ra-SZ@GagcZX5#eTcPUGa4x$sArl0#UkdL^B210URT8VwEEcN)$5
z%Jj4^xGt|yQ<$esW1YVAmA42NT-WYts9pLTS+>F$3@7bg)NF}XY$yAV9)6PWX|q=^
zC+y$N8ojX4YF?4WT=(26NBs4g#`-17K)}o{wGzElYEcGUJU+;MZ!ZG4SNar*_iH*|
zWs3j`=hfX4@;;>Lu$#Y8>_r6m`k<Hh>&U_f;V+LU2G`zykU5)P7!a!W0lZ(+W}3tW
zVW;=oZ__)-bb7m4Rp5(h7xfD?uu*7Ck;aKgG)dsnJ}*-4ns|&eHzi74%CSM}amocu
z%;08#xTj0&C_zEaF}lOme6c7%%;!tR2uvj<>3kfzb=qs0ZatXxTaO-o!q=<i&t|87
zPV`6duSI=Y{7zk3Ulg8KPW&L7!NH}MDd<|e?kJzyJf*;%pK){o66ozV1q!-9WwyRC
z75M~aD;<5iz2`%}RYl&`2D-Tg>x0HVNoGr*4r^)J2AnkgEv<gGTU!2WK}*Zuxt5kc
zh?aJ7y~@Bc1<Jq=C7w3>eNaYf)~?wfjGAQiANN?j{MA&&GgOhQRHQ%?q6{fQ-?*PO
zocMR42T>fX@I{|Zm?d&0JZW3H`5CBS8C395RYz9^y+;o};;DZRh(1Z|dj&P`QML7>
zs9CjpqRxo(w6DVI+q|m!Aq)!)D%plgrVYtdvibI-hadA*lYTT?`SXHulOF0;S9y}>
zW$T=NrOJMc`Da;O3y+>q%x4*HoFTZhTwZx+M9V+&k1x-ARm(`w@1o@D%{~GZe-u=l
zDv?p~@X^E6$eek%hYy#0PAGV@NWVi;??)W<4yBehp|MWW(>Kk2pWgVs*G}%U+D(gn
z#!gms=IdVirFbba#{_YwR2Z_lu%~>ri3~&#Mekt_#^M3wal1#)VU6082bHSl%u+`6
zY+w-l1+j@I4%%TxS8gX-q1l93P`MlmWIXG&R0sT%Ud;&%g_?8@!Ak9X&z$>xdKmmL
z5la`^icb<J-_!r4IYe9#7JPd<MOr4HB5Sr}l5Ok5^ImDv(m?>I?@DqHR?ii1@*zg&
zEmkNT3Aw;zQ6`hHhG|vXtx`@!XOdN+A_W7ND_*0jKxG!Kn1=WvFhuXPOBdDnrrkbN
zN=&=VZ6%qe`rNeJ7s^H)m^HZeQf2ZsaePnp0pI}BmuPu2X^ti_aSIWkuj0<DB3aYY
zGlKzquj61dcN~Y5dEQLv9la-MZ0#S=P%ugQW7;BalF|foC)koxByBqjuztkg0QD2<
z35jZQoKcnJsP{1px=Kvac@xZzpQU9i+cQh6-^?-Yb4wx;?;F?T_s%RW|22{Yq5k}E
zZ+xAb9S--#C*YKPKvh>A8%PfrpilDX;RhIhPqAut)<9i+FQsQJc0IjR{$wia$yNH;
zrz`Qb(@VuaMxqobinhv|y<tqF?9>h<Wd&g6i=8O@#>4>u&ybCi7Y}o4Rtk`@7iHgw
zARs#Q&$W`>cGF(4X9WNhTQ~cLVr&8enDX5{E0SA`k7(ZAlg=lD+IvI?OxRJG+%F{e
z<m$rKw!ghcW2U`F<4BZ(O2TOB44Oi$ER$CjN9F7TzyS$9H9E+)8q*J;+pU5c`BA1X
z4TN6{sNc+v^SSi}X^{L{OMdT+GV)&|SrDpvtZ<=al*-kKW4ZB(U!nY1W)#iQ(b|nq
zK&c0yR6a!4AKV*=mA)r%ll^R}_PKQ`B;pM|JJqE~!5E^`KSUY<)La}lbIRuGz!iz}
zp<m@vT~6<C0CGv#OPmiu{hv|lS5wKS*)WNm{>4)=G(My|Be=#jK9DWY>~x?}`JO^e
zLqjRwQ`B!}bAE2CsHe#9ot`59HIfCP35$D*Q7Ts{_7vk2f3xvD#VDGiV^1+Yfu_0z
zO0^HjC?}0}fb1onbwI}VovG|I+gF!DmUoUcY65-!%_AqWprK(dvI{#$ij9#g61THT
z3%U!sOZ<Ya6e*{mZ@UVtf8;LYF0u=`3KXhp!E_Z)f|2pAAyJ`l7wam>pG;GFlG@uF
z>=Iv-$za@6{G(`LjC7rXsFX7rW`&$M7p^0+?+T(^FB>tb3WCrVH3Z4jJsC3tV?P%V
z;mS~)YUz79Gc!{<qiPjKw_YlWZspRl*p|0ap2!yOZ!y5=Z)G3AfcGTWA|Ku!4~B<r
z+R;hp7@U~WA5D9FdRrE_>hlUz5i&}Ppez-og`pi!7W!QZPZ!LG30#BVSN=KWmFci0
zwt^}7W!zS)by&+FPA={Fg0CXPgu1=J4@IDiTng?KAQx!Z9yBaJ$UGPvv^|52{K;(8
zC-2b5z7`c<JA;h)$4Hcd>hji`eVf-5jU)R2D!C1nOlRS2l|F0_DJi6%sJGwr|9bQ=
zO{~#&tbedRGb?{~b7ix!y0YFVxLTNpA+6}yn*SiOp+Q@hvR~I5jiv3l@%qlz0xb)j
z_z@3I()C2ee~?Qz@qxlC;&|<5s!G=Ycw7>!`vd=hd<;*cg1l?upo%VJ%nkaVu$chI
zei~0lS_6v8V~+Ye#V6^!2h7XP$I@S`z5iynebA?;Ann`1`L9glIw>BxTTsOd^eZzA
zQYoZfJ-D&>^4+`?5|n$Druby<ZrrAelZC|QZr41CQVN!LPkqFqw21HmY>~j}jAX*K
zv{_}dN;wOsZ~2NuQiyKYC!k@y<Awy&74(lA_?g+XuUw&~-7N1rfWY+~_&};cqv|UC
zmb&I7z2_u-^3WX$B@kUjA}sYu3}5PsDtlJnn3eg0s0`f~!X1KP%2GA7U&#C_>mb3v
z78Y`2Ap(WgDGBB<xlQPsfeWP?R+$O{9xn@YuSt4EYmc#5v?#8d?OkAr_Is8YPHQ+&
z9aglP^`0gBk=bzP<%rsBk3?#ndf-=7%|C}2(x?@kK4$!^hE0G5Jsw}$+(^fvf}za-
zBzbz*oU2!q<AIzDWXiNd57*I`l!vIXe*U1OjPD})^Fpn&OV=5iWzsV#Q1?r6XrqVP
zXRf{DBNz>0Xk3k<sUV<mS+20Pab>~(lKMak=rw&w8MT(zl7Pe|5!ukc%)}v752hfX
z(4d2<?Ma&&<g2vhH=;+$)hWhwX;4zPd<Z)mRv==;0iR3qw4l}rH=gk=oihn-77zp+
zUX|uKi)@|JTcS3c+81Cz^-k?n?|UcmA=S4ez|YJ<^_cff7n5!-gPRTkox5&xVJk_W
z6kQYJ+rsEz+DrNq+6gNhyg{nIpq{5icV)2>pb1~0QTN`7pm7{TJ4*npu!#!t7D`6*
zT!<K*a%Sf9CMDH2%lSt2#wgM6?s$jR5Gwlcucrlo3W@}+T+W7nc5n+}!u4iPxhRtc
zBo<328xI}~_Rhl0Zy5eb=Umai;$x9T-MBf|Q^*`((!7qf{MK_h%u|wwoPIsSoCi1`
zX0I><zNZjHuJoBA5Dm?=L5I)o(b+_CVr2;K*{lqgDlqg42ac<B=SX*tu0<XX2Yrz}
zpY#tVhvq5VfDV(ZP#+TwC`8UbCmGn?(%*=iG2|PoDM5O4wcqSQs$_u3oTom<9#AOb
z>EnQm$CTJX@vZ~i;_cSk>NcQJA^>LAY!$|39Dwodxa{*;tySO47Nq1d62SROrJUpa
z=GaV63*eAjhkk`Qpb+JRl{C0y0_-oXCB4Du!FF>$F{hp4wOKw8krPugh4A>m;YRrE
z&ZEIpqR%yZ6viX?X8aADe#_=r9vtW`77v>|0wpM#jt6M(dNwc5ft`L%y_)he20(1B
zif6#r8#Grkp7RKPZ(&Q*+iz1kCY^3RoP+-KB_GxF=P&_1HUixmUIF<?!mT0mo*ePT
zO;;!Bybf!rG+gM$6vc9Q+N8(F=xi<ZZR5~xrkDH^a}5<WbjgANC@V^XkIuSradRk8
z=9$b^-PX>4^Ri%o&i)Wnz~d#$Bnzn$Pf&(|0@*r!zo?=zvmzG)^*&oP#+C_p0)9*Z
z-w1lPJ0RaG9yR-St7|++a4rCAe19;a`=j7;&m!%Cx`PDg0zmL7>%NI#q0Z`<Q!{gk
z#Mxa^!QIDM(_8@5^zz<_Ze~nmIH^0J1DEET#D&$)R$b1joUOAh23Wij_r%o{oM*^@
z<>O(Hx)nCado@L6w*v|#I`J#!Yu$;{4S{TVsfE&OeS2#4LH`fkhdC1CY~{K;%S(<)
zTqqNCa)bG{l_bOQFo9cN!;C`T|5EgIR3s6&ozK4<4$TWdir{lIadrZVRYf9v()-=V
zE}cXu=}G5Y*j&9&A!)gN<)5qbPt3Xc#qxM(R*+_Dou`s1J^PW1b4JUWR)HgsA#jdB
z_G1nO?NT@)GKGt$2d02;LM-w?7#NW2cpyvxp9hkC>hVCb-*bh5Y{vs(3PjnzNg<%{
zOI=z8tKHn;fF7Q3y@9N-hk>&j!xV@tDL~=O!DuiY$``1(5tD(-hXKbUV7`uWe2{xR
zohQ{5@3a-v5(n1kj4kDNS0>`DZ#g0b0)g2{-?labn(M+zHDSIM&zoeM+LzT>q!Qjg
ztd<1MvVn3^`Ne`_-0XZajA-7BhU}4pf{X^^8*@xMKN?R)bcN}V&J@z7T|AxapB4?|
z-(1OoXb^I2B>y#Hkf9|mjHkolU^L0M<YV<nbf9c!nlfbb&>;S$^3gm8q^jymA+q5-
z|CJlhkFE66T_xtiR0e}2=<I*y*C<DWN}MZL+!_4x(lEg3YpjX|dWP499N#m(<{udJ
zfq>%+vqkckQq%@NXyAO&GLMA=2F@M`Tf{fBozGWd?%>WM%_2M(f;G=qV~<gIE(C<1
z_AU9!YwfHF*dh_L?7I@Q`eNboq1@ScW516aQBeJ}Ss`vV%YGc0hM%nf1G`;@0+JVK
zyiS_E{5ilG2sr6snJAMmr5U3n;BjtynCwlv^fcYQWTo;@SJdrTV357m8a4MP>7y3>
zzO+#z77zNN6K5?e9oYnc-irD)pof%GyI=>;1F12+XLGzapeH57cIarZTVggmGy8TI
zCwN}U4mhCJ<W>$FVkkp(j;$2+akggk;2v!pVg%wQ?J>97dy}{?p01Ub?$1SO&VEQ)
z^&*!DXl}bn>FM;=1#KgbsW)bOVl%PLMyYF+Ezd&(xHhuN)??>{_A}g3#J6zv0q|M1
z2f2E*s*fbLd|>zT&S218sXV$mIk;KfM-d`|v~$vxD)c3=Gz_9evl<pt-V}(wm@zET
zKO*I`y0cQw0PQO0-Xt>Eqx7;kc{|yA{0uTtib2xZgmqr}CbqL&jN0^)rkrqNN*2m9
z_g>@|k)F@Z)!*rWH4rD07Avk7$9>4HjFwjPm$d)BP<rTttuzsEQp(s5BSsjs`)1+#
zK{GWK&Ot&?aPowLnfcZk8%4{!nK-(WP$-ONIhbuOd;GCbc+jM?xg4%Zy~8Z%K!fjK
zW&tzY%|z9mE<<68QXOuqyrp7SH&hU)*2!oJSFOVd^-C+pyW=6fB<x8s2o?l<Yv>mV
z0om)`OoMQinaAYBDT6W0oP%`#>q}~U*Oy+UQ1JX~9<~X#fQ+k?WEIZULk#p9$E`t3
z_mqiaH)?>*=3{nxtzLm0$2|?ji2U5#I)JTzoz;Qo26eb2>YCoamiY{GYJv57Enu?(
z-Xei>L<zCH3CVcloNDo<8nr0p$kvH(wy&=`6>x+)eJO3H%(l{_QFCgXI-`_o9Ge)h
z--=^Og`;s}p32wn;8Y)-DXrb0qnIY0?}W91_XM_#h^0=C{{%LCWv&FiP`(l{%aZ@?
zB%+zWbAJEHEehI|`}-ygiE(z`%-<se5?c0Bkyr0S0Dr?-Q2I*weD;e4KD!SK7&zO{
zzEU2Ke8F^qT{qW<1{|CX8eb{p!Egs7Pp_;m&nVqlV|t~N3~2wW%&|ajO!5Kdfd%J~
zzpqH?NyG&>AUGQvCvxv+Bhu6Ku}4cc@~3uxIGyMZW@A%9H?BzFbSd8JCe6M#QNdTu
zyq%n+d|wedQId%Q4vTTCJ>C;XUHp+~A85`)-?E$y|GpyRhJrZYay=HF^RCG9(@jJX
zX!8-1Qv95fvo7&dVx<DJRfd_|pq$NwoPcQ+Zc;9=^IxRl!_C#?<TCh*n5(s<a~tMo
z?`hkSv1LbT8otxE_?bCF2b{Kr!Zdd}mruY4+#G#zOP-Tj6Gv@VLgLPVNFv>7)2ufg
zwH3#(EO=<am7;0091CVIXY&l$@?<=d=|;Bp^Ib9!fDPoF2wVU_9=m2TA+8ttJQY>v
z?CgcG%Lzb}Ue6O47)RipAeIwAVtI1I*N9k70Ek`7tc@L215}m;dS>banRsBeQk--U
zU?KTnJCl9o@`2a&IRBxxzTC;OT-&s8<281R;$Uss;cM?sMVedoQEiq^^fo~G#F8Zg
zy4;B4d)=;4<=Rx8-H<S<$O1UNLC3!blvzt@412WYody^O(+N4Wwdroy@t`oeUP8e9
z#y%y?(8B_8m#m^L)$f1iiFZ9A{_ml4dLV?eNd-u*?hSgBaX^Zl2g`@9hi6#V$TJB8
z1;sAud=NaX^qif$OG=eb)B8V76B2h6FZ;suuU-g<fTYOe=X{zm6DgJRs$MqSo}H50
z!!$RN;cVmU-%O=RJZQf&f``Yr6h((lHe0wpa9U+YBxE~B&@#uB+TE6!PTJjZL?Jmk
zgely-v;&J%lgB(w<u78k9pzyj9EeNV{HeoUAX2oGjpqHBYu#I$w4<1e=;88Yi7K@?
z=u#&ijBd9lhnvk-i*6$*IMLyiM6{ei-urP>N&zf{=-o~FA{XCvduqr4_x7gUL23X;
z8DFTzo6_HWd{uCw&vmLGxdMF(ou}3@M^!-^Y|jxmO4&jccIm;q^iGh+*9^DjS&N)7
zRHQBbora^{aPrve^X9zk1H7SH9r+@-Ff^6F7`OA6o3j8?4XXxg`|>>N<r0UAC8;%9
zkNC{JPIu;6C*Ta#Ihc&5{XI(U8uX<%`}jK37w1_ai#1fK(;n;B;pMB{bX5BB)oQM$
z=MsmC$)O_Mmsa;RJ-&L)eW(E5P_4aQb9^@)0v}(6ZqL*F48~BEo^}_~06Lu+7gq*y
zE^(c)FLRoA=UFM9JJj#Xd(y>Zef+spXs&)kN)~Dm$Kq<SF1$5Am$5r*6DK{WhGHmi
zqS&fBB%K77J1>MRchZ|s=}!27g^vE%T;iOnw!}#xXcggH`e14z{ft}IoRha>?Wh}E
z16MG))WRi|mTF1>``DUN?vF^_kedO@f(9k?2L}rNiIYsqDZ-=_u`Gb-HJSnRZFt3_
zW+^>qGqseSAsa|73C(o}WA?VYGSoao>m*p06%{LcDHqsm<$H>PQ%qfS)ZVMf%}KH;
z<;JrBrfc~(b0O(mIZ}aSN<i^TIvZ-lQOC*c*XwZPbY8EsewKquvvU!~uvd!5u%5g!
zkM_m)Ww+FKXwg-8g`=6Tm^rRp4jtD#U^K!xTmMSWo&wxh_u7_@;H(8-GF&KxST%UB
zz{kqN01WhUkFL`eZcJI2%OiwXIY~u@)rf>ZZG+E&#hebe7%P<j@YP_Wu}YeDvp9?0
zARej1IVk9=1IuvMvp7q$AS_XFQmsm<n2qyY&^W1+;fufy6FNPcPDm7HE^H3?)@*<?
zn}o6x=pL9TR97d=tm<lh6&di*39i#HM28&Ziz@P36~ABa!^`J5XSc7Ih`4VQWCOFy
zg7S9r9z}c?$-CWc?$HHs_M59X8b>BYzY&;UNyVTPV6iMYR!|CJ<+?e?4`3x+l!}1S
z8bvjvwNXHzH3wpzd9{p~$-?263<s2I^|j`gu>gM8h`F%4IjGLb2JqNB+C-ERP^^+p
zqsm|-v#0_G8~D~7Y_=#ULwhRJM0CoTF5nx*^pbq?)|76nnH*By;PLqQvGQrrgs0Bt
zSrC|CMFDc5t(LO_U9p3z$sMmu&N<+$T8bx!vVhQym>BidsnNaRG!`$PH}5su6kzSr
z>*r;jgei|y<{2-|uP6`JFA%6*UTQ2a++5v>7Z+AnuPiLyD16n;tv~VcxPHLLkpd*H
z>7O^ZHm+_pwzo4jm6hXJ!Lei<X^)H4yW6c}h`q8MZ!cVJtS#JVY{hGv@#e<%&TJhh
zq{3clKx$!pmoA~&Z_}MoDrlD!KJdZXg2#aZ^}Aw##w(#tr3p%A=<L+H1UbiXfnG9P
zSzlSd8ZT^J-Hz86)*6+KE~M737^t<PLB_DXv#`^6%rrm@(73R%xwEpi@&<h|lO~+|
zz?eor5h#9p@mgc)=IY}Vk0L<)*=-sHXO->or%fx3415rLW{ZyHg!=I&BH4Y2Tmv5D
zUKK74y(2vB({%2Ootg6FVINJ@eCNjFsQSv%T`##{pmd2pyUy5=Q7MQqI_b~FWh3Ww
z`jx{W&DK4dnR)NqS1DlOoDwNvk;(*JEM=G3)}rDZhbW<vNd+>(t7@l<h+1?)yegw@
zh5M)F0-@^^#uTqeh;_~0-J166<ieKd$dV1jmdA&bU@xvzYz^O0O&M~5(53xxO(s?Q
zln@9MHiunt8v%tm=zT<5%CwSu?Y(+E%t>l3X)=}pkh~VO)e+SFL)<dc$U!oalIu2o
z0;!fgRV4+Ifz(xT9rau|Rh>9Y!9ZzA@7mUz$f^#OnN%QiO&R+AQFAy<y!k$l#gafW
zkh&~?s%1eTYdzDfr9>ccU8EILCcfT-szs;jsBOsxVy{!awa#XzB}jWRj`u0Rk!}Lj
zqPbelj}Ujw`I!Le8)8#j1UX}hz_jRI{#rakA`52g+QNa_25m!%gkUjy>&Hmz^#td%
zR>^?R5dey7bS(!xK0cY!m5j8(J?-^r#apedz~^()?8^u`;(^|x*!yk{lecS;@@%j8
z#vZxhE`$P=tAfhjv_-EA_v-mHRmXc%Fi={Szo=+hEd^EeM-90^=sI2VaxkJaPddOX
z-oL9ws_MvQ$p&JV#g(D;e66bEzLW?guF!3$lhL4CODk0uRjEuMa&>PyqJsjHc%M=s
zYO$$m%BEnTv`UM=ancl_bupIH4oEFt&QYY2VaSmV<Tix8i!&?6c3(ZcRUOAU0zh$D
zWh#p`dp%}VhXO+`5W2oc`ynHG;bcU|%qZceR&!Pz3M|<`Y&G2kRJR)NuT&LV3ru)b
zEzOY*<d*3yhRpI4oxs1*=9{@ncTIf-BU3Ii8V-hE6Go<7V6-x#n1=|RrG8%>HB~h%
zOEmD>prt+0Qi~6|-de5i>#(c3yk`jj$2Vw@owzPv+}@|_r4tI+)S>H~SuO5Hv_)Uc
z1(<KrU?TN9?bBdLokTgSb@+Nxqr82DX8<g}B(EjWnWE~sFvWW=j6Fxe{ES5^VIdn}
zuP#+lArE!=4Est>akJ)QQ029NU<=ZL-Fp7B2o?F?y$j`6brR;}1H&7#^<{Ww4(h8s
zvj~2saNs7hO0&pOedT$wU%aN6k;~i8<u^%a<2C=FPUoJ5pCumny`H|rk-Z-kQ?!BB
zIfscj+g+=ltW{fY>o-9>On|k104w{pj!G-XM%3$(nQUP8rZ9-?C%O!4EERf}F6KEX
z$jbKFm-F_5=hU*Vd{jgSh-;P3rRir>X(k((E%&EAx;kDyAu)>zDb9B|XwiOZ)m5D-
z7Z_cS^Mw^n_EYxeu+ZVUs;+F=z-*l+bMfjIrE}%q)#)5nClgLSFx*fpPPI#XnC#sx
z$ujj>c0M0hb|>YM*CGHMSA-9qdb}dUSx13YU2#h^@LE?%dY!hu_E}7Pb;wn1T~0nQ
zTx(N&F&qDs3hNN8I^8kifnOsQ8#ncNRShQ?Vu8;k=@-_~P}TUI6bKxaXxEnFKqA+?
z4iV3RCSU)Ja;d;(oi-Kvlz}SwP^^`rEd0%kq2p^+8|UN$!xiZcs|%6n&Zk;FQq|GH
z5)Hh>whYCn_vxMmBcNBONh^=690A~XU2)tM4zY|V)S*{-=9Khu*}!av5^KlPBqm3#
zPi9AM+rCJI7DeiCT&d^;`&j_nEqX`%!EQqI4%DFuF%wK_FQ5)r&#^y$5B6{XrYow&
zN7R$$)<GSDm3xb9@JuxDs&LT7r-0lAsceZw-p6gFv3wCRw-s5{-ReRjpv{fL2g77U
zm%>w^UUVC^=TWDvtByS^0pNInI!hf!p2K}U>kui)ApaWO|3EjZ(2lgc9J&q_&p{I}
z6_gCDR`9o43#`ih6BL3E3Ek2e9r9SS*zObW1ooQ!CS@Mf(OuOQbP*5G<m=*vzCH0M
z`M5{#JyU>kZk(%zQi_=X>#LcgHJ&837P2$tpa~C82nJ5erq1iMb=4KWDHj-RP^zlB
z=2oZg(DiPEI@GF;Z!7_aqq_0-Ysk?h034U(#8=z%tGd?6r2?Bp<vY>Mrgd7fYF=R`
z6u4X}IfUpN+jA$R>Pk=w1P)7MTHq%#eWH#+Dz{E171&S?mwLZGKa5;!cB&dnV8mPe
zMgx(4V=T7bS$?i~;CE#nYl5mdW?CjNS)h4dy!BhhsH!Fir6PgHO3F@L1Xnk#P}SG5
zL<6sDv`bBsU@JaoP7dpNH<dj@kPNIg<jsc7y^*+ACY@4}S{7b)&SMDx$Cb31MT^$;
zU#hmKB^r3GQVJTK3MS`9hPNpqT1#<Nx13$+!0xIDYR8laEu-~ydSz9g4+I0J8|g-*
z$}<rU#nj<db(7f<4&10GOvi^`)5sb}05~?}7h>CI%Dz4y8ZNCG**3%iAG2q^(_^jC
zGlgj2wU!RDa$A;m_=fHM2X!p1vR9$TcvC#^yB&+1wQ-NeKDx_()E751h^|WM?Btu(
zVOlx*Q$$2@PC75cWfkJJHcA`t-Hbywtcxs&bS@DWlc$eyzvPLM+TWX(R-93X5CV>L
zTPU;>94L=FRlcFFcVC+EE^=E1M8-EOTC~$U;yr@FhTh1IE_u^aD}h61=z3UN71Q<H
zIsSTjiSJ{e?X&($ieRtl{$hUX^<&`e%Ra2eJEsVSvZwB@j(gMYq&@5=@nHWkRKasC
zt=$z$EXJh@RUubKCS}w6AGB=r=Ve#Y8yLRl-qwoMK$=j6*Y!=P1*3d5`dR9WFMwDH
z&bg`eDh&S5mJa?h)R@q)|BU+7KMd<H%>n<Sj?SeA_k@;8FW8|6Hq<3*G|+l(O{E4*
zzlkgf&^o8+I}wYnEY^4KEEk(1gSazlp4kb!0xquvRn&KJL@p7)L2hX5)q|CXqkOSf
zPr*$0LdU@9!X{n(282uReFJUy_HA+^jsTFKoom&t@0|d(898{lR(v~!1!Ne_B%MBV
zy|h2Meiz<@NImaqPtUVIG&|mja!<=DPB!?ottIOd&bL}z$sR=%@4Ia5i*;<t6E7Al
zNp|9~)+doq3bgtLDT*#M;O{I6MTGgoNkj=n@x6h(E}ZFkGIZ85!3;Adpu?T)NjBfJ
zj-diTk=M#h%h$k#4?Mn>48&TC&gil4c+AAqIf@aA5}9N`r)!x`IWHj;KOc~0Z}W@@
z@Owk<wPe4MTQ(-=l8t^<C+j%Fp%Cg;B8L?+m)RoDH}g^R?96#q!2Hau)z!6?Z0szU
z9xZ)wzJm2hgxDbWx;zddqnsl1gK@fx!Jt<=;n(7_)S|tiIE+n!5+NIiZPCApv#kT#
z<ASe38WSs=_Y@~Rs)V4A12A38?5z&9(kTz?l#heJWr-3~(x=wyka4#EO7xUV1vZ=f
zdzy8al@7<2ns^4ta`xd4h2?ful_k#rS@MI(JE=>a0kB+Lled0t(>?f>Zv6K=TecPA
z{`sT;mpfTk7}3R<6d9p#?4cVQvBXC{Y#8aB#ds)6-fs<3K)afE;ANY|4NAQOA4E=7
z@k=_t`R*s4c=OS-&#G4zUbKP*X~a@K(iI)<>qo6rlJKL4=_@cgU{L(ExpsNgyLKtW
zAUk>?^)d^;<5n@}s02j{JWpm8!jaf71c}W>K=Ys5r6i@hG%1Rn?ZigPL<ea%(jq``
zc12gbPfx9hm)2%tpf~vonP#!$Ocqdijq>j9(d9(yy|)yJl3MEU(o45uUd4iSKR5#`
z_w<R6Y~^lCkXetTgzn`YkM;X?-A4~k;xaz0l-VyX1sf>KfDBb%92g{<>nDp;GQ6Mo
zXy^o$FdDf?u$k-HsA-sQQ?5#o@%k9)J@GI`mSmu{sDG@*N8e#wjSmpn(L`cdY4WSZ
zfAs^_r2~ymQ~HYdYRnuCK=muC1x5C4-BG|td&@IKv)|K}k+7%|K2U`_kCD61ehyQ4
zF%clWnj%&lX<41o=%QDHVLTn#K<!N_#4uVF=AjpwtIY~Nu6k`B$W`RA7m`urlg^%+
zDU`$X@dAhG(Zi4V=L3b-?0h@SJ2sq3525{!-k|Tm!H?!F^0aprF^f=hek|=W=09_D
z0=>a>H;KoG37x+cp;wi2BchA)r5AP_G%-n(4)>e?OfM}J!Gg7%Npper2E(?v>9>k(
z?K6UG4;~1;nBD@DNsGl5+~R1)e*5i^TI;<nkX!@`;?IdxkC|m_zY%165kW9rh}otK
z+jg7c%x6eFKf@W438LVW&NqU&f6D4TnrffUr1dA`M-S6i-|VXB-%X>xywiAnC*D|J
zy%R4j>?~|IcH*VR_TttGU9i8g&iLJyFtSmds%`xe6(WP;*5#)=8}aty##Un~7XNc~
zg`S$2m6BfRv6RB(0Kw~<3oDvfOywotS;^_yz$O<c1BjjHl)SUBvbr7AbaqEeEre}4
zi3-A{?VPBkiz3_cA_IxV#f|k_jjinkVXreAjCQ>3HYo!KD%wiQ3I#PyiItSe0Ag2V
z8>Mzzt7GYXa9d9y0)Ub<T)TzmW4J6T2wzxU-B_qO>gk1ut%+i2pt0n3-qj6y;YAI=
z!Q-!vm)lHA(OojAx_Of+L<cfwnXxrqTi6WhIXYp))`tfZ1fN=2r<ZNk7aQ^NN@I0t
zJ6NcGI@uP=;DNxi(2=(`R@I|AVZ*ntrXEZXEXK_pk|!RLxz<?N+KI0;s82Lr-(1;}
zV`-4C3&zbtaG-MG_QsZYV9KDeLN3A9t%cPf68beCTlaog5dYlj#?_^jtsqgKv^+LF
zVJZlK{ld!Cb@FE7mGz~@>p|e3^MIQ`AmZhfHF}+k9FiN2^$K!y4wtPj2MvUs(yyNx
zp9NLR*EJYC5O^Lx*j(6Iyr$|JB-ZEKcu_(4nfx>H))jRRspTNa9=|){Awle`#y5?1
z@`Odx1_`xSYuYk7KoHqjU86bccD%Z=wGdNHuhJ4=Z9}|nB}czadUGs@fZi#!t*RUz
z=&Y<Y);6~8(7SWXjaclQkzp@4wi;?oUSHY17B8<Zgi+Gx$81VrB7oMpt;WvHt@U_u
zW2q50wzf95;wv|oiCB<s^gf79#0LtZHwx%B)^EkP7PiPv;w{qKE$&e`ODL__^bUaa
zOH@S=P+T=jE^ThyIuS3d(y$fJ#!nx3D&kitMUhK7?}K&o7kukxBet4Qh6Npijdq*k
zgmO<Y5|#1s6);!UXG_CLW=z+VOD*Kg>>i0+*lB4e1zV&iOz1YO{>06pDXR%laQd^4
zf+G%0<NAayQ_1$4%j8B57;H8>i=tpok)Q#@)+ioK`Sqk6d)f^K1#bFSWA!RhPOM}C
z^se`sV|o*MKHO^PP0#>hTQMC2rF%i->Bm@vxn}2>ZO!X+q#2Y4bb#qvERKYZ;(3vD
zqDsMrLWd0$6X3TR(=p%q(W_yADG^|Iet9tJ$zyvn4IhqE?o(DK2Ba@f(m^cTyfPr@
zmih>2sbyuwBP|*TE$&NoRI&FAF;9nb*6O8NiIiO6vU=F0L+s+Ey2C-Y1!wHTtC^Du
zHk_<NkwpO5Ei`(=$%E|MnXE_WPLx&$WdV;37#XTG9D?SG(uW%>vZCH^jF`ajnrhv7
zkaL_}pJ`$3IQzP?J)xb9gL+Nqga&+kEMcIZO;BAGHCd>6BkGLA(nhop3R-ccdMjGs
zIlbVi+Ct#k@$zbWJgL`;PMFiziXaNw@M?AK09dXYos3f2bWont%+*CBqeA<fGmTWi
zoH&V~r`z;q2Bkvgvu>zg#UaBSsPiOB^~NEIm(~ngK>15eO5~Eej*o@+W*rSg3^ySE
zMV&?WSh(vvirIATQUmH=9_&)1+*4VBkHcbgR?uwhS$0r?JL3m*!%gq8Xnw7h)!4*<
z_D#|?JrAtD3HMC7ZUr4wU#vTux=m6*_w~tOKyNlY4g=9Q5X{Ed;svC)#^PSlCS9X8
zM+NKTyxHhxjDYCtX+}y{-jbQ*?@i7m1ni-nb}e2=dOS?_+Vm(gjoKr+l4XuMu1DJ9
z<s&_yhl;-*NlRWp`j%cs#^S-hIqLa(t?><50pXi7``YaImc@4J@dZ-A_lrq?kB&7%
zGHR6-LNV)lz6V1MxNpSji}~>lUYl~&j#4_n@%8i{DMeWY#pg%1-V|-~0@iQHrMA8_
zR^%(2$8~2Lnln?b=hV5(fccx&xUD9^`6<1gPsfr2{$G?=Y~l1WKlb(dBSWo-yBy8G
zKDoOS&28!rWAQe4yA^k7H8wxq^=5R77U+AO8-<n!qaAWsXIh3kTbjN&MS);oxXTRT
zbv3Iin2jg5>rCY-TEq~An%qjq%ek|!6;{yfQ*yvKPXBRkGmoSAop~tUW(jq;(pcNv
zxkH)4Yn!VJJB9o59#f?~*zhno2m>O50TEfJ?CQ<N;>yCRN-Ei|N$pyw2Ft`NvPtI?
zuvdSg!e0H}VRLfO?LB&UJbb%emoazjuPt6%*txp8c5+tLq@dJrAyL8f>&TI+HS%uI
zNLsn<aAlDzBcOY9vhr9pg-{Vw9}U55jqGJM8&)I&GYQnkQWu6qt%*CJiS)c^kS1D5
zG8`w#T{=n=u8XXYsg0GTMq_ik(YP_IJ~S$6BxJ8fe@<y*mRzls9nebJ)<If1NP4{{
z9nTI|i7H|0<J?U7EfR+`;ABKGgP{DD%Wzq{D)@6Vc{SfblL^__S7f|`q2n}}DLHWV
zEfJINh1?1le@nnhJ%`1sjkPsO3Y@L~OX<|qAsc}yQxK=k{L;*3e)yobTg}W>3DeBi
zt_hI(wkDBD8zHtPzm2?t=VfapVyf9%30S$UUAwb(WoBEG(gSTRWeVc7t$hR7+Vg6B
z3>i`%J$x>BFopKJp!}oR`WV$XB+mv#fcOpd+q}F(A{4R`D-#M7uhi~*fri1!0VZ3x
z6&e)N<v=0UAF?&UffVh_-e&eb=fgH+&BwB37=dnYxm%v|VH^^EW*NJLfbbo5mvcUx
zL*k<>XO<Jt-my10=fgN;Imj}$=>XZy{i^$|oDAr0A#rjJL}Za<TfabZ+uF^Xk3vH-
zVp*OR8=$(D?PSh}WJu;9OETjD^tP=a^L$9otVxRvQnmIgzs6K8Ha@E1yQf^MK|h*V
zt2tFSl>M6CY6{;oEhQR~@5g~;X~2>TXmqBmL{3KTV$aOdzdw3w_El&x*O`3)h7joW
zdbLP(F-v_pc~P8<Ltytpx9wPO;BuZ5kT?3wlcvoEw0P6qXPgV$nH?#Y5Ae0~Xe#Y{
z7AtsWKgt^V6(wMOi`i0~i<)QdT@0Zj##Qz>i?uu_#_n{mMRq>J&So*&nI|rb4Cr=+
zlc{%|oC{wIh9p|5hGc*jhyz}4GDB_+E7dj1HZVhS3JSQEMulEWN4j}(o^`>UR-T<7
z-mpjHc@hrEIw?}J%?&K%va|hEbfVdf(v!hDm-x6}`8K(p?flw@-}S`9i?w!|h=P4j
z8h5nTl7U)qHd!8v791%+;zFHpJ1HR~6o?{Fe4Pkvpm-Dk;uqwdAM2}YCuZDmAPOO;
zRa`MZ<KpV-!rDR&4k4y!$zTY~x(6ND7bJmP1`xPv{TMIPWd}PO8-<s#xLux@IJZT)
zUqwD)K<WJEwUwN0<*w;m1;X$E2_XE!YquNgH4Gx8TcQXMzl^`l-Wnk%VLh^d(1klI
z$KzLK@1!C6KoKDR;v0?4Yj?;ZX%nG(W=&xKuw(#%OPgB_aZjQqP<U69YiYzl-!@&~
z2S&CKMIq^Y8Fr|uq(ac$l=K9v$djp%3Xv5u_e8I|V`#NLNf+uXE~JbQcUONNxq`J`
zUIV>UNrvcln-!BGR0&ffvn4}F98y6@GKBn=ONLlq75q<d4q&xx2+byB15dH>ONI_p
zF#)j*JCC1)EIiB5Sa<%Wzh-&~RAo^YNfOl1At979!YKjJ@SD|^Nr57AMTac&0zp}Q
zFIiqd#nD8oH{^DOl%{a<a*qO(xn_}>amLLWJ!F%FWt9mAdP-Nb`Xd>uin_|>0U4#M
z{Kt&)j4fDESBzjqFX?;?<^%gJ@1mJ^c-VH{@-F+{oC_|4D_KzyB%H>HkQFt38G$5S
z^9xY(Dwkfpbw8o|(kongB@38!zk&{A#mWmfs(nb-v-mVZVPKdy@V8m*GG!F9m%uQ}
zr0S-*0Zo&>7ax*d6g?oDOrIHgDIoKYJ~x}?GWydMo8>{MNj_*tC`i})b5QRpnap2J
zhKCR6sn-gb%xMv`{^w`PRaHP_kRj_w^)ct5ZM-4tP~Snz_=Myzq<s9c!9c_rB5{~9
zFl<>0p1TeqXNXGSSf`<3Ez&AlwGv9L8*<T|W{68qZQjp-d4HPC8Wax_#Bv&W^f0}^
zA3L(~foZz0jFU+`YW7=$UQCJX{A)gHddd-~R?omFk^dA~cwh!!kPq95kA)#PPvJ0P
zp7e>ykWj#GbFz2XrPoSo3&@Z_z^l#v_&(iNLhq7{n*O)Dn3_E2tBHdM0#ECsrtMzy
zfZkx~Hu?9;7>UpOB)X6w@)A5KxFz4D6;DjB^oLXLU5~j1ka3`Yk^hA6qgu$oVY#h@
z`KnMzQ9K9I-knYoqJmMWKq?eJpsE6)tApu8R=+kQj9P{=QsRNqvU=xzO}xFGJ!oqw
z)we{?#gdM^i~<AzBYIJ>P)7lHkOYM8dd@it0Akv3y{rubV1XYd+gIdgwrJA%ER6AI
z*)g6v|M*Zmc0<>S())JebwL`nt$(EnhszYVzVehcX9jV1-2LXLXhsMEbvj{CDUlB;
zq=S8Ji${Lw_Lf8_p&yI8Lkq_FAUVVibEASvMJ^E_O3xU{`x9f;M3U`&8bb5N;@^B!
zqj(SnB7PIbqM%YyIWbj{r1N>O$keE;#e^;MU^qFE*HM}2qlc$M43aHs+M~GVt)e}0
zn4~xK$U^GCS|g>#rPkO=gmXK}yJl6>^bmcmuz|Tkxnc60z)GnjDJzvoHI0;GL%QCs
zUWHOt+Cq21LeB?{nM5=0Cj-|=&jlJOTiUeKYjSX^FfMs0k@C=aVGNZc6G{2d(AQ_`
zIvhBW!n4Prq`2Xn9%{|(PYSDVf*#Iy%B+wpmyMOWfjD~F7OyK?RPPTy6=I5y9)2#s
z<P?6ha~5np%s-G(U)R!u`~$u$*m-rMVnIvooGrzU$X}V?h|CHls<>5QTR&T>(wLli
z4ho7hxe7xI%ZF-F>^ZQdvfkh)ru)JxV5{lCTH-m-ifuCJyT+PYI>s_=wR;Mi*18t&
z-*3ftX@@%LR!|_1Od7<HZ5W}*-~i+E8}TlY6W+vVut$$q$E`s;pg;1i9Z#4fOe8VI
z2w7OM$VUihZ^^I5bd_Qfx9-NRd-G5%1|O@?HHZMsFG`xK>0^4Zf^Kx{3dek}%HqLm
z2$%z2g)helh_A~x5a{{auO_3xJamf>p9FF>NdUbyL$C8-H(QqyJx6I0$lc6`5*k%1
zuN=J1jp^NiDrC3-*<0oZyUo4371udlws-<$2DX|E(7m}V0uc+e1b^+utxckvw3es+
zJzuOTOe~F5n7$BJEaJHU-<Hw-g^nPvM~{fl&=seX_FldfQS2I2!e+gjL^p=$Y>%-X
z#Y?to_0W`|J?>;p+mGM6dv7Q$NIoy_-4w`5q@gdK1=-L{!*Pag^8wmBnW@dy>hW^s
zj7niebCWcw0NcHJ8NDgNvH9W%1>Z_MdDi4x+J0)Ad|@r>^;{F09`;hBH?+m;S*!Nr
zIotb8-kpyQkY3HqeJ+DL#_}@;fo3i!0J|^F%e-af?4i6~pA?X+Ie$r8M{RRvS=XCC
zC2LI(biI+a7P(S#qOZ9o`Ti%Lc-Ir+|E$|wGC9Rok$tA<D6UE8c{2p_-CPLf(Zdfg
zAs1D?+4%}ATVLP+M0(r3CR_<kccSh-dYE42gUeLU$L6y1!nVGRetYtuOV@bsk!v;n
z4CxmpcpizSyAuJQMJNNBwy{m`Z?4kyk=1cvCk#XCJY9=Nq(lRqD?7J0YIDK-t}ren
z7m3iNmR}PhBo~M*(IJEtdZM*v`(Qd|m`<c*AhRq}SK|}4>0n?xj7};W2%R&to+@`9
z%)9ciFeQNOMfRDF02Jsrczazy=}3tMDpy8>ZZZyQ6|+9TB&ge%r6Va7h%|Z>y`nso
zx?F;W6X2o+1D(ae{yya+)l^LNCQ@R7$}(L^Lyuy}F+QX(o0WrR3hFv$Y0Ah-B?F<$
z^v^XFgK0iNwwnxr!~&5B98JtBfkN<5W+00bKx0W9qoRz8n)(PDWX*a+saznlJigl=
z(tgz2b?tTrOVN^%&;g|gQe>K*U3C1`0Cnd963jCSC>rIGfzkCQJ=4-}#c8Ln#|p!a
zA*@h3Fk2RkWvZV0EJOfwU0o^3z-W!)#|NWkt38OtCCIyKQ=#$@lwHYyIZpv~iW~yK
z@ABSY_@G`3Rv!U`Ok~k&@71*K>MVpzU~zGeo(+uewfCm|W_<6MHyXxkJ#;yMLJ9~R
ze<uhW2Lk6MIjUc;DcnrYI@4SKAdk)z0m857UucOpGld{G$`u4ePRrkEysfT*7ejis
zh6kb45y|+JYAG#)R9{*BK@Nc}21r~mex0QQ;NfrHPQ-iYx}x`vf%rxEDem<q&4cP{
z2yz%1Nk9cOF+)$RtO@Lhpo!z}m?lz5poyhcvfrF`ClGe2-YP-f87B(Js3BV4BS`C<
zszX#z#6}uNCPWcPI$IE+`WPFaqJK>2WrmM3aV~N2j(#)4PoFP2xFk!^yz)^PRph^N
zN0CnK9VUBs<HJF3aFFzq_C!wX9t~;irx<JpN2}ns-~GfBZ$5hVS(RtSABE|Wt+WZf
zNqS(9!WbWz$~Ze6QVKVv2TUoGpeL+tzg=|LB8MRj;kcXs6mg)?2AAYfmUP@5C3LQ&
zO{vYIsf1VLQ(jg_l8s&6knKeV0?W--ld>We11UHrmrb=wJb5rII!luaBOchp@}HtI
z0lIasm=Rn)085wswMFuKEbnX>4!ZJ@T2BDh^RaC7X6Fmgovu^Or`^O#Q{oZSDd%`J
z7}8^{d(Cb_OOZkUZt@@=it|U09<C@b3<uAKWm;&mIOw<N&bnl4K+kJFX!LfIR*POz
z_eH*Dl1L5t26V}pP$Caiu)JY^a+eODQF~e|@41SEXhG}oRV~VD1C<;M+wX84_a0*%
zb!DK64|>hF>Ga@RaX0D5drca>Tklwd_8wzp-D*L#d%iu$`I>LG^xl1pRZXh`_3P^V
zY?>NqJthyZ(5xs&pn|z+rrOh}^W`~IuYj?tTGBZV!#?jqq;WYKT0Jg=@qsx&KVRZc
z<;AGJpMv`4KXIPbX&DK7Fgc`|Fa5Hh1#&DBgU3MgpHy%SAfffo#qBZ8XM1<Wxl`Kp
zY<HWx-6CHghbE~w_2oZBEGh^-Z+tSYE;;h4m4^k<3r3Yw&UBiBhBbPFJ5=9z`<7Aj
z0&F00hCWOqc9qrsg0J!nCI~%6AB>t0s;e7!Cb7Bz2;yEPare62UUhkRjvD15A_%-d
z0*$<Z%8JE1v`{RE3&KyQ!tYkEuDAEPkRVdD=Va8aF0iOQ0T9Fq_2v$Lb*V+_<#3ho
zr1RY{n19If;nQY%^zeh|kL!Oohwi6iJ+Q_xv+PQX(8%_&w2P(x5*g(*tCnQdX{!JH
zhFr6D&a40Rg8I*k>OZmiPgnit4fUTD^`D#SKa1)=-y;7roj&|so$;gZdGu`O|Dc;~
zzvt1zZ&Fng|8{u##uL+r-_UDL4$Uv#ci0@q$y-yoQc`6<MPNsr_shS(OaA?m{QIr)
z?>EW6-y#3LB>$d}e?KDsF3G=Nlz&gjzw7evv-0n&^6y>w_rvn<?@;xRI#0>}|Cs!H
zP5yna{QGJ7_uJ&(r{&*|%fF}P-+lRaS^jOyzaN!<59Qwv$-i&PzaNl)*W}+%%D=1f
z?}7aL8Tq#(|K5^+Ps+bf%D+ANcU%5_kNo?D{QFyy`Mc!*zaam<F8`jBe=p0w-z@)b
z%D>+!|K65=KPdl3@^2#lZpgnE<=^Mz-|v!t8}jeV^6ypow=e(xrj&k0{{MUO@3+gp
z-zfjSPyQ8;njUq;L!L(+@rdP7M?Bhi)Ddsm9d*QOV@L8v$D@w8zwk)jT6NSBH#i-2
z#AQK89dUKeQAgY`awN~7AIWpSM;&pt@~9(DwH?W`XGa}zs^>^1Sslq#sUsPWI_ij!
z%uz=y-H+sw`bc_MN3v@kbwpP<l0D(5BRay7><333(G89|q8A)>L?<}vh(2)C5nbS@
zBYMD5M|6Orj<ElujxhhDskHFJPHz6sKK$0SJ2poLW9sN*y0@zTc0|{iP#9$#4W)nc
z=;60+>~@m9Nuxg*J$UrTsP_m<@BAI=LY-fyRJ5bXqwjt6Q0H6;sAz3tsj(U_Zmch_
zT#Xl37q+(>+mAXIpMKX9Y2m*oZkGNAap?b%D*j=ENm_j8&ZbrBAKInfZ<bQ-BZ&DY
zb^$&9NR}Hz3U$$SC!_TyZH7X@e=Z7sqs%8u2}1p_WVCd1ZS4+}|5u{?H`4($b(cnp
zp`&$-^k`ve;Ywq5rLhb+|CYu1oLOy)i@@2qxxNH2zs|yZx&UT%VGCRAH)nuYy8<kJ
zhlS9$)OeJ#w(vTD{H~91nk{ObbyCS1V0>d`j4cP_y_GQp#3u0gRu-eES+=h(A;h;=
zhPbnFbp=>_7Yk9;EZZA&x!nSQd^ZbHR25RkIzW7Q76@$EkFgL%-DQ2_*2Yz8U10Lb
z$}m(J8u4~eC!b+qK3mZ12*x4vITl4P)O@|>cD%iKt+8}-wGl6GH5yR8r&&ZDgy#^+
z><aLnWAR|9T@xYsh0(#ZN7>0Z)V?T6!%(Z2zmf@-RNFA})u(tc5)tD=sM;4;DtaBj
z0d@zYrrI%Z(N3^vpEnzjr6=P4K<gD2NUt|JT51C&9+0CvYPYP}=`;%mvx~*6Iodmv
z+YvzP9E<aGLH$x%eVQ-<=^~4ymz!J#Cxg35UrfI02`|$*SYW}5swcyyajQK7QcEmO
zk#*4<qVYb3D`*cVWe0#)S-_%N_JrNXqvT#&q{so-bruXJGOH0KDZp4`F`y^g7$e%C
zBSO7_*sLRVfVauw6&VDbUna|HITpXoBEd*(F%!CPQD+1~w`PJ7eHviyurNgyCD^>>
z;CzY2fmzO~)o$Aw?qU|AXiTBSXP+$69#eK%AbqtU*X3G+DFq>8i3jRQSSauStZK<o
z6MNes3sO`u3DWEi53#ZCvOqB5acpQ)PsDk9awig}m_CbGq$)|RMSGNV(HN-eEfz{g
zU^uOhoA)RLd_ecL9MA@ND|U=(!b0kOX%12t=cw6>2k8PvfHqse^F9k&q(nLP;U#H+
z_*E8Br|WPuCnGxVB$5JXN1U34212nXMB7{k)~x+KEOgPVO?NzG8^~g#!JIyRFN>x_
za~x3#m2J+MSihgeDjM-6R=$~Og8e5fSlO6Jd-bQGen%`!u{AeawC5kcOrsG%J!GMZ
z{F1_U{LfjqqIOJ8MC2rl$;|<B|0Ro7RL>zjvq#q8LVZ6ARWz?91F?T2;egZ+ut-Jy
zs(sKGcYN%&DUi53-QTBdc|iO@7O}`&@mW#R!u|(=^&Yk7nZ_JGX?k<F+G89#fyw_Q
zN`og3P3IJ*y|(-0N{k0ndVLv?{uj1RC1`2tkhCyB`(If!eI{0RXqOvZ2qUwB#E9GM
zA56tY9OdDQK$fsJXXWs}u|i-Jc7>#r?@8}Yjd8&I&#<66)yKnJT~@2^d{*8uyVUu!
zEVj;&^I*S4Th+bP=Vnp=9E+-RQ$46+FG>vzVqw7I{&^M``i-kzF(s+96DlzRbo&=r
zM3{+PL=(-mfxpN?!f@z9s?;wr0oxPNUt-}(dVx$7NSg`j`<Gd?5(7;wbDVU^?x`a@
zNYH8fD=a8X>b@H54wkmO)1m$<3#~J}J(eV$r84Bd#v*^h9O#xPvP_597TAcxUl-+J
zoP!D7E*pt3wYZtvsp8*YiG+bUOX$DJf<X^)n9;^zIs~O0!D9XviwG9uASyUom$y`x
za~9SACkv|24#;M8KvQyQR;at+|G!yGnCTr%mFl_Q?&8ez|FTf9B6gteHM?!1D5eDm
z@9(g9#jPT|Tt#o(><#H~DYVMpWf37N;V=}43DJx>z<<C3f^*{ljwhp}*|P_%f6O8k
zw}ZS3j6ANdVmtgZ7OJ=%M6l~VU1djs36TZ|TKktQ9?YapC5?6+54&yZzzI1jeL9wa
z*#CybhLOs_&hMKO4rhPbi|-Fcbd?si^}lD4b&?C;Sq?^n=`fzq5k~SCMMkWv+}~hf
zOZs4YEK-mtIVOV#1^+gSS=?ZE6FME+ZTFL1vZlLO^>=-g23cnZtk^0i3B{(IF8B>B
zQZX~t{e+(<k6uDezlp^I$H{4KihGKb1vN<Mvju?qEi5Wb_71Awk$;~+Lq{G|w<c29
z$(3(*ZBQWfodO1CI<P%IN|e5sFzC|N%mDcTwx0UTxislg*yeaB+!2|)4B#JP!NL11
z0yooW0QaLT?pf1gD8LnNsL}^rbu*gKb1WEZ`2>p&-LU{YKVfdJ<UY+}>a0hpW_so!
znN0i1Xo2-|J}VHQ?SYY7I+8O?%jfKTo@E4W!(fU?GsVBAeNrO16oc-x*T;e4Sr!+%
zH-o#MR-T-)5ymuLVDZ4EW$>iq$t)DXj3O2q=0^rvm`|5RUw}H!LX~M#z{KL|lzS?a
zDExWU=F2QD3~mfXi7O9<NGD^G3=#EJ7Bz^iTysz+YL*INTKQQP7se=ttx%}pr^+ZO
zDE9&j3rj-=wh;0%3t6TC8!{Fh7pN|>xMe+6Ryl4>d%XwoZZc^e2h8QUV4eWX6&ABh
zrHUt=VCpe6SLXqVEH_w4n1q>zS1esUZq-CXZZ~V9orLyyokfFHgtw->f!Kwg_5kpU
zEFdfvy@2<V_QByKrdJc14*>MbEGX;=8PCS-_$RZb0D{R3rosjdTa=<X@Jxj*r!=h5
z-O?1<q?2jnp!DxTeGb_AfYuo>{X*CwMSrd3YKH}c;4cFx2EhCD?AR`l>#<m1{0x>R
zN6(i5+K@#n)0=vi#|avYSuj{dGt_$Rx5;OqmDZlSdw!2aEK^*I7A)j@T2+0(VwEW_
zeR|B<m;K!=8jO@oUDLoMt;?r-lj#T-sK1AWgh`!&OjM_!tp{;kx4Vk`eJm<;P6m}W
z7e?(pFqJ>Z0$nt{YX+#@r>zi*audI7l`B^0G>p(C9;kr-l$D?#(~#aUBSCbN=H4E~
zK53;v*+LZimQfTrQa}y<GZw$h0>nB{K-<mrM_Du&N*T6v`YM^I<GFM{PqRD0q4fJ$
z(6XUN1djU2eU;CGp2s5=5ONu^htWLa>VGXDU{r=hDpNdS8u!0tiGcDsB+n52-?M1o
zsBvg*#u~vy{zn!U*1a6AJXKQ){y($eu<^iw`wZ|;uy8P5aBw9S_@`J@7{oc$0u%hx
zEMjT3eN)ztuy8PCaRm8E;YV387{)oUqB-lwSjcjIgEZ_5_W0v0SSdeiOa#cEm>qJL
zqRDiKS)>^P<WI7YrM*$aQ&U%hPDW3~EP;d9PqDZ#cX6#V84Tkg5d7)cfzUPl85XE?
zd=M!HV&_4`_60kc;}^GESlypx!C{@uRhRBs>pqY-mr*>Kb|NTf0G0na78WLM4whyo
z5lL2$q)}dpp2sN_fWnd>;pbTirR`6TG>&C@@2J`BQr?iWz3>YxE;s~5%-iH=$wy8_
zT1-kGc>f}czi7HN9Da)fmb(LVo@UZ?%=njBG2q2=Vninv6AMuOWfrT{0ObT#s_$Q6
z!S$Vl(gEX$O7Mf>lr9_W4Mq>-8_Vsl$|yQ;|5X+qVt^dH4B9EH!~XJXv*D#s?ndw5
zXYs)K;`sUF`~OG)VtgMmf47?ZlZ7s&4*8<H|5N~g>jk3$S3XKd)KMZ{NO3FqFW5?g
z*H8pYyNO(G-@mSj%4ZDzI~KKE1!7_uk#1w!Wt7_wn3Me<SXhY0u_};|`7#D<`?pw3
z*j;2X#cZIPir=?<;$vh3&KBEcbC6<@+j__gSb4phl>q}dD}y5aBIQIT_R(hCBssw4
z%zIc|NIAYj)Ba(vIl4<bto#<oHwzHx1mKOb09iJNlYsGUEG1|t9+9%261CPA)nyjQ
z-@zioQfp~*qp-if->E7rkeK4W7}f__GLQ?;p^5M)*!4$P92gTh9J&r-)E9f^!YBan
z<18R7n_0jST_Z)8=Ft69BR5O_Q!FUVEF36}kUiRkA3HlCPtJ(vi;zCW;(-}+{A9>l
zo}O{yPtUN3FjaGidE_O{UGWTO&$IBbm0AKX`ie2~fDOFFqL-4FH-1n|TZDV2y{Wp_
zjA^}NEIf2^j(i3$uhtt5Xs66hojJ)OFPUR;IC3l;yir0GqNOf6)TdZErJY19!qX5V
zlYf7P1uY$51So@co<)OYE~g5C=JvNsELiy{GdMU9gNzG!g#|3Ft8_J8w8m^`vJ1Fq
z*JeZW1l@12Xt228+JTPI$NQt?E$El)ELQ0-I&Svq9!pw*(#@8D_Zo{=Y9kb(8c+hp
zWB_10ELiD`Hn#TxZnHS03ZqjlZc5`DELLehm4@L>!FZDe1*eHx^xla=PvR>A1e_+j
zH@N_jFXXnzGQzYxj!~4w<;5vPaaX9D_qESr+fyS$mV8F){OWQ^FqAfn7{FGhU5fE!
z*ldqzPbxiOVWu*4S!`H3ux%5m5tMd5+?fA>#k6}Js{wXo9<g|^YT`K-PajhjHo&TS
z&GB6>L+@=C5cU>$o{p3HAF)UfVdjy#1A0l=@vZ)277oLXYGHXVDeQ#bD}bOALW0tY
z4<uW?h+K4$RfOKbA7J@`f8&Q1Z`HQRH4{r?ia@>%UFQ$6$S^i$$fVK2%Ku@f@?e5R
zI68)5ANBqtEWPWd`&9^=F0|Dl7H)kE%J?rHlQe#x_K&gBE(O<3oRv(*!vS4=8;hHz
znYhP)#fk`T6%sMt8;oqH_`hLc^;-`rYN#!pF&esvC~gf0aZ?1spaK3n78v^B=E^2r
zr?cLO7q+h6Tx+cF#A^$ioAh8C0R4~vGVdY!sHxV4>14o19)4JqgZ>LsI}}uRMYe~D
z$dr~DV6nPb$ecd@xB$d};D<~qDuG?74EX;^QS80qT|p6QvF=hz{fMz1G;~REMOo)h
zQ%U3cCY?@Wm(llWrd(yw9}z-LrD?@Gg_jm~7PcEZF<j*&F@Ds>(BV%9<3{6-ED!8|
z%r5SwAd0W7Z|^LuFXp#NNeVx1V`xW;rLeuYvDE+$KVg^G0Uf9OlFF%t(myFmg9T~t
z*H>INBV-WDYW$P{a-x7bDb))m9;-AGnI+Uuvru|Zu?#9sKY(if3=64!b1$Sh^Ga$z
z%i?HP$cvNy0KsmF_j4>>sn%k#;EiLb;?D~x?7BK=?&YPHL{eiSEro6!1jt`-Axo`e
zpH4bZ;uXD$WA@cwbdj`A>Ls^Fr#a{l4iETCE?}wF=={G!?3ZW5F<l<nkiX*M=o9tc
z3ig`WCa}c#RTo3Mh+YhF*96yxe$B<w^JoYb+ll_Zi&@$!1;2fokdx7n=4CXBf8^q6
zZ@rMC_|YAJ{;7+m{lY>tr;q;&7pQcg5OEnW_<!w!=!JuqjXG5aF#pblDeX7v(2+dE
zB&TZt{0}amUdnoTi3><*gWB3TmL~gKE?ntg$VFo$*b^UjMlih`@p3!d*QQF@DQ=WU
z;=bF({f^YB(YRVi;=<I<H_`WoPvVR#o_lfu<6DHxQjcf}XJje;eWE1H$c{sh7ISA}
z3Gkf)z?q31uRw!Xqlbj}01M$L!h+B(0gCt#3#Cs<x0C|Lk%lsRqondtw!G6Jb@>z!
zW?G5!2^Pnz6>El-5T6zhIK%3Ity2&G%qpQi>q2>3*_i+(!slHCXSmZ!@y(<X=vf!Y
zs}*YomE~WkR^FUBW%<Z0|8c8#X4YZMn6m6~x2!YkY9%l;rNnsI#c*a_27@*BS6wV`
zr}4~;lG9li$5~mJoLcsbC{ZrBD9-Z6L~;7wWf#EPam^V}N?&vfJF7b_9cMmSeAzAT
zwQt{yCxKR6AZI4lJouST!mPS5-a&|)i6qE|3*vZCnoG&dCh@jhJjecNA9B(bw_{LD
zt%dWAxEQ>1i!hAYPH%FipN=D@e6%Q^*K$si>VaS4-Vxkj9&6a$x&pMC>rD4u3i*-{
z;>>mR1(^{JDJW(IU8_?g?rgJ^wad!VS%0%M3|(*!3q~m>5n^yUTdz|ex>=3_!6N&?
zzO_<{J9Ncets^d`C?XwHNP77?SjxHU%DEIGC%;O;d<8*Xs{?l37x=bkr}G;N8z<9C
zdW;(`(v;la;;@6P*<i0;`jK9Px)eyh1PXPSaB}pkz=d@vl5;OR|0yeTpA)ha(iBTH
zOf6F8mN+i6JQ!_HDJz7o5K3C&2f6|W>he_~2F95nD=w2kPaa!$ugXe&kC5gJIBWB$
z6?IZN?bo%;`S%Ke*h}A|q-iaPhaEb9fIc%u@eikj<Y8)&)t$a{0cHM>fHniD?@vpa
z&tard94X`v+XX(77LXej<ETl2YP^0-JU7w#qsUF)Jna;UM7(-U+|a0L%HrQAq(TGh
zC1wGFjCRZ40qc>#LTy@g*d+tQI*7+IIzLcYhrCd9g~~%6Wbq$#>Y#nxv<@nZeT0(5
z109_oLdn_*;>tx!iMUI51o9=w4?FU;pOE6{d)`<c;)#&XpFkdZWnNT1YV;)^j+Vsy
zQw}ZdBBrpy2~u@J&4}kpI)4TkX%E>`)Y?Lj2!B?{H+z7NKlw1OEpv5|Sbxr`i{9`I
z1Iq>+ZLkvc&kIyESe^A62&-BaK>tE*P}L{^`WIbL%SP<Rj8L0Vw&4l5f62wwvu&UT
zV)t}~)a_q(k@caIU}QX42gLu13$J&A1L4JF+oVf7+x_8`u3-`fM|mCnRaew&I5d~(
zp3Vrk!U`()*IbFKbCAea@2|VUbb5K9Ch4+Cx_DDwr>t%X$o4O#CHxInfOb>@1&G!B
zFrkcoad9GL!r`smQpn$Qh3E~MKq2NDTSERV7gDcnbI8J>UA*|z`M(Ipa30)2kcF@Y
zREhNe2tww7s?S1|(4nEAV6KqRe_KF<8=zN2GhtgtfTf84Pl#}wJMDkZCPurFdT5Y>
z{+<woqk6r{;2<w${KI+6z%gIS_$RK6WoH!NjfZV;cM6bV{<$k=qm~%|a3JOVD_7n^
zEqP|gl@k7~E8#{h31a+l6QQN7Uw37>^H7yW60)>eXScFRQNQVm(u=X#dXEs5Zcv~D
zD8>DbD^ABJb8+(ji|IggJ4@aX58QR${RtTKEPt5hOQWDQ)XDPiq4F@)>C_umG!1a^
z!p)rx^(2{<10?aj*~ZiRashZEu|*Fs671V-Fr8h>gRO3CEyO#NqC`nD#E_or*|zhP
zB<y$CusR)phrPDECRTH6j_%(ta9}LgyPD-d*1#=!eNf;z1GnoaOLS!m)F3&z^gSOl
z>L+nO!s5DatrwTf$Xygm=#R6|u4n6o_8g&?;Gbf_bpn8d97mRkgFaexqp!?|2CI6K
zg>{`6FK5c6ld*b0eTqe`K-L+Hqz;~8VRhDDfDV{9WhC(PEO3RE%M{P3_$3zHT}gUX
zte$CMkdLv*74#w9TL$<f3tXY|<yQ%UU7TWZFF4brR~7l1hPRM2LI8SU6<RSzhi8YR
ziJW(l-33bkZ}kbI^GkCQ;I)=3t^gfHm!yh^k6brCqiNK{Pn9QM*`Y<TcuxV0=bDS}
zdZxiGAXeUT;4)H+WkcU^#pq3qN@`L}*ag0nv+l}qm)n)9!>{e7u-9B+HFP%|&~3_e
z=#TDB$rjyG%uX#a`ufq-P4lbnwkt+&+XXaHD+w0#hAXJp>w+1Vci^R>-gJR0xFD>z
zE^)u&;_BVs0L^$^bBVp@V!M$Eal1Gj3ZOTs#^PN%^zg;|)13o?DU3|85SF_xkvf7<
zab$wWy0*MSG`BLAYvu-pgzgDw@MhhRK?Q6l5G;iZU4^=xvWgTZ)F#D@T`_KNsUn66
zzDhy&TtRvlEjPYug7_g#N_*f+yW=<#RirV|QeL6I+m-xAJ;^klro(9-4NkAA$_7|j
z>EGju-k+Oj{*EDa1{i7lK9`0YO{t=LGnyhR`UhPJ?}LhJ*WklqOgW;jjWNaA|C2kW
z`%UAfuiv33&pUqv3+X(n)Kv3A5nY8YuBFlvpAr(k4+~pccy8f{uAp^59*IiAC}ZuF
z`3BNsg2E3V1uM2-7mYl_orL;9hlmwBVcSwG<eFgfL&(OO&DiqMj(-%Em_O{$vc|_?
z*gOguiz7|I{ErJMPNZQQtSt(cM2c5AGuhxFMaolD>CU^X+@BO;aj9bUp>Ua*D5%)~
z?n<!iOFiOrf^NdYiTb^wMSp~f8=cB>eC(oW0-4zFr+1xb>OZlayaf}h1pQIL2~6H{
zdc$!ebq9K^N-!1pF(J!oSZgA$Adh=MOIGT~SwYsSp@N|7wE#I5={A+3eu5Qc?QB&L
zb=YpT=-iQc4W5+qldK#o$Xr2=*R_<=eo9EgzHk%!Uj<R>Q@w}~XP0PZmrK3;c4jH{
zr(LO=GfUO)-^nB`kHDXC1>T-nAR37}Q{|ESv#wk#hF@L1`*f0vt|z9eImK;UV52`b
zPjT6BnqD#k4>-wY{CQX0>~<;~k@VsKd@D+d`h|Il5)b?NMg5{H%9{Ktv>Cf#lr{P#
zSBw?bsvw3|?BW7IFos`t1z65<1p%p(ERrUA;@M3Z1SFT4;{W~j!Eo|2i2fB<v^B$4
z5S^ZH&%8jXy<c@DSlcH`0zK>_v!<fGZhNTzhvxceG1>lILNX>>YmdDECEA^~yDdEl
zP6qe$SCu62KM)jsxvi<MTLizjy0UpoPUt%h_CFTbFqvCh-W7l&I)Wv7L_S7K`u|Lb
zaOQApW3`Hy^joPYxLiky`<JXZw&Uq$gf@*Vbw%9!m{QWeVI^4$^C}ffC8gss7}vjN
zWwA~}SgqvH1<r}A({BhlIE}MTP?!+aXH-wmmY-72Z@Y48D2|`Wq@Z_w(y0(Tf`nDb
z8Xl#DZ*V2lP?>tShckh15^~;4<0s`^)BXQPlSkkC=%E$<uVQVmK_YANEt~{)E(y~X
z-sdkxyl>7Ta2H03_)bm)yJQHf1>;{!IUks_9C7_D-P}X>p<!3~5GUu$oYNUrA1g@K
zZSE#rmBuJf{?M4v%u>B&ML#MC;Fw?~bfDSkf$Q7EtoTVPY|M)KU==@Y`6oh&^%<!N
z>p~NK6N<d`GNR|Cs5<RrUz2!Rka!A7=s1y8BwA=SuegPu5mI4_)Vtu`@{|psosAZG
zIPv%(^Qa`B=LMuQCF*PoKWzS%61~%zhqEN+ODv|2YWXqg1)WKIKf8ceVjW|#bb`Mh
zYhOOfLn&$bQ%@4_B#Wmr8vS@NtxYC$!ksRh$gb6pfTsi?n&i`IkMN_>hDLr$MZ%nM
zVLo39Bd%ry>pJhEe9%D|O!J2!KQ3(YGL<qqh7RlG^hF72l1|Y@ArLI`8CG6aI+8~r
zy7_Z}lEJcoajZ}uv-J?6^Du6kNVpYO4LStr!!lJVF;-m+N0pSjl;1XdTBvf9N)@S6
zmJrHti*5^nph|}`mX1u$Qu-~I^LN|LAufs4HnT4%zk!tXc$z6n6cCKB+6CWq2tUaQ
z(`A9Kl>8N!?swS;+2aArf@=%8Is<M`2*V-=4oiKqCaa+;_5YLiCQy<c#ksI#Eu(!!
zLJ|@hw2q{q28|@pf&{IZ9;u;wrs-~tB*aiucUN~;P1mBTu8{_eF<w~g!fcJf#s*`s
zdA70fju#BZ$2>582Jbe;HpaW3d1m>|GylWD`y%pY-pq`=_vWoC{m1A0I!E``bl&_T
za*K?NjEr>nV`@#w--Uv|nh8MHAP(CuhnEGuXCA$5j|wiuID+?Pk~7%xnfjLGSfABT
zOPk`@9f<QkCiCkvUuk|LaxSF;zsxPL-Yu~{XRhaUpW{s5CW<R$W~;06Rp?@nb{n^A
z9Bb4cb5LF>xxGfzw}IgsJ|)O^Nf04-EM7W&ijLNAv>KZ=+yQPTD=yZnB~}hY(4WaI
zg#w}_rhEqZ9tm<qW~uBJ<lL{5xrL6=dd!3(rx3cv3b8@$`9=U(5==YqrS0iu=2m)e
zF$rycz=zu=ejKT$H*-oL@|C~3l9<c4%3|s*g)+E~vmpcjpadR;VY^)z_lG5}kaeaX
zTucq|F66KpLdd!{)x4YKJw(K*Y%ztXTtNxKRNpK4)aNlx#U>a~V)&kE{9U&VL;R$K
z5F$5$u&asl{<O?1)Sz|+<FCDrt9_FDU}}Cn{4AmPW-2iffujNpWWoGG2*#C%yhhfg
zKSIs+0c5h8&6S7zd1XIci0fqfHKOD+oIB*Le9kn*6n}#<yBg?Ow!qMoE+qCo;<D%V
z!OdFB+|@ZpUw_p7+U<k%b!FDi@^MiVxBr-2Fm=^h-74KkVI&X}{ar#3%AnhYQWP}4
zneaLLCxhAD`b@L0)|VSKv&v<FKL`S3(YL!WXa8f$?ly;C%c=Hz1=aGOBZt*x{95ky
zE3;Dzi@F&(%V&s)(_!3^j>$>o3XXJ!_-o0hUq+2#myz>+PUdyTMcZzcSt{f|xP^4b
z1%F;S2!UqWk~96YD5fh7tbD<OGRVJ5X5HRkqu52|-2WzXyX$~<=UB&9s@JKsUv^9D
zE~#{m%0Xnc)oRZA6;Wna9}&5OC1<FwO5WXWBC+i9Gsf2>#*=O7QsHTV?)WgHXeZUc
zH{BX=y9bsMX-hZh^OK(Hi~+ZM@aL5UyL)h^Q<2FX1#b7S@&zl2K~5(Scf7jY!$z^I
zhjX7LbGwZVgG*T3I!whr&n>ogyHQiE$XZ#vZ+}SBVyFv6Ww^$nLNrIqV#teS;k4U+
zB!?A8!(z}&C1_bj?N(sS%O$2Ob5VG+Y`KLhekXEd%3S)I>e?(?ILp;U#F4qKN?~$^
zmn@8Non+LV9xV{LU}1<GB!u23G>xLT2On9>6-L@EkusT)ixr00BO%;LBrT8KXq<Pi
z%<C?%`<?Da!`$ZdG-|eIB9GN<9xeX9vdAwJK%4X3LA0E<bIp;8l+GFy7L4<3$@7hl
zH1vQ5T|~>NS7Au+H&SJtp`Ih5+?ARKW!E1A+$#axPJg_!-q2b@i)!xxa#*$JDjM{4
z<zYX|3=wfE&aHEk%h5}j+N|VH3)XHDW?V&t(f`uCE)uWvF0WcL>Tlhg6%$6?J2}Zv
zi^PcS+O6eH5!k`$>HUk<>4%mUri}W<h=+Yd-3c-f$&qb7DYH|`hY853UcK6kKyo6s
zJFqVzSa(j-c0Gg3hFFeSpKkJ>N;UryWVU*V9w@<#9L^@WML~1GAS<F^+6$o&hr}l?
z;;A-pHj+Z>+n%nBy(UX1x`$mlo5_k3rJvy%5>E6}hSQwuf}TLN)pBb~bOa|Sb29IB
z1cvE~`qHLDO;uUW;Xa%;*VoDOc*|u#&q;x7OdYPn9>8OQ^fMEYmWyFEljqaSY`v6F
z-7cu7seD)}p{`*@dbvc>GRPpA`xnx(ok*?oO1E-EH?i`m(WEZUfzY<&h%wZ65gWcy
zL7Oc@j8xLQB6N8e=+y+|jQN}W`nK6hQ39us8Sr}uP%N1B9$o=!GDt9EzAlI<1ZM`L
ziCA%ha5L;1g0NbtEMk?FNk2#><GwkF8_TYJot4v6-yej%GqSnhOU75#U~!~Q2YQ|J
z2MFCABC$2W-c7-p_#p|ZcT0<^*Ab0et?!gzLNuZa9NViTby~hD8hr0YZfhbDlF7;G
zoylT=A1Ctm#C=<!=2PN0m5^MLJx6D-#FZ$F`4c{-kk~F;7=GsFD&032=zRp_4C4as
zY^1<YX3+N&s2Ixn)Un-wR<Rx6UR_6~u%i_)_|FHyLtQQloaCZ2s2TT{g1ADF$*xy5
zDI+qdVEI>q&>_*5#j?|>ncZI}On2Dbo8PaT3Qo^1vp9k!TC)8iNn(Eq5^yC#JVpwl
zqHl8ZCL}J$M<fyLPZrZrD466*dgYS3-_6)?n8I&M3fft+3<bW&o6kFhFOg$f9~+OB
zfAt+M22)<_>dGv)T%VA%^udv+a+M<Xdy?1-O7sO>ThQv_N+xrupI@@-S}na^jjQta
zCB=e4KF9aoFV@!Tc2{@A-5(LCBkr`rUq+RMF|!2UpAx(feEae%DuYCh2>f)A%snHK
z2@Au&3{of<m>C5LsZrd*pA8bxp2g)FL=#zSU`LJa>vNj@Z-bOV8jvM`^NQo3{{x}B
zqWDt3El4>VP^4d-^C8kL{slQL-9nqSP0>xR%5*UQ=ZUh@owW~G9)whunVByHD@kvi
z=K>l&ok@gIzZ^srOBjQYmNYQP81XBkBjV0KL<z?Is^)@RM*QmNh>vx8IBH;@6=TG&
z1rddk7Am#tfy-#$2%<&R;j*~qkP%P3G9;*k>KEisU06kOW3vMfFt<maOu+6~h$=P0
zy1`S|85}z#lvo3V7)<_Vmfu|9GbET$mqK7wjYhYfbJ$DzfdgK+*wGuzXG;>=*E@Ic
z1W2sb>)n3Wx-yx|aGoR}*4+7J=yr`KG9z9%He!3zMZ8!d3R#Govg(cn=q{3lmm#Mm
z3$+#5RGrD?t+6=s6-3rvV+m23W<L-lzmpIgQ7Xn)03slxn4hZ&O(+6FIEN6$gv@}~
z1p$R{4gs3KIl`GyZwR6agM&e+9ME-F*@V8POq|^g&i3n(%doqHuu=87-h(GLTw)$~
z5%&<HD^`T!7UD9w6G$D4h6Q7_YBqWuj`vD@VKENj+h#s4`fU<O-@>JPvL^go>h*Wv
zd#l$AbE!GTZp7^Al<jUeo2c3}Tcy|E>fwYPLM-!z!-K80+cgIa^<3iEw+ITkVgdFC
zUAh<^b(yT^oU5A|b01;a!%_Rx>N-e+5A?j@%=r)a`Sp%?DE~}5v6r}2G(#B9fYxUv
z@<F6_#4ODcoEXt!#<c1w<;yCqGX@#>P!L!c2g1O7>_cKMjDsnaB<AyhaWJKt#F!5S
zF~uw&F1xEJG2SCVJYmDhz{^r>82ZHo?Mk4?{$!{*!>%Ur#D;~**{jT$PL3@~FxpC^
z;yAFY5;m_ZEq3cG1KdE;I95+7IDqQ}=nNyV3bfb^;#F3AHN*_h=&Ma4AO_am{sJnJ
zNT5@+f&|2BxsU)|x=rV+>43U%Fyy|`m6e{?=faib-VVmNeTge<359Iqcp6<ORc%uk
zGCm#dwC^!V<NnfhPtjj>bR7f>(8eBz&~ZsfEWis3V?J=8zg0q)@1(iI>J^fL*atMc
z^(sr9&gP66u@0}_K}gPc$`d0lkkU3CxbG%7F?xgsm&sc?H94@ak>xFcNeBC@^~dWb
zs1Wz&@0FNxTGuz!r+3k(3(R@ad|nR}OJS08MF_#uF&OAg#Ez|_i2Vs%yM<LAM5E12
zWk&xMM#d@H+4mTH3*kBqER5#Sz=7oj^ZquXAzEF^2F+?w0i@2uWZoVmBMfx$#nDze
zdxwN6y(36T3Zm!&g{`Md>s>)wVoM;tRLVFXnBSP(dxGS~C>nIBqx~*ow>frBruE(+
zEwNIHFB&(PmJCedCy9is2Z@b<czm_vs!u{B`O>3AO!22B#U(LAMR$xUMdwEUqniY1
ze^u&HO!Q|bPSoioMjktL`wNoj!SPC(95KPBrK)L%+%HRVFB*d!R+zX{EL^wF5``=H
zSDy&t;X0PYKPZWh(}`8dO(lC<z+Bu9ONu(EMY<e~@5H!meiL8IlKqyXBYfsq8&5>B
z?hxnE-1^^<*_vC04b9HzO{tv!cZj$>r3xF0gRp>Afj5IP%>N)5X9g9zM;1<C!eq!#
z5u}(fh50NKRv6_N^$&xn!X}W38k!Cn_)mhsLf6Rzo;i5HY1lsx;t5~IOuQT)Am;Zo
zgzruYVlkTyPTL6DmgKPg*Alh_&kW0#iUq+xC&9)3ayHK^oAp}Tu~TwI{JjL#CpyhA
z&jv-b*%br}Q|l+Qf5E{2B!R`Wo(-%*4Pt{u*8~=V<|F~*|4)f8mQZEz*@)yc=Kqou
zgxNBaf}@8p!(Wte!l|ogSxJm>@IjJv|Bl?2q!UhEPEJqMF~ENid0W&;5oHsHQy2M<
zut%3%7&7AjAw);&2^)yZRS;5Y@(liU0vGa6EEr-47}dp{eVEJ%J40lIsU?Ptv=%Xi
zlO{(&+BuoRsgt81t)5Kb^dN<iTH7>+9%@YBEF$2F<B{4x>c$Lzu7sDesJQpzd`g)8
z3nac+oyByaYBd<X+o`IHBrUOAkE0ci%gz;ci6nB!SG41Z*w;)X>lYP_+$q#kBt5Yc
z9Yb#omSG$#uWllGvqRvil7O5oB6`3|rv>w5A7KuC3Qv<1^l8d;B5hTjZRCqcvR{jg
zmSn#&P|uvn6Y89I7ZJ6Ey0Gz~)>Ljw_%88Tgty#t!igu)e@PS!2E2&?9hoktyeveD
zv2Jt#Myn8-JFEqRgRRD9zoMK}Noj?xnyw^CT@@-+Nqd20Rg%~t$VBmKjlpgsw(MFH
zgEbp&9(I~~4`B`K`@yWf4gEpfE;?UZUA69^VM@1)>U_X&-Z4m-8!3#cz(nsNqGIY4
z%f)dB^IWF~7!$oGNOW<MrR@ePcq&Nt@FdB~>l?Yi`-7B^Oqw!=L#w-0RUW)fjo%k!
z;Kh?<fFlFKXh$LJjzX+9Ob9Ekg69P}czBW=ByH}7t6<`DLE?{0lK55+*9^d(o?L<B
z)c$;s{6uRX!F0XrKOGJhgB(n}119FNKnFY=WZ=O`wm_@aKbnNscFO*6ko58-N$b1m
zaL*|%L*YQzs;6E<+y!4mT)5NQL<QbTXA}C2nNpc%Rnio;<#8l?mCh$@rok4O=&~d#
zOl>w%KB1BBGSz7YUjv3Ug9{(aKz?uTW|#)CV$a>VszAaB9SJ<Pcd1u{Gnknx8E%u{
z9DPS@7KIqI@hWuVCz(>&<fhsz)n1rNhp31dSy+Y&DPa)eHuH9EGYx>o_`M*$*d8pz
zr&4To()-{Td_drCFBgX3Lgdt*L-^Ge?s90XnA0Il=W$6#IJ}mi6C&|4N#dfQ%+SK-
z5cS+wAeYs1g-@hns>NqHU<q5}-k{y`HPStV^qnq}u-@e&d82^=UquD7N5gcmIF$_M
zican^jo11#gb_NQ?G{FzQwu`^^!@iThOy0fc)fZ9RqPwwiWR;rl0z5Ft6Ro=vtMpu
zCND&_i_VzeA1eA);WpzdIz35xJ{9%{-NK3mnqte&UYW12ZF%C4p?;Y7am1gni{zsr
zCLCS1!I1+Tw_1k(5rTI`M@A(IZWn9Eu$bD1xrpu=*Zqf4-z}KFtGEWa3S3ady952>
zlHm*)RKRcrIwQ9oR5+tFGxt9+NfPyLhjbH<#QP)(VXMn%R<DLo3Tj`cj@~cPGkP!^
z$GsjrO+EM*0AWegj49X7PB%-ai=S~TC3K=*759hKaZT=KbqjGyt2b9~eun0fN&GT#
zX-h`ox}|C$Ok|TxJR10B2l!Wg;EWOD09V0&TY2WJbO%)mz*jp~Lo8G4N;+oygFbr3
z9O0nvpFS}4!0b|W-_-2vy;J+1Z-S;U<cEF8480==Ilpk|zJ=+<MT70s#Yabho>{Ce
zPTe;>H}(AKh3edVb^g%el5D?^`_LJJ!{NGr6n^7t4LFc1tv!vS&o%M8K4yk7%E8?1
ztgoj^BjbG1$BE$$!E9y@&K$h2I<;`$V)fwE+_Y1be=ss8YH@LCYH3<x{;`iKoa6&z
z;L!Zi%-qZifhu_Xvr+IC_uW6e|AE;M-e3558B>|l9*gMZA+Wz11#AHu`rf?AE&MkG
z=FE#3Q;fr!g;pW4i8x?`He{94(0@nJVl5yx;{q*X!`uF7%<(@4Va1+y0qjQCxlV_X
z|0RelHi-+6#Q`zK{(KNSMpWCC0pW`4qj;ims`?8-bg{8iz&RExtxoSa;jbL^8uQCR
z3T654Z(ygDPN$^7^0;=s5=1ZCRo40*>{o-ZViT#LTDqI$(5E&%p(eODy)gS<3(|?1
zeeA{!uBsXT8xz6jZlk>nobGet(?TP)tRSHCjY0U6gYad?l{QnkLS{!0yKL2~#(k_;
z>vq!{y}0Vn2;#@+Dt6O4y9kVXb`ZDhh^0-s0mA=zCIF^zUXa3UcmdI)Qy|<y>b;81
zIo+V6_NibmS#NA5x97N44W@TtkY0?=XwxH`9Ib<@&~;aFl-KE}7Y9iQSBZl0Oa+0W
zL!wkiPcjP^F3F`qGGZ6Ch)kbC%kiy~FqI8jZDRytB9{k=WUS{in};lNwQk*95y8+`
z2BFIiXM$cCtRfsw%lIQP{8d5tvQq)Ws}+LN#@7bX%MJ{95>v=z1jdDXQhVK@uq#NR
z?2zWGdK}!Fg1BWB6P%LT>UGS?ZLaZ35V>rFt_*tE&fTi6!7%ACesd6BxFr<ycW6}o
zdaZ|5I*ngzrOrjTHAtiE(4|g8d+(MR#x2UTf@I3}O|=`?uC=^n)ty1~vVC&}t9V@T
zJLtiC!!kj0HQgPgP_`)&gN)kGu>)W*jpqhwl<iH}qUH@0+i1n``v~4Og_XtbcDiBg
z0}{Jz?f0PL(t^daw=wH8627dm2t5(#N}s@a#DKCmD~Sk?qk^GE*HYnNhBv)XJkvG;
zQ+OK8AxSU8cAh!Qpu2Lh25k&LG<#q=TaZ*T{4lbq(AJ6GRZ{Uh6x}ez2PDPPQ#4Uj
zYOy-v>8cwKNeX2JR!V@hdp9TxcezI-jnbo?h_Lb<D#VK=feb@$X2Tf#j55?FLt+cp
zz3}=WP))0n#9T>DHi<W3f=pDQV4A@BZo{ujYNhp9bz<F6b0*xT#LsZd%54(^UR}Yh
zmQ<Xh@MK<hXT+AIHDX!%$rVg)U3DdXS#eLdfz}_VD0JR@+n3m7d9S8w354dr)nt!J
z{IUZaqLCtQz+|6Na@zX1q)~cKVU<VO*(h+ev;D+U<Xa_?vhyNMh{?P~>msKNuaNL%
z>q%{CG#mP8AIU3+#&<{>W#tsYUO#Sv$f&Y7B)(gcD61J($wm->evO1KyP)8C$l>>U
zC1{4HL#CLhcAzXv<nNQnWg8T$XSkRsYziK^*-4MBaG!jmq*1m{R%zx@PSo%hcjeBu
zgfBg=_!!a}{0^=4gwq0Vl|)JlASSX*bDOjR{h-93DLJiS!>vE4S0VV@Xa-CJYqV!b
zM4giSu%wjX+mhXvTD9ZLb+VqQt?(qCnb|YtR{s%6r*x}xjrGu{<Qn_k625eU<O$Lr
zm-uBZW?c714B;eBrke2+l33X;ym1^iq~nY$ma5e3Y60#v{QD%447ZldhDVD#knfj}
z8N2S8kZ^UZgY7z`2n?7G`p--B(mlvQclDuPlJI49FKu;Y2QRYMQ23TstzGlCPk%+y
zDyy|p_?8ul=hkl_N*I<$+0=erQY&i*=b3_}xqTN5kLwRf@Um*2U2RT_e?(%J9S7JL
zQ)kAsu^QKfbsQqUJrN@6*jOf!k4YkBXKa>Ea;Er%L@v95qNMg#ci_~^?@8dYnv3H(
z!@cGAC1~k-ps+>KR{2K~xvZL4#gQ8dk(au1xf0?}CqSat;hWsENPJq7C~Nr-&c1&s
zam(7DAv!50mD5K)D`Cru<b+*1VIVZczm@Q%=aV)REVSu5!b^LbM*kqu%SsnM7*bp5
zkwb4X_Pv$rKT9HIYbdFBsv#kL|5d`5wROSj54S9e@=h^9p`W;&{@)~pvXcib8>`qN
z<R!7wf4(Ha%Nn^7>y0w3(n)-Wz`siZ<=c4B-%t_#(^2aX`435??DDZoG{{siFj4AQ
zUj82mJx2Et1MKS(zO>L(Ba5|J*lW6tHFLo*m*9k}gw?;a;!Il#{T78zf-c)8H%RJE
zFmq!ja9ZLdN$k3eWmAhSipii&4=dbC&Y3#^V>xGLxVz(ADcL;@gy%ucNM{g3whv}j
z@AI+LS=Igd2P~Py*+jx|)H)bf<Rrd91`T<#lBu0X)P(O=R*RM@RmD5q$n-7@(#zTx
zDMwEpYh;oa2T5ie>JptGx8bOW)O-2%Z&)>}oi*Kmne?SW(itbJN|H`?7S)yGM#eMQ
z%Y$UIRwU&r**{Ii1g|85?pEN=%udT|k96+7k4EU#0I)`A#?c8n`K?YoRrDIebBJNE
z>xdiwG=>O+lx)4FkU+P$X&LH<5LA{=X9}e{&60aMivf2NpfkF&R#zElrsRUxJp}DG
zLY4s_#inBv+1&0G+-4bJauBmQzD;s`rQdDbQ!U0c*rr}{2l89JMobM}UQeJi#IuP7
zTcC@@mJgDKj-ey+ht^2nvQNn~>~loL3A1Y~ZXlc%lem{i2+1v``)D#W_c^i5n96jJ
ziin&TO~phTaU<6-t^0$t#2$7ut$K%43-^>a6MBA-(CAfX3Z;cVQ#nXf-0>*px#+U6
zwZPDU=G`A`ZW*c1^j<JYdTg4_r?({O3CZY}@9^yG+>8jy%j`c$aE@dYnsGiF9MsMI
zR<M)0mGeSE7o$a3#tSGU;T1bDgk+OCBI~74sJQha0|Ijmy+i^FV@p8+V1`X%eK=ip
zMIsjtJCEi3I6ut)+W1r=hCyR|DiNK3Q&K5x<T>`br@PVIx^tSML%_~35PF;6##y*&
zH|kNWN;g(5w@8nOiBVA~2#X0iMH<Mux+a<C6JKwRNKx5XgBumat?s!yX#Yu<6CVeN
zHP_!11K>^gjPtF;j6LD+i9!OS&<^c$Q;hZsA5E_<a|>4C8=+vM(bdFQ-$7Wu+N!O3
z`FKex%%}+OLD4-N;O`CrYkedi7~2rIKeDePhU8OuO^C|wjLPMm6dkP|hA_tteh2Ax
zL%><>t&fQK>;{3nc4xfT6MJ^MYddTlBEIt4pHbiBqiS1P9IEaIFoiT=j@7Nk&~FLV
zr#_Zf4&K3iTL@R5S1pIju{UY`ORZps590#7Jw#}sOozbonyQ61!y<Yi6MaXBsNVOB
zD_sd{?+Q`VYD*lo<SRLJxj63$5zty>Bms5jglQwC64Oo6zOZ15?+sDBFwkncQ?1Zw
z{VC+KMyvi_t+BCIZ>1vX9QywK?)R4^`1beSAm(g8Pt4haSTAs$jIG*H+<&oAho2L9
z$thbeEpJwEVU|$~8Ss}vrPXtb1FCPa>U10Ekzur73886QhJ%*)*&3KCOrbs6D_?7v
zBOc7)uZJM@B;!DmQMDNFLm@nEt8wtUJ=#84u95eTZfj=gBO!>3y%uc4rK6oLP)?yn
z|0ptAokZ_GYN|<=%F^^hOVvXMXCJQapIVw)oL;K#pI+RzFoRRPhYo5^81Uo7iQOHv
z<`RTlI#gZUcW7aHf0ZcA&ftu85rN<J3Fxuy6PTZx(ZE&E;UW})f6@o{drDq~EKSYK
zE;^O*2R>Y0{qBw8@Q-~+J!ksG-*@QX;pv6NDQcm@O8qk*wV+Z<f&YaMuFY#c%Ul6T
zQ$_WKzw(jun#-?W*<63)!|AErXY|1Ap{dC6@OM6@UXu8jTzF@Q%^t!zel^a-F#E?c
zux`ooVE@I3)z$-_+rU7o&P~lbz5nw*n2yHfgUuX7Xn}+KrmF{Lrf2sr3ZO6eKv`Tl
z-E!g3tcq6Pj>#|kU^xS52^bUx;Qr~Ug{A7f(`fYRhvsJ%cu+ZI{fdvOgCO{2eelo%
zMWnD0XAUk+FC3nlb@0CG<6RLJ=9Sq)_wAoqaPof5&wF)7-kJLjLiSc?4(^|R$U*pq
zkDx8}eraar5Mu+<_4(5W1289E?ac*Q9YBW`vEUqjPWFMa#<>eLKee>)e#M;wx5J0?
zCvPp~Q#cBKfiA!$b+FFxv3yM>fHi#(^BvWhgLJl!qzy@arO(bm{ybQnJ+v@Y1y}cB
zHl8~~!7q5^%p>}TgL|HjtNlBC+_~wwLkkZhY|4S@Dp^C&q6elIrq#eaIJ0<v^}y_u
zi+!Pwtyk_o_QLei0}BVM`ws1&u1+s399pQ}`@jL9JDgtZqv-{)kCwyf^ufc`!&3{W
z-|7O&bJ)Z`VeRBnpMrLU$fJNTe24c|r)Du+s`>Dj`|#KMBhOkh)nTqgjG@YMt#5YN
z&nBI0Cr~V%ObELS3{N>{sH=U3M4!q6TT0m5hg0tGV{aR$B*wjta2<=7P#B_tD_iiR
z)%}3cZy<DG0@F@6rm5m7u;->%gxx_RV-$m;h&sjC6QrQ++anaC3jZN^qq470)giT)
zNV)qOLhp{Rt}1yE_0s5^+$gt65@WPT6@FGeh1(?sF^gsu;y|Z|o6@Z)ILz~1gy?ir
z?KD(a`v$PGm5c<Z!tWtu(M5%BSIBN<jTv{ar(``u*L2GEMh*UQ6g_35vuWK%&-Joj
zQqV@t!it$jpy=a<jKlbSqPn#!PDxZ;fsobHpC?%t^I2A{yR3(*eomqb^)L&45y3{=
z>*1Q27d4~zatdq4gUzX$Mag`$IP<9ly5(UBFV;s{bu){548k?@u&5brt}Uz?2Q<{0
zFOnQbgYUEKn|dyEI(t<xu1%VSjEAr(zAv_ff?{l$NC{yxAE{eZg5+RzLz}=JltHr|
zq^ZrEqZLsbH#10X1W9V^^k^hI%c$04YOf?*xkrOk^#<Q)RQt!_Mcm3LW;;kwA3Yn5
zAc}@CS&FRFOCJpq)4qhG5!=Eoz8s1uFyn3o>FKSE(dhN*v{DU65}cAf7Nj&@$r2MV
zOwu^>Q#a?8dCq+46F{$zYsMQ0+=YRNQxR_YUjBsA>mb;cOYhr*^qx1CkkP7>6Yh^|
z^E;maYS~J|tB9Jb9*-wI)r|?xCYaMXUn>blt3p9>o!rsj5E)7$<B6?=2)3ov($t~$
zdP!}(*}%E8iFOsuHW2raH%WTqO$FZI4v3VuNRs1q6LzZ48RKtzVhPGY{&q=lJc)@e
z(;v`L(pBfKDsI7dNMgdElcg|*1_2aEy7%4bmG2@N&I<3raU@nkqEa-z-y@4QmiDLV
zhKBijC9S1NlreX-|D+@uze1dwUwRn(ZFBRp2pH@%*H06yy*sGARtq%)paJwC_6z5y
z_svYrDg)|bn9R@mWQ5^4(C=0@YFq1@EzR%cWQzMm09mHEnAJl9Q`vX_)Y5&kbGHZ%
z8TbP}hdM}fUPTae%k=+~h{0cX8H_RSDHP^!4%O&26Qh2}=S~>41NFzTck6Xy1ACbG
z`z@Ehn5DYLsqCMgo?o1ve!k$5kw5D5DE24<Jg(PUt(tD(C1UV*Tn2^xIt#?JwldcL
z@L7tn!7)pf`-riS&~upm6cKQw-QfsHP(UG1*m1G`kf^%aQ;bn0Av--chiwY4moe_2
z_|+BDf*Z$M%ccdNLoN7ez*u86rp_6paz8=k)j7le<w$j2Kq0Trnd)aptMi2H!*ln_
zI%nL!_3Qkq;tr&{;K@|m|LE3sjDc2(ZodmK+P`@040a7;RD{^cfN6Z*qcM!>ym{S%
zQNKW_&QOe5<c44~EI&-;%S1(3eqvOOAf+6m50m>!kX+0fFi0-j-ow<s8l)Dpgb7m1
zHT5vTuLTLls31XtSympV^^G8{m_=ZaR>;7^bWXe`)I(x6nSyjOY&%Tq<RGb-t%)Eh
zVb)<HJAy<;E|0h9b41x0lO>g5%;8d<{X~&+mUv9+ydbHV?S^10X<L9}!eK%ePL@#6
zdc%Y+CPJ?MF|q{AS=ceHOC>FRyfRNS4_2MGm}C6QCH~0rCu;%6gszl?Mwa8*mKr8?
zl_WK?5ckHbYouXn*Gg(5_l6w%4Aa{6gwc}Y_9jVdWT}Uq5wgl~nJSV_%r0T5hkFJY
zhQFEM9la>VYL|narz$hWTV;VpRmC#QFihrIl1z;KGp}U11Ncr!Pk0Xp7N?nJ2QI|j
z1Zvw~d}n?8P)A|~&axCR;B$Sz7#=eh_6)hthm2Wc;HIy@GL!KRjDmN6a0$hDGd^C-
zE;u`R9-N)qE0)5HIqPG_w9D-5)ZA1C;GvNLrNN3hT<`&f6L+99GKVQ(^#J_-mJS^<
zuC9#yfR8MkI|In`_s^ubj#l@O4;RyYUhv@b!Hk-G#K((qA#(A;<$kdb9K-0tGdEXn
z3-_U#j~CO2UN}8}|HEh$&pyKqv+4ut1Alp%&^+$(gF~tYG``o2xb7py*cEZ)1rdo-
z!|aR63~>}lmeD6>2Gcl|?6JtuZJ)^)vlOn-P7RAMAP##jhcSjl2JsAwj5Y9CiZPNX
zv{_<oZ)g_MqlPIwP86JJJ4UH0B2n>FU?%l4BIT;{F`8)+El(L?Qr{*@6{CzqQVJIC
zm9lU#PJszY66jK_aZ$dDfbE$(X5Afx71D_DU+v?UWLrv4hW#ENwj|prY?$rW`S{lr
z+cPWOjfU?0m(fstBS0)+5~IQZq7s~g7;xL?>iP(-9(}BC&WPVlY`x88D`x+VFe>ik
zqf5$=Z}%CCS=<IdMPo4fJACwOBZ__8w8itO&_C)HI;Jg-H!9(2kpX|qXDVi=?WOhZ
z#&P2xb2Tyc6E0&hN{T}91UiGg&u1#eFqpu-ka^3K=uG1MM8c8iF+2T*G<>PYw0@pw
zxgtGgCsGp&>mZE$OMXQ?+38YrxVFmbi0<asQ(OHia#;h~9ert@3SH&XLyT|)q3`4J
z9tY!tMA;Ah@U#r-Xk$d3BV-5T!=*6z%BW480(UTeD}>?pVXH*7cK<Q>k>Y$g?h&=4
zPH??^G=$~ujayhc(sHAP;M#}=Tr;sT80_OAFn7FKU`}|I1rC)&vASmL5abN|y9Dd>
zTUX(Z#6+MKE4Vo$f0B^JW{SIxiA7fzBRCE92j$>Zgts|_R@pSsm-6c6k43$>Q>n#y
z1azQ!{w&D9yPaB+NT8PfB1pwus-UYOR4X0lXj2!a&@33C-d}ayUqu7{uYqa}c(=!R
z8N+boKaP7Gj0|Ad&k?h}3~){9ao{H8J_?1%D%MFMrt|k9IuAIqCXNoy?eK-Us>Ybu
zV;E&=-5Cvo>HbrQuDcBN>8^LT_NH*Efpk+`OHJ{wZd17PKn$AS5{&w9Ax>YATSFmE
zE2bfIR3p>NoK|6)UkcG&EKgH63Nw_~CQS9;LsVU{8mL8FKG(@=mSt*|ue!~WKSIE1
z2(Ut9()1vP{aT1y*G(#tTff;E{u?3qhl`pZ$h*}FoSxSZO#8%Z1B3taJlc93JdY~;
z6lAwVQoe-ag<4ko-AQn#yKwpK7liXGm|@Qf71CWK`|TG7j~rM`^xP29d~p{hs@qhh
z=Cvr(y&y!_HEQ{V#hI=x#KQ9atQ-0^YMG1OmT^~9UdBoouToum3>V13%vCq;eT;o+
zh+|iekEO5-3UoC~a-lRX>BSA19doRg$z2{I=Ppgkkn8g`601k6tH+!gyD~(^-7+jg
zM%5eNwuKuks~o_~p?6h?o_q2;k{%zuMQ|<qh7+#2YeOVP>k8CRfv&JCM8Vx5j4VXj
zo6)gdt%u8UQ;3qgOFaT5x~tY!u$WXOM9Q@WMv^)}I^z_krx!f`@I2jtym|m}_ndBU
zGZAxE7WslMW2}ZL`O=k(cPmkJ*TwnzLphScDt(ruInvNF_1>;Ndh}SAtCYi&Idys`
zQM1jG?m0FCl4$Zp<;!flyG7gOD}QAO1)J`<vKq(hEq2im^C2)8_DPZ>iF#A%?P_+z
z9q=2@T}>K{$c7VxZALWMcs+v%1RH8r*4lUigee6Z=#V7kZp1{^n)wX!*B4R?e!HAy
zYn*!Q-AIbZVSH+7m{;2$R-h6ct2qbMa6rdgeDV^?|3WvvtD%PTR~GeI%*Czao5oPZ
za7T!9N4&agzbM4YBEr|qnnNQD{1O5db{p3m9tXbk;2~XujJz^2WSmZMkk^99?w(2<
z&p1RgV*&*+*Nq^yyCR6g=KcBVUWfC|Ai8Vr&Oo<q(hg>aFrBX9`d~-kRouhPToJt>
zt5F@pX0_7mSeG9&@*s#D)t?>YR=wYc@zk7bWbns>;N@$iv$h5Uwo}nB3qnVA4To>|
z=HgJYKf=E)i0;}ZqM9;2TFn*m${?=mRuhTq_T}#i0!NKPr#i?1#fsp}OussaJwjC<
zZFFIfc-(Es?+K#2w(6KFp-X(R<U`CBr=Prz;9bSkU97}l*Kmd$_p2s-+`)f?#2<}!
z-g|I|&zmK-Yow0hovyd);f|0FZX&F0EvM^XZrksdB+B>3l}^`O)5WMiFbP!qen3Y3
zA&D9_ldmAaN%dG`Wzeowk3GX_y>}9%t<;wXyx9T#Q6DgBI$;jcn5~*OMg5o$7bWsF
z9Mqx^=zsG;cSq<`xJL)I=#Q?S@}Z(cq{SbM_zh!c<yQU~AJVnbXLNZB(re<M^YNml
z9|y16YHg)kg3RSF`lzM3bo<Q*e7w?J-W=fa*L>8d31PLqR>S=+j!Nxx^55`b<427?
zWZ??B-}C`5aC)Q@pMQZA`rk$dYe8_S{M_?J;9MW`$y^i2$Ja7eH#)7(dc9q5Y#GVM
z5T76jf1m8hE)43f0hVjV!1)cjLgJGy%kG%tE8I9wWzfaeW4*dEfK3rM#&_Cl4YM1~
zKz|To!R>w)9B<dFK?9p8Uts3mcfgo`9Kv)vi-lQRtzp|lfpXhW6n<3)RXfBXL;qO_
z+Filn)CDeJgz=dj7rPw<i?no*d#S-c<JO)#4Tkg6!Vm(r*Xbf&&`NC+fzuHP{%HL;
zUH!PSp?mZmg8!@w|3arhM}ntVRI50gP+!0uy}jept>yZveNh*a{adOVXSJ~U#8Ixa
zw~oRG1Xb%4{U3-2vF2QP0{Q4R0&M(qkd4+8$Oc{5*sniY-K@8(D>V#zvm?%J^RGdM
zmY+a|JPS2*@o&V1tG>Q?0xY=ONxX3mNyGis=^S4o^7-N}Sut4m$=pmM>MO`;N$96K
zOs9D~G0j=ON<?j?%=O!`GPBNqd}{-u!&|s8(APrTKRt_+DsFct<1e`Y!J^dItSzT@
zAIAGe2+s{#YcXfx^)moZye?3^?lzqTXz`c)6}!~vzKG$a8;#jKIfUx^pS!5Z`XF5B
z9U(N=QQSrA)sANr_lyvf>)_==@m*74w$2V=xvu9fmSu_zBc2yR%<Ot-3_?7bQPc}V
zV0pE+)x&)u!CrT92r93n=}eu0cxednoMc=iwWfR2lV~tsfehAQzRdj;fjW~+X&_zB
zwv*^s19{Ngi@&^ldsp)=0(>?lpE=iA9c<R8+grWk+iynFnMCmKo_yj78+UBn@!GBJ
zS8g}IbOXMA;Pw+w*wx%YsqeeYPTjcCq@GKuJDy^vekx6UI;9?VQqQ=_<lRfD-{quU
zl%`%nso&?M?o3nnQ|jBC)a%pKM=153PHH7heI}*;xRZKYntG5@f7(gyrm4Fq^)pWD
zyVKNXQR+#T+qH4sZd2Z)l)CPuz93C~GNr!ONnJ`)zex)1>zvdF)6{Dy?;D-e7pJMu
zpwzcHsr5AVeoB3(llpj?dK#s^-%0&&ntCOr{-cw6>eEem&!W^TuCQz4>@@W{O1;KO
zeMXx46iR)rle#ZWy`56$oz#PA>KdiK&`EtbO+7%Vt4``_n%bb$wv*aVQ!k~|mpQ55
zm!@t|>KmQZccrNpQ|h~%)DNbqM=13JPU^?f)C(!~BTnik($qOh{iu`rxis|(O8s*u
z^^0lhET#T~llsjxb)8bb?4+JkG3~pPQore>o}Z>RDfOhM+O5AcO+8Gh=Q*jjq^Y-1
z>P{#1V4C_gN`0o2T1!(~lzPBP-Aq##DRtgSeMOpjDy6=}N&UVw^#n?7IjL_+Q@>6l
zbjwM7Pnvon<$a}-`l&Q^ic;V0q~5s4)bm4>`Y%puBTc=FQZKvGuID$Tsn4a<JDk)v
zr>Un?>fKK2H`3I3N`1YPI{gg8)p?Zq-<{NNPg5VH)PHwUKb)psM5)*Aw0Ztynp&aM
zy-w=+HygTjN?mhOpOK~>qSRM7skfx5_fYD0I;r=jsW(vS8=Ta+H1&K+{Xr*nGfkbQ
z)L(E?Z@R_sd@H5yz{Mx&40|w5y^m7qM)Wkbk*2mOm2SRHQ$Ludn!7U7)IUm7`Gz;T
zSusuhKWXX;A=6E6Y3kS0RKEL*?rlp`FS*r}=oHFJ7ks6uuTN7sL?c}_k)|%(X7ZZo
zcWLUwY3fOYOrht})R&~GCh%LDx}K(T=r4-8mZrWmP30g}6wxb9eQlb`he#;`RhoMH
zGY!w?U}&29oHX?X#4{brOjFm=)XON9&Vi+=tu&QSH_>sdH1+W`m5&+GxurDq6=^CT
zW1+)CY3iHOR6dwM$5qnQkEE$r5m)5fo~HhOn#%6N<Zhj&el|^IA6D}EO;i6ZO*Jm6
zY3i5KRQ8r6kJB{uglCz0X4fY2dQ4Ln(^TV{kft6^Q`t;Rjt6P#=h9ST#Y|H_pQajv
zVVe52G}Rcy($sIHscbeNdsdoyDOBe~f93@ut<BTa*)-KC`zCd;ar^U67z|J9VU0<O
z`)@v(6!_$i{jF8{<M2%EyRCP+WVGw!56MTvQ?Y`sZ_;0O4NrqHwE+VI?ial74x&Ch
zyN{EF?Z;sNAluyb@cil$4k-1N?Qg!<d7PF5H$eOU^eHEtFx)}L*d9!(!&7D$=GEKF
zuihS>2J@qW1rx`gZvM4#NAm`d(b=0f;UB}Z)KX)o4RC6|Uh9!D@HXZukn7Av{$l4m
z>;l`0sza)Irz+2d^K6LN=@A(W&#1LK?X)Vm{7EIeYP-1yxtq^m2>gq<NLazY2swd&
z)i$@p@D!X7$K~QL-yRH4-s-G&sB)<_kbr-2T{oYKv2<j+d4YO;i+a6Cy=ZlXG+IXC
zMe8NJXkmmGt$^^Nr4C-SrooFAFL==^1ut5T;6>{Wyl8=e7p)}lq9p@fv^Kzt76EvX
zwvQL7^>~r~ju$EFc#%eq7pdNOk&cZQDbsk7){Gaa$9Rz*j29`qc#&p{7pbs#k*<mt
zDW!Ok_K6p%nRt;ti5Dr3c#(#P7pa1Hk<Nz~DR+307Kaz9Yj}}fh8HPdc#)=s7pYWu
zk?w>SDM@&dHiQ?cJ$R9RgBK|>c#+0}7pW$Akq&|vDI0i^R)H6(4|tKDfEOtQc+t#{
z7ftba(OixfP1<<T?2H#p!+6nrix*9-c+m`s7fqFT(VU1EO@4UMEQc3OXL!-Pg%?dw
zc+pIR7fm^M(cFR;O(J;FY=IX|3wV+I$BP6#UL?`+BGHW($!NSt7~@4!7cUa8c##~%
ziv%WKBpLA{5r`MbI=o1z;YHF5FA`IDkvzg{RlREJ^);2@+3H_6sn>4xs;bwfdc9D+
zX4LBe_1dRiXYgw<7=BB;OM@s8T8V{13AmHn-KVc`;?F?rHGd8dNEG^b{?rprIN1C}
z{NZ5pm-zY1{QLkve}$jlJd3{hRs4Q!^Vjfv`F8Vz>h&SKUcWt9wx#J{cxvJSFuWj<
zjC9%){_?mxy}vy??*QKITC4s*yYcA$#>$qGj=P#Cf|b{UC90%140p6QH!EG-?G5p9
zf|GLYG9rR)v}(PhPrq4z@h#35JGxzy_arC9%DY$RJvI4)YWEGpvwO8Q+}E^PuT|!8
zD;ubv;-sI;>AgDnMe8rlPQIX~c*F3VnWbvKhH&h4!`@jL9~d6a$@t(F&Ey#wAKa=x
zI5Xpe+w=#gWqj~V{lV!OA3Q7hU_g!YPbA6zkRD&8$Nxi*&(P!V>G415@xSQt*Yx-l
zJ^q9qpQFd0<MD>&;f~c!)fJkbz#oQZ@9#X;rp-O-7k$+kh8J{tjrB$w_bjNBtb<+y
z-F0|6?YgQnyZG0cxaXJl73tb+N_`S`L1+%d?i=Rk>Q-Zm0Fm%yH-V^A2jsMi+_V-v
zWM~Hfxt`*tFdR;tne<EC^cwC-PDy&16-RK)?KKqHi`~qf<tFVnGDkAWFuMcCp5bQD
z-#$1zvxN&w^segg$^0+exq+)w$iV?ytrhxmdw9;=>OB7$+jz7A%c=+|GUuxMq-h-C
z9LDDH?bkxLS!?Ov=c9_SZ9vD8vH8ae(d&V>+UWD2`^Vc`HRwRUaw2q^3#jayjrP$#
zf8Im80qtV<FL0Q8th!ZOM#`B3*ogbss8VgPtN9y1IoSLVKR?XR-=yb2HE~+#3)V37
zv8mCouHr^>9F{&--v(C~v<KLg?4dSR>*_2CcI7_`gyFfW)v&9(QLoXJTCW|ROTW;>
z)f%Ko2R9f%=Qv;gt-m=~R}Q`hho^5KayMM1=z#bq2<+^&PG_s#!Jc-%)7#p9&G16=
z2W&^9l6bTCHN*2B>-1K!1p|&+jrL%xzWw}<LrWYi_rVQhAt{&5PZ8od_+u5Pi5sm3
z)FJ%!kLa%#QpGjaj)RexzVx`NSWvrQ6_>Z*jP2%T=do&{hP`HZzWvuewtedUr>huA
zxI!B2s!vxt18Uet8)WLJkaLoadO~laO<yxSH~CrRo4(O2o1Y=RbeCYqg5}QRNO<i=
z?a21<qK(G-#wPx!4w*K=Yldew5S*-yBSos$50-~dstvX}HEf*rI|I7^MoEGr=#N+L
zfvZA&Pj5gsW$!^19^HdmGf`Ik>i1T*hkGNwO12dC(5>YaICMdV-*qP>%`;2nttMCY
z+4*lPkw4i-u(Lm>M0V)^Sa9s2%2G}aHqdl=pp7+jGin=jNrJZF=0<{3Q8|0rup<<g
zuGN(MjZoZPeZ2=aUW_yecX#>XhC0K(@-S!wu6#&^#~W?C37%ITtg!w^sJ3Gbs<OT0
zOE95cLQ^wtrp#`Qd&`$Rvm=jCi-r8%SDrui{aJ$`?(f-xrzT{r-PHGw1sLjyv*i(0
zLno2GdV4~s^JAl~Y{E-A>85rKlo>hQMi;YV8`A=5<u%MPm>g`dGX2MZt#&&htf>-3
zHdu}JTD?bFiNn@&M}evzt8EVG(lon(`$s`S0gvqs6|qcTNs9>G1m0H1>eWeB91D94
z5Vzt+DlS85cJ0b5GY9GX!O<{MpJlcJeRibIv4&AFncC{LpxcBx?MQXcDvGYVvFjeZ
zi9#M2WdyLpO}Lm3j)KPltM;_&txoSaR`+Z5s!PExc^QRD*W4aO0OhB6J;{lGR}z$&
z-aJ?(57=kc9d0wD*G0Dui5!4j8XeNEUQ2}8bz5dW_9}t;Ru5}pw+710a~_~#F|)nO
zC}19PXhI&Nsd(}Xl>PZ7nrf?a6dNBeWryZHdCQ41qS65k6?$<D4hwa=J4%4lrci6o
z>Y&v+Zg<CfN<%OZ0><^6rD>UNtyqEUdh0PZqL9~Fb*)!-)XY0eGe&^5$L+%1UK-<g
zXT32V0=RHiM$azI0}z%wgZ8S;N136hs<*$snq9f&nF{soyGAJ*dAQmoE8m9!P;1!`
z<(t_@D6^zCy%#&Gjry9+Ncn1QcaC-Z8Yy2KoUT~5`6!Qo-qN)Z%9jLH3bV{s%|<Cd
zhWj0CMLQVf`$yYv-SXA334^m;kn&6r4^EAiQGRGf#wqLm9`+w5QwH`Cl0VvPm02Aa
z%+N?KGYc4?07GR8sr1E+Lxh&mvkY&El538A*_BnM4HF#S&{#%HC<yFi@P+qIZz?k^
zG)xNzrtH#`8K4Sd+;&$emM8yMZA6_%1<|!jRYqG<{De=01((2$64O;Kc>ZZNyF4!}
zP12l=P%&`!WW*G*>b{IcL>0@Y5$YKakIlKNC6AU}IggA@A}eI1RXYz&f`|w{Y8PYt
znhJ)RwMiT-T?oh>8mYY4F;}ba;h^eP<s6z2uDW&A0xsVt!we(Jx`jF^cDZNTzzY+q
zhtNQp6t=mc_z9>Gb%Uz~JgE*x21L7}%NUlFtP1xFx(UHz_XQJbKYfk11zpCvlE-lp
zcK!)KPOjFrkSEpU%;hzVT_2vWx4!ATZVTpw`u2TOpZ@IDkAD6K9vWV#6KidF7_i`B
zAIPCTV9uM_$yq%PwB5GO#KBudAIanI=1BFL>a7o2m<Rhf)`oc=PEGB`77n)IV||8_
zKR(>(7k#e}5Ayf>m3o~6uVZKM=X0b^f<SGvIHNw&${9_F^<EAsea2Mf&>Iy~YNIG8
zpAJ9b=G#Ais5q;RP{3KIutT4c;bXkXmQ|I~l?Me?`NhdmL(bLjZlWpZCnf2~Dp;ss
z)4#$Nb^cws|G<8=YtIdYc1)WvI#Xg}s}rSD5jbJcgICa%2{*|iq+dyXRFXE@?ho~G
z;6kp{79Zeu4Np-K{D&98x3ANFq`%eMy>qtF-=bef;Qx5VU2wyAno8uqV~iY0J~98s
zhl6<PYBw0v6?g{a{!q7CVR6i+0hb6qW^^vOUS01FwudJ#cRKLgK5c+Mko{l(WiZtJ
zun?ZwFxh;e7SW+8BJ#N#4$R%DT;Fi2%qaHoTy{hxZzuSM)!=VEyy6hMx7RiwA-u{W
zIVvI@K4M7LO>`$WM$$)EE^`M-RvdZvLQvc3=);9&qS(f@9=Rgn-Ul2*?)03>+0`7v
zp|ZHILU_XA^EiivGTNGMn&IV%BO5%|uo_r~JJza2C27*;P1G}0^Qw+?m(@$D_wL?_
z<KSEidW`JeN#3iflVj6%WuvkLPq_|t^ubVH#Z_1bkj=6dWK&00Me`1;9Cum2dE{MH
zSSDg!-^4i@>vr#sWV%1pA=nB@w(30uHQ4O*YSqW;<Z#9UR9_`mn4RS_?ORcx2;6j0
zeUzYR9O@`Gg%JH3d=8O8Ur<w&fxgutkV4-s!9!-jILOSg&BT)PyM||QE;U&U^?`Ms
zMfl?>J<D$o&!(+Y!~n!*eSQ7-_V6V9M0=K{ZcP}?xB=MD6=R_i+P!2~^9HM<ih;Ob
zb=o!f3DbmuBDVVw7w8WC3fB*bICXVN9|2Dv9gs)HK7=*sH`W^U-l4%3Bn(|%GmjV(
z)XDRDT!>*$Z0pE-Oxv5X!`TBv`erJ2hU9MELWxnr`sr2$rIW2at6EcQk!%Vv9}rM(
z`^EK=rY(;|GvBUCQJ|MdDNF^LO>)Jy25G~|7JE7s=<Y)kYh=G4p4wmOz|nAfc&*h2
zx|4FP^F#fG=Y@yHU^O^{jl%YMrupET*N1+-J)B}jvu6B=+li<h@|2)Z$I`M?2f>PM
zzjV&jvZ}|ewm@ShPeiJzeNHu&BShwV^(A%=SY1*Z-F2w0MPD39gh}4~6DoJg|0+ir
zqaGK@`<Y79Q27Q!eX&vbn&TY${6G`~!z-}upDs?SxQU%T-m$sDE1BUrMztixh&sgR
z>^(ZDwaxJcDEEcc)6Z2q^IL&kSzs!m^AS%JEmhxm(r$gD-NfmIGJW>sgB>{Q=j)aE
zg;WfXj*V`ObtE5hiRgF3vz=zg`PTanRHq(TIy84^|MYBi>EZe55kzo01X8|?REn&a
zH_~tqa5FrY)@QYq6)fGL3w3&WEif3qprrT>u^E`j3w!7ADVz(7NE|3(A2M21CGIph
z3d}Ra_V7~viVobMnP|nfxmm|a8M<F^coNpi6X*bJ2}&kE)N}4aCDftlio#<Q*D>5~
z_Cl&Z)VEmhw-IU*PTiG(zq@2>lKM}^Hyr<3-`b$k>uaxK86*UI&Kc?}Y03c5MFkv%
znUsCvBG1$kXah4mnu{0#%_Vv~fJeTnKi%f@ADITFyk$ZtudW6vfR3~rs)bNqXy+(9
zm*Z(Ph_jU7czaP<4r0wldG?u5$!r-m)c145R!ewwo2ONUSkPKk-slMBD>pi_5OlZb
zZNKH>6L4w_|Dl8A4U;USQo|$wLLD=ESx<)BIbLmZx%E)g;pmbk6ON9FebhF^?i!v>
z*3(1@(zng=3O)JSTBGksN@Z5(-q_C}NZ~!%dav&^DCAK^0A}WBv0UR*vYz2Mli(Uc
z(BqgTt0}|N=tmW&R56wE-}=5;Aroh%SEuR1fr~9PPQPgmN@UrDrH>k4?^+bq6VCL~
z=Q{Rt(dX1(bSrAJ^VvrAC0ik^=bn8bvvw>n?V?1y=<SIZUT`vCz258gNIyVUz4;Q8
zHSM>^7~A!U<z(5S)>CX}>YGiaBEpG$!my#kpHOo!6)RiS@YJN6YSWe84=RSF<Mt^1
zuHlK;-C$K|crxD)t@S2vddqAOR*qps#(o*^(l98_ozuTr57WBZj{e4AZH)rHBxd9P
zpqVVu(!A*?liDhRI%73JPEI&4M$NYaN5vcTdw4SU<G#PD&cy21u)jxE@cPP8LOLq}
z$El*ms^+pk^cJg`aFn2S`f%RCL2jP<Y?M@$x<9<mRl>6Hs2-cYN2<yv>G3Ig{5~Fo
z!VY`?fjN@Sb9`vhHjNf93L&>TYPqA!KRk7{jvExvYTl4-z7@>qru+{2Etj*N)lU@F
z<5b(t(kU@&o&5wEDfDF?Sz{@K`Ee~2o^2<#&X~(m=80xTqm9VG2nGmG-HR=1dFw+-
zLaHwD9MANiuahj<ffV)fc86xx@_sPX*V2v5iLIqam_RaFv4zH9c(6bcW{j}3f=00{
zG=GR>RWd0x-$DOK=Ag0K%Qq900xPwMe_okdD`_Ao(!!h;;dOvx9Vk+?MRr@#554tQ
zzBsV;K<*GUgKIaPn10UlE7_r%=VEqlo=1=K>2U!)E~E#|aLp&tFA|l_i}0&>_RX{x
z6ftLKq{0;SpggYvEVp2u-J;MpY6+w-(aoxv0x+l(xOuuaDJ%qUX(_CI1)aHwuSUXC
z(WF&?g8&Y^QuMa+MHei9QQEkI8j~aJ+A~MqR$W<+Fifc7*o{L0j@c6cn6}=ST@+YQ
z<gwP}l8#$piL0LHOjEYPQiriIyJiCxX)f2e48cclA{B(CBl9z6xr|k1%=W@HQ<G?R
zIB-m&R1|A#671+kA5)O@>Z=0;dt#1?l(zzzlt3ZA8>5Zc8{vAD^u{sAey>gk2H;vY
ziPr1sY9CX$;Sk)!ra54xf^y96K!%jg+C-`f<~D^h=EQ<PY#i)tqtiL60E-q`x&3qn
zPNU?L;2XZniEqA+5+Fr(hg?=LD-k1O6^B}$ty7*Izg%%}V;O^tRh({zjkRM*ePB?D
zs9MJA$ruyK6@vsd)&!zr@Qn>a5rmW-vB=z$*Jp^0qHO}b)Zu7}9QaTh_=W{rAgAlB
z*q5KMD~>XqoxGoREe3@PF!yjw;`URzJk57w9X0WSJVdd=war?2V7u%h5GvL%b-JC;
z993IN{+Rf}>v||M?Bi64d(~A$^CJ&iE7F^ukf~_bBIo!@QP!N35GAwf-iJ=k-KF_d
zEQb3LR%m-cTdBJ{*v+eGMNQk)&8zV!URHbMakrzJ*D*`F`Lh<e9)@ewBA?pkjr1T_
z%jRx82E$^5hEqt->9uI*z~cv(X>D#=v8GVryPD+l+T2SIa&T?lN)N8K+wf~JEH(x?
z3~CE|?g4`4?L>qe<C}NVgWOP?&!z|0%-!^hS-1zkgo}`v;zTI}Oaux}KC;azqQ#ut
zOTU<teTM#ireCDU#L&09LX#=q2kd^KHsWE2{Sk^(ZVv?d2SVY7eC~@jTs<J|*HE-6
z=w_q2Ob#S9eEsO9$#%W7xgXxMk}{_V(X2D2B36(!xQp-krJ$GQquAT;%|D|q_vhRr
zq%M|^6WYx$TN&1_{s>w1?!N3j`J5-a{L&<xpN+b&x7*+RJFYsZ??of&omJ|bQmPii
zcw7FS7h7RxpfWjI|78aV0oX18&Chd<c%Djyd}vRA<`<|#evuwuqQ{r<82C<Q5op>8
zp!pR-`QP-QP5I_m8Lw<t4|xJKzeaffNsq77;~Vr~8sEgPkjk0gm}39G`7LChKM9Zi
zP^$3xV4-dL<_?CE%ba{DRCeCxfAb7Rl|DJSsH^}4T>#{2E(|nX;KLX@xC1ye6s2h@
z%*H;x(wscVaCVv^Cga4>9Q%@4?g<;03@>pG5ZsG+xO<_llb3maxFB{4%F?nVlXAwT
zA{-eTHi-?)N~iMLf`Y?i0ryISf_H)?b693q(xbGm04p%hNiKMGZ1y*6?eziOe4JT;
z!(-D(s%Q<4Z37%%?ZeMpxyg+Wjab?6UPI-;MZ1N&>G?iys`l~w3rX820G6)q5U_G9
ze3ag;muhTDlLp$dkmz=FYYS=<zGu8-z&$T;v^r0nshRDx*VPGL_)c{jkHhmmF{ccl
zV%+=Grw8a9VshGdyZJkGjY;k&!*gJVRUwJsfZjifZ~z{n1{<E+#i61qOl5}n)6A!Q
zMoV&V1kOy!3aE~IFsMkm+%&HthA%U)e(`~KI-Q_Ueb4fPqSNn4JYOBO#GTzqnLsdg
zmP$T)Qu5J8V->ft<5Vh)6`}8Nz@9=9C_hgLbD_Hbg?j=V<#4Zyi$YZbmM+8TnB>x(
zDo1^TJD|Ec1~lzs-S?`qUrPgg;2rQ{fnlE`-B%m*Yn!uk^7QEP*doM4_)RXj(FLR%
zLmCLobSsVd@#iCT*&J>lp|$sgcca<Q7ySnpbWuWHe`sD@_{FvKMcNBgyELzN|El_i
z8Zf!^XgxlZ<8lyK!IqQ)sELhBn*Ui!!AdSjS>b6vbxMn7hDD?b)u*SsbjFTdM!<N6
z+g*;*)Rsh%!vMrl7KjV#E8bDN%xU@l1Jeac)Obi*W)n}bYF(Ho)v(-i>5xeCm+8TV
znh&t9Js!Gear-qu_5-^C;ocQy*tWV14@HWIFjUy`5^FjLV`}?}7n-7EPa&<-JwqjJ
zc4aD!my-I(Ep?h$)gfHSpi&&WJ`O?>PoA$62lnUbqH-@{!_4`r|L`ie@k8GetQV6;
z+>;Up-{pd5=OPl`%3gfDkR5u$KbBUMm&nPj2Bn6p8*jqpxy~&Q`eQH;!AtwK{VBJD
zIcozeo)`5>E?4c`!5s~0WU{k3{(siePq~zqbi|YHf;ptL%~?y^^!$7C5=?DMOkRp#
zA)9{$uIu&Pyxd@3!I*LzI093Bf?X_i%sQ#0odnJPh))|GI^?X~yassvz}|TTs{xnn
z=Jn;_5=ueG>*kGAaGuU?;(|*PO$4Lrlb~~TvqA*OiMqLm2}oy>QUvVTEor;GhQclM
z>#aoVR;DF4F-p-2?4Ph}>$8XqIXpLSr(fjl+`NMcj!}EQt8)pc7)R&kJz%gOaDUCL
zkvJ#k<`m<~jetyCIdyxpgC}|&@h0YJX`Rd^EKYOYkSq3W+K~JsNmOrhA*_{z=Ok~B
zsURy&O`b>$=a?=~y}kB`1|bGKWkT7gth?O81@)AP>|8M<mIsZ^)oNlL+v+@q`!>Ot
zXOtBr3Klk=(krrO+ZnAH*(~`yWwiuaP_ul%?ZhSE)Th~2%K4FzbIhv`jtp9CR`vAu
z@>L<tsiQ+eR}Gs{M{ob2`E+!+pxM-$dP>xx5YL`GDgkAgN!c>gyZ}ZVHXb#fM8C+~
z*}RB;kyWVqWO`gek4x!s89kmt53({gFQ;Eu(BrA}xRM?_>G3prTt$zo>2VD`$cEIs
zjviz;YVM*3Ye+ZX*I@X<k$Z2HwQY1kBdl!A3b9URwdNjra78_XesK|Rre9pdz4VJK
z^cMQXg}oKO27W}n`1XUyHPTL+NVb9(a?a5THfQDJ%}dY|!Z_nNP!u8NI8YR1jdY;M
zM~Qc!$j6CupvcF_b)YEBn(aW5pEc8gB0rloZ=!3eE}rE;5sjpGr%At%4iv>Og$@+O
z5Cskt#Q;$b6!|b^9Vm*>$~sWw<4wkaBA-S{2a0^KQVtaPP!N$LIZ))HqPJ(nP>>eR
zFp#~ha^2iUI65i~3Q<H$*sfydFRxgdLV*>_Q{FOXwG4E%+%2+ajY%ZOvL$y1OI3fi
zjcY{UWceRw?edm4qt$(!-OG#lH#(_O7BJ}wQmS=SWl{ml+rlPNx4u<uB+wG}G3nnD
zQD3DjWpbl6f&ytT?sk{AOo;1FVtFPvStFSdb~Nd9S&YOJ$p{iZRvs_q2Q?qY4vr9x
zS=yW7Xdl_}qrhf1PYBu|lA=l{6^id32u1n|*MEQ@X|K{!9?ssaHE{HmoDa-#9}|hv
zt^lPV6r92OuYYDKc6|K*r9PyYKip09hTwmZE<py3p+1Ket5<Nzu*Rh~Emr@l^>e3C
zo=G)R-b92@7g0`P#bg||6ZB-_(COju*qJ<%YaJ3LR*j2;)e%xW1OAxJ2Yn4uHr4s@
z$?~h?r##}?K*!`4YKa$4WwQMO_-VhG>_`E$6>Md~m7O%<;zTPu4aG9q5n>{<Oqg90
zsAaO1tK+szwi1kWc*4v+A%j^5d6{e#n3yk<tpXj3WU`XdjtC=9l!SGgILYn>h?(r5
zNaMv!R+_otV<y`#Of;Fv%62EX%*3J7X=Aee&JH~@*=jbCXeK*F%fK|TmYA5~V&c?D
zcjeK?pblH(ZQqZjt6;jP)$%0EPF;-S(M+=JI>yZ;%btY1nPhjhBH&DNe7ON}COHqK
z@o^@Jnug1XHPc57D(a`P8|LtR5}{XOm-<16<ygDP{RfuvZFGspoqxl~c=*~nLH0Ec
zOYsow1EFIVVc`>I+P2!My9Aq$kYgxE_jnQgVz-JT_%#T)VU{7~1w`?#x5*T6VZO8D
zqbw8C93N%Hu)oUeKODIt2~*VjV)@}x)`=#ECfT50T#XIGN0a&RTyaX4oc0zHmz<pY
z_nIvt+NMW`iLQ-Tc#CM<O?u59(eBe@iy4SF{f7pZK7SKChlSfPy^%@}Cd>E#IPuE8
z{iV!stZ})Bl(&)kazp0Zm`tpWT|_3lqxwps#wP^6bE4FOTdS`EiO}Kva#BJMw9p>w
zYbQb>ce8a!HOAfGbzEcPNsY9}rJIuM+F@TI*L(xf<kJRkVw&Sgh!9Oz&hzfoTMW_f
zr(b*k;cZM_1`-&p1=AvD?ZdvEDD#<wAEsY?CgGjT!(^-7-;f;-J!)t6Jt#(h7`vVc
zkM77V+V?Ukc{H(1t6`1c?bXI=bz#;C&la+QqlsnN6T7$Zjyp?IUxONzp{7;bsZt$w
z?I~~bj5CYZMr9%8_U61Vo%n2)2ma7IY2dPC=OPkbwS=oV@9g3&E{{<FMA*AMDV#6f
z%{7XR6wX*NJ)gz`Hi8bWx5d<eCY|No8r2ROB;?v!ssb#(*0TAuQ>%9XHNWB#k5+H7
zmY;`&LO33n7hftGMmq{p_wx=_6=Qggtap@YbSlcKxw)=v=V$47yi1`al&+y-fIXuO
z2l@W1d&@!PJFk}N$)k9#ddHmdn=aRF)iVI(Bjxz4dfqzuI9S3_Uw{QCc->bL@yk01
zp9_hOk>RcColOa&EXn!K_E;S*#`KO5NPz>9x5$fU&el&=oYNimkp2`UUIL7gJn+Ps
zIt8LbJyyc!?Yy4u<yk=#vH}3lQ7#W)-|Wo%VUB^eM92M{2J+&BmVx2JQLK7XC3t?u
zWq{NPN*xIZSY8yVGFb9_n%7lIftgrT;}mao&}tp8CX7*4d2>)1j`SMeKgn7OL*1<h
zM!X?j2E=$&`!Q|GfatJObY&yh@ywQGklKS5-BAf0i#H2i?24T-NZKlpnBO*Q%k|BE
zm3(|TWEJ{*jjx0DCh-y^sHj?n=r+3b%|^Rk>D3T?7k5Cew=qByT`lLVm)5+(`Df=4
z`|L5T@mczPhN#zbcpRG1e|0d7sBA&?S1a-{2wB>1epR|v5mJ836yDN(OJ>umuU_-T
zlsM6RpP6*#m@L_fuU@Q_MA7;Nby}@F#X@*$uhy7ivpZUEnGT<!^y+CkhJL=ztJl_X
zNck#nneID^#;fI246nHguhx)afO7O*t+&L$#HqVC#*Yv5;tUa(-=q|lVJZ>MmN+#Q
zkZ@s~TuS~m)|8O1E%QytCVEpb0M89H3bl!*FbjR0U#ta8I!irV*yk#RyP9vMi!Ju4
zD2(WI?LPTT;_9Rdt9ulFQQdYi->a+qt8;WX8L=8Wy_JnhyIsM*agphf?dGcxQh&JH
z0l`1uW(Sa3Z@ph(0Q}3$b_kToLw}`%XnX*~l@)0eOq&_{s^WagGv~e}-w^;YDI>#%
zSEWGAs44~H>{sh5WIC&k9p^O9L6aw`kBpqVchZd}kHEu3S3>~(2x^3{A~|4C-KtSR
z)@pQDi>hPTG5YmNTG2~73>)<>aC-(f33lijTZrXnQ1{hPdFW)1JaQx}f@ZJ@!Qk-R
z6-2ci;EKE!LJBul5S(*5fa~(DdQs9Xrcr_5m~MqfIlutb3?ZMs0=CF71I)V(<}BRD
z9|^+g&s8U>_o!5qovQH*)Ulq!F+v|v-LBK!YP2c@+k!@EnnC?}vsdk73{tP`Rc33g
z<<;8k0$93&6uw>umwETAo1OK>R-)|bqTgfKOsiCpQ_)~`V<)j;y?-9T4L91x7ap+o
z7_6=0>_}2X8olv3;VR<f%%5v*n!S3fgOI~E*?d*O&J0^M@~R$4x{+B~=u-s6Wkz6T
zy0m@0t->^8&<|R;Koo+qU9YdIXx4hc6oHl$XV6A$TsnxuXe{>0m0@@ujdf7$YaM%g
zc;;%oztU@<*K}|zy?5dGV5oOHB50G|p@mlUvBuU$RdM6;ar<*{+Tt|bnYNn`(7!#h
zY^$?A8Wb!ToHk+ghsI+PYnQg36=06|Tda|q(mR5n=lN>AO)8~C)k^jN4ko0CJw=TY
zPYBMckJ1xak@d<lxv!Yy=6p6)+a*C7Xx|bA9u&;teu2&GV9L9z$;{-`u9;jSs-`SI
zh}f(w#3?_+v&<i<E()tDx2Xok_EFqa$@+>VE{Wi3t-YFxT@q;*&`O<WWl{hTM7@it
zfbYWYo4}Ff=@7EE9u()MZkw{Wg5^E2c@$yC*U5UYYZRSDHSD=+I+|)9-@OwLlK%Pn
zx81pt$}_VRv+^i)bFQD^>4xX+;YGn2uhe8i4N-rbM--QOc!o8Ihq+xs6*jrSJidU{
zjV`2p4H_Z|{Q+MQcnUNO7*vK=E9DAB9JFhg?(s)kp`ywxCz~L9yPY2G4X0$eM!%bC
zG>Y8tWa^(u_Z;pp+R%BaqPK=YhtTE&6W^q`_f=6s3WH9Q?nZs(Xt&d7v)gR^oK2H;
zGV8M5hu$AAw7KV0n+<Hklg?fxc}C+$TcnFi0Grk0W&kM^0?0S!*gd`aSg+PiGv?bw
z?2Ifps1=77_v-yYtFEr6u<Fr@kgj?Z>*E~K96Do(wto^-k`C^TAYWlN|8YGHFFeGG
zLT&RA)>?s%emNNCn<wldR4E9)E=P%GgNBb~S}Hm>JL-<j_+C?Gxi{E?YL7tv9QCh+
zrbn}jsBN@*iQ&%>%QRha&!ax0cfX1|UA58g59(D2X<9l$WUtZDCxRp#ZRC%@Y1NJ|
z*Gbh^W00D4fNKv3r9@qFS$aH$C7DZ9L}2jipmGrfVA5tTkW*}Yqi$elxYw=qY7lLx
zOeLFnc%4R~8j<Yl7}lo8Yn}CxM{F^jEbr99SII+K>Yc9_k)7oxE?#LI1Ju(4P`)e$
zTmrH^JgtrXj+sEK2*cCSZwZVv^^-96wws@VOVaSU2G<(%!3Cw#M(>|r<Z0+$EG6sh
zRbqDs3fx9LY!9zDV43Arqh8H|SLM*T7kzzU(1v{2ool1A<pVk<x4_BxfpW{aMv*=F
zAmj*u^R03=Fs^^8qpfk{MjH}9(V!V0>Y;7e(5qTjUlgtE^fUrqM89?DD@BXLH(Z@|
zfh<NKy2(>=81?C(ges^z>1GGZf9z@TMslmx#sv;6AoZ?haq()nXljLSh3U6yP&)!x
zXY*Ap*w!Vx%&NSNrEjigEvCj3h}NK~EtSdcT%{Ho`&$wqBdXd|4WFeDR4+jPiZ?;n
zRSAwtXhb+8v5C6dqzj7J4uh3L2M|e5r^y@3(s;23yvI5{1i8({i=VPtT<a-o*YJ$w
z(j1yY{e}DR91JhW(l$y%TAIGsW-!d(FLDMd)5A1PRq*!#>p;K2Y%IXCfl5>YItJ7&
z3>0eIcxAELhZatz7^CYFfNI*)jShwOEdkl<&_zW-NF^8hB@Ew|0#IThKRnaY54Xc=
z;qZL5WZ;XUApO=FkLziRrJ0oN19PKyu?PU*DqW#XgMYKY81ny8dQf$3evOjUT*^${
zlrx{r8=J1WZ3*%rUS&$^8FM#AJL|5yn%_>u^Ou@z9%KH7PfCicARw?i?akv-jy6k_
z_J*9&PD3-C1*fL0CFa^TW`1<~_1e~knn>cM7yKAjtF@IL%}9v=Q%()>Rc+N1uV5!L
zKb-blG*)s=S56gkFPc+^5Ti=jvkEl4a(|}<R{l1rE%8`;q^Xe?)>x;qyouK|%1px?
zEdqN6Y-}L8xC4Pj)K-kDs4{lW#QvjJ<txb@cpEp0DQ3rJW2N5iLsj*x#>11hYNRth
z#nOU}gx$T9G)8E;$%c_qc723%uxAg`OdEZd{4FGjQ(!))-J5(GXOUtF8&5TYp(Y#V
zBk|ZomruNuEv7r@Z-iF0@wN;`AS2UYnMCpM9OywUGOl1l30kDA!PAzZ>>Nc0)w+S2
z3t7aKk*C%mA5ES%c|f0I#$s~sTh7q-=ZaF&v&8(a#3rw7VWsRg){fKIDzQYSjkG>{
zgpn@RCnF$6shY6?l_-|Hgz6V;s9U0i+L!h0<})#Gv6Z;_Ec!)TI?X%iaVI_Qq6cjW
zH1DPd?e{kCp<mCT$8+g1MUQ*wv5y}6=|LNs&1w2|fFAeZF^JG(%KIfykxVX<R#;-g
zTCZ$NYP!nzIdTBl9FdH7YNq0LHnce_FGtOGI{=Euo&li92(S^r37IGtPmm@f&X$sd
zO?JpAbMYRd$xL>%WpXyuT%&)nF}v6bm4m@cAmXIh9I8=j<dQ;J@_j<A=Aw_a^S@Cs
zSGyhKdg%D9@KmM(Wr_JWGZQQNw13OoX-5}*S4$yLQ>N>ElSu|eDv~g-*xzC~xYgUN
z){fO0>=F^IV><)2ZxYEd?bo=Cmi|%gbvoElQwr&6{1;vasz03NI6fuH5_vtKQrPC|
zPuWTm$zADzaVig1*#7xLx)0dyNWwwSE|r}pJm|&1%gXK#`P}Na=$f{@F0XCV4AfG>
ztZF>?WBg?GQ3z9uGPmsxpWz%PJj{Y9#xGXbOhTWlVZt3q1*8l1<3P?VKUlGp59AH`
zzs3}fd^^-ya~S#gCVYCRM53hwTqS<4Lhc1VRic-zwm|36(^U5qh#4zCw-(*RqcF{6
zWcB=6g&fVu#%-MJEt^X!^ExCIGH~yJ5QqVKWI95Tb~V2Mb7}L7^!O4zzKq9Uc!NLe
zf(ZHksxjS+t&^CR$r{?-gku_U^flt+Kk4yxdVGT(|3Qy$;t^1E0=(t=rwaGp0DvQ&
z(SJ@Q)oce}(I#YPAbXrYs=ik=r-`{jevb<ONqW#GM)UW%8uI$${Riw5xVGHN?ZRF*
zeOMYisPXeu5bjH@z$(9opw8=X8TiiMtB8GC_BXHQGUN>m4@w;{HmqF7tmJhuFTY`u
zxw(OOxRD+l4`nxVlP8TmG@47->OGW?w*&LKgvc3)JU}*w=3aVmn3!AX7gy75%$Qiu
z7~;K(2amO0x^<(rqs_-$a*oq;I}x~p9(U4%qr*I#esOKwO~07@J@_>k>bN+eZt9Uw
z13g$lX#xdiE~bbMb8#>IVlMU>+WVQdKHxPfZM*PnKWN?u+<*<)d?>0ke0j4z5f#B0
zYl(H7+96LrYzb5!-R$&gXbq{rXLFIr>s-m~HU<&Mn}f;f0w?L!P7C|F>^{j-R{L08
zlQRYInub<_+@*eoe85lUd3<<=TJBI>KJ5;3jwZ4(SX+ZLd!p<HwmPlS!;>5b5aD)&
z8A5I2qBLS@U}exdIE@xvslK*Zqs`IW3hrTZ7KbM>TxeMfhvc=}i?Yn<LGK=o6y{!w
zFDpx;Q_(C>da?3=WSiENNnKMCnx)Ua>iA)NPy15K57RJz<(7#9AD%;3B2c_hmiT55
z{Y*sE!?r)W+inj}VRI?=nNicNF6Qg4-e6^G(4%&y4ff&riASx~0(xC@*2+e0Yq^ss
zE9(1!+{84@_bOJk0lZ#oIRdx&H>riWKh&WvLPIgdp|;IiA}n_6D+3%Ofd4_7$(cHZ
zW!~a9oV&k5TVc?NQD<!Q<M>x_y5y4q$<}G}FR3uPdQX9cpbA&A<liiZy$pf9{9;bn
zeUCcD@MLDj!yKMTUVh4LR(AkJTZelW4mgE06T2gN*5l+|2L}r7lze1F4`sGt_ba3?
z)mPwUrh}SzqKq%!^I1NJSQgzj4GI_P58-yAEqR<Hho@K{&9AC~?5?DZ(zSmcWwnJ~
zwg`4JH@{A8&VG?@E$3+xVctzMj9j^JZ12&bQf`Y>%EmSgX?DrJpk2h_DeJXX3r-`d
zlz(9gJdi$ErVPDGxR3!EFcjy*3Q@)qz<IiAkZH6>uC~1dZ-trggp#M%df7~&M{fS6
z)qr*@m0*HKF6Qd`3YA$H01b=1ipTs|4U#uhJUW5y<4z%Y>s=R%N+`6dWXHN=3{CdD
zkzpzlBtsy^QY0*Hfv<yxB4O8(8MEv}VudNQhRj3}1+%O~abIgAk(lZwONot0OJ_3)
z$S<=^L}DeFo7uJyiDQ)EoSuP5SR}%^9Q#m2C(JMp2}_%#mTw&rn+PF*eB)3=-x|d>
zB!sl&EY~z7EQ>OKj%7%wmoj&pVd!c8+Dibjv=7>aa(bYvRM|30?jEKTy<8bqp`0%4
z0Tda9;*o+jp`4!OF=FN=uFRlGC`asiV4g)Nrz?nz0sBaSK}ein%51GC!XBi<&;~@S
zC|iJjpt#ZIAmL@3S@57G%5HJn8YEPfOxUn7NNfpb=FheT<xE&!t!h(Hju>`xkj|1b
z&t@2cn!gQ|y+6D<TNX((Q1fHR)BHF+euo~PpvUjxF_8NiX&nWPJz}dc{eD`V#DKbG
z?`h^t+G&QdG|rqUH>H4>rP%PNDvv5bQYx~fagU6-qfo4T5)f2~#fL_Qi{OenWQ-qm
z9Npzm3E)qdFe71B>o|+qRVKr{i6g{Iw@XLJqDz`HnT62X@S`q)p3}7rIyIzN31PeY
z*k0oafo<qxdzUh!W89jYRj>n6%om2k7@iI{V7(+yLko<8b%ac=xEjnDm4>JDDF)s-
z9PXdR%^5u$Q@b8#>sy`1>h-&KUcUkBN1c8s@6vjEtJmosr#~Fsq$e)8sV2W*Ik^5v
z620T8(Am-uAY3H?O5=5~|I_el<4ERwz8}U<D<a1Q98imWTv_F)HgFO`oo_uGj<2f#
zu96%ucII<okZiB0fEZ-ENbHjAX=NSuGd5PD2hj)K`B1iB$NqQrAyYr?2o46RXdRVs
z+0yo2a-(vvr&sXg6r5k3>PaEy=rkr#B#*HFBGEC}6K$B*VR|CUwsm?jFT8)EpPpe|
z&TDzV;G_f+&!@!1`izC~`qqZ+GiiOZg*|j^Ze)FPrX9z9n7TF6=^~K10B%4^N9}2&
zq;wr0o(sn?a#^a<dIFm|dRt@or1d(EU(%_A)j^LsDM?kG%I<ul1W8HV%ac=dbQJB(
z7F=wT8VRs?CRq=<I7PI%1?P8dV7`=8GIe~0xFZ`X&cGZ+%${U;1<t7E$UE}<A!!0R
zxodb~w*!~s<2Y~8>4P>7rqQAO^Hmo(fK89B#O6J`DEYEir!#+WBF_01CwP;+)(GHB
z(vM@X&q+UD>($NX)bLcC&025Z@ZeL@0nJ8xs}vZ4KR1OD=$Pv7vE$5GAM3uJ|0i-_
z7rucxwXaMV<Y34WT^~HBwm5~gGkE;@hvvx9TXQqpX|Jm$q5gCR1P+3Ms1AFHnHTu4
zB=so)vOW-ZB8o+eyXMEJ;^J9r{tkVg$CA*^vRR_8s=RZ*2dK^OrN@vSucOEJ(c|^>
zcmq8sDpm81^oxCW-$cLOOpk4PyoDa$Pmj0K<8Aa{&)y%PU+mfYgY=6%d*4pK*t7SC
z=okC*{xJPwf8KY{FZSnsC;j>ndc2DsKT3~x)8jq#_%V9?I6dA=kN-xGpP<K2(&MM-
z@jiO|G(FgR_-E+X`|0ts^q@Z5{CRr(0zG~akHDcbxB9Z2Jw-5qoNQrtPqEt1Mleo+
zVl9)IhqO0gCw9N)+0>{gSW)oJ418XrO4)^%vYS28B1Q*j)sAAvX`_xn4G=kS*}(QH
zTNGC|Yl&DrpY#p>XH}h%hEP43hVeH0Y$f}zXZ4R_JA1XhRvT=l)6x0LmmLR%2wZK_
zhtGGB$H(xZ(1{Ql@?wwMtp@U=Pr0O9XfZGGIX03-rzXFXZ6c<m_Z7vm%>_;tY`BE3
zfmy<k;OXY9wMLx}Z(0ZDhG(gYawuYV6T91K)sRR;nF>g|i|G8drpvUKGs9$7u&nSv
zip*x^(1NBj;Tf4*D%+TnT-4=zR0%)(TtLr+Rv=rv>dC%-E8DMX#y4%`vPXK8ITb%4
z*Q)ta90YAz<CxP#$g<oD2$HvZ0Wa$l2P<I`%Y_b`)r8AIE=IO#O;{0fakP0&m~ZlO
z(M%yW58MpNOkvBBjIBXo3d{mB+k*uI&5#am5f)6rI&;=GVS&`p*+W}}kp;E33*&R>
zEyDt3L9>@^8y0jG{k77&L*C`W@E>kF3&7q!ESQmO^4=C=f%vy`u0oq&b)ZD2C4K{*
z=0#e!GO<+hw-XBnnr2-&C(*>;{0dZ+=KrP#`-OiMzXI-L;hI-#oN_MbVjCz7ox8Ot
z_NBrYIopc`6QZuCxGhGplOAM;Ydg5jSk&LhCK@<-q%OD`UY5*zq?A$w;QFRfptqYu
zZ+N-=MN+i%o8ei!7=XJNA#4xzhAHE*JvO0jWuc|MGN>k*@|Xz3SMVyxE}&kSB*9g*
zkZMuEBAnZ31!|mIV>8h*vv{*W*)|LPONW;kTcZ|7aJZ~?H#xAAI-x$09#=V4k}3;4
z;K<a(O+~|(BRxknnR1OQGAa*-`f3uhA(_J$!tETZt7w0LlY8g%JA>W|Om|9fZ5LPY
z@M4xrRK5fcW{!;x*{b1n1go5~LP|T_S<1qMpqDjriq_`2(-4lfx`yyT#$?E=rNS1B
zvm2Qa38E{AMbPhd5Fa41lqwtE4j3~L_6dV(CoM4_u@;1>B?(0%oA)q1zE4+2*~>c>
zv4xGYfzY@)4Mfp!h&9A-TUFauUJcTMa)(L>lV>uMYh+slKReX7_`5~Fxo|Iriesg7
za+TxHZ|=tG<7&;WS)v}kOcWg=&3XD0_lf3V`kBPX;MPr|fugo{?@X#}Hh*Pc%eabh
zSk+x{1fH_^c)=w3P#wFtg0h$?Qa3D_%rh6?Dtaeso9C3~UG(eO^thWI_t4`xcnoB8
zMYn^_A@p_c_+o9X?$&#H?0G`-8gmyjn=$jb)99ek@D(nUp3CIv`j|RNtNmN4c3E|J
zIsM{l|2F!?)&2^ub|JWVE0SFBs~~1JSi|>uT6_N)*Jh3dZyRmcKSf<`E?jTQuxu}u
z3-2};zAj}bdA1ZGP0C>fYz*w7lzp@LL%!H3Lom+L5_7Tn!(*{wT`_K2uBxX;l49I1
zTC@(a?lrGZisp#dnhUp%GF7jQABV;%gr)oG-R@K__?Mas8&nx~Q%d^2vqTiUoEn9u
zk><DIF&HjPraY{iQ^>}(hB_U>Gt4WAj&-H^gox$bXf9^u(JIZ}2!Hu@lMACRG#`_I
zswX^gi3TP~+T%@ofPbI)Sh6DR0g5Kwzou6lDM|l2^ReVYh@_DZDb4KL%wIgo&PW9g
z$Y*ZL?90q2TIc><=3{}ExXS#6<J4oRo6Mg#4kfakS$-|B6pM|AGQ*cmah6;SoX;kQ
zu1GD)Hbn9WLI(~U;fY5~2f!kU#U*+?z%p3aud+(LUmXRUBlC%EE|+S06oReE@syzA
z7zn4|D6pb!xY1C9+i~)|X+|qJ_2o*UpjUgehTs5uDrK}H?12PiRe`sA^jzdnic=!X
zMM8)cYz`;sd@-S*2#c}OsF)4Aw=`LuRq+v`7atCZ@HSqwhUZ|BU}F75L`@J|EKvhj
zYqr9cACsGeqS)sOXVhkRXL%Ylp^MF)B0#g<7de0pE5U6);n9&#K5P3gQc7uFLfe2P
zE5h0iET2<v3*H<zhBVvB_F!b!)48n46+K9;M^q3h;pCTJ`4*6|O`}Uj&Zss!-nL=n
zbY&x~+S`XE2L+WQu#G78#`0?;e>*X9fa@X^Z7Y`S-=l0V3Kz5d;^%HN3U9Z3teow}
z$O&H8W8Ai*-WoGoHM#7uE*X@ak<7&!ZksXIN^$Ha%zFk+V^@uA?J!KLUB_@7X#ic|
z%@$U-XMzC<CS=_no@>+%y>YAewRqz;C7y2E2JYwxS9`b+h9FlQhaS=O&#N8Yo#nec
zbo>uQo#zxp)>=Z~l7%EJNT<#96MJ;2c>o4(y~jgQwe)X_a;1OsHixbO605=3KLZ)h
zxJ(T=fpd}#>hY=o=MHFP24K1#W_Y{WAx!4etYYlm*{dDrxJEjt5!~mk$_QK@SgMqh
zFJ|-neQ2t}=24r}*Q!deHY0bd?NjS7*f1RoBX#$1Z>}P^CbnK1%eZrQ^@%F+d45CT
zlntE{X|(%WH3YoF=Bf(mLdQJq7&cQ})}$Y#w|OR4{f0`kfUDUyw;n`Tv>ED4R6mhB
z6(KY_4jNcpPGZn=SFkQ2v3suS0M2IbVEAC1dYM%W+oG2Etf+kl5@W-&)c8yOGQ7w5
zZa5cjCmR)&#rmTISS{-N$e+9cm+9Ui-r=WepC?CyU;$*{y`b?STmXb$+uEW7K;)y&
z9Mi9AXDx9+KZAP;Js<>F3SO*^&Cs8cfkXS?)FC2mWno6>sv(N%lyY(&bAly=PSURQ
z!*lzCWljwG5HOcE-+`MZN(iiU9{(0d8MRM(68$=v9;e_DEf7;h;qdJ2Tsnu^ma_>>
z;LM#7i8*t2Wbesj-JZ5p2f35F#v@t6#VNUvD*f2un8^<uzu_i>USoNG;4(brF>{XL
zHgPP|b4F$bsUlrS!m6@Jk=furG*7qd_2WByCa<A+L95*?I3hgKguJ7;q{AS?O)@b$
z8qGLOH5fh+Cpoi=Uob_ER26bg<VdX<HF5=pubd<Qi`5e?qSLNIzF?S&xkEP}A>H-q
zmPW&4kA1#ut%m=~Vj6<_EvKJw;tBNM@Ek^7YhdTEi(8EnudwkOmSf4ML(O7%cvkWc
zo1hwFB});w5+*Fvv$|g@*Pl&-oY$^*(=V3m_pn@#T}L`@v5@B8!L-Q5U9vtmD=G;T
zY3ogAmK-c`pe0;=Gg_H|Kh+f7O;PXVfJT~!x;So4_ifPiw`d2thdH@fhu!Y_(tkCY
zxhG;SC66d`ABUPMa%yF}*yD>Lu(H>rQV$7oyTC>4IdQ@^&|DQ`&oDlqq&#g8ueB%Y
ztkodh&>gw06QZp=!>WBuo8(`hsts4YlB#)*`Db!XtMbD=KM4vDTQ|)C3GqA^@!|$)
z(uRsrm$K>tQ~uKE%}`zK;O;Drcc6-9BqWQXk%FzpvEAewC%JaVE`if=10@*GU=i64
zPJs@N*Ca{TG+vBmUKqDa(Tq}XIKnDN^X1fwlQw$M7)44H;9y7d9Q$=xs$!8w0*kZ`
zxO`)WN>1K|dh}p8ewo5uZ{#v5wIhP_am1f$4b!>wF&J`5<g*p=O%U!Yst^$;G@=)&
zK`CB^k_g$Y>pBpgTTh6NJ8%HY46o*+q-?R;)uUhRcHF05>~_3Gzeut-kKr*ehi=Sv
z#4>EKRl3%&SD}~(OZ>~)%9Xpo;T$H|0kHJVR^lXLFe?#9afa9Uqc!w3eJFM^`9q$b
zlTsuamaW9dum;(=Oj#IHmqwg+%>*!-g6@`XZ3j4|VhLY=>|FANlVb(h4S|?#b+L08
z_&T_iE`sdJj9J^~LbsJ!LtEyeFUf4yHhJ9FtiPe=ul0ML#Lfz`E5EdMuF<M2Sdx2~
z#NK^wW_#OP^!RX2{1un8^QwS4lsM)b&P8gPkw_7pFk?$w1lg6;^0%?YU<m<~46-XB
zM%k_wL3Sl<?iRI>95R2-wzP1;kh$ZwqV@D<q>%tZugN%}zdL(~+#8%Wayq+}!AM$b
z*2w7xX-++9h=_)1nasyD7KGbI`rSNks#(&(4B3>!T=5;rlr=|SQP#!UF$_Bj|F&FP
zIZ8*<_?|I2x4RU*Y}kZVJ+t0k<yK&|X1BA^!5@GY57r8^jDJn_)(0)bE5y?G6zo@R
z;Z)*{G8>c*95Ylba?W~hQGL7v&ix2>h=^?gYJqI~lFtX_OIYuA5V&iM5;ngGC8kvI
znqQ_IWlNSbF9w<%ZI?89qKyW~<uu#>SB1(M!Uv3|VQYRlVk^zC-s=py=AKz$UP(KB
znng!A8uYh1E!>;5ikrx{8uW{;SUKZA#n9tFN6e%jBZ^*q<tSaeYKfT~$zgKR@YgGG
zBQ<;0)jUQuGow&i8-;_C3out;^%Pa=W{)ybik5US5{s#=E@w;wdM_crG5};$%-j8m
zBZG8M%@i6;ICGY&DTIDFdq~}kEU2Y&#^=y#r%+PE>}3UN34fifehNLJ5WrSJ#X!r>
zy#`@tsi7ipMs`j|6^-n)8S1D|WW!XA;11pjDnw=$NWNMc*$;VsFL^hpn5n`X=BlSc
z!3ksJsHnm%H;fRcri#^A5R8?ipr>Skzz%x3xuTDclKF>uMuN3Z;_zVub*8$YzQQcJ
zR$@Nq@(66B>I%yj(Jwyha|FKzfjirsSu0GzJMbd*+uVl8!U%XI6`Y+GduTC);*Pt-
zoS7De$td!pos~Y4KOKhal=eZB2*zQ5IC)09NByBsk1g&G1=Dbtu3pNB{nD5UN)G|)
zK|tDFa$4Cx6D<PYm7`E_7Frx;EQhYPnTpOpk4Yrw?DNbheI~7bK2v14x6eF}2(0P@
z&zQ5$BaQ8pc||MkjPpq2I7G^ov^*PRpKU%oUhAX<49LxbGtCpJTif%J);lACmN>(_
zFiL%uI=d{~O=5;SsgdrPWpQFD7RHu$Lvd~+7KN-yFqI@XVxVCP^yr`nPbIFe&J-+H
z@~_CbKEDl1U6XS#A1mbMy5M?ILwO?`79zR#s~o|4ZW54i_~)q36GeopMoaNnAbaDA
zz1rw0wEHJf;`*ZsY<3c05ZSmFXoB3~l51{?0{IbFaTx8?xO}ca!A_*l05)ymiF9lr
zvQc-E6`8b3fx;43k!jfYiFD`&Vta5aPlh?#YiGELj1PETG1x>(Y7AtPVQrEltv$(d
zQXKlpHG%FP6A@9p`sx6oFPS?L{>h!D;|4<lSbtn$PNWah-8BksBJu?HSmDeJOzxBG
z$vxdD6A__MJSIHt0x=cJ6X`g8Y*pZB&*Z8Qp5_W|B9j}A_3)Jt8=cNk1zWIykLlej
zoAnxR&m0(EJHXu*$7!iipc}`qh53<?CQv}?BXk--9zHEE3Xax~Gt>kM*zI(QsR_(E
zK%z1d+5|=cMwPx^1b`Em#0Fhh|Jf020$_Yk9ydY3DlVdrQH2%kMc@K8oFT2&j@25t
z#1GMpk)^<193t#o@0nQ2vQNtu7?HyST+;?QOE*s=??5(ApH9E%9Cz~!`bDR>n`hE5
zI=0+Ai+-_*`E2|O92_YgCn#Wmf%6EM4q!Depa&DXkVzEmeTIwZTCtOhuP!z;o=m^E
z7?&^&5v(Vw7-2gs+C8E76hrTFrbmyb(u1#xy^{Y%mAVtZ0`a`#cnDaGxJ9ofc&^ZE
z=oeS$wT9$%OmYN;hq|Y2EfVK+n>P@V8x5tKn39Oy6H_?<Ky9-^sCx|bGZ=b=!QHFq
zbFk)!Z1Wa^y_Fuf(c_u);2!cU{0iI-7*PQ332eU09>8Klwl=_ive{aPQ0c-3RKvoT
zbhZZEOt7HFG|WegrFWmm&E3P1NkWW?JX<~7%4LhO;w(b)J;T=!Tq@?KS$oq>Ium7O
zN_OjMtC$YRvCoJSU&nB(KwpaLETHt>;%J8HKEQ#Rr4-4RF-f5kWs=kYnO78xJ+~52
z&NanizrGmAx~jPO9bB*xK&%^yo8QeK!dRHW2~C;>EnysUoUSAGuU!IZCf8R<>^Ws+
zEAUs^o!{M4^4e#qVBDt}f__T{)oF55bqc>>w}>;r*f%nadS<TdvG}J5mC={3lbJ}a
zTEN=KrvcIY3wnHp9)C%XzoN%q)8n)B_#1kBjvjwYkH5ns;Fg$46w9u9&Ds}7l_|)7
z1W?}`RWuSg3y7*(1ebF^wp5iRal}Gc_6zbR9f_~>G<qg)B$bZt(=QUdf^l~c2^&WN
z>jh&;=n4gJ!4Xn0W%(td))UHlCf)FuDX*-@5O4>@WR>+8`eQ2VQ6xe%N+|1bH1Z=P
zG{;od<9L{;vK~ir63TiEr4g0&80hhp^$04E0EGj%K)z2wb)Ro^>j-R8ufQl})OeCE
zH~1O?77!%EjrXvv&SZGr?ux)LdkSJgkpzSOwl@|M$#S(7nueg5NPZhY&afhv<zQP}
z1Nd0`){4V~0y%N3722ElkXkd@j7MXtg`tXR8gjmL+ttczJJSj^_y4l@CIE6A)!n#!
zN?Kh$@P)BG7PhsPS0h`15BXxP%d(}FtX<o(g*=+w>D`^y?96%&Neei{oHjJ#TJAF(
z2@o(4LI{MMBmt6O;7jsB{`n^)oCzO+K)4b@^1b)!=&I_f?&_Wu*!iS{cI|Zat6#l(
z_3G8Ds#n%RC-r@_iDAw9k+y|4TksH|B|fQb52jwT^o_F#BQ=3FD++10Ko!|8Yuiw+
z*DH0WtZmbT{&~tKS=$yo{bRrVv8Aj|O6ttEHMTdHG*e~6Soe?Uii!Lg=0=i)V$7f@
zcE?~}Qntjpxnl>W+eR6?VQoh+80+oCbr+&_Gj00kmcc$_53J1%nO4TaZGW{b2Bem(
z>_(NXklcYfgYHe`mY}^(v9-6s0&IG94}<n1#eP@Y2C_7c#Cr&ZaHni_O@vp@=r1wx
zj9sp_#U?-0G<D3p<fl0Ol)bG9f0s8m+}m37(_{>Qr_YYoygfB;Kvs<3X4bsD(pHM>
zoirk3>|?d<p3NgZ#jfS)mznAm2WlDRrI%yls%?8i8iR7}S+y<Zr4dKFP2aAXaA9Qh
zB(nU}TXhn?7Y_o7O{#7q)v9u1!8FZ^E1biH^l+-iNiwjBL?7qZT~}^J9DkAFu5z2X
zF$ru%luY+zA<k=cXRXfW1k!g;k8lnR@slfXPj9<EL962Kk&$`F_z`hhE5lD4tqy(X
zHumsbbo#f6n-^pYCBLNu)r=eR9KL%-fZvZgj^VrQO3PQ)Q~2R`aXk=Eo`>JV2c9So
z{|z3!C@=Bk0r>DoNR5d9SQ7sU5qGzd9l4y5Vd#MHr;_e}O1eKIy0%4irfOLGZHAS#
zO|1s(er~l{$mCiLW}?vBFoFPWC>o5GOa|3%529JkM^B*mckxxp!_7Gek4f|7Xr=x$
zM1D&w)<`KZ{jys3cI%bIXS~CacRU=}f}VVL-%EV0>j<~allU-0Rk)=;LCLBG+*5h;
zdsGGq2ZKH#$>VrROqt!Jr2i=~-*SFqbwee3y6bb*nG}8zCmdgbM<>}YT?A}SDenp^
zzNtLl2p3gigmZ!9o~ndCaOGE=P9O`a!zv+!9wb!1RYLuzeLtzYuM%s?Y2RvIEN@|U
zII|KiK3B<7gIW)RCsdG|O!{%~#HisSRK&xTIBV`Ap6cjIC`=cX6n|G@oN>|cb9*Ih
zEeAOed4~%<Q+jXRs$IyY#zM*at#Aa$4#=)x%<KdPgsjUK^U|x)MUxWLWq#Vy0HseW
z$@R7tbuaA~9D<bPCU&#243b$6x6IG=3?u^?mn<e`iaLolT!*7cgP+4K-}Q=#?cNy}
z;z|AjU96ax(-r}s%0oylU3R5nVs~B!ar<S82@@=X3JwCZFd?WUu2IY!K#dj`?!ZjN
zuYU)^7iCs`yWDz{5;-U_r8ld-Gb46TVv1`LKwBSna;IR(Br#Jx7*smb0Za|55P-YF
zY%-urXAr0-ei`<Gh5vw$@1dUfM@)SMA791C*YNRme9)fDZ$Qd&35tn4?E(y(s%Qs7
z6FbUtQq#_YE&#FkJW9e>nHQNj0h-v&mXnY;{JG8aHsqwEp8M<<lQ{C3siiW60aSu<
z+Oyv_1d4GRCeCK$WEl`s&Up6CG=3B?Q*UN4Ek4=VuZ+q=oqa=+=Q;ajk~qrQZ`I_C
zw{Jka)%v9)&TjTwF?npWUqIq;V841rJ*=77beOZl8Rs<n70dXDX1|(1Kb_fc8_I*3
z{c1byOy&X*i&ba8QlXy2?3Z2g5N5x_P4fI@VnI2Vz{;bSnO0@0N*AXt6YD&=z`$UI
z25x^fLIJR!&dY^G9JB1NM*M_jf8wJYuIx{y^>dZ|Y2WxrWq)+y)0CMrk~9fKxyf|x
z!l@^c0qqb{rU%e=05h|<#B(rQ=GZkkiRX#fN{2Ll*uK#8J4|q^Rn`q9cpH|y7zgJ`
z<+yY?`-0_>%zhPrW`M4X86~oLSC$p;B!iRQcu%%+OIjZ^ofny9JL+9HX<kV?8pgK5
z%#n7x-H9o>cI+-l^^$wtQk9G4@a@m9+4Jd)b~){O#&ENjtg!_;>kWsNh0P4sslvyv
zTwcdTkT$(|+`x#Aqv#y|5&Hi@6L!Sa$@9^L0bMInVIItcgu<n!<1Gkcu3gSZjJSLk
z1eb0#s||6NrmOeCnkAl;6@<_k5%qVoW8$D!1{dR6k$4xn$_kJDM(V}APr-Uvn_1)4
z0#rAj9S!|0{aA+F(%~yXRQ1Q;#g%#a;#9$dg)XWL??sW|S7PPG;#A#(;eC=UQzf}?
zp3L;CYQ}`JrLrnj{<u{66QuGp$9&f{mg;5$eikIheU@gwc9x>*N#^bx1}<8z2K1(w
zITD0hpafRdR!H@tOk2>2-lCtv?MR(s4cvw))HdPmn4)!|9hf4w?oLScvTX==+3Ct6
z+>Ip>jyo_#IQCHK_!z|pg<crL6k)rQ-Xh++Ak`1Hj1C*#gC$`v4EOa}Qbu<SD_98e
z`v8@cxzz09cWhs|yFsF}^}6*&r#utPwAyp9gKH837c?5m`C2iW>z&PZQXxuA>2lDP
zPPxPqpFa`nzLbP<?oH3NQ6=Fp@}_9pxl){I%2t-Xh={!{iFu-Tz4&IA#BAJ~05~Wd
z$F&k$s*BK6x5Fe{VYvtz`(qM8BythcHp~?7n8-ZKaj?YhWN&PI4RtWIZmCIZ5y?R}
zeUDA?j#XsSO=9}yU60gVI0<9Tn{3LqoZY5tz?-PwKApsn<4d209P40^Um4(WTv{Bb
zv$!;Y0P@K&WtrG?6I36#$o=59Mfc`rv{6{LX*Y@HFAeuoWd7!tdZS3yKIM_DLM(%J
zbsWT010OT^AS0@YDV%4AElg3UsaZ@Bs{0|;ixSH<xw&qE)x|={-s@qC?7c&nB3Oqp
zMG%fasuv}KwC9G9VSYXf3nVzt#uUMMfIuK`ehE`#em)0N1o4;YEt;PXLTUjZW?7-n
z#{$U;eZj&Mm}QJUBy0FE6_{V!>(Uw9r0!0aSi8w=Xi^i>1q(f#juc4;aB|9$Nilm<
zO@47ig%*RvVOm6`VY_a`W}7InBQO*03<OR`<>+7??9WHzg5N*VYM0r5bF5_6DS_};
zVE29ac6?A@d<Uc~H&`=-aJloL4eSy`2#m119ZUvo*!~!}3KBak!Px8Kt%8jJZ$U&E
ziB1KuGqc@!=Jbe8otSQQf~NR$NjnIGiEdEiKM=O)WIU+=4tNI_@}ird0~B=dXT&}Z
z<fW({zaYluJb01V(raTotfM3%z$LtxwhuS&-gRh$I5nhfoaKYai*b9`OYjll<E8jG
zijSA!<K_6EQ!B5))U`Lm<CU0t6+Y(i@oId$1|P4*$LsL15%awsQ#WAh4VZez4e)p)
zrf$aH-h`=5n0hm&zKbVg-h!#E_}g1C^)`I`3O?SBk9XkXo%nbcK7JJ+zlM)@<KsQ}
zcrQMF9Ut$*$NTXyjPxGC)CchKL414&AHRW*598xE;bDnRmB~%K^_-u-8Vo;*lztl@
zAH&DT@$ozO_yj(F7azZekKf0~AK>E;@$pCaAUvPM6!peGhLq(x^GrN!f`xq!Q7Na&
z2E$Jw@lWI9Gx+#3D#0;p*5zN5pSXnh$Qmd9SxNj+BA!?o%z{sY2!&EQE|Wqg9RoA%
zqNV8&DrDGFEBiN@8ss98r}Gj=OISlBXpB0}={ck%>-)SX_KFvi>+GIm!t7BSIbqH?
z#h2KZ%lNH2-&hZ$+van4-W<@g2}`$|hg-x7&FXV=$YvHCm&47&%VNW3^R^Xal;UbD
zPC@wD9Ak49ces<y!%t*m22;;;I}vD8Hkh!0NFMXg41%hA&BOiSqGtHk5=$O_odqk3
zwlA63Xd)NvbV$eIRZE<b^g|^zgw>_y;d!+&Eh`*T!G2T3ndaf`vDF~eljczSS^Xr%
zjpmpqWKs0<p?OT=EG!_?#_B-J*^~71EUMEhd1XVo%o>!$oK30?b}BAs_cVxRossL0
zy7k0d&fZ`E*1cKe3^LEp<s3qd0*upiIYl)=Okcz!Da$xpcaMF_>*^dviO1rk-GROk
zE%>=Ru@^}=amu;6krj?ARm@+@PT?iC0_G~pekO0UFK*&+NS1KogkDaA#E=_4FeZcf
z=BY6b^EovJ&0U8Z^(l<#srf!=g<vCIs6lG^dQ9Dbj~nrE6FzQ+N6%-uTT}SxZ4o7s
zfls}XrnR>qRdTg#!W214He-sMBwH{=SA1-PR4+=LKuzbBQB~1XX_$wdv%*jsWZs<H
zurTs9+)jm!<r5pNFgVZ$w;+HW1veyA>u`cYu@ncx5iDsJK6c~d4t(sv$0$C=@Np+T
z?!w32_#jw&A=Qf#k+1Ub0Aiv{;4azOR<i?fa)RbmcN(|057J((@EKUvetg^~ORZ3;
zIaPqh42;Brr5>)fYjn<)7YyABHad3ka;3p=91EPlM-3kVK8Poin4<bmL8=$^gM{ox
zY<b=&_1Y130+fMcoFF72mP)O25L46=4T)rikUXOg0f7xCuq5mSWmZDDUjk?ofa<XX
zP?=;`ckrg^Nfd`9io=8=@5DgS2GsK2YR6+`PE|WuOC=r~EL3vn*-t-woD*Lgb{@g*
zb)2%%dI;Q?0vD@+lA5d^A;$bO_zc;i0Li~}`scKdMW`RRP>bs!vHUU^eg_?t-^Itj
z;^TYx_&z@V4IY;8N9npotj#z)1sFT24a+wk{yB%ERH0xA6E<k?4$4Q66tTe^9zZ2C
zTR9S5L{Xmj-r?y{zTcf0W6Uvfr_!*+8XLD)oJlXrSG*k7&q`JJ6<i|yDn7o3kFVq7
z8~FGpK1fIZ6Qp{P!{RCiNLie649UglCYGqef05Pr4pk!&Mumxgy9)6dge32QKZJ~E
z))bjHSK~qp`9`n76s;9qhbc02uE!J^IyYj9R)cQB6s-n59aFSYa|@<uU1&3=XkBOv
zrf6Mg8>D)XOP*a=tF;<MNW++jplx!04@bzf?;Xx2C$G5n!ys(}g2R>KBmp;if&QGq
zs1j&s)q+Fy30lK+Od{NbK(ALiU=dT0IEcp3=*__G-1W&yGYD$1^a}iOC|x&AUcp|o
zhZmS~nxxlhd&(HL-<|lli`vg&*1759yo7ZPIJAs&6R@z5A$nOParelw?!^>!)O}Q{
zV+xaADv3TbVAy?DMDRW!*l|rpoCL|)h=T%+gSbaSAz3AojxC-}k`<(hS63+z2U$0(
z$?6AGeTSpXNm^1yr7FR4;?t6Ny)f}^&|ywog?Iz$&Pd`-BJS{#IFWNBparLkX5c>9
zcBR^^RXVVSfkA}vutIo0me7VrCmBIJL+_cyvA;NWGdlqyYh!d1jSe9^2y=!hW|or{
zJuKO9gxKI3g<V>YBULLt$ZGy9r2K4rJV2E5>eFU#rV<=MqaLD6;)Ldxk=BFwpziTp
zqM8?PWqP*Kfvap``5JUo4=+~@Uw|}U2#@@|4OGNzi%m-<VFF7uA4bA2#>Y$W5#i&d
z@UYzYm?02!Ux>|8$va9=3{R%vagP%Umdj{zH95ou1@47)h(Jsu1!@MQleh1~bty-r
zgfd!4#xeVG%6a8*yE+?x?%3zz{G4<zHQ5`mdEST*YP>gL>dp9g3p|`VH=KkkEeOUT
zsOM%foV5Y7o_&v#+D40A1e~|MNq4t;U5Ee-=k4eQ=iF8Ss!lem*Vp`W_dOysjvARH
z2V2{*Wa#E<z0>Ih70_Z$U<>H9Nel{ugprzKBet8`Om%J?!y8Yrz9sHwL|VByO|@%7
zTTwJ0B$1mz=)yF@9y=!0?kW+CSn->~Qk1v(r{`vaHta>N!e$&`HB+8YyTR}y*g1X+
zA0NfXZ{y=*`1m+Jeg_|)z{lrsm+S9h>i6*R`}iOg{s)k9&WW8hgw~?sKZ$hySQ7ga
zBIcSYyObHa2cU9z<{i2S(fu@%{EVdgXGAx*G-!0$07k1OpOu6kCBk`!b#>ZIVa|dh
z7z(}#_ry}~&TGfu{a6#=n5uMUsxXjY;540L_?L+Iui(*fxz?RcMd}aLSEn1`P|LRi
zei11WH@<`^l7+v4RQ^eq$_=2OTexAvV*{$+05NDy1hAbAEV&jCVwnD(DxJ5(0L{a#
zcB5ve`Hw`?v2nx6NKy<!X%2>8L$-Y#AK$>oH{oG9w4Y^D5K$Nzu)aj`TR_JWE!&5V
zr139ABX|A4s0b<$fdhL=ZcC!l0UHeGY16^PLE4grWLO)1pYY_hO)Bjc#EesQaZK@t
zL@}=`Lm685^yVi(C;WGO{0BaMijV)q2la=aL8=!e!ue*j1!$cbZe+%;M$?={crhv!
zassBPkP|UQg`5N_%T4YXh48B2jnv#M;l}x?NSLPJi!rqXA4~B;00uBc0G45j04#@8
zFUqa7n_xpgdrwun(@g8-?W3?>QG-jE+qi?3cYT^=<R0}o!;u%djyn_UNILEbm?B=B
zg(>32*_a{@oCB#|GGJ&%6JV^GE-iMn>P?*Ik$6&ZPXvPbH8mZ0m;##~Cp6S~5aESH
zE3dzR&k5P=U;@^tP#mY%h8H8r)%bWaJ}$w>rTDlEA8YV2h>y$hu@)a!;A03M>+nH-
zo-(G^<Ks$r6qriQ_Bzuxb2jh9gXn;pY`Qm2{q1Rp>1upjgO6+RK}~!er1JMNMY8l>
zb^{WoUUnm<NV0Ci6iL?2n4(_xbW9O|VMtkG3uVX{?_T6KfXVo5eG(TIqud55Iz~G)
z(BrybKY|w!SO$Uy&OuQ5Bxz_KZUu;z%iGf-65D^yZecOS;Cq>%7q*&Z+}1^*wcBiF
zQ{F_&)g5{}7Jv@}29xC3B4Mi)pv87Qp}7Xj;;bCv2D-eX64+hM4u)cdcX*a&j8=(i
zWAeVAXvqv%J76<fF7SXrcyNJ7bZTb~G}_F3v_eME!8h~KDP4%P1K*t~f2SKj*k>eg
z(WRro{XIOrvu~^i{+oT20Tk8#p^{8O8L{3gz=4Xg@yoN;;i<O*cgzb*IPfL-&ZuiR
zVEYmKlz$b)B4+U_R-hS3g0O|7CGxN69Ac)BuT601349K~<#}59@DKymS}a@;HV?Fj
z=-vh3s+V!mh5$0=#M(gtg52ZZpy>%t)9_9jV!{g6_+VqhW)P4?Pl3zF7f+(`p|Jv0
zB;`08Ow`f4-h~+$*)xOTg(!HQ03PeLYR!TNfwa?OpjN`+RG_Uu<8<K8qe{Id?in78
zPHK&Z^U+y%Hendp2VF>2aBK#`qjB->C7iA4#b<AgSjShW)9H-E<n|z#iPUEXC@**c
zVSd23K12hQ1D5KZ8NhK!z&IbB#_Trbb~*lHIoiyOMja1NQf~YagC+ur8j=8IIv=g-
zwtLM98vNAXnYIxpEydI(y$1X(%ngAej<qC5w4Ltlc-D`Wc;AUBvUJ`BsW`+@I%#$6
zWc30rSju^~EayF#+ItEdB7Cnb@z<$Dmy#o2N;{Z>9bz4vBM@&rt%ExqO`L;fnw1cq
z0xE_^#BedDaQq9G;sXc0a2Y;`r^_J~N8~qongfkFVu!1+3>=fg)9^tEPRA4>I0I9J
z;7mxxdlmafpbe<u*;vjw_#k0AR~CC76)Uz#^iiyRV7$Bk_H_=BG=78^$x4-Er7osQ
zEr4vLja=a+vgAu;$(J3QlJ(Igyj+&NR+fAPl`J*{^wE%(5g;sMG3#YnS5jHRtLs-*
zwz?O`di>}KwNKU&?Y0;Up9Z*u)?dyxPgI)V(+A_}WHck^!)q`HY4z~BIC*`XypfWq
z;lrEbBw52ksf&{<`EZg>R{$;u6mdu+9~4jx;7^AsD*m^Dd_h;*LSqx#uKlp_y|upM
z)$*H!DqDdm8n0Gjigf8JNcH4q0zUx0C7&TnK2w(b1X=P~RI*q|@hh3<vJK1U%EHc*
zMLdy;$Tt|94W4ESTp-K5P?mWSm3c8&=CtruuJvYS=E|GtsqSHe&=k0k27T_b?O^Fw
z$6;<v%jxB26NCD)HMa6@k%HOwi48EQP9c9E{C6C#GQp`gn^x1WLnZnTWL<tZs83D9
zI&Uyhos%E+q6?hZxXUtHJOQ3}Fm30fC7WRX7~-1mMCZD`qdxH<sDSXf_;?;Zo{x_g
z;Nyk(co9Axf=4g9Od}&#CLy?Qf<K`*(befhdBwt)AdCne9fJ!djC@sE>neS?R=p)?
z)PqU&AV?%4oigE2idM5@yklB(r~sB|O?C-jQ?g+EP6o&1MMLXiYD0pBU!GQYtJkcl
zwHu`Nmm<0Z&fdaYR*~9p1#Eqm?@tQAS1tY`F%f@6OQ2_rBY7FknT;(Y;MD-ngqZSJ
zxyVszOeBRDolKv1qDwO?C74vo%1OdPQvWL`UvJ09JMe)u4d02Wcj4n#@j*S|*D&>N
ze7px9JvW#g!+S>4xI5FrFgHDL%L$#Zv<7>m0^AKDe>2!ht0nE$A;H$@beNyPitj`#
zn4GLn)G=<3AV<K(A3e7uaGWXpc8VH(DtP@Q7q2l_%;(=k)_nvYzlD#F;^VjRLEQZq
zraq33-@(Tx@bSC&pjP-jNTqSSIy3I(xW*5iughT9|IBQKZbyju_XUh&&VrFgr`Ce;
zI_hg4oLm!UH-8XNHh&;j%%4xo+I|L8#Fan8)Su(yFYtlD!p~yrQG9$39=%jG10DdC
zFVPjQrUEs?@@mo$Jvk%4sN#j``A{J_HQWf6t)X<PW|36~=^RH`c7ahJLKVG9*-|0y
zF(K*-(=A}C`szi_dI|cT^<vpG>qXgO`T8b`4E3ge!W7BZKa&JGn+B`BTWuQOdPk?g
z{_}2CcByE|;aUY;zx9^<i8{NrSEULmz?qn?!p6Er<xo91jLnud)FJ#vbq+Q?iQM88
zjco|R+!L)1oW=p4*G#=pZP)1}US<)n{)_?1m~Sg`3MAuaX2VgoxZ(ayHrx+n!~GCb
zKf=e4@$nOU{5w8Kh5iSmQq<+>w#{j+iaJ^hAEc(`A1FQY4-_Ez2SSm5AR74x0<t!<
zvo|n!W-E(>Oo?`5+4V+V@6tz{wyljb3ecO_srjh@(ZbG=5p%^2Jg^-ajz-31n4*z!
zIi{$$uD}%a)|Hr|Mqh;~YV^}EMU8$sq*5CFp4L>|vnxku;#LA&&ZtgI^k#an@(F7;
zvPHFq_ZttX@Jy~?u9zzqW6g*It1<Otd|ZML;>D$q>P7B0Yt2sAtJ|~+!-8q`2<`M2
zqj@wiQJwAfXc8xW>qSFO;gLjjMeUp$jPeiIEl!nb^dMI$QsX6AWe8?nl#!A6wGP=*
z#>aYmTuHo1GnShwW1_{d%p6oFrUeT+srnAT)$8wU1yE$oGR$#sQ4F_<1K7yfb11W@
zwzKD8E+7+Lj}^TE9w}f)T8CQO+Es0;<DP;jx%KKf87aRIsGYM8krNr<B!_BWYnL=u
z&19RGNDm848y{(y)(TC7$8u-Z77Av=KNOxO8;Q1Y_6nd#nMKjgUcobBwKuqW4)W#h
zQGU#4YdgSQ9dWQ&XJ<}+CQeN_N1Nm)@-RV;Vnn6W+EK3%R^Msoz}Vxgyj%4dABn0v
zC-i11vf|!Fh-y3CC`J)jX>mY_83|Qr=)>|r7wi{<rcO4A+W2BeXrAu{D@4@vJl{&L
z6e(CRNEdXkN+YlIxgCB6lSO2YF=y2IZ|2igrQR>~yaMhMwSKAPL0CG8IkZ~rrc0P@
zATW3NWi%X~uN5296|l~>zXwQc(&JHF&j1f~Ya&Or(N-O~5=|hp>ul906fs$2Xw|D2
zXR}vPzK1HRp!5NNa=~n<IRzIkcVYJg9IVCPh%e~sjn1GQOw0v7Mj7^ltvChaE_h9>
zFP8*IW-HZ0aE8(Tei?sX>D=$ivO;93)h8!inU=!-%Ssv$^2J{zEnDWB-(dXxl=$z{
z&=mwT#CDra_LNV}R|=K0jIW*o^=d2bI`l%?jcawj6)kQ;+?;mnh_KS4vnCq#*~(O-
zH4c_KR<i=zF~O_2Tx|<*y6U4UYfv!iyLJEe3||L*$l6%p8=P<6<b3lM=bN`V-@F~)
zaEEF5PW6vpRsVRm`p0|KKi;SQ;m7mvL--B%{CybHCT|nUk(}Sy>2I++*~4P)W~RyB
z3zE_^oOsZli83|q!9IOIeA-FseB)E0DRh5cqKH^27OPEsu_qfWJ5%wwYc@<9W4mUd
zJw?;h+24u*S{y6&EmIE8c?%R=)NQ;>pukmQasZ6SevRU^&78;RlLIrSfS}K#j9gft
zP6i5rIf!saN~4&MfZqfevUw1hgnf&IQy)EadYV`oi+O5J0bXCcRC2`f=YnJzif72U
z(IhD`VDdZyfNm2IZ-#+E$T!^>hgail%@E#ORu6SE*`8bHj_#ASeIeHlb(Y(*iu&mD
zQ%g>)nXb~BFoz}?6jNA@fx4^KmRGNpZtx`Y3!u9GN#sMAx}6BiIOF9V)@7iu*un}E
z>d%u<>v9s8CE)6EBHzbiS`30I#GpZD%b*!8&f9R|e2A>#XTWG@nhr|&rgv7iP?e0j
zUsmrmN;a$4-APpU#p<mS6ScRFd>^;Y!tu^rje{=6d?v#pq0Tn--jK{IDxmj<L_X{d
zx{cQ%zhharAC)o(L#wY)_tGRj*E_N9xHmE<U7fnpF|vy)8f~cIr-rh^)(5X9aAu}F
z0<6~EXu5vXbLlnHmk8CI0;fJCDLJqaOqB!&ckFus!v5F7b=iTHDp{6RCDxh$kc>Z_
z$MGkkZl)IFB;$V4KWe?fu)St!RJLifV#{8v!apUooft#o(oFrO_a6<IuEF#XPs?c4
zVJU=lEvCz!N?5dyE_+IjxP^-GEIwC=A)HWt%4~iJTis!5$$u0VP{emwB`(_9%q{a&
zGMixEN^52(PC$1eeZF&6wJQdr<(vc_Ih+=^JmHanNLaKO11SLsX1kmE!yvD~80m)&
z92bCnrx^T29@hi?u&p1G6$}GYhl67k+FW*hAGAnbw%!KD{5E{hNsL<|)sxXqQX3Gw
zDj>QPYJtRrNeOH@JXA$3DB=dB(vW<q)MwxbBjXc(y~U?q3IZs0rpk^26M_nyvZKq;
z27QlIMN6l_sY5V?YsIreq`^3J6>-`3RO4P?KX)|5XVFqQxEZ?YCc2`m)!y1_kM`hz
z)}cCt)x#IihoL`FpeSuCqNE4`E|j)|mV?a>xMLJrK#bQiZ%j~`5R%Y5ipPE_G|cBA
z6`QM;Pi4On)<ZH6g?=DyQ^9kuRc*O2k==DYj4)*qAMhjX>UVO)PGpAN<2(pMY97A3
zJQ71|f)2l?MU2*o_ZZ;~Lp=ttV^V~>U>C87fhO$}p_;MT#I((t-?&UgSVY#m_@jsd
z?iL0FZn+ReXy*iF9@*V-i?Fb8EuUfVIO<A2L}B|8K7NdkpOC~kTbrCW;NZllI1p2s
zodNhMe)ymG_!&O_3m-)6zabUxa!#N~y_iId@QJTdt{xF9hKX<dE*Ck+RFa!npC>+9
zd*;>V7rf-e2nuS$xIHt0ReYMFLlVs*zTrT%!JZ93rY=dybCQJADPjeTXX%s^d9Aud
z%toD`sk$QkB1+4b;L(w5tIiK?x+b$iiq7%LD|Amnx0sdZIw+w~%zSQLBpjg5`YNBQ
zI>{+IoB0@p-x<0|xXMz?!k1D<WeS@^S0y?wFH-c0a=OkEvkNECtP7dM3cZs+5gjIc
z&Q7AKx=bw2JHJlRX<`Z6`K_OB6O(8c8MlrT!CzRw5oQf)$DSJX@$#sMK``2#Yv9h0
zW^h=ysPla;O(J<Sd9LrL0tgX_QL37FW{vitlfiZa*Vnm-%|Kv9JjadZfS4cTB7l*D
z+Kd~;m<b09`}rKATe8n@EK+@LHTKh*FcX+&-#@(&0tnsJn+PE0EhBPsAtJRPn4M_N
z&h=A~_<PYU^9TYSDn%g@TUzpb%g$23$6`><@v*nyZc<|Alu^CIgcPrfhGM;Vu6<vH
z*cK4@nMczpj)A6-6ffl#pLW+UU_d2U;HtN9cYAldMO6_eaZ@>{k-=ui3grJ%%yz^G
zz{-s0h-L2BognWG1}&PbV@!(ZDy^={2NcAV&&k^X8_Otm0)C}Smhu(2wPw2&_8H~I
zgS@N=G2e)US-~-41miNel7dExYf23E>Nm_NoNmpuPz~c&|5u-|=!8Z152N&~V&$Ww
zi3JOZSmYyKVRAs}#X{5`wjX*N!YhF<N|*?&PzZE61A(SUT<5ye5K02$KJhaHdT-%b
zXQ~I`B*JvifSuTgg4v^B`^1-(d>N0c{R2L}hxX||V(KgS_$ofWhL5k~1964lfK<GH
zKtshXRuG`v7orF|Mx)2W8sI4IzQ`q<*e0OWAQhhv)JS2Jf($y}l2!aaR7K&#(CEyz
zA>0YfwlOL*$Y40L2?}q*qa#n|u>8DPU?_DomatJ1*()$7nuX~+y0lcM3OF>y&!k&<
zkd>vKgZ0^2a=b|Ak1*iUXhY0o)nl`S(}sOegDi>-ZOOu*SVK%?*;!-wVW@<nxgP>|
zbs}NaCrVN(!C3*<@WB04?CvAE|HSqkS7I7GVV~Rbx2tn{UBsNsmYHrb)JW@^KxsFq
z!X(6;n#TKsntBPs0c9%zzd5L9Peb2rL&sEGA`MTh<n(XET#w;FL9ehRIPu`jRI}Ea
zf&KOr0)ja!{+QzIiKUdpXQX$L4C>jXlXcj)r^sAK?)XT2I9`Pt;Tsr62=_cbf-!PF
z1x1CQ#>Z#y@n`t>b9lt3N)uEdYTaRQe1kdLuF4ZpI?t6l3P&>Y0t5SLOc~b36easz
zJ9g~c-kc2D7?^HP(1AsSyWy6P?HKqO9i9;R7`(q7n+fMn1qKq38g7Z|c3>_dv9n&c
z=5r7c_ONcv`x6N6w;7`T-!ng~%@Fmc0jiK@B*L?wkeG}{h4B)bqdwk)X%bN?+9p6L
zZj+)i21JxI{zV&Rs>`aPPjvj58c#cm%2pjV@)9=-L%{Vd%UsjH5}=vr`=V8l{=cHF
z(BQ35^m{lb+E)b}z6XB}iZ;|IJ^tVG`6^XHpS0wrA?CbA#VK|@6z!Sq01W_$Dcx6_
zv;|L#Y8e=|`yW1lNUW-gwl0PQY)$sZ4(ZV;s=hnDahd>3w^|1o6y3GtD`*p7f70y;
zVtqN^I3i}o`oz<>a8UYaO%+3q4M}c+V#p5$bM$pF<ZvY&A}^+5;bR1k1d7QZh;c2L
ztb&<_G>WNN=6ERnJvpR^gJJ`CF3G`yZ*|$D>^e+x0(hIH+xjD>S*QKeinWZ`IFsOx
zUnQmQ);W&&(Ge!=*8bR}TCRN}5(ey!{tzYY*gjiNGiUo0QQfNTmtowj65GKFN>ps)
z_9>du)a@6Lw0QeukT87vte3QZ`-Q{Xt4}h*8t$`J(kSi|j<DnUR3xg2yuCjSV5|=H
zDUGqg+-Kj{?dCpPPnyqtDmTrF?w?jX%=IY_s!iP|tE6e&=fIM*u$?m~pCMQp+Rh1(
z7Ztt|<+K}ph;XyJuj)wayRYg9BfPKba9g}Du~AI&z9d<<%==Qcv4P$fnb=Nu&J#Rj
z2F>*eO-;jgHV9?}t8K9Q1GgZyz_|;5m0^hyHV(I2v#`Pj8)d<50E=_b%(&xxVjAPS
z*WfUFum>&?YtL<&84oB%zPfQO8b5~Rfc**=WYUe8_a3`qkX!dH0C^aT8dcc%z?e!G
zZ-XzM7>Gsay+$s~K+sBT3GnT9p?S=oE0Yj2?jT$@6yVjggVEAkwhxb1#_rj@Wj;EO
zeN%2!n^QelY%4=)u;2d3EQG0Ax_z{=Y5303;T`GuR@|~<*QVPmw{98Uym|Z1^t`8T
zxogWv#VgZDI;G`A$*a)C3}8m7JXd<L@trWt7B_kAYJzcS0%ezbtyO`2(X=^XF#Oxw
zVas|-Lxg{sks-eKD?cbWLISZJY6mN|Lns{KUuTpRelaV<-(+Nnx8>(Y%cPiY6Na$u
z6^iXnGCeRnwrl71O%)KqRIv#^mw_()LRN;X*7!NFFk58)E~6H`NFT*L<(OuYu#be_
z3|dehVP}Z-(a)htqSHsW+;QiYkxg4FyNCA-@7S>=U8OxvnozgqJ)1Bd9$=)1$3}hX
zQ-Bs8z7PZL9&IG_B(@c|84SZNgF9}&Yd4<csBML7ZpsSA(FG){!WEZc<kCkHt_y8{
zOwg#RVD+hUUf!B4y{#{n5;YZ^?$<ZLF{8%(HE!U+XvK8B2F4D%H5_)~=uu+Avu}iv
z^6Xu+Xzf%R{I&oYDrtFapkl<NCmQ*7w3<^g@5))cM-8o1w#_oSK;~v!eALiCu*oO9
z2#Xe>tJO4Y2a9e?elt{(1T|91@5zsFJ5hYMk6j$?OeEt1ukIFCFT%<hvkKup?9M^7
z9mO>_dC2PKD9~(H$R?F-NHI9mTYU<)AvS|ntarGzN;`@Wn7*VUN%L&3&ttKP?nxD5
z7XeW>@!nNCeX+>`Z$qMLF}A_*2|GYQR!d<}8+|M6FzRxV+&>=;umjR4ne&<QNL}H4
zNtVf&N)YV#=>S4mQS(RXInQEuW3?+r_j7i&v&9F7n5CB!UVt5Er36ho=y=yk{7U;j
zuVd_Dd-2qMG*nU<DR?+k+R}hfUY&Xe`%$`Dhh|SUoJTiGAB3z_6-nE`_R;0;L}2N{
z?ln{R5^|>N%V-INwyD>8)rP4_qf@GIhNEu&%oMCnB)kFz_zKx*Nxl!2)RsYIS;kLX
z!o#G1t5QX+_T-UeyN0r}K(-ilU#{y0Z54@@)p2(lYy+;&!sRZjSVh2>F;s#F^uM$9
zF#+}}2hAxuG!O6jhUC*&Rd0d4pG@dL&e9dD9kvh&_SU6(^sNFgHOGA!>?%Q6O=cJA
z-hVh~mPcXU*__%+hkd~}1I`uW?BPwCip)U?M2sCE<{Pr<uv|x;z-LXBgB~6cCl|+b
zt5IY3mFs3~3av3-XL-46vYxu*EGI2!6t)qI)s^mV?8xiLGplNYCtTASMcwF10l7(O
zaX~#<=iR<-YbCdU*COicFuCi+kdPOGE+5QyBjV1!xWKPtgmcdE?ZR1?A@1s{MCzbo
z*Dkxc7i=MY(vX*hg|9&DPlO8Iez=f@dS85~FA?Qsu=kyZclRS3pvT8xfGMYSO`i*`
z@ix)?H+(O|Rpz%ZMD@)pQjcAg&7022_Vm1Px)+H{Q8Nb!oBkDT?NmU=<YIH(G|t%z
zBxKXbxN9L=-L4n)DPBRq>FV(pe8c#HXpA&X{6DXLy3+pz*H51p7L-ue_>$vNb$v10
zj$<mAnW<(^8uZo1Jf!qlLwG052Rp{*&phNTlLp;iokmO;&MC77eIr3$V*0cpd<Iq4
zXClIy!G=$pHiQ+T=rg7=ikrNq4L)N)3cWPrB<FSNKGWP3B&MB^^D1=_pDu0q&MrOI
z%)!U<O)1hoMEWS86r$;aW6JgEeacbVkd<<E`hu3K4<sqqru$4XQ+VuB?};nZeJm+f
z&Nfu#)(&s)pfapunz+x^vj+#=Yj4+?7CP3E!AAjDXjz5@Pf9kOYk17`^>JtYY?Q16
zg3n+Ib+pm5jRrpBACR+3M#4MjqXJ#T>|(}(M&=v`e8x-{j6MeepTRYw=v*U!k0NoE
zXIK9|V~UGnVcp-4;1pE+3s<DK*7xz(x>)P2@_p=fTKmmb%|5EePRgpCeN>H|oJB$V
z6SHYTLr44A?F^ZXm{m>t=pZ*qo2K?rEsUa{vi7ljQs^z9zV;bxQ;_tjvV9amcImma
zwvUcYDblXEeeCuWqJ?#LpK?@CeJ^OK-WuFzxJ}`)w-Vp$;Q@7*BOI%#U9=0%fDk?V
zVGmPFFm(u11DHCDDGE__1XAX1WxExd4Jgr5pM}K`h-YJpKs<mc0`W_jA`s7kl(|6O
zCkRW2eJ++qONP&rB|o1^_E}EzsJ`jw_VLwbleMb@{g6UYW@TuFmq?C9l4CC=jxCG?
zrz(e+%feqF3xDM?E8Nf#ua<?sMi%~BD%@uoC%aLds^Sf@xHnR9{${}rJYciIZX7UG
zA=k_wt4n15{m@-mp&d<S^jNL4k9A~kR8!sfh>Dd`>OgV|ju)iE7h{Ta_)<)f4qt{T
z(&5V?6<=tP<~FioFq<o<ch+aVkXEi`xHt!^M#IIqvXbXfC4DA(*23(n+{fy$lFQKE
z7s(2gWCbp!3KW)<RE2$sEbmfT-en6}o}r^Jm*uUM<y}GL`D_)la@47o*UJ*Fq!N5I
zm8}G!ol?78KHJU5Baj9q(ev_IZ9dzZJc_q;K)1c-vpXlXOjC9GOgGX9+B>$7sgOoB
zqkp@tvwo#BwVS)`vV|?y=<RNc%x8w5#%5=yH*GQfOwEL{d@7~4jKRbqdg#Gy_nVIk
z$w|%Xf%8%8PKp*MTrPSx*BNz-IIN71rjqz<^sKHp9|J0ts?8hcBU6&7O}LW#Z+)~u
zI?2>QBcGk-V=twHNemnL_Tl<yTrU8!>(T7yY#gWEmYa_gExlB`ruSKhO{bfo>)p0o
z|6-M@?cGM*LKkf4d$)Dx<6cc?wo~JqW}c76nx2~(U!Yx5c2ZVd@6&<p<SdHbpO{T6
z8(Q9HwS^(G5wogzpP|xC(x%^i{8x;kO}QTfE8fSX%?6ZE%K2=0{fLs?!0`rQQOs_O
z-p6$GDA}~@eLS$)WbHcN#{-*9J45fgZTy9i;8fl3w)Brh;fDTq+xtEPPd4M7V}NP(
z`;6_ZxLa`~u<47(@evIlSIRite`?*4WQ+_xYTQZ7Ixy&w%IpfuNzpbgT<ftiM+9LA
zfCG<sh*BS(-RCg;jtCmWErYes@j0Th1#W1Xi*Y8lbLo-8Q<!HEGAevBKkMmZE7@>a
zs^~MNBT}OaaiG;oNv|Z|n(o~hOk9>a`fQ9OWO%ns+=@D`NHE`<V>&epE+6MM;j${a
zgpm0d>uz!jIOFy1HuickByt5U=kkP)i{H%!Zq-=~D)&s+$1UqF(Qfhh7-?>r8J3QF
zhTESArCK=d+3i9WYFIYznXS(xz|CN%MPtrteXL?<F`$YvPo0MORxo`0jG1`)ocj0}
zmA)7(9o0Sa@o|J@7i>;^e2msis&*~lbH*~0bcQl;PkR<bc2jkNd)9Ld$~IJkd(z|M
zxy<CcQ$v_@9-qN-cG~0khmVbEBcz)--2C#fFm1#vR*o+zn~Tbr)%eUo5S5LTbyDMV
zq{KzgHmC7fwnBP#!^hMtK8A&p+5*~hK5j-Qj>KS*Z}GxsoX#rSGDx_sIiKYyXMuJ#
z?lbN<NoJ^Tw=LI~^rWh7w<))PWg6<*ZOQqVG*0e1)wF5I`S|Vee&eaeO!j)Tj~;Uo
zv#M*KnWT%LMP=t9XVbNY+V+_-3v@PeR@Ln@Zl)2nsc)aTqoC=h!hLLsREi6z#(i8W
zsaX0{<v#N~pYn65bDxQHYN2+O?z4Q8O17|C?^~h@s`dpg*IT{&3@NFc_Ezye{o|Rz
z?DSkaTf_SpmJV801@B{6I_O#Sdk$(geQc<AA5F}$*{E5yyN}74Ler+)eJn{%(oeVh
zsFpN>3#i(C=2&S^`qb<`ebl4)T#DVts7@=>uGf8hQ)yHStMz_`si0P0*kZkvy3Zsl
zjm_RVJx8a%msGpG>l;fFkJSS%Ia)KVwx%D#XA1&vtrW6p>h9PvbRl9V`WSoJ_2)V^
zspQ=2*mNewhc^@+aVai1?xV=by~xLbnvL16Bz?ByWYbPpmA~NFaSJ2zX-fHjU<Gyc
z{|l~Qj*&b?X}JcP9O3_brWH7WnQ1J{w8BBF@U+4~ui&(TQ?pG33QQ|fs1=@8q|hun
zt>7f>wuL#``V8V}1k=XZoZ+8*rWI*Weh$+LkK&UfZpyI*pJ_!}nf4LWM;WD2O&=>$
zLIL(GOlc!nN-V&IEmj{_Qi1{cOe@mZ?9%p$Xn;Nr6fnf#hDV>}0H4j8PHI-Y?z1N1
zq-fFXx#)T5b|1we@!9BEwY$$cLn>99e)o|nNz_ln`)Gr7k_)KgeHOOT!St!+eKf8Y
zfLwas$DN&Cs$J9j>;_7wTUgikFIEM${X!S)t?zwSl+u~)tMPl$s%o#>s?OFct?@9J
z=yvAAAD^;lQM7DV8!pWWYMbj5-T7C|U%m&f^F5XAY@d%#<KL9!N70JibKU7y6R%T*
z%YE^S2XUwC^eH1@84grp^3&1cS^j$Zli~A6;L3=l_rsM@m1cD&nCCxC_8Qs`C)HXL
z^V8Rbx1PFa(f;{Cq7#1l<`Wi8e=<6a@c<~`ZAEZX-+UMW)$kSgcq!69hwsHtW;_b7
zr>~1n;g|9)UUOyn%JOw<;BK^Pvk4bS<HdjTBhiXhyFOKiD**A1>1nv?j}znt!Vz7b
zs#NBC8n|WZD~9X!64x1=jFyK%3SR}eMxy1DZMcT3QEvv7W^4ZTlcE8*3^o2Ud^P6k
z1>tw`R=}&*Pq$`*_3d7>8MN0QZnY1t-?0;}Xge(a(`iqvXGN~Bz@^&tZlzM5otuxA
zkivxvo#vxcdfmx&*UpD;JYf+OzoH%BrQ3%BUT+Rq`??n`ZqWsaATwWqucD=4y%|<P
z^bP^%mh{>UxU0JbH_R<r&*f#Fw>x78N<jE!5H@Oh(Zb;i;OA(7iPPq4x0;Zp@a6d7
z+wp<LhVOvXNcc|t`(5yWGDrYoQJ^wjiSL;dMFVgn@&pWuSaA@7CH0xvR=XS2<{yfd
z)|=A-+-?njJRGg8Ou$v4oytTDY#zLs>nK)pWu;c{z|B%~m5J)abTI!AHCZ^11yHj@
zXU6m^a4}k|aR_Pm#>00*9=&6PufxyMDnn93m7x?><_-7}t4#PlNJ9sCKm6z6@DX}`
z0AC+P-h2>J4BrUD7JdkS>5T^--$lzhy>Z@UJM+-{!WY5EI({beFM;2oL$2atU4v$W
z+v-6s2=~X~E%nEcQYhzpcLTby>3XNc8oD&mss*K5FbM+RDOF3v_YyWg&?(_1cBPpB
z6ilsC0>colQG@&8;bOUGLJJO+;BvznCZOxqN;B1FeYS_f+ng#1ASLWgrCNP*5_DR#
zTf&<}gC-E`^d_cD)lTWw-V}86snS-s61{|EZX4Y-QXVJ`j7<ln102Tz4y)drX!L4;
zQ-#Dq7)4g0&<VOkv^__3tD>j@DvARf0@$;&D&ir^-3~xVgG2bD3qUw*Q3fN}p6bm&
zojT=#dC`ZcQKBW2aH}opF=(1|>Yd6=uhFf~v}&Nus)wreMs>UadY78$_hl0uxD6WX
zg(su74Y`s$9boj@&5EWO2jE7j*`VF62PhhF<z1)J0$qq~fS$8dr153=8l-+ETolb!
z<WXpRz_gh2m1Fp{#PASch)%)1sPs7r%p82PT{JuJ9{It)FR`JEi3$Eb1&XU#nyt2B
zV1daBux4Pl{9|C9@^1hepi*s5bxIorxnaRkP8Y-g{(%YHt=5~YVEBMBz0?lw@6}<P
zgqsOT4ha`;RVY@p*v1m5o1qd)cokaV#?{kI;3h%dRulbU%z+J!Q-Gj@cari}=hQ^&
zmrI)&5rEPvX@sCaUKMePBE!DB)q(zkH`$gXa>$FfC{xAt+_wjFFoe|9tcGNr$^;+O
zd9*}~C#(^p(|DV}$jGC1u%162i`8IswzMDM2GVw`UaL^BulZ>4aC>TdbFu|Ge{dxH
zER1d8qd35wJkn|g^WFIe=fl4ODS>yN!*`#@cPrK?yj-JXicVNNPm(+`5`P(;xOVM)
zbSkU*JbM#xK@a~BXpe+{1Wyp!KZgH29R3OWgY+ts-e@Tt+H2Rxd))xXGXjZyC^}tK
z9UF{fYMxmk7aMA^Y{|&F8jU2pwp?E;C5K;!VxUN>I~BP~P~vjN3%tlxpv<NgXE4xN
zZnTPjUcm~2Ugk}HjsYTmO*Yt?3@~wa6%=c6Czk0?lirG!i?X|OaMdPb_mf-&vOGCT
z3Aduu5~rUa3gUHVb`dM^5?6tGUZr3$7(aqC#+Blu)VR^&3f=|^<|Hduwy^venHtvO
z?vzV-Lte@m6|h_;4ifM2a(29VlNI<xQBhkCDtpo>JGyrh((07zlQc4wkk2$WVf|69
z^(J8a2w;rsLaCjBfj!lF2mTB?9hfV0nBJZS8=Xji_04nEa7gp&lcHKn$!f+Sm)A!2
zN#zAqVmhwGpwKB_M4j>_cqsZJ{99x<j)$ri3I85{{W3gM4YK!v6BjMo)DEg$sJ<9Y
z#Nb30h-wai0^UR8s*;xtTr3x33u~K=G(<d$o;Vn-Mi#H$SQ@)$_m-jfR~VkU)#gOd
zfmi$F&-g+7DgK53R^yfAS3$<RhIib#rE*pIs+bVV!SifT@@z=vDVNz8Bffb+&yrd2
z29JU_l=By?;@VHVc)(=^3=E*e;9M4r08FwP4Xd^rP61%}H^<~U3|82AU~oNVgD)48
zPHeag40M9VWNfiWLnUW-Ic<W3wW4$mQg-JG9guuaohNj39<7Vy=s3}JtS7`(6nz8d
z0vwF!NNOK36-2a*CJx*@sXE7Sp)VH0q*~DyQrXI%sk)80x}<FPegbS!6Y)9$84x4M
z*p6c+BkDpJGmrOVT|c1OP`fJrZKx3Swd~2mrCD+~F<$ctfHa*M%lWju)S3jV9HyGQ
z2~45!k`jtgU<}FRk?FutLsHT-q$Ik4iwFn=evL4sg3gWMB$y!bF>RUXo%9oxJsX$J
zN3%TRR`hz;?8rJl=zBJC>=fFvsi0etMlR{HGX=M_SH!h;%r)xQQOVM&QPY+=Np6mW
zUqwS9W11jT%$Z8Lv8hYY&bL5|vk=6t32`t3qWC_q<+KW6g7;-AXwSklRM=0_wokk!
zc!x}QyDadw##C8Xl#UDmwsd&94t^NyukBWYcV6_Z<*mgEgo)jpPrr!GXQ^yfUQ!K?
zm^D~!sR7A={4y`7B?yYmJqVASBFDcIlxLbygaR=K{xT91viR*BvUqZ*dWd^`;+Yfs
zjSxILMj(x1bTEhisvE^4;rH;@@53WHbxqi5HQBtIO{Ak2j6yN0BLp24)e`0_P1YL$
zSV}I1gRg071;!0b9GD@7J~(v0BN=q>U?vb5aOG)*9IiJSrSYJI70N$#IJHxS3)uyC
zRtlLBzY>3b$-<u>A}4-i^5@6+>raqBCq8up`6M`RzB@|Al$!Baz?_p<F%&Vd5mNZm
z1inD;jQP^+&5Q?aZEQg{G1xZBHt{A%vNK9Ym^3SKP#!KdgTn%$u*D3kn3Y<KtCoA1
zJFVLWD;rikJKYNS*y}<MNsBc3o?7OYEiLm?Y=i&AHaO|ZGHV1e)B72|`Y(KS@>sj4
zdzPr_{u|%vO}F<jSbd|2Ns~Hg4UkGijdKHZ9V2*5LrbQ49$6jh&kUeRC&6vn9g^Nb
zfE%|k`dMLh0%jOOdvY-W7*+I*GpX-&zzK=IabCQfUdBc>n5c|>f<W<>mK9bkngOcO
z0?_9CQ%L8W*Q~4o&%po4XT#iXI(swkUgleppy$RCB&6iT8$e1<f(Q8ns?0`s$!4G(
zOSr9zxs+SI2@?bGcWfkkP)sso+mO8_Yc)_x0Qo}1mXh+ZC7gLW)Q5&3>CNiEYDpIs
zve1FU?bzoh8F4ByVlh6Rz3YTU=uG9zShG85;{+JAJEK%0N+pn0O`qih8l5veDYQ{V
zM@Z-|5)J9hMlJe6M@0rK6?sZ}MfmcNx@6;{+5%b@sgA=)Rt1;T!)zhsHvyMexxG2Z
z$}R6%9c>Xdd<%M_-inVmf_@qa-v-I(eCCl{Z~df~y4*-hbjW75SVzl^^c!D+XX~>o
z*|ObMvBG4o_lLiW6w;P8X~l`kp%j%vYfs5#&4(Pzn&_2-mCx8i7=98<<t}@^(8=&F
zV|X{=a&B@#`BPb7X=#eUQi{OhB9&NhV}&edh3uq-%48{@%_`+{n1M?9Jf^4=S~p3u
z#J<w|FgYrt{n+ee0fVNQfwARq87FdL*#XvA_~O-#C0th-5-Uq&a+Pi@X$vYtrM37m
z{AVp+G*KD;6$ENy_$tPWVYyV2kdUusu|uw`gx`?t_$H=&*})g){|UdLg?X{~9sU-m
zF?}67{O2q<a@8gLmW1<v3c&fc7o2~|g7Y2BK*I7}Op&nsE2eUC=X>}~CU?G%xl+0_
zUylj@O=9^$0W3fC!t$dGEb*#M_+ts@C;8!sMH-<<K9Cfef6s#QADDq!>8BD3t=;&y
zQc}zP3`rTC&I0bgv%oEa3NdgeV2VVD7IzeIQ8cjWn62&%=>Pai!o?Stz~%)*2`$?3
zxe`oqX6InTPU+%{=fn3yrR>V@AyR(&cyKb-E1OfNVzw0N4i`&KEs>m}6(ogId;b@#
z{tO&n{>iuQ^mASQ*%({q8wUvI+O_Db$EBaOYr!tZKbT$4ro8w!?wf$kJH}L(TUKDE
zs|!H*)mj=aiJZQ#0$)0Or|dg6zB<0N?8<=Nvyo5dWp&V1MHDu($&NOM$6}z9DgPFt
z%S9In9x?`|TMVcBb;&ccy5tjNmpn^$Nm`dux@6SjRW;benFg~KolAV<5L>{B<=_~w
z>^#A;bY@-X$E=I9n6+9m>&cQ?w5p{rYwz`_zt}%3zkpqcI(1<SB}`xPwJ%j;zSVMq
zo>;C<lEGi_pU<+j%$Korjrr8YCHx!z8S78@lU^tM1^*dfth!9*i@{izvoW717&9AN
z6hi{A#Et*yYjF$-;J<~JdVOrZ%ululvt;{nDcftMY|~ntBHPh!vE~Fq96MaNe@N4a
z)BzSkE$X#h=U3bHS+%`V*7hn{TUz~7Yx{Mqgl(|HnVe&M`5!qltxB7_UhCH+8?&0^
zdf6m5$R?qsMYT!xF~w>Q*xKqM>vzYn=fw+`;mv-PAI_@$Ewb{PWaVkmQmy<%cIAb?
zPg;-1TK#PGW5BIh4A?FiaGPWREsLrQc(I)U{j7{0b75f{VZ)t%eA<=8r`?iIcSt_b
z;;YK1{dPVX>#skrs*m|q{jRL4-z}@YS5}=Cd)2D%O{qGq_a5_FhjJ-7+~-&G{aH1?
zPu9F5Yfg*IYRx}3kiA}P9nIuoF%4;QxfmH$Y{{560sqfJ7y;2G5+pwYz^04R)yfvn
zRhR0NAMfi~ybmSs4@%zC0=B~Yy_dn70k9uezO$C3txeD**2@)X!TQ{*(zcnbwkn-E
zr!BZPtu*(7wDs9dTAUw(mEWL7NpPma^1s@;9lzG?Wwq`h*}8{i>(U~)+PWX7*3G|U
zZEb3~UQJEmTCo-qP<++eA`057b>EU;UAsoyYuBnP*!`E2T*qE$iMv5I3g1Hj0tcH1
zKmw<Glaq~LV`;qAYQT@U%OO=RpXVo+FUXS17fQK&k(5hXT36&U`u&0IMRcnS@)dMy
z8l1+(KAK@=S_fVVSY^+#j;=HdBo{;%tl&`DTEC_OB0p9i&0@7&&JSNESxt-gDy#1u
zNMG6Je#~Pt1&+?+TP@&}bsaRKB8(bG=KVPGnk<gI7BkRl%j+aZXcvIWk;x2>Sl9Y>
z30WW+@#M~%liXpaclUk)^?M(OBCUN4C4v@@RRtKY8MCd{amU-l4o0v|o8VvYI~U2+
zE#Fp41%dgNdOi}rSfDeKtni<sFGP_ZU*+Q3^Y9s5)cZ4hXa~9XK8T|5*gsJHxH?a*
zG#e&;o6fa7`4MO}I4L~<hw9-FcCSH)n+Nr+BlLS>>j*n!|L@Smc#_*ZZOBzP+vOH+
zfRpv?Kmw(XLZQbG+aL48_Tf?1IAU+N)R(+ft`@Bl4LX`Hu{&oz#J<i?;qPrPU9<?!
zeM5Kz@L<y+C$V-Viz@Xxxb0MIyknVa!9=TF?Y3YmHO`6Ik9;i?!XE^1WFJn2vp61*
zn3c!m+3wPbY7@6q)8<mxhX@}|)~E0+C#(hW1xDO+3QqNEbE*fYIZGhgK?BVR{$AP5
zO};%k5Q;eEVI2XQws>=^O%R7Np==;0s9~%aSg3{3B7+)bv{NJR06&Req5v=@I32Wl
zqSxuRW?(~ZZ8a1K@eJ4x!ZQh_qW!{)#bf}?Av2E!n}8kZz<bz$OPEXSYhqK`UgFkj
zgX$I+%p!`wk+il~9o0?o@C}6TX~9Z0+Zi7dl`Tc#O7OTX-zN{baihHCEo)s0$KzGJ
zkOn&bSaLJGG<c|9m2!#OqQO`nZ^Nqk<xo1C+|izXpLSL2TyM_yU^lkde2p@~4hIZE
zalw%;{efkW{Yfz4j0?CriWL-~tgy<)ML}!x-`Qydh=~Etlq0aK7I(p#KxV2(Dy+h6
zuRLxB2JZ<IDwL*f04H!%slx6IU)5)NGq@9mZ}yjU4vvtU1k5icb=Vtg!8i|R0)Iew
z3W#+AQpvf*EVu2GxIKM{8b!R3av}r-4sJIm;4m`<x1bYr3;-|`>w(pV?YZbX>J+Pm
z!q5e%heA!L509FFCJCxDsDpC!@1TjXQ&0=>Z&B+3TgDqLI8@&O9pb|fsdbqgcOaAq
z#A6CdaQ=S^PP<emA$AG0D@3ZPKa<3c=@D0Zkr2ggRtL?EaQI9ZG#uh$M+BrS@ZQ&I
zR={Bcmp$j;0l`2SXI~Ow&<Ge0EAAuI*+r0@(dlYC*tUBN7^P;E@$GziWj6qM!vppx
z=qeQ<y?Epz9_c{CwxBohcs(76l2w9Wa3UCr_h^TmP{`C|uZ?ORyeT|(&;g8fa~NYr
zG4*|5XyAaNaOfCI1=PX>#B_lX^1x8(07Q7<TiihKD$uJc7_<vGhc6+z6iQcRs@>|%
z0&|XZ+0VgDtHQepoNeF@h93k+fF(_)-jo*u2Mz|oECmf42RS{6RPcNXl#WeZ0qJFd
z84hrkP1b`3oFU<Rx^YYt<32Q_IGn-;OgNVflz`BoI>c?74jQv@I0tN*08a14#srgF
zu^WoThroCbP{e>Jf)AvqMB!9RHU1cJh@Qd-!SU`YDPS?(RG|t?N;4~av`W?SR<Apt
z!;71;s#OKI_%qJ|k308uFvuYO%2wR@jc$hMg3Zhbk~UbGONkmD<Y~SZpYbu8?y!jf
z<H4yRVWJg+k^7(p!9IKhGCUleJPj=YLB`lzfgz2S3pgyqUIu2C3C~VM`7yJPRm{uR
z#5Ty9dD>%a#~wZ%tYZWD=?VLbpW<1aV7VnH)nTJI-z^l>ikw1*8n3}h$$c~d1^aDP
zctTwL*skNXxHM4<duFPLxB-<nW=F_LV#-jq6H)7O?wT^0f$JNMnRSz0c|s(21Dz>E
zi&=p@h$BhZC5g&e^HSZ!9roFEv(iwjri2Xgmovru<V_<^WygJ*MmjPad4!>-cuSW0
zY2R2^DWeLSu^ez>B3Mr3BUWGQjXLpt-_TD_7{1Q#Y{{9kJ6kAZR{(~iw9b5*Y4LLW
z7bj#rnUUaM=gs{j0Tum<tFDy)%2y`TYDP7dN|Tl27vY{_X;*R(@D1L;l`j><;qB=A
zz!N6X$*3~<fMwB`;WZE`mJO`QP8gfkxjUiysRGO~sAh0*m0|5UJ7FSU&Q6&4UV-k<
z)98GqGn=L;$3B#h;nZ)a9^!JmY`vXIOQ}<I!qaj}bNJz5AunqaRdAZvTzo`u`s@m%
zP*UOBho1~-I$JUFA#e%AN2A#zZ<sZSTDh`Qo80W(Q(a`V`k<oWEpif6!`*MAGc8jV
zUImLSkeuG|Cdpw7*k~x=>725c&WTnTlO$(<PUOqkpA+8~!@HE0{FUK9slk;TgwbDv
zD_<(;9EwtzeOQ!vw2Ik2(zW1ZY$9Jy#wNa3p!+Mza4$xpCj0fWM7L8N4bok~rS>N8
zUgw~qz|G0&M`2Kf6`>Us?XDeGi3<~zFmfT6!xt-O<T;M`I5pTq$snk_se|yXuq@3^
zv_~T1Cd|i^r;O({;oza$&)GaYHaxmztg?B_=%zi}caLq~HB#9*yqgJ0_+hzR!o!w}
zDUUMuaS>41DNb-={Nbp~+lfNWmZ2?h_di7Fa3WD>r5yGJV_Wu)Rd$W+xTj+IjD!EE
z1n-<hXOK~<5S_S&Pt8UsdO-A4M>R4?x+=4QzH?XR6MmIRAyc$6DI_hGxp3=_UBj78
z^M|rFSG#&`7M--}i1s!`EQ_Y4K4)$p866uQ*|epyb^Debn@5$-@u#v9SGX!+{a8V+
zb&RzRiLw$=q5dAJpU{ZpKzb$0qgP~jaGRz$wb7b_B@y1Y9+h>!&{cPlU#spL0qX5*
z!Z)%1&~bO}DdRnRh3EnpQgxJ<_61qmrLNM9&y+%=ckEk(&f7|>*_c~1RH8pN;4k<G
z?Vpsd>NDUNCRZ=MQ^yJXd$x)Z#_<*c>0DQSrZONDp`fh(jeF!UtN5)0i@`t%|D;!-
z;fXVA8E2l#0s-ipS#tZ9dn(8k&Xm8FOcB8)5+-!~h02dp*Wf;O&}LNR)vU;?ZAGp?
z0z&*mWBiRQ*V!1xXKLZg*2vR_0>L#5!8JAnOOe#*rd@kPt^ZDzo`~knt6x?=QA=LD
zhPW<Dy_S`Ftybz$+C2~&VMeGYVYC=^CqVe;Aj6D>8;R7;yHC`;)Q>_2j1rFd(4Gk?
zng|g)8h1+>;7vwZ5C0l6V{ji4|A-=9V2t(fOWK<}u_eKQP0hW7Rr|%tUTCX%q4}@l
z;OZ;{2-YFG+}RbGFP=;V9{*2ihmgX65SE<`%TCq=%zv(zYsC7XauadPMZ0rDO9Y7&
z*Cw!ALB5R*K&6hbQibY$1YU^H-+W!+JD8VRj6!-)sc!p+*ESK8U9B?(_n^{uvC?-r
z>#RlaprR9D&CLq4n1*V3Ka{oI&C1*DEH7R~VNp7$$RVG$g~ceoyrL96!kByqEA|dc
zv5x{a=oUsq4lFi&dE6~TG<9NH#HhX0Q)qb!Mpz9h;gv=S;YqlulL~nmE95=-3Na#5
zP$BbiA+mS(8`C^IO#*!%0o}RugnUSigZhN@H4daj14xG#0LoXNun5#B{3E{ZTyX-!
zd$xi1BFLH<89H2z)uz5kC*}$D>tfU-CgG)mH#^Vch3zy7OU|FTqT#Y)9A2hk-e49p
zB#YV2i`k#E7~|A174xRJm;y0_!)qj{102*dbAmF?%@U}$#Gn|9Z<574z>9f4D<(e6
znl_GbHs69&s5R-7Di!rMv#2eys2A~~j^?b9aTb({dAnK64q41AcrmZbS&VU%l8Sk!
zS<I*`=8e3Vx92RzIJQW|{Hj^Zy|S2h@nYV`+ee-z^k^UBoFEnT?zkwa&AyAqh&vc<
zI00zTf?>f{!$}G51038R<>Y~Jh>n20*8(gf3T)WYfwd*DKjFZ>mJ^t9{)>RU&jL&a
zK@I0P1YzCD7^pI+qj)1DpoY)UVLw;G{w9Zg)`|HB1LJrQ!G0u$EfCluZ-y_CrJs8u
zQ|*`KEZsOUL#2N(S-LCSW%w$I;Bt=OX*m%XCsYW*Z#WQG;#7ujk_fKl2*CQ!@Yq4D
z$+yC14~v+SHx-FFDQrKuHP9Y%lHZIdHx!5j-ZPjUagz3_lX9jMg(0I9-+L3j7tkBg
z`ksc<v9yR?wBeeFQi_OPgC7Z^zm+6<eHKx<@h7|ivl7u8W1`tt>OIUR1vlY0=!6Jw
z#*_#!XMr?~@6+me3ua&=8>R@-W=ttZo?s64ekja-E`dYZg5MCNt(Z!Iv<=^<LAn()
z5TxyxB1pGkDkmhlZXE8wuTvoH#P?~CMlb_G+Jz~Cv>Q`7A>D!B5SR8~Dh1LgzE6WR
zh8YObotPp>cVQ|gq`UDOf`rFB(jeV~@6#aNix~*gK1>m$XJ9HPr2Y5}LAno9DO{@H
z`!q-gFatrVVu~P*V=5=43H*j2)i9L;DZuw>kR~w$L7Ku8L7K)?PDpk9h9HHQN`Z6`
z-={%pU<QITgDHa4#8gg5E&PTc&0;DA(*5{84N@C35Tp*K2vQeQIU&VLpod?lKsto)
z(;&&+CE;NS=|}-cbNCH$>6w^Hf%GhVp9bmKn1Q(T0H%mbzl5p6TzU?EodW5X@qHSk
z2QdRddM>62((^Et6VmhX8{*OnFqH!7h4?-VlHBGIz6i4tq=zt-6Vk)@4MBP_rcxli
z1mCAYiZBCl>7|$=NJlZ16Vl7@8-nz5Or=121-?&%^h(S?kY0r;f;5k*oRBcYJA?FU
z{5l2FYw&#<q}O5wg7i8}5v12+sxYKC;MXaT-iYthAiW7Q5TrL_iXgoOQ#m2M6~Cc|
zdmE-wApHuyPlNP!%s`OdfhmIYPE6&56syX2;nyjUeih%RL6UnR!e5h+-dzCFd+;0L
z(t9x_AblWdd;dDVPlNP6%s^avKc<LFw0psyOUX#eA3#!id|-aG{vB}Ma^P`w;F8bK
zj-R*eR@pI;#0k&i=Pi$)w>*B{k{!-Fe%=xez#czunR$%<_<2h_-+cVMC0Z9H*xyOv
zeU90A%YqSC?7?y#rfoCfwP@oB%k$mf2Q^&Dzk%=5EKk~uN|q<3$nyM6{3vO8x`5yr
zPO<SL1<G$_LHQ_VASjd~D8HQ_O87DS2JP~ZG|v1^7MxE=IFuqdzgqy#?|H%bgDg0I
zDB)0w;QUbmIG^-_^Cwww{#3%D6v6paemJJwyfZm@`b-uSV~Zw1p)H#JBAlE&{W+GZ
zPo6?c|N3keINH-mLQN?G_c=SC>`NHzo+pz}7AT+3f<hZU2@0hM%Kyp_CHzbLCbNJ2
zRTiAT#ta09QUvFJ7l89cFF0Syg7Y^L4y6dr|5pIc-+ICM|FhuyorFUvg7fzU;Cwj=
z$Fk!14_R<N4eFFN7NrOdZQf*kY%sKVJ-(Vn)Y!yHL}?Revg6wO#?SvsxPQSq8EV0@
zviaSts@txA`&azPx9YjBfBRk*9Air;wZ!)e!1*?ciLM$f-1)aGINy=d7GMANg932A
z=LP478E~}Cqr{pY6~OXCFDyUKz+!GJB{)CH562u&EZz6tv!EE8O$iEZHuaa`<hc7&
zEY;{>7H~hy0%z<~C2+J)RRQ<-zW(jM{Wx{P9l+4E8Gy0bmN-S5Z52)(zy1xwsv1`(
zC9h5LxiX3JWL%eI-hpHktiEQ~Bf-^3wi|K8JzNxn7XOGN+YZxpSI4h^1FdYgE!d5V
z`L2IkihV6xc`eK8lE$ub>XNjpT<MaJ`|ID%^kddpS<EsvvlFvuGrPj9<JZ5T{=$ET
zD#4$UdJKQ{(Pc7UhCai5s?x&0@t*~ZneFj;{o6TyvVC5bY#S#iNVe$&g(BO(`0L-!
z_p9v%S+z9|a!_sQActDp<JZ6Okty~1w<r5G$)#CMVjN?kCZS_2YLh(Hu76wOSNY4c
zDsLRDp~}<28nyC|#p~a$@MFNbECv`yeux2d<VR(|i}|{Sajoj(;rh4retfzri%-Ux
zCE^pESyK7**t-7hDSlOdT2|GKb5vAyI!C2e{V{d@+ckbQzb>oh#<4D{IUVa#YkvIt
zH|;K*oBen{oW*<N<QnmwPOd4uKYsliSIqrh?SW3Z$mjaEZGNr0J*#yO;U-;L89Iz9
zYF#@1r?&3#>)%4S5&@%xcdPZLxMqUFh2bAEAP9uN#;Ml-%h$i%;U|})S#oKd?<BdT
z^PP%Z{y%>G+ueSwz9);-#$i@sH63PES^c=U{%yY>M=DtyF%H)fN9b^^%8|$2^=}hN
z?i|1V4Pvr(;5r4N)A`=_5)Xz`F2T<IYOeto&A_!NtEbs*D3jHR0B`nTx*lT{lG9(r
zbB`VWyIUP`LxkAzhPO(v?+zYL`SSK)4lZ)2$&GUOnf9+=LxZk=`$t^X;@il$D0Cq5
z$z%Wax6<S5`nP@?>m#so0>Lk?faZe#yb+ox-fmZhAN7rh+9hvKHUlBefHc=DaLo+d
zxx=NM$737W1JXc&1SoYB3XQJLl!cv0ue1~CiTLq}>n@J6sAN~8xRI?%DJ)7xq2y3H
zWedLYX5(@U?MAckIwZ?9v*UUVcFLa08}{HwqC62aG#M=_ya@}(T^Q^H{v(ijSc^XR
zb@&CMMcpFnz6nzcfF*HHMz}>{*@`JA7UNt!VcA-UFX{JP#4&T(!Fwm55y_lgk~!N8
z0m{6$BD_N~X-_JX#w4aYC8k>oVdA2VVHe@fJAS_f_AY)WOL(`0v^N!!ajc%&EfK2Q
zr~i~$q+=HQSw*s6BDpUW$pL&zJ-Lc0LXrsN?uA4N>Yju%A>q(rd#C(NN;q_qz7K&;
zzwse7j@ml`8RzH;(3V2NlX{0k*pdvQbM{U=ZONbxrU*|W+_YDFWnIk>_OJja5aZ}P
z0VyWJj=LDbImr?_RPRLcY{`-bBuk2ku;2X(Aswi9!g)}_q0{ojnB9dWM(I(>YZ{I!
zYR5QO?}TQYnkQ%@g`oLefDjtT>z%-iv-1S(_Cmlst}zH-CIyR5*gH|ZQnuu)q+k^s
zIO>V}+w_>A#tC~Tq}NJFual69$-3X&10kKRcfxs-g!5(zr<gXfUo#LIXX>3mj05u|
zK*jjN22|yWf}=`DGtSdH;TVVI3C?YW<XMgItXwmof-+9jCqX^R3|Hg0Jc0^uQ38!|
zni-+PEW!T64`Jg!9UkJwHXcEH`Sxui5HA_xHgDdtm(NX}E9NHAl|Bs7DJa#BPF@GM
zk+921cpQ8&Lo#En>ohDlPv5?Cc-xlB*sj~RjA&qAB*BU}?y0jI3mF~p1_QCU`8k0&
ziN^x_%*|W24&S+BtfG|5xg1I0*Z6SRN=Svibj=NQ3`l;He)E7UPP~<#r3Pom87mOm
ze$N<d173+&$Vw#c?Xy}qk{dcLwI0i{9&pWNwcRzUvX1k~Yzh}bOKqN@<lQo|S*y!y
zWL*+>u4U9kp`_MifNBynYeqf91pq1aP$;UkI7)N-=$uV->Ge+Z29Ps=&)IpSyirsn
z+=aS>T}U3C3l~REOqW@q*hj?7d%j9RTjjNwbV)1G=AH)343}zB+~X&(xW@-{AHbtb
zeaUk#QOSX@T!BNm^=_rYu3-r`$I4BH3#E8{+u+4WR|#uQ2TN`hzp_xFB24)?4Ah00
zF%`9IHkV?(;MVR5xFiXuSJA~T&`L%cC%k4C9>f%3;gPV<XEiyWpEY>|U>V`Y<pG-v
zxajZXcir?4mk~?Cy@0{JzzH{6PCBVF*=p3pbr8`dF4pLIiPc`loFBa;{0oKLu#-J1
zw;yE^Esugk7r6?lRN&@kDbU+vf!2bOmsn93dx|J4K02cvOu=OkLA$~)l$JbP>w;tZ
zg6xRfV`Q?cz2gSxWokEzR<SnV_Z+v{^CSFrBkYTMuX{49(q;Low5(E@3c7IZ@(l8f
z`rVUVb<wg?C3eP@;8ACz6YBG<3QUa&oiNDI9mqr{;;b^=z-~v0&W~qL6;l4phudgW
z4^_FA{hZXYcf_CsXQRcPRuAru?r>}(L@xGThF0fgjiH@$qva9EIli-`JXBbN(cc`|
z-$-PmNo3(ywh9OWR8k5~SBQebt_4TtM?$o;6pb9MYzN(5y9p+5v(surO`VeRY*xEh
z`qu7Zh5NwrnQCPcF2Al2_bYH~cD>1K`p>eacUo%t1$e7TcZf9&?}n_Z%!hmN+k5bV
zlZRN)_dZFi&&b=gLl}7RE^@f<Iw%p7U_%1^0SMu0?ABxnuW2a{48uj+wN`Tt%Ux~G
z@e9}C4JhV~v@FF%WMc_$K4-Vd<499FT!(w)kqlkeO{K79F4^ep6Zt2um+>CUi^o;3
z>1u}w$Hvm#(*fRo&QLdJdvGr}76dt1CCUSPs4Y1#iKkMpu{*;X)$yRwVb{<DSm+Hw
z16Ll%CU`Y#-VgaU?}cg2d*~|Owm+0@d#``nKEv9!`z>vId!lVcvjQjKqIKduFs)fa
zc0qzScXblfG5yW*l?O)QuKI4N83f3y_?r3&;1M>=RL~6C)h=F<4*!FZ6mDfFcK%Nk
zg%6V`>|B--7o97zqs)}0Wvopt=-ObOZf&(kd$U3NP`wk>HiLx1r6F7zP8fymNfv>(
z&mf&Dw~9|~jpKGBVFRyY9|a{DX=$?6E_EcR(qz3Etn2o0Uy_jjovV0Q^`R4VtC+7w
zk<y`oj_x~6gH@JYnwxIku0LO@2_tCd85}+Q7tNrk1n!k)2dX_2v^(Y9?O?183OA^Y
zwGIZ&`ZLi2Ng7Jgk_z0G-e|}%{c>!+R8a8;G_v0pcXBbZFO_?}Y$N>6C&X;eaEO?(
z)Nd;iOWigHbjDS`j7?S7qnZUF?yPrwogQPImWFBNhf=gPhEjY1))r&^4pf_I$6}3z
zVIsN*KN3VO2ixz=B5I4VepgKN7{pk=I}4Jr2u$_dTL6-=&?^D~S-5mh79?XWm>}I-
z08;GZ+<|Vr6iEBBAQ`K{1nC(CAQ>yYDUkMOK{6JD3DSK9AQ>yYDUd2zkc`D(f^?t&
zBxA`p1yVH&(gAFF(hamKOpwM4Kr)tnQy@)bK{8f`2~w>9BxB(>1yYa&$yggENRtI1
z8Ed~OkfyRAwPnN6@-RV~E&wT30<;R80;!$_$yg{RNMQj;#&U29q=Q+Ij1^;o)F=Qc
z<`OLlr$Cy?f@G{36QpJVNXAld3ZzyRBxALhAk7wlWGn)wK)OE*lCe%qklF<x87sgk
zkUCkAUMYJVEfN!?ZUIQf3UCUfUKS){k(eMI>I)LB0;fPaoCV2PC?-fp3P3W}gHs^Q
zWkE8Qj0w^+3qXpMKvs<PXJtV$mW>J0vkO2nR)htYED?bo$bw`n8WSX1G*&jyXIv(f
z3|I6VELD&AaXiL)CxD&dd`d@K3n4Bw?9n1;UKLZgv%MXFdrRHsj(bPRF$n%2{xSN6
zdO(dle~F0at0!oA59Zo6-1(l^_X(wrw^|KeEGC0$6C8-Jlm%g;0m5P7?xRvlj2xBi
zDDilV_2V(tkH=U)9%G%aLL85=?xEI?$5=leV|^y*R&hv|N5>vlG1f=X#uJw3yTSLb
zA8iUB%(6UbqagX9C`Fd%^YEjj<(U>^J$ycX!}egLz<FU79NG#<a41D^9x4FmVJ|o@
z$$~>0{0I)E2+m6jz&Yv#=jB;&XsaH<p%lS+WdS&^@`Cf~EI72=j^I#=;Jmf~oYy7c
zSmHpxAqx)enIkxqA~>{rPOOSpW@c~8A}aU8g};JXi74%OOZE?YtVSLOdZU<g%vQV8
zH02l~YrcZ}vNo3Zwn+R#?CFC|26X2GYypec#=>7i6#Aqtw5(&iC#zoX#SFx(UzhcI
zAEtckWsm)Cf%5(=C~`Yo_=tq^fnx&YgIQ2Mgc*n{zagP~xPK@X!TZfDD016c_z?+(
zcBc6^P|_vvQ6yzZj|JSvvcP>DGZ3GCM}nixY5BlK7s<u)A+|T8B6qFbIMQkcn&^Mh
z5B#5Gf&Wtp{!<b>ZA!}rKKu-Rt#r-5z`u3oGJgw1LEOw1k?s3(wS(&LN599P>8$#j
z5RDj$Emv1Oitp1FnQ3zzwK1hg^U*fBd~Ax&;CrMLeWx|+|4y<kwBY?x7GByiM|dek
zcxm%oPQ16XO+iE8;NiO^gly6eY%kr7%3%O@mSG6IM(SR=r2KxvucH5yRndRO3{+7{
zQAKI*oKn%f5S(Zc9Qv=rhOb_8q6@*4OUF09(f%#I@ok|y9YvzrDP>&>&&JHnI78|-
zKXe6_ZY+^EPwOV%_miF<WJ%8trSwpWq=$CrsnYXHN$EMhwGL0_6x?O`coVk&vIv{S
zBwvI(`qIV_+VV%Lic%zOC*ns;*amH;>*6e;ax-4I1hW#+r7_X$jY_KN8V=x>Vw|+t
zLd&w?EXNE~#}$|&IJ6T^=}@U_H!NmuxC*~k%!xDbZ>u?RU?A5P$AuO?GA~X-bewJ6
z%DJal?32FDosWRq!x8NzFoM6lDB&StM+RO3zfXTMTEVAbG(^C>i(2)ASX0{HNGZPi
zk~i*m;V?jSV~)LujP!bOhN|q`es~_=erWk1ea9i8jFFQ((r^Cq$T&I}X0YpAja}t>
z6Sh@jtS7;_k4er^%EGx@^K&jP#e<2L=uC8co6b+dPW+@4@sm<{`6=QNWketnz5v$n
z3RlapXSq7wfJqUd`K077rI5SP$}yIuG|{MbI<UE)VPrynlx^2u%=2(EpAuvuaZJ`(
za9MuIcU~{9j!%?_Z`mZaUXR=wG+^&HD;<ojY8THm9IDp>*j5kz_-eCOg&<vrt8=(|
zI?)brPd%)A0tr}4g=NoydK1n#)FI;p47L1h!culNVL+?Q*IAEUTB&pzt!_tdT0Wg)
zgj0(01?@TC;1X8BR#7|kOD?i_i)3`-;N=|9Cm5hlC`aZVidJ&yL5)tj%xgS937JIA
z7O^BjlI-K>JJ}h|W5dMX7MQuzeioX&8om{#2@mypB2f)VpKa@_S9X~Qx-}a=K)#-9
zHl<T4pQiUolK8k6P@kmVc<n5<%AcZ~r`amh5r`(Gh~}s9BTm!e+CnC#(tH#%5Y5kF
zifDdbq4_?P;_Jfq!+)6f@S*SnkPeSRR}DW%zoP7gU%+oP?|PNYoan)Uh3#_3#8&tW
zqUW{($9)y=rd&}SOB<k?&jxsst|#Ce3(TD>K@-mc!MO=)%x__Z$RwbYHn3@D5YHIz
z)f=@6C(>;lmL&c~lK9$9Kn!P(`{(7+!_(9+Dif1a+8M;~?@=bc3=iWRVr)HvJ@m&U
zKHPwO4Sw+rC<*#H_3<6~`?$^ukI|P8D59gMe4pOiiODhJsTa|2yk}axMc+pDruBB>
zKhdNV(WI8(G~Wjd<2}>9?!|Aqc%s=-LlDlGg{su1?>|65x;4+bg4X>XHoE^qp>Et}
zm=2ex$!>X~QJ*yg`J^#GMS@HzO^|tPKo(>qG}Z}e{JPq^xUwD`L>X4qg+mh37vR)W
z4Vb_I7%PJW;QAZ@lCeHi6c1ymOu-V07u!)p=f%M%FqX;g;jvA(Y7r;~1V^Jy{@_xG
zDj*{kRs$B0CZ>&W6e7rrI~+v<t%b$L@*=VL=|yVcj0_SgE04tIc{`s&V`-6~h>%Cz
zKGoHUCf((jp%o<zhr-KHS#u;HuQ!3PQ`(`SClCnfQA1;g--Z6I9#nSfXgY4Yf>~o{
zDD!<oW9vsOS8dv{efM276pbl^$oWFp!bLX&E^d{tVo^ex&GODxtq1c3{Z#!~tn9A}
z9}F9?9z}ik$xtqLM&}wunqfcw$uev$SCyeQWQ`ENR0fSi18@MjI#+45sx>&8jvIXA
zbK9U5iBV@Wu)Z-D#H;}JTO!$@#*Oat%*rSkNValx))Z2zfYF9!s>6A#4(BP;1w9~0
zbh^qn9`F*!CWsHBw|m)=k?@N!ZXSKZgQQXq-=HSZ@=nlzQ@9mqQeosoPcqBW^AK_o
zNVNcRomvv~lNjTl<YxRzPGS;fqP;esSw|N-V9Q(tT!b|#4qVbpLQjdx@+kGxXaEHq
z;}QoAF62)%VXHY1a`>&d&uWoPco>j1ZgpfkFT8>o#VkMp_xVH@m>`TygyY*W4q@_!
zpSey~pd-<ljcRkM2Tpff6vh+YV2E?2qDS7ioVD}%eDGIt_E(q;a3&_&-poZ6_)c5_
zankt=7H<O%T|=DmTBX(Nc5FwUZ@-Fxy)GZv#hfljd5+0t`I_MUYXY9aUVSvUfx+44
z2PawrO#vRB?%X`OOKZKEnYoG-V^UcO0WdNV!0#&nx$u*!yp5r`-49Jzf$<M;oKow~
z&5F)@mdP6alFI#oqudcz?yg+rhG$?X5~}psi9&-z^$CDPrT)lK>K<0=Xs%MH&vg{Z
z!X@(}f8r?eE>`5dxr&UIP1kERh>+BQ^9)=9&o^tVWua>ShXccYhT*=P7@EBqj3+)Z
zE&R(-0)v)?F#M+jLzQ6|_r$<;?j;!Yq}mQ(4**0I0ULutsz79{wi028Gw7)%gZ#cF
z)c+NumiqDZ8x8#!o}KvR+{7=Jm|9g)1R{3Ot<a0qz&}Ext(5wgjWWqRb|W`>%BI%r
zTnXd-(1QDp9m97G!(~7aOJy7)DL}N9=~k~%D~$)G*;b?8nGR~w*UN(_eKoUPM98Or
zxifb6?aokV$8O){5StD;TpA{gbhidgxlw(Dlp+@drI-t1FtqIFJq0cPG{1&1PbZ?M
z;zz!IjfX?>IQs49U7Q8Sn6(p}B?aIp^K-FQYvIn)EI2efCzex+;4CWuXE}bO`D1O3
zwK5BiG2tdSs|vtTX4NU&IV}s0G3h2arx$>u%&i3+i;M1zEI7v0o8X`e(;~|Dt+zQ{
zbmt-<-9=}M&ho@8+%zjEA)^%Grun((*Opxd7iAHZ6Z5cyS&1l3!IS;k&J4Z}`4V*o
zIfREEI<PTMPQ-6aZai9RVT)xgWGJf!>o5bcr7UZ(9#g(GVB0q0OETg+T`X{}%z|^3
zgtI}yc}jjbBZeBVKzV8wl&483S4$|@91|$lW<j}5LfI&xT;D$wi^SiM1?5Hw<t7P*
zCb0hUlH36ObR=c;aSOOxvcPSU;5JKeG^@=AE;^rPjnD<9t)<L$6(cCT-4FQAEZ`#&
z_$~>YX0!Q#hj-xD@p@A@hJR->_AbnpX20Bx8Awa)#T4l|n$G6LhwcOw-!m-uUVtvP
zmAmJ<)2-%iF|U2qym5r;zAVTGB;=}uOcPvx$jp@ZBeVr4kcJpGzXX36Z~}av#=%L<
zKpdRH6mgK|y@lY^kwyxfgIRFoBspwII5Z(H1gDt+XEp;4pFM~7ODr@~E`+6%fu)y$
zB|L-~sHF}|I5cT41ZOS-&a<-MJX^wfK*FK<b0Ij-QQ=rr?1Ncwo-5%zPr{*zu>vQW
zgsqUd=b8(>$$_?h!}cZc31GBUu&TNcP#4{mm6>Yi;81C{+UZsf26Gj-Yo-^Fw#Qhk
z{9J)u`_uUWWn7>+IRDYQ!u)#VlCFW-#>|WrX4=mAB>b3nbFZ)Tb!qL2J!8W~!1uik
zh8e5Z#&*JLuiH}n+u@H9Bl$-tg?_{6vSG+I1E>Bma^)NZYQ<ZJ+ih?bvjDdx^bDeR
zfe8&?(>}^}!mYe@>*j#=b?l7Y=;2RgF(a_KlP{_bn!Dp_>_Y~ytdWU~W%%hyma8&T
zKT>a&Z{13z?ZiLkO|tQ3D05pHkKDcYWq^G<#NV%18{mz@eRX;*;QT*sw8Q;#1G}qj
z2wVcYGCKq7lz#&QBusSl5JE;!5E8;9&p>w|Uh2ej31SOV&JxByK{4)5<1rUDg%~OY
z-3cfJHf;rP*8|4Kotk6OR7wzVwF$?!hDtECp!+~?T-}6dq}2ulL&3Pl!_G)6BD#v#
zg`%|%za$QdEwOrR4uaXvEClk!+o$3BxE@7>!0SM9sss7-TSpx?jvl1DMiaS{w~S(C
zOzC3ILS-rS9KD6imDKZ4%=8UF;U4%+TP+R8@Rw-G8pg*p^SD+jJatPYBeV$K!oi0_
z@JMvZnyJ~I_Fec|q$yWkS@!U6(F#F0YWa3g-`-{Z!nhF}r_~uIPyEm;-rE35vf7?w
z@v@n$K|8T?ul_Q~X={=Q=WZy(FG?DT=0*@n6u}`bnvGVwih>HK|7W3#3~(Znl^a2J
z4Os$0fR_&O)gY#SyFT9Qf}}wKbjvviLTnKm>-m-|-ndhFi8v}G)^-eOiC>fMJKrKL
zk<nTs(HU@Drvm+80@MV%cmP(TLX3VE%T2D9o`sek;|Jo%t2^0iG(;?<6jdYovpnL=
ze~`zVk7>^oG0}DPCpcOY7rKI8Ahm=+)8$K1xgioxSkHr<%5x=%(^SLwOdvSe><vtF
zJ`)N0yi*!vvIavEgbbepy&9BmfqNjD>vm198wJ|{Rz)G8RJ*jLSyO*fNI*0rCQjjn
z*y1=h@D{04*`g~KsIv@#5-&-dF0zQz#rWH5cqk$Ea|z4(6BjMo4QBzMzm=#o<PuD5
z;b=fdgwYawZl?$jTV{#J6j&^PxcsPdsiiQyzKC<fem;gu-Y!ZY7Pu0`<)=O=#x+7d
z#7=}#cM;XCvG_bNAYnCD!|vTiYY48cZqzPbz3!&czN@ahYN&M0)2`bvRD$G{L!}K@
zUVAO3H|(b`pono?1Y=}G3D1g+Ps8@4aw}*Bp=tU6(b%UKJH-CHc=dh&@T@hG)EZEs
zbT7o<zI`P`1H(B(#-VX+_-@1gp>nx=`Oq4@c$EZv9{>OTp?y~lT{(0W{@?!nmp>q6
z{mJ=dompZISr;!!)-Q3$x;D_|8U9XG<eW<CSI$wuNwpB)6;Vgc``D6o*#jAiYw>y?
ziJT_pcA*8$SYj1sH*PT5OzQzdX)jTF?UF6}m*_xv(xkz!>kvH%!Qg!z3`9yVgi`eG
zBSiPI+@cG?@h5{e4h39#&owl)@}2~EjYWX3g5OjD9tkhUTx;<m_S6YO@Cy8KDEY==
z-Hj>Y&9A^f^T$x+WVLxHapEjWjZoPbKCy;#j4G1E`#NC2AZ4i(h6b9Lh4A43jt*gK
z@p4g(!rdrMc7UjKsL_}K+2eeLL;QFIz9Sc$Wr#<lE2WQxedIb>A;t#rRB}v1mIvCf
z`wbSiD_~>8c!j69hf1wkGH+RwM|pos3uaK1muj(X_nq)JJ2qZo7n3OH$LGWZik%2A
z4*<_9RL87BAY<vS9XoasQ&UOio9y_EgV@UA&Xii*-fXbM`BDKh87`fYwcy?5iAHzP
za!`j^hq7)=t;`OS%d)ux%wS>JC{o&Qmsq=2+6Qab#td7q4)P1Oi;jgnoK~TJkPwQC
z5<-F9Ma@Q%vECsW;Rbl2*u5j+Q!wXK;i24(#iz>Fb2xkdLQ$L41Et*NQk`%Xl9`sJ
zJDZ?(_kofcD(wIR0aN>#HMpDf=3wXYAz^sh^;)M2f&O6BgAG(L30K`9`$qXI@)dRV
z8(NE0jh>~%uG<+ZDar|B^}rCoZ=e#htpP%dEZ8m-@}8grH(uTiw_|O`+Ob-&&lG)}
z_6bbK;D{*P?o*wZrXdBC=?u)M(BIRAd3cA^D)CMXD{$<Tek9LeXSM}Vc5wzrraPZE
z$Wi2g!K!jC5M2|(a*VZm!2ujpVDd(vmkx`qOo`ECh|oz?CS5S+kn`Zd6QiCEj}3|W
z;L+ke;N$a-7?Lovz`fXEezHMsrJ=jJ*@969{GiAcHU!10EuMxx4f{c^Y3P%TEpNC_
zPWA<^!>=&9Yx*bDb+1MshEhilNO&#0(7Nt*m~$gObkD~~cs>4j13s)a(v8VC4jbtb
z91z&9KyLEt18$08<l!1Yu|(x5veJNn@*>&D1?1Q{H_c8>e&hjt-ir`<!-@H=$p!Aq
zxCt4XONgFaf$iVgu1gdnQ>Wu9TC~BCC9f|@q;5u7Hjx^J7fF%21+!a4Y7@Q@A|-dJ
zqq!4q##gzjexbDjvgvsbTIW-1<s+)_76fCf@HTj%RM;-{u%e_|aM8-@IE+cvi7qTQ
zOt)%q;7B`s`Uq72VeL#GcYU4#_mjc)e)Pmv4hQwAY1lJQ3!>4*Mx|^SsUo+=6%j`o
zRA`<+&W44f%4{&%t>DFtAl#Mv5fle=sR727nE>6k5P-t_z*EaOai&ptr+P^2E8h_p
zCAJAiC)Mh3;j7YuBhf1Sq=NohzMR&V%m4BPv48=k;&X++V8yNl>i{sfL$rL^i3V;F
zg^h;O0i!O7B^qU#MTE#rEvBfC*_jU8&HPYEM-l%5j}@rLGmiqx;g|#XamT<zu_u^<
z2_v3L->Ixxe)vxKhy?HvlnQnfa4-`0(h~<dW%O`V;Cf+BfUjQKs2<xz%-2AT%tcHK
z3K-^&uVp^A)-o@&3Zc2AxjU(uXh}01g%_%FqGiqSPWaWnv>6V7Tqh46jL_MeW3#4t
zMsC?WzO-o{-^24!DUoUB5F0VI;-{AGvWN{?_IpjSDWO;xCRi><&hhokU(|Z$NG>Bd
z+#+QNJUf;#0s>@7CpwLP_gI9;^!afo`991ltaZ&v{*2@syD7Z)xRacARm3w=i_iVI
zll=d(_a<<1RaO3YAgskI`%V#-bV8@I00Nx^OwtKNl1@w-XbG{X?ta}}FV$5|Em>GZ
z7{nmYpb&v^K}W^m=O{Wej^i+nj_aU~DDLxf)EP(o&(B@aLEQL%&$-LH@77zZs{=TR
zeu~#s_uczF=bn4+z2}~L4(*ylHqD7fa(6QcVwPQ_JD>*)Hc?>%rqxXTiWEa8Lj3a`
z*$yaEr#>T^PJ{UKb{7zc#8K4UcurI|+C~sof7eCyr8|Tl0!uG8Y!Bnxw1({ijK2Qg
zZP=F9tP$e19Gtt+Un%V#ZdLa!Ojcn-G{oX%PjgO4Kl55?%<gBlb&gZ<(5bG}`b+p&
zmsTIwr5CcOB=6M8YP})P$b1Vh#YmtN4rygMFM<{Eud>W<Aw*%;tO@)OkogLgtg3GV
z20FDaK_rs~ITUBKWz8{ZG`s^N>Af_W-h^tH_Q@#PYl=#>u&GnYhWa*?cK4*K(o3;#
zqryRKm=u!%My~9eAt)chC@?KmS=R5~LsLJ6ZN~<c%QksrJL!HBc2dM((l=(jQ^NT=
zVg#4)E(iwb^1bHaXdc!n>g8S3*u{L=Te$`sadLLLUq(PHF(FCw)1d*!b+N6pz9WIm
zVmML}t7ZjhpeO;-G_@R$!ptSYmUz<>9d$8>gIK$C-{3s~VWu`bxg)l*Dw?wpc5h{n
z9Qgk?LhWwh9pqPLX*BW)_fC>^(}54g_j1b1l)v=ufTlYVRzz<<azqJ!A7@bU4pZLo
ze8#+635M@8+|NGavfh^Y<TgThxUQoQ{@t5*x2`U-xY*qrVvS+46K#VbmmCNX?50z9
zI3KAN*y&rk`iNy~Rh@zH`7m2t#o4ZO4=2YyJ&X6`w4*sH33v*%TwzFXJ8wCOM_@II
zK7)GMXCP4da^It97JQXvu|rXDco?VXXa)#F{fdmY_>37YJRGk_Mh~h5j)OwZT$=78
zOe{j3NG(UwER+Z?u!JcpW+2}bki+WY;IK;6bY=yHwo^B@AAzzGYR-Kzzy&Ps65|^*
z#xuh)Vrcs?B+MllVOn*xhm>w|2yWuADvX=tH)eFR#A=09^l6EhJX&fo(6TmYnL;9Y
zOL|Bkj<D<K@gcEZl##E-`vyv>jnd;JJt}ZUD1W63BuwkXC6T56>jugUGODv-coBS5
zHdX{L=J`1*AC33hLg*dB@hA&skgf>JM4%vp?akQffg0L_dxXBPry^>x1v_F~wG?L9
zS;tjNE$u-+^!jG8(vWxJ<hE*7xIPG$WYAVL-9*U^R$(B{JjUL=OpUP!!Qi8L)g5gK
zMxcf%VD)lPqm9y5Tq#q*MI+`fvm8UtI)9lJ7>LvA*~e*t|4Hv?--*N+$t==y!bKzW
z1){*+Mzf4qm8wl8PvblAJug0WY(e&Txq*sD849ANUJ3gMABCwnB1U0LVTOBS1l*g$
z;l6STaF6p5h-HToR%Uc}5~G9fF!)=1@XEPE9HHK-(5o(>Rk!<R_%K%5Vs}6I?_{x)
z-f7&1x)p0IDaS|iG}PfVI0olZr`8>QZ8*4`UK`kX@cMVn+EwN1Hs!gYHYGi|!1|_u
zdx^Naj5T;!T|W+P7cBqNjp7Pp5Vn6i`8taDC!+9OhqZP7x@Dk*v8}4sHagcg>2k#i
z%7s!I#T9XC_&VwTI9#2Ws8(=6kA3R!Zq1><odQ2RA?M)QA8t5Ss+7aTspDcf%7piF
zx2UpS&6jW?P2V-Zt$(`UfY)MKpi_!zw^<=OScx+wEBGEWvlv}v_M*BWs4Q2l3-5$M
zW=L&NnWvN5;5#g}xBF77#FI6|9ZF~e8lr9B$NC)+Q0$BwE~Q~B;D}ck_G=<w-`R24
z+<#=;Kwxf|1}UtqoTwvVhrYtFUmF423&Qq3v2LqVt04_fzFwVB&Q$_KU0l5$Zq^Y;
z0XbOVkxbsy`quo4CZZ|8yW0AtH#aY@?#@HPp*9f8QQVI15_s7_fyFJw)hW~wo`J@c
z3^o9bR7Rete7RhlDCV7sAqQHag6sL@eMe&*!A5a?#s5J`y8Ak!xXmf$6kh@aL~<lm
zceiCHMA+lw-{ij7<@ZHUU%y^;h8hsOtGq{j8*Ybf@t;f>$)|=P)W(K*3wy$=Ox?Xv
z)V)#B5uolq`&-_hK45>#wkTdWbcPGF!bq`C$yi5Rl%ykk39;X2e*Ae9Ki+K9_!cVp
zR{Prr>D#xdZ#zSMi+t0Ggi5HKO}>L~GR1Ev#B%K?SEKQcwh0n@idsesOD_k}wFNd}
zG{(QlXy4(Z)uWMbKfa582%dw*o_4i8MV&T+8!BR>5JTbl;c}StF4OpsPoo-qOyXaf
zB#QZV@Cqyh`!pcKKuIpP=6ysgb*jx8F6!grLjSsfmA!d61C)1ki~EElRi|E=-9XY>
z_~{~fTXhsmD}<R)Y28NXdS(gNO60YhTyW|@*k4*E4i_u5v{FqYdQYCNPU`0Ab`gN&
zt3!3ChTRrcZav6n+@$M0xX~#N=kj<GS&j2Np^R6{NYcbc*U=^g^tYz1p8Q6~!5p;d
zGR!m0A-rC1I-9%38;!|&@BH~={A~|bhVyCA&%=o&&CmSDd@fs`zwqJ(7tKk$!I&WX
zhm!~SL~(ne>{P}Y<8o}C6)p>QS?>FXjlQSR=v^o;QK8m6+4&|LlqwKYGfr3$w1S$U
zl_(}^T>q-#!Zo>7qmz?Ps1hhSjL`Y}$~W#i1A>Ns!eSO-H4VNayq4c%6Lh8|i0)eH
z22}6}*3MG~{~`h+0}Y8WDtlsfbWd2txij8tV?W8jE?*D}_jVvl-|(h1P`jJSSbZIR
z4AaLbeYo^7K_8R!(V&m*^l>wN+)5v>rjOUr#~bP6FX-d{(8piX$D{P|Df;*_eSC*L
zzE2-NrH`lS<B#;Q|J9ZmUuU?BGC$gR7_*umjrZlWL-DCaHbxOP@e%wj#MC^mbiYa&
zw|KIDo}^qn6ueWJ7G-A!{Qlofv5mw1{#T0hbA7GR%8o~+Vg&sQNnwQFKL=rPY6TIi
z0nzGd``!~)YxaO-yFyOrLTZNg$>z(?JKnVf^aUK2jEQJD(9^rCX}PusogRpG_c+E!
zAhi<?GD14Y_Tm44uhsMtiuUrR%SL}(ShCLe?)Q6#ip{Ci(Dvn}e(!MFHA1vyu&$Ok
zSN+~$lch-q8&Dj9;n+MRMp_ovlf6ODrV{il%*Q267?qm}O|Y>(E^#lL6NU_bXB_;6
zT}cdHSqQ6JtwqFXn)^2)qr?u!mPNX}pA1YT${On%vVN&6!&t^okCF+$7p%CGF0@+9
zawv-v<+=I*8E#TxJ>~GEDf{VzA^pU0&OSsc&3ACg(pQ7~l6iek$2r@h{n>iYTlcWo
zTj(u>`}f4*DAQc%Si2~x2gPzl_TYyjdeEGYGWFi!M(^qP2hvMbZ8Q|MsiM?Tl!fS+
zUC(vDu~Og#WMr%^>DK&c=AH~((g_=jhGmDhO#c|tnf4`}Or6<lbfzD@f+B#JK?mi&
zG5mxJH#A!AA#^rY)pINYk+)3}tpK6J_kKNmyTTQ4<fZ+8@_WDiOZ&zhdH5fP37z48
z{4u{py_mKAkNe$+L6N<_%p`@Kxee2nWAdK47~c}(Cn_9Bhh^<IWPISwJgj74eI05g
z`?%lhEN9bZB?}!N&uYxDgGFK16O5eYK50YlgdHpt-T`DeV3xw%6$LNM-1TW2;uOqX
z3R%Dq@-n2c3p0cqtWWoaoz8tg89zMBgWT|4GTsr!5v6h1fi)-c0G=p!%;pU&?_icG
zGC)YP;L*CS9^)sT&pzw(MC%uOvz$efz%!rm$wbj71|4iqM|6;|yGBi6Ub?d>v*KNv
zJu9BUxyc>7-3PJ#a^FTDZ^i`M@4lV3V<!uzJe345_)2`vJC58>LMNj6b8-^?PFgI&
z5^nvL0`k9u-uWIdxbMVMzxyuw`yqV5vcVDixl(@ON;!7Fsx&11#J)6n#Z~Gj+?q*6
zxzZD<mqn%Sd(rkFEOs+;b|4GWN@rNT{x?Jk=xk;UnX+u0S8Z&%!QEmr+rUy44Ie4Y
zR4M?OB16~@mU7JdZu0Rly`M#vv+n#{+8X0ObH%M5^DM+YfqMPZ=ww9QR{3v2<mjX%
zInpr@+*7n1oovg|XGo3)VmAJ{UO$&ZZ;p5QbLYyaV2BtQ6J4D`h|#|uSOVz3bD3y<
zM|oa}Jq!J)Tsm-?2iAx%{Q5n?q=+{oQX@bhdc|Q#UBXrqo^K&wSkyX&5z{{_14g#+
z>r@R_F|`ovel>`HXZ!;b>8CT!P#X*Qcj2HCUTBj$MykX3VYmA}$V_YszJ;>ia6;M=
zq<<+z7>p21KM=gzl}<S4CJCpup}5^zI8V2Q^OcBanag742o+YbwPNYPtrWTo<8POV
zr`-9pM^#%cEq63XBcyH*$g8w~Xd%rU<DG~LMvP7q=bhdP51XAW);rY`9u@n!jrZQ^
zK{b!%t2aDWYSe%_lSG|NR_+Dq2GoT~14^5;+>7;qI?EnVSDFJVjpZq6K#B04h@EYO
zji-*+3ibIusPy0$7QaXX6%j@M!be5}^M{Ib+JGw2V|G(q-1trem6w<j1ItsnI4rHf
z<vTr2j^l9>H6U~hdv|%d0nwK<Ahc!BU8M)aY<oa_z#I_LasY=PHOVSttg_z44hP-o
zBvWcG>0GFJ>OmtyJoyP>jX#BO^Hzl!Le4X#bLSe7=XsV4i9Ijd!1#cUm|FaK*(fcj
ziT0E&qIk++^s@fWtTpYqn?yp27Zz?>I7G3$G1PJ$S?H}0YaWWntdj$X>$M0a&G%@<
z&V(3ge;>Lhd0DzaxiM)_>JUNhm3mNesC;v8dk^hN)UvOWgK|1!`$+M5ek~dIu-F~1
z#4c`o`-*g;bX}4t=`cm^kQOD5t#69byqSh5rHdRCtK+fXjgE;=%5HtixcE|Btk~><
z%|=3id}@s+fsKJU`tVyN*U^@%`MjI7cEoEarhi<>dl4m{T9ST+eqi^w*IFXwQw^Pn
zmQVFC=jeea!%j!dr;2e6M8xyAdZ#B#<b0|q3)Bn|^r`9$QS^=3r6ZHeE5s;|7zxM+
z5ls2uC%)W7;@T4w*UgxaD2Ib>3v+eQf47`YVN{Y7hK_yZPHKh05&cbtF~)OGd|3TZ
z4Ux$tMtV6evi|dSGLhlIqGIiLTGSqpOlq|<${%|AiVXznfx0VBPL_8RhR2KW3U_ds
z4A-Rf%0`&BoC-On7eYf@>7tnyN3KpP;NF~0NN-6J(pOOdK2f+;3+biSt_4RliKf1-
z&kg6KA4i`yZ`_%VFLx#J#nXIwo#qRt5-<jD8on$_&zI1J@V)8ixi5(x?ZV(bpy}a^
z0@>&R$^Km^_f6@*y(J0UTd4pG;e#3+XBU_XxVNVR_l_iRI{2IWP7RLp5g6bG-41La
z;0!T$uAahKWv9v(AeFb}77<w5`R5+7Sp#UIS5g@9e(KYY=XghYF1SfcgFrrrG^b%7
zXcZfgCVudT)W<20Kyerpa*X94?-?Dvwr$6M(OrRaLhdzr>JXckoZjtFZgK%bA4?jf
zMPCOH@9g2tYa6SDoidyC`^j{^3j*MNfDgUM!NUDL9oMT9)AztrVa~0)$**Gnez)gx
z{PZ_Ts)bWou-@Y*K6&S)7DBajxgp1}k0w>>ALsw0)>oXRz|>bO!(=%mmcZUqeUj9v
zor2!w4|~0?B+0je8d@j*9MTS(m1#Wq|6;4E3zN7fk|fSQ`2Vt&I8Jb2O5EOqf087c
z)&Cb9{BKCV>dodEATFu9-%cmF-%XNS|IFq;wB&MD1ygeOp4sS}!+XzcFoyp`R>++@
zv++|mxIaoaFn^LXF#U6zpX!0hxf{%Zc})j4UBb3#tXWAz?}2lhKZ#?%BlY=9m|x+q
z)5*fqNwVM{y!=MX0_T1(WdT><{WHA12QSvai?%9|Z>0b41!SHd(H6da>9RqxyXt;a
zEY9uOH_z`+MS1hZsU{4uN@L;6VELQ|i?B{n;U1BWFE2{s%TZK-H>F2wzHr)!jC{E|
z{Ti^2gwRf19q{le;b^&9Y#>KB|6zQy8BWTE1@gpnM4ysG^r@QY(=^eXZz3bnH>W3B
z`_dpyC8Y}Q6|~AQz|BUfT<@XU3gU4C2W@0y(Q4&|v%$(S9$?K`=~#145^H?>=WNXy
zPI6(e#`~|S`cF*J_}S)k>Z`)e_`w{*R>wGRgaskJ@}XAK5e1Qt)m}PvhI^jwmGkMz
z=#_x>+UeRhEQ&7UUO?}N=rYQZ6#|DN@x&=px&b7V#w#?OLUuSR8h8v8jt3M6MTfF?
ziyF$%v_RpgL~&4bta^7uLvgZ0G2<Y|aE9X}#lg{m=-n|5XS@wKC3+`OW;jw)92^~Q
z-rb_%l(WN0Au}8=Dh`ScE$>z|l<KrV;iy${P_CzDdHStsDD`Q9!m+dBpfoj<ts2U<
zX@SBKyyBqj&`@sBP;Sf)#l4B%Nfcd<B^C!qhjDjbso~tx2Ao%=g2PeG;^5py)v(O`
znTB(F8*nIEq{W?(69|s476<1J4d*o)4hK|EKQ~eMP4Ap^hI8oypkuJRcTua!I|L39
z7e~h~O~>7u4vw@wt#s(%?C!nk=-?=Kadh0L>9}9haf{{SF?a-frtgiQ{wEtj-B|1L
zW46tI>++F@=(XvFXm@;Ic*A9nGiga5y})+nZRrLBM|+GL4DZl`;hlOg+-?nqgoSf<
z<s^?S{xTg69IP^qhWBV1{z}upf!(K5F-aQUpN<BOwHZgl-)I{CR@1;y-VGYOU*%r-
z@!YL(*O6hDZD!P=<lPUa>pYH;8rOOMs5|c?y7M?JywQ1S2GQkhsgGXdp(!l8a$w!s
zb$u(>6b4?teq}&ad?Fo{pQ2)MRDN1h`58?mhlkHdrQ0bgTdwyDH&P=eK!6WlO2>yU
zC-LDcnh#&qeBiL~)5Qn!+b$qTc45MqsMdC{1*~)f#}e($vt_WGuOh(nwo*rGgUMnI
zVYQudp<cQHUbx885a7eVr{lwSlKAjl&4>TceBfa5+4z8gVOTJ~mk!+blfXTx!Tmsk
z<Jj=oz<C@&BK1A3AE!glv6bV7+RrrlpKJ6S_1!=}cr%^x?2YZ-vDog@F*6;f?n?$e
z+ymqB*Xe|YV{yj`jSiLNKBI+(!^Imy<GC_?XFhAZc#At34aDK=<>1i&eL6n;k&4Cf
z;aMuc+y5C9CW=q8ILf@qhpQAH8iorJqEjG1KEB`?DsVQ~u{}bN=9z>rd2de9*JEkc
zF}>UaHO(BI-lTc0NwXE6e)_vN7eZ=?ZJmOp$C9c8&%1|dQV*vmgVcn@7MuEypmzec
zFX`C)i|o#zd;Ocqo9|EDzxjjYZlMkwLj0SLOOjj;+0TlJpI9+*@O~k=q5jRMCROVD
zH=jmDxl)ecZVZRL;mU@Bq06VH?e1(jJ7g2cL;k+Ta$cTlxH2;izPhv#jD0V|M+L&&
z%YbNc$FECBWUT2YL?>U8B!vF)>!n%<UqVlY5C(K|D+^gCFQaz?IyuY(dwCKZ-vfJv
zhQsmN4LE~~NX-wl9ePds5LrRjNfGX`P-`T|mT!}Gr!Gj}i@J~0Sn7<Z3$TGzPBv>y
zYbWwJI6*cX5o<;o&el-RSfUqly3(K2*M20kb-J%P=DXR~t4((}hr+`n>!2NFMsbcp
zeuv&yQqGZxLh$^IWELV0VfW4yq3RlO=S85cpj<kRygQgqHm*sM4L>N?CM_ErP2Q9Z
z%<8_YWT%31`8N#p6$c%52JC*Wl|-miq=*ww0w(`;L{-EGrPAT(-H~(xF`6V0{so6I
zEf5@$-V}&iOo8ZR{2Lv22u6XCe6gM{k_wubY;fW_+IJeQAx<Jn@R9-+JS1~N0z68v
z=WkA*k3tj~WJ!q){kJ2clPcKFWIDO2CCQC{UA(U4hGW?qa+9#Wp{X&r_iv^Hw=D^r
z?~%A&gX6IG2Dm||142DTr$p=aam0=4&fmx;k+T*_Vft^+iEi7MvHvWc#JoC5Vth;C
z9a>^IWWFIW&y}Sxd{G%3Id`Vx!(B;y@GXU&<^#v!H~Fv!EQM)kzB?VwZ%Crqw-nx^
zY39KGCe71lDNIA^8(T@0rn#Sx7PDgxgrqPq?rDcF<^8R?ao1_>-p~X2=?x7|s@V9K
zb|DZOcHjh(mKd|#{YfK&lS!~yj-S|q_UH7H91$VA(zhhl`Bo~xbv{T>TqmcBFh*JI
zf?+48-1biP1vP#<{cdU|xScL?s)B@8IC8_A=w-=5G19!%l=5N%XUxMmO5TBmX#;WD
zYBhwmlWWq(j!6eQnY3LP9vFS6`*-lV#~qe);cAKhycy%WUm<aO@MSQDgh<oz3`&iE
z(mO%k4Ph45dxj8*%1q|#{&Wyjagq*xqMU%<QOi+EQZQVGa~ox%6MbUlko69xa;y(z
z#*ts@%C4j&B_gAUcVKa7xM$h&6}v^YZLj}wr#y*Rt6Vh}>%|7G@VAym99ryC8q|?u
zZA2x^9<DhElSQetQ3Vc~VUs{2D#U6nq2MsM3x)vn6$gu4C5K9&U1qWpSKEIDZH(E@
z=(&*~(*kt!ZkeQ{os=YpC>Z{6J-X~uKzOH;yhrGdroi_ko{^(NAJ=$CYn}pMy<Bb7
z3k8w<#_Nvi%AU}7%=C`?cFmD&`iQ{y5otgqEH8xYz%yR%X9q_9mbc;d%TDBT`5Bz>
zyf*T<4C!IelQxe_@Rj&HE<BW2vGEftHhz+dE#zF}7c_L@4s3aBah?3cb@C_^b%vgc
zJe^dj9vAL!s3=#;;~=dDmPtsCh_j#nk*W@AT{pyjPF)<!0X<0EeW)md-It!Cz_y;J
zv;BM~y%VsXOLcy0uCtoSn8f*Us?z{ZLr)Pm<uoKFFYf{x|3Q8ITuwPM0aA~gazxO2
znN?d(6sJ=c@T|j6q^btXn6Q@R5oU89<s#14n-deo+783npjmb(F$U&}v08Bgr$<h0
zqEx};z*$CMCc>Oh9l><F#z6=#swqEIE>^Z6>V3uO5ph&-ZL%S%EY<VKNdgl!8M*7@
z)n<7_1y#jlldEoZ$UiKPNolxNnru+Ob>HtH2nBVk6=8WtcfN$#zqlh`EZ3_!8KxEU
z+EA_JjM~*X+b63vB7o;Yn*S<@?%%8oQ+xc^d{Er17ss5Ovz=m=I+%=e13M;Z$svBo
z%O5y{%xJX=N9YwaX|JkQyXL@%I#jITEBw&2N>;qQSi>)BN|!U`E?zIVmYU%{dJm;u
z7n8aBD~O+%$K79(BuecQKHCbd!gV}MbxdV;O_)yu8PV)nRVWO?;mr?Dq5<JdoJU<l
zbtKK4?*3ZF51=Q>KN5BKAo_w=Egx3XvD9Jn@QX-OwzDZo3M7AHjsFrX`F6?$gy=T!
z&_2#xS~FK;=V<_s_#J5Hf-|Ny(jwyhX<eutEl55tUUa-;O^WQb+Iy!eP0ajb6_Nzm
zidw53XF&`u*A~S|TGX2)9@vb~JmtOM7j4T7Vb+T_nS5_~_T+mbLd3=PC3@X<n&7+2
zzt{3s{L+o*BVK0l6E8CONh-Rq6nwL2pk-aVu~Jo)hU}itPO3C)?Z#SGs*?8#QF<T#
z6TYuZvvxa|P$aM2NNvS(AZ+c{LoY>veW9-1Vy!S|62&oAm_<o!WR=d+$xkeutlVP+
z>7t}ceL=dIigKl_V59mb+&t;UZUQc)s*?qYR7OfcTCTyapr<IX5rWi5@4R3)0p}C?
zfGuk9?!C+BJ-U2;E_YaG2A3A>tK-xPfvHaJw8|tYyaMN^y*sVv^-k+9-)!3UPV3I8
z+A$MW!)bR~7p4%>=fRHa1`LpPc3ek?oT$k$_bNT+?)Llh1+(!wC4?}0|Mj}5%2-GC
zU#Es64Q|7hr4hq~ivFZ_WxEW4w|8YP&Auyh$Whdue4=+}=LC0Wk#EQE(s(Z8bud5i
zl9iv-Iyl6_UZqi)b(h8pSydXcORFVS>MyQ+3p;D*sP$?)i|g%#B6)F5sxFoTVY{>&
z>7^*J5sPcXOcq$Ade0XNd+aW4k|>VZrQMRmM%Mo<o&3bo$umKWAiX-NQeTknprTwU
ztKq1=X-AMesybPaNQI;nq`NiP*V9uJ*a$&-1HJQt-KE`1=mP^_?=G#rXAe1^%Uzl@
zM5=SJ@zKx?+CPwx@CJ>ac!Tx~GC~fF<Ae(TFG@cNaxuX{(dDa5({X;=cxiZC6x&uq
zdNH_F6l>KcPLUe5Vx>M>txY&J<sps3rJTCfAl)x;oRnYP$dAK|y6*6iSAN22j8{kS
zcM~<^`$C~qDK!cO9JtAQKuBaJ8YoqS`$n<m<a6w=jRvV@yn$0SKFq?|+dQP3p}&2<
zY|d$*DwZwg8^tXU4%DfTAmK%G!l^XsIfaFf6bqCH!@4{sAnANEl-!XRL%<jpyuuM2
z7F>aAIP$|WL~t|56$)}BamWBfxQXUcv5eRd{EnsxdE6SHg-_v(n?i~PbeL*Saf{z)
z=X9J{Zf|hyK+voa^Bm!}o?qQ3j2FcQ!c)-UUv8pU-vV!T@{SpX^UIKv9|d63yInZM
zD`tc7Iyx*J5t8@_51V>(l1>?M-dP&uRLqThApcLk%Ez02kNfFE@KkYTcjsr9tF<Ej
zOA<7@I8vO%*(?5(OfHwJ+Zc@Bz^Kvfie<+6sZ;LBUtZlt5{Hfi7eoU&;=Hb(?hoN~
zcdSGz107kZHu5{*VpXDJP?8IW4tpus4~7O+qbWAriEseB)v1(F8teh;(c&;T31*Qv
zgbJxYZiIA*j*)$!g?wufht17;zJJYn2rhM%l2IHvk2P^%D-Pt*1fWsBb7z!Bad1oh
z+s(9I-^6(|9W7(5Zo~0)Wz2WA3ZBb81~b0!mE$K?7>6vucjmd`=c$lfWJH|Eeu;|k
z+n=WZR)G`Qe)kLX-Q)D}HI!@YzIC_zRXlX{yI-U)zeFG34E&PON2(3m0%PLU)DIMP
zZ$WTVXRB6~aZ$uVCUvpfHTH1se^W(bS@8%S?+i0(Fu~mH4|TIWDVvpH1A>kII7OpB
zr83;;&vc_N&eo{9(HLy}7s-u}KPFm!O1FG&%9hE?G1%O%lbdr_y;RZqv~Kadlr753
zlL+F|7!S0p0=u?$>*Htyo_4>bwwS}eQ?0sBkXVb6pu{ALoXVPKz$Z=AybnF`82gG<
zvpbW1YNPT4b>#=?%D?7UE*%#8-EZ&+#YewCc(V*q_sBK@9-{#ss{#I71i=1w!JXOy
z?sR-p3~{dvjA+?GvoyprHN<cE5V`B!vs;>I_1z4se2%VsuCDyscIDk|R=z-2zED^G
zUBB`++*RER+k~=MLs_Ds{HG5^F!2&qx%*%A(<d;}LY-j47`a?GutGQReY=5`ssV9N
zI(6l%b>%PBmH)u5d`+8`uh*5oOjrKjcI6w}to&+S`Jk@+hj!)HsLI_R(@&ZB?q85A
z=mvh`H=s2Px!k%x$A>Y9``usAx4)#1VZ2Sc#}T41svu?TDVN@3J-tQ4d5Y?{RxH1!
z@BRlL0XR?7x4)r}Y7kDXO*mUMoNXG;GgQ9?=Xdno@9ATQY*;Oz{y<;;kv?v+eqpfB
z(l58jUnK8tqo&%JEbgH9n0K$CC!UFB&{|Q>adJtwFMTx=AIv)G)OS~#ZBD#IwRyL0
zbHCO$51_9Oqz}af+3vk<wtGLl$Fx46+dU|v-CRh!Z%N;-aKyXQy^j_SAEebwt0SI!
zp^mG?Pt5zvsa>&7e@;(<Vs{4Kv6ru6@4vOB?WKx+;B%PIQmqL%f5Q>@Au#kfLjQ~G
zZz+vHzqemMwv)U4l}W$LF;R|2X>b482;Mk5@Z<e7CXS(&Dag3_+HLg9vGkf<WtPRx
z;JxsfurJCn53pCHdnk0hB<JVk&CS#Q#MY245eRg_p$aGpF}8B#vrJ)Uq(8Q{Vd(v#
zFga9;6Ha%Y*00_9nzOZ3r#)fbX$<$|MydriH=yF7YPHOj4^+u)ftZ!T93kom7+Zq-
ze{BgtC1QJD6cx3EpTZWKHlS3Ok#$}hak}$EJK(iET12*)Jau1g1Zc5`=I|cf!OEBW
zMQ5QmXOT^)P#hk1%Gjef(1MCR8h~WH#hLBlnVd!GJ&ECng}sih;~gCkWC2%Rp|N~2
zTlS5iZ|<T7q)(=l(xlC!PHo~HAE=W1jip8cO4{5-zHNYt-t$P-=<FumNrC!6ei0Yg
z)O|40Ro=x3JFvJ4vS8w|O?ulqB~Ya$F_KM-v!H6J!oX#wMR^3Rhi?6K&J*vr05(0S
zB&rLtpfX-I@JRiYBu_f+iKV5Ls*r%r&jd)SnUwoyDN8!*iNzab?X0!K8{=hzH;Hns
z)1G)I1o{b@W_Y|XBFqvj)s`<x{J0Oio&}H2ePYQpniB&b#>mmqSUWP^QbiVAIuVL@
zVt`NAH;a`<Apq*4ED{ua-$0a7s&z&btN$jQQ43b4{tLXG9hFXuV(D&7WR22Rx^5&p
zFq7^!-!?GC>!ZhM=J}+nbE9}T{#i@+@6{Sjt_cdp0OA$rakOX(uJ(CKIsxAst9{IB
zG}F2#Qt?c?MV;ov67@nsK{AG)dr}0PQ^Mh#ni-DHR^pu*!J*ax=$_`I3a)AM3YOZt
z@pAJFd}}N>#U5?&-?4;Tzc&NQpKJz{Y->?DV0FHbNO26RP9C$fd~sBA$7A*^C5>XI
zEXQm{bBfOX;mwK|AOWq(fX<EpG`kakbSjS)Kw_lg6PGM;^z0&<f>5|_R|KHBK|tOo
zUib+>gxI;~62+Mmv(B>N&57t}!C&n6#4ZzOj*`9IK6``uut)J%$t7<n=BsaoCTveG
zb-E01ZX#TLye;uAh{C%_Aq(I=&;FLTAm`iPvZaFO$f?k!(_eV=6Y1j26*h4fL=ktP
z4fri&3%SVt_CfmgV)gA*NYNQA5+rJ}T5n+XuMRtP9JjIt>Gg@2!oix(Qegp5>oK{w
zCW*GWxWosbMfzOg%q8?g(5mEqhn7O$#fLbbnhMW!Y6|b%K=+S0DL`N9gH{5}n3tKD
z#bhve1(qLsvs-j5yG2{Gy$%%TogC<FgaaxdfkdI~RL1C-nuo-4qjPB=;9a;u_yT23
zHPIGS9m~zK1QAqw57B2MJyRkje;r*o=+u+ZbSIZfI@+3dS_=cED}`yJa#z{NPBM^L
zDI2n}Ng6q{AGBrUup{zhG5{GbQ^zZdBH!}%eNkQr!FZm%CjQFIYv^`J$k~W>dzHt-
z8tO~w!*l^$A?0z<i^(f8_PoO7%c@+*OMQ8eT0YAuK*oiOEg~-8LnRzuk~o;2n`U$;
zLmXV=Ug{&LGfo`bRoZp*GI}yS!B*23p%EsI*%aS-e4RL?^ZsQcYUG19eqw^<WsWEh
z*Pel9r8rcEZ;*ucIn*IL`14eq=-}{3Sne081UHl9?IoLuV(Al9QMwW4eoZy%ex067
zmVT4I2xaN1tt^!{%EVdWWAvuU(#I}kyUyR^R7<dwpP1Gq2?n@<m4gF?b^U8zUg%pk
zux#VXfkNNPjVm^+UO%vUU4Ouo>;3~(FX-kY%SY+&5EviPF!(7oj75wk5ynR?7#|CV
z;pJDX>%VH{hK<XJl5px+{APpOLM|>%;8k*k#>n&bojrTE!7fkP9Gi_&x!waSDV+#5
z>OIM|w({6Ja{Y#t)L8>7*REf)Y+&U^c~!7m6JrIYX)#f^R$zcvrgG|MPdU+P;P8;G
z^C=&4-@$hlCd{7J+Dy>9u&PAtz*DYTvu5q;3i}KXI~$xj&H%5*(dKH3v@Ho(pa^gu
z^fkG6Fd`H=BXlAvgZ>THQRN8!?wXUIcS*kDY|8^m1LuYM)HKIGPr!*CTT)%h{~c$t
z$|V~yk`NLcuJN*TzB)u^g4rUAL)R7uQJA(C%S~qvPFPE|x~RGY_g_*+Np-{Ly)>HN
zS}OMP(d!7JC*626Khh!T7E}!=i1H+A6NM<u_{gR$n^AVFXl?G?E!)WRG^sT{E2bqC
zod_>w)Dkq+$sYDj9u#e>A||3bO-4wodZQ<wUj;xa{1X5We{n>+bRk@hw^f%eY6<cr
zC$BRWOU2{OitFMY_F5MUSI@df*&?55Dr%8fZzg=ace^yxg`LEC2Z$KdLe{)9W7|+g
zS${20Pz|YSYK42ol@%P8Z(0NHZvyho-FbfG54e|a-V9H)R$?9pc^FQI7U-;{fy)sb
z(~%E9=fnaDzCeVz{r*~qx71$mK&^>e1R^DyF<R(K2Z>dE(iw)=IqnYduyD!<701Wy
z<Y^v%&fdEC@b;q?WOvAC=J#Pk%~BZ`BkDb(fS5SZe>^bvDgmYRj1Dr@*>o!bUbf5-
z;jYxOtINFvy4$=a;2l!H6(!6MjT!|VWdZF<5KyB$%jbGqK2Kk*u4}-JgzN6AJLS=C
z)^~H#3Zc7_<{5L2l7mV>hBBVW0$G)3$Qx|PNA<yB1u<s{=O&Dr%~M8sAQ)TnnR-?d
zQ$<-uH(;YXtxtO6phnE5`Zf&D@KZ*5K%wAVuUQ$daaB>qd6kXx7`Z7YE-n=l+mzAe
z3I)u;h`6i~&Q4&ZUz|}6+9*#_+!UIM{Gd=h=JqLL1t-TGjWdapq9|j$#>QxJ@{$g5
z@<>A0prKx(v~L>cyaayQFEQe4ZN#U@zL%vtIpW>VY*34rB;AD80|WgPHq4W^W6!8d
zb?umwU$S)RCHd`}c5F^LIY#OwO!lZ4TLy!-Nt3<#0`v7c8}QLGKrN&+j}V|+${hbX
z%`6@Xl*H5@PKjd5MH$>s3piuIU&0I7sg5BdZui*<a!Io(2{A((v7ue+dwH@}$I^MD
z<K)K?h`fv3gjhi5%waD`_$xN)-eJC88bfs1W^L*`hWanzbVALkN$OvgWl~0MQcPuV
zQ-_qnLBI!<rvWB1TNY-7<2J(M{28BW6}+9QD0q-loChlwCdsKN%J^IxpUJ@`o#CK=
zt;|d6T$~1IIW61Jj*>$Ng*wQ#^uxRCkD-c<?Pz0oFX_nerarBt?^98f`*hL<XZGnC
zo#|5#qZW1wEpka6E6XyznvL)B6)KIv2;9B-@^_J~AY$2J>!WGYMBuogh>}K;D9RKy
zY>G^AxT!PZfYn{i8EFptB9XMZ^NTactt}usKVKfEK|0&DY|KrVog_~221DJx2cSZ&
z!!s?tT33Y`;0-MR<@l#k9a2tMB;oD@tdq`&W@XYM$$Wu<-ed#qkwA-dv4ScWwmQ0h
zKvCV<Dnk2kGKg?$lPRsApe1U?kV?`LvMeKirHx#TDIQKY(Z*@Y(<b)Eh`*3&OK+5C
zn6I*7p5ssOvMkGk*wQP@Ylo1l1$-)F1kkARjQlnm`3V~lKSFQoDVj#dG*=y=LaL)$
zjFU8%U!0NNZX;dLN5@&RPGK~szv#HVkF3HI&P26F$p#P_WJklPPj$ZnF@`}?!%q@p
zS(dTiVPjXT7!>M|qQGvo-W=ix8spXK77a6Lw`#w{cz0TO2hYdZ#Ihlvg>#P1#6ST*
zNL|#Bmq^4GfwYJm<J1qOx~jViMl2C-1@T8pm92;)Gm>|TH6(<>y(-+ElIf4uVi;1N
zN;-zeona6{ZzBMYOh6%*24vFh7FJryc+p>N=n;pQC;9y9QA#o)3fFVNpD7Ru#Y%x0
zql_FBV^t(d8lRvLK@__LFR#PzwQZ$3S|~L*&jpbvjKwXlbx}kXsu0%o;pzm@jc_!;
zJ~DOF-TuR6uu0k(;}0TaN$$SaTHWW(>+9j)#VDaWYM^oh#s`vsPzn!H33YK|)Hh|V
zwBK-*G6u0aIwN-07X=xzXG6{dGU^eCk@IRCI1xjKzT*)-)zLf?rUVHx9}AMpF~(gs
z#y&ZXQtz{t<lQmmOge1f|8TG-`q{T?;|e1kFvtsoQ|9*m+;^!&M3!V)UT@Q~w6BMM
z*9><W$TG79V$J6hGXEIP9dV|(wS>b-jNHE2ppQgns-3YPtus=Ij3~$?++&llxUYwP
z&PoBjGgSg+PF5#VNfVc2_`7ZRZ}02DuU%U<b&qW3BhjsN@0G`B;DpK!kHz`)>XX$9
z2WyhEbq)%U<POA<LZZM>qg4bwqbhY^Ep=%X(mGZ`45-EVD&7&AFNdgE5YuA*3TMa4
z8rEb}med159Z8GiQ~dm?%QCO-vw3xoQeM!#lAvO~<y2eb#$JTrY6yqL_mpwc&4IGS
zJD7#AI9RiK)w+Ba@&PrFZoWG|iZYFPQ08+mu^?<~Pi_NoqOKWXo+J`)`ILUj+{s~O
zCi_D&D1Lop7H1wkVDsqC)s?za<2AgvZjFIUC;p^9jh;6oL>{I4x(Ymd9;mnjnv;lT
z1Rp7$QsQ^yswlJMO%_WAPa|L89#N)i)5`Ar=DBlM@mP~{;+5PI){pExL5!FWk!7EV
z{7N2ptFRau#WOk{_qgg5HH{Vkbe3GqEBzw^1oAqgf*e{-y*DqW29mdxz4<=j2eA_W
z39YSM97edb6!l3@3%|;(NNbySu%4mpk;qm=7yuRjAFa{^3Xvb$(d&a*1!NM<DngU)
zz&qMh&Jw0nd3;x?qTE$)vAW8;u&+lHkm?*u&G?B(lx~9n8MHzrlXBP6IqCp?Y@&}L
z`Y6%IBz<h7k5|&i?euXMecVGI575Wk=;Ph=@jm+aTl#pEK0Zz#kI}~$>ErA4@ooC}
zK7IU%K7L6b&(Oz=Yw>X)eH>07$I{0s^l>JA%%zWo^l=e=TuL9S=wmH?4B#UWhU0oT
zzz2?nKPr4JUZfn93o4tM7+~(TR1J5^+n3<U3g7a`{z$FLmp_!}Jx8B=2Z{#1<y+A2
zvcKJYxw<<b-}VkG7l)iO+&;?9iAsHUpla_pg>gg{fN2gYIatcgQ{)V@aqn=t;NGYq
znHwMZ%JSm6g17i`3|OqeIObrXTEQ-am+*){xQN3L%Ft;jAOc=;9qipCrOSPY7!!8s
zh`k7xgUB$Bp9bF+vbIP^GZ{_OW`FZq#r}6wE9uz(u+3;@|NpVSZDGGVi0*XXqsmDR
zEl?aX0q2w1@>c{nk}Z7^Y&o)2FVu=V3dOC(5`=-AVM)=uufA3>;=NR_ysIlcj-m}x
z+yw_m2|e24%S3i#uhoq`lDV-%A{sj$jg4YmF<yXisnSB<-MYySWNz{>yGiLU`?z*g
z<tG}&-q{I`Ughf87{-*kesnM1=@(wS=LvV1^Muc5Af2B3y8-kG_Y*&{zW7(7L-;jy
zxqqMnf25CR@gXG3oq?e$>R!QhuYJe92|nyI=?$*^Z+tH4n@V*3Gcfes0s6*05O0~$
zaQMoadx-VzQK+%sJ>2@1F1_{JH=n*m5=kOxwibRL$~~SSvp>-_f`Rn}FEW+tSUBvs
zC)4}P+EeI>iT_{-@ty^zUuZ%dH%h43ERO!}Yz=*@fc~_&{(c&G5<0RU0beZm=W6)f
z8vchu;IoV6EMm6bQ<=b?ufhIS2-9)<2()KrKzor!d$C6Q;Rv*WzRw~N%M{YJC+;Jl
zUMQiSlL6{V4Rw`<`e+E$g~U^Xsr~L8YGp2c1TXaUyL@e~yY-g{(O$0!>6L_x3PNHo
zTT8tmg?m5~aiu2WBPoc;BpI7E%#wupri?I)8s?CO`OyfN?jmBiFU0-sdBo83>0=ac
z`@ckRb&H1kRtfhrQamo8I>Khuka1VlkS8_de-436YQw$AhI}!#)Jq>K?3`NPEv7G*
z&_}57EN%TIg3OyVnV*$pJ|)S#gz891=BqTBw`wv!5k=-QR2=9TCUZG87fR*|lNf!y
zm2~;q|B;wi?JJZ_Uk1#c#{4>s`BPDtFAHK0XC|&>F{;)33VqTqQ0ZZ~8|{QeeGLxA
z`!%iyG_KG1xYUw>TY8yn=^Cr0*5I$fHr}e+cu=?TSVSAw$~JDx*oGgO^PRel&-rcW
z<-$YwHK-XMrf)34&ZKJIPYp7IZkOP(q={agig7D8XgiTxh+>6$E3@YmYPnoMCie$v
zk@7bBcry$g{qEcG=$#>!e=6B)*o*O*ggEH}FGu)^)2i^3ynGmz@AV^ex$Yf=<a<Ek
zz7tRV?z`ykhwy=bJyx)Gu9Sndt4c#M34JW7Qa=F<$8hILKTpF{RO-H$+P>I*AN~`Y
z?{zb6bvOoHZ(c5*>7I&;nZ8UVUBFcqvKt`)AnY|xWrPF%;mg_M&9O0<$|;gQasbSw
zm+>EY-NiJqxSu7OJ`F&)7NHXAd}!T1Gv?g7oBYoA?{|A>+w}P)(yb8iO!~jlOX*3c
zg})j1JqFGflHh!ixXv=eq2U?M<88qCF<}xH^kU$AF$vC3sTzjELE{<Dm)d}1#E6f9
z^F#uiKp1((^5r(L{5lnDzLJ2&jxo=0zM375x?ZQ)5;A<hmINgjc>awC&4lW}Kat4k
ziR|MzaBbQB1~qSuv=Eg4mW0v@V$TfwiUrQ|ax3WY8irTc+M!JH<U-44CjJWt>CIvC
zo6W<M9!JG>*xC8GWde~wkQk#b4EE!=FvV5S@r@`+`8<OEdV1J2b;cQVL`RWD*u<N!
zG$$r^j8up7XPmJ+QR4+U^{sTA`gRhhtcd)~DUQf*=%m3f?>l2gPPzW&`@k!lY~<Bd
z1968j>_DQZ%kt3`jOA2H@J-xM5Z4ZGWbV*K!CBM?vwHIbFJHg1+y7R@v905$P5Ntk
z!GD$hqJOh+SW;Msf>$kDbLGmy!k&eGMWTc(v&bm3sI^Rw>@@Yx&A}px4KGU3@S>jV
z4V$<&GcRt=<$C?_zP&mAa_(H>bpf3;cP;|f(jP(yiJoP~SvQXB`8FRN8}li+>|O++
z;Y%dADF}I{x90LB$<?jnTsy_GU*-C09rs!hI2i64YVOtVi94hbZ{qMWa4VFO5`iLd
zs8!eGkiQex5R^Y30s71M*NGwLovwH&7#P1ycrx9P{Qr`Mq?Ln$hvdK6%Bv|!wb&$?
zhLINJk~lk-18`bdcA{cgJhOh5j#<A*VwRN+gPFzIFiicmnDm$U7s@ZCn8+uTj|+tx
z5lsV7bpF-o;Fu5BRWwj$Me8g2Iam#SL;up~EnO}_cTqW0Ptf1=&(dM^RVpx{x{Tgt
z&_0+@eMaR{g0YJU{YDi)$I*Y5<tKL_uNw|*X(Ev`kkxIZ&*;b(N9cU6D!Xwyr9L<}
zKb20lf1M=TRx%cr?Qh$PV7)y3r=w`PPETZO@e61~9sr$Ci$Q9iPDkxCNz_`|VVGLZ
z4r7dD?@1yeY_g;6Tg3eGCpj}65w~G|HeHwOg8)de1^_Ed4tL3a+5)&qC{}yIX18y-
z=jNeH!?*KOw3J0ZLytR?s7l^L?4LxumBELJ=L|lkJPs!luQHiUKNEyu(LFF70}f7N
zfR$y48NgYFOa{C*oB^w_nsjQDNChbMu%qlRiXAfRA}Q-yl@9vh2*)nDdyG$qq~p_J
zNqn*rCNZBlVUo$G&Eb6FQ9=_GTe9{T=|`j^{Y6QnTj`vbbWZ1Fl0FziI&Jr+pK3pk
zlzVhK%8yN=+_z?Po++lB^GunP|3hwL5#HP8t_t4w#k^amR3#<$cMP4)=r_&^$Wt0D
z5&q(30ON%6t*9KTR1{;$kttQ=Vh4L{xQrv)Y8?Pt>=B(<1bAm@kTQpg+|lO&m+Y%V
zBtlX7W_I`Fbi6+`iT74gE$01G*0eNu4h&OZKSk}xBg|1nB*Kvn^hS@+5-Ds!QDN3Y
ze+5|&lg$v^_RK{V^+a7X#2$%O<@UrlcdjU`6k+V1b{XHURGli<yVizXR7Y;IlE%QQ
zuOoeZT%{M)H@b1@$bki&wEEKg`HKR*duF=cJv*s)tsG_CyU*CY`zZJBWEuBI40aQ+
zs?PZ%J;sok7M2a{<Jhhc*44Oo)Zj}>+!dopsS8S__wCXJJ&S@801F=FA*O6Yyu-jX
z0ckPeEXQi#(KQuz;iTGufIG$7ea@ZB8In3C30xadYl;G`yU-BPD1lB{;v`9dW~7qG
ze>x&@OLV8e5te!c0*9`E1dcaHN6U^#!GtfxC8HR*oR>~6=O@Xfm5z?(at57Wni~IO
zxju-8&lusOY=~%-x+&pi01U(2HMEWMWp0_Vt>ZFE>~trS0ve%aL)DMpF`em|!z5H3
zy8&lF1)P_T)fXhO+RDPmtlrON^)<Qm$b$_NB_`B*5pgqi<oPlhU3#$?Zm^(J@_8aL
z9D%yw(OiN?qzq?L7y}%+I2}h8Cvn8e56B!j$mYms0!J<vC!VV)ceqpqfvJMo)REDf
zk~^2Sa>wBV;K3(Izx&Zn#P*<1>RCAr>S$x)Z~<B65PA?70V4H0(ah^?Qvv!1#iCk&
z;H?mLanZK6Y2zx(_01A)!qcdu?CX}W33V$szT6PjL*0mKD8gMWon@Rm&(cfZWwdG{
z&8=-&^oJqFYWvia?-LaPRy7}}41FmY--EUgV`QZjoSibX;QQ%pbQk*B{Q#csWS>;`
zpU85d&L`i~dd9viX`on%D|w(??hlkMH{>{RRZ^wC&%$L?lq=;_l7{LZ{H-mrMV01A
zPX|%TU*xh7-KIsN4;D^3PnmKeMky64$r{U~&d5j-GmtSWqpGkjlWuX1fn11G>(o;;
zs&+V2qd$~dS~4;(ihPaoMJt1zCfrK*NMMaBMDj*MusbzneBSTopv%fy%7vx%5SE@Y
zrJYJMsdMfv6`m|27i4c<9l~)8Ps~1?dbFD^NSoOz_2k5ab?4z+<flt46h}tnv7Pj#
z(Up??Q6N~+W}LZkGYtyM;uhi`Q%D+S<mxRvK(5ontVmDBFoV4>&NYbrV}|IRM2D2&
zBsiS5h#gW!G#n>89NX(6gfpW_P~JfMABHlfp^Q%plu{BDPB+6)Tn%MQb|~!5m)uF^
zBse<Br8}YFRN8=3rFRl#W-<xRky?bW*Klgt;iQn6dJ+_!;?iwsD9vesvNZ{c&T#2&
z(@?fg3zQv6P;_!j_XZ8+#%Y0aQxX)N&(ghFLwRL(DDExvPNL|(DhZBGW9i<i;oQ~+
zoIgti=k_ExI)A16Y7OU(HsFvyyv3c6CE#n4;OOL)?oJKoIDcK4+HLimbc+D7iS#pC
zx*oM^&9NbL>`J0TXS{Up)^waS?R4l=nC`vl=y+oi9XeH}d!MG`G>eYGBgpS#Z(2)S
zZ>%GBy0~Y1ok;WEg)lf1e6N-bmALX=a}el%MC5zN=5vlXLLje3c8Y0W7nAq%q6#sF
zB^sH4yG51tDuwfrA%Jl7Fo|ZdO~z}nOHRaY;Aj_hm6tvzYK7ApFO0DbM;pAPA0L(J
z=tYTXGG1JpZiu$Gk`7ph9!xhF-kvlVbfQl8FZ5uTWetXeg)?27^fD<YdD{F?IvV~m
zi3Xk8(|uUez^OaaFaF!6;l1f-cqEAioek7|zowzf>Id&vxfgz3&*1L3Gj`=n!sn+H
zbw8M{^FEZ+c{=5&`}ex@x~<MjGl(v4OMUdhSX5Yc<-oeN>-tu%DUknjKvn!xIx0Vw
zM5WGD>i)B)azPtZx}Bmj!d?L@h%zq)`0%-OeE8QSKIqJ*?iVy4I9X{%b(5M89MNEG
zvRM!z4-1smsB{B|nUE)3h!ayTPawHbrH)9PWX>*^5p{#&CGwGLfDd0w$A@nu@j<6W
zb-$_k(3@Qzq#Kr;6aH^HaQ~hJPNz(DzpcS7$qvqAlNd4AwuOaW_xi7N=$}kNud}DR
zKhWqoDXJl2gEy1acyH>{j-@`Gj+yB=vZ%I-PhdR$D4o#!BuQv=7FPGCT4<JABJa5>
z;#byes}^@UGI0lo{x8z;;i)7(=oGK+uQVU}Y(8A2_&^6`Dhv@j=g=2CLru>fAxQJn
z>1ci?iDsQZ*8Q!fd6iA`T9algOYHP_?;k=)4KZjxn?$NkP3z9s1j7cOsO>{fhHeZj
zwsyMqc_Ewn_oa6Nwl8sj?A@A^G1EEPE>2S6K?2X6tsEm34g*MnI6~XqZN(jQkUFg5
z+UJ)oUxDcY8{k~20<XFf3Jz0TLhD~gPCJ(qn_^Bo4<_p3ZfIMnXqk;`e2ED??L0K8
zQvbB`VmgiGN?+<%n%dMA>!N-Xp|I>(A%a9{USm1n3lb-=Ww6K6OBn^pNLVX7<b@<i
zIifg5kWNfuqm>qy*?5w{#*nU|bg?PBhI8O@ou?Ufz8A;-X~wvb_*RlwWEp&6jPa9+
zV%0SdgJdgez^o)TTIqF}jc10lG0L@qRi%+fHwMmGNpP$*yA0>-HsBcfbYtM0lLUv;
z=<;zSKQWxHHsH*mcLEcM8`4?lCc){Z0t|<r7|y)xaP0HH5Kox-Nl+GODE!1w9`x0C
zmm4ysFG#BNjKzw+7gJHL^g?=yQnz=8&f9`M_ab^z^?nGPi|O~c`LCA>FuNDi6T?|T
zPi?_jN^dGSA%`oM(C=|@E~Nqt=Oy&SaQc0g3#~PgBbQY(K(X}lNs;Q*T7+a=zqhWT
z$YL5?%dQ~c*1R0T3XZ=T$BKT<iglV5>upwS$mxUt@Q(}0syY4iZb(NjhnkI}_exFg
zRhr%nHoYsc)3;M{PCtn+Pe&pL(v2hWT210+O(LiDGzOdZGLSf0M22nL&<)G$IsN2b
zmyX;aDi%lXuqJmzlY5ojqnD}O7+vJj=|QKT$gy-pmXe5cHIZ91k(^xA?9;WnPo?Lr
zxak8|+xGWGBDPZ+jZ{fT>SPkB*K1O1n$&Az`%@<=o&Nr8rX!LATgM6Qc1`3CP2{yU
zk?WMuB5ekxR5H>YQ4$#IU)r5+`X0S09kH*ZVsXUk6sqnmn%Gy^#P;W6vZGq0PQ!e9
zTRK{APonkJRDhQjcW7F#vuRzWj2;-EI<e~X(|Bh(8t+P?(bF`(PSZHlN~6qnKpI!&
zXPtf;U!RV~dy;7MjVt$R8b@pzYsS@wwi3NgKR@nE$Bze+`0?jdfaUs4njfP!KLWF?
z&K5iUM7}j0k(^K<PEoyG6ZsdK$Z?y<)y86&@>!^yW7ALOyV8-#DHGzz{7X&d!<tOj
zCNt2NI&bat6Zu!^h~$(IaYXt7@E*}bmTe+esu|YEfIIyp{&hMMIpIYdi65kTcufAC
zCb43ZxJr>odRQgWoqifWoQ}pvsaPD1|EOvFh^BGUrZFI^Iv?-!6Zx@pM1DMp$WLe@
zKdFhV*+kx`Bm0?onve!qhF3xsx<Hcg=%l{W&zR4oW6WbojQOl)%;z*?I5n`j?FmS@
z$_+gIG=3o+jhvz;PQt&WX?#M{xYh3b=V-<F)pRs|or=ZL_zg|tH#Lo%k~r9pA?cH+
zzaPJqj>c~%(fA!r<99WUH?;PnTos#Xji;Z+@1>*h`$;rDscHOyrtv15#-f^WQf5S+
ze)9icI`V&%ME;L8`9IO*ztSdOt0PXLCDSTTKaD?6N8>M(Xnac3_$y80t85xyZmfY3
zvBNS>_&J%UpX{g8k^M{(*}v6f|4x&An@#o=xs4GCou{ABKc*v;)1Aes;(e}Fi^hHF
ziO)W6w+S82Wz7dY{S4Tj;3l7V97x6D7;q33;8pX%^u!FfV+o$jxkzseaNU2;ZSW2y
zK*=O>BD**ekI*C@sY%>vkvKR?9*ukRP?HB*wmj69sCVA(dUQH5IF^dViNSGHfF=ES
zdSXd`?Gikhl71bAdWL+~7V*<yi9Lz1CJzG6;TT8#shaxJH1(cM{YIibHZ}J26MRNG
zg3qL4aRi@51-QG<rYG+1T}$v}cK0^5I3u~|6^h7Tnk~=v^fRT4NJy3f&fgiwlyfyx
zx;0Z?Z!={QQoy;lv!=V|muzNC!<d+#jvEW9SR6MNQ2`c_^XQ31<envXGDV~UZXh;)
zd%3@DW^`UVzL=0FO9tn2jpN0|nisvA7rSj<d>OoWekT&Q7~Xk)FHOhqOQ~2Kzh6QH
zSoW6D6U*LxOYmgM-lw3RI7>Tak*$gB>d$pD@@eOAAMqnuqB$>d9EUH{9KKw0m@}E1
z9R5d~Sv^lunNK@ouSmz(wNxyQvHhB{>ojBEv?M6HgD>M7LVFXU$0S6zI(!-?$_>;%
z$wP?}d&l<pl~jP0)m8Mw%IYmk@MI1p7n6g?l%6rK`t<YR<%BxO2k9q&p51urUf3BI
z5N;uvCzMa-yIKRnZ6ZK%V~jJUGspRfIsSH=<E5O+rcKGk)8`wvX-+#!Mv=@n*uU>=
z=V-!6N2ene1_Or4K!`@jFn#%{H-BXX{-m1_3bs-wKP`%fh!7Y9wPLB_j10(x=7>m`
zRAN=J-q_rY>_C+Cy--1HMF0vUf=3;)x?t+}?mVK373!1DaH&`>a2dQxm*|Xi?;=5`
zPSWnyS|!U3>F-8Te>bTB_cuRrf4|%A?^o#lMkEQH?p+1Nq|k~;6)11#+I>7_x#mv(
zf6wDiHhF^U$OxfbY~TvA%(6%qm+RQv`h3%6dR7?U4#*e9HA(lcs1H7Y`w#AI)I&iX
z64ukVQ?a;_{c0+}J<U(t)9<BL3|%Vkm&A>?p#Sr0(t*1(2^{CDXK?();2xPOI4>Qz
zT}j~XrUDF(pBUT+rV8#2A>cx;P~4jY?r{p(#Nhae!R@A(TDvtoLFi7G?z^P8Lh;5V
zD8Cy}Q22?V{Ob}8CG>jT14)(o9$+&mEiG62H-4pjNnr4juwbHn&M+cH*49$ENF%0b
zFGs!=*Ii{0N5qNlOzcouccWQx3NTqi4~*^T&EroBPwRgn44vIDcv7a-a<x{>!vnE6
zQk;Y%Q%+W*13Dm?n&V7jen4~}=n+KsX`*Hds7`*1BZw+pD-AUp6k?ZVExL<o|4Td3
zhD*quF6-cNKZXUSGLyumvjs&+469KGf^RMjkF~i}#@LLDmiaIVh^78QZdkl0Nsdk?
z>EUs5ik71f+H&+6k|Q<gb3EbECK5x?x&r5G<m?+35MTbx0aK;cmtX1px_+2sbcKw!
zx`*NpQ@mD^2P>GqzZOwg85WfuFs_G>DB`<04`sF4K%Rgd`La_SA(@5%lF*e|zQqbg
zS0h$3`7R+FWn3H{hk3Jz%p1!cq!OODZgk#8Bp!2cVVLS!Ss5|EvZ@JsT2D=$cM5r4
zC<zZ+e=QYp@<RpFJ#8^WiRb1IrW4K&B?;%bS~$D4aDK=Z&R24HHW|bnT&r>q$PB3}
z%Dw7VzH1OBtfk6ivmpZKLvaw}gYomZ^RcRjBklD~>YL|@yh8E_xE2d!p(Vu!8DxsO
zrIm{8>52ywOpV_02j48Xwv?;g$e&prIb+tmOY)l*E?C%|zwm;KktzX?3%c`*7F=`@
zJumW;>7l71%4r*H=|--+P#URs<KP{F42@ivDAu=hbNdC#)kDADtRZg-Hs3A&JY&{o
z0J(9tu5LDF3=V43wP{nHP$3+;Kqw(fjP%>0&D}@>IH!Ad5Q13+UibgMxqH)s?giZo
z>3^Fy&q10%|6fN2)jTFcZL82j-T5_`p*ZF;wnyOu>P$VLKALVo{d3ZQx=0VGi}ir|
zusxuzGzZiOtrkd}rZAvrB(X~5?(@Y?E7a%vpwfe%xq%ghPLaVf-7=oSM+S{S!3w05
zgevg|K-_Taj{A{@NciwkgIPC1iUIwIJVq>v0mTpi6EpPDT5zoTqhzWHTjb<(oB$Uw
zAkNVP;<0oC;&Vv@Vx=AstMq_))E*EYFbBk#Q*q=v(;5;wQcT?ZqWgnnO3fu?)fmYa
zDfuSqoS1Bgls_n8w1z1OCAbkOa1?Tcgi1C{YZUr-^|(_J!ACe07z&RfuNS5)EKkQv
zbtASWMKIyHg6h_?DZVqN2ctsiix5|(P||vOtl_RIR*3}!EsG1Ck2XLGmQ7V0tneF5
z+GXpn#NYL5vxe1BaTNOilboR)Dj`Omm(03R6>2GEizuGIYR#Ip%yxse$oey8qNZM<
zp<=By?YWyoLUuUBK?{c{#ucdX){!}B1&=k@)Qm2&Lz^i^+TVwUJo{?8LHYHhK{=oY
z<&}C+e#9P>_t2h1<?b<t$&?2rvRo;9B*aJeLPv%ukN!QD0R8zXcE>BR3tjZc*ni(i
zCrba3BuYgsN<&(dK5C27yqSh5u^!Mmz+U#G((#BnEX?QDI#28V^{nK(>T{5YEKUJS
zapBZ_n7cWe<3un)@urPr<;RcfDREI9Ysr?xCKPinA1o#>Wpw8a4M!O|pg>!+6t7c>
zE34H!vcak2YW~%w#`tR5NccMf={@dGC2_A|=gj#go#ODg<_uK-1ZCZZL8k$ujmVnB
z0;6U|I(nx`Z@5~+#4uS!#xvS*^ESv|TIvdU?z+z6V9$u7#<)zkbsf!(*l%++`EAu^
zd4%#_`jp5?v0GQH5!!WEWEw{#iNv(p3RTST?Lp;P4p_DCfm%<h{~~o^UBuC*!QUm9
zr`U_ZsD`;xuI0r<>MPA@up+!djPi()0J)c9D`nb9Lqp=)6BJh?fy`Vn^~&Rb3-p5U
zztbs<A0{b`s#X}2T4DUNtuV%T?!hGqvf2njgG&b*M`5eLOeQf}4#*e80+HlNYbL$c
zd2a1xVQUyfdMC~(J3}jL)G{XZAy~KpR_+-dkWA8NOY-k}k9_Fq6NXAyv?eFZI|{=H
z<EKxX^vXt<wwy|F2w8It^D)x%$@Q}?D($oB7Dw(ST1fvVosj+}Nl0I%h4fY}q@S>a
zG!PTbj-MnsxAnQ<oE*G;nlCdpr!r>mOU2^O@;uF#*J-|dstvv@O3#<jh44Y?=*cC~
zbHAqN0Zq?mveOehn>;ifxWkjcy;XyIP=kAHs^E@H2kxjOaDHO#cWQ8-v%n3y9oRx3
zKGEE{dQreC8)+Sf8&xFS!(NsWXUjc1FI<%4ggWe?-wSk@Qs9*(?^jjsG=GU;Y>v4?
zK7CP3QJUzL6h^#%g|h>;U04`uV@+CK)o~m5`M4uJ7u=+!L4YS|PQy-V6&vADH26d6
zV>1OP<x`9}0u7(Dna?uer|#&re0r++nXflt2iN8qa&zhC{OYt++ghr_2Hu@_8pC?h
z4xAHmuVH~nHd1l+smvNnX9<JLEMFEnn&VR+OByoVl5vp_RVf{+khDF#-q34>saqI7
zOMWBwU5gLt3MOUb6vVF^Aj=s3U(Ab8zrz$Zf5euRZQ~_6@1SDaYFJFb9967Uo3te{
zJf-{rpHth6nyA*n$XngWQ;fd41COC%Igezs<JA%TEhVUc^+u^tDEMwsShk1;kOPM!
z5YwtN$My)`VaU^ME;S1Gta<WaowgoAN$}C7s#vy|ZxpvUAQ{^gi9}0tPGRvy-#XbA
zkhJvBR*o1$z!(=apH6Cv)gT|hgPS?7P>`cEp{R;zhMO0X6!4Boo!A3ggVLUp^pS2=
z>{K+MCscd-Cn8(y7|F=ajTwhv;+NHZ=t3}69}nf|go<{X2=F{ytkB+X6o5?wtJuyd
zW?M^f79Sxw_2wjHcCW()VU!n+NG@-B1;`D$-{ZcQ0G=uiyai^H?*sk|h`HI){{eqW
zCgZ4X8-wv1FuPr`OdWkn5xBg%jrv*ab43F=Vu7w-IbxI`71WWGvuOt|l$Ky$;7z$R
zf?lAkGZ-3F4VW(|#*S7+#Kx|8D}3H)3y-ZR4rM7zG)XCm18vl)NfWIxZiIA*j*)$!
zg?wwVR;o7Z`TjNQA-L34N=D1Y%2=~FCPf`h02=i>L(}=Hx|vp-+}7Dc1YV8X5S6<!
z28yO?OZ_`&mCd`|bK!7=|GXLVO~2AFZV$c;M(VwJch~r1?lUMg{z>l~=@!9VRPRB8
zD?Yz=Y6Wr7R1>R$-NHd*TLGHub)>`iX&0UbpN8q`1*W=Qh?;A!;)2K4Ud8piQe3;G
zD`S^XDZ5yuxQ~&3V1F)tVs-H=+W64o2l|P$hWmN?5RQ=Y?Z@%W_$OW8tME%ygx`Lg
z-X>4C&AYqYuj1>je)kKsb$^^bzJ@OvyKmjCe)%GO`6c@JX5bh1TT~%c)8FrxcfUj5
zeit9&TZi!l_N&1BDg1t?a&~)!|9A-$>&~X2{7HNR?2L&i{{g;W?A}G`6s$53vB8L!
z`hrma!a>jfn<^Axy(61v!nye$>gHdZv3Z$@BiQ(lQ#AflD#MNcOgDZ>dyN;Q?IQ@{
z7s((ntucb9G=lRoA|ThGV6(qYZq{A(QbqgIy5+?gTb4A#7~L2TAW;6lt=;}O8iUmN
zf!uzlT6e$XkC=AkS=KxQ*Hi>Y_n{{qi(j^Cc4yL0A4kjL+$?#p$jT4Yl^>)l|EgcP
z9_ZiTk&F*>&}JCQ?vZT*JVpaNRs;Nc1i*ks$pY@w7I3HIn_`l}kvyEGA)cure$$62
zRPEU<O(bYDS@|4Y`CMK3|FtXcZnN?Qy7Gm(@_+X$m%X1wWxE%)31zW{vP47qjt@mJ
z@e)<J`(GrF<YSsz(wFN7R_F%)!){=uY9Qk<TCFR8sjmEccI9i@tbDz${AIfG@7tAc
zY_syKb>)M)@*mihU!y8_f1H``z6G|R8~AU(0j*(vMm7IDsA2luU(mO|q>o{|O$!FG
zG;l{1q>Me~(tA8VZP9Rk7y{?l)aL)#aGs`be?uSDAe>s8aJFhV+ccaXhrszAwfTGc
z*dZHM3#~uUmw%*>o2*|L?6dUCE%F!1yW6O#HYSTZ=so7$Yv_q*qM!O;6{3CVtC{%V
zxuPhB?p<xRIq?$J=H0r@pSQMo0DW~JeJC!-cJFPo-TUc1ru6~c?k^+S&4sl4mh|m9
z?%PoKPWL`qIDC*+G3KgvmsrogijrqB9)4op|AyL?>-pR8yCnI~sA&AE{tha_b^MP{
zQyc2Oy%oBjW$SK*!|O}AQd>!m{Z7ra-1ig;^>VdQFBAexHunuwxxZT9>3(_RJ~INF
zq_5%&!PI*d5O+5{nVY2h=!-X_Z?5>T`xg4oN7UW1+;zlr>b#vz*=M3JcZ$Bednw)w
z=}Rj;7WISIlal^{CLc`iRuDf25@f%Bo~RN19@r51?B>i_++5lXk!&uC#YQr%M?gg^
zi@!$jSmwS?PbQ1MNneDr_~=#^OO6W`KSpo1ve-LuMYG<hPAqFSs#o!`%c70W^=s*{
zMjFvhM8yKl1=LFdO_w*<utU~+mX}MDi&j-@E6BTuufUZr({8`Jy;;FN8zT2sVP##8
zN70&4xn6%~^9Ndcez{iEyLN9@B>Ct`ajvjiQ2*PO@!vmwg8O@XYpnG9y^~54u&rvd
zBpqxOIIS@DN&Vh_<TJM0-_G}Y$6OC%FK&zTP4C!}Co99PU|sSHhr;Y#P`<#s``s&G
z0`M0G-s1!7`qt%}li13MGlY@qa6PxIHdgQDgDY63aYax(pUH;|Lk=0zVOcJg;PMeT
zyr{y;y$x9#MNQndbW$Sf;Oe6JBAN)}gZL$e?p_)G3XsjTu*qv(VFK}v9%kt;h!aQC
zc<jxKmW<GsSX&GP1wG0>-IKBIT~Ha)v=Y6)mLwgz4Q_@7$xShay>p`53zlXg)=rs-
zV5J$*yZ5N8zg|}~=Ab}`@#(FbGHScpEV#F7+-FDQ9xsiII2Fz;E<vxEGUz~^sSrO-
z!t_a3XT9~#jYf=2H3ey7Q-gyira$<~uziD#cZxae3Uo!N;Gca;(S30=xcIjj=d%V*
zVTtvQ@S70UD`5-~OM{HdG{H*mc%{3-Nmv+7Lgb5#ejl1}g&|pO8P;Mpo;u*uf}$S5
z`ir7*C%w<`XZrB<=GDiqC|W=i>A6XM>m7-MOb7#A?I{ey-`Q!IzuN1nW2MlHI>kn_
zhRgNvpB`>VS(_hCaKvj&=l(vOil~=C9ZbEOj;-N-a~+4mK~C^O+=~Gl@fw3Y&<CqJ
z#mh-X<C5ua(=^!c9VLIE_<;ox8%dUs%e8M@IRX=~nDs^3d<k9xIJ0&phMW;T!KTyf
znu89d6#+RLj5dk)Ib*Hi10Wad+VFVKnrdZifUiH!>1A&WR9nS0K)fyr3kRQ8e5!&~
zI6!X%YVZ+R75R$@a@e6ITE^9c)B`VKM=S|RNUi9r7!4cp3is2&em|)(u^$`TO|kMO
zH<G)YXwl#}@xMgQ2mWe~-W80bC+%tH^E2#qaXmq<2dBUn@?{8<;TXPzyuy4rq=hf;
zt1O-9Ro9@sXs*ODg_<>lZ2FkuK6@3EJJ~6`+0kl;=3nKfQ<EE!!we3{3BL^!)zx~z
zaz=FX7`fY!tz9;Jgx|1I9qx<hyQA>Yy0vkSrr(YU|LwCE<F{kOe>;xeJ3jok6X>@S
z!+$%8emgn*x3j3dQ^J2cmEJoo{I}ETw|w|-XV7o6!hbuHemg7tx3lTDbLh93n-**~
zcZ2S1`l*XPf}1rt*|~G*r!3Zd?+hr1VW(cl9#rhWSe&r6Elgm0J2u%kUrv6Dqs1Ua
z6+AZQ5<2VPn*<~{ryOf79>jH!;KbvS3<O?tiEF`cNWOyKJoXZN<L$GMc`9b8_Ld0;
zsavIzIWcAeza;aZ8=!0+@IdGxw8qX~44wJ-He~4V5LgiY8;j3E`Yl|17U3r?K59Y2
z<L$h#3fZH;J3^6Bkk^F-V+9WfH^6C%{Mc%x5%{`Yr^;T3nY_Xgk7z!j+h1fBz$4A|
zG_wdh0AJIFB|uq*)Ro;T#TGG?acR3#mWzyh*L7kUV9esz(fnT@FHSo7E)Ju?8!6n=
zYEIAvl}g1a<JQ)=QyLp@P>`9C#`v5$!h%54`f9p)2j4tcXDHAd!Ig4NZhBOL1(70c
z=+^y3NL1tMwrc?y$HvegB2%dAgmaC?2fz$fR5)NVQ{===3nXHf<STri1t9Hg$>Bgc
zk7_Oa0*MvHtyQe9;G0XPDE2v2aY*tA4mF9Wg7hUT4loHpt{f3QC8$tFobBLj*HE<z
zXU9k`;C*Miyx>G}Yq@67&UJ}4<VAWc!wP)q4Yoquo-DzsQmC+cZ(=K><>HuZL|#P#
zgW;)4Tow+kA`nZR{Ab~oxsaEvkWsornODt8X@Qgm7DB@auZ2DWt#pYH%hP3cBmk0B
zArK<q3rlK#t`JrEc8^?Ry64k)O{deP<BqmdduK)~ZNt`2iphKFluZprjdI;VzCPZ*
zU2}G)JU#GcE?d4rnm!IJ4h{G4pN4_E^XA`BWBE_8VvIYF_*?1{(^&pdwwZXx2y1C!
zSXd2(byG|c?r%`1uoPcNZwIHr@e_g*)dn0kni@8Cmg3{C|FDm{B%~90p<FH(w-&YW
zM44*%6l|ZO!gX;gF8BWZgZ^dx-VxYr4{w>Q!W-GNkEEDQ8Rz#RaGqSMqq^ZOWE{s#
zLKi#=xKv1iE9CauJwzj29D?)@s5rnx?-1xJRaanFn#|4pudq7!u}j$~W(7gyfio}V
zz$sV9#vCltjj_=$&T5v3<FF-Slt?GMUE;X%_^W}0JBpup4e>G}Ummo$CsKiv@F9A@
zJ(>PyslJq}U7OoC;SBRMdV?!o8p2m!6xCW`fP~YXg}2P*N)Fe;`rWgwZ;zt$54X$u
z_I|EGEn!u?^H8eA(&H}RZ`EP6DtjJ(dw!taOzt&{@v30cRf0{k_ena*4|+!uWujv)
zrK(sIR-_b#%U6Y+zK-Bx)3D{Qw@kyV3n8f+s`O5YNSGH~Kl0!Tg~+`$q(?p5XY^ap
zUaXMJ@e}Sz>RzD$t(~)vVBAuRal0W1VIE&Wuzn=EbsFrl5U^*E+r@`SUVLH8)X`UK
zz?VwETLfVD@2Tp@i8Tn<kMp%j!(9;pH^9X#ydBn1D-!BX3H8woQ2hu@V;btp5U2}@
zsqJ#Ls-fN`p+0VjL916@kfZ)p+`5K&nGaK`Nz!C%pO9^QEn^$L%hYz=#%jL}ZKZgK
zbeF-3e)nP4W<d>=N$hUZ4Em-7|MQICf1?aQ?j0KZ6+U?34YX4?^OS7n_Zgefb|%-;
z&8)SX*`=F#RyH$p-<0Z7DkpuM<ldv3S!Xx1TQ_sSzIyUXFkUsKrwIN_o^RgT`Qz0I
zXMU}T4PR|OZs~5Bzh*7AiQCkF>b2qdLQl`<{bzlC<T32zj@|Bqq#WNyA8*Et-0!{}
zkKUa06*OZn*z?6Bp5FP1qYm?vbUSk21E-kag(vq0y0-KVLi9Z#a^H!ke)nDU_e1!=
z(cNS8%$4#JSISowRHY$tZEsGhRQp7@e3OAI<(mwmQjU*|h1L7ec6e|J8Qpg?xrWPl
z$a1<Y!lyvp7{GpC+^9wz8L}uNq7GJtPGzJwPZxTZ=4X#L#}L(Zj4s0A3SV5zqt_5k
zp9UbD_p4k5A8O_2ok^ry7a^GR*V0SrNw;r3#K5^L366C!g5h{=!1*zWiE!(Rf%Cc~
zI6u|W##bd6&aO7#7&jtf;M|=6Cva(kvAn(wEXFm77%Xo{z+&H|U^w?=hogKt6qJzR
zdv6jH>skdv;cFG?Wtg_CfvuDKMrzd>U?Je{PXcG%%wTYQGs6JqdASv|^TZA7QI6u1
zr`s)Tmm!Tp+Mzj2=Z5(SilGY=RBl7D1VbYdWYpQHM_woSaSrH&D(E(@C`kul!GBY`
z!73D8kW+u2j#F<=;*@pGggM36Obkv9ewo5+uBhP%n*wh2iC8`P>X3`0Y!PX-h`ZD(
z6umZI8D+;xxO9xasT9ZKCSw`6CCsHavN@DYqO+(EX5rlM<?C1ak-TOh&RC;}_?UJ4
zv`H7H7yMW0FZwqNXUBzwD0tPfHCL`IEbLk6S0qZvGK-8di(1PdJaM3<n}bCX8(x&6
z;YB^!8#Zx?z-|bJ92&|Bye^=VICupy-lu1o(OoK%n=*`&@Z*L$%$A~y*gDLd*kfT4
zY4BK)IOooV=fo(J5|lrmDD;={uM<PgJDu1@JVYu3JjJgL-kEMlK9n>ht&3GWB>7^M
zF(kdFB-LV*v>O{EPmWS^08T5*PE;(5XV&}DG3)(F%(AX{F|+uJm%*&T#iYN)zfgWD
z#e}G6a9^eHJ8Za@Bf&q?36A-2UBxOWT3^x6!D<3JDM4>><)q&dbQhH~^@M7}(qZ&<
zDlnnCjNWEt>w^i^XH+gF7`vDsxDq;!>RFbb+<|B;6)}-W*|xfkPE{zp1v}l~tVMR?
zbV_}2ZvN|Zvi-M7vTfZkWZC8$hK6i=>(yBigxHTUDW}VTM&tp|3AGrc_V3bB`}axI
zT2~&KTE6mVQu`zk5jNS;mM<c<*7|B)UJ-x?xeBkPVkKpa6V1)*!#O!CJ{L375pny}
ze@fRSA4}>I>*^<W319s*yJQo~Rm+4;OzDwZZM}}_{1h!^(Z6=;emouVpG+d&x|Yhs
z^R-ly_~B&Y)ph;pXM&8({ONQI_?ILGSXXD60ep4VWWZ~~8L$egNvAf606tO=JId^$
z*db#jk+QB;%VNT17sfrtr_ZM2)8~`;WZeR0KJhJJlTVw&`NX4yCMY&w?J?57kdE{(
zCXsI4QD)Nlj<QMmU<~QB-J5=@{maqr6X_`bN)qM%Nk9PxK#|>zFHf74|3i-5WyuLd
zE<)J*CGXZL2U@Aj(Y(WEGy07W!&TrxfAK<|J<6k&DyE_`PPf_4wi42R!R3hDeOv4i
zomd2TXK9dZN$plUdSzB6B4?YIq}+c?$NO(3@!q;A&QixW#SPvMo&!@h*iWI+@)c}m
z<%-Bp45OEY;R}iivmW{@$a*+`4I!o=)|P&i@Msc#L!p(2>|`00QiQR4+GY8QvrP++
zT<=P^A+=!_ZBs1guvK42`uey^FRI7>w~kzX{0PJO`HKR*`=9B0_rH>Q*SdJmy~`Kx
z&E9>KdlzB#g^QIMUBHNhD=GpYd8Icfo*bOXV9}rug`sOlD#IRA+!dp6Z~>(w&S(){
zU`rSDEDA~heCQ~8Aj~3ShhdCu^7v8%PY1@62zV)O?LLvng0OZ>5;WzeDA2YE4H1o!
zO;eUQ#Z<EyspRpWj!4`R-6_ZiPCWvFLsvio$D5<0Wu&Pkzdw8_f`Z4$<&V<I<xi62
z(u%Rba>+3k47v0^mg|Ff_(L&NBZO&u>E_5?L)$oC=6q*ASw@MS<~~g-Bh;);dS&P~
zzh64kGlxkS2<8r)0Tu9zbgX_ViPctU31&5imM~d;jm|9tTYVAHNq6MglZ;$QjhS$I
zw!q$r#Bg}E<Bn1Z8qpBO07srq$B}1}IAR68V2*Im3zH+G2^_gx9CWUth@TM>(jgg1
z$(=v6az_UIeS-A+Y8`S`Y!CW^2M>JwLOm^Po17ij!R56`Jx?^tFwsZ=BWv@|hsEsb
z54;uFn&=EC+H!!KW!E<mK%7I*>f1X$JB7NH8($vII7mVwL^YqxGLD;P>7{Q+V@B^m
zTZo}u(yINa9Npx9=-x;Yp^ha(Tos;88YosY79J=ZjYSNU&{WR*k|%AP|AX(Uz~NcA
zQV!2zj6N755?vKyjW!1n3hV4M#JPY%o5X_E&e|@=aAB~A&`VKZce?K<*6$+Uj}PGK
zP7zYk2+|@uM2t=E)?8;pW*9QWj~x-D8ANf6ARUp!Mk^Eyv++oSjUin_AyHCx4aa-o
zI*&2xd@r~kkfVsMF@n2D5{v9E^TOzw7ZJrVU2{Sb8#zJ@yCd-vvyo%P#EkLNlPdMc
zc%F)Kr5u9A7&e3Z(6scY6~?69(HHBJbq}_v6w4(LbYBGR!K{j3y2VwO(6BW!0lNxS
zU`Di&KUo+0!b%vCXheF7z_RRo>f&Tw`{E>_xsVF5BwnP2=3;s>geEYc?KD3^*V_I#
zA-bhE3Cdy(Wr>EebXuTXk_1J^RB<oWP+l@EP?jY@(P2{D<r>P0sY40zN$N|2;ybLa
z)KECANqSjnb(p%0s<H$*1l;N*a4)3-EXY@Aa2(4d8#tH3Z5V#M?t1z=8Q%sf7PkW1
zNCg<*06npG=4d5T#5b4@-!)12Hfem<YJ41yBpW_&u@x+)8)+H*m|7~)jvcEcM8i4h
z2pdZxY+Mso(uD0#Zr0fdb16WHDHs#!@KuxWO=^7CYkVAX#K1RrFiqQkqJcxK`8)06
zId=L!P23ZvtI0@`B!c3Y5n0TRydhmj-jvjlH|vgkrS3?MNn&)Q_YQHOBCj#~@l~cT
z{>U`2m*a<+EWAA(3-3r`;cGMtcWM@L_!5(af00|WXM>gO2@bw49S84D;-G(b<@K6_
z9R9@MV8VK?b&kF#9k|^|;B*8P_l+7Hhe*i=Zm<{La60q=mgrXaBiPi%J&8#pK7x_`
zwIxo1pZk^X)Z@0i%8$1xrfe|QK+8KKkdj~ZB(Cii$n#AEg)+*offYKWA^%mTwO(Lt
zkrG_Fh|rU*W&Q9Be(f|Dv)_UNiLAkX36X#mmygLSjqdR$l~f`28t}J{E?jfqoQ|_2
z_@zcx$Q>dA`oLDr0TYa}#5sbApfE^E9TWXPTDLi=s@O;QfnnT#N+;eQOA_yYrUJag
z`?wZw4)Bsqygis~5qLw)UDifXNCL8UsjlES7iur60@W~^Fz*@SPnU*p2Y?R=mo5x!
zC7(z~{#TO7|EebcYnptH?P8EWc>7E&WcCI|35&|2qMhswj3UN)V#pHK2YUmfRQR|k
zut-U<hJNFW8Q4?cU!u`}@Kh8ub%~mBjYvb#KuTvB>e-%{{TU^wzf!%@p_xF>>}raW
ziMAtV{;&yi?pzhlNrismwXh$S#{Cf~K+9+lViKdEDzQcb|0Lw6TBZJso(z>5Sj|P4
z+kZ~)1nxDykWNBQBJ|4K{;t;Bs!!AJakdu@G{X}!Ke6qFL(a6aG3soO7RBCC!YJG=
zqIAh5UjAil_m2d`>h=)4Gl~g#IRFjg<tN6=p=dJVJ&td4cMGoqBR-9}bvuZ_BxBAc
zVdhXYjG3PpGY6%~i1|bn@<t>$HUrj}u#XCX{V)!ntc&5tB*C`A*D&m3>7`cK;Rc7D
zeAZ~&ze5gwk4u7L#kFB59M?ub2|f5dF{x7jP>5sNaHSm6#-MNT60$Ctz6vo3tb5r@
z!59JJVi|S?MWiEVj|LgTPTf?{X;j7cqF>1Yv?X!1tUHj(7L!f0!v1gY4Qn7pWH0Wi
z;^HMDrd9YdJms;atkyDzZBACA69)sI<4o2k;rAzl>o-vmIloa5xe@FjYNerOgWMsc
z+g)HPr9V<kr-`$u_YjlMO2f)@ywuOmt5M5yA`t**>Flw|CT{lAPOUgRi~!7_pjuPL
zR2f@~>sQ}#Jb6v{QsHAZS(<dp$llB#WD9Y(Fy@nVS-K%)8e5nogjQf67Q#jJk|BhF
zY0N%k5`)a;&_97`%ndmzI4=o~6%~l#a8w`z&fp>%69Y1XG`nMsHsO%$yUBz)V#=S5
zRObVfDA<21kAKhQEmYnjB81<Fa-|e;2?^E!9Fg?T24sIYOl=(O<bX=v36dG{w<{MH
zTSYIVua{D>xT)U?s>FTGL6ywDUQK<yUIcT@b9!V_IIz)D7+B@3s8~ZGI79nh?9G+K
z6se@38zl3Wq(N=;P!aL{xfBOs*@=<?K~cc}up=Tck1OuYUs-_~-o)NzYy-p7W*Blp
zRgxc@%3YaGHZDt&4J(2b%LYfVGGzl%k9^tSl-X01jS!(AE%UsW3f5x|WynHyb)2ZM
zPUJx$_I|-#Xh-DaIF^$rfzT1a+;!;$;$=w!VZ{k!f#5h{ra;_c3WOhj)>aoh8z-p>
z?LA4lL8*l_(QbLcQ<u?##V{owYVsvj4d?Oi`KDUbzIntaR%jr9j+yFeOU^RK3#LTK
zj=X31@^o^uDM@auz-=ry9JtMpn}qdE>!Hi$bl?g};QV`D9Py38al|(R+@R9|p&lvK
z%ykb$5#J!bN)`J8zF>sB3oM)*p+--qY7*qM4z$J+u~>N?RmK|MrTOY{2SSeE&vau6
z74o7iDDqRk5FIuv?m5vt7i_~V5l~m~*`e0;p>v}Y=TbU}*^(qNR+v4O7!I>%NQ~#o
zTl<I@v^`IolF0_gGUsupLOE0R5~nsH)=)Id1t$p}^_`e0Ft`je%tSgqRFnANpQUmf
zK;{F-0W|q=mEr^GUM<vW(WoTT)Ngq1Jk8gqqq&|$v+v_|4@DkinmH(;N%LBhW_eLS
z&pXpjYO|G8dFSFMq{VDIMo3&)eY+`Ye&%kI7cKN0y_Qlu`T-cyn>FMrKVa4{w+X_I
z;oq+i_KjJ-((CBm3T<LAly#qNAz=dv4c0vV?a@_)AwE+bC43*_c#;-rJeFc|5^T8x
z<A|$*{}=Pa<svqD!^LtL3){<Anb}|E4Sz1StyY{wynpn6wMn-!YsE^P!b{<{jz6pA
zd<DSOI`y9X>PCJXsSD~3Vt5zJh`omBFG!eBZw`;=@x98K37e2KHsEwuG{Cu^aBr}P
z%uoJhO!IE&D2zGtq<zd^`|(a*)hip!H;P-3YXDU$Ao@N~PC@ad-ZD@LL|Sl=Dq_4u
z=n+|tabYsn_lSJ=zZ};Woa{wSNTbpy;o2E839!~wB;krHU3De78!)t~A_4uNnp3>u
z_r;XcpbZLfPPi2qzpU;<KjQ~|=#-m48V9<O1?CPHD|E|zl<a@T0-BA3${<LqZ3Gi@
zYOoE^jdvCvAvg8rB;BP&jGodc7r=GiCOINVYh_;Kw($GgxAozvLh(9hlLZt11;pHJ
zX~o2!lE-CAASx`CssXdt70ZGXPX6-hHj+2=Blsv9$Po*4{dC8j%&8>^1@$Av{@Xz@
z)9AYS1l&CM&OIH}V_;A<V7{OjW|6I&)C5DPT0u~}5``+ly=B->=w2RSC262-GBs(U
zHAan)4$(2P544OUUuvn^tmpgJtcToES1A!y?hs<gp$R~verIT<(b6z0u5PCF`X=3+
zhMgB9b(@1g9%Jabsp^!!!x;l&r=E+%u=tONf2)qBdhliN{6XYOeg@Oh_$R$nT0)_V
z0XI?FUaH8H2=5<)w&K9^{3HY8i@CQekYD<jgDA=8E&Rmq6v?es=7#tcewWFce@#UL
zb;t<B{^7uHs0i0l@FNh*V9;5DJO>tHX>-cD{xRo1ln|S6cnjjV)uWRjM`e$mNfilG
z@gt}%Hhc8`s!{g<dXfwz-n$3U7o0l4PSE`k^i*6zj4O$cW<m^+FTtCI(93JqU0qnS
zZo{&|z_JaOtsE$<UDtQznw1+1YnQEGzq<c2DGQ%avfy1B%ZoG>N<pHSyXvTla!+^{
z%bJ^wQn}uP@k{|25N9AwohEU+goq!`Yb1RkRTA<UO~|>igxD_|9Wj_Yc*cxAhjfXa
zvV+AjHIT+3eMbgnp{vMV$44bSd7e?kw;)qoi&ayh(9xKj{TsM}$wAFJ;`o;WsutyP
zH=h%FERb3nS976VdFjMKSP7So@TNd=1Qe)BIu0Y{U4#M2;R(kX;a6EX7)Q6PShsPm
zCYs0P49fZri;Pj}@+g7nAP%xdVX><ZgBi#asX%kM*xq~!v|JJ>1t?6EU<pN?<?4wf
zvIsJ4)doSLNkvW<BbeU|3S~uRs0pAF{4G#sY>_uB6*hS<Ssw0TeIphTG@YxO7}>UQ
zSXnh-#V9sqD$p7N_ez>>6x~-70c|*5A0^3B9jP!is*C_kgQE;$_3E{PHl#9R15yNj
zs8x<gbq=vg(NhV4bd?N9bYvz>>?sPL&00D*W}qM69E7gGATs#G<cnj7k?OVztsj(`
z*|=>E1*HYM4!aVsQNYx~DEZcE(twPddziu^KRQ_CanJ`07YHkwhn`>~63KBfPw(sX
zca4&zA;9`beh7xEaS`1~S6;`d&QJ|!-E(MF2npAtsfJ`Fl%yveV{jlnObZq{5+L&Y
zt{hrja!FKVBG6t1>uu^<tap&7OFW_d;p)T$BFL0R<|uJN>&j|ICo3ZUWWPfvS$sfG
zgMiF6byk-&CQDl>Dl+%LeE*=5eYXlD=~aX%3jBI2Ae04mOHPe7RI!O)OQM(I4PAGj
z6j)z#BT5%gDa2LAQm$OyflV>+DN&%pT<Pq3UTQ*+O#)CWP83LsPw>$uBgiNOa0J)_
zTh`>H6cFSzb0)}iiG?qEnft-1RGVW^pR_6FX0htMD2uc)Fn;O!Cg$1F4XnOdvm5&h
zuAYig&dKwK(7y&PhU{95#@>z!3v|pTI0tW+-lV7k?ecE}#Kq$F(nNDY+3BR>;+G_O
zh#g)Lt7-TQk2rW=&1v*Le+5w&n-twIaBHnL%OlwE!ZhpGkOy1KC6zQVS46hrf)p%O
z=~`)A86Gb+U^AvUmP+*GWG=mQAoM5=08#1qGzzCmnYBcYzHrv!SK-ToqCzo%eaI_~
zufZ$&m7D6xmBwJLaSB7FV!begxISbKFsd9aiOhX{ed{4eb?it)wQJVq>6Q7k5zVXj
z=AjLa#*(i?_>0g_;6V3v*xr_DV^xLuK}`3^gBMuNKZ&IrHW){VPABD1K=M*jy<!FC
zox$wY>$0A(g1qX!B*OwNv=(EbDU)>aD%?9Ppp<sIj{x$Wf%UN@;t1Leiz5J&g3~Mt
zT1qke#aZCDR54}jaxF;RCK-RC1zRv)6kx<J&Vtx{d&*#sv{zda^63`j)?y5NNfzL(
zbxaxl!CH$*+$UJLjo%sY1z7+aFPlgYpUoZsNH1Tlly@LDAnh>I%l09X4>Uq0>XR(g
zff9`S!YsH0ubasCqZ?6K@gA=ADX&Z=<Wnukb|FUHn+3I9!&HI0)t6xNO~w(=mI}Rm
zl{qytNrmz9R3UmtwXRzwe{E$QR*-p|Nm+;0Vi4g{Izb`M`_bY#MnfMyL34FMEq5cN
z!Y-6uD(o5zwB7-;CvZzj4ZxXLMZ^H!pXE#9JSI$fr7$eaEVO~#mF-P<FTy$ryO$bP
zn+;sV18*hLyp6QA95Z!kCe*E!7<}a<*r}wzy<X*F^K2Dn>>MOtGpVq^>jt*=xO))E
z{Roy6lBvgAO!a?eUY?(Ym;Sp3%=9eu7#poZVz7yy+ruW_Z4cYQMBMcTI9rh#zcac_
ziqv?yJ$U<Jqf?h`JKE|}@e_lJHzJ3=XW;SnIbY70GqHDyQ7}2)&GD8-VzSqLW&uW)
zY29bOZQ|R1c4hSZluGc>k!lmU<_lbxM0v8sfnX8Fc~O=D5q#f3nqhW1fMl)!b<WfG
zu@*x0BV${X1)F-)K;`XoQOf@1my6}eaWz!`fA+ouPOhr@pMbQrAc81OhETEzo7p6w
zz$O7gHX$L&CME#_Axvg>W_RahcW0TB5Cj1c2sof(!~z2L-hLJ=SP_IDQboZ6*svlU
zQL+5b_uTu=eecejci+tJMt`3_`{6UZymIgNoO^CN_Y}K*gD*7SzQM19__jZs0u=KS
z(;<L+a0wMU(#=<}cd8EM{oXM-ii7r>3_1RkAxSYo?7;=6V}5=KqenPWGQ4Uan&DTP
z7tQeZL7+R#NRvM_b9qOO(;L&#_hhTA?7%=NKR`;eTn5Ez#0Oa%ihh-mAJ>fFMwb(y
zpotl0FATi(x-w0{Ar=LVUS=AOZ-$0OB?YNSz5v|=#r2cf#-IsFph>W$J@U=wr9HBk
zAb6z3O|)*R?WfgZdD2&FHHXpCX(|r08Y2D^CTCGI%@JQ-kgnv7vRkg0p&m?@52U*y
z9jYe9*4xy*#ps*T+td{i1f0AfUdm`k3)COTN<hQi&*HrG4ukd2Ooy=zDyJz`E?;H%
z96udjKmclOl0WP?2KSdSh>^;M5K?;=ZokA(o7#ojCA5N-m$XoDYko#8x7B0D*Yt8*
zeH%d5#NbEMj8Um;eD(?`{0d`iY6TSjK8Vxwq?d^*lwHspPdDDti+nYPt(4OeC+dB@
z(P~O5S6@Z|Z-d<y@t>?ZKGQ&X>m4hA?S8L^96BCjO}@t4RKKq)q7mHq<%Ty`K-#UZ
z&jf-9n`-^=_d$3&Pbn1!1}h2I>Msr)P-sE5H`$>#SWcVTWQRTv01b4f#9~_8BjON5
z!?x$+$Tt~&Q>)L&Vp=BSP7-Iz5&kL6;k@gxXMs#+v`8u%7+qvCN_4$GV<*LV=M!5#
zspRaJdTf8QkDj(^<zBX?j;*27k#8!Dmz}OLdVB@WdHH6LpjPk2OXw$PzN2iDJj)rL
z+vT9xg~d>DfU;zTv|v5TUg|K;R=?yhUKRCJmF=3gf!cG?!pyvrRxc;Jrv*zbtEUh2
z5jL(lUK<x(L_h{Kj<mutGYZG<p+b<cSBb6nVF8hjB|qZ(dIX0n1cyf<*w<|a_A$W{
zKNiK7RFCSl3e`~|R9CPWmv^ME+f!F8WtTw8v_;nZ)5Ny0n=Vy@&*lP`&D}nHHmN(8
zF*Q5^UvQBFMsT&MVUM?RHa*>K4w<P{wqCH+yN|@Ry~lew-SBp|@3xU|W{<aw$_u{B
z&gOgIgB8}3ccJV(@dT^H;5To=lUL%&Nuh6=F@5w-2Vrb0I7jie6mUw3w1w2YLY;hi
z#!G%ZO5#2cufSTgkXM__U_hEyp#nl((d!jR>lLK+s=3t+9<H7_&jzJAsXl&v7OGyY
zCNPeLAn0D0Rn=!3SO0c20argD)yGe1q3UTB4#N?(6pupCs}S@yM!@312s}X$zt#^S
zz#cIWoFRlkO{**$eF{fkV;m4`#ff3Takyg`3d%v54iONcJdB`-vKT>5A;>jGz}<~G
zai$=MSGI*XK^rtSvkL*xf{<60=Nni4L3Bf|ydsr1LB5<NM%6x`Y9DA^yBuqHo22%5
z)pm&VMu!)g`axCyVB`9$GW4D;^*14evKfsK0ga)cFccbN*en>%6%6tE>JSf<wmJfU
zcOHsk8T1qguQ7;^%=5+I0s+xPLyYG8E0O)+Lcy>>Vc5_Z!<&Q=7YT+YhC~_;Dr{Vr
z3&~<&Fc?x0h9V&7dGlg{u*HYq70kTpPmnx#znXduPn*i_pH`&I>#C~cy*wLvd!u`O
zm!{RY8A(@ug;f4lseD6&%J++|{C%qOFRIG-Z=&+6r1Gnz@?leXc*T(NYxfz#Gx~w3
z?!ELi=K4G@)4lhhFy_k*3feCd&eEDE9}rmY7g$9D%dejs<X|6FxIuoSD*k(O6@Nl1
z{+LvJW`l~SL|6Pzu6SAJq~>b=AF27XQggXM&C|j)dtdO`y=>MbRr~R(cJDAWPC`Na
zx*)hg5L69<uq>o~2!31e(xTJ9tK>j;j%5ops*9WC$<2c3+k$9QBSf?Ox<iN#x9{=-
z<OusNuX4Fthml{Zk>A=R@*fKFI|TXWM#zt}(OiRN+EMo1YjO8z|85xP?HcFLn#B3A
z;Cw)EzQN%1H{IyU%O28oJk?|!k4PO4N*(8#Iz+Ax1AWI3eb&ywdx&cH9;Tn3h!%QQ
zqk75q&Gom(1=Vi_)p-V$FGLJ#UJ#V__xdBgV!rIQz3OxAO(HN603H(r7X%Po0kWTv
zY7W!YbT(Oyaj<_zs=3gv<}XssEM3j(nykk7wEtbIxyY{OA5zWHx*FQanxF!4v!&j@
z885c&))ISo|JKC9HIpe16)R;Zn802dpx~|LZI7~ezPeFO8#e|P3$bo(FY-}%JK`$f
zP37$@h1zayvvbC+t*TU2Tj>9f&n8>C`JIEs;e00<hL7Y+otuiK4V~mCM*gY(|H`Gl
zPPI|#<n3U&Q`V#EaAb7zR_|I+bsc`LCT&BHcRgL!4sd-|jg9if&XT-m;46+=z*pK#
z+KY(C_<4Ax#f1*DJS@3T_OSE|jfh9~MOUFl55g{&TqwI-szNyl8a4f=i7C<XCNEV{
z)M69yeGMpXrXomcedpAOxP5S8jOBolX!grwu;0K(jlebtKO1ny40+=J-^lj%KcLuW
zV?>P6@3`2=j+`u=e8p_MDVmLq!p{aBH+?7(GJUfP3cF%56ux38-#1X&y$CV6-Bsuh
z@B$Zl2d=o#J58ZUWSWP%yYLMRF}%BR<<l7f=^i|f>$~^j1%~uPTrs5kaMeh1?TC{)
zh2)R$4MX}du3{kFkLPia9>5C>=_j~iNI%6@laPLfZy3_gaTNpUK|GIx^blTPNDt$R
zA^ieZO+xx5zF|ne!c`2UU*mZkq~G8LhV%%o7}9TX)g+|f;TwkZdtAjp`U9TFLHZ+J
zU`T($6+?OyS4~3tGrnO+kKrl?(&Km@2k8mCz>uEA6+?OoS4~2C8s9LaXK)n*=~+CF
zgCr*R-e2%4LwXKZO+xxBzF|mz!&MBVzvFovr04MhbLk(rVo3kQRg;k9=-&GmzK((P
zZ#<8Kv>gs}7}6wMF{JHr)g+`H@C|e6MYxKAv?HF!LD~r~Fr=Mv#gKNvRg;i*#WxJ8
z4OcOcUX15)kY0ip7}85|#gJZxt0p0dF_5<#zK((Pay*ZNWc0$j3#3=H0BH|=!(7@E
zS22)YiRW>U_QDIyrB~sKx%6sWwalfx@pTNOeegUE(!O|sA?=4NhO|GfT88u*d>sSn
z06dR_l)?)P=|EgDq{+By8PY-cItJ3gcpeAo5WK*U4#gEinu4n)A+_ThmZhn<ih(o@
z&*LB+h8Gx82d)^>bX+wFDUEL!QYWrrAkDz@I7l<`0z;aGD~5D9u9}2oM(-o=bqu7}
z;&~jT*Wm?*bR@1A(owi-64KH5hPl**s~AYL@jMRF9K67g=HiMW9fPYTAsvfv7}9aL
zih(o_&*LB+j~5uyd|WZ4Zd^49X#u`rNDFZl1L*`jkAt)bFEFIVxMD~r;;KnVOYjXt
zT8gU}NXzg%4pI+ZU`Wex#gJCus#!?B<+1mAd>sR6C7#DYT7?%F(rR2Wq?2&fBqY=0
zPR7?UkWRt#I7n;o0z*0#R}ASiTr~;FtOVBL>ljF<<9QsUb$Edxoq;Qcl)+V#kPMgB
z<LekmSv-$})QcAwQXj4uQVv&5LNZ*+<Lekm{dgV+X#g)Uq(NLUqyny*gyi8H?r|G%
z6$5Dq&*LBs;{}E^f-8np#8s1!M)3_pIulnhkV<$S2dRt~7*YjS45^B%CLx)I+la4Y
zAZ^0)I7nyV1%|X4R}ATFTr~;Fi1!=tbqu6)@H`HZm_d8z3Zyr-0O>q@!wq*nu3{iv
zfah_L#L(G$6JBL5U5KkDx%6gy!;mh*RScwy@jMQam^FKs;8ljS1y@Z%x)k3qq|0y>
z1L-Yz9tWw07g&}q#}z}m0#{8!GUEMKd>sSnZFn9B>Fs!dAzg_phV%|xwG3%1zV;zS
z90|M=&*LD8-LZF-Kw|6S<dEv4$G;mzg^vJq^xJldq}TRBC_O)b6vf<mA(S3sULlCn
z3!(JN$Z)K=fiNuy@}@#HQm{kfP#|WVSQ>IvtrtS+kpb)rq4Zt|rH8a1$wTQ)U>Lo&
zpfg@DD4RQko(|+?9CMoqpVwwNs*O|Vycfde4W`BzHZPD{0GCN)fj2soheeHdYw$=Y
zBH8QJuH~QQ(#0!McIm2L{xt$pzx;b&{qoO?FkQIL1<vvaV{q5wqk3@B<|@(YQ6r$d
z-v#9Zc!5{$e8o^cI3Z9z<bv{Hfx=e|<s%aU<)bbr9}_5i#ZW#zb|?|HC7*CXvAo6@
z3VV$yS&7VF`)OC9#$Jm3#<)=S8w&{Fn!l4FSzX>2c8TGrofJG5SsNmgHX663U^<in
z8SalqLR(%LPMhXC#BkCV2~CVJud9n#QBv4(l?j8<a5z|rQ%&p7KxkOMK;OrQOGMyX
zLaPEEqtv=|E-G#)7aW!FwgZul{k1;=1^RhkluwzTtENLIvk265y!mc2<*1$Pe`q;F
z3E}ZYa4Pl|2kGeHM_bF?Zko9(rCq5aTy8s3eoLtkuJmu8(V=|2R*2=)+*EXeUR4S`
z);Z=wPHGthpj8dWnT7&|c9Z<AqtDP3@fqyCh5i%X=0}-ueaF?!EZ;-!X5Ym}fo>LS
zmZ&rIvcn;iM?Invi7?-~$py#qMPxYNYaWiD>cH2|L_oRO1;z49WGJ^x2$b);pjZxy
z4CU75p(y_&XCvL_f@Aq9GMxWy0S<d6I^o>zf@3)?GMpbY4<~`l+~I;^xiB)6J0}Fn
zT`nlbZiW3B8Oq%g0_7eT6w9@dq1-zmP=4rwVtF_+l>3^8q8u8XqWdEk9Lv{{;rzG-
zIPBZ#gmb?Oj^*~qa2{v@j;QU;WDubr|A`BZ<pRlY*ab5A+=LBM?a*Y4fUWHHVf-Y$
z2LaHUV<YHz*hPosP04hyH|2!VAugBRuanX7h>H%(F_Y<F$IJj7Yj)cnX|wgxRxN9l
zM{qT;79H$QaZU(|rdOcg`lXJx6<LZ-m#52e+jRNU+88X@NSCq<VYlX$2CB$Xrjki0
zFBFv;g0h+EN9nfM^%fN+ftYx~WvxwDAwdV_1PZ}6naI|UREIM|*<K0(ww_k#srB<H
z`~YKHSS+o|Q}Qv|b<9U@K}k!tKE=!Tb|mdU{BV`cj=nABH)TsXg`bqqQ&o}^l|3g<
z`D{==#h_8#)X~NitPWLVneu8-UU`*#fwI}D>>BtimIY~@$5WWCGw}Fap3iXv7&J9i
zuobbL3jZZlrbMsPb=Kz7xemP?j$5b4l4UonUP-Htt#jr6DOqRuv#T>$?!ep`*c~{~
z85|4e<_9ND!;{Hqc-lpS<yp)$uxIf$py9b>H2lp)gXN>lG_a3yfQH&LZAfT{)KaN3
zd9rDVCYeLLRF#;ck799CZhJ|C$O$))kaB*tMCl&`2@=(luD0o^6~!`d<fy;V$TeCV
z;zy}=Bg|<lX5rK>)o-asb83&Fq3m!rl^^cSb5ar?OOaCNjL@lyc~IBy#--aeH#3Uk
zw)f9ujrVU?<5}+A+<5HX9ca8HooJEac>Fd$G<<YX*4IEdDisFGcz$|2l1pTQ`G04;
z7I)5Xxr{TF>@waYmEL$!*?|ct7_+8M>k%3<cTdKLJzRXS9MG8$?0`N|eAqZNG@K#r
zR7PcV${^XHPQBEX;*h@`DSoFyy^Ry+An7TkI4Dy#DbAY|;={hl_^`i=50-;F^MM`Q
zo8tp@jDWBnkPO^`E^wA-JcDD;_~yXXSj50wTNgHM+6G$mAX-!a{gh<rr@GKv9`uZ!
zJ?I1I*PIKjad+Mi(!$64+_gAuZ7DyL&(he7L;q5~T&$KTXBrK^J<C@w=t}9?U}$qk
z>g1(#DwQkd%Og{e>6sE=QL?#WKc#lV>Ty$+a>y0wITchWcW}{EWLtQGp$x=G^GiwJ
zu;}xvCLv6b*qZ90rX?V$VsANLqO5JC^-WV=OUjhX`FT}pX(TN3Xcj}hMNBS(#UV<f
z#=77BDks(o*PTwS%9E70tv#Vo8LalIG*#=`2P>7)a#v^P0N<s|d&8Y*#_5#L(W3-i
z{EtqR(yDXj(KC*k8eR@2KV6Gm5OTokdhrPf&GckK)9Df#E9e0W4F^342u;n?OXd`v
z4Rap7(Mp!W>eJLeTCMo$)M(vJ3xVM>CuK%fZwQX{Hy#srxa-eM#)rdQe6XT2FdsN7
zLy!+A`+PwE5K9;)m~L=4+7>jwHW|%Fx@fjSL@>=9A|gohvLMZzB9hWuZ!=PljvzHc
zp^X3zvC9zS)X!lmnAAD=C_rjxvDLt-p952blugRO@V~~XztO<L$M&DZnpB?)DBK~M
zZFlf3-j`H3g~Q0*f}$#@&=vl<f+O+wp?W9FqQU{gk+YmK{!khLQV&o_WPYM!kDi_<
zWE>xpuXHRB)i~i4;FdE#2OOynkrr83U98z)bdcMOkF2in5h9)<<XVpYTw#6S`6Ly(
zatC@=EP&4tfT(nVp5^GzkQOB4x5a}rr-<N|(1{0wVOXp{01fe2UjB@yKHWnTEK<1w
zgHWsh01cs7-u(<^VQZ{W$|jSTIUJ(Ka{q4#$nx!HK=omt6Y(ZGKc8oL{WnBnx%4xV
zd1GRYO6eDDCd=KwAr{M*pRshehDGLKovtqk$8z;=2*>i{A4?-er_1v!5C4WpEcbmz
z(&A`?eQr2MbXbCovkH6oH-xkuW(~GMnuM#dw3QlYOcK3dTkU|tF|~Rx!c{{!I|-bf
z1x|fP{X`v#MJLRZT?Iv3EQ*&1ikAwC6IvADgxTqoKYv5Mm_yIq1yPILyHP+_Zx58r
zZL}w@8e-WCPg!JMC0ObcHzsN$NU=&~7ZesbW%Krq1+uRIVc&d~-7x@BnUA7@u+x4+
zj-&+0fx?j%rvY$8C#j1@a!@Qw4iO}W3X=MiDT&%fC7la`Xh)GOAX8&O93~(-1VoD~
zh6Z`!ytF{+jD<2&pv)2|^?9BW@gqShIPZv9Fs~CZM+%tw{3eOO@X9WnrNBEHg>v(-
z4}L>wwH)lZ+qBp`3`1v;3rfnd0_8YdHH30Jp7KyKA6E=zaccuDL$R~K1u-oUObcT%
zEfP$N1yhS6!<ZxsToB9>0kbp~OpkzBE?`b<jX_OhT~m(t4O?xMfLbk}mbL~8!(CMN
zwji&}Qs!h-*ASNFSw9wG;?<R%jxC7EOi}E8-w@L}VHLa9Gp3cTv8sg>XWsf)hV=@#
zK3p-l7I&J~X<&HSWtf$Z#WNsy1_jUj)|h2wISV$Ohq8H?+z^XqSkR0Jnih94F?q;5
z%lE#aB$WkDMc|BKu@jSa%-e_x8UnH0=~-%89v_40%Yuz$`aJvEHw0rj(leMAH3Dk;
zVjQHxGJGC*!cud7EO#t7`myvW%jqhN=gorWqF6kZ5B*qLt6?rM?@|GCSuB_up7L~m
zIj$H?i@O+ix-XAC>l;eW+XV}|%`=u3mmbM;cX?Z*`E#ezMp}OJz~p_&3hNv?A&HFP
ztn*+6vvPUwMZLlNSXZdLNVlR2ZV|rX8XrWBLG9`7cqyc=)p(67%iVZ|zrMlXm0;s{
zP^@L|PW=3c-V3HFx(he&#?J$xC(WcOO3V?Ju}f=y852|fTOWI^J{?u`V!S(Bp@Y;)
zI!FI!)2?ve{5=Z9SkcpX6lKD>`aeqb^%+UxtJeoM;mRLPQ2Aqchbw<vDo;9>__XWF
zGvER7A`Ie5H;BPo6@sS)L49QLc=GG@$j`c~^%kD!)Bc>)oK(Z))2?Z*SE~D__VuHV
z?lFHyoy_^?{o1{+njYiFNjC-raXnngvo3kN<$lSNaK(M`>sHC$4tUf;;X6y=yGY^R
zFojD;zn^<D{RDEIHPe;7m$wLTPXYK!0r;&3099fGm2;peaQoMRJAm$Fz3)*|Uh3OZ
zHEjCH0`VY$_#Fe0o8CL5u8L%ZPnE)_N#Qryg?F@A_zWq0rWAg&DO|UHQ>;9)MJTfc
z${d06eFH@?@fg2w@7IvW?^5$5?ncK;1@onX+w2M!_!TI#)x?D_mcmbz!f&?=U)o~f
zE2Qw(OW}9eg|BL{@KdDlHB$IpcHyV_g?o=S&G$2;f{avfkEuXLn8#7_6X6l2$9ocY
zpTbWceI3Y+)bB%TvZXwMvq9keFaplAsPiv2oab=&ulOm3;gnj0vr*t|5;#AKfb%@+
z{0DwE>x%u<D*wdIf8hsPtKbrh!TuXh&eKnHUa2>ssum`Ti}4-LahKqVXQKNJSRc{$
z$n&y;pJ7KQdU;EWbq*ir*ZEed^C$In?u0sb#*fbhUGF<utoL2`j%j_j)cdmr^|nRS
zdrk6sL+kl3Vigly&+B}*4<dFZ<KZjtzE+=cGkRUMWWlMKRr8mxq?EB&(08<~<{SnO
z`{6iyIKi*?5dIyk`U<s``2H%tYQEyC>vPRUSG~GA0-Lj~{<h<PsVv<xx1(6K`3SAw
zjnU6tc!R4;x_;Nyr5hxuxd#QB^?PhenjiU<dq2iiaNG0%Zg8HPx|B5iAn!VA`z=+?
zO8rm8;#=`$1dFYB1l+U!4QbQW+F5Ml_iQF@?ZZZ5WuRq;<^RKJYnvIk^p?<a8E277
zdUzJf7uhgztr4n7(;v%8oB<@=mxTZ*U6wLN4)o>5+Y`+J%Qc(<BvsIa0EFBTLgR@O
zgXI&>7?Ns%Ll~gMk56CgXb#IAoRQ2;sE-aIk(di1PIv(~@%mkZZV}4|oG~OlGYn&h
zPGRd{f#vtzIt#>*C&U7$Fc`ZBle{+R$$E(VMph3_1eT9?>*7dVLUl`Jde$5tEcb3k
zk@U<q#0N(zTZcGUZru!{rRIoFX6ry<d2=(0mWIjLbheE^+=;c(O`he$4G`Wh%{=O_
za(PIvPh*SO$v9*u-auLo*IenZ8&q1KzLqN$@9D&qTHeuI=_97n@QR_yw6)F5(wgy>
ze=5WIoq-c#e0{pw7AmxyNx8y5G^nsXO>GMmTArU=;h!2*7@3`x8(N$i6F0Qww8-`T
z+0+|u=%hJmTcF=^D`XsxH^MRAnP^+UYq|e1-X|O39b5j{7SLHvd5rF9gU;XUwNSF<
zQO6}eYf2V*X*$<z3s@}wIL7jv!Qzh!@fl`YsMm6_;(GrUs28hml~=Zf!Y#igF8q1B
zaLy*%LgAK!5EuSWyKv4S+f3nl*=ly@CUa~{OXmh#^nj}lmd>$T^p>*5-c!Ft*STZW
z7M;TY)^E{kd#ol$n3duOvkaj<v>VL*RHV{|?Q#rVsXcqmQjyp$G-<mmr533@y4Vl$
zO@98_R79=hz*ny5K!K4#NTyG33&$bhuo#Tkgd9ngt9^s1Y*|MZfKxyRhmY<cC(exp
z`1*9D@|8Yvjl*^`5P~Gl#rcs0$Z<3f#-vSo%GSNTB)^%G@a6)qBKTk1*7(!s+I|t_
zlX3@f(2;zYMSBu>j=&V_w#vYS#aXMTLA5Q-v_KaHM9dB=ChA}yx$z3g7<3(V&-JQ<
z)?Owd7&HCVkv=7&waHP8wBH0fYe`Kdk`#zBQfnJ)FG7UCtpV*@SW%+ZfVF+8n{a#`
zYUBPqMX~53ZwMj-{==G-&ZwGWKVvc3!H9`JXZ(Tst&$Sm_YR~!$+e^BQIXo`E#)If
zlfX7JW-6k*1t*b&LU&u+KMK20ss(z-dZaU<#ow9k^YN=jn>!(2DF<N^-0BeBDpIVX
z)Ieclek7$2w5bz!vD<ICE2Yo6$uk?y(G*Dt;eDV^8Lgyn@}8Ppfy*CmPi~fAdy<GI
zI570H$R#zti(+1&te~`Q-##@cYjM*^ZHIZs&sVIk?VRoHOY=Vi`rca4qWsV(`K9+E
zL=yS0XDbv5wYZUDvp`NT7UvKXN&C<DmGX$JPM-Bt5XBUrC`oOFk->bapa*So$s}1=
zV(X6o<(Qa$1+fTDOy5NBx~X=^Xtq$w=Q5;ypolSLcD2v=S{bYJc)JczA>1F@jFic+
z`I#Ll%X#30zsF-^nwR*3=XV@d9=+7QyKa%+o&5LKP<h_V{P)!9lQ!z{c0>J9C%9J;
zS>7fpD}q?BdOF?bV-UV#N?wkVtO=k2C9m+Y_z0PjJ?y&;DB070Pf=2vv4~opf@&e|
zmayLMawI5lr`N$dc*@kHhxbYVZb<s(2Be=Zr1KS%z86YLM*6FL@FCLow(r)H-cR+0
zg`!Y*Cq*M%_?;1XM>A`-&ny{TRDcN@Yi8kA0>14N%eOZ~^R1?avkGZRsg@xc-Wzs~
zInp`yP0rW-Z4R<Xy~e&<&({PzyNu$qrAidy$)5{5m}j{n;&+VX^R%9-R7O(6SxRC+
zqL@-K-qJeo2M_p}Lav;m!3U8+XhMZla-0x<fXJzE5d0d=8wio0;zLZEMkP`(<YKi#
z;Uav|Js`FT&b6CBA6|B>^?NN}@t}WXLVf#+l?zraSe;qDVA+bL^Hwid6{%k3F(Zvy
zf&-+gy*6dVyu~XQbZ6!-?>TwF%2o4LFJ9i0>7KWG9?I=ruxkFw#Vc@6ujW_yQc&wj
z3NdN<p&4q*6n};S`kJrjM8uym#W3MDex;mF48ERn??yQ^<}2oV;>mWwn$@#b(@!+3
z0;G7~v<{NLWgjGKF}OkC@Kw`rGRu3Gp6Z74RSVA7qT$pIX6jZiC)rxFdg<a-4JhT=
z=2Q`iB?*NhS>XXT*0#4K!Z%*DB_dhL5#lgKx*=aN|C4&Zx;syR6mKeOQhIjon$zsA
zc?Fvm)DBw7;Usx$y4WA&R`N3T6bk0E7?Empw_NYHfH#UUNN=)tT}><29AaW_RjcAP
zWhxmvu=c5`>BrC@xrsv3jaDf3yX?x^`0;Rdbd&>F>bJ~u8US^gEgE3^OB7-+OG7Nb
ztryKxR4mW-4S5Qxv@ttW%}=GRbAcj(a?u5fr~|79LC~OV+I$!KQyUA}E?yqwDEQ+W
z&gIK~8uvF)(FBE}WvbHDOV~E&)^0$^^^K}FKRT7GwoSxP9^bLmpUbE<Xq1UKKs5Ff
zMWKCq3KM2JmBK`nKPx*tIz;i8n3wIeotrzeBURZ{oI7i(y_44)fdgmjfxk92K})B3
z=Fv;>6$$%U6XgaZRGkjib|_cK*YYzo{)Ts>8h#hVmN#zGl2oInN{#>Z<A~ocxSoK?
zOR0U<c9U$cB~m(4CymfXSzU6pxOMADaZpd+okTGmbsa18%}7FD278Q3mefz8d`Y1%
zss$+Pn`>RGOI2DzsFg%O?gvTOAz)>Ms_idO&*IKQtLMB*IpE%EMr`613IQ)dU7GbT
z?X4DuC^cF+t#0bEVkfoZePN*R2gX421c$td9w>PCyPbCp$>0*-0<7H|rtv*w_ueCI
z+nG!_X)8v{;gZPaWLq{5Sd8%tv6v}?2*2B2&JXo>q%a~*O)7w188cCEdeyz@q!s|v
z;65yX54HinqMIW7kuCu!*HV93H)f2d3&z6|GPPhBdZ>->HI$n(ODZRt5K1)PTEY69
z$}-iV8}hSHuqGOPGnxFE!pe3V>Ygj}6i{%4SvgnDeq&~)Cjd#AkFeOW>lo5o)3iRE
zHJa;bHpqQ^eyWiQfe`6^n*YZP^-O`9ke35_;bn&nGsw$h#)+3_3BV(g@p2$vL2obg
zRx1W-+J?HHZg&0lxc2v3R{E=hgaZ_T2=(_FHrjp9A|a8t+BZ*4&6zv*nABNoH?JFO
zXSzuEn{cWp7^K+@@OGAsb`Px<#bR}|(6?cTGFQ3u<Bl$06O>8C)byq!>hJ}+{4u=V
z9Wy@hi>!4dlhV*h)$448$C{HS9++sMO#x0xT{AS=4plsh>(r?n5S29a$VWhv0ePin
z5gc{Q)VxQyop3TJP{d*X*zG8rnBb^z&Nz(<*9*d={HDQDAa>ah_c99r?5Y%d$IQl$
z3dn>)J7fl|IX1)~7w3!<7sWTRmav_N@*~2;V{C-4(A|n&8Yc#Rv1taHUVoeoZ;wE)
zpEEwa{_D-+8!-)hyp1o|zz2>~1AkWlCTw6*i%0`^+XxrU_fN2Lq%`JLQ<fDeYE@zn
zQWG{j-6m8e>{o%&X`xM2P%O?FAF;Sw@FiUS(J5V(&a?dybWt6~@t;ud7mPELO;{RZ
z{w#rc;x>T!kibkh&QmrDGtMuqL#W39yfsc^)FXm0;rfY;Xy~|w&Sm}P?y*sl)acHo
z-&LwUGH6c~>04HpV2ch-2}R`|<I?I;At2!xQ3`aY6*hLiJ8@4s2m8}81AJNlCR|Qa
zaDvc&f29rbP&3`@cUc(Je&1Kzjv4pgTfiOK2CTMmzj_trQIb7BPVB~Kz?$TrAlDKO
zSV$^S>=y|>*#<hJoAxXIOG@F6f58By8yaPm#G(V2y_+lf@>pef7a<^_43kb&(66!4
z`wJO*X`BX$-2~qe$wu39wU>=b28+cF0`xQs=$a$wNMl|v!KC~T$_RxFPbdKc4-=ds
zh;u5CoT!vQXab}=P-VB1W{WZ}P^ty|H?one5^|}0wnSDxMYKyw+q^8J?u6#lpY;vq
z`v?=hUM%^#^l|_?-VUX<P0?u8Q>;i&%jal)u(%&NzSPU*w(w&*F(5VY2s9gHIXE;>
zB)gbFI^r&KiZS|lIXy3JDwL^)LZuBOAxKo(<GRn<aqNu(NkvY1A3!&8hOp<xuFu;`
zS~Jn%MIR;Mpu+^5s(GhdO<0@Wo#yANlfV>?C|#K!Ch~|0Lm2rf2Lm5Ab&q*O+FyXT
zClSNG;^=0L`wSa*ig43o1)FT01_>;(f6zVd<zwy92Mg3GNuZKaLL*&oBkk6+DTINC
zOWrlp)QQ6g{vV`YCRI8n$(HE!jt?MoKIt(t{xG2?k;HJLYkGQZdggYg`MIz@Uzt8s
z*g%rbc6-_ajZQG>baG>Xjyb6_cTXW2AxUFxmRUkbA_>9*MbnY9>6qP}#^Ywmz&B&1
z;%K2Fk#LO`M>P@sHWAl%r|D_?hP55Jb$rgc5mwUNdqW3WY=xr-NeE(JkutHBM=3cP
z8cNUNZ8SXMp%zM{(djQ%M{+2V7le4|r#L6!WlPT4sUqE0<6>JWOT#0z`;z?T1tnUi
zjahJy6HX<XScw-9GtI3*n_E}<0~?Lanxt&1?ywxT=h@V(>6I!S%Fq<G0te`1AP!3<
zuz_YN>aR-|FI=8#C&SCi^a2^A_tQI-=``3+rHO}f1yNc^@S-fN2p&CRGT_7iG4tmH
z;ZLHK6RkFN`)L+=Hj6G@Ov&;}ytG%z{HPPdjsSydpy>l5Qa5%H%T^kB`A-dbJ;Iqp
zoAl}^StOGgIM0$p7FX81h88;AY4uL~+65h{b<?IT<o>2-%#-+YT3qr>iOo?b2_hR?
zq!w`JTS&{5e!8ZnrzE6hob_RefaB+E${&eT_9<S8kcl+q!|F<@*#Z(rDYLqjfyDy)
zUu0>Nt7=1{^0WzuDRN`m;*nIIRy;5#rtoUOKv3zw{vu{p68hfFl<AFDRT}_iAtBRf
z76Q!}H-RTL*IF!8{>VOJwN~xO?zDPAkL|P`jh~v-+bz(bBcxQ{IcYnR5c<={uin`>
ziaj4c7vrafpDXe6Zv0$}pAX^Z6ZrWYe!h&K8}aj9{M?G4JMnWLetwFdU*PAr_<0mR
zPvPgU`1u!pUNlHQZTNXPe)hu8e)yS;pLYDD@pA-zy6|%xeiq_qDSlSsXAOSNpr26i
zmz$Q*-QdZxGV%5<@|#Uvapx&n%@G?ck&Uo+ZBY7UQaWF8>6NBSf3aP<Hu}3$jQ;ov
zrPrn>w2c`m4h-bO!O(k@@%$0WeBcOWWL8y%16Po6kX6eUe0K|V#q@jx^e8)!cJE7g
z;mi2BfqryQq_5!LY?^Q+m%Z$Flj5C|yszU6E`6inme|s>fB34|WEUNb$+VhocqifK
zTl97C31U)EEGz3SBA4`dH(PfRNu$R@^kq{*Q&HOPqyiPOJ&KXTF25GdVr}ZMFTQ8P
z$a`_cOgJlo2{m%d^48c)V9RF-)THdun}Yvyfj_F+YKKY9;TMk_?-zpq><Ikuy9wL1
zPet3U#ZQ?1Jsx|Yo?idNdk8zihv`R^q~-08J`Kxunxu$$og%_phccqv1K5r$+z_i#
z3OB@KLdZD@2x$Q783A>)1~u3uXR`w6uL9_e4FGw)L}{pd91ha^@beG)Iw0K9Je+hg
zy@@`%Y=ee-p5|dLz{9OnR8%a-Fei%wd`FbZB7J@Ya2DwS8~AM0H;A7|u_)9(iEaTJ
z#hIQrX?i}Y>G4oNLYa64@Uu+pA(XtS5hcS^aHt7JP*c&rtAy1@^+}EDR!b=Mp?cgL
z-d}LNMsU5k5w4Ptz;{<E16Rer8^-kmjqCoFaG3~&Qv}z=5x71~5)vE?ni{e?Q8YL0
zPc>MYUmU)KT0pfistICsxDT_IJbY4=?_5+gtJnHCm#kMa!}@^BW`qT%iClJ+kIFk5
zSK5JYHeD;rY;TSfJzEWUbGLWQu*XV~$KlFybmAHIczO_$UyiG^z`m<y*F|*O3@SYy
zvvdiHUTQtzSXImM#O}{;f*h{IlU1Q_nv#ao@I5a!<|r<goU-8kkXDPKKzU3*_wzGQ
z97e2KQgi-A3!?;L7*6f=aj*57E32Sp>jc945kU(JtBE(36>Le<n?z%a3p~c0>{al3
zlfW~f$Z`TNX`+*8c<z{F46;vw>`MaDL@65(kV#V@M1yP)uB?FK8D37o%O!zl0+?+O
zcuA9tG=LW!$&8WZ6|#I1WbA>*9b-h0jg4`>w_N02WI#b1NCM4-Ju3;cq=^EfnbtV|
z8S`yWfg4N$&O}7pDB#A{E8;_=F~EWXSV#iU1WP+x049xB5Y4^DQPa*r4UEuJ2)!f-
zO&qm13c|6qZ^M9U=L^6M3gCt$08NOs3kBfVCXLuwYs|nQg>NVppWa!T=xdkw_`EG<
zG3KuZdG6cnuLjiuL$6YoR8@8F<tf!W8}oIem9sWS4a{E}Mj#%Wz__ddjB}%5yidXS
zqJr_pmSC6&ZSN8oH3K88Y`DP3;WLSsDi@;StLbYjeDvP)S`@}jbAtl;OULq~A)idl
zwhs!xD-1v%2{+fpKE7~s{YF9fd#eyk6t_<bgts+-aA`Dze=>w+ovsymLpGSeZeI`#
zS2n;<3uEwZ@Ec><EWI;0Ug7XAN7FSex+d7$je=#X!4j64xA|cFmg0rmmHyrEHfw>#
zcX7*nG6CRj5o}jA!uC!dMF`tf_FZ1fz1zOa3x3`hglV``(_n95>#{L4Z0jEaIj=76
z6B^##h=!|eT-VSI?|t^&YjO8F|85xZ?HcjVTIRY56!!~3{9c3D4-|(1VcA2vwx?RI
z%><A8t<-jnsZHdM^<^Id;o&V-ykg{8jqfGfw>n;!fO1a=zH1FWU)UJlM}zS6Huf2O
z#k|^Wdo^yfw~WOEnfseyxjum93ThJXd8zO)U14X-6`H_v+nqv1vzF@vc7@yH5mPuz
zSNOV?D>T9Ac9sf1WLLPWRCu(mkW>Tl%8_9A)bZP@Rqs(8t2~CEC-CzWexAY4U-0u+
z{QMn1|G>|`@UvZkezwQYF8FyFe)hu8e)yS;pLYDrz|S1~EX2==_*sdcHTXFlKkMly
zv_Utq@?PrGUK`s!Bs$uWx0}zuB{!<c^v1fSR_KJ}K7^s?A#8UanD+`?X{8N|-JVjY
z9jPcsg!)nLt3LDd-s99+YRftY$qcKrR2><~mpaLpd_(8bWee$K)Bj(&)Yqwwe>z#+
zQ0}Bn0z$K7Ou#J_PdN><q-)UwuEWpObh^{yT~C*_smTkddLO{YzoL)557JeS_aXfI
z!}LQ(MW^TKni;`Y4<lX?U+Dn*-Y1q%+HUFvTfw~@d%W}_di@N16IbXhy?&v5#f4_@
zkzeSS>kI9575W1y^bTBcp?#)M?^7V?Xz$bXze}~l_h&)q6(Ia`e3z3x+=Z_ZnZdgo
zSD&Y=E!AEx0#eQe>0W{KLxGfU0n(2I(vKaG`dyG75J*1}NCPcE`k6rbxdYOm3(`XZ
z>0yCXXaUkM1=6n^kUSTp-w32f1k#2UApK4t{oVm-$OY+-0_jfzX}ASQe-=oOIUtR=
zAUz?Fo)k#M79c$>ke+cs8g)S`N-~_k2&6Mxfb>^^^fw2jk_*!FD1b-Me+Z;<3y_eu
zM~zGWazLuMAZ;fJKqujfhstUTkaob=JXF32S2087Mi-==1k%m|X;TZ3b`?l%4oGLY
zAiYE&y;LA=ZUGV&?}|&iIUt?wf@GAVy9=Z@v;b)je9aBFrvuVCE=YT!02c3838Zsd
zfMmF|w*%4}U6A$_Nc#z-^Tq_}H4aGUyC9_m(t!f$f-yll$N}k1E=Y$6q(cSLg)Km8
z$JZ=NQyq}r?1FTdK<W@k7qtK>Es#1LkS=yXnkkTG38YI}fMiDRBOH*nxFEewARQ@?
zE^PtQ(fFDhuFC=GG8d#d0%@*5dP@tCjulA9IUv<skd7Bf^99o7EkIfzkQO>1UEzYX
zNFXg1NN;Tc(h`BR)B)*jE=WBBX}Lgpdn=G+j(fcW(v>bqs|3<&f%J|RAekO_vIEjq
z7o;@;=~RLA&K4k<Ic}{3(p4@<>jcsn0_j~XKr&of?|}4f7o=W+)F+VM(*h*JrMv^u
zdtHzQ1k#{Dy1E5O9=_(Ga)SfXH7-cQ0%=4by{`pGqXOwn2c&CVkjet7B9N|Y0g`FB
zjSfiHyC9t<kTwgX_qPDai1!;DkUrpobTm|3ywE*YAbqd}Nax{emZkF@kUr#sbgV#n
zlR)}#3y|I{kS=mS`iKkCT!D0nK>BD4kS-NSmpLGP%mt~20=VHW7f2s(0g@5#w>ltw
z!UgH=0_jSD^vN+n+8TuP*?RTqr(BReC0+<u38XI}Cyvs|CQ+Zl`j9lN=2*h}3~IfZ
zG^i9~Q)>tET+*8MUANn=tvlae7|D~{k;)su!QtkW2C4`^2ybV&1hT7Nwvtt^<nrV&
zm?evzbc#H-$$gL$4){SgDI@?nLzCBZaTMt==z?>PP=G*$^+r|#_=NZ7*^wGi1L(v4
z!a$Wg>L`FMO#7%<ofrlY4>J_tVKhrYVY^b~Th4K3k<Fk4T!@^(-cvqEev!PA@S#dD
zK`)R;I#HP8Pq~5=dyS743*>eMAEr(D!oVOAE8!(MJ3$%_;_&F|@PUyJ963P~bX}tI
z?<)5ZGx`R}(TxkwaCVBWR06d8Cb`lnP{ax{&tuO(00v3$A7j{l9e@N+Or6+)oH4U-
zAoKC4f_#)zT8>aD{$`AedLnGTd!PE68nr^9N-4n(q4Y~)WSa&le+SnOz`6x(#%KuK
zDoL7AH6O)mL%ZvE?*TG3rb>|!5SiqIPY#KNe$K^0ZHj2fkXh9rK`A1Bh?QJM<4Mp`
z4nR^Q15ypy6ebP^kfHdLB;^U<Fkh$|?w6ESqmMe{<`gvrb)o>Dn0yrkI#=8@GE~gw
z$a9amGguuMpa$#D_T|ZSUIe>L1+mb;P_&=iStSbrko6XeLrh_oFZ0xo3*<B0N8N?m
zo~kE26c9$qaW;kT>0g{)g6_BXl0PDQ4g$C4lzv#I2+qAFa&?EpEF@fdlb_`&8Dq4Y
zbD!=~h;2h-kPoR{k33nfoa!5)+)`*-@^8!z=qsw9LNp9hWQYQyGbnMP+L!C$3}K7c
zaREK~zI?e%ph5ZwC#Na%aEU%y*oY{2+yb2@y-mOjG8sWR9wUgQXgnj+DRRUnw_*$j
z*(yCPs8$}O%rAKw6j(0!OMfiDTNI>|+G>a*YS6cocSnl?US^!+maj)bA{zogDWftN
z#fK?WM=rw%niPUufpR|hQOTK0D%c8KNvV~YjroycbpXRBs-z+LsAeRMLGedyBdPP4
z@Te~7D5z+>MrCJ?=%scI&=8PWi^1g8ikJOCeK>nohC;~=@H|BIG4j>mT=uNOaCJDP
zAL_x1`3d2n6J~Ue;s^zx$>phPe-6X9e2c_$(_n!FQQCpxR}`y5@CK%^8>WO51$Q5!
zkknLcw$wN1&oh33%P@74V|w2p`5O0;YdU>P!5PzSquCt%`jY`L-8Be87hw?Oi)EUW
zx!eZqq^HVdoL*If0F4#pN}g^LkMNYMO1Dv*NE&l8y@hN!(_6(5jx266)btn5;@0f$
zUO_@ru96cyM}A(qEQL=xIje~O-E^vh+;<DPRG9=lOJfmnpkqA+Y$so6S`birBsI0S
zS$}DjE$+YrF|(pqt8r?iIy|~LSL{n2c;Hq(wBFL-&>9vA-g+#w{PUNHBl*v{mVdXR
z6ka6r6(0#bXqJENUc{R63$8*X7^rtQiswRaFoh;vQ`XjD6w|rmv|}tX?xl62F0*8u
zs~1f?+>hi~<S7rpl)#iTr6V;BAJhM)sZ@MHk8$rC^()q>qyKNYNWWvY&(Gfq>EFgj
z$(Pe?jqrPXjoo68G;f3@!gpM7?h-hE6gb~)0nVd|;N0Yb^B9U@X?tAYe6IyKPbPwM
zvjfiHtqRN2g5{PLusoXx%l92vyoD$Fta(o0+}b=GKe@X<CPZ}K+gwm2DysLqK>2F(
z-8Z}?ydAZMw}eq+%N;Io+u=BzMQ9SPcx?H)1+G?Wn~xw}w7sG=I*lPJeILSI@V{vO
zs;cRV)+e+vD^tJ_4jeO##npg6-Vf5G%SB-Oqu#_!82(E-j5o49s^2`{m5fvOxHz@5
zaB3If)Hf_nt+{u5+K>BTO|1-wYT2F*B#0sEb!3&SysOeFs<+fUse-estSoQfRSWW{
zOvc1Gxhu8$)D;Ul%w5dTNZDGZM{DIaJ}{r+U-+9$QC~7M>BW=hEj?*LW@dV(DM-Dc
z-<cJ7XIA|?Y29f4H|K_5aaKGlLB+Gu%~u@6rPllBQhG|0$+oMF88mGg;=5!hrU+#r
z>*D57YYEUZFVI}jb08}Twn|v)LHKD`O)D1FM6^bv7R6VT(~RauF3#U{m?9e_v{m1!
zxwOHT6ox63G>3w9(qEU7p|X&Jh>l!>p4d|=T#7+J?G^YNk1kg`z~`Z2VEpRfC&{|x
z&s<${PwA4clrH%#t4r3ZnpBHTnrWo(2y;p2GgTdMLRq%A&$4)C{W=-59&s^ivM}o)
zVb*snX04fx@k{+h@{8C7)JAQm!X@gSU@}LA`+W3R1%aR3c)Y^na51=pza-<9yyqCZ
z@Lq77z`yZljbXTz2u$QyhOZg40TVf%;k|@l?AO%zh5~4e!~f=`MmJXmX=}>ur2{63
zwpBG7#xn%2;w>-j=XEnqq|}F3F~3VD+kbG$_EeGWX(HP<S+ZSQp(ooi2{E#bfou`K
z(1^$#U_6G!FtvY5M(v+n)XorUX9~49+tfY+BBCa{|KDb&-QxDCPbX`VXI)J)Tbg8!
zG|BhvCRwWnR(rzcpw@b;Ei1P1pQ5!adcVdV7S4X&bIFMRn~V74h4}eG{B1VzeQx42
zdi<GqCQ$q<FyQ%Q4EU#u0gHtJCkg{@w;6C*Gy@jWYBFCUZL>cP=Tll1e0FH%A_nVH
zk&-}T>B*-J+l){DPR6H6lvq60VP%Ez>Gi^=J8V9!i{=ye5(o$@u(lcLJAfwlK6^(O
z>8A+kYlQT>Y|_`nkdE!%#8d4(Dx<^B$td5|MY)N3lo87Bu_^y*+bUL-2X<9-CX!8g
zWm>`X7dUQf<ZOm#e7MXT9@0dZ#S8TF!DBYOigO;Q)dThT=PP9bP^XU=k3~T3U;)XR
zR0=N3>DEN83L+^-bKt;#w`9EE-Nk!Pq;7-o{)ZOt*BnZoTf}~(EKK1zP%9A38LTV|
zqZbq-%z7HX!mKCFdoP_C1{Ya4#Vf0d<bXF&xIHmWo2Fj&2g2Ar?L5vLqf65}UVFo?
zibJ?xO=D=)H!f|x*GLx?U+JLZ_B<)hF{;l^9Wg7^y01#sx_i4?ccZlKCTZOt*{%CE
zZe2QdqJ!Lk)Dti|eD<vrIJkYj%!k0F(O|to)Y_3!D^xfS12+XHe(mtX2+y67o)wk=
zI*1u9mT+{Yb{Ixu!bT*1$D>0-I>)3#Ta-?-VYE_%v5uQ02<1Uhz_w}HG^SC@(U>Jp
zlQfYMH%A(U!uiq>C<Kn0f+TQ|&f<_p87FacQzcV~k;?;<$>l*Vxx859@)D8D`z^Vw
zeXFgT#KSNq@o>ZT3VEem<D}6tE6xjS>uA?YdS~9)l0lB&-!TB?P;KP!qi0|IZ(PQL
zmuje8j4Lz*Oi9M-sV-Izi=*CKh1EZ?S$$gDihK!k4o#@#Eae>Etjwnn$v-eAga$+A
z#kg=`Z!nxr?)g+VB#a@BOi#v<P8UbsCBpY^;mFTyj`TY?vPjw1EX0v>aDoa*$T-PJ
zLhc-1&mC<6cn`++#ga%~Z4bJ&W?HL#B+;@uTrk#rX!59W9yIehJ2(KE^I=<`JYVU>
zRo$dt(3Xd#{FzlUZGxSEF@7=ZRKu;_>~KZ2eCwd3?|IJkJ9l#6xOsA*lha$!z>6-+
zR>$iS(<!oxMABe_RZN8nBeGTH+fWx05S^smD%Mot^VpO5$C4kT-K-To;I*!f@}O9%
z@)e)_y$&DMAFVe6!bj^5`#vk?Xg$Jm;z$>q?@Oo%zG65>1>i&+*Uxs9Y2sqsh0?jq
zIe{`iMQkIrh>2ye#fyw(a1Xu?Sr25{=%RK(o6e-C3}|^-qYXQuWyC5oN9meVo3fiB
zgnfR#!_>Uv=PSe9w!#SQND6#r*ZKh5sDaq1<!DgsOG-r39Oq)A3BAf!%*KD4AvMyd
zWxlIWW7KlOX@1*0YzkH8&40HCw`^+BLS-i9&*_bZFt*>)W0(vc-E?+A%TYG?qudDE
z#?Xvd<tkZmQ3eP8SX&cf3yF*=8w)T*q(nZIJ{do9I*=rfUSjZsv1N4;W}GzAs?DY<
zkQcfBnwu89LHjl}AW7g%F44EYHu&@TQ8JLEA(ga2Y;w)UP`r93Lp`0Png(25Ne9wy
z`F6lCg!TtM6_M8rx^*=jZ@!y6Fj}4Le+Uad$`DNH;@N&&2=5SMaRD1BVqtcuOm&C;
zl8%CLC;heCO*7LfYnax*o6;E6)5~B>H_xAnXp^E_N-aR>-{p*A;ap`Ot{7t3h{EQH
zh$mYoi)>lOFkv03P|Vm2tm>5Eyf(J-&w$rL{?<{cs?IpY^hxMHmfk612t38r%}g-o
z_0rAO;G_Ccz7eB&l<&pYp&>BJp6gT>91}`7CvZ+{9*%966Tz9aE+{5)@PI%$eL|qD
zb3rjtf<1w9M)OcOzL~p`GA=kK9`LZhS>FPjBEELYOx6X*#Q!~0;Pf^RCxOiLxuBSU
zzZHR!n-C~@7Zel2Vv|7WpAaYmE+{6P@7V%na6+IITu@Bl-g5<t*E|&OJbdjG-3=}{
zCQ9!G0%xcNI2R^@Gwgz6V)9-ja7J2yBQAVq$sb`DS#-fMA$Ye4oL^c?z|ivi(z@k2
z%>8PICR+q3b%VDzWx7$ve`Nr)=GX{2sxCSLF?@xNUr#t4`wJbLlhN@87ab-D@Kr*`
zBNiQNc7uh`3ca*d%i8+2=%99Lp-6kQD$SQVDCr)#qOb>H+jRNU?sizPk>Oo<$;Qh_
z717ULrl;Ft*IQInc)?|@O;?TP`wHaL7=m3+`D_36ba0Uw%Jx#i&h@lHPpzLvjyH^L
zVX?G|R2q3QubdBu$wEInP8CPirx<NJypD<^<ekGRZThy9--PWSvx(FcQ<ayNa{nW*
zN0<(#G8w&5P{9cGva~}LStiFINs3Eui^@F+pT)9(vK4a1zOv73?Lz2tc|PwR7&J9i
zuoYY1gEm?|au+?F>(I;LxOI9gS$4zfm2^mtctNtxaG|R+n1Ivom(K7zt22!I;H2#m
zE=fkir7jvw#OjX;4S(21G+dsHhPS$CFcGgmD>VGcqM`Oo+j5=0IaNk-W+l<|TOIPP
z&sE0iQ~}4FaE#`SGY-1uS1ENW9n)D)y4t3vd`n&GZ#0OF7KhYsyWI$Ln)a2X-p*&>
zt%M%lrTQ(6kL<5BG?X3AQf|=RJbPMc&nFsra(T+j8F*0F@5W_&+}zA~n9L~eNY;4o
zbTys{5q^U--k+_;OVWuJ8IH$q^FzZ&7b%M}^G;g5eA)8u1xquqln)IRS0|(LeJ(0Z
zu<>sTm5;YTr8izwHqciDv>qWod^8y!KJMa!30QuI@Zrgc<ip0Hq2UZEXfnzrHB+q=
zhP3~vJ`1J1sCb90T+1AUfT9`<QRbmCxx@0|e25R9PsWEYxcFegpg$mdc)EE$P{#<E
zgnTI(xEoyHOw{xT1>CdEgR8Mh49vB4VbT78$<Tkxh2F$e|E-{Z&O*QDTxgBE^I6jM
zvI{rPIJs--tl2|(a@*jOPYOLztd{7cn1<h;<*O-uuAU8sC}aD{OUXkONqa}8RFo$#
zdCM33Y0-z(qgLtX;KE-ZH%K<2AgB27G(w>!IE{UF1dF~BsZCKe31N!Fmj11)hnkjv
zkl6|)YbMO>NwRvdiW4=m%T{TG;c@L%$!?jP(=oXW7Kg|SoOQqbRrVGR*PTwS%9E70
ztv!KM-PK;@d%LcEuu>T<cXf6S@LdWWFx-h|oKEK!9(f(}KRT74Zs*LSXB;&(q_w4g
zjLXcl7c{E%;uGqRHzgCAn_WU<g5N(TLi0CEXlkDCibd|=90*{vlI<mnDDph6R(#)Z
zT6fbzU|1OzAS{CNzzK0^JlGTN`rl8+hud6yFfsX`5k5R`^WkKl59l?<mt0#?a5t)3
ztBi|CglWD#8O?XNXf|m9{w_5C)24Y@kY;^+tVWD&LF!!*q(&&T?{Sf8PCK_-ONtFX
zrn?s(HB$9AIPKgXUx##G%GBWh#%bp~P#kmG`C}Iw@5c-55zAN1#$Av|La{OOwDae#
zLd|LCgLst-on#7Cr^#Or*>Pl35t$QdI8vsI>9#I5<4|I%H`Z|V8K2|L(1<opL;k*I
zc^mKX_FJ&X(#u73h~K-I{s+9kE%`@b`k!#ss6*827y0b#QG65X5K*qyf5uyJ=^7rx
z3k>IRT(J!7V9UU28j6(83<d-yQ$?zu!jPzSfz>wSGWdGTtH5&e89=n!AVN3#7Z=ec
z0rqo3^o};sSGAo$MyEk)X=TKgJ2dTNwMV8g9O~jZx!h%WZjzc0x;LCokCL$~*~^oY
zH1d7m_T;*ed_3mq+7us()@f;w<+P8sTUla-);0f#;BAC_|I5W&GvNPQc)Q&IwOb?k
zZZP0a!Z#uL*0yCw)SF+#k4e$iWTxLwn<mx)%XF|C-&o97zw*loOfR4`X6plA*K2H#
zeYFj((&;T_j7H6`EG(LIu*OGLhf69ru|_!$pvC-Y>O*9yXqj*Af;Nd8O?Jf#Jeu$o
zk0xza$I_~mMnG6I`-l8lY>`h}7|3jdidKtoz?edog<YwWMrguWOZFxsWmuR}=$|wl
z4y=a#Nj|sQj~qtvtQ>DFq)21yI!Gr!R2?4Sux<p1^kj(t#rq-J!;|WcB65iW2)q3<
z?O!6^Nt1$GzFe(8RDICf4NPM9uUCSF))*Jj6t8kMMfG&QDfo(;;^kIT)Xr#IL|a5U
zR0vV!TbIH{kprnQt(L$gpMPIyX%-RP(yx@3J~&xRAL?r9)1{^Pid%XwyQK@#(lYxf
z?EuY1vv6x$Hu`V=i@P@;ve9`Lx;*uEe^kloTNE~w^c$pE$>lN}6)-!bVj0rA&>on$
zYP!Amk@h}1S$ofRwf7vnz=JGbaeMD;xA!;N=wL8K*7ah}r?f#i^6ILPR}xw4Wh<2t
z3VB13^q?kEhl^>gapDz3rjRSA*fp(`=lE%4EgIZ_!)J-~5d4}rH9#>zkw!*xG!GW5
z75eAI3>@G1FP5Ri!f}8I$C3!)h%lgD<`Rw$?2}nI_=<(22OrfhzZ#9p*k0InNfA(%
zyP&MV3k-#?7|QFLhvKcoH=&iF7XfFr3(k#VTFzGt=cE?ke6!xPJOa+iE;u&`9KK>W
zr?dd)tBK&OalttiFK}b=6~j5L1vqQ*jkB@Vx!_C@G3G0Vb4Ckrrq-#AqLLbBTyWOo
z1?CQ4F`Q8|<hDndo#tJI9yr@Cv>&f>p#!+8@2ZW=7))372Jxl8c8}P&6<lz-gyo*V
z+0Zhap&*=yJ^HW<&anb#MBt<h%aw9!xDEb;9NQXhq(;*;NcQrWs&wF9n(Zj87@5hF
z#dnI$qw99wZAo`fsx9hwt@T0P&IE94DIdX#vt6v{5mvlGSTWgV#mctDz7+u(*VCS)
zQffzdf)3ucBI2<1H~#d#F&Vw*yXZYh=)FMbJ=msq0Uf3dk?j`wG!ON!rPf`yEl7NG
zG7>L#k$Ae0c!`jBs7>PQ5kw!(VkFt6_NkUJ$hIK&vSj4eT;%o%xt9yM?KZh5_y-<n
z(YbTUHM2@#hV8%c7u&ZbBl1cYk)9Cw4k2=yP2@6ZQ|$%`1qF3hin?u6IsqfnRmn(w
zw~N#>h1B;5sU0?{q4qpx+qLI4$%wqxMdVpR<aI)1+9q<jFSG-BvfkSEW9|o%5&Iz*
zvA3+Gxr$df9~NR~*u?g<ot4gHh*Am#KAg$4Q&{~ubLSqDI&1A_KR}KV;_>gOA4^8-
zCtS2%Bn|sXp>>u`>q6g)k~(N=w9vOfg?lmd;cEP8{7f<$Kj)&cCNzFtXgs2xMjh!D
zBknQVY{30uG8(_^qS2gR-XJu-&Ze;xIKSkzLOKN#6w+%9jke=FhF3{br?SZk9n)0F
zg%Bb12S&0mU)dI;^4F5_<3<-h-Xm@P4dKU8Ha|k6@;TdXRQ`4{BERb*@_HfiCLyxR
zCUS9LF-&8XS_yBHHSjIT$h_4>=0}Ch+l0(HHkqNeoU`rP@(0O?ywgRb2||9C5P6JE
z<N|+&rBoQIuWpMzd2ccj?{ks(MQO($35mzqBrfzx#5g=>n~lQ{B%|@CE*ifsH2zFz
zJl>`;B&!E*J6U}w8Iiwm5&2yq@|Qwnw@u_5B+NQ#)JyE`LZ$*E`As_Zmp%9VCug1C
zBxB5PU5vR+81p+}%tD(nAqhWcn@RW|lhOF7i^jW!#y<;<i|T3AaW}T@it&kLG(P2`
z@qVH4X`%7N{}LMil8nZ`x@dezX#AVdxU`-|y(&g9pKUke{398S|8mjzh|u_Np|QuN
zG3(DbFz13%0x3~xorLU*oQIE<r_iTDZ#rgs;v~w_?FSPnOJ9W7;!IE;74mn)tE^RD
zVUsVz5syUYVA|ali;7*6(b(pq@oAy)#X{pso5oWEYhd!~w3HL%g9829HdzzDEE(A^
zcai;fA$xZrd$mpWlD1Xk!ipAUr7$8KZL{fS&t!z|<sx*FSaiHf2tC;*w6Cq3lreA`
zD)y5uW@>8sF|=(xoAZ-))5$zJXbe;OM#>jSX;C**JjSr<VnVbR`y^w)el7;=A`IAH
z7_i1>fNTsX#L%`Irc=pCoa`cTHzDyLA@MYeM8w3|u5Df~5lDVRDj_bpjw0QppDb1R
z;vkF{IpGwMQITD0dO7x2JG%F3IP2iQk&R@{l}qKbB{E2*D2wFYlkM$GQ`E0zG(13-
zs_HXNV+)Z=w=EVwZ@rpT@jCek4JkEyiWNC_qCmwdWR4`?aP@MzE&Ny|kfgMH!!Uvz
z8YohTu|cv>K{{amcsU)}l{OVP@oJ^bCsCc`)qU2kott1~LjFRaTSXw}^qOH|AAddF
zp~=Lc-6aNli5N^3F*w~4gW7uP>V4#;g+n0f<p|V7tX@g|n=(1VZwva^vWg$ty)7oP
zj%3uQUDWR{)OQN?XV}!Q0`<kw&5C#mU^hr47dMf~;wZ^+f2up&))O+Y%}Pe_5iWuc
z7J^?Z1h2OV-sCUNaDvT~7b_=eC5Dr3{vW!ttkINrg+ATZfz`H!6daX|DP1n6941Ve
zEllaPnKDXJz+x_9kgrT1Dr}(nt`{zX$gsJwfQ+QH<K9@nmU|@CO}S_JqirA}$0Xy%
zaV~Dm5^l^BZscrkj1V`lGH;d{_+}gMqB|Kc7P@$GwD95t;YGjAi+hO|qs7to4Qo4c
z>v(|Nh|q{UIEL&$w5Khp^;O@Ybo1oWF-ma{AT|Cfk!CRTJKfFTA=_W1w8$t@zVxFP
zQg`I&wX;)2nY!A@%pC5tsU`W%3rZzQXKyon{AZ&Rlkt10i{Hlyzn2NW2W@_TkA^2T
z^8}S#?ZhyDQ$Z17DCIbr;AB&?rdO&I+=6(pg6`2Bian_qNlOtL!<H^yxIEQPfom$$
z3lzt)pWdlVrx_gyyOD}oC2cGVGlRB1nDNPszOs26aCk*B4zF}^_ypnbD&eqaa~M8%
z<a5oNZ<TJ5hBR!~Fx>@YFCe3J(*f~hr4iWinnlK6bWcvk*flQ3_6TE76~+!(j9v43
zcEV3rN6D)s-@bN1M{3=)X$ymks+0Ips#`6lpaJP*#R$Bv7w`bGkXFV0bWOrXF_OxN
z(t?#JEY7kie^6G=@x0)rWM|~A-<48xIAoqY`n%U|K*-Gn^uNf`$d#Tln<glax?H%(
zjU+p%JT0ur3%`&2`hkI<(trI$ESsp?^=_s>6tsBXfG-ykGL2>-(Aa<D-{PkyYw<H&
zEq;=;ct%=$#A<Q!pzltz@cUCGZJpxR3+io~4HIzG-&@ZI?bCdC%e8NbOC+9!KF~K2
zjqoYTxj252aGbALw^p(^p1{2;>{Q=h&jv37+@K5GWde?`7+htn;JgTM5!T*AE^v2>
zwKrcexM6&h)Yf;4XKZ~<fWwHu*&{9}UpR@XWlJBvVkjRme2R1>KhssHF~#^giswQ%
znnG3J>;?6nn~N64`)*F8Ke;ZuQXQd4S)^g9<OgVAq<>;HXl^LQ1DVm%uDgR$29&Z;
zd}MRkQBs)u4(}4VnKI2b84kPUzTU_voINPtV*FdhHxb^cbup+d@`-K)m{vqnjA<b(
z6q0&wbOh?TI`VW}4CxCXN^Ya*3tbYouZZ!zB5`Ng5_bzE&Ug6aWqp5@)FtpKC6zdq
zjg1mb&|`MwwOy#S->LEIrTI}i@or=1Un>=^aa7aPmBp)hHEn0)u}^#lFDz-LO{<s9
z6iOz`n;?>G>IoFE3SB^jq?BT~h;pi3$p+~VX^lZDL|V`vPtgHJrZ4ZGzKT{uh(w7V
zvS1_^yk!*=K6q$ukzq$f*y6FsuhfK#y!(tu>x*=^YWFDby}3%U87H}yCzIT_x+K>`
zANYkx?%B5FevTuW>I^vp`4ODWk)8)y4vGvVNrbqUH<Du&1|Z<x4nZjEV4*4;C@)&6
zOxi{9l8~A-6{rmZou7C>m4@>ayre6JCp!L48O)DRiM)@dxJ|UMB!@vrf?j@>2aDAq
zT6tn$OhXLbL)nrWN!_D^Mpay>K&|<WD^i70lb)zUS|JKsTrL0tO{9>-7imkXf?$EP
z6IPr=|5ib;QrUhQ<%8sq*4*EL`Hm#OJZ+jEJ1Sv~^tuY>g_U@Kwua@m7$d+MCX`EN
zcx=!OPCEk;4)t0n$iTSt(fLY>piE(gqlxTTZ9XII`{5XF-n!W|k+94Bdy;j`t6d#)
zwM;3GNyj|L?wAy?i9!ueK*yvcZ#psoTxX>va5SP+DP!oS+^u~6v4GYGs33LKCepbC
zqEkSwu{!zh5}}v-BIMH(3S;vAWU}x<mn>vO7XB)-@J3q}C<Bj?1&+)o9l51}nj#^q
z&=oWscBGb4l<Ku~wa%nbU%?}FdwD7)(u*5Kb$fGaNg8V@7byk04o4<gy9yCD=phHe
zE>oC`;wZ;y8j>*>k?>OC*YxJOE`<xGAX`qP94wK{-?)q;XDA_Pz%SyCCQefxh|vt9
z(qW<k$xo)KQC9oz=VXZaWHPb(v`egtB33(|?C+D#x5P^Cpre=fdabiD`C2DGf{=lk
zKqel_d#7h-dx%Xq^m@XV8<O$mt1iC00^HzD=^nzDH?_c*Q<C$gOzMB=<y4R3&c>N4
zsVDG10(aJ}3ym?~N=Ec|Ttx3LM88Ieesc>%pPQU$Y55G%PM(EbidMbLQNe3a8b_@y
zr&n^Q_+^NSwAh!KmHgo;q+wi)=nrAmd_Ng$Zga85<Y%5Dthv}?P3_lXorlS-(pnwU
z9Xd-Svl%Q3k$gE2qt&0DOa6wh#TA(u3Y%0aHrV&<Nuc9n*WDA*DiJQ?KXkQ<$=h_e
zw90+>sF4<{-tZ4vEbq1WIvgY*0?wl@I3{n?83O0e&BL){T|_{6%mu~dZORIi$0r2J
z6D}wwZ&Oa7JlQ;ygn65ua=|fqn+62V(=EU$;A^MMJmZ36@-}S{IL|f@CxOiT#RbLW
zZ5k0M&rJxFzq+89yiFy6^0x_r@^=>$lecN3KzV*bp!~xH#pG=|TcG^2c_`jF_}VGD
z|8l`Gd7IuSaQ@u_obwaG*{(pK;|#4$-lhu$&Ln))<YbU6Z`1ZJI3{n?B?4!QwPFaZ
ztn23{)*?2So#Jd+3GW1eZoT#{E;<5vn}m+bCY+8p3LP&=M#sxsbeOzN?-Dv{79B6-
zZF(VZ6U-4(l-+7{bFSEzI`F`)v6+v#i*CETO?xKm412jcgUQ?ULFo)vSe;?q2Pd7k
zX`f^??B}Av<Zb$-(D1fxL_;bW4U=6on7mD25E`zuXm}xS)40rxm&=TDXtKs@cQu~L
z+jOHe-d3yel60bp$=j4pM&%3_l_qb~Ekfm0El~NN%G=bHj1O~Md@y;N?h`(|dm{M&
z#qc=hZR$?OhlMUan7mED5I(%Oc|K6b2n5n!lnmU7E^sDq({BabHO+%-GH=t0WawAA
z(3`wXPYC*JE%YzsZBj`I|2y(FHJF)_<ZU`Rnb53p3606y^fwWj>n)-APvvcjW>2{5
zpO%adr@Q!I@-}UE3hjK^;_m}CAGSf>rX)0<k&Nc`E}Bi=rk#c657{(NT;8T6r1nOT
z8llk6xkxo>JYFiK=J8RJ+MV6-bx8N6(|A1DyfJ9BwVp9_44mcgk(DuJkUYx#a)Lx2
zA*)C>n4t4^m@FwPP~QY-?qp@0M2TErEP=my{>eO|N3eSUjGbUtq)+KmI+s^E)vcm?
zBF#c-@?dp<jOYgXDTf}?!l0;W)5Q8RKw7tLU&@AIh|!?F*~Vsw<9M2N8_tG`&}eES
zbtNjHHGTaV*q7)lI)proEUf8&SCd#7`{*>?4CgR8cWe9Lt=`}0t-+gWuhHk8EEVOn
zNo=IDUT;+=(qA4TM3(~_UtK|0wS7)tle{6aV;Z1VF2gjBOG&qF)m@}rRdoed<vq2A
zzUKpOzT)CeManl2O}L+-=UESBd(Yxk)y;Gnchc>_n>=G={a3ugWt@UCpcPrSwcWdq
z-fQ=MhC*<wr>C;@g01@QL%91e{iwTnt}8S(5Q^{7^Gkg#=4bdH#}Ny+R240c&>g~I
z#)e6!Y8<s$@vKZAPX@Q#DdA^6RIHRUnXNtqJ`>ylMX7-T6>stGCF`0H3-6;FiiJD+
z)p|SOO0$rew+n7uO$99ZigykEccVXIeUF+k>s_}T<ammzy!xki6}{-41WpF2y+R4e
zZj}2<H!%~SrQ&}@D%P21tW6HHp_iyuvSb*rRW)<1M@I`{PkkC7vc3x|dHBOflAMZ=
zrd_A>>?`FFmykUF$zNYtTDBEN$a=9rFZPjt9f>~3-9=IT;&=h7z1kv*TBR^jsGw0;
ztk#ZItQbt255{agB89Ahc-8i<M=7!cn6;zA>^7ZIGd<>eNDGNSWUsb;`&7-o8-1Zk
zCW*Bjs5KPJYCG#-3R{EnI6<v!%k~2S*{*7-rvE=qI+)sm&OwTf&{?XEkP&ewnfPt!
zT)J!_d4l=>E0_8@6(2j*K-gKw$I;DOy=#e6-gWp<W9;?d+=1>7tFfm@Jf!lt$XC3K
z<SRV@MuqGDD!N|W(UdjVa-lrf`h`YBVZYH;s5Sa>p*;FFS`pb()uqZzeYZFp^X~$M
zb&3?Re8X54%K>YMX0SKoqefus*OvCE9Ga8=Z=~J)8YuQ1-XeUZZga7b2YHrGzG61=
za33Q`cex5Rf^;`t<wAM9ZPYfc2-1&Hv|Erca3mC@p9t8W;;IqY27>f6d=s$`_^%P9
zM&_2hD7_u%!yRC5;}zhT&AP_0q&+A+BXY7YdpOrCm>q_RD4y;43KL?jeV#UW8_tfF
zgPAIM_kxWI+ekKA<y!rP3;!Q%ko6Mn2gFCc%Duw~QKMBZcMmS(b136Kw#sG6=k><i
z1l%Q~$<TVS(SkS7Bb*D~2f-}n#6!3WE_i=|8<7j%-Kl-!7Q8%=)h~FxTQSG;pvPBS
z?N0-Z$m{Ie@sb_|)hhi?x)a)@+>KZG>o3+dUypYO&WG;A&yVOmvr*~s?!wKx@$*3F
z$;5AL7!y<eTOWJvwMm-1TRRl8=`1$j;l0i8QJPs%UK-kF=uH(<&2|4#>OQ=gy0uy%
zT=Sy|YJLpwaLtcP&9hsmIinREVFXXQ5zvO5tAARmpV3V92;v^D^I3PD-og`os-Kf;
zt?fkQ;ORRN`c8%9VNe+iRqxvdqF0BTT6Yl>_WF6hYVXUYi}>4O$EGPHM7redXm_i4
zISE(X`MzS6?CpR@Efl`96uyfT{xws$RzLK3_jA{!pP>5TXP`8hQo?(Aivaf&fUgvQ
zH#Ptm(v36&ZvQ%P2hg3YOLd|fXtF>&NFaXGKvW8#L+Yw<=w)=_Q>E}}Quw#+!aG_l
ze1;T0QwsmCDZB;clXql`P-Y91IRfQ-28v?hF@E9RuOW|c6pYk8M0KO%rGoiV!7X+L
z3;YU{olWAx7fazMO5wNKg)eQf@D)<{>!tAjwF_U>V&SJq;cKMuAJ~PT<`?cg+BDyd
z*GEPwxYJZ1BMkH!-V@;wrpJ2{cb~#fAAOA#g;H{Q{XV27Tgt<CJTq+&ICn?Dc@}m4
z#fI}7?*0`&#W0*wi*PmyoJ|7f-Uv9)qt1WeXS1$YsepUDf8yr9@N<s!gu(tBPtMa%
zH1FPos#=&VF2;AvyGwAzGtqqptdD5>IrMM``r)}kZw=nkVx7as`E|Zk>iltiojak<
zo$=#yLD&0^7VCW%zGGV7E%iRopx(BKdap@duhQ+kbGz+GH%)(+D%JHDv6hKg$={EB
z?hK5txDkE@V0r`b6MC*`|0!POZN!U}3iN2l>i$7|z$O0NkSaYCt~o7Fv5yFr{S=l%
zT@n~pApZg1GueOiLDm)~Sy4vBqDdQ%x&G=%UudD{Jt{yC^??Rf`5Kg-oF5ZVkNcp!
zhYaWZ4gu_nak^oimO7@YI;JONhOXB{n|My@{Y8DfGtu{Vru)H;_`K9T!xsw2P!Mhi
zqlNspUw3U@s}0fM+-MSlTCw0Zsk`@fz?Uq+FTz!DQ@xWEYHQlR9N1Lbn)VZ$*R)4(
zF&e7d4^_<b_4EzDj@P-fx`+D*U$Fv`uY9d$gsOBWD2w8B?w0C@3uV=hUucBReOFhZ
zQL0kb1vSz>wxTNC4Jh2IQm9j6IbfvSs7hafj~apfzp5&Y4KlYgD2~y&@8x2nrG8>T
zV)avuAnolcG$2UWh;pH<m1@*B4FqYOXm}*b5J%P03ZQwVf$A{DKq^xVEkx<2I0+QU
zF+Y;)f>j6tYX!_aD7seMQP4r4$t{+LxJb8@T1@((_$WE)d=wN=V)@M*aWXZ<1;^5J
zF`V`m;5>?URI_aioT)B2kD(YI3t9QaaHh2YC!pqvfpeGxPDm5RSUOt35>S7|V43c~
zVr$44PP%zGzB<f@5-|^Sx}aDZGls$%v*a=i-BZZi#hZy*tqv9e?r;}4OX<eoSm_o3
zSF5$nAIu}BD}{{47?Oh4k^{%`zi4!*_95S7ijo0S{sjusy^_O750i0Gc7RMmHc`NZ
z)L;Q)3l)K|)auP#D7Xr&vBnl)MV(TIIrZ9PoI29QDNBRMoMH`PfKzMkMdY*j+T10j
z2Ub65Yd;0RA?p~E>7A(8VUDx%E{EjSZz;Y}j4k3;7hA}^K}_`~qYWl^rB<K1VnK(w
zTNr`y4Mka_M{DIajO)y&_!s^rQwZkFOnUL;c}q`PkeQjDX$n$r=yzrX-kDYZPFgpb
z|INAKSDY2kN>K5vbn_Jlaj7sw=SEv$qD3+kS(j}k@sFTIq-icn+AvbdlBtnp6G^))
zZX&Xy+t!AR=rUG>t7+4aKZr>Q%by{N+AHuk9$l_>fX_pUAxaF4UmYBqtV_;wbxBKS
z&Rvpq=7BC*t7=j$HmQyWYqc<!R7`x`0VkAYd;2VlXV&6m%v$1NmZiOCX0i4@z^pa1
zF@CAPNPhVPlh_BV{5F^dDCnpfqc9AF`+W3Rh4&^m9<S<)!5#c18Mov;$JmAUg5w1K
zjX!G)!>uq)xPw0nOypQ5?-?+W;~Cyd2*!TRgju684&}^Cjc%?Cl4UQX%%6d*X2W=f
z5bzY3pX03SW}HZ=56{iZlF9aRmuy>yM=aZXcodNB+6q0{qR$Uecsh_R;ujhbxdV*H
zuo$NH^~tDR<)YR)&0=c#G%HB$BOoFwK=3v)vFGabAlv1e_x~qmrdX53Q<F8xT33@;
zr+VBZe5x00lC>;XbrUuxaMW9ES+R}(6s={^oG5zhk`bSA5pSJ1GVy%k7$m;WO}uY|
zG4V{$(FL-}7|`cpfORU$4B%7IAOkLoX23#PP3B9Z6jfM{!+Brr;<F>QX|T{Y$cxWX
zaY#+LeOcOPZ8JXQlksW5#V6~4m-)m8yg@$sSv;)u2lo<8P#Fwh+l=%=GSWA=NVkrk
znRGsW4wAkmT11$1Z1*OfYIE-G4JV_#=%U;@7iY@(Ts%nmSKC%0USZJFjLbE&DX&Z`
znEnFiV2Yg0@QkgPd_Om{cv($uYp`#E;U@~ltJIe$w0EeUO$7?7PqDobFtJV_F&>M6
z+Q9;nHK~*+O6}-<wJM055(?5;O~(68F5X*40L*(f0toPa&7pK0OYEnBy7|<ISrMru
zLV*^+5-DmyF~Y2;@hi-F3X9WA(Z_>}EESiMD`H)URd~csmQncwVeFoE9%ri8D_DQM
zYi-y$ur8<W(n^xhs&8D{dasc#Vj=B+%lSMh&M~UbO&u{S)Vk*;Yu)o)t!vp-aO<*7
zMX+_h#;r@ik}DypCt%boC|d$+GdLS*zFdK3gZ6QpGdrrODG_glv-WaRur)Ph*-)OK
zb7!Pyg(ZMuevcMQ9QmJj7{-f~9mf2Q$3emoNFJpV0&Wy*^r!gZz&dV{Ae09;0N6HB
zMW9j3(U>JplVp0;_)FXzX%tF~HB=-|ytyfA3X;IVYJdMwUcw}+BoQ%kc}X(4ywoL^
zmiY|JC7aI#<g)gywr&y+!<fXwO@T5Vl75q$2HQ9zbMpe*I(|oOY$>^}uN1`qltZ<V
z!;hYQ@xO5y3tp<B(rFvv3=IL7Cu8+nU97gOhM3iCH56p^X%cFfTr;!eDY-etS(1@t
z)E<HZJQyb~><xyKgC?aWE6|A8TPzNdghCiY9C=4Fj=a;w5zC&5Il}f#L5}o0II>9j
zvo9oncHi?o)Nz?h<02yox$~ZS?&wei_h5WqOpRI~{+>-|EojkhrH}KcK|X%LeRa6N
z0rm5g8l>SkFYD~!0BFvKRT4VW@#ZVNxT>3$=lU4AMA=#>By|qGjs_<BKS9tVE3s1z
zw|e8t712RbM@r_CP*ed~`i^Uef^Db^33RNaf)i`1@G3}ze=Hec^K!MTqgeKJ+)>!R
zPIZ)aFXA}yT34aQRDn(GxKK8+3-&%_2sQGWO+{F)$&j7)la#S{FYo=-34_WTtxm}J
zSxW&k0<D6}q%9NeZ>N1^_cc)CHTxd{H?1?=2&;sTxtMO53^LO{j*kLNC+*Lt$yy*9
zxXRC`ECRz;2@x{z2^Sp8Sdihcv0wmBZ8a%l2WVELL5Q|_#>p=QLq#kZ$9ue0xhFj4
zm9Yc#J`>Rf5k!B^MYLt9$V9WHVvy*o+D@Rwb&y($%|M9^I3re8kIrU*?DF%3gokrV
zb}~W<%(8ubc}jf6NAZ){m=3oQekJ@jHksBbJEr(hv;m|BS@s=?e8-~Dl>WsC-bTnb
ztP&I2rm-a^^Y#XO6yR-0z8l!4O@d)$NWS$PdW1cPy2hmFjj-wWfrvNDkbk13u}RI3
z+h~eHWnsA)<6yXjZBo5!&`x@qQrxaBN$RFMW&~fN5szG3#v!`S#sZ?`%%a3rq3AmK
z5u%Q*aQW&As*eOwiKRP)&O1!(m6k*3ynU>(PRv;7b^>|?^xc#sgpSQG6G5htNAZM*
z&?!+{#d#Fhh0yt?qV~@cxreW~wclCC^&anWdQTa@w;Z&HL;3s$_U%D`2#w>#S-Q{0
zetg9^-W7r4pX5&w#L?sJMqjEs2?8fEh8>q!7_OiZl3IT6^*#o^Vhmp}QrzzCix>8z
zAMJs&KmN@I?gz6G{<6m>g}a9~^f&9q6u#hs-&0TA;H$lV`2IG&)4{>?L63I`{RaK8
z+xhNP>uv|%oo?No!FLhLKj3$DINemp4^i#|Z%pXApgYNt_@0~VC|rS}+S0K5KwT%9
zUoBUP!}F?@;>l{fS(QJN=Z|zBKgp;s{2@pWH!yJRQBYr}pdK0Y*A9C%%obR41lEF<
zVPRvLITPyad@1BzqVXK1@Ep|`&y}lvHXSQ?PG}j=P<CXXN*0HWIW(#ejW$NKQ4k$3
zh`L)wG+4;x@*}J>jbhNb3eLF^aCFZwVH+3taJ;Kc&+z@NxsRQ%O1mJTr0N<@M7b=f
zOQeSP1!~~HY(3sGdZ<Pb<=jIP9anmxs`TRKDqSU&u9ix#OJ3;~Rq0!rt8}eYdb(8l
zep9J9fEiN#K_)yz_;-7$lpfACWCjkOEte~VA8roepdc&=!VemRe#NgR8I5X-kE%*P
z+gzoiQt6pe>4#0F!j>{U4bLna<&GZjH=_U+i_a@4-)s)bx%i%C@r}6RdElc4ickOf
zQm=M2yHM)=wyO8;=IXsf>fIvsemp_FHL3SrRqun%)oTLLzD?@=WT0L>#4|@9RyF*h
zxf)K9;q0AK!>3IR(zv_h-Yf9)E_yFGmF|I?d*bK4)|1<grkk(CllNIq_QK6q;pckm
z$*Xa5Z~T1Fda@60?u(y~SWgZ{J^SIw$E_#(<K}Dd^C|1e0l1mM&u6VC2jb>r{QQsg
z<RIKU7(ZXKo*aUkhvMfe){`l?*^Zy^0>U2$e#Nb}?9@rzY~ED7+0dUq>~;BVs!DZ+
z@8F6%!)Nh*D40->ho%en%bV%_@EtTr_zp-4-$6r%@7#JE-SO_E(&`4dWosuX%C1zD
z#ki-3sk#rea2Nj(u9&LNQ_VJ2EOu!?3JN2OX(wL)iALp*2`q9m@Z{&#6K=ejc=E9I
zgd1-bp7@RDm&T%fIG+3_^o_@%NPkE76sPpK_eZ+H<KD7&YPQfq7h5au2$bP!@VbrF
zV<?Z;jK{aU^l0F7`h*XaqKoPvqe>_88G;)dx%4R?l=n2Qf<C9uN};yT>8p!o+hzNl
zwh-O1Zq3}#-PDTi81-SirzvmgSy1_k)lPh+_xDlJ9S25;F>#)zQPCaW0TJ;FjZi*%
zu0qZ3iDT<B(zlyJy-!ig(s=Y~`d_pHI7xKJQJ`>ZtFg_F<$&=t<p8`4wv3OO)aC^g
z24Vfu1#ROSoYqo{CDux9@zA5{eKFB*yg^XxAK!e4Vkm?oJ$sw*JWh}-wFI+~)e`k=
zYy<?;NNpdzr>qn=<x82q{sB6(2qlu%wvy)pkk$MVxX*Lpw$vDmo7EUi;@*3doKvyY
z8_5ontLiWrhis(71k%!lbkG+?z>YXAyTk?C(tI%NFPJVCd0KXvt5DN#S?j@tveqNO
zi#3y=oL;@WdwHrlN@{m)yjJWhx6y%YS-D)0ol}2dKwTfi$E>}13X+@JNU75clt7#G
z&v}Zwtd3%f#HURZdRdjkN^3f9gMBHL!RM@nVK6p7(nq@M5;HhcEU}fHszUGhBc6me
z;MNPq8r<qqV>_bIgV&Xs39xD!DJz{sIeq+4VRY7fCaojYn6A0X;W>&h{V0Yq|Bd1%
zB&IG_OMUsd3$sJzuz|x>5t0~j8uD(J&{&EkB{cXbAT+Uqy*ll)JzFE7yw?TAQYJB!
zt0x4?H7+QY9*Lp6Z$hA4>w;n_kQmB!V}}y4Fu2|Y#Zn%r1~Q|TdtsoC|45w7F@g61
z)N0KH5#T=L0%z%+RFgFijulfud>>1O?-MS3mWoQX<2c~^Ofr0*bK$eJS&WahS;+?i
zUIEOeR3B^)na}YU-$BMY_DI%&*&{;+ox`J*3_Bl0jBsB`M%dR}gjq^3CXAI}%@O87
zOBNK2ZzjX{Z5KXEU&i=YUlzc(W>-wxC*{ZhUCpzKcd0^-{pRyz8=cF++!ltfr95#L
zb1)2~`m1oIZwt)sn(c*LIn`b%mih+MON%1|tMen}VreRQfYV1*tTwS#FDu(D{UNz|
zmosF3S0(E(T>yp9&sIVu7(`Y<zSo?Z`Y}&hvv=YJarE9pTNu2?P}z~H2Oe6~-IA;!
zZ*?`KrETPfWNl-hA!{E{t_ccK%CHGG(*-yTm}MBe4ODyheg|+W;8tz?xvYxl1fJur
zOayy9XiS5B;RnfBc&Cenmg1CI$cob-3*X<i6y|>oc@W1zF&u38^|levIu72OjDz>N
zIB2PWnS-nX4sg)1o~sWJ^5bOS9&mxPw8spNwa3kYThm2;<!nNfp;4@h3_`=B7#Res
z1^8&;to)E3tGZIlM_~z;9a^j24-8uC<O^QYQ;|vD%~s&<Ohsmfv*itS->C7>6g`?H
zPXNToP-AD{{iS5N76Hn9s?4P_JX$0KH<YD4Tm8Fcs5hl!F6K!%010;0=8a;tGFq)<
zgy3Mgk#G9vl$0<r-}q%(C{#=2g+VUm4Xak0tV)q0N0HTcE`L_2Ja7&$*Unf6JGOjA
z`Q&7V$qcSQ9!MF#J++;9U>{|4=iW2P#QQHU@wPPVEZ(ePZ%({x<eCC@lWf!CDRc%+
zETAWJTbhDp&)<uN*hKSQK>QJYi=yf9(ed1wq51Dcq*G0>0^bR*#p#eNRXmf=s`vo;
zYu>yAEo7FH&Y)Dt(E?)q04EQt<kyR-ChWV(fi4S016^V*gN2OJEM#HdNn7(AS^T0e
zr}X#5Qh4eNP@ZaEU)SN+V|`N|%u?K;cEgtrBwx?;4Q44l&=4_sFkcuLB<HJ5g<NHD
z>QuF4!MwT{t)TY*WTnn<X+9x064HsNwUV@;)NXoSqsqw~tV&vvd}b&=GC&oQ>a3`T
ztZ9MDu|V@*RazKkik{He^NHTjd?6+&Kw*Iye^kq82m~v#8;i7h9L<-YGQo;eDu#c^
zXjUy#Sx{+-%6>HgB~eI#scC2wis1r}2fXAZmz+GghUrK;-02AnjTz*VfCh^j8eFyX
z4`tQ5TEd@_Jd$4&gp!+C^{iBx{3TL_a%v_o$w@{L+$ZIaL|R%%N<%BOzAU#et)$`9
zL(6aDbwut9L=YSEDuD>};M6oFs`~2*vC>_$KkYe!Lx?}o1}07#c>JL%FoN6jv6)Cj
z7^0_4(?`rF<1_)tgznE*JQe7HGdM*x20v|@KUifpY=R-J8b!^SoVQoRNE)FD!y#7e
zNNNr>_~ed9(!KFfU?dGK;u@GJ^WmI2j{EOC)M&Ja<3o)tm~TV4Lp0k*AzQpJ^)xil
z&3jjqG@ERgm(Z)>GZK}#<t=odvSGeNACqGI9o9L5E=L3K`7&Q=TMFGG2*X~R(?rl)
zF1gcBHJEOA?)Ykjq_h70_~S;#$j6O*#UjL4jYP=7&UVj{Ky<Gl%hK5#)}MuZzh>UM
zgn4`wVqP$_Wo_4ax$N=5FW)O){$(*W03>7G*4hE8n=)6@{7a*6DZL!+g)2X^<OdhS
zKU~Ejm6B80p@AYDSqu)RQ)CcX=r7Pw2K#W*E5kITrB-ZSjXgzMDc?^H{xm1k!0hAV
zy#ZH3%di|qV7NI~@}rml%BkMXJToZjQC6{+<|YaTkQyjua|J|1?MU_HtEI&&R;6am
zm^H&nSNRqH>^3@;Q0)~0#7<QC82Doub6l{<H!bi@5iZ1?+Kw~QGiJ=(8nlJDh#$P9
zeMb7|4s!WD`skylhMs>|eK<2eedg;z4=&*clV${OQVPi7G6>#kEZQ`J+HRY8U6sk@
z`-)X+9eTx&KJUNs;(l7d<?u#!b3_R{m-6J*(~n2?m#@g>{ZTrXsT4Ei0YI?ddO1f(
z;iVoOO$`+X0`G`bGn+}hR9Q822DBZKmhEA?*c&0V=zX>QOmjsbQ$`tnZX+elBvVRw
z((#|_J_LJRrcW8l?W6GS6JY}S2#;;v7X;hhrUJW36fi&cP-9?q-$kR+M$d;dM)g%e
zwYx#(H?u<19zDVY*d*^qqlk%2uX*!jLGen1A}EUrNrw|j=$jxKSO%$<nxa~2N(4{t
z<d&+vglule!wZ{eWTRzy+RNYyKjoR}2{Q!v{r!pg{s<5@w4JG{KO&)EYe+lNkz}5n
z&l$^Z28+JUAf5_<(4=!$(&pp~nyS&e>{!1mz06dH;Z~>TeI`(wwk>`O4@fV94v5Dj
zMR@I<E7Yz8TgWmE+O)saGY6*)=P^7D{i|{~R@=S4p<)@aXxsa;L!?s~BI5vjguN%P
z*8S(aB`D?dQ0_qKVfL)N0l_8&Dqf}XP*163+MeiayC<mT_==M`D(%Ik)vQWsJGbdI
zM^?s{hKr89LyL~k2l1;8p4@rW!B@QM;48iAh_ID-54v2OT7ee;Tqa*}nY<KGWk%Xc
zT<0p(*h*ZFSGiE0Ka(zxw3)z{*!vxNni54Up(&O0EP0AvIX5_scBE)=FqbDzQYOgc
zm)$&f1}!1fJgE-Af{%G_oU6iY1cnp;MLpK(7wYF0{lK2+_;yJB^Ib5&I)MH1a(j8~
zeb2>Gne)7xg{8OrKlZ)^K+dAdUv3x(av!452}-U>PZ9zofq)?gM8XhBfMAH7o}QkW
z{?gMu^pRu)1XL8X#Z`l-2&i}~=&r}EuB)r=f$NI*?H^ZHJXc+p^;)n0d%vo$uJ5R?
zdnOSjS-1J-JF0&5>eZ`PuU@^!S6N9a&kgfjEoEZp(G&ycUwv@YG#5Om;CynA!TC2I
z95ufMpHguCeUHKUv=5G&=z@n7oX_+C2h*9g!i+oU|5+cL&+!Ww<DXY>FqdWL9aek;
zp6m<>8hn{cb9g((x!|ild|y-ezOL|Ln(Gdq?fLM{Y~UXDfm1VN@GS)nv!Vg6^I#aG
z<gqpT?TOvV&Jl7_h9{LJ>mG@cTplC-a^Ii9N^!cW1vATCOcN<-i&=blQp_wqBf;?z
z#kt<D4NDurF%tv?8YHfy>L{4lvJo80ilFh~@#@2eSQ+fnMDRtY<GEnNLeST@u9X`j
z6Z=rPOa^QYLdKqtw6NbsFh~`oE%}S&%072#nSvyfV+BIwtezxi(n=`@sI=kGx4v9#
zAVAvXq#^c=wOSL+<pTr7#;(eae7!L`P#kIwELyg3aem>lrAyB28?3j=LJ7S*G(*tV
z$1vF<!cRxZhS46%m+IpK>)Z8eX&@v_s8t>&w90!tHU%uXxZSQv;X$ojtD`;fCWYV)
zq;&UUR(K<Q7u?4mAFwcQTO?+zq(^T`D3{F<!CUZ<G^>S6P3uFzCW+v!{LnT_1aIR9
zZ|9%zh=#q~KUNUzd`u$2JNU(q(_+lO<nJ6hL@~<?)dpFGu{hE+)eo4tJF{cQ09LW+
z0{yxi0mg~k^Fw%I%6)*!yUjiiqh`_$R<N>D&1ED;*uUwH4dkjRX`%RZQRgZ_lKZoe
z<RPqIjiuX+kTLVd%I)8y7!p=)lbKd-2}?KLQ}M6T(pP6|!f~V>7vovNS)Fxn5iZi|
z44sy^II}In=KJYuGYD1aSh&KH2Uk)}h*_H*$AUF&Z3fAUGI516A%jJkv1_v@`3lu*
zvs3t06beCP=rgH10dj42uU?y(u?!OzX(JU{)YAwyU#9ikXmNbJXx>8sw4&4ZAb`xZ
zTX9l96l*+ms)|=<`XMMG6P(G-b+px(rO#PD0-nt;pjXaO1YF8jSqT{L(;cHYFY`fB
z;wV_Ipg{Iyhhi)Cff7K$`COtyt{99@_hHof^GbygvdO^Md1p5Lc~8h7N9&u~@doK1
znXDpT6s4++YjiOCp#CK)OVdnXXw@e$8UIvTHN}W(6{S`^%0MvZ<0@UNHa`s1hbM7P
z-R`t%lFzXD9wH{r`56>Y)a@#u$8pse_p}aJl(+Yw4rq+kcMo<Bq|L0eEcH-tWi1q2
zS+){*r~ut(SQ6rAduxkMiIEO#JEAp)wsV4RNvLtfYdaz7jtEJ&H*^jGH@nWE-cQ##
zz;c+0aD}NHS5mab=o}ShyEL5xe2Fq~g)+gRqRd#Gqvk7A>l}4{6@`NLj5#2S&T+cd
zIaUxm_UDKN$uPX^EM)^Gjx+cwxPhzlPhvOv8lbo@xJfktESI$b-ihIsC{i&oq|!wO
zZrjcu?<H|~oV&onKD1DXUdK2f4RMt$4G*C)%*%QyMekMLJ#3g5jyJs!E^~Yj!f`A`
z;OQO_c$BQrm4IK@t0d17Ja$*#5OvWd@?f|;o?iKPMA#HV{C!h5qbu=ux-nSdg|M^a
zI=7b%G3JnDADk22?Xd9mbm0dMgP_6{WD2g5h2K5QW@*9?e26k}g)+gBDZ=kQU!hv~
zy@_8%p<uhLvO_Mo<{QFKZidN?uZ2tP)zONDmJs}FuGTLE-|K6L7J~0n4FTh2ZHQM0
zelUtOyAXUmf4rB3;L)xS4Bgm1%OD^Dag{6q52R5Xe)I+P(SI4@FQXHdBM9>zLCD(E
zxGM#Jpv6ZoQt%eCS`Bux_%=yH;`9ncc4`|pO<zCK&FM-S=DU@KOM8)q<H@2>&aKiN
zvB=FLr7>L&g3lnca0T&#t7JL&dFHn?IS7VCnYcokV9gXc_!VEFS`L1dUqzu{ysY9w
z%E5uC987TvAdWM7HF$)p_RGTW`x>KV;SW?}z;;<1<CTR!i6YG|3%|%8?<HAyY(y3o
zcS*hOz-tKxq#>@7rQyLmgw<z<sL!51WqG*TBM;fjifVQx;tg7aWRr-CttD{R5;2rh
zmbYiDm^{v4=XjOPS~n{bVdq%Y(#U?rjk>iVkfi;J6U4gt#O_e2;nF!=qA--#o1?X-
zuiZ};7*=~RILhntp2RgVN8-v|x@8*zTfChvvN0GSUU3CUh^u6geJ~G+G?9%lgEDc2
zGBJ8mME2pnLbb>~f?q|U7#UfmoD|s?I3k-89;J%xRCV^r48<?PPxUoUjf~(l)i?}<
ztc?>QJmM8|dNzDd^WjrtCOAXk!%*oCUy8rR{3z1wQ_P|K@m`u@j)_b$<ev&=U&|;k
z!{91;hB=7k&Ib9_I`!4lyEndezG7SH$2Vm2XV^%QI`}Zd>iSL*JmM<&D$-%<#yFjM
zEH%os#Z8_MeV5kxL`8di0q<20>jM$ZZ3$itix)0DJHK%0;zj55b?4SV8@D@(#R@w%
zjBjwf8t(T@oa`?=zQLV0muNezXWk4RrBy>OE)8sR3H}!!C%7j3f*(4r3BTm8e#H+W
zju5}*N56@E^dWxqfBX~g&~TiXiB@1H3VzEke%#~G@DO(IJ2S$QuoNe}yi@R{n4*cM
z@{p{21ol2#6IM>qo&9$07I7_Y*K_Gnu+p9#pL8XMcKj!HX$Xbq1N?{O-_SX{P%TbU
z7%>AAds=Wmn-73!X#`O)PU9*L(rQ5_$2UD2q+<%DI{A~+;d45BR4vlMqDe`ulO(d3
z=K#1|15~`I5h^OuITtHxDQtS?oz7uQjshpUk+0f{*~OxD8+s!+`8)-D-iBl0u}JHL
z1UDa`J?BN+)7fvrM&CJ*t!K`?L+HRT#R3?zZwEXi?56Wbu<vMM*loUCw7F!r*~4Cj
z%SxZ<J3|XRU$-vQLF2W+&Y#?0r7RkMzqj#@{<*9t{c|E&{ge)WhWd$@)3r}blzQ@n
zQipp|uZY<-x|rw7H0=`;D$2wa%EZ(v%8cDL8uS&awa?4=RTPT(C947?wa>e?_G$Qy
zq#h^G@A;fhuvMpDCnq~nUWW?}KJ?<;u2Hhma-+T@)=x#v9l<C!*00}IeB=xG1@y^w
zMLx_TS;;5$TNarHrw0@Kw;x}FUrXbeCcglDExtncVosSNzFpbyJ;#UdI)(3gg%49o
zclbKT?ds1>VuP|jrzaAVqPJ#4{9+&Cmng&?g&31Ymx#}pAipe{MkvOKm^e6ZF!?-a
z&PH=A!`j{1+HkM04SLn{desI@4MrOtKkh<riBeUp&_u|(ztDxqp70oY22*k`Emw1q
z<?7Btld|{@w<%`Q2pwaT%h1m;Ca89}N?yD|CA?~*EMl!L;ftz->kLb2>c;uz=aZvZ
zSkwY8ix1zAU)8r#oSWnj&ABG!)TP;qPTtk(xe$7T<FC{xRXQz7WtLjHArE;{cz|5r
zCW=rqOb?5V@ixg2ngquvt2`avtWhW(Ui%hv)zC|y9#TEBNx8{-J6EFD0?M7#NA^h^
z@D$|~SV?SBaGM4pqY0G4v1}SME2d+g$X(0$DRN96@iFJjh0k%GkBrhag*@UTri6T`
zOtEti^r&x$vRc+~WPrmcn)%h`K~qx7sqp3m>*qVl)yX{KH>ws>jH7zHI^0L$gGH30
zD&-)?;RXI4F_m&V$LL%GMHb}NBa!n&ozl7Mld19q{ChpC?I)p+!fxYUb64I?ZADe9
zo%q89gPntU|9G+OwKY8^>u^<dgxPoGC1$V2JNq^09xcRekFqdyL6zcHbA)q#hmLDl
zmi2R~nS-6v=*S2~yJ?6Bnf<xh1@6=$Mc|@{K!w>Nw1T~2IY}t(3yq1_Z$9rF$UP|+
zW_&n&+!}@8E_qVAt?6&&d0H&$J+HfZ4+1;NicgGIEFVL7>`bUx9%s0?tD229T1<P5
z_1dISYx>FXS54)7!5%*v{tVGS@AXp2S<GG3(_+q<2$Gj}r;9nnSlY#Wn!eD(qyhyS
zS5UBVB~_`Ih29r<K1oyBzsN73OkAN%Og*B^*oEHLe1+<T-q-n66bf->=zOVeaB`t{
zhhFI6xcFl73+!s2ffQU!U1G}N^W!*ouC^mW!L)*;Qa^`AW0AAMCyNZu3Vy>4^e>)%
z>m%9k_yzRC?-j`)fwGcJEbfS^+W$59V>WR6HK^`1CfT1~0JsDA3jGNoH5G6NaZ!HC
z9pVG07Z`^saFAsNxUA=S6V4;-Ihi%hnlZb6r|b_M86~+~eY~F!D}>s}J|*%`rbJFN
zKj3V|mskqyrIk{|k4_rFg8^neXg9dBmPS^x<_<g3g<maJU%isGlkN^xbnZ8n)5YAx
zB;1cVElZ_XYjH?Zu89Mo(tl{8pe2but}JdMj+r>LJSdHQ6ar1`=H*s~VbKs%b1lUy
zrnQlDYsS7{xjexE(<nG<K53(G@xrC~g=e3A4sG<w$TodKeR+##{lIR2HFwGI60O2{
zmkfy^&5q$%EaZSnzqGASzlnj(VonjqECObbH%@`APlXz7jK(Yzg1;u%k+^38FARe7
z_$LfUX6yg*?D3h#gy+})1LY)uZ5gIu#ChhWnu_7ZMUwN(S)#L+c7>sHI4wb2#YU?@
z3s1QsV0ZRf-nx&$@-hR995Xm5bgrnl=cNF<Sk_Dz%a=P9b)J?<$(KxfNyxNX`H1g}
zQ0EpdO&<|kr^nJ(Ajg*~P=IjCs!%8rHfd2AQ3Bu&4bbB4;7)pA?(PMz<~LvC-hD0K
zz01A(I=*|id-oo`d#`)<^?dgY_TAtc)PzBMi|vj4-hK9agTb5l?wj4aZ{fRdb??57
z@4nr=`wqVQPWSG+`0oAgT@G6qJmB8NIsbRNcd`He9``N}%HVI@yYJ<@?+fqR1M=;H
z3BHy*806lx?Dk_s-0_%Rf0^p__j7$Q2ik(a4Kaj66utg;?p^fy2i?2q^$)ps(d&Qj
z-bJr}*uI<C>;GWCCwu)P?p^fyN8P*V^?!8lqSrs>-bJr}+`Wrl|Ac!Nz5Y+`UG(}t
zyLZv+|Ki?7um5X!*Y5RiqSr4yDJ2mpWu8R9*PgtqYspLMh<7#KO<zMp`@m8LS6Is6
zN~{*<#O0M-#O#>1K7yKpLUDycp|FI7#+<W0$5*IcL@(u6Q79&OV_lPF5q+bxh_+&`
z+4H<Ubr(5+!s}yeBIgA;Ta!62(XKGcnp{Dv4N(rqX6-6%;HLN&eV6-)qo(m-qaqIG
z?X1K}@b$YY8@{W3_@2ctfFrgje3;9-!)Gt&w`K!Z@PX4S=4}cb=6?fRXLH6A{L{Q*
zVOejtx)g1-ShRK?Hjmdo_7!}aQd5mffTi7;OdGkh3t|$2mVK-M{R(`FD+n2AS_Uhm
z3K=nuHoubW2SUy)&l&kyWKtKg`%Do);vSWEO%#edD4Ay3(?auF&kGcndL-c$O0k9o
zNW$ZJ!l<}NdUuGw7F&OvssE}Jo8-XOKs?EM&G^WEV0>>sIFp*vpad=>oCJ&mVX8|r
z)TO$q1E(lqE2L{bMd_(b6h^D{AqvBy)>mQH0YQ07_I9EJDUN4u_wfw0Gw=+qz%$U(
z_VG;Et8AqHi{P2}M|mcbNBI<j-r2t4!|m~L=^q+=i@{)7j4NIbcaC1oFN9_n-b~&t
zEhknCyo2zXTvYz2D(9uVOp;cm^iB3-Nx-eF<p4)-k&6xL;W`B{uGL#(w2CEj(uQzj
zTjmEAaxph&d3`8YzHJ>@i8wIwmKtfR;hTB&&DL#3?wasjwpFQjX;-|i5acL<At!|5
z2+NHM!5A)6Tqd?YsrSkAhhpQ(%r4bNhy(IxTYqjF?|yT9X1b=2i<AlqH7qo6a1I@N
zpPwL5rXdxMS#El7@EEO*L@&5js~fo!;?{W2WAD}0?wfaqYdoxPUO;yvTg?IQb_Xxy
zN6E)lu@Jh|`rKa2$SvhFl2@?71H^Q;kF~ME+9%p)Y^vS{SftaCImFmh4LXb>5_0>_
zOg5LqO$xzat7PPDt8X&~-%bzfHcu?@9lz)pAK}o-BZF>t?<R|3mb~z@-DK6v=_^0T
z3`|?N!nB1eIc>%4o_@zS6+)7rOkAN%$P`g#?C$9ge1+<T-w*j!6pAsDb$*q*r?)@^
zkkS*15t%eELGT#52&IU_>JEFYlv9(_PHd413;x6ThU}2tG0KZ5#Zp0j$qn)^|9<Tw
zkzT?4Mv(|Z%^*?dlNnbqNg?d_*l@BJ0@IVh^&}ayad|Er|3rR-^f!cA@W&|q<p$sy
z-T>S;!ST!qN49W+o94d4s_nbn3)Y5QX9;EwQ;nH2TTDR(vv={T#qlAjIi1`j<j}xy
zW$ofw`qqK#^+!)uA05a?Lb}|<`<}6h$AxA)o41}4j+Jv4at_Cks3L0xS*UWWd1Ee&
zF1);NQ|Kso;hM`gairpN&dyOTzBZ+$NIIZLVTYHH$KA5C7k7I=Z*qt{up6XrRR-jS
z<N~H|xcet#slC9XNX=kMxO&tR^nn{|hs55UfDZK_s~?FwfXNyZjK-hD$9#?=n%F4#
zH-2b42L_*_d%?f+gZ&rT&VQfdDD<D^$Ki3jP)mA<Z+^ynVykL$_wQ^-x8Sq<;>Srf
zi{Et;toaB%ANzzr!*h1$C{VFP^q&w}W`*ZpXO`ve%iWWYzA1U1qYa~3B7mQC4qP0Q
zrAAhdCZqbpP$+G7QP>zBQ=L;1>J_gT6@pr#!ZU?Vc4l&gpd>0h92E|ihsYu!t9(jA
zmElYAl`avLE{Uo1Awgs{L3cze4V8F*&8A^zKUU9$Efwv2fiZ=~Gnk|BcE8a9CQFP;
zU~hDU@20D}5YSLfaRn_BSIO$`DJ-bd_BkNQQ6{cXCS-bwx;xWXs8)Ap@vA5lqRi0c
zQcaN4bt>#^<rauUYSIM9S}fbzl#tN(7tsSAn`*m5Cdm|o$G$7nU@fSEW!zXt6^@Bs
ze6Ejt5U%Kx^A!13@Ksjw$viUW^EV<+dT3CNfpdWmjuLOd(-oYRJ;34lzz^p_9~>q8
zf>jF6>gj<Kv!}Pl2S<s;V6B1!d1!Rj<L5BcdZvLr1mtSuBwmd?J`}$_vA}o=6V_T_
zqzcf~RIt7rCGJ`w%;*r*7zEY|xRRQLV034#u=4Vaa)EKJ`e<?YYlY2&q)x8cMP5Xd
zQ)J^st67(qsY%88ru9@29Z@P)d6|a{rks^}g7TU-`nHKB1Jax8vqNNfT~)8`s8`#<
zNJ^<lTRZEG?agh4LY|grZ1ycL7imY>N?M$ls8&j1sX^b4hrWV@gsWmuWR;<0=uJLN
zBvPpsOSGRf%#d1jWWVD#L)_#jRvRuhhMW1EjlM!=S~_@CN4a!fO4KRpK`A5kstiyv
zQf!WqiI+^%oL7%PkJC9W37YR*gL6~Y5m`oS<>4Dt20nf^TDYz?f6)!OGiV;Xrd8h6
zx{g9NG%4)=4U|-S4t?1w(!~V5bwh4`_%`KYzvddNm>p&@*NnI68@^?H6ZwS83~LtU
zF9;E!a#>`*s^^+(sN;1jaGy?V(b=%&Xz;cPGKtE|QGK~XUtUAVw!sD}Uqk!0p`_t5
zcQvT3OzU)<jyK*oOD!D*E?4U%3T8om60+AxJD}V+!ZJp-bxO{PlI#O$IDsaslce3o
z>;XeQA8(t?t+_UzJDZtVajQtv&J3sAQg@UaO+Ma6pBBlQUmvZvo7G9hWz_l9@pKx3
zHwR?jZwR*^WCk7+X%1)h>bZq9a;nu_vE8Z@5s9^{R45(3EAJ{*+jMXcEy7SjJdp)E
zD-`o){pH~(;{$eti2g^MO5S~(8<KLpHz|8N<q4h?!x96qiIS?7?KDa%<Ey9$Cf<ct
zU$&m0_Z12%i;qqKG%-9fu;?82pgnWx(xrL&v&+@E9AxV&qhl?m%K(i4akXHmPWvDO
z;j&x?C4QNASxy4RU6%Lj^^R{@{ytj9a{P5C)$sdyb^8H+=AN(%{+1v8UF;+5c6>1Q
z5u)JyJ%8#hyFbh~|A9Y>!ocGHBmC?j9jC`$mCqInre}H=Okx6%i$&$k(>a=DT%1pC
zUK24~G$#4vZWI`<5|)9Faf5u8ftW=r<1iMj3#0WvRu`?$677GPw-2Ua(b{MxZ_>Dn
zR`G87q7|9~7BjfQS`b%KKaSa>nP80}ZP5yq0%hU~WkR_SWybE&?C=$;7p*(_RTPSW
zZYc7pI+t9uzFX|kMD||DNiUp5^m-F}GPK^DC_{(~X9KazlDKq*lU{6`l77;1=x{bU
zvy?|hDkYx8tf~ltc0Qq6#F#g@DtH+;-PYgZ(pkRTM@%*TgI6eGV!&r5rt8cXs8Juh
zDjUXE`!MR2+iMg?2n7S<<7ee|Z<MNX<#r>f;C=ga@UgN`jP~i^IgXgHPX`~Vl!yPk
zb?`rx3ce={Bn_5dd&xjz`?J-=W)Ad`4>Jl_NN|;GAvtJpC3cDM)koD=^9&P7s`nac
zB6rcg%q4Oev(B@OITF>m>zJo%71*<mk(pKYL<DWA{CAEFwW9*o)pV*5yO~Q6TEt$H
z$Su_q49z8e6DMI!a!g2#$U(s~G(~)i66slXa(5t<($VP?AdD5soM`3oE0VDDPZ8^a
zbzTV0x}baSE%12q!oc?4QZJ{k3@}3UsDZ+bds6?3Ss8qU2TIz?03%6{DU^vZCCZFl
z8GOQ5s9qWT6TgZ=F?0+?G0V!}3cWHAu`jv~ePIs~6tcwf#`-D)^y}|m_cdAT@83{O
z#@NW(WLJM5#EL!mRyKU!_Tf{*DEN-ThvCy5zLdZU-;W~A<R%Swpnv3#_tIo!XB4m1
zchuQ{DUI*@w+IO4BU~lVM~4p5hZpG&4^kigfes0V9zNYnNt--V($Qg!?yU4{t}w=V
zIO_C*%!ZC-jlG7uN^}#-c{Z*^f?yqqQy4|E#1<(I+34=*nfCWk@_2nWcAh%UE1whQ
zspAb2MW(9%(_{*pkYqoT2_br(s%9u0#%%SwX#JUHt4&^NII~svDWq$15F}4nb_!|p
z{q#8tqX_d0t}qqgO3Iy>IqN_kNojKyh8oJm70Se*6J^HES%>)w)pJ%KzluUJu(AqB
zY0WCxbCz||RD_aPMq5LPNzTM>3LOTM11LOIS{!fk$P#~AGuUYi*FS^J^btVMV6zkf
zFl4e4AYleOBOAUsK74uxo2&3)NOgxVWd_Sfk!GL44(5;d(hMfE3+55DQT$e(LwjFx
z4qL}iFo)ra=djLED`^J{@8eB<M92y}6FAQNDCCvAeIPtmgGb11S48FyzQ-QCa>QBU
zSdS$5?xT#CaTzgFRFHhWCHM(dV4i$Y$6JE`cAlt1Ey1JmiJDeC!8n4alaI}<R=xmz
zmL-5=J@CI=;}7Q!mfbkR68u(HDhcxk`qL+}WxY~x2d|<o3UFw`#B)1&fRMB1YiYt1
z=csW7Ja-#?@;Z7htG=6GP4AP}^D8Lh4&5f-Vm5daedYGWVDMJwF7J*F2Jdk0LQe?p
zckbG1KuFa0(p!Tn*iCr%1Gww%(|s`Z-4Dm!{it=fC+>KJ`iOdi%NM=z&wRy{>@2Z7
zNR3fOCW{Np)k=9JKU^Ly*GQ(f%EN_Lz0e#bg;Oe*>^_Soo35l5W6G+TIHKwYRn?39
zRqJCZbOx?Uu?V`YmkBNE{gkSAuD@O!4<4a|i`~}8-7V^SNY%H%U*AY&7ZF$3LaaIs
zKe`8q*0))-eMoexp8Y-_)*0NX2S~_?F$!Z366R~FfnQgZy~e0a60C=k^(6T%Rq|a%
z$;SD+!Qk8U3J+>2Jig0cWUJGXTfbzzHb!n8&-wbGrr`=Q=RJ%?&Lt<(b6NFC{A#+~
zI)z_B8Tab;c;r^_27dK4`Z6lJW;u7?$gkeVKTmU>%veb`-^5SmI8WdW@D_eD-+6)=
z>8<=E?>u=M-+Vj&g!0UiM!`GyQQ~I3MQ14%&ZZg!ozLMbNTk%^S#Q^d^PGo?T$o5_
zFox$~q6mr%ljUJu!8;kTnA<(WB)C6&<79=Jp;xL3ABa|ng;cPLUNIPYEq~DiLqC+j
z(9SM8nDL(&i)82~88IY7x7ePM&d}I2M;U*k+v8(s_Kpdzqry#wR=zR8bDby3H72;(
zc>)Gy&zRr^&XWv|F~M!lHy8-)787*j6N{;ZEi2aIY%z3%$*>m=8hupL$j4-A<2!SO
zn#r)SjS4>=txz!8U9yF;r|+RZdtoxTU&KhwCv$m_nz#a%FEAF#WT(<|S@miBYC4nQ
zI04G|7u_Bolkv%@;0!9<WHQ#i2ZOoJU35Gjh6?(fCtxr>2^9=DPd>#r`6yJd*m(lS
z`EHLw@c`Fn3jBR8r6y2-&gT&dAvNbDDH%94WSoDngl2F)7a)!eJP^?@WOZm7s>7K8
z?vl>T)at}0+3l>JEDN9`rs^@W9>>`q^~kW>k>#Q|(Uo}2DXB#f$qYY!Fxlt0SM8}l
z44R8&PfO0dgelg8WX3_8m_c)hiX4qAVEG1Pk%Q(idM>IaTBeT}Bn3qQU(${7jF`dT
zD5@YjT8?$@@@N?hPH^t>NEr-HmUpK|a3cFBTJaU#3XkCIQc7!~c?Lljs@d0~m5KhJ
zBU`6Bp@+^#=GU0cceb$5>i=XQvhzPi%%<u!JO7Ta?N2Lg`WHx+)|LzGTcM=?r<>&E
zf(Ip+hlCa~^Ha`U!37UFcLf)G&bj+V`g<_=lDs=TynziWwC!8EZ6W08S>R~iz-|*N
z`*yT4!5iO|ty6u}18-o<GRhmBgDy9`2F`DXeEKuZ9-Y$=64@1w`O7FI%sAW1<nn0q
zEaDa8I1UO9Eo$Q6K`{)5OinI?d?o_^Ppiumdy9@klsWYjMUWxia0(kkuJXcHok9oh
zsBrcfa&nix*mSC*Z(S=#A^6G5oHDH#A(?aAE)y>BS0aLQsmx(<*d?BtDPlB-XS(|?
z=Ug=KA1_bVYs1@;;xE%VZuTBueD+x;>dVU`<grwv3Q82cdAagDAFZ^;+C%wLeSBbj
zWwKqYt#6l#&GnV`00^QY4@tA4JFq%O;x%^%>k-Wxo<SCTpI+_x*!B<bvG&{!xgN^O
z5BZ@ywr!2H;71HE+nP)^SeX+#$Ak@z>_`2WOSLD<X&!x@K2f=`3H6g`r5J_LKs<lS
zjFde{fbS(N+MFzo5Jd;`yK;88gk~lod4lykkC_uhbY4KxMwIb$O-LWha#G-6qnrUl
zmJ@TsmCjwB8;}P$xY~IF7X1Z3zQ%c?vH%AK=ZQ)J91J;6QgQ$@?LAq4;}&gBW=7EH
zcbZ0Fvvsc8cyXVUKR2#QC)pyZ{KKBD(#fBQDxbjI<m_$eW`+&FiF&}^kcS>%#rC>I
zy<^EArI>X?94BugknSIY^mbKei1g)=JG&!ydQ7vMZCqQsNm`Xa?nJcXK;4cIv`{79
zNI+Q)mjn9E^k*;23}ktY(IyV0vx{N6^>-;{;wf61WK|}tbn9Gbwumm%kZnCl3z)Wa
z>#6dyGlLuJ&i;q$?(~V$K6*87_J;}VaDM2-U4aGch?obEv@=j#f<AC$%vX@LPvi&j
z8P|`BeFSNHG(AeN`5wcMy3#L{wCx-tC>oPp{a7y5)-SYJnnvl{IZ8l?BvU_r&jL`g
zxO1YUK{Tg2#KnmkhvF|yiBN+4$pmyrKT`8m(sbZHIR^KijkxX{hE6e5&r@`BLYn_&
zsGi5L8*<!qadmyPB7HIr+ClH2SLZSrgz}kny4ZZAz4qjC!;oH-Z$~BJ$tN!x)%uXq
zjgHkRHUI|!qu6+K5P3&=S6CR<v9wZDsd~h*<b16{!Bb160sJ|zbkWkY7m#@p3(f_y
zzy(y`f?8PM0=Wq;7RI^LDMrAx#V*&EOB${6c;&1CZtViK94}l#yXZ~tI&%hp<xVuG
zX%19^;AT=7ZM_W9!DiB{G3S4wwWowv#jG-mzm173b~;!8R6OJ<Cnj>()A(WR67&rI
zwyP;ZDNdckKnr6e&gR1Gnc**EB;Ljiiy4WBY^+a=Rf-qM!8lJ1#&h_C?gnG$g`zOZ
zi<R3*u>d%bac*O)PBU<mf&=6>mT624FQ9N1Iu#aO@pxNbtc|vd6qFAcv2q(4<pnJn
zcrV|#itH>1$dO=dX^+)wqsx{Jh}!vBe!8l4lx(v*2<Gd~_Y4A=Q#aOMcm{_)J)3e-
zMw4Uhplw&-Xe^GRgu)&cPFHV=OCz&@5$5?P+#(N)7~q==%_s2-rD!}}iz^~Vn&ULB
zB4(OJT%A2wro~LtIkG)5Og6g0Sh?AxY;1xzJ12xt+^3j4&NRT>F1CiIvZm@B5!Pl%
zhqH7Yia}iQAUmY9|4@_89UXYqqQxSpVVB)u`H{?upGE^Gvkd64iZMzH&JyyslMBAH
z*!3=TSOiEUgEE%Iv@?D-7!1ynE`A{+Vjc)CaPG479}F&Z?t<@wHO}3O>F>c{opYCS
z4{7K4u(p&ma6R<CvQ3J-F`&f?O^dM7X<_ofNdkk)E-)&a1||-W!38C+bV?3}h`zm8
z0Fb#TI*|m9>p3dDge#T2cnG=Ps-;yrBf`xb1+3Ai)qqQ4#3QMK=tkqw&RxNX$2oTe
zBcABoeG>g0WjWhDv&&&^8E`ba!LJ-GzDT!tdf81WJpr7=QQ5j^Wx`rCUs6JKShm4r
z2`!vEh+&j3u<sG}`*8&z&omh2OY~gTo6aa-<)To)rMe|vM)@Xv6=jrfId=u4d`I4$
zN;hlD|3KlrOydo^Sy28*vO-0H9w`4`TzCxS1FjyF#}x>7l|lI`Jy-RnQ@*ZRvQf9h
zOZld3T-apjL%~kxuAuyL<lX6^{7nk)6&h~{Ls0&CvO-0H9w>iO!bE*9R}aeL3Y6b!
zQ2q_{T-BRS`8TPST&-K;rTp7u<H9Bj8o$%IE5zUf^6vCd{(TDX7L7NAAt?U=S)rmp
z50rnH3y<M><gEwgaRthc8kGMs8Nx-q!B^NOo=$uutVbCI-54+NAC|WFu+_If8Tz(!
zS9HmDox7q-zVF=q5&b<F{MfnsQ{4TTbN7F6_fh9AD^7#KFXi2-bhVS49@Hsn>V!_T
z+4|I)ADzVVD7iE(C2Nt4tnetgY?N#{gZD{IRI<s89u=qSxcn%GcAiMqz%aM^+D4g9
zA%I<v+)1wk*tPtjT#2Win%#Ue8N<S&M4EOgyKg0h*CRiOT5-IbYqm#6Dc%!>Cal#d
z3g1|HxLu_sI7Q&AmCNjFv6H+Q=ykfG4r}CdeVghO%9tJ&8{_SXe$IhFQCsLW%8cFS
z?7X?6lL06!IyYHw=SuWiK*6N?Xaxx<qMQP+tSGH^7-x`-YX}FZD$X@1a%F`AB63Zw
zTpkvIIdkE<kuUEkR_U684RX{$_7W+RAps#6`w+1bd5+izBn;rs9v}Rb42Ee3TI6Vn
zy5i<oy<HvdD^{Czu3@5DEKv;<W0Ak-sEK8cTT{z5i#xdWM90$hi8=+`q(n+S>%&P-
zf4f$QisnO~Wa%8RC~t?%=iMG;*<OZN7oo&~6ba2|;Y|n;_a<aWG?E~MNM}DgXa{6S
z=QOhJ6kDyv9C!=$=VlkUABv?hPT*UZJ=YGp(SsOJ0;Y4EOFHfj5vxAfuE{nXBByh_
zTNn69RyIB@l_@C{ZJ1$ns2g>eaw+PU@~WvNh~ja?CcAzN1Wx*uF)tLHxi}mwkFZ9$
z)o#>;$f?$w7!cVy-C8|-m0K;OrOXd}{=gVTP912pYjj#{fWzz#tiN<E1+fkP(`=Lm
zgx~N0MzZ)7hdL93=RYasi}dr`I6(?gjVm0B#+6hd;&KJAr*&280dD7%DGJ3Y)3DH(
zxZ;=j3f0b=I%O1&or*$(cL{UZ{q&#PL<+Ou-E<w?PSZ*79^7RsU*|EHGU{q>`#%#z
z@@|($jN^g~Zt;;w4MlOX6?DhRR)a*3lax{Nw0^dfQH3b|WvIV>DWl?ZBK8=t?AC;V
z$$DXMDfdFi{xNYQiaz#t4pM{tagaJi><s$~)nW%Ht5N73q@Tz&R%WsDemic&!t}rq
zi=0FqD>w~@86v_AhKggqD>x5%f6h208-|vU=XB)@(SyqJ!9;^JzXJQ2jMBlu`Y3K&
zktnW@4wE1zf5vltBz2BhgQPd|M~0a4gpYF1REKO61Lr0m9Ou|I;M}~&;5^R<$2pA+
zIM3f>a9-ep;~drooLhQ;!&`k$vXdAs;DtUo+S-hB+<=2~+}U}DwSa(kKa7oo;Kf|A
ze+usS@Ht1nfe%N&yTeE9uhq@q_H5w($_GyC2RJ<ra5z0~fa^S%Eu~~qM#(*;e>*eI
z6jKLE_23AZQF28ji{$Po$%RVL`Lda`h1_mL00_gulXw3`)(hkepjXC?EFFD~@Z<$K
z7^tr=3s0jU+Y<dj>v0$sNb#prB9kyU3?bA6msKBXmK!^0G0!oBII1<7LI%t2#<~p~
zbBh-)Ug#1lGqqAD=pm*_C~0FR=vVpp&xsfS{zJq7!GE!VonGTBR8P<dA%H>=Lcm~Z
z&uoHq!M%Edwqk@Q&D4~Gnv&A1-1dl3PLuj*aeTZODIgkU+?>FRlg0;_=!fcbhCepD
zmWm$`yfez8anterJ_0)N1wg<D_@nFu<mot!Ef6wDjC6mu4~mxV2tNQQ@9bXL($ZZ=
zBnaLcL#`N%@AqNULUbi>3J7vJ82@6Tq<iJ$2cuM#OQ(Zr-R+7>=4wMyrW}JK(J0Ql
zT_h!3?o#R42;1)JNiqbJGuDMDu|Col9*fKIu{h=4sPm{4hnfR)5^iyR;j*Pm&NPni
zWjP4fu2)L~EUdXUSGMLwqMDgA`5a$~e0|u8JDIDs$0sI->!sYxnY#lvN~DU`JDH5(
z$cYiHfA8Z;Cw2w660s{%MC(U<g=*2NGe97iMRv&m(fVhd0phb!4v#a=eBMV)E!%Vk
z2t>W;j+lmQ`xr|d_8ASnoDJhweHgWD(-|NTD8s<`_>pbjj8av~wtX2OWU8)x86YT^
z9Eh+l1B6`l&<5zAaTc~evkVZaO2vm+ex)cC5BsJACsq-Z3dAZ(Q7XReD^x2L2wsFj
z5xmHlW3sH*?(TBE7FmKhCRS?+X2b&0_otK#`O)ZP61Q6Wg^zrn;z=JHc8F^P^8J!O
z%1%BQPmQ2PA&bNqPk-fu<HS7zoL~0<C&oM&1LrqBI8JyZ!1=%FffKV<{H+g;)~9qX
zWW-Q1I_vSHPyLa}W!qS#KD93w@|0GO-(tdsqc~>uxF4Toe<LpxzQdd0_rY58CjR@)
z^g}^!5mgI91y>k0h_96*=nwQ2ss%k_ZlTaGIfjxfg8l+6=zGc%V(noa#i(r_j9bke
z<7=E=%^?&Q8i!C^#>mdlIK7&a`9My{hVLXFJ}r3=y$kpdy{kKXDbduQ97QTdvut*%
zYO+6noWilE`r79N1w;D~6fA{fXZi}&9INx%AU;@^MB58_Z3d#$kK^M-z9wrv)_H9Z
zHmtiQNBH=xZ1~Rc;nRGq^V)p72O*NY({pYVX?8w7jXzG|;}yR4A!ZpQFRsAHh+dY$
z$E$pWYCc}gucFXLdLY`fwB*%VFlA^>+7gMV*h>?4IfPKwk+*_PjMOifuk^KDGxSxe
z?LY3m?Gc9Fk`3QhA3n{{&sO*lw96Q9j~|tAI7(Hi5^i#~4-n}rbZ*`kk1veJ=j=vI
z%Vxv{+$kwSZN$gpPE0n)AH-x!5o#4*p<1XRNE-@8kTzrBWYD-eQ??r<3$^~-aJkyD
zc7JI{c;|3|jv7?zrU<3mwb+c+QZD|MB^FHk*F{H2+(dtak9;bkP_Uhj%ZNDv@!Ps1
zU%-KNOy887v*CNb51$k44c&`iZ&SqgqHOqX^Wjs`h=S`C{SX<hJA9qvcJ=2bkszu+
zx1&<tSr}>8O4btPmDv#A;X~|1hyz6tA+AfrXV6(0_A{aIrsW2?#Scj~*PM;!SZ%m3
zTN~c&YlEIG5giU~Ky)~x4UeDM^6gQo%GvTTXSQUgqrDe&E@8iYnqtoFi79TBDVZ|G
zz0=2-2oDIo9amsX1PT<48SBdQ*S<pa6!$m$DhfsLJcHY_OmXi^(c8m}sm5$X<GDCK
zQRbs&*4-}D`RSYaRII$<(@X>Zir_Oos_L2Vvx=&So!1>zBQxI@v*G))51*d-zM}9U
zwB8i)eIpyb|L4P}XTFCOK7`!s4qt|u?+4is|Hy|}&wM{th!JeBOT@j*e80)ohTr<y
zpl80{sWzO#$H|N~Jbq@r{W+qSEejH6K05HY_a?qQ9Bp+^%zXDVH>J#c2QsD8_6iW`
z5HlaHFi9cip<v9|nXk`RsGj)_=T}iEqWKxzo@M5{Az|iAK5b*nc*3w%sa0AP^6V{^
zN^Ls(TciWQ`W?je^BJ{NEWhB?DBH#@-f}+jXkjo@kq1Hk4DvjFgu(16Ri!X^x}yf~
z3#93Y5OU3dx6;(SRS)J4NRc6Pef*37jgTR@!f;2Z#uOPc;44(ikcIpz3PmtPV~}K#
zAv?7U!G5D{?N%`nDS*USonKgD%LBSa(q>+NZoIe)8Afb3{(g$|FhQ$mKBI0wu@=y4
zxe4}!7biI`_7O#!!`CUIAmU+nM2VQgFU^MUG9SJT`~o=PnF=3*Aa;k3`qeluwJ{sG
zD}3Oz*u7GLL-fS%z;%u{j~VsnhMFr7P|<{3h(Z=Jt@FV?5ySCM9glR6NRs*7>=a2-
z^mUOFJQ9N*!6QW%#cCtNzCyJmK{QDeifEEu_WO;FB(by*G9y!L)x|QxYD&uZJe~a|
zSyEsdT%lHPjI+?;&^HA-aIh;p<txLLD3QX)fh~!$Bg$cM5@nZ<I9j4?<QPIA4#G=z
zN1TX6xjq}d=lbwD5h;NW5h=UFXG@fuvVnV^51f`L2x<v%2x@77>uk<;0<+fvnHC+x
zZuJYIt(HRRVe@29F*h-ZusOLM$XeCnq*Y{rtdT1)|A!(jQGldIeWK21V=6Uq6UQIL
z8J|J96+$N`$LTO*dt8S+M_lE#6nR+(TjsNotIOpH5oMV|vy8P`6V2rV1NLd3#fv%b
z)!ApCb55^>O7&nFV%nrgt{3`P!3i}ARzRpx!3wdG>o#AZT5=)$C<;aRQG>s-=s+b~
za#?~)NUUXZQ$&{7Fenr$30iTK`YRMtRtN)2I=v>!YH`x(E*}B3bV4*z5CG9gyCXnE
zI^CNM-y3}Rv;aT=Q{Y1Y)9&!4_+`C0ic}0*+3%@b{UZK2g{$A{Yo8Mr741V@)D*6M
zhp$l0)rggfLJ=#qOQKPJg9fo+f3DhqX76Kbi$mWRrE46Ae!$ll%>f9cipC(2YL|^k
z_GtQ06lr!2cpHD5!T}%lwa<yqiuNHsYYGQ^)K{qH0K{xXp@`YqCDBedIDn$gg=@HF
z6zHCIJ64Fa4W?fH$tW%3di_(r2IyW#5LYw+L0r3RKyt5tCW<tBum1yooYL!`^R>^3
z4U6_6Hf&0-f6-T{?sdeAMWKin+a=NF8@(>w*TUX*oH<r3w8s3yQJTi}{(txyqI(~K
zWYLf<jxj&Ww4B`g-;E;8-us{Dk5hX8d%pG|DlJy=xI+67pEjlUf9NYz_dWvFqEG~^
z?XvgpaC(0+4fdgOtJqw@IW973GiIny4#kA&Lz#W5jlx8IXSuP0PM1>zCJ{Gsg)g9_
zDP?{i?Y%fY`=gI!D#&E;d*!5pD7M{^%w7fT$1z;}rl<q>wX}@`ClD_B6M=B20`8z}
z;12PD(<2nIbO8>rbPaG>L*w-QiDTnrwehAc1XC2u!=oe@qGt2^+5B#Y?qx@mtM&!&
zEz+(M2N>Qv!F!zrZ&rRtT^W3zWlzc+F~c{)5DFS|1g<c`5ENRBu-NtPvA#m}9B~}K
zib4^3*ci51CRWo@2fZ+Klc9A2O%XfGmC><QVVLrKPqO^aIv?~l8!U(pu(;V^z(*WC
z8!S}BL4e`zh-1$Ni?e}S;sd8=gR>Mk1SFmcxO1|BTjm3&XM^Pm93mJS;P%Z1@K1hv
zl;m<YcyDwzAm@yIbHShgT<}Ehb2%4?&hDJKvRVyKXOtUcl+C=;9Iu7k8oZy5rcylQ
z`4l&j{&N>G(_mO<kf)n-Z{E#$77iK=^5iTpQZD?4-WvNv=NJ(VJC6X`dD=E9j>~)m
zU*1K?k&ggZGCP$R)uM>oQAB{#SeUJFg-^c6D>j)3>gnW0BU34?WRIG=2<;_@1+U}(
z+%J8!=2&?k^i+jISt#*8xI_+-v~{;9N`CNtoMED>qDg-)aqe;sj^GFUGw3|w2>QVf
z`RAGPiOQLQ2An7wunubzeK?EgAI-Sewpyp{t~hpu)_q5}E|4+bW0H=c;!8xu>y1TG
zhDx=Vel|5n^y_z{m5Tf<+hmJnua)=<)IAvXn%+T<uUm$8Zx@4~olAK<mvtS=9h{g?
zjN*R06~$e>oF2d(d<;g#6*M4R$*C&lIQUn25Tvct9Og%vV17|%tg7}6U!e~Fqfqd_
zF{frx)%ND0#P4&X{7c^-`r7V{4zwMkBWv4lg_MZwO8g`nzMuK<IRgp!Fpvy<e=+`L
zf|e-@+()BSl`74?Ly6dE9i{X-?V)_BK0dI%GTAQH*0)Q==K9JM4<#mP0*DU(HSQFt
za}@ni*QcaEoNDP0&C%gPm<gybqH`M^HA#pt(cvmtmvC0T_yn0jq%Lt3sZ(-RI+`^r
z%t}XyS?L~+E`en)O8l&z_&mDAU~m#uW9kt|xD%Y>+!a~7PIc~ztX(smyCQ4XQ=Pja
zYu9Y&E;8-}^W@!F4dl%36ai<Y3wnHB(<4l!1Hzsq2zx<yH8{&v)bPcG8di!LF6piY
zErYS<L=9gyYB00r1m{WONw&Fw{uFF;X~;HRDl==kR%Rl@B3ZGSX(=|bAEqlg7+_Fx
zaE18^S8{%eQF4yq(V3>?IHM9}VpNJUW0joaeT8ZzN86(?FpY^ci<0x0vPaG4R{OP}
zxjy1JqZAa!D9uV7F>~AYsQKBzE%1SJ1~|ZBfKLV7KsIoTeBd0}0C12E2DrzEJ<6-R
zofRedK6@00q<)Mf7VL3vB2_m$miv4!>4rzUy5WR0EnUeVFl*x~Ssiqa4&#$!IOVQX
z2Ol8Igq+vkt&9}XwDbm#I*4<_DDhw}F`H?rkF|^sNb^W&sbQ5cIGmq+(0TH={NxCJ
z@?qx*3{Nxo$w!?hFgzW}Pd@HE`6ZY3M1Jzm&XZsBlcV^_gU*u=@sp$Z35ShjMb=g#
zj}uDJxlq6Yb)q=ls$-e4oU_89<e{39A#pee#^4}VEINv&B<Pms$Lf|(x{^E8XF+q0
z0i)`^>{VYVs$OL)o!P2(PWYngBXrejIiZ^!65=LBi0}>)0E0B)om<p9b}y|JhzEN?
z?40lg;v-`a->&Ko(Y!ozXLsZd^Q7qB@GIs;`;Km>-Kq_uRYkj1zRuSZJWl(fJxA&G
zgfN69^AB=3NFlKw@63YzN}}#Jc<Xj7nM-?7{wI?4{o`DHAMd8C_z;><@o@ze2Uk+X
ziBa)4`_v{!(xOaAT2W@KioeBIs8;cHkOl}_Ll?`U;{Rm?X{a-~{A+`2eN=S>FS-hX
zH!D@8h13mC)Xs))rw^Yaoq-S1d5ZY1%ZBd;A3jHf10O_qcla^{X?RgK#JBknJ0cv2
zA;P;v+)I##JF~UnHNG}DVjFFM*f!em__3Sc9i^&VdL72D0PDsls%1Xk)6DG)(vXm&
z_*N_ef~%R5duc&)ti7PAPmEQHa!K<b!+<V=tK=n(bDkTY%wqLLE@}EnMUmRd0;R1S
zBD9sZXGw#@+$ix$S|L&@>CF+v?D>2rSVGm97VcYc_Z;W$4$9y)7%X@03cL3T>u&H3
z+SwnpRlo(j0?C`pkWfz>;F}B0C$^F}Lp1#<UPaAXoakLdoaF{ho~&sSR-UxSewwIs
zPIr|$2eeV?sYa!eDWHGO7d4#MT@6}@;*>RN$mtqXR}1BLv(+tw!4~>^Ft|qEwaI)z
z$YEU?!<Toj4`bLQYs4qm8nJpiT_?t<gHDVqOfR^S(@Ts_{BOP~%NeRD6GK&$8LJb2
z+E=L7iFN1&j8<bF%%T%NCPFWKi(BhgX}|3wiZfh6X$;q_M3E}3o2vGaZ1}$K!{>}_
z;KRu74xep~|6w+8KlXuhgaW`pD0BxdOX!84M<Gk6wm<jK3#mraPjR#Nl8#{;O(*N4
z6{%qy{Tu=TiI1yf4a12#g->qd307(thtLEdHH<^)QrJVJ(e$|<4Fe&kP~t6IVm71c
zOn&u=a@q-}UZG_i<J@IaQ80_&Ki+wAI^TRM|D5DJfx+}?{Nxn*#GcxY6;s<1OtpcE
zIF-S~sYe!KNuk{@;)+DO!={MY_H<dfB<&gWry%XwqW<T4r}`<InwVRXCsy0k6uz52
zt%B7st>Ox@1y@qG#7wKlGM}dHCOiCzGQpps%-Cu51Ye<gTGdg}z@k}&piJNT7>kPb
zRBpF_f<41W9*2`bb8xajp2yGXdtQ{Pa`nA0D%t{-U%hY%g`{XYn~+oXRp&`4IF37D
znkLk;<M93>svyqc{Uvmt9NwQO*YFdS3Ca;)qrg89Z>ty)1U#;+*;V*3U(m%$CoFuR
zP~aGeg+j7aK1D8fVn0b2&OoZsV^HIsF{o3Q63X+>Lo?}ZoeweQ+pS8qDJi#JQ4Uuz
zl<US}Dy(7vRb-%F=rEP?VLBCo8dSs;sEDhssVL50%<&z$nCoz$pk{t}qP}BsVYt{T
z%H-JfhNSS7io&?EC@c?7Ja_?;lRZRxyn>%`#r)h^n7O)qwLUsprdnkUB2_)i719bb
zBe;W#9F*r7ZWl*LW^QPdN6L+ItyEr3=vURNhy;2SdA@Ea4iD2|PT7tvVLLjPWXF`D
zaH-cy)WCbR(78TgtFw<l1O)e;C#lq#?=3yws&A)o-_K#PfSgxMBXqOyD9w)%Z8!!r
z?>x2DDAt;EVt$;@3=!^44LTiF9^Mq<yRg_KVfFwu@iy5%vblF2M2YN}!8^14%^j~(
zUJlvRXHI8RWx+->I|mNdS^lEKq>_HDXzUq03Ds-G>Xw97U&=5^gO4`1bI63rbe*~W
zHmeWN!5gQu)v9Qt-NE&j5&xmp();2#(dwhFs8OT04Atw^{#>)wh;3i+`jsK~_WPU0
z!SJf(+GuM`_UV<=**ddyqk&e(9xu4~c+tces$(m)R($*3sM>dCw)T;@9G_^}&D=WO
z&2%bBYV8T4wI^h5?JHGlm-t&-Ee@5d1>&g!q*9?WEPM6Y)7jj(TCDaiShSF0zkq?|
zDYz3wdk>&ajT@7}9jcWB{#ME{Q`k|ga$H6X%B!cdfldWx+x80^c9Q7GlROQ(Q#EXn
zzhSKg<*z|Lt^!L#*|aUw*|bO%X5;#1Z>Ly*g>r3pw(P!>MJrF1tt6Qf+kJ0St<1or
zl_u3tqV_d72mt?gPcUg*Ek;}6?Iu|CNurHU5^X$E2_P1;@e=4ws<{~@P^;J&E$?Xw
z<Wyob){?iUh~}Omnwvr1zFW0+fuDuPi_PsEOO*t!?A5ELlYOn?jTRc+`DD?)Cuiu+
zhoJ6tP6$=F!cehQ8Y`5=*1AwiJ4^kIPWqlDQ_#MMejGdKch=Iq|D+$wx$nDt_YwE*
z_xSGj!@I(Ax%u~`S%->Uu9$0*4qYwh06!<;bHLD6t~FL^lB!S~Epz0J+;F`%yOk@|
zYZOhA9@cXkCO6d^rLkPATx-@F&D><UbuPacw41Hm7{yqjVq0VM|4K9==6R(?p|_jj
zk?8H*{9I{_f|Ay9^#%thB?h2djdD;fwa8_<QEpDuYt3>qS7}mkQTno8&sFQS(R^-g
zrO|AmRH|3jW0rhwt};T9X46MaT2@xor)ZJ&fj}@2)=DI}s+=oUoAq3&Q7*PbshGZ+
z2!_fnbH33?sra!9k%rMy5LTfj<;vH#iwpsP`THh?Sf$5&DPNI)Ng{BZnzVYT8d?aU
zxah>1Q!r&|U^NN}Amy{%hrUgiu7AL+!jU9MY<avSJBO6845LW~ntui@SbYGt-l^F-
zG;{we35qyR-i{cN@AcZZtsVL!Hd14NZJvJw77C?ewOS|$nG;M#!M$M(759^H7zrCZ
z**>78d_VyFgrMS2)2SF5;mgA6v!NHOkFZ_DEa;QXK>b3D4mf6vgfh}C8n6<;qJb+|
zG;k#>8gqiT5f9lWvS2f>&C{&*j%fjfUd<n6H<Ji^0xSrj@h^zkKzNo93hW4IAFcr9
z8VzMm@Bvz2IqfU>3ROmdU>m=RLW{c4EXQbd`+_0<QW`kp;FNrD)S^8YR&dJQ!wE+C
z8zy-$%2z3n#`t*}2UPe4kUiilP-r_}b;ki>`3p!@{yGKHI6qH=q}IYgO+l*n0BM51
z0i<jBDg{!5pQk}m%h#aEuYyV~zUq=nZT<$3cJNgSq@Dac4U$@82D|uGK$_&ME+IXK
zzX7D{_$md`_53^yl3EZ3H}I=~^jyB`64H(Q4ItgbS1FKg=I3dU)KV^Z9={4m&*!Ty
zA-#aV0i;{_Dh1LD`FR>7weAXD#IFL<t$fubBtH5hAl>GH^kN?*wKfV~q9Ap8fb>%S
z22{G8uTrS=GJc*$CAEMF{)%4(m0r$QJyYowCZu;qCzDtDAnoQCAS&<RD?q|R!GN@R
zV1M%T-KXXAa^!KU<vV+;S%Xg|`6}@$v@=_*B$l%QnGz^r%_jc{*G>3S1aA)1nj*M|
zIo0oK@OmHbzd`Z-8x`+kHDU0+XIGLN67X`vSW*OU$%gN3K74B75xiaD!-Aqae4SHA
zXvx{8h41p*`bx94Wp$<0+5)1ltW9pEB^rCu25Pkt+#kg+oDMw1z=ru&{#XjhOkcfU
zEE9VP33qy$MThodXxE8MdXR2EVOUr%;!0X<1R>?l<n1CzoTIbDsJkXZMZDAH+g!Bw
z|E${G9Y3MmqxQ?DrS`-vJj`ua20TDraa-_onuBjm*q9SKK2W9Zj7sABX_Albwz(^T
zL{yHN4^)><53Q1S!$2)82NH2Px;}7im>ygS`&faqEJ%dr==^||Y8UOIBWVLqNOZNq
zbvyAB$qZg?2F@xW5uvjL0K(1FL%gU}%W@zHoK-_2L}y6=h^c$A*@>9E;S)Fuh(vVG
z>Hz3cT>-MA)3#T+Eo64Evzkc6=PVVbvx75jwO{~ab&&|sSvmmXG>($w{gJ>~Y$PId
zRuVv%>I9L!lhb!LKs~HFqSzkd;r@xhSyC{z;6K=uIn8?y!QDKkLw8ZSVXVE`n$aXk
zNB0BBA8L>(mSoSdqqxc(MGlqyI9gePW!Tov(po4j!ojPU<ghcEbNNTmJ)NUyl(ZU@
z3VBDdLK3km4evqjI{FCEer7<soj)@3aJu!QuA?uZZ_T<=yu9XR5$4JMUDmFl<KK0}
z>J69eSiH$n{AR0W{f3(*G_cN*M1efL9lmRu;ZGH2_}|HOc|yqF&hch-`Otw<TK<2o
ziKq^{!BqNUS}n+ZQ1dL~OR%{a|Lmpw>~?;30Og_GZGQGLe)L!T=uq01+ELhEoozkG
z0_ElWT=+1V-!3Iryx<k_&oB+|#xpoknBQUkdlf$;Cj|2Wrk+>x19Jo4d}BFR`I`7D
z&*NvWjeiE4_g(SNF5vH87yoP(Kf62r+0S_Gb`L*0Vh;Ov_2>9^uGONo{QbT0-$TuN
zL;N%7S@*?1!|d|r_-AjS+reAopJC#Bd;GI^@OSUzA9mR?DgG|Lc|ZSzVRb^;wvyZS
z06mFe4a_g^iTmoW`R3p7S5D`@mmj<j4+M+aXItb3VZ~S|yP^a$inZ|Q9N%u13$(Qa
zZ_|R<_zS&FJF~3um+&S_w9eWrGAB!EQXE2~)y{EaLSBr)OgI?h>KA}&yb$f2WK}~O
zwS|c?xevA`rI*5aT|>3XFaUA~iEYJ-z*mXE_mG$r4rLlw>rLq@?T9x05c9*MF{15n
zx%dB$ek@0V5Axj)xp)7b?|#_5`wx8gBktXg^4)*rAD4YU#y3AsKQZhJa({xKNhX!v
z3fpPOuOg3tw%mcIzM#GTrr3ioO1A~tHJ_+gE2T+fc2SG^99F9I`Y3x93>Ake)k>>U
z=BW6T8mPG(8|UJ~f-q(gK}RXHZf?E%Nw`ffs}Tj0!>qAAuXq3Rxs7AR2}&zIM>vHv
z$;G#Su31}XE$gkm@as%9Dr#hHqWVCRuZLEuA_^f?Z$s55POV`9ncB+f#FpnamB%OO
zbP3zfa>zMW=B%}0GeQRH&$T8eD!lbSnQKzMTk>|H%s}Lv!VVYY4m}Fb4Vt{!)n#G+
ztbC6AK`Nt^_*uAuxPU3QE?1vha6Y|_zE*D_TJiGSx{=%_$_5SB6Qdwrg72sdmxl|~
zMrn?cL}34q3We*alqxNPN%E2$cHEa*W5rgk*lbdfP3dByxqXt+QyUHP_4Qp|Cfd{_
z|F6uo+7u8(FgQh34&8Z(dwHCdDNUyw8;vsM+Zm?6!#<(j#htmLy_?vL1)s<=*jcDW
zCcP$T8WJ8oQ7)2qOnFz4W08bJ$6(P#*|o}%EnI@aibJenMWXKnDAnPj9P4{c4Q@<x
zg8YmsJId8b<r}kIh~E#0waPz1ZjsncoBpI&K7pO}#IR>t6o|1Jl~O3APq(I$q*o!b
ze-R_HSsS%uU%-(4AmOuw_$L|gzp*eqWHy$OLc_GD_rDWu{7A;>UAFIN4ccznsR>O}
zZ=qGc%HFC^$F}OBgjRJn$IcQwg6cG)8##$XvBfC$cumhAl^RVk(zA<KBkYRSqI}qj
zCH_&6>vvvqIZi&$n5vUcXa`8lS8;~WKjPhcjIYxjdthYOhIec$;-0j5$9dEIjF<Ci
z+oz6Y9)<ole-suP<4yCZuh2Men*Y^>`n_pxgMD*QY!UmsX@22@6X#9y%kJU0HuRKT
z>tFdGIc9c{{ntG}g1)TH=P8hW<AdZ_*a7MPdVu7Z%~K%#)(6S4tOL^TdVu8E$x|Tx
z-UrDsr32C*dVu8E#Zw^t(Fe&fodePn*g_+QaMsc08pBf{?Z>63NfpOl4oLg=0Lig`
zr$9Qu2gxy!1JZ##K+5h-bC3^`V-g3XgL{DF7`RiYbchd<V*>}ILwkUvspRvfIm`#i
zF?|CPOy353@AIaCyZz(YXMS3}X=FH%p1o;~W~TOg(;VyLeP>4nypJ6fgZEP%NJK*-
z-ZUp>!*{X|pR)-AeAtBP4qw;aG*5}*xBOyK6ZHO^C)7|#zA5V8cbc!&*wuj|jw`ep
z`#aHACjsFt1ZxW@m=$eOoTvU%ec<9e^`FKcMZqO`l4(zNok%jqQ~wMf6l@cLPPhV{
zuwf*i#D@Br?<-V?`swFaQ7HC>vbIkQiOuh!l{$=s%X8x3HLb(qPS{_{XU)kJ#RsYK
zwOXG3Zdc1&b1Sg3!PbcpY}{9u=hWfCUJBGD3LDu-wN6%}6=bSf)DNRgfvjO<v0Psv
zTNxdVpH1eZ@p@%=c7JX*8^+3fQEIS7gCfT^>JyWEgSM{dS8ihQRdJ*>o2%0I==37l
zQ?1r-;`KF^NjBKFh(jB;mIG9%Ed!&)*5_s%2;Hoxr^s@$D{&KfGpwHsfru6%4*rIZ
z=)U710ry32fI7{}uMl2rXEe(#0~MpCJIy>Jq6-bCOvj63J`f$}e%7DU-hJ+~ty^`W
z;Yq=U0l^{+d9#7EPTU4v0vHlOV~iqJk#&&wl(%fzil!!VsR4zprbETjcIsKhmGD!v
z@PKY~Qa5}%FVHo(M*i9;uudTv#yJoMmAHw`kz2%w!2LYefv4d>5x1)il@G3wUtb&_
z8ZNHCoLg(|4N}iH%482Mm5Fu9x?gEE3uFoo?Q%u~G$BOYn{cc#tWx~NB*IXYe;Tw^
z!4QlhISv8g5nGk)2u473IDxbmEr*k&7bnUiEgCMhk;<s?3UiaLHR@nvyozJW3gB?1
z$&64Krr}+06Q}cIdMb?8qh)f<q;rsk;dX-tZ-IGSy|R`LN|IrlG7<7dO*(N+K^U?K
z4z=aiP?78F^@+>cEh>g+ji%_f7=A7>jgT0Oy|v{zCFV>vVDz>2cwuL~u^k*I<isN3
zRXd1NkiqpV{v91F##r@S--L1_zkpflCN-hl%vV__lv{-}o-)PgNizn{^L%jBu4eFj
z1?L4lz`2FLVat$)6LXf~g+4eh;uk=jTlospE%qu6>TJH7w3)tG>n`u<?ArsBr=wGL
z9o3f6zCA!|D~hD~zCFNPF1K$F(A)#;eyxAU6P`dP5$IoVKZ(Sog%7XFHuuqsu`UNf
zuUN}^d8m2F_tbW)bWVtQGiiXQzS>o}oN(8j?#D^FuhfjGI8vRSyW^*2d5}eIum*SW
zG~i6TF%!$(zKP`?HL={QCKl|d8WT&ZX+tJB9D&Z<FZKzS8=AV%<yq+sRl@ZYks?$J
z=^hphLP=7KD%|8H6W<pLC2`AAJ>Hg$2Ji6E;GK#F?@~0t{;NTQ&P(EF)@j%`b%xQD
zgM0Gu>HI1B;{(zDu*^Y8;~j^q=J3bT9F*y+_X|IR`E;a*+zsv`aiLYLJl(Wo4~VcA
zzri`E&VqPHwDIzIy)jv67DwQiM}Z)VRdN=Q$E;4io&aKR9#_&EO|Xc4%<4fxCX_W?
z;hXs!EyFk?6?}?cvc49M1A(@ze3oCq*9YlX8_MJM)92FDhxq49^q%P#`5C_XS^oK|
z^#nOV;Hqx=Mg$LYHQ%BiaaTD%bmI*!k9YwS)YR{Bp^UwAL07~MZ3KB<;TPB_VjpDr
z0hcTE=?CaC_DC1n4-`LAC|2}_f|uVEs>{akUl}m`l-~h{pD7FpN9_{WPEkT)2<;>`
zmY@5vj6EoTJgR^!?~NF2`Lb*Og}?S-EvH5SC4Qv<T-X}`Y4xGtdmKtpjzH5)8fA%x
zs7^Y#@qb)3K>RiY5geiiRajOxEH6ub0$Hv_$@}pYhU;NY$-x2qsE5K2QH39>3O`&I
zu4KViFxctG@NVj6v<FA^2=F)s@OTAqMgl-f=zN2_foCe~ZmcK@o)Q6f8r_M@P--32
zkR3EjK|EbSe4>Vkt`D9TsUjNNDX#Fjs_=QL@T1+r`+F>Wp(=clD*RYoxa|F|`uUt5
zp`5FroTs21ub~Jco*xz-{3r9{$Ek~B9RcGAd7-Lcm8#%Gw}Lfc1;X}}x$t$W@MoyP
zPj(Am-(%q$RN>E5g`eUUzOl!`uTq6?R)s&sE&N$w;lY1(P4{Q33JR)%({u$&!u*U&
z{%>2t3<f{vyZ^&KCHk6m5TQK;BO#<Ndn!<HwktR@W8nOP>-?n)=U069*Zfnr;WT=L
zvqQnzso<O*1LwC~=kNGuQdS(EMEO16`~&~6Vbavn0QQgk<VN{K((ZX&RS%QJ3;8?D
zaWCR4%tTMqU_(UvbKva*=m&F!OqBAH9_t*ZTn(bmm#I2uN9#P8>pX;iLMq65U)f{5
zck*{=>#J40a}(<Ai>dc@+3U4V9^T5zjCmUHcC09Z-);)=;0m<Ai17%Oe@^gwe&G-N
z^GEuTJ^h3$=<oD3zybUUid&$Yr$*)F#6IQ(Iv8x?JHa9JmC0*;^!H$JxO4XyW`f{I
z=kA>-!9J4G)f{so71#^^p2FXuPoK<JX!Jl#qdSh)Fc{?M{XxZAg3-HzJUfIOoXP{T
ztC?%Qg59rRFN%RZlS`h(#MM31704wD<mp|5T&h4WQy|Yw0BJH*R{&QkfM;|KaE$`E
zRslRK2H+wlph3aGU@q4+kAG~Q9Sr94&Hm_<_!V!kA%x%MoV7_ozd}JjCj<1ZXuGvX
za76`fNP%0H04`Wew6bV>CKGQ7|BTSr2H(aj>f1w*!CCxGSHq~TfKDi&=f(i#(H|^z
zfiC0vmh(?I!GzP@xqS0H{)uIe711XN>~Xzn-1+=XSB<+#HST8BxD`o_TS>)Ry@1AD
z$d$%6Zk5>t%X$;6?iy)FA$_SrdO;G>i)^IvgtPpsNM+wm<z`*G-Ko&LTA^8~(M)04
zezz+59#!(HX(?H2)c2{9*XWYf-2G;HYD==W@g0XvyAl%b<_fcEjzX)sRm4;Hch-=i
zT1`g;9azamP!(-z-VO9F=H@NjC>eBMj%&~vB1a!#J}o!$oE0T3MTH|s)qCVAUHOq4
zfM8t=Iaw77mS5-$9T+6fb?=@>H-Z)J-SheG1>s$5AZ48qq!|=gSd;0Th{IlCGlXA}
zFlsUhJKb6YIR->rVYgFZG|tjZVG<mg32>)a;FJfQw7LS|I=5ei14^RF(gf%h*bPW!
zNba0qfi$ebz_)q|_>8i;fd=LQ{Z6)^6~!<ysdv#7(AmXh0{9Sf!6!Ng;ZqrN{Fo?4
z8O7t0m_dXiS{H^xAe=%Gzhjkq7x6AuyLS=kVvT!uE!_w%;-7E^>uGq&2A)_T|FANh
zST0kv7vUl)sg7=Ws@3GgcOk<r)kFwe%SOrI8DR^QxMHyyT*6K0X6EmlEb{<RvKTU3
zR~mGBSz57W2c2P|O?ex5FVlF{L^8;rE(-yN&Dg+qpGiL)J%x5D>*IFxl*{?~Mt)9#
zSL|3PoA}Wcc(llQbR|E!ihn|FN_1qWzDnjhphO8JI{HngNo#aOx}F<vS6dZ6yga8>
z-qm6<i01|Vvr5~-b8+&T{7#uTIZ2${(4;&)P05&YO+Nli-e}^?Gx=fUC^kPjWHUqQ
zszXj-H_H+oj27NNGN7A%?l=^&Nw1<So*B~>TNutY^keC7Tlwy@!@Cye2L*oCRReM%
zDn3$P;JgO14LaHehQR`hvHTyQQ5$0#HN+50A^ecPhWT#Uy$h}y3GYhYc;!`Nbb^>1
z8z=)-2r@^f|H;>o<AabpE1ZouQw5zrZc_FKPGiR`&}<LU2&Y5m6mg)>DPUfh3SxwO
zMU4aP`P`Ki`kIn&G>ep7B3ko#9CtwS&nUN|JMxFj7!K#k@Ggp5P#+J^+RJfmli&Xl
z$bUr=`Gb(8A^9=9x6|E(;f;Y><<{8)b@L%)up$R&mYyJ5IN;btkT^#8V=%oFMj!`w
z;9vyoke(Z^j35w0D@QR`$>(*Hf`;&LZc=`fQVV06!XQ*?rD~f~5cUZ{w4B36kU-*q
z65_?y&N{tCDO9N6n|<s6JkI_yLv*e@w<F)o=g0+>@|WO2ahLg^N!g9)=sLfN$6*cn
zQ*kF}aLKrsGeRC1@0tf@l8=CpNFFBOI9ys|%kUMr;Ho4pn26<qYolCX%a}&j4LrjO
zwJ=g_QuZN8nlrm=rTi34y*jC>?bxR7@HKU3mZn-3-Cg`ebnO%T%Mjh?=7}pt+ervu
zl%x&Q8I^KX9p2b&xqp0;i?^0QQa%57Rqv<y3iT#L$<0=;2t!+_701g3^5>82od=)c
zdf25R_$*&NKv%JD6`xngzMzn;?@q(DDMG1E8%1$yCkED+6s#|&!uqO$^)&^nOU2uP
zC6;MXZ-;LvMBhwB^sqwoErlo{cyGooBpF9%7YcQn)EX24$MK2zw!-zDR9xRxxE@is
z62gaN#8srzSU3=6d1D0MSAc$y3g|})(2o_Mi>Jg<l2FX-Hl&{@NIy-5^xq27&lRME
zaL^fPq>dwD2->I~RjB?q6;)krv;9(`np)45D6SBBxm75LxUsejeysrgCKb?c`6;C1
z?-Zb^NhywT$8(Cvswr~)n#g$JD5TUMQjtBO6e8P?uOL6C)`3DC(BZvGlNKX3y$;|y
zAUhA_t3=s(utIf+Le-@clq00TRyLT!6qvqLFh?jbGZdJ?DGfV197sLaApc1^)Wqo~
zZD6c|3MxG@71*^`(tF^lqZP2J$pbiA7lxy<{8+949d}$Rq!ailAf2cnb!oM8Iu5}c
zY2r&F(aEWJPEmNCJdJpGB2L0{YAT+b!ZTChnVMu``dpg^==4-TPg8)-P=HoWiMfa9
zD5OoCIb1TBXKpH%`3g(F!ZJ10gc2jxNJK^)G4l%1Kq{g|3ejSPs7o8K!_LF?HlKwA
zFq=wRVlPR>bhg5Dj>5ERN<9X>(wW(oaoOmz<*8`SQ)pHwG*jy{pea@-#+)H`fkO23
zR74jlM5`2{gq#Z*RT3bQCueDPql5GggRfE0)}}(cSV3E-piPaP08JjOwb|(sg=u{%
zra^`2GKDE2etbr9HH$lFFKcwPQ6425$1t6aPL`I>OvScQVcVpzC8Sr#h>fLXqc~m&
z4~8?kqT{cw2jZ2f=r${KS1WW=V>Bpr)naq3P;9lxU=-1xv}Uj+71~w>?b!-iLOSw{
zgp=_MXbYh(m!L@rNHqDjrD7aX7)uJ{)Yiwt^#UC+FO$WbtfaM3XFMpSs+<aIRKXfk
zu%0m`@d`{672-SUP#aXhRY1yaPX#ruK-CnescGo|RcmdJLp7mLU7L!ksZg~Ps;Mh`
zbZUO7O(o6sJ5urNQg|j6o~er0;xGlaC@dkR#>S+%{JK<3%PtNT;u{pEsTntD1}awz
zNu%#Zt_2)*Q!1_t6t3qfTvHo;M#PgQ8C`60y+FabB^A~R1?xo$*3`N%4A2tAbeqET
z;#5o>eu_2COBJT6Ewjq?CRy+hIL0QE=Jc1PLVCG^^a=%OYEzF099<Zy6q|*iHg8in
zI*x|6J099YWC*3BL+@fa3u?V8gcfYoX6w*=(v92AZNly5wOn3|>G^C`Bd%b=IKMlI
zy;?-tNiaSy;|k)8&-$#^dHhN*#uK+@9OR>*KQR{G^SKTT?F;xS(RgwpKSd9$ns!uC
zOW_)YYHcd2ixsMM3RRaZ8=(~ip(&9ig*FrAwp`L^TAvDSP(iy)LF-b@j6f5($SyZb
zTgkDu!)J0`=+4Vi(QQ)bu2ASMpBj5BCQG#cUZucoP6hWY1#XK1*QLCUkPc2Ng^>#9
zBQi~zTNS2fr()WsFclT1F70&@OmwckLOezac?__U0ydlqY(xPYRlue?1XvFc1j8Py
zyE3gPXhABps)9DIpmk}Wj?j#uaX<pGIqD*-D`XR?$QlY+Qz7ee_KhH;-S={lw#*Zk
z2<=oz=O`s*r-C$9rXQ--OWO;a(Zgo?Nv;7>;yJ09uIHx^`Zp*{Q)PPIKik=$$bCYD
z*}$|uaAPX4n-#F<DPYf<qWCgKW0O3cXtFFfO0-@sjylwP0app?-I59t8H+%@7b%bl
zsV0-w%Mm8i@!^Qk%)nqa9dA<*Uz`fD!%qS6r3zv~GRMpiM;rC_1U$<~=_|Hqb27?+
zFH`vbDiz-=6uws~d{dsLWaKylr%gP)3cDi}+?@*Cs};DZjt)d}Bs)WSh=j>@M_bmw
zUYiQ+bqd(s3fNSo8>g3unWgUK3NRL4p9<-X3etTF(o|_BW+<9GMW8~v6SX9Hb1JaS
z6#G<+fVV1Oo2Dp9NCAT98)o(vCHSe5$<l`RcCHL`dq+IH$inCT5MEH2mW9t>b9vE)
zkL~nttX&`_7Co0f!iuEJ%>mOp$tYBwTsUSeaeS*y2ORV>;q=>Cz=?W_D)_3A7-`31
z?I+hPH7XNGe-g8z>KrcI&kD_QnXIzhkj~6lKzZv$w6j}%#qc##ZLy0rw9n4|OBU}2
z8+2xH+|W9CVs7C$oVybjVx1F;r4l*pA-bj1w1rlScJ;@%E9o0FvD=9=wuwnKn5`8l
zBHi|q-jveb0FKaTo5;r}{em@dMPM)3XK9e3McwMZu`RO8HQR9jdt{4dJW+KUqIae>
z1m2=)o#gs6QWBuCIx+0@by89dX5)^S%__DS2(u&1>FF%E9;1!c0RY`uw*RnfTemZM
zY<IaOn{DDug)D44U38bq0n=fZT;)Kjn_?PE(M8?PXWmlHCLFkU;S#c7qZL_YMWB+#
zbY>^F!3y&$d!$<(FtI64;Z}<%)OFY|nOmfiawJJ`BkZnh%d|*vt5~BI@az`7a-<Rk
zh*X>`!h@GmS}-vUJt@9hIPAGZ->u#j+Y_QZ1GY%?+x3`0Hed|z)8c_6i=6O`z>w&2
zQ^H%Q*@VMX*!Re=NS9KbqhUZG%L@lWmo`zfDq)nwypGmQ%P5H{+-#+9&b&4F{KoyB
zAR_DCE+Xs2X;XsTVxOQXr%IK#NWtFnG1^#o#`IKDx^>X?O>S`f-BN3N+REYLM2q#a
zl$Xi(|1J{Bm%eC!WUIUM5BrDq@E{;PQ?yBVWOt5dzc%s!XxC^vo2;MpYJ0p!%OiB#
z{CK`e_yq0ka^JeE3XJaai3ibp!33+-rY<g9J!3@b#l)nI(9>U}WsImn&Z}r;mo69a
zt#k@DTGu&#c9n9Ew#h=mYcNiCNh)lb(%@<DN$~QBDN}Z|b(BNT#K~WQ3fFYGoQoBq
z$#ob6F}hpNXj&L~T7M{eYS3vs68&Ink)$%x(fTd4uFL(^_&(NU#<%7`p3=mO=4r(7
z+<*cf7uzBHbc&PM^-CM537<^gKuxH?XlrM`a3OX8t-mt0^jMTiZ<iaeN4Lrop+K4&
zPvzJT(8{Z(gGFPDH`}=X;!xkp+(^-#Rdgcz4>c-7lrte+?gB^jlWFeGXa#0dkDCpT
z2665tFWwCW(*ZOhGxhLsJS>boH~4qSw$F?2;`lGn^lnYRv0^B`e52)t2+BTak4k8y
zbuL=V+vwY9V7F$o*ak++HX7IIW1X*7XpRnd-I>R`C7Z|96Eui~Po|HP6Dly9Y$|}6
zIrnI=DrG+zAD=9+_D}gj(lp$d*U`vn(Qso5HycPt-<L@Z{va4pGL6kh3rT)=(?WWE
z&$W;WH(PiRwQyMvZ9Ga)t~gA-rDi8449n=-(>E-mWt$D`tqa}eg*}+9d0_`mi_~$N
zH%<L_xG2`_x&*tk^&)z0TC5gUu~Y9bqdj3?dD|0q(=@kd`g`b1DL=d|q8Il>4^gpG
z@1Y|_kB`?mGC+G=HYvdt>%NOtO^YqoE!k{Xufp&cQRvQcWpu1n7%rEJld^ep;#IM@
z_tC^Fr$gC}E8T3SGC8NuY4R9+Tdw|tU2F-sfq$OQKezJFOZn%O{PP<AxtD+5#6R!g
zpZD<32l(g1{PS`Cd60h|;-4?_&)50qKj_Cibl;h2ntWnOxZAQxe1vQ@6xry>(yYSf
zvl1OEXkuBw!GK08yF`x!Ki~?P)<IG4Rfe&0Iy3lFa)1`0CU)EL*M7*h(Gq=y=q)mM
z$s?D;`h!>*q9PqiLM=*i>_*tk)ki4HmI9|DTIBosHWcZQMwymt&AtWdU!Mggw_JbP
zCu>5M#Xb&r!&{N^CdJnfbVvo#q7P!BTu!^4jhe#I$GPLeP#D8Zvbb<4uVR8e>IT1+
z-Ev}y90qXkbD|>S8=-e(k36b+q#V;D$J84YN{%D*K#K3|Zuu#9%g^Ws-4X^s`ES1a
zbNBB5@ZCr0rz3P%4p}KWa`VGvp<D%^)Jdf}Lp2D*addLE^3L2yrCc4B{mIE)hRY*G
z3SpF^b#}YlOzg^NE3E!RsghCY0j&aKdX*fgId}^9+`M`CV%|K>S$x}to$nLFz{4(9
z!T3TJ%p4nEE|G!+ooKp9=hL5Kxx?GU3Ods(d`ZwJ7jBYAkme8sNMv~ADp82@u|yNn
z(&!$JRzvI0VWPjwzlzBv6UWEu0|WouW#9)dfQz@2tQV4cNq3T(Qch9uB5NKUgo*i*
zlVD+{|M33VU|X1$Ma7*HFE1m0PN{64s+1$=pc{#sfu_q-D2c@y(}aX~ihy^T*pPXW
zDVQBh61p7p2XYi0iSw`!XE3Tqc&Jjao&AzoqLvGjqf@ra0mqo{Y#JCN4Pv0tu8~W|
z0H>ZDSbr&H``Q`)r`aeC2$Ta8m5DMtEDazk;Xsm`*aXei_pkuFmw)aeIWQQ!o-R8J
zyDq04yphZJ4wVtyM^}TvoA~cH(+?dLKw=<X35qLZX~aG084|>#Z$9OEdOfXBCz}uo
zMK+<Z(3rGer}_$Yk_@3xBpDKg2JhkqElbJLCR4L-#T3RnH_LM->#TOpAE?}^GMSA^
zA3W9`CFwspQY>-i<at~Y{b$~G3U0=!fApz}U=FwGp9vQ^I)rJ%V={Zr^EF+CmJE=i
zNYIl%%HDL2vl$qfh*IG6``|dqiU4Op4{-jAF+s$nQ77+%^HVMcER4KGfHTko93wAL
z3UwBG;8>}PfMrn+SbmX-G>bi0+(brzb7uE&2Ay7t8NN$=P@Hr|fP!>J*?E}5Tw0--
zgR{9-XMn|kTj~SnWJCfuWJEH+bvk{k#>$i!J69ROG|ve}$=Qw-E)pHAAhk=Rr<QZ$
z?8Zw$;rcEwR+{uDrKhGnzh({!8)aN?p@>?&CEsKf!Ty&QVIybGBn2Ajtx55t;C7VR
zcw%z6Udqj!x!Wt~Y)UQ9Myd0BlycHLfl^5CWKe4J=lAEVzM<kO1#7N8#9N+X1wd;K
z(&1S(YfLHyiIm*%-O<Vpih?NK5_BUijp(`@I@opi&C$PE+y}ELvG~;+*7WPU;cBT#
zkG81W{DJ<I|KfkMwwEUhi|EBGSFXQeO<_@fkuFHQA>UbSyt6p^PF{9e_{|OWE8dD1
zXQ+5_zWa(zTw6hl8~XZ|>($-zKHQu)k9(v*okWT@t?TFy9lLa;(Osm`lb3&`MpK{_
zjj>V_T|}KqJ<3#=hq)2OT_*K#UF3#(vm$hJ?XRl76LV?8mzR_$Nyi$tKBE>%C0jLz
z;xvw+gpj8M|7ir&U&g=b7;>G{LK+GJrfUbQvkl3$z9H!(G{cZYLNjAXcG|M34w)p;
zjFBftsW|}mD9KI^NtRBm!ED6Z;3L*7m4yeX)j+H#@(REp*5-3reu;mv^a=$N#U!Pm
zyktA12%^$<K`v54M)3f!$9%l3;`e4H%d2QHe24#Kkz49LkL=?2OgX`S^FK$1@h!Ve
ze24!zU}9yNdQXFim1q23MlkMcLcVbUB;)u$D{~W*<Q80m{0e6vr`uSbQ6CD!LZyk>
zoVsyOa=kq_U!IM(H~Dzm$?ON-MrJ>Qw>uj`YXmVdxgMa|iv28!hygGiV$p8xmDyUm
z+1FYp)gW4nRD)(~AK^yCO?JFiOPF8&B4?&2rma(-ovll@`MShOxri=7%0;tFwg|!M
zPS`wM=jBoxLzjo|^iR=}7j?SFU?^MThkcE|P|Ykz2Z_cb9i-X#lE3lcIr=?M1Yvdz
zMzRrL%trtxWhDrJl$9m{UK~#V@}VVL<V1rON1=h&$stT}(JQ4fD4&gbRZO@z)xX#1
z6l9}Q)kh~M`zGjw?3*T?w#L&5qlA~6aL3teO|NBZ`h>6PPHs>%9l1fxrf*JZIxqM3
ze5-X*(x8#8<t<;!b&4e<Ek(<bwA5_*2m3Zs+%{8}6p1E+CTTEyY>nsVxY>-K!80FI
z8d)N=^0J8(Qx2dVZqx9qoMXU=rN6UMtx`yO%3HAmTNbhp#DAtk3FtgkfrKTspZ8sO
zMbCy4vbE0UMv|MFVLp9MHrikBqrH=%6||q}>ilPreVoY8c>;1ZwGZ0V8<srkpcck=
zK@FrQ`DK%yOg2N5A%eDvl#+#P1EL}o;om6SofzlM6Hz6?c_DR8yAm6Mvb0d&br$TR
zxM2;*()y;QuTP*%2t<m0N|sO_#wGPIAKo;2_j%cR_XWP*brRm9caiYc7+IS?gx=*i
z1=i>yjI?<vk7dA=#|O9BvOzt~JEyUu4Trn=ls>uwPI&#{rs9f)`NcK|kOgm|-e6Nl
zIJ1S_(;6$b$haj29&Z(jtw@0Yo#L!M=gkAwY2yUda>bieE9+fU5g%ZeK&Q-cAyLh0
zw1Alwjyq3v%JLj~gcy#xf;e!DB5762lz@YL^6BQLM!THKmz`{UdApA<o!rLYOXM~-
z__FiSzSYDYdMIWVH<cRMP0C^k!^=3$b1RKy9o`AqPEnr)5XUQxtxWezU*EL!%nS*Y
zcH0Q2B?4ZNjnun+q;^s|gVadrY?As}eH+RRo^xp7+bq^9t;rl&=}X%w>ys%HEMW*@
zObZoGW`dI!55lHc%xF>M)!8WWS|3H6WYeGsl1-Zw8Szl$VzJS=mNq)IQN-jAzTTVE
zxhG1UGWLBpe~#t*I_lI41&MkNXY?VpXSJBI!)XxPFMMCvHtCF*^lwU%^DO?LXPW|W
zr3$X#+LKoaySS*E><ex7Dr}PD)3t5n38PWRfo>gU6XI4bc2E)NL;X25pUg70n`bH2
zcO`30JWnZGmTQieCFXOhg-Z;<s^_f2l&s&2wh)7RvNF!CGPuLLc}w(m_F{MsUEPKw
zmcf&mBf{O<7*~bY`v!`W1swwgS<uBmnG?ipC*J2PRJ$r57di?>E_7q`Z9d;+*<$YE
zwW}#N)Nr{0Hx@N3<olL`S#vm~1CA8?3u6CF?2qA@-V>obWVBi?HpMq^$)en4v<~Pe
zmJw+o`g7%0NtMm<L<+=I<P|qMxjaXILJd+kYOJkS`*Sr)lGjg++$iQK{T4afPmqgX
z2ymE!0V3}U;VaVtuL)BBc*#zgVA=@1OUoGgYM3ShaWslXgVeqZ&JdSb`Qx?({@~r>
z&X^`yKetIYam8%WB68tSs8<MIuVErksk~zL=tP^I@);r(uA|;qv9?%k5(U!iY4457
zsW?7*zmJcc^zz`N0e9i@FU&`q>f6gT5_}LYBW+SCrS6s{+epWSz>!`s@{49(%1y0O
zP*+H`a4(4p7K+RTj{Y0wrBMkfn@S+^GFLqg8{dziQE#_MwV$NidqvLZ#%#+>$Kk1S
z#GxgcGE^}rqM4f`c!RRN(R?vUqpF;{kTS*97F;&6U?WL&B>SUs*3^d0TTU@^nfU78
zwD9vH*-8C0@`ei%y~sb)Tp?83SjG%a<y^z;CL4S-8*_fl$DH#yb2Hp-`uPgXxyWVC
z=kiWksCN<7>gWOK)DSABVZV{pB7VVDsZF$7Lf0k1!5wI;yy$$&%}J{^k%;jz!mq@6
z3|z4VIgZF_8!hT@z^@r&HlzVB%lR8U_-0*Pu2$>)wB=nLo;hp5`ME8N7B1?~opaVQ
zTHVkk<)mD^aOqOMUaa@Os47I06UBu!U5CO8mEmT;if&)vU>*IaA6Yv1=?x0<7Q6a>
z@#oB0TM5W@vsH1kX~rnmn)Sw<En9MoiZmF3P(qX#`Pt&F{j_g4w|}+`!7PG4*Z<qv
zzhz<n!v014e_OZCrJX(fZ(0V`0yJZJ?ur^ss9UIq`g7}PhT^NOTN5OGaJm{$pUgI(
zKII!wOVxl{rUulR?tr?&98h$uW6A@HM-oILx=*v65~z>$L8S&isKG{*Fu5vtIx?Q(
zM+V7xwQlA#zj^?q4Y&TZhde~$hfg?|RV6GKXgVU(BUmvUt_;cslQQ(FHhZk<Q8G=d
zv=aolc-1F3Lk);8WE&7)@(qYJYCx=21L7=qK)lBs5Tj-Cz($DR&~YxDW1t86<`>l;
z%v0f90w;dfq>EhUIK^pU6_&fksEzI@T^m%wa*JP*6|+Xe|8S(B&7GawQLT<se@kzA
zQcuUo`Jg#gZ_{B8vc$uqQ0gmst6t0>#`ItmD3q4eB88Av(_@1|<B?IjK$<QzLAH@O
zhbAOwY6>-)kMZCOwEv<FSJ2<hdYb~j=87Y<4lwH`f5q^u1XN~KDJ#_6_KY>s>$xk}
zufG&zH^`M(ewWT;+&~^Gj;`s>-OLiQ!%4;{<`A!;fN;D`%bc`Ia1pQ8R2M0;kz9g=
zgmmx6`l5d;+o1fmZ%}ShgYpVBD9>>R<tuni5~_X1FmV*W?gu3qrsYa!nAk`40#Sne
z0jD@bG5YtK&*Ec~?M@8YE_M|m9R_}wjg@}vW2K^Er6I*i%Uo7kaDc%|kON8v^d#fS
zw>4JJQ?h>pg#4W5T-qfNHZ~USN;yyhSh9-|B$DijwolO(>?e{8hqrB$bV=(x8nXZ<
z0sA#+uk2@!?@)rWVS;BbS4u%oWvZ+T=aI|HO*;E`6`c)S#|sI)A`q^gMoZE@qm?tZ
zh|82WK~aVz5mKnj{X}8t1jB?gBRj+Mq*tmpXkwV4H0PMWunf|wr793}*EWZO`QdUn
z#^rX+Hl7>HIVoy)>IqWluHmApsU5BrLbvQ1nf`f;m-cn#ByK|$GkkAQA(aWMT-9lo
z31QN1V^bU=<er}+)l*uFkx`9$TI%v*BGp234_FbNEk=35NFct~sp6TZm*+zF?Jdl%
z`Hk|i<Z9|AVK=kqo!~5`5I&ksVEn=-FzQNROeleIt}8G`G53&NoUJyZp|<Lfqy<XT
zN;Z=is{`@_p&&9pDaoXCoj<L;EKY|Zk7cGzXlIC$CRAglXJ_Ztl1S>To&38Q=06Fo
zjbzc9n5a$`O0+3X;qXNiaHZJ@(^k1w<iwqZ`IvUpr2ed4?MgWc&R0yj-wkBN@}DI-
zfL}{H{CJaM(wh~Nu5g%?yb$llCdLE!b+<NPy58}pp`6%`3|1?;9G;CXGkkRED7w5<
z(dB|3=(0FFU2K6QwU=YFwdZ(Wd+t>2d9`ZK%I@1^J4c_G4cy5-aCa+k_b70yrV8$q
zY~Y^a1E+oA?o;5_IN&x1)3Ah~T=DbfDQyEHn|5MLlwq3;`*oV;$k`5S5aIeGR~C(M
zvjagac{4?1+r-2W&V*V;ASH&h6D*EVAa-BEl?_kyAr~fmA3e^*AI0c;angJ*u1Mz-
zHCZ*Fu(FyaS$4x_N7+X1+W(0A*xa>G*#eE<833K-9=+=5OF27ZFI%A~ftHx&J|TLI
zlzz66ioMTJDPJ*57+hv)UYOQxd9^KR@bGGPR~5!C_EZy`wwLt{r7L8&Fs5$jzkxNt
zl<oYvKG8CdU%;jmuCOV!MT?f$?fks2P`#Z$z^|gvb-K{Zkq6$TxmBq{++lez?8?w%
zgf^r@#}oUt)xw{}c#)i4M#<tST>`9V265pcd2KQxQwMX!cJVd}%Pnacx7*Q;VgebQ
z>!al5tW}_U&Qp|J!B>W3ghw~xKnCu4^cX1T`=FeEVF={{1?A~`1j<Su6dm5-LIq{j
z9)Ysj2jx6Pl{E^=+Nnc{aiY1%2jyaZ0lczKLAj)dQTI!XI>qYvtC=4g;h<>jYa_-(
z-uAsQMoQOT!lNbhomF2;tu<!dV}`WP@H!PQqQr+)aB^hInN+fUT&?kY1yj%&zcGDs
zm9I}Wt3J6}^~t3@^vUd~^hpS*#t!4bCCrs@Uf7zgAq8JUwyA~`RYNXw8q#@9-zuc0
zBKMJ8lk7yaXeKipYmVb&Vyo~;fjPWA$NAXBbL4SjbI)9Ui#Cjk@cff~&j!6lS;l==
zMi!X$jEOY+RGRAf^j5Ji6EaB##xkrvjJhBa>ZtP>GRF=1ewAyZtue~?M{%`U<SM4!
zt;rR^Oar7JQ3fF4NHtPy(uoi0ij9@JTyJhxU3_h}E^hj|xTU(dt-AP`P8WOT8mz<5
zlX~}Z;S^vPk5^NS1mP|~`r%e~6_6bbKMD{|wyvOBcU`vD-Qa88b5-kZRIS_Sw660n
zeXy|F9hN&kN1cJ*3mwL)bv}Kkw^N8uo9!VIG!&wPHY)p^0_I5;&^zsoG6g>x&gZTq
zzquTexv~AY=CDfcy#LSMo50CcRQuy$hhdXdMARmVOk(H(T!6?nWFe7ch{+(47<+nV
zI+^awbdTLLSwvJ+lt_bH6qQW@*QbcM;l2Qh3yX@1%fsh_`rN>MfBwJUQ_HPeeb=7J
z^S$@~<&)3MoqOxn{nn{dr>ah!1t$3+PUj!9a$I(XLVy*GXvBpj(H{!9xeNWPz2~v7
z?x&5mpDxYTPp?etr^{?VwQWD0=k?Rtg|5nys%|4Ns8obtv__I~66AbnI;qwc_55g+
zsl(~f4z{E@XO;HQ*zQ7qkN7K;Nl-ag>m%dSRL}RQfxWPxe#7QX7o6JCKq(b=vem=<
zO-R>@!3)oj!scSDSb#}VYf3#RZ*?EEFtg$$fF<Rz`9kojy5lksW3v&B|8G{<V=$UM
zvPwK2w8l!V3ClBY$i_2oOyik1S)O^b<(cz6p2;M+X3}asBAT{!Z_U=a8`4^Lqix;W
zZ0nx=FKXQu+q(C3woWN2G@%o>txw|13ES42_&iBx{P|%NkhQK{$qw^<e92g$N9=HJ
zNvrfLw$i(JC6#_4s5HyAm7A^FJ$##C+G_Xm%Cs{A(g)KZea%Apx`p(iE+BowLi%P3
zqz|V-y5B<jwuSVOE+Bo^Li%0`q+8P<Jzyby-$MFm7m$8nA^k7~(#O&u{n$c!&_eon
z7m$8xA^j``(kId&{oF$Og@yFVE+GBNLi%+Iq)(+mde}nxjfM2-E+GBRLi&9Qq|c;5
zYMhB7E0X1p7Sd<Cfb?ez=`SgeK9>gRZ>&Hd{oO*ktqVv&e&_#6fpmKsq&Xu7(k{Fb
zQF%ufkapwSA}V+1Rf4Ged>W)ZEu_6Hq%U*<X&(!zCk4`-X^{4_koLEbzSsq%AldYR
zDUiOD2I+avG97o2h4kevARWTDrQ;4wf%KI$NQbimVedy;NOyGsDWKA$QXt)(2I<ij
z(oq)DJu`yz*c3?jra>xLNJm>pU!4)8$EQI0S{kHdETkt|NMG*)QZL^YUYeHz>Ao~b
zPqL5}SV-UK0#eaJ>Pvz2%``|SSV$*YNZ;xLQXsvbk^<@eG)PagkWR9YzTE|+Q~0)Y
z+`<${-${dXs)cl#h4kGnAe~_$otXmZdufo)vXB;8NdMjiq$L*8(iBJ!q(NF{AuYF%
zzTX9;b1bA4DUkjn4bp&xbgqT;gKi+LGQ9N66i7czgS5s%8nlpp)CHtq#62qo(vQ<1
zt+kMzZ6Q6_1*Bk(TbBaqCuxw@TS(8fkbc?)q<~68DUg1a25Hzr8nKZ6vkOQ8l`1Ka
zex3$tqlGkPA^oBYNHxAKqH<FTq+g~%ny`@S7SgY}fHY|#Jud~)uhSs4ETkz5>7gzl
z1s%6J1=7Q5khWS#+bpEtbO9+~?+a5P{WcBKDa*hEve3QALi$}7kY2>Mg_mBO0_pc@
zkj}7>USc8rp$kYavyd)Hf%L~TNT*pyudtB*)CHvN7SbzIApJQFQkxY>$6anA{iO>?
z0einH1=3&BAic&ydaZ@@w;4g&;X;Zy>;8Khr1x+J7b?BMLRz0+PYwOC6ZGuA_|@=1
zytakw-3PXCh}DAJFgQ4tj7NpAMGoxDc^YTVhg1<7G2oD=(JU5vXyB;s2L2BfoW_;M
zt=|bwe@xUHRD|$jBjq}s4aI>9XGC~WVs#LMSW2MU9=RwnUs$QplrV!%J}B#^*b#|F
zq7Kq(4lmLTlm3$xF^;H=in*H9tAa31HgGs6w&*Hj9Wo@CU_N5uRxB*vpuG$<FdH~c
z*Qm~tS|cFF$QVu;6=+G(u&}_BF!^LWS}6&X3KU2Ik}%VQR8P9otyt<X-ykF=ae&Xr
z=IUr=bU~<vx*|8c8&E0wmKx}lCf;d*kQuNUFCg_c&gZ6hE-Qz6rY1oqXRa2oA(Z$-
z8_*Vl^opEOrw~SUUsHm|7;oC3BQ?k1d{L?a(Qveg{22BO(t2ZeauMiZJbjwNUfHOT
z0?1Y=fK*53!X!~;;bhRv2FDGqiI=o_F}ecoCk~yY`6?8|Xk$yAk)U8tBGhprJQYV8
zma$tH)e?)-96zc!A2_rpGTKn+p5tW&<5&W%VYq-<@p)sEOg=6bczlUYQc+@>g<c5{
zB}qk4PE$%hY{A!UxQWCZ^OVgiV$12#WFiM;BaB2R*-#zPBY|cC41%bKR<D*kW%)S1
zS4X@eb}gK}${Y0+q{m2OVggYctE2RTQZ7@DKN1Zuq=JD8qT#gwS{WWNU{6cLC8bVv
zGmgO(j6Qv|E|oP$#!92vMT9B}{5Fi25*E{Sn30KuDo^8SRrT^Dl08x=h`g4M7+Jtl
zh<sWdDUai{9q%I9i4(R!Dm@nIVkCDbZF|W85Zfhs4^E)}M@xbrClkPJW)tjlCMPw6
z)RCEHf~N3rJBJ(-GFIx1>5UYh{3Jy3$$2<UK*>wbn=aR<s?QhE-Qi)xVzgO%$}sxX
zX@kSF9)YX%0t*{9q$z^3?7-{zo*?JshX>V(=?P<=qI2^JJ=xz#&^98FN-fA<ylZ9!
zzALxDo+`{VJ56=2cE}%-z`V8waaKY)MpRoS7ilIk6|U44=p$p*DF_561!9R}&m_Y3
zfc)721jE$@p{ocGR2nTz%2IC(Iq_5l2}4mO1jMid@sKFc`BXkt?142gWKL<gT5gqw
z5jPV<)~ymLOQkm#FJ1+vX(73!sus@qe5H?(WD4fJ<M0iMb_m<K@^~2z5}?3>A$a7`
zRuL-_9jO5%4FAh&@sW5We=`?a=KLihFx(tw_)I#If0|wX$(0-l?ZTH5j^uYwt8^)|
zIm(*yR8}vQ?h#a)`AGhBt6e&SSEij2NAi27K|0H}bCHF#R~L|$SV&7#Anlz7X_<wz
z+(O!?3rNBKWJL<3o-{}U7Sg#E(!O0ldM4kNz3%E1Nc*Kh8nlqkvyk@h0@C>w(%KYA
z2c$u|z(QJQAsyHSr1cikb5kHaA`Q}zg;chX4(bBZh=nwo0_os1NE<AqjTX`&T|lZ@
zNVOD5ho(Uqw~!_*q{F&^)Uc2yQy?9l2B~QwwJfAZb^$360@Epwj!1)al2zqwv5+3s
z1*C0!TZF*#Qy?9g2I&Q?K<VHtq(^rFDWKAeQXn0b2I*o8=_MA@W4eI!GQKTTx+DeC
zW78mgpx;pG6&BLtx`4FZLV9Hiq(T~`HY-p%I1B0ME+D<iLV9%yq`7I3UTYz}&O&;8
z7mxxK^XpR}Js}O!l@`()Eu>>+1nJEwke-+Z>1qq<8Vl*zE+AdUw}rjmk^-qW4bt@%
z(hU~Uye=RGLf~yFkmjdBTDu40Gz)vb!$Nve7m(h?w}ndYb|FRRa~GsR3iP?}wUA0P
z)#n!Z)kuBr4H5d>#hwFaibi*~6}qlh0ix5C>l>$$msag_u?y3W^D20uBE@Fe)UC8T
zPHj2$jFGh!xIs5cQ0ytEp_y$J=8jB5MJD<XxM25G5qny07?&q^=UNthS+bH$y)7*%
zMt@Npmhh?I!zEW)Xm<bcO6DY8VNj_hrc};7w$9CC6oRIwv^F_2iH&w&>1C82HHzSc
z-p$pr+pU305Um#lb4l_D1uB$mZC1getcIHHki15PUVM%Ld|4|MeWEx}x|$h$h@O?x
zs<FYRz@hp+lrrEe=po=avz7TEp#JfJP}igyEY2dLu{SCwQMtmuM&ceK)ds$#&u5Gx
zD#-q0+g#bI15l}vBKnlk%CJhcK!q!^1&nNBJP92bT}2nbSp(GmN}Fh&SUxcfqBUGl
zCZci)C#e8{plpXYn(Af*inBbea0ogJ?Qm){;v+*#Fv==rcp`1gGrm>iAS(RyyHxIh
zEW_iCkxkIXJs-}q>D_9gx1lIPtD!?wv}vPnRFS*1)=t<?$&tAJ61`>6XV7vAoR&Ie
zVK7iS=t-b5%L{t+HQ@3qaq1lXX_|}^802yFrc?}xAntmSo=A5BzN>QwtQkY3sIylM
zf>3TjCp=TS$TOjw;y;JlDc{w}C4%eI>x{l>7;CKtjt;5X)S03)pI*9C(uV0K{$%VD
z-K-%Vnr&%9HLJw;(1M;JN8q%WCJO$6h-!*N*BdJc={CfuSt=7dwYL~mB{;ufl(zT~
zigk%gwXxZlo)i~gbsO6Ck{%}Fqb(iA7b;t!1Fb9ZM(1t)pl4_k!h4hun7%R$KHbDx
z7*P*jXWUt4XHXgDhcwAHR4a%Ty8&u#=|6A`)b=v?AuGZYW=m>9MLeo3j8(=bgS_<Q
zFb4Vy4u=)rnpig~(LEE_^K8R>LZ3EKh~gZXzRvn&nRBxlDXJoP5+hLrWIf!zB`XSx
zn0gE*r<)PfOLD{UxTrh~|J$B0yr3(~p{9T=;eWHDO5T-gJfmWbP$eIqw(>lI1!OTS
zSEBVjk@pf*$tR^%da|wb6kbWC3xi5CtCFA2cVvZqDz8jCBUH(!@p+QU{B#zOcAmj2
zfpjLX@<sQk_46HpbQZ4?AT8qaBuIm2p@2YI%%TEm39oWOTFQ3>(%HO9fV7OylOV0K
zke0KkKzatRazZ+X?+Bz7yh?zylFySMt+tQ`SX3aL%d4D_R`DHy^h{nQKw8b`Ns!)R
zRW@r_R3HuVDkr4#_>Mq&7OxT@ozLe<kgm6o*0QKTdN!|eLb`zO2&8qqN`UknK2L)5
zJ_~6*iwdOY@+v2!VCP>-fi#o`>2eFHY#|MI0cnKq2$e>8l|ZEmpC?i2RTk0)78NRO
z<W){8jqx3UROM9yq#B<mLE6LuBA3Q_C6Fd~6$k0knja)md~@r3M<6wLl>ligpC>`u
z#sUKA`MeTH7xF46q=1)Rz_$}1UBu@}kY30F0_jD(5=bxRRZd73^BtkmOL&z4>7{(0
z1nFfgAdoKMl|XtquW~{PsPqcHodD@lK2L(QodpEaD|sc5F5^|#klK7Z0n+7so&@O%
z77$3U;*~&pHLr3)dJW%^j(aVy5+J>f&yyhSU;%;jdR_^nH}EPaq$~N3Kzbvu5+J>a
z&yyg%nFR#WRlE{NSMw?-q-*$&K)RM!36QSi^CU=bVF7{kR$d9D>v@$EQo!Cf@a+Uh
zH}ZKBq_?qvKzci`1kyWrl@roC`HoQOUA#(w^lm;+g7j}JAdud}D}nT0Ugd;z6W<X?
z@8eYhq?`FX3DWynKp@@1D}nR@Ugd=JLB1o9KE$g8NFV0&BuF1&0fBTYuLROZd6g5=
z$M}vw`Z%u=Abo<*lOTPP1q9NkcqNcN&8rNMUcdVsgz3RQoqtm1-_P)6!S-2RC1Cp;
zpC@6vjRgeT?Yt6fckn6$HW$+8`Hn#P0<RJv-O1-kkiN(Q0_jV<5=dX>RZd7>;X4B9
zE?y--x|`3FAl<_P0_k2}38b&`Dkr3`@g0HmbzUVvx{uG3Abo=c1kyKoC6K<wtDKPT
z=Q{%F+q_DE^c_A=g7jS$5J=zSl|cG;Ugd=J0N)Wv-{(~Vr2pXaBuGDC0fF>GUJ0Zh
z@hT^zAM+i7^dPS^B;`cW9uiHW(J}=|9wfM+)u_WDv00s@S+n<VR@6mLc3{<RJa?Bl
z^Df?T(T?7+JqBt;8j$>y&yy(iGZqj^$(2y5>>X6DJ((fHRzt&x5Y}>vfyl13&V;5=
z{zh&+m0?vZsYRt+mh%FoW;JoWNP~r|jOMPj(8Lj@whlT=75qy=AoGsAD_ZLU{()L@
zJ%2v5qxiUqlxz!7tE(RIJff+<CLuIh+fIyr-V=u2v)>_*;NMI;QG)&*86p6}<w*<O
zkba}=k<T$I(!=bp@ELGazyCI^-+yQOU9P0xD}KM9%XDXppdE%mJKTgClQ2Sd`c-mV
z*DG6akVe%okNA}4H*|VRoAStRZ#1jOa#|jjBUI^y)V`MwL4}=x1EAqm2U8WqM3?ta
zyMm}qwf**34ihY0e8sfEJ?o792<V2Oup7z<?kUYClSQQLAzPEgVUntI8lGc#eNt38
z-&Beea=7JNuL4a2(+5tkGLs3_<-+G)p0en1Im^RgwE-0{#sbA+x=&<eEKp}S;ad(-
z7ZylSoi0Yjm(w{J=DetL2)z?y%M+cuvRIPnyqlICB0A+thR|+&DQ=S4?x|HStwqI`
zj)1az8k9X)K%mH#K-sffD7C%#rkQ>tCY8O@;Ot}J$d$m^=n0p^n1YwqC>X=ujQg_E
zu;dE`&5I6Ddui=M=-Tx=YEMFM;n!7|$_8xXintl>c21sm*8YN0W4E=B*FnxFtCJNT
zloh9^s^hJq@?<Hs${SQ}LCjq`FU$_r3Ck}A)XOQZ<?0Gtp)HCndes-MkT2BkWFhG|
zxsrE2w-oojhzEenS8HG9kI>t9<IdP^wIEjfy(}VcU(L5kz4beKYxm)H??6pre&5Y-
z=-<GNsT~*ZFi-B`&3pOd{?L=!cUd8;85ju4*B;>A@8gHMTM>MrZUW}p@qD}Kshi}F
zGePG|r&Y@z#2sJ{;*kFYHw1h8ti1F#`Fte5j|8SyBm*SbG5w5HbNx1;^R`C-5fVrj
z{M-ULvuhyY>Q)~O!}w(e7{6v2f$@-qv23O=jCE2N&TrD;U<?SB-&riDb&Z9tIAH*P
zNC!|0yjT8YF)ZmC11ByVQ;WG-2#bmC-^r(XjQy2g7h3$y;HX_2jInM;fUbECmg}nK
zU3ev<_bpz{wcYqA@aaw;Ub^zVY~_2~%C8S9x1!-&GWPMq9p*U(f9(-n0zA|LJj?>T
zF$Q2r2IT?wm=17{!<}-Ph|)BiYau?~LVSAwQ3>Z`IzEvi|8(W^Y~}N9<?r+>U(jXc
z$J@$Ju$8|%s9g7c9wA>lsY@tNw@^;CP~H<jQA9l5R9^cL`H?%F%$$Cf{lFsoft&mf
zEHNKYYrxEvFSnIH!&ZK?U-^nID_><Rf2OVc7QgZ}T~>a+t$eMm{DXew7nsUxzsybd
zAhA}-e&EBw2dsp7h&4YPmM{ag-|+5l`C|lcbBUoA6SWNnQqG>L@ja2#n=G7LBjEgj
z-~6Kw=TE%*XZ~n};WWF1v)RJgV&Qx&0?yy~&A;=<HvM6<H2Vi{{uh7H$01mR4b%ke
zKl$WE`iZ99OZcfSCX1KzJ)zwzcqKE@Cjzhr(XOZB;coaLb46JV-OIZC=7fzE`zrg*
zPj!BCPkwVR{xDR~-+f(|-@TIWNn77&zx$b(@AgD|_v-B5t<<hV;qA3IaN%$%y>Z>;
z=M`!P@W%5k9$u05?S;A5IsNeR7eDYF?#{t(I@HQr&pg-IFJ5cDSi3!-fE~EZpVaNl
z>+cCtrrokjsW%Z-C#Bj4gmMTs^<L}ycCU3G+o1MM?IV_(Em@P<iO=1@Bn@<i-=Wy2
zJ+KT`i7<Q?osF_8)V6_8VZn44rjST7QP@(J<b5Mr2UaiipS4IiQuI{ou=A{<;0SDc
zcO>u6+K&ulFP5-pgxzOphZ=03k0F1pj0eujOmF3e{kyv_Leodg-I;95o=maA;n9Aq
z2ieJAz&yPQdHdYlteRHs75KQ`zojkC+eg3-6tN?i5+|HRSTRTYSnp%;#YN`b8FYh@
zd5cNe4TiFcc{Fp^HnMoGwaOBcDYJcqhs;qS2$U}cQ0#!bl73D#gT)|hW0PLn<li0V
z`xE?O<fbV<k3C%H6P}y9^NO^!fhX<VsX*M}K6^Ufn(TZlS|zadT!30tY0|P52G&|T
zq6UsWK~Q0gbleLX9_uX{G??^2k?fYin@I2tMJWO`ZRi=2tdRP~5RH3je+-@fV3V#o
z^}>*SP35Yi0dsKb&~u#qrvlBuvOk=TOH`&SRxTZHZ1J8H8pBMeWSp1q=3=$4bYIc8
zOxcJ;!`hd?)ApVpMA4CV5UUEN&eoz`oGjQrI0J;TwX@m99+RzD(u@TPsgj~o>K9dd
zvVxxuJoF*$kkmwG7wn}=8z-kztJP%-EDI!!X_&bq9gLG}UR7FvG}TNZS{zio+}vpP
z2e}H3rMJ-AsfFy?gA*zQKdHA&5BErRp3pd39hI+sX1VHTBTv0F2mKG;P3gzegv(*7
zPu;%C6@|<=(>Q%WT(uwxJdF1-*XbPf@TQnTz8j2_u#7C`$xojLqVy>+X2?ySBQL$q
z5%WCCpgcmi-4hH4JuN&c#1ze<TR^1mSNIW6i7L`}3J*f_@j;v)rv1Azi*EMsisaqG
zs*U6g&D6E6e3YwchexL;*Udb}Z)RjJbG-z)n;(yA@`Zjgq{%Pv?@E&|^6x%|<awcg
z_fD?;Uxd5uU7iw}WrRX6#)B9NUCc-MC?vB*AYJsl5ksN4!on2y^5Y(b*3p>#Wq1%`
ziVN84OU&I+C+FMMwh!0wtBi0=h96g0;D0TEZ<YejCC*C?&hX-9JBw!{wbtm{e5-AL
zd`WOCt)3^A_A_R>YIJLiSukE&``Vga=D4RV3A+Nd<c;4osD0VVKJ3lo;}e1KXcmJ>
z%iDFBeaU>khG9;n_I&9q{IyCSHtn>X%2nurbB?cF&0mXVv~Ovj<TuABh%bRaEVb?N
z#u0g^eWb39`8vHc#XCadi)TmequkPZ(Pftk;qu%bwIAn}4?4<w#ST`X(g})63)7S;
z)3vi+VtCCz-|ATEbD`@FhUGpH(@q(S9#x$R6URbr2|$K+)*7p9A0sz?R&c7N^lWYV
zI;!mocY8lZqZ1d{QWZyT(t<~O#0eJ>>9b;{1P|Al6Z_rV-`-ZsycKMxN*%Aombg3g
zY-P@<gr!eaEPYy-rFURs8g(q9&ulg36UaY8Wq)arG;y{&j_p#UL|kP`W`tUEhHA~3
z-L&TEv8_p(M{~C(uCj>MoMnV^NQWESH%h;F>9iiwwdb%wCH@%3kI-rR-IxmX9OrRM
zsni;8Otp+Esy$<A<W%U4jP|a*^L98#@tw1C?LLcfH)z?ksM)JLJW`ZD9gB$B*WSs{
zxK)w=6M9apZ3PvTX42AFlU%8EU&$p>r=qgvC0G=hb4G5NBkJ5>c5?N)wJ#3i%OQ&p
zEy=w!jPmK#GyHT>Ew*UYW<qj@j?n#_jW<ESZ4c2!gYyzH3b_3G!!TWn<A_MEvZ{9;
z?$^B3jj_nQyo`b@{~FaprpVT5{oo~Uj0NRoW)x7XX4;o3DoYR(JlvCB6319*UWP`2
zmS1pq^s=#S!<f7*j<J}$6pe!EoLM3lc|VNGOXL`f%FEIys0L<=%Es!IK;~g!UOLBE
zU|z1q+37+XpeT&V%jX!2$xGHKm}V@!V9P4`l!JigVQ|!sD;s~w>=+Af&WN5uYm&Bc
zc6w6D2ovf&3~V=kkj85@$?q5o%uC)Vz&e&(se1}<Fba!>@p#D{WAW@~2_=~u1<x~Q
z%<@fy+k!TGqQO8ah^G_A7PtUPmd99ZUJ6ISHe(@yNLePfdl;3M=rI=6;g({O#Zge@
zxBc^akbFOi$uSHmAe1C}jD_T7ahyF!oUh{`NutMCNCitJ$>J!GW<2STO3zEe^mu+X
z7Sk~nlcaGJOp9mA>xe#&<JBv;Pt<m-UY5sLI9>`zfivTEjKtTMpz~o+MYdh!ljM1f
z1$Ba@ktA^xQ2Dpu9(Rv6riaJTb&UNVh84(j$@Lfu>uDC2Bytp3OJ~}3vKobHg4=ms
zrpH)7UK+>Q={x~c9-kcZ2A!AdF&5NWmP(6wC7{;K3YDgy=7U8pX29w3ipTFu_$^^A
zN&6U!&dcdIJFF!n8*MfwalRJrNiX$dEUW=bG|B5Ium)#JG|`HbH*RcJHo_l^Svleb
z=`#r+7M_>mQQ$pmrtm08JJW`Q=@$&mXC(skGCs}@)5$t;yxbZq!Ceu-))4P2j4qfO
zB|l^=`Ml(hg6=spC7+4RM_U#QiOP71K@?Tukuc1Fh>|lh7Ur;J!Vz8x%o*>tM;j#!
zj7o`FHfT*15_ACpD~a8^(IS()kb-RGObIBkCad5<^lTWHmp?L=VqS7c!8PN(n&7HW
zZHfZsWs!^p=B0=fV6(D5M3lXg!?X(eaC2fGdMP3W(u^l{ttypdJz-b@(@Q4FSW0<m
zA_dlrbtcK<(aR_q3(ZRxDbQv-9<|tjER!rhVWN47C1a6!StDnsFLjcj2v}Zn$yiuk
z=174x<GmV>ZZxzou>!`I{F1S#USoxqB##tSGhPryq(X{AC#fbo(cP_1m4dkAZy-*h
zt9{cs_`gDJDnjQAvHmI{`{g&Yz6fQvm%2#k(VttGcbwmb)ZM4RP%%>79XN?eiX-`W
zeu?dWT)YV*y~S!?E~L(W(5h}P(a~%J>(obj8IfXvdFhS>Sbp~Hpjbyd^f>li!LE!&
z=Vd+;bop1N9q58+=RpYdFhDO=QY=6(50U^pZ^qni8BS~GJG<A*n-q)BONu1;@=wbh
z1e9pb5`{fXHZONlEUY2+IuDL(WnKxa{3^i?SWX;e25t{y^O7jVV)HU23AR}n5goDI
z!|1$xO0nq1E#)LxlAz1467Qg#AE7-A&daP63$AIwN!la<mw$%t0A~ZIhfxJHv?N!G
zMdf8q5>&G=27>7AVOU;*rC3;A_9THd3uC|wFCNAgkW4Zy#bUeIcBG_H5^U>dF4mm!
z>Bl(_<Mh%m#p3jGD+$hZGslU5GI0LUNnG84>tU2$My6PlZA;C|c_k><&K#u^@x3(6
z#M6>WAPmvV&=d>twHBhJUlNG3F^_cyrw`)`_HmM~DHflXi%Ia!!eH@2-p65+yiKv#
zyrfKmZ5E{7R0RiIozs|?x+xZwmzQa_D5hq9n*kzBu|T{?4yRabUV<jUcK*x<L}Z$X
zFhVb-Qzyc^{e6?<Ya+th{j;)fdOt^6=f0`EKV3E*^B!3p?eS70FHX-*9<RQTxfGuT
zNuWo@8V%;@HhwiKklRuOxS5)+R|0<<CWu+0vf4C)yMeQ#>s8u*B$fHbkXg)IOX6E;
zEbT+^im`=-^55x}@~$6g`%#*NXlJV!`uvVihunDv4KmwLT)AZBxvQUDTGPLDNr^^s
zrR4)lm#khguxLqX>GFZ)YnGLkuIN8owWoHl?Wz2Xob^LjsLeVf*|d*7HsP(s=MF46
z)>zlk+qi9_wWwo%3>12_PQj`=V;fzDRrIMzkSpO*xspNLVgH?KW-#8^h!jSlK{`+q
z>$}G&>$}IuxzB($JpB|&a9hm_C?!nty-p^PdJ2&yp{|l+x>93Mg7*1+V@UbZhdkTh
z`ab4Z=~Hfh=2xpZ(x;fH&v161w6HHe1XOWy+m70`fL*(eKdzE^Q*U9PJh_{yrV~Su
zNq8yg6S<N=TXLms18Wf(M9!hZ^!3E{Ltv}jfUAMpjr{v<_yJ;jF{z|diAiNDjYx<$
zkXES|s!A%AP*tkZEEz;-Fxy@~YOLoM7TodeZz6QyH~1Qsk>IL6Dly}5^Vl(!{AvTz
zf*-q}FrUH!zvge+!mCi6xV3E49iR!Kg_vk)5t&LZNNc(m=}MZujxS|zx*yCc0nT&M
z;CS(_1kU;{;QW#-!!z0-oLU5Ro|^{e*Q`cZSmIs@oKhEXoS;?-)EP>F6N-B!Sjt^s
z`9mhs45wi6LtqJ<k^FFs7rh}%#PA(WgW^TO5-1V{D?1N!_=dtl)Hd*|-T;dLH<kv@
z3#29BB#@Q^uHEigG*%heL>F>#HCK72jN>^XKIn8;1+0@s#wsni6PK%4B_XD13wgPi
z7;$Q>z`RsaMZGblOR7MbK@7~Eo~%J(qazk>7q5*lrE1wIHJ(N(FDjT&N}_@}lv?|>
zP}El~SPRPaT`d(6+!wJmap+WTZi92Zb)Yt7H&!>pmsyokbW<s|7K#_~h4Qn2e{*RZ
zJ$GSY@Y$=DEC}wJP{1ubT4!(bh2T~G#eZ`*RkoE*K*4AAuQ+c>>4f45K}D*BE_0$&
z=ETl2McrxUor}UnQa^lRh7X@u%>Q8**LtDFMbXi8*&`)%l8T{=KU6A5wx{3eE<}tK
zZ*Uymiw`-cQ^rq#(~!x6h92cx^JQ+7>P&97pvqN^mPwOmw-NR|HJ8S`c}`^;R$8OM
z?gOu(McQ}Y#)NKF#y6<AxEi41zW_mZahQ{aT>Ei`h6{yY$*Y5@i!%wY%`BF*+VbMy
z$&i#dc+QY)Piyhwkx3H`Bh8FFJxbjHxKl~?C_}PjV!bdMv0j`;thuz+mc@cZ<P%~^
zL_UXDYoAW}rT#_I%Lpba{VFY30=3+cpk!15|H5NFT2`^#+<1A_Sq<;-UlzG#%caOJ
zmUHC<|K>kWhVfRoO}xW@9+*g3X3GU&BIOy&Wd!3FQ}T@!K*sS`e_?Xl)EM$IlFQ5t
z<aHb68T!zsG1UTXXL5abZhpzdnFPVhSS(2pcmW}Ww<RE?!`tmuCdi2s?8CC3NK(!D
z>euYAL|&e)wU?%~){9Unt(6FsZfhT8E21(6aIY3Kzx<z^nGQ_ar(Th*OJ1GUC0?XX
z=@N<5>2}FF&0n5m<#b)tTWz~yoBb(TkGvpCXzjJx8owj0@n_kYMWTpG<0Xoy+xU_6
z#+!)CJD&(Tn*AHH5#Wt!1n?rA3IQb2sY`%YMiT(4V<`0|5pzh(;fhha7;+T0j8&2I
zm%_T)7*`XnIE($a(do_E=yY`&oxGr}LMI8@>e6X_G@WFWaDo#3_<vi|ug%u<x1=@Q
z3n?p2myoh<)7K_6o!h;g-|8TiZSDGOEx$3X<zB2?X}QF@bzA<?o;78iAjj;g5VWma
zsI{<y;lUo~=cw6?&%}9IrZi|Gg2fBde@vF4c+lbBSQKNaf*c);0A{{pGTGBdphF)q
z8;gMU6D&xbeM9f)sW_opm5sy^L||B0dv`Y4zbB3MUSMCLy#)4kXutM}P-TPs6O$lM
z9h+H|O9(lcOo}4AER61gx{x07D@=OiULQt~P$WwUszgsnAhk+l<q<nsL1hGC;-0o&
zrFseNc>N8#`u1o5D75OEmA*c_z7W+T%3!Nff#RG}{j|bUP7L+#E!leagK53%#YvXl
zl{m?6?|w{rcXGUnc!&<y16U)-vqe=N&(NvafJP%W?U0<ax2n&tNLMf}ELy6w<D7PU
z@x(9(z%FF6(WI_R?J$BdCz<HYz=Lzis7lJC5JW3A)M2`2#Tj#g)W+6Y-o#f6_&0yP
zv~kwHqlxN%MhkiTniY;a)pp9lg7gR&4qX8b9Ggby-pazlaHBDfn+$fD$d{kU#+RQ;
z<4Z3jw(zBd#CG_y{pp^?V2^<QkrE+F;#cd~(N@$JX1v;Rq)vZ<e=CV2WenR{>T@cf
zcUt7ff_|B$p4neR#c3OGg+##TvXT1sG*Ww!z=hNj3EU<11wAI5ETS`)5kzbo;-f;J
zU%<e0WI{+73NdDdKOaScBX*e>w1t>4M3FnQQRGW$6!Bt|3q>SGxl54^DHK_z4my|O
zpwp}q{<YM(yOTO9fz#TrQob)or&i%r_I#YRutyh%r$IS>;eDmsL?ma#<d~{7W92->
zpRBW8LECR_UVMG>A`>N*->^Du#4vr{bhQcLj!{SVB#+q&_&0yP-uTLgupU}au=B}W
z=eT*UUHbNO)kMi{by@twVPms>Di#*26mcL)qY$dh%KHCCU5F!HvJR_DwXy&EcqDoS
z4GwO`)pl|3sI4GJm}AKZQ-yo^JV|Hf1<#j(@>Ra12TH_g;(cj#dg1Y<P6?0i3_j=~
z-Yd<xNq+i`6HA3hE4MLrPqK3SdqfmI1}(ex5Q!xG8edKz;R9*SmH7FhQkN@fu0+yT
z&F!s4(D27;l?K)x53;CKDlzVJ((vMdhGBD$m2{D{gBj~LGuXq$v<z_y+TeI9pkcW9
z^3IskA^H9*HYVItwFsMshtkMngY4HHw&eK@ud<Ve79kl9_J5lO<##L~-T!+FMZ(}`
zha%kGX9+7ZHzQqx5c;)0^Yb2UBUtaRX$`V*6>ER94U%B`PJ`C|7`6y@`n1UZh4JYl
z^#a2DGO0uas4w7OVvu^h72wFgcF3Is?L4I4*P}!qu$VnWMBsX<ge!3`bUm@mEPTBN
zn;iW-<L7h4ve%z=LfU8iM)tt=`V*GPf+b0Ww27c=N0j@H!}5$_?%7G!Qp0RALct)t
z==ck?qN!3I8M7=i8{zJW#8$Y|!rt>M+Sp-cq3^qan<E51tIrVzzl$|klF`H3!?7k4
zL1RU>{M9OsZ9slvM%rqsfE2~B5K?n4B#V7dOPZER%DJGBJe))OC}s<=c_r)<A>#sU
zuuc4PuBH|ij6I5R{o!3F_O_kw*F$5VQXv~8*u$z!<)cB^ZvSgxW6+vSI8fUk@jx?7
zQ3tYE(yH|lEFg34LA;XHkz^cjrl>3{%*(s$H+c)qM<S7h1rZ(~%6e#3!s#Iv78s_t
zC`wgqPb7m1R<g0(q*h8n5`)^KvytmDY2<pWCD-FDxg=XbPI7&67Rc2sZxd(TfXA~@
z4T@w?Tx+GZb~W(fvO3mu^4bgtwPH5%9-l^D8`QaWq9w1SI><@hb5uGK4A1lGP?xR~
z&a&(U{t9l$6|}eO@sE7B84()I8HkhiUY~HDoUNe?(;8~yJ=adP4V9z_IU9PAPQoBZ
zHV)jHO_@+TGXiqNMu9=&6D5UBsZd*FAur}t9OO&A6yw_?Q;bJIk=W5mLMyOhUvHs2
zcc(y+2-ZnZY`FT`kcCp-DNq<kJF$aCSU^PisD)D5DNq=UIuVKu@m<?yp*%l7l-h-S
zCuG;Jv~=x@duBIhp0kq4rKwHShy2IIK%bxwc}W@>z2Mz4IlnX@V{g=k|K(|wdSSVx
z(pLnPy4s$&#hSQ)+%v5Ergkr{OgkgS^QCE!ykOh{X?qutyb#+7kY1Sv$&11*kS^;2
zk{4h*0a7~+k{5$pAYI-CBrmXb0;DU_AbGL31=6d!faJy1PJs04G)P|PZGrTfE+Bb9
zwG$w{HVu*&cUvI6t_w(BIPC;TJJKL|5w``>>$`yDMbl1z^oBG@UchaEbY&Ni0=IsN
zp`8HfjcJg)Fxvv@O<h3pqGu;SdUF~iFTl1yx~dCEUI6U`NLQyp@?vZYq-(l>6nOAU
z-0TEM*QP=8LTd}8>$-sC1<g)?^p-S8hcjlGtj;8^wm^Do7mxxfNzm*BNY|%9^5SX>
zq#I@gNrGl4K)Nvvk{4H7AiZrykR)by0;IR6LGr?C3#4~+0m+M-odD^bX^_0o+5+ia
zT|n}pW+y;;cN!!wu(m+@w=N(B(pw^CCqQ~n8l<P$5hp>l1=4%FfaJx@PJnb%8YC~Q
zwm^Dc7m&P&*$I$tPJ`qH)fPza?*ft+COZMrEoqRvNZJDF16@G!LS!dE`d}I)FN(H6
z`cOBJ7=cBCiAz}Q1V|rFgXBfd7DylI0#Y#Ip2Y!^0O{5=NM5XLf%MTXAO&-rM88ge
z^szKZUVv<Y^zkkr1yqtq*9nk5kp{^NhAogj*#)G4N)qBa0n(?^AbC-+1=6RxfaFED
zPJr~8G)P_mY=QLIE+BbvtrH-9E)9|w`dT2})&-=X<2I*^xZBepd7-Zb(j8qu3fNm>
zS|?EH^J$R0u-5|V3td3+Vp%6Zx-$)u7w%dheX$EjUJ&a9NMA~W<i)ubNMG&(k{7@_
z0n%5}AbBya1=3wzKnmDfqE{zCx;qV$7vNeT-7_Od60+Jrim>UqHw}^()mk7)RO`57
zknJ7EAf3@izQ(VH4UV)qYVSU<4uat&0;0ivN1~V&R)!4D97`dddaEKlEl2i&MzdJx
z8H5q3x*M1vR5n+d<?%veQX*%J!>`Afu)_4SJXIJe*9*fHRIRK{@#2Q+#%Xx^lnZil
ziE4Y)*{z$3u^=Z<Vom|e=oqt7aUnu9H@--N@^S&L@8UmD8I@P1US&}|*`UJ*+<OXJ
zD%Fi+40mEIbD3J9UMwu%pik8`FogIRA18EOqq0nIjer;<W3W?Ipy~Vtu*8^xR!Rb;
z0tLfYkc44dtiuDndZZ5X4MJi>7I<rHu8vkl7lc}<D>5iaV4za;4Kg8S!=@cYXjy{Z
zp`!^MPQ*NykyxAfx*&s(77WvyWzY_$y@pWYj$}Yv2+}KAx>0psQ+b4eK(rVRI?`a8
zY6t=}6P6PQc2nIT))VMc;|$fwMW9DGsbDs0qyVy23Lw>yxiHy$g4y7>o^qqG+8)7R
z+@`IV96Cw!TY_n{v87J4ZhG&KI*z$^ain1xyM>B`X+TsjnIMqSB2H>6%Vh=QSiB}M
zT+sL61oTsr2HfQ1a$!Rm4jz^?g<i29A1%Pp3*|JW5V8!0@XZ#yE{2<kvNMmd7{TFo
zG?|ov*$C5`tc+AQR7dnk0He0((WD{uYROZUkK=oFxazWN7Zj$-8}${wKhl_(fD2Z2
zbe>@seBIjO(1=7T7^pCkE<m&C2m>^t1MA9|tZqi~A_b#Q8w^QhHiKTa291i00Dfze
zuY|>P9rj8hp~?t(S*WUBo-DU8=Fu9A7xku*1uPW=#>BgbT%=ing{9JC5&KC=LLemr
zbf!7V%LJp<)+U6Hnw-=OQb(An3G!;S#a<RNR_cxEjTE2!Bt-Jbc|Ia!k(VSxTJ`xN
zx<#s6V=7e5VyIgZuDI1ZJnIp-S}!~W`RR;M50Vt&b$m~d50Y}_t<{O?iGs>tWSp3U
zp6qYLTpSTdk{<-`npuJGwjf05*Q0#KZ9l*ntT8=4S{M^ATU$echN@K=oi>B4Q{hTZ
z+;Hd^8LLh~AaFh<?-YC71SL-SXaK=*bwTJV0t8NW5magAf<sQcq9S6Us1gDoTWhL<
z+Ysn{DxWI$M5F|9tK3lCD!sXQ@hUJ)YuW@TTd}e*TrIcyCd%NjVK~oGh#ZG+i0e{;
zd7H|RMWDcfA(()T(>n+Wog`~g9X%8NH!EHqFXKYXEdL_BJnl<d{&^`zWcl|EzND6a
zkw^0Pr&a2u4v|W~9aNh6NZ!i`V%iyTB>$Z>NM6zqf%M%jAbGh!5+Hpq4U(5GL?Hcp
z7m$Mei6j6?fb>8bqyZbc{9IlMr0;hD$xHc>0O>!{AbD9q1kw+>faGQQNPzUiG)P_|
z5P|fgE+BasJ`y1PI1Q4Q1VkV`*aajnu}1==pQJ(Za(@V<pLPMs%jS^)>1Sz>yo4VD
z=|8)G<mK>4fb{b;NM4!`f%J<mAO%7|5_lv)`ehm<FUN;K`c)T@yyP7Tkba#8$;<B{
zkRIv+Qa~li*^vP0;WS7WTTv-#Jp|Hkx`5=R=SYC`+cZdC1`mPsyDlJkxi}Ia{XPwn
zm##w~{h<p;Uha(qNPkR&<fZ2jNPp@AQlMg%bQ=kf{+tHM%g7;+{xTy-l4T<S(qGdc
zd5Jg#(%-s(<fYh1fb{n?NM05Wf%K0qAO%7|GHWD2`mZ!dUh)lr^v^CJd8srEqzHZP
zoR?x<mSn&k=yN68hCq^R8#B@8?#i!5>T_?1(C2EON9!t-qS2jgg|1e8$QG_#7%$g1
zP9w;)(l=lirXP!kfwMPFQ}8N{ps8DFcbwXC>KP+zD{y~ol%UKF0?llzFgM7sHy0o5
z$Oh={T(c<2RCVIjU5x&sc!uFq!H0v?e%UDo5>dV2>k5NPP2>zH=NS~dQz2-2N^6rd
zlh|nYm0m{aQKJZ6=-pf`E0D^OH&6+p^`c-F-<jNP7XPn6L(O(bkg`HAK3CZaZ81D@
zsOS^Lfzs7T^szu!lfsUgxava{@``F~@F|qg_f<GE=phimYb%4hK>gzb>By*uHXD_b
zs9dKih(d^4h?|2i>4+Le5fy~;vRe7dRte1|%6qD2dFC-%8CG#qsBlH-4<wI(x<u>o
zRcI)jH9+mJw29`4<rAY=)EX{zlruF(y`IDp13}piHoK%kdBI_-n^rgkorQKdwHfh|
zp(Rk-Rmu~jw4>}w8{f(mlL|lmE|r2HOORhL3qkBDnqsdO!n9QC&}C_TV|fEY!x$yI
zu$_`4as4HF%b?Go<rFwArMN`zF<_W5h^fJjKARgW2MweW@;DBxCMMNs8zcht^1&DZ
zli|D5(8ZTqg*9X4X65WvgZQjluryUc%sZ&mvy4hpR@&j#Vz`&FbSU4|$z^%G1r2(A
zWPExQlCOnSIfe4*s0sU?`Sel*tdd=Y4bx5hi2ymKSwmj>!l4B{Lyo}lvjz?=C=B5s
zCpuXlA40KS9!4}pCs&@_hIYNAhY9uFitKz+*$VR2h3M{so*^$)9Lh4l9KOyDD}j5b
zMVUOLNw%R{foggK)Y=j&4jcoCB2k0Tjo}HiB{iWU9#s~`D&vzuk_n0+`r8MG!-{Xs
zDT~Tl!7LO!+c2NdB9B58=g2tRH7CoQo6Sg36(MF)Sp+bArl%(uINO10Fge|f(*lzh
zpimx$i)T+5UeJ~0P*Xsb@V{A6C4ZCaR<rVqP$lo4w(^t&Yhn>DSF-ZllP~F&XT)l@
zC#}wXSwQOCk5^KsWH6h_YW4tD6<*D%pg?fLV=lAc&}ydH+``?IJ#1W}8{S@oZ#wFo
zh*dN1;vGyEx5q%OxD1aD;`5{}g5*~dqRN#J^$@<)Nz^zX;xTt9-wAm$)*|2>o(9KD
zekO1p84V}ef$!2tN8B3`b=srM5ou7oTxbGCa-nrnCJsucG0c(tYM9-WF-&3{f=g>3
zLf0Za){~%U!Y`HV%yfGZH^Yrn2?PIvh!e=SH&>_cGMUwiDtFvuWy4fS=0~Q96ApOq
z%K%PtzRA@U_-6ZvL9WR#$5Udi!aHj3)auz`PKDIRu23J_mqqj|y4LrbTHD2RB*oDC
zDz{?$P{6FWC?7Fz7ftQu8UccCl8x?Qm5uI68dWG|;v=vEgzLM+F%+vCoh$Ju=VQgr
zGD3;J1=;U<Yxm)Uy#wuCN;?JyrgmJsL*J-fg3N~DJ8j-+@6MnYJ6N(JVB15B5xsXx
zJn<2Fj#-a*YO12qOwbUaP?;Jj_6$$MN(sw0oZttb(F5Z+I4|_p5iOR7=jaf$i>pG+
zG?Q3|%`2c+gc^q!Y6NkD!8UGgJ{2x*S|@VUSpan>9!$Edk*XJ!$H!?mq5JalwlANW
zu`m5<beA3%>Qekc9>dO3ea-Z-{krPwc5zXoNo$kwZHk1emMk4qB$TP1mPs3?VIn0|
zwZu415n~|Cuy>PrZfox${jM3}h|Y#9zSyCL#XmZkp?2jSbHaV!-cxmy<`LENhE4kv
zmM1#_oTED?#K8Mu?JG1YA;X@=`nU+J;bOwS|BQf4u~d6#5jfOk;)ijg_8b51Z+Z83
z;k&g5K)c%a@k24aX70zKud#No)?s`%j99j&XB2bV?s%#q!O8haP?>BX8FMNLoMJ+P
zUCsttwt4*yGlb>O+8@Ks8mI{g{^Z{k68t%QH--df2??-jtUy+cR;>w+8KJcCgNg$K
zA4UjLtT~O5i}kCqU^63lp`z01`KYuPIG}}I&^&TvG>@nfLYv*<X!Eyldo^wT?%x&K
z{3Cofnl@$xoM}d5nrN!juEVUo-6RoR-hC3$NoT5yPaArSl4yPgQvc=TM4>Wq29v2Q
z8swf@H1tL#3cZ;^HEE3|GpJN5S5oOMd`YiFA{NshPOI}HEFg8>$}6c;<~nC}(|%pg
z&YrKjT@)v+xY~!8pfqYC#_a-Z_;4}OI!z{GXDrojW2?gpu~0D3xDRmts(pYjo8?!;
z_UH~iPwG*bM}_)wCDi`{UovbG(WPHWtMo1wkV^07l~gJ-meZvv%S=ki_WpWNqNg#J
zYTsbRVX8&NE&Wy+)cY;eZ(FD`Q#nxEt5u$d_z7jFFwff=<ZK}Ya<Jqy(ph2(s>KPG
zUG_`^d-Ew#4RyX0v8Y+YvI`_+9#XVKeJSFlMTnY=(lHryf1Isbev;NLKegTRGutgP
zd%E2+t-57KlidI7IURHBciDRG4{1I3N859MvOOnryVG-7#{U0C?%(B-4xywYDV2Nj
zZ5U>oEoHK}lO3Wb3ytgPgE5=xo~+utxan#3^#o0*_&q&A9|<nYN4e75aK&l8YU84E
z*|#5$6H1u$W7WP<wzJJrqglcc0uE)MLamohD4lS;-ddh$Ds3-~-degqika%wz*Gol
z*%vQYi_hgsQ>Ayk8tWr%o#Qa8xA;!b_?~L<EzE<@ud5T&GZdz0L}5Cr1W?qPQ++Wx
zUBWq9sl2&d9jEeL13TFQJ0k<I<O(~{y;-4qa}>IR7B5-af8L70(xUzqE6(a)bj}(L
z>FE~IQ!+q`sHGFf)e6VeQ8*rDYFM@U+_P6NS+m9h(s)j{cuviMC!)4aT-PXE*F@l|
z$!BFjwi5wduK-*h0igY;<!ef7`p;gnvj3bVt4k|al~$d*X3!mUi*gYrp|(!Kd{W{1
zWE8H0TbrtrI8JZ2rb-f>v(nN%wbTNNQ80z1KvXpj60LtzDzV!Yp4%hv)I6;-MO%Ab
z22L4oY{W_U4x@;41kIi5Yj;L`t$py|>i*>e%LmRb^{+mAO=+Ni<r3Zc6}eg;RgK&F
zLyirvTZgO6942z`D+=CM;^Bc`)(rL!E^*<l$`8-0#)WsNf``Sfte%THvzg$#OX0gK
z3f}?et{Pmva`^?Q+9BPb#dbo5j`OQ<aUF0hPx3??Tb!IS6uVm?x;qllnnlZ&EIw~V
z7}45Xi2N$Th+0qsk4}%n<tjjQk3w`$6rz3CKqQCZJjcR`QNxEuR_J{f%f83bwIu}P
zUIpad5RfbMdBAEcAM|QRGx?AKQu`;BPf8&yLul=j51%u~ERAqn3cHu8WSXNsd;`?s
za4I?u&T+6^fo9w=x2S$;R{<Y2W^3X(#kN5%Is(1A(w0Ab-Hlfwvz4>4>%ktbr7}?~
zopJBFp;gU_!Z5m$%K4%7v``t6RZOv7FXART0>{+z&4KUyc|k5~p6aWbXqCfN9-?zA
zlN{*$n&O#Agg9Jhu4Om}w#19hcd8wdN^nwMof;}kOhcIpw_oL67vxPlb+9^lx>!^&
zw)wgt1-l8v`Z$_QENA1$1iwkSL(xhgs*U<c+PFs=f5>9t*^O!Om`1`kXx4EOZa=uK
zF^!acTi{biI<<KC4lO=1&bK<r#hc>|9#y~!XsiO4MIs^?80R3<VtJc>Qt0R;`#tp;
zr}=%h{oJz`(04)scWXX`xW};_g}3BrUJR3_wp7ZSgnOjhP)V_HKAUWe@X(<!xhc2c
z-Dx6BQ;Qq<QA=wJen>@aF2MwDh-OW&nL%}dK^!g9Mho+&u)ve=SDN#wD<*|Trg7*2
z1A?G1yp3IkQ6!<YL20Bh4l5NfSntqc{Jlv3nZ&NvmI8=qorPaKYLmvouUz@-!gf-p
zOVf!7#A0VipqiA6Lo!gv?e>l$p(TPt%5_8f4=HPLlXuxTa*+1T9U?ay_M~-)OMYAR
zx=?%1uTG7vqIE@mBgZeO*xXWW!TN~<B?KCTQbSvvijs&X=D2E4$J6Cd?sNG~`l2I~
zNXy!EUz4&N_yKdlyp2x26m#o_UZ{)Op#EfBbmsI#T|8CAyH`#M9QS25MhS`{s@_s=
zj`q%jw169Q1EY`u4MGMCO&I4$+ETv9e*+DQQJS)V=G7_q-<HkDb*SODSMj!(DM6Hu
zHXvk_Efwu{AQR-vrkUhn<?hT6!fME*1)z8K7{uQKz5-)HO;lqcWqeR8=t!J-l6qy-
zg;G3SDo^QZ3PT$!I5;H759yYm5tb=**Pb1ekuJcP3L_U97%2;5q52do8xcVQ*dYMG
zd-SY^ZC0=3zQpT;djC$_J^QtZavd96o=tLiQ?%ir)*80CNI`M2?vax!aXA-XTTmab
zmgKI}ld6Yw+sC)r@1d*+Q|CLC!s>yul}VK1b$2utdR4&qc!pn4Sd2+&9XbgXTOEvV
z7=Hb{7@4Xph&y>+Pk<(-#Vl}rH*C+?73R}=?uG9>!X!*7YvW*=u1oJ&I$NHkMI25a
zAylM$gGvgUFrz_kveVUxgE;eRabpA-C#Ya?F=%d|1rd=rO<76EVIb2vfheX_Kq$1>
zyg*7Cqf+xmSQ3c)Gq8$Hr<(nAL<EjfW67&cJSDh*?9$k9^@=jmGnB*9ghz^yIWoly
z6}TiINhZph@E!7dtBQe)JIxAQ$|1%reIw67Au5{{4tv(db;WU0(_CE0B0v!~SRz4)
z(^P;`dEGh)&Vb4*iDP}7jP4H%3H2aj<_Kg*M4D99nWh;!UW`X&8N`qj(ozU<Ef$I*
z{V?%JFQCe%^4BMECHh5K@(;VKh=68TAgRoYBIq$anE6>Xnsp44B*p@A0!t^69<Uz4
zl0q+Ou=H(0Vz$s_$kYrg6gh9uQBwFZEMBMU<-XN;dk}=}TT|V5E(~Pl*i0v=!XcW9
zaKdkb03<IvP#eMp@lIT)e~8}^gy&I%q)`}y13crX*a$qm#v*mRybVuA8q<;h84Fde
zC>)-k&5@KrE)-@3#rlc-JzO5HB0q5z#bGJ~yLp1+o?v>aQqre=A$C}T^$7Su4^}V-
zMO-XLgh(qq;g2N!EpxJnxyUT=MZT@%T78O4&uKmUac~4D5JUu;ZU*enRiNwy7|Db4
zBgG2_j+|T;xOb?)URat5PNOjnCL(6`a8{p{Ci$PwVkJqC@Rkg#s$pjWFQ;%d(CoC$
zt(0E>Vi|?8&`>KwvW{5^n$?`G&-UIZj9=l^z~tG5b(p^^>k;}yeN^`tn*)Kx<-!(>
zI3*XjJP}^iA$K}HAOdiKTBQPr-SHIF(#F~kfzWCdr08b%^B~;N;54?6>(t~X4~pP>
zwcRJfW7}9%x5uHFu-onx<%!|Z@`}|y;}rS}Q>@T79p>I>y96WB!%WpJJ<mI~PPQ2W
zE@GDvTH&IV{iEfH^SO?#tHqKCKKn(o6h^v%8c2_ISaM92#vwb)K~qZ1?^?oIabkKB
zRT3`F71>viH$&V<!P1(&6U;FssJW_BtrFHwkt0kIZLm^rHLz}Q|H5bs{h3t`2j3Pc
ztiE4pGD;Q%5pY19!FXFw#Mr8s^_1FaV<2K6RlvZ(Xq?&-$g(Mc3b`WOERD0>gGmH?
zSv9|@f$ElO(DlKr2W6)VlYOSG)kfb92t{#<!M+|-pN>&Qa{AI-%?*&M%?k2vkI)84
zZ}-r6Ef89GI8u>FIqV!y%ISV$<-tN;S|m*<^tKQu4ZLB6?>zg>A(}&AUln$Q8f`$R
za$l(ezl>ETiapCGn3bI))ULU;zT`OM;}u3kCcop3EA)?J508BsgPNHY1-J`6C52?-
zoaS@{UWJsgXY1;5unOZm1D1!xR+Fhw!f1AHO%IFHgyK<)p|qum=ronl1S}H`NKE)!
zsU3uXBE^+47UoCHmxv66-yk6`spWwm8tNqj=+8I6gJR#N8#PFWBm5`XRh2Sl=-4%S
z(V+HyD4{oLqQ@x~mg0-rRRSW7@stWfiz}7MHIvvjz(0c>i_I$DtJXOgYRC!`+Di1!
z3x<QSwAu&(Bu~}ggVJhou<_z5eii0pjQQ~)=@$D5{Ymwqv2u&1s>2n8meSOf0fUdW
zD&rf9Qp(U>_UNofv^i7&1!Xj!vafKMrCuXX5`gNj^L6Ad5QOY!g17;J`g{vAi{gzE
z(>yda2w4if!D569EZq?9sfGRdscLJ2_|aJt2pp#UZ1hB<TC7$ZvV9eQ7|PqA-m_Po
z*V2<0n<5c?bZcOAH(J?@MDvl&A?VOPTm=IPSO%ZrAF2w*5YihGun)F)h7X_;QOOQD
zY!zTj0}yT$oP+3NWDB+blNF#D;VQYcuBuZ@R*Y<obgB-vhdnW=uO*^E5{yFhEZlu$
z`4CA2(`^8!)mhjlV;>+B<&eBLROqb~p>&h|#*o>K2?#D8&@#lSvNJ82U%fWZm`r|M
z^w29GIKWs@xi+*ln<}||sH2lEmPE%|DrpPf3nT0yY2jOY8VtG}dv0U-EvB?5ML#<k
zw`-?FKU>IWPmg|fDxaMe{p^NQ@$B^IXJ_!eGozpN^VwO^&ld67;^=1&(NuT|pY7MH
zGCcC<Jh5pvG+4^_&knyIvU*;5)Gn%PUbU-ZOqOK#{2)%?V4S@lx0TQhVvk^)6xv6)
zy%&Bi=XOueaQqWrg?rSo3rrJu8nyCC8a{q250{&8ce}P8P*9aYG0%G2k8o?}f=aPg
z`^h<hsaIV<%AG)AIzeN??^sNqV#f%kLtIQ_lQ1cmfcb#X>uEV**|+16Jw+qK?>J=b
zeXO`rheGYcT!79)fp~IGh|c?QAfKv1;&&{_eI22uVLsA@8G0;0vsi}2*5E*`NmvTu
zy{BoQ_#F%MKrG7{)sXU+e$*Q5quoz+6cNCu=cH^$Ee_}b#h9mP!1x^tc;6PhY~jOF
z8XcXCqHO4KSCkFa;DB^Z0#;6HsQ4WVb@vS|i!n5^)qMe%cVw$q<4|3uQQ>zis>32D
zY)#6e-F}ZLD7Y8r?ERR!oQ7x~xK%d+zvG%vli5*()1zFxT(jBJAT6S@80{p~;ou8(
z|7{v2e#fGedAbv&tUSyqSC=TgI^s~?u2JH5EJ|512+Gjx$TgG)GVkmdt|AWO9U3Ek
z$6}O~i(m}NdaK;lRLm>uo%iF=eO{x(?>KbY=G|;(^gs$RW@x|DWtV$(!~won1IF(-
zz~XkM=JtbJ)|W46bXjcgBHoQd^;L}uzhhC!GE+J^H2HX%3ym-D<kL}09MZ3Ar1%|+
zR2Hp*be>CtK=Tx#_0qUk<Q1Ftv?j97xK`Y!TY=wkt!O{Os8vg_rX7V@j^^B>+%|QV
z5$T`BS~R&D2e}sfztfoikbwD6QUPXj+%ZCtUmmBCipE`VKNww3urJEl1<_SG_+`cQ
z`d#P1Y6<_;O6D)RG58(Z0S6<l@<tfw!ZL}o7LV(7Qyv>?N<=AXNPpgjL{#K7BQ(D~
za#ywM!tYr8dqV?VDo<9nYSf{<aqvvAo|F@7@VbN0S=t_{@!@wYzWrFq$n<ELrq3E<
zXpRd%7o2(LxNsHGSRbvi;&&|8eWfC0rLCcc_6woMf;8`bAyh*&(4#a^{Eh{B;AjK(
z_=reOQ$DC<HQ12I?kFOF^NQ?_TB1QO(xCA>4z%*k6@eX^@Wghs%Lz}v>tNDiY>US8
z`~*DW`6_rqn`?cbq7miYT-&!hF}*-z!tXd5wfEDywbF+1hLJ;|(Gq^%)o2M<;oxjr
zeaVY7RQ!&GdJyyoEgVe2HH2r8fm8QmT)7!hNI;*NQxrwi<05wzD=*gI@jDj$p&W>`
zM{r_F7<z}Mu;_BqguGK&bX86(9LajIZUcVDwqaj6=@LzlfgX}pp~r$aue1u)-~iQw
z?4=qce#e455E(N{O6iO#yk&rfBvnTd0h?D+b=1-c_+=U}e#ZhnNa?_pR|D1t&o{34
zA-0Yv)D>Gt)YA$3QVkowJ7M3g6|Ferx2NY((`@)T!PYJcROMO|+}_(PHFfL@6}DsT
z30PJ2!-@bop0pJJ&k}g%p@oSxDS!<Pw#DXf&<_JLw7wV^Y}=|}zzHK0G1g3bJuuYn
z#~EYtr*v>O9Ib>cPUUzq`Bj{%$&rpeZ-My|99`t3ae+F5RhB<uXT$@o%EU0v_v9=J
zC)xUZQ|q2zqkHd7SyktzM<*MTi9H&AE<O69jvlQY`f?NGL-*xh(U+^mkksu?oAD@b
zhsvp?8sIocMcN%0g$QF9M!|s*&jdH%3;>B`Oi6u!E%ul&7DhfO^b?B0EE6VjIHv12
zcK!dG*#uoA#8CVb#U_d_JnJRVjA4#xGt7ooi>;A9vQK0Ag<@G|-UhM7URV-I3H^)u
zq;2I@nknC&lPM4M>~x|Qo*hMmH80M<nrmm5HUEz=k)r2J`AAEncj}J6IcLWohLduf
z<SG5vR6`$^>+TQft78fZ6TCEF0vpJuBm7eBU>uRUafx)`nU%L_>>tjFy?}Kp0#W+g
z{8Dv8Nk!t)%~;?rT+`o?7F`mw$T$(osn4w%>!)&JJsgWAq4abe>qSiIA(h135W<*}
zg6kClE=%b)kAeqrq!@}Cbso<cpOtj@`x)Jq`*XHsZi(h=Xn{DiD7!U<*>6`d)UFI~
zNgKCEv{4Qvo*mi73!>WCKK*=cAS6cOFnH4LH;~9`wIfEe0-Lr^s#{o$nj$MtzoSX?
zb48-sIuiABSY#!gk?V>6U?Gz_Oi+Q2jS?`dfg)MFQbLBy0y2bH@Ru6X3A@ReeeJn9
zFdZ6vjORH_ZDv)Iq5V;GIl*<ghimO&BHn2_tL&|$F(fLdHL=@toK|S%v_d}1MdSI}
zb&ZZinntvuQ3fRDgk-2-4dBG<P=dA&l}a#!FO^`qSk_J>wuW96c2CTb9M2XNES$jb
ziMq+r2QLpy(b@}XjkaQw)sMEYp#d&dFv@O~ue!Fe)p`v!f+{w!j@dq3ftg+vNl?wC
z{N&w~y05G|tem664iYAk!=5k?g|{v2u5cmgs#iH(r9vdu<V0!|2t-(fLN%@&(Li6Q
zfmU(=J;XxjV64ZvQXsOJfPGC2*fF4R41Dx*_eziH*m@maJ)P}lx0~&fR_x6U*TSMX
zlKC-b#aY{8iS0RxSQb=fd+wp1G7*GH8hA68qng?L=zN>qdF&vDAnoZR#-nBX7i(_1
zNpn**FE_#0V@jwQ5>p-A)KN^h>2;mlG)@2$_~Eer+UI+G#M&|luARj}wlQFnE;wiq
zn+<6J2O+D#b{^{uu_o<;Ros-Uw0Rwpbz&}`#=`Opta7tvmG5a*spnvo$EtX&CF@I2
z!hu`hAfVTP^IaK~_J%Of>pL0fd4$#U80dH~&>C#!)Q-@iFjl{a$6s!WF(sTK9VY|K
zePXaKA<!{{!40cXmo<rTu24=c@Wy8L)$wuZLwD*yzskIW*$Gn`Y~vBTjL)Gk6GLW5
z1@VB;o6w#?<r)cRQXcPNQfo53YGKGiZb2?#w{9%C<LAI$lp6AhIK=XOjb@K|VjUlb
z<`DusA*_kG-n2ZRyZslc+g0SP3@fWg_}|u@96nr?xJCh6D8+y6N~c??l50N@(Ru%y
zJ13gdztW_hyL(1MPrZmeiu1MQkkZI_ONQ4b_^Hr-QjS@A-LaKiUPjiNoG#F-60z3Z
z5*}IYd$o>4w#XKP$;|)<ir^j%uCYL4s#ivD#v`S8u}e|FRS^Xo<pF*2=y-%eMzPsa
z9)cK9F%boa0!^jpWI8`!oeJzDI4CpOw4E4_*<Gy%FI0U|+v-TWBcHX)oN_H99<*yR
z4eJ8M+;)g13;?H$Q2(lc`gR8?g6%>>eq%oT0^+&|Kb+N<jwpP0FlcQ3%F`z36K%pT
zWT2M>*9+pnBAyGJoFFr`8@cu!3N5r@kglxDAyZU<4Mw7Xa5H`Bb|rGls7(~RjxI*%
zSU9}QQ<}4g)HkXOrpM8S?U;*n$6S$fknG2=n;bN{D?+Dx;pfr~*96^Q=>1~S`(o1z
zYMtA@Wf{_AfP6rTp`~-^P$jLl0}{7bmVc3pO|#SDL>{CehtHUCoraSeQj|-&Ew#!j
znw5jyb0HlOdJZC=ZXejEDeq%MPeI@jBw>Irj<|zxHAr20-E4bat*LpnX7iV@c^UW4
zk0=EPgW!~DV_^`I@SSCZve$J|mSNy)FZU>W*s@h{<KRIjy*a>(Bk(CU;IGx-@5spt
zheFRo$1DWb3*c#%2yL*V%Si*?8Z^L;P-($-(*naWuVlL}^Y1dmeeH5{H*E0#1>LT1
z<!sks>Qq&k^^a9XHo4*Y!+JtFe(fkIExRFTSxBYvZH?;3IZ@#ha(rAXFS)4Eg!WD$
zy&{6-Z4MGW2a&{XW`OYC_$M0TZ*xLCRK{hgR-Mu~xrd8li0E<x_8kFO%MhLFRf-R#
zUe)Z;-rGz;I|k$oI>>imRIk6+tvX_le9MAEs<2G5>jKp>-Kfz1EV7ui>0O!H^bMYJ
zS?q@GHw?0;_E!=Q78T{GC!gpfb1;oNat}4n_T_BczG7_}YE<X|KJ-``^KUsC!;GYE
zbbsl#jv3ap;RM}=hw`<dwrdkIpwlQv7(%av-x~q?%@iG1;fDjrP+dgO)zQzc;d|Fc
zK5Op}cMKI#5c5UFVMSGObWr-W!@JFyvy$$ThKCCX{sKRIkvs0p->kH-vMceR(=qzM
zvJM{mDp1gHqcNVKa$#^66kUKYSP&=JI!%pnw#9x$LTA9gQA<0{bLt*vQ~F^;%{R>%
zn87LFe~ek#dcH%uNk96s)wT=?4B1pkoK?7D4mPJ5uZ$C~tZ~Q7m_(iLq4o2*MlA**
zs^+JGu-}US1H>&*5#E;E3Ub#dF%%R9RR9d0R>~&7&o1Qp><SOQ+cMwFN?{b)u9`M*
z6Rbr+a=e4s6|fG#pNjU3>7c^a^w+ofKKd4`s4e&sJrB9mAPk96VHfspEC#si`cKtF
zuG}7RAOwHm5<<Ty*LX@?fMa9^Lh(>o5J4OUzNvT31x0#xE%w&N*S8m1V53e6VnoaX
zY?uiIKE?_N-6OHYqW*;cHiEw@5=#QzVg1S3Bu9lgI<2BjP@SXuO0`XGyn?WHu--8F
z=ztor+hELLwmK4&fvh~nG;5=XOjAKxNNkR53A5Vyjg~U!?wM~I=qbVHP_4VQ^grC)
z3%XB5PQEWf<lK2?c-l4MhFd~+IyUzyqN48m)3+&4(d3p7uZLFespg2c?Uk?Phr&jh
ziE&KtzzC{g7OCNNUSv6;#{2)rP@}!!c<g4ONVdUiT9xKztn9%&(n1OTLGj&lLgBng
zY!Osc67eOM0#@s7qd<()fE6RcsnxLCJN8mU-)rxDL_bgtLjqwKwcP6{BBcI6Kx)Gm
z!Y_^{*vz;G?XAE)J}0=tzyh$`_2wZBF}y;KE+{}h6ackLJ>hskYOu@V&zYjbJbjFA
z!TOvnc!X*JGX#YeKjBT6fOpFX?vDhxLo54|#`30|SdK(E?ACOJ@lZHXLF>ePrUFot
z^ob!!lTcdlebm9HlZEXoE#c%hJ{2;HGn{ah#{Tx4*pDb7^wV$!V4;ejpR^Y2#^Ue^
zA>^@T1^34T+=d_8drgrLu(=Ga_B-=kb!I<n?;#=Vycb163V((9=$*R#cjavVfr^is
zqKSd1h#aPh_(2hA?I(lQYF2Oe1nZ&5PP#{f{9`_lHJfIq9cJ(7jq9faV5S?g(7P6$
zZO$;*pEdT8ee%h{$0!2Z6<XwpN633B<pm-BXM;8xIkV{%m^&RA*?tO^wDz;ImhD&^
zin=IR)FM=k*7jHSQG)gAoUJ`9yb#pJ=~03;rl7R$wxD&k2S3DJ%ZDR(EwzvE*{#vf
zKFVhw<Fj2)JU$e0;p2Sr2|npL7HOd+!ql;$68AhEB|ga#K@LJgZV3zU>8NKho1_=_
z%$c*IjDT~<kX(;m5@Qj_c~~#a)uP<QmQ#PNkpOczPUb4ybnYlpe<QdM6Om&AN7yjr
zO=Aq<9yi`mW0=Md2!n)83_^eag+9BIg#)swn^1VlI#jI(D(*U$Vob#qfnreyR>Sw~
zDR>Vz0@&Yi{S8%*E`oq`TtlnCv5d8hP-QYroOU6y5?3xdy=PEI%nb}T)Y%yA!em7Z
zv24GBy_`DO6u-A2g@@`Sae*UhS^7I$yM;KBg9VAy)zhrlvT9W%`zIUvs$SW`T@A^L
zmhpd`{*s8y%BXz|`X##^c9054*1}1e4Hk$5)I8HO-!eyN?mkn7p*0jB!L+Fg>>vi!
znP1UAFbhizgd5}jL)owNSRld1uh?^{3hCT$tal64&rX)H>2)JK&c(ZS12|WQ^i$hE
zQPZsV_Fm3DCSVrXM1hgOZpxl0JglwGmS69P(Pp<<%A^)`G)vn0NQK%Ftt?-q_6k@T
zy{#z+{)$v_rJLn#QZ&a0V3~+-;4N&ubaYZ2O9?M&&l<I-Gw!o?IE82$?4UIX7ZlV*
zoJzxqovCUz<1VNl%X%Z7*m{X8%1DFOT3ii9hB>2I7x)~*o}3j#|FLvX%+}En@g(hN
zE)BuBGiI0~3^^Xj;pisRJVcsQ-6xXegyWC*j^W$`9<4@Dn-{EQRply1Hv`Zq&5#=@
zH(dMBTtl0=?rEQig$3G~5mL~(b4bJS#umB?h(tu>_(hP%2ovSj#ZvgYyj6zzT(plP
zTWbp@t0S8%NkeZ*)Fu5z)00qWRIv$TYbZex6WFO&$uH*(1{uY-Tb0cS$ZTb(B8z53
zO%xV00|(#IV`15<HF8XHmU0J+4P-e81~K)4O0-Xi*VJ@`a0g%vM<OuO6R*|e#17I!
z$Px){S91=VCt?-aDMq7GdN)iK474q}Io{O{9@e-*dXch9Dugs+<ekvK>kwwt9d#wZ
zgsuCi+Tuq#hYBM%DI7|@0~o=mPtuzuYJbd`sS5pzC}oPmW-e$=O^<F<D*(v0B8Xop
z0wUG{OF?%=+_R*v*3pm5w+@;kbao)Mj}xIV+8B|ubydKL6^qmd5#L<jJ2o{n*;?2~
z6(f59XW63=E~utetTy^8^*;0@CdWQRV1$Yhn$<oXOuPaE6ZMXu4>hGYHZ?Im&ti^-
zR#<1+8K98wk_778FlvU-=wrFQ@?^Ed$9<Kpl@UxMIK<~y9?yrzfmrFEQ*%J@F$6;_
zs(VGq!9G=aeN;tf#Ak5S3oQ#GHTJ0~d!FwoBH9)^shtI=tI=ue7b4E%xb%kGS><5{
z=cVA5w7)=HEG)6#;KFKxv2AHF1YsqvBp5M5!yyzd=>k}g*tqT-(Hx+52GgtiYxU}W
zPLJ$mYM<dsIOA#WZrCa7_56YE-dJwS>iv#j^==l+xA6nF<A=LHxn)HIYhApDMHHBo
zE#F*jik4oBW-JvEvQPx6Rjj5P^Ff?xOWzJ7-(X3^&xVGso~OOJiiWC6wT2h$?r7H@
zwL3ac)9;QP^gEyg)h;yU6)a#8ZdsG^NSBm;38Cs2I;qN(Yqc+WRNb^nRgZNyWbF_w
z1ctycU<EJ@R>7nuEy{G9b%y3%K!(CS*q}*FSITk**M`}}sJKJ*1$uKX?seo>OOAsy
zIgZxk_)-pX92jn_F;fbi3U?F{f_$-)ASZCYc9%zxmuZ3&uuuWP!E7Eiz<3iLJ>Wr(
z`FBcB;p(XhcT1acynr($&{N`Uj#DJ!@A_zugb(3MlgJ6d2<YSL6vk5O@tRU6XiD9k
zgHlJCwHQ|s@aRA^zpc>ymFs26p4DmHm0VtE_+^KNVqo{wew-AE4U39as%%+cd<cul
z25efelY{And>SiJ`E=lb@nqe)lT_<6n+QIZkv$PY7XCpHar9P+YFo%EH0=#(-Ca)W
zwB5khN$~sl!yN5u)16m$Hk{`}O0X+Srb+B{C$Z>-KHVdMES-i7CT>a*Vly<_Alr0|
z;~ukLWJGqeP%`qbRh?3H@s?vw(KLNVPMYp7a}Rh?46L-2358<v3JAsS2`FY2FA})(
zTV_xhi7rF!esfppBG>M>>zp~ODorjxC``1>#u%tfvV~0i!(2EJA!;ZVdt#({774C)
z?!7}nHYrbbs{#~s1fte#>boJquhMxa`-DLqew*vK_3nmZ5FM!B#Xv`B`X&a0WbeU{
zP=~PU1Kb<2up=>T?E5;lvYWU+Q_^@Cx`>V75>avl)f84T#<GA%bbc%V;IhJ}=9?<W
zG49yqli?^#rV_Lo%fdxG2v^%zsR@R}SiU$bKQ0ZZq@kgiWW+I*Sa(;oavX9AGo&o^
z^emy&A*Pc>J0#OM*7wrSmWNkp9^R~_-nTjaGA=jvbETMO<B+*t__=WLS7W$XX-8cS
z4&7-VpmxwDb!gRctREDAw=ZAT*dl!)jYT{4$7RbsNJ=23Tj8v2W&;d&C+fjC$C~ZA
z^SWB{D7R#t4=Qdzu(vuEGkP(%Ra5u%IVHqlF!c~C3QVt(ny19v7n;POGq9L~LhG*w
zw3f4^_AAWZ8<06a%AunE0=Ir-Wo5h}fr??p4kH4rR3;=8Jw_SkSHn|c9_=`VnRG{d
zNAHk(^n0y&r~&2mmn`pHsi||d=Di1U@E$3IQ~L4IEEI|=C=~jJPocMGpwPx~<mdqF
zI87?_iguwZ;^>&&Ab+EdkkK0vUtZE<Q0pOvEi!|#8#Mjk{BQOOnt<=r1pGk`0`|xO
zjqtzbtB}co`&3Bvt$<W^`uQ=x_aJ^m=-_|CXFuf+hI$Pw=GvRYSCRD#zX<~5msk2(
z?x30#o@sH>t5xoD6pWR*khj7VVA|nP2qrZJ3OA=$ES=u1>Gac_!*CyU>^L<!tx0rP
zXe%Hzl7~W}ZwC}Ia!d&H^Du$J=8*rXq3yPBuA{5kT_zoRI}|KR0BgIn?*@3nJ@%_G
zAPu(NBUVYs-#3C|T&G9FhnaiqtM*q%<pa#xX-tHzoY_Z&`OSGM)cp4Vedr+kvHJVR
z=KTI4B_!{_rpd8wGeAO?mXXB-(f0#HA$q#BGPFD>IBn$)X+w1@NP5_+)k~HxUt2;h
z)kWv5I(PZNU}^F4)$XZQdmkwKN)w)aU$_J2X#$*~2vGYi$5_T0`;bzzLDRETB19Dv
zBK#nN2v4{a$^$#hUEX&|6K+MnZpB*Nia+FN#onGBKq#bN@J#UjC<3pz?*2J)5Y+xc
z9Dj{|_BTHJJD=@&EI*?TBvrfCZikr_G)V5D99N@i*&<%g%_{VAJ%gBIX>Cm@-x_IF
zk&?<RLSaHNiL@$mp2bxK<UPFFSdbxfB<CSk%z9m8j_|OKk%?RFdf}|N={U7@!6vt-
zm%2+wrN9HW&NM6bm<%(^v$f+Jjt}5Tp6bP&NuNVnS-U2cwiAv5p>qjmd$5X-kupj<
zUSiEb^*+uh#hTQf&kc#qlv{NkM_#)=7d3%dXPd(rnUiMfoyo<Nb*{JH)2BT0w_{(4
zUC)Y0tI9w}{I4{tRQo4&m_>U~Ez^KnS^NeGX#FD~c2MZ}ryom7oY2BYlX)j88L8jF
zVh55>@(q!0>Ps!jI;aYTvdO50gQO2@73M<2Jw4YBB1toA$vsyXM3qq5A@fnJx8&hm
zq*hT!VxEe)-`yutseqM*EX{zAwQ47M*>1?b6ZZq%l4*h6h1eAe+M_q1INpLZ5UkFY
zR7Rfuo?4QNt&#CqQfD=&7OGEMr(Jl;6mHi&7ib}Sl@_x9pa9D_RqUm4>XlC@1f+c=
zBKE-u5i2g@b7+z3n&^vg-351@)rFZZM7ZwCXL}yYgM*N8y$uOUC|m>D$zi<_uEA|;
zgOCu)s`zv_6nhr?>Kr3qSYD2jdCbOKNjhI^$orcq2z8erWgqir;c$$5M5>|W9ob6Z
zYe-DnBl29jQ2{jU3}c=ub44Pti)NsTtY!PHF#tarAt@c{xH#}f$H9UO1h@4#SaGs|
z;zPQbLDt=!mUOqUj1v*GL(E<-p+)paX0G0_ApHM>;A2<*D+QmhQk>uuaE{<pdDfR}
z+x!qiF-IjCUb&nG*`iu-zTlCfc4I8`a>ZgjIEG;%&jE#zbu5(eqq3<T-3H-|<l;!q
z%2iE|s1PK)z4Ar2ndP8QYi*(z0QbCBT3Qj$_c2%Wibf0h<(M@OrQvEa4(kEgE@QCL
zL?3ahdDaRmh(-^0(B&r2Cq)UULEzbFWh<2*$H5xH+UKkmdyUq@!u$c8G>;B41LAz7
zLX_D>MFX%FM;j0lsj-Dq1N|M;gne8HI256i#rmjx9hyt^eWzi7IOGm}n@~(hU#gDs
zILxWSwYkvM=}w|U1<7Wln-*H27SmbV>|SPea?i&jjbRj2D^8th3W{P|OWa5GG_0@%
zYKv?1wDdJSE$s%WpLJR~FqnFwA+xRmp);wDA~H?=G-8?(PvJfNX-Xn~?d9Lyn<)3e
zkBG@jCbpjFXEL$v%V(XthW3_KmJRv>*igbXCZQ8S4jJ?uEd>X)0@Il{WTvwZ^f*zv
z5s7s%4BI{o8T`P&<Zm8ftFEvMg!^<)+_;}QY~GJ;%(y@}0KwsS3RzN)IC`BBJ}mW0
zNI(2%XFtFaTb<yA4XVd(>)b!KFFCsm25+jgu{^H8$YEYioE`RR<u}^PK#?##z^G*r
zeO#&w&J)UEWbhb~qeadTVfi$P(|A$;_c-E%$~awQNVOEX3ylKq?fWTZ$cGgDGM}m)
zR)Y8`l}DO9F4Sthicv>|^t&+yg_6IBpyZyakzqo%_u=P6(owTRN$0qEwl7-^ql{(?
z_NUN4qNGR=O)Rcf!spFG)#y`ao%<Pjb(QK*)&mqr@Uh1aTE?{+G(}aW60PzEei3Vb
zLs-k0w`#)PpOdf$2J?k`xE2x*9Yutgzv?8WN<nf6#zbfpE5~bx`gcV<9OmB@!EiY4
zIzmG$3?_GT4&*|519K)fB>T`IqXyMS<Z<T^nYQ9mBPWx3ait^x^$*fi0n}wA&Hp=_
z%X~J+^F@cbxHE+39yq*L`K&t2c>kEeCZ6PIku4=()U*%WFj3S0Y&C&eH>ov{ESlDx
zt2g^PA@!gLYgjbryHC3;9firRyGSOLL>Bfx48_5%oLL7+;(B$8Yt#+F0^~bdx}3JZ
zn)V)v0h)E%J6P69SnZX>)(qCqMsA0qQAU*#IrC5;XUx`D1j>;{bg1)zweLVUtfiRA
zWI$36fURb=PLEf_iiVFi>`PbqDTF0)xGt;qqFfalg&|ohu<tbhl_NNe)nU`h#WJ)C
z_$p0Ym2Db40!9l@;!U}tLQ%GuI@K}sXPk`%B|@2CnSsB8EnsV!E)`rUau<lX1kZbv
zjKNn@^^2~5`c5?uFzbC;u@2E{+7EbR=yjZQJ9JovUUtqpV}G?6SgE)w^I~9sYGq5{
zU<qZS*HIyJ5U+sn(Qi8W=rLsD$MT0ca?`wYK6q)l@xy@LsEicqQ&3G$(w0MKGKP&o
zt2j12tfOmGN3jVB6cgns@#!#<%-w)^8b6i7H$v2fJXpEmq86OXTV-)kuER3k94>=5
zz*xd(O3y3ksQcXl{)fxKG>+Y%DXAhdLA7PIefqqVa$ra!mh8IfJ`Y;4p~Ui$qk`8$
z0UX`Z;|74-PX!;q;k8c7j#V_WQroWqgzfutb>Cl@v+obm4ka=q5Y4`6@<fHgl13B~
z9{63r1Lj0(Y<}%%aDcPiFj8hN8RPNM&z`_%$3#DSBA*=_{j8VI=0!i7&u34Hezt(m
zj*EUKQJ9O-&-(b@@!@BBIdu+Hgo`Aamz*9n;pfVwze&qX(&~j2dp=~`ySZG&w1KUH
zW(%{a3XOpAa(&}8%=s_`e4V$$?1QM=mc-}zEhX-_Gl$Xgcy)xIQU=fX1JhxaDs@4(
z#!mBhe#}&<OJX@9g9_tJ!Jw91;@z{bN%n48T9w&|od6`Y8Y7k%oJWk#nQ;~&JhTF4
z+4QhlsZ5o{aF_JqrZX;5L~y3D2311YgSYTC8oJg;#;2*<Q8NYS4TK;@j&rx+5FE`>
zeIh~g9piD3(k)cogz8aM&^?M~=*r_fy>~mRQ3n^m&zg>7Rp@m{6Y871PLJ*~w?>lH
z_J(e>=_TERF}<ThX*iw49!A$_r4LnVR3l~YoO96tPSV=lg%uPK00XwO8>$Ep2yxm~
zpJ~-pDL1otc#MjR^~?~<Ovqwj%pzqt+2w&0Ll{(}LN#_n9c^KBk)GFX$vLk<!7<$$
z%M`>Rx}eN&e+cF`J&Uzn!wM)~8FgJ53Wm*UKB#ejFDLGO1r1})O%LmpSSVyw=&@k@
zQ-IOT!L_GyntB?ise_JPg2f02vR>MRotRHz!ISyJm^`a?#<(e82^-$?CdE2xz8rsx
zyhbP^d_&!^$ztal`Tg*9H(seF2ijyrwuo=_iY%@0$S(Zm7kgHfO&Cxmt07e;7<df<
zIHRXPZxJ}xYjx-#WTmbJ;zQM|ZL&#Q?5mVm9)=y_ChFAn-Z^2#DMoV69BlU1lz-N9
zTDu2gI)oplvb0ksdnh!-)iA$4B%v80$6;&$|2GlDxdH|Gu40{@!_s52(4)6zXxlLe
zDAUbxGe>FKfOq70QccZzUQ}}cI}ujt$cs+P%ppUv)6^0ih9UlrS=BDhB60K=$UQK-
zQ+_?dHcYav7R${BTr**hp{FAKM(Aroexne5K##kF4^TVnQ^?yH70RyeE>5){Qm;_^
zFJtg85o1tP`={dbA>Ey*_D}Qgitg@o+>KZ^iw5tE=x3tZKa<b?=c)FOI6&=2-=TY=
zA274_#35RzFW&7`j8`{o)6y_(E2s3f;bPJwe~ajmO*9j8djtA?`BSan|7U6T-=)d-
z5k<btoB79-a5xEH8R+e&5bL4dVgOUf7j@*Vi5LzklKg^@^B(~@l_mEn{m3@A2C{f%
zc@q>a;!+@r35rbfX^!Qh{WjL8<M58v6Z=PX`ySBju{4=IDz)nnZgKl`rS=f~xZ2D6
z#;|gyCoZh&`(R$Zsc*%~r8osOzgo?aK9#|;ZxWkkSi4pFBz!<0H)bl@7wqvT!B*EY
z)!uddaTU}Q1GTp>nD{Z-UY@R9&li7)7i%}*YM^!_|9%^OjNMl2y9C$ty;4`-#|}|l
zE!%Sr63_J3-ic@PF5bbA(0dHjipx;)1iq70>4~;dxspns%zLWR+I#rzQ)=(UukE#)
z<mdbFvvvhB-i)hDYp=u40iCMw{k$_h90lbmX;7YOp~#g$dD>2aa#9+UlPwgv5-6wa
z6etVRpgi3|kt>06>P~@jS{jtoEfl#DC}->xC}(y9rGKYDIV%myA{LNgDOW<3#XAMc
zk~AnwEfl#DC}-z~Qd`D%RPND)0lhp8&fV~oP?V7?f%A;~a5Bi2bJC!!U;*hQxe_QV
z^Fyf(@SRl3oSOz`m4zc$0_T}sz*(IM&YCnhgBFfl37qq~fb*<OaL!MIv)00qD}nRu
zF5p~{3C_ASIM1<g<VxVI?*h(qJK?+=s}OI-E~UX4vT)=|;QW&dDzglDUuUJGX_elC
zd0UN-3X4jm7Y3E4F9WvM7?^eKHM`B3BU`(L1?}wWCee^5JZn^;m%IZ%>U9uxzi1Eh
z*B4GJFa`>?el#hi(jjv~4*Mz<S~e-+Lg$>;t&D|=p_0YLxV1=U2MmCS-ou!*y0t~+
zvx8Sa@FooqB3^V|f^|SDV&T{E`hvn@8z@;3r?brZ^}$E1zQl#*HdGgoqfylh$b5}N
zP(fD=HjvvLrY+60a6hP7ZS;9^=TyB2M_yF_uUm%`@KEn0_q2l;TW}nG!AVG&zeiRc
zs#$u#20B>+ssg(<<kOMG-0FdlAxXeJ_Jm#nK1bHcBRrP&^H@6-^7MUz<Ux(fB0%+v
z+%NJQ^3?FMx;@?u5>FLZlqZHq%PUrsw!MnI`E%X^8e3@NDcAyHqocx9CsIb(P8lJX
zPg;`%1x){GdUA*3msh)#WBm>ceI!>h6a9xfrN=ANq{n3}AhT=RN{`EVm2VSKyMph8
zXV(ZguVSgBJ&j}y5;(86a9+~|oY!W8^SU%R624L3?67cN-vyjEWP)>L8k{$>fWUbZ
zuSC?pnO8YG>ngsJ+F4h#R8nVM!vX^5T3!j9>v)wD4s8U|;k+dc&RZ><>n)rc^271h
ztPyN{V;YpVStxI}P~MRrO6{F|C$*#A#ZpNf^==lBj`}xVNk_ehS2;WCy?iGX&P^<p
z1m}G$AaHKxmB4vFukyjk&{4OfK{=UzNCM>p7Rm?nL#cg;@1#=Z!)b69S~wrEaBl4a
z&POxB`B)kpaZeTMeB8qML>F*AnF-FP(%{^HY~!l4K5gNArVBWq%>?IjX>e|10in+A
zyb?CPgID?3xb}Izlgc|^V5y{0B@uW8&Yc#{7rTJ-rA%<XoCc@gvhr6foV&V!b9W{<
z_oTsDYT?{#;e53VIA6;I=j&;3?qdPztZ(p2c;}nE%EvnyMBum5pxkeveA`0#PJSq2
z>6gyN-%W$_JqzdGEu06sfb)I6lRB#YgQb#2)el%eI_ihKl8*Wjukv+NM(X@H4bFoW
z&QC0ypYAj`KTCu2pBB!~Eu3HMG&sLZgYzp3=hqg_Lpu%5!)b6HUSxRZHx|xs^TWw7
z$bXjx<+m2f?=6%+>=Y<}OoQ?#7LXD3XA9*o`JvSQ%6C!+_}^G6X@LKo1q9ANcqIe;
zzj&2%fd7;4q{5kl?01Q9c3}a5vn#Iz&ThQQ2`Ag0YWFladssMoS~z><hm(PY_fCVd
zkA>1>q3pX;pzN0hWq%9h01M^7nL~-t{yriN%0U*&!4}Fzp7wX`C{BEPcurpyQ+Mjg
zj8PWc?GTR{?wFDhmf7xJ=;+scZ5l0w`b~@gqjG~`)YP`giWu6ay&f1x^{0MHr-g>$
zCv1YOKR4`uRC3%vHyVLb#YC{d4^{X)6(2!jBB<&Rdlg|i=@c?C4BG`UEn*x6WbH$I
z9RFWYnPDA$G_-vZ3Qu{&in}JlCKir9dPm}6yXdwKWiNeJj@>^;#r+Zd+tcGk@WGKR
zmL$d=%>p7fj^dTbjTiZRa2%}mVCgjh_XU>~Z19e}%pm4|%8-q*eTqHHm7Tfe3>Y6X
ztc`@BUSJ~-4N++dCs~?%AxTK3*+d_sC=3%}6Amlk{}U`W*+4X08va$AjMN#yVu~)t
z!rY8NLp&Q|Q#<mf4K3v+WZnj`+0DwrUg-bd#L!2R{n8ow@hq0a&`)3iVd!IcB@BJB
z&(N1qu+1Vv{~uvJ)%P>yzW)~aH$NM@EJ$OQ<1D)rExWwbW0w?TD9Y#d{&M_9BjU~Q
zbO{}~JS7{{r=>wX$wEEZLcJtERJ6vKM;2xScWN5A(=52tEx1?Y2iHE8N3u8uU1$P;
z*tkEWtqL~>A5Z6O5AGn5CtXXjwe0M)mMya_TW(vn-D_F<c|GSh%Svvu0WeI)F)2>t
zD5TZmN!+t!fYHElx-P-OP;7dlzLVAr?S+Km9E2~@XSbOtSzAZ#m#jCQnT?uj(x^FT
zsd=8I=4Bo=GmJOQ#I@&Ifahle`0O-*7g&JnEWpb>fNNoA$@znp_;gl=zm)O&<ZK&d
z7_b5XVJl6pgC%$@BMV^(gQyY+@BuDi5D->j%ms!Ja0sqyTm%KK$TkEO4dbzl9fc3I
zsi7BMW)f)|7PC%F;~onPpQu+x=(y0_HqXtP784{r$d^WR62P#G^**ZC=`R?Lr!#Ey
z&Jg~lv&*)#pPQ|-htfK`Y&(0{cJ`~h&d$J%?I(s!Jt8>(LQPEzwTf-WhHUK^OKV5f
zwxed-@fxokYro_(tqzdE_)_U?JG*RRSLza_!*D%!gi(VBq>GESiJgNAGHUZW%~E_#
z8G9yvy*QjOIL55T*R<n#=Pa??Cxy9^mMRQdspD8gm=$RYD$S-<%chgr*tD6(rY*~+
zQ<hC%=d<a1eKuuWv40_hx?#<_U{Lx*NelGCh7uIg-wkre#_6YawZqhEdUX)uM9qPg
zA%|q$Inzk5JDU@q=n@fDeKCVvGhW>?-nMLvcVQaiy}&ZwMV9ei?=#*A8E*|Hb+wKY
zJ1&}iiME21{5W#oNdQ&E89^K<1jrtp#_WwLT_W2f(lON|@ZfJ@n9&>GrkG>s6Z)7S
z!)axFT>9RfQMnr6bPBI)tPK&WP=)Y>(hUJaRWM<RS*5r9J%d~-RGLV;LF+s$K#)Rb
z<Cvs%Gl@fy`5;4d5=0<`le-SiU>l4iUv;EcYxbBYwy8t&)PmN8J=sZu>yglnDGwW3
zjxv}OOz#lVB($x1NjBbiSsHI#VtM1`mN%~SdE=lM-uU0#_p&wFxjm^@Zp!zpwmn;q
zU6$5kZQEm)+a7z9-($P^(}-nntskZqF}R1EX^5@>e>bL=VH{^28N#lgVUMhQ(H&in
z7%6j32z1J=qIsi)H(E-?3@*lkN4@TcD>KDy2y+#_D72|)VcCwUdnb*y6A3h*skIi=
zTdkmeZMLr6k=C`Zw_W=N+qGBuUAs$oFk0!P!lDL!s$G7C!Vu#V(^yB0F>A+g1<^sU
zs4PI#>wr%>hK~~j*$MCiQE2?48KcW~mCFj9A`7uI>Qg$nDV~~>0<%*TvBuTFXk#aJ
z?3taIM2uo?Fh<K}E|wdCEpuogs7LR$*}C{GX<htQ+r`)0F22U=;`W0*%T%H`&<DA&
zoTy)F!^@cub8XtTg{vj5;W2zLu`<G%US}bW$ylY{!@n4e0F$^`p4ol3WUuEeI)<pw
z*sG6M>KmuV5FxAugKsPVzzL8QT$^GF#!T(+e0yf&+^2nP0MYc7w{TaQ1wA5PVn^gV
zvUS(H(z@&2w!8k#cGq=&cP;B#wh9BP3||h8D-5d<#I@Nu6qzqMbd9qK7?+7-2bnI~
zb)AqLYoYbR&DjWYOBz8wU<vX;OOUsE1j(W^Se3mU2&9WOkRH3%IUw*8k;rZ^=y&wG
zpUBqQPo=f?)3&vrv8}x!e`~`FfzM?FcY7MR|Bt;dfs?DK)@M<H0hApTg*K>}gvo>;
zB19G^E0JUplL<ugu%~CbXVNz_(?j=6LJ&ky1hmC1?(1`T&lUHl&+Vz`a|Pb}`&1MK
zMRDH{aYg)p->Ir|>n?q7cPH~)p8WWA?%bv7t5a2{&N+4JoX;q@|5R{q?i^ebRx%8q
zv$(w&A>tR?!+B#K&YKkH&5HA1yTJJfx5=PvOU~dx378r`OJZh3;NQ|7{BPvJzg6LX
zQ{lh03-}8yd?)f4$goD!4d5)J6v|M3%qfv}+}U0`evsFWA8I>(r0sZn7wvddyd4Z6
zf@Wy1BuCP(+H1nE^O~?<-R*APqfK~ctO;9QvOg5rbx;vQvou^a5MxE+EBHbSBBMYm
zjR!e`M5{5jMryT*4BJ}?AqNpF4S6AuVioC7Y-qv~?9$joE`&LxxLBHdT#|Y2X@}%u
z_Wd)qYNWOcMMW-!Rt|w7ia3JyBQb?dEXR6nv}gEUU>Rm&N)z4^tzRe-co=CBVTQ6n
zXnl|z<6wZe0V;+Q&II!}6b$p2$k14<^3`G$PQk)OXb^Fuj6Aq?o9JD54@mB>WG47r
z$l$Xwzl(v1{Hwrp8jx40I&8y?s;f@p=Ye5t)YwzAa2qcbFR?`tuNG-S1UNZB$DOw5
zSWtrfB~$y@aIc!rraW7xwC8J3T?9gs{W(aV+i2B6KB_wMK`>)A#*|Uy=0>{tSo)Oo
z_B+k-QS=28Yh!YH2Nl~FDD<doT6cSOX}7x_Rs>$e5k-2Y6q(}9aqxu;O^~Sw+kU{+
z;8pn^9S8EYERT+Z_=0$J^zceNI^MM!SM9qt$`<`oC2wqzLEMEwn={7tGM()9VJIG;
z{b4L?oghhme;kqKxBcOKEvxOv^95=93A~cFzh^bBytX%A(Q`7gT)+`3j1^X??|_1W
z;VWycL~>+*^(-lDp2j-G0FN*^Rwl-bxk5K!tQt&jATx)&Qs-vC|8Mrn64EfgS02OH
zvU=rIz97BwSYAo5yl*wGyk2QT1qiO9w&n@#XRx$<e2?dAS@_Q63xe+~UJ1Srtj1Mm
z_?q+5XGNch#+MxRD_C1T{#AS}3;$}qAo$PamEiy2YFu@OKdN**kwxW$Tg%t7z^&s8
z0&b930`A{e<Ek@o!3KWqb^a#)osVxbU(3RGK3@=g7w}5(ePlJRI>XmIWu}fZ=?GA7
zdYgie=-3F^HwrUI8Xt_X{|al*Z_hYi%WBUyz98+X@=DtCvDLWhtUbYYe(lkFiht+h
zo91g-_&(_J*ajEzRl)Zkt8vvCzUD(!s}M3<H9tAYE@=<<rFpn7;|oI9r}Ij1e{yvc
zcXOPsxhl<y0G^O&#ZVSX;*)b@%}n&)Gl@FC|E}U|S^f8Hz94kCnpZ-H8&>1W>%T4U
z>w!wOwlq7%WB^m?0<_C=Vgz|3!f!0?Ld=M%iur3(b$tEw46=MuW_MIA7ogD~B=HPl
zV8TAhJ%~x{?v;C(HOa-wFrh6G)f)1*iS};y7G;qyg)6`Ds&?|P=iwnKIV4~xY?)?o
zu|9R4U{V{TGnk||J63=D3gTwsejT|>p{+2-S^+r}#N6<-87RG%o!v7e+u4vOP#B6u
zk_1UV2xlV*%aosKaCfyp^D`UP5{rKrO@SevdzcvV)k}z4B87Uo*HY}5y0Uue6@Ax8
zFfwHd*j$xl8B>*=)PYD|fZ_jBq#q)ma5Gl3Zcx*5+oUSaYyNJFyT@&e`HG;7fAudR
z!Ay&3haNI}vi<SRR0%Dqb7uHL(N;#W6;Yc+$_fi4R1-CS8#aur1NXM+Vq?05JaxUp
z&Vz63)_M3=4WILD6qK!^6&Pz^Oj8RZmJfu3MN)ELAYr%}H?~y8upqT3B7dvyM<cvx
zybbbWhi1_MM}`q%Mcbi!=D&>uck*fn6*Q{jr~_Y_#K9a$*uRsp#6v5UDH{qmF`Xhb
z4fSnH!oGsJh$_9Qq%<YFd8gBrGXS$$h8tTyS-RL<>)8p*=j|Z}%o*omWan;WSm|H}
zTcRUclQhv>`XADbX$rgRm~xEK%}gP!LlYy<R(SMA*p|POg(5iNjG}>^5K}mET2w|Q
zYhg0c09{YVO0yH#i2-GX=Q-L$!YgBIhD?WVxEWGF5MQV*IBSE<XtkZZ7zt1dGPx`V
zKStnvzbw9mzb&ojp?&nWumO~(tE1bMq>-<L8`&CNPs03Bg~CK1IU@}D<r@CweE|b9
z>vjR>ZQmNQ*ntR=(_+Tg+*CkacIKSBVUuj4EHi_aPC#K|;#fzx@<1i}kF>Vy6vX4y
z(%ljj(~$_=^2BSiII)AKF<*d5iG;SBHHXa;whCJoGoKhM9Tp2@`2(}Hcd<fMC4riB
zbXcJw1DJxO3&><<v@~_az+_l6s;lI=OimS=_zezYut0Bwvp&nphEs4@r6lJwa%94n
z$Fl^9j}A>k6E)hMFs?Bq$eit3(3qJm?=lhqlXv1#r5_L>2do0!Cu?;*Zw4ZwAF;d0
z@z$tRLRPa#G<F;p6XTP4p*E&nR48mssq@0#Z8J0QomotSBmrR(l`*wYOryVATU?o1
zj6TESxEN`yVUR>J)Wzc!coxyp<-QU0k*4%-o0*(2`9H`^(xDZ$x^{*rq+XK1)C~`L
zFbyO*L}=a8bhXIGiz^o+l^@$LwVcR@i$JXW>dYDtR)$n1{pMaj_~2qw_*7Y}!>k4d
zQdmT$W*Uo4H^q`?Z>3iUy{zoiTn4Bc(P_-%h0)E$-|wu_2<*&qSGI(KxWBMc>!4sY
z3CC;k=mN8ngH`xm`fnI77wrphy_&%o9zlFGnbDk}x`M?GFneQzdCL%-DcnNsS^h#v
z!w#JxE*NQYMrayLEH|m;h4U@8%7z8G<w<rtVXHjlWUwW-sw{;iy^fpn-P$o9hET&V
zz4|;WDf%%MLr^<4#&11as%2ANi7hE2%t|e)QB!E}3Ehx?Ewzh$edYChElXedXTBi%
z${Tnk`pRck<I2-l;6JgpHeMJ;OgUr~1ux-n-BGH`mWy3EA<55-io>A{bTIsuc^O05
zp4`p){~H+=OD`n+Y7JhHp~T8+Bb?QWQLz;aYwN7BQsRTT(Hfh_5X(|8rlNsjEEU0L
zw0HiMw9N0FxAL{D-gz5ekluMaucUWAyBb%1@4U8W1LFQ+(KI=BhPiV%3`&CcW7<P;
zqVz$>PLsgM8U~nMXa*Y8j3glBk{wNJT1MPRu25hz^n5ZoW#@>j#}l&i!7w|wChA~c
zfvcz`?-!K3?<O(w$@^ZumPOw8@dY98`*|hg{rqZN`Q-heBX42YW*m*v7Slme=5ktC
z>e_6&q`W>gB;@%Q+`_gcdWR=+N)ud1@%L1@XB&RRR6eaKre&jVj_7`pX^=ml{cxgw
zAGl7qNju@+NwWM-_y}Li>V%K-1?hy3@k%=3i>q<vcfz)w0c>(kV3jTG#L+nZi{U&D
zPi44Q*WlAd37^TmD;MLGb7=~h0P1se5H4y5{sb}SQ{hv5EsF{_@CBj5r+Fn*xN$YE
zd@9_KLIpa58<E2}6HdcAGL6+0OMZ+VA_5GTW@oUu-k71e1PKr#3<Bd9ry{X5mV0z#
z(i7nw>qm~2S&<p%XjndSw1lnqBC^#(m@IpXrtDB+ZKQ`p#5(+W(k#ChzR1_Idf`ia
zL3-gvUP&+9yc$<tFKju7@e4Lo>Xfa?Kz$cXlBfg$0&5}JrkP(YSHgByKVy_?CquCS
zG3v;i<_4vd&i{ruT*6gC)HJLrvr{lHZKvtMtUxB|?3l6NVV)Wm*<V;!gNuc<u2G=;
zJZc2q8H~IN6#CgI_)(Hms#fBLBm|Mt2#z&C<5^A*4P%*+l5mCrlDVhUppV4<=#E~g
zg-+ud7^P^~Uv+G*Q@b)6OP}jxF?@;$1l3jgd5$K*a%qtB($tXF*eaPqyyJ)vlE<^Z
zt~~pHII8pK$yfPW);#$dUyyln3$J9J{P${HdGn<CiJo(7J4IF$qJdskq!+8ykkN<w
z3(IynD-O2MX)0z8LxkZlnW;En%;!ezG;u5{6BR03ttJ+GJt0ueg}&1~QP8?halu7w
zKGsHx7u_ACSAGwCkFRC*z@2<Sdf@xKk{<Z-YFzm}@Zlb3`~pEJYlvMP3j}?bX?ro$
z#r}Mux6(h}4_9vH0>cyq1BSLLvWgq|;v(%G1}==1ke^IGa3-oA`&|P$V(BnuYnq!G
zV%Oo|HXYqRA|>)^`xCyFMcbe91)=TFcqO#`%4%HswEcQJv>l(Qjexj5F$&19%WR_E
zkQKY8b}vHGsE^k~zLcz!aCV!;z7R(~JJ3=dQEcM>?F}{Gk^VbM`d^Yz`K139U&|u>
z-F!hv|7%_e>A$ubS3c=qYfin>0Z>M;or8^<7-o6QtzHpNJ}5XQE%z`D<VAR03E3;<
z-UY~ID~3!Vq?8oe(C2*UHqQ>-2tDzJs=sRf*wbRZEX4fB^qg2K-dT8<^vk`MgvuxV
zeS9s8^nd0HLi&A>n9Pv=>oL-!<FgVuiDOhDuWb>@?f$GQAN>J*EerjPx^I4<qQA8Z
z^h@&4(|aDmHF9l_&Lp?f4{xtIN9Hx>D83+_akMt)Tb(y2s;eB^9=M0(fqSTeJ5Isf
z-Z{ADnvm8)Nt{tuk6TXGvFgQVZP>VS)tW8E<>#(ke%^-lYX*mkE7ol6;~qtEGV~d=
z7J^?XKbqfOQy1hlb)h!(ByH+<x@c+~FjiX4sP2uT;2v$kDebl3)VvlvR$FkIw&1(5
z7Hs(_f)4e-9xd?`jBY2p(?$rBt@wHsR{owLh-2hOTzE|Q9J4(GxgYkY>Dp97cE`C4
zVF|(+X1>VKcHGLR%5FtK=MVsqPoxlI4q-~=MP|nc(p^uloK_AkSKO`-ofzzP7;N0u
zhZc+7ld&;J{*TQ_ghQDfkr;aMeRjw_A)b8mqaXHwW=NQ-G9pN9FbAw@(BhFm7wiNH
zRM~pn?H?9zNxx`lGjNVU0|8HtqS+$S&hV>BB=#emNRwDCZmU<8!#zF2Psvcv@oRmC
zZdS((T~|StwT&723uJ7wxyW+Z@RIX(tienIYRa;^F32TGxT8a;UoOB>4^xNTcBS=F
z>MzT#sRA?!I63IRKkUaCwp3TmcqyWK%UggNh%>gUhv%__BAdwrkdJNP3)TP*#crk6
z%TvB)EW5E?Kvni7vo}|{SOWAInZ=~5Da?b3V|JcSB5r1uwKY<I(Y#^IhOSZ!5(BQ#
zYqw%!8%_hhYEDBCgc^|+L3Qg3K-pbXE1q3uZ-hoiJGb0yEr%x(e6rF4AXvj*GBeNC
zXY>p4@N-8l&CV~F0F2ctaH%Xjm$uF;{K@7CU9F-=gp!!&o9KO);>?~HVY<1*y64+P
z>QBnzm$k>@SLCtymCE8*DT}`|&f<TAY@fp7!yb&V`0zIEh~?y9=<iwK_VTGFa=<t)
z$M<Cd({;eaH<dplkxC}9`Dh15!5DDB@mLNGm|&JxxiKuJN32Q0#;qa14oP?hR=0Dr
zv@mRkaYcmr0d;WlgM)bd+S@I%ESxhJo26#xhui{<^I0(}N^s)23#M`|SqRL64CF-%
zhk!FAiLqmERQ}zvUv}kRlKsk1YQ=s9oX39Y&v-T+TxIC{(+R?m8Sut^v0NP!_XsRz
zIDSSq340p6&`Y^V&#;Z)j1VUj%A1e9qOxfF<y}}Js-ticfeQ(|YQ!odN!d&wit1Pu
zCyU71DNJ$Eq%lC3LS1P1Jyxt<JCp0x@oh7ByG9cntaf5St4;Bc(v&F<>S(QpriD_S
zJ{MyBwIBewT)CKn%OZGqsn6j?6VYyIVc{TGHA!~1urMILG-VTkM!hJ<5aF=4lM58Z
z17nk26afwk_;mkNS!zer3sb+>Ff?S?pZZYnx#}-6q%T&>+(Y(?i26cFaki6s?NaNw
zn`K&{7NhDw$~5I#hC~*lDfUtw{0vmx(DN2c$URz0t4ggfV6e@RJ*Nrz5ofS$gPqMe
z78wBIY{XB$+qIBYxwUYAmt{M6QhRIRro6RqNY}#6x)%N*wiY(u#w}gj4~rwxP;DCe
z{WfgPijCTxAno^-8d)6mft^~$o+KfeM&RO%S%x3XW&R5p*R6MPTio6W>hNy%n;ZnK
zx9GaH5|n~Fd0xm<$i9*>(9%IpXB24;^E%dTxY5yVuXMgI2pyuGUum!N$MZUWn|6Lx
zJO4-V&i_!)>JqqO9Ud%j7@_U~zA@FP)#2OHYxjJ}u@Q<M&SY4J`3hW`d0f-JxIhL0
z_`;4#U6yFs{)aE?H2kY%*Q*B)9rFZJLw)kfu+dK{<>9#*3P)IgeZpit!-9>vQg4sY
zGkJubRYLDjLjNRA=#L|&DPp-daL!-?ijX~7Z4pv}i#nVLtrthRqkCCY1#S+RfXE^v
zab-jJ7X07{CRj6%92l3lMrn*SLMwCRo*OfG^B98<)2LL2y%vk$SzG}l2F_=+NAfH4
zNd8PE`LmSdKZ}uk%NzIix27ZJ9LD^h+&(Ny_OLUIo83e#mYdS8C3|;^+VN3qL6Duy
z$O77OD>EMG%I<KB>S}k4F-<upjQPVZn}7Se)D$*C0WRdyAe)3S@bsGaW*^j>*d{93
zyQTf#koCq0l13;LDRS$2YlE@DM~8%e!=1FV`*gl7!~{_qMEYd#z}1f-o;bpsU?}H;
zj6Z|30?m~~c8v^Ux1M*m{oiVN=eTKmHMdraV|B59L3<4CMR^SE#mdljD?|H5oS}V#
z!P!<}ITmuzpag3{bqdQjEQGQuY#oLyXPgPj>*zYbV-8T8?7LQXhGOF^&*Tx|u@jNl
z#a;;YY6fSh=&la+$;8d8V1YL-(>SPBE}TTYNUms^7At|-Uo628OOZlQBB&R{KKvDG
z@r_xY^mArsEPZ(1BQH}*%r#W|-mr7(+T&C<nE=B)%wrU-B)t}yxF{jL=4grS+(prf
zI<IJNn7+yl)0f<DAB;8p7u<n=gEvZYoHsE!&YSVG?S|mhI$mF+<Mpoiczq7X>l$|i
z5reQ)02cutp5{V|@HPyRfc2`;zioD;C?f!>DVE>RepQ;0<BX2K`+6!Rc<;QE>;?M^
zqgw#POF+XnKriQ(4OJTBkyV=775X1w{9*9QSoY9yqhcrKSjD=sV7n8ijH(S?R;|AQ
zx#O@!%;y?OhqKJZ@8Pvj0Im;JOa#6b5+d9PLp<0?+94>p?W|6FhH$(yN8{^g3qE6m
z{SEDnq&K;d)M@~|SqIR+=m5GqK7gK1GwpIYp2opbu9wDUsDOZZkh!5J!G{dfAz34b
zMs}2{SgD|}fZOq*(^4bDYoatYJ`49)jIa3ahbKx5;d6b{7p!D=1`fCrFf*YlbC3pq
zU<GccNc*B+V)9rOvAZJen~;F4>xQKR_gk=r8Ii9B^HmkD5;{5!t;V5wfYru`!G*x>
zJ}|yAW~McQojW^Xgrk{~oH!tT$e&xqD+<q3Z9<VSxZx|*#=WVj(TQ2w*9@yC<1bN5
zXHT+qVPBdG4I{88*>bu?-tx;aMex`yk2jSi{#JhPCW<jzSisMc7O^Nm#Z;xuk0vp%
z)6=`mFOg)`-mqX@FWJ_&=p9xCj_Gw035Kq*hBuluY07KAbJn8)TxGeK!x@VO6cC09
zhvdhq@Dj(qLsxxft!2fI2TKf(Ek}mW9%9)OvN#xfkuqH8I)T)JFet{v$m)hVaXZW2
zaDUsf9jm<Kj-Is)Lo(bov<$4(V~!ek;sxLJ6?8wbuRv>CSWw6tT9p-@`mNrQiPK({
z{?P8|i;iBFkk;*d5wWLyy1mu-Kl4`O&+2OYIbDtKiLJ(MxVr^k;MboNY4KB(7H{O=
z`3`M2^R+C8w*TS_;sN&Gyb@2y-^TE5`2vq3Y@QZp$AuzN$!I}nLda$2hU^V7@xVg2
z68`daZ+e;K!4N{-;wKBC`6MBqF6{5wpb^c@st5i&S)D0jC2gb&6ERtJ-bx_iLSX$~
zBinlGY$;&Grd)eAWqKAK#ym+#BghDXLT;y~W(^-1g0c9*deTfN7vsDOqOQpPnDr#F
zCKB-jHJ-UxZiXqev-V+$*Nh1U2OgYI7a-=Wcewa>beKCC+r!Ln%5{vV5YfOWh0WF{
zoYY`x11icEqkQNj46~ZD6LC7H2Bj6t=&&Q1@k58F3r?$uMO*jXOzF}rF0-}<jvjbX
znWLRDZ0b1DA(9Kcx)3}8-s5&rw_ikR6Y|Fy@7~$#*l6l=<1WeT=EuevkMr7rqar@>
zVe~|z3<hVYf#cTpX7IQ2X7Ft~gKyUv{QKAp{>N4nv+v0Qm}cbk`gcyZq6W0{f6-p&
z-<8+-ztYaXTRZ=c@y`Ey-ikWZ?%kNJsK^rjhxTZEZyt^Rq%^)yY5b=+jVC1K^d>A)
z>q^^Ul#?T4V!RTo9|9!6%U<-helUIGRRa}h0)ai^|LqOg(H)&wr6X?NYZC%PC3pu!
zi7jKtV0Aa9O&Akt4rqu;v4DUMIJa3s!obQRFhuN8G6ttRQ9|y`jTSifA^9&oIX3JX
zVveLRQv)3Oa9acclpn@3@XSzVb?HqpuZl`++~(9S9r2>{R$i`PrX;NGf)LhGZoKk*
zDj?9mg;2uDZ_KK6Y&8zVRP$qTz9jVqv}_t--J(nh!Fb^CivM|Gzeqe9>?a`yi5D`n
zCz(xpu6k*0bd|dRN)y$QI?lrP7oKcgjFGni`;oveizh3TMY*<f$~2h4*C?2aSLkq@
zTyzzc_%6=<Sr8G7Y-#R<ubg!zO>~mhCv>&Wt*rmd#o~d)OQU&N3SgWMI29^029PRR
zvgGfql(tLXxPg__`gzi|0cRw2S%OfITB*`n!4G?8Kd9!|MThH=4a-6Cg~kMkurrg>
zCUq0#S8Ggh`N2&L4=d~V1ICDHRAMw;DTKxq_<Olcq?$EcRX_wH4p01p5z3Rk=2)-U
zi`G>2_9&tVM8iTV=SO2~M;z^X$B<Bo;6+i|I&A-tcNI5+kU_LUj6IAX1RkRtFJ@LZ
zJTW&SS?RUtt&#R#fDMTa#$$}Z&@j$y5Izo;2xt(>jMb#~k7-Zh=^8LlngU7R499qK
z0;^Y&q;BGegKJH@-tQ)ALe%SB+dV9^q9gBj{dN#OS`c!%!$d4NObo`8+g>xaC4Z^O
zqFTMgc+8OxgC}4zddwMPxTF=PoTilEV&L|cs7f-ct=OF<Qm=&hU`~j6QGSMPz@-|7
ziD?NMq4n9utTsWJ_zrn5xlR<heCfEXF$nRDD9ohv!NAz#`BB(Vd)Y*OKE$;Liz~f$
zwKjxdKyZCHV#xF+Yhw^c(Hpn};2{OYWos4M%jh<wlVz41ob6*efXYl!?!v>$ZJ*ga
zK-2b#U0NUB<))usPo=PNa5c`7>oVcb#zL?0Hb{1HIFwhYYhUZPE?nTNqTb>sj(;g*
z-Y$UHW%M15!8xC9UrsyokvZzES4T&~-al><dfmXLj^u(|+QMZTek3>#taR2fiACE6
zcN4OQg~-VVOmcRHad0#N)T9VyWQB-uG35B$GH{X1WhyLy{aQdh1Y!?0vr1A?C>wG~
zX%R#Q_XVS1FfbZ7#zNmDySHkPYkMW~A3~X$ludHp#4H6AAk8Oa!!Y(+%Ap>UoWh21
zBNb%x1_e(-2<3}N>j>jB!PhHvLPy3Y@?!oc)4D!u@-I7~4s)Vj;k0K1)k0r31)Bv=
z>Tr99%h51$<*i$>9ljyDL!NO+CtUo@MmtVqz)lWf&CD!qh9zZ;$})*bQnt&pbO@8j
z!Xv$aBI`5SSi71s*s3o@$jA|zK@2js4LRNf#^E5EMw4JeB#ATY@R)&c4=j}sDXicF
z(U$>^n7E4$eb^DOUcKx?RtE79A?+43LfQ}`$eH+sSQ{CY(ilQ$epcKdfZJ*t-Bzoi
zV?wuXCawYlAEkT__B%R?_<m9k;@2{h1_W1457=v510W5ScHzlrZI)R<=+Z`sLaVDz
zk-f@0*sv%g#0%X@F}_DiBh?A$EqJ{$Givups1qRO8>}S2KS~>XI2=|&9>I99lhuY`
zYdIn!q!q>+pGmeZJI~QvjBO*}3x8Y3G<h=|^gPrStsEnQ)5cifw5gN=HS`oL1D0}@
zkFe;96?!?+7^@8pr1hLjiwga0FrT&LBw>z`uo#S56E*_mmr>!Q6H%qC{0lJ(&)DO5
zokNG7qO99Vb6TD@31OJAH49#a)qg8wugZnJ?j_q}9L5lUC@$=@iP7toN$kSQSd}#s
zBg}06%Wl4DH`$qZy9f|%JR&plTRAyyh+x-FGiRJCgzS{P!^WY?NICMLf4z$%%mI&~
z+;47A)QcK;xPNVFa->{ZyD>g+oI{m1Jvq%xuw?0ZQcueyInkaJEh8e~qKgO0rO68@
z$J(I0!j-f}4k^rZ93_@~x*_-23Xx-`h@-0rimFX1(iB>RPtWA+G>RnLI+O(T75L3C
z^(cFgAY5bMgGbLW$=8uK!wk`kPnsj4dZ_si<HXY@#Dgk_b(;ZaJk#XNhV$uQ$Rx|l
zpsF}|N+%*rY?W!m+|k)MvoqvnYN?UY9J;;9pAu@vtZ&v-?!5&kXC5yx@}OwCHhz>j
zAA*k2^C&oUi_D1H#5gxMYXi6{4pmNob(+b-s?d1L5Ev9@w1Q22(>XCNXZwjWDHP<x
zCKP%b74$9+wmIgzK<galjvMqUvpMf1HqG5A%Lz(CEc!K*)3rL+#8Bqe(<%#1MzVzq
z3yT&ZswZ@K1VpD-nOK*I1Pg%`1!Up0zGGCt)C6M|^iT@pI)fn(j%{RY=i_Dx>0HOm
z#9joWwvhr%khF3JmPv+RLbNoAF=6EsuqU3xGd#z{k~PXApCdUn5>6xd8poimKDwnJ
zhDv8+1qj~7U<?oKiu~o24*csm)2+e^ZBNL8rAhRs;T4t2^rmU34a3GK6Ur(|HjAMt
z9|D1GE19eJfnzWbt;QM9gHO%mgVAaQShk6aT3!KRk(*nX^AqL_wpL*qX(~eA&Z+ry
zO@d2_tr3%(p<JM(z^O~#D%vNaqq7{b=0Jq;-m&A9eTD2t<};2-0x<pMwyQu8nl+s3
z4>O0FT(jU=WN+k{?x8C|$c|HKdd7?bmTd?JwhVt(Rc%ZXKj=4E!SIu_O-4C$U1y{X
zQC}PI_rV%B<q0{fH*9X$#fwduo<K}_+nU-`(fH}7Hiw|2(l<Yqjy06F-Z-WR9N_ZB
z>WY&bQOdCKf8(?@A1Vz%xH3jggCmEwKeJ5YuAAXzA+=;NZc<>LY9;GWZH;uQJ;4w6
z5mMiBM294th3MHT<IZB_r4qq)8^Bp}7Ss^v0dhv$n2~TAg)y0ozA@~yF#$p00WFgN
zezp-Zyjk7wg1Fgwk9Xca&wcIERrq-iuCAFKG4|=4-SH3LT||vMX@mUpD!5Dxy6=O2
z++1+Q?#p)fZoB_r(0}f}`|$E}!SC_bwi}v{o|uGR?#}*cI016K?l;D$`lJ}g?0QqH
z?>776zk}0d@H}FatLtzDXeMA#uPOI)ICBjb&YMuJ`9`kzO><Fub@gM>M@-4xfi3>U
z_;xUO6@Jg_30`gfc?s{$ni{V&EqVCH3RxVPjI%v@wm92>Ke+}pSF{Nxr&M+rNN=GE
z$Ojsbr$p&;Ctd_y-gu^^OLKl^$^EK&?-qhL2lobEztkjkA7p?(C?Iv45B1l|{sHh2
zwHxdXuE!hKTKe0TRWETZdsKGI>`_+RuC;rmU0SBss^HS~KqOsqm}%Kzre&uZW0djD
zmkrTKQ_T#pHr!2kG*cO8WZaZCzka~Axi`DbFl-c!QNuQS{a$IdU#!>i=KUyT?KW*c
z+_e30)AmEDaxfbHR!B>!jcOhNVisp=#Ytp)sA9vbSQ|65r+IRABh$W<iGRUfX>3~Y
ztr~rVY4j28HafUtqwC4O>_(b(sAvUw+nbyAN(19XM;bSOnn!tMG#_aiccf|D{usg0
z!FiKmHHU0+O?q^8lZv9f7K^4a!ELMX9pj5(c>E8F>KzRJc_nVW0Y8H?yRX<?im<z*
z4VqKLGb?x_-=udIKK?YBgmciko`1|%g4bbE^50)P9z15<;%#u}W!eYWTo!X<Zt>c6
ztFRuqe;W1C#fG+v-3ql>=I7%03e`WoYj^NkR1&<7KV}hpJuaII+Am->(@#v+Jn<8g
zEAbPPE4u&&Z$Wj@wKn(z7o05Tv)Do@h5nHrd13f+mVdKA%95zd{@Xz27?gYSpvdAV
z?UO5ka$g9gH~6<?q5D2Jr+vEM1^e+;DRh5cb<@7!e*DskMrm;Fp9e=*so(($=YY=P
zn6-r@4-U+MbWk27T||N&g>-NikPhKD0_jj*Wl-sXe4aHZAH)}ABpk*o>9oUn)mf(n
zNAMegbR@4bARWc$S&)=L2S@W&fpiS7I)rpAzY$0e=9Po=uH?{u2%l#`QU)75l&=aT
zVV><nk}!_PTL%+(DT{0J3f&1~NPaL<`n-mmH8`HNMg3meQWHeE_%kQ8ZFO5;eE^jM
zLpRDrq2;nD>KbKy8Dz`YAW<vi%C38cidL<-PK5m6JsNvbu0*=%%d4e-LuG7c$n2Z$
z4y?rfew3sOnW@UC6L$3svqOG}bqcyFJ)X?U*z1<Y598^eEQ{_Alfy>uG<hfZ54_3c
zvm(o9@F^A>dNZf4m%$UDMqM*oGOK3rQjUZ7@;h_i0RA}cLbDiPuqz>_TnXUqAz+KH
zc@WiGhGi@7HXxM<B3m3}R`bJJBePP!skM0k)?psxwLSS1?a9foYNgf7aN;UTnI(_{
zqIYh)nzT9(?G9;$_0Jc43c29SXlWj5<R0yMp-#m?Z1(34oY?nf=roVQ=v6CXzbidT
z#}f@5A7$B7WN2wf@8&}z6)@eAY$h3kN)OiNfQ*lH*m6dn8+TDX`*6;t2yWZ=JCv}e
z<PtWe3`^V=oidF|oqJMbK1a#kYnryulHD>7JEcRqAD^f{jrJGK0qqw!fO)i!j2LA`
z&3Q@WMecwX!+5S%jBli%Z<|?J4Kkd9&EXX_8f7JZGhETE<udehghhE~w7-Y`k|sxn
z7hx8@FB42L(90a1u%^)}je@v|g0p(qeXlPl^hA}LlUmPlQN*ta{vXsEL!|iH&^(w#
za`V!dPdXh|cA02vu3KgEBqxeI8xz5@?Af@+vwDv-9xxs{|FqV1PBb$wMOpGb&DL^c
z2@aQKk!s|gU5|ql)nWywewL^dTeqdqmHfzxN-5&kxp`&kqzGh%mNI39wq?Ep>tgf8
zJ##&7<2K+ub6vhx<csDpMeG>RAp%Lfjb}kJ!q5g1C@Kb)#GBcwi`N*7Rq!s*A>bSd
z-i@nkWn>4d_^}gy-;@-7H|9}4CT<AzMcnYHpQE=@k_v`cQOFE~3zDr0Qt&-F51&XP
zGK}R)@QF;)8NSZbPK=~9AZ5yw^BN#>iZnp3qyc06sP&Wy^2*c|FxbvlrA(19=9(xC
zR+lkWuT34R*O;+7&X3(#O(ErMUb96q5mL&PG<zpMGR^J{QrOSa@(K-Uc`09&LPcVj
zYg#^w)kSHU)&Q<TnLIh<pI2)GL}>6D@K?`2uSu{sQK{SEpD#?(u`PvI)aZ-(@m^A+
zE7n?Riz``OyEkv|?%{DUo148wXD78%`4<ROR0+AFZryy$MiZ<GThw;jFSbui0T2Ge
z*vf}T1<wccLp-ySC8Bg5TtewGv*b$Z`RWSXi)qP!dSYD9dl0@fnxD~!u~vkuKwKFX
z#rOW_q1+V3_c{OMFvyi)xFy8kjMX`%AKaq!!qZ&oawVmIqvO&KZ&CVHie0Xx^lx@t
z`a$v1x!PlM_aHg};{1>9-`??DG)5}<IV-U$*TVQXV)yGf>Zh0R20Koy?SIQGD)3|P
z&;dhS4yCqX6Cf8(KZwm~TuPmHDEtY%1I|5*y#Gv;CON8JSc;rwh{2+60@g3o?D{XS
zNsM*lC#wa+L*SxWCe)+1fE?2%C~;Rnopi`>MRSW79_)KwI15EVDBk)E)atGdrF_@5
zz<U*X(xmU=2~jgP4XtK-GiJ3+!{!lMaeq-9l*4VEa9F89W@Ezqa%22X^MTQ}S*uYS
zllFR%HvI{EC+@&f#X1byBSW;>L3T@RNqkU7m|K79hh%QxBmRC!OAtTjR3PVSJ~%8Z
z)>NZ9p4Kjs_}87lVNV!1M}=@?@XCm?L`fJ{S`o$nvI7{+9vt9?c@Xxk?PmRGdT{(_
zWFpPB;rL&61P5D!6Xha(w3BfBXPv?Ezw7|cwg;OYIxOrVxH*nh$8GN<Yz(b>$9>g7
z-NEDFxsw<ma^BvOr!x}xU$zC~CYig<!;NuAyn{?2WZN;vc|?(zT9H>|&G~l7ft;<p
zNfmGhyeOg@%dOYy>jC51h!kLfJ)np~CZpRd8n27oeSvUC2*EMI=3&<D74e*4*4>F+
ziHJqIc`$R|PL*fsvoqV`kIV<f8JsoPJ(w_>Wy&<IS`ug)A)lFB-hyc$?f71@9YNB%
zF4TRRqpsTlc!04e0B_HHoS2wJz=>V+chl78-Z%i^n?~Svzk5*jIXx@jt6efy+uLv@
z5o?}xvddZ9v8?072M=?|>K<9tBM-2GyX6mI*Wkqu=V!BqW4T2sb_I_Vz_qc9adUjp
zk+uk87$0?}Fr8p@N+&oqd&AM4_rCPc_uKBD$XE*&HM#pi$EA?yU5%pV!%}dnr>31H
zkIrnB_qMd@hkx-l9U8U?VexDuj!kQXJQmbHZX0!QWo)MG#oq9cPfcVZm&bzqr|l!Z
zsvYE8z6_z=5)1z2Qx$~lhKE07<#8OW*$($loUXE`MxA3aeQ@FU916OFYu6$*;-V#s
z2N6qkviP*>SxYW^s=W?)ed*3(_0qz`RPS`}PCV`lF1r-V9g~G)ZcUYII~OeI>*+!G
zANc1F(qouXdKmPSHPMwDerGs+WlM0c1&g3hh+`yt%zLW*%e>lOIG5$ki`Fy`NG%FE
z33r(-Q2sb+3I6Qw@9*oG7MP2cTxRqYd5SOLTi1JXUm<{7v;l4%_z;?Q$%6jEyvWUY
zJ!r`k#C5m}BjP6&kJ81}$?zy6S&E-J%;_aY6SoH&a2N$^DvmZv>`tDvtTYfj&(bKp
zzo?nF$M&r6mDxppHbzWj8NJS`jH|H0fb)5}H(voqGHcEsI3BIa_{g|SEGnaXC;B3|
zIZ7~lb6dWcrYveb|M!$do+%YlIb~64vB%1yEgCy(P)ea1H?UYbk58jKZ(!6a@c`8m
zo7Bloq^-o;=T&w=SjKT97u5o<=Bl9DoVW`5%M^~*mI@lBSVRa5ZiPfB<|cn74Q}Jz
z+wo)W?uokUsb*ozd<Bc#I!_l$|MEI~*{;IsY0Y6_9cs;~A9e!bB60(xHHT%5MaHvL
zhH|v#)?ar3r}?0X5|ZioTi=51Az|r>7o@Wu7iV8lO)cYd51sR}rZd@~-M{dy9XGLa
zZ)o;VeL!dxtbkTxIVvpQe<q9muaZ-wd4xL)P!^qv!51P^U@$lX%Ar5do0qI+rC2yn
zD}HM@j)@hQmNBs@iS{RPRAS52uo>5(UpaDq>bYUF4uxkBWvwnXdi-O=?Jq~ZgG2K;
zt5)in;Brxp!ZHerrP`xHs%5?`Wc+m+8O<7Hw$O``uQW!+kQa5uFKmb3kG$Q2eqadN
z+!b`cNuq0^z4<(CFj%$B*AA?<7MxD+vV`v0R8<WX)6;@Z7jvhknM$xaSW(f$Y_xxH
z+=>5znq%9QSt5hva5K}3)v+R~qZ(<cd|XK7lvjl0zw40XZ4Q8>8%pya+RKMR2`X}f
zSeoq#Pfpb8+&bSpWgtv+fCKHm0pH4&l)DWHjZ}a|-FN>xD!SQY26o_CnLUC>tjx`X
zGv=6E9QxC~iEI;fxReXvLySMwp?Na!Z165DUH)(iUaxCe>MU*c#utgSrhtkVg^^|q
zX3k|DSiVQ&QBnB2H`9Tr<?)GYW+RfjwW{Bw$32epDczW;;Q+lo+tJLis;7SFFM0d*
zF4!&J(`La-4*%eP6vqAo%a&VV;DFM|Xutev^at6KYYv`UnP7|#lR8?CFJo~*CTnP?
zPy%uG8#xGZ_Ob+x3gIHes)*&NfP!$&6g&+n8L_HW@#3iLPzp&uWJbhCou9;4(9Omh
zOX+4bp=*pLbRao{ziwY=G$G`cu$Cw1ru=OV7~v-O<0+T!ch7dFDMVp)|MxT^HS+2H
zSxAjN)`_%mtWLSg%aLKlaqZ>alqDpT7~E={*gy;Jw>&DGb!g8^Y?>YAL5$1)IXg1x
zUwRXK(3{**GrNVs#ux?Qp<ZECohtrr)#-5t8kMrMzV8H5sGl9}6`4@b9@91&|I<!@
z{7JVIeIWkcmn{G71MxrY1jxR+viU&#U9S#-_@8zF#BG=z?)8h?`w|59Y+rc#yyXaS
zpS4x{O`>@P2P+vK<^yQ4J9S?;z1>`v{gbaEe5|)S)x2pL_n9Ubx5$aNRFSAx!E}cR
z3RtW7!0e8J=YWA(`<{EhcZG2rKyF={3f*lJ`>LmB{p<{m0fZ8?DXGvaK`~je_nm!a
zKP})tTbk?&BAdw6jD6e2RC58241(@9O@fsWQ0!);N7<_50(O&XAFh&ZiAKTK9g!XB
zzrmorx8kR{Khr#VyHmk8`RH5n=;Z$BqucoCc0M|2{?jC0++~rSt>D{y;XC|sCrn(0
zckt$S`DxGmNd%!Uj`P56xC_(#q}%$^X3nunrS+&%m9q2I6Na(xsmZu;HiG0XSyxKe
zVK`n~DNRDkx6+W)xz@NF_5S25A|>sY?EmHg)<hrEypwyE8~_pRxA{OZgA^e#78znG
zRsu&w^!=X76kG=+s#)~i=Akk)9HF8Q%@FxMM}r$8DzbNBXa4~UN|7!PGbd=)(aXtG
z7K5{%hxdhghkG)nb?XoLWQr={tnHsn6E5SmSrHT5i2-@AA&l%io-@|pvMZL7##z_n
zZQo;o($+ASM)o`#`$OW5iVLd`NY1^W9buIQUVj>OsFSxhLt3ceM<*I(v}>6T2c|TL
zTqEeggUI<yuG)SR+O?aRl^}iR;P1(rW+nMhY(6eQcV7Y)LG6re;W@K_79v}U?LhV#
z#e#p3S|Y3kp)C&RRPy!=Z2#Du^!#M4)QANk!9TSt{uUUgGh{9jE6}=K`{h_N^B8Gz
zUk?YK$(HK<i9sJa+{T37{{_Y?rmyA3^mR$pnwVlo3`QVk7%rSJFW585p=8p}vC)BE
zpP-BPeI_pmDinTvVab9EcO6ei6doO}mRp?FA-jqf(jClaslHiAz)atVJi9}jiBdeE
z$?ZtPVUtw7$*9NID%|7(kl2?zIA!^5#$<)Rt$oNG>byj$V*^5Y6C!{A@z4}<y*xp#
zkT0~UV&q^BcQmp@sbt)P)-0{#?G2|+cd}I(6glt7!PdQIU*+35qNF`otv3DOQWkPK
zf3D=uv-$G^{6vJ!cWbZCOH9#LTZa!y>C2?P8cLlV^`${)wj6>FD>{D@Q3SB|L>eam
z{Wb0x{x7Vw9V#obX12e9_?Gx!`YV`jI@Y8uHZa`Y-Azo&1I1!=Q!&TRpPXpn_x}3z
zH@LU>B@BtIy~XRvI(&h~O!TJDo<lpDv8)AOBj*yG_fh=3`)$V5LwaF#|JY_+S_rAJ
z$I4CB*+Ua&Wv#pwx=$SMO^Fae+=O#v%DN*;`+nR|^ey7z8n}P_u58uaKYBOay0ND_
zlD<^#wa1hDWesazDo-!OV-_9kQ`%^>BEEU$R$X_QEoZAZEN@#N&nl`DprIMYR~o$1
z6SkQ4m4+6jbOJBj(`*6H-=maX?h2k?N+<Bboz52U{GH40;rXSs1@9)KVukx;EvAT$
zLc3*lbjq9;E9ZgHc{~`%-sC~?yA->F9^BP4L{1<-i3dy%<`cKA;hsF6PYy{w$>bHm
zq5R-eXlDGZ{`I&Qbsez8IP6M%A=bSXedX^l$@+I}D{35u$|D@`aH2UPepgy{Bp*fF
zoY}I$&RR93TcM9=&DC}5Oq%tOR?Rvl-T-OVvF@%4dsK~r2eVzBbU^diV$n;3=TlpH
zWl<~qEk2UA9M`HX$2rt)m`Ph67Qg#&w&M8s-Oq8?>;!i=+U^3s=%n32NxR4fLXNKm
z#o~^MiO3eGwC&-o+IFHt7HQif<9B8FJ&JJM@Qb$O(Oyehx%@?>M>CO{#}$jnJwAgV
zj*<<=rGk*%U%NZIq*Y&Ld?;-`A#Ag2+ycT}=o+W-_GA#A<TW8_XG{Zs$k+rE0Z$=2
zP{TRI`?|QdW<}o=c#lZJ>nDW8>F`eO7~Y|b!N{!ZKE4yvsTq4!0G=2EbO(%s#}LJ-
z4xuCKV|n*9cQ-mR9>-_xjd5!t4ZS~GtTb_@tt(;7E95$^a3<D5^KVdd+jw2J=Jr2v
zuTYcMqIuYcH5*p0T{F0{xNcy>hBbq$t<rWO$`hq6X<du@G$p_HC~4DN*pU6l8fQEk
ze$xUtS_>FlxpKv(;>s;U8wWO76vK)l$H}ZkAs&?Gep|7!i&^2pW?SLG5jd!cr&|sg
zK+GYH4kOEZP*xNPJ5-bw(Qkd%;59?=i!y$}Omj8P<j(M`G7<9O9NbCVB5A{8O&f;n
zc93rrY#zd7Wv)caKp)~`>yN%56McSiqBjq9g%lYxDdtxa8hafFNV<wX=7K%ieV3Ll
zbuEn;P*1>-@>G{{7pT%k<X_QEe9CCAwhY8?!$F6^DHGpiN2ZhjRuokNxsr%$a^=Ff
zrNjxmkv^Gk;5T<*zy@!`)nM=@{{3eBAQ(+7f|V315v;r@Xy9}2#M`X`Dss24`H*R(
zKrD?{r0Ws25jlx)cDjL+mcH9n@MYrp3<|^Kav|+fV1FeKb}Z<Xz?Ps_oxzT1{=wJU
z1NZejaIu(K0#0IP?FqPB+XMHlJaDm)TLMl(ZtV%U+f%@$L@NAF9ykfmB|}`U1e}EG
zGT>4p75*TvP&GUSKjf=Y=xtozk}Xb;!)Sw&J}NXtATg@j6z+2os%-G9$Z)?ag)S*s
zB<{{@f*yVieyvTohga=4!H){%W?l-E-{e8jz)`_(70U1C1j_I8ps4LA_=7_E<D5Xb
zHxG*1iGuqS%Ae;1%0ApA%i`#26AJd@t1{g8=ha>xEBDKRV*THP`zwwI%n6PIa&Uyf
zqk;n!%0YV$B_(WXPaYI?$`1}vC=wW|{oaxFwwDYDN1>OxBop@UhPbORD&Gq8LWCb3
z#yVrmZ%QkU$ZN%sd_nm1QQ8WLE!ANw7RO^x^%dxPVWUkk7c2u(Nqnhf0O&${jhUC%
z7!5cToT!bF08?ILn&0Yi{_i$~Od|AC&(hGBcxg`$&mx3*S~DU)@kO{mhhd>0=)=id
z4yvnt)Rkv@8JQlF{_QM?{~-c*^uN6ll6!-!erDloXWY5S-DibCw)zS-u+~{a^;sMn
z4JGtoNqc?yn7qC`l`qJ^e604R1f%l$(u%X0YyHEbt<6AwR(r_L$wU4GMLwX&CFInc
zAz#xT^7HbLuT|vh6uE?*>Ja%^QJUv+wv0~M++Ncz$ZOgbZQ7HyX%d&pYuc9k@3+rB
zJ^u(SRqgvb3oI37qI;Oz*6I*@ag2{7$GC~#z2!fk>f`t};vAKDT+YNTX*)b|>vc?=
zU}fo9*E%{eSD@v!&Lb4_yM+t6zJ=9T5hmR?q0@@&(uu{<R8?`op~L#@B(Bx>s7AMV
z+X^5kbh{w8vbfMx7<`xNh129xl<jgJbJODnfn15AAVJOyb4yjWgS<jRWqZ4FOo@Bt
zam;oW;L~I=kgka++fi5LzQV#1j+f$}g(iYmM5wqpIYZKtCSH=)rclp(nzrduUbWw*
zHns1|@}Pu<fTt^z%jX2jGxDIQS9)-TLb-BIpgc1Vih7<0S1FWdcMc`En%|gsR&HXY
zXh+xN!BNlh;5iCM0$X|GrTNaDzg&!}*vhtdmbOPG!*oFh;6i~A#9{hpPw0OUsgtjb
z?#`pP`l<&nReDPVERWvp@SNk55wcth$YmnD<u&bLe_bB-*YgEYSpHeDOMEO3d-GOR
zKH-p_2xirK!Irf;qFb7S@+RmJm%+$GPw_PM=vryl&ZP?%%Hqi{?X6^NtH|KmE2`)6
zsA9m>P&NUfJHrfTj^gx0X|y7mrCl&aYZKUOuho;5gLk&qc?_A9X=+n<`rtj<c@k;M
z>%7RoYmu(yfzbv=2HFQxu%_5;zmkVlcgTaUD%P*@s(q|n#DZJ+O@v2Usk5JIP(td|
z({nlQwyg(oD4wnRJCHh~@{3GRCDC4yI&B5!bT%jILJ^Uqrf4WsK5P*V)qH0}suo|S
zg+}da>SqeMV;pU6Elv?~Z|CzYYxuYMg78<l5+PT@SD9g-D&*doS7<2YexI*Op%Nac
z)m*wZJx$Z2&GKkOSTa;A@2c6)Ip=zd=FeGUY|%{N2fxf~Lnuk!rET~Xui9@zn?mi~
zc~CT{Qt)eqa?hMV`Ar^_P}KjeLiydCK>2+hlyFPw4+`ava{}ewJSgFo(w`K{eRBfk
z&pA-smeM{vBqH-}UtaCmm`#!K_hYG9d?nn+*<YbZ2qh0n^UpnhU%`?h{C)3e4c+Zy
z9Y~tTm|#jT_T=@VI;aN+YcERJCa)K_ETTEh?fxihjz&LfExvRwgKJgJVr4~cDRG6_
z-7CZedT#!h)!T*E_hk0)VNR^1*f)`W-zy^h=b{pC8@7JxkW-bNMbR=FOOXL#23ZOa
z91H3|WfC^G(#&kVvec#!IIDozSbYV!2v1JWFbdM_sCkS_e5TQ{`ruPHi7XGr>Li_)
z;Bq^@v@lYuP2i<frHMwR4|$wRGb1%*-$s&~r3lL@--jWx{9UX<d&6>G-mpB8FNo0c
zNL~q_lW0=ju#D&xE#!-xHrkZcV)0Leh@f0;J6w@LJP9ckp(OlTG=5YcsEv?pPTR^Q
zT10_ank0(V7(SV5r(UBclm4MHU?wkIsC>XkgEIX?BJCBWL93V(OZ9iSsAmZAXXecZ
zi3%q3L9S#zNPI9eA5w+*WqE~$Li}=_WD+*2!%4O`EY<6ib0jUN?UVA_6k3xuYMUf-
zl-H*J>(->_%F?4Nj{Mc=--e*-Owv=YjPv*dHf_wglKyj74dMPqy}gb`2z7(#$`;jP
zyJ9$H5Ts~!BYA@$)L}+-5R`f44T8w(k+e#+so6=)i7Y8OR5CA9C{LdgC=&T13re_M
z@eGA>#hgI7lBH%r3AZbrsZgHPIh5ckeiJdpr7;bO1d;_O+^)D<;Yhq2Z@l~sZ&xHo
zw>Z#szFl!0sgtjCFz8BV_lMgRFI0L<z#Na>IrCFm(8+ei%i6=v7$2F~LyPxY6}tq8
z=@5IC#rs_;SX1~QgHL2)4K3a`D%P8L)jn3Sc;C!#BGb&uuk%wa-riw3?5<SmVXV>C
zO5)$WjXO@1y&|>#eVMw?<}k~UlE2F5S$tdqr-+m+S0W`#2o*ElQ>El@<`o)B$=~9u
zQm91cXf=sisG>&dl5wu07wz8tB2VdX@O(X@uUUJC_oC`7MlFiIo}wSMkh*@rX2rOA
z3b*-D9-%@}=f_H@pYW=ELbWM%{WK3sXixu{LizceK>0-;lu#D^r9!!DPN4iM4@ziH
zzgwaFdQPC+lLsZVr~gKw{C3Zwq*y?Imj@-Zfc{>gNED9thph-B-`ks7mCj{r@k|_Q
z$Wpu1en%(sK4i+NNO%LiZOwWW@wiWq2oZlb>RlGb{s<1ZF3CJndf_0xmc_a?Hb~IJ
zSB1+;^p6gE;S+nM7o-rlR4@xeVe=Gp;jK0<Q0xAvmv+g?6L;+7LfQ{`rlA`On;0a)
zQSDLpm^|ugjFRA3rLF`a>5#e??S1Oz(A5}H!&oQ5!`mat33((b@CA_-9-$<WSSB5k
z<m$aolK4pKV|rwaMkd4>tr*LHUqO<ANRkiP3b2II;3&aK?Gb8G9-%aNO3<%_lE5f`
zc|xs41d*-SkLYNsSwd;-mEg4Y2z7cMq0Zn7GNP6$p(L70hlE;dxR2SOLV_6Q>V?DP
z{qhK&Ek*OR(Z8v3(QE~yVmTrMH>zXRN`1YAfw`bMvuy)%quSPLn3mw|_FDUdyw(nA
zYnN$jB{)lmt$i@Vz!Ym^5^iN_LAsPbI37=*k3b{11glt2OioQ%YR=_rSxe0tz99Yd
zL|(~KBT-vAM1G1OpQ#mH7#M@TkEs{skx96N`6`XVcAaaJMt2D|q_inT_GYk>%r=FZ
z{w8hH5U*ObDQOcDO@A}LNz?TIu~SvYOsm#S7?O!v3E@b>%uTxG98<~u_I9dnp$zPF
z@V5-^yU&6vb~9GW{Re}7<QNYAi322Sc3sF9WOm7w%&w>MUaOH9SC_8|(zL)7D8)P|
z5)4P6$dy1Tbq*yM;Wv><5u`Zcjpo5A^96w;R|2Qv!Aap|)x0tTEpt1sq)Z7+(>}FD
zd07z&T;uzEd-e!;6I7=w6UgUkz4Wq<o?JpaQEpud-Z6;uT9@|hwb(upw+#HTXPz0?
zNT}UljL7X2Fm}%7`Nox`qqswR4CVtM7FLXtHpoA(f~&xwgBA4SW^`tnfJVQ^TL{YZ
zuy_<x@$1J~IK)wM$xPq(d^A%c8OaqRKQ-rT>f)tLz&bl(A~3~L2q)s(f?yH~%Fi=w
z15$m^)zVy$g{eRKsSSCoSz}GixxJq+*31$^x;sGtgagu*w0Vq=39)&jF|QNxwMGz(
zplC@PX^#JCB)-<)UMk&>p*LgUuI-<aO1GH^_a#JeTq8o2-I3UUb_5+vk`F|N>F3=5
z3O>k~cwGm>Zl2msPHgEvD($IRv5&TkSg#*F%*%>>lm|Z^%=tY;PhNZBZW|YHPJ`9^
z2%ATSnGLjoHq6-x5y+$4Dx=$(Ufp}mu!scu>?GGV!ddf>dSxe4I2%s_8xKt*9ZIya
z2bpMPM|fUjV0y_=k~Rq5O=gy!dl{(7wX}J--vEJaBk6viJIKN97;q+gSQO|{2I$dI
zpsm7oTDS?Oa<mUO&QlV}dq?6tWx&jhOqL>K{is3vQ7gAx4^pM%+xn}^)%^>P!DxaH
z8<}o-H<9Tg6l@uo?gX<og@NhfpJm6TlLb^HJGl}mOs>rO(winO{u}vPmLM%FrxYqz
zQs{^HkrnMy)I_r9%rbRx3qH<QrOf~E%A|ajv+*u!;wSh`L?X7b-4#81H>G>G0^T(N
zM|@nhxYh`G|4#Hy$4QaoKgZ`;T_bCfbd6jI8D({H{hcDp-;`HqxX*YqUzI{-(XjoU
z=$mBgM3jF}mQ?-~79Ew!(>!LrmIwA0z91C%x`KTl8E(5^uaUK|nTRqISuZ?R8{E+z
z^6%v#*Oeo<Q<2M3F=xnsmV!KG@{;vsj=FI2{!)?O#j94xTiAGh#cv{$*J?Q<JspZj
z-OkSRz(ZF3`U2ra)vqUV(G*vbY_;Sssd7C$RlRYduhoW5cmdib0+n1*8)!a|joV-q
z>rk~a0^D_?Sp~7RKuY*5>10FZei+wAM+!c|+adxSXheX6?D#MR#sw0y5U{}JshKT^
z2=6WAk3iOiZsd>PQ-m$;y1{D~2^4%rTl66()YuXKD3H^sBeU+J6#N4D44PfVv6W@t
z^;v&QaUl`MHt}7yyda$~+*7WE)ytJ_SIU|+k0U2*O^OW>DN}}sEi;AVkbIeCh8!mp
z7aq-zI^;Ni_2NP%LpqTywhYONIZL8*_AkkI9-2C95xqbSJxhsSrj}d@)7c|KPZ~RF
z&Tow)!a8=h;#p=)<#MSdZPt^?qRi??5iv_C>W(Q5fZUE3SsRCMBMD?uCR2iix&O?S
zblAHnQgv_gIHjHS97Cn0!Q&`9(_0xlj^I3yg*k5ScxyuY{$De91nQ#OJkG<Xwhh~0
zYbJPm0%*2nBWe_|){DbENwp2@g&tBdi@zZtU?C!;bo)E^2ve6y0Iili(6yG$2Gm<G
znpQxH>Dc;F_KGQ-OIA#|lGRPF4A)52u+Jjj>vYLHn;*4aGC8?3%S>A`mw9C-w73>R
z`d|H$nW<r)!xm>~*sJo$sEJSwGnD=oX6X7mMZ-QXuh3A#UdvadP~lUyzY~4ajfTCE
zMdxeSLwR78i3OV#tnjLK!CDPF>zu*X_OM=<hgJDm@KnVrJZvvwEhn*BwRN~HwH1uU
zEOT?Bw5}$)zdfb(xVSJ?p02q)EUU6!%a+NiBv+IkVIj2BFruW!tYY%bDD!w(CH0HY
z<E)bUC8~6hGHznIH-vjBff&K0j0;)Dp6r~2v13x#=Ez2r#Syb_365d{N4BKkvb>~h
zD0G1+X15YVr8z>>YEvUZse~m#tWw0)Ovvcouo7h;dmAJ0EL6EN`mZ^Zgc|qo%8oS-
z)8MzfA=Zev;oxl;)G-;o61;9b)q$7tOEKp9GWy5O@Qmd(k!mYpCj_h(v!iwQO6iB6
zhboe#H;<4kLnS&Hxj+MU624z2;doe^rWdR7UVxuSM#SyBI}yJ-$-7hhaVhECFs<?C
zG=7@*8>~ST_43O$52OPsv(4=uZ0>i;sq)MS!dvjcK5OX?Wj&)2vA$eD)CB}OpD9#l
zBsv}(rKT$ITy2!<yBIZOvNU5orsxUcKCCv}cgD%8R47+RXW)CyuXe(pQ9Q235Yw)<
zlOX`Zm=+T6<7pb0Z@vk7G2c4giGyhPaCBR(QJHG-uwDr&qcfFq^6_ORs>lY3r_m^s
z>Xi^cxln4LT?E!xYHOPJq}fV?=$97e6;3LgX)v9Q%XvLL0}|CAPvqsEXqv6nGJb-K
z*dL$5fQm6KyR&$m@ww<(v0=TPO?Iq&R>#UEEWv7zy>U)2CM(%5h0snl$(5q{{bc4m
z(U{!cmd&l3M}{!snqZwMBJ6uG;Gr0^$LdBO4-ZA}oQX`PY5TC}8AKa0W*M2rwb30|
zJ%-af+JkC5ZFvgIXkOWztHp<Q(i^WeA=+94ajm!c8iS)}rJ6^1rE(^SYn4TFa=g=g
zKAQer63jXSW?gGAq_o*gKh`VMjtw7A!sn$0p6I7ZNP`AaB!_dfyOy5gAU)WF1XpN;
zxwQPk(kNlIZh<NGtq02TJF`-(H_+B+P)nit)MBXX7_xS$75%gounh*-h7@2oan-p2
zULIAAcwUs|Iul^?c#nvwZ^=qFlQL78dpJ5T-*{&FyxeF2KD*6%X+a-aq%UnqIqLXL
zzt3aO1?Y|ts9SPn-d$rLyrLrrfl5}*!#py{7`FD9i^KYK(M=MD37Mm$N|F2o=c^3P
zx8>qALdmu8gpfUrW!>0j{Q1ywN8@<IX+q;n?5k|+jA#a8oslb1tmVpz_-XzR&*d^j
z>&#2`*#}D*{%e(>FQ!*W>7rM-(o<{!*SXSL6&l(C!rgXJG)$r5sT%Qw=*AYXn<(;a
z0WZx18`DOF0-}v{2)6y*_Sd$D{PlUrb#8gP?V`2J8S=NJAWzx*q6W~WAurr*e>*>F
zg}j9!&(me1_JRMk-S$jNU6W06mb#QYl<&%;Q%oh2u_-E1l1{CFP&kZeO~HFuYi#1A
zfO}saIMJJgG0ByH6Aj7(w`CISr{^CkOxG$*%%HZXKKSCqG%liWy;j5dTj+205JI_H
zw|_QD5OsDLx7PSs<<SzhdT{6#@49f$a!3#X4XN0g)09+Z&w?f^OK_-b1)okCc`4iR
z|Cu-PL<^IVCs#7^L?1IFuQ$!>>Wg`WhKtsh_^K2t>XkQ~+gY?u4E3JK@s6n69$GPi
z|4E@r3RC@RUISDM3%;fexP@2kHX!|QPzsc<=Rr{oEck{(xphvUd@~P<>R-XP6v}OL
z0_FBRD5`e_-&QE!=^RRM2fvB<n5E^N`EDMZGo7v#d{5!r*#(^MC*h<BjX%hP^FzKM
zlln)zlKCe}Q~SeLxZ7be|7`G4<fp=s9B;Dyf@Q}RzZA50<)QtRqP<(uissZA+UUxB
zPkZ2gn+Glw41cHKM9u09Tn7%M`@AqgR<nVW$uX@{;^zo5BERK)^#N!$Fx+)>cq`#a
z2wiN)DY0#t^=yqO!*XS<28MXnTjM&h5C-#~hBsx<%9VU`^8oHyQ~3fsSDRm?0HdON
zdoq12N(#7!1;mVc)QE%U@IgP`o7sKEZfD$kE^jvZvn2N9dAxZYe@=}(c|LExfIp9m
zJz=%M3;ASe?8)h57cb(IvtmzP%$vLU^Mu%wm+<CG`LjIs<Ym11a{jD}JrOSX3O-pA
zdm;?$m3*=`_Jn4W;8lFGKKA4sqJA}>Y>Yiw&L^+olg+UwzaR}>%O_i6Pge1p*YU~L
z*pt`u=0Ef2sj(+-;LSJkrxbhgCf<BAf6B2Z|H7O9%AfJrleh5ZTlo{jp1h4W-_D;)
zvDe~{kazIrJNXmemKb1ruIH1BVx_%{H{YH1&3kzBz5Lk``{sST`F{TFvQN~XC*;5}
zgS$OS;BVO2y|2qrhM8K=;1y>^^9d<7+i>>w!@{)Vlp)$NVRmO^9vWIi*mcQzb~}@n
zoU|aJe3M^f-^jv{=E@9zrouR_0~m-?fC$)ArHPqc(aNu4aiOo=wUNp#ceq+<aE+@x
z_y7qU;|?F>y$|6hvPt#ty!+w!-H-6@N8@)t#=9Sn-~A8X{RDsFBj=O6`6>SF1~PxN
z3BhjQlb6{iW)uXkWD__Lf=|oGhKVN+qoonzwft6k;dQ(cHu)Kr5+g*?aQIoi^*Q`R
zNb`B#{Q`g9903_wkMoDB?WVUYx_2nLFS79V(A}7d?j{!S-Uzzj11!G1xn!-IX3O>W
ztMHva$L#VIv;GNLzs7p^(5v9T6OHfCL9x9y8V?EE2{AjKwB^eQ9DixAg0IXGqCNF0
zNZU{GIIFjZ*~<6Vv&CJHsh#Qj+m&Cz*C3U3<X2z^`G1aJKnw4XD_KbTvN`y=NGdS(
zs=f~G3O>)`BBGpKDZZrj+{i1b=bTRJF#(8*Q>DpDu~simmu9v_myDZPE|-kpzj#G5
zN0$wY<o_v>FDsHWJ3=y1tC!5V;uIKPQ5av%gmH_)_`1S)Tt_gBDTS#qqJ+3r5qvWf
z!EK7*c13VjM+o3-F;+!HD0p1dtK|x|vN;H_Oo#yS9mR1+CXVkZjyn}cjt~(Yg8ilX
z=r+j_5(V%B1@Oa606$g$KT!aucS8sc&M1VRDTJSALinXZxJw~Cz9R@O187~QNEF50
zisILqD8vy*n89xpMQ1&s7SJew-zk9KX9D;mp9+9`6+mZPf`aWBqK}zeHO9i>4@#A4
zgZna(>@(_+?8__RSDkf;nV$8^jw&-6MCq|V%M@O9KVG%uRS!@U2PlffQT>RL9*gPQ
zFT)~Q`9WHFPiEzZXyu1$<!5&`ChN2431<13sUTKaIW{>cHXABDC=<v<=i@yY6^AR3
z&gLgFk%B{%lf3ds7APHYR3?OD_*5Vqs}LUB5iLaH!om^li;x5l$*li4t^Z+rP(K&0
zR`nmBS${$6e}vZG8E0VoFUkUNVkUq`DS$^SfHY@MPb`Zq5+l`_*aFqdlBDw&WY&F>
z*4?LdceYetG+UZ7S*)Vu?^gtiGZ8FN1g9v1Gddc6<r+c*q6Y{hiXr5Pr)FX}O))%9
zG2{rwu^te)2^MFcAD_Xpqz{&6)_$heewNnWSszI4rHScnu_<tlB6vb3f@O+exgy9h
zxums!20nk48H$gy6LE=Yr9xSi3FTacvPPkFHXMW)<$7%z(LAFaa-L#Xn~7miF|1b%
z104;?MhUs!O5@1+GhTw@E8MMGbEhX|V%elvh7`;4j<AqB)k~Ab+Vo6y5>f-vl*&=c
zgcIfbOf*{*&65>PX9Goq7=%^bRxHiTRHo?58-)@s8(TA>T&Pf<s!-N+L>Of6n;J#;
zK%TNHkDpN9pUEM4hBGmZD5g=x)S3QOt`#wIE5*hrwvDI8Tal)c31eJgY*QFZJ0gu>
zn66^npo5|y0v0J;ZF?q&Nd+;bAUcyW1;o_M_B0gJisGV76b(f&qbNETZV+Ym^e7d=
zQMx0u{)@H#U0Q#q><pfqaL_#!i+mJAI4Yl(iQx=}7!?8UGR4rDG7OGhnJBiJ9hb9I
z84=IO#Bq+|xKeR+HaqN$1<yc;?zqVPEQN7ZCXBNb#?=a=vo4t~mCFnloy2ghVt7s_
zh9;j%mpo4~bSCy!Y7JOu1V?mKgd_9$nGjy65MHDZI%DOCds(YjiX+uhqc}1<Hil5z
zt)T2qhjJ55DmMi0pt2;?c&URDd_A<6I-#mP+i=AG{YsXXV*fr(Ym_Tdf%~&L<O-7+
zS2M*#je}>fz%=9FswcX(oyk|)QMThXNdklh1goTB@NAYTQ|}yJwN$p3@u^UF`5vQ4
zT7p(8idC5?&Q%m^6h+REOVFa)D2_t&u1$vC?fyy~j{dcoPzDvsdWDiBRwbYq93oM*
zunj$lWlQ&L%tSM!Xf`XFWt|O29VFIpne3$t6wH=PFi%l1TNO+We@u|ZoC_JNPQ=V2
z7b=FQW?~pt3?;>oW1mT2z}b;%nexk2F$rW;fs`|Wj46<D1=8t|<+^DI(kM+%Cp4p~
zLJ2aVOemB|g_2`HOwfc-<YY`LlA0o!&O}mIBn?H9vw9_v!2h&TLb~ggBJOM^gvaS3
zzEdG|I!H$*YKY6)zyyviTf11I%-TybF<i>0My^r}oeom&FYK&WX+xxgZ4^kz%r4Ia
za)ko9Qh_|NL-yg#md5l%6^l)!J_`K|>C2;_p2aGJI#*?a5~mQM&eaO)+zvra*DG`>
zX5w6kTRbkivsT}pkTb4TNYBZH)Z|lv^gM;MvO`GY_1f&T<n)EmQ<{P4H93HuuXtXN
ziRVR%=f#SrW42#yNIE1BSG2!EE#W1ZU|yzRUanv|9nIS-m1(3LnW~Ina&M1GgCUSt
zW&(M&0(p%B>6E+7;FNF**MA)gl(GBzObBmK2yavfozlW^d@MfZjB>P7!kPW%Odwn6
z{U;;#UlquT4o9pxHr_BR;Y<Z95NsY%IB#X)LYKFt!$}D0?{sj2Tlb2f{vMW>BB(EU
zx|_0cCE|0AWmcTQ224ojq_lo23bgSu9?P=RWcJgw(ldA^mFBIq^gxSO+K(hdWk{-x
zN+o7*dpye(f}hE&mU8>qisT$clE>Lok(5S9D-#u%bGTV0JKIT_Xh1<N%LFyF43K3|
zu24{UGEpk1N!W{F*`Tc}HXv6il+~F~HYk)e3Z*Ohh&;qaX4*uGn5@iftHnC)JO#Bj
z6VyfpHK?HS*m^2`Osb=FFTxE9=1G}g#6ML!Ym<V>V`ZseT3LTKE0FUufrvY-0NJ8I
z@)qBgKnzbU8UjUm*;5qH)=WI&=qh+FR6Jcx^2t(Tdk&K?W+D-HZ9!5}BzX&7DrH70
zGqDYzQ5GnCrkn{u{K5som_o=g5vH=){)t+nk>akhO>tB+act*P!7-sYx*8<@@&6RN
z$&})$IXo$TYZob+x<eB*cxAmLXK`)Uk{!IwW9aLy-cQb<ue)zQx3&a3UbOE%kr14_
z__j3j5?;vyle2e`PHy@iV5ezz3fn4WaQ2DW$*Gvw6^a^{vMw$)rm4ZjfE8JDtrywT
z72V~EE@%5I9i2_2T|o!bW47xnSh5VvE89f%EJbydqFS3H3Z>2x*_)dvO^we&x8cNP
zaGH#5sH@uqc&!3_jsnavZ>N)uTVzOe7n{mqKQ`M0be#fvz5?oMlo3p^1S0J)o0+9u
z2^H*xtU>zoMQx(nt>|8&=vH^rpMnlMG?hA|g*0X=)3F)wGR5}tHnF`@vAs&Mbx*ko
zp%?c%>(!Cjxa|EJMfch^(Y;>L{j;LW)2vfx0tcFB#FP!<=wccUx;M57?#&ABUld%P
z#*zvy8`xV|xQw^Awh8R*3hW&UtgAka1GBMwd@J<zitb%)qI-{`d#|GFYT=VkES6^{
zCwCPo4=jmx==&Af2iipTK}GhVy^HLWD6$VLvX8Wh>|=`T<BF`Sp=M}SD#J-4g=C*l
zV4rLg*bNHo(|Z+IjAH+(z&_h1u+J;7FDS6CWL`tD6k>fzG2PfErkfShf9)+yF*^M}
z#q{MiF?~fbeRXePdQ5b*-=dhl-X^A771K8rQ&(E0>AT6=B&@r$lTlc=DXiPug!LVT
zb%(-Q)y*<)VC}3_$G6QC%azem?AYA*6xyAwp}EjbR;&%4MiKWi{#?$VEBNzF{#?bM
ztNC*+f1b;q=ke$H{COdNUd*4D@aJXxc?Ex7#h=&k=XLmrY&CzMsC)~SiEi0EezLS3
z{yf_%6BDvh6vq(kP&O|m4Wx}P_y(3qJs|j2Xiw9yShw@uxA7A>%XZh}arZm8+dO1G
zvPxP@`FuadWN-(+{4RfJe)W&*eUCTq<kw+jN*jIhTAXp#+??2`fnB6ft0Q+MM6H<u
z@>ZIA4TO#?%@jsUQ-zUAff@)SHcCqo$#Hy^1FV6Nlp9KLKdV4KYxFGA|9Y@BR}u;o
zYa^7v8wC?tQi3-QXjS@J!M@htV`5$6b-?H=aNHo8<%E1^QJXjV9JG0YX&&1!FC^4F
zWP9ILC|AZxvlBCs2mL*b?bT@vLi5OkLW$LM!8TZkDs@KU<jVz>?iwblru!H&DMD68
zsv4@#R{9G)JqD~2rDyXL8p5C|eTB6kw1ii=5MeCo_Yi!a-IWr!(wy`A1-%SX2H!*A
zArg)?5kaBxRK}_o+i~|^9d|!q1+jCBKgO7j24DLT-}*6rWF(sL(!6z`Y!Ww{qFr=S
zB}svs(G3n9L?3nCUZ_qrW=cr1o;gC``DETgck8D;wjLP)((jCT8di=*XJA0Lg+=)&
z;{CA#fl?j$f09Z5d(z1t+)bV!dcIuA{QhYQo$pD|`R9D=7mm*E$p0^S_pbQeU-9nU
z{J9t9aE{x9w|{2!t-WR>Z0=og=+~J52Y(#Gw*p<R1o{JM*fh3}-r!Gs;XeNS89%nh
zeKzCoh)Lb<9E3r3_U9{7+^-2aHsX>oOozbb27?3emCwC;@b_SFaP008jG`DkFn0H4
zQUX<(u!K&Ir3S~MzzjMjt@pOk$MHL%;lp?(js8taqmwZX3wVDp#<DwtJl{bMj^`?o
z#yDESK1pHsDeT{+z<vZvo=4(#N!L>q<cS@FJY7Map&<Xz0;JEvI-=r$0(^AG09Pu&
zRSNLl6o5-eK#zih!2;H`kU!BeI~bhAn|;YA!EIb3ZpV))$<D?N4*p<g17Jv@Z&v8{
zwE?{&+FsZtxRQb!QE>Z^F~v9u=^C7Zp9pOq!@H;QXAEC^<2Idw+a1W@vHYf^X;f37
z(+YII6rh~_!Rc|JXRyAd{D}*0kLS%Z`I9<4&PqOMF+48S#+}V?I%?cAv~gEx<L=k0
zaRVqm(hJhKWvn!{am)QCn6Yf*GXyI-M%q-Q&r_riXoYlj6lpr);;|yyI)ZPPgqJCr
zmn)hBLo}Vax<<kvyhclYt(M%gS4w^=qov3)_C_uFkg#OcQr?WGQBL+2-iZy<j+n%|
zSYc}k#$UpWGG<c^e2H;^e0kVPHJ>A&l%~j)tjMRaDRxKaIzIm*f3k&`|Kcn1^<gQ5
zkSKUJaio3)?}bceFu0jF|BFAjM4og)l=$$=38Z2-b;8Je-*?!Xi*xjJ<MAw>r71?v
zAF>>pU<X_KFGzcZ-apn_7qwUGNNIF?Wh%A>|I;?={5ju|I)9;c<{bH0VOwhJOm{85
zE5Gh-pEDKTtrah9ud&I~yZ7YR7KFa#ztswJj!?UnQkP0~DUXDYk4z!U!T!LiguH)r
zl?F$IlVA@x*gnc<_T`mKucKlm2mA9;dmJIc9uLsM57fes2@6*y_BEND`0*C3PG)Ct
zSeF2gQh-M*zz4Sg=yTbQz&$(x?gZQ^1vxy#cKtksc%njlXb4fdK6q53iuMX$poK5g
z!XFkdysyi`Pu9YhXyM0)h1=fmh?I}(63XKh%9#qK5JE9TJlho>e3#>~BhhS`R<K+v
zm=~{LrK_Mr)nJVl{zNVOk@3RUc3JoaE&NGZ_@m>6Z|bt}3$*YpT6k}~@Ta)KgP(Rx
z_fSkKY6T0!3Y5bz=|J$yD2Evg#F+alJ{rZ>lo_lNH|8L9*i!+&6F$9N;q;}zxrcTB
zCJyJfy!$);)S__eUBcO+aCR!3{uDTWWS#f&XP2$m8NdF-oA>c2t~3bPKl91u_8UvP
zD_K<+i^a3~ozU)TUdc*yatPKT+K*D^{<Jx<zpN+FbGxila!5&?&(}InN!IxQ)_DMb
z92IQ6FYdD5m+?EP_vKpesV(a5NvZeM?bjRGFL{65@c22zq{LysZ_yK2@W>TupG<=s
zjSF+$$|n~$EnmNJWpp#^c9vuWR55kl7WXl>HGV~Flq;$6(b;kvYaDv=hFFDPi&gm5
z78QP7E0im#@TB|-*9>kN8W>!@a#OUKx5jGyMtZHy1?yI>TfY&M8d$Zm2+yPQPbm(p
zU9)<yxN`9P;`sv`i$m*+8}Z8dXqC|an=c0a1YC|^P6q{3!LmWLwrqo?dF%Z3>oybz
zH?J!Wox5@6z=}=t`wH{du3tW|wzzR%@I3zI{$0Is{pJnf-Sul%M*kQbShq6#`hxWv
z*RCjTTJsdS7yZkW6uq~R<wb9;8`w0oa$|8|#fptA*Z#R-{l+2wIkbM`@^e?N#XD=3
zgKp@WjYGxd>o*T3pROC&;@*(<ML*p<SX{FrQR3!JD~kiGR}KypH{o?@gnU*UT)up5
zabVMv2bULDZrr$jWAWU9!4+%Q46aTTVIHqqvv%ctM6?Vif<{xFovBVV`r!}GEf#Fm
zb<mH#Zf&R0O**RON(Rm&vq$wrZG4<@y1h{`Xfj5fV3PTqU|`!cR<@%1U+y@BTuDVa
z3f{&FPA6Be*%S?1l2-FPQ}d%EHD6~oMHXJ>YL+Y399&FZY0U?Hcy8D+#<(OUT40<(
z7w(&5P*}{8rM%Na-m4;mMEGBd2cA+g7>G*$`1oCkdw*8^?%6E$9Cuf3YJzd0Y4NFf
zD`hm9m|zk<ai+uG%;^Q+-O<K8#6EG<ThI3Y5`$@p9k$nl(y?T3(juw<9IZd^1fb_=
z){k9K8}*|VhEoiNQ(9vPjVj9&Lt5%IFLG0a2GqinmK95a$C&CL)4KXF#lcFgJ|}Qc
zq<S2X9Y@6EmIzLD1H-fO*g;@5!Z6_c8iycw0*6}67%x%Gm$~8R<R^(>zC3<c;()J+
z-<25PE8}-p;YP5UKTaBo4(-R87F^oq(6+7G!nVlJda}0aTvlRQ<>2$L;E4`WblzWX
zDt@I@9Gu6GTPkr;k`8O-C9QmISh@0qLEc;MYIm6AK+Kc)sDiJ3)1#b9*~BN?Vo&5e
z&t^W^9(!^=Z(hJ3x7A<;>uEw0=J6WS(9fENZsEIaHgtw%2$g4fCCu;1{602zw({OT
z;U~iUF67;(y1UU{|AN7Ei@{Xnn{8rxx?;LqF%9$k7^V^48+Eli;+J{162B{iA9Hst
z0k3B9Lcm)M*gFl_aTd`g*yk$Prh?tZ@6*5r4uAtIBV{{Ze*wPsM~d|N1fRru{Ur+e
zhX(ZT4CqO|+a~B&DCk!z=qY}m26`F~BE2r4FXHR3wV<Ot|3|^K?(lsLuzCl;-lV|Z
ztiT$rfX$`?+rii0ng(p$QTrNX5Co`?20OEnIX^aI!ZgxH^O$uw?NQ&gsWevc*|?R~
zpST;v$|SmyWqTVMRu);e$YQuKM;K|IkX}(d2kj-SC)RBu6<mr3k#3W9^0N3{S-zeg
zzbotd<?e2zd)M{b7Mx+Im|+oVA*(bW(V}T4!jX{#NW^B2TWlM}T!&(;s$*94b&s|1
z@o#cPwfi(G)-)g0qK>%WB5-qJm$X?_67p#lGX8A|S*4(6p+!BOoFiy+4j|^2k_7rV
z3l#sh1gi2<^TZZaL?t4@nsf50t%8zpPq%RKZ!5UX!yOtP)wPG%xL!7Ub1L=TmnP)+
z<F;Q~vK^R-uLh)b_&3>(n>Yh*@KV7(EUPM;7WNsK`j)FWy^!nVy2sm^z`b%TdPwZb
ztn-@Sbw!s6UOlP3Ws28=QE($EUZdY-CwY}TLm~vpl_bBCE4%-kmgMzG%u@6Q(4<uw
z)>v{PDfDLg@3=x!5?>AF6&iZ*Nx~v2R1y}MLfg&ly5-#nX&~HoX~8(WB2dCqgC4tZ
z5bobluM9~Zm$GQ;%;tsg#{E-M4E~I$q8DnuL1lwpLc(qe3`^v3hNXouM8}*#EhqOc
z=_}}QR{AExgSj-Zb$+8dK3S`lX>QpD&*%y-M(aHH3V&zBEZoH#+b8&o2-!t@RT`U_
z&#J<ETx^yWBCVIEqs#&4J*}-34(gKGcrA{6z5?px^4bLAJJjl7vnEO-l?kwqvWXjB
ztd^I;>y=GhwB+QI`wA|aop>s{N4W0=+hc!;vn?ASK|ZiEj|QUlFm<?xifFOB+~_AF
zT!di!bGhCJrsLwD^%dxXYdoUk&n~<$EL7vednHB)6OO8(8k`CkI-Q9v!V+XaN)B(Y
zh~*`a5|M?sZoN>N+H#b7P!P&sf?fpZ+4a^jeXBTgNAL*Upm@8t`-e>KBG$;Sy#nP}
z1OcPOp;<(a+G=Km-0x={jOhxOasT#29YGbgRXAbl{cB5;BjwWCjcjeN$<@i<1?YJV
zuY-`V<A@<$#563zRli;x^&1cp!c;O;nQGLuO7j;<BvaM?>CqN3r_4cpk0uy^(KjL|
z`m<iNL58zSoUV+`*gzLvSsYDat%xKCiy%gJoySf1ulMRmtu_(9TDC!UY*b;*jY?1c
zzQT9~MifTYD9`e!gap{qS5}oK8hB^4HaU%>pV`Fq8KD~@#zwmmlXGpYHobm!2E~BZ
z(v%RO7lR<B5f%eT=3+^BY|bABMDImea?Ai{2*V{eXmX|a43-}lm0gbGlw6_L<t-?&
zJYTY)Jf9!6zo5jEd1b(PK^~k~zAu6E!Y<%=>Af7Bl;!qCd2nJGzyywD0Q0D`<u$NC
z^(;3-3x{zk1?YvcW)(1zao9ovRPo%!bZw$Kx~qV2GzF}i#uQ!<ria(Z!3b(2rI9Ma
zepM?xv4i!eu~ZILm=BQ>xYoev!MoF73~S?`unJaht1$%=0&kjcQEy`!I}IqmuxVRq
z8tX}K7(c6TfnQs_!UI7PsJe^#+}CZ@sMKh;iTZ&hUl*;^B9K0!&4%hHv07aLZMD_N
zH$#X3MRqk?M}N7KSu$Sy0R(QoYr4wyX;-0vqk=OObs#oKkAhE)SJ>x}k$BMN<o?I@
z7cQt2Oh8!kN^F$jU>K_}EIJ!+OJ8e~a3HRN(XeK$VA$!-NI?wHumiDv%f-aV<!G)V
zoK<zSIs=$EfQBKj#ac?q5w#Km$c;}m3KirCnD9pXE}};pJ@V<XEHnNG3Nu`E%>W0^
z0%De9?ZYJKrEJKPU3PBND?3Z|GX8dbg5H%o3ru^5Fcr)vwhXKZ2(O3pO+v89%&j06
zV4-rcM2mDpm&p<q6|L3IY&)84#SW{h$aai?(hj$?({Kj!eMk*9rZLU%7&|HxyTBNt
z|Ls{)@@qxQ+tP!rwNx%+#EDfD$DP1o-%|=(ew6a24ujad&V03_a?PlvoB+AXuCOMc
zmA3*XstcFw^wrp&_R~a(!5g?x#8%ss_2f-?>q#t;qO2#9NYPtQV6A#5*7a+xKkvJ!
zVqS;t3EnMtY3gf^33JDJM#er|>IvvZgr}_QB*C?{sc|O_;-2UVl>a?lN^Vk9;(enZ
znOhNw;~nkM;QBlo#L_hi4J2KoM}y{b(-%}ukQz8-y<p{Iok8Q695?|>dXP<2ry!v<
z%u_a*r%oA;Ffn_2_Qv%2i}c68CHo_i^!b_=1Bf>1U?V?vlF{2>r&(~tZkoC8KN$2g
zsOyLFdhzqm$6I1Dkt^v%$z5rBu{TJW9v{goG_-Eq%;Kd`NoMH{jdrHTgBDV)?<*`^
zxE*|!d)uDm{K+KU(!k!32R4@RQou^a%MQUl%+6)2Ta(*!aWLEB5#(QN5BZIG$iw~O
zJC)3m8FS8%e<cNZinhWWlbJTPn>M*o_e({73qNXw{F)%nKSf*dvSr#yV5UxUq-UUj
z!|8G)o_vG`iVfB23d5-uc5bVJyHpzxRv^gGDA7GT$()`4&L(Bd&Tr<?DVB6o#^$#|
zI`sxA#Q%0)p|K2`Qs{TWLff%D3h{3qLF;IzHhDB%J5wOtkq0T3YEvM6w+l#LQAm<g
zGXv81@*u_1ZVIG3yMT17Li#4JG9Z0F57KQcKzQ}-yb?%1=mOGr6q4lI%z*U6JV>#W
zoC4`bT|oMQLXr%e8IXRQ2Pu}MQy~4M3rIgxNRp8=1JY0PAjOh)3Z$QP0qJgqB<VRb
zApJZKQY?d~K>9@&kYWitGa&sk57HmmIvL*g@=74x)deIvFB!gcAIr&r^s78b`!M5)
zK$7I10_pB9AjQ&nW<dIN9;8^NPl0q#7m#AfJu@KvCJ$09|EECuZ5NQ}AY(dC(tKt>
z`duERSSC<`^!qL##nOIeK>9--q+@g}NoG)i^v5nBg;bJEpc#<v&4Uz67Albbv`3I6
zIcNr?`|=<aluD96R3QC%k0439&<sfXFlbNK25>B+s6g76AGH!zuT2+LV_8HqAnlh2
zDVAtdAno4;q*z+f3`qCOgS13DPBM=Qr2BUPDdgUgWi$iQ1M(o9rjR5NsX#iQ3rMll
zqZyD6%!3rmPAZTN>H<<M7ik8ho;*m)luDAWR3IJP1*BL`(hNw4<Uxw1Efq+Ib^$4t
zqBH~21M?uovX}~_2XzBU*SIH<?ir8{%Yzh4W-5>l?*dXd<Ic~4bVMGcSVmKUbYvHh
z!Zl9PmS#}ts60rq%%%eA=q@0IRFbTv8IX?2gEXQuP7<36q+`2)6jDhtm}WqFa2}*s
zl2d{7kS-v_@|k8pdT1V`Sjtm@bX*sZV#!T2AU!M(QY`<eKzeu=kiw4Jku&3t&w~_8
zi7Jpz=mJtWye0i<29*kVkYX891=1tBfE3GwngMBE9;8^ZRDpD27m#9EQ8OStG7nNL
zajHOiR2Pt9Nm4T)Jvt9kERU){n%@PaaCl47)C@?yd5~gxRRz+5J%S|3Qyrv~&5eb5
zkYf2&1(M`fZC@SW-XHcqyv>Y0)*9WIG?wfwy$8W7Wd(wa!f;(G7=!jYW)KeD8fTf%
zP4=U?_h4S36BhguYIVd$<z~6LtDEGF3R2}^_nwY#jQ1KkFQmkw#kRglb3`wPl5quK
z$W__!C^8&r5uR?$r)t8;ibq1JU}N3VEv^E8ml+3AUc^2gKcsEGMAtdlsH~LbQz;jo
z0F}}0(=|F*(k;ZqGHzf;5Pmf<++QLoxcceoVBGL5Fa(T+3j#vu+7b(l5L{@Cf*5dp
zb)Xp{2w;gZ2d%sblm-+?LC=UPJao7n$tJOUDZ|x4NK>^LdPaz^MPH<awjhU+fqsF?
z(6`70hIRO7Fnn(t*j!j#^K}8fp(HYTuvTy-39_Bru&1hk9id``u@IzJ!dS4ayU}!C
z9U}+*4g%5PCG1E$VgN-DpqYK7TDi*1#uyByTlh_K6oDS<get?UJ!<3t(jo<r>Bx>S
zX<FL$;0Rq%4cl0**5UVo<3`wfLi2q(N%I?msa)GR1qThd=NRg2n;l0+uko=G!upQZ
zbYLT2Z(Rlrh*}&gAvd2k4an$2A*I1#4VM+fldevc;d~_X57h&Y1X|`eE*1DaI?4L+
z(u)j%!pe;0jQ1?1?864aNRQM@^<4|N3y%@5vk9Kr)du)TdxW76Om(a}YG(q5ynJh?
zR<DhKU6RUC-xT)h*tKwaD2>}I)DIuf$?8m@T82lE87`=u9=v2{675+E7_2biB|ww%
zQw3&jll+%5CnW`ofw9<b+e%?@k}sCqXA$OA;AiJae;Z~oI|Vmb;ZUVnJgu5uo`(Na
zrHs~aTu3F37x1k~^pkNE!?*B-utR|_OQG`-5nH0V)3cBR0MWpU;=T;W<uZH<fuO@d
z0Ds)=g~`(NG`X2jX;O~HafcMWEM%-q)n>=ZK3OGL^68yaBB10Wc#L5kFsxg+dUeYB
zPoikPhO&#Nj-X$?Hh5z;Mqv1Z@MXu;!=!5X9Q6tEAyTe%admQbvS8et9kUX8vcK!I
zQ=A^iF3BA9_|~lo{8nyZXq!!A?FU3=ugy-B3)@ONxYW233LxV|Rc2?2qsR^-SKvA>
z+z97Bx((+aM#XE8Z}j)LV*_mg2&bzL!WI!C@RHiMASXUC@zwFFVFDl<a{k~p7&@Oy
zRsB827cr8Y-Y*gn(4{w5tk{4-(?E!JQ|#Jx1?IC~Oo!@%O=@3QggW4#R4rpsE=`oc
zi$H<CVR-sXfYCuBp+6-$13eS{uU!wH%q{ubXiIkDUy8#_f1dakOLi~f-{MdrP2G||
zC9lvh_MqgomqH&C7TWffd@O^#YiG)q{Hb}6VoB}=(qp@T6ia8H0qL|nNat!y6v=om
zkRI0sq)<PRwDuX0PS1lhsF2q4N+6xl1*BMR`wU1+^B@f=q|FNH@m)ZQWw_6PbY>o;
zrzoVY3hAsaAjJ~iXFxhT57MwgDk-FMx_~swuf++b%&QDYPsoEbrjW)J(m)rGstPH{
zfwU|S(u6{qR7lIafK*dR(>ah<<Uy({q=rIT*#)GK3C!j|T9pUsarCqlFM*v3X>}Kn
zcJXT&OPAz8IyVo}r7S=oU8a!MbO9-((&agjo|p&e3Wao~LOQPtNYCQeLZz#6Ag#@V
z^gbTr6)Ih=kk)kp=~{*KoE%7ld61ecKp;I&A+7HM((@J43vwWB$b<AEh4f;D^rS8z
zg(l{g<Urb(2kB)B>E#M((;h*3We%jFJV>urNUu>yo4bJYI({v~`}H}H&d-DN28HxS
zg>*p|kU}Q#<{U^{@*r)&d37_q|5YJ9xeG{d<<~-`xA~A#?72_LgB03x->Hyp4bS(t
zBYP+9x&OpkQ|-BLOtI&#FbBKawdn3`gRV7whzf_aWHy<vu{S^qvyVG9@EVJtE8AFi
zywW<fGp3eSpn~p<pwKC2VVP|d=7l*$=b=KJOg1-MAbMYOZO2GvzLqRk{pmHzfhANG
zR*b{wuwNHp_w&pNt!=4Uw!*MbfdC^fGO?j>>e$B8tp-8YQ=OV}&HJ-P?^e4!Z5`3>
zEA;NDmW<)b#w1XaAX+a5<|4zTblhe+<`Y_|S!+QG>P2;xi(xHBsWkM7;lS8x_Mjr3
zkD8TIq=BX?j8J<Y1|2jVb_kqRytq2Es{r#)90+Yqrojz4SV)V#GdYRLb*6$~oVcZ?
zS1}z~LllmMkKie-=58Co9#9nHJx#OR^C(wFj0u<q*NbMWFwMeTV!tq_7qQoX5n-%}
z?uquA8z9y`1!E!_iNmL`>j?l5l<i=%i?$ivwZbFlUTB9`nj3#@v;;P`N@<cdOwV?h
zH@_=#{DTHR`z?)vAj`-^ZFD=#ftTPcZGN}vE}9v~4V;W+cvNR^M6?nyxr-g$*tUd^
zwCXRhTZVlGE2qGboPW}L#3cj7m7uc^HcS)yp#vI7Aspl8ODPx}fp+<Dj$n`>8|5r)
z@i=^pBj@$X>J3Av)_<`$gVWH6mtvxuNGn>iJoj$v>QFlQyV<!!;0IXHr$#4cxjV)~
z)CD9a#8zNuyO%ys8=I};Pek5u%^K#W+0Hslvxa>S_w@{W4Cnu|_ay*w7FGUo!zi8{
zqA1OPWHNLn$AUm4$Rv}HNHRo{Kp-1Botd7Q{xUN?c26gafC}pY+D2ESpdzwftNTYm
z*IgA|FT4*{@xBj4UBv@eMMd!cd#|cq)mQ!XS6}x`09OW`eDfVuzk2oR)vKy^)UdNV
zjhGMBA)H?!+x@nDZkrAI0mgHqJ<vlF=wh^5sZW787f;vtHg?wwdYI6Dw<0?|DP0Wm
zO)Bx`Jm}iSQDVq<X*>9I2bDq|l&Bri3Z{{zy;|qZ<ZYT{Bjpmdrbn<TP4uBT5fX(5
zKg}Mo2@_6GC<z78QDqK!Rj1&Ri35{SP?<y!J5P&eyPE&ZDGTr1=HSbYW0)G93~&`%
z<dG0X=pP*Jid#Hzn319ig7YWxL_^lYxsMZ^x=;-!r<-wFfXGIQDT~`H_35rSJlmDU
zSW`fj_|KedB`+c!&usW4Y$ab5(mWSSLQB5l&BFqHq$~*%x01Jq3O!W{J&mro&>>rB
z7Il`ryPig0vxa>-U72<!Y$XqeAe|wQo-U9|9YE?ANXr9|MnaG{N-Hn32L#e+2ar|?
zq_YB$#zK%d8Ye?qBaq4+Kss9>tqnkmLXbGvDnl9+NIN=!v|b=RGXQBk1nGL&ve_Vz
zCOUw$Ng$mUfHWC`^hSZSSs+z9fOMfi+7f^?6@ql5K-wyhp49;)yYeptAXP(<UMi4^
z0;$#kq#^p6<<f8f(sT&Y%LLMhK&p2DX-puM1CVxxAnl+6c*Knhq+Mx{S~x~e3ZzN^
z5+`*}bAOo;NY55Xmv#Wj^3tUNNStCm4bpQ2(sKpUb2@-@IepE%bVUHtb3+~Xe1Y@=
zf%Lo%AXzHCC;;j55TvUF(u)Ps6|;iW2tfL$5TxA#>16`x`5i#|XZo5u?iB$@F9<=}
zBamJtkY3mUq^kweH33L33PF0UKzf}(y0QaEuNO$y1t47&f^@w=`WJ!p;tn8L_I^_U
z(n~^+-Xf6RDv%l-KzbW}&AjyX0Hl|OApJjqbb~<J-2tR`3ZxqYkX{yo^e%z)Zh`dj
z4j{cpApJ)G(m#hFy;mT;PawUb14uUuqz?ojy)p#pzXZ~U1k#=kAbmt2eKY{+RUt^X
z2&9h-q*r$U>5~HKQvpa<hai1MAbnOKUDE-i&kLk41R%X81nIv8(ia8NYde5+n?U+<
z0MhG1kiIIAz9x{a?Eum@1kyJHkX|2x^lgFk9f5RR2avuekiH*)^o9_m9}1)&38d>g
zfb?U5bcX{;M*-T#8;E9eNrw@`mLey7kd1V54k%B}BBWTV!QnP)l-WIbGeT2%5xV{*
zM5#B@3%n%YE8fugF0EV~f0NBWLSVVBb4#!&g^e?9XVX47w(oG9hZu5n(4g)mcz0kw
zw}yQ`GD}sPaAaBBwRFANwnOYv-pcLmrL8*NjV)BRQ*gstZ!h-RDM%f^ki$>E)F&_J
z^vfuf-stUBdwBdUK;>mmPzurQQ12lmB+gUd&(X~&yQx*(2<?RJK{Dm;M)15D+0t|$
zNPV>(<tYRu9x2wxr@2<Li%+Qbd}|v|pX7?uAmet#8P?XU@qN=+y&Dm_m^@-YHwE4h
zO~vu!W^&+aTy5Pd4aGL3^E2B`NscIZnb6&IlIo#6?dD?ZR05A%5+Y)tOCoi0fG~{I
zC^$%rPLjI;r?n~&GdWiFwn=!Myl`~D#h>(pJwHZ~p>8DRjh}-@34`EWp+WF&83cUA
zgW#uBi#rJVXb|)dV;7o^?+~;SX32cug|h<FmwdX(Cn-ffCR3*iWc3V31C9gMiu18m
zuU=vv5;Dg+Q!%)z+Az4hK!+I?6}Gigz?pN3W_Uh_QS54|VlOb%6R4%qc8+fix-uK-
z-u{L3_WQH-_6I_}{l79f@D=y=oqlhhvoD2f#(Bsz`LPUDreJpObm^`#+>$0syKuNm
zhaqHcpw2qt!?=~Kt5nPIy}`Mtbab%WxgwnA5Z`6*e8@urhz}G*?MdagN2Hk&!0p$<
z-|1S62Sm$CK)0C0BqmFcn8eXU;y6k9+=v*ssrVd3?4)+rqZfp4y#h21OdmMCigQOk
z%VNRl$y(_tDRwUmhjR@m%NP)?)PuU`d1PSH8i$)t{FXyh73wu6G*$Yg=o1M;DB*1B
zQ=uVr&qfGG)*ASVhtMOJi@;2b>-pJGp*F7Pew$Df7y3JYT%JKvc#ZMWq(PTky=p+^
z1W^7jMv(S(E#<>OWog7S%u=|@@u4MqY|#*AFCD9!$YHIzuN=e3e1_s}DlfErt~5O)
zWlz^9OLRgCF)=!W3?X%P+O#(?yG<0K4oqVDq7)_7BI`}V#VH)x8-T;GA*xJ{lt+QD
zRGNYZ8)Qw~Y6jCQu%ld786exWnT9jLtIMp@d}`p}F>XGS47ML`6PL|XEyAoWsFx`Z
z*<lc9sI+X}=u{m~$&-uR;gHpW{Dhb3sLN>~SHG3OoCzj(-wE;26N!76Igb`T`h(9$
zpCmros67*TN*}31OQlhcQ|wl9TR&#`tJCG`TarIT8P4SJL-K+C?d3-EH|A{HzO=ql
zGiC;^hOE9*acNt8d3I@LSEV{^nYWKVA};VGmvQJCHYRGyQ;^;!5i5^P<MZrZh4XJ5
z-p|0wq&TeAePTM835@30$Fis#L8%xnqL%AXyn*=c5WmP^0BO3sgn4ekIU@@;z^Z|*
zR=~(BCx_izB2ZN^WQb*d=7gUY>rR4U!|-rHqgQ^dwPi{XO=QgI$&{#`Rn4DdW6qz2
znDazo&b%<^ANxge@&Wqh=aBNTzEmZpI>(7mi0XPxVy0KqZ`D@Js-jyn_dsxv+#v<q
zvL1InnZFr}q6Z1Tg7YXWx3C2qyGihY0^FiJ-1%uCN>jX{0g?-FC~w=i?$uayR`M7j
z<HN^xBbC9H#fy+6>XegDTawS=5=oPmEIRd6x?W<JOsGl)2anZs3IBy;Otm~TeTp>z
zLn!h(%ya$hhDlE^6UbZQ)#ufp$98W8kW1!Caq}=^aE^kWEn9MgDxcE`HAIb(o-Ns$
z@9XQEpPv_p;1+?;?f<ssw=Bvp$}gt>wr-t&nH^GHU2`(17H~66W-#<no)_!v9F43u
z$ph**8Bo8;HlThR8c?UofI3YE)SvwUwaFb&oIz-o1BylxjVRT9mhD8KKG+A941SeQ
z$#Dc{o^BdX$s>ctpdtlqO(9C`07x5d`Lu^LM3RS(8O%}%2?q2dEP`Xhl?6i#Ov=zl
zZSk=>=i}rE1+K+_I6(%)pR)~!zk~+FN*NHVWI+7I9}qXW141XoB4a*pj$u78G`~oH
z5Ko!8MA-yl_mASQkR73{+nm}kfbfRWwV6t6pG^IyD~4{tKfJP2foOJ)k58b#wPQb-
zCd2Sgd#Ym}FLis;j#H;VrL?3LX+}A>w70hktrl8h0h%t)uHhX&{RPiiXRY5vsah&^
zSbFA^@vhSxXS>nZ{SyUxdr6_Dwr8vnujkHN3m17>Ib<n+Zgfq5?k1KnIoIHlZ%mif
zYpkrA@ir%O(h9{zv|5uck{w6A1Tlp4?<b~#IPjHF&Ac7Pgf+>*^jext$wnEJn`BVl
z*YBx)7}-|sbB4*R2W4-sC`w=-=><>%%ZFJGQ9UNGcu^bn56BYd0cN`+4cjFuJ$mou
z(b-t(u_0C}3M*|FR(hD-dr>Qh1^YRy#BxAnKu0njCo$=V4eMFR_tcQ8QY-5u+(kK1
z_vq|LNi+5b!Q2hEFNBzhbIz6x<nw}ud`MJ-E@|w#Wv1atApV*@2#Ot?(8dJKUjB-h
zPZo$OYs@@yZmEXzoAa?Hw1yTEc10kkUPjGH`wS~*4!}fq<AN2EXo5mgVME8J5Mzch
zGb7o^)1-%(YLvsP0<RuAU*lzvowB4rp1Zbr6zm%=nK7;pS763Zg=$f|i{r`g1Zi5L
z`y?*~3qrr_$sFI&M*GcB)ePSmRF=xXs#kTEWeiL>0MzVWZjV+^DSO*w1^Z>l21?XY
z6R8!Ndtn#hA~niej0Et#W?RkE&Pf)rYu|=ASL^LrKmAa2vM7X4$R;q34hf8k2#hHa
z82kDHW0dC}MAM^P8`aRb>d;CHY->x5)B*j0QV?U>iAaYB-(-Rl5u%LQXJ{skQDf$2
zZGDX<(o&tXyifmKGZ!XXGj_BP5^4sS^NI-6CJs`hk!mr3{c83?gSRLRcsg;WI9OVL
z_LZP)%29NNFlkRVChZL|>GOn1FBc}=-(%7&NxgPHFqG4Wl><VTCE4h5Qiv`Mq038!
zE(dl%mnGThlBftT&DNgNL+!a*+H;Mx=iv6+6OV6tMmBI~hQPgEz+ETcx@HS*c{XqZ
zA#fK_;u${Hc$0v8ga>YOGzUuvxRv+zic-KLTLMhN;Gsq_>C~E?6&DlT(W-Vk5jYZu
zDJnM|uUDHF1yEw3M{uY}E|5=O%yxs)5&uuDNKXVcNi~SkB+F^oU^Cn3n7u!zKF)Xq
znrv#tbmrqdr=!>D=z-cyEBjF>vkI1&?l~d%I_4YNBUNXg<`{ojx7u()@C52+`sp`I
zhNfV?T4-TY9?t2L!+p5{WoX6wMBPpVQwqgEMZt{B9Hu%=8}Lt6*~!zcM($^QLv)4g
zJMpPTfS;x<L1P4%aGbwBBw9XA##Ag?_=->7pK3)*;&J|lP@(oX|BF;S7y4*hXckvd
zwmI8MOrVeN^_eU6m(hUwja{KA)S6ku7R_N3989Mz(YH2_xuhJ3g;3MKKxzQf^C!~u
zZS;|2PZm&S*h}_j^ffsT3`+MVc*qt*aQ<81{9NE{?*Pu7ncxhC;QWe;Ve0%^;0$*F
z=dMg}N&z@ypH^7@SFnt9faUj@u#5(<L^e|A-2!K<eK><2(<JOzmP1hfOaQo-?hz=D
zZGZU26}25yYj`_#A_VSXq6_UqS1el&^T0J4T`MTaJA%3+GB{+23MfE23;%`LNyRll
z0AB=eFV&DvzF5Y#xkM;nBR>j2z(r8db@e8NRFD76#pd<cX|H<hKA76^<Hqi<WFy(L
zLL@s#NOrK0><Eu!n{T6IA+dTKP{uHNXH7*7hh3|fP=O4xJ!alj8Mf71DoigKF~BmI
zr{_y#&hMtUyxmK48!uSD((XWY8{@4S9&M4AX=b;d(qHt`y`wZ!Sd16X>tDNRWnppO
zVq1`UL%*}cd1p!UJ2+O4)pS|>m0-n7GE}^zul<T$TxvrShxm#1!{`yR$7ZFAy2rMs
z-{~&114-Luw1-EdfFgEaGl07Y=YOI_-A@`3jUvxlskgUQ8Xw`N7@krl{<cx*FXv}2
z#$MwXA{)_AnFyw<2{YM-<RzgYd6*2z!(~W5-W!sQx^AjRCN=Qr>?Tg7oI3#bD#?yA
zBugjOKV>7<3qr)|7GfPI#Cn29tj$j&`K5jm=>>8TYHT)RLyxkCkjILQ1SO+Lbi~Je
zvaF)_Qsh;0F?5H1vdArYFCe?<Jy%Z9-}K|jFuE0Q6WyU74@{ygllLr`M0rN<Wd!5D
zW)EB-<EWhe+|<nU7%URFzmx><yN%8(=wyWrx?u&N1=`-^`nWptqHMf<RfxCe3vc%d
zZy)XPc4NJsY-!Bfvk~Pq!cJ9=OvL3}h{bqoUy`l0FAcSJk+gQPwD!q<Yww`eBu#dI
zZ!^=Q(iW_*%+@8Z3U$fTq)SehF3I^_vPCnPCs}E_?$c{-(WB;miq^as-7Z_~S7&Sd
zYeJ1*E{$IyjqmmwKNN1fF|XeHM9>xmuggY&*M|tOMhNf>A;9rI0bZO;fK|{Y5y}$g
zxmpgF%x0Y-ht6$6!n#@+R}(IW7`@-<^oDG7`j-%$)(f4UDRer)r_<JCI`Jr>35s{J
z?zg7DDO=Ou5^DPS()7*J^d7(In^T%j%e}qd>gZ094*#01<!=wQ+#b6Wq~*PS%irI%
zfiv}3U6Q;ni@B%<4TeI=(uyT%Hlt@8`q*rkTjhlgUfjsN9FbPpmpD4B4buw%O&b$)
zp#(IJ7m(WfOt0u^*B+a$^Qe*Y)B&g#OgCnu{kuZ6kA&-X2<`J8?Khu5$02N70e3+x
zX0=9SQ^1K@nA`=EOnS(#IO%N=U0Wode!(^g;Zb4k<Srd}Me0qASjlqmULy!o*R=i0
z)iJ)~q78D)f?X9yHV8>t-<<UI5tAH|7AIElLE}sG@JUNzz5Cv5z5D)9@9vb|-6g%-
z=lAZ-+`AMT)g?DbK(5!?366t|!jYbH&`~JaR-=vH;jp{h6(h*^1;1169lC5$-;y{7
zz^ZAgQso$@+GSL85|AMFs0kX(R+MDNq7fvs0%M&sPLP(ntU${q9K%r?HPBgeoFS@P
zjTZ9wXHGcobm^3(IqnfK9J&G=I94AS8Ar6=m@!KVUw$+jU)~bp%NGh?zDW3Tk;j*f
z4|ffKJ?v0Sln6-@p99#Il*J=0<1EkF=ob1N<LM>Tr%7d)s!bF>wqxd|XAadcV1xvm
zF%j^oY^457h}08jnnQ+{38|O(q`pvG;b7cfgzN8&%HB#rrJb3O$DKoiIpWWwh~O}o
z;!w>oW{gqf3)v|0-yw=zEzEa~P~;?^A|nBctX6T|S5d}0XZDK`XKr{TBXz#qOdV~^
z@Fm}vfqHZxJ`M6Y4Bc0In`|;%g8y_&IZtiY<_$1h)}`5lVt+q*g=?9M{RIvdUBhk@
zg5s0SA4Z*|hP32V;#MzwS<*ruf&-tKPh#c?qWaFMt;7Ae7vk_vR<VC&^iG@{UKQO)
z=ZCw|&(Tf%_h;zuXA(ykn}xUGju*J-tD%ANT}r3I=aYQJM}S|WkDAZeTLIxS_V1fq
zFZPT*0nXP$aK0*V_=@3tqXRf!&jjb2AvoU@IDEx$zSRMoZ)bw@?GT*r&<osIe8q6S
z+X0;K(Km5}h=k7iK?u&RLLI(hI6q8=lN=}E%7}Q!KW0yW@}m%x1D<Or!&eOD_Sr*8
zumbvV2ukl|1`1y>l#A`ulV}C>(@>$-3TPn}&xM|93(af=bc3~76QgJne|>DWCbAvs
zypjF!*ZfHUhSTE+K2@qFSr)B7XXlcB_1GyHp?8!?W5OqK6fw=<E+Ua<^N?W0^S>cV
zo-0;7rwb*2M;|o{ympv20m|<~P|lzZ<l+2uf%1pF0_E-y6l-d=T%i1MuR!@z2+HX~
zm6ZbJ&$EY;V7zlr2+C>#z#X(kpq$yxsJk*^+q*yTsFRVzf7SZNMi9OUem59Ae5WrS
zfror~^hjy@%h)nmzSc@tg7*U_7t)V2ypG~0YqWi3R>)B73&@Hx6kle>#_vIUn#XTK
zpBzH3rMZA^mOi;a`ebzneKL>4p0t%^AV-|^1(s?=bT(DNAxsX<){sYq8nR6qQj~_Q
z^%~N6O&7FkB)p*<Ej2i{L5+qp9;wn3aI`Da9$F7Y3_ui?f_ly=Y-pu9U+tYv6<PV#
zol9{Tt0jsHNp9-A6NYWExnA>Ww*Ob<oo$#zh~1hG4Iu3066ykS&x}>Dfu^@e!FI!y
z30R-19Pvo=gWzw|2pDB|JP?p%0-Xtdq{QXmNAqZLA0Clk&Pb8MBkJ7{cAL{(yji;V
ziP^gNNue&DmM*SK7hm9YF%IP3!TNlXmahqUx^~P}Q6jfEE~Eo6!=O+h*yo|3m2mRZ
z#N}uzOgP=Tg0$|qY^{4rsCCbg);(8Rx6Nx^<C9(NXBqFX+!;A^2KOE%$)Yvsc!dtY
ztW!1k6u~bcXuzp3wCeH-=rw7(WDBQS+EuK=wEa9p&&qLHEb=0q^jIZ#I(EgJMVPwG
zx)f`r5U`Mcx<UY%vKFIn3-cZNLjgB;vEeHBJoeT7G%EeHAX`5z4E57Xq@NnnPh(y`
zZC*+n2<x@uqO%>uE_x7*c9*z?1UVo6j<l!GPP)V85%!EjP$YCAg1VLA^a`Pc<2#5R
zoX3=_a6DC+!qB~4YG5zTsecjqO20V0i}H!-ZF)|Fhum=(yx5%0?d+@d<q((%fo&;2
z9{%q>s9|P>i7LIxk4+8sQ+3Dg0ai;B#WKa6+dJ&BF_}HGO1xCbH6c86N;aN3EyOdg
z5uSOi@XVygGnpjUY+9{HL{(b%^lYu`54G;i(z>@u>uL{B>vl=&&T4L*+T9Bfnm2;(
zAYImM1SKq6*Mw;D40?feUB2QK<`tGEiObe?p+fDlb&y`=LZ595{X1B;Zl-pJm#r6u
zC~23iTZEGTNLO>dY~31yVwbHK36#QKfwC<G#V%Wm0%iMNfie_=VwbJM0;RN9pp1l|
zjM57{l*a^0dG=5eR%=lRid{kP5Gc=UM*t`dSX-@WyNC%YDn6#N3(B+UV=X8%ef1yE
z&bRJ~PQ+#>{&^krB)i4khnw*o_sjP5z!mCGcx&vI#!=eQZmK+0qQszmI%#gDx<~cT
zAy)&;OHo0@SG#ew`Q`gR>|q<B5wBGFk8vW-VTuu>jl#Jgim|f&a-tbk>ysQW*Gb8X
z#U0FBV@bj(ObX=>g;Io|qOHRHk`jm0db}l?wQ2;EwxJ!!II5C@Do9YVpx?p$hmP%w
zI~ZBDgu^ndO0d2xRtYRcMly#%OO?kn@BF!b6!HQVr{mBUCF**-3C^};rmtx&WCLVa
z@?h5zx}kJN6PyYZV1(T<s96Ulo!FLOqW@qmL4BpjF@~m+#;R7WQ2B>A@Hwx4>Ex?a
z?ZKj)j2`GO)r|~MNKJ?MZ0Lm8)ch@3Df27pdj(v@c!4(+Sx+GPGC|{P!;XDLtcj%_
z-eA&|n?Pa+u+On^IOL!{6jT9qXh)$hw`Q7Xr(h<vL^ZSNSPyzV4i(XfC?D%EIa`Jz
zpM&;ZUcjM2OVIzh6DDZ~OjTklx9#GvttHS+*QT2o-zrn|XU?3Y>LpfMyooCEozs!o
z$|c(t3(`;z8!-;`Y|-l7On?;yIG3>^fGLq|-q=nJBqO-`K78P@da6`s)ELwh;iSvs
zl~J8=*>OXnuA$Fqm{zb;MD#wtR306JzXn||FT-((o$}}{T4C(M?d`(&l<1ej`u270
z(P~DI5^uX%yhWcT!Nv0oU$Gn;?8iOQNFggpoJF6bUrI^QU*C#5W4A<BH)O|V{<??0
zrq%S;Jw4G^@Ltbg#B=<uv}^xW+?d{T`5yCx9df@&Prea*5`7zgaxsI0_WkI4bocxC
zrS6s(Uu<Rp=8xm~Rc3NgTQX-t$X{7kEx!Ygf!*pI?3qm18z;6D_QuDWqt39fQM5tb
ze0KUXe6Ek<TGr?}A9IcK!!_!2^>}@EW~lF1^bYs!ujz{GTb#K*v*j1B*YPxN)y|k3
zF;{v*xKc8Tjo0M(5x1(zcM|5x`ofi&{WY>Fja6!nF`FyA6%+*1c&@-yVbp6Gz7uxT
zC3}9nTyowkIogj5VLpmVl@=A|bzQjU@ydnw*}`R<ew_y*ez{|=jgc5Vx<i1T{~rTf
z)B<2kAhZMSNloC6!JT4AUT7M2)7Lzj*z1lVF0l}mq<l(K6#-$S3-{cfxbTzw!t)&#
z?s*w;;ZL=N>)vli;zp-*2*vZRVJN3sD2j+@n8K}xOJ+`APTz7bu?GuRu+*<$rKzCJ
zVdQy-aN(!>g|F?faL=WJ3qQjzd_#wYdrk{n_?dp;7n;JY??7g{d%gi&!E#%HNSI%f
zg!xTe!VE@t(cRzDuOVQd`CiTJ(TIW6W>0zM@(gDn0nYEK&Oi8Y?xwqcq+gXdoN9+~
zJo|2jvnm13Jyhpk=+}&{*yu_3(#`wm*QMSQ2K!ff^4vJsD>?-08T~TY)fTKlwD0Mt
zazFgyxuU3s?n^qX)3bi%I-k*8=Ydq`LG;T|LD&1r4(s)7Ou62*E$Zz`sQ23J^_HSH
z;N`2LSCh{0o@d~ftCa3m>iBzKhWjkO_=@l}QnaV#Q2y8Jjfp)*Wxp9^XtI2Oic^Yv
zO!vKYgQ@02^afY6sI}hYR?}A~)W$2*wL&4LIYl3#f|NblJAwMj=r+_5TWWkoD*c$L
zG`fYZ+@;1R=*H_%z}fGMuBCrIXm&?mg?*MH)9}i=CY}CMk@QpaWdcdP1Wnw<5p}Vu
zwkNuGBmTILe*G1{wBUId1(r)+@^ES<E^fW$7!h=|+3RX5gV9F16CH$t?Sf`7>cZcH
z(ZjvFkEL-HJ<_{-HJ3oCooxwiP1{GJz)ti>FOLv6`rL#@H?ZuBH#*17Gow|>j<z7r
zGmvctn#_}umtlrsZ%lxFEJ@F9B5sFtJxM@HR*%dey{sJ!^1K!xUD~!q#eM-OSr#$_
z^s*>0z|9E&7ZU*;3Jyl|iLSl$E6%fn(TQ|3-~1$bfgi0m@Y_@ayjb%L{lW~;+oJ8(
z4#9cR%Ng81wg49`0j*-R{Rbl6N%U(3^*el<xJ44}Fd%JI<;)D=MNMX)7bO6t(I1`a
z13ir}E~Q^_DKHp4jc%S!zY^JFS@V+?>~X0yPNMN<Zk!iwmm9aORpa^rDb@?zxa9;i
zv2iQhCd4+YB8iWi8L1clmXU66h4idA(qzK<8)th!$E?d?FM2Jb8MbI<u@mBjp5>B9
z=A>l1Na47#T=JMLS?2DyfDUm<_BOiXv1waE;zq-wVF{)b#VeIo^LbJ_U9ILrP>OIY
z^(Z)Z7L@r*fgAz#Vbt;0;B{`-xNVow*m<S!arl`r3o8||arp+sy`sHiZA(Li@nOj}
zbp!F}x4{8;YKAVbaXmC97kxwZ;o>$NpKx{pc0=k~;QVxPS?oH?PCpw_i%CFAkCeY3
zHl|zeXtA<<_}R%cID}6j#$kS@Tc4NKPukW?ufFpDh^J4P<{P4Xn*8oQ<KO)(-Tj=o
zt48kTeW9ZpQ_j-G?9hxof7<A$4MlajlR^#o1V)tN?KnWkTk2144g2K6Q;VIjwM|t)
z`+wp#nLIKr*C5Ljr;Vnrv(+0{QSMQ;ASTN@dJbSZP*%K;?6)fB69H09Q3p6G37kGL
zsUX8%u)ChvqPrdggX3K=l*Yjnn*>%Sb+?@=-8R+SZH<FfB8XaPPkcxo-2fvu_NT-a
zduaMooz(c>O^92_nUF^6oSu$)NLPsy6GDLe^yIT{Pk}gJS4%F&8Db5jXR`O3^TfU`
z0}GFEfXYu{K8@2J{6$1J&q4TP$}2FM>U=td6f_Jl*HfzxjnS!Ve?GT7pIed74dinx
zd00-=*<^UZ&{&16B%j%j^bWx;d6M5HyVY{@M=t@J^Kz4~Sa7_M`amzqevIdODR~FI
z$_ul{^TKT1FZT(okA6m9aFNyKZjGAPM|2<)@8Vyf1a(K|PDU_aEe+SC2o73<Ue)Tr
z=XVBKV?pah?@Dua-{Qz;P*LMVh2=zN?d#P`tx!b1Tk2hsrQZ1Ac5Jz=E6=OS^Q|i<
zGONn}ES0C6kl04^X>;>Im#%n$s(3-`itm+H-`8Hn;}sZv)6}*jsl_L$ice}?@jj3X
zs>S=#6)|8VQ;Ru4Cd=QXiceM*pWL$IvB#@g50F}Wvve{Jc9HHlsm3!@jc2s1F*;?n
zY3jjJX{M%b2a$KQZ0Z_S@tTB+^>M|+g~pFC6-U!{kej)c7X&+T-%&WAbMhEorz$CC
z9K*We$4SM9NyWQt#WL3)fv0g<a#YhDo<O1};l7<d47w48uGSv9ldm-Wah#x=NkDfs
zG?6$urqm^dQigu>33Jq|5t_RT_^w%RD9IQtq6(R?I~DZTv<Lkp`krTklj(}Z_NAGj
z#~YIb&=B@oh3=Q_p*vI1^$WV^>;-hcW_0V8?b9B-wE}OQz<XY1c+nMpzo!sz-P89`
z7|U&9+$<O`5RCtn9pekKV2p}XG;!g&aSan%LTl|_P11T}yk3gvaHw(S!m-LkX(1WJ
zm8uINly@v#yKdEZWtaJ<RvlWXB<n(Bl($f(%D$<YJ<)YQ5WRtZy$*WLV01mr0Y|5P
zQ%`kRW6M4s9el+$(0rwLh@%8+?MIMZV44@4XL`ehvgwT}G{IQ&QK3RTiySVLEpk+$
znXR?&$5aDHQJdnXph>9)o)~JnXVAk<Kbk(u-gMuT#~9-$cm_Wy1jn=XVK`6j0M4Dn
z$Le4u1<o-cIKQG|n1$IGh~eZqfa4edrNB8h04HW6#8|pJz~Yz)rC>QOfW@~WVmQaQ
z562h}87K+E_bDMLo_!HRVf&)&JWQr{v5mcG9@Xj%umo`PL*P6UB?iYPN)EWcy;K9v
z{1ly-jZzCjl=AGYm{M$S<xuLO3<Ej*c+t6=BXeho%FXWtB69m-w!Mno#}8!~a6+-c
zuh#jQotN!ZDwIN783s<yHY8684N1?!kcT8&7&=4pZ$87o)3Xt)KSV6g){%+DwvG<5
z9?CF4@{4{vQ9`$x<rv+~BFp5xfIOr3T%kpO(@z2z|Fyr(3<E2&@%G9PZ+oV;%-d{g
z>+tpipJCvvY^_}rYOQDc%dKVGU$?anWf;&iQ)-5RbFy{GGeceC*+6rbunn}^B@ea?
z0~@k6ep9INo_#hqp6#>U#y=P{44j{h02hP^;MtHf0oaDzCBTcd%H@U8{#(c}uq7Lv
zwub2BnZq-k*c{%a(}OF+Kp|Vxi=n1_#`@fJHr98W{-DY*FqEz3rBKVQfsV~Ez+MDy
z%OA=x;GXwKJF?M!B1C)7WrDemT_zmbKa^oW3+6dl2WQDJP|Mc4^-%A69w^+q?1AF;
z?n4;{D5{<tmVm6Q#EeR!SYb<qmRf($83r!P#+T0x@ulZ%#C*xlMh;*8oo5*Mr);Et
zL5S3zj}?=eeXLwk|4n2VxGEb(UJ{~+=l#VLVec=OB7buk242=oorf|E=p?6Pm}x9#
zDdDe<t<aYnD2)^mjg{?rW9bS+){_}1=QcQJiUZ?}+jJ%CE48w^Cvi6@A%mJDb|9u@
z%P-F^&A`F|cAwU+)Y)yvWmZL_RbYQI<ctm_jxfiP38o7F92zK|TOtpXSI|d#pd_4+
zle=E3`Ge<!$Ys9TDf0#($2sOZ@Q;ewd1X`j1U{~J91nF9i)`i4xV&q1X?zONYlg;)
zH44sxpglT%5{2ZUh?K(!H=5g3oFN-w6&?o`xVir2D-cPAqDYn}5k#nr7l$})*<R?H
zS5k{py5_YZGJ5WfOh$HZR9%x8^Y8VcLah}EJ2`Tp?BwVO#?5EIwx(+hlF*b9KmOSW
zLenxJRbSUq&Q#(=j#9~`Il$D6GBjs{Q#%k4I#!uLtNWA1i4wxiB47u_)ZSUH;nRGs
zG(99`PuC|)1s$ps;e_!g4R(8Dk}`Xb=W~-V7)MCgO0}3n$gARTaSEZ148Y-9K^bHM
zU#T=zn<8%rb*GMaNEC3H@>>nVe!p7YUgs!RI`vkZIocSy$q{Mwj^voN!_8BhY!jEw
zQ!PR$WkJ1+tl=rv(N5TgFPk?yRmW2b3RD~#LfBTcpi*7N@siUsRlkE6By2HoV~CGD
zzftBR_8WEi=##`p8!J0XlO~YxNPT>qofaUX>G&}zc@+6A;4w>NZD*@*)g)FAOJI`1
zbcm^>af&RSODRs35kq(zc+<L-!zQ>+O(2qUj661t&-2hg=qQM#8H6D&4iisPq&6@e
zqS1lBr%FQ@e^Lw=QOorx-ar&y%ytMBS1sk1BZBYbf^$X|Y(S!Dgycd<s%mcK<gk0o
zD+aCw&QI2<D3zZV>rO)4tl{B;Mvutn=>6I<3RKmCF|GG+^!{wj`GF8~dfvgzoa`O!
zGUszBjr#ghl_KTmIFK1p-3i>sD@5I`@Qz7lR*s6Mv<C<dBE0ngB39wfhm(rL&VG>a
zs|s3XnY27&yj!?Mc@@QuH7d)ys<Bo(>c%uYUdgk0{juE(&d6<9yl8Pgcgo49A+;SY
z5k`E;qEk<$>m~MR5>+YJqgYLs@Lz@Ua4lb)ri|y1$c2ewZAYH#=ZNX_^fKY(E%EB}
z>d#}lw*tr|^Q5?Wm@&YDm1@tHEjdDk^i>L>hNv;pvn5;eI8vRTpBIPV7J<+0|F-70
zEXps+FQ)&tZk>;`?)GO+2Gs&?#?qV<RG#CPgV!jWd}7+{NIcb}Te1zPPlN`P=gH0k
ziapug0kz2;P#kD)mII1L5{pFcKFf9@P#^4rN(R3QJLp91WS)jA;8dLwd6J-@pX8B2
za!8Q^o`Dc0^7@vv;g(N(NJAuf_?W>gm5^XSKO#UOqm2oM7?_lykJ{p6)sB+v>!8%h
z=LiL^#ek5=2+<d^4T%2^4G1rC0uKm|oZt?Ko7@4ReG*B}^yV1W14HwR^at@&4N3cW
zo`(X66(;HDD05;8(Md^zz#C3&xPv#8uB{l3)<pD@u2}J2F2|`LRH)3I<Kq+PZ_W}h
zgo&HIE^A|z`Zxvc=Oq?<LQ>hno1{2TB3YF{bqZ8UOKOoyNXztyP_redfdy!~tO=5T
zY(7`fD|ELGGzE>WP;grMf}3{M`c34fS*cf{4HZYQ4sh#6h+X5Y%qp`~N(wc#J!6e{
zJ$K&Pwd<Jd4!K&E-{y`1^$ZOaPuKM4Zej`R@+!=s<C4z#SX-gEh*oRTMe&p^YGp!%
z^zZlfDiXxbxxSfgP<}f!D7}~<JSaIPh&w1>L2DAD_Bq33)`Jpf24)Qa_K{wglOf9N
z-pjK+v*zC2sCsa+-I0dv5|thmd?C6$8!O!rVkIwv3$qeOaB*2_!F~=au^bQ?(2<OX
zzj>mbC&F$$3;CYfd`>7t+Euau*6iYBFR2zM;#-_sHc*rU6u}W5YS1N{L@Q4M@z-)H
zt8AWvlp6_38xu5p`BIAYRHDinGmo5Gs$nDcd<4Z_LyC`G5vcG5&Qy~28CK2^_z3r1
z92yhKKoW6k8*J!I!&6@cY-WZrGb4pmrb!QJx|ILET5xleXgg&|fjoC@^C;LiTry)E
z3A|}di-;Lis1~)m>;!2PHa#ikrC>qmm#q_iVYkgp;&G^IhVKk2&vL-3S9NyIHZT#C
zqNd^?(BSXU>M7P@*i=JLBfvIso0>?i(A*2V2p6eQ-eM$x?={;Z2O077(wq^6TZmoz
z;_SK=y`=Wd;`5G;lob6ko51*WNMLx8lvrSJBqdj1jPl&0(>SSy##M(_T3}mSVx$h}
z4`_!_#e5OTB)ZN6yI+>1!{8(by$ZmsNF%DuoUE;{(L~CS`K~xMH9k`q!Z`yH_(5(W
ze!gfnYHJI*?zWGY3eNU1vMy@<SzguqvqsDG?|V!t5wD_qvN7qsAtv=AVlk6)L@b9%
zV+S+Md968_+Rg`ta!P5m@wK7@D7;K~Gy5QVEzMuoi@(Km;rLr^(q&0@y2SfhYcGdp
zYtN%X?eQXbaeFv|S9|S=`x-tb8@R`Xz<EKy7#s%yn+>?bvw=G@1kRc%avU)R$8p3Q
zaGRq!SVACML~pMs1uU{n{s>yj)`~mz`W)v2ly60yp;XOb2R$#)VTwvl+hS1wB?j6F
zFOIoDK7BFU4K&dkE^P6A^BFdMl%VS+N%Q-0MS3ErNvc7NCRt9yj%{WeZBX%Z>f?+@
zpvfl#OlJ<W<8<^oq*EF1<S3L`1xrl#oRE7>pE{_sPotDC>sA{s-Bw-fH~agTWT61l
z20Ju5QeKU&(1yi?97{S)8}Lt6`B2qbOWe=;hUf~}cNkMGEnY{Ht2xep^ErNIspz<n
zXyE{U?559GELu2}pCeitH+9WH(7wb|RSL_;3<Fg^E_alD40bFiZeRQ-tDe=0a;z;J
zLM3zDyV$PJ{N}M2CD8T65M4PmAk&qvn64Zo(4p&Q1gL(PBx{y%eGs%>gug`6nfIA|
zv|_TOYQP5h(<9Wz`ZLtFZ-borEfe2ytUfX_Ug9?;03gZOH5~*2Y+8ZSz*YhOR46(s
zm;k2+0dyE(PI~`jLp>E65nBhv>7g->P{llfMJZ;u%7h7psFdOur<|OPQ%(tS3P&Sk
zPT?!&6pl;ia>@eYlueyR8cgILCHZ8~^*a_})Fr*WYD1!?_y7W3UTqn6(BEuvBMx#m
z&noKRT=T#aOam#W7DtMtq(E%a66@&xgpN)yYFiQNXfOC7cQgk-R2`jYIkhTOsI{Ep
z@P}L|hd*?<C(GjLDONFnJr9N>>j<!TGEoV2Xg(WUDH=?mOM+qE`cM;W93766$W3@A
zeU!BcS8_@$KO9N|lygH+ywHgZWy4;9vM~h33#7<UHtiKC=Y^nnp%od*`FjP*<`5Jw
z*djx@V6Q;AFaRYMevxr(*(*5yF@VDl%E(Z*&K^oaz@CdjP`ogW428oqW*^C{oAr<Y
zg0Wq24=g69w-dbZzRPf^6_1n1bsY1NTfs3O+ib-`tY|2!X=%8i=9heKv|6c84kM5<
znU$taI^)?IGZ|`(7b21y!yzJ_#x%YwM~^xltFb#Rs?&JPyh$dJJY}$!xUtR~jcOgK
zPGhWo8G8uj7ADec-6lJcM*Y;i&LxsY2oBIkJGL-VR%1U|)xl&#4V5amv7VcvCcZ3u
zm#5UC)$-7?c`$T^q3+NO>|T+$bbF;%VynU!V_uxCFP|OiOD|?7_a(>7bow%NuGg(S
zI^Nn8<X2=v{`?SfFP<hN=XjcX4S6FQ^4%fiUZ_n*&Y?EjM7}Ie^Ppr)>6BMzYuYuT
zrg;%MxoI4s(`nk~1ISuPO9dPPPL@Wofr4E?B#(uO%?_lf9;;N4*A^>7Y}#-mRSI9V
zgvDUOfE`@B7H3`y7B3uxmHkBFlJc^}mtCZ<Bf7qHSD}1qZhW$5s%IA-=cCImr4vb5
zaus%!u@2ice}2BJOP#O|Qi@rs3|RuUV;d1wli*NiaFilgmtbCt7Erol+N#6}PFI=!
z;^R7GrCv=Zw%zfXx`9-qfX7C4`Z8T0|8e4C{0UpkeAg7iT(I~u<xId&@#RGE;vU$S
zM!5yJPrwV~rO8oLd17wy{Jvaw>}GctTH+c`)92-h`UEVm`6IHThqnM{-xl(Mo;t9;
zT0i&aMzC~7TIaDc92jIDP17j3zkDH@+3hD)iq0XAU*F8*dJFW!{uU?^W}D{FN_Cs|
zzTqU-qu^8|HFs9Ft>-blCkktab+j8Q%i*ROXoY*2FJ*)9&#GguoU@SAR)!r9Ahudi
z6Pwy<^b9vQ6;aV@tf{Kv%Ql`-g+#Ne(M;aD0w@YTFU@vBW)HFf>VLhjsXPAP5odBE
z`dZqdd-|aq_pm?|)^eT6UFM|2VdhO5F7&oNgtuKuI_X0>?&AsIl?+ivI}hc!ud;Ad
zBM)5n_nzZ^*7N3nz}plFxlTliMf{_phnwS^KHQvt`5ww?dB9-QhujM)z2%=U10o~S
zQGT|Ek@94TPU5OM&CM5H0~yMJ_4#Tys%pHHBq+i<sTVU#P+c^g&51=SCf3||adNa?
z95sp`nTQp0vGtybVWHF{HJD_w+7r*{vRgImkz_nI0pGmpAe=C3x;0-it!ebMY<Qpl
zrA{kDO~=AW>9sv>X41w11Dn>ZyFk^^I4u5spMI8{(nHAY)oQ-dm6FeJ<DhxN^@)j@
zd3)Rx7qK*3cM!x|8i(``^e<nzcFoFF8q(uENX~PH)s)z!B`oJl7uv(;!6kJ&^!U<9
z7~a8D$T^z^2Q<1P6VR#03~q7Op6avZy4boH8*P~cp+(QCf!paArSUiq)Y|@YH|$Y)
zS(?)Cw4|2$$OUuYyzyeS76j7Re{KJ|<(fXtQ0NiUrX48skq>2Jdl<@rAnUsR%^J+%
z9$m~shNCka^qC9?eJv?tK63#bOdW;AOBm7tgY>;d<c~M~i~%*}m1zk`f92viXx=0N
z$^hrw6gc`>n{f13aX8hCa0t!%7Sz$t+JvLOio+o^Q$gxPH`H5=sSRfh0477p&dBN`
z7s-M1YNNwpAdUSmIOnW2Eyk33)Fu@5Q5?!lWwacmN#no`=d547roY9QG7lL}_AzBX
za{(PZuU45FD;L9%_D3HNwxEuB)FyS*M=qEHX@E~=gJMRvFZ_rT+vXYdQ4^F5L;K*t
zbIv>GER04?8Go*_kC{5z=PLW53#$v^5q4HaG0?+QqG@zs<=Tygi+pJIahuTWhjD1=
zJnU&qXW*rSP=rIYBbw(7^^}1%rEqU4qoq_A*CE7I*(mln6tM36!iE*AR}O4iyRxwA
z+?6YJzjil6Y<`pBp5Bf+(p-iM{cxV}b`@$vW2NCbOsq$$rBb-#9|IAX*a&tQusUyF
z!h76=rJ0+MP=ecl$C2dC0AD9d;^eT6C+OO-u~750-aRpGb~kJCk(Y6ZS`dAI1%5et
zdGtBD8&l11#htNRB5T_4MS6w5em{LpXM|h#NbWVZAGnqB4t^Ey+04{~5vRoYB0c#=
z>`6PBbYtem;oQkekB_WP!%qzM#)+-_Za{C}(aK00S~#5*{fx>`DS6(B%d29v)-`(R
zf0$e5hila4l`;J?x-&z4zoK{8Y>D&xaD6E<)U;67XX25>>%A*nZ!D1yS9(IYk!05s
zugS^e(~_dAo?%*eH{s&SQnXUjLh3_o#ljV6dMs9{F;Ho)@K%%$hWC>3P=(P4EW?{6
zfHlDhv!gEA%cR34|Cd*Cv>(hzo#Zr`3-^-GaN!@eh08eoIuAtrax$8?GZLdmcL>l+
z#lZkS+5(`PS+Ff|Pig{p4DMvjZtf+(V2HO^h}`wjQ<|#yI9TK=WAo6Gue=NdT=*ya
z!t)&#?gjkk!arpT*S+5sDNpGTiWlgfp?t<dQA9k$6dv79{P;0YFf*rn!Q{Du&-oRs
zG!=w1udAIXJx0Ai?Oga5{KD6EShyG6oD2VNzwiwm7VgCl=fc0_7k;5BJi4=Ox?BGj
z4%5vQeA!kY66V)b@^9i2W-z*o?*5j34FQAAy&8=eNNx6%7rC0@d^G{i@2So|_;Bu~
zyMLr#l{lPghj6@5%M9o132^SAI{!kyW^~0S!|T0t^FI1zvb>rn4EC?|<hgOMS9A#0
zi@(cYziGjmdiOmYRqlshPA1>zB^}o31)}9TzujEtfmG)~^vh5|ry_r4hxK~lUb)`y
zwy3u&q26n=*Xye3(R)a7ITNP`SL%f1yOlbAvlzDV7329S;nC{&>X%FQ^)<K*DCKME
z6)xk4woM&~`!*G9EiG@7wrmm{w>RM!jFM9yt65%>ymz2cT`%=tB=z4hC-v{yl&UtU
zp1zV7WYLXOV?s~A>-o6P$~j+AFEyUhT2EeI8kycmVI}G8NHfEGP=w-tIXQ_<SMQZ7
z-bYu=2>+8nfSd8uRjLM~7d+j+)1W)ihw-K(oj*p!X6u<^^ac!-s}xP%P2VNZ<dbxt
zX~I`blYBZ&HWYW_$ZJJ$e4LgoLYmK@BrO^~ONAyaUcVr<-AY$Xo?j%8r<p%*!~3)6
z&u>Wm-<0})nNUBirK4ZtfuocUM!%sq@AB{dmhS#<{BHC;@K*GF{Bkizw^OCrIxogm
zKNZ+N6WG6_YP~f-!~R1YiiUl+f0tqZF@CoN?B8|>`ws&9Zh`%$R<Q4hL(#DR;@@T1
z_r~ug!_F*0W3$Ak2&p?xK8;a`Rp_njai7NtU-3A}wH_xh1RN!?u^TJ^AavY0s1n2X
zs?_VfYUd(86_rGM@)Zw{g9lKtmOi6+u7}tdy~>&97pP3+Gy5ivIJ;f6gTCM*?@Q<d
za;%68B-*Tr-+2~FaPPFUd;t*{?F!ZWXNQJ?CXC+3v2Guim{v4WHXT-(qx?QLy=>k)
z(tWW~fm!M9BE>7VJ66w;ic^kuY{kiYR=VC*s@_$t>m^bvcDr2aO*z!I^~zpz%SNB2
zDnF}r<<FPOUm%tLLu+2Ql_SFUG$L=5hIUL<;JO-P|L8?j7W2R^g<w|#0<FhfB?w+@
z5JV57f#3<e2A;YcIT)P=gAjE`=`qniOSPA&YQK_^^)!@M36xh0lzkJRT+;-_4k=No
zxcOgW9P3`UPkTVDZuKt$Wd8&p&2wi1vss)<tCoXGqR^Y=#i^`tyd9-G3dapp8R1^{
z`h65|NSeSLrpHi|Qu;0deqfux;}m?9>dD6ha9kPP247+|lkWDF{SOG157L#Euph#;
zmareD+uUJ?sOk<22-uHFiMP;Ij8!`km5)%>9vW!nRrP4~>r68>$t|yZUO;`J2~@^)
z0Mp=ch6b^&Z>5#jwLV^tOH-R{;A;j}Ba<{O6l&v@=~|%>7b)MMQW)L`6}%5RA|;{s
zzXEnpQuSM={^;9urF(xj*gX2K6#70>s5bo{0-2RkePvx*Sc+(WrE%uMvC2egVYNPq
z<-$TlS>Ca5?YdRtm0jkaT6IXDd@nTCBnw4Y_f5_0iLRsC-$1`m4~FOUxNID)<Zg>^
z>Zze}*uaf-7QSLv2ENj&d6eL<`$aM+ON+n57H3>2UvZ(g(MP7x1P{}%h6?=}y}*Tj
zovyghZ`eXPlLRz~cYq?vw)R)CiWPm6zEMX)#$GT1&bR2Tv^IX5USK%ip(}>-UAoeY
za075;f%853CJ5*I^i~?2AJ7X7=ZAE~aDGHrZNj;oz6rwlF};-rM{LTXI|R;8I)L-j
zOmKb{g7ZK00#oPbbj8&91zojCojd89Aa#C8Z>3S^SM&nI`88cJoZrw@n{e);Z-Q`s
zOK+vY`Cod0;rx!S7|!qMs!ce5pl^b3?xwfW;QWzZU^suGD~9uDx@r^7J@icw&R^)Q
zG&uLt3k>Hzx?(tgrK>jKJnXgjCJ1LAdMgc%Se8Zm(yI(-Ke}oY&i<L;91wzYAicoU
zIf$+p&cSrmCY(d)n;><%=&dyBJe*!&IFFz!hI1%gwF&2u^i2@Xqv)+PIFF_m7|vtp
zis3w#uG)n2IQk|C=P-IJ4bI{80>e3it{Bddbk!!D$I~}KI7iW2X>guEFEE@Z(iOuw
zny%V}^CbEv2<OT4RvMgR=mmz8qbr7UEM2t;r<=YB!a0uKN`rGey})puLRSpu1iESy
z&OG`i2&aeMN`o_>USK%Abj5H^q^maJ<msCroCWk&8k|0Qf#EEqD~7X(uG)mNn7)a@
zQBHGOb!!y}ixvaXmuVI-I%I;L{0~%iS|dCwFKz@|d8tMu#<h8Bs8G9`$Hre=C>wue
zrwlDi8(U?2N?fkldQDsFSYJpJlUExkEzMB~yJfj~WA#zwF&Z5y4wc~JpBCl&Ols37
zKvLKf7<wk?<jX@%w>xocJH}04K_6vrIwz`gtj1E{420l#hGh(AWe0HXBxZ~8Xp}&m
zRUtUPqGEW+u-zHMIjaLWcV&XJIshkTj>cHlbb#genXo(~fF<$_))>y&?ZYv4YKCbN
zBEhT;LGeu47z&%RW#?fUzK~yb7bY5{TD<|50B(H<oM#Qk;Mf|@0oQ1Bt)T7Q8l}G>
z8ImI-LiUg4zaSmzL(_G{5163Haito(@{45zXoJUE4IXhd<?uGfA<R+4WMx{v*%Z!j
zmQI8!abKKL=VqhS#t@}E(>|sYoAx=B+I$;@h*(iA!NVBde_F6=xB1)%k^@jUL&U*>
zuXwRKBjF32ceS5BzoqC#0b+{ND!{=d5_qng`k;GhZsP^(SK8S1-NwhThDTfEWpaeK
zpVD9S)4iiKQ&@}_&+A{iX=Pz?-(p*kdPBdn#Cc~)^E(Iv5v%F4_$$GRmt?4TNniUF
zySP*;1El{-3f@CKQb4F4jtxM6*!J{0-R1g|8`p7Iw^H63?jjWcfGBsrB?6)FKVI&c
z-ribie58%QL9+`|pmSv&V`!+7i>7M_|Cnt^UKARVo-r#ANj7G6hGe6zLnL`*(s4ta
zk*8uc<=g>i`2a+6e&&Q^M;Vf(6Kf<JvBpBg^2}<PShvubutTiPPb2xIeiG@W#g`vi
zg{A?F1RKOOKIW5U6%CXWc}0)o#n2u4$s)Jpy@2eZ_gpzaf76dA!{}BfFp08^zGl!C
zOrkua_cDU<U$d!NAmgZ<{@m2e^cZrP5to@6$cu=>(S9cJlp+=({*La(y~*|QxjD+l
z+v6eL_N=;@x7n)O;qAtH6DtTzY_qCRD(ayU1;O@y=3<P;TRWMpwNs(idS>R_S~fFx
zTYCpJCuy>yB_GcY<SPa>WH07Y_e978Gc!FZZJoL^TbEoM>Jrb~p1Xw2?cFZfqB+bX
z3r*L3dZ{gX)ZEWEnis99fAs8ZjlVS1c+VAp8_%u)ZsUi-jW>=^d!Gn8o5ypq5#V_t
z0(kBUOaOLQa0&3@WCFmB1~y|;NX)3^a0&6ZhY+Dn@>J_+W+beum2nlzYzV2fO7|O`
zuE<8G=ZEOzd0sG`*z>}r)7E4<@hG7QicOsEx2C@^Thp%$HQn<I;ij`+h}-ndDNU#4
z-rjGubx4U`oUP@JP|H1s6>d2@thg<Ie;4eFRbo)Bgiwx^VlJvdgBdB0bN;%d*^HjC
zdm&G0*om;p3x)R~$3rTKmTUB?&P?a7-gcFd@E$2sDLF~ge#BfT0gdAYB=4l=DWe#z
z=*_k&H4^^P&hE}DveAA|i1wa$4%42!a~#@lJ^^+oAU~33A&w*#vsxok+Z7I_NJUEO
zf=MPl<X4>ZHYBAj5<Q{fO(qteu##m|W?o2L)Al1C=Pskc8hw{!CQSJ$RueTu)lJ#U
zlC-`Ief_GH`+!*ZiPDTxc(eV8Nu4}Fa%*4*{iG$aZhmdHZoW3u%{$3ChWCMY(G_<y
zdp)_`d^2}5GD=P}EzB@OzhQWg!vj%?QyG<NoOuQdI#!F5`g|JiaE1u(6ArpbA<#yO
zNFTXuQQwj{D<DqRRHaJVrU{N}YM(~JHtce#5h$tfsC1xVo!nEFkYI<f&KXNc%UzzJ
z^%E+hHfo@=W<A}c&RH+aai>eCEY)F9m*vnENV}wRm`E}^l@Bk4FW;PvFW(yCOV7cL
z`H~&n9KLLPxN89HVF{Bc8j^SrF=fkGxpCLTbH4OD>pOnOu$^==%?ZO)E$_*CLQc=Z
z{7(j<5bl{nHB^;gB_;y?EgPwC2$9<J>tj;0U!P0r3njKWg$SN3PtWAoZaas8nL>l?
z`I%ao>9IsSiU^LJi^!r;Mz$f!rUQ{k%owA{P1z{&?hr*hk3*&idmOqH83|BiwL0-!
zg%eLRx9pvrf<{K_{AV+DbWGSUk$iXap;3<xsEs)@A95sMy011lIYVm+sn=V|c@lrD
zaJvG~9u)iG?N_MzR;^L6(NL$9zqRCez*@3`oJ!p4RS#DLJuaV<`J~%9ckUJ)cuv!$
z-H)3hj&#W?TvBZoP9lyl=aUIG3-1dJ6weiw2MW8ws)3Sdv+#jXq1I-BePg*$_KkH$
zA9f^yYE<g(%}1LDLzHj`dP@kb=RnI~Z}+ywvw(dhJ03~ugqT?SOg5~a3t_cJ{_Jea
zSlQXOO~R&O{ZcblC1e`gx-jhO+lS!ugV4Yux#igQUtgX=$jnJvTT#??ib$_gV011i
zQyKpKgf4~QzZ@c==Ml^#{0e=fNtnPwUk{b(xd3yS>;mj`OXHPY16b)z^HLMLJaj@%
z@zqskcaC_KD@_meb#1Rt=Z4fFGSAQKSA|vM_%LGp_e>&LAL3V2fLW|rS!*9IA~b$=
zJ}3Tk<jWjQ+(#6UL%muq0;o#$QEN^N_I)By+#aFm8t|v~=vV3Elpg(Ii29yCGgJRZ
z^pT=|q6oVqRH)~+%!RVsveTtmq{5-PlU4u4gu*WfL!4?!0_>L|sGf&2LuC(V2WsP7
zwZ{VYlOk<+)L?|6a-Ayf*-OmxR+?Iu(h763M*AM5Y?wtCYZiSQ_2N_13anVBIe&Na
zN2O|ENV}k`?@HAT^lqNd=CM7RR8zu7q^RI+oV@n#*hL6q-Ew!fZuwKFTRabL?iTjo
zcDtpnx&`a-5~wp=Cex8v$H44nGB-rh0s@y`tJXOah9ToaJx6|xA0xxD{oq8EVL%-~
zucg^>dQR-zbL_<K^jwy)|F_8fk7}l&p8bnR&|0mQhNj17)Eu^c<*JQ_S&=P)47+Rw
z>(1FSb+WjfMxB`~-iA(($Aym`6Y5*fsh|7yvGh^azD>@tn4o$;E(FDM{bwlb`mdlQ
z&Ra)>3bm>?JO6W`?EIg#ecGhwsVs$9MmtG0uzPL%?DeCx^VMzw>rHS8mafN#+T?`-
z;5I#lKFZpr<UUJ)azY4-HA&{+01Rc`UV+jRg5t#sU?}tV3Y6Xu6fb%JLpia1C@Nk+
z+%h^r9nXj0c%cLs4u=wO_@(iVuCtJXa0=@cB-g_4%4i9DBURcPBuyo^eY`?Oub6ec
zTvXfC$iQ8l(rO!ToL7-aqB^NIKXhsc=;^f2kt^1^Jj2hJnm$`aT_>|Uo>O}vTZEh)
zBL$d`HKyq-AwqiNGlAYGh3M_YOJI6)yab2dsq>|7!INoKRA9x1v^H91nie!^X-l(V
zKRtxquBbW00%PY83l8?iw`HRci(6WA(ei7rri4`W$o81tE?e5$D?4EN7N2jD4z0S{
zUv+2QWJ=z<)%~)9xkz?4Xg#R)d2>R7jokv0VGpUCk9xuws*EGqQl;9wUR;%}^C-qa
zs-kAMZaB;YcOHkCa5^u>cvnU@lA^d9yQR@hxYCnd+IGlb0@j3?g8~JlV)X(cFxFxE
zC_7emX5jz`F&U}1fcwaL)87K#1`89F;^v`zaD1*8*I#(QSx6*nP`zVr&oMJ>ZJT^>
zx&n2AsW!JscyUd8-{{VnSY|u)*(dHW;!v$`>vh0Q8*zjnG<%>kCw0xXng0PX06(=(
z5>J^Z@FuuNpCs}%nWuJN3-WOQF215*BaP)mzcrI1NVy5y(`t}_EPS!_S06B6C8mpt
zo+|Om_=>7+4z+H|DOLLh)bWsTy@d7Sm?)=FX-YQ55=A-ARj1P{T+OL|sHl```t;1z
z92Yz~g9;?s7CoJ=-ifQk5bw+BDVsd3m>q;k<kgDDmsu$gR;5B%EfCfSgk|kep<F8r
z)oat02{VVrahxqU)~4bZ6ddOWjx*cAp+K<O$aaN=;c`5PUGz+Wa&9V=jRIwpKzVvQ
zPz+<haA%}ET4?T^^99M~R3sM)k}ZPdsqG+P9kf7}Tk&4mDgZ7@1+Yy36a~Q3+5y1F
z<VY}FnM^=2Bq)YcQH%(RQ9;obkJ1`PkwO_yVWX5}15_$jMaqIBO2siQI3@%~Qwowa
zi9t9hIcFgeMnzyurNXERjGDkWxt&hI!E~uO0SO!@gB_~%)cQ|BoH}i&CH1$(mAoWn
z!>c%e87hv&=Ce})TuM(_Y+fb++TzLzGQ01pmZwYDhsW7e9EByrb5l`VE-0=L6lb;5
z4bIr8kwseps@D)q4d84VNA!FupXu^~R7C&YZ|L$OLA0_RL{sEiQ-L{B4a`x1?PGzW
zf#I+tj^`@D^Ws!I4SLG`^isjIyd6BFw6js*jg7)~^2$wMnwJTfm#2bxg@AdbfN4*v
zmutoCu+w!w#mPcJ##J}HDiz7qg5(-O(iZ#fD3zuPbXdI|<9LVHITpohQ&C(iC|)lp
z+L9t{sMkETyn#w%X1zYO{x?efZ<6}kqJk3Wm{`EF6H{KqPCJO-l8R#U0a!sZi~g&i
z=x>KbF~>ug)YKF=U4s2Fj=ti2@pdYiDe`a0SZ<}c=$7aPddgVdVX#D(o{4*8cm~y8
zPg&N&IxWpbEHd!Bs600^@NR4zphhpy2P^n*rW)O#f^DU9dXB{_NgQ~u`EMYdQ&a5l
z=Bb=@em*gf?qChViOkVb+4sV3@Oq;+%xcD|Dqi88U>WAA!PcvbU(q|vPaIf-D{KnH
zo6ttjb0=Q$UEzvjAvCzwc^R5HcPX8kvNR^N(WvNilhf&Wmh0OarLj=yUK9%+jaPUj
z&uLDl=i9<%uzvk%V482>mlHm|od~dVn$zipEdUO73U2OmI$dcYa@WskPNx_9h0kS9
zr-m)OgCrl(DWvUWn~P#x?m3w<m+!Vv6cOh-r_;;*3g$AW(<}VK=Q5|$9>4Ip%<1%M
zzwo)v>2!^)KqSnsNy0oZIh|gc0OtYB>2z%ZoCj@Ar|T?O6LfvA=XAQhxz76~r_&o-
z)O&yAbb9a0jMm0iLH)@!UY4Ejn^B041^)pmPwjszIeZ5+ILZnl0mVN=Z*V<L={_CJ
z$*rfaMNspPP)W4XQCewg81s*rTBBR&%3X1Mf^KkZbU%#wGhR>aqTE1N*7br$|EW?1
zeTu$JSm@ld5%*aY<}03bK0=+Ws_lvHr5Em_Uw_3foqyqBNZb@ZI?yXMYn-$ny~4%4
z)$)vpI-5qp44I9jSw#ootLZ(L?=g3~@b_T!aPRJ8kwI)QdZc&vYA%7+x3+|~6tp8z
zU?&sz6X`pq;n8%(jec7~qnlGU<nX?qvY{==^9<zZ$uxP8k!0H1H>Pi#D6sPa``;2^
zA4?^76LCAF>q!FgxVAx_Dj-i2kT<jd*|GB5F91(y8{kR-xJm%NGXdaYBA`RT!Dv3!
z)l0wPJUbYjNH_D%Pm*&YMe7axHj_JU6zH1-`b`<2x5c4bI|Nr0aN7mkyIX*ZmVj0<
z+WrF(?<D#)g0CIEP0r!9!+?xVrf=FBMil`%B|zVk0F*|5bgB>ZG^%eY{i5wj3v@7g
z8r?jdekHQUvgRi(*yB=Z+!^#uTaA03H12X~+<&%eTtAAB^#V6;IaQk2xD{>_BpD2?
zoT7oYkv0VBO9kosS|L3vjx?EYI0TEW;#%cJiA}3|9=ux6Tq9_1wrFOt7xH>3`8p~2
zgL6`{jpXqrDfvUTWSP6)f~Rpw_BOiXv1waE;zp`4M(md9o?}TNKToMdU9ILrN`*WA
z8r<g@nXh<8{+Mlw(Jt;LF8es{GuLXhu!p{FjpfzEo`;`}UkUZ*rCz?`dOu<76%xLa
zHrGF8DmHLFO>ch2zx!Fb`#E!0D_|#O0QtlgAWs#Le8nI?PnD?MbOE^(@5TH1Xk^hD
zjF#a(2`UZc%nVSTlnkZuSbeBgC{N&QuP`!R#K|l7+g;FSl_qs-GqkZ+#c(ycoX-av
zht4~F4HdT<_BUtD+hd{n13JcI+^@&@ICY}H7o7_<jEk>`WR0{<@ipuE&stg7c+S}?
z2McQk2Uc#5Zxvrf#i`W*SNmf6yP2IDkE`N9x;zCt8bsH?;dp&|<Wx<c`-DDUwDd8I
zv<!xVib9H~(QsZg9IJ94b+=}3o!Ut?yKzLVK1Go|tif?XZALRG4!BNtQ6p`2)T)Tx
zG*XqBLF0*A=Fw4Rp)xr>gUI9aRCm&?dC+DV`tb_-JVmw`%)~T~$EIwox6b1@U)3@{
zR?BX6B=iGHA;k^nD;_pKCEDu++K=#Di;&wT+ti6%;JW^O0;{E;&=*|fZMHjQBsQ;>
z=;SCi2BQ5beJ0B>nMs*&r>iDyZr$hVoFtyqi%+K1Ks3l2X{QHk#M-UJL6A<8xJ;p0
z=B$OUiq}l-P-Om{)SPzu>eOu4<jn_By8bg%{b#hU-@2mQE%h%-VJ4@3Ws0v1*&X)Q
zRX<%-{q)vV-}@5N^gp*-HC!1WA7HeN)V<AB>-?N6T35XfRmlB)Z@X2)5Mq3|+5AAa
zdO)>$pk>u#k5~0R?4^1HqzhKdXUc{wsk!H<I?qX{Q>$kWm*(zgs*S#42cD6mtg?R%
z_g($qAo_xb))rOIOva(5>$VHVhfCdGw{^>)Pfqz5ebc}(AiqW5ew%(h2IaXci0{zN
z@6xZsyeHqIo8PBjM|w|wKsSF#zx-9lkLc#@^y_H-jaC4T!JmpCmoh=tJ+{pxn8#58
z%zVew6^p?iQ$?Oaasr;lg<dm_h#K*C7BqcDbcw&Sr0JcLo9-}WenPMeWnzMM-C;VD
zCS-kU?eU$M^{&!s294lT^GoCK4O69|I*cUBJ4;rm=;?mFnQNoN$&^V=ISn`i7$f67
zYU<hiJ4{yRmPQaAzF6H{-nn$ilEn!8e#+w0mh4=zwP$R4daAZ`;lg6|;_}YEN_BK$
zaeHl{eP@1G*ZN`=CbJYYrfY%o(=`pbR+=tY8JAm{TSpG&VqF}wsb5jy&}_<E3ZJHh
zicv)y_ZG6K?;5TaNA^x-T(YUFi%z?FvOsGFJZtHW+oO-r;i==zNT|g2SjTiTeh5-H
zxdbUSY2AQf!a(Ba#*-7d4cpfU6gN#NIe8TRPr@i{Qbrn&_Iqnjd~{`2R@QCS@QdLw
z6x*`DkcJ0Ikj4qg?P@~E?Q5hNG-WNQX}Vo^p>75KwOjsXw5vmg#v>FWGl?<w?zAe+
zo769jG#lS6g*0RSM~xZ(He*&ZSL4vyP^Aj@y$TtYZC8@1aa?kJ-mA<_zewPwNm_#c
zk_2+o%=IhWZS1$GZxJuy%yjo_D&#l(Xt92z<nZP)Ry`0qFgimB7D8If^$eb~F;}aM
zOvAnyzM9je$yx<*RNw(hEhug;!??6;5}4?f3cLQ1y;`YSq^u`!hO101n1V+<OxR%k
znP+EEh%}Umk(VehyL9~sS5c`BjrDEhin~a(t(lxc_!srOugj?i!a)aGtmw$y{QgFZ
z1mRA#Y-Wt$2nupRm~znkhAR`3uutvFtuBv}n=6cH>uB0Uafc4PTBgM$V$V>V#tD!o
z2Tw<Ggw5x;BBZtTzCtL;DVR}~;YpG^f2@Qcat@Z$m*p0h7CaTyU`kM2APVr<m)kyL
zB6}e`;i4`OuQ1iBZ85^8G>>AM4b>;~?C5mf3<BGsqsI_CC-2q_)<7g33Jj~CVrJ7R
zA<VP}WM0>-7oI>9Txwx10%?lF5!^XA<XNAQw<SGnUakaIbQf`_-eG;hd9)ymZ8E6o
zKXwWuF<cu*w-`wrBvlLK-%|OWTJ$@*_j~+`4WU2K-MjItvCm>>52Uehg9-(!v(=~}
zK~B)I5y2e6GqIjn_UMLQqZ|51D!!eD(q0e3pCA*<BrWyXm063Bp5%rnz0Y#wPgJ)@
z1kwbfd+;DegvTtyn}6}|-b>%#hhH9PPF}ANa<D`~iP4Kl-8z~c@nD$-Z=pPlfO<q6
z`Z1hvttQZ0y{Rh`7c`D6*1?XYQE+Pw`k!)xU_S(&85Sf{sO3jYroU1%Ju)#t_UZS?
z#Khazzst*q{cyKg2r~uur$_so>@wFgXQ6uLHX!l{$tL8^Xrk`hHFa;+>~{d_HtZJ~
zVY1+A9Ft5&QKa-BJCF+Z2*)Hk7!P9nbqKxL<=^E4(}(+anVgTn-DYwgO2s^q9%*u#
z$&S}{kK%6@FH-ZMp6t%UWVgOl<v<34Kkl`jI5RC+(52iBm*!Tl8Ym4xgRin0AJcX>
zst9lD0cK0gspS#KS{e?u+-M2u&|%CyOcIxdMNFs8;m6hAA^S7eMB#!Vhjx~SF>@8C
zaCtHI0*Na)P6O2yUo1ewrs<SpZL2Z4(F{`#`;hA?_m%oEh2t(tgEjnoHNWF;ri1#Z
zYqU5qQA9Bk=9-GJA+B}9K?y_>PisD>^n$B?3End~xY6i%<z3|8gaFn`?A82=@U%Qp
zpU7NcoCZwrrrE7hp9W<-!l;{yGEL3U^w0wTzx0S(wr_Ab8=KE$G2coRS3Xu)XXvG?
z9(0cem+L`icCcuZW!;(ks!fG{4@axhxcF|O>|?eX`?3I3Fd9o!w)&E#hf_k6T77oH
zerJO1&kxLYRBT1BmMBk20kp=@5dNHcOQ6Yao+ECE|FUbVWVQ@;VauQ?s(>vz5LO^n
z%9y1!3zCi(yyKN3bay*q|F5mu&(3&CkZsZ4I4y-^l3CcU`n_4NX{{e+Udx}I`QRb?
zFRb2U%Xv8iXgnnuun3Q;-s$I6ahg|j#&ViRaKed!8o|*=06Q*i<m{l4ORkSatP45K
zuPa<cGunhhhL&jMxS<tV_fUZe+&)^jh-S12=MpW^$_Yg)wC*`XhiKg*n$ad4Wwb)8
z4>ejX>0*01r?fp>&F9S!6HY#wA>N8<6=tq-2GW?9%xNk_Ee9_k30%S?!|RI2Iy3cG
zP*%@i#0MpZ;XZ4}YAd>N>vD2T9Y^2+6I-2kPDo{inn!sm5TqY_uBvcsrC@@nsT_lP
z2Q}9-p!Q_vV<QFf-`rMl$-BDFQ92H<7Hmnm9`=Z(ik<INEn3T!FOUR-kFKC6PEPB$
z$4ir=NGpRSu~VnnVLdzJp-)S)x4RnYKqC?9xF=~96W^Pqo@*SVSy0orW&(Ja03v&0
ziym5(H1-3%_B8f0s}i$dp+!#P0N%&KQ-AM{j-49&K>^&;IAq=>th!&O3^dI5bjZ@!
zpK=H6kz*F~j1Hjm*Io1Aq+c$KPEEV^9gUC9azzwd_PD|CV-INe&|i6Vu}(tSV&xn`
z7F4qHw6v>BDc~j`Nb|ee+F`Z_Y3=FW9c+U)EAgJ@@W7wd`D`@n)4F=}Wjbc~8vXhv
z{rV35`ab>o5q@dIG_4WrW}}hDBR7JTtuC!k^NxK3zF}E<0?E?Wr+Jqospjz_A5zcv
z`k~qiy##$(=uqv*c-Q!+X-C?LQ%?D3p<%DvsBxX|Esa}lh-n8G-pk^B#lt+nte@(Y
zaGJSK@@Gx@$T}dqmKO2&x7DdqWGBSOd&N5Q7!Q`C66nWmd7{xg%v8ZUa7X)hc{lAz
zxZBu&@uDRsIUARo3m9m#oIEmbawSe5v);#2hd6V*(Zu`!%}gJi$OPtZxV5pbbb)4x
z%d;}AQdb*?&{k!kba81&pBnhw;yh<=d9vje%@&RQ=^cGI9kYyF_fK|+hm1F3D^<I{
z<ijCd%3bOGLyDU3@xAkD(~x>-M+u7ru;045F+DL1E+n`5+w8Ziv^tIPBa%$*SD+HB
z6LefplM;%G^l4P)9i{XctC1EI!(-Y9cn|3rP^n@g)9xWX*+c4ehMw-ycFjTsG>_8O
zyGH8(wAHTBQC>yNovsi$DZ3DHzir0XdY`BnpWP?&@wvM@By$^W;B&ui#@Bjhs2QK#
z8S?SD`#J66bH8oI*Lp9g37^>uI^64Avx~#P(kzTyeAf)B_3n=Y>Q?kqZ1*SUm(Q7r
znzwy&P4j8Q+dkb!&3b6t2hP{gZR9=POz+HEZleuFK_uBj8$InEeaH55{-!s2j`2os
z^TrG-!^hddD@)G@ev2l03c;GCIkh|FbZobJJ=#WFEJO5bgnpIj*Es#E(63kHmwC4o
zQQVbA|H84#L}_8QJ_+N7g^>O`7Oq{l3Py3}pIUWDFNqhL<J5&R0IPjdGkc=zAX}q1
z(685FJvkU%Pjljv)STVwqk3vgc@8Q~OZAAaIHo*b>A1~NLRy-CrjXQcq;Gx%6w#Y-
zH5k2_{(cL7LH2uLq`A;n&_|}wgy{FL3KhEQ87P2}a?ofl^wqXdH(6YADuSz2$XPE!
zw8;dN0O{%wBrk|GL%OB|NM6L~6iBZLLGpr1Go;sc0Exy7Z>dBaJURu^>q3yc_|gpN
z+72LjL8Mb4y*>oVi!;rTuIm7j7fm_^(i=jMynxdT>G}>JdGV!FApJ`S(l+7&9&sFk
znjyWh14v%5=@dwB3PJMXQ8T1BcL2$YJDmdQEg?u=XljP^)(#+fF{o1@{c8wPRXUEt
zR5PTvbpXi<0S-%@0_p7`NM7V>hV*Y8K=J}rr$G9DAxK^<Yld_~2aqh4IBInYq<4fM
zd4a7N(mOkV<b|zHfplXCk{8*UA>GsgBrkGx3Z!?1AbG*98PdBufaJxmPJ#6AAxN)~
z5yv5}8Pa<?fMi2Rb0F&!NdFOn<i)yXNdGx2NF3EV1=4#%ki5Xx4C#FxK=J}zr$Bmt
z2$B~In<3rY0VFE~IO=r@qz{B3d4aJR(g!<$<ORPrkP_0f{Z|N*7ayA;T?%Unt<Y!D
zz3KQq79S(}FxC1AtQZi2L4(tIsNQr4@=y(wC=U{Hq^V@#1HSWQog(Fn1&w-X8208c
z9IwDCn(V99O>$!@mSNysE5Xbhxr>eall2E8FH)Ec4HYNJ{+is3$aY9=pus~6AoNzd
z-US#hPL9?Q0mPVi8{-U4NTbfK*aE1S!*ZG3rx4x0R6~XVgPEcqdlo=cC91|IyRMAW
zmt-o9@|?6NEk}eHE{zoH<I~Dzi#(~iZO)c%IzI3-$)HxUk<pEy%Mxf)o`io5tmQ%X
zU1O5#5SZaRMR{nLN`1*St}2dm+ib-JNfwi-jZ86+h;63Ilts2uogp0PA6uw&6c=8D
zx%@aKt42S-`(@ant?JzDBsR>z&H}mC1kd&CEEk0>VX#xnjhDAqi`AL=sNH6TX}itp
z0GA7(4j>*uD%J5)t_RhTFA@DgbvVf__&J;CXJJQ8TvZVzOR1~I+N$t|zOP*Va6&XS
zUb?tEJ(Gi9jSqx;5@NYu&}LY5DJaBE)1^_ky*XS(eVgE`pc8!ON*A+h8Kv5!k-~#<
zxU{`KI!Y8lLd9p*%R@+-LD>aizBM_joG;dlVCYXP9}V+F_{1>CG&W|y+iK)bk70}`
z?+g%>+TrGo>8uqFLFb?yPH9FoWoU`oydz?@F}_Qalzf9EnMz&5LJwrIxn!~sM1-zl
zuUhF@bxK~A)5eg>1Crff%FkxDQ*xwLe=!r;K7&V$&9=(DS0l4fO0sE6kkKc3E83s|
z7eYL)e0;cINCe40Vb}?iOdXCn#aeE|Sg~3<YyC!4>%NGTDk@P$mQiWmZCxEow-X&C
zYz3ofXuM9|O*QN`!liT=4#<^B%zEP68=CA`z^d66*|@9tvznJ?ZSxZNY|D3Ta|Di`
zO=nv^w+%BNsFsl0h91HV)=%xiw@VWh*kbT-=ggq&T18Jv7bDNdBy))Splh3#ng#E|
zWtnOxr>p|Fw?0JP*4s46M#`n}VYF{j`2=e3Vv;OmWEq<<;VFiaP!PF|<;F_mQ@IJa
zQDXkDanG<XmXKHRt;y4vqj*-iF~~YhyIn@J&?1k7DAbOk?Hih#%}7xNAyh_@1#nTv
zxG;{;1l3@2x*0phZ2wG-j>YY;o9T+fvt3z?H3eje|76XIn3`<;m89bt%`;&u`J*At
z^V77g%zIFL#hNGQNikc=3F_>}LxtKP!e6A~xzOjdug*5MQpfP|`O=&|I7ctlj8(HW
zyjYeStW3huU+4GWv7cL(n>VJWt`VfYMOd=5MCi9RE6Y*Bmh~4yO&5Ej=qJ+jFVRPe
zo>??`KkbU?-UK+eh2VIZS-2;@+yR_BsWHj|Fa^$6LU4X1+{SrZ7|vHafOA(SIA06E
ziDh$PEMM;c%kMK``9=VXle6<~Dv;rPvwb*&UN0pK-*1JW{8@VG9)a?__J?nLHToT@
zHNF~+62N^g1kOt(!*sgB1J`JDtr#m&Zus(uQsjA-pjn1Ry-Ze6hT-8-s=+|AScbwY
z)6T9kmX9>6a1oQ0Y5gXxaN<8%#dxbTv*vaC53*6}M<Gfbgm}bCv>!}YOexNb<A|co
zw^2&!6>Kqrd2&FjCF(>yr$bCo&<E0SsxxT4vLtoh9YweUOoa%zF*MEI(==y;uF4VU
z5>_O0`01uT=w6!Jc)|LWd3zTd2h&9;sx>^?BDd)S`zifJKixY@Glj)?@x1=En^qPO
z7t0o;-q7zXao$<d{7#?lH1o}6@mGQsFUe3bf*{08rOE;oyN|arEiUWoT5271VZKYH
z^!9QX4s;TytfoI~IE#Mn2-RJ*!$Mm#v><V{PDh79kE$YRV!fX-U@x`S$g1vIX?%q1
zj`N>w6#C2gnTsLUIL6R$DHAMRJNS9FA$ezLNFF9b@^BfFoC(Sql8w4<sz)YGG-u@L
zQR)uBy-Kp949U`o^}B4u`a_6V-9oJ6gjk%D$|2U~r;+?pKZ*1*f{E5hyg{e}+R-$C
zb%>Ht#DwuNpDe5Bz3$fXs<{}tLqA#Mmb@2`UG$zSC+Khb@njg?inodG(2oZuQI^Sj
z7EGc%qxUj`@n2K&jS7H_qksByQ!~?JSXB|1Y2Uk?*KH)vu*s{oVzHFd-MBZoKCZ?5
zF&l6HImFxZg|~Z!w>c%6!`qGZ#%%*kJdSWy)NEnDm_+0OFc)Gm-rB!pYwdlZ)-ICP
zE|%7EjyAWocTgjeCcD44ndwn!>(u=zY<O6WK9F8ZGa>VG)v<E8+gGEvD8bt0NmiP!
zP3}$}x-@*}eu~z-7~M`01}Rcmba1xDcZC|iT=;Q?@FVBZbNO*7+<4;#y!VNq62nA~
z$VPxih6u1m2=ELc0O$I13Gm`%0<3~IiDL;IyJ<OGGTSbO95^&8W0RJIHNqRJ373zH
z?l(F;IvbrH8=}*Cq0=*kPMo&TrPJ1AI`Jr>35qwc?zg5NmaXYWgqnW7G<~x)os%lM
zP2Zf-bXxB1{Z>bJigb8<ww6C3)N-3fr64WmY>#fs-`};NxKqdF)Qe{97Z-C;4H^s`
z3)1|YG@H>gKE~z+4|XD~@<JyK?kOyvE~*1qclBn@p=toov>!1SN<iay0m(b5d9$LY
zGb-9vrAE%%<AfE9a@lC#9in|CT(?7L|5|Ta+I#}6qCtL~^B}~V)riz~g+r-UT`<X{
zhy03@-bRZPC5w+MoN$|7;r_(f+siR#OGXf;u4(&oh$m+>SflTH3wBl9ux(>Z>zk9l
zK4Owj7jbM+Dh=}`dibOzvEJ>;*1Nr--rXs^yGwfaTEBO1=H5l5Uc{Fn(+WMh&|R=i
zFelZtl{a0gv8e>v0JQ3AdWYk~a98N?y!!BI*`mHBaSp(74`oZCqn&v07<G<h1|FX>
z@zJM3z@6ghNWHy`b<Q|JS}weaT1oFhMbt)3rde~GA*x%A7V`LKPB`v#=@cC>84QQ6
z00)lMM@Gg=>ev`JH&*L7(@EjWCE57$q!3@eQ26pi!k5>1eA)PL*8teV4#mvkh7Ti4
zYIBX2ahB)$on;+Xy*TdFY$wzw0W?h2@>Z-f9%6!Eu6pKB4Hc(tz!?((r)DGd(h#X9
zh^&0x_%b2&^**UD>@uMoVBl3mJo}j(cIk(9kbyu74d#eHk0OHO?3<6>W6T(%$Qjuv
za%PAkR}1r9BNTa~Pmz%TMOLeW&Q&<*G!oP2ko%W91I^UY#)n@b`R<3S9MDo;Z)x)J
z3-z?x+my5amU3S7Y*zr<^I=u29e68@$s*N({e`sT7Fx23xYcLOToLr4e9jm%EbDfT
zo4ZBz?QdzLs7p4>r@kCaGRcgP?XO~mN#*zBUWk?YnZ<spslvx-m&hC*CzvX%3Jnxa
z8qK1EuXxMiEc!_6<q7BGlvFu&8^lW$&1IhLlz9WT9k37a4*a7I@UDz*q@9S}<d}C8
z-{s?aCq1-oVv&#f8<%&jhCK^R=9H~0+oNc!5}3V!HK559tHZfnMT*XxQz83c50mR(
zz5+IQSZgEp1T2Kgd`Q;$m`?4Q)zspYt~n<}MlT69lkp=~hV>+!^qw0k)LM~nDrzp2
zQ&BtXu+3-q(xm7It2T3KU)NHWBg&>&Tb0mWvpy?hnZrRB9Eo78&D4FeO3m{UEsve)
zo#h%n&F4zfLsIs1eG;+Xp{&95at1LEk)Z)+Z}!F{9gU9XbCYbYG7e*EGOL6E(iB2X
z8-T;G7OqT=lt*ziRVq!@rl6UNv9vN7CGAr(9FR?=R*aoNt1jn6V;i%vC}8F3HgVZJ
z)goBe2<m0Z7H1d)8Y(TDH#$|vQ|Qcv;?PiOoC1$ms>|41IsF`IJ26Ojt8+NSM_zVs
z<|EGT?eftliH|l?cyzLBq-kNK4$EdM{1X@R`Z3Q1+CWL)5+fx$>7XmL#(qCZaatY_
z7kEn(Z3Jp_Le^xi+K6_`W7GINpX$TX4Pt2q_N2vO;%RO9$;nH=p+iV-DaCLRNZz7&
zW1M5m&tO!Qa?49FC0=mO$bt<x#$o$Zl(TYj*uCWyLnhbuXHNKevF;?8Fb*U0okFie
zyDw9#j>J89o1;CN%*LEkA?EaQxHEHd4tJM1pF@Y7MkNN-D%=B@M#$*XtQkzdVM8f-
z=ZOu2)iQ+kaX|R^YJ@wV4Bd_G@PmY3RctcLq_F7$e&RuHQNCJ&-Kkm_7#h%y34-T4
zSw6Zk4Ubpy92Dl*?geM$wk%$>IG;P^<kPV5#O0!VZporkPo?W6_Gl7SaVXk&O)Bsy
zWf932r|CQr(?DUOSlf~3`Z>iaJ-tjgc}u+dy!!Ll?yUfF$vnnNAI!sy!HI@?wrt4}
zDkzr<p@yh2(z7L7^EgtSpPv_p;1+?;?f<ssw=Bvp$}gt>wr-t|Ls$DVCxdDMH)CmT
z(<I2T1wE9{t;H&muC{J%A?c&zWcT*sYy;}qp#gO&Wuj&+@ie;P0d<2_MOZ7_yvZF<
z<kU3V0YxK;MIv{fmH8r2AMArl20v2+8ii~)Fi$s)r{s|_jrl{70<o7NO6&kg8*cfu
zhcrZzhmRS|QV9tLhz0BlHw%Usn3SQ9+Tvr?j*>ZAKp&yNC0ip$C&+-fBHMs?erP~Q
z;G$@i42XC71L7ukK<LCG7{+7(@6Iu-2ZrVs=?~&5GnbG8I81&UFve$(0k-TXesOBU
zl!P~wuFX{9l`y@eD<+MGemI$#a!}kkK0bl|*11T?2?=%~$R}2Z_W}IU#6ytE7T#3z
zhci7m1uCT_wMZqTWqPbatA#ZYqFL8j6C`)Oyto+%IZy|h0*9Fj77zRd>zQY*--N$w
zm3kF>H_A=NX^xYO&3O0`0eX8$p{BNHtP!v0&Re^79g{sv`8#%{XsKtHW-ybYKh4%j
zAAzEw;^~_H+)XTD^VSG+sMojvGv4N8PFkV3h*oRTMY5x(mmr3a{{7xwMS|F--|lRK
z^5vmHxlv}RO)@BN@(1NBXiZ|&K4+NBdQjrbK(BO$!9LOpb23Dk-FsPmX4W*0I(AN6
zU#S_NDa$`*Y4m@B?T$2Tm$-`1wj9@FW2M)ISg9ziv|U)~-99TV*w0}lmIERKI+F45
zZ%owlMA)roA>UJ*j{_W~tS1X#%`SY<iMbmm<H(R1H$>jTN4<E+hR`+WlE#BeYU4>D
z{+d2fjU|<&jR~5)d@03xDp6&PnMcko)vytJKD+?d&_cql2*mu|s5xn$Vdc#3)8rv0
zSRsih1*NggR^~CxFlJ`t!%YZE_}XJ)n5w`loEF@iE5i;WDUj!`Z5{>thD&CQ>ytD%
z?9oiBMeQy-L7JB6Nzq!Z>%(YFr*s5vd#7(w-O=uORB9Zmn&CTx%2F9v^{UR!*#;&%
zWk}<Dw0erQ7&g_gU#4|=HIZ7Oxfga3E>fes#Yh0($Emh7XGGx^V%NSnyKY4<seQBf
zyc3-)3gH{G35+*~1V%*!#*_$*_xJ*1l;<9p-_ov)YG_<_Xr%?VwIxRCfc`)!h_tUM
zl1UV-6iws-noP*&7L+mj49%o5YE0&{lQjRJ#x<Hq;s&YzZc6lm9j&RU@tML9{9YYb
zXwpdKKwbHp7o6>5o6FBvg0d+`(HX*|H)UhecZZntdBUWZ3zPn*$E2}?nX8%(X2`Ru
zvGakUoKhO20inyy+351Y5M3HVmzN4%-q!(LmSm?(Tp(%f<zv~}^YKu7u9o&(Bkj4l
z{r1G|)<2mI+^0j}UN7LT6L24#Ex6BS1NZq5IBTZ(CIR;$58UQx4wewZ<!W!QOt>tv
zB|Hq)vSfy+*X*pg=#@p)X!P{FK!+&>Uc|Rp6gZ1xE|5=O%yxrX@&2D!8k`7fl4=m6
zNtV;FuQjucHmLYH^|5PEk#Yo@d@{gv<`Yq;qt{V_2D`$S2o?M?2k2>W6w0iEC8m2$
z$h{5)llDl}*{8LdP#Z1?lSJK2KXYm^A;*$V(+2!gRX$WD;}WX=eqX$bu8@6)G4(iq
zJ+X>8&VMtEisoOwhf*sXFc|f%#-ne9M9cBlm_2m9;yv__d7`CpQ`ZIr@*w5EigAzN
zRAi{fiqJN7gQAJRrgmzEE^y{MG$t2)L-pYzMG8qgRi%QMVW8>@Q?T68VYcW6`q=%+
zx?j|aa;z;JBF2WXU7znXkG&{?uHOsMH7|7KE2irwJi2a1_<49_bL^YxSVSfjBu$bn
zCX&uXK4S9Giph?u0hn0n6F<%7e6(e)V`B!<Ag6vyb!I^uXL7+3x{QX)8f^8mp(e2c
zX%#wAqyB(7#F!`WbM85BpU{41T`~!<E{kn5)JdFP(oqOH=VwlO|71fwW&cRwL=YW-
z38ZbL<Fru9a*R`cl8sY-7UGmsg;V&7Ipy;{rz{{&*@TT<-t$JN9kyJ?l#IiUY7tJ9
z#mU;PQWZgTuo>fQyYk6Q^3YQ&Eg5wEvcD_bz#9^^k`d_g>eM(K&COnq+P2_$bWNKr
z3Y62$oWmY_$qY8vV@z)P^SR~u+=_f|AfH>wa%G}EU53s7f}ybrEK73F?C)$*pDF>K
zV4zcYBcdLk=CVmktfPOGt)qVv>gc=Z1wO9iEAHr9y^hxFoK)OTrcTZ0$`gp1R2UgA
zjxL*b`WpIUHRAW35qH-3Qv&w4!IsUNChiGgf6=)HJ6|#OztBgH?Lq)M8_!)u<-~)_
zBuLkLLvZe+7nqcM#c;luOv>cN_LZD%HCl$VFxR(bUrG?^*&!v6Ykztz&E@g{dV%rr
z72`XQ?lqIE6+R}{LG(?WTnTUv4#BxqLhAAr!#SjVIDWvH1k28@5R@M@3}yI=q5RbL
zT%!5lp`k*p`QT5ecrNrdTWFR@Oso@}V6|fyxnWLXMHxfQ>=mn&U@_@%IDzw6Cc!-Q
z$WR+>G^x9!4Ueaf+G82Nj}o996@nt^xT4<)lqc*JC{GMQku2@e-2&z4y#nP)At;hs
zEBdoQdGcO?a!d${WYLQ56)3sA0_E5M6rC_Dde}?Nxa*!hj)YiI#|3cM4C&E+0_7{7
z2x$DZYx)1FH0O?1jLWQwajO}JMPLt!Jqnj%qVVazo3@{0LU0~+k0%5ZdU1ZJ7p(!y
z!=)F$?)Bnk)`iVkxrk;~8;fdp3dBhA$7Six+U&+U2VzzG9=^v&>>*TanDkSaD?1Aq
z1zD4ZOB_&S(?^kZDP|W}I;hW5XLG_nR7Z8jwXZ4Hxc5!0-{+*;Kh3mJlU|nI&r@vS
z0=!zaDXsA)MUHu#Jt8hE*?uELsO$EJC+f7&Mi7<gjV8f79R^MyE^47H+O2<1LFXOY
zD>(9i)46<aQ_Q>S+-cd^@@XNqJdT)=)w$z^Er0B>W$FoqQa2mB=$Mt2cTMg%5kWcH
zQ7wRYAW>;5MoBmyOu~Sf57<c~!BMqgmmtw%V|86KQS|QONB!MJTIO$)gYs%1eR=CK
zr}rCSNx&mcmYPkY1$!YXzC6{81!|hJ7L!o6E@&*;YV>HFn~L->m#ReMC~c63za`SF
z>TD7<v7L`OknxnhK6*HDfY~%mP~y)CjfeI00*?p2VtxDYe%w>KYN8UqDO9Lc;?JX3
zxzPV<XOLaRYr4j%%otj8AFjeWK%GXIJBCrx)%qe~@Mg3GU3w_gstf4_Ce^UCszg_F
zuGWu)po|KXF@aK^J(PrbJ_<qEAyCEz$}haW-uz)QR5iP!HJF{i2&HH6oQ(*_Ju(f+
zi7jYlojITEEHSlgFK#a*mOvTr@h)PmK1DwMtQk0S5{cQV1yf?QLEArU--t44yd|j)
zs(K#kyuIDGk*&XFOKc6v<D7a+1cg&4f2m^<G?_GQwDYACzMAYKXud#P9rnCY8<LZ7
z1L@1HE{~FVDdiPFV<(C`ba)c%0$0o1v3CZ0-U$#a2d6aBUGgg9Jni8kOQQEBqAyI9
ziaW}sYM-*Ub#UiSUzS^3TJTiRs5%Pzn26bZImcRcQQsn?5`nUXsr1IiXrIzvngWLE
z6ZLVN*x=aF_AmJuffvRq<HIqNuf=_f$PdXrolkm_xBHyfd+UXFW~{3n(NLKd6^A3_
z<RD8fc1osF`DacvujPtq;0{8^CYrBeGry-pF^TOxv|3H?(OoovIC$1B8dRR}N*Jay
zp<!x6v^-me>2JMZ+PJ{PXVD>K)tDq#CEH~fl7?vV@0?JthwAWYhm)uRTV$1LqL*qK
z{ZVLiMK&5eKSZM!&<iZ!Unn&ClTV{U*LqkGQLsbe>Pe`8O4ZX&?{rgXbnp!Fr~Elu
z*p2YCD9bwjtJXI*QczwXRW(WWtzI?2oAlLnbS7*RO~6Vba)SOJQ#D`y|Ji#JIJt^y
zf1K521VvU+z$OTpgh^*%O<1!K2qcq00z_iy^i226^i9w7&^?n7c2JgJ69mH|`wj|%
z`>sz#5fp)ks3<%T1fM^iA_^ji%l~{&Rh_!`_U&7@?{xh6et!l&H`Vvn@;#?cZKqD1
zO2*?$T|EA{@OZuO_@7}OpGqzi`^d3yzPhxuYR0r_Q^{xWep9DUTQzNY=fKFwaBarq
z$@%KZrB&HVwSRKHr#9KfnUGmbm*AkqlkZEBP!nj2q@E=e9Oh)Yit%;Qwlkofm~10S
z{2Oa#AKd~r$2^UDi5rvE&dsjcv5WRwq;}q7XM^e()}8)SCo+R>Y0(}O&4#r@c-c-D
zij?|r(w>%7HQ$yP4P0n%JTo+hP4xb!Z??!|SW6}1`U&cdaIpm)eHo_U$+U=tP}y^v
zG>FnhI#;;N5?C&{I2ydI;$VsC6Zmvhrn8hSW@%WUx0Ewo!!bckYv^|{yZ-#(V4kxJ
zZ1N&I>-9=Ag{Dn(ZpI$cXk-mfq-e8`FTN<z!F7QhT)IXVNO4Moql1YH9trZNVenR3
z5?=HN8Ko_qv9@j`$MDBA=Qeey-$~bQeKMA=GuD3Enzw(J05kCI6QijK^wnQV)*F1q
z)f-q{$ycQ}*ecW;B+!*ml<&}5RhuAwJpx?BYV4aXaCVdBTLSJs5Nfs>i}nGMV3Xx1
zE+}@B<uQTs(+vXUXD%pqljY|E<?#&y<rgj}c9Z3o0_BMf0_8~;6uZgtUjpT+4Fct9
z2Nbi(@+-md%-C^6tOuTT;P7p-{92%F9ctTeZnDI>5E{cK%WuK+;Hpk5Zs%fhHHO(Q
zjf_!B4>cV7v}ZiZFTlcJOj=G@dhtvqG~Q?Ek1o6jsqETaXynAtC<VE&H?_y!-R$ow
z=>O^hkG01~LkwfI$C>Ntdz^91u86=ZW)88UcMsc(L~4)saAj(>$Kw%|Gu^>vx<O4q
zCZoh)!;HF6Y^sZfD>f%<BxcMsp<yqCn$1qa2VpMoU^Nt;okaBXd%K|6HQhb}WycUZ
zm$D<JMd97i4Y{Uk*dH-^$2KjZ=vX@SLDUIth}Df*ho}+~YzK2^)UYUQajiprdjkE%
zWt(h7n}77@8@w8&iwCa{rCaxK4e7N@kh^B7LXK|s4lGFyf#`+WRfI6lbP4k;MBsKc
zTZDO+kT54$l6Z$9wQp5&1b&C`MXXAWM6|dL(XL945`6O!YS!*kuS!<9px9MOUZC`B
z5GcJaD0Wp+5GchB0;SId#jZ;F1<Jq%fl_iov8xhKpsd^=P|6M{W>qpMIEKcKBcgd%
z95{Tdl3{_82{rFGS0%oVy|Mck!>VKyJP&n(5w&=^s}`-^=nARD-9oh(=%kxa@o!;c
z^T|f0=%w2CBW-M!(4lQ>Lh`1a4NZ|IPLy4O|E+d5;~)ikrKGCAUW6Xr*0gpNTr0up
z23@1AJcN*f-+1EUxt3TSl3xzezbBZ^`5JrUEbfc6Mp(Yy+mj9j&_7p`Kz19HUY1@4
z&xYVM;r-t06l>t#`tI8E>2PVdNJa_8EVcxAe?C_juH^FNQvXn~x?WATF1eA&zlaL+
zIJAyJ)PU8Trc-6Rj<fYXFNd!=#A&`TixOhNxqN-6_eDhb5*}ZsN6|xG-dFHDkljNl
z;$F|L1;36coZUx|9lAMW_ZY?N?Dlwuv6#;7caN+;d%bzP1PA$##p6FHwf{>8{U(KU
z=<U`A;_dfCZ?8bM-ciI(n}hcRU;{+AuSD-DynUJ;>aC{kA<ED<*`%8K9z|Jex>~hB
zKa}#nU{vu^6K`GMK8hTe`12x&uM-S!X_)v7rRnxgql2U=VLzD?wq0=kN?>0Q0edPC
z`?%NbO+ZyG!z0M;Zf{S#{7~bUCLDjkK=+m-4faSAbKN@mdQsqCm;ingD6beOYX2iK
z4C^53HGxrY0>kSeDg&I{3uO1=@i$8C6M$$5-S{|)iRF-2@bPOsCCw+4RqLK+bL7Rc
zx-J4P%W8iZ?g1!u0FTIuFExIN=C<By@;oGHsn)j!YF=1wBltho4F4cyA7J+o%Bz^S
zs@ZESQtj`he3No>XF;=zpt-aenyM++bbB?R88L6ij>h_|7%ynnN1(Zt<i+2hTk0ey
zBWt@Ncej)0NnB}!K%cEDZZ89_J}c4KMg}rVXqWkzVn3On!6`he!^s5~JfN@Vfr;vn
z5AW@RtkvN?^wt-6kJ5SXz__;`VzB(|Z(w-S5mMuxcL0U+YOp&Hudk+Tjyk}*hW=S_
z&?b(t?-0rIP=un!zFG7kAd*{wFNcNSs&VfKdTslJZpFi+koEk~7mR)0LVO7~%yaPN
z7<^eANYhfKpMdntiUXMyM^#-SFzUTVa}mXQEdf@$a3Zop*HfQ6775^)8rv8Lq`)X+
ztps>~#k|Nra%)JJfH^D;d4V)<ERe{!D?d6?Q5!gEQE_E5s)C?8YAmSWXUcayKmZpD
zOhcP#$}tG0;#j|6J8Uf2#Dho>RSCI=TXIuW+=!TCD+Se@F`=sD*sDSSR11Q2v-EaD
z0%%NZV)G%oJb`BV>!{)A4zC47Cjkeus49pK9}A1vcs5tSnV__oMg`L<!89g8r{<?R
z1+wC%8$QzFS|hkl5nM-%g;yty=7&b~wIcyg-f1Wxv7FyF(^TRa0&Gl`I9#N&8WGI$
z&T0wQIf%`|^<lv^rmmg3D>6Z+dtM4bc42f&z%IQoJ5OMpZ(w<s+X<1Ghw;F)#>~Ui
z=x=7Ir(lF=HbYhOi^g6_7r#!5zgUXDvPto$M;Cv_7>d7KioZgNzseRLT$7(|$~R2m
zdBcZM{MD2ivk<*Xz7CniPXHZ!xk-TCEWoa@zzhPemh%j*K()wZy<vgwDEd=U^qo@l
zbxn$1+uou-Cq;i=ioT&q(e+@_-j|Vr*L@eEMpDY-*9E~h1i?)fK~Nw*W+0gA;JMPJ
z=56$9rPb>Az970!5Z%%Y(PgIW0HVvoZ+Z4~W%w;msCnYu9^^j}<c|sR+nOQ2I*jHT
zdf{CgetR9>UT@yE2j|m*^H+lN4vW)FCdSmBpO-RzBV~NbmLYP*%As38WN?uZt0PF=
z)c+u;UJ_JySyV=d7}V{4P<r|I3Q~=UFRuxL*9F1dJ_MM^d2fuln14z!8y!y>vP$PO
z;bJzyr!i6U7UaNV{9C1%&xMP5dzxa3-u0yYykJzX(7bowL?vy5$98z^fX9w_ybq6^
z@pwNTyW+7M9=qc)9*;?Q?1jexc+AA(a6A^^aV#ErJXYdS!DBTZr{i%p9_P{{uxz!L
z7;b6ku8(OdpUc(Cm62L57o6I>3%Q49f&tydW2DSbHSlhO#M>Gnt-D}(+7>T(ss0tU
zRR4ytX;@2?<%6u&ocAt=B;Ix~J^YL0b!w|1bp0U9*CVr}*iP^7OX$Zx@c1V^^fX~3
zEEMCn3i#qd&hAUdE<#=6*HU$R*e2-q-bxw!l_E-2@{Z73wMFpG(AzO@5$ph};+xUh
z01x$4&~6dzgdCXYuS5_HwQG>*F>eoyH<0c9YHp>Sh8?;;jbVQ+0`{)xKgXeZY|XT!
zVA@+i%AHk-L5A+6Vvyfx0@BahmH^HafO3aWVt{s|h3^PrfZvJ$*kq$+qJMA-3|n1|
zU!uoSZ;^rDd^nAiTf51^H>oi6Zzq7>5^tBM1ZTHi_-+mc_uVGoylFw{?Dlx;Wp8k>
zkIMJSZS>a5N(0i~8IV{2?4}Ff3BW+V7Xg$vUG@(HoeqpM@Ca@ac6+=Xb0EG1CscYL
z?x4mmO~m6gAx?bFCnoMJq~&?R*$6Rl_cbGKCPDJ`5#1gWH>)Xev;8ClX2jO}dt#&^
z@9vEB{$@xI4<e0bTzK!Om7TbdZ+1rWLyIQa%sjHuh8(MN&JVUFXS>71UeP({hiuL=
z(+}@3b$fS^Y>a7h{vIQu?j9^Kt9cNb*oaxp{h0Ew*m8(Tc@3oaCsRK{r1(k9BZ$JO
zAB|uH+nGKDA^s`Fvz<)0_b^^Qg2&^5FRk>PVexBNeh0-vZ+ckU*WVcI_5G4*H}x@%
zEd28{ZoGKEMSgY)8(BKj&3u1BO5dlo()Atj!Rhnw5|sTS;&9o2kg^X-Rkra16~ypI
zHwH57VE``+fN8A(XwbKNe|DGZ%{|gk{ue2DMr#FYpG<Uz)W|@9Z2u-iu*&r;@~<e8
zseZ#0?mcE3%NTmPjcDDda^4srZiYV#<?L;WPbuc?Z<Wm7CYe8OGndZp=iG?t;Tzam
zX~o`FDFJRP0Jjr>zia|9^~87Q25|4EH%Sd7#tFpT1mcqxB3He)dqWY9j!etd1j&4&
zWd2k*^RAScPm#>0O6I?^nd{nbsh;;s3FQERa-cwY)<RKCJlJIJ{TlMvOhp{oKxat-
zv!#IN!UY^=3UCj=N^7BOX`W<$q-6e^aOU$<X1+)=KUy;XZ8-DADKkG_GG8j0zYxy+
z1e3Y<yO#NGr&Bp8;CHqF>0$nWoL>s|Fx}oC@%B%6^ipcfEwt*T&p>LirmXLwl>+C*
z2snR6nXiW7`~`1c!=n;}Q%wnH71Hsrv0C7~6anWAl=(M2*64!GBKhxl`42q8D<=l~
zPkcF3r_sDS2SueAEY3qZ275k2JQDrMf;EUX!EAI>dhl4GHO(JQS>_=8@o<?JOPQ}U
zmiZQx`Bpp(7j(HFPg(92NXO+~DdoP}q}+~(a<55VuA0iPRSL>4U=kBCl|OR5!NVcu
z{Xc-mH<hQUa{N5L8<}v@zgkkH1@d0xY4pK2#&>%>lRiEC_CTaRBmDMYygk#r4J>7o
zHq!<>s$~XqJ)sf60`l})U9&jfSDdqG;i_pfmGO8sfgx3J+|WpvCKiMf1xZ5tjV4H{
zMPtE1erak6r<o1wn!KP%XoS!NP4qIS`FvLcVF6&a6a+{@tFk6Qw3UD3lFNyu9|#!B
zfMDrPHt~00=_{3sDHa1qoNfTEG=RLnp=~Jje`||@K=*JCa*603elD(rImB&z|Jf8*
zcMrdy@483)C8Ean6aR%MoX1AUBVyCS+k`$e_7j_i-}1rF&4X`~shJ$p<A(O|g29}0
z<La0q-A`U%J6n5lrpc<`DT4NZ>+c0e!d@a8N9dGNYrO>b<lG=$5;O^WiD)!wJ5PJU
z_2u>eDOY;)<sA7eUD+B(?D^JL+XI9yuP>9|Y_e8N2t=<G>^{ls28j2Th(^Xv8{QUd
zINcsA&)*(?I~Sco2Qo2}A2Su>jsJJxOIUf%YV|D|zr;39Z%ZU-Y1y^aIh2A$v11*D
zxKnv2atU=ROz&3oA<(J38!@*Izh&&(;8T#QSnNqBAiAFHrqB+8ds`G4g?sxjLdLy=
zc^jArdGA3kE%7h|1kA%-1<Q`ju)NQdYGi=1>=b^xGt%!8etSP=+wV7TgSazD(GqUE
zca;^~yCRn;9`0sJ4dDJj_$}A{?#RT{eGugy_@q&`vQsf#csxU}jz?Y!s{u6uuO`yN
zx3)26lzaM$c8p$H$Yx5Rv!u|j#zM1rH95M_DJ>T|PYOL!3Y}^TmF2)*c(=DH)i7Wm
zyxrHl4f1|Jd`l)Fdib_>CyNc1u?^o%>`snF?pA3Zac0Kaog8nFcuNuTt*^Wj@FLRg
zBy$VqR{;~Bzz)r0b0}rwysiEoI{`gz{`>`YBBBqw?m_vgPBRbYDlqkThT^K3UA_B=
zVwqgo=+nO=LC>L?+6eh0rx{OXbsp|ZGriR!*&dUV^32Rxv+3@1ItAQO8Y2JSC5qTv
z%yz7gpC#6JZH7IYqwBy+BQXBq{(c6ahuS<(Q4Tb6WTcB1n+Jv8^5p7Z^EM!JNqcoA
zZ}n|`(<<NG9(`+lQd4rG6M?k);#QX#!J5lPXT6THre4MGb?I(;6<nX{K13!I^>J<j
zbcGyZD}>h~OBa$!PNT;qmI>n<77l0&$|)>6w%1!F<lDVeE~_h?g&4YRr8Zg~sbxxi
z9V_^zV*O$T9KF*;Fp$3`bmb=<LC>s^Wa>aZE_Gd$ujU7fbO^lGF-iV9cos^xeFXjI
zvTLRJ3+9ps`p}n*x`L-eL#32+Xitxd_b1V>OkDD?TV9jHWiUCoLM1QpMM{V3O1m-v
z5Bt9Bra4UpY50yBPD2F3sV=)G3b>O8+1ZXFUlgRBWK(PDqdOL<(d87MoFyvf+S-r2
zcC<@}1_K4n&6ja~X0=!@63w~(;ZbL{I&KDW2}xzNN)7`eGR;s&VGu)yK^f08FRJ3h
zFT^mPXaDZXP_ctN3({3SLsG*X^XZy_Wr6l_42#k6rICSoxCmi6*PBX~Q8&@TAADSr
zuFPuc(TeP7m{Sus;BlX?K%nbJm1hL5CG7y2g&S2zH$xB)hATq_>KQZ2J(hA0%o$Pr
zlk7U`@gSCRy0jP>Y7bkhE2HHC93)!`)b%^HF3uR2GcYptHBV)e0)kpYElYQ!=1C&S
zn*;7}XoRR7IJXXw1%)tHxjuZbaQiP9Odg+W>ZXTIKBAQ&zwP7?yh0an_g0uY>Ouo6
zh;*&0_6t4}T}?}K0=8M=9=Wq=hZbUu>elKl>wE%|Q_VKy+N2+jm_9sxHj&Ao!(%fK
zf%R;7oLE&Z588HL%?6>YHzWOfY`a1;SlEZlKEU$z8*0qw>CIeq7<m7U+cX8J&ven>
zw5sh~B4P$nf3MCLy-=2QInw>C53D3Hv#9UnkF7iYCh;-nsK4JI-_}GZiA+eCb%!g9
z<b|#c7P17Vxm{ov@nsRBsZH5IeF1gmbo#bISb3^7%Qv!XcO-N0>M(jlELrB_TeH@w
zj~UIz6x2U;O+WEyJPaRh22JUf0cU-9CJD_!eUh#zo+3E!p9qQulq&TdbOw5^P|IrZ
z2u3Z>C8^j9vw`&ux)wMLG2hJE)XZy92lWXHhvBwAUtZ=AtL5wKn_-At@3*DG!Z$km
zXtjrT5bBLZvPX&JJuH&-89dHv2y~}US>3^yQ^7z#6dr%qH<gvOL2ogl^{MaHv_@_3
zr)Q;yLl{GioYg|}jPM$EvP$E~@LL`@jtakJKQ8l;wW$<ouh;_h7T{CLCa{Z9Zq+9@
zC$nH~PbwA~NbW}AQ6w)4lffh(9e&Fs9}|Adx}C-0x2)S)LT~FE?;VgV=FqY9p$UhM
z!>5!Snwo$^yg-olt*5}WFUB3(@hCjRp=Ho_evm!{diWC%bD4P?sO09WL`NTYKmzy}
zkZr|=;D5-%H^O@&UgZo<HH;e-(h7V_&T6e<kV8cuM3X-akG0(R@wkN^fwMlcIS|t}
zxf|6xalFoLd227wk5rTFzDCTd<_i0p&x6$bYJc6Ii3hRpz+y-5&^*e_v?hjg=)tN<
z@|#7!7HRin8+PtBMfKLR-<*2UiayH|J6n0NQ?IFtb2HpiuRFtyLhOZOv?!Pw_}fH6
zZwXCD248-$3k~mXv}6AaOlwbBFXmV&Nxy3DnCIp^)|FC2uIUKqmG-@RQ!W)NHFX(q
zu0(zk7=6Nx)r48L2>st<_G?TCcz&n-6&)0m`)0cz!I|1v2k833aK9Pud7D;4vvFUI
z_5lhHQ7!w>G!%|?3?Bo9&87ws6wWfee!woTX&r_~{_mlS_yHb240mm^#}d=E`L<2C
zg^jgq^X2DjUiD2nC#+An4O8Egb|rLse`~&{H_JbLn=$R*?clZ)K2ZELu`_OC=PFte
z2Kwf8>Ok>*J@g)Zv&SLk>fYg|66xkNti}lJKs|tL1I9agEO=1L`!PaXUQ&a(z|AXq
zsy0OW_gtmgX#C_ojQp{e<~@SY-4qILOlU-p3ZkC~qJ;jcf{L7Mxl+w*pGOS`GrXS)
zoS((Qd0gQ9Lf{;drb0A>8qcUaAxNH#Me>v&d0LPpJZv9gjB;v}ql@jURoo94Fh%j6
z5j@Yv;(1Q+JTG_-PQ{r#T{W9)WXyjHkl)4vc|m~uj{rF|6%frB2yu`nzZWPk#zJ{X
zp!`vw%uWSGoSNkt(Y!2ZUWr9h5hLhV1x>md>dp5K6mx}A(8ZefngDq{7RVd;%Dvj(
z1W3AG2JZU$@-^CJ!$q16PTyF&rh1uw#3I_L*C5&$A@29mRg_xuR>|Wg{BH)uW>e(N
zecomWHSP1>B52+!Xc8Vl3pH(KOAy4{1w=<Ih%E%fmI5N7^<Nl7l`g0zjmBVQq(~mr
z3ZcdXg_`2ZyJ7)Zw|wswK<V0bGk<*EHprMOXxmsQ+v6(>@D2hc;pxZ_Zy>-08kmB0
zWLdIfEROdHj-A?o0|R6;9J|Eg$Ow*I1xLD`1^hTE4#;k?Kz0`(dkB!}sXE-A(nx3^
z>_m<%91~(O>?s(!1Vg$hg7SsfI1*+B$>dlhQw7O1LDH~=mSpyrEmTH($cCK(1w{n5
zqKYef$70!6u<R#T4oFoqJdX>FJkyaU*UgMr6bA~5g9Jsox?vRg^6)^Y$s8g`4vj@J
zOOVVKBuUpx!I?HAv9}?P($;>kiVhP<b7LVLA&}+?q;%rKkP6kxFc~WZu^c5>=Eq{`
z7Ay+|OH%D|kUgv)$oG>+?0z`O%mG^nlMFyd$6{J6n3f2pq)Mb9CiHS-jFE#W&>+oP
zfGQvI-)S4xaj~eD3aSqZs&s<Hlc{pPHjvAYjF5Fp!~De#OUq&*Ef+{93Z$b`we^we
z=uj_J8@8&|%nG-&m5W3Dp_BE#ncj+6Y(0XlSFoj9PZlaUlJH`#MrR3NN?}w@>|Win
zsu&BWU*HS~oP@sKLo*4+GF&3LqKXTG@sKt5jw@rq3<{Vb0h4YP%wUE_Rz{&27BnZt
zqNxd*5kZrFl10`>`v<vVt7ug$j*|t)8o`maTjV{y+}<>~2x75~{nS`2(~qE<<ca6$
zf+gLeKfX1ciQHLY&Wgozh~PO#@T6-^x}{T}L&sslON5UIoO5I093*hg7dYvvYB*nj
zeN+UNwSwisSS)pX<*K?!u%uhAVDCcX2}+HX;EQ9STp~~|6)5S32)aS3QZ43sO8Hu@
zXS9#3H$!s`3u%2cB)#!*7sg!X&gBM@*E^fu1(n(XHFnd}5P`FeS0O+DHkVeV?kgp7
zh?n^LCL5Hr5s%$93@B5lBjYG#s`VsuAfmLQR0x;I2MWpLI6{$nF!JW6dI&;Il?t=)
zl{r4UZD<;GQHKedxv^-D5H#}yO+t}u;7zHP>m?0CWiZ;+kQK@Kv5>k2(n5ihaC+AO
zN#Ws{T@zi|(I|tf?U-0pO9a)ig6hb0Vk<(_RIhowfLa<0>I4C`Oh6^<XB$|>mq+Aq
zj3U^%Ef*{&#$s6^Sn`4;p`NM%%PN}56)=AqiU8CrfC{mI`UFtF07@&`nC&VCX`}5R
zk4gf`i-lAcNP_|?p~AI+M}Sl$Mk&+L@LIDXh=yYkRRvK^5G5QP8xWD*doe#4GX)=w
zg|eSa8deLGw8Gp&CyQ4$p6B(}AZKpvr^I484PVv#POzjEX6)yzrej#7bA`G-C^@!M
zJTn&1*#hVs0d!2t?#I`<YILI{4Q53;8A8Vt`a|G8f})sP=f;Bjyxj3|zJObl65MdL
zsBgB!v~Os1Fh^&zS2oOV)(W%>W1-dYRn7hc+JcnO`m2@EVLC^2GI_%8$&b+HSi^yF
z?_$CAu~=M}3a*a}uGGUxsg{RU(1$N5WOh|^SuCh41k{xRDy`J7EEb1zxc#$-I)#;?
z*{KC|RV<)u1kkktC~faFQl!m~#^LRHWXv*tLoAe=1j@|<B`t5%NJ4{~I^z?pEZZa8
z8VhJC8OkfkzFh##Pbt|n5ur}1rbb+IP%a4TlPHAw^r>iA4GW4-8(3bkEendzBEQB3
zMSYtndqTyT@~xDF=lcmaA0o5znW-#2@ogJzG^f}{a!n_zdPzxFi#jW(&foH(E}F*V
zRXiN8BBfBJn$0Z1fizA`oY%+0ic|ynvOSf^$He&1B;`YA5qs%qFC8-@<zRu&`|@>B
zqjVT9Pba$c@x3Y?wzNJq=o}vn$$ScJE*&xRS*~YT0;z+Yk+IP`$PY(>ChxT(l`sRx
z#JcE6SYMuwer68JO!XZSw4Gm6yaKiS+AwFrWgo_eBPlEMnu^H`gL63a1``m_V?TSf
z$`B6p(QQ~6wAV~$G25Tb$gOKxM3|5viijimGIcM7HCzyMw*;NjbHaR$m{TbAj&ux`
zPA26pjV-;r9jrLhMTa~qbRlD=uUZ+*EF0I;H+A1}UE}(O>7V85;2o!<ql6|Mv^9JP
zk)YF|y$Y%$gLHs{gdJ!-xa}A<Gb4lZUFv-5Ou}02ALY}EbO?{tK3RRljt|ZbjQ00a
zjr8RS5M>O(s8-5KhDvCirI9r_zGsi+;WX%Qaiqla<Y79ySf&_bDU<V{#S)zzH2K!D
z9Z*O2(8P`e0~R4vf}DK8(HV7&(|3NBSwivYWH84sj8=&*E2LeSS!+bF^s!cq^Xfn)
zm2%Q34$tBksuV=RsLJUc9qb{_6Cd>XOB<b-rq4W@1C{;6w;_EHv`Za%>>Z_JphF{N
zV&xD$xHZwKWmq{jLAZamW27>i8}`HDYE^*LH<W0Oa0Td_nw_H3uk@!Ea*%o@9{`O^
zJ!xvDlMZHbEINb<T^K4GlYYEos{%5l4X}#O1EBzokJ=bw5XaEmiq!TfSR)>Q$&@wO
zYtZpeUzIE)*^fRqE!7cO?-A#w1No|>(*E<~ewIXLiO9BEaEKY_&;z;3xoG~xdKrJV
z)B!3!pv#>&QTm376IIVcypY~AGJl-hN^rnCo;8J{v5VnKXecq8>EIr$d*PD9W@OM=
z%Wg3BaHQ<Zn3D@cJRN~Hy^k;70yWto)bo%MbX4|0Ww3(oLiZeaHC!o|de`_4{^Mnp
zeP`*H45YJ^EoQsuz(4S=$qb+#5R;@-%M5IMNF>oYpz*j4)uU6uArh8#p+Bl7+2ELj
zv&QC%G$fXbB<@t49+E2h*d60sLYP!ss!kO*gScD_mnUUqJ2+=L44)fhpMrcuu5=1d
zdUUTDI^&Mq4_|a_M#azINS%%!l|PT_G@~1?RMAM$%qTI#)ERWbod>;}U`7sf8)1|o
zKONfqQtl$IzjvOD_M!-tn}vC<lW>L{m(Myf1Jy1o>YI-zo{d+QzV%J1L9K7D+!POC
z^<AQ}x04NR3!HF|!j+dF?k1C_8m!a{HplAo9uip1VX;{6j4H8Y7$1nwEKhg>BYCcV
ziZ)J2(2gwN(I}a!Nm1p;Bx2GM`b23l>lE~PPa3<)`un3w^Ub`pbS2!8b7tlTyXlnd
z6fIwy^E4nv!nU^aDAKX;{7g9ls_zk1W>jphf`rSuq<Q+u5!rGP<NZ>&KQ5{us3V7p
ztJR!@mir7mp}I9metyuKPMu=vVAI$(?(KgDEdW{eBSpry;u*Wh1Tyo(6&`OXvjoh~
znhTD1<+%p-&QZLPwUP#%&^Ex4l^?P?u+HoKx<!r2RCCnzcZF72TGn@qDz_mj%S%Gb
zqC^?`$$86otkKLEmYATolWB?4_iIw3Yr{rjBF0i1HY*=|KFF==;+-NvJYQ~J86NJ9
zQ<5Laf6Utg_p6VU{|y;N6QI2<+}-4uN&Q{O$G6&ylc9#F?Vur0b~Fu%E1|w^)Vfxa
z6gkC+j?Z+BVTuu*tB<()#$=NN<({vxJG!Vp$Z)V@RCDBIvF@G{8g|BH<A=H_EvYrw
zcS&nd3-ad8p}sk<p*8E9A|^Mq%<C~C4tle#+Zn0tdE&WS-?8aZUy~3ym5WKlWQ}1e
z7n8NWayD5;0|X1wHc~fMA<?{k)TL@XP>+w2g9v?uuBwEaIf*m%QDbcx>xrAZ2cI&k
z-<=j!X1s?nP1Dk3YWy_$`^l4?hUDbcw~bnv$MtB@@yFVuMd#`xuHO2ji+$0u2u_LP
zY;*rF_2b4~Z_&B?2um{=sPE>}{rjAoU|b-ldF&vRrtD1neo?e1O3rn<JUt5%72@MW
ziYjIY+Dzt7QDkC5f#tAnh6(L|<8$%>_Enu@8szc4>I^ipqvthybQ>6JM<0{5kCYU=
z=L_VSq{&en<8q^wbGkZ?O)pxpnfuAwY)W?eKCRC#w(x7Ubwi5U7bKsAtBhArvqnz1
zi5$hVU~WD-e3F)Y<YG@oTtSi0wenqCKC+mN0eUbuA9U?2Ej$}-w)A`mipqOEF=MUo
z8pWtwZn#3b{{c6~{q5M4cKZ~J+Oghg9{Iv0z4I#EX){yZXd}8@B|hG!x6wR;xtz2c
zip^8)VRzPdl%gBTt#9YQVaMW8zPIXo$Qz}@H^pTrmD;4BJ<0|*(^`G~+#zOGG+qEi
z(4N(MfPA8viBCGcGiY3y`X%jc<Fk|LTuZHGy}+(&!$hBiDMhM<xgi}KOKgH%m<NS~
z9bX&|5|rg<jm)0n^=tK}obcJ-je!%sTsAl-d}aALvDvgKdz;0vgBjc7Bog}=rg?<j
zrq|=aUa_*!EGd)y!8Qm$vjTjy`qn#+`ANM4nlUnGp^be_UVOXuM{irZ_Gg-oj4ddO
z!f++Wd#sv}39fqbC16q-Tqc!FL$rAptJxN^O-IbTVQOT~CXne}jW)ly?Kz5WOvPj^
zX@}4?=Ix`cZnXt(XC;9h<Qu&Q#*JKx%$T7~4Wzqkit_L-EY;C44jLDNk#=R;PxH8D
zgCR?j9th!qXu+wQvn!|t=WTAzrg#%+X!rYYtZ#*hNr|jFAUR~uak8$sNdhuzSZhqg
z^_Eo0V}vwW*8XbS*y!JEf2bXAPQj!_nds@WK;q1D{}9oh-O=7HmdpaV@aQGOWF{`@
z20<KgCPP0-P4>7O#n5AuLT*-(@}Xe9KOkvleqep+8;9WPx`sDx3hAnZH*E^Vs)y7N
zZ`u@QGMP7R3g5IT#QiP7yI2A?V{h6N!q^R#d9n&MR*Y}j6q4P#whnEo)cpUAO(D-k
z@7z?)Ma!GXT=d}Zw9H!Y#jofZO!?ZOsqts(pPtzr)P^bNY&+%j&#ArBaP))cX;WjT
zm(p&0GY~!5JiZMppy=%Vv_+^|^lZ8rY3B%O`^J*SQXdU~C(|El1AQ@NjEzWZ?AIAf
zWTVwzQbUQjH9{a)KOs%b{~R@Tk<{4aSYn!504Cl0h@&Ygb3bV!h<;N*2VvGIaL-u6
zBCP`xx*}>RV?RNWW`YiatifQs*aeLJxHYvy>;k#^Ns+pVvLJDgWQ_^;h-GEisxYb5
zX`IaY^7NDAGB0#cbdxlvu2_nqjSLeWn{F*<KXq<vLkE3#2z`zkv*v;<f^DfWo2QQ^
z*M_2lAZv^mUt!pQZ?sBmM};-y98H{0@9iMzvr<QHk^(k`TvP2LDUfqCNs$}C4&uHf
z#KljZq74g~!gfraqO<pr8E?1fAktbej*D&UepAMz2JLZuqR!7xhSR3eLCSZezTBi3
z8^?B$(vq>GgN`2v9q}{U7;8ipzIM!PW6JPXo!!G|Zt>`#db*fuPl*-8=0=c74bS4l
zGBS4`rM}(4G+m0di##Bf$V3K``8KRvTm|__cUVd~>i4H2)$zS)j6qjBdei9aePqTP
zbvcN%)|1l{)o-++Be&jM8*4YZ7$4DNoY_1rb$e(m*@=uBnfksqP@kwY9}~t@^VX<x
z_gEH$tSb|`bW50}j~HK#CZEn(gUk4lr>Q1U>x|kl^2C(kBXdl3W{olTk7Y1JiM7f+
zu?^xHTZ*4#w|S<cmwH*6YrL69(6W--Wjo9~g4y~g^T}+Q!PS~>#*-Dax@;R+!EAkG
zIgL4+Ro6pdv(AaJVj5-NS!&h3vk^<!hG8Cb&E0q`#$zcS%kk*JqaTkl9w*_k3XfCp
zI0KJ!@HiijIv$tcaXB7W;c*=vH{)?T9(UpK89cs-$Jg-q79RJ~BcSF!SPr<x+n33z
z8aIEywT^a$#WK(i`?6*Qv@Q*bzIE#2t*Js^6YErEaIvfIh1j%*ELv7gYiicB^;~k^
zG%`9=^qFI5tDZ@4_&`~(m!2A%xngUL^AndznObkPG&}-JSKd_V)doJR#lR;}LFpBW
z=6<oDxl)w5PE)nvCVNlQC<io3`nUNxzftf0uzW3<MN-GxDDgeZAfYUg%V_e!Zfteh
z(P0L&^~+MyCG(HbZ2Vrsv_<kFAuPV(7{5GZ_g<=!52-;i)2NuME8Ed<nib$Dc8l>)
zpsQ+*vsTt!STmZB5t`$*^~$fmby>ijL_4(g;f(z>id9+ksZIxJHy9U`_4sW|@X=gb
zz+|;!+tQz@pCrHSaE#)<S#{kcG+ArKsOFF2l_$X<Y3;D!jL17cA$ejj)_^G}yhkXE
z-)$Mp*H((NM6R}W?6w4Q^-+~@GS;lxE)t@4jh(dg6i4U+6s9Jj58wpZ`N*i#UZwFI
zld4T!EQG{wLzF8uy3}{gu<7{Qu?^wN(@)ad$Ja{zgOyUDwO%khk^2P(g{ScNH6Fjk
z<M(*{36EFtc!M5+fqbRRMB-Z``(*Z;V(U6?M{A78+s~b3+iHRN)k#8C)VxSZ=Pv%q
zRmAMc7)<X%UL}WW<kM;p2GC2Yd9qL*>Yr)mQ|6xa-Dei^!y}ka#Y8r<tMfNeCM*Bg
zL`+sTx;<1gYlMbru`L8Q-_h~A(IL`A6lfY;9vvJqr|^;<xsKDIHp$;tLkWA9&56RO
zIm;PBX#8>r?nyDjr8!HL4gqi%lh7Y?ECWON`UtCUH@=)7>K`Q+(&(qXVlP~vC1|A0
zNd1VAx=$=5UQtIBt*Ktt2JJ-iB76++)vO%{!{-^iX4I|~YTYU{I@n=`jdrQfRF=`y
zG^)^=5wT8)h;N1{jdpqI_@lZ9EWd`j3teCBXogMl_mNlMn3hGn3~DeZYew99AufK8
zh6^X>9>prV<{u%Sx0<we?9oK#?k8-MX=XN-n6j&cvINyu=E}moWYtp7=!hBd+g5Fj
z`TMEccD!<%tDY*Z>0_*8LcLOGR+Bdfz45y-P*?k|52@gaI9-+P*o}$K-bbykM@ilW
zZx!O=&ki*qtQ}{E8ngBhRqwzAj`RfmGJ{K8ad!%B@q!u#kb?rHjr~iL{QXq=Ct``|
z`>fCxKfz>4%oRol2iK6NtdTs;@p*)A#{@GXZ$DkMqL{9G&IlvHW-ac^DG8%adV>*0
zdHV_5oCuqqg0yc6Y4Pn%@iku{FWUYpYe#!)%-T;_x*EGT9a*NrHUL>|s;>uxuy}=o
zVyeHY+M#d=<U00Rdbnj`Oa--<m^RhTPg^EuFWo`3mmJD<>~-^m5EHLzRU$E1!My;Z
zW`5BQRckn7KS^nOyJv)`cr$iY_UdA(e_$j>*Iwk;=!$BG8GBUbe&R%r96v&ZRYF@}
z&Ak;K+u^Yz9y{YP4v+D8bm1`tkA3i%fyW_u%)w(G9^H5>#$zcS%kk);hc3;#oc1;A
zoql~v-0F6HKnW1u5%YL&j7P|X7kI~RjCp%g?W&D1Zy!{2)K!YIo8?VM6fQiW@-njU
zY_qK^?*QbFHJW!ILU&VWZGatmA4fJoy+Z_1LhIUSL<u%By_o`TRxCK#J@e)WxP*3>
z(cqfyhk0`a(&4d?WZ%o1Cy>U(uLS#7-cbT>ek?fIpz^u}T+&;yqBecAF&j?aA^~-D
zEGXGp@)irIgr2XORifR4^7UNau>$D0SU|Eb<Si9I3AeGD0}5>*`Ei{fxR%A@lHD9{
zx!@X8=iuI;@p7?<WGlwY3!;R#BSdpb?b-O5(~FF`pDM&ck<ApZPoO09Pah2>dWXat
z5Im(=JhGkPl?Bh3geZP*!W$A?6@yDDYNEFzysAOv)ezEmaE($}k5Q}enk_IQ_6W|R
zF*mS#BYOniDJJ83!f94C`&&`=d#9mH3}IR~bq?kQ5k?83;P=iHlnGs)M5Ao3vG>kK
z2`sqhBt$1VdGB07mr&^tjjlz_x_5p;prTIqE)<{%ZyJdP3S}L1uYr|;RnU3$gup~a
z?p-Xv#w3MqUAK1$O5%FHG$BgSb$gcy%7n%|(d=`nt-UJ*<CO_9iq_h@S}=~Cld<Y&
z?^;25T|$(iYxZswlnFQDqB+@u*4Vo_AxO~~d$$S5gj@L0Aj1mMV5_+Ug|Iw+G9f@w
z{Cal@z%kWz7@*b=`-k<r1?4>nQHuK3`<$R0(}1cLx&b|}AK@1S;TIDk6g{l>l?{V1
zs0i}DCJ4Wt5TU43y>AJ^F^Qt$p<nOnXW_R6;CB)N6ve3by$u2|#KHRn;13c4d;sxa
zuj&0z0FG%6t~jVQp?>B)DA<0S5S!>Ly+_)OEyTA+1=~*&V*4rLG2edHZfrq)qW8F9
z`$a-*PYAXr1>2Zb#i~xN($Qb7PYJxI6T*8&;5{qw5^7|l#}5TBs6O;Vd`=)f-yC9C
zzjrRC+~?!57LSkOaS<LD<8cWdAIIY|Jg&gw6L?&W$F+D|kH?L8+>FPqc-)T1C+XqW
z?|HugD*x>MThty0?yYYzO#V_w$j{6W8J{Z`kKSF#0ctdFrSH7LWAvi4+Z)8IA$oL=
ztUp_uPtmP>-=VMF-goIqmu-BH{#omNU;Ti3(fiAh^zavdKRI6}AK4G#{$v($4zcd8
zE17*vxzgVc$i7<W=C#*R?`88W`UfL=_o#m{M;9|UT1#oNzbIti4jO$oEE9cL)E%|6
z%J?5aHRMC|iAZ@Ebh1@?lq>CQF6imIVmaICY1STtGq)!h)|$6J#@lN6?SU|xu7%$o
zjJG4<x1;pJTNQlkJq5T=)1#X!QaRK%gzlmU?TV?^yD8~f&0DSZWc<v5T_U)gWeW-a
z#&^Qyk1MZ60T{2~k5>Yq^t3^!%v6TDb5!;)mGmP`dVN--Y-N2}H3IEaKs%cQ<$7kI
znJA#zwW}X%LN%Z}w??=V6x<2T;j#cS+^9qU*=n&*J1F<rKO|H>M*$A!*9du{f;_P~
zWNsY{Im*tBAg|6>>4*hgvR(GGIlY}*RF;jv_f)|5Yzn;o&efIb%Gxm9&8_c0XwLU=
zOzvmL!1r)|jqtk^{H_T24^huS{pAs-QwPM5t&?OLd-WNqEK9c9qVrax13PcCjWD+8
zzH$v-onrblrGC3?vyJG07X6uSY_b9|H(@uXfYo<UuD9ua-=Y3;k^LX@eZ1NMa@5{w
zwrLU`Mo+c)rL#XB@6Ip;7_!d9+q1%NSwznczvZ@ePWWw|*yMfKybX5i^AtZ$Z;c=C
z&5=j)5o)l-k?`#Tnk7-TLD{%e%$1K2NL{PKl`A$U^yeA$hAT|d`Qf*Gf5QdlZIIHl
z6s1?Cpfu3n#Z#S&(q^fc(zVSf{is17ptK%-%amSZ-UcbXR#Eze)+pT;wW9^x=R2O8
zW0VyN*JfjUrfyvXbvzaLm_cl)V~M>a{FcY!OT%xO=8uQpa;sTy-Ud1GWyOgHTI0kx
z7^IUbHCn)RkI#IGI)tB1lA5`68G#N<JGwOQ3PX+I(v^t$3G+5c<AaLE$C*a&D#VPM
zbKmWooatW8&>1bLN$HGD&2(N3Ou@dI$-LGOr21;pML(ekdQK5^T?+*5pjj{COK$4b
ztb%JRx|>#Y-I#=-UvHsHhs4-#G}uiQv2flLe#^pnvw0h=xc^pkzM|;71tle{xGi%y
z7ELotKXp;3xM+zHh{!bEijq`y7@F!G$_E`YOU?c@UG)y#3VO@6^-=4BzbGpH;cQ}N
zrCap=K+Bi<{K*4kO@xbWXq7n`vxmv^7tAeJR-1on)!xZ!**AH(G+ZpxVUfutGn)=v
zSusp@JJ*qTd)MP}H4XUP-VGEq7Ay(Uk9s#D?X#5DyO~1W-YxijD?Mmla><Pph6t(9
z(-@KXowhykKDBV;jV7GE9?QzjyS?lY6!YUq6PM}dk2RTch%@~H7HcNcdm1yn+?A=U
zIK8hUd(QMDHdF5|B4X)bL}kZ3b(aJAWv*rNnIq;dDV{uXOtGJQB-H3g%mU19$yu|y
zbT8X6gIY;`Fk`2@Gctt|AD+oq*JM_gMh4V3-?8-CXzxHKU(@f0@`J^$%y52Wpo=^|
zuPULCu1s;HSMuhQKy(y^PFwV^nUSHNG}+477ejPlD_`!)3{{3Gr*fs5&s3@z(wPpE
zP#A!Pn)Vw&_=<F>VYpZ<WRxd?QCwwJ8RFyeKF2e1V7?>#m(6x$=x^6uGh}qoOMIcM
zde&sf+uQJ(LZz2JtoJ@m{9LQ8e(r&Qtb+u7hTo!z)j#TInctv_IXu%GI_m<x#hj&<
zLYady;}y~4yD~!9LETKhx*g)6@#OS`zGA;c41y}vgV>*RS7svGGyOAhB`sKBTy9k|
z-fa*it&jOE;k{tA$BU4-J6sa?J`v+bMdHp2OWb{sxFwa9#UbMdyl=Ez&gkQZS>pVR
zMG&weRb_2P2lIX6ME!Z<#LSG0W}SXj%g#|(wx}vXHBoKpWFVE(ETpp#6}ryeMp{&<
z(G4beX9ZQ#DJC^Zc~o%?^-%?1G877tnQ;sqVyg}r*dv<CLe5}iScna`4f5F#-^Hzm
zE)>h;ht_1u#e4x`Pm)b7p+ub?HG$z`FWtH#*>D!MOsA3!@;gt?AJ!1CVrEv6oEuD9
z*f(i0nQ0b>dW*>Cu%UwgRVbSYV7~mdMdXjsgf~{2{$MJvi&}3Xr^(AHvJQf){CLUz
z+>MDO_X~&?*8v|UlKU%>+zY~z`<YEhau*e=I4+v8mt~nQAyesNxAxr1o5^u$Xn1sl
zL{eA}L8ui^8s#dYwo77#PFUGCi9ZitY3S%oPpWbwLGl+2@xmY7MKMIZ0mKI0imbbR
z>R<*+wM=cGGFql?jQoe7Yt^Eb&(PNQ2xw97=zctvT_8gxZ?Rb_-)+*Oszs7#>YAvN
z;%0@*;JY%=o1+Q(%ODXZiLcPmz$uut!xtTkO9CpRRTAWUA9W4>vZ*I76#Md{)_XqZ
zH*q5RLeQSDM2em{ZvOlQ>b5|02Tm7y`sn=I(@m%w1Z%<rh=i_i5(g0u6^;8D(@5LW
zMrSK2!dNU-G$Vuhp;1SV9)^+O_rc-BTMlJ3Ca&JMZcNlLzl~^d4ReV!%on9$UKnne
zCtyg~5DjzUMCqeRgrp|g(oD68{y!l?bQ>7DwI<Vx`;y7R{VrL^i!8h*vQQ7p!eYNH
zaAOuFO{$KX<{QTjQ6II8y2Gx_eCit!TE1Lb>!NC9aCoG%HlZuN7uEIcvpuFsPH=Lv
zY;k^Y7~?!ofNIMl`bX}`gV8dT9vof|bW@lZ@@O8KkqBx`k?dbv`jHcr5H#TzaYqw}
zg*j0^BV*=d%Rp2%NYrxbd(`YU)WsB}i+LoOSUu_ztBQ!#=EqTQ#OtJWA+aKTgmZeo
zD*0T;IC710^%?GB6L!6x@a4&5e0j>nm#x7KUg>TleEC=kz8s&NFSwtxNRpvi95=SU
zawR`PErEad-Yh3&zi$ciTr#46<05)zA$k`f`qC6cpPrm(>7Q^DE6rhMC|dO@XLO`g
zuBmN86+DJ_oOsDN&rDH{;e^N2(t92{;vj2&my9(px>$1}=^&Imj2G6d53#2H>#>f*
zbdPdqc1&-md#jLcfhL8pzcuQGp@J--s%F8LayYeJVMkl~!fgq3{L1Ckh+6qlqn772
zSFOmMl-^!aE3YGwQXR3DC(wIc;DwbFAIF4kAEaiB2+ybhBH+CHCW;j|Uz1IK??i#K
zH4?Q3r`ru>8yA#I&?9pHl@}=6ZV)KjxuD2p5?(=|Y~MN*YrfqbS4lg#;B?4{*DrA1
zlLAf&shu*jqYKVf2bzjnDRAD~I-CSD^F9}pn+3{{K-p=7K-t*^<s<-L@vRD!T{Z}m
z_q(8s3Y1j>C9^@G?COHDMxdM`P{y?m#XAkDoua#&3(jpbO(mTna6XU%&RL1z?Cyed
z4zgjH`LMv*BLy6}Jizud5&GcqE;#22obv_F<+BK4^1*-==X)emoT~|;cZ$q}_XI$<
zUVD;@j&(xE#X`rG8=Q_agpR4n=-A6e$K^uD6+*{VAv%`6i?y6|^BT0s%=m~+lYzcW
zew1}|{j`QDR=YZ^&c0)k{B_uhqjfaTS!J7pm!fK$&rImd4oJ3Rpf88~ja^ltpjGv}
z$w^ts(s2&$JcZZEi<Oc*2%GesdJ}2|?&vA!dy3`S3eqBGR?MW57~9-Rb+J~(&MxO^
z^Q*7a%Ns>2GK{vfw=y_b86q`1Yk(-N+VEERY27}76!^54F;p2LT|<#BwnzTt>05DA
zQ=_(^x>T06N~Sd8Q^gKbv842)WU(yBcIcf3!|V=~tKsr&c9_kesjB?77+Cl7w70bh
zZ{zflo9KsZO^dD{n=jd5rFX@X6|PvV1PHA3rYCC*2e=x;P0|={md0>Rs4+Mu&aHP&
z$_CmyBpD4eT{L`3Xt+~oxb9y>!(qv2INU|U=Y)pO3k^4fXsEx^v2azfS}hf5vxH0q
zv_zB4jh1Q8n)jU538_qpG-yS*CgFxlyB8g^N2^u(9sV+-V^YTK0aAOTPHebRR@=gz
zR+zJ72vw~N({?Pm2}cW`q1r9A=rWmBX3FLKU_MhE>?u+oHPmmgRE7%0A=*u_9~#<S
zTbf;KBjZIfq8yd1@)o!%@9R=|-;m0?DO7n$8qpD!<1KU_GsspUH}lvf3l}V$bJ+ac
zk`FFAEYMXfPDbUiE-Jq-RNg04-jae!uU%9&?D8w6q0clRz=vEiKIC0|_=)i0G2z2)
z8<G!D3}eTTT-Xni0w@O!Sm_jQq%$(C8SLV{&0N$)QG8K$fDff)d|2t?!_&ftUkM-X
zXq^w#Fnk)!!DQenE^yBaxZeo4PqhxN&W0qYfUrV3sF2qq+GsNLt6k{-An0Eb^mm2O
zFD29PjXLHOjYC$u`Av2c1+$IGZ)p<7d&U?Cx)&}vY(_?p2IVzfYOA+UDb|L@k7Nd*
zSj-R6#y3s+$iyPEI?oEn0`}HRw6%q-X=}P8IKa^MIoYXYdg*`^P5R81b)F`HB)0Th
z7mqDu0zz9gHMm*Drgyeu$-rohGta0S8Zxvg?~T&tJ8l1Ca2cqS3$)Qq<_4L*QL-y3
zhK8_gW^u7d<I%FiN%1?{L&<BS#pRs?BO}AL8IvdX^V=-VStp|!Cy_}3_KEn%WR6>#
zJazh%{U`X8pY59vZXdmG5onL6CKH;|T|)Dk2+ivvG<S!Drtaw}bB68*$iP5qcqHFb
zCZ0lLPR2Gv{baaC^KP054A$6|1J*kboX}r82JT?fKQkF0&UW$PpTdWYj;EC`?__-@
z%!lI)A4r(3{=C6>B}jDpY5s6Bn$LC7{1#BkDv!4c&7TX?yueR$PrjEbwxH&a|BC88
zKZ4W<-TGP=sSnZKwjy;)A@xEeO0G^}RsXw?I-vV1Dr3+;qF-&z7&O{i&(3pvdpQf(
zj+7RK`mQ3g9_PrJ7^6^L#*!(9-zt(VdT1jaR&B}()R^cTG)@^OjkAVJGtJIC&Dyan
z9~z;lI*$)}moB4aBAc0`?1-peu4KjyjP{ceU4I|-#gv<dA=Z~Z(&mKiOJ(tqp}&c2
zd&!86-oj!tOOSS&A^O^G(uR9M6T?*;bM@9^KgsM0Mc6m5b4l(SODF?wq#WXo>5*CV
zPDyTOl(p$4u1u|c^QK}B@+F(866u4ZM0!rKmt?W31oBP8W)jh{88#2t_8u}`K4=44
zE#C+Macg&PcFB>oTHZ<I=qq74dedqdJsB($#Uk$iU907(dRS=HDB0UZ6fB=kCY<-U
zgmX{Xs_hct{90Hz&+r>2lOb`1?Z)*<1v9fV^;X4{C(6WLnQ!?A)-vh#?MTIyPbWjf
z4pN51|GQT9Ve{FNF<n+u`>rb6<nwTEcmM@E3>GbM<Blkn%aty&>n|5}9XILV%(AId
zrgmla+h_W;t_%gIbY-SZ*?)h8r&-%SDyk<xLT6hzJ5$oB{3}-~)Vk=_3CuPsLpeBx
z?BeqI-YI-Noo5tiSIdIMcd4Jdj$2MZP8~1Vji;k9v?QriJC`lX0998;Bh(NzMtqyL
zyepf{PUspRgy7GDlH326cP*RJHKl7R{#(9$!s)&_M0-qPCNUYtwku>!*~C`H+f5qO
zmy<Q9ueuu4{?eePON07GxIrE3Z&2h7ddv+9Er}0uaP?WSlMeOYy-`WyXKv6ARUetX
z+BVhDo}ydE2#p_#6?9aBx)R#};+k7m+($H#=;mV@vlN1kfu{3j&2O43Kn20XG<_;9
z*jC%#PPSKlN%Vm&Zx3k@-$~XW?sYW?d-P?lG>C778^oJOU*s&!n@3-0z%n}ECX3p{
zC*;CoH@cbsl}BH+8vMt}8s)>TMrn_|94n3T+u=ressE^h?}S)r;>U|K*Q8H6Nz}l<
z;MmK**Oc*JEOt9^7mwpIU_+BrGe1uzO22T4l07xkBck-(uqaL1)OUjF%~La>Q!{+*
z09<Kma4*eAndQ(9j+y4fC8P7cbk>oq1oFKDk^>#WgZToimqzG1QtS|!O$ut*bkxY~
zL{SVfY^V*BtpU`Tsw=fqOUaPWfvyOhzR4DfW-&xYHaKXz0>?t>l<?>X_Cr?Nv!R9(
znzB<lwJyn65Qg&}QfGb0G6Q>8$TbL6(UA@+A4nvudR1o`ZD5iqM@`viU}e?GRA#Ui
zBRw(ci1l1voguZI<_0+X{I59=wU0~)pGnqX{MyxF*z-`s(qVir++p<d*rTlkRd*IV
zHQQ{!(W(`mI5-dTqhB^EdN#R(Jjr>Pe^dPfKx>OXqi1MjjVv3ID(GW9OI)d$B(^j9
zw>mgPc8I=X_QS*FHM!n_{0JFMkgXKC*pLl=b>4mh98Nn}g!B)|g!GRtAw3JC$vX;X
zi;&(I64HR-(b|S%FKoTn!lAu<8#$-8!M!=cm)Dc=<qa2K>cW?cgfI7};LEh+5@pr`
zfvx?GZ%(9~-xSf}&J$fB^js<Q{IGR;g8S{8Cj<9Z7r1K$+;sx(!LfpSdopnEaDlVO
z^==k$4~4)j_1dt67~_dOzEJ}%V3c`ulNBXRVrU-0ej)e>Ap49lD;o^-Mhcsx|KGAA
z-IEUvhIo?1htGD6VxxEUUua<qG<s*iZtwZ_ycfK=oiE1_>{RxD!*>3~H&Le>vz@<#
zt82L*Mgctda)`I{U$b3H<aYjhU76bL{6~;IXZonkwDBNolM#J6MlL=lBC{c^vds)=
zrr4ht2LsynV+M>wXe~6}XZ1nuwBnZNA82fIUW8%K9xl?w-L&@`A$>d&H8bpKB%KX=
zev8xr!=5NOoh~>JV^+e|`+~rkkOI!{62Y11g7YG>VK{#fID4jm^T$MRx*Tu@?pIh|
z7A%ueVEJ<*ELjH@Z!Y}iGi&}La3;47Cwy`?;;i2k7nHvO07H30pgh)k^9@earlC}~
ziO1e9a2tv1f{hX4zU5~jaP@k}?15tM%36l@iO`2+M%E0&B)Fh{QSUI?%T~C9*zzsb
zIy#QYm*|YgAh`p~_nTuV1FV0bEMO0;VivF#4*a!a(HG>@zR5VXzl&3E6;8cPIQ4jl
zQ%fJk_UP<tk=Bl+>6)Y5?`bzQYGW|(a|ub*i&o6U?JupO4HXqj9p9mqJ#lLWE0<0c
z71KD>!MGV29PO}*r*Xz+t45!eNn}g|ZBqQg-?){<HMyx2@wl1uk3B3mH9OU2q+;ke
z(|mEJHO8Uc^gv0c2ctL(o|d5CY1!5b_Ty5P3AA?~W#0iOojLLq#4}|4u=LFIRhLmb
zLZ=AT9h!nV$*8J`P1l(d|1!Hqw4IPjcH+cZvE0X`7@1OTkStMDU%tO~G`afw4G)#e
z)A-rJp~;%$ELW4<R+{8?(j<QwYLfL)O=^fus^P(*V3120e*@g0EZfnrES_0MCS%ro
z7qi9*vvw0^JsD!w(gV<cslP~mnT|=!g_ROsw*=Kkq3Z~?`RKk1@nZU`#%%Bge@Xf+
ziRb9M5YOLF;5Yt4eHdN^X~G-)g}_AiWfISViR{l1FCmz4G;4W9eH`+cnHgR)GC-bu
zA!TL+3RN4nDX^i!?m6j*jjqNGDfPjzxjUI`FLKHD1d;8DBHK@eWV^md54LFYc1uM0
zZXWYtmd14ll&C>!k4Z-D5*M{ogxaY>?XSYrK95!$HQ4=&_LsSM-N}lt;{O{X)4Sr<
zsV5|>k`KA6<N&FX1Eory4OhuB)v<;LY%nFOm)bIqYX1<eWzqY!Xx2|mM*IpF@w0^Z
z*+TqtVd8t;#2a7G8=eW;CZ#7C0}3t%%o7G2DGc~cm;o0?Ghi;wCh34ZE|W9%$XXT*
zJ7_DsM258JtgDr>8gP4QU%dD)<5OQUJ`K3|v`F}LwD9S-VLmO7<`cIP3{b4V`j?UJ
zB_qA;BK>$FeW{TCLYVZWF{ER;x8bSwekZ-dP%_GgU6kA1mYh)jyD;Tn?O2Rc#{N|m
z9W%>kyc+HzD3#e|Yt(3lZ{#+0WPleuq=~ST7v62AFF3G<j1g6vmH5kHzcl3P8#@aP
z`iQoyjXoeCS(Dm@J#Vb&jan5%jt}e+u1d!HlU=;`MCw)w?_Uh@e(4_A1z{hv#mo@d
z5a=0^wPTiPVf2Jzh*?klSCIAOOQ?sorD$$xv!t3}ix{d5V<hd%JUlScN*0cBi>4!t
zUDM9Y3>8<K8LXM_YC9aWU{~3}^dM^1*Ot28eu$kbkWXK_Q-gz~INzIYwDtKjI3ro@
zp5?0DRZ_dFrFLHm*Y4N2cFA}0NT795by2IJ%M<wi5I(_QtnuMb(rEB(II6XcP#n7b
z>Xaf^g?2Nc9l#$nB|9xB0c6ZHT&d!WhZ)&)WfPJ-8B)O^6g50CUqS9vRjix{tv<=_
z5?I?N2|~GF6tHZff<U7NI%bK}ByC8EJ5VY`JN_VnqpBbY92o8ED~D}ZV&w9IWO8|-
zOD@k7xjbLw@=qbTtp7*H91;)9n8+R>N(M`0)JONsaMfTLXJu}tZ&}B2jMzyfV<;$~
z*vRfjkG}Y)Eq%e38Y=C!5zastur3*^Kjvcf;Gt%3;bLL+D`8fjAfAWmI*UBrXtX9n
zv&Y_*8ETlmkuabXWL~s|3p;}0bcgRM@)AWoqP{%=22}x$tWU;~%Uv9~LWJ*1;mE6D
zj`TS=a)dhGJ{Nwh`~#FvLfRxF3AuAsBX_h7!+(Zt7*L~gf*TL)1q|<%Zj-&t^Cd#4
z@~Acjnt7h>?*Q%oa9E!_+i#Fd`UUOAph_MC$wM`KQPWLSd3F-xW2fqF^<sw$qUjr*
z+cV=yz{!A2ecS4>;9sZ<@rh2-W)<ss;60Ft7*mC7Tn**O^``6K5byl`!#0#iQ-$kY
znOakYZe-7y{?%rh*i_*zYgiMsRbbm1kwkLqWh6BiO*a5vMCyI|=?=Xp^M^$XS52F#
zKKl{`uSgDNA4C!8$3EHEQlm@_?sRd|x(i<>oV*K(TC;ozZ(v6lVSL&J<zQ4GZ(yG&
zQ10F!Q0{R-vGyQ&f%2IR0_C$VC<h8x3IgSG8wASdT~PV~fGcP~pnPHMP$G;SzUYD?
zms5Ky1<K!A?*M{|^Dm>+ps|A*{U{nPABpWKuSMdRj`FK6l74=X;Vy@`PuqVsy^C?t
z@(ov}wxj$nWY3vyWHU|NQ6|1QF>yyJhX|tjN;&3J_4Sm-z%<Yg-Uki@>ny4tyx+wr
z+u~mkMm>N;ZSMzv=z_A@Y$}$U+3y9)k2VOD2VGF?=<t$2`SAvU@{kM4-#==&^0GjA
zc!NNB#06#Jx`FblK>5$HLy71IA9X>oXUAU`C~ukFlGySojOTbFB(o3xRm+kGY@Exd
zYjMm~w8}(08+k*mnkA85N3ua@z<M4XFeadM(5;#uszE<#R3Nl+%dyYJz|u{*S&j06
z*~%H2V8`KY3I6aMdA|aCL!C!NEj;V0g{{sp0@1OKLfp%~Ges?YyDhcAnGiqe6rpkU
z5kBYDP#@L&8ouJl)DU@ePtRe!?e)84+<no-U3(OJ2jT9vDY$!5`?>4js~V}i_W&#l
z+$+gg@~Vp^88Cu}h+TyxJEmaC`R!*(Sd!?}Cwco9)tFXWv(p|{@%o$udl_NWMz;`w
z?v}X;qQ#wFvS+We!l+&UJs35gly!WQVJl5dOVl31sE%Ze+QP-C=^{}xgi+&DFlxT)
zebgal@|I-!dAdqE%b!T|p3_bzvf;ZPiq(aD_!;-8Ey^2Rrnje1yHzr3w{}rGQ>dLK
z)OMwy_C3(J<|=(T{Hw2xL%r(z-@GY;eEVd`-{V3)Pmmue$fu@2K8=yn1s~*{7%n;K
z&OZPrb^d4%e^UN<_3LG{k7H=sPiWdDf~JUtf5t_Vb*s8qXxbHtn$5i%ufX8b7sn!X
z;Pi!7UVbM^dAY(aGFD`IaodQpgVu(bM!ncfcd6t&Z>~3AfWug_SkW7UcF&*{)g|Xl
z#VW<6ET}fTjx@9+;{q*{#zfq(zO-A42=4;4MOpc^n@Ip-*5JFlBykT!;Bl2hJdw-W
z))lF&+|!k*U4wTaDrY*wW}0{nezp}knPEfz8g6(IkqJhdqh{k$k6pAY)<@ni=xX|e
zYw*Czdv6dE8lodw|GqBv?1u=<^ZkWA(-CUc`WtN^(fY$0$cTCT3>Or;nmRzB91^Oj
zrO%PUrR?hn2lYMg*!`gW=v8!2iuP|U=T~SUl+7HE&1gP9Yg1D1OP7w~)Jw+-K6k2L
z(A`LMXA)guMEe@p$D-=WtdL}?%2vd7Puzl!nCthpD;pg8;-}zO-EaB(9s=O+dBW+n
zP=yjkoYdz`+;`4%O4>K1jeFxA!5G$Pw`{(<;K3E?D9w6>T?^BMQ;W8~(=}o7%DJAt
z;NChX30&>W4p5|6-<|CV%P@Bo1nwt@D1k1TS3`iM1<jpHnP<rJE66ItFoMXJrH*{8
z!d1@L8wwdbV9Z|@=OM=$e9I9cOb@W-9hdB1=~ySe;{WZO>p0>5uW_(rOnqmfi{$m~
z!RR*KW5S`Xj(9IQ06SAU;#mka8xx{ugbBuk54xbdd^Q!!!_x^CN@z^b`vb~DOd}N5
zuuTrO;^5F6(VX>4Z?5q6L=+$5ZNwmVnu{Q-tU6r?Iy?nI7x8pN=-^+hGE{wNRaAxk
z;ipSo`e3wE-2KpYJgJ-5Xby`*b2&Y<exW<*;dH88B$XnrBoaj-w-?Q>1;VcLld)^9
zi(R{jZtp^2S9c0_eX|YhYSgVH<5~g@49o}J>^)A{`-x=iz1qdzslwiCguTb7VDC{{
zr(iUmELkF6f$Qt}Ef%!&2f15S%mpoH1Jm4_k`a1~i_imv&|8Jj52YaV-FkM&*2rwz
zt3`i>)XgU&AV<v1?{q=7yFYgc<WD0swq7^_%J*DQ?DpLE1<HLJ1j-LwQ0(^H{Q~8I
z4FctdE-1YsQ$G?Y4{i`BPrIO49nP-=$}_D)@t#GR2pvxS><)4ePih3Ms$tV!z0$%-
zJ6Kv5^3{U5HLO=3jvva*oHbi5&N@m%<iMyz5qsIr-)AaC!L}$M#shtf0^<MRdz`-d
zz8;FevkeZhuHXgB#?Gj{?ccjHeNZy}F+!Z_i#F3F3JBJ(J&e?-8}AW>44o0}`428g
zj|!xp2&9)%K>Ddb`k4dLA6<|h7f8PlNPkKJ=?Q`Kqyy5+E=W%aq^AYaD=8p7BaohT
zK>D)_(sKgod4cq53P}Ggkbdic^cNST7X;G(2&C6iK>EEvdeH&tbr+<U1kxV`(qB_R
zdRZX7;(+vq3sOZ)xL*}We@g-BHG%ZH1Jd7JklsKB+}Zt2ApIi+q|;!5&Hd6p9FYF$
zg0xYufwVC~+*NLLD}|a(d7_n(-lj;+kTydormNf-xyKC&ZxKju6-b+;fb@2O)Zu`%
zsSDB;0%=Quv{?#BM=dg3dY1#z<}OGl$<2%J7D#VN0cjhg=8D_a0qLzSNZTU=7VjMd
z(%Vu%vRvBH0qN~7NbeI!I|-zYF+tkJ0qGqsNEv~&t3cXfOptbSK-$s;X?KCNhd_E~
z3P_zu&9XGX0qI>XNP7yTE`hXF3P@RjG}!^^-7ZK|1=2Kuv~>zdw)fuK0cjf-q<sa_
zegbLR6p*GPHCNmW2c+#>kPZ|`2MMI@Q$RXIARX#}w1W%MEP*szAiXCAq{9T#TnD5b
zU676tNb>~Jds9F<N+8X5Kzg4GQnx@_D3Eqa14+iXqaBcTc0pP!kd_FfT~a`@E$%o6
zr1!fZEfq)~6iAsAkn9+@%mHav7o_C^=|q7vE(IjZr4<fHySX6s2&7(t^nnzRESHK7
zNV~fr^$Vl{fwV^oNFGviSGm#wX}k;4pg<ZDNS!Gl4GW}`9FQisAk_rYh(MZ{0+Ow`
zRSrmdx*(k_kk$yKt`v~0c%SNkG|2^N`VnTLd%8f%rhs%NQnM_b<$yHV1?dohbdErp
zk^<641k$+<NK;*q4iZS`3#4f&AgvWh7djyA<$_d42E0PKNFeQ<0+JQ)iye^maY4F7
zAYCeu_8k+X^?pbZ8>aiYAl-!#oVj$lK<b@M?~>~T8lALFN2$TxTy3n_`F64|qFc~Q
zLq*bSE5pPELF>X%+@dIF>Qxl%x>oZem1;KAv4pIU)LZSD1a?HLidDMe+}Ma@?CEUf
zCZjh$BwjMe;D{3TmHJ2NZlt{0vZZW0cndR!i{f*YDsMn#cn69NPN0Lw{Y11!8@l5i
zeslPmcuEMvkvwi^Vb>VN0w?8sm9v213hm7CZ7r*drTzh!MUf|g5*@2Og|BqaX6E%}
zV9e5GfQf?-ouhKPQ0paT^bU}jjsnfDglA-c(Sep@5=sS%SV4|M;E|ZK2Yqnrz<HQ5
z013{9jo}xdSqgPVem5k9O7S;Wp<1buu~>~5nE;zUv#x6g#$oVeLcY2x8Iuzuqg$=W
z=YqIh(Gb`hg#j%PP@RdYkt5ReKdL&3A#CerD$-yYsW1X!8rc_;ALmjZ8)8wN8dp4x
zQbhbPcVSgoypnOl0Z6iBK&m2J!o=Rsa&#X=GC0f^Dh7O^M8#fejBB(fsIWNH7*nCL
zdI%m1%x$hd$M@OZiU@X%GGd=m#ME3L*~RCCG(rZOI?gT6!6LO|b_hbvg@lzPf$;FP
z9gyRE2I)y&Ru+fIzfLvN$&L{U88Q{4c$!oCV~re&^i=cJH50h?R>%j6ZcWwVs!|OO
zcM?f$ZMfK5>MQkXF(i@aw7O|^>YgXhg;RM$<b@7ZOQ*#0{W?TqM~3Qy<Xxy#z)fLH
z0F}N_o4oNs3eKRrNM2iN1RA6dW%9lF2~MC*mR7-IHVfcnSjoaB6PZ<e2XckvU4)8C
z_{H6X>MTZw3PtLFsXNGz($|uz<zX@{FBa70ZYrhe3lNJubd`GZ<?<SGNJ03FC_sof
z)A3}Mu6ja33IV9*2XpAf2dVxFIY!W;5EM$amE;+HcvuU_5P47^gshUiDfMH_#^O+A
zv>)9kili=i|5hZ8L2;s!M)P$0w(7bY?%+Z+$~(7r57n!W2A|A^7)%zhh-}1gFn@AR
zol~IXR30N=0?y@6E)9+jX7oqu&4r(+@9OAKFOMd40;%Bd{E(Jkk!UM-h?0){9Py2O
z1MEOfa>&47wZyG64=K_@WQLgq`i_>~fzk+d2pCThBik{YFW}O^WB_b)^@Gq^SOmpN
zjRs}Tw+TDxt2*RBQB@~EY^#kF={4~PU%9Ak2RX5zJ}1{x%GYu|<ll*!tUt@X(#c$#
zbLK1}p{b1;-~00yWRPSs?pn{7bd#tI_R5o~k)Y?PFCq?ftsryWG92GjXj5Me?ZFNC
zHf_oO54L>F{>!Imr)a|2>rY?bIk0)Rm&J_y0N3<yUzz@Kh_{3ev@_DkE%`%Snc90J
z4n+2x>7h2$#9Q(Qi+1S{gbbY#Tk<nqkY)*;vjx(u6p#)RNOK*KX1gFAA&}+?q&X=d
z+4aeM2c*MXkh%rZLV+|l1*D^qn%BC=I3OMFg0w^+9V?KINCD}1fwa^CX`Tzx2?A-E
zKsqu7q~!wXL<gj!T#!}>q`W|yp8`^^Kq@#OEpS2V6G;66sXGOvl0fntkQTZil?BqE
zKw6XnQbiyQJ0KnHf>aeqHGy<Y3P`pS7<E8e?1HqPsB%^dq$MdJtwCz;1Ws{4I@Sg0
zG-SX^iqi$saVa2KE}iLsbi51F*#hYtfwVLQq>msqbLm_Mqz}3veSW6l()j}EgcOk0
z3Zx4ike0b1)scbH!3m@frGRv?K>C;i(sCE1O9j%$1=5KrAXydjWe!L=7o;l$(v<>f
z#h4&n<$#oTLApjDT`Q1!Qb4*Msad>la6syHLApsG-7Js_DInQS;8q8uq6^Z}%{S8G
zeY-&FO9APVNX=aOlpj)rKDXZm$?9`IEs%<1)#sK_YNS5*<_LXm_$aRL^eEI4VU2FP
z6}npW!Am&0q$EGoKT7wfDSZR2kM-wP6})PZZSq#y9baxO>KP+*%V2?Sl%TYx#G%+)
zW?b(msmNF#GL8x?)pF{E@h#+Exh7GPsnqQAF&kZ?IRa`6&iR#vX7%q-i4%*UUI^<7
zZKlSHX@)b0e*)#oS`>n&rnELWF^i3MpV!^2VlzIBgVmSmTveiL_yAJ7auSsw>rvn<
zoBiD<-UiS#^I2Z2p+@e?aegNiM=q!M2NZp*I8eHpF%(pUtSDk_Q3OT5S2qHX9)i5U
zpA5gnr2Yv5fv$-)>N(}!9r_TXa$=S1NRckspjYG)no1&tBdd<IBFcKH7p;8pWcE7G
z%6q70$r~1lc8_ui423Iswe4LA)vea!D{m5{TqLcf(k7ZOrglU?w1!Lh#2p!cUXR`6
z6GK4<$Tno8qr%6~HqhbA&B&h-^%zR1l@;@Y;IvP7=xX2U^ewhpbXq7`i7h?lO7BYR
zkcc&j&8t;*(Uct5$OAFlC>VWXzHfw8?yN^=u~Yp>v-;xFETyF7WH>d|DB%TwbiFJf
zCI@MK6>^(K8!N-)eJ1fn0A~UjS1CDT>Jgxqw{3)kjLID)U3|WlSv-)h77t&vgo^b?
zB=2x^aRRB-lQe%5w|Q}tJNkDu`sK?t(x4CZmPfHWR;$n^6;zw*gnRPcOJCoW=^L%m
zPxAe2$XbC-v(<DYsHnQ{6<r-Gd>u{=*BRufTj-_R%V>aH(Un<28#%-^y{C}F!OGQ@
zDz(xYK9XWR-$NHk^j9mR!|Xawy(aB?PB#<iyA|7!pm;K|Z-|-h|Io2wWwAI+PKAbw
zJtU_qkqholpwdy*Ig(y=6Dr(4fMtE9V!1%{4fXOh2qZDoqp(A4s>a}eS>v22BlxJu
z4A7Y>dw(jrpdwxQW7(_Z*6gyV3t4f&D)u%^5xUDmCrVx$XqzQZTNLp7jj`FZ6qOOZ
z&T|J$jttm)H=>w)3?^SS)0t_$!vtGR`5rpr*%5?it1=%Th1e4OYs=}bgD}T4D%J>9
z@=Djtvy2G5*5eRseFyPQSrSC5l23AFy7hr3(<-8JrZt;sVpZ}8((nv>6d^-rgerLz
zzQ-L|u)aE1BPv5W8KLAB6MW!o4bm{AQxJ-QbSl2bK{^c)lzs#uhI9r(EkZgIX&BO3
z2*p4;8{gv~or4Gr>B9&yq>mufBBXPXh9RAYPz<E=@jVXGlX6Jt0z_p<YY}P@(uGLF
zkUok~45T`~$3eOX5tvKs5MoFdBh(_Kk0A|1x&)yZNSES!9Hd_gmp+cD3~4<=Eke2s
zX&BPw2*p6U0^j2x?SBSEU@l#Ws0`^72(<|5Dx_gZS0fYy=^A{GgLLJ&6oDaKi>M6g
zI)qw;bUo5Aq#F>5fpjCj$3eOY5g5|V2r;Bv5NZ+9tw_U=ZbK*r((U*j2k8z(U`U@t
zh#`Fnp%x)omCc<<9Rukue2;^4g*adNv_QIhOpxwzK>Cae(q|EYx%4@Nm`k5WC^eV9
zfYdQu`Xau^ap_Blz>vO-5JUP3LM=l2D$=kleGQ=)NMFbII7okxe(4*C%8<T^P>Ybh
zg)|K5KM;z6^lf~PgY+FlU@m<ZA%=7>LM=l29?~$R?;{ig={|gqgY*MLU`Y2P#E>39
zs6|LWL>h+lBZOigJ&5mdkfuwE`!S+2q=yh{5t3E7KkR_?hznAAE)~yQ`cHxMXbMQS
zU;2py(qk@2KSczF^fQE*OFu`bMJ_##G+c4NKq#i-eu?jK6?cGC+!Kh(ke)=SMM#!Q
z|K))6lnc_+h`?O>6+#T@8H8Gd^eoaamwt^<440n6_c$)?DO`FUQ5n*25NZ*U<<fsU
zApO<_={kY*e+1GCDIon1(lD2PhfoZcevj{QT)IXey@;sHr9U9lB9~r58iw>ogkm85
z3E$%&y^IJfORpfrkp7HNi;!%^y^7Q^kp6=2aga{LMh!!HO(4CV0@7cRhPm_xLIzUA
ze$wCYJr0uHPx?EeGMD<tx}Wq<WEHue^s%=!o)3udALrOo;(ZOr838g<)S4XJ22Hx1
zm@@+6-HvX??I2w$DVWKUcO|K2&1T44=(ovrWVO^YIs)aCR%??7If>puX$qCW9Bz+P
z`gOWgqS%)w|ET1kY$|#1F<w2b1No?U1yY}p)iE+*-RfaO(92O)<~6iwu*+p7y)oRj
zpZPvuwMnA4C5qYpY^FdvOMEUOiwJBLO*-OynRIl8HCzzvRjE|DPZ-M6Jdxcy(lJ;%
zne25*f7okGr->I`bn2l(_e?75_+{gI`ljwXu4`Q1F#WS!Df~GdscKC+D6HANCUlvw
z{Kw`2(qYd42|J)^g-s=_P4#7+3~nZ2HF`I5z#*%hoU(UWpZz&9CR%Y7IP<7LrpaWs
zjlGe+jCG8f?I3+<afH>fq$wrmqZDH-WpW<ScH4aEnph$+*Fy?>5)4?`QVB*SxeQuw
z<)rlZps73J(^d(`FO2fZ6cL}U%q-YjE5Xv6dC=u5CvsGZ(&uB3M+zdXrz)pgZX8z8
zsSp#Lm_{dCMyqt0G*t$4&-yAv*4uVvxEgv#tK@Wfge<-*L-b(vC#n45?~oMUTEGJc
z_s^1z`fzU84~K8XRe;ntlxW-m22xMWPEqO68XL($>XnQ)GWDdXna&~Fs6;H<kh98J
z(n8ZawinKo3ISGG4|_~7_<$JN47*5ekAgMg0dJac7PSE6WhDC%K7{@FBkMiFdvGA%
zCG0TR?_}ATWwf*AK(1PV{Eqqj*-|rC^#o1*#EH^3kRxAnlt!wZ#D|?h2lB^pr)-DA
zZ>+CsBUw`@8Y^;^%Z3thz6<V#yB97wY(@s1HEoN)Fq`g6=SvMU!=>ROwj`*kd=~n?
zd<)bS456Nfl%S*D2P%UVbQikkz^ma(xzxKx9|l23jF;8oYItwsBM&NdXDM6E@_8Nl
zz9ysW?kOfoDQz{F11uzx=p5`7L0z)tz~qEF1^6sxSr_^PoYpmo+-Un|sK%$<=-6Dj
zNc*MKEAeJ&sDveiNyX)z55%?mbzGj5mF?i1$$4o`ow&%2j+D&FHr;DR<bL>~W6cQ-
zUwkGbls}K^w2$Ju51SG*BlHoEP8u09^lpL~InZr{atUsow8;*C$Un`B*gV<fHpet;
zQ$*mIABT9FwHe;2$!O%}Nrx-b=MFKMz5`J?(=BYKi8oKSL>it>z7rucos8H#c^AIN
zO((ZP1g7)d2r;Cs5o&SLy$#YZq-_z3fwUdI$3fa25g5`A2r;DhAk-qH9g&71y%(Vv
zNbke<I7mAo0z=vvA%?UILM=kFDwFpkbqu5ozQ;k@6%iQHID{C|ZV07@^Z}%ffwVim
z$3fZy5g5{VgcwpMLa8B5K<XGs6Y)I`(w>OGkh&0JNRtq12C31XHt!8)k%nbyGD0zs
z_Q&@)NYfF4A<aODAsv8Fi;xaP8isTbLNSmI#`ic#c2oBdL}f^aBGe+JnMlKsW+4;<
zX&%1EL9&~=M<OaiItrl{A<ah`hO_{o7)agt9tY_(j451k3lWteEkdY8NJk?LLplbb
z7)XopJr0uH)Lnw84Cz>eT7+~Q(lDgs5sHDd6yM_@*-hOKA}T{V0ihNlEkhcH^dW>|
zAT7uDI7oI=_e4ZxNI8UBgtP)_7*Zag7)U+%9tX*8>h>ZkLn<KDBBUbHFr+?&Vj%V7
zdmJRYsXKtE45@@ri;z5|VMr?xih)$d_c%y)Q+E(i8PX6!Ekddw4MQ46C<f9=_#Owz
zZt7MMl_Av-Y7x>1(lDe^gkm7A!uL2xc2jpXqB5kD5o!^V)tjzyKsv<*$!_YNDv(Ya
z6Qt7}kj`*HvYWbR3Z%2f1nF!Cq;p)5?56IA1=2@SKspy`SeDL1C`Ojf$M-l{vYWaW
zAS%n!T7+7Zr3;aUA$=5~7)W(|kAq}4buU6xhO`c$79m}XGz{ru2*p6U1mEKz*-hO`
z5tSi*9HAB=tw$P$bQwZ1kS@pfI7oI=_X<R1NLM1%A|$&)`Gf<~RW3;7xl}yQm##)s
zhI9==Ekd&W(zQq(!=>x+J&sFuQ}=p-bVCYAHzEyJ+)W6@K)M;<;~?2h-CGcqxpXT+
zskwBU1Jdm-NOn{A4uSN^6p%iJG|Z(t5sKl`UHBfyCA+EnX+&i%-HlLcF5TmR^cfc<
zyQ%wGf%Lf)kUozz%%v|N6vL%2;(HvI?56IQ5S6*~WrSMf(pQj%A$=907)W2k_c%y)
zQ}^qL%8<T+P>YakC-6-Nq;I((*-hR55J=xn0qHwP!(93<LWWBbo4WVndmNYSrtbF;
zmAO=k+th7TQ#3k#yAN3fH+7>f^tj&z?qIR@d_chc5TO>U@JC3){CW@}!><Uf<uC9(
zj$cb=Q3M`Zeu=2euO|>{2I;c5ZA7>7(qALvSikcm5;L~{LMR5?Q}`Z-ZINJm8c`YB
zuMld6tqG)OkcJ^Wi%<-tU*mfmq+<lqbBM~2o=2!fNWVcEhV<VE#X$NkzQ;kjLC(hi
zA4FwHFCf$+r2j!0hV(mxVj%q<-{T<NB#>T2REG2igj$6364EfFKOz(Z=}-6`2k9Px
z^fIC{q*oAX5t8ja|LlPDsteLZ0_iUT>9rJ)UPl_{(q9pZ;nExU9>=AN1=8OTmAUkH
zgj(d%KahqY{S%=WNE_Ww-{T;yls<D~L}f^uAk-qHO_7EnZH7<`q%H704$>Ndv?Zc4
zq<13JA|xwI?{Yxe$_42Rf%I;Hv~>zd+aL{dX<LM1xU?O<$8qT+0%?0hWiIW2P--r{
z#{p?a7o@cU>AeE!ePe>OlLOMuE=cPH(k=q&{V5=2kcKO6SA=3JZXCYHRor@kv>T#w
z#eD#w7AtOdq+v*VAQS^>Jif<4x=J8*A}T|gfKZE&CL#?(+7qD|NL~0I2k8cZGzn1|
zQWl{WA=yq~vIEi-7o=MS(o}&oEd`{#kcPRmH$pL7+6Uj`xOAsL+80roOZy?zBA50@
z8iq6-p%_Rr@I4OFX9dy$h{})-M5skb2O$kZIvAlCNQdBi9HcJ`q(c#vA<aanMM$%d
zh9S*HC<f9Te2;_lO@VY6qB5kp2(<|5aHL^KM<5gfX&%1ELAqBU9f_z6=_rI+gft&%
z7}5fSVjy+ndmN+(1kyr8Wk`z<Y7x@WNW+khK_~{&VtkK-^sqo$f~XAXScF=HbR5z!
zq~j5afwUCg;~@Q1Abk)~8PW*|wFqe$(lDeCAru2?Iljk1dO{$bh^P!Hhfs@<Rv-;S
z$|Dp5sR!TVAUz|HdJ&Z&6%c9>QW0qwQXfJZl5$dC|6s>VZAn~H=5%5!IA_s9WiLqP
z!VSmp+jGn6=Ki%ngelno@W!!ds<4Pd%%YNIQKTtZ*_El?t(}hSInzO#X<}2d1CW~e
zejq|IqA=uwWc5Uc2&761NHYb}EC-}v7bL4Enj?@-N&#uEKsww3sp^7c)kE_HQY{6f
zqXg1?2c!`fB&#0k7D%HhAT1I|M>`;`azV1{p~V7ebqYww3Z&y4kWO|%vg)Cw0%=VO
zNGAxSWe!NExFA{e&~kxvY6?hpHlK4qI?V;is)zCd>GTwkdXbv@rGf*}87@dxJ=7<V
z&P)MmKp>SIkj`>Jvg)C-Ksq}PQWG8VkU*+9Af4-iWYt5b2&D5;K(exQngi1LE=X2A
zbf!SMAO)nek(y=c90#PeE=X2Abgn?UFa;#brSlz-KI(#G)k7Bwr23d3)g6#7azV1{
zp^F95x)hKuL29nJOC69dc0sc0q00o)$5KGLLLgn~fOLrql2s2~Es!ov0qI(Sbe#jz
z$6b)Ddgw-hv_1tSE8aIdAYJByWYt5r38c$YK)M5|S(ZNOfOLfml2s4gC6KO60qJgm
zbdLklCtQ#|iwt-}^m78~suYmEAdtT3fONGBl2s3VMIc?10@Bw6($^i3u604O>Y;B5
zr0Y^Z`nEv&jsw#5E=X2A^gV%eLkdXu38Wu5Al>MK^Z+toS^A+sx+w*u2L;lP9guEz
zL9*(hM+DL>DIh&6kbdHTbgK)JRS*44Al;S%(&GZ@7Y<0byC6M*47lQ+6i9cZfb^6=
zdfEZ$lP*Zl2&88P(x*~DdQKoc?}Mb(Lo3)Tr8#V_R75!CNy?#B)8yno-3L@H4dVu`
z&|h1;HF_Y*Xisiwy!hu%7pLw*1g<;|u|j8Mv<c7!-re0BZ?x9Cho0U#vTN}^!@pqy
zRNrC{7n1cWC#mbb@6eZS@4JZk8I%yR2fBkkgsv$1oD1ng@MF$MImAdmk3`K>eqp2+
zc(A$Y=FzxYm=>X0`+^I~7ZHJ>aEPJYZkg8UMHpj!#g*y9U>0ZkRYc`XD>l<4sx_{O
z-|_L{9LZ&kyiN?#5&G3x{Kzf{+^<PKC&(XlM6@}|3AU&{$t!D_ZptcFBjV088c~w@
znJShqUA5SkOA1CEjZPeW;>4A7l)AqbkhC9wZy`yGF!OrARTIaElbRn&P5cO<u}NA4
zlm}f<ek@RaEKnZWAW$B5K{-&YxgHTH|G7b+JnDjSut52VKzVF~K>4W)$|zbWx7wcx
zl%H=9D380K`~nde$}a`V6Jv)G(U(2xf^x*^2Fia4l+~^DWfyqdm+hrJrIVNF8ae-F
zmm9iE(_d;?9mwi_Yk8nccm~CXx`YUVe(fUYIYeOIJ}(5Fnu4GoXnzyr(H!>}u;U<H
z0b1=ZjwJ5<TI?%Z>4r5kW#DMW<p7ian2gCUyO{inF!|5I<d39aGI`@D6w5*b|EgsN
z`f|8Fnl2z!jLsf0caHka1EU%b$&<LGYeqXcuNWN}9v#7@xm<w(X8b)FGydsf#zuD-
ziQ5<<o(`-{!HhGSF#|VH<f_Gfa>jz2YK19F=>0MBx~;r<J1B9t#&;lE+}7q6!kaCH
zH|tXH=0@gC#Jq;7PC1i-*hFMGr(s$Zk|Oc~qh?e7z9Ar5J0@e_`&{hXN!Yiuu<x=I
z?3;@jm^2ss$hszk@cW~0k%3zvU-R*<0s>w!?3Rqq-CcC<A#{!xI<HPaCr_V~)2Zsr
zyH4oYGZ{UTT=Zmxp2<Sb^(p9C(rUdWmsIZ-p?be$R8MzNJwvEIK&ZYo1=ZUt-*li|
zM%><Va0M~D5wlwC4@~zyP3yC@kxSqUya%v$`x3hJfCY&o)fXuhO;`4$g)RMaHLa-a
zvNTICgDYg*=X<~RI>j2ex4x5}1sQi$*(zPflG6*d^=hW_%w2@%A|T_?IttYffg>lq
zQqy;k)Jk-{3Ef;y7mz@EL5`Jh@duoj(f!XuxOs-N9VAa?*^c#k4b-X7tYb7@==NZg
zS)UY#&`j258vB~{%tGP<P<qp`)K`lwUC{INdf;9!bfYjAG^3@0%yg0O{ZhqIE%9NT
z&^5zK*>NaAxnNXLt3ibB5FQH%%mITSLKh1+1rfS6m_Z!U8YddA5DOx)n<?{A9Zivh
zE(>NPvsy!9F9WkTj0M4jZW3+^CUi|OgPGYH7(MONPB3|z9>fy5LbxfG(EY%SW&hT&
zu=`NHd@KkdbY*Z;2%-Cc8NxxWLGWD_guBUtI70UXH^mXU1ekHOD6I`wlLe84t_N<4
zqzqZ+g&g1g%SaAxjU(a9#ezUWmjX8h61wr1f$Y~B5c0XG?)M5Junm{*0B(vPbipqp
zXi+}xg<?Snp*w$@LI_>x%Me<W&(Q5wK^&G9e9>=H9HCo$8OOZVD#O=K!dGeqF@<jW
zZHg&$jW1(5yfsWs@9_#E3SIQu6j5E;&P50@qB*T0@?9F1v)+w^a6*^-HiZ+q!I$B*
z+SfPUWfsI@{ZjDlzD==&uI^<lEedkzHne6q_)6cVI70XJGL9B|m62j?Bx1-3-Qe34
zLFjtkG@MYQo?QAD<b-X~d~I)29HBdT8OQ9_gf;ROz#y8?jlGR%9#Rumd(xh-;{_UT
z6S(Dsnp01EX&!2~&FJ0tDAHP^F7fAF^sc9!zO~-v$mO$WivC&ZI<-RBa{hw3wDqIa
z4thzj-KQxs7d{b%W7%Tn+&@BQfjPH^Zug}gK@?8CnT6izJw~rPyFK2jcnJFyKcyEV
z>(5?qzC4VVkKplm;EVSJG6V;@yKQ{$DZG7}9_nq8*93`~6PT*o?UUaiQx<}~V}&5V
zHC`;RE+<&)zeQ#+>gjGiEoP#5qVa+hJ*|bJd-8NI>QEtA@bCUV_Pzv6uBuu$VWtHc
zL_nE}$dH840b~pbkOV?PNGE2HNQCO@uIlbO>F#Q(x)UOkqD<O?LO>J|g^POOB6?AA
zMnx1SaN;70aMg<<$~=q6`~S7~Ugw<Mr|O)lPVRg6elH*2q52Ga{d?`Tr?u8zD+A#t
z_>K^MQVBQSBgN}%3FqxPI1b?HbO6|-3iW4|`t&BMJ-lwnTVp<-PBOADd{L=39`@pO
zc58m>q+%6z*5YXQGZJQd$KOd$>ideL9BdKxj#gyXmOKG{+>|_#zog%7<&_+4%17Bo
zjTr1LweYR9@V2mU?bP>6uf-+WL7V8d!S*cz+(`kxQ2}mK2ViXUXa?LKW5Df&JB9iq
zZ;XeN6yjSH;_E|*MwPMe7!q}d^;qFkwD74~`1XF`(^@S2AT9i0EquqYaNGLL49~$4
zEkZe3q0Ce$JB3gT6OVC)2Y;kIQc2usevC14juM!w1a|fboZtu;pOwsoFVMnI(!zK3
z3qQHV!k1{_Z_~ox>=(YQ#lla~!dGhHyZeQo>Ix73)->NwR|0t@uxCg>N0=vB@>B5<
zrXzTocc0;<8((vAV3q+vkAu`?O9cvNwZh3Iz<HiDU-03)$h$A`Qi;Q<wg_j9!da_u
zCMCdmg*5-c%Q{QgEvjDS&42R3JxaKI>IekvYkYE^ePVfcK1sDOSzO5PWRAOtzhoxb
zCj{#dZNePAo8lsKMZt94541>gK-2tPrZo2(OLGg-+>#f^1xxpe7U_P3-$`AsR=Shx
z&}~bg`|<2_W6St2o&>sMm+`?5FLQY0m+<}&;<0P@ALF^L@=y3`;}!h9d_~Hb7S^eS
z;l9b$xLwiwTI=$F;%Fa(qa*mp=J6dtM=<K`5juj0QNlR(2NqH2r}Bs*I%p#yTE8O0
z(!!b;?pfwq{V9iQwDIO)PN7z^cAfTozSdtEuI2NwrEKsFiwbuw7YDxp6*E^}YV0__
zQnJrEvcdEG72PPl$Q##S3SRh?;N$$?H>}VFU&ZdSAbZn`N31Lx8Nb+URd1)&E<wgS
zFTj22bMi~%^F;RlDA_K))L6Ppd_}%Kv`!lahx0)%-ir>H+<^v1@JZl3Ig&Sz;$^^l
zGJ`jd=4HrxGLtuF@lx}i+`#eT7(Q9!Jvo*)kK<*X_hdG2&f(>3@5x-=Jf4^HyeChx
zJ}2<W`QDRxym=xo7uYA`*<F7CWtesw<u7u6^st6|y``{HE@w)#Qng(Y6m?fdkRYiV
z7t~8xS}3TO#AL?y5NR`!nGY%L#wN!oZ5n~&G%wFU^IK$`&|I%H56WEMzM%wL(p4GA
zhT+IRs$?4v2GOc&zPEUNugO68TD~Lo{e%*3tUZYmhIz3(m=|4MoV0528fZ^GrKGP{
z(v58sQPM8)5QsyG5|^&}J9;ZwH!A5*E9vaDuJs|q6H?$6i$n2Rf0l)%2=z@$`er4a
zU8s9%KFwP)(7aV?eqL!dHi$(zJ=9<5E?LJ#C~S+RHOTW9t9UBEWsDUmyG;Qcp1D_1
za|2Ute}7@P5^w0>OBre}!?+N?y#?Y%QJttL`)UT_GE)fguPgEF<7WvLr(`AFH#5+d
zxkzZ=p|rCPDw>%r*3`jwG7!F#?+D@VDdFtRZ5;O2R*?g*<mpDgpMf}6mWKG<O1wR@
ztO@aWjenGZs7&Ef<DV$e>|<GXWdOdW=wl@$747zBt$Q+%mpNa^|4hlx%UtKsgqZKI
z^p>%zL@u>tFu%wE=2v`2z}%-`j>-%M30M26p3e6nqh5cr1N*O)_5(`0v5hV|(y{uz
z2={|Y4T(VE67?=$0JF?%4=RS=DTbpn*B|NmdIu%tDTlm~fk96k9%h*-bIk7*!5<XC
z!I=>-4PjxZbY{F4WU2WhO7~AnH~U~my4@p1IMxn$^J*5_e^J_xDedf2y3j^8!<I-t
zuB4w((%CzIv6Ant7K-^ABp;r5&2#-JZn>ntJ+0)k4|uiV!f=zb`7=uJIVG5Vjck+g
z#_E~6fOt_^{F0K)E?G8FZ64?3yx#P8mX^|+{-LyARoab>ZP76WTY{S8?u9Qzg0=6p
z3;-rx=o)wv{*pad_CbxJJv=gqzSURiUOiMn`d^O;S-RKeieU@IkiC2L7rKyml%9}r
z(y8sfwUTX9ve~;9(tP9e1Jf8|WUn<zdRry^dL^B`+0CY?p8rezdRwK{4oZASM?9F%
zt*P}Tu=e<!@X%<;tbTA8zTJZQcQ>WCyV5%;1HD{}^!8DD`zpPYGtisTBE18ZUc1uk
z2<fS%Vys${2@J=tggip&AF1@0g!IQ@LY}21AEPBN4NKOVQN0>0$AgHKw<B1=H{Z_7
z9DE(xU|LiD&Q~l86w4_ISY$i+jzlb{^7TbAEWr{MMxhQ?@o7`0!Q~2Rg+h8~2r0_=
z;Pj?TKUGU#rKRV+(u33ZNZ=QmCQwiUT}q%ULcop#vdb*ue#9(en=mLfO|+~;10~v1
zhiD(~_v1Cnjva(5{7NcZZkl#YX%8#yU@Yx^-WtFqQsLlCr90R(-LsYMIZC%OmhKt6
zRrTn;$I)%lUU`AiyijS@A~Y|h$Y0!I;iFpk2et5#Na0p;M8~?;tI&rX&EQOaVLRNL
z-=4v^Vdt+amHJgmeO*XhM<wCsS&qE3GM>#h&*9}`_&OYwf@>YHCOQ2ng>${aIX8sk
z=)Rk`-h)e2sy^*#2Iuo}Q=`vKO7~`^``)p1FOXXv-CG^qChK~e()^;*yeOoptQVq}
zIHHaP@8_Ez@b8ZD?g!o7mh1^uFz!%v-&S;&vUIz!q0F~7`P`)?e_u=fP^4rl@T_)6
za5r8FN37Ne-lGKXRf3m?1eMWa*A$eqs6$gna34zO2)GarrM)#&zg1KZDysFFP{lDg
zf#_LIWPd<e<3;wUBKwOXyQ&Ff&ie0=2^WJWHvqCtst`cgK|im^u1-MqG2Xj|mlrKn
zhebNm$NA*%-ji#2^Ao(h>OGloQMAoXNy(0}!RC0im1X_b{7x#?#$RGLy^dH+Z&MMy
z8<yCPU>m%{Ms}6rOC7;>-d!@GBiPZqYr54Ny}PV@N3g5CJDzc4Pn2P5vKN0zO+Hn(
zCVP80OilLj?wXqH@7*;und;p&H964UwazvN;m=l@<xqYn&2kujNo8&bE93YtNFq~F
zYXibeB{oZm-53&684zv=HAi0tu3$F48gJJ>Q8CO{44(-xL|5TW&UTBm<PI(QUn3>k
zQGW>@MmNf>vGVOo>=Y$-b4W~C`3|KPV`YrFVdJ`YI>&q^LMLC1r|a}7>awD~B}DDW
zuU7JA8Q&U<ZhxN>?K70vtsyaG(b!3GEJL{J=m^%Lobk%vIm99z?p#Ing%DLe76j+>
zZSG{6l*$X0-bG69wve83;Y+;tWnM19d(pA?cHaC7FQeX*ukz;Cc)84b@^#+)1}~R;
zPvoG&H~HjB?}_+3+`%VT$DY)gh_1y;W^4Cte$&)q_*48=8sU2Wl2PwFEX6MUY*)RL
zx4wr<WaJAzt#t2dn(j?X_hzO0{jqfK=B*!kbZ=F<KWdupZA$lxO83WO=^jik&Yya8
zzoK;SX`1dgl<qf`?!6JZvaP(s5ol7f@6^J-r-lDKQn+ozU-H(kcyX&Xs~fnxMY=yx
zx<6IA_l0z|M{!&U?#F}JU@JWSwYzI)?_WA7!2>L)DOtW>;rv?R{H6|^_)4Ep;&&cG
z3I5tY_`L(yWcxg#!2YDb9t?pQhuueA;lV>q*Zglv;Bh7JuutGgCGh*E2|TL=o>KyU
z2nlFQ{}=B)!pjSIFFJPqi8ueuOMF8Wyy9Rr$<)^r(u9jqs*GHZvi$hSC5Rs5{Y~&1
z5shai{i)8RTPe}MhD76hc$^Q~ELAt1ZHvD<g4f%-<5@!NL<BN)zmdO$0Z)Wj#<6~T
zvzEM@mi$y$a;%me!5(-g+R{1v-4X2V-Q7p-#@pVmW`bu}c2iw_DoQhTKY+ia?$0J5
zj8k=OK$++Wrdx!uHVB??8p~md<#5IFLIReVy!R3>M_S4b&&#~|cV3S6p1i`F|KQ~q
z`^1iuvq?(Sxv%nRQ_MSo--_VO<1b;}KUs?DIoiCU&z-<sXta3+>_mDApJXAqY6#d(
z_+*iNV!<vU83DT~pEd<{IlmRKEBH&mZpKnPu%ag10uN$fSFl=J`ggbD_gnL_$^vy&
zkjl03$?4t`sod-Mq+p-e$`uJqDz^=vHdVPkek+wL^Osa^TbANgZZ)39=BK_pJAy&n
zWW!h-XRw48mc=9dPB5(DFTt=~0){j3G=?EKi+6&vagD$@kA<`V<GuV&V7!mN1jY^t
zFk~(920Vzhv#cd{^6yGJztO)dZM-v!9*sl4%s~%!;nSvC@(P8&Ug7V`QsT3b;C!<~
zWS3u8JGxEIfFD<y*DB53#?st_xAx@4jlP!dr&^?YqtgAf(%mbh>&B11slnd7`&qmc
z9Y7}W=398V*?Y1NZ|=*>=e#HT@#g-#e8GD%nK!5K@<s2-RNg#*m)pH3(|GehUcTl%
z5rb?ypM2AMGMzUM;^o`+iS2iH;?GtD{s;U{1paRR5`jOM<;NxFM|fzu-FVh5KT~2q
zS7L7siMb{V4&%MUdHE&Yi?-Skym=%qZu4Nq%sPA0-;j)qibwIArux!D3ie?IJA<Wo
zqoNG?Gw~oc6v&W2%fBl_!7(i45wBu0930Chk9tpJ$Ulxx{%W7tCVzskq;j+Qw5iHH
z!*8Y8pXD!U_BkxYt6b8MFZHw-UUnFo95h~4`u|k=$A|RIUSz^003;Qhm!*QS19h*R
zh;O5Zjkt$L2stl|JS#r>mWKY1#_`khNs|Lmlgt6ANybaeMv1%B#NT)w9lh*{EE|0z
z_j$+ZRjiIV-84GcC*FmlTrQMfd*ob~-7&2EIS4FJCsY%)uQNie(+2iiTORJq^_JI^
z26H&?GQt3UOjfz1P=&7*4&Bt+4vhS3Gp^M<8IzllTNH~GpR05QrEY}(%i-VL%HS|g
z@EE#UaIRW9W29Uy6?4OtTz{p2|09^7lnI*RrEiNq-V21<a^cqR9>rH9!4D-|2Wk*u
z$ZS{pW=Rc|s_=Fmhz3Axv$a?}G>o9Rg?_`6C6yYSg86rh)h0$gz{KBPSX#HXQZ2?_
zMf^ud;=+AJq85E=lg3BgMt6!-59j^plg#6O(v25{-AD&ihP02bniv`-eun|xqj#D)
z|Bo${;6K{*h$U$=Q9Poe2HYRrIa;q!B2CRk-|BajXv4SH`_!6^P4WjpIn$t=8TIar
zBhtVzqi>8Bt8#1srJtB^XEGySj>R*};F(n)j|Sf#-8G5_Ucs7~!9q&+e07lGACE<K
zj6rowT~wo+F6fwN#<tPzqOjz)jI7f#x0rkr0WrGuk?x=E8}kgFc?o##LEpjH{^*A=
zta4Ht-75+w;<9gCFL`o6y2`?Yt}@p8z_T*Zj?po?t(^RUN45zLWFDFsadgG<zA}=*
zj0|>jL_lo!)i$!Kh;t4n7cfL3v=0stRC5J0jFty+#zP*3rwe2N6*@R9iNLEOxyuX>
z3K{#C<;ye5mp8@u!f`rphf&LsMPoTKx_zzGkKv!E+RWI!KE{oN^V(xL7GxQK7GQpi
zvGT|`D|r^BfN<r)84E=cGndwss_Vk=*~+b=;nssqaclGqa0mupq)oKONn(}snuyin
zu}loc9$^oeWF_aZlR0l1S#Z^xTy8J5Gbr!C(D3v#Mi85gLS>&$k=i?1#6*aht~Ntv
zY5T^E-0a-(rqN=PA188#0(N9DX?K-m-YykU&QfuVjugeJLR=?z;*?!?U;D^#x!+`7
zXlLofU8;6YF{Gnq+wU-oMjK5V=wZ`9FE-aeugiD$m*Bq52=;dP*(D~|(Z|vtZwnj5
zs#yB2G?J#f8R=dx8$7Qzhi5xv0>BB4D$Z%(j0j_-${4g;43qzg;98au*K2}n(Pk5y
z1GaOXX(ddYMcdN(;R+7W2);=%e0AOwm@7h<x}q2}Opd<J5EA)~j}W7q7y5_#3hO&Y
zH$ABRkoA%M+~}4JdY>;B@oeHj(I>&HAHZnf-5lM)h9V!p(c=EnU@yXTb&PIu@Zr%f
z!<PzYMqWDf$k<EUmj`=5ITC~t#Nw~)&|AVm^#Xfb-liJ@v9r#JmJ>)X9^I^?GFXb;
zm%%LIz67oqdw+CCTQi&7Ln4uNMBcMw#H1M*0t*AV2(M?@A;Sz3H9J)-+G3&+nhTl~
znq3owroVy%NLFn2Nf8_St_aR4Sp+9|ua5h+bB<12NO_30M(}>V<;g>$kOUu=H$ABc
zt~7V!Vi0`Plo}^>t$B6hv~+Njlpab<(=B_EZP`hiG&LLq8tr6shiDthI?zu4@5t0(
zW9KqI9LvfBut{CwZ=YnkDwthi%0?8MCT@KfaUZHPmmR}qBu;4yWDVseG|o6?!G)>K
z_T!RG+)`_oxz9|rT#uQqC-WvJueIg=eoj||oD+LW*wf-%R0TU+q#P?oOC6R;&E5w^
zEkRgnBIZYv^<Y;TPpqOfGt0Y0`)*?TEcSmXy6X|E?7;sSL2GpG^uEeKX*xspm#WjT
zy<9!rggA8nt5v&g9LDJ`X~6VqWZ&FAv~GRy39Oxi>v*{a?bi`}Qr6OD4PXDWk%q}y
zB>8}7n&g)xvXNglm`iZeVn5YE@Lx<_a6NJS0XTvi@T(*E7yf-CE?DV%83&}$NBEH|
zG$GmAqiKbD*#)G~zl4Qm$-tMCNWi6MOn~%Q8YC~*fI#|d3y{3D0u3PjEe(>Fc|agN
z-U1|A0c0x|NPd9^ke*0`)WM=e#1`|HKzgzTNM4qK29TaggXAS55J*qA0Le=>&;Zgi
zX^^~p1On;V79e@a2O2<nE)7zrHe5j)?)erVd3gvLKzbn!l9#GLxb$KRki3Kh4IsUg
z2Fc4=Adp^e0g{)RpaG=6r$MS}!_~CmUTFbRI0#6tf(DTOkp{_2XCPd9wFO9C?t%u8
z{+R~JOK%{MUTXnT$R$Z*&;Zf|2Ka6?aeA2!1kyx)WG2qcbDWpOpaG;!(ja+>4g}Js
zEkN?p7&L&iSsElS(}6(Ryah;JDuV`)wn&3?x%Rj#w8w4P0wkJE&9o~S4H`h&Dh-mC
z^gy_@^|&BOa)Snt+R`9-`40rr>so;1Wjkm9X`3`iUP1(cv~3HJ!a+b1A2fjU`ZP#h
zZUlj}T?>%B><13gtz#GT+owVD5+n$uP3Se1eO#jDI(9H2*pXE4!j1s}oh>+%&TZul
zTThtKc6_M^kqC1IYcHiH6Zt<wYvD+Z+0SUl<gT*2zksx8Od$%DPz5S^>QBv0hAk13
zXpNd^7y&ERI+zn>jTl=|PGLBQs1w{&BJPG5O7US&xp$;W5Ye4$A=h6R>>WWq^_U42
z=p=6$Hpj^QBfHH4e-S2h<4}#m#_KRsgTyRu4nIDoZ5E`hL6)U4>_(>2<i+IDj3T+z
zQ$TLV9D<UKlyWeDObU~FO#;C&$;Lpnt5WH=mEcdPbg|Jcmb;;GLxtA%l>i-jW5Jv=
z#R-?p_Pyv+oLA_Vx`l+pY+{+}C>#Sh;TUY$-Du`@#DVq+3)P0|E+A@3&1ehIrZv1R
z%5&S0(F9_d$U9*Wg=4Nh!*i3@lnc>jt>ya5UDZN$-4rldtO1h0B@@wTosB}onG>X*
zIp+MzO$IsW+T{<@h~zqw=b6RPGS?7d0J{-WS+dNK!W;H}Nva4FJT&TOmWS8nVE6EW
zj7bWHIzv}ugo@HMTtdvQs8q3S@l|<XWFR*<GSG#mlxoH#rO=hm6b(L1Oh)RcMk<!N
zMtXbs6o0_UfgvUJ3uFIZ(33-NLRv|bu4TCnE$So)$S{;MkZsfRhiN1<0KiZ`rC4jV
zh%j^m)FV<_6E)GCWN3-Z{kt?+=z^Gq-%@i2wuI@$vM@vkT_avKIHZ&YyJ6K5J3$V*
zE*VY4^eSR!#*upMB|;RoSs4R2RNJd1Gp4j&4HjJA04IgzHFZ%FJdiIb<3^1y1!F|O
zd{PVh2!yPbtBrK`(Qv#Bp+-w5E?EvnMPGzj3`Qu(S(aYu-mOm#r7zcs4kK&@y{Wr@
zq==EP290B`P%Jvvm&}u$46tUlg`vEPKdaNUTAf}-x6Wy8osj{@Pp;NEE!P<aDMT;L
zqgUmr00rmx5QaBO7XqnOMux;@Xl|DVDtX&YVC&5rw)2D1nP4Bn?9uEKdC=DBW$41Y
zR33vmjk9@3??^Y|NtHS+%aBsFzZg0Kpuf9mpHp#c`l?AuC}>T&SjzR4`iF7|hgHM;
z4^IM&Iweds;abATt@A0$F=bA!4`Uss-8M(UPNW<}vFs>>A9HSYJ;f9RXRv|{z;JC6
zFEPI2%5j(?%?!gCoT7|S7P_E2Z;QhVo3aqA3fL0=*KP*Sz4xZfGwCdb#;$c76^GUp
zE8V%h_g*hqcP~!Nx_gN@&;@_ue%h@(6I97NrL8<ACzdD=<(I5Hr-m!f$w9($cGt8*
zy`))E=w@M|S+?0+&W@hOhRKjkJU1&x3*<D|1tpN2^nKjizt2ovJqXjg|FqmxdavPs
zQ>{5OVr2*>ja@7U2`cM-(yFemQ-S2%lB(~^kFu}iMT;IuvgMjL0nUDDaO8NaG_U00
z5;*&}0OxPCKA32L4d6^pgYyK75jc{aOW;gt0nXEz;7m<{6U)^lSPp0b%k!DAOiRJy
zC-D+E2R0AK#m{rCl+b<K)1Y`+y#&e@&3E7UZuB5hjqgU2#+F0Uz<G(m1l(30xY5zJ
zxeV!8%awb~LSAMG&Q~(j>tuz+@G!SA@cAi}Ygik*X-B+=Ijcw!gOy?XCf$SL|FRC_
z$;s0%5s&jYrw+}=sl(Ga<z+b&PPKWQTKS7jpbD5<En&rkdGdI>EHNI>xgN~z+?$uh
znGahm)ZREpMc%bO*76p%gV=VP3DUSjoh1uN+Lb4<4JOUVEq}+76Q+fC;Y&SSz(Tc#
zN2~NUKL|hNU;H;|b!lDxV7&PD*(a|!A%AfD!C^t>4g1a^k#`Om`%b%UH22Lp@mEp_
zACiIaA??i*j^a8^cyUf!+l(dZvOc3tZcd%b7RjTLrcT91gMXN{1nV<9(p)*LjH#O8
z?k&3NjI3eMqWo&A%#Bi<TUmug*75H)-<i3z&X-BP-l&l}*jd`}u<Vf%;_??FKzl|0
zZHO*6x|idjVPNCc!7<so<Z)?T(#z{7U2+?*OOB4%s(Ng)OpEk9+e@PzaHF#9O^#)a
znKeHfvrbB5)+FlcWw9U`28CI#_n5WvXpUdzUn0Nk@D(ZuWB#+D%VxaKCy!MW_(}E0
zD?W}F!#n(!W!%#DQpPU67ab?~H~)EK7;j|)lQ@>~YXKdCNgU7kUPdtfYi4|70T|=>
zzuCE=b;Esh@5Oj#U4#5)<9G%ouTkt`pT^l7+sTb7_3^p+<ZQCtktW+-f=iL@?LFBZ
zUE(~EA;kUVnp9iGFE%1d2iOq9V!XCXvekBJT5Y|In^M~y{n|dlN+eBo+_lx2U;bCl
zOxrcur>@M_B&Vh|iI;g)nq()xNmiMG)t|6AUANn<wsc)~-wi)STUiXVo(AvCR{7J@
zDnCbO7Rib#mEYN~e0O@~-T8=(&jfS6H0aF6fUYzKc$r&;0lWGPcz-ej=3zAnvjog;
zb{sA_wToj1j7DW>>^N9gEB$7|6?4&l8=s2V_|%ieCoc)D@afGypH55Wlk^f!P@=&4
zZ>xG=wyFndRrgZlO4WDwtG=>9)w$i<__Ypm^9KFdYCf1&b1!$V)O=6B=3i}FR#;<<
zbaoL!qgf#r)Ubk~S&;K{(rm_OV#by!4Vs8>@j^>O)FLU?MPq=C?%te5jVT1j=p#17
zB4BhM1u2@;X>LbP8x_^6tmOXdqw3S)Y`kBS#(OVOvG6|U@qXog@QMcedDtjNGpAJ`
z7<EO2Q}vo)P*{)gE6)1RAQ~2lHZ`Z35)1by#;H@y%WfcSxTl?+8!WALD_FPQ^)~FL
zIC|jQI>uIg8>Fpw9u(^$%ocF<m!D|iLl23y?%CN|_uRDBU8D1^<b9Ubo#eOf*Q9la
z`pf2gf$LppF8C%mOSPOM<#9l0HfR>C*VMy!2YI93Rg$JamYV|X)YkYk3#YN-5&%QY
zP^Bto41H%0+L2tx6YCHSQ!E__q*1)x=hUgj7;52<TN1S9QBmN!3xrt54MxW@HA=<~
z^7!8dk+_-KC^K@>A`mz<1thR<q^GAJmRp+6adUZf1U8)pa(Q7kxx6?{F1-}rBA5Gk
zayfcO+wl;Ouq!5x2uUNp7$Zw&3&U+(D08zT+dAxeVeYhIr`+eHQY2YX#d>2~*e)BW
zWwy0YX9%v?5HOmJ)t9ER+RJP%tlrOO^{H)ocnZTWqC2e1Ns1;q2sGfq2I9h-$Z*)c
z#q1sv#u!H~&&HAUX&mu#t_w#d`yA;>;mCYr=$r>br<+n@Q-z4i2I;*SxpVbc?pWtT
z25@Ox;6-pbK5lprGXh)M=r%*wihAQbhd)_oM+e%_moXZ_g@fi?=dwr|&hTG@wkaWs
ziyc`(*$lUK<12))9-5ZZ`D9YW+&oE_zO(E4DB6<U_NhGwmrS}5ivLxtFeU!qs0)dV
zE?EVqp{v5qln7^*OmJ2BSXxK<;bl%AEWf0q91?bvL|27t(+Uk;74Bv6Qs}g>(9Et1
z4V5-G5JkG}^Tsr=zg7-BpkUiu0K2=K4aGvp^tx#fd>-yJnT%tD+%4IV-<pP858nol
zDDs0gD)QSCkSBOhd^HXEJ<FWbJ*CLM#*gYbbX*+VN{@g`d35vh_;pDj4js?%>zD_H
z84pLh+Au<lbQxc2%>O+2yQDl+LdwgM)ucevEoW-!$<_L>-mrIG!zwlCo!?61Q>W&F
zlwZ=3PULxN!>7dF`JJ>v!`@k9@lxoa&5O;-Yd4)RL1cbLZp9!x+g4#YJy@fi8g6p!
z@&xPnyb0{Fq*_|eeMYIci07~2*D6sz1_#^m_eg)~G%QVxyAem@8QheFWw~{_Oh)Ze
z4oVJZBKN6GST1vS7$PeUj4!pv-jskNb8W)!5QtE9dw^(ZN%2asUkbmZpKHR0mP>U$
zpO!lwHkws3`x)kVAyG);w2+8K{E0$~am|DTGG2#L7D7|Nojq)d<t}ItyGK~EG#%Dz
z<>5%tbCD>SU1im()6jTPI*nU8g4oMDomhCRWN86J)G&-V0bB)kTCyN*KtX;*i?^wq
zAwACAzUpLb4dTu3%5;gJjX0I6tRGh)kpk({IKY@nC~^4+6eFulyU6`^(vYW4md)+{
zDpKqAl_=BI_LB<(UB$x5OUc{GM!b2QG7avDI9wqfiU~=uBNYhQZfGwr4c01HA4UIS
zgNRU>hgyT!r1JoW1#h}c7`lI(HhO$Ohon&*Js#vo^@i?Cl85d~`E_jQPI3+WT^by^
zG#Y#MhZN33Ex@@v6P$<B;7EQ8F|V&zIRDuKoU1ay`F$Fkt67Y|`KZGALkn=O$pq(*
zX>hJpIG<2B|J4E<J-61OvulDVKavLLQwryLh4ZK8;rQNx2~hr=2IWSD@@a+gX!B5l
z&+_ZkUiFtWI5#Prn-$JuEx@@Y6P&-M!MRo8d|u)Ft$8>Z8tU;hC`WRkBt7ajh4Mu6
zP=YUI;>?q2aAqi++ZE1JEx`F|COA*0!I4A&!kw=xoM&2q^UX|fo=t<pKw74;?oc?-
zwE*WknczI12Io!|Bi#9(!g-+uIN#3%=fyNQ5^rAM+^uk4Y5~rVGQoK{4bE&8<)0{=
zzqbJAo=k9FNrN*_;rvYD{G$apzsLmV)igN2VlmQK_bHrzHV-Gm5cpaelwT{92NcSL
z&jDVOE5K~C_(Yc6DE!2O3g>qUXOkA-Jj}0Cr_N2&;QU_U{6XPt)&iVt+}S)0&Laxv
zPYP#?jRt4SG&p}zIFBiutu`8*t<&H<u5g}EIBgpZ&g;_PJT=#K@~0KfHqFDy(8;$=
zgYt|*c}}6cexpFyE)B|y+DI=cl<k{`68xQCr*`li(%}3<;k>GFc5DI8YnkA@Aq~z%
zr1>=y!zTPC+Ps}wfRk;!d1D%!%@xiT3TNl$;baiuUDBXztx(z&%B~v)%A3-lY^zXS
zuTb7Rekcj&&~{6MvV%g|QK8J|S|>7luPoA>IJ<~^0thc+^kOuc)3Db)QmxWk-1$zg
z>O421v-7oSou{>j5u~*!#C(PyTPK4bq<tO+fS#TDt__`nQYGw%@$RAZzmWQAdv1YP
z!JUDzr(A_=csJCd(~O6TaUK{hn<HHVB@Oc+Ay^p4rF9!Zqf>0bPHd&`bTC@wcX18&
zVgq}6yaYMeJ53IDQ#sgO<=`Y=4i4l&kfk`*4zIk3D5&2=v%(-}(yJyCID)IUw;{@f
z2$d5u7-e`au%|LISfm9Fhu@JFq?1b1PK4^fS;Ij@^QczQ#&{J51sw23_z;)^Ol0x@
z$s#rcj}$V)$X=3&8NR?<M&5Hh(A60dsAs@vRZr_$P-?S3JJwgFnZb1&?-c(3HwpB<
z*#vt3G=bhn1$ti<=#zbc{s0HI4Jgq6OGMAK{dlD>;@q&oTw}(Je+Og}mjlzp<v<me
zb`_TnPh4<V;+75?n?EO4-vlk;6t)%f7yTdCz}YNja(QbusE4INJwl-#sZf_R4>f)a
z>4<FLj!FYJOTitZ;FdNIZgi&go50CY<3Ldpvn-6B<ZZ3HT9yE45z`nfrOsv<WBak$
zYBoEqX7jaX3$$jZcr_b6qwSPx!HnC)jq@2WlfoLuLE6li5{F1S82xHT2F1}8uZ;{0
zZlpEC=!}?odq!?9#(2WW%*kLC>nZDv6SHx1K^ixgD>qjtH{a=TGeduK=a)`bfD5w$
zT$BdzR0X(70p>k`E9aZ}!z@cUS&KQD4a?a!$+B+~nP44!3S+>d^|^#GKIo8=o?NFC
z;oJo}bHxf=R=Ef&(-)r+-*cULCY{LV86F6Wp1~p_2)Gg^Bhb!}jR1zYM2OtiHG)4)
z&^7`CiNO*>;V_z8t^&~2wOBT_w=Hx2F<e$LLQ<8Hf+P+V;`j7b;340WTUQy$tt|ip
zmjC5Jrff4Ynr!U6u2NrNO_`w!{eM~A+Zqk7uCWW+*o(6@_S@1LyP%ETrH$R?HFgGR
z9NjPO>X9e`7{bi>6qK|cE3(z&l(c%3wH|@iqsObq${+bc8;Qe~F18z1Omw9#G5MU-
z*UJI~)<Nc{(zAk@@TP{l!Q6r#hInFdD=b;%5P^ztr<@XtwoE7XcRko}aMK{0L4S_v
zgb#%slcnz!y1Ps4Rc=4XAX4n^oif|E!^w%heUqnjBD#rm6~s-;6-I_D3}VG4^*U@`
zxNRzeGUA+!U^ki2pEpjo89gFe_sEP{Pdww(Xi!!$U6oBtPfHWinu_VLifQ1BDeMkh
zn@7f09Gl26u5K&=jUEAMK|r~=Vi9gV!x0ptf2)#cB~S(Oe}*LP7iA*6`jTeIaC{t8
z!7F4FyzVr?J6i?s92LBZFL>P)yk!tPvyKx#E|W?R6C-;);>cDd-V4I?BBrFeceR~@
zvxPRo9wz2e0Y~^`x#uGBNu)dm4n$1XQh&d+eRM|UYJedRG3jEaE+N?LG?{T3(|3rm
zfe|H@j%Rku-?m&_D69J;fD*j(fDHi%d;4;o3Yl>TvG`Qi*Sb^tWW3ecEg_2X`gJGu
zGt-fD#y!!hc1|%1T37H!8n||0phdT&Lioed5M1wYNG-O$@69F~<uuv2KxN}Xm5rJ&
z8#~mI4J_yW$F{v}O*U>z+LfEtS+363VgqR{HmWW5L2a=Szr{B7rx6uzs~?s=;hZ<a
zD3&!G8DY3cdmPG1VR(`yqU+(!8s~(lAFTVLj+e5Z%h0_Z65xqo8E^Da4a;_{d^S>V
zJA}1p&ehs3WY5%r`iyK%TT5%&E468_(xzSKH|@lDXVlTj1RV|A*&g^4HX=Z;tiG^M
zK}>}bGH_wO$ziB3A*vc4c9p<&Z`)H7tte%o+QwlRzYrW3V=@>f;Y_2VFm}kwO~#Nl
z#i&=q^Uq<Ua#-VPpjg>R4SU>JO!hpo^2p7_awGI*wqE$L@pfIdCO#{zi9e-He7!dD
zxn2{GKGHTHp=u?nGIFd9M*?b_4KHUtvxHAvOSs)OS4&*Oqx*0RjMHCs#owMg1zFf6
zdpOQ5BgLp31)xaiz8j&zv1&tC>_ektuWo`pPs{NTFP8V%o#TENS9X|!F;n|H-wku1
zcYgvl5M5h&=K)Mz*I&8W1au4h6Dv`0zIM&?vbEcL(%S7NZMU1X-QMfB+x)iqOVG0l
zNae~7hW^!!a~>OqH8=%cF7&%oCy?Ng!2%4AxEDP`F?(>`X-Lu#F`ZqIjUgAMG2}L7
z$QP9%7kLcHqC8lVy&gERjZwpPVyqXd<!+4J;eio3@*S`3W!Y-`;k4S`p|$<C*7nlo
zYa3q?T#*gjm1*GaQgGi_a35+O+$hv#5T9AJGsZ;bquFp?lZNvi#d)vdyu1aRdpm1`
z4i0h{N1{>Al57<*_%~#Oe`6Z_-zxkE75@4b;7_&i-CSrOLoH2zfLUfl7g2s|JXXdn
z+3Im?T0I`sdi+J}aa9ZT*w?RziK-9|*<&T?_H0%7YFZUGsX{Y9+JVdST7|2<Dy+P6
zQ|xFL;kaF>pN1Y8nMZ~y5}_s3!Xd#6Ry289^z#x9K;FR87Mf&n(CB2`hz3fyLQY<t
zhD_hkduga0B#~hx9TQ@}l!;l>+~eHL15i*Lk8nN@9W_>UE;l<|_>E9wR|wS|0z>=<
zaGo4V19EtEqy%7`Bh4<ofGIgr2h?7IzO~yl$i;GXDyQy7Ae0`KkUkh7W-^?x!x^=s
z;b0g`LPTS_F0`iWDiuSxtR7@iFCvb86-sy->SgQ!%cC=y;e9y3b2h8j3ZiM^_?1IC
z5H4J(v*<&Qs_Gg4X#864GQ6i|;x=f?y@OZ~=MfdSmdC4k>JkA?PSVlj7Du5B3&U>k
zORmz0!myq`I8b}I^tqE^S8%#mdzw#d6QV7t^;JfY;{eG`43HwT<qRsBUZfx@Np>PM
zhyB<lGyqPQ`(;GlAOJVgn7Wy`x?zA)xH!2?6F?0`4Ho)#i67=G=%ZPr$XrM!)r!8c
zaaK3qOzx5qHLf{Finc&(esoSf??=hZx-in?wn@z$p{334bj0uvfn3l%Dv#xG$&KKN
z@&E72Hs<{xZOq&HA~)u>@t62GUgwQ@Sr%NUhpE|RfQC5@J7D`WneFyrxI94p6FSJ)
z@&1?D>V98Z-S^bG@1=GBlvnrBkG35IXC4?gdXNuDLM_!w)iv0u&}ok&DzqBt3uv#l
zgL5l}Fi&yE3*94wPRa{viLv`(+Az3v5_6y{b>j&9e`l-wK3l8&F|Ab&)mAx7Tjd6?
zRWe9qZ~?!GY*GH44c}kV@Xb_wvlQQr&Ep%Ll-w((nH+LsrT)om_@7S0f1=`_ulPUH
zJpTBm<JoNBo=*d}NWpa|xPNUPT(E>+M;ia-Z1`SD!}oT@cZ%Y>xp{n}hYVM7IK77K
zF1p;6J$7lR(%qLl4#svLTi%NLrMs$b&DXB)z@<)R_CBpoS?hC4^YsZ<^Xmw|x6Ouc
zyEJ^azYpcglKc$CcWd+bM&BG~v)$&8?PR-T!~Lc-+~+7?&sE%C@Nkdz(nnWUv$6mm
zv!WG^S-27(*f?s<G3~cUHa_f?#)k`)4;LvPZu9uC^0qc?sZK6nenqCV!Q=%r?7N&Z
z?Lyux;g=THA$Y`K$^5mf+`D*a814wmxj42)h$2z|!o`LW*)W9{ncH)Kky6^QQR;L-
zGfcFt@)~Y=OekP$WRyuBsSe%DZ=9yh++A{+4`D>WMhON*E~cr}G{QxC>P68eD*_}j
zq{lp5m8|?lU4~{ZFAu>TNQ|Mg&O&yw(mBVpEjM`$BaTDHh&hPZmt25|=G{@j&d*q`
zCB8_f!9!yvMwarsh*~xZk-gUpm`>rrBxSW#OZu)^!AO_MA&#UZ3}f(5?19K$)QtZP
zl7H~YfcG1=YsjO*O0t1$WDl4;?SdaFwu%^E9W7@ei*5u({?*ZhWr0g{m4Y9Q-MY5>
zR@FQVPGL-t&eO!qWBjMu6iB#DY;T)69c(@(bZXD_4drV?j6yfL)4B4kS~UsZs^W8!
z4ToaF#~5v(w^NJkdO~55Ef)j{;s!ZFbrLs*IiU{KIX&3awq4OpnejA787@?deJAgL
z)5Hv{s5|za>2FsH>*UoY2-M2Gpn<QjQ7a+iD+CE5FS2A~%b6YEnC=wgX}I6EEJT!E
zY}bdgu+o(5`ki)H_5k!|5jWOeGKET-N9F;t5=3>bJ?#KFcfLl}O{>(Yd@#Z`rXlLH
zbWC-5Acs1KL73Um<!Ho&MkuzYr<Z6LGVkDn4vKhPEr*CiO<E{|9nNetU=lHfL(>9B
zL5?{_j}&#`ikIFw$?;}V)IJIu4LAcuMnwn7T(3AQAh&82BiUL!G%T*Q!@X59AV<sM
zTlkw<S;6Z`s2|-ftZrcl`Bs&sv9Dxww0+XGB-9FJY?#~#ifoRNa?SV^`GWf_&~{my
z?HFp<Vg=DMg|E4;09rQ4(flRL#E3G-B;BK5TB7?K8^V<bF42*szFx=gxMV82BN4di
ziNDR{#0IimNSTC|G;<ECC&MaqS^fRu-DV~W`oD=Lk9Vy@UN}-%f`$zk6eJUI!@-Q2
zrYyZ~%m5~0J)Md$5*gc7sNy#`j4}wV5%&5<TDFF|B4;J;uVCOodvJ!~;B4oD+VDtm
zoml~3Y-`7(Qad1G9WW1kAAr71Ssh6u1t}_pqxOL^SRjd|SfGtSR(kY1ocBqfdD`T@
z;bAz@Os7JUfY6DG=vpYI)?ThmFAYvdOJZ`IUM=-NA&Hz}(|b!8$e|&CCy@C@(8sFM
z-ZwnZKSgOLL(8pl^$bx6U9!N?hO0aVjp=-EdSR%X=i}+6GfUktb;9j|RNlge2ZCAY
z<jfop5-SW1wVQkG7zd}D!Uv0@jV2i=NTCryI=bnmnWi}JL&Nk{ouV1hNo@|GE=8lA
zHX}D1^B}kgiASf4<}H=GVjBkH_S^|d11?zu1JH8PU`*wvoUp?8!oyCyoVPDvK{5;(
zrW3dHXm(JY!F&bGhH$c8xg6$WH&I74f8k2QiotSD7)k4lP&GJax=AcA>~G#I8y4iI
zCl@>Ykb^KRxmiUsk-E79X-F-LBJ?6!QnaHNOi&xu8*d}JR69&XO03Hf8LU*J8drq|
z-{wx@yWHMR&$eGVIBmc3Dc!GJultqnc>5JNQk-1DR;0huTf@L(wDD^SRWZ5PnG-8{
zBx;m-jygj8$Z{wEqnkYlFy;)vY<CP3Gqm_e`7Irt<2i#-WpA{U_+aDg8b@~~mU_1G
z;4W>Q!?Lx`5oxV+leW&y+B)C&TjvvPOAr7MlP02*V@mC$<*?KFBK2Ui!&ae(r#W<O
z6$3PiF!M6fbs-2Kr|bfDM(_+M%5m)JMH3p4!wg-Z<gJZ$#%NDMc0L%!a_eFfauzj;
zpx{T!-lMazcUBsEZ&UVuQQ7-rpS`y`_U1Yzb33|#)oVN6V)Vx9JH}n4iNGNk$$p?u
z4GVb?2Df0g#43kJa|->ONAdSyv8@jmQI!u>z;W(ljiGNNiSq%bLQ#Q^#ubgWfop_&
zv=L@!YlOLJjqnX^gl}pi-0L?&A7bz_?>Ls~YjH--Idw5-IyevKF0SyIo1=6lGr<lb
zH96wO`cvEn@urIFfltWBg%i`baHn$Nd&-5M`&_s)feR8dK(;f+t8fU~5#)ZNRmS6o
zECD)sumrIPhN&(=rkk7-Vh#`(MR7b5Q)98sX)>U&)g`<e`fKH3V;tQ$3?AJ_a#7_=
z;{T46CyVWoomFC?#N0@0eM~#NFk34uN^6CmXe<0wTj4&h6;>Y0Kn6=nRW8>!fZF*E
zJ+(AM78S67z+A{-)AX;pGhvq5&nTrPcWQKFf8H}MaQGn*vmG&XW80Cs9osL@eoN2h
zVMqaTx5xwl*$_&G0Jr_iDK6nGp~+VUp<Y@|)r0Yd?Sh)E7h)#*JBpYoCby^p7Yq4a
z%W?gT(|~;HXd-9>H|>4lZpMiu79xc%oNvI6XD*XS8_kTAgu@J2nMXD?+{aS?=+0oN
zhE4+<nDu#Qbk^xK?zU#mvzk8F$fEd^B@ni*(#{*H6U>zgId@HMeV9rHhen2Di;%ia
zYbVYJbj)6st)Hw&>nHbXKl!!xlizs#Wc1r@^DAqmO@(RTmtE3}R%$sp3YXd3oOMn|
z1QR+{#f<6gG+2?(&9c)3v?!Uwto1q;TYb!e(w_`A#C~V=ErQlHii<LmH?AtR1TVVN
zvb8`ytpy&@7Wk94z=M7Ze67tXzrYZ#HAJtD34+ed)V;978DFc(rS{%-ocA%-{ec44
z@>n5at>PpfhJR)n$Nt2e3eNQuYJJpX;XkXQYO~+freF@n`<Bwi=^=I=4jxkJE@b0v
zcN%a1ro4SzdHb->+Xu4nwzt31g*9B18M(=#7;K5HaLTc^h{_;Nh4#V$Pc%eW%Nq|i
znc}$#u$)IY8~@*i5PQe^-z)1&*;wD3#`<TK_0K8m|KPL!6Xy6keeXnqZ6CBI7b-=n
zCGq6Ln#Nku#3vD6SHkuZ58K)^^ukCfNdpl@TWjO+Z}gbZjWDvj;nrVw{;+Amc87z?
zdAiSNmZPzLxdYi)UrA&AYs&fw7em1yj!}<#tVhGA1~IXIdU>c=lIs>*xz)0vA4x<1
zQ?;9KrRe|K0{Vl~TSi<jQN8ZrvF7x%vQ_7twCe1{N{Ac48?`!5G+&+gUgf-O;NFu4
z?#&8rHwE`p^Wa7oguE6`;@CXbbD~qtqqFvu&z!Yn=?U`|tjy1yf5O~_OBOHaSe`$A
z!P058QRD|gpFz9Od4?cFJ$6a9s(v7?s#CS92WVBFZK0|@V9c~5!&t|$mqer}cwTGp
z;cPXyBCQ68X$=n78oc1uVC6TE`JfH@=%|UG8vtYfuQ~I5d)sm>7&snr;xTb_jN>w9
zAn2clDuXq#j&lKqDF_i2_+r5swy{aEOF&m5mVB}bG3pTHieG)8*=|Dn(KcC*Er*&b
zn(IR+25SxrjqaH2-T}iL$A1_j5f>#jCJFGQs~@BB54fX(stSjaVph(GQFDnoX-$O|
z4}xI<MPng?q`+GM21pkfrE^?iCDd3V`Xd`&LQTCg9ga7w1XqR{Dn{v9o3Z>Iaa{#l
zPDXAN&R_NT@_*B%+iY2lywDfjjx9h9#4*6WL78nIGbYc*a@T4jAmALK*sPF#3`laY
zlYFz1nFA^NlCjN|&NL};qwh?Dy=-KXb|Tst!$2XEME1-ZMs4VbV(l&uC*>wv%cy5?
znjPJmQxRm8^0HpCz5vb&Eq*A(6e=BE4Rph%iz8Ip?)hOXUv78j5iNl9JWJ2)FT}&o
z8Twt~@c}5SwM<<L%gnlwmR#?N_T$LHB=P7cx^B1<iwJt;y98(UF%qW6CDOL}n1fe!
zh`%P=5Pxmj5PyOW@$+<u|EE91zp%L-;=>k{hId1Jc$<2};#gs5Z#&-E@|_~kjdFZn
z++*lEU;>@WpRtf7128^X!N?f}4vyt;QxAtL^G(F-t4*{?B1{IL>s0UzEU%`qG}mdx
zxFn1D&|DOa5)v7&BV)83!r60SER`-1tv!szkqAE^WUHw`<S=9;#rN3|QB~%1I>fA`
z;kxWtLAS8@a2%1!|Fy#}oAUqW@GGKJZ}=5(k>QvAjK|QyRYX4~Q1CJmjiqs4ES7u3
zJp$3saQuvJ682blZZg+RdWNkFXN1_HP~QHq6%}Q-UtEVNqTG$MaX4-?*xM#r8A;z}
zB2tul5Hd;}jJTMZGDf(amX_aR#+oduOf25Yy?w)YyFwKlw05FFs|@mx(x530+tErJ
zRSShGeJ(`%Ye4{Vv2-T&Ne9BiOMMQPn(#|Aa#K4vt4Y?gsX1g-=U*lije1e`Ak<-H
zEf<SC@nv+f2XeA=lMQ5iaHxH-DAb{{HuNLUQIXDQ*T)8*O6mg*>+|Iz-P0mP#C@Tp
zINL}(QpRrQMmIZXfm;l(gH@&}*UgbppPxhrKLb?_dXdQz76L7$9#OmYGT1hfK4%o+
z$-)%nPUq}}2*8{e^V4s2E+i>87jDY@ld@sEG22}DnY6iZxz2?vbS~U1%yKE|Vn=VH
zN!PZ+^w?^sG6dJ%KA2`jN9|6KwnwHKnH=?jl~AKklGsdLh>B=z2I$L_Z*273eCO-p
zHi$iP^z#%{IS2}cOX70oWQk6}#cd?^Db~J{I?(b#PG{t)4vRFbGq};wZG$v^bRo2j
zHvVnd8vjdajo+t@U)ILo)^Gf;w4GQ$_pTa;5!$veZqNq;hW*Lb=7XkVL~J;dF<=&6
zv7Qn3#VH~L7#C>c=U5@;fB3Qv!M{qZUTt{j22bi}?JJ$;$Z4tAxp72cM+<P846+f$
zf(@SZ^=yp(W*VbMl+kOH(cAfq{uW%|5Y)Yfa|TdWW3G@<)gt5yF79w5yiRsIF<S*1
zhfHK-StD_XMjRJh=mA};@W_F2iK`WQND;d-$L@`^r`Xw42HmJoguWJ&;4IF7F$L$H
z*;sy88q43SEPtP}d<T!^D{t5|YMPFzbDVX$Q`7^UPBCt*iI^;xC2LEvXh@_)&aN&2
zuWwQ|FJN)u#!)ke{(mbq4)Jpr&eayBJIPAT#YVUe3P)-1Tf!Jb_BFn-58E4Af=R!#
zhU@-ymiOVoW0ds&w)Vy>B+a5wR*|dJBP43JhdXI=_vw6VV<PH8lx&(NYc9C@G1e3J
zwr*i4=YotsgVO?StQFbC(r4U!u06|Hnlrg<)tlZaH<@R1t6JYhC+mB&4QM}48_=%M
z0d2hwXuJ9Y+HV-8Z63OZu!9OEXbZ}N4BLRr9z6B3suOFu<>E|GTtwFa9&^C9$-ayA
zHTTXKdz|K(^G0~=L{{wFcZk&iXQ=3|jy~z)K$PJBNPl34=*=C#o+`hnnHDX9u`d>&
zh~<((P$IgQ2?rPM;%g&3>Bq&sgu)h`Lz}!z$IDbjbu>iWMjbTWmQ6Z9ClB-Zk>?!0
z4m5FCLVj(eDYkhRMZ46wFPoU&pC+asS24X-#dJ4cOy5s2UEoe0LiP$d3?3MMhd6&C
zstsc!z<;UM-Z#>fmxjlV6tixqwkizEDMmM}N4KY*R$5HJ(J9uA`=US7vzzp{!KA{S
zjB&&&3?lBd^X7oqL1smSBj{+KSc|SzvK}lnPjzBrZ#SVi=Vrh`vC|^vb9<lu2~Z<r
zdCd;T-@*ZSJd`2-d@W=|X#YYy*a4~^xL{jb9*XF{@y<pPt2bKk4eaW_%_cy<OB0|Q
zRe(ON0<@PeK<83BJJ-2!=8Dxq55fjA+msn4x;gxqv4(7oofTPAC}V=cKE?9fZBAq2
z)n6Fw9f4OY<jXgDL{<>XQMSGb@}y+|P$>ex+kR-5K+2q-!5`Q-Whv6W7%g$^X(|$P
zMcOy9@Ub2n76#2Mn7RzI3ba;bcu1&p8mPV4aA1Tr0;tU3`0tDkjIWHIX?0)oN4m_K
zXt*E;4agr3%k{z(nS}_~Wy??`92W5vs6l#hLb!i~YBltEVO?n$85NMpBU*9v%4*RN
zGy?(kVd>2xZ$-;7MewvOPc;=K>{biK7|Rv#v!nxA6riK4Qq|`*dUBIRfsw7+8WyZ;
zCEJ2^TF1t5iZl`#h^DdnHfl4e!OPw_i%|hit|g3V@WEmMIYeKASA9<z&fyrVT52<^
z70fp+EFK=GU1T}q#4;YXI2ap{Go0s|fmDYvC`KH4TjGwdUfB>{Z!1@0mUp9LL`epb
z42^|$0IT<yt?drF;Jax#H6u>TQOOn-6pjt`%8GaWdT&X`$wo3N+D(1Y)XNgmI?ER^
zHOkZ3X5(kmX5$~|Y<#!Q#s_<|aRx89;79!WJ2;7FmzytU!}oF;zMm<+pDVt%diYlU
zh$j$M42h59R9RAqv>+lO)@9~~n1q;kV5*x5pStc}K4QmUEJEGnCljIhWX$?>YJ1y~
zLKU+U91*Qe2IKQUc{q=mw3fV>@Ybr6P5>fK1a{pSv*s<brC=a7<=P`DLnCl3<^e*^
zv0d;_Ud;`f)ps;l7)ubx$Z95B7vq?V3u$7{?OV-=xPKalTue8eT(pyVr^IVU2ZP@Z
z4yFr`rsDniXSm0P^f3LKYXOE)h;6)<1IkFX->n)fZ@@*dEXs#&g<)n>HX;t?RIpdY
z(x}t1%($P!TJ17FsFuT&F4f{RYbj8A;5=ncbc)zivGbLc44k<TGXYNHc2c)r#Hb1T
z{egGGm^hXidQR@fj&=Kca>i4<R&aEQPkfM`siq;VUgIeCEwMmN{5*c8N0{1_uU+3D
z!qh|BgCEu&Jj3h3|K9>t85HN3dMYYa<ZV<7)NF11?Xorg4rz`5j5hwW+W2$)#{b`J
zf!a1^fr?GxyJX|>o6>mvs`B`s%H!jG9`{Si>1CLt78O=QGbbm-L}w++KEy?UbG_`_
z+A-)GXBupQ`Vr5=|F<F5j_$<7Je9cJ#&V-1*6Ec!J7ab?8coSejPS>zNz5UR1J2PE
zkkqPlAO=L3CK-CuX>>Sx)4qZOJxKmblfsg9%b6nyzQ!6HjJIVWfbzpI1`#!snO!C)
zXjNrPZ1nA_j*dA`MrK~lV5TI@?Sc?WQ6A9eu~a~yTMMCth7F@lr9Z3j8|F6O3-2YV
zH?Ye#bE}z@xk50GH~htak!WAkIE7UQ5@t^_wt3EaNp*C@oilBJxvPpJ@a?&GSkGeQ
zZGfQ$_+|1WL79{*YX?n*8FP&bb8!ak^w~vkX9?@#yq*OS!H6YuEnMWxld+AY^n|ZC
zzDU~qXCr*^r`dT~4nWRlyDe11&_BA-k|}>}sjynw#tGKgT|W=F*5HGro=R90Bob9#
zE7;D6+7DZE7|-ExWTSC#`69kSxv9egLnd_-*RNKY;_`!=&>eQI6BQW45^iGhnmN_@
zYtOZbR2zw_E)ao;;)(x3LV3W~oaPl{LqXgExCpW*5M@EsDXel)Z;kGV(yn(53zZ3x
zSVpTl?H}^4;zsN-EUge)4-&MxG{|+ls9BwOqHhdF*{?;f6905p-&0_RSYjN-7y}J*
zX47?2SYn{Dx;C;p*6GL8XUx+D7(l5CJl=S^H)5^RF<8#B<A=hvAdKIQMY4T0bhU&r
zXCwJqD$hFLnuSF!4NJs=GsGBpQtNAEn;u~{i)#5@#+b`-7-NFOcW~9D3>TDO!XZkD
zDTdMBGF5%g>~6&xlqi=4l*PFV1AQ3Wr_;t+sz6LkO;8D?XDf4}iN(ZsqzN$UnEbB}
zxqRtHS)&kw1%W$78mA$9JT?j)>SR`tgVu5}yV1?5Kca*!3?imJ{4b<?b7*6{McFsd
z0uVv!A}wO~(%px2vdof$BYn&QP#Vt5-AGnUYx*;F?_a02!@IQgMd_4sOFK?9LD$Wo
z5bce*$>zND*d`7|`FI@eGT{&FzjfM_=vFi``H8<@g3Q|q5XQx*e=$bRmZiS66ry2=
z3tbVRQy-RQj7-yTAHgwT<+I8pCT$lMk!%kOkpmCt<h*PA8cp0ZDMAObTx@Z%+>LK@
za9*b2G!+(rZXgz20<mYA1qenW;cPfcI&W9^MLesjb4ju$RfAmXD*^uy!PKN|GQs&|
zD&PXre8OQE%6`i^EKN*02sq;zpRaHYG=N1YUqo6*D4%m6RV~r?92uWD7W1EU>*|Qf
zzw8!u=o8hFOh@3UYUqomU?bbI9&on9x#)Qul~=98z_}!D7;kVkC!G9@eH{le7<~q=
znx-*nS0`vxk*xH%EN&6BHS(2+rb9+QOgz#GD6%@Nm9?`eW2{D35h6K449g&++mHa3
z7&sh6Q)x16Ia%V=Iy_w<qX(u+EGaY&D;UW2^I$Li_+SyR&b(|x62tNkk?f{3Fxqkk
zjx%8kF+DN#j!}eAIj^|E0B36J?yFQV9*5rC%vXg0yqoKD48PsoNC6>q5Vn>HG_Y`W
z%Cy{M;}L+>U|}7ebXP{041}I+Tu~^w@*qMan0FY61(99oO$zz$Ds+|mv2VfarQvRC
zBjKI^q28b+0e%DljRr+L0qYUSgYB#k4i0g3B_hUJ8F>AkWNE4YGRJbWz?boD5t9@T
zQ|Or@IZhlB!ER$Ta9UNa0wvBTBThjzZaHVi5f)vti(X0^T{2KXYR@^f=%SzU&u4W<
zlJFc_SPZ1rM2ny!A*Dj6ZEmIP`WMS6JY$RFb-qLgQPOs+IW14Cgy_pK%^GvyDy+Rq
zr%iS**%qT1V*$kFf=i5UQay{T%A5%aGZ9q9ns4gOJ}6Eu0z@m1fQ(ENEeFO86RgLH
z=`(ICRHbAXKPt1zkq7OIT^wTyJVd$O-0rW2@rK(^E(~-P3nwr21;_bOY1PMeGZQ0O
zTAths<9>sks87N!Beudt6VEOd22SBR*2dzMflnz?q|noGg19)+k4~eb$0{s2hVwYF
zidd*xl{{6UdARZnj0~Yj!mVPHfVN_sZ0@7%`GIhbf$JR|!X%GPQV%msTQ07|j)mqg
z^tR|jpLtq^+2L>|eJ-SJ1kO05$x#jG)4@ndb}YkI#YV0UJ0dz#2?8AU#^JT$Sf-j9
zk>=3bP5zYgDrSDOs`AKN@NL!*6JQy}Y8y7n91B6i=rI(W9Y$tEt)f4go4Em_D$Y{&
zW9T%Ug``k<J0LI?Om_)JebYGJC}-P=Jt<tshgHBxTSDvNJewPQrzp)%THLU&iXZpT
z+>J7wpd^H#UobFKsd7#X*W7wYWvWR;Hg#(5zylG?6MJ~XL1$_#(Jm1QCIYJtk%`l~
zjB%7v6^xnC!&MmP8HhYaY_rC8u5Ff)&ST7U>;s|HHmd*=8?D#?%K)P;AxN6pLC9B_
zfibpVVB`FV`H~@I&1R9P(3ueF9WmD^gO+^Uq|Zjj1dab#SyQCQZ<v0G1D0+&u=$Ai
z(9HuydoUJ5ngo67Jib&KS~i4jL#J`Y#AX#Go5|2ZHkjZnMD&pl+yH~6RWC!(@Tuv1
zP+A=yMz7&opT;fp`F;eF!XT_9Y$XjL<n0{XpRP%8DY4u|Z)V_oY*OIVC2!^J6WOD)
z9MR^WTSVTma>}+s(j)U3Ws(3)dpUCz7(%s%bNylFP?KvGV-|-uPC)LVD?!+f!)SW6
zj0-GV5pG39{7I@@8z6o()<6m3Cr6sh=FByntVpnafe_T-?}IsRFtW%wamk9BoxE6;
zp?-vtx3#GZ=8b!PVs!`_q30OjmXjb31n^Vgy5Y3-#xX_U+*Wj1U2-c&l+tPZ-`H)<
zhspymt_aDgQF17bW}3uZ*Tc<3YT05ugD`limaILoHqxl};6B_($bCB^IwWB)M9Y>L
zZWfZ4$OPAH0B6-%U<CF7k`TUAzQZ9DhGH`NjZQn1bCgndK8f3BD<R@da-CCrwRL98
z+r`1>V9vTExRpk&OX1@Xd|v+kA^*PP7F@o7Ul)&b1)s(Lr@UwVIqN6)ZQc>I&&Q)Z
z(zNyWEWuk6$wYohSeZTf(flnYOh8o0e?yrK4el2QSKyV7;7YkVC^s!017)u?C^^0$
zP~?|DSrr-%CI`2UEp*?sLhl7EQ|NwtRSI1Y7MfnaOM(j#ePi@cGbC{7ft5bGtoTV_
z2JVM(0wIRY14%;Y8!@d5!Xc#Z88E**wnm4h)oAI_u11GxjSlCp@wHe2lq1rh{8XVF
zsZfsEC{SjkL76$naph=*GIOIqnUx0R7``BKe5^t_u6ZcIY<?5(RtY^|4&Q2I@P70|
z4$fS@Dm`GK*8@f`M(7v}VY*-E-Y~jg5S60S1vw*#Xjx=NXXqDf$zvNd1P^R^%)2CU
z%)8ih#l55}a?x&e@AzpIGSUrWR?S7Mv)AhqxVSiti_a(*mnat(`CMG>agjR&3H;1O
zhf_IHxl5-<ID=4~EYa^b4x{E{2_%blXXyyDdxfm4Gp^O=XlFK#cBOH&n=eQ=DJn-7
zdmLT4J3aS#J{BuQC{L7$lgsVclS0!nFf=?Jr}Sa><2nfAGW5+;;s&y|i2c;Q!IMSD
zo0qK`DXzD2ZH2`Imilsd7j8yzHz2+vPpa6jn>DOwogr>mh9bJ><yH>tcTUf5$&KkR
z<1=mB?y{pTJ<?%fJ!EyZ_8drS&p~a^iniz5y!K34NpPm^yE5vxSIY)&Bn{je1-Dkg
zEgLVmv$BCZCk@=W3hq1wx55KAx}t3cSJDH<-idWAx&u=tuveIVP}?}8$b)vc73I32
z%Vp0{Ut!ik?T2trjO9<cI16bTBvxr2`{vdB%i5BWxgl?q^s^2=JiJ>t)7?GvNE9-N
zm>YRRf@;=TxMxjK=-7GV6E^Ftb0R`)W75S2DtoV5;ttmAd+aR>#y;U0n6byNZ_U=8
zUr1}uf7kZBP22O7|7Y6sKa$#WAsPO^rzIcE){+mUwdDMB+#L0=w&Xj!mRxxm=O_(0
z&iw#0&Bz>NH<aj@GnP077c^6{I}8<5Gjg=9EfT|uF=~Xf5vBn9wjIWCv@||(O`U2t
zo6uFq=19&i($?G)+s*{dkpwn0(ZTHS!Nr}qttC-kCpHDqrC<(Z9D2MiyK7MoMM{<!
z3)pqbeNc^+A{!(!L@N^QoilPvOX9hTULIpWq*i1>?B52N7I&8ln7tg<42%sO4jQou
z^3k!vWL?1yrK5NBTp`zaol9{zY_{1JcABU^U?tAn8I?CSissw|YgJb&WFZS3#SFw@
zhjCAJAY)c5b4~0xvsTx#QJFJFm?(fv&&5Wl$kAq?JcVSeP;D4OR#cS~kIUtwJ7uX^
zgiqrl3)>xp4*GT1^^`97Q4>-$o>btI5%#Mk^vITbYK>$<3d+56L9X-VLO**vK8yVu
zuXu`LXKxCf)utF}8|~ui(F&H8EAc$!fhp7eSO-Yxa9F0C)+)-B@dEo2bSY2?9k4|H
zcMXQU4hJ!4$IZCziE603gO~AE-<_j-^bg=@^4j(x93Z8wr@cCWrG!mGR$XuQ0r%X7
zI)9J~8~Jr9eqq6A)EbjbahFPQ!Qr50zL8)2X0*K_iFRMW*9YV?szWltQ7kY<=@H!S
zyEB+~lA|<}uLz~1o1@fjyF;F<4nq&sqSh=phGlY#9~{eHM$va1eqKCc>T6NV;ZxZw
z&E>Cgp`Z_69K{JNQ&7y~uevDaD~bh*;<zSx0Trw<=DIy;9Lhq4a&ki`9SUW!LYdnn
z6g4UrVUmUBo~NnfXx^r1mNrDQT+yshG;^9n<Aem(u%2?Sq)m*2IYq&&YzXF51+z-Q
z9MvQkSybogZym>QnqoM;A%;%HP*4oVG>HK|1vVLH0+4P6QfvsMM}hPzkY+^|o<_!I
zkcZs}Ufs1gPs$1<Xb7cWp$sULW1Hj&q0r`#jHIGSh8iNNDw3KaIkHJ4Z~-YH(&1ol
zoEc&N9cc*R2pT73KwGO2niWzxL6qliF^*v!OO)a4T@5ju&8LFl9L3PAkcJT%WC)1^
z37K(TLm=-_Am=NPlbURgNN*9dU>>U?9RJ6GR*yUHBMITo1r0&nHrsLMA_X<SNl>AO
z1Oz3Is4jUr)Z?^boKKf1r1v+3G|H#aU>{USCo~DEmpRGu%tKa-s2k#VE>k=oYKZ4@
z#dC$?X?oZ%*XXg9kKl@fIU6P-(_U9L1oIIEbG3qLR+?ACJS(m^UFh7ay;cf=e5@gm
zk1LRC6-cwAi^RJeJN;b80!4N|*$~3@3gHHY&@3;^6bRjuQLe=sCG6QZHUzQ~aYKy6
zenx>D-=xH1K>%w(O$-GPT+Bdl{EWl-R~9aOxhWaWJ!UccZ+t2p=oSYjXgePFsCczY
zVJuC<bJMu@J$!nJv!FzmhNa*-h~_21Rb0M)f#pRPu9upH`uDpWMfoKQ*J+Iz5Z~~|
zm+3!baY@VcpD3N5@|Vy_nOx&^tfvSU$N5S%VX1yE%M~8~jK89sh%MEBp-6tINM<%u
z<M#eawV)m-{*K{3h4CK^Vf<QQJfJX+XbOhtNAkmPjg1T2Zxz9V4G}z~2p(1hvzkI+
zm%`S$!(Vs(UUB@PA&&o29FHiD!<)iE&*20{{8<4!+7Q5F3gE8_AZ0-GWV7y`;BkfU
zL_-KqDTJpLLdq`IgP_LNgq_K=isHG3C@R#P%JB1oqG+xq>g>~AQUEVE1n>%<3V?qo
zfaXRC_$=j-SZz43{D^msaFYCILnITr9g>Osm1RTkHHhh+)zX@B4Qi}7icMLjj8&WQ
zSKYB{3q`S|qBx|ft|4rQ6W>~iw>2cbjS}BhiKk4`9t(!6aAVKYs~TSQ-O!W9frOLv
zb`61?@pji1J1CIm`lncb^Tn}a<r`R_G{jB~A?(bjBAB}<gu|L@2nuEq6GuH3ys07m
z-IV_B<Dk!ps~-J58`94y{k@fbb0Y)me?}t!Z)pf%Uj?wA0ywy-2IzwGdA#zIS&}sV
zl!mkqP}<X!c5_qp86)t0PMD$E6~Xj|2o6>RhbV%S-MH5ki<J@XVgyK>3*m@(XhRH#
zD~2N!!&{qb0a-VB8oZ-embAf)hSX;&^;t^2xi%2$Viu3r|5!zETtftN6v13YkfN#b
z+CUJP{JdN~(la~`<phN?uOXEA3T1&pX-*sjiaie%$FNW_oZJvYhhkW)7-lym4m|9a
zFZ9B?h%n;_z(-6T7H(QZINsI}%QD5XT(P7mo4uyt=mg)5yqwvA!is1L#W<R<qrAN#
znw5&?9g3zoL6JpFKhneI8D|403H|j$ICHLQ2<0?|a=JoU&{QK~3YS3qSWnq44^^DT
zU8Ej+IvZl@QcT^7sk!}Ev69D0z7h^2K?7Z1!C7huqgP?{DU2CSu|_Zql_581pg1oA
z7Ab<ex*>=G1u>`~np^e@h{562NhpRC#TgAz)D*?AqG*0`1K)!q<6H=(bWKD0XDa=5
zO266ROcDs>55*)O#}Ibfvl?PJO7-36D2C=X!x+&^{rP&m<2;rs67lYaIF3~u=PQop
zdWY??7&D-H_m|x7Qy3RCgfUBDT%<6XYZ4p|=jq)943{W|_cz2a%BQmC{-9!LZrNX|
z)SxI59C4`#CG)a|5I(FBE>{T64dph?SXa3Kb@m7oS9x=O2xWaTlrgHos~nVIn+>8G
z{20qiP!0Cq==d+cWDA_K<X)g21L`mm*AG@$W|Dqz$C-{;m9Ge~l*UdV)^1Ke#T$E=
z1xnW&;jg-z?N>kGAgomgDP1oKLg-@|hp|p!ysIIMv-woS{v3rdukpl`1f%SlqOKL?
zJO%XbhJemjK<`yR3!4YDroTVpQc@Sy1&ZpzhNvD?UDd^kDrLY<ViWcCbn-{cfp~Mh
zUvYe(A&w97sWjK6iX+8>kc5LVL59WQ35G#9LJhBr!}*ZH`EWxxzfrbap>R@0tRy&`
z#t>Y>Xwke8>q>=kRYNF8p6fVswL)p3g-nc{Jfq;mS@JQ(aZN)US6|?8T&p-z7FkIw
z(W8-Z6xS(=Pc}qxJ)er4+@L5@_De}92Fhn*a{{|<cY<2+M#b{!hFCtUSpHS9w8D{!
zb#sVwB%IN1ZiwO|dN}PCMbQFBxLIQ29|SEZK+jhX%dHKue1T7;b^d+4Si0bL0coX8
zq8^qnHpKE}#d5o1X<-7OHQzebyF8g*NB$~{mx=Lf4FNr=W6C!aP>Sv)Npy!RL;0aN
zOTMKD?r4bMJA5k5^WE_xSRF_3Jw<R=Lj*ri1a~Wf7A8S1MvYf%zD9nCdY;KO<EVb5
zsD9iK)lu4oeyXTa94?YthKVWG<+~BPsE++Q+(q5n5KsR+*G@lIJWHBiDqsSnva*QC
z6U9>Kh6+nUZKU7Z;Dw{jFB?L;k55H}{zIX)(5p<)59}&T<bee5reCvo=~WLj1a!2v
z-ES38N9()_Lo<X7iwjlAsoymO_As9cr~Xrct!N%tBuZb1NPLW3#41nd|3GTOxIZ?;
zcYreP5yh9Hxk(aw&Kq2l)??D28-n<Rg7}MqXkk2%MRB@J{gpI?HGgY}>f?&)2}QN2
z`KA+8y+b32goR)x>+(D{X@eZbx%HIddb%O5XZch_>N&-gVoFSEGSL{DLBNxpP_|!a
z2;}s+4#-R60b(&Ec%l5A#Y#)Pk_=?bcK%fdBzXM>+0IY=G!P_g=kKN)khFpP;to4F
zoD`#O_CMy`*#6w&4Z}42jIYSo+lP232fxDY$sK`29Jq&Kcpq*IuYb>a_vBvQ{24F5
zjy(x}%L3`()X@>XA3Vgn594C)mgE>fX5Xin$4?jq;ZH17bPLlP_X>L$K~4tPGwYY)
zTX!trQI^A2?x?R$kopS0A5*FaW~0jBW2HfFyYsgUG@sx*Li0(bnWD*cHMKO8z5bt0
zr;Vc|LikxFJT+To#~S6IPp1|+_VJ<;O!16xHFb{sV@`Xx7%87aiI>E>knIG2Cn>4r
zD~@QeW7r8=5m8(61UivQ$rJfYy4Ox#$-$<4l+E|g$iFRoOD%jWEqv#&aP7PIOXtKT
z+7FxPox%1k0^CUfzEJ`0S_fckkZ1<n9%I1mg*%1%{*y5tPEv?(QHXC2A(}%<`;H+|
z*Eus*_!KRCsusSxU-+~Z3qME;KUfRjGc4S;elw$Ra72qxj#el$6-q9IVwiZ0D?Iok
z<&n8X8hN?K7&%7?%vA!Dd;%vp0w(TH=E4_f;U{U~`}l>Q++yKNwD7lS;rsc8FKe;z
zQ?&4vTKHtY@Kask!QYzZ`{_y`uLP!s1ayRXk|jSCA7MIzr+N1oUb^u$*8pa=4|*J=
zCR-{{II9)Tv;;WMljaLPoELfbC0;6VIMo*6tWh{?6;68soL5NmAH1xygx%WdRo?t3
zFWiZTtE7%Vz`n*O=h-KgcjuE-3zNl#{7z<~i}*`sqJu)P4$&sG*KCT5%oPRGbwAJ|
z%>h*~U8Xb-8B22u(%h04#|2CGiWcd9gx^VBuU5K;)}h;$K=<R>>BgM#zsNJI2}|_b
zRyaKJOL*T$JZ6bLIrtJ^_%biI<6>L-EBw3hn*Hm1MT$Ezte*Bt30SraG8CEjJA&mD
zsujH4fv=-7*M`5*BfYz{;B^Fdd3UMg?FhI%3rlEfuDJ&Vwj#&7@jGeTU+|Yyenvv&
zN3rn;R{E9C;hPQ>Gb8j(!G5qs*e59L|5VsB6JQ@q#Ig38!lqiqOPt#sfhbJgIrd5N
z9xHgvK@U#jH%(2z%N70;3jdf4@asT%&OvE%0w^eqmlVcvbzlTtsAY_kZ)LT+dHDyv
zj>v7|o-~+9_hgQ=K{3Lw_4qXF;<BP}H)mNQt8)_Iimdkfa7B^WN79K6Ump7;ncH@A
z9R!hDWLv-Ps1Dk1Q2fW&!$08TA0)ksyPLexe>2L>%FR6#&7O*8UOhBbN7qekHKG}I
zcgK$=TpI1KXyzxNxe+oD?a)n)Hq%*jGn2gyTM&9(yw!mlO;M%JpSK)mGrgcNiGCM{
zvB*$Ka&hpBWfLcu<@%-O@YxZ3L&oSM9r@rW{<12mqw(8L@iTe*8g#OSM+YCr{}#>K
zBxTM!R*O82zsw1`_?$Nf4`MxkdrtBv_;<~GHy^jdv7jRm-YsO&Cwosg=LL)T#P5*j
z^T|>^Sr+@Ism4B)-wP{d3M&>?)0%!Hc-NO4>ut-2K}t_&aU60+Q#7*4EpOc37|122
z8jk8xOv?vn14C@pW1)PQOc5z4D3laCa59v>o;*V&8$Z=n(MaP7nxdjf@g+`1lg%z|
z6o@Gtq5v8W^(v+mPl{wrMFjVd@D(xc1Z9@$_td)WC7nb#v|7=m_(|14Q!!y1VqlsO
zoEz9%1<0TRX^Ab#XT_R85}r`1Dw33QYsoCBcTlMbEF(l9Tv?-7S`uJ%C)2qTM5YNm
z>lDws6i<prQZjGMA({j*!PzW0v;2Nq11FrTfLdy-p%S8ICE<B@Q+Uqj+tOg~RXi;X
z#c+ceF0bKPidc`5SttX0thV?<g>#XE6Py$-Sll{ECXjV*on&Tdw}N~ZNA!C8hI(V2
zC4Z@szf8%WT!;KQ$>h%+2l=a%{7000M@T-tet3_g@0cR<`+G_L8hp*wg59Hkf`v6+
zpr!ETdIffa0$UOSa|onaE_AqJ%~F34l|uR^C4IA!URsCrB^#FXzbon6l=SjCq(|eV
zgWLIq?8`pL7HLr)zpe<rp$Og{B8Usbha3dg9J02#+}%yyO^n%Nr8LJ~is<`_Xk|S_
zS324;L|6HDWz}=Fe^(a1vOwN2$bYHGf2GJzt%v*?AB{LST<hNzM~v&--3^2DVa55M
ziu0Wz&TMwRlvaF1Y5YlPoF39pxf12gzo35JW}@L9zU8j|q@sFCQFVr>oDd19&qP7l
zZQ67EYFy*oON!uSMbH&Na4B{w!7Jk?^O};Ga0&_(l}phlvk4!Kt2SG(1ex2nR5Cq2
znXOwPV-(64kJuema0Pc)SMu@^UOvjpHN0HQ%XPebikBOBxsjL8@ba&`+{(+Bc=;wT
zck=Q>Uhd`PeqR2_%j3K}&CAQYyvEC>U&duiT+Ghi?&<EqR^QZnTUYPVmX_xEe67EN
zP*C~!l4d)W>aX}>x`kgUf*l-!U`PIny1VSe8?u%DirGqk!zvYm*P%uQ3814awj4L{
zbUu9yzD!Wz-+meHOM}ZV*;l7HLNK=Q5*v0?Ps@A_<=8&;brxy#`*nEyExsY74sGmn
z5hc}Lhqd2kVHsKbT}L>$lfR;@y^A*zS-aC%)>=***8YHBMp^qUba_Pj+~;cC7uL!z
z>9RqjFIf+y7db3_4;CevxGWDR^V|A8SwbYeabYsuX-Zdq3ElpXu8!27V=Sp3IKpnI
zyqmVIANqHH#JfLsckLqa&<r4V^Fdx}RB4^PC{UOqzXasB2-3Dm7oJ;Qbo14iy~uXj
zy~#N#qld4EK%LlFpq5HZW;sl)4tq<*OP9H`CERtpgust=U$c7*`bgz-XxzB7jYTtK
zl|vIKnw0o<S<o1l_#`;}$#Cqx<XD9>=->o{p%fSc=nx;(7c<t?EDK^o)jTbFSc@Lv
zFWYu&@!PiDnY=B%YJwTCv?YUsbPTqLJG9udwAdjZ@i}~y1#xgF%8zwROJ;?Vd5@B*
z`efcaZZhvwG8ZYCVV}$;<0jLsWIm{5*7#&D8#kFi$y}ji&h*J#Ic_p%D4CBcnRoeQ
zt{FF(GnLGBO6F{z%%{dp=KD(_figbbsASId$$VzqWG>L!+^l5Y?UT7>++;?T%omi*
z`97K3#!V)4+`V1Nyw4}|)p3*gsMhA2O6EeJ%pKz<^GPLhr;@qYCv(@h$$Uo1{7}if
z-zW3qagz!CN$*uMqdu9RkDJUFwKn%DnM-{#_m7***Okm~mCT2HGQS%)neQr@-z%BR
zeKLO>H<=$QnLjI;^*))ujGN5Ql+5Ew<|?1eljA0Hzmj=Y$z1J|d4Akv9#S$dE18e^
zWL_CJnMah&Yf9$hKADMq<6YVOP04J|vSdyD37^cC<0kX0l6jqy`J_)~+i{cmyOP;K
z$z1P~dBeELOnj%Cq<2v=|KgK*)40iOrDXO{GN1Oz>@{vO+bNm1D4EatWcD35nVpr)
z6eV+$Pv(Geli5SbOjk1h=94*i++_AqGKVRd&-rAI7&n;%l+4je=JP(8S>q;ih?1GD
zWd7YJGk4r%j#4rwDw!|(WEPB@%yCL)k&^kcPiFDB$(*QUmMNL9_+(a$n@oq2d54nu
znonlcxXG+gGI=HQ4WCS*6*AT>ShcuEaFcyBFURsSmzNWHS;)&`UY7B43NNd8$@9|9
zOD`|0d8zPH<7F)`XYq0#FYo2$B3?eg%VoS=!OKT@xrUePc)5X>&+u|HFSqh?8!xx>
z@^xPB;N?zUe!$C*dAS#tn4|NZzP9S_V?~wsI|lZXot*p<P1F93HBAdoct?KO+{H^z
zh^cOZhh2!k4x>DLh<_(24=>OrlwUHJf7@%qhNlZfMYwgH`bcVSsgL}U`s~xVKK+&6
z-cmeTMTgPs9Aq>*2N@--%`;@xeD{CQF;0F-HGaivSXI#jcy3iizu~K*DtfCFxTtqx
zT<d4`H{k~Hcl<(%{BFn@Rd0%Ac@vzUjd{$;*~(4*yPNTt?dI{jbq!X6agM3<sVTas
zOX6vlgu3wkF^4_P%#frbtkyZ_X&YpL!L}fx758vReFewkiX+A6CK*TBMQt((%VV-`
z(VS$^oKzo;w3MKEO3|b^lq936>)R(yw$K1sSRY6jQR~_90jU#H%W`C~0kXJ0kcp#^
zN8!i|;{$^9DE-B;K_4uTB?icn1R%Dvh3@??J0QXL5<1&QXiL~C-hny-@&Df1zdLUZ
zD%Zvy=Z?iV+k=>CU&klDBdYj}Z!`8u<4$^-so$m2NrP?qc|ETR*FkK(w?7~6N&~cE
zBr}7|_v_VrdlVcSWOg8>9sRpfhd1z1ympQ2@L^Mj&t@EGEdKCZ!!C;d?t1ul_K^zy
zUEE#uQIKAAv%zz};pwidPDY;Qh(vh0w_^HHJxsef>P|KU)9(J=JxG5~|E}1u_HuXQ
z_<tk#7tNYzDv)CV87mN;(rDKT{G?t5-r~r|DzJ}#SA=3;f^b3+gNuyt2JbwJcfSbU
zdPg?woMT1dC}sN$MSV|*+QFN`f~PuoW=_TegKfTOfd#SHf;b=oVvIMC*OO)|`8i7d
zXJg5?v&iYm<d<6VD=qnhnjpVG$)BX;e-V<`)xuj@=An+hW6EK?d$_wBZ<$pV6_zQq
zM+Hal-84xW<8H9bp=xa%P{$^|f~87#{f6nT--rzH2^PK3`QQ|XCs@f}Q49I0TBz@Q
zFz0rz;+VbX;*k;O=urxIdUo{q>3x-f(sZT|Emfy8UDx!J7tKTJPWQiBwc7@_nC_B!
zPLCz?Y#%~w!B2Dm&2_w70|RnL@Jaj}&6y=@gYQOy>-o(e@J(<7esu)@!oP3C1r~bA
z3MwO#{Cb#^nEbNV=HMn&DV~Hh_!)DGHY(IhB`SsfoFBPDzdW|kU!@i5Wet@=?+XhJ
z{*4tnD!2vzyTqEPZ)HU<WyL=)ca!pi+I+A<0;K;)gXE<U6-f8D0Le=h+5pn8(;#{I
zL<Q0VEkNQ>EFC70q@fKU{U!~PmseCE{k8>2UgppSkRD8f<RupsNWa5JSME3g(t>|h
zQZoycmq@e$wujQNdAUXf+rure>oje_)A_3br2kBV)X4&5kdV})0_pcHK=Lw-Hh}bp
zG)P_&Qi1fx79e?<MjJr-uQW(rGE#x`NDGj>l%owG{V5GnRU1ySk_x0hw*V;|4I~9=
z14xgiLGrSc3Z%cZ0LjZp+5pmHX^_06r2^@%EkFvnB)Lf&K>Ax6BrkobKzh6dNM2^r
z29TaegXASI6-ZCE0LjZt+5pm1X^_0+r2^^c79e@aNgF_VCJoZ%+T$dDsX%(R1xVBj
znE6*yls16$TpA=Vi>W|*eq4|wS!n}EFQh^8@|g;x7h8bjB`<9N>7_JCUVc-7^l}T3
z!a+dNm^Og)_cTaerc;6RN(+#@RHhEntz#GA|44)6Wjhr}ZQM(mm3@}=5>Iu<)b_zY
zN%bzQ&=J?kf-6Z$D{t6(!i2WtOFiYmQm&BeMt+@I&c=k(Yz4WI8j@DzhWko6hKVyT
zF%GT7mTqrbjtAx@&ixG+5Xm%GD<S)9p+CpeK@~)9&oMUaKxr769#BqUIM-bm%ypGe
za;cabuH?*tLByK}2pftn*HfwDul~Yd?+Efi=S1(?-qse4RF|8PLrU+q9R3yfi!c#1
z9+4=SAjo27NV{C6NAcH6!z2mPk<yq#HB)KwVsdFYGNM@ODIgwi4ymn2O1W|^H>n%>
z{wf2^pEL;s%Q$T;H(XFbQ0=Nz`fVlnvjpCh2aDxy<j_GK*7lXa`l`j8GhYjr+LI~H
zEA&g<Lc)1XL7AzJ!ZEN&TD**~>NtVBjyTXhVWHYkQg|)5w%p&(Hbrf#rDBA)rg0?&
zXD0Ga7$kXcu0F$alh>3B(PpjX`paF_LUr8~P&cWNnT|OV(P*8;IXfo>umLD>IVor|
z$RW^!{6QL#{N>{yRHCey%7&!CetW+pRYXFDq5jgD<>7U?%3#h1!qfz&!nFc&EthJj
zGjb0Zp`tVmmta_nN)_7{UzG<&26Bjs-i4=>>MFV~>s&h1mScK~bjD(-YoxcAPth~a
z7%3wjC#2GT(Jr(W^q}hxPZPxMiOPu+5SKnndj4=9`iKDlhO!>4wvC}*YpoU$hHij*
zL`ri*Q-+#gG(yaKGOLOHZDM?LB(vXg&;wh#VCz{e7M3gw(LvY9u?uvcGYQIag&qvU
z7)=mO)y6M2<4C>sl7<c2tc-!1aB5~uMeEftwuyka614YeBJw79AcauIO$u`<7$bsa
z6L;bSx+C3vxk4?stglclow#H<$VOk(u<cnU374Y0J~@=m@!d>x7-1{uP2K$?Mf9~A
zl#jVEihcIUP6k+vE}W8jMymL;I!&w9$rRFSt7WE%Ep+#lI;XXDMh2YPP=9$Cc~Pp}
znCmb>c23K6V&((ax=S@q*b-Kpx=>8Ie2CcN5<$MVS{Xr*eT-w~Ho{rwZ8t$oa+^7x
zAC%4<!t5&pQ}jVw=ju{vD32_2r7p<nYL-$MERuR<U}zXI{Hr6~%pKEdS=Li7^%qgU
zL7c=#yFy~zw9k6SCd`_Ygo4%}OINP1)IXFPKy6UHnzT#_j0o9l<aV)JV=!oLt`B1!
zNVGW;b|U2<ie*P3{FpdgPca2yeNvDC7=c?x2Ef%QREH_j%&<#Wssm==E_C4xO<NpZ
z*p!7>Rlt_`zjiZt?!7l<o~?&PQP(<-ibLy)mG0c$d#@L*>&11nu2i50zu+3rtvnNS
z$`ihb=Z$pAUe<eAc~0a<4Rp$zr4{O>xR*jV4-3t*oE4q2m&o4LGeM`kMH(b8*S$d6
zvIR(9Qu_vwwn~HK<-He3Tekqo%W~fUQd=6NhnNXmG?EYVmq2=53y{2o_YELzlLqO(
z6w)IKY1<Yc{h43$FlX>6e>H&g`ZP$7DWtzDr0rUO^teKLA_da+X^@^$NKY%I9a@0&
ztU`J&1=5abkSaXbDYEo}LV80BkX}+qFQ-7-DGkyqEI=UrLm|Df1xTS5>7OZ(c20vd
zq1!>4$Y0Xqc4-09ru<sOdo%uOAl|#CLE1tgZK;sn)B>cf6;fLYq&KHQ+D0L5tB`hU
z0aB<%+Aams?rD(1Kvz2`q&-@I^ag${4YyMYq&?Fh?aTs1ymwJZd$j;5<kFi`Am!2^
z?WT}+S4ex03(}q`kS3);$|<D171CSA1?ep*koHM~w68+iPa*Bw0;I|OT4ZTT3Z(tg
zARVBPrYWTTTY%KAkfx_Vnw$pdV1;yuLYmS7q;T{;GzHSsG)RXlq$3p40WCl}ieF2^
z%}9YXEe+C4g)~bc9oPb-V-?bIDUjOJAk9%oa~0C`79gFVkmjX8Iw%d&e1)_?AsyTT
zq=gFU<P=DUq(SOXNQ)KHTU&udQ<=!p+fpDMng(f^LRzkn4r>8Y*yG-w0_pHHNGlc6
zI~39pEkFwAxK$~Tj!c7ennF5VAsy8Mq>xLUDUfEQLF!US-3saG79fRODy2Z0nFgs>
zA@wPwSuH>c__c`l>J&)Fq(K@`NP`OL*cKoSDWo$}ARU(msiu&I71Hb$AcYOLCI!-*
zG)QMEq;(2uZVQk?@jfdB((!4Kj+ze{kcI9!3h9IvAf3mrMV8*30%=|vq+=D*`3mX8
z79hP(AzhFHX?_}{SqkYQg|wgrNS7$2_oqNQDGkyn3y_BUph8;M0;Ev9FH3=RavG!$
zE2PU6(xP!eS|5d!;0@4`1}XFgxJn^yGhT0iC8V0@4KO*u8(<MkQ2kkb0df%=32ZUr
zcLp$4{~RoXgdsNMe5EQmCbOG$VXkymWEcRdB{4CTielK3a&s`Rbgv$&&~`w#gSDk{
zZ{KhY79J!fF1R$G?Kzk?J1T=E*gK3ng4(=oiMbw_KytNiI7W2$LCtSKi(O(yE`zIr
zya_X`0cB0E<%n69<qK>_gOCO*!!#VwWKtxwu!bH=+C+fL*w^zkFIDjkZN(X|IXjK{
znr2>lC%~Pfl5-^qGPns0s#RDS<uS%WXwvJV3hPbi%Mq!j#WY+I1gPe;9Jp32_w>MJ
zfo;lW5sUX2DFQz<(z$a{$iNLFPjvxlkpjpxWK)=;4jUQZM7<Y^p2tRnPa;<}!BniQ
z9qg|ZATowKea5ukQ|K<WcUM%f)tLiK&G!^)NQP|;$1ou`VbsOf5Cwkk4A1y}vZU7(
zQy1FCa!RbybQ5x}9CS8-glliAme!POv?FGc+S*X5yWCUmwmpGXWPWSA)@193R1o5D
z9T|jqkxdH^hC;9X1^V5Ufq^p2fkkmFAp=0&*&SEG;CY%9&{3lC4*n<fl^M(xx~(Nv
z`lJLQH!!AKKLU}xYIk40n0<&auL3_CFO=!r$Y8Nll@W^R9%upmKkU5;oE%lPKh7ee
zqdZU%^!d~>D47hC&cc>VGQeb(BqTEllSv4~(CL}(St>o<LoW#fgiV12n;;SvSp?a{
zja5N`2g)WOn+S@aq9AS$S!5T{-#O>rdv4Y3>bliElX*V>|LCWvsjj-`d(J)g-0j?R
z&?{;L{JGF5&_**6QLYR47Vcg4CDQ3p*o4guty8MN4P`Po^$1G{NKpa6hbxYIrsT~O
zCj?b00c=9A1;3M#5%i28DT#E-xYY`^OrTSlY;ge1Cz1q{JgXW_BT)3wk;Mc!fgLUO
zXVI$jIbtXpU&Br9eR`o@Jv;y;V4s|=3;~1R1HP<`I(n)|hYIbJ(c^Kdkv~dTF=}~O
z-*83={kp(2B9B|#+BX5erAMHi>j#HG5xE|m{9v{STj+y{b+C`kB{-O(euzwZ0<}mg
z5vU`Hu@h~bgTH_a0d2@T_(rU51oyDPKOhnS-Q2w(WDy!cDw~H+naZsoPWTn93Yc$r
zRTu%NkbEHpk3mNGE0K!T4Gm<#fh^uTl*q?>i)g~JhhiVO{-L9&k8W#Q3QChNLjOg@
zF79;jlWD^V&@lDgRG>jV6VNFq(g~<nkU--aaEwWVp@aQ@@KVIefSU3D8<#El?Qo!_
z_Fn?F<hyOdzci-7ra>~@(U#CM{3v8geub@2o!=;j#Z#e&>Ov!L$rtcznlcsfPVpJA
zCBM=JNuP;d&yWtQ0n#W#I@$v1a2up!u>di#WJs%OfOI@VI>7?zvo=U4F{IBiq}4S*
zI+-DzVu5so4bm@pXLln*iq`<?G=}sA3#2tRNT*`~RB@XbQlbV(Ut~yUS|Ih>AU(-a
zI*TFo)d1;ihIEbvQql%#)^SQo=P{&I4Uo=fNEcWj_1hqwcd~+X5knfN0n)_`=@JX1
zK^vq?u>jJg%NWv74Un#2NLN}Qxi&~wF{H0Eq_s6b((5MQv_MMRAnEL2-)2a|V}f*z
z1yaTa={hWcD(-rQlpPbK?^qy>*dX1=kZxi~N7ex8`}j3=yEj`P<!q2%WV>`LL(10x
z={AORy9H9g2I+@b09D)_45?THq`Mf>-4;meY><A$knUwj>uZ2?KSTPl1=3MANVDfC
zDLu%LMr(ki7vdkbKswq6DczwU{e&SMQv)Pzmwsx2bgT{1<5&Rc($5&uh8iF}fnQUL
z`-KJ4aW+U4iG?6N#gLA#0g{%|uPl&Gut9nn3m{0pVMr&|0O=Y0nxyos1=2}2NED)p
zAU(&BK34-IEu}wLAbs8ji83k?q(3pFlWTzVKln9C>3IvJQ*4kZ^%FsQfgzn*1EiN2
z(#sY|8*PwY!2$@<s|;yV4Ulxjy>5YYnhnwstqRhc4CxCsK>8bgO%?aH7gE5cPRRyI
zZ|c0mkp5$=n>w43YT%~M7vEEU#FO)q<Rb6v;QEy1lqak#MLy-pi*Y$xjf2%0SY6Dl
z^(=^Wz>*y-7s9E|-Xf%L<UNHZSo&12*TFZ*>~K7vf){8Vu8|LvG&?ium`c1Cw`AxD
zRnU7$IMD)2#aIZfowaB1s0<XP9^(NQTuh}!IknoZe|Gb*YA1Ig#7V(aY#`=vYS|dR
zFa?*g;cRCj4aa7ZqeKW78G$N&Pand0CO85+)K{n*9y%(OgyT4{fGyX%K^Bd$Rh1or
zr7POfST(7)f5v{38YlISz<*YYV<hw)ZUV{gaOtrT*IHpckrp;_F&@7|IQU^59)7}w
zFg$<T<n#k3u@Kt=Sk+M&wJQ}HNI{Gx$fpS_YmR5r0GHH30EJvOEmoUhJsH;8@wj5Y
zBfp`AHaMb`D!{fHo(+TZZfSU9EHtSMJb<Rl&BtivxEIdQfnwkSC{W<ldR#~p%k_sE
zB7yf|TZ`UL7U?W8t52gd4;RVB(O!AJ7LMn{MdBQwa%3@;fz^I5Ay|Z#+X8&NQoIgC
z)axKL*p4jbGPI<L3%&y&H+A-{(IFk_E9P()3zq4#8Mr|8B~ke6(&=INFKu!4K@EEl
z&WgdN)<}HB3x}#z02069k~p0fg|%Sc*E3?%#rN16S#Hb@e!Z6P27WzthEtz`v%~ln
zY?tWsQgAQ~p3x$2d?@J?)naJotB`=i2XeqS<l1=(+8*5EnQj!aSSQZ=Ag==5$JOxB
zcpn_|;*)<~{DJjO-$0Xsqhaa<ps(DXToD$`^RVt7hePFXuG<GYut<DaR|`}xeda4w
zDv<n1ZDzBic|QqFZbJjxhK2^-|Kw^9EFbr%V|C&%X&evgCvhJ|DvD&pDw<?UE&Wq_
z=I<577;MVp7E{-fp7tgOjWujD;A%Rs4_oE~pu{qN8&s9Y`gqDsf+@&gJ!gcXqy9r6
zw=9|qX*uxGNH#sxH!64V&=BLpTxxwHmy{0)Nc~VO6{Ag7_<dAvmcW~!rBY|WOyoc%
zaa#!+kJwab<WqoDieVI1EQWmr-{w(JO=4-FbeTN4CP3hWH25muu~^w9;hq_WK_Ch}
zfJwXy)hXW9M`(#2%Q{v^B@c`g<6@gEUMvhLcLZrQ4g6U8{H4p`*yJ!C+V{LqY#8NJ
z00HnF7|G@kAvQDkm|)5R>Y-=AXH(F?W6y!YFiP=noz&Hh%jQWB4rtV-Lcr$97j46=
zd+~%94MypX_Br;`!>oEYVC;65txz2p`VlOi3f)B)8fEhY!*i5Y;}&ondRI8&E@NWj
zAF(!Z8d#6Ynx)4H?UiSev|s2nJCi+K@8m|O0Y8TSG^~YIF_7Ov(IQ2{Q}?3skzr6Q
zzKd<F>$uXtXVx#lkD^B=p`)CC#IJEA-Nn@#uxIpD8=QxE4dl-Z=h7PBJRb?pWi~i3
zU@@d^FEX6VYk>1|Bsf=C;0)d?u)M-puB-veYmu;g&4R_%aV6hiI9F8<M<htpN)yn0
zzixw~lS{qLP(E0F^Yus9`6g2JPrZU{%eQUdw!w&0q(a-`9od#$4REDWT`T&kfo~c(
zIALHC+V#g?k^X|QgE)K-P8KlXN#*P6x)VcSw)q%6ZI<IHc`5>Z$HbfH8Rq{N)r?mP
z23j76=$dGfy3Qu49a&QEVM*;~NNVMy+rhlI6{cD+S4pU#MozZZ1!o8`eG(Hb?3r_;
zz<swEiR*X8SX{g%oD87z18Ipl;xW`s;t*&O*1@DEr{}Px?Tz{=x;KDFQ7R8Vt>VY{
zf&LW#!he(2rbgp4;Kda!i<h^@XT)acg2Ws0otd6@W|qGbla;2v+2DW0N_b`j!ZTyl
z6ZYZ~gZvG3bxlioy`V`9CL0=X>Klhjg28SR&4b0H7WrwRzeRP?TvNuX!Qmh(KSq;)
zp_7Wq05ytVHBjG3#i<dtE^=uWrl>SD<C!kZ9_4qUFRk$9qSPqF)=2t31CE3a<(Zv)
zD&0>d$c|m=pGE-n<@q-rO|JAwB|{;=@X^8dqczD}Y)x`kZjvA7CixMgNtTL|RYOct
zXyvWc+W@z!$Udq>7A~wiqY3M7o3JLau=ZeK{ih+Um2=R3iGP9elIuh;6N0rKoE(eA
zDcKT)jlu@tZ}Y*n3g4Sl*<O_wgHP~Z6uZUmS!@@+=d}~~H~uqh7(Vjzgir9F0VdFv
z@p~FfpgqI)B7!kr6ZQ=Y02_z@X>mqI3xhDdLM>B0kWp>8c80FXwB`ZMkg^)Lrq=sM
zF+Yl?+xOXYdkX7z1MBvG8M<9sDmz<T>`cSr53<eb=d*~^0LH^C`nmmaG;SZXaXX#4
zox$9G!sPZjWFn}u!?{|8{^j4YXWBVzp8C^hRr0v4O6G8t%;hR^%qm$W46E5;V|N{s
zQ*9oxjei#{b<uqm=dhU0!TotO=6_*hejYR5%FItPneVeRucA|KeIdwIil?Fp;8!*Q
zEMx&3%mVndDS*?01<(O2;SgpKmj0z3hJBI*L^;ePl{lnlw%q6t9WEW!m@qj#9ZgQZ
zwaIBI%jpo7)1IcBRtL+8S_yVgw172XtUnu#_2+D?AIhw+WY+6V)>nqGj?=xZ&$atJ
zTZcbH<NQxH&W~Wu<IH)3$@%x|meDx_&#VfTeiM3+68q<%-VFCIX)}YmG+2qygBM(t
zR=cJ+U8Ai|*c5;bNAK)S9ZtbLl(O}R@fZY@KFuI$CAATkvvEd`XRL5p6*pTZ`<72%
zj3)b+ZL)V+_G?-8jfU)3?g_~;K>QfO)xof>asfdfVB=w7a1~@o^kBdIqSpalb&*h~
zw0|eJl0DTx<1b^%(a<1XR)#QiPTS&SQtQ<SR*iSjW_l*<qBz{hlsx)IedAKsyJh6|
z3wj68E!6NnGkvxDMzq>}%T~MVxOUfb?Z(X7y_sqk<97PA9;yrWOI74u+Tcp%DfBa}
zXy9^0U~4PDJG93{RYAME?EBC>JvP&?0scK6Ph|rV*!)xtkM@>vhX+fC-Y{4d2<y0M
z0(0(F1)MfDG!Tzc<;SdXlBKPwadWv+<mMbG98?7a0v&{dOljB-#4TfZi2KU$w6b1q
z`&}3m+Sin~$JfGwdY;01c`EDWbVDyox7W477N8bPWOGBoRa0ULgVQ*zb1j}}9n5;c
zGlC|zxHGP78Yq1d$@1Y`k8P8kFa2j+Y{5<vD(toaPM-<bF`B5~YZLV_dsCmmqMm7r
z`tZ7?sT}q>(4poNaJ+m}oIS@qa8IlB8N>0D!befy;0i!T;(W^JlgLidB(k$jBImO5
zoyQW{$COCFMIsBt!Q>7+uUA1sqVG6|ZB>*aO6No6(vis&euDOWp^6y*I*&ElCRe8A
z#sh7?;B)Ed!z<vG>^$;J<7}@1)ceDtSl#eiMK>#G!}<aRGS0&yAp~rPu>81-&j(wz
zIEA^DlN}-i<Do`JxjZ#b@@zLx;-PN~j+&sC&`Xf`hewUo^eN`F3EwGzm34fD0o#K3
z1g?d6tV>km5*m8>OdK7AxGH?u)=+L(4h7K8AKlT;-=J<NfvyT4u@$Oa6>h=esnA)v
zP?7(m-Bj6xv!J5~VuN?524Gbg9S_AhVr92ubxq{uDLika2LrBa%6%EqZ#wO)w?;Um
zxp*!P*yD7d5j01f=5KUTg+4C3P|Tz-jyP=H4U9H9@FyykegFrAhZE^WI5CdFIMR@a
zgj_3=i4kx?QUKZc61G9uj}u0r{X+vq^iu+hizA&N7gLXdh})=Y5^7(*K<o!NeUOB?
zNyBM<Kl{KpPE7>U;OxPSHH%;*NswV*s(CWF3Bj-E-Gq?`(&&tp%{7y=S@^a_J*tv@
zu+d=CxR0Tpks3eB8rNiM+&!pqJu-*{8H@fRBo5KGC~&kZe<sr;{bl4M_L<Q(5Nc~;
zjS5;YR1{;<q#WpY`A7t1z_)v#2jHI-^WY~32jsy?6ug8-AzecPkLjak;}99*?L!m5
za2C>!42z*6^5}4h4Fa4)!T63s!%I2yQjn=*>XQDc%aq$Bl+&I`dLOYN7sX(33;l0g
zRK8nQ5<59@rO6xR@C0#dl8{4OuqT{E^UY$E9JXe>Pc+Tl&!)LL1^I7SbLX0xdo6Cr
zso|S;X~x()80bTdaH*A7>dh%72Mm~5^r#5RA6U5dk_R}H3Aq%=Kw*FCp$63mZwN~&
zO2a-CUy_7TFYq6DGG?GeR2$P(n3Ar@cn*-UhjS-k**tI;eLKuQMRC3Yg}%XHp1qID
ziWDx6+!AwOkYI;C4n~R^F;0XG*F^~$SRprc`313t9*xi)Ws8vQLMAZr(1u(IhHtiP
z{>sxcB1Lp>408CaSbmir0>~gXFvCm;uW-L@kQELCHdYSlJg!^lI1PJSg%)97%vHp5
z1KE2hL7LC~4u|e)Ua>qVTEm=YYnVF6{0rPL4=@|%;n<~YjfM#;5VE66f(mg>fHc7I
z1NSb+&`hO@{#&TfE=q;``XIM<7DUs9gKfHyKn<Yb#ha`P2bsFC%&QB2KjoSXHFzN{
z&%js_qdYY_bWj2A<Us{&&`Zfu*XO6edXAV41zS;B-?Z)tHW?Pm;jEN@_jlDYT=|6`
z$uBCuI>0`XCHKE37g!rHWsEX|;sijssbO1=LnQrgT>APWgc4Mc7f44kCpq^VEkUMY
zex_E#h*ykm{VL@q*~D~5Q>z}ET4h<Sc37e2N%IV~f(_fSwCCi~y)OD#M;*<8r6_s2
zO`p#&O>9Kd>#|&qh$fdcHo1Hdr9sW@LoAoJ8su_lbh+dqr(_B|;zf(2#wLy}$(HAN
zvKmrZ{t$M~rH$#2#`K_#>5nthpJ1jtYG8Upbf$SABTwcu2`eLKIe360AH&k<>_<M<
zO$T3S%D7Mszq;v=(V;MTSeO|p7W|^gL=(-3O*DG*a59T#fgzgGvt#XtA$p-PI@a+F
zVVp~+tHj`!HcC`5LN(#5u{+hKF!JG*(5e2e06KimeN;fL1RR4n+Eyz%n%CZ3E63nR
zl}=@Bbgw~&lg-diy4S+pCig7-dtdz8dMe{A8=P)5Lu53LU^rW<hhzHs2T0~@8<aB{
zN`j%BvsIv+YlEUA<Rlr&dDTO4`CPm{1rZ>bFWcbMEmxH^z;Mp50nQM9ZPl3zY;bm&
ztKh6<I2TqACxXsgWP@@UL&-3duWS`47u%rdtU@`4a>-VK@>Ls@A_AaBy^f(=x>cZD
zW`i=yP>yCOmsbzPJr=*Vs_qpwI9IkP$sEUUuB-vh36bD@%?9TrEQTuTa}4LI8sG%j
znV{2FHaMp+oKqRj!N!QeH?l78n`mUcCw2JNg>x|=bdTnd@V615eO>z+8y}mQk29E$
z#ao?^<Cu>dqVe%v8z1K|ALlY3T?QX3ccxWN@J`27WHHhz-!vKQcPPwo4px~VhHI{|
zj;5@X;Jt1t|C0gGacd9e(KKe|Z4R1>F7E6|XV*(I@S51IjMWYH$FcmltaykkD^BT<
z4tS5Qx53VnIZs|DR*_{q4!PH((*>`=zzyj{Zz`Q%11p%$niil$*gCSgWim8FYZ`)m
zVBl!lC|V=Vr`KZ~XkQEl?<sVcw@(p{G$J57oXNoRNKPDMc%U>TH$s{+wgaXSE~xmv
zuoNo-HsfcZB8Vyfc_nNa)Lls#c7b7=i$n}n<*h~Eyq|`>)lL2@G||t>nu40<wSKJj
z7zvguy)m3E&zb;LtXu@}&Gc@G))@ZV))+43#&8)ohNVVh7<c0&XRAMq#>1U99=^dm
ze3N<To**866pe@bY&=}gJlw!M^cXypUaeaqvlWPB+d`w^TEtxUwCCJcg!m#55CwfF
zF?lmJT=br672#9NUz+NsI_i)Gv^Owf5NASc3)gF9jwQ2k@R@<#Xb3rkExbvX0%*|`
zK{}mI3@4n_aBqrYUn?wGh{m3Q-30wZS-TsTW>;O$cslne4@RrJhi#R23s>H)TzM;u
z%8Sy77HElAVLqC`S0UcAyk|-0lD786aXfzOGZl|V<MQV=F7IY8?_n-i*1)AZUR;*#
z@{2`7k86NW4!?~ihi7ea_$kZbF_y#OTT>2EqMy(^6=SdX<nTYy<nR}p9G+%5{D$T5
z+3Mv04a2j7`C>G1FWbO9$KZa?;Et#sT!|b>P`KuTw0}Wf_GoWJL;scy{fmtLB}Tu-
zK)-SW?(wvx65w>S4o~;<<~R6FB-J)1Jym<g*bj(6MY1<Yk2Z>}-efkF&rB|ea6u59
zE!z)+K5(&c)+fN50$fyaZ+!@2tI%>Iq$%-tFb)hOVE_wxL~zijwyePw!-0)%-m=c)
z-VAJB!&Xf`l>^MMp%jBuNkw$igTtP~5SuLvzVLa+g<cHz6|r*}%%&kP2l)W>7b!TK
zpLWbymI9aJx>fUG;kVc;vLURlA1o9`@=epG4bW3a<}y4Dt8psqRbuD^`eT~NoG@+1
z?CG<ncosj$Z$P+P@V<qwJ-!o7Y5r+bnm1W#-eRTcHI%00s`w6&xr0)(j1(xwAMIL!
zvrT3Ia#O&#8wLWy!j}WrmdX9zFQD=0xcyCk+Z!XzQ@6+0!koST$#U4{P<WL*{*tB~
zRwy}Ot5MtApqY?mO8b!bIj+{xAb5`8=Xr-{Jnv}Z`Q6AVEqUz7JolSCcY1l2Uarbr
zX6v*1-T+numTli>WA!1}+ZL?u#H_v_KdN%6e`oyKxArCdmtRJ^>h)i?JA($FP0`~V
z&t6Wa%txbKlJi9f%*d`xAx@!k=iKKNNgh40dy3w1!V6UGeK$`M&PkKB<I=eLDYN&b
zhqX4-zmz8*=pis+5lE1jq0Ui_Rj7aSZfJl*{F`^NY3@yDD9xoiGE5)vXl_7j*wt31
zZVk6#=~U*2J!O6icC{>158C+p0e<ak4GD6ME}dAnAeF}GcOvR7os)t-6w~aX>kl4d
zNn!{>IZ)`6k%lr(%e+?McL(8T!Iy1_f&i7x{~~WRhE~l7kz-Xgm%}NGDJO2(u%&)*
zhb}jUNk2bo6XU1PgSW_j(jAG>HSOoMb<1Ea3;PEVZ<kIk(ZFPlc(A6?!DY3PQM>`K
z<Gw+@iS-qei8aW6(3w#@0!3{ZX5j3c(gTE)Ckwt{%5JPmO^eF!<<nQbG0DeKV8)tp
zfLuRmlWQZ(mF`HcX+y3n$yo{3>R@f7;G~YCbDwlX0%A_Vjz0wl#`wsAqEqja`0!Md
zHh69DG5K-Ax55JH@W)Ep`fU)Wd`m@B7N^zdP4v^`ih*R45MK;j?okyqLUp{29`kWN
z{GDpfaCg+-=@>N#Sa2&Aa|o{Tdd-pu*szUD?eD9kCk}*x^AyL2mWM|DvJjIXpQcO>
z(3F5Ul~Zh*vSvWFK)NGMxlL<IeNeE*Mq8nu;?3`VES?ITiXT-PD{KmM3lMv%?jV+-
ztW1Ehj@jTetx`O?3};%+aME5l0UJ}(ZEy}`I2neM(IOZ2f4C0)&oRy(X}dn+G6;hf
zd_>;r{g8mqf#7X$fF%tJeloN#E33STs<XnWB+B@CCh}<5sQ^*zZxclqi{cm-#fT}2
z?z)9)SrrU3tdbUT^%w*KPT)XrUH`cO;&}Y|ogIzeIW~TmGruP?zd4iNc353V!)6xr
zbLsw7P`eOb+Wa^EEH+1D@jx4kpJf(LWflu2i-!P<BAhw6D%DRhR>voT-Ii$Vw%XY3
zV|GtxcGsEg&Q~))tkLFXH~@=*z0H5)U#IQSn4E88(q$&kVkVC=ne5~`RZ%V}xCN|e
zOa!Y7qp`Zk#_EyG>X(_-qfJ(Q^?AU=)n{ikCYRWlJc^mTn3+7*WO9j8+JO{oXHC4#
zJtP{l%WTZv)2({{E122iOlG_4j*7+Oz$t{0ACAZCN2fS*o0|`Cj#@RU!Vu}^I{poH
zc{E-Rwefl~SL`>L*Aq-$JJi$?51@_=^{q`$h%q`W8jq`NJeHWp>zK!r%6XLWbkX7-
zFkuFFbu=F1HXilq>6@6x&zn5vJga3iR){&cChrWw!a+SoiK0=`lqob-hC{jNivoiR
zNRB|sRwlwKC!)!t&nAyAbDjS;%i|PN9zLsl?8I5+R5T_BY)oFlOy0>%ZZw%(=ot)S
z$_r7)6H%8#(b!yTWAkgw=6%fOX(pS#x|}<4bvYc3$*he@9qRpIX0l{5*{*t6*#DWh
zdOR{3i+LN1-{E?EoLSs#ve=<mL_0ip!t8J{8jtI3Jl?`QKE*tK(d5ymtDl-UT^)_a
z<S{lT?_ef>%S?XBWbzmeW>1^klr<_rk=A4k7^Cm`-|TfZL=(*MHo@G-g82grW{W8p
zpN1bhVH$p7G#)=^<M9#Z@h{BdIpsXc&4`H|F`gWa$5U-Q{+xMyg?T*hUxdd^(RloV
zjmKXzk8d%L=a=&+N5vTad*XVW)1&cthK<MPn8$xIj~ALeCRC4;00$V%0tkz!Fiyf~
zG)@-&=JixuH=Pg%i!-CKf0m8?=b8N-24PA|tLk4d*=KV^mgrdQb|+$Rcy=@%&$aRR
z3iJ3r=J67f$HP2hU~qFc7AG1VVtqo^g})q)?F(#d|Bc!H0JD9m$@U`J2gMqtMPc+0
zo3QTYqG*g>Y-4oW&#J|k-I>wLO-B3b+F&sTGuUPOVJ&8gGxY$GqYm7J+R#f%WQT)n
zdl_(TgQ%B!X-3|S{|e1lqY2<Ln*iR!0{A!!;7U^fJTU+tlZi9aS43m+Yc>{lW)}Bg
z7Oyf`T-lG_J}te#AUIMACkJqB08d6T(m8N*07DHfiJ;`dw;r>sqTgUIIYiWB^X|bH
z@c)^0;4PPQAbTpMDMf$BL~maVTp>EaJfJSg&mbpN-&$0uSlvQC_>vQumqffL+CoFW
zo~~?xx1GQVz;R&v84k;cm-BW0pT(ZEj7yYA4`jjbVi1l4<SD2Ue7pp<>~iZN6&5IM
zp$_{YWKrzom3>w(y*<Hcvj7g1$Tvs^uqVrn+W2ew*Q2SyH*IRL8>_(-R)cRCYEW7O
zO}!5S|KKJNv~r3Z;yB%qK9Ul7rm^n^he(Rj+BOj#*|(!{e~pd%k2CkvnEP*;+%H4!
zvm>K|dGP8P1m9hB=!IwK6WKNahV8m&4Bud5_|werzRd8|Cd2F1;0$-Lsq>=6Nf?RY
zPB;Arn=_BmL|8|;yDgn^CPE9o8%-!T*@Ute3uO)q<yuoHBcKJO=Bx&(!qoK8+LQ`j
zF8sdW{0}&Ots8>i<`6uxZ88W7+r|W_$oHd3;})AVX0kL|SQ^)x(#U``aAaPsFz}lR
zki~yTlf~^eS<Gfx%x77A$CSmRAd8XgNd4MXjmgzyAlG4_G%}7p{||ZefL3L&X+b0#
zID9Yy2f<Kl)F=@hsNhH0%<#xSf3}!OVv+o%@_d8lNHK!vI9cwxAm2SEb4)mkQlss;
zoZ%UT!XN+M=!endeWy*{2eG_6S>89A^1chqljwQ8i(K^}Fg2-wBeFvj0x{vtoLVTt
zX+JPWOW_$fWx|PNAxIdCfDK!`uw#i+4<~sGQ-?B;>H*#<OobkO3XCq`WCa|eMQP?)
zm|kBW?D10&f)Nu#`u~l;gzt_f;UC!~Jf9`Jj3xX%Q^L2vo)sKIDUcG<$w@G2=njG1
zok#aPI$G8pfK0lz!j4xjGX8?SFPdO~Y!hr33-&M;?9GN?R~|yAl48XXaBxi3uWD~}
zRyQ<scn4L>=_i;#(O?RLCQqY<kzUC5(M|@W14hODa1Y8SllswP^f*g6$|#G|LNwro
zlyY!>smT$2IO;s)VsBfu7Q?5t!+!!xgR@1DKQ=;n(B%9@t^?h1AT>Ki=<I`No`?aU
z!vFiTIBbGu*9-BQagA~<e%S%YB$^IHWBwce8h<cajX!Lw@#S3Oajx-OjT(m|MQt%s
ze$`dN)Jde5tu$T`Z<}J+3P$@U<#G^*JW6}lOg+>Qy#Wa-7tyQ0_ug<s|EW#lzgnu+
zCFqXU)*i!;$~{&q0U>|>r?DKLKffE`zx}uk&dIC*bVqQ0U?_k*UQh`Rx|#bAJ^dW1
z8cv%5vigM$+(RrYx+8E;2FofqOxZ>X`ejB;3h-Wj$_C}<r>knCJ5g;~PW3^VuzzhU
zR6Ae%3X7*g@6d&kf3ELrN4fVh`S?eUxQ;&Q0r6$fEghCOJ>e@jSYIySb&kiE^L!C_
zT=tJB0Yec2{Hp(K6V!uql=T11g8Co)sJzisqX+>|p0`1HSF3{Z7l!hetpeo*8x$Rc
z?j?ru;#Pt3k`2l`Ur<tcg`vE>RiOOU24&llg7P{;d1dTS0-VlYwL#J0*56_%cU9BA
zH@eime@6!mIbsC;m5)K>oC0L9fN+{Bg@uT`6AR320YMW)E9YY5-y-TJG=CkV9+pZo
zc^KUctK?S@wlz*?wfT^|Si=lH{0yro89Z|w_vD<Y9Qf65INY7EeNnE^?QViQ!*~SL
z!n^Rbu;aqJoTL<@ZZq7GmA$8iTKK`Z)B+U((t!A-DgdmR=s{MDu8&+|ltQDbClCp$
zb_vT!2LBOFy1Uq<t0R~1&eFZN2I(F-{?fI`Rh)8iKf(&PTQre;#3m941wm&GKgA-s
zzXp+<I{qRtH3?3(K$I|*^af^Fa6Yl@keC#d!-AR=O;De<2`aH%2`a{ddf;C}P#gih
znx-ZNwI>UzF`A%aHbKp1jcQ^+Jye6B77ObmQY=8iA!X;mc_BtW!6IH<{F;xMG2&9W
zCB*`mEsVds6c(E#w|g<SGox|4kB!?F=5`))`$!Gkegsd8#$jg{|0}~yW4+3_(e89c
zJ}VmXCL8jFjQn6m{%8%zXA*J<3<#TYxIh986M;i;`C>2qB;t6;$E%>xDxUUdo(>A&
zDPX`m&&HEJfWC}*YQ>K#jlIi{_0fcNIezV%u*z@@KQzNJ;L`CLJ;_)jPo~o%o=7JL
zA@@`W#o$B3LeJqfk$IpKV>0k$MXp}xlfiH_8FRLZO;Ro!)>s3Q$x+F$FeuI&V^>Wx
zBsxUgX3V3;L`aPS6e(j$Mg+Bx#>5<){>RZC7=M}P##{`pZUM`jrEYZZ>w>!fB@UP6
z(ut`0(7EkK_q*^F==EN(3IPAP0MtasgO0&N|I!PF1Gv|H3*H*Mv-A<!_m2#Xq#&MM
zDh6X&46KpcB0A=qAxEuyI-;Vx&2R@`ghtnkk~ULG?!EXDb>?(O-z<T2a>NtB!cW~F
z<Hh&&L-1tqPFK(HAHi4X>nZp(j_X%%sdpcP$Msz<O>iH=S?=TTps?k{E$Wwt@!=zQ
zdBXRL`xF+4#dLM)_uXIP)2HDgo~8(&Z*>MR-vz&KQbX&j>5qdzo*RpT^XK3RVAqrG
z-@^mKUTUZ$S2Z*hA^fr73DE#)YW;u2B1G%2NRU`1QU93<)>{e6I5&QB&qpBl0=`4!
zUSx99qmolIBR|=f?PNhmiRLRz^D|bOc*Mp}=rubbx8q>N-y2M8s+E>BJCxwuM^mk$
zl~><GP(0|1K8f(QqUY|Vb-fl%lqGKi0Y8M)w#7SYMN^HE-R<!w9ca$3=Cbe|x$yUJ
z;nQ^C+(ds)Z4fTr=2k@`ba$x{;I0ht!whgn1%N(VPz|_`mx22vJW1G0g_PkWhPVeq
z+*?BwM)EUdBrJAF7CwawZ{WiBH4AU7vGD0!_zW(5e_gn&{c4P@yMK*P<}j4G3}v>4
zB7}H=D%^b*^${0JB5V3QCeX?R=9mQ96#=nM5V`P$T=>CUc(Ymf;u;HI%7q`og&$}Z
zzO2T=59PvFa^Wpz;fJfj-RG;8yUr*RX9BI709%+BvE-Nh7N*O68K3?YFMaSecJ-oD
zcl#BjDr?Hc@2C%2%W&EQ;Jk)3UpL{rfluGWOV$r3S0kKt3}-#VnI8b>ZKU~kyo^f1
zYKZg>KKut>jx~NEu>Ztgj+ehk*`0)>YUnJCJoD5OE!1EYqV48F@!P|N`ig{Tx?iY~
zrjc-+Xf7(J`EI1SBVLpgB;B)Wq-&%#C%T;#=+*_$y)Zi6lzTC}yvaQW2M!x?+~OUN
zZWbf)8+f&d?g-C`hzDmEIF|k{zHlR6Zi0)f>F?p+;X`xE8%@O>qS@nCNxLk|K!-g2
zewW*WO0^s>w*vt^0Pk`!>kPC=<0<AZ>T>Tko}P=vQo0*mLREduLr`EXdTgXHCgzs~
zFkgcCg}c&Z-XRv1nhB+^3idB+gsnGJD1R-%ULF8@1|s&=*A%R(Rd~UqA6nX7?q2xt
zv*o`8Po>;HE9g24XyiUpua#4lPlA7F1o#!8ysDs7*#YQfS4uERFb=B#!|esh_$0YE
zimDGUe+T+nk`+L1y8_-sORG~Fg1olCr%?x&M&d$(yD9)K>FR(9cMc*Q#7khs50(EC
z%!r&ryPSY9a;<+5B|!b{hwzT@udalD*u<YffLZl4cs~E1P;OLd8kzqHO}rACoT95b
zwmhOKsHbB`qlZS6Y>&_+0?=Fm`r>WSRau*9N(G~~6S)9l<eWorrF|<-paXYZV<Kt2
zZzAo@L3cHjxzT+T0wnqto%UyynZ+tI8}CF{=gxt9*}Koh#}_~`i{`i&!hbrOx3hG2
z2Xc`I;hnJ(Mcv&z_`%ocDR)`B`BZdu3*fOf16?kOZ4nl|*!TszJ9i2GVz$MT@Rx4<
zWts1rs_OZ0{GLQHmqf8BXX_e^?v=MF(aTCF{;9PdfyGg!&#to4_3E>);=L;<qSD=~
zfV^+4BT*Ph{RqX}DpB|{>(mN_s*J>uWN{o&B@pxyfT%DKYK{)a`k;^|coqgNazTv~
zwv<x(kz8g~35O#a`+<0}{OML%v4V_begtIyYJp_q6h+jBqY5Q4viQ{`433OmD+nWh
zAAxA9QV<kLH=Zn8=Bz>%M(#d>QLUargApxPq^_z}B1Y~$f-<K{i5yu>WD2nRhv}_+
zIJ_DCBB-g6xUZ(#8A<gALpcF0f?m_NG~JW%ZL%bkw2z=vYgoaIh+G&+zfiC{iGV~d
zD4xPdC|@5Uar^Y3LydK)Cm4l}iF!iO-KtR_C*KPi1y&M-B)^%-pMiHoK2?GIhG6oX
zGp`2n=P>zmnf!nz@1OUcsOT%9P)Geac<NpNUt_l~=h<Ju!opWnEONP&fl;P1lFN_=
zrVvoIoT6~~s-^Py&O-VdO!}L6N2J$QAiZgPNng*TZ(!2H6-bx-q}`kF3z}P<jx|zg
zG1$Vqk=%-y&uRpIh4`X^psIsLDqm7hgJ%R~>--i(Mj|RgbYvw&XDix1MCX`KX&`f+
z`IJVm)V{|D`A@kvALH80S3-V)iRMCh;9g`t{R%$4SUnvdoJMLVl5kPujJ8*2sYN40
z6VX_&X|P_=0?g&eq<>ZrIuEb8slUiI^b*(5s79rfh(LYK3rfznUd6A*War*w1eC^y
z<Z_G$!Dj4_-X1d<BTEjE*<g~{?$EJvV<e^_GRK=_-dhV9u_(N;DCbA+?YAISyxfVG
zyYccPyxfPEALHdgygZDTpWx-EczGNzPvPaac=;1vUc}35czGKy+usV8o$>NfynGxl
zpTWysc$tQmz2V~Xj?ivVl%)pwE>$}dfp8J&Y#|?y`$sdL#5H<|=UX3mj8o)%03}O8
z{~+GU^#M|$UGV`;hrcJL!?((X0e633CP7Q?EL|yVu>ybP0q|wO3Ve!P>*<cjo{n`b
z$qvBpvY8CxtKrM;Yw;B-<8xYH*eYN*?_Kc`J~I9%M?wkSC)KPL)aDaIgAW?S9;48g
zD*IG)faei;9SP@3aunNg_BPwVeg^zgMi%{Jyfd9mKGBys;~d9?8!JvKG+}m7^QoAX
z97<LU)=y+|Ga``dR~!pot77r<O!H)>IWq!HJop@^cq=1W@Pza+>Yj>4={bsDwsJte
zi3y*k2)ic-sALI#^{QK!J9~k8B1gezVgcd3Lm-5+8NxXX;nV;K=aoUw?IhyJAdvng
zO#iD)e^UVcb9eCjn{~Ok>h2dx$2`<$|L#ziJL^Dv`gJJXGimr1601R6*D~De815IU
zgd0&@VvysrzBe=VTbTN-cqh%^ZE!Em;O+RBZ18)94c^sa6z|{?@5DRb;%t@XjmW-V
zklo29<)V8e_i(aVXq|X~K|NRos>==22^KPe`H+%<ugWX!1r?0}K!xu2X9`>?iao5*
z<vxLhXfqK|>AK5(QW12Y!aHx5`zw49*yWZiUG96(OWfP#-Uht_IMua4vk``v?x?#w
zLu<MkuJ*wBJJAXwOo(ZheeHaM;AXMQp5l{ybcX~>dsei<-Kq7E8>|Pyb&zkZl*J&F
z!8pVqKvbU8te>B%m|Ba}h^aF*Q#=MJ*Q^<3qk>;US$Yed_06YYW%EyE1H~Xn`W^Hm
z7H<_uiL7IZ&>cynvLEbHSS<^d%4oS%q^oh6SdL|}ob|sU7GGuE^m$cTr!uOu15n)z
z@A*4|lhNe+wvkHI37ih)cshYEVzJ@%70@29V^~`>EWDFCcrM(t4pNi+ps+rBsZsC^
z+$1-tIfl;<Wj&S_pbz&jWx6AM*sEd-L48=3SO*7dS|Q{XIxBBc5buM6WIu5~mKii$
zevpYhgm=_b&JB=$c~f~5-WSrZvQhYI{|To53#R|&0Q%Tbxi7-u6JLG5gr`wnHlO|#
zpT6RM8a&x|pGBhKx}n;#5B47z_8%Gc1(jgG?uXLl66`n3rv&>=|I-Sv|5_vL*BSO3
z4Ev%=u;2DWk+A=6J|)=i_@4&DjyxsuHIpA9(tu|2&S~(Rrb={2&1Aod%|w_?JV4V5
zAyzPkt~h3v#JE!2Vm4!^Xq?lK6C}MEww<yErGsiC7Yo}IbKix<UkyO~iYH+V9j&;2
zH^ZpB4;9LFY$PGMmu%BP3LIC{NX30mD3D896p%Uw(o&;5AeD4noH+AOuHKI&<EY8q
zDHO}j_$#TxE{tVijaWb^aX1(^Tuc{+ASf0b@AG5&AY=MaD5hsPS<i<VQ+th=L@+Km
zji1S9bH>4!-5JkEg7L`C|NXqA^PdWjd$Df%YM}R3d~_*X$j)*HhbGd!41cS%381vq
z*LMGbZPeR#?}ZK;Mx93B6>7VmtSFpxc3=QcR(aI4ON@ed7hUe|VrVUo7i|^;&1MXV
zKuSS()U#cU9LNs*K=@tujtAkZ)QA74DDY2#eWgBZ)rucbiI-!EzFAb4tEbFLst=Y8
zE62FBL||hA?0&WmOv*(nf}VDDGZ$9|aDhF(TL(Y*dfe)!<A(zU?Ya}(E4Xo%{%j?E
zOONb7N2b@z`V>Q@eJU!GFo~UbV6z7g?0J{Y0yL~$Kxz-4yy=C@eg(lDz&ojHLvSy3
z&Be!5eN(CE&QY5`)i4*C!Mpz@Wos3JINGXg?q~knSjtB27cFsvVx_cq^a`r9#7IKy
z5U@JR>c^c7F#L*wZBF1gfpJt`QZG}K(6vVbGhcw29|eqFjr<&gsqC&$1}1b-@?<QL
z5Um%8)`udJhCwe=o~jVJR|Ob`2+NeGGk^^Oz+)D*kz0M9I+LLE!sVF^>Kg%|A}w5!
zNPb3$IxpC^R!G`_p3jgjz&oR5mQR_gGoF`0nVxAZo5{|LwiPUZ<**q_wZv6)XF+pt
zoA5FhF9+bI11`R4jb8WsI<xkz0BMJ@#<M|^_RgaIKVU%`Fag)Apw}~&8}Lrr)bGN*
zw5d1ZV<H!E74-XD;>~#HtGKPZ3VOIHa;;5wGN`-CKuy3ZsNYuV?Z*cdtZHm_g=Nr(
zu`H^nTZLZST4tDpKFTqiUKIU_Lg7A&cTxv%MCd-og_`@2MG&^yhve1Xl#A!!N(W9G
zgn;zZa>Wb`0;a9c=GIPI+}Q!aH`RagxxQ(llbWVZ`A<`){9_}dTimZ8@r&_t0gO1i
z+)LoL^s%r{ilK3&C{h3oSm}=9NYR~Kw{ZjF3f_uw2Eu|<8Q}$}Pzo=g3Jr*Ob-S%l
zJrJg-15_wQ9T0^^jw`sDyLJv5h9~>gS$`dKqM739W8k2^wW&<9$$@MO5Sn%JU~ymo
z=I{N9zSQJKrvX2P|1_+HMPm#Qpsk~OAM$iJ)DU>IsqmTs(Mo=7V_lDBC};t({s4Xy
zopm!_K?s}&ZEzmup%sNRAUF@z0OxsBW3fLG0_R~HoENYdQeg^tKyV(Z0gflmK?t0m
zSm5|VArO{FYryhaB+>lTg2jxMKyV(b9*){sS7HfhzK`3W7=aWB3I$S#uEW@4`j+$D
zpCeVHfdzp3g$<k$i-EvVECvs_QmL*L^PfP-WQes5YkCg4c2kIF`U|W>u@3^lr<~zL
z0e0>ooIZsphjcsnd}07Xl~b&rLD<cwA~1lwcr$56rl~Tfuo;?W6OAqWl6oqdq<&?S
zlo42hBt?NGJd#@Z=ys5urZtzsgt-Z)O#~2<af3ws?1Y06v_*aj;9dqK*YC=RNc5JF
z8^-cU6q@QVMA0#H#k)yZ2a}qdp2L>5H|nPx`wf0t#gF};;$Qf05(G<+&wv+Kv@BlU
z9-k4Lp$igk$aiLX-kDkcPE5uVQr~RwzoId#7YxEPBM_b$tDdkI*J_f*hT!m0SR-+W
zflWb}@DCkJs>M@XkR5>nOF<VX5{<~}g*Z!8MNp|wqsU|f^^H{Ak`b!xv@UXKHa{eu
zLFj)S_)2^y`qBztLiz+4IVF9c0Y^fI@=c01X8NKM{Q9R6Kz(`sjV&6KXmX`bDj7DB
zz`{od&qZsJf3P)4Ba{j?NeZRnX_BR)WYrLp6k2&J^)|q*Dzc9%k%bHE#c0BM*(NL_
zs0#^;g1UHwwQ>&HFYzx>ULY<+4YAZ@pnnl$4Z8sdIV)@wDzd-L2iq!qZ&GD@RbC7}
z!GDJN2((-Lp2c?IdtN(%f8#&HhT$VWPxu7?8DIi!8Na8&1lluvFCrN8HDTYd0I+fR
zpB86iv@i$>f2h7>fPbUf(4M6-1x%qq84+YPZcVNC_sy?F)9u%6x^2W1BHgB#LLS{N
zEtQ=uCXv9Xi^w*spU)yv0~imp=;!v0XxzSK<JJgrMBGx4BQLklArnnY6S)K=-hhCh
z+&kc0twR6uZ`m{L95zq=XS6EW_7)gzht2zqU{6#f6zs`cC98yCH9KrdwJN9DJYpOF
zE?VlM4m9O%k6hX35IfkIHv&-+^Aw26%Y2`mc@>0j>kB~~H*t52CV=<a1YiWqA^}ja
zEUy4g3l;$QcEfJ#NDd}P!Vag@vWpUjvwm=>Z;%F`xolc=xRlj#!sPV6XmZ-gCMP5A
z7s-j@etG4j^6ePo4{9aYL6Ix^gt5MJG}b?0W8H{KMyyj*GB4{ZgH?oB$LZeI=URtD
zb3YV~^AFoN*D*mTTpDps;nKXEf4>f#>M(O_O64`c1g;6k&BF+$e~7ZB1@>n68wIzd
zE)7;9^x&lj3L8lj2C*b87+;09l!A2%FP-&6>9o_Ea`FJ6Y&~K;1_7l{Ge}xVg_Le$
zMz5Atk;%!v<<pNvll{kSvNvMGk?bipoJaO6_k`ti5dSc^-#8gPA`;6L9!cR4DQG~E
zA<={V@{1lK(f2~aE$<);Ya%ZQQ7$o~aI<4<Xb>+eLl`=zZE-S?;8KoY)ryNTVHd?=
zlD&K(k?BkMv$&5-UGJ8WLHF~i6fB&hRd06ondz(DJ)+g_XKb}=gxaIprBHj`+JylQ
z)b2<cQV@H@jliH)prbdgW>DVXRK9@sFpnd-1XVSp%q~!VZK{g?bT$D&%w_hy=IODS
zehq+hVk6lcE_JA$O;t}}Fj0WTEz$5Go=B3CuVLwcfQI6R^>1h(tmCE$%(+(;aM}cf
zkVgr0%o-<I+L{_Smn#KQ*kO%;!a-F)aQVSve}6h<<ai0u%c;@ya+*yqji`#GmlRde
zqnD-I>)Jp)v|u8en<2<e3Qd=)2B&da=UP0|I(kQ`9al0Mpd^wFG(WQU#s7?pE!asy
zh21v5=`#U)M-%nFHc=bFBuUg1Owudr!|RsHi13gpI*}PFj5>6_4$}d9Y{J%=0+&9D
z0*9O=kVJ3@EMjjqn+6TVghM`w%#J3JIW~zHVJ=A`6z0+^k$#Ir7Kn|`4%p~aUCRHi
zbPg<+j!d`k6SVIOp;CuHzDLW<@NNE$2iktYnigxDBfKb3sY^%mH$iz2<iwYnGYsDv
z9qvyidA6G;@zA#gM@`sI@md!C;Zb8XeTq44ss04`6~l}v>-Y)-wuP;qz_pMXj7y~q
zjp&Wx{x_~%sBOsrSA`Z^Lowp7QbVEmtD>O<ra)}76{=koC<-eTN>Nxnt#2iGIQdmJ
z;RHt8<IW*g*F-w&=tyj0n?Y=&iET3aO>d1rZWWj`Lk3#$JOj&MjWiv=tkvs=^6)9-
z0xI-z*?6)$4!Q+fcLR_O7XGBs5I@M^pgj#4ma-WrDV@zFARn`nNG3*Lc})T2U}S(V
zl^TIu+c36)Q7FWHDniLP6oSD&tjVGh_ZAEAs~U=U6hz#POCJ<Dm!!cy;6slO!8AB~
zFk{VK#J>oH*gSb)qzJznA>o=k5@}4c8@BS+Q*P0Ml5VcBX`B)ImNbq+-+DFf9@My=
zEI6}(i6mpuUxdM@9OZ>sK^P5-k;X8Lt=3aI3>qLy3#$hoP+JpgRM3K<q8KA(b=W1T
z#;|!dojuS4@Xw0*Mru8fBnl=0qfR=Npv>=}qo|jVYXv3*BdI><T(}r2BJb1--AIq3
zJVz;u`Meak2TfhlKXn=GDx_e7g)O8#lk`3^ikX7K=YQj(^4+pp*~v-%9grCjPUOTT
zdvFQ`)Fa<4G9iX(?wV+t+iTNYBl<CEE=51~YVNhTA*Y6K11WHx6IKM~Rxk+8s}U|4
zxk`FW*$H95%%Vp{$U`<Hsv??xQKoJv6H>>EsTd9xG&MAI%94uGuusKKR}w~L#D9kl
zrZQj+P;J(w)1DMWGUs-FFk4JRx^481ko`E8%>#Gw4CbGrIA4K6-=I^FS&agU6i0Tr
zCFUSTCG1Pa!AMaf#<6Fkla)L2UK+p(XnYpB2Ks_no4<59{GHDhbD+qHez5Xh+P$M{
zpyH6VW1?sak@tuc;dN)l;>DdLb`NhA%dgTy00oKd%P<o%D;&!JRYF!c^hvB7;x#Hj
zHPUfuqpd=VurKB+;<<tBJ(M8L=YEGn9k<&}M{AfFTf;QsdQ-!sxZd7|c{p|{TccsZ
z3WV&alAuCd6XViMsk)W?NI}cpQ_7u8{xwx-7o|dqAK!Gc5KR}>*>u4Ou1>l@!PULG
z0B8NQE>L6UhV0a0s2!C`7R+HIr3zdHcdPXt|5CEl_4z5V$|Gh&!B$k(H?4buO@@VX
z*besZ{;pbv3$pMdZ58Js7j2n+BunmpO)jv4i?B9>I54E%+%Pq4V{nM1|BZ_sIYKBw
z1$lvVBy*DUCP(S}Dr73=XKFQ!c*W?}uTnbHy?b0VwK~D3Rz?tfQY#9A?@=q*u>Cg7
z^nx>>OSh|a(Z@Q~q!^B9dO3K$ds;NPlx%V_g6or9R^y47D&=x$bh+gFve3&b4O$E{
zvsJW=3N3;D@H|-!-s~PH_b)|bdW(%|BZfaQO)>ndWO_q%rrAD0y74@$k;rta@|0k)
zFq9Uf<+!-@oQGSehF{(EAs*{rh}p0GqB%dBXfCvgMsFTcLIDyDB^2<8ru6Js`{CqJ
z&KMoTlT;r>Er7`lZdt+Bq+DLez=U0P&p9>Wt8ql8P2u<~H}49d!{^+W2h>WysSFIF
zAL{L5WKy77`5J!YsTH5~s*po=UyO=rJqAI6BEy_-^oeAO@lSAmP(2*e*FQiq6s|H1
z%9(S3BRY|hK;8(-9a{wo1<eeDLJ@Ka3WfS7D0fv4#T(IIEjI>8hQfD-!7)Pp6P$Z$
zfa8hl9|DI0iH5;30{#=6d#i^NL1*s6Qp2F=P$wDInftd2lpoum7}5VpDi3TGC=}W?
zOe#hO0D|(+R)InRX2YNu82|{%Bh^Fk#`+IYT?%y@2FFMOKyV(d0gfl$e+V23x*G<^
zNCZG|9;*S4C+NR|6L6sEaV$9uj*$(3;7~Sz=zSBOwkqvO9X@S>@j@bQ7sWb#$;ZWj
z(B%XI?k^Ca-Kmly7>Ds;q!l1OD6PO&<%2U3xKBspgMu)J@nPgAAU-HRfrpQkJJTvB
z?C2%%5E%+MVAEu<-$@i{jcx!2>8V^}9ZgxuySZ*E|Etp#2lHruZP8RzoI6$nI(-7i
zmtZH3!J5P=9nt|Pil>fka00@dCodC=<bK$B-h^5M-ca#$qBoV!uYpxWXH5%mLfAU8
zxn*)GwiTjG<cIo)`e=!FjYDYb`?ABs*$gbR(+UuLo1-9&0zb^#rwB(HJXA+#G74CJ
zJP!9LO<xL>Mp0xw>wq_qUWy$7vasj}pM{Dbru^rXu<OLemXdNSZF3Q5s48zQ`sV#K
z?5%F{Up{=KCi+=j(_u}tF`O;4V&x)$Z>IN$XpP}dHsfUE#GuANIWatq!7^~J-Z)8T
z=f6bbfnuqLHE1Kb2Jt}2H70<ESEBLonyns;EFHuHW$EznP<pj)$+}c7H<X0U5{O+O
zH5zoTm{vzx`-=HO7Irz%cXAl^yr|)#_grf+m&21a#xG5EQyq240@@oGv5_o9$HQ&m
zdacZ{WHt^yGq4-o2R<Fp!ka|9g%+Jgj}!=~H=J-%;9f}q<>c{NXpBkNz)E?3C~J4)
z((J118Bd=B{ox(xvG7*3%A*keVG3v@IHAg;1Sg)#i_(Y|Xo**0KAON+A>OjQXG!Oh
zw)VwwJbvpl7284>(g^)O<tPZ_(ny*@TvF1MD!FvWi_5ZIez7m(aSiavVW((vpfn9(
zaxhY`kQ^ul%hr+uE)3(2VSG3>3=2TC0ycCsHB#wd*+_evGCm-!ET_|{H2KAn$FENg
zyGN4)WvU31gOTEe<UlE2s*?jW49^PY$D)Cw3>snJjNC8;j&j3P2d+epBv=9Adw#M<
z+anr!O8pUr-pC<C=qZPc2mQ(oxX06$QW?wi<~KP{d-sjio-y_VT}yh}n;h92q(>XY
zR&O$!%4a4QoIzYzOrUo(4En&u!dai71;-@rtq(zL74pj-Ox5|j5!gNlUlgY=k-<Tq
zA`Rh)9Z*}|vd-h)3~XM*R!u&YTSp2UtLqso;t8T5*v(JD@lY7`<sBD#G1!Br&Sfy0
zPQpevIw|!RDW-#;cFb9p0+-^tRr6utx7Z7x=ZmS;^@D}NNWN*>v;le=gE8wgtj4LZ
z`{lwyD*Z8y-px;&F?;%~DW1j8@f#5C7QAoaYmZZ+DGjBH2~!#)?GGsprTy_JP03a9
z9U$bYgE=!t3W;7gW<k4F;B1o_fa57J?uLQDu<+#|CzP^&yUGRP;PJA*=}(O&2g;Tc
zCI=%g5y@eLv13zOq2z!rBfNfI=Dgxpit!k%_<5cYjb}^_6uQP>BuOHkDM^yI?mN9a
zOD|V3hMb7%-8X>MfMr`u9Tdu{4opUQk%-k<_>qTI%V3KFFnN+7$<-a{zx*=VRj>cD
z-5E6a7?K|6c=mESX)S@nZ18-XUUov9LTMTct{7gYP;!`n{dV+f6JDUo1*mzFa88<}
z9hb({Pno?hJ*>5v{-r$mK!1%1i$J2BLp*IaYQqinZ=Q#<1+@?u;NLvgrnx#uEZvdj
zQg$n$xq<%82igkN{>_x`iVCH4SCyJWaD<}Lep*{$0Ij_3;bfpM;Zpc+-Xa8RtTYBR
z@lG30di#u$U=dGU_)&D8u!%?53Rz-<@*qwYY0rw%ViA<3TLsD?HYhrJEv3vNDBW8H
z$}$@i9lQ2q45>>{da8%wO`fG1R=^7TavK~SxR&y35gf{|<<XbYx9jHTvc(Y?Yo?B(
z6Bq+9+!Q{Rc>2}pO=mHn7@Q7;q0!KAaafIp)`*s&W7Mya18f+KX2jul=~4wtme_K@
zGquhLs^zgocpKFog?ie2S`AyYV|AnG&f^-ZZ=|^j1!=U;0C}&r$y>+drDS3xZ%QWS
zk+-i4u3&RM(Tl#Yblf17i}&GTv*G|ZHhkFoqG3<ju<LNEl!%P5Qz9}C_R>@AzabA@
z;OrwIgbURdbP$!4g!<#2Vg`rQoz&5bFgLmr_WcN32P5(voS*~aFws>izZp;ODj%Wy
z2T3e=<Tfsv)k@5WZUD{}fLn7ek;!8#Q7-&&dM-}0GwGd99Y&u_tU(7de3h4tR(Y5q
zEOhLr!xLY}c8Aikc`DDRc)|0I(tr3F^kH2efHj~K!+cwzSoJuS5{QwaosJ(x$4cW=
zPv$WHI5mb_RNiqaPG#0W^bU~{a*aa`N*qO!A;k37Qiu}kV|Jv%E$SobVX1W)wpR*p
zTw9nxJ)+IQ<`Fn6p<5ChN=m?gwO77Jy)~IOaW-&#2+3x`Ey+KAp$5t#8WTm;awP#+
z6=#|t3%^UNmYG^>hou#k7=Kr#+g6R8O6VZB88uY(#g{`Wm?<Z2!HCQ|bh$CeMd_Yt
zQ+p$y9H~9!lPg#IN<eT!$;E_o-i5)lk+9Ks1q{N^wc#_;&k?@!@FR&Ypg+98R;H0|
zj>@ETbLBjT_UL;^v!wjUW!e{cGD4hT^VzD)%7v2f24AiuVLGq*W&BuGb$}?Zu(56A
zts}Ot#E%5qf%BQK+X~gPq#SirDCMZz+OoV3kr@gWpbVJQE!6HC>FkIBN@wR`pmbJU
z8$?)vEpJ?J#Zi#x&4sJ`FdPZYiIbpkAk*m+!CQ3#T38FECDHl}oaF>Ni-Qv=2rQ5~
z{5i#UK8EZ>bpwM6{pN0Q91Fm@FbvaDaMPDB@9S?Vm!nu%ksWiWPJRtP4$;XU*radd
z&Lipn5I+jh$-8WY8aeW)P|A_#sZ!stqr!N1C)vJ5{rOt*{fNP?eGk~68tL>1Dy7r&
zKrMBP*hbKEK{wP3(M5#zDeW1n@>JmdO6(Tl7_%BnLI}VZjbuS_X)qA$k<nkcU&g~E
zaQZpuBb+eEe23z@RBjo*+o-~lMf?Z{F%ZNI7B+&S#{Vo@wLD>~79(9CRSTsH^j1qz
zR159G52unt7_yl(MpX>Ns?NYB3s?(^VFYXRxgwnP)+e_A|C+;9pF41G|Cwkt_d8q7
z85t9)<}NaJ6QZ>J|BKrHLb(iuDK91JTC?C!(pLa4FzUmWws-U>WrY(7xD7}L+J*ia
zBh}~&AfcVb9SAR^vv|o?w?+m@s@s?GBh^_1%==!o6{?-5DNiI7`kF4(J3Om27}SfG
zA{nIOF#yi%HaJGUNP_c5^>8TNI-QIVX(2-(y=jAFWR4_AZ`A<F$PO6->2EejMgmEK
z^mYx9jO36Zkp6CiWaN(|Nbl4D$;c0>AO#F<|6zk<q>v;?ltR+ei$$@8pXi5l1dyEp
zMgw$~4c%AqJZYoTk{MM;+H|%_cRNVqV+@X>4=MkOTq&-vg08`s(Y^}{58F>Ra#a%7
zJK#rAx!&mN2HnL_0%lzAwm~scS`rjWX{iiWK&`&lR;b=VYQ;oKR48S&jLM%p`8ggt
zNFPS9_F?euHlFk<FXgx-o+!tqho}F~uJUfAoFpX?h=&~wsi)4zz~LG<NZ=Ia05nhx
zN+cY%V}SvY#<oUWVbHxtfD#0BeVAr4wBhP)#6DK+J@F%t68Owj`K&jn;RZmV+<IX)
zL~kQf229e&*;@q)W%vt&qPGz#2_`|AvsIu_p209EdK>X`TtUs%LvbnHrT^4sfFY!e
zhhcE^HX`M}B*{?zOOL+%AKpeR*KYFMuYMb`4Xem*q%pZ<X#3RLh?Lrs<V~qPJ@U4+
zPvW4OZN$aVuw$0SQ0#gqiV|fKc1o04CHAnLs5Jpt1B?$QISj?BccLhjBVk>KA4SJX
zJ5iLp(bvqR$JFG&ohZ){ec1I(<>ZDG>~>V%K-HT<oxVeD|2BJ1rcijPs}>Ib&u&2l
zPkc90R!?_FR82Wx_YLzaef3=!)I+d`y_V8{E&%sEJ<5*3L;rwZ<QIJl-Wt5KRF-p|
zeC;_F1>~G3-=9Nn2S}9eo$fH}C>SQueCmA(OO14!^v%z3xyYNa^vAeB<@;JDJJv&S
z%Bx9VUyEO(om;)7-kVO|eF$BX9*3fIg7q$!@<Bd=zdYglrOKqX+9MwXdl!_nNg4dB
z>5qdztQ-9aWAzY+5@RCv(hM`nw;^7Y=#!!tQrq9+O~LN|5eo4nh~4A{&^0n(5?z_M
zHKy{m#wyX3Ua)?u&qttYB)TN3`$y$gQdQ0re(EpVsr&L;66waMq%moOpWbVBdS>oQ
zBDqggl5$6$aw7OhEC0N5m6FZD$Ww`A-RpI8P^wamvdWT;B$HI~cZ`zV?P2@dvjuH$
z39@h_FC!Ixqb{6V@z1H5!lg^M;A+~byGxA#jVyHp@Ou>icGU{*<7MDJ2~QF>Ya(Sh
z3BM-Wy9eG8#G5rls(SY`Wh5-7N)~P;=c2-IH4AU7v2Y_778QP*E?m}rRrS1ojZlmn
zPz2=%8j29&0jhAFhbgkA&%<x2mRj+S2;5;3XjcTp&P?ROjT}Z)_+4h<i)$?0Nc=;E
z-(wcOtj5BP)H+o7y=LKutHO1Lo5-TkL8vH$4H39s6JQJTB3hW2{1&FmeHowr6)$~&
z0c}r8s6xMjRAo)M_#LUiTD&7T4+Ow@4Qak^!g&LqzKNHtA5N}DIO`bBdWQ2*0Gzjx
z=HKx$DhaFc)jRm`A9y*|_=Uj!6Ms2g{vu^}5|XN+Co)n~knA4OU=^b6=0fq?!-dRo
zLe$q6YNTo8Xds%8meYJU(%ca*N(z$hSvAr%G9eJ%$12dR3!r;pbh;_`Vt9Fzdk&5n
zHXaNY@6d5GjsD%6&Vc8nw{(YiO4gCoJlKC<Iy91kbMP>dlEeC&p(H7sTd+_uxEC{n
z=b+fZJ6$~szYX6YV*A;~P+nrOcs!rZ7V`19ZxrSJ0E-jT@~eT=M)y&O{^^^PJO&R0
zpLZzI?wxq&os`^-4=#WL7JbjX5dU+lS|9i&%zhFi&y7VXM){`@?~m}y0P()vqsY=7
zRrg=9x&_&K_Z@uUA9(pETx3JqW;y&FJ~`SRU!mfj(AvXB50`ztqn0kW2cNh*!dH5r
z)#YORuP*m}##4+W*X90)@$_6O0RsZ*5~}LLcZULNF%LSJ{C{By#PpK^OqXDS=x5r2
z_q({|3OiC2<jD%M`w6sncYsbr?x6O<Z>eQ9GVEUl!2T4LJPCzcqg?l4kb6`Oau$P}
z%^-hW0i;*9RY|dh0q$8fKpl_0g8}{~0N@N1phtpT?i8fefET~ccDZ}u!^ZMo++Rce
zx=+JJlq6f@QU$-tA%Tuqy_}&x69IZvvRz#xI2}T{m%;t60yuXj$jT?%y-|4k;H4kF
z_ULV}1lKB%?!Nd<Rn15T<sD(5zYhS4t>2wx0-cTYn((6fZ#9~jgAeE8B~U$@%YUh$
z9>+3s2jDkVF?RxfOE%&}X6}!bm}`OJeYHT$%|oJr%(Z%15V{O2b~rwB+p0!dVx*@t
z(mz*1I^T~pSa1>xH>xdAZq%vqxs2vKM)SNzLqkKe{Alyyi@4;kaLF%>OUZh|av7KW
zk}jG1?knI|za_g0pBQRdm65mx2}T>3i$UW?F{*h02cQ9?n)DU$oYa=?XdJh<Z5+oj
zigUc+&tu2UsFYbO)i>&q!$uw1HixeexuzP(#YBD_|Io33MdHxWE#e&x9|PA|-@(He
z#PfOvaX>8~;EZJ=9S0BKwSFL@4CLrgAjjgb1Y`pPSx^fQUgD4UCJKFn6e`CL>3D{8
zLMWt@7}Dn$QfDoY{A<(|ah=S#P6@^J%Y&4tHZraQYr$1W4Z}tgIc)hQbQ&Z1LMW2c
z@mH#}&5Wd@79<dEELETwi185!9-l6Z`T>2B0i78N=t&mMSqx}iEr9TxEhKrBnR@(S
z&So&@go2rMoRZFY45p@ff>xEx4#zQ_rXR`qjO2n)B<G#1kX*z_nrjgYCmQqPxR`NV
z5{lze{FU_OGRDzb3y$HTqf$wXJK0Cs)%<9#U^G{TqPdFEe4WwMCY5ZQl0O*xFE|1d
zHu#&NIB*Ior1EXXQIk|KBoRjbNn@kV!WJiXKrmpSJ${Ymnou;?;jdIj*N+`dZz_R$
zo5=MIRzmZgP&79(nwuC+P2B<pMHR>G0FOvV1;^060{T7@pf2;~P)IMbjk%Q}&8|g>
z3y`aH#E;=N#&CNmh9BauR7rP?9fQ$h-^Cd24#n^z#&9oVsHuDOgujIlmm|XF#)pzd
z{oK#EejJL6<_{#B2N~D=TIwe;GLjyR_d)(>aGZp&n{aB!Ut2go5gPMwD57+Ss;!?e
zqV8G{sk|zwKJcM|lln3{nM|zV2dCTnPeZ{yj=z$)e#YQxYGEQFnm<ONAJP*@fLhou
zLLt#4mLNUFkd}@`UOEM)7MdTLme#LAp*@Yi613kiv_orw=1I$`F~!kJaRzBW!e@{k
z3GdlZgtQDq2%lqwb8A7U0^IpwXhHoU6vkJ$Q~MLcsL49e;MuNQ|ASOWJkN*Xx{z_b
zz_^ywA|ApO9J|vmv6mRx%c00#!C$H3US(tl)q)I<ZAp`0m{47LuZKc8qE$h8b8Jw+
z(HyfPmwSk~f5U>QzTOUo@(_+)?sUI}gD8^9I|_>XYCAj=j#H`!uVxuQ?j3D=C6p4d
z3~*1cqDXf%FOS*k!SB@QUjeue3k+HTcz`KAh<8M(r8-J6a_)~~bh&so7a-?{v1F3-
zBY0PFA>dKQ@>9lQT?p`V9m7P$G6?)gqF{{p9%mpw3kC871Nj95u@0{cAVMhR9#Ky*
zhF^wa_%&mAnlV^sZUzR3kqw!c<5?Nk+h~cuWhBppBKaL7d5)3Hs!~O8^c^o3#vd5M
zA44JhnIZfSL$Iz>1q%b!)NdUB!XRD<1@RJtc$q<1=V}HB6;&%o&Bc%76~^&uD2^--
z{$FPt)z(K}0wO-dA%m75!kY}?txyPW<F8~9|IQGqwHnyB_b2i&uvH1h{8;qJ_8*~G
zw&_z?w#7TLVAWQOXsjwxx*x~(SSnesci~+{>-BENu_NQKx-kZ+hoAg=nS5O+`S&yV
zotV6}>o&>fAT%;`+r!yHDvslOKNQ_{?;Hw6r&am@L#ej4!rB2?jAXgV_z)IJ)v;?R
zh~4m4QqkQRgw==EkO!)05;~MhV)$q%f{!tR{~8kl?9wVB_(UiIhY@^=5maj_PzFba
zLD(Y{!e<!5o(#ddp<z@*?@+<$y6UkkQi3U=)c0cQjZD4Tj{L}CA_GRM%(Xej7^a0{
zn86rkGKM)-cAv>?5rY2335s78w3XZ^6vcjwVt+=lZ<Q!$Y-RKqv#~U)iKbBcbD4fK
z)33HBh<+kHGH5i10~y0Xp%~^dhE~R4U70s@kuVe>*i5Q`!%WPL;;)c)2GbD=W&wj)
z$Y82f3j&jb+$pfC=tr@LQ7jHc(Zwj1FbeCsuOSke>m>%@MDGABPbK4s%0R=b(1(Pg
zS;lC37)^VXN{aRhvOdQ#tuw~DLOcm$&q6nv6`^=mGM>X2Pqhj|!yE`cIv7tBU<m<}
z-&7LLs!%Yi8O#w3W>J;GDddV7NVf%HQL}QA<PGNPU5VBrS`&(@mr?aGs%qzKM*g2l
z;z@-98DJoT48*$TWN0#FAsQJ1{ef!o!*H=svU6)gVGJ{j48y2)G)gcsg|$IAMi|GD
zp*Zr4qrf<78U8@Ls9AnV=uzOhPy|OYf>B0Ly@5+4VKFkk4-5hPD71ba6N+Ls#+9UD
z;08uf?KB$fJmkBn)KZSea!E5z2t{%rBRPqYRNGQy`^C5kc}Udd7s=-t$jPBVni<Hc
z45ZqM8Nm$`h^$tQViTh{Efhrwf2E2!ol#UfNl0aJS&xwTMWJ={j8G6~GKeoRh-wY|
z&;TSuNX2`H68U&<u^;wBj1hqbvn3cz*)rof3XJ>uxGXcCkL3j{Gw#O=n{-E0=jpaC
z#SEzHUuc|-MFuT2cA%R))%RR{CE7IIq(w(LA!u~KBI5x(#yAk~Do)eq;je_FbsTV%
zuO_uKj*d_q3mC^j#$i>xGI<Q;<02_v@NrMAdW%EBbTODE48}U}Dgz^s(3q>j`qCj-
zI@!wZP&_@1XF20(t5z{tS>!Tkd0idKa8`!GIh^6FVmMZdStcF|M;Y%QN*i{2HKRBp
z6vY}wkzf?o6`nE_>mbKsl5(d6K<Q&B$xtZ$3}t|!RIj|)qY9x2*A>HV4lx)v6ik}I
z3^N$(Vq}?I5KIb$Bi(2WELp}f5{f0qSn`a;+N+gefoNsmv;zHWrCzNV3Sxim;MX&V
z>eZD}Rs^^%j$)bAhK~+KaV-8SMqZ4fdUeIUi}kr7T#v*h>q-JSJ`~D{4CN$-vbbt<
z;%Pbgk@OIBLaAIIW<eb=V0|8mkc3VSg>^%VlF+FP>)@(kVdmppwhzN;!GH{R@7HH@
zYs*F<n;6t-p`c3ms~DFtD68vAnFbGFHWZ5W1%sjrKCovnqA!Lb`Vu2LixJh(84Tqy
zaYo#O%&$y(Msap1oO2n@c?_p|y<Q9Htm1geuNRv7TEp6CDCdVlxsah;#89fYHU)^$
z9MBhBj75?bUlI!9QU-AugQ#8}q91^H4e0Xlh%wpI!<Z{Vp{(2iPO6aZevP46eSyn#
z7lsR9bn>E)4EDPEIu=i|`9?6HvLXGq6(IM`aT(HIhvj*P^tTCLg-gGrnlIhqFx&l`
zrpRX@o<`gA8@Tw+z0l$9-{MbKLOJ^MQJ4D+K71B0sGQzkeuodA!^=12H}Z7X)d+<S
z#rz(B4LZI<2WVe3576FwAi7<YKGNOZv%I@2-U^wA<L%wuOS<FpmUnctcgwgr`~MJ1
zrFn#PPor{4o;9!14y$JDy$_s1#ER{ra~dTT>!ny2O6MWZX9hxsXA1dP<pRBRA{lAy
zZ0}sseOP>1OGkTr#p1=C@r7N$MtfH)@YS^<zM`c&-m@g$4R5Vzm(0y&=Byo6b)|b4
zUO5OSl~+w((z!I=wY)Rlv!J`ZrES^dMrZQkC9N%s<J~P?i|{Y?-}&83mM_&$mn?4g
z|IyXb*{;7nbV>K(w)nDzhto6vU!o-cvu-TU|ERNNSx<X+yrr$J8_Sh{E?v^ygMaoc
z>26)nz8KzF*b2g1*3;761In_ztNho_mX+!a;?Mu-@~-&8wz3kJFKdst%x~}Ni7$iK
zi3$2F-qqT=INq}Cu&!3Al?z(B+7>VDnqO9gP}+`#i`ysr)k-a~n`%{EReDuzvp94<
zeHF9`vVe3)&3`eve#*nxGvRl6lzeY|HE5<z$7ZRF{qPlPVb;Mtsx_ZO>L!<~Lm@ft
z!pumq5TleMzTQ?u=W%Btfr?Z0Cj6Cv%wZr(autIwg#r=Y*Fi{{gONx(ejp<`NFj0m
zuB|G0fVuPVr*J2*faCKWT=aY{dI8?a8ao*7WsNPu$8?B(8*zx9>(8AJ=0cpxmx#OY
zU6S1rydy>Uhl%(Q{3!}zcM+N+uf9n#%bCmyCbJFdo5wZIT{&hlpJg(unap-3nfRE=
z^f8$vlX;g(rhm+2Tqd)Y$-LVnGdyN8M>3fllX;IxrZ8qQM=_aECR1mUIcCgc?&$`t
zB85Jl$-LhrbK;oEoQx$9nNyg|f0$%8j+sn}$(+t)b}`AEF=jG4<lk9L=7T1gv&T&4
z%UA*l?R+M)t4Zd<F_XE3$$XW`>}Ha=Y|Lc7#$>KyG9NL?d}GXHbo{xinaqEhWUd`E
zneSi;R5#ycGXG_gxoON~ZecRFGMP`9WNsTXnLC)wolM3t$=p3=GWRi=`<cunlgtBS
zCi4iB`3aNxv`OZtV<z)+Ci4W7+0!KR<e16)n#nxPWa>>aza2A~=a|gznM{L8=8t10
z^E{LJ3zKOy$-FpbGOsY1SD8%AB=h>1$^4DUyv<~$n`GV@Gns8atBlQdct@konI@Tc
z;ZI{4b-stmyqC%BW0HB_n91zSWOiXP`<Y}uIA$`tF`3<&%q)}4N5@R&<4oogOs2^s
z<BXZir<u%Wn9N+0%;YhX*^9|EGMNKRGO;m}naO1KW-<qvWcD31nb}OHiOI|}$;=%y
znS+>23zKOx$+V7{%zP%ZfXQ^2WDXuPnJy-?gvl&0$s96fGRv9F3MO-~NoHj&WMmjY
z752{U!bx^FURL1caJ(FWmtMT|<Hf~G1}{0hti#LEcsUL)C*kE3yqt!Y&3O3|Ue3YG
z`FQyXUM|JUm3a9EUarQ=^?11vFE``mHoV+{mwWJXKVBZf%cFSt8D5^m%dheB3|@W@
z7hhn+BNxF_@0x*JqCSlUN_W&rbQuB-4KYNk#_y|8`pH~6-BIaFt1kV$X6a%n*EP~L
zVd<qkZCiHf?7%?E?>F5g)^QxMj^l_WXBqNbF6DKc1mdHCJKYf@tC110FjQ(V$x^<1
zUKjHQcibWtkV;((Qfb(<l)1YLaH`ya?Ur@mNtTBM^tiO0OyrWz`ot)v*OsYa6B(yv
zUaM2Y?ZvvG%wP(3V-uOaRIF}`U}CypqSe2cC>Mx$23q<QFAwH$WpGJ<KRV_@p%8j<
zC<EK65P6cA?AEo#+ZEk&fbkyq!M9yOQQ?=HPv6xF@2*f!IXo9FoXrqen=IpxP5$Wf
zL>;T5v`0lIgp(jRhQnSSsfBf=GEz~SfvOCjA$4Yd!NUG>o8oGC!GO!b@=KE{(nvyD
zbQu$$&{`LGqm)z*c9sA;t1?)sEs|}0MG`|Q%0SKwxrzt@hQgJVR}Or(06x1i@IKW(
zOP{F-932$tcpHS0kZNIdtBP1yIpij#+hYI9s~8WV7TxJ`$j1=HVA;slc&KiT>Yy{V
zq`87xv%e)(3N<I&lf*HDMK`S?w{)t2!RWK(WFii$0O~*i>Y!Rc&4~aCEtV%#f;?I-
zCAF#osCfcZYgM4kxV|AZ=84=p#-{8k$)jGS?dqpUQ-7;q<yKFt2-wo#h`^=oq34i#
zzhk1eFGhdLjSZx*m3kyI#_%Waz&#X<*}Sn>lyUg{o$b>ns?L_~NXbsa`jlXQ3%|=<
zm}l@+Y62Z3jm}xyhWC8#bNB@n8P_V}@7faZLwVQMYd$4o*{7cRySDkFZkG7ftUNZO
zziZnYJ60NHyrfW;tX}E9J+@qoy!X)gXe0qAZ~VS2veR0kz*V$OuON8>@~Q&jCbhU!
zaM3pG-20UTM3q4QNcGVzh{Q<(?f`xq<aBVgr^l2Ai4CLsHWMGz#JR4{MF$gCQCFBK
zK<ZlaDMj2F#zIVwi76=a4E}<SEnZ&=;x3E7>;Trt`^y~sW~BU^;0U1Z`_Y(@P<n+>
zt`R~x5=)3ElwFiTbU%Q1q`*1+-pCh3LMgxxJ_Y^&R#ee^O6s=Gd`bdYk3}ni_*MIQ
z!NVPbhokVth&+6PdH5vrFpA#?@o<cZi+DKJd`diQFrR)L`902jdM}>UKOUaCQ-Hp=
zLW$=S@Rwko?<Sr*_t;i2eImXYk!kK;-I>hvN%+0N^s0IA?DO!0ui_3zMov~weHGEU
z=e7b>OrYWjvlXq3Y+u~Bx^o%PDH@S#NEGvBqay8V+?_KdX|T5BQ?v;`w+Te{qC9?g
zTNU||_3)`-<h<j1?7FOj@_>#}PwD0>9PR?VQ+xezFMF9q_?Q}SwCR=G#U(DmJEN*e
z0*Am4zN)5)LXY{BI+;V&Q=bAXiadXEkG?~BB*;qOS&6$E$`*4ZtK*hY37%O?c^w4I
zHAEe_#{w3Aw?{luZ?lHMRNL+G;8e{@${_{JA}T`S`GlKfNLH_HLq7ibMme5<DMC4(
zhm6UBJD~8mgLo%YYy2kz?l9BIRG>31<9j#HbP5&dR5@f<L&~DN=?)Ok7w~dATq>mK
z7)7cyRIgV5g9qsDF<2lOiz)}`?gj<LJr3_YLv!~8F4P=OUvLMQsk67+23*eIU+B1c
z&&#1yYG-suV~F$7a!X3}@Vm5lQ}9(}qH~Tk^~1J-bt8U4WtMbj!h<!kd2^vRE4O*g
z=2Nl<2dJk$o7Xv4!fA`z=}CjPKl46I;VfCh&vioZmE7pEM1RU?+8%3zM~KQ#z=@G|
zC!oXxEBN1vV{JX$ouiPso3#XmJ6kjUMD;OW*9UdIEjU7m*!Al9MT;U;vM$_I6hdt?
zjewq5&-dCCEVmu+q+ZU4d#RTT@G&*6sPnx=T;yWB3+jry;D>UxJj8rTYT2Wn`qXk!
z<oTX^1GHD4bD*AU9gYC01%1&||BrzJ)pxmVSkLXMQTT>qi~Mt~jxddoJ<N9jg?|}H
zt6_p+^x~Z%F`8f`l@!!qk!D>3=2Mz=xmZZxtZNv5Q3GyO67e~LzZ?k{?;xFKU4`;*
zD$KgL$t?2Ex<;`CTZ56^uAX&mz^|xozNG7hts4ntp{g9k3e8v!HlI=>Tx33_8OUO+
zIVBLkg864%o%muU9x7I`o`0Rj+-#}D%~BH~adU|Il(^|OpVItmnfa9FUp?^D-2(Kz
zRZHA2$6tcEH|Ad}@Xe~(*E6vTnf-GD*r%D;VJ3K*iLFvkeU%e^CRVX(^qlMp#&({@
zrZkG?WN`(;X9UXTWNV@dNe`Ir0IT3WiI?Bw<#oLL11`Rylb*kQOF=DJw^u55BI7%g
zyG9{$uf;pre_Rjuvj4aNA5;GkaVB>omv|H28OD%g@qPHAy!*J-d`jKN?dqvdfvT9v
zz3Wb+I`mBLZUl%Xs$|_is#qO0&gtF_s4C6r?qyiD4>dgFEwQOrKAWmMyDJB$XLmnl
zaMrD$iZZKxp65YUHqU!VA#)$byRA0QdyMHkUV+Z|%=4aPI!{%gQ}aCUL!f=`hw-wz
z**bJ8#@{IVV`v=}W_-_J(PUJro$>unfpVY2JMWC|4_v4@<NFLSS+;?aaxnvJsbkt;
zb~rT+B1>gbxoLRzc-rF5j&ydt`cFRBCy&uhQ{D;F{OM?Nu@OkN@)fWn?!|by0A`k5
z?j>kfCJ8Tx&@YSbrTEn|@RfTR+;zE^<KI`n1vWhofdrFc3QBiR;}SmI$+J)HH(+0A
z%86UB$JwFFjp5<>W%x~4q4y#In%B}D75XzgUZV=Vro7M<wnEug!hHmbr$VpQg}UED
zhGx6phW~7mr>3sK7dIp0*V0pLv!&H+Qkxv<B!VU^)&_h25Im(NH&7ggbL)Ah+3Ct=
zQjLyG8Q<hU-~o6zd9XMz0Q))piM|w^X=}ib;Xe&)*9%LieK*}T$kW|G8qT*}2mjev
z>~(J}XT8_Py6#eb52Xm!`|u+n&nT=@hLJzw*J9oq0w-yM^DyqVQ1$+q;iPJS^L!*Y
z{Wdr+U@-*eMTRp_1Duy5!5Or`8N63ud4;hI)qv%-NLXA87FUNAdV}GttsagzkJ!aJ
z9MF8zHYoZ)$J-3$`s$l6v~W)Y%OF*wfdzmYv4PtL8l|YVZSjt5%XbWLrBYq%V5)Cz
z-Wlq5(1th|_0$2cF!2{yhhkr$m`gds7}6n?udC}$4CUcZ2tEN(F!IiNaO`phA!ZyE
zk;xY1o5@(6@h=vubKu{nK7}4=7`!5tbb3b}2wyQWn#}gW4_mB;&M&E4G)Wa~lG>3a
z^&XbgjfSLFKDymD+tjr}=!OEw$B~BB$@aRbzJk*aAq?PYa;OiZJtlIaz<qZL1efu=
zJ212^g@5B)LT>%p9LTka<Z=vMQ9_fj4kk4@J%=rAZ`4mYLkIk{iXY<#`cwQ1|4mw(
z8ja6@7gw|_Ufv#`5u2e45^u<NW_sS4S^iE;R+{=|gZ~vP;h7N#&x}=1*o$j5$znrY
zUDHx@Mo1MPl0#D+J#1(|S;wK08XBA?61-4C<fp|`T@IY{FC@SxF*D#422-T`&~G|0
zC7>#TN`)H5uNtUtq~ex9yius^v@UXKHc#o~5@`qeP0{7@ccL$?@a3Y^D2$_$zR!Ro
zp+lOTd@9{fCHVDEBY^ty{2PxZSNf!qp%7sB=-}9BP4YNfliZb?<cGOQe$Qx<rJ`ij
z5R(*|r{&33>TQ5qRb(GkA`2JRDba+r(I%`(EUZ0PST`HOS~&;pm-rVbFJ+i8wooch
zgGEnE5H<=MfWOTL+bVo-Qe}HpUJO3Le^Klfzh|*s_@38J;NSSquwnSf&l5hue+HO9
zTgLBcFoE_A--`&wd`;LlEC6gA{-?zm87&NEGomArx@}Y&+B2v_C)*G9Y$&e`+Sb&1
zf8TssG~F)QbbAWxb_477t%h!wmdef+ecIEIb_>~N_48RoY5?P57X93Aj>hd5ZQM?0
zZf7vJx0&2NhfD-@b~sn7(7*g!_Dnm6%~Q{bRwd`zs$>pV$y~0IADC6LN*GqN!^ZA9
zCa2oebg}u4e-|xv(R~)@u()I4o*#|*3vJBLW9D0#`8!PJ`|QlCSfyKE2*N?l{Yo?e
zTw)WzLKeWmEP%UA0h|^rfDRZ<rg9@Wm>dZ^oKnj!N*vDm!J)oE8hqxm={yZRDd^#Z
z$?4K)a=P3mr=={XLs(Atm~vVjEGKFuIQgWb9uvm;mC;zg%EtPk%=$`Z{a%yxl_9L-
zbZ_f(?LN=e;TzF7|CWvOBbf6zbAG?c`S<IVL1sYTtO`z=B^);oBN+5+!Tvd@H)Ed{
z>e65(LJwYgV2Xyh-tf#C4pQ+|XiF(rr|^tfVUC?nJH06f50;dzM~ufHp!8`5Nh_(1
zj-1h}WmRNy@)qyv&h^n`{~ep`U6%b?mi+^U>{spy%fujl$Re9^GI~TLmMc7xs#FCT
z5<S>2zvyAfrWei>dIwop6C$K<VBu!R*w7$eR)#QiPTS%@PINVbRVyyWgk2Pu)I9n|
zedAKsyJec7_(G$D8PVw$7S8vX>8stFqt))Mw%T3CwY#2c_aU=(Z>HMCxU@d42aM1v
za&k2Tm+~;&C$4P3iblRtRZ~dNJVf%Qs=$Mqz-cnopMXr7&C_Eu{Tcu(cq7>yE`O+=
z4ed9sn9vbZ)$kw_N)nrdLO_*b%sv|$L?BREz?e0GIrmNgaNGriFzo5L(UC&+l95Ip
z{xdEbH<v4=$)Oqng@dX94IG551L;(g)0@qv;bBj%2zxpqdU;ngy}ZY!m#45^p2~Xp
zh@qFI+w0muJ#<qHv<N{Ke<%a%CMhw6!D*b<xfai~j^0sf$I?dwgn5wyiRpS7r<&PG
zLWSKn!09sq_eT@;12$0)bI!*zSk#Z2qCUKCX)1?(4s@vb1RSXzbzqggZ><9j)3EPe
z8H|$@K8gZ|!;usw+oy~^i98ZbB9Gc6axN?1c`T8~Oo{YcB(gwkbauc-r|MFCT?L70
zTv~5L>HMr*Iw{)xT=^5U?+c+)haiCe(XiKoHQFZatX%e??H7D5);8%l9<6{^vh!&E
zXq@dev@L3}TK2BG;kAl&aY38cw4Ge)$l_2Aj61YCJRxf+n=rR>;!A{J_}1ufe=^Cl
z-8_khzAZRv!hVX^vdrapSz^p-6CD%4$~wNnfNep10@p%3Ei0<xY^<=EkG>gh4{C(k
z%MNf=c*53DZaM=Bpv_~tqn*EB;74+6Jz%^3m$ovskIQXXI+gh=Pnq8W<ggB*Jry>(
zrl$gJ*Z%;&_HEZENZU&%){Sqd3Qw(6)uuzM=1Jr@q-uU^6XV-GPykg8-H{mos%==k
z8?e*+J6oaJi)8!ds&1apg$iqOfN4z<&Vr7%6y|)*bz-_J?yzHZO=L$L?KUa4e#F+M
z*t(?O^mdh#FTz4bA}_viF!4$wP0<>mHLn}W!>5f-s?f(}7mArw9NIf<rw)uZIq)Zq
zQ27C5SQ<{G8=VZft6+jb2gZ?!WMTyN_Y}Y+<ieqZ*MJYA>*q&cWR9yTkOI3%8Qoxp
z3120lkIoJC79mw34~$Lq7@rZ_pcYvp@}ya8o-8;5?;1wkEPQd2AfUcf^W=e%BK(Tk
z0~39HsdNfB$mW_m66rk1GHj{yRa6kW$H410J^C1G4=Lw=u^v5X>d`%@M?EruGui$A
zB22U8usW>Uf$0JD1=2xDe!)&ky3+wGI<Zz9y9`taV``$(R43J>A<unP4|KWwvtquH
zng&Fi1_z2!C!L~LH>kEaCxOflFasM&^+9jW#ZVD>r(Wnr8iMc)rbeMvrJQ*w@W-0E
zq<`u%*v3gg@Et6tJ(Ki4GK!gk!smZcyP{CPvKg|H8lmMS;}X4yZ<22ovyMQ;bU(q1
zP22n+l5%c`uZ8*F?ZwL3$jbRkQ#p^rO-?l=2B~E++K}=$QF8`WZ?K>gyz)fB%f&PV
z3<nC{KO3RSPaRc&8P}0TtebxkeHF3JwUUBQj~r1EM~yk*k%GfTB?DYBfq3n(EP!nQ
z|4o8!IGt?-Z-jL6Q<J71;H;W4eMY0R|Gu+f;t99Y8=aZcXU)R<nR;syNKw)kQJkW5
z3>6NMaPp0u3Kg;{=GQh7eM+&6ziwdatNi2}#h;&=v>JdMGntE<3_V6Flh5YrSFLgo
z6$~x~LJ3h~#NTGFZiFq>DUFl;5WGde=lcJv8&^$loZdJC|F?Se6xejt|Hh?JO(iy(
zoaGtl6<0wGH9Cu-hr+wnt1DP~cMon*b<rBs`)v(s7B{Hb+@OAKHmK#^29?Co!WbJA
zwj?r%RDD|Q*r5KaH!5!YBn@2OLf<cLQ)TTbxMkq`LKLt;52i#nfUxG)81^GJk>KW|
z8Z#3?!vOUNtHPv;s<{#qJ}@CoAE^1;s%|CYw7jsJ02e&;arfi~v1_yjv74<y=);K}
z+#r5qHi&Dz4Pqb#4)ByU$n0Zi9BAubxIR#)R9}JvK=1=YR|b(7Ni3kk8y;%VCGl9B
zq(n1etUpOu*mB1?bP!w)DUhFo4){>tkgXKEEEt}mUr2s1TTFu=Ci*nWz5>f;-4xF}
z-Giq<F=z=bQW$CO9&<3Ng;nA>jC5%vh|8Ca4(9DfKROl%&;V9o*p!9I1ARg5YyQ&Z
z@OM62gdA*6q95h~UfSSFDwnGr8Zt9c98e?g5h=p!4j#lNv3tZ-vHU9i`mkoOsTgBT
zv+qWgkQEN<z{(+BqXJZW8<#$*RcH~;*0_pzMNdvalpxLLfyV@#XpM4`tx@)HFSVQ-
z<uhiZgtSw7*ym{`($QEA{Hkx1uxB7=I!RC;u7z=FqNMH<=1%&rsdl?7wF{g@NMDYI
zXsXm`Q>6r}QZK90?@U#iy1hr0$PTa#sL95I+i~DHkB!|@%6?a$p8^{=V)--J084eD
zjZWyhW#lz7YRVCL)iT`bg&)ZgIuB#XGSf<(fd4hQ<_l+nv<1b53GBU0D@Ef}UziFb
z@spkru=~3j=7fC)GiTbpPeH~{mJC=TJhY^Iv&9-TDU9kFX>$_{WnVT29m7Z#_K#q?
zCv2o1T5^HZcdaob7)z#98<#t2Yp`#G`S$-~?>pe^s;c!7L1a{_(p0!afe^w)5GkQZ
zNFjk_5|RO;A<j%@ZZdN+jkz;Ph9X6fc8Q3gh=>S?SioMN<%zw;`s~kw*s!7Ue72|m
zwZ6Ug+UM-E&)MgmnF)#i$&cUV%qjbPYwdd0Uh7+<BlKld*<dF~BZ;0AtuMUZjHc;S
zSAeda>VV!9ugfH!gsNuva!@fU6IT6IXXk7KlQx3u6%l?qq^$6_7#Y=QuGDLJHIdpu
za}ulwFH^ldqbE?iPik#eMH)0W*J`r4uA*8}cC+NX6TL_lgvS;eFpdii7=tok49S4;
zb9ca4i@ArkEZAzJ1Whg-vQ~lGw1OTSG>P$YKtG@sh-}xCkxAA%TWQy$iOGa>)skk6
zogu9nvy7P-1MF>DNfHLB|E`<V;3g!JELuZDePiuiM2SsVYiZmhOSG(PlxeFrfc&vB
z^RZ=um0O!y97U%}BR#!XBb^&+q?bt}y<8gUFP%o3*qFICwK0Pb()CY_=G0DcG*|Sp
zsF+?_Lwae5UTzS*JX%68#}wB~azN6nm$qW^oEwtoR*~m6k>@wf%ain5Ur`L)OGDsZ
zFW~MJaK9TbxbusFdszsabyK`Yz&+-GTM<pbCq#Ge@QfKU;bO?9P4X_<#R0oMyCLfR
zF)Up4m&LHz>;V5YZ>F^4LnX1avrB~7$1w`T?u+@d;epw-!i?wBk!DSd=3DWL^a#=>
zFAdh$+O1BC?1uHEn~ip;c$Din?og4l1?t}!fS$GcgjDom?4DZZh{9W;D1kmP<NJiD
zHCkY@ja2P@8nebJ)0Dwwnr#ac>M<ePlCd^C*WFcxv5TDuRZH8->kU~e6t^(uZs$MD
zJEA$;`B7+S>3fseF~t$K^Vji1h7`i%YB3Nhbgz?5p~GA}3LUhC7G_2m;iuWvA{yl*
zBc}(_PzcihT~7sIXLy4^x}XH4F@bbp0Mc*>(#2c=AYCGm>LnmuDv&M<KpF`_x<Vkm
zLLiNnfOM5WdSw98`VgcaoMyCijX>H^0@8H?>G}YqjUh-kZ~>^eO#*4G1f&}U(yIfI
zE(}5X-f0HXYXs6oB_Q1_kZuV;x;O;sgqIsgw+W<6N<g|@AiXXC>E$6vw_Rl*y<Q+)
zS_0A=1kzmrNSB2m-OUA{E!`uKE-wM;O#<n@0HiBIknR^qZxKkZC;^E%3f1WSwg99n
zLy&Iek1J^D0fBVYxFEeF0O^$>NblkT0O{QV>FRModT#*IH6cjv7f2rvNY|Et^dWv4
z)9!}@kgf|s`u%A}OCJ$P*O!3wF@f~)0Hj6;(kHk8RNO-X>4p-JJ|&Po9e}hc1nIK^
z>2m_<RV5&OK_GoG0O`gMq!XpbeOVyAx&$O<v{h~CD*;G1g&=_@7Zvw4f%KXZknFhh
zjR2(0AxQtl1pv~w1k%kVAbp3QMvwb$0MacXNTB}(r0)x)TT4K)TKew*q}xJ}e#8X;
z(vJnwYfC_Sn4gB0ej0#udkE4I(!75rkX}~;lGV~L0+8+qLAp~Q{YoIcz67LS^V87M
zZvv3+3_-d>ApK4ty`co8#{|+J0+8+sLHZLHfVT9wKzd^dNVeks9DsCp2ol(a0qHLS
z>7Ejh{>D$E;{G0o<ngD!Hw5YJyz~O3{}V`$&!cn2sR`Mno;0=I$8SwKX=}68jZY=B
z6h*Z34zL$aS6^qnURh}TbcSb3Y>sl{;jKwFNvr#K8xMZs)e74i)k(V@SzAy3Nqu%3
z7=HbT527tj$_3g*sXoc~2mTD?ZDU;L8Ha2xv72n)U{_~f`*7#LI@-@B>lDmC^T|XN
z+Zz$rdfI!D+gs?F?!o?cj!mRT5kG6(bWW)pP1!}YKT<o~N8339)pwtD?^jGj^RP7&
z^ZI9t>JV#=^<!eg?Nf+0SK_H?@#(ed+NyQMuksCtv(b$-(%DBeZ{1@MlB1glx<soh
z&n)J5YU%IYSnF=54X!T4Hz>`_qyW4@a_3iWfh(u2UUT$u(`HUvGerMer7U-Nj<&GP
zb8J+^1~@^7(-XTO@pBA^A8g_MDRwktB0e1o3UIRtt8O0b<D#s<#zsrZuEgT3*qe$Q
z1`$m_N8Q9mH%xS8WBjDu{;jsqhFxt0G-mBXv;)>hcZ}zqD1&R&mP__S_9a?f>*R*f
zIZTlwu{~dQ3R7+bF;h2GLidvo4({(BMWmQCpP7}@t@ncV|DrJUcD0F5Z&8lg_4tdZ
zF{<~pN<6#X^S8Id1r0oYyy41iO9$G)Yh$L`Uhf(u_k)2Ea^D{upda)mQoieO7ejbL
zwPl1tCx_y25aFu;scZa3vo)TO@963a)brdMWv^Vs;iuPOqt*NLkw<g90W}A@i#Geo
zv0ro6Q8*5rQ=Zx6CJ-+hejoe-jgR!ecTme=mDH&3(IJzDAyNTFkefkF;FZ+&JsYm^
zcX;a^&rp}5vC3@HCd%cF+G-1S3CIJ4V$~C{y+vZzL>dlz6!K$<Qgs8;mymcRj+6#K
zrH3{We#Q(L8_0H#k%x4VU%qm$G26?C!LmEcI`L`zg<d&gek)uFsW(>1OOUPftxMYG
z&8qNV-AgW$WWc9gC<-WDJ5Vv)%p^Ry-%ueKePZQy(@-$L^*o>&I=_dgp9guk&?5(*
z8XD~D?HY@3ZPCf$8gU1A>yrR7y|-Gc&Sa)hdT6ZDV`xFBmD+%WVQP{*IB@4td+d%E
zlQSszTs33uPS+%tM!WnRF#y+`zP?%?<{CXECH$n?2)(+Os*|Qqx|iIgX$0dAnRE%O
zwxDG4XKn9C8lU#j5sLJ=5Zfer)O7H3>Fmj>?ytn|_mz)XoxFr%Yzy=-)Ts-Y&gs4B
z+~GXH^DyZ{8QVDF!=f&uxtk|_yr40g3Z5|4`$NmD&#X4fEF58<<1PG9&gRM6Lxp}x
z3jH!4q0k3xp@lb3zQRvqIr&vSGIDx0PaX_G`no{+hCq5p2}u7XkiHdw^v)2Z?+B#t
z3Z!?Hfb@NV^n(DTcZVSTP$2zCAibvqq@M_+hXat_8-nzRK>C?LdS3}hzYs{j3_yB+
z2-2eh>DL1310^8+Rv`T@0O^AvNRJ7mKM14`m4Ngof%JF)(uYHk22Z0dix0Rz3#9)j
z0qHLS>8}Au9|=MFI~M>*{}4zYEdj||1pgU;^sx}69l8vpC-D)6%8!?T^kjY-L*-6<
zlrvQRX9&_$1kx@7=@TU&?JAI30+1dGL3+ACdWJyyWC=*tWU_kz(x*a@E;!dz+@1pI
z(<LD7%}=A^o*97jnGmFXxBxWoeFf5IOF*((+CKp4b0J927D&$#NS_}Uqyqzxz7T>`
z5l9CMq%V#O((?k4z7&FVh(J12Abq(6r0M)L+S1_xNLxaXju1#Q1=3eaK&lF)qXLk=
z8iI7RKsrVseXRr}J9-}*fb{hcq~iqA@dD`^B_N&1Pov^y1t5Jh1nFdfbc#UwuM&`6
zERaqMK>Ah)(&++ejzIc$2}tt<()<9V?}Q*N5J(FJ(sxThI!hoe3PAc^2vVy+S|X6X
zUj~x+ESw#H^n(zj<pQZqApLg<NVdnF8-Vn~5Tq3X=_LZ`M<pQHIc{YD(vL%sRtcn+
z38bHtfMm7Q5rFh?2-0eS)FqI9S^|>QQY`@Kkr1S{0;xwJ{j3C}h@Zw#xh??d=OIY_
z0%<@X{h|b<A%S#30Mai*km>?yL?HdD1SDH=>jRJ;4MEx{kj4bkuS-C(&HJJNq~C-f
zoyf`Ju+Y6kApN!kq)Yi}w57`ekbW0}^kRW@g+Tg!2}oB8q^kmu9t%M_MIc=*kp55t
z(zOEVx&Wj<h9EV#094!!0_jgBAlc^qssN<NLy%rAkZux4|1&N~o8yo?Ti<^UL9$!l
zw+N&^k9X_)FZ@>T*7x*X(_7yNrzpJZ-NN?tz_vHc_$o0JZ9U{~d#gQE+B(5OI}_Qp
zW7B}x#_iU(kY&)OChZc@#uo=3)3%1*0HW9p-qfItAZGEXfs?q_t<3g_7U^P^)}cY#
zQiHvIL#=mh52sKwY@gUbs8$xP(Xp<bdLzX!E9(-~E_1z$wspFCXrEqzMh>@G71^lI
zMvFTMC7v@I2)sdK=%bn32C&UL7!qfBqdkfBy|gn<ag;F$9hAHiN~qMnd;1`5I1ke^
zyhT?4n{gA+yY|E$$4Um#i75%lXzPl|ZVh0gD6sMv&;kLqQluJH^$p8@9;#L<QaehJ
z!hnFp#I{y<?;3bB=zMVeOu7}3K1{s%Fq&5}Z3F;GkpQH2f||mFv=i+Vb5#tE@vvYS
z=^-8DMpZ!`3bw&3@f9Ii_uz&B?M#8XiMb|Jy9O;+KSj%-(eNhVnoc6@7SRY9Z-hd^
zP8`<ASp!VW)Lp3m#Wpc{BdoK+&r>C>s}0ch-Ed_(_+ycEkZ@~F>5FxuyjeX=nfu@b
zLM>c+li{x{rWkPKt?BL3Jpn!zOvbGC=B8_hn@+r(-ZwzL4_viGIncRQAJHyj*I<7?
z(Yp6`b2L5#pts9TzwAG^);NzMfo9PMDoh($XJRMx$=>yxst^tEC>_I!!pPmCy}S4j
zp>8Go!t<*Ji_rlh4#f!7IZ9W1RV}j%4i5@w7kG&ovG^`ZcGydfMC2nu&(c10j64g4
z9!B{j)ksK4xdBi<zjof6?Wg+dZU=%U8v!{et)rBbLqocO4492qB0r*9Mm^L921nQO
z@X4>FA^C)6S{gv%p$kSkX&z8}zZlhJOP-46+o<gJV^>qX#$<@KSx3R-LBcniLA}3o
zV|!n1U@hh$dLNLt@etFwvA2J;pZF$mkw$Y?KJ>};O<9_|Fq_bRcQ?Ba8TT!I7N@9r
zsysCEoS}tMTl7)_LgstXJJ|^b^nI}c7wb64t{!qUP+3E{1yoyx*vEt@O9}z7-Boi_
z0fa8XBH(=J=A>VpbX8H+(yeL`ps}JpLY`ebFY{Hrs@g)X5H#krukP)vx34B*eR7M6
z*U*~Yji}AJbC*&puaBCfr$iCOj~vB(h86SNk@ODGp;zkE&^yVYf;2F*gLonOm~ng%
zZd}B6MKzQBHz8Z{Z{>xS*?;kD$^R|1{5wuOA8~{&p}+G(Ia~7o3>9h<_MXheqtJiZ
zLJM!npDNa+7xR&k)3YVN!zbu^9+~LrBIg`|^dx?$xQ8d$tmg@&`2k2fh9E5vNDBqh
zlS@Fd-zSR#kah||Y86OJ1k%nWAf3%m<6C!G0Mb)JklF;&IRa^y5|GXlNGk%6o*IJm
zQh~HmAnjTL(kg-UvH+x(5Tp)))G3gjRsvF&K<W-adU^=b8iBM{AU&f5q+Wp(1t9Gf
zg48FF`UTSNB_It7q@e($JwlL%1yWrg?O6hn9Rx-LkoF2eI$lgU8wAqcB_NIQ(-;IU
z3_yBj2-3w|fU>~}q-T|YWVLi@0Mb4oNS6zwD+JQMB_LhNPeV&r1t9Ggg7lu*MoU)<
zr2R`kx>g`v7l3p?2vUO!P&PP$^z0IlUL}xj3_yBL2+~ag=`{lBxg{W36Z6dhNC$=>
z-71i76G#V*3)1ZYNR<$zI|S0}1=7JKAiaT~M)STa0BKqX(%k~-9)a|{5|Hd5@TLHy
z=Z7Gz*m(!tyx%O44k-cYE&MdJ^wu~ek3IL$5F~5QeLx`nWxV#>>HJo&J@+1uJ$J6k
zSX9)aJJAMRZTi3`9C0Le4y+yRTx-9N^=0GZZ$6H*O{T`fpFfsO*~+>@-*l$5Chd&s
zlieu|uHVK8O5c>D<l|qjOzRq@Fb&v4hzO{qmolK>gh^2m|8gy2BvVgA5*0$0dZTus
zZ?P|Kj~fdMYK&9_jXAbLTc{`#g_9x3B=0l^L8Du3O)jirb-&36qwJ_Yg;%Dp@9k6|
zRg1MILD*3!p<+7{jam~*pp|Hu6=kVds96pjgGR68G!k4Y8~R{4P_~+Jyr{-)1*Jfx
z>Qw|qpQqRcYPn>GproK15yUZ2neG5#TNCiMBZ3GPdt-8f$(54!kUceKdWGpo8=`D_
zP?5Q|5lKj4yl2xa$0&5yR;yfW9Og#p#JbjT{E4>Xt4KN8Yap|~vL>1<(oReR(H1T;
zCPLqA*F!uPDU{`4v5gpr_JHB687X24od7vvrI`+2KoaU@6#d3F%$Tt&)W21L5gYvU
zvuqS3wXCN2%yl;WC7=)!8$;F2e7Ghy&e$6jf2_9Jq1mZ%B&)vgHd|+8<pdlZOvaS-
zCe+51AiZyJfTAr2`-euF&;bhJ04DV$S{{uEY?rrvgc=#WS7)$V-a~w)XDn@_*T!#*
zk%WeaJQJ&y{W>bWjryV5xug_NWI-S3>Ko<VvHBow9on$G;=7j^VAa@FSu;9Je=>uf
z7OR6d%{B~^X;uy29Wz@xVgpW{^T<-bCvoc!&{jl^h=n_7BZsu6_Y~SWoRYG_P%o|X
zMKY{|9d+&S;3!26(m1A0Qy5gc?j~&SR%+)5Y8%N0I-rKP*oBskbwo1PPPEmv)znVc
zaVZ?kOK<0@<{V+8FeZ0sm96QmQO1HbWY$LhS!yvfqM-9rH5#(K3J-#$ON0fht;$MI
zt#62||FmJUmagy+#B|eEy{g@sT^3kwD?K#q@!p1cg&y*F5T$)&+WMpbG3+e!+-!P^
zDu_-~x-uH;B}6v{m4xH%(|9%0ZUN^!gHWBTDH63M3C~t#XF?QGOY+}@n3Dg&23lic
zy_aZ64!?Z!C7Y-B?A#hv*&;9_wDP>0Ncq*b9*$u3{hPD$T=6h7G73{TRW*-rR56y!
zTAkytZfD<~@?X@cM*Bp)%u47@B+iw9e%7v%W~$6i9MX!hs6zi{OTr7eS<|9hu0vHj
zhM&=i>f=JR9O^!b1E)ELV(X^?<usORT77i2?oiVbbnEd8Bn1~n8M4>)4z=sm7~aa8
zpPtvAXG1snkqV_bQpxUk4Mn>}i3J^ndakyC(rp|)>&r%wPgze5EcS1_%?<3smx$g?
zCCx%oq*O+At3t|^7gv#*6C@e3+SKU=S0@cFb(%9w&t8dWW@=cg441ejA^!PnYO)V8
zVdAl$pr7OyPINzAa5}+pd|Tc=!|{9r3?4W_Cpdx6rS}=KVlA=Ha3Vj$UxcmFCp~aZ
z3c)#9;NS>wPALKB)HEE=ZqsQYI4@Xfw1XqSSu|g?Gd=Qb?aT=kda$sr9LL3@(7ApY
zvr^x&%3IY3_z7Hw83sLY=7r$Q5*betIP*)xIVBFqV-GkZ1n0#9=QM%y4`)<v^tbT9
zpaZqZy#o_1dl~f5pv#L4?c_kS?z-9-<@KZtm6CKPHZt&G(a}NGMLKD0nn-0K%jon?
zk1AS2s%RBeED=@w(^bW?mW4zOukuh(j#M67r`KzJYbb{T&HMb(=KPy@d6yQGcUef@
zb41>AMcy53Z^JfhV;&LF^btV}E%W=<tfbl{-Uj#I#EaNgOvH0TBAzcIt`ZS<bVWRy
zxpq2<^MM_grXODEH5IHrUR-G>p#`rfChki^;&zF+-6HNzuDEBI^%d9XDW_mvyfy0d
z{9;1BEF@$kLaq}bpW+I+Sn5>gMOTY3;=9-yqIMJ$b#+M83q;gm5%sCAsEPVKb?en<
zcQGN?goNBELXL@$Ev}GDOru>(A5l-hnA=lK*eE3IXO_`ih4tU1BJ9&$VOv`^R@>W2
zQu;RUZ*QMYE{G?ea>}WdjVs4iZH3XOub8X@Az81IihZTXx|=KOe3Rv!Iw<kabgk=C
zd8QI_PW%KjR7}R<kc<tH@dlA`kF<>XOOQw0Q@7fHJ5o%>^&uH;Hu=|xjC;8<4#)B+
zV69NCRF>C>9f)i!)7j?@tE9sZhna=^QrY~@?uKj007NWiTVqt-SWF)mhV=1Tsq@=K
zAJ264kr<UP-g=|*#l?htc}U2+M94Rako&koE{rXPnM+F5@z$uz%ZiD4MM%syi<tL|
znESb6ChGF!tyh;<78CN7AtC?977R=}4~mcnxI)e|Gc36cZ@qfFrkIG=g+zR>)Z_a^
z#OJso&Nm|RIDGO}8;2XkWZV>z@jpbykBE#1x-ur(>cLyDt=?En$eTh!J|seZQiQCy
zLSED|hjs>N{$Oi91>uYiQ>;T}pteEVXx+KziJWyd7gNkFA;o-N6!Qg9%rsXqi57nG
zR%_w66_fGykc?jy8NVhnK0htv0$Nkh*6`L{G2T&3#ydkYep_Vxj>vfEHX`F)#bmrY
zB;yZ7#vh4{)6+8QRWWZWZoL`j-eNM|7n1R3BID0R#u=`Roo2@Aq+yD;n8}C=&Q1E*
z;q6x#PuZTzc2n$ECBAzm!PInrG4bCT68|?M{%=M6nXdRU9APB7n5W&Xu{eBtF&Q5W
z$@nLc@o|x{>dN?%*czDjn;qlCOttU4RqVp=EGG85Lt_6;#QwX8{Q_6)vs#wZ<~r9X
zjKb_{v(=`X_ZAcM{UJf0biP>^@5o2k89K%lw5w$<8DrQ7WpEAIVh*nyd8+bAp_s9`
z<fEa(2Pw9ec6Y}{YV~9froSfMdhx+x3ixnH0Z$bL>?#U4)>VLf7|;gk)*GfjQcT2;
zg+$z4MBGC}JkAku#TxcHn7x_=?4+R>t^wo_!u$0mu$%^Ja@L?mF8+xK{acNDr*rpi
z;~V(Df%Oz$+)b_|!{Gns?e)&pUDesA&sj{v1GlA7^%z3_xk*)97E)L0L0)man&thr
znQ9A-9R*qkM`YWH!YwNtbWdTU>gIY&^0IP|Q@J6D*m><B`S|qoGd6Zwp@)~SBhZFk
z1gVd-7!lP@Uh1*wyvF7|K5{^zC}jAY!Xp!*$jQb&@h$y7i#3CXLd{?wX$JdBGdRI%
z28|Bt>Ro+QajKVtl$*Uc`bkBk$UdIC`^iy={E#Yhx5h;FsbbQ9CM5lHMfw9p`dO~@
z%USxtp)n;qd9wD9$b%aw^n3^@0p=#_dQt|q&lMB+3n77@F9IJT0-x*(yumEaxOHRl
z!Y2i-#CWG0|Ksi~Ycz_|r9E%GyRDsQwni)XQZc1$2`S|WQOZnF%BikWhNu;wnM*UM
zjU3t6yN>$*>WG54*=>>>)z~eov$DRIKe-1gb16<A4{e1e^3`J6_<Bei$A~swDB3v9
z)y4p6gIDIwDubWd3cC1aF<pEsq>B?p7qdhcr@OlNI_Y9)aA^9vl{33nVHvc3Ce1(?
z925Q@ajOh>ElNm3lsH3!186m7l}IxfU)0@<m=~z?5rqU7340D-G@feI9g$_@q{<+j
zR#O)xlA#%l8sS;Bv3bM8gXxXkiJy(WQ%v9A3+ekb(f4f8_gq)spQ7PO%{*}<*K|^t
z`KTagQ=(@f7o5(@F-MM!l3NN5M@#7(ogvy4rAS(e&=|I8;ru0)>6A%m<jCFu@}Z-9
zMvkNzoykp@wxeYkYqKuOjQjfV%7OCDjEtoQyNTEE4~l8{hanB0AsSvF8lLZJ_&;dR
zs$Lt$$G38gq#+G)HMF*mwY^vyO-%=+lV!HSPSh+i@rwO%F~vR{Qf#Xzc8Mr<fuq<J
zXY+b`Zgq5soH%RKSI(PRSv6zEe2f`-QFRV3(FYV3Q)~e_3Wkx@eYMIw3?TDqRlJ6d
zsqtw>Eu3bLv(Abcl*y&KXHc=r3{<)YYxP+bHHUMXrz9tHSFU51&Uy4dZ)wEAv93^g
z)a8;zuBWzBp`0f)9MOjYFQxH?Kxo$t^0JA#-Rdz)_(6;Jb^PRfLZ;EoXEg4=iLdcT
ziq-hfLp6Sm)cCnl<7YZG-iWBW(fG|&NuN&n>jibUtA<JVSpH>N4=T@E<AtT)eogcs
z?WtWGy@PkwHc@Mj-ig0I&wtP67aA8)kEi^jA&p-r8pjdr60MHLDVN*>^s+1<X)np&
zq}334!2K=+?kWL?BY;~vUT}|jz<K<c{}=+dg=1Q=(}^R1`x8HucCYsP`d$-RUtjB5
z?eS%PJOpL_6$T290A-cc)AY#WqWafRq1F}SZ+t5XU1kd{!kh!|`pdxv1z&hd&eu<w
z#x_(bDy3cpWs^ka&O~gT=eD^js3>-$P+JXqZeM7U`q|`#%EP`@dDrjw5DDXW@Opf!
zcjDXfyz7^pWOTkW-wO3Sg^$Kp{T?X0grIy`pgdKe>^dn>T0&5Enq#!`G=cK;NrCc=
z5R|Sf43ymj%I@Qb;^|6zgrLkj$3WRrpq$rCSGp!j53+l6QBFhlfO}R5+|gGUaQg_j
zmy8$Oe#O8Y5CV6SfP1!pTRC2E&-H-w^t*#X;I=F`nyU!7gZU_H2uMd5Ai!oCKa&Ws
z(Tj%5jl-gf)I=DX4&LLHWPOC}rJUz(L+2Rm#l}&gLIz+mz>?vFa-7ntv6pV_8n@q@
zy9Bu=!V=_vcy=&L5YOy&cu0*uy4<J{M;J-3vvus9y=I0AwX@gHxOf!WZVTOBXRl*f
zZl}|Gn(lES^}JA~jK@Sh$MZv3O*dWnzQv1Y{5c^6<xf)mj|-IUW@<_ej*X*aS<uX?
zAX>|>LfqZS9^HBbIyEGa9qs=v0`)iot$2iTG`B3R4RfpjeY5FA#H!Q6Sf%^9Nn*)6
zBOInk+rHJxc}!2?PbEsml$MToO41Jgs^#Eh+E7xFRRfg{d7)ZCadfjcdKF7I#D6ZB
zKNqP&hp}pt*wsag=d;~Dc~yUonOIT{8^!3wocp{!?L6qo*;+6b)swlE<!3t|hy>Ds
zs0M6)UetoG=fv0?U%iI>FJkG96K^%lCKdWOW5zHAK2hDaQ!>wj@6b<F=n1mE)dxo@
zcs?0mtKOg!vCfzwb&l?Vw=6X6&aRa`F}|^wUACz3hwLn6hpNv~YC}vbCnuExmm#t3
zd1SL#$ixjQV=arS4OlkDI`1*0h9&DF$D&J<Ax-rYO#stij}R@2+DcUP5Y%VLb_m-I
zS9Q`DtkNqhZVyHx1yT$%b&33n*P*>c55NVhqADA(i%MbERvH~KK2r)8lqMZAQo#1t
zs{_=5IrvMF0i@w7S<1sj?Gs*QnT<i@cw{D_C31nfRl;;d^h}ABDY|fwERxiMvGL-x
z;iPv-bC#s^9Kx&G75k7)0|SkG9^$WhwPy{5pmR?eZpWEwa+pzHj5M^#b9$tySM_*8
z`)32-iBQF)(i}Ts+p$h4SD;3AxG<^W`O`HP&K>`tmDX>&lL_*+NJ#rCLY^6!b)oLt
zX3SVe`HR=qb!hknGDWo(o1&J6rl>uxph*w5i#_=WtD|9Oiqck}EZYUX?QbYoZ+MEt
zr;$NL+OH&~tfG*5?h(}LbRU_8Eauha`s`AsyP)QoSxs798%13&EvByXL+UzE)OC=k
zYpjI2J~{#F8txpcayz$eyqMOQL4_%(`AdKrF!g?#jyQ9J<>H}1?%1$;FBkPj#njsu
zQm_32Jx0`fMG5turN(8FPEkl;TC#i#GePR02pDjdbg+v4al2bjeE&p*1tvC7Y$a?)
z!^H$02?=VyW=|GDUs*!XJ!!({yzwsF=vt|pjULFJ4}}Xukgb_(jzGSMAIkbr@PFef
zMgwc*^gy{b1jU-URtc2rCI!m%At=_&)ge$ClLF<25EN_X>J})QCI!mtLQt$dWK5vk
z(L9vsB7S<iwTH}LiF52Bck%T+%krW3(hZoLafHeFb~{RV?IHJs3bpo-nOr;yZP-G$
zS9{3YSnjYr<bjZSju)MuAnJLLAIci$(rej*_K<gkpjh+4Ndje)Q&Uss1O9@mRJ8eF
z%6u?oKB&>U*0DNFnGgOyFdxuo>L=g{!8?zz<G!odobaB|Kx}Q~vt=NDZ{D0xZ0YMU
zDL)i~Vr}Fv6(~14b3)>KJq-ol>m_aE^F_Cx@#xkg(Dy<DSp&o+BG64G1llGI5Q{}!
zj}%ka&qM0628ioKU7Jg&>)*lvajvNM@nY)zb4b0`0C9(?_m&dsoiqc)KZ*(Z&yb+j
z0P!Xf^tKX$W*Q(mq;7WnWTBP$PJCP5*N`<pyj>vg%nxO)%+o`kZ)NT=K<piYVhs?V
z7AVi06e!OML9qsi&k2-$CI!mAAt=@W@kN2M-=sh}Bm~78Abu=R4s9Mv^e{iY-5Ma?
z$fJNZKqyC&#L^)B<@!EaKdssv9YGF_^uO29c8nP6YRNyNtLfGx4=wXDe=fS7&i8z>
zvA5n1i8HcqR<?{=K0~N(;FTJVHt_=|noj(Caz|*)IjV<Z{*I!Ij{~*gqnPdcDD{13
z{#PIFI!e9ZsQ3$xYA3@s@rSophsHKXck(OVz&}cmyXdfSQ0TGI+%X7<VUfWRj3+qC
z8iN8lo*q?6*FHV|m6T6<e7;AIsqHYW*t<lrI7%pXx~iQ`s?=xBiE+=jQJ4i$iq>=~
z_n1-|Cl+~k)n61l1W^<6VfYzcBFT26xX|Ht_IjY`_1ds{d*gY&;;Qbg+<F=dO2bQ)
zr(xRZp|R_H3+zF9+qKbYR8iNADm7{m_NFwmw2MdiR#S3J_sZeDAHpmyFO0X)k`nD@
z2<=k>9p)9tcRS=+R9+rmnK5x1JF)M<W{k>>y@gF%>Wf-kW`#X#a5XhL%1BI6l_?2l
zcqPnu0zxh+FJwls6C~#z^)X5@+n*Bg5U+^on<3qq<)urPI6;CvoyLo<TDvNL^UZU2
zpz)0UVPa_<?3FI@1m>N_{PL0|ON&VrNd~XRPE3)X(+YPVq3iWj<viaDKXotaW?^{=
zQU%6@7*F5YwO->i6qF}P59J4X<%!(|N#>N7Bvw?G49}MF!OU`V$T5h<58Cj}t05I(
z+RmYi?erQgk-vsg_l~!QQU%7OYo?hYK9<uCjdTv%8P+dE>fZ4Rkt#4IM539Mts?t^
z!672%%B`Z*y%Q&$e)$AT_Y4}nu_$Y2f_E+3G;fD;?u2A#l<!ndaanRbLpl~a?D)FU
z7A*53N&VOG!*6dN8l+C9*6o|scfpf5wF{dm?Fy1DyDDyn&V33CGQHXNU}|-y-rU%m
zHUh|Hti3C?*43Kyug<*pvp%CITR_vJZ_=6Rt<gOv(}kVsLhH!p%Qu^|PvJXvanC-L
z&+bY;y5+~8XyI#5^IeO6$nX9U{j}m8H7uuBe3a(A#U_6c`ScT^tzvN@|3k_uI84jY
z8U=!qmn)s1AW)9iCTA&RVPWq=gkhTuM05{3_wjjNWXoy8zfvc+J90SI`HEt9t0~xA
zsJ8IVTSaGFnpJ7(=;(;*g99xW!T)huybjuXyQj^n9HTrVr>S!21Rs$f+~CpE^c750
zbYbN*K59RjTmb7`(^l!LCW8{+AU>ArZE8obQkgb9*jM9{Su$O!Tq8syGz;+Mp5AFv
z!X?&WGovtBZs(LpT!DRq2ggc9`KwwkVZF3y@3g0L#q&Nz;TFA1yK17k6+dTAtYw?|
ztt$7R<vxR7qPkMD<?c=w(k*un_bgiOo^&>=<?iLX=5M)XgJy4fq&dO&&_YS}Gpzyt
zdNzGCRPjuHt)q&4_}sqqlTZb^_I}B;evwwRP-mQnafXs8sv(=ajr9_PiCoy(oT!-w
zD&$ZLq6D3(tKM>?{A-a%wG*VNOuJY?12W$@)Vb5|s<voT2=(Gwm8HYAHcnz#>u%$>
z6Oq_3ijco}d{DiT0|Mw44_NKnZIHHzuN;a}V)H;F;^~&XKWnwAmc17y0rGU_G;ie0
zMkAK4V*hui?{-x30sLyGk`LsXOjU4vOw}#3$qD9J&O$A*`t^TJW;_At&&hxu9mM$K
zQwG{+#hg{sg6Uibb16;fes4<h*KSux@wK9Vp7h=i5vFB{@O%T-%##q|5OX#u!lC08
zA-*6iB7(jjA{>?_!r=yNLWCLSY*K_HnigSq8nx=gNl&vQ5{2pJ?_aH_?gjLj7P`TR
zBzZ>}jGb6OG};z-60;J&OvNxL$^;4yve64PZgFnadbM&+y|!kw50nx;v?ySw?0Ani
z^QoECUQk5vTJ(k`Y*M?0Ccnn<daXJ-(%V-jE-;%EQ%3s?8)Mue_qi9tsJQQ(C9UVq
zTef_5+rlNS?aP)dn%CYsd-1&GN}JI+pQX;hx@w$UY=xXvT5MTt99khv)fEQT!Fn|l
zR$PWnFwRED37r6r0QW_O;A-zD36S56xJ<o>j!xuqm7Q8!gOwRGntvu;fKmHoz6CEl
zz0?cgk4|wIa7PyFFPl}PQ@Lo~Nr_&}N9k3&26H-J1(-P#1x9^B#i}ftCt&91f>|J7
z77CbR-&Sr_R8uLklj;%cQM4!*QL7+YB8Z9^>Qab=RI=M<(oTXpn_q!CTb2u|O+cL^
zpo&?jQlPq3WFvpHt?6B>zkcIYcb;Hck&Ef2f@!5-I;~uzM)F=n1tit8N}#+f7fOdf
z=@cmQ%0VIinqE>7%-0m9ki^p^c)D}(tPwnG1y9LVol4Q2MAR#YqFh9Mf~a2*6*JPN
z8jzc8I|*n|01f2=8Wuox0rcW>RYvv^mG?S{#CG`6TqMWygo^QSgCHr{j%~v2B$P2O
z7~TBBTqqawRY18!pp<OKO#d3o0-eNTRdZ=Bp34Q#6@usNa+MV8S8fXGB(^L0P0-a<
zx!B$_+vw_Q!B%X$DAl4f^Is>CT`S10%SG1UtEj~r1X;1&x)icl((QH|L6(H|DuH!l
zF07jb)@uY-**PSg6FiCOVp@HwO1wE2)2)K(Ho;V~!8_T-v+&%Wi{}o(^LoKka#SM9
zaN4v^O(AdKa?!l+%0+UwAh}17l+=uxU1>j7#W^Q6V~2t_<>Fbfv)%-Z-Yj^|DmN5(
zzd(|x-oo#Imfq?|wT0HXJ)g9;@8|)8Dmps$k*3!*#cbR0kEZBdTweU!L4E&3ARt5M
z#u1kCM-<XS-{4yM;7iEa8oiF>IBT!y4)s5lxK(>V=<hAIIklTd@!sp@6H9yORAC?L
zy-?*fO+|9Y3y7`r^t<;~@7^2QHt+tJ((du6kq+=q9I>p?3$1p=-1#CtcO3mBY$%vl
zPjJtk$oI~2&z{6*Pj=6q!e>u)&%T(?o<={7T@D-WUE4Eq*k(4}h<N}i-A9SFTsy4u
zBOa5DR<#!Qe4)U@5yU;6pLDEU4^mf1TD#z_csFfVBoqJ(%)Rc}Ra9c~={5A;<eA>o
znF`TbCGQ!Kdz9qS1$08Sr*oC?uU5jp>Et59kMh%~YaBuNxy6OQpI(v{{w?m=wD1^L
z-fHgCqFl~kP>-9HDBo40yp8Vj)?+P#TiaP>-P%%}Ceyv`FxA(SUU9z1^WEIwKL&N<
z2ySjT(&mH6!|rr9`;Fbro$z6@V?_7xU=iKRf4_--Xf%Xl84AS_3cZ&fGKG3Pao-#&
z^nShpg}#N4Q0QB2p+%w@t=CudHh$7XY<l3lov-J~_yFGkI1lm>;Jky6iuwm$8>51Z
z-pNnqz<C#6&x7-Bz5#IF!$*MgUOs9LPV_#01{2=<`6vg{2l#p(q!029fb=0g0;CV~
zQInAVgP#GUkML0rq>u9TJV+no8vyC!d<01U$wy5>`UF1%NDuK*4x~@=^*l(Q;u`?z
z(|iO-pW&k>A=&lQXZh(INT1{Dd5}KOHvrNX_y~}`$VW{=`Vv0_Eq$4fav*Kt>v@pA
z!Z!fYSNRB#zQ#vQLi##C14!TCqZ~-z<m-8m{)=w_q;K&NAbp#UnuKI4?mPT+4y5n$
z^*l)5;~N0!`+NjQKj5P#A^kT$11<fKk8&XWh_B~C`Z3=CNI&5tKzf*unuPRIeg=>p
z;iDW#KjZ6pkbcfL0Mak`2#|isM@>Td6+Z(=kMfa$<Oz-YHDAwzB(5{jZ}?U~f@@51
zNW3;oY;H!s<D#_dOb)omLg4<uH$Y#16mW2piGf@37_9_arhHYlV_y}QtbBYtU)SPe
zrE!b?B&-e3s?g{EK%Km5`+2v~<Fzo;xEK+c5Bv}25^X=TL=jmJd7d$64}6_tOk8}P
z7>w9WoLJh}*EV+1vH-~~4#}4I<wQ^7Z`p8Pc{1OY*O7PP8_-vF<|A}uIOfFq3iZ+l
zXr1gQ+0uHu@d?yhLaC0%#GxnJyO_qF71G!~qOpBNWANXJX>0{q@pfow#`(s}Y-wC+
zz=Md^k>P#ATipQ<9Qeh|m?71KGChqN_i=e`#yN-DY0w?V&$h_iRZ^bECnwf}C#3VP
zFbd1`RpL&keU6IsZ>KSd^H5YN)|jS+8q@QnF+E=z6a0i?jj2fY+ph*kuWsd^6q7PR
z5jLY3>X{+bM+)kypoWW3lc+lrsApA{*De^PQ?-?CoV!Pnpm7`0q-cD|qiPdXsJ*_#
z^C&vDm<o;yso;1~!3m-QxEO7#DiF`5=+t5=I4z`t*`k8eMFntC+E!H{o=?$&Vk$T@
zq=K_V1&c%laCM5Qpz-pS33O_T4VgHcka6<>N=;+<2&t3ByPNJn;yV?comO2Uem5%t
z3uxHi#OrahwD;sQX$vVFUQ$@D;0Vi=bNQi|!u5EW1;pCaB!JjwGh7jZ;`l-V%1bPi
z>5<1scYdf)$KMHsuCj#|wHdlDO-AJ&NG}UPa=e}Zsl5aw$FV5~Qb!1q<NE|ioh2Yy
zO9nifav-e^L2{g*0I914B*(8Q2U2$klH>XWNVO7>9Ji(%NNYln9M2~}T3Z5=<KC15
zsV4-<@qhxP-V%@;52qYRQ3#Ub1O-UzN<ea4oN^%bg&;XzP=M550+Qq8lmlrX1j+G(
z0;ItbknA7;Z>JncLm^0xLlhugPy&+U^^^l?I0VUYi~^*32}o8;aD2*vG!la3_(lQJ
zXbDJ;zf%sR^&v=(OB5h&C;`cFbjpFWF$Br+g#x6p5|A7xryNKZh9Eh9P=It%2}q8Q
zQx2qyLy#OdC_uVoT#(@AlmqGIAxMrV6d+w%0+QqJlmqFq5G2PX3Xm=@0m%*maCyps
zbVUe~;}r!+uP6b@@pm$iJbM~fh9KEJ4Y))B5?rFv-xhhM(Bz)REBUQSJFWU$T5->=
zoD;UTh63Rzx<w_=a@`nXsdl=_`&`|`c_Kh&P;F`B1R}%)P-g?3N;Fchk^gFEUuAHJ
zU9stanGYx}EZMLrq^6Tnl5`Gm^eqLRA`n-3vva^rr^(%eh#CsG=^UV#-L<uf%4^Zm
z!VpXZ1;z1^*%ih;&ha(Ou{Xmt@&~O`<f;kV9qJtE;Vd2;!mA%<V<Nqp!VOJn90F0O
zG~DPft-^{h35t<Ol@kh0vtb0&=!K~;2YM^hVQ3-vutvJ+rL^2#<Pk;!ZRnwNx)igh
zFjrJ@kfiE=qsf<JUu@$S3dsnK(d(eG-eJlwKR7(bIH-PXp`y$#$~i;HSvGL0C92ap
z-CgUB={6RsQ0VohF~A|NlCAK{^!2@+@oKGC`g&IncMgwn>J=N5Q0S#`nN^A$6u=e0
zZG?itDF^a&dR=WJ<wU1H_#H8?Xk0Jc9JO-Qi)!3fl(6<ybca5#g3yS;fvk}mdq>79
zg99lbsW1bQnQSPkQE_WRv|7AXaoy5Wz5S#86-s!qni?3lY7U8InRTE#^;Yy0bjI%5
z>e01p`6{JPzF@SsYaO>r{e<GrrQFceH_13krJF0#PD}$CA>J}-+p6ayKeGZr3Uz}M
z%1IGZ=mf|SE6ofa1tg)-NJTYJO%(3mDhCgbWcpbi^hhl>FrkP-WYAU3tFG<YmA=}*
zS{e#ft}Si8R!*lFN3!Y*jmXv+5hegmU5%-nGL3?#(LhWI()+||xd|Ph5N_i-MHOv*
z>W!be?IYC4=)I#|Bu%}tyr*-xcE-{+dMyndH1Vh#B|5T5uQYo$kRx7-X}uhq-bSrj
zO>~`o^+D=QU45g(xlpMS8DE8hetVmp>@dKp`K+>LbeR6+U0W?yhZ6(RF|(y3HsH8n
z7?i-OBQp|$FLtNB{&Bk2KiICj39*c5FLS<9+ei#615kSWLQ99^t55eb8wS0dbBYn0
z&hTj02+if3E00ySrnlDDP4W$Nv3__?G}As+V{*cz5L+sV3R>SwS>bwWeM7Rljhf-~
zW_B&1=08KUQW+T=qGxEVqgS+;SLlh92T@vfblQg*hv_M*AX=YvVgROuQlm(m7K3Uq
z#j2UfIEIOwkA>=7O&N7slJIO*b|yq2wIu%)HJ@i0lVR0<DX;O&%F|=jz9zKtbX?=H
z^1PNGD!TIYXX*Apxh@0+KJqZI;|Ne1<A>r&tbRiXisM8NC~%@Lt`+t%YP=i+Zab~&
zD3KStXi!e6zQL|_Hm4Lha?ohs*UodG8me{fsiDs9cJ*{`w>`x<=jh??8iqHTa@DNL
zk&kkgJUEPxz2iwA-NAw<KM8z^=kE;Z$ML#{e&BT<(@&$fWdZf;S|2SZX!+DWIM6qy
zQV-H!$6!UJXI5TsU}qe?D%nfuuc95r|7E1JpFU#N1Wn{o*QnSZM<vXnSt{U_6S`wQ
zQK!*1a@SzrX#W7>c!`6kuX8oE0CkT3G_R!qWOdu%Fs1mT;WkD{fqUp)8t~d}8_+vu
z5<C#Svsit<J5=9JpaJR|fd=vVewoy_cPce<O=2bgY8$TPxK3h=6di@sJk0}=I!J6q
z^s#hJC-z3OI>-CSyNaj~j-bdX`d@<EI7GrbGb4fuy?wa0mOjYqJTPH^(^L}|e~tzq
z#Hrv&hc>BJlpy1E<0AN(#_VR^m_d&bq~ML5;0e5W>A2q<X4h4577xPREgr=7Gde?o
zB2EwmV4X95SVG(tXWmSNoWX4uK^K|WoQMiwYa2f{vwF;c<_2J7!gHc40Pno<!wa+3
zMov6NCPF9J0tlClAE9yr%~GEenUM+33A+He*71X*b+jd|HJsRBMX|<S#AjpzbAm4b
zEGyLAJ1Tj{2nx+~h-f!)4?GbuNRbn$k%`TT!T`3c*xM$s`Pp3~Csrd9l@o|DQ5tgy
zyb|%UmYX#~&_*UUCn95_u%$S>;<f2SZ)9R~LNg{FTbeL6a>6(=u{kjsz?Ma-&_qi!
z#t{$Oj1$d~Mz%#wS$10kp&E=Vns4J%O`k6o>8&{Cb;(Yg1xj9EOO_~Sg!48_M;iTw
ze8o6C+3<x48EN$Q#1%jLVYCCU#jrjTr}?5}meEf9taXZqU}t6t)+}3P^fW3xB^Yy6
zM!O}h_yyD3OOZIX7naI#T!&PPJW`F9crrRjR4)$Gg+V$F(E!=%0Xa#;iIL;ljQ20i
zc>g4`A4ZOkF|aJl!ZNuuhLPj!3oLC}SSFW`Fgi=RC%kTf<y?!!3>ZFw!Dy+pO>t?>
zYnusXi{<=m{G&iKTR2XsjHfZTxusc?WPZ_kE|g8#1bwG81!b->icbVlUl#eBfBth{
zT-AoD2`22DAW#!a+}GSnA*HF*&UV(?ZxM5|q*n_1<JR$EMLTZp`zM9W?pfQ7@FY0L
z;WZPS<IM_i&3$I1z?lX{R-uU}^vg;CRIu@5FV4e)<aoYjs>^X}g}Sm5`4of{+W||n
zGdb?BnYbL^R^S?E>rd^#CmWdK1)GVikNXA|eU|92xex4A19SJ(lfWEb*i2xKqbpRF
z#fez3DrK8HiO7EPH&6UG&h~D1WRj2^SJ+G?IexAa(cYa+;3O`qs^*CRUox%<xf<(-
z?oM<PqvIf(sXNE}6}oHAGM8%kncLw>gpR*#Cc=jFO-l^eoM|(KaIy&k$2&KrT?Ue>
zbDU^1Rp)rfPDICWoDi~bHBT5g&QWa=hy%wtWvW?Rd~uSRvO`Go1cJ-QH-vci!IJ14
z$J%uJ+Txm*ERo=f*vHr=zD=b+59csQd~P*EB>2Ln^kG;5H8vFy7$Ovm;1FS5BhZr!
zLwtPioO59M=V7`_7zxww&LI?R?2h@7{SG(*MXg)!<Q@EZTC*^a=%rTWVs`_(;Q4eR
z!7s;5a?#7&vj|;jch7=zu){qI&cROiEI0>OyJx{U*hOaxa1Jg}_PTDm+sjp_jE^}K
zi#2}I;T(h#*1E7DZjXBw;`X{{A#UWJg}Ce7vk<qB&gO}`Oi}Z_N{QP~cL&97>@N)f
zO>1NjO{|YneCnn!Uz)-I7wFT%kc$>txWGLNEeyM7p@q787Frl_&q52MbT&^5Hz_TA
zN@-y|-5u0IwCh^B3tbRLX!#rXNk`laqRFS9>huLzrk~-S<sOoxj6}4;ZsxuuVdiHc
z$KpMnqy*WW?g|Uy_;NsyvEqX4WuO>a7X;bcJ)0rOGtGUPH1c_@RUZBp@BJht%nB{c
zvkD30I3huqi;4@gpMjGQW`Fl=T9^r<|2#p&U#Z2vMT>txA@NU^Mu{Vce@SuipX&;W
zx;)T5n-;$Sv+0>4+H19FAJd{8lqH%b1YJH3LQwzYmq`q&$#_%5pqa(w6NJL21!Dw`
z;3kD5?cm}I_J8^3s0ZZHB!Lzbh2jW>Udj)dLOqf7SA+_6d|XiID{P@fgZ)*Ommopt
zl_5BeZwugDRT@r+AoP_XIF36D;9T829M?a^`008o?3xfH$Atxut}Ow{aY)I5bX^FN
z<Ff)t*O!3gxS`}gYJ?y;o+^NJLkURs`zS;Zx+w(7@lgS!SCxR|_?_fv>BbNw$2A3z
zUR?r`<8+b(>821Q$1eqtUQ+^+tvL9a<UraSg5<cP0MgARAUQ53IgoA%L2|rM0O{5e
zkQ~pF97wl?AUVz_fb`lDkQ|p11IZIidwU3y<8=Z^@H&Y(E0bB^0a|tXy<!!wV8U_`
z-NA2lJm)>&?hJu*99IAijw>;6Qv{)iKASd05K4^k??f|2G|Bct5c<|)edX<;zT)^o
zp|8LfD%MvDbmYdiPZ0XSVjBB!NMnwl6*LAvtC+^72tq+;N)dzqrUao66>ChN3N<Fj
z9}bPFb%A_%7WqcRuvLs8^b5sMe<_68@uCB2c+thWK#@V?2_OjldNCDzGo%8?c@HXp
z^WL_qf<qTjt9pp{624bV1wRO>!0{P`3g9!it*XEhgg#tM1&@SO;J7M61#nf2so>v?
zAoQ1M)g^-9azm*v<A=?6hbMmd(U8I&UrZ?c*ZfdS;d<oB0#cA5^fw_WK7!ESS}0zE
z(8ofB`Upb*U<)lWA0WEN$$4y4?y-vfF$Bp+5c;PQkctz8J|2SPBMAMU5|FGV1D=yP
zTKaPclH**7hWWoGAUXb$IgtJmg5<bV0@7bgKq^iU`nM1yA3^BfOF(kmBy+U%j}Ros
zlM-6`zY>rf56K)z{|rHLoG1b5UnL+pu8}#AcHrIRybTh^dlHbI#1Ey%-7I?##wEuo
zG6&L*AxMtjBp^Mx1SC5Mga|@+3PHk#JGPNFutysp?OXzq;}w~srKf}-IgXQnv`Yy{
zR!bp*(5HqV`3ORHEdj~#f6UQRO9+zVA_*-$tpp^;;V}o&(?gIPpGZJ@MhQs82|{-Z
zLApuYG?6G7kajNt$?<K>(b66vNRIm>AniFWNFjpIy+V+D1fhGEfK;3y^qC<@K7!C^
zm4IXife=CHJ|Rdxg3x_SKyv&TO<VG88|)W?WcM^ILFkGFP0ev^i%ITj9Kdf)?rBUB
zgiaBJQiKN6J^CqCrwBp~Stf^YO%a3+6P4c-K`6+Hg$vixW}>$cpP*2kO7zndK`60C
zP7#E9*|4VwLZ=8qo8UdnGA4r{^mlBVF)L4xRr@)im8Xv&^tsI}h6@vf9vFh+M8V*L
z>Y(vM@eqVoLQoL>0&5D600nU{MQu)+mhvV<?<5F9<J^!cndB2f8Y-bM(J`!y<Y$79
zH1ybz&U~b*2%BkAXWIs;s>IGjbBgI_UPwO(qd}X(5!%#1Gv9c+O>G<cS<lF^r+ZE@
z{hSxlj}w6d{U8FT$@cUA9ZRUh1w~!Os-hOE3MYmLRe>0ySXDHp_+WLqUoYveHx71I
zMF*3X<b#!<spx`YHC+$Yv=d;3nnr+Cyr!p^o`-8~Y!|3?x0xKz-b$Znre}!~i>@kG
z-&cp~+leSceIueQUf&;Wp^V>^;eJYnu!hpZP}*db44`{(;INU(K#h`obdHP+P^yvo
zI#7N?Zc3)mx}<F$lbH{Vj<ol7*NGsXlFJO&s1W2F;SV1j&P+wv+jQm~mDys(3}O|h
z6yij^PsL0QTlhz}VIt|mG~g0v7u}q090|V||F-BgQ4TqD&Ebl^n!L9TzkD;tN$=bm
zRcZ29A=B5>`JPWUo*p09bl7mu7ifimkVPEnsC3=Vet!~$@Sfa6Bp^o2quMEFMej&^
zdo?rUZ$E|PsPz5$v3<aZ)ftY^F>n;?7`c9hMxE26fUc&+byfAw6Iz28hz4<#(4c11
zRy=Wy-R3P`+V+z6Hu~S9*=_Um^CRZ@#?j$-R^1{JGy+T#-5T9Ozm4Mxl(m|&dAS9S
zsurUWSJdlTm$WZmI&aRx*^AoSmYg-Ob-63UMv-Aof%k?B&63KH4CT$tF{yHJYdJS5
z+B_$`6a4{ff8~hE=NC}IoLIh`wpu^sdl{)UMNh6J(A;fK@Fq*pFigjJ+Va9^nvOh)
z#fjVm7DKz&G?p^-<4HJ9)F$8<UYDlfl%yL^B65N@fv8B+_EuK&^x{b{PPit(81~Ah
z)zciEcoK*cqzNE~5vXY(P1A=bu{hD0z+w^~G>xSRy6_|lCm?ft8flsyJc-4L#2g=%
zrs=?wSey{diNaEl{yT}qiM<4td1c4{Ch5L4k}WjYQ*H|Do?wZ-_ghTMJ4KFCkmx&E
zrV}a2(nSZ@GKFero2K|qisS^eL8RwoiL_0Udne^_!p$JhfgXAGVO!w7^pjwNLdETu
zJnJVOFHGy5gy#e>0ba!e??A?of~L+0;ZA_0KKW)R2nlbVmht9^r1DOpaN>Y~;`v!9
zCXmECiNXoR0g6MjP)s0wcM^pY4FeR@EegZ`!ym5EQKn6HV&E!F+&voSkZYV=U?j!|
zgKufIS)*4U35{p_?jajY8)Yid-*Uf{@1Ro)yg0W!_2tocrthZBN+<NCtG;{nXQrM9
zWXx_H<$G8ou9C%wlHO5zGn=<_v$6!tuvL;A;71lYqsLRYrbgYfPvx`g>1<=C*1>_=
z=2WtS4ZdssQD?<R=?iT!C9$GJ)Y?>4?-?AWtd_KEPwA;C8Qch)y2#0cPEdwa=EtL)
zY#Vwh-zuB3*3h3LCY2@K!(LT<o|3i=%V|!1$tUZSTAL1a%zNx^wa7LXsx3?>YBKfC
zs<d=;bVT*RftHI@%zRv@WMZ2(t8&atbw!m!C#KO6XR$nbn!bVsD_vMQjgQ)oZm0C7
zUDH<StJ`rDzRriYk?P#Xi>BzvljDR;(T(TjwcvKugmtT(nbm49a@!|b?Zxg{wAxF2
zEweRh;THXJ-!*@$)h?dVrSwQqp=BO#%q$?ZWUtfGS3S~W#=pXq6EprRl4t!=tZ1RW
zHxI)EWsg)Q3d)+;KSZ5+fU*En^fqNGXIp!P?8uZ`b*ZG9Z#h!_waBAtBYmZ@cFZv)
zsSt1)mnfqva#N?@Rc$dQ>0xr(TAH@>qkD<npN9k02}cw_57=cj%9J}$H%MEO0qII?
z9#=#>-H@*=Qm5J~BMe`S8F_WeV^KUm7*C9ivL*^MzPp8{Y-){Njmn2NqZeKieSNtx
z8LK(ZCIHNsh%jM{bC!*gMmF%UzFzSV-ufgn!$r^eqJidGyz-|xuPqA-HmMDo{hl`p
z_Lf{!%Ta~o(W9OgNX%vpZ$>opTAx7-@)*YHvhw%}j)|iu^N5YijlE6k=atfiD2>X4
zbV)TLk9cB+YaE@|7cvZNkkC+mmz6N1><JW7EPMHbc-m$M`Ai~&vht;ioj}pVPS}{9
zSAS~~2?ZfmSv$0}_(YL0k!)7G#!joT3eRnUVqk0>ktadQbO<3#Du}XmlPWbP$9TJz
zSSuTc<OvqDLqepoCTCMGvGTHndA5ub=~+K8UWoBCzqV3l%?YWkRH-omn{0JrIL((M
zWfGkbIZ~y@<Vdu-aa5F8V~Yx??W_|lVA^^+Q38sIw{du0_cH5TovRVzEgGL0pz4&F
zReR5nZbq>UnG^9Anx5l!;(T%1+lL10<ROBseEpU20{fLw%$%HF4QLSKHBvN8f|9)P
zIMcFYMli1pM48JImXAgylv93l!gR@Bn2l%Ri(+nX*T_n$m6;W;npMgB_?xrgRGx0N
zuF)j6<U+ZswA9TRUvg7cluD(tJ#!gST6pTcXT4XE=Klm-^NmB6*De^PHe6fTHagT-
zTPdHgGb?gPO|GZ~qKw(4kdr~MABrVs8|{0cF6!@fWCcQ{{ZxI-?yRWSC%V?gWjAUR
z`J^gZeKDJqQ;%$kYV3<P=~)LRhE6t5H8<S)l|N;?a*S6<@UV+c15nj%aaS293Xc-9
zp_gx_z~W(Oc(99X_efe(*K3Gl>>5Jm*a_JD%Uo&5HekNp!CCQ>C39qu>|(r%i;m5J
z99^F@qqasct~Sis#J**8L!RL<6Ta*p2FzUD-R-HxKENNB1^z|@)9kwe{Hx8`B>0<3
zftPK^X28cj526#Yz~5|OCcxif&L+X%S_-`ER_1@#(A-rY9j?(nPkSHI@j&adEO3Jm
z+nTo<glY?~(YCmA9ql_&0oZ^VTpiWAMk*Ao+*#q$D&%!dX7<`za?Y++&Z*bdjP_N?
z#8T;@Evr84Wmy9?EphO#MO&(0S5$1hWjZ<<9YR*g=y3k2^3MzT=LG&anSW;U&wTzl
zlYf@*PaFTdgnwSoKlky^JNV~={PPL=No>d#@ed8nH)dZ}JE`tDXTj{YGZrm=q3f0U
zMMi+nuw#pShNVl1eQv2F8``<H$gXl%Pv^+mzW!<&oVNOE^t<y7i^;Jz=@tAH!G26Z
z?1?h6ke^tfFh<_r2jKBCld$)kJ8$07<@4s9<pTeP0576+Pbpc@O&aUG55DeNZKz(W
zt*fTXO~T*pjCqR}&vvnYTd)^##deFx0&a4|<pEz?>+kQZIwd-g7rbQg=_$zHZwj(o
z#CRb0jP<YXtU4toA@8-YwQb(A)<uhtaW(xTf&HQaEz(y^7W5*H7d)a44^Zdy6r7Y`
z_eG0lFP@!(|5Jfqq<5yu$ijX?0bTU<by7=li}ayRjfcNz9(BBo2|^t!GKqRW2xTv~
z&tI~P;tiH8(#|!%7WgL>&{K9{S#lKd?~Q5S2Q(?rGnOt`Xz$We{9dGZK>;b860)Ed
z8ES`mdaKFel4y53`@DIrsh0jHfm>uqbc@KsTO>N=f(^BSs#9PR@xIBcZhQG(fj+B%
zrZdaRlHtV#3iZffQ6`1i?IjD3Zhw(GPX1kFD58G1h%D$un#)*k)hRBCclVdhTe{#S
z)IpQ$mVXK0A|qp}j4Z@O)-4y-hI+<^t8SSI*wM_Dv?gruM;(r^R<H88#G7HZ*-P2l
z4#AFhEBO^79Bx_D;c#6s4?8iD6bO>5ss^Tm<ie%oht4554T9qcp>sIWE`*VXm!!__
zjCriDh+RXWI6|SL{7^a`&rgt2n3rTj2ns^b00l>YGB$oFZ(ZVsR7Mwtpg7TFfP!c;
zqiPS!_RB+siYrm{md}uaQRt<9`LZHOEwOnpS>R<MI8MA7;9OoBPAFO66(Km{I}^P}
zwDXGQ;WUvf@X8P*@pp+nB#^Eu0jYShz*mMKiFZo$ae;Jo2}s401zr<^Bwi)arv=ir
zB_LVm(NMC$>q3ykO(gn)K)SvJq~gf}8zD&I;1PXQAl*;`Qt@Pgn?jJpJtO*-KzdaP
zNVeia$pUW-K@v}k==%cc)g>SmPZoGn2$J|lL_ZNouPFhkc(TCFAxPrQ5dA_R-CP3F
zulQ+p28<r%BSUZIA=bSm1j&h9!(@WUwe&I~Yi{EWC+&6Jsu+)M<F|gA@X-cn4!GMx
z;KWlQ`lILzA#gEpQ^^8b*z&wRZ?L-@Ce<oLI{4s<-o&CHlhtEPB04nIJ=j$__~6Ys
z*()^**!Gev@V;Vw<^E7#5f6fB=g(5_Lq|qzWvs7IFMWV|WnM?#4w41Fx0uG>AJUlY
z-bee2#t^+3)7Vt9K-r~6pX8y&6Q2?5!4uMXR~YBJ%(#`ADm<k3!rJhxpoiGZ%8VK7
zHgpcJrHycoV~zIbpdyg-eXLkx`p-~flCAmZ`O=sWYa44!MY><HWPzV4hWc|M)UrVz
zRRuNTbelx|1SAXGQcMM34XHpj`J)p=1qc$}R#kAQwB2tNQ^9vaDiBwJ=yXv5B9ymP
z71(5fKP;w#ABR*R4hPX9Q31lEV=CB|lLhKU&jIAbFspLTK=0tdO4@0Q`y%R*?*-C&
z9!aY%L5h}TH3Y5l#r&{Y+ItLjzX&ND;o7hn;t0!?U-3gq;nO3}$INd-h1&0h-|?*|
z6cN?2TAcb`nEGDeCBP;Mutjq5?c#euLb#*<<CY!%w*FVBuh{Q}9lmJ#3gWn9eWl2P
zWm`=yF7e;dZY*C|V|(yzc{?uldtpz}7-GL;8k_oF_&57rxOt}?XrQ2f>cc@I(4zx9
zt;(~x#Ie)4t;%jLJy%-QfqWEeRf&FhP4te$_;qa*+gv(`pEUEb2M(MJ^BTF`Tsl<X
z99A06^Z*>VJLbW$n@ckU&JoSSiDvRMu$UehfCT5uJV<tP=_rBpf)bF97D&egAi)(h
z50c$nI#wXPs05_r1k&*VNN`NegJd_CP83M9N<gxE6ek5B!A&*~lHFW7MIfD80@925
z8MLL-0+8Uen+M5mE}brr=9GXmS0K#`K$_3j^B~#Hr8D?etf3e1QFG&wjns`61|Y$q
zIS-QETsli2Eh+(NF+YQfYYjkJ!q@X4+0CV;d@HncHXk*qrDgmKAT8&koRJ&O-g%Jh
z=F&L=>D&^K&J##0;*dPwV{kpsgJd_CUMi3f$68!V>CL6{xv21R4leO|;H<^&P63Ai
zndZPXidpRLEr$9&zAX>6wb;E`P$NdBNz_}-Vke<D(F4U)@Q#oQti|q~q5?$PY^y4e
z0G{am#Z-X!jl5Q1Ep{Ig6(C+`TUCLC0Yx7#rUH&$$?XW%V)u}!0Fgg26>LiuyY#Z1
zhqgx8Vke<Q(Pz`DD{irSlnnxAe(^XQej%iAg!I5-ha*gRU*d;i3fCVhS*An#q0*WM
zWlIRkSNH}%!4aT*byA>wJp=_YH-Lg8K>6mRK>1b(3PNZA1xJALok@Z6y$}?H#sCVA
z0Obdh0_BGxC_myG00l>Y^5aQ?@^A<WLRSC<M}YFkq(J$32+A+`20+0Pp!{-DpgbCa
z@@s*DBS87hq(J#y2+HpT3XTBfu}OjQ#}Jf12^1Uw%Hxv)<<B7~|0_^%1So%*6excS
zK|xFgnk|k1<sXv*<)0xa|Kb|}1xJ9g!<XpXginV%a;bT%K_{>RP!L#=hLYv_<C&Qg
zON*@M-bbcLa`HQ3DgSSJRHgs>vLOAHY^an4>97KEI~;+m<Fy`T!4W2cBONB9gi56M
ziO%6oTT}*)5O`c9NJxb{FKIn@-m>Mh+ZHZqZC|!z(Y*H7*^B2b-<*s~;g56`gB+d5
zZ##`dy(Ssp;-rcN3LQyO&5J+<ioU_M6lw0f6u~gNsbHAhU?*tRujK73(&(NkP7kfS
zKN#ILip1{#fv2b#J?M)5op;!wjfk52iI%Frv#&Br>{J{h-zfyOME%^;+tpL)>KtH3
zXu7}w@{}5l`IjhQerAQD=WTF3X!)|KYsZFa#5cRBm*Pb``&J%?vJRVBQ3n+A&&R4<
zb#}0?ewd=OJ+jiq59_RH!kSyHR9e`72|q_x&g&uaTxOn@8&vGSqMZd&16$oik*fnE
z6-$BLp)Pk+Rf(*d?%p*zAsi`SWJ68AVPwM~mpww!-I~>QLvLRn>qvzbfq6H^=c@`N
zmq&gv@UWkgPK7nnjx%-oALo#pemGHMkMs<7L;HP1XaHSHGWPD?VWKtY8&hQ})-AYG
zCd*bAq_Apb`B1H^x3jNrjM%e!iRTDdRu6KXI;Q1i+V0w59lF$Kh(}?~Fj$fJb;N;7
zS@61s2kUiF3rVl}g_$P2#_pZ#d%IQ9(8NG5Q!42~xrj=~Vf0QV@nIc!YsX>qUO7W-
z!^5?q;aZ*RhDGY^8X2Y5$w)Oug}6?Qvo?UDJ4UE^beQ(hk!trHTv;8EiYk`~%Pifv
z(xNJ*qhn}{6tBum6*8S)a*8xt5|5kDDKT*lH^t!p_?f+iT6+a`f&Rh*=_jQ^N!-59
zF69y%6IfM3Z?_Rks!BQ2F72z)z*^~Mb@kW!SJP|I9DCPP#s)`;eO+l!tA?~{S~oe`
z#T%i5F)~bTb1>edaF=d+F&T+LqdrpW1S(bAh6FxVPk^pqLJL`a6ZcV@SCEWeVjYrP
zUZ`4!D;=^4*`X3MqFB@Cq(^iP4b=v^6@xtsX*!2dCGl#<#)~#$Ii)s5t$bjRhJn$c
zp}}G9xMQ^ueV;*OJa9-IGnqm}yW!BEtKuFxD144;7pB%^ETVBy^N#6Q$g}jqLQi}@
zqv@#UlRclqtk>8l)%>d5{1k^yqexu7e+JN{qyscxsOF>>Voqut)#NCx=tr;BQ<)N_
zv5Qupn!p<Sr$v$*FtZkk8PAYP3Q7xhtP<>4CD<X#l`2q?UU7Ef<eyp=PxYv)5M&(H
zqe_WM5nW6{lZyK6MM{zvDM@w{no4Ht)jXA^2DYFv*(^qlgVM_PaZ+9yW7^Rmuwpr0
zy?P$@%vDSXz424@mTQgW{z3jjHvs+mADZ~!#sR+9^Dy7uu4$nz&v^40>dgz(ZvFX-
zNEr{&n+m-Bspsn@q_5?*w^I|4m%k$8<!7pwpQT>jp8s+~{G*aSZc@v+I4wMuv(Nl8
zp+YMu5`>QCPV*JW{An$z6PfJNxw@-5`}8?FcS)mlL9K6yCV;L!;!#)AfS!q&`=~xb
z^JeF8cV$B-2Klbx8s$f%$+SX6(40&IcuVg9Wv=a|8@p;%Gfpj0(x2}ery}G-V2K)g
z3BKxxUJU4Ast8p;Qo>jXy7}>UOk>BDRNrWbOBJT2st;tc;DJ5eX^nrN1gvUs-NPbY
z1KsL2cvY+!e3RZg-DSg*R58G^Twz(RTF2G=ux^^0&w0`gouTg0*uAHd*xY#P?-}gt
z=1&FabAO92bptn?=$aVJcqcu_)K{_0PHL2nwosY^_2Rwli>Z6`4UW`J2Wb=`32|yN
zRz#vIrys7X@*@iFM^KfmF~0kF2a3Ka@Se;KUUeO5|M7GOlhg}7XkhAZ@0rwrJEJ$y
zsC><+&W~ii8aJsvzuUb2p^-6Brv83Ao}N?aix74|nQl3{_s7N*O?);v{l7y=K9KDH
zdTf`S;Kp;3Fsr_K<+_X}+o@jB)!e*dd-AiKcPF}YBlgmq?%a4Y3SlNVHOzt&j=wEv
zU1-koBA31)I1_lSyV}y$Lx0eJ^g+WUa=4j}nM#gE+&&s$yT>ZD8tv<)^`tHs>ty;S
zq;+XGWjvw<IxPv^#3{^Ehh^lF;oh~q1N7Zv##9-WYo=Nd__B$G49$qTtJ^VqUq^4|
zCAxnVwzOmxlj!oLMsG#pm>3H%cVdi;b%E`4GL6op@jywZdsMMj-6%rc<{3SzrfVoR
zS;eE+#Qc`cIBQy8p-C?~jKo&c&4Eti@lSTIigjY$s4)2!oa{6e=n{IX4A!GjyXZbO
z3GU~hicMo!wVhVb;#tiuC($>y_vx^kRGB{=?!FxMtNCVl)$`#XJ*fI_5l%m}=u|;l
zc|sgkBkyYh6XJ6noZcDn)r(1d-52BSx~|?aT6RX<$9=i!_gFf<Cid2xDRD)WRx+x>
z;$|SRWsvF3qy*I88Qqn2wXqF@^pQb9nWO!j(u}OAiz(@Wu^bYP_)MtM>#)I)7E3dE
zA|*>QnII(*l*td6oeRx`fRa(Un-T+H+C0le0c%?)S-5EeU*Fj`N{J7Zg^}0YWXpkk
zX8N<1V}5Jq+-#aMkwVvWvU&5(w81nHOwQ-ze=(}zKQ%e;=T@P4i=Lee4?)@GB!-}=
z>3N5<+;83J3L2g=rsv!)R@^q{){g0@+sp)gkc-eiL*E=5Q{p38b2IeSr0>`aeXFQH
zHAC<3!g8nR!gj=*f)X8Md!M84AZ6yy(a&J}x*68AAtHJ+J*H;pBD}Jgp`Bv3^89?O
z8hl^touBRU8o6XdSXmfjg1mLHotASZNOFtAD)ct<Mhy3_<%gU87TYJGw3r19>&?Qp
z<g8K>EqQ9F_YC##rFNb%)aNvn;)yzwHM7kP^UqD<^AGa3$NEM2LH<Le>DVCuc~N|7
zkcVa0_EC3;FKnPl$`&go(Ji+30e%Z9GIxNtxo6c8#=4)g1F;&Sjb{u7wN$>?#5&c6
zH@<}_yf8==NlJ<Nfi_3iQ}%_<5#lpwGizsF`^Pi(#;T@O4YPNuVfId#iWT4HgJcOe
z27cXoY^39Zf^2Eg&vh>{`54KE_XY*;4GJFkH$Cud<RfS`p|~PRO0yLgNHVJkI8mi{
zDTH?^gzsndco5n^NU&@IVUfhX*$Bf)Vk1y;0^}PN$TupGAISh|LnFbv36Mq7J7<H;
z&GZ_9pA-1*R`~8#_#Vo@XX7M6!U=o}$JGjH#^|cGE0hU$G8K{sZ1*T^_b6<j?etWz
z4VVNQC$JUCs+g_2yiBYSNIL=VUIp)71@Frls<Xk9VCDq8`QuWZKdWj4>P`TAlLGc8
z1?(Fcz-$O5_&Fy)Z%&4!tZKba!Mjhv1F5H{T5UKbxH=~QUMl<1CK_&O(aXEB5hy*O
zy*Dd(Z&vVrn4vuzS_$6H35AzVR2qTa6Y%a=@a|Xe9?5`b!z{t!33z9YtJN8MjZR(~
z*WIre1i-yTfqRPr2inhg)ox*hf6{tzM8DxD8C-Nn##-!cs>F9UX*%R`;hF$SPQ2kx
z`vyrXhJ;o-6WLHA!yG5)jCOI)aw^VfSN{16mF{NWjGo3<{+_tv?nJ1ZQ_&9Wm<x3!
zk~Kz8=Mq9+T3Usc7X>_u5b&-n0e7c@(*o|{o`ryWGKi-U!jyn}@s-^YSNsB+@5p3U
z!)R}QzMy!P!~~>=5bth}c-Z{g*A)b{x?l3FA8K@vsf)!<;yX*eSHLUB2I+!U4{-1G
zz%7);j$6j!7w9y6^HmzY!rn<X@D@~gKyqIOlIWnMiga?iia9GO(tv1zM%1nlMF-Pu
zMXS@s!h*dHsP4}~^?dVwlNJrtd5AfiMAfNLt=Fi8cCJ9J+3-{l$^q3|vrrvw-k(4<
z!<<c`+Ne?8piv1ETme)zViX*2K=pQu%Df)A;*K(}S1uFLQfbYv(g<$V2!v40kHF{V
z6*=T*K=S!7QmyOHt6$TnR>FB`@$LGphwtdOR>`Og;xiQZpcO4SDv!`MLwe8}b<bkZ
zTF+Y&j}x5KptXUo{BPn)#z^oF{%}M~f1Q@%OImlrYUS5m!pdUDW$;x&#CK+ic#$g%
zCZ3Dkvk>tT7UU_U{gj9==PSD=uJ}dN<MQGyTD~22R8C&PK~+FLJ23CbEdlbq+an(a
z<||xD(34-0JnP3A?a#%dI(FO<OMu1Hs9=y3a-%%B`s>+#*Mqi*;Wozg@kiDh_OPxt
z`+0v5i^EGHj|Yg&26C|IbQ&KOF=Ir1Ta2MEWidF^6gM%s&`TK(z^!Bjj|K>C!lOE}
zCWIM+_uTO;RZSg*ED2y+;wE-h(vkq~rCi9?rf`%%e#t~YE?`<Xn%@Z2Z&XWTZCU_5
zQ!Z+2WB~sXIz!t;pw72794CO2wua;Rd@f*XbpSgP0FTS|B`u?A<HJcou+|I#-X?%u
zJ`vi@wMM*Hl$*3hfb%I=xz;WLQYYXYm&wT}H^V#u@+Y+0q<I2VP`OC0q2dgdcH%T!
z%3cAUs9e0(a&gwg;>|T+fHn&HO&TyjB9)8Rnliu=HF5g2cs&*ka7qDh(xS0E*wd_S
z0~Augd-jC%G;O^1m^jXpewH+GfPBiY-7Pe;B#a$kmtwThd+mxTu}M+T*rEOZVpFMp
zmt_`iH!Uyf;3MUJkz8P}rVHr>cFjGD)n^YE(n~+7)n^}H=}%nASkVphov84n*`v<q
z3)%(j@_GY51!ZpJBcFg5(uK5um$+vk;ALFM<@A#h@JhaNRpN?Yzznm;b^Lrm@$Bk)
z6F&v<Ud2a_cvxM(nl2>7!|Hl-@~j_f^ja>I>mp_LxSg*Qglo+n;3fmy_uHl|&219Q
z9+_Rh8bIF6k^s{OGcZNhCqG5B0i<EhCO_(31Bg(D`MyIk@mW&{Xu*K*!&&%lG)T->
zKdSiE=4=w5YYGuQucq)>qX;;=fbSz&_--~x68LU0XOs9`qll1gHHFWbM;?;;{g}mP
zfWSO*y8)ujBcG90@0v%1y()yo7)nN?3${@6s`Jk%|7_r&G5)!Te=gylOZn$={&@xe
zT*W_E^Uu5Z=R^GS3I6#!|9p*pOzFy`pz^;*NSp~<HjYpu>)BYbsa3Ww)Ryw%nfg<C
z){Zd18x{s#kdput1zqSb=oDShLQ{~=ZQ43PxlOV6HZ-<5x|3vx-oQVvBk$bS=q@^J
z%uthgZeCMULViwDFmgb5I0EMgjx;@oFQ@5mzC>v5W=Ibcr05<xYK`vYzu!bZG^0B-
z9w_v;{7{<6#7{96=R(e<82eoaio@jrD8KhZ$>K)g-LF>FT!ih+lPck{5EM{(K-D+`
zls{T1(<2Yh^PdBS>cH{nfB8O?_!nDZ^meX}{9I91xahC^q+yoxVEJ1BOZ0cX0RsMm
zj{xWY_$XcNVO>Z6<R^3B{EM&WwZa{~Og8||llTa5cI2bx;HXe+rj3c7%ugFg9wOD9
z_<9~B37?C07Dx!bE)IzoI#j1@=+jyHcHyEvO@GtSoC9vx5IBk3j#>m9;<HPEizH|}
zdU`Q{yM+KsXm_-`0DQ$_622Tj3G9ycOaoM`Nh|(IWH}V}HwkPxEwh$(4tEme3pqy3
zYH4ADiWxJQ>|w@?l5{y-M1V4fZ^5h7y%SBH!=<vaCX_yh3!MxK9WK`}g`~<?7s}D-
zaKVatgh@swkV;1tJZW?~T>Q8xb+`~|VI2u-9lm6||K?Eq&)@*X3l*i;;fJX>DOwID
zk7|csIeBzDT)Jvy@mQ;IQ0{PnqLoRa-r;wYrQdPz<<js#ju|sFzY9jhLJTe<Klycn
z<txP<7RzwRhvUt1$Qw}6pW-VO&&qJ`(8MMY&Jm8VQh63Xq$W1+%J6_tp%V5TJ)3Vu
zq0g~}7F`@h&*f*Z{5g=1a#k1z@%6mrj|5{x6}}aaz`oJk1m1Y6i2>ZKHavxdBYHlU
z8(s??8bW#)-+;B_bU}K}Vmj9x(&R#I1{dXk^Q^OH@@;u)6z-1bNC8*nqbzV~&T4q{
z9mUUZ&B`a1TJ<z8Z<$zT8K=`tOt0IA(VN!j>2=5P^*l9#ngex=Becd7tQx(k>rM$3
zYFGHD@~tTJy2W&qRX5i<X@FE3vt+|Ljf-*`j_32^^bk1P|K<od&}GEHHGa^77<Lo2
zO98rTD28_{5o2&mnWZCe^{TkfaAWWU;^D`LHLmfF=KO7y8eBGU?LEZf`~pukc>3>|
zp=z)r^I1|2i})y34T;b8jQ+crpTWr7%15Rzcv=QCALRB0JKlBht!TF3U?>hL{b9PA
zi*lN+M_<e(kPFU^cQpaGhL4Ja!+5urpV@B5yAv6`sbkN0H^A5P#ye0iV7$W->UhYi
z(L3Iah6=Ug-Fm(ig@WWDt8R*ncN@7Vr{Q>7^@SmDcD%btz<~=P25v<U*)UsZYo1tu
z;w$kdjP<~47&w2%`8xA87Q;e(z+>V>k{{%czDi4AxiwK-vC((j)4F%~J!cWM1q{EJ
zd0K^MIJqL!D(vw43TYKr@=>f+6d8W6;%6}YzLJkj*Yq@04r|Ztns)fTjc-LSLV$g7
zNa^ACb}q{4MIL=|n0PKYJN&+0z}?A5#ld0teFHzU-44GOFnUwRp5gaCzMePyzL{^p
z@QWkV@%?;GsnI+9J`gI@j&Kk1ttb@1>{)f=AAVDd9*Jp>-p%EPTM<W7=ZbH8-TMUT
z`}rsfX~v@G1N;nn-3R%|)QG3oah!23B-`u0z_&tqw=brn;*iq4?n_)$SYKO0;B2q^
zih%nn9~B3OUiUSAX1ncmZ(;PNjy=8ZJA6H_*ICzL9HEX80j<>NHLZL<RH${DMlduA
z{Q*CeRX6Eg$91DN6;)ufk<l|i|91$E69El4KP(Lgfy_B@eiVXZ-IWpX3^+e-9*&A*
zMo&lNat@@QgdkaWWP~^a(!(VnA!-?r5T~32>8Bw`*4G#T&4Bbs2}lT31|)<g=Ro>d
z2$J<CMhr6`{k#Muo9h)3$vKdI5rSlWh7r09NWUxr2{Fmg5~7fEApI%?$+`t2Tp5ra
zEddG9$bf|S;~Yr84nea1zX((Yq~DZ)WGfCa$2pLG8-iqAd=Z=sNWUuq2~o$;65@<=
zApJfB$$If3AQ_MzD**|S#(;#_;v7hS2tl%rx(GrBq(7E`g!p1WLPW8F<XIH`DFn%W
z&?EF1knULA)ZCU{6#Wms)mapI!2NFsoORzt)H1-`>401D7=O?&uGQ<EYl-%c2+k>&
zO3&bEA7x9aRo0Hut~>1wscaE+g17*OI(vsXk%=KUTtk137$QP?2o%1wtqr!qepN77
z3z^inUZY*AZqMDC>QLWbZJA5#zcv9$3vFgbbxJyN5%#R6Q6WN+La8yP&8i$TQ(aNz
z(1~eu)H|?d@aSp!3gRB<!pdoU)P8h(cdgzvZI!;NUtH%KWM53Zjjaa!Z`$x+UyVy<
z$#kje3SHBiZ+y9@cbb%Ni7K9N%P35upmIv2{e`~4gJY$l{8cTNuwLRe1sqR^UH2C3
z$#VjpS(OoL19dWM_mlHQb?I=eZJ4+sYu#;}5{K`YY3j4inTzTpE9Vl^<jOkd0imB&
zl>S52%B+f{W8m|ZQ!1^rt4(=LILAorBFey%a*MNtC86!nleZM=D?9UTc`IM*3621G
z^p(4uzLGF=q!+0$$lUk@>MfyEM`IE#AMII8V|#}*W}V9sQxA>Z?PzQTF~IN8GH;-3
zkn<gBD_D#`l=H;IGC&cZsMnrnp484h)WTAp5h@23_p`LwE%ris1K&fwL#FYfOi!bh
zE-cNQIZ@TXL#9uDoY>7%@ZJ3wv}L-@;#1lnT2y3|qFh!iwfviq&bz`WO!vpMwz$*j
z-dm)9JB?|7iW65>#%C96OwSE9ChPhS#sf5_d!5Eqr27?1Oa8oKs1FIDw!r~lM*!;k
zN}%3qX~~6TAv&U%3XTk^z{VediUBIPe_K|8FgippET)1Ng;Zce9Kibk6})v@R)Nq*
zL?;zf!6_jX*eD8+Oh5&1cT}(~r6tdpyu)e9h4&(womQRl2MG*Q@PYd7muUbs-yI%9
z-JFoZ9WD(h99$a4Q0Mtxm>(+CelLJe1BE{5)Z*0l0u~0-roI;_mgSwTOhAEk@3#HD
zAe1Z7vUInNJbk4t)K~2H0*G1AS3t~C+^D@xe=i7$OVm+JW2-|Nv)>D#dV$8??T#)}
z-wXWx{O|U?AQU@M<Y`QvO&lhn$^BfndrBbUL1XIYhqAh#e}7c(DS=JLOvfHL7lh#0
zJtc7D0M2k}IN-R+frBW~JY$C4QvwAJ;EXg6Cjz4l45i?($$>N)(vaO#0>up=tuFxy
zbTxnkf|?vi8$yumo)Y+L0BK_hNFbyEBv8=gKteokUd7oxB{0+g(uE};*{ul>&E!Br
zba5UeyQc&m8bG?Z1SBxaKuh3~$$^Ae=R8PuPYF~rfb{YbkU${=NFa^LfrQBGJV<s=
z37j&3bXf^Vw&K7GlLP5;E<F#D-BSW>3?N-m0uqQ|pe0bh<Um4HdLAUZrvzpgK)SL7
zByha|64+gGAR$IR50c$e0{IIdy|M%(5V`;oC|nFA&(abR`+1P;o)Q>d00|5)#pky4
zp3=1}Q+QA5U3^;}IBTc_M+d-xqoX-+jbetn4-`WU>VrJg)=&o)51<B%N0X?xnxRg}
zK%$QqQ^6-f6=w}~APRvBKI|;N6Q+ywiU}VqTTKPRHxhlem<m20Qh_zpfr|tx_{g@b
z0--dCzEVsDUkj<g8tOoI0u_ABQNgxksLN=CVMCoTutfirR$XyJ-CTay4CeQ@K52);
zFW<~;Dm%AEpoIB$NZ}6K3KR~um8_8`ZJM;sZxVUt;~Du*2#Ukk0w~~X85fj7x_lpL
zj`$XiSc?BCY8Io_DxFMfRngV_m9`FhK558Xs-q*lef4T$<KOIZRl9mRN7nZB>!g`a
z{VLTN*l2L1X>F8{(^c7<P<G_Rc<?igBKg?$`>M;`q@8TsB{_$`Y?zFrIvwhdQvGht
zt!8I=gn|I?54k7uC`)B%*p|71&V7oCzt+s+58{v0NP8x=wl3LvIU|k(CI2{9a!$&w
z=u(Eu?~EjpApG8+(f!0@V||tpDFAKGDZrzbq0Ly5FRcsi2$2&<z*_Kd;My4BYgi)a
z;95S?HFP~4YZj*lpJwi|bbiDAs3oZelTz)gq}UtzNLN~pYH4o8OO`8^)#XadoB82~
z=;fs4CUXNvw+V!Y7h4EgSNZKm(}^9xJw>%3MHO|;sQxp0o`|k%*LqkZt^u_>cO13r
zPcs$8!X7fJR+8OEZxjXG%}2UwSpm_#d;<D*Q_~+tf|)c>N!g=Y_{kxDGTy5;sj&hq
zL<mU05#&F)3Heid@S78WALpp}#2BUYo}etkON(S~G+iq7(kgSN$lzjU7DZ>o8oxFn
zysmtQtElP9Pvg-3Qa&i4GRlOq1J@7g0X#($ppyR1`nm)Tm%}KHWeX`P!Ns{Vqhc)4
zBa7*0++(=IE`*YQ?l0L=2Zcv!joYCHLg~NEC|!7e@=Di4bE*e{rOVTUoZte~T#+1g
zrZLZnIsKUimFPq?#Cm#AWNpQWKfrrio9MZ|_vRC}!w&E^&Q6eSI=V+$i?PkaR)pR;
zPZ?G5CCp6424pt*ny;^0<FQS_R)pg|7g>lq{e7$~fU|kmeBi7Z*=AomXS{V;fF<1t
zWyq{nYg4v~*NRY#=PFNmPiEAq%Do2F*=%b*)mhuJ&AC=2uURfM@(RsMy9QV`(V7pI
zwJ+OrYeh2t<-&qTX_msA=!VR?v}R_TX{|_-x?E%#Z01gQLnbC`XSPY!#;vU|Z#n7_
zQ4X2NthpH>3>am{jm$EBcMcQvTXL3mf*mrES*x>6tu}$kJY3{X%tIzJYkam@)rxTD
z=eDqsY}Ron@F9(Bi@UC|3Dq9;uMcuENhNx_F4?*mAtV7SkKfpmh2j~+#7{^fHGKSv
ze|g@bC-G%y{XuL3wclk4w)GRJK`QY9FD#Yy*?^BXs_QY2RQU8CciuFB6yM^)Anld}
ze1`$@j~<YPXv6cro0A_$x=q{71)C%7@r?Hu&XERPh`VlKU05r>&5ZWvEG$JcqXCQf
z%@)RD&FwZH+F!D;6wQYQEaJdg7>l)g+bn2*vse-|-`*@}c&~VW=Dj!ZErj6d__vVe
z&i>*TX}ycDSRA3vewoj0;t^Qo5Y>5|cTonT{Z&{xWd{HKNo1xxdxDs=%~X;}o_(ZR
zUFVW##|p{k&xS!3V&e!VXdLP9w#efP+rmbwyv;qwii<*@#t)f7J)5w*g$i}dvnUki
zSygD55Aa$T0jMgu<!J8VJ@~DkroXBE81uIW++HDYj)fQCVBw8{n_}uQq^T;vCXXjA
z9Wpe5F*&Lsq&P#wz8>x4X%!wPG&V-(_5;WMj#jZhKNM>f)E*z8_NNHN)0=BP3wusO
zWg~GeB+D5BV>KYbSX~?vo6?fK=#XOiVnc5(ILD$5aMMfZD`v9Rx>9>$+YB*eR@MME
zm`6`%iA^1QhToYXH9CfIsPRavMlWZ`3j&2EtllW`Xj@{DvDdY18v#B2>X-l)FH!#s
zOT!5f^&cC8;~2o9ofkC^$2D>r4SD9H<3f<^WC=SsAi)lv)y>^aZr@~i0>3ppS<VW9
zb4=*~2UB_}aMs!$om>pysUd)lH68$AjV}jK4EE7!X@F{Enj-3_P-)^RpeR`4B+y$?
zG1`L;b-ANzDZ;0S`Z;b|#Vk(|^>+-7jr0&51?z<WneC`4qJGxK6j49-kb%LGN_})_
zXpk(|-IcN0h;R?+0cRT^>d##nR(bj`%L32J@bu8c=4|+4Vw=Mcsfo?IGMpbO)Y;8N
zp=a1an_L`Xlh(-SSsX41!EyF+0cT-pI3YH<GedBk&0D}Zt9dw0u*oe7L2|Zi0cmjw
zNX`~)PNlVmAUV6VfV89pq~dIHOGA*H-C007y96W~KoVk;TNZ-k?92kv@)D4o{nng{
zYYRbgwqyb6oDz_n?baMf=Y}9TJF<XuUI|FH;;_$}18GGFlCu>HNG~Y?$=P7df%MW4
zBxes6kXDv}<m{~GKsrAJ$=Q4bq*Wy#Ih(2mlIJ`6Wg$rRJKD0zVc)gXm!LK}cko-o
z-_ff>;GC^n=nGr7F>q7t5L4_BY*gGtKC@tl(8lj=n;k-S`=fzkePt-rR~&Bu^c8pm
z#QI9Y6z?~PHMV^=xr>Ww?ByYiIi3d47(5MP8k=I1+q!IWy7O*5Hn~?8YfRUK8k6G@
zg2n`okXU0X(*24BsctHU`o<7y$BzW4;YZRW>a7-`DefuJEyYxDTSx_t`wCP5_myo`
z1>zMG-C0ZpcZF2oc+Efs@S53HRUpnd(S5~KaDPYzj<XI_0B4<;3bv&n)r`qI9Hc7#
zMA6&Rs!Ie`<{Ike^26r4!xN<XU`XMPHxd*MZ={&QbH~eTBD>)#L8=J>vQn78kCuz%
z4O1pE^zedy|NVVBFm(3UNKbIJ!jX2hQZj}ELr*OHx(Sa9oy%+LamN%nUPDmC{8-63
z!HW|f7kYYWWo`pup=VP&N;v;o&2Cd+p$lS_s~fv@M)SEa>aft|+)tF#q8k=^zKE{l
zm>ee>Zd8q;LBIWdZcnK`QBk2CB-cbmg`QNrZKw7P6Y*?O@u1Mt#_`Sxj0tU`=2N0`
zY(HnAP6T;qg0|*ELdT_$iH!(7v)OLw^)h$dSTK4o@|SFVRD7M>Te{<eg3?<vN*4!{
zywVe1(9J}Io?5&m`vNl)9f;N}iP>F`S<V(B#;l@YH#BDXUOOrAphcrDvoK9sIOy5q
ztE1{<MuWDE6Wb&QgBA@u%c|4Kiv>MvoQj*AP|%`*+FAOVv`Eke<5brq1cHuNSCbY8
zI$mIt76v+Tz$Q2fv}j;<X0xB1Akd}$CBXy)fiANp|9b_2wq*&nT?c`l>yc`k3j$r?
z0l77UKwp~i{wE>`^!zL=Pe2gp%d)UM0YRW07Ry!-0{w~vf#L`sU0=%$0^Khe1iEwr
zgFtB}$&3M=sn$Bd7*JJ4eh4T`uLuFf5p2yk(qC%65YRbnkZLvrbS^(+3iWv7oDnM2
zdgFkh4@hB&RfQJuG))JuE#$X`y>ZS8fwLw@ko*B0jJYvzQz4++O9<!^Ppj~FBeUT-
zw;%XIK$r1Du~w1T$QgGCC>w-xA^Ac;VVEusDcy@+R!m=P)y)OxSg+BGI!owl`wjt}
z#}b?ULyx0>Eu=<Y2<RHCMz5oPZ=le02xw$WEHd`m9Z5xQXaw~1t91b^-Vo5f(r`i{
zp#32@z7WuX=HWCE0y-FiWG71-0t&--RyX%=n$zH9x2U2E_^si|vK|8G*v`?-VLLAc
z&YIVw(P98MgaA6`cL0R>y&OQX&_`owfKwr$mBHakC#NQ!3IXk|R+bOdx_UeN`o<_t
zbMHVG<>{rE*wuq|6YNRpx@&`V33Z$b0p$=!%DX)k0*d6{lnB1Bk5Yt>QG2oB(o-Rz
z{}v&jYkB%G3n$OY@S@Pf<_iJ6m>*ITn|Eb+X{b<V{}!|5Wwy{J7l)w`(91(`d?BD$
zl!g-u0ewXXjxPlC%I4wN5YU@%*nz@1=wAeGUeg>r>*%XO*nF{|uPlMBcr57EAxOSh
z&}&LSDjo}ZZ3vPt7WBFjkZed{C>Hek5F}qLXrlzA;<2DNgdq81K{u6vR6G{+RUt^e
zSkN0wK(ZAViUoaj2$C-r^rjM!ipPS!CIrbB3%a=kq~fulH-{kkVnJ^y0jYQ_=&d10
z_Ji8Sf@1r%)F+{i1$`~QHT*&Sx)3;DEGTwxW8kJ@L8oFtd2?;kb{Y%%=3;&2Eup^R
ziv@*KK&-DMZ1R4~*me*L`o3Zs`#?xzzF1H=9K<v>6$|=r9t-;MVvXq&p~mEk1%+cs
ztT7eoe#K%zKU)m-=R>G{v7m4#X%h7l5DWU1Vk-DrNCm!FP<XLyt17Uupx-X0g71b@
z;EM%?^USuY0vikZqhc!fNk|30SWx)v#8mKa77O~bwCWO}l)15>m+-^JMB|AC{Y6OO
zj#CnhYj8@6DLi+)d_rPDk0{`c-Ce5<)oV4%^#=E&3%^d9g3lF>nv4c5669rG=tj9o
zj7r%&p(n>O7ZeKY^Tzx*E<{kpsj-r`W;|$-Op;mZ+y){-ucNlKJw$|_7OUPu4M#_W
zHn^|&BSLQ$(RFl{<5Pp^s(DRxPq()`VKJevCCMf#CiIkM6p#vylmUb5tcj@5B2m`T
zt5R>Bz_`$+>OT!+LL)=Rsg#M04L!Zt4(T1q9sd@L<a7NcTj!O$=*`4f<~XOI^!XX3
zi|0vR>7`>siv;ybd-jECCOQ(WV$_;Ob3f?u%h_GT_*EooFZach5+7P5)G;5`q(z7p
zNoAT3Z8Bp-&uwNj_O+wxBu9xB39!vq-{i%K7KxP3$2K{UqD9iI=Hr^QSkW`b-QFf4
zT6FySnzVS)@k5)mh|!6JHo-BYMS@ua9pL}5_a$(0RMq+<5S9UyhYOoXB~Q)bFbQEb
ziV~6_kz^8+0V0^!>6z|iy3*4<boV3#LB$>I7nc`s!4+{=#03<eE1Rrw!(Bl@T%Y3l
zl;`$8=et$+R^7T)_jXNBGT}e)`}I_ARh|0ox%;{2J9lDOqi32Y3teE1K2J%0bg@QP
zq-AW6W{vh6ay98#qpJ-dcO`3dP0IIo2Wxa74a;s|jh>T+WjC-!*C{N!oHhC_$r{BK
zmSf+}WQ`sfXN|7u5^GeBC8@m8J~8jf<c&t9WHU!$rd6g}Sf-H{tl~_}(Tmw|)hToI
z5`HNZYG97O)FxE<?kMIcY__g8bkligFXOM;eRp1A1E-mNp+=Z}BjBc(qk9T-^fiWB
zVergl19fIQFfm78$1g={MQqz=+{{rn9A`o@F-Kv*c88Q?j=tGVUu^r$1gBZK(Td)d
zLtlF~bM(cW<8Zgpzz=w*O^qhz=)05}jr@SCEJBma(f2Be9g+hcxhs-EL%X`#g2l)j
zeP3=kHs<L2ZE#G?(GPSErvv8b2W^nlV5yj+FpH-(bF+UObxslH=!f~M_QCR_HgK9{
z9nBn;^<3bT@!h}14dAslfSPe00AZZZ2T<DY{ZA$VPBBNPn4{%6bG?x=9ez2*91W+4
z8@x)>+W=lHofu1J$$RAsizB1e@*+<WA|-!ziaA=}T%KZ%_Q?t%FOKyF>h)^VdsHw-
z-_65E#?0{3wt>yW9KDWT5(ArYX82i~P;GA)qvZ`sXor(S8*}t?HaI5c=zr&iV`Gkf
z-Ui3S9Q{J)a1?XYN*Mj34U&m4`lTF@+zF#!wm~uxM!%8+k~?AaMjIp(Vf3aPkQDFG
zMi{->2FXMiy(I@Ecf#ncHb^GI=vQ+<awm-5W`krRjNYCDk}5bGVe}3gBokru&K!{3
z38Qz}AejiGcjtiQP8i*3gJdF%ek}(icf#n`ZIIMDS`kLE^O|cV7!XFk$zQdvqwleS
zGZ99yc^d&YMHt0YaMnn<*(`1<H)k#K&X<$J7s)FQKQpUQua@cltg&XfF{?LzFYuP+
z+ryPv^4=xtu@qs_8!SWOndNG=J~JW}^4IrV(nAZI9{s;QG0mMh(;IEnhbUj=k~g@`
ztBs9}ZY$M?yqPn1_#Y#mR`cmld)VQjKXGd-57^p@i7*O(fJj@3nc~d`u|0q=`dc@R
z{m!N_6JZp-1`&--5k?>7gwemcRi^*5RVEW*6h0x5%H-1g+z6w4eamTpoz9PC?PHq=
zqwpl@5cO^#jPCEIf*zX+OoUN5uuN7JD8lF=ZYns`rUDaT6n-<4RRxMLdbpbkj<l)3
zL>Ps;PDBNdGQ#MSld5wkj9$<EVQ8o`5Jo+l!Zm*+*wo;U6j69)f4N%-qjMZwvNx4S
zMv4Jj^i!+`_*&trL$>HCu3yv1Z<4T*mn!=96@+_2jt+fiMDnhriZ0Hp%Spf!y@_g3
zj9aCgu_o2Yq3@2AZ6`d@Tey{&d7?XI)&riE=0U@4tL?(gL#x!=$!;Twev`7*Et2RW
z=bD|MNXmYqLx$*c@_(|6^icO=Ph`Bir-qJ49o;2{Ue##}bS8#gZI-NjQ?jU=vCd2L
zJ;CVs{uJrbu_Q}+Zepkly;}4eCN^KJ@z7L7OkC`&P?tm6(!V+(lu#GqU<M><0W1(l
zZ=~32l^dqxc(K!X4eWNJvlrDO=-9PwM0D&fw5beS6G;Yjp{iy;n_woW3#lsu-NeyA
zT{w*yz$S(R>cUOU05xF*(C0c2v!?#3Zvxn#<J8oIQ9s9tYr=S+UBlI7(x=O*y;(JW
zVi=zvG|vLM!1(;IlKkjme10@7V|z5?^BP00CLQDRS_8;k$@u(a%J+8%<MY#LSat*B
z^ZGO_yMgifS%qbnGd_2&CaVgFkGR5wdcO0FqFUdyDKU+|k|ytG_4scWe=es#{bOPh
zzn#K058u>2;<;@6G|ff=gXjzI2@Ikg7;Q3S-cAtK?`S{EAzOpRX1ULZtmp^EDrE05
zB727i>$%v{Rr;f}k4**DCl0YG4ApB}%8h2RRjJo@2pjO{RBZ@GA^(Sz0T^2Q`&&CM
z-61N8e=n!rM?WI990T~+5TGKi7+1$*!oYM{P(rsoXBw4||J^&oT7fI9zQ50Dfeys^
zchJO9;zP&AdBCf>!~+)7i&PSD5SlTQ1S~~lGk{@wM?=6Bwi<91tk_Kq;Op2VnYD9T
zzdO_;t|0V!ekl}cU;y7>6RMoC6ayHhc-Ku469f3a`KxwktS{KWDWkJGI~Lab2)HQ*
za7gcH_OVyfqBZ5#Sfkcl)B}g7YNgr21$$!=<pZ{olGh5R(|9FxhU$tF)c*74ZQfdJ
zY-;WR&CNfXd!QS+clnB;Rv4Iyy!DXT4onQ-oB5?kt%&X8jhg|?TLPJoOblRbO}Il!
zGJwDCrY|<^XM)rAFVJs%D~G=JYzFXWILG1kv4H{nU7H$B4B+o6H5z#yKePx<GJt=i
zBz8#hurYvtY{6n=0N<M%j*S6)pAC+Q0sNEB;dH<NzTXB(4VH=ljQx+aW^Q(Pq|QlA
zr2Gf@tM<Y2AsaXo0~lK<xxlGi7XRmN0Dox%sO`d_AHpt7K7ewg#{YE^;1mORiUB;u
z08VT|PceY$(BNXrt8DbP)yKSMxrIZ7H?+2mmN)RZup7i-%Nu%;4yttJqH<&E>{wnt
zD3bW2iUIt69zHT=hQF~5Y$gWqZ}}xLuo-8DzqbiBF@XP|gmySNv@wAHXoF*70RJgB
z92*1p&o(&P);;DD|Jym74j90Hu|YC1fd85Uk~;(V5gQ~E1Ni@PKyqgQ|IG%;!~p(#
z4oHefYGVNZ!v@L30RCqVNbU^af7u|J7{Gh69ov1X%vM$3odG<JrDs_mO$^}aIUuQm
zvoV18wm~v6fcMD($(;c_!v@L30NytTBzFezel|!Z2JrqlAh|Pu53oT}>uALQ#!kF@
z^-h}04&<-e*U<;tz{wpK{|~p4L<}w1%#VPZVgSRYJjDRsLywF-!mX_wWos)Y1~B|X
zB5fsRiZ>g?_5cR(Y&VU~wQ0=60EX{LL}OD7;72(F_&B%9biA!HnHa$E5sFkMm*(fj
z0DguW>Sx+en;5|GEb0*TZeRc}c2mI;n+i+}U^p~QRuw1)@Nzd5oMlsii2)2hs>!MX
z#Q=W3n+n$2RA6EN!+k5Ff=3wx_}rxG+!??R^RhWK)EOASFR&?G^H0On{ycsuqVUZA
z(#QZdIrE%vg95KKcnIMNP}WDF7<_q(HkoSWGRTiYCLF-rv&YM-_G9O<|3<GD_v8oy
zEe*5i_%f?#H%7k3XTLfBKltU067+3Z)O^X1Mfj4rXHmGD!ILnA603sd5I?W+ucHi*
z*W1&{$C3VANi+Zc@O{6J66apJgFW|V^!t5hhUD2tgpZ;NhkJP2$Y99f-v8UACiuzW
zD)5sNSq#`-+acN4e8S*y>B8a0*K}^Gq?hI8aGw`BZ6pA#HoA=-g8>IO?(+Cd$~~F}
zyuP4^7e*vcHV(H@my>|QeK^&k6va*|4)=dV%GRmY1ZVvo$*si9;r@4-^@pg{k`Jop
z1BG_tLgqD>ZK2St)?3X&A;vlMBMws6&}RiqQ3648y=QG$wco?eoT%DY@s;o@hDWS_
zH7CFw_Orqr_VeNRG@G)uD>>XE_laF(*-vn3b#ThB3u!s!le?J1orZ!utXt(;D^~3N
za8~x=@5IF(=QjuSPjhBM{Tq)uy366-&}j>d=W!?Hn%bD=fP~6DXqK$Ji=0OV&3g#a
zODWQ&YfzSSCl2>7?w}fuu?SZvH<u&eMFleX{*w<bN6(gR=-2cGNe74x@qt>^tnRlA
z1{vR|G9D+FwaD8x;@02q?@MzLe@su^pHsK6XYQX!rlJA#`#pSTE8qD`?2efr;QyT=
z@fh!4$nUV9Z<fwu!;vx`?!{~XWNbJsV=vOPfQ((Nry^sQFqC~sW+G!R<~s+(?wB(c
zxaj#W;TJl}l``=Co6MDO$Q8VpU#8=K@3ZK8FOR315&B24c&Ppr?jVEiNI4fA9fEAT
z(95mW+96iJG5kgDC=KJI@}!s<;__$lRb+gmqQT(!4h)Vf-#ksaSyc5=UpXy2Efm*Y
z=Rn-CXVWqMz?^tq{hFGxh^7#fhjM3_CxM>Z6`;0&1ikz{UJUuiI$~9R4)Z0}^SXl7
zAwt1%67E7rxXK6O1PS--uE5O>S#Xj>da@%@<r?t}iPR-bhP}4B#4vcKBTD5hajHb=
za=yDAWoA%==kUi+Pfv3Ms(dI;mq1;DRM>&0L^)V0Aue--sC+7(D<O7AmyTf%mOBDf
zo)%|y8E9r4gq8d;6neiSQ00HIdcuGj0wb)EK-W3~RbClqOQ37I(k+G5u0pLiQjV=u
z{BtGDbvBs6<WqTSoEO6Mt7<kAQaH1%Ji&TBG594WxCQx*4SXe{0mdh|gLES~!7b~l
zm<JBCkP7`I=7CkdGZMR#GRtc4lbCOOS}exz7jv#Ar>9N~0|IQ2K{e6c%5Pz!dm&$$
zGI$Z)NM`U7Jrx;zDGRxjei9jc1>bpP?2b8u;cV(bF`Jq(O=R!2{JvxM)QopKzlH3*
zp0Bj*VaEGLx)IABX1qJ%sb;ADTUjWV#mfDr3AlX$+_&)^N4RR!^qu?`;MP=?mes8b
zffXzjUu}ebKq47ULE^tIz9{JTp}2OKDhb<I5&UP?iok+O?M+@QQ8m&~y&?R57;6Bi
z-WaCFQN3v#sML1h4H8u=4b|o0_hYE82vg&z-aHOeY6tH|iE4{N6@H#K1pK#!pBJlr
z|5knjrFCUz2m-^!SYZ3~FW^N*lYhqeXDk0~<DVDt&n5ix68^cAe_qZ%ujHRs^Uo*v
z=LY`y3jf^BKi{CAm~rHGMjDNuwq0y%9C0S^S(uC*LFk3=u{dsEPN5gX#4AH)5PD&W
z!!Vf|dLcBwNRtct_SvFue?hR6jF~B<{iS_uDz=OW3>T{EAiEYcl8fj@tS5v;<3)NZ
zI{J$#wY^V&y;k0l=;|*q-81)~>%Kq*w=y`j!y7I)%3gE0K31i(F3R4fv0|f0=1n^J
zf`~AbP^o#N#Y%%V0=HIL!yv?Lq<^;PSj8rv`$5lCYMYSGr+dicVvQsWljt&;OG`3Y
zJuMba;f=nYrNvfJ9Yx6JbNJ1Atw$XGF^h!As1}7viB1byC=LjjCFJn&A#_!#ZLA*`
zoCh)sO9{GhKCTMK6-wpi(5&^rdwcM?CA9HK&FQQ;VSPpa@nWh08Ka*uD8~=EO~bH>
zD~y6a=Cti3pC$#7pd2q0AA1pI;{9V~|6$soYj-OshiLWV3d$ixWC!KIUMjMTE9@QP
zDzLhm_VIhT2fK<?;y*x5$A2$f_4`-zzu(7tx=G)Fh0p`}rBJA0BmXfrp=#Exf^uMQ
zRS30jx0`};9Kv6{g;0}$Hr#?YfIHL%&J>gb+p`gHQ$acQl%O0>G}H=1P!8Tp&TI#!
zpd3f=OOaYZ-O9(P{s{+*aR=q#J>E=6rl1_yYj%gkD;Pq}y=>tA@>Dl{@lI_fIBi23
z{l=Ue`iks12aQ@dv$QYmp%bu1X;>brlChcwWTL&&*;ZRBG!%M8ui2tuy4WarTZ`Lx
zydP?mi><O(ta*!1U*e6?aJs3dQllwCg&rI#$Ku2wUW{w7nAgD8pY;2NKhQy(<FGR_
z1m)<psnHaa<5;CeBlU8DMQAc8$8k!cOW&(JjC@;Aj)fL1#-JR>=Z0eo%5j1Xj<%1D
zYI|bma5@Oeagq&^8Z1>%4s2+rHS=U$Og8gV_^bB8a*++3wkHmKVNX04IJIr=Kg$i^
zvuyyiZFK;|wt7B*a_8MYEeUWcD92P#j;Wv=33sZgpd4a$G8L4=D^(ihp;mR9R~#KJ
z*Gj?^!?qL=;gc;ho69w?QleOYh{*A%4a(8S!$-!<@O0b2rg?5*U|Yg3iGgj7Z&>@D
zVG}BCfj)e(AQZk>k<oIB8;q(s#SLbI(Y09b^8B)Wg^VV&$ZL_MwMkwXBNVQ;Z*`+Q
z&?r`F<<bDRZjLw98yZKOi#%{4msW;a=TmH=URt;{)~}~O+nTh|yU1HL%F%v{Rld32
zJJn+wF$FIo4?}uY-h>RiDIbZGJCysLzlIxued;@4Yb$Ek6YhOzD{$|Nw3V1ez&tOX
zwA^6n8t89u)7YR*V`?K6u7l7RTn8f>o8ksPD!IYZh12&9mC0a`V_(Ni1C!?BiOMv>
zFGVU-tofM@a)Ea{oHoPZ*Z>E6kXd$e%@q@H8oA-XuQCG;c0;q^XbzWv)9M_KuencV
zK*HW^79`ED5|Fm!fTX!dW<bJDaTX-acM^~;%mGRBh0K72ed{bpnv*0TU6cco3PS|P
z#|%i=MbCnyc|-!z#W^5pE{++Hu&18|N%MUKq!;Iaq`5R^K!S%s79`Er5s+S*1ClB@
zI4@>Eg1<u+B+aW4kY1JplIEtE0SVq0S&%gUML>E*4oI3qVg@Amf@DF`JQV@yRXHGO
z&W9l+!!it>DOr$|$roOUfCR5Z_pvQ$^1Y1nWS{rKV<ron=3fSI@Gt8OT-z-u$7kG7
z!^I~HwdRWk)bK^?5cMt#${}6c{4cqw01i%BD$pF?paM9)O;#01e>(rGZYqF#SC$Gi
zFFU9JUUrjJ1=4xX|C*Z$*tsop-mJOwK?QK>i>P2S1?5Oi+hJ`uvOef9^3JZ2ap3$)
zt<@{c2z0fc+&|)8hyJ&ds*6QL$lStR$}fkVxnT?UyEcWxNf2&1xPrpJ&o2cEHw+Lz
zw#ig>x_kLi$b=(aq!6d<bW?UZvd;P2Lp$AT;IKZ)?R3)D&;JeA?D%}eJrK(e_~o$G
z8bawjY-?ef{~%fz{0Ac~EM{KNt$W5$I<anN0s^rFM||~-wJOynpUAE}OGDaBAN7&t
zMy)y$_%OfjJESSNN8&1Qj}*DI$6|S=gRgS<!LEm68RY8CqXBjK0dD2V<V@4n{QaPc
zQbh8u49DU^y-!o;B#6jzCDo<4Z@CP0GpQo7Y>X7IQ^g4=CGX(2Vvfl2ewp=$NO(#_
zmSy8;Eq%KyD9eW_XWa_Qa%v|ENH`M6m=auO9mHjE34|kCRpP5%49(I}`6q!;f8M?5
zEaOy4cLTHxciJGG1GM;N$;#8rd7RKZ$uLe-Q>05bu`KB^4$>d)l3k1FLATSijk*yP
zF#=co)%Izfw4x=ENkCYoVhxFAF|pA;=)ctd%7-{iI6k~;ydQQ5pMjG+uAywki#R}~
zwpp%`s9YH3vU!`-(dQWL0;hVULQ`8VpOVmAVvb}(v&FZ#j=u#8`SpC2icamm+#u1N
zGw$Mah-jf5_mK)v?Ztda0(6NEk*z-AMo?f^gkp>4y>g<+KFq#$?nf#%wJCFp#I|zW
z%5#ZYfx|yiajBh|J0z~v<HnU4r$RdfBo&(4q4~Om_PlXJONmVJP5uNL!MFG-6_nbe
z`HloNetkK{qQG$=snFD3%@4Z>Ei?Lrb}UFLG__}Q|Aas@#GBBL2T6sd_HKSIp}E9B
z$?h~-!QsJ%2ot~L&!DFLim#Huc2d7W!?xPN!3iLkEY!Hm8UJ#XzYBBXAHppAqe^m|
zgFAzW*+~xWetIf6xCbza`>!N3!N5I`@BA%x$IQR=_xdV53G3@Y{LYPpD{BT6TPrGm
zsu?T=Dr=-=?6G<lKrTK+PesQ5jiDS!G7}j)l<zzycE_ACnj;}skK-3Q%9ZkYI!xxO
zWylryWrykbL8LxBo@z$uAH(7S@`#Q>sx7^zNRV3$AiEGUB0j4NT8rA8>yzKVFy;IH
zF>!SSfg-#xH6Hw6X2-fvyJV+GEElC=c}n=zFcJa^?5SaD9Lv)>#-g^XPM26NPQx-c
z{Avu#yf8J6MFx9!Zk}r2XobY`VudCAC_+KZ4?h}&f>_OOpl&X3(Lros^fQjY7!A&F
z+jG#v-p6Ea^n<uUcX2<bZ6{FFULyqW{RF}Lh&aJ}&HUl|NO^u^tX3;G=5MVxHqT$N
z@{DSIYxt+x7@99;Ec0Pcoo}?K&X;3JD#yF8Ln+xbZ<uM3aa_SNjjLc4XQFxkCmXJ^
zHb%8L+Yq{)UkZg9cB$TI6RHM5Me~NuRtR;W8YbNs{)@jF-xe^rx?gSsr<r|`W0-v-
z;HGHadkW3_t%h1*aN}eHb!Iy-(Y)WzFGXraY}+TfJ&l#+EAOUvFi(6HUxhnShE~Le
z<4i~<nl}vC?vRqL=!0(hV%u*fI1|nLBRTZ7XVbi2&p8gu*x)1Zahn=VH1BJb8jUpX
zPg#T}Y2Ke!5?%UUF@mBJ+i2d`S+E#s-q+`bW21S0#s<ek^Zsn-a5|uQ-(Z8J21`Zr
zhFLtVnd=+QrkE)I&0n<-mS3=eGts<ZS<eMd8Q=XcxdHr&4WMS6$7l}Yd_I8Ee(&Fu
z1UN<Wo}zhA(Y(XGq$S}Laf;?WMf0vys~lNxisntbmk0p#2_8N&W`?)e1~zT~7X#a^
z{E`^hj5EXAZ9=uZT?oBH3GHxlh#lK7kA_9poi;eyrY_*zl^YIrX*1y5ZG)q2<O0sl
z&f)ml)@%l(uh}4JySIS!^&F72&Dab`->^Z__G<y@n>iq9d$1XhzGZ`?ZPNnMJvksL
z3Kq6sGa!B221(nY1*GrffTV4^W<dI`4U)Dq3rOF~0ZH3*&4BcM8zgOC7Lb0B1ClB@
zY_?`V`k@VywjT>fKgt0~+homv^kW+&Z66kp?#%&7+g#0nbe|29w*Lx9Kgj_}+g1%B
z8O&w(+aRfRG<IGA2|KU3R)RrP#0UAS_I30_HgJl!fz4Zh!{%)S+!SwPinqc0YTHBJ
zhA`|*THeOx)Ak~y^q;oz@BiMd(frZYXf($FG#WStL>f)ZCU3TkP1-;!(y74zmz&0>
zeV3|7)^48WbO4RP=^&!9sX!}_%0Mg9`NKbeYl(f#I?z^`H0KahCOC&gDw9j|bK`se
zn;YuK*-&dPB|r_Ak`7VtGQPL;RPm2+Q^8R-6=<F;PyswyCaVghgN#4ZO$GmMQ-S6{
z0~Nr5X0ob4`sMg@-Bj>&n+i0)9jE|)I}sI3Ccbyd;BDu7OV=WQeo}QY;$<e^`>Q+#
zhYJt`bM!cy!ZpVvC>)MS5rt>=m%D{II>*5`yIL)d6a(7m&8!AEUEwNlx)NEm(?(Bm
z{aV*Kqd(xYh{o@PD|wACV)qW{;YSh4yOJ}yIIk`z0b%qWszos;VC9uHsR*O@M#?s*
z>g>)atOF(&%78HX9d0FN!svZ6>kqNR;!VEhO2c)^=~yYWO1+)zHn!-4l&x;DMHe~O
z?8IhdjN#S+Rn+CodZF;dC%ecKb>U{Fp&%#K?vX^tqmJ$}M1Rt03p6^KYkn-~w;wP|
zR!%CW_HC4|IiW!MPgA5z50fnEF-Pc57@{sbYf*tre8^bip{Yu2R1bP9Z3_{-)e7hK
zYWnJga6?@fgz1WyIAW*^>nsD;M6yC%II9`ZCYTcH!tTmIH*tJW7jk0;u!$jqx{y>e
zKus7E^tt2fDJFmhI!;YZ7zcEmxF(DM+BIBVW`CYFzM3^b)X$%pX94N!2jl!g+dahm
z;ulKtql^0atF(;m(bUiXG308}Q9plc0J$rvpTA4_{_dcD{vi#^ZlHetDGkeRpnm?Z
z!m{hApS&ebmJ;pg8>~HCBB!O#bk3+Z2Ixo;x$yZlkxu0a+Og%&PGPK3T`>;IL09}&
zrIL78+@c#v7XFx?3bOhZ7V;<he1fdLmGAr|b|-}j{&%L#f^K^uzta(4U2V4GR1;+E
z|5%E!nZ`5rBD#^x*u{D(GIj|I*_Q~qiHyCN?;H@jW6l^YMUg9cGD+6jIc8Q=8{#;5
z1iAV<e^$#Cy5g7V_(AG^c|6sO&_9A1$N(u%ys(2zEUAxXDX7-tTZ{$O+F@mV41e(&
zs@uYFnG%9(W{At5#aH6I`jBvqgS+1Ws`bh@PYaC({*Ld3Cf*>M%G>v8p}6+Can0Jo
zY&wQ<!r>Y!ul^N;Fvc(sfGR)Wc@pTkT>)xazH6s$IAB!{!t*88^SXl7VLh*%$l(B2
zIS-#8;hx<UxY-MQ?Vt_^q{^rG84~IEj7ni`-B@o7l?wuIJS>%RW5wAW4k(qI@u?D}
zOIS4Ph}f5D0(cI83<S~B_{ssOay~v?0v(@8Csb#OO+Y*0!vUglMt-h@*d1Lu8VPXj
zhXYXMqI_1Dfo56_w9`NwfGWr3)e{EPU`Ei+2yp<a+?mgoK-YAoTgtF2Xy=Ax!Q3e(
zpUTA<$Ae%t^G`LK2?u1S9m>Su7kooaa0{U-{0)4i&6qGZ8l)S^d0<&j#XNABg;eM#
zF%PWrosrm`lv!4TpTvCQ(;~FX{bJ75<n+|(?pSO}s3y8w`7J2%LcTI(@FKdA%-|(@
zDl+&|7IG>5Br^C4zVpi19dib0qKn);C}vYLritvmmfv^Go|^G)=eLl(*YlN@J<NFD
zNH=2HL#ghFr<$SqZ)Ks3cIAH41l+y=?%Vi|BV6TV_)dNcaQ9l7gxdu(M6flVR`HeN
z;0Gj<=_yG3*TokFK`g3vm>Q3{Bu{iPtq3frlrP`464gFwsNN8MKU|rkINuni#!<;b
zR&t<H?sYdvRQsl(x;*@T4Am83Y8;h3_ap}@<?(i-M76&{6@DI(s@@iUK8RFxE5CvD
z+m&=Dwe}`UE7+<X3l<XoTX|7&CI7sGf8NDE@8O^K^3VJD=L7unA^!OY|9p&pKF&Yi
z=b!ue=a>BRd;WQZeq!6fw=+^hT&%M>))vA;;fh<5Q^4S0D!8iuoMmg)E*@CEs=u&i
z)rw_>{>3Yot&Po3#gU2q{uLzJKbJpefU$FR2m@Du5k%9<Je@$msMa@a(t@@2i!SkC
z(cvE)#PW(RRNF_cU<b8@rp@KRYi*zJ6DnX?;ykQp2^A~g_Ol#>4Hlc_J|i?~XhNU=
zSdqcwf(5jB%xcd{N7YvlhiUH+y=afB`EldTL3AV5=_7>V!Fnpjw8v0tdmlb~cSmAW
zd#vf6d9Vn_wL`)@qNN=|0#aMrWn;pd=(1q+-VPIBduDq6g9SlQ;z!96ci*6OkR23!
zR*j1n>+vGXrR-vzYyGrVy8p(y(}r~P_K%|*$<lqio{G|a0;Q&v?h{S-vP<`bFsmrt
z!$^R0=?1a=e$IUx`Y>E!Ht<VM+fH2$Ria32zm4K!7smR7K+LbD7+gWjZ_HvY5@N0}
ziFr_p!4<?jWES&QA?Bk7G3~=wtXRBq@xsEuvX!e>EFM@^IAhhC!oaFkD}tJOqO7Uz
z9-ZkTlbX6lVEamdt?WNcPOR-!U5bKO<*v6cH3VXCFimD4MJXaX{vrI0Q8{o0|7u(X
zc6w9%!&})8^C}*%9$;48d+Dm*zncI3KCV$W`KI`XU*(rVp$01I9X6qAou=X+!v9zZ
zbqTY;O*=4q2UOI%_^Y=NYML2^Zt4bbJ8j^Ur!8EVp+>kcN5D<RKMaH0H2Zq!zeQ`x
zt+7U}xu~bdqyMUvW(ybm4;pq5V!q^sYkj=Jukn9<eRy)_OmDPNA1XJS<&rnJ&8v-#
zjBYE{hrF3HclaNpQZ*k66;MF;*ABHmJDj8X-!#+;LnsdRx6W(_ruc{7=9eP1g6g9}
zI_~&~><66*$rS$({@Cu2lC9`|H+``;b0#>=X&bHRr#bYsXU9LhjdL7!Muzx@zp$y%
z6#wv-N{z-SH~(W1nv8$=8zs@D@9iM|;cqQijPVa2&JE|EcSZDVXy(7O!7;@@{C($e
zI*5Px2OA_cSgQDkaQaSb=4PT;e6aizf7L!%{;v(3<~NRJ4!`kS;FN2*|5rDF|7QcJ
zxt@!Dh)0Kf0Ht@j|Mw)osrZLe@eilsA13%eQ}GX{;vY`MKjh{0wOGd>{^759_{f+U
z{?j(Fnc^S*i(e80n{j41oekz$t9^~Ugwb+ue#w2bj7$z~@elX0!BHMjIMfbsX5@xr
zi+{MU4UY1A!a;U`vtQ?MI*5O`zYUUdZ^G$yfOJ3(Nbd0udu))D3lq+(1Ed3UKyr_N
zc#sW}@>Rkyb%1nm4oJ%X$QJ+bF*ZobF$pKs0n%f0Kyr_Nc!&*>azVn$bb$14IUu>m
zKRnb1Nx2;1TslB{Tn<R8;B4^^A8&)CoQrTC9Uwg+2PF6ShflOYQf@>zhYpYq%K^zf
z{^8*^NXlUd=g$Gs5jh~a$3HyM21%`>Rs2KHkKD($q`B-#{8jrp`pGtM%Bx7lKLmp*
z0&Xh)AuPqSM#xB9+*EGPTI8KCt;{cyS8Q{h)u>m?bbr=Zv)q`~8^1Rs0s-*t;mRy|
z?-KP`iZJO$ogwkea<y8Y8IcP4>w7Ng*|1@Quy{v)yJGylqE+8qu2n87Hx_xVMzPi;
z2ilQxquIB*Q66X%E46ZIAbg_N3$?e%LzuUvm7&)8%~qq=qo4Kb>Cd(%+36Q~t43Ru
zdaYRHo9n$(J>KK)_2Bgyxd|C~Q$C`Z;^ldV{yev~(rarg%0Uh1A);>qDK64hVy1Yr
zL2M5Qdw7bQ#unK$rhMKGW3Low41~Lg#-_p^KFY%$p6*temf9+l@<@D&tV|$AMk<p_
z^K%P(_&hh%D{QEh_aaWxL=ym=vP0CnA?)EAHx&%nRG?fXaeyaO02b$DRe=h7c%GXI
zUT9N+^2@}zp-=(HqmxwyD(vB)n+i%c6(}cA99IezfU_D=!J{nf;c!xQ?qLu2;+HcF
zb-qpE8YveF-^?!wh0ie^(pj?!RqF-NY#|gh+ej%+trsx&o?0&`(=#t%CvCiR;a)D8
z+5Sb`2<&U?i*0R1trx)DMq2@MJJMEMCM<8*jq9w~B1~IhN!GXj8aIu-)}}GFUH~f`
z8UrglqOqy<!lQD%Aldo;n+%o7;8MdhcE>#>u=i1!F6Wmbl_}Qz%${?CT;YR{9}dR`
zIG~qgZE>kRC9v!P=WV&+fHI!}2Sk@FIBHJ`TzbHHd*^U`jTxT-3ACFmNNP_BlzBjU
zXAVdj`#l2^NI+ST)SeP(@qqO19FR1cdj=#>lCmJFJtgqp0qLq7kkr-$IPDpbK#<CU
zr1q3Rb_b-Zb3oGA>lu(h*UEyV_LM+t2c-AsfTXe2Ga!LHmIX=eDS^fgNFU4rNfjJ;
z=^2nfWy^x3_LM+a2c!?@fTS_dGa!Ljmjy}fDS?^}NFU7sN#mMlKmyG#3zFJX0tp?E
zuE_yO<CTYy3|s+_4znPsJtYv%0SSb2_pvRxr}PQVlYQO`GGrDwWvByn7~nu1?hIVp
z%}{rr8)~pOvrsET9cact4VrO>sCSv6PO_K%pS!65+|?`<C_^1M%}@b2&68CHk}B;#
z?4|-Rce7NW40Rw-Lj@pEPgWI3Ubg=iHx)2%Idk5u40T{@Lj_=KM^rGG40XwAJFE@%
z^GPJ3-2X>Xb?%0`D!&|d=7#f0{$*3RMi7U>_hRm0MB%|gCGDUT<^0y41!Wpb&8q$y
zNgPn7o1vI_QO2WEOa#EaZBW1r2dfZQfHETj#V~9hV3jGU_&xk6WP*;J)O1?*P`4(l
zMMe_0{lhtb@v*#Ur`WtkoRPgRzZ}+R1AFL5o1Ve(M%m#CdVUhW6zJJdcAib9vZl@C
zM<Ekr>qkl1Nm{l)pYxZu?2hJ_!?H8<M#tIotWmw8XHdP}_3YRiEwn+==;D9^x_H;2
z9B+f7QO*Gcl=H4bIl%@6v~)BsTmi~S{L;9f#Kya{&hI+Ga(+665+Snlm1N&n!%ZJ0
z&b@R8hsT)F?{h2{5h>#ldTRKl_R}09a*PgF0_y#<e?a*GuOC+dcV1-BewfusZr{?6
z=#jx@8+V7nX8k5z28s>78-(LhD=<*Wf%VHAS%C)#QceJwBQA6rRNQcs)r}F!yYeV2
zm*Zxr)O1Zwf`hF3sRqSDwJ3W1q&mpz=19pp)tMmD(Q0lX=7X$WAhZ4u)xMO2tWL|<
zMt0jVR_iHK-8#m~<*+-h)d@zK>>;?MIyl72<@6up2fKKLm1}_~GTj&Wt;#-44evg9
z&)1ML;>RNd^-d!-b068xuZr#-Uv+Dz?Jl0fouXxFG-Qkdx0xj?j<G3%no+vOA%^rj
zQlv|QaF+B~ys|$qXFVDyY&%A3Un6*-)Sl_sTNuM>X<?~jqj*qLH2xR9=@K4_8;$9k
z4gt175=*VRYPSvyFaTOL4kb+7!BsB7dQ;V8i+g?DtyD)4PxyMiN>!KIx&s{y>T(In
zoC?e#$hAftOU0)4@W2ZLHkX6ZQn6)6wAScjsj$>O9=KtEb>?_#7<u1Y!&ieomI_Jj
z<AECnNG`|sq-rWNR<*_;OGTx&@<0y*s%7I*R7zlKjXsu&NA2T*8#aC|IYyxdgDe%5
z+R6hxZ2YJ)Lr!ZXvQ$)RHxCT4ZlW^8m)3Y>si@S39!O$9wR}7^iij(t)ncn$s1--b
zv3ZiO5y_I6c8ZyVT2F!{#+dxO%wv70>q+jpVJ`s13zBy#$?+JJGicqOj8eFto{A`i
z2T-d2SCXEHKzShF`CIIcIYObo7h7ZzmQeC_vL-OG6`isOf{%lYeLXE>d-Ney-!$ZE
z(j8KDj{#&CqC*>J#%i+`tO^wUcT&E;I}WM(UK*C&a7fh;(y;7?L#lqHu*9N1&ksA8
zAnNmK*)1&KuH;7Q4YRt1*t+IteArTST@wT!efUmFM?Z)w)S2IN+IH@NV=+wk6OlQO
zj7R2N*`>&wJY7nS%Q;t!Lz!_o0|D7lIbk?Oc5wwWF|LC3muXY(G&U+_?GKuwaxUhV
zLZOD8ouxLRYUoo@Ibli_LhajDA{ZkTp|ZC6XYg0!n<ys2!<ja4%512ja>5!L0XG$u
zb5DuNxx!E@49>i4JkD$frl_2&_@zj#aPTemUwJpZgZ1^R_$u7DGqfT$>SjVRMdgHX
z*&R~SCG31ReX(&i6Pzh3=YQnT*Pb1f^K{N}SjL8Z^+B5&O;I_Alp2k(mo{32CZlq0
zQW9PI-iSH)5la}%VG9;xRL)9nIJT&qz739Mz{VK5xpO!jMCGj7AgRGpMdgGoJFS_U
zgAl|A%Nl>xK3I<0z-i`i=nLlXT;P<&+;6x6Y}o*67IgrGMLi!tX>|9uBms&Y^>zQ8
zvDaQb1H)x+q*AMl6sz7?t<v)98@-~8OWo^j9j*)wdqc&VH(2Hy95$nuE{J|rscq^N
z;Vr8a-I`y7RgGg=FIrt}kinrvF*kw$*z@MKwvCqO&GR_Q2BlLNjoA=3yVnyJg-VI9
zg<z3duQq3|2W62%m>la3kO+K-Ko5l@d!OgYN94t^-nql&8buk5Jm^)LVGgFEa!y6%
zOl&7kMdh4|$~hI4Q&FQ4m9xadN5;(XLfgQmZQ5dB+r}@6fz3EGyx1mG+m(gTOO((K
zCx^DEoG-S)(e`8k=Oww}*rIa2)CNZ}C{<L>OFM_tK~&C{*&u1#u+Y%Ub3oFzR>O*E
zSSi2421(m}1*BKzfTV4!W<YwC4U)F=3P`Wc0Z9?@Y*9I1V}qpayaLi?IUs3!su@~(
ztqqd4?FvY*%K=H-R?UFawn5T%UIFPpb3jrBhdtE{NZW0Yv~5>FdVLN^+J<Tdq&L_g
zX?w1K^u`>JwB6JUNN=)1(l%TH>A!M7(l%2=NCtD+4jUx3j#g1QvEQ0&B^Vq9bp?Ob
zzK(v24V;RygDqR=3tP4ka8prsune7wvZJPw;3n-UQ8}-6Yb)=!wH3_^0Br?c0Fky5
zGsT+?VtYVT&QH2&?9(=lsVF-tDknS)A{v{Dvh%2nvU5bZo9=(!tulSlR+$vp|0yym
zCp<wSmC2>~xsm;Ec0+xu4YlS!f=&beBORjNWn}-Evf%D=Q^8J~3KZF2MdgIs%4Ai6
zipqJ9n+m>TQ-LD;tEimtmYJ+7P*FL5<fejqZ7NV?e-)JzPC5}4OeV5_%HVA$`%6C}
z|AD0HV&um(jy()@7xByCy2C)P{i#jinimp!9(W-|6rR~%8X5Mn3`_ufbGCzb>;+rP
zwSd04^@pS-I9TB-aIg|Nw9_|FcKn#mI+X0Jy!_4Q^Z9%e!rxpQk-RJUn=Yg;E;n73
zlYqeaYN|sqcVFd+HK_=k=R}G&sOZWK&E&Kg5I8U6Hex1lzDZ_1phs!$Gu*P;E@zS(
zi?&c`R_m>1p%8aekGR%cL-GYmQ2^ToG3=hTgcu%$?_p;pwC}6<ihD-NDd*n5niH<1
ztFt~UoG3pZ`b1qpx#|{!^BJyPL}L3<b`n7QCw5qg^-DtIF{c=)``Nj_*hLPfbD1Zy
z-CYvrIF!*{7U#N78=ujeTytAN#Xiq0S^1=xnzm88=5qq+>r<plhmtJmF>mKiSe(yt
zt-~ftWvt=Q48_oDd~L%J4abE*hJMQQ(Fx&kt{#sXCXUGILN85M)I>5l*NjVH6HMh?
zIWBM$$K`Y(lw}m>gpfI1Xs_v*CXCH_#`ub`JW6hGt$HIqpV9Umc}^4Wo`boSwyB6d
zYkYc|Fh1w_U`-gIvs<vb%;<E9ACOVmCx+6w!94JHfzmmsBtN<+ou#yl?a`FZjfPxJ
zI!foT0pzZvbowdZ-yM|BY8sZ^K<TWdVc89o&QXPB*HJpR3yXwy#1Y2sCF7Y)oKbHK
z)azBb*7-GYNd+ZUU-9gu`2>a5Ql1((4+)FAhSEs9CvMS=B)NV}PX)Pt3k&%ZeLg|1
z-^zFX61$T^uKznTVL`5!hudVWD`VC+wXKfBYml*R{AFPujc4pddKUKO?W&9QRAlTD
z7P9Z1hK#+K?;H@jW6l_r1#%@1pmENX+WN);Gsx9T`LkNC&=bE*$N%1E(UUKar<xJ^
zM^H8dNO`b{9b_;gRjX(mT!LCXt|eGds~r~8IOT+=$_DeFaxtFCnS<_T@m0k9M<oN>
z-#cLYTlwZ`p_#yc?vFxqZjejm&5QF;kjrzs&~{xH=r}<|Wz}C!0AsZA0H<>E#X%{+
zc}_QQ+E(n^$tn&|mHRIaPyy5zbOWlx5?(uW#R0AI493|iKs(S4wApKV?I;!ppvqqu
zC$0eKxNJyadvuA|gu_`JKq?PnoWTN+YrCP%%vHa3Vv7S#<yMS?S%7m~CY(^2DN&uY
z16&*sDuV#dY3ZIO9qj@*)x`m)ay`c3E#1SJX(-T+dU3$1{E=~@%S7NbSPr!FUmS2M
zCuJP-0-P=}>{4wAVa?2l4y7FllZA4pm~bk;Wt{xNDE(Ke8B7?=#AypMG2{i`P?Oq%
z9O*akl{QPld}xqvBqz0HJr#4mVHQ%MpTwlL%6CR$cT#3n4So`nj8BUV#{0!QtI6r9
zfnvr|CbwJpEhzCqzA|O-BD#^x;3ax0GWb#!aw+{JGWZI<^UByAa|XjH)PrIQHDj8{
z-fQ`N$Ly(D?sk3)*?T=-Y1zXp_l<NTmOYf}j(DmWs{d9N%4M-~zi9$)UjX-Qe8&;4
zav{WdL4f;eRi))pqYIXYVDCMx;w!g6oB;$Rm!%-_Ul(5p^!rd;J4`J{o1!ld>oKhY
zEU1*H9**S!s_kj0-VlC2T#N(N8^hE%DtUfR4pho{4kzUR)qka-x;*@T4Am83Y8;h3
z(k2Hg<<Ex0Yk=x43RU=d4o~mDE&O~Cbm~^FvshTllVRKt1RJtr!Scah!ApsL{#ngG
zYxrk?f6n2bb^LQ4|GbcY{)2xu@XsLswE1TT|Gb@luI8VQ(obyD_jX2V2yyjCPKSpE
zu8{jbbJ}(uJWL_3jtD|riS99ol=T-W23HXCh*`{0q7Obf9x1E8OOdiLw5Ep2nkS~W
znW3^m3E8o-;H3t=;R;T9xC%D8O|i1JvHwn|v9d1WmqMWi+R`O9p=$V7v9jQ{CWN|B
za(NI4EGPkO=_UNtTd0aqbq{?T4d5=dfm0q@DpnRe-y-0qVr7L9HJTj4qi4~Ya%-$n
z3j=s?jE^1;!@>X21ahR<YL!beQV3q**Z9A_K0G;drZ?KC50#rudT?-?R~s7{-Bzj(
zc{6A3@IOYSYW4>eLDSm%Ylk|Y&7CC=-2N*KwZgyvW<R~mc3_H?^%{ODQY)xFDx}E+
zy*R{B2eGo)e=ZY}DOMKz_}n2SThR_TeX$!{COFO651rAQbLcDL&=)jn;U(O@wCBun
zb+p{@hN{J8Q$(Yn8nmTSDs$+TW~*2$6&odQYjGRJn(>AjWeRQUQEZyUr!VowniP<x
zr&1d(H!AetP&rm5JF1tNj@B>Y9EY8eAy(FvHZ_`JWxZXg(HIfoT^6A+w?GukyOl(j
zzPE!|S?{r6F~-WeDmNTktgQFi;Fw}%UEMjH4q|1!&jv{imMT^jJO<O6xtU!MA1pt>
zU$qaGAF_edoDNZ&;B=S^obpTbKjH@PV>W=A-y#6QZ!sT0>DuUjJPA<rOjEJ4!gyaI
zqE@X{uC=`8aDA*wBYWA~AZ~ByQ=zyxuvn?GabvkrroNaSYHckC5xH7h>l}-#HC$|o
zfMVX(Anwjky|$&?z`UX!+^UqCN+?Yt7ytr?RW2$w*m;;=C?KUte<+GrbIx3kVvh~i
zO9)?9Ez`ik!OG-UN|i==s8!vj<PyYsNla4W^2H4)u+LjNS{|wttJQ729*sAYF@Aoq
zPPyU8Xfzg;ymGxMCrUwTqwH;LlrI<~t<h)4cs^7dYSd|r3spp#2>g1Qgu2WVCm!}7
z9+|9FSdS6GEXLa~o4ze_KYIhdzF{_fU#8F~*(i@T%1tgWhFTnIjZvsKMjT)j$eR@D
ztz47hH*mDJ4PmUc4GFL9saRR0O{%_?jo!BUnAa=|?FMQit%oh57h0DUR@A{(1M?TD
zZBfhcWKdNj#p+dR%~rWs;=&NMg6pzYmiT~}b@TjON=}P|hv%uS2zXS%^afq2wW#<T
z-Ui7w-aujhsCG-yVVjk$sW>`Xu5l}(>@&QuQHeG{?b3*ky&9-4QVXfosY@9f9j&7S
z*j8@Q1dcN^Z(e{8(F4z}RqD0#Nlm?^m{u=E6KuBlpX-B5N?fmE1}+J5#=!{*k_sA#
z`Jno7U9n`&4Bx@SN5;(X6Sjd(ZDivL1KTJ0B{8rWXNK3=gsM&Z5Ak^m5PH25+TrBT
z7AxyBHaKc`9p@JT&S!JOvBk=|!3IZdqN`Y0pX(e>2eGpL+XhK(mgC4GXz24fAi2lN
z`hpFT+7ZV|MS%3h9FW{&Wqrv8N$q{(lp;X-at=r;<dQ8`)>mwh)J``}DFUP$b3k&B
zm35O1lG@b95k-J>a}G$F&wAL&8LTe1*dVE0Y@AXANVn#IqzcX!E9<K^NNU3xM-&0l
zZ8;#h$I80h21)Hr<9s4Ox+4c9_gGnX+90WIXdF%iNO$Fc<Q^;QZW|=Ej#ja<Hn5#A
z*Ge!jI)9D7YF|fx!v;=mR{uc`yzqsOfSZbyg(==ttgJmWR@RT*+RA;lwxXPRabgzw
z7SJmqZ6#)kHygzEfLK|-cGK8zY#LLp6e?C0s2dTDO~uNRt5>;fT^5*nBCArh4D8O%
z)w<A>!?EsZb(*lA&{`FrXA{;Uuhl5ln&fdgLecX3RyWE6jbf!%E)B3jiJ$1D74YV*
z#m1)Q4&el#VrBi=tup<^R+*Hei;9&6icO?4ximkwSXuvYL;WusYULDzv)0gQfSA)E
z>fI14YX*By*ykAg@nczA`N{)F#mWL7XtJt6#mYL^O$Cp&sX)2=s90H`Ax%~ls90G~
za8toyHWet}A{8qO?5Btd9%Zqzj!LR77HTTfQ1>bBS;94}VGH+Yo5D4E7kZv2^GibE
zb4&-i&9n(s>jf~jAQX(PNGVRO7pB$=>~gYwk6JHClAnKUve}y2%6wZ}QR@Ye0MS-J
z0*tg3mkG-oI+}-#kSkMhQ@I(;>y}oAu=PM&fwccXlRClgS>&x6ZP6wL9dX3z-l<-H
zy;kNaAMJFc-PERdl0fJ`(@kTi+BBxt3;%W}RUl~WrJAW^YQ6BNTrWtXr@zclnG7~u
zrbK2=JJp^Nh@hxUXYxyF%}@87yZqki)PvD~m-!LTBal49aoK<i?#o%zVYRsg>Lzfl
z>;zW~4v3f;aNtOt1xIZzfuRXFt2&3{YlO=TNO0@Uf}}Q=z{CWk=jVWQ9Cso(A9NvK
zWk7<Hc@`wKxdgf;Ag#>-N#jpuK!VGA79_Q~1Y#v1ot*=c+M@tlG6NDE@UtMP%_Yz%
z0qNWvkThas1|-k{vLLC=B`_ud=><6;X-vorNFXO<K~kGbphyDJ`8gn|f&=L>0}`kb
zS&-D`64;P{v_1zUjo+973B-*oNNRHl#796X<bb4688aY(Mv?_dZ7zZ12uQ^okTl+6
z2+2Si0I4MllG<DXp%IWkXmlOqOs9XBI8S!N56CuI;FQG<)HHwtHLWvnZ8wYEo7_-?
z9h8MyS?oY#18UINIz+w8EOwF|=fBNO1>i<ysX$rmz?p*zz?qw@Dv(q>|0*{XfEkvh
z0%frSK@TbbL2t6EK=S_l54)*=d2yK&Xl1bj8xSf08!)1R$z-w9=kEbW#~$5Y5*Gby
zld5yK*d4<!hn=}0y8S0@3fBmTP&f#P5rt<O42;q3eFKzF*`R17MnD0Haa>Shx-=83
zUgKX!wj-j_J$*I)TuDo?eyP<yN}PM?4(40V==Yh2DyWx_&{M-VwV&z`vSV8%2($mL
zA5(FGor$Y}<SFuJ53_%w)2H;v@G>-3-XQy*XKRQcqKCB9>|h1)sGVVmf@hKxFg8Io
zgc80Ok-RH|?7Kw6-d52SISE4Sf1K)1EG~i~a89Za`(KU}Y*5Lwt4u(s{{*)WbBO)V
z$*ey_jg<nqX{<Z0Q0?dBY9W2QE5QC2DNo%Bu<vpP2i57sp1bTGxSV|V@cJ&H$xYwu
zVsL%O(oST#yP@@SE26uB^>6I7$r*cTjpK%1`ew6aMPM_<Trf)42x^diYl?KqKFg9G
zi?Ps2VEsjI9R<-Oh(^^c^0q}-jbaUkrbe;RJLsE~9T?mK^i3||#i_xVKIsq#5d4E{
zs0qgvAK+2;Uyuucr$4_6TqE{@g^&tLS%E<@04SHhVrfdUh3C7Dy-i`7y`Ha9aVb+U
zcm}}b5<M;rmqRQ*jgpWGO__wjHvlx3;~LVSWe48VSP7}9lvNm%1E6~DIBFO9d<~Tg
z0$V&pyfQz6l#q%@8HK?(03w&@;%R!y45_El5>i1avoJUZ0JU@+YD$Tpr%@78;V6?Z
z_y*(Fkz=qtkP=csDWfnL2jd2n85vKbC8UB<W?^s+x(Lb;22Ud<q=Hh0VXzJW>dbM}
zrq;xS&%ez)ymujd{vArP3^<0+Su<#f9*d8L*v9+msR;je0Ed<OD-8#U@NWn5oxjEI
zq{LL)i;q!76sm*xof~No&YH8t)^EyR1ODAyivC?`8QY`7=kGM+YSM+z|GEKW7lPQu
z=VUvnkZJv@Z2s`lMbUpV<@>uMeEvOYSaw7B{O_b;*$v_Izo)Rog44|pJD4Ci-D=q_
z$dgR7x`o(M=4TuS-ViSCZ|qY5BO0zyXa2!y+qq7e!o?jPkMp*oOL5+4qLdotZH^d)
zGNZhS_p@WX!9<G8;tKX)Tm>sGQ;fIwumLG+kEPD19YQFsAoMDJDHLkh5q+OcsM--#
zG2UP`6+&I8<jLq%AK<UrcSJvA1E)-cD#jZOuMu!lG2Zr+7;hgl)Cz+~7#oT++kq*@
z+qL{sq*lcCeUdv}9maTL18ycHQ;atllHDOCThZs;^u?CgOmLdD8U4nWa_DQ%j`8+h
z&T+WiXdnpOWK*Ll#@o$GjmAhGU$qEL#(29;Np$IZBc7Iz+!e{7p<Uf>!D5W@c1LbF
zwis`B+TfUCyxr9~oDO2V-ED)U21^y=4aV!VW^SgmQs<<$l>D#pSM7u4H*DZE`!||7
z?BBV-DHFN>EjNJQwgJ>k=l}>4dOm>C%I<$R32-XL8%KKL_?jFhEj1*RxIGo)Z7RlF
zB*0h?N9f}JHf-21x{ZSFsTgwMM$(dS&N+vJ%AGnJl$+VT6pRZ|$4-sJI-466C$Htj
z;*Bi|4vV;AiHaJABcpJ4)#8wf<rf6J!$-XeqNs5sFd4E>1tnWuEf*=IUWr5Ijg&_Q
z=~GiN-l`NUkOMge5flHvkMVX34<8vb!|&S$HdBnZAMi_JU^C7Pe{2(~?ZaZUyjKbB
zaB^si@phjLj<yX8I6uh^#}?!5ej6NZ0~T-|=p0T5G2R}uLDKeK0qLhXAZZ(_VZ}7e
z=pV8{()L^d>1R10X}hTzkbZ81q;0qY(l2sA(za7GApOz?N!xJ+q+jKLq-~~VK>D=}
zlD6LpNdJ=qlD46m0qHk3NZOt&ApJH6Bvo+OPR)Syunm&7;|fT>%K=H-OwEAwdmAKe
zzZH=FkOPvom6`$Rk2Xl!ZYv=DDF>uqN*|P8@l^=PV08Yo4U$?%s~B(CZOye33}U?f
zg}-WFM?Yc%r(!nzf%6G{VZ$~8ZYpL2Ch$`+-u9@N4RVj)pK-5KTiK5v%i5CCd;ri^
z-~$k8D=|~N*&wzD#CUt0o5r4C)0m3cpklnizaXNqshACq%9sr*#@o?umFXB;Wm1e$
z72^$lAd$-C()`>Qqfd21J==y_^BzH`0q>CxQSUOw=*-YJ%0Jdk1@mnxP>fL(;|)$L
zlT`&O#@mT*DmdAu0>v0rG2Y-SGg(!jV!WN|rh@0#RG=86D#jaJbRsI4OpMW#!Q0Ll
zm0m>tlBDWlEXT|kZ$IRzd${f}&@`9X6t4Loq33}QQbggI{iTsH8q2^0&@|^dxW-;o
z9v$A+2&kIhV?DsZ3Ri)HmB^!=s`(7(&*>~fNiWGu*ZliR0zM&h&3{HD?@GGn>A96T
z2`HPtpn4Q@`BnZ{lZvvr7rP#JQf&gt=C8P&m?@iolvxjGNSYH3x32b*{Dr%_Xq$hb
ze07Vq`AqlhC-x0xSHYcA2h`2e#_`QA(l<LO`9#*cOW_=sLb^-ioZe|OH2RinZY&ta
z_AyIVJ}Oz>$*~?x^Fe|1eN&`M2a_!6G2iD-Xq+xYele;Wd6lsSM01wd=pOW0+7=@E
zEEkqy#%Cvl&bh|DQ5z|ybg(SNWG~iQ1rE8+hZxgLDCSlU>SX7X&gYL4*hJDg2gZqQ
zf~lRW#tCoY=$$U4)2tRSF%(Z1N^=&f38Q(s5c9HtO#sz%yt<k&y61R-O&H~~d%(I(
z`&=>3ZeoI{pZlAq1YMwh_9)4ZF6!q&X&Kw2sh^KA<Z9AUKMyg0+?CYNLsP!LJE)(J
zPs6essGm<v!?GKwpNA_fyPW#Dlfuvm;v%lF4q4#d@mK4cHYH~CS5kh?I>>*!_`~!2
z{xPxA+b)9J9eEy1qNW8yU?#ovK7s`>E}<-fIZl?;+X=<`9c`DWN`vH<Y+lz13NEea
zki*A{94?Hn?{?CMh=wh9-`lg&Vf7Wnq0KvlMfga1HL#V%$Db$Bjb!M7qxDpT9(XdP
zw)f#95OySj4;*8<XAVCQX7Z3QlZZs%k%06_1VJI|E8O<n3=Pn!>W%2q5M{pBy#jV1
zI=1RW3BFTg?wuerw+}c&tx&I_j0#I+wBq{Go4EaWbL9p6@l`TkRMl{i4504-(3NkV
z7S=WY&D1~yv_Rd`3Ao*tp%deB9H7^AMdwEONyWcQDE?hpiubRiEXJ#p^ek>4n~_J;
z${txfw<}pRiI3zkXnUfIc(=&syG1@9Cl4x-HI2+ys;U`sEFG8I3o>$?VlP0Z&+AI2
zQ@)Uv@Ars&zbALTrFV4u=^6P>8X}P8v%8Yzr0}%dUL|sSRqotM$Lsd&jNHbJ8^~z~
zMlff=kzD$42u2U|UXjQ5iahQ!NQdiKs;+e1ZqLj3cEUD-e05+z>qK@WYq4f}waDDn
zxeHMGgtwoPfhS^aK`uL3`MXI?%isG%{@y3@ca-dg68j&v?npYCXO?NghSS9|O~{Vq
zIZ|EUFLL~T_v$Jg)-$t~umW{4YYEwrti`J92SnyRFz(`%9`TNO4h&2alINuHDwgLD
ziadXC+<BHR_w6|ujY_j!A;%qTg}6&jE8`D|{C-H}_h8v%`G~W$j&~<bT50JaJ@_+o
zSSXCvo2^2nR%sOq!2)C^Pw(TK_2~RRBEWnyLX9`=y9HbKk$GEyNV@H3&P~Nys6aeb
z85}jh_3sK-d|~udx{+KMJxx!=!e}<7230DtFq&h!XI>bs`vR?DR|Y53dBf#K*=r8h
z$EqbJZf+VY(&@sjGVx(sbxx?%ywPH%(d_lMR$9YEa4mZq>7OkxINOe%A!;qsc_k#1
zne@yLH_N1sl1x@li`hwdqpxRau~qcS7ZyiGtK~&r57C4DCXs_*B)9-kofaykS&O`d
zz2c6LLkY9!s#4onKW<iV2O~4xIDZyj6^<*E%FUr!>w~-D*?jzfBn*qM34*fc&1%%E
zWtPl23#7_>gs4@nwfOdMWtNn1i4f0^r3jNNj0}lqmaEnJ%!pLTU*B^H>!oLhn45oE
zRO<UhjJ>&315#JEsTDiND!<)<Sv{j0k(L9|Tw~3mZ9U_9N21ZR9}mB7?Tp3vbKH#D
zT33|PqO6aZ8&_y`co1c?a>g_iPW!2A%NLAM_fS56V2tYb`O*cwSDg4=skPRxryeOa
zMq;px47wxf=byQ_2sB&>hk~Fln#UDIQ#!29y>ug{Rz!_FHlAwE_PQSGQ<mYRbT9lI
z@%2YW$%<2JHK`Y&-Xx^dH~V^eR#P`KQl`E1X3w$mzaIH&>Otse2jSwAdYB+MXXIOb
zJ@SZ7Z;@v>9vvbE!o$+Jn-iT&O8_*0T~lt2HEI%Rk1Rci64~bdlFnbJhyj#6(Lh)3
zH%-)nK2Zz&zFbXI`r&8V3@{dsHtIvvu@}%|2gN<#P~1-TdmcDfE(o%4hR8zt9#v`&
zBOUTH4FKA#m}*C>PJR|VZSjvw$$@`7-AGp06T;MT)B?5g%pfO)pfFBkscG#kH8;|?
zK64^aER_n0$uRPAQd(Y42~mdzbTqzagsJho6ob5M3G%WiS6-x(d<S`nOrZT|rsd_-
z5Opjs&kj@LdATsi%YO!Wc}}jpNN@Mdg#h)4&9O$ACJ==xIrRkf>ok>*5HO!+<DVV^
z7FP52z{=ugn$VB{%xvm|etD?nRhkqDhoyP-joi<bX*;3W=bhawZyc+7Wb*KaX)aX7
zBuVuU;WQ%bDqbIGfefUuQ!jDVLf|4cHs-&ZcO0(fpAYfRHT?4_{`n05e4c;4!auk0
z&mH{pHU7DWf4<K@(|<xgkKv!g_{Za)x%{(`e!}kvlkIKK!9@LEchX%1jl>md_%x0b
zww>V9P$&Y3VFrGpn5-<va-e^#>_1HB-n3VDDeUG&*p5jLz1b@aYMG%om5l80n{aza
zrf~&7e_RD#a;ET`KVWCztGG-bU{dFMxn}uS^S|H6n!3q1h2Q)kzZ42JFrn|Y36<+8
zUxnX<7rYSa66J&E2~?1gh+;qCuiiqaX}S{9B@Eymuz^$V-YWbi9O)zArowN^DHP3_
zdU_TGGm%9-n0auogA49Wl9M95!msgveIjnr%$Z&=Ne~f_YGWg#+e-BzZ|2M${>P|P
z%@mX1__qDEQ^(KdkiR^J`ad(&3PVs4cKXk32d40wzvP!9wSwxSLOSm7o9v9A3CR?G
z6QlrlNXb_8XE%MZM|~zZjX8kM=&w2SwP%Ol{1N9k?2HWIH~(%^qbdC6Ka?7cL6-L7
zF*Bo2Pln%|#&4w6nk4_NgYcWvEm(}<H}}pB#}<Bb9~&G~_{|xe!|5RW=Ds#aYOqw{
zH>Y#g=3ZdQX1+gv)jnAE*uZJL1vGQ;7IJ}8)CT_`H-L|^0o14u00`<sK7f)P;UAI&
zI2C?#D*R@7&RjAbQ~1bIpSO0jJX9%GtJ``#G5`;W)tQXf=uL&+oC?3Wfz5myf`GIl
zV6CzeA`)n;v5lXR(`m7nBepOQ=GKdZ9i?@CvsEr4I4Di0w^T~u9!L;9RMtKfEmTTA
z);o8Y428ToA}yqBMF1c&O{B)tCF8;dZuuLe{c%GtYPS>}#x$J@ze(#D_CoXjX_*-w
zY8%+3!<=xP!@%}9en|{$#+l(0Z9+Bf9E2XGgt{!uBa=g0_|3y@aHQv*k5do<=ZM^J
zY~eSLw84>1az4&K1e~Kfhtom$%_rF)NiQ}ZM;`*x(K#Twhu?g%4U%*)^Ks-MARUtf
zl6&|~&jv|)f%!P^5Rhi(fTW^0;xxl>VP~)>%(6j}K3qNyJOreF&jHCj{N__^kff`X
zkK+yj>8Uv&xrg6;nhlclit=&fAt24p0ZA2{E&S#j8zkw?<m13YK$@Eal6&~gc{WJW
z*~rInhk*3-9FW|@Z}!?CNnaozhaCdau{j{Qhu`e8K~n2z6@C+pw_Gd1ApGV6{;GW)
zz0d|u`tkTG{3b|q5pYxCH!+`_3ctxaHP^x<&W4@#6>`O}$m*Wb8~Qah15XCcErnL9
z+h_$bvdD}1#d)WC%zL2~068qttMVph{OmcmUOu9k;#K&~XS=nP(`;=;x(4|;f)jlU
zc$ks45;MgoEtt^T8+N1KSY>G%dz=isx57<h{WgtBw;W%E-+YW_LYxY}DOay@*=q2h
z5t~-s+qqg7dJ3g_7IC8^2cF7B<%Uc9)|#IRzj=;ZWm;#eOw#SgSK&9o#qF^9xrN{S
z4>!~sY^bHTkB`G!(P@Ce+ac=R5Pox`n+k?)Dv&Nhz6!qy;_zfufeOD_b5p^nO$E}k
z$XDSv!B?KFDp27!x45a`LYoStLz1t;Z-OQrQNg1u{N}|;)y1M$W`^H<Jii>S^bO%R
zUu;vj#sf#s^Adh3qVUY=iZT4A$(iS+HYnhJ!$SyHfbud0WsYy~mwA;<sG5$wnjeKw
z@U7kRH>sy%l9}ya&iT_8dj{JN6Nt0+-+g6eeT&rCTlp$YV`g8Hez>gpZ{s(^3TlA!
zZW|nBWqpr?b5(9Q@3p{zBVdMig0HNs?~`!e-#Hxr1N;s`pncE+32up5kd&46!xGX*
zazOg1g!C~Bq-*$o79?e5{Ww30R(CC5bykP`Pw+d?(kJ;UqaML!GYgWkvVK}Zx-JJK
zWvsj20tpVJS&)>K^|KPv4LKlvj^BZn{@Vfx?yOmml$G@h64DoQK>Ct|^koaAukigW
zNXp83BR`6|bQ527SeI16-E4sbSKlm1%F24Hg!I)MkZ$95P;j?fAi?oC3zD+3-YFs7
zl>^e<64Fi!B)CCmK~h%MuS-bZ$N}k_64JM#kPNc{IAv!+QdZV)OGqF;y4R(ox6*f6
zls51hz`<EO3!JjD{y>5Q12O_`%F6oxxs^3J+1dlFtdeBu|I)3kfK!pxR+P<fZ>Cyc
zPyj<O+E!MPDY~Z<dR@;VMX(E;#plfnO<?op5it%{xPgE^Cr{j6FKu#3cXH*e!UR5c
zIVX{!$qX{n$aehfK=>qk+1Lno2~YQlwHc;V#fi2VN;;r_5Es3@;{`V-t1>B@;bUcG
z0{<^knPO|;l$^WS44>eJ8f2m@)XHXfxI}%NW;S<0y~}KdCo;Gj`RB=QDgc!#O9jei
zI8&+sRKX4lZkMS*G7kOOZYltgEK3E-W;jo(01U&)ssc$=^cT3PfN6A@3j}2|JYK2*
z<iv;yCX>xDrT4HO*(2GF{wYb-#cYOVI?`)|bKlE(3=Wsl2Akm`o5D5zA_iaZ7b6Of
z^*s6(r_pBUl!5E0`ydA0(<yo-{kf6``2H{}<@i7SSJR_&FWte+%NhMXQ&R;A^L3Oy
zd{g^Lv0z~vtHtJUfy30-T9tY&h(`Ie`$<!vXW}ZLX9l?h6LUq!Oib-d^9w;JmUI^8
zoEQr;>-}s7Ch`FLDy~p$xN<MHnEL1VeN+r46+c4dLMmRErQ*K9YP~*E(6SRKet}dx
zu40N0PD>J8+V&ArK&^1W<}IVeq0Qw|fn)UrqK*$mX#qIod!oe6aQfs^N$CM7<x$%2
zUqMx?9Y4)M3J5+!PSLG;P0x1Lcf?^hqV9eXb=N<4fa)bm5?839&ty^|uLaic2!n#O
z_i2`^8+Y`_$o`M#=T4=cSn9Jm^*MU#X`H%PPkrL4lzO_J`gDGF3H=x%gY@yerRIA_
zv$SRW02^qc5qgf{J7+}i#3=A*Jw|Y&&yCP*sZY3>RCHM|D-6Pa2(7gbEsQqGqeUXJ
zQ#-E}#6c_2;_9%D_|N6f#kJ_4LU-djga($=qcI&m4^Z{g6`Z;<OpWK@g(3$V1bMxm
zpXB*bW-ew?9;(HmazV5oWQAi4sVWurUh8G9R<UG}tB_+3cKM+A{>BUq28{C;jJuyR
zHuv;S@5IbbFI8A}4~WhLC5|iffVdJp;2hsD8ZKllX3fCR!9XakAoK)&DcP;2ZRJJ#
zkZ@qWnpUszzSTd;21V-^00sR*5=z<(fosXSL#Y4euA|o*u9QmUT7gb-X_d*dilU1Q
z4fhRDJcxRu*K3rUmC{(Ts;3vLmBC6PF&tab@btUY;x>v0JX&nFBJNuv=uiBc(SJqU
zx`v5X5(t*wU$IA}I}-(^Dhw5ehACPW2eYK~Gm6zFIgi!HTBBnvH3y*VMjzM_X(awi
zE(vYrX(+H&wgN-Pqq-mz7`mX&3aq_el$M;o7<F8!asHaa#Zk(;&@FLedO7;W`d$iG
zMp5B5m$wzR6su$9W}#ZHZE6iK;=3a2zU-g;^W3ukLR<FHLm~UPLiW*Db(VdaYeph1
z6q4Y^JGHb;bdJlM=xiEoEmXa8Y=U_aW5w`|E7LsZi|IrD9M!FdL%w0K!hk<79Du@5
zEf}tQDb+AsV|Y8(d=EqUeDl2pl<Xg8zPFI?9dEveDZmN*fN=_NV)RaY0DZg|z%CR6
z=t=ybbpSm^P6J95WUoZsPQ2L!^TxNxqB)txI?V~<vjm<G#8Zx>8BIKeX9prBp0P_r
z#x9Li?95rg%y3qK%9b%H_<+pTGgP)>lY&=>ProMP)4`-*-MA+OrlyfTCpaN<PLL)m
zdrE+Ihb{<L=z?%1x}Z6}p<BP0b(l3JFpa}#tey31>)L7KupHx}L*yCNda)(O=JnK}
zI}WLlK{VbDA_K7)Nu%=RaGE~w8Xp(@&_+R<SIGRB$L#Q}JOCr(z5EJ~>+Im1KJuF%
z4F*El&-j;fO|Z`c-fXKd=xk66afMolZl|*f(>~KQ58x4Aj{LztnuYSnV7Wx`NzLPa
zY_@Z?TQ1*k%O(0A<PulNC3>OGav2})Ka>PF(ZhYv5A}_;D%GZ1x(6$K>W2!oay@0l
z{|dv!h>xrE5x+eH%hX^rx>>$9OLg1ngMNDk#m#JNi!c4#d(-qTo+Ltz#a6LA`1czo
z1w8Oa)8pe+a3)H0B9=8?87&x3amks_)0y}fnueICI8nsm)JM-rn9kFXe&ZBGrN(pp
zOfgi7^}8|1W84Dql*lNJI+8I^|5ldx)0AFf19h-^FWCp`^Sd!nn<{nsSbef>tWFb`
zJzS&FpbNnjx)5At4A(cYCbNcXG%g6m6@;Rhb=EJnpCx)<G`(Jr8e6Z2#T11bh%J6k
z6>KFrbmHH!9``#8**>m(&z3XOL<G&O!<;n@)_{~;scDBXo4uYKkWWn;e#MQLLVoS0
zkl)x8f>wp1#ubVh?W;3|#D}2YCBaSf5R`3NuU3mA#gqYPe+s221|VFe4?xB~`cTyu
zrH?-7qx~mv_Zzk6Vp;h_N;Qlwhq2tl&G(Mrdq<k@!A5)(KcFuiv8a4f^iI6*euC(`
z*NLU$(fpuo>F8_IGz{ZU<~Mio>TtclzBQxgkM0+=rOjAWjEqsIPu31k<tsMoTqM4I
zamKg1W-~U{<Mc5n+h&|9EPKR3D?yKeEA$w+${2BaSc_RBj;UWp)9S1nYX9EQFYBh&
z_}G(J)ddSixuBzn*cILarI0#NDy>+Ut*F;FC0>Xb4kPlTYtwL9DnDbfwnl~tlNr{R
z?bSb;b!{IikFgbr*0xd3XxkkY$;2$dp6-^*V{N%a`$lcX6>^D|-dQf=<KKcLxQQPB
zj?c8Xnm1X4rPP^>7F{>4cmQ1Yll}MFYu>U8N6U@M2rc7^)p=fpPt?FaxHWK;>ND6!
zU~xE;qeZnhfX&>#o`Ev4(o4ZnA4og()P;Qu=>JlA%fbZ<7am7H^cZe=bccdlbNuUC
zd8;?bJJxj6$DEZbmam!X4UV<w@Q?CHlY(PY*z8KnE7i-*+U%CMt=#gq)*G8eaBL2v
zRbsGBLf7o6Q(UYD1;;AZypv8?)HB>_jW!p}pTCJV2FC{bC}7t7`e?aUtjrgxnNJ&N
zgY!$rpM27Sg$p)5W5I$23r{}r<Z`)qe7X3H6HY#{bi&4yhc*r^EDxPXADTb9jY7uN
z!guSV&H4Ja_UN{MU$Jn($vx+mXIHD<*l3CNiwF+I=hE~jk3-?S2MP&QP#00?_@yQt
zmvUjbL>`mpmU+gAxL<k!#7k&nqdwxT9<Efqfl9SX$|`!aZPjRTjIq%=ZDaH~s)aU&
z>4XKD=`rg1C{A8;{(_T^-AGVM$I2d+hN7OHC8YFCv^myuthYQwL_0@x>dR1ZtXZrg
z)>y4xJ2w1hScCeKZ*X_6#N*AOM!nfoY5daKvC+|bgEH6{th5?M3c=PaUqJgcO>a)E
z-tzo0K4)gE+Nw|uZgR~%Zs93&dr@zIi=wM<EjCIJ&${P-#N<*D^5P^?!Kg^t1C<0t
zzH!`1z25N)d#T3tcKnG@3Ep6(NRF|TsZl;Xhd;w-<pc=HIK}ORvMEBg)oWGCEER~T
z5G&R$qh=^$%8ejLRhEXB^DHUMA97zhl`}9hHZ<&QL?L^F+h`ZFG&aPQq*2~Pk)f#>
zHAkreQ3?wR4T1Dsavh*>iRCJB-2qaH7AR3rK}i%gGcVpNm8tk;jUS@(WvMiJnj<8J
zfVBullZDOmYOHSd4_;Cvqf))bm1zL2hq7Pu=9M$LrapR7)0jG#=z_5dHN-}Jkn05X
zEj`O;k9b^3j$5$csbNt5BAww8RTc6Rc00qxEhK;(4%q#;gsB15v(gypfX*hN*MdgO
zLHR1xa$_!QO<)gOLlHeFcX=qk!7|736I~|ttcwW>zq5R!x2-<rk#hsxt)tPiR;ZKn
zV!ay*naZ&jj|W+0aVTO9X-^hXs#8tsUTZk6i6u1W^RHSs)wqGr($Wf|3oY=M_|Ph$
zlUaS<^4X**ewgYpN6$nL=8Y6<RLA1_>!JEhjcemZIz_71O0=~(CoJgoPM{114Xf8X
zg%VDoglC*Y?U;uMF2*LAAFHY`3ChL_ZPvGXXc$0=QY#^E7kYwnSS@c93KF{1nm5(F
zr2!l|355=)YHdX!aTN^8m(S+nfiOIwYRTD47kfb>s;d~>!pFku(+v8mQlmafb3RlK
z@nqP>=@XmEwKBJ@ZDF&-K!z^_PtiGA)U(UVCYxTR*`r==R_j~CQ4lYOkM()02p6}y
zAT!~&pRW1`g3+v2HVwD9hk+p0W6g`A@YM5>jHG5+7BF?sIL3%dfj8CptQ_vu&;!h=
z^p*R1y){KTP`f0i$@BYBiNrrddx|SJP%8Bev2HX}QnYBo=>wmx6RcQbOTz;T8@HBd
z-@j2S)4`sY8Lp5Mq`S{btZm!Vp0AmZZ@5-7$hN1Q6Mv+=&*{|3YiWD08f)oEE1G3J
zX<+4QE$QH9ePe5+cwwPARIKXHAHcy%OXb1^gW6m2#&Ft=k(3*o(?o4f6ScXPB5F&T
zs4Z!twxo#~OcOPjCTcJ(gY`5~^)yj+UDUo}0MoL)znGR3E=Wn=tfe0kK3{0m$7-c=
zNxQ#aFdk^hhb~#X<jiG-vzEP}uy)z=S1ju<XjQbGeB!LQ4F79f)4xtXF^knVIQ5%)
z>bE%c9zFHjocbL-^}C$<Jw5gNocaSj^@p7LBR%!UoO*AV8n<CB5w@70W!Pe3js5V#
zAT@UOf(%Z)D%A`1La9z3K4|%$lP&)~{;FoxdI8;y?Yy+7FI^SaV|&KZa<x^AC;FG^
zFZ%67%Io!%c2CfhONsia?&uX%-O*~O|0Z3&e|aM1%0$XL6DcU9`@`b!@1fP>oPHlw
z;(;(VUP!+Zh4e4iLOKM6gh_6pw7F2)5){q@ghi}dA50cbdp~+~wAIji+5JjeOv%$r
zdfS^$--#FQzV(e8n+3i95vf|s^xu-uQ&MwiAP<E)3N?oY@-saZ4dmxxYD}Li_YTf{
zdg3%!jUEAwiZQN0lOdY5RoQ)ZXmg<2y&nu!`wNq5v5eF;;rA!<oEQoHwTWlWNu+#H
zPoZX|cVN?(Y8@6mqYtus(`ra7!f8w63I3q|Eq`+&rIkpzB$4u}M9SCo6xDt&(AuK9
zlNcyv`_;!iLu+-ZbHC!^{B>B5!a9e}{(tmTboRf|Q_-@2tEZwaKdh&sF8@wXMP2^A
zo{E0t4|*y(`#*-M@p^w`Q19<^uJ;Gwkl$9lK*8DSK~;agtm=QF&xB6lq8l5&N%R2X
z;5M@Hv6J}92ZHZD==|LSBke8t>^k|`|5cxrJ4h4Ol>9^KyRmM~e?lVV@I=Z{iIgWN
zQj$%ntTzFFIFXWUZneZS7bH@WO>%4EnTrxB$!7b~#4|5Xq$Hd69f@b|PNZZte|>r>
z8~Fv=@a-R~YdO+Yq)O4J;r@bjDTxWV#G=o@{o{2hPE&|vr{prJcuzu~S0_@?Y5g_q
z8p5s&oz^3IDmtzI(^LP(E${DoD!Q_N=&9(+{;8*;EBlw8imq%g+AfUEozay|3sd7=
z+24XLW!ZGsA^dR>qf8-Ky$4;(?Nm@=UNl{8S%zIpdz!xZn5NHsrs)%(eb^+q!{9m8
z-dC7#cPM|LupH1EME14g`8{(w=71|Nb<XeOqxlWZ7(s^bml@v2km2d2x;E(FnW)J-
z6Di-&Q`*xsG3~vHB@vh0o?g<l>;EcIZ-0|Wxre_M(N(C!?{Ru!A3s~OuLv}Iw_UT_
z1u5}~;Vwe$`DdN88g#)k6kBkgp9Ynw{k$AIOnqsx`B07ZJKBedT~xC1);BeZr9S>J
z+eUZ9N8GOmnfl%IxXETGNufh8<>qnw$>#6%$p(ULu6+%)WbdKYg<4_mP4vlllHZMz
zT6YlXokTD4=D^@ZF`ThkL0bpCf!&*EM#9iLw}+eFBCkHmS}9iPS!SzvgN=HzMCS8W
zWsq!uWE>7f3A2L|DMXXW%%lI#AR}w}!XjItd%eXgDyLV-7~2&4XTe^~*7{hrM4NGB
zxD{WTbDRQ2moa&oh{%>srkF6hfvoW8+HVDg{Ge8Ik347EUPV93V)U<}a=$EglzDsq
zc%RTdCZ#5*$50jaPgVv0Agc25Dbj<oGl+L%KU4Km@omV>GZnJ&^jnQD2V5tz2UXyT
zB*<`>dj2u^ApO0ZbkHB@V7q9^Pmp;wG~~ImAuk_SLyn8?tR4Sv0ROPPE$#SdV@|YY
z+~ZVsiO#+~ePWt5Im_;Kf!?SK7pUeV+jO$+^Y$3+8-}e{J~^(|&)*hptgyM;J%E%r
zu09}!Fzg2&2`b6{d&lQOsf|il^n$VA$z09Zb{827qE)s}>i`aNszhlWz=3;<c??zP
z{2f61@Qq=xg<8SH*k37|?utB3H(h$NgE=zD5dAfDFbB(K8JQuhmgpAeL8poAtV20E
zfP1Q~Ls4BLBiA~T$2ir2L`UK~%o@8hV>lz)wG?=ZT()du|3YroVMi2yn!1^(K88#<
zUQlo5ySCeuK1R)*QLVJT<q)TOmDaZ`RCRkCeamAbt$(AM){n_XyChGeO$bf)B&P-N
z>96QX0w>t>#!bBl+zlj#fMzthZHNC(>iGS4@sEgv^KKppP7*ns4|bC7s1^8dsnw;D
zmhA=$UpZ_N;R=p}xC%CvO+3l7Xz7>f28*3$wdV|>*mLg88*$wWr|-4b^LxFsdcAXc
zJ>F~X^=Nmw*Q5R9UXON=sr!iT8Ph+*U1D`PtS(37<>o<ksUFx|lc`(O16$MsThs%C
z>VZM^z@Yj>T|H1&57g;_u<HsYHIeBJeI9G5!Uf?@sa!(!ddb~TLmH{-^yQ1!76x9h
zdKozk!@Z!7Ja0CQ6*ra1Uyz-uE*zy856jmUmMlJd?cx=gPo2JE)snLcXD(a3bm{Vy
znNKb|hxTV39_hDBarvA8mi;~POx6cuRFuzft+qQH-(ai!*lbnXuBiOjcI~Y4w|And
zFJ0B^VeeH|&Gd?;Hdf`srAXZ?WmY~@m2=m!z06{|zRyk9_uF*+06zd-;|jXQE^3Fm
z{&29%+Uo`Ts&qk<jM5egAA;S|;4hs<8!km6EpwOF{g@v2q7}VEot@5)W$_lUQ3^fc
z3VOtrX=i#|*Fzn4u&dhZX+1Y;GNmo0K58lT=d`U!7o<>g7p{8JXtx~cfYk%2*kR?r
zAd((=vxL4o=v|GaxOiaI%H>N6)YWBnQB&P(FoJa0z4p6R)z!ADij7WGRa~K}V#~9$
zs!H}&H68WOb^~{=4IH*Z0S;FHhmFzBz^$9UcVgnxVe?V5j|<f7Bhof@!3>BTeA34u
z*<2&fM39yuP)iZ4rOhfgrC9Th$cfC@(Ci<T6E1RzJ9}CGl4XU}i`Oh(v0_>FL~fT3
zQ8Wj|ym8#qv&;A1i%bUeU$m{>>eg0ov$a+1FQTpD3T<^dFLgU>tL+nL0#9=VnkulF
zGayHceTa|=9UJ13tnho?a`Y2hj<A7<9N`K%!d7C3Ia*Pw_j)o{bd5M_B}xe=MR{{N
zjo0hi9Q(t`fOvcJV_EYRZ08}TxI#{6@Jk}6b4*7{9bgkG_uu>;eiTBn#}-lFy5~@Z
z?3vp33TIGrBPUae%RaG5H*=;JIHMA4g4j5<v60bjrTUOJbLI~JV`R_B3RKuWk+w0Z
z3T_@pqdJPqP^)T&(>so~)e32e_m7da0^4noT0t&V)-3IZEYT)@v@d&V5~LHulFiHh
zzWj0+R?-mRa<(mN+KwG6Ja+8dvu28LX=u=MY*4VRhX#!+K$*ucWi;qMn^4uD=kucw
ziruzI{??76L5FtcuwGAXc$xRkL;IA})8+~40ZtY@01d6<=mQpWvD^EACANyB`hcaf
zB4NiZQjv1^0nbZ<bngS6&M#;50V`}-!=@Y>9<I<v`uQdItU2}pt87rP0|zL$0+i<~
zD8@nkY@1Nk2b{x?LMS%MBKcdlE!GFb_nK(d6RscRel}0{!X6|3G~G!&{r}bOw>}@}
z%5>lvvNIUhJiEd*z}_8fw$&-s9aLqV!X{ayPUY?n8cC4u-N71uIiovh*|LTWHMBuo
zp}B0~m)x^v=ng~%*6n@3R5XUEXbd$Gq+%)>Lv<<|!&EeeAPCG<G=`~Y40L`>7<geS
z8iR^nGZ2rFkO)9AxuP-jjCaC&9hr8>_qq!wVf}l|B}&DAt5GbR(RvlpTLS9I4|s|P
zk8NB9ZcxFT4+mnc?bzSR{6U9vvF67dogBnudpIFSn9|}3Q(9cPPif<2VmTe_bO*Kf
ziJ*lPJEbhbUC<#X^%|ejG0MS1f>QaZEEQbEN+mcgYdfu6=|rx>mQcJJ3KUgdQ>V~a
z+m>+*@DfEj3IzRJ3c8(0UN|j|SSybetd&}=Z)LqSMj<z1Y?ZV26a4nR&;%Fr9sC*_
zAX%wxFXhvH;Q?qnM8&#=zZ{P|(B7MZ0L7^(?fpk96oH3mB3k+Zwbo|ijl(!(h^!=t
ze$F`u(QDy?6HeI?4OM#n3_4h@J%b_>Em#n{ap<$++@%;n;;j9JDxMTE-D**=V_c4h
zNH4U%{3q5rbTl(fM=DcLQoA(%y3^<It%RH7SqF+79w?j~O}Y8K%uQ;E%c2vR8JfV%
zs81#Ha;V75p{aRkABr$D<qP;Q9C2t+Nwk#qyfp2Z-rx?iJu-p0G<`Ig-{7#6#~bpy
zQ`AIh1Zy9YmcKZ)?5oMf>W^HY@rl~|ku{iA7~@pNTh5ciPDD9dm|ISqQX~$?X&)8F
zRTteQjzb#?db*%%JK`Z~WEi>bG+^dR8|OfAqZBD|pbG@h<I<b>Sm60v>DEY?JqDzQ
z!(Aq7T;K>SyFQA9sK)jAcH!BhCPo~Lnz+KKi7PQ`&hZVqJU`<uB5T_g;}C@63PLd=
zg+dJ*bicL<l|zyLKl~_!Vz6-?-!H?k)Lun{5gnTm3_oPY4F{ZFFJ*{1mB$y__NVxN
zr(#H<4f67q%1}AE#XpU!sb+e--mvk_k7a34jw${={3tRzgRfjOYdoUWH$d6f21O1C
z{(ch5{u2V_02`DZegO6AKndlb@k24x#)EB8WLN7yMnXYv>Rv!RNoqfq_P>MfHCdaU
z$fD!5$=yc$3vMN$jo5Gs(&4r&qRWN0afLdF&bPxXKF-kAV!PwV<jiJ#x*j`BxEMc7
zxE!prtViBHE;h<jOiY@t2P4ojMu`y!SHbu!bP>$bgW>GZP32ZW#M~_e!36^uXG$5r
zc^*j<)8vpin?7{%xwQTx{6_RJS#g4{zmtw#6^#N{n5-OW$VOkaP9X{#JHpQi`#sJP
zjn4{WXP)?1y(turD=6S$1wIVa;NQ2Hp8E;?#G*+2{+X0|Ki_&k$%#(UAK=6X`R8Kl
zAES4E%83v0&qsoH#1QUZ!%zS?I6C%u7{&dc@v}FQj<hL9JR^U6F+GLqj4Nd1_lAu8
znVyT!C4S5DAEuvJ)_%vSzt>a$z^Q-apTCnF&Q<{aU;KH1Un!;_SqHR5qrUmSkic;T
z;D0m#|5XC`FB!+sa|7T3kcTCZ*9jo+5?U<Lb9OC$QdS3C0nDG>!Tc)JR{-X-vPiEI
zV6GQne$B|OMf%YX&{HT<Tmj5q6qqnD#mEwRtVTa5zxe9+2JzW1+2@0cI-U=T?jwDI
z_r%x)Pd%MIV?*zQOydgOGOmKI$i(D&uC1>}#{{9cf>3l($-X}A5M1ueXg5^#V*c}z
z{krLd-W4`*S`P<sEBU1)xHNAw!)B#PPyIG1T5kv_=nW%K))84{ubwsK)>wnX#Igr8
zMW?1<7Db8)PhpNJptneCQv79#^gBdRY6lS|o#GBt6x-o4?HCmso5nbLJqMNUS+tr$
zSC5ow-cMo5dp!DY-aLxXUK}DKcC%0)^vgr7=Dc~{B5wtK01xm(i{<y^!(Q+(Nflyf
zJvdTq;avhn^p_xXhZGXzLp;|Tf}I7Fvp83?nqz~33*ao8ByXd+@x~%WFOS`j-)nBJ
zj240?@Krv`&+yrEhszY4nnL2&J${8F<u)o5IhmpxBOrJ95pmjW?7IVvv2vfs4ho9|
z_)uwQ*C;B<`N$^giLchLml-94Zsfo5ns+g|yF}_eEF@7vh~-XdU&IxGj?|c~R5+Wa
zgAo<U!!tNjboSq-8pTd86{vR53BQzw$Z;JiOTi%>VveOoM}M86JsKv4=h+&8Hj<(d
zU?hz;f>lHj>RHk#^FabdiakDBAqP>Z4^fDGhQP*3nL@bKsCsj#bb2JHJ)&UKYCnba
z-&COpGX*ITl}Y|9R;ocw=MQo4REp~z)_3_BhhP`CxW-Y`a{j~7(dW$rKzczyq{hq1
zdVXa67EQAAui(AlE8Ih5euVTEYeP#bL&TO%G+oXVt2nMwgjqL=_o&g;JaM=gKOnWE
z2?*eYWe_P<4p+`cETAcAt<PLGs>yYw#jOcJsft03EmHDma@38X=M=>hBt?|E$YYJh
znvpmi&@Z_)p^dgCq|NEjgfORzHlb6*I5JY_xb##WK?TJ@8T@bAC#t$xj1&f+tWcyQ
zInvq?7f2y|$l83Is&u0auvrKhxU32|{UOAtch_77vI<T>?|fuT0b~+F*MQm(be(ca
z>{s2&y=E(SZTgII$MiW`?tNVDi@6U{g&uTAlds4leWd~1o>ba$&EdD)N_&f~w6*mK
zN*n8wNNKw`5FSK>W0Q#ek-8c5M;9fLI$4M^Oeel&-!oXX1gGbcnVx%@Rx_}0e2@WQ
zA%-iqlC+Ol0y_#F%)ny<pF<*ENt|gASTsH(EgHCjw9j!`WKYDuffKNv4mL1@eUV_%
zFGRJUKCo=vKw)52VeOJtYnCl7@IO~9Upr7(x_Dsm+GX^2+1e#*maiUIzN$aS!N<ZJ
zwBtLx3U)^_BbeLMXV2atFdrtCoqa*D5wYG6Fwb7SczKAGq2qJuct0WWK3lW;glcw{
zXU2M*z<P!W>k$CAs(-}`A{rf7ynMylK%3V|bPH0@Wk@jNSuXI58u0j^msJN>s5)QL
ztIii?)oBV*FExppeLht%EQxUiQS&%0g36!E38;lH7ot986193$C<<2)wU*POqSkN%
zL|rdL{mvxn1yU5QAnJv>sPm<$KSI>Xo@oYIZ4}-@t$IZZ?+Z?amm(pOH&yXh9vp;@
zJ<Wv3|B5U=Tp@Qi>&16dSbQs&1j1IEgtccTYu1WY{m%`We5A$SD6@Hra~(}cGG}*f
zfbi^egmLvH<e~=N>V(iD$&B#a0O1Qv2>nmV0>Kpu<T|ZD`cZ3FzAzB9!6@ifDF{~(
zbembwU?8Y!6m<4i!?uJg2s)S3BH0TN)dE2+lOS?vS=B$VW))GY*Kp*UH3Nmymn|MB
zEL*pF`I^O%+HtqkeQ+kWKJxY11x2v6rysvS3<4{+2Keaj<up!);W2|@;K*}VtvM@>
zZF&E|vNh){UJ>-I-;l_TOR2Y6B4U`1J2Aj?VStJL8Zhm<V%2k(E?*OT|J(BWN2h#0
zd_4NCy=V6Yqu|Pmf{)SP<j4G%tqDsSS19RMa#}>1{|ZjvUUTIofiU`O5cX;*3|A2L
zT3y&>QrL?FVf5D^tSyD%3c_Bm3)?P*y(ADue+|NV2EuIO3c?QNv}iUD;sj*#rGYT|
zYY_HEi40c|wnInuUlQ4+fiU`O5O#$WhARkrt1j#<QrOD^Vf5D^?8d<`o4A6on>j6#
z&7j@313_;v3Oe-Vp&(pA&=WW<BFKL{CqQ5S6$qoh22>}1B@~7$2wTKyQDM*E1PI#^
z2)n{4>{(J6t|06=y0B+UVQ&wFecUMQNB<Qf!xe<x$7xYy_i_S|eKHVsODHVxVmUR(
zu0=R+Obt<YglO)8XNsB_(+k;q7g!TuM8g#pM7R<QqB*ARfsb%6ljSR+Ef*jZ%LVsU
z7H<!T6$R>u+%`9`tv_mmgY^UQh%3OkhF|Is&d2#3E1YX>aIic899#j;Cv(90loOm!
z+u&e5064e;oa=MI`HT~s&)VQ<>jl8UdZEL@!ji&TSU1?<Xlo9@!J4B(I9Qfg;e5^p
zM_ac54%RIl!okwU3g^FVaJ01#;9%|3Asj4^tZ+VWgM-x(sxqzs=ZpMOhj6~c?^xk{
z*#<{jI{^;XP94I*^2!S5D>gV-R6#qq0-T$2Xy;}p?c8F6gB29u;0kcQngh;lPH=9w
z!O_-LfP-~aXNA>ouQzwtplB;IK*7qa>rn2rLDANBfP%GM*P+~HgMw8b>I<#_WhcMX
z85I9({EoFXecc8JOFh8B72td`2b^y?!MVo<2P;3o!4=?qCkLGGI>Gs#4UV=_1RSgs
zJ1i_LCas0_eH$EYtqC|-Yjy|+%TX(wAK2h%t5v|kYPCZ+SkPMG{LltRTkir6*1H|T
z!IIbt=SMa;+KL%)uww2I4i?o`I6t<*(bn96gEe=DaIh@5!nxN5M_aW64p!}*!EqR;
z?z2JB*8PBjb$@42glWL4nV;C;XqE%Of#sk>I4~|);oNV7quCw+2eyX};lMm$h4X+7
zj%K|899S<pgaZRd435~|Y(Fb?PjModxwgI0!(w(&_!l6Ie_$|O@-vzYz><S2w&b+m
zHh@ijK2U6g+#!alV<qy0YL;=}y~k%D@L5`Ps8+MwYOy0>(QA;ORDFaJ$`q%%r%GpQ
zaXNp^gXcxjd*0er{p9w-2WR2XuJC>@T&&6$wi-ne%ff2St>jfG-rH2Jm5Gx{TeKAV
zx0TqP72fUsQ{F<Ez}Sx~WakNIQCi^59vrFs7xw(V`LvkRtv`p~WqZ|G2a7$fqayyI
zS>D_212~&->IrnYQtI-s<j0TTq$BC4pFOaJ$M3oxa;sQDQiP`9MaU|G^9=dOC?Ct7
zV<Z3d@G}K={v!8Ea@P}K54q4-;Z!Klb0{QmIE+iE)JDfzO?sU@_SE}GhN_#P$5$Y-
zuctoNGQK0ygO`9lc%|9r(SJR}(xdMXMNEB$k~X!56JH26qK_8k@GQr2@Ox$H;|isJ
zlqvv82e^h9xc*ne;txHsZ=(eNc0KhDPQ5cs4Lrk7kX7Z`0^$q*AA4T{XGc-(oj?L)
z*a@p_mnCFjG9h72Ab~&@0+~rj!e)S*o0&VA+qpA$xJwomk$r1GAndDvfQSe#xZ?uP
zPkoAt3oeMFA_^!fE~roa&N)?mYU%3kx-*l&`@Z;luhPuzs{jAg)>Ws@Ips_72&PC+
zL~&0C#cA-KY3RR42!9Et!{0T-{bQKOGC|~dUm}0J3*@4aLQh0w?>Ld3n@kpefovVY
zUm;a2z)FOM&5s3zUuIIc7CJllCLhFy*z`mc&PEF2;-X**{4Rz^a4;Wb|5||jM(%Z_
zbBpwL-Ur7u33V7|z!Jl03DaKNDf8#I&+F?hR*PVVwy0PGv*1%p^`3b+Alj(^yWCM;
z5%{mrz9~2YVIBjR=n6eaWE%dRh#bj8jvNpXF{oTjeiPLMeu9GOHj`)JuTaTHGm~Q?
zCc!?cok?RK9E(3u8x9kN>cNh!sgK0HB6x1`@L|F-<XX|1a~NnY109b~vNcbDXW5!3
z;-5`8@#!5GG+^d5ocufRU8>0i_(Yv#Kb!wa@dA~pe=`4IEzZ-r+mfublNt0B1|4IA
zo;JjwXE5lQ3_8{Z%?~kXCxdn|=y)5ndx${;23^UZ2il;$LkxO0gH{=Iq77OfV$d}V
zx|TsF*`VhRG3ZzLX!HUGoos_%IK-frVtF)9FJsVyZO|)*7&K(iYZ&xU8}zy%235Y^
zH!|qqHt0=540;=uN7Z{fgHE+U?;K*#dl~fo40@yudjAlEKE$98Gw5_1^pPP3{Re(R
zsd^t{(4%e8$A=j7<5(Vneu6<~+Mu5tV$jbr=;s*pI2-h-AqM?2gFelmvu)7-8e-6|
zG3eJBbgm8hjUfj8HiLeLLFd_^-yLGm=Na^e40@ss`r{!6eUU+b#-Qys=u1Nk`Z9z5
znn4%XpuZVn(AOFCcMN)x4f@6qgTBe2e`e6dHt1i681x?u`cDQu*#`aB5Q7dsBbo*`
z!6#bbpK61Sz@N-5!=P69TQKM-20h&d-ExRQw`I`n7<8!(y2B8I?#!UOFzA^!=x##{
zx+jC~#h`f`G&jVc`!nbm1})g2V}}^@Kn87P&`uk)ZHPgqFzCSyTC_nA9b(X_3_6WL
zSJ<FO4Ke633_6oROE&0nLkv2XL62w9l{V-JLk!x%pbHqZ*9JXlh(S+g&{G(+VuPMG
zc+e=anpCy~=b_@|y_j5r$rYG{n7j{@b(q|W$(@+oi^&6+JdDYsm^_Zj$1(X7CZEIP
zi<mrv$yYJ?1}5Le<a?O>5R(@$`8g)P!sIuYypG8qF!?hkZ(;I3m<)Rbk`b6}fyq{w
zY=_B?kVFz0I<RFc;8$AF(-VElp2q5wI<V#NyVQYo;;ZyQdrhd^ewU%q?8XeF(KPRd
zM;&{f4szNS6;~IkT@XmKSZ-di`MRL-x=?KPAX=(oGhjZZD_WC$lJ%Rvi|U2+n`N6&
zOuyL|5ejN{5AMgGnqNH_ZL9Q~bp~qSlWg17@GRSQO&l)LZ|?m{qJHyS%tc+MVe>yf
zUZB_NXFoto*Kc0PpsQ`ri-#Cg={GND(6u(`l|u}w^qbc(=(#rNbwdoQ^qV&_==nD2
zO+yT-^qaRa=mj?D9YYMN^qcoE=!G`u`-d1*={G;XpcmVq4-GM>(r<o<K`*sIKRm>s
zO27F8gI;cfer$+Am45S+40@#v`spDCRr<}(Gw9Ve=of|<ROvULX3%SF(El1@P^I7e
z8iQVEgZ}prgDU;zw;1$#8}vIv465{--)GPpZP4e37*y#uf5M<Q*`PliVo;^ue2GDC
zu|a=1#Gp#Q`3i&HW`q8Ah(VQp^LGq-hYkA15Q8fH=9>(9mks)tAqG|Y&A&6~JvQh+
zhZt1pH~+(+@3%pR_Y7z?8zij;m40(1mP|{~`)$z8hZt1pH@9Ta57?kv4>73HZ*I?^
z580rjhZt1pH+Nys589x+4Kb+FZ|=#UAF@IB9%4|X-`tNuKWu}J8DdbS-yF}NAF)9X
z9AZ$V-<-stPuQT7hZt1pHxFgdkJ+Gy4>73HZyw2@pRhrv4>73HZ_Z@UPuifz4Kb+F
zZ_Z`VPurmLh8R@oH`^KXvo`30AqG|Y&BYA*c^mZPAqG|Y&C?n53pVJ|AqG|Y%{+sC
z$p$S99yHPhM*7VIUV#tk1Wej6nS#k-m`uZD1}4X1G6$0rFzLW#5hkZ#@@`Dd!elun
z-IxTJ^kGuPWECdoVDcVJF2dw8Os>M@T1;-h<Yr86$K-BI?!)9kOdi4HF-$&+$&;9T
z29u{Ci6oM)`ZJn#)z|cZiN5+W6!;`kUrsV}q&w@AjXG&s>4|Lk=_yjQIbw&-L8We*
zv%Y0LPKEA-y+v~j2p?}*40hERJ5Door?{2N#2=Qm=9bY-^Y}p58}UP1XQQjIEGgz8
z+RRg^LoktUL5pep9~pDYu%q|_MXywWDF^LPK(Sy)Vaa5gZN$Dm0V~W5@$U*uJU^>m
z$sC><i=L=ncd_dAw9_%LV2TO6Gte;Gzf--j1w2=r!wNnYd3VADVFOP^C@c7=j5JGl
zYNZJigbrH-IgS?~eoZ!umxX{SeAM{DK0YFlU?V@&ePA1|%XlQeMSdM&NmfU&4D_iZ
z>~41iFf4}Pyp=w1GO6c^YYM&nWpFyd$hYLxfW8(*@N{rTo{rvkH0CVyK@rU(tDM|V
z=<e)z&0|bYY=H?A#Hh~KdK%r`<s!X_aiQr~d}AI2#H9epI49s7CKy7jY|21W1^Bd-
z`XWac8p4hMv?AX$E*J^RvdAHX=;ov&Oogg|JHe0ys68~QJ>d-^{d1^-XmV(kQ(jaT
zA^@~8Wt`#HQ_jljDVCoE^^`qNG^gUzCSp%HxrPB~OW-Ed=tFNL>?zWTD9wW?q5oj;
z>IoCL31(`M&Sh^X*80KaN%RAlaQM~!GBhzd;b1dY<y)?%7H&;iGZ#DI#16u%le<%H
zT&>aBgR$0Uw&sp)%^la8o7I||O&^$E_)fxLnAq7<feU_eC(RsB<wu=hNczCLBI@Xl
zh5lFh`r3ZSWkOwk4c(p9(3T$xHT2$AL;Hjw-3#5qAP9Lcgd)v%mkTSVk3DK0{>Sm)
z;4uS&dZ`1mpzPr8xJs%oi#xfk(07fVi2K=!`#9H?MrUUgYWc8Gp;gSJEkZIXL|rH#
z98{u<b{sb>)w1YJW0SBW@rmfv2TX^&R3bW#tBuh8BXx;pxw8<ReN>ePkAfj@*xP~;
z>2oC!sk{q9HN(6BS5oGBU|8WSUF$D)mLT$WR}QYbY|RzRFo(=3Kx&L98duE?sw;wz
zpd~T7!~~DSlepFknoH|_t5HU4RpqFP^RlYK@{FOX*w3ztMJ@C2`-!(TK&W-NZ?#?>
zS1Xpg;c7AP8Z_sR^(60m{NyETkhImg<&;D6j<MyP-7>3KhRZ0x0kzz{6k7Mp4bh|*
z`H6T*R+FkDa&iG85y#pho&+L_ODwud)p6rt%99ODm+5W@aWC@|m#liyT2vj*lk*XY
zJKh#|Zgg7-w&?WfboIi95b`QNA;YXTRmK41>_kEyXbaiSZ7S|S!OuP1!2y@5YzR@W
z^%IqBv(nmAnIMph6^T007S(9a85^!WZ}1axT~>Q4lLvC?A|WT)LN16pE!?*DcFegq
z`w2@nc4;lDOfATjjD($R3)|7M1}<6#NlW!&Z$3Y6?fBeL)2GkKty#ME%xG=O{|^24
z^mad4@62jgW#K^{Y$WT!wybknjxUz`;qD(8pm0xT=gRU1Su@=2CnH%(rnRRsC?SVA
zlJU^EjPhP3oN;GtunG4*KN%m$YDZ<iLjHLq<Kec9)s}gErFyAQE}bLR3J{`Taj}?#
z`>p!N;f{w^N#n;;I~O`T8yI^cSA#o_3UGDS25IvL{py2kV$<4OnbzFSEBL8)eHgRy
zc^ht4e#B46N3$ATnevdUA~pJvwvh9*#W2iOVkNvm*1(VXiAh$?X>F;DhsZgS#GG!6
zX|&~08?G%s>L=vKv)WP_E0HrM33;?F<m~7R3+hoZRyV{-=}A8k$tFFm9hGSmxqgy}
zGi?#)L?YrmeAEV;hd<*d<L9!PQCVV<rzy#JoGqi#SNGX)ef23nA<3m6ts#}=7kRdl
zkh5(e&*i)3;10QJboV0OekNA?_P6Kl{OWw!uVS9bYJFv$M&7qnF>~#TF?#rU8?1+a
z#ZN}^B1vm}Wei8oz9i$kxQuVhit+1yGJYeg9hL1J`4*FmC%!{ueAZ7!@&QU~M`h|q
zuFNE3dt64jDy|m$H{i<QIX@Y{m(`BSvXDHdNyY`XjD_fn1999i$Q!6dah+6vQ|ugE
z^anSd;t>~KKM0yNy8V^+b?Zra_ya%j$+;`7iIt5b`EZl?C)wijbVM`JdHC7g5Q~Z*
z`^or1Rue0eN^%(|85i3!p02HdAwZm^oM_FF=o@5B{Gy-O<gJ(1w92THoYqO~lWnn2
zB&!W<QBn%yA>ju5-2B2%&|hV>uQDVhM|l$TR9n!_mRX>Tslm3k(ha(p@wtgJVA~3t
zT(}r@76jkR^};onRXFs3$K`9`%3^Z_!+#CEhUV9P6+qsYX)UdcTFGgjD&Tax0(fHp
zN3k1jWALh<h_7e0u`;2h5CJ6OQcJ|sy7An5=5i1aB4vu}V!=WJuj<B0x?1XzC*sh_
zI~%onr5D@3RD+A0>1x?=IJ}qOkvaZf->S+=I0!2is&EG)9G}A_*@fkuZ8MKOt{tWa
z>`O9RXHT(nC3dN{mU(rsZm2ZiZe+Qd6*oZB5gJ~*-BGFYu@jgqk&Q&DE?=&-m_N7D
zR|W2l?k&L8;^h@^OK!af!j0C%ZN0hN0{H4wSC?vd$8^0V5>cGw#j=gVdz$ex1!gyJ
zYdy*hH3Qtc;*;9Y_w?WUbp!G?PV4E)sF{K`P&YW!>IUI580wv6XgD#-@or?O>P6z3
z+XCFvj`~;`K6TcH_>ld{Px?P+4FY9dP0=Gr`n)auVwAqpzg7qjSK0S~$d%Q2`!)1q
znCv#d#P(M|fyt#jt+|zDH$|=>feW_4tE0sk_HO!kk#Z7NVtCR`|KM=uHJUi{mZ#hD
zmh27D3;y9(DgTqz3zX$KMb@B7>9i}QA9?|GbM6Mk`owZ+C5->&K@n!h{%U1aiDD2J
za;r+X<?hSP8VePLV`Bq!k$?Nu2E`gk>mte|o<e?5Z4~X==!4q8m3gz3fq6DSU2KBJ
z#o3k*Bk;Afx=@z-6p4iDVuf87&p}=ESNg}TT-w@oCM|<jf&H4891Z&quyf+?7!Iw@
zN>$iEQLl+siD=~qKgwa&SAo#Vl|~=LxRzf){4u(U9){zJ(yC(Lk+}-|S$w)$zyc7q
zQ-$1##kI4mRm*%2D}U&}Mw|K7H$}Hdt8e9`KtWijzDss}KLgVfU9hfIXCXHZDlFPm
z<oXMxDq6M`a#JSO8(_5n)6qis8~mjT=G3GZgQW<}Ve{wBS&$o7F4XE1OMR^n#kvYl
z@I@aFs|&LJT!gh*<I2=+eejE4?Cr0wjZJn#U&AeaHB6yH(rQ?FWKhHzs^OJ(4SyES
ztZHZ!i=B`xLSafPl)=&oK6hH*)B0%aa{zTB)fm`ehw3Xb^cA~}U&T^5mb8jhUMCb)
zhbp$$s@T&`!u9m5wnjf#?-s``o!y!{bHao<G-t>~)yeb|Y@jHHJ!(KEkz!;yglL~l
z6UZD`6?elkbiRI=KG5C@2Lf806>`y}+*_>oRAA14y6?+%Rf-TfUwp%{&(EHgo3(W1
znYrn?+3=6Ev~r<yrB`_{<jf*hLEp(0VPOqQoKA=gj~sw1{9ktkmrXG2mam<r$jkv|
zlFb}sWB+gHTYLw<7RT5-sUHXB@<SnwsKqN*i-!R;H+BB#QwdwA^!0*x+pdQ7Flp}`
zuLpUp#KD6^ztS>Jyn;g41QlGNL^hcYGuU0TYMf#nkr^yKQH?*XYTRiCyL(ol$_%!9
z1@NOn_rOd8GJ~aXMv;sTGuS<|=qNMTfa&ZtU^*1cD5B%A{oFf?jxvMoV>&fe<^LPZ
zVD~{8vloL0WL1SSgFTn4qG4CXJHQNfoS(cCvgB1}uoSzA2F7Yz-nYjLw$)F>wk#2q
z87xJLA`#cxBEJ1*um|~xJ0(k8Wd=(zsz}^(ZE=l*@(pMPdx)Qqhh+(=%wQ=d772O2
zE#%v42782`sME4URc5dhiHk(Nz!ueL&kbk>dz7D$GqQwKX0R0bi-f$;7V_;jgFVJi
z*kiMVRc5dhag2n$*cSF}HiMnzC+nOnS(O<q1v(>HFSTWTo6KO3_mlC2EE$y<EX7qL
z8845^D0j*m)(m#OpNt(@GAc7z3gbpHUTMqtzrYN3p<jJ0%Bl}#21^m<s6MW?>%*9p
zH=r5p5<ekN$r4hT!BRv!67m{b$hXf7_B1~+-<>6<GJ~ZUdL-s`wwOj+Za_2GGyH@+
zD@#aa21_yfNXYALA^Eg>gPOrE^AmA-mWav>mLdd_h&S3IvQ^H8HG}Q)ld(HXMr8&|
z0SQUQn`{}4zPbU;V0-+846=k&X0Q|hk%YX(7V>|S8Eo0FV*0WwMw!7<kVdMQ+w6)l
zdiaJjgYEZ|v6>~LGJ~a<kR;<BaT(thGuXPHjH|L_RA#Ugf|6vs>m4HF8b2A&$&yi-
z!BWIalJTCnjB-`HVa;IA^ONyCSu!d!SPHO7GQQuI@ttP|`(8isFUk^MnZZ)1PZIxr
zTYR35=*(k-nZaJ-C*x&VGAc7ziY!Vpe!!OTonr=jg`e10Wr?lKU@5jKiT#i*_S<O&
z8~O=)ZI+<Q43^@llAs^71$`%)!M@M00&d8v0A&VCv0bSGK4ez_Zwxl58SFYg5pT{C
zQJKL~Bv}&i!<L8}&0x{e_Z?{ld#hhJxIL>IC^J}!?MvO@BUU$f=b6FY=_mc&S<)*r
zSPC*s(m!EK|4uN2z1L6R`?3U9X0R0Cm<0ZqE%5&qGuQ|GD&@hfN>OI86o8p3<r8+L
zyyMMaANH$_N3v=|nZZ(IXsV4*+O_cxH-mlDuPz?TstaWXO988?E<SD7#YQvO;BmkD
z{%BTxD>GP%)lK#NS-ZafzchpWxL*xFnN`Ee43+|uQw@LKuHpZm%wRv|SFxYTs#s+P
zOTpKvV!vQj>_#)#cY+!0=lojysjL=PX0R0bo?84%R*S!r&0xP6uZQ((2D=*qN}EiH
z^!X)VBM%NlE2;(XN9^k?H*mjIL(h0{RfKrM6j&6DxWSRSj$wbnZ61uiE4}cCBA9u%
zVD#oF>a#E%FXT>Gyr2Ud8p>dTO#z*w-!a}UxM8j#r%o!2!tYA|JsI^G%(OAiEm-k4
zXMi0y{rv<)Cp~jraNt7v=MJ!d?}(%i+Tf4lFI<P$Uykeu+K!K|to%E?)$???U%9s%
zOvc;9rK9C?VO1gDN%6XC6oN8eE-hatu0aid3zbS23enTG01ClSWsH9Pdd+Z~#l`*t
zMH*OKUt2E9{MT{*=T3l<#8sQ&s3OGIZ<`ItI&tG>xcjV9cQ*w31*`j7R1U;qhJxB!
z)`^=xHxt*1Ze^msjEtTRH(NZfWA6Of0yW&p<kA-XNVl&}^^m?Tp#Ukrh|f#+k!~Sa
zZ9%uLb~tWa4L4nK>H^7Y2dh~6g-9JX08*4Y&S;ciG%AHr`@$s&jOdpEGosvaMlFI-
zOERO83+K<A*C9FWXw{PZjaVHqfO?Yo<Gi*MytYi{wfX${Guvn8r$}ZyTg>=Z;x=sn
z+_;1|yR8JftsLy06`g+bd>&#D<?{*b_*Y~XJc;8EFN=Ul<9qOjaUB75<Bs~e3*{(;
zk&S)2eeMW?Za;s77WNpjJp(^p8JT=P2LEwI@EipDH>QCRG^%Ox2<0<XjyB|@3wKfE
z(`IinY}j$tA_nIxfahMh3_hP21Rgx1Fu?I7;o*diXBe(zwzy3ohw}%9HGwy$wY1Eb
zF+(`f=Z>qCE7fsGVf-|hVZoak{<moMEbz&h%>RAt{F7(HAExjh<}R8&y953L{uwhn
zJBu|^$@QW?O-swf97fnEEpH$}`f-{>B&;?lyyYl<oR*tB1M>*h<w|8`qhEi0+~o0&
zw^7hs{N<%%q42S1#xkIIQg{(GV0jef1->rmT`#^wEBnBix(EKhx6rp1qed);crfU>
zgkiK-6(ub+J0bexTR?oQNxWep`qNTVwAP&bZ>+*;xhXUF4^)NIatF@<Pa_CB08Ua&
zt<Z(R@iC~s##>N~NMYmy<b>?N!?3ctuZ3&9mYX;cKgakRxLkFD?0pdR1p5ev@a%#Z
z2c<If4e^y|-&^+G*GEL(iQr-Q28li@#!bMMb>S}7fX?UPmWMiyC>k@LQe$SE7&O68
zpwrSY>Y#xm|N0{QH5{>Wbysbj_<J$H-W?673nZ!c`VFYyar|l?10V4#8anZCCkWEo
zhfQCEfxo7nr*0Mv|NU)Q-0zZLpNR?95ss|Y3)Q;hFQ<rbWM>6#Ww8FZCHi#=a8us!
zb8{s>YAxg_hr()fm_w&q@~efBjD3EZcnf<!40mi_(IHxhiTD-U1EPGwS^&sJ-jW!g
zqOaIO++-mQ?A};RaSileTz<*U)uNFV?_vkXH>vi(8?S~@L5p2c54uP;*8PKQtWo)1
z2-yKlq6XU4ZXoqLHOl8yqsZz{oS`3-93Qec?!5r-Fi)7Lle$&Fb_0&_fK~!L9-h(|
z>izqJlFo-LI?WQ{gR#*5(n1uIy5jN01RfujJRTb)5B6Vi@PL@@TpwSyc!X1hZ3VcH
zHkOx5ozMrxRERqRawUgd2*l$miXN)wIG(;9mlFK<ASGaLjc_-+*X2cMau(tlO$f0t
z2)C54pqsrM`(m9QO=#bgG@l(D&2Wo+9vp0YkPG@BI{IJqGmVh1XjJw7_iah%JA<PW
zu#r%>N37PW#jZwYF<)H6757z*jjXtlu-U9voK^N+$?<!G;}~w0&vzAz{TL;Mt8Mof
zTm2QPy06E{j({1`&GiS8>+^%;3e9WqjDK#UpNZo)G>&@n{z%gK@!;t6JS{reXsfpH
zQzSq5&sq$^J(s{&iPmTG3x$g>`Dx)n`?BQs%Ha6XY%eI<TOj&WNsiyPIqroV#fme9
z<g1b-ER57$CdrF>{bLbbB^X_f_(@!zd<m29VDckOUc}^=nEV!#KVtGXO#Xw(7Oz6G
z4JJE65|w9c9JUfSh6xRXm=1#7F$Z>y@F1VPm05z&HPAXNoH$c%5b8Yz48Kt9hwnZF
zSkko~Iq;{4IEfH-LMj}#6C1FfL<>nlS125UEkLOvyOR#vR$E(xz%-(%!o!<o&Dmx5
z*Axvq2sS&=b}-x&K8NdE+J=$U3f#6Pr_&t_YVsGVPcMCgD|ejOXhCeWAU0|~)Z$WL
z2VLtiXt^up3RYIzW?XGM8Kw8at_$vZ0^&r^OGlGXK%C)@g5i#W;kLw(X*zj^n^n5I
zK`AKfakK#y{X|TgZ{jF>oaau0=T5STVX5G%!w@ce?1$wLE7`)G4SL$I#L!DEkyJvQ
z<j#WR&Vpo2>lbfx=&QwkS&`e>II_JDX?~z!nHsxr<k<c8+Yhb@Bdy#ng5)l;CX)us
z2w)cJcDA>v!4Fh_hYo}rWY)=z!2OEkwyWT_YdW`{BjlIBpaq-z8ti-}GcRSBJHJkD
z)Yy~zP+8Cy)!uG`;cn>+p&za+_0vXFwyxC=)vp8XhpGgPZ*U@Pj72l=E{N_fh>pNp
z9nBH*EN;S|4qM=ElRPZVAgGX}x{FaQ`p?KU_}&SSNM)IXrhc`j0PM;5ghrE|gx{vn
z-(T$O%F9VkQrO*~V822%exPVb{X<7M0;h3vLmu=VH++~w?{H!u=&o#zF4Y#@!xw!=
z;<Xk1Of-L}XhwR2mMR5<%J-Mb>pG~#Vrs?{dj-2Yn`5_qy^@FXoUT$O-w7Y{<&~&O
z_cH3!@fLCYvBp(z*}Daq&o@OTU<<F%oIY?_D+X!sLIU6y1mIVj1DpZsQZX@@fEdw8
z0=}34_;msJlT1K!dG^(#X$Wo82?s>^Ri(~GUm?HhAkl||HLt=r!Fmm+_{@@a8L|Pi
zRnCio<UeUdcgVGalfq<4;oLz`fHL9~{zVkpHy`feuoZ~_M`pc+=u{wh4>RyurJ5zg
z@Y_cX7X)^664)Wi4rg{IjSHEE*W%7qPMplnlFU9%GFwm~JgWwmFojm{?m)TY!23zy
zRwwXw8hg68Rd$BUnSj@dB3??INSh>bn3KrXS`j=du3{FGy{kc-!eS9|3P(r^$2lpC
zHYntKd+UW2vO2C|B3{c>XF+i~vn8E{PC7=7M1%BwOr+@=agEXTMUui9P6|eiOxA1U
zrk#et5jgxWxRDt(T{T={ylT#pyh=`9JIWY;#$mf`s+*aQ*Q(B4Qk>9ANeKSE38rc1
zj0*cz7()<ATvEA>srXOoS{ZQ?Wl7?^h(s}X<4E{soG_^e0z6NuHu~UDW76tMb>*b_
z?Q_5~A^N9Q?VKb=&?K@|nB=xqXv2@~-MDVM2NPinaW8(A4)Dy~NR3EGrVpr}4CskW
zap+0fK?IJVYuBRMBQ3BYSw>Ky^hAY{jYL$aBg)yjtU_<X7pTyi@req(1)l~IbWPcl
zIH=s3Mddc8aywJGW4%zhGmFYyOyzE-a?g69a&H!u_cN9In9BX@g~|h2R6f8|9%L#H
z4W5c48Xg|NrjCUVGL=V|3fU_pbzAo}%|d~yw41?WSXy?7I?`vRja5D{4nD>F=xn$-
ze&N;=@Ot{z+=K~qRIIJUgD?0|C9P{3)h}Z?Gy^=%ROz%=qZ<Cc<$v=ztrbPHIYT;k
z&k1*<^W|1h2*YuAYfd{eHZGq3mR~E8T0E_lp5s>fF1HdL|7xuSXKf#bZ_bsZMs?GU
zb@=D~_>smwjUS%~2R~+hbokpGKM+A1jW76-qmzU*a(qf0{EW%biSS^^z2rxZPB_xY
z@zHVcD<((B#2UG<EMqti;GEd%{7ssXy1S9tOt1LW1Re9F)dU|w2d{EX&{?xq6Fz<X
zb<C}Gqc{Bc(Fs%<zdvC)G*iCG{OEwWIeup4{l$+Qos6ZC<CEy%Ehb0j&CQVu{(-r*
z%KMieKRUEa<2MZV2vm8)@rf3ibi~{wzfCc>#&0wH@9Zx&ohPR88^!!um>(T4H^<Lx
z+^zh`(K%-dIe88pY|G5(*jZy1E`;lSQqG;%|0J-3pQw16o4T??)t=}lJF&EMfUQe=
z5=onUg1!F5-8C+*bRp<yNwG{H9I?1fKOR{J@53P)#g=yLEZ>z?=X7dLW0Rg}Z0>=X
zM4dasdFQeU<@4@fAAFSx-B%S#0jrJKJT9G3^P(Ce*bj3?GCFpb`)AP^gD;Sb2jCOY
z8H-O^zw)Ou4s)i^8IQlGjOc)mwSx&v=fDBeY0aQBF^f(czCg7z37@ET4#KAa)y`zh
znNmAb@b|RZIT&9cI)~sB(K!^K21MsD%$Y*xaQr=u&Jp+m(V2=*L}wa4HAg2n67x`Z
zJ_?^wC{4%T(<sfr7l_i)_(YVB!KWrE&BQ!J=~#S9p>!Pno<d2UKL@k$J>oMPpPJ+|
z2lEi0x%iaA=Xm@*g-<XKU!W>F0iTG{iTKncB|1GOO7k&yM9HDHXvg2vD0ScqL}>v&
z5hXe~_NRnr%;qBFBrGa>1++Me+!7{tGL!qk2~c)(<if4-ytx%j-pD}5*bSe71!q$|
z3tvmC$2`73^|%b5s2=G|xk;+qsYB=Kbd{mOw&=%pMHbs0W?N#mbgJAOTeH<y`jP9+
zBF6{PK_8Q&6X?N^>-Qs9%_7H#*FlZRy?6q&_+ZF2;^gGgZCOiGR)6bdF}EkOi{_Yz
zVd-PKWd^J9_q4G=C+svf=!wS0TFm5MXKLN0H2MxI=VVbi7hfPM^h8w7TQ5}3&!R#{
z?nH&2h{^@)h01%gs9cCI5EXhNDi^I6Di>!_q2qU=LQh2H()B{+vMef>;|oNEo`}j7
z>xIgdSybpWo~Y0hQMr1(PzkfBT!SwV6?!5n*RB^T*JV+m6M3RSPekSV^+M%_EGjqR
z3q*yUh|0S4Lgl6`Ds&1@ROpGQ+_GM%+?qw@Hhh7o&=XO)eb7|Ir!iw^{0aC@fm!rG
zLJRC*7`*gWx{6?tNm88j8puN?c=RN#1jJWtobCw_d}VngIX#hi**@`d+uE`8nbj5K
z5emi&n!$hYH7vs@{ACadG;iPF!oP`6RJds?M5L|4+vtyZbOgy)s`>sxy~i-r3BHAe
z(XQg#`1BZnU1@l5O8Dm({<{osn#4Ha!K=BD7aZc)-)HO}q+<V&v46zark$1(Tfo-W
z_JaYlX4?_`gkfJug?*7>f5xz;VTltKjFGxaEAo6>xC#Cega0BG{8tSAGJ_wO*_YZ1
z)y|%LT;x|6^*5=guQKXuj5;wJRf;V9k@Twm9pnB!755K}`$xt#Z4#YTU55+g`{0WC
zDs^zPqTgibKc_-hj)87Z1O2ZI?NQNmvwI$VDoxb481?U|sQ<)Y3H5&%)nha_YQ=mv
z`Y^NPCj5EeMk_}1zf<9dbw=>R@rfoAkNOrPs#;uC62|;`Q*Vj|(L^!=pAsjM%@}%f
zh8~;U#YkXN*inqwl8U$$BW}%z2YVn^8&J`BZ!>tUcfyoyqAPXGwyEf6pAuE|_KfZ^
zq`;vIMsintD38WcsGfIB#oZZyr5?5m<4(wy8hcn5Oxj7p?v@I=JA>}gEGT{{l0f%L
z1<f(&J`CzHf>F}5(@^(IMIFPa2QaF%+LQ5#oKx2FQXRhd`oI~7*{MOtrve_xfUOMZ
z@fAF~QRsvDBK8$*W7tWlu#*{f3d6Q#_oJ=~IFdsJBf2TCnxhU$1wM>{4`<+U*}$~o
zux90{SO8V=v{cBW7;-v8dQ>q%7RvoSRtG+sVUJ0LJ(gjQV_4I9%h{C(7NSd_6A--6
zAooeVXU%5(IjQ)^GyXir_h@3m@2Xb%;d*qliBDwU`KiDi47`AWO&<WK;5CT2k}s@S
zQ7x_j>))<C60r!Nrxi8$NvQ-DGl3;cz%*la62PegBD3Wy{q<5WtUr)Kq06M8y7?)o
z6i#Cbr!xhQ#;274SbO#4As7P$mLcalz46ueX=y6{nT&rH<Il(*W_2)r?!<U5;IBk=
zl<zI9$wN?{74{B8Rm8GXBFmXbClm45VRTjU&;^V6T4$jQF-y!wP?b<j#a_YKJ&Y|4
zxYH(c0`4zCUx4~IkpnD;df>`b<X%SZV`Pt&7$Nu7SGu748T#y0=o&-U8QOD&03N<a
znDVNcUX=>EhC$adsMlOm=z<VO`9t9QX#%V6b#5x~)X0H5IG=$%c5~@{`vNSATKT=H
z;72p~MGWrIx8)#%v7N6$#C=onOBnmoRP5=DeK})$)ObIHzrq+mao{T%_^MRk5Pzi=
z=rs)NvGOfeYM`eea8qE_#jZ=my`FJzU|f$G6=Eq>s>S^BQlXY#-oT)a)<R9;uXE!+
zD>OBa;>U;NzB$4VetiP`&GgGv#$&Ev7LgK0`ZNYNL)pRYSe_nUOE?$?&oUxC(ROY(
z#~fa+tXNSr*G3(}A=!-CC*2pHB6-5?J9~<qEA!y;Qo+bvr8+pRka!QXX6!ebcrr%c
ztN5qaAfa$$MiibCw-wHkyO9B1@H)QvWx%rPp&`9(rZ(+uBR<-R&NVxUs*$ca{BpW4
z4JrRfY<eQKCOt`Ioy(VIF;-C8A(^Eyq(YZqrl?SdiP5Q9g<6_ID)cl}Xtu@EmBKvR
zQVm9jfeuQiXHl}WheYY!1E6GS3{xmA&7x!}6^YUr1E6GS6H_RinMKJ`I1;6^20+Qu
zF{V(;XHl}$kVI+O04P}s$P`M2EJ~J+k|-@703}N|nL?>Ei;|_jBuZTapk(PRQz#X)
zC|PPvqSQSAN|r)1h0=;FN|w%(DD@10l3MYRk~4);DT~rD{GigLPI^$H6byiprTk2x
zv@(m5r5`0q<pEH#w4*7Mdb2253RI%hHvmd%0Yxg)6iSsWN|siYDD@A3lBG>ep>%c@
zrJcDik;0WIRR=&x)e`AiQz+H4C|T-QqEsIgCDO#EP-<jR%5g1`vX&^V8Wbf`)22{b
zokhvg<PxPd1E6Fna#JX+&7x%KdWq6G1E6Fnc~dBzn?=de0~4k520%&8-lYCbp>%#0
zrNg){k%pKky=MTFEcI{-r3<nsSsG-b^xgqbvQ)<@lrGGoWNDX)(nSNHWNDLAC|#UI
z$x=8IrAr1t$<i~YP`WgWlBJ3!N|z0UlBs!4p>%l`B}+L?l&%;6B{kws!2y#(>B=lh
zma>{CT{Qqo>KjMO=@d#=XHl~B*F-5C03}sRq^3@xbWIi|OOs8Mt{ng+RZFD3PN8&N
z79~r;O_bg@07{lxJB8BqS(GfzH&MD_0F*3scM7E&vnW{_aiX+t0F+e2t;!g2H)T<>
z^yNh9<^fPr-J8_pDYbM<79~rQPLysP03}O#o<ixiEJ~JsohaQt07{llJ%!R8S(Gfb
zJ5joG0F*3kdkUqyvM5;!d7^ap04S;MO^WywO7~<@vUK%C>E1z6BHesM$>ALJ{wzwC
z`kp9}`d;W{eblFM=cxM;)$}+L7Okg^8#QcL%dBE|sjr9^$-t3rE!S=sh~^r28H2b+
zq);!2SGtOwaOw?KLTx#6LKc4oLp-=uq)%NHh9}gOKDdSguegGs{^BMT2qld-m%#Na
zaEwe>oCpiPMin^-g?O%|g&qvs(Hz`3R<E|;<unC+LSA^$La~NN(2^t?dgkDUlPLOl
z0d6uY!sTKhVHahka&vU5&|iV0M7j@Ub+NRfr(T152I$_5(m8ZhN?UGTH{F@h8WGE_
zE<yZjpe{=kx0}^Ep)xvq;C>E48tq+xtOr|q6T*)RQgGV~R7tJ`w|XF96sa4+uM+JD
z11aH+AP|>z6`ab!b!)nWvY?11Rxq%#jY9pn$zW+1?Hf!=qs!jlE}RNomxH%(6EnJ`
zgbB+Qv@1qZpjPo|M$vp#=sDP&;+hlHkP)YPg&2UCtypVad>a~8o46VUJG0S?pg!1s
zk$&*2jSNCuBoPt~*%T)>>GY#R^#U+>{R`NE!u3xOgFOdL0fR`ZC+b%<;?xD9iqTL6
zh8)#SPh-Uj=waQ3&SG0<g*!F|bEi%Nm1<dg0yB&iOr;g*u?XY1)CW;iVYonN1$f|7
z#U6tFxR66@QTQ)uCz~6$ywJH4@RzSe-m<3Tk2SbKX?e9!T|1seZv`*=6Qe1z`1Gq7
zynLh7jhBdE%jab1=ZMlcIr9_<9Ps)&duXa?%|ZM}`2@2^XQj8dRL_;V#z)-+&=<(*
zjN-IY0UbrW_X<d(=za30E363|lciN)p(1EZ>Vl~g0;5x5Y+C6j5Lsa&e{4L}r*n<I
zt|H98FdY;c@M}r5azD)AI4MxSpxGsw3-GO8XeY2#DX)bKpnzZ074T)MuCcY8oC$#`
zc7V8@QL~+(t5jQ=>xDSYvV-*DH4xZWOEsLws2YoXmBtF3J`obFJDLHZYN1SI?kA^i
zn$=5vbU7k)@D{q0GP2>&i`V+BRl%jc92$H?-B4K=L3Y3+=1Z{V3TsNeaJ_{vKZ+(R
zs!wchVX72aQ9$148-{t&z99#9a;@$uK}Y0v$klo(jdGW~-HS`eK{b|3G&<STUMsMj
zdOZG8s>6hU-&5=XZ7sSrh4#LifEunk30Z_<pczU)kvuPG)d^W(R)OgoUKNu7R2y9L
z0Dp!F9e*XLwibu&gI;8JX$`gJtXT`8)6^Pp7bz7xzdeWM5|erh(8Hw7L~bHrKuDBQ
zR}M|G3NRNz4YV#R^+Bx8PN?dBc?BG_jQKy`cxYEh<MNN7NyN1f7BzSro~|S}n&9Nu
zG!=)#S@Z+=d)kq_WlKTJzXvgs@AB^oIz8_ovmoWS`T$twTQ0N@Wl<sf3T0_QRLIi8
zp9-!-!^t!f@)*JoZb+@wr{(Ap327jLf401azz_qiQ<P;!@F;S1tzXxlN7gQ;0ZXF{
zww9HyLgbcp=<C1;EnNb7lBSQ+ZbG<a3Hz1ct!3~MRU<u7o}Vj&l_)@c&#wB-cpbqa
zi1g};O~1J?UfD^D{$v}j81Taj5j;H+{4e70!7U@<CR{R>i7rta2{~O1<V6hodol=f
zg|0$>9iwhT>}}B(z-m)eue1!O>N%dP^jC^sL}u$NA~L2e&jF%3rlgROc04nhx8vbP
zPmYxPF4LID$dPW*L;Z#XrzaX-^dvh`;25)qV-=-MA693hLaDPgIc70jI;{C2eg!R`
zagZ88A*Le{b3nTI9-YP4>Yc=wdS{b-cP5??hZ1eYXjdnl4V9T&Fx!vw+$_%2PgRd4
z&eUU@;~Zd@*1BylKhAG`yPcYooLnlyI!#*Gbs-~~G3d!P$)%V8?q$rNs_?dCJ3#Zs
zW5W(GIR0fIBdQ&rh}Pe=i9)iY`$o1Fv%)r7^`=dda6xDZtv)+f!d*&vn5IhQ0=SRJ
zTIl6k_@{!4rtRQg__u#UV$7ZY!M_eW5fao^;T;~v&88=AF9`1COK@k6U@gXQ6*C-;
zkkZgoT0(A$e;NUa!xC~IL1U~ZjbKaun9vio!AJ#ZREqgA**4*0G9i2Z)<0+AXlN*T
zjfve(mvSv^t|&#mPSh<96<~+?ihJo9P<7Ph=!wJ~l~DFnQT8<7vajN@>50nTGNJ70
zqU;&IWnaf-(-W1wbwb%QMcK!CmL2W@4bTm0!j-XJcr(}kwh36X1=bu-tRQ!8RR8ou
za%`Va_B>Jc3BF~20oQKSw)8}0k4`ANU6kFCQucrs8QQ#<XfbRqGRz$5iwv5#NlJR6
zZxua7iwwser5oEZZIR*HqpZM8{vSYBY#qmXrzeYw^-&=C=%e6IC9W+G5H)*oQ_dns
z9}cQ~dLnZ432Ba8I3fBnp$d}J2dz20L#!V^Qex{3x3I?}ch&slXk^JjpBIvYo=6V*
z&@?H>6e&k-TJGdN(1$GL%LH4+mT*a<zg$G+S}W-j&OB4@1Utw+KRCxvp7XNgvA$v?
z4}Ha&lxH`+U0|sU4nlHl8f)c?{Agd2MVr2F)C}}QwCPLN9Bngz+U0)auFNFIpB(yB
z5;^)*4u;&-e&nvnB4>STi5z`v2Se^UKXTV+k+VM5M2<ezgCTdLAGw>d$k8{Q`W!tG
zIr_Q}hTN@w<ZjO*XDtti94!w9L+(yLa(8Euv(^_xj@B20A$M<_++eqeMS1U!JhLv!
z@I~W-HrTkBDnqdl|2wEW;cmz+<?c?X2B?et6xq>2ik@&ocG?)Ycy9^Z5DePbYN@{t
z#}0VZT+88k2pt+1;cL*2N`{$HM4}wt`aVmXm+&83Tj&a6P)pzmGbzdeS%fem<*-tn
zV|IDmA`Y1GeR-S;THkqas?*zFsls7ztqAiqoKA!C0NvN&xUR3W+(5rWonLFNyHo^S
zyf{)^i{}`vd=zB-6>e`ly8&l$(Pu#Z2!~8Hycr%&zok2`{7e91`+`}s=jZ1vSd>3*
z=E;j^&d<+p&$rK9xNu&_+}0M9WQk<XHQ^|<r9~8T+=7ld^XBG{n?G~$;@OMwfCUbN
zK)DXbJOwyigQGZkkV@K;=nPdh1C=Waa`I`DWS5-2FjjaSzuac&sCTcm6xTr2cY>R|
zqzy+NF0Ocav=|sNrj=BlT9YGnu92V_jrO>5)33BV);3=_F4A-wcU%m2bWxGJ{+FZ9
z9Oks1-sa9BX^2%VZ8eQ6R;}y9wiJa>Jxa9SFj}ONf!+;zaq1oN#W=_wVqda{Q1%<Q
zK02K;qbOrkcCr>`Z&B>tMzPn@un)d-8&p1xc6y=)nQ}Y)O^YH1D-5cGeMLEknl1Y>
zObZZD><3)X^Z0}XnKg}vtr*>)+5xNHTHCCX+uKh!J#v4983}1WfwZ4I^>_>li|&$@
za6e(V7Z`4Ty<si5nX_gdJA3}T*>g<D7a4MY0l9xN<j)!MC5GHD19HL19kWctUozqt
zfjA}^@$6dEI4?8e-YJOlXD(W7qP@asQn7wOGTLwPUFr(2dZM)-Yofi*Xk&$JW0TQ-
z&uDKjS`^)m`g6Pi+h?9;;{B2F#tFP}$#`!v-k%xIm<cSr4(KuMCfr{cZoGgSpA7dF
z!~NY0Zc!ZWpA09JY7>&-{>5<r_JTvSg{E}FaMaW2J5We>U^3jJji{?^g1Kn0r*+K5
zb7xs-BN(k!ptUBWZHDhs$K2ct?eqn6=b5sNVzh|@ZDKOomW;L)qosAs#S0c5KX0aq
zw+-X93B0ysyzLlod&WzvOK7?d6Kyo3O%iC6lF@czw4Hs?tWmHlqa7sB4oXIQ7o+XY
zXlYYs$AVK9%!Q#~%C{#YP8NuhlM(l3#GDr*)Fv$0mRRk#FC$J7h*Odg_h-a0jJR*g
zBn^mh5yvv%!2<B$WWezZIDrA92pmZ>=c)O{#~nX=*2(i{=jSY%J=?6`Rt7&rfFF_!
z-p1gQ7(9w;<pPJ1p9Vddp`|YWPy^b$f?F<GtSASxWT8+N+LA>CKBd|02$HG&?BUL3
zixu5~X0j-%2+3paW&8r567G9MU0zB81u7h>T-M2Tj1|g&@S~_XE_{lTMaz9h^i_&y
zK;V}N@aqJ)iic`~TR{s5e2m_8ox<Zr!0bR2w19BmCva~RxGEs3iEBkBAlxX#o=a$H
zYnm1VmC)A-^g9K*iim2WTR{j2-B>zWQnO{2DKy0(AlSPE?EL~(g+w)Btv~|=8%6j_
z5gK*GCNM=BAm9fC@Ph(a#Y8oMt#ASaJU*qB*<LIu3@TwiB(OgyuvJi06WfX#kXhGk
zG;4}XVFL*E5dr&Q0jr{-ny^-E0D?^$)p65U6QA5&2>&sG|AfF-VNp$dD-b|tm5Tu#
z*~OaJ6byi{KPs?4A+S|kR1@3s_$O>*p=)(-YHkyfoc#&%Ndfs;0jUC`nvj-9KS3HF
z1sl>M7AnDiPQX7c;K|M3u@|(w_X$32WTgi~CE#ZS@HYkUj_7vf0JePe3D{WPSaM@G
zONfR_sLu-2?+8@#*LR>=ZunlP2{8=`^*Mq1eSxZiqnd4OdD;_d+N_DPCdES~*dGYk
z9|>4;yLSj{dD9cDu{yS@I3XY^A^un(z9<mKCp4<%EKi7OAH~#&sD%47f%{8=J1GIz
z@`xwgD3Y=3`;ZY5m5_fWkY5qV<RR~<W6SZKkc}OWRmZdjG2?y(<n&IMzY&<P2}~6s
z)$C4|XM1KX6J}CSRKk5-;Qm429+ObZmis#48jEL_teH_!iNGHPfj0#K6(iLY+;UdW
zY-SQXIV>t6|5+gaO(4%t5Z&@gC*&w9rK^K^#ziFtZwUtf6bw#EU|_kR6N4yjp^E{|
zA#sNV^IN4p&Hp1f{6}yg=X6JBw|vesI{+;WG9#m6CA1G4P7K<&7%nJ~GrEI<<zr41
z4o_Q<H5VL}xQvoqwv$}Q3*EuRaxNz>Y5R}t_^1TGy@cOG!js3j1Kx5ZC-^8Tv#UpB
zg-9jvJtg@55?qB$HHV_*GEU%W3**cvsf0d8qPI%)vl9k`<sDAwS&IksiZQ!CInNUA
zM2UNl#3f&GN0nPH;Dnp@ohgHtQsif2V)M{zpDeKtli1`G?!dNuy$L&Q<JU{p%F0p4
z#3zq$lKpUrKV9O#H=)ig$8N$;>+P1Wj|orS-2^{F!q1iPmnOhlPTZMQue<q}*yP4d
z*vCuklO(nZqH1<_%UheU(-z5^JCF&jdfXxjeX@im_iab3Uys#6`dbx2HSL@0a34yd
zaEhdGrldgb+B$`2VPb~UDa#8Rr4KfbpF2{GE|X<~#wht5Qee3BQ{ncflxpTGLLLgP
z_#L29#qX)l_Kw$uCM(>=Dn__UoYn$9-Wy%KBgaNK<lpGxjV2CgF|pCbTMr$8x-)b>
zPVhNrMORA>{wI8#H@bLlbn)Kk;@yOcHx8Zk=;lqu>q}7Gky!Ih##c0V>&VrcvfHaj
zD8k#_n+mX&fslVFT)wGjK3vcWrLTT&quaNi+xI#$T4-;`RoiFX=X?i6Uc2F5vGo=>
zu&suhxoVY0wG+--XoY6%3bJxsi;K*uBFwp39<P#&&J}CZa2vER!RR?l(YYQB+w^s+
zJPw?As$43&Xm-L4VwGOua_dLa%;Bbao}_t!q&WiFxyXh4sW>G0a-nZU1HJ1YY)t^M
z2{0T2erCl|H#{_6WnP)7?q{a1j@!Hn(qs)sbX5AFU1@RuUMcOxQrb;nB)fzS4(ARt
zCN7Jt*Hv_JozWnMI?FSo#+BR)1=C9;)628kCD@;-w#E@lz1T{{KD=^HTx_Ur?xfm|
zE+`^(OLO@(;zlK3A*o*{sc+3aA8Om(@yD?Y2gEs?Ev#*gs6TQ5M0c(vuy^N7;{86!
z`xbBB;m+l<9FZ0~-0z0t;<6;YPi+z1*_a?rQr;waZk6)f<1J6HoZDgFICs>A!g%im
z?|(X|m2F_`<$ZE#e6<dLuVj0FR)-I_*GcHp&VF%p$E#G;hUVKVo5tGc?GH%a4|&&e
zxU~qLfQuIP&JOpEv#?$x;)cE~GmA!AtLld(<qt{9KA)x|xo0@Opcua;F9)07r^(FW
zrunF(`4NAb;pkOjQmz%t@L7WI)LOivhDM`fB6q$_)toVrJ2Prhgquv9B0o8gODR4U
z>mRTRb<T$2Zld9%FP)6Cp?zNuOlo*%781xfVQ4D5E|&f+{kRnDlQF@PrulGJI#hC!
zq$ZjWCzhOp#ELHXOPDf~b0<mlDJc~!B=IV)(BtF9Mr7gsasFcWfvO1W!yDbpmcZPV
zFM;!khOL>C@fE>yi-!*r9zWL#(|@+3i|HA~5?FdE!zIwMh(5+8;~Vvn8$(XQfJYT%
zEO_TeedI=cq;bKxuD;nAauQWQ8$(Wlz9BvCXfq)vNvnn%6+A0%rniv>kus%)ouuM7
z>LWLXp-c=zDOUX_#@r5qkHhh=Sb*8K|KUFIWh2>ri8Y{D_tR?EFk<$|l(b+p`~y-J
zjKQaYv}s7dP6Oc{aRP#{@DL!vhA~h2`$=34pMvJNvZF!h&}PKF3W#xrHY1)ZiMQd6
z&>U}eQ3!XB*BKuf$c?sPyp#I6&UBlx81o%<f|b7V+Oe|N8m<h@ac2L9aL+giS!=?X
zfvVXsw93e!^VT+?CmQ|_&9P;th%gri=YwJC0YQWfgZAt^Nt_M8iRL)7Gli>HXr2>9
zH=|c*uO~4zTrHYo%KjIw4KZ0%8jC{Nkux^Lu6!EzfX&fPG@LX-?Lb59f*q6Eu*gKo
z)g8p7ro1zJx`UlZBN}ctwijYvL_^N->Il_&L?Tyq@CbKF=!UXB4Qs;0S9NCk;4)l4
znrnXc2MHYtSvX9s)#13Uw^A?WaR!oXGZeCpY#MW=+dR?m8S!d-_8)Q0T!yoV&&*|H
z&?xFU;}Z@4lIALxole4c#oH9ei108Ws)m7I+S?jWttt(F6tAje-;^YBxU7*4?m<qx
z?Ivfqt$2~k@gR46Rk7xzzlzWNslD!|XSla`(PLMaaOZea3BML%W;hooy{VB3haV!p
z?uG+QxG8LQ#L;KsRKvb7uC<=-_(H@o)_VH;I=zu&`??Q_W<$eQ#%DHUw;9)ox#=TA
zBr^{7%^Vs{?ek)I)p#{6d)S1##v6=Q`fx@;^bD7(<aae<;czes$FuDd4bK~&=4B@w
zSMw4r>2c%Jy!!h(y-g052O%dK9y&h!*jXoZ9NiF}?U_SFGmdVY8Fj|`SI}_k@#&B3
z-{b0!1kKJoKK;?mp|jldpti}f;SK+vaAY^=C$dI1I|GG|^LDrA57EsyZ+GX^Xlq{=
z!)M5A7_tA5>tG#cWS6rL(atznPs*zkSIbMo(a4K9I~}<?GJ<0S-kyA+UdM~CDt&S?
z$mqz<j5=jyi)6Vbd68x}rLZkN5lNG{Vje>Oi?y!qr9pK*E!~VKy#!qhuK9M$tI10Y
z_Hc4(Sz_)?#5_aG63e8MRdsIpKzWg6e<+tVfaUx|R5PprtSlPQaJZa3+AVh}pNW}0
zr(D}5+f9miX6&BrJWisPXO$08cCt!ZkD|iYe?2PR*2t;ptljdz@@j7O#d3cvbjK^I
z#f-&^u9r4FX4DX}T(#U*iHQqBcHnX;)Dq@_424=emsY>}rm5ZX^3t|V&@R%DB9S?}
zez~RrITxW3BVOPnJ4D7b5X+Rr+j0-{Sw*m?n5$!$Zeczh!_1-c+<a^mFG`fl?{>?Z
z%!evFmbrG%h8Gzz&CvT988o8WK(pM=n(V>sjOO|jJ3Y^ccE+dJnbA$(a#{1G|17NV
zbw@VW2O^*EuYhg^9isAi(TRgmgY|OLM^xnR)(F0UzHHB6@-<AJ#pJt~{1B6$V)6@2
zevQd%nEU~gKV$NDO#X$*@ZUo+5|b88w#8&LCc9#?Cno!1G8U5qF`0zP!I&J5$&r{G
zgUM`6=3&x~$s$Zn#pDc3mSNHbNhE<`33C|UvX1sg0X<=Z2T$PN^oUYZ@egD}`~_(t
z)W3-{q|_X|D}+ir3bVv}y&Vg;O-8qMM0A?Lpl?age+aXr;As5iF<@#1#yl4PPUy!q
zi5`zMG@H@?DbfEW(dQ(fAJ6FX7~PovW3t2b6VjSm)`I!rQp>4-OA5m_5#IIl6DV{r
zg#}E(7`-tH0v|$?b_uiAJlwx&vN#D#qP8EtiSY3sAz3U+V6lW*oXji?{dSCn6dzQg
zcvF)j_R^_LVx%Ooxg>E~0*QAsiKR@!I2VYKpt#EU?h;;>qfcLFGWHe{8~#P}xZtb=
z>}8ByU~FR%7{exa54xt;L7<ZfjFJSlk_5UE2y`=n6-*$#E8-4DR6woJ3-OQPiY&bj
zN=#vENns~RAxNN5W(vJb!B|wp>HsMep(6551{G$ovt+QBWYC|$pvnwt%)rpQ#u$Jl
zSFr$BAFMEiSL1MR2@U^J2WTWfAAWo^5v^wE^v<`uT<Kg13$Syn&bJox)9A}d?C}!&
zoCNIi@b5GUp3m6noeyKnt**i~n~`O^-ttQL36emoByd3jfeV?yMNGhOw20Na*86L4
zCultn7h86MHUPXg%$B$W3#8hZD9KEbWG+o0^RbyxZCuV|471r788FZVyLr3{7dk?|
zuhE-dU8$~wc;_aSE1Al{lFDI{%2f$eLi{^5%r#8qU|mlTqq3q}Y4qpo5O=(mUyhe@
zJ37d9%;Rv$;~2@~eF;2nU>-LzkBsTLRKx38;Ib;_V^)URV$GB!7E2O0C6Ks<N!-dL
z(tGL3VzD3Y-0UkZhe5j1Y7IqTi6n4}Byf8IfjgPNT}&W-cBmI&eHs5e+=JPvPo65F
z-z}l<O@O|Sq3>ttblJtsjBaM+fnrv^8lg)ifwLrm4<ryc4erMjJ@FwXU??|ZJrO?4
z80TMnq3T8WLcn#<CXEkbQB)6kNn^RB@rX_1S@?MB>-pfr5shHWge7~IEECq90}4#>
zu6Df6vK?<dwS|)rDsKBd@TUc<(GAk_$xs#B#Y&XHDys2d751xes{<MjQ~ZMH8VQJG
z&<RI{aG2WGg4Y_;r3QGz0bUoOZhELlmr8K)2i<S64($Ww6zVy7EjXx1@m6AaH3D9s
zT;RJTz^M$cQIL*|!s#1t5$+OKRnTf&^U5ut)34LWSr0EDGs|cKP+U|R#UfCS!^x%5
zi}GFvtF#=}L}XG8AB;v3?}XS_tzY*Az}SMi86!xjube23s;SM5NLNz<lmb=@Xa5?5
zW}zmxF1uHSa!Zl13SJUjF6WlR9ox_==+5k@ZnY(|7rc1C0o@BUK6FQXbk&7f;kj|E
zN(JthaNWDQ(;lQ(Yh^D`>LpTUxV^ej=SxgTPPmi=Zrm6LaPX<4|G=B-wN(NA?$d$d
zK)9l82w_n`0R>d#-{Dqq=vT1bfHm)$QhhDnUv3kLt&Kru(7I9Nj6xqi7FQIj@NXB~
zM_-SIGvuP1)L{j&9Dc=KEw4Yx71xM0wj3#R=Sx>{d1J*2%nU;lZeHnxB?sqdBdmGd
zaLY@*l^AnbQL9BABsyB4p<l<_Y6JnOP<w$rFe$VO9RfMD(wdNj7wXUwd|P{#Nqw;o
z7cn@KiA6rmdQdHD>FrY?JnM>nT!UM3z`?kfD;N7#!1eHjZkUImT~`+SO6O3w6Z1$?
zdx`KXGRR#Q^r0Nlsfj^}cMZXoKwJu>uKf|CkCy{Y)B)whJ}wp#RIvVSkER{KZFGV(
zwcO&KLbW(|;S#7Q{Y6-niba6WsEpoChJ(`8m#$E*RiGyD`XabpsaAnNHifP(2y}!2
z1pL2tG{K7RS+2WLh5v-R_#&~ES(~0^tu4#63CG@sENjgzOWYXXk6@;6FOlibVCt2Z
z!vq7#iE77RifdpE)ki%<|Dk1>wa$fi@#-W%C*Nb}y^YQ~e9V{0DwFp)HTozH0e%ys
z?>@{G<_EJzB|$-}K+m7+DVF<la2rhxc3CxQnPLH&oEH<K-$vf+Ikaoe^}t*Q5E%1>
z<`+5T;Uo&{j)KaJJfe{z3W5a++?@^+Fu2swRaZLIh?CaLr9O0g7ZZ13d8JWrG3lwM
zEEuBT?Mt)@t;-eT#coaI6N}_-EF7D>#A_1~1iC9_Fcy|4BgU4^JFB*(HqLb#lI3c#
zR@TmSprzdovO5S;g<3y4`O=46I{8k_q+@2H)$UY%3)*A>x>O^_!bwYXma;~Qb&kJS
zTf^e5m+zdH5{*kEWWG<r7`6`nc{MqdtyBBCQi$`V5FVc5;b<5ddC@>MFwm((^ocGh
zOIr#bXD*V+uq0@<1h-u8k#fD)U#{TbSEJ#we+r2jY|(4=wWvWQ@`hT$FP}y^+<Yv|
zk!&s8exc+IBih$Hna?9qN4Z>i$TJ9H?mNoOtU7tklB(T0z;KRoan|HMn+fAvOi{sm
z=!18b_}I;i?h-UTl{aoi>jn5E(&2*Fre))16hkiI(NMH11kFNc4=fg<y~f7PD7Zu>
zYcn=(MmKIoQ9ZzF0WD_V#?2^bz;H*qaWh)WZQP9hzrGobo6Q8DhVur)C7P{Qyw<A1
zVr4*URgqO+t8O(`X`Z7koMrZsvStmQ0#QsyX|{z4TT_<(F?GG%HW;VH&E?mm(#+;8
z`<Ya`v$!y9RTEB1A){(g9SBHyj?`D_O&6hw6z)2<jn-h3%H`gP-StUw`%#M;%`(wx
zuF=?TCrsO_Mmrwu;6aM%yH$0Tpi8N5oh*Bw<|JYxpm0n|wF`BsIN7BEWtdEPW=!I5
z*$g$upACt^2`Mx*y`?6EVdc<F?j%Vp<D=#zVT+{H2_bgU(&U8T$e$#QWyhpR6Emoe
z&j@VX6dssTv-Ww56vhZf*Hm_0maQir#Q$qMNa{Q-)2QZzVtc7DRmBq>pGaxERXi%6
zCWbagTeejWAA)#0u$@)v_*MrzlC$sl)-!6P4F#@cmepL9v#nO@__mHJq!7O2JC<9M
zM5}(wy34Ej+0rXKETtc7)02IyB1JjY^ZS@uk|tI-{7^X{EHkm@T8C}N!b4KTOFR~m
z!X4u&TuM2*X!^I&vOH@}IyOfOC#OiKnS2>0X!M<&?kQSQA%<KypUGCUEZO4DPVt$5
zjoQ-JF7Z=3*-GAP7gq_I9NO`LWeVptqS!1hbwaR?2%DS`V!7S)Ekiml`fOB}`U%kt
z>8Km|enPZ-8f9ZPwoLG}5uuLLq*F({53yZdYUNvp?@d;IEVm|!HXB=(d|s8$R(;_?
zDFfPhh%bdW%5yv?l%a`YG>T;p=-DXGT1J9l+F78>`)BLHniA?ACGwqV1ka$I;MXwu
zCMM5e@;oLlVDb_szsBS>Oy0oc&zQW0$-gk!<PAtR$7D-Pw#8&eOm@X&Pe>v#<1py#
z_rR}YK1@$Ea`w_2Y!2AQ)*F4rYU}_g+zYy0Urj`kueB|amL%#uVXqjDnb}q7C)=!)
zm!tFb%jId@zuo8SD<it${&8Y$X!etj_JOD;mp#y!3jFdV^}4Gizia$Y3)eE|aSm(@
z+q0^eZ}g${1GuV`8@+vChD;4%oICsH)wpYe|60lXh9=Hu-^0{%4)|`_HQ<D^3egdW
zvSD$Yokyb@?ld;C9M!JCh*S_aO7^$<*ZU<*{y+y2imTuNlyVr`siv}PB5AdMo8*7D
zKmRM4|LzU~c(z}LWrO5v_<^Lp?&f!o<oAH&r-k^k+~KZeTCVPll=20z8d?RW++8AA
zyl;0-$g2?!hs($(?H`bQACY`j=pBphI;MMoqgn_OO)aZ9b=2y`{)nt$x$c`&r@G0M
zXyN%G$@TFpuD3JS#Hyv}`!$SfQ7tWRptZc$OhIzS^Xhze97}%VZ6X!#Mx*ZuDZ!JG
z1hDRL-tlrTOEA_^XV_;og(jME^-2K@%LwnR8h>r9eNwXhTxMVX0Mp&iftf(n2&?9m
zO`{pMjIDudbH28Mt(rbB>3%t@azDguUERZGn+KD)s557MND}huy!9R8(~|L5vlu_Y
zj9s5=YX5v!qqlc$emNNOLHG>n%vqmnM?Re?gz=dww|rldM4y#J^*}z>Jn_k<sKPWf
zAgYj0r#cd-PL0TZOOpLwHrdZJSy#Um^#gX1y+*ZW_1kzhovK!=-<L#xoK5uUW{5_W
z+9c7mI{k^H`txk6Ut_9!IR>?;9KDwskwY?a+;rD4xFG`mXSBXnr7uaIzs~0QEl-|N
zZTj;}tH@U*!Pm11e&3Ve!Ey}w6NDTIUGjI5-k-AR{e<c5lh|s#U>)99Z$uv_!^7Ak
zlTOkY;%`ctZ)MYbiD`~WL|k3ONF#Z0>?*94m1!uUJu~WbhewN+ze}pH5hJrmbLw>M
zCBxko6;~IkUEmH^EHAOPjI_nP61So>Z;sKom?_z$FWe4sN!H2T<=(vCBy003Q4p?Y
z;P%3H+ru{%Q3ghC>Wna<_LT=FvPs-t8heGNvg<Xke_(xfnGVZN-pq%t3KKQ|&f44u
z*Fl#s>=o2*c~Wv!rfr&yWAUbo&{{-WQF`s2rDA6tNMxEVs9p5fWIo-F&9t#7_Rgw_
zF!()nqEKpgJtdi0#wL4)*DTPMT#6>D)>=M$?L@iMu6y5Pxvb5%5d_00mz#IW9z2n*
z+Ks0SmxPTq&BUI=C7}pS5TCtyq8MtIJ~6rek~iyyw@H%)=py*+*%M_@yZ5Z7h;PK3
zZi;vjx*P-B%O^^vcJ*z^m7B7;H;%!YEkTn@8{;W^{6sO-Zht~D341Fn*Hy;3cr&>*
z`c3Wq6QxjEfRt`#OQ9<RJV$&&5xPA3k|kLukSK}L4x}u)9GiU_ES`%lM|NETZ3~vB
zD@YVUX$_hXA=;S@P=vH?TAB_aQ3$0?NEsJ#?Ep1{=eUUHPH2TR-9n-OO3UCcK-35W
z6CkY((sT}qA}H;{nB>_-)S|A0Ft&T1853$gO$CuiTPY*FXh)UaH0`wdPE$=J5?4wJ
zFXEA|qG{p-Q&}X^R?3TUS*>m9Fboy2e_p+_c&joJ=_{p1YZe8EZo@F(ZZ5wrhL?&%
z7a>x1G=e)&fpHHe4`A{zCXZtB1SU^n@>xv2fXOqMd<~OlG5IbgKg8sxnEV2hS1@@U
zlRrWd;phkS;SRImWC}u!RP+OQ(x&_k^U)-1?CD+aYf(2(%WXPr*sLOMH^IqF1ObX?
z$p|1sL2NPt2(_@!e|7*NJh&$rA^ZS@tb?OCuq;@q`BTn;@FfroK^}TRlmKaLFfFHE
zZ)t%{U6tN^t*6o54KX=n1PhESkY5cyE$6>PFBZWIk-~Gda!iBixydap5EuZ$$zaq<
z!9o9dY}+LD^Vn7hfKplmC!Jtu*GczeM>#rsDkX^E(G3w6a!dDJ-aYxSeOveK?uUQQ
z6oFSLAB4<^^5MZ-Yfkg#0^T{yhp}5QBZNMJXl3A_kE|)g$7~UNMEx0qI`kK6SmO|X
z1}q>j7>xQks6mhpVPY?%;}qLgwB@=WVkpJPY{M7EBOXMcDU>0kN!MBe!Z_G~O8F^l
z4#e<)xKE|dIz;V(m>oq37=W>$rse3W1j;iLiin~zK%5$JuDGHIrX+AQj4?@iL};4s
z0#p;w1Pb+PrA*YjN*EZ7BIk7HWF83A0o?=Q=oRY}wFBb$LF5hyt}_@=Dg%R|sB&wp
zG#n+Bh7Y-whX1-61|SZ_aL@P%YEvCkZXt{~%^4Z=`D&AiQC#^0B%cYJ3Y~1?q4(F1
ziJ0x<0E&9Q7$%aPq2u<*q%qXCWv!CoL6V_QaD!1iK4nOQJsj}bJ2d4PH{+zg?`Aqh
zGM(zrG&q8F@_RWTqiTiWDeKfY*~rSF(Ns4HOp`2UN|x&CbmdQA9c2FM=&0>oy;4nG
zp1o4fpi|t&oCoEB>#>sW>=<8VDQWS&?*kFveH~R~O|Fu(p>Xrbr8D1bEKaJCX~P`J
zxWk|Et(QiOT|Jd&Tgf%6r)ux(EcLzh0?F}Ie~x>E5l2^71+%@?a8IRnt7ukNHFD@Y
zHyexL)uP%wO)@>(pXrM{SMBF$NjpH1tW@KO*ejdPdDPf*DbCjU5$&p^eXcj{aCA|u
zS&DhxpM$;z2RH~h-=>+&Ff{VbsY|ozp_5^IbO(Hg+Ru}6ToRKbak?1K9c4EMIK)?~
zmO5pn8;5o&FYDx5#uhp|AueENbRlZ{rIO>-F^)-7M7TGEZ7NnVAe!bafZ=@M=yMe;
zvgCJTii1;TPHF?=AYve;=pyqqFrk$6rkJEjtr6~t-#MtUVi{LSm1-WWYbks+t&XLf
z2RRsK<Rd8!L+s|u&}7uU_%}--Z<RtO8xFp}1Kl+xX{$)t-Lj@6BZo#V<PF~KlG|O9
zn|X$996MgaU!gIMno*0xXwfM-8u|dwj0V+cX?KsLdS5oxA9Fpr7I)|#*jHQ)w@5*-
zz?!6)wYal0=@d5w7aBReJRnKJdYEhi^!2A}^BR<+S`|;&3xnYgLZWR*g9^Sx)7OV&
z`Q;JLL;4+#lHatTidm>n8&P%VMbm`ZE4M{6geFo@XT%pI;xB|mK?8qz3}{-56kh5F
zs~OtppDt*92cW~RW%yr8_*W$SISKIR;ooV~e?G$-3LO``Yw9x$Xo3rv$ZsT(*Cdhm
zCJ?!ZiCoM?jKen<kyYSU4$%nXQ@F|IQfBkIWb=k(b6El#o>77;n2m9Lmy1o(v=Lmz
z_<xZ2Z%X{D6YzP`2(D#(<8$u9cg-bw?|2{6_=}|RmZWig0u7!Jf*YBJu|jsyNE!0M
zO-$h*lES|vg_{#7@Xjc>l__Mlgt1}LtKfFV|BuAqYy{0X!5s<syd)3qVtix2?WzLz
z$~(A+(KjC<jE1(A==Ua|^A;w!pV5sKstY}~v(and1I%GN$zfN?;lTtB598lyiufRN
z$gY8kycWY8i7NK{ZW4VjiT<Gk^jpv?iqJpI=$SQOtyz*d>@7L$BRPB|fx{E{cjEBT
z<~S^e$i4Ymu~6*<hatTcR6Xn~IgF7UKAyngN#^iL=8!p2<82Ufn;ixIg%7t*<kMIf
zeIgH#L|P@0&m<6eBU&T`pJO7%`rXwf>y`d|zlr=5BTtmb2T9~FBp`nY|4z;D<>ruA
zn#j*E@??p8h(!Lc1myq2$X{jT%#V`_4h!xdAU~~}@}(}T1-{N~4wY<<kZk@tfz4DN
z65nJthC!FB1q%KB<+b@vh<nSSkj+|AS5i!sJdTn)zLmhEJSS?8?=TNzzvJQ|0*Rqz
zDJZ42Z+Ep=gJ7yAB{fQ>OG?K`O5aVO^nLt0b&(%1rOZ(R7rEA8GlZuK<#G|6bz-;L
z{16MHQ8H5!nI(z*D1pe4+(tiPBE~K$t!~tDgsPyVnwc#L9WM#}G=b31@b6SJKW9S6
z=~i+sLfXNIVnTP!z{L*r0&JVD&io53n<{Fa<TYRN`eg#I1G%DJW?qJ3%f$=lcU=1=
zRmuVhzDR<<k^p`;ga4MnGsiV8I<x!CYgi&x#bU|k6v^iG1U7dvo8L1V!=%Pl6~tym
ze<R-u?w-}P`9{4|mXX}e+W7<XIaTs`x8(E31U_%#->J*|nfVxAm@YnKQYi+r)i=~=
zK2s9NO9FpQAaK@k5rMb-2}o=i99RF={R0c37Fi|<ESCiSX%l!BHfqL=FaL@N1lwXO
zT{dFaFo<?7OQ$lsWSOwV9MFv~J1&A){4>ko!ew|wQ~h842f(CVX-*%q?i|vx-P4Eq
zXzRq*Ri->cgRAcFB(4fP;Ruh2@_|;rl>SNg2K$l6WlM%*T>x!zky0fa`^31$-dVL)
z4@Zx!_x4Xh(s@aGQly-%OVVpaH*h-oTif;PS&#$L*ubT1d2z#n)RP#SH{YDP7@Hh+
zh}4ywBUxV*W1Xb@3vTjc4H}*Svet6ytT&xJ1+HwS@o}-FeYH33a0gt!Qp>?*ZxlEl
zZd=Czk<g1=uhXb8R!7d-cV*QDQTDr`l;e7TIkqJk^c&Ggi2a>RV(*d!hUcPRKArxE
z$p=k|`p*qgfZO~9*a_FK(cskXx+5_rIO!zhAYn|)ILEZtmXKQ)2@EWw!-?CaME6OF
zwp}2vnp#RW1I}v+f<tglB>@g`3Q%&Wm5h@C62*v?G_Z1Xp;*IzzZB_F|3(@OAs)(b
z#RA=o4Q4EG%MbW&k%Z%%%$)C&RK}9oKd&yy$g$9d(M0=UDGhubQX=8I6use2$HM7M
zPj8{R5}dBsBTL8?JcVsYjQ7#MiwRjY)o{;LF)4jU68oC_v!z0u?&~a*Zv)n%n;QBt
zpImya>1VT_mE69VT50-OL%1WTLyO|{5PLy%#wLcTn~OU~e3&+qTbC$)O!p;K=goG8
z(X$pVNLV60%Uxk_y(>5~XdNLu5byDctck?EGHBrh`PdXJSL>{n`s=*b`KA;gKhk-a
zz#9E<F9&+uddJZAXXVhVR+;a7M>6`p>?->4iQYkiYohB#!(ClSXaFo1Ec<8FVw-qZ
z=L0ixbm;WFB=mwL<TE3OTbDso)(68JFNukcUiWeFv|gm%YMexQWYU>$c2FBcbBluT
zy(r25vKjK*m#g_2+&0kFfCE-(|3KqzjFU_JWfH<T_4CZA3$n%7&I(-6Sd1p#UrAAZ
z>ph1Aubv%!7jY2Sqpc16aC_mBS3RJKu_w&Vqx02{0AH0<e;=coxU9WAy0$oUQ;j<@
zXHzxu=u}5QABlR%8<Ojr&Ge9M$jBaaOGfwD*HvNmo76Xq4~tJOo&9LI?PvuUlVa?K
z)%k;`EHANT&7Y+(e~(psqS?gm%0fck!W~=OPI56!%|=2R`i`cH&?O!ot#|(^<r_BA
zUVJ2VtnghLQT(22#vyV=)t136mW!k_+|`;b!4NK*XoruK-sk?p?Y^Q~=qf=J0yGC{
z^ft=kQa3rGY0h#sUTQYK#!D^2kS}Vy&7^!=`pdU7qri0Bxw5|k%?uwE*>DqF)t;1t
zq%w9Z&E(bu(!y+RB_-NkO5|h6y1$!X9`&HrA~Yo&c-EzWNBC2;^OLx<vOfHNTy)`;
z=5MqVepi3tgZp^R;{252akI4LGjtc3pHfFgjdZ9PzwIVj@8#VIgL|2@a}kKx2HVU<
zpq@jgsqgIfmOS_O<{6H{a`mCVr>pyDFHs*cj(;;USY+AcXN)8}!Jq8b<qBOnYY=uX
z5X~2fboK(#%%l_FWO8JwbfBbtkU#Af(uwPp?CcZzTSPdsPv|*xw%V#{vZQ-h6Ljg@
zl}T5B%a5*}BZ2PWlI~I7bOU9T?yOba%G{5po<pa(CD^6Q!-Z&hJzdg0Hbysb>9hs(
z;e2<sc($Cb_izm#^$t-s&co8**9o&S={QOA_!!Nk6;RL{o$|Yv*0n+(xX{3HGJIg=
zhB0$#ZRXL4YBrzqB-0Mb)aT<k3hbULaC2q7BCFIng=sHQon}s9S_X|P-0?^bUr`TO
zAPJumtJ<U<5d1AV=5z1CYs$+ba}O@w)#<T?>Zy|0=`mtSZ60m`*E-X6gK1q;z2nz4
z)%!Zlkz>31D^VT3TT(qEMm4bxRa7PCnt)~+N4Vb1HGw0ePFq*cohey^7N|)zRxmm8
z_7P85k-@L=Z&ERJ2t11*RKcP6XF5OWlz6M%S|?>EHx;u`<Je!kRa0Wy`9w>(BN@0y
z0>i&3XmAw%@)$f>M`JVa@3?&kj%EgiV=c^F4hC$K4iP-bRWL6uh0*Fgn8{>Hl1y2W
zIW~dJl4Bz>vzUy%fp(AqJu}>C4~m=u-U+Kyn8OtMBn9}F>SJyKg@sIE9#inBk6HmP
zDn}()f1y-`9Xlv-dn(}frIR_4$y6j6_?O7cPaw01$#gIo<4nojIAV>eOw)A|3z<Z}
zBmw^tiIWmY@S-?a%p~m3hO<%BVp%6}G83pu0`M;pI3<AqZzY1$n1FFM=&l7^{#siM
zoyWVGM@{m8e~HJ^1RfuWb_c<k%)?`__7-X@B@wGXEQH=pvqW7ofPaZWK7j!%f`bAx
zFf88PH4#(#2A!ClpjSy~_?Mu&5}>bR;ky~yu;_I|mn*fJ)T`>1-^2JYG2RKYPdofe
z_@xAV{&oaq#`ow1>7Rz6k9k-j_JT^pLwNq{KKKF0)eEW-m!O7E@=MTwXZa;qg@2}s
zjz_W&YJW@$b{Cuz6&xB-VceCR5dKcDaJ8AYV%G)dA<|>e1Tw@y14t#rs|zk*R)#UJ
zo0S#8PG@r=*NPR#J-8^74F{SFE@d{hhR!h^(XPuG-(4rBs>uo(7hIl6j3dPbS1~br
zGw2|O+8~T%mNYk5QQ(3wlL*Iv3$9}#9=#|#q+4)3mPftWinkuzkjaWex&=2et3xfV
zm7{u`LEiKxyM;Mfan*xcGdXdHw%`ut<Xcmz5!Zsdn3WYPJ-9oQ6^B;~-p{OTb*7`H
zup`+)x{^AT6)r8fKa&&(OA8)kQigeydlE<|^e`4h{mBY?9(*v95XU_W9%VuvZD$jb
zk<j!W^%%3Vf|vy#$z=5ezCp|Bk1{Kd&y<MdWQ0D~8GW1?Sy9P?Ph>LUkYvH92E)jV
zxD<Sb8CfC5g3o3$;`n00Q_RStD~USN!<g$;^hG9Qg$xV6lu3x=g$2(Hf{;~1UtvO4
z0I=ZyWD@!szCpd{>rBXFr73Di#skwU=Nrt$is}`7Gm{I4^a{Q`FfLZTJjYzDKwZIi
zGr4>p-=KQ=!N9m2Y)*(jWG+^~@Zd+8Tz<k_USKXBdo$5SDj1jEOfNDWD=>KQvrIZK
zF`Zv99m62ZJwFRNW^gT?(yy436#zVV*+nTb25SUsU>O;lgUNZAya$u_Vsa5Cmtb-k
zCRbo`6(%7j*JAQMOm4tr9VRzpaw{gcV{#`X5stAepNyWw)H~>jlqsf<ldwl!RkRf<
z>*UHk9PKcUk0Oij%XVxOSzH>ot^zwGVO6c-1f47QG?%%RJ_klT!vp6fF=87v4BCY2
zLpUfTfs9XFQ`xm<4L4Z|=9To}y+CrkNbJADQS$+7X`%{mDs!KZ3|;~?a|zDhv5LTT
zYK^3BtYMnTtutp;>BXv&apS06-)L#L*iAglC4)w8;Z(<lb|X8FPH*H`@al+Eq?R{%
zxG7YtxF*N3sbt@HcPA9BZ&=B>`f9WVMNX%kwM8tnn<LK#j_t-4y467ichgj$5bz7M
z<9bZC$t{ec8G*RoW1ru5Py{v06BuYiziG{_hX8V#@B5&lrg?>GgQJ=CAE^yDL7M+A
zNVOocDqmiyl+UC{bf@-!cD7hW2N`%jhwtZvzi+OqQ3Wk7^d}jepfz{wT8;%cr%<jH
z+ge=yxjGOY)eufcqjx#LK(IS;Itd_WH!y4s#N(?#ggFS2*GJzHv|U{RA@2Gh{9mQF
zANn%ohd?<H<)&U<o2&Fe0$jT=P98dsB``dqtp$zn`*k|Pnw|<2zd;O{T%@ly<eS`f
zkjRgr7P;P79L&9v_&Re>oSYliR{_c3t?_LwiH@{-+4QkN08wFB58!HxHHMExh+YVf
zHK+ruH7FWmm>86oLhwX~ldO*6Y<J27L#hB5Q9`^zu%?1oi&|4JsWzacNogcck28zJ
zmsPVs^P+34$yRnC-Puhq7QedGPD?=e2@^QF;)Dqzb|cgUcT8vb9hE*{o$jow13Igw
zt4Q^XK@h1uqM;BWw&hwx?_RKE_Ou*E;>7Pfn4?8Yp<I8dzX*pRMG#&KU#XSb1yeyE
zwsW88=!nx(>8)U7M;YW1|I}Y8mpa#qNRbeM62g4KAB97U{38(=S86M^wdP9HY1Za?
zupRJC=%upHp!p_rNt_&T@Q6U5tvOu$Rv>01m10c_r_7(<F4lRG{6_zfEt%+5DnCgE
z!V=&lJW!!Ih1YTpy_N}v@VIr`T2eejWqeDEKbAgs;mJASQJT{{N?qks#8#MW@C+X6
z0EZc5rizdSj00;*{M-0wn#5_tW+@ku@jyMnW3Wm$jM$Zd{r<^IDm_FFEyau-m0v!6
zIi`&5t<XB!a#8BNKL&(Ny*EeM3EPs9xyiJJhFtF&TIps>(4}CL5GyQ?j}RQ8Cp<Wr
z0(G_&NsiX_v;!)}DUKD_$449v@e__o=AkWXWK$R>1fCgn%HSjsnUYz-fI>ZpSYlV=
z$UtFhGJ1T0B-tD1&&}o6h2U;v1rqWTfujnA<B}=hnQJWCn4yO7&#P0{Y~!pDL_VG7
zh)6e*)@ig2MRT3T$f%R%`nIALX&%eC^wD>OL3|eCP)6Z?$#uuu7)jkYp7YA46W0x9
zte8g{ccmjp?DiZ8DIA|n!rJag35=tmCNk@Cn2}wr$Vtsf!r_!sTS0|SqK}tvE2wN5
zd(FOXH|Yx@qb`L``?|zN;8;v)jetudw&|&WmVyNEY=l^5w-{F7rcetFnMj_4IAxL7
zHhs;LH!{1)TM?g{Yy*x1m9~^{?JK1$o=XXL37SCR@W`>>3LK@429G(3qe+E(Cl5{9
zY)bxyrATHLo$#ixDV3{CE8<k>c27*~N*stPi*#fg+XQJdR#m8$omQ->=Bk+kSf#Dq
z68G7%>OI$PDaGi`p%lkf+^cXjD-6`8KI7BIA+f@V$$i4HyO)xgF@K^xsIA*?a~7b<
zq<Q&Fl@qOuq1EId=g3;&5y|4DZ_cG;hBwc^%hH8ZmMvCPt|o=#SY2r|nLExBeNsGU
zGIt4@Kq@A_72K=2A;uBE!o#xb*&X4F#5&e*hZMEGyCh9K?dxyF3Tv{xID%L>Jy}4n
zh+!n>JU=<T3e?0^t!YIdYpyjpELpg3vapdBNRBnbzSt+1Mq6toD|ngL;)Vm6rOoe&
zDyHV<_mpBZfmAzK;m?|D2absr9-3UEDH<**+DyOBmRgW595{ndjhw-)aB9siCOFDk
zctEm*y82j(WZ1d*=G0`+dYl!Ht;u<TqqL<7pew4I6Tr-=2@vX=ODm$A*2&T~m(Rl4
zrXJ=k9GhH``Fww+R);W%rFuRuL<qJYY%Dl^i5XcUG=fi~al%uWJdMd$G5IDY&tdXB
zCNE&}5+=XK<W)?5kIA1f`5PwxgURqeL9!VpEtqVB$!JV=!DJ6ia+r+4WCA8_m>djA
z#K*9V*fct|59kSJ@?gb7kO+h2alvNz!sd_^$zvhd0{@%r1Fn|G><_wa8Q706*b=kQ
zH^Q)Y<E8=oli1QM!D%npn&GdL@Ha^F@NMvy$AG_e$iE%_ooqk0XC5awrw<no86=CY
zq|!oW>fsTzE*y+zdN)dXH%oduCeYiN>FqLv^gvs_(h>J`H>P)sq<6cd_pStbdoaB{
znVx&!;Htf@3b?b^`YU~)Lr`~fnU+>6PVF6%)7_HO-U*!cVNUxpC;y%WoNApFJj+eu
zagXG2pX9NB0*?cj$5`g!){7+64fxE77=}qq?w3qph>;7#_yi_wHx#ro6Zdl3$)r#(
z)QWWq(<lv!^gi8&MY=44f=Lk_G3^->i!!;I5>W~c#wY0saVR`XPl&_t&jb)Rj|%3(
z%w?Pwv7J#cEh;#4@Aq7?#to2kQIBFKmJLxb9e=^uMb7Z^fL0@#Q5(U0n<bkT1xGVW
zx6;qWGO{_+xy@v5mhn(<tjW!22DTjvW<@-L+4v-z0TF_^T(sl3uEQ3MkL!XHq9Q~8
z;isSD>Eg|24%T^G(2l=oRlg9c-slp192YELM$^5gc=zd+&g~@TW*xHyi!!<KDO<3F
zxw*H5u8u$p*ldEQFhT3!EI2ijAfKBBr!ztKxr~dTaiFHx*;1xt9g78LWK!ZYvEVGG
zWCRjQT{^f9%XDVTn3;9j6%;a=@ljXM$;{k#m#)gP&%kt6MP_9kV+GxrtoRfw=wVj=
zO_+2_6$DJlI;IL%W>Vs_si2oBxpz3Oiqa0P^g60AA?xfZ=+7j?2Tnni3As;CU4&vM
zQ97+U)3Oedf<`7SK0OLnGp#1-smb%9U~MKLJ`)PgWkUY-G_doY;QUNleA*LSFhE+(
zpX3A=X42x*o8V%m<^Sc$Jd_D8#ZqZ$WSz<cmt}I}qnO|d=H^yuxq7@jqtW~MRZPY@
zZwao>B*RB6!8HRUV;-ag*D)FEASHNTCK*0G32tCA{@-N3^N?U26SU4lf}1i4^6^J-
z3llVNn~2UdTunFdV~yapOlsvh&?9L(eFsyU>%G>BwA_QcE#$6DmV6cw+`}yW$7SXb
zL-2kqmByuY#t__>$&C*df(Mw}?51mQ5a$HJgP9!pY#?};InMXy*sP8}cmzSHIxQuC
z@S#kqtl1Af%v9Y*U9PdBRQ~!*r;fxd<$dsxOg^l-51wE?{*#NJGCuei7EKjqY2$;B
zXA)!;eDEX_bRR3aD$G}<9(;-^S_<{x)0q@mhaP;EDf(+xOy#<}T`d)P@cB$mtQHTx
zFaS=8TJPXXSQ<5@rSlHHoXqK2*tYA6?%<h-Q!sjnbl9}bRvM1Qu<H&^#^h8?PRC>^
zCTC)j$E1KsCniNqR$x-XWF;oOm{c%18<QF)4NO*JvKEtbF*zTT3oyA5lZ!FA6qCy#
zF}D3GK+)GC=|g|rB-=El#3}l3ZW}vJ(KpPtF?24IBcsCL*@##0EqszXm+!!{RDC^%
zf2PLol&ZeWa8TdpGJk+idiQ>wOS5A<{Sbe7jFjWS)pk9q*h)W+$`1WUTXKk{7nqJ6
zV(F(59jQ&b1}K>-9TjZpXUxgJZ6hy~w3tgTF*`fv(l0XEspw0;Vs`#&fV3b?zh+i;
z5T;i$S*dVLzhzebiaAtHL?KM|PW2iyvZFG+p2<kXXZk%e^4CjeglYN%6SKoK{V|i6
z3fS}}6Z6-Y3?OdPUot7F2u^=vN`_~Ds?sYS$m#D`3Qa9`Ag6z1a#10j{)f5vYb6pQ
zJN=sp*^!<8<07O!J=f#xe<LP0VR8#5w_$PzCU;?S4<yF7K?Q{xj;EP)`sm;DGGaq*
zibYb@+OeTV;4e7OI!;7XY^cp5O2Ov%BtHwI;8}haTJX;V;D`<Nep}YH{Jq&aDmXOn
zZA{xO4k|>{wg{-?4ekeS=0yG;CbvD4bGv)HR+}l2qDFHa*^#1l%p|2EMeWR_+$JS1
zQnbDtM8K$BGx@22QSV}Y&FyI*E`WF3s68_IsJKykF(3aPhFrje9NYlGm$EA_!XueS
zo;fU*I=me|YM)GUDty#_OwM0fLgc`(xmd+}0c4b9{rfkDso7Da4#=dYqDYNnYE9N#
zdN8R8OwJA_bzmkr6-;U(lk?Y+;KtdGm}Iu8+C<wir6y%^Q!%9`GdKT3-gI6EV~Nz)
z?9ft&Wb#s>r4D0W{tasLazvOqg2~wtrlw|+QxT?)WODxBJakCy%_jz$n$E=RKvOd^
ziK#$S#|!~6D{9cOOw5iqbzCMf6>n-b6Z7w8qQ;WKPR(UrcG#)oGkK}7Qzs0bmsMZ$
znU@`Xsy&mJiaxb~dHFA!MSVGfP@Tk->>yN&GAXGbR7(a$$*QGOn35ff>eNh1Di+o0
zgQ8@Gq*}_9?2uGvWKvQgsm@|b{u@@&aPG)d%b1lNnW~V<N=2sXWLEy?{DM_-fT|)h
zvjbFhyO^o{#)G({co>sMFnJV{$1r&ulaFHZaZH}X<Wrb@29wWW@)Rat#N^AEJcG$s
zF!?GbUxy^Z5!;oo9*v5;aJPl?7S5hOuVZ$8=AyYLx6kfal5d~6aN)d;xgvIZ@J-B4
zyBH&GRZ2E(4bve$cOR50rjHA2vJSaWzjpXA)FqC*7!c@Shn~n1bUXZ;ZYi_{{(-)o
zpOl4tKC=E4@;oaPlUs6OTj3KGwo`&)@^h-Btc#Wl!95F<f#*|%6}HqMzBXKbR9yiV
zRFeSDN0k-qqG5gyULV&%S3P%h)Yn}oqjyDDM}AS#_|nMa`<}t#>kTL|oH7YwQWPga
zEPS}!dD7}ib>*b_?Q_bN)zLq-YNxdJnuNUC`r+ozy8%DA2a`LX#yf(0;W6As9IEhJ
zjo?0f;}`Hoa6deC1P|bUe*h9_qo2MG&m<x}{cI$>MNh|XhWA&G3m$;#8GqqAY>SZ{
zK^xwu_H)dWR_JwSM1|556*>~@O%xhDir7<w55qrK2pwYZ5#)3&a(o<~uB1AJS5)%_
zzd+Rgf*-+&`3d;vipKKbVffb~_p2;&FSEq-MC5*re~Xj5(n;=$;6{M$Ap6OVG%BxT
zQTYv1p(mp9+x0@_)hsH{VtjDw2lPZ#UK>1>N8-Kb^(-pixh<kXPef(P?c6})f+ypJ
zzL8bv54h0h@rerkL-X<lKf>H-IjdUULFbQIbbi8gUSK+Z8UURaz39A|Md#;C=Ow1|
z=K;|9r5ByQWYIah7WIIana*DaK<5=NI)BTe^II&2#>T5m=dA(IdEJZ7-?Qlap6R^7
zbp9~_I)C({^Uwdw-hIbMQ9SVfrwgGY(nL6lG()H&2-15Ay_b-3X{3<?gx;kIQl*I?
zO?njpQ9+7GQA8;Mq99EW1nCI!o7Y@!CQ8`hme2S5`#pXS{bPJ`Gqd-;*}cu(-tMd^
z&P^5PmWuPxMw~m(6z7pC&OH_9zKZj=jW~ZjQ=G@9IFD4Ezg3)6SH<?^*<YLRRO|qa
z`<>J*+3GLz$^`eOoQ4I;-C!CUane0goV2Dm8CZ;bXT6}}q_Ytx(=)|MZ;F#y#mS=L
zWUvt@+cU*^!4xNlijz~t$!H_ai_a7%lPS(iDo!31$IC{Xe9sgovnfsi73XCYCyR|Z
zuRK$ntfn}HRUC(klg&n)qR$j3yD3g_73WnICx?wVC7&ryPE(xHDoz;{Czp*l-p>^0
zMN=Ff6{nnvliNm|3QxpwCjWBtwC@jm{|&#&pAPpInG`-Jzhrt5dji}$Pbs-CpB*S3
zbANS9;e%;D(?SoZg&st5q4{l$%(wKGSC}0_?$?<@qynZ$hgGESRiu|~L^`4(9W_KM
zXo_@PMfy=idc{VhlPc0FL!?5cNT*e#pH-y7HX@x-k<J<-IZTnxsYvHlq#`yVT~v`S
z86p)mMY^maT~U#W*@$#iMY?8)RNNHlx{7o|MS9gnq?;<zEkmRdrbyB5TYHAzQISg8
zh;&ayx^IY7$`t7@79c+<K2(uP+lVBu4VL?*zYUSfm?EWOUn?V}MsipAnvF<l=`ACr
zLl(PAZ&RcUD$)xoQdt|3GO0*jhDbi9NLf^*tSVAD8<FI1zjBna8zPlAMe5zbz2b7J
zNEK{EdXe5TQf@<}il#_;Sb*G?@~TLcY(&zdl;04kvMJKbDpEle>GkJ|RLBsiiYbyq
zMJl2qz42U;iWwqRHAQ+=MJk~pRkIPP6usrPRN4^fO;e=TR3vW|sk)6wJ}Oc<L!=s}
zNEKA1iYn4uHX`Z0cV$DQnx;rqRHQdlq*^v2Rin3Dac>$T)iy<{p(4GdBGs`Gsg{aV
z+YqU)DN<b(sh)~d&qkyMDpEs3r23{vja8&3DpCU*k(#MU%?*(nnj-nCNG(;QMs_0c
zBaGaZS{ouYHbrWyBDGVIn%Ib>-*Fubk(!z!byAT!t4PgkMACobx*8%iH%02MBK1&_
zTG)uBN2#YFlCLRJfQl5TBDJ&;Nsp4#5UG_ZQizHasv@<v5h<MBa`WzGh}6auDN;p>
zQjyx)h!mqD^)^ImXNnZ3BE_pn?QKNTE3S_rQU_C{ekxL;iqz3YB)xeLFhuHPid2mg
z2;>LdK`K&b8<B?4TW(834UxK-BGponhO0<jZA2QWB8@Uc>Sl`cmWuSYiqzdkq_HZ}
zI76f!rbrW5fLw9!s7QV`BI(V0k|9!0Q>4i%(i9cR|G6SfOCD*>lcyO3m?EuIr%}vM
zkzPpo%$KJf93GucoILf9Ghc#u)_<P);yTRSsZ^#^sl4htxu(J4?+A?WkBf7(hz@c_
z=-US-#Bq-)Zc!5?CgpN92@)5}i<{^~$BNtQa4Q$rquf}_KP)Ov+ykkPxW`U}x@C>L
zgG{7ToX{@rCB<Eg0{x@JjjUYvh!QuK5*>rXLK0%hAg8tCu05{#?V`kOar~lV<wRiO
zW<X-5MS0s7e?%@ZvyQlqRorvQ9g<rRI>gOi=*W#?{KZ|Wobj<Dmovz9#~5CoxS5hO
zuvbj9ILlt%aK?4YUYxjNk-QmTSbuScpy((cN0VR&w<+>=2Xpidi-;h4(h?oP;uctr
zxIi%&fuZ6Sajwwui7`$Q_8HR3mLjF>=e~Ch_XZ*`FNx*L*zV^bQdD$2_ulFg7UT>f
zntp{z3v#CvXT%4IRIc&$ixz7oR`iK`wroMpV1IE_sHC+cR%$$hEbj*9UXn~E?wS;(
z11GJ)e-BJj1i4n8Wi`51U#z;76x(02m+4xZ7%-VYyi9pVrJ%52c_%cnPFXGVd2%Zf
z<Kez#RIFc8o@ZE7hCphOGLY+DZ60L`^cS0?Ur<==v#lJtnTw~9@5``QXP|g9CUOHS
z@g_<(PV(i)$xK1feWN0x{lzBZ8lBLDkPxvxgZ%@YK7r9{!;TXRk{b<IQv~~q=}6T}
z6Un#-mt2~^q*J_)!=i%10>#@!Y^=%dpCx*W4U_(^8=-MiD@UmS|G-}2`2!NkoitM7
zE@@Q8IU>RWV*O(iOUrj}bi8^u#X9?h#c|uGXL?7)#W(}Qg2Mun-U(upmwjDp#8E0~
z&*LwjTs&V?V5r<xcst_#Ly{!+9)Z!3kzw)T_HN=vT&~SUJbO#`&H0$`RW88SDQ=G!
zCqnbmqa0kn9Vd5}yeH+gMXng-*ma6GkX%@7V5ncvv+p8YO+|iIp5CS%2~j~#vHumj
z1OEh38|GTeF=7wSPC;%L@=4tL0$N3i^%UqIA#Tqn`ifhiC2b1ojZ!ksk+dfiN!b8m
z{UiP4NzGyviOZx&;NeCfCf?}fh!pd`ByAv3?t6ZP#jzig!|05PP6%Q5$y175a#c^B
zM23=`dMEgczXPsa_doB3CKgRIv24G}0b;!-dkx99*?$eNxOpLs-8*%CF|z0`o=+xk
z$1cXdUsz;9q{DTkGJkW_c(T658M(YYqNBv*r$PK#aQ_XXuY5>s+I>UA#D=K)IpRX2
z6C#3=?hdDx;9we^F!}EEXDdqDh};WoA^);b6BaLa2>d;zg^yPZ^8-bNJX-+$c26E6
zX%RYsGdfQEQI^Yn%5aKWu~!j)jYU(}PCyKsxSCBoEXIgh`BXk$VIfgspW_!0<{#%5
zkiagSzffvd6CBo0uFd-OTZyeaE<ybLl8Y6$yrWAwE`4^m`;`^XA+DMW3le{n{UiLv
zZ8OCPc=rs85_e4$H)V_#_la};Y4`fqe?C5Pg%&9y-nr4dja~aGF&|b;Vo-FTqeziy
z@_gf`ed>}w$>-sNmizaY6n>HqHhugHp@IC6EG2&u3gt12pX4J=3oYN%z0fF{%7sSj
zg+9v(2b{3%n$25&Ws0G<`*o)9lYDPeq_Q2|kz!S(xaW!#Z-|s&iqwY&a_IXaIZFMI
zhbP)4(npR`e`GOA1E{w;+3pRs;s(-GMjC`XjHJIZ4K_p?Vv1B;MH;Fi4YLtxIDO<O
zjX)NoG?IF&Q7WY(jiRX>rP0X4QF@y`GSV1i5os*-R*}ZhKyFLpk&H9}c^K&(`p8HV
zkwv6Q)LTUgP^0uNO=YCX$iql_Tbg2sG}RPo8V%$qO-C})4CG;?_vj->X(qB5rCHQl
zjnZry$VhXLj5HT{7)g)P`}DSmG>>|#NY&Mf`#?pS|6Gw47$SXWinNdha+DS!IZBI>
zhoiKFK61q^MHZv<5%pH1w2THa(sCpttw0_|T1g)nX%(`Fw3>RWNNZ>yBdtX;(mLc}
zr1kWXkv1TUNE@lQilk4@`<SLO(kAq@NMoM3u1bCq_=G+((q?24=?m(uBI%R!w$oHb
z+JQWb^d)^{q@Bnj(k|+)BI%R!cGFZw+Jiid^c8(%q`k-@($~~mMband?W3uTv>$mG
z=>UCXq=U#J(l^vwMbandeM?gr=@9ZT(s%Tckq#q^NZ(U$6-l3*_XAC3q$9|~NJr@-
zBOOB)k&aVu6-l3*_ajYZq!Y-)NGItdBb`DPk$$4yDv~}q?=($iq@R(8k$$0%jC2NB
zL^?~oRV00K-mf&3k<K9xBkA9n&Kn|KFhxqGejB)`B3-f(={NewQM!yQM(GOmR->d(
z&ikFFa+Izj4@c=5ePpCRkVT~H)LTW$qDJWkO=YA%k%y7==cSv5NViOp^vQX*RiryM
zBHg8r9Ho26VwCPvZ#7DJ)F?fmsT`%hknK@=Xo&R46iJ_)_qU4l__-pbx@J8}sj0Ub
zC5IZNG&Gf?lor_@rF4c!=}nRJ$$1%6q!(;N%19r%EoDL$+maXcR->d(&dW?wIZ9cO
zhoh90J~C1^WDzMl^;VJe$$2?wDkJ4Y9!Ao8?_7pRFPb7%QQJ~(73n1#k@C<-j#6G^
zF-rNUw;CmVa$bI#%26tSJRGH$=_4Z*L>7@=q24NzJ~^)tO=YCQ$iqku`p8H{kVT}T
z)LTW;C+8KTsf<({c^K(c`p8HnkVT}D)LTW;C+C%-sf<(_c^FCmjVoh_^qMJ>J~_`@
zMJj6}l77ee7$TK3Mbandl~<7}*odS@siGlLB~v7Qa$aQ>>2({C^e9y^M0&#%NuQio
zRYj_1Ba$AaHw}@hn<9m%U1bdw=`9<PYSKr(<7y#`cU*1it=@6^<h(jGmG8K^$isJB
zJ^IK<^^rxS2Gm<c(kJINq^XS52zeMuueipBNKH(U^vQWmRitJ%BI(V$xgk;uQzU(I
zp0A43(nh3K^pPvBHL_T7ZK$_ear)%EwltM1t{w7l#kHr8jMM>HMCwSrRV00KUMHH$
zNS%>~k-E@FM(T<zB6XwQDv~}quRBd;q#nq_NP6@3GeqiXilk4@^H-4qo-0yd@<=J1
zoEKz@q)*Ossz{lvPR@Js@adB$=Y_DS-M-1ErTXvWy#G$llV>A5{WQn_PR>hy(&v9C
z=W*JgI%DR)lk*(nzyD6o6DMn0pP~8R$$9^socI5elk=vhO(iD(6#tpe(hEy|@H~aH
z^ukP^1FNWuG^FIG<#jyf`m~(l1d|b_g*I#|7BE444h*lsqU1s&>GZUZ;A8*w+q`R9
z(eRq|mfs3$A&YMXQKm?BS%8dGPeqEh5vhTS)X)$q#uTZsiqu3!>TM&E{-V;{5GmFa
z$yY^csUpSMh}4?ia>cbVM2a^>YNsN#SCJBIMCzy_buvWiV~W&8Me3>|^|cYHyNcAq
z5UHOjQco4hUqwo^5h+kb3Nl3MZ;BMGB88|(18hVJQ<1_Akp`L~MW{%TD$*buk)l<k
z7(=ANrbw|WQk;r3#6~2&6G$*b8fuF4hPvRguZlFxMx;b~%bh@fL!{xRNCR1b+<OmF
zkw(~vq(^CpA<{@wq+u%3a208kjYuQuEk|jTA<}45qz!f4qx80l^tO#iV^ySahDc*f
zktVPJx#HeYk;dAHG)YBz*AQu(Dbf@bX{w4e-bN&Sta-X2(gaha_f({rD$+a86=}91
z(nM3Fxhm58D$*nykv^ce+`Q)-BE4&h^r4EhP(_+-Ba+?;EH*@%Vv5v>TLj3>d#Q>v
z)kdUc^p>NvJb9!P4(v`dMbZa$SE)#utq$xy`Puu)1G_VL))WuyE==L7>-w%M1fTV~
z>;Ls-*GU&B$_uUh9qv2HxvpLmSFt5kdwLnR>ry!PvR&6;CR<w4)wk~J*Pec~({+3=
z#icp8jLv;DR9yL$;1m~$Ig0B0S``&f7#5dwV=zB=77ve>xfUm{fcHuH85{aX#EDDo
zyz~?MCBLNC{2HyU=+z6A5yX|G;u3msy)ak8xvoc7m+iXOl&2S5yV@tXu8wm&eZU>1
z`h@wrE-!U$vHHSY@#^Ju>wdvu5y>xfeEND#{+h)ldI{oUVfTgaybgz4P$}_r&VJ&W
zZn0D@$Ca1&xvr>ujwj-NnhA0#a!6fICAQF{$Hk?SV!IMo3&un^`-R0PI>beq{|>?x
zp{^G!Z|*Cvt9M@*Dld$VcZP_oo|9c5Eidp)x?EITSr{N{xm?u!nw|apBgON{OYFIT
zI#^tZ9OMjeU8v1v(Y+JG#0AviqQ;~iu4@*>CBR~fa9#83u2I{Gdjm<j4pm&$C|+cI
zC|BRg0AfIS9rE^~NiRcR-s;J#NH(DVgV&L4Y3{e|GrSV<E_3=va;32c?{8<6e}LFt
zbzd%;7sC<|5gph|Tr=C>`78s%RnpN>V$uuOb{yyIo#2cTSDYqY73?1@X1{P>3h%lC
zRqjWgw!Y-gKYE>sYmsG~IM?OHV(pPjZXoU@sCS>Z(pX%=8W|Js!5GMeu#LNV$_0x(
z0$0!L_lVfY#PcSI3zhxj9Bo7WW1Wp!wG&U9yrEybQ*6TGrkeUX=cm8@-EE)#bYki3
z-(B;Ih&`;h&OItHB7r$9;>7Ka99)2$H109};$rn@nPY@Ep?jMOPKXu%jP+J8t0z-i
z^o<o4P`h^DJ-xkpCfnh-?&~CW3xQ(F5&sZpPj5$0G2KGaCBs}}E#6gr+<i%m9F^h-
zJ6o4uNNjXMj2N7L@veuRk<orhZxgOccMUrooc+Y`MY(ow$tt{h_HsI7{366OCIMm_
z@5NI5qk?#LR<*cGXl#PK<hy6mkOhZ1#RcTS;`lb>!{0=)N0D!S{(Ai54`!TPl2{NE
zia0`@5i#N#eKD;;h!|ydK~C|fMo;15`upde3qf9A?FbdSJ!T+qKSk0mkDaKP?L|z6
z;9fW6aKBSr3lghL+^$gU>%~o}5+cQ@CXedQlx#JJMTtnvV<29Xe?W9Xyw^YD=~e0f
zkC!5b<)8okPcHZl=QE!B5NisDl4qJedCsDN{4gvfkNVE$F^faVAD9-pu8n)4^Jyv<
zx<D`VnJ@VMkUsJg_Cn--ohe-Ky@-0N6HFG<K)%i;NJd(UJUqJn5q)H&Wym7Za_X%j
zt)PL7v=YfktB{A0^mp3T^tOn!hI*?=YiS@OtwS=>dSpA&26|gW+DN@sq)L}X0~zUK
z6={==NT1M0uDH#}BGMM>ts?1PO19Eej?$;-X`}SyX*g4TMe}X+k&!+_7Lj&QZxu=Z
zQnH(-GSVL8VWh9<BO~oa7LmTD-YSy*rDPvXWu*Pc!$|s@{{cg!gQiIOmy&N(q;G9R
zIz%5iO5Y)iQ94Y$)hOv-O1`J59Hk$Shof|aJ~Gl#WD)5Y^;VJeFD1umDkJ@fJdAXL
zJ~GltWD)5U^;VJeFC{<GR7N_DJdE@+ePpCxkVT|3)LTW;zm%M%sf_e1@-Wgl`p8J<
zkwv5n)LTW;zm#00sf=_9c^K(8`p8I^kwv5{)LTW;zm)t=QyJ+h@-UL#30yNo`ok1S
z|59>YMY>@l(x3E^qjVEljM6RYtwu@zQgWN7a+K~M+oN>X5b2&NlK!RSzKZm~Mx?*!
zBS+~WvKXaD)LV^`{-xw^n#xgnj658rRDXy*GE!<}5h)GzR+02CC246YBc($gMoLc~
z87Tv@i1Y&WR+02CB^hZdBV|G!M$(@uy$q2un<D97O0uX(S)VIXHbbQBrbzmik{l{h
zP8*SO(MPVh7m>w^%T2x2iqpT8yhKyE;_@I5SDfBe<~2mhXNshMDao%Q6|fQMW%|fb
zDu~>pl)^71uTXC_O8S?QLNt}5l-25&k|#fVKlw|EgGK%KONl-#_1`Zg4sp~-9Mj@Z
zZpu^q%A?p~#Qrt+>3_eJ{P#=Af4`KdNvh@P%_;rnqkrx2;8%kGekl<@o$=G#f4`K(
zIsW^l<iB4^JeV4HPV%pxDg07W)bz=-_h9iMQ67kvlAk<_(aH7sJjIjlOPLnhOD(iC
zP31z%=!HJ>FD0+hM}ETgM($78Dg07WmU^pC*oi8V4^3sHa_DK_L!LYrNgmzS-$Tk9
zB2_R&8muB!RFNv#h*X(Aa+F?27Nb;!daF?ysUp2WQ#ndik?m2cW{C8rDbiRKsk(|(
z<GCWeWr$SM6ltQ0R7*vwZ6i`0`p6Ym7g?;hdemF3xM?a<eVWP@*8q9A;u_LNMrwpC
zA~mMoD$;BfsR>PGq^8KjNX_UYBQ-}Bky=o16=}YT<V#Z-sU`9>k~&ASl_64VQ>4Wz
zQX3Vit&K?S=p#p|J+c_34%AzX(h3!+BTeNfbwVDFQfK<eNL`Rcq^{IkMOvpKb)%__
z)E#*ksRw;zBtK*ksVDVTkv>t8{Ans91t1S21=2@G3PKi<oYY%I`b<R%rm2h+f;@~A
zN*@_13|T}9r`{^kP8F#aO=YAA<YA;p`p8I8$RbiS^;VI-R*_<8DkJqq9!83#kBk(D
zEF#5IZx!iV6)AzHGEyJpVWhtFk&*f#i%5yoTSYpeBK4=Kj5Gjw7-=AVWTZjJBGO>$
zts<RNk%rJzMjDDdj5LfsGSYBl5orYVR*}xANF!+~BaK2HMjA~Y8R>156v;JV$%IZ`
zb%Ok3<Vn0yah#?lPJ;7kmvpYGI<x(sr(ubI{ZG!sDxNG(cnaSy#!zoHLStzlzh6kn
z5gNy1t`SP{`^6;FLiP8HcWEjYI$1CD$?q3W9xj(3=BLm{4*OJOu_er*-s*Mg?-z4v
zDkHs*p0*`CC6fMrG0za`15+gZ{bIg~w7^EB59uREX(6&0rA5?RjgtO;v6!ZEl$Idd
zqqNiz=_6Al{rzH@inRQ>BCRk)T4{=;zhA6UkyhJ?w1z%%#jQmaD{dY2Rx3__zgSOG
zx#Bh;4_Dkq`p8HhBa29zsJDuwzh8VpQyFP9@-WgC`p8IIkwv6Wske%xzh7*lsf_d)
z@-UL#34Csd^o1#s{(iAtMcQE_(wFp+qqGxQjM6UZtwu?Izt~MvIZAtwhokfrePpD)
z$Rg6$)LTW;-!JyjR7Tp5JdAXJJ~GllWD)5b>a8N_?-$?FR7N_4JdE@mePpD=$Rg7B
z)LTW;-!Fcksf=_4c^K&^ePpC#$Rg5l>a8N_?-xJPR7N_1JdAXbJ~GlNWD)5n>a8N_
z?-!?ODkJ@jJdE@UePpCF$Rg5N>a8N_?-#$)R7N_7JdAXnJ~Gk;WD)5i^;VJe_lrw3
zm63i!9!9!M9~tQivWWCM^;VJe_lv7Em65I?4<r3S9~tR7vWRqpdaFqK`^BF$m62{D
z4<p^8kBoF1B}Gd5e$mEDeTk44ro=k?`o{*jE>aa|3OsR%$N%P=#rR}FQu=0bhkC2A
zxl052%|c3!%{?A-jZKQ*EFPK``e?j+p^s=P7y7qe=#$?pp7JxnWBSMePjy|$VjD<J
zz18d7S6eiYuQLrzWu&y|Y1_b4BIyqm=?sz5n<8bPfsFJ5l94hZ+mSNS+hUZwsJ9xW
zs_L6~W)&%mjYwJPBUfBDWDzMl^;VJ2)e;TlDCMB39HpGd!%@mb9~tRIWDzMh^;VHS
zTr3*MNH5V;M#_UcjFgu?GEzQd5h*|QR*`0o5)EXe0yLG8UPc~9Do7t0=@n!VsSx#6
zkqS-_4P>OkG?kGY$iqlQ=p!Q)MHZ2YQEwILwAu+2r>TteD)KN=3Hr!LC6Ps>Qq)^T
zDoq19N@b9Y^cwOok~e*1q_W5&k`MKkk<^MSM^hQ8Jn}G71^UQH6_G`xO4M6Ls!Rhp
zO0OdssS5Hi(i`-Vk*Xq#NY$vfiu5K8WTfgyMyi24jHLgdyhU$|NHwXqigZeC-nCSu
z+BPE9p^qG;y2v6@J?gC@y`y%O^=T?csR8nElp4}UMrwpCA~mMoD$)rxN=;}gBQ-@H
zMruYM8L2r+ij*`xcAK=RQhC*P28&Cg#mS<95#rA0js~t-9in6P2gL+=Lw5H?->zms
z&cJ9f(?ztnf4aWBTwKN;FV5`}SC%`)HQ?e(e)omS;{Msscxk!6d-@2mWNtYwZb2<>
zp&r#sOh+J2xRclPH)KwSFivd^NQmdmyrje9Z;Jc7yPJD?iB3V$k$!QZ;&#|_E)6-i
zue?8cUr`m{p5H(<bc#jEX$rj6wcT$yDtLK`3#7%g8h+w5SyzVS)g^t(>FSc+Vj_sJ
ze&R0g;(GSLusE)-mZ`lRfuYf1VkU!NaizJVYtew<3RQ}F7Y&XP|Lg9WfI{{YQwO;F
zF@b`&xTm?3YwMYBg2^oC$C;UQ6cfsbtJ&jXWNJAq4P28j=!dzuotxRa=Hc)cw-Fc1
z(beH`BA;u{lB8KeoIW8wjvz77OLSClScne|O7nbT4h#PXamjp8qI@E5@-D_o_LtL#
zh`9{>#Tp3|H+pA20`3paozvfR1dFDwyS&#Ci*R3^F6Qgt<wf{}IHPDQrkja!gu15a
z2<8^!A|_i>Y;=T-9u&qT3UUUKU`J9PF}7m!5a%^J<K^rHelg<A$_O#Z$n!j@Tn3Xu
z=;g*e@r~$z@>9!wzB0IX56fGoPE~r?v_aEKg{Jon_hA~m7Q9!jb`KG1k0m8{52=@k
zPOjZUieHFYnHHL>hFF07V-d^Z<w9H2$-{k4JiX;UCjq%%=cXrLXB$%_{T;2ZiqzId
zq(l{|zadgPQ>1|`K#tNN6{)?ANJCVlp@v8uOp%7GNF!9Fjy58VQjta*B6TuF(tqX0
zs7Re{L>i|ejW<N<Vv3~y%1>00y4r~Ju8K6-5UHCflKv|{RYmG<BhqvgX@((E4^t%l
z<$9)y<Yyz&Y!zvaAyQ9MB>lDVeHF>yMx+l^r1^$O0j5a$%k@GPDbPlw#VXPgL!=;6
zq>osD+%GLtk(@Rn>5uv=43UCOk>2C7HW_J^iWKr(k=7U@g_<I*V*xVKdKD?`xgu>e
zL<%=W+N2_Vq9XOO5ors(<)7WHhDZ^nNPnnrAKO%<NE?wpSCPIjM2a#++Q9<kiu+PU
zinbAHmx{F85GlqK=_?g!uZq;$Mx=cz(tbmvSW~2G>K%7bMT)Z#X`uM5=GvCNHAIRx
zMT%(XUUA>4NC`F~>HX68hDd!(k&dtcxh)-4k^0(*be!Ju9rvRlQa@9q>T1QERFM*G
zMAD=5lOa-nQ>33+fE=Y?RHOkmBAunT9Hn0kkp`L~y{1O#yoxl)MkGB-7Y&gHn<C9q
zk$zK=hS-R7h2C<Mem6uKYKk;hMY^UU4YLvHx{7qe5NWt6(oGg1x20Pu(g+)o^oqM<
zh&0j^sYg9`q<bpTC>xO;&|9v!zmi8<^W?YP(WXc%Te%}WQjxN=GN1hm_>)rujNw^R
zoEqTWj87i^YZ4XgjAh!*|JRQHiRlTG5<J%dK=<*#rySN(&*B&0A0HShCn*&NkHjpD
z$&c?<`2XZlw*Pwv8UL>y&eKQT{tq6_6UR~g0>liresNC!STQN*n-2FpoLxCU++7~w
z?D((auA;p-q%GSACCC{S)aK*usGF#c7bneN%><@NhlSmf*v2J)BC(OVp2t^BVC-rx
zp2Riht(?C&)+r97M#<AOIQAYQ4uiU9m-TkYoPi0k;=LLl!Q9HCkaLHMW2t>2A|l0>
zob)#ON5ZN;Uh&Z}elf}8$fLBbK(0DwER_=(i!(z0(Yu0Ac~^f9$E$HuFNfZ}WZo2e
zm#yF^6(uhJrj?k5TF*!-rd}0~#YgvYM)`#W{bQ(_i{t!Vd4i;TVjO6}l+S`rafDlt
z`xwfZg5@IO{t>yWIKb{76BCi>7bqrDRdeem&!6IYmwm)>dolT{dyd_IEVo^3f|%hr
z&cBb-FV-2C5FzIEO*SMcE|SK2wTOevvfDqFnzR8WRk`PiS0gQc^Kgv=Ke?1CqmICr
zDI*T_i{txZEYxaGarwT{Q6jZ#5&}Pd!U=MUxpR{?l%yA8d>uhfInEp?XC<iq?q~FI
zc*(8Pw`IEqZ#ww3M$84wtmxw9`$vR@h>bWb#>qUqVpS#ki6hx^gT#UTDAse7YeVOG
zLZc(2Ir^XULjVuOL`Q@LCc1u2kTdY|VDeL%T)j($`8a*N9bs~#Npys=9%w1HQhsa@
zo1J(qv0{^C=fIoCHKBvI{`rcVA^vNp(4l$r7IIrxuRo<V$<o9|7Zn#BE7qp?Ni4z{
zp(n)t*E6&enOtq<bicIKlT^#+Q_tn&C6^o$li(LACPI!)^h<~jiwNs4XZ20`4J3u7
zH)_>hOiWCBN3!<vW&CUT@>e}Ed2md0tehxVY+9nTT(3zBC?$UA;jz;EUc=8*>^A<j
zlz(5S@;~|U#Fq~=?jIFW`0-?%>0{P-8psc$Qu23>2|Q-;<H=;xLYJ>~FLVk`<wB?G
zg+BAyWYg#)KTb|Z?$?>Zk0&#zxB8K2@><bAzRve(DkIHA&;D^zUOF?2J~Gm5WD#i&
z^;VIluN4hsq`5Sek={ohMw&+-8R-LL5otd4R*@FaKt}oy$w&*4hmjW1M@CwVEFvwT
z-YU{k8puc=AsJ~I@-WhJ`p8HtkVT}G)LTVbMFSaWHIk9mAP*z0rH_oX4p~H6PrX$n
zeY(vCn#xEUk%y5!rjLxY30Xw?gnFw;n`t0NX$z8(wjvKBeM%o0X&Xw4lynYn?7vUh
zXd`ZoP=`OP$-lI@enc0)<MUH|tZ3o-+w0`Cihuq0KiMn4lPqXT=lFg`z17%#P6Ii1
zQgZCR;4#<OrTCNfm!^f*ZRK9*PMXSv?$Qf==DqT6`p6Zs2U%<*Ur}%MI``5*zRs_a
zjI<AVxQ*<mkBoEx{bL&mt!BQ5>-J>5Qd(gLskeF|-_St55GnaWzU48C752Spq522^
zA80BUdPFbunOE3R`pD5ahAdXtaq6vJ=R4~Dc0bZoMmm8!Twy2aBO{$c7Lk6U-YU{*
z8pxsl8Ocb$AP*y*p^uDo7Fk64m3pg4=V%}!okud#1>|9*i}aC^E+LCZzfo@$>2hz;
zKt{SuQyJ+B@-Wix^pTOSB8y1ZsJDvrwz|pXA2gMbt|Jd4-Jp+*^e3{2bd!3kNVjMp
zN9i__k?tT5Bi*HsjC2oKM7mGCRip<rkdgjEGSWljVWda&k&*sJ7LguPZxt!k4bebG
zN{wWsG|0n9Y3U;)r9&2x(o=60DFY2;q!*Bklo5FtDHDBUBrjwUDKqs}k+RT0M#_q0
zq-@B;Nc!i;?DV#Xl!JP!NI7XBBjrLe(u>H$NX=S_K5~?D)7v7_OVnFM`cVBr$)h6W
zwGk;FePpEk$Rbh!>a8NpSCL+(sT`$($iq>3g+4M;A!HG$F!ffE95j$Ct_YHmiXsmq
z6{C-gR2*4EdX;*sNF`_>Bb7ulQYqwNq|)?}k;)*8NUu?E70H_hGE!M2Bl#c?BbB3%
zj8q<3M5;i&RiuhEkdZ1O8L2YzFw*Pvk&&t(i%4%!Zxu;jo>i5mGEz0<VWc<dBO_Hu
z7LjUDZx!h+8pu(qiDaZ&$iqmr=_4c6K^BqfQg0P0Ql0fvkESwGedJ*zy{l|sh}6&&
zsSyq2C^behQWNB1q^9(dqtpypj8b#ztw!mt`gOepO=Tos<YA<i^pTNTA&W?@ske&M
zh6Zw!+9DaL9r7?zd-}*o9gsz&j?`O4($n^IqN$A38F?6~3w>myt|%!|(vgrJ^45m#
z3kzJg+T*}}{Z=j2RUwYJz}T>u`2RykJq{)RT$#eL=x&6z8m9{CT2?7JP8WF0HBKoW
z_3$$-RG-6JjYY|Y_N0@CgI}HAa`0;)i|xVR6saZ)keg^N6)C_*q&g~6T|=ZmQ>1ov
z-I406NI^CtHB^xr86r7Nky@!pO;n^{8<Co+NX-qALQIj`s7SslQmBnctyHAehDc$i
zNczKSTNNqXMx^#CQU^n%UZzO;BYh_oDZ)miE-F%2L!?Mkq&4azeRmZp%0?vp*WAw#
zDcTfCA4&69kz#B_3Z%E(F9jJQ^)^MCq(&)NMT)f%DO5!YGen9rMbdvdBUGe#t4L3o
zUM@;SiZ(>*XNsg}yz8$bCEAFjx21uGNc~Nb^o(~yRHOkmA`PRr+?IwLA`LV}(lg$T
zQjrGPh@?m9Z9}BNrbv3myKySgkmrgt!4PSvDUzP?Zjy>L%toZi^p-1biXqZ)QzSj(
z-E<XcgpEk=sYo*okw%&#=^5|ls7Rx1M0#IEnrDbK+7wC8c(*`BdfP@My?HM*L>gm?
zq-VTaq9Tp85$Pj(%WY|yA<{TgBt7HZN)>6mjYz9iq&0>}6HJlTu>kqmd%cSEj*Ung
zRiuv%ktUiV=^5`ft4Nb<MB1t%eQJpGt|^k9@$PdKX|jz-+f}3;hDcLPk@SpryHup9
zHX`j&k-joSnr4dhH4Bj2(moYwx{XK&RHTE3NHa{4^o)0hRHXN8L^`Y@eQ$^~(-cY1
zcz0Aqnq?!>aTV!DL!{ZJNGDl<TydvVq&YSsomP>4Hbk0higZRrI;$eRZzIw<73qAk
zNJ+<tdOrOaQJ;$cmyQy3=S8`Xo}_S;Xr5`LDloN!TzOLRfX)Ycq*6Rew8*s3*VIB6
z(^M{WiC*Y4pK7s`KJt&}N67s;Q#fOB8TD32!>XxB%V{bjtw0_gxLQdc8EF-=h_sq|
zt4MlMj5Rcsk=CN8MS9A&v32y3k=7%NNL#44ilonx*-BFx=~Lukq;2$(kv>Bfkv^y1
zDw3WQ;|rR~NZXNzk#^8WM*0$2MA}KcRU|zr#x9!5NV}1Tk@nC>M*0d_MA}QeRU|zr
z#@95Jk@g`EBkiY;jC255L^?>lRU|zr#y2#Tk-kM9Mmj_v8R<J@5$Q1XR+03i7~j)W
zM*0DH80iRoWTd0WBGNJHts+%bf2oesR7Uy{c^K&gePpDQ$Rg4y>a8N_Nilw+sf=_Q
zc^K(u`p8JXAd5(6sJDuwC&f5RQyJ-3<YA<9^pTOyBa28EsJDuwC&jo(QyJ+J@-UMA
z&ib1n(q&U5Jt@W&73ufqigeWw>6$5$o)qH`73unOMY>^#^rtD3o)qJzige3Hq}%k7
z+tM9mu`S)D-fCOYlVaSXsoa+CBM-Nw2lSDV{z4X!9#U@=Nl%LLh^8{q-^jyAkLe>L
zrTSCIB2sGVts?13G1AafMoNo3jFgT(GE#bE5h(-pR+03i7%$LNM#_jhjHJ&>%Vdb;
zWs0OH#mKB8Ww8-S@0YR~B4sm0(vxCjSCMkqh?J8)a>eCB7Ax*W>aA9so)jZDP34Mv
z33<5U^eE*qM9OQ5q$kD5ry}LI5vc%u<S4z2EJmpy^;V;#C&hS$rgD@DA={%=*bvEK
zilisSD54@2wGpWpedH(=M;4>>D)m;Qq$kBFK~p(OC6R}tREj<_QfXunsSNd2k@Tb(
zuhCRS@<tv;(wlc#LnI$lBt0ocITfkAjYt*fBS)zsa*t99lVVh&-fEQeq!^WHDn}`&
z)ub3tegS`SQj97rDy2y=Cba+89TNU^?K1ac``6`D|D)Tl&F62hd!SM{NL1A{P-)e{
z6Dc`RAxp(VU7J&i2Z^ei7OKyj%*^8DLTk|JY1`;i?nsc8-g1M;hTN|+g@Z(InIh#-
zuQR8LRMSSJ7geO(hDfzck@BcWc~zv^HX`L$kqQ_h)iFgXs3N_hBGt7KNq^`mY=~6P
z6sd@cR8&Q(|6GxZ8zMC@MJk~pl~j=$K3AmDhDeP}k@R6%ZxyMrRir2Pck-%jA9~Ao
zTscFe=B7ycu<V;EQVScAYN$wW86x?bBI(1jwN#{*HX_wgk?I;EwKYZ3hh>|pNbPJy
zYOW%+Fhpu^ilh(Awp5Wi*of3xMQUS+)X@}4AC_&WB6YG6se_8t(GaP#DUv=c+gU~G
zVk1&l6{(vcQdd(XeOR`Kiqy?Uq@F60zadh0QzU&@Hc&<CVIz`LMG7`V@-s!!hh;-m
zq@Fe+g{w%t43Ydzk@R8NNEIo-Mx<yJDaH^f&=g4@mW@@Bf^0;JSCJA7k({PT`mk(Y
z6)D(8B>l}N(GV%b6iFYJ9iSqGK3AkchDc$iNcyns5EUu>xgrfSMCxUVqz}uEP>~{R
zL>fhJ`Qc==AyT9%l0GauMn#IU5ow%?G~N&?+7wA2mVHM>im?%Cl8W@MAyRKsBz;(R
zii#9#BhoY#X}Td&oGFq%Ec>2{6mKJv{)RHk5GlbFDWaj+EaiC?b5x{0HX`Z$())%;
zeNB<{Vc8E<q<%IcEugo2$9-sslxT{i56do6k^0++q(^CqA<_U-Bz;)+BNb_&jY!Ms
zEk|jEA<`gIBz;(Rm5MajMkGB-YYdTwm?G)Jvg=f&p*A9Iptl^QjfP0WOp)|q*-a|a
za2t^}t4Lc6kw%yz>BF+0sz@VkMA9qnGee|Nrbzm*>=!E1Xd97s&|9v!FOx?~;jrx6
zrbzm*>@F4Q#pgXNJC<kt=V96K6b=!NH-)RI&e!-_g_~ewy!NZ{I*@$4QiwF!6iHug
zctS;*Vk6Qi73n8Kq^YJz`f9_URitS)BAro@&Ke?3H$~D{8=g~<X4r^yK}EV~i1eN*
zlD^vTHx+57jYwBiq~8sZW|<=Cs|~NINV9E3x~?MKFhrVTilnbLys0A1wGm10G;bRs
zy>E)7uQt4^BF(cA={~*XPV<2w(g&tU`f9_6D$;x#k^WYZ9vdPpFh$Z=8>Z&BBKZT)
zhc+UmrMDcVbjacdo+YM8`f9_hD$-ILk@U8d-4N*`Q>4M_4@yoIX_<{kFVb6%Qf@<}
z<)%pbYQsD#(h3`q^eE*sL|SQzq^~wCpdzh$u1Ez9kye``>8lM3sYq*VL~_tuuDBwG
zNNY`z^woyNRHSt_BE702l`uqFZ;GU^HY}wgZLkrkjEeM{A<{-uBz?7ESrzGH8<F(p
zUCt0`lPQwE+OUF(^ofl~mFO+ErOJj#n@y4Q)rM76q%AfgRaKFy86s^pMbcLrR#%Ze
zwGrtp6{)5n(l%2heYIh273nh@k?N{Q^$d|dH$~D{8#YjpzOWIgk&4vV5NW$9lD^uo
zsfx72Mx^E{QVT<*FHMp3)rKuqq@6Y*wN{bZ7$WU5MbcLrwo{RI+lbUbMe1mXw8s=l
zUv1b~Mf%D{q^>GbH$$Yorbzl~!yYQq*ES;cRFV7*k@lG)>8lL`RiynkA~{v0;AD}K
zt~Tua)U(H!lS<zG;{V(kWYu^z?hoD{`95)1=>w)AI!FU~_*+VT7x>0%h*Cb;%UAw!
z^Hot&IGOERQ<U^0+)<=tlpH*kblzDChpY~pmYG>CGYgW-e97vN)srvp$ukLm;9389
z$jV<{IyqsOSDl3T=tyx(4slm{aeE$d6ZmLxn~Y@Bm2o@xhT=Xb@d^K!x?S8yqd_14
zhy?%mXnhy1|3haT9wNSbn_2YaJ<E=o2K5*X<TfKE2Q{zfgBs4rhVKv?t#A03q6|Oz
z<hZ)tW1SkONtLQyXjoj*%x>I$iF?U7g8bwC10%!@@11dPI$Cgt`8r9roEP_}7k3J(
zFY4o*@#;ZuuXt|n!>#81#a)w_I4;?`5i5a5<elnecNQ<YtGg`4$BJI!4oPCgG>LbF
zMRKPUk;y4;R_`A{zk~>Ga1$047?BVp+9l8D?FbHYMg%#6*#hDdW8{sN+_(Ap$D`sN
z_q`LsV#R%F#C}8E^**Um+}$ir+;1h4oe6hxiStRS7th$DW&H-t{TjAx<5#av`?htO
z`!#Rj*P>3VR!w{xd3!Ms?UItKkw{^oUS6)n)NAS6ut_7odd=&!ZQG!&LyWo@U~V-K
z<nVVyh`Ten@5SUAC^4AgmIIEY)gYIfw4fC8_#}mG*SXab3y)8TiE#dN>0+duabC`T
zV(0_oMFQ9GiM5ncR<T~5_r?Yviw&{+wD8x&)W70?b5f@g8>jod;v*g`J#5;bX{AEb
z`-c07dA;G@@cW`w=*|h*A|l0owfeeO%QV+p<mO$mWxvDbEj3X{Ox5q&rMbHAttB4X
z89s|P@?9k*yS&F^NnPX_e$#khjwqU?4xj7#AKxl`6a2&f5bwYVua^_|OmdcsO^6b^
zmvVihV|$fr-lCyM<^Eq>Y+yN8&T>hck55eEwD5T>_X7&oOP(({A?K3~6D-f$makC0
z1Sz>=kV@Kvr|<{rHp63eF8IHOMJ`hgi-%_;%aM~~<zAnzeYWAg2^QOf^Ix0CPrr+o
za)RQWXi<!#2;C<NC&k*fqF6;CN8~!MyIy}$)TY?V(0@XalYeHTC>JXVZ;C@PqBuyA
zKTZ^RC^{sHqAf)f|LjHaGmTGC9PKBH?<uxXv4x^xZ&B2vXnjQ#z7#X|h+-;5Ix13A
z>}8?5DH@y?MO})yJozk&=0s~kk&Az3r$`eoipM-h2>%SCc#oGqjp8E~u$ZC=?HW>4
zY$b|v6t6R26)A!!{3+^P5Jhc@yX{1Ai{e@*QCy+e(pD6kD7J1D#U~V*-V;Rzii5L6
zv5#WnTv3dtc=KIRyg^a8vnUEujOCxBDb@stVkO087J8B5%t}$5rr0n^6l*C;oDoGa
zik6c_(VU_%3n)nO%?45Ir`R`M6kkzn;fXfMB0^{bg)>SN0TekBMUj=F^%tV>rP#@n
zZ>Ols0%}ow$4dEzV%11dET<?pN)+A{z1g^;C?a^GFp9L1qDVz?l@)cF!nchmno&&n
zO%!7(PMj0PF^c(Y0`F5yULcBz6d(K~in$bh{X`K@(akA}&J_2?iQ+az%@d-iPSN+I
zDB>wD9}vYwiVfe3VlBn4W1`qWk@1Kq(o>{m)uy88K*+Wf$4`sm2t}RmL{XC>;~`O`
zr|5rJ6n!bi9~H$Iih+AYkw}q$peXWCG#Ms}h7|qJh@vk=#<QYGPf?zKmZfOSKO0c&
z;GdsUoMTANh$3Oai~iyRoqtT2pZGlI7ab7p42+MP7Cv0OQ%kp*Rw`V-=STA99Ydkt
z-+JrSTY?@-y+v<QTZi6D=&e$(SH11%^`N(Cz1H<sq~9xg+ty>Qw{rb%(C-@kp3>X9
z-a3-Kr__5u@6YtRM!z3ZdWYzDuznBdHK(^Ry&dcKfL_=7y`r}wy`R$W6}^?~{hr=-
z^xjkNr}TSI?;lgzL+gE^epl<i0($SM_se>puJ@FBKbq2hUhiA=9$N3+^<M$KpV#ks
z{r5%hdG&rk?}PPU2EEtTe+Bekvy}GLdT*lMKgr%*dhe;<h58+*-=q4SsrM#&AENh3
zdT*llfO_Ac_b_@-r1w*LZ<5l!M(<bBi;q;{zx*kR3%`rv9QqAiL|4#dRB^T_uA=Ka
zb{+kRZlYW04!VQxqkHHfdVn6IhbYZWQT&b4qf}Q#krAaqnNd2F4P`(%QAU&-d7->0
z3o3xJp;u52<UqMlF_asXK(hlyQ5xmvF>mxTDu-S{6;WaIIx2#yqGG5zdKJ|~B~cwz
z8r4Uyp+=}IYKqFC7N`Phg({)8=ylWqy@5KTYN#8kj{MMDC;-($PE-elqI#$oYJj3p
zBh(u;LGh>=>Wf;S{-`Aygj%Das4W_S+N05^BN~f3qY0=hnuNNeDaa2^NB(FgdVy_i
z4hrS5`6wDKL~&>d>W7x0L1-l!hSs3bXg!*MK1P$#W;6qRisquv(R{Q6Ek?W03iK6P
zhxVaO=pfpG4xv5hd$b=NMTgLj=m<K6PN1LB&*&^Vht8u*=n}e$uAo2BHFO8vK!2fI
z=rOvB(q0qA1M~uVgtDMiw?vT>rA05H45$FggbJf9s5r`wN~2t;EP4r5K>5(?=w(z5
z6+&;JBB(AZjvAqos2M7QTB5S39V(AHqe`d;s)7PgHI#`hq6S*mN)&ZaB(?QXG-`x;
zqoybhwLl4|73z!HqD0gI4M3gIAk+;FL4If$3P2-}6OBTl=xx*sjYUytJnD_!LGfr3
z>We0${%9&1gfjEmhN9U#HVQ33W6)AG9<4$X(Rwr)Z9>z~r|3Pj2hB$N(fjCIG#~wd
z7NQ@~5_B3ZL%*Vx=n`6kuA=qm2KpG?L7UM7^eK9bK1b>P5XBCZ3GG5z(N`!Z+K2L>
zgQx&HB-=3>-=ktYb{v&KCs8?c8dXAPP*rpe)j$_fZFCtmKvz*ybRD%sH&I)32X#XC
zQFrtZ1)#?$1f{txiU^b*^+p*{ACwslK-thxloO3YxzRY34^2idqxVrEv;Y-BOHgsN
z9F;_?Q5m!Wl|`FTdGr~ogm$1RXb-A}_M;l;5UPcapt|S;YJh%5jnO&O4E=_D(KXZ>
z-9+usJ=6*Pjk=<=*G16-Wkmic3kpIxQ3%R|!qLkp5;;%|dKJZ?*H9l+9wnkGXdtSA
zhM?MLIBI}Kp{8gIO1(rB6Hsd&n}ph;DX2Y~jyj^5s56>_x}tfgJ6eGJ&?4lImZCtk
z968Y{6oS^GFth>nLYq(|+Jd6dHq;w^f#T4YC;{z8ebHW&i1woa=o>T$eTRmiAJ8y#
z42?i1&?xj1%Fg@k7c_y#&ZDX55}Jvwp!d-=v=H4u%g`+(?}d98ZRD{BNZ$GN5&DA1
zQr{7Uy!mK4w2#MLK=P)UUP#{QGAoidn9PCXeH~v!@^*@O&=vaTNAl)@1<@TID~#l=
z@`|D~H$?F&%7{v#Z0I%gBJx3bQ3X^GRYpb88>kd|6ZxRGP-RpbRYUbqEz}UzM@>)@
z)ExPumZ%+SgF2)3=mIBLbw)jTtQ!hIekce9pkU-gp(qrEqh2TiMWHCv8^xe_6pQ+z
zc+?*yqCsc?8j1#?5oib+jpPjFW09N~d;%Ip?IiRznu5ln=}1ocJQK;emgk^J)Xqb4
zX5j@$PUyP`O{aD#lGE2NN3(cr6`F(AqW94T^a1(=$%#9+A~|2?XJ{?8+tH_JC)$qo
zAUPM}*GNu5cmT<n|Gq^((C#q$5gkEtdb#6B&gym&$;s4CBRPlJ8FZZ<=g=*55#2|Z
zk({6EDoV>2o$Dwgx{0!&J17Ubk6uC#QGWCo6+&t5ilQh=k4mD9$Qxxw<xw^y=M2e-
z<fIsRPz`GHqnfB7s*MVxx~M3sk6uL$Q7P0Iy@pyKA0)3puYlU|SY^}!y@5KRH<7%$
z_$}0p$7-V<s2=Kx8lnKy1O=hyC>XUwp{Na#R|U355j@rj$!p-cq8J|Qf#elm{wSWu
zf>0k6g8HFwB(LU)L<4y&1`S4WXejD~hNDC@5)DM7(GVmr2N;ef@YrZ1PkA4Urt#PW
zGy_dSGtm?zPiLNv=JMD~G!M-|^U*x?AzFas8KR4jJTY@AT0!k{v<j_4^6bC0NS<`J
z0m*aSHX(V6*%l<v?AnIp2~}U9Pg%;BNS;2l8_BbF_9A(*%zh-#QTYbRQ$xN(@{Eih
zkUSCM7~0SBP9XU+`)MS9<vxStZ`0?H{6Y0MG{sjG%P3}3tf!E_fZV6nm%n8zI*KAc
zT8&;t8&P4j6%|L@QAxB1l|=_oIdm9RLB~-obQ(24zoN$I5^9QmN6pc7v>M$-t<VG1
z0j1)&eP@&b^*~urPm~KeQ9jfQy@Fy;QIvp6qRJgaQ5N~3@+bgRLQYf#g`#RG0@Xm#
zs1}Mtbx~i`01ZHm(Gb)OjYPh(9SK^?cBq4Fhq}sk$RE9pg3&nC3r$45(Nxq2%|Zjv
zd^8k&h(@BtXe{~&O+w4jG_(@ULfIJNdFU%1`w*=~i&1TASD^Jgwi<0j8_{O86@7}f
zqtDPD^aa|7cA!IOCpv<5qZ4Qr&+!ZTn#X=cnQMrmNOw`}=dt4G06$EWM&BTBbO@D0
zhfziJ19}~ON8hUGdsG7*L3PlNs1Z7eTA-g$8}u{kh|Zwy=o|__7f=Yggd)&o)Eiwz
zeb7xb5Zy<^Q7T^B7?ci8L>bXE)ELb{S<yV?-$fLQkP|IMp=dekg;t>`v=+sp4JZL^
zLjBMdGyrWwgV7ggIQkNeM!QiS66{3<&_PrP9Y#gbar7$s36(-;kvF=8%A>33b@V5C
z6Wv8M&|jz)O4UOYbx{UXA7w@jQ4W;4t0-PV&3P<8YKjV?m#8g*3ZPd}Ayf(#L*D3B
zR1TFw6;WCAI-1Aws-ib|tU9WJY9e21>!4PsA!>t~p?0Vh>VVp#PN)m&g8Wc76oh)9
zFw_%8p#T(zf>1vcj0T}lG#rJaw^0O|fTGZ36oY1<STqO4qxq;0T8#Rk<)}Yeg9f6F
zXfWD>hN92WaI_PRM0?R_bP$a}htW863{60%&_r|wy^AiODd-BChOVO-=r)>(9-!Ii
zF`A3g@tfj&<b{@?9B3uVgI1#gXe}y)rgalVQFMUEUPbFsDYOy!piQVU+Kk>rTTyMa
z4K+lcqvmKkYJ<K+ozO1S9qmB@XfFyz`%pN#h5Gx8;u|ywjiX{Xnt(>3iD(RZ7mY_#
z&_r~IzLU{mG#&kbW}~BMK01z;pc7~XI)&Dv(`Xa=1$~OnqV4D$+Kn!tedrSU7F|Y1
z(C_FZx`xi6>*yl-6J163&@FTa-9?#uisB*4fl>vCA~#BlUVlpzFQ5nX@Iuw7&4J!R
zFQVEg52}aqqet{8h#n&cO3m_$qqL|LN{`B-7f=P13B7?bqZ%kHs)Mql2B-{bio8)v
z^alyrqc?c0GpdGqpz0_Py~K+SMXh<P7kUdtp^DVTqgp)H7u7+7P(3sPH9%ugBQyy$
zK~qsPGz+yr^H58)5Vc2H`9s<k<v?jDUZUvCW9K78(GgujyD6?xbmy@fs4KdKPEg#Z
z@aM65L89<O4bf*Z5ZWnYps!FH^o<OJen6ej3Dgz+f_k6}$RGWVg3zBR1l>j9=n;xU
zY527&24zHXC>!d7a-l?&4-G`GpdqLz8jeb$QK&2$gDRo%s4AL>YNE-gKAMJ_qW4fM
zG#j-?@1rhgKJr5gQ4m^!!q74ljaH%rv<CG@>(OBJF&cq3qcP}HG!cD{rlK7vkTtUl
zInmc>KD7tYLUaf%LEocg=qP%d9zUWGbP9b*?OC)7okx4nCA1e^LHp1(bO7Bz-=JIQ
z5W0&FqX*~*^aveAsW`9XI7*AcNRR>bLYYw{%7LO$9@HDXjN*_3C7=?hFM16nqVi|}
zdL0cy)zA=B6AeT4(FoK8jY7Vt265Y>@jNz%chp!k4-KGLNHK}WmZEpia`Z068j7hr
z=7<)>WK;~rMTw#mn$Bb1=si>cZ9rAfEL2U39<|UMR2RLE8lVqQV_8d&X0jHwl(ncW
zT7WvDg{T`^jC!J_C>Sk6z0e92gI1w<v<4-jb!afkPcf3hj{%Da6Gb43MotuqLQnz<
zL;X-MGyp}S!6+IHL%q>R6o=kM31}SZi{3$r=v_1bO+|yy3^WAILc`EpGy;8qMxhVU
z+h{Qwi#|f*(F*hqT8$>5b!al$h^C@X&~&sFy@x(Sv(R=l2kk`fqdn*Y^fg+54xokT
zTeKJ*MoZBVv<w|bE6_>wG3(<r`U(AtK0z1IcG_JLn*N}LIFD8Q7ycy=U88arx`Fnf
zTWBx3i?*<k2k2AuH~I{v4Hv~1C==R&vZI|SH`<Nzqpwh5^ffAu_M_71Ao4-qqDtsH
zR26-XYN8{kK01b)q90K!bP{zyRf0v)9aTj=Q4JJ`9@8!a{fxp<Cu(C*7ZiuOp+2Yw
zN<=-;Koo$6pdd6H1*1_Y6pcaQXgrEQ6HyeJjLwi?8u}H@Lg&#ubP+8=zoBL53R;b>
zqK)Vev<2NjpQD>-C%TQkLU+*tbRQi?f1zXO5juqqlkW`5SW^_|QEGGvU7_}xP(rEj
zv^;u)N7JJ_C{>6k9wILuOVvvhIZ%3(2W3VtqnyZ>9(hq~Q~<R{ub_^|fx4h#s0S*6
zVpvLP6vAWPs23`S9Mo1sMbR6mII4k4pgO1&YKY39X2=`0LO!THDv!FLipURDMnR|w
z3PV*<6nYcIp&F<ks)+`n+Gseci{3`{(FD{GO-7B;4Ad0OLCw*8<ck)gR%khDgVvyS
zXd~)?wxCYvbJPXxMBUI{)B_zvJ<(wlfR3RcbP5HdGbj{YK;h^Lia^&<6uOOK(O)PY
zJw}nCqDadJ&m}yT0rf$dP(PF%^+zwEf#_v47!^T7QAso$d83i20ve5~pdWa)8t5vj
zh5kTw(GAo9-9(MiZPX0iMZV}hYK{Iv?a(9C5j{p-P-=FU-O(5l^hD#36HP$jXd;S1
z@1g`W1@%YM&`>l3jY2cgI5ZngLUYknG!M;0^U*x?AzFwQp^uQ0e5=rM9`og6%SWg+
z@}cNJv6{!app~dQsz(t(v7X01j}*mPv;*a$*h6uG$M&I(=pZUVahReM#Sw}!6vrvN
zDNa)OP@JaNMBg)LGdhp9qRVI-x{f|achGk97y1&V;*-)YlmYEQnbBU91MNdEp#!J@
z`UVw7htR9&FnSIBfGVJ)=nZrny@k%A`sfsDgq9Jg1v<@Rt<W#1J^F;&u4r#tQT#-4
zh~gKDbF{mUGPD!LJc^7I3n(&EETYIxv6Lbw#d3-Z^j(Eg)e^;ebcx44M&&5BQdFY&
zjN%Q7?G)81c2ZoX#~$=M+J~;8Z_#!1J^B;<h;E@%=nncB-9u;519ToeM3>Ot=n6W&
zQWRI$isCnP9bG{;(N%N@{ekYI8|Wdri5{cdD9t)i+(qfpeUuUXg)*Z@C>wf=a-!6$
zM3Eb%MR`$rQ~<qzUO}0V17$|VP*zj|Wk;n^8phrml|dCyI#e0GO1o;PB&vZ*qgv=S
zR2P*+4NwO9HbxmybL53uqb#T+%7(h594G+gLLn$OibQ!(EXs%aq5^0jDu{-mLTEH{
zpb4lbnu3a>nWzMshf1MEs3KZ{ywPgpgEpY@XfxW%tNaW#sx68gs4}&?&?#!aMyJsM
z^b7hHokfSyIdlYFK*!M~bP`=gr_t}|47!HSq3h@(`V(D7x6oB|2VF<^&`tCJ-9Zo0
zee^eah*GT<#bcBPrCBeEbSOQ_fHI<tC^Pax*-#dg6J<lWQ4W+B<w6BeZuAPugB&Oy
zDuxQ65~wjFSsFD(KBzgWgnUs|)C#?Y+Mqh9BWj4cq2|aBwL$@?Jqku$P&n#|qERr4
zLlGze^+t)PFB*&np%G{VdK-;J6VXKUE}DX-p!d);G#AZ43(!oo1bu*3q6KIjT8KVD
zi_xcODcX*fp<QSN+KX19gJ=!<4y{8+(FSx9?LcSH9&{1yM_19e=qCCO-9tyv-{=HN
zw?P!AkQe$HWk+XGZgdG1Kvz&<bPW|p*HI~S6O~1GP(}1Ns*2L_E~|;MqWUO1YKn5A
zRwy6pfL=!3P+=5+DxgqQ2}PhcQ7o#9`l0%0Fe=1OdL-(~V{fA-XdJ3S?IctUO+hu#
zbW|J7M9t_i2em*8P)oEFwMMH@TeJ?fM<1h(=u^}g?LdJj{bW&eM;VbH%8L9^F0|rZ
zQRGEV9(x&upu(s>dKJZ@Qs_86%A%8~JUWdkp);roI*+QMc>2~reNY|L4>dqj=89tV
z2cnpP)}fhbBbtpqL37boG!K1-=A-TCL$nhuLVM5>^fmej9YD*`w`e6gj8>x~Xe~O9
z)}xbXBRY*Xp)+VRI)}ERi)b6Vj0W=BuA)WgPc#_aM&Hry0s0<2LPt=l`Jy<6(xM+x
z1~inunb2^Q6^%r>&}ftojX{ObI8+QxK&8+`<b&QtmCzJa6-`5Lp&6(ynu!{r*{B7Y
zi`t@js56?6{LqKUi5k5xiU`yUMWgN1#-Sys5BdlVM9a}|v=WU$tI<TX7EMFz(QLF4
z%}1Ni60{kuKwHsTv<+=UpQFv_OY{ZWjdr60Xg~S}eTN>B{wO-cV?Uw{^F;9z%7lJF
z7peUf9Yz;WK5DO^0_Yklh;E=l=oWIIyQnC7fPSbgibv=eO1nT5=TIi}QXNrbN5An{
zF65+dUKEB3pexkAg8oFs(0x=IJx1kFhB~5n9pymP(HAVG7CM6JqT{FmI*A&i)2JCb
zgM3l0I-+QeN}x_CFY1aavB~#D`FSi56+&UCIkk}}0L36DibJ8O59);yQ4|`89Cbu7
z1QkQ0kS|(1Qxv7pN2uU?qF99rp|!|?HlU(t6Dp3jpfdE`hP=^s<b!sj@@PM*h`vLW
z(NR<dokUg98T2N)fNG#Cs3y9OYNOkzE_#6Kqg1m*(GX=ojZqfV6y-wAQ9d+<^o7w3
zR1}3%TLQI0rBNGH4z)wCqYkJ#>V)c`E~pXehFYK=s4eP=I->y80|lW#6pTVpDC&hG
zP%Mf<{ZI@Vj7Fo8C>FhqS~Cz6P+K$!wMSD>M>HLEMl(@YGzWD@^H4m?TY&naB`6WC
zKm*Y_G#G6{L(!*b1lo?iMX}RFF$N`|N&G?Uk0zr*=woV!qTOgT+K<MgV`wt^4oySf
zqgm((num_FyoKl|v=p5|%h9iB6*`aBqKjw)`VDPDSI`!86>UR*pfAu3^d-89cB9*9
zFS?8Nqx<L^^cVULJwiXA$LJVJJw+5JP+Ig8N{@a)FQ8vhCUgO1M!%t~=y#MI{eg0#
zKhcZmHhKx&LwV6(C_nley^K;%7sV?m9V(1oKt+%jDu%M6S5XdB61|8@qde$G)<=F+
zj>ih2N~j2W9Ti7!ppvK>Dub${vgj>T9@RpXP#shS)kD=#15^VwLbXs6R2MZv4Nwcz
z7_~&rP;2Ci+M?E|J!*$KqK>FD>Vi)4+PWh@9t%K$$cdaN6osH(C=5lRUZ^*UMDZvZ
z^+ml=e-wuXp#(G(^+hAlKr|W+Mq|-XGyx4qlh8;s1&v12(HJxnjYD(L1T+s#L<`Wn
zXc3x%mZE8BIhui1p_yncnvFJ~xo8ubhqj>kXdC*8@!O6rqCMz$bO2?mD~iMD7LOf4
zchE_651l~|&_(nRT}6MRn<y1axsTGI$0!|2KSLB5kQd5`vLi3_BFcjDqAI8m%7KcY
zo>N6p0tKMbC<u9@U{nr;qKYUNeP2hpQ8knY)kOJFeN+H7MFmkyR0y?04%7t|MSiF_
z3PL4N7%GLLP#F}5yip?ZK|@e^GzwKj<56Wa86744OmrO0K_}2W^b=Zu&Y(r;SF{xM
zV=2o~RkRA-qINyHgFZ(0&}Q@ieTr(+<8#y)?L^JdUep<VgL<OxQ7HNmMWLTj96E=F
zqu)?}^gHUt4(|r?L$^=>x{I9X0SZHp&`|oO;-kSv)c7M&j6}`QXw({wK^@UJ)E!Mg
zfoLKML+_#}GzG<>X{aB%zd#g&QLYa~F%0EKBT*joHp+*_p#ta~R1m$33ZbdUfo7nh
zXcj7t=Ashl12m(qC_Y3p(NZ)UtweLtdNdDxg65-b=tJ}+T7<qrOV9!I5&8}-M@P^~
zbONnLzo50~JX((~qj4XK;tzC)$8Mrc=nmS99-^%%%`#DJLmAQMC>z?2a-%O%0kjJ_
z&>mC*?M1JleW*M-fGVSJP*rpf)k23+T~u<RC>o*Cs4034wLoQ2D^w1(ML*ED13HSj
zpyS97oj^`>3iU##Q4IP8C7`ot06K?;q6=sgx`f7|%V-k%9Zf^m&}?)a%}0NtCFmAf
ziSD5F=x?+IJwV&gN?y?pv>NR~YtdI|J=%vhqJ!umeGj3k^+fRlN`;Q0uW5G*?MFYO
zgXk>!7M(}mp-U)3JyBdi+0b>A8{J0v(F0TjJw_!_y5*wqMww6rlnoVGB#IZ2Kab@_
zZ=wRI8nuN{Egmb1>hf4gDIP0>8liH?7ga_bQ8m;J)kLf4TMz9)4bj)A2|9q9qcf-_
zx{TVQYp4Udjyj`1Q8#o8`Jp=~0Nq1Q^Z<pTho~3&8%3d1i$&2Jr9ts19qNlRp#CT$
z8ic&iP?QCYK-th}lmm@LxzGfZ8%;ua&=iypO-BXLOjHofL50vf<Uk8hQM3pZM@vx&
zv>cT}t56xV7I~u$$Ommg<<S;Y5p6@2(HE!+`Vs}Q-gcuJJhl(jMh8({bO_Z)-=l`;
zC~Ay;L`~5t)Exbce9>9d3Y|x7&?VFkT|phtHPi{+KwZ!+)D7K5J<tQx6Fou!DAfv4
z1fjGj7-c|CURx#<$z$129LkCMq1<RN%8N##0%$CH1x-Q@Gz}F)vrq{%50yrXkT+V6
z%AwV$BHDmnN1IVq^ckv-cA}c-Yg7k)gX*Khs1Z7bnxa#v1v-OTp$n)jx{NxYKTv0M
z8+AkXksk_SgaS~yCHxa*LE$JDibVNP3@U`;P%+d8l|qTAEE<Taqamm^8jhNwQK%If
zgW99<s0*5i{Lo|+gr=b|^d5>rvr!y+AN51?(Ll5i4Mj`PNVE)%MJv%nv<6K@>(NZ~
zF`A1uqxtAlv>1JkmZ2SJHQI&Nqp#2>Xdl{!4x;Vo5DH_&zeoFc>=-(UPM}}VPv|`Q
z1zkqJqCe0DbPN54?xEk&BlHJKvs4s+qHO3k%7yNsyy!3VGWr`8MydG$xj0IPN~0H$
z5As44QC9Q@%7JR27f~IQ2Q@_bQ8QE!wL*nadsGy4L9Ze|R0;*5*H9QLhayo$6pda-
zy-`&ZhpM9lR1@_@bx<Ozj|QMdXb@_OhM*Q`7;1$^qqb-a>VU?f&S(Pah9)9E^ezfO
zQ;-u)L!oE}>V;;aC^Q@OMsra-nuq$L`KUko5Dh|$&``7ljX)ov(P%jui+Ztzy@NLJ
z*c7ys$EKrbG!yNo_I)97M%*`eK97Ef7NQ@}5_AkLLnqNn)R*O~K?Bf6G!kt=W6|em
zBHD?jqrGSjI*2|%htXnm4AtZpq%exQ6cH3_sEt9(P#ju;`k_^55c+D8D9WM|6y+&4
z^Vma*FDM>Ud_s@3)UHDr&<2zleT;IT8#I0i9p>O)uQ{SPj-t>{s5d%`;?V`v7hOU9
z(RDNk-9|&v12h6XMx#;M*`gSWGNK76E1HCIp(!XY`hVEF7dV;fKk)xTa;YRqr8Gum
zWiPIgQa0<FbzLMhJG(o(J9c(v&CFU$LZsxDqN{XMN+s!@B62BIbWtLtlt?5BmHzM7
z=UnE@?95)|`~CiYkN?l(`~CFXGw1R-pYu7N^Z9(v=bSSKU5p+?jnE_L3N#NjLr<XQ
zXg=D(zAZp+p@pauzrKVzqa~;-dJT0)Z=xH~QnZw1m!WsjO0*oUMJv!p=za7VT8+L!
zYtc98L-ZY5kM^LC(0=p@I)pZ%KhS6BFZ2bf%+0bdQFXKhosG7l^UycwBJ?9_guX?W
zq3=*rv>PR&&)Lh?=qIMNMZ1{R0qtX2N3<VxMu*Xj=ospY&LxtAQ1uiEZbm2gbr`yw
zUo%h><U&nRE^3DIP$ic2qRJ?Us-j|44UI!tEHMdfdPRaBZ%FV3N=09yK4=T-kG7(L
z=o>T`?Lb4(x5$CMLm6l{a-knlF8T@Op?%1U_M;K#01BdCQ4#tLjYdb%So8-PkB*^<
z=ub2m)nqTHpwnKJU<NuJ%|d6O*{BA30M$YdqqESXs5W{W)j?08y672H4?TzKqZiPH
z=q1zuEkT!{*U+WtP1Fc2MUBxibU9jqnxIvvDO!t~p>-$$eT0(GM$`g*hFYP`s14eJ
z+M;dfYP17ggLa_~Xg9hZ?Lo<CAL@jDL7maBs4F^*x}!hPjp#T^MSr6+ISQ5Dl%PM;
zPDg`Kbu<{&LPOBmXc(%49OyjcMD<Z7x(K<@B`62kkQ+5ddFTq1kD4MMibo?*5(=P}
zs1UV5MW`JrM%SP*=sGkOC8KdD1x-L*(L~e(-GNfkov06*g8HMWXdpV96dH`KLbsrE
zkP|)5eA(zp<VN$62R(y)=vm}P=dx@eIu8}2^U-bS0yGI-gziQcqnW57nvHDeVPr>-
zq07*GbOm}2#i2#$O7sefM{l7-v>Y`@t5Hj|9<@fBP%iokwL@D`d-N^37X65>L;F!j
z^c%VX9YZPT1nPpSyd^<5R2_|ZO@g!0Y*Yu`kIq96qWb7zbP<}1E<y8<4Lyz;qbJc7
zXg+F+o<Z^GS(Jo&5QCPeCu)m&p=(hebOY*#x}r4H6AeWD&`oF%x)}{cLy;5Rf?Oya
z4M!QM0A-;<l#NECGhdfr9IAsRqIzgD8qR!oBM-U<EkpOC1m=4PC84><%dd~20yH1_
z(Q_z>7NL>o6*LOHiAJM$(XD7Dx(%&G<Ix6mJKBUMq0MMA+KTQ%JJH=}H=2g_q8aD_
zx(6Lb_o8EHHadasM^)aI;6YR!J&evmbJ4kI9;%NXM;D_fQ6n@TU4foK&Cs)`Ia-L?
zpchbkv<O{~7NZoj1a(KRqF(5A)E~WxZbEOPVdx!{fj;NR<f8YOmWRG(nos@8G(TF2
z3ei4(9gX&*vFHFAkA6iH(QjxnTE((c&>A!YeSq#m>(GN}19}vFjGjar(E{`-dI5Ff
zg7Gri%(QVcB=`zVLSLi1&^9y;{etd6+tGdK$W#d)L`Tsh=ub2c{ezxBRi;TWA5}vO
zPz|&Y?PPr~qFrbS`X0T3en9V{J!mD`i`Jr_(FSxJeTELA&FEIPu@#L&+tKalTQnJc
zj}EcKkLWPkkA6po&{1?0eaRAkqpF=HIE|*qU;J7X9pcv-=m<I!9YwX#pXgk40@XvO
zO_$&THI212K$Tc?BUBkRL3L09s)m}Q>ZlFMVlCI8+t77r0!l`cPztKW5?#?*C>7O4
z{m{qEHwazMv>~W2x&`gy*G%*a%0|B;H#&?w=nv#W^;p)A>Z2lbA-WYcKoig<XfnDK
zO+}5+J*Y97gVIR$htVx)9&(~5kPFR6O;}<9>W#*}D?w8<9&I4Y-+?|xccG1FD*6=7
zK%b*os2R)7MhWOal!P8dEznb_6?zV}L5omZ^a{Efy@jqp%TNdOKDr)#fRfQis1y1W
zbw*#Iu4p^zj=o1XqP-{;9Yhn*AE+-nj{2iY%On_ps-Z#XOf(psi-w>J&@glfav(c$
zq9$m;QV9~!d#E{DiCUpm=qj`ZU5!3K*P?akdb9!EfIdc@(MHq_eTr^GpQBzVlNj_x
zE;JD3pdrYOoG1@vqkNQyd}st3fkvVL8iNYacvOTYp<*-zjX^WfSTq}rLl2>L??^BY
zO=sGZXd-$V_2Soss1JG(^+StM8hQoY!4j{dJJCC63R;1tqP1u`OKd<hnYIx<$h6PV
zET(;lo?+THbRW}ppgHJ!w1Qvvq6e6^A3cN)p-0eB^e8%ker1WO%OyC1&Om>ln&=oh
z3;l`CL4TvV=rPuEK02N3*Z?hJT0=A+HA2szE6^3p7muz+N$48X5_Ldr(6cPj4lP92
zp%+jmG?Mwcq0LN7cwd6Ws5#2yhc?KC+MyhD4RWLFP##J~OIS7qy^6Y_*HJ2Z6ZJ=L
zqrvDMbPHOBGSPd;jaDKrT7?2=4H|_$Kx5H5bUWIB?m{1<8E7M#jXp&WqtDUfXbZ7l
zfWAQs(II|aguX&Aqp#5$Xd8MLZAb5;o#;cf3w?sVM_-^H&{nhueT(*@AJETeANmCy
zKnKxpDDOQ9jw2uX8;wAfR!I;*RZt<ShKf)PRE*9<V^D207M+X6p?YWnx&Tc?4bUB^
zA-WSaLQ~LXXew%grlTv-Oq76Tq2}m5)C$c(SD^>c)#xE~EqVl9j~+!gpvO>W^aSdL
zo<cXG!=z9z^c>UrqZiRY^b#73UPeRFE69OfLmB7|<U((uT(lJBp?8rNEk`5J3KT@|
zqaw5#jYezHSo9$pkJh7!=p!^4eS)T-O=ueW49!GepnK7mXb#$f9z<KwBj_765A8rt
zpl{K9^c`A&cB6&pNAx242`xtZ&?{&^dL11=Z=qk&JLosG934R`(eE6E)#y*AtwaBy
zkI-o=B-n_mpwH0hXfrwkZ9z5AHdG7kKxd&{s5aV->YzQSF4~9cp<hsa^eeg$9Yzh%
zALtTv99@e3MvYLV)e<yDRnX<A8am3p)j-G5*{B`Ao`*W1`lx3Y2`)k@OuGbiK{oVP
zX9*gk6DST<Vu=J)8MQ=JQCn0EU5l!tWK<J%L1&^H(b=dEItL9v=c1d@dB}mzM_K3s
z<VF`EFS-~7P(w5d+0a;IN4KNP(4FWCG!4a}S?Ef1KZ-|>phWaIYL1>pEzv^M8oh+B
zLa(BB=q=P9EkoC$_tACe1Jn_Hgl<5eq7?Kc>VmeRZs=Ro1O0${p`TG7bP)Z<(K?KV
zG3^i34;@F<R!Z;>s(~u6k>E^pI;xGTqd_cD3k^oK(GXM@HDbQ{=x(MpKn~OpwdL2w
zD2r)Vpj^}p`B4j0jIKhrqifJ))DewgEuGQbs2iGwZbUOsFEkhRMGtq8U;vtiZbFOr
zbr_n#v~=_;(_H8trsbl0Q9hcD{OEpEgdRk<qSLu#O+e?NN$7lZC%ORLjjlk`(aJ6o
zTsl#LxyX)&Opu@n8iuY!4wQhLs5#0+t<XG{y$U^!+M_4Y^=LlojGjR^qGwSbv=9wI
zFQ6f45lTmkkqa$B!_lkAhh9fP^d=gO-bUlmJ7_XmhNhwS(7k9Av404yV%l8vIeG%E
zLG#hC%=aAnfN3wFb?9ZZ0lk4fM(?1FXa)KdtwAZbORyf@f<8u0v<bP;=V&zg5=}%~
z(Vb{Jnu5MXQ_=TmI{FdKM0?RJv>)Av4x%~eH}nAd9X*7Op-0eP=uz|!dJI*bB*7Et
zbo3Oej-E!f&;oQedJfe=&!h9ui>N+&30;I<Mwg&hkPW?t8lyMR73eM06fH&Z=v|bA
zmZO$v1!{xdNA1vRbPZaIu0tQ9WV9ZoppQ^j^a<*LHgnuk(HBhXhqj;rXe+u2eS?Oe
z9q1PHEpnpoP!`&aa?p=xIQj|YqkX6V?MDH00F6YyqGI$Lx)mKk<Io@Ic61Egf&N5y
zp})~o^beYWzGB~Ip=uoD`_b1-dkEEK+B{SrJ%P6I>wNU*-4ZmWL+u2*0#zYsN^k~2
zJV7mjB!W5wEeXyiXhX1_^|eC>M@z7Dj0A_#GW0uIfsUe8=r~%7{zB`}3G@-FM6i*d
zGQnpARS7l|R3q3zP@P~K!A@eZ1O0?{qkZTnbO8N=E+;rdaExh3(68tynn3V3!6br8
zw@UCEzg9s<P!044IvXt^5v$!H!Jq0^w3cbL(cerv7yX0kqtm)da0#k{8lz_DoN*GI
zj_RUjto{O&fEu7A)DX2mjZiCe89IYyo1hw~8LEYvqqERes5ZJ5)j`RqF6x5np`NHd
z>WeN!15pDs1YLsC(WS_R8lgPY7>z)eqaxG<jYUn-MAQ}CjS|pwl!Wd>Ezm=#6?zP{
zK~JN$=y`NCdKq1V-asADyXbnf5+$Rxs1sU`I-^g}5VRR}M_W*bu@Y=Y9nrU_6Z#%?
zK|i7!Sz<3rMF&uCbQtwT$5DUu4;p|@A1}cmR0|D8=b)EZb3OD5x&Xa~8lX2&L-ZDE
zgqEVq(7UJ!T8^$nD^LP@A2mm-Q7g0-U4=eGSEKdlTJ#aR9({stK$}o!^cm`gzCbsk
zFHtYF1@%Q+Q5yON4MIE6&FEV+41I^v(QcH9eni>mC*(%^kO%EYK6C*2(XXfw{f0)N
zBWMiz1Koy>p$X_uGzkskDBOuoyG?>==yWs_oq_H}HP9SX3q6R=LXV)@XdbGAo<Mcc
zd{hrDK=si=bRl{XH9(8eCFm7&DS90>LT{nQ=pA%9T8^5am8dCNjhdkkkb`|&k6JQq
zBWjO6Lmkm()ER9-J<vAP8|^@8XcroccB5O+9+ZLhp&axJ%16JVG3YQFkN!ZD(Qz~t
z{f+KHm8h!SkE)=#s2X|_)j$i-ndn7S8@+<gMQ@>cXgRt7tws&ddejhYLXAW+BJN9k
z8B@POP0)AfO5`L36OjwGMpZbW+oNiz1Ippoj;J#i#<>J;ragx8(9<X%J%_Rh77_TE
z_A)9Wc%5Jb)80Y>v<waAin)YfD8Xw4PJ%ZHTm(xA@(7j@_y|@I1PN9V6cel^xQ$>P
z!9;?O2<{@-NHC4yGlE$Jn+Xbu%NA6GcA#SP9csnJt=bd`u0l1?)#yxgEvk*KNA*w&
zx&U=W4Nwo%5T&9<s4uz<4M0uMV00xKh7!;}#HBeJgIb~K{Mru9Lf4=<=sNT;N=A>N
z6!aA8iWZ<A=mnIDUPgV;tEfME6AeV~puy-pG!(5y4)h_)Kp!C&+Jth^W|W7%MqacX
zjX>X_Ao>v%p?zpHI)KKaLufoYf+nJ4Xfis1rl2a5C76cJKr>M-bT6un=Af~p<b!A&
zdK67SPojxv0lEXdfbK+#(G>I=nu?a9>1a8ciB_Rm=tFcL`UK5EU!VujR`d|si5@{e
zphwZq=rME%J%Ns*r_kT%X>=OZ-UX-{dJfe>&!cnDi>MxY2{k}3BO7`JU4dRhSE4si
z5_$`@LQ7Fw^e(y<Ek_;E3e*|Bk9wfhs1I6;2BHtqP_!OpppQ@v`UK^pO(=jqL&fL|
zbQ}5-O+;JJ6tor1K;NMI&<^wf`WDSa-=QbaZnOaXh+afLp;yp8^cLEWmZJk`HTo5;
zN57#>=m`1}{eiZkW9VD-C;AcnjrODYYb1Dotpp3v8uSwS5WR{vptsQ{=sol)T7$ko
z>(N(e6WWTtMBC9e^ex(jzDIk|kLa{+66{4)&@bpLbQqn7j-iXtKj>mq^#cj)s0M0+
z&PMU5E*i+p7ox#xF*FQaiX6y}oal0tiQ<q8HA6Wl5xG$dl!scQeAE{CP<u22bwB~s
z5f!3Ns0ejI#i%<PgL<N|s5csi`k@JE0GfzyLM^&UFa%9uS~{ABGEpmj%|<t%Jd}b)
zpm78be<;C3^eDOmJ&x`~PoXL388j6=ho++!&`k6anuV62`_OA>4tf(kfR>_%&@$Ae
zn*=LRH?$h{K<iOYv=Q|}n^7OM74<_qQ5xEf2BN)a-v<&LMElWi=vVYR`VAdJN6=sB
z5A+W@hAOX<;7@cq`Wsb8|DamvG-f^<RY7&o>F7Lk2C9#0po>r~bO|~O*-&lN7}Y^n
zpt`6js)yoHeUyZ5>Lx);bP3biqDxVGbThwpKrVCx@}jP&2=zq8s4p6W2BEQN7@Cf%
z1|%4V&OlSpndm;$xk!SUs2iGvQqiL*4b4G=(72!kL(v4}Koe00x&yh;11yn?9zq`U
z2r58}Q4x9!jYf-Db{u-1X}7CsXfj%brlTe39<&_IK`YTi=ri;<T7#ZKzq96N(NXj~
zI*t~hP0aT)+KS#pU!bMvOY|Pvf>xpL(1++7v;plvo6xuD3v>W|jdr7N&_1*a?MFYL
zYTfx0{mQh1=r42xoj}J>CFc7ZRYsLZNl+D?fi6L3qw1&*s*BD?_0UDAK5B?AM2%4c
z6o;CkBy=fiiJG9cs4cn<HABg$1?qxYp&qCW>V-O?G;}o@ggT(1=z5fnl2I1A5e-M3
zQ9kO5{HQw`iTa{3C>7m?dZUSG5V{NXM^n)NbPpPa=AgmoK{NzCigM7C$bp_lPP7nZ
zqD9DsmY@;nE#yY;pggn!<)hWehdxAOP+h+S0dzhpL=8|8x)c?o#%LmHK0<=As1+KA
z+Mx;PS~LYE(bl*FwM2KKw&*_eHtU;;-bK^VN;DI#L9@_0GzV=&uc0r{$Eat41RtQ@
zs2xEX!8)c5LL1N!^c8AfEWt+90ey;YK%b*7Xfx`8wxeY%`!!mDwxKoX2XvH_+=>1~
zyHF*T{T@|CKchn|u?HPNd(m<9JKF1$;1{$Xtsyu>@FBqwf(-;m2|gkCli(mroIr<A
zW!85XRYNCGD+-yT=qhv^U4#BY*Q2UEB<M?7TnVM24g`Y<k_m<qq!2g=Dl=aOIu||3
zURFa-qv~iOs)-h%Gtm-sHhKe{gWf?Gpu36ld1yL1AKi=W=y&G32pvNgqZ6nhs!YDJ
zp=#(#^fU3h3>`pMpu;E*9YrnCd=!rsphWZnYK|78j_7%Ec5Ac<U4>pj?a&*jJz9#c
zMem{O&}!7Lp9J;$OE3^!gl<BYpqr5m4MmO7E$9lAj+&wj6pyk{63Rv`Q7&qOhNE`K
zgRVhdbR8-{$;gjVP!M%RBT)}D3Z<gas1Ldo^+z|bZv)XprVT-Np<B>&<V5$PEc5`%
zLA_aaI7&hJ=mmcDqnA)2dIgO_-I;F;x)F^-sb~`Fi>9FdXa*X9?n5`DhfrmL#|UZ=
zJV|g1zdnu9m3oWrLeHZNv>5%^OM*AhKJ+#^h~7nq(R=79dLL!6>>8Af)}dVV2^x;R
zKpwOedC^W(fOaE4+J}PZAR38|pi$^J8jb!zx1y?jB)ARLMB~vp=yr5InuIP!laU>@
zCCKe7K?jtFQjiz*KqF8e6hs425gLj{qYN|_<)ZP(izcEVnv6!FDd;vd4NXKd(Ou|X
zGy}~+v(ba-A@m5EhvuRA=n3>Jnvd>g-xi=(nf3yD6TO7qMN801^cq@=-bB+_b}9OV
zY0J^4XeHW=R->=c2WUH5k6vZjkI@XY2_59u&FB!?f)1l?=y$XO9Ywp)JuJH$-HU!g
zv(Yc;esmZ;h>oL&(LZP|s>&%e57k7EquS_6bRL?IE=13uhUi(;7%fCi(F-UEEkbS3
zVss5!f|AjvXx?549{5RuC(t8kK6(r-Ku@8CXaRZ=z1~BD#pq4+Dta5eh2BBS(K7Tt
zdJnBdE75wi3T;Gd(C6p_^c7l%wxJDZC;AwDk2a#8(8UA?2<!xh2$~QaA!tT$l%P4m
zp9F0PP7t&wIBlN<9Z^-(8J&S{L^aXpJtQ~_ZANv_SLl57HEMvip-a(r)EE`-l^_n?
zikhKuC=uO`TA({nYjhWCi>9LXXa?$lW}%K~HtK{PKwZ$os5{!(LxR3PORx(KMBk$!
zD2-WOCb)^<Rf3xd-XIuC@HW9M1n&~06TC-|LGV677Qq^VY=RF7atSsN3@7-6z(epU
zftTP5f&zlC2>b+F34#RM2}TlpOE8MydxFsfKN8$Zu$SOAg8c;J2@Vq6PVgJSB!b@w
zCKDVZxQpN~g1ZU+A(%!`nRJ<fPDl5k>gZln3(ZDnqx(@E^dLG9J&fw3x#%J^4_$&D
zM>g~WN1-u#hG}tV57U~Vw<%SA*e$^#f}aSM5d2K=I>7;g<;-^oT{%R8BdC9l1V_<O
z^e0+IaDreJOPuzD1n=`_RkRkJfnMX+n&<;`78*p~(mCj3=BtZ7Mdzc<=t6V{1<J+f
zZgeS{iR|dCGzl(8vzZo$9zxB~stgGdQ9Z_4v_S1wNYENxkJ_Rg_eszm?I*3qvGG)b
zi3A6jHW~ejrlH@^z32#f5dDGXp&SzXX@X-+dlvnPUO=baD8b9<bo4qp11&{0&~j7@
zy^qd9Yf){q9@Rmgpt|TYR1dBGQG)tvKGXnhLYJT~(WU4cGy?5Hes+I13Zgxz5x?$3
zjnP4LIXZ%xpyQ}1`Uf>bRrg4cfNG#5bT(>%>Y`TY0@Mawg4!ZGx*9b>*PwXR0kuHa
zqpMIdx(0PZ9Z_f08FfWHP<PZD-H6gqDjJM>qgzm4l!^MITr>dXqd~}z2BRW01l@{;
zp$W)=CL<@BiZanX$c5&h9P}`9qsLGlnve3)bI6Alq4&|N=wcH44b%|5jRO4oE-FMT
zQ4v~;iqQr%25mxP(PlIbZABB%PBaniMt7jS=uUJ1O+kmzRCEkYM<>usRE3NBEL0uc
zht5KC(7EUVR3ANrE=G@_M(9y=1$qoMLr<XQ=qc0&J&oF<1?YP897;jYqweTM)C;|Y
z`lFZ8P3RRg484Xj&>JWRy@fnzDH?&^MI+I2GzP6e<I#ucGRlM1s0liqAfBK)K@!24
z1T6{HGG80C9<@gwq0!$<umg=l-=Rrp51N7wqZ#OTbT6vhQ-TLjRWui!PVfXlHKxr+
z)zL!K=35C~LRS(jA*ji(uc0&1Qq+s!J%YXj?-QKOuWQgbXagEau!-PYrnT86K||CY
z+0gaKj*hVL%g}Lj1^NfYp+%@IdIepL-bAelvIwqWS`O-pJg7GsfktnY;Ein(^kv%H
zs6SeU?&8-~Xdu(pq6e6^9?eCcpvTdt=t=YinvcFhH?izibTit4hN17!66V{3UPt@T
zo9GwxHu@F4gASu*=nwQBI*wMNztJjG=^F{wpepDCR1K{|HP8lhCQ2tRwNVDDi?Yy#
zC>u3Ix#%)995qEAl!&~j6)HgOP(vbg9r80R85N_hXgunPx-nlrl!^wR-smRO7Y#xE
z(Jg2Ia-u;f3k^m&Xb2jPhM|1qKn2K&0_YB6FcRH~Mx!Zc9GZb9q5IGj^dOpnj;xnp
zHadnLK(Bu$f%kI><}%+1G!GS`Rs1>zJ<hb-&}U4$9esr+qitvk+K#57ooFU{l4b8j
z^U?k28T2svh4~&shtQMgFnSvOj-EwF(evmyT7>>WFQXIaRaA-K4T8!9Zxd7{c$c6W
z!FvSN3En4ombk1z3(-3C0@{ccp)b&4v=uEuJJG9XH+miIMQ@@5=w^Z=1j7i961>f?
zf1)iMg-TyY@Fa(#`*sQ5;n$vM8R~-;^XowL9@7S+<xCrfRv{<)0A-<dC<kpo!_i8X
z%}1-y2($(jqOX~6G}?~FqMc|w+Jz>f@6lxR1Db;NplN6?nu&f!_o83W9CQ#phz_Ag
z&|x$WeL!5EK<m)cXajm4eT){Pjp%jsDO!p?NAID{Xf^r@twVp4&KuF%FC}QdMS`#S
zwFBCQlF=9Z+7)eQS`YL+(|VzwP(QRE4L}FbP3TuN1np$mThK0)fxbuC=pW|GLsh<#
zz>7{tBhVQrh-#oBR11wpXQ8pEHX4uWpoyq1nvCk9DX2c0hAu=i(GSGsUbF|@kM^Ru
z=x6i<`UNdO2hj`Y5PBIMMz5pa(NdJhHddgcOk0JHqjl&n^a+~7d|!xUMBKypE2hpx
zThTnU9X*b|MNgvd(R}nHdIs%9&!YWkAv%a&K)<0y=y!C2I2=QjQYH8sRYsM+mY^y+
z164z3qUxv)s>ICoQDt-yIvrhtYBHY<UCgx0Q5~kmp`XZs31~lRj&`F~=qi@D3Kg!C
zpxOorict;Jf2{;(qH#>CjV7RT(L_`a-GMGZccKPp3TlX^qDE*sx(v-kP0%cKCAtqK
zpgE{Hx`6#?g)TyE(Z#3(YKU$?Hq;H-Q7XC&^+Q*nK`0IlML)5P40I*aTqqvpp+r=G
z8j@BcQ4>^*nxR`!b2JXMLARs!=nixpx(jteQ&BfG1NB6+P+v3~4MY#1=EUG()Dq1@
zt<h8HD)cOBhh9YO(Gs+QwY-ToqNV6_v<!WPR-kQY721i`qVLf<v<H2JenuP7LG&3q
zj5edAXbbuaZ9|m^b`T^Ekzf}Z`>6z%ak0FXm^MMzp=M|*zqUXfnbsOT%(QmsF?20@
z3SEz$MmL}Zs582OWxJsi)Dv|<eb8&nHvqkbZbD1Z5cDp(1uaKTv;t+J_fZa7jfSJO
zC?9=@3eb8KKp&xz=o3_ox)GOKQ4cg8^+c0VFEkbPL9<XlbU#W%kD!6*adZ=U8ht_9
zKaXm0S^tgTW`6x04MoS%g#?v0OK=O*s-QTgoq-ZiEz|;?jas2Ps0}&~rL$~(lz|$c
zEOaUA#C(^bY^F6qxhMe*M=eo4^R+{MbPWok>(EG)j7FgpG#Yh9x1t{CHk69Sqdw?%
z)E`Yk1JPtO7<q`nP~=7Fr~tW;9}P!A<U=D-5N%;CqtPg)jYZ#(_7l;5GWxcUB{-nY
z0CWiLLPybV^f%grDt{!wJ~WQ?{emW-LuexU9o1vLKheeL1iFJ?Py0lIyHGWh$ged~
zJ9IX>n_ugoX{a9R$*&F2Or|wN*-W#e0(1qM#jj1#eJBA<;@4Ja4%4ne^O)8iJ&UeG
z5AbU;dYx%q(8EmYjy5ta75zY2+!t+QS{mAp2BDqkX0!_pL*Jux^aIL7dr&soi`?jE
z<Uzk6A3BKq=nyJIb5kW4h327K(c|cL^d!0y%}3MGGw5FQEP4PfM03##=n3>8d1C=u
z#I%KIF<Ok4px0172Xra&qh%<FR-ln+6&i)sqS0s_x)ptdZbKW<c=Q>%9c@OF&=xcq
zZ9}gTgB|E~^c{K=?Llv&{pcNZ2rWZ@pphK6ztLz^X`=+mR3`QiyvKa|&`NXw^(Ht%
zu!?C%(Jcgj5o8inqN$yOsvtM2hVoDiw1#ERL?57Y&^mM;DrUY5(O7gb8iy`L6ObKE
zM3<vGP#n4wHA7QSBASX?py{YJnu*$?S*Sg_4|PBrh)YNGF-k!jQ4jPf>Ww}}1JGu4
zGx`cS(AOvnZ9{JKE4j&sYLVDW2)6U<YiK8W3tdRCoM0EzR-!n9H3SI+>j+v9d_>TS
zU?V{rg3k!PXW7l@2lO@CgSMki1m6*KCHR4$JHbx`Hxm3zkV<fXpf|xGg1!Vt2>KHo
zB^W^PC&3_s69j_^PTM5G5L6WnLua79#HA+s8J&%OLFb`^=t6V|U5XB)%h2zrDLRT0
z(Q(uYeMCB6jkeNwyKbihfAMQFI)OT)pZN7gRH>H)z0gsn^+SK7f#|dy5)4LF&`@+b
za-hmAn}MpLY*Y;mM|GLchw39gx)2qj251zz1dTzLqT5g-Gyye6lhEbpPSgb5jhdqA
zs2REkC7}CIb>ebAs)-&(XQIc^+30C>4tgG)ix#8v&}-;?^ftNxEl1<n#%gps`T$+T
zuj|o6OxuW_M4zE&&}Q^J+JY9NZRj<$1HFZIp=D?{dLQjUYtcTm0sVqDp<huWVsIEW
zMn}<==mbhaRp}3Hfoh<u(Anr3R2Owb7ogL)+BZZgWT|}wDg62i>VgiTegsDex-sof
z<RtiqAe*2HRmtI~8uFkT$cxTIJy^Cj>WS*2UZ_60mH944<I$z)c4SAB(B)_{ibHpy
zX6SB|h^C<yXa;JH?m=zQy{J8!jXI$FQAgB=xO77OP*;?OQqe%v58Z@rLN}vfXei1+
zx1b!9jyz}+XRRO2pfc8m4v!3eZHKbZwP-HC-hi^1))_s|wC-pzN=2`tKInDSAH9hN
zqFk08jE19O$b+0{4fAEAb;yl2AP@Q&`Orq>N1viX^f?-ZHls1<D|8$B8cjgk&?K}S
z-HCRhyOEc;Oh*N17V@J9P!P>UBhiy+6k32rqZiPv=w);pdL4aBI=>?-oOWTLAlL03
z<;Zdc+&NxHW+33m%yoKmTv@ZHOsnG<>2?)m`m$V(d|y_f#}$}8c6QvfTAuuDkFO}v
z>CXwwj=SmLs!Ej_H%clD_>(f+-XxcIq^+PhnCtV_Y;2Qd-QJuIg~9BEb~WvGds?nL
zU<+jW-338gmMh!sbp>o$&Y&~X;|v5`f%djk*_bEUpHrCc@&*Gor#H)%BEJV*L33h!
z&0wxGD694coo=r!nCr4hO!Nh^eSS4X;_LFL#aX>9ZYvD9vTV%nm!%|9i8X6>58B-M
z1wOxQDBv3D@;f~?eLv%6&P-1sk+;S#-j?lld9rNTeqX*VSX>}K$J=@r1l>NbENh)w
z)9)Hl==QrDvf6<BtpDhDdIJ*0e3_V77<79AiTdw>cw1`kl+HaJU3&L(bV?r3Ke?x)
zXR0GLxlf<&y}HKNBq7rDn42jPWu-N1YOCqgyH}U)T^*fzCim~(xxY<PT@qN1u%v<0
z=5YstHea?Sp(HW0-8nY>FsSwFD~cMQL_M-;gZq?Nd9bj+;|gE9q?9XA(=|$xJ~JpW
z&=N|HTeP?)l22-C$HnIH<>W{r*HlYbGKzlOLdQlwGWvfcHM?-2;%)iPJeR{<P+L!d
zGnngfXBg|~BY$c?Xh+nTZXDNmTTe-^c-sK49LIQDiaRq{Q=i{*IE|&^YTD#qsmGET
zN*v4hnoUctG$M}}+la_pdTjKwGAvt>fAyqv2OQaQmgPE}!JyaS9O-m>oEaWhj11F3
zWs<m3_9HRRRUC+G+EmXcT2|;_Q-?o}{FJE9K?eul#%akkPQYjfy=l#wH96KUzfHER
z9rVN;SFop=7Uytyo%t?@gV|IbRVQ7(d$h~nOMZ}Ov{%WVG9Wc|u&lwJRhXY&Y**8g
zQ<87!+_QV<E=+f3IWt@yw<|mJTkiqAQjDK`h2AXfm!8S}`ZLwz^aqS7sW)g-@-xh(
zQ<HDfm(6#M(x&#34W=^DD;vw#ru8$X`9sr~)W=x3K%1|B*Az7=kdviP8Qi;TcV;T~
z<+$~EdiUwpJ(+0%pOoEBZK|xTm%cKuIYmVyw8dV%2lnnNQDHu>Z=^3rVq?xLWlx$~
ztIRJ&XE3y0V-fiwY>CuCj{cpxbxs-3v$La1zs{ZY&F2?60-3okE*Fk$zsse^x@$_w
zxn(z_&0Di$=!DgdcKB66JJysEoJlF_G9sl~v{|Lp=M>-?VXjjyJefYfGwAat+SD4-
z<X3HB{hAWG9Qcf~y};?0j24t~TUN%^$LY0YxU`E((5LdYJEzbumm)49E|0!ywSMh9
zP&wG=cToD++S>v_zxt6~aPk-N^t%fDu7F(mj7Y1Q<=kOm%2cOKW{_f3E}U8B;%fai
zN%PFS0-xI(RA*a}%bk-e7dE-p1>H`Md$g0QRHChWwyl@X>xwrbYAbSkJnV|TR;_TB
zOSChXnJd>w;-~F^I+5Gk${BMtx5QYBoFtK4e{F8L=I9qO?Ha?{`Dbj8=y`oX*-PI@
zca}>EaO)oHD>Guu6+KZ!PzBmG%;A$Os$Z53NFtXru2K_m7J5{bf~%m^7B~QMq<uDH
zZDwK@<+?(32s07Y^~u<Y5>Zk|ue^s`JJf!gcIdT>bm*Y#IZdKw#z>Musw>&HEO&Oc
z%P*NHbbT$E^DOBV^yRs{6rrkcEq`j25kYgcW(2i^Uy(SfMKhg}r5#yrfBB=P(o8m|
z&UdMOWy(ojY?DJKC#EG0Rk8%k_+|NuydIyFdV0+yw@_6&vYnZ(#7v)gf&^q`WH_~)
zB)TKp83;Pe-IZ8o_<SC7txol)OUeniH_M$Vr=OhQ5*3M?R8h6_!D(Y1@|S)jY;hUR
z%sh!lMlt(fuFISuAeZF~ztdmb)JCb`3x=Jce%DC1s%(}!=LZU0neJ?NrhaA_l~p}@
zae67|G$SJ0_Gad);wau0bmr(%SNHsUx74Szn(9$fah60es!>Cvj2mXIsFy3#6$r?V
zkR)E4Gm}3_A<lRANN3QcMJb75TTV!GrT)xZM^?p(B5iirCl0d88(N_x7kign1Ek1t
z7Rs;gP>v{|H6n$N%13HrMw!E$`Eo2XogPoITuCIZMz&#Qwc;iLn_l`_GK*Y399&JP
zv}8F{Oim)hWVr)*Ql~AT<)NiqzAK+RE){&q4QBGXyuQL5DZaJR)uvtN$g^^JXjjpk
zsv@fv98u_$Ym;^<QCAtaR~0#OqJ(dXMPyr!))|r~BKFgA4n!OfsT{IE_^PxrOsZE7
zkIS2*uALGWbDlI(-8ssgUzl&xXRx+xBH|VF7kW7reO^aEt`QP7>vfoQsman_E6R0C
zMmLYCEs*Ogl$Kwvb0n8aV=Xp06jGhkFGHlZikOpJs%KK`22!<Fsr>2=O5sF(fw{Cu
z7C5tLZ<aq|W~mSsIej(ZF?0C>Qt?phFXeE^98%gz63gsbp(JHjAn20GlBWEsHmOBd
z8k15UIx^hOfFq-jOAlAvup%(qJxb-06sgs^1A#)hCacx=OtrCCQod7iv$1S~Y-^-c
zRI;Qp;FM~&N}2d{x3|C}1+mRnAnkKig;6O??N33=9$b?gdgUsj$dPKPYP4$&GY*k{
zx$mIrL1sph+MU#JgU$|P%t_2_&vkhU?Cou##hJ=K>CDhi_;S|0QheSfL24IVt&4qy
zwjw#3xD=>5QS`cmR>##gXtOmnqc%2PQJdhA>MT7pE|02R{#O$~`bqpQI%~Af6X|rJ
z9wNsw=$9^|Ozm3M)7LLqj)K<}EL2@D;d`YL)0XV<QKpgoIO}|Vx3v6arCy&mL0?o>
zruLFdl}|<7C8w#+qc<^(MAB9lu}EV0CM^kRNk=USIV1md5*pdYE?1YZDz8!1jhFK&
zAiXmB9;w2@>=&VfroGK>Yi6^@+w6(MrB<&u#Z@!AY1q+^toQ!wxSN?wrDW{<-&?O~
z+X6mmow!Cxq`54`s!eMbdVPQ8dKI<7Ve34~nJ*nkviD)pP*L$ZT=^MU9a=PxmsXNH
z%TbW!_Z4(#p4c*;BEzNiZ0MKX4%tqRREz`C5g@e;RTy{3^aSLhl;zS}>PCjKWCrV5
zWftP)PEb27OQ{n$d|pqnU0W9$DXVC)r^dK0*qeq{qEa9;*XNd+l9We7A~t5TyS&=Q
z^+=F&g3{;a@Vjy(xu_R1Nvk0`B4-{J+EBF2MfqCpWYF4|p`A9*g+6B6)o8|qwI!rf
zr0vIP9iYAI3}~4}wNLD0EhnmZ<x@YM(ukR>*4x~AL#+I9Gn*|Hi<X(&qoms))qu(J
zg<UjMb3`RpiI$D6^cIah><E`^$!J&c*w@W8f1xXsU82WV4x-u0k?sbgKe*K9PN*GJ
zvY8VeXRV+%nmeiz7dyK4eCaNUdN!IFGmR`)zNU~ZY7>aoSf<Zx0fom_WvPFi8_W!%
z3Id~4wuc&0QjyU1q1<gMdyJA#>bKIemCL#>=yFh~l$9g8sIu0KVJJ(Olfoh$HdXy1
zIo!xGYU9!q#>OR0rIM~L95fZUx~M}R`)aP`OKrcTy>2uX|3j%?a!1T-BBhvdB}`W_
zNmT97bahiIqT!=jqH-?FRZ-2P8l~y9HPdO?S(~F#E{j_G_9R=V;p;2ydP8)$f@H|e
zKWDJTRXCEZTH0e>NhoKsq@sG#dC5FmEt!QC#@lj(!Gb{hq@*0FJQik1pG|&JuYq!R
zBaoCL70<-*vypgEVLqktV1-e(GvbtQNMdhOYmL~NG@Iq_snoh-mxHZKEwd>Cy@mN1
zQfy0GkJYKBo3+KCXemTFYqfuNOP!>3ysh1^cw5-@MXf>Azse}5n|QNZHz_lY)(EL4
z;z>l)E?ZK$x1hCT*YN0O`3mK(m_u!*tSnN|3!%1dn{^i=4x?(x`O7$v?Mh7~3LfdD
z%QG&ytWru4<Mu_OjUye%m;Ng`)Y|qaW3<k$Y*&#?4b81)kWP<4w0Jr_1-Wt>m73{V
zwTu?ouyWHpuTx!qZuUy-rJXOgB!WdQ`r0MBH0Bbd?PxXtxgE|AMk|rMlAVzMaSO-q
zlbutYsFppKTQ+K{inT;qt2#NQTOr!M1!YKqD`*URC?n^_tng~}dbwmVc7m2Z-9^PV
z)f^7`5^V$IY^K`a<hOu}bf@D}?nSt{iLNgw@ly$=UDebfqv7K#ls2L&fMcKY%@bQh
zOMf%T$qV|rNOrkH<5h<|(IYADj7*t;bbdLUIntcSk$LIGJjKN*8?c2ophcx*@m;bn
z^^~2dl&@uUvfA{{)v%+&RmHr$&J8@Dx2!n*>q?W<i{u>et2=9I+cvqiuWj4raTd!j
zlB?vFh4jMMy7n0$%}~`<OczV|RiA>~Zp)P$hAiWjrdOsf5R6^PvgAgot<j<ClKUWf
z4^*h9wyXq*ReVM*9isHV8T)U<L7mb1{Q`DfRTb$R);osFwL@7xhn!n3xi2k!k=~rL
z_TiKrL}}gi>)ofb>inb-WW<EW<CLO7?M6i0)e-Bd&QT+#HtC1U;<7D!PSM!UCEdMh
z*puIuRG5_KbaR&}%QZ6Es*D<x(?7~RGSTPHNuu`DvQ_h|5?i!sb9I|2`K4Fdjpf*M
zfNf`Ze3@6bRTW?POLF08x#ukPrLY67cFBlGQBhH%?2hb!s?GV6divye^z7+OO-`b~
zlw2N2YSG#=y6x~bN#0<dv&boZ(W)0HOFMp2*!zkC4wZyj&J2)8>Az13->{_-WXaen
z6BM`|>S9#JMJ{R?8@6*sfg2?kbF8#09)0X?O5lFJP)3MIrBWg*$?#u}tz?0diY|qv
z)Ee5I^vPu;r`;?9^qYIEHLQN6&Li_KRalWt8_c*zWoaLB2V+?5D{6%btrGF$U~9#a
zey=N(6~ijcY95ERC^c`R$}W}?y;%1&j1&vcfZ8pLQtqDFq^;#uy>fc@2zNz8aaQN7
zYItT#mCIM83LX`Axd&>rek}#CEzX_jN|dUmal>05BNp1xrls^rSnu4Gr$%O;ffetM
zNR#7K@=l-`?XVke`sGKi_~t#gh$K?U6fGqqlg`|LCHX9?ms^Q)sm&=)jJEy~NgG;F
z-BOg44OInIj4TZ|tp;oNxUz#ZjO5-+ZZKv`UcZs3)>_7KXrk5>wN+;x9j)Lm^)6d^
z4~21y)3_@eEuDI%b|I%q85FRo*hzCoyD=(*F7!N`sO5WZ)ydUV>xhZh27-iVeie7>
zjMg;K$KRNF-1=K$X-7d%cR2vYflz%e)*cnh70Pl?xBSJIR$tNXQWN>msbQ25)-p?C
z9&tlBtOSj63%H^&F>^sEBIYFG*H{ioMxhJ~(MNcz@?0wDGF~pDkg>$Z&9$(l%zgRK
z+<~_BH-<G5MgG~1ii@QS$})^7Nn&XZ{7%wtg(tOa*}7fJ#_A8fN|Vq+s#<NYZhrOE
zEkoHX?7E=c^Ojs{b2tj5jxN_PchKR83%ESl`WRTEBX=v9B6TN*bz6qSNduW%ES&DD
zI;IvhW>Ys~taC^eRINt48d>kL)6zCpWZ5p;mhpXZJ`8kv3SFK3exE=5io+cCgoF|i
zlaWSpZ`>GBDHlgc5_y_H-SOA5pt+E7&nO|GWV}Mj#)d+bCH^vVh><!}L$n?`^C#xh
zZ}q4>XBO4U4V#0xw9(3mQt#xBcAVW@8A+oT9ICD$0~}N>T&>4QH_Hr(lAF|cLy9DA
zCMk1RRLY7Hn+=PV)KZwQ#7yP{>z+ooPR*Gt`y6A8xG}t%IgNj2L?YHJaZ+VW*y>p*
zGQqkq7~Rrh)pL|NFx}{)$t?7%L5=cUiQDh<QuR-iLHGd~ysp|w;j4+3W@c23<WOBG
z;XORYf#J;1#uta}Xp|+|tH%P=s5aT>fD8i)I;4~eXs1+^1sJ?uVr26`cOaFUzveR?
zap8aJCzo8Y<bF)RSLn~Q-iIxtu3E9iseOQ)x9#ZqRYil|wI6=>#pBUOw39m4Ua3sE
zW8}-%8e0FpS}%D@YXOJq%W7<drAXDsKZae8LN`Jk9(jhT(3#_M$VlShGSrF@UyKed
zqrOsB_1UgQn9?>^WshZ0Y0NE##ZXTs{U*Fx$Nwy~-1(%BF%ZC7x?3*da-ltm)`oe;
zkxh(?ag+-I)oi0Uw~lvKQImc)HK2him=uk%8^T((Rx2w<Ff>RX()?c^5GH3wsCB1$
zak)fEKT@z*n;z@(6n2f5wdh@G|C3{KYIcHanVfq01wn@z5b^JGxwN08G%yDhXe*aa
zeHkPk(SdK>?x`6NXJlb}A0FP2{8-A?rN?QcQ?BFccD3b9I@z+?+_ohF)a_%*T5?-~
zq2jt!5g952wJaBrHgx_v^poU26gfHcHhl{b8`q0HsX9oBECr@?QCs3rddEgW)Eh~b
z#9_Fu+3T%sf+~cz&8TUi9VsKh^m9(TNSM7XCs+QN(K7Oj+LYyL5j8&Q_R4=HI>ttf
z`)QWr6s3-m67c|Tsn1_U&MZYY)if^Tbj@qeS%sd!QdbHt1E=r-EaMgcmuY#}Rj&!?
zNiLNzE=u}5s(6SP1zy(ESlS~w#zypuQ}|ldHBD7D%qwB+wQ3i@sB2aCYK?w085~_G
zH|PR3yH<<fK?1w}gn?At^{RytFRj3}Y@YI`^rum@HK?S~?h=>SvZ_$(n+!!lE8+Dg
zb>xJxY?O97cXXpZ*Ct7)vPHy$I-wOCM@cK?89c5%suuPnj&UUQCS`?^UR$P6b-znD
zr~Z(gkyrIb8s%BsNVnXh)$_1S4-HT=9!%k}Hb=Hw3gq%0cU9rh(?CYKODoan%TW6*
zH@#(Lak5z%#Hs!;o~%)$<h6$w{^bT;<SDI<O)_emlql)b<>>_3Ww}ExH={>UiIFbO
zXc40OR>j>R<2Gep9t4omJd_W-{ORK9UdyvktXq~;Pi}-po|tLi(gp&n`&k?U&SupY
zq{axyAM$*?JS!y?0{w}?(4!vR>Hd%>w%LO5tC^RKOr&>@(uG7mr1`0azvu^)eIgl*
zT{`(avLx~;+9%5zjnloHdnM<i%bCw9S;6wb<>g^<o|e&<r|qIA6}R{Dq&Il9G)>EQ
z*m)32y)r?|_W^lVf>ich#sfUgY<bXD4vTs!PGwK6Jc@NJ&5Ra0rgG(1(F$n8VCa*P
z_^C`}tR?KsrFp2Ts})QawKC2Ny+%kbl~RQV=|bm~Dlh_tQbCa+&HZ!bR%O>dY5L}^
zGdqGVIenyyzr00=F~8a`>za(*9?-XvCTD_ng_V(-@~E47xrLmkYUDoM-}>>ejQlD6
zF=tLKBkyGw`sGi5yv^9fbe@_m^2;+$+NB{~YoVoE3NEd`M2eP7>CcgNXnMRYU1}wg
ziu!|q4mlegMLxfr;Q@Y>K3{U6BggM6ERcj671Son-DHP;T5&T^I}9w~8YM}obs$=Q
zP<bUi&*ds`FvcZA@^c<*VnmU|K($yTmlvw1v(hcer@#3_hC7f}T!*DpQiTWCiLgth
z_TCa%8P71=7^*A%?DDi@j-)sl#3hZE&>$-<ZyPmEZA_cIlSQ5q(jG`OHlr7v6w2}d
zx^-le#KkzHw3Qj1>7j0OOLUERSq{70AeCmbmSLS4GO$RcQg}2&hu0alOIopOK96Nw
zjw@b|j&?7Ce5<aL`ZIk-mXjh|l0~YmDxzA`-`FyjaZ9s84rVDm7}mO!ye-6h2F$F%
z_t&2Llb-3JmU)NT!!})_`eoUEx!Qfvwp<k6;I-7w!nPf%E`{o0md%&3`-!Rnog|I^
zgzTX4ux-FH&@)PsNJCT}T#WfRLMTFJ386L;2p`C094@U6c8VV4Q01Ju>GAI>bzv#2
z%6L_`i<PeW*{?lO80}aRuX5J$>|?d;Y$+_M<;B{xdF=~}gE=+I;Y;?KrG*}+i26ju
ztAXK1wnTN%5_c)rN;w;2b-$XsQ$~Mcm#{iwL7`vnkx7{wrLriMLw)3$swI|D|5Vk(
z@Xk9G6ROYZtrl_#C}00%$!euh@DC|HX<E4Cu{SdvLWk9y^q)@}OE$4qWKUh<n6=A@
z-a;d5>o>1MJ8zcKougGxjWfwwKP_i!|J#gcY(_u9%2UCOlD#N3jM}PMmX@sP8+F}M
zs=CnzLZs+#J<*$>ROsdXXEIt+E2#8`>8Y>KgG1$Cs%om5u4V|>;)3#WEty+ZAn$LG
zcPGii%{)XU5ACYDd!T9g&WVWHV#L8WGkDp#fQN5XwNstb>Y@`GAfRe^mMl!AJgk_v
z6u6PK({dMJZOVO$s-ah*`mZrsUu#cLqg2&CX)lDLrEV*8Yg}H4!5;cuGEB(p3!8)6
zkfEFuc3&Z8^}5`97wfDrPeU?ns3?_+CY5<9fVJwH*7~AGXRYX4`W`95EcJUw2JavW
zH7sKf%4^ZQ`M)Y0dzlZu7#$d@j%}Q3Do(1dFV6(%_aJiUG~gv0q0}b&{IlEAEqjq}
zd36B;M{MZ{3DFueq)xPTWl5b>uK!u`fVJ76G0*hSTM@z+4GkO!t09+cLzQ*1rNwyG
z(>yEH4jWG+hD}w!7&k47)UbNhhoI7iJun{L;AvmgYZUaE=^MT~TB_!_yu1TK3L-VX
zoDN#IfokLH#Z`K!uuqS+jvI=3VHj_FlJ^nO#8M+DWNepu-ABH3kV<ZqEUI2cpvNoP
zdZZ0+)*w_P(7Iv0_=~!nG&AK@B7P|<^_^x-j1be>l2KL@QHW}3p>=U_V(Aw}NnkBu
zd68wYjWa@vhUz9Rd0;H5XC!;PJo%tntdaovs$7q@2SdgLee$vg)jrpX`LVVn+n7+g
zj~y0TcYL%psCK1#(L<EIjkZ7>M?{_o)n44Gr>?YZ<@p~Txs}(iL|aNd5u~Mw>O#$M
z$>49dT+Xz%aCV_5+EQHaq)8%gGANYmm~o?oDx~3vjag1L>y7A$9)@zAXVF>?#t9@x
znpMjCpJeEk)@z{+&Cu?zq}`OJs_m+>D<iYQo5n$x3>7=It%%OleYGBgE_76Lrz^Y}
zpf#H!I@V&B;V}oRm{%7|QfQaSlj?R0yXIT_%q+ccM)l1Y-BZCA$e1xslYCo_J~}f>
z>NJ&lPWeXng{sS8r+38LLF4RtOFM8~hn9)W6Pq)@mx!5*+p=9w^)4cP$$<2>X}A20
zMbc#?nz;>Q6qV6{SGP^%oT7H1Dp_i1n0EBl{fQF4g{m~#9TR<q@Nriq#u3V+?py`s
z2DWq#JI(sAcGrd}YLik`O$x1FofFnEVJ>+oNceV)-3gDdF}N{&UdmqSd31TzTh?UM
z7lSNA3nr?1W1L+!I3%IPFf+wijDNgUOiKRH8l*Be#GV-#$srjY&|c6H9L30Mn>{_W
z-gINUx%5KGdoBYbc`YLIQ!L4T>-UEu3MOO6LaAY<b_Mn>V(C-6ixDd?|9(cfOP$J7
zlc@~1Nr|{$9Tz!4A5^CefRZN{<drR!I&kSdW~vSxt)S+V($;0v<+V{voC4vcv9?_&
zZ6$fj0%d5qw=Uz)vLv%aX#ul&wJz0xY?dfwJL#QNgOQ|#=k!=MRYuRGs`^FSl#JXA
z8=a>Oa8fTa3)`-=U>w@M24t{hP==ai{TH_`kJCB=!Qu+^oY|#*&6`4^?Q?&*^s2Q<
zy-0gNm+>Onp6dl+=paYQon7Rz%>&H-&_K^L{l29+MbDPTi*1v$obr}*DFw`t<B25}
zQALcg0F7Jqu4`#Np2WDPXuD0Wvi85T3A5zYT6Lk_Ca>|{4tYqp&`a}G-W}p8%=gNR
z3Cs&>8F?;dB_OOUHLtHlp)PW5!>aDCZCq+`YU3O*A-C0R+1{`WYP=&%63Z{oQk#)1
zy#$bVLYEj@8nwebdYt9*#ZVN6Zb<?fQLvQV))t!@5)@^7<&NyhC#ZTB+t{>JevM(Y
z`XDRgG|)ykYJ+R!Edk~}Tla#3KwfEf@~_-5*Gd+PcJyTnww~SlC!2KuYB$=mQqiMC
zbR6<FdiAPNd6J#6Xl1=XBkBmW=uzV!hdM&c?PzJoOPN({V<eY~n{=zW{G8D8{zZAY
z3C*$Cb?27l@BOfcM@_Ar_^3u1@9Q4T=m2BuT*^pSX)Mcwd+O}chSo_<tkA1w7ip=d
z8e~#hW%=^u9R~qv;PD!88B0QYbWD_;8Y>4$_N}y3(ALb0-L&U@${4<FmlI3gy{(1X
zvK2Tpr8J>cFNaPo%CiaDRwbVDs!RE(fGaU4Q6EPZwhc2AMCSU?y)f<NuVpm@PyTe}
z3<zya-QiL1$B&h~p?Qs%$r%$nhuPKVLF9$)6-ZC(c&gCmLQ#ux;6t-&F;h=flyTr=
z^-P)vCamD_8N*NHc@6!$A=*84IsAb_{f(~C%MBLIcc{+`IOGi_&KBi(1;^Fp?-Wj^
z+gAoq$q0h52j2LcK`7zkZNuep3#~g<UdrT`9{2Lawq<!@d-eZ1zVaH@{QpGacJjHh
z<z3fqxs9Rqu4{cg`7-p5GEeRY@+g;0-VoqdLp7xA<AVpxUt02-(LLp<dlC^2$jgP~
z#R=t#_<tb(o<xkxlf=aI|Hjm9Ri5*${H4QxD0TDY^{w7uVSafcTam&cYIL>gueL$y
z8}rlA94gruS1b1v<vS?aL!U;}N<06w7rx3DtTLU&W$!}yQ+ts5f`*LjlJ(l+dbX5-
zu<}0QpuBqD?5$UI1)r4GQj?eN;gOUq+|A5(t0y|6$GYT0fc;$Z7-?22b(gg8He$(U
zV`L-Iyn<1f`PGWF8j2KUp`lq^{&<UEv_}y2o?Kt3X<F(4|B|yyVM@74V$c+h9OADA
zv6nHtUtOomJ^0XR3{aL&xM=;7G6Geq%IdXe;bSMeNUB78T)e&9;V~A`R-x5g9f4eV
zm$|gfw3?ds6YUpyS9exM*i>TydL>o;T<)Lfx70h>T4+nkP&&D~goegjXK9j{6#7|S
zk)YBb-#tp&&61ZgjW0P!wJcs<FwT2_wHFHvkvA~%_C<Skf&6D!XxOt_mVBr$OBIzn
zTjo7W>WC~$w;>B_o6M5i3ep$FUYOM-eLO|zCx%C=7glqWRgZvV`Owd*Gg2GLrHvkn
zQ&r_Gc`QMV#!6%c_00<T9){B+gD$g*&5iLilWbkBLv<L-U2o~%Ro~en<I;o9jXz$O
zub<%NvND*-tSQO$TC0_tHwvX_u|6^?QRdc;ydF-=tn%#!`8XTn+T}{G-CUHRoqCp%
zd(^I=8W-!(PyBzqakU<XA&1s219aI%qYEdLDa&Z_B}9Gakl8dYPqq)0PQ*a_XS82b
zq>K!%oe`EUhm-BZF{4-Jrp+_%6l~SZE9{~Xg(wfB%Izgy3Me-t)Z@qUU#$mJJM)dP
z0NOx!{kvIus##t%5H{k%NI*HaEc+pC>r#fmMLn+N-33WDb40SXak)hyPYH!BALE8)
zs4TL?wTzxHYY7&$VQ3_`y|*d|wUUD?6~&18TBa;HAp+8UA2C8SqAV<bw6yZk-04tn
zxjCtftWe>SG$MsVl?=5Z%ZqmOvRmGqs|xSXjA0qaELh9E&lPF_N6GyEjRHSBkyOga
zRXA4JA3k@COe=lk^1@R6?Xu<VzhyK))ZALm?9rsJe91&T1fYGjL9(+J5AN*oiA$aj
zFu&!XzmDIgonMJf!Zt!8G_m!UJ_i{R!nw<rGTYmRdZW!|v$beyvm0**mG87tCy?6}
z>QXCRbT`D?I?4Ab<86tFiLutuQr5tGRpgTh;ma4(zvcDG%@1dBhAY<Mt%v}3`t_~0
zNNgT&OK6qYre&-pTFVkzf@|@#cc+Y%*VC#V7i|cRFV9(QkF|m}O>O;+QH4@(2-0*=
zT_46LTlL#7fmkcB*%EB_oIrlNyqMcivjj%Uh@8@AaRy|BOc}FeWd#Z|N}op?Pw5&_
z`dsQjJLM^w+|uWBN|V3zIWhwy%9z9GEiY@ng2Ie^S5{eb$t?MqWzAF=$d@m-L>{yl
z=egW|;8g+=P3^&)a+iztF>0{t*nf4t_fcEtmV&l*b>qgF>6WpU+*Q&dC$FoJClg)r
zMKEa#Wfk(DXeY6G(bJ#QQKN2E<wEN1E~$}5TPQtp{}bgJD(7LG+PMxYY$|V-mpcpU
zlaJEoh$v`cq$~IHWE78l#>w&wpIMHUUQ=n)qWm{RqFiyP0jTP__E@j_u$Da6B+n>I
z*Jda>#%M1p4OHHxh}FhDN$V}MQ10p(O;7pJe2k4w1dlt{=Tjf9a(a1kmD_FdKw<c0
zMY{;aN%E@CO;Rn9Mg}9xEHR)(C9hVciOU+KabkQtmyBR;SGLhKeC;t3UFsu#cV_vs
zunex2r&%&9@I@JQlSkW(s%cr$S%#1p2VD}!`dti_7S^hYbc9PeZx22H%U=>1{Xcq~
z<Xf^HHJ{g&Bd@)X|C@;)^9_RLNnjb`E3GWQbmEMZYnml#O3#A-s<N!zn9{bW<v|rK
zo0J&(NA04=Zfy8L*0PW>np0YwbYsYy#^n>RX+|ZN9$CGLYd*v-`GyxKd!3Q>b8Urc
ze55|mp2oM_jN++5r{2e7xp%_+mRCE5RU~Og@ohzU*E=7)b;wzyo@$nJ-X~*TXvoFB
z=cC<>iZ(1jno680WPkb;x`jTNX-f`mO)B{2$Kz6rPKMAJK~)j5wCzu>@6K#tq*;0i
zwBk3^z>qNtT1{IY#ujRS8*TeCHy$;snH%p{va_J9IPfHp*Hxr_K(wqIyfU0O)Kmzy
zXUy%VhQGtLv~8E{e(Y}K_j2|CMw$PA&;@9orKhG3kh93rUXRwJ$1LVnEoDQjdCgYb
z|A8HA>`O%ZT6RNfIRCM}Mz&}<+y8ZE8kK=Ed-B+5$&>wGZ##BJ9tCcOyduF7kk2Pb
zd2P9rsa`X)^%qfuvW)cAsAVG0UTN8xZ>H6T`)i$sGQh;4_u7OWg|KXpYp1!r&?7F|
zBB3$*l;ZMXwnD!&TJ@fp9I1yIuQZ4cy*N{U6{C#q-~)HsJyf~ijpUVK=7|v&+jd$^
zgT4Ysfi*4E%LPM`(|%{`>N+Y_t+2I7D>Jm7JnMSGc9FH@X-z|(t%zJpLJM1*JT0Z>
zl2@saHA0WTOVdJ@R`olFJBt^Xn6==Z^0ry^aHb``T-aqU={KetD+3#$`FxyKO$^(G
zA$mW(`T&SDy`A#qX#Hgm)@{YiBie^USk4?!9=3kEDz$x5L&l8ChGjQmu38QBG7q0o
zPYGMQstIVT*Vk;kILS;UDLO*wL&fLHE6oo5E3ec#{v?e`-wwT4+GezFR&MG^14-WB
zVARO8&6g|5<iRQZ70l`&nd>n2Fwv&n{~#mu>YdiPy{(I6KzW>mVYGaQ#3S!FmN)Vj
zNK?W~?}+72ie=qda_RSy7rdbpg-ns_%a_NQyfR*0-Ctl*fzRX4)L#XtM#?kMde1Jj
z2K{Z`GKO2`Db_zbEOX2Gp*LjYK7wD*#uPf7gj$6zUfoXR^(yAcA5qv0>?vR1<!m$4
zB-YVpt&&l_0rhQNtzH{?ATP?6(j<=B{AN8k^yU?{U2~h-fmU0{DJXPkUng<MSPYN;
zg=xJ`7Go=xFPd1EQM(nf9aRsPCLVokbj?XNl!eX2(Lak4C+!cWH#KUtBv8bf)H24s
z8l$ZuIX`wC+Ah0R{=NtKrYG$mV#ukU;j{Krl25H4+tOAolVl6SLbXG)rpx^Q^ZQ9g
zwVTM8N8VC*D&sa+>9ub4d}t_pnd=YL_Svk;VdihAYTIP=i0wyYmaw75mYPqQO$~WQ
zCv-TYRQCU0ODFWL_sF_ZiI$!kk1DmX{~K<>nGFWZWt;12?BeDBr?-6~S}x^IF8%JC
zrRcBdH{bu~cLsg>3+XKTtZH_#^T^4LDW*XnS4-*i=E)+4b(^(wy?;AYqFcJ7!md9N
z{ZIen9w)8_5e>9ckc?(0jHP}kTak|$#BS1AiYX!$ttDrPS?QgirG5@FyyM*-dM2U5
zt-$bJPk9%XdM$tHZ9z-62>)iiYVcY1P-_py&cobijB?hRNw3xy*7S==Vb$uhY_Rks
zEn{?zT3KmZi`lNU#30nHjJey6Y`K~2^A~rJ@4#hb$>TFa<iXHVlR+vaWwuMhYe!KU
zq?SmQQMAzNEcLipPKhWj*zlN?-sCHPq+*tLW^S=GhqMM}j0R$O{GvC5%m#4jIjZ!E
ztZFsM1BS*cDy<_H<J5iN1p3mQ@{Y`=)<Soz$_(0VCL{Sm-^K|woBu<JM=2K`yT}Gp
zSRL51SD~n>_edB!C*9zuEImU-ow;^(%2pu4Ds>x2^lvfV(0nnYrP&?xEbdUD3`v@*
zhD2HI^RT2XrJa7t612R{`jQcg)x0;OR#x%fJn`zNZ&otTXilK=cL;>VP(R4Ya#U6R
z(0iDqS4=t()sNO5HA@C9qiocUdKq`mqW8VYwaap1TYKCrn=ePhAlk-BDdUM=vr%kz
ziG^)g#i9Ifqu}GfdR#*TiK*QfFCwt+2rprw@e)2KbH{k^gLG1>my2Z=8<EwzAms|<
zm3nDt+r~S6tZ#i_y(QoLz*LDFFGEl-a<IlPkv=`{!OGhl=*P6{<L+6;`mbbcNnq=m
zq=aNZ18;6HJ47Sn#u}r?k6(DTiuP3@>w23e%8h0j3ze9km*w`yX@3Mds7ta8M3c*a
z!<VQ2jQZiBe)5v#WO_I&@<_V8I*HfM$U7+Ht4je}v~|#XCbwqfvtta`_RChZj#T-~
zw#rXN@0ot%LhJpk*wgCvV17XoYj&tET9-em0zGWj{IA!6wK&9jP*RVmyfoXPy<^I9
z4win+zp8b#x{*o+4lDnR0xZ0mu6~gXs#l}yFSe2{!Jy<UJ<6(Q+q{Ki3|^N}1`In4
zuc@n*=_75l`#LfNUkxm<)XteH)))dK0pxyl1rwk}%cyOWa;HG}P3=>g0@_}cGYM#)
zsSKyz9pK~M=EbY@Oi+<Wvn^E??Jk5*1{>+S9d;&#j&N9OTSd|O(!6rUu|@Ni|0=7h
zyMfws;kGdmk&uVH^%tF~C(PB?#i`o#PwQentRHz4%nTB?ceJ{U;=+7yxaByMU5WTt
z<JHtoN*rwR{<A0%3GJuc5#>x(2TUD|s4qd2p?h+p#GkJYuwDW+)xR++7YnHd%E?SR
zbxBjJ)1GVz9r)7DSo6AW>FoZ0!>dR!{hwWXy6|5Quj0KT`p4y@I;P$?VabrC*U;>e
zjmm%BigJM%6J_gTsr~4$$Re*w22!_x!t)KOVCP1-R1mcn9?ETHd4Ztx@-Z&kc)?=8
zmgDhd$XGV2TsiWDqIxi0FORelsex!KR$Gi%wK{uBtX~as(>G|0qYYnUw2M-d*X!IM
z@1@Pnclz_B&Mq~1?V*0HZmACsKZzHNo5fJ*9*j9jC87S>OM8Z~b@SCY>eUD4s8abv
zmnUC}r=rllg_Ue_k5wO&DYqx(M(F<?!$U(OK^<Nhjm)S}`9Oo*?c^?{Kb!x$)YR-2
zSQ||`G%N-6(^36Lo)YjgpTDD;-QTe#f93nt#QaP(F+WpHl+QxQGm_?KX=J`qR+~{Q
z!<)lqmIok$0aYIIThx1}QL<P{pWc!6Wi5jlHQF>eag1jXwWcIFTNX2~bGAUH-(3(4
zuL%Dib&FO$x6F5dLQmqV!^PwGmU6A^3vJZiuCOC-RD9J)W!4|srn>BrY#w>C+H%II
z<+Z-9#`0ZgDQ~5QEQ7KmA}a&P)hBs*vzm1zaapl$<4P!hCCKjMJs=FhV8FWCo|K-F
znarKEw%|Q}M}T+X$UC1Z`m9o>PaYPLhkVTSmQ@y!=R7iw%GjRvIyiH2YqaE{^L+W1
zivEJ3a@Q5*iJ$UBB2<~xwxhl5Cwc;fKf-Uuq^v_GH_F?TFEZK&v}gIFN5MRx6?~Yi
zKwcG+<=`D4MUGZ2rDUn#u`;(C<3On+7aAI`e`o6?D)|-2-0IY_j&5LRj5ahSdg58)
z$A!#N?X(mdrMHt>lzkvsbwy%jyjg@1@P*RttKL>y!R%g<RFg9=l<AF)^J;u^<0Z}k
z?Y!0sHDd`G9q7u;EAUBrN%3Y$z%n`<W8I7~&;77Ei{;J?UCzeFW!#Zv<Cdds#8j=#
zm^(bG4q1`3(4<dFDtgxU>6H=jnEm7Bin+N#b=@#xR_1<?7!1MJ>ai8=*{H~~Pujyc
zdS8aNIsHRtQIgk)Rz-VSD$;@vi;yvLxa=rZ_&O2wmTmN7%4a7+=Z<-4SDTEHN_<aV
z&W{S7&lUKHib^JZE7Gr~y==ZXMr>3MbcIu>VwZK<2KAk?qM}OR|8P`7&6P5vqjw)l
zAGM_@E2C+gn56e~SM*}4y=6eQC9QI~aUSDf@st{=Xo*l+7bi2@mZ$1d{!HVQ(ZO0F
zYKhCKEiScwD=N#3=271T(0k-8siSsJy9~+W5;9I!hQeE7RL0>_Z^$VBp=v@)N!|x3
z?@D4oxO^Ct5!du}swWZHyw-!Jy%a)?l$Ipr=S0i1@_Z>#ddQ>)jpsGf?atA1$H9^B
zv8TtEe1?{CwMn`?Q(koy)aH>T@;$0i&esC%URMF{=aq%!5A9ZGhEMuDWP&V4C7(5`
zxTM>)9ZpZQ#n{jQxuQyh<hjUn>psbMa^wTHVSB9ZO-m<ImWqSi5R)b7_LBz{tn*o(
z<_+sPsL;5^u!_6c0PCj@?cxPMVP8GrBQ!j2AO|bMB}J`jY|;CpH@qV5QWknE*2QAa
zD)33mJ#;>qF(P|d-yJD0<_q7JD&dTN9d2fpJFBYklCidF+!kezRb5lxL}=sDy9GwJ
zjO-YQEAzcw>|@l<f{`uF1diHW&{)6O;|YyJklL0E&MS3{tNvoD4y{{8O>y>ETZ`ON
zlIKzNPtZx#ReN$&zT>HG>7B$kBuX8xq~7+VlASys+MIp|QN2n^hW}e1HquAy2b|gR
zwy^@AB$v7g7&=CBqD${Ufzus&m?)G~(nTmaSz6=Hu=j#TG&GIWQg642l0iu>ZB25B
zWKBkT84oYX?M%4>WzmzD>$AT4C#)PbwjwXo7{#m0<QPUjAK08cmo1;N)8Fl`71UZ4
zQTnmeTj?yPK#aYKe)_X={T8FXCSz~(z98-0*&H5Y+zat7y?$!!TFd6G+r`XqvKn=R
zGol6>Z@f>TnSQhL|6}jX!{o}c`oPMR)X-`g=%sqA?y9KBs>(=JDzwznh1}WhEG<<c
z^duo=Elsv2DwG*1g_IeQ6_Kf>g)nAuGww3jATY~d=4-aW;ISEdU}nIBml=QN51Nnp
zhVkI>coy)0#~3he4{fl`{C?-2yS;bci%3Z<-R|fvWyE{uo_p@O=bn4+x#ym1>0z}K
zS?jiTxfLok*iHYL+GB%U1%IP6x~g>re2wR2TT6MI2eD{lY#YE+FSA%wJDyg%*lb+~
zo6$1aX-qb#%891UqC>6Hy<zKNw%?6GujHx$5gEyPh-@`%SwbzIw8y9ls_~>f%jHiS
zSLQU#9mqn)w%C6f&I8Ocj)kX$Rz_pZnQz)cA6bb4DJ$Huc8TLdZ56Ew%4O@a@$e>y
zi$ZANq1-}t-ZtTIuJFc~<1Wb+Q$&g)w8zw7(cxO)?v44{LG01sket|1?4Y&Y5C)c|
z7{*_LPqE6R+N|uzg#Wuy+RI2oboNYfUYxL_>HHO~Ug+$*Tuf~D$6RT9<}T8c>0Y#K
z%m(p{p=%B%TzObv(n%h5wxIG+p{7C&<#2nvaW|F`g<Xc=iuQ7%L}R<rl9>BA9wax)
z&gfblFg9!xMn10acBvM}od+m$X*n(hk-RSHqpgP*baTDT(V)BMq4ygUjaHC5n=UqL
z_Hv%=gbJd>G>s}#WQ08!_Zz6e%wPhV?k7F)%yNw++fXv3m_946nj=5_?A?3z<k{n=
zPM$vAUOe~0OWE<$OYLm&?Aep2UjUD8iEz~Iy%obcj?H3+BEH+iXw<rT32?9Bk=k*S
zX(-4tt3XbW36+=}6f8c)R}DMd@Z)g|WDrG$>dz&D;<SF(JJ_W>LU!gSHibjo!V6+6
zOY7@VUOQ}K*bLI5%2Zf4(ReWtLYxLpQgj;{-Gu5bB2WBcyfJR~E%}Ofzr&KnPQMzO
zFibQvz9h0Mb}>T>ihb&*W)Tz2wpgy<G8#QZEvfw5LQ%s#<5EvjPXr;BnvbAV5SOFU
z7tl;rWv*?dP1WE(icfJZV=1%IgjLYnY|P0Tn-4I3iV<e-wVJQbU0_fsvhQlZy`Y+B
zYK%FF(G>7I-|@}6rXJwu_MOI+fA$+w54^EwfKflZ(|V}&&?oL(`E@+H@}|#DJ<xi8
z>;3rs>9h9tBdtg9`xn0(zrO`2TJ_d5pSbgegp<jJ;%TsB)I<jC_W7eP9zXWdspIVv
z=Z+u0W3W9V&~<K+1<;Gf7PD89@K3j%2KILq;a@nncua9V)p`mr?=Hd|JAP_uQGpJ(
z4kv(4pLzLAfay@{5P;rM#FQzPgRO&UfYS<apmhKM59TnOKXdlQla`WZtC_&@(K9cc
zR2=(T`x5}*=vf7rZ_NYXow-WLL<{j`>&XNJShi?evahu-0dYEjm}|`eBHt7OEWWCQ
zG+GUS<oR`KF?(JCW?QpKfOF>+;EC1~07&%R^JiW<Z9Dk!*5d$rPqDqvL(LyQp{O2f
zJqD;eSB|}uWgk_Dnbr(M_a9ueqWfI-`V@-1KVuR*r(K%uHm+UYf(q_!wcB^z*oC2a
zCwpzm0M~QCh(N}S*}mB8jM^8GdK)I6sV#eXD!V}R@5`0w4iLP-6IT683VCiSdx?+_
z<scOb2@6)c3i{$y_EiLZI0w3AZi8q1O#uDjsq6(pKazvqV1wI!9jwSdIF&7==wOml
zu%rDnwucq;&{Xy;LEoLj%q=unjn66A{;BLB!9I`!Lo$PIyT5gnK@Pov0zW>L%@a7^
z9|FADi(o%6l|4q-JRfMM5jXuBp22UM%HB`tCv#*d*Shc)^E$K?Pu@J0eL0Kf`<)F%
zp5@VHUhV^%1f17qJTR4gF-tt2t4wwmky&S;x(~NDfv=`e>)*Bh07YhUMPyf|bLkSp
z*N|Qn1%@3h@Y&Yyqd;C_2+&>y2e8mkz<=HP-30jph+FMAyP@TX>`<|7{j1hz0RP?`
zB?8}B-@I(=|Bcpf1MI;R*idW!_13>gkhnV7MwnF_Nk99R#9wXwTB1sUkmbJO`#)O0
z0<d@Fs$?*f0{r9FKLx;hascG}>jCHfm)0)>DkZo9(Err>r4pc3-v^&={UU(!;~N_{
zxb=ILsc@C9|DpBsfUD)`g787h@ZL8Bcri_kLjO$b?*sasd1&T>QHY;x{WKu*V`@VL
z7`He`=6wbIuUh{#Kp)K^6X8XkN|P{%VL%H0U$%Y%;CT+pP4v73K%xF>>#rs1x7lij
z;zw7|A8Gw4K=Uk8Z!9e*h5NzQ4<{;SI;l_szOVHIfXZt##m0*(hR~E&8t%0IGXUoW
z*>EZ7oWI!mc7Wyeh{(HvrME<>{Bx~u1#F(128@kqOPU!6{<W=d!hCb-U+R35ZBJ!C
zP0jGg%e_&qN8&l>%2Tks+?4Zp5q$it^MPWa@`>l+B5=jr^Jo#editps3mBf2&Oix*
zDkh=#74cKT6Z2552w6E5C3>oAHcD`&dO~Uz>Dj@Yln@LV*_G2$qgW*i4XbCSCyJ%U
zPEN^AESaAYO_3p3F-5(<NRkvO&Qk9wfwcO}OjHSSOXsRY-6hl2yNY!eW~_&bFzk}z
zq*Ws5g?Xz4t}t~altkI=HB+RqdIEcQqCjB|ON>XO8Isf3V?~rA@8vUDg0tz#EI~-c
ze3sx_*_4)`AU&%klw{?^mJmV1y29L+&`MR)TY}35aoG&_#=c^|h;l&UY1wR7+uBd0
z*?*%m;giLZxasy<hc^h76#ZLM*-wdiT`YP{EFWwn^H7Vq6V*)I+25GT{w510#yHP#
zv|p!_(ON;-k56TPy=0y?)NwBItBe0)D*G{lCx(db_ZgWB+W&)7*^jVDqSGnSVKHsN
ze?FD{081p)H*?&T^-%U*Q`z^kTtb0sqs3f{Etl*sPGx_YrRIxFHaBA9#0bnyW9sF~
ze)ClJtt^{R>E>$8w)T_Itj+tzsq7Oh^h~irD+*ly6mKq;qqmCW>!z}AAd*BcYt3Pm
zUo(~cnG$Fmb)ma|usvL{3-4=F*)2d{d6teaI=-@>oys<+S=+JPeeF31f-&Y+rw#NI
z1!#<N-s5DVdu>upZBctVyU03fg;Hj8V7YH(Md6pHv;Tzf?<>ON$`1*DP{>!Nv#(BL
zzKT@W*of?RFXFn2tJa&bG@ZRn_{R#Ekrt!F72T!b0MYnNaBs^jPG{#?X0}i!fDMAJ
zm9!5|XCGm)eT8BW1vGaB_lsimN#+e4rMc=|E>BHo&$8rvp(MQby;Z|v<7LIh!U~q{
zh<?{v_fKbsS+>Y(E-bF&9t*6}kiy);@0-r%3I9-`E}wc|QP-xkIf52@1zuTXS@Npn
zk?HIaLOxnRN>iEp5t=^BviM!o*~2VQPZlsjqTzB%%v-0k2U()n&VWY@Y*F|(PiJ33
z_+mQ^_5k#Yr?W36XhKHJ5V2u6e&Ltxna;k5WgagyAh<_2H)m&4)7fqoDz>_Z8{4p~
zMmW)`sZv|<AEvSgSmFbPY7{P5v}V6QmHm5ahH~IQm49|B`#ma#@BU4z7`CRfZ(v=G
z40l5HvcYO+y-;gvy=PU#!HRN|D2sg+<*U`kV^t+f6i8KV)vBafQQLT>a;T#8MC#?~
z3IvlXnrAC;1jA{J%Bf;uweoqqqW)4fRMpF63hBv;ipHp>`7uSSlvEXG0{#X%$f~O<
z&Xg*yN{Jk+!VXk4bOL2o#iwy<ZMLH73I#Vwv7)Mbqf&riJdCz(y6NN3sxqu%Elply
zRw73jVz-gIM%DJ$PG?^?PL+Z;8&tHg?K<mP?$&g6o8|Ub@Ws^FVOAcvVS{~$8`Igx
zS-6TjrtsQe9Y(O$xjmg-N3koff32MeJ~5sB2O@gBxc%IeU4V9Ya8S|wTP^fxX(4;5
zuon2tboN=+@^EPZhSoi$(EoBe`<?O7hZXwQrnBD|4}D0X|F7xn{~`2DDJ48-C|9G~
z3jVk0?Ejpwh^^(9rn7&{B9E2U;;*;2)%?PA_CHTr#@6$5)7ht4CJ_wX42HAmNbc7w
z7ykZq_7BUNlrN`6{`PeC_gG{$(IgIzl43tKo&8K1g^2m<m?iLtJ5<Bo#qBL?(*4AA
z_LD64M4~z=$I$(u7@J$TTUUwvvFYq@RMy6u1FxBVs?JSsl>g9l_E%Z%p+s$L9Bb4n
zzJEIVL4rP+02PlyVgaZ{zk53SD=eCb=4q~JfDhJ+L%V_b?bF$Jm9>U@GDyg^-3NKA
zuvyo)H~N+f-!h$j8_T7nPwaE6ZUepf$?5Dbpv;y3^tY^LnA(-?*##`~r7<?+6+~lP
z*%CeR__%T@bum#PJKC()n5d8)ZdQ5h8%JKL4w<~9t<$QKC&$%Uu2FVS)YfVh%YpJ*
z^8;Q2k5tmFeyM0~zPOgOYNlCU@dVnYg6v6@&i?Ys$LpSPb&pX&_42w)G|}Pm5)&$<
z*>Vh3da0tNlc=W(u^gwRe%EQKH}A@Rgj(v+M4%jvet-R!3fK>7fyWaCiq{xvq3_?7
zeTua_oG27rJD||tvn%^PLLW>(vk%=NocwKZTK+qCW#7f}hZE&pC}-f4k~}LWcYtDe
zV^{Va#PHq(1_T`AYD5M6rd`=53HU?;u#4F>cX6eb`kGzYH?q`BqLliDu}Cu)d1{g2
zuIv_zq&OJdFs#t8?#f;#^!pPPxD5UZduCVm3SsLB*jy5OEpcL3cA6zVkSO7!-xc`z
zUD@$6!h=&`3Vmc(wpd0uyIxK;D(vB1*;f!YCAt=~xffmw%<sw$meK7~+$iu9yRv-%
zzOw#9c0&KLUD^6>)|ZF?PWyv*n*ynj;U;{~yiPpyZNIiVyT}TX@S=*ekdk;|ch)BS
zffPX%m)2<US9WJ#&Eoq@i>J~_YvJ>|vzJ-;{i(w2Db}yBXLe`j2>Zbl>;xH?70t=r
z*(stqm_k#PT3X8=-<`e4@{Lsail#2^&W^F%;Z!+WZ~77_t$AU0_7P&3O<_o6H`iic
zu{-+^izPd#G=IAmer9*}91FiURVVVQwXZ7Rq21Zjm4KTHxPNzckbp`4Au~eZqBkw~
z#O`dK<z`ZpP*j4;=(I?EclLN?kzfkYB9H9O9%Ydaq-x<o3Q6&(!0+9iJzR=Et@OTp
zclMrg{PyZl#s5XSvo9m?BPn96Zo(;61^<Ix*_#MnPk|?A;osYp{d=miCsHL!r{RCI
zEBh=<CCMtz!@s>N`wYR8@)sX=2xe1l({Jp`ev5^Z($cdB?26@neOLA`SnlCe%L-ch
zS9WE;M(FpYpra}MpX|#1Z$Mu8<ZoMjzPUU5NpiiBI#SV0WG>W}{WUH3bh2EkX+;C$
z2X<#a&bs#{ad3DJ<zxm+#(P-bvpf3~OFot?Ipkpxf4oa6{<huOceBiFvJ8i=l+yzo
zF3R<d-PvDau_uznbf%|xnv=F#**EOYJ~_UWOI56;KE6Bq`f?JrF3S@at(R@>&TgXE
zmH*!L>VLO8`#B=om&~`=Oyx0D68^51tfx!bMSrld_0zkvzr)%;?P~vdOv!st`{RkT
zF3m38Ef!7SJGijN6ncNE5Q4tLeXFnDlYNX8ye9?AgYnv!SN3EdCEN#6a4vs_0-xEF
zy-eWLnW5&W<1>IL{K-ApX~HK7wwX&5_UNAMMZ!LuBBWp~F6_ykFNGHVN!A#}{lPui
zhf1NvU<#oR?a7`c^rI<CG#`pd)uZU2+>;$-fg}$p_aXMj_hkD@X(<enNA_fo5jrU+
zgw|z1G_u~iCwrKMo=Y{rW|gYScctxm$DZsx#ItX5JX-Qg_GE8k$pe)oF*&<*ViAwt
zv?qHDi#Jon%M-L|<x_jIJ*AZgrOnn)K?(l;?(6{;ucs=v0sAKRp;q{>cW3{Wsw6oq
z5^nw9+MRu-96rcosQCZi-PvyvJ}Cr+gp69~S9fRsoQ2X-944030{?V(_A4xq)EM*w
zD%?NXo&6KSB?oF@;{C<l*)IY5$|1KCva%<;PX0E)-{Kj;@xIQ-s>+r4Imv|5x>7#Q
zu?6z}&9Mcn$ItpUpQs|K)UP>NDO;EIY0g&FRqoH6u$ZmQ`ZAwP*Oqrgru>*C&C3^6
zNwhvp>Ze5Sa>-Cq((aa&kUEL)axPuzc)w*;onw5K9C9TMWxdK4yK_SAnSh?qS2>d|
zUFD}NY1t$`$`Vm4vxIaP<&8(uM7rc4%8~1;a^9B)50u-rJ=rxXr;_0uBvVn;TYIuC
z7AR>z*%Uq6lZ{x6Q%0hX3U<J7Pc|fciSjekzs;Ycq;2lWHd$sqT}L{@kCq+m$p$RT
zNia#D&po6i`g^i|MQv^&&ht&YQY8#P?Cr_cS+<1h#f9+YJ=tXxxN_)+?IP{gp6t&N
z#jHC;o<G!H>3VyjyjX<;uA)?h8}6C%x+ZtVJv9bNg`dj0<m%-pN*r?w<t29Ro_n?&
znRS*SSB=@M=8;pqHeJQdI_a9_SSnaxU3Jfum$o!a<gnXcj%U2vuA(a^b>2NvUVqAk
zHwHZK$gAM-1n#`Ka@wk$dSePYjWQUjOjy_6OnHrE4!(-Is@;55b(J{#9xo@asOhZB
zuRK3dlaobZ`^`H3<|``eZg#hj(}{uOkoNVho61M){;MFg+zD7A&6BzUE9A|ofui%u
zIs_jpZ*gIDV%>tZ@-ivs;2ZnO;qz&-l&9aI^V5^%B}0#oWdG8le`Zhi^_=W>xykO6
zd$J$qWOu+J!!%u(@5Y9ed0NVgmn^8fIF=Vpt*a(2mtS5@TF$OT>?+GE=5jhq7h5|j
zY-_e_E-%F8HJ2~Fc2?TfZC7B2E9D};CQ2>Ds+*q2pn_`GV}~lMpI}*5P5&gTvxAi+
zj9;WpK*5-`npd8w{wEU>OEzyBrV35Ca`Vb5g`{c$SKaDK)^XKxJ8n7W?dcTdWvjYJ
zEBiXJxT{ts<!ij^#*JI*`ISkkaiuH1Y9(E{@bgn~s>X@df8H)ip$JpZst&Nr!0Ni4
z?N+rKtd`^CsTmEVsHVDNL04U{Vom245-C1SzO3_$qEb8+%No0~8?Spvk@lF4y7)-t
zFxvV0?kg(s+2z#)OT6moq=FrGmG`G6qGY0lUiG{(eyvyCzmqQa{K`EgC8-tP_!9X=
zUp3b!SoisrM~cYmrQi6nPJ<Lyf7O#n*#fY-#_BcTm>Nr#fz`86`AV>QLa1B}R<9Sv
zhJ#d7T)rf%7Wzq7h1CK%!NRb50xn$}`W1dkj+4v7YU_E-3bESC7_&%xqk0B|^t<Gr
zs{VgK*OJxT;jl3oX0_P&@5w&JC1~|Lk7Af@^Qm$b#rN*X{t8iixKcf9EC)_>AthVT
z%jhqY1z55D<vrQ=+&#9=7P3o?ko`)refys5JBh8DD}}IxZ{3r98;V`|`QL$Gnkh50
zAKjDvEk4=v4c1frQ+u+%P1KJ^c}$xnxkc}olmS0u3SId}L@XJh{?4B4?*h#>l6E3N
z*cs%vY#%*y`ozf>+DA_<o<D#5{GBUr+5Nx+Z@k64Tl(l(_vlRwV&GXd0r=80Q36hV
zT9P|aN>pTDl=tN+9Od;p;tHFm7MTLF<Q3CL%Bu%+rEKy^d6}nHb5|zcY39r1J0eBi
zOnH|l(<I$YzA_0r<yD?co9;7N4%H!9XxS^I=ag4@cKSr6$=5)t$xh{UX}(l>J5lyk
zft(<#tH6C}#*ykWXlffK$&)6<$0lDBsH#kE0x&)=oPg!C+Xd-(<b7qbI)TV%-pgkS
zN>&r(Fp&3o?$9_|%Vjc3)@HP#Pmm-)CFug?-2{mT1uoz4C>4}Iq>?6@q*A4?O>QcA
zoo75-tU0daeLAzFyvp}|m~2Y`@`*6x%v}Ofma8&ZL6}KQ0P^`i1t6cDEJ=z$x}2N>
zGFhHa0i869Xp%^gseCp#f!tXpR4I}lrJU!ZRI36tCV8vC<j0So39`WoP)gQJMn8F-
z@9;Q1pn&A{ff+cV(?!7Y?JDFlOx73YM--6!_=)l@3PfIrl4%?TGS9}G=R;uLR>+m4
ziZ_w}D9&gk#S>7*$7_;FDXzOs$Rt%vngU9IcbIER%1oTqN}zX;J4@gv%FZQ_lcoL=
z&_qhfTv9OsOZ3k5C{vk0B^a#9wc;db0=%<iYf@xKncbwsmmkYz!%{gKmrF+qmjju=
zY5|%k>5o88no~%CCrnx-u+_Pc1f-VdL$i>tMBs|8p@PcgL=nhIazY7UWtu60DAFG$
zixH3_{e?_40=L6dTT*66iM^!6<Qc>SJkgh_BxM3yY-k~Ck-!zXy5%z>m0&X67$j*E
zxQSE13G~?Xa)!F{L?=1`C(mXg2zQz}MG94Ayvn6%5%?V^g^^Mdrk@emx1T6n?xOsd
zFoA<<R}l-&XeN_H2>b*|C<Jb2c`l?#bry|MYGwDcxO*w^6Q=>G#=b5|Kf+`q0zY|9
z69G@<jwx<Emr{}k)qQS(PxZKom+*V?*j1kx7V!$DBpbXg%D|#cwJry0u~w(^u?VZ7
zCd2iVwqufvL(<?$^BD>7drR{lXti%YZqC97S1ak8)EV;7Wh?k6IpHmUsq^k!AnCDj
zfk~a9kG%2%m7*(j3kuLgN9Icj3l8F!5*8ip7pTc{1_<cHNeKkHD%XJk+*LXSDR~!(
z9Hd;TtD@})fqJ|sT+rOb7&rk=iBf)VMF8J%(nhBGOCVKdD-W9*!V`s?1+m`2(OAK{
z({s2|X#81U0h{<Bu|Q9BP`SMN#BfV7EI&Y!2gnOscv9`SQ{V!W#7A`zZ9-OqMrYRr
zG})`^eN}-kJ)n=B+iC!u;1__TTvY5iOQG2m$CDe~0)AJ?52R$0AL-qFfxnCN4pQ#!
zvr9<vF^Mb$Cdp{o=d5VY&a-n!T@xe$DTj+gVFYfHtR({Y&eUmdm9+E$a-A2W)7gTm
z`rvqQ1bhN$teEX#Sb<KHk~+&S;1eE~7uX#g%9j#%nF>J4?jZeuz(1Hgo~|7<(b;v=
z$RwJQ^YK#R9f^H%t#<60Q9)2~0#)Agm{1|wc|AVTDu9z7m=)mk>D}lwZUtPocj@w0
zd$qgLK@`2fO>}-(Ad{!0D-SA5u^pd|F0afT3df9AIFc>Jc60_@O5DTgb19$fJepv+
zBkuy76s&@Kf7}6mL6w#+?3gppyq@GQuTx?K$3jUU>{zJ6I$}&71)(Y_#&K>2flKlt
zNcbQ?$&qNXN67mmK|T|Oz$E)I5eY274`OH+&l~E9n&fz(U@1BA=*+cT)Txx2=%iy6
z_Of_dDKp`@O@W=@OsK$3e&kfZOSl(>I}2dtDOrISf9zJk-s=K;^4eNE)Q7U=0h)<8
zC&B51f(u28aaB^{{f>&nw4sDd9_Aw`?=E5~h%vecEJ56pV7a>hEGb&u>M-O>z;_Z_
zCI#MGdgrmS{jE_5ls=p^E=sD{>+eZ6#ka5gsVAl$*utfb>l<s>>9m40vS02rUfn+R
zzymXn9@riY4_xf^4|Mz2YMVE=E)V*9XKJ|026}#Bduwg~b9-^E^wQ<tsCHJ~@?E!v
z8(`r^9@RQEa~rkp?bNs<S3}On%Y)UCT*<8B*!wu}&DZprp!vO>{%UOlsd6^A*O6x&
zce<;jqtI}$QCsct-dNm5QoGn$!8MOSHrigfT<eT#FK){mR6EgG>DGwmE6<-fja#bs
z@*exwl#bVwtX_X*eH#}bMkFZ_xhgxT6zQ4={m$^F;O*F|_DWSs;5^9Q+zct67iiwj
z)Wc1adOf|8ZR;=y?G4;bvJ^L5gV859HU~ptK)=gdTv+*f@8YlnWZ*8jWbPu}+Y?A|
zfAgmOer1Ht(&w6yXRol2*^+>zrti+;=U($>1^0qqw$Hoe(1Hf{X0bhrg^_3A&bnf9
z9p(tYmALxQ;LSrEGy9;@i6``jB2y`4&Tw0u0=N(=gHr*ww|eU%Bps6*y|-MMhTKtp
zY;kGv{PCssvE%2Do;!JV>ExNyyzbuIyS{Y%)ur~C)2BY#PJnrk!va?da^lpP#c^m(
zo<6^{c={+V`#yR6)Uor1gt<gBc-Ij)8?ZXs8EWJ9uiiQ@6xzF2?lBLN`}=ndy&GLR
zt|{sN8MWY3Xj#4<mCIlXn`4kh6AxKiz$ld=#$e1-5+m7Yn=7%}a{#kUkyuJVFALM_
zyymDq=&#?b>jn3GaG3wt2T(+v7{uHe5W<AFx!7`>o!(Hgj{94oa<8Or+*ll93ZG%G
z$5COQGmsZp`BPr+#hvr5;~#CaS!h+Am&VoQiS5Yqd~7d~ADMUOkDfUv;DQc=I#P#6
zgf&Bs_8LAXks7SmSx0>BsD`WM?LKR5aPbOmTrj$m&oA|^b^A4>X2%7pNIO4j?v>(a
z-2?f#KlsQ4jyKuesU5MX-RRLepeOZnbbza#&=?^E<8FHc+ICbMc3<Bnw`*GiQP%RB
zh)3NmCBv$=BQ@kDn6Ggv=pZ-iLjAJQ%xh50roF6ckx&$E7}fBz@Lp(Dk&4x7_-5^x
zK?ITpwFsG^T1XcSaYUjpsf&Dj&JIR$h`Y01bFB%=UA<mfvemtbJF`~v#cZdcd9^k|
zu6gx&12H6P^P<#+(M3KloF4SM^R`XGqL?xBwSA-nfA$%gblj3x#4Z!G>BR+Kk?S@A
zu|2S0>q)-0${12hqGA|Ix4vM_{LIwO40~AKfrIjPuoiZHTwME3((wX~lRRr)$O|L2
zL=Cl-2C2cUAaQ1(5R@OxrQ5@PRNOTgLok%HxS&+Rt8Ydi%IGD7(foU(*||8v-343S
z##|G()8kV6-i^kbYnp?_-G{XswO+r*nNvDFYS>)!()!?HqrR`n8Mq!JfEsal@C8#m
zhA{4c05fqN+$GpB>tHZOxuKI&e78T^9ulB?vv#?2O}^Ezkb;?okte+o;Usyy?+sVX
z5eW(PEsr@uQp6@hTw=cw6q@5S3>&Tp&LV7=$Pf6)@7Rs$tTq51x0Rs$pnsj}><_j`
zki*ct+Pyilvy%aj{~dOEqi*eGWdH6S9}WjYGl|DA{@2=U{g$okb{#F~uER#coL0No
zt?}lX9v-cV=}Df<*SgJ1&Dn^39+~+zd+THUYKXO2KnC?5!Svv#yB(BvQ_1c6k=pXO
zf~=LFTc$b`sh@`e#Z8ml+My#gG0HK3I#3+uX|rCteJ@B42T0M6m%EU&9iSFAT^LKi
z?Q&+9ikIj;QA<gUVfbYJ^`J&}MUI!P%v?x{LU$b^><d1cESVlEtaN80JiAB@&>3A-
z6(G_p<31kqqh{?agW3n%qjHY1dhp#S6Qew++UvJmYi`#uA8pH^B-B_`w*oD3d%o7&
zf=<GmT)<i}s$eFPF8i6Oor3xZXOr4CE0HI<o(tDlx&pcxA;EZqo`juD-Ch^P8@(TP
zt_L$7^_BfuR45*ou~jXh{OdL8g8y0s6UDPYsd8hw2qeq$g}Ma2smW@qiX~k-OXoOC
zHkui<DQc^Lo7ANO-R5PzovmOh6M+sVO<`Cp!85g`Gsn)<8Xx(}9&Yj6&gu5e&S>N+
zEugb;(Y$8NtTXC)<R?zPnjObn4to&iCZw+|L5y8>xXBTWKAAVdi_3~@8_PTXoovXN
z5I<Q5Nd;~xg`tJ563jbd&tp7Z>vd|cZT2?ZgZ&sp&1G3gN%;%)*NQ_Y?!}<H(0L^b
zt)e**7?d%-9Y{rYeUvw%WLiBv*gCnfx!$F9-(5BH>)Px~FgP~pc)(0aA%}5qO-X1D
z?WNoGSslm51jc@%TEKzHYD~2n-kEmgWH>~Ci(+wB(Tpi0u54d`3l{fL!eBXX>_ww=
zXd$r_7ZT=@54d1(P_aAmTIQTIc7qO-OECV#;<=2o5rT@6WlF62^@26;DN8~Vy+7zR
z;W?I){27d=goWcu#ezRxFmoo1HD0ixbFH^(mR`<l1{as@q4}ctLGjbPNzJ%27uRl8
zx;t_cu@^pVt<!8~aH^F_&fIO29Y&aL<Jz5UE}0H#HD=v?Yq!%y(x<vMa{ZgI9Yp(c
zA`)@oq1Bw=V}Y^O?O+)LDW!<<1Bb(qH!HfrmK%YGXxUNV4HZ?6Xm%q;G1u&kM%x$p
zg8v!3bHRvWt&_+jD>aIbrlryc->D%u*%7zsG|AI*8;j_!^C9F61p1Ch5VZ-tRAz{g
zflz>usW#joN90@88reBhi0UZripI*SW8EFrlOj&A(4niY3zj)2i_4&l@wOOfqdClM
z8q4@C*NsHNkMR(}Vur)0IQ0jPB3Sx`!bDI+$*26(v1V!?g1!Ip;;EO8w+}TB6)P~1
z3VS$!JzNgkY<dM~%k%H;bdiY(JiG$}4>#{0fvNWRaU(7Bb41R4#<!sd&2^ELMR?Kp
zHRjl2xJv+MC4FetlI1AEIK)6Q0t8#t7k21N{ZJ&R+>6Dx4o8JIi*N~#mMdz+-zz1@
z0?#N8N$)}|cn-sv6RJrxplZL`LZy2qHHq4R?+9cA_!+Oa2zCGFQiH8@KGoOwv!gkc
z#r(X=U!`;Fed2_jRL<Yk6iW(WI;bd@PqPY#KX<{7sINtS?3AlKNX)pb&`pGJlNw(@
zzh;{o%%QR3X2F6it<4tKv-CyMRCK1XHGheLpLp`c<e#W}*)=EU2c*xght;&~SYeWd
zi>!<AJ8=y*tjdvgTde0V<Yscx;A{trNZo4@O9|+BRhXnFF&c=0p913w^%yt`%}$L7
zU$VAyW5O2Bc^bBcH;cW0c}+|Wbl9!i1^U4!E?&xdp!VU~!6HW1&JM*i&5kT7ynOJ&
zSm&qGwhT-cKw#QXu#L5SZlFb?0&yy#qR?WCEhnIS9P0AG4e_Ggt!;X{O$jfWYEgaN
z8!O$-t=gi!Zdn$f^gMTO6uCN?u&vb%k^y4$hNC03TiqMCn}`O0YfE;m%nss~bP9Ho
z{7#f@H~+5HYaa_;WFM<Ll?R`4U7QuTdA+~d!N!o8&8Z$b*!2t1*zYKTi_Z~fF*@?G
zekll`Z`sY$+}?DcHf|To#}kNjQquVo(WWbNLc8bQ%^6PE6r7y*rbG~|h&hC0fWS_s
zz|%BM?zxESW!-DDN4K%4>Tj*zta;%qXmS~>w#nM>`V(G`2&-MCd_*>KO(cr|aM`jV
z>y5S_1uN^=$+KK3js-8{(Xyq?2J<GELIkT6-6IYcE8e&aO|tIB=GIN)W)V|!!7h##
z?Iv$SXLs48lMPERVAh6pr5ZVW9#lJ1OL*=(U20iL&RLHNg(aBQCSuF}JYVP&fz0|X
zWwVQ|lz667LW*mcFRa_#V)1ig-L|M<X{OJ09Jpu#)Ad}5K?~Gm&l-x_3Z)y`;RwN$
zAL_R;ixxY&i{^uU<E>S|2xdBjs1XAPB4yDVB4Pb@qHWqS;)fUuMYVX_kj6r824Cw9
zM_X;$B)J`bq6y3uC8N|f3Go1%;~~-Ta}iqbsvnbl1eW?zYDX9M^$E!%C)SRk1C7R)
zHG&+93o#2=u7V8(0tsIn_k79^nsmf=g;-L{y$j{+Ng^+rkwh#88SOzF*t^tiL@g~g
zDA##bo9YV4RXg#NZ5d=`Q0F%*T@E2BEwAi;%*8b;$@gO(*2zP>jk(DKW7^M?1mN&@
zvemr%A#ZWW^#dGe$>G{=W0?1Fz!tkAz117V9ww*VM}%Wlf=Y~dlx~^K#U3o}tB5Ao
znG;J#R2)rW7-^FyM)@VVb(3F<x;oh;vvr8@rptqE#9tuvqJvG_Tn_}RT@M!o?^rF3
zsI$R!jvj2%M74}*`hAmCp*+9(m*KoKn~}1L=wM;z#KvH?TXMf+$mf;c5K&Q4bQ$6S
zux+QI02<nDW_Im?dGGv0bh-_pdX3zkd#$#%-Ivg9GyivEM79srCU6YhnHexsI}g_`
z<F*+89qgMjHUdv`wOC59MwKx3eB~TO22p<Js6>C`ew)okK6KEvh3HaB3dqv6L?v~c
z5rKPN+)#FXW}B)bChWC&^%!*!)mPYfcDPBcS?njEiqRU;A`i{EU(-skZ6`x^uD1Wf
zF0ynPyUiE;QevkIm9R5hyTchN;CWLS8tjh+q}d7Fka&M6vN$KD%)uM0k9v+2A|4P4
zbT<*+%MHkTTG*ZNsO->DuGqS3*Dv>Q0)&?SaImgsKjKz=Ajuf^T13r|Zqp#)gKZUu
zss!NPYnwy{9JaB%_K-1`Wg#K#rF=px?@<PbZ!9qmNFW`5c&^hAP!bg=W=5BZPeCKY
zT_t8}*I;NLxmD%|IB!E>RcSIFydwdn66O}{<xV<xI?AxE1qOP@2{q9Yv^NjTuo`M|
z0z=qv{78OUBChukCBzt0P*)O1!c~x1F|wpgL|h@Hsyo~=QN@m?x#nnlZLN1h3Ym!F
z`5I1KDJ4Rh4-xFt1OyIL4To#S{z4z@>nZ@Y>xcnF*r++xH7m!uW=#YzPS<*QB0KmY
z=}n}7GNz0*=NohLXx}!1FM9n|thN`PCQ^njw>zViUT@(<2T@2xJJ`uVaZWKuPMTeb
z`(3MtaY@QabVQ##&uy~8?`r^VH-q?VR}&i%2rVO^WHis$`bR)Ln!w|`@^#_q!{vg=
zg0;S1RG8CgloAz-yAZ}cgb*YYg%AtIi-@=l7crDcIZ+E?;JLVT&F3Vfp?XiZ!=_jj
zSh2fH&Mnq(k)Ca#jj70Ob{nczJ!^b>LOr}QBTJDwGS!`un+cku%NQ`yqutd7b|i*S
zcO9WM4PAN9#X@LLkn8@uVl6mxBk3U@(mn+1KNR)OY0!9j(jgQRgn^XT#5eq`4A$lR
z6_l5Bql;`^wr?(2ABKH^e#rylxZL{oMqe;_Sur*Ee{ajWr^oV^nFYQOI$osmN$eDf
zdqAz7xFa0iqF%UL>rltaJ+0-~7W;7|@`3yie+0<u<K!5;Z4%+Ds>M9O>8>L+IqJ4>
z5TPzuB8#vdsa+fl))$tB+c+LHk@DGGN9c7h;Hfk)L$PiuLxx4NMt(=si9IAnLMNRT
z9l_8CLNWF-^e`7l_XC<ZIUgZHQBJE4`qoCXljGLq4$rTd84_zba{xoh3v}cJn#yDd
zwTRsbt3ntGiJC-FIfhvm2sp*E{J1Z|iThZm_DiD5dr*!P2ynJUDH^_+E+wQl(@a}y
zA}w;e$Aas4C!Px`rbaeJsXFGcd_0`vKKfkNr-rA?8lrz;j6%&p-=#(s&tDEzZpN}*
zwlggoN%7}JuV5V2C+-vhzAt?unC?rL=)@Ndn0Yf2Ro%E@cK>F<yehaihl!O%{&mE<
zADj<)m4n~Z(b*7dgdQ>)91*{8nT34}CA|q*W#djdr)jmA(dgCns-6Td20m&b&Q4=a
zPt)FJYZ8<Y0YCSkz_DDmzMgR=uI5tK)-?35*TVNtSmMsR&!QsoLJ<j5FZr1%=Ri<W
zBrSYsFA0`15jxu)awP~4fvKY==S*s_Q3Bh^%fUT%AGa#aGNK(dDJ1<5qnE<U?sFd*
zu@21JvNpQGFWu&Tg-yRty_Pr=I8hJUz9EOW{aizgUg<s?{VloJKC|oyjw0K;2#8Pu
zVd!meEw_tXE&z#b*5;P5NhM}u*9v!pbqU10c2?`aqAMeEdDg(sUO+IRmn(Le6+^j2
zsmv&Z4+y8v8|$5oi>sa54Jbuyx8GQv?e#aew`MQQJ-Tr4K4P`B&XE}<jlCb$F-jL(
zK#gU###Aa*9+b7KdB6oM8fr<pkekh2ACi^mM`DUNI#<IX;IeKBbP>z*gZvU)mZ4ME
z2Ivj1<IHRG?^^xDU>obUTcwC=`xB(xMq@-N22*A1^|O`KSGOhEM#&JKufYj~GYUpW
zmc}a9f$mU~gj|AKRQyp87fs}WOiX6em%RXgs0@p;s7GQt9AeVWffE+Y%A?}8-dJ=`
zH7Q!^C`+RQRq$4gF#p3#3DU4#jpN1$y1tJ8bCHS_Dr~U+2<ia`v@Ulre36E#JB*})
zJdLH5ParFi!YdRZkhq3&UMmPiBw)#c>JL?6YJ@`dB(g~7g@C86NDcNQLpLZ%;HPLL
znCEu)El_k}O}^2N#v-O1rEB%m^i(2C%1xbsF)7!AL+G~gpz{tcYrOL5v$(>McSE*j
zTeG+s@)zCjeXXbQ`<LABXIszW_b<EOA8dUPzyEjl`?=P0`27!01K+nyJ@7#5NcPG!
zE|z@cn3<=WC1XLalzeX5fWOm%hcm6h%uQ#{5aqipj7k6_gkj<;)O)70CkXW(3zav+
z74FxjvIhzGr54UOt@Q2xIF<eC6yAP`e|rtramr<ve|;+Zdwgea6q4IAIr-@OADYU3
zobMmBW$B$<wcA+=^u1HrrwH`!5GdHmQ?S1@m3<e%?8P`zS5BlA^b=Frw-EFlAw~sN
zklRz)HxlG676jY4`s%f*?Bjg(W%d=q;Apz*>$6kYh5T!VI_c~8PGuQiJErHox(fH!
zsq9^Zdz-Dr#qb({){kdz=6idmO|uwat$^R!`q7ZuHuEgE+xqg2)?dI&CrhhpDsm6U
zT&ddns@7Kl!u77{JsAgQTW9gcbxVQQ0(7QzCam3YSzexQoyJSoMAj}FnOb$$%JA9=
zk4<W%q@Qa2dA&#VO=fK0x$;H4IrXi_G@Y8>ohlE%*rwC4Kv5zDefVZRlmuD&;ER;`
zcdmRHF{cO>>`oM1^W<?kYvgr~r}C_%C*oRM7YbFC5)jiJ$dJ6K5JGwH+NF`6%7^f=
zk$i!U$BH~kk1%;vJx&G0@h`V`uVj>I4JqrS#%_3MA;)xXi|Vt2ViOFGIYqT7?@N`d
zymcKJt7myPUegn7`^vwXi&A2m9D+Xi9#Ud{@{1pMK+_qTKS-iz6D*Eyj_g;C5120|
zl^;ehUGQ3LX&V;SeC-qz*nI7!K8}DID2J|zBLy9fOj|+;n1|PX<tnllAS6bNL^<Vx
zjDIHW<L1>%yhD1hJ;a1Fd9FS#z1d?Xv1eQBve}r_Ifa%r^^KUXK+ke*WFg!%GzF<R
z>=UN&(0%^V)!ybN^AnoY9}FF{5o{Jey64@41!m^UMu-Cb=B*+3m7R4Xiz(QgAry9d
z`8X;$rw3)CBVV4)<{ML!Y-6V|MCQ#KZ9n&9#M05zO2_5U61q+fD)}WGOuufk-IjwP
z?KamqnP884%!AX`(i(&x6~+`7l8%WE&KhfZQ)_k+X9u}sUXwFW%+@En<fh3K##>=~
zZ^O>N)kR*Y_$imPtL6^4)h3QQ+S=^jB<Xxvv<3OpES-H|37<tc^D;Fm3Kg7%GGvxH
zd?WzNjHG25ReXR<M%YEWaKRpp1mlW=HNsli$d&I5ZH=rND+PN_){x3SE+X~oNSEId
zaa&gB7Ds=uIGngvkZb!tA|QC4q$Lc2EjTYq#}rS^h2x0~!*DV(U<Q9M_N{6_?oKWq
zx1zy>bhy|GRS<1?G$O4eDhQNAq{_ssgcxEyh#XZIa}oAAtCyT4p~D+7g3ZnlchD$$
zSK3nK3Q`{V`B1L@xOmu@pdVSVFdkG}#MB%Mnqu`rK?=P(P^6-P9OLC0f<Z7mloAC_
zi*MbOqI*h48ai;;xw%s$C=z0j%UWE{q};SoLI=x)IM(cK%l(}a<4YoX$&C-$LJwL=
zsH>Bp4i(OFU5Wv(`O?PO?ok~>+%a!!xIPKe00w)A>tVxv>Zkyyv@2uk&YnC4YH|zZ
zy{)}i=)U;Ru+Zglu=Uo~m*V%vZ`j{&YdwtL*WB+%Wnul6j(xwcbr|oz$-RF@7S`YN
zG5h{2T1W8y&%56Xtq<e(x4Pept>^LkJ6-wXtrPfNyJgFtY<(qu|11~cLSM$^$Tk-H
zZ@jxCq4Aw7hqw@M3l919_M*X}a1SO$PG7z&co{HL-#%2K&=lky0Lc$Mef8z}(WKAc
zhUf7Z(HCs67hm81`ugqqI*{9-rbzV6^X)5__lfa3lh5a#$n*?D_GjlO%<Vo?Gilnk
zH(`@%z~f*=U44?Ld~E7>NhM8F2#I5LfoBq&tLZ8k%O!e^X0KRJi1pW(?h#Btrt-A*
zE+Y_)<&2?VflqG0KDH2pQBy)A5r01zZQ7?;T=YlW4?@~x_iQj+xva)Izc!8K&#EnE
zrTbp8ky}*e>T38OIRzoNJrb!h6+6hxKZ|QCNojuah119|jU<D|j=wrzJDDxMaJ;>A
zrgi+ZeRS;jiN%*rEwzJJ^ELmljU?pfmQ0c8`SH`o%u}1Qx-4Iv@<ufm_RsWz7)oL1
zlcC&RlQ~yb5Liy+(k3HPDF_iR%+BH^EJ>{#%*RO4fnALff-zSx79Ezl*6wXvQfPBQ
zNQ|qbwUyMkd}WEltT3zx8wTRe4a>Ukt^p<pzK$^V-srNu=D^1+(|FSwLPxDftVgv4
zF~2*+0%G!{BgMcf;zPr9%i)WV@8Xjx#IKvoS8ngYwL-VF*k!}+C0H=noZ1@T+6Bb2
z4Q+6n{f+2f4ny8Q#0v`sLvw=M8#zL;D-Lqq_f(@C8URA^caSL}-aB!n(JQ%{F@5Kr
z1T6`V=aD491`@=!qekBW^eG~U(vcXiEF^QwR#2W6q2iE~ZN8wccgbY}$i~zmhc;|(
z(6#Qd14R9^(zsbGJ228clUcuMyM(J7nE(W~_SieKo*oB@=APucrpUXii0}D<5^Y`?
z^fZs7;ueXsTxLyka|p1ZYTIhm&%N+G>>X3u{0pKzM=8clgQvs9%v&%>p$=uEbhERX
z+Z!bU3SjrfCL$XTk6z!#=0;<)x8Ck;;NlQ^0xuzdc7GM7+WKIa>k3Q6_1@~%<@wrW
zxrYo5*7D5}Bsrqg768RkhE$nSF^?@A|Em41IFsmPb!ZTNcmjMQq)&C-f|zUD-X`1@
z@FF#|Ir0Zik^&QwG#FBe8WSYOvPBAf2Q5x)Gn&@U)ean}9S(F|*x;bi@I3)xvc9xG
zz~V6ls~Twtv7L55n%U6)-*5WXv~j+6aMEUy_CixdlF^P>K;<a{y14L1l+IS;#=gVP
zJY&}Ia@g06Y#j<4-A+G2hsEx}aa)4^XlNxy{t$P}jmy!(?B>?&-16*Z??!jMjh(^m
z?nuG}BFfX9^M_TNYD7<ZcvXoQMzL-uqvMmv%iZOAXeVJ&CU%BPa(`j27-%2s9Iuj!
zjaLG9T&Lq{QNxY8o460}=DI|H)Hl}=o}Ejk!%D5y5wHgQ*yxyCGR9_<vPA~5cuAZ6
z+~^1-OZh?0ZfH)Z`-Mg@k9bb$eCL_PLjG5pb(6H>+Mz!oO+Z%#e!9{b+_s4f!3aj6
zL%X*+GP%EnNjSox4-twA&tieq<edB9q8{)rEN-4iS<hdGWX4S6P%E&zIMv&zH3xxq
zRP|syUpZNGh5%HS$brLG>{?61>HrHfhIJH(!Q6uY5R%53X`Q}f@7{2j$%GZgf^xb=
zy(t5y=>Vc)NxG3X9IG|hpo39Srk<MjbuG^tjXw*&ufxAQEAu9zoLo&hgS*P$?h$8M
zFe-hs+D2jNo=cdd+prvKcn^%$3RYwEgm0vB^db2g74aRXWVziP@{!>a<^hF5U-=^P
z8qqCs7lZ@w?}-AH`gNdNHwro}xjZZL#%#{8WMO;ng9DNNG!?pMyf}Ux*ZmZGhbY2N
zf{^*z>?{s$!SjnV{H}{^p;Y||B(*1Nvq%GtKZODlp)8atB?#QCZ)|LJF4f7waE{{o
z9=SB+79?1we&KJu*KxO0wXjIeSPmzc0?vNuBz6CX$(RHHx%Qz03KH{P>oe;=^m(Fm
z#8|M!3qEgy7iEJPu`}Fi<Kn(z+skVReH9?_V^KwxZE^9id2R2~u)zH5cv=@>he1Kj
zq>^l(RZDV$3CMI{WZ!VgFfl)nMY*_1T3I8%8?fz=sI6);apkIj47UXo?vB;&jfKDu
zq%NW)8x4nYCM9ELb!BDfLln$*9|`%91ILlmkY~s}Zl^`jRq~J{!xU_qgK@JT5&}Ow
zi=;L}2A6o)2*&!c4Z{f+nnA}m;Ml^CU`eNZT*jeJqX@SzGBASZKMrjHc%<anys(dp
z>{sWEwM|$mLIZkX`N;l5TBj#ea%QpRgS9m{d(f<dLKTFVsG>H{G4-dMvMapecA0w{
z72*o~AOtLMdNeN%Mn*<pSNPH3DDb`kM`x26M*6EUJqEQpIH>+*c~RhHL=N~=5RePC
zeYJ+x{$%Y?m@eKEAg9Y$5h-i{k8o#%d;_Cmk)b8PnJrHCRRjY^xL)z?CpPV7GtHZ2
zjkmxs;c}01!Mt4`^e=fuRov=~uC&n5P8P>bXH4*I-Rjo5U6YHJK$U%NjRIkc8yQD<
zI@7tIn`ejJB~aS$t}Z#}nG|rgc`{(ao=_0SyUT2Ig;!j4R#(LclTM|y(;|awT{GXI
z*G!oFBCpIC^k+?pPXDGIl~Vm895cQntZ5-9JcS*`u_cKF7|1STKx*_c_(7AGK!W?p
z^D~e@ItCG7Ysq0>2W@`kGVX_kBa_4j74nv$pqg{$P`f9T6}j#*15<2m(GbKXrm3OP
z@-Uo@MT;4~*cjs~aN!y5h!gLyFpq7~_la3SG>`=8aJX9Q$`v%&!dGKv*pPlsTLsiK
ze9Xf`zCJt*3zpj_P@;~uu*Aq}pzxs?elxDKqdJVk+ILSlVWj3ghXo-W@c=b`4e3;)
z%gED^j*;m@7#Q#uvs2^MyraR;%dR${m}G=g#K&1g5!lD0VE(c4yvpEC)W8|QKEt^U
znpQ@oH;O$1XOJE3fuF}zwtU3C6U?+k!u=hiV^uTd)}yXjxJ213F_nI3W=^WX5~iTI
z?2_CNJ2Li!rDXVrskN-a(w>wv>rCju;{-DN$;w)14Lc6<+rXJIe15>Y$NEU`fW*tI
ztM+zJxqcdI=2EwB?^T2X5@9f$Q5E<fhmw=w%h4g@z%M<@$2>=iuD6g!2@8ItMaPwz
zxQrWVVTq|JkD;Wn;c{scnW4j{hc4VUQaV#x+}uRgW$js{LBvWB2ap{>cK0)<cHxKd
z!_a5ISJ4c{AZ-ZJx9o>lox8~O3m14aC!*9(5>(Gf)rMdyGdD6ND$W98=78PF1-Mu%
zvKuXJcIuA>l`yAg4;=}TbV`NphAC%BvK)2H722?VJ)Un_&@h%@%fNVTpj(V+`y!%r
zvN0v1c3-JIUOS9lTC3mc<+6qgvBrvQE6ojJ@I)@#m5;ZoSjhESI0HzKmg(pQbefQd
z43Mhkq$^xGXD9jz-!;eSFf`qmbyy~@Ty6NZMWkCzje$PBi>ZRyxi{<~x7hO)Q)72K
z9kJ}Dh$HbbEH`P@F;HRvY;mn37IUFa)0GFx<JTLV8=PElzzP<wRxaB<>uvukE`saN
z`YRGqVA6;<QiHXaVZTldV6$sva|LwRgv2USA|<c^y#G3Eu&9vpc9C|TF7D>iD`#x?
zXS~2tD-ue82o+kKYJ;dUIdlQPp-IfCHIA3CCUmjU-#Rmg?~q?zD+z|s)?f=4o13|r
zn=S^TiS~J3g8y#W#S~i`PIS!x#8#p)9=y=&qSZFxa4<BshgzLqu9-@gk08Y(-xVfv
zR>Kcl3$2IEOIjXYG01LaXmt=+wB9xskoDXy0^cSwFJlcYMi8(m`RE5vxZ<P%Bam{e
z@o=m!J!XEh9|6ZKP<xW@l1<a?_9e(njBoqa?9-a=Va={nqTvmwU~C7rl10OSao+eR
zAv9y`FiUNO5xbn;XeZWK0oZy<$%V??4#rl@RhAw@p<y|A7$Om(+|VT!6(p4-WG=AZ
zp3GNJkg*7lE5?b1JLTZL1Bfd*QjQ5-YY~0Oo3aJf6pK($2y?R=o-t6+)?)p98GcMa
zjUzvNwN`&+@!aW?r(b~OYE-8vMMNTx<eEpU`LzYNYu9jz?kz_TlEsx^tTlR<`U8a0
z&lMqb#tZFohsEjF1^3M>?buK^&w*_3N9Zw?K)Ai~uY}DV#RVethMgD;&ojK~8g9j2
zJqAEHsv0Kiy=p)#ZVfOF-`<dfV7SQ}N=Blhq!?o9`rZ9o+dPphd9a0EHw?#H{ARRs
zU4ios2%6W6KM@I{hd4sL<y)>x6Hf24A--(E(>K=Gx4NPW`?&a>Eo5$c=+H*6B?lu_
zH`oFV;wW;4M!&C7e?xrZ)G+G4zRk^(c<aIy^qojbF7m!nqz#2JnVWj~R%d#%2anWA
zyq(HT>CrUg#Kw?30d)p2riOKn{Zx}Ywm1Rec7+R7uL&K)&=&W7`N(|jNRAKUoC!=D
zAE4xTVZQ9Furw2O%WB!Nm()9IQX`W9rn-d`hY^Jao1F8`2CN*9fEGrsT~Yt~eo~zw
zghczdPUbN!#OhJp3bI<VFn$kLYzQ5C903+0H*%ncdBO#rVegd{PuUIROyntd*cGN4
z)UJ<++PIRr77xcYiRVXnth6A+#7V6&-VHu45iF6OXN!1DS=Vi<VFhmQo$oj{i{sq*
z@VNN*o<mf*9PfT->ouHWohb?aMARp|LMAJ}9&Bw$fqi8KiV=@mZl?7FDte@>oI5ot
zvF^3jqX2%a3|Jb9Te=y*si$!tX0-krkCv5in`#RFfz}7cf^!fk_(QFS0A6+nwTq3_
zroLNF4*UXnEUj5>-9#k+FT3Ag)A~02{^#!ZceH*2zyFQ<{Wr6>;PCX7sq@-N*_Taa
z^E^&H#|f-zX5)D}nO_Vg**mr1zOe<JiZw0H-adu@+w%q`T)lH;KhGOfGu?1;0zj-;
zaQ0sSC+v8E9WPM=wj+U+5|fPy0evWN7y2Lu6Y-!;deAXAwAw8PgYpG;qr@>O;KNmb
z*A6MvSOZAmb|e+j(47b%3v6Hc1vzMOW$%+y4>Vr=O=uu-7TK+Qs86QAgju8~%y1Bt
z`xfy04EtD<>eIb@LjX>V`S*e~+Afj<02aR_v1$j$(e(U6xW8mh3}}t+M8yz>O%ExV
z9VJqeIW}UkQ*uo}3ye#AlgSx>H+!3iVC}<AB)IqPovFRTd@1I%DP|YhW<iYm2F`kD
z1<VhFZ7#R%W#=B5qSkRKUlUg@Zw#)%&AU~*2@ia2a2>g2d!wsJ#93%&qyBXB@CTa*
z5l?aLAWu`}n$xILF_P*JFC1z<*F4B0TuTEOyhv^Y^3BW$lJ^gtaeM%$pBYJXU>*1J
zY-5t!FYSeGd_b{pV74DMw{C1na*`|G%_B^bQDIkW*C+0bZX<C;(4Tj%9EPNfL`1fb
zdP0xn7$!D5+c@1X-*FD-zuh(AIULyZiztZoOON{SCS&2@BasWkHRg)jndV@Pk%hQg
zW;yB7H$F6#+pfmljplL2z^#&!lD7&IA%b;BV)d9HFC*$No>;<8-Z}TS3X$X(2MHyl
z9PbN-5k<NLV(SI-5wrG4-oFQ?nAYi`1Yxrr7pb~#rGvhgB+al~e2J!e?CcMkPaYO7
z7~70NfBk089iz5N%-)7aUnIy?BSGSEMY#?rt1HnFcU@(!tLn^^R*blb$;9H4;*l>J
z+p&%_iQb-++VEU64N9}&j{~(QnZ}&MwYw?_65{0>LD%Zi3DT3W16Q_?pT5Si!y7+c
zb8h3tPc@R=eZTR_e{QcFzx=LOtl_!0^<G#s?{I=DUU8{Qt*EyWK+Ze9*#X_T@*Y|o
z=~8^PedR&SG(|OX5lYUgkvBdy^XP%?(eS{<UjINBdp#zd1GKYd^p}MGd!a*~M=4bo
z;#mmwMd#v*RSJvGAB7vj_#Rj!<9Ki2DLf1!re@_Io!-jdJ>xqarwm4hS-<nDW!9Tp
zUqWU%<goKb>w~QkGv$b8{h8L!0O0K*fT>QI`4%DoGv9c#1G#hMnCRV_Vc0{l0<>e*
z^ho2?+Vl=oFSm&HbLN8+Wn-f=ylFFo0vqy2aBN^V0;fYa;9%?yF)w4U3-?~#MA}LC
zYI+0Q2(kB6mls!Xv*{vcnyW}Q+V3`N5?2Am8eC2|SYgy7^LCp|YH&F79(k<P_F|qK
zLjU6^%xDFHkz9k8k^<ak1cDO=Tr*p^>jwm_j|O`kM!1Z7@Z#jq;S2+e7&Jgc0DFMk
zn6;rM^6LOvygkBPx!1aw(7+|6I>GE@UfAhXBxY^Eq<2Sg6nk`HFp%xViya<S!;9t#
zy`Aa#&Jb@Hnvt4@E;@m?QtCS1$!a)pG~DZZ6xydgYK8V4t#?5t-xbzUBDW9bU?S0d
zkZ35nUlKyyxw0(MJ8PuZ9^@5q9W`?*)sFmEpZJ0lV_bZGNpWgkymw)i+3M?ZviyNW
zT_qy@{xoP>vfq&azH{YS5%8uF@P}jdoVQP596z)&xrUZtI84|bqq5`*IOwm~{L(9;
z1nl~d+uuW&0g`buTxsb937xmst#Pd;Y}QV0acc`(UZRO><_2i|-p1uqcy3P>`;De~
z(VnOP8eHx{U{=Px{t^Vfpt2Vo>;3X(EyanqDN}?+tTkd+=qiuNBcfF)!J=(j7Aku!
zEuuTr{Drf>i+x-K)OIgnGGJw^BSxdI^>BRB+?_?jgf4{7@Z__QfCakH4sPxhgr+&x
zfo)Jx3VNM&9_!!|PzuNd*4^R-6uf+5iO;o^Z3BA3*5_VH*bhU8B|uOrQ;X5RIeE-X
zNzM<umws4;6!Myhl}?{|N7q0w+QCAGY&tR>Vfc<rthmDoE`A{nq!@cg%{?YRS`z>e
z67sG#kU7=&_$>B?u-=_TRDB<lJOAPUY6pKQL9<BawF#2|Kb6V62VVALd<DUF47$^>
zEST0Q0xu4(v!5YWXppI3FFVoUXHEuoZjaM7I+9VhH?c^BL&93bXy_SZKM;*;F)WvB
zJy-x7JPh&Z_h3I8m&6(lZwRdvhoq4DmA#P;(=plyPROryU;}L<>GG+w5L|YZl~KFq
zHl2hjXh0#TR63)#277`w&b+i0J2JS>gn766qwBD$uz6+71$DY$TwV?`ve`6j`gQ;Z
z25zy~>gg&MHa}aWZ-b0%^UiESLIk)>U6M8N5)>UR7KK@B(6m2lhLxDg&eT6K7#{7w
zW?Da$&F`HvXg0fR(nE9^oyGS|7U7ZkWBnOT{bFl7X8arxtO)vu8BOII{6jam8DQ7x
zf<5E2RH@aJdmyw>Qia5EMJa)MWV_EKZ??K6^D`=BJW>!cW_m+df>LG}w{-l~rS_TA
zr#{-Y86}S$KY#Sx$+Jr*&zv?pCj>bxaHSwe&zyew___0oER;s1tk>zsk%WPv={Rhx
znkDb-x#Q=LBSpsX?Chz<rQ_!lO=91)1(hxI@~KnVNvmPb$?51<dRv!IQnN(}RPl+J
zOD@_}Hy93C9oe>~xL&R&qt;QcI@SSMS&JIX)?y?D0{~~wqJ&UbFyxch$fH!+8W`Xv
z2^}nY9d%s&o<bf(B#3Csl~V0Cx3=Lz0ooJ}n>|Wp!h`JfyG5pwrcHm0ZEj?7X3h(B
zCgPSNsDL)6g}!L5i_oQbi+LW(9}$eu^_r1D35k>QSQ5k|%2&2C&vO?YygG{mDH|q)
zQ4vHev*oo3cqbab|CBN8Rx7b_(g=szJS-PKTIXvzm~?~RT{e2sO}kXyF+Rf+GriI8
zA+&eF4nzO`;A7VGMrF|Tix3kboW+C+OektwIK3^<HD-B)TNvqU8GC2kKe+U7YXwsc
zOUE;(<b&i|ne*hl<a6jCYrsO?%%F7>nagsEnqde?VSg}il9I})&uT$U{o1S@GY9K1
z`44J{Q<cn{R7i?V$Z(bpS{5ALKZg6Qk$HmaTv6EDG6?nlCn>1hu*+yq;Mu{*1{^!y
zn72~yDnwNHSS;D<-n>2-u6mYYnXWX+(wDy?_lh9oQ|>JL9c15K{#nHC1!tHz42}<y
zlf{>VGaYxaEFheBI|xI>?A`9Htl;339K$l#H~K=x;p1@c+h-ar4^&V|H5<jh{7fjX
z2Dnx`f~c*|hOrDZ=?^#GpttCOkpn133|$vYE}pqMBF-WHMM7h&Np<rGB$hTFu7?!K
zrF};^R@=_CPYHHxK7yXii5OT(3x>w1OKV6=RaXHHRmiqb7hrWpV;ccC4wOxaWrI(=
zd@T5{zV`^SI%rG160^e4!iWnJWT+!kcI-{s)*K4EhFEZo&H)(FG7=WQ%F9>ocMP19
zHj*Dn93OCk;a?6PJ6C!(35&d9Q$-ToPm6$D+}GD#!9vDFhK@`yk1wM>PRIF(0=XVq
ztupWY4Eu4(E`O-V*0$-!Ge%sXIY@s_Rr67xt^JknRu|&hjt$(;ciozZqo$IfT9^n*
z^$}l|Ycsxe&8^UCgsaeO5t?&l*O8);t}feu?k-ErO711mErEv)$TAX(EErk1^#MzS
zn>dzZ7A+WSlylDDHSORCm7p2b8b;W$xrY9MqZEs%=dqsZ??1D)|2(%8Fqv}HAMdY*
zj}oZn{6rD<|4i*H=QBC}mYdIlO9zp1PfjJ^G78&1M(xP#$E8sKiXm$yh+AJeCpILA
zH<}A$)`1t;2AF8Kq^=|wt%a>zt#4%1FVx}wV9u2cL^zpB-g&Lh-r@(-JT3{@-c;W*
z7N;>x5EG@f*j>4>k=&{Y@&S0=2HR6&uly6&GsDhxd-V?ks1%)hjrCF*Lbh;}s}?~}
zH!@AR3sY(N;s|mhhpV~P?Qn-_<O85syUb=N5>_jfxtFNM(Kk~&XEyEW;TU!y7vKTX
z8uF;L7`ARo(cYqPdGCTY*VujRh#CBMXs-5(T$nCt6-O|%BrL#+ob)#=xh9s_EW!_q
z$)`>+JPeJ~;prdwg>80b0AUX!V=7;i6=PYHx;VhqF(yv4E>51hxvM2EcX*DD56GI4
ze~d^wBv#ScCPI_O8@vQvv*ofiz(~O9(MsCeX2VuV=e6mk5o2gpauHFWv`R?{B&_rN
zLaGH%mGF6efFrP2oWu2l9g`It_TNTwJD3JLAbcFTmP)qP!8!&qQ5wY;#wgZE3+6&B
z)4?;f7vWpt;V*~@y-dMT2Z@K&vK&mOxuKQ&`0`JAx%bLShc!0PDGb^fjzHonB@#Dr
zQht1ND53Kt6@X+%U57zj`C%iDlU8o}+6%xqI<yC3u%ev-Hu{{;Y*CuvaGjqE0Z`_O
zUeOGEId4Wh1+bFwWhkI|??)3s)+4*pC7SRPQiw!k=FA^jmq#Kw5i5v|8O9pC6CkXQ
z-*+&Z*`jOo95{y$9F9upEvJ6dJiJ|q380I$<j{bZd{&#LKkv}MB}pL*J!R^nMX=OM
zxF`YI-K=^Yfc0zG7V>Y1sj@R><x;I&PRI~bG(2)Nm++E@ECL?8`fmp*A&WfOL0_)f
zNG}(>5f1QF{04GAZ6jc+zlC6(s6}il;#;64t$Bj=8Qt{4X-wB<!cE?Xw#F;Ws3vb!
z`Y&<4La|O$7IU1GEp=>Gl*R0FYn;ckP&~3MT(Sjhm6-E`xa$lj65aGFj+Hf@jpg;?
z{_LUSAeU|a=wU&XLCbiHtVv+IdiH}&@0mb)?ZNgYL<L4}tgZ@mEZI9b$bCnOoi^HQ
zJe42IU*0WB^%tcQl)O!$44Fl*$EjfosoL7QG{RYF_nMT6jEbIv$>Z7^uWYS}-5ml(
zELdCRFb5dgn}7{}ddS<3fUymXd2DM1^Rzi}szo=1B|}ooTved1>f*>wVUOl*!5Ter
z*vQFLdHYOqzB*Gz5T##@#aNu01zYj)SWL0R3YPd$gn=y#KJJ}uMj&5N|I*%eV%5YB
zV7G(Vqf2qe!o-9dUJlG*ROlvIb7S=x?;Ejo7oYG=VU-eK43<L{3I8JP2og2N(i~5S
z7W&EB+U(5Cj6F<5fBu5r3#dbO$OBp2eMH{C)$Z+C#~q8pRM;Szq*yLU{x{H*d^<O(
zCa4EfS6sEUAfdz-0)ry)y0j^R@8bGk<*KAg4AJ<fiPCc#C)B7*#9xlyx#s2Wjny6x
zo9B8hA!Y?8*Lh(pofn}dYx?kg_76ee+`fuU2CtATt8maA=jyn(f+HOW?8bSy^)8Ub
zGRy$C>q7))8lVIk#;T!M{oHn6_OC4Q2po4w)B`WJs~-82QKcISKoR_k>(Lw*ENZa_
zeyH5C9aj~`!^ReYB8}0?!KLnM;)tML=WR6lXd$iHNsiOaE-Qdn*KuY0_BlEN$2@Ta
z^S_O((2?dF8F*6R_QI~TMpVlJ$X>&QA+8rN;oAjdn#jd~*2E<mT-gK@0#I^$ll;Tq
z7lLDMp|k9)-rFtJquX5GD7Zc7KeeqJ-lg+YTEeX^&`z%&y*J+1K|YCx5w?gkh2qz4
zgnX{KR?VU`?@xub08Wt6!kKw18Ud?U&8xrxNjBAA?7~)_$GBf#eY6f=%JQLuhvsV^
zeC8_-&)4u3NyQHzeC|1ZKOC>?P^AosxW-#jKa=dvyZ9bCXE`rjF6(!`^%EhfSdPh<
zH$NY(UjRY3%zf4@G>g3*$&cMwUWVCDLX}b_RcYlr<W6Zen{)HCF&QBW6!-tXFu#0o
z{^0x}{tq{hdHO=JMw`Q|!wd&N#wN&Kuw};fDEs3Xv_aF@95*ctwz(O@OEX;O#bwrs
zj*P4#s*CX^I-5HA6=y?{C`iL3G}$SqinO4r$s_mIX6fdv)y-98A{4lUjNx6JF?h4I
zD5g~;?DN|`s}emCDysscLal^0h8oP<O<|W=O(eMgFf|dyfU3wyXsm5S8BtCD$Dxg!
z?lm2X=yTW&UhgLrQCgjFt}u+nyn*54WKpz6m9CnvdP%*K);U%aMio)JLUbetDTjQh
zT?<52=uf(;BYo>?2_-hEWKI>8j0`Hxps8xnj5CLon5d(gNR!l<TUf}YhDmCvrYIsk
z<7E<d76zeeD+O}Z71xb3wWB7U)Ak{325fy`@4dDpirkeg)tWdfC>D$~q)3V1XAxA~
zJa(a>-ms%04)%%#7Xk`h?L&k~@Pq<FDB`T0-YVnzk5NTs@8jL%YF(t+?HEd+8;Wtf
zsL<tzn{-3jv?{UD>h@aD2JG%J+6dJmL9qlfaEJv2v2&3BouQ06PGPLc*p+9oUW5Os
zk04;GRTFV#W9BqORgb_x-FK{gBAsu=!PqQ0lO#ZD_H=!ZCeY^9ZP1l44|$>JyaxSt
z)*VYJZGrDh%SbC<IQtT&Z%r0wau+&*!Zz7W=e!l4zh{S)kBKWqfg0lsr4uHq;*W9%
zKW%1GvD?&)!5*Fg_QNTs%{b$WDTIKet=={4_64bu*%k8AzSHPeY>gsTMuWWlNG3Qx
zmfZ{Xiyx^5z(<(GZ(=`&lL)9vi^-2&V!{j3UkIm6ZK4<l_(_S_H3t~$W>}bn9l@dL
zv!ZVx*odXY?x;Xvm^qnnEzfTaKs4b7GTbz`6FN|%tfAJdD)LP&YUT*fBq+#>kXykx
ziy0=gt#&KB)3Vz2o+mY`M`tJy8(4{qp|Wzp$`IJfNsbNYVlHUECoGgmYY_%Y?~g(Y
z`z8JAzJ1OPWS8Gxo$Na4^J8Uu7b?djx0dTg`UAQ3Ge1}{s6|XEhe42}mlsMQKUNLu
zkh9SwW-7Fru&A9_RD_bNR<q<w_|ym$W*Dl<E)^>=mgFZ9S7%;$?dQb7zKDlfQ8N(E
z^2Y@JNh~3wON;y=aE?a^B}C>+o{j~VaJ&|gF(#L9noUw2$M0BH5pr}BVO*=U(=@!V
z&^$u%b8n0#h6?7&>o8=N55iB(WWJGgy6Q&8>k3m>CS~eQ1dp{%CVYhaT;^!r9l`eS
zvxALjQ-FyDRMwG4An2vkPIPfAm`Hq&>!hW_LIl+RV}ipu-v_gp6Jj%+xu0fUD|6U7
zu#p~cz}>4n;c&#AfLfl7&pt?cFNXYhZDMT&^F&;}Txu>?oq&mbzIHklTh`6YB?}uz
zbLFDOK^X?Xq&9BooWsX~k^g69MbjAJib1ogFw4x+l}3(fsKZA4FUL%S&)Xat%&i~9
zW{W{<CX-{dOdz|1NZxbl@LAWC@yvIiY_&m$)s<~$Rlo5NBMwd0V4GczuweDev!~p1
zXQ~1QD%#5%bfL&FGCq`~HKWZzU(OR65%+5#QWPPUrQB;dR%1nP-TyFU84K_l*L8@_
z-QL1HP`f_ZUPtx?t{H8sbe-tMmNhwQ^Z4Oq3le7*$u@1eH?n2*2fB5zdr~dQ;^~N(
zVKU{`=o-h80_#YV-$N7Rduy~L{#bNIZmx}{1!SNLmP{qnt{G7=BXj&)$o-CiX)<lS
zzPgdy2lMI=`*H}u*S;)dQrG7QO$*x-8y2)aVwDz%ObkjnFes7G8va+*XA`wI=`WPU
zwS9tFQGuJv#|t@mO<O~y70{zgvt?W@kr<HV=t0OZo0$EUMynpN1RI(4v5WSFH2b#a
zKC!1hu0|Euuu^&F|KSwHoneH5V~*m=DwZs7$H{c%Fl+S95y|(4gV0`Xa9Dr+CT^SR
zY`HKL=SJiB_mGGK#wk?X+Q<=)m}J;`Zc1@QWbFAeb}?D@J~71wyPBv78orL>esTcA
z9=<?GG*1sN<296%tR;-cvUJDOa7@>tu~WyW8po*0xi2_Hp$x^;RC+=r?CK?`P+LS+
zl{}{8xL2a?K7SU8DzbK)Gt~p+R_F<UqH)6GdWS%Gp-!LN5mAY!t`yP;4{;@&LO^O`
zQOkWo1X7n!FdTAz*1UVjb+?XO?<-1-=rSvmRP&?4B-fpCJjZH_5Q%}kka*_DDhu&b
z#lnTEmoX}@_llZ(N33wbUbz`U9T(z?`7+VaKc>c$8Yg$E;9ZMU66C%A$E-tkAQXT6
zx;`I|Ql`0s<q5{I3Ng$fOg?h@h%4_|8V7QAA$MyXN+kNVXmnUb>&_!J3z0<Z6@G9N
z^u<-|e~3+r$ZFS9=C%J9rz1=#`zpelWD#oZ<RR^X{1jRgRgVfG6}RfV6Q+Ko4iz_K
zB2C)Mu=0spNLdyp%wD*S`LC2FqO?@#8RBYSMq7GYmN948XcA+wv<^Rt!v@?gjyS%y
zfoxVQ$*gS-BT$;%`Mm0Iwqlj}@dC{mijK9k^;rXVv%B8P#oiZH5$V1g{5Q68s2vN#
zBPKb~mNPb~lQD(R{}Zwo$B`i5w!K*IGMN4~vHe~%$G5N=6O#g4gIrGq+k=G}=<{U&
zh2f?PlZ+<e*B6Eph5hctU6o%>=I_^V((Z9Xh*Y{C72|0(6ok8-Ep$m`v{Q7Abi*j)
zT(B02<0q_}wwx+%dsEr0xe1r{DzD-xzMgyhFsHkN<}^g{y<>o>&i|Wyfoh5<mT?vD
z@+{{XxECMA)XMoL$0v^$)1l;*acN+Z6<PTDX>7fAYPj0DCuu(XsV}Z={4@ifUytXz
zjjty-9G3RgI`pf9?uZwu^3?P8ra2XjfEBg?S<XgGk=d;AzPXZ9VnkG&N-TQAE4o6^
zb+@9~y<YbLh8CQE)52~cDiGeQxey8(BA7i3)3dOpC}$c}c4oO{P+mt6hiR*{f0K2o
z@#`5?fZBoK=Ix1S#HOkJT(wIk8IJXjGPAqo98%A{7D~u#<_sTCkiV8NQ(xm!7t2R3
zW=+hDUKD3L(*CVWHhPRb8OCxSj~R)gxnV%X+Axk%N(|8Az)aX+%VA4!Z^|;&J<Vh&
zGwPV|HAxs_6MDrQTrOE0)H~Rq!U;`#?y>fooJavL1C#IeYeJJGypTW}m!g5>%O-_u
zc8*EGQ340z{RYmFTJUZbfbH}Hwjr#yE7zy{U~xAD0amrNj~7{mN_cQXeROgRh1bB0
z5c8s;AWHZg5*;AS();fEo6Y`eXNa;`=D@9rQdj}WgU0eKLd22VG5iyTemlzVaZT-r
z^h>r$m(w}trRPz(2-Op-dPjltJHgn}RO>is<U_DGy0~~*4tch(cYBvEZ;d1zJ1|dZ
z1n?FtSp%4b7sjkP7s_oS5yAw(>pVH?>F1~yJVEXg6Z+MdYZ@wmt-amn!K2{Blix?r
z5s<*2({1&0T_R(&BJ)wZRZQ*Bv->aNj4b%JB02D>s)PiaBdVsnXx2`jSvr2CX7r&X
zN2+b`?)J+_4qj<%)JNNE$cHJ(IX7@)22x<55(H6!plC@2RO@T|Jt;ON&g<Jqy4d@;
zWSPjXxf~yvk+Q8N+QN{4b9c}2{62f0ucrp*7F`l7N*K7LLmg+{Gl&#<E>OxyD~6me
zkR=p8f3``+8$DB$i9}o2&QKCfYD})fc^b`!0-FFJecW8a7l_{Q<}uuyvI1W|Zb)W~
z>(<8RY7h6iY;<F(G?x?3uHZN(v#1lj6h~4PEf1C93)|6Sr!m(V)@2}a3ZWiBM&J*z
z33(e43R{|gWxJuhDhDVQYF=kp7^xwU{s7)-EDM_$tpPY^wB<Svn!rKCoocVOEdown
z><=8H8Da+$H)!yh;duN97~@Gd{)M>|`HF+numx)C=4>GtZ9QN?Gu^wQ0amx^3lqf*
z1`CiVD>iJRpgfP^W24=)2e!S>8rzhqDd^-ps|+<1I;3tvzy@<K?%Nk}CT5xmOTiH;
z-oI<QDq<Ttc)$$v*LuT2pLRiG_SBi9i>KP>7EiZkaXi*v78i>-I~a#lu~&;Mz*m7B
zJJyW~RP(b3XM<Z9t$alE-I~SuDvvZciAiwbMKmgq<`mJdhJhJF&f$`>X?6>&jGFE6
z1p<=bZJ~_x&(_)k1H&8j=E_xE?aWayzMtk|Q6K@P+qhg}9_~V2PGC)7UqGpQQ!Hc!
zwNS<sx37x@MO+J|5<*PMgAr>^kQjAS#Ok|u!GA4+A%(<ftIdP3Jo}5DM5pFiT7W5{
zVeVI4NT#NV%ykRMNTq3EVP4|MMEfv0ENz42AdS-_j(1|?{Q!%)tSFvngv;w(5XU08
zNONv#V11m?sQmR{8?+l#C?^9-zp;6&YyD3`1T*MM93pX(AMF-WujZ3P2HCUz5kK9J
zTw8-st9Vm7IFTV0h;k6I_izsB=j34skr^5{GAxmBA9tn{>;iDwODH8h78Tio&W|l?
zfyXLdN|xIZq6C_f@a@pEj<J+jeiCCW&>DZVmP5%{!%e&mX&yqBFr4J;7=E&4IrDX`
zexBm&rOo&O1sk^OJnHJ~N3L(98Lq5<<wB27pS4%fJ=J;&S8V?NlD(|vX!faTUa=Hs
zg%(lKyEMOX+JHQs14*0CTIgfb+1C^Aqq#!GRi764s_E>jSzv#zK%0haX@rm#J~^G8
zX5qt=6pn)t6vO{Fm7O4lCvzAksQl-qvd>JR=p(tJxauzozSWw(Zz}r-1b<%v+%26I
zvNx4|4<X-^gUrJnnab7(_h1groM%*I@14p%m;y4VqYY5&_p)~p=G{3&R);9mwbt(@
zfNa8H1^VjN=o97^vp1PhaOcWt-0Skjm&UmDX?cxODBm(mbn>Nbw!Ez1oxHE?1%Vjf
ztpc_)!I2_&C9P6&r-MWaO)?Y(aMFP;z-h(7SKC*fd2;FjTuJwY29?PKE7wk#{>$C(
zh^u^hXFqSR^7&l6$>$3f`QQ?yFI?n<2;S~~O?8h1+4hRwKmLn+RtB3l_v*D7oMhnp
zvp=_b8(njM3Y=T?2`bnr#QV;hLmuwF9+!OI6{hpal%{XcmNovdQ*hB84c6gq-DVb<
zOSraZ3DIBXm$Q+=mAiW=3ejEi6R<XS<qc9do;dkJJ6k+^_T=do3}){M^1qkt>c~IJ
z)Yoyo`}$%0$2)~+n6k8e^|E-P`8M~?|2%8L2?m;J@Du{gln3_Kn2>nH!{h!X<oSwj
zx;cvbt&rjRlx)h!S2yZb*T^QOw+ugK(%wh+;OHchoIz~4<C5`Zc+6apBkfG#fEx<Q
zqb9wSNGmp%mmEf>1GApCb(*W<$yd+MV$J3nA%9i6Exo-79>AXg-v;xZ8S%pM0h!Qn
zFZ#8P^X8gLqt+Ol$(wpMJ1bWaSzs^m#*H^^9tY=DRm?90uVDIpqaOArL;9X>KK(3i
z$yT4Z$+F^A!Kg`I;$Bm_P0)?*HVYkfFRt+pmGitqMTbu9#kD1DEaALn?oDuwPqCJJ
zLOk*1L*|0F7f&R_sF8R!@wGdNvRK^mjw`ubj^i>HupeFLAOlyVQ1-$C&xyzxb{any
zFW-??xl2KUMo})3x9&)Z($fU#J1V41g(IhUCxw(?2)`i^jy1c$@!KR^h#MuzzVlt<
zpG7yN{9)g;quN90$lt=ljVff(QI_V8RN=yVY<A+}M{+fH=TGizvfU*Xt^_+|;bDg_
zT;H?3lk0ngl@y?v+Id{sfE`;LNW92cF1v@!OG;#q4mk~}E)Zp~%0s87!)>HUtwDtu
zVB=OLP5_!)zOkpc0?TKUrAKmb>I9DP;(9II!i7yrpYLj>c6n=Sb9Cguf#Ajt<T*dE
ziu<My*s1z}(q^x_z_M;@V?8*2I)hzUY&iQ3s!A{JOhEQAe<aIo^&XKdi+Rtu-VZve
zUE0QZWXwPe=!u#l@x{D(K(>n9QDNRYiAmPw`NkDT*Jb-(iF~4V4f|CWgXW0O&Fo9S
zS8u-fZD7o%(MXrTf0I%u7oS)jOyM{{QQ&8-9^=G$@#w+$6XV{wE3phwX(bUCk;a}}
zG=(nS%3KvYXoP8q4=m~$)9kQu(jfNHq<=D7hyKzR*;fx}C`LT^wy}yC+)}%8c@14t
zFf|%(a*2!h0%=2}tQ;5=>C5VUPpPE|Yl^0h8g@LYfSWK!jR;Rnox~)V83h5U3x-c7
z28Pis#tf8Q-AHa`G;G8(e6U8hJI<=*S&RfzFfBJaQF%13E>C5v($vpML1S>j-DKGv
z2p}jm&?s#lVc)=p11wk68w(e^D(4fAs^@!t3=J5S$Uzi@n$^xNX0jX?{GuM4vXt3?
z;fej_RKk8iN0ElY>|uq7TT~1`V0hu)r(B>Kq2rF=ld-Kt2T8!8*9a^&!)OMT@F8x9
zLWRSqxuR4$SZnHKBVk4Eaa02RlP`q5dL2}nB@Yn|Q5lkgazP`^*QhgjQfcYMmyTrz
zczE{G`o{ieFXl7@@$h3x+XkWOZZTP?wiu6SP*fqo#=KjsMhJa8mg5bTSlz71kP3#o
z!Oo|cW&(vNG?R*)s%{~O^UJ%4I+oi(bC1eOFST5>94bdxs(vGa(B<P>dq3qu1T)%_
za(c@OOMU`xR0<_FxbvI+Vz>s;q?TDC8v(MkG7^;)YgUGk!9GaK!Pf9*`UR`SZCi`k
zey8n3t3(}OY2)9!;Y}$rA`IGUzJ{aOuRjQy|M=*N1r5lmtVt&joXlUx@LH(r??meK
zup(Ou<|_Oev^{fcju4pNi&~dYyEj(4n_K0bFv(;>Jp9y8;BG&L*_rHjcR80BT)?)Y
zImMPrw~}u$bIaYKpBr*iY)857gN`YTIclczaYugasAnbMa{_C@-|Ul^_gSm`)MK;a
zKKAu})dY8;1>&BYNz7U1{=R#r(_I5tt*c^s(oz)9QpO=lBToGKBV&4{bp$TSJo##N
z9O2I!xG};U_i$G>NrYaoK$U2+i~b5cD-8TQQ<ErN*4U}x)?ZwQSX|ut;GwXnbP}e*
z(kXvuvtrCJTTz}x8@fUWrv<wPs2+H7Ufp=0LG&r_>aQY`feyuaC`~F!7FakH7|w|8
z0-TU{H3q+u^O7mL-sxk}57)M-HP@A)U6CfOdWy7MP5s4vUkE=mX?ue!lbhl5bi@5z
z2*#R_C`z&SC+4d&ZIktv7t^+DI+l@F5N6IuMG5asvWqwo(u>6yF24lmLY=7q(*!DO
zK~vk)+VEdZ!ZS?eve+r&NRM7p;Ch!9?@`n3`93QrpFRS!_B9*{g}O)u>&p#?pv%&F
zf@Rb_4DSxCmGVv@Xb_&JY0)*dty-+u+c-Vcehz0q#+XB_We0{@nhWLqLG&}Z7MC_E
zjhrZAmKEqt7ewEL<FmKHVQg1#Vb@Y$QCHDypF}ub$2l~_m$mN&M_OB0N&<&Bn;(Ae
zNy3RJ80?cuxhuA#*`<_~4qe$;3lbhDkyOi@<4|g|lXq@0v^`)$T6!FjRc)>wy;@bW
z2}I9Nxm1QazZ|Ms5u_wF#G86wjnPdCMcD}vr^R~=>Ogx&xPs^9eHCls38t~htQ7W<
zpO_|`Pg#hcV5m(%5M*+~`O{!W<``Td77=$VKF(v|?B5uEc~<Cax-RweeA66ilkBP=
ztCU}&_9R%88*()pjfTO{;i|+lIcXProzeXv%wQ^z6-$y?1=#Kj{Ndx}XNJo;Tor>I
zG#5%M8%KKX4k>BGMa&!zsGuX>dm}UjqR40IE{LE^?<u;<)sYkpm0mMdUd+4n5Q~{2
z5#)Ep6W5NyA%SF*3C>S2lFI5X(p|o^jOv;}@8GPg4L*)t=V-T}iCw|giqPGqgb!E>
zIh5!$8l*06s<6a~2r^BM8g1Gi8XV<LOIw_O?}slt;JpzV%HJeQ;*0CO&BJDf6OX8{
z!|sg^Vm@1<lYDbwVUzKVxB{Z@mUot<SoG5~+jwO_Y)}Xie<l6_FC1K;$RpinM-}N6
zK5@hh!g&#7Y~`^Vh*<gXXTiG=Ko+x~fT3v?6vv+}ntme<F@}xqQ$kPb89UdiW)2|U
z_+ka+wjjneO@1$M>?Fj!4WW~9K9XA3Z0X=&w`}Po2jTte{kZ4uP9<2D-eqr;GmD7t
zCxOQH4P7qHXuY&^nTU3L3UXl?!@ucPmhWao{Vq(1>as_Ai*KTW5R@IbpNq&K5N8nt
zoY4s}nVu`+K->-UB^c@G6&JHH*FkY1RHNR<LbxvKIu7OH;9+LtqF-Z%N!a_|l#$xa
ziL<AfgyKMreu$z*o2ji~qr`5LxR|jbAm}tpe%XdCzqZ4p;DjEgT+Zc;u)-n?bhmq;
z17Zw4B<cEi*Z2_3G2}=t=GEH-tBIEsEco~jWZj8GBVC`pO{9xtU?4CmA(VM77+yFH
zGV8oIvySPc-(`FU5<!TO*@20%{}OzLqaJ+F&gAr%OQLf+s|cAcWH>7wFO69(AkqVk
ziXb$eU*KMgJL%k&doXRo(Sy>Womf$}jORjkFfzwXk)fR8nD#H{SQiyaUV5XdDg=!B
zqDY9liDhCcTnUn1nB=0WrEVUuwhgz3tWA*mq<2CssbL`VoSX#Jkh&n)EEfgL-ijiD
zZTD%(AZV6vmY9vch_p0fmSrlUX8F1za+a?X(X+0n&ZV!RkoGtr%Tj-_biPe63LKDc
z94qH1j6SUx9~MK|kYMi`n<Km#w&YHbLWhSJ#`2iSGGgM92}-`Y-=#_2C%?wpacxpD
zk*sDry~5hn@+Vnt^m2ttVKJ@uvsWl9*wwq-UAbDjJh;yJ?lK(BxO$ZqXwF1++<?+4
zUV+RFgPd!4rnYqE*ct3Q;(Q5ATZ5BV9GE~}RJPH%ifWJ%a=m*Av8$dy3bYw;O|vX9
z*X<Q-gu2i;svBwUV`2gNq8T7U@2>K*Ltxm*CxCEGWc_JV4UI=co8;$-wK)!m<Q4>)
z<!4M*7*=&x^cfm+^xO*-hL@NEl=GW|&4y)g%z0ML&t22a!mc{uV~%%CcyM&-RlamI
z80@t$xb31p8VpAZvzuG9$QZrZyMZk;7%<yiIOWRh9@B(8G?-wJM-oA*(^A1VCX-d2
zckt5Oz==V`*%XIS$-ZN$%8n?|)Odzy_eN}|;*ea20?rLdEgwtn7p_YG6Mj{?>*yWN
zb`*hFt1ek90#@(q+nDLY`?5Yq=Q%O<<e=l&%g@7^FFB9VO34ixnN+@T^|weN%|o|(
z3oXR^e<AE$OT8;dtxS2hl5X}qaP;S%uH<&;KQ~8<=JAH3(Xqvtkgi?f7=d+Y65_Ub
znjF(2|1`0#UgMWW{qYx%Jf0nS{5)(Oc_U|x<W-QpZbs+Ewfe0P+wCL96^?_mnpPH<
zWo&hZTR^t9zCF5}k91apvX3LQG*xj?54_T&oaGuWXI+yp)Md`w$s#|tKSQ*};UbeT
z-P_IZoi~*3X@5hTiY9Ka_$I=A)my_kI9e4r>Lmxa$q5(Q2hRgMJ~@@IYo7)9gf@}W
z7Y88uMK7@S^m4Qas9dtf428UntVt_y);*t%;bR`{+h@weFy<L8Dk3Cu@Adhk7B54u
z-ReLh`Z7^7NlhoY)p)G?uV|q=x8CF{1iQ*p|BS<69d&k8R~XBd`;|8&QwS!%q&9aj
zmW9JlbWk>Y!SAt<@JYF8oojvM>O>aKKaLxZCv18;tR!x`R}_IPr7tVK2S&KjSK0rQ
z^?gO-OZz<E=6srCN<ir%C@t3abd7M})&8T_LKS*P7iikg_bXL+>=BWfkO^3#aWpOw
zc)q`KhZE}eEiJW7+JAgmJe|!j>0;|%GI+YmByz-OD7~w!oer7Ka{iL#cnNsVu4=1=
zVzHv7ZFTWv#-$TPXG)bNaHMkJhj!fEC+K7%?MUWj`?L`46Qu+#PD+mI2vr&7#@3-Y
zbHr_QZXo+i|I*fFSvv&_&CuvJ`vvfh-l*~X_vRK;wj1+_HxHpSejru)aAQ<MIt1el
z@_ihl0>kQ#yUtB2O>mE!Z8@CQZpZzZ^cCqLXx|PebVWvC6IH#wlI&+ctXJ$>Zr$`z
z5Ap0cBZe(oZ5x@7iRkq))uZ9q+;P-90w^D8njLJB`nY=>-~_!xBizADH{gNnFQ<DV
zNm`)biwc>&R4pN-L8Wlgsw1J@n@f!Pq1K0x81>z+*u<!RI@`;HptH$5Kyocggih0J
zzG>PNn<+2m7^g-4U@F_injR}J;zdr&d}b>9cg)m@^p(d=kW85DD$L&czfNVpO~^(X
zvNRv-FHU8@#BxY~nX17pTa}<sPGz4a`1{k~CLm8?`%~FBSHPMbIEDSwQ`uF*KAgsD
zd8yEIQ`tfVW@BRBx$>sZPCd{XF+ph=UYqO3;BCE}%>wQA?K_P(TsG4Nvza0>X@Wa<
zuE@($UR>*~bQNZ7yQ(l*PF2BJ&I`==&J?g6$cn&MN})hZDBA)%+0YTt$p<IFx36qI
zIrRV%lD@x?kW_S6d!u)w*GC%C2R}Mt8q%5CQFs%OSK{(k?eM|FPuI?4cMML{-U|F|
z*aFEqE0>Y@qjsv>8TJ{Pd9FLc*Okk)7lwoFO(d_tr36xZ1oe%(;I4I7n>GvKr4{>a
zFtXohcmj{f{}yL9l$*xqYiE#tV}L9d=2mgD{cKB1zC`;&j{DdjF2F`H2dUi{J2Dt<
zX%yomsR*C%ZH*TDtIyv=9>AmKScgnhQKX}p{>~1&OIY~yyQ@n!V6ybZ2_&vFFNr8)
zJQmV1*h?IWZ;(YoeEdg|)qjKoo96tROSi~~thjA=m)TK7$JEY!JKkk)qzMK<`zj)=
z+7ds`UYj%AWz9{KwGS~a@cQ+)T*AjCT&8jU#F=y1@pI?f*_mU<Pn~ZsooSysb8fM{
zw0Q1?<4b({(y8NkYtli+sT8H^5iC7qW0CK+iO=InMxG^d2NtL>OU_f-iBo4@sjORD
zq~l4P!YVFO*P=~Hm#8$Euuw8)Bi~_foo1g+Qy4#Os9hl8xqM}tcnaAqYv*}!C73NM
zn32ex18fG@o5OKAM8+8GlFBg&i(?h3D~E%d?mU5kfvIAaeDLJy5TpfgN^)5?+9ras
z&HH&?Jkef!Y3Ypgob4eLGFLIzH7;GQgT_5D<#H@n>*AQHi5eU9bHHiQ-iupMqWYIU
zdiFRf7tOGotSllSjTP6+RY;#}+f}6K+Oh1)EizZ3q{d2`ft;H&7pc;=zRGm~Zm4VQ
zyxO_i*GS3n)y|Qpdu>8Uk?D++M8gdy6-F*{tkK{dHnd`mYM_BG<5AAm9jV2`!^5l&
zjJh~d<GnG#^O2sA&ZXxwCXHY1F7vMPg*t;VePL!p36g0U-1!WFm)F`I=v9?`T%p-B
z6NwUR=auDS2zI%Uqax=Mi~L}*6`#)yD11OqQVT}gv6UVOx9;&Qh1bA6C9_E;Cql$v
zia%Q*?<Pkkk?~`D1{JYuqRN&ASl5Z}H_JHcS+fA-<i?2~0_#j>zeaMQsX)q1GMu<0
z#K&yLMvZA5msoJVhO}Og#m@G+NwDFtse6^XOt=&#S=sZxs0ffNo0A{qz<4YsCzfHK
z3DR>YzmV$?Y?*<SqC0xH(3HD+Dl307a~;}V6!%Fm?84M^u6G3}QYe@wqNz!Zmt~Ks
z>Gmgmkhb8-OWy*wvO|3qSL??V;>NAk%z>zq1yOcXI{^1Z>bC1kcLK&=N8(-uhuOWK
zUrRH$o4poC9{KX}`2T0`ZDZujvNWNpf!Ut%bkES!(>-0&Jv(n_PFH5SGQ-ZS>S~v#
zvuL;Ns`iGn%GmDena%M=MP@{1MmjPgD?V&zmx@nm2So3#ggg>j2@s%VK?otVNbq9>
z{6VY$p%oylfaPOZBz_?zB%~1tiRU@ze&72>MB1+Io|y_&IV0Zp-gD1A_uO;OJ@=e*
z>+6DgN*i^92C7N@@NEEQy^84GN(6a=q1^T8b8jR~gl~f;5*QNNP={skvK|5kbJBYR
zZ*}@$i_pN-;2Q5lrbaPvOJx#ExnXi&AW?=mrFWk0nu{9Ss*k3)P0{8`1=Xg=ipqoI
zJ(?963yWrPqCE7X+Y4wz5T&)&MFIy_ihARauxIU-tzxTULSBW+ZGxTp0Z?=IVKkVY
z@OC!ZzuuJR8O$@NLyrh}wLsga(q7g_)_A|IDvzaoT2*++J=z1`X@`Pn$QWGTq>qr(
zcNp#9rG5?LDqSByV!9Z{pna31h#sYVK<~$!qtS5W6*5aN-ZL#18LQ?o>494$wGZN*
zJ?Khy4&!|UQtTf!;7$V<gWfpLMXI)CpS=`o@PuVivv)da@DkUGDQ7!FI{XUIHu`YX
z9hG2dExou6<=4fMAOXp?GAbf*?-V!DAC>@YY{bM#a$R{Yay0ej&GCudu{e6Udjzo)
z3C77i_FTm(OO((D;Dxxfed;eIORCvB2jgnmo_&{85!B=I@+d8qW)e)Qq_2H-AG#0D
z3%U-+7e?E*#y#9Y_zk4br1qSo%uMOJyA;-1dJxZu@SPhP_a#e4=zs$XH+T1~1tp78
zb0uY}F^h+Pq66-ELNo5cPz7(LF%(jnK=Jg_t#IS9+O!Iji>G0}1zj2%$K&Auj=?z0
zVWPId%7j}mk-5n7zK@tLtuk&@KOkc}EmobFuNJOyU;55;@bK;R>N{ipb7}mxtQmMV
zKsINE27P!)VUJ*3nMR1{F~4ALL$g?Ry8W~CmSczexhadrmASVSxnL7!r_#K?K-hs8
zzx@mVT+|Y908yR+LWw^-A38QecFuts)pQcX;yrl@`R>Sye8^GF0&JaV{@f#;DcRcy
zdivIZcVEnTltV~f_<wXWZ>@Q%sJXh&u!f4MwV~JiIXJO2amugc<(3?1-ujbBmAIw=
zH%cZfBGDzwh=|xb*nAVNR``o0WP<O(@@!c#1Yt9i+42rw7S7K(`82K$yI6RGb09H;
za$epL06+7D3d<ZU-8!hw%*x#vQc8@v5JMf6%lN&XuZLQ(x&;%%Xu3an_s)IpA1DT(
zoHuMYMRbI~AUP1MgT`th!qTg2aP@J}vrhyP>Ww1YpkX4`wz_qQtnfAWw;wF2)?nl+
z&-=09Q{{lW<Di3+fgpS$ABg}5OqN_Kmj@ymUXp5Y=JMM#*5ijrxd<V_CrWJ+U>J{o
z{L7(aSfbP2n;xzr8$JijFglhJ5)_wa22aCVGaV2&)m`P_+}SJ3N*!IQJnham*GYm%
zUr#q*ftfC~qQzD+Xx2PPP_{0Zevs*&+6FjGvx<9d<r&OoJG54LOFesGKQNFYt)W2!
z8-?i!Zblya?a|EM!b}t!4QmWWxE4s3e<i-La6x&V;+PEn>K_iKC&+oY2j=<^gI;||
zT}N<!M>2&M&UikIO@YWQc5)&&L15xbj)`!hLu#zSM2LepZYv=4sg|IfGH<NMpvR)o
zkB^4aeuo>%QyevE6EjKUDR{RW9_!fG?5oWN?Yf+%xtAO!u!8We;Svuy^n_6;tvcnt
za2Yn#<C^QRb*6CF*bsd_zB5~803*OXPZG@~w^N7}Z}&JUZngcmw6<;EB~@fKkIT!W
zv{<U@CsoqdNp<etT9X4*N-uT~2%}%a?Q3wSd$ggL$tD#3WXe3CJ{nLPCOQ*Vjf{=O
zy#Y(aW0hZD7<Kr1Ie%x+{{Rkm8|G)W)#%5q=Tv(xByw;waY=nVU6<n{@6ckYE%0WZ
z2T3N2M4OpuZZHkmGmzNm5G?teMlrz*U6I*%G{P>F#)%q!VO@h0Y-&TSrtfPd>QD*>
zYiT}$4s^J-5*R;P&v0K$CI+q9gTug#Wpezs9H5vF2`0wtJs4mdLi#0a+sQt&u4@T-
z>i>!L_sELr>Ys0E50>W;93}+krUQ!MA>2mU{1*=%h37}S%Oz%VQY2|2_owh03X9v1
z<ex)&S!Q<$6E)%|;=i}6H)2tiR+Fr>0Snk_S8A#u<uf$XE3L^9*32W^4zP_I0_T)H
zYL8n7r<-rKG?8I?25C{^-@U<}^aR%F6U5G#OV@4qk<;eN4Z6yta%VV&GeJwJ8qy_v
zJ_n4B2xoc-z(38adp;Lc9R_V@%y#U;ZstruqP}pR;NZr4Xi|nv%|e|B^5ZZzvo2&%
zH9bp-c;1Dke+S^1r(FQ8FzW)gWxw)i6_Z)6P(rkwxFzd}Ku+;Op_qb&0<NE=TE0i)
zQ$<zf7T}4zz{(vyj-ap(u*b}iZFN`<+j<<>Gy8<Sn3ILWPgvR^uWiEsFIAJ3i?z|*
zlGY^0JJ0C|7%VPSO~PlpVWonZfuPfc@fOX1UHEeI;x=x#M{w%JE>hrfJ!=OiH(hwO
z!8W+xEf@m{JoyCZ*p=~MIK>$fwEGZm&6B0(9<GOEHM_Vi*wVf3L=F%XEU5<jLbala
z16R0>X(OvvzfYIpa2p?D9HJDH5+k>ypmJkGfeBH3hR2S0($u%LE-P9PAVd`7$#>h`
zfSj*G6I_2Rw^?o5i;x@Ful?ZOwc9Ow^>El_ClZDI>sEtND08L1<{sh@lss?uPmkJJ
zX?dE9IjPYZ4Nno$k<~HJ^IEag*cKr1v$&352VxPV3sbuK$7Kq@KNC}a&_5GXpNT2Y
z4S|-zhl`6WI}>YeWcW;?jq4<u(FOxji{|40QH4+!8~h8;3+p}MRGf1vgmg)W%^|RJ
znY}EMUC;wWn&ZFX=poL3CV_9_8{%zc4jXfCX6i-arS67Hl?N>XQD~>bzBrmhS6X2E
zhh?SP8skn%nimt-L>PDbxOHBy->kYPdtEt69;>4ww6hM9Idy{<9?{UKhs+Ygu{deQ
zsJPC=i%Ei%?-FMVMd@u`^#hx#^{AICKt(;(0W@kOPpWzo9Gy<F?Teuhl6W^5e{l72
zw^(ww;R(h>=O=AXC)^hb8|}mH8IzsE@*71R5ODf2RBS06Ns4Mo7xODC`c>G1nNdA7
zU&uI~_HeT)yfdTI40$t{Jk(ct18x{>i}E1eA}L`b4<J|Vd=rC>ME1X{vA-Q%!Idn5
zDTBspM0Us+TayL$6!+(Ovm%{sze_EhrgB?R#Ao1YqFFqyBEuf8Y7%wWxgf)_-x39k
zw{moLve#bmTDjPkl{BFVHRd(LA9DQDXm!p|a_FEEdi2Jap&Inj0o#sSu$UoT8Z>3*
z&YfQmfddkU8kF~a2n|?EdUV5Wg!Dj81Fd^kV9bl>=Df}-sO@-!GdGW@8zg~(ts!SV
zNrI!a1<Wie4zyshf6YOQ!Jzje5tY`FUDz(PB;?g_VmqE74B})6`ww%0hkAPUGOnL&
zVPb-F&95=O3DFvdBjnP@VOL8MM4sZTnfp6jQ$#i>Y?<u{=`ON11V}}T7Z$mRzygXU
zvktXVT*7l33B^4`jyvxIW9nnD_jr1;N&~^_>S=>HVOyt@(drr|a0AC`UF0`j6AnFf
zvrpn;PP&lPW4vsr_YJ;9%oTF{!P0%}<WA=~#a~td<2YAsRQHDy`1g#FFlhT@i`=EH
zix;;pUu|q1e6qvDJ-|1ZAHp0_fUQmUEe$wq+?RNKdvea+;!Ps0z3Bl`EG0`7;;ng{
z%sfiY0s9D>a|e%fgcEA-iO}c9jXN+QjFC-I7wOwO6=7dG!FCsp0@48b*&Bdf$Rx~#
zzYT{M*a6n@P?vjX1hR?@8s46!t=#K&uU1Es$?5p&rAvps$<cHV0_)_`(Nv=SkV3S3
z2{L3b#!7*J-SH(w<<i#H8&|fL<p+;A!LA#d-+Jp?-+t@zTW@VeJrSZoOm<EUYqa1?
zYbgVFqm~j346_^6BYu>@LIi%HTLgRS`Grp%s8uBVJTOJn;f0206~$T9gm)ltZN{T$
zO?ix-kRIs8R90b+7X<C1PEe>COt)lwe_Nc);AYnfdlk1NJ8f-(SLs=EQLJgQ8b8a`
zZQQgGa=PUIP?NTXoM}P*SHOa@cVJf9;aU59!Oe<NEf#4Osv;_KJwd!&z6Xyo&QZ#)
znq5vzoLnriX<Si@2`+to2O*otQ4f(A5tXHEcWGOdoq;@1!DpPBgAGI6t`tT3Q%GBn
zv2v`DNPg8_u722s0|bSNco7DAw4C)z@_V0%9p<?9qahQNL%n`d>H6;#>*eZ}cP$F8
z`qbV;;takg?P`1iYhLvTmn!aQQho$+R&7U&Wi5bt;?NTD$9%2kCQ{fo5g!|*E0djy
z3&d?8x!SPtE|?L{z8&w@p@gIn&oH@b032@abP_$4SCo#RoDLpJuvVb<ag+?Rf-RNi
zGvT>zv8HPePJbZ%wguG~AGJ@rVX9e7F(k9l^~$;0tyOO)AWG9%$5s6xTdu|NY5`ub
z&gYAU?Vsy`6iHn>utNc_j)qLy4>2Ky<)1oj-PtX@0hECK>=UBS3H6Is_*!+HB&Wc0
z%JXnk8yhG%2Ll~1)8E7L20t=0LEC-YY}Gl#bifu8i`I|29kR_MYOa_oHm>kvA(;LO
z&tWmP_C!K@4hKU8uLl|Fc?60Jpr{bi@i7yEUm^7Os@_l<vjxdefG`;<duh#Bn5cEa
zjXWW^A2T;6EXhwuzM@|5*dS0`+%p_P;kHljauc|H*p!c95hSE~2yRW&6q@9aVM*Uh
z=u9m}@||$>!Xf~q5y#pQxZfy&kb~zGo-3tm-|fBb=AL(!D&O>O9|t$wp)j{mp8XLt
zYg@+1C+u5T&GU}LqhBPdb$iJx8>)LkS!D1xM2rVm3373}BKJeh3+Xy>r-rI=aJSIR
z2;V<om?qrVx}EE}gH~s}4<9Xrzon&CU36NEsKYSx_|GQ*;6`F(zG~qmlSv3f{jG%<
zr1BNQKYM#ZCfY2_3eS8TY)?^*XNZT@OzipqhIf#F0B(GcgJ6&9?zJ5Wn3ZV;snG7&
z7=<BncR;P`%R`9L@dPq8uz|#=@YiQU2qM%BAw^)_Fp_p|V&p`$o5TE#tK_~wA_bEm
zQK0KJ{~R<X-#K*pynf<K$}ZL;mJ%lH&T2uBt%+JHjrwI)!n)(R;1y140y8dWQc$p*
z4Z!l$`{J4}>{6z<9SSMf@f+qcCep}$90jDt9G%Wq3jaC)@}*AKljI{#<UEhUCy<bt
zp>={i!04<+^SH5zT}9GoglG6cma=-OgY17aQwi4js}M7$T%hTJr^y4@=t)JFLB-}u
zvEA0l$Xl{*No3dR$_GfBV!7VcYgBjC!Yf{h-0Ae%)rY4&+KG^=0S`9dgwfb#YY&z_
zYy_#19aUvmB83mS#R$6=BfY^wXxbB$I12xVMn0Vo^=p*dSDdGkMB;=Mf>kat41{mE
zD=z9zqUq5#)1GXXbgS~1Z_nTd&j_SSi4U}P>5Z@JS3wz+S~idNKODAzvz9A(_UX8c
zFF9(Dk)Tm;7CS;pO76cw8u}+mE(7GlhvAGniQwb-!tMneu!f6RBOm5A7h$AQY2!T=
z@tC#rJD7{KTYGU?UDLsJVY6FDL5Bc|;eO;)aGBH>qJl4Ns8Ar#B8@HDb=O|d^+J<j
zf3xO;)s4er;$7qQD51D(7SbYlm5*>rV2Z?Wml`KPHJpM2FV_gLPj{|~t{9aSnH#&5
zIlnRM5f;LxLg}zZTd60Qbg;-z5tN14J{$<2WP3jF;tZTg1XBYmgNlitNfHQa2t#t6
zxA`WF*JZRU`2~r?;e{zTij>&p>`6A|k(LkSr=90M$3@KVDQSPtgg0XIXG}2`S+R1;
z6cTH^TSqlgQ9}i-RmaxJSS&G$gj3XY&^;v4geOzwiUgC@%OT;Z#!CG7$*yOF{<o*8
zmaM87E8(_UP{eC-X-2uFAi^dBTUsJ>prukNNWYxs_VHRWy&wujyL@-(${;}H;hd(4
zVMFUA5A#~iCZXls&$y(g1WE1qMTy{e4<b7|U3X<&Mb5`g*H(HlGcuTe@R+&bpa)Zk
zxnxGTdyj4mP+xeG0Ow;^fc=?(!}Y!Q@87&y8Q-A(*?J{9vrS_l8W6I3z+BdoK*q;V
zjc7t;MwuRTAvT3395|Sb`dyyZspk+P8xdoP{2*MeEJ{?dp*DK7nAnTn(ZfBa36+{2
z)B8tLdQd|W<K*^eh(nHXb&R8PUiL#*BSvT7ECSQEI+)UNCvk9URB{CAFx-7t4j};s
zN1hK>3rCZji5oDPGZoGy{;+3~n{Qpd;@KngS2BB)gh|dT4Lz&Kuq`MLeXZx0h~*TK
z$?+l+z)4S|6*t}h!3FBXY^RMauKNUOH?ZYsbVf)_F={;?jvjVLZdf>H76BYpINXP#
zgY6@H0Q|TRl5A}n`_oRl(LFhxoV6qtGbmIe<)R)Tion?GqedAA>mIl-$<RYSXoff`
zApcMpNKnR4I#;}LWfTm(TB~c8&q5J|juT5)@Xt)j^gsM-Uwru`evk_L_`kY?T<MZa
z`?cn4$esSL?c3jt<|cms4fp%?=Ii+VZ@S;#YF@$bZ+~EOt^dx>HZ!1q<@@jK@PJ%W
zY$7xNxIdhXTP>5Y`;VG`qzSu&G8~17A59|gw|-6`A~SVT+Krhhx%G29Z?f&LWhMQv
zo1(A%+|DHyeJv{LMN4VzGF5-&=XO?EbuFqoYtb_m|L>P~KA%@i^|5dRZz}#zFYo+c
z$XWj5OHqqy#axjy)&Gl^cm4-fzc{zLQ*cbpf9~a-zs;I)S4$0xPXd%&Q}O1@JAauK
zmy#yUBn;h$O^vOWcfMDuVV$r7KLf4N{Qq_?u>w+z2W;VS3%$`yjsLCr|IDbMxT#~7
zsqh~(|2`@p_j|DgF+G`*|5EecM@jf;7E30K+MG83I!ga~T-tI_$IOv_vyVbwjS59n
zkm>rc`4FXEiAu#?mtx1wV-$;flK7pOW_!(E)@)!oGo_B2M<^9f@}QT}=3(;?WxgDt
zW<5*|m~OM1fpL1ODbs0oP$ot_(S1$H{pLPO#&RN3woSRc<{rw$<k!)sAC4*1ZnjbA
zOELCx$ZZOI*!&O$z8n>>X@d<{e5W}|uNwF^uNwHZQvOT`QNH_P^p3v1=Ec`$1cI3&
zpAV5S&&oZJrs^pmuV)fcO3WD)DHHQZFu3wICZFeyLe+r1<ZV1NA{LeRF>)EROPN{o
zrqD<<{p5X2LT?x_#9u%l6ma!I%!n8%Qzz(RxizzBEEuF{Dxr(=wpdcz#lJoj#Ps?+
z%Y}eG*UBU{&c75&wb_`(OK+iq713qwKq{P1p^z%G)fFicPsK=IH8jWNBRgK-D(F&#
zGd=#-UwfIiX1to+nvtX_(;njfjMs1eEZm>rZ_SW<FO0JW?7>I<-k!bqg621<Am(BQ
zc-!Hf_3OBUpZWrhyfnrc5WFUNT@$0)TQ+Q>BivY^J|QAn4Ad8N(S&LlYKY-jiC~cl
zrO)gv)Hz1Yu6X-I<jx{x(Y2Pn*36)vUIow{y+576rS^xg3*13o&-l^_u7H*}!cf<2
zs;KDR2l}5QVq`3i=nNe(jor}a>s9mS@3rn-fB){ysG8t77S9{G^mW{T=cz<vnELni
zXQ`5t;BvK#Z7^i}QmrCi+sgNG1J_-5F@tnBf?<{%wa}pbnZVR+dfG;Y;tt~?cqxVk
zN!XPOo(%)ZWN*9$<`j}$bY52<19SR%6C+-Z?)I;~4SPjN6u=cX30na|-o%2$Yhmu(
zy?GCplH9+!bLaN8`#0}Jw5Yk{M0<Vb+O4}cZ?uqJ=L1}Ja_#=D_up&XxOV><!QHrd
z@A}<aclgZo`~J<}y}!kOie;pK^7j4rZvWn#BDhb5MQ`7_Hv`xLXvLSGXkR79LcUsl
zxhok8g->qTqwl-jF%tB1GTox1dVywS1Y}<gqz=yZNA0tgJEBKmL3-M6?@NiwSr-5^
zS@2GX<$ua_lxG)UqRW8V`r->mOP}pna?{GfSESh6Gzt@GweV~`(udY7Guw@ccVsP$
zx11j<b1i)TodP8FUGUdTc&BRV{YLOn|3#5J;>C+xc2b6RGa1Tfxk*AGKz1Q86hat&
zgl7PYuXKnUFk>36dvbNd_7c)W!oz&Yn|QCWB-sIhi?CJ0ddwfRTpO`!FT-n8Hxvj{
zTvkg#42gjzXgHl3C<GRKIx-?JE_~l~wV8@AGB$CkD6-yX_VWndF~O+thC`V!Q=rr{
zj<BVg`JJ33v()0~D!wo1ewB#)LjCO(rS*zG;&8-Ct~pUhsyVl-q**K$zZ*q5%C4V}
zmsx|Ks}e*NoV#5`bkhsI1jV$GP!T`LjFzjff}6>`>xQf8Vn-ftciN|jD|6j6z#O%K
zZ_0k^X)6a8VWgI|963-{3Xl#as?|?kf3pN#&9keO){|tZ$TDlfxC8{a*#|sJNLQRG
z&1LTWq>&_xIas>+ggK_WovU(hXLz5J!`$2Cb;hMRLt!NUc*&>0gsb(Y_aqM$8Q4q3
zgRd1I#nPhfIh*wHgb7@r;v#lWEx|(ch>8YWsAvR~=6e(~egeW?y|L}aHz^=?k!_t*
z@vFpAvz`2o3rEJ9lzQ7`qQ(%C8A=-2<NRQl<g5I{Lkg>6d14Ep{VHLk^3}>|w+<E(
z$G4b#94?cpf{isO%U4#6g5MWad<uy8aMMA(Z}M`>n!wwX?ZTiU+LcRIJ@z(O_%>WQ
zyR2NTj525*j2os$4e#V%yC@KP{ldVvuDuxx1k24>BqA_(%L!Gbi>@s@dYzW-R282?
z-VZX^cPt^-CH50LC7kG<A+!k*!jO6pY!pXK^iY{mb?p!{X=_};N0&0O&#NK2XL9$%
z_mI7Ju<`!E#yza&xXZIkST_fq@KFKOnqPB6vA4|oI7|>zq8bEe+RN2Dy%7$EALG>9
zG7Qg}sYq~N5ER9<H`I}6Gjq&>2v1yg5UCY0>D0yaz2W4RJ?8Fq)YUZV3rS;dXBH)c
z*9g1OkXShAL>qznVig?HF3*DnjR~0y5Z;eX3DlRf?j$Y~go&9}Buy2>YuHIf>;@;G
zUR0u-oDKRXt~aChW6wqooO=!VR4x%x)dC3G#-Z)>WFR*-Fp!%Yw$%hy<DlD?G#qP^
zUJy~#+(%M%bNfhmVuGo<Pk1XL6wJpqQSr+D(GW@`Qj@_sVP(|SdvI19)+CB*zT3D?
z#dlDUb}s{TT}{Mg9>EELcU8jyLKFQX*U6C6!`#Rzn&|;OmIf%}&)FOd3Vc%)n;`UF
z&xy%1{D3q4I7k*^t716Jc-13q1c?2`MXR#qO(MXvHpExOp|FfAFy!VSI#!Ijm>pmZ
zxpEu_UgGLY%t%}_k2vQ#J;p|3He06hhv3V=dvgu6=^^SIp_%;-uYA?Ji3|jp2J2O~
zap-3_tI0>ymM!rTXKWVdD$@EhM?D!IxtifV?m0fGCZ*}PhS26EiH%;8fnQRyPh4+n
zt@=J3QY0ld$I(lQ;AVU97_NRmv^6K&WC(i{O3uXHUV5phGmnpc3n(6S)!=N<g_owm
zRce|(yyl)pP+qU0bALF7or<@RFlI;diq0o`jCj;gFftZhizQ=_U0Ew?TL)xjdN`L}
z1;K*uHCtG6T#b=;-(wxdic(xDgk24EJ~(C|+7kf-Nv^!5P_CeC_|t>@#0-V}ame@;
z!N^XlO|!F0@iJhW*Dm{C#+v0|vIWvks2o_N26fb61=m{|%|uKNM{saLcXpMj9{8@X
zJU>g4L*A4h&n}v4XeggQEHn~J71W9!#5KGkPXs-UY~O8!<1G64{R3e*vWm+uc?EXX
zc(AsVQ(eiVY4E2<q;9YT#)}v%oyUH1FE(ywj+F}MgM=|CY){y0xuwc4GcRzk>gcoZ
zU_nUjVy60aoL9IM%*5$!E~Ob7fr%$+KK5bZV1P@EP-d|blJgoZ)S1(E#J#4|nj9qV
zfH79*2+Xg;g(5X>H;|tU83J-}BxJE=Os5pct`p&mKyaEGbQHC$`?M&8%1`cyr_b$8
zcMB=ow#vcM^`KSAUSrSycn(^h(5W<2vXfv}+epmlW&Bgm_xFYo2|Z@m)I36Zdlz+U
zjHIqW8b(1Rij*x4#tQmBIa`T|?j@iV#qn%(V<lI(72Aq9H0f&15Wucf1Qq68VBZ|*
z@(h&KA~@!my)bj>4GPbhcPZ?kxTO?`9OoV&%1b{Ra=E6b7WJ5G12U&~L!4ued$4pv
zS7F&eRKLF@^USWwpZq?uEGE-x9T~4j+(ba9ayXEP{47<H+{SElvH$49d0e(0XvOSN
zxG5<ffL#6kK@JhuPT_;C$*wz|dhqegqqI-z!C>!@f~h^Un*jz2{O)}anP%k&A1`=N
zI0N1rF<SG^_GUo(Rt!n#H0L}qgGM%U1M5VMv43Iqc(o9#6eX^g%rLQc+b5t(yaZuG
zbBNTM%$tPiXIvZ_C!KG3^CSnEpZp1*W?TFKVz9|)QTus~_Wm@vkL#8##lNpr2P@0V
z%ZOof7iMm2UK!i@5wgztm!EhuL=0w?V2UDgdj+i!pW<6dN*1eAT-`}lI}z26y%Z#J
ziJO)l6l@CH!LnI+UE39bx7Q!;Ka?~|AsGL(P<yQbLm_;2R+pxegN-W*zH5!6?vqaM
z5P7ANQL~huxvSza11*TqggwjAtwd8Jfyr7dqEQJ+R$!8ANduvILz3R>#Opo*EaqWe
z1@m}>6tY?<-VzBIL|71qrMuIC94uMnaZ>6=t#Ga767mnW0>%Z1;8)yEkfFvx4~)Ty
z%?nv`;%Q^6U~?OT-DB?BZl{oBP4j-)?I+G9ymc=IabQ=Y?VuGiY;j3{O{aP7W}jck
zOk9Hn<@*?fGB^)z-^t`KL$+y%WO5CdFr^kklr$a^Vkg;UapOyOfx7q65le8@G`*{K
z#hmqE9briyMfba<en`PB6QM2qPIp8b9YRS4Cg&n9c9Zm8v@+AyE%w7_JQ-1H6iy7R
zeUqF1j$ZnJW}#oIoe<3i2w{=NwGe^_gb}<z#44KLW`z(;sslj943t<&R4_>t*3la%
z5dH=_q^46p0m2unF}tkK0^>db2SD|3VkTIPko1<xfCb<~z~qiK_xlCrU>n`WUcP(h
z2QIp;Y9HWWIBXu#ZdT0~zVhu_zFN3ZRM3Du>O}$J>YMmU9!04wV_e`S!z<t<$5&d|
zPs5Psp(|L@#L7Xk`EYaC^45}O0I?jE>7&ZL*e77P4&i!2za-^ba|b!(JCC+JZSie>
zCM1ijl^hXX?uXHU#G>gvrQ;6D%UN!=Uar|b+s5T+nNlj2tSn~bQ~btDR+b6Tv>;>L
zKreJq(N9CBtyHxwdD!`eIfCPuc(ooxie4~XV+;ki*+ERJwr`GH9@VHFv)TaJV0ocm
zf{|8!AlPcKp4fgkf?Ar$gTUYta{Nz8%KK%vbMc~|o1oE)llgy=BoH|OA2)PzcB{dq
zgQ4W!O)X}a)WUUSt1z#n1ECNr1U%E}*v+6Q($DjfEzQ5eNXUIL-Zm_i!DK1RUGd!J
zo0zfldB`#r{h4{SSdgEj3vh<f1X?DsRk;_exL8kEa(o8gfev*s<498MGfUR@n{VU3
z^NCL!67+TdbefoySd^cnuL}$pYyE6=G&S)MGk(rt>d^|krrb~q#vUVGndh`*>&eYA
z1mXw6jU;$ZWdaZ2S8-sEbN_*60A5*F)jN=q1mXwdV{?Ye)M1=X{<!6;E#*xPu0<ou
z4H$Xs|Dg+1D_zfA)uM2}su2^rP({}(QS$NaL2!mMrjG*mnBvDc_MW1<E3vEb$^$H^
zJVf==H)iZ~Hxaa^Ez1M>2NshD`1mCe3yfV?61}LrgK(Xm=~v}S=?Fk~{lNfgR7wn2
zj01Q_whpUfWb0o@rBCRGHf&VNZcIPytC%%*h(qwGL2>Smiz3|wJwxDV;Pr7*KJfjz
zTXwi|ki!*w!O8UO8fkDGQJ6%66Is`W%IIjsSsA=U7B+VMFxDdP(C6Nmob!Q$s@Hhx
zcE47ms(W9{&^Od*wGOA9_L;cuTYtQjfQun6Tni3QGSz}3z@jf?6bQ(V1b;bnoK~n}
z%&qd)R<UX*9#@Y+98@!j@7iK5(1eDlG>#%wPFPf_^g+8n?bf)ivDC_4BziH{dl#0G
zWlqE})n!k_D@mRhp-ISbm!sDy)f(eAA|H^v<EXp;aFy2<f}C;-g&3h?NNHur(e%+A
zuQXzpRsj^pi;NO!!M1lY#@@te{bqdE{LmB1G97ttz(HMw-~cz0blM}5B<k;SbyJ<!
zs#P(r=`A8#$zr!$eY_%*c?EtN1hTrj(r7eR*5zLaJWcti(k#go1z)KZUFCdb17is=
zc=FCa8IT1U%IYLts}yPKtFG;|2x%2x%Hf>5;Onk^HGf!i>)Qw;!RM__`FrKc)_Mi>
z*Q?E~D_h@sGyVG@Cx%q_|2z-uV09<J0C9ysdsHkAf0{a38r_sj>Xor{N=4A*hy3$h
zq-dS=_B9ujJds5Yw;UK9#;;E{@YAn)_3PYte~6vERb51VvUXapQ0Wt^*9Fs5r<hzh
z41w|S68$M@<mdd7q_Wt=h+EbJ-+zc~gN}$L_0IJ{9KpGDvLgu^gGU%!qRuG3VLStt
zo^)IwWUPgYnXvPO7x#dUbj!wBQJ2>xF=)%KN@Y12^w)AP`oLQ&EU;bc$M|5a`cAdw
z6;E59e-o1(BT@`1ANM*>KH&);fWny(w<Zs&Evv;tQGD1tWvFt=B_g7UMQ6P9pwT`(
z#k~-o142ylqkQ`|*i<NPyXHwU>1}!bwS9-59vFZwtQr~fiMKI=))qHUsv3}<QEVfA
zxy!fi^{P*DMsH8(1D-25zmQEv65HvF_-4HeiSK5paFypNWkK;q^{k~6Ea4$mJjTd!
z^2MBJ>11Ff^+=MsA>Zvi@WkCcez5BzuQ^SmDTlcGVu6`cQ;GRVFRZY$m4C$_%MVDQ
zx(x)Ey9dy4<QO}vJo7=L{K3-4^GbbU=DMHaaScGZp&eR{U(CcFwc^rx?}A*kW?w+F
z31SYS-L&=g7t;EF=8Ns(7UMi3+%F)I<Oo5&s{7#uVwBAk445qFDCpq@y7duCg=`|{
zQ&jw9wtzolG<tKy)Tm=LD*)YWt!FqE^+B~`<nI<Q4(P}|SY}M<ncaT|;~j3~t%)Q!
z&n%lqq%n)!OcTjQ5PH@d70$ze9!wEPoDT=(if9|TE9o_@v%BlR`|^CI^sv)n%VXvZ
zyF<lIir9WV>cVNFD<P@f4lY|9d70;G#HP@iT2JGodT%t}K$5(_^#hwE@7p{7J9C%)
zR?)SqCR^T0xiC8sp`^K|c5K=_qz$@{)j3IXq3fZgw5ekA=lynl6(8VZ>exhjZ<OoI
zK8!LjY(~AUGK^Wrfd+z2tycrVCsfi{qd3ve>wo<6OFK5l-jy;ADFm$KiOJMpVw3H?
zzBn8wJq-w(b#KAW5EUClZ6Mh6dq_r98vB;XUj$-=xtoLzHV<EIRcmjHa$&KmO~|)i
z*Ql)-XQ7g+nwf8IO&vY>g&p)vropDB==<ipHuGG4NCj1<jYZQ*1|_OV+E@0*Z20qR
zlO0a8_kE+>NJP6JjUSS6m2ip)7Uc7*={7<OV@gRCP3%{LQItMSife|y8VME9Mha`H
zKjfe)p>HG54P`XvUkzXvU~UR)^1qsXCTEVOie>?<5qnXDxv8M(0BZ(T2}uRnl-9g}
zNG@0GYMvkvsk=%gxz?E3`rp5-tHD27E=gf!_xdxHMg#7@e0k^Za&f7VKp{9Md92=P
zK>XV#?`Mq!vNNP)ayJnE>6dr@CXslx+^6)s$<%uI^3FfOS~W}>cpsPwZ@s*8G-Dxk
z{WHb?+{-&}vUm-nTCUd8&C{KKXhvb1qQMk?uX&QDdVI)qj$bYk;OVivoQsE)s+lT5
z=cL#oM3aIGkBO9iCLWNQ&&GgK_4%iQfSEb>SYmqo>bfMvvxyO8x|b4^Jzf3rVEG%D
zrsL72z24wbckrk>Jwtx>!G&c6sL&aDTfBHLK+C*ua+*nYBt9@H*4lgfHsR8>?_Gym
zjiygB#aHpnsOXY%TU5yZ?4DH_f3bN1ocl#nxU3^iKY9A&&zTZWkAE8-93THsJNW10
zjDOz!%Vzrf@!wgrd+&-TW<PTwYxthVjqSlTTbr@n{~n&~FkPV|LZY9px)LCX=>K|5
zZPKFXsdWAR#uM;A?^^=?Li3kEz*mxn%7py!SMy@pw3+S~2#n<Zm89U)<I6(!zocYe
zjwvx=h|b8l-dWoJ?pJ?)!7Pg;@intqf-Cdb*g+*|SM00YP>J2P5inkiMB2U`l%yw7
zCNpCUs>}E1hOw~JttM~{kJn`TDgbasL~~M|Y(1llC&DUz&BnpG+dK-t;grSXvqFLi
zMDIJdU7DfRk#%8m#Cx3|_2?Oi82HJ)X|y&iN;xHpv^rzEIwJ*&%pz{HBaV07h)=S8
zR|5wM;u`#1AmPldDC64fF|ZbZpG1x-(!#bU8*sPiNuEQPRK(8cB?NdIKIgb@AudP{
z31b<FU)0z{z_J2{`R4!<1~&1OwfP&p2lIRc<MskyO83GcMu#5flTd@%yGXhjqNKJz
zNoLWm4sp|ji>e6xI<Bgw=3qNl5D7q|+aCPdd9)IWlX53<1X3CZb9Bzph>t;x?W$16
zIJ|lbOK#@(V*KwOSTX)f&9@-Nf3v`&HKM#&)Dhpv^UDJDQVHq|mNJ#snj1iAsZ`mu
zBm8Q%5U)00MftCn$|qXo*I5$F%<7N7R4D!Q`1@2b7iw5ap1*%mZ5{<#XnNc&DjF(D
zx8U~f7Q27(?t+V*-t1jXPqF$SbIB3%BM$dnTHY`dFb~<Fg1L;ny_2r{K19y9=tpBP
zpxhq}bcF-oqpDp5dHT8Tysem`Qj+b6i<c=+fP`VvZeAZD70J!P1nwGVxAD~|doali
z#<0!cs1Zj)nn_sC<*wmc7+H+o8IG>Er{i}2_Ku3{@D0Lcw(x)UN=tsP-#gu6{o$a4
zOW*1LrD_?!2SxB6wvyn%KZZ&`N!kw{e;^B9`t;7IE1{d+PVgpsXmY299o9Q=Kvui4
zv=p%YeSby@;u9){V_fi8uQf^n(^#ufMuSH#L+TJ5q7>ho_WGTcc{dq8M#{plt|odd
zCTKlv?>}rGBJN4Ex3=1o7H^nBx@3qzI3Cr#UtI|a0f)|&HyUre*|@x71iCaI7HPn5
z5lf6DuYrE$(k7Z7pegsPg%mKu*U?9&L+V`c0y78d>R3w5fMYiz*Ky6%@PtgL>_GmL
zfnS}?ZdItPXPQf_u-Mg{QT^b|tCfHb+0>Qbm);4-zznJ+(6aZNC;2>Gt&eohEmXGO
z9vn{DfUw+RRfdnD!0Y{DCOv4dQm?G|c-Q;y-*`WfT|R}>HK=IuLTTvX)E#zcS%jWm
zD)y+2LvIW_G}x{Nr9sno@t0CidckvLg%yFSCuEF9iF|R18AS61@2V%Me-Kgz5%1lR
z@>D8JM7dbQN^UlK+7quMh#s*od}?R4_IjIlP4U*+*!Mx}<G8X=2i__YkCs4Z%tOnN
z0ScxC>mY^i0fMQ<JcMQV2l}8YHv(2`Dq!JBq$93HA~3iK4r_@F@grnf(++DnB!%i!
zD$)AzTl6}c0yBb7*;}?Sg$`wf&?`CqUi1ATIH7%57}Uw-E5o)ivE!=xtYIqaFVxN;
zGK&f-JG!}Mm3@sF6kZgix=9%aL?R#R`QN&I=DO=L=UFhbGAZktaCj!<aSOVh^}=up
z$&V}73m!W42)-Cn`o)6C@}trPyUaGc2f%R5;iuwwZofomf8ccnGwAUEd%P|Chf^ox
zdpf;VC@C>VX8tiGwp;B%*#WPNdI-Q%=I6MWA`!03_JYQ$&*^|PlZ@n&I%6o0gO!O4
zsi_*z7Bp?1M+-qcdyt00!H8?d74Tv`rswlu&~PS-SQ@bgKN?ZR#eBR9-%c|-7>^%<
zQuR!0l>apqE!yViQLoeK4kQ;rGJ^?4G1f&A$ex2kS=%-h-s0}ZnI&>Q<I_L!bP|pa
zm1oQHmLf`gPYYxGK@w4T4+R+<z-BX6(0XoTy+HSfrw>USHyBL|zUC)5<wHgSZezq$
zxcO$8)g%Q(Z_v83*P9g5ASg`QsTdN~f;n<~sSV|^kj{Zci9O=YS|{C;7B)SH-GWnl
znR(E4<L&RX+E^3LiNOYAf-Hz7EAmrIDnG(377t}OpSU?Qo4fhKKL!)YAGq-t176&+
znd}o<l=VAWkKv@NX--{G*1+LeYNsoECr#fNgUTr!zn|H2ZK5m+^KR62ILChO0JuH0
zj==Ndv0VK?jS;O$4`G^*yS>9BTx-cxa#&bsMv(OitBzl&I)0e}xve@GlMf1N4iRb?
zbrNK}(X3Lrf+A8iuZY3eJwT4kaC6i6_~9o@;+l&^c91ek4&#B(2rjL}&@xtY;s#xm
z^7Tu73=S$Cqtalt08Ff$x1k-i6D6MGEVYcRHGJI&7P?%3#m$HcP^jnURL)*8gmWPp
zs(~*SP0+VqQzWO5OGFF$eK#hVoQedDtrSmj>Ad(!aVHk6U(3QoP7u3Bq(U$v7|xi0
zbwdzCF!JD;00|_XnqZQ}K5=4V5BBMSRob5&goD<it{?gWvM)FvatgfoIR_A>uOS56
z|M1!Tg@CLcev$@rHV+H)yV6f@d^fZ-80atK$4n|&&E=W8YAV}a+L@Wca)X=VX@9?n
z4%&`nJ&c7`c}(YGXfQoDqNZM`w59jSSR&N?<(dq!gQP>6UQ4S7OlJ(_nvY7B9<+wj
zAvI6o3vc)8K=ab;4IuZqJn6OZuHVABX;T>2^pUFo5fpYRmK(9s<?7fT)~Z!GXD^_A
zW;$Ke-YnVf&<ry4;k0RDEP<WgiRJ8=hs@>fc<-7jNIBGX`>J#4ftF16@41<-F#Q7!
zF(<NZ$o|cBO}n5QVG=+K7dBob5z{Ybi)EYEDmcuslx03Is{{1gXThyP#0vY5eMCkL
zus&x);w!?4-m*6+xRGJ?cyxRP!1CHg(8hcc5o+|2CT5f88KCa<+L{6+kl03&n~t{7
zFjOadd@Po7?CT%5M;#u!i~~WTDuX2u`feTsrQK-4irO9TN~enLwPcJQI4WBPQEaSL
z*5W9q%Dk!xt`+-7NoZdkTIMQ@-Kz+3z3N-?DNN?HwcqFY%(>fhUCMRAEQ<VK<p+bV
zxe|&@hJmO|pOO1lsArHgPd{1l^2n$kareC<;!d1H;ctf7JVKN)wGmkxxcg`!Onll6
zhdlI5_(2EMP)nxPsd~t7gxf;8CwpB<)+tKes0*3T-9fNQIGW@q79N#qRjmoPaWlyM
z?qEC|tzE6|W2r$iCP|+b_FEy&pd5b-=p6|ENd(DbngDbN-c64j*aWp<S0W^dj8M=6
zh%cokDC;VOg#s#L{^Y2~m1O&43Q7J<)bW>+Q2R2$CVq9d0}L0WVCIH6-(#SR0u4rv
z<JiL{(;-k8DZtbLp@S1$#MrtF^L%iN(bcuCoIv>ThF{EJt`x_HA44K0c?yY5_O2@%
z3`A5Xcpg<vqT)$`K%|P65f0S+VYm%-s(u)>UsbipD6sJtItsS&utVTH9TT>;s@=Ec
zgyv;&6N1=X^5cZ$2@!cMsN`?4dN4#w4}wA9e$6^xYaZfY1{h%}sJcR4Seg~jGaCzf
z6vgw2Nf{8GEEZ2Mqd*vn2VeHI_oUl5k~GM;qxAxjl;<oGNgk#-TA&UNE0Gu+h7VXy
z#}BjiRQlyjiIxRI=6Fire%^2n!uairNIfUga`8%D#}hSgjr#}@fk79UVl*#l!Xq7F
z!gglXu+{=mUFJ5U3?<nM=9bkZTG$Rp(w*)TCQWd;m`HJUWA}0}DZo-K$=?K%I1bi&
z@E>Yf{$MXqK1i<{<hL@eM%HQ^yU4@|fkIBT_8VEywV`^8VN}tmf$&(elNY)iE$uV9
z;EJ+}WO0#$CUhzU-54LWPrD$CE1s)VQRPinFk=FM&YuC%!b5K-)g0g=GmK`Q6;E4?
zm<Ww}iNda$>Tw~DMe=+BM4}oFmsCfzIW_{G<r!)8b5>)Vd_iH?ju~hrQ0`8VZr(b$
z{tfCJggO#?h@Be_bUc>b60SoQ`(AZ;I@x&RcU%i38d^!jRmhpjuuNk|`D|Eb6Z2^s
zT7LwGu$IA@)qvL-Lw!N?Fchb$(P<$KE7gCqT27XTkj1E~d<27!SK|R@8~nsI93%Tx
zLW}9YN$++yHXrbRQ)^@E0aDp-CJKnHuL*OK=Hy|yEj4y*rxoPz&9e=m=W~ruRg|gF
z2O(tWL*rzb9O0~{XT0Ppq%NnYca1)cCLu3pL!j-(XvYJQRY}opjN#|9y0&{2w5=fy
z_nGEQllK$WloDcXhm?TB9ZxS~0!sUH4>DMapYY7y5;lC+z&;U5dYQfi5aDAb=X$jP
z*MN=995dgh3MFS3F?P`Uh7>Dhf$b2*GP&50b-fFR7ztU`Eswj$f}+lP{Mfqj*qh4n
z!+Hc#mMfRcsZ^c>L`ysFX%}`TW5uvA1~>dUUV!<njmy{-Rr9h;;$A|o8UxsB*g)Q?
zNON0HWEH?hXFlw!d}OK{*^OXv?XT5oS`^-p-EY*hjnkpav}YTxD{{&MRY{VQX4F)C
zQXyLpHbzy{ulmA~sH$z+s~t7u<zfU<MwWw|nU>*Bl&*=!a-u?24}c>ymK~Y6401u;
z7Y~8{i58BnBQTi`#aqeJ))W#7a}hKM$aFT|M^STr*b+uzoD-zB-$XE%5eKZLqG26)
z!s(P$$QdeIHySWfpTa5C3_8q%tSgB)MgIl9?S7fL>M@KuV!0;6TAM0!#C&d$o>=1Y
zpvBC=GQ7C79%jnFo{hj0>Oo<&#W7Fg#@HiBc+M(bIvX`f3`=*5+1(I###bs~^*e;)
zAcQNv#F7@WI>HGL2H)N&AZ%oYOFZXf%6RjTH)btEeS@pHhTV|KQ~41#2+b{GiBNCz
zx|t>cRIr*c%y1A~7ozH_5QPoI;$Pp-tt4b{r(Kvs^jBd%+k@T{>wMAaoXREMa@$OA
zJf3>{UxX5vJNk`FR%0eD4S4^gb%eXZ8|wJc5N*x($}cH}x-oERnuDOZ&1E<uo{#C1
zXniq<Ceva&`5hOI`fL)b+i4#m;I{2@QGjgVV@PI4<PaTiXHSY1Vok!-i5|w)LIoiB
z?vVBL2h*CFQ0p;+K%t%_p*Ld{Zc!;_#z(ya1bWy~`ej#=ZbJE7udY6z6Un4^I2}%L
zd!$1msB1;1qM~(iicwz=_qgMn7RDf%%d#%}SxdmY>$r7*1BUG?4K^yHV9^hI1K-^4
z<CqBvo1GGFEgWDy8t%9E)Q@ylM@=izZ*3*jT9NowLRNGIyL8$+#ZeC2RIqH|!g_ok
z`{Pt16+%R(!=Id{QMXF^yIc+pZ#YA@r~<mHV=VX@BP1I&TjGGlyrAQI<rtt7gImKE
z&oWbZSj8#|t6E(pChNYWTw*3j28^mE3kBsZG6qE1C=?GT0d7l6(P$|T;{`cCo?3~d
zFcWx;8{A^`Ac7Z7RaYf?8n=>3h;hR|@<juC@O)jBEbFV1*0VoO-kztD(+y;upR3g7
z7`SeV_2E2CpG)!08=kNH<aPMOE63ZLfeGFbBJL38LSRmbn8eixNmsv(S6hezlkK~4
zD&8)`(K+uL^kUemEuzC?Zc&MkXR%`hPGkqcQbRHD*6h_?JqRxd$GzcLm2R~40EZbC
zg#FuqR*Z9O&2%k76o=V`VvuLEIvLmSztGcUFpAwx%7g>f>h~UYy*B7Uvj<td_yMJ`
z!#a6o1+t?E6Qm6Gq2~^G2rV1jKtI)Z56pPpttR^qCR6bXV5G2cF~H21>~%VlzQ*cW
zsWWNy=r_Be=+0y`Pnx(Z+`-WwCO}(zQ>%kVt+(WmSjuv%wd;-LF-7ak0Udk2Z1uPs
zUneV@0kFSXm)u}r!!d%rl}8y!Z!zfTS2TZly}Im5wZ{mznP{?OG4)t$fo#FoXrc{8
zjI)wD0EjJOazVNkoV1aebQKRksdbofrNe7a-s5*Z^J^Un<0PBen$tF+st0MCv)7U1
z1KHPi>^1A_67;}f$)dGGQkzRTszD!>fTmYn%ZW&cDup9zn{m@Al$g69$Aeu;u>@Q_
zhgf(~C)v@s%A$D3wMvL*4Y{7gb!+<)kPm$52<Z7@F6|-|qSEYUfn8{%zZ#G<sski+
z9e~V=mx!oXNDBr`o?yvzh{RnFix$TxF1{erais|B_*K|^hABnbNf$Y-)E`7GK&~HA
zkI_W*`vq*7V?@zF{NM$&U=-*G@znC6toB_Ub$8L35JBC^<?7MEp(R#7Ls4km3<7rn
zulETO)J=jj<W%LzkHuq>oU3Yq^@IB7KE8Gb-}W><ILH=RO1XW$&Rx7{Em_gMxN<C>
zlE!Otp;y%V$j~tA;hy2@LksFd_&5(Vs0606J;XyWa2{Xj-^Nv2WUSum3f;tJ(6C0e
z;b^uuRp2I4t$wIj2IKvq+|tSeaC6Uz(f%0UX<b>@t0RdWT8F~WfkRTDY?F}u+@eWV
zg_m3>i*A^GQ8x%FVoAqV4jZq*)*&-CxA7vW#o!PKAnM9qBz3WX%zY5eRv{JVU5^->
zvSSYZ^swB4BMpbS{b&f!LnIlX`a#2>Lx8tVoBPR8kQxe9A)NnteNzaU>K|aPRr34-
zR1(!n+$oXMA4-(9OUhK#Hj}%1vAWkfj4ayfLBD-y8Z|c;fX~}V>V$C<_Ry`J+_*zq
z?xq~RgZC#0N3X1(is80^`|p;VD)-xR<rruJ%gqF?p4ZaLi8%w5v4SmSb+M8uU}}37
z31>6*HK8~qu2g^ilsLB-8<QH`K0UV!VP9Rzv7}0Cvrqa%lAr&SUsoA7#|o|%PM_fH
z;zz3}iuGmzmds2887#Adi$K~<C?VJnh%cm&ZI?MK_e+W`r3bJg8~nj~5a2)aG(W;&
z{z-m#*y|gk&_aHcte?bCYbf0oIKj6k@R4<Hfzx|iFCNo#oZPF{CEiOlk>g%hX20?S
zR^MA=U8VLJFm^<qkdP=B{+xG8?{+KUzRS)U_tKvZ&N!hLLF*4uW8wj1&0*?%YiHC~
z5{4hQ>U;fCU+BLM^{`*%qA{Z7GD`Yj6WvX0$+qoCloS#7M#h*2T!%oaKLw}n@j}+X
z260skC$Hx+M>DYqUAU3tqIZI$OQ;U9KOJct(iemZf2yG%qPe4Ky5;$k>H|Q=(75<A
zpN8}(myk8U%?>}Y7bz`sNn-##?}||nn-JBuWWDmb4G#Bi{qD}qN)PWJA3@`^PK@He
z(es*TBt9^-(eG?0Y*xLvH;P|+N}YJPMR4b~vq%GUzVr0(^14*l?$P4)Ey8Kmv2-k}
zW}b`JY%Mv7JhH|-U~s-5Nw;75&?X!G?VbOVDd3l{*JZ}5$vL`IU%~cYbBFHy;mg{@
z@6^|ryFM6DyD#tj3&i#t^`M|yFg(!YE&X42evjo>>dQxg8K&0%-27iq>9zVwSxAkk
z_+K~w$JrILkSJ5}KWY9QRJ>H*VljfnRQ~sx|IzHq#UMvh`ENG=T~xkO-?Eb<y2>Wu
z?*__WY5omB*{p|R1K7*e|Jmk0kLqvMR}XdjG|1hc@Xs{=GGNRYv@B(UsrXMe|0z`b
zW_{yUwS<SRNpJf}^Al8At*;Uj&D1(=o}$*P^|eB^-PF0#yn;F#^>z01J(j8ZX7k&q
zI%A&N6oiHj-)#QlCKK-~waq;}{srdaJuf~|Y2l7b>2nZQ$}c*8QuusaB2}K5Ii>cq
z%nYgc91~OO&k!j!Qej@T8c8J8X3(`D`Q}wANLi_}xHOmIi!2sW@c9-f0Wo8?`Bjl6
zr^in;ne^JM9KVfy1fba5uuoTi_YW?;^b#c~jzahLM#IOjzG26BI)q{d=@zAcj(+%j
z&?doiA#?Nf+qdp~;PV}>SA1}b+tE836Ennpd%n<hO=DTeH4Kv*&btQiL9;obZ@v58
z`*&})ZoPNo=I@quP--mU=hn`(cW<`tzu&z1-g&CsxcSbtAKbp*(&~4v-M{{Q1+rMH
zzIyB4{cCscKV#jS@7>Uv@b}^2N+E-&nR}A1xwU_A7-R&~y;E2K3M@t!YT;-#e600W
zjKyG}iOs^Sg~-gjG{j}O?cOEw(FXEgZuF5NwYz~3+5Mx<H<naMgw-g7+mO@BF&Cp%
zQ-5i#!ohWIp?(`No=bsE-Kd<NTkbP2b%Z^3Ye*2h9l}fqp+{E5v;+?a!^Z=gKvsx|
zj07PlrXs+#HQ~ktz1+w3!+I}c|I8hfFzw+7!uL0ohUU(i6nW&**pb`hOg}lYb_F#N
z)gcDjKl(bu4I1v2?M{~4Mw0{wsel7lJ=s?vp*1VwD8d9(`U7KmAGRk6uTU<0=Jv@_
z8vh`LgXXqCL6YBQGYwxAE13)r0VlFLGJl)q3;<J%DZz!sf`kypF6_w>+Xw_h5%{8J
z22P*3-Au}2(nuCRpyWqNg;JJ-N-Bhe5n=`2ctOJ~8*s8z)hXsSDW=I-HZI(#6>}L<
zQBS~X-Tw2#gp1ZkQ`|&4ZXx}8r?pzj1|4!)oWMbCq$af2@VL+dBkwd^Z(az<rW7;m
zVC}xETG#JP?F8i!j)lh)K$|KDMk1?iz*Hb7J++#tV_VoNB+ssiA~-<;anuAT@ubJm
z8CIo{KI{T~yVKt5^?Q>u2LPOEkqJU3%HaVV%&f14+S+v&8{tV{BxMXkr#S_Tq2Q;V
z2M65fgoF3e0x5(|&08l0w(b0mCvn^h@N@itP9LHFaF*;QgzpmEtemS2H;TE~D6S-*
zMuMB~)_GYN+e9>z!bx)=H`Lf^LXP9<<DC8nPIcr)w4r3@j1l@LS@;8%^|_HSR+&6f
zJOT$oDWOI=tZNdEs*l7t)|d{)IF;xkjP)|k+SS|?#m(AePsAYnL|kGqNjTq>@ffcW
zF?pKCs94YuSU6;#LKW)76VcDZoL#aOtr)Y5rsy4JuokA$!i0&jUz93iPZnIXp+-wL
zC1%K)??NQPnZQUvN}kU2<mAi>EaBLIQw=2LbC={>m?1EhQ2AlxvF5O;WgO$0lw&~j
zYg*2jRXbY?O7K`5mK3i8hW!l^_@}_EsC27oJZ%lda=aCq8>5^L!<ts<sY5Lweu`fN
zMtb1Vp~0?I2aw(jqRAnd0CCTGKpr#l2Q+CY(=wF3x^zh;ig_bI#7k%m*`BK{ZIFz1
zzSv44VS?x+qRjifXaGVU_h}+1jV$yV!NMw?T)5&&h)SBgv}puo<N~sUw}4#7eMx$C
z1vQ-8x37J0&BXH|eX>75xpy?2_B(XqIvqmbABE0OA-GV3wE`i0=9owv9#7=vGjU|1
z|BL98)&@}_Nm3O55^6VIDB{r0dK#IfOTBGq6!0E36JAB-gl@c-BwTTBv8xh{0>P6-
zGGkz3&BB&$Gp==uMM#yfYsJFCw1O8C!&q0G45pmlkmrmmxJly?483hVKET09shq^I
z-p_JI4lAr<?2E}Gj4W{!S>WVV4>ffzI?`fjMaJt0NEUD^JhV0FqXs<JkfwCI`c^DW
zAynYGY&Is2z;tHbm5L=T=z;nLU-s<H@si?pnR6?Ayzz-zEAM<<mk4uwiv5JhIs#OT
zT>L;0wWxZAlCqCLjnx-y!-BRJ>)-AZ+=w7!yoz{;%{9U2iBlFGDf}iNUy?EmWm$`f
zCS_{&iLSeY?qiKPGx513mv5MFO|&bYGknqfgbZ<WmFUIJ6o?ox86Xf_0FXZDgUYe`
zW=If*6tpskB<Nzj>*AiWFlZ>|JW8Xu8n`pFR<1S1Q{xM9Ij&FsSXS8yk=*WMRoGdF
zg-}n{O{@!Y7oH%emX%yS$ABXo!SLfzalUZzs@0I^CU-)Nc{~wnpk)obR;^xcz|YY&
z<M->4eXkKIkZXn_SqD-MuxfhRZjNk`8**Z^U!IXHc;%yA#|uHA1S{|(7>G9(xd=fv
zK7p4OkWA!oehtF!`9dd$b0ebXt&>o2ju>_0QJ{3o`6mt9CNe3h;>jXSmnKL~5(0Dc
zDx*X;M7C{LC77a?5+<6;Lsogm&heRBDfDsF#N(qLSzRa{QLVwr*oQ~a+f`u_30SFQ
zc8a8OESPqE83q|et)?R-Hg5+Z!Xq6ER<)TXmZU}Xd1j`I(S&&+g1*QzuXI{^DU7PZ
zrEQ~tJg8hCSCRMaiGYY+RkC4qolN0reR9$|>7TY>**e8N!7#?E)lVFZ%hgbW$q;%c
z+6SCCbzHTae~3KMOc~NouIO5hdI_|9H6s<Gme!ziOxpWLtA(qbUR=|KmUC<#JOc`z
z=`3UB<M%0c<F(Da8?SBEb>tjEG{jj&fgNx5*ocw5?Sf<_Bz`Op#n4pSIhgkQt12m@
zN$&Khb=a7qb49Da8fdY(!9Em=NO?T!F$(&ZaZTRg*TBJ<sFD7fE-NmOHM8KpwiRRr
zx3zOk%)TKgCu9w1_d767xm8>DgG!__x^{FFjH(xQ#1k;*n&EY<IIBJBUc?7woIhh{
z%MkLHWLu2eLbJRD;fY?W0D<1+1?4BY4mn0o8DF<vCWnEn#4mDM6`+^M7>W!Ni3L1W
z%006_2*S(o*v-pqUFm0In4Vi*n57uZ7LCu}bS`hsA!|5PM8T{=<|M+;&KS5Y9SUA>
z3^wjQmxE}1Mtb`le5q!v_LfRGZlwLBRG?^Bs3U0!K-|9t31<O00j5(XL{eWB(W4M6
zliWWI+%14pdvLb;#B5t|)l1qx>;lY78WF@`k`_r!fhX-J3_S&T@y&t;7(ZSa0JVx+
z0i3>Wwfnt8!b>(_M95N0mSUPR(#p9SA6Ky=iS%_`Akc6_t)!LLbDmY(t&O-%ub35|
zLn*>IQT^JE+SPL#ywa1R<efZRV@>29NGx}Dgu`i~6tY@M1T5yklsaw>V3pW^xSGCJ
zxDJg-y<3o-2~zpBZW!62KuiE7Ungpj1;a5Huy)L~jeN9O$r3~>6h~ZM*%4PIJRfi=
z<Oe+0#G`T6(lYkhIGcC##+k@_Vw6W&K*ggRS1XLM)EOtE%UbqBxe#4qU~O`=XXivF
zj!F4!wi2`*y|}Dy{(V2w--dB8xqs9<T}|r-tg{{vD-lYzV+K>bnE^+99f+y}rr%4q
zUmhJd;MIn)!<r|^SPd4S>k!6QhCaqEnd9k89sG>LiD{1Ih}V(MMALAG4NV2I5!>72
z8{B6Aq&2(43&z^E7kAVlEfO~G8R@Sh9@o>s#Is<mgbMXyF(6%V|3)en<7V=i6Tf%p
zdkGmAaMIDTnEh*PgK=qP9oh#|26K#%t8^?Hiu(di;p-ws=N8w6MpSInfsZ@7kiZCr
zwlEE2fNmNpV}u=c^ASw1&%@)&PhpP6m^n(#<nXwTVd5aK&^g$q(D$obUg0&zlzQ|S
zH(Q7e%|n+{b_>O-hSB{W_Yg++`X9giQu7PVU;Trp$6vc|!|J}={09iWdBeM=iU;&U
ziKI<9PDdboErk$xWSBUgc5@F+Wd#FY5L56$^Dm*`S5k;-JT_%MX#T}g8CB0rnY+z<
zDD&kMj);C!<h|zmDDs<Gk<h8g6#Wa$CW`)gS`<rU5XbkO=DR5L=d(h3jh=CjG=$r1
zzEx^FI-WE|*P545^fywl-W}MKTW+qDn#V4TPLsyYYx?t>&EG`Xuct6gS;>iQN`Afh
zb(H)@S~7HA5JzQG{A<mxq4+Clap!PkioMc&1;xIa7PGGErbl0Ce!~QN{c0Yb@bvie
zjPm-G1jf_j-{Rw07)XIddL#ws5J}2pW1bRSO3fTXDUtDF!Vl%fVlPBPfO=jnw-lX(
z8W?r6M<8TV)K0(hTRVZ}?QA(vE)vNUDe{bhi*={R|FK3keZ@pJG3nAtgMQX<`0XAb
zjOqXVt(RU}{>G*0cytL_*)DYlkE+wN$<c6tkd2LtP}qh{ka@M5P7XG%@W}-m&4f&g
zNA|mDswSREHmKT={0k2>7a8?F+vkG|CbHvNr~N%aW!1ehTJi{adT#OP{RDc`xN-%n
zcK>iVLO{$(quK#-2R+1ONFWx9Aeq6Sx^s4)IEeBx>Y4;aNE>9Ek{h)I3pmU0V2(+X
z#*^+To#k-R(3yBvDTN4cxQE!TM`(q9`#oZVlmOl7=+>Qk)z;-Lb!lK$A!)G?i^9ay
zC-lD<ZEtorUWbnX4%NonIJ|!gX)qgaz4aEBUNq0Qn{O6QC^Z3m3wit7XWMWU0W{)b
zE_g-+8Z~jg9!^0pO@YEJ_1%uCV{~g=WHjj=oYmw=!N%V@rFCGaZH*5*29)I7muzB(
zU8XCCcEK$MM#^c3(B<d`a=CH2`VMkokKzqk=I_09Yk)hHkT$E|ZI2{*K3zQ!IrPa=
z1IrC~YqdWLPW)o<pw*)AIjYF<t2Z`9@bRX$0U03LhrUy7Hj)$X`o6H1WEnr{;;<LH
zRd5ednBvGh?3aZ-r*ZhSy^zK?3ys496d2r$0W%$Y406s8;dmoZOQH++yKo9r4wn>N
zFj<fd*Vg2e8)i52TD%A3SymIIrV780^j@w7Dxch(3vA9qZ6q$hLg0-RA?k`NC9P<)
zyXM^k4kwI>mi+W=6NtV5NbU-NM0rbW&DNP7K(egJeSDG$#iR)l_b0V0wula*I}f|7
zo0r!f__+cR(Z*eU^pcQ!N+B2e2<Q~dw&h<H2jI|xFd!26_(>0$5dQGu!H5a^G!fbD
z-Ng{~+fR*(bAnTI_Xk08IwBZ3)1l{OD&a)m7m;QlQu4%ch!i(64aLm2_;nx!Q6%`~
zih+Gf?Z%_8w~nm*P$Vt-E0UJ$20}WK?Gbn44*UWWgxYbWYy&~#F>n^j8feQsBi&8_
z%mOkUje!^dcSXeau<b(}=sfHI(C|nR*0r5`?_!y$gCZ2v<!;=az!ov-wp!wYE=h_f
zTp!pqerh=>tVsx-`8f9vN8ka{Rz^tJ4bgLWmdJm3)&MNSiK{E7U(R*LFV2X->*_V;
z0e<IcOoS4Vl2&&SUP<A%65N@oeMYa@Be0!dFF$rq{hrvCGN_`Q0cdd#M20Q9T-_6x
zxJ7Hy!vgvv8D~dkTTI(pp;99v%DY|K61XJe$tet{xLpLEkBH18GvgZN5&g-LPtoc+
zbyppl77(-5YCk%ZMuMXW&GvrU=(HzoKcb7ajo2mJ@Z|Na#c;qCwJMKm@yN(j$NwJD
z9VDz|(n_SrYCj=eaDzyNPB%Qs4jYnz^`nN+tWk$7VH%>*if;(FAkv&=985F2_HpmV
zbkvnFh=7TqgWl0Bq~ofM<eP&-c=H(@T;%~__+W=?qeHm$L7(;Wlvo<*aIQsXDcuk#
zB`G-DeaXIQ-%HBRQMb#>Fu~_~$Cx;?WQ69e;c5cXQoJ|B;}K@$<QvQc8j2hg43lj$
z@HUQ-HuX$7#4NkopU%_<PNxk{5*iiDClEqQUf+|trSgSCk3aZ9pa|kXD$W@3#Lo&G
zwfhGMg6-~0fIMg<N_|#L5F->d_0z1Y6XF1-l*#6sG1;{3s>*;%^s#*#{?$p1=Jg9a
z*a0?(5s%9Xg;<tfT-glixEQg|ilAQt12&d1B<{Fy<in;_JHI;x*-yGRN26g)YD<<~
zaheTow*&uj1A~Q1a(HMPP7GrA29*|}Lf8=FlC=)h`+I@~KycHEiPa95YJ%tIJhEM+
z0|*<h78xO^AmSP&{@Frlvu23exbo~P2FOcIU2u@Zfyu5%yW>RIjFp8YpLca3jOSTl
z7QtcBRVK$}zR+f!^aBbTZYWPdmS3)3?1Dp>&$8&SUiitF5Kh;+(ir8ODslS^XJ!)b
zkFHu{L?<ec!9mIS%D2S?g=LEJPL`R<w5?!76oNKdtJufmD8|f|J~f4l_LtvZw9i&*
zJVx^qFxv?K);2T`bUj#f)*u_h;1;!JZW0rbk?juhODIm*4faou+S^4~!79nR{4CRp
zkjJzmC<e+q&}c?I>H*p486^h0RaFByF(|;OqVBGkhb!3H$4_+r_!U}PcARQtOEJu2
zuh$;8_J_bu67FxJwM`7G$V3LSnuZiahHp}v;?vgN%4=uUm#ZI~Vr}YT8#%&KO{Hdx
zogn_4;8F{aYcw5*Ez9W98303(hA^byyar<|#)_k6N6KPUIEP~~T(FMyX9}gAvLE`#
zhNXU8b0oKKa!Z=r7Ze151Gt|4lv~cXEk0WXZUK|dlxc~k>v<!2frZO5LHIWmhYZ_A
zt=`G>#OmQ&CCWn`pY))lBP~%8rYb{#B;Hd2<3?O;hXq?rF!CvOcQDZ+kHU6>c!4JD
z064|o{m_2W5Hr?*R!wZ!xX<!@qMHYfBMr+%gge~<vW(&Wu%zk6_@faNuj&%WGC9&;
zDT5={%kWpwk*t+^A>`T)=v09C;!w$OTdt-u=du^!Vo)@QSVmf|Zm4;x4iXDeU`Uv#
z%qk(4tPz*0grs@o8LJ!l4_fbcdXHMGc1qGThS6KsP2u*2Yo@?AG74lJW5U4(Si8PX
z{&|R>a>{X+8fQigBB543l=DiVF;edGCf-OiR^P0|Cdqo~2b7J=Vhdqm!-!h1y}Z{N
zpR_6LJuO{5I3{@@fnz|}cs(Dolmi7ju3KYj1ighSN^*-%O7bn2@wQ<4SGs~U28>zV
z!W3KeQV;^S%mjoX^-#yx-$4CjUBYjRSO~c_$xS5|FV}g7x-0a2{j`M<)H)pLP6Y>7
z*!d&v608CH2<@A!z%($Z;PyNkLY87N>nLv)If?+m4ce)<v(Rtbp8XpRyq8I}kZm<j
zTTlSBS=IJ7sy8Uy^p^~UJPF=!@x2zl7*HH_=C*0uj>ZWm7QuLIX53OgJU)F=eq>TV
zUx-9VGpeZ1j6mIn*!DDRXrrEi9iwgOEHhKN%xnwthFwL%MMM{`!dg;0bg7Ir&}v1j
z*Y7FG%VVeG=Hyi|y44cUw;#%P)vmo@pVw;EJ<u>VzLfbzTgTitXt2(LRu?T=`5`pU
zSTvbE{lmZgxtCw!=aZ+a$A5ijeWre)`5(e_<(Hj-1c@lSr^a99U3vE_{+(FV&C55N
zH&O8`{$&duLH2OKl-X>)fihqAWrEF|@tj!^FP|^^B4&v<uU~Dx3J9<G*M5(LwYzU~
zLw~dRO%(gOFP6_v1MoMRzk!m!>Psd*I$xl7(iAlM@h`eEPmf=v56&F$Qf3yWQlQjR
zc~M6td09(ld7F=rJkAD{Pp8LU*m(J+2}ULJ6d6tj9XN{or(bz4N0AreDMC~haTH;>
zz)wWxocwcZg2@IKk1k?CnB;26yTx#l^C1QmCT^^g?;2IS+UddzrZd(nR&SgknQL#q
z`kp`@9tg18)^*`7mQi_fG@0OJ>e8k5=t=KUV>mjz)ZQCk+I;KsR^#$pSKfGC3cWXk
zql0{-WVve{R3sbG)yjgv_1Q<$JtT8FxpaFv?C)P9+p%@2S@t$@KxtC(C&zj|<$uFD
zqY|qvq>1>WxO&t|Xnl`7Q{3Bw-k{xYg|#m?Ha*zG_Q^@R4l=e-(W@)Io}aqyGhgPi
zFMu899(#VZ;-zRnPrX?LFS)rdHorrl_D6I=fiH@uGihD#xv{tx3OWM6jB|iSg-_c7
z#f=a<$&d4{;ee{g!a^P?5g@-}f}(+Mk1)yZg6=61#D!3+@mGleMW|V}TxeJ#FcCfi
zW`;YGZN|6F>i$<uB$*87bKub`GFdrTk=u^Eiyel-X~VA1VfFfsM}WlS6Ez*a_Pve!
zZcg2uLrc_{dbdQ6goTW!OZ37LdE0*RQh+TA8Mfu$_ys1}HvW$vhM315GEtjWylmSq
z9*Xa;_g|nSTEG#8=<*}h?L8bEL2!~jyK0DC>W`MxOGthfT^!6|4$po=B`Wf;s5ahC
zP9N(q^;(!Y<Z(ne&jG1m8#-DnS*jslipXVS*(je_g8&rP!I?GPJQq5&cIeJNwdVDZ
z_OQ7AfLd&~KlKA?SVWXDIQ)zw6J`+!V7y$E(HfS20=kj65=WFJUFeqz78(er=Q^zx
zL$f=^eV+4Kc&<{xaHGk7gr6m==PYt0yHpQm{%LKwvY+xHR)te#bR;ipCj+`@=DB^c
zR5+RyD;tW+u&c=p?m=0f7WDa{-m~}#?h-e25J6!|wy?)9?gV$qY}BQ(%r__NtwrK=
zy*lZ_j)lax;t#y7Rp9_3-q_}8v3fo(1Uj!r=RXg&J!_3W`e}C#gKs%4Rs<|Ll>&>N
zE*HtFAjXS0I<crS^ip_gac(*h&T}vVtCuY8dRplZe8wMf7Rsrm9n&RuYH6RyC})J3
zOZd3|zMc3v!~>BHKU9)3KaV9*ooa>2uASC$c$$F(vPLY0x)FHy8?|T3M)LCyjdjIS
z<)s)}f#N)afq#GmRxEgUiv4i+;Gnl(JQp(}#h+?!x`VET4pOZp63pz*V#9+`c2|!>
z?a__5|Cgvbd+we47&B{ve6bAo%x62?9DfwDdVY>M{-{o3zW7tjnC;EMAvueO;5>I*
zf*!wAFxS=}4yER03~Fv&wqP^TrssQ@p?0i;696->=GCt8D5$F)+T_bBk)a893t8J_
zii@)je~UF1IMFOl;A^KyrOWUrZh*$T*}lC$p&LSEJIR_{hB2l!hFef)+K1B|uKl4Y
zp^bQi{R#Xi)S`rl5#9=d|Jv1KoOQOr358VBngbhnoYBG~9!FVw79@tz(9S-An6n-~
zy88kpRQ=9?^m#>6OL_`8767W32Z{W$$&(w?qHIv5IkTo1UF$XE_|4kNpTGn0e60iS
z{MPe!@+Tob8mCd@{!?debCB(d%&3jqhj9ij71nTV8WQNXRs)i*<6PVIa7_hpMiw%+
zk<!__0)K{wGyX8f;}{-!@TB9*r}m`d7n}bwEdO7y+m7&{<EFe43q1Lex38v08Q*PQ
zM{Tj*OA)(MGsg;VH{Xt4C#MQu;%hlo_yyZO0@rz{P=Y8g>adZQX)pM2dVIOYvX5Jh
zCT&>u8$bW7mVM!oXAJw4)-z+i8ao)>;6h1^`5IGx8waH8)#zxrjiAYIVNKJJ$v1;-
zacYS&p;fGd|FsILdB$+&62WHy!;Wxc*}9SX<NsOlWnaKLAcBy}FMT&vDwejo=i9^1
zif^I(&~{dIV|V^Al>kJ0e;<*wm@cyeT{21Kn}&Cvg+Vf?hopZUI7zy41*bU6!)|q3
z!4uZil}A0*atnfO4qJY7pu9KmxTr)1;J{OmOPn<BHRFO59#80e6hPvRfzwSCvR2`|
zT8N4|uG6x~2&?E~0?1~iAOeM^qXj^q77F=RMC-5VqLBnCkaP(-c1y_Tny;B9MCyb~
z$miu1rMbRJ7Jy%9UI-QdDf0CA3Ksy1@(>yyPLHpvDF3zekd{P&B!3r$o-LUFc_jI}
zXna-G1qn84q@nz?g*wjCE@Pi-g!z^AAl#GD?(>B^YvTW;Vz?(Qh<hTvfV8I!;O8@`
zpUF1z9A>Tw?Wwm3i!)_78W2H?+Z&8vi|Zp8$AlJUSc#~k5PGo-&Gh8Fk{@o$Ue2?6
zj_Q=DI6CTzjTu0K>c5^}mHW7qiDX5Md~0j`lQZ~UB&t*DhM^_hp=@iEtA4#mMcls<
zh`42lxX(i<apJ@hj&J@E`=(aFea@74dVE0nmVn~3>G69i-oBWMw-d;=zyDLwQcupQ
zp&DVP2b_@`g%fb|1zLDKD%H3iy5V&iE8L37<@4k5Khsw2G~(X{<d&ngt@kNO71hV;
zj$n+|W_H1|<&Q-){2SMY`Fb{ptkb69(01rWn!^fA7=`;tI~lB^G23S5@B+x9Z$Q%6
z0Hn(btHHn<aIFp4neVJH&9E_5fvT?^ntv^XEh{F;Gl876Xf9?uG~1L6i==jq?a<cF
zsyj;z1|^W}v{Ge|5gB#4@h!SHC$Dlx;DxNI3okhiNykOTUpJs;y6XHl!K%~1+Cq;@
ztb5=E<3IZOCLdEnK7ma}jRoeFSE&DCzJ7ZA4i}pc0v}9|e@|DMFJ&uD`w7;X<qJOz
zJsPI*bFMa9^Rh|J(x$Q34!pWP%h>d@pRvh`op|Iz)i^%;t@n2f>uCH(z0a4|JZg+<
zC3WR|n=x?yd32f34jXwx`R;oqDFp4!x7G(rr0Qq-k>xD@T|hsQb7X7-&Yz2cDHkx>
z^PPKwv_}>>Q3~d7?LHNEnFURaPAdYMY7vo4{j<~Iuz!IL)f*f<kD`X^^RzT+A=4*5
z1r<&B<R`oTcQM*jbQm)!q1q-syE{f!5bLaO*@}OkLzi?7GC&z{Y)&N{QmT}d2m_oU
z_9kaoa7TIUCLgvCcjb77Ih%L0@v(Sf^dhY_y6e4Ii!-``NQBije=&RuH^XOzVHPMn
zRYs<EVlhRr_3B}_d&;{YOm0aAN3W8Oz)zJuB8t4tXcY+<4{?QDpz~ScC+a9jV0G`{
z@ZGAwwvzq3k5k}S_s#U%d~1<eMF1sGPE8N0*Q`#By*H!B9IaAq^>ZKDlf=(8Z$K^m
zC6#cYzWR#1_>y@Mu3koK{YvxKQS(<^fkdVKHDe?Gb6Vi(agR!E8J@g~ap0rraYr@Y
zU(Gb$Qz*PIefe1xUdDMoM{JKz1AB5OQ-1RluqSuIupSwc+K`@mIC5sN{f*({fjA!h
zP?Pt)hv`<`8tjiGolkEdP<AAN!}Mf>2e=)dA$I(sytYv}abVBZ<;!n2E??Q&{I;^k
zxg&evsQjqi?p@-z-jRlWDuF#<aUXbth@kTX=g1j_uN_4MJw~z$<{hhfjXSsz=P-Qq
zSzM0_v`k$Wj<CXxfB#Iy^+bD=pK^c;6d7Uk1p;l{7-@w(phli2u{>s?oCmy}T&3-s
z0T(W~>E12kt@DKFKv+C;sEu@eTmJPyHe_tunQi+eer+Yiw*BH?dnLwi`R**bnfxx*
z?%qDqH6zqhD8j9vB)T4S`>jVkXfIC!d%aovpEUGMg9*c-N3B;IaO{%$H4x>hfqC$7
z1T+uM{t$T_x=2)V((5BB(*)O~tyf69g+YO%D$s2mOb7b`#>`*go}({r*w3ODXMuwA
zq(wCkub(vdWtBRfHLaqZ0E8`gHm6_*9)F1WcmmVpnO;7wFf*kJ1yxUlq0*pnsXCKp
zw5$wIaLYN*KX4dm&9v|;n@mT0G%><g12zaV5rjYiRUY#&&6_@<^C6d$P5-isdyk?1
zF~Sk1R7qd0m$@L;G<4u&^o=7eBY1Y@%rW8~WyA(<8#{wbA4?~}IetKak*S}JG4*$Q
z<e81NCn6^j1eFKbpw7Mqp(5k}O?^@hhq!U*#;}l=H`f9x5gtFMEnak|dNY9JRYNx6
z+p?ErKxM~;+5nvNc@WwXSCQ<U8PsY6H~^E2Ap-K-6bLy>CujzW&CNt3t0A0X*(fG8
zff<3|`0jDXm#KSKwl;ynpGC%<Clc|~3x3O(U;&F*(Rkv5HOMOA$1Y<SbUnFQvQKIw
zJFIw6Ls7|h?o+XQ*?5xKXOuQX20qUPyH`d}O9~Y!FFMXkDP&QSsGXlMR#}IanHxSg
z{Lt2bm=0%s0InS7h2oI1HcT}hlVKR1BY9$7utH+0sX4B+wKM!oXT9ml>J+-VX&xPC
z{P9|bWfoJv_RLnn=)dz57ur)!u~=aY2=*NL^s?R&OEZ<BBNc)*@(#J9QH;5%Dna3y
z4Qtb{DV7*m=B>o=XCC$?|Jp?pYL|#6ESrXaP;FeotxcxeBFT6%EF{S~Rv&ibo3M~A
zj10(fb!Rj@)|8#uOS`&)3|=L(eL+Vf4eae|b7qIV&c~TVA26iMwoRAy5+Rr0P}r93
zU*V_>Y5Uf${Ivzl$chGrn8JDGEHM@+DtuN37-;hr(dEjRzsg}VptHd=+;&5d)TYET
zc2sest?d=(bI98+04HhXqEpq2po;yu2T7Ne_SRh_z_q)9G)GqeF16;~a@O2mZ2mj2
z$zBlCTxiVg%5!@3p4<tU%b0TSHuq8OtLC-W#m#=rl=^=24oX=des^GA-;g9)V&Fv^
z`$q_xGTTv8;JeLhsLxDJL?U*9__gM?u^0cmynlN9iZd5y(BxGK4|y@88&O3+#Y|%n
z*;sUXJXZU2mD!&kyYR5q0d5n3HG1?4tkKo=;pv%PF}u3I#&k^W>Q28sVKg{4yD;E3
zB#Q$=$$I@B4s~#$0M=P#ew9jN=;nHD7a!xl`;0If52qufLByqI$b~qr`n~<`fc9QQ
z_u%-SCphL2ax35&cXfD(+P80Azxm$1n>bFwl?wW_8V#|p?ev(L3}Na_N-I%pri}^u
zf~KG8YBf1=iLB}k@vOFw#ONNF&VP8%1lYZUMrXQW)1)qw#@5(i6;V%qD9^9y1r11d
zg`APo!$a6;LG8h?@-ciM+@Ra-fNYPEq)Iz2=<#B=hPq1MAL42?6ljl5rr4gt416R6
zM()WeFQ!1CUlcd!z*lF(DWe3d<LOxQ;o{I<Mn++^xNMGnP9wcVLQt0xc8Fm7Kbm52
z*dN@))tJD%I>2t2S(_WYvjX4uzzVqI0zKL5g0H%4M?md$4||%*nWM^@f-|a`+b(#y
z4q%^hO?%YspEbZ=?ye2=bbK_N_SK}+?vICT>9pV8M@wKH!a#~n@ZjSkBqAHP<?<Ct
zoqjq*s#bF~h8-#~dYCz`?hb;ZmF)7@21(^_VB4cp>K5Xq@|Wtm5&Nm;r9g0AO&1&s
zxVA$%j+^Q7y5Xo53kZ6pih8LXqDM~|jk%2!YK2l9wf6qeHnO_mR@ssA{(7~<^bmZ}
z;`+3dD_<4t91d9!xO5+vd+&N{z9`k?-x5(xDN+sp7mzjuM%$zMmk*1?GBqQjhCk}z
z2@|)-bal7?S=dVV@$`gzQx6aW1M|}M24nJSDP`Eh4Px?pZbea6X5?-%ILjZ)3g)jL
zb2SXlbdx3k=>+GR2nXtpfh`O`RgvNR6}>V{(-W(tp43jWU-8e9KmgIs*Q=H<Dtwic
zf_mm5#H|r^Vkg&RsjfBSMk<z;ouXBbb=z3Dm5L?N?8=YiRy`?}H_Sr(!1e$sZA^rd
zVfCbzBaDKESa1XdcL$wqMF7pf`9RrYdvjy+lF-GEp*CapSP?nV6TJz?4WN)pB<0W4
z2-4f~Pt+)$J?AK%YYgSWAvc6v<&A5f$>x;?*S_GqVL#d1t{h=sSctHlAP(|*Bo;L)
zYlJW{q!u3k8X3huwE7q!!=G@CaSgJh$ZKCD)DDu84}GbJ$*aJr2a!q<{BW)F0K%tf
zg;<LS>clUSYEh$*f)8<s61I{`LQ#$ZeD&dLu&G>%%0{qC<`i2QD``|V`Z3S?CM`!b
zQs=0P(#}PH29$z~rELPs<ia;&uZzF9De{bo9#qMj8Nie)NG(D7;>D3<eOH(lFD5!d
z49L8Mc9IPxo(Vz*5nqxKBxY&xIv=@pGCE5K61U$yJmsx2TK$rT649gkWWRelss19|
zS8nw5l6bjmtQOqmxMi>uR|A8pUldB0NUB@NO=%I^j@2c&8Oauu+Y;GEd;<z@gdQ@p
zU{}&QML%{=YqbL|?>;La@!<(nJmDg+PR+3J^X?jrfxxVyWHpSuFvmc2DO+ThY5<T{
zk3$Il0j?IVxkR#1*_((1C<O=Hs8hh$3<J8yEd%Ouk4Uql?Wf#aoa2v!ceOYBhS#ef
z0e|LlytwOfWku}~?|B~`l^J`BIsgeZJzmtV&W%O_4P&!*ejpjsgwB+<A1^vTm@-hb
z2@EX#=JLE)vJjP|nhVbLbhINLoil96%9w+*hYSf=IcM-7^Lj~nIT{yO>c<ezb>h8@
znX9)8b%i4__cp}bAah^SC{{L`+J3IqB@IgY9_^ttn@AUJ@rSa2niU^bl>I4;bFgo~
z){-}XNv5hnec*S3gLGPPz%)|}u7D^a1LF(sTN=XREBBm^&CNO0nWIF$)m#?y72K9#
z4m>%X;+oFG>IqLtU;^w-VA9FvOM;^AVSwF_+|GrI$718@WQTUGKf0&~H5XSmHv|e?
zfr?%I5%zBd!*_z)c)HhG1zt))+eVC9Nf!zBl2-WTa*qr^{~@#ABlppd3MMgZ{h(j$
z(<s=5O*Eq*YZ4`?%r7F`bB1O)Srkw(l0xO8l(~K%S!~tMH(%C2gZek8mOr-)14Wg*
z`mZ^3tP=+8k`)NP;V7IRYItGDS&^4QO~K(pJH4W1!KuyKtR+rVynhT+lqx()@3VB3
zdmo<}(NsEflC!A=CVBxQ_5PVV&smd({PWLstx0qBdFDH7oi)6vV%|L0@=kIEB)1Z5
z(P^&u)=YK2{82+@crvoSnS_swhJ&3J1|NGA$L#6(*MIBIzrN7?TR39k4I?Mf@j`eO
z_7|kgujge%qcvw$Z#TbR21GlXDYMo54$AyS4k$U>`fBsCIopyms|c~D$45NpDqyS)
z$=iBd>nc;)D(LC)U#-bcF~FhK^;e%OghK9io1C3;73!VH6~woDH~}%=MZ5=@)@+&k
zIKLW>f){#Fr2q0f#ujurc`o>Kqd^vN0i(*T6Qs_TTfuQgb9yEX!^apmfZ~EnoQ7bG
z1`U-L{EDqEPGkI4D#BUv0-4zDr7G_kPyfn>S>T-Ra0vg^qZ1Kh@8ITZplWa=32q>`
z2&HjsZVp3gA~3ctUuj%^^UZIY2n-&X2nOdkGjVf)^GZI<G{?y5{EgT=ws~HH$ZGtH
z1OfQ>t+hf?n*SF%G%4E8v8zT3cq6uJ4Do{@eH)&b0IVEYv!8L{6|}Gg22qG=Zqs1o
z=Vd8TgYlBD8_*AM%e%Qyd*#*uQ@7W_WyMc)NumcRBZyZTXcQWN+*RrDkDc9vM&|~d
zyOyHE!l<SGs1`hiV9}nSXnaeH>j@WSuQ$a+6L!Frh_DX9c(Z6*tmO|dsWA6&{QLx0
zc!U(p#?v-72b%>iW^t6Q!$$65aa5?mX7j-uY;cR*3!%-{BDC2Gu}Kg0q8|LD0v@7S
zC3nZVNfC<-JD1Jh_@$Z1+vcz5#syBcq}wVD!{~SMn-`TikO(e<V&DTP!b)^Z^o;`>
zOnyZ7I!jOek(iVVuMJr-9bJiU4f(EaWAKu6GWi0DJ<i%E?a`!#<NfD3*fDw#*K|>Z
zVuLGtwAe8k#K398Iv5D9@#MEFNqG-r--dr{ymuq6wdLpCWI)<6ZMudAjf*hMcFF8S
zYV2ixux-=)JFj`o|Jae0;e2So3-g6y;t0{{A>lM`G~dtfFlUcqVzgMg77w9jEuWIJ
zQU+b{q>8!=k>LmYCEH#r#CBIANdn|Dzr0TO5xkkWxuw@*I2Ex#?Ybjj{Ag%V*BBc-
z><{;Fxto(tfTL4U%%`geP0~sT-eNP!_98;gRh@`#nmb+2?pf9n(X*d<GAHU2D=j{y
zSz4wwX+Y;N>M8Nfax4;lq@>M&m^E!nrli%)F6%bn*|G|A0!}Uv9Ut$((^`(IVm(aX
zt|js6F|7T5@3fUx0l?qSaMm%Gl-;okhcm4d0Q~mh{{g+nONrc~RM#}mx?Jn?M%7Z1
zf7Q{c#jU(>E4r~8A6byHa?@7)*hsqiHxNEEUQ#9$UgJ6%LBqUqT_UQ~tT1u-2m1v6
zc^%lAC5S)M(^K5=Fa|Hc@lL$&AWD!IcZ6G7%{q%K?P*~{^dD@L2fR%KTcudMFVpio
zP%bHt#GMWfXoz?3Au9z|d2TNwRF|v0GYscL*?q!GL1r#4iXbR7-^WfAt329%*d4(Y
zHL8nAqX`<Q2T1SowJ=_Hv*3iBZ%Dtm61~Htc9Y{v9@>Gc;BYK-zj!gKKC7lgGPs&s
z=cvi?^liSL){GimToOf6s9pl?4+pSWPxBQmogxj~_6<*))l)xFNnU<ruDJIN{!HW$
zOD(3w0PS<!Keh5Knn$uupHo{%YR_Ng*?0RPm)ps{B5MQBH>nYGmPEW(ZG$WFIY3C5
z;k@-?gy|$RlLW_&7jP9FtW3bTxVo-~!BWRj+qC^Flk~%)MEx*gHD5s-lChQz2blPw
z=%5G@E05Uii?t28T&HO|)Pu=*9<ntnTrDfbq%PgQZE8^HmMe)am~*VHC-S(arjBf0
z*LEds(LRszys%pj;q!4w(cbCpPjW>@C)##8-{Q?1;m9=GmVfh?mUi3rOKlMn%a-Ez
z`hr3oS5$uGg>ART+wNOlNQ9|e)frs%%RceLms|33t6cMSd0DWerNVnVeNcdhgifvf
z;Ye-*9}duN@?dV2*Fm(=1l#b*k)FFG5MD3X@n9)1lUvdzXEVh(h^V8gDH%S)!FyJR
zra@B_UrQydIlz>%o8<zmqBz(tK8+EHrw7EFX$$L=dO_q8Y#$LBPTQep^`zHQRiV{}
zDR}kbMV&A)=<?zo!(vE4zgq(=qMt0I>o~GCs8H#kyG1q$eUWis;!wWL+)nHSkX@gH
zec{3dv78jJ6P&cos}&9PyK0FyBP3J3^uo^Z8Sgl=XT6r$Tq<BYQLKGYM_;D1$;acL
zz`7<Eg8L`0$4}y!2Gd|zJbszLieKssjH<i)#I^Bo?nLD<Q=;EyPN;QT>lV5GYNaEy
ze!;MT8(yQclO@PiJ)~G@kJ=~XsO|u2ID84<e#A#12g#DHC#d7yp5J4bk;P(R$GGqo
zW9tiWpi5;qz(FY)@35R5XOyvD1&*610g8HTT_lX3rJf>oK_vv-+(iGYaARsP6Zjjh
zao6tj6t@7-1R(`p)8zpXrmW_5216cypgubgm`c-0%WqdqburR-7<wp=OQ+D3Jq5MF
z#9M2G!8ZjkVG8Kig9}fzFTARXeM)Zud|f~e0MB*ap6v&IMn0lAU1($>DkZ&eB(&^Q
zhz7PYOv|b{K~`cP<SW#{o;_(7R4N|&L;Bn4<od%_g!z@oBqlsc5vZw-v&I+4;^Iro
z@wcU#I-YFcjK;yumji8OUOnG}bYT+~j~aZ;nYQO@E64K{OR&-@UXrNf<DxXRQ@m8k
zGU#Vwe1bcZaO-CzwqKYS2hSG8T6*53Cau^NP{tKXM9D7J=uNs}jw8c)9)eWcJ+vZG
z)(f1|)7}9Nu@|+g+)f&Vp9by)L@6vU)>x#o0e(~O-@kbk)|EjQvJ&YE5f8}t(*AI$
zmKewsq^Cl#;33?3$0EXL#h&mu1lpIbdJt_>L@o|>O=ctey)m7!x}(u>WRB05P1q%-
zz-yRkC^hU%_W`r#+EADXm!yS>>ez-c1IZ+@08sfgnbnfQ7rR#b*2?~*`zUX_pS%Go
zMqG>VG6#uH04AjaIH8Q!sw#z(8i0=oBTcaZ;)~{^gii_rbXd>~mWnRBhQO=etV7b@
z1VY%9h+fm)*4*}N>99|>&1l`4+bs8k(#hBB8r6$t1Q2;UkW+q{vx3_=vl(kzvfSCF
zE}=r=*J?6_C58O@18hsmVIB77LdnntwvJcOMmX?sE5<+5PChZU_F#7H!uVRX_lA!U
zwN3KC+QQwfQC|vcDK9bNu%U#wZf+IIsG2bRsBJ)M?DK)TyFBWSWS+wGYR-8Y2t<c7
zD!p$+#2~1BA`_BUSOX%Vr!cP_wd5V?kUZj~vjb1$9iwKWv9%T>p8%3}9WT`&P}f?C
zQpIr>Zn40H?TmS{q_HFz%B0k_eO<B&*Qg7zLXrTHZdxD~k%Cr}R1>^ywR5Q|PkB4(
z*2v}*zR>(P;26jID+ZzCT~i+O(xm~8p<!*)qjBR+_dDWoM-RC{>CVz`HLn0Bug~D?
zDA(|pny>!>E@)!&umADOFE#(I=6{FZf6e{=dplo!**M_wevQG?<FC-?uGEygC}1eh
z!w$)VFCqJ0Y6398e>Jx>=$q;Bzgo&^+?k%7oWUXQmuFn!fdmj}ZhJa8*tl{52d_6+
z$NHZki2<Kk*E^bv?047$nF(NQ9PT1_z3a#%(BA7~uW$>_Z2Il~6lT^7wze=y;HzhS
z^9F<9p}OwbSDD+MU|RCqleu_p0A+w(fu96&MAH9HvLAGYSn6@0$pa(fymyBi+b%E-
ze590PtJSN+z2mDN+HUL)26~>%(hXPd!H0=m#g?a+pg<Dn)F&)9G?mko3he|Ah16)e
zOpFeHKUZGOP)2K^9LB6Q_=wHn!}aPzf&+v;zS{$Gwh~?y$eM5zPLfQZ5WJ{WzH@|y
zW@6&8=@pzr;~}-`R|KVmx#~u?jwHT@GiyL`yxj25xu{@EA&jxBRuom-XkL#JF*sDh
z9hcFi1*nO;4@Ln}5#&Oy+_}zZ2QE!*>RHzxeS%eA^H*RAtllR6mxB}vB6^+Qp!8=^
z7dYQ4tNB5PbC-n^Hx3PZmstpS%C<HKhXozq=HjsVU@;sV>upyb^Fnbb@(6=8Isagm
z;y~kfIC>}({`|xZ@gj~~Quw>N&|tStH8pUMa*ik5)(pqfQTJ+PB~l@D+O>SAJA(1+
zgg39N1h$h^13C!S2n^1HYokLL#*m>THV&H>?L@WhMssxy{=gPB6-r1vLp2lT70o%r
z<P?30z0)1MK4pZo55)<cWL{Y8v?B)1fMN#c)-Lm-iTf;uXx#$xhKnT?5lcyaR}u{q
z#)nxx!+9=88WY}}y%A#g0dy94hPznIs%O^*>ljdt(G?SGLieq4b=IAv5vIgIPe11d
z=+t<&g?-?6$}WUl3eYtkp(g&cw!9uKzKFb(h4XMWVj3`7@)mNSMZQqxE|<bF`EvEB
zy6Vioj;So-Fk|v{XE=qD&@$k&Qjg$Pg9c+HkPR+*0<^NKlsMMk1R(T0r#1&EXSPjw
zY>G1dpf_k*ykmlw@8UaED<p2DG`cZ_wkq@%3jYeL%47+ppeC71S1eo4Kt>KJg}iER
zF{Q+1f@Xty3@~s?4gy>*h;IE4oNoOENy2tPG^$X!z9`RsS)V5g@>iR$+T?2T{^{`p
zD!>sCKA9f>Q7KlW-#*!cYWnxS_*}6fH8CLzhko1!7MTi}T(-e7i{G_z8|UQR>U$FU
zXH-^QKPV+7mg9C;Rc95V_iak8@7=t6pF;H8Z&oL8X2DYM|FidQv2kWenpoA^bLpC%
znZ9?AduAs)=^Ca~WKvXJx>_@=X||-Q)>4!>lDc|&*xFQ*nIw}UGqdtSiLH_6#qcid
z#;hMS59_@E2KLFY-wYTq2Jqv;`>=j7Yy*C>fi?WHupb974Ey^c;@tlK{K=%Gs_v?(
z>}oOd|IdjNCr+F=apFY8QCEh1xX0@EF1>|ZR^NN)va;!|!K53c@|~{vc*1Pg!Sd%|
z?88@Vy3@qQdS&-?2T1`(P#$|L7TdDSF_F3S92#NJ>tcN`4IU9i0P``g+y#3milE{<
z0c6b6k=_>yh@W0dOf}*CYLt$tjo>XG{-O8uc&X9p()^Npi#+G=6+9m)oS|5XDi{dP
zIa&%1sZ=bV)?IK&F=PWYm`iE(piv=>*6^V5CYt;}JBl`ODrV6_ob86ElM_U~ae@h|
zYax8YBjOrMrC|7#Lv2nC_WY<xEJpz~gl;5&!Y`l}vmS@$j4@9Dkzk%!2_WOtCxH{<
z*8u@S2)}?%#OUW^6sefGAtM5qoNwk6!O2*PCkqsMaM=8}&C?Y-Y70|5`sJx6nkMHe
z-t$lTLMJL->6frT>!yK%I)rPeHOEe^!}i#;6>fFQ8wG%-RvPUixfZ<4jlt)tKOM?n
zRQ1<bc}|Y_gXuAk{(n5)$2gaL*c_<3Ecj>c1+by@-07;izjfHUaC;V3-~JfKN~79c
z*+{@nNS_MuUAvR79Ok+M#lO0;+?nDC?AQVi(TR-SEea|+C(}r^)nA(Y_-Dn+aOmcX
zk#Py_^ZSv?L4o3|`9F993RC8q8W3<wDA{rCX|{TPSkCIv`J;ba&)^jCNp`N(XWNoJ
zS33TfarvD~M)h8-)bu=ite2f5Cv~fF9WWvU&zjlZIi;Rp6C8Qj6rf^+1K>0AXxPys
zj!-jgY_%EYaPnrjN0!+!bb%BTw`<t_g_IQWU|MepB+C_!BOy_Or>H-Zt>G&5WVG_L
zqzG#T%Dwd{y4ccPmf)~-f72r=_mqE&Z~-<)pY`Do4&N+H!#GrT{M$YmAY^-IG-!8r
zaR3=?Y7TydMF5-ZLaXH#O^;(3ILPK~l(;;~vUjNx)PNyZMj=vyGQ&$1NYXq5l(8cY
z4DB0PNDi9CdkU@vVo!ilRxjwY?3JvIIfKJ@D$Ps2akXeD{?7Q2H#U^XAwbH+@UY7~
z156Vd0zV#>M1U$Wo{W$wTQ6D!!9r>eH>}mnl-7*fUkNRjriyWYR9vKX^t-Bx?~PRO
zzp9Ey=;^u;I$EmY<~M<+n-xcf2%3<o1VV>ehh1>vd~hSb945y_<PG!`kJvY|G>M_<
zcgOA!_t!W6LmuK%!;0-p@5<xfkjE;Qo*n)wPhH>Ic*~b3CieJE9?t&!OHw|XLH73Y
z#$}%B%G-j`>Z=>?xI<gMes;J?n^gvoJXbumU3rjV7?mMS6xj6eU**kRqcO}~?|$`p
z&0WP4)A>zY>v!&EO<dO>!>bALfp<T+M!VD2>Ak@S`Klz7{Y_LtB6_5@z=@<LWWf>Q
zkZX-_%L6rY?P!K8XTrj1=hZ>~WN?HOVmy_s-Re#r4Mq>g4_Yl1vdt#V5Zx<*F?lqJ
zs4Qd}Q!^Q0=oQt-YYD6-cufSMKO^pL%p{M!y&jXR?IHkzN7s<Co`G4~^#hZ$-)WCJ
z<7O@GO%HZ_iQr>tWX9KawJi^6Z*XLn=E6ng0KJKhe$WXnhfyauvwkX>@}GcY`~7Yw
z9B8>&x8Hg)d1Liy?GG^Lemdzso;<<5h2s_v8R0gA+OVZu#Jz4$CQli{s-<zE;!n%5
zsV6b0pB_(9kB_2y*=mICFwl0E9VUW=feltQ>DEudB%xQHK*-mRTLP{KAs!5IzGYwc
z!`GAW^{2q{fgEUAG3jrJ)uKZS{G~Y}a(X1^<(jjNdjdEe!zmEC=#Uu8aR=lPa#O0T
zg}wvtN{<^f`sg|WIsLX~jpPr=N-zPy-4J@fY3=$)&DuMT-Gs}7PH+!6`K&$8Q3=)C
zVDQ2IU^<3+;keS61Po&C5fP7Uc`pN<Vln2>1rTbAb9=nPwH@v0WB|g_-;uftFH_$=
zn&O%X-$R*_a#_mqqaO6Vn|Iu7<@F1|4*vljf>&|;soPM9sQxZ)VuuM>CvpgvJ?cHg
zEbAR#1sD?b2S2%e6J3$kS}kuq4p%-8J9{gu-(|v~H{X5t-6no&;d5J|ZL;3%?H^3&
z{jh@Bqu-Wy21k?eis0}v&CyYKmgA&gG3Eigr!kfr1d9Txiubn6$Bw<2+6f4MGgWP8
zFgU{7g5_D4&)tDbNI6NGio(pG)rD6fmr`V?!m|G<wM)BUIG;V!*D9M*-K*66?T7Px
z?SyLOhWxv<rI<11YRt2oTPt3&(YVBBFNjqcUYRzfGO4B?X^WBfNIAGPlb*;8@0$A(
zrYz~_2W8EFOAExqGvh)HH+M5PEwh<Y|JF)R(q^5t+z)mj2<EP9P2_LG^-Rv0#8u46
zBD&fZy}L>WPreBv%1_-dU>>qd*#mDE|BSMqSxy2CQ!8aOX!E$~uw~lWTdg>CQj~A0
ze7Y9E##J=ilJC`dd{bnf8mC(d_lwS%Z;7gXdjA6C_}`XO0?`)Kh|%rSk<*;5g8t4y
z=Eq!+%>N1QSbn$`ociKa6ir1V(S)jwK<$ZADEo@%8RvLXV|=pa2@ja~dZzb<-IIbp
ziI*Y)tbbG}>L;;>>7Eq5N!Yb0zt&51nMyjzj&>yZeTB;uu92=XRFwejXUoco3jzb%
z(TKBIeimxfxBalYs|QqDmn~hXkzoB1y-LJgnb`;&B@KIuM3I|@A%ExVu%2xJFq&3<
zWyV<$D;U%_T=_Kc&!-!zi{-7<Awu9MNPGPq{&fgl^juR~Ns8<fD!cK;DAQik@MfBL
zLR7a+=0LlKLYT9z0T9_uiy1>9TQ=A!HRLrtd~%TDzkphIU|X8Qxi0Pr*n~&0s>cv@
zIDQl%+U#}-Uhq}Z;N8lBAz~my<$DB*Vr#y8U>>G%Klr`crpUMRAX?cnD(e}6qm_lp
z0$XQ{qb1!-qx>*~*iH&n6O?>_V|8O9H0&DiO=q#3-FXEHU}e5CU$&?xu`^pFN{Xh$
zkGv`I*Ear<m<C~~icN`sCXc^NlTR+<<L<_f#Wtv~lZcPSjX&|WLA-u;_<Iu2k%1&H
zz928u@W%(!!~ZQ8gV7iBVdGC1e&fX~f9B~%dyC6kd}|#;Kl6@D)xOLVg7eY>`{BF+
z#riTc`>J=gB=Xe^zDVIKo*lBwbf$O_#<+*iO9PE(U3&{4(oJaOWprjvTffaIdfo@O
zn4*JJ1fF~1geyyWu!oPLAGep3Z%Yh$BARH#I@2jCOj=IVkcC)0ZMNt1gF3VE3b&`K
z#^c3fUK+=BGJZz!>vX#ub=pZAi<X{qbkkayw$lPp>&307?lx^o$Oz2i<y4v#h4y?m
zbax}u_B?({mHVeKG+D?MTPO;J@}{O38SzUMQs8rTh8}Zu<p;Co<%JZykY7>8OJ`BW
zIvmYom?WQlF|J@SOhoO~FE*q=lNeHiO0xndxFHcMlCF18D$>IWDa+j=2t77kJfo(c
zHLg`{4o3mN%Ro2`TucCAmg;dHmmE+QWp?K-J2B9lA?R8uK_js#H~~n$#j;C-QL&6q
z8ig%67oCeo@TqXMOVz>fy1QMBMJiCO-u7^h2BOJUjhlpQeW(UDW~C<S$skn3L6edk
zyPk&k;^3Wigk7q4l6P@&6_E;*)%4#;=L-YQ>#E^<QVx<XpY5NtwD3u}u(`-Pu>9-d
z_N7+jMbUD!fyPkEfyaueXcwZ=t70*Ip5a2JTOf#YCUV38hmsiO8KV`gcSikEI-88^
z3mG_#YFT97HVz?l$ZXa(q*<5;a&QmWmFVu^5rCCY5PhRaC&xE@OJCgn;5ZuaqLTmL
zMN0k+sOztZ7>%|3Z*TDVSM+(jK3d%Ow>EwU?|;j^PxSm>-}r_LT#@(B4sTM?&jOP-
zN%MR%J^a5_)&Kg@F#=Z~*;#yla5CT+GOXo`FGbr|%bUauGbe+5@!CIH`v9#q<Z07l
zO}Zmf$7ielIHfX?IG$VeXV+S7oPf4*BmBPIwNXP#Fuu^ApKk?_XQpwcHZIp&6oL>g
z#Wyzmt=tN{EM>gFr|D1FHH=~HZxFbqL)*0acog&!@Uo81RfeDQSieiB9E;Yw!le?Q
ztNuz`{g<496<oIuFJJYyv3@S?--a*f7j8VOqEb;ZwY!#-v5zJAc+Q=Id8|J7Rv~!2
z7C|(&?_X$61{y5A9a3rA%?@F9#{&=XYV<7cCtW$W!@%uRFVK2v6bx3@vp!(&_cL+R
zC0J*5gLhE`0OSx_==B1P@I0fdzW9!*gllrl#9JOzwc)i7oQ6v^Bv~q(iP6J==!{7C
zPzbTDNQc{CbLFwGHdjXgBv}$E;1X#%4ISGR1zYOeSp%_v>%ri0i2uVY3+`kXnBFn`
z8{z8+53kx~9J}lOp*EWKB?(OXWWQOP&=(9h9hBN5k1G8DNp7bI_{W@uYYV1K@*J~4
znJR>~Q2P<Q$PBlptUc^2oEd*c7s~`KV9g<;UtSh7x3@~$ow3Ox0`aGMm>lTP-5Fp@
z$Pw!#>u203zhfM;TjMjf*6JvotvO>;9U3EEL9xzVd`{#9dtpw;kjURw>Z4Jnyw7Cy
zO!=^vYq@*^vQ6}C6xAti(|%>*+`4~l;~q~{ZrPHT?$r3|>;6?{v-S<ytbI8uQrNG3
z&Gu`*?utA++~=093aY#;V#9~i!~ZXLMg!8owY>4CU;5_r?%L#7#*S!wW~>)YPH;!g
zcZwUkLUsZvlxC`FyBj6V9rad{8{k@(c`>mNK<HYSA|r;JXDiZSo*-mV++xi;6TaKR
zXE!Fb1MEU~x?LR8hI=k+4{ocU_J?vcCT?DWgrFz^`bbZ80`FAIcpy;j@RV4*OuMIp
ze&>OD+PVw{!N;_{GhRWK1NxX=e&=n)AnnN<Sq^UYPN(hu%_+isZuX`tLM*}JWxA9`
zN&UsDh?RD}3<vK953$&m7d$;(UaPXBlw-0czofCO^@JsG1SP)CVbGD>2xHU}cY1sk
z$KpaPYn=!2cHwCsj>z!7K5f<mwv<PXkKh~sV@+t0zK620`|zlXs%w$YZCHLF|30M-
ztI>g#jQ6}bze>XTc?|^#poF3&)aE=uOdxV}&qS@_L@@!BpcZpD{ZCWXyL5J&07@`Q
z0F@n2<JSQJLWq_!Fkd=pkSDkgj>XLI83P);>uISrVdBE*%#e^MzL=s(j(Nuycm-X8
zFW;9Sf=yuhM9#l}Rb($CSn&#@Qwu{86n07_76R&)$;g*dr_lrhmgGmEY`fG)<Ww9J
zIIhjspl2F!2$jG?zS3&>e5KF{Z8x+$<$Z&79wu#X(`|A|oG10)`_knWJ2Gnb_q&bN
zOBbRYZqXu?;G6LbmVQ}wili*8`I4z3c^EUw3_oD_ub=~Kb0QyPU>IdJEQwV9Sgc9~
z$Yqm~bUSbR!LWQ<d|CbA{bG-%v1oZmJxLSh7@mJ)o(jd{rv;Z$vTW&a(oI+2V}b?`
z2Z;Dl^a=d<TuRw`{NYvts6=+Nrljx8c*w~n=j@bxXQt#(mF82KlVjMTcWi5K_$E(!
zZNKH{m5eZHEG8P1JHwbq<>`}65m>_1aYWQV9RGEX2cy%Mm0&~dRWFbfBar@OqE>C6
zND-DrulT4rIUROe?QtWbTUaU_Q9Ubc7ki(=lSCmN<UC`PEfZ-*4g)hJcsl@EB3p_D
z(cx51#whqyS|ymVpkQ9pO#GB6&KW>7x;1CxuzLd>R{T5D#Nqr_UHzPm#5%zPatu-4
z!$TH*?1H_hS2du1e;bB8vo^cTo3GbIZHaN@cg)eWC1mncDBTVev9e~zO1E0$<2KE|
z;qczqS%v=Se_SprV%2=<YeE0>gQ)$_@&i|;%Psm|$hY8?QkAB|quzqm57l@BfOimq
zTfuTB8-uTd^3rXoyNtSpydwDR5Srzy?}i+RXFHXjWu7$9at)6%{}NRsiNSqv5=+}1
zoyew`d4gpty~uE2hOv(Zu<X*dqF#tuJir&yUFMU1JI@7dFJ^&sufn(S=SXAn<45uL
z*x@6KP+T|+S}WS8#3kbI2wc~$EZ5%3lQ%An!HdUvjv5jCj39_%^ZSa@^~^qA4D5Dh
z*#gdIU@@Bp(Ud4>zNxO7e^Wi#Ns5?xps+qi%YI&5R*_-ys602o&ar+w!&V-xcxp0_
z9yqzZoWQyW6H#VudMQ(Jj%j$taRZd_{MbbOVn@|W$}gw_)m@_&e5cBws)xAHfWNcH
zQ_+uqhI*!g>OolqDJojD^uITCmi{kn{8upNzb5*4Z05fu&%dJ2GdBK>jhl4Ha_<uZ
z|2OH(HN|ZoKmQeZ|LpKk9I{;Du%Wd#N%Jf_J^XU>)mJ9Ze*Sk8hph2_r#0R^=ys+E
zT--yFb=dg-kFUeVk2EF2(-Bh*)Ec`NYMWT@YFkI`3HDIOh>f_>-))LbrM<V;JL*B6
zBk%e`Z5_vPQfbUvr?3cgJA91)B;~?*FdbnvV8&gVT#?ZP=@+mPO#93LNQ1aNg3$xd
zEZqcZ-@I}4`mMXyaT_vjyn-o&MMndK0(5%3;S~zCTGRp5+L)j(X!<)V)A49!r`KQU
z_D^WAJs9)>Z~5Xfv5?FKa<2!U6t3Qg&nDD;&H0rf+sa*|+52c_W_$_HB1D9`8q2K0
zge~%PU5`8aa)<kLf1h{w@DLRdfjTH+o9=hJJa5oYQ|7J^hc@2W(X0uWd&t;@i@IRx
zKAs}-371hK_XwzjC0B0|Y8NiIaNujFgQ;BkeuykZ0t&1o<W*RbT5GK4F1gV|rJIbV
zM*?I0(G&y0zQ7%|If1FN-?$+B(PY|weCKu5ILSpXcDmppIZ2iFBqxlV=<-a2HF=y$
z^PnaChxMGy6s0S<htUs>>+KT)t9>*cu%+QqdlxN%Q3wNC3HS}!SdgZ_eM0D<?e4?j
z0D7w37T^Yn79f`oNh6vCF4qhHQLE#U65c=M7uefy^+fKC^L<^r)cjuVeJs)uLsV`w
z9*t~)5kO}v>J5xXqZS+qV5Ex<8a(bC4}|Zn;pnP8nT(X5k-Rgx2d|#F@3<BjHnCXn
zuwstuM7-gA{+OniKIdyAsd<JeJghJ1j2kkaV8B0sQd+pNy=0oNZw`3#1soWTF|Wvy
zP|qLnR8YfkS2f^!jAJ8QXA)w$TpO{cN72M<5jbTYvGH^;Lh^?Ut6JI7+9<<p_VQtt
z(d=cyknxXdcrv+LoQ!MtM$>L}K^rspz*I1`;Q*$_UTSTbldh5+2Fy=@!tXOHYFuB;
zM%7i^M6$nupVYp6!QY_ePf*<8YEkI~WsnF>8n~hf2B7MDLAfTzFOft@HvsqgLY!Y9
zHFj*~b`hQx{FxlXgre?Ap&({O08-wT5y2>8e?~+S$4@r?Xfie9*~lh;(Cp&R7Nys;
zt&|&frQyGa#Y>ep=5~5q$rPsG9d$=TXomgHnwlZd3=9Z5)@!TFt1FHWVI*B!8KR;y
zlDtTlNDiY=OV*KJwsGVs=%;EdL+Ua&0`Bj8V|m_@q=X6s$<0EV;iMQ02o2(8CP9bN
z4Q)y9g`=v&T!C|5id15H>QcI0QDu{dLm_#ack>`UU)s5m8yt_!d8C&ct1|^sbkHO&
zS-v$kAB*e&DSlcIZx2JMhlN-10vAffPf;Oh9YX6NG?lnWQE-V44rq;sZ{Wt*m85I}
zE4Ryam9h3EWs4-aC<Rm8j0SE=j5?Gmiv=Pi?Svz~MTATxuXI@sQ1!RCC1bg6@ZvL*
zp6*R5UK3!7E2$)8x_EKKSULdV(tq({p(`mMdkH-vAQPSmLWW54)Foj_i8dLX76%Yl
zK0h2Xv{9?C1Sm-tkeaYNoYejdX``<7PUO1f>%48b65OS<Wv~*(b!NpWLr_ihy0i$P
z(Wu`-O>UUEv679@cZ#=jwQ<GbNmsI4Y!B7Jf;Fr;-R*!@z#>Eu6CWOm7nt5_01jpu
zL0ilTz^O61_6{q|{fU0C%@rC1SJz%mTcYwmGR<)EqZ#V}aN_0$)tpO;Lr$Dh$`~Sn
zrXIHZ3=RFPH*8$ow#?LYRn=Ezq%Kq|*~EvR5_OP-I$;zRf(=E|aVsxr9-vYo^v-zg
ztQ0^>D8G?|k$~}=JW4`0^^&X@QWha>x8eObKw?|T13Rwi)wZ=$jNr0twaXguf*32{
zTMaYI@es*wvFm=!b14{jdK0+HmHJe`R8=z|`7^JOak8ZooCGV>lOFmZ)E*v<F8X|N
z7aOa~0%&<P-eeIHKTsi&N|($1SqS2)_ErPPDK%X~DSkP@o#BHr4-%DAu%Gfau);JJ
z;!~}~=?Rh$!@>izR|!kjBuY}5U-D5qEJpq=Gjf?FS#~bAWWo7alcfbgH)M0MBWux`
zcz%}DnzW&vhb>v_v>7%EXR)J}kLH=<jJa7*@vLM+zW!NLxUa~12{V}ZgC>JvhKAu^
z#+K{9`tVm?eT5%hB>h{DoC*7j8~+@p=&u#*oUt)`Q%ZcjSVEQ=3ou^a*vJ9l*49eA
zz46B=@vR~dXMnR(-()6pv3>t`q2#l}4`~H2;i*l?yDALx6*=z}==89r=JGEW&E-Q_
z$p7cBybLdRkzA$AoYmAUw=tUq`nJ2C8Tp2h%6%;P$hWcy%WoXR6I1S7r5xLu|4zQU
zmdg9u;E`I_f2`>SZ!z~BjQv<cu?F<|!bO{A+*;nj!%nyfdpy%q{qm(t?=&yHdwKP{
zDq<kgqP_eap6XccPujSPk3Roff{5R{aI@{$CY%IQxP_(3?6!U$RA0j9_8%oLbIv`V
zODy{q2VUWM$U@AD@j4sIM3yUDXgg^~Z<<kd<3}C6w4vg)PhXbFqWqb<q0bG?`HaB6
zR@ngTUPbQ9Si<}upRb)zt=y1*m$nph1sh9sfs$d_b1ylHA8$-g0F|4MU#Ch+_{ENs
z2j{@UH>(xO?aA{(SB;0y1K#m{CS{d-6{~U1oQfsy6G@be>Hgea(?X#&`8NnD4yo(d
zn){Nzc5^+GbvX>B8l1TX%<E76eQv{gyT@JVN1CTbgz7HN5c}Pu)=97XsPQ;5W11TA
zxT$ZNV=P&XOUR4|#aab?1w_foe-=DA{GJ78cL3j_F7me=_mG?u7auep%dQ4v0@nec
z+uED<cO#7Hb>*L<FMcjj(+)3AqfO9bm(cZblfj;iPVaclrKL%H8%|xp(=!Tou<?a4
z<1vgFrx&2cIhaB1Ho~G{69Ha0I$fz58Y?wzM$5|Zz@6$42jJZ5%W7A{>94AV*se{c
zBi#0=9Ok(S?uu~A2x8_Dw+cZ!#1;+vWS13Zdl^1oZ`7D%k46{9b2P^XxZ_0p)dB^%
zwe=L_#(rBi7Oj|jlo6Y)F#d%de;80;Wc5p9Y@L%`jdUd*>p(=#BZ#vAvPsj^1&AIY
z2WT2n{CLFu1ysVz*dGQe5gtEChzsOwTecGV?5oCXLcl>AQ2_dHL}kx~?!BcjP1wu5
zvq7yb0|#J&OHCY~!&#QUtw79KIzda#5adiWN;QO2oJ>kdO<=~ip6?!YLYb;}<l(Yy
z9TfhwL|m}r{Jb>7xu-g(l<eWWoviw}FJ9b1Jl*=$<EMp~3a9mi!tP0-px0VLN&l!&
z)U8XQq<iwo#bJ2r>8Q+Q3dtWtmLPpB$w7QhW^sg9Qx>lmz@<mHa3(K1*fkYEFXFME
z>!0}~Iok0rd>k>JYN3sgE!9Rj^;>s*WKt|PQPBI?Gf!AMTyqfDj$z7jmxVN0G9h9~
za_vCoq3serdCOaX!aAi&3$-o3oR(fX!3T1L)iT08SsdRhdWg7!lFm5-rORBjikI4T
zHHWDd=pOU05V^7<o2-0kqCyr$#44Os&S5@BQLJB8r6L8(bz`CS<J|oQ?lpKLmS(d&
zXhkznUaag*c>rBQ%T9Qf;?`UV{$(=9%uSJ%?I9n&?R_eXnE$b3>+<wqyII2B%^9mz
zK&TJ#`BRAhmyt8;e-0Df7r2p#&40TaeEcPOj1;F%2Q`b`)<zp|zo~D73TA4$6?%W;
zRsa!nLVNpzjjMS36$26NLG1l|VmkayecwuaWxuxZhh{JQcljC>#8oy6if^nJ2?~5Z
z1OKR6(s@2lXoUyU!#CBE_|?>sI2a!E+OQ@5n@|3}!@JV<2PF%l^Mcg(yodcRyb$Pw
zP`f)BK$jn5g9slR+}Yh(#?^vwG4-}dIo-O62<2&ezgydclh+ZXX?IyH!Z`iW1D~t(
z#N^FvSNxS}`!L8)caW6mcm??<;GVnxz3;6k2D7d<v)ztdks_(uVQhOVPSyU-+nTH$
zJJQj?08$(Nk#9u;E|w>2sEi%wMJ%YyJA|F=m5d>cGZ0R!Lu{)^le9>*SKva1T^`z_
z@8HLr;+<hX<9X97c&iO-1*hImX9jkRYN{ucjAg#NWB}2g(|q%3f(8^y?N!kr#>_Hr
zqj8;_`AU5)<N!Bnw`jr2<y!})YxsZ8kyv;%wKb=K`_;$Pj`f)hOXk2*#O!N<@59n$
zMkbXe^V*rjY*BUHRiJI~m<+KE8dMnCKhG9P7VP+zt3}t%kO0z%Cx6Q20|m-8L_uze
z6_g?9Op77*7_6D_SZwNxMl{Ua^tO8A7H*szW1|O)kPlWiFc(D0Q73xc82I8!2P=oU
zzil}N17%DErL6?wvWFxdZ+TVa-hib=SZwTo14l(eQEq5_wtwxcu;150{efMBB?=G6
zT%875IJh16+G?$eWafE>lESN$5Wr_EC(`aisPlQGf?&bBg~o?6B?uHNA6yyW8R%MY
zf9Kppq8UHV7iMJZ(+yWh9U1Km8;#W^tZ5wvNw4X1844QuP7qMD2Gvi#T&R_=#dYWc
z57~J*X|uf;arb?>u_qnvOU=Ms4ha7mqA(;OJ==cdn-lF~VUnC9=zS)9<C>QXzch%T
z%A#vYmT)&anh(3Du*Bwh!bspM2KS`rAVCSGBJg~0xN}Q#e?{KJ0pbF;a%z?>aN^wx
zZEC#s_Uh&MGR5b$bG?jP^3Ee8^su+ZW<?M|dTpm(irh3)Ru>){CuiQsP*^-vshW*`
z8QyGRm5lc%C#{OqX(BEgUhs`!c@icVa>jc}*~EIJw{+y^g{otBrjVG&U2%A9EZWZF
zxf1@$fO$;C5wDD{3=&2`q?=BjVLI;Y!^dLF`&?SP)!?OwLXK1X$uZ(DR$9l{@q5{z
z=!A*mT9f8}wXEP@6*7fN6w<tJqu8nuyHMH>LAgQ+!fa(AKe=}(AXWQ-vNS%2QUZJJ
zIHeTKuH6xY-?7A<!aH}0m`(FBWv^?9Wu@;CzQgp3C{cz+(4o{xllDHGo0D#dD0L^H
zM((BBfo2V%(o4VIq7>BfI6QX4h8n7^vL5t!9hZz601B!>@cA#Cfs&-2f>}UW!{0UJ
zC_NRL?e#{8GNga=Sd24f9gmU{^GXT^Ui)j*6LOKKtkXfkaOd1&T{h%g0+Ea#E=Nl3
zjqdzE=2XU$(+^lPDBIv_XfhmG9i1^U<OA_^Eg@)?);y-TS;gn#uxt@Pp*QIs$2%>D
z#Q{?5jCl5L7rXca3b~vH+v(iok{)St0&N~FA#i+h0vj&g8xswZop(}O^8I*5AIpm;
z&d2iCHm0y}e$jY1Hm?4$y!nQEqYb-_li5Gt*!UyV{+cV3Nhk388x66q>h9b}9$Nk1
z-1t2+lKz^(cy{<5nlCFr%DYOG<ZYoVd^SD2mGh!J9qjjD`TUQ+{PMgglVJTm63vWj
z{r>rUE7i5~C%~;_7%e0eW|_$HxgL@Uli4bB0Q+RW>SHsOEjad?*i@c$A3K1;{BWcc
zzHvE!hj6(Vjhp@6aNOM8U6G$F@2<Z4&T`u_2rnDZWlW&uz5}{!Ve;xb4!x&8_^7R*
z&+UHs)(A7J_x1{VzwGV4tqq74W%8HljTwchi-i;OcD{tcQKn&@3iR4Xt@iHj6gjop
zSgwNmBrLw~1Z!|?s|#f=X@Sp_17&2?6vEDHI`(4B$y0A!C77Ibwc-k&1W?dUu!NRS
zgV)g?)ezv8svspZZvv6%orzj%_*FWr7IdowwU|HVa9^y(C;?RVnv7ou1lXr&adAc)
z%!#?29HoDA=X}k~nzhWcFn5SujCX2p_M`~iQPVEP$a!qW)GyKyT=f5QHz8`eqJxMj
z7Y^@DQERhV0YociIaONfssPsuZg?C|zsYBsp64)!QfI{Qs68qlfh0Qf*?CVhRs_#d
z#5DfYWnunBr9dsAmA6Hdip^<^Sg5@}I6{;L0xo!fiha-?Lg}zmg6g9&AG*k6c-2A=
ztMDUc1se1V$stIxh1~gTsJEi^R(6gCJ1fUH_w9~W!rPX;HRS@D$;v5ab8~pQaC3F_
zU8iQmL!)iJ4`!7VfxIuicf=%7NF0rQr0ilkz47i8rvrn2D)1}uJ?u^<y}i>F%fs)K
z9pKs<kEh-7%6Gq8qBhv>92R(&zq0C@*M-{k$3tv@VdIiKLy}!cF94J~9ykwp3yhxs
zg}p#b8Z`=pgrS%>{eF`o7l-25^tStxR<$RX?!mjq!^!CxK4Wb2Mg7JIBbNt=&+#$2
zl36aQU4gV_`tMm}9e*@!k1!F=PhQEs`nQb2r8(dYBK8GD0ZEE{Xhx4Ty>1B>m5b70
zoX^f(DsGK%F|SN4n_m7~admK2n$^O^!%Qgcj26h6T1cCk+M&pUvMOa<yJmQt*;>ax
zw14<m@z|VG@WqXkImh@5xGm%tY0fE)6#E0l(1}K$VssC?D2X1=;2&6tM9`eYC7A52
z3aoK&zuyRT;29dXS0Q5JBB}oAsHg{D%ib(>IRhz!yvt%4VPv;vC4mA(-nw8mq-;M;
z7K$5WsX$PM9p;=7F6zRn(ghLi7g{a*E(Tw}dDCcmd?GuQn5TZlo2S08@gKl^^aW~E
zv2E%ddHkF5*acjgN$IN_Z#lm^zJ7LiEbed-1U{G^w$&u`d*%)|><zoH=60Ly>15Dy
zzhEBv`>$1-hctKNnxt-g4R(h+T{vaI>Qi&|YlKEaB5lQ|H4RinBEjASh9M+dY_gWx
zhYSzoC^#B7?T#l}m(8T4OE1p4hPbQ;M%0!SX|*RUB(UTCCqhn?I5|2x#xXcF8%^Bf
zo=-+FLX;nm+Na_1YhX*!>xyJyF!PkQ3y08k_aI+>544<2`%tT_1Mj;B`SLKi!n}?o
z&A!47|0G{yblgfBxYI?Jn~u2r-DAU*waQIJRijq7Zfsq@dE?gg*5>-w){R>qK!!gN
zTGl-=>Q`!>4k`-jzT<f3`)uz*%Xk@DvbX3^QXq>}C@Y-@oT!-ed+D{OuX)sOUB7<q
zZtMD=-n+AY7x?3p`8BL9G;z#erN$!MCHrIS(~hB3!_8>%t>&9=H!m&PnM0OAM(w>x
z3)c|tM`L~m&(`&k5YD4z6y08oYl&M$TqStV`-*XY+*wl!L}eO#S(*yYbs~$u%$h)d
z$F?8dytx@hs-hQ#y3&zC#d-Tu1eJ6rnZYxA(-{tmgK=%^c03kYBB0vI<xIts%W+jd
z8h#j~=A`q5wz3sML#|3_<t%D-A7dNXX>2dX4TDb;{M|`P1uw`!6?mjTRQEU$i!*7B
zqaLQDLGq#q8^=F4+{-gbk_98A$-y{_Lgq@bn1>1}Z#f*+u|RFv+Nb+ntVp_)AiGIT
zPa`^ADoWVtV2f;P^?BAfDi>Vm`1kt)u@K(7-<|j+SB+~@mgU__+TwDl!jF%cz7wn2
zXn!1@$?YxaBVRjxe6jHJ;zKMQ`?@cV&Q&2pJOD|JEON77E3$KHCn(i|Da6-^<yTsb
zy;3XFt-<7mGp2Mq>aybL!D8;Sx46ZnUk|0BV2Jsm-`&M&aT}?qJ6(FyU`_2ICGuhg
z1dJ#zm2~~A)t|VpLh+~da>cUNn7}?qC(r?7y>h!W`H|i*(+&qX^Xa6BMC!9c80(v-
zvy)8jopk1YYQM<w5J{WiBR=+^Qs&{S0zb*4n5Xz4P|aqLgZG-Ky!ZbSC$hi#H?O|3
z@#T$g{)1<S5B%?MZF~#AJN%9Rzq0YRT;cjV@x*P$6Ug-7D=a4YYQ9vY%Cp0})QP|D
zpT}HjFTWVWw`a53B*f|AQB#t;|87xxq~hzA73kD|@^|LfscSz#T~TbVK(R)`Ar(pR
zzu=24r?!hComcs!aYt<@m*dJ_81kVlbd9)x$plKq^-2_w+PjB@SYX>WRUYcz8g1f4
zpSqmWD%EP)5rQ%f3<4^EBH1%Uf*rLBNHs|n-s<fs5eeHM2ouBU7B^SHLmfdTq6JHO
zV`S>|6BD_rZeP27MN^K`yg+vcshc1yi=V;b3uaBj(uU?rkg&Z@8uJBG%!L-`!?jv~
z)SCj%i3V?S?skw(P^Qs;|Bu`>`NGE6v2*--G^B7o7#|HL<5ugL&a}VY_-hpTa$H2<
z+YI`eK=}RmH8KexWqUFXR`|C!z5-ZZOA4m}VPD+%O`DLv5tRZeoQr3-BE@pJ^2PM<
zFLdUWwzE563%|DF;)b7B-9=E`%T`XW_Rg-GMC(7eO3@~tg&NXCVBfHdTzkw8<rEC|
z)y>idDi~7=m&lO)H-5-|ed7zD;Wy$M`7wi7`}r5+V$Tj=0TsrFf2maX_p<UwHcXH{
zul*8B9}#pfB74xuk4f?v<m_CYThbViIpj<#Mo=#=@uHSz$qm+@AvXTop9islJIsIn
zTV$?#v?4TF)0>v~?%UyA5C<qlvU_3hWfc5MRIo}Ee8HD_cKHAPI}il3p-Yhxa(p^H
z{43@CZx?w#+LVC#|KYc)m)`t!Bn!3g--GKkC;64ycURt8xqRvJTeZupS5`0M3X|_c
zPdI>82=9I}t=;VL(Oqmhq&R=P&%08jE)j9dFb?TnGsDnetjjhQnuDX>j*~50_^B@t
zViFm`uHjzZCGk>XGHSC{*^-4oqGKwHwQFuUJQ~1!Y;UcZ?jDRETI}Xv)V@0C@AdYr
zrO52z%G?86@h*_{(f(9Y+Ep0Hb|`3N`3NQ~4(__yKjQs%m65Nw8x><s@nNzT#pcx^
ziVddV3d>j69O|67A>GxTM7_9Y00grsa(l<kK*TW^CnoBBjY^0pnQdIK;wc*RX_C0b
zt8lL104KW4L7wAwoWkRA8(Pq*PM<V@HEid0kuDcoYSyKj7e{*=uo>7wSAbBkKRssf
z_wL$d1lEd|lnuY%&)x(VI2B1mkNCQQPjI)oo%)7t6U|=jiFv+p4{YqC`Vf5NxEW!!
zw)njeF>JQ%QXGU8%kelF7RXNFMLe<~F0~zUDl7t9oRltJ!F(A{q*I|ci@OJdakpRb
za@1v>C2WRDb)TxXxFw*WgThzYBhaBEq9W!G;p=~bqiK@$O0AAW7JsBTy@#*$pZ0&+
zU+<4`PYd3OUguR<4l^ez_geVV@Z-+>P;RiLiu~Y3NfN{2wZq5ECN(&M3vL6)7BFWc
zwGuKS`S+v`9Yh|}K}2A%x#m7R02=j+&BO7a4~AqZji+D{-W7ys$7ret#$@d^fkW;L
zWF8|0@MNtHg{9Z8hsObHtt1@?uRTv95Y@O-Un)uiTsqYQpSE_<!{a`CZusF0olj&N
zYI;4C@MIcz5`-?f$_OLDs5kWH?PFYOMBf2>i6FeyOP4NPaFLIksk+1DqyU?>m`IB^
z7vU5iF7o3xZ40;xefhl_WhvQ@XRM>h|HgKxfn!PUx`KnndAq<n9~mc=ct=Lkzr}!B
z<J)!juztUg?W3;jk5+tHhkNU=czj=X(;-{@VO-p%hxN=6_P9L^tTKaYfdh7!^|;$9
zTpi;@koyuUjBwE|OoPS|Yeg1u?BmqsU?vnT&{hxAaxy!}$cUlSMXVFL1?`T+LKdk=
z5|0$BfIq_z&wm<{$(c@A#>4VGEn<tED;57t3(6#Ot1#eEe5WONpi(R#C8WpZ$boza
zFxFzec1NiaIvG6d_L-_S>gf>6Jc!P<9W921J`6W`dbC!cNb?9PK`UCoSW2GhIeNCZ
zyb2@kt8_+#A(A}j+pw2uwFxK=TPOwv7vYW9wmp>ylb`xWlIJ9nSo$1qPAcC#tb7CP
zD&OtxLTSx4gcqond-G`0Ej%vFZXP19)%zI#LOBTWhb@UEu3ugM(cSf%t(%)sk+41p
z17LM^Tz9Xio}~i}fAyBeAuH>&Wd6ztw+>6pwapo&({YLl3hT}bwNN+(icA0++LOV|
zze}qe4fe6%>{nJx-ltV@2c1O{Wts~w(@J!Gpd(puW`_b%+4_vq6=21@A~)ipR~fuK
z#H3&f%w6ZkOvT*{jO2Y<mBWKWZtWAR@1L%ThtD|TMNT*wNBT;PGYy27DRfL4gdS0&
z%~T36(@LW*H#DuuAXCS`ORL~g5_hlSNavJDt%TD{Wl|9FeKDDa<JW0L@x{#4@IIP?
zP+DhaZ}pwRdP*AG!a40@OVaV{r(A87)*nq(ugWy*UZoW}onZ^d^qrQ$76)YJWErnz
zz%O|JKHaZ4_H3!mc1@h$fg^M*%naygCh!w_=NY$^R4X_Mjd9y%rb~Prx)TH=H*F#u
zvHe`Dh0S0p4`sOz)s@bBYA)7j`((+X0<e~d+_Wy;?lL&?aavGdxD{S&B#+Y4GJ{M<
z)57r+ny-bUlZV{%7B+_Lo!1SJq{6g>w5>={Sc|@=g(!Yan}+JWk1a)cN|ig!T9w7a
z%g|I9O{b@ZL5&UrlAR6_iS(WY*$3%GJF*GwhC9{;=Y;Iwl+3Q4l$XuC_ELV56U2ia
zn=RP^2CPZuc|HP%(yz4Znx{6)#QUYp2*j@gDYLyUKWt*F2N94?m!tOaPN!Xad_}yq
zA8)%o(ESV8>u?i>N48<XBc$09xXH{9<DAT_&}DZYP#(J9;hY10=(}+PCyK6XKrrxw
zn@0)0@uZT04u?oG1dQ#7%gn_>vs-u}R)P_uus8?OR)33GMb@4`v#MUo(8fK_atv#K
z5?oUgJC?Da?aX7Yd>+0b?sl%ueU}Vcy3;b5TzSmsWh}TtmKr&k_D>@_kAH?hp$^m=
z$Y#_DPrFDyE5H1#3GaN3R*&I5;f1e+7rvg1TPu;7k@LW#5j|x(x&k_^eO>8;dn$d|
zf%_%ADzrTM<BO*2uQr<5xatBNa1MMI0dQAW=YTgwHy1E6HhLs#Tm5EGBO}xnYTs=_
zcAkVycPNI=4aOJ)*w_wn6ZK4xxQiKT6U|%x;%c+Sk#l1Fb?gPxA(k@Pw09bzRUn6t
z9RK~-|0#}M-lXHhw>Cb2qr<=C@4uv@!{6KZFK~+TrC<@!uG$&PKascJiryw;Y^DC;
z#y<v(zZ;h-4~mt&v+*CH>^Gyb@ocv<<E@P?6pEWo7E~*Bd*gNiR4f6l&@D+l`n9M%
zztUQX&5ccf`Fd1B2rGMED{^z=CW?GDDpK%=-XOx_2lAEjQvdAmRh~Y-7OLBmsy3x~
z1<LX+CSTfTc|C)0rRpj5<@K2ZS!$%f6!Jd9QhYu={1+96=wR*Ot%yUkfARJ!uh<dN
z?jW3G=_8zBHDR(GOh<6UqTzk`aG&;9ec(>LY-jF%;t<RL!ovmPoIYIs$*G-?r3Yc|
zP-LmbQ-r13y(#RvOJdVqs{N>sO)QR&^qdP;$>FIaBN*C2*GHXWcWyMncUMrNB3}F~
zOwYHJYC8G7S2sBA<nleW;%P(0^UCuaq=|vjI;oo5CLIIU-C3P#7T5HQp!k%Stn2BV
ze+B>d@<AQEV2<guuf)`Lj@u7mJ~`+fsTX34wPb%I&JX5{4<DKE44(Rv5q1&Xy&i4|
z5z)^3tdP$IJk+y8HFcdxk&rn*ZXWExk_9gfh<bVA@cH5U2-)Fd>3V{`E!SRgcU6oa
zOSO&bKWW{)di&0G{d(^wTlVe7t-JTuZ(Y59H-V?)>cRp2Y3t_o&D(bXX#M@`EsXDn
zms{&MZ+vj8b^X?dtq<4lwC>$*-9eELuS4dO?`4N-6M&2AO?lR6$>~U|#l^wj!<CLF
zR;vn{>8N<?I(sg(TuT8M;rDvoBjlr&u~@??g2f)ROp}zF{<X)Y&gcm9n-xAb?m5d+
z@0W3E1wq-BQ50EzMKcw9E(~SZ4=ZAWmMIj=*(==om-#wCL}!(MX0#mbCq<D7m^xiZ
zDkuGd%!>h|VEP~}?E8gs{(D*^?iEVKKQRiTad;&)sTukzcwecdGbFtB2v!$K7I)BA
zB}o7guTh!TsIGAo&h!D3dQ0vq)l=`t>=n@*`?5yTuxioBw27-@%m;b9K7kWFRAkut
zaJ4K{X#8Zt*E+Pcp`Oya68u}o<^ad4X)ex2jQ-1MB;74S&ciD6A$FgZde09O$B&=A
z`j7hF#n+vd)_4+7ds0P%r*f_+eD2tyspr%RGj_J+6W{0p&N-3D>k_nK#%$w^ei|-V
z&F+TtcDuBbD$<%TiqcOcH6XD0`dw=***h?*8i!L&ngHc68GC}7m>+AnP-?dePYfdH
z5KzxeXM3F0TVeX#UZvdPnl|X^Lu~z|czzHbp3!@?S%avBmDT%Fgb;L4y=s`5n6{G3
z(;-MeKv;;>GA9qt5t|T{G#L5v5UFzJCaGFH-LEj&NgjUWWEz9xz-lx@)*ywm_ENzz
zKs6G!;I7|3nw(|;r3K0Wj@u)g13?tDk8(#I@#}JBa^M~!2y@2ouswCt!%J&kWo3tI
zc2qxOOq!1OZ3r+%$(R-2BSm)P@*u-GqhaU!z`+5YRZ?w!eu^QyI6KAA%H)6P+!T*?
zhL_^kWo+hS1`@hB%P)snE<+1HG#2A0G7W&5BDnAA79gF9m7*)fbEb$&izZB#&g(?J
zImrNTngwT^k%K^beIa)VV{F7{zSuQ(FCO2u_(w$(o-VBHtygQ>d;dG_2qw6lHMXdj
zQkd%Gqj=%VvgM|5o;3Ga4bufyo*TM%DRvYEXPu9f%9KH<n9K64J4q>4g4=*LoUP^G
zrq$5Sa|+u9S;=&NPRi5Y=h0yFkY|SlDaBFB79@*3>M>jk?qb{w-GjCx&Y6hMz%`c$
zdK&h2AHr>RQL_}Zt~POM3>U554*ceD7P@+=)7(<nAMggC#oAzpeu18`g4p(tY%PAX
zse;!(kLKn}H#>^9_0}()?UB=tA5R0L&D%|AKN?XN>@`{xxy6vf8<Takdw{Do?N))p
zf+sI`r=50F6XA$j7umdhd-6q{x$<+_mn2UG-sL=>xxo{!J`t^8&UI%6322J+auNfg
zxKuQi$B~<h)Qr7nh|5cHm?;7h0U8`?Lhm^eoNtN>{33P_#%of$_%y9!&MfB7Ar+EG
zX=$@dm{W_CP9CMD4G<kcV!-&FmUC)+&T&R6&L~g;DH!1~pxU|zlDtdnC+;s=EqRod
z=3o>(Vg%pJi*n7Fh%&6=GLx1IhXRUB_??z3d8`=jMQ4|^R_v*w72+ppS@m6!-pSYW
zt$3-(OMXk=LDHL(juh}e3#FW&iWah8g)-_bnss`?vT^Np?bhvkwW~i|zxBcO+Pxp%
zxXZhuPO)r2Z>x1jBW&?qvR2Du);ySskJw&&w~HO%%47weFWtvdcDdKb`1U%>xL9I&
z^_}J3cp2nh9_%ge45t0gShBeC7TV=YE30o^{@%MQZ+giTmKfXrcDg&${l;E>T{CAp
zQ3FqO3!IQ$JC++=faqMVJA5aPo<@gBz8Q0baL|I?NlM5+6r^+P^-@uq?)WmMo4i^s
zaSX5+;?RC@9GyF-8#K$^x`w;A!tlm=w=00GqM?9Oz0EWfxgwQ=8Ym0llz){udlAQg
zz*w;Y)u^e$j(`kFlOB!(U`NVZi7v1d^QAJJ!<$`pPO$*mMb3Vgt3j;RpdS}MO%d<s
z@}jT|KeQ@)12Wvj!_5azeSK~x<wV47vqe{?-3K-<=-Soxt;kQEjYWuTvtin>86%Q4
zg8AnC7seXs_Ej`|oP{(pcolZ94c>gy9q+EW0RRh=peZ@)kx7r3YBThmRX0GTTp$U{
zC?C=3TC_=6hK_c<^S(I~IUoM*x`Kk6do?Y?G$nYkFkn72AOM0;ZXQV*;zPZzB<1K-
zFCZubu(pk<hh!WRA-<d?+?R=Nk4aVSct98UdioO-Ki2Uz`&(JU>+LCi3B0t=>YoFT
z0GzrwuL%BfG%ikbywcMg>tS`*OVrF@U4-szMU29r;4{mVJ`S=#iBe-YmfB|Aml+*R
zUX__GI4`m15(Tnd@Abipg&UM}Rh^SlRMj!xb|CP5owakoa|0n44m)H#ME91%G(Sug
zumaJdTms7_NcvN+`$+h9M&qIE1>Cn9%kYHU(WI7%;Kkp4m!+so)xf46lmnx7N?Ghc
zah@^bHm&ZEUy9<%o5cdS43xxm4<M5ej9KL+@5vg>^v0-9&MN#g*rwB>X!V-eU(*-T
z%_{6sD=ppBlU7771%Hy=NvQ5=X_`Y?Ti%9$WDk~?u^vvUHz+vl!X$K0j@dPWDvhOz
z!<x*tBqo(84pJ6%fzs<wqD2jfT%4ds=cC0FOJ)q}9Om%?NQh3Ootv*0L}Y8<D6y@V
zVd17c@)QZ-rA|!gInP{)Yk1)Eus?XD(S5MS+hI@KwWr9X0e>(K3=`(d>WV;27YJsH
zMdg`~N7Ps#>V6n{`4*4=*-Kx_3AQ{KC&=Y@@xo>cnZ4!^N2^1+c$qe+(m$|;f()B?
z=zBjVS<=e2_iC4x{K76a$|Kxf3$xO=izqvuBtfl%R|!<;T_L3>PXW+HgRF_2nIJ^F
z*sOhs3;k+LIM~*KDOCI;oTz!AE>x7Pfiw`*PEovJ)RNl<`CCDZp-9ykV~ZX94p2r`
z7c1!88{um`u8j`{)1yvpN3U&xB<)6HZ-cz86(P|J*oAIE)zd<_saL$X$?Cj<$J%s{
z#}}0G1GA7%M1V(2cD0@`2cLLy6``skP_@l)T0q4XL0o{g!D~zCSu9g9C|hbY76!4o
z=8jQyL&jNl#s=WWFoV0yR0w|F6SoPZXaZ{?R)%S&pai4$Fhz<kj7yAglzBAm0JiV3
z?wzh6*%kLt$ckwdCXWUv%|z31e)6ShHAciMIt)kO+DUWVtij$o4yH~i(1wS0C}6h_
z@j(K|wH!+7lbGR;HQL|Ti`f|h*bjg&Kt4AX<hVCm#Sty?iRg2I<%-3X0m6Eg5w@CT
zbiUTkZYcOtoQZl+h1qz}R&D-NiNbsrQyv$3s`-+BY&j+-AtCDchuN5URrYd7L+b~G
zJIkzgEeocOcs)o}j$Z+Cf7sMwO{lgIeG+bMeT2ZGL1gVRHxuH?p~t52fX&9hIQmQ!
zU^hWgH-2arXz>}GbR90f&q!T6v)49B3T@)bUmhpG9)wPE&#5i&ak+^MoznW-%d4^x
zjbUytUk*CNJV;g#k4G;iiYI`>j*L+nzU`h!x*o+n8rV+<6y^rx3T0+(VT0)YmCV#D
z`m+RW#_=33nMFdB4vbswNxH|&TDig&s<cOK#LOwU+URd9x>*0?2e_<8BB7)jW{4CD
z%;^bcBCD$5$iwPPsh|SQL!Lu4gtFO*`pYXBAbx_xnTMY-TMLse9?mRJe?HNjq>H4V
z=a_h+Uy_;2RYCINE91xGGcMepU2;eJ%uZjDi67^#x`}gz(dpgxUN?3R)m2b7L6)GZ
zYK0fy1lg#6Hhop^I2Yi}#DBbT@nWS9H<rlQ;T743@j!E!w-(`%qyqr6FSSBy+|Hci
zzo$iF2UiG&_-DQ#bJJjLkz_nci~$4Fkm;pf^sE>6?K(5{)a@qpF&Khbi(PaB(>(=!
zP*4!RlE{)B;{`Nl`@sixA~o1?b*)x+g<OIQp;q9wXwRIdXH|h1&7i0zFw*h!G-jDX
z+6qkxg;W6a_m56}SQM!~#s8Lvdbs)qZgks=xatp^AZ&^TBg;X;B<g!(#GS(|4zExC
zZ1r3F!|7twRN$H2V+OXR)Z43juw8^0PWwnUy1cCL5YCRno+m=QxaZAK#afn<k&TXB
zK*}flFCoAi+|;EmMV{R*6pLnR$4Gd<QU9hRB>SJVy8RRM69H)W(i;u>IG}`uGnC`$
z%t;%;BmLUkK?bR6zZ+U=4hO>qpe>2@Gc@Ay+db;>tOXc_Ae<&r2&Ch{fnLJ^rCc?h
zbYL`F8=sDGPX)i0UgN~I`ZXCVat7z9@B{A<9M$f&_WEle5{zZzUZGtOZY}C#Xf^cD
z&S#O^et_fI{7Mk@nAi2z5a-tew|BUK-@{4a#GXK{H7aHGSZLDr8sPxE3xk0vnM-Si
zBiJVzt|$lw;sk`*b6u+M#Mid()vwC|cV~*)`?Y*W5(Fi{@Lz8spinX^|Nh4BBf#)i
z{=)sPZ~Os%UwrBU4lit8d=<fTZ^ldHocnnVA>91hs|uodJ`fgw`2TEvne8tpm5aLw
z>W^XqH?805{+}CvQ>yAt534MjR`q|~_&2D!khJQmYSu7kkG2Z`YU6)Fg+E9tKo0RF
zy6kLib^h7LzeJsXKdDpU$!InHV&gwUjjtp%M6B5R?{54#-v3Vdy~V;>faF`e59a6J
zNQ(It*vh=Q@g~ZAJ1LW{4p#8;#$^;_jH3^+X@<qh3axIeq7Wk}L!m-Y=0zfP`1hU}
z=WQ;(x_O^N|GRmnjssse|Io@dW|ajEyu4PlHox~O{zDrJXVjy7vnubty7?}vRLr%Y
z)mZuOzPkBcmVdpXi5yKg-+2}Pq0)CMDi!Iw`Sz>$4;6mDqC!gR&8GHSLSTPYW#O~K
zb%wou^Nv&4QmxNke=#0GS0#Ry481^{%|cS7k`v|aEOwRx+4&%E=A2tncJ8?=rRNh3
zQZ$uaPKs0upiHH+1)fwow}_Qm&nMfZ>NoQ2W{fcp_X?PK=Y|xXW8z7<`R23~J%`wk
z@@I%7mYE*@4;ncAJH;J`c?k|i2%4VWf8`Ze(|Ou+UA+pS+{tzT7FBvfASe?D2zYI7
z({dU#?K7xKT<k4_2_kXZ(0}{;Q(S{(IqXs6q}PEb6Am!ig?K-Is9@2rC?z4jY)SET
z`EQ#wBBv-}J%a{}{H;K8f>>W-24yUyd?m67?uN9~w7^%QREl&*!_A_Cxi0`S9nLZ}
zb*!|VbU{=2>lSgI+FZhT8iCFL!{Cj%BnM}n?j?W&G(8GfOKoBkAUN_P18yM40lLZq
z8Z~a>_}jv8i3U?^AaM>{$_}6>@|O;ypy)|G;rR{Mq$R5h+R2VkYILEMGZBtzjfgCs
zgp9hhx04r$CSfLB>^5Y5u*UbhNDc`ikL?5-chEkb)b?>V55l+U>LL%CwHp)CPSMQu
zM7zn;EOsYqS+CO}d}urVIrk^9^!1iCJ1CC5*cY6=Li^Q5LgS%d`W|$%)~t(Zw5qbB
zaAqIMI69N<q6P~_14%c!v*FT>Eg<PYcM|Fa&{}p*z@)&V5$L4^P>%xjI+V>SNJez+
zInqMm&=4jnor8vX`)JG&KoZjN02#=+6=7~>q&6TRR_8PwVC&1)e%$TtAK+p=e*Xay
zO`>J&M_7udc8~CQ$1^^|dhU&;b^)J7_J<@4;^*&@B1ld5gKjwEA#$*}2iWnsuQ<f^
zX%-A5__aGh_H<W-`HPViT`%L%40n)a$UCKR_F=p3LiN{!DM5?itkqx5<_qBxr>mOB
z+}f26G^3yq$}0X$vzK!^aaZZwgB-wX`aB8#_o>?@L4*6{YB1HgyXGOOCe!n|vKLqu
z382!FudXQ|jNFbqJZ?!cJY1`YoV-Vn5}bETTiwE}7jFj;I~W(X>-TPNT)%bW&#vFO
z+uB;cvwrg??)SU9DIR0>Juq9h&MT?!VZVDaKx!eVEfnG4D7ZuOM0*5J5?KAm$elqB
zw0v^@stdoYRb1d8hK3u+e2!b1fvrxA2fo$=TY4nYL}iv=JS-jg7Er+4fG38)T8=*Q
zAn3}-$Db;-79tcJHNY@jqZA9@5G3$NQ*jE8g`dFHs?rYmMwb@vIlb-{!WU}D7}C01
zSjM8u7GJ0}gtbtRL~9<}lYY6={=_6M2U^kU9xc^N4h9T;AmP~Tqze#hxMExm0tGHD
z8uQwL3!PpoG?BuK20Y2->`O5N9D~gKgWjH+_0nb}3|&PeqGTx<Ba<(&mv>F9agzp=
zE&7PBnSHrskRX*K2mS@~??AU9D+z(0!qm-T5>4%35)$Ga4d4;Rss9irIvS8Lw~;;s
zCa^~R=IyIU8hmH{)&>(GT&foh@NgPI#t%^?N_BsH`_9d4t-BBmth*Z5RXHj#UBq2e
z2`kzGn+ud0h|9PRwP-k99`xbVa*X`&$W*O?eh0|CrO_G6W+Vlcrpyrd1po{M9(A-G
zj8dzn8hXRh0D43VMPxu#qT>J}0*p)rxPnTdcSolP8Hg&B&%FJLC!-$+BBEL}vS<XW
zvF6+6<zc!Ysk<iZm=MUIM@Q8)nJl(_riW<@UFAte81TvH^h#=HmOd-LAM`r7*UpWm
zywIkb(HoCi_wjCbI7tbp21qj~LYQN4#Wzmti;Kq+)dp4Zx_fN|RLD~#+rRu)#OI#C
zIm}oVtg%0wT;@DDW+<^F@DPTGaOHZUJqQ$46!+>mz-s9)y@_qI3%V>S(xSc(=s?%$
zo%F`M(2HB8Fe#R52rXswDa-6=SSr-{epJNMQ${ih1*E;4XIuq#hPGy>VIi0g-uo^e
zH@vB(R=~kGkb_3`I>14bMJ&?VJD8mh9fK4Qv!C5oDC8?I$PPnyG$!XwW|X%HBJer>
zmSxk{uE=$wT!&p#xZpUyc1E&RR#v7IMi<>;hy)grWH`(Y*-r4FVjG;JE_2?OUwvKd
zDH+RyWD)bPgU%Bib5sW}g43rc2ua~vnG2rb?Z{7ePirFdz?IpV?y(yI8EFmC89*G8
z9=<xRB*!5l6bbKM0f=+6Pw%W?A&YUi0*KfEj_EcNow>0DKQ5N$itR`$%Lx<)Cg9Eq
z{OdV~K%Pbdk+>-zp?_XOO^=cy{KLTw0i8Vq&MxoP(SzyUUUYw77y>6??oP3gjK_OZ
zUQw<r8_fdUugWBN!jTO9y6|)<nvp7!3R5kewNIOo;PiwuPox0X&ktcBn@(=ebeH$G
z2umWL0Tkid6e<aA3o1B8al;mI_dphDI7TDB4YLVuZn`yCLPj0NHo~?uIq>@h#+^y}
z>xjX2v7>=k1n%M)Lt{s>d=`>P;!)>NAwRrE7Jt+(oDA1@kx;X>chnmeUks-^$K6i3
z!Va=)j(9B+y$*|&P~&*FTxU8yK8nkRnFI~PwnN066Dr@ObY-;RDAElVV_;*2y+72$
zh=T61CrMd<r~p_1<N;W5iGGYksT0~^0*NT~)nqLUlfq_}m$2atwDM7J7ju%w=~L_y
zPa{IZsI!CHm!gx{h#}nOfTO-;TZrpEK3aIPt=ojKh!-|&tQzgvmQ(bxXTzOyyhm=6
zUMQuF6l*61utjrw8Nb?wwv7gWdQG?US%9H@+&}P+5ejUxO;S-c_7sc4RfrTy=Oh-*
zE|K0jOQ9tc6veOzAgA|AeT_$q0I3e?#l#Xr6J*%zZxC?=Tjt2K8}mBv`I>SP)o}$3
z*OD-oT@zNr=`_B^Y8r7dmd6{<58_)GkHByU#}fp#c1Q4zevHf0>V+{tfim^*@Fn#i
zG>kp~Y(*dNrn(P(I?vSV{q8=pvf}@}hv??PV9+V{1qI7}!NZr-7a#y5BDQ3}q86|~
z$IwPa$M7cVSl+QP7yktPT?Nh){rlA2Lg|y<VNHzIE+^+WamtJ@GEZephW!iM_Tq}_
z^f8%XW!2QgMVLHpRs$2~&D6wYgu3r#oCGuPP;*EZ5L9Cd8|=*`t2CH(JhALV);Tkl
z#L22t5S$dKPv0^ahR>|*2S(~?c(s_Gc@xF}QMzz@$*L!+mGVZGrz?gVtQdO992oUc
zPF+EqJx$fj{DX)k@%Tb31qD|1B1yD>ofXv)F27yWk>iP2iAsA8;zf1Kd!*WXCLuXm
zwu?rD43Sqq?Jmw{5_*s(JIFKpFnQ|6HW)E3<|MdZRif=)yoUoe)CmRUXr*ME{5Z!X
zrGTly+gF~WrJ&UiZeFD4Ta(6KteaPg2A?-pFFX&q(oSIV@OC~uWyCF!PU00NNwHWu
zjB=(0_A8cZVjBfDxFNfhwchrkd|5;iP}yIJGH9N0GB_z}Ixoo9sHr~|Z68>!iSY<J
z1V_g)Fpt~ZH2`+jC0H=Tfxcl#S&G9OB+jZ2k+l>_{TU{vmoYRa1EqN^d(`+A<P+Nx
zfK<w(My<2hqm38SejVeMv0_<dWrimrrhSWV3l+o^FB}vj(uQ>ls`SL&#7acm$2PjY
z<9-I|P~d2gu{iziBfZ_gvW6ALJT0=6qPRvpmnKLtMOgyI(Zpi~5Yc;yRKh8gbTY{?
zS*U}6rb4l$f(n;ys!%3X^U~pxY$PC&CyyN|8r2Y6^{_W=;e2fuM<k9yvaBU+ueBJt
z6!VzDQkJL&wUnY{Gsgl#!SWPny=a~!jivk)b!Xds-|pRavr%E<!Ifz8s8NHBxEdEM
z+|;p9ylKmu)}j&#UASqpeJ&^*tw{v!C%$VpU3pj}s_*P6)>Jr93mY)jaGze-V7W*9
z8zH(DeJ>Y|%=N5?<;isU9FjEs@73LmM>3-)(f6k`8$HzJK8XFNSAN<DTzBp0+7scD
zr+jETT}XhnCyq3D5v~2ih##lk`@OOCj~a2H{L|XZ%F*UIR9F7Dn#qus7iy;z6eh11
zG!;uVH3xE?kSe!Un_dIPxs+f~vDj7>C!sv61!j*3bRp9^Yq_pM5Yv7W)xZ>5GP|67
zjaaDy7U4Nft@g$mqPCJgE#M`Na$#kKT$i1h`UK&87uNb|2tOtqrVoIWpr#Ar=!uXn
zT@56^QEd?Tb;O9IuO)Q<O~i=mnx$ehaAPWgQuw9S2748_>@=cMX?2GJa2rpL<8oL<
z8IaH@OfFX;(qP?ecdsHlCMgieWvq|Mf(xagz>qj)iYYsb(HdAi3Ag>45w<-4*$O|n
zU`y<EXaM$AHyzI?Zz2yBBA%I1N7)C(ZQ3QQ;m$|Hxdb9CS2JuD4T#?bJb8NGP0vt2
z0nbj-52LBW+R{jvIF$*El7ZnAGG>O>QO}2mYKFU{0dH~{daZ-NlS?UM<R@+kI2MX%
zywU|OLEX*ISQVOMKgb?C)qFC5Q5-CxQizeq-N#U-%QSOO&L`v&bmMmYESKnDP7}xJ
z(IjACREj8iEO82^DGT<v=1osj#3W7{GhA^7NZF~ea0xqIPw-w`gy@5#UyvOIm!o52
zut(RSM*a1lEWdud{CcPM`VX(XzIo;KyLCi^V@sRCZBC3mPEXP2C($n+6z<?2%P0VZ
zlV5}!r}C`6K80tWQYuJQRB<GAV?YB-i4<<u6~>#Y<t(oRjxj&#od9!A;|cN)qD7i@
zaT;#}=bO&7ENd0{(5*1s;e?%fak^v#0THknME5R`iGP|Mc!P~?Vy6?675K4AYN}|;
zA14R|&>OYIpY|6coS$<B7Rdk%pND0g`m<2G8W>^*An*g7sdUCiOtw^epSOs`OailD
z5y_NfDXDgmqB?e3FazogT)^g@%V%2Kg5{SBF?MUT#D5~THX6-D%0s5o&qkfZI*|1^
zf6a;iR5b=~e1^<!UO;5zG}t%V+Hn3x>i_5Fwux>4KqVo|=jJvQwnYmW=p(9jVZ|vH
z^NdI+&fefQ_}+H+37gxrNBXbeWce&<uJgaEEp|zt+}w#RDbHc8n`fgVsC|6KI$a|O
zUxYT?)yx6u1x#;<e2vY6pQP<AT>#xy$oKB~EN2T?lEpvKmP~rbo7+FOTubn(n(l^#
zdoo}Nw<N}}A}%_<pIhGI<%yHjn?u|nrYkp#L>pJP7dIov=ax4ikJ|jqI)JDrrOxM;
zH!@H3@vQc_<*j=HC#u<&H_@uO!-LUjiac-8W)CVqtu%M!#qDA9?vFNMm>Hk;cUyyD
zG36Tvz$vN_iV+xZf=i^RD;pUf+^l!6qFo=M|826sMdUlfu$B;2w@6|v{Wz^^w$GB*
zo^Pin+!C9Vz%rj3#7LFT4PvB%d_Fgb&1n#OL9<EG?1Z^EuNlb=75{#gZDVvI%o)SH
zUukf_%y^)=Qy`k(5c7Zqu@uapdCleV_99Qwuyuu-@gmjb%lv0CRhX-@jn{#z3#-C@
zhQ4+-z;JCRr+BYBQObP-E|LKL;(GrV49|A@#9%Imdvqqus@!VE0T_JtJn<5;-IhR8
zw{=BSn|eRs*Z@w~jrvdfI8<&P!l7K;2ADTH<3tm-jJPTY_lzU{OaiA7TEw`IkPX=x
zl4YZ;Ze-Z`^|AGXoBP1-ho%>bx-g{VXJ}Tk@^S)%!XOIV{?%2J;dwAe?IKdPqfl|M
zVWC=Z9~{uL9=MsARZ0gy#dFJC99On!hFzTvvV1%innoYQIFyMH@;wI)vpxz=6x_qs
z8W4NO(_;X$RP0JmxMV*Gjl~|PC`g*4ea{rIW56~;Z>A<sks7O!Y6!1XVZv}~haJ)h
z(yDdnrU@wFC|~TPNYdSxH#p<a9Rf0%TATcWD6Y`u!cj*6%HiGdpx%ha-UADq$Pt>1
z;O%R@Qp}dQ;|+*HzvVSn5})R>ip%kGzy@XuM9L$F#L-|7E^Q`sPoF&1wy(e@?6AFJ
zy|_;gvw3V*6GVA|he3Tn_boMX0+EDGN-up;HhE>)dd<s-mRB!bN)NrxzD|%Jizcni
zbW-CM5k&g>_71puE#RCE10gb24Soz1jCQ*MHq)Bo;pzx)uJ<P+<lwu>LXQr5a9l;~
z>Ij)uEaipV7iJClsX)N|APTnm16(jk&esSAN>AU=Sliar8O9RJwN-=0QPO+N5a5ID
zes_dJB2*tA^!6ZX2~_)+Y1Xnq9Ci05jB8vCQSOQuE)Fa?onu&xya-|$BZ%J><zG%L
zw<RlAP05;YGAFLEcml3UR<J22FRoiC34#W;wr)Z~(B$D6ccnNvk+pGEPH$Pvrf{jy
zQykSXVgH0#SL|$-OC+GHqe}Pju%X|uK<qwju-_M&$K7@x(l-`N#1y^oLVyBNL{)hy
zNWG2Oob#HQ*_tPqnI=&&>lhCdgAbbMCBCK}*7={dYzIBXzcImmcASq53b?-_Xuu_=
zF@6VNOKU@2C-Bs*n>R%atlXlLPU}f;ZT0Ejx9^Ai{^yTcy+5xV^&7*+BfMTZeER2D
zC#g{&3IMq-9$mn7z8-&Ej{=9`Tb#z67ivc!D(tddX#{y-kX*6DWx2u>i!tVxht=a#
zNxIc-)_zF12waO>t^;$!3QRJZ;9OS#%a1o!@w3@%26vmGpuY^}RPR|<0R*gf4J<Z?
zxS&HjWV!|U_(BuT_R>px814{450PK>SR{eS6<bklA;rl)V!I*2Fzdb8iRgYuJdFzx
z3hOoT&ocQ2uU91fgb%RMG9MfgBuX46MRHdwxSqrdwI3p133BpcveW6C$yJeB3ckC(
z2<Dd4GDZ1=7OY-Ab+4or5{VnA=QzglSf>HOP-Z<kobLRT!)f`{htw(#h0`yKG4=t;
zg?p$8fS(X?>!OmwbzLPqo|c}7rPG=MI}7!63lT>-C)MtU31qYeUplH&-O!zhCr=Rb
zF4s%P-Bw73<_ZTjLbJfQFl87CJIFJUdN4~c<7{XSj&C7Roejx2PeUJho3Ll_7P6$L
z<zdqyqnbd!I=QLX`7nB|=6+_R>bmBZD(T~5jdYlFh)j8P+6l*sckYxrZ2Tf9W&*U8
zX)^z202$8FRSwb6FVsFj#v|BkMaG#hvI~Jc5lefVpssW1&%3|za5MO0?<AbWX&+jN
zH$S<JPbg{l%--~P1j;)G#(lTSkFln>h$6HAaocb^5@ET(%^YQ2J}`9fqH@KcWX6tE
z?y=TRwT8^DC*g+j#)ot)=v`r}!<R3--60UK?Z}9GT?|WBAb+&w@kC}_eqLl(!}@3K
zn@SD(sf;0W1Rp1)%<qO~0`fp-?U_mtbPlF?xOq7z5@8=h4wSu3cH>1-2^S|70@`k~
zR>G0l0yyocQVM+CM&W0UOW6!RzGQxzr-jmgeHGUcHgqm>65c}XLnIFC7%7m;uB$%S
zmzWLj!3koLb|l90L8+sx%?T~Lx0aF;sOUPdi+l*6!JrRY(;(1+0Rsh(VuIlZ%8KvN
zHk7(c_nG6Ho7{9>Yia$r)ckVEE8i6dc-ra|J7|6pPB06ynSyD;Rpj6KHr8>Cc{bTu
zipD@Zx4Ct0%~NP*-5^I}Xzu&Dck46TTHxLzvyad1+Rp%V-aY*p;Lh0blfPyEU%kI~
z6yx9Wh8?^f@ActbUS!8Cf@KronawEEawd%%+xIVUj%PtTKzSQfWtGVG=OFN}lUTD%
z7A2TY@&&1htB1J%uPj**vT#Cox<dvpce0{{r;POyC$=@vbDo~}c4aSh2-y;rkZ|x^
zn+0g@G0Yd0AWJX{Dn!n7L-Il(Wlg<&$ItA|TZ;}?h0UAR(|^?fRKhJTf~yaLb*RD;
zaV2b;<ErlaUfpgaLsA<hu~?5HH7_z2oe9brlQ62VB^8IJaKDH{fq%=)#ix?)!tKP!
zmplbJWB<!~?(=CTAsLtR4s$>j5&N#whCk*Q*wO&hR?Zy4(x|~Q<KToj3nrN#QdF(F
zQWXajb85J;<=^TPznRmfz9(S^G{ps`dMXzVcihIGDZF7p8%(M(CI<CzbXZ+5BLyKE
z{4QdK&6G{3-PCu8M%Xi@W;A_$MYEr}`ma6vtLYbBeTAQ=&l-pS`+wrDJNee;zvVRw
z-!#!O7&SxM9k88Vv$1*f>WpHr(OR)Tes!})Xn&A{1}mNBGV#$Bc8ACRwYkPhjhU5b
z<Fnh8tlB@___wI_l~Ajz(y?prPd5HB-hU~)*N{?sySFjM+i!=rrCcahaC4)Lf?o>-
z!)cUV^>Jh4M%s{`ZQ9%SHr`9?hnPSs@%F~sX^9XcY9;jgk_ZPe7RZsUl~~?b28?e7
zEW1h!%d%pnmNu49>Z_rYSh<+O&<eb<@dgS+G)=G3*to#1$|V|K@Iam&{weR#$U~J^
zGtiRv6&)+Si^)=a7ZX+9MdR*<qgW>%NO?D#8;W2ZJIV8iWJY<gp}aW<TTA)5=Z=)l
z&Rdy~)vsk*^Y!%b;>xS9OmKfr@yZ5z4NRs;qmL^a-umd3R~EjrG98arkeY6Vw{;Cq
zal0u~n5+45Z93Uoe)l!l#P8u80&!hhckX)~tTh-iX&RRhYn=0z?suRyr=8|Lb4Fv`
z@M*eir$G>(*IdeVUMc3jGZ!-2az8`DcV=<ds|rqsM8?vnulG-va9IEjK#`Pfsdjrv
zcZD|M!Q`fG--Na66Tu@RpwH3n+H$<1&F~(9$zXufduwU;011^0g{ZcBt(%B@l0$c-
zXDu*ZFe|NCq;a|15z@#H>=hO~O^yEA1qDeOI8c<qLKE$8rm?M2S6k|AYV@1@0$J8w
zj_)#}P4pR~KT1=zbJgXcJyIHWcOMSnzA~9n6`RT<I*T|=lTKf?m)d!w#8nt^M}wo}
z+1&}eSe9JRftDYNW|Br;p~Ho9jMHtVni{o^hj5+XP}?PII2Gz3mpm`2g<lL7A?ENp
zC2SitEWbYEzk6mAY5-n2Zpy!O;(?lhoj0E;0jv}Su<I_l>zHM4AtSP}qZM-ucK$s5
zGL>BmOwx}8g6_6~;?&oK`$KH9TQ-`m;ufA6$(NA(aBp(AeFA=PV}f&9&McgL!cM1C
zj0)FHdDdvPB*k+J+R<xzxr&z)k|<v=GK-EoTq|YC_f;PJ(r3vRMa(a+A^oDCJ*7O1
zWMj;pxl-|a87Iw{`~>d`CtV35_=Q??StWybwlh_0g{(?J7*0jE%$JLCy-rG4orkzb
z0bG8?i_cJNqkD?w8_^tvDvk{#ovTE{wpy_dqV!gya_wtMh^5*^wuwI%wRNzU*N?@M
zBB)W|+3+JR6b4bx38k&HNZKx7r4K@@3$?qDCR|&*Ai-TH2vmP4&#-!$pn*3IFE+xB
zki#*)!z+e)Zhfl@3@?DGchZFi;28G|V9}&~j0yS3L^PBg12O<20M6f+!UJyS2LRWl
z2;8te<~2ePPkt28!~lqjDfN_B8Ena_zp{0Dk0*yThK%XA7X~sfDI0Y-6}EPdjuGD-
zXdI1Ov|%EBCexvlA09y`PjKr<Zx<5<3FcFan)Y$6{(7T5-etms3uEkLUvC_BPr65x
zJoY>K3h_!BUl>QJU%3Orcz5@XakIhe$1AUQR$l)}$hZ*v8(0vT(Q%xQ)MeWay}J|S
z*gMu^eq5aDuK~Rf9c4znMK!oO0KpQ+$WGOCAx)FXV$sPjTkMW_8R$)U<3WXuwaJLF
zE_iTn60L_-T588a{?JCIUU{*6DRp&C#pMaAuo&^8F3o7y>cVsAsJE-v7BH_}M9PKQ
zO{DRJ*4V|B3to$)xp+*s5!$?ukySaIOpqKPpxy3LV5a((siDvt9>TPhPG^t<N{U=T
z@pLJ*1Hitfw%l?heV(ZxHvpH*%j0r!hMbIg$kUJ63W^~|Kz|>Vcc4Z#D$&$~8Kmki
zVY~Xy@=lLVF9B2m%HDAGoye3I_{%6r`GQ1T_Y6RkLR0R_fwDxsr$E9a{RiEYCrHCY
z5h%``K~D^_rJz7!#xUiL7Z9+pxLl*XF;aOJX<<ewyt#m=LEM+-HYp;_o1K9wJMxal
zPKEdMR+Q8ntpF(S5`{q-zl_>D11sMX0z+?#@ZbW)4hw=jhevOj)@gQEFCc4C<}?mN
z1<&>m+eg!`T&A0HKM~+(uYOyys3Hc=uH~vV3d0yBZ+!~79zptbaCf-YPyelBWdpug
zUqF5F=bF@zy>NoNzA7sG2skb;=P1vWj?Z^-A6T<oJFqBG5UX;AnjB?M(ZX<JO+}Jc
zy<s;nBhhA!bl}-WJQcFN->YBsdml&$ayze8mUd+2x7mY8^KyUK0uA?tPCUi>?k=gC
z900eSE^ZX*Vf$CZKC-{x%|oioA=(2_R&(THbRRIe2t1Z*e463(2OxMQFk-GJ(BoaC
zqTD^0IU=HV6*_{sTzHZyOz@P$^+D(!7J2G3mKB9`6MZ5G3Ag1f&NJ5M#)FiU<9=lS
zA7U@}m`<}y@KSOga?uEpGfs5e>@-ut`+%?@KqVApGQh0o-HPB)sw6eBF=5Y#OzNr6
z;saHMf(yw5nHg%f)jfc%|L8iGa_DBBwQy%5t}~$Wjl1&v(-^B*X%(9L2fld9^B4pv
zmp%;xVW(RvE`#MPKPRI1POlCf^$3u902>uXjx|JWRL-yLVofOgi0uiNCO`@G!wH|H
z>2Nq0QS8(*6qmayd*dqJrvA&Z#B7pDW4cT2K|uKdZl;GAt%87YXOD2#<X#OrG%nJa
z9>H}3InT)<j4y(fOIwM7R)K?NjA{=Yw~p-cq(IvT(h6_z;XEGUHdG)E(NG1@^+=R_
z%>K}^+T+KMAIzM}N_L!BKppb8?69|ffXgT6Rb2F*?F>Rs%1nT|!l7Xg0dv#5;YtqK
zS!$mv!@p2ypgTrgvkg*1Q&@3d83oB;@dJ^M@U%P1lUiX0RzgQ5CdfRPAuEK#q;Inz
zG@caY#?uR#QF2|uvhkxa1!d^TS{cGj0B3XVVgSC#9&(F0s}GLj{k$p=BZTM0{@exK
z)%DN_%gI${d$%IWCUzD|#d$D`v1;PTo0}WNB~?BgmVrW$v&9kg^Tm}e&l~sV7t6j?
z1NQN8Z<%@hIu=jAW)zd@NX5Si*>g1$(&x!OKa7RiCUqgU<948&dV7#jT&s0zJnCUL
ziW4c?Zu+#}aCfT815NQBA-z44u%Oh4YTz#Bg?2aXKNOuJ;1;xiRKFQF%%J3MrF7(B
z{=N(n0S*gBczP2!L*SlZQ!Ghk<-nqG(Y3Mt@cso&nho<k_Tp+ZA2%O%Psfc5sb9F{
z(RAS->RX&`Zd)2EfV!}K>3-zcO<Wgn^>3|%ur(2+y6eZ?x)>4MOO0R^UK7m4mwK+H
znWy$a6M;-fPk99vgpjzUjVIXJXqXQqJ)2qbOYQD-JQ>ht4`6tS8)E6USb!%Mv_ACE
z$X>>l3S@vsJW?e(vaGB`EX)L{VZaOE^+qhn%fKAAmZl@h9S1Hk#1JL`hzZ7Yd$Qew
ziM9vdtI$#@Z8>~BV=NJ}=QV&r5_5v3aCnFsWQgnlH*pMh4E9J6N_oYc85H@TlOob^
zfEgr9j%qNH$E@3tkgf-4l;d&S9=hoWBkAhWrAu&?rMB)HTtI(RZ-*cEKyJ3u-2wIR
z(clrZ$43ZEZDYS?{3N{+Q<L1+4W%;z#^m1E5m5IFjWugOu_dbAIRwKf)|z)slE_Lh
z7w=zeN1zUHnKd*}(J|qEj+?Eq4XTah8ZID>ej#p2PE5lvFGTh44Qg^vb&Z;%V@{b2
z)IE>E3O%b2N!5h$mDI#E8tlq__HxU11Wr23faamQR-B`ItlKhtg;5TIE~st`wS&oI
zIKHy7axfL^<6e8Wi_q^CG%#K<;Wl0w!u}0QEE4couC^7D875;IrEzQY<oIa86>eTr
zk%)U?LIAJ~e!6qP`*1f+YY!L5Oc3fj-d(H9@K7+PTef<%za!sDu_%o?<_vMr4UoPR
z08tT7Db`O^U0CgSqS%V+1uC&LKbaO+iy&#W2x!0*2zxX7QU-Q}I3}eO1-9EU_`o9G
zt24~oeIguFd$(d~S9{nbn^8i%(ASEc%2%KV6B%(W^ewP9`p1iQS~w)a`Rq6kfwwH0
z!3Gv!*lv19-L;fj!L+f<B6&E%BYX7^IEMq1P<HuM_eqd<PYY)UBvC5J@=C?0XLKSP
zyGh{)_h3;m?<v0&c90;;4zkU}A{WMsnHlGlGPUF5plI^x>F`yzsCMK;T+GYJf>DMN
zAL^0(--loEq_+#Z^T8n8In6hs8R7K9V+J(FNE_d2>G5EBI&hnZ>-fw#x9-j-%bX(`
zL^#{rE;vsx`k$7=wBm)@@_Xt-vMnwq_@{848?j!XLP0|Tx154{dH~5X{DfCr-P84x
z?;(!_L~)1&T3LRuiU{}iP<>(GkV6cz)tgv`jv(3_^&0PXtHb-6H*_}YOGT+lu<iq`
z5J_WbJ^{uMM{LJcT&vLNSvT}_8z4wq0TT2A*FYF(JADRW>PI7lF-ljE<YmO;6P)Bs
zx=+j1t0=~_G`Aghi<?IC+K)Cjf08F+KmyzbT;+H38qT9m6XEWVM#k{tQ{zaEWVR3s
z*KsD?W*}!HVhyd$H>(Ij4J;F$U(UpL5?q;M_%vXzBy3KH5Yf1gH^MR%c&sHh_bj_)
zXt#cC{Rh`?-njn$dFZ#^Y5xGNboa94X?GYWhPTg6$1(-^q>6UzbNx}5`WX&5ciT)m
zACF4>s!T1`J|Fdq{VT&f_Xynn(XDIe9ROaTTS<Su{0x&n>fQ@d|K|FgyXT=i%V(($
z`{Q^niXY?LtUs9^7r4|tt`HFCV)AsQV^x06$K_FfUYNw2KR6eO+s8W<BtG7mi^MtE
z{4vUumqTvfw-+QgB>Lw1pPq-{5a%#gWJwNLM@vRn!Ly~vUD*Zkmz?NH?_QAXw@CN%
zQ2iFko+W3WczsrB9(V2qNq*<NB)?Nh^3i9R<Z<U-kmRKJ)_J&`6yM5nIVnE;ta5qW
zxfdk)-4Cvvi{dPwCHQ#1GatR3D)LLNp+x<8K@$Ju_6Il4!_+?+?Dw)HKE;dADoaOw
zdqHyFy}k9r8|&vG_}#(qpx6E&OY^b3`>awu>D>#`{k+R1Wp|e3ly098!Oyp8i4c21
zTHU(+;q4EgV4R0bZw*ccA7HV~(yWh{`*`^Y(M?qH*z!>j=JBhFE$8`2U+CWpk{_&k
z?fT7o=MoZNSEPfRgcCf{z<t0O^71V5pRdyes4w2>^RgWJT*+}jZXSj^caI~WFW%wJ
zAGhva{o(a%Ke~Cn_5Pjf*UvNG(D~yZ-^F24XL{6qf7A`fz;4v}7i2ho)EZkcoKkhW
zXE6b*eOmXk`L4d;KtZQ`{$WDpGY1K}+;a@kr$<rEcJ2iS1v2v1Ij2iMyEI{j^k4Rr
zNX|cK270q%m65p?uSOsDLKXEkHvm;RQ05EUU~eoO2c9GFiR2-;#>wM0QMk)Y`@y%#
zAMaeLa1mB$wI|uboG2`a<=m&-Cp$Uzt?}R$H|@GJG#mN!1dgY*cB<?P%29H3RP0(r
z0lioN2WF9v@S_n972!UQgS?6eZ)6wZ0ew^{MS=~Ww!rjQ<bB>9E^_|U!kI0C4kh%-
zP+~Yx;hls;<91+&eVK3s>f2TCk!~jizMY*=tXO??$hUPtk>`ii5Lk|%p&SMGaWMwH
zxIDg1MpMS<6r8xUa`|6Sw88AtIFZL$S==242MuwW{LvIKM`Lujjkt|6fbb*=tPBss
zx8k9k45aXKJm2%yb3%Dq{py#Tx;pDWc?w@@UaF>91v#U<ykY2K#?Oii%=NsJhy>4u
zSyOQG@GKw#NX16v<PsyWP`mA-p=u9wx;zj!M)>5`dXq)u4~GAhCLYgn=|b%Rr+pPJ
z4{F^eyzb~`@j?40z36<&;LaB#go)#(txZJD6=4E2zZ4UIhntrn;kBLe4+j;_VGbLI
zMz-55R7ukJ(c7c;X?L{zh*5~*ND%PDeG{O9qk0xPJfr=>A>L()2-0}<2cZZfr5M0g
z0jqs7Kr&T`1aWwjjL>@sGM*INbX~tW0x9hJFM1>j=w$mjV?+vTn>na30|{5@vQLzw
zL}<1=2-Onc;>A!;c`b`?S?GiCm1}sO?@JvU9}k<?LiqG;$3>!5gd{EHw%KKmrs*<y
z(WBY}RTrjQQO`GDQz$2%Vg>q`qh{eLeC0H0jdSEg3XGPxSw%NixY1CjP?rE`jm~C~
zo)kw0ZgofW)Fp!SZ=~>Pn+Lqdbjxcm{?T&u@eidla2QAO7P_}?>UA$WhX`*?1z1vm
ze#;3G(NzsbigZNnAswE@fl$$nYQ_B%oW!XuRR+20d&CJbDdQjbBB?vEgOxt@g~6R%
zaI1$b-=Q^l#1BuKE*X=%J|L2v3yATRy9LCvTH-+Hb*#R^9KU1wlJm+W28TnXl6VG&
zp_pHHjH1!8X?Jna7pt!Rd@XSW=VI7CDkLc_RC3G+C2b#b2~xuO7LUrZLdL@1j({i(
zeb76mx3?iU7^M>~xR-+F*0}FP*_-xUuHObu9M}zQp=-`d(RItaOaq;?KVi81EVM6R
z2R0=f)||`tn-!Cumzwu9aE+B|;oc;GL=37IHaVv}NUm0?#ezI?<?vVN5QRpsDPgGV
zSk%e><e*Wp-Hy0hX0zic&^I71I-I*R&W0`8UBm(Rd-Yq>;~ngE8J!@}=z0+WJa?Zk
zQ$o^+TxaA-iE&CNG!sEEM_W2w`NU!Aq((Al<zdRkr8+Ka7q&aNYNwbYsF{>!5sxJ4
z9=}E1YwFqGh#zNg_(jf*7`2o#?9b;DWab1RiQFe9<awqD>sL(^!gLoz*(`~hNhb*b
zQKJa~8O%t8q%ke%4lO3b>+O!2NC=Ny-up=RV0C^l<UNEU!#lqZM{!pi$)1j4O-0M<
zzsw6Wv~*b#KItx$qm=-mWJm&q_M}}%bcDWX0yTskV#*`Br#~K4MT;P%K4b6^)IDjz
zhn!a;OuN9I@mB`1eT0}V?gx%>!2-jFk_qYXw89Wwf(OD9OT}E+gcN8(4S|Skbb7~Y
z%kd?S1gNycAagLL*jOUdP`lHpBlqnTIl@li6Q4jA6B(|OLNXvf4l<4ys!lQBlMTpo
zH%8VD)?pM*;Ib&f4)`2Nu1+*23_=xlvV-;zp$M9vLT`|OnRO3pfG8f;Idly5n2-dy
z7?3JycZ#%>-SVO20x#@BGo}e-*h&TjJ3<CdQvH)_0HrByyo9zs<~)&`LU=jH>XrLY
zRs{0~NzUHIAMGb@H)DVGHjmZWHEcw6xDd`JBhX-d&B`%e6IjgeVgYSqy$AwX^?mc%
zwMW*fbn!_<C0AEn&uopOEs~mNJ}*tVICL0XTU1^aqjYSD3@bEJrdhJ4g4)S%q!>HG
z4xoYk%YL^}h87WFp%!vH;&O(O#@a-Q?gTJnG^oC#p>(-UsHsW`%QX?)2xc0!Q%b(C
zIiCRS9@PZ!?3$#pv|oN?$Q;rxx4@Qglx=U9b6$qoUe?BP&CT6Z0PM!!?d|f`XHhqO
z5h%{Nys!zmq=$~OLa}0KPf9&W8|vv+>Rdv-(pykZ^w3#hVhIF}LDXU#DwdRscnO3m
z_X)w{1uoF=vTLWXUH5ix(h#+d@ca|m2e?$G4I3x-JF}co_PXTQ7U70VbqNm32$C>%
zM^qZIK^o$>ux17bdx^ss-tOG=z{;~tit$=r*8<4tz9!UQL4JhRDH5oXKB%rJ*rL4r
z4LqZu^<PcdHi9hC#ew0`5YJ4~jw@r83^m5;A!gq8q+%Q3jSks{WmI_UwARHoVh9YX
zZS0FTkKh&#TjJF=GG!j!++4g;TNH#fCr(r0ELQJf7XRc?w}1JCAZ*f@)INGa7}Qo?
z48{jtBsY8^EI!z}clqY#50T?wlj9VK3EW)f<mSOFYX0!_c;~rs$UA_@aB8TioRonf
zgt2v)TlS`fnZt9aRx8U`kP0O@u#RN-Efse$w@a(WnC*$BUj#N8Alt>pf*_9}mRF1T
zZB=_149WK>iy_nhwcT{_SY>G-@GND6NGRY=6@a>8V^Q8NBIQH2S>*G8E@<Ksi{UBM
z>+UWzPR@61_(d5*bhd54l}dS`A@JK2wpg774_GvzG9fDi7T3%n5M?Z#Wq#Iep=GO6
zbp}iVDpWw5t7m6)%(@Z}jQhUl&M;_k({k0^!@_cS<O?_Dr<@KqDpm#8l9WsjMA1FT
zKvX6&={4xfqSKOl3T<JcPK3-Bk`i5ZNz9fWXGM0c#eb0?hA9hW-^QaZCe;-c999Kh
z2rlDjo#%?Tdo-2A#z_CfRWG`9K8A|vD0npYj|MxoK-ON*FIc#Fu9~XCOk=^NmGD%Y
z71p&(!eq-5O2r5)Ad5D3eNc=wcLPkW?M_c_2<C_~ZO;r25%kKu8>uS}H~G6zyMs;b
z;Fvd^i}LK(cU@LRLCLS2=jZu5-Q~S5$55(9o0Mwdkx)3AwS64c3*t>=-os2bonLw%
z3!DpJzgU6c)<};s3b*3MRB?BrubX}bYqxclDaKTCz<MEovj{$w(+yh0`?Q&hGS)<Z
z`rgrWd?1G?72~l`yDkSB(U|l4E)Pz}2+GY24j_~8jx9)tY7DFSyDT-v+W>d}g=NRR
z(X^Y24sOC|mQnX0lFRLS5{Brg%B*5o=bR|bevbsfbcR};z~fLmEp{!lw8@atZeW6c
zDdr7)NA{n>vNLn8q}p$;!L&r&9MY0gq`3lYN8{jAF<9!JC<Bpk+!?gxd1TIJ&kC$4
z1c^7G?pD7uham!FK)7NQ_rB>lz!`|9S~QoFlz!=ZY1P@wwV?aO)g&EipLv-oQ6@=-
z*t6^aU@iWU1;qmD(N#{j;m$Wk(SZ18#ls4yJg~82vD6z?;8g&5k7b0Ht#J7H_Nu-7
zac^?4g|uSWOiTVKrW%t3yzmi8k3o}Ny5_oZEu<tgo07BYtrdQsyYH|^RU6r8t*U-|
z?zWupFdm6{*iVqK$Hf7sX|?`kC6A=i67JM?n+*7y!|efZIu$Ysqlif1!sDb$sa@t9
z%cp^^Hn}BOrL3L@m!KeN8wxniHt{9x6>27+bygAw$^RxlkpN%*Hu9*g1>liTA$372
zl0Y{*GfZwQ>slJalC17l@VbQb#_ch@v;xbGpN04)d6R41_V^ScmgeuI6Zw4;u!%5{
zcG~;<IH1<mqhtsqR{Th%=5ZHkTg6=1Qm4RTF1k>Zpo_VsLmKVaNY%_?5xAZpaS<<V
z>qoGXPVGFLjuXrj$;Dqz<d=aWB$^=lhY2OnXJP0)73`R0G7Qpe24p&Q*nFlCb`GEz
zrivRfB~YehQ%bgI?=CE31g+QK3b7KL`Hr8#c-TC|HBjjqr0?|KT-PE(XQ8rYJ%{UX
za!{{0i_R2RBBVoQhg^;_NL0sJ1kRDOuT+~V*}GAz+)-HxWW>tlF|;#Z2aCThb!BUQ
zLZ56`w(By+xb}+IT8-<>qQsMd5;iz3DxIMhgn{E09D%t7tD)d|<BQTLMZ(n(ZnYBU
zsUqALZ}dWTh`O5)nhg%_&BDYwJM^nqi*R(wQ+8ylLK3Bibc3$bJ!09itYQ|{gH<XS
z5snQL%@a(uY*vi6>#p>qy{}XbL+f4<P%>iK=Jdexj8-`ZiY=4NWPHmlM=%ovGGGM4
zUOh^G9i#uWmU$6DFGLFD#drbn3Y>*n9$Sz~n8_bx46(JN9r<9olgd2Db5VhYzZlBH
zD`V#B+0liv&~AIwiE8`m?qrZg)MvbC^<Qt{qSashn^#}ixVUlgA3Qtk{@h)>y1aS%
zcX%16Y35jpTP-B^#IbtIF5~>`jlV{LZ$|}WJ#O{(T5#T@!<}7G`r8{{M#0|-1@YS6
zZT_=YwVl_)+Zl&22JnB{{3pDT^=qL{a3HcGPdENYl=z)cqM&<NA0BNyLa}d!VtdJY
zWu-pY_yDCMQg~Li0`G6UpW+iHwUxNOaUCVT5*l*nDTeRWji1@grN0U}g*@&5KYMQ*
zBWaSP33ZR|7PjHIo86h+`PiL#C-V9-vO6oXvSy~YCwHQ6YpbfeYooHOsH)wW-JBvi
zDl@V&DkC#uA~LHgw<x$1LU7XF0isVrz`+kBKExlS;|_$54~bt$=zbv%0TTQ`Li|TM
zkYM3?%*;Pue<LHadS>TfqFa>_@9S=EZf<UFZf<V&+0)--+R|V2HS9su>9`RoT*gpd
zE$E`WUqmS>5Vac<Tiz`@cv85=2Yh<^bfl?1zn16wL;jxOYzUtW=K4I@y;!c#+rIL-
z$v%ac*g|V-n6DjrZMuAkaemH_kW&;1SL8ViO^@9VpF`L^VD8RBK1O>4ej3t%>54p^
zBYmm8i=UG<fSRSJOtOI9gs!t*-NBKMzJ+wNMA@DF)00!E1TIhF!vp5QxrdBQx5QLF
znYpIsr;lx;^_BwhS&gxh95GQvbv@)cHY~ykjBnxk>B;1VIKVi_s&c_p>61gj|AIf$
z0>{TdR>P4qN8h~q&=4tPqP(f=+;!&~ujssXF3Ra>id_z!Rkg!B<-KRmk|u;<5dC+b
z*QOv);cFykh*t084{z$@Z9YwEY6?q{A!)KtF57Ca)Xb<jqJ$lE3}^Rr`_1*Q_`zY=
zb08ufMxv_YN|eT5?yFPl{&?`j(ybb$6d7=fnArRf-rm0c(XIZ&yS-bV^lyK1^VSdF
zLPjU#)*a0z$2%=~e)|t@-G9)(cm4kLJ9iKu@xivNh^_qrEF~@OA0E2&Kb9B;ajG9f
zOP14y@fcK|AJP2iw^#GGXb6W<dp4T#))fqXW29=`f4|K~$4|@>Zi|>c!`Xv{_WaT5
z{{H@+#`k>dEq{m5nRfa-+l3aXx5-4A>eFay5pr8fAL>|mg2Nt_!l+<#jvk+PL^8*@
zLiMyhh#AQh>e$8bb>82>l|;9dpN48b&lf5>tGUcOA#9sQIPeEb`RRU{{G`sU-#35;
zMK^LoME-LNvmH*!q!#r&@<HN(l3SMKU35qzdqb}vfPwsXuZBsH3<3&fPtxK650nf)
z!sad$9)f-+18JurB3+_glHe1v1_~-COyLHe$DD7s?%cg`{Z9Y>^-p@7UsqdeVZS7G
zLwfUi-5=b&e+PHtAuU*UGp-BmI7d*^5Q^*nlj&*Z>#wWG?&;n@(nM-m?P$PWj5nX$
zeRylD+MS+Wk%N&V1k|QX7c;523?lmw*1cGW`Xw7gLk`Dit%^(tb^SXQi`&;9JiK+k
zfBojo`zi|sE9892vZfX4+`D`Kq1M^_E;0b5$zX_!V&$Z{s9^PNdke>w6Ob8;$YD1d
zRfN@_%*F^JfaOHrh2IGwa>meo5(YLb>mB9-p=YM<KK9yyAuZ=qx`{=;8hElIJpu9|
zAPxB5&`!4;{gJRb-3<xu3iM#Kw>QM8F%wugH!aaEI?)^7f#HkY;q)X0rU!K`$tB1I
z2Qx*IIcoiikk=*I6^(t)h%W6hT~rq7xe+s##iI9QdU6ddAY}yZyoeQ2_$|VfC@4+C
z<SonuIthCHYcfl_Bv?dmIXXczix`#^&I==}s3s}JU4r5445n$?Dujd-fVTgl*rI_G
z-d_Zu(hsEDa_;?+%$q@g{<hOq+^rb74FppmvM%7H0n5C+P>|8(C`j~q+Bvq2hN|{W
z{j#<tW?1)lA{V#<?e0pVz&Z$RsX|P4Q=%d`_Rc^#HsoJxMft6bBQ>i9GZTE`HN^_J
zL@^6UxD|vT%szrH$Tblv9xy5MHWSqg)jI0vinS8sXrOT|sbs$#9mAbk98#RGI!t9$
zS}YBUsByb;)rx=w#7@hVrJ*Hw{D?K*HOI>_5vGwkl%nbpa24qk2l*76yX=*lv$W7@
zwXkfsB9Bk^(PwzqkZ}A&=hBd3aL{!nIUysKV_EkKK-}?T-6ap52gzVSQxxQsx~@)Z
zsXmUdS#mFp_4+CjiEPujryKP1R1`;er7eYgD&5WWjZ^y%pv59Hgt#4q5`k`}{01!i
zycB9j)7Y=90XLb)p3)*-LmP&)pWfjOjjqST=X#S4*Oo9Tx|>&{d-O<<1J!E4UFH_N
z7bb&~ZgS5as}@qJ!g4)3-PJ<)+FgAqBl0S8+v5s>Q<$(<*{Ya1I41{=FHas%lwY^t
zH!<tBU>qLxA$v`oj~FGMnfL5or>Bju+n?P1QjbRvBsL6CdT*+X@rrIJb00~_XUCA5
zek|SLgI1~6<!s?6CPrCI@!X7Sq$Z|>cY`4$!XXs}I2e^P@sry$(lciFaIYa@@2U*7
zo3OYiA-S&f@&@;1FJG;8bg2m3su(x_TShTbO^LZqRa4y)9YR7QNbj_a6Ex9W#}*bE
zfzsCDUL!f|E|6fU+-XsgI-7mG&2+tqByt|$B#6Q&=L4mp3J>NBdx4ukT+}#FLpV2b
z?x~>Eu0-6=2F(tnnSg++)WuaGhYaqy2s9xNE^O4Y`a~C_5?DJrtgDacy*JBRC3GEi
z-d#xrfE#Oh-z~!oGLMkzigygZ48ac-i}EAQ?W}UWT;++<VziRd5myMLeRXma9cNYK
z)qIu1p*`jeu!v{|$%=tmTriu+1)}o)@mw&31KVIW8=UF-8D6Et!A!18&RXwRtdJ;E
z(gtMp;7RrU+idS+SXn<{56m1RzEkO$u5~<?zM|E5*_(<77UxHr>PyCzzBC9$C~7m^
z>ac?7C2=bkfGdrCP=dKQY}F@|niQ)MFkuL=Px9L`!{nV!2a3IeP`P{aZWs$pIPQpe
z*UX%@5}$)I7En~oE14kU4_vJwh;crMI|}ndU>WIFuACw%15Ia1cuY3{<HC*-(4PP(
zjRhc2>%FU~Z(jLx^KPxc_(UwU<x1ux7S*jrkF+j3B8m-+l_YEPauV7LnvNro`>Wa(
zGI7gY&2G!<7p_437cN%d^5HJ6mO|TkmkyTD`>z&}UW1B{oNv$ImzwX@>vUy9LUNA0
zQ+7ia<|c~yb~Vd9>2ORrqX;7{(gX<NCZ3%l1K@}DY>=62T*LgkB8e@@*&+gjO^)Y7
z*m#-g4v&Q=Q+KqqzRVg^9K<I^s_|wIF4n`JZ4XxaCkjT}fGneAf>J0_S}<&GK<0ss
zG<Zl&O*-(+JVXq4YNqnAg2z4J_X`g>D`l9)KmEl~WK$|!9c68PYK#-ZDDxVtrQF~c
zrpU`I&X1A}7KM3xbA5ABcgxKdB-6R=#<Re0<FM21$er4q&BAm52a8;iKp%-5T&s}e
zebIW8l%|BL!`RunvRUSV%gh`<cupL0jyE^KQAh_4_0k9B%F+(SXGG81SZg5~6tE@`
z5wTgCU3PGyY6_o2<Hi;80#7;!DM}fb^korNgF<L-XMSU4erF5K$|M;-o&w7)AX1z+
z1-KbY)`%L}2z}9Hs8tXZNbGhCdIIxygT(#m^<))Dtk&mpd2~pI3v3!$oS*%c4d0Ji
zdYht5lUF}NtTS~!^IOBZB1nLlGS5etnMHChw8F9?Kc3wBh3mQiiwv{~qQr$BEhl4*
zW|z?16)a|5UDn9NDLhYF2{`VwAe+srPJ=38=dSh^oyzs=7ni1tgm<aYVF2HNUBtC$
zx>=^X?V!f_v2I?4l|0Q_Tx!4SrazmjQ4X3_OZ794iV2mdKr<AaPs6oC6mkR$8H`WP
z%E0R7nqlHnf;vW)LHXCs5HhwdMQjt!=3KEa@~1qnFF0n4#w#71rN$~6hB7^(0%eNS
zX<r6ZFIR@=2cQhbzpmHLBoksu@?Mu2eW@k~|4*b3hdCA7bf-XbNlpgK&0IZ4qn{;=
z3A=DVq{Z!9*o8lXjRG4*-6_qSA4_3P(Q^yg1?t+UE-!5xc-WF)0Xl@&Mz;T<G)WnV
zTx4b!J5kZQz6n^9cZ;B|M|M8Ppb?ht>rT?vR;rbinVH$vD-FAeh$+YtiJNB|N257{
z1L1-nJg4|~j-23#&oi6A%kju3EUzb4knFf8?$k*7C3(G*aU%Zm3Q7d-0h(bU-O_=?
zJ68}YrdHAQ6x9+M2Aymtz(!12IV_t4+#_Ck{G7*3I?xREvV)d%T6<7G*mC=XmwL<w
z>#HYCe17Aqn{hn^D>HMucQA4GU>6WgVpaxbuE9wk=n1)8KyWW$D2}GL&1v^i(NX1q
zhp@1xB6T`!FXm<vRzI$&L2%$8ncWNzgbAXb$=OYKuG<)_Z*m<wdL9<N@DM?+cz6TG
z56_gT>~J7Fu4^;36P(EY1Fys>JkNNr4`)j^j}KrSoy^YgFtS}in6+-;+~=b^+uQW0
zleo5`UFcc5h_=xQZm7`LG2Ow@+Tfa6uN-*V2M9?HX|t=3xIiNGr=3Hc8}ukwgIzKp
z$~w>Rh4c<!*8oTTb3Jm6qe~_CHupMkm#cU-8jklO-1TO?ds`P^>}emugdQL`JH3jz
z$a~G6jZAY%SJ`S~L#@7j#}9=rq>PG=q<{l|I^_6Q=&Sb`1B=83jDNZf+!Pm0aqc!T
za9kx?_{%GCRG^Gsjh(x-C-sWaEO*%O_>{r^F_FkKAYV$)aN3_7)>G=+@xGrO%^*vx
zl{H-FFvok<y9nBoVwrAL*scwe9(g5W7CKNvG*oKpTd8f=P4_ox4fIQXN76R5Ru~4_
z9{Z0mvuPMl=aOru=8EKAh|LVu9KkTDU^#CG*;uD7t0qm#?#cmlf7NBF`aQ<@ljGA=
zc)4P(T@ouaTBA0LFM(>Uo(CU#xU(gI3XxB8u@pLq{jK`MK7b{harF>Q<+fYIX6^9R
zM2tI-zCYZ@LFi!i!_l*?Yu7e6JDb0;`Mqn;u03iWoSaPOTN@jL*^AM$&SdsvWAJ#s
z5z5#w-%DdcN8zBaNMBE&|1$n7CBC?9;)Gom62IX4gLX&p8yc`ty>(IDLa<x<2MPGc
z^H#A#D|lN_JaOgA-V6~4e_{;_^tfGZ6Z(YfWv<w~RzFQ(|IB;pW;8bk01TppKWp^e
z&4Zl<c3B9)54bFfwE>4ySB6gztx*kCWFTS|I9ld_F70j^BF2;5BP_JbvA?3N_1d&)
zThV*p`Ahm<iVuzeS2n(&FJ?@U2rH2BH{N^S<Hk<P5BcY}ctdlvi>sXnlRduTN8X6;
zy^CBIyrjqdU3<=7@Osh>)k8T7Tu-T|_4#k4P?#qa@VMHk>j+xqWP*zhq2Oj`eXY9k
zLDE!M3?M2Yk>Ih3P%(m2)rOmMHJ|LCKsRClQo(JF`&VK<csv@z@PxBnwZy4Mk5D^8
z6J~_T^vYDj?NY;+!!k@1z>K^!Gx19FgNXH1HoQ_|`M0ceTMm=8TsVx2@PSws4>HZa
zG=aq|ssAwSPc&*3!&FJ1z?dfp93`iK>P?4JsClRQc=Y4|MG$!dA#@Rl7jboINb&yN
zywHBc<R(Zs0QDKW`J}<|1UG7u+VEwaLW3CLN@Vqe1B8%`vHyOz+8kbaPm+6qf;33O
z*0$H-@xF5{hHsp2LO9^OJ|_dTpxf-%?kR5O%d#6W0iqw`(iaX-$9r%c3Yctmt|HYD
z(iXV4k<N5qoy<I@Hr)PqJRkf+4)HzDduRw-x!DL|i%De-#5>q~3TJS7i$yt1oEpIT
zgE`NI6RLi|Nfd}?CGe>FKH0ZFtw{H%ghvS5#=`J~NyP?n=MzWvoHCfKA9iRVKbV}@
z#E&B)Z(`R6cPowTH=K8kjp;Yl-dtPj2;)Z1*Y*9GO^^*eWi#utN0!JCVeJt+lp@Oa
z2pY|pLkegen>}|{%@>9<)~exr7fK@LbLMGJ5d5&WG_oh)Iwngb{Bz;#F7TF6OAZSs
zuqbx$0-;C{Vi-V+r*&-7{84H^Fm%%6gbb_zQm{^EaD3KuTHqO%Np`GD`u82FTZ2is
zrAqZ@*b}>1;edD8NjcYCuMUU9DZiFYoNOjxE1fOKS~Ho$UHTBYS44z2>=`^mcIL3~
zGK&SBOK_x~jTiM94!fSULX_ti*WuTBlDP7d;Okgh7ndf6V6}wAW#~&l_E5|&7OKdb
z@K3UI%VSxxB~FAeV7mC|gtagd;<^l`NBnQBvb-@H<`*Q;0|A1Nj*}zgoYB>krdu+U
zUrarsvB*;Rdr{;$9d8Meay&VLOOW7}VAG{2G0c3xIm$)lsX$mG$;(yOjMzR91l9e<
zHjQB_*DGIwD5C)XbP#3+3!vj$+0u2hnviR&3}G2ndlg77-^z4nAaRfp?%}xDAZV$z
zY^zG|zw*MIc?^`sC-W|sAxehkp0)!9w_thFR`mqCOO4;_!3U)x9w4>00Iq1a!5$Vn
zEclTnR~J9UQ?oEm#X5YLnXWkQp~p?SVFXPXKuq3XvA&%F=_RTqs+*}vr}ToOzNkgq
zt?*}zlyFP5!d^Xw&3YbteHiO11{p_j-tsgUfsmk^%o%o7*b<AD-8edTMEt1dOm%Ke
z;e$Gto?9<cCW77$1^;|5NJrg?qVJYP(QBAPss|RO>00NS?@l}y=15=3fC`umo_@Yy
z;7h!Y%<sNHW?@cymD!Cc9RNl)mratwtiu5nSZWnN6_x=Z@D<z$4UUG)gsKMJ0STx)
zdgNhBG1sgxz2D2RGL}?s=;X{Y8qIF>JzX_TR(gnGsX=7A$>UU}XIfLh-W^XFw5*9R
zB)V{5)o<?vLRh68FW+?qc!Xouf}52mhFdsV4iXN<+*jPi4anhn_>oI~&QpJY5bXRM
z)gTti^T}fpZPII#&KDAHl>0mZZZs;%eDjr_3=mx0&!$i@mAS5fzqS-{YCJc7LKlS6
z#js|8Bh%(7=&;Wp?K5gUxO?u7{V|=edz0-e4N9@uk1+W0go#Lw{pE!*z`++M(z6k^
z5*3avW|J3iGvYpO?vArlo2PGBh7nC-=NA@ZqV^Drx4x7!koKrEgy3Xv038>^*3>qE
z-?~bBh)KU>zyTh&(CXkJj$EoOB^+!a$Uf&*%o6#;I)W3kA?}Fdo+@_r<#tnsYC-^|
zMn|Vd+$JuCNJ>L?)O~`9@cubFaa~;-2yU8>P`pNljLz{)usya?`vbVJd#YetCKBDt
zgk^jzW!!^TmpSLl)1nLIQr?o6^pboXTrz*h`hMpE{*9lTUn8tGesp9v6wY3PN+rFf
zYi_11`I?ULYJbm>?X~jmK_r240Tkr4D_r~AAo&*<NC^Tg{2IDbAR}QFY@U2vuzT{c
zr!Pc<F5+q{PUstv1jQ1WU$~1dgqPdZD9q4KJB1qA@mww}B8(GE8c~de04u{yORoTt
zcOsS1yVoaRHEa&NT_2x5Y45jCnX84Tjb?KT@Lpxn6%ts8!TRvlJ-Fi|7AtR<R)6<z
zR!?BWonu9M0b59?gXQA|9?Fh=`^QyV$g8WzTDG?spCzzuPox!0MQz%Qc<jwhLEPJA
zY`W2S*3?xIovEIeQAb;cGKy%OpiGPtR6WxPS0Cj>S^x!xKi+!U1GES=42Rr>YH9$w
zu;kR^aJx@Jt>lLd%eAfQ{^@bGJ+0dDBDRLX_7R@4dT#ykO6a_TgWhm|gbS@4&==Rj
zpr)Q~wu-s^J5O>mkDaiS?>2SQFElh+6byGMT~{@k9Jp8E`2j-KJLVEp{(qG@zb8FT
zei+%ljmz0Qv%u9PJ%|XJ0^P-@Fz$LLPq#JxH8y7AyYQ%brxGSh+TjLm#6eIOAmzef
zOmX9@>|q-VHQiu>W0R9oFKi1|AMK1D)pj6B(}Z1u%Wcx~kXd>)9ex59P1srKMFC0o
zc8a$R4)WwBa_imEmCf}nY?q={CE4-qz*VY*`e-kcuSA76X&DPybXO&zp0N>iwG^IJ
zo7I(S+J3RN@nif`Ba@i<X*%aom0E;D2YCr8{Msnt#g^)WO60`7rPf|uQDIqr99v;@
zF-AoXl>}NkOIneihl**KZ?+^D4OtnJhBkB8#@efmm%@+~Y3X8X)8jj+D+ffn%ZZPV
zRLUdG0fKuY-(r@qkxvo209MS<TAe`7$muupv;}zc!0WM4LM2jG$1Q5|oT=qX7MNcT
z&FmZk$m{Bj%|=_lMF$HG3T(RWD>yO5rOuBwMbmaRaA}i=b=K~3#~)=*Gx$r)+jAyr
znQgv{#YlEZm0e(;JEcI+v%R6n9#50O(zzxvt#m1~P>!Y-=nNMF{@P*zsHYZg<@V!*
zqiHedq~q_3L9lMEZ>8cYnKTqz+|BlD7zX=K(jT+?q}ccpVcyYr=ho!C$RrKK!(l5C
zn+=|;&W1Pmt!ph!Xn~QQAB>=(HA4J;W&}^CW-3!wyZlt244=z>9QTXy$=2DlM!1g%
zfg|eLJXe}z5Fwi^I66AQy9#zxUf3n6t~Xy|qr$14epEM<e=k*ECu?29^h207xA0ff
zf|tNbsi{^*(2Bl}s+wlh%e9`x^&_QJ`sCEn45(<K*)TWq6PzH)hy@T3b`2~_Iw^B-
z^tq)7{NTw^|FVJ=2M4c%Gu7}LCd}sdRMOC#CB3wTwfc52$?<9l&3S6{LIR;vM7j0w
z>?OkVox<lER;z2*f+@<TbV5Jv3p5uI9)6C#n_8-(IjS>XECl)W0t)8HCfaPpXg!ds
z<PH^QVzi62fZ*a;yNy|$4Nti6tB2@AM0N^=n_WX|aZ}TZKaZXI|Do8am-WB&vsM%g
zoqyfT=&g5h<(<ZC4b#D!GB0_&Au};MxkzWB*sBpW?b>&AMTq8FY;gJ6Ie)-a5PHKw
zV4AaRP#dJTut>eOfj>w_ie^8CDdo^m(bSJp7qYag=nTCuLs>0|T2seN+FjU?etC1b
z?#C7HKY4@srLoCO=BcjWH8-cYp_y296JG0ML1*yIt}fpwQa3)DZY?GZC7I{kFsZ!;
z01&BWR&?#~9LahqS1!Dn%v+zh=p#D&XwUMw$xnC+&N@T*fVr0ei=axcNX&5*bXHXD
zn>RpZBr?Wywz1UKH`+_IbYE7#_~_|E=Ji%&jWvZs%LyUmA1V*DB_hWBaM#>4_&btk
z$WG}}@In{jOHW7DC#ik;Oaq<S>2Z6fH9dnX_Aw(wU)jaKtcA_Z?qJI6e0{}2T~}bj
zdEu;g9`FjhP9Y`4XrPVwSD1hL1f6B14uY1TOz9THf&@P}l^o&wr%WYm{eaIbwK_w7
zG~bHIM9$Ev*2w&157>smWC5dPQo`d`-!I4UiqC5cajX}#<r^Uu{)hike(GO`FOn<f
zkN*a8!+e{$V!qXDAy>?w{x@9im_OfJ#qW?yX05k|-~Y4ECG+m~b>?gNdeHXV{YpPa
zH4V!4m)_K(Uy6!mWqSYQ_Ae3IuSG>_A&V`m%1-YLCBGP#l+e+Zl;v*kE?$2*dL2Sy
zTalZ+n^}<%Y21q3=-tSQ1RcYQ{Kej1M3JvW4UOEgs)wgm?t|V3C>M90_iaR5u@c>0
z7bSi<g5=`WTbIAz`ypt5k7M)K-hAVY?VF$|6ZxzbRrJnc%KchHr8*I-a>)jIHk584
zZA0k>D!*OEWD&u?S*}v1{a+{-hBA~~*fxg#tuH9~8=CM~S@Hk$&FwBn>q@zih7&5Q
z{O`TF{hx4Vw9A$29#~fIUw?D^-<BCwt|vQXm#^uwH@E*4mR)Hm8{1P2{P4}~f00#U
zQ*p$<YV6Q!MatycKYA1YqecU%BlAc!HMalYP5h5q-zcM&&XesQzPUX>>8IcNH{DFR
z@rCVwfwcc-y1zV6R3XhtypKI+MtwOiU7t!{&5AV4p}bhb<jIRQ%$jdxje5o`$fvJl
zWy|v<FU56Nv<72~f5yn2m-8)Kx!?c7_Fv|(d?{;HI${j-gD-6VfMtF)E0c}+LoIfS
z%x8b}%Z1z0pFRCW=G^)fA3iDxer#`jel98dR^ViN-yprE^uiTDO2ms%JXE!p@o?8(
z#w;T*OG}`<IA;}=GI5Wh(UEtH*K{fJ<>GuaYArp*1=La^L`p6vpQLERBIm@Ol(~RN
zmGT#r-%{!PHA0FtkgUD{N$Gzqof3{o)2Dii)ZQ(akODD(A5Cn&KYjWyYbv+5YpL9D
z>@7Fb`g_w!e=r_BIhL$$|J2GGZ&=nW$+IR0eH1EUCxo+yz1(Lo44+2r$*#*C7H3Q2
z?h>|1&9}u2n&Z)9SN|S<>I+0bq5;$3(z9k=u7j*s*N<W5y*s6!BW@mA=^yXG4x;J6
zPVha=NhY&Y@_=z;g5w4CW@i9iNW8JD`0<Ai@%p}hyUmwLTCe~}w{XiR6eF#WZDQWB
zFRsm-`aEgH<u~hK44z$cqu^?6BLM&G?7?7vc=O)fXV<Rl7?lCcqVP0uIZ*oCA0pY;
zo^&oq#&Z2<<8cW>K;if9+}Umb!i!I?N!{K{lh@rPee=gR0_2Ya(C|h!JGUK&qW9MK
ztVB$f=v6tV+R+hiJsM$TQwOJ&WAOa4*~QFIUcgHCuH1~fB`z2_XaWsp1>q`Dlk#{6
zp$yr^iqt6PH-R=^wOj$Eh#K5i@&nhNDDFXX`!ivhrHZ<gsET}KbqI!E!4Niv&~x7h
zDdc;LGX8s7B<?Fp#Xr-6X%A5}eUKOC+=#j<<+EpL6)fb35^Kywp`r}4;F3BwrMDe5
zkl=<;WDW3;<O_L4`q9KFBn5<sxEq?cp)rH{;zYFOsT@6}%^h5i;^O+jB;k<|w<Oeo
z@d~%1aY_$C8c-2#k&s0)L@~i7@fBlgvqjd{W4O>h936Zsgu}b9-InLj3RAYaPGZ5X
zloZ5;?~`uJ0}27CqOKAkEJ&{Hg5iiIVy^+mei=h1E!S;6O;FOH>lw^IW>)8?hDaek
z1h@{mf-+IcUmN>@8vU29&#zh=EtyN1*^BKZy%OwnML{x}<0p(P>bN8wYF72HBxiKD
z*z~)buPmVpYq(FXlik{rAlqD}nvBtiP_@UUnz0mf)#7LMIyM=)@ub*s1bvj2*HIKo
z%0D92Wd<yuB557Kq_oTqqH4tXTm*3{=$b(q1$;^xMLL>c?tCPf^SFHoYpKMBa~>+t
z$Wn_fa;*)qm?HHFO|w#Vp73M5<hJ)$Z!0f88;V2__7(vq{P$|B`ltpib4HL!fZ3DY
zrKxjvSa_CUOeVu-bUpOr$ys;x7_LgINjK72u9FP(Nru&L%Nnq0iourgY2YX~ZNkFR
zZ4w$O^R1;YQsk)*3w3-4{g);-ppz$LNf>%MUvkYfFAIC`;Sli)4*dNBylm)WF(DjK
zQ05y*FY5uq#%1e#N+x1mVuX6B{GdwsA(=Z_i4;US9%UXf7b!7*xYaCOhm37xzab?q
z1=ej>oNlumgKe}HVx28g3(JNcb{@l?atL`Fl1wHvO@_};wtlFU@Zb53GQD94J$iom
z0+GESDC5?^jR_;BtU(@h^5A@o(1UvQ&tF=^o%1yIUz&u5w=V>-!QQQi<d>YKv=Q);
z`&>|DP^WZoGEu>{<yGJ=4E2TPPXWYb=SoaJxNw>;ht2S8u6T|z<;_#EftOUWC9E=F
zfVG6MByg%LfCbN2o)mV05OwGmW;(CV)?uY<SvUQe7FUrceU*PiZChepCY1vuSq+<S
z$Cb4w(lqosQuiXYN)Ma+ySUXlj0LQ0AVcM~jg}OFj-L!<&MF-m*&=@N?`L4{CX-Fd
zH<4N-Uu|nrXBl?d#$WfOVVwztBMkx)T3i_znd6Z<x1l#|Ek%of_*519Sn8*Zv0RO)
z;AMKqMr2MaMcPnUB2foK3&=;|NJ1e$iHrs41A$sE)6{|pQb(&J^U2Y7BIJ49BCdLq
zpNRA0AvTc4x&j*l+Nrb2!dhYPys#Jz6rH0({M*JTF$9?TbPQjh2mQ%mb}Nfa*V#Kg
znx;b*ktEyFPY`0aO+DgVG&Wk=Ut??#?>Q(aZdO<ux~r!r`&ZsyjmD28gzqlhQ3mtf
z(I`(!%J&SegZMq><Hsk30ZpVuSw4sEHb{$t`WYDS@<Jo<Bv|0SsI1$xQ{Q$A$1GsR
z-u-l^beUE%_3I|FXyn2eESPzIcKCjk_Dn`vx33R0XDzJy$8$PO`eQ=`>zR&`DL}bA
zX@XihE#M_KLY@%h6~JTxCo^~+?%(*4sf#IOx@#_;VpxRO7jvE^jxLD7&JwlKW<0W*
za#Ymn0^SYdDalBsPJkMQ&=awt_Bm2)VXYv=78bjx6az|~`QRC1zCkdZ3po=^+H>i(
z{NvP&W*w~{GGvZ4+h|gSthpSy-$32emHE0qaPi|3bc;wa;q*wAspw^-wb=3t=@WE{
z_{e3k_z~O!)38yGvP{`Rfq21J9)-3<1;T5;jK23;T0@|SznVOnrD)davsm{`Yp3Ja
zs9%RkWS#EF-zZ>Z#%34Lq*A6My|${`&fta-ysoDe1%9y!8x+-;GZ?2YEIz(=*w%Sb
zqk{y-Q6;dyKN|DApgwX(BgD#;=Uu)`DQRlM?s(oUG#knWQf3IqM7QmgDmQ*&<R!J^
zb5YLgdZe%BznB+k1||%dd=v4Y<JKT)Qi@z|+$0D`fj~b)<8o=THP8FY(TY<$NPvjB
z>vKMOsbS%wGx!4U$D-?VWc*m(`c_M<AGaT!KmwJ}Uq1{<VfcKD)g49mTQA9xbOTpI
zydFG7mie$Mdtf#dVwmzS{3G+AECkYF4B{YkF-QS@7A;WW6@b$n1zmW#-#H!}A+s$a
zFEG{#BB}C?+k-%uASlLu#WXCVldvb-ozOoZ2D60|4Fo2KmJnf8?Q2{tfn0AMfDZ=q
zLZ#4<Q7hDKzw=Jg_}XWG{flqFSr{L<#zl_YU+aAhE|P!i-!T`--mmxm7+!YYDhT#Y
zKPI-i?)Cn@6o0o^yu8R-#s9VUzoVV^7F1jy+ziJ5(EDG`fnfyEO^g?-vNRZs|GM|z
z0Y<xs2Loj4A*YuZR_j0N{a2{fX{uF5-Rl00-ha5LZW(2(`<HtED(aTWEmOrIJaVV9
zR`GW46)JwUgs{}JVjH~=Q7oLShh@E!o+F#D`g(62<=-xWO=!URZ&lvyz1!48s^nPt
zcY5!jd`Z(#-NrNkD=qgD{^-|B2zfoxine=g0RBd)XxQ{x*;QVDz<q?ex`SE4Ip6Oa
z1;0Uxi2JGcBE3hC(_b#sg%dwtor{8$I<HSsXbCb&x#9@t<d>4mj%_p8oN=V&B2JV7
z7vpiM(l{NPrqtqE2?KlCVPkdyfg!Cd5_DzsOh}zuyNqUByG+U^=`0N{BmY_O^ywd~
zC+M%`o}iKq#a8dn+CO~p^o=)GzO`{WpKUxI9d8VepH+Gv?5!2v1fxAp%yDn=sw|^O
zUf113r5DNPaIqVYu;^;09)ub}Fl>oZ38mr~wbk*7p|v{#!a0sQXHj#}gB@gghV%o~
z@zj-=cz4tJng4n+IvT1=;|Np!1n&Rd7ZLg5M#pct+7g>3`4QVv*b+@EzJN0y4QF!&
zPoYG=PUl0YyI(qKakqqyA$2EBKX#YV_T;Lw*bfFdUw_No6TN{7Eu=7^E)8hM7BRSs
zEE)sgu7I4wsQzN(b41@M!t9(3=gddiN4|$ZlN)-8?+!e3>G~+%y~Yrf{0<RA*lV0g
zLE=S9xMQ%a5=nVy`$OdFFepi7*MrVuz{ovbe;dz?oywJ@F9_3w98Pkqhio9u*A>qm
zB2vP|Jv#5BrZw7*5HLN{T?nrQK8ULQ>_<J|_D1w^;VBtu?*!xrxfTi+8E#HM(WP>3
z8B<u6npmcYfWt-Zh+wbPq`M3+B1`3nbj^C@{KZ4Kj(d~2<a|ba50ufwVft3;9l{v|
zoWJEg%|p%J@C<9>UR+J8IHu5zjcOHk|28Njh+*bTbd3Ha4)#Ke_39nAh(GUW+tHRr
z3?=~vHZ<XfG|iEeLLNBE`0r_vxKWgfe}=2~CU-d>N4psD{n3-uuty{M0KS=Tw{g@$
zal1B0SeLik<KeU6n0qbz9eq85$R9kxi_BpVcn<fE2<vF$?Y)h+f4_SB<E^*1x88n`
z6+A#7wlM@_cgLakK(6$Ctk2t60B{vdF84+bd{M$2A<^wmK18?QcOOq2!o-bk$gn%?
zL)*21@^>Q`wsE#)dkSaYjEHWZ9hLBHU@KUtj|@Z%jK=h}!92quoy<GThJ?eoc^lu^
z7?!6QX=;g`VaF1$CBA}}ZNniQVZpOvI7V1Qgo?&R`LoBvD^eIc;c>YA&4R`BifCQa
zYL?pd-D7pjm3a3-L!!hf!WLVEFqtTvC-5DHTaOdo@tncEZr;)7rF?v)B<JGAi4si@
zH}Tl~L!@}E_D_$&lkn_60rgZ+$%K)_SsoylB;Fd=U>V>cd~>+S3tLRW%#Ry#D@|8K
zm@P8t9FaClXH>+$oAgGPZgzkldF76%x;J%YjWTHBh#5!92noMawea7g>Q!`>MgT+P
z(ppqMFYjrhNn3^7$*77faau=rfiA+<`GNHU*+hnYBygLJcIRC(ciKr@b~Qb)kEw=m
zoqfoZhVsFf>ntU1Q5P(Qk(lVw^*M|~YD&T{1ciignf<{`4;aZ<g`b`v;ZYE|;hB&{
z#I~naVOS$pU0n^HqbaMB95`@SmNfLlUIk%<Ov}_`#o`vYF^0v^QdWufuDqW?utNeR
z7<T9nLt>_q2Sa^&q~<=tqxV8PZUZ@QS*i7?vp*b2KIbq*b=Udxp@CmVz#ntA2$JmH
zkDvf0odUGlxo^i{x(I8ckvu_|jX;N$3Z@}s*f^oQcs?4B8P01u+C7A>4w+-=<@<LA
ze|WY%**itxJj8Ke+}7jj<H4Ni>cFk2ud4?BXx911Lp-*O7Zhnu5(V~{9FpC!(x4BZ
zp&mzLQR}D{>sB4c&((5mk~$PK&Q$BG)WH$%d<e%M-v+0HV_=U#d-y^^G{=II6IC(~
zCK~_*taLacMDrY5xRf_oOII;=!HH)Ne<vTTD7+b~V#7FQ&?1Nj437m;J32*ZfB9!T
zL0aq?#$|9^H8^i_%pbB{K#{acK3Xy3$}tQL#x#ZlDKE*$cgI|u>%9pb?qD_L5)%Xn
zQFfo!KaqW^d*dO9UqOwNVaIA>0II>`3H;cvD1#9h>$2mu-GkF(EbVv-Zkrz=5GZ>6
z7-BFO(_ORt+X8<IYdszUJw&fTjBR05u3#dXI`xWSP7;KPP93_yEy4gYo88h1rGst}
zx=Dx8827N-dS%DSrK7>rjl53UNW`%zj1V;%7MI1^Zs?<dBJpK`_d=NhCzdrpC-W|1
z6Z|;l^GqHjvlh^Q#a&^zS8(xgrRZP5{)QvyPFS8u78al(3=_57h9w<W2{iA<lBirk
zmMaMts^_ZI!@4Qaz(BL-Ge~}QUMs`Vfw^yH6i1O+1?|U1E7PLc=aez(0-I$Vs6WYZ
zxe!$&2?+iRyceVh<w9M{j^0P|%zPgiQ%Z%37q27^8i=ZWP%&&=owZefpCeOfE!#bq
z%$X+52~$L0nof~6kjna}s(1uMKBaIxDabh7s}mbh^=ZKEQ%#IomksPii;@3{dZ)U*
zU!6@(EwUxk9|9TXMFj#TNfV%_od4CQ9^o3bJ{54ggX6}Q!Q3;Xu2BK$Fc)F;>6I&?
z1J#QcFFrk2wy`%+oo@Br&5&BF&fiV^VL`pA|JHhRx#8V!aeO1xOmeqhmKgLEjv=vh
zL8Ej_pon*iqi2g7mH`mWq<)#a`>poN=COIT)~SA%Cu0#_-9_ckC4baE*d|8D?NaBW
zy<CmB(cE3Bwz2%k{AJc58j6Xk)1;L7Wfj$+%`mXyei_zJ2|H#{r6;E|C;nv9=kKA(
zvMZM7fOAl(sX0BFO-`rcvpe@XO}4zPR4|DkUUb%UfJ1hSu-ija6;-~lw7opsQhi!1
z(d~EKtDVC~YkH1+2&Dk~XsyvX9G=bFYvHVzuumo@*x>n?d-&2WHn$^+dj@I^if~OQ
z`vLMT&nB4FU@Xx_pe|4i*yG0gITdMYYRGZQz<IRyQv`Z0bn=S&3?vgg6^Da8O+p2J
zCg61^4-zp7T@b;`k$n@FdPEb)76>8YCM-OY81hjDVs@O3#(hWH;V_Q%8Jd*sA`M1d
zcVi=}yM{~+0aAwK0~}tQFt1nO9uvpukgLBC$2b8%#4+eQCp#mC%B<dvT6!dVyeJz4
zvK+)%A{n&5_q6t?29~nm5H-j#MGojdeK-RDBsEiJQml7c<H2J{$`*4BxJMe>5!qC*
zE93o`4324(S~_TyD_CSPC1sUTfgWpMp8{QOx~N0i{Kl?#bN%Yot2TpACfNO8i^W{J
z@pMGJ+2lFa#ODJX0tE*3eOu8DGt-1&VMDk=tY0#nWx_{IbgJLSVu#hq!sjcl!L7AI
z>yX;mp$w{nC_6c1+mi!u0_KQh9gkWD(*~6>u*FJL<RKLODNbf8uAU?Txm{)#7h1_g
z8w6gbdN3pnF)t@b$c&RQ(X_Q=C=KetwsJ7plRis&bR2SuRW`YYyHXvToJ{9i8yg3w
z5_n}F-p&}z4S=0*n2I>xm=1OiVVs3M(Jmlu$R@zzyL1jtj>aplaOb93q<};i^^5Ey
zWKbwXxL1V9!v?&S$ypV}(Xb#fFxbSC)FeV!Y;ag7j@FS3uM33|)Es20h{G!wk>h}G
z9T)<nII>{gxHUKE%*~sZ2*<K!A5WSECDeC2?^L<M>+f?u*KNyw>mx3<fe|RhxG$FZ
z`Z7qqQjCFAq}6piGRHqSqocj?Y{+#kwstp|=}^oJakAeA`RW{4=UD)Ty!pMs_;e^n
z`gZHe6=w#7RQ>Vk2ofEVPcU$5RCKe9l$Cdm^}+cOrM+#?_l6aKYM7^7Jr#n+@WK&u
z(-{cs){Co_NxdgQ1V<nX8c8{ZfRoCczofbA6^?>vEr#6_7O8d(hqXv&jjhtIdVI(s
zvql$FlNNT!VZXTM+c8HZ0^{R`J1{U}k*Cqj>Qs`QlS?QPb8a9ZFAny@rr$Dk8V}h4
zEW3OiXAdb$7(Ar)scx!Zubug_B0mqg99bDGcvuJ7Gi2-y%fNk2g;{SxO3teGr<xB(
z&7A}$%L)58oP8SgURw-pT79?sF*L$T8&gZ1K3$186`4JQ9^;uaww7wPia1m<u}~h}
z@qw^aNAe?v+dT}hW_7{+Je$cBYy+_>Be+TxD^#~|(1lh~hBYSBauNQtm+kSSa7xRq
z*wUai4o+~bbWiUt1vpBWc}IQk+vCB}<Gn%k0y`2+r58yT*Q)P#t83qVFD4bvS7Db{
zEu)PPZNu)Wj3zEvdwLq02m>dNNZ$S~%%sP%r!u}LjiRxJaFDsor-cAMZhGOhZw6b8
z7rbt{^FkCLjaY<7ojH~Dc57v2<>S*M*dN5R_eD^h9T{04J0Y4TN7MG2{cWwFLWDUq
zid}|;QmQv%k7`uG2{Yw<Yb{CIB}q|3f4^TLQQr!TH$0xLhM3J~#&W*1<iZjpdnUQ&
zBBYh$!2nJcA((k&LQT)a#Jm*L1z{XbB@`YG#?ymA_iE=_c*FxK#Cqyp4Ua;=%lN4s
zctd@%4QqGvy->%!^=7v2cis<g#ceOXq87cgEj?sjUmYg8FJFb4ZYIpKlSh&8P&mXa
zoxO`)@v>KjW^~#f6^Ix)7D8K4?*~UZ6P2LMlRycsGCPFAiEFKJw=L!kjH2lOG9q5=
zJ{t|3m<}hxo>As-MuZm>{4pqE@TbFnVsVFsVs?h9cs!rL#SGhL3JOJUN5vCHbsi-+
zlM3aMQ7H;m(L@1~tgW>Yu^^5LIGN_8eWWS+5IF{i7+6%4ux@jslP%CkS58hY;t}P%
zhn}KQh4ir}0~d+WdKLGT({l*@SHck~dX5Uw_;I31EdbOmAnBe5%eN6OE@8YB3^G<q
z?PGbyMt^gydK*=+r;4~(BwPg)+oN|k`NI=0!-SB)=A1Xc=Xffq2q#DK2j@o47b8rq
zb?07U<_VmIrFa=cJ4?2Sv(sYnkVjHX0M>5Ocj*~9n8V;_nC119;qeJPjlsIDDX+bo
z%}DQ#cTOg%wYh1CEDFy?yTdxBx~0T9>p|yu(tk1=?6nINB33#-iI$#)IeQ=OQOJOa
zUsX+F)LV4*kQ>6d&36;NP$E&|M6CcjKO9Z_*w*hd0G<@qOJ&Qz-l<VH8XtHm5Qj7Y
zgrWtW#BF#iRL18i)+wi_@8)&1QuZbS(sl91&}k`??#JL}IrOf6j6r4_e~4u>nhond
z*q0Fx=Bir%a?oyp!oVfop)N+Z9FQ>?Y`bn!BdTevg~{>wtZ>>T1>?XAWxBK2OY<f`
zf@sjbdL<N=ueQ3Ymx|G));)rVT+jHEpI4C_)F)idMIM?$ivHUJ<zMfw-fp5;eITR3
z@+vF*Do%%(&tw#psvNnrZOlj9mj;Q~sQ+OFfnNkw+`{l#MF{Z@$TK|Nv&+&2w(_Fc
zl{}XbBVGbIAmu%INhCgcsVX=epG-DC<vh^2Pj4vS=FZbck96#wN`GYs$7LyWU@c}(
z9PN&w<IxPwupf}D^A46@xVg7mKe&7U&dvUV+kbG2@jEtK*!Q_HwCtTcqR<@R-JO@O
znDaNli;`iAln1SG(#I^mJ3D+JQ#oLTpEZjx@n%tC>{_N*I_}Sm`%mEbhs8xxlecq(
zY&&d?2E?_N`9UeM9c2GXkxE~Rya^Ny-OR-D$<b71Hf!TM)NGwdN5~Ben=9}(L#L39
zo%TH=9v%mF+wqTsnaw3W4ydxR%}X|>P~ztFNz@=Cqsl@|pD&!xwmqg+bZ2Yp$|kaf
zl7~2M`Frq&=+e<p`XcNs36?wnCkyF>kI{S}l(wj5l%$%zs8{qPrv8s*MdS}soH{iI
z`o$5pX*HY!+ek&Vk-#AOe72WYjd6<RiFxD0VR$;8AQaBy&h(7m=v_1&hdo`*>J@er
zg<T<Pb_hi_Wl?|DVK1VmMz=M|j8<ulxFey?JB2#qA%xc+av*I>ym0L$rNMn6BGuLh
zA3lV^zP9gYiwrD5|7oEfPP-7E22Q+KPFFYlZbU(LrpHgRSY-sTJH|mqtizjM0qy<m
z{T6&p)T>eSNFwtxVLg!@amKE=`cBiRBZjZ>Lo%qaYj{37Z@A>-wZ3y|w)Kg=Cjo-$
zwJ<^;es^`SDo$zpD3R{8DLDAFBQ7)TRr0c}bX(Ob$j#Uuu1|iBY3Fe8is6WAP)xbO
zfp^(S9YR8th}Ve*KM^XeA2BXhpqp<07*X^T2}W_y1aHQdu#<@U@#OfG=Er^YQi<>?
z9Pv}u8D%)AzzbDV#VB|N=dE-gyfFjIo^6VAz@gg$d&`2{RkguPLoF=0ih+NwKU<IJ
zUa4U6e^pNn>W3#jJCW205MBjxtyDOimU$yv#dugmG~$Z(2(wB~5wMlKe;rfN74p!z
z=7=f<iq*-uv&whTcGjpI`%#nBuROd$0RVwTOpJR1u1|F|xce4sczsK%?YvYxU?&h{
zUrGP;yIVM=6M;@9S|!1PqdTo>lU-~_V7Wdl2+lCgY9gtnA_+RmNxx-t#w<OYA}x)@
zwO*}0U(G^sArcqG`F?#;h7e%|60BFq+*=RY6-gOSa^VB!mB-c-A|5VE4D|(`fHM?r
z2(Tr<Yeo)!@yh%y;8HCw@msI|4|zI_lg{I??Bx0gM!K%YIpUIMvCs2bT%DQLBkG5N
zdQFXOS<MS$pLYl96?G!m8Lrj`$A%<`lTz)%%zg|7;4sWlGeq80sl+RdZEmG_`~bz=
z9tV4svs_rSYkr3$*5bPPQrJ0_Tx8u~?&Ijl&MAEq8Mo3Gy>4VlWaWZ6laIqZicOk)
zV}uK~m&B|YmnfSySt2(K<w)6<slmia2;MponIrj}A14D#rcS=M6->xhsBRs!QoLZ4
z70Pt}*oaB+%bP!0I<NhDW>7*r-xSJHx{<_;k66X*pwV)`lH^slMWlYEH6uaALPjlT
z<@B;EdMCp|5fg22`MS3K<wN)JJ6o;j9`o-E1}Pj+_>M&8QNHGU)}b`pX@Zl86*H7_
zg~=?W`DFnD%<z<UWu591fF73iqWul)@RPL52jK3XTQ+}g*&N13L@|f_<Sd&eP|dU|
z6QahJWKGeF;#14CqAIlMGcterB%EImBtJnL(BD5dwaP8V+{8Ht7ntcaa*U+Rt(7cn
zyoAu{YR^aipxR&^r?|gig>l9geKN=XD_e|M1;^kEKQ;E3g_pYuxCqTBP@3`W<jNk5
zp?am3_hju*+$q-6G2De@Q?ZL+8+6{@<*k+3Xb`+J9Qs^v4#mOQtrT^glx3mE>Lkn#
z;Ik>Va|o~LG6X_#q(LbxoL2-Q?sL~Wy^uyL!HJVV`KdvGX^y)BU5wEd3=Oam;#{rS
zLLRNmZYcBKm4}qX!6C651;B|%LL5r>CgKog=x}#r+wVH!OukK1AZA_Mhey@7B8rts
z?6s06{8Na*kf=n|e~qC<Epl+m7&X<8^dAR6w32%HOm~3>me7MLUR}W=%a)H31n0E^
z{~56pi-GI2zVr4Jw3ldbu-hg$v6`!*dYl~3re__s8tjRqh1QM@P=pz5_rPHJBM(-M
zw}?uz=It`eFR}F{7(J}WWLVsmSYwnms=`SNn{uBenv9_F)Y^F`q)ZbxFQYA%k<Sf>
z9Hd7?IV<JPWRMdZTAgS8Ylmv=H9;cz0*XRhENN$1yF&s~xX%s<V+!&)*7lUdNlnGW
zox#L)qv2W%BJNNLP)>7!mh>6oRSJ3kgkS{_E6RoK+U#O5y45$<OWC#wBr+<;*6O--
z+7GW1^b+=M4<V-KT7$k7;iblIBI`aEJZ>mAoa;8pfXTY9J;nUIz=0M~l*+D?7dX^n
z-DMB9r0h9dRbOJ$<EqfVa9EWeaIu}VR@|{rFF)jR^)7$V<siJ)VV48q4!ro=eNtJN
zA8tSbE_Z}NaCMU{s;9>&EcRMQDOmOVfzIc8n1ajOVr?-MgS$>~0kFgTWL~frsWuTE
ztMKLLa<oD~l+?00110O^tsSqhiitsy>s%P@H0xE;gF>Nmj#(N&)23&xEtfrNA>`6=
zi#$K)$c1IX3LqjUJ_-Iw9J{c|*F1V5h-5f^Vn;A+DPHY)Z<Sh2&7Fn?-G^LrnWGri
zUqV474owa;1aT47=%U9mgm1hb=0;8+Xf92MuG#Sn!NtpPqtigzrkl5G^d3kB@aJ*h
zBdo;}ABjO{zV&*F<J_ISLM)G?omPBaaBn!nSR5%B-jtM~L6|doYXr3K(H$Fh48-mv
zJ@~m2ef&eIhKsrk`?xI*AL?Bdj2Sj0{$Q(kU*>jlS%=J`fN^-~w)r;1^Nw7sk<DqO
zTY7{PR|*i_=_4=>&N6urihbL}_AW7WJr7ldfwZVW5`t@#L<ii|3&9a#8HZ1>xu4_Y
zsZ4>G{Qe9I6IQCzbxvJj8At(+E3OS+5~1wiBq||%B1huRa5C4Yi)z3<or{dSFBK}`
zMY$fYtK;Mbu_mA;jEDOtt-usvc8O3%%q3u^W&(R9zIRxh;yYTt9iEeCkzU(!8zb7r
zKA#PyQ!LLq{6-@WA$cm4dyK2Na5vKOvD=TaO*jrje^Rxj6^<iu?7Ep+L0h;4(m%xW
ztDWf52FFI;Oj)D$hB4ITC>?Kki6sN+#}ULwiSnpqb#Nga(K;n+SZnKQ=l$=7_mO%c
zENa2Q(N7tj_?ahQvQ<OLVj-cj9E3Vp;faUD0Z+>K2foM!ttNrb$S91;+y2fq_)#P{
zcf$5EOqJUYZ{5HC@b=wLvL1DKg;xTp(!wfN!aZ{42uA*ZZ;@mx&TAKn`08`S&(%{<
zbA*j4so{0Q>(DbSZ3U>@D3h&P>9$6kA9JM0mnMF|7`wdc{v?e_xs58cQDrD@#8#{g
zj67~B;r5mjH*(La2{$2nhF#B^+1yCFUc@g96`NZE!BC-oS6Of5B5J~I#K|5~Fc0_a
zYA$@A=ttxtY*Z)DC+g=HI!4bnjyA$;bDehGmxQV9?e(e`nm*68k&c!^h-ncZWjbs+
z?_3O2_&IuXq(a!lg4EGTwY-#B0Z!m1Jbmqf$c!}h<bjaMXn$fK@Ecozy$~_9gG-1K
z+zs>RB=8r`A@(LmkY6u&9w>HuaToVZ1(<lLr91P3)BXMNkavS!<%se!KczQtNOJj(
zLU2(Z=u~7-w65T*UDy>d6dH~JcBBlB73R3i)Yk}N2gBXN=>#*xa@E1bQscJCRWM80
z=!qzbtCJo)x74E;FLEL^i@fAT*L&~1w?KB`RU{qX!R02`#Jig?8~N4=lFJ(iuJp#5
z)fQiP@vc};E5c!VJ2MzYo*DpP91vaBb_25G5g-sKbk=GHP6naF%S9v1yFPBs8ZD_c
zV4J{z5|Iu=^tB4&t-!6f!!&^s!VSpCHP57{jb12(<31M>p{#`IlEg9fOjdWDg78Wm
zr(B~cPr<WPxiP`sm5oRq7HjvCe?|0`U-?lezfL$lS_;dhZGRHfoNg?goY)Dk7efi2
zSw$%Rbs0*{dn*CrZ}3EGY9Gs6UoX3rLDn?9ZtY3AqF20{0{FVYO31w^2C=IU?+cIv
zO^!(xaEc9(Th8e<mXM|XzTTlA_lU%WS8-OlzyuLR;HNM}di9PH)BpcWj|KwNIQ-{4
zFDq5y5;yt9ePVMC(g9hh^S3+K;5b1CX)k}EKH~JBDm7gCxdaN}E0KR>I^U87AYCM=
zoJ2P^MQPI=v#rAQX_QaI`Xor3XkLfBh^vDTc_b=R{hg?ByslM$mB_;Mb*8Vx#NKal
z%2qG^#Eq}kaCDYh@>PXs&^k0{%(6;{MSn=fO}YkEnyJsB9HcbnFfP;|Q-<(71FK6E
zQNS@G|HkwwF{7Y!QAUJTl<1J6#Z?~!N8EI?Qp4je_<lH@z}+9gg&2>X@iq23+5Ik7
zD_Be8;OZ=L-o{DncAh10>Hk`T!+l~6(xk(s1yXN_wKjMOqb1A|U&oMrf8i`vH*6ja
z+)%p>=MluAHe;xqK|l=1%nf_79o$(=$DG~w428Zoq1YI<Y58tOQ_YOZN!{wL&(e&C
ztzHWm53B#cWjuU)`)_>#$=$wqKV<CD%yOF1ZTsynXn{W;7wFd&Wzhqy&Oi6&c8fq;
z=hT66Z8d)3&Fz2oP1N{$QUlwED0+;gnEQXW{}fC9YEm*p$gxuYXYc<;p>HOI>c&Yc
z{6F>nJ_>)eUfATa72E6mk2RR}ZK)N0&>NuepN(O{b<Ps%eJF_rzY@QUS;R_x(E9)&
zz8shG;@FCGdtDUymAFWd=2qtWz3-#Um*O&6*1q5D-OID~{W+%Q`$wWWpFNdFjZMkh
zCS2v^g6_)uI<4e!I)){BOR+_xSAbLE1u53dSMqu>gG!O5xL!&wIj5v-X@c^_>C;<R
z-+bfbvp@RvS`Io`Wf(dI@wJ9CB%$klXUQaVD~QiKJ)23a&T9y7T|Hp%OvFdqMOZud
zuWn0jwqbS03_>{WnByx}3W-P9LH4?FCDrFRU6~JO&luv&&VCsO)qM}i=Ohf4d;A=Y
zJa#|!hQq0Z{d5naJasVU=K1U}L55srU1_=5XHCb?x#Yq^p&8L|@etqL=`^ojtyg#C
z`0{s;VGdre9-d<LTd%GkpILgh!R{_>2sDJbTG!nZoEN$F7<$E8`0*Z+Hoz^I@p<vx
zp2Bi8=h+($)!f_brzeveVj8KJ(YVhvtmr<_S9*Y`gB<1?V<a|UqQKeR(-Q=m`N8Pq
z;NBq0z$d|Acz8P-Av`ONz;Ne~KNFNY$HyHPFPizi=GBMIS?yG3y{R4uh>oYJz*HUF
z=J8+l?6zmg=!K>4GY&SX+ms7W>0)I?+RZx$`{F+VmCYTz*?E!C{NoQF!l5oZmP-Jo
zIrQ8wH!3{Lyf>O+Gkt8C5v~0YhcC?usAQq-&upBUQ&7|;2Et*+CMD8+F$xH&JZ&(c
zpm}oi<QUZFB0Pscy<5NEe{kdO{aZ=T6h*{~=Xqqr=P+{4250@jv%v_li;>3aEodtF
zLl-f~=!goU*My&vuuz^!-a{0SR4ko2<L8}%UN}31{;KwMN6vd=MA4<4UIfMiU-W^N
z#4LLYAw&>AWHg%`@3bD??cMqW5k&4k=-<13|N5Ofx9;42uq}?ft^ENK)wJk~^U$T8
zfsuv%gs^&ydBQNkIQqxX&6zhP9J0;dQoS1rzDWa~AKsDdGVQ+P>hAZUUcIGBI7{O+
zljMeuG0gDm4tH<+bXb~|=u<zuv!DIshBBgF+ko{qpW(GbdkC#dCN>PKlpY^$Re@4o
zNL;2p*b6CL!)9K65ZnOCZCR|5%wHU~-c;__!zox$r?#q0o`pi5P<1**pt61zyBYS;
zE|H-=i4GNH_u*Rv_xkUC%OORw=1?$uk`@mgMal3Z>|-)vrU5xJkaoZ$XSLDJMDPjc
zw1P?!CFsuG8`tmj?_d9<$G9L@<0wH=H#!>8qs-+Z3{jx@^+uS>f~Cp%>XYedc7iEt
zNN4wSZy=F6)z+*X4cM<y*nD#L;jOJ|cY1n7ju($iB@!Q#18x~5-<78@<-XTsI>K_%
zDjb<A>W7;wr!`#A9|=|+7?!a4X+0TFCLkXcoWpKLaolS0F`D#U_?;jb4O)SCJgo7m
z&t#Tb5TwcYZ58Q67;+6WNg3N-jfx=>LR7AuZOfJYI8*&VhMu#<Fkysex2ob&A4+ku
zfRp`14R5gas5a(obPx^wDOakk>M?R4X9<u<i)Z8Uk-kL##E=xgsTGPxU{2Wz>UdL~
zY|d9#K5%k3kQ&3{n@)yKS2Le3r}cYx?rbCa{cwJ~dg9N&)oNkhPM)Oe07=aNX}FL|
zE07xBBo{?)^>GNeUMG0?A55|ZeRl-VFzP`)ZOag?mxl<!r`1wZ%@15&v*K8WfG!{f
zBh4=uIFAb_hz&iE<|YJ^U^tbm4E8)V;rN;=3e~&Jd1zU-K@2FPIZCWQPs_lbQJaz(
z3v3tIgD>IOG(N+T=>Yy-@ajQM>S4xYEVYP{x{F?>jkz)Fx-Y(m4IOV}G~d-=A(gq`
zm-!I5Jf#G?BDA!Wo?-Rm8-5~6YZ;|)+6@+5)aeUQB=oy3bsPKWUo7sOCob{j=cnB{
zwwa0(KH(>H+armgUPRL%d|2PkAj52uu620P)vZuPq31TOb;22_@+4uCp6JGL-TXny
zBcX@&fn5Uo09*^t&&RmV%yV3;9L{TqdXlqS0Z;$fY`7_w@;3Y<daFk=Aa|MNFk;KC
zN~|3_H+6+Xlp=2CI5sY0$!Zjwb0NRTdg&I2Fg7VaMk?)v_|K$%(@vFybsu#MyDfGX
zbqHI%kK{LIc9&%dJHx$>PejRu3dSo&iT)He09pU|;K~ren|45tN3BmgS&TV9c5KxT
z3po`OFu_=y5MksTzA_pYmXCGjqnD#wbm!ej@oE(jk^GFlJXFZdEFm)S3(kuPz|i4P
zpa3DTdVNX;5DCrv<Ijsu$v_2!{9~JYzFiKX`R%){qM1_zMz1~R+-wsp)~NC6ETsK$
zXCG$j6U2;@AJLOXkAg_XbciOgI*%ieS{s4;k1I{5z*`=?UIY^eLM(==9M-5Ve991!
znY?3i&++d*G!sN~m1v`?!|BSL_*<J0*D$zHiwG^w<4_;>tZB3sJy`;_A2{~I{~#W;
z?u+lB@FZ<*VZ>P2rF4tc4C%dx!cu{ub<5-#W2Re4|2y(g3c`*Mob6BD9i4X}a>?K2
z1`>*{xItFW4<s3{F_*_7uqekwuo0Q!JU6wMp%fw@B5IwE9f%qEhkB6oUDR$dH{A=L
z=t9SnBp}z&LY)#!&O~7&gm&d(zy~@*bN3Cf*OGv(8{wS2*PN@j-Nqc!BB_1~gG{$2
zSO!%;S}fSaibTnmGRSxljnulFP{<3oK8cNy;@+6kLIEVrMLdT|!npnbsD^+$$UFQ{
z9<>D`ZXxL12GD7ZaoAfh03dUP&A`~UATP-BRg30??9?1JWMS|eWE@4+JlYzkGnM_8
zOK?&(D%Kg`%wQL0erK#@$@nA42gZ;d!=z9XW^o;U*X>O*Nt8m_h{g5X?!uToA+;+P
zSe8%ff`LyjgY!l4JrCCu&!j6EdbZ3{2lla<m9ca>D*xCJq+sJ&>n_~K&_ijDfMd8w
z3Nv=*vLZij5xg$A)4>DFod)7=$0d>#p@597nsF6DH2kBBXHoMAQifOr;VHB`*On7a
zUEJZXX{!MoWs%o!m_{Vph%2q>6d%2ig;*BclG*=$dOQ=YsO)O}<m0Trf1J<Fm5L5F
zk57;H#t0IG#mZ!lJFTF>4pEr>m~&1{pJ2GyO&;kWc(Wa%AhiS1zy(sWC_g*t?IwK)
zJK;z7ewUY(0(smU{%zgRw@Jf%A(7dISG?pfsim0r9p>tXUaN1XKxf_7gNOHT-*||l
zoExzACglQqu|?x739g?M<I4h-yXPG%)k#E1(<jKVk<t<#=~IRlhHYkq$o8-p9%@>O
z8X76a=`UXx6AKAh5$9erhQRiSBjz0tM2?ghTPX^S+q5>6s7y2VX1YqOR3BmmYb-3`
zjpjJM<DD1;K<3PV9em#14EHlhlnmvamMbWyZkxVqb>R)hW=#F(NRlUwi8UM^l}T$?
zWtoP#;2G`}dSGmL1<qFP@ELKL5S&(%72{qD^Av}X($u82`<W}1gs641{UW_eGh@~$
zfQg44!4Yz^iu#8+Ewm^wB3>nZ@;_EqnB|lV9o;I`<Hei7PJGW5Nk`W0LDJ(Y(u^OG
zpQ$3n$0{NR%vzTpUwN!pF4M30NaniK2ZI(ZvE}&`nl#OUtv6^4C#T0lcRqa9T|I`g
zEzScH%!mp<nQkP8(ptfE0<v1@MQ4BzY|p?&>Kc$)ZBt#K4ku|!hXOI8$2d7u%F8km
z&urXwI^G-1Xfr4f7iW-*Y9y_(8@Du{b)||##pSAdtpN9><;dHrr0Mp~YNyj#U6+5c
zk=&E4b9y1Wu*a`fRb5&2)eTAuP@t#FpBk7YnyS^Qb+C%EXZhB5b~dlQkCe5rNbv7_
z>m<ed?_XQ5fR8*xhQ0i6WTay8<Z~%_3jqF%AowUtaLzuvQ|`F`aFReQJ7GNTijdzA
ze(ifq%?&mck<3_km?x(=1=3=2Tkr^PC9OI2lmjFNs14r=lq%a#M7v>bCiR{doMuNj
z0b-;*gcK({T*8+Ci9S&}?(rAEay7H<ItKcx`)&~vuZ3puO;4HBNv(uV_dBV&LDH-k
zO0voTCppf&7MkIX*+H(5VjBSiPLYEJZgA)dw#uMa0qZ&mamsW2WSHe+CjweDCYq*z
z6Y>{sX0i@Hx)U|6v9M#^siDPMn$jM#TEpPr3jpa*TU#^&tUlV2=JEXnGU^l|MWzH?
zMFbESJCNwL)}6X2eYDmwi&7gShJ<^2qa(xzm)jNmx@mjs`o5Vx5X59B?)VKWwcEzX
z+7oSUV<N5Ldfb!INeEv-X#6xmEI#+!{x#%GnI8<W?W%Qoy}GgqW<^p=xaAFDd}6Pz
zNVjN$!#=G>(UofFabkR9Brfsko2Ma&VuW?irQL7P^@RY?_N6bLo6reGq&^=AKgtPJ
zLzM6NDYKWpLn5%2oPW56^Wj32Mnqr1dP1wDE41U9eWb6`P(}{2`z!M@upMn;)<;{y
z2qkkSaCP~&fvfq7preSDwc)LEv;eHr@)e{DLWxSUdKG`qhpP~OOtHe_b_f$Fw2&qA
z!i3p!pAlAS6t``V<3>N|!80ckzP`3<Vaor=Z7UA0seu(wol!AW<Vvb&+BwMd-P0L8
zi_gT)=L94;c%^z^i8rG{>?tKBHW81pgdoWTeG`1GR7sbP$|kZKzNYQ(ANK(r8z^Le
znaq3wQ{Q=x2S4giAnH1}Ooc;GPg9t$(x30VlWhRP>oESl!M%M`D!ui!K!40SY?3+J
zkN~poxTO{xCoyE9Nx^gI#c)@AbGcOd7R<W?%NmmwMMIPZu<^?tcW}JDg#E^|(eU}!
zkB(nvNOlf}$OyT4Rl*Y~y&^NZ;_thYPWO;WZZ>)DW3ggBYkAOGp4%_WCD(^*(&Ra(
zY=+3r)}sy$%y<jR_#Owj=O5Wrh;fs5;UCq^>6Zh8PtCw7s=9|JU@m545|a@4E>a0E
zdJq~+OM1E!HhFqlSqx(#zC6A|BB4@G9?k#GeZ9rMsLqBbvD;2gqGnU2r%!=eX2IOT
zBwy@tuCt}Nud%rhi+4PtwoDfm+IR`|So$wbhRB4qr8Jo&In(kmbT8+Ba*^y$_T}Jf
zE1be9XG-VfX21^F!cOU6TZ~WV=BJMZ2-YWLla-M9T&nD`#Uj))oEdS0f&UFxtfG7m
zaxpf;k#K<l%InHIDFmQaCctx)(qlaX3k^NRRj3gB?@|k~CsB!FkaMoIQZ3rjXN?$D
zNi_i(IEFNhEUV^BhNvBmc$N=)E-BARR3To}Ha*H7oE~wxab$AC5M0c_$Jpy&ebQGf
zEJVf=*N09xhH{Uhp$KGEWmXQ>Fh;?eM18`bhtY7@hv`o~b$DV6K-P;o19Kr&Wr4nB
zmt?i6cMW(zgpe{!eKczw;~-_F)d0GBS9TxZRX@VUEg8W4Ug6c=RopB5dw<p4EByBM
z6JFE%X32g&>-76`q#m62`*ug|@AUo-3a^w4yOxFf^?Kd*&YKF~atLnorpaw5@Mms+
z5!bl?=JpSX(Pp`ke!o#7T9dcm+`dZyf1wFLy6RZ1-+FWVL)N;|RI7L{qLq99{dSji
zztvRNuLxHD-|ziB6vurmY7;q4!h&d}|E1o)i_#4vl!JcM`)kdhgEO&>=RMviY(gZs
zQd{Y}y}Qk&b!ngt(MtDvJ(PaC+)0nNo=sSl-|77hsx<VJnzyKkR`_=BHVS{E3_rU$
z`P;oe$Zt;mfHx<<T&(@sQ+atl9#U-C9!lYcF^`G2s9ZcGi^|0VE#(^e;)lVy+<;!p
zk^*iSPM3lgola6|>6t5~FC!?V;!-k6N-ihxSoHMiU(#!yU$Hx2?uJ@tcGRa8oi{vx
zY4wda>}FPUOOiVW>^AK>VRF^+?(_tA!Ukh~M`vhvL(4fw<5HO3TEy0^aMzNq^mA9l
z`(ZOisN4pNpE$^<A=g9dCFEGs-)Ur?=e}9~;Fv+Z_GETesD(ywD{qD~yYZO8A6-Mb
zsePz#Z`jMXaKS-P%dXP7N;e@4fMHTbCei15#}L}@PzKF4`=NV~H0;M%ukdlCwKuLB
zS4M0V`y>r+=dRl{vp%m^_Ykd9C6PT^1b&Lbc>@ISVDJp=;|Um3uZ|#vsd)j@aH_nc
zD3G)_kZgB=ExTS!-D7A`c-ZOQ7CMu$Dt*81EO8I-A^E=TX=leFFnrgp-bLbAD%l0R
zj;gm=Xxk9gR&<~sLvammJGs(&nM-N(+7e@kz%>l~8uXLBY3KP));qM7)8%uyZfC7v
zY^`@(ywRAGHo@F;Z!(7^j1mZz54$>Sb<}CKSgSWY!#ae!%b^OjSa;x2J}1rgDPd5f
zy@?-%=vhj#=*2tiRs8!MMOn8oL9L&k!n>or29?)t8pcgWE#?gweCIPAG<In~{$Wok
zMuPdSuw_5jz(E`!@`8N}LZ;y&RCYT>uxB7mPU$SYrNcWX#|q--!Wumv?PAY53Rz^~
z_HenoJZ`0O!7Kxz$Q05%zpNlSFO6aFQbFz~lk$AQPjIVCN1=0srNujR?T=<K-mCYb
zLBL7Y6q`o`93K6lM3re}XR<Er(w@6i=8PRMZ%8d14k$7y-p&g|NK8)aV4uKS+Uo2h
z?!5EBWZDfx5;}_7aV_|^Juz6k?zzFI2>@~vLOa5_NZJc*#**$+Yb7v}E5g6Z^~mUD
z8(;Me5HH{Guw6Kl4s?;lFX~^wRlvBw&qmM4lBnwokA-ME{lpug1)%wa!aqXtz9)@n
z<aXB$Hzpa6AjY!-;-R3BEDRdErVup@G_)JF)O`X^hjjSrMPU|0q$yf9Qlu;Z(SlY2
z&4_za8~m62jv{~J^sa_Pbb{i!r<o4SMv?X^&QGC>4(=<vi+3rpQlx7Y<t7T1N9NK+
zFQjrd6iFqP^Ei_Z@mm16sq?NKKNb{IM1FFzLB__UfS;pv`4HomqDBC2FakF$B`8=m
z%pWpEjz~f}7B35o8la_w25Z$?EDH`6y|nOP<&<VIHPP)rgL7mEAvMVn!py`-+)Vj^
z1p_Bz=GIS2hPqTpi*!w4BkLAa>AHY+G@eD`);}vr+aQbe&O=aEH26$E6i65(C9;3D
zQe6wHuE?FprPZ-+)CRq<#*^6~6(W(A3Z?R)?SZ<5?F~oCeXKdZFBIc@eM1w>rga8y
zYFM8eBEGY1xY%fDf<=6J*Z)k22@SB=r!+7x!L#GR(P;P1;A}X%uI_A5pWWW9Z5X-V
znJ&V{6Xlpw74F^rF~LEXg%uvbnsA2%m%gX?Ed#UeuaOqO5oArlO?iPu_z}6_+QQ6p
zUp+arnEoq;Q*=-zeb8J{@J4z08;;5225<+5xlFLYiceDY93DwLjz<)y`MgqsP)&t*
z)I_~}6e#`YiVe&7&|t51F;AwzpzM`FL{{&N2S<YsC$k%a)A?Y0XB$3utI~!}w2s1A
zZ5E2F|M>Ia@wL~2u#H0o1jl(z7%yHM#z(`W*QCWq_a0umv;8qnw70oK55xp+F2+{p
zpn;mNiX$q6fh^WId6Hm27*1hJ9i+a!>0@JuIXUe2i$E~PYjAJ~RN$6$y2rdKtr}b5
zBA$ls0yYI8J9%TN=y6c%lFq&1-=_BBd!#s?8aP(>IAyn-987^Hm%^+DaHm$wHL|@b
zZ&wkgv)C+BFJkL<Q^g)^cxTQ%g0mH7qRg#CsZ^fw0w2yiqi0xSMB?!@ZG;IAd~jK=
zHhywUJfj0wpPs&2N8EiIB%zi7w557>PRFb(j)?JJIZ$K|<k(lb(#$<9%&Qp^{7Qf#
zeUMA#l;28;Tj6&p2}iddpN`;9g1skfR~G4{OfN~EGK6Ch@@e3hTNW5Kl;HgG1UCZU
zw{tR^jL%(F5QLDa1gPJRPF4{?3qk!j=5y%dh%~6<Xo8w6*Y5a~*2D@I2Do}f$FFgp
zxa1O#4r1jzwuKJSW`#A<)ukc+tOF)B6c^msi+Bp#&LSff%MwZhFX0{s%?)tb)_p&&
zJ7|i9xt#5TA+7VZ)`+5@w_oGk0&Xn5Qr*Yib#f$kLYNxCFV4Cqs@r_Ap^LNp{o$4U
zA%{?^!cED%bn1SYz9R{Nqe&YQitHXtFp-&01e;jtd@OI^2q4d#$^l2Dq1^b<ONwQE
zMS}tFx2~I-1>UmGg2ha5z-AzTiwM4wffBbjcsUAJ&gn$IKR%rwh$XFYIBp%&)DsQ4
zcqMZt6C9e9^c9O3fMuB?gboL^;uJBwh|rMy=8s-mbktLPtRUJO=H&6oh!=d=zg>4$
zLk4i4JzWA+3hR^;gPT|HA(w2KxI>BPhAq;LYYgo8LAN@Y8ZIlGkPUdj^G3w1TtMK2
z2}UGi3Q-;#CQ2$<np;cDp)EL5QkJ>roQ<bbYzf6jAM02X2Z$CBZTXyOOrDpveZ}Rx
z8By(X-!J?LWizgZ0+t-l&7CXCxg(m+OW%e8#CR9CXXz&RLX_Y_RqAxAgG>9)45>Ll
z+%Ip9$#}_|2&s6<E<XhsTMN4Jj|Bi(Ol&d*cz%l?@qmApkAGrfkQhAh-cp->ie+}%
z*v)=l5?43d=Ss@xAUCWCffGCztW@9b=x&>5h)zm0Y^fdnQ-K$96@iy%dhuWg4tjr_
z&Im<BkY<mQDh;jE6Stmgs@W3E))U;M;f$rqKofK%^-G5D1zOE&CZLI8@}QyXPH9Q=
zLZ!n%Sg0;DH8H}OwCX@*Ofp5)s}qD7=PdYZDh2r7eOkY*2_m&Y>Z1T&eM+l~<e`$4
z`BTyfN+r(MvC42Q7BjHxPBc6`<*XhIPsE`=f+EoXKaE@mUHkkS_&I*MJO#jvnzkaY
z&hN>XsGu@Ga5MW4XNwb{1->hbaY?zLzV2{h=K_Ubmm1&J3m7a@6VZK!gZUX2ES`9$
zO0oGXpc^rc9ebRWtsS9stNQ556cI@#hh?>M3@@A(mI-GwDp~7iFA0IuVKnR!a)&FP
z<^4nm9Q2asI5m{kWfPpNCOB<L8jzP5GVr$Z@n|b}=#kc5iiE4dv=x`BBD<q_4`b)>
zFlDuz1lp6dd2sR<q6^o`xrteb#TtTJ89aaC6(3%nN8#2?yb`ZSdCZztu(~+m5ypn!
zc-BK$%_2*2+iAJd$kL<q&bYz7C$xmV%?+zqIH!HixEu$vpWX|n+FkEa1UNyZ087B^
zw{(k9hs55zvJFW6ZPd=I;J?gP%Eq6pAB(ymWc0{88anQ<pb&^Da6ohD!Rg~vUONtr
z3bdWGK;Y#%lCIvaL;|quHcSr^)%Hlm7e%rJD8JhKRX8ely`%EidjA{TJ--&|lHwuM
z;G!J+_E|X*nf(zW(4We-&Ae;>Z10PJyIzB9`0t(~gZ&Y%aqsmf@WSJ_nbrJnxBoSI
z{eHbxH5|oT$ydGq1_fh1SaHghZgXsX-g}PHvB##CPDjoPf7JU3h3jPToMz=Tl}R1f
zAQr6Pt==sFj2&aObx>_+-T9sj#aC<ncCIs4tkZihce?#Doo@eJUh%W1^1P`{zC3;U
zpQ~TxTJBeAk^QHWXV*l%_g-yqw5-1M#v9@XM{0-}7Hgn6yZJ}XEs}}}byW3;Ec#!r
zYtXlK2@IkaNi5rh0fm2!PbDEa9FU^HoC1%JUCYcS?QV@)Yr**K@zb^Q-fZ|#HfCxs
zS4BTOd}yb^GbTP@`hlgL^nNS!UkFnHs9AV!mQO8JJb3M4i!}_l1Pz(o!dAhRY$uf6
zrrn>5t=Hs151IsBTua7qh9L--w@^sz5R``DJTrod4R->Gg)sg{(-ZSvSx^?a6oz9g
zewLvfSEwbf5Zb)5c~vik7-D_w;7&E_5Z>x5TxmSsRl^3m%9|dHN7g!o&qN;{u`0?D
z`x19A@t*lG<!^)#`$_DBhA65B7A7pHdJiQkUGM7r(E1vcN@1#!BHm}lCw6_^BCZAX
zj0<ttL@EiTPvId<cX0L19?phO#l#gomYYEgoD-H4c>FmI!9`j|u!f>8cSf<u$vrr8
zCgeb0=UD-?dT2OLG&0TJgpskaP$D-5UQVieXAe2KEPBqISLlATdOob4LRMF}zt2!j
zG*E>>@(T`I;JFT_ixknW;#5PsSJ4PB41`CnbNDi!;u<Su$9#i+X;hrY2s+nM<8J~T
zcv!n}3%^OlR&Ny}K@W_=4ur(sNh&BN8?N<;G)R(-EE>4*q(OGrMd4yKHH%432FceA
zn4_745$*B7o>!;jnp@>ARV5uD9BhPHOj^*Q_REMv7pg_=p*bW)Nx;_Uo8I6i7=tRZ
zB?`UZJViNZ$}SRt$%Az&O<|h(GOP;$D(0M8-f>^be^KH?A%fGNk5(YRxbV02HG&uJ
zL4;3-yCa-UXxZFE?S`)rlq;|dFr`l;D1E&Mkni7cGFLE=caS}zfcx%s;<vuaO7#)W
z@7xg_CJ(Q?*^n7ZcD0xxDWH%_G$>&$-EiKz`vB|?f>!OxJ_2D>S0nuq_!wxDX25D^
zB-_%_dL@SO5X&;5L((XssP3aQ;%voz2&bwDSH;k%bYjL*GCw5uf+t+BAXMUdzVD91
zSj9<O|B|eAJbX^`szQb{0U!has}6tKgb^>p#p2BHPb(q_YrwazSC?B#kRd`>Nr;g&
zTPLlUHK&(EE@>)YdQP7}jymRH_iTX|AaS^Z<Y|ZrVoDe{MSRN~BgZXPZUMI>bNNEt
zQpdQz%W?IHG*jBpT1e>(u^JorW&)MZsZykJe>c!Mrko=yd(<@3T$$1YKX0xTs7Kvs
z8S1%a)vVLRsjieqZW#)mTd7RLS)d_g$lhKRq#1<<xD+d+&Vf)e8!e6QA;Qresw@z^
zK{#e$@?JVvn~^oD*2~MbT?4dC>(q&$>pFeOv{(=pnyolUXFZCL!`Uybc(rEZB2hmP
zN@X7^;n8E9S<#zYqa4r#6%?yPtFj^D#N`LI*<2>P;>Mr=Hgx9$GpKnV`Jm|<XdT^j
ziy*0tWP1S3c7|v=ewGVD_$NUIM0<I-RXy8*YE27OZEN#Lq~+my^^C$ksmmxGM+njJ
zDrrKF85U#4jmxrlT`8kgZCwJfdFEqJim1}|d)#_yN`U>=?;al>PoBfg&r?<o2i)m?
zcq~>YCam@Z5PqRz0vDBB0Z;hE@vg7=6XcX+Hk1b3)O2JqUxEb^Xk`}2W*3|E{d8QI
zGu?%7$Qp21B#pd^N9BfYWpo?ki&<CPv^F_!6iQfq=kSr=yt(B7kElq%CkY!p*$~;A
zFdsagOE=+bsJhQ~eIwbs&`h}>J1y<5*lwGOM4||Bcg~?qfCzDMj$EV3VMq9>z9V~t
z<m0UFn<G=C^ln9n`C?(o@}$TJ8!$xk&4nW1aj%;zI7<FW_1lv`m|>m!93Dt!^Qxtz
zjMJ6!a0e?3362vOwW=^alYR9DLu2KvhK>=+CxaCVy>c@#mF`5Ep3?op;hEshO)s!#
zFUmEDA0(oyB#WBNh`v0cj*~XQ3jDx)gQM5e^jwE8WMjXB<CDWz{g)2zE14TQ)nL=~
zRMx^R>n>SYzvcyu-l2dO5_w5SU};c+MW&y;v<;tIB>MFCil%X1+yM_7k+0xhg}&KY
zh;-4qOw*nEc&N9W<$;J?5mzd??TE<O@c10UKz0V>nTD*RVMTUTD#1N&v})osBQmIP
z82Ir5P04)qQ3rehJt7@Z4F6#VngTr~585qk$XphYHJZ<ocbU>yFJSXGYl6t28Ie-w
z8FyHgADm&41`uvygzrR2A%-s?t~G|!P%H}4zk(+*;h^32VMlZ{>!EQQT%5Tn#mmt7
zsPBa?5F9^=9T^x(zl6DIRD!QwutED#(wh3OXl`+{naLr)%o^9t+YaH5;!@H=)@_?Z
z=r%?*d*`?p2|(p-%gozIvT<A;UX&E|g{XlHB|Xt~qo^+qU%*Jm{Mle+qNYs03R$z-
z-~>;4x!J)*9ug)xu4yZm**xsa+|$d}=^%F#vWMl-AzL@^xHsHnBB(u3mUF0slauLu
zYhwfaxO+Ik{6;Xk=kT;S+Q5zniY(018^7^wL|49c^}Qm|z|Kxw%GjiH*|o~Q=S`rU
zt*tAYrD+rcao?LexJ*-px#|GNe-}ui=58h<^SMmMW*O#?Ew;Hd*XP~Lh3W<Snjll;
znZ<*5iHTNhg?RkrXBvk9izF2tf6zeRZX}J9AT&5mF@n%|0PFA=p;T~2m8CmyBf0!^
zDt0gKFoK_rrVKee@9tn;B{V9JKpo_cEN#-g^2%mPMO-t|EpR!CH)dRH{C(tZKsF6r
zvA+Wo|4y3y0yCd!Zj+1{f%DvI7W9hvrVhz(z$t`$J4v5S`RcmYJ5Fwe-491WmnBSG
zgR<k4>$I3#q2hw>Su0!mWmV|gQDiO>vLjQ)68B?dW62u|s;j5tmq6TU9pO-Sq-iKT
z*6vwCgN%&M!k2(ctQtP(A02_QXh=y9(%y=`;q=I?iM7n_=Fjy04BT#Z{?Nq-{9^CF
zggeMP=XwXg4@Y^)oIn1y)LT2hp5bNH{;2mi(cJszS99G+hsU6Nv-ej3<;wX`qH`T{
zfBCiEuc7L<m#K=75%FQY0r-0F>j3bb^8wTqTZ8bm-q!$O^Lz-I!ON<Dwf9w2-#EXz
z$vt&%u*zTQePuc9>#LVR5WlTu38=Hl?8FA(S9-q!6xPqjpC(b=+gQyn_r8ppOYofa
z+^T)4_a#(YU7|s{u2`jC?)@?<y?cHo$Effc`-T7G%{O|#)cd8}E2=}UsBfJM>a(X(
z@M3LBmF4>;#V<lZsc^x;k~&LHv5Z5c>JpQy#GX?3;`2hPU1l~ei)%>`NZln#lu03}
zdQsUYm0m;aO8q7H#!GvtaFJ!@umAJgZ^%bx!Rgb#tNws1i~Ir27Zsj>H`jg0#n<rj
zTdyRJ^m=szY7?*i(FYLW2@si49r@@~2Y+bq#79>;{B|ye>K+*T3U=TYh^zv~$4L=9
z?YQJxjJY<iKD?--pk6V6OewkOSva)V(a#=9X&8{MyO?e9^C5yDXzWIK8mgOQ<hZC8
z7NUDW&fifn2z8rlimRlnb_;HIcP5kR-P02on9^{NSwO?Ry7#Apa;{X>ju;oIuO_zH
zbL^~XN3*fQ6g`=Z`-5kLk&HXSZNkeUd*}>h*1xf<eDmI2Bf(oiuh(7F?I`7S>)sd_
zG3b+m%|1QEW@Btkb`RkOhrK!TYE|ebuDFk$Ae>@nddBmui^XIFCeN-JF71Gp*=v%R
zBVLE3@)!)s=GUpYj{|p!Qzt@{9Cz^TbUdWpg#MkOBD_o07d-M-JQqqG%;Vw7U<=zf
z<Xc7)S36O5#PI|HAZS45_C+S;PHR4Tax@w3F(uXkhI7a-yE7Jq$Ic9KRQcd=%zu%8
zp>N*?`$)eG@cx-62J}7bj%t&n)?3N}e>*SgaXjPdovSZeu}f;8WumybB3HxvOfMAY
zOm@dRr^ZEj2ZrcEOl64TNgc=d)?-v&hmzexE%lGq#EJIk5z-UoEa5ey2u@(Bhp}sO
zy|OSOY|o<@YbTa>c966tfD-O3Aixa^jQ|)@7xPk51nguIt5DV71t2@%>eSGz&horN
zXmV44hbr(??daq2tZ!KeF@Nt2jvns~?%ZeB(gU}2;&9KKK8b?wO>sefHeoCrgkxgT
zB*~U%H;$rS1;Qvk4<%c&!%lyJWbw;Fr-Q;BfsP!L16$R+6_%-x=rQM2kSMr%c_u1K
z^F*Cj>Izm=#kowUG<Lz+_8=sX9tDY+;H{03dN@0cgxZrKBr*2g{k_u}#7m!(!xt65
z?m&|47H)(=7@Ty;!+=PC>K%%<tq~m8ldNHEk@{0)KN+8#MUpch7I=tVfUja-Q`&;F
zF}CQl$ylrsUJhdQqs;Ikj<z!rFc^f$wrLT>0Xy%CK{^iz!pn$&FCX(0xUn0)mWw5u
z2q~RBsaL8?;%@y>y;>re(=&j2;Y>>7T6J=zpkATQ#!bn)NsnYUNuQG^b?^~Wp}6;V
zO$!DyS1p{FXAT)0Zk6>`dkVno0bTzsEx;AN%YV|xes%BqZ)pK8&0YSJK2CUW(XILV
z_?|oDOAmK8w2O!O1gfE5lGC^aY1iVVIEhA(7V}f6>%XN1!cwXOCaY>%#IKZH|1B*L
z*28Z2nidIirm$9Z7rspah1uRX@mVC9;^$5J3(uHn{pm*EQ%nmpw*Dxsk}W*l>~~tY
zwElEU57TPdQj`aXNDFfUL8p_2fx1RCQrUVD>P1q5j%&QL&#g-`7i}^1QbUZ?(OQM~
z(Jx(kxI-Zzih!_Iq&t);QYP-6A&>#DE0NQ1erk%%huk)p>f_tB^3IOGL=C@+E#F9}
za;bJ0A*xBQ&~t=H4NxU6>T6wphzMI#38HjaHAqfFjDQy&McJTmgXf0kK9J)#Y&qdn
zi@=RM5E|HNJpHF-%*4IIQ_A8#U;ydk^LX%hG)4@1W@hz62O~umTe~>;cXAN!YLhi9
z^3NX;+WC@m@j<rR2h1M2WhtE{;o*8<a4vQYEOfDU9b@1`$!?o1(Xyvj#)^$i`wB*0
zcQAU1UVBjRZqhR_H+L@Fx&v^bzN;zV;qVA?`{mpc))I@C=Lgd<9Jmx0dbxHXLGlFq
zBU%G^6v;zjiAF}AfH?t@0)M{Ssc_d0c9sb*`>}|-8*5~PMUKt$MD%J%lR#eZB@ujq
zllJz$?iqMLF#^cd=@|~0_xhoRgAFGmn4KWSMkhd*cAr75kle9C#7lsj+cmy0OEg3(
zOh|#1!lb#6j3%?=Y~EZ)egbS_bz%T-ceN5+&-pq;kXaL`Gl>p?-)_t>`jQiKEv3Ul
zP)(~EEb#z@W}|n)+`#D3^#C^$ErpC56gkMOgG{k7iDBGlh&bF`gEoMgxHIGT_?A(i
z7CXn=eGC^kJ_|_1_YtQfayuArG~qqLg5asHxTl^}s4)%ElE5knjKDh9ODPU8;YA9G
zR}{kLE(|etXQ7I)p%<CPqDi|~lbX5?(?mvFnmBkmD_Xj{19mVtu<tP`1ls^UY1sGy
zH>^Kos1yUMaR6mH!TVT%2Jsp$EwFMxNv(OG8$!C$&4%OC_9h4}G`Ju}sadfJv$}GU
zjBh;fKPl49Xx_~6;~d9OUMcZ0{AXT!yhhVWt<8(F&k6(`?vQT&27C_nZt@fh^9(ne
znX6ugi}!y-Tt7J-4`!C@qp&XAHbn$sJBLVTLdOXR&B3XYJ_0yQe9jurypV{3`ITz3
z+Nj#s;4|;}1@4gVyi;BCL)XN<Ff`{Qa8YV7Z1snoW#$mEXK7_}#5fuqAv@>PX>Ksv
zz<M}39JZ&x9_!inzW2Q_0i|;P3GF~yA1;-Mkwkx%HFkd!La6@Y-D-1q<=ZI~PrxZ!
zfg>gDq97$(y<#BbCmI78-z8{BP@)tSwr6n#c_fw2CHu*<pElVqO1TD-DYXWV=j|zp
z!rq?S*@NMD3@v9k`)T*tv&v6{)dC&Jsh-E6!;x{?(|6wCSEZSJ*BMIw@(f0%XBmA!
z;&K+_nU)PrG)-qaJ9^zI4ffYrjW5eiY}V-?51)uMi0bL}yoNO_n$$sr*rfJ+mC*AF
zd5#oidz`Q_t+;9n625LY=Oh$<ie967osWYiH%i<^K0c78{uMK0d!U)_XMXh7#dc}s
zidCn0X<h5=L3`glh;k2k&B|$!W$?ZjyF3RSf})kaikk2pFn2s<((^TH?Ky28j0mGX
zfI=EO-)XXd$bL2$W2E9%<hiM5b#(D`OF+0Q@hHleu7*fG)9@#%@{s4He6FXY_DH%J
zx&Ft_hoJ)wcSTZ706;@*o&;&*o%Y&Q@{~+XK0hNJZZdXolQDr>GIOB-{r2<;HW=<w
z0#q8gO=7`ZNDyAFIDwacC*F7%Eim6-hKq)Q6*$)lQz5WJR3PQpVv#Uwpz9Y)#6XfV
z$oH*^5l-n4DhVf0(6e|x$(>j`9{(U<C3>eWcE=M5WAB_Qh&Dgi+laP>d8hsa<)bGS
z6<|`B=|aY|5=>^D)8pr}!Bp$&)<aOie6_@FdjA5c08?>Ji2S6a)@S-}5Cws1NYCP#
zHZIHWp?plkMK$`(z#3jH9t0syc_fHnU5fK*nE;_P=XXAoU8ds=Ju5t3qO)jPhu&SK
z^P}S!E80<4u<oQx;QEGH+f~`++|Q$HaF053S`6XTj|Edm9KrPMk(8nPF+l$`+%8)l
zWYiQ;jPT`B<;--$V~<9CjH5a@`}GA2_@*z@mLsErDi**-CpIL(d@^gxvJ_n7u@j@c
zp<9q$9eJ~cOpH2T<Jr77e1?PqL4Yi~wmR8VKTTji3rM6h&k6*JQX5E%1SWR`T~uZf
zB&8otu(;_;8;NM%jlPFMfikMsAR_<n>1?nILpn%c@vNK)5B%=&@K!zXwTdIn;-qeW
zgiuSW0`zg^51Vnnev83KbOVjy(h5OMZRO%!Uxo3?8re_^MHguuMxEX1DRQzN4F|`O
zuGS35p)a7^kIMoJq%v?^xFP~x6geG0jXIlCKAN4L@=NwzG*KV&O;9ber`LIx#ajVK
zV^f$GmjMbb;aqf4=CYs{A>0hs%eh2JH_r1xoZrfB%gAMqr>igBZJBTM{x94L_y(-o
z-p|rPr6ZGyi}P%kRQ?MoyV6iLR`jjHf86`e0Q}2I1t-6($YJkqqsUj1B9UCMD+r_B
z2&Gy{sp1)}l|Sem)EdrWy?&D*PhWl3{-a+2+ra4X+0!9!HT<J-Blc{mhNbp`j>!8a
zTFA@gC@V$J9Wa(Uefq`CH{UqHy`r+q`O)xbGCM;&@qIb%MmSn=)rIr<@BBk=ys`4F
zjnnyT1M!MBhR4sUDemb{j^A3rAq8Wfc27_Cue^_+H{g>x8$CHVsjgkU_8p`Z!2tuV
z`R>l;sAC&@))6w{-5CyM;+}jTLF@($h4~S)|8}bDJoI7pIVAZUNdkv^9p_ojaY0k=
z=Vb1_&CmSTT{Lf=))46~A)9w+SJm(1eH^?Zq}}n!y}=BPFzl@OJR4oiW5CzCsU8GL
z8Bsdtor8TB7}Y8Bnu-474{vI`)NT3^B{jY4H(1}%m>IUTqOjX_b&*c0Ve#|`X*UG4
z>j@6DL;@JDwa1*$1~$BNUhs+YIGihvaE;XPnSm(~;hiMWG;p}Qi;(NGgu()#F|tAR
z5w=FY(DH8G_uaW<+L^5?{f))Cp(!QB1Uxf>hhRXVn*uQE?2m@yJ;S~W?`R7c+;6GE
zp*_^jbz2^{mc%9Ij-F|{-M(t267B(GyQ>B=064=1O9A1W%i3YzQN)dUn{P5J;e~r`
zc*C3W2otL1RyOy!BXXp~(dRLt?I!FuN6@~?+{a}I^iIymq+3kO1Ya0F!+rr5?){Jt
z5oW{VB41M2{XKZLf+~yn3Bl7+bm9?G=xGVSO9_P#?P&>V)=V;S|2aO%6m7?ErGr_6
z^%`kQT3q!=QUxwYsAA(Kxz{78)Gvj0T8dH&+QM+kFKH5P<iWsE`Y`a$aTn2@jh#8T
z^Wku>8xK|oat}Gy5O8O>Cj()_FW;>oF^l*mKo88(tq~+B<(^Ga@4k?QAQX);OVA8O
zQaT2=N=x%VX4Aq-VuRR`zZvEU(uEcR@bEGjva~~?P&{A-S4G84F4ed!owASUBcdvN
zA^UcMq2eP`!Q{s3(?9)BLj{xjm0$0D6Bn3&k-xvq`<AWVZ|9oH2h>b{^}Z8d=^Xg%
zsl5JEY)rqOKK-^X{8!Ik_@9lDsTv9FaUyxVJH*=mZ~eo6qH8}?icfI9iBnNrn31^J
z66FAXC4-7?-UH|j$9o93jwCw6irAxu#~<-6u`V#v+c8c_A$;_P(nt4x7p6>VJ9N#$
z5iG#r{1mbII+a~w!tBxJaIA^c5*48CjOO#x;iI+-X`*i=mI>Rv0#LM1Z1Bg1l$~#E
z{=e+K&2n5xvL+_evS+2yMm9EMI-Rf?CfHz-tm^LROEuaqiU3)#A_y1&-l`f7?I3|n
z09gbwaWj!c;?k&<R$9wUOKtQ5TIdz@40;Sbg1*l^Jbq5ZNhHAPx<7W)Xhh~Y@#Ep)
z;o;$b@i$NY1jiDSR2m{oB)CAD(Xgu|Z$I7Ve@$_wKL<}29$@?b2A&9gCBlf8VN!DO
zQq14IdbP7{L^MsR%6{U63thb_`7V}Y2+oG>x4bVPmfJM=4cc17pN&Hd<IdL3?*5PU
z!N!ZNnt;{YuU~BKZ@qrLRlnGNy?yX){o>We%jIR3Wd(uJ70hkpq21}*Y2Zrr3VW`U
z9~4+~GDtao8NPMr6)~mjpYfU$x5a&t@Zt6<4@uyD8IbY50__P-XkE~94|Dr7beQYa
zS0e4(BPbH1lvBN*W|MK8@d+0*GPOLs9AHM13n*dH_fi#<ZsHO#JDHBIuBBbQrxk7z
zy!nUk`xV<nCNTuG3TWIIR{$2XWeQGkw<`<6j8jeKaNRT#3N8{t^@3k$@`4!Tv5YF2
z9K>FcLkZin|F%AvK}fnj@4?p1zaB44k)qo~s<FIJzl5bEf63Sh{6gVFdURwaYfnf_
zo~l*DVSyK>;D&a_2c6xlKNw?H9_R=>f2Ew8uJxpaHat3N#R(7fLt{PvwXv;-Np*t@
z4;)Lc&#vapU=kQ?5aI03XJJo+Hf=V)xL!|Y7WBEOaOls<;ZUphV-G^hElrk+h6Z7^
z$y*X<b%~?Jt~P75yF`u9yho|Czs``ct<he}m!fjoij!@0Nzs{8c40V>_7TR~KjTd&
z14Nj?T7|RwG56XlCgD+!Bra0N;e<H#vfJf2@|Orz|KYp;kO)!LcYoUyqE;zHIVp+X
z=RbX;Qc|ZZB}tHk2#Zh-b$|H3|ILze5p17_)Gp&tNaxKxre~7HHv0|DqPm0C_34~g
zt83eM!9OM!bA;yUUa7^<eWufJjpqe?JbMzAzPG>iV*7{s`L|oozuVj0etlSPZtpLs
zZ1FM%mo81$>2nu=s+3|6-ObxF;qDprKyts2i!BesP$;3-5w2ssNHsj%Fh*?>RM7e$
z(C-RTAaV~}{pL6B3LxOtb=Y-(<Ir%qt1etTD8`tzs;2cG_h5J(={f@eBjYJ%soIn)
zT-X4tW>^C<foqiy6{>L4bl6L{R*tbGRA&Jr7A_dSnlKpgbaH{qbgJ$chcl+g13{os
z4Af=Ko32)|tvu9(-C1fPZB*w|aSlH}(hJ-m{G|7e<61VqDI!D<M`E&VvFp|JYrx`V
z0!%!{(G!Tk83_OD>f=bldsMUS?I#SU>sq5Hh>g&}={P*ALq0H*KSXzm;l!^B2n?>_
zX69~Wl;GcEI1HE!8p@|;Ol=UUWc9vJ8%}#u9sv#_g3)LsNtr3<y;w+LJ$@RM(DSKy
zlGL9XZyJ~3xwkm?&*7777<t>M)X3|_qX}*PsDLZT&qUb<8QXaf$Y|Ze8ev&*Btq5z
zwOXP01%j%Z!&dvFQi$Sth{--N7X<oi2wlV_L|qgn*vckN6Emo5@OI<Tx;xS1CZ*HW
zXo;~tKUwDrp5aR~Jq2pkN`-1om_fIm$U}9|@@4=15QIPLhWBgm<j0;|Pu@50PEQa&
zJD<#<OC(E3X;T;0bi!q*<S9#k)}@8w5^?oImvA$5x$YJcR!v}J2f`PO!x}%c-wt`?
zP0uIODbAgY3?YBo+j?;rMPiWBPv_X0g=Kl}!b6Mz@3^32Xv$@xnZcMH=P5==d#)6j
zNoAvy_T*|#5O2aXv<e)soS%xW^(dRERMq=mp^Ckhy>rlQwCWfPhXEN}3h~-9HySRD
zLYfQ8Dwq}bW;i|2UCGFZf0^caIzbS;+drPW%?u8edUIuj8zLXGglk`@dxq?TYlRfc
zZfG9cY+31oQ`B{LU_Ow-4-h6FFOCN7rUM;bO<;%nM;j=%u;KI)D+X9hhYH(-yOm7*
zz));Dmn^U{HQeChea(|iLK(u)!m2=)%LQ9qzRQIjYZ0oo>rhtsm?I;Z9M9O;N)V*m
zNGcRxL|ScR<!c*xIhi}Kl)!m|2#{dSH$fu%LKOP8aOJXm{DVa(%|MKH|HiUyGJ~w@
z0={53$I7bdz@rFDiME_p4&lZCwI3tKY-x-7_5b=`%;NoTzxy3*!~dRQxm)r*;foRL
zcDCvNo+W?!e|*hH-@Qi7f5)2O0w8hY%{_1=0_W=et(5y8%jNunc7>mP_ZKwUQdmaH
z)$a=b;k!Sf@c&R~Iqv`QcVD}+&%cRA`M>@2-|+x+%fO|~zs);vwHI_PHE$osqGAzl
zq}W1Kv&8(TorOm@Mh@v-AxF6X_5b~E{~M=?<oX<byuKJ6+d)Ipmmz10H!YuCO#Dk6
z%^6;2WiO%VvG3u4qIMX2J{{V(**Onn{#P(wQ#P>2aMhuK+-ZfGb&lbDP;El71bI!3
z#XJ40D-_bx8`T1;0p<Tz1j>c$2@|VmA*!S>NPD|Szzcfk!yosy^oY3q`lZ|u8>ft;
zR(h>c|La|33YuW|^O@zuIoovBe?PeFQKxMEUE={1t<d-Z-5Pv|c*;v&dkM{fj*6Ot
zAr3~vYUdSlh(ozaJwM_-q1T7s9a7m$PEXOs3ZlB;#>H?chmPa94~b<mSv2Yrq5bB1
zdZQdQKYnsc^N_fPg8B&?Lm$eb`uqy^G&(9kJ#vDZ00&`X;1LR0i~t#?f?7FiPV&i0
zdZthq?LZ&dzl?nCAi1CRA%dRtp{()=+Cmk)$1xLEKDWG__HTG2bX#4-VW4_T-48Go
z9+%!xAPr;I{sTzqTL94ZN;_|RE4}N>)!z(1o(!Si-4@E03zP0XV@pp(V;UAsfl3NF
z-JHYOb1HK=gs|hN!zi(xY&KrKs=vkR+TYvQ-`LqY+}f`Xwl?;ke_MaE@%+23*P91=
zKH(xUavyv9QLOWDv?@NS#x>MIHBt~#Kf<57$j|5EL37$a8S=h#gLWU(I~$=c?oe@V
z79?on7|h@Y(zepJrW7S|kF5g*Ng%^RcWpEkwpMc!-(OAoP=FvYlxnHhs(rCmABdG3
z&ky7#v}-6J+@o@nbauzd6-pMc3)lT-cw}{r4@GJ-*z3lpqcixBl5y3kZp;f45c^Tc
z^&Qvdc+}3VnLu&sweIoeIruaW>^eUC^*v08dq@@!Br;x8ISL!{>8DB^4Z+S*@HY8K
z87?~Dk2hl=6ArE)cHPYi#fFAg*sTJd=x87hz^q1K<e>iW2={qRzJ`cFZTO(r8d7l~
z<`fx993jk{4#uHjfcvA}8+}n|L?pNqIHMZwMDhZ94(69|GQDZ5LI&HMFcsR|?cxaJ
z;$;`-Xd3=VLAYG2;;4zX=CQY8uTr-Ib`|ejJG+}(uj&`O`}OmUHwPQ9YS<BXHum<g
zKi03YkPp@>Es{Je^aNhY)OtdFHvEl2A8S(RhAhnI*eg_r-yN4+iUu;3R~tWW?Z4jL
z-_dViBbZk$YdWIIw3b_KYZ%kV_0#(46Nuw(uGF)~1*+D52J6-8FU>NYr@g07a9;QA
zQ`6+ze7a1Ve{OWkCZD3ooXxtFPtjt<4h#jXqqe<re%Mk&s)P5NvbCUly@~GMf*T>w
zMmMERU8+bH^TFQhi#;UE1JiFlL^=gvB7Lgl*(hoc@1ZEiR)gI<Sp+(Qi2J>g8?VKq
zayT%~Dkr(9kb6$v#3Z$D#DRrsnxU5v5a9~q5VDOylQB5vjl42bCkVabW%LJ^6A2U3
zw%VZxJU)j<>q^TB<bwcNeLTe~#|`LQDSd>=J7Fp=0w#};Eo}_i8F#CL=Drb3aGik-
zA=PWGUVkJ#G1Y4&z7M-QP3*201e-GqEbhQP(vVE8?jCW!9-TVz$USqp)SRW-J^Xml
zhj;bt9FpW<e&XIuhvG%T3Af{bcX_yuHzoX_eW^qiMN@&o_zgy~KX(BcThL60WW`#5
z&#=JgxreP%pS?JM<Cw_&`up3x1CfiriN)e?=f?+tiFRWxMR9+M*3ze;?OoB%YT149
zIK|#MzZ}nO3)NRb7Q?}23W8frJL1rTACS)-y~>-oZQFb;f!O4j9y<tP)b_Ej>4afJ
z)F>ZVOZAnr`%;rCFQ<$RpIz!pp|Ih{;mQ13tjHSbN>i)8Cox@+UZWjJsPMdx%muPq
zwII4b#yn0=X$%t&Wk+>%=5uK+wInl|@uB<0{oUO27to08YIJ09?2|Jod0t-&fzzWE
z7QNxc4wR5M(UGiBWy4@8u+;)5dcOPm#kTH>3f=cx+b_R8Jb>TL=GG6gTc*#sf9ezX
zX<-FWTZ0TNGdFg@u4FfK4~R~g2ui8Zn#G3XL#wjm5MN#mKQ<3LcuYwQ=)`-=j;+Ap
z6@2%gKOf@!{60J@w=zSTYg!Yg$P{PtldltAUgj@#lO{>RlcWHIz8??gQ!q80i(8Ap
zjMseU9L~xlA#(@k53ml}fK;ad$(Ndto&FdDn$B=qr)>$buUI8DB6?~_>Y*2G_IP`n
znDLw6Y;L{Sc=PHI!qcl)Jm?&t;@iUh`p%c!+uwb;zjbintDwf-G~Hsw;WC{j^UlsC
z2_3CXrw#YDo8G)>Y7Y(U5-{yGVKV_xDYYeV@s$ApblNa-LbO^YTRTd%3X1n++7Xup
zQ!H2%!wS_S##t;}8(IhZ4Gu!nVNW(5njCmi{CFu`DJwbPS;N_A;r9WJ4Xy+re)5uG
zo2Eo5%mhoBXFk@u!aW0uor2242AiEAx*cEInTcG#{;>cfQ|=$}3IaJ}85joa3$XAT
z+ev^><9UQHdAGTDmu6d9w8yQDF__yx<AI~*gHp$GaIliQ2M4WXU?}Dx2%-^??l9%R
zc!@U$@&fBxC!)lfY$jag*<~bJ>G8;rikMbL?*Mf(&@o9uW??}MgD5&<u=>HyUj27l
z`v=>*uT=rEeE^Tmms>j<-$6x!mwUShhvhR%RCU2=B>;C_5}letHHJfP1c{iW4v$}8
zr;DD_Q&FFsiIA+{ojj-FVRT)>qTxusYfQ-O{Ro{x73h*>h<hBJCGf6zPXg1Ffk}7G
z=NxtysExRnTY@P%C(f9HPkl{-{DE!#H;4Nh@PEYtc4PnLf$YoRu^pgCc75^3=^@^$
zqXns{I@~ziN(*AYQ{$z0b^c~|?{Ir(`@f(pH+44W&bz$u;kxsi1Na?nzIheB_lI5g
zej|NJB9=F18sCQwUj&okR276~S9A}-df}Ni1L-0iAf0gBg=LSp!NMaH{%dZ4;(|g)
z4`>Z0SL&Q~b4Yp)jy;OD|L_0!KmT`1L*DmFU40-IMQI2{sZn*`9)&Y0#i)y;cXW+7
zl7&L1goikF9IaIki5lw8A4WE1E}=uHw6!(*z7ksaBmmAit*&T`PI&Yjy7IC)l6-ET
zif;{^&?PP78jx@#jfj@#$Sc|DWy_#Cc4VJFAC9j6l012P<j@J|>2UAl!~vaXsgBp&
z6x6+!#%n-GQZkf*#bFAyPOtiLGN&&-?A>%0#5j^pZ|Isk`7qYOLve{X?J+19>>)!R
zA6`u8HzgU<ZWJm4G$Z+c=!<I@e`k}I^eHGpPF6z&|Ni-Lf)CnMIj@3|SR6ir<$mEL
zdLf=@yZ$ijzn6KFDMdp_(n&P9LkY_;rR<Edn7FVbpaI78V#gl;3E)S;czy(b195ro
zL;p+n5C%K&7(1vQUW1DX48!R|n(c`t1l1$r<2ovg?E%6P10;l?0*Yi;J#m`Os(ZAF
zP1Me5w~@773dF&ZjPVDQd?^D*gt6`(eGm+cHyFFS(hvlA3KHN-Flqfr78O^8f1>D2
z5E7yG?UDN-<-O*W^v!{i!eH0oea1(mcpEm>5n>_3bR^BcO~xLndAa6+)#-=PY#0!i
z3;>8zmY{Mek=i3BF>N}w>}^B)avQ4IqN4$07=iG>E#MI^tn%&A|Lls+JOChUOfV1N
zLIXTii8BY244$WpR+&uC?UmhCa2jN%onY0%f=7V{#!;8JR%8DfyoX{m8bx#u=0QCB
z!MJW6yUUG(Ot{Q`i*Ps7sNzQjJ|ex_a7p9+Wa1)uraJA<&Z*vDHONa&#U(KSM4$S3
zRYT~>AWjjJuxC)6L6;u}!-^52I&xXdCm<gk4Yr}APdBKSdgTVok%x2br6Z8q-Xsc@
zgrJW%yF!{KUGlaRZ0cPOmDLZqE&v-Hu#+Ay`{Oz4z%Puf041p~P{{x@A43f64<&~!
z-HYT#>~s1xVJ1+u#wnYr#{A6c_#6%{ZjwGxUQ<|xzJ}HQ>z68ogtZW_+^VY?z4B?d
z1)&#-=!HD(rQWSB;K#D77S@E@SU~}@hyFsV!tK3xS{B^Gt$U|)!=?%QjPtDzBEVQ?
zE5W;`R|o~P&|rI!DGv?F72@27KnUueDf>Z~63+lrA?DEd;x}L?w{Cf%5iAuE7YL9N
ztpY?w$DsxeYmoz1Bjp;nq(UmX<YGF4WbC;6fEht6)ub_<s>?}6Bxp6Vaafx|lmh*8
zz@tQj5RAwf7>!eh!-BZVY=lLHCs2RSvG~SGpTYu~E^b)Q(;-;|Tcqsrh=%!r$}<7c
zK`g^2Kyv|=hQ%wN7V-EExOI+2GmGxXJ~ix(u6eowy$A`{3Zk}wby@H@3tn)H*aRFK
zQGa}FU@|L^?WcOwHvvh`cl~NS{NQ?2DJ9?#0+~w-$ISwa#=r~+U|WYokqzJy;+k+A
znjd|17Q}F7*+*#S8B=>RjuXl$J)H=OA(p}~;_*Ie9|F;j4@NYPsylzBx?MQ;Pm+lP
zGKg*t8i`hL#{xv2QweuPa2aHOYyc;W!~pv~frdmE(xeh|KV!)tLV@{%{m$lk@V)5Y
zfKE;jh6-CMb-t8Vk}gMMvS%rWG!5=XCenYn6oByqQ37()PwP5GXAzlfLFT~`!!2yP
zV`&kkB&ZfzX|6cepW_P{I_O^y9KsiO=%91Wh0zD4qyT_YxU^Jjn+%(hr_d+37$kmP
zFh#68-oxg38G_LjDN&qdsxCc?xJ~&(2-gVTE#y8MUTO?YagqT$#RFZGUL@d=&&l7g
z%!8Up2`ElFHtEx(N8g?YmO#_ogVQy*%D8$<BiXyI6^v$*CBf_{lEB?XBytslgq|@-
zc9?Y}>lH?8$4WdRdM<r}bp+%CoshNy9_^?`5<U9Do81~jRk<wt2^#VxsTKvH3J%U6
z@QGfGsBr2@By3DO?BSQ+Q-&U^P$rx#>6Ii0&H#M|57X1Yao$3#fw2X=w;wL6tC>lW
zHnhw66`Cc;@bTu{L7+Y2`R*tnFzi=UuOQ))M5UBgqDrZ1C?%f(_vq;_tjgqP1Chy4
zntVL;uf(Xyx;3YcbBsHYG&-`u<LG3vn_sY4Y(o%tShd$5QXZ(bE$||tcJ08VNoQz@
z7Ij<-76?T2IT=(==7Zr3=SyrZhnQ<j)aVq_KJ^;V5<~5q3{{Q~(T<>d3^+1UK+wNY
z7ZZU35*+hjBA}BYj1M|l4bB+N8&_I5OvuP+lqRP}Djv{_XWpy6-y>?*aNcvv(4CkD
z6Yy}JePv5ROP5L=+hQT%D)w@Jr2lz@5CI`nq+WjX2$_tagTOhGVPDvfm9tRg?xX>7
zg*mr2&@s4@S1sx5AoX#-!jb^T*9}@dU5?&Gh@H)k#fBYd*ux^)<g4Le0Zf+@a9db>
z^hBdVgZJgfo{K#gnG*09h~QT)xLGA?Ydnd-up~S-z(a2S_{XWxW@Ybj2P!9&_(PC8
zKw_3=&}O|rukJz(`tHztaI^`(r2`r|kJPTl3C9I{Xk9m02fD(BF_SSzm5d=llskhA
zA#Zh&LMgWy>=rz8m`&hFWG3H5G+6Bn_S^9T`mRFaBu3DyEGl#?4)apV=0NlcY)dp$
zRBDAX5gl{xU_R2Srf1jr`~Dov9;g6vNS=L%RzRU3%)sZ!5=n0M(pr$RRW}Gb6y%}-
z@n|GVg!W%rwt-np9y1q0xzB^P+pKitD$U3-dxooCuwRcOK?(^rX3-|dAzfY2>rh7p
zzdJQh6^BLO6TN{{jmQXH5slgeQdq}gy)#TX+JIU|l<Ts@6>yIaAkn+MkxVHfi^OT%
zK)rt=6y+MgUU5`1(a4%LosI@LzjHQR2!cVAu2i*PifB?qh)pGcbZR=B>2W5J3~H(t
zJF2?F9u$gl_BpbXq5wi!0VMI^9KuR87%mR!@<=^gsYaLABH+gyNlye&^fzrJ5<QBH
zD&nbL)8AtaI@vSthhX9)H&AGSARRMe!|kN*mlzMUV@N@jW#tPV*l84w%;B?2^#u){
zxbXUBZt{hQFjy$K1q6n|W#E;A(WPYM3Z~@D?;{gLnj*4M3mTEU0%xdDUdP40P%Bn?
zxJUmqZz0wT(rjX*;XeoI>De|tGuWXe$BPNJm@Ay*AVu^q{*i7p7P1iH_(!>}I5G(#
z(!qcH0+y{q){~q(u_@p-asSqrmoNS?IGY$Vz(_c}5M)Mn8Egk)1q?QVkN}s()a1xi
z9h2^9=QTL&p|huLn48d%Y~tjmD&RzBiS<vRo6kqRDC(1K3vj%nSkL&r!6-U+J=6~|
ze1Z&DnScC%SE>W5JWZR5MP;!iL0clnQZJ$i168nv*Z8Jd9KbZ9nw|Ej@mx)#vzzW>
z8k`L<#CNM%*Y_pkL5u?JHI752@508V{40GM?I)=1tjokNKP%wMlKIL`G>;TL#LfJQ
z<}@&L>@WC&sDRlBx{|bYtaHf|3Z6$#JM<v)HJm06c=6b1)^?R#r;Vwzx2{Qz#E(-p
zuRwLJDr`v5v}TbEq3^@>X4u3{-|6InF;saTWbQ{=Ze7Sr(<$izhUYtd8Rae{eS;)0
zo7eP$K4xUZ4Q~z6l2K^m3T0OuhnyV315|8RKm))hUF;#r+HH6qE}<Au0m?X82(DH-
zLltRAg_Yt$wYOr}rQ||cFkjDXObAJu(!vO&r=C-O_};8n?b7Z%0PA+r8ix=eh{e+s
zM%Dn9Eff}TDP<s=gz$7b${lD_o5J@TCq>Ygno`U@t=6sp*AeGQd><q3kau_oESU3b
zk6(ghVPNy`7r}{C{O~j2EF9yI)JyH|RJElX)cWmIlo3Bnp>d4V94f+bA2h~f5+2qL
zH9a_JW6qrh565Ai4^JU6TvQM^A=E;!qjWsRk*=*@mo%y8(w-VX_V<OPqWNCsx*!_a
z>#%tFWFd)p8nfBz)z00z#y7GzCcGJ!LMPJANlEMPs54Z;@+})wr=3I3#HnqQvpcTL
zM1F^=`$@>F(D~<8%MEDfQSd#W7b}~ma&d&gHS#>(fG6ho>^ObItU6a&U*uURP{|BT
zjiBaJJiCDsm?{Z%gm4F8r-e2lpQ<X%==l(4k3xU5`AB?>1wye{LRNeal_iz4=)fJ)
zjy7<D+TmKqN&1J6ACE+-2~<E^RA|*w8b$$Pa1VSun4HWW0}Wm|@mQBBo<HUht(P_x
z@kY!yPNh_n*hMKF>`Hd_g0U9cN@rEjq*If<joxXeRyidMAwN`3%gClQ54%9lK3eN#
z+vNr~9*_iP_Dq$?(if?ue}R-N1I#KN95vZ0$q|Ex60$jXEd<tE|LJ&ijT?xhr%29g
zsY>@-=#c)TE*oG9=3U&<1e@U|2EQp0(z2Dp0?}f@I21H4#6LX(Y}CG;jAo+xKZN3n
z>p*d!Bj@sB0wYl~KzYv4HkH-6y-+=r(gi!@2RAK@XSg#~_>(qeMwI>lS;J^naT4Pd
zg6c`L;x{Cuz-vIA5?`x+7)qXS%h=i0rMSEmXkGy#r*wKI$zWj%A<FF;t%=73OIDs}
z#MA;XYpen`oU+KA`dH?6mc~-iM8!4)os-2=M>>q_c!MPHsT19XGePhgymes&h3<}q
zpBYSIm^ZKwHYv<iL}<}19D!(NV0p(kus!}|{YmfZ|J-{5-C5p##3Cr0Q$ljCsTHC+
zJj?YoE6a4|;`s&x-Y!Q;$G^uAVfF2*{2hwH543sn4N6KWFFL1u<PJ0PMzpq%9>rjN
z{D?^O306JTbHc(OJ@P%7jArsQgFq4PH~Lxp`KV&0{+R|SCe*u~p5j5D&IY*7E*tG%
zVL0sRml%nkJHlwQaxsC#0~ONvJ=2Fidb-5qbaWovEC~6?CN0?l$jf43-1hS3iBm2_
zq~r!B>s`k~c!N1Y6IA=?x>Z|%6LL!=9OPZ!?z8F$$q>l|A+23NXme5j**RFCa*g_6
zr{2N#&fcpnkux#6CdmtvZ1}Hpy2=NGd=SaB*jPL9$k?fBk<}CcZU9eCb+k477DjEM
zo(v=fXFygTW^69-26l>!KOQX**IaOFU4QOXZ?5Rq?nGr@$o7Z^ZXZ3GyW+kiu$+yt
z$)`b%ZD=o`JSWluyJ~4K0i`_O@S@(WH}~qVNX*UYzF*(I$vbra^)KK3*Wsuu7!%gA
zfU~zSIb$B4(dI~5a+5WdE`a>nyj)B<R3Nc;29_f8z}pT2b8r<B&k-{9_h*nYrsX(e
zLKQ32toRiuVpO*y2D87cBV7Bd-jnEj&b7Nzeb1D5&T0x<{}q(!bUMT?-$SZS*%p%q
z8vVgU0hgMQnA9L9f=Avs!@dX}X68|?%hhl~4}T(0WkyD(ibQ%bGv8tu<PFwFS2%@D
z;J~EDN%MW~Ux*Quk#502!;yi|94B&N^64+NB1YV5wZ|BB!`*6C3U?R$EkUpy8JAa{
z>}iAonW>+IWNd1_5nqI80~FfoQ5}g3eoRiHsl!fH7M8;}UZ_b1u&tcJCkdet!oYCs
zu-M(Ob{%R$NO14P8)PCl_;zD|(=!c{Up~xu**80gT1iGGTr%oR`l(;;Z*11PdtcID
z($K;#U+;eT=&X$tkAe^jG?E)?h<%8=%?ar-eDZr>xJGDq0xYPsleNGzm+EMQjYKR)
zcuQ1aGV7RZQ3j@E7PBrF1C21Q!pO0vA-Cz(JZ|865GRe<2Llwynj@i37cqiy6DMs4
zY__Gu7Z+DlA5Nh%NKh&zFW}IG1WM3{YM>XBmfsb)tc?YS_ts8R_*TS&%-O~#Fwsw;
zrD7g#=Oh5DbHQ-#l-24%;K#Z`UH|nJPF9ur5K#B1$80onG9Tcw4S3ZC3RnU6sdn~Y
zeAkONkiMV({>ZZ>T1!NAfLlXaq>tqBh{^~a9>>+w3W?ga<*4(ode!OG(?33PX=kcU
z^YjqSMC?HVVm|L*sJjk*Erp2D6?JX&#1ORvFWRJT=>{G+L4CQxspozQSY<n;cbyj<
z5{)0VWYi2?Lc2N+?yoD=*8mhev^qY0`bT<U5Nh$1^oY=EjHe|i;`js2B9m2cZ{l0^
z<dQ2vfh7d{MTq!)&*V=z&j7o$06`_Wqp_$Q!b<fA_{VUwE4~)$DY_+qY-QyXUwR<0
z$Wj%#G?;1Fe0hQdOi0Tfo6WafJ)sj_8o5vVx35NQhR3QE^g_{KqKdCnR7a0$Ve*?9
z52tX%l+9!~2&X15^17X@^zQ3-Ev<{}1Sx^g@E{e6HS~51vEEQAS{#xOp;`wS7Lr0y
z><I4k?kJ}g#Mm3*dnk-ub}6LqaWPB*P=O4BSnz599i+d}Pvb!tvG!;18OVNqrFA*{
znKuJkPkyIGW4{NKjDMyD{YcoJjVdr_+X{-bEUTPTc3&;5j0!4cdaH=Nj5nQ0M7?RV
z9l7h>y&iI8BovF_nAcHoy+7rD{hueJLB%QCa}n{nqSc0jARW^y(`(ANWh2a1USPsp
zdIG9IQ){oh^4K?mJ5V5kYuM6}6ji1UVhX>4e;Ro`(+809Boxc}v^j?}-)vqNvm`AS
ziuhOQyM`ktN;RIO#i5dRZl@)$+;ys~+ux^6c@Js0ATa^`MQ{|RHQT>V=0fQ}tho_C
z5K2=<<g4*iVf=!K%ln85G}wq9R}%}Whn=&{PiGGvlH6s@fGXD9BX)~Bu;7aW?^^~q
z$Z-oeLG-5V$=G_1Jdm?sWi6>`J3m6n>+DnJUFsY?Xv*a->ipYOnSbQc2Kx4E6bX6E
zEOM-&feB2C^k!*9(sUNO8wd`4<1vQQ)c_e#d>Aji)GPBQsVHDRg@K`@(bKtgk`YLW
zK_0k{kBd9Q{xK{*?~|v#&-#panEIYQxZ1nEjr@gmHsx?h6}I%Mh2D1L2QFG}(d@FS
zZ{BexP;Rk_w@==+jfKtR<H1}c2>ETaBwSx9(%H<CdX_^Gswf?igeVR9VJP$(TSHNj
zOUOBNh!CXnu|_{e$dsUK1kXS}p#G__N#x2s!qssJEulpjM}6t7ak16?;Y0PUhXW=q
z!(b{`>)p`>0j6&<-$3A0VXm!H@AUv7IkKKGv;+wvsv{^(DXxp+pl-{>BI~TR2PJGs
zx*snkiHm~ST}7BCI015#f-V{D?Lr1pT+4EWeUHQn+<{5S?OKiXNdzWqv|Et50<)92
z(Nqp3|I<2|B&7?xAZiARi1_5KjMOj5FX5;Oe{{X$+&B3EJYreAFoiv{Z(34KjL>v4
zB9x<=JH`~S`D^o4FSlV7Jg4*MYB`Ob`&3N7qo8zK4r850nO<zQpk8Es3dG_4Hqk-e
zDR37G06`R`8-k?=@OPo8pHpG~)<TVQBy%*Pr{}KLxB4U{CM%CH7pX4`#(nn^rTgt^
zt{@V4D8gu^Jy^YcGRbZuBP6Ls3sJ8NJskZVYG%K7^1DlimF1O6?4#c^I|NRT(4dmv
z@c1tQOVYCQ9aBp`wl%m>@diAcCk275a3lp$0&jytQ)|0=B0C`2E{@UzAhdB(S*f1O
zoio)5;$+24NtGf4U`PTg21w>U*xD?|1V+x2Z|V*aaB($9l0%5oNI#tJB2t9%RSny_
z5M#H0E#SFpSjdg+!vx|yR0~X3IFVytpv#l`Bm{#s90c`_PsYsCBSrQc%EYtlfDyI|
zu-hb%6iyZf%SEz`!SEk`p3G<6j!1=oTCmbkii^@xR!yYA$Cg?$aT2-E93yN1Iej8Z
zNQg<LGRg0s0v6w96Yw3}Wf$IS!`+q(hIFbcJvk97v-Sn-TqWHk1#x|Yo)L$7j~vuo
za&9hi<;a~&+#HUQrjASi#R~8VeGG)EYq7(~RDslrbcloB)DhgvFp9c9TBW7eFI`|z
z<^Yl6NLK{bom<Rfmzho|2W}<N%poAl3%+!>f*J#hQsbSmmm+_oYAZBn!mI&9axjji
zSXoM(1CXdo2$^03trO4~!=GVLO=%z#!!(<dKbw1(G_l%JxrK-eNS`$>K?`G+butmo
zgIf*|&^H`329k2x)?s=4bJ0+Mf>V)8rIHTM!X=|zD2;3<f%#fnf(#M4>*E`TG~eN^
z^hSB;=+UD_eK1jpI^f)hj?kxs6#by2djU5_I439Tr~zY)Bv~5BH+zQ6iM-bjt3uML
zI4h?IbfwxEeFWR`E?4CCb4{hjkxDzzR7yHQ&!PE-AODsrT4rWkP(o8T`U1B>AOk*t
zYJvjnRPR~$`9XbUWt)p~$SI&yTYmEb%F^HI;X<T+7imTz)OYkO3UMNiXY(rx8w@5;
z;)WbAey$z_c3|t9n#6QP@NK>-e0Pm;A?_`>ZMv=%?wjGYd73d&ta`DB!H}56PcvdQ
z4xH9}Z#j6dn=)esC~`V{fQUN)(WgNI_qlywU?K9`F=QE5VOc}6+At%afANymF!8t$
zcDd;s7kdzt_+wZStu^)_WV55IJ#Gs;IX&y~X$~CgJJ#sjsMEC{w}09Ztf<a`A~RvH
zzulAceH-q&g9EOuPkGm$*UCp7c5vFEgLT$;6BE+1umz4~af4W8a>#?2R05_9OUi_7
z=)Om+t$bX`ER;O~5#4~5`)3+-pXB5xvjtST(WYVxVZgMSuWiyj_F~p8-D1dUztXI>
zIQkF*nbk#0u?;0|@p+WA&6dY~Lisqc$&}^H!iv$vL>sI+!P=EC^i1o}?YWaL6RYWz
z1t_Ub0q#`ZLE_xDd9bSO>yNm~dDtPuL~+3-J33<RBZ)(R9Euf|@J=2L1Okr`HfbLb
zQ5Plo&4+7M1&?jmf+(naPI-FDYr$H+WxQy);&vy27m}Np1C&Q@@P6mUu569k-}bI=
z8<5*(qZ?J(W-L3NI={qE3!3_@Modp($oeOj3S|NWj9)&PD-}<6LbHK@c7kP%IucLV
zDln1?w179wu@p`^<t^L2RJ~L~5!(={=pJM`5Q=tf>lkvS88~q7EVDa`0oRW`nH*aS
zZ6t*SXp&B05tgJOv6&mX?5`9A`P6EKwZ)2yRQlK4*4=4AI4^0%E#_laAOV?e!n7<F
zVJ}|GNt81O#TpLR$hB%++g|}8pWY^94bZY?0BWhF@`|;>iOn?V@5+@3uyo-P3a^`m
zI^+^=1c%uWg|A4}rt+`7r$ydN2DV=j5Q3Z{wOuA5FwQ0%0H*zo%DU`D@D{K^*x!bP
zXOH<e#ZIx^MHpIM?O>e-B#G{re9++`3^AO2t(N^w!3{N9VVpGrg{>j}O&ilak1MnX
zpn9!s2c^iOh3+<woLHn&o{8N40=9UUWnNsy<t+g#hAb3QRL<reqk`Xc*V+zun6(q8
z+a^Sm5#7bc-qsJYpUD@ui}@G}AB{o3Jf@{9Z>j{MQ&|+thYufmQ6$TlrojGV4yB-*
zX4QPF^thIs^$yXV4Z+~Ox74BT47_keb;HTUNbW2Qa;~HkrV(unk4(63LXoFvWDwlN
z75r`Vz!wBx%u0&a*8R)_%E57$^24!hBr731D0LwO8Mi+mIjx%uDzcU-j$jO5tg&fk
zQjIT20Qu>(?V~)J;uLV%&=ydu9dKz!)*Rh>ex;%;C!CI^L41&m`oXj62|E<?P<q*2
zGeRZejVskFMmXX?!>fSejn6h8!LNit*1ixzFPNaX!kM(8(31IzOM~{;wGX>(s0%4%
z1pV~7lwHt#Antszm-~?!_fsTg<h(2l2SIT|y+Jx7MpgBiKLT?aDsZ6F{8cx~FF^`G
zAuI`8Encjjzl!TeGa=BrOlJY!=bHK~;`MqLr)0DJxj!9I0ZiM0WR*ZX(u7_GnjF@_
z168}TOCYXwemSQ$UC#PP0YUJ<Lx@Yw>P&|_9D5|LKuON~UG(Vk1Rg-t&@*6MJ2M?V
zl)fTWhp0*&kfy1YmNBZZ(By5nvC;|DD3YvNDaD;bF3c2P7nh)`DOsBlOKFdh%CwzD
zSJ9}JT&=J^s5y|B^psXEHCb2{+81~ZAy;#eG!{{|6!9pPw~d=~0BhsPGVlg+ac!#A
z&@65}<ZTiQ4QwC59Oz7DvM}6t*G@!@1u7HY$oDuy@E8xDBNw^svxpsw5{yos=}4V|
zU0%zDUKp$3MieA!c97>m1X`i{K7+3#XG<Ajo_Jz{%WdGgqOl8jjkJ{~u&%{zIVnf)
z>+!H*CKU?0-b39hxA}OzJ{Gkes@#kA5HsB(5*HvRDkS}eKr6F8BcOoCw{t|rBJ|ge
zk{~+9=LSn9#|@i_vz$izlij009<;%zUSEj(fckQald$}lNE!>ATX1;q5TJu4K6gzY
zy$d_`2F##|9B2I_N98pxQia}qiX?Z7n~Pv2jQDKaTE`iO&^q+SW_uv1F{j?*?vVOT
zl(K$?wNGzjogv$ma6)YsSC%5!1uoA+Ug4$G3^(FNV!jy3<u#T}mZ8bfA1Ct)6l#xU
z5#BnPdZAi8pL&3}$BYU+D+9;A5SdW7)g{{iGm%L`{h+xQmJh!dvNWVBqa=ou`z~Jv
zGCVSI$RJ7>_b-}WFoUw3)L37l7Z)OljhD;LmYX3rN}9W`h)tH(t%Qv<Xq<YtOfflh
zbcKBd98BgbEZdk`bq-a4wz_^}NQlzjMnv04Tl=fiZ5YzOXv$JF|5C&i&wiO#HK(;O
z^JOn;-GrP+j_SxG?k>-<tE+GY7dKbJb7B}x;noU$s3lFo1_l29!>bJ_CF!=w*l!3S
zV`Pd_myc`e%3a0?2q|*P)a<U3n{gy=tS-$f`FtXm(XW!WN^u6K6=2Uj*umkecewXP
z*rM5B+^9O}vUWW-RsU{ap@0b(WbkzC$qmQ#=(JY98d*FMwhYv~4!JNnZpt1gisAZa
z<j@i@Xy;|X9|C!tt(x1t$Yo*yCkKLHqnA<A^HH<eL`;bI3YBpopf=z0&s0{_LA&09
zpe(iNhFySvU3g#DHx&8t_TCR~AeW`>xYH*y58u#zfpYRdbM8>UP;q=TaEXtdJT9>h
zIPd7q-Ox_cyNp!RkM_4kQ2x%qL=@WFj&uUI7yVMBx6V6Qy=4JK=K(-2<SaAp*3o>7
zqY1B;6;}brB<T|aig;Fa+&6zq3pzpaWpUhuD`%825xLeavtmjoViY2-F`+ZdC7$-0
z(u!mHu4ljgWH7hVX_y;7fi^ZQZr6qpoxKTyHr^oZutx`~a3cg(uDcE%q$nb$B#+c=
zmEll$m-Q9V2$l+6ts5;DoU8|2gAsDkc?Xe{h?5Bjx7DRbv{QJ23FG|~75PkvQhiHo
zD)RUwb@8J)tWH6SlMiHX<}~uD3D0M!d>HN~EDVA=lj6+~z9*jkwpw{`PdnE>jy?(Q
zbv^o{tIx;v*)A<nKY!^$5*$0UJ*z%FSgRiNenMvJ43(b?ipA3ob6qd01O}rH-$<0F
zX~3D6eOWNE)-p_XBu$N6HP$~J!cSa*#|(id;w(?|R|C}OjAwYq9i@6>7b0Sq>Z0k)
z1A>q@C3<`pQF1eU+^^&~X1o+7{(A=5r%TfMB8Ae$yMLXkpE$=nQmtD-E%dg-hn;Y-
z)4CfZ^Kr11nql8(v#eV3wiBlvZ>o!1m_@E^OA8twa`0p%74b>BDWe!PPPodRN^|&$
zituXOBcOsL4*nR&y*(3uVJf(mV4Olx;3<FcmBKVYF#4Xic5-WE4`K5FmFTfq)Mx@6
z*Rir#Th6^$umTBv30L;$@GIa~;ms!a=}zT=0Gr&G33;wB@JHq<Obe+qnG~Tn-#}`A
z%NE>v_9U&K%ynJ=q^}~bydHf|3kge<?L_spZ7ed>3VWO*>K4p2h*(ush-*O#i>V59
zB(+^jd913Q$Q$xNi@pm)2*Q!6)~4KI_j?kg9g<b#PMQ~PFcvoe3XFku`CH8}wep4&
z$d7d|l7uRyB_10-sg2}KaT7W&e+62khGv_RP*r>(D3BuAxeI0+A-_EC!GdAPwIvS0
zboQppT-CL_A&iy=LUCs&Qrlx8hMt+4mj?i_48Ksc$0Hd4-ycctULj8Y1S+}1nloEC
zL2A7h^C^};^gI$n<F*y%=M*?iw1gxbbs!2T_zR|OC76u6FUaofTY#gKeT{HmUO6(n
zyh4Urm;owrXIR&#c>l-WulAsy!|ix=ceRH!qnAyAIp@t$4?;159uHN5JdfU<UNi@R
za7eu{G&zBQ9JGbDQWElAkc3IPq_(E^n~jC0;ZZstA>Rzyuv<z2vk6ukUcO-wakMB2
z?JWDpmKt3GFDV2Pm*sp3=?=qsc6DN7QLzJvoM+ePG#{jVkni}(n|&M5rIp?OCafvl
zg0+dpddw?{{6fh;VHb%2oy@H@O9Dw3zP$;Yk59^k6h@x#4pvZBE0?p$sDK^CgT;F5
zJ=3r^ARJF6d_*#On(+ZVIk{wAC8V&BRS_`924Bgfy<|k^aNxM_g6WfGqPUF=6AyJv
z^3qJQwYQq#(5Ar%VLKrJ_Hh)cJ5e%i<?;^M>+Wy4P!sLz%?IUbeyOkfZ+XFGD8gZP
zGhDaw&kxZI@<9C41@1YTMMmd9SvhJrr%Y}b6*ued2ysLo%QgO*GfGwpXrTAa)_ugi
z^JRUn#k4P*kgcMe->rqMIuKZsoX~T=00W~;m2?oq-!`>$YpLEYx?O&EbL&;T2M_gE
zueM(0{8h~rh(++K6Ru53nmes|onM_;7CG64ZD6H(173=|jqS2=C=L;}MAv`c2DzEU
z;86#shkJ$q(h>@llqxKBx5JhP%V^l|*eO5x0&{5p9<t9Xbe-dn)e3Wx(FMx|<`$DO
z?uJplYi66xsf(@rYAa|Vl@k*C<b=*Qt;qZ`B>vMFzRWMxk33a}{YA+-oJlr8>aUpJ
zdK$#0_yD`_Y<rZ#q!k``X<HT9G&;hBhGNUG#JEE3Nqe9Z;<(8l1%M!39R&}k5V(d|
zKQKsy`r&R@GFOPXzJ;#llNR@QMpf!1&B2$%1B$L-B%nTMdV~0NiuCE7-3RV&HBs^V
z(7#CDC$#X9^0zyV#y?TFE;nYK$la6BL}TTSCeyH5V+5jpzWT@rD98ky3q8sO3rX0d
z+T5Q7)cZ_<a;;prFx`6uu%QIIUi4yHCJ6i0Pgdv8=T<a%=!;?`OG?s7|F;U!NofPY
zsebBIeccQzjE?&28Vd>Xj3NT^b)O8)cXqqxDXEtzn39fDWSK`o6YOm9G$IWf+l!Hq
z2%13-L6B?XEMA)XM$_UGUFL7eJd`<mbAMQJL+~45w5t%I%{{i<Dfy`rr-pZt$Py7O
z>N$CycNxh?vM7ZCN`5EhlR5HkLEJr%rdrTpeL7yGKqGiW4oCtqoH1%5<VFSZv0i~A
z=+!+Dexl0ZeOfodkq+K?yBU7MrsLk233LgRE<ooUQR(_#Phd<(vK*%GN5B#4%jk^?
zq)-LCmn(+%X+5O`_2L8VP(E_h5z4u_6U!v0Ns_eLEODzsCIQ5~+fMOJNFTejXv^=S
z!dyWb6~+OH1L+BD=`nK*!Z!UiFNSCRlN-k5xoctJNIPldmDTM>SZCdcT}rpnh{O(X
zDG}V#kpUw*k{&p{tOth`>=O?iVc_vukePUp7bP%Y>g43uhd0EeTqoYYXsA8ea^bLp
zt+X7Z_179qak$9<8D5GM)jWB`UdM(S{?${imgD!fa?OP*TNknx*+D|k90~>Mw9X0`
z3nX)Ut_>&JFP#5)s+--Ff?~2ZjgZoYJf7RO?OcbxWXpz|T&l4KnEKhE@^JWS(W=f7
z9wsAVId_>clG0tb>0~R8`*k<C6|^T2`|{i?>{xQJX)-X+v^F7gfRlOe2^wYiY}o@4
zCweZoU*F*f^w&9JY&2bN6Q)hS+tLAzhpn~hp=QD8Up#CGnxvF%_@GKwLHZ1e1FSz5
z9mxuesa-En){tazx#dit6R$5n3KnTm`>S_Rls(9ZQ|=B&ptIWKV3nqgh-r`XS~PJX
zNUsf6QyeZ#+lH$XPGSs_yBJE+6MuGWJ(9(Pu)~;~0ka|6+mtPfOKVj;szpQ+4r_?9
zh*lpDYc}xsInE==PjHa*chPnd7xB$<B5pa?`LepR+SJ9q>w=Oft4rL~t>?P9us>-z
z18U<~7Z+_DuJ}sVT#Pi>Ez$`oj#q9lRtc}oqadlbsOuD4{$!Uzo-Mk@>8r8c+IEQI
z!7lCz=xnsS;9M71%=td)@QZmr8@Zp1%$oe%XZK*ed+Z>r7S<nsx$Se)Zw2`oZ!r>!
z2$U+}O^*8I&K?Ai*!f32<;hswA;GU@zyG=y-55#H5%CuE#<fx;xpvp%-qR8Ry^-`F
z#lxGl#&Sr%A2Hv5n!<rD8V7lkQ;~%CVKfa%h)M6{(56+KB;=oBS5A*|f!YHnjTM*s
zvE~<T{}U8>;=z4^$a6^DHb+Q=15dQh;8Y690vb<Aw{j=hhDjWuY4#fi7cf7+ifqFn
z{PO#LjObhU{kX8_73F<DF51|9`RZ5S@m;5W>O?Z~dwmM+UvsaId%DwJA6E>VLg`Qc
z4}3A=Gz&SV)xIB3wA%mFH~7VvyC2q;+iQrS`|ljA9d?IkPH)*gSTEY6W%##l@G+92
z%1CiHl(c!Y5~300v87F0G)YXxg5#VVWWsUm7Lg$CCnaMO31WED5~1q|84C$~xoNr|
z`QLwH?=z{9NU%6%zmpCKohZj)HdALM+=e~Aj*cm@L~sC!K;5FLK)8gOka4%_p@+L<
zr(@cNr;L-G!J&aS1xGefP>9nj(4LE%XD4IvqV~R&ynL22E}SNY7Z<Ym#)elKk!f|6
zv=icxw&sHXn7AFS0y<<jfNw~=`yQ$Z6`2<$kMH$4e7??*O4biy28{^l)8B)@5qXtI
zYzgk7RCW`BL{U7RO#PN&B|Hk~cuK!<>mV$G7`LK#XwF%rNybjpnsZVW7q?`4L|hmw
z(a0sFSIa$-N|FK#wrH2qT%23yV_pj)ZkHZMests-{;6&5{z1=w^A&Pkr7x3J_myH~
zcszD<VqMQ4VJ?j}*ZsEz%R)UN$`Wva-oxbm^QY$}s1g{|D{^eE)SuAqjS=yv#f%X-
z_e!PU*o#^+!G3zmDz*VEM=%NTwhp~EVqIKW`~&qG?RpT`D)@KRuQq<%+JC*fzoXv{
z(u01X&9{sqlpRXG04Z4^0Ngnq^}_{vESdU|`LXB~A27RlIO!Su){{0D5I>)?Cwyu*
zgFGwL;Cp_q6j)@8B46>t?@2Ud)GZvG0eDv&*s~##$z35d=kMpp>^_rZNGYbYuc~^N
zZm#yYqoJ=b=U|dL&vZi7KSAzo*B5!+&-tTY5fgMgFHP_-jg2S0C;2!VLyrQ;h}1wp
zCl?4{4L>z*3pC_5GPN`skE>ByGMZ>H;bs#48A)es-qzfV79vom+_mm={e|n)^b+d8
z4q2C<#)4h1b`K76B^gT!-OmE=Ei8x}H>@Z_j_MH)#YOT^>=z<_xTv<gTuRLOCd3Z2
zF$JWfZ!#8@aY`eoav;(!*#VFq8bOmj22)baZVdXDe>xsr7cyN+WxYcN1!f@6;~NOP
zfrYYweKY|l^Jc|mbt~?@yud{|7rPwFjwI@w3jO#aZG%j#&DndTaYK+yoPx>T@h=5>
zaCJy{DG$<T#ao!HI*<|>6tn^i?c?4S5~~at;TlPwRw@2#upuscHoxqCg3X#bkE?53
zs_+a<4Kl$HHh6(iSfut51($bMKlf-=7-kXXaX<WiHDw&mX~+8fq-}jhkgR+3_^XEx
zzk2d{$I*SHv1qo%-)^qJc+cRr2y?i$SDU=a`{#gEguE($MGW|-j?hre$pxeR`m4`5
zp~YRf2;{-=czy=RW14v8?0Jr`D;o{x@bv2IuUdd_AYswi{d$#IZjlU9?vkxSdLcyK
z8bZ4~YAA2>3o4>qcbH6D(#7((%v&`&Q6r*2XD+e6p(YFEE`}T`JT`zBVQ1Cc2;%;7
z$P>ofjWNR2@Lv?J7mM<-;?(!zG8M_;!WfIvj0!irQ5xqC4KlYQ8LXbJ_aJuypfuMK
z3(ctggg?fN;k<AK(>Nd=>iAMqH^qZFDy%Wlg1S+Mj%69~Snpw)!JQVdEde5q-A_EO
zn(a^N2+w|H^a6r^A0C0ZeERI4gJ2p}QZf;a)7RXRAlF7&EGW4uvfHVco1wGXn<0P$
zw+62@<Eyz;kXvHSUG0G^B-*G|ST8Js0P|Z^rT2hvk1%LI%Wgm7&^i}WELVZk&c-#*
z#-BsAK6ZephQ|eA?BIF;NSx1HerGAlM^E7nP1O1aUp4Xzc91X8I+j8%e`44pTaeC8
zY1c^eH6+|wZ)0=g#m4i)-Tibf!f>&vVXPa2zGiq#Mi5G=Y|_)DYu!&BNi`fKRX{62
z5tz8?5CO%wpIvkj_B>*EA+xNsUIuyR`}*4#Da4u)D_$TDI(6o7>2)!V%kRVVa2moO
z{7_MZZ|oCFq{9^4>!ZN1oA85@-|l8O#{huy=ijas=EL6AFWww%ZAKi3@Py34y7Vk>
z@Lzvm>^)h+ty3c-Q|Qw<ViRr1{M0#x{$~J+U?$*u5QO%M3_@w2xX?O%Kp+AZE&p<F
zb~8RXoem-Gs8m)!XbOHCwWOt%pcqydemogoU02&7?{E%ut3yh4+8;4!8a6ElydR8=
zqB-1%|23(#$RI$F%xGMi(RS=gj$%LAP|Yb?UvQEMRk{dMq$z4cQNE2ozFEgwjqmgW
zw+#!D6~7F<2tIMxm8*|jo<QVat~`j6_E{1vyWnRLzmalk@Y{P}`<vI_q|@Jy=2-a4
zIRXzKw<DU<AthrR^ZP#qJRgsbx%&u!f@MDVC6K)F`p3Qft>@bZ+q<uq#kLp(8IJT(
zaZRDslQzV~J-cy)5I6Rj6rqlLWL54p1<7v37FMdm-Ob(VIZ|PrA;Sq1$IM``L5^1N
zh@JHpdtZOG2G=&8D0!#TTJ>EYMe$2R7bjyRV1q<>!xXek!-Y6mn#|!1!i)*n(jr7;
zowrd<R*;aERYQE!0jH?n6RsMEK;*Z9M2g)3p<vgO<H_VgPwy)I0<dneJOmCIQfIXx
zL$L?@bS`0Z7+=FsBA7xaXs>^b)20m5scNF08s*W6+$+K3)uL`BJ2OzbY&VVa%XY)0
z8^iHT5LO@0Pmygt>#;Z&hL4s}x3>$(EG2)$?W{Tb+k%=PtNMZobob-kdyvR+MsaM|
z=hKgJ>0jN*I3EX6j$priUp#oZnR1Ol_PVyEU1bwJ6!v4fy=lngFZVa@&^)*elM_T;
zG<_#C*#U{lco4*D#u8VhDbCT1qkwI1KK7Tkg*^aDfp8@I9O25H&6Dxx>P_8-cr<q=
zuweD!sa^vy>;xJaDx;PVBPkxw(o=8Ogle!Y23khv-T0^;EIhu$o}^Fp@LJw#=^O|V
zPvej@jow1Ai4GBcjhFs1W^6&uZyDd1Hj`)jU?Zfx5>H?m2L}b-M8gmc)EhY9rR0;_
zxvH+%X$4rkwVcFue%D(t!OsVi&~7~PmFVM1qtMO6_3CT?KB}|gA42RG&CA6!+z4N`
z=;X!Jq~md)Ym}vmP*&W0%{?m|XpJ@aNmIkW|CuOa4K%e+`7(KHL<XGDFXxDe;hmz9
zHvt_V<5vaKWPzXM*e%&FJzhDgAgIPVt{4?z<*@nTM~Q8kIrDyS0a4dKjdKu97N{qX
zSWFn-;!+NfGDrubq+g07Ep@!06&<$h+XBGp0!LSWyH9;)u5~@R4AGA<04Pdat|ouU
z((ofhDO%=vUn#=JHC+Z`J7!R3gArlzKy0Dg9(?q`r+lOFXfoNDJ+d3zV~ua>DrtTD
z7Kr+l>VUVS_Ae;ksDZHLaxErvyVYQTI~;j!)-Wt}HYLhXuX1+~tQ9@DBR<8N{J5FN
z@Fg*33bKJ72<6r)p{UzDyHxgUO{bG7w3-lXM!0p&ejj3?$t$SVff7uEArKSOoDt5t
zJ|ou*_)YLy9L{PL=hBIyj)d)IN&h8wG28}$`J7)lk_Y(xi)YoFF$BW5RGrVqeds^&
z5Q@Rg(aH5&$cU=FeHS4cH1q8q_*fqqxA2F4E{s@=Nl91{yk$rdxVMXc!|BPGU;xUl
zsK5rm4TyxSb=`xo`ofHPM71}EGx)cA2Lk6edaGP-ckl4c>xN=Ot#E)-pe9sT{gWZ$
z&;fNs02g=aXn*6!6l&j#VuJ&*kG|?5b9lG&fd(v~Up8VtMiF&9crv2+D#qq054K-o
zpe_A>dM|!U-Jiwg9od(DMmFf{>8djOpGBGd9=cySr4ScA?8M0XNeO=@Db!#R3Nq&%
zLHvH>^=9L2-oTI_Jk{Be8X8R1A#8*+ZkhcIv*gnK$-%>Kife2mqGCh7vjJg?q6s74
zRauTQ-VNT8(~}{7LQ|gwFHrWU7XsP3;BAbQb09ejM^5w7;vQ*aYa5O8Y_06BQyqy+
znOGqjDOy=B5GQ}}!`nti<9X^&J;4P!#04naj$<iSn$*UYaNj+SsvNWE?>TY|hi+-I
z<AK%r`C)Nj-FG7c)~t0Y39_KJ2A+Txx>jYiE05Fh0MUgRbjFBK^iHd?wEP_bjK-%E
z*Z>bn>5$N6m8u<d7R%^&>aMg7)9wyNXS+~WZiJQP`;J!Iy#Xk!vb6v2|I;_s0j^fQ
zitK9!L8(^}Fa2$o>GZB3<LR+aL+A^UD0yKsC<qs}T`(gIDM>9H(6yQvE{j?Wd<kXc
z=7`X;Q-yweD$8X!-NZm3i3|tNrOUA8kn%ju_s&n@asw%qCE@=s&|#}+X(jlH^n9&A
zez&%)=0#`@v?B}q4Xs8Fcg&@R8@_QrPVG0xd+gQZ<HIZA^0>=h@9lw5>o;`KvBW!i
zJ5k?jM;1&^j+y)i_9@#~sr7)cRQnfQ2O3f@Ufsm$INQNlaz}S`p%P6>1-h;k&pHPl
z9z|O^wE#tH%+4osxrl5&M!<hc_?`yr(g%MT2+|s=RJQarQEV&zsJ<bV&Gx1M0pY%W
zmE3XX%_I|9VbeYmgR5nM7lNfovk3tSUr#3&J%@~0|K~<zDIu5H*IUI{pbl4yMKH4<
z`=K!!6yNoe;Etn4JP5#k=*T2qCvs<5vvqbH&CKFqVMr7c1X$`vrNsbhL#LpExW~9$
zQYqkv8uefVw+pO09f8~d);klIA~_S|7GkJ3&W9)O<(f4XPpN1!N2U%$rQvGe^=dN1
z-gAAfndQ@3$Wq?<1cm#H?H}sr-)=qsZf|${^<llaz5n@<=84bxjybuXB0(Q~Xy5=8
zx?Zc05CzUQV<a^pHLuf#7)XtkjTmGrxSbLd80idb?I*ahcay=zeIM>`Y`@;xuMZA4
z4!7#p8#`OkjRowRA5~qkLAQQQebtk$fWVV$W*cek1wz2~&h~#nKTGxy>kyXh<N#B%
z`Q}yVM2JVUc0j8-3QO*pn7^+#)Ne97nxO{R)_Vxoot~kdoKlgl4mXO4P&~{SW-B~q
z1=mV;y`+>fULC4LYgYJ7c#|~NNUqh^nDEj7T<&YA>@L0XEp%BaxoP%Cx#I0<Pzgej
z>KCYu+7wvnN?WcIvIlnn=d>hA<J@zBOnc?b3T4(mjhr92WD9cBD<w4=P?ZvEUhr8L
zyGq2k9BVScD1{v_gR-DJJT8*PsiqSQ+hBH;-IE7Ug*vLi9A+xqbg5TLHhLJl(X-w7
zlOQ2};WTTiiU3KChaa!IIIK-0rA%XCH>>C-f*ePX3Wrz$;Ec0wQYcUnyUui1eCTHa
zFvtY-7}^w3Gz;+~{2VglH2`(4pR=yZJUzdj&}fGnTKka9xC5e4?VJ<4Cw(yI+J%I;
znZ(qJc73cj)^}H9QA<12|1p=ij8PoKFwEJooBq;P;(IZA^n9ybDcscp=c=TB=V|Y0
z;hu~}pV5&&!6?WRtTQ6SxSoBFajF2~31wVpO#C}29#y%jZUt`s*6pE0or$sm{W+xh
zrE4A8-<(I-8~f|K#WXiTYHc>FoGZ1hi2XJ2#S_KGL{-2uKU^7^I=(rr1+@fLQQkIO
z+pY2VFOzf{VitpyE&^*~G0KQ8T~NT++jp%PY8ybxkE~DoYh%5J;S_osn-Q=OT=+YB
zKTH>~psG9wYcPP@gZClj=(LOW+My=Fpf{>esE&PcHQ^Q>cr=t#cxe4LN@mHtLAppG
zcbI>f1Ef-B%1X+}_l<NPJ~ZqIvE_{tt#>GHaQiA5xp)aW+(0iI+Jb`fLBEFW1^CTN
znw~ut0oE$7TBoxGoY9>IK2pYSGk3UZ-zjJp&Lxd4@pQR85z3GuZgPrbi)7oAQF8Ae
z!HcQJF-^iCbh^{6-?nq*ci^?ufVm4QoYO<ZdYue_#COtg|7SR+F$A=E^i)x1J9xY0
z3QTup?lv6H#O+7EI31mLuA<7FFA)0awv=F8f&mGNev=KH?0}Febqs(`XE@FEm>tR!
z=hxR)vuBSVN6rE~U0ght{~uu^1($&F;SqSvBj-lY&~rwJcgIty3+tV$%bTBSSC=zS
zd?kV)<v2rrx_nw?0+}1X{LLeSCm(-i<jlCky@r(fgy_eoPybk-&LQMe+E8t}G?=W2
z6{^L4u8+rn|3dC6!<4`wijyqXZir?S+Qf)BySY4`Ajv+*aRQwzuaTY2FGM|y{e?qD
z(q&O#&yrqj`Lcv}v|`}I@WLoiKwE<?>gVCbdS~}3Y!3~{jAFq!FR;z8mv9gmO}_a!
zv?;dBGeNEty>EzU4qo9jw#2x#q@zj(?i*|j(^qepy>aEM8yye4F*=fFBWf(Oftul-
z3zxsPA7uFEnGe{Kfy(h(#2E;ka)Wk(cs-4}A~v<opca|ola|bO1nnS5p|uoWMk7uj
zqt1l*Pl0M*IN3HqI&tEfV}A*zWzI1?smG3*%+(Qv(Ni+dDbP!14@F`I%pEcbKvfO>
zG(~Nt$?Ez8?@<+FeiUr8!&RhFLdu2VpE9InI5?4z3W2pq;FjSC=t3!o8-w(KARc)X
zhw@14(T2dL_;F;Hx07JcN~PgJ(q>E|;28|Qg@etkrXe|~uY~#*+|v?ac%|xJAV<Lf
z`or;v>0{x`3ELKKo1J_ZSDTxA!XGzw_B@0}NM4<u5|IH<j_Qj9mcF9g{IPIXXft3)
zgJUI>qVP8YOT@PH2)i&`9v%_*2Ad1R2ARlhk<;NLK9NgY&)~il!=^#~ZV}7Gz*4H&
z1m#x|T6q%#xik6*G%%+xrk}A&AQDOXAYz?(IICAAE<1eM`(t3Bh8?ajDaJS`tDW{O
zoq~XfDbLxkXdqnm!erQtr^Un_49SGa_T`OCIg=c4rRj@gmUoEVM`u@a4@}i{`WWE9
ze~R;cCGcvZ1fvD<p3Fxly^U)&mu2viVPO~ublimF$)3AnSjQ2;WSEk+OhT1iH$@Zl
zf;6gtD9A7Ir^o+Fn~X6W<(Af@?&OBXz`BDDCybA4RM5%F%9YiB)OC-H8aTp4K=7_s
z58qn#`)GVC@B+9V(vIU8B^{>S{SCsJ002$Q5Zd5uB5Gxp6;d}grgTZD_*udG*=WGK
zS0YXpBbg%9hv5TWwL<sB0VegxDKeP8<K{#ECc5t!Gte$x<vXN;5cT|vkWw6{?baz3
zb}0?OMM=ZC$&r5RmpdUcGK$-Qp~;m;@m+}Z2;yst_cn436^vBU6_*l8E+-L|OL7h7
z_I6yuZGe7lQce%6m7-=c6gI(g*yNOZDr1rZP;L~vFJ8Rb-Pk0dgodJ_XYIi=e^P&K
zD<dk&u%)bdsMUAt&AmD}LJ}r^R;nL<_~BXg4fcxnMx6qI;*UZn(~pBy(syDT(8*0G
zAUNoFxoXI)sgxK`KBQYLTgN13uxz|J+^xUg-`Hy%Y)T_+haa0z=FpXG+BlEC4*y5s
zI}Iby^a7&)O@Jjokohjkx}63ItT%p_E8#pq@~(vq^GL+4eO;%nTq7cr_2_$Ec(sRX
z=L4za7tj~i^L)lRHbwJ`gUzu02Ycn|Y=A9Tx7K{gI_46(W^rm@;Z5+i^YNnJX@v9<
zu`5ZR=|boG5U|#ZxM`0~_jqW#sRfZt`yi3>ZtpSM=<~fdySld4D%s(Eph~uGzObi+
zjG|qT3_3w4N5#*g{#4dc^Da}?;{X7s>e>@{eZ}J?B7GreSYUMaHb^(xp)5LrvWT+~
zvE@}35gk@dKy24BFAaK~6)ceKzRW)PX4!Vt?2-?vi4*`w+RO`rKBLjMFq`)A+qR>;
zJtl3T^Gm1~=%y@gF|*D)h)GY%h~%IPK%dUXC)Wt?p(S}rxxCDYSeu`PWRU_B%Zs<d
zovjP)uH?POH-R}c@xo*r-HzZUrqYZ44V`O{+Ql(zbDnHxECvDp#!W?y$aV)aax3)Q
zh^%WIpwb;ifHX$61jGeCwg5)JMyGo0Qb>P+k)!*Xi(isFndlM0);<V!0y1pifQZYv
zaSV`s*K_Zdgxq4|?slHbaf~A2DXt@sb9B_WBFtrXfO=55lBRKxJ(qe`YrFKfebL9I
zqaI_rKpFrD-VuMpbhmf+<qInAs@Yn;%f<sh&|v_xgtUc7iu*MxnDJU4&)6e6ZwDkt
zO+yfIdxo*ccmVuvsvzN=&6g?f!<n<p*{lt<>WP}DAn-O=28H5fnlrU-hD-pb9k)s*
zHk?+Bq@qDrc({fRu1nKY;11>%z-Z6n5JP>IPXd=KU1w#ztX98;yonvSJl?;q#7!iW
z1z-TwzSjug?PMp?7eq(sqL=~X>_#^mGST)*(U!khmbHm%uE+J}gZ91}bKVrsO_Iu}
z%DQ&9gN}IREYvo#%jvWBKB>-CT^+40u*PWJ>~VVw?nJ<oWz+zl;vy#hw7vcX>{{0~
zJ(telmIkK-g&=!eZXqS(G8Zp;a1d!hu9XxMO{v`^^B*qycNUrN(0t^BryFl^f^UJf
z7t7)8OJtEKR_vDJh#CBH)SS)=+gK(-nB{V-V`pci@oPvMA{+?}4(Vy20aa}MPi8K>
z?7v5D+BxDPAOO%8NqGAsoaX4x@>TCiNd{<+r)|&i<*3Z2`=Vscb@Dp_k3Z=ibgLZO
zE`R%8!;W|d*?ZAsLmd<!0wRu)OPFcVPT8-cFy%h1>8~$u$0QobAPZ&fUrWwz!1l|@
z*xayF*=srZ`qz}L%N#jgER7-9S!;Qj`pXtc+<mj9@|Hv3?=MHk{oPZB&X=03SxsXk
zByYPvcwZf@=ZwR*Gb&$`mbx0A<<}*Zn4)6i>gwXAGLv59i7vX>iD(eSCdOQr$iMwa
z88=S0Fi11H7cpxgF+xlmbbs-a5_8_}CEWMPR_MjA120%Kttn?Npg7yjcH8bHJi420
z<tiDq;Wq8JYH>sUUxc4bb$1dQNh9=_fZS1_T_DVQjJRE4NfFamixC;>#!t(H|L|c%
zXL*sxykv-n%2Ygjj&L<6On)wsD{a9?=5Wmuy`ES+1&PW-Qob|hJMRc)VO3NHM<NOR
z_7o`XiN2p1J<*e7+l}JTawR`04Pg<Wj)8XvP^)#wK)VLZ{S<korslR1hD1_a|Ma<q
z1}`R1;~?_FFaS^g;T0U}`~%5blP_zc5o^TtT0V)f46q9q;`d@?Z&56{B|u1W+RkPN
zz?PBWIf$H$6ZZ!w(4Dh#)00d{8BN0I%Ht~1a9qUx2(gVYmt1unAfEqccuLfx!EF64
z+_V_z0Zr9#{ByT+u(P>wxUsjfkHq6!NIky$eB)KUzw!DzWco>Em4M;K!G7^grk=4z
zi89^PL<cR5pPDJVDqRldnpAY*VP6}6+bP$DayGWpAzyS$Cw^iGn427*>%id<28Bbq
z3JeE3bTBCp>xwAX3PU)iX)O*iB5TJg1k*@*X74K{=e+az*%|5}LFMpqv<1!{U%@i+
z`1gPO!{4kx%%2Zo*Jc#>qd)%s>#zUt$0uJs&6-~VVdN<?<{l{2xIz>y20)T#%R<T$
z9ns?|T~5i4%IP|upS*`k^@P4#J}FvR2WlNARSFc4yiW##-Z5fzF&Gc$QzQwSAjqNB
z$CJ~Q*wXpHK7zq7Teb;sY&TxUVeN$F_fBL2uf3?%jRUHvC+h7;#N``;1$qgOh>kL$
z-Fr1_l118Kw9O@`ld(#y4%b*<WE^H_f!>vuxKW-;^&pBolN6)4isj<1rp$nf3cC$V
zEbr`=+|YzXIxU%71JjqH3|SUP5eO}SW*_dYbaje)e+4%XrsgPg3&J`bO%bl*+)c#D
z6`)Fy74eohdB+Yg!!wJzD`A~aZ`8RYY{(l~)}=DakZKul9?@&zh=^L^Lfw8PRp5e&
zxSI>^aOYGN;7%As{0$<xm7GnVF@X*}t$rR&<`hfdPs6->$gA>J6DbU*NAKj4oPI{@
z3YNI2;ev-g{B2QXJ_MlRKtWGH7+8=r$473`)FYB;&nS8{eRvx%S-%C~FoGGd9lVO1
zbD<EUQ-c8<S@Dm(RPxwogrXl2Q~gMQ*`T6lP?#~UYoSR@3F`VxFFH3NHk_%tZ=0n<
zK%s5X>ke~EEH%<#Fq8-;p+0DbNBG8DO&*g%S0Z`<>xn&uDnsZ-bs;&BPMPP!38$a|
z392P2;C>{@!w;X$E;+Eft1!U1iecbdrQp)001c`S=fiQ<ovJ~(!D8B6BjNHcck%UL
zj-2@FPHO8Z9kXsGxy-nM<o<%b3yy|5S|8^)qt;AP)SlCkYcNFrC!%SCc&pK%9@uzQ
zW6UjU6q1jyi4=^M!cAlN3K9%lPB6<ZN*nHMlJ<>hQji!3YZTdPaPXuU0<m@k_jV7S
z>@Bs{XPi9t;3?y45f2{mZ%Xa4d=<9x6=?u0&wJwY=8t%@U%}x+Bk;`{%=@%1Mxqxq
zd?8CX=mo;a=<FO>ydajq^^Ne;@sfc**eF)9xJGb_qf$|Pb|-!m@J9vt!;Al7@`1Ar
zboS+PF7;XpSj7x<IhdX}S<=}=BLuKK<Z&7}fOv6qPiW+&jUzP2740yH1C+X}QZJ*s
zcN7uqvY_kp2(f?%2|FQU!pGqHyhFZLqQ_Q3(88LwC~7Cj?}X685_<d+_FKN<r*jl)
z9uajhJ!AZGsXE<zL1bkAfU#el&abIJw%Et&4b;d-WiXGbLTCzi2skYFI$>iJVDR4y
zY5IGOrAS|gaRG76Fs9>27-rKk0~7;@e9RKE$`}!0sSc46V^E!8UXlLbBNQuCl4pw?
zVMh<vss|eXPsZkY>yf!3R|an7h-~Fhbe#itO28+YtCmCw7dq&2CJ4YC4rbg%$*)B|
z2y946$fp~rLHI}zChR?oSPRK0G^$sxD~wM)&6m+Kt#-s!9zNG}O7;s!XCBKDo{E68
z84U(3OSf8#HEj*&f;qf9IaJ|1Zt=SOk}3@_8>ew%X-}9sI4K{74gyw44|-76P@Vw~
ziBPeRW|xsK9Pukzv;<IH+g#RBX_bh9R3H{vJ3>LJ>v6&U!f1`%k=wY1&G9#9g_SZS
z3IpUD`!FpJYpZ8Iyz))KAq>V#4>>5lpz|<~(;%40+Hj3psb10T^a^gv%z%2K;6=id
z?*+3Pg+-JXuL{@72+Ji>Zdfh@Q0Xp1DJ~M`l@kXEF3>vh2IyKE!!pask5zQhhv-DC
z`qvYtyBZ}z>#2IOiZT*G?Fjh-PrzC5rs}$^us#jQ=hZmua3M(CCJ2__Gq5Sn`e#F0
z+i-DCr1X&zx}#VMBrdCbOhSz{*Jkq^)GebX+F9iOM7Pecx4MPzLPHQah;d^zy9->V
zvhC{bB+zMQsAxG@s7y>eDcz@ZV4Oq3)!h0)CLZk~Ey|hGk#c`oeiLOq`-F9CfZF;M
zAY-VL<5>#HXp9`fDyx7g0r8Wor)WnsPQ#DGlle7;mWcc2o2d?@(2!feE*FtR1`7kX
z`%&oUEFFbc9%_y<T^#N}UFhd&4iiajo&TkqDO)X*(R%rDUR|~*_%(s7@|UH<#Cqvr
zS~Uzb*ESW#`_;WB2<ud5;@6s%um0ss3+YBkQ~skOr}pR!qvS%M9l-8YmKaMVcK5YR
zWbUO>7Fi;~LJO)GDJB3!8k7~i`b*Yeyi*2}awQ~B6qRR5sZcBkPSTbf>Op6RJckUY
z2R!3P=7Ug%YE(I9tIIIt4N1_7h$k+QSa$Mjd!>4bV@?QY@-L*-j9mbVfKy*~rJ`*>
zwGMj6aY%$LLLp@=+EPMjvdW5~Uo<Q|a-VL#i^l2@*xAqz9DON<H!C2sjEC)dg1p70
zamWs+HN%i9sBw5c^U7$XWu7CbPy@JR2v!6~a%6*}??&hhz(~-5$g_-3CIxv=uqh!q
zGveyTXV>RB(CKE7HB}s_Z=0)vzT&!biiJr7I+0urq5JE^Qw-YU$F~l2`L-|nA0?-E
zx%0K^X&`Nb)8(k}$9lqU&vMixpa6|BN9a|@nIFH{OL&cFNY1H$khEq1DJ2ND2vl05
z0=gDp_jfpyUI*U8ChgN`=RYoL2B*jjv=nn}b*0)tJQR;O7ZjVISQ*iJKpB!kD-V65
zt`wr?X_-%CbX<keiesX1=u|JM&w%k2ge+lKqw5D@nc6T^!y)OzKky0@T>Bek%C}NP
zk>b(PpJ#}Ch%;zuK_&ywcV713=MU3^gcRpJ9HcK)hz}jR34FK!&pAr7>bG{VsEsgo
z;b?5UowALaG(qtA5F25Gil}ZNA?yz^`;egnW~iu)=h&x*mpd3#DJ9I9Y|EytiQKfe
zq!>@x*K&^TQ~81f<|vFHJ)ig9;U5|Y(8(E;_RvWnW>jKsB?bK$`j6R+N@212lQ@|d
ztd9dwdiHK&1hYOv;3>u`JU0fOMa#TLNX4;<WFV|!vw}ipqO{5J_|2O5QfkjBJv&4u
zeR9;Dbbb<*hnPvAWyM(^X;*RdL{>L2g=p0`DMEK6<mJcpiCI^t9d}WQ$H3>bP*?}i
zWf#7gmBJkUA?%jooHZK%)|ZbH6L%i+Qy~eEgv@&eCXXY<S6m0+R<GxmmTilC{8K#9
zaT3f@y)EvQzLw6d67-^)+hT2k!D&d_ld@gmrZBoYLn;B0S)=og9zrvbWKkD~<eAFE
z;|{tXdg3fYPYt9sh(g2v6YrDog3BuVFO&G9USGncYy>S%wGp+Ya-1K~0zqj}(>frh
zUJ~Idv>VY{Ag_43JiTx=xH*q+bP?5l<kx}0^XlWmUB|+pub~`~W4h-07GDs(4ZeBh
zdqn(^gei@PjNTJs0dB%RF18Fa#YQpzehf-qN)4g{jq5^Uwl}nJxCBgfshhYcQna7p
zcr$~lzzd<i!<FjI75D*U0<d05-86)$G;H}Y9I71|Tj0|et18C{>8VnqT*0djfetC*
zA9zU072`9J)s3IsQe9VV#PE3f!MQ0Cetiqzi~<8x-(y_r5g(9LGs-g>CZKV6N(J>5
z@nM&CT+w4#<HWBW2-fWGk9Wo1TAg1_uwP@-CVs1<km5Vhxx)&g<_Q`tnVW?>wm{(_
zC*{lC!2p{cjo|x-9A&K3M%sdOM->>AP^C2;w3{A>165D!K*H*yT)7wvgwkCMfk~6X
z@nsR-uOnC)T!JE|5GY?>p?dRt+}SW>sgrG^asA$nzzL(sH{156l7pKwt^~U<mkz`{
z*<0m1kK}Um&hZcenDa@Xj&rQ{;joa9A*T{Mfm&&y73ZabNB~5k8`Y3U%}+5zZoO(k
zJS;W23Y+2CGR$i<0?#*r3IE+U7Zoh*%FRcmYEUkyqYl^$Ww@HT1#E>DUq%P_sCyLE
zvvW&3`ODZA%-lbKgYmXJzAyzG-47sX{yc%6owv1~!=GyO54x*h1s(3af#nMi=V#|t
zk2i)ya9Lv%0dbj4PuuH0={=F82eN6K*_4ZppOp$mRvuyq25#}{LkK1`DuNCwB-EMS
zi1&@_O(+8(;?Ty04>nK5w;Kj+yfbtTw+$gB$S(kz{DI9k8FQg$Y&X4V(-5AeS;wib
zP^*V*pER&V^AKew_tU+i8=+*4;MN{tN@|9^GA`n)OBqBxGex|0mTe_IBmVIau7+}e
zb81lGP42sECE90k_IN+K5+sQMvubRe`jeypC6exL!0qG`*yZX3Qi<PaXMn2*pQ!ag
za_59~1&-$p?@j^8n)M7IH2A-2{W?fjf_J#0*3Gd;3Q2|zFuH4lBOn&!3|UY`*tG3E
zQX{fFKUgOyi2=ZlbBUZ*+{$rgmm5)<TN-YSbcN7ADJrt8C4gwt5-XPMi9J_+t=gA9
z546jyZs4ZY5-<fwW%WR{g#`g0nsNQKe)^PtK}2FUkQ9N|5IpyX)lYMHTG<6*LLeB9
zJ0UVL3}#pFJd7qdTI;fc6!b8c)&c*Lm1Ev<+QDkXtb|P1NSyq%I-h*t>h1N)wK<Ko
z=u{daMR`Q*P&{%oH0E0o%kr~Q?P3tq5A+|v5!z580s}|);e5i+g=|1V6<*lTn}%k%
zfR!Xh%I8fz$_Eko-V(d9Qf(Yohu>~hf48;&W3}~qGn~g1BK^Mx^@$fklk|*o;*}a5
z2JzL8zH}aZ+F)HEmZmR{fnmzuL!bd0mJ|799SvvCDky`S+gPC11R}K`uvz8fa%Mw7
zfvy2c5&@`NS8k!#lp17#s1t(RDV7SOxMmG}g*NjcF5YS^-8NkFZZ>#T%GGlNw^R!p
zq^)|7&Ip1x<w5~X#EEE}7BCv}K{Svvnyfola_PJK+8W5=^&~}rV-AE1G#0y2Q$XZ-
z8Jjmzj=8?N=<T_OIt`ird98XI2?Fm{d(kmon^84nQ&1pq{0ORRQ**J)6o_OrW!8{2
z(Zp3ytFTAv1~Es%Byoa(XB+NGa~q-C_Jp~LZUsBz5J~K9gTl^dI=Yk#0?(?%pdST7
zD3Ed%0lS5iG$IB-6i2^JXF&odW)!Y+85D}-s&g7U(jlib%4Psg!>9on%-kRnl(IC@
z+SKabZtO{#NkLdv*3n_yQPy$ea|l{*iaKNyW&0&ITaN=h!_w?#Qbs|EuwVj7+VxZt
zG0@^z?l5Be1e(R+N<c9j6PqXZ*14yo#2~|%(Yhh?m2+TMwk(3aP3tm4%AE1oa!O3-
zIPa%HaF9b6qAsNd_R$>M;`#|`h3c}du@_-^w5xkqoSziispBi(mv>CqBo!aBXBL)P
zIy4Rg2z!)<_;6q38%djtt2rnY4YB`H;BdmN=Xj;9Ne)oWjrVu!RoV*&fps2@Gu>PH
zo*(5nOA@SBN&k;}@N}EakUN10W1~^SGRwL-Qvwip>9DU_8|?P@k~M`vEhdk$QB%gc
zkSwc5B^<I|)@7n5jQB<w#d+6p{hl6vyA4w>EAf-{os)Nw5ozsTV19X=%&x^`>mu!>
zQxT6V=BH)6wAbyO-~=CgYS?J_x4jzZPqa^MG*kUY-NzQF-l<e8?90+G^#e;B(FtFD
zk6nv>V7BuuoiC8?n0k}uvR`6A3@uC965Hd4d!dLWQ4|?UH`tYyjY`)@Y0F)TjJAl$
z4E+Tz##V8D63r6~RVFq`%Il^k3R+x5B=KyALmUGMpbbLKUd!Mkie7by#8Pyo$yENb
zF-S1GaJ-<7aieL5^ja1EbqXi(ncxOULJ-ac{-hzvX`9B`%V=K_C{$gvMCU8F_bzR<
zcUqQ<zieRH9Jf-gqc91Mh2DG&vN?sAaslOg)(BxDRXW{=szw?Ilh#U_5geXe=mZB5
zr(?uoT$L~@QBqR5bDF-)C#jWOF-Fs$q7Sx;*5LmE9TT>+f#&Q(q}>$W!(xH3usWwi
zY^WtUTKkwz6Z^P*eOz^)R=;b}J2Lb20)1sZs9TAk3SA3Lr^h0~DIJ<<g@kf5R*}SV
zYWt9{ToQUo<K$}Ab@v4|L|rxpZH@P(C&v^;pU+tDP4hwPE4pD}{t)2LkneyhGt*6~
zy&lK0B12MVr;Kp8wGxNE*OqN6DT5d*WHu-a2(8_lM|0C5IhA?;in=8UmMANVUmEi~
z>k}yRFPK@}J0ww!cqc^2<YHK&K+@*_?EH`Jf=@QX)hjo7f^wh2dJb-xG@MOB>tcIp
zHuRGV9C=ETAo~N>g~8BlLs+pnO)q27BkXc%5u9~At67E$*cl??yx+fggV;q49>i>T
ztj3S0HNDqPY#$g?nc}nx``edHhkh~<B*_3}Su02u*4mrv(QB6*EOc?NtZdUxjUIAm
z)ROnk@AflY@7^w*a~m6h{Y4wl9pQ>*CT{F@Amc;l=sUrI5$k-o{<Qa`dd&Vl;>v;G
zB^qCEYz@YLAxG@&JqyjL>UTb7qiu9!^=0a%_i0aS4C`!kt$4<)0>`%ni!fwyG0FuB
zk4{LxILTi^2?U=PB59^dvV?%!ah_-))0ScEkhc9w>9j0ALl4evREm4hW>uFL(TR4<
zlFs;-t1z;_cM6`7n)2ub#sc`SP2ln9CkF{^7p!LMHicU8;bFc%GMt>$z(Ktv5~Q{v
zAW4_Fx%bNrUFE`bmg+kWdDghU!O&`d$yXqyK*@>lM(|5`NZGVh6SJYLVK&DDE}&|I
zcVuCEV%TlCm8Er6up4B;nIq_T)|k7xyu%q_{LA}S9PK-ogp?XW%pg2P^pY3RP%7f2
zigQFt!&Blu?0dd-EUr`7_SsZZmMLP!FOwQ3w3@-BF5udLiU0cT!uFeX{y{evf?$#o
zG)jzZr^SnLP(@gTMcjRN@KD`l6O-9<9G0>`#$DraH^0Y~ot|V7<GYRUvougk0i%iz
zkph<;xQrJ{0I`#r?*R9>DU&+M?%$@~APVAP-w9`N#ar>EK&pSt-C-=zb24m*kWmjU
zbhv27WH7#)R>7MH17-Sxetp?VQrhv7Tl-3qa<RkQ!Y-wBP$R@%9fYX)k>ct?7P*h`
z1eNP;SvM;0Y#C1Q|6P!;c{&v7+F1owAwB5>GZ7WBE=@11$l|53WRpudn#6;aULeKX
z91n36hn@tof#hsJevhFY!Ck8!jm+P|m_!{bR2qc{vYY|5pKX^O>in39Lw>h=jMmtq
zLupUF-PA^hqvzy`duI%Zm56jmlZJ1MVKR4uU~d2wH=h;sl+M}LC5|X>wjFo@K$C_T
zdKfk#0f;sYnKGcr7@v9X2wEj1?MrI^p>1d4F8ZS<2gL&>egQ<Of4dEUj_@+}jP^hO
zRx>Y%($3{_z~Ip71AwRL1#pnvW{d-(cWLaWVV0|!M)V?Sqc?go$Ra>cxOe9J3;_)>
zKlJzH;h39a@PqKl!|0g^$B<d$8dVM@C%RiF)b{rT(72YV61j6HF2nHiQmijSj)rLd
zmDc|{Yc3fDrfG9{JU`oo3I~^a!B4Q>**e_cetuAIZhiCSC5cC15dmw!qB+||BX<))
z;J|n_f8_G#Jwvtx(JA6gj^=#|9a`7i%}I}$PrK`G_vE$e3<QT{#0w62gCG7wKKNcf
zwL~RE1(QONhv>QsZ{eOx(}QE|)r5I=@|x|OXqlDCemg%-<}Q22AQ?zHOsapdQgE~y
zKe)DY7v`amvW2bk(*b|N?N7-5GU!#$nUYHvxN#v~T`TX^?+Mc*{<Bz~A=ye{Ww7ZK
z5vUoH`I6jDIh`jnG$O=9lOa^bbeZ7FH}4vDfYYqFN;J=JE7fycudJ(FJs6$i29}d3
z3?1Y0eQ^j#<s<%qX~RJY>gEp$-$}RgU&sG-d}!>C1a^TV2e7)y^OUG(#E}wS$mf0V
zGU20gKB1W+N)!om(qnh5Twi6QQFx9aKuoiYA9M*8`fxzPIFt=Sd#QXiDiCK|3oZ@m
z`h&+{A6X>CMT<VSG;CulAk?}!NE8}jN4j}omkK`6kNqMzY9qOwx-`R+HbEz#Uw(08
ztxYke;XnM0IG}E)JmLs7oG%f|g+iLvt8}C?w{ZFDxg4{RU#@SaRfm%|#rtMMOUaOh
z=EXS*g1=2ytSLrTU)TtcjmhnsPSNih(FSq{>7ba;eX!{gCpHNuz!$unArtcKlI_2N
zwk`b_+PFfZ9b^@Q6n27{5Y_-q9qPJEK-Fi@?UU=lYtTJ$uPKyjmP+l_@56ynIGC8K
zk*0)bT3qjMJBkUu4Jkskz3|2(FVqqTQ{qHw$q)oZVZ72eCLOqw*Vu^F-^RD5Q;}s?
zn_SgccNQ^eXmWoe<$zQN$Kb9(Oh&J8@UM}@!1n%{$H9<E2XZmQ-+-mCmZ4_Eo+F+=
z3Y`9ovhvKjRES)1d-Lg=eLkFZ*KY&~at{p2w`6%Y^E660tcdK69(jk!`codXB;sN)
zIz`9@L)`S*4hU2d;~W4M4rSZ{5)S->CG#E|4ZH&F$#7&PJHeSH?#$mp1#OC_jR<sF
zBA%wpnanQ&7a~jeW!O$@kYNDeP>fdy3HVBuVcAF1k~hI-;_c+LBNIVU8rRuFMvXdn
z`zieR{H_`xDiwUXIK&2nWH0>8&Z&s&O)syf!(o&hKB5E}!V(8UT}#UdrG6o~d^d+&
zl)FNWs&Y|;MJHjGxl>juu2xUhjif3mq>u^c8;;=DPFnKd!2>_cCwi2Tt4ru?KO0Ve
zW?0Gu=UI##+rG-3zQWTumQe7o$aRjtSWKci;?aO~A(EK2?i-0ASgD@=^GZmfi&IA}
zZY`D;Pl!8ve!Iu1#heGjQQ-Iuch`SSFG3sgzor-QatS=cD9fnIRMiJHUDZ|f_OIzh
zbSn{8@O(cDSiCj+2i9mio`cxqx|6cX31lK@jlLy+$)Yx#jB2760kK3Dkh>^Jll$`Y
z!GIFSaDs@dAb92*4j^It|C(MT&5hDN*D{m;nqDLj95gu)7^ppPMs#dhNOpDq^dg>6
zA+g5Fx;L*kxAt+3{rB}Q;@0@(R&8A(r0azRe`!k7T(T=P`%+J9Q9*XoCKP74A{%;t
zjnoxg!kUI7Qo0!rKOo}@PCiic!kY=&4V1wBynYDA4_2bI3-kiRxgLiQP1g?GL8?Z4
z=u{F&Q_~{0?CR`l4o#n^msFIJH1MV7!2=NL5~U%{03aZp=X2zY8Z&>?qerThn?rdq
z#X<E$&D1>nVlGIt`4x4AfKkd+$ShFpK|6(hnCJkR2Fi-;4n2MuRvqOg>0R`3E%o_o
zK?#?#y|NZkKn(z{D^<KP(03i=Dxb>~t;&>6{=OdsDXPhCqWgm+q?8hGGZsmYMNoes
zda}kfA%>77Jpj=e6j@PWoU0D>Cotr+%m>>$d#|>VDZvS@Rd>@YC;;(#ry?W)fgpB~
zKwAuk{xP(02*O5|LvMIQN<Cbtg!39{mnZyCRx%JL5OL^wYVkV^s8dCcb-$zVuG`Xe
z8ZC)FlODp5tgoIb@nXPq5PZ>G5P~nM5M^c!BhrdVhf>3rNXFk*+T#wT{r9}l243Tz
zNnwlH+16-i7w_B;O^bGuk>=iE!q=Dmocy<q#8bgX2CE%J%YoV<PXUtIPXvp?%r=kT
zav%m44k?HXf0<O3er)iN!;KO@g8@g5SpHHDaz#)*1E!-$XIls6r5C9*P@#Yfp_P#T
zG3YHgdd&x(XhV${1(QN!s^N|pkWpSBWhv~<#lY86wFCcJ!LV9chVY{Df{U3(82gG}
zS*rJKTzb<I_y_fd+)foiC7O6zCDA*QJ3G^h9r&vwsc=LfSP1$;@$r$TkPyy?cnVgi
z(fN8kA1ob`L%`Dj2Hrp(S|h#7Q)wy;%#TZ++cX%uJ^sT8(E4)g#^-!+i<9OCw;&0O
zdNFyRJsiKxO7q$%+;hRE`GK`aY+=L?l!m~Y_e%!`{bHsLobt=?BZQ7nS;<c<7QUY7
zh<(KtB8b-iio1rnKfZoOW4b7N#D_ji$VxO#3OD|D<N%go*VE7Rx_|&*RL##Ks$o~{
z=GAab<fb{Cm0#0d8YW%8iFU4`#wJ>vFGu8IY9yGaGIFFy;XoFlsKLu?j-zjH2x{JD
zPNdzv!#A%t0xAuC)QBVRjpbazHjKOo9&;=e9KUji3mxrm{FwJ5>hUT3U=bw(L9oK@
zK~dV<92UG57cs#A;EN0+_W<c&`z3~-L)94jD7~EYj^qA*<Mrk@c^_`TAFwAL2b#%7
zEbn^&sMmmkFiLVU$K){w{OL?2F%4uX$iyM^w1XBNebS*u3Frs`AF34w*DUe|7riV*
zR<lTd!6X9_%mEU1p<KWY4VK#QROJ|uD;=#L;ULZJLswCGLvovlq|;89n6PZF9j>Ny
zTJ)8}r~C*LgEd74Z1@<>5TiBZEtCO0ZIFDwwf*wj!-M+u#?IDixO>Oyx~4o5I05$V
zrwI}n-PZ0lj;udu>%Ng_%rc?PquXicd7rKkOwAA9rtB^W@Yzr0ewlsIcaz#_UO_iR
zX-V-YOS`QRUNPfa0d8%;cfsrP-PbR+Uplm=GpW8BU|xG5DG?b5&&7SCCq(qq3_e85
zPhirKhnjc|Qe2Jm;Y?3@+BriY{k~&wlgb!&p?;Is7mDEcS6hp9osOJPxmj80=5{Kc
zG<0E*Y38&EqWDE-t{(xXudw+`Y-VJ0Gblb&C@UY;Pdmv@>O6B`#ZxtbUA4p&`6(LJ
z=bm?e<GMo@YPlGcxB2i$d@`9|3|e~Z+>C~J3G32uI-N`r1viCy0gRM-hT|a=NFzKg
zstNr3+><2JrJl}=++0HW)N>%{1^D~r-W#}b(o2xhyRgr^pZ(SY!ubR+twMR)hbw)b
z)zpE)Wm?nB3FS=>*L|-xcnI6($B*Mdqs9Yp6F5kU2ndxDdA1S^2T3IsyhEz~)hL*^
zwL$&g37R6<X?&w-`XYB4A-<*Bq8mz?dj*?l+&90IkGJRzAgcAx*5kRwMKf^ufG>i%
z5~L!oK27G(x(mCcJ;`@VdvhD6MJM`t_fS#Z#5j!QzZk}EZ3`qC`=0eECBOlVv0W~o
zQ`q?VkjtHEMRM}B!7szatZZj%4CN?U*NZ*0O$NaU@w|+L_vZ?U*q$UzZ33}ePKNHJ
z2W6ycBMZF2y2|H|Z{+j?kqN2tE#V^pxQBcI!38%3(`k#oqwAfMt9gs(0>hO4evUhy
z268;+7{T($EP2uUBiVl7v<DwG9?cXu;q2!O#TC-%Viqjd2<oqvS7xaHwo{{64QAxX
zUwnx2WITo}SruG}7sQM3y`DS^O(GiBpra!d+<~swNf>23Q^T$5VRb$#E-{?6qcJB%
z%4<y$1QIS`$I6}sETq21j5;>`9PXpQo2CPW|K#rOj&3kH*20*69dcNSE`=0gcyX|~
zhf5dEN;h#|-H7V^vHg{6w`Y(4^wr<|@#_RxoJF*JK0*qwYuxrV)4OS54}o0)$ikLy
zm;Pq;?mF{AL;0NFI^PTCn&1o50lng4am2k$>5o6FC8S3*Qr!s5(0NWIDS=pIXOPU2
z;YV#^_T`tW5$Ook^wXxl_A8-v+YiuknBcx#_pPOKE_)}$$nZpuA)^U?T%wDN#Dwy3
zKs?{#Gl6t2Yor(2ijeaeF+9At95+>FMaA>!GmJa|15q2?#DR2^f8_X-H6izshRIv2
z;+;J-s}RSh!Jr~qAdra{K%`glv=NDvDk60ofJcaQlPaKllR-X<Xi|y@r~$e*w`R7=
z4<ZG4r>9<gsyb8L^!o@2K;I*T9ygN<AH*ZQ^}1l4(01JVN_y`NprV~lZW@Z2_6nG;
za#kWhImDiJj%@J|-ys_Z?HdHmkIq;M7fP5#FICV~na`bHkP3pzND2Wu0*ngB+5wX}
zGisYrFvY|P9^t=$4U6OE6qN&K3$`tpPtmaiE0-rB!y^Bd<=fb;I3llz+Ox(2#ar`%
zrZ728Y?O>)tQabJLd+{?ol7m>MWMTL4X;e*VmF0!xoB(+L+SWuQX%L192HjB9k{Bi
zVeF@*%#Z;u;C&F98^@~FgBW|e?TB#C!<aVx+}psu^gzb;AUWRoP{w+4FuN6P$y~Bp
z{K<}Q3rnX*yM;w%KlB9X4uh>V@|8v)%Y<j(-!{Qtcm`xQcQ|?9dJbfJZPIVVk2t5x
zpLEhj?|KMf?Nla{4eR~~k@g8o<wO-A1RM7~NTeeN`%<qL4yP$SERRl4QVJj>fn^m@
z1c#r$-g(2dey=kJ>)v&uTh1MK7pU2}FsZwrHgdtxHjq2YuaP03Axoxckza)~NyfN@
zKyVld0aDQ#j${fuxVLVVNy{l4ygRDziRNZ5h1(QNFp-FonI|jJ$7m=JG2m3VG?GLJ
zAyfMdO|ArK!%GUPb$7{FKe<E8>)CUKc?T;Iw}M`e5Hg7)9c1Y>iDPv)#PrQg<>=8P
zi9TW|G%{=>t=C|9)X-Wv1B*t7w4Q*8Ow}ABZkLLhh;lFzCh?7kr+107G$iW!{qSbC
z%J`l}ls&4*^eBs!VoclY5)}YkiCzNBAy&OT8v*MJdSR9{6i8^uFSmsqB~_w<u{J*u
zsMjNC2!_1)G7h$5p)$K~P!Fse5*hk|xdR-MDk6#Aqtsc4)#8YUSRD><#BD?G=lG0U
zn2Hc~+&UW#AQN<OXbIGzy?X#2Dhn`xJ?Si_#kQOmN{KG&)7fvAN_Y{@#YTO!V&JA2
z3#|qkLK%)5<-n?n`7_gkEG(h!)A!<g7<*{(9a_c(5u4BwzQw6`P18NT!#@4#b1-lI
z|LnbcjGS3^A2#I9N-M5*S1U!)N?Oac)mpKdOIDLJtCg1CUb%<t;ZV&Y8{Iv-E47&F
z>h9`hk=<3Ts%}1(rY$8797KX+17jGr6a_F8SxJDz5b{s}I|yRQ@gLU1k$^acq5Kmh
z$P0M~_8$TA`<-*{z2|<9>T1r+KCCUwu&ciBKF>Y(+;h%74;F-TD+0XjgF?U(=`(QP
zLJ-b*suSr`47(bMi7B3{lD!gzgX^$M)We1hJ`y8GlCTL5cZZ-#uq4$ELimt*xa6~+
z;1`=M2~}$h7!l%WVk`{Ol9dVX5c=7^=Ck7TH{p<M&ZNkMga9Fh5h6~f3{hns=hAjs
zjLyrA&*mqE^@iNj)Z-zC&2eA!&k6yW*g!R$-WAtMi<#4-OR$vuhYhG12(0mg(RS~E
z8iTW?J>rVl?42ubxdTiM1aRMb!kUjCa7ol7Y>A&56z?;{CjgS8ui$yi5F!I%L2=~3
z+B3D|jwkrqg=H0T1;w%i;UL_jXcg}t3Zj92=A#I$rWwVNNU)m>9HB7(`lc8Ck7{iS
ztGI|bi|%_D?t&6sr@p3+j?}9Q%@-6*i+fG!^SSR?An~GE-As<I)#pYX5ZquZ2cEaB
zSgC<Fc)mshEH(ehST?uiKvgr(QGuA9*ae!&b2?u}?Niu}beIJ_w%B9E<`-i(>>+%l
z0qX{6tv^BOiospjh0+YYMU;X<K5SmrhCfvF5ETn?>;kC`4~nS2a2Pp<$SAnHSi5^m
z1y$2{Zl%Io5h1aE(qmp(mTiXAWGRuwus=nGmwYV|b-&<WZNA?Yyu-jDwX!#HEQBJH
z3CJCLLB*X0i6f}VQYwQ#&zgFHTm1)4Z_XAmTykPT;4qC<=*NMLh4}%Mf)7bs0)1p-
z)7RuFZVq;nY0*$1ErX^<xecgM7+KCWha?Bc!bjU3YH&q$oAVT*!zKa-vOjhIz-=fv
zuAJfC>rEcuGL^S2=pQjw2it>#!7+>ul@HxNLE0SHJs?I&xr=<`u~S>}(0_TK2k-&O
z17Jmx^UIW0>1(nob&E0^;F;B|9=v6k5j4&B2kH^$5-?LL835cVNWwCt71Zoy%=#qz
z!@2vXHa45M7=17rHt(l;Luf8xR4_V}gi+J#LZg55%vaU%xY}`d*7VK`f7dlOs5Y^4
zvfD%S_WC4L8Ra=W>WgdjETh9H{P4Ka<qg26$Kz-D`N4<$oYHX@Cqf-)DmrJ2{ii$q
zqvOW1{GkmV(6w-Qu^2z6k}npDBn%nm8LOVw65y11PKhk%&JOWK5ECv~mTAH~d$c!V
zP!G9%q7~ZppBXlu_X~Em>@0i`HEDyla!({=m6@<XaD-c~+UOl17Y-&yguvrS86V&<
zBMBEnDTLyJ&D+TlWPu7&#;y`lqoLV^AP7L1e%c{5tMl|gFhe$|C>@Hd0?+ZHTxX0z
z(M}gCN~|J;pdGjZLI;8r7Y8R|&jVk%1^yQd@r@;js!R*^-UF4E2rNvEfiM$*m=<~B
zT^nc<D60#OBVbV>ba687fCz`QsjyLpE#Vk8I?`Tn;l#oTn<5W!ha(FyMC4%YHD&lM
z3@lY@CLRJ|r8yw2n&@04eRJs<7+Z;nf=mO==c{YWZ{Ge&_r_bxH&#~f-M+igy?J}>
z#T9>v?wCP`i!aq4h6r;(+|2gJuBo~BiNjU?%@xnXdd3PVpJO0pk*!5$mz_t9>t#Gx
z?gdfi%Ct*RnixJQj;3IWUKwF%QP|nERg$@T@kqDUJg;d>l4_^<E6SwSE;qpb0e(P~
zE?s*HmQMi;5!rKsiUi^GP8G5Jb;A~6v=Db6TTNW1$rw?^z`cljZ%Gv+5sJ1)`zMo6
zHQpXdo`~2(1QTfmq9%*cbYlxg-hTgeQ!jL{eQtYz2=uBg>q4m~^AgqG7FS5}g|UI~
z5eURSkl2%`6Pfc?0*EsTw@nZ)*JR0~+fW>5fKgCq)0Y+s!}V(S*c^Z|tk?mQ7fnV7
z12~=cjen+i?9~WoA(=j88zxMRMeItOpA>j;(pf-8MW0w!Ba?~wUJD*b8AF)>cew#a
zmfM}%KfSzW2JsO>Tm&(l4F*tc*cL)(9Dc&T0LC`EihDd^aEw6>8sFzIV{6$!{e}9h
zV_gS{y7~5<>antiabIn=j?yq%g<+-rCH^?IR}cB{0amxRK?+U4R=!E4Cs}>fsC5T>
zUBNKD;!S;Xu)Nq#LGSn^e2AR9_1-;V#^p!Q_yIK~l?fk9FYTSZedsCS2wp^weBf+S
zDiK4nwGEx+Q>)fc%!)nLwZp|Y_jh%P$FDEWFtc*q9H3!Oi!|(_o|Pf;-r_YOsoZJW
zhhKm64(l7BvO7zZ_2M)n)cAPz{kc@JIg{U8abbiulL^fUeZF2UJxuWgX!N$B_Lf6x
zq6gW6Y{><o_ZhYqCziIZ+qB;xbh0?s_Z*kbL?($xGviQk=<f!9;`%Y9BDftSu+*zm
z`h@6ps%Rw<p50xpCjKDS$LNFl6kJB(1`#nl1ueCND16i1Y|!+SIYlcD9jbH~PK5fV
z;%E$>97drLZnN?6gv6<(I^<#uUe+yCLeD|rTzKQSYgdm7|2`^d;mKy9()}b(u#zTh
zhXQV~0cFPixDO@=F7c-OmD$=uTK8}FkNKK{J&Th?EKa+82Fc#nGfe@ni)tKuR$oSB
zLEn9if@N;t_N(fK30Vx$u7D%cd%Hp8$4iA)vu5l8y4rea7^Pap@k9R?LyV^NFwT}T
z>4t{0P)&4fKU<dC^Uw7z3^h=gevP@$N$US{V`FXU_FeE?>l;fO%YjRbo!Y3PZi8yn
z;b?fpMc9c-hgc~pvlx5qd3?<E4QjOO4O1^c43V`47n>JUo!NC_S}LksTB$6xW7qAF
zc7@|Oe1#XW8vs-Z+Dj|Zz_(o~DDXK+(Nh3Q-o677C-f*2t%C~OjT>*@TuOHijg}<p
zq*DdBZCDIIXXFDz<@m~u;skNu%1Ow@r<+q1EJTx*H&v~wdzr>xN-XjniMj|_17U(!
zxAB#47=c;hjlm9FZhGVK0DOX2T385B?xJw1aFsr2(EaZqiGk{UxG|FQhk7LQXQ5un
zZB95{J&V|aI9;IW&mc96^=#4rfqOp0lEE{6;$&8a&WD|Ocl5}__t;RjkKJ7mQoj^%
z$sM;55Kw#7vg^g$gS5)!6+Ua**#(ji895%Ru9BR%p3M~@#U9AD<jHo<14?sUc`c6E
z>5z8-b*;`dI?Ry5;3*8nEOz^<@Jcl&X@RmqOedV(F){{2(0A>)0MR1hIQ%HYhfcmQ
zk^L-0HQ_|PQplY-&1Unq%g2-JHNu^wpp<!Z0>uy8sO9l04Lcl?IiTyVC8H81FYb7A
zsK}90FIm+uc&-&>6q}VcEYTTPwUpRcv1#bE1}O%NP_O>+$#|Gfz%+Wf&P1F^Y#<&X
zIafj2+2%j_nU-pov>q_6X?8Ca^eVuLTJ1oDDTgWiPS|r*#`{xxPCGDtL&vE`HdA`2
z&e~B?SyEg%xy}asBPv#6R5nqCh!BSEsmU*qzHQiG?hG7j;*w2$I!`Xyjs2&%i2w!W
z<p(`hC8{%M1;WE30w8fo=&zjo(a?BIJ3AoE<H3##cFrZ>Cr(6p0HB4EFg7+bCg7rB
z20z;OUVq?s%bCVsSmxQrKa&8MEj<A&7di;V24+f3fk>)#F<J%K8$&GzfR;$u&6fC?
z(pni9RSlNLC#OUlNy$(@zESv=M<n%q-=!w}%c@q&rzmIYF$!2IzS)$aYN?A-6d8^K
zVQIe8n#gxMN5Kzj%t}7a=uUJmcvCsjg+!&e-U2IS@f@JyYkUzr4E*U!bJ6iJVT_z{
z8u*Q#g}f<~XprtG42At)tZr$!s8c-S?)SP%WFo2@XZ0$F!T7Y5iHujlLPHBwcEQG8
z8WdRN+w0x+rCZCLr4<Nnoz?E@z4eXmweGco*>xDR|62vf+A;cy?-JTfG_dmsSA&Nz
zh%ek|`ELC8@*ehYr>|2eSvB8xIh8uZksYm<UKUOeo^2X#7unq*3y7ttU_C~a_CR@>
z^;0{A2-dFTyFV*EnJbT&;4VLX?+?<`iTErtEu5+qImu=^LIaM4zu1{FzJe}ZIImd=
z^ZJdYq-3uq^U{Z@UP=G-sM5Q@*=Fy$Fd|55^xzrTGf+sFmO#(foxp5*=RxAV*=^F%
z@Dl#0w*znddy>ztcd&T_ch!tWy}8*0exD3`?F}a8Y2HVK#DXpSeOkj)?V1`SWk9C6
z?r8gApAIe6EteV!*7fBEnWqTd!xA62>FDmYtEn>x>+EDSybjTd>dz+B&&byE)xI<0
zHR}R5!?%a1ve`rx<aa8BOstra1>bWGOaT!Zt!N^I($JyLH?DlK@dk|g*G`5BH@pyd
zN1|N=;|i+@#N=vPhR7{lpDo6~+FC=L#SM6B=&$tw6BKSS&Q?lczT5cO6BxyqapZ1L
z*QWr8+s_(Xr3Kt#lvRca0dn==LnJ~#4+*h{0(!HUR69->+E7=kS@gu?L6RvUJh^JH
zDbGTFc#JH~yJ;Dz=0T)qD#~aUWJ>Rpjyvf9zOXSk!XYhw0E^F$aaH-^i;-Avj3lnO
zgg-zV56F@*PauQ)LdsD2CiD(X(v}i=3bb1cBs&C#4H)-n@Kw%6GS_WxnHG2J{&mVo
z)>iBj1SZsRx+q36rR$ai#)1_|3U>w@4LLdo*kDP_=0XbEjwf4;z<qLvFkRrLM}GXV
z&&kR54supa8g6h(oZxCQ4hgUo0x0umZ3gY>ab&mdOK_Q;6C^o3jD5v{>1`c@d%M+G
zz;Yz`FCi>cLT6IbKwTQGttnfkL=1p&m)P+LrZbl+ILf7jd>Zuu7X9#%5m_{+eqp}|
zjcYf)%ouEj#=wX;%Ru=7EC{@eLEY61?p9Vz@JS|ljSq?lSYgYZUK+7arRYm^G#Tup
zFVx=cAHg<?2nkKJ;G_|>v<@78Y1?KnFM&AF+cSV#dkFhH(s%BxKy?WfM79uAPO(P9
zepa;=_9Dm(xBQmTihi=m4-28lTUxMske8;vbw7tsI7R>#oFmz^;VI)N+qJ7#D^QMP
zRHNjQO38^t2T-Af2#tK4cNWhj^o9drN^e`~#M-Lp_RALD8ZZt!5mC(beh;u_V!3fc
zmtum%rZ_L4PxWoe(a;QQ(!@we5zS)pXt49Bgqm4xp^DaXSa}e5no(hQqW=jL*{ES_
z-ZVPuYA~uXMmp)!4KC9599+TvPl4sP7W%ZESU9G)z>Q3$5nQm)E}+MR6Le9iBoAdg
z!vwI+Ig)U!^1B|YL>JqVCEx)jdRB7UIP1hc>)?R-QE%(v#tWhN6env=k+{m_xUz<g
z|Hw9Ui8T|xBSW+bbecvu1*aE)(+!3cE`x+@UOuizIEeQr7*?V#OT%aPPaVYWE>2m}
zcDkbC<^(}M?Z^XjKr>uCU^Wv{7GrV3*)=&n*?p!Q4rpr|u<A2>GWeV~u|9_|3C$o!
zazba8F)}9+Uy=7aKcHCPiV&mS(T<$DT5z;iYNyfIw8gP=RK{|@_2BsUXmWjlHZ5E^
zY|dQ}H#9eC4@L|9;R3XyIP4dq$@0Q}9|sJsCw4;}X(^jx`@!*{ONGf{@>4z3#%6#+
z8jbW%5a}2gDsXbVz;X)+(%I$f1x*SKxJYkve0c@HEX#j&-Rc>YI_(|7?oG}um|Gq$
zXeYy6iTcUzAeSIP=kR#4pdd^=t;roifqU58$`51Xn8}GP;T-1?;ujZYMo0|Jr!)>A
zb#B9JAoNKF^RPF5hAUYil6!P#Mg44@w@|5T1;L0NV^1KYtx^sWg3`(Bd3K16f#8@~
zM2>6WB?Bi3Bb^-kf(10LM|Gj~-_u%TDX=RbZEK97b=Y5~GeQ#ed|S;WI8h<@4kI4Q
zm2@I;K$CDH;Pi;Zsa+R=cViR^8Mz{?8(T+U&8@OM$F@xOTG?P+6YRZI9}fko@6XhP
z;#|2EeO?Tmz=0$<r&TGOFfDtNin`^1#XzBWyDHRv#o|b0W*)qhNr0=i4AvRcYrtWI
z3kC-9p4u(O+gRpG7`qFx=_4$~=&u-FhJS=qalG6H8=w>c*UvKIi*Z>m78hk!C|3pb
zub4)0)P^B10p*B8vyQMLBfty;luQ6G15>F8Xr|$bUWt;Is0spUfpWxVAD0-|B4v{p
zOtj-UpA)W%;m0<JBK$P5qse1c@>jVS&Ldt08R;0i0HFo;A3z{?0}0xNf}>NKWyu50
zW*6f0=JrY)V$Fsn*>IFt_hZscuqHpn+T2Wqs=)5k>lp$Usw$S9UR#l?P+3qQubZr+
zIQ9S<lG%s`0;o-Ubw4hsx!c_Z4=ATejtWb6?sVV!snzAR)ulBgJlLSY+49objkmgQ
zEZtaHzI$^$Hsm{ogo2JwuN;R{I0&mJfL6);#U4v2q0ug3JvAW<h!Fk^+n=QSke0+B
z#}x8og9eM+lD(t?8pVnZxEx#Yb-JgDTa?|QR<UG+1MdQ(62&&B-aCe;aj>CHNeHsY
zTKty3E=kF*@A5nJUGY#Ugbt$g`a+tcSZL-LhhOHrAEa>xsW25T<he^KQKU^C=2RkO
zBvNr|s#g%MuS<N87XT^5(|zlY;Wx^mTtp9>(>u@5-5KpSwpbGahsOC~qQ=n)EFDC-
z4R=oZ+K60Cyuz0xAgu=_omhh;#T*vVG;@${216>zku8BbH%vgY7o<kY8Z{0llemy@
zv3GpGN_VgU?m`S_MM;5*sa-$Ia0;lYt|t98?jr{cH#EE7Y)T<9Qq-U@vP`c2<;H97
zQdX^0foab#U1EJM^m8cR`$sVLbXC3gVxnKJ04jIkM2owTXKjcY%*C#47Z7HKC=_T)
zJ+Uakdc$J-ZcWfeE5>ahMlpH_u2%Z6g2F9_As3V(z3XtcgCw=77dIPyIGZ725XE6Z
zvbvXvy80zq)lO(p1D=n~Zvi<#!yt)3mraBNiwOK711X0lxG-?;2sVVUy6l?TiXODj
zj2qNA0L&J4adktX8xjx|SW?Ov*ZJ;&5d%0AVnJbO*<$yXVW}tKbPGBK+j+;haguWL
zt>=$yO$(5EJTK~(N0%Fqf%OeG+8!VTFyhI)NXuu=UWL9}$0vxGhPA*vig_!_BM^kC
z?mc#WT91Ae3sW|w2e7%RHZuW<iAf;&w0cU|%RLh|STvy%UA?6^g>{MSU@E#qPF6T_
z{d5p0Kq`CCf2wZ^!(j(=jk|YN6%VL!0%F2WR@6+#3n#oZBMHKF2-q`vi}rROg5FU4
zPy{l2EiR&<#)Cvz**ZUSjXn~DL2m;J*SH6)qpk&Jn=l+X!7T?2NS^c_zxKGl^V-F1
z1R?!^OwopMl6xCy^RUp>tCh)7e+M@(xT_iNO3nui63QENLR{bFAh9jfOJ0T$PXQSy
z{?P?$YdZ55R*7}8MS8FYgQ)%yx<XJbb%Je6sSFp{o?(|@>+z^nKA>EIbnN8FH!&)q
z3*!%=I-;N?BPrEt%T1*ojV<l5L^=RS!~oBT2Y`SRN}_4Uw@M+1y?FFY5Q-9MSW>J{
z1btkrqY?AVaQ*Z&h$S-eq8MsHadKd_bj)dgqI*Uuw8!FMnPx3ZR7`#q`{Y}G7`zWG
zFmVX1Mjuzwuyg{-BL4XCGZ9S;n=Lv6C@B?o0wNykJp2d8?Z(Z~2_9b&Yaze{SA23C
zQ`T_oHHFP2nI`5ix&+S?G&RxSd(?#Oa0l8&h#)e81zx;-(dA<6LMRyaU|UO1{DYIj
z;bguXc~7U9h>TDnNLa@fdI*+WS7qZFS)>A}_H<V$Z#I#BW8#9ZvAM+88Om$fg!F+G
z<dpz7)iG>GyBcC+EIYsvE{_dMF!(=!lCGh)HY9tv;TWAT--%fxTWg>L4A=oKJ}$9u
z$i^k}WR4U#0l-uTD!=2>aGbHkDJ9t&EWL4++qX2X2<mt*SW0Y<Mh7%lph5Y}(<9=j
z`V&+dUJpmKr-g+B)Qp9R%89uZM#!rbW?=Fh07+th!YnZnBtQVsV}q-S*A#($oeYlk
zYQ7qR;h=&X-=!N$w+B-|97F<Ac?gqIJj*vAYT=BM!>Sv+%ovXirUnOHY?EP)McdBt
z(|{iUVYSjyk3d+mw$W`nKmyL<Zao&WUzj!Ny++DMrLO!;^d#&T{>|pAQ;T5q<!HtQ
zY-C1qaWQ(Y$C8h(+Gs3?R^4DyOXXJb77W!62HWl7$-#mBgbzD`@@&$^alnLkpdx!C
znuE4pzufq-`5^Vl?q*%4ct@m3MpccVvyJHbOH?8dMzNu@gVCtpaM<ohCT=?NkcXN5
zs8nz$)18ph6Ngn~riTiE{IN-DIHCH$Zfrn18^Ba(ziD&S$Z#(sx5*?SxW!r4No-n>
z$1?|-#v%x-C#Sz~tK|*q+SfTMF9uoxx4n#=$~$DIJUyN8AQzT6B2R|UT~LW_g2^@v
zVUS!fv6V8BHgfWYh|0_#%yUmTIrz{@-x}lu97aT+u=4^Tg9zT+20G&=;D#hCmQz<+
z2eXMgCU<RzH5Ln^-VS3%VJx8$cV?`~RZ~I}_P6CRXzRMn*#s$dg$YO_CQYIYz#%F;
zMFN<|H=VpJp;KiU>?=8+N#JqPMpDW;!I?+Lv~d^IOjC&?F>XG7Idl)VK}TT|OYdmh
z%<hWjmtR5{NA2DuTtH@$5|#sCCkRa~m!GmX5v(*U1pnHewtjHQi8y=xt_7flAsGlV
zBGH0H*5fycLXIv7B<1~7UZG-YoEHvaUMh-hA%~Gq%5@W{x=e7`-9I`hhi8;!WcsEF
zoc=UgFJzn)w-?s%60;X8Edjf)uoaqA;wm2(kA@A|N~MX%r152-awM|#$Vp3$D2U_>
z>Rq;kyN1*wgQmSGHySv{Y5jm}MVz3<78R#Nf!xBzO&@OgGmX4F*MfCyK7MI4gcZP$
z`v`zLnam@SijK!^iWGQ)EJp9B|1@t((1u5@GzW8QtUdoRJchnrE$5JX7vd=U+Lc0i
zD@0=Va{54O7exRkbW7W^7)__FL~JhGQqq<ob}W&`#n)*MmXCA3Po*9HyR8a!gzn6A
z0B2i>s{~ghO7h{*Wf)Dm^L16THnEU637DSBy)}9w#9XEVf@qD4T(iMV0%O6(8_$Az
z`?6Yhs^Y+g3CgdC4UFaYJ-FY(b`nXSn5Z-YtokG-xv$lo`hf<TASVfgg3$<lAwohS
z&O}pBnkE>5z=OmQ5P|eHsF5)WN28-ws55n~<N!mxOl@rt#$#ClXcx5E)RKAaQ~pwc
zn%o(*eCkmdhc?&PxOej&9OoHfkOr!v@WmO;L|U3r_YmRe8#}oF$C6+gunSMH4ZsnC
zSr8r5y2}P45k_`M3?~FXm~1BUk?Okm6lmSR=AfN7iy{?K_u3M?jdvL2N{5Zky_?H-
zx^LcF>)u#;dwuCn_fDtVSz2AaeOI$YueVj%T|66Q;jmdSWEpCX1ExU+HPKOmsyrC%
z!K*`-tir*iyc5sXL0a8%OI+Qgcs2as6$#>a%dw%+*v&G)JeUhcGaL=J9we48F^0%A
zi;xZMH$gr0pkXF$Q}%RA#$<7UaTpEZ^(nhFhQDMMgzGg{9tc_(H5xXVHp{o>p~w(}
zt!NX4Q~(%^_CpMhiW=%MQ!BRbVk2eUO9C$eAt`%yCFUkCr{QJ>Jdgkb8Z3{UDg>y2
z3?zx#W;NZjM3^0tta}pdBW)YDFYTqR5Q61TT}Li^Sh~%tZl_rHcDNb9L?F{Qfh4OC
z#;y}thXk&{&7e%!fjrh_w0~8*(AR*9inQlEWef(UprqtUgoIcdB##7;G~8(Q+xu;}
z_)#4Opn>57c^B<q;El0ID<d<rj+uW(l4bRgW#o_=HtR6~*ALO)lITo6Q{=esKNZDb
z1`dtL@IHXUVt)8>BjN|W2LfM6YISBva0dY3_)m7({)k*2#6Xa7xFDRst%Ii>VQ+JC
z`FuUJaAueaU7}#H(mEuF4kZW(V?gB9z{kctfISufrO?DE5Y;YwcG>U)cL<U$Aui=y
zsEcW1CDIzoE<oJK)5Vvt!eH`nSsx<l#U9FpGF0Gsq{o(_eHAiD{2bZkrf|@$1^b7w
z)8)~A)=ubR;Lhu^fACWCN!4s58uAt+l9&SrO%9RTnC|kRZ6|;`E|w)k29zg)HGPP5
zTM2aHryzJBEk;7@kuH_y@<(6+>S=yNr}8DPdv9l=z%V_|T8qlPsj&oPcS?{@#vI7v
zeUMP{bCyVs$XW+kOSS+gkX$CHVUK5TK-fV^m2WgpO{3;==Z;N?B{3xKI}HE_13$i5
zqP?(a__$FR*aWFULd-1uUR(}V#*!rBRU5d3B@+}K1i_NTEZU3dYktz9k281^&owUM
zJRGZH9DMfC@kQ?50b+;Z<dbQ+PG|KFUqu3(JcL6+smjckP#EC)uqX@Ws^C(rR|G-5
z*S)#geT`(+!xdsKL_dUx&45q(R}2>$ns|NYCZwOe3Bu1}NAaM{IQfFgBCS#2IR&CQ
zf@FhyzcD1q*iq6<RE&scNrN>JcnDfWmkmVA=Prb3Ef)E_>BaRJ!rvnFunSQ9Ue3nJ
zvK$&<0}>R+Eu78kb&4>Z%k5)m29HN05l9aNtM#@U#t%4BnS@E)<aP&p5=Q*EadZGf
zEX+-KLoz)U|G2S>kpH|jue!9Bpb6NW2p}(^mmNHWanf+j&=WEmGs+EcjZa}%q8;%@
zaRVK3ADZ=b21K1nA5jMJUJ6AJ_}~qgj^+whivcpul4EKQO8qzqRIg~8EFy!0>>a)S
zTKfnl;821>T*R&-0OXoeEE3W|!!hwQJG!!TRZbI67^)L=vu609zk=U2+brz5P?&$`
zfUPN$$sQ~<?7CmWX~S@{e+Va3Aog>%guH(643T#pAOc$jr=f$hvi05ZI|isJw_7r{
zbcBY23_SKEaR~4efj1QSF#+eSA-F{%tyoNeXb(=(Vgw{AdJ-w{j)J4*6@<!w+~y$+
zL0Ae}GpNmYmh#9X4_3x^^ibAXYOGfc87`B_&aG@^bAl6~V(SDQCC%q@(jUt3Z(71q
zmy}*fj!MiIH)UBu46QbK!p#=<d{IE#X5jw7lF0r_djz`%_eniC9AmFd>$AA&DsZf{
zx*TzS+J`yFu@sTJ7<kvMg`=tIED)}*mLI9)MO-^`ja3AKBRwU#c<W?aU8uLX3>q_j
zi-)D5mRUNGkR@!)=!gp|7gglAeuSzj1klehNXF+XwV-`GMCAHIh`(6Z#b!q~7l(sw
zD?k+sJ{cc*R5k#CHpcL*76u%ZNkwt4yB{P^u@%oCsH83fEppgl1#>!O2&QDgVHbe>
zJazFR%!iGjbHr%E33&*$99TNI(+vQyq898W_+7mdBe>G7&XlNy5`dJ-xTUpQ>vaK0
z?fw<CM(0RI9e2-rgZ&eD#Nom@NJp|tHPR7@7c7z}RRQ(Hn@goDG+QfN#a#txRYdPk
zz;-m=NS@Ql$y&<LFE?PNsY>o$!8C^`wl1z-rr%7BAe8m=R@hQqUES^uMsbR)6+Q`V
zau0#<UNC1c48pz{PVfTHQHJWOiO<zq0+)=-t*J)}5ga@g8DR_$%0+XMS5^XW27dcA
zbv4FqMJ;305$2%)V)&;9(VEnp_hd2?(X)C-ngF6S&nf(TC>;LUv|6e1l)aN3lxg1O
zIVaR9t-f1mGR#yJoU*8LVwM&nbsxy{@d(*j5L2amVUkFjLzW6M9b1?(>LQJvkv?bA
zF{LuM4<#46$}6}t>F*$>0~TELYytfwo-dWmJj6^_AdH~!^{A`bV<8WsfurGHI>8dq
z-@7B>jHKFcfxYJJCrCh4yJY$Xpv>PG<d#?;^({Qj%L=prI#<qXgfbFXBFZz9^MqR&
zj;Bu3PUV2g#DY&fcD!!{^8E&x=S1Std=1HI0(qtWZ7Q97chrzeIM=u}8tsBOF+n^d
zb@8}V+v9lS0QOR?KEaPeTS1+H8ZLpVnSxn4u858vra_H91*G-d9hquT9ng_MjomX6
zypEoN)RZEq1gD29qGEs-G$SH_|KpVt3}m0+_8&5p2^Vsp$B_qc3rnqcL6lgTuf7<?
zUCt+2J*XOwVUNloGC^5c&~%aTyW3-^>eYT2;!6(!!S$AV=VT1l@mN|x{1J8ba2%5b
zQH53HVam(RB@CZ@$I9q@n>a&K%+2zCuANP(p$IE>TzV_rND&o?Mq`-DWwEkjebxn5
zMCnFS`T5Y-G~Z`B<;v`IA`N#B8JR%sNTgkHh`Iw`9%fa)pY{hN7sQ8v9}I*7HyYR*
z96!L2H{De!ub^BJ-tDuo)>HK|dP*ghYNO-G=QY+e?xYi|Z-akv1wo}LNF^l`?uO8q
z>yh_pG7z*}>^)6^;6^);a&^}E8@`=3$>7u$b?Dg10=yP`hO8ieavUmS341}j%08{D
z>GQCNU%2bQ*VbOMeW_~tt<9@fK7ao+^N5BgUBu;QTl`>7zDZ-Pc`%Bo?%u(q5Vrfc
z;`efJ84PT6CQyI|Aca|+O>7=F=LsNjylH7CU3kIK0u%?cd}&SVF2S_2h0XMI^O}W8
ziwW+CYw3OHH#t-@%01(0!;gcybLA&Z2~4I_f|aP~6z360oRTi_q?Hm5=j-9?iraN)
z#rTlp8JD$?PR6ZOAeTz`Q@Wck1lQz#7;x$!z5D%Eu}_%Oz-y-r7VLfahL-W6Tyk>&
z3S0FUQQ)x#i}mlStEC08%j0s}PDxaeKR`)g+}cXeO0==Ur8q36s@o|0>REIEyG3!v
zk$nVs3Zbqpg`^UNpujqdq-MsAsaQv&H(;5HW8to$m{g(VEDOtE;FcjrU^~d++{7hO
zh=9mIkWPxUr7)s8WWcoQC!J$SdA_&7cNCzxcum<J38(6CvUiLNBTa%R0v|G)9_%)l
zauj|>vKIq~g1Cxs&xxBGr5XA%Zp=Big`xNyQC|+NzFfhei6j^nM;3ruxxBG?fGX!q
z1Ugf;rKbY16%tn}*6Kk^bD3+U@@d+EifvxORs&yk+Av~Mg{@ryh1U|wDmU02?RgaA
z1&mQg7)JJ<wysX2CB+UB*7?bx$?d+i$`irA)6-$|`YD4+<b#wiHBhAfSywp*dr=HY
zsCf-YtSfHL#-=->Um*S?5`huXL?u_icc-i5qE8+$YGr$Tcs%a+9rcBA@BR>%dtKan
zj>Z$8$%Y(Y2*a3AkL#Tja(pkT1mMZ&l;k(nw=r7HL?p1z)HE<U%SDc_TtNdwc2U7N
zvKY>nU0Wz$p{&U%Rs;ygJ$f6cHN-iwNl~TDy`bbnVlRvb2+bA;yMZH^cNAQT$<BCy
zRO7+L1-tqP5{ZOr75+~G*Rmi&)ku6<E{qvHhp4qMau%~W^_YeC=c6m?hX=tSb_#|L
zhKQj4Cg3oJY>K?A;Bb*(ovGJPA$lVf-?;0;we$NEW5TaDV@>EfpBC0{l~aP*n|Zu~
zlrhrelY!n5#XI*cp(VZxIQD>)>rvQiAgHP}F=>aM$qA?jY=j`uogZFFMsFXxWZF`0
z3QH#5FY{YrwW#N#6CtAka?NS-iF;s)&WJpJybR?Xz1W_MLU60w#nIL6=I#^hSJywG
z5uZ7{2iWeIfO8t$mL#qLov^rJNKeQW>Osn<$=#{Z0#E~-l1tXSip+N8<2F~v0Vii%
zj+6kVDQ}6ykld^hw;gP5*duN`Vm4qDvx!jpP0^sYBoSh>%I9oJbJs|?N0k+c^cW==
z2-v2piK$^cA;4ilpj7+d{!=)9&=ngH31yma!aixnO-cjt<o(Lw^P`CwwIqPX`qnvQ
z7aeS)Wiv=;##&Vr!X>#E03DIwf$GOi@UcxtWW*Ghlt%iBe(+(`(KRC4@_F#{LMvOT
zw7R;}>7NOOzCuf5w3seM!;PYOn6JdPqh3f^huE|ylJs{%z|3b6uvc^k&|Mu|5mpBA
z!>~p01z8V-3dVM1596kl^%0((j)0PWcAO>Dj+>n97&KDGM?k9W7GAaDZ}gtl1EYuc
zHq{5H%Rg=Y<9$t~-oa>uwBX95B&fn58jrE~GFUl2$+I5x_j@~X&KQif9OY;&$tKlH
zTHeGt+m2k6MVD9Uyd=uSDd=p|@yVC)(w4IYXtdZ|jy<?2^(=)z9wPvoAv@AGHQWtb
zl{3O4QZoD>1FOTnNPJpAZ&l3f!?FoGEe>DXPq4@U1%q*3iDiYW3JB*AVQM%HJV2bx
z9kE>&QAS|NZ#fJcd0ceg^FrE|qdkkIC#iz7Mo^JTJTehC;j2WLG*?*~_wJH=mooRo
z*#p8*6P5aX(R~Q>0#15hj*`SiaTZeCfGeQ@B4_&?5<~Hrm)tx<KZyg)+X#_P;_$iB
z8r#IiljA90m*$pW%Gdm)VVB>5^uW}YlzFmfgK>{)%ER4yvWSxxPW1i3f!U_O`8+lS
zSH$8f7Lti3%HURBmKnTtXwfF^N?<*2_P0-%5CQwIjcjLKgsxk=ePg|QbNP+8Z}H4c
z$g<0#`4Yl9^62CWR-R62*Bf_WTgDSWN=EWY4srF_2Ezm_eC5tkgZ3!E6vQpkt4Ew-
zKo{=QBmC1p-f5@l5a75lHWLm+gMiix^9TslMLZ}dvEVn8geMs=kA?B*ARs$=8hTiE
z)n#<th6BHmv&@D=>dMhT!#8oPm3D$CANnnJSEl%*Bhl7su$@2%>Q_R!@WLiYY8N9p
z312xlI7T3hRgyo9DU34BsbWSM+oO2FmepEIq>I(a5#1BLal!(Ba8kNR#4(dD(qqQ<
zz@1Hz4@eB-%@9L8Sg;4ine~L{!)XOpe*y%OpB!uCuIa91Jb^i`fr%#P$lxMUM{c=+
zL&$;=DT4?$B01PX*j|BzJ%EZ5+h8lAi+4v@g6x@KL1gty;zWdk_|KBITx9wXn@_o;
z2xdCFc<vtqFfb(MKl1)nyndV-yW;0%c>lM^g(9d!#`-Jh8r2ejJHGzdar&}nzJWM>
zLB%>wUj|#j=`)W^N?dZ=q%6OELCwu?RnlT6Q}GER+w_@%7?R;WZl+o!{2^i;1RJEv
z#Ky&zy$JqnSKPREpFv6)?BH@mLa0Tg#fF2~1+8n~n+x%Iut!^Usw}8Cpj;<nOJx0m
zVuI~JcxtgAibw!Aln&ZR4hY17&K!sz#pZaE8he#nn=yjbssXwU3T5a=2l40%GGHV+
z^@TjgOAwJ7m+n&V+MuuGB_whdF-+q0sRowg+RME};{do%uQ(FQfR;!PKHNnHXe3OT
z*Z3RRVt_DslH**Zv@vz+9l-NBYbwtXK`ARNDn)6IF;HD2rhqCFpE`w$wr_-zq^^(x
zE0c=D&?DKmrL(Auv6Z^cGY|!Tb2k`&488yITZDL7ek;6c#oyn&@>`sO$#1#Lwi?eY
zT#`ESpGh)hPNOWn#3fGLUy<t#?3~D~-;!Op1KmeAk&_V1^2Rrp>@ourZ?w?cVb9LP
z2iQrC_lkv2`$1XdmLu0eCUlR$VKSgX7Zf_YjSmokMCIo2#(tXg;ch5!lob>&A|S*q
zMPjbI7qP?e{f_9J5`H)>jEDf~O&TgXFl~st>Ii}Ua;1hN1Fr0Qa?i007115&ucpZ%
zX$cmLYv)l<trw&^xIE%4gF-yyuWLKbUSyN_!sebkAdkk6=(q;{M(oJIQSdK>{UB~c
zp!+k46w6buB(7Eg7$qwo<8-Wko?I{0vqA`DRngS%9JN)PIyoE4$AHR{1&gi*o(Jmm
z=szQoA>zDHPufqfDVddBpmM=zuJM(xeC4|KxI5}k=$Ir7<d(YWb;a#F-5*N2KS}zL
zZtbBQ4*^JbN77Ur6!-F6((|TvPz<^Wgj_6W@#5%l0S=`Fgvl=c1LV^S0wIaqo?5`?
zAL^N)OxCzyi@1-io{HIqAz~o3lVFg`u4?JXV=qFF<f<kcj*gQ?{75#KO*~ZyKGzi$
z%QfOmt4hb>VIWE+=}xXrOiFX*Ei@4p>mnF`B!aV;zB;XmV4t{(?HyhF99=ZG_wYC1
zBN8`AkUF_F0Df+5rH#C`#o}9AhTix&5M)vuj>CpQf<y(~+LEzJF$&uBri0{6YX+b=
zM1v?OXn;%$v}6|+P-tLhgL<e1V{9FJAR;_`3gA&7z2qn#HVUT*Z-RpZ(K+R1xZVN6
z@)!1)EsfNvhNyC#d+l@4sM59`qiN?@BkpVRmY{s;{<f!f7y;mv##0o^iQMFO{FnG)
zZYHiq<GDM2ULXcOKJNY=P<+rs6c-UHCDgZ`WCfK<-a(^r7im-NmW-75xu7q1ZFuOL
zqT?7>f}gVCpl`aFx6J7FkKn#KbPf38dno^gSR0z|{dodh=B=h{Pb~MhI5Oyx@{*-Q
z?Wcg??v3YroW>l&G6T1<c`Od41cCs#lLQ;v;my_}ow=FV%fSV)3D`1TgSY1q?{+Yj
z1&A*=%D@)g-s2v$|4=s+9qiKB(8tk1J49Jg9wh5+(;@f$`8G_SVVgos2k*g)vX*<F
z4@F+UcECq>&NrHpFBZWtIG2Q*@P+}Um-Mz!Eb4_KqP#j#lJ)6N7D#d?3&NjKLD#so
z`t|~p=8aYiXoy3UN(O;!d`Rw$K~<e`nKsx(==nGt#lD7|@RQ0n+HvJ8jsKLrI(_&z
zICLb41rLyBoi+lFOyT)(e_9C8#y0|ssuu>HiYgYxs_+J(s2CweFTOz~)ms*azHkSy
z=s-4vu^jt`i;OFv#-iMLc#RMYLLAy@Nk`l8r-l7O*Y9|YHd&F=`V?v=u2js*JW3HV
zxQbje$g~<;mSxeH<R^4UTcAR7jkk|5Oud*7gCMgIX;EAju6HPbh@>z>4)=3d62>SE
zPh$MjfFb;g?N?P9+nwhU@jFL<)`=A|pz!W3Kr3~c4U?7UBrcg0fOR1?!eiXlqzYss
z*|ku^5Wn5+eX8_!A2V@AB0y}HA5121sCxV?>dIA*I~zD$v7>rImNR?HO$hmGv=0<A
zIV>{i#$AN1n+H{eHF#_`iAA;QkF=#^vVu<_Y&&qn21*)X%F?GO&TTg3FJ;3fQXc?H
zLgwJTe{iyEnO2Y`12M&e{w!WFr=1{O&PuCUSgzZ~k2Daqqq2M#(L2!#(m|5346+z1
zfg5n5dO4K_MhIw5mag9KORP7zBOf87FY*nBtx&MF=8$qqYHeEoeWRm5O%6e2BW0_g
z18zwtR)%;$8V}p~-NE7F)xw&c9;;Et62e)A!)CYJ+Zm%>P)mtE*GGg_DU&x}QMUk%
zsg(lRH_FS&OL={sP2QTt@(}>8SipZeI*2Y_(<CfX=baB0%`18a6rM_R@}d#jdlit$
zispS1K-udWtjWsMHesi3S+py4lOZ+V@{<4{4H<kZi^_h?VQxcY)o`{LAWsK9d(USj
zh)()-+E$Y?Q^BuvFmIk6>hl_a*Hyx}WDkH&hG2PGG$U72F{j-3)W)!xJsmI4gCslh
zRW>OM(><QDQ&OpPV4Sv51OOzIrx+rzZOuWYG?zr)%nvF|5lR&K7A!|#(FC3>B(KsN
zc1`(<<Y};bAd#k=OO6k74$PIriR%wF$)pQ(E=Og-_7T<O^r>NOJ^o4m@Ss5&0l1{h
zH$FYRCQR-NE*xBX^yf+Y45&i#4K%gw1|H{Vu=mW<_%cqgqm0c>=ZAZo<HZr8t%obb
zeyf$rxg!Kw-zUX0#3q$6$%CxStg*OZX=zTWjru7`8Gs{UI!CHc^9e~VG)BDRkq`^b
zj^{QpN7zoESw^VKi{{#@pP?2)<!VAM!}7dVC55)ZbK2IP0jIWf1x-)2Uy7u24CQHZ
z1LVheJ!j7r?DQ`oi)*CZ2`h5+v2r~utqAErs&EP>X`R`53l4{$0+Hh*L}P%KLWmiZ
zcEZl#?{*)$Ti6N=`%h@?9#b`xR?pEA<=*uui0h^2v4a)xik+z5!~dID>p&8n-hpaE
zXsbkzLi!I4ii8DkmgK}6j5#huS%^Gxs>=_$URf5mI?<HSfFn>vA8BZCjZen{Z22MT
zQq!iXJIn;D=75Nxo0S3^%`79HFx0|?=7=ZwBQJ$M3Ed88rH^y8dKrp~_7at>r$=X{
zupIi=hO%M;jtbV#zzRiivT*#lHp+`&6t9Yc=FQsbSowT)O2>zmiaSe3b~<yLhKJt#
zMnrejl+P$q8SxU?OpS8aIKqX4QA&wa{hrdZxPM@@euCtXh)qp@Ofd?M?=VHD)E1t~
zv|9y2K~;@a=4OHKa8@`rD51ZAF(4*1!IJwgA`Rj*jy2N(FWm%?h?XJOVOEHmF7k%V
z4yHNl;T|I_C7%YJ8%8Uo#H7mg3e5uyDJNnv1^*OgVg+}}V8pyzVI*O7@Wx$>*jI*j
zNWBKzC&x}3WMPoQ9I}aCZrnC4;7wIXOXymS{U7w!9x$N0UdaXZ_0@Ugm?`j?IdFY_
zF~1Hs1#!CNtATtSle(+0ibTG=GZ7B_5WP+@cF20tDbGK-bS4I*ZrZjy3W)q2B0fMP
zpEXq;M%4WIp~O0vT{dbk>cZPs);bNmm6iO7bvN!7J5h)$+kMyoN-|}|C^#6N92^hm
z`-E+kG($2jVFg^gl6Wo281lf%&_SHwTvTOpJI>r0q_Yvx{22F%G$u?MO+d_@Znc!l
zm05HZ#Mw4%_vF@KjYi(Rni~3}$qs=qKF`&=1x>^;4BJRz9l_1$XH_3cQ}uS-s8*?y
zbcoc?!^%X;o=qg2FWF4ol6KT+=2P(zW@oavX2}ZLQnx&UJ_&+VZ^C&|2OPqXuE_Q3
z*|LxqMq-MYg5OXzFTRW$i`A$y$jpo41dQVrtD>}M;YeQOBZXpG9nuDV@+W1A$1<l3
z`{hUI^t<JrQ?Jioc)GD6UXwq$e(&xTT>8j09<Pfeo<TT>)SZ<S0P&nHSMpfG*<Haw
zr5SxAzNC(JdF4Yj0#tNb3LaE+K%rZg6H>zdiQH>7YFot`=O$7;NiHiZCQVrmwl))j
zyJ+y}4&=%U0F@3{g+pk1@0BgkBVD#m)my&?KuGwel9Z{luGmItht41o>Y$6=ixJM`
zJm}a8vds+U0j?nxPQG0*X95-F8AWb<JYg3I+*9R_R+P3J&ep*GIof?8z&9whnZgPW
z-6ET72(dd^pq*|1S>tB>3N4?bVT<m&^-2rBuF+Y$7g3?oFc$S;YcA$RZmqrDnMr}X
z<dg#w#^OT+c7O{#ER*?-wloC(f|{t~Ox6CvrcllT${~b?7YS7|Wa~as!-BY+Ni8O#
zrX|zf(t-C~raeDEEUOKP+-muvNP!;vK=IkE4u~#*O?9Kh2YI^)Hc{ml9c3P%It(lx
zLr)KSxN4g`fZG|<JQ@hOjOZ#6PGe+oke{C6CH27smA1(Cu+XAxL^)!n$979iI#wv?
z_exD0R@-yk(+HU}IvUWHVLUo)EFotoZhIr2BUlOLclmD1Bc}*_{t_u|XTSby2tJC0
zx{v#Cs79MAiG&R6s@Z`edDr3SLhd>cW-CJ_{ZWe%aa<zP2Q~!%(D8MWJU#G2!z537
zd+Zs;2LVkJFXx9Tco!uT?elV@#hnA*lD;1EZU+*-0$d?g8J^6m#T@iO6`(^id@jBw
zi}9OAZDYVc+a(2*aDkM=m<XL2hVuLo41MY0SUsG_Q-<^Ca^rCtk(^BkYxbfrG!pBI
z;py$`hOBULn39}r0|2YZ&9>LP(WBXDz!RZrnxz+&8qayI2UsI>fy6RS1T9fH!62=t
znAXnsy`%Y9cS=<XTm*hb7S8#`7aFf6WcF(<twmau>EMWg$)d=N!4w$Uc-BX>_Op&H
ze&QzSW0DJ`&2T~&&k6K5-$sIPZ%&4&LBYm^0fZ(YW}5`pkk-;7!Yqd&#F4CC!fQf#
zNZ(x~^gMcicnUZ*oWC~=)kDi5;#ZUPIOMAl<5>V*B<=a&Sn`bY9*>Y$Ph;++{RUWj
zICF5wuN_*R25|@S4GvlmL@hf}eQ*f#8_q5kLu8=<2j63S4{dUIFN{ejpP&;;aFA(?
zSp(WK-dY!|!XyWime7k<XkvaIY*F0t6-{hjzudUaXh<?9^vEu?(OBB34coLg@`|LI
zVxzxajCF)s$6sQAc)bzRaHmC7Nnxg><EfJCm3*y4QP6Z@smz(&1jG+{gOqkNMo#j?
z1fg)w_Pi!UQcQ@R*@nqwCN5Fci*_qS(^t?qr@QKOtk8rFh7XzHyEZB7QpuB8t%A>z
z7@Z0%*s?iw1c{T5F>}!SR&MmO=0+zVt56zXr3*}z*DuH?*bHH?c@_c(0!sj{gnL2`
zs6_r`tuz`L<PQ34escy&TB}GS|K9Ffc-0CCO%AmLXQC^Fq+RUrKF$<3tv*t7g43gL
zGLnE7y@MJLQV7@SMpSYeia7!p0b!Y>!7h1zr0Jyff;rtB9GIzQ8!G7*+?2f<NrYXp
zO{FFx304!!8+UH6UXic^LILQUb0p@kU}Y48I0Hc-Wq=Y2Nm&0aEhKh{TVV(Yd$I#3
zYB16`xfKRmWF}AmM9s*UMt}^#GRZvS{OCd5+DP)FZ`=(WGqq%#Y)_5ZF=NTFp2p`Y
z;|SAU1y)V>CYqXa8TkQU;9gftbiP1PruBx9(<Eg#ffWbKA6@g=N~_FFqNgk}amh8P
zi!RuaG$BT8qoI|z>YS7ec_~%69bYK`qm0+}XbultZ$R2ca+oSjhi@`j^$XhR!e~{r
zK{~^*l2!x3K%<WmR$=pg;&-?XWsRJyj+;OpiyFDcRG-Y49FVY}Q8@wvGWeP>k0I%V
zQA4#xY-kR0qBwU<5V__};|~6Fn=;~|y6Gg;QZR+}q*9~2%sC5bgNwqwrG!`J%;BcU
z_A(ez8C31y98MStWd?0u2*kC#XM!#Eu3wn%n2-4>!#Nh_mQBbeN;c(zhNCV(gIuY3
zPed?;ES*6Q!H&nhCoz>WFc|s>+rq-V<tEVilv(@iE5KU($SA`~x7;Xbi(8UrwT}KP
zi+s2ZV&ZCTTDIv3P+x=J^=#-$Jzeg{vc5sEb!`RdmScP*3P`NK-(a6<gzm}^VPECN
z<kRq}9>2qdP%!-ULwCd6ymsA%WDkY;mpuSH))6eKsknj?t2aa-ena^nC;EEfdo++B
zI+(6Z6@6xK2{4vtUO#Iak05gB!~cCZxIp{Q_hePe6XFo-C&jllD^rPtX9sF^-!H%A
zgmlP}5bm4=L$I>PkoXZGe+Zf}E|L?@GkjP+gvE+p6TY#WUeGsI&SH2<!=*BW+6}Gw
zzlI|G6Yf6(q9)qsE95w19n3&G%qF^u1Xv^Q`}VgXSa-6G{ml5idg=Mb09Ct}-u)um
zxIjRJb7)<<qz7a52KoA}L-ly-A*ayGfnT{(VODeR=Dq7+ImpYyszvn5U}kXJ1zv_M
z;UPQ|5v2$e6~Z2@@_Zgg_7gsUj%+y5;5(B+=(q}IDWL}cN)<67#3BSVsch#8qW`D;
zf!?a*)to)hEeSQY>}BFn$;~s8Dx>|j6ZLTdJGv8|M{Mw`c%Imfej&dT0iF*-3j~`{
zcM~R#$zTk-5(u=5Xn`V*g<I{J6jY8s^6LFJ!7!S%C6VF=LrCs0S&~bI%~GB`=$fF^
zq(Ky{HgTs13n<F>M*ZXt12<%5`W1$)@k`^RU1A?Hx{^{_%v&ilK_h6a7c_$)gNy^M
z2^@RSB~Ys%qeZktO@o{wv`vsVJVoIKvx3sGPpL!+@YML&iLov#rcB215ue^ah0>aw
z8igQx5Ll15G#FMH&ZWdjX)b=rfw7fxRamZsWkngw&{Z~}K(t-1Qxf&cw8)yds{4SN
zTE?5`f+$H;OT%rE$p(n6Ni`ElfXJIQSe(3=8E+g7;mkdv>anqvD8_@6ZH?am;}EB~
zEHAV595h@$Vi>O{T#RcO(=-)WmROJiyu}+a&-C69w*nSGg4hWZ7mRnKW|Z_N*pYn`
zP`b!N1l>iF>I2O*Ed&Ykmd$|WOgo`=U`*u`BR>#AEsYh;VhK=6Y^9_5#^GT9fh2@}
zGJp&lH&o~qWoxmF%0Fopl9<NS4IWXv4Jfi$iWRW~U}+kWUpmf-1iMJkk4B=H0;Z(w
zHFDyl>CCr#+Y>~MKZdJ*n=)<etqMs~gVh6xl`+)jLNmxn#5@^E{Qw#6TB2gavPfuT
zwXjf({^wGNaq{xa-Ol+8LR41?y7%FGbL^a#c92jUc*A!ywU3+6@!&D46gs!3(hx(}
zqL=RIk)@0Xh;Tf9mQPrKCgWOctYHX~0R;AEsz)rW3!;)7U_~Vm&LXP_{H8)8blZQr
z(?@=W<>BMOcr;`%Ub#Zehmzq^)rL|dI4XfaaN@d#KEMhgCVFq5Uh};@usKk_9*HtF
zrV=aF$@6X`1ain&uESgD_Z?!KRT<s+6ohEvlq_)p+oa4#h~naJROcFRJPX7SjELyM
z<otI5h!E`&u`3d(!aC%ZeWVZ}!<u4mEBt(jFqt*T(&T`&$ELHJ;}l5!60o^*kfi_^
zz5k3kYu-B9Ubmm^r8{@JH{M#lv9fya_T7#3xESt)VS^$`fMXe+d0@~oK7lR~VUI>g
zzrDZMJUQOG^5e)V2Xk8?=;Jb`>GQVyLHZFmO>Z_6Jg;0@fH!>N9_JcMySt6%C1lVw
zbDJiKdx8#iH_AM%qXct~jpID=3ZU-mzEM^5dZ!z6IxgS+Qt|W`xOU+q-JKkQFF(dc
z7>W`#u7<}CDJ{9!HR<p^B1BPI7MV>(!GqsOlrR(B9q@k%KiELJ**HK>u@g+fMODMf
zmh;9OAZE&Na%!>T9>NzRf(@b^FiYU&#vKqfNj|7@K)GQZA`gwzhJYXzVBLlY!9gba
zw9*Xgdd8uc>3jF=ldzg+n|bJtM0@s{ZXbE(^M;&Bk`-nQT1r!P7`7uG1R}AJ{a1A5
zkp_kQXw0-MxO|Qwb^`MfBGpah;{wirk;57nqO!(~z+KMI1}`TK%Yq}C4vc|edq+n=
zE#5{Y(T}wZ-_Fp5B52k_TLZy_J16X60w6RE5b-C3T45C-eCx0vi0y*%>RTjKg+E~L
zpfoKlyAL}vNJQFj1Rmu!2y<#craXkXOM_A-wvO{8uGu8XsBXpmavE6I(S_DV?@%9@
zk@zl|b509FWRDn@0Yq>iAArn5IlP&dQ?4xxd{RGeGT*4PFJY-!%ETfEv&Adi`<NQo
zRPZ#CpvNIk^-}v5$pDmA(IydA*)X7HRV^owTU-JfOX@CtbxlEfUTRVAVPX`fDZ3+v
z5@}dt0F&`!-n240&bwRdKyh9g=?des_5=;zqp`ggU<<=~A0S6s{W3$W0UVmsT+;6b
zdj1#y)p{xEFm8+>HFuF%`H-=+7r~^fukX4(X_KsJa6;(wV?`o<&RXzEApDErOX57_
zwG!<`Bgy$oV(B9P$sZ?zwV1x<MREePnE%X+3Y&mluw(NVey>QxNyMxZ&Y-ebC{g*c
zORNbc<VPx&PCcEB9M&+#Gnj{EKnzFIu?RDlYMqUbURz$e*;#HM?luvX2yC=4&3ie&
zyr*okUbYDD=rpDfS759z7?Aj9Q!J$ixaFmQ;#ifr#yuoGLg-7RsFP!nBzy?=B%K^H
zpsvzJ2y+BzJCj0SG~=a^<-0fYSrxP(V%K6pOK1QL6hZbp9yVkU7}AeJ74sF#B5a`#
zG45>0)$BOVC~iaY<RUu7Jy~Sj+fd;!W=iE?<K9ZbEw-cdC;&*0V@ZhANKGKduo}W(
z@@%;C022J*tIF|_FN8_EARAqIp@hP*0!Tb$O1_;DotXfFJ!I3-h2pQYvQQ#(Wn+KR
z4mIekKrv+V#=X05-oDknyVO~}-1zeH?OShctjn)tjzH~nR*^hQ_Mo7h1!@HGbueL1
zxzLz~XNU*uVRF7<Tt!2AQXbPIaKdM73|EwsHWhwt5v9{2lzNR&LZpe_V=6WXOi?+(
zb{M88U=fLG6t1xXE&|~LL?OhD=`&de*iz$802VJ2zHoBXdjc6KIE+I2L!wd(JPS6E
zxv$Fzn)MklGsihWt5^=AFOnxG*4Xk4QFobW52pntl*=g-3kuhGszm^(NeSCbs)s3X
zjMDPb+AT#MpN^@d5d(UpO)-rsB)vpFcOnFKPX7?8pL*i~2o+OR%A^u*=?>u>G$JRb
z%<UW}$yWF|hFycJhD)|Od})%>TuEhJ$X9KO!}_wttpw|5K8NhNR27CTz$mEvk5!Y%
zt8^3`OO`s;NZk<eBo`-DCcV=5VPb!>t@}J-QW#IR713oYpRFVLp$7lvTDwA((g3V9
zhv+iv#=Tvvp4HEAL68_Csw%2*MBIik!WgE9>-Ma@zOlTzUPcD6_(&K|&n&8dK-Tt}
zfeRw7g1TMQF4%+&Nu`m1CFn|f|6sI@Sh{@vBKHBQ)(y_w6Qmd1!#!uZGC_9jPM*#e
zdP3N7DN;Oa!r?(ls4XwSRbap5OdQuMAJt<16CoYoF7}AJH@T-l8eSxN?H=qgC?RCD
zX~D>G&f@qfXtc96d-?s_;hxoSJ~_H$=P;h#(Lrw~-Ar(r{CH%`F$>!Bn4COtH0f&1
z!;jEL=?rPg`l;qJ=sX2Zqmxi#20+HkI2w)OUhL8%+}ZBK7Wm2+fnrXf>3RLPj~K(a
zaY@!%4!lcpl<0#<$}a`=#Q*+p{pw%UgRuc1hX8J(7uPD0tCGfWn|CSgf5upbN&!M>
zzFYE1n`kuUX0HT<s$0H}?K>=-wr`Qw33H%b8Wk5Xzz}_l;rx`&3Eisxf?++$<Q@Vr
zh;Gc1{Mx0exPFPDDGP1!DwLzy;+1dY9`k|e8s^~QfT3K5nq9K=!dw;?c(D2ia-dw!
z<A6bK4YCU(=4ckm0~?T%PIv2m>Ts><R(k_OJsBSa6_CLaDCiO$o4QC<9z+JcPvTLs
z6+o-|92zO4(4xOgGLzTVGYTwcjcf(MMuemqZHuQ0W&_km=s=bTB(yNLtXEmC*tSP~
zVc&aH55a2;1O?|=Z>S+?w%o?t(t&Qt4iW5jFF_6UnpxOiFck#1!dONq8w2rV;sB6g
z0DPES+cFMHf;|RJp2}bTL%L%9N>zV;DevwbblZu}YO$f%BTnPZ^rFg?J-rQAxB~bz
zX~kq1?LIEZ870CSHhyTRjJ&S1sTabWMT8u&zWVsfnoA?*ArQvWo<Ku<+`>)p{N@#J
zkc4PR&V0%;e}ck7tqlpsuEeQ^1%Jwz>VfWBNFs&^mt9S`1Zo&+?cxd!#O@G!N2Yd(
zGXZbJ%S6E?2iAlO@>=)~8Gj2`pS;fjOF%reyIt4Zdm$F*t7MX{>jOTn@V5q1FvB$D
zL7`lpkKtyId!4&Z)72@xz^xiVMVQj$9z>5~@(6eB;4ag|T@wlYWV`u$s&^?E3v`Q(
z!r@qBA-5l4pf0Ed(XO%vwtl)UV<}$5KIp?++Kw}lHIuZtjq8~uD{qWqJ);z$JDpeZ
zq7`R}(I4U5J*GoBiKd{G-sgZ;iPC>pa+@N#Ns`tf63YEgtPxQm+wTs2a|9BQl@1%z
z2XS9}y<6f{VlM<}Nk-#CX?Qw3S&W|oT$`Hp)&n4^vJxA$I2)L?i&9{kYKsTTNHB)T
zLBtkdRS1LTw)h+)YKWLe$o8P1Z{y{Wm#tg7u?o9RxG&?n4t``KXp<@ZW6zGCwSQ*#
z!_U!Gmc_%^j|v1;R&-9lj?^58ii~`i3~azC@cPtotI1`TS*sGq&oxQfNGd%wF~+yD
zdVPas^}^mYCgaq^HIOfWVP9f|7QxDv<Xq8pMX)nahh*K^Yk&czEZ69=-EJ2;h`5uh
z?0rEOD<5Ng)z6(Olstfjw5lqa`u$cWF{E}m0&{apCNbH(67eVSHllbypy;WI!_qNB
zIwn+EPZpo5>}#7Fi~|WD%j*_3q7re7{<Vh-6G-H$eT!~|%FUIm=@E#Cv=G`^C_`{J
zo-?D!>jtLKqA?1I{^{4RNEbmd%cy|ZC@e;!%GW6zaHr6$dMJ<M2#{d+3C*^j6#Gzx
zt_&Z14-bNrA8orgL6+jwWwaNS<%rMQ@mt)K1XFCs5F`~#-vW~pz*<@FVB8g$BLCYs
zRERwVq>v#=9=LDiX(Ou~Z*y>Y^T#|-TVfd6DRO2+AjTgHAx{H8?Vq{0qlqRNUqv$|
zuvL<~ZN|KAdCfen8oGb(yY~6niuK-@l3p$BCOWuSq8Iq`jK(_;+8YwAC+wn7kSCo+
z1`jq~zO)(0`9~6rMZ}ZPggWa=*z1>WLtQLhu8p{!YHV(yh8XYX8sr&X#ikdIdtmSo
zOZR1A$b`n+ghikR>fKd{-IBQCuq8;XXc)VIIt&K%2ri};rnNiTTTIB75XW|A6>E`m
z@3`s#E!LpLDTJ%0SkoAzkg<l8ub|}zOU>&o{PyTW>Dk3X!y$Tn3?>S;A)+J1*+nm#
z2sp0f9JFULY>gdeqJj43F4b7HINc&nu?%J#6i#UqLhWP?@@gM~g^QP!O5t)(vTa>P
z0Q{dlp><L3z^%%<zOA^*sTfze)ha^30qaJvplCd7T$eA82*n{KbYP#hbH28n+E#o2
zJB{ci2eo7!_LjoH5FAiKTE5xyDGCR2wyc#AIAq~?9T&X%xOHd9yQdzEI2!&Kq;eCt
zU#FVO$)#*0IqtA&S+pXHT7)(W^W)A(s|8IgFw#Fqk76MsVxb;^Oc>)6>@a7{^dxGs
zK-5=Pn>BDO=?Ipga78&3_c$9#9+M!I;#d?L5L=C|2Vnko`%kMAnpuF2spE3zItLcB
z2l(exM&^nI-&TpSc-qj6K3RT;sUvh;14fv%MuckkWX+ZEfTc)?G;PiUW+$~NRTdRg
z$QQs>4O{qKq&HJa@~sqFjh9NHRmIY^>ABJyk9*J5d%A7RUKjialcg4OW`Rd>%@iGt
z3}3jJx$`W{Ac=qm{Hw$o-J{{aooPURTkCk_N!4&TB-?9e>LS~pHfQc}$-Y_euVNqk
z7_y(KpKOC@KlX@eKE36X_CZ$k{mFck&DaeKw1b4+=RDHyN#zF`p%=8Uh*G*L{A*xw
zxov?s61JdtsfyUb-|(4_9L7gkXeK;$=VkrI0q|zqFz??+dA~Qv`@1pX@&HrgcZm0K
zABx=)bc618QDBMk1(>FiJooCBMXtm35;AN`fgouE+AKozr5jCz0Ie2bc;;3k@x{#S
zAf2B7LSf`s4B6CartVG7a|UwobxeIBV>iZ%%K{e$Bi?}xSxnH)#ZE41{-nnGLDN29
zX{w;vT808;H6DUlUV!?+xr3$Lsbvd5hu??^FXGOtXXkAKH8~eZEG6YaY$=3$vUC>_
zfhCRixAIFY*5eVBZi=1tsU?Db&ijWb#msm=J-aoCG^~<oLEv;t-mL4EXaZeqP^<LM
zeQpVxC-$YuKL0%Bg@~KZY#D5dK{T+rdnB6b3r}<s!&tJE>C^xo!LSk@3FeUBxDkza
zdAl9_>Uodn9Qo-?()K9m1Ast_D+Xe_(f3EKIT)|POpfB*?5r0$ewMH-v4J>1_|W*!
z16N>h2{T{J0k(e?FZ#Pp7x)Sp0+7=b2{6V`>-vdG9KK;PX2f`dTS9Gs2r<>s%pkDz
zNAC%65GxYsdE8C&Lz~(TnjKgAp|Mq)2vdy!O^&fY7|4TBNh93Khg`W*vzY`94er3p
z*a-0qC1S&=&q9N-1T~A+2$hPVy&NzAbSOh+5aq1lyclPTqV^8)BX%_%QS)bpMR|us
z`3VF340uqD8WBgP1?r403zhfK$8&1|o&>QGxSlXRlHhz=d36oLP+<&9+vgozjqJMs
zMrjg^3{n7HKIeCXGu6I}oSEI`?e@1%_Hlv8IgkJpX3$1L8o{aSjpqhPe<c_I7@Q%k
zh`ZgAoi3P7+bmG5R<=Dn7}XySrq6J~84cC=TQ>}<LO0zzK1o+aFmvd)CkOrhQR`~^
zM_q>iED<R63o^t0yAmm|81*R?<M7m+t|b)Gg*6FyBOwSAfZ7E=q`WN}CSw2nm5=_&
zxmWmk=fyj%hp+vs=U#dEZU6M#D=W8GZvTaM9{y<j{gW#{iQhjSfB*HBzmDI3D*pcE
zm0!m1FUH@0W94t)_l@}b@2vbC{C+e3{;ieY!tdW*4gLN7mA{YQcjEhhu<{S^do}+4
zot59g?~VBTcUOKFzrPZH|K7^);rGwP-@m`|`}n=Z-+vJkT=|2}AAR&Myz`6mYlx~5
zEb*rXYkBA4>wrV&KmDjy{F7P5GJ9mj|DBI^{uB1|gT6Sqv#0?Z9a+&&e6;g7S@ip|
zqF5CM-iGD075k%)biSL#KIMy5`tMqW|L`N7fASGj_!GWD$qv*CzxYV!H(2=lec{Zq
z-3s3ONar~V{*W(N4sU0rf8ryZ4ola8ffqSe^dEm`XOTq<^Q%aOWEFn%J3IgAJ5k{W
z{76CsVqF+3`D@?V`HxsK8wF(w<iddC1}?SZV*KHUJHN`J-{Tv#b+gv~%@23}s}G~d
zzu=2lYke#8(;x2q63gV+G9H08^qn)|#ELI{xbu~AakY7>C_ew;&QGxTzvxFW!Ho+-
zgF$ApMt|tTor^5>abHTj2CTpbKiv5w3w&>>0Ph#A(EoU$^MP_-FtJzLAuILoUg-P}
z7tmC;c$IvURrr+)onNo2pzMipW0DQWD*V+8onK~!Px!%t4kA==U#t?W+|vu4zru1E
zB4_4wR`AY+&SMt5<Qt^t+YVKh69h{joyD1wq8(e!&tK^LBx|;OO+S>HdaqyT{21$H
zXh^4A7=`ztBxTfN>tgOg=NgOVxSbWXk{`d&`4mfj%nt@<roBFYq4P1m&KBkY(qN5g
zi7p;s-ATU(dl#$lE8o$1g%xrFNtPAyRB@QmpiacWXlG^>``^*|<?lenPx^rfx;YHC
zvj6l$ogJ3VAt9ZNaStcHkXH2fKGgYtKZK$`oHYyg>1{4M$`Ta|yYJUO)cGCO`H`&7
zU@|%Bcj1V~ME8@fM5sd(y<O|+mp;__HP*@}AcNY534Vs}-4AvC8p~#2mBxy}6s+-2
zf2ebx74jJ~z+rMYx1t~aP-l)s8~Nymh#stAIxu{66-n0L{PxbbvwTJkaJ$mi($?zQ
zw{`yQZ%3(&8pyt|0{_PcJNLc~1+rtvSqWOPYai_VZy!XljItr*<<p2B8K8dPgPjEy
z&nBZb49plRmJYIf*8KndYUdLypTUWi0#=kqk`?{$U+w&#ucBza28=3HgLVJStDXOj
z6|%7i&xNhdv-T+83jgJ+o&TEUbA*VcP#E5GHke;|we#C7n~^7PVPxh0(yN_cF12Zg
za&bq!c(wBvSvI3Hf--R<?i>J-aWALdW!A^1U+p}v>m%-AEBn8Gpwnd8j0)O#V%f_5
z{zo6^{Lddi%?uGZx?^sZWJhEj{jCpl{z+wV(e0Ir|J(;U|2~Un%uAQeOL?<+vbJ|V
z(D}10oGq*jg|4~fvAq6)PLJg?9!46M4FU8Kdn#-Fe>vZ|#?tL<KeG9q(E-AFkH(TL
z28bk1_<VYDI1H5hAD{31Pv=p)nbp>^-IF0BVijuq=jS{Bh&3)}H5jgDf?$piE^t^_
zAKk$o7@q!aKhUt>Jm2}xSU2Aj9GC4Mpt+_Bw8{O(`ObgL3fX?M(FFq?EBfo_JAaEs
zGZaV47V(6mpSG3zrSqL%E0v=PsL<YDKHvFkESE2_u0XN}d`HBP&cAjZ{}smxt3E(I
zCh)PMfBt;u&$DPo$W+-+T_qNVf=%aNIgkHPEJGSBM^`9FaupBdfBt;u=S#y)#DD;R
zly8my-1*MWv2X^fAUrFNCWH8M=R1Fn<uZ1YT*r8HU`_p*^PNA#A{pVf6H?lEa=!Cb
z7RZ)xSJS<~8obB2Dc07F^PL+knt_IMMjFU2klu=Y@_gr$ES8O4ik<BCT)7`S-}yn7
z%Q#_{LlVrV*47W4@B9GE<a@1z$yWON&Ue0#rL#rhIS<z?8fQcH$9%p@e3!ZZvvZw)
zb`Iq;5*_fY3suH04B@Tp|8=hO|FCSv*^s4DZ$tLN^dQgzAy9<m@3xA6c&_t@te62f
z@OydDKRDO<0~XB|OAa|}>mQ!${6iMV*i$_PpCEH4LPuM{-#^#+eHQ#)PX%HS04wzO
z&UOAC3uOxx2`Wg2mHX{;o!>6?symI8Hx>M?bDiHR6=ab}>#W7Ud#>|$Sui6s5|P0O
z4aBNIXk?{+^IYdQS>e0<eD)^00q1_<T;~_~K0{5UQyqi2b$4U2qjQ}j7R%NG>vf2j
z$>evi3B!7S{#@ttr4}KR1QUOo$>+{>KF4Aiprb>E75dS0ogZbPY@cHDlb~3U&z|di
zmPN7&(gwobUpv>i#`oD;3HQ1)I-ygsmHW)O&SzLI+f-e_{chJt>?hB4{taxC6R2u_
zaSl^bhRUp)yFg&crP`&LbN5>J+SPX+F2DN9E5G=Cu_%H&p_aYAX0Jcl#d{nuh=Cah
zLN`VfwK5;Sx%}qR+jlm)HxL;3jinna>-OUN!;96md$-n>*Vki-P~i4@cYW#Ba%X8}
zd9B-7?XKQi-|(aR?#ZJ85=srmU>}73lT^#NzfB2%{rc{5IHBwfT3_E-+E|t-Rrcni
z_f|J<cW(bQ-Z7F@c)E@VLO0*OlRy3NI_6e*<dD|tPFY47a|Ig6^J)x|S9KUDFSC7-
zVX?fPz8}ghWIIb<X9L!aDbHuw;ZiUoX$6{5E>NI8QXqqAPR8VY1|UZZ%dHnEsuZZA
z$?_tDU5Tzsk+Yl`QsI;XNy^M{$Vs_Vk4`C-(Tj8l%exG)`9Uu4Gvt!@fgh098GuTh
zg%rtvYut&v$nZUKHu5SPRn8pA`;0NH;IyPr21E|M%)UvvY>F8-D6g|AmiWr@a8$Nd
zIwe_`tH?sm^hzVyoO9;6JdP3{E=95piPN@<x0ixv6d<I|Ok#?ZJhM0?wQ^ieGFlY5
zq*S)3sgRcvrCf#~71E&;Dha5OvQW3fbGN!!Nsg6bXA+d9MotRE11aiCLg!}?@oU?3
z>H{g0p<1qrD7TT6A3?1lWwM2+R+UJ(94D)FFjBCv&kA~-a-Y?zA}N=xYo!KCie(!=
z(`U)+jBv@JWcX04PAd-*%vptUPReET6%6(IJ}HvJPV|k1fUGv643SjX=oORF*vu+V
zDLtc7wPGeROsA`rBKZK7w~baCHi_&c)k)OqnWWJSJT;nYDOlEk%j2mkbSaUqQKn=M
z9B8&b1B)P5?4>ILX)>b{r(G&Yjf{H|Jfxae@5iJ-ww?VD46qrTz#vxLq)0=>g6gIj
zq+oi-SM7il1IgZI%$RCi<<?00$aa;ytZWmzJ6u^wwQNgwtM0!lidEgJNwJJ0BGJcn
zAEY&Tm$4;Sa^B0@s|+2iTDm2a+Br^{0aUnhloA<`Vdf5(ik&gN9t_S=w#j<4fv`8X
z^4XbLb=4|8W(ZWcxs?LhVOenxTs3aqLRS-`jI2rsr*K;>?bJ<6?=7TAzSvba=~65w
z7-XsH(yHkCojg3acJ7s9-2NOK0$5MlM>rdpB&LmY(g$#MI~0Ymt;1;iPU|cG-nYH-
z3PPwL#DrLzVB=kopw{F!b`aB*nf&a1e556gpGkrvd$}}xb{TH4JBU%iybqTf_biM2
zWlf=SAvFt?OKekUDI3N^Yj)flPN0Z7gw`wcyo9UC4W_)}4P3dH&Mp*Q1)MOA5Eu2r
zg$pV|!{Ulx@QSHy?ucHk)fF{Hw+rKx3x7IIP-c#lR7^K`6QjFf+U_qSqoB2~_CHY)
zyWYqLNvj+4tqYFs4C%-)t1v$+jFG8i7^_U|mEeqSVR2mCO@v+x`DUaLn&!2Q^CO>9
z*!09^op_x?K!?qvW3!NAmYuGssJmSYfW8N3aVXo7%tf+zaA7qG4%YepeX&{0pCxlB
zEF6rex}gR1w3)}twl}mN3x!f`!mZFko4UcFOZ&qz{5+5LUHV)9ZOU!3cqUyB9c-qD
zxRD)A`ore^6e=pKP4hFjk%J5!B9ea7HIcuW($0%)T*<zV{k*?(8}}KWl6I8uc6QTR
znl?Y(Ye=50CTlkY*f2~J!>hPk@SdT)TDJb17vq{4STEj3B;Xu!g_hTBs%L3COsdej
zhB7eEg)~O&vD{6xuy#3!ZL@1$Jo}NCNE=%VrQdnE^4^Y(7WRBzFW=@`Cx4d8C(wq;
z=bI_*7wiBj;<iu`yPEmVNHq%23rLBuqVfI;bw~iy027XBDILTGvCvBva~D=%ukrBz
z``3eo+ecSkgZlVKrV{RUAa$xxUin_Bb9s859FD54UiqGt??G=@r`Ig6rq%xLmG4IF
z*7Vv5N37PzS3ZtfpPpXJ(9f!T*UERH%EjqbBHCM}kF9*nl<ZLh{aNbK|HRa0-gzj6
zX75Xi&pI0^d<K9?`IlH8sdq+%kUBFX9!sA*{LJgZtTk$xHIap&rD=WTV;_uc8l4Ro
z{j@PWx?s$gXbQ3)?IA+PqR6n!Ed@Slg9ZX%r?f=-sJ{pM1$`EZtm0WLc0rQynbnzH
z{k24&MAqGG=)p1|v<=RDIA!s;LkwnGc~Qi`kvhShEW{qV1;&bP_nL{XS!>uw9_!g_
zA(Ad@NqzxvtVlz|C^yGq?JajI4OOwx7YohqA<iY@scpyHl;lMA$GC~wJszJNKaj?B
zBAXX`JJ8PdcAi}%`!zgjqxEzs)XeUSH4~U`EJz~o>S{`ojoNAVU`h#eu<ew$HEe0{
z*t^U~n<wL;8Bt9%uytGpa;}84+y<l($8**k{&K3`kf{@vIpvOp^T+VkVA=VwY}=Es
zqQ|jt&674`E48lbtR4<_pW>C^X}bpwe;=<9?r!qc%Z=Bx$RjNxB|p=+W{V)x=srx%
zZnA*(YaMkTxwmXZ-x6YYab-AchdAiHgEp85+`k`=2D?ai8yde7#@u~0Z(erNJ<QZ|
z7$-K>mWiEQZal|HW^Lgdi{I%EV$aD(#_um|Zd^TxB#5na;er=9Y{#|iuPv$HeS!_-
zs4P#*OgO=%3fMIZLQy2rVUY~1$xOn~9AWsn4Nq8srJumjPb2k}@8Bu-@TY$v{QjPm
z27dSXTa4f5Rz8hW@CUMav~yjOwDZF(_waG7*IK!Rdd(6H+SQhJU<y8i@;_8AU(we*
z+m@T`kA=6<rs5V)+i$n6|IS1C`H^h&_HgQyq}s<zgR!coj*`_+9^Tb+>Q5HWsmBKg
zhtU6mo+5*-98<S`?3GvMKDBT%882)Ph6_j?+c<i5{9rV^Fb97Tisg%v;ub$|AYBm>
zQX+BqYgb?UY-1hf;m{ilb~K7`r?-Rj2>r$#*yqw_6|%UR>NakHP#?7$ONiku)h9q|
z#Lh=5n|3$>nF%R;*Awz!unlLc$>?OfgVP5G<I$sisPyeYDEn}Nf!dKuX#@R+pQFhI
z_?THL+|7$$x^t(~{qnsvxz2}4z-`?8-{^LhH{QB;^Wy!*i;~&sVyKK9G^TQiPo7Qm
z*cH+Qy+BlTz%-e5K6Y$pDi+nNNJ3h#UQwkB<njxlb<ZJ)GUxbIor?Bn3vBwBI3{X9
zqJE}`1c?5L>`JC*3XH=Wc(`iMvZm!YLcX&-i+clmBc`_s_1=2(rU-`P4xeT<X-*P~
z>3McLV%}27IdNG}FsWU9XtPL<FU)5hsEQ2x@DO454GSbB8=}PHu7>@qK9JxM!dO{-
zh#TNxQgE#;Z@j&Bw|nE>&E@X$+S<Lf?i+8v`R4N4<wj?@b8qdZy6a1CE_Vf=Zr{c8
z<-0df@a~toUs_u0Zrtmxq12a_D|+H6x-#Z4ST+)2g8c$UjQXnEG_J2`0C_nl;K{hT
zY_`yx%a)FYFF04=nALlhOkr>)s2q-?)wW2DHjQp|#rA?{mmnp(3O%pW>J%ec1#fVO
z6BRSLkh~!^2=r~yQd$&f&NaB!yWt|C2QUx(92Nltf?d!9?1By^n@Jdp%^mOLSWZ9N
zgs?o5;i7=(8QXT(h|xk(E%}%RNk_WM#5_zfFJx++F@8@tv8atqvr;&=!8mXYTDX0F
zRxu+VnNcAYuwHW+(2ow$9VaJ8)Zu~%n${$o-4P+i#D#l%)S;6OxBM1$XLzu-N-U%D
z83G8h5VBuoH(YNN_E{_=xXBHpq1Qn>+^%CI{G@}!T1!JMl(ylqlBki=5P*g0N9<k-
z#WA2v{GJyni~t1-Px9h@$TA`(L$`>snZ8#ui9)ylMljKar5GR#+B!m-%h5ay-l-4f
z+rBOJuOxpLyB962o3`$NyQ5=lf&;;?{;utWcB~Yz2PhoMI2D@Ag`LQ4R;<KR5QIYG
zUCr2Ibqh0gFB<=M)qQmf_1(+Njg`%4<hXS^Y^K8QIDJDz)JXXE_1F)*puyEmgeUxy
zh;lRrBDH`B!uYUtuQf0@$cH3KNjcs?O%TI&9-1<_OzS#qzMM7_rH;V+Nw-O<C_7Mu
zZjwoHx1um6kMWD9N+WeONbg&<t?2@C`xAd_>&GQwH%XHin?04xFhyB@5|sqL{B<Id
z92|kZ-R~WmCk=#VgHBo#_c{%bb%({$G+HrpiHStE1>>4mIYy30q}Jv&O|i0j)qpJB
zzw)`fDz%D2$tJ~`d^7AnvA21x3A`<e<-u6)B84(v<%N&o9C1`EEf34(9#xfll$V2x
z=TW4-lf_LzB~L}~W2Oa&^CFOHCrM)Ht4B;p7#knSmz)^KcR6Co&cEKsc~;{<xu!gI
zaP-A9(7oypd)o(n<AIY1WyCm7p%;4}_l3LZ?jN1R5}LIz7RdqNt^{ufTpY)<HQXmA
zIEPWCsMcv{bv=gse%gv$49bmRMraGn=>4><5U*slKjw9BT0CqdPzBN6L2P=A)aSdF
zEeFUJpxSPHnW>z36uFvqm?8W`1j&}1R`z0xrv%o7p;JinP_?lYfu_WBtDOQK2tO*R
z0%Lnu;wL^5L}(N-A-*Oq-08fD`}hK`ST?y>&=3YK-Bi?xZ(I}hs#q6yqzBJ%dDf#7
z8W;eWd!QY?a;A|HcR|)%OcaPD%dE@j7l2EsYcah=UkF8A(v`|l5nBo<BT`$!7DMEj
zM_>V5#V@t7--lXSywkIe1=UlMF1H(I?#Ajgu!E>IP<22>CrJ+)aYJZ0I;z-6$WB`Y
zU<We^fgfSECVJZxCc_XazCV5c#G=FtZ4MFH53p~UgK*l`LxMo0I_o$3(63~zfVNO<
zkw`4yU^T~wtUlud9Uf_FXU3X=1|P6HV)PH}Qaz*6vT#IM!@x7_u~YCw%^#4JksWB0
zNKtZf04+VD+sBzf-a?;+bC6bOGU5Wi(H@%GU(@E5hpzK__d^e<7frdEh8q?H#v79y
zC@ndC8wkaRAYvo1HL4tI6gR+p8l8{qL~%9{bmeL?kQXG&>>byqc@MJxcp&RXu;8@o
z@zy4`>5_O#U>g>Sq!<h)ZdgRcF>EC1qcxxSM|P5qR{koi&Fs9OOfR*S#Amk3Pfg2(
z#~BRbk3u0>C4_5G3~)Ni#x~*?6$Et4m{`iYEwHO0z=8PJMIh(QqtX6F`<o7B&ar~H
z>-BzY%=rF_f_Y0pFsQB`-sW`zAgb7c#c3~7Gid=MxPZgc>LwO}08o+;MF`|>Cn-Qi
zQy3aW7~}wDHCYl(@#mN?zTw9$n56-<(=Lut%aw3!v4A&eo}mJ{cTqw?FFskc{HL%w
zL;A9+ku$b*D&L$hHz+0;;%t$%7U%A~F%VnrvlT74^K#LB&$~+NRNiRCHHdvrhkFT_
z?<3gVyr#`D4hc9YazvbpZ6nVz=7=c8tMwl(1CV7bFScnMsOo>>2#7H8haDi+4pgmt
zOHrf>$Im8<IG~$JA2!f^w<!uJeZ3$ECp2*|u9696l>j(3e&Xa4ddXPb!4iwGyll})
zwqLXNg?UWhs%|dz_!5YsB)m;<;9#g?p*R+taH)<Td{tQ2Y63f@LPg7Juqo=3g98Lx
z=Z*T5cB_h~R@2sw@G9UNXC&|BAHlmyYL)i;yTC!J(|OjlO`drfRLrR6b)d#^nr<SP
z@&f1iXVXi!?(YVNBOvE!R>FfmOQaMjK$10F4Y^c!aD03;xxTOf?|Vc$<8|}`QRE7k
zEgEvlj|F6c-Cj5Zi@Ok_)Gmym$A_9Z(FK1ZwsHH(wd7|GS)TQhB0b~EmM<}$b@zth
zPDJa{qUIj%V7FlygGA|2HbFcb!ZU#?I9#vb+7A8`a5|5@vEI=>U>z?@#4VoOVwVOs
zE?H~nVl%fp43C?)R^Nur#I>|$;SK*bZ|K{sW4?HCzj3LWq|45mzC+jmKLKEi&Gn77
z+c!44tMK)1<|9eFi8X8^(aI@p4!+>00@RDAT@#gCVM|B*NQx3Jm=vI}+CHKN5R(7k
z04CuuS)K@?DuE^+asDe7jzY0U)qx@+GyKNNnq3!%Q+ImSg1WJ|*tnLTr*n-rflpQI
zB84TN{e%}9&qQs*9S?dtS-h46BMzl7?#*T>C?T}fn26=dTn$hh(#`k@(#%R{52#Rr
z@`FmE(8%IdiS=$0Py`kmWG1v=UZh56l3`q3Oz<u)ml&8GG23RLJ|%IGF5VjN^v^P@
zt64)O2w)3B3v)O^h&WyeV0Q~FDzFi+vZmq>5D(BbVznE!ngrcUZ0u?%lCP|sA0SYO
zbk!}Rvz#31@Fytwk)yw;NXvH~ixm#}dS-=%`I&O=!oVLFO77&)8dj$YGYcn8C|^2M
z)FLk*y}Mxx^@OFAiDP@wbOmB)|A=4o+9ltrtd(lCk@;`%n1L0Q!2^1wGocK1R-VGn
z@HIbEptU&VT|QH7Y6=*&sVV@M8>)dp0S^Gb!jp)~VKf3dw^BT-l3Kk~JbR!R%?LZu
zO-8s3PbFa?I^2~ELia0{8akY5HE2o5=>9nPq>0w%#df=W@p2>ogJ<xL$cD!0c?D6J
zuJq<*WGZF}Cftzey>uO)9a=DcTN;9!21a+WV8$Ox7C^6zYT;pdWm(&;fYx#ks%xBr
z#Pgubl13TYe$ibaR~M)^WO0NXGi3tBw7wD?+;a+AM~1Bg+W<#NGh#5TVO#`YSiw!1
zdf!}IU_Ok!spu)U4xz=^CtIz7Y*S?;->5#BvDGr2eE|T4Q1Fsc8X2E2f_;5r5vFkf
z;_Sk83ryw6rSCCbW(g|B_JpZK;A92k4`f0QI?I4E@tsmH0Ug|~j>NGrn0Cn1HlBva
zI5}ObI*D?mfr3~gZ0tN+%tVIEfWkU2gG8$x3-S2pBb40+&kK#Z7#rCUJ;~2#nY>D7
z>1*Eq-VhO_P<wpBgVf>`E|$@!DNw*F%ip{WLQM0arVDeFJav0Dr_gKGCmo2(V%m=>
zRVoH0rK_m;)aI)yR5b1g0h)8a^0F`TpO-EbxS{kq9V8L=cK1Tqt`j(M(|4f)8{4o~
z(EZR-;u4eYVq1&tH%K<FRi5^DsFZB*k|p+F`gp*ZX`mg#%1P_C1%^#MJeaP>gZ`82
zh-F_GT>Akt23)(EImg9*blgfBFpbUlg4lzjg@%9x<LHT*{lk%~k%)^tH&S!|h0v=m
zybN${iCfabKsK-6Z{si+<L$&{Fi_Kbb{G3LBH_m-@6taak1~{PL`2d;F-nT95K+@A
z>L6iVCdROva$zR_P^pDEr3dnj&a?QW`l7&Nz6{aEi9@WubZ53)e2n0D+R94<Vo#uA
zg~hYlGhY;g=8MQtjC-b>1*Sc5dl&9HNi_(IeTdK$*m{$l!N6EV0C*QN^5SO`dS)=p
z5$JX#qXs5HJ&tmhr0aP;-WS0lFl$i&OTbqfQza6c1Kv6;fsK|kXIhH{s03%<IDIE7
zQ)z@LFE#|a9DZxB25g3URwd$6`l7N-M<VF0_xAeP&FQ$6|NLgR2@|w1SOC0lf*q-w
z`Q^rzFTM5t+KI(&KSl&R+mj$)nTdXzZzio~(fV8rpn)h0jiv=coM0Ub>>Lzg>26Y1
zKcau^TP-e$$3il)rC-Z!sh#;stgc5V&km+{ueBIY`-;1foRy$pceG7)AIzE}Zt^`p
ziBlIb-l<9Ty_H~7QwtPEnrI=Laiozyaiksgju)FU4X5}@+id5M7O+kA?PxS%dIicW
zAoScsT+MkDXiBldFyeGW2sC<(EHa&sxmrLa3OP7+OeXqqZ_E?vEcWw>+;D!bvo_+8
zwa}pE@J#)rfgKEj%oM#i+T)X9zLM0iECMJwSSQE3IH?vV&)^IvUrVoXfY!gJ+d;*p
z1GodS8PM7+0#49VO$LR20lP&v$JA?@Uw)$0&;eML2gj8Lri^qs!vSz&R}4QPAspKi
z=ABV6d7u3VLg)$YHGFXB^oEmDWS~0Vh<vr?8eeR@3G_Z_tcoj9gIve>dZUXI6Jg=E
zPxfeLxr;B8XV;-Xq16bj;(I$kdoln|hG0OCgqit5cQ9(dVZ|jM&{#6KEnfVBF3%S)
zelhVaL=<+7S%QB+&H#)Z(!k`w36}E75Smzc?hUq~nB0XO8P46^9?p>=BnM`7O4dDW
z=Tw7?e<^+%tLDP6KZK{O$RWLL#JSU1UnZA<L_4@BfE=&BJ&;t=I<*^#(y?cLU;+#|
z7-<8-NPCS(Vmrr>+R)+S?xArq42z4MOsrkN9{@T^gxGMDiRybwxK&oNR$~BiDqYQ&
znxj=y_certMfRX&E;V>&`HSJx{HN~Js>VV#$xp;a;~ha|gtXT1Kk}|Xtmr{Tu2i_6
zG+n=ynsqx#t2RtqAH6iPkr}&U11v;3Xox%GoBwifXS}*{gKmkp120%GV24U$axlVJ
z5jK}Gh*$ph%HKx$h1&AfaIg>wZ%AkLm4#V|&<fnURsM;UKRtD%v7VLx!pb7bf4X+0
z-ssmRcYWnLs?=KZ`!k7@+G>4%<?~Yq7SB{G|Klq^j`DN015**lb?{>=KZXjQs;v;r
z>uuuKIE#n>?azlLezWuA5;QsEMaDHN`-6p2wNUz@Lcs`rWnnVde4<dc8Y16cC{+W8
zPZx?!!#ISC*2xzuO~bU#kDTkY39(`&e>s|jOoRGwcjj29IEhlZ-&pxCP^<`%2t8|U
zd*wesDe(_Y%ihL)X{F19Ra#r=qEc~T3c4Lzn0qVtQ1<%@uq<HGEf#wCBfn~x)cI5A
zI)|KO9eJuls!!LDvvKCXWFxapSX(86sshpK2Ud-8bG5zHAm2i5g_+Q>zWPiESzmn?
z)U2Ou4U%4~o$O4A*s85Q4JGTBFfP2Jd*_y6<CWTeVhtw|DwcI_p2L4Y<xkaC&i6v+
z3tB<E>|-dKA)V_#k#^{pGvCN}9?JXKD7_S|KZhOuO35=EQc`}_bMTa5M2Onl&VH6l
z2ldtJwvkl(Xilzd45t!2sXIF@mEuzfw7j22=cT}`heu`MI%-f^cq+&~QHS3Kyt8f|
z8Np0@lqF9d4m9S&51Ycp?4rfAt>nGWFp`b0Yv6|v`GCAQL<IA8*B`;h{EkF?)K_`z
zN{yf(`pXzY!G}z+IxdJpn*p+U6O%dM-%wketRq(`3<L%f8|o#z5n@^*B4qjDn@GlI
z(V1)&q&KKUhV<Q+wMfoB=D2>tz1%Qoo(kuGxsdX43U1uH`=#Zz^`(v5_wIJr?%mN8
z66;x2vbJFoHzAQj_plV40cH&h07ODCBj&V$1m=NkAxdop8C1Z^wVGKmkBJI?+zU?n
z^zxTrNwAjNd4#1=1&x%(A-h92Jp60=Gbtpuc52L_;-T^mDdd_n^w217h(HS8@BA^n
zDA3Yw-(4f%s&BZX>l5EBUX*Yp7cXCk(KA)}O=AfO`e+Ug=%|4YZ!FF356=#8ACcy?
zc>psU_q+Y5{|whL49Ij}1yt(G2Go`T4-s?8HRi55(u~o=;b7QGL$UxvxReb+Nt579
zm|#5nP#;k5845J66i2ByPZE*y{^r0&M$s2~fz(yw3wH6An##2Usb~f$<cv-RUTDz`
z>kCB%!yLW-xHQn79FIm&%s`)@7xlTuhHwKg2{|7k3wrZH2rN92j0%&cQwwu${9rT%
zRG?LxAwoVmt<b=zh-5_AP4rK20l4Ldo7WSR(xqEy1TlZBdp<xUr;G2d<14A->s38Z
zTq6509{yFm&X*90+O#@RBjZc)`yMuYwtt{>3%+1|Flrzj2Eqnb>7vZ(a$_GVzvqRD
zNC94KoWBzaSE{NkdZKLVsp~bzIKGNV7$My7t1yWjy7epUg*H;GR>rnvwN0Y3)<wh#
zqGUI1<%H(*<z+}SWIO!$Arw%UNH%>OSKq_V!)`gX(<|4mzMFn!50sX&W&^wjD~WBI
z-w7z(&#;QPB=M`H)-Z4pY;u80h<xTHt{>o3Fc(Hufjeg$nnQs4A)U7Pg;BXSPZ>>l
z$8()#BTjqG44-DEO1@SFNz#?f&cf_~`=UR^ly{0zgR<G~G1vz#4QPq-y!y0!L35_(
zjtzJbh?Y7A=LB=FnOgBJzNk2>i!~e1i|o{k%QQZ9Cw^X~DMSXR%is{~kL(0yJA|74
zXHOVkj$8$JjZ`V^0D*$3&(jt%hA_0$4R+ePO@l9mafDG$kIwgzwblEx&<t*#)Y74X
zW3Of7?SfTPmrBTab7B_64Jl3YjzVr!1P#JXjJ}f;07v~DM2O2mbkH6$oqZ)16Q-eW
z!I%%Y4^w;O^d1dyZN*>;{Pi0D31Jm@nn|2};a&x<!LHOH4fZV=(E!3tKd;1lrj<bX
zWbWopKv+p`s$rx%*jzvJ?qa&gc%7-P(T7><ypV~@#aJfOk<~V-MC^Tb4iYcL<JrN1
z2X#Q1d?iyAQl4N(1Q3Pd<MKID1!)HxHpEXEJr>K4(si&_0&fu6t;HfDg(Cb53CsTk
z!79u*L*{&%tf-5CiH13Q&ia1wrk^#ciFItJY%Re`n>;({2P7!U%VtPuSywnt6YXG?
z@R=elWm9MQ`#ChGR4qdy?Tq5Mrt)ExaC=X9gBiPhPY?glr!HsZnh4N=XOTF!;gBei
zbHtkr%8#eVlLX!bRv?+2Tu~1YRyY+|OkeXIk&ZlZ#uuxac6yi>E}n7PD$|r$*taAt
z<rT@k>j(zBu9AOTst{ETd4>2muaI`NNHVTY;Z3S$I~=iE8JdyiqGnHlC>)0;@Kch1
zC^SO`V+vJwn1{7$Ztws>ZOE{YvJ0m-<L~(Tkmi~sl{*J*+<RwBX@s_#)oDh93HwrE
zH(`+#8)`ps1p~IuZEE7Fg_-R<I2k?yqX9Dy_<bGHvvJ(mMn2@kY7l*zh(&^34hE(V
zw1S#JoC(ZUT=1ZsDkyy^vm|Ut^;EeQ&lG`QUmiuZV2-F8ladeBl#168!8hq4gQ{^F
z>*1UE7QewDje);;S$@N5_A-79wvdxaZ@&-Gao*|qh<xD<kKw0`y4s&?y5hyt1h#8%
z`EiA*nEn6Rdl%q1v+O)<xO!{^YiFh1)xPxlx3MEMW`HIMavlp0W*0+(GgRZjCBT^-
zflRZ3ZU7Ae-I(qMA45`><W%I8H&!fBvXsP@E!mb`agvIZIF%|TDXAQ*64_2tj!SVK
zrBYO}l(L;frK%)VNh<libMEthbT{~#*;#bW5Z(WO?!D)pd+xdCoO|xM<&r)|81y<*
z8*#A>wYGF6@U%hm^bKkq@^sWHun4e+7W+vUAT#v-*ay%$*C%jsd=p1v5LZ;-&0e2?
ze$$=E@zg;S50p0~9Y5IU!)0pl#@G{^qRhx0*2uk@EU0^mkx~)2?;vc8_fD(O^kPCQ
z)qhK%B%<wc*qR+<6?%7p>7vbV>4Lu4l|bUys<N65Fu`7hxm}82kuH;=>~o-OerVPp
z+tW5L`6j^(=H!eWAB%nSW)q=v>>TbE>)YYeRJIVsh)9EkMR-(%BJ0j}%T*2Mvtjr3
z*=chZghhvcqq;<gLbPBsJUn;lWtrPkK@ciK6(iWJZZ4}hSA4(1`W4mUVKJPuU$P)N
zrx*^38JB%6(*;dnD#b6;YB`r}`r4<w!o7|&5y2%~M2h+<V=)1FdnzfDnGj8Yq6{(Y
zVOs&4Lne@Gk_&VbW6G>-&2k<EQ$mIfIQLZo2feQzj+=D-(fF`v=0T(aVmAoa*yDaK
zoajSdGOshRsQnll;mA$rTO6#_*BBWXzGn<_K3dmQ{}W$-U=)Y$SVzQ|{lkJeI=R)X
z*H2qB$k$U`k#<m=jkW@PWEd<OgkwFN8+(4nBg9K?9TZhNo}VdgJ1CQS<BJ{Z;ZIl(
zPX*V?Rb0D&_O}o(C&99EvDfa)U%_v=6Xes?Rm9U>y=s5IRsHk$edvBasQ%;hWm*5v
zw;;l3%&Qd-cIC;xV9iqV`@8Mx;WU5q;s@dtDQzDp)Zy;0YnaAVL70Yk`?}3-gda24
zzaDf%xW=D+;~PF)gU58--J&M=Aq*5a4%*NwV85Y)HV|5mBOB)0MZd~Lu^XX%s=3Tm
zaC#TbaZHEmcLAlS!UaDArx|q{@1%m2HhKUEi_&2Q%WQZM$d*@eYIg%v?u*6|t0ho1
z{AlLvHDz@^4<2ouMX@177c~bMXW~N$GIERvfvHB@IFaHWP9dft*3;4?$o7b}1KX62
zFfbm0MfmX(XGjc*Cq;IDnH4=u9LZ4-nDz+d#F8imMe!tjJKDA_A^CaYaV2adX5@Tx
zLYdIHZHJEO$k7QzAQ=2&h74_tjJ3sD_<8v76PAt%!6*Dg*bSZ%@FH0|q!`74p$oai
zvUnXSQiPp&GUEZ!n{_U*IetbSi#RZ1Fb9lCAGUwAaU0x~fn%1&e<+WP*opj^$gYhp
zM4nLm2a!;&jMlwqA$vb#K*F^+2rK!93Iefam(DQH&5>s5m55^v@}ZSNzob)_JdO&^
zYIYBy)p>?}1dZ-tw7F3m9<0ed9^so-g-4FROBKus!Sf=R-?Hc)MPnOTI|yj&ZKKHl
zn!C1*tO{gGXH}mNg?ObM0gr*}_z@e=aXx2i;LwQg{Sl&(kEH88hLJ4Pghn34SRqO=
zN`^*99^zsI2M%uvs~p}rB5DBx6r*$;6)bm!%K4yUjquUJf~`g<L1YNUlD~4EN;ua`
z#&1p)G0@ybbqNaH=nyl3CEQx3-*M2<GlDZdff?c>a7KIhF^Iz?gZeY^Ef1%DhBYFZ
z+2NCl6OHV%542zM(R9pcl_{hR<n$;b6wz~i)(=qD=lF;k>Jvo!L#5y5>$rS+6g`*l
zBWp+mJ~FcmcmI)@wAv#ckRv`>UFk^LA**<@{V!C6*!1I2APc8<MtNh4dGsps&L2C~
zh8(Q=FgVq8vo;Ek)7gk<I2X^L9yh~j2uc&yasmBM=(Lr#66`WQhDhch!W9!Uink3L
zDmwq9F}|wr>>cj6<rG-)wrn6Q2RQM8$gmSsZYoDL#s*t@H}WXeO8rMWDn&~27SY24
zrpmU74k%(so*<|>AbBdnu;0Ctb*zm{TsCDtHXsvtSSAh6vvH>^43Ku@wB`9X6D6Il
zAr&>vpy{kMWliZ&>HLap=yY*CSjeAfyfks;2TV0m+_XYJ(fDcC#N&yhw$<g26W`KK
zh~lbkkz5|hj7V{uwXJ8Acxwy$q~fk^{l#PmCXoG_ABWAVa!HdHQH<#rN5Gk5C!@t=
zTdYOq4U5gTz!v)fG1?Y%NUXL6D~#E;0IAq*``Cr+QV?g&B!W@aJ<<5>EG|b5rnXh&
z4tlg4xeljSq96#s!Q``IRNvs`EvR&Oh;(iznzuKbd7<a=%9e>fVqIoZOA=$3*;WGV
z^`vUV_Q~DWede|8IE^99uxwFFRk>tNL!h8#>AX@I@dXJL;~%*gb8ugF$2|OMZr4yW
z9#xiyf$#-db=u%V9Pf-G1}#B6FKD#2<kRut)dUMjip<@fzNy0%eoOfHFzE4NP|IJB
z=`iT=K~UsN=j3M+dvy<)+jB)S94DT@Ju!=tD{olhiNI!x4AbCb=kY<%Ag$r?K~NE8
zkz#fP17#G$<VhJMS?$j%IZ&4E*L(vc>h9w=Iz}=!EQSM&j3LHFp;3`CCY-)}d=RvE
z^fRd07$Cd&Yor2eJU$58dkBp^4ssqrRX^6jj}x*yv4kwqnji^;tr&lQcQ6F6tX(R?
z2hMZ)1B|I^@r|>LskK{_{R^kokMqT)@*xNL)ND%M<9u<rtTOhSa253Nw8#14ingmF
zxk92H=Zlk&;tXA~`n*>O*6l<L&#S5a1t0ElUJCM+{U8*^ohS@)OoWf)ne1oraeg@5
zulK>o52r3cN!t*Fq<%o6DhhOxhgUK#U=yp$OoP0^5h;F=dkERD8lCn;KWR4+8-RVB
zHVzT>ZpaIr`Fj8Q^2*u*PidKCM{?mdE(k|O^dw4#DW@0(V90*V<Wo{EdH#7c_x$sT
z-nb{dduX4d)*%}W@I!<<HnaJ01+s$|D0sjr+LX|Ek)YZr29_IxLsI8Tf8-zT6$YVR
zY-vzTuJMeZ2}{9LM%rxNRlDnBaD=1}+Pj9s0v<qnl!AaX&u$G9?Lx<9ZZ(tfViIU{
zl#*FK9hx22%g91aC-Axy4OUDHlmE8ahz3*!idf2*(Kp3{HZfpBg4$U>KYoRs3(CjK
z!#B0ntz<HT5#7rvGkm1<L0|aKZMvXO%XNG{ozUg#Pa%EK1^51^tA7W-zhm3N|JKrX
z;fgb#Uufx*Czig(CoTig()x*|pW~BPZK0)KII;Af;7NSX*_U|FS(JrHo_yT#iH7h!
z*p<gYaw7TtIR}<rkW7K+Mv4G5z(X?{Nt<~Ik>o`wjYD=()P@ds|Dh&D`ef?Xq<vh^
zi`+*qefE9LePmJ<(S?pQvYMrcdth=AI1=WX_wEsLhqRh|xN}+lAb|3M{*;Tx9hBB{
zQ19>TwKi-Wm+;f1LArCmoVBLzm4iN#LDl!B<gRm*L}^OyN6NczikD3RKB$0~GwI#m
zHS=<8O!ovJgG|pD^0ug(S2c-stkobY#lten9qF!T_HJ~O>PO`xmImBVE;laWnsRqb
zwYjUjccXPsYc#j&xUF|K*+9y~;2Ld{Vp7sN*LdD%HnH?rXHKbg++=?3+8jgcOfp+d
z_m9U)F9p!(L#PXv<x1<M&KuB4A}m_<g!S^K1$`!IKe2nxe92AbLWE+2%*edo8+SEq
zBe7z;->Yao@k{k?eV;c?>wQGi?j!3$hAN4Y#jPImu4xCFn9Jo=Z&J#knDnP(IKh=S
zy-}zsi?X>1o_&j`oowJvX{B@YLYoSQ0S$5-rjksHZ|^jrks+7~xtB3+RQbBCjYHnh
zBvGZdexzMJY=ZrRCfB-$%?i>fQ(7zH5z*v%?%>HAtLib2Gw$wb^=wMlvLl`4J?x-h
zKqqn6Q|480H<2H92jyt52~Ck&TKEpL<gSkgcF?9d>SyU^LBb}gMK`9Bn|WZk9GK}O
zYqd%XNknlKSR5n^wIaj*qFLWyWtmrTVNy90GsK6e^@iNIo`W|Lvcj#V60BCs);pdV
zdt?{K74D4&=1SvEyT0Gr1o7@TK4rW76d}Y5;BSu7j_?>Kd!h_Batk4MXB#pKoCtL@
zthuCEzj3W%IUHAm<)B>H%LvnO3@n%BVimCYO<Y%k6(YqWLRS1^YXOE}S4fNU%~-o8
zpd6TqtCX3NZJN+=m8GdtWw(Rd=5p&TI#?3daWf)<U?Z$-p|W^@Q1+$t>}WLCJ{RGp
zOQ1_wzg4;**_<1#EwOLGks`x1b`w@8t_y22dl+KnZph_0Tt5T`Pj24FoJ1<%dURPl
z5wM@AABmDe=z%^OF%On9vfEQ)tA(YXseLi5rNoTE`c`Rr8d$Jhy@<z(7~6`KDCS6{
z&^m0zFT(jv1#zsQ1WQ-%3$vRJS9=u9;)%TB%90M9a8NR*kG4uz+Be#rTWv!a&{uH}
z!xDvdM%i}_8<l0V+eGeTY`h$_kUZLnG~LVW;ATf5s(WsZxro3+p5o?K>$X{Or4bbi
z+2`%%Ew8tnhaeX<WX#M^ELi6R4Ja6jh+igjtwecM6B9^G46t}p%=)anH<Ff-?GKi3
z*Ei`w*fUio<T1x=4V<lA<C+c;&OdazJpa6ug2P3zGrpK~S5$cjNi=!`Qpz+y{fo`!
z*B_7?)sxmax(T~Nnpz2+n!lNv6DTw7L{vv<+R-Fcfx*nu+0$08?xKQd#;HiFi3h`K
zGPXXm&|-e;id%LsTsalxAra=-Q_FqS^o)kWr&<l_Gn1GTSe{XBaFr{RpTH19Swo>b
zSwVtHlz$D^cAoOU<Q&%&T8g6t_;#ywS6VRh>AwD|keBX595`ANmLCDG;HF==6*Dg{
z(1^;m!E$D-<^poa@tq;<YjdVBRK-k_LEW{j`ewh=y)zdTvQJ|_42$v8YPb3Zo?r8U
zy3K>G0pizuAY4YUV}STIA7~S~)p3uUfiu7IL1C`c$yf%~{>%ryaZ7F=o4Zio>tSHV
zML>Xy4(^<(fFF{be*d61d-7zXv)QYtjalh*w@;oq3B$Hq@7_7-c(~cC?DY5dGK7xR
z6#pw5(sRsk!iM=H^&wz5KlYqK@h`_VjtP2|`oRG&V|7oogo=5;=tDZi&KDfxxErol
z_wPEH=$IASKP6SIMp@+==xGZ^=}0$ZbdPRxzjL$c+pp0?Hd0V*Rb1S_`x>}^66h4E
zYo^6j!HQ&MG7Gf3OvG8;er>J?%aUziB=v!Pb|_^u2h*~sLv9RWok2UC3_^|zyQ8fr
z%IyD;Rg<f{G-(%1j)Utt_2IxO7tzf1O}_MELI*aJ3<if5#Z{d1hwGWV7QF%Qu^jSu
zTo*RIIG!x!In-jNr-V^+>8^uXXNb3wadP%rkj=~%QuoQa{V8?Oi1WAzdh(B9JqxQv
z4dZdc#1Z+Sxo(~sAco5fyio>;i!}B@V7Nd@&B5!Nkk@pq?t7klf?Re0w`sq&QE#`K
zU3<ZkNCz{?!0nn{JUGzIyH;Ki{@sw<QNi4$@si&82w-&y8gm@G%bl`<>HG?xZ<WTe
zGT3u+_wjwal!Y>-Q`<g&qWUMW1M^EiZMSWnt^P26|M8F7-%Hg$gWrGXe*f9(FXH#F
zI{06z{%!pJTfbuA|D)=En%=*8misq9FxX%IU{|0%Q(&T+*WWj26)`ZME1d8K;5);y
z(r1Ue|F7=0d^)w;B1=-ghYgostBkneLa}_V!)+Q|Pb+S6xcB5HB6!F7a%^qkYiqTY
z<;8bu=jYevR~Oc5=NDGbUA}l}?c&NZn4m0Rx>dwlO0%$)DZ+5VNKIs^yIa2{SKqpY
z1Ab+31a5Wq_ByxdhEeyxta4Itc`fY;)EYG5wI}2s(sj58h<KCyu9jd)>BH$#a^DxE
zc9Mpe$v{*pY8n$-4+Y<s?oL+rxX*A<E?sSdc@34;R%)x~RxU4`ud%_!i>sj30Gh4w
za>N_BwUK7Cr5GMT^AE-Y%VD>PdlegrTzH5`K-|Rya-D?0pyL_aM7%#Lxg}d0y`;P6
zKq_NO#Zp8;1iAEZcoaSrZtu}|X!dXlXz53SRM3P^B<d}=izj-?oz7ub7L+fWY6!(o
z$;1n!$A$0imIMV<bGmX0K7L}tI+yN9RlK_FhdVp%Qfk2BqSL1;uLNZ;PVAdPP7@%!
zKg%6dR@A4rci=w3eUs?a!*=Jz0HvTi^ZR!O_a-I;J0nE~_vPAH0Y6@vzi9ce2DX$5
z@Z@-LRQ@x2*WtGEk$6w9A0wj-8mP^kPOsTcu@i_{Ni%4-x!c@CJ`{plg2xh~w2#fg
z7VA1ydC3h$F(Y{dht7^%EXnjk(Uh4{jOU9UPx)5JBYBd2ZT{los*~!RKNsLQGND9M
zDw?lfzph>asa{M5vk(~}=62N2>oTCl_|gSyIUHlzBnyb|T7g6&T7m3-^Kif3o=x6_
z0PIwf*E`r=ezWz3j4aP7(Ya-3I~=Bs`kgOp=?g7vL_>kDw{K7q#8ug4fJj)OJ(rlC
zR*XhKqDWuW?x37j6$i9aYP{1qgh`7l;rDvjg{>PKk%#)5y@L*J@^uR$eYv|mNO_@I
zOnehvz}K0<7<da2vHxZsIZ-`2a6Ob|f4M>m!b?!PRgzogx3G43M=4)c8i)G_el!sq
z0uU33CY&wgc8oA@Vhp4NG$G_n=<m!0Z7Dls7)V=ST?#-WQK}~kVP2X76)fU>QdX$(
z<az{nJz3r99PTxe9m`m<@5P;4#63`2Uw4&Ww`^6c*vt%FK4bx(F{^k={nNnDA2K1I
zkEs7?8>31HX<DTF@}(<lwbLA7r<IgeuB_?P9KA4AawfOrBj}6KZMlOM<`M>t0Y7yr
zdgw}uxZ&HOq)p1#LwIC=y*`y(*R!wJW#{U;oOH!^=s-4X1U8|tNJ^IsYZ`SCk2-_q
zIW!R^E}jq8XnW-`fL7mGU0YbHUAnw-X=yD4(X?DpF-vPJD~q9N!qLt?#6V^reywyZ
z;3#`s7|5+l$0;lMiIqk^De?Mab^?}DH$aQ;_+fuJ;|OwPb>VXDqT$1+-awsTp#jaf
zyH1C+HNUzF<70k#t#)}~_0r1n>H?_okZm5VYcmhq%p-Iy-KOj45Y_)FbK^H3kL3-7
zJhM6!9Sj*$+L$hpTo@=Jb0T6f;ZlN9fnZyEX*yJ1MuJb~vo)minhZq90H=hb3G3sh
z<P3rA#Mk!$%0p$QRB8ShL4HQyk42WA31w+ieW6?pq4H^9A%@9Y3&+2s$=<L;N)Csw
zMnr3n5DTjdam$m{wKrB)&c~88EdJ<-76|pQkr2mV?MeyTu*9RI^_b{4GmL(jNyu*a
z)xvfvr$eQ{XOPbnM_W~FQ~Y~8Ux({TA-!^j$G*-V8vE9Xk;lYur)#gzA5T|H-R$2`
zl;nuJnIYsGb+$%u_ua37%7$3B#zFdo9QN+0WuZI%{qw_T9PVC;vBM~y(Hxhf7ni|2
zjOU%-D;VUibq4vm8#32+%7I>h9UbNIs|<1jouFy|#{268J&KDXOEKrTh`gdH@{qoP
zBgDB(V<HCQ$^!A&N)?&LG7i-d8cus#lX>ij<!r!YXEbrhc+kFSU9iV|a}>Tgb^zP$
zm>~|}3yUlBqXYnK?RDyX2{g?b-x2{J0FPjE2Mu^RHF$?eg23o}(@x|@0Hk0Z!Ka(u
zPg`7Be#7`AU9<#+Ru(dhsdypxy(~&&*l}Rtc*PpjObnTACjH(uGnV6XsV&W4!X|{S
z1VKE;Y%azwN<@b%+9SwzB0$Z+2LnEv3}(YA?1PV8T&B<S+(PZb#f8Q5tHsE!YrMg=
z9K38c1;0^Xp+e6QSSq}rf-%+VMaQloj$G;~kgx#!Hjz(V*6mWB7l}0+j}UNh)~3b6
z6p41wS;TBSqsV~55kc2W%K5hD)_#2(8Ok&e17zUO9=AQ@k%bByme3J4u4+J*WEMCb
zZc<PNLl`;zF0U*u)RyO$7Puj8cuvX8DQ&))F%*YJ?1(RU>kcpc56fMLZ3&)XU0e3M
zdDq#1KmVrPaw&yB$!O!r;^O>U^K>R^s1)&D1G<ZF@xQq+e|fF;`T~UJ!q+ZcygYAa
z{+Ux_Z?9ag#t;{m*A_0nHNS`so_=v`Y;om{^A|7UN2!9ytMO#1(`fD`=Ug)v-&jU?
z5q1&IFMJJ9yqk0E;?n#Z3yATkE-d>OGk9@+;llit#kE=hz?xr`k$tU{dVgtt?cAGM
zje+8p<>pQVSi=NE=cF+U1qyWgG1S5`BA|$ng+YWjAV{}Pw~jxWNk-zIZgslIG=mv~
z-S7iBG0&4OeTHyoz>_)2ef^-h)xVBI!Ez!232T}?rmX259vpOVl=w~)hlzOh>Q)OU
zym90ktswBad#l^(!>hm}%&eet*l+FiDsJE^YnSJjS1+twURt;efw6LaVR03x*A`bU
z&)3k`HxTZIr&ksiM4Vi@xcmkx#wi;HsA@bGLX`{>I5%7FHJi!KR!v^j_FK1GZJel>
zthheQX{7@)Blh?AK#dy6$0Iu^akz)`WBr|O=Wu%m(qcQoX%O%sBN3!O8q(~qV7BMq
zE6Ek9j&Wn4B*F!HO%%o9Ig2K0)|lU^q*>V}HaL2i*ZRb}2y*Z2;;6Iyohad~_jvMV
z=T;N6M8F=;TJ;W^o2{)o>;MjtOQ1yaPH-lhVgbQ}i%SsMZx5m*xI<_4#Jjhe?U@OR
z1#O>VM=`M8<Ou|<HoNV;eFeMq?v0!7)y|#$jk=V+Em&7hv3#p7Kll|Eru*K|uYjRR
zh|b;I+Y8bmjV&!Ktz3SmwmN@dq4w6|;u7Z6g@wxtx~wc;Tz#{4VR8NqGWGz-^DD~>
zrLoHkYgaBW*Uo{fY73VyuUxLZe&qrxL*aE<08XDdJ9Z5yxU&e2W`bw4c=TFw0s7fq
za_J7HLOYSGD`t}#_7a=j+Qy-rZf)R84<}hSWvidZuj`xd9=5P`yVu#g!Enmt)mp1l
zdEJ1&jk$nR|BQN{n|QU?Z-6eZO}xetehq=@P|fffQH2ER?HuC$Erb>xbik||hz@NW
zf}c7txNtV2O+Ljr*&fQSfjjFPJmnx*O1c2X1O`%biUED5Jem0hCut9d-_b)+#1Gpz
z{n<i#h6b@w)K20LYD_qV$!rLOWLckjV$fJEEWcHIYyL9l@@0(ZTMLfprIn>iwdE^I
zwY4`dFU+4`H5O`Vapm0nVh!S;YM+GP-?+ST<x>2Xt5XQ#mMdFcYlP5WJYQSA_#+GU
zdG?9co_#J=W*;xjV>kSAZT|fE%eGGQ=+eq%yFHF=?#t)iTv!C`#dF*T7SVlf<;rqC
z#M1oNER>^3T*VbEFm``Ft{wXWHLOC|Z(c<a*HI~6TRwMgu{OW@&hj}k3*MYxKEDV_
zYH^QR(TQ^5;v#Fg#tGUu=-iw^oD|q48u9a&R^FPa%`ak6tQ`+`JRsko!~NiATi<%(
z8Gi0RDDVD<mk_4^?f=(_XR05qehOjx|MvfEe}AT0!S8?1{XSj2fZzXt`+c!`fZxCF
z%J-^2hTjwGw)~&2{u+K?Vj?H0|8H0Scl>_G{r=;nF=l%Dk8OoZA3L%16+Zb5TjkQZ
z6HAx)<ToA6>WQT*eDZhQlebSSy~8KJ=1N^Xv2<Nv+>?zHOM85B<+ltnOYIX&eLng7
zuAW;bmTvLMf9;;!KCyJ0PyUg6^4^K1_xR)|F57zUpIExjCqL<){OF0LALWy;yC;9@
z#L}PQld&~h>c>tj{TQG8c?a|3CzgJkPkzol`H2%tKfx!zxnWCv{lwD0#wVZn)Aq^F
zoLKr9K6&{!?2`vJf0ldlZ=P8CcM0=9{EUV9)e}qq8J<W+t?wUbB|q5xHXhV?L4Y9T
zM+m$(MJ7g!jW*z)%R@t~WCI|0rp;()`x>AmvTSlp>lAV$f`8F=pMQ(?FT#$b@eDii
z1A}dBQ_UB$^ho(1LcqAae8A7-kb}m%sppe9cD;G^nTSko0bVt)N%VncnSuK}me>Nr
z%s_m*T4v_gcZPPDdG_6bnWc7^dGW~{JIuWLT>L69!3@BsJwRxBnYW(`-a3=Zyekb?
zzYXLM4bWpRTyz?*f&OffQfuB%4X7=%NSiW~VVQxex)wOx{p{Hj&-5RBeIkAS$puQn
zkFkHS2MegOfpbVW5cR)Y8}UHYILz&XJIF4w(@$nj&74hEaefpQO>0xVI!pD<9fXG`
zi_Lmh^0QrT_V9IcCwT*g+d(CnhiWUuaTEuyx;L3w!o+EJo0=Q0zOiW&!_B{b4u<On
zP7@(N0nYL7bQ&WksLM+<cOFr}Q^_jL;nn85hnnJbk%~W)&q-=7n}g2gV4F(T4iEO4
z7D=AfgT8Ij(gCk_k0-O3<<)fKQn$IrQ%KFmTIWWy&C4G1lGr&b+m<Vsts^J0`wBOE
z<ka_={F|(qQN{^2#LUAu(wJ?8mC-0c(hK-H_L4*7Zr~BCIu4*ISJ0AeVaQB+IJ{s#
zFb*(1m{AUSNy3zJ;?o=H95&;_aQxcPdwoLcotR4W3%r2*u5IX3I14BIX61CU#tK{(
z71OIqaEYhe@tfAnCp3v*_-prRW|_`t4{Uc7N@7t1OZ~dOz78XwUTJ<vH=8_Zvk6B^
z|5g(=Kg;)Tby&Y-36p*~Uke5rGmjSw^Q19JnY0=`Un!g~Qh{X_`d?T;(@FO52b4kP
zG~6Xb{+H2#$3n4ltu$cSZuVh_I%I9u+)VxY!c|yFR_U`q-lIFnVbsE@b-`z&LuU!G
zrQ@jC(M)!vf_Vx{daEmwh|R*q2q*4l7xu6tQQxr?k2CVXb?0UaR=lD2O-v5XK!-8t
zgKCoN6KGp|KVbr|C(yoo0x;$hf%rhQKI?*PI0PpXV}q;Wh+}_n$++sTlmD(;{<)r+
z9@ja5*QIwD^x$OAY#xWjM@@t~((|p%2Im}`V_X@+GZ%DE4=AuOkG}9PKkm!?7QGEJ
zUNY|t;aH0sc#vYRz6s)+aW=%TjApy*hSML}ytLPZ_=mF)<lJxWZ!qT}lib09hYe=9
z$&|W=Y@oWG5$bc37AV~m6<mM~s(aj(%Vc~Es|Nj+jV^ddh@El;eHf{+lm<Myjw(fS
zTi`KQ6g@-ja6-d{gW>Nc*B7y5t>f1<MEH|qT-DY;>SmKB;4pI+tfsx<R5tTC7r!;#
z=D-Z0-^n{b;fGDZTP$aF@nI@YxP#hQp9Dwzyk@=<T`vpInZqg>VW_7O7&=T&A_G30
zz%_^+1uv#W4|10>y(W`9FexUsfW&bvR8tpeYg6*53tx+yn>ly7oo&o(6*bZUq%TTp
zHF(+jNZ`>z9!dC+_qM`p7n&@SPboM#HQ>`MA`jOz4hdY~n5=yqdUQkYzJUFuHqtIZ
zzB<fbu0#I76A23+bt*?^i~-%e$hnkF=(H8#CVb)-sQp7^X$5YYxN|Dm2C8?7hxFR^
zZ!*jgCu#3Iz93?-uIj{_xOOz$QEE8qTJ=et2YEeEfQRXW(j4=P8o|*}$0V7QwS8}|
z9Ats?l0g6*E2059PQ#%a-g8@7pjfzZ!_p2G+=fhyR3!7TS23v^$U)iUPdP~f*Qi;L
z57YBJu}wBxkebMX>)4ldXFgjcw+dMWix?W#HiP82DC93z70g5~ymk^M)2~qr@grnB
z^;@OP(yA-rUEcyS`I^;GADUnHCrcHxph@xoOy#SRLm-6)njMZsTysIiV`CY;87n_>
zIb-E5!~8I!#_GnJ-FQ9IK?*uawo^6<*KF>zIgtcHLHXw^DQkuuqmTnnqEVUWQ%FpB
z133+F%8?|Ng&^DDY+}<dxZy$aw3;XbJDojj4x~lp)?R%ZXiB0pDZ{oo7yR_KlfiY@
z!4Vdc(Sb@HI}wjM(Nh!EfOne}wg1Z9(%rk|%JW~Iyw7}7z8W*|fkbCf`!OhU35hVV
z?>Df`a>Ko`{D*!b=|pB+7bx-l;JR<7?V(GJT^Q12PT`)J#wBUkjIeGWxjiX4f*OpO
z6A{v>!Hhx9@{<ose!3lwt!zrRJGOPIS<rHMmkBLzOGnvk#&f9OTZh<n;!Y-m25>|c
zI`tkmvv3&u;9w8QM#HM0@KI^A=WbQ`V5OHO3Awi!OxAGMAaQp;SYXAZEOjP>N6soU
z1T#adf&@!Ay%<tY&Umkiwd#hs?7kggz(s$fZgJ#6*VZkuKuWVU{a8m_yA&$fZf2Qm
z<s2SkcQu&W7pfM6ZETKrulg=Q$Fg=D83+GU3X)+on1t0}22CkrP;GWmBkr18;|AbJ
zrSit}kxLyv?S|e;I%Z{qZ^k_*cr}f!tQnf$#bz#6TaX427jkyiiY_l>Y$kiwd1St{
zOaqnQuv<Q38gfbA7-dWDsd|N-Bz0rslnz^`pT<P;h;o)yV`LGQz--Q}TUU=tZpw8w
zz>jUZ0X!XIR>ky-@ZGy5q_#r!__rjoTUzmktJbFf0&(WVa#c@qfk@?44dG4VE39>N
z7M4wVYS6ilkO`T+NnK1}LtQZzkQ0hb32)Sa4!h>09;O2Mc1sr-p#o_~K?n_-+F2Xo
z5WD$m5pV7^5g`x?wm@(5_2@U(S<n&Lt0NA!uvzA?nH|>*Q=PL;D&sRSi>g<XTr;UF
zg|ccS8!c|o4Rv+|9qMLS8qJNv?Q*s%U(znPMqy&kBqNe@#4+`l3r@9i-vVK*P37r8
z@lwYKAy|}eGdK#|?ySa~Mqvm#x3{IWPBt!YHQTu^LMSb2LIx+=ia~5xTnJ&E#_O&&
zDu&FzEVTwfA4e#q;`ayz(5ua6O^fL@7eK<RePYiZp*j;U)d7)Q|12#X-$wm3sonPK
zFxyc&2M6&1AnUut$2CW<*_Z1eHR|Z)EMrRQ$)F#YXA$NyOe1T8o-%IkaQ0b6(Y;n>
zS!E6Eb2quYQD%+Gw5`tSRH0n%^e<v^!BYou)mZ3uJ6)uCb@W#=QB9t>-TQ&l15D3l
zo5E9r;5yyVCS_}`C$1Oj9d<+I00><G^t3x1qbRKxbXy^@x*#sI={oO*UHyAVtMA2z
z_u)|M)EHQHU51SQ&)Fnr%g4jnlikL2<Q9~=`N(@Fa+vkm`Re9az?;4{nY;l5Q>y9(
zKt|AbP^()>=fsq>dtU3ZrKPtEVfBk#jDfSLJ25fgMtcc<IR-uJ#fCo4a{Ca$lwe8q
zk-~XJiCk}WZ3aUwqg(jdw@M-$TwK5TTucl@A92L;VXq15BcGqP7eE^|)V%O&*m>-G
z*(6-Eo`2y{cbT6^UyY{ZH-2&sBF}iz352$1qN$m*6FE~HIRrumL&Dwpt#%sO>U*0u
z@{$nzlnlbz9E^M$W5p-@xHGt<P^wjh&>qm;Nd$LVS;gaij4X={Y0NEpBeW>_#J-Tz
zYD6Kg|K4}c!<~+h!O(ijrD_T1Z#!DDOlaL->M1R7<LX>-4U+<CjdM*ruuSM9mY&8-
zgO%y2Ps|1hw#4=hxost|#>|MB5w7euwxIUlQ1;r9YnVdv4z$$$jGC{Hy|$eo0Z%FA
zPaxW;LGYs`s%3)p+qFYT`^kDCkW;N|Ep=;2Av#+S^Rw*`7o8$RTAw=gh^n%+A}a+-
zu5H=<rKh`x{XXnqsgqJ`Fv@$n1sE}+Ht@A+UJXvo#D*OyhkZz7wwA90n@R%pi?EJZ
zW+w-6{18r%o#IkJ2=OLP6_lN9pR$yJIMWKvA(p^NH@HgKyFgDi!MfU5Icj<?`o-Ge
z`R20X?@Z2NDWiEuWgQ;!MU)uR9rxygxPs*_bC0}sA@X|V{K{-_H9WS7kw8#vM|chN
zc*1`DHZ4|WXeM*;V+%6JF}Cg?8Mr7>l})yA^NI`)6ev6|!CLIL>l;XNg$(*IB)I_w
z*&#980>}LYJ&590E7uS<x@r&^00`U!?rnk0P#$!XpVbTFrk-~zI(5rfO1VbN>3h#e
z*(sB%VdR=8CMKO07#bcjE5ojt1`Nwu#LbLzmSZ`MM=&Ei`i@L!F}#u&*>Z!V+&ygO
z%%r5Ln0;np0S<PF!F|83PI7@;3Tt_=@P(X?;iJORhLxR;kkQl30EKwrjGGQy$y_Ey
zjQH_bWopm9$~56YOrHwd$@XU0s70OQ4!o6Oe$3(y46D#1>$+N4YlN<uR@TAdwRfQ)
z{z>uS3!;S@HpF%yZX^`1mGZCj%ix-Gg^tJ4z(lB|Ts7{U5#9z<c`W|~BHPagS4w2J
zU%#wHTeu_T)ju~FLoAi7&=u=~7J}DYxX&Pqr1Kzp_aR(rZEgwp;0TBPXzpO!DTxw;
zVY8NYkHxd@f^|IQ;o?QE>p!Z^G1VN7<d8MXxabA7Bt~Lcp<md!VP0iN#;7SJV{R--
zxaDCu3u_70lwhsE#PFo2*=4Ec2a7!dM?0tAXuEb4p}vTbMT2~~X{~x8^x(Rh;FUat
zVP_O*i)>n%Bg*`+*;YGe-!Hp#X#wbg!j}!9)C<97@fCfM?GvJZaHELzDG8hO^dhKO
zY)O!X>xU<zl*r%t^w5TWvRCQ@+u|`Q=)t<<+=?rS-o=xEWm5=-P5Q{Wqc-VE7Bm@N
zaA-Rq$i%~76>I8Js7bCk_6l|dxaLGVX2_S4rKt`Fu{ybNBcq~~Z8n;*`K)&o|D11t
zDC0&=-GYci!~z>KrRPHi2tx7?mx~<hLe~j97mO>$F;buoQN<L4(z*i01tX8r4%~<e
z;}RPw3bYA$Wf&+=Rt9cG_-!;e)@oR()pJ~(Ka4SU_O&>F;MJ|;Dmosyq~eX883w;J
z5m@fzTM8R)MQTi$M$&rMW-@;WDQ*-98kQ2@r=649Q#FZ}?}{fK#==qqFx9u@1Tu!K
zBezuh1tw&#$ev8mk;jhVwk;f+2p#vsjZ^I9jHh#UP`Gu^n@RA`g2cuO4_LZxG!e!w
zA^M=MYl<Dg4R~hZ)3nN95}a|xq1G~OoYXt^7FxCt31gY6$c_s4>LkS2`dzuW${k0H
zH#NRkKI4c(8pPKskAs&2WC^1}BgQX?)7EY@x6>qo|7WkhU!R_tO<wKcR4*9k)sy_v
zPk4kIr{K{YvRhPts#8gFvw|&UwqZZKvPWw^gPQLNy0y+0l(^gdd#{0yu(59VIqC}M
zi_kWzEn*LACxr`@A=snyk?x@!wQ(tR)W%bSOlFBREdNlH<<Av|oQgH~Ga}NgJIyiB
zvpuRtER8&j!(2AJyiKs*W4u*_>H75PYwLmS=;pSS!CtvAyF@V$JK-csPL7TfOLVTT
z8WJ1wj|hT{qT?e;T_YYDDU#B39GJmHnuZ~Dq=h~o-(a;Ljux7K7|aS<_8f-$Kx5)(
ze5#Lo|IM`xwxALoQFs!02>39|WWkzFiIhRJv}(DF`o><T%>Y$i;34u>7Cpzb)*})J
zLxf6tgd}%(UD+BKetkNN_$;<q>8Fyk8=WrMB_?dPOPDv=hwHOBmB-4kLoPGxP-85!
zWGbG*-dDCEul>9FhRd&ThFJ;iij&#A_=%AFpa?G~iefdx%Yb}=6hjXdOEK(23a3;h
zvFe`^1Vw!^VsWTVW=%sgo~CyrnPq|8N?{x}?rf|S;+)4r8+V0)S3C{!4VC-~A3<#A
zcPw9cr(ok@LkCg>I0CLJ$W!$jBKw(~B}qn37K!sz)oQu<lO#}nZBT_mT8%bWYDRN^
z{S>25%6XlAA*DE+x&~lmC^E*iC{TC;Wy1H-iJPNJD^=-S@oz|JC4CrKS_$-s(n@Q7
zI91B=P71v|g(@qpqLz;!t-{_tw6w}L?gUP1vc}S?pmA6D$kHmT@Nr11sAEH<mBNlL
zt%l;7l2$P;ZU#OMY2{HL77L}2f!oRB5s4cZFU4w9ESc2K!A6;vZH_b#C9u+lg~%wR
z(b}~~6yco@k4pNpGx7cfLQa;VtzA@slhf3UHJF6<KZ@`aSC7f#<U|C0?NFE-dlI_j
z_czG-{Dgrln8v}`GgL1pI@3zJ9FR!X@!BulqBEq+GWIpUFv|LY8pT>!(XI+N>$yTy
zF<I<|%$}h+7X-%H%^B>%FpQ?0_s$?r`(F9uIJ8dq<v6qs`#;B_b;1k$$L;tdnC}}V
zw2m^9vF))juVHdHH|=9bv#Xf{nwqHWc3L<y!r(f;FpwtW?{l<Xe)lwd0L!2`Voyx7
z{i7-;5pKt^eegA5@f7dw<B#nN2O(5dFg{jOJPznH0fpZGWcTErOi221Kwl7~!Hik$
zO`K9~<A3qD8Ue!1CzXQgSMz6OorE!8UQjrv>N(jG%Uy;3rVkC=>yZRCximTxF4JQ4
z363jlH_O>j!P6-%8+5i41>u3~6THrJLR{Gesk+uDxDT3lgon>_BMH6k&vk?!0X19J
z@GBWP7Cp^G_J!3K2KBkhjiF|%&O~5FZ57Azu|AqQbW0SR@syf^=43^5p>}cPonvp-
z%TF+nHkdE0FV+>raysg0*%N$uBWB-?IBE_&Gjf?XnbC6)2LJU5<7%1+Tzg6roa{8o
z_#GAFh0ezfIzWd}Vf<tPwOtKn1Rn#%g9DIxL=Rq%NcXUW>!{`Iyq6j3hd`qYKnNeO
zHUnugGbx01*d28k!b%?BNVrYAym4?ZM;rwY?|Ny+Ap0sa0v^QD-&sRB6Tm|cqrQi9
zSRIx2zKUGRbK)>TJ#y1AM<n`O4y2a&>%-rtx!LHKr}dfC?=v~wO@%W{J7FV%`_hVt
zpb20U=ps(ULTm>>u^Ms223?(I%i%bv*=D(Ll(CVn$3e~hxS*L%W&<YRQA|}tU%8B(
z$3e{@rH+G|k@%A7;oM$Q&PEi)V2WtL7MLrU*@cpF885mOs;L^SbKyv_PL8i_<U>k$
zixTUBhHX0KbTk(&U!&10(~4X=&WRO2<)G%Vv4y2eYwy(7@Mm#;jTe&SZ71ky9xG5K
znZw)!&WQByOsIDwvR%4!tEm)F8JRq%cg@&Qwp$wKtOOb%g{ya~u$H14Oz@0P=rVEl
zt{IN|_b2>_s6jZXm3sBVYA}zlOG6MEtyj8>9dyb~uWXliJD(ZlW!y+`Y2n;OTqz?r
zVc}M(haLA?yOTP=@nkt+2-zHieNe_RjnLTS1Ku9;59}2i)n}{E;*yC!`g^!!;?LoB
zk?JR^)t`Lu_0MJaxKcv~5M*6^uzQA=j{LBIeLfFNb3_`rH>z)<x*yDg3$nTz;Fqfl
z0RD6qm?vu==xrn~R$t7=&&(zRcD8ypzbZ;q19PT&23`Nu-?v>~sa{2yPhPN@fY{}|
zPQS-a%S|Z1Uj22p?Ez9*;Y>HKNIl=>!F2YTx2x4vRPqV`wr>u(nMN*EFY#s&ht>6J
z8}Of%GgZD&K59RZf*OS1XHu|vXng&9Qb18pnu6K)r9h%|y{6Cg>L&WsaeeAme};ry
zjm2{Ty?-zbHmFyhO#|n4>(gnds9(1PyYElK<aMl9Jw(TT#dYj2RDS^-`wS2K4I8WP
zkAVht>eDf3ZkIk4!$ckWd1AHu%YWB0`d?i73nzf)7xR`_dND$+Mn`bzr%q^@++F@k
z9Q!JP9;O`b;`vwE)+ps9C66grKe2R^<(|*QG54&1Df#A!r5a04<d)1Hm@>sqomjfS
zV$bCkOS^VVp^u(es<6=a4JxFt)!$$GHiG68Le<MCIKmYB_p85$Qp}}+6DoS*Wv!`M
z)-YePGW6f7{%Rq-WgJnQjJWxm)xQb&d{#)s?wAsPqxv^SC}9Ru<mRtbf2FWQYA8(`
zf4%zGQQ`-3d+h?r4eVd5{<Xr|WrxM>iF8egzgqpPDDj2d+A(ol@-zef%hg{7cs^sK
zc3o@>zf}Dt6v=1a(4nW(6q%?@Q-7uUS5PjWb)7#pUFa`Y|1t{YGggiYX`1@Q>Mx?y
zcrFoQHQE&TOVz&wDe{N^g^?ml|N4og$&a9l7xJpOpg}=Hq{+!aMTZNOBNWe*EvE;y
zp5jYAx@V@n6)}}T6%3U$lY>eZPwj!lif8!1Q5rPSM=Y3~?*j{_$ZJ?kr+w~Bvka9#
z`v<l)NC@Q48rv2oTD~lp8Z>kRMZ!Ry8!j0Ja@-IBF?hD%93oCRwEG!hsd6G$lvz3z
zi;J%hq*TGG`jL;QV9AwDwoOyk%*cOmV(A&mqJe!ZTx6I2jT1|M4~z3|pTdnx*3#5R
zmS$1m?rryE?juVV@kDOl`*zKE@L>02yk_sa?Skyw2IfVMC8Oqt6w9%(Hubn(ad9<-
zr1)@~O27u1XhnE9TTcMK12or+_y)0D<3&Joj2>;nRguo5ZK`}Wwj#{aA-087{XKC>
zlljP0@bQ!qZypcR?WN*EBSGGNEYi14y#-3R6!=u0%y_W-t>mCwPOqe=l~j`}uqwIB
zyugpHO9*)GJg*c8CMl6CtutBb41q?`xun*7g)}3{1$L;KIH1PNIy3;*uaai58wu7}
zLjs$x#-|qT0gx%JKzJvgA#!D%zz?xi1SWq>v*KES^EoU}{Sokd4oi)LRkx%>eh*{y
z(2U5)@Mcs5KA+dn+t>~k*ze7o#b!-1llkatkkp(j12U_ffF5yCl`_x94t9fUqECw*
zvI=*&`%`+U<ntq3D!GZI4%>VCxL5Mzx4f_SN^ar8f<*4Pw0A6~Rs-DaXiRCha!=*%
zJHT^&?#=nNHx`%9Jd~^2kmStM_D#Qfk88-i*RHmc58QUctM}ZZH7bH>&HAzJ@ASkG
ziAyJq0ya1C>OF02WYpY2KaoUp8OAC~>_4%-jqB;F*Ls};?3vIzzx&(2ZM0Ba?fto*
ze9*i9;OpO=VLzRiz1_<%o_Gegy-toeG57Z{FF*Sso0m)2M|y}2B+knHy{A7XTl<<!
z{oZSwombnz)O__GULxla*)F?<_I~UN3Py7N{oZSC!X6ptr#f>zeYM#mV5<}GzYHee
zXLSO8oD(qMGdum>(dqZX2-C08Y##KQ%^R42*FQuP@chEUrPYOn>O)OIOOvNP39n${
z&4z2@)%FKv!p&PxOoZQr2?;oIq{4t5c@-y5acTyFZV<gwc2?n9nu$JeNfPZ!`VWJN
z{CTsIX86@k<?A|?%OgzX?dJY|9rO5K{`U7(!8JFI(;0=MRl{#aH}{Y~E|plFU<lPQ
z2voM4{YBt{s|Rbc#Z;>?<|A_v-y5V}o*k8I=nXkStmnX==XHrePIe#Dida{39Tu=F
zPbjxfq>g%MPb&5zDi<_iXWPU)lKwFV|E;M+u*a>*NCp_(8DMkP+cO9N-)iZt+Lgqg
zgme|=bzjpH!mj20%Sbw?De_&7xFK%`3HanftehH|7th`DEV=DAnp+){znqGOJ<TEs
z$~#E(UvJ}@Et9eYkmkC#Ty$ZFB4s)%Yc%`y)*eW0?#k*mx0p;yQ}?agPP<f%*K!ao
z@AUfzz1fo|cMgSbw{VSDWwUc~yVKd;Yo3(3HGR5rdU|GKy463qvDeu+xv_QjOEYJ;
z&YV8e*n08wOD|#FIDPh|mruWRy7|g0FEw9$aciczakg<*FQjGqeD0-gwYDp~17v*h
z^oyrno~*>k_@FBjyNoQHVBJZRE;ibXEvFgGvB=oi8w*QI^Wo}O%->Qfa`((lNPnO9
z`uD16bdsa+u-nejL!pDZw;z<1F@>~bAdPjq(Xf7hR=gs0{;{ogVN};xbqt#qSN|Iu
z@$qaGdhFMy3pq27=S={@$60jucic+)z18XTF31{v*q<w`{K2jO9JO^R^lUu4tUGFZ
z^tF2D&i)41>Yx3t_jRp)e9kE26V&A%`$;@iZ8QO1E;WT@nDWoNhj0QnnG*%^1@*lt
zy<%Q-q)fT1+~5P_*%hN2^l><{$F%^lCbrVNl<8yR$@0qD!fc{v)o%BD6LKd8S46Be
zdad?WA7FRyPP~dVsNFgOu4|`WTa`s^;{JWUdXBky-LizEnQpCnj_+|{(Q`avXpaR2
zCot_go)|!`t7r+53MpqCyz%BtT&vK*b5O?R!&0q4OVxS;H>ku{6vINY027Z^@Eb_t
z7keVF73Zj8(@{Q?^yhMsP|)bzy9rQ#mG56m*aLpM*Y9xc!mkNrsg-vVi9sscWQY&x
z07^VEjj1$D*gh*^l3*}TzeX!+7|Dr0VFJwe3dYpA7X~)!?4!rnGO0VT(KP@(M0=iT
zG9xxk&gLG*cC*ct*T}G3GZ!9NE(>SQ>DR)3o1vX1<0)H?jcp>2WiNS?(k@TA%+;-p
zC>f$+J6%U~ZTCyeuj32wjxWHQP1Z}CLB|*1K}g<r^Mex8pYQk2pV9?*9Fbm)c3zZ=
z2jW8BFuM&`>udrau4DvpOO-dK*(H8_-8hm~a!VJlik6bNE84DE{>+~v3{Wt(ZEgY~
z^X{QD?BGZ)f~j%cS07jM*&Q9%xOU+9O1*t`GBj{`W)GJg2Fd>{h1@t8wMu_Sgv=&B
zVqsVC@nz>B9|!IohlGYXG~!$QvY`iY=I6(E<sN$r57(RTEWI9i0E$;(!93*w5DVq0
zdjOnLZuPjxJ6mL}Ivvf&2|meP*Z|2u{}~C>Y&7Ht6w%z^s;7kOn&`D7(NkyjQyz9}
zFh&?c5%<2k=%Aeo4j#N`08=#B>py+Q?5(6?kmlC9hfT9b;&%5XZfJ1rLDj)QWzz;0
zR~J$&4VVLY|Ls4ru9}JJJ9O0)ijV=Wnlg*-{`bFa!w0Tb?;2Ol=L)NPu={0tZ;slY
z6nZumeAYSho;qhr!#7gkmifxJeZbr@ah%lD&<Lsgh@t$a#@<rbO~ph7SwBlA#3KOP
zhc<?!2)K(bwGVAz>^`c6%4~8OIB_;xpxCu$AY?JJ5b^jL2qwqvw)gfe(8cAoh0Du}
zOEb;~ZUE&<91FF$IKMRSLEwhs10PP6c~HBsav8^iR~Aie2Rq2#r8Kzo=0#u7yngpq
z6IV&;`>(#eupBlO|4vtY$JP7J#nZJH)4h1Fd9ZT_w~T7lKeBM?&3Ay82{*ywf9dkV
z>cX0PvI=!u@7bPkQD3t%YEkr<$`t`Y2#Nv(;RWWk9aaEjM%03|dc1-b5GlqLO=AM_
zZ8#i1=w;3cY71T*7HAq5c0%njCz+S3*32d`6PcHmiDpnb%RmN57>E#Q=Fz^%>ofMX
z=ZLWM87^2g8YNjc#|WkXIE=ML++|*oyNF24N#V<|ofs<J%$*i0N=;)^)14L=5N~5f
zG%uAA@ixta_Zc)cwQi;UC#uisR`++MS<~)x|0^n>M+sW7;{-{)^C1da$+Fy0GEP{c
z%sh3ao#A077>nN7c=zi#cFaWcj+oUWFY43NSM}W^Q~0rM=k5Nk)8kwi=qIv#`(XE9
zg#OmsHFJ8D=oOKhn9%>@55BLemswFw@Q5zs@7H$L<J+AZtoUg!(~7YiymR|)1=&T%
z#xzQ0SiFty11$2Q!<msc0w!n0mBD|4Td4iz*&jsDKc+9Qqo4y*S4d)?y~BDHmy>cM
zbl}y1-o<JCqy7o2O$7**pvcRKxQ212))5*LL<8vtsEz@04(=S{qExzLHNt13nY6as
zh^A>&67L>mGpSGMBAm55aFX&rUgXzkA|I=Xym8m2O2y_@W#f7?JD|yay$v7e-ks3B
zC%O|GGO7AN!XW>edz&)%T_)Pp^Gr5JeLu<FErSf5gkck@8X9s<Ca@@#G!^`MCSjFq
z6T7GMOQ-8#a}FSL3DC7k_=Mk_KUa0Vb6=nxB8BaMtiUs=P%$pMBc#Rjelj~2b9P_4
zx88!EJh%jsgW(Y34{5_}UV|{{&MLu7`dqu&T{7^cYotZy4*M)vwnfmwF!ktJhfS;A
zA4B|o=QCAiUu8JRgJhA8+sSsp_=8Eb+>ES3S%*|%I#7?ounIlo@LD5uQ+YPMs3igU
z5mZg{smqi3+Ba-qqS{lL<=neemQJ&2T)bFZFn7BYLmaOh91VUj?L^4OD@XPg3()jQ
ztR0M(i9Szc^<WD|u3N!mHvN18``m>kEyU~;O+C*@1AYHNUr&N`y69N2o<b@NUSB>K
zS?Ar75i_slr-xaPd=_&jeQb64+lG9bYDj+CRbW7c0_Oy<Wv%FGhOxq?;$_7lGEDeK
z4GhkdA27%_ayEOv>R4)L=<4>}vg4K|vjZMRY6ocBJu<mn4W_9WnTNVSAlSLfOQRmV
z9zkOy*M%iqM@M7lu;?h158;OqKZJE?6)I2cNDjM)l{~&7a15C@0Pf|8<oEEdmzMb1
zR~g3lAeR1KCy$wp6!Or+sP7>iR!611uOgT7#OIqZkG)%$(pqt9YYIG&>qwFEsaTFZ
z`d!|KbN+gsW2aefz&vf?{4m$1SMR}PPzxS1E;!gq$NNrSpo=!tJ>yv3dy*N5ap=B}
z<MScL4uMDSOIp)>k|7q^j%e$Tp%*R`VQ`{>!G|Nq#!@`)HmM4j{em$Bi+SE}fT=ew
zQjVT7*E}BPcN)VtKJSN%@B7pn+QyI0{!Sh|oj|1QJaz~?dS8wO5;o>>IUyDwo)b=O
z@<R|6d!2GG0+6-$3Gc}c603d}f(q3UGQeD1UR}EuW+O<}Ps+0o#<Cn}3?GU#1h2lj
z`p)WW$gmQ03^DAlo-~j1;tR>4DFzLf%!lICr?g@ov#e9}2hGb2nLbFFGaim*Q&&BF
ztokv8hkolnwvnbEubxJvX{nfZ?TvQkB=C0;e)oG_`7_l9%1;)TH%eownzJmn`v<>m
z<CV6mxAFU*y5H|re-6KY-~IlD>fgfe7ul?26aROrzhR<UKc1SA4|c!GSXO=P<H5}1
z&lah?=DkbmDwTY`xLXE(B*IBuUns82kjPYd6k1Bn+9x0D=wy;scxb~yekdJs;F8I^
z8zU<Se?w5nPkxA^yrM|H0>dU(laNTjr*6j3YGLB9U|A9+^MjNO%&EUeMfB+^yK>6>
z$S6Jr(^{Fv2Z25Oc>zsD_P;qX_|dA#VDzzwtsm@uK8*jG7-6C_m=!bq-bdfp@UN#9
zay1r*;~qs0XxQ&l7X&^Cks=`_5bzRSy$_O1@aQ_}aAcgHs(A1SYqLkkwt1i#iVIHy
zWn)Xls(=*{bE8|Z3)s@=2f3@1aOYDpBC`Dn6GJ`}n?7atnjiFs=hCO|M)O$He|BU>
zbzOZ*?lO<AYx&4Hf3t`JvCjTDSl^NBclZBvzrVNovN3a>jYY<T-8ou0pNXG3L87mn
z=KtIX3c(#J{{P2EJ|yY@T#51dnF7(8D^v9Z&hd)hZ1)Q@R)>=R8)BLV236Lhs(SYS
z?dIJ_neU7_x1Wu<=)rC^RJ2D)xUDC)Ft>hl?0ubEAI_5&nCGLNh%iyEKOzS?b9eId
z9v%m_(p$>Uc(S%~eq}c4cMsc}lBo*GRW>`_1~bng^`|6a#94#e$zHR)-QOA0V!q&h
z{YEq4Wg0lm&`gl0s-J8%Z=onQe~_pJ&r5sFdbiDktfeVANWqJaaEjtqtG}ag2vdL@
zpFVx+RK?ZrzT`%ny0<BX;_)A47#6{kMj$tQ$Dxb;*});B@7s|CV0PpK0>fgVT;!>w
ztd%n~;=IWQqK+rGc5ud9iD}apA!XXlmK=*AYi+f<;4NAs_OM0L+vyzcHS}-}gG-d9
z>PfqKYp>OAI;vZqWnrf*H`Y&GLuy(cy<r-q!gm$idAx@-wq>TunMx+`|LHj7%dL~X
zZd53T6tb@#A+#*&kKj9uIpa>{vL<Qcu^d*Rhjjc>SZjpt&?6CM9$5s;2^jm+bCR4m
z3OoO#&QFjq(uJ|qnQV;hfaOMtqyopx>F%ejd1BksT@e#$Ns0&i$4}8-_R;suvdr-$
zLxb2o*J8>Y{NdQj4z8(r!sBNW)%cs^zq9(CFj#;5cl~za47LNGFXsCpm%&yBl#DO?
zGM9JpBKjqJ!Sv{*C2r2#8)>hZJ~g22R9YZ=HJP+-Z-Sj3P(GWE%rtJVoPJ?I;q0Aj
z(yqODdQh)bE=Fn0%|@=?DP11aY~jt)gGxmr&0OiK_e&4riqvL_DW|tgk5<l%XmHo`
zpmM44jIN&j;sBbs<Wi=D-a9?0B{>e}7JPfhv)(>EF`y-^n=VO|X-w~+9>ggrCDXR1
zH&GAb?_8&*X;$x}9>m>QFRm%7w^9#cP$!u)B+@%)CkK!y(pqd^<%a67mj)Cw)5!%i
zKw|Cwf4^-b3tp^Vw2=j=_58u^IDKv|L_mIhzPKR+KN2de%em`56S?b{ow0V<$At@U
zBDBBnnP*C+(i-$!XuaFrIx;Y|H}?)3xQU|I#D+eyUE0y;&EiqNPXn!<Ag6Z&H{s##
zAsm{4&0H75jddro30F^kcy(nN%JUwwg>TgRn>#GW_fw*jOA@|Kt*qU<#nj+Z&SWlz
znbrqLWemuTjTv8qwrPL05BCr5K!r@&2li1PJicRpFwwRB-fiZkg1FK~f@GV!eC^Ib
zv!S`p`7ay+Oj>{r*R<IhyISTyvphF%PH+=FDl<rrE_P@@7CB&kT_0SZO0EmO{32Mo
zFO|*1M*VtaOsctrg<!vlR9JfOaXR?Odp4g)`|?xNC=4Aitsn07Wi(QjtVK}qsJKq3
zCDA^p<i8LcD|*!5_go9fnN$#3Ft4<Fwfaq5-$6UO9Gq*{8rZq8ni?_(2obSqMf#F)
zYUGMCIYlmb0hjf=De^7?OqfS4Ey5;kP-;Z7xw=6+RPISczGv$gW?@xRDd{Q%zTE8N
zDvN@_XtfUx`#n^-hg+`WI^~scu-x%NGtxApaL7DbSXSSe!qQVyv{!L>!6%|@5I3u?
z65cA4Y0>h5LE4=swV1dEyi6LFhT&`Rm2?i8Z9hI!ND#i+;dGyyIP7mtzcMk&S-9nm
zYtB3wRMf-o6^?Rw%Nwtz4F`u*&uF#e6C2ZAK}(uR6auJLn^{dZo4wLC-x`Ucy|RZO
zS#rH3&q~*Ps{}ONjUn|RL1oYnx|rYPtrD@e!<O82Jaivkr%m9{t6XiH47aufWeCRQ
zE_-u-O3NadIUiwY%BJzj`zR(YnKMF){gg#+iY_k)tn)-9=pO1SH*eSXk*x#5m~3Zc
zYpsV@A{VBNWUlpg4!5@U@Nffre5vWAuE~C4>!QQ0F6gq)O-v9$Z<sW4#nV&ZJeL7K
zD+GB#(K4;}UX~DxC|(g$dBc&q2qtE(OvlQK8hkI(FL{g`AX;s^uIN;d@&l(WXOq_D
zd58#``i0{K<HW2*9V0O86e6v`*{7r_UmG5oGumbv;tnf}qa~7)on@7{w%>usw4zvh
zm@0W|heA)<e%1IyJ^m2MrqO8yc}>?`<qJns($ynuCxOLqT`Gr!ME_p!W0smh0)!0R
z9>7(fJ%T-AV`W8tnsrYWv}a%v?VwWA%pczhR&7~UaN!74ikek2I_)z2KnI<65By}_
zagNS0&yV=z$D#XyrH~T`+79nc;FK`MO^I9uMJI~oj4O{)$*|eR+sH6U9JS}W+lTw`
zA@-E<Z3h8@CC)*?g~-%4au+SL^9T_zzXf36JW)D%9y{NJ)Q9?^&yzBw;0^`<{8aKf
z5>!nk=aJV`FJFOn2@y5DxzlNFHj}T;Pb!Kele9Z7cX?sLScKL~PVbXMG-fHF?H52v
zS+IIAf?OcVrS0DSD>vZeFZFIhIPy!q*J$08FO5d;a6{ffQ0z6E@ACUTw53+Pi;V#J
zRmb089-3zP$2nn3Q-lmMnj4^Xumu(?=2Aj$DqF2~gDOsUV*1Nh8!ud~;NR!rxh3?Z
zu5WTGJky2xyN7kQu-SW8{tAX5VoiydgI?6y{Jq!jox6Ab-hypQXqDd-n4#WpZr{;{
zHao3$ZELS}kkf=vV6TIzRQAiL={TAPVauhfrF&N=?p<x)yIQ$-wRi98?7gcXRVhJd
zq|<^31&=8wb@mSSTKzJBGSwoF5$P%ElcWU3TYi*Fcd+h*LhVjzASJmKRNI500xdeH
zPx*KQ<06FDiN>IzJ2>3fZ#I;g8?AP|dnaem#*>3OQEG#qODm%O%_e1Z>0Y~YZ@+$L
zqdDo*c8z1g_PfIQ(AewHwacb-nyBM;I>PUdB9-lK=kTC>YI6ORS;^TX^+2lGTNx5@
zW0#gs&imE^#LCK-_vS@@-kV=ueS6{ZJ=I(8Dc1K^uB=_UvUYDCa`S5WdFjGC=*Bw>
ztL(}${#(%w-CJU}?k&x~^ZG*2xly}@flp)ag%KHQAD2HVzMxdpd&hGbFkhx(cyP8f
zxjuc`-R2Y(b+re@($gWk=!Nv;0FOYOvP_<UNJM;8LJ@W-u?Xb46A^2DmfcTxi}2a5
zMYY6SWY%JBBVYQ=f%P#{_BSzC{7~l43?tAcE1pju%d80cnn$$9KR>KwuHW0Zw|TD-
zjPa=BJ9KP4cMhBgZumXCaDIB|v%?;6G^c#JOIIhKe>w~dN=%d|xYR}*?0Jd&X+N&J
z5$t*aud`IP<3sAgMsAvT#q~IR2F;t|iF4T7M<9dm>NnK1nb-BitRQyjNb`}O>c>;V
zavuH;o&SXW@J|i)GNaIPNui@Sm>6auvZjTZQN5<0I`A66P+RWwFVd4j8>-n@=yp3@
zIDM|PabqtpMevpmEk$^6bpF=Fw__bRpgL`QYKW7`7vK}{jkxK~mz1+kvl!=K!9j-*
zC1_@qq^_=G9KfLGPRqfK)7<ae1Vle5*Q{&<T;>ua{Yz)P_YXCi=fI*AP{Z(sz9bq6
zNmfZg5X#kSV5t_UxyHBrAe~R=5LnW=XrN&wx1pdm5h255vbd((q3ci*lI0p|fz8GG
z``tU9etf`>Lknh0S=uje!Bx{Ys_JAaD`oTcX7ivwpqf$%{b+ZR9xfxJ;f6oFce_2&
z#|>W~_|5trqQYQ<AS%iACBt`Ob{1J0M0;>w42h>{(=!$b5pXbB80R+Y`GnPs&>cZC
zUb-C3j2Ac7&D2GL?H)QYYpwnIb`u6*eH(0i8xBSPV6OD)MZSF#tooWYr3lmm_sRB*
z>d?M~b8_jd0OVxyT5{SChk{CoM-8TO^EURG;5W12lVbd_YVwOTCF-d)zrtBuP?ds?
zC!2@+y~DN;Cm)jj$uc%T1aQBc17ujs_SohUmfw*BN}gECMV*7AMRRArwYa!+v5kAp
z;XH3%Zua1ScpL7Ai`+vfuR+4d-jkUS;t=+J9JRk@^{(CZMc}=T$#6|~uJku1-?eMn
zb=%_QCa#`uEWu~9_oxj^liU?!yXMYc{VG!tEn`lt-!1hJzzH2CIaPTPhB~Q&A0<U!
zMzeJP+Jm2c1~*0UbN@kk_lrM|eNsjrR6kbzUhJDL{S*89`>U_w_mq48TJ<V^zv6yh
zul^2x|IuHx@c&))_woA_Z>;!psK5G8s(*;ck?}!vuVAhpHhX5{_L=H4Lkk=pz}MJ}
z3QP=a0UJwZb1o|M#eszmItRF3c|R(V3@Z^qe_<f>jpm)W)1`sn#3hCwjLdeke*|PU
zI|p|n@WIGzGN?C#pB{*Otp+cZOu?N_=SHm-6&*YbDBAC)N(>%tlz<1jvz01T9)@AF
zPZ}S+6g*041c!AAT288+$syQv83ussw^PNQ8(Qoj-O}?zi!=@qqtd*cYHAqKk@TMB
z;;|h(^J}#X9fna3!M#UpqJ31H_WG$7hT(*DGgV?(x8YHa=`?H{dC_~S$S|y08QKnQ
zq}@t)eOO1f(*=g{F~kyfkNfHV51X7QmNPkr2_2N;<ZNvmW*QvYu{4Im*k5(k)EEtG
zrTK2Ez_3AWw9;%ajB-#saws=KsT{F3Y;YM*kBzQgs#Alhpo)nJE&D^!7X~jrqL&%?
zKdAlz^o{G=R^Rxi)ql=N)e*F=Jhfw_ayi;ZWt8>`l#mxjDVwK@OpI7CT5HRr6v}8O
zlcSU(m&FRoNEoK|l#wvZs4F93nAKX&j?%G#%F9T@m#4#w#4`n|%*hea&J~)Y7tWCY
zBX&2d-<%o^<pMQlYShvhZRhl;CG(Y@5l1Ok_Zdl=<fuR+(m1OLjo4sDAsVq<MlTu>
z(X4tjYJ=%DctkeHQI<wTHLEj?+GJLB8nKfZ4Qj-hQlLnUIJ@)psS(BD0JUn=PNzq9
z#CevhWR2KnM%NmVC$cKnh&0aEyhbmaBYsEZjRHMvBtk{%*gqF_YNW+A)XM%@^`F^j
zduTmj==Ou%2}aw02gCd%IMqIoKsJU}ng`#l-qRp>%)IzS)@5qmeBU6$*1(LUSqh4y
zDXj`R!Wv`RKME<O?jwyEOC0Y0^h+n6=|A}TOmyA=uI1iN4J?Iy76jH+HugH5eL9@O
zr#Oo6-p8MLMveerhlMenCg`8O$EjrHfMx4@Q_0#P_H2v~(wz*@z<({xGXS=5+noT=
zaP)S&Vox*RU;~2JF8C@Cec9tIXYjq&6g;WOdGwwluRnleP5NRRYjisY;Ncq1m5pf)
z%~rc}fY30#=D7>c6{_5Bnzt3$hkZm!r$}MJ-Q#xc-5WO#>YF&bUL*GEzXz)H?`pyk
zZWYAWDekly4LAqH;zg{mc7F{=<2zk{3={h{tp+ahD|(Nukd0R1OFN=9o~ti{+N<Az
z|KI>IS+i~s&8>+MP7#LiC#UaV<b?yKA<GNvVFlR|i0G_s);D)d2sCQEP{*x-zV|9`
zJZFMTj6jY786E1jfg0TW!~2N16?y7lw$N^aBL+A*ijA8h5fKK<rY7#@a#VPdyK3t`
zmUaqd-Zfi6Q^}d+wrdLlD%g%2mv@B)an?jGJm^g&Lbyh2-|s7x6(bp$sf37`uY#Zc
zn6%i}WQ#{DE4#j#_;s>fh`YzUHkH&+MbJ{Ew^KjxN8>ko?@c8F53syC{MI8kSsfSI
zGy06&_1j8Bggizv<1$Nrlg^-SU(^i|p|L?E+iuM^TtW$gzaeov=@SqPa|I$NZUgtB
zO_O4?;g}`Q>p<ra9pt@fM$^+@n@2^0;;^b*1$jfss@wpgvpOTNn^;D`pe2U`n8cBU
z*8bd#54vReVAOMJkhfDIkf{Z^2pmB%LT<KkfYuPW9~15XqDJ!_lpQ*8t*)|oE0Y-d
z!W?zqEb1grAqtRk3dlPNTb3tCno!<Ib|w>)N_x|XiBER)JECNF059{?4Vx!RPQ_76
zX23V#<a21dFK(q1afMdOrz@v0+VOMWSo|FD@q_i*DZP%?{5yNCLMasl<i^-as|a3@
z;Oc-pIEBURCpDa7*(=9QOd$nI^Gm^ax*CJ|iwc8ZuD1@<4J!cM_bgb<ZNV=;tSUPV
zE6XG;ue)Xf94d$H9)31)C}svs=|EiTMF2so5eRlvF+YGmW{AUFR|z~UT#x+{OlIW=
zJqYj-UNpl3{Y4J72<NrhH*2`E7s^n56Ww)9vWh$mJbNor;O`AYp6?`$1kA0h-b)%m
zIx1Hs#N7^6kK|?xNphlf!|}h7#|T2m6E)QAv7xKW1IQsgz#Uj!L=SL{>6sHt0pbWw
z-rnvuw<%#UI@t0?VxG-zr-z%14|q=L29C6?Hk(+Zx>%)m`u&66?8%ec*cU(CKv=~7
z$%8w>2JUaG*TaVG$(LW6IWyz>2I(WnKup*JK{eEmxB~=_H1u5$ae6g4<Du1q*0F5r
z?ctOH!w{qvyF6pPO-|-j?Aqlp865(;EfgzMp<Mf36P>K>wQe-a5!w8tXIW@)VNRZ$
zKBVGf@*0d0cp<q^;pw+}H_MbP9brSsg8z8ft;ID=$S|o#w90sd`-CU`^J`vlVomlV
zSVPn7H7mE8t?ivYYkl}CJ%zH%K-^U`T+(!!%5I3#MFJJwuq%$6h)zrf6<wGb*xpvf
zuz@LH9$NPDSf*Mc7o{sUtCueWel>VGG2DweQeXEH9J7kMHWkb+6Vzj15oj`=oNx7b
z4A-8GQ+q~}Ynd~lu0^EQ>fK<Fjiiy6C=uF24DAv@=T{-r+r@R~Ef7c}zpUP`RUiTk
zcaO7N4+nB>9hAWbFob9%u_ZZ`RRbxI<F-kayDg7Bmy+l>n`P^kw<((f*dka4=jO2H
z&#E1`4o(wxgXp#jk7|CHei~0jW|UEyq>$A4`8pO*wS{KAptnI%5ejN+nM_P-?kV;>
zCg?4<wmta_4sllGAe*uBI;~(KdnHR2a*^Y#>}&0g{Cl+a=5609#(G3U=9?rAeS6JD
z(>z;!8@pSVe$W2CTD^|n*WB-o>Xz}deAHI3TcUsE6Ipa0=$q1TDk|jR?r*3c;q*X1
z!uCOb#*ctH7vM(to$q-fZUmc2AS8IPn_!PO_IWpf{Iqs`{or8lPOaNP;xNn+tP{=s
zjbZ+RafB0scWo736-j0)w=1_X1-5v^2Y=Po5CEJez?ch16wii#UGZNLkB6;(s3HhD
zP1k}s)ul5%c0_Qp2jvgjobhV>hVF_mhVdbrFREHcz~fNaXyU3rBY12#q%e91;Xx_U
z(68pCsNr5e-9!+%$!E0z1`fbK;9L*lg)$C^3k*k9adItKCB5NUP|g;?B&1#A*VnWZ
zbgioN3fs$+E%uWPET(yVy0xPIYRd-WOry5+qaf}K$becjbm-U_?znLz$|)j(Uk6Mw
zenfsF<4}yAVHA@}rDAdn_!R<^%=R>*Q?XA%rzHZ5dmv30i7&XwmEM&l#0Jz_O^Sat
zGW{)vC2ByrW5f`Om#}M8DQs=Jn>Z0wuXZD#=jK%<>m}E<$?F>8tV=fH5}_$JDv3XK
z5jAoNk#-1^5Atk6{zw7_9G!(muNcU2`Ri|7TAP`-i%S`_gCy$J1SRiZo1^J7c|ADM
zW!nq18a@&m55q2sm|~vtnnh%Ru3=mWw#=pwxSTs!ljh)`RIbh~2JPj#MjSOo9YO83
z#`uN9CabebSH#Jdk7C3}!%%bWc;}hA*`iu1><-)m4mu)d<j*tM2`K_IP<~s#q1RhM
zXG`Fy7NP%$b%^PmTZ_CgdUi1`u2bVoADgW1NFvT#%mgc^HooitF!j6yE}o%QYaGpM
zs3cHfnDw$5&)9UHgAIL77c)JUKS5E~1EjLTY8HFeR1zkoE0~$$+_pV)r&j!IPPxtO
zWRv43PTI13?5=~46CYog4m-oVo)^Ybhvm|RVn?0niSS6bi6p<+5`bx5J3QE=mlCt7
zO+0Mp27pK|S9d_?PP1<^3J~ScZ-jl1%^T(Q%=8OgJS@YHmczf;kb(9Nm8@Al4d*+2
z^WvPRr`ouN@l<^ybgat`ORsi-&Tz7L2d^8!@ni!&n;Qo>V&4xOb|9$LQLJDo5Gdo~
zN(&7s@R86~Fc$=Kn*jG`S^<SwG&VK0^WL+B-&mpOR)Q2aZY3Pzz>napJ{w1B<{Hgi
z@!`83^UdaAO&;>C@;tIy!T02KEJE>fY6@Pf-h0;5Af-U-e)C|?e3!kIIjsdmJTn71
z-Mk5Z%^b%|zebIF7G(o!90L!pBGw9Y$t-q43fGw_?eP>2HvsKkXR9w1*sSkq8j&f?
z$HQKIPgc26Q1UcaEJ1d-@RZghycq+eCD@n)k$lIptIRn1*E*mzFa7jpbc9ESA=FVD
za#v*&8zEdgNId(=DQ8^T={G~mr3+N71NSu6K1^2na#Kl;+u~d#qhY|s_Ftc#@pFwx
z-o{%`P0zvooi3e=>K-LI%kIf!ikojMThN!#pf?bsX}Q=8JlY5hi85@ibZ6q;n;Ik3
zBC-nRw%Jm<{a%P38Gw%*!pu&=?#xjD%_&i$>f+XwrL=aS@G$J}NRp<+lb8ZgS4bIW
zr)B3QU;E(>dx|MUqL~pFlHxnYIv7{4J7wCO^d=bF;7253nU&ZniSMN>oE?(*jUN+P
zbO)_}UMlNS$~CcO1%v#sy1aG0+r<;>s97+%6NX~CzZ)u@X~&xJ@6$0Cq8JO7QSo?$
zCbfUj6tP_ncQ&|f7w>BX)In*XYtB~0P;FAj7-vj9v2Asm@MCp$xlv|^!BZ-DXu-vM
zTr=L{xRhfv?V7lVgBqZCqA=W7zJ6E2=uKMC8&ZfU@ye2+A!<|?n7qtUlWbpw6EiiY
z+0Wui540d_xUbj&O<f7OJxyn@uzQSPi+lq)ryG*z3JR*FXyB<Pr)4Ko;FqSLVP;1<
z^_mSVBi6o%z0zZ5XNR#LfqJRST7e^4T<F$^D$Y+Gn0!YxQf%e8qFKzlCFPg*|9eMt
zJ_*3<DvVV!I7e?jv@r2F_A$wLBjOUd5r+DxKP7kk{`~J*ANBp}kHbg(EAIEdRQ)p<
zAN6zeQUBl&)F12$*cXNpTLBMu|10%ZpB?6}uC<$;9GCTX9wGA0$E$_8ZbSMG2|YV@
zB~#(edZ{1DmCRJOy7sDs$k?4}p&^+EgJsxcA~chxECEsa+S#6dUw*4uM_SruPv@4j
zg}EtzEB#Jct)&e)ntl2u1%j>$5d2#N5Y>WRU#+C~ZWmW}k<%(zUa-Y2F3gJa4H*n1
z*v*s>!`rQ3MHcv+Zr@K@0oX>&0W@1U=y!}E4K<#fN?GM+CdX|NV<Dmx3)^BeCqrA|
zl*&{Aj$W9A9D->q>H(&G!PPL`reDX5t&R(W#01np4)Iwy8mG;AucpRLh>2aEOz>aN
zD3+y3ZsNpX*11EoT{BZT+kYejUuo!(@TSoK<38W8cSeAtPM%T!m<@WRD;E%><s99T
zkuE#roe?i8qpY7sXRE+`HgO~nrkYij`Y_G?WOREU4DE&x(H8sA1$@k->x3T>eKeBk
zA8Y4B%;HBhiJk5DSj^#}rm$=7(ahj9yF9e{>+pF*#yJAtIEk6VB<>w&!T`fK%~p|f
z;bDy8ltDa-F`PAobyTDeQOJshH6A^f-J95Mrio<MCDqZ&JO5KJWbvZun|khC;&a2c
z8$$r2$V;=vTyqUjd{ba!dPmP9>rDOirYPSi19<ZTMubVB5<IgjL(1(RRO7mX;_iDv
zF~Qzy03sUSRr5F6SaPczmH+{|Q3w=m^W4k@%eEP9DhW0t8Stvw2trW}cs^432-m0v
zvq2+O7jH>%Qc7pAet4#{l#V^_)*1u_Jt*#jlkz&db3@mWwa|fRe^o;nX%EVFFkQrS
zPyPh^Peg+sY0z&2In6^>O}lQ%kF;ls4<Et_#bc%7;01MpM$CRD@<7d8!%l%}pl&!A
zpeXTrJRgGST{tL+d{91B8v|g=uBkj|sSh1gwhd}31Rm!7QECnF{0PpGbnFfEr?89s
zq<sQnU<$@Zh);tXfaM~;fINUXzTE7rg1qzA7g7-`2PBohf_7+o+K{&+Rhb`)_IG~I
z+Vvl=ehGHHk4^iN5{ve~y7wE^hOzU%O|W{fyFn}82d2r-F9z0pA(^!3;qLE_a*$1;
zx^m3<H$ORg1X{|RA9R8Z`RH~$c;MW5d|_x)SV(X1)D$<eWKo?$!1MvPY9)yCgYH_8
z({WEO+$<IErf2ce5p9nafkxS*3Qn~2rXpdjh*OWRj7<<VrE)$mRtsC(EuN#qy|-a7
zn-*Ba5vbfk<{*)JFoSl&7&ucfK82UjWbzK!Io%}vG6>rWd*si8#xYyL^E|{%Fqhx`
z85TKXtwDs$Z?(27SO~3MXbu;6mw6^?tCWSd`hZi|B>X?MU28TaV@ppoA#;!uf>&0j
zV)cxx0yKmae6glRc+mh-C59<&9z4XaBCBgiWI@-h)<z+7{-ZeJ!o<x)==&c7>L!qO
zf;B^7mR1M1lInE$^D(w0GZjbpO#BGOFX)ip%@#0loYfDtH<o8&9Wtm5yEAbKUM=iv
z<=Ad^NDf2>7|t<q$Jm}Ihg~Mg#78#gt5<a=9A4e?)-%PZFfWqX94=Lu5FR_~1{G|I
z1aim(s*Turo<)GfNNyk^Nr%}RMM6aO*^?MZdaZXoL#q%!jSLD8M~3tnSkY;e4S{`C
zf>PWpYLX+6mO^1caREAk7<l}oo}!V68%O=?juPG@fOrL%@3M~??qqAi)!PgmQCF>*
z*CA!%`qdCIN0D{s%N~mk5e!a6V`w8dJ0{|wJ~QLx=y>ubR1I@}1(9J*Ue0YUc$f3{
zoDjGt1$hMCYc^abQGXa<%Y3?!*yt}5_-4((%PE+eV44oRMIb)%is(B~Ph^LY`UN$(
zrejG>-pi{@rL{R9WeDS{2#JQ)!=K`-50KLdnc;BH?!^T{E<-Ef&uk&V4Y6xY9GbJ9
z*IPI?s?o!0qUjce<hvBvfzttW#H;w_AJ}4f_EHduLj7rgxCn8Kf~&m52y8QBB{B&&
zL(0rqMimj2!u7}VUS<t<14Mdnm97VEUAGs&<KPp{RWaT+xqF0G?<1t^5E+hpK2}p3
zL6UJFZ(w?K-vgPB2QV1Q;8z89C=W+vQL)M-FwGkjVfjvhy^HCURpcCXly1v8xztP7
zGsUjk70qiwnBnc@I%S^c3<2ZziqVIpi{8l~o(Ms}@08mh4cTXF*KD*xYI!`4Loj-h
zJsmeN4#CN7qfsDtyggGJFKM<mrj|d2d>IH1<_(d4<HL_rV6hk}vxzl|rwPR7lp}@a
zMI0HyMN6L`G`Ek`RW2g8*WI)~bZx@63$9_(V5p!y%oXRu%I}(&2E|omEaW&oEer*5
zbA*#S3JcLSz&o-DV8U8ca1!ep1Lo6yF8{GF8cEHNELc<Nc8--yyUA*(eC-J2XJ^!D
zVB*olJ_YqdS3+u6H5E2v=p*Bdg=H?FDz7k@<dQXrx<WMv;FrzfVU4THK_47bRN8p?
zRb9&|`a)6307Lm|rb$6N=D?bDW2yH^MPq&lP@ZE>N_kJ8{#@i~#nf5U{Gc9=B9tmQ
z#G4Nx9XVRd^Uk@L@dS>gj-`H2^=ohi`jDS%)m!*o`Jb$-@V)B2OvumdXdyoW=FX@V
zQ}@9xArE)|NL_?y2FCp4XL!OX)o=M(xp#Z)`Gwm!Kfq)&xPbI|o><fiN!+3B3(S+{
zC`3JdO|nK+#@2A@29AEo0W582Zf>T6?P`X_&zzc>IgNh+Cg3<HdWu>yu6c#;8wq!i
zO$xaY%1etEFHed#f}5fDd$_dq2JakkS7P9VMRKdty&>KTs&zgyPH(J(Q{P>LCD+@y
zi)3~zCw~KP4z9OODtaf8Eo<Xs<IKx1otind^`%p%PR+dh;>*ou{Y<m|rL!-;*f_iO
z^5)j&Omp)^G{l?_JDs*Ii#%E<ora*|t8#E>d~xQ~%VTdhC!m~ac3hx?`Iw@k(g;V?
z*>^(_O9CVq;qi4A!fDplZJdZ3LybIkgU8JzJpl0tiTCdku4YZvpdf=-bsS%)cjaDl
zIrdQ*ku319;`FWS^dZdF7SdInJoVDFWL=uZOn2QL8#^c2JdieEI=N^uLiVcuj$KHZ
z`J&xvPn&hsm(G@;_UEgYoEdU*&p)f6(5jxgN$TB=7Unn3BV%oVa7I);myg{;TmfS*
zt&~fzPtUx9T_sFu#mmlZB1YOXd0A91?TeVfFr+Zw6I2wmarz}h3gH;YOJ}F}=15Ig
z5+)n1I<8Gcrw(|bI-6nM1I?&Losshd-Gnv?Y&-2ebQS}m46(SnfN6-)!eoH<zb`O*
ztZfO-bE(PpPkP^&V+ZzmTk4h!a<Xv;UJPV?WtKanM?r!+Orzcb7zj^Tkv+|db|GqU
zpC~eE!^R+zu4=_nPkri0)Zi2Wxf)HRwsRaWW^`w3thbM9P`VE^#9(>eSZm<+b9G$U
z%2Zd=$(or6n9|Q<DJv9B6+osa*3+|iu!)c?f*D^r8{`RCI%5|n_C?1yed^S6IH@Ou
zG`Akb{fTbk>GCK#4Dp8etiflEL(rfjWH6H$OC`wG?Bar1i@Ti-DnTjoWc)UourqLm
zn<5jp`_2Oumbti<-02)9xFuI^{N<!K#%mHuHyLD%<CK7#uCh8A;uNN1gD?e8A(ov1
zU6AFhp?8x12Idds+88cRiJsL|gd(9OE;?{DDOHk-6S%X8z&r&EW)%qrHM$KRi|IGU
zsn}?3$sIJf>?k$Y%4ZS0bru~o6Kg7Y1rJ`qgD<^=Nk)T}!`SQ75x^NQgDHLz<Y4KK
zVHuEntT4Dp@j=<s#NAk1N+6|Ws!4F(d4mq~?;xAjEg2-Tpc!A@JOR#>o{{^{T{%CW
zn1{lyy0}>|%Ra`uOl)iynL{D%97hv+<M1Bln3<NckVy*$Y6|`*Ub=0{I#zA<#(ZIB
z7P70L6t-ZMWCeI})}`8h!B1kKfwViwnnMXA6_Ji@?We`n!wZ)W>9QMO2!%t~VH^cs
zEQR$AbTks6+^RGyQ^{rIXKFTd@8;yPFp>Vy`4qQghzijdiBU_WaA>UhP#KdepG`W6
zcmq?ymwi?ZX2OhrRWK*F4{`kzCrp1Yu-z#)A+kFCL+}{6y4|@&PftIgqX!GiL2vFX
z!}-iqo_p!6aFBodO86Gj8pjj9(wFX5Y{g|f<3~M#y}5Pso6ijBIdL?W@0MO)Tsc>*
zE#a~w5fSG1vi^p*$2W{ci$9<vY&Uyzrx9SKt+3W{mA=}bQP$!)On+!QS$Jz<xwb@O
z+Z>o}NZKD#1)7_oM<uGb7-7!#0N<C+j)X4KTqxe~J<hF9oSQ%Q=0dHy@J?-Y;j4?l
z?yE?!mwj60)7LM~uh!Pyxx|~QCoZnm&dpz0onMqULd7>1=FgwMxHNGMEliHtQQ2u8
zcCk*jHhXHR-F3E8Ny!0P!;0Bx)FFMgvAM~A*Gpz$e7AHhdCoktV}-3scq_bpto3lr
zitD+Nsb?8a^6D|Y$gj<?r50606)RiHWD&VP<_WNiivT6ceCt*NSuVTn=AKgGt5ZpJ
zDtQZ+utTT8CMc4e2yGU5*Ztbkr71Gj^2*u*QYI~3qLK?0brUy$Z|)$J5OSRh0+&^)
z`E|YEuTG0q7FN99mg=Gdui64`M`|_TVBf&pGhY#iZNKhEKsP+l4G(mq9Rc0+KsP<m
z%@}CI18sPqjkvjv2kLmBjsdc6YResz9aLqhK-p(V(1WWEn+-&kW1(bVk&q$F=~GGO
zSyf>jr(^eikGm3Bq;DUX^t+Dp^Ik^WkO!bVpL(N$VLA7y;jGj<>5`ccsaJ_UPS<7~
zhGOK@m@5%o!%4aAZhaq71-N-!>g;T7)pj;AZ_UwOzU4~;b@HT9MSW5K^11NcxD+X@
zS6?%+XN;8DS2nM0%NqzCGf~tpC@RuE&c`96K5Q>)+`tgaO3xzJlOEf?KW}KXgs3z$
zXyW0aG#QFIQuJ!ve$4?{4nUZ(GY7;BoC1y~7d>ubUP3?O+%-bLbYNccuF-UWVzz_4
zc9`=zvE|A|E)@GzI>g46Q}s|7VLAp@4W44r<WtS4F`k#5F}Ch)ZpO2x4H}{@Hjx7x
zt91g4_zq1fkkl!m&`ruOPQQF;53<w_`smJF@>LL2c?7|IR>+uVI}G@<(uU!H9HQF3
z&60X8c)hPhdbP$notsE@cumQOXb5ennr-ODMz*0FJFcO2PD3}^MGYZ{LK}K3+tAIN
zhHegS=w@L<&~==&;d%t4fEh(6yhnS(qQYe&i_Jz3HXFo7$g9_A$|b0AGhmQz>-J-4
zrg#`P+%UFN!x+s}GaEP=j72V7h#*cK0j_;!nsef(#k}q5zLdhbj~ep{V?IhM&+@vN
z(%ji+4)8g=G27P&Q^N6jWB@CdTJ0^fV+C3GD8=iWo3&0}1y;5gsgQ?`nS8wFAbEF|
z8li06LTKHg1+aQ~ehC&d#KAV~b<K<jXFvUg)2B~OpE-3cSvwC^L3Mwu{^%YJRTy8!
zm0<Kw$#WrUwJb)<CMceYHPZ2jI|#Wk_6$0p7cnYhD;(nidlE3%beDQ>b(wryHw(WY
z@TkKmxj$qVHF-`g*}k(YntQ>Q?<&#4rri7NUVRpt=Yw!CK$zZ3jpS9Q1gKYF4qjFj
zC2^%k#4T(skYKMm?wm?D$3DhQ=J(pvoS9X=>dEAXjO@(oXX@Yw8?hfT<vjt$6C|=c
zNN`nkSAuecBaAB+Y<vipU>%Z3MgbmYSqZJS#j&I(!RbP}g#upL#~QqUxL+>YwbtaK
zHPg_kX(oLEZ61Y09-%ff3nSYM7Gi18&Y9&Po2~_xDQ>-%wa%7lukFh?_8Cv6r$sHs
zi7p&V#INaT_tFmhGN@{GGpOAc{ak2tl9GceF_{M=Th!?enFX&|uY>fdvAyT&C-H6R
zT*fEBjez;h68C~5i(55sw3TyNf}boq@#&~_g)E%MNF#FaW9yq`d^^B)0h{#ewwnVt
zR841h^FAp?w;+#5uV_eZo%XK?AItP^v2W)h|Iibdjpg(WES<C9xrtD?)UY~~MJOWX
zZKKmKzl+#0@HYe0Ya5#L&vhCuj_w^QgkX8I%_&Pj{}z^QNM5+Zuni@Pxb>wYrUi+^
z6Imo&J>VA#4qbU9;$WGMd@lZ|g=>j+@m9<1;u!Ol=5Jh6guWj5wL-w^1fajwAF&cx
z;e{+X*xk^De_B?4Kd*qElWATB%V{dDyuOC$r7Y{()`v9dRkJTZ1Ed2sP|Vmk(DE?3
zS|XwNvGZ{`#$I3I&JXI4GT^e(8r{+f8`+-^G$c1{wtW5^;@crL7f$bJU_5tAY@1ir
zGKiSw_{WWL1T>1yYkOshwqzH94;=NPu35BGe^myh8PoCoyPnHa1v5XPU<_d9V8jM!
z8$qR`jp}%^Y8~p<tq-993DQE=Wsx<JC(G)#;8yr%Ovod`rDac;yHDx{iwK9i6u4VN
zS1h+?rBb#ast<fcs1t$+AY1!&e22AqxG6VZC9gB;PjrHvI<{?u`=AK7Rd%Hsp)j*}
z%XVIjIQ550R74~={=rfoW~YM~V?z-r?2}!cz&PzMniiF=`t9<&5Z&-fCt?A`Lw%qH
zKqf%JDqtOMRrR3Rxjm$@G`=G=p6+2GGJCF{J!q2$?3>SvIu1W5viIRi3Rqi?Jo2li
zZ$!kg!?fqLk*K`rghts_`-PQRofj&it93ki{mL7-;)27Dfz{n)?$@>U_nQso(*<`@
zn7R@e5P>SI35#t*qN#{^Y6lMkfviPRP<SuUpm$Kpji76I+(uFelzVKY+xx2GG9?HI
zBdd;auDDf)e58({n<J^Xqt^AbRNN!tlOtMj^9EIgYk_Eej4JNP_&Jsdgtz<peyX^X
zU<XtQkU5lIob&|oIxHl`&1j~>_A~Pw>RDfRqFg^I9p*$Hx<?Q^IAnTD*_9GRT}noV
zKqON#d1&Kr5=kk}<h{;z+0<e1FjL>xs4^)K)JKys8Tk^)mZ&wEZlO$3IpXDsfsSPf
zkLtibh!B#<moDOuV$epu=*py8@1PBZhjG4-L)&L83isrbJz1c=htGHh7qNto>=|@a
zBi&g+v|8nPB*j0ki-U}}`j&JU(_03Svma%w^Ku1xgQ6(DOO<k+9G7j!JCFm1k`s&_
z=S%grx=sQs-Iu(=Zy)9Q8yp=hmS%nOQ!1+m6qOk!V71P;zK23CY$za4o^BhrM(I_E
zy_>~O#_`KgHF~iB4U^vOeY>sS4Q%}~`nKCaybQKAI?xAbaSI0Ek*VbT3jS~I_3JA4
z&3ZU^H`D+%0;oo?Y1$aNYx-`%Ua4``<!>xnt|DE7tx@Bn8!io=%=Y3F*!ml7U(cz2
z#_ca6_h{2TZL$rV^;x~(5g9gg^yOLI@PB&M4Vn8h;tc#}#jw}pz=--Ve~S;iX<vJR
zmhIZ3PR5~q<BZep(ah)o|3(J_@pg+LJ~BJaLNv{VNFF$GBGc9OeBjm;Q=g-uLeU`R
zLalM5*0`x60bg~AYN)P^3SizfaBx<51zv7ljJ!KN742YM+0h(zUbmd5gJ-qdXF5H7
z3Wtff&$P-r!?4SQK*pR+rse;?y>E|^^SbWiN`x&_isMMs%Zi;(vq?A<$l>maQWS=(
znvxic9S%*Cw3UEnN3%QM?#}K!mh)J0E3-|C00s*60fZ?Kpy)pZ`asd5e-|kFZy!jI
zB1Ma~KnvSVg1Y{rK!GAf(6qnbIrsTJW@eWwIfAvY#reMP-gD1A_uO;NJ@?#mGn-6}
zF!PcmvJJPIyofga#X45p?57y>Qs7y)N+ej)IALzT*&ewM6z*>DJYiuj@{9fRCEOSU
zsg|c`mjGg@t~`6gxHl0gvs;`GLX6FUJ03O?Y>zvw8uXjgDWc2)K*Y$#03DZu!vk?v
zjMOasYezC}!jI+=G*P1BHM*;bl^h~{yS#J`S6K*m8P~8%xH#}jzKi<D`*-U2|2MT5
z!qxGhzazhpd8NkW-5!L1{zOog{fW>k`!f?@MXSJmY6q3uLG8AS+|uYOYz%nkc@$N3
z>DJd8SKe+S@GoO5NUxmFy6B_#vu7*`j+<H_b_y31I4S|>8*5~coDoO0n7rrwf8bf1
zACU0)HJl2-4T&19XZI%|t!-$Lo`=xxnIIxt1woDUF&6BrdT*?aT)K8TLGYEa(|FgX
z`6*<N7Bj=ay>nnjQ5aD`)LkywY}<3iK^Z1?tzCoigHXt>g~He<vn<ROX{ZF8dqoxX
zC=8#HqvmJj>{GjC?sV2j1$2(lNV<`JH&`?{>k0%ATV_U>kwOSWkpnPi_)s~3@~|MR
z5X>T1EUqcND72)MylqE`cV>_Rf}o5R;7B)(=MGRp+?;UYJex4t_@*g+U2|AYjVNmG
ziNBz&_8OkVevRURuG#HTBVtN7N!vx6t8>?8V0dA)09Jyy?zN|<eQs&U&TeNH^y<qd
zE01JQsEpuj(d(Br)&}tsW{Mjzj0kxvoL8&M%dehER$hCJr#`T8(3F6uQzq2Q7;4x@
zm2e=~R}yi~AP@txhd?9obCXoi5h<K$^mGS^xzkCq1EP?z*D~qiwq~3MzsIPh$*%5h
zU_%G7$tI#Gy^k-r-MG82AHh~u_F(JO26cHKhOwqFbnh?(uY3+cSCz3YWxKJY35PL*
zQ{a@(X2NN52_M?&k8Gk7YVJ?f+N%C@#u)7S2&R+QhGX@mdGH3Mo<ueSx+BB4hoL-B
z2$?nQ<@g5K<R0!CJQ^!sR>x|J%^ON~U>iRl0prO-g6G}g{(Ig?%|i=q-XiNhTy)An
z*ip_md|*&9ep4;}>4*pV%-j2i#C(2OG5VxSDVDjJ>d`!O0FB>>o^M<}1{}gDJeX?g
zQs+S<Sv=S{{q=Td59w3or~`~~YxhYnd<Y$^P8&jJWUHQ8vY?!m=plH48#(T*;pzkH
zuxN-{<=Cm;s<}QH4k2GH0Ao|Bx1Pn=MTWFWbP*PKa47M3^HxXojur`Jj-)TbAw0SO
zBo1h^vj<J68qh=)vDa1w^8BL`2P%*!J(rh%!B1&BNnCJ;T~$OpkdG+!5b|7WVHSd7
zhw|XUsooQsxlI?p>jxT@@Ms=~sm&<ombc?OM=(?Wrbn9=;miFa#$Pbrk~?>1@hTw-
zIFGY>7I((oa@3dL2P`8#u@PuuU1|WDdgoj~k{?)j?--zQn5eTX`0SuE@+{LXRO|HC
zyhUFUaK`aS;oWPlwT8n^I>dDsn8!DwbICqX&y6J+33J1PKk`_$qmEDur)57HFFz5B
zyyD@M5nC8*=^uS8+L2Qy8riVV!SBFfm*q{Tvkw-0mSN{$SdKE_Y|f~wqBUIBDXqtQ
zoj!Ww+%LYaOQ$v@y1MRhkPBH6d0@iFek~@Kx+AQS<QPS$(Oy6A3t>LdB~;ZlfsIhK
zhQC+f9FhFAbpxnj41Onv$uO*{7IoxRLT&~ZM5M+sWpIf^foo?q49#gc4?C~NVo2pR
z5xi$~N_)@s%56;e12C6@BQQ8cV-KWcHY5!F;<GRA*b8a7NJGWb%Xnl$c`?iQcV}gE
zfU)7u3+4jcVXko*Zja|=_{Z9N7;^J-CxNhkZFvQ&F@Hq<{8GY06k;4=rr@*shqVuc
zuMN{Bxjq;|HIF0aD&(t}t%?XSZAjgy`0XO&<A6#j-`X(|mU04#;h8<nKSmb>XMm<M
z_qelzMVOf(Iyw{RFcy&GJ5NbHcW*!BOh=zipr})x+*&T4N7L&bgl7ZpENbGA*oPRD
z>xe?*_t4#)fBZf5$WkSL2WJ8}9)q|v8WHV*318zP$ey%l<xW~SIY-e*vse~QAqU*b
zc@+<+kq)GdWKHE3SW8%7t-Qv}GMt!x14z0NuurLK3l?ijD{xpYFRh%<Lz4^{4jlOI
zD3oB^C>s)$BLKpd9(N&$3-+JIY3e0)7Q@g*=chh-YKm?SR2xrc>|%$d3RKT(*g%06
zT}-Y|#*(Ux@_jT}LoXy!J!x$)B^BQS0Q|F9$tTs|c=%chJ0nH9>L7oOhTFTIykkzl
zim&dSG9zIzK@t_59TY6u5-WFCxf|<pm^zXtOZ?X+O%~#xMzTdlwo%_UT*5(iEr$Ru
zMTbUHcZ1c_O!Z=CCR)Rq$5ZK@jwcaF7nver5gWw98RX^|9Z<p!quMPjn|v8~{{8f2
zYI>M*$}Cf6>|KZC434FJAr6kzo^&eaHpv^~)>Et2neUpJ!C3|iKTBhtwBkJH8!1@r
zp{#Z0LqVz45TQX-+M(ma>_6c&=Uzv+v$;FycGmY-EApa=1c~F!ZSDp}O!!P3g6Xox
z*t<4pY0uwItlJ2Yf0i6M-K?MWzr|gYMNr@na~$axFQIeCn`j~E@b~<Ev9*BMOuDfw
z1W^I)=ITIhylvb3_h1+oj{xBm5_oFMtOb#R^WQU-{`BCH*~8-fT?b(|K@k6Wq#>G4
zf{&s5^9`^JEak@)3XD@w1wIp^WE9cNhoZPl@*ba>H0q;e1QpM4AA8|Ug4)@7WWoy*
z11H{lo%JTpf>i^_61Q8-xfYwLvZJ(x`3&S357M9*7k(*kDxxuq=Kqd@nLM(L-oWNC
zQ|00U5uC9@?i+`M4N`#UG8+t3;Tq6{umo4m#4;2W3rg9ziK1eNlyJ~6T284-RD;<h
zkJY0+9$SFgn!?IOhaSbg^-=6ue+>K8Y62W$uUc;oWyE%KJZLi5pxQ??sfi#rxwK6#
zx4tHYlYefr)E!Ce(yWHTv_A76Sy|KYG4G#cY;Y5j6xLv{;(C!;S^tn#54rHZarxTY
zmz&ow-MDo1>gB7*1}E#02Vq7)tb@XgfVdn7mzHb|20O@l9(3*0J6}I_I`CT_1v0#1
z;V5u1HmD^TyhwV6pBrw`ej9e*cxu+Uq<&6v`(pC|TW>yzA!K^brZIHHUxX^COWHdw
zI^Y1>+)t{x2IK-A*q6)&sdAvl?7Mc?u`h8hJ{#jJ=U)jLn4hGQH1Q)H-bE}D=H<3h
zY>sYx%aO_DkDQzF5pi@F#-B!+d6CHx<`B4;9E~4g0?ac(`o~Z6U)ubajG!vde<{=X
z5yGH--pIQ?ZzSTI1+EQ{_Jux<T$VErW;rv*<JAtZXLxpqemls;*$z6u(Cw&PkdCt*
z<l}T<-Jkq4nmG}k3EmcRsDYCCa)?7PG$V{{ngyn>%W`ePAMnwfcB7*Kiqaa>qH`rf
z{5g6irSpjR)jXZS!PC112es~QNpS>6=86|*RFvX~$;=fm&Wk9;8N{3`Uv_RxYZT$%
zF6sbgjMgaX0;Y^QFq=X8Nz5IsQPhieQ8zG)q(-!CC<vW+3n2gnJuS}BC=eIzAo6rI
z0>I9SJ5aB_8+YJ8#U5c%kURXcN>C(g#d^tvhS~F+0$O%>XB9$^=vdFsCjmju_rYi;
z7it)chL{#J1^jZ4&Rcs_zy+kiP7DAA3EkL)ELO@7&#;L7Gl27sny*LMlpXfAUXf<8
z-}rBM*d}}*Ig|6P8VMepqX~}D6db~dK*Kge7UlyT$cgH^ctA7!7!0#9kw?^ffWwn`
zAvm`lUUk3nVBDddRrf1jNIAlK?xUxC;X|$>^MJIS#;W<(F{XKOEn*4f*4TSuc~<uP
z^O=!Jt2r69kV&U7=rramV$3N_d02+DrpQ4~hgt5i(#LM>F&lRwBL+-JLp~lu$8}}=
z=Qys6_aaU+!)~}RZt6&QZf<^~r$Q6J^F}DS=)MUV$!l~>$~|mDGI(rN)5Yn<KQ=NY
z8fz(GgeQ_9@ybZ9-7$DAxYg9DCU$@u3T!%9YmevI@9HJPG8dYz5XK7+O?=-8zT+|-
zZ134u^pfe-p<y)#TPKvMvZqiEQl*_-T-7@mPG#<G(o1~I9Y$m;?=2q51weAo&?)JX
zY2R){0_7dECyAdXcML8sxYX_9=9T_vfTYzuUUj3{ow$|CQ_Va!QwR^*%v@bAft<Y1
z2pLk>w`yzV8OMlDeN8SMGWc%nF264O<ZBztuV-L(Mu@xOwFlS_-Km97d-h2>6>$K@
zD1L^rUB(Sz=eXIsRU6$MVrLs1$GM3sbechIBHb_4_2e@7&Tn!?2tYOKpuC1jU~77J
zsG%&Y7>be@hR~J_LIu6bk?n_dO!uUL>^H;&Bfb<Pbl!ai4$_p0r79wZc|Yv3$A>V2
zzA+QUxU)GKOh)hw@+1Eo6i$@eD;MO(hw8Ft;EWv-5yT|ocEIE;n<5o~%S*p$P>r8C
zfI3R!Ug^E8Hzw#$FTFe*4u;iJSFux?tb~k@6Q$C9VPE8+je4bwrZg!0?<t0i1qTYO
z0de}2CtgTh<Us<L+Uaj6eL>HqUqVJ5IoYXKqKUPGYX(%-VGEMhPn9X_!tX1m#S>^5
zQ1c2iBQXDG_}aE>tzEFMF#mem>AFr~M!<i_qp>!001=QVsS`Hs?T+`WWgW7Bft$J9
z6O7K?FQlS3S;S@93%|c=E-4uIB)CLAd<5Vn7SRK5^2}=$;R-_WNb?Cf?pKZHub<vW
zHX`11gu@)o6$l4J+`O*CEY(olQpovN@lLMc5vh#p1ZXHF8i28hVZ&$Ro^M>fb?HK4
z+X`0(P(1v2+g-79ha)YwnAzJK6p5z47v(mA->JIE0T@tCMmr)p%xiJr1EJH~^=~4n
z=wCzOdB$@OE5OA)@Gc)aOQ8RJareOD?jbA=MdypV2Nrh^VewbOnc*89gizcAp}2<-
zO8NA2f-Okw)AsGydBdaQyok>(8+YuB#RN`KaV9A9ATo6wO0nAQ^bo28?pQ-?5hM?T
zXFA9Zjtjob4G^RsB7$1|Q+hWAFA3ZnX(Z5EN{a?+Rgbya%^F@ci&)f=w28$pgc$3<
zUV)C%`fgaTnQZSHT2=GNn)h~<5p+|COdW3T9?z8Z1c^*ITk4uxj=q$e6|3%7{1N89
z+zF*`m8;Dz7GlhVRU_^ssp1!Uf>u+D@+LLSSl2Ev9PJ%t629i<hi%4(n;*`ilzR~4
zYyKFd;}wqv5nEL&;{zp&xJEM%@`_r5Sy3RTppVQ>U9^S<^S34+WUiTfAV?Zh#9Ep1
z&Q1p_?L}_FhMJ*93@8k1-pb8RO$%Ci!qjwvhr|*SH6WhYsqu3}TzuDCT5ON9(ArK=
z8Srf<ip&eGA2Z;5R;A8z?ARlX;6ze8h(eN+^+q90QINTMq9y03XOlE#wI7B+35!tu
zs7!#6l#$^+q>em#eOHnUF01@7rx~IuJ@fL$F#`=L_fV=U{LEx||HKOZX)f(jtW&HA
z4fblSNR9|@Wupv*v+nfJ)F)|SVy0bavPRR2Q{>ZS2dPl-_sL#%AWN&>78zK~_T4qS
zdtH~Wi{j{(yDafi-_0vTVC{xRGwvvq)sVi+`RFTm)G%N6;vpn@p|R~ZVYBka8587f
z4^d>0k@%1qzz}KZWEQh6F>W#<D$*>zs8dvwMDBrt$UTH|$f+?FcMmM?9>U^I$+5V5
zU~%^l7B8{Hy-1u_Tw0}O4Irs!$YtM&u+6*+k?|^9YP<ePJ@*vVaGjDG?n!F6r?5uW
zGI2`<3~1I6!Vvpmvmfmm!Iu89T{IxAO*T9%yQN~KLiuKdr*9xGiq;qUxoCyaz6+ar
zcfZL%Rk6qDXI_W7K1!?ywg2_{H?JdaW!nDbNq@bnt`DxiaOq>a#&=x8!CfWu`73Az
z_q6uX@zy~1km<Fyp6d1SeazIIL%FIcDR`eS|FwSr1;;_O2*CC5-dk-p5Y)i4l_;7n
zB!gX|-RgpK#RI-JM1Bbzq8xXSQ;GYi7HrJ)meFxVvV~{={bjhr()%smKz=5<bhYzV
zN5aWvDXi;FT(Zg}N3ulIR_SrG4dAEFlLFWoa5zhQb{>!l&%t#lhjd)ji|Y@#U}L9Y
zqF?~5fT9JHcdw!V6c~%;)@}r0*PLLCS|svODYz5eZFRUFuri?XgarP#1fRKoT8^Gp
z@Wdfs6_o^U#AMX!S_eP`FG657dUEAFD+$WTWKP_W;b5KK2JFR<xoxl5VqC!t+Nwdr
zfwe3BX_LscA+xAG2t35ORzW_CU%P2`DO_X5{G^M7g?RtSsM+b`Tyc|g`Znp8-9X{p
z>f%(fm#edH35$0ak-eXey}y{(JTZEa2jA>H-Rw9ZFkoNr7_T+$we6T|>$vrN6SIc9
z7&yy*eP5)E)MgV7a~`J#lvdJ9d+WRVwd59VTG<=GnK<UP@d*7KPUH?a+{lFUW4Onq
z*&DP07&p1EZ%Ji3mw3=^M{qGX_{prf$z+ij!>9)2CiAsHb5J?(1e-<ZP9_&@$i{96
zX%KmaoyER}pBx$3e&HXH3Z_Kx$T1BbF)3z4E0{H47iD3{l<=_rs&HHQ#`C@5ZMLpA
zRD}P9y4mnMnkcd>s;0RCy#n$UE1_Cz7uWgZHPuloDwv5!r>e4wstH@!ivB`9mDg?D
z6Nk%nYAp`hp|u-vGPc@IffhP44c9D!{zy75!rZa;(dao4MdteL`&4;QRO+s`OLkp9
zVPXWkhobaRoZFkMeaCNTTIG<NF>|Rr0t;rid*J*on6IMjvU4?>xsFzT!dlId(QG;!
zEc$FLuJTdowtTx&wHt;Am$_GqG}~#WzIZj>DV?PZX+C_a3~RHLVWBGY{6THiE>@WH
z>&t1pe*hIp2ShlB{1YaTn+}Egl`jis*&}I1K_^s+`a4#PrFt<`d1Y$xsMc+uzuQPX
zSeOy%7Rq8i<92FYVl_vdzX%h}`Uqy|H=<Bn6JI{njNyXPQ`G5n$KK&+9DW5z-G=eN
z<UhP_ME#@Gj-Ep~QSI)nUGHVA*28N;Y34_w3v(?`)rPaw+pC|dH^$x_iFzy2ftlmx
znnygl2RZkO(Mn*;c)?b^o7t4Dg?}G~;)3#;xv0x7<*1HSRXu$0u)2Nqdke?5U;X2S
zWA(S|fB2n;+yBfx|DgW=@ccdd+<1JU@!SI5JpO%Cqw&Rs#((9Tue&$jTWEZrZ+^wS
z`ICjl<;U>mU%5B$KGyhkzWJ_u^Ba#feur<4|2IJX>kG$@HGc20#{a~d4}Yd)1!`CF
z4+ZtcADZG{E-7vtpr-WiKi2p@n|t-UXznxUdgFgRhX3KsX@|i7eys8Q<3PYej8uK5
z{U;yS(qD4zN2TND|IcHMCjff8!N#QdFFxM*BHvth&420f#wweC&H<ER%QU-UN<ABw
ziknLwZ=56WiEHjFk2k)8Hxms1hffBfTMxHiJAUlghc8Hke8E*`=KB}2=hDsl7fU1S
zO@XJvGu7tZlR+i(>Y31G<N6|~oeV#DV#qg7s>ow*fckNazSd&+v)N6WdU2!7XJT6O
z0}}Fq`8bP7qzR922%(ng^9unYk&SAa`iY1!a>xl!#T4*V%rKl_^^-w+8;~Okak72v
z7Z#3<AAY#l?IE1<UTre&bVs#rx3$;8B|6M7HIhIxxaig%R)6K!E*?A9>Fo{>n9d77
zWWOUFo!A)SLa`L_S-6&0KGJ)Mg!y=*-w+SR<Rj+mRonx5Cb>0%zu&o{Zl=sr0;k~3
z@qRb0va0bL!r*$Rvmsu|F>GB^3}>8Bu4b*Vn^&%V_3CBl5w>h%kZ-;BwfIHj`mOlI
z^{bbzaPK&%boJ_`#--+pe~HV6g09xwZ?oC+c0eRu9|NUJiS7iAd6C3Un(o5$ilFz9
zpb6Zr=<UAi!Dt^>KJ+f2UwX&HaL`TRk0PyZ<DI%edPw5MXgBlTeblddE)iKb`(sgR
zAQmo70GV)Kn5CPL7rl#L)Rca<mSBNGmBqzH-8TZL75mmjup1w+eP8)Tf2TjV*RQBA
zezeos#a*uV#}|^d0j|J-PqKv#YyHT*t<4mnvEZEDF~Wws8-bhHb`y2@rDmU<sy40)
z`yTEI+}=YUI0Fx>_Vw+hJ8;QMwzV;tE%sm7QrSjiUoQw_r%52<k0WwMfA&V3sAeIc
zozhBLQtIO;OX4ERE;2b+D;GOGjOCjZocynx7T?%zw%Lv3RIi8l##3~N2`Mbi9Vx8E
z5+DQAkaP}7Fk*~jrvmHQI<Vq+*o1-JG88poJ|GL!E?yBV4YC){>0{;eY-r0dG&<TU
zH&Vp+?WOHn7C?L$Y<8zC^}f}w;E1WEuL;Ozb+CvK%uNCY`xSD1%ocFETaM5v3GfzB
zM07}B92q4`7&3B<EJb6O1|Rb=|IK_4nYWB<S=9a+c<%U+d5*BOW!Wv9StY{g5HhNe
zY^Fk44_UW(T=*pm(Qh!zSq)1wl|Ti=hj@-Dfsx{Yh1<3!JsX((6g}PvY+E*xAsC|F
z5g0b=J4dJkCX7Z2gzby2N}+DNtO%qJo8TWf=+E?sp~Icz=aLL@i1|l1Xi%Z?p#4}(
zm!p~|jyva@A;uPo34U+bf>z(>OfjcGpjR#=8<mR}Z~p4dH{bkL{{y^u@U8yEi|5P>
z$Vk^R*+W1;<WtbaFm7MDcJtPqL?9%$&&g9<SH~OCBoV|%pJ~W?igZSCSBWeCj4Q^B
z0<_3RX9(o&sEoEGIE1_MIK_Ynk2FSEfjBTC0@0WV7d}vKlmW$A3*5j&^A@}C!#vy{
zv5Y|ThH#D=t2(^k5LM+WAy0d09}S$PjR1lynKBA2YDsPyArDz=D{l{2*hW!UVLk(s
za@j?A2394bzVz9)u#)RyIZz8Q9$Mvc{I}NSp5?9~WxDHoIo}kP6-La}hgWk^w;p}U
znZb;)vx5eP_!!I)%TF}19zmT2!EO1G))SrIY;`B*Jm*H`N*_r*5c^^lM;}<mQCsB!
zvf?^Yg9j!<-0PH%{lW!76lov%@O3dJym2Nr1p{kBV4I>NdhbxRdOI>n?wI!MqDa;S
zI(-3G5NP7Wm&Q2;ptT%LCU5s-*kZJqv3DYL)YLM|#ybIIa={Fd;21=jK!8$yh2?lj
zodRpEQEH;J?HBv6NZ8K6Itbp{AE%=$Uvq<Ci}7D6di}~(TUOR1g(aZ|v0UTo#r(t7
z<~jyVLYqJ~6ez67p0iq}Jl4b4TkT5md?9KaKRv6vGC~fSVOqtgH9I|^LGRDza4-Se
z;5yjBkXL0PAZ3}yrIwe@pU%*p(89lao%R@cwYJjE<`%WKfm`DSU||<*WpqXAQm$+5
za`bH-!}6w$so+HjY6EW`06y>3L@#9M5Jh*-(y68~XI0Z(`3EuNS>kyiY*EhstJjp(
zb`g~t41%n072C50Te-*=r=F+wlF*$TAnsF3=Os`UJB2rk2s&JL(2tS~Rl`STl{b_I
zJm7s)WRTf|0gfkwye)1*_^4#gy*(W4PMLHVW1r48r1UhHH&V`=DNt<><X^aKSiD+&
z^;Jw){HKIa(?@=a9?JLpLh?cWB^6WHL;Jhey5HQo2R{VvzTUb=!PdH8Rm_PuSO;Ii
z7llw-BOVBDVpR-5EHDTRnJolrVHGxvkRfduKTz$BB{-4JFD(_8howG*WFI%v_9n9d
zz)@KxY6cD(3z2YOZO=-`-o>p>8K7}PuS8&rY53lm<XvJ!xnC|;Ax__gukpQdA^Pr_
z<UPO%ae^Ns>~|q$rn8O#8iB#l=Ahq&K&3OTQ<jO@(-KOed*0kA@1wwVwv0R#6Los8
z)+XZBXFda?@H6gD%isLh3=G1?tF|64^cxJ3dSbP5{n}ScriNC<MS8EFhjGV?GSv#H
z;RT|c5#sE?)!>hoTyNFWr;%?$z7==N;S<onHQ}omm7&z~)+3TGOZ4}kiQw(=8ty`&
zO%@mcs0TcMoNT&7TWX(5zE33|XJhtH=G;LfpOq;yWzkeQ^yjZ%E)zsp1stsqB0!aB
z8>g^2vzd&B>#G%RPhJol&+Ya%e@TVi8Q2o8^#>K~MM7rMpM^{payhV6s3hm!^hZ%b
zib5c{g{x^+LoKmz4a?czn9iUh{_FO-h#~j7P9Y@tRPkPytqdK+C}h-@UA=NG6k;U;
zg!_^O7?mrbc_0+mM$Ox2m+$cZ+EYgde98F>cu=rF$g^Cyee5&ukQVB<Tb1>?{*FUz
z#dE2j<!AiWIviGS-^Rhg{W}^~k6+yVh<pH<$X#^?C|vy=1BRbQy(9)R`<N@`zURKI
zxKF`diR`l3+tknIA#jbfag9-yjD}mJN8C>qHbWA!0^Ttr^MM;GHp#z=k{5eorNowE
zLS|d4faX%j5+3E0^l&cIl_JixeZQ-xXCDqQFalji#NtpiK2i*WWXs@ZJOH-9N>Cy*
z4i6lO3TZY6y(+V4RSD=6v15#EWp;-v(>lwKB^`ah&=Ul2G?7?r(r-2QmN1j^)6q;r
zhIBBPV=W!uV~U_wf_v9H$aj})O?oXnBldv*;6k>I&U&ZS<=Q6gb~~FqOUE4y#BsFx
z<606qnk%<(tj~Q+^rrQ#w7a`8!TH*DI&O8kX<Lu?@%AoU>D}s&kWEZ(eB+rwX&&bV
z?@q7;Vn(Q1MM)?luv)FWOZ*sPaLM9OF&Ve)5UhH^6BK`s-Q}&hY=w!b+9n7G9yCV=
zoD@v&wRrBA1mNLX{|29ixNKrHm<&<n&)HVVf7Jk)%z)o;5aap=l>swZ5wmqr8@oLW
zrbzpp(N>D)TM)g7vk@7G8yQpP$l6Qs9MK%}oy)PL)q^{3%!FD@^@l~;1P2{)G;=iQ
zwuU(P*y)4tTl?)HBAGT*xsSTXQ_t+_T4#VudPbSiOe#=8r4+#FR=sdmK;bMi;gi4{
zj9yR*-s>PtDBZ)s(1~;eJ<Q<VW{1b*35)OgfLIMW1~o&4WD}_(lXrO+Aln#mAFFH>
zGfns*oHeKyvO9~F<yvxS4@}!y>wsJLFC<gPbFVWr9qk}cXbWeq<Pf=>It5R~{FK46
zfIX%VlBZj75pOMrFmCg`+y@?mJ}*e(yzSv2C1RFJ5tl>gI7mx&LC74esI&k;cHF?U
z{Y;0<%&=!{$<bXVSjDm4R%a-Db`K#6L*yv$b`h92fTLw|2<e0UPD}#~aFX^qnAMo|
znBPuli-hOUDWmn%O_>Ktisd#kzU67~Y7K&KG_B8Ow)Jp%r6i8)k9b855GCtE+F_0{
zBmQI;&e{o%32#AOK;w0IrE_lj2?3RIgIz{pNIC&T$BmGVuAEMJywBG#O5M(UyoMvT
zklWG&NCs3gLd$dw6qN*Kra*6i8JQtzcd#zF9t_FXYB-CV-tWvcAp#^fq*9DztKHhg
z+$Z6!G#+QM9Ca|IL=&h?<{9<k31#?C<b!s*vx!YHat@;UP4juGzg|mjVxl7D@qC0A
z&?8#|9U8#26RkoA7BT@3dltV8!Q=TKqKmzL6~y2)OcfQ1U?Wirb`rsb)8~D1(xcJ9
zXk}ooL&2Wu`o-u&wKZ0aF$BxO7`iv3*hv#AscH*_RwYQ1ch1D-#PZ?H30anL2o7RR
zJh~adW~W&Z-?(}Ch8PhyZ(g~1>(aGbR-Id$y|TA#3=eNs#B^P9BrB#X;=x>rRcz?H
zhn3aJ#d9OXNje%MKDpPDAl-6<ty1~b!9;kwg}okQQNagFCpWAsVTvuR7!gR~3@|9K
z?!6XbYt+c1HDIbjis4~*0BH-sVfl3}WmNO1wUO461kF)st91qwg9%F*lY-m#x|25b
z`#zNZuqUS7q>nr=6pz^8ZLLc@^aw(Nana+!{mwd!KPGm-jv*#5!`QlAh;a!ah8Tkx
zc^zC1C4e;A6UbySiS~&AdkB5NTo0o5$@M^f(&r%-hz_zAx&f&T0gknxm<ws`#EB~#
zFlwz8so5fkJ+BHD)JJFv9VafD&td~=QmvHr0q-2{19xorbRig!HG)PMHs7GmU|=a#
zgjTRa&~Z?Wz}E-EA%>&I%PN6AT4j<61`&&VFrz?y=rF@QZ`yM$i)1k9<^}_cDYV2N
z3^~B4YULkm2G3?-WvacEM`U9YwB@)L%qzT0t<2R(5i=v$ejb+2F*KsL`kNCtR%BTB
zIy41BZjR)spARNa4>qH*kie=03I#Sks$)1Aq)rY~mg0MnNZ2{wL&E5+oly}<B?}9I
zA!Q{KNDKF-3_Iy~Ukq$q4vTE@BN=zK(jDsA^~#^!fkL4ARGD6>T>N&Y2jPZQ;J4q@
zh2sgV8-o)9ZsF8YvKm2u^=jiRU3Iw7{Yp4~w?KX;oZ{7?jJTCwDs=KpAL7HIfyRTO
z7`}DTsSK58okT}}A8{@qS~Q#oqz<BAzMT=};C_`F)7Iaa>8wHX3fwCkF@X=JbtFJ5
zK$0X1^$d{v)H5JUcIQZVxAU^)6$-UzF`rc4?NAa7)0dNXA;xK4UlAoL0)9MbxAx%*
zC>W?mG`sZ3(%w>2&Pz<4r=_kOkzQs>KB6ZFI#N~`&|QZUvqvFV!xf>Jh#3(t-UhfB
zjBszm=P}+AIlh4v=Kvm%O|!*%Au-25VhHv$NLGeBR=G68=X9{Ss)v+<$;#tN`sjp~
z<$(MwKBPe|4mOdt2eHUGG4S{3hPgEViL%ToF_%A`hSoHz#YPxadOV3R>)`bJD5hbl
zDdjch!6x7wOEhx5pNxhNbTT45vW4Ke=v*lTHSHtkgkMH-et6con!u_hx_pSX=-#pg
zu8VD)n1tzokRrg*jh%{V1-ly&V6*zH>@AgkS<ZejnSCpEhoH28G2U%`^LR#MJ}P3|
z1mCQ^qmm04srgf0!s@oT#Gdbx&bI7leQR1VFkBbxvqi#=lp&k_TmS=~xcH7+GcRqI
zX*wE6_<808H(Rg?)@<?)@g}aiLM&~w*~3z?*)&mjyF*5U74ZlIbL>G;1TH=Nt&@w#
z<c~=Z{>Fbdi3jSRsXv8d_a`&n8j0vJ*#mxXyz#^1DDzCF43RQL{@{4yk67gSOpzwI
z7TI2#rr3Wx-uO=}_EN4`8>fAvqW}JQ<3F(IPv?pb)Aj^;7SsC<?C&3M{JY}Tk+Xfv
z75ly8jepBxFXlk%0@xM&o#Tz)Wx?li1>2pWEA>y0H~vj=Da04JQvdjP<DV4chCLWx
z>>nO){38~7A={*Qs{5Okf`9LL<L|TF&t}VE!%R-*$<g7a44<v@SB^LSZZQgS{?iuw
z%f}nP#$x$iW3$Tj=c~sX4~j~e_4`A~?@)iLkf~(RU`jk$e-dpyo#_TvKIZ$0`iX+?
zkfi4O=jxxs_h&Qp)zmRXo~S>8ED$gKz;gN*>i-#4e(LRDVOitmxH4z`Ul)8POPbIB
zqW&*Y`^!=7TuhoN`*-U99A%%6%KB|~Q>a(}+bHx>R46_oVG3^7cM75DA#hXZOZ6=j
zdM<+Ij*=UKyi|V)WllzAjM6bBUaY@}5;2KJ@V=POFVtVa=PyJ6@Y#HQzW#h31?vtr
zWuB`)hceGZ5Y&Ne3OrkX76qP-3Rt~sikz&UM3JYXBI2(y-=C>JlWRlW3Z}r*^{2rY
z_sJN{MpFMT^*_P$2k!a*Ha@$67k|mVXnby=aguL--Mx8kq46@`{Dyn8xX?JmH{Wq@
zY733m`R1>>H*YL7-sYRX=H6UhXk6!;zv<rGSZI8UZ+_Fg`3nn;4&VGO_hx6IvBNjN
z?cQ`38sEkn$#C-dcv3vvexFHB%BG5Z%+5oZXN6y6hRT<$$}qEAzGY>IlMeFrsc`5F
zdNE7b;ZhXB;(Es3_9^<}*+i<v5G|7_dQm-au)5X_)B^sNT1C@BK9tO;qNZ$e2h&tu
zMA%6!m8S~_DUyevu{@@Mnes}2#644?SMWiFrF{7OdT4$wnbzXB%jWG<Qr4I{=H1iw
z9jv<e%M<bi=8q})g`lLEgy#KE1n>7O0?$M*)e0*rZ&8cN8(YqlKUri`nQz)pXGQDV
z$@VujDa_MRQW&1i=^_WrcOE}>Z1LrDlhN?pTBm;wo8QUqK9bh=Pr!5^wBa;Zos2im
zzJbTLp>>9GWNBq-<$Q9J8Sb&gzs?PoHnOL$V<#X@uBNSFpT340=?Fiut^QT)d+c&E
zrkhCh5w|tcA@fF@FgawpgUwCUDM~N%4pbrX$oKcpNcI$yY35AwjXvDt$ePU4yhvo-
zAJJN3D2)Mm38`al4R#Ptf*XuXIYR-tPR$f+kPerIn-c~Mj9}Oj73^p-=wEQ6y?RyN
zRjt8`%|W?@7WcWW?=-gty#dV}1V~hcs!hU2iXHC3OHP-#4WuwR`=%sO(P&6GZFw~;
zHg>9T`D2@<g&3m%60R|ca6B#1f4(K|`}ikm7iUDi<=jP;%txMI;BFSysI7!p(x)Fc
zkqp+hCAdIWy9PrE<RmJ-gAB|=Vvw>XG`M2SKnOH5zBfReTq=Y%>u5o8Y)47f6OuGS
zPe5WsgPs-i(}$Bf7#O%{<ZEOQtT$4*);Dbjgq8DYU0uW=cl|Ezd_6-)escrZn&X*{
zimm}g(nip_*EOsucwa=biC?-dQopD`Fp+npAR;Spd4%Ch1%~)WETNVq4czZux9VSv
z)$M=qKkzJR;eL`vkZ97+D33nezCjK8Q#m%&50mZh>coFOp7>a_cfb-Y&i<c&!exNk
z><-rKbd+rF<x}e_*Nkru`WqcPWAC<IvZyhBoB5$P@8M!z<VNbzQGg0n{RfAD94DE>
z5?E3CcSu^*lY^b{_}JY_L}mr~f5yC)Q{iIx1MG5x?}n3(jhU_acxVM)iyI1_1g<GW
zw`=pRN+X~Xv=V*^YZwB-kpz*Zy^bpu6p;Ec$qGvO2|{U3`q#o%WFkjRIRcr|qP4Nj
z!}Bx4XsV!^moXV~me<O{H`pL(6!NCoZS9Nb4QN(Sz`;QGA)H&|@i3UxkVbKxjg)(o
zDfXDiH4?MS-RhWy*nyZp1|?HFB9f(HTtv_^6J0@>X~3X-S`+jX89AhbtXLyAEA`0k
z03H(&LS{Z>-U0(o(Hj3#&3scvrMRRJqoCw0%$4%o03|a{n>3sNR_|^gmQ|I3J>(>@
z;ZduV_1#Hu9E*$YsJ28gneOShVfm>NNAAO}`qF+rEpl3PHV{?@Hz1{@>?`9pf*RfM
zE?QN-+g|&n1ph(upmCFzR;0nvGSd_)Qga9fHQ!|Qbd6tv@}v?&CmAgf^~DtQL?Tas
z#aLJ)Ka!xWxEPD7%d1jS91$>t<Vm2I3U37H#wZ~y6!|LFBV}YbOscCHa*z>FwBuGl
zwhuH&+|ULG>|sQbLLRdkUxALk-|Q&vgirU~xezr2YBGyQ+|#vQb7gncxAwEAL@A&Y
zcjV&Q7Ai}Cb9_N!$A85QMb+nMu17mzK*MJ+jIcLvA)JGWP=Ga57H_bK>Y8CTU*^D~
z4>Ps^g2IuH9tgP83cE35S1TMdAm#^&bb>G%clr~bS|A?&+o-^EFC5@dhQ+1CIQirE
ztugXh`a`xCTK4&c`p@F|-5=WLpRcbNW94L#74UHTPqFU!@WoQmXbL2=cw5U&wtuLm
z&RdbGqmfZisf^wqBJNfqDVYxwE^nMePn#peY=OmkgFc8r;k&*CtLUW@$BxOmf~hsf
z$ad%!6-Zaf09m-dH3^0qO7<aGg}4I2JLP}L$JuM?1J*pc&a}6Uy*)q}4??qU3=ns+
zf0+}2YcO-&SifIu4BC@!YNjF9g7CttyHw87AkeIX_{-gn9{ffo`0<YY$<vjNA~H7!
zd9=E+bp8!yWf-Na%S-s<Nd;cdE~)K<`HmZp(jnxjYtZwdo0@X?!?g*fa=%@X5;T%m
ziV71eK#_lN)4A<T^2#eo6+OifbN+Ld7u_$u#j<*XlI(L&x-HkJ`_aMf)W$Hq3!uxj
zCDvEifTwB@X-g5Q$J`P^$(7SJxtJdPMkZ1v3^FY6V{qzPx;;!cv6zCDq^A$7726hO
zv;A2{GCN>E%l>pIPa#;YtRTV)$aK%M{7t!ZoM*C2ptx4NAz&y_0?KiRe7F6=32U+p
zKJsLg9{KX$_>M%&q05n=)>hIZ{Bktufo^BE&%W8&2HCLikD4hU-(up^$;zDFZGlZS
za2@xoV>t)+L=az%kG%;!7l^HcupA6w*D?Zou*Mi@u}4O<M<5v1!9gtNXoN}QkVWu#
zwJP}I;5xg3H8+@I6}REI@<I$(gPwNlypb<~Q6jy}F%V+t%>aTHXuO6O)zi0WxvDWM
zm%d%NBdjSz#?*%kglyQ4j0UQK;x^!Mh}*!6gSR1FrZj>DA@{$ru`>w{D;h@33X{d;
zK91C_Ap(*2z{7+G!d(1Nf8wR7#dImKotgH+=S0G_YP{s^{m#bw$r{Ym9WDwd{cdN6
zhl4PbaBT2^NDD^<hP(@HZPJ0e1N}M&`f<?Tw<7HQT2jXjpNu-z)CewTtW&a$1uM(p
z8brj>l6L%qvNdSI=^@0-PqbL^mJ|X<7(fm-;U9Pg-k)#Q2^P7j#}*Gi{a8Cf{^Oj6
zY<$kC$eCfu<mV*;2vr_w#sTikIal7&LFV1OEx8dfYXpB1(Ba`lin%@Of}3i!TFpz6
zRoA8i>R@=(4-5mIQ%VVwG}*Y=#VrCORSX{_kFo}fQ}@OQ3BA12Y^j=ffHD>oCY>?f
zswFm-H=9-qa9H)&h$;L>_KJY|XX`cW>-|g#RG-*hoOoZB!mpGS&P$YNbCD-ni_F8C
zML66^{P5+n(yl9}xNdrv0=LrAZ0SBx|7n=|FaCiw^_S|)#?<GgPZ4?#x6jd0IS)Qk
zuoM&DO(}OEGD?lvqa?*%n!=#knaTE_>Wbs@X2oGQtg9bB*R)r9GBhTsEY+53zgRf|
puG#+WYap8J?Jm@@EjoJe5cfX#(hs*!f=;9Dt8Xxi|75aO`~QC`f589%

literal 0
HcmV?d00001

diff --git a/.doctrees/examples/DATASETS.doctree b/.doctrees/examples/DATASETS.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..13e999781b1e9c6366d500af256da24520a45b9e
GIT binary patch
literal 99117
zcmeHw4U`;Lb*5y^zy2jl2FsGmV{DJ`Opi6PFbE-ANV2U+nh}~YmLpHPYr1Qus@2`q
z?yA;kY-#yVEGQsAlMo;=KM8@81PBR`kYuw-NV2;*$=U3(fdmKy=Z7RqHk|B9NOrTk
z_r7|6Rb5s6s;7pr$@sviyI#Hb?)&b2_uYHneXn2J`|b_v*000=ysb{%v}#v!dZ|>i
zONP_(21<6ZSv6{I>s_t=uWMavjd+6%ebKR-jiS-=*1;P^)2fsjMy)m5^0wgXu2pez
zw0^&16kW@%@#<!;*=P38wnofBuh+HQiXjzw{U>x+cMP|MufQ8UGsYF{S`@sg<x|mD
z*P%~%al1E|=R}$tygo|Xq}NliYDR0aIj4I=y6ZNqxuy$a!ZB>HN)FQ7U##ek0}tVI
zZ%wcIp9V_Jy7)4u7cVu8d3>ko^}G547PZW+=DySG&Ar|O<EC9T#v9FA&1j4-+Ko%&
zXU;xYu@~7tK%6t~6E$vJ(W~`};f$X+HgjzH<ji!gT56el%uVJk=7#1RjCkW*vr;kK
ze8qAE<_t6_AH;&Z<r-B9Hv#D&z@jfS^tve(dpj>uTxUmaFZl@X_aR>AZK@l!l2u#C
zUx7hNmHalL$3Ck@fMf!|iP_!86&T409ErJi(%gr@o3{em=54^s+kxd?59!0}sV>3b
z4D$|nei!_IKOk$~ZN3t!-yg0dzvT4+i2!HI+ujCb3*-_DtGZQdO?v(C#a*fcpQcEG
zt2Y)vth|0Y{Fb*7-Y~AzD^}5RTTf&EuYmqv7EiuR3BG<G2mp-K9EEE4QvT)qIl>?s
zF%JeGm@j4Dz55tXv<K}*Rd+S3rafeU4CxiYrLlQZHfGhv7W^{Gx1+2b%BJI?@4LMh
zQkJ(j-=cE}=yu|VJ7P|nQ;|HXEAhzuel?D~kaDDbsBt{1#1kFGqaMm5%jZ#0Cc_mb
zfT^s)JY{N#ha(abUv+W1S+Cm-*C=TWZ@mVe4Oh1+@X+V20~~eOi(_oS@ZwmBzDP&!
zD6{uYwT>lTe<&WdcwRq~DX;HT10jy_?~5Nnn2$I)dnx1JL;2Wga>|U|d-1qkyKFQZ
z9ak@4Us3w1%%n|nBN8kAAbue6toXf5S+PkTP#oaD8&78x;C&SE&np5hO{6_U8`9}Q
z^&g!R%4l!I8AzLu!2Xx`LBxaoPnm+f5vM#IP8`<%5Km?l*8LRgzxT23tkqbBZ&Bui
zw3@_=@qp}WGX|I}x@&|c!5Ekz<he3`cFD7WH!C<*-VbJIJPNVQD&Gn9GUeAUd2;Z9
z#&E7YoJe*Q^9Fq8m2agrlZZ!C16L(l$L!3%;}?%N9oMc}Pa0Yz<j6_(##rP~m3ew=
zhpr^<UPy!~p1Y4^&D~o&3_gy}uSy(16rTqvpC?oBnbu)^HUcKJ+=<ucy9zCD{TQ*I
z$3RNhET<|szBUsprn7&eqgvQ?5uLHAT4f33$AynJd{h>c0ajU(h<7If8ZQy=%vvG_
z*l^-l|F%R*Wv(#KV2HB*Eh$(}>zJK+aTDRi1Y$~=F@Zx+=Efk_B-JH$d_HlI@$C4G
ztl2S$V<7{I<HxThk{ZR2jg%jsOTiCXM}Z%AUp%K<4Y2oReG8nalcxb4tIVUF5lu?m
z`iDda;<@$rS#xV=#31AN_CFGdj^f)U%D2Bx!8clm@hx5nmfrxamfwhoz$%OSlc_`M
zGsIVA`ja^s)LD$L(B2pGGimT3zAEdBMH|LD*?jGNv8pW4B;pW{J!)w6hJD#88Jc5&
z%cEB`{W5s3`dr1(sITg1&05K5Xa)RU$Z3z+O|7agX(fBHR<ZSx=9)%T(<>EiZb`#X
z7zC6KlL*s@x!W5}(}g7BFS#LT4?-}izXG8zAW;fbc|Ay!AI{oMzwii0`GHUyiH36d
zD89Y#^cE^~i~xwt?e^|c9k%G4)o1arWEP{(N_&<9f}81x^K%uucu5}j9mNvik9Oys
zT5g=NBIlTd5%3Vo$A%a%if{YGn|sRN4^TLv2%bDay@85eTWIPF5C*=%F(LkTrCK@c
z)OCyt57ZlmV>B)sXqc2QAgUNB+jn{^1fAMt&_;PG>NTxlfJ;#{v;v4yL0hz37>nlA
zjiNPg6`3$<7=YDWTSEgx7C{K2QF+xyjNC#_D}-SI>h!|?9GA>(;f^KgtAr${d_O#3
zVK#3AP{^Pqp9R<5P(>w?+d)N+^-4tUmCJ`fc|bQtVZ}xmT_}e9gP9Ha2H=ipm!Y9P
zfhdmcLUv&Q)I9fA{Fwtg4UPu;2%6(I?AiiIlTj<e)&>Z5X;gEp8i=B6T8_2|x~>TV
z?GP!|>-Eag5U4z>p@9NK^;xaEP<H_~9E=9&KFu|XX3ct{X*ik<I#s)@H!Qp9X#PF`
z6H_Or9don_bUmays6#bB%8TJdCdu=51vV2%n^5k@gS9yWo{eISB|_1hq3Ok<QOAn9
zgQP(?x!!a&dtQU_XmzV@U=ZKY4x7ydN2uo086qO_?I%L=0P1u7P;&X%Oz}+aD~1-L
zp&*g~)Sk(bAIl}?w$A(%Qzi_3=QZa>U-j@Qvymw|wym^{JxJPAAUk$y+YaU54T(ph
zh9o;}H&gZad0G>gButn$&e(L3SR#|i7k!v2S`~INE809rzT2!BqtYB^CXcS+h8oV3
zLayxCwZhOm1n!x=!xtZELYBbLFS>SPDTlE}jqXH^@-^jTCe%nW^WQ^g1s9@TLN-&>
z^qUZeuZAioLe1=-TPWJBXEG%MlZKTr#SByCzHWgN#D-h#EW3<SAs6MxV8mQtAGtcF
z)pgg*yLO&x)FDlShxjFbfM6#*n2;XmH5=w#gH)3q=W;o2Q>{?zJ|JL&{WA<DjSW^V
zLxu7s_!Ba_ZiU7KMAFo`A4S-z%^R?pUW5s<XlTV|19#S8F(mdC-LgSD3>G193Fth5
zSyeHFB{h-0AxXpsHv}#HR3g4TR9Q*H@1i2kdL|;?M+YS&k`*&K?<|q5;Yhe2KdBc@
z4fPD^Npno}nN0mM+8MaIv$Y<S7w!p<xMst^wcK%a;13ugXcEukl77ipa)#hFrhxVf
z4K$f;FA|^jhoBCx>k;LfgER_g!|!&6c7NL<5SZ*2Ft+m)%;sU-`5!*44bMDw?&R<&
zKH&($<BRY>)9^+Azv0s-AIl%W_g{VN%p)g3D;yw|q|ynwbRt}OP%b?fF1=SSy_c3=
zL#mR9m`EgIB9}zOL@tSliJUSK2NQ`nm`KFIL?R9biI^R_#+2$B7pWLNPr1k!ulqq)
zs4iYTs1jST%NL~BL@!^zvIr}*uR9Ak*$bd5OI8tfD}8s0IZ>z(G)&AW8E6dph2Z_6
z-A5E@;VF%1fl>0$#Ha?#ymqZX>iCXH(S~KwsoNx33c7(X6a+b3k{<wm+zhC1YDo_9
zEC&ty^*u~1BhDFB@{U+Khr}QxRXY_@wZSE^2S>|{KL3l*tdEAODlLc)Qsv6}B@5zx
zUE-NAj_{Jf`eQ3>aGbj$`N9SsQ!0?CW`@L}u0)|L93(eXB+Mj>k{EFq&d#eUW4tg}
zL>(_7J4?r$K8y(RmsoTHGTCUhT1%G`{DvR^NF6y9XHzxBiF%%@uM4c+lSVp-8~S~8
zl0eEV$y9(zlHe@|(|;(L!H0sFkEJH&3?=3xO2h=RV<;_wXM&W^r6y&PlJZ$Wig3>x
z@}?N<E}&8-QMaO9tr8ytLTe?^G@3K-VlWL1UC^koX&_t?=7q@F3alxpIBh}{@1b)u
zR093@7HFE)=vv_TA5-Co;s0~(I6}%V3>wl2AvIKr{<GK&<mWb;a}BE~m3h05X);hO
zmdMm*T(O}1t7#}Ne;ME2Q~nbE<A?bnh!!->e;djIrnyWXv56B*^kMnsyj4Pj`uNo3
zt52SvJ~ne|YBGQ71UI^e9mp*<s`$$EW79Jy&*smapE`GTM!ewKcEy46i^DThQ)lug
zPEH>`f9f1*2kid#hhS;9KfEwJ>s#>PTR_&h0}*4=CU+`?dx3WTaKo+;j|79roD?j9
zGGs3}+(_~x(<jg8PfebCWCk1(-WGKh@a33hA&DLA3mqC^RN>3PwvSCupPHUIHaU|&
ze{%ZV)Z{dHI%)b)Cw(GKvt5!d_d$W0CI5yztM8)he6LAF5LWmmdueMEdTl~e0imjf
zskpTX{ajV?(G%KV^<n+~TUetra@eP-InbOFz65g@m`4Byi=2GPy4-r&+XiRNtc6-0
z3_@_Q@mLoeJZso>%xP$OeKs6XGD@xMUJp6aQ{K=6XVnHx-O2k0$e#9kELcSG*iX?k
zF2hzXeB6%9DzTdK(jl*Yr!-cb8#~z2z0IVuT)1h@E-gt9Fo^y~!l*e9gNih8eKVT)
z`rPsiJQIf1HXuDd#aYg;z@#5E318Q45_USsAjS7TRS4Nmg~M85s*Z;m^vcC!wWV3e
z<~df=Ap580LXrh+Rb#vsgVA!uoYvNk2HWDqc*pngT?=;*FreZ#;KUr;u$EgPF7%=W
z5pUR9fjw;8jw5zIcE`4;X<Q^22G}Y$0~R2&WZRw#q({;q%{IPXPr-f>txSR-HyqX^
zKb{pu$164muw`V1o(dulXGY|mYHU;<NE@=$Wo&yc26HB}!92Ed!X?NJGcsONv8l+U
z<2EiGiIFvuw$JM9<8>0w1U^#my_diwh1?XFOk|#FelbfpaZNln2z%8yjUXQiJb;GV
zK6i-}jsz!EpWhM_Q1(@z^F%kyd;^_6RWaDzUU2%bqUh<fjZYssYDQptfgtq_nUVT3
zm_%J-EXWCuu(Mn-*vcHX&T_@-D9ROadIMs01{}tV6lGtySRw>@7sLpJ3v;7nB`vGr
zNJZUCDl@LpZvvZ-cEhSSQmOh$4i7|jf>OnbqNQp#m#V;^8I^Ixg6z*_M)tiZVX1)@
zl!OP`S-2Q_ALQtD7A{sNUby75lGOr>fS`a8h)-BWHz-u>6%6yWTNh-*n!bX@1R8Gs
zCXKWuZW)ylOI-_UvjNsRW2tx`p!-@kOneiSy)Se0+GLLvMa$lhz=IJdKN949J2Ubg
zk!6e_vwYkIso@cK7D5K&mjsL{jM(!9^es`ie0>EbD{i;pLt<~jXFANBcFJW5kp`qn
zV9GkitqK|#sRizt79w?54>rh{<bM%k-r1YNaA)S^4^rvbN-Ia0ldLFOdiwkUGosj6
zLDn6aku{nIRLP7?9zYMG&&c<`2;%R{jQAepuQE}CTo4##(i1`6sm#b*FXuFe0qZ<_
zm>6y1V$@lMvpR#kPJ-Qe4|;bPgKnE+^m%xi5!^y1Wj}qFVamS?L~Gq3`a^UQmBmQH
z%RejXauU5ZGqU#jlPC@~fkc?%&O*RoSl}>p76MjB2tj;5?PrFV2nnSfUmIN{Jf+MO
zy|5hO>$LaCoFysBf;Zye!I%6KP1}|43ZfUEh;7D4VZ#vZRlI@+qYsc{M|TbJ8$a_~
zVrvCS{2(C-vXGDqe{$NniUAi8p(_Qs#E>&w@|Ox&^;fKYAr+raBZwqAbe&F)PpL^o
z`IGqGzSFFl1hfFMcgCtyI}3(QzX>A$Q)(jLLWz`m3$6ZP-6Wfr!;~de{Sxq{2^>cI
zk=r%EavpS3c~uPLtEmU#(SiI4M;nf$Z8d6eLskQFj!MzDHYMExv}c<5?EXoo8v0t0
zwC?XGa=}Y{M-1m5QV-{?bU1&{htpZ5usYx3b;7)pqK{`zF=HQcr;3nm)qq2UzKUew
zKFDaL@jwXO7B3OrlGj2@4$YSWfy}m1f!|PH3JZ@+zPC~Q`e{v-YaCb|CH#IG@Wa!!
z5NVzFWdn2Axb>oqSUDp{jsa;EIDG`E!5Ds(XNP0j%}NPlt#Gunwjb<c{{>i~`0X}u
zaH&Kyi?J@spx<VYM-YaplE&f6G&fpOpEvXhuHTlL>u;xA-<N{xtd0`b7tRJW9k<Q=
zL_!Rm!_6x$96zDF<S$?>kQ7C+5AO8SAcr5%XJV|-a;XmlGipn)L&=VQX$YO5+s!RJ
z8XM-{1-uW08c0+y%x(H%iZJV)xJ@sJ8-AALp3jWj@wChkn6Y6qYc}43Vu5J}*r606
z$?6D^jPJ)Az%`C+zIlCc!QdRE8+m<@{yJ}6X_<X+=VEgnKYHNk08Qn;#l>OYX|`5~
z<ubktW<atf_KSRdi5Ho61CrGimWJrj3*5pavKPr?nmE>;ugEgn)ROFCt~jFeBc@bv
z{+(&Wk0d0%lTMYlbIwc4@6*R6Ee~f;`C0&5c5u1`^O5Ha_zVY63<K|ggdh>92C{(Q
z!8wQ{k(H;0XH6H4yWo@xIoI93#{lPD{mY=p?xBd;k@I)LmV9%z1bySRKh`}-0`(81
zLLK*HTkWXhZKil*f|Q0wJru~$a~B$Rv))#Occ6mKtl$Cnw!-*(e+Hn+qE&Lu*6G`D
zw*3;)zo-0ZJ`9pTd>&Hpei<IIqi~cO`9e;4yBY3rUDZ|rie3;rz0?lGptqr6FSb1=
zH^@TFqOsIgB+Y@%?D{fol3d%0_sZ_}LQjP({1pc=>IPF8FkkoXDgQaI!C;S9Sjeq!
z;QM>ZUu&yYeij~uxFwvjEf0d73={UGe}gOwvV<Rtk^EoYf$B*9H?OgxBiYU`$7HNw
zgNY~+<Jr}p&OF~!-oa}ub39@V3FojNL%)i!?J56BO7y<uS8JcOWF}xHN}E`Zl>Zbe
z1w~{eS!{+1McBt%s8kOY2+BbN4(@LD3mFH*0|ERA4x@q~F%#}RR6{;MD|d2@?gIp0
z!8=ODufuKC5+S+pFq{%BTwFMNZ2I)9W7m&d^|!*W9xWW1YaA_*yKOLzq-%)I9GKfB
z@SQ4<gapgwlICCT3=5uK%^kUl&JwTrsA-c3)RNu2|HxJDrt$Jqnhuxbq;B?Mdr`B&
zi9@2|n=aY4yI{k10goX;pu}ueb=X2H+2AI?5m7wj4p$oJkY;1}SvrnA?=C`<aL<Je
zulUu;NmQWFhEyK7(E_|lY>;l*L0(VE@-v6vrl6W*<b0V5#er{(!j=hGbfFM7gEO-W
zVbi&=!5rEE*OW^$$m>DwGBW%xqT8l&w_cxbJhi+Y)2&uQXVx5S$4`NO4@k5iC^+x#
z?MBEL?cG%tV||nISW&yt8<E(2mPX`4J4_cPqx&x99yz+@H$t7@c|b*MZ`NX`Aww|f
z6@k--l&Q`Nr%x*Pm;p`)VA7{kM-09$Y6?_;U3rXIq5AvEJ!XLFJ}psOGYC7qCh+=-
zGWA*E^*72rW`NfsUwXc~D;HzLs`W28{LjjGCvrH^NQH3qp7Iayj#kpX@(x`*W7RIr
zj^xJqIcP}i7UAYRixt3cU);_N&`!5vdOg@E<C_0PK)81x$v23<3-e}PuQMX@uNin7
z$gfb~FHW!@mKfo`NYP=`{?(X%HArN}-%p`7IRAMI_~``s_bZ})sbb%0wj%GQV^FcM
zccvIG3LEuR+b!%wVPhWAeQolV-hUY&{!6tI<-PPFa$thd=o<Sf#NyR_xG9XuAETbj
z`4qTL@s2vnk%a4+wra_q%O}utIdavi=1HdS)uS|9nD{MtlcBQP0!_fC4|7c9Cefks
z9igBZ6<>q<Ies;Dt0KNhK}7p%8egl8kkxRB2$lj{BAg)yX@`t=Fb_MbNXYXdmzOT&
z+2QhBR=3>Wq}-#TE%0No4oQ>*uGBo#l?b!x-(_uqZcBNr(mKqvg)}C0bU&orV+QDM
zlcf$P7z3`oKw$Fo$`odW$*(K-m;olcqjelZCazfE^9Rb*W`)mJlzYqopIx%daP`XI
zi76H6{H-#rS)udK$~|U)&P4j9jzQN=Re|fkV3N5T+N59+oMmaM-O4>?fNQ%#q846z
zfxzSrWeT&BwS&q%W`N1f@rDqSsiao~POns^IxC#cDfgHGPBA)(ZNNwCGl5QCnbxe(
zsVeuF0XnZ`asKaau!W#bY+@#>t51mF`Bh~OWX1D0EBDwH&l8<<5(%pFWpX9`RqtTh
zNKm!1yg1M&5`m461N|(o*BK8v4pg21AOpWI+I1L+MK96?f!Lz{Q91yXAka^x7%WOW
z4Fa_XD-z`_%OKEDp=20!$1pBMg@FcGAyY2mftcif#Rr|jb$%}m1wESr-LLbGI*XEo
z?pNEYC5M8ZB%z@H`D^c=GvG%DzVRckv*+Ot?8?F^o}vNIkEU>@N#KSAksblR@Ni%Q
zO_UvDcT63In7!i85fB1;KZ-oRr?6EM)7z>9a^h)^-1dI>NJiXXVw_!rVj2t#b>TU%
zY0yI}Lt>eFjAOTQkBV#Edm)0<6(x79FeQOl){gd>#P}LjBsg@hGL@1wvMdh0Qn|;K
z=FqEFn34d8O5Y8A#(4hS=}bsbK^e)cq^PFc<4Q}>wH2l$AVu#FNYQg_onq4W+$K^;
zt*6YnM5#@*ONsxpo_Dm8Mq0X^aAw6W&jb1-A6b;08}9-A9Iw|IkrncQgxQ|X2V#rg
z$LJ7Le4vk{7;d7Ko+xWs`au1K1*2Nkqa2`KQo<%X{U&DBH~CnjL?R|y{66aX{CNsc
z|BQFkS%xG~|Fo@Ivg`AH;`+?lP8t{I2-z{u+w*z1*{B`G{V#vJI<Nx-z6yWEFAmis
z=UMy`Wgn<xqk@>QJ9MOD-2g$Wvi!7>NCYvo<$e6f^7+LCJd?J>^cdXOg-gVyM=!0M
z$|br<xkp7)>3!f5Nt6T(lo!SdIKwmH4S~{~$}~$$^RiI7SGmUwP`WW{o-=@=3k5dE
zl_|{%n^!6Km;pA2y3%wEye?}CbPLL3%?jO`a*r9HyE7ztxEL_(BFh9a*OckZ3YoVk
z_m}}PdlOefhEa#N1X}M^raddPKC0Yf253bqDZ1|&T_~{mtTLrpVe{L{J!XK-8&+xC
zQ#Os5ygnyr5uE>mG9R+y{5O<)?27Y=4jFO&Z{P&1q?_VRq}lu=9^CRg|Nc!$c5J-o
z-^1&5hGT_1f9kTPbNiWbxsKAW;`aY*R;*8yvTWUcbE!HP?e>!rrdN9XZ(?TO&-t3d
z?SDUY`(K^{)R*#(Itvgg5xSpzsI6ME+y5H)RolW5;|iQYg$v;G_Z*!jKlI}}<7qcL
z3rFyt;-lJA!+7l*{_+m{;r09zjUwEAF+8db!;caa8y2}MAK!QgeuMyT`GP~VuR0G$
zWJ~1N*x}oP1y?J<f#56f>pR+T9qvb|8u06Y_<KN4@t-2K=gM#c7I^`F*uZGG7Wu&-
zcmeClZ~9k$Mi4LqmowqpczqB4PCXDnPv%m3hOb=%0vg5>P56mw{FS7Oe(ST>u4zZc
z$%qMDWUE&|dEh8+R{226-lG3Revb#c{UTMzpuHi{Vc5uPf48mERS9#9J4Yk8eRl|a
zm`JiI8C4|aXMxLq7xVLRK0i}z#nzO2R9t!A4{-;Hl0e+y?i5p!0hX$cK=PXMxFj2f
zX*!91^-b-!DEF8FlHd8as;ih-{M!XU$oPAeB`zx&|EO}0-O0H6Z2RmAQxecy+k`&%
z*F`3U?FE86zoSfHR^0i5a*y3{CxFQ&*(zXo1Rn}4o>itSD=faP++zk<?2_e!n<fmD
zm{Nhxy3I*mTFhqLvev3C$~|U)&WNf>$N*EXATZpkJff^5>~7^AGr+K2(eUFh-20)b
zNnmnVnZm3vd04r}EHDYX2~2*rwG{|V9#y6=D@+XK9y7osM%VD8E|RbaJtfd_m1)fi
zo!2Y(m;pLZu2Rew{6r?3t4V_3^E;GLPvmoA#EZmLALGmLO2&cG#Z^gAY<Y3jPbcCR
zA6NYpuh$ulm5QtSF~M|kRc35_oYJonSN)k3V?`;84FM)fS(b6tjRm;s)@al!)rqK-
zszF}LCbRTJ%&l+nai@sk{5XxV{&fm~|B`pqS(qe%pKYs_9Ao_yNv(2@%Eu(v?ii8L
z(Agmv76DzAU?g4v1DlKV{uUylp@gelN20jMggK^E2=<mOUAQSsu!m^nR7p{{DEFvn
z9Q-&0CnZV(4sDlrg;3L~%i03nyOqZ(Ey>G5_hrgGW`J({F@5g9NOu>JL>Ga{!^#w9
zg~_Xxd&~fnxT{1cGI7NMA48eitnhJ^d&~fzWuDuAy)vy?q4UGaJ!XJTH`j$w$?B%6
zkh*7-N0}9_Kd0Pd2Do;mseV<N!mKd)&&oY!fJxFdB2=c5UJ<hM$I4V^h0|Xt_m}}r
z%hXu^s7z~C==5w=wdpcIXZ0`mpt8F9goqijQ<(!<@%%RB9=qasqPtFlLj7FKSJH9y
zUXmu+j{Ek@3k*G&h;4je=s2&}8IlzW3`xO^bipC!gxpWZpb{K9l486l_cS=v9=k}C
zwk(4~1BFUOuj-LMnbGGLF;AL(6_fnm@!_V33H=0(2|bwt*G1k@XE~B^b=#^X$As=7
zd3RUIJtO3B9PBGxg$R*<pbmer$7n!89=nY8$W_dGmL8KpN%|XV4cjHHMs$RpV?sCb
zW;MJP-wu!EBs8YFpPATE2ZF;}Scz?^5xqK5DTZ}qqTI`mbT30}wk@j-F-4!@Q=p3}
z`aV8IQ>5PgoN|wf>+nxNbgL^$eszT@3B=iEQzQW<N`2%EVp6ffy$HGaJ!R}9`;Dpg
zuDJEb$~~^M+<bF|DGA8U@eFi0la5T<5;E{l$~a^t1HIca_S?GF9XGBpB>@@u{*_yL
znP{wJ7eZogS4J%>i8-L$<4S9d2UeJpfW$nMks-yzBeV8|EKDimk(DgWD)+e3vQSxJ
zN&>R5YGEH{12Z97eL{rvJ*CWnMCnWP!%5WVJ$#W|N!Qf7H%-)s`0Zl{mlyMSCK18-
zn9uupz0QEFP|T;F0sueU4L@6izh9*EGegXLeGeUhO2p^gDMpKuMk7A$k%>e}TY-pA
zWG;{&;+bF*ntc`1>Wh51Dcs3tXvF8grNH&S@Qyl5k%a5-w^d7y_#CELKo?I<PS4D0
zM~+NCHhq--I%*1sj-KQ@GA^o<FpzB9YZu4;PLz#3$ghOd7D`%p!Izp_l?X`OA;Il}
zWQrz}ric<DSl{7{>q4;p7Z<D)af-g}8T%T~K%7FNB(O%OI_^ieb5hk2NbXV|m$dS+
zDJs#gerV}-<sKF7t|TM}j?h`g#7aC67)&TrlobXKDEHVM2I}0HT4728a{Z^@+Zv9^
z*!TJ{#mxBuW$8(rIf>Q`v75im75+*t@ZH$W6U(!kU!RCzyxsg7UavDCD{D7Tq_msa
z;_(z6fr{OHCB<kH?Yl%t%hGP{E7VOZa-XyZ3z(|#zlaI-IX=)7hRp}4#r&BRi2efa
zsI&Y?h<>uITC&CbW}Y|iXUnr6)=Xh7lRZdQ@5oi$@L=EM<s|eP-wkB%D%;VxP0K0;
zy9*oDox5BWTRS)a7>zxTm!rU#h}4Dh;Wv|fc*Y$jm!YRZK)=idpbG*0A{WpUmiJ#P
z_o!$qeE=+PiIRYYGLc>|8D!m56}Wy!d6d#h$|k-v8g<LBZ%4)+O%kpHY$h?Z!~=oB
zE@g_c!r*r09y5@woe}Da!4X*|keN`XGb>~sQ0_4UWHukUD(P#6N$9CSV@jE}tk9TM
z?lA*2##1g;46oGn1fmt?@nwbRlgd42fT&P5G*}`Y2n?<(Q<N13KdRhg1{mBFW4$ox
zi7giheMFhwtPuJ|<sLIYD5P(wsDz#hG=58&wye<jL**VbK;zmfT@&D2uS_aenF_(}
zuPbAp6}SIRxyKB+4N@*S3Jm?wQ-Q|6D$|yTMxr-P;yz#CX7Nh8sp+mx0RMftyQR17
zOtQV>qdxn1z0Po~P}HZF3_oSehdC-YQ~Fh6J~yQpD@qyND{YTZBuZJ9F`xdz6N^SI
za;vnLl(2b4zlr(u7$0j2ulj>D?DI$pP|x#@It!2l>Qq~`<gm{tNZ3a@)4X<9!t?R2
zgdH-`J`0%$M->Tqev->e7xH|W%X5k-(bLL3Dz26vgeZ|jNgztp<vj<~qU^G^K=((L
z$11J9Oj}4}QjZn=lyZ+5pxf2i<6l&!Fe^-cQ@O_sFp0Ycfr?aIv5=WRRHim7e7>gK
zV+Qyv^UU$zDbt!2I{!zx#|+Tv=BfiKS>03>Qn&HOBs(|eQqN^+suwEvm;tU`X{r%r
z3bT^6mnip`0VYXTAW)e~dPT_2NoA_D!f9H$#|&^<rpB66rZp>c>dHN4fX=EVhqL>`
zn4P=&goqjO8f6Y-#q+l+_t+KB6FqYh|8e*-xsv{>cQ8%9Ji78@%S`+ER3fnP0iaLt
zdY$oDp#V^yKmL>vAhxLg3>|<<1n8qF28$9;BS7uJibOffG6K|7ShAvS&AegBVhRO6
ziYfg!e26I=<qy%o&tIm%@>$+dXTg!M{CZop#K4cYZ5n>!N7EgxP&9S7TA?W?=5B9*
zl}MxS-xV|u%K<OD&u@YO$G!^iwuY+f=B?LSBj#<hE%SEx{~hrEyWs!(XIqj!WZ<#B
zn{$cML9|d0boB1sZ&*kEGjB0B5C-f!eZGSfGl@)#;&>W$v}7$W1My%Yit(!G9$v4r
zM6Hl2!nu_0&<^ANC?&oF0}1!{rx<AB)JYVyEY(mysG$S-Ls6Qjmz1zMMZbwDb%~EP
zg+BT))kg~{KrQi(I!liPs@_&DQ6F9SAq+&P%IN|FkX+z$HSVX-gICNJ<Pa%OVqtgS
z>Q(p^G*n}AwyTwl%SHu$;jE<1Eoo=YK3K6A;aAJ_YK}A!$U2N7qJ96K*{bwBkrZxk
zBV=_F5d#DgnF}Q!zI++1VKTg|HiYDRT(Y{5{HM9(rwCcUSGh+;<NCu8vX&?bET(Z0
zLgfJB$CRm+7F(tnB&yVd&YxB8F#~*->5l)7GObym^9AJ|GeD=CC}1W5#AlU9nH8?z
zR_-waT)PS&uG<~7nw8AnWzEDb$~|U)Nm6t$RRD3XGSykh&fUsAW`NT&HP&HeTC+mu
zVdWk(Kxg&ic2rhZpAcc=JgUrrtaxrH_t+KB6Fn;8yuFF5^Of{7(j|~BSz`wdEzfCx
zb0V(sPWun>dY$1|A*UUDP8fblznv{#Z=m$6`0cMvF;<j!8a`}yF%zXMOTT@f@T5^U
zml{!yd!Jv#ydL&dOzz+0!%gA9{3LbV|9uKve}#9{S%xHBf2plnvg>a7Ib`f)n(ySM
z@fY#V3I7MZ;|~o=?{J@)9sW~3h9NZ~pH6IZ*roR(I$ISNosksfoDabbv*efLuks&m
zlOGAmdzN#y3(5N$m%J3N^S70IR9sem5?p7Al7Q<xo>J)86ir=EAiD0RE=+tjm!)y2
zJKI~7d&~e)ar}lFN#cRPV6QSolCiQZ4DMF$F#`;?1x^FPJY-_iULY_ztW056m^`f9
zV+NRn&e~An2t5^OJgQ7vR%jT?J!XK$Xqp9xNl&^;0!vqUWLaVPdgUH7z)~of1da0!
zWs0)G;2Gr}Gr-`+D1(H_O?07<kB=);niV#`s@!7+*o1UUC(ZMF%Cu#L#vd#9m;oBA
z`TM*S`qd;sNc}gJQO}Cc|ESz!SA0%%Wl6x~E8MhRN$)OQ#+C2-E-zxT=LJb7b$rC+
z1-xEoJXR=T!hHR7sSeB(zL5?<C1kQa#b8n5X~?AAwNI3@_>f6*wL|^D!ae0r_c2Z~
zj>K2<dWv?%X`SAH)cyii+*AH!Tm9e;inj&oELtVkbjI|iYq#`pD$&QWg?*=gga%LE
zP8)2_nJ?Y{x)$aj^?J>7=;PNGG}km>o4;DG=&qsFFg?n($-~0fSkbOsHX4rZT6QhZ
zUMgsMW1(3!YA(L1*Xxxf@_x~7Gz_P1*Gkx|T$Q(fm@MZbkZyHb$L^&(tXa->-+>S2
zVcir%Co>&l%Y_~_LLCGF55s#Depm<K{RoAZQ3vqu_Fldma0EfDO>?h5;&=%H!2!J}
zs1eLAKL&Jiet-RT%_%mly6cQLnzj7ARWsaX&B!_C(Cu0-mm4Y>M%}3!kpA(ar}*bQ
zu-dt%JwBvq#uXie#Bj!W)%p@C(6q6!s$DWFd6b^K-N@H<*VL+ptB;YZi^tF4k1~x-
zjNLakhRG&l<~*-h(p}v#T)zr*EPeI6#nxq6;$o<gwkMz^pMJ*f@BM-LVgj1t8i=hk
z7cZ>dKt-=DH1!3e<=x<zt#TP8qI?Pdpt`w$V!W5=sGo+$gy5Ki-jH6Z+BFO3ed}E<
zCPWwhYwx;s584flvI(ZW?rMGzPOCTUDoEETC<VL1zG-@`#Am+)a|4nuD-HPRsU@N~
zEReR5iggKe4rn2}W@vMeg0#4QRKq}}R@7@62rDUMQl~klQK?{xQk^QW0w^zcsh|N}
z{>OaH&VwWuExnR=?Mp_@DX8nT9Bl-uFr~O?L5F&&1i4akwgXK;ldM7V>smEbc=o&o
z2~>?GtzkD|(C}3jTg*9KM8e@;T{FVXKz3D!1{hsYgChjiAccz%SzFTR=zUU=Vyd8p
zHXqf>O~*wNi=g*w@PBv>ctCJRrtt_cOk~Fa5pr_cPyzmXs?@OSC3~@UO*^K|+l{L3
zy1+&%68K8lv1>z5fyx+mm+HpwVQpAOVt5owtQuso)ry7##TVg$reOvAZ<zJROYkWx
z#n13g!>*8qkzCTC|D2RTXD}ic`wd=%{di48N<oO~^G#%jV<9DO?(rJCi}&%`<F$u@
zSN3h}TIyj+o2Feu>iz#>Na8If(Dkr2s+o{phx1{{Zr-1!eYGJ!WZR|ON|FCiMTc2!
zYw%-n4Uih$%}E2G_{@@jDO?Py=LT;d1&Z(T#eLC$r3y3+(LptP@&1*9?9^RV%>e^x
z38xHb#U-42^Gyd7?5H-^gsBU9p{5->HL9_TZA|?#Xv-qZ5i)~8YgWcw3+ywTYMRq5
znpmmW1gi^7F*XutwnX%{6VDW2&Fs)MoT_9-Q=@DfQ}f<-(D8IypYRuCwzx)=2&3g)
zX{;!7D_wg&g=-Jgl&#YX{{ow9YYWZ4)rE!%<ZkaB&p8>8ypj)w#69Sh^b)It%^Uhn
z%$1LYD#sgfXsS_<<I~jRz@CY>(St^hGD$ZPlG+UjMr3TcAp2J`A$x#XB<iHcI02mW
zxN<@I?`J~#M(*F`)Pe6H3t!8GvUfb^3|4jSE?+Reoe74f_FXci6QJqr>N7RqubSmA
zZV+z1*@yAzk@8ohXWq^TJDp9O$TBf$HoXvhL~@`^C3WJ<&?~IwgILLTEI~~L!Vnie
z46mRyxaJx#Hv;}e|A@IPYpwxv2oWIWIQU?#yxX_t8u+dORS#>^mBa?tJb51YU=wo9
zlYcIG^3<GwwfAONdsRciJs7Bq4+?*lF6Bul(7is($(O9lt*5<hCBv~6YWX6BfgzGx
zb6e#NJ?qwaeQ>0-W|Ugjy&en)(x-!_?&Mv|tr$?X$ASg0HR<&iP2+NdeB6#pq7IuB
zjeM~JAvMEk&9=%Lp+Vi-OzKI`y-jm=X-Rs3KGPqeG2&YxpWz{fpm7Bq*XASk&)$m0
zq+gXpC;0xEsP63o0{Pz97>gCdUK~BzAAEt^=J0|z^KTBp#?3paQ}+?17fda1>^8DU
z3^nIW5t_O6ID~EOs`aFy9hV5uW@xaQ*>xS?6E@yd8XFJ7Hv3Nh8727)ttp3Yyz04%
z0b4p6{8l9d)@bPYVh6HkM{?suHZl+)0WII}$yH1HN${*{L)5ecyGvHZ$&vBmhT@G;
z-AZQmY-p6`(ICj?arWKoh#k!V>G+<&*bMa^bob+T_YGc;?9+TEX_H)acfSM9EpH>R
z&V`LX!?j8<D^vaI=(AKb7~CXGu_J<S0vq<K`d%mNM97~sF>mGL3+x+4O;Pb`aIZ$1
z&7sN4BwtKcspn0y+)A1t{5;g=84B6El@!3h4E$EkRIReaX1aJFApHOr8wIzGOzFR*
zMEp3dDNpGOJ0Kc6V_J>||HDn9s!g_zs<1`K_I)(e?W5YFX+c=p2J#S5RN#)*(D5FD
zD=t*8X@*`jha9q#oYP2~SP}QW*q$P8`puDzDBOf1{kaWj2=~}vr<jNmHU}!82)oXW
zA+q-j`;HI_L*lGj9m46@pkEHu!rru_jcZ1YR2T~u<g`;j8bswG;tc~Ff<EVM*!)>^
zv=P^A7}z4+t0H~VeenIGT=ksJ5F1HjeLTcOP;%>slFQF#ifnRUF|;r}1=LPv%bDep
zvtOOFVkU)|Af`K8=;ies6EVKoHj_F>Xm%VUGla|=IuF?Njw%x5eK8~>!F45^xI*JI
zU%*Py@fq(&0>K%aXL+P5cGDIkIq(v0woV^Ad+yB1{LIM<Gx-C<Y^U_8m^UUmzHyMg
zfm>A!jpr(PK1aaJ{0qP~Bm4XBpgCmK^WXmd7K|inVeuw$e}^j6QLTWYKTf~q={y)G
z%HRK+KXgwHeU^d--UCtg`kZ|kk`H|CEsVj<fp#UuF>teQWHVbMxMwq>N&pi_^$KFj
zZVQo+JZ0(5&%aQm!ssVv)lG3Qh|bviY`mS-7wbvzOVsfOaIT|1kgOZ|S;(?sMunPX
zGdKf{Pv9R0uk=I=>0lcCh;8$)6uf(A&46ulcgVJ(Vn^&5V(5S=11T3o#=sH)p9ahb
zutk`y!_6EW5|XS27#>3i5;u0V<u1vjxft7p7(Guk4F?mo#`K!A2rd^P17-xIBk7eU
z<ZgoDL{{P<Mn2~91dJXuw3Lh<k-#}A8$HVhO`4;}LJSVuJrU)T*+SkGF~sos7MYl*
zkep0dwp>b%XTfqY_rlN%)evV5{lB)k*V!6^Jt+imf}lu6i>ZXjGC^QDBs<B<mD)oK
zSSi{bx-bsvq0J;Bl*O7vB=HG1EFj4ZToK7MIyHIj5wwj&4%x8Xbc63qJu)M|6iT3q
zZFEAVQTCE*qX$(QWiOFNC0lI`I)Bp&SQf|RcIbRb26O2O$YeHl<k{Z!&>qV8^WKm=
z+k-=iTE@LI0gErjl7w|~91>mK{DFhZv?Jda(lbFj^3G^GGTc4WX{xUaHTU6A_sMF`
zw;|8oyq@Y}M!c#K$=+8sB1Kn%S7NE~2Dx^Z%*epQkR8bkI5j(R=Q*&$z+gKrMUffQ
zk!1qc-${cj+LmYcGEjeuR)vIzx54%S4<<kSC?s)0z<SYcTq0@WHDaIveR8yrN53%j
zn&hA~>mWoBdxHFW+pgs_jSOK#+6uVw#V>vo8pWvExrV&}XK$PrznH$Y+w2pg`ci0A
zkl)|uxUkQ%Am#RZ+ro`8%-Rb?(|;CfI$03EhHvjXt1--wFcd;YEKF{E<aD_5m}ZWi
zy%n{=7yLHF!KMeZ0i2CKtL6k|-$4{qb87Hx&)H!Hm_JxCmC)+yyou!lB?MHQ0=sYv
z^M1lnSQJ^<$%6>5?EmN$*x@55zgYV5#fGt{H%h*P0rEXNV!CeKIXphDH?CNhb9Q54
zT%U8sCnoMWn7e1<z@bBMxMS$z37DZp7vvT<iNK*Uw5dAeHd{}Uqh{L3iE~r?W1Bj7
z&!OCbmrlIw9%#yNi#ZK-f5Uj935_Cg4n}`56fDm$XLJZoF9kNHF3RO(2F{M$E<f8Z
zA_}jFj8&_mB$@G>_XN2QlW~xvAkW9GA!M0Awmlb;byKMA_}NG{b(#&ho?7d~FLF8g
zE^K+<**lmZvQdyFWc_H;_(D-@R+B9z<=xOU*Oo6*5e$Ze+RFl-ab<YQcfhN$&~W+d
z49!?73GE6-^gCgO6HaUki+ECtu^XWNt~0pZxCzhqxXA>O+B=(wX8e}s_d&hHZA@vu
z5obopuEOpQW(VN5BxVqpQ)J5cTa+y?xV=VpFO4hpidD4S*3)=A@)fX5gvAbn3OQ*s
z;qsu44F%GPQs#3(DAvxd?cQTat3MO$4sQvGb`p6IqPk4o-8xo^ad&Uo=)1cZv<Ils
zcg1j$Q3~wIQ@1eU5Er^UEJet7e};>u?HDy^Dfk>q1Lwl>RtdX1J~jF3ljo<8&77K=
z%%3_Td7yrKP|H7XNc`%@re{u`&7V6zb?)p;)T?+FksL7_o|&3DlRt5C`uO=%=SbV&
zhPOWi54ru}!Qolqv2!XSkEyXz{I`9)tTZW)k-dZe%6XxWOrJcTX8{BrVvygaUcLMX
zKP!+CcVta&N~Y!5^z^CenPZbP`ST~I&rMBEpUfY~KQuLULL38=t9DOV#924FVVNcg
zpNXu=$N-;-Bz)S~=^vhBLk;0K+%U>uJQ+A$+L^mHr`nH%G9dBa7@0Zn9F&;@`BxrW
zn^OKs-UyX5kpNGp8O6(`xfPsC&*61zc5OA1XK{pRB(6rB_e?E3f3@wJBn0$%q$I@a
z&`yHD%de47^4`XpoxElzGxZVtcR8A=U9V%?!F5UQZJB<=nw`w=wp#N&Wx2x7vK&jk
z)ta5`0~G_$b8|{%u>=a;6x2s+cJlLj<|IUIxcS2dxMn9)B$m1EKHqd`CqdxllAY|w
zs3GwFI*2xBl*sa8-0#Kc_47ZGWj_u@Y#fN)25+YyZx5VZlF}q-abRl`ih~IgA}c{;
zOh{ea=``zgy8#izF`8WmsiW|j9I1nckUI)zVH~MfQ6rG3z-rL^Ypf9@<n;IXDJSM8
z|74>?L_bUH6DKC6N1<luwj=zoF*zWaw6E#d%|;RSh(^5|A#=Ruz_q4$c`4jZNHQew
z_EWFdwOzf^I=#*tV(-8K)zj;mxLdRy?pO8i)XLMgCSi+-KB>ToRM?6dfP+YwSL?LA
z?IpX|bnz(SJY-bBEe<94>7Z9b;w_w5Wj80n8Ca6$*P1l9PI`k2a2~dyyLJOw8pJFQ
zI9^n(!<UT>+lH$!VV}#o((=}i!%;k+ggo?n30@eYUz?2zeB12W#r&LJyo9Ari`;hW
z?U;u);4VNzDtX%5>_Ei0)~x2QAami!?G{J?vFdnOm+Y8BYkT}BM#bMl&o5f8$&b`R
zXS4!6j0>1F3>2b#1%EcSb@~Qx$kZ$RYp=K6xMI2ZTeWV}VK4Q1y(MF=xd7d5fwTcU
z(*<w9&1O}2G6=jV8aN)LkK|6~$&oF1tK|(g-TASXj=>S*mUm0AFmE&(aO<cELR>K#
z06&sibX(q5A9ksgUeZd-+Z!qgHBtTx&@$c@zg<BCt9?tbAXM4ghGPK1bKvwCT&oMx
z=xt^faRTBcyp7YSwY;5>17SBx`H}%Aq)V7q4l)DhEn(Mn9)<~sSS2P6W}mkOFZ+dE
zGWN*30gh*Sn@i1VJ<sJ4fF%zs5XDAKhr5L*y{+U4k#yv~cvN?cX35T9siHJ(gH2P|
zsmseXiIf!G{AJ6re0kjpr@bA0u43SZpigB#n<p^mHo;&Z86EDEgh>aDZl<rneY0?N
zEF9tXcH<}L1fuk0V#3{%&{_#tz7Y-z0;~M5r`LPiak7Er)D}PxE07%Gw$5(5%iB^n
z>Uo|qfrHtEPv#)CBaidKdjZR6fX7a*zT}#aKZSWHEpHdTNeF;9`N`szHvrGa?zrb(
zz;jDuz8GpApNQt~WX+=jLy^5?G;)^V&XYbs8;qOq!}8GCpsUZpEP{KQTc-!Sy&#j-
zJfEC=0B}<&O2~C@o3jK5Ca>g8Yr%vx8x!QWr-C_J-X^E2!$f<c3A5S(pmvh){?~lP
zF6tE8y>PiM&;yM3=Nq6*AcWeNnsw-63sC2uo%hwib&wt#4l~Sy3{btW0~LkSoD02X
zRpACu=u5}DoslrqM<6oQIoKL@+$EsEGz=H|*j0qLF69l_w6EmhK2l6eBUEq4mQfWu
zMrGbtXxF_BM(r|i@C8-)AtwP25mo>4*{8i7!7{i+5x4c{ya8YTz=Q|cwcyf2%qYtH
zNdZLCw!<($7G2QJ&}+$<cZl-e3YmDNBHYpHj^QV#1sOOug<pl^rbL3728Ro?H@xJ~
zy=;S?eC8(39k)cAI9^*<^7>f59jf8-2xHkUXsTMZCbGHwA}Hp0G3yVquHh5!_)uKO
z5$(lv-Sl<@#*SAw$KX^RhVCLsy%!3a$dGwfo#Yo;0PeMx*N+eV&m-ku2E|(b4E}j1
z9Ct3i9sm41e*HN9c`qDyEB`3|@!-yd@^$=k3co&xf8L3I-j099Acd~{V*K+T@XxQ|
zpO0Y!AAvtjZxDOL`~@-+(&u&3-*wX0b<)pu(#LhuK9oYU*GaqANt@S6d)J%hZD24E
zg3V1nwV%g3lnJ$HT=~@Ead4m7&*CRSE#9Z>Q;WyceQNPgl}{}u#u93Y71;DP%bJ0}
zwi)tQeA@+Z^`GDZz~1vkTz-ODi#!EQQk}Cad_IwfH-;Yaw+90)4zgZJr5%CxVJ(Nb
z2by_TxGh=&_7p5nWNzU)yb<oHulSQ545j|#IOpK`yFe)3)H)BUwyrloX|qHMTpf1s
wRVwg!BY#Y$`$nr)tTaogkQT5O++7X}DP42WMD&J;(r_wGFww5Hnsd4T547Bfj{pDw

literal 0
HcmV?d00001

diff --git a/.doctrees/examples/TASK_GUIDE.doctree b/.doctrees/examples/TASK_GUIDE.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..901135fbfef77780b345235bab719a794cfd8551
GIT binary patch
literal 20428
zcmeHPTWlQHd6q<x5?2x>$}y|hqH-j~wuEq(wj9TaA`_u)*-=SRhAzIqO=f3z4|j%~
z%XTij)KQzD2w(vNq=mO=^ID|nL(s<}KwpZsZ*|j$B5j^pAZ;20eP|yF6b1Uy@7(5`
znY~h^3$&;K(L3k*pZ|RS<^1RJ?aANyvpa|Qf7$7{<G8_2&2F`Vutno;cDxlfQ=bM&
z_oMFB?{pt_uVmAay&Z>X)TG_)5K1&1*K0*I=zas4Gpu~#dT~uQa4e?H#0`VIzBB1e
zImb}ena(B?H}R;EmCdc*e!~m5$*VMII=&rkk~dSgMZ2s9DjdB}cNi%H%YC;ih@9%!
zASvlOv)QzjkHMM9rX=ZW*-_68Xm>4b*jdF+lE`hO35LvuIN`Qp3@1C*^z1lBB7RQ>
zwl97iZ>1gm(XgAF5nX3x((G7bw^>%#Iqh6nIqb}5uP!>FPZy&!2xzpp9Y&jrVu<R0
zATC}ML@m-C+wXWZUcA42Z^e4^!L8SC*8Eo2dD=PYoOdSD23RnSk=gCY?l>mjzV{9#
zaoXvGQ9_dKkZjw#F-by#nIv?}_EMWo0*PsoxIsH6Za^@FB%+;gfz;PlSIK=l-X!-#
zex2Asi_}+dS8wXNg#tCq5Su4(nIsM+Et=SvbrP<#Pyaffuq)17c4I$;b>(z|0gH?Q
z&itBl0W$_R9P`|SMz&`%;Y@iP0ycA5Z-#F8D`vyC^Eu4Fa}i*>jDOGI-*fnP6?5x+
z!Fd^T3L2aRMGq*M%O*Lq29o+^AvVrr_S9g0Dnf@9(Il_zmt-<>`gCUn#?ztvsT%x|
zfYlYUInQ4QWU37%+SN7}ZCkNgl^}TuqDmN(P!WQ~Ct*TL#R<h@rCQFYgzSkIYGe>B
z`sV482jjDuN&BQk#-=2ymeY_vh-nm)rX3J!$F8?anh~`VN&>n~63AW5AtdFL3zQp?
zmW?c$K&ApxVKUZ<!Y$}RQV%?jTv?-S$n+K^t6`h0x|`H<oiJ=&trMR@nKf&<2!3t9
z-5c2cmdO)<|8QkU=1G(4f#tM-%nu!yl93Um^!#Net)nG-O3oi2BA4vrIV*Le1qNtz
zD-4_Jf{BA_RDog$QYWE&(~cUY{L&r&;rOFrXq=X#`9TjF{WoBQxYG7*m;qq|nqC-C
z;wB3mi4vYR2WTU*VT;MOn>f4{vu{k#S9X~NVel1|+#pVDAd^k{dgwvaVQI7>c;jHY
zHDL6|CP&IxEN4D59<`d^g$71X^obI`r?XSiLRk$jY;Kx%>TfQS<#kvuuqoOElWMhE
zrR^p}OoEdno%s6VB2t{RQEP_&;*Iz2Jh)qZaPQ{bwdLD4m!#`ZLy1bWMT9>gT5xD{
zGfWdwZIBzxiKy}lm1Y>UY^8Oo0THU0baoRb3|@J$_7zgym4$4$Yg<e?i83-(O|IW@
zzu4KWbleVH2r=JPN@^7a+y9G06PSR_|8d+HtVXiqo*lGPyA8MaXv2=3ZZ^H+d)MQR
z%{=AforuB;Y*Fq>e9X~uVPyubN(t2v%O;`rp$W8M%o9p>Pzm`M+GYMVVRGJMLd#vb
z*M)r{Z@6&5xqm&EjVPt1<6lhK8_X2uH8E{;PD;loJ=*AN@DG(?i|H0~>4?1rU(#-P
za4VG!OoJBWrOv+LoO0n|E;WRQlg>QNg}t<7sFD9PhvH~!bd+_b7$ZLN@oyYDeCUu_
zZ#WB><M>(00^S`rFEgcq<gGgek~A;|85z~I;_hGNp2tO~2R01y?6ah{*b27;FSJ{=
z*kK9CtLB0`MoA?H8|N_iaYf+>1i|@x=N}jn6zZP>>J_cpOZpL@$$ccOrZCr7yu?^J
zkv4oB_)U5pA~8YMfm`7?Npr<DJcd}SEPFAQHTWjJy$cH`+hfq@9wm3F-hX`=8uI4-
zZ(LCtH|!*?Z+mgm_9#=XO#Ql{TBzlVGg<Fm%}IRks`j&P9oPUgIybJAfL5||{qir5
zG5p8s0Y3MIl|y{_HO-j{Prk5vKPFkDq$dB)p}nzG8qZ+C$)?ap8oAx<#(pM<>D!GP
ztk(-G7laJvOygpBxHg0tqlHhl*$2@7d@F!6<i;dUXh*<t+l7M^P}*YfZqX)l=Q(Wc
zAl_!6-|fKg!kH)a1d)zw>`$FkLa09W(+UG`w+`PwOxz}0C)il+!`*785sP;4l|h}j
zEQY{hHe$`s>ZsTNVge4#5Y`ij5g2{OS8D|7Q;-+J`14ksFox(E#fTk~`i<qad+*%5
zYu$hEt(!}A4gVYa8Llw}@0Ji^sO0IKDGFgmipE&I)WTqlsa5?Y2@KV@0EV%)*-$xo
z2x2aqz_jdvZa#Shx<!hnd+T86erhm4#v^wE>o7$q|E)><@L)*1JO)UI(OO)18k3W&
zsJZO=<70eS*V;iiRgeV5uu`gK-R~a^>k8_sjLOmKN+P|Fy0XNY1b!+ajBu=kVuC#-
zb*a|t+~x{R#)26gEHtoeDfnhBH>$AUp>IQp*`&@EA$5XOttdV_up@)-e^TO>p#^fQ
zBAwR!Tl4zQOSMO9fM=x!Q2lZZaDLc;$0K&A7x|NeA@eBX?=W&o;-bcA=&}>WoDFM9
zqyH$ISp|CWq51!>2Q#oq0UX_Knex$im-CwA-SCb5O|VTYlb0&~7P%o|zNFIB=q`>f
zDM)ml-k3TvzHcm+iDmEjs`ifc@UnT~<W`xtP-Q`vs`P`a%nvr!in5@PURF>oqBxdO
zu(XC$OS3W7kh6T<c$c=_IElD(p*YkhM^D;Cuw?wT5?e0Tq%dBPH6e_8o$xZeAK{>M
zY=oo`iQ*oVWCwOL*snz%{a9Me=hSdiJ~yHP1OJO9(hV*2WZ!NSLk8q(ro3Egd9;Rl
zUNS||)a$&e93|HpK7&?{s?ENP{;fM5xPLsH2Z)h+z5vSO^DK7N*{vh~hixW43k-12
zvF)Yx5P#t0R_d#FURz$ZAX`{#S#RFG^WZJ(_VQb(RLtp@COl2%X!0tYxooLFQ>4t8
zusbET8WVPI5dCV-1{cuN_D-q&(bINKPMe~PPuqNnW&il??;A`THlC6ySFNfkH^w{H
zpYdw6gR=El{ev0nkp<IB9zyV$0>Qw8BDFU|6ZspTE|IxEsSV%L1u}~eRTK{`qly3R
z10a61kShfC74j9V+WHA1q**Qp<!UwVV7a=!E5y9$5ysCEriyi}e*%pfApV2WFb$CQ
zPOWt6dNVZJ{;1U6Xd!$_3ZbHe3nA&VCNGR~ydmW7=U{jAPY;IR(<0^tWg+>kS)vA}
z^uTCTD3%zg(7w^_0>&YV<dqL91pi&aUd5&B{O^KD5QY|MeeL0;h04P`#(2Q3iqXU8
z`YG5uZai{I?0NLWBwt251}n6?@otZR6h$6}Y`CmQlUOMQP6kTLCnc~L(lS)7Ncasc
zqXF}8rCvtM(2|rPMIV=;E8~owSvNzLR09gD<&b7izSg~^#kCxADOC@0a$+x@CyUbe
z9pvHbL#Q7uX{t>FURu)b6eO((tuchI5P~pe?ZDg3rEMNe-LJGQ50bXcTil&TfF40h
z$HwsnwlB>@NQWJmv$;ukt87aEn@2VlaUx`TDc42vN4?^ft($g0wE%u~ADtS=sFu)X
z2<~hTY06XRlcVwbQmMt!Li%+nq>2`?NH7DLD#wt%32<q=aSU=Ar63N<)+1Uupb?o8
zSsu}7iBw$bE0M1kBr?a#IKs$6B1;96cQB3Rve!HS+56=|viFx#47P1-C1B}>lLrkP
zk;I`vu`SD_FcxCSw$;|%C{BoSm~4WR0<65)bl`_X>Rb%1*Y4bY_14<*{abg|mc+@z
z0TNuP3o$9wfM+3^d0}Zm{J6HX04JnAM6`9R@s)jz$q<UT1l+RG>XSd)!;oR^X(M;G
zmu*U0^vlx1^_wLcN2`mwQe7zOxw<&d&j1aWqhUiRtxZ*<ybrN)s0aClJxsM4{hsP*
zwtW9!*izDxdUWzksvg@QH8{4G{Nce+T-Yio(>q#ADsUKDQoF?$?_!J5Z^(^$wq?N1
z3^4^r8oQ*vP}k;lerQnxP=8SxwqZaIFDMY;m!8^>OEpF#{%whP)t4}kra4hL%0P~0
zlxovply>mG`B>&;EEQ<_d!KeUxe59wuzbJn-#dd$&@Fi`&JElcHxb&T>EKWm`yuwN
zapv2XCWvi7;Ruk3NAO)Lt~Oy|m@}x+_tkwgX&~+2OGq=YIVCz)09BtFrT<l`I~t|m
zlqglba+FTVk?Dq2_V`d$4fliUyAABxv?%SsrsIUdpl~Vq6)gO#RdFo9;wKggs}Px=
z#?*>z-{OaA++U+Diyb{8Z{U0ll|_8qG4U+sDp+A;@o4_%KhTK}Wic7q=LPv@Mo(~G
zP}GO?R&3$u<GS0fZNxZ6Ey{`W(S#i~76~|i=64bcn;K~|NpYj$(&fvSzn=PbaGiYM
z2H56&h=k=h#=Zx9qwAM=y>F3(Qu6r^ys&LyBkAEH_C(yEhEJY-ZK*epMiEibkpJiQ
z{|vWt#r4F|@fBD9ow<5JZE3rZ%`(@x8krko`tT;t#33|e*Q{p~PHuQPaud#I`V3$y
z_WN)sl-=Diw))up1%_hQySm!byTPaKC5&-dV*XkUr-=t$`cVpQL@BMFz66f%ZR5|E
zdhbP96M8RqIo`bl^@Kx8*jA#r{{(O2{hJ<5VHl$2TZ^7w)n^2&jvWOwj`h(#<1qE8
zN>|m!72oEv{b-ZzaWSt3QjW`P2u#o&%5J#{v!u=@;uu!wLua(&8WA@xD$GVf--QIq
zkXbN9Dm0O-#h9Jd!gY|`%HdcRw{f5W0mIeX*3GxY<{3Mw$Tq=X->~C3crQ*beNGj4
zWtg(pOVhIP3K*Lo86~6mqhK()phm)Ksms9@aAOt3WG|`{e|wlW^PpcSaioUI?dK$)
z*ddH4ujy26WnK-Nf6@ep;z%)3kG?tf=+$|&zVyHN7#Mu4(dL_8bJ;8Vg7E%&*9)pA
z`$~dh`+!`{|L+_O|Bo`s8#W|O@Zhavxdt3uK$HE^!H`{81(qSLEzO`K0t$-#RC4xv
zZ+W=UMq%CL5o@J9a0a0AKDUYqJIMc7ig*zhXBvyXjpMTs&aUH{ENjr>4dPnt)e2%%
zZJIDgwTZK@%Kc~lq}GDx2T@qsl=xwZ*=1c&&Mzv>iXN&$rJ1DX^hc#`Mk&I(kD&-h
z@XgediPV#7bYeAZhKp4WFL;5U{+`QD6*zB*gf86^6j93}DFEtIv*xeIW{uL4eW{-!
z0sy+<GEt*3uNv-)`H%0pKCY)k-mcz~_}>sJ?e!&$Gf~U?9+ZKCUzBFDEK7azhsH{K
z%4<d2FCZ!r)-Owg;y+3>4i-d(wjC)eiW)8}<Hkx=rQDS$#Y%O4ZhZA1_Z0V(qu)~<
z&6bmsdvV#>j%i=ElwzfNbUkX}i9SBAQlWkGG_4m7hSq|8EK@pKSE%v#)fLZ)t2oiv
zu~K<=XJ`dO(!Nw;f%Z^0e$0YS$MARokY<KVmr7Me1OHnR_^LBuB+QVh%2DVZ%{A4g
z!8IK+eZLPTMr^<BIbQmy_8~BeXu$(8d%wZbUq?HdO`5gYQ^hqecR_lC1Z=flC^#Zq
zAeZ2HY=gfzqWdd5i1|}0&O-JX#J+;q<9mM=!f9<>b`?*lOeP_2g?CpDWffJ1w(yxU
zwR$+4kyn?j{8^ThDn&fck{y>r!ecbE`QtL{QHUr&(CTJWZ?jVXaelozo5J;HMC`h2
z&gr#mx=r!yik*Z}7as^DQbh3l4n9sqVVHnay6*0Dv%`zs?1Uh}Lcg0RP?4W$<lze*
zPt#&gy|7%6GKuR|(QY?8yN+f897r}YKg>?CM{k1Ew{~#D+wXpu%^<eub`Xp7*}be7
z4L&VWsJN*WTkLwhrJkEXf3kvne`YN^$?oY#iREE47VJEdRUF&PKPR(Uy5q7Crk$j*
zDm0l*wrC@5qq~`HL_`p<U_nl5`$(C_bTla&5~GhtIW64h3tKG4luf6}diBfI*v0Sj
z#Y_uPOoVF@vS@iU!VEA{n@Kl2EjVwqGRa%%X7i<-QWG1$3o*%NM7x>>Rr`D~qf|LN
z!^QyY#SU2H@yphy)KzS79FGTONANmP_5^}AVbro(6vvud=*9(X*0L!PAZ8qKTM7`)
zR5oL$NoeV$?1&4vooc0i$I2lI#_|LflZpa_deF>ip2ES-W?v_DVw$!>3zx!C>I~iK
z;Ch{9*5oK@Cf1f4y8^PO5m}De0!oIlQSq1(6^PwQ3?@b#N}R<9YGg5s-;HIj7R|9=
z=mb~K*#khB8`fG2Q+@)^IOU%!hqGBGZ$M7a1`s{CxJmc+ndh>Z4((X^WluJklPm=n
z+^NMxBRi*7C`gVGDPu1R)WWC@iF%Ti<OHCkg{>P*Pr~oQ(@yo7=f4Pk&P40YQu8cD
zH~*|^p6voK$Zpc8=F()H_W^~l$Zp}HvuUggAw{jk>8>2l=7CAy%7rr@fF-WuKg!O;
zySN~^V>xcy@$e7$J&HFl0QZwI-knOUUriy^F&Om({}!Lv!(f+#`GZjNh}wZ3OuUdc
zaGA4^y;_Ca`;YirSp0!2sRkYaJz<j_jnDPMS*9rBv{5RCTgYw91@=q0FtV4bHoJff
z_FG$!;9XGQP@15RC-G*4y=hTA<>6V_vS)giQ#~u+-2zs<b)nN9WfL^m!VI4CcjH7e
zheOrnk92*Qoh|0Y&^g@?(?)h&=pP6?u<HgQ6pFQYgcUev$1H{cEGCGH0%tAuA{Ki}
z>olIcYBj?&NUH2t%wF;OrE*bh_wnx48)JAZY2w<;FDXBe|ITgdPY2l4-mc6r`^Mig
zzXxwOVJx$nCf^PCt6v*`hwMBCdiVv^HGbtCe=?}!KJ5+YI?c`&#?F?i1NgM@6Li;?
z)L$SYWfQWls`IPT?9H-=c(aTpir-fp$Dtd~W2LiEWy_tPJ3rfa8s|PY&appNSo$;U
z&juRYX!BO5S#7qb$Ek9hN4)wY=jV@j%}2b}BVOZC>Re=xWl`s(XnxcAnezkQw)2T-
znC&w1hPeZiW~WSn<kX&Oy5gELQvxD|>0MP$Aj?b<<)GOhz<wk2av1ZXpD88grKbxm
zPODydrL%?hVL}0UD8whrZOI&XGO!99(#-swXyUCKNHI_kzu?$w4C+Y#2y^k>?p<g&
z9F2gk#5~6kYSCeC916)N@?<V$C)}Xvr7flh+Dw&Ka9=!>*RIx>bCy-Otj1p2W}L%g
G1^*2U-8j1d

literal 0
HcmV?d00001

diff --git a/.doctrees/examples/checkpoints.doctree b/.doctrees/examples/checkpoints.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..5b81eeb958e07c417adf20aef273d52bb8250ec1
GIT binary patch
literal 8439
zcmeHM&5s<#75Ccg_qVYfBy4CCFt*~E0SU##N)BLbuxzhHAOw=2c1?H9RL}Nwb-Jp(
zyC6%6A|#CJM&}ZVOQcAVa?C%FOa6h}bI3jaL8K_ZSKZa!Gu~aVjSn1x<Yl_P-mCY1
z?^my0{bu&}|MrgA|IE5dMU?DzSr{fV<f<?Wp$zgdPjvBhap||k{o<lo%-D{Sc^2@(
z9K(n}L~)q$r1-QjYcyO(vFcXy=M@ihB$Il&m=$wk{?p>3STwUb(lKv!ndNJO2ivKP
z60HgvhEdOa$ahKnWak$}sm)x<AX^t=(=7V6vf_l9t90Bl&%{x}i><uR%o5W&i~707
z-e@l;qEM0E^Fhp%LL+|9CTv{(S_t#B`RKD?JL3Zylbd<XhSXJvb@9Tj8F9{B>j^pL
zy(~`>p7nNQw%xmZ=SD1d-2b4Q>Xl0M_%0i#F;~5y*>86o7UFraBF>5vc^?~J#Mam_
zW2tDH=$o5~H{_7P;>+Hi%)NjmUKnMtbG+xpl7-$NlVh*7ve)19aup>*?{23vhGS!&
zboXWEr%a2xUGF`)!|(CzvZn?2Zr^5i-omnqdy(=~91R669Y%uzCtG`A&OI&d*g&Rx
z84bWtmGiE3+C_23{P^3bSybMIIeKIcac)bz00+V-^KqnU?`@Nu+HY4Se9oBtI~xH%
zY71cEMZiLw2k^d+f3M)*tN3>bE*C!#SKw;sA}%|<pwSs~>uH*=(Q9Jk{7}6*ExjCk
zmdj{u!fnGfn2plGO{h6jJ>1c#^Wo9WL=iDpnd$u+&Fee?kz`Es(2pb45K5yWAWuC$
zJUnjsw<HYFtvB$u8&RfoiT;H1kia<P+5(aV0awbCg9^qL?qx*$a=-%Kml9E|ILiXj
zi(@uspI#JNr|N32H;lB%``ti}d;1T))b-wxJBbZ4iWE5*L_x&jiC|+%=qYv&AtLzL
z1CvSb?#rKBQ5_1m`*)it`Qel8q!k}=+b={KpeTw`u(M-qF>B83zaQuLgSs2{-)oCG
zI7-YM#OGO5m`g`Cstqj-%q&!xv0JjCb>#`|eV~2btTD7Y;Sy(vsn%xyUmb1#jwK%6
z%1rklr5+vIZndZkri0B-;)g9uhUEky+_v4>+r-l6AMijX;bjkhaRQV<v@fowKukgR
zUOpV+Sh<0?@>HItGDCDv1VcSZ3@Y%`VMK={=2@%LVR7lmO0<mlm$pm96MOcr?PU%(
zE%E<S1yU>Bg7m`CEO7+QHykFNu^cQP?r4zt_iwBJa@!-#x)bJoG)A3W{#-$>LD&~}
z(>*O@;&po}$f8su>nXA3TT=9QI3*Cu*U}e*Zo21nI#H75+DBI7Ig?oCL{FK%sOU59
zU3nYdGS|~Xo%Kq3?_BA3#Ncje$_6FB`wl4O-20u-rSbf~9uH9~RYd%=^!(u=M5&6K
zP(kHzY!RI~Y#l&nt;`_!Q1HZN#kduCLn+JLKoZ0i*P3!*I=GwU*dt(31V6-76Y%MJ
z*OQUl^CEq|X05AFF04fc|7}~lMPp|j^7n(YL(buYYL|}YW{w?e%Q@Uh#B*zvSFL)M
z=a$<xsFxNK`XBB~a=u&lKSx0eIp0lqwDaG+C{wy?viOp>C3h?gM3snn61S8HU0#`p
zSyJ%QjO&cyb{Kk=SGwNMkpep7KDG5~V~t*YEI*LFHcCH?Uan^17JKMjfJ|1+5tYdf
zc8Xb`q+IW%xXRDrve}OU{D4&k(QYB;aEH#3;qDd&*$d?seJt*hktVZWt*&H#KbFCE
zJMHWSp;r;KwXgCnT);69RylN636I4wHDX|A$r=7}r<L{<&yU`Nk<=juG7B+FhB+Hz
zzsHp*%;N4izN%8>FDPVS-->7VIBj)wgX~RLy%%n6;=1Q{+Z=1^lj`m*`x+3kv{GI+
z+(j8l(E!-eIpGTggqn%!dT;M}%-dn$qi7t(EVJCx9F0hge14C`IU^|hJHVqK%Z$NR
zcQ4n}bL%MzZ@6gmiz#@v=;)Orspvd-#L-P)^rs{r4ss%>Q&nI%oh?VrTs~ws2gd<n
zn}V%c(|LRJF0^u~^T&jUYd?GEE&rpNAKZNV_I3ZA>mS{`cKw5&yypFC$-{pk=c!8R
zK3bP|Do=9CYRW;K4olPy&wPqHPouO|ca*73yE0Q}Kl_~l^*n-yUrx4%*+j(k+%aMY
zbO@VeB3@4qnMSEc+ftB6A3YW$B3KVavPQWSHKkPhN@sbXa~!TO-J~4ts`t6g)$YI4
ztoN`PLggdn8$`qINXf*-`Ts5Y^ekFa;AcST<51w}V@w!*f<IHNcY0sed&3b=f&<o}
z+as2;1laJMJ!Y#@)v>~pO>RiyD`<K2EQSL5#VU)b=(}M6vr*;&O(Iw@SMsa+T1g`M
zT4lf`EbQuARk^LT@N%=HYOOidJhdDO$hN9@;!#;qUDg(t%_-zAiHggE0`WP-`Bj#+
zXSLKQk={CHmfRQ~;xi|rn=xxao>3v!uS?`BuB9xMn}sS_#tXn^UD8KcIng``ks^GA
z0}3^AS79=9CFc}dVtvajmgQ+JvjQJr6z~Yj-S}9}qy*8F`GuM3p{8G|fQn$UjR8y5
zXP(9QTGcY}sq9X@P-PZ6XM2S?IlwZB)V$UCrCFt?2Bb6o?m%Ya;!CpzvWikL&zK%C
z6c&8GY~eAtE#>LzO8<~oO{m@iw#-V#sixt_D2~IN$IX&pas4@KHu-KuoXB*p+@M)A
z8}fcWgt)aGmQkI9Sd+6cS{C6B)F81X(yzSbqE;tEs)m@wTn{>Lc2tDlXD2&-o@FT4
z3ltY)p1}j8YM=|VUa@_%%B)?bFz2Rvrj{7}1JPvG%5@)}eRi^AYO>j&Jpg-3z(ldl
zpH^LN1&fE2VBtx~Jd1Y;nT3AHQy>RoBEV+L%$3IhStJS_Adqmbu^d%^rpX+S0JqgJ
zAE$l|Nf_2PptQ;}r1Z4lx^1ywC+{~+sp5GkedKu<wZV5&skq;sX;Bhr|6Zh`60++k
zJSf(Wd39)E;=BtLgxv}@qi~b7afiOuYM4KW@S;O6NN4C5gdm%=NgPjTwGb{}#@m_i
zK5xyKO^P=lCmB+jhlEfUcQ#%%Ybj6tx{ynoS)rCb&S{@Q!#wAzt7uk8E~viJmDvyx
zb(%(61u)V_B^~b3HofI(=atvqfIZi;K`^yEwKU6LaLZFVK#<+$SvTT(U}b<1dP0w5
z2wT*wk0=UpX}Yyw&H<BSzYflN1AdudKQJ3=FVSq*7by7HXQcv<bCRIHqQ(q?_Ujx`
ztzf8U?RWW!{5N2gY0m`$_mlvRFXIhXwkE{v?K~|W*cV9qtXR>&1E5C+Q8EAqDqc8A
zM4|Hj)UarbdjVvzG8pOAaJv9x#(lh4D7^;_1m_wuP6rsZ?ej5<Vjpj%#Jg7OO<I}K
zcE#gC$!QPF37*`82cH}7;c25bj)iJOgR$F_lYI)DQ+%2C%|gjP2zX#OO3GJ0U$YFa
z9TsVu*aonu@#YGtA-zu0v*tRUAHzWAiSE!ZJf$w0?@bSll2W`QaXb^8c>T54-C=Ei
z*J;Jc<NK!-4~uFam7@(zqoWOb*=%$gM~Qub9c`l9<=1AW7L8xE;Fk^(yw7?Eh|bN)
ziM^Lky8GH`v<`9ms(KzBIXPtDrrWm;dOyC8_hV`-e_tH^i887`(w}eW&mZt7H;Xia
za(?R}JLZAye~^pw(8W#=D|WPaKR2s2QXtaGSPdcsBl)>*A+~hmN<6e%%5mT$gx$C=
z;~IQ+=&7m3di3JNii=LFopf?yeH?rMDroHV^jcL9@(*yi1tJ|<%YlsJGD2{g&tD^e
zqWx=F@QdQ7;3YhGgW>S{t`Lf)z?#x0nwRTl8ym|}62y5((K{ppUJB{dsR)`<DzjuG
NOvU+-Y>7v{{{bvM6|VpQ

literal 0
HcmV?d00001

diff --git a/.doctrees/examples/customize_conversation_template.doctree b/.doctrees/examples/customize_conversation_template.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..2b1627083a2de3bc810fdae15bf1314126f70e60
GIT binary patch
literal 18349
zcmeHPU5p&rRrcS^czXOdUVE~^_WG_J?~K{$?!?|mtHJilalFf7du(hc-X(Tws=IDa
z)lB_#>c=y~#>+-1Xh$kU&=yEYgh)t;HxLi-7C;aw5-$lLMMAvgg$E>fLGZu>eCPgD
zRabY<^mx6?k8Ek$U3Kq0_uTV$&pr2?{$T94e*djq;$M0=>e`OK-Y_l84=f&K>3%C{
zCm#3X?4#`bx3VYMTsj$=Yf+GdZJwpO(4%cTt`%}Wdz7VzsC(?VQA3TNh<H180zV&a
zkJ;n)#G`D^o=nGLCw6(MDLs8HiQ>R>ewnjtf&Z9?kx8T3eeQK#Gv*l$#mIg4`8uJr
z{eW+066^7<2^{0ho=zu?d{Oq^bX=`vG2Q1nKF=1DmYLSfI1Zgw5`zY`=Dm&;(fTIZ
zt{FvmiSJ{+>B(>Vt)yE#w9NKu$UD?0Nhe}+nVK^Du>I1F-S(;ULDLRA-V77p=V5a#
z2v?goZ@uXTYx+MR7d0hPO}=ho;XG=#HC>E$i7rO0=+f}4%sycsuus~1lNJax)pCRO
zs_~5^0MRt-H0{wo338arq1m-dt?&HF?p?dy3_{l8%MPIzEwBgn2Gg^=!5FJub-f^p
znd!Rh=FM9Xiv!EtU{=6P7CD40TM1gM&blGDoOaA&n=>EWVrQ^c>r8{)5141Ja?#-U
zF%L~4A(%8?<J@OS#6xEK7V{v2E{nV%0E}>+S^P0~gD%jbJ`4<MuL46fZZJqJi@Ln+
zn68{CIs?qq9Nh$H5DIf<%!oa8?pyO{Od{s^kDbVA0d3V?OuuFeL>zya0h`B8n@7xb
zfb=!njNiI*|2*|R6LZ(aG+3ZSbq$RYecJ>&rek4Iogn1%Vm)TGTIvh{A&<HN7Ahc{
zK7(jEQEd7#h-7mt@r><Q7BC^oVL^~!)&}9ZMRQAyIMWb1F=x-DC%0l<rW65E@)}8r
zeQMEu39=4Z*b9X&NkKYssTH#3m6}wxnc}$sait8xv`<4v?3baLXYqFqf3x^I4;>lD
z>OC{g(tUR9xmh~7?zsz5*CgAq9}+^u<ym6CYF{c4v*)2IyMWqdX>shabnnB5IXdEd
zfy$koqB2H?WIeMN?c2j9o_XgFpb1Yc=|BD8cZp*^{ouQwg0!p2BFv^m*zrFhiN{<d
z^J1z`PU<mBS?+1JlH>~<r0f%chsiY`DwM?Z3y?6V03_{%-IPXixD?ZXdTD^RvKMpl
z{LKd(_AYqrSkkO@T)ytK9M_3qJOh>+Z=wEW1D>1K275P1pxl0`^9D9q7Ode_$>irF
zSZjlI0v8(xSt&9?QOvs$EDl~RXx;r^h*_b!-O*H6Gi+QjOw)f#(_ecONA!#I^cNcJ
z%e)<U-2i4(5VB0fNRwyVeg(4M8XZ~M>h~*T-<Q&57s7J$9`;M1#MWTxNj&o1um{Ed
z{V?Aim>UGEV%Ny|{A__1tNxAAS4}!KSy}aDdJsy0E#5VRLsHsi-kF7oz0?Y#EB7`a
zx!yvJG=LP{WNmDutsu^KeC(qeOnOO>f75H!YG1+Du~2(p!|%Wlkn@dN?W*2C7iw=a
z(;EfQB@P5Mw4V>wSm0X2AX<Sx3&&@f%MII3JkyWh`83$u4Ejz^QG_|W4sUZ7bXXL0
z;x%A5k8kKAhuh|3vBJP_^KMMt2%zH&4UR+8a>(I@7ZY(m;th$P64&P>+X2a>48y*E
zFKP|9?dg8k^p_Lx0M5O|J7(hIjG$bD?+R&t>BbbG^y~4(w`1l+jGK`#@@6Xn`;@~g
z-5%KI5bbT7;h+#<j$B=zYHXpEf4noTJlUs}Q>_GUWNf&Q+LBIcLyGeOwHQ!cudlge
z%6?X=ZoWLWq;&bA|K5ABW-(4&Po6yVANor|gvm=iQXNy;!GBm@7d+DmWjsnaq%RJe
zLfiD=9SQ>eZF!>M(m^}Vk6_kLDFO=olGgUVg&yz&ng<8UEuE3SD$Vd0a)ABUdMlGG
zDbf5DvV0|q3q!aCy$|*FFH!qn%0#e#tY=JT27}b}9oAe>_utEaw_ZC@_a#MLfzK$^
z)dT){RO;#h+f(<8`5satasF<hkIuqA7sCBQ-nAHBP<-VDY<Xqcl!(sqkwYfY<lnwy
z&WM#0mHD>k#c9cl9Me(Qp~oCivyCF}YK*NoGF+q5acQcuba!1G6u)26mVTrB)>StO
z=Ghu&YNLgN0Uv7>hiIJ5$xD{84f2_>_tJ~9*M(gdK1ATlLp7|`GlN$!<?@c0LSEI&
zz4p5xR(jIXPHvr?+<JK}QD*A%xbTPbu7mtxQy<yM<kc{z_FqSEwohJYATAP+r;S4p
z;xlXskq8|7;rJD*G2Pdzq!?eV(4?dolQ}G{0i`zuRli<=u)V0CRiduv-cHoXtJh=h
z<)U7)%op!+ip~h33i&q@C*%}4jf5Kjr?z~Gv&`6xcw9IHPcybyeYm65lT3Y1t-iF9
zhTZ|xp7ZCu&vRRBN~e}0;lwX3u%+8w>A`*t{t(pc(UOi(6+6th?5!E9AB*+;`i|B^
zo5(q}p3Q<YGwGfMRuVGZl!v|wgpO2^)~NM26i9yW*&w;Q6vJ4l;#laq*5$II5+n7(
ze=T45)}l||h92{)`Is$uduAyP0ypYm?M--xn!ke>JSK){BC4l?azEHH<z7+j{mTO5
zDtmR)0QS!0?Cn8Z<2S6iK=>!m2H_r7SMlAR)tU(3&xx=-tM!=Stez{Gu+J{3lZs%s
zAl8==#2>`s{ko^7V)3G&+kftuZnKKf|B<f|yKt`_>!tyWt`#V2w92mv#18K1^~AO&
zBpH~2i1jeEir4lG)iD39SnLjr?(|X^xSaG-TG|1Ib|3Z7l#mYv4VpWq!JJ~wt7;@-
z^i}5QrUA?umuPBq^kae0)n|iH4^yi6Y|j+E$hmxxTMGvnR6XWgK4uGpdTdD@jkjdo
zgzBWpl|L8s>g<?a=M@861)fy~>ZYe*;QF&csE2`7e70wxUZjyPa(f2qF$DvM&$IG#
zP}h}9_#kJQ&Lf)rg&q$dO=C!8wC)#%dH9(A8TAhSL2fy=UL)OK=rJXiuZT*o_J(EV
z<vpgd@%sK!T~F$>tq7D#rN$wPZc}mYqC1QPM}eh#!u`v=_a)&5<G|%<e{d$f^Q<Q=
zd3%3Uk;T%H;&25utzv!e3sw8G3bd`MP5Z^GO4T$T!dBKwL@0pkE?VV=3jzrbIei*<
z!OCJx`kA&kjMe2~td1b0Jd2gigFl1AnCzvGY~^0+_EsMS)uAWPF4*GU<@m)0`%;2@
z5OE07chs7`Ls|<${KJl@Ob4>MqVi92T`zevx31zeCT^MO{$Ji1B0`V*0whP^5*U&G
z7@KF<Zn|EFGB8ByO0#@W%QGzU&2=XI+$dm6lK4-RRh1<1mC+{G1RS+{LD7G$%)Pzb
zHWWoQ)<SN3EV6DHU0gNHfn0k1ffFO^s7GR0$8m9Qf`B|ya%8-ME<!aEd+)xU9za2-
ziM>1V2T27#-!FN;rw`2UyfMCOS0@ZS)@b3F&yW`rgfXj_o%uYi=^9;B`iK?G_oO}D
zS2A0x)!x1RmAl6M>$mRQyn6pSyF9SlTn+g-IgJcTJGwkeVWC`z#k<*g(GNB{G;*!K
zT)|DkD#N?{H8G4{(Laq_j{01@(Zzg`u==y}6z-t#|K-`i(d60lIk-pS+j)tWsqgSU
z72LK6y3Q(u<Z_2ESOgc#>&1<tAzz+nXCc+NBU*D{h$uyg;*DtUjYuLHiDHq+ZVHje
zq}Q)E*_=#-$XSL2E0QQAm?6%F%VUB_8DoO%I;6SO$DJi6GS_HC1z^Nw&VuE10FvWc
zGQWA&uV6jF+lZixD|M38sFB0z;h&7`d@9V7&O?8=ykE2gs%IH|5t*q)-U>{=jr-c$
zi#OlJ$91-KpjbhuMDbL#n9cdPtgf_mO(#6B+$i`&3V_H_>wwL9wi*(?LIg^QGg3l;
zSm_<{mjpwNA1!(<d3=`f<pnj4G1oOcvwmTLU5c)VbhJxN`VuR-BXlK$OKF$+;KvQc
z@3Rm>3F#sreN&)g1RYqI_4u1tSm3)@+kikj)+;0H0;I~+Gt@dS_lc`ACVjdECTZQ8
zuPN;AIiBO1xiaW1Kn1(eXt22zWKRmu3wnsJp-3tv6Xj8W7Jfj03_%4ha!3Unakoos
z^>Lez%Uh@>j3qjTteb?CRswBS_L-Tj;4zm5?W|@j78i683?F@QCBsoWuhZS0Zza#S
z67J7F-%6lKpPY~Kd@C7t5vtAN^Q|QOzq^&lbB=vt<v5NxqozZRcQ;}il^fM;s0)BI
z@F4e{H#ze4J0UrgNO}!9j@SdvZb;>Gd?fTE(M=_;E?p~KjcQ*&d8`d~Q#vk;d=Tw>
zAR|6OkfBAXbjv|tA^%PmO3=*OH!k8#uRkqjrBnPlkZwrs#<4;8>ofb{#^U5zufW~@
zg#JtF<f$<dC-g5#2mfP{t3XPF9{q@~kt!sk6p=dGqL@$}zHD9lOr>BHg@r^1ahNG}
zsKlXEtY?%s)MsC7u)8Q?L1JkrMKA{|Du`y%EqWAWxU(bPP(13P;>|5J61*v5tnXdz
z5I!QRL9$P0bCKY!k(;Mi@Hol0Bso?KAZJv@M@9@*H4Ve@o!BrM-3`QD5lbm_;p)dm
zqNEgGuMnsd+PW}WXse8>R+*<6S}?6s!D)MidZd`9S0fZ^T+yunteZzy3Jtg<4S9?x
z{gtx}J4#K4iF(5{#4#N->WTay5_IS8^?TRv=dqxB2nM|;k1hm_f&_zNJ`1b|&V22O
zK4ARM;~^=GUFCgn?&I!G2H+IrF4>?NtxhsiN~>JS4v|UFohnA7FFi`>`6#yN*cTfN
zF|Z0mWrhYf7BcrcI}&EHKr2d^e_M_0u|U80uFkPXKnnRsFSC18G>3o$#ThhB7=;dD
za^}U6S8c7rHib|MpT`M;vQay9x<vxhCjhyi!XNI43KX4P?WaPadH>jGO6(&l!!-1g
z8j;N&3Jt2)MM0kkml{E6bj{di^nu`mAyics{OR5*n@jc+?LBu0syPSV)0h*z*8~U_
zX!r<EprXr=NC{-hZ=z;_Qm)niuh*Dcj6dCh#n^{Xyxr%`daxITs+8=-Gy>#9O$Dhf
z+ABKUXC;&_YAaEKembuS>&e+1<Q^4WDW+c3L~Fr;1*<qpC^Ht7ZF{%NkIsvq?fLW=
z!eV}e+DN3o8n`-KMs=nven$FMY`WQvU1?4C;T9efx**)09%?5cl~x;hA@u?MMix}3
z`(bIw*+)J3bY38ieUB~(EEI`;)pQe7QtLA1bX-bzwrC$-OedE)Dv`}N2s1oDl{g}M
zUKbBjAu5A`DlUoEksjR4(kY1o${PGtbf~GvBy{m~Fb>)VRsT{eP#FWx;YOAo?EnlS
zeZ188VS11X*ZssZ*6EHU`!GF3A@(lHuSLu-0>BgU4R>Xlr?G~6EJGKFV?H%NRg5pD
z2SQHQK!!_o%-M}SY0Wm>{BbOuMiCLH7@Kht=}u$mn8jPkGNwC(S|T4*L=POTJiM8t
zf)h?lBK$<vr-1-|V4)XkZIifDf4v?#_<pjt(cobSvu2~p+T|fQK&ZA+rh8b~l>*9`
zfRd%BDlHWdEB^&eOb<!8{i9FzHdF?uM`#U@y~qZMTu~i-P}k*v;ueZ?_TUEw(icLc
zJBOBG@h&6>(>M^D#dI8c(ZMo75y#R3VUMSW%mntjc$MyPAZ`b(#OoTlNP@89g(yWw
z;S~WqEZzuVC+-)ox)D#Tz(5r<dL7~G6bFW@4X<8}6jIWT4Js;^B74}doXA9hIaTgU
zEGM+8K<y4-F%gtQ=pBF_RNb&Eq3)l4XJCfDVG`We3EkvAL9@W}DHLvl;qvju?sS^;
z4U*$8Ll9kr#pCSOk=gW6mv;?uC<es@!4A+Hx|uXcHPRRDIPOLZ%_d$)I5N<&zY&DX
z(5U0oQy_qz`8^^PZ}V>b+~;2dJr9MQb_G1WDZsy?!P7oSBD>1NhQs5Im;(l(NyXuq
zY%(@m&>}0gvm5)<Q;<my7fDJv^92~Peta`M5^eagxo+6bvhCuJ%FNw>ME^kK;ZjBZ
z`auLjy&#^&k0C&_aQClBdkTeyRI`PfP@$BdvLGQwuO{7WQ~aKV_>l@_1~wr*LEG^=
zkO5^Ej*_8>l2)ag<DuRQbD^M*Frwd<Sj`|ZUW;4M;0<74a~@-k<8Aa>HMnOwu7N8c
zvb*5Yhv~HX?F`7O+mSYHGu_Mm#}JbjJ>**##1Wz@DhDy{Xm48to0EN*w9@_3{y^a&
zyN+)mKOK^a>!)QN3rU;CG9ZgF(jc%0;Fv1y|6wX-ZwHAV*ToMJ&<}8YrrMRxn;4ZF
zaVpxsaPc*Lq!ZusZTpRVgB|^}Z>sX%wsIN6Xluh-rVE@r#}~iTv2q2?`UXDT{RKTW
zzU33&9yaj+ctfU6(xW|VCr51vomMVkx`C)Jq9GxMboB6aoc1yN%E%KGBGOCweQxFN
zU|3gvL?3r3?d%qP{5`TdR(?z$zd#Ld(Z@&h@g4g38~XSmK9Y2jAc@Oc`WX`eYg544
z6fic$_)Rf(Q;ge8?3bx0O=uqwv)S*<sVLV-PDL>uITf9>#8kqEO45U6browl*mj`w
zBu`kMd_z`Q_mgs0ek1$A?m@l1R^aABF1jA8yv%z~_CTD}vx-4Sd*F+U!3^f|comjv
zfuDmtLkK)=j#F(^*+WyX26zn<*GLiY{v%NI+u2>%Km3#mENPw&4b+e3TSw39sr<Fj
o!71D{x`{<rVwntmjVg5H-6UKe6b_~}p~;b(z;$|(C9TGP1F<ktK>z>%

literal 0
HcmV?d00001

diff --git a/.doctrees/examples/finetuning.doctree b/.doctrees/examples/finetuning.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..e2319fb2e1e890b49fe09d521cf32bf1062c8263
GIT binary patch
literal 12319
zcmeHNTaO$^6<&M2ml^NH>)4K!#HJ&Ny@vFRcWso!NI(KIL~Ad>u@Z|Uv}?L+rh2z;
z^kru^QX-H@<VE$C&I7+7fPgpN_z4JJcmtjyMGE4PCms;rslLv5?;ArBY{~X?)j4%J
zbuQnjIzJwJ|J3CX^+(P|fv~;x7PBnRw|JDwNy|4AmwR#gUV7;#>09Z%Y=o>I`AKN<
zRE~hc6t-i9+)Lj`<qVO>wiC5-{)vd2vF&>$yBHJWV&aW-UNq!bY{w3-b;;&++v9QK
z*<LRtDi9sJ&DUx6fxdUrY%Svf!_09i=48Vtmnuf(c)r4AdCb9(>2lIxvdQ8&v^z<R
zozYH4Z7ZTxPnZsiA~fRnn8)1g*QAvM)klY!t0C_aO(G{^)}yXe%!>0%N5nb#LR<JQ
zZ-<HJ@vz<Z!`1eUo7Ww`U;Kx~MeS@+ZNARjz~NE5Tkf{yTB&$kOo=DNXwnhKGg(){
zI#D#p+3ShpXfLr4^T#}lGC7^Ozde@`M!ig<igU~2Jc&Vk0ecruLbzYV->2~RBK}^&
zzQkA5en2Xo2AiHQrXd>;GwX#c5cR{_P%DW;X483TNnkNZR%2LREVxEI*%Gzb*Yd4$
zjcE?^dI{@s&4rFRt>$-m$Tb^`o|a2yh~Yag-PRYi4vPvFwdZ+pMqdA9mXKDm`et2L
znBhosht9<VEPb1)_>RZ&b*S0ceaABG*w6sx7#+tqSL;XfN7K@98ce%zL2I=wzwbFd
zvs#hB=>*JF?I6*c%~m@yLpz8g(p{rCC8B)2so{^Vt5Y&a-3>o90u~Fc7sQKYKg|>I
zIP0O0b~Z;_X@+6?-Wt@Dkrya={3H&N*vN)fL)KeL!{C-dQ#A`6|GaH0(*>B?ooENU
zl#_+Ur1EGaQrTE{oof-!5gvFl2zkWAHLlF}6{RERmnH|yx5+Hm_v|4*EZ@EEhuXmL
z<VX3m76sh2yBjdpf&Q4oZ^6j5E;u_phW2Rz5Ao$P&*WMM9F|4S;I6LMIgNtzE}1rY
z2h(@KW3@CbH|EO2o&Hu`Wp3wl7CNv^Z)H|N-GX=XcI4^KFU>(*4&3fHN-z7<PlIeq
zH}~2DTusfry<L|aLhnTGU+SLa(QOX4IG%0$s{^`MpJwL+dKZUtD))PD<ABb{6T=F$
z-THTTsKvlI`9Wei8rwn2hGXMOZd{}-wb%$fj}ZcSSQAr)amK@>6WV62PhPy?Fqi39
z^k-ifG_TzJiYQo!bkFpah=03x5&s&y`DMPd5s3KJ!Upcl92psTuyF+NfawlNqkWTp
z)&x^TpMyk~Ej^E>=DBi!5R@?*&vuG9^iVQ{(Gi?$k|{&k@AnhRgb+5t&LVD|C1o5o
z$GRQLaXT_Bdo8^uPar|DdmaFW8~7e&0mxQDKOnS|%5gv7SYo;<kHs4SkouoxL$Jum
zQjUA_m<=sSmx28SUklaGxzdu2T0$@pS-RWNucS;)E9QFVRL8eAY7MlP4{~FAv~m-b
zdy%tUz<0gE#&9z^49~i1uo+!7WZr};ERgvg(u7Ml*ai>#HZb#xkWUz^0R6H`JS^>b
zi@C3A^EY06`T0wQo7o3P<!Jt~cSl31(wA~y{xPxD1MZzsIGfjKJ#Wm5I1Zv~?KTV7
z?X{L4_S&oywHL2mzS3HF=DCIEE~#WebGhq>8#?cHZPUi_7byn859ajz(2j)*tifVR
zN&J4K0gzhU?{*2g^f~wPA&_`j3X^z=+X15SyIP1Q-*q;$#OsEfzsW6REs$r_(ryc+
zDj8<h%p|1r6Le_vJ^YUZC<d|ZX%W`pa6sh*OfNZ{vX>}kAR*IH2ntCwh9#CFd3rJr
z`AE5tY$7iLs;Y;MV<e&Cz_O3>ELKJ!!o3}b9LMNMXs1%!P1p)n{u@VkAwRl*<wti$
z{^j$04y9Oy+HjFQSC(9uZQVvGO)KP7sCX4mFCRXh8ht3W;n)$|B}1!4)-0$l>d$m^
z(JjN@LE>sXN6XT^dP_ZPQ-NX8c&1a%M{711?t<2!@Q57BnzV(%e%nJvv4(^eMifNa
z;^i{pK9t=7WLL!GmB*lYhY8E`2xWyn|DwQ$1;+hN(NF+esM@}O?e6rYF|O!)_Hh)7
zfo)g5OdMI7HwbTk@L4y<9bj&CJ5#sNJ=(1DFN2180OeEHw`cA=lrnd=^o*a3luHx;
z+96H+YHlhA2HybKpA_a6f88(EzEU};DniSPD#{QB*6k7l>rsY*bp{OF*C{U5F!-gC
zMXOk}oCTL#VA1A@=hFS?ya2a*K<AI$@ca5L=B*ME58~`{bpD2atAgV@kyQTB-j_cl
z2>wEb<5s#kXB~yhAJP&sSXZVfyCG~s&w#%G<6Nu2IB>V8QLa;i<*fp_KImyG`SkZC
zAHr|nzF$5igp>a~<b6nv-~4bl{MN<gZpZgm+dbxx^EH{i(7IB=sYAdw2ea=S3CcMQ
zqAf7k|LNvS5Bts`J?wRO*qYI;maysXH+hKHFlBmuZubNd);uGv-n@`f;^rLp99lw|
z<uy%f-o1M_5<?F-E)AuQZI=Y$`+e@U>7RwxSD{+CiY#v2lzwdcGDE#3>2Ba68;X`>
zfd}i&)7$E<f3CLoSJmDNAF)MGW!696ms!u|`0sl-f%m~V;PS54R<LL37mGX?bG45T
z_O%akPq%XSl<%W*O}`p^$@?@bLLt4+(vnLz@Vi60ftO1qtdI)bCG7I(K)PG=9{<}c
z>`(hrTgt-zKpb0TVdv2C-wLV^<Wl6pWnr18&5h;GLqTnuNacbmP4sVvG||^{h5c;i
zLaz;Yipt5zv$zD;akITjsZXAC=oF%B`Ojko4dpszVnRe)Bs_U4QB*C}rfOD|O|`)X
zTZi5@P(`Jxjz+Wl<?Y{v%i1pSv^<3?eJ{ctI*Ln-AoP1EhRABW#$rEaPP#NAn*|MC
zgO)}Tx)(YkXY#9iqpTB~Dq6C7F*%8vmKaD2MR0RvRa+M?MKSj*yf3}NoCJkpWm(*K
z28PmQF}o}qJv=^zEcQb@F*N|zamSCe6CcMz-$$(s?%(b8R32%k@_4oc+y{88AZX^F
zN$B8fI`&P2YR{+_OBus+L%5O3lU)qs`7y6`-jmZ*ZHHG3W1UJe@J2Qhq2Mi`Co-zx
zi!k8hSqpc#YRM9k;<*d+<qUa2+_IbsIlUek4qxLC=TX@d%qc&|<Q!kOW8BxWIEf0P
zF*#=OPSV44Gkq3Pbr>i@&Rn!Ka2&{Ouq9fb?NJ>No;iFf8$oGH5_k1ybv*2)@`=Gt
zgNGpskdRke4j$W~2H2a4SHIaDhE}7DsZmmSZm4Hyh?Q$cM@HmKHtzl0PYiYpG0PLQ
z2k2fTAR;7bD8`vC3UwfH3*Sdk6)7JHx#@?NVR5`jTA0R$+APa)1dcAY35nQNp%7wR
z&afo*jjBl=wV`g)R^kRmsge+^YET#E9&>2GS=FMHold`M3L>6ZzJa14P@UlGfgf?B
z&Z)G-jEyxrvNM&P#T#qHIu57ut!$MOg{i>orm&d^#-Z#EgHGpUs2@XFBb8B}p<kE;
zrAf**xksZ}IP&8tAH{)ZpG!yN9N8N*$Lpz=4cCs-n<p;HnScjI`36RtnW7d-hzv3f
zdA4}DLUYs#He%s>>aj4Dr-@Q60F-6v5!FiZK!57;RmgKD?3zQvQ%g1cg<^Qhv`J-G
zdDyaf+*NbHA+$y8Ixycx%sQ|lD;DX}q&x?mbdAzD%MBP=xz?sU5pAGcWZe+R^Bl_c
zQdFq=r212l%V20fN?_Fy1oepe&OVXRnJgFWIa8pr(!@(iW^a(P5+S%(lK>CiRD_$o
zRA+8r6WZgOw%3IY<bL5KIf^Lh43XI`$}%w*%Z!mqXIE3G44y?TSnvi`AUKaP$7vH(
zs|I(O?HH)}BEPFvJ(m~!LRXz`=F>LiDEHQIf@j^0D6WK~R5hzgyC+W$_C@eH>Aj>Q
zCo}&8gNN?go@K+DVk2hADdEqcX>-^HbTLNRF=Bv4CAw7GJBu1z%k&d3*6A0@WE=9)
zVNw*X4j09ZE-YLv@)PyDL=|tsS#0%)a{pBIcl(ufuu*8kUgoMbxwfaukXJU)UETy7
zStzE)uX5s_Z8z~2<83o_B2NzNJ*#WpQcf#<Ot<AyH_?&MfpiOY^^QzM)Ni4Pni{j;
z^DEOJU1{PU@sap&#RMK(xl8|yQTtIfPJ`&FG((-2;@A`)Z7SkTMYx%WC$YqAoGJ04
z(A6N~V>N=xh7vhlw>7npY17UwRLQ=x7ILWtU8XekmW&Rn1lx2wzEf(bB0V+KSW-6z
zMr;(bDyEYI<0Cpko8jwD5091i0Hna-D{ZC7cHHzG2Wk$5M#!J|7HKis--DRnNpHb<
z;cgag5OwMi>@}OP^!o$N$IE7AK*w#*bP|h9yhpavv?w&?UYu~3Y$`L1oCLV%tu*Ph
F{sY2Ma;X3S

literal 0
HcmV?d00001

diff --git a/.doctrees/examples/index.doctree b/.doctrees/examples/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..1988374f51db852c052f0a17a4d1d0b5032fbbcd
GIT binary patch
literal 6059
zcmb_gTaO$^72frFc4ubqzN{_T-b{qT+QPdN2oC|t5)rl`8Zt()fd??Hn(msZp6y$@
zs&{7}ECr+>k5q~jr4g^Z@W>N?fM<RIe}ES*;tvFg?^IuB;x!ZFm8Q4*RMn|-PJQQA
zA2)vc`)f<)PjAaexWVLzIZhBdT&B9|gm&WdK&79jcYc*VOK<B|%uZ#P#5Pa$5@y)K
z^_-Xo>C;qi(RAf{@+e=wCb_NLFesOchFBGAPt)6?r5nms9<L4Q=7S0IBM&>%989_R
zgimPi+Q=tqCdX>Tu&YYNu5MYSm|{h*=F;``Mb8a*+D`^dx0zD0J4h6GlNKwkBT3>l
z+hbCKi1$Xo{OqmiBvJJ-VD@p$hcqYAYl@9%C>7gczqc%I>3dxf`n($_LBQkgX&4`O
zfAqmaFFY;&K{(mXgzEBK!LA!Pd~)PFskk9F#7(i144`idO0iMQB2hO>?`OOd#o>wT
z@Q&msJZ4^}VAN4zM~d*YLqEJHl6P=Aod`m+5%0J`=gfS^;lShmQD)ga{aShEX988a
zq>8xJ7yD#(@dj)y-o#OU4ZpYWJHYP_>?pn=4q-n`6?bbESkFLKp=75Z=$-paF_?jq
zi43L;M)qDXP>J)s#l@lHY~<qHD-Wb;hkg_$fm1t=48dXvSO{$u^E!m{vGep6H}qC<
zav1~wL7+Hx4e<JUw;a9y-jnwpKX~#u{Y}dBh8x&k;_#tM&H)=Trkg&SI6P7!?X3XE
zEHXv`>PKEUNT+&LxQ+wAoa)VdbxCLji9g^mchcUHZW}k0Vc?x%ZX51h2}I}g*cfHM
zw}x~2@%*9c?c4^3r$BAQ<CM*xb06!BK?^=W`lI4lb9+6R`##@(;+MjDn)daT(onYG
z_Hh)tfs$k^VZ%JXW`YA;!@d|zg_sI=D#pDfaXb}X2EIb6O$e9FD<wYqpGs^d174zA
z?>NBbVI7;SKzuefgIiYjf_4pdMI3mE7qlNDq^TrugHhqhby9`=-+eXYU(S)A=Ey%P
z4n#K}hH>Zd!zXu({VGDW+aDI+jokj~+;0Cr94?n|7*+-<(D3I)(QplD_|t-DFqX<8
zeGQPVEVf4$W1c$-D05Qewz-t0{mvV%tofI(X3hQAZq4hZHDi9tV#o4P4j`G1N(cMf
zq9*OYq<?({lXhN9l-R5kVMA3)|HY!xA3*wlzk>7!GvT3K!(<g0W~sHrNFkx{#|1({
z>-}JeSRSAt%QIq)Qi5u{yS%itIIbug9%i^2Z{;gzK>p}Mj$NtZ_<3RDr?-Vtk^FAA
zJ93ps21m%9-A71EKJy;i%;QBtG(dvt`pgZ=a_CN0m)sV6x>YhxE8*Xm6P`*Xlw`Tu
zMQth?x>*QQ*rFm--0sYAJHJlRtJo!p-Bf?)m35<HXM8|g?)SR6Neg{o(mlO=^b$?0
zEK_8ZdiJ1TTJWrnsam63ZV8SGc{9X0;N^#2)}$}XMGsDxm(0{+?HZ}FcYk4fQ@ugZ
z`(|#ZUcU0R1cNF^pt9;&KlsV}_a77?VeDe3MWezy;>P$Fn7>$H7#B(yo%Vb0%*F;@
z8L>K{_6&9Ih;ri;M^roQ3hWiDI&F>-x))P@uVhL8Wp0r7RN}6_qGA?E&m6mj+%}4l
z+)Pi^P@!U8R#N73Oof@ysV?g+JBg`dv&#N#qo8Dcnr=$26tWB&&)u?<9L8awI0|sR
z`cuFS<!|BcdNoW`l&G{Xw)?s@;^@Da3S(4<Efi#2Dm0V$SdYU{K~z-FlT<HvQ@x&v
zU}4{5OlaqyN$lZk6BVsRJs*uilm?Z<XQ{q41e+jKyf*k;Z&HVga@v{<!`M%s>n(|D
zKC%$|DEmt=ypd5j+ApG5r-+VL(RX6Md_mrb_VosOoG@0FhgJ{fyrA2HdF5w=dIT3%
zR7@qZnAFe>hYykww%bC+ruq;wV$OU}S~xPBlS(9?8^C53@<r%SM%S%G4G-TvlrG+H
z&JJ2Uj>FgzEO0!O_Gqo4o2^p4omsxdrD3?F`qtdY9E<TE;Wy}k4BtwIV)f10fw{$c
zhjak;q=1P$=#=VBj|HOyoh~fy;QInP9(_6Hc8IbC?GccJZK%EK>s31phENk0ah(DX
zVpVUkM1@vG(idF7ZPQ8o$SNTT!x{qm3?2u}BZk|CVjO}_zamAFCr)Thd;-)CpG2YL
zR(+*GiLI;?SGpNww=Kt&3_Us&$)#K?Tm^2o0c9kdBMY}cO+L*kT0*@?Z`efgm^wk6
z4RLdD<m>3^%Fo`i-X(tna)J?n=()bD(hqhH^j5?ptIVpT%mz^g46>7J=<7ue1X&(M
zXG(;@Q5cWlQCDfE5dbqS#c_)R55&C<=!b8;{cYHDD<0Z&?1@sb|9Zimx+#L}F^`X2
zu7+k01ca_ozK6|PiVfgJ4z<2beG8cMt<pJ51uTg>5RR61N1g?WO)TM#1gdfY{9g1z
zg8qi|8654S1YRv+sLSR%`?S1}Y8G=dbjudFbg*&onZLox#)MdXoJ81yF1}0~oreMR
z+@K)$0t0zmxI}?MCWE<YD2{+#FJzfB(v{V53do>8ci_QikU(&*u*X#!vyLr}!pXC6
zMMrVhNWGh1vH`2!FpFtZy~2YNoZxl;OsdK_2CA;PD0r?f&5lXLob-7z(9JCVz~O;i
z+*7&mCS@rWk2vBPF3NVH2C%4b)d!s6Rw9f4+qnC3Y+RP8Lwds{UQ1t_pXvtL_C*j!
z&355i?-tb>y=HX-QtZDFD&z|;k@S}~s&~}me_$CUHfXY2N#whMx%?ad6T@Y5yZm;s
zGv3OL|8w5PFW`61)``9}qj`3lNH>pP5`O%5Z1+M@zra93$1yCHn|hS4E}!Avh=|$y
z?eWK$I(~+S_)L5{UZ<}nJ$^`!2lSZGBf%rlEn2D0BA^EnnYB~#+0-nXnk7>+eVT|j
zX@cRqAwCuN3{#P29I4<m98HCj=*_wh8WA^b7oJQYfesmkLTxcG15=%n%>%^Xv;H9T
zO5~cUSLTT2?A8p&R<WyDbZLe^vL;ZD7<hG_S3ZLKFeCzlV#;{ycIbJ4KXgDTu=agI
zH|U?kaG#_fBB<du65$c|#u9#<ZCK`m8S;8bHjcHP-^s!CN90p&hf-D^-KieaZR4KO
MOHdCzOVLjK7c;NW-v9sr

literal 0
HcmV?d00001

diff --git a/.doctrees/examples/medical_finetune.doctree b/.doctrees/examples/medical_finetune.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..cd32897ab1b11ed90f8cfc471c5981596ecc01ef
GIT binary patch
literal 6312
zcmeHLOK%*x5w<Lk9;26~U2hE7PI_%@JVK-%Z;(S^K!=45VFSnBMPLL00;9p{<_sD2
zD<s>J0vJdR3s?Ye+d1Yx<na%ZKbEh^e$0%W9QG^&V>OG#s$xA^_2<D~|91C=`)Bub
zDP(>TvLwligzMUFBt=|hJU8|C_2FOYclF3_Dt4}mO2xdkH-HffnI?+o^_$x661tJ8
z4x97qn#V>Kd3Rh4#F|)tQ;)=^9T;g+KDT8zPG!zbmGhc#fO6|KzaZXoJKxn_j<u43
ztEt7l-HbXh#j0Ivq<dv=r65<oswT`1nK4REDuY|mEmmcsN#gZ5Wm*G>??KKo|Fw}+
zWxtrP_)PIB;Z%0ruo-RDVo&_|#fo@fzaEPs<6~9jIalNJLY<9&{qi&|&f7l_PLI7%
zV}8N1GUa-l@kGWfji$|w!z`)AJ+UqBi`8m^tM8=Ja0QG>TEu5_TKe79+Kn4BD+^@;
zeW{1TsVcI-D3<GKp)#&?(CmIa{YuT8`MzaJbJd`x*%Z=+OwBsX(J;WjmoR$T9??e*
z_M91}xmkjKRheQAbht=4@n>|{e6;d3w7kHL6dCc<A&kx@DUbZJJ@DE?om>m2N?CF;
zZpa`=)UMiNB;B_S#&9@H_%z6v%twb$-IX2%P8~Xc2Ap=Z1GiKnGH-AaJZw#)9=iRO
zVs~%gEa)Means2Xfw+#Gca0V(Lsz70crPph63a#!jh4(1d|-m$15pkS9Z@+AQl5|C
z6=dDFhr!87@biG>2|0=b{r<~0A<q+iE{zyH4F8~u{9&UOlIl_Lt7&k~A)VGik*AmV
zk9T?sOc5Bt105?_8b?=}$J)V4W(4iCfM^C&nR4$yo@~kq(Oo)_h7h5w0BNKnu0wYP
zLU&H!6nGSn3pJme8cDU#oL6D!{zSwT!Ibh*_jOWeY#I@=iKf4uUZ8|VVQ@X}o#Kb`
zb!L`1ypSYqvdO41@}uDOH_yL$8YG;Ty5u|w;v%2QSs-(1BunL=7-dKA5zzf2ltk)$
zuyMB5@(aG^qnFS-6_F30EUbEK^y)q7Kl|-Y<0su$zuA$bd-n`z<&nH2(n(|u2>slD
zjXGK$!7;}6e1wmBX?pz?XFPAz<Q&eRAOtg2RHe*kK~`xKOn8uxo66jIB87TTsv_pv
z2R8Y7uNe|GB%G+hf!8Q)MyPl3$-`FgjFoWGp7(r^G;p3cllyBr(ZhDf-4+}Eu4I@r
zn?vr58oG-_bf?~igx(VBSo9X4ZLg?|T|@}?nGxQO&L?`{=&(Q3>GRZ=8|cnO8`avx
zCtVn$my6)u9G>*i+yzehyMU~()Sq1!rSq_jjq9T1tLk-8>PpddQ9|Bv539arP>Jd@
zRG<5=ixLXgby0E-@k5J}ZxYc@jrD|ypP=ItpQE9Cgx?qVJ;v`49p6UE@>#`ZXl2*R
zOC!*z+RckBeTw%Y^|u?P;&`RM<#i=~CXU4eafGICweQn-N84_+)w5<Tp9NUMlL!rc
z=Lll=#Hw%j{*RDtKP!&xr>@IO-J_A7FEfbe``Ljh3`^@5H|(&*;R%)%@msOGaiyqR
zj=J7>yM_23*>2#uV(=_L)4$(!@#rjbgLVPx`7KK;G{@Z-V6CW3S(*Bk*n4F+XB@pR
zGli<Lz#EEdgFY9FEmag~iqJmGi`uS?YrEw|h`{#@2*YMssT8XnQ^XPV+q4T&29Q}@
z*7jfuGI?S6+~#|`L(_%4%AyNueCzjiSEI=<qqxXuW}!jwo`>+1J4i%tLEn|(GT_%7
zXodi<?6%_6SmRywmP4I4?NG3^TMq0#zmU+PVW!e8(!dT9KB;D)w|mZ14lM#hah3tH
z33J4pu0-<596F159v2C);B1wd9{=)KOMKs7+>E$Vg^C2ax0EYbfJBW=ZTGz9=co)E
zmD)a7+F2qYwB7YYdjZ;`_ZK&oPTJda0Szq)=qGitvD3y5)SVz!Zej#vKT$j`6ke^o
zgxi3PbZTDnZ4aqKCapJ!HM{HD_a0<#O8DANQe|b-`6IOI0BEwf%2`Tw+jEd^=Fx*n
z&8wt{5W~Q_%`eJA^JxC0bCK9YZ>5&rU-zO!YBzl7q+D;~1TnXb%V<P~3px<p!F{QR
zbIqok8+}G?Czv2pP&Ws|5VZvM-omV`TfSJa`xI==n;ij=NmIYP{n+l7yo?Yc93hIX
z1x&UHGC>O!QC!%&!kAJ&9ghJ<PwO&OSg0Ao=?-B!0f3240@EdgAEho{9)I!V7trUf
zn#N1y3DT4QX-l5EAM)2Tu0qMp)bT(z7z>l7V76)41R<0dQNP%*58#kl)J0}@fyigi
zD|=gC<_1lQkTa3u2j{(&7UcKaI%9~kKUN574Mlz8)_xgb-o-E?qmkT$STM)~!~0-^
zmYoW5`m8GJD>rd;W5gy0T)}&aSmsmsK$8>>C`;&SvV<lx<bCjBJ`!nGcviz@vI#mO
zZ7v~!;M{=6ofueW5l7RQMwpFK(sferH`7h{YC81^?aHq5{4FeaH@noPSB`U4^!%tW
z@9n|jo<_o{-&Yg6;gb)d9lk5`L?W6@WYDp4Br~XEAJ>2{8q8SXXP9gG<iCfZV-got
zZjR|o)8zC=%cyST0QPML7X!QUWz)UUS64Fs^|6v^Kj6XIDoPdRA5<D~gZ*BTOy+Ld
zD*ovQ$HebR$L~qc?>`^q_Yq1T<5<~)g^SXo@_#Ai?;jQK-xcY7+8(rPRI8xFcNh*6
z(0_|l@l4xI0+`Fvu@I+MT73Sl7SF`Cwmb82;7;$v646A40!0s@bl2i}gU&-d9?Xom
z$tDG6!%fk1sGlwYJM7H`w9S^QJ9Mx>A7u-khCFm<nO3ud*8`HV^CT)cTXB)5a6#Mw
zwRrs*c^Kf|LpR^mzd>U0qc><=BbGeFE+<Fx;H?hqoO?^=aatu5elzlGjA&6dnhZ-1
SB0F>;qSFcwl6Q4A3I7YiJqiv0

literal 0
HcmV?d00001

diff --git a/.doctrees/examples/raft.doctree b/.doctrees/examples/raft.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..66395c72715b8d98af92b56cdf7d6710017f4486
GIT binary patch
literal 101107
zcmeIb3zQt^bsk6&BnW^Y2)-W>sp3#1fiZdp7!Z7n9`ukPfFYg&M&{5o-Ky!XnW{#2
zSF@{o1|yj^6-zOZv}3O{(po-|9mh(PUB~OJljB6TCE1dl<k)+hO*U&=PPAF;IBR=t
z$8nAwZ%*Rf{l5FZsvpx+#lZ|gYbbFzQ&s=#{`cPhzW>*6{M2>VzvDXo)44Tigi-zS
zm|rf}<8qL!bas^EQo9z^TPvSjIq=by*H-p-b~gQ`ByKlL!Aj>kd{GLcYPlKIS6*4^
z?BVCFsG5xF_BSO#sTIX_+dRB6+#cTa%F6z5XXnON)T#!lSDhO!OdX$D;d{KkVJ5iD
zarM)Wuc-02H++oNS_yCK>@3>ph1)vY^{mczZm34}VC8Ik&hPB<Tdig^*KT1ooc*?_
zoN%T$m8yP{;30m$vF_K@Z#&BEhWlgAFI{Q|^ZcgWxvAwZ@YPCqYxv-)cZ3gg-a8(~
zwP3v2uGfR+_)^@wG=BQb@oKze{)5RS<7!gl!DYYJs0PV+)1PmR)yga3zHoQ=&Tw0M
zF1%Sm)`4cPdC=K&*qg4mnsK?Uz$l=(yYI1v4f;I9KEnsjh7WT2;k&Tr@F9TnVf_0D
z{(TJp9>99SZ;^FeA3hKsM6-gP%FfPCfY)DW`i(HXF<*Hx@R~u*Z(j0Rq2I!fz|%|e
zqQskPN7a@W*S*uH{WDYEQWz(J_i{BV1@$B-zp_7UwHnFf_&6alwh+e))nKd?*T(1L
zW-S>nj|bz2k55fEK5+Wt{Htf5f9?bIGnbZ2Q;mtr`NzNg>6zoRhe{_dJl>vva;`Z4
z?B&W!#dDSLh3V7PQzxdz7o%W_-5lUN@QYWE>jm^(3!0_KuX=6FB(7JN#{{kRhj(|5
zyot4}I0%1h8p1%+oj{QQ7zuhXG!ov}*<lv1u=(!P1J~6H>8@7atj=}}(QZa7ozf-%
zQgGuWaQp_18?s=a>D?V~O&Ghib%7}uiiqz|4Q3GW8AAfI5t!43AgW_;O95GtcRKK!
z^{Bq!%{Swkcd}jc>)!DoD3e~j{TEK3JbnPk`f59B28rk6*CoGM_U7Alp&_podkgKT
z90(Ho<w`qgNw@L5JXr`p&VDOe47?hcV08=>h!?;vjqnG(rNB$Vc**0x=#40*C?7Kd
ziy$lD^H$T3>TUwk8&I?g7Qxw*h8oSd6qMV|z{8h(FkMetO%VHu^D|z<Zv@Rr&+NJ1
zyc{i##m$9re=ZrHm^gG~?9jyFqenfzUS?YiRgD%@H)^cK$d^vZRq=8mZbq%J1`sKT
zG^gs9g0goU%yOnpfF4jp65hQ9^i2i2qndIyjzIQJs@c)QzAe>ses6SiCWW(JDXzDG
zjbPPrD>Nh_SVB}J25BS#1_?Egm4c1p%xZxkl4Q&~jwe9%X7U(&TMX(EnVwKX=!07P
zYBTW5%f#tLkXN}E^XGh=1w15x+l8}1T$mVp@>$}2*j`w`Eav?Z51B^nB{4n=yb|a-
zxfC@7hcy-3DbGjEq%}4=YS?<X-9fRLh9X$>cF<H@Ur6)k-}z>MjyZe&uMJ!5H(vkD
zwSx?)L;53W3|ReSB^@n;j&^H062#$AHvpR;Cp~!Ty7t`KXPN2EoMl7gyL%3g9rlj-
zEs#dbgT!5GMh%gdwV=1^dxz|w%Dz-eu6y)SKfz{Ls%06oui>DDXvEWuC(-CJKaBlk
ze7;?+7Q(R5tcLTWURlr8$H`hxorobsATRJsp_hoD>&a<~L(?<>%SfW6mGFg69#Lzw
z)XB9JgX)t8-s8BIBb~!TI>+aJy(Q-_gwj@2YFGW{LCp<R18)f@npgGbf~q%CqS#p<
zp^(xLp{z+Jln5Jx&VWM_XEM4rqjfcI1{J6(L3w1%o544OrhWME#3^;ClV_@R5)5VU
z@Wj|ryb6+*d8~SIhdt)x;MD~qJJzIOWu2Q-+mJ%)A5ZTz*xHVu)(CxoDb3jK7_e=I
z4-9<gFzF}L{SPOap8zK9)tDsHBGJqSz#t2`YY;=e@%rCXIP#6xKl`Sz1RWSWK^Jc!
zrnpV(LW;F{li4&w{yRIjRHGII-fD5K8ka7mP-*rru$?1HzMk}6q}q2W0<wdE>v2S4
zEk|YiTaOdJ-Aufie<^szn=*7esah%mabKZL9j8Xit5U+&sw8L$Oed;An!wo~weUd#
zGWDy~WiO=g4$4jZWd-m#l3AllX#tPtWlHT9lumSJC*~#vL4PT3SIgd9VAisXxq}fb
z%K}mpMt2Ks$T|lVV`7J@DcIGVKZnU^wZf~$ijz^|F#?qDm4YUPEYGjSZRjZgR8;dJ
z+CZC#>$0?^080a>tj8_0KCmiUG-QosGAJJ`Pn4QTZ#iyL3X*TjY0&Q4^_(y}%gZLc
zk(boR?6>^P$gW)%G)GH+gH<;<1?&pfz$qrJ_WV3#FlbTO(O7NHrFy(HvP<)gEkyXH
zM+C)hU?}=$$q=g&j2wP_q3tigG<bayu5@-@u2m<K21pf#LLfS{hs8h)48JUg$AhPC
zN1uilSy80(4`)EweklTL@>@zA5oH>rCi(_WcbI%24(OL!oS0koV7eqde&|Er@i71Q
z*kgvp%mpAsu^Is9DFu|kSbUuL;9TH2)~cm6XW~D21SOLOY)2vz)Pp5+a4$vGD*1`-
zijz^NH;g3zX#I14_~C~^qxk>kP2L?SU&}59PzAleN1ptr)^M}udK@2Eisofm4-m9S
zv1Dm{O`RcbMP0{8_q4B_lO#3WuEV5zzMFfVY+=T=AY)upm|+i?-_3KtgufB~CUtmo
zn!I!1Y4TT~PH8C%MyjfCe`vXoU9~c+ZLeh*j7zP!x$Jpk<9s(BhDEBqICH`Xsx1vt
z<oa`z^3WdV4ZS}+aCo9PZIAv3>V=RSlwsItVAsM_9IOBHe60S_)mVM!KuH19d-PF)
zE?%lvW4}C>gr0^VU*Ma95h?SuFsM<v5&Fn_uw%c&4yPz}?$B6JNSj}L<1p)p1;!vs
z_a`Y|+OCvkMVoz6L+p;yK0Mbqyn8)wdb%|3xeT7d+a^41);~k91#r^f+<-S01uFXi
z)9hG41Janng32ShW!i?+`4NzsA_JiaQUzqQ^}9H5o(}Y;v=c*oCr=jW`LQ%D_F+S~
zQN2`c)1j@+YS;QS-Tpk0Ykw$^9qQkoot>NKV8shs#VTxSDHY%O;&<{*5te{j-)m<g
z?7Z`08!o9uDF4cB)DpLmxBuetnFH@Jz#aaPTiV&&OPkP3J7rOCsFkWR&F>yOWtttg
z%_=Y8S0kt@Af#6Ls3s)qYXG08E5vU`3EUT&85mz#C+p{K<+PJ!CpvEG5xw#`yM#<D
zH_hD29al0O&V&SXwksT2>Aa&r6|HiNV>#%h=zWMm`&|<-Xb-^>gCuKy=b-hmT+kwm
zdQwB{qZV4J-TmxInyS0)vC!iTiyu`{9hYITSy&1p8_SUr2gZ~6mJtn#_<gJbv=|-p
zwAGf;2EzTE5!zUsMb%siyaibM2U?RAQZY*5!Qs<%X5q>wdf*fOeS_JZC*R~^%r3p@
z_HOrnxMy}l@iwGg`Dwj-W?eKbcaC2r@Vbea?oF`F>Xc!;t!82=v~t?+S8syt+uQY#
zk=Ak}7=byih3KUsabe^jo{zxy3U5X!NbnASzl?wJUk~5oJAUuP@r$9@D{1_jYr~`g
zxsyJbm`7;~5-QDtaLxu0Wg8(JrO$Nha?|uAHG+B>zLvWFHfkb_OHW|T^QPyuX|V}m
zS9+6<=!}JL24-O3C%nD*9fn=8zeq=%fU<)w(I}}VhVGt)`885q3i8(xjKRj7oW$fH
z76YLfkab{amLLq8Sk<y>K2}(2wZ+z3^67o5#UM~vXKJ7#NZg1QDq|3_AHYCLu*d)+
zT$mxA@w_Rt;!6w+duqLPkW0`FBqT$8ON`kfz|vL;8LK2fj{v;52cT1WC9)0F0m?qi
zy2>mCk6jSJwo7+8SBpTA`T`-`5^ce+!nQ<e=ftRHG~@PM73@brOZ?_3__x|k>J>2z
zvg|OyTxre)dqI~o4Dnz;VtwMLu#|fe^n!h+x`98Jv;i0E_Up@bHfF!^-8_00V>3-#
zG@?g<6kw_P)?3$$aS?3NVgw@wRD&A%r<lxYF^%y-j5q<0P>0b9h`}WSU;D?xI8jr9
z6Gk1ai_!2u=C))8n|?{fazSKiQNRNjXp!SbDcjIXaT~*7*O!7S&gWok<TaAoYGgJ2
z7!aztECOTfRT)Htna03fVp8H){Te-j(DM*6h>hzWOp%NluNuejgdqM1do+jNu@vb?
zT9+8%r@J`lO$!=8uVh3Twn(;AwP9;)=a`TW_tjEYEdY|?N&jeyG$5)Wl&6H~HVKQW
ziGLxrmWw{OD7_Zq-r*;no=~2Rq!E;&d3J!4Y8VJivtEM1POFuWHZ=&a-+5swB&|Fw
zTz@sXBxobufS}VL7v2HK52yo6L1YW{bWk`*P8(sxEu<g|<OH#JFUn?~hlMv_R4N?I
zZ2MeXjq8jIoI;2oU@+-Tc_&qWq+6jB1^DhUQ=3cIrU!2{Xb}B0Ks&%b#2p?K@~lM*
zVap5Zl^9|GoZG==vT6pd$?PV*3sBD%Fdgk>3cO=Mi4m(@$r<>IN|WBjR=XU*ONIHK
zj~n!Rd(Wa3f>HVCVsyC$uG0z&(#6F9!8rQCbNH<drv7|=0k(KCQJf0sBcB1|6PIjR
zjd4(cN3hlSy8`bVNTCU4SuF!Bb^N~-Hsb~$wy@0EiC~u?Y?rf#RQv)X?oI#)U?Lv^
zI=x+9H{M|>bZpUxuTAX=Ed5aAW&#Wm><Q=LwGzgDUblQq9;G(X;6thbVWt6ckkdXj
z=DjfHJ--M+u+;+B0MmE{|MI_D@|oQ_&IOmZj`OYK94ExqaV`?U)^R>mMA|yeAy?z5
z+B(ja=lohY&Xv9&?yLMX>=0R<2RG25v(oEadOv#k=pc(y<0S~M)C+?$B%0B26~&2E
zpLV^7@T34%(g41YQ4!W)RE*}eeFy;vHH23;5nQcYR0v;WP^5{gQ~`=z;nIxXaYmwB
zx<*dfguc>abxGNT?p<wu#^|&92hMo%D?R<MZPhuYtvY6AV%3?xdLVIgGb3a-?3lT^
z-K{*21sXG&wvrDw8W#!$u@@JaBvy=@#fIMsC%xyZL7Ou5)b#j5qgANuuy+INv2VQo
z$vn;6Dt|B64$y{tM(@DC*Dcq`4>Q(t)9f3z*>FIN7Yf>zszLDXk3g|yaBUF|*4Mfm
ztF)vVP#gj%@-=k8`2TVP<E#e8f3vF^ioymMrdd{CJW(jXz|d&R+9&07AYO=UEjBc0
z%~WNzF$snP!3G19CjpZ;)!u>TZbaea^EI5)pt(c0%!ntZSys^8RVdWkwW7H7ipcsX
zg~_CM*sT0+tbEXGx3EI)g`L+68_l;c(=5lr#DW3Mp@>AXVA7ii3P;lm8}!;O>{#xF
zeLyd4D&N9Pv#bkyp-_PLuUHNik=Us>ZqA3nte8X}#xhkd9jv(}SQxdL4R)O-5`I1i
z6CP|=%di?Tb6A~VFzcVfHea<q9SC2}4Z;f=gtPfTXqq_?uI@&td=Gd{<#qf>dDjeg
zbw^LiRrLf6A22GxVMOTy;)F1e{xB1|27)H+i8T+NSPR@Y+uXsiog0qdrs0@ckM)ss
z?jKuC_Aqvrp`WOEK??g`Oe?D_$2inpNp5yEN$zt-o)Wt~eG%2Ti5yBMauT6oJEDN!
zzXM1xhdCRze&v|}Q-{r;pY|W_BJvn-qu;>U*no1`3Sf#w_Y@&HQd)K_!t@vxITJ?I
z*zJ9LOx7XOCvzjycML<OcMnFUE?`--%J#8Eupe6|f_-Nff;pDZxi7H+6a}psV7Cht
zgIrjU_{QCbOmUzNLA2P9FAw0;o;V@n@XqF=zhh!-)nK84@7x~#e6BsDFTSkz@C&+S
zSGdrfqsTIA`5uh9@Y#tQV_-&S<yU&f4L@vz^UmJh9?VJG|H7@~U-tAjRQQt=!Fjz7
zGZX4BsVu?N46`Q+nmz}?X10!lqBjsFsxQXXMVQN*aFGd{u!(avyOvyM0l77G`je1k
z2j|{$L;HGE`U)FLzK2Sj!=IAe3ltMJBnOU3<`oYmy%;WpfqpB=^i%!QxDG#KIz1lj
zN<1?z03wF1UHBogkAo)S>0=L$Qw#@33ghZEKa`=tGL41|FBj8is^&fI3ObfcTq~D?
zWq6WD8gT-yxfqNHZyl)z3zEMN$2&QvX4y@+dl0?J#S=tsZf0eMvCb}TK_0I2ERaV&
zoPj-4>gaI7BLjat%DV%OLVqt8g;Ku_hjE(t<iMyOVuCq~M|JfU{%yj0#+x<sYq6=7
z71E1D{2XNJGqY5QQ}jZDOmtSL!X?J>jI2Fy%5L+YdsdyY+YF3E8Zte@26MM_d+`7E
z%w{;3Wb95=?}1r|Sb5$#6i{wB(<r;)#tfHYpjO(PJ$CbT*L|cb?&f738(?s@JM3nP
zPHpBYc_BFy^Md^O-r$;a6bjw+&v?&4G>4WWJePqA;-W7H2obL)DZlwN-uV#bDaixt
zAncL%Axf2@E+0mYV^7e@bXP8(_0lsr&<+Wqv1*4*`SBx!)6hkWY)o{|$+<XLwKhfu
zsnI0{1kUzjDzZjZa;U@GOGd3Bp;$aZQR29UPXz`DI)M8eGB3kEdOpy7%?HTJMih$l
z;xfC{4A9R>luuz}POufXQAkSjX8nZ=2{ABctrnzNt5qY0hI5p-6B!1FfLeEgczwPQ
zIIFmxHK08MK7jnkB`g^y8F&~BxjKkS5`7uZqXHGwQ^F6uTaUH@6qmv!3A_(VftapG
z5^S@SPNki114tMR$_9=ydF4RMwz4B~#;Ofa&U*P!X3%aTrl?MRTWtlSD1aj`N%#oD
zu&9-&Q#_UwT`ys|8T71z9290?0#MqS{=jTEZ(4lwiG)0;cqE2nsE42l9>cO>e5#Ul
znb0cJiJOvDDv?Kwevzmu**6VeC^;s62I&!O!w5J*0tggRIixojh%cFghwsw?FW9Lt
zs5VfbpeHcsMAerFCv3X;029S}YXHZRzl?TJyr9b=L`l+>#~>;=v7sv%j})}U6{Nn)
z7FXEf3Mg>#cI66+BZ8bwxS(bU2$T|TK~{lcBT^awUzzYpFQ~!}1P<gGgFQnLpJA-w
zP;r9A!4QQRfxWRTN^Kd{6A+RMp2I>3j2%P?0dO=EkeUi?)aR!O1GEFHp1?8@$QVn}
zJ%-sP5vw%8jljBL<n8zX2@8-jLRLA8AO@UMv80MAi-3Z#t2r2yTBz6nE}c&Mr}L4=
zfI5q~ZkY$zw$Fo+^A^FLCNBi<g2#&XN0l=XFiD6bM)s3uR%fg0fX<9P^pRtSX_B9J
zLG>XOs;b}ufk6f2aiXy&Cs88;yxI*LfQsUJb;u3?9nM*X6jIQ^F*rlDmZ8Hq`BDM9
z1UbU%)Mso0AVE}$afIV=uD0;oLdY{+4GjGN0*eSBb3A85gk1)8R=i5@VvA1^a?Kb~
zOa;+@O~DZbT`W@O6QCcg2!Fb^3X?EXEDs^xIDbX!<u<4d$4($rl05@6fE8rssD_7)
zZosI>d2lzl(?l9z6OLhq1tT(U*DJ3X8`_pUy(LczV=={4hG{RL(Jgs;OP*F{xGi~F
zWm@2<-O{qRwCukuTJ}rEgs>$~qx9jHJiR4PXOyRv#Tzc-ySJ?`j-ie|Ql@WLKWjIt
z|LT~9!T!`oF>*R8Eagw#VOkmof_vah^@X1H!#x8mvAU!^17_r2|EYN{pJgnxYWf@m
zgW0Fx#hWo2k)(e5DjGvsqLC%qL?9ECHV-~^C{95`Cmun!l0aT4OWtUAkbx>fT$G>=
z=uHyTtD=2Hy=ph$I}-id8_`}RiGduUYQ|_1c8M|+F#6l#Gf4kK9l~JVM>?6fKP9`9
zAD@mdOu5h#sF`YXwFqi$NasRm&^)cb=$Xb*{2$9<h(QD=Ese$j3hdK26{D<dE&8yy
z1!S&DUsq)fB*8;R5HUyDd9S9{4Z)Oi>sSxcq?}u*8k3Hi?uWBM#<f3mNP4<wyx~Yn
z_tmmSQZrL=UwzFRyu~nvq&Jxk(6`$*J6M?&XZO7sSUKX$A2dBTIB0qt_Lr6l6mBO?
ziKrgeVX#7aN!lk1>MbI)8j;n-iQ<#bj+erqbg2;|d9!f%iK9;&#3>=tus;}U(z~i-
zIe1ob;U3-BuW8&{&@D5D-I``)+iZAv_K7^<*%{o+Cjr+q8BBV^;=PA;=z#ap+`#*8
z4ZQERP#8+=CY;+e%L=@wbA@xK`L#jANq{RyM0Zyg4nRMY8$jQq0raVS05Z+80%*NK
z-9MFkVXy0j{r!9kGtF`=tS4&w$nc<UdF^)WpXOfJNA<$Knr~sIS=NQUeS*5bo*RUJ
zMT79S@`2Da%L>BV3WZ=11+*zcB8TDeLx&8t-wtZ;edTcCPjfHmd-a0;B;SHev#bkx
zs8GOqBCvG?kra~#%J3E;9^y79T!Mu_|HF2HBUs>CpSlI#d_%X^F_cHk1t%ZV3*Mny
zW;_&4Gq+$DgWFs8YG7EtIaQr-<HGW1FHp09B+nX$MSb?U>xBed;zYEOWspU5{W=lb
zA+8fCRICak#U=2=6quBN4S0)6h;@k}vO`r!#f4LYus0b&z)xK)hHS%LxfwAI+Hqxk
zmr#0G9Y?vWP^Zz(5fL+zVqmLTDxp{3enhoJR1~_0P+WwVK;$^6Lsz`f(IIFsI%?wL
zsIzbhsr73BTAgEd?-qFbz?;&d9_b;Xl<K~3$Q~G$H1r!ss)e3WhLb84{5LeIn)!+D
z^l+AqccYrN*|iWe>L5bx93YjZMJSS=R_5sH1eHP2CVd&`<Y?h${d_K1QP%HhSfyrd
z;}$y)t)&Yak8GAN)huBrx`;`xCuinw46!vmuP-+oY-=j@fG35pJ4{OhfA_$Fx6;#o
zZB$&*;58%n+M3LB`7EQY$@H1pGHY*}y%Q`88QhX5)P-<~iUbs%r|-~l95<5Kbkq-1
zCLNpta(!aq$!8C$JF=cXCwF8i-@T-P<H_N%BhRXW^rFtv;v=<AmEA|x=48f5fobPu
z7XfdRn9PZV(*t*4D$|$hu$NRAQ$RB!x^V%d1M3!wNr|`Mz~aO%)n%88r{iLnE%-3{
zVgP&`mVY!CmXnEmx}S;gPE{K7`fIr2613I=0!15ruXq#dO{KWX;!J%nSG~6p_rAu}
zrKsErS57^|e8$gk{`)GQw#zVSFqI$1Ba^M8Ypt$!E$Lg`D?o80#1p^V4~U(eZOwS8
z@44J#i{rYv-1o{!jGY_LuqLx%a{4c~_y61iEz7;FtV8k3wuJ$6H&^W3%%Iu=CHA0+
ze6jAUUn^f_^L>?H=<B8OOL)|?S(eRpZh0O8hBs9zq3&79o|!#Q-Ln2q+kMt$eciU$
z%w^qlL2kCS)uHX*ZYlqdomkeT{F!ZWb(gZbB10<t5=Y%v`MEU)@5P5NsAm1dD+Doz
z`fh#Ly}$BhH0nl`{ln2`hdJI1A-D$)((k@8pN_b%@{nz@9!Njf*URw5Y=83Uv7-}@
zn5CFM-9p}LJIsnQPumuoxDZK5*JaUB_}R^W#?B_|{6An@T;2I+Liv4tgKrRmw(LMz
z;qt0&u^zb8`+8XmK@X2T@puYB)t_#IJGR5D3%O!jZ017lyKqLiZ1v^h5Sw8^XQcx}
zd_%r)%l(|4eAeaum~C-&m%C~ByxZsuylK|}tNWpY&oA56W`)mJY>V~4=gWP)tbx~0
zJUup%!fXAfTgcaKhglc$Teig}E@V~4`)77GS?B+sY>TTq|IJ5KDz|-zDi7WM-nBii
zgmAlUu^#x`+Skikh<bGF=u?gXKcW6~3n|zRvu^aLZLygPdElJfXrL_t$b39}2>%!*
z-)z?!U%6#JZx@ku+0WV*S9jTO62FHxyN|>;ND*GWS?31o-z%rKQG{PnviN_uh6vx;
zV-)bD;I~_GzmoR@?BT&vKh$R?a1-fli^_*nl}DudM~@Y#zuJf?>5;A4$iow3M^r4+
znV^ZP`Lu+bKy{o8!4hwF?b3`hed{pkdn{gL!6a5A`7sTQ@6s({8FcDx0uvqsuMqTw
z94PuZl5rG`SO`?@oyPJg1AOF(81y7S%P;#4=-AHA#%MAWViMz=)ODo0g?F;5d{Y#~
zM2->ih13T^J**ROlO8-Jfq#F`@>20q_pUZYLyfC{;K28(p8mT{2t7KYT+Q+08hmDE
zV&lKC0Z?(1G(%-~37AQ`Jx(F~yI4l4AbVH<CKhlvLqQ?DD&LqQ20=!t|D?8@h|8-S
z()7_!90tQ2aim7SH9ZuZ5vkE|OINyM^|GhTkG^^<%y1pp-Xq?HspB)=6!DGjI)fPf
zFC46YAs0fBo<6S;;){dm>GgA_h)BbS`*_lOnNwo@d!hX1RQ(bx%g13~o>A3}SXt&o
zP{$=9tmfh!XJn&_SQIYd7W2yfu+@UEeSF++UXB*W;^xA5qdY%8apchBV~3u2@`)$M
z@aF+bTzh+GlA^L->sfR-ZHyY7lUHlL2KZO6L$_(a-ZTHzsW_bNygLir<bvLl*xmm@
z@2&w+d(Po<uNxglmw4<=n-oS2xi0FSBe*VVs=9zQ#8!w%)pMv5v={|T24JhcckBNT
znb*%L5@~-zFY6BmA?@9xqi5jq@Fanik+Z~s!H%u})-!U-v2}ZYA7)8|Ub)q5!!<(b
z!0gp`ldGZM^(Xae%xrp{UFNxbmeJW|`Wy(M2OfI)mHlI5<A`?wahAq;-${|+8asdX
zgrYf`4SDtX07t~CrvGj2X0+0|YhbS{oo&dgT?w~$cEN9pE94PoovgfutI9@KDx&~d
z(%B6Zl3Gzk>F3_nCMZ{%JxC}678-Q-YoM<YT_X7XQ$vtv_Iu$6e2#Hq`UbFtD(*|F
z+G<@$(Z<Aal?EJZDhqoax3EOu&RwzE43N-gfr&Z8jqvdjuK8#p;G(XsUf~rziMKx(
zL-j5RsZb&d!`fL!8l07R(g);jAKjfQ5CZ?kVuT5zlt#53AlzF5OxT4g=mkF{N`~-$
zc-}B$DvjAnTJY_eFYP)~Wei2+p>w?y#<2-`XN@<At5wt&;msiy4hU>PI<(G(#sos#
zoz*O`6r9w>#x>q_Ru)}T-sde*?+B@B#^pAzFRG)07z&6a?K#}@!v3w>@%y&6V`H}i
zB~|eMOk3hu+>^Oj96g;s-A8lo9O~TR+vB6aGIMf#xMGTOg1Aq-)F(aCl{pr&om_Vw
zK@r81-S}!fNP}NG5S-5i0-jkvqd{=SvZ|EgGBP6>9ET_SCS(pULlT`VpSpQ|XRdj2
ztNuyfJS#ug=O?Oss_)VMtKlY&?R1Tu;BD3G&-OtjzYF8e^M-k%W(7EHsTnn_>_7OW
z+wRX;_+`<j>4*NbZDAljRA4cV=TG)E8*WP5b4*7cJ{QKQJ#9p(-p6hw|G*BAbtOM(
zTWsh`Zax;MJ;6C;meKXht>K^98D(9=FWDCBT!Uk;1H^peWd6aNqiHAe03y8~CNraS
znadV(dzWt)gG$%+3kJZ0UO71bN4t)!aQ-9PA{U(h@4jZkvHtE+2uh;|ZRjnGPe^o{
zkX7wZ8H2XrGqY6r-L38B9eLCVrojEYX5csJmSW<$t$z}34D)4H3(daD4R%%=3cp3l
zw1O2<+{T1JBeHz~fy=xVj|F%17l@0FS=BFWFR(PvJnBG>mGA_&Rr10Hz~1+8e+48=
zma#|SPZW5RSZOK#qh#fjNe>piw`zE|GY2i4vtsrE+hUzNlZk`;Za+cfCi`1O{dCt5
z7((cK?!a=|E;1`vrfrK{U^(8`Y%PL^TGLJNhYw9W-cRuO%&c=8e|KwJutR298wvi4
zn&3;i<p6@0Nw_i0m)QyaLv~gh3coWXc-$H?2fBvkMn|El7;oAZdupmPLY5KX$eflE
zZE|sI1Y^F~9IeQFm3wt?e8WN|%U)63`nYYePB?BFdEV8E8zwEqr4Sk9UFEF<(ciNR
z&I+QxV_W0`(ckWCHk?%N8to(1Lt_Ju=G7_nt9HaJOCzOzNmJ^VbxWbty<JB$=is1c
zKFm(2KW`_sk<gpn1;dO`n-n`(P$qxWwzHSXc;h&62Oy56X*y%*t0TqZ%Qe2n>1{*e
zrHMMJQQ*(i4`%Id+M2ArOWUY;2`|TjGmw|CCkWQpcNRS7>{<LkJ2ft-!u-g10#~tS
z2Wxde2g(FScBG3$D@H>f;1}MuFi9cZ?^qO4c%8C}j~pX}hXT%UhAf{tME}E{F^4Ll
zVl_9kWM(V0^euTd&5dsQ$WCUzZRfjAGTSshemgJY5>d+@kY=j}y>h_0duJXEgooT6
zwnZ-B?CondoP*!l&%q~DUTc?bwK@YoYzNHBz`v{+ctN)mq`S3?fny2|cILb6h&5{G
zw4tzjuv^gT7Mv0*Ij!}XTj>Qm?<}w+mYlOK*16J6BcWT2Ge>TPGt|@b)PbaF7nl_!
zv2BqHB-OrV!wKi!(RpM(_vk71in}!N+$bNKmFmCT3cueDoMnZSfPO&}&=uWsfUY7F
zabuV-vy;$YwX@n#D2|bY3`RieK;0<GMO{TyF+j}Q%t7LRvg^tMZIake*cR)A#HNwh
z-rg%C4cPi$IKcdxU1nCm{Hkq{3o!q@ui0=yyQBYv%4G}qeLG&3g^|8~QPbDI*DZy<
z_H-RkkEhNj^I3M{`i7m-hC=QfiA&un$`WZRFPG1hH4tV#{-NE<@PAYC=Z$@o%7>W(
zu6n-?3wQ0xD<|D#TdWh7n?|1dU;$X|`cOluL%(t08nuhhg4^U~kJ%Qv!1eCFX1R%0
zU22;?7{zF(ldgIUc+cDMvMh{5`xQ;JQ@Z5<qCK1<TJu?UqJ6)e(uP9r3DP9vKNcEo
zWePx@{SV7c2|})et#?+~e6!6RPR-kOW&u5Ms$^TN6Ed4dg6~4q4{RQY`YAeV3yb?`
zx-?|C=64Qkf7JpZE7-o*w#Wsx*ZZ1nEGhn&9WTqmNQz(8r1)9gat%^6pJgY-AGTB4
zP{@53Nzr0TFENUp%*AA(L=dl>P<LD~yk-f(oGnxnLUb3j=lzGr2b%H@$Z6XhWX9v!
z67*X;35VZbu`thq24o9gwk_7VOPj_P?r__Q08AC7Xjs9(cMd$?u#3$Ko_}Xs<O0vX
z?Q1ri{O{{`pN|N-jmXNn&g0cR=zn2H&9Xc)fnU;0;J@gW!UXQ^5~5K$R%U{lv8j{U
zG|0{l{<EFjhC=c~&qrVgO;3Z5PVo`Y^vIZZF}pn+(MYP?%rFc1BYLD1fH)fd#CUQO
zA4g2V>_J@en2wj;3~TeFJ!7tIZ8j)jW-BQ100XpE2Z^IF(<(b&?6WgmCtkcs1~j~x
z&XDU51X{h>pbWWcU-UrtMpPPz&+jXJqO;<DRuid<r*wmN_V&afC#gIg3GE#4z9*Q}
z#A{{?Cf$7mIdZ+Vu(}^QbO`O{X5mN-9{m-)*g&@g`njyP(^K?t@zyV9yh#aiiI5SQ
z+f?k(in9#@m)s(tuU!4rA9>xmx}CbUFXmc115*E&{<ROuN7;#uKt@Wcwm|BMo5uI$
zng+Mvf7R3Y*nlmV{k$q4CY2vmEw65GCGYm`HeH-NS-5ce<Z)xTkSOn%S-SLzGEiw=
zr2LjQsK7RZnfRTmr=VS-NI%3Ko^+_fap@{8crRBGa9Kx{JePA-j!VhdLLB1?G=xai
z#$h$7CFAAsVEpj$sp-ZCPG6jV_3ZP{eV~5k(sF64F;O}H__sejbA0wt>BNP{+w)J(
z73ZJ5TzRQ@t`fd5eY$$;#MC&0w%N@ADQ5<bUythr^o`8B5-ypQ>2iIt&}M+eA*I1D
z^O9aH#w0aKk$XT9c?(F3z?s#a8;fFd0kX~MbQQvrv9MLEiXfW4wwHY}(~nYcHw7r+
zl)l=}0AL%jVJN67JJJ=%m}|#Ur{kx3@HCa%w!`eJ5%b&whv%Q|Y2VFY(;1@3Ei0LA
z*g*WSscHX8)00`DoTz{6n#UdoS~Kyihx&u3%#s}oNgY|+kanCpr8L)44||R#!-`U;
z!4Qs!^NyJg!tnc&Ne91li~PmR_+xVO{<YrMFAd_8XAPfZ4t1J13t-rqK@K>ST;3@0
zBYpoO?r}s05qr}4R<^Y6em=Z#+6(H7Q8ShdTE&}uT+7&uYcy2NLz2by9k?l5x`5hL
zj+lQ0esv5@dHrHs_V%AYf9`<9kO&N!8YLGm;<65m!5l7>)K#6Kt?)Cy)$&W1gxBe8
zF7R|#moCgL7x-UH7uPk*t)=ClhE}4o$u*AbA0}^ige#LbLsb05IFdUc!TX~mX$RI*
zJG%whs}5O51^*%!6;NpT^&wQyxl;!qotIM1600~kKlST9<EC_@y%LF&>cN}>{V&`)
zey69u)lUPHoBCUN9cCuv(&2X3qj@f$Wz0=AeGcRhmlcPIkqX#E2Du0`%w$BACquTG
zC%6>V%ed}T3|8>J3K99dzkP8TSL!y)FkET>(24UO1QA0pSVlr*8|BZXN-Cs}xT)l1
zC2aL{jKI^0qfg+(z67yJ=)N35A|bVcrRWq(RW3qg4*elS4%`aFghqb0>O_#-DN{)}
z7jQHDjf;aM*M~P6#kB@9AtAb87Eetae*8%rFt^|2$3jSC3D*Ov<m)`5*kXWWfvnM~
z0c`K4i>rVcCkM77ck1H4qST5og9bQIFsV^!c54sb!aZ=b^<Q&QE4j|U?dLkSD)sEq
zZP#7rjXeDD!zbIgc6rjfh$3v&vLsf_wRtx*^N}g10ts6<7*&P&QUJ4XBgCD`&!B1Q
ziffpR%RwWk<3?;yo8(ZGqGqXGLsB`;WT3Y<t*%@K>3{&`3+aef6QFurx<oW*0!O3W
z<j^c&kIy)PIpjm~W@sc<tBIkzXHhdQQeBFsR1e_zMPp7*G=NvzQRxyb)(8NXAWbAf
z>t*4dYY)%bxTBdjYm@GUc0(VPUchAP9@L>S26>U%6K?(HJY)bNA~OUo0G(2`_c<`?
z!qjo)7nYzK@TO@cT$nq-HLRsq8LK2fhv-m_djLA6S0dX$A)xFt0bi&X8D~L&loz_U
zxmvjvnh?fq)c8`=80tHmHRUJN2}TiN11-5PIt72+PR)wY@%+4WDA#>M8B36O%`3C@
zE(m|+r@SFx4&n^=nNkS;T#D;R8x}yBTh=)nv)}k`9#a+0j5Dr^R2;ODM0T~qdh2>I
z4mVcI1Ox+cNdZ}!oB*{LHK`jx4Eu&hsLSAzG7Eg|9}9aHA?=5|2A&86<|g9$v|!Ak
z{|q)A*MbwG)u7cP14UXqZ-(bb4$((UvGy$j+9DuLqFV%n^CJfVsWT0JwVk`ycZb1m
zqmMyiW9J{!m{T^|n>M3OoU+1`Xnev2q`Ey#L%7g|zw;rrH`uOtZc%O$g$oDRG!!u&
z2TH`^9WRidLWrKillTkuv_=ENiDrl{jj>7eSz(qnoMs}(8UtWj4YaP(x9$9$xN;x%
zHton6u3S)D`Q3hT#aXo<938z#MHw{_>^{L7qKbSCR+k>@nL$cjx_|Y)jN&xx6SvB-
zo-u~6lI4B=K(Erwu2+RJ&-18Irq4mNFnh03p_Cn3-_xVSh#ovA>USZmvNUMr3Fl+E
zz&gF^LI{uy@wrPeV+X1Yq6Hubf&h5jqfQ=)O6fA<ai~GXiFkNJFcFDyp4uED4HD6Q
zMHhmC0wUD@i!v<1(u-4X8jt}B=oGRgtUkB~MtiwAD&;wm2|2J@xQt|OD>5pMp|>p<
zMNRO&Rbb?>{rOzjPMZFsewwx_NM{*$P<h1?l=ga&s{H;WsLn$y#IFY?cX{{+%n=&?
z`@EMgi)40Lq**)~lS{ewTlHlGw!C@o@+;D)8HkuFZC^h0$`~*sYV1Ft|2y*V$N?k6
zy_GYthu}p3HS0w_+Z)TULDpBo-w1z`Hkn`6T;h*cI|%R5Y;yu^ll5K|eGcZArJmhQ
z@yk1eUv_t580Els4reMoy{^q6|6JpYnNKf=G|%&JNYm#a4%yi^I;yk%thpE_7wQ0m
z0rk<IC8Yp$&ng`mV6FPz0ql*Q-q!}$f6)Ll^Xdhdd7cMgrq6)@`?VSHRc&Ec3?G^q
z@0*F{BP!vFSNHkpB;8C5*<-lK4t$!1tu}H!dwYW%Wr{n1m5b!X3aiu)yRXz{OR0$-
z#B#Gzf$F=iYfQx#ukDpcXc_WOK2DXV7nMEQ$`WJf>ilI}mcywZ&4p8B-2YYMRBHbn
z<Nigp_v+sCV%(HV(I>rgxRadTl9$18Ujf$!GE~d^g(vIV2H!Wb6ui0w+d|n--}yXp
zP(Ff#1=V+3{`)>6fARnr34>~5WYSZ;@CpAJ>B7A_3)t;X)%iF;8%zWvDl2`A*uD;^
zQC>T^%N+oJ307hclApvTulFGNi`5R2ub;KS1RyMx;ax@XLS<Ig`(fE9XbiyzmJ|^<
z>I63uCC=uC*tpZ-QU}FeC8oJ53KlAJ8|pL1h<;Eq<cB^4tCp!gU<N}8LG!DK2kf30
zF&{y|)WV=xyHXc$_RQDFft}yJW>={z^R0iSjwyUGFH<<P;~o86!KrA+pmaCeit7rO
zXPIK<7~!5f3}74xl?VRY7kQ<#D6URtZn3|bd$GF*EY`8W2YN1bWX)7{v?ZZu<;yj`
zvZVX99He`F686e}XZF0Ki*75=L+9%)!1Y4kvsBR7_YI0Gn)Jq-aoidwPabb4O$lKc
z*SDR^x$l^GwF>iefwj^-akVS4fJSJsweSUr9v1j=1sCB`N+{9gP1Q}P6K8U}p{vc_
z-FahR*vwloA%{YLoEwGiG${0ky6GAPY4tVsXKcIQx9uuJx9^EUftv&vs-yVbAAx3E
zf!C&p5?8OO4GhC#vl`Bi^G8u><3=id63~8A?HzEo-;&eZbLK7soNc;kM&Q_XS%LFx
zfp&{VTY!-Vg`B)<0QDRIweG$g&>qMQw7U(^?#~A_+b%26?kW^|b8m4Ix`D&pfWx5I
zZfTF_UfMloX@~PI&9=+2G?S+-nfsHTO?F2dAeT1iwOiV$+)KOHEbU~zrP+2_m-a%T
z0HfY&DVdX+`KDMe<Rjs}{h<+Nt4{$QSFKM6#JSu+yw3o!m=B1yT~;8TEfj33)ZnDL
zHwDOZ0OVEe*8y`mH!$xvz+B7+OxrFiFz;1-q`T6L=jqBUptt)pD;e;GTi_4oUf@2n
z!0*quK-(_s0$(f?2ByP#hsF+19DR?sUs)Y$Fmh_|HQh!Mo9{iva32Y_QJ!Z>RaZx|
zm4s#ipN$#7f%o&d!TU}F-p}O&uWgqVymuE0f*CHk6}MIC5{kf{djOq*@7?0QlzVXx
zn8p2EzQx&gSr_*}0WtBWioVU$Gih0d1CL;hgFZ{;RHZ`1|M%R>eb6lTU*%h_ZReKj
z0+@U23|GDf>W5sB2fKCU82$(!baHOW%4GG)*}r!Eb=S>6aYS@&&_uKblu4+Uu-?_6
zl!S0AQGcXnDHcB}5?ZZC5HRCc5bPnk$9LnlL>dB)j6I&Zb^%mqoL7j&MF28vmk6Fg
z5K9x~=d|&#jB9I}Nz__q@NY9J>FXH~E(hyysa2Seuf@}aN_`V*yhCG0o<=ktv|(l_
z;5HnYq$yr$isQ*84YX>S7qFTRC{-bgMD_?*iZ8TjRr$1R<t}w61j^U8m|Nk}AyiOg
zT!P_L40C`Prd@591Mgw8M8kn(AjI-0*PS@}WIz_sD<eJ{4U|y_RqmizK!Jl*hx6~T
zr5H17is5mO>@cZHyAQi)C5cHw)Mm60nhv?Ub13Q&;?@ZS9o4aeU7HPMM?xME2gT>u
zrM7qvndyNmC1S>qu@CtTdMAFxg&TDh^mWh+{p!4liZqi$FuAG&4C_^tEMv{EVriG8
zrp7I-lp8+?%`L3F2C&kdbAOzh@sAkBKZyNy9?-7CBa@&=<vb*viD753U+gb>W{~oW
z-6syjL0y?sZpasIjW_JcV`V_N#hFLV8qMT-onz*C4(FKbbP(WXZ^7X<gE&MS?28fW
zfDvqv>JWm75JujXgE*E66+VPi5isR+__aRi!C%e6pls~6Mme|=IIoYW2b@_r^bij3
z^B%@}umDzt*doo4!7mv=iI`Bj!x5It*bTCD$?;@tsuDWM!xv(|iceL*rg6PtRuWUm
zP)*wG(w2B<%(WJ2IMn>=lE2Itc8<uo9Y-jtEda3vm_EQb%-)v^v*DpTv)^EL3JINk
z`r9S8LL*jDp;CW*1yBp|5F!d7szSwF(7~odC}x`NK-_2gZ`n?3+}_9`6kuS~J^;9^
z)2Vf4wA(tKGJAh|2Bew7<!GUecy0#uQCG91RBAK%1fhTu*S8-+0f`!<1R+x8j3hOZ
zdlS+VbH9AxKfXwX6bjco$oha{@y079?4XGb$%25UDffVNNMEbdb7=c?ZnQl-0Bx<z
z-+9-dyXX`{bGOk|-i1m5-R_NWwQJ}q2k>p@o~~QyTz@N-Z(-4a?UjT0&uc4{G5n`8
zLzJtGv)%o@x6Y-SQjh7mckYCS#u}Dv-aB{Z$S^p}-V5D_oLePSIL=L_g%?%ecu*b|
zzZ6CgYr%UZWx+6i%4d=Wz;u98DQL8WmG|m16xMYnkV?WP(o4n!Vqq~xDkqz7Ov(`)
z>%ejlj;nc{y3Y=28TvZBYIF%QI1W6&DG~LJ7=4qDc;-~_qaLUXaFZ3Oat|>1xQ1g<
zaV^fm7)dx;Inq$K8@Ede4U#4c;(D_{(e7z<(G!TMq7M;PXliu>NC>EEL+jws%n1;}
z#|OwOQ1|jNiwoD5(v=B+m3XZcYfzICMK5VH`Xzn70t9n`$}CAS618Up3x~lsEnGN8
zfS4tUnj&w~Nu&g*&M^(j##z0EBU?C<N;K}l5whq{{ku1=lYd@Y*^i_2mE64gXg`l~
z79QR;K9NdM*SX`g#(}CJoXRc5e%O<(ZW6+?WvlODwmM3wfSoZk!zc^AkKKW^m}`-!
z<Z<Rnv&h79@05h_r!eC*gPM;_F0l!L5?y%|#eg70!G~fa)>_DJqpU5mg3X14{2bR%
z{SbB>Ke;sKy@VE!GnjIRgt%b7%}g|~2NUYzgc9ClV1#VpKBK4497o*Rc+xD5S8FKA
zyfog38bKAOg1Wm^F~cT;>~QytGCxDQV-g%DoHE7OEC^?lV!W*mh+|QVPh5MePnkMD
z*0a~CE3|UJ$lz3-CJx^^J#!cyt%-7CXP!3rW~NV#wDZK9+$lE`(_Q8Q6mBLC^YcfJ
z+%q_S`#lD%tG;(L{~vR~kiyfXf#K&Z%?<^;39PW~K4;qv53Ja?x&~#=;m*)++$w)D
z_bT6GR{146+f1vp?c6FI+s<T|w{UWpk>ZJ$rVyXk6RoE#(ls9fUn0s67)`$^W53@9
z8-$qx6@5g&TLeSYwssd+aR3Sa@g<dlYtU#@$8?+aZ*p(i`^=^d0^RIy&^9C@x5$?*
zR@p=>+%j@6VhFU51R?h^LyU(f%-dxBO)zc5v-V|<K?FBxYGO=c>TmihZ3j{^r@B`f
zhNHG*%kUhqxjq-psCp*m<dRfqzr4r<xU!D*6sxb!vEdy$`_Yv;u4w0BcUtI<KCG<p
zu%ypZomPKgk8|+XQQUH^&Mhd|(3zopQ`eYpi(XCBl}OzVgv{?=41iH(=t9-F!8ytK
zOFd{Qs3PJ}cY!2vNnW-J5-tACXs*o#K7`6Lh&&Ys3XgEADeAC=X9qE*9GPl8!`cK$
zIG4F+r%#_r9h!gsO}fFUllk|1$mXp$ng66`4ntXI$dmaQL%L@Aax(v;w<R(<7G@Tc
zT`sV27{tOlKlL)>{9*69$g*rY$z#%+s-xf`)a@i<sezd|u3mySkz{)uf~mQg59By)
z3RVl{i0S$`(*PBtID?A8RU`nOmKXAubTU<ifqK8L0S9tE;L6vzhvNfgrUNOajlHPz
z@c|!g{`Aecc8$!&q023|W_7WkMBJmMMBMN_Fjd~pOc@H>sYl6TDYK0nI&R_}<d0Z*
zn9{3P;|}so>-B2fLB9RG<n9`jHGevsIcH&%rGhCdYM-_(3{;1*XLHZ^bYHXKC8KvL
z87E!-(>-zXpSRs+oqx%;*wFd!QnRHfW1ydY<W}&Soln*keAu>F=L%L=EodSC(hXoI
z`zptCj;8MxA3!Eo<!<r6^Bu18%SjY&py)iU#JXi~A!=6Vn-NU!l=VtfMJVj`)aN-#
zsg!mfNRN<`VIxA24&_b?>yQaq52B7zt0EiEqnZ#6KggYDH@XbaG&-jN2<3usRC0{&
zS~F(LKj)fU!;^K$_;>AQXGO*zw=HrZ<KOCQHk=(qGjQ4=lS)-a^AX{Ll0XMx<`?**
zCKNn-j5(q@BQ=Hsl-(880zM2?L?0E2j;V4`tPRACruK$-H0-)lkDf7Gw}~S<_<zxY
zB`f@Y*0$J4_|KjY?bmF9a#%&>z<<aSd#J5lMMV@cUZW~ef0n3lnLD6Ng2W>g(S@Q_
zhz+C$imJg#6_qcvK>_k7G&vTry2$-RXc;2x>Xn#%OBYIqgZVbW`}qKJC(E~>6epuL
zSpADeT1BGbOG|K-JZr@8i-ve`-~wh_)i>QbQvS<Mgkgwq72`_!px#D*&*^cwP@owL
z8Hzg$YNT3Dq_NXmU`v>kRhl&Gs!72rEEBpXpkRzoTOhrPAaxY<uPr@frJ#RlTdb3U
zG99O%?HjG~WgN6d#|gJHi*o-^W;lBVF6N$N5C;)YG@3V1C@+j~Hi542l=bUq{OCCY
zouNZ)SdRZ4M8Qi{evYV4a)#4G$TUO@)KVzvym0U;2Z)@1W8PE(;|kw{B8lUH=F}*8
zp_?w_N!b&xGOaMIXCXd}GW^P7ghH7&GgTPR+^~at<XSmDj88q4V=FRbOb6V;3?QW^
z8@45-oSLsRRM4PnKw%f?Pnd-g2-Jdf*j2#!!{7(fSGGCG4dRg&?+il3qDAB^gT)|F
zgr4(7@J22ZPHB^A4GWiq!FG)cs}*V-)-gxgz?5N@>tZqFIv(=UEdeZSb}IhYZUrC+
zD4;lA73m4RER}nqCJMik><NS;;09|Wf<vKPdy=Z4cM+-DH2*@B!GS6pNS~zm)nx-Q
zmbM6RGTo@8&&%02fRx?R4%WMclVJZ_IBAh~vvAU#2Y+h$ZPxSPPi%`^=fQvIYc`yT
ze-skcOJom7H7K=sp91V_AksR5&{&>9$rJDj#7)#;h5t7u_o2(KsaTItzhd?N9Btl)
zIfGR(Oh!1XrD1ljMHCv!yZRMTOz|1Rc#MohLp}2Wu!C@Pa}wkL=<GPyuDz%PEQXZc
zVu{(*v@G*P$Uamqbs(XAiUnd#D%6JU6CXv<>7QalIYKDa`yMhmz`n@3?DQ<#$5fzM
z!FSLUMr)_RX-Yqwu;bpQ)Tp9K6O}N#wHe|dd5bs0QSHrpvzjtlON(Zy9k#`WQtf9A
z)mjwQ(R)VAghF&GVkzZ$Sk!@gQT>u<$WtP?16>WdDOwFQQPE<ZHblI&R7($Qa2fX`
z)Db358D@~YB$ALQB@IwQlXR*jpSTI~cnOrH)Etp+OpJJn=sBDRA&-@Wwb)>#1{~t>
z!c|!^xSMi8r7@H;C|ONi+!*vGBB%Op4*C{GKWEiEvoJf7BW8*`qIwC{ni=Y$m24w(
zhY(^vC8f_nduNOytpUs>!%YoF9NF3^BTiqd8|A2d#L{loGx}Y&#X6~d({#4acI#}@
zo?p|%jQZoM;s=o!6}CVFpmSt!z~+ccmB0b;#QPvp1fg_XL{b||vOfokv1c8qSIXP+
zJMbY~u|Qi#*lr^4(MJ;F5_BOFD-KEI>8Vh%4~|_N<BM)#-kZj;Tf<o^5*sua+)R>i
zASF`>JL4<x2@ZDB4sptdF}3b<wiAaA9Ze}0FM@j<^rj!HiHQ!z5r&VXJ2G|0*%p=9
z5n<9sS`5MjFl*E8YAYMGs-6|iP#+Z#_>d+DTd^E6j0Nyy$0x*M<JN-39e(=AQ%Vex
zmU9uzN2qTP4yobWOs#MV{c|sD^b8!LP^OrY*o48U^$ZNuVshl@dMdgf{0HTVxpItn
z6tlq*VRi_Sg3YMxd#3^<=1?h@x7^D)sjtepLlI|$HWq?XYTOv<5BM?7C}nltaJ@1=
z%5oAbuFY=UfHI8pN4oU5Z#p~`{HNHtg{g{Yu0c$7C&#n)9LRcdJY!qrIyv6k*KD|U
zHx4<`IEfOdY^YvI@l8UHP)dX~${Dt#tZihv5V0gvFTv#!nw|5p#7`l_f@+Sns{uE8
zOP}Gwh=Wq~(&exaMnX<S9)@PqEs4oXwDC=j2n?S<Cj&Q&Y7v;NlAWN7lFhK4Y`tN1
zWKSX@GDVO*3b%TS{D)KJ4_TMK)}acs6^{sYZk1x3V@iZ0ukAjRG&mT0;&KhQK$jnM
zjv`*Nh?<omzRkASP>T43)0|d4as};(CWE9H5npiNoAa?pA|j&maP%P?wSom+6?Ie{
zhYI^dcYxXtC5Lj*;e$^dI)tD@h+Bfw%zS`dM2rzgz%u?Fb9IZ@J#e4{nMro~Z0_w@
zU0tQyfNHpu=X}H{&cTEN%9%0Lrv?)NME6V?go%WGlGO!VOeg%~80d<{AWVQI1Pop;
zsveqlXQdaEk~mkQ)utpe%Tx<4(FUj$e@~SmC!h$1ite=P($?-=le^>Sx?!m_D_z%Y
zi*?fVrs+uU0gFiawIIX#yaOOWG;#=C2!DwY5LHpys1FQM=$2r$@<QQFL~*MViG>xk
zj<OeWD<p{pC3svvxXv8%I*&GxEG;%x<I&D9<XVE3S(+GUa>B-(37df-iRmRZPxK?#
zM3tH>^Ah=zkVpkV2XMzY>C3R^^mMA$#`|_#T$hP7!J$qYsJ!Lm#9=Phy599034Rh_
z?CM|(ia@2zP;@l+gO-A_(%kpk7P)Bduk|$>DqVJ7HY(>}^z6l!UIdj;hu)72N@*<G
z*fAW8KW~?w6~>>lEjALyv)^gtmAQaK*G)rm_Nqt^OOxdow%_B9U4eT+G6vU(;Y?Wy
zb(D)1k|7+hO16>-P-9z+!Q>$j!ttZj=3Pc|OPaXJkq8cta=oP;y<#iX@7Xg4N_b@+
zRE(%zK^BkgS)>}?W=Wd_<&g$1l%rfAE+dWzx66Fcg7cY5gY=9Ez$}4JX->wiY>M~t
zTg<m_h0c|P8#?D`vI2We;!<Qg#jxW2c77Zpo0UDbAid#`cJ%foOFdcX?dNQZb<*3W
z$q~19%MmY1s3s3!GD7A!;Lv8O7Dx(oU5as~1;_(*$KWv0(@!9f)Jx#ofGFYDAob%b
z1fHoFP<N%EcM+t9f@EpU-1Y&iucfSn=Pvo6F;gxfYT<-Vb>!wK@gq2_^q0_AAmPh(
zmno@09LHiWs6v+41%hbZlhJWM6Kd#n4Pa!0JzcVa8}*wQGLAO>4~ywpY2)A67P)BS
zH~X3mKksi>$b;)GMf^lmmSLWqL9g73{>%=Pbw&Tlw%EuO%`WhC6}JKmU;27h+>Gi$
zvX}D%x$$jy^l7?o0ht@IQjHT7{pf!>=SzfI5|@2-iuE?rAK*MIBLGueqRQkZr=twd
zCKz1KuJ(ej$JS430m8IU)3DvbQK-fb7EBzgYY1D~r8Rp;RYYXh@-#;5Mgs?WYGkVB
zssholukfzXz;qhW@Y_ZtNcH00!~5edmfr&T=7HQ%;O^V<Ml$TQE!Ihan|0oHel4xE
zKL(RE>|pwmbqR!I9He&!R|HkP6AWLv7*`SKbJRPJxCSJZd*>MrO50sCkdsRi%i-W`
zBM%w2st~viH<!q-a!m7{bv#2O+%3<PW{K8@?0ZteE>Fc3Bsog<F(vm?xY(!`!JkNf
zn2t#lcbpP|A;0%HL5&9VFH%ahGR42m*X6MCpvAGQ=fVNoA{SOZ+ShF8NjF=^xiF1g
zQWrz3Z4z6o+AJ0{q>qB17@sP3wQJw$*Wjj7%1d}#xM(_5eTav0fNDIO#0JBZGu?WO
zmL}0o+VwVttyE1~aodQ<2pLH=RQjM`HE>YI68vN!wCGs4x&waS7)QmliIkQIxH;VY
zt$Zbm(E=PV2<t-!v`3>VmY8-h%U3Zz>lSNEFo#j68jMqn(q+LJOhjr}MRSpS%SrOu
ztX_(NBls~Fb}h|ix;Q`(R@jC)Y}rVL&9?;dvK@bEVUN|V1${@8&skE*N|WDeTWlyz
zehip#(x^CMR8{c|liqvhn%?;P4@2ge(d3ndjgRSe6VeWZs{J6V@NG575qzvaAb|@d
zNd&%P@1nJdW9P^rZyAS7GN#%+j&>(i&;>;#O;olVo<2E5<|K<qol@#-(cK(X-C(lJ
z9pjfc3nX5SoF-mE<SytG$5gu<A%IoS6xp}1ywFl@rt<{*VJ~psD#|`a7Zl=pSgac?
z0Nx!|7O}iWopTe1CJrf)pS7Lt9=sc=Dh}UnF+<@}xU;jZ=`XE(a>XQ?bhaz_n8VKe
z>F}}K)5q#*zMbvLp}o?%A#ByE=8mh%9D8_xxOwI@3wfGZNpXgW{Ae>|ft3=*%D%SC
z`PWuS*hbynI(%i1Wkl?&x17Jek;|FLPkL8xWk0f!E1Oz3bGUj7{hv2-q0g?LeO<kk
z{?bOS^u6n1u#Wi|R+Y8%%D&2rc3JB@j5kef`_`1&CdWF?CG8cO<7p&ks!qj)u#FV8
zRTQO^Ve-xO<KX_s7K&Mgpg*)Na>4!g`<e|^Ogf)Yk+I#*W!jXP9mBHjP$UFua0=s$
zKc;fu<fZQ^aPqcWogavgzb&GM5Y^+$bR$E4$Fmo|`tu307w7Th8Ol_W#|TFZ3rRNO
z2IFE?Um~xlEwPrsYj}okf4ASB*Br9Vw%E}9y=$7*b6juAgl)4A7PhOrb?bV-&N3_B
z-fvs1b6uOoFJ_;D9*|-g(>9$Kc#70ggo;B0g*IMWmQ-Gq;LF0y)WrE;qsAii9Ty(J
zYeJQmV3(YHh~}uFilPAUb29!tgkgtVsEp5W;Q{)%6yW*==&$t!MmoY2ixKe$lw3yB
zn<q+QhK-!POWfiGWR#!`wC>ae&oA)K1x(M^a}b7iCi-yu!|k5P60seDJ-U}BJVQ!m
zmpL&O9)VGoT0QJpl3(i_JVwiY5-?1em9{`Q$+$g0HUkJ9wY|^MPF8B0v@LQ`+mn6G
zhMy8IPm2KvEZb%^u)H|$cH&uKGMoIeq~}wORM-!0f0}vYg04v;DzSumy9A2C2?Co}
z&??!&BpW)0gYk@AdR7>Jz_!>(7|(thBm~!y1*WQ8$`rOiI0VSb$5g5f_$6mTCDEe5
zM|7E10ZX^Sa~6@yuj=l{5vteqcLD8R<<a0hZFne^e+fe%#lDTaeLNDR4r3kqESsO6
zf?xo)-wjJ1i3h8UNB6V5cMZh!C=K+OXbx*9aBm^5B3J~H0__rToT)T9*v1*+)&~ug
zJAmOtI_D^99;R8inwH|;eJg>PM76G&({gKv)CoYm&|7-V?2x1HOO{%*(syWEtdqVs
zOJeCvro<IF!X;3OtO6AO!tqnE`%4fP)8%-nIqMcBAq)5>#xC|AcUlkXLG0J7_TixY
z*DdU_Li-KdA{Vqj-q&pCi8cEuxG5mPBUq&@gB?g17x)0Z3N&x(&_J(*U~{-07^|jp
zMZg)vh0ZU~KqoFF4IkCh5MhH#BQ@cts+^Kaw>yu!5)l?3HydO}CzWkhdf|N|%wA+u
z+^tGE5c`xMk__cgq#xX07}n!zyufk^XxcNligRNaJ%Waj<bsbb!WxTN>TE7IPv65W
z*oao+2oe^IfMQ<8qEhf++U4b3B3`mq0xC3E?zRQ7%n)<<|F<n#X2t&>wJkOj|F1{7
z<EfyIYk8QLF9r}v2ZbXVu_6+;31PkghaT?jVZk^Pb{WBHPvX$|7t~Z1{VJ586jIBr
zQMwo26#=ZW&Par`sEksLlQ{ov3Venx(zGa~(bLVGC|5XsL@`pbMU8q%3(F&67{VJi
z$DqC#fgs5c2aN$T<%OaUb~5w_dJJ&22<#XLKZTqEcnzE~qFPvbQOy`%R8{Vq-kj97
zqyefpj$E{(%a#x5zfJ!rHWlI@2It8P2dMa7Ep2aM!zutKw7z+;!SSsxST2;6Z~csI
zu};3VX;M-nB_*AK;@QUa1mJ35Pf8*vGl1P@#PJdwnqPEsI|~<4FoGyW9AVWDkg60v
zM08J?!M5O%-R2Y{TfORJt2aHa2w>Sqc02nuyH8nn_SbETTs!-}^)(wVktLAG=-?Gq
z4rvEW^hMcr_>dW|0<S5n?LiGg)l#=px&o7}jsl42ig;v-5?{seCp1gNgCK@RpE`_F
zhS%~UiwP*jm>^YK-8+<^L)ZS#pjM}?zJw_pWONQ^z#@FAdJOE=0fQ&;SXHh_&t6>y
z-M1!jgzM|25z~l@ypyV9F#yW6fN2Pz#QhR1UPI#vef+2fDhG&~gkA+@Sx$*7g+p^%
zDpsW@VJvj*jRk*KNXXU(q;NRpVf>GlWU|ucAK4ZgN}B<s#TizXmckyr7%RM}hJbv8
zX+c2(hMMRvNJ^4zfKs*U5)&~QF^CKD6ALy`A}L3Ws=8u_926qeORj@1Dkg2BN8U;d
z#>@-^Hdh6mbfsIe3#GpyT&0(SRbgDHHN38-dmK@y5PU+~qy5}SrM+FCF-E6ch{=H5
zREa8`o|IZRMLfk&!l*_{h1H*(>cQC>W;>RBVP}}jj3}HmmGLguzNw3P#{Q|vBmFrj
zqr?q^L2-L)HC_%hEV@4HT?@yIn+r&f))(F+ZQQddDO0dT3)c}atUZB}-;jI~=#X;Q
z5n|nud1`A`n~ntJL0gqzRZ6Ca2*-UwRaXpQ3*odVvW0NO_zgrjCmG#zM_wQ84YtKP
zx$maQyg!kWd8b9UM0C2UbBuJ#x;Mp4aIgvdKrt>x5rf0RsL$JL!EH>ERyiuyAEV9R
zs+HF7gUDZ$vm8St0KXv1f&eBpc9HiD!?3kvL8b{3ej~h6tZReC<$vJB<#kRO2i2OB
zmT%UawtThf#2rpOW-%n|iT-ZeA{S0Q)Yojdw7rkw9z^9-=Ko_LG;J?5NTJL&6fnj8
z4BQZnXTFKcL1}OR-=}~{U$T;T>66xd&`zQ4K0MKIFIxJpJqYDf(@>r$xikC#CJ9N^
zwf7gGTAIYUhHaRi^Z^^Gh;=OkNO3jjYzYz_SX@m4*}J^q&gCs7S}{$jyrWjNM8$xn
zuin*Q^NrJS_K|)tVe7L23Lo}&{$@cXo#h0Uw2?A<6q$6Hh>Zf6!}<4HRLqL=@3Ad5
z6z5m4nUX;YhDK0A-b?TOQ2qp*1LUanPiyl4uV1k3!5a|^<(QjE*I*%eh<zA|l{<_*
z(fAsKo>HygjUYqxtE@N$I;ut#vqViKE#Y4=#m>VlA0yvf#4EuGU+Vfg)`}r<-I=+A
z7dmRaR7Hd~$Tfj&V~IV@wHH8DsJc~_Kcggp4OfY6xcp358`6ML;uMD68Ay4Tn12DS
zQ*Ie?3&oV#(7+Q?U|i)DnSe&S!5h-jV}eyMc7??b@t9)lBD7m;h5Vs6u|L3ZeS&P-
z4#|+;bbI9J`-7HRv(onk+hU#c{U+t4;mz*55pv(Y$|HS+%vW#LxdFMX$;zp1yfA7-
zT^Kde*S<S*pwifW-F1HW$SW(U3&txyz{U@rdEQ(z=WdL1qjz>7IIn;Z%pi4nxxP@G
zVSD;b_qFBjmT-4(c~_U0ed@AFs>|KHxLg-E=kx9kPyLGIH0-N<5$g!w8E$LOrTW{$
zTht5bTht%MLb2G+-Xq>xH>`hN^&IZ*>-GjVO-<4K-7_@}y>OfApd0XMkaxBxuppLD
z0R3R+PUfB@NJ)m#pa|{(cXpDj;O6<xjjb3G(8{UnI=jp_K^cFxS#|pzojoPge+4fu
z>b}l)cAF=>(tPD?XGa37A9!4{(z&f1mvB`pZiqxh69n(o%PXDj-zFY2o0ihq4p!S}
zw^q)Ex1R0n)R&vbP4u*rMODEvYYqHyGw$4NVXCO;ba|!oj`5Yw&1woojC%<m?9zX>
zn^pXEODirF=TIG*Z<7_dlCE<*ZuV*sV^VM4=-iUPAOp{E@iP6#D{pl6K){O{MHLx}
zuz<6jed-BPRV7fMNa?Itg0naQ5@9giz&_aB447<Jtnz}}l~dPuc7@1dH-FyPxh=RH
z0fb0!Z71&OjpYC*6Z+eOV~OSW@dZ71c(N0FQ3|*o&M&G1U_}<<=DD`g+1YN*7oI94
z5q^JX_se3?Y{t!^vdcFCeonR2TIt-Xz@F;mM(Jgx^FYs=o-QiC3Xay<qxyAoFs<L&
z{i3IFXD`<P#7jbesEReMbZ$ZT<U$*#Eg)XT-`5BAmChZE%x#v7Fcbl8Fbr~j0f+7C
zqGCcMDjRGFw|DmNs_LS9)VV$auH8~@*BV8Oj{vMZ5IGrgGdsLho(QIM_uZpL612;4
z@iKCo&UW^~7Kgy>V!EYZNvTz&GZ-QX`zo@8Go%(Qgx4wK(M^U<Ky$mX7)BQe(c#@K
z`m^FZ?vY7Wz%K6QZx{seRYI6SoxPQ@<u@af3tOfBJoS#wZ6r2ePJIE0h>N$P*2<Z^
zk9GDmf<{s5{W5%li`mU5b0`f`BwcjwHCKS*aWdAxWj%3y42Cg~(_Q>hCV(%CEgIKA
zek+|FcwTtq(BpvTp5}b1r+Yqe-P;>RLAj_7jjcxQ1r!CEmodN%#+fXJ!FD3j4pdZb
zg)66abRGas){2%m?E;G8<6XJZ*_$jQzUp!@L>x+$u_#D8t0HP-rL#M!L8L0a+6Gl8
z0Ms4wxB9bKjd|&<fc61cD={CqBNV*>$^t^PzSM4Dh&`BHD{2A$iVa)=_Mo5<>b6?G
za61`A(w^)242kE!-D>PBV!T245l{x9CS_3YGG-74K?`HtRl--7iUFMR)gr<H$#iAb
zx3Ook;v}ffD~5KZvklHy?BKn%<)r0+Bd97R?XcS0yWhco*IId_J=fWx_)kEr+{>E|
zK~1e9ginMx3OH`VGJuOMxkwv$Ij@-it&n}oFcsHZ1^$MJsGXgAdOy__^(+SuZ|vM~
z<nYnM#>J=pP%;x_i<|Gf?z&{I|5)nu4VFH2D?8DQM<Z5qo0}D?Wwg1`m0fskbN{YC
z{Dc{tzuKYqtv2+49#<K>-MPJM{Y>as1y+?E7;fNX@5hTaH)Y;5?`-Em1^~ad(z%Hb
z)$jW&KMLkr`J4PP0*#>ZF8;WmKkng=pN05Q`N#b6Gw@1O{vm%X^Up>8_-X$5DgOBH
z*})2byl@kKoZyeSoAIN_A3wks9sc+^{`pV%qk;RkD>eT3KEC)EezZF~IkVO>ZqK7=
zAalDS^SUB)x+3$rB6GPS^SB~&xFX|Uk+HAHxL0J%D>B{{8S9FSbEO?V#Gbj}aJQ^B
zEXY#BPpEa$2%*-=`&!gGzrbb7I=_pHm32}uQ|tT#{wC`@!)41lDF&)_evH4#I={r<
zxXudY&UWXPH2;x}xuq0=i;*m`S?URo8S|YYr1TT@9p%1cxi%M9E!E1WclJEAU+?Vd
zai<wo8r|O2zvz!E;Qx2^_NCuIYQrHY6b$0;+ze5@s#puev74SFW#joLv7ME#Vd<x?
zPd>hK0gN5q7#tBm?Sxt{&!XGynb`ICX8Tx}%gsu=Ci`1pt6gOt+xmo~K-$?QEGnr&
L<;JM(xv~ErhfJ&I

literal 0
HcmV?d00001

diff --git a/.doctrees/examples/reward_modeling.doctree b/.doctrees/examples/reward_modeling.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..38c064cf51884cd78575531e9030e0c740295836
GIT binary patch
literal 33651
zcmeHQYm6M(RUW^_9zWN+iP@F2%iT#f?%7QDJiK0ehd3MW!yDtrTDFxq+I5=huIjGp
zsjjNjV|qFpD@&rFwNz4mq?64H@dF7Vq$ns7`~iL-fj|hz4}uZ|iC`rp2tOf!KuCP&
z+`6}_yJyDT_UvGVv1Gfe>ORgn_uO;O>)!k3p0C(nnBYI@ktnboZ?$5Wrstbh)Jyl7
zem!YhUfla~Z|0YKw|k{@Dl|HgpM-U*mrkHW-F93vw7lM}UV4~{$Br9Svif@?s~$VP
zr>on0>`8m?tzOBVO83N0>{^9S>5*$z#|TYv8NIuX*X&VIl-zy8S|t#M3w*IBp_&W~
zz!3NBqv=#t18wh0Co@>Dq`T3g)w`0^jC8*d$Dvb8V&F(P?Q+bB;NM$!jVQuP{NCdk
zZTV}TnFRSm&8RPjR)fkU>E76A(x;w%#D3!94*T)+l{wpQTXSLJc~&^r@x$f0OP9~P
zen<TW;G#JR)SR_yv;)_Q=0e81+RpH<w9TG<+&*Z(&)${P>;sbYIqe=&FVce-yg2mD
zL{gc_b!^~eMk2H~Pwm-{U$LJcD%d{{)a@rh;P>P2Q}}xVe`kP}{UJsNMeSKsD`#L#
zrBk4?(F~2iE{xjEZ|<C!xMn$C!v|9kvxrNU5qcOu(Fpyv_*l|5JaOK#Oro7AUAy$L
z^D`o}KAkwB6$t}BdDI(;$7qPy7tO>mEfL$6Fw9mG#oVml;5SVRq%~q^#S%=gin!q6
zgA<92p0lE3iOBXlg8rhlB(j7Y1we%1qG6%;acDSR4gj|%8orC+BIv36AoS~&nS_==
zNqTX?i{cRUe&y;75g36LE(tr1gJ@}P&InhXm5LuW=Zso3x41ZevNFH8aO#w(h)-Ay
z1U2xIVf#ij*GOEqY}@6~wHv}TVk5F*VSpT70S((i$CgpIMZ}yUm;}I7RjB)3<V0~q
z9}EJv9iSD@6miD~U84ZsL1P{Vre;LN4KHS*VK`%_p14Lh%ebiuqT|H2aE+Sfis`!T
zN0v7&&}#t3x5Cge3D7$F!lXl2ASCcwvvO#)AOS3Mx*~3%%vxYCEG%A>BSAEmcoG>*
zv9MS<g-=!#tJkuXe1rf~POm;1>Asev5IJ`9*SFd9XP8ajdGlM6QQvv<SMP~c(STxB
zH1U7PuDQJa=;-qLBUpa+49M%}^(wLVDhmcR6;l12_jP5wzV@!y*Uhi&n3$Mmk-sE9
zVH>enb}*G8o;=?nQDizM{(637Bw-}l#<KOiIH#o2l3Xn0^v_G7j`z-+Uyp@LLLY6C
zyn&>^3gNV^SrPfdiBTYe5Hei1D{RstAeSPeD@U?MLL_jFI;!}h!5}5E=n#M!upB6?
zJsm%BO;NKXL`$Lz*wJm55hwtR=9-BWMYB@uQiHN8gmKNN0hny16Rt1iG++d3O)hVU
zx)sK}3bp+NOD72Bw2h|YVS)2KMy+E3HLL@kA1nMIh_&oBN&FfR=o!(wej@6S@Kn|;
zkXFgTw=qCq_U$1uX}x@~hA5wpua7q$r`DT~sd$msZ*YTD2A0J1Cnbk4gFHW-@kRQ9
zp@QI&zM_zey&S^Ucn}^~7s7MbG9{r1Bc@tNd=F7T6NlzS6eo>_ut}SQ81bb<SoZvm
zM&ac(S^^T|wAGK}>UcOFS{KJJohxc8te|FdSS-w66j1{*jg--X3N#D-=OReLy0Nkh
zlp-Wb9o7vO=AR<c0K>L`GA5$Hbz-Qnp|DmAH(?<%LlUzxN4nRXZC6R*tf)bM;NZ&n
z>sQ93{gHm!TiU!y4?0B~$x!Cb(EEb<vmeW;GJi&}!ltjsRv;F{^#u073YIxjoQJU)
zCp?~|>*sIGD1whtG^fF@jZOnHK#yiL_!d=z;h1Xx{{RC-F&vmhE8?(>OlFrT2OX=U
z0=?FS{UsMGR{ot;rS0?b!AqE43t=Gz36?mlh?yZG4Jsvh<f<MyFnOIuR|I|(mr<~e
z#S?6g^^1(1evKZiZICHb&deWWKcEZ8bzv964kR<ti8M$5KfpBSk8h98AEY5ZGQb~G
z=^=;<7}IXG=KA&JqE-L)?um&v_6z)(j=O<1jdcUY=86?YEL5gv<u?cBu;F10QQ$Rx
zMOEIIhVLXYngg%pr_gt~Iq+UMJ<H9;@ETAL%D9H=v&04?EKhuc)<#+l3AiRSKu@%t
zrX7=V7W+Qbt<dbwvOVfLbuffED<A;jO>p6ds0cWfLSRWkQt0-`z9i+78V&^&AkoV;
ztWbdOmn|!xR}iPbyN*c>E8d3d0VE*ajhvbb)w2dIYgzVft?K3@%mwghJ8@$N4xYk$
z>KDGPHYs@U3BXH4{BML^Mk9h*-e6p9gEU>QD?Sb7l2k#=o|+GpwGD-mssckt7G-S~
zVCGHH^ziC{41N!kGp+15Q=sk#7EgJ0nP38vg+2@An2WpcLO`#kWwkPWTMjY}8~u4>
zS`8h%WYUc=IDVr8KMvIFdSEzDL?R*t)`3?u;H?7ao5GAmCXzfD!zalCu<pYpQjY~3
zW9>G%Qlh|e$$cqJf`HswawQgUp7^02<P&70z{mi-yWmt|G^Hw!x2Dk*7cxTvlyV{)
zf)I@xu}xNi`ULst8cEZ(%!+`j77B>*zz<^>K=9E(S89TZ8V!QZ$E>lQyw@QufNI<;
zu*}kT1=s2aw=kmo^Va^``?V#rEuywX)V7F{&i|#pb#7Zk-IF}sJP{>zG(=DSAle|*
zlJ0X2ubDtm>ZQ9|5dvlD)N0#ZiUN455L|=G4kc;D;+V|uvG(@F#bao7Si}G6b;}a3
zozHdm(s@J}W^R=#b67w!g&u*W_jBNZ0y%99s$VoS+h6O_h3n_msO1`6E9^iEDnozA
z3PU<yrj?sz!!+8Rvg<?!S|bgo4~|RzwKZGxJLS;umZ^3Svr<`h{m__U<)rOH*)uJ;
z##{t9WauTwgA5GPo21rXvX7;^3H^e1;(JBS*q%ra29`&LOm!78jDnsk!;y$FPC6b3
z2vwld-jg03qNcLq*7%Yu<v%WV-5+KvrMtQRba%TO#XV~<?DlsU+P{(25qf*#;>1RA
zI)q}ahUvHAHt40h5LX2A;7x3VO)KuDkLc)PPPblqK$mb##8U^Dp_fi_n`r7$U0u2$
zbZ`=3QfdP>LC^jPw`Z-wqN_V`?=1=){s3@)Ff%lG6kv!ROLxw$KRQ2Dj_PbMex#oJ
zM}RqV%@L4a^a!-}z{R=2AlUzG|70!Fw|44!<5A!K>DZJ%G2(>EN5uZ`dm{X#oRRBJ
zjZukDlXz|7^Ov0XB~8S&0?nm$iT{+Y_pTHFQ3QC`ls)%N?b)$}mnNFsC$4hemGV^;
zL-S<FtNE2J19^oMUQ#P4FO+kCPC1|cW%wC;xdUvP%EE1xm!-&v^8J)hcm|~h++xbd
zG=f{~7`w$5NiiAD>Nu9J_gPb=^-0YVYq8Rx@FWjdNeZ6U$FrJ6C)mwD#LK6jehGe1
zag_royy!grv;t5&Ssic1cUKVbq4kNJ0J#gR2Adk;EV&>_ElWDP5JB$-a6eG^Nk)9A
zB73W4Bpr_xaUM%&LZ-h9j|H+6440xb6tIH1KM=DgCoYBlW&1M!0G4o`<Um1s-($yK
z7MAzPIYCywkAL9$xZw*j{oyV!;2!8Gw;y*6Y{?DBr9+1zPOkb+E<898+=Oy4kVvp>
zbrIj1hSvjbk`-&3*Z67AYJz}xL~{fRwFJ*|oBSt82T}PGDrG>?a6LyKRnPo;<jmo`
z25N=e+khcU4A3jSJ(=@KbAV51Ifrl1M+Q-0Ie%zy<drxZ{*X5v3^d)po)55K%cIgy
zTzop?LIoGCfH!{!-?lkzBu+!jM)4Yj^GTz`E5&bkwY40VbAuXCl|+>b-t{E4VugZ0
z`hBI|-2BZc_*kchU8vIUN-j5*eN%L<()%~a8LPaeSEcI?8B<=YrgZEF;uPc;W*vE{
z5L1R8oux3rVd|O{a@{F^9=l@2?OCDSEkm8W<g;J0Z6V=CwMTJh_R2!Wu-rp;?0cnt
zAq8EQ4<t=!a!1Z*Aedwhmz0ZvA0YXNyqnJ#p~=0A8-sL%1f9om`C&N5#qR=wjctyx
zN0jY#fnw-DH0rQKh1^w+SOFQ^g=CH#7s&-w&2<sHgCs>J0OcpJt6E^8aO#2t@`?k=
zOOcFh6mqIDxrn<#9*^r$r&t_Z@(E_{UASpIWUGPH91d;zK0L^t53gTEyzag3!C{Pr
z1YnXk9Q6!_Uv>sZwW4rDPHh?nbZQVduspfruJG&iB*1z^U%jMV1I7y(w1c?}<Z?%;
zR~$+(xQZ5xiu|vyCmMa3rAJ&r{_8f@AkN4@1JaMZSXqD@AAVA-rQA@?o=7Og1h*Zm
zeH0F&$Wt75MFbvp6@UeLA`<EWe^7@=K-8jTL-z7AGTdyY8_f-T2XPvd1wvGisUaE0
zi+918mQisoS+0mz)X#`=`WP$DOTp~Ag5}*7=dc7q6Hpl;mV-6fkFmVwqaixTpu@s0
z(+{$F39S?ggNJNd6U!~qQ;F7yzyWNy<FW7~Mp2LZ7(1UTxAoM$mAk~IRZ3G|bY^I7
zZ;}9c@Z`sa5ZxG0d~Dd(ITs6?jWCK%Tj1293s8tKpi65@x5AL|;AJt3F{S^<z#7N_
zx&XOB+9KjZ%SdNtSwQhEXk>`Vf_Et|xlhcp(BBZNqRc0$)}dKX(wut&bGGTJmSDnS
znsQ_pGmO(?Nrp1_l0>s9nb+~27q22h0WfHiHj%uGGFT&OaQei^0s{)gtPWCi>oMd9
zQ<A#pwTD8|qzw>WtO?LOX&YUZOb{0&f+Lot)RREZM%0J)=1iG)k?P<}5Qb(wGF^ba
z(H$7gCxwiSQY4+0_d+yOP_ncL9K;-x0M-RoCUAKpK{|GvB{7r55c+<*Lal&clnf2o
zTa4UcD7_|{pkkynfcPQ!2Mhz5<{|`z#fJi;vSWyM^h1JO)F@hNqg%^^K>-UO=MKm}
z7Yhq5I!NRRO$QG&h9t281SRS*R-CDa$GooOU=%`Y2sWb%n;SVJVT7@I>2;KCDI}AF
zfdV)b2@c8wCk`TJ1l2+;%Ygsxl2lI1{l%cuw$7$Bigjkyw$7FiPjBmN+dA8}&UOzr
z--a3q*2Zy6z75;j*4egow$0L%w{<pB;T|}hO?vL^<E^*gd>t<kZ~uk;mlV!aQFrWw
zTsx|;6Z^_;D9w~N+96YmHgd{cckI4caiBbNmN$i5W?2`7%_1xl3NvZw*e_Zgh_hCZ
z^A*)YC)?1M1aOW^@5^-NCa(ra2b(6MLo?@QuxS@g4s5WI`w1)J9I|)GzXs(SDzZ|6
z3$A};i90BbB2~^EuOfrF4qpjU60p57OA=ts+j1$5jDW=^S=bnPIrw7Kwms<Th0Js5
zU~zrJtY#3w6v;Mua&OS-(VqIoIg>gKp7i-Vi5)4FF`OdOE5xoj7rTC7bkt$+_-(ir
zu~|nFlnoCwtGj{)Q`l;|2Bm|pvVW9?t8%5;^6@*KhLN33w7XaJB8NzZCE(?s+;VP!
zEu0aM`GFkf+Z^#Aa6}$9_}Vx@B%1=fzc?Jn%)<s}_DoEeRzo^o)XXZy{<7s^9v}e~
zv4MuuoLOS?^~8x2(yfmz0PuZaJ3l5J{6TOUv!zU|F0*OKmsyyOMu7K0=s@hi3lk-9
zE^$RFljY;XD93hZ_?l_@!J{d6=0>vPH}Zhtk_uve2xdCD)T=7KgCK;8>%d2e4rv3k
zE~P#}kCXddsBcwOM`io=dsY<E{6CRN5vC-8&d5SeQ1!(maKu^t9A-sdJE8hIaSQHo
zOnao4LsXu<5hm6QchKJq;9avBg~x`?Nbp|ELOxaAO;Lq{GJUJiCWHDKVn*0%AwOy`
z2I}LC1M+!8u(yq9d96kUOANK7yvOUp0=zbcMYk~6S075;%wP!(4ZQ#GUu7(}pI^hj
zc^2Ek@^z&&P1$pBvyv#RHsJq?&dOO(DGEo*&Bl(0v(voFpm@WvT=Q%p1Sdz8+c%?1
z+X(c$f=Awn>qxaedmgd0Z1`0{^oV4-l54bUrXg0B1Uz7+)wRZQMtu0JIJH3G22+*N
zR4NRsDklM=g;33Opne5TJSGI(b{U!*E#w~&^KwcOw>su(wP^%s1?z38v{1f#19J=W
z^E2XU2G@*c3Y<@S)(UWeWD)BLC1g%Cty0!kp~e6r>r|N4U1jKl;5i07czwM}uE-dr
z@AE9{5UTKgojblQLphaWTZXcL)v{(=hHlHywT+=|8JaDmkfGYF{*5ur>dBx>o0ocn
zhV`X&4D0k$P*{~=n;-e&V(VRQ+QHIcanFjYG@zC7-FJ)+<=D`vJ!O4tkFwa_^=@)A
z;CIJ116muH0qLu>nlm}hin8Zbl)Y=X71wV@PP6ShW=Wokk~tS5QHI2qNAjpwGZl)f
zmS)t_xcTjBcKM@CU>A2icA2>W+n~q-WT9sc6ml5AR{(AzFM-PY8>2=(h~FQ35QB{d
z@yW`Zgb{~;t>Kx7Y&|R^l+=WNXADB`jO?K~8TJu5&vHwTaeDD$S<<<zGG+@&#au!x
z4vxfdB5{_gJIX0%`(DltN>d}3zsLvXf5sXZN)}$p2IjxAntjQ_AHYipASqpnY(-f=
z%K0;ReisL(k@68rAq0ojrAx-;bLEATFJ!+?zc4ExgUVN4ezm+<p(nBf)SHFM$rq&4
zOnMMuUpJf(-nM}h?J|;x!OA-AmZ2n|G8rw~C_+siDTGc6J8I@AtbhC#if8^eRLC-(
z=}4~r6S=Y>xw%cR5$Tn4>;d=+*WacxkXy`dq1BBz)*{y<WFU^4;9UR$4jgPN=DZ1V
zlw$pJ3|A*g6f)_oz^~glQzhYK393&5lefnkIPVBMa&fp~9UI`O`8LjNAa4-uzryyR
zY`KLu$*<8tM#Prz1Yyr|jNaHW#7=QN5tI`bc~RI=1{VbboblHnBU6e9goy-@)8_{T
z3z~yqM+yqM613G!&I+6=h}?b3b6j|~f`d6iII=ja1#a+cZ@iksT!~9Vn>e{hJLa*;
z-^1}MXONJDWFmR+QiRR?lng^bv!>@Xkl_c|vuqMRpAa7l4r>Ehm7!5bUdv@9I#D7h
zgNhv%<Yt2dj(k~IogHK&0!(%;$Q)VSDf{P!`a@QfnvR-WNb?1F*je81@aIwC9fTrs
z`OE}@qv6DS+5vhc7^JAl2A^Kw&`zJ6@5B`Z+JGEUn$yi~hTCSitzo!aW<EBy%>2x-
z%%rntaWq@;?Ad%DXxt3_Bq(eg$xp1^TlZZY?oc24s1F+&+y;R2aecSrn6cfv_!z>8
ztp$Src&nn(P@COai+H7W24v8zcS7sWW~+Q8u#N3d!{nVxcZGgu=slkzj1?s84t*+|
zeYEE?MPrrfm_6kj)zU*T$|kdqVDOr{h8n#7T|wyqIvb=+VN@%G;@f|=Zi~bwTB!}S
z(rVz<01M<tI30Q!E?#jiTU8V~18;M*KChb}6|K+d8e56hksIpVGaRCVlYA6nL+^9c
zzM%mf6}7i@jjcp&?==fYwk$^qCH<VE@?8zssHl8L*VtTCawBh0hLPC%CF=Ef>(*U5
zA7u5{WVK;w$8){ro5K^g)G|=1e-@NBo*7eGw^Woayvb?(V-5VMwEkON<9=xU!=Y9-
zMrZQOGnG@Po>I7|#~hJ=*A0$}$iM0uTZzbNI+H;ij*ZL;vsN*jIcy#J%;Nmw{2Wdv
zq9k$JjzhkWXs*_L_VnrYs-ZBy=lVIR4(}Sz#5&$OplfU;siv=8mgz#w%F={UW)G?0
zu*novMk$&T;DZ{SQF-P4y2j=bU?c(a*F%i@u8D(JG@heU<|SR@ekgNpsFk~kgN4eO
z`GPo*kNFT+b%UcK^180Em57WxotKU0Ru;M)tr6UF{hU-Cjrgdfigk^xB-N8^h=YaP
zATLOR5lZIhe@P=VD*CCevAO6!Xu{yPHJ+nV=G(f){ZQt&hFZCcFgU$XdG?uH7-Wz6
z5dW!ea8yM8Sl8G}M8=)Wt`~RO_3k|j16@BS)!%ExM<vydbd9Yf)nmiLfDQsJ2!Rbs
z<mmpV27Xj@f2wP2F1lMb`S(e~;PeYhcgkPnba@P?NRH!$Ki)c~YupcA9vN!oZd3oc
z%BjV|)R&JrA`7~~Q3tr9YiuPVA11S4*e>W;&MLAMo2M2}o*uRh;*&|IeWn|zk-Wb~
za-v<(Bp8)w=XH&(B-)O$!W_SH@C^;jsNjD}*VtU}AFN6Jd5z_$bordFaX)nV>`*Ir
zo78t%&%dD?92Jqbb&ah=<euhZu-HmYh6~!@zrRNES@B&>f>DX~9bIE9iMGMIon7U&
zL5UpQKhwaEiteB28k>vmmI;A_Dr16jD3oOEf0@(gUo^I((&wLajr*a`KMu8W7h&-1
zbCtz{a-Tisi0s0~kMUHM30-3=5&4BR9e*1y@MjwJg|OZ@qXzT7+Q><EToYi_v42$8
z*g}%E>clH_+vTyeZ-GClp&1qad0k_3;othXP1#V29rWDhm-+Z!*Vv9qpR2mY{m|#~
zP%C$v+ovk0&g9zrv+^-VWLY;jDk8S7v6YD2%e*}8TkTHQOVnWAR~tFWzMu&(D#_l|
zHMWvukH6|+1D#5}L#7k9)6CDSJ+q1^cPW*l{Y{O;sAzvp*VtUN?@=UdT<xzY1h7_y
z#QN3J-ISQoySSS|m9w(`%upm@AV_~$WoATWBZ>XtIQ<D~;faf%9m*fc0ZI3%Q?d&g
zu*G)+hv#M)rz+>Q(9n_fGitL>GEWNuv*A=)Ts(|{J&wzL`QE2~M2j(5eI2qro4yAO
z=|!&G`;Cp4P2`Zt>yY>GX?4Ae6X||c#xn7k(0=$G>0#ux(Phh3eYNsI^+sN?obJPB
zZOSmi4bn&TmC(onq!TM~>iFcxaWxt4R#s=TrjzhzVLxN<ihbltI+fk;>xVr&AfXjU
zT;cLtJRHDzRB;viRU6J~FWoVR)669ZRrI@z0{gSaBy{m~DE8}Bx^kGlMm^5ENFQ#X
z4J7ngg)eWVhv=Sa90XWhMZRIX_f~os+XbAUiaTy8voAtx$K@OBZsRvq%Dt_s`=ZfV
zRw27U`bv5b+iG#VRn<i<q1(H-Gu@AKvGgMjm$z0OkPrvoCy^?&C*5OOwWNve4&%sV
z+GCCqIBu|wH&Ymkx<&X9d~&BCvNCasFIXa-O5#TOOc|+(z4U$kAFF)pOw~pJ7AJ6l
z`~+$pC&V9-WUuC}?BP~=>Ei=m2AXK?nw*$O56gCkYrn7m!$9TqA;JNS7ug^YUCG(Q
zHB{L(2%xx$=bgAuJbgqagj7w7&hAGy4tVBDIw|)n5=9*BhLzl4Po{@y7Gq1SdX?^^
z6YdY0Njs=&egt9p1-3p~Ax=f029NL?=5!kU{8bQHiRo8yAp=T1gww=u5i+*I<~5m1
z>TwlUzBw`T>yfI7TM&?121IBNHY)B_^8|D5AdtavY4UPUd^?mClRH-F+VNwwcNLu=
zX_NOMSst}!V#p8R;&lv_e7v|LJxY@eoZ~gYh;*Y=+`IhHiS%$_1=!k=rR)+ixcVRf
zw*z*X7wMxaB9GTmMQ+OBORIb7d#NM?fRfeN!ufKTesm87&J=%Y{u$8oaM-91G*55x
z=0B*Kr&M%e*<~C=?O1Vx`+zi<v*WglxjTgvJj^0ajoX(#4xVgR_2kqD;8Y6YkUQx^
zQ5UCDtX6HLv2rRaZp?7$gy-}i4hF?U`*ea?9f44f@UwiZ;#xBUn}Lb6kK<%Iz=PqF
z^9@vH5>k#168`S+_4RxweI^F(fP4HpU2Lx9!owseqNFxZjIM0OSYSgkPe!_Gez^yh
z!C9&%W^fk}*p?NekN4J5YPpKUAjhrZZaosZ4E0fJnRaVq18qYJv^(i8%Ui(+K1#Pm
z<iufC<zqd7?Zf@wBFG%v!%$22N%?~b58g%2G0p`6r^?GAm>fre26!<>?lgMEMtLdx
zkI;2$b-IhTOuvx4G?o6`U{TKQ#$GNZ!Esr}u2auFqf{CGEfp55<S#;4tabKEI@DG1
zu=&<GRHJnbxE#%4!o_I`4Rh-Sd{$XP>5h}CYy8q3Uth1|HQHOJ>m+@+4|fl@rmRD!
z)<@CZnxJ08hlB>Cp{nyW8`yc;yN%m4=%xH!YW)D3TI>6yu6^D9YU?`optP>iKmS4h
z{2TrAJu30%_$NuHs3u*+!4PrXJNDP_aJ4&J<qj9Wlh{w*?v-S_2ko!gWp32|mTZxZ
zrsft~v=Jsr4;5!2gLnv+dq_P2ILJ37Y*jhQPQ^DQA^0t3ZM){Xns@o-dj?+W(o=mc
zPN`nG(!+i2^Er?`NWu3Gwv~N>?t-O&`Iy$c15ozp{2H)2l#;zK5`W|UmoWfe>|KLc
z!&x^Fkm}Y#5*H8a6`}9-0sWe%)`9Gtd`$i()!L8C_UJrQbOdz<>R7s;Curm*P={{!
Il3L||0VmimP5=M^

literal 0
HcmV?d00001

diff --git a/.doctrees/examples/supported_conversation_template.doctree b/.doctrees/examples/supported_conversation_template.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..2bced7c3610922d7b364c46b67fe08787336f750
GIT binary patch
literal 100788
zcmeHw37lL<d8cJ*MjG9+u}$zn+m0hy_RRD#CmD-a0^35C3|Qc?$D;kZ`}Oo2_2oXK
z8B4PXak8=Ckpnsjx03*i152(YunFPXU&ukSgk%HkFXuuMZ?c4BlV4!7xpu#*S6^NI
zs=Hp#8%eguzn@#v)vvz#-dF$ns_OsM@jXL7y6K8dSCIewJ<WQ_t1gc@g+jGfaGM=}
zTcMV3SKMl=^R~|IZ|tmg4)`Mt=WMgqZsgsLe+4{|FL~uc!>x8sb^P7r@s?L^j<Nh(
zn{K}4)v7#uX{a<@+Ip&UpfutSwY*l@m45PH`e3_WuQgh3A#tQuJ>xc-4#}N3=~n7x
zr{#7?R>-;Kq`OR1>Yw5JJAu~2bq928bxQmEkt~0y(q?~{y~_!IOWCWsofGYx<8OCb
zt%jFtx4;JEoi}@hCV9WD`Lfe&LL&S<RCOxBU)u`py7-cF@~0bakv!A(x3-)`@>8d@
zr*zY?O{E+ChsH~_iaXwDSF3Jg{A{gpdi?(5_mpdA@gGpGIUXoA?k+p<;x2qgyJYi{
zU9v4^m$6ErQ@Xyivvfmgb2|qH+L3FQ%Wf-M_L`#V(Dv~A2>BuFL4L{nqGm?GTFzp_
zsh6am{UZx~s^`>!Te+g6vZAA?=<PubUg7VoyVZhMUCb_nt)xN%9YDijuS)9!amSUv
z*1FQcBCjA8DcyLYbQ7u9(o3N-OE&|AZUqGTTWHz-mdXkk$t~Rm>HFdT+o9r0ca#o8
z_Lu9K<cIuWP#tQk;~%iCrw+6plBeQ$)y@fjD|~6K)Bzn2&>vb(W6^B^s{2^6<L`i+
z?sC2C<-JztDWdLOpzf^sZDqyD<I?~~Fih!A$Tq=XGZwUPwDf^eMolQq;CKJJg(D@W
zb?^PhM>A0AdH8#=Tp7)<>OF+dvD!SMeZSTp;n_Q^g@PVUv-$^0CrS^5R#Dxmib{{z
zS$>4Sipj=hvyx#p-ok9`1vXY8sAot+rC^34s}yq`?4xyX{C-vem9n+<hl%x%-!GZ|
zr_66<HvQAFntqs=zEU>X{at1abu$ud4>7yH-PZ1yp}*a)S$Ku()*p1;(_r}m{Mm#*
zt!=*zGfUS0UGuw{t^b8st-lRl#AN@^n{^Dc|1h)vuiM%mGxWFro`rkeO2q-=FOq+)
zEk8uF39~PeVUPK(%$EPxSS>$9-@;_}|7%t-%<Nm4*}r9LcFbU7_WM>UIb!w_`Pbg;
zG@E4h@xjc#bSLDC*Yfm543=Ly&{NGY<8KR$zjPbR*-Kf{2NQSC@|619|GI^v)t1|+
z-hVs|b>Pvz%aycsJ&f?|QeC{E=Sv!!z|un8buq$U$W$K>^;9*iK1Nu5oQzN(ERRKf
zT)l9A*{L|Aq&CXIU)FW8m1UJ`;w?R2&s-C4in}JZvX?Q{!@uvTXjnaLXZ7%8gnD3k
zBGdyJhz6?L4?|f!WTbldm7cF>u7_V*r+Ubk>fslADjHS~J6JvZLWFu?c`WMTr3=Ts
z<yONfCuWw@>BCSPmEij*`JS_G8@m)SQq6q5=L?%_<}c%}nO(|@nd;}yd#W2&KRa3d
zd^JM-usjy^bE8y0(`xlh+t<!cQ9ILzrOJ89l>?2Xmac<*z4}Q$$+n%r?G%+Woio+V
z)jhQhtD9Y{ZuYXAy>!ODo#nBpn>`D!E_rk+4Fp_-Y4i}yCRM}Ho-bx@BlpBz4MX%r
zO!hzAQ_V2@?`HOYc?A2jJQnufxA2;?Zk2ZBkJCS_z147#QL2LFp6_O^f_B_hFdV#!
zsSY0Rsb^Rn>|u3KjZg<Hk3}7fEIjN%r?KK$+y2UjJ<0AL?D;-syT3nfyI;BDnQZ<O
zJ#`GT`Cex8_e8Kc%VS~ly<qdv)Yv4o`)F#?+VsOgKFR!F?)hqF^Z&QF%|EmPuQX|@
zf#2z=W>^jEV>R&G5o&<tv8aK43y-+<(n<qr;0gMNeH8>5r79TOJXjxmCFJYXpGEG2
zhl5u!)xp0-u7j&s9sF~II$(LQ4qS!&mn+&I4uT~ef72*s{zoMmmT}|e+}PP8nLgf+
zd09$=xtCJJ=D-1vvM?)#Z+573`GL}2OptkV8VU<%(wFY;i3IuOi4G(*bq%xdA(j)O
zNP_MEwu1*B@>->YlW0Py#H}PC@YQq{-Gc{dIOjm=+R`wm@s}RZ3MavwW~))FE=qA8
z|C&&VL3rp|)32hCd99G1#QM_f$<sF-yOzBqY6vfB&>Z3-HywjU(YxLW(Tp-B!;Sd6
z$`D$EsZcgouH{e5uYK>r;_<s5Ja(#CtIwSeta<*>qt(Q@bHSG^Gwtf?-1&CXZLlB7
zpZHTYxw^U?4HcWbz_g|7mmUXOHMLloj2M4g*{LqJq52^nT5yX_y9^<#k>yJHWzD)n
zqQcwi4Hxu1<C502R3o-0fu@^|T?gVNI|+c@5CE*!ZYAt6_9PK1h4z*nf&|`5UQDnR
zCJe1+en24XO<ECF2)mvk4E56yhUv^8RtQ6x!g5P>`E3u#kg~yyeFhGSpM(M}L9z5b
z@a)nX$;bD?hX#QgOc1#K;K92S`5Fm#w_Nx`ME}FR%+G;rGXkzN$U*=Ze;4)tajgKW
z`cJU>NA2|b$Mh)mk1|d5UzY2CaF;T79+o!1ssN|jVoryD&o<n8c}4L%7q?K3!!JOk
z#Dc?{O*p*z;KAeI(LBImqETyC3(au+{jmlQ)dTI(WFYRk(goiBToY)8x7!%rP#Ya@
zm>vahDAR<upSg&5Gk1b1(nE5D8W0N)@t*RzrZ&g?KgWgn?=WG$bMWAjS6y-vxJ>8Z
z!9-&F!oq7|kqNAt%Dblylv=HN^JU}XrS{??EZZqUM?aRYji0>lb+0%+&K6B%MyGP)
zFyhILr;?ex<4h!z>1j7tEX+(yCet&sc_&>=PtUkcexg`RPfa8z$6K}2Zq<7N1~=gG
zi{9ecQnOaQ_5O)8EW!zvi75lL?OS>|{oU$IV4KsEMU&g7iL+{Q2U(LtUGyf0>7vQO
z0C+Pjnv;itTjz3(#P}f&PX1je_wClo`NPFpqvF6QY)gr(pin#CTZo+Nm%7zjx)pxm
zW2ViIKFYPBe+x|}2GC@PEZ1AMRi$*TziC05$qHqIm0M)R8d>o~!6@5tghLc-TS^gc
z@Tv%J0OhnBz~Iy`HstAO`97lQO~=AO=5IRH!$kdODk#(B*Z=9<Z3zGpPVN$375{7|
zZd-*Py<(zVtJTMfUZdH9B)K0rcMcz3*x#s?-TkK$tE<c_i(jCTw_p_}{SgGZ<)#aA
z=FSh68x{~21I=J9A<-PQg!QZ0MHkjr($%BcdIN@)L}8`FAmuX&9QZF}69laVJVgN@
z7|R$7))t0O_pUm4@E#A=rxp^g3>LVC53!nBt5t_syHyd`{`F+x1fI~uS$X>;^Y*BV
z?(H#Mc>Deh5oH=0;(P|;e{$QB#`F)@+6h?5oNyAP&&}0ZiGq8^E!XN`%iKyLSV;`s
zxKkOEhE5NaYBbkMt^IPP4FK_QG~rYtio36NHc=}VdWbC4s{3Iozvw2)-f1^cYFC_U
zGf}9GL2>RfOfnKq6AG<+Zr)AQiiu{e*g6aEbtn7<j|ACE&KdfmH5h<E^M%KVK=4Yf
zlAC})bHO8P!y(?>bgNBw3>)!=`35u}cK|;Umjg&C4nQ;vUs`7CkhR!^Us4mLh1lDn
zKV<YIU*X<^BE`k6j333X!Vi2niXWil=oE7T&#;`}1O#^#jU4?Yt#H++t&tS{j3(U5
z!%ef7L(Oy#hv`u~9Lf|1Qz@(b+<C8(1+Jk1TtndA9<7E^pzsuQ(qtSV=O82(d0uBl
z=`-wkbO-)LDT}wAuRs;V(s|B6+Y#oI@D`aXeO5z-3OO?s${LXQ0nwzsp%q~Tyg3Fu
z)K3Q<rbhuD$`qDUftMN>Mg!#Jpx81L6DVMlM>r^=vZq&rM(fI+zC<g)3K4fRM4)y$
zA}~D)B2cCY5$}=_u|BRyE`c*v#DDQGf?y4GVVAXz9ObFFP!2n<!clI^&3=~#KGoxO
zaj?wMb)^gccCRMT3eER0G@~{;nlU{Jno*_!&GVaK>LA1I@&dH1S!f?lN!{>1pSP8;
zhNI=fzJJ3Ubm+95(MHg$-Vj$Cz1P%6-#svwTq!%b8tjT4f1pvX1q<XeqcevCMmaN?
znJ6SDr_!)bDdS8|qzcK@q+6U#&N$PlT+(siugQs-855(NPEKdSIpvDhwpC}LlgG?g
z=*hxWv@~&6uHqPT6{w5uDllER3YEO&^f6P9e;0avzqMYRPF~8uur8EL?x#-=GPyH=
z9!Ko;(3xbRX}}~e9R*pO^p?nCW;rQE*zSYYIxo$Gcz@YI9j`mWWePbAQ!0h5+eS>#
zaZWDphbD_%P^O{#@qYtI)gBb`x&3kvus?B!V1^lId`DvcqgCYxK^enuVsu2FA&;&S
zV%n1a1(sP+K<ZNV16v(N2tR*bC_poK?au(1W0|2IH+2D)^zQFytx9veQR&?;YvQaN
z?`xRjMO}2qi|NAg_HQI8)6hufpE8WmB5}LngQQ?krP96OPfFOgXMn>+1%-B#!r@OV
z1z!rDSon}ft9+oiQ73fa&!bf&-!CqAuht27Ro{rqRlQbmRiX*|c<6v&p+$;DK{FEA
zhJqG*{Mo2NB#1%9)?t-`>FKX14gC;vR$Ew3a8`rT(AQ{%tG;)QH1rLc@SbmmYjfw>
z%b{kvC&u(Bo)~2cgQ;xbtLOgqlkd*Cu%P8@Kla925k4T41Oe*23+act`RGThk5*p?
z-1Ey|xi0Z~@`_yY8@dxee!;v=WWczwbl(p{n-GSPkQA9<KCIQZ3MMmu%o>OJ0nu2F
zYDHK<>J)<%>Zd~r)1yENWeUrwka}59NLhhrEh0Ef9*YYmZ!p0mD&<_$s!&CVaX~Hv
ziSc)VkuzEWRv5`KjG%TpMld}JMo^{+BX91Bk?3ycZd|uZ{1P8uNt`lxGngWU`8e*e
zE-Rflir*3!it{EE+fvH!*1)Brw~IPvsO?G@Xnmh1&<d@K46Ue*j#f;Mf>xAiLhIKq
zAGC(LPvCw@-`leh5PsOA5xfvymA9>5j;n2zOl|AwfvI3TjB3>J#__zY(buO($+=|#
zA3Qsenw*|?XVRIe)YL>SKRY#(o=wh971GK4RBCc+W-5~`Oy}|@J~%TG&IbRc)-2U_
zuG7A(nd`}-m48taXVuEftd*lKdMn3t(aIyU!C$r3%YqH=r%w+y_<s+ehmj5L)<b85
z#S?2_gEw!rowx;T@Hqn;d@(>S6T)GzQVC(*JYu43B7{+<q4SvkB5+;6tHL_nDy&ru
zUzh8_%0RaWXRj851-wny^yFQMq7y6`mbhAAb#t|5mhL17Slk{AG*yLdvE&Pt@473o
zpG@cWlk$Nh4!RjL3r)HkB#g`OYGLyDFbsuC{gY9$tG<V85*|c#nPgtN!j#K8MqVdP
zj3D&ISJf6vZ`5t;d04l6cyI%Fxmb9Hs%dy&878<xbGKTL95qak*2G!$$PLybqb_=n
zjOn6B?(aHLrpa|Kk}yFpekww|fOvzztI4qeZP;ub2QZJMgEI6`Kvy}Jg!VUUjg#Q{
z4U@;ZBERR9i#wX5aXFf%;%GzzghTAvA-x?z^UG-((a&gh8={{bDa)n-T?fqOqxc$1
z@h>w+^BBtsj%HAb|8cEw)o-qm;$PN;TY1qH_HwA1?nN;@iWfzhlEFsc`2IqrR;AlR
zq|Ji!=fR`i-6R`Hs<pFsCg^d%u-C+G(!nV^b6q>)ggjn^Q`Vhwd1Wl|*b(Vmf|Km<
z{l^k#VKWmM+$LayO`}Ud!&@xDh_(nPg3zriFj`FnqgAI0yLMa{uEKs6*waGBpg5QW
znd}IQ>Pn(ncZmjMX$5TaY{CB9vBW)Oryj{!a9a-CxNv6zcC0(FpN{U6gaW9@1=OBz
z?KN)YU98m>%kDTmjTp{^N{yy-qaK@W=ZX`v>4{<{l_?Y_Q&UsWpi&c4)2XSHJ2Nxo
zPEHonZf>G5VVZ5HCX>^*kFg!S;xwD#6Azp`dgPV3bql|W>GO|9ff+QQ;Fysoz!Cfi
z<dhrFM&*+YxtG53uk{bDtqDq7VBe>q19k~-We4v_&E9`eSffSEh5slJ_z|s)-j43W
z<msD^tujYP6eLd0f7fN~B6=89T`ae+%a@=LLw*q!{Is<N1G?m_fi78Z|0Gi-T@|=#
ztcRJzP{Tx%puE5<sn7dcVdrBNd^(Iy)GJ7qeu$V(_%zYSsWHbBbI_%PgsbQ12CHnX
zm~FKi)kCCv3EedXu_6AD6vU2OEFbaxDZExJQ_JrLM=y*bjg%sXu>VE_sXACPN2uAz
z&-{RB2jA3+=<P|^iDgeRPlEdCLncf|l-QhDhBAfaR5;C1IIR&pYZJnuvU6KZQ29O+
zRHBBluLq6Rb?4MAS^-u#d5Yl#wbOBe=}~ZkGEF#nDpt3n;jIGqkCnr=dcfg#a|VZX
z`6$HkI~x~%FPQLa8^FF#gO%#xx&qbx1K1B}0<CcR7KT&QM#m|pN5Lt|G~x6gFB6<<
z2oLR!NuXLqx0cJ;F1m9ft>wRP4Zz!BF|Ky_gQj-)u7QKy>pidy4|KP+=BxSxJ)+$=
z(0yJLXXOpv%De&UqI&~O7v3Opp!+S>dRYXz`{~mo(EaWK^e_gxyY<im-QtNg1iF9L
zTIb7W<`xDiHOQ@-MxBjmQ!si@BFZ#$F!K+>snp$3Xb~80E+kJW!Dc0X+i<+5n{}7T
zAvbWI&4K+724^gj9Z@nd=@$0iuI^HTAb7t}_W>RqW*e+{to!!F&37dZ(DS^!YQbGj
zCT>fl6Yz8L_8<)2rI$>jz|oj+Eoq|CCP(>{$+c9#E>Qw=oJYRLNe?G|9UB`9c>^$0
zsFGX^<e}$!k!1nYCJ?Qsy&#Rg@<YVH7}68^Yf}l(u|jD<j9=?kNl2aw@3jH#Je)e@
zRmjFOt&bKFa#A0(EGQ^^wF4=Bzg9Zy)cIyMswH;c8}QnI7j90B6<&OB^c&#kW8q`q
z4&rb=#xm6Z6|KY3#w1Zg{ePl~vl^4UosCIQ7kx~E>0(UMzXL&;rVeD4g!&1G7a^pt
zxsYHsAxHN$AEiCdT#j|jzvrYRC|nQwTPko#76YiS!(Hu0MO@@V`!tM5CslkiE~oSk
z#VLu#>a&Yi#hL(~pe+bsL3yhM0l{UbxFA5k`gG5TN*?y~NOVEKdzm}h$#Q}_8gxOx
zjauQVcU|LxfN$4?_k1(l^Y=7+In+$|wU{2o*P={eFqPeY&74baDa?9>Z1PYr)Fw1z
zFeJwfeZiIj@nq`I>kPpYDPFP;l6e>XonYQry6B&P)?hMCO}ub#)6BJMwelg@Y()Oe
zdevfWbu}>u!GB04i@k&#E5lv%a=BSw6_;7xr<hf0qLCv}4{HrcZKdWwQ-ny=V_Feb
zZS{k!t)hN<TgCJ!Z53rofc#;IlJVnNo_GbE#9dv)17AH7CA~92ohr;a&pInjGUaUY
z^)Yl+O_NTF$$FZgP2-B1W&%#it&*ka-M=MZ4LVt%4ktyQ@d|EXED?RJ%DE_3<!e=*
z0pOg4oQhvxxn%E+5&cW3LC`o4fL$&`vM4#9=EDr6fkj1MPvY`+aduK45*(F_cJ2~>
zP58&i-foRjxbJ!|SSOY);Umf`OTH^=X!IjmeW|X?xPzB@F5~Y45FgbFuyR~K&oF`7
z>5dE2nd5rKCWMYMO-OlT7*f_U$Pn^UuzreN(_;#cjyJ`{5|@D{9I?L@7h-?Ggjm~P
z=pSf6Qjt19F!ZaMKr3W^j3E=X(UFPiQILr;P00MRK$7rb7Ymti+`K5>Ff^pwdF7{i
z!kt7tWw<|IUyrw;e~haQecaTB-ZO9@vkGG;&l#uVGjgl!;nAa2xcIT1XRC<w$+VN1
zOy{PG)A>{?GvlOYrwh}Ln<*Bi(}jGYSe#AGx{hmFMLdxV4`puMuFk5})*LyMxmy!w
z)s{cO+A`{*w`EKhZ8>r%^QG2$S%fnC>C+>WIW~YE#!zOr9(pKKJh6sQ=H1ph2cgUr
z!y?r6hq$~9H4Id0FjF^<I^+!tW}-|(2QhyRZW|*JsxFo!h)(qjIO>5n5^oIVVZmle
z<MEDT;3>TNf!k@QgG|zZOn}W74RRGE+$7bacMzSEk9?$2GI=9#*Flhh3>uFkdXum8
z{tzlUhy%j#ozfNw5&<NCop0619b_cRhm=N0c#nq?>4suiMVi114po@w!m<!`<lM0F
z7Bpfhhp1BQ3)+V%hL7~pjTDYYWDZhqgx1F`t^yC(Z-?(<mem|!m$Y1r)bYnr7O9i4
zw>b&vL{sT7QfgdsO;z~YU>I4!+bCfcdWC#>gHlHDv{j9E-9)v~3^$q~xkHUF*T7{9
z5{d#!Itp-2=Lv|l79qvFFq<JCa6bYghgfKqPnrhYmcemP8^UO#+^E6vx+czQl>1pW
z%0*rDQ7)#7QEvYM3Cc7LNFE`<aSnSiX^jvo506R6!E%+wlOS&!AXfzhO;ZnutIsHf
zT#_KTG;yBau<|-xOBc`++@1VDT<+w*DDFgwan8RU79_5@$zYK0ma0>6Z+Thb7Ivei
zn|-{Ihus;s+?lwg;5PFxYA5R!$rJYmcW#nf`(E=}7}pkD8jEofiyjHRkYPN@Rd*U7
z5>|CJA=9~AS}S!bzm{8a^T7iRxX`fC@?7=+$tOQarV7=?5Cu+>n*rfzvi%v3aU}|{
z`vRFhx2&#$0uA@^w%2eAAavmts@17}0nE$HSv<PN>lI1<i1#bXJlh7sHW1e50?&aF
z#$Dx)#pNnLr?^TnBJqbxE0vrhoeIJev=u;X6^_jcppm0MS2U{kA5UW}RZ0oojl?#8
zF3RwWt_1o#8+(0(<%F@<peunsr4_CYCD*tT=yRHItAXWjv6n;5^noR&M;Tb6Okpsk
zm-LHs%|r5iBmqqowirl0o?gyD2omx<`4^WCWyuCba=Rr^MC9H|Tw8e2isS9)kDwxA
znYsKn;8hq=LS5v=EnnBLp#m?K#VvoY6=4OxFEaR{emeXxJqr9#rm&m}zX=U~k-#)q
zW)6*!9Wg=UcTLcUx_02jpwYT+tA43gfE6yj#BhPy>A1l3D7ZkGCS1H%!^QfzCE1N}
z`p2}OAk3Xz?rL(ZPsN4xFPpG#yXxST8U$7EH^8cc`!#`9nEnHXY1Br?G^R(vG|Dt!
zdVZL2I2WV09Te7bzEDR@78*#+Kz_bw^U?C*Xkf!4Vtm9~YCf)(`iG{L`rd)pHh6SY
zU#^VfAL-FN97aBxPUeCG$tR|#3lq~5GnsVSnQ=0?^h|Looy%m3$>LO@0M}RM3&|Xu
zoMu|wFl)G;vZ1wc)oDat+_0jFvvL`K%3KEOqPq-C7cL|6;)XX_>t(UHp`SiI7B{?O
z06mP08@lz-7dMC}*08wYKU(V?EN*zhu($!Cy^a`{S2u*AN?qNcTSrXYO$TkEOhZTH
zf2Dh$c8|#TbML;&=k5FPPU3R__E=`MUo~|SmhrgX*IJ$Cn4`wy{!kNV<(U71IcC&F
zcg&bB9CQCBgfb0HX#UOSv0Q|ZoM|u>48CNx-Snh1?6M8gnM4JJ>ft)m(@MdYBycAU
zMEQVnqguqW_{-`N%fFAyVLhujEYXnt5$w3r5|ynaYIHWEl5Obht-(d?oXqCm1BXng
z?43JR>ZYgXqEz<RncEs+Il*lWN@c%9D_r%$YoxMo)`a(bGu+hr*X-p`Gu<a+dK90G
zGKIlZF7X6Rt{0A;c<|(@#N6D24?lPai080H{d05U?AI{bI<YWLWuQ;-;}*+1@fk2|
zES>l_0Dr<L5{e>I)Awq1tpX_)YWjp$gcY3rj=>4_)8T~aQQ(9!h2>N@-6r7F9o*<b
ziRIJ0^4+*9|L;weAC-J=X%(Z^y>Vlj{l>L(S^-uy|1DPYsGVN(m>#9(QKqTpOQPmu
za3qH7D5Z5r{>3Yh<L|9;;qU*L@MlX1|D*;B6?X#=!p~>|t#J0w3}>i~jx$V;f-{t9
z!r6bgNH`P4XpRhcS#)F}DAl)>Jy+D_82@x!82=X&#?KE-rNYuLXS7^)D$Y0?T#ioX
z0yec&49L`q<5f*WX(}DQ3+xLT_SMD|nJE2|CeEfQ9lv4|YfGq%-j*<3w57;Q>9f{~
zSumyj^y$Hrescglj7({_9y(Jho>&7@x^0*3R3%_a=M7BhnjtLlpy8lViBSFJ#pKUG
zgbJYs4?I6X_Ib(IA@(|@GB{h292SYoT$+iy_`w<rsh6D!PCX<iFb45zd`^ChoPx=>
z)&v<`ew`!9YAT?I@UROZyN<<F<>L8_p~p@VDhN)Tyll?KB&v_G13Jm!oC$JgkY-{~
zRx$%IB(pLYe?Jc6f-O#3PIB5MtV+jOA*_=(?Su-eE>Q~K^`}>_*eV6Pqxe(gq|NTl
zE#ekWP>aoG$?g^=Q{Ib(T{rH2v)<SOJL(J)NlE(sn?ie7%thh^x2@vbc%%J<AKSXo
ze!@3oZL}Y3KYWaOGjzsrkU+<;++-T0T846x+OR+yIz|oUOl#t-hK|EHbVOzJp(Cb?
zp=1A102gW+1>6w?b9C-7LLf)#K47!49JoOzov;qi05KpYY<f6G5_d_0E)r`ozhO4$
zl*xhR<T6EPk?*pBgB1((+C{jmu<2Ht?wGdy=|o&@eyiH%MN@z);K<Z+RE~v6Nm?X?
zMIs_ZLNd6lpG874(#2q<C=H)PLXPjmR`0T`(CP;bg`Ck!SH0L8Lm{i0a4Vm-3*Qb^
z(|sDINAYP~s4$tzUcZ0te9m3;s`O4X*bH_aR`~}vp_N>x04w!9SePFSgy+tagMg)E
zvQxA>Z~~9jS~OdTBr@VL$RR)BL@3x7fO2pc=|4A=<$ZPsU=T~6y_>XOQG*ibBS&uD
zr$JN2eJqii4{1eMq5mpGKPsrBAJe0tp9>YXQ_=s34f@xH+4U@fBki+sA?<1-(xQfm
zzNn!<1zHSYqOWKLSi$T%gc+)*!wl1-z>Etu!R#MwV0MX}Y>2bwYj+GDoBy6|xb^ai
z!D+uRU!FJ3zlp19zS!6_Z6j#^s<jxkNe&P}+p^o1A#RuMU`sxZ-+)aJ)zO<ErblUl
zT&ST5&iiEUcKHu=;%<c@@@%^^x)}q9U?FmN>$mg{iaZ?BZPW`NWM3xr!X4N3(70n6
zz20baT+hT27ClZPq(N*nGdiOmqnXZ4rzd99Q?pZ3`P4+JFzHMsbEye;deWJ6i<9}J
zlbd#?ib+!dHJzL^>@_*4IX~5#Mh>B-G;vnLnVZp<qB6QK#dP6IBZp8ATPtP}LhYwd
zj}Yo>2GGM8LhaT=521=D)(}E1SnC{wP`|H#2zCA8E{9t~0j7pqbwdeu$rNtoLQS2u
z1D!RyqE3xQiZE&rBG9&@s1MW7r5@nxCO4E21Piv2TkFEuj{RZh9wwx#?l||s?%RVv
zURO{uL@zm-hk-)+7(Y7m97gu|t%gB}0Ji9Ws$Gmt@6MS-0(lR-VeD#)1)MO1xemtW
z-Rdg1gjl-OTaDeSWgy4boRijpMGfS<O%rF;f$hf*43*J4FiaO6SbzV*g_``!K62#O
z+9m@Lf<E9$1|US`;80Hqy_>pqIB6gy^{~^L6ul&YD*oEsA8vr7kjSkh?mvD{xpo!~
zh;u4qJ*(9?Z+TvUgZu9f#pS;bsQz2HH4MsSqzy_eC`Xdipd1g1*+k`b9rws}Zso+x
zayoq&V{=++@TO$1(yvDG=rlq%ithZgENeTrq+C*!J6VF;qgrmIUM6=S^P8kx7^k@f
za{eI0UB2CDxYbs9Wi0U!9Ff>?6K89U)8u|DIBwj7dz4{i*IBPzPPFR<Dhbxo)oRtT
zME7M>FFJVeA;?RFzq(OdY`D$l!GrYW50tJg4U6~vvi81ehqzZ4g7)kFL5TcgO{7()
zIF42%3dtQJ)Coyy<^)*6#@Ks;&hq#eLD=sFGDMv8vGuM3w9EPzB<%bTQS40aDL<Y<
zBL@ajE$VL}_w+$yyX{m*YH&+G1Q)K!#x=p*C1-2GK>iFI+v$)4KLzB;uw7|}q!BdB
z-eL*Bo`5^Ii!K3vqS&Zag1g+Xv8Xwx505bn@8ws(4WD@6<k2Iq#0tQ#!c4pNL?CRT
zMc-Ju6HGNha5Ms00gD1h7x5ioZhnzU6Q;U8lBo!yrV&wrsWe1wJ?k{cm3%!xMJ+8r
zC0YBi!KpG2PJngFjyKL=H%h%fziwvqOmZw0zDppf*+6xOktwceO`Mf8oJFgmGO|M?
zy%eT%utJ}6>lW!^#I19=Mq*st^2NW4%D&H9GYgKZuR=XYu7?It#7J_5D5A4m;*mA5
zTuatU2Q1f|iRH3Es?1J>Vnb!8^cPfHrRLvh$H0Y(cBm{GnEx7Kr_OQ8hTg%V5m9XM
zHob<0ZX0+`ET8YQ<a~cZT4&CmWpov+P$!qbuuai}EHw|M)`(1Hu{pIiZiG0OXgFTe
z&AQ8Ec^!1K2lhWug<h0QN$3cX&y5p=*n>lOVI0|P*Xy+gj5cqEtqXxT<b9+&O89he
zqqd=9pZ~QT^aaqJJPFP$7JlwK$*4*A5aD?IU6o)OMmyph3<%)DRpN>G4|IM;e$eK6
zIK8r&>+KYS%xON=Qxir1_a18x4uO0+)B7&$mfy*;KBs1SKdZg3I<V=U(6P<y-9HEu
zeoPZ-HMlv9RwN3^0~>%2NwS^k-A%Ce;XsBMp|Af0jM-)V3letzuTktwJM>qekplyX
zhDh3-#Y~SHw;Sd^`t*e>=zpu71EnGR3Ci^00cU(b*$gn{dp452aLRWt0o2!YKv_@1
z*8YxBrhIEFO+f1(BAJRz`Cf(43QVOz%WBF;EiKR`P5C|;>y&R=tc;*DzF_Zo%ALwg
z7t@95d@kwC7N&~1)NDFk%p@oCx#^jS+1bg|Y|@#@n3hLO81|Qs>{W>d)xk!dIqucO
zSvlAHv2suu*};;TBc`*N<Gq^(m^t2Tt(nElv9CfsW{$}L6fw>mLln_xj^dFu%p8wc
zD;>-npNuebTra@O6VOnIsS{A$PO6Uxn}Bkm!Yil~(67)5XoAl`gIUZAVhXx$Gtkf(
z4&UN@(Py0-Rje$_iG^giT&g%2a6td_D0shErkW?nXiazv;bQ!qVyZbU&onU|W|}lz
zn`q+f&Ss)H^y-o~$~XG4Z(>k&#@fZ>)bc^>x0@`h5_b4Dh~{~%65N^=Y?wf9B0Qw`
zeOl@2IAo36{NAhyw;Gk4#J5A$^ic_>OQRA|p-2_T%m5jPa;d^*N}u7Mg1y49PHd$K
zu~%}_PUyb9;1Zs>^MpSO9wL8foAU4>SejNVlKbG!3u=9xOW`f&r@->Dw4B$GmLqH?
zjlLuAtowNl4Jy=P*;)6WwIZwl_b>tu71RNT>5%}(r3&MzfSaO#8yHE0X6F$3owy+K
zh!G-Dx5WL4R(&c$V%QS*7g_;U_;>^21J%><f$5R(!KIq;@ut`v5-0lu*TugoyVH7k
zuggg#j^lrc3&&X_j&1jk?bz3AQa`}{v8yzJR(Q=LUQrz#ub3VQuUx7LuYYuD@oI3S
z!KNw%HHUn73s4^p>%X7hFUi}~?SQ*jh7Sc}n|gm=^6u0DBbkwJ)S5818mA-c(TZ2~
zMkh{>dR17+(ayI5<}TyRq*5~zlZBbtVkV!;Wr`E<0Xv+M#o1zVGL@W|o`PQ{P0U>;
zGjlszK%^{!o6*{&+DS!b?q05mvuffcY~rYl-o!CoYT}XCe7xFPGYjUfuR=YTyX*jp
z7@4~;MUKj=VRkIM)}>3#@-9Rjx5EF|`bt{RN(et%wN@-(=f3As16ro$LNL^S8pSi2
zdVxQQd?k*;V~FiV=MRw|_8U(S#2rY)ljCX%JXS9_iP9+)vh0>|Sox5x!+JqDVhDB(
z(H%ibSAYW8`H|;LXZPWv2mks0n!G1?5+EK+Pq1X{2`qQxyj5##n%|AO8|NLGI4i$f
zMZb&6=zbT|CBNIhQE{n;Mm0}%plPG4XxnRW{DXreZSM^Jth9*;8)Gm(iVxY?fRAWJ
zT%s*4s7=TaZ^PJi&&|8!C^j<0f<?LRo$wbNyBE||$vFet4anMUyA#{UE>Ys6VT+8+
zhj9n<b8$JCn(AOgv-5}Qlz<h>h<Jj=>LRgf!|D#d=B!(VbK%$s0l$jD!RMnm7`kMr
zh2G`YSyu2aVzJ7e19z999%!+tvB_X;3S*P5bfLn3kE8;jNze3DAsV&(tX8_}u-8Z}
ze?t>)<-V8k?NBw{ePeng_syk>YEzlMJiYsiZb8PQ;==i5anHM)gLu`j3wGoi#rVMF
zqLjrO@0P1#8Xc^V#w+Y6!9Ft0d^PA~zWX~Cn)xMK5mw+^Mew15I`}X>68N}OVLTOl
zsez$1Ku!*fR9wLLcSc}DC6~WTt2EUe#Xv6Ks}*2{itj~Kpn5tgFg+3~xKtA=-X)_V
z2ztfslH?NjsE-uFzsS6^aDCjBUDi5sq`x69q(5myx-I=&(Ez9-d;t2nr3th`^ba7S
zQ5_x8m>vnyT&e-l^P9;WiCks5uKlcB0>`lhsO5a2{W+zcMI(@(?^%wP4;#;hIq0Ng
zJ3CGp1{ZIRtDQb&Y^U$+OKF=JZ8E4ICm-q2>9kmf%9fxy#i>kwYIdeLIh&tIXC|kn
zX59R2K9!lBoS2${b*TBmv^za(B(e>g7vHJ1Zq;2xCbEB06KCZtd~_D5jP5KjU2+zY
ziR@pn*35#)?yFD_BKxxgC}JeCLln`Mpo&M<umtt@t(6W)?lT6Gdz~>ZQ`@1aQdgbo
zrV(_3X$27%YU+WGlU1i(w*|39bz-L@J?^ULHJZU04fy*4xzC8v-*5%b{$Nl>PsyTu
zb$8DuD8Q0it{2<o1n((m_rZu4<hF4S-*fI7`uPpRlBLV^xHC8|t#?>Mz=VTAj*mco
za_bNuy?z54cP#Y&4;p(#OM3rbG&iI5Mp5bg%~z`|fK_kwR_u*X8ND~cbg4J$?>M+r
zljC^GFmx3SJYRBJxmqhxaL>5qS{)98={kTYeC!yEHC4Jl{7DHe_RN2{sG!j9AVB!j
zO2L;V9v~Fs<L8Y!fq7K5^=K7&e{wHN**onfO6`hMZ9+6)3}OQAG8{<&{SHI~Vh$U~
zb;8}%OQ2Q9;;!DNx+~F${o%(0USC8DM2Z$JK)&84TyW*X9)=3YpTQfEH3Sn;92E@{
zydAw%ie&{aCBg*JIQqj{>8jUVBS(K!6K>^$-+^z3s_8x$(<Av{E>+k}rAaReZ_Z*x
z7o{xTW*&oDh^5WEle8IOKM8Y@Ir^H03>9{<aP()iBCNpoUIZU1sDlsFBY}@g6~<G+
zHvrSlpfo^E4ve?N1&pVSz=+Dxzgw#`6&EpZ^zYLOutLTA5EZDNjtWeVgbFUzgo?|W
z|7Q5Rtaao_|K+%l{(d9UZ8`ei)Bva=d;pIAi<&?yL_dRwMs;*VV|pY+bEyVIZ;;ZA
zlnxrthB<VO{(r{RPJi0iPCEm0^vTpH91-|1Y~`FdJbDBs<N-UMpPZb^CsQ-I$(h-i
zWHC22nJN}1W+&5ynW=nXGCi4erc;@M@&1gdWcbm7-_qK(>L?<!^Z%-evvLz3K{tWQ
z=xzejB{vb7o!|W;Tc*Z>o$sqq4|e{>0TeN^^C61p?7Vnn4eWf<TIqnDuN&C;bwjtz
z%ZK7g<>hswh?%*Gm*+xFz0UrCmj~*z3KUv6eF{g+7Tls&1!j%kJ`w)pU5TQ@mWk8T
z@F=6&oA!V1G42~CBbSB%1;rK<NF1dWPLa6uqtFy%nP~s4vG=m%(&sg&p!FS5x%8|i
z&Z_VDPuO>$GJ4;E=~Ca(zp-<vrpCU(+g?VzEgR=<)X8O3Cwx+TCN6jNQPo|EM(kg?
z;z@^h2H!Le9m#AP4t*E~->~u>d35|5yb<Bje>jSxqFnkfp_h6K%L-m<P%izwTIs6S
zT_czN0Zq7-5B_C*J5){g!I&P&2Xm>yW-6U|B<$=I(V2#>e@m{e%8M2-4(G=ri(~#*
zz>2YS(Vrlg7w<1&CNdZPTUyPku!)5W|2?e;EAV_8!Gj9w;KB4r;Nen*@l@~(z-Y(7
zh5uDt!1#<27*V<Kf7B{X#YGHU`2W)iutLSJA}UZl9Tk`!2^Czb2^Hto?otfndHeO~
zmsAJt!mhn0CKvWOBT{X-?At&U^G=%qxa>PMfmVq8O++NBqaza2BO#JYH6ilPE*&Ce
zB!_mU+M;mzd1)rwjeIR1?}T3v@5)=)k+@pe=Z!7w$Jfq6!$5d+N+zNUQ!@o8H=CTD
zn07MxnW<SfJp-%6bLpAn)a>k(Q=FZJtDR?zMD&D#h<>fsDAj%_G7<f-CeEsje*xP#
zDx<e?OqbetWFmUWS~CkGy01b#i0G986fqLfA&Tfkw0L9<M6_?MbU;Km`w`LW0C1U$
z4uzD;MeBAEGg}iE&4n6zjrmV;E;?-8_g(6`@7ogOa^_$mIG!<?5H~#2731RISM@q%
zdZ~_<RVBI4>5}!UUV5IFtX>-(a&jy#SLQ#lwe92`Nv9RIvYmEj!HF;Iy~q(T(!7uy
zBOoS{qAwFmoXfz+hZV`6W#xv-4^EsB!f?ra*tItsHTj&cN#5gs0DM<0w7?gQJ-#J#
z|5440YMpIV=Kd3!IIGU~_p!4@W%SM#)1}U~ziZ@DO|G#7t}*J0cNU%r2EiMv++m3}
z+~L}St2N<5!-7XH{DQk|n{Kt~!WG5xifYEpZ`k}&tYZJYxZKT`RCgm9p+CF=RQc$n
z#J<xMAkVw*%t8v>ih+N`t(R6BoPlRw14l|2_`i<ga3}-+2k347ie&|FGbjW9Pg?1!
z_go_bzv)_)?XdEte~52~s_DKI(<Av(E>+k}B~kC~bCUUZlb*;#{AO?xD)|eCm|4Qb
z(<e9q5Su*S<-QoK8B3S@N2Co1J4o1x%*5~4YFLF&EKK~ER)iIZ{s)2x71Tk5>5)Lh
zr3&MzAiA~ZIrl0+te)kS|H`;3|EtEzkIJ^6)G9`;`|Iw81JB?}cDRFGxwRS{q49U&
z85gtytZM!*u;x)cz2-4JQq6OzrkZ~|9``YKp3l0%&s*+xG?$=cj><+{sC?FlN?T_A
zjT%@~G!DS5Kcxw@Lf`*F^r1RB`Y=5b`nXgR`aXHF(P#8?gb!z3WgjH-CwZ6o9DfVm
zDxQw3ReaspDxMyg4Zp9wxCmFq-{a)n`zFSppo7!V>D*|}Z8^d6?A&zHP3C8vbYa#_
zP2>tGcP5oCWb&EW8F#XnNoCyGOm1S@cz<7JB7AxF&u9%wZK07_@n6)$S+&o<#rBEH
z=<O5JrS=(_75_DB%`8~)z6$kV#lJLwB1Tp`L=l}87muug75__Xr2|&{oPia;7(kcV
z@la@~?6__pG5a*J<6Nl0^Uwcuz>b@VXE<q#65C-6FPXv+a+jROMMz_u+$iUZ3%dLc
z<&1AC*fo*I;w%;v9;5_kj~j0P=6B=4?Ss9=^S(t^pC<-|nC=+wE`^cSv`8Ss>2Wh<
zhn+SPzNTFmO)8ZvbPn{ECBmUH!C@rW(QQ<kggME4B#LkuRvPcuH(eLg1pM!e{kkRR
zy%#hwFC*C<mGiz%6KB<xeiOS=R7USgF<t6P`#U@?)#UIFgTs?YyHWUWjl)sc@329u
z^{FnJ;I4CA!i8UwlUt}N+!JkBb8SbpGi!1eF&3AL_?GG-M8oligE70zQ}YyUr8pAB
z$Wu#FUfOKe>$OJ9EsR13f5vSz9TNK(g{vIvWvAsz;>Dv1k&}=fRm(lbJVR+S@loxZ
z*`4^q&04#WcRQs!{TD$1rrLzE+07%14Y%3stac9gL#<lNDR+)t;cv%h+yZ=UliNKv
z`MdM&M#HVPvP{+qe<!AtVS8HVgue|!?=6`0G&}yjLM`8JdFAF<u~DnGpl|N@!>@PB
zZMWHZTj%yS(gKHT?N+_r>YOO;IpL463sGtfkTg;N+dv4dQim@)8ns#rE=hr1aLXNk
z(|E_<5vY)ba!<np+u7H4qYU46wQBin&dHx9zk(`5pmSrT<6l(-F|g0hm41B6-_>;U
zt!lfHU53F!rSp`(8*Xay>RBMwDv+@W0<RBJ+_FnkvS;BA(`+6FE=_nDD34{}2|F7u
zsq$=@9EsIAcBQ|)<dpf>A%7nnO$1hKIjwdR9~$z93U02w2=aEr#RgTdB0SM>oeHFk
zKy~C@@=8QMuL?HHy48HG08e%Nk#?&%Iy2hz;O`sMAG2<wQEOyNPPI^W8&Cs8)qJbt
z?+Gj~Ng1M2I{uB?FPey@v(WPV-GN-81LnR#{Xxs@?<Ma5*lU)+L}hrVj=!tyR2SQD
zgcDf20N<~4s~!L9hMTW73fY2N2joC89<}r^G@&BACRoHPARtP^{%)t;s%1rzf29X-
z+f`^+>RAp+Ff2{*0F4a@K9Z{2LsKZ&N$rcIdedzeYT4xq0n}c1xn66!Sve;~Nxqdm
z<2Ah&h3uYe!D~9XGI<fE<yLGe(Cl`?%QT^J&~^txcd^H^MXwAmcdFxGOa6jFV6s5l
zrduV_3Q*-c8qQg$s^IIfP5wU8-T*n(MF3*it9Y%>@xA-~-F3GP2UoKjqR7kaBq=!u
zbSF*2e=$zsAh|hKUul(U)v;P*5n9wW<VmUkJelP#$KM9&qqij|z@EDs#k?k-qzLif
zf#RX@Is~$(-Nu;bwu-b2a0uh2R;3JujkKH`w4y?*)H$}zzY#E5$@0d@Um)A86NFsw
z_cm9mEoV7f@)k>F_#g1QrCcM>-`T7<(9j-lL#u9rp{}OigRj|gE$=YX-Uz4Efj&_2
zL3;x$Qxih&)9pGGu^Y5&c`dMi;07)LdTM#ES_BL*zi<_C6wP){d(5j4?3QZ~)Fv8j
zp)Ls^Q^~mnXu&I>LCJMnP{uWRc<OZ4tvFse3$q>Kcd6F<h-Bi8n{K%n__PcDX196<
zYVgGsxb{~VheDM`&*5#aQh#fL&*?nX&iUH{{{syku<OAT30hMt+j1I<F8DJ*+CF#<
zz+wv=G?WUviv$0^=Pb0je63w=jgr5bozjSZo&G3XXh(8!Cte!zH)m$~9n$n~-c5hp
z&`tYYMai+>rC%iD<Qq$O!n^DfI>B9fZBorkAA_HR9s~;C#P2zQzj(pFZYnrLo}uWr
zf0gp?RR^Z_wA<=Ze(9r7T&SwQ20ye(g%mNnKTP^DFz#x{-%1jLzYi?E3Lag$mwbEx
z?tEMNDe|#t7(V`O2tH12g^x$b$H)kL43m!!ZHJE!l8@n?@UfYE<aWVFmVA6_H++18
zeEj$x_~?+2I$TJ)R3RU~3YU~DeU^NT!SM@AcaV=a!LEy?A0QvkLbQ76&&bDNqR`9f
zTZ|BGTPyXz3RK?<RM!hs&kIz?3sk=gRJRLMuM1SC3sj#ARF?}>j|)_X3$*+TwCoGC
z+zYhK3+>X)Feae}DeVlbacDE>Lap%(F%GrH7Gfc4jmL;_s5N#H<4|jSgcyffV}uxo
zTEivAq1N~;F%Go`xqm&d2D#@dum(92Gq48P&>L8TED)mBpkdv%zf1O7v`Tm7J@9>`
zSwS@iDWr44=Yr-Zrv%SIqXaKk$<@ldt<y(eq$TpFM-+)8SSrnQl_H-mMCw4Vag8pE
z{Q|>1=sIb;Cu4*i`C7RgI3yT6Z9PK#0i?eT0Q>!&*FrZ?cN);4FrF(YkoTZU@(!L%
qUC|D&nlHBt#91y9FSNZtI(gO~k`b)GojSB;xeX)8)lNG%_WuDinXT6V

literal 0
HcmV?d00001

diff --git a/.doctrees/index.doctree b/.doctrees/index.doctree
new file mode 100644
index 0000000000000000000000000000000000000000..1a526e3dfaad7fa34a411045dccb4ba88534508c
GIT binary patch
literal 58437
zcmeHw3y>Vgc^)2zd*Ka|5Gjh*kQI`L)9oH`hXX(`lm&tW0Uizj35Wngp;ogyb33y)
zJF}d}JxEiQY{??xQAA0r*paQo4_T&Vr&MB0n<|wPDV7yUmTf7LT+}1wM0RYaY*&=X
zj$W~o@9&=Op547$%;F9sC;_XuxbB(1`~Uv_`|rR1?w)@0ruQEHvvushcze*WY<ID!
z8;0u{X3&fW4X+Z_O*d@5r@7}X&1ae;@sO`C1YYD<%x1g}Z&WPXF?`c)Uc|#~^nPeN
zK`|*{Bi@#@)~5KM{bRdc)dIhA-yNkO)I+;6?s!$N*l??NXu1>LcgM+@V~)3QhZclO
zj(Oi53$_tjhqZls_uhH0R=3?kk}<h=Z)5RZ&9H-pqc0uS%8pl=!yCF^wcW#-9);e$
zT3N5mRedjVjY7q9JpZs3`nnr5bnqMAd&g^so7TX^rZp7zoii5+<2A?M+nOZxxSzA!
zw6+9gJs#G>(6`G`h$f;Y*=QR9#2*h-96bn-iN80wdY%6@XhaQ_Ad&dyEPJPa*7NIn
z*o^zEu<kVDp~bp$7}BAp9c=h!VEXfBGl~a7y-FoCt>L&|wEg~Z%d4B?e&o8QKfd7k
zbK|_}#6M7lV4S}_Zo7uLSgaeU%CTLuIlbQ6ZEcK)N>ab9jZqm5F%X!Q(DvMv&RG4_
zNF#W-DYQe!Og)PGd6m2wcgiG}+6d~Hv|eLvv_?)@qo<;>ZoLBHvTi|>--`cl!~d_w
z|F>UkM%LZdK5KLuuUdORRwOdU&=4fASAD%<rR)9Z5m$r8m~LQ~9dk^pc<#LE+R&;o
zO?M5=oSn5Rl+;4cbIRVL22pFXfF4G!?N&7hdTK&j(R*A|b4=Z5*>!X~Cm7Sv`RuA|
z8X(uorWT<k3bQ^6G@PX|EohjgvBWC!O~=+53skLMsh9y{i3mf>)WCx6o8YfrN5$b%
zQHbh?7=dctquE>c2wHbjXX~QjLLE)ur&^aQob@O-c-AJNfNdG=FSMse+B4!CxqU71
z)-47W>=3%d?lb7Mp#?itwa5-V=$I`g7iM$#L8)w_0n=GZB35>Z%ra?=P7g!h%k7<G
zB;FdpXzIRE;GHhM0qI#ttFU0&RVyr%J;!L8=}r{)+d;{&=bO*QJ2=D=^seE#P{L+B
zoGOgE+asv9gf4c!5OaL!Y}9CYe#o@dbD8xngF4VgPM*|fj*Mv!>5dNbP|@+{nRA7S
zG0hAs#XUkKS6+Y;;9FJ}V7x_&ghmTJjhzr3Nd&ruBQsm9jUq04#!@SBcfk!Y!7(Z>
zH!@MY+1}9EYy%08&3G67;tE&r=1t$X4KnM<Tqs~PZ8C<Sf`+$H2<pAGWFIlaOy(mg
z6?(jrmww{K!@xqi1nqIm#e(lG^itPQ6?_-$(5xpF?zSF-(yX>CT2-laTTU`{KF_pT
zXi{sY(vE+-vUap<gXmt{pba1Ekd3uuJ(a!lSZ@|R=B1V%b3e?24`bmhX|v{nMnhHI
z3{5}K7VOZ{PI`|X(UQ)iO-|l_So2)GIx#hKKin%1<C-b^&RP_u<n*O3a*9{%I6ELJ
z0KCc_hZ^L?&IWnE^)J_=cUT`xcT%eBh0eM@Wc_*%jsFqR_-m~BThHmiT;Vji6(PnU
z<_hnO+j3=_)`r)wTi2yo+q@~E{=2N*wE<9e3gOCnK&_7zmKrwp9K|35&4(An)&8iZ
zhZr89?ApQm$D#KccdMlD(*uQquXTsNvR2;+32>&bvSO2{*DovTtD@Ve&iaSWX1z9a
zyYF-ZQ_@7Gc|tYiW8Ir_1x?-!@t$2)yjQ5nXINn=O@81UJoW|4gCaNV20VLA4Mq$-
zgu5dq27LN3rrGYSiAik5+@sln<^gPhPqn0yuT5b}Xij&bt*rkyLv9@ERn~uMpqXVG
z=&I;HRrm0+)m?cPXqC_iX4?kyn3x9z*NNPzps=qQ@8A!3da*Mrs23*M-l`jg$+j$~
zTG-!<H#W$;F>>M3gWMnt!PyQ&!2C}+y@K->*_i7r>y(nuS}*s;$~MM|EWd*xF@rA%
zM^DV+CPYTO<%Apho)PhYLt-mhGnM-PW_R^(vl$dixQ)*?YadA{)1KRf%7%68sJFv=
z)BQO%c2O+AoM%%%^yt|nL@-OyE5yk#m<byYL={WZ13tfBY#^Xz&m)GQyVVFDoqWyG
zk!zTB&kYdu!U#_UwvV_)#cWW4x}#NeM0*expl|_#N|*_2R#eyBf@9B_FfyU(IChon
z42bnm3nIV4LQF}-BPj+>r{x(zQG37)8gSN=nh<(hpkRjxOV|#o8l#7SU=x0*?Klv_
zlM%%!C_++5LL^|R0=9kP4H1uujgX2K!exPXi<l!tPrxAlrz$u!n&;>!?#Kz6LpI<F
zyOpJ)cHY!5)7BaYvjjY*0z%gEriMtS28dS3r?sfebLMP>aZm@uJZtJ<gh&`Fpy_!y
z8#(ZxWq5|gdi20XUQog0lzfmQXHGy8=PcM|DohLPdcy(ptPpVFdlaOh7EYt3lB7|m
zhmEN0*cBA4&+9fXxkDF7G<AK^uG>!unIrN<5i~@jP>C1}uB%(pE3T_s(z&jZx+#14
z?|ZVB6rFo@7ke51B>FHjXy}J5P==0y@He+=Nl!z!WiipVzFybKl;M%=N3f_cD^EOR
zhgMX+IAVoiBRD)hUS;VbEXVk17|6QroJF5DlX&(X-kC<Mo$;e<&rDNE?(=PuL*RP|
z5qs|LrNWePKL;SZT#-~IeYLHmmE+%WKWd14yBXh=O&T=b)?TLaN2hO38cr~dfo^BK
zzSsp)tDDmvIU!{bWvanb(ZpMX`DQBxx(McX<fnrBV`@N_!Tm0-!QipP?hE0n3?<(G
zlUcmvEs;xbcOt3(olrtjU6|NR^K<PSTQCrOwB#vPcEFv;J~t;;MiW-HCX{LKB>on;
zCRGm(0tTWK7c|$vkYl=*PA-Ao=5*X0(CEfjJ}OHX&hQ3i<_OqESnt4aXF3K}FB12q
zZW>r4!4wB8BUpv9;0fVP44wC5fc0ol4?N~9aN7+pg82vm9@5B<*k<ArVAy2qSa4~u
z`r=^qkciRa!K!ZPjSx#HVd6-j8u$=cdm&$@T*FXFb>pEJD&xPqGVI@Y^S}6^Mz3_~
zM|#pF>a1hkbm?c0p?z)mprkUGxXwZ;SkpthXl@&zDWFzZ;xgDG8F_e`nC4P*wK}%&
z30<v9j;bhR&h-`?(}3s8TvU3FB{RJNr;9s#9!i0Vk-DimrEe5py9}x}8SA-Cirw1e
zJLk}e>kSK@D;!5V4B(_zU2?E_BQ2!lr<#NM6I+N8&J^Y(XQJ|qX*_a-4ZQek<|zJ7
znl<1{q4T;ErKTSZ(`Vfk+Q2+>ugsR2jT8$|h&l|3m<yHBfY6`<@Auk>sB=t9D%`^=
z@}jWd&C0hJTZPx3ojLj79>J09<p2K)K<U)so^@((r%utLJk5Gieo3-?Iu!SZwD+RX
z@rai42PlvM1K#nlWzj6{8DOW0=djHm&FLG-`*@W2-(5Q`VX$Q}1e$ASktJsN2{$V(
z{9i9My9C${49nDw&PE%GH&RqcY!Xsy=FktUF8E6wk5t(6GA%^D(-E}uyF1#s)t!wu
zhgg*@RA@27Z(Uy{ZF!Ds%R4$sslBUHThyBLpfEPV0aF>OCp|9<&9cvg!2F7&5K5Z|
zn9V7hDDbr8DM%le3}%J&JCY(#NCU|bAmGTfg7UItEGv{RONv~f9HHn{6}zlxPuk9S
zK%^CvZ%W3pLirm>v9?guY)s6)Qd_BNKTWmnu04C1%G2hZ`shz_vQCT6EDzfO(P3d~
zF^F1?{YdvNz2!JXIJFT9WXk35)=VYwS8eQNVntp17D=%hiSO<xr7Nh6HgsP{saIll
zdgLwcJFvI7S7<d)C;>ktS<0%?4@im}1swNHr9rTS0#TAIWd-pDNs%Lnj9L<2CEhaG
zj2k*C@4bg_OM1Auh{;AM`*?e{YQOi$O0%}#bCO~;?e~t3Qo8o?xbNVhR8L|F1>)Cb
zVOc>uFDY^bap+(Q1WPCozbjeF3gY9EB3BUiq|E_KC=g$hEM*1pvZTlr#G$ElOHe|A
z_?BcTD~NAOinRrim)*!{9(VQ$mtZw8w{?`#wd-f(?H_fNdL;w=)#2^nHO$+?*D!Av
zuM%&0b*p&C3EF{xZ5FgCZ{NYG`-Lyq`9PVUACsuFnx7>}v6_B4-%(1}eu*J@vN$0C
z@q_}=l!avl@h(Y`D~SCEQUd`?C=kCPS;`9H*Ca*0Aoiz0P(p$Dq+}^8h)+n0TtQ6j
zFYcG>#1aa`*Ck6?L3~A0tSyM#Y{x*dnOBHJrj=CwUSiBDmG4Q4)kx*r9i?2xQXH5p
z-Xn1Iv4jF~(<b3mWCgKHQsfC@aCCGE?)&H{1A@&ER=)yKkSuW!2;20-@_TMI75>Fc
zI_>|ngeu}dTM@19OMAd*KfOIVeNVD3A?l|M2+Q*&MAcP)S-5#rG9qYNS;-b|?pa6n
z?=K#dI>Hi4Hs>WvS#{)`q{tP-)KqFnWC;brlPqNg;Yf<L1<_RpL<n;&!V!af(e$A7
zdsK#!%X1QCR=K=GQmjTU|Fol&?pADLs%_cd3c~ZUsH_lvMN;Gmp>N_~ai8pDETKSr
zT(XoE#P3Lod_n9lPAU+TP#|8GEM*1pWl51Mh<%4rLn2Ek5Z{z6Wd-p!l45N^<Yp;`
zxTV06CDF2}93`2Z{k=L1+a$$mB(u4rl*`zPeZ@lp$VU2GLD(aU%G$VhNs2rn^zENa
z`Qt31Kpc@QWd-qCNs%jv$-OCC!4e9@<C3MUARdzxxq_HTjXZrUp+Gz-S;`8+l@x0W
zA}?E!F(CNPHGvs-l+v~1XXNddI!e6~7yIh)_Tn|n+o!K#-oAE~c+0C>#jm98&U6F`
zdq;)eu8kbSu^+jP*zFZ%lD;P;k<}!9TT-mXB>iniDP8+&j2ZpGsp7unXY{)UT0M>0
z6kIH^+XAsq5Q+hwClvl$+DKdb_6jrk`;wWwsJDG4FY?)Iqtn>z*HtmtOn$p$Bu}yS
zVe&pDWy2CmHisokS#{%}q{tV<Jt_ZuKP421M<q*HL7bKpxq_HFkZK8*P#|iOrK}*T
zl45N^bXDZF3`gABVd3Z#2_y8e5NYDNzN4h~(^3{$rT46)SdH}F)=^4#8@J)0jH%O~
z3cxSRg0cem1xb-7fWE1HDG#0{6o`*Wma>9)NmAqqf?~&eixbO_G+&e~aS&}I4G~If
z$`T6ySKCN4jWn-KMjBCX`$!}5dl_jyFB!>OtcQwI%Zv5<lBMOv3PL4TN+_{z7;GK;
zdlIXtw_U6vzZbE7M`Bo8v0kCY?M}9nHo+`+Dq&@^xQ`A)G^aPw9@rVa2lnC1Y@Lm_
ziPI24fxdOwx#9gdl`uLzndtel=lsGM&o}ANJdV5L$d41?qk%Y7OQ-3znVFNX*HVXd
z==2Ych_bU(ps;6g02i;dp1Se9aukGg^t}D#U}#n>*M2fG<r$nGOyx?y5BH+q@FYj4
z`3K&1*OV-4CpvZ?*G_e0ja=!8lx?ID$vKy4$*$-H*Gs)Ff!f<Uw5|93vO}z|)4@wI
z7Nl}ihk3u$z^s18`y_?X+15cM+al`Ebd=KF+Z)hbJ@o2J0*p*6C@;z)vqE`6QsfC`
zR|Ow|V#ikHDOZ7Nxkn1%A4qny0{pC`$P?gp3}cRov%6&*A84-1HtWb&(Eh7rG%K`k
zNQyk6?GVuF$w7C~a-9zpxWAXIW(D^>Ns%i!1`5Zi@o^xNj_pfu9S;<^n}&MuC|T@g
zm!!xO+)eZq1Nx9M`$kfG^R4hy!7NCov$o-=q*z;+EBl<}SR|*knwL?a_E24NN^2Q^
zWEXs25gg=e03%m07-d@nJugL;RmkTg#cG87NJlB%2cy2pdr~u3mQb4F$-=UNa3n>p
zAoiyh5LrTjc)MgND~PvBihMz&w^kw|n+^p1E6Gw;5I-*|as@Ftk&4K&gldW3mMmoj
z@tcw&PY@d>4~XG{{!{?|NV1a^z~>}Io&ac9fI8905(>m$NS3mK_%lhdwjeT@|5F{6
z_inSd$`7+^lY8afm9|_Zp~2x^{eVr9Vl@)l&{4|eZ1+SeJciDe4v%TFu&m8|o21AS
z#D)pg+3=@og+r2^tN^AYMV<iquwg|7DOo~+I3rog3Svf5<O%{CR^%E4?OsU_K9*!D
zD+p6k<OyQKq};zwf2vk^vt%bLfM+B{o&fqLCQ}R6ETKUBtYj%Gi1$j0wFQxz-QK3`
z_Pz;axZAUpj6Nz6XO+=!Ns86T=-+gdav94#)n#Gh3$n1RApTHN<O?D_b(o@rYKi|Q
zS;`9H&m~2!APyW#ooHYQ1!Dc?ULApd_`!g{DV5anBPA4wTY3huTT<lR63S^sOQg@o
z?UO8JRi~n)$Q1;-f%IqkSVFbL3CU7c5XU6N+JeZ<f)Ac<J5#GNlw4{OWmdUVCB<sw
zVsw;p8LK^!J`TVV3dB#y!m@(+aY>Och<&LU`UE8ui1$gBvV!;-Ns%iExY??kv4jHg
zf@CQxh!0DOJVC54h$sX8QtW?LGLx14{~#&yWWR4R{dEqOP$0e`S;`9HPb5XIAaJ}&
zP8#UoRk}BPPqLI1#J45I+JeZ<UJu`Y^7QE$d<Efv43ng@R0H3-rB~PJW=XLc+1%7o
zN_U&RVN#w~r9TybNm)?V#@#C^@&wQ~agREs#1jg{v}7qOh~ttXUl8f}4o1#&q^crW
z$_nC*k|IYCwa-!K=HEVSG1yN_ZnE;fC@FH}Kkl1Me?@~OlqS7fvXm9XJ0(S~Ai4#r
zJ|tPn3gTBK#oB_%&0ueS$ZOE&!zT`@_()5hYU1CQShGs!(~@E}()qoPQo5V$KJ0;4
zM~7KLf%sEdSXL1KSyJQ*;y`*z&Jqg5_a#eNL3~G2<O<^8-qhH`5(>l(TYL2Yc1Vg`
zLE!VA>J$)5C=joaEM;wp5lN9Rh<j2idIu<>Kpd4UWd-p%Ns%vz^cRp0w0;5U3CU7c
z5EmrH+JeZ@cAq&nS)6M5=GbO@giPkCCjK#rHLG-%B*ki^Gv86lDr|RC7M2ynyCg-f
zAke?$hXMLnLbb$iNS3mK_%%t9D~N*=shBWJC=j2NEM*1p2}zMFhy&?!MJ%B}d|k4X
z6~tF0MV=rw9F$>g`cnb;TggsV0RKZ$<O!e;BfnZ+U<n0c%eG$KfFVh-wjlDd+ZjI}
zFuEo%@9rq2s{zi)+kG9SUdhJytHaxwYnZpnHO$-mHO$+kYnZp6ze>F2Wsc&mhoO^z
zRo8cvG5?s9Mb;7NB}uUwWBy`CDP64`zR{TabQ4b~5MPvqWd-qhNs%jv1L-dYv4jHg
zSCXZyApTNP<O%}rfSO&igaWZ~d#}F1I!Tc$2yDkmMVu(1K-?->%Gwe)ONv}UK!enp
zC`%|1lai&ZAofa%d_kl?hH-!r3dFQzDJzKMl45N^T%m36PL>?a?mW1%GMPA<J&-z@
zef%=#WoifTg5G{=COMA%V8?N6RXI*WCf^>y{iQh2eJiek!oBi5JGl}Q*=@Jlii=#c
z-6^|Ee2bkflt0&Uwd`__ly1CLwsaQVpc`gipFn*}LWxfS$a~c;d;mv<I$5EMKy@5J
zPmYT7OF?n&Jh={%9XL-PNyqK9t+#^)S{cXEeRU~(ogG`p8AESD!>P~+*TX9k-1Ll_
zpbVT<=QotnH|il5aeKS%R-6b|y=x6Gu<1&5de1ehxTY3=n~OT`bvJ`Cy7RW)2=PY+
zw5)PU&<)%C=26oz>$sJU?vU4U4K;3}r;0rnS6H(f-RYuSi`_{jz!WtN7s&G)?&Uqa
zbcbrfA$^~U-EqsV<4sGa?ir?odwBJ-?bsn?;fXW+IN|T$K409M8abg2&T$WJFvsuc
z#9MTZ55l4QK<oG@fF|#er;D~dz+NEqksFw%%PEsP?y1QtmTk`CgIOeelTx^Ump%@_
zxFK?O<2GFZYcRP$vYy}Yd|V5v&FjAHMM0`nB8Z&cOo>y8&WAMTb(dZ4OH~jE4PwQO
z#0JEc(y5e?a1F7yfO|(HC#77hHNZXYlr#a7T|kSL=+F{0Z-ov6y45k<6-?l8C#|Xb
z)B?Pj!4tdS6iuu1`-^##2r-DpSVmntWixvjWu+^TN#+gk1#Z}VxR&WOxR@QUGRGPV
zQZgO(6^3+&A+dtaAVr@i{a|-AlLRzhcX13CEV7$(D^POk3CRwc3m?4RU$0(&Wp_sl
zC4ojHEq1-fUhlEY5cbw%)tLS6p2lqY!pZqAV|M)Fc^Ez$pN9#&x*Gp**)|5(qDG%c
z6oVpv4OS}9X;{!<uCyl}4lv~55_reFIAVoiBRD)hUbRCjDi<*xj-PIXc3pSQ;*M@}
z{N&6r6gpmZyz)3*_C9WSm0&#ZBEMpemoSjv$oY81GAnbXvWe@Y&5}NgYhOza#*`8t
z9E&xoZx6R%xTqk$Z&mE4+~i+q6Km}&>(#mEcv#^S+Yi%Os&$@kE2cF<nD!rQe~AiW
zJ2#;<$=iJybwVQ^Pc%Y+rVl#B>x*5)p&VM_!KPhS(lSxza-Lzl|Mn+RgLvYjYG;|V
zb#X+r{fTp>fMx9~Hr#6R-8x)03>Cr+s_K@|_?8t5<E^ve#n2lg-)_dcm#?rHZ`7Tz
zS^EsyFdoLO>4uF<&&{CuOfz0T+N}LY<Od|h-=jtzoqi&b;wO?ekIE~zHptl5h6K%8
zNNGYzsrD)GAWrFDv3P1<Kq<>ehgH<9eGyM<e@Z`J!p}3!+8@*3e}$j4R4yh``A^8V
zUSn;{A{9Zi7E+naODe3QL@M7UNu;H+en+cRN{Ljy)q_+7&00uhe_m2y6(v&H4k5Hi
z<rQsG(G#f*AwR3G2%5E!%2ZxbVHG7(xwBm=g*K^F5~;kp2dM~}wUEkzyrjY^N~Cgc
zyHt+0NySX0GS!1r1kGAV<zQY?VHG7(d8Az`kGDysnn-232dM}eB^5Qg$GbWs5~-+F
z`ss?nD-k17!K+mHgR>?@Y%r>-SXer*Ef97$$7dC}d9Vs%kV`&AV@-&RcndiNx`SCe
zCb_~(Yt2-CTA0Yo8p#^@swa|uQc7AyzTy+-X!_pJD|7Hv*zCQGuV0y%i8GeqF3%#K
z1RtAxU5ZYG7K^3uxB!I|lCafb!T1`Ms8IhzPpHYOYjmO>ink&PMRBDPT=~je+A(_{
ze1EOtc!;KCb8L$s7#>DY$k4S4A}xp=6<p1!&r)o!u@qXKdmnCC(+W%CEuEqq^d5gn
zWG9l=>J58;11~r1hLFBiFty^a5bU+gB5x^tY4qJ_sGy5G84rr6Vl(b5>w(p@?zHYA
z-}E_Z;t>|O_fe42)>I$dk_e0U6x9#g^V|Aj(gU*e#mz@)8rSe_L`a3;Tk@1n{_CD}
zl48(5p6KL<63TQ8IzEO8p@j)5B4TC(u{(pz6n^;!sCevhq>LDD5@bZ|jHHpoU=y2`
zOfWGiTC3s|-jDT!m-^say5T+bs5y(B><BPD@#up`9+-LX3g+RQuvj-z6YC*aZfaKe
znKs^AC&~T19MRq)si1$MjeO<l5}zC&&SpbW5e11PJ==Y2B0s@>=WJIz5u<_XD=W=*
zMVU$m6-~UE1u+usSKX^EyRCO6TAihL4IK5{5Q`5&(lV_?{dbu~O*2#KRc}wE^gk0y
zNNFe@#IV%x5a>@C`DC}$x_BV;Dxq(tXXcmLNoDQEWZpH*#b!JJ{(T#6z{TdNcryl)
zYB0_p{ZvzrVTsr>4F#6ymD=u|SbmEK>-wT$HbSd8y*?hO=!h<1s6Q3=RS{EaUW)rI
z+b~QQe{D?)kQsw+R4<#b2hC|r?n667!L(X~cZYp_L2*MybP%;Mag0u1f)U+>WcxHW
z-kzW@Ecghnn?C#*tm~zsBDIZF%I=yFs3<Th*ZS~8_$^#ywV{qIYo#a#%De$@*KWiQ
zNpG82wNe7C-OQfV4&ooZ_DcLtZ>Zghe@@k2h2KlH+t}}qu-{kXcY0k-V~N}Gdy(~?
zln~yDDo634T?n<ikeLzpJJ%fdR{BP^-mp7ipCYza5Bt4<ePore-@E3pH|u45pSu&Z
zxi+AG4xlfr67<iduW&&}P0W=tg{iDIGnkm6EM~C(F`K^TDXgGKDkJz5V9>llnDv#%
z8({>0F)@M*31y3G@$zvjw&G*zI5}qCuEEcsRbO+l0%sUrhKH!W3sYIa@+FoqqXtYT
zIrg-+sRwKk4&Fcfd|H+>+3K_BV`-I^kL98?oM0GS$Ao)(^5l$K^iGy5$28N`%lLY5
zl~y)=wov?JWLM^})LD*f2W(iT?Q7lwd}u8FMKl{>J`Z1fW$W1VPCdO8%%kaWcK+W5
zsC44lo^*oTuV3n-6BjfD)K4$dte6eDUZ~aZ;e0HU1Xw!8+B|ECdo@gvWP_8c$R?M2
z92Pu(j%Gkuo2I3I&7)-$`zhwH^VMnkqCZ=S_HfHUQG48r*sK#;LB&HExjKk-_;9I#
zX=mqB^;p6QpMko#0w7?OBuX=;;oJVG$a4$28(^`W-sAct*jNmcT?X`)dWM}4RlZX~
zEjM46RR&opP5tLRYw8EOY3g0j)Yh9JY_^XknRV+fv<$mIXfuIf*1Zb$ao81Rg#rt^
zInbO9gjTRDSJxKE0%AeiBc;YViAHZQvj^Lg@_kuh<|3RU+k{<Y=8Anko2KxzJtMuz
zC781vzPBOi6LoBd373jwXtB!$odgEgU82y!f`#2f0aO=J2&gBlFLpdIdx>Rs-3Q-x
zgCYckrRL!ocB0^ybqNBJ#yU&98Q=vK@EUJV29yieC6-xYN^3vRv(|pFi`K@!c-}Ih
z55Bep3r9+(Bdk`nUF>V3txIT0re7hu>jaC|v}K8psYF1!S;iiyS=dAtfMdICgdP{Q
zlV*r*Rb*{pGON5IX*+do-Y#PU4GcI761X11PP9pkG_6Bh;$ObnoU#oitK?O+(+fSf
z)AL>0>B9SAU7oo8i3jUt-Jeq-2)kY_2G~dD7(vnY#>=`<HOCD*z)mrv!1n_b?wpuB
zIC1D6LoWvN)gQ$~(7P{rV~;8*$thN=<Iz2jJ$$P1lr?{3&oHVC97J6;R<e41*?P~?
zn*}lmg|ZWw^SGSA40(mDkh`$<oT`#uEV2h|D?>QW1CHa&*cm0;#a=6O{Hd_8Z=n|5
zg9{y&3Jcza@~QCbo2F*;`QvAgg!as0een^-=6Uldd3ss+<gXSUW{moF)e5VaWxJfy
z1?;2L!!}lfeMNn7-y-94V(-M{_})X~dk>5^&K!B<+?f-_S|eFZA88BN@aAInu4M(e
z_9@a1`fA5Yb5Ooid`tvrmX&wrc1{*MsM2J)Fkb(o&Ds}`POmI~C{e&qCaT*$8%@u@
zc>w7)!QZm#la3LIvOg}%rX3N(so0lzA|?~9qTHNP!)4_8U#Qs2%kzlfcUgI6(ceUr
zRG(OR{UZJS*Z5gZmVZKwU1pV@%kmMy+nV)@E0sSBZexns+5%hBJ#3jK_&y=Md=oq%
zU@qGP8M|iE3RGoFEA;PLz7@WkxfMnQdzbI>-NmcIl`DlfMVC->VMP=7&-1yrn$oMB
zwi$ggr_M5?A3SQ)E-*ExR~c&L(jdg%X7wLV499H=B~<5Kh;oVu+=VAGc)hNUQR!)J
z;>U(H4Y4H)4#s`j(`POFDa_IyuyvZ^8``6ht(}2mL4P0DUF@xXfF?VXoYv1;k*ndS
z&e9LtaH*@>Kfyj`{=#|F)-C`aPo1Mb*u!V2N*0u)8u`F@SUcis4=y6m8`xzy+UGFU
zn<GaCY#qaX<z%}!)qBFl8UVT2G{rlq`<V3~qo^&Dr8Yxz$beO`ge}fg0}T=^qY)vT
z5Rij~+=uM&cvNQjHN+}hY^#^Ksh9<O!SWUatwC?SA@ZH4g>tg_DsG;mZGW`FN8150
z{3ja#pFsi7i1oPZ!vBE?pLZvO`if+dGOjfCm+f+h;QcadjUPc#s(Ymj#UR3Cyh|%n
zk<M!~7D-plVl0LqK#z5FyN=j<a$qx^r3}W2o(;yoOAN*_X)x4U|AmLJYrbW<A1tt`
zErN9OwCP^gu%2k@b;Ri~HKqPZ!McK+X<}&_6BW#3sV6(ExAHJgvd2|Bfnz1Jh$NCF
z#OaL?%On~~z|@0ny7g(rkM(vcn<hmx6UMR%b{U{ViUHA?4GKbns;C{uF$s1ZBn6%@
z6UK^;k5d~gOoSj+GGC(iF`I7SfPkHxGojKbn=hquV*&$1Suia+gNA1;Y=z2sU<liU
ziCy^Cm{iM+4xEz~u^ArTqaYc=lr+sUdEiv9RB)iDvP36ucqo)^o<z8*(Qt5S8fR^c
zCV#kMjHI(fMvr)+_YoeTLpjN0k4pekycB|>_IegNty5?YTQu|qk|2s;!VH3l;tbgJ
z?AziXjbrm#36~xOUd2Xm!BE&zg$q^^QBaX2DIP5I${}_=bG8{eXfDcfL@138NWo%i
zx-oCVl=D5%S`^~Y&r=k(;FDLr$Ar%5pkah2VNL>;Nti%{gCmA&N{k%TjtHG-hV7db
zOqH?Dmi!eoOwIuUl2Bb2`)8PF>6DI%*q<AMKR}=jxB-<($Be*AyL?$72XUUXpoJ_}
zRuvI=zA$~gOZ{+lDHe-J>HYOiwbq@AN%%^;l^Xlb_cZp=a>pmSjeQ<2COyGM5q7U|
z1>0uGKp<vjAx_~?tO8~VaSpP_+*|}=eE=pi8J75HM`jmWmynHx1B{(zW@WM3P!S^u
znNPB5+*)Bc5ohvHA2)`H{o~e9jU0F{vDp|#+6E3Pw0-0doDxedBT~XL?d?lfg%KFX
zQ^wf%8sO|DE%MMk%jw~SNg(qWA~^teIE^{i>nH)lOJtZKN~)MixWJEp(pZHBV~ma_
zR*Lu#yzICE`C-x!Vu3M_A%nFMo3bp=av~8qMm%|4^=Qn$u6poLR^zQO!<t$7lqEHL
zwxqw?VM$Z-qG1Y2CbJhcGiuFkn;Rvwk}PwhfwNJAynHb+l4*tKO+DeE9b%tO@I1ZD
ztSG*CVwSDZxF*aEIh%BR8OP}P=7%6E*D?FRfkDI=5Did~`Xo94>$Kcq!9AF;1mb#m
zI`4N^wrWyw+qey9Yt*ox^CRRWl1fa?Hu9DCknk|+T_Oh39auUz_t`|Qf>X9XaX=tf
ztcd3#0<E-Zr4XJdS+$s=i3fs2N0j=dvf6YnY5Od?hfhuzn7<@DPG~fdTY#5)r0~6z
zS=3^?&gT-zy{I&D2%8CUXw?i$bgH#u+u{f|H4f2}5*Baj9go#CKi>W{Ozkso#2Cu9
zCBBiX$hvq7LV5;<+R`GPrwWL-A4bOEc7ma}-*#!XpL(_SGV#kliasbRNU7}Cy9#0?
z?qgNQ8-vimMoRn^h!r!v{z1DKze>gIuKh9o>9$l7_k$Woahq&cVLZfAbUwFP`wNi9
z1IVELBRC#O)x?|ei$zyV#_bThC+RV!&~EctMjBE)BW%VB7qRjsVukNYr~PH0r{rZ;
z9jJ}TAN|9mjsGAUP3`yTJw^6uNmlzD{uD9aWTUG5c1gUIH+O&lX~p7=UHfv1Al}NG
zx8;SerCz9QzX28VM(yWsrJ8gnJzQ}s+DPrsQBu|h-HNiVut9GHY4!%KeTTRbV3uje
zp#)VoZbUo&QPPh6*=R&N3NqD>%7nF>y||xu{U*D0fcX6s0Hl;b*$Gt$op@sQ3R|3v
z0`q>;XOqNJ79NsM+pJ?^hlF)Z72eNR6jFzsR8CtV*rte;t=-tJ$a;n<q!T7TN7S_T
z#5W@N=3?!ULWLzvqOejN(1GQVcvI*hn!|THC-1P$&k;4q`gj|nYzQKTrKGG=@fMN8
ztktP_Ffc<3_5{s%C(HxIp@Jd;KP=YRjQbzcv4|c>vljPblF^95<|%96sdy-vdw4!J
zC=Ve7gnc?#UPofH?|C6u#ik5w17AN5gTh&W!FK2H!f=wrf&x-ovCX<vM$n6%2F==Y
zm<GhVW>E$vYG&%mv+>pd(JgHAE-hkqR);;>hB&O<D8aX2pXxzrxAGkE@tzXJCrfa{
zf&fTRSwbK)w$M&ZA7cr9c%?bLAs$AcO(r+RJIzHKB1D8c3dBpB;!TEGj;bhc8yQ9#
zD|kQ{IYVf#ib;5gKih@CN(kk8G@>-)p(va!94Z7h{(e>K!;<Oyo?oIb;y4I+qgMi3
z1%`Gz7kjFdO{|n=d|TU-wjydvWDK_PaupAP`c<tD+KA&FgaL{dSP&65R$zG8Dz=<L
z;s(+iuoFMN0U<|h_9$W7K82_eW@M7?M}%<}Fd-4!5ZYk%$J=O6cS&W%8*GemTaBpR
zC`o;UU|9y+0EP%THQjcW!+Ha?zRGF@W@LD!#X4!z4h&pgU}EzIwvHu~OiL<ZiB>d2
zrmx#eaL&nb0g!~t0nty;=e7V0?Y`v88qk{LwURik5bvhHPzeIftYflWyXbw-Li_av
zG!+$&lL+hMoz&T&Ic^n-=wOp)*qqsMcf74(HcB+kvN%71f!oo5wu5)-7xB%)%|v!k
z<Ok)7o?k_Gx{2Oo4B*WYyd2u^gytATe&No&`ytP5{%obKc*;@5OD~p!x1_RjSV*wV
zaF*2pYcOtwbqAFV!6c#=8FVCLFun~sSue@XDFKuM9E~*LEbIuD5MWp=$!HOdui>;H
z)O|}($IiXdlM#A#072cr(maVTfY3@J9R!JQ!}b^O!@dhbof1-JA_VnZ)Ib&6z#Usj
zP0Rq!f*{9sXQ2a$UD!p20#lN<*Vtz!P?yd-V;{3eP#K&XG0=mTz=37Lx49JGgd>jn
zT*-u&<dk&7AfwB;-bp2sDGp3$mK(H7@kY~~M+@FuUxM2yg~L=8Ydcu%em36K`YfQS
z+p|$Q9_0219Ui(X!uBQZV=-~;1Pth62<wbm4Re+o|Lxc(X0T`u_9XcR9Ppv|745Iu
z*k4K2r#`H)@zA9F>^A$Gd*^W3J_=*8+&W?0QC0e5cD3K8(H`R(u<TSkVPkwm&0Kp4
zkA+JXua|cz<1bn9XI83snetYsI*NC-V0U$d!q_0$!F=tvP~Ebuei{!VYLHn$&KzeO
z!o~MN1}KyNJyM&&tF?#e=eyWhQ2Sr><6^@>ZH|84MBj>dnto;}S)rd_rJoPb&wr<%
zzows%c8@$sKd*!rQ+oyd9HpPv(a$3NMD+6l{d|~y9;7PoXM7G3KWx`G;ba_LV%%I}
zoLpjDTw)wtVzpmlHD6-2USc&~VzpgjHC<x0Tw*m`imY3xOaf>XIl7M!M23zIeRFhM
zF;a7MbkLlmqvNO?9UYwG=;(YDM@KtNI67KN=ID+|bTo-&=xXG~Me$bImCz=v5w?o8
zX0mNOhlUgJ9`7*e9R41x3hYw7>^ZV`vR7|x%am^qwUjs{YGp*bTFQrKfSwFHe^Yx|
z$rGGK!(hjHDvj!!#r9~dzd@;vQal#<??EHJt@$VnH?}up2!Q$qv?50aahDwPTCz9G
mY-T(*CnpYI-l{}3Y|yZrj9}z&#>3161x^Hq@0lhxZ2Vuq-qa@m

literal 0
HcmV?d00001

diff --git a/.nojekyll b/.nojekyll
new file mode 100644
index 000000000..e69de29bb
diff --git a/_images/IT_sample1.png b/_images/IT_sample1.png
new file mode 100644
index 0000000000000000000000000000000000000000..705705509999310e1323b0cb6742ff04c01effdc
GIT binary patch
literal 523440
zcmeFZWmp{Fwmyiv27+54xHJ;ngC$6CcMrjWI}HH>0fGb%4haOe1gAT=LvVL@r?G}9
ze&^hK&olGofAeKN%=A;$wY#cz)vmRdyz5;BpVZzc;9^l=As`^&zJ4XEfq;O8gn)n=
zhk*v{`4T&*3hXGclaWz-Eh9sx=IU%?=U|P1@aj`i3c8x6K55XAA1E_4nIJOpPdZ&D
zg2v0Yxk~7i&oBuTBZ6}*6i9U6b4El2(Qg+=e6UBfDvPi%SFfDHt&U*XvuOR;_Ym}~
z{&x7`;Ve6iZ_95h)9<js>oCBL27%bRTrUHW9XXD*FJAV_k8gFAi0le62!THb=@mlv
zhIN$w%1SFj66>v(4;Wo)?XqCPyBrLDFqI1f^=u*{gym;+3~yTJ`BP1hid8*FN3asy
z6*S6VpxN?Fw0rVXD)H#ETi<j1&u*hAmcM0{wNNOnDG*S2Q`lc41`SBTqm4|kzJ~9I
zRez2E5$AHq&oJ=uMa7;HCJz1~8&{wi9#+2LQtexEW$JZvZD4iHGU^W{@DH6?#W@^(
za1*_+v$A<rxXCS@l);7>AyvlKXbT&y`LNERh=rSO@eE8K_(AhK;)v!zzYVi8Pxc<s
zbJ}!53bD@Kmy$7Js(8%rB2i}`@tPdHk#WOG9Og9zsqAT4x?gEHVTD*ir*}R@>Q$N<
z?9rH+tNZ6YIXh&v^PqV7o;}ityOzSOi-ZbszT2h?EaB5C;Uaa!(_e0dz#+Xc;hMGe
zUni&%kTIT-Q*W1;I)5$~`Fb&vxD^o0c1ymmdou^!)z_ikAAhVqctsOunhx74hYb;j
zAX?%YA&x=s3m=5QGugXHHucNbcC%bT<=@K@-jjUfo&uhi2+>PZy=%wG5-#}9ZYJ!O
z44xny=2cZyX^ug1h7qWy(1V_*U#49m<V0lM_8WR@ygvNqyvk)h;vDoO2T>&l8z-2&
z63O#7W<4Tk85>EOECT7o3~@PFEAEpHgp32-Ug{YOT6(L18e(xP&3gnYGzDosA=IG|
zrDckGM5mB&DYTnbC2pL>5Os_X#{{T(FTcjoSmEfu^dulU4f4-5V?l}xf09f53gva~
z#vg+pWNg7~ITSzeZcs%qGsDZJ7XAeMAf*Uhl*%?oVQPP&hb|Co+p3(3ir%Vz>@S3B
zg0j_$dMsa0RUU@ls&edeMd44G+zMJDM4`jNpdL{uC17fkZ;L~DP99Ik9#8)br$~+>
zPUjnP!>69OXX`|0;kg!MWbDzDObA4uSIkMwbu0wT*tpTTRZ=|%RFpwBWM{Zat?w*3
ztIGB81n|;77)Nt<B%8y`Rn0&a&X%`TM=nBC)V7gsU8u{eE_U_S^|<w!_44z=MuO$|
zf*-oq_)n6(z5L0tqJ~>ZSG|_;SB6*pj^S6aD5xK3T4R<KUeQHhT!-?uNw<-|#5WJ(
zR@7pmATs`BE-T@}>O$lKs!H<CoK~e`EQ(VG=7mLr!Vb46V_d57m9MOCNluPg&T<~N
zg7l9{gIuGPh&JmmlQ2~)?P}CCl`MtiFM?lS3B&P62}>F*#6*##jKN=IW?mKL8|8@Q
z`fGh-GhlZmzK$#*W`CMa>_=?MN|+R#q^F3Tc$(Ok7{h9;OQFHUmd;X>)X>MC#Ga(e
zwxms?o?AR!99m>m?4Tz6!&Pphfz4z>c0xC)!XT$n;o`$P+NZe`8Z%ln9-JPa3;JK#
zAM?{jg?_#;;5pzfixMa;DBjE=$*+|y(>VC~H9F>x$~KiRS^yE81&#-=X^at3GR}dR
zv`Duoz(cxla6!1i^6~QdU5=lJzv8Hv4!RF=ejWT8`-|Iv#K7$*iT%N7;}8B9Ap-4=
zXwLE@(fNEdZ`)@{W(jBef5P5+y^G!6-o4t3@bKI=`a?eYeb0BNU?+JuYlM0{f7fRl
z^Fr}rXt%}e2ajGfj=XSwCQoTa8Ep<|qGIG|l$M{HKSH2K@a+q4hgZB-6ik%K6t(1d
zlsn{MWSoM<g1%<TFKj3ZDVzllx$n&^-aWCe9{M)3-ca<Bwfn{J&&Bh2^3H)Vhj{(6
z8BVth;W4oV|H5lr?CG`<`iZ>J)ka2#9V^zK^L8`hM51HX1#fK*2AXuN*ps!(wOh0+
z_!mT`MV1<8ZYDVL^)YK|2X<302D?U>+HEQZWem0Fq~`qRbmjy-Nj$}x6q^h$iTxyR
zwJ$j@E2ufBQSh3m4uv(SWvHfvYlQ<nTs^-C@eARKPzp1P=n5}ckEX5+;12Y2>5)38
zXmb@BNi>KwY}V6w!21XLI2R6b=FVmcYEWxnLx*mH@3$JK>ifmWXg<+2iCbmeWEl+`
z3>TVwH1V$pFhS;U)1lZS*h{MWT-R|M3PZ1D_d~tSz+YWE(p(?gDA}bvFkHFq82%0F
ztXOoOUK#<XJv1OVNU8l!4r-#J#*5YLCdYHfGa{U(k;RL|DaP=na3JKsbixgmL6WKc
zD2Bz4_cTl(Jed2<^kilCO2JB08#ps0ok*h4b*KhoE@Fd_I)1z-EFPU<Nx?_KG18aJ
zk%aTJQ<QpS8mT7f*|SOJ+UGhXg6vLm^S`{yyv_ZUkR+ZlnVYut7%er-ZC(z$W#|eW
z4f&5<D{Y$U=lr6xY~2Y{#BRWjWwao&B*JG3T>0o?-@x!X<6H4JbDH$JOxTyBPS6HE
z);ucXhhQt%`j_=B)<af9O*Bn$&3=vWPzUp{V-Eu<1CQD8^U8DhjwN{&`RBCIfsTO$
zdTsT_AL2@a_OEs7rZtRMk~v+Ce>#p-K7UmWqU_7cUicy-Lz(}kDD52G&beH{$@;Ln
zlP4;cQ(d3lV6<V}V&~>NvQMCoB&eNZjnm$+9O^%Ky7;v8DL;E|(rflP9rAKEy$R?n
z_#&A98~=`dV1HdNs25x(;OS@|{HAsH1BCVt`~#%Lkx0B~yj5F%qc9NLM%y$wSFu!e
zZ%7E;epjX9qJNTt+l;FVu7(%g6;^%jrEC%qcAWi1UZ?-_dWdA06J1eD@l^4q_M?L%
zn{^^d5@fFS{ag}nc=hhVw}a7x>eJ=ZWNLX)fuk#pcmtQF(WdGu#in-w6XBV6nc_{&
zp8Q&STBN$*Mb7zviY{nD(d4&U2Ye^h!z^1r%S+F@fx8B$>`BOKfGI5MvL8>D>TQKv
zd5P$cfcKi>MxZiUu@^^wKDSGR8&Z@~yrbCp+^O)Ja3`J=`uyH?>+PSBTFU`T-0sh@
zx|umyEbl&iO%}ivEEMQBST$@^May!q_plc~DY>Y)Ro%%h7Ay1g9U~u2`@m;udo~o4
zL6l7?lI;qqa5SH5$95rg`Za8K)JW!ja>@u*A{Ei2wgokrN_cc$^!1hulnhQ~l=FLt
zcxGJ<p41T;zcE&s%B-lHf8Q`NE8cXg9BVs#oh~5=T3DTL^b|e!gk^4v!VZU;*qgec
zu)Dx6_!Pb?ZM1|D9OL%KcGc-xHS?_MJFwtGi~8ns(S^OSx$76FJM9uCOZ}474_Bk6
z$I+*y4pPoJhYbs7njwK9f+hn2BM%yPuI@jMRg)x3Oww9J?o*~rGuvI-;T4P(-HQ(1
zCD)!0qTybS3@Z$(i?$G2xZ1h;*;=sbfGTyir6hhd*ZqeFjr)l^`@Omx=h!aZ?ye3a
zOs~ABQfgrd35eb`v<Qk~2;-D*AGp1$v$F`^p3<6q_zhm|&GLJhf$&yp(A%&>0{u6k
zmWob(gF3bIr@Uqa2WQY!(V|zLDlIA1O?UO2ZN{PVXNF*Kj}`ud@xsGTGqDont3o&_
zTo)~<aS}-vK(CS3darF%R1nyJHU<I;A_W2}&_V=O2}H{OZp$OGARzy<AHcogb_gi{
zKIaXvef-1$>tmmPZIQoxL_i0AJpoql9Hjp`8w)81`M=tzk7puCYs$QS4Qw?nU9GL1
z+}=97-%RGX0y{9@ztVR@Kp<g!ToGSuJU<4mKW(R_=dP!sEMn>G$YpNjY+=pi?fCxj
zItXIkB0$s8+TEPa+tI<vO~hNA{+}~MfcE2V5Ix;Lr?}gT)9b0I(aAWwTGI(|adUCg
zOJLE_(TTZQ*@$S!%Ky7N@JpQjt-Jes5fI4B%Ztm4kIUKB7Q`bgEDYl21@ZE70%vf#
z`8c_odviLuG5qTy|8*T%Yd1?*yZ7#P&Q5fX*EP3r_HY-cr+>WBe}DcpPHS(w|GATs
z+rJ+RctFtO5fBd-H|W2w4RjTI+$*AH=WXqvFKg!r$PBoL1kVfJ7h?Z(_<tPwpIiP%
zSH1t~%F8dn`=4F^<Iw-TtG1i9tBkWFa8q}Q|Cz9VcmB_V|L!OTdVKc(kj1|Q{m))N
z(h^u=p#PpV39N3i2!CKAsqAD`wSX<4W{)4F9^i}RUt6G!f)iLNcM*bsAcgQ+R$9v&
zaX%Bi`MF#}oqg}CSyWF<Y$ey;N-F}b(JvtwEErEv5T!x%(gd<!Sg=`)WMvr@AzrQ6
zKj<(@E0L7qGM?L_GQNs|x2RrEsEv*B*%mYnr3w4p|AN*mjO8}&7M<%%7&_zj4+skr
z5+MH75-k1NER`4$-MW$W4?P{y-z>|RuL#x1rBP(1{$@erL1ab3TkX*NBllPRWu+p6
z`jG!-+`!XvM*w4gPr#h@f$OgxoeoI@MT_RIMXwBuJw)h)*!{1j4v{d9KIb$DBjT?X
z91CFVR2`oRbE5vLe?$;5ViNk_&AF@@0Z|<NGTj?7RO7$V8Q`wpFn);rox0K?g`@&w
zM|W9e-Tq6pp_7x^_&ak$BqZwq#x4{=q$*ANmzQ8gV#EKtIgj{F$3!RhTROuVtMM<@
z1~7G>zqC`7|F=yY=Lh|=+pzeS#LafTbhOgnPC)Pwrn2b4Pp4p1lI^=rfkiSkx!?R#
zMq@4Kqe_k_O0j=CH{&v9L&96`bqk(`P}#RqPZ77W!GW;VDA1*n?0V5(&eu6xa6E&;
zW%@I5R;0^!#q36Zr}CP(!A0_f$GPD&uC)Hgty&4(kr976qMtHmjXMQyN}v{Z(O3Y@
z2N|FYRF|$H?4tgq7fhwVz@Zc<HO=G6i41~2LH1$$OJ6SY=oiA@ud?s{osdGZAN>lL
zf(`BO`v3p;|NpxDrF5SnF8S_p!Zy>LJww6cKh&M3E6u4}^27bYf3No@ET{-BtV`Tq
zS=Q|oq&Gg*HLCJL5_}paYi-Y_I1m^edHPziEoth1r+wT(D2>G!-n&JGjhb(Szvpk?
zuQN-!7&=tRDI9IDEYiI{Y}jgS$CsSX^-r`;qm=Nk7u_qZnHSkB9iv3dh<uq_&Ht3q
zMp9nx`Wx*^pZ)*Nh{>dojI~jkk07`67gMH3PE7{_f-bAFu>0F1Qu%-w;m!A^@Lk;n
z-hPfXf!VhVb=!GSBob5Dq!o?3Y!&Mte>LT98+3vjHj>?7OJFc{a#`Kttg-LTcaN#_
zd=bQ?vR$0;5BQpv666u?_S(a^9k2`B|J?w4h_;LAm(gkcmlp$I#?H_1l)%hc-WC%6
z_pqJ9>_auObf3+%1!ca$^@4P##<s1-0GGZ*R;M|~I**bHNTaU9zNa_Z&28E0!;t7v
z{fuq)LcrBZxJRpj-S?#VOs_?KFBQpRBDI;~q4z9T+`l#wf?w6>T+B=Pn=bwDU}Z4<
zqa>q|-4{Hj8cx=_yHnn64U5AvD8+r<54=EG&G$EZiw4gW>Gvw2klV5B`|+sO{W|W8
zrw_4`_tS>1{x>^?du_)XAn=T_YcHe6RKX>!9>o9lpw50I!?AW|{^!^m!`sXoS!;Iw
zk)knZYnCb+z4lG<78*UywR7{<qU({3^Ezm+4kJQ}5=DaFAvAq)aJnnOSN8uD;rV7u
z4YMFE`W>}6&DXnxbpS1K!FC)z;>EqND3*qUs&1OY=CdI;y=jVfkwFJt4{g{Wci`;0
zTNw1=Ai*(Sj^tzxsiJ;0vL!AWndIZcb@qcxIBlT!fu954Z1$XSiNl22-Z#|X)(lkx
zS@xMS&%FjKT8FWGuj(n4r^gcl*^`!l>F@;4eEj@^|F{0ClF0ulL<^Qy%CRLzJc2$1
z*3G;2Km8!vP_A2t%=EjP7ThelnQA$o?hM&!mmDWXocsJ>=-jeF6L@8=U{IrSeA*8>
zA~ATIoAqX{4M+cEwE_&cuiMJ<oAmeD&JCacqQs^DpHj~Nq%JpG21uRbMDRWz<4V#0
z^%gK{$0w<d(=4n=wY`e5pZ?DP{ka5?dFd$UYpRM*P`h*x9wzBZsCSSlMh0be(Uuqh
zHd&r-^VM!1P;RaG`(=M)K<M8M3w=t&O)s?gfb79RzN^7xC<<s~IW>+62HR!_E{b1V
z@j0!?k-B4ZHA^jll&w{P1dB~cO-|ua4cMNfzOeh!Up-LPblCVbV04cagGF3JEs^%o
zWpV$+5QCqO9iDX4!bB@LnvdIL*iR$N{(%`%x|ly6Mz!y5>i-LMNX7il(M0iI4n}`8
zqqL|la9|7tf1hLXN!MZa!(Cz1)oRqjLEX}!iSG`hpzHdpx{fDG^~Rnvot%|YXP=zV
zu(%r5pM;0irkDoK3<}ISIPFhYISukpxt|S*R*0T-5G_D%5BI?N1D7GRcfT_OE@$rj
zT!NBKeC@oBThXVu99u3XV-7xpA8wbX$~&2}Z^jGfYv-Ky0CdUV)%jGnp1q=Zv;`Q&
zefpyhgX^>>Fspj1ik-;b85~e`wvXg1H0hcKOh65r;zeg%`#BemSK8flC9ivL!FOx4
z3xk47`_R}hnbpq|5WxNlhAb?ZPqJ;ZeC8$ZPhwRih%-5Br!AbyeoaP}k7qyGV%03T
zdhyIy;_g_cZh*J{sxpoXo2%Xoj(u4odG@NkThVCSTJHC~#Z<Pn{iMEanG))%l)m3)
znyvq}E&Fc6=GXps{O)M-HON&b=tj&K_65FOQQy6^^DVmbD9o*|?R$#JE{)$lPlcM&
z%(oZ{$-w5iO7bjrHwRrpZhnA52c*@69ir@8Rx>05!822G+(qW-c(V09*`Uv{x|a#p
zL#nT_I>9%_S0bAfn<42cj~I~%nc8A8(`VP`FmVmM0`XbSavBcsqPD2=D%T}Oe7Ko{
zK8B=D_Ut%}a2nqqDTd)jA6VKRb>)vW(z(h`g5<@xrZL_p3H;z-WqcB^kM6^<TR6$%
z^(S4F3u(4l`q+mq-PBt^Fon!IXGTbYh|C7gb3*W3E1>s{_)J(cfEL<=(DlugWP_>}
zF^QPQZ(g>}J<gU>7r8Bt<eWrx9MMCMYV=C~cZT;D6S_@NT#Y|ASB3*G`Jij62Or+Z
z=3oZXK3sJNru?|N1R}sLpaD%+Uxwt&{t%J%-1p~%Upz2y^PjgDB>z+slg;4>&ZTGA
zyyd(-f<h?;os9Gyd{t4<LLXIG3^wn!@GP8alv@g+Ok3A@f;#YO<<M~HCfV<_mw92%
z8G7)vds-*86*&<>Rb=ORg<eWw6w*8xJ_T+7!}m;0T5Xc`ZCqHkTP|nqW6pr+w8;Jx
zFby9Q)<fvSEwqL4H}|Wjw~rpRo{&L)oGUjjLW5c{631zSdPSycD+n3yJrsW6OdYAj
z(sZnHmnP(r*R)^VPvrt~4X5@@#TN}c{#+^Naz8WnMszT@^oy6CxRlG6;3veF(`}@v
zzwHV$C&h2~Yb2;x2=b&n+Ax`~_A)!bHG=~7#VFU0aYZ$7t6i3gOigO!^t0gNvE2uH
zQlm;p1i$pO|B>I48;cSi0OL31z#H+i=xkZ%)%0fI%DtoNr-{<Nq=_qrP7R&-G@5wo
zi8AC@N8pF1;cg=akaPXYM+uN-SIAy$Jvy{MvFeyhA&~+$V@vC`6ac%fSI-_4ht1!M
zY;}C{jlI%z>aX{Fim2&!1B_=UM(nhwO(fvggpM1aAq}FJ)0P3ocL4Ob-HPA?pAAbw
zE9wB$k#CTw-o_LyS3dS{{ax?ObB0a*(uoX;6+8aXw7<%ZJ=3+4X1AF(FL8TxD0EqT
zJ6OTxO6e%PILLR@V}iRH6voaNl>olO_S!dRu9t~fT%taPJLBq6`EJM*5hLo$zJ$+z
zngevT<t5*d*xhlvNtQWbpRw1%Mmb=Z$x+Wmu4A&w<d@?WXin^^;g@r|XRn+A-UbH?
zsV9X6srpK}{1yd`R~!ILWJ|H%#7zUQH_~HFe=4D(l8{*Nmh-B`pw^t1Y(I?F$rR}M
zpLC+{p206OohtY1dV1ChP{{T~kAEXC<i|?vk|OaADS(ZfcA0-)i7Wa=udHQ(?|mfY
z-z>4v)Yzw>U@35v=9xU)!=OutK&8Z}0w%5|Bzb?e)^dVyQ~8Zro_t+42ZRK;G1YmG
zDPvtf90ua0yG=p3xAQ^X8{aGhHv>Im$Ia9dJpg6E^X_vSiS*L$l`5v-gQKG076&m6
zSo`V9=+P<4I6nKZ(~lNmEvt?{&bisH-PB&*`QqE_J)ZzWCUcJrV%ano4wHUXy4i`x
zh*Ns?%$++Z+si@WQE3|iFjnN$Ir|7IJ?9(+E3`t|@cQVL|K#O#izi~UP75*lshh~~
z;Fnl}@6WQ~$;c;l1nyqb=ICHx8f`B#trFcZNc#f8H)GkW-7TwlSYoyrF7mlgXbn1X
zBFzZ*=9ca~7n<%cLMT~!)d+L-LfR}BbE#^8*fC|y1X;Qs^73yc<F?sbG<KZe3}M9S
zDEm38cnFx|$<cts?S5b$h1|fK)nT(mT*3KeK=EQfx<3>1`A!W%2Ty<GOi};v0ePhp
z+7E)xvp?`GS5D{{`h1rrW_E$cD>Qe1y59p_`S?Cso&yhe#45{%>leo@x%*a&`3LyY
z!?m=`DGxGUV$^cw{Kq7vKB^6>6p_OfL+gG$?`%AQ3rr$WmM4}Aai`%HXZp68+LSMH
zYd+_3ZPZ2E|KrO0t{m$YHunnDhCd!_g1_jnFxOOnc#X&bK~u<#Nx~M&ng1lS!>4UL
z^IE*y>}{I#E<s9lH((5!OiPD&lE(^%?f4RD1A&)2TgsnY(FETus-oH3W?Krt(wFW>
zUH0W*t8QSgz4GP;(Tjbe`74#FgHAd;y;O}Fp@#vbd3=2F)&sAe#Y4)anSkG`BkZ(_
z-!ZO!RRj#y2}f~MzcL<ITeacAyzqSg^YtUX6@9onqsC@@-{ah4Hh$Lg!aYk8)-W8V
z>%LEqXk~Rn-Sf>A2pF1@h)m^ri7)^%D#PQcQFy3&&dsD*6j=ZwCxJHm6LX2EoSf*e
zP@nKt##*H(pBkS1=6O<i%cX%Nr1PBz8J^7)XL3l|j2W)z8dEen_hxJPvz}fk5Gv8^
z-H8wJF#bqlt&KEyk@2h@M<XYZTekADMRi@vB-1LFPu<c)*NqNUYGRsO-+h+5b*2A9
zis;$%30JqYvJJ}#w%`SGvk7P^xowd_U_fDU(YP$aU*+N6BrfVFFR^`wb(U1-iMW@k
z_Z|23zYOKz2^(c>BD~VGy?E<}DJM{GDf+yF8RcVSG}#U3_D!Yc)=?m=Y4&c2`Gq?Z
zlDPbDG(+FEnZCV&Kg9R?E5+FPDycT6*L}JaS^QyH1Bny7H8e?S)uy!o>J$*O#PUf~
zflLCJ;1^#F%T{)#+mJBd-nb1@^NsBQg|SVTFyDp*c!?O%ernLy#K!POz`wn=bg;fV
z8}m0m^mvQ|OF&Gu8MCg8WLUYXYz2ZVf9pK^P1Ytd1{bDMSspd~2se*$mMU8l8@H~^
z*&X)oUSJ!hsrX-QVN_tAdj!sq^xxAMiF|^3T=LG1)(wH$tRjOJp;yor*kV&xy6wQJ
zDW|fSEtb-+^Hm5dmUY4UQ@{A{Lku2Sx_9R6th++f2mx}mhS0^=Gi`;5`)I`uX&O#8
z?B(x_266mslEvPcImo#fOE52`y2^X*mDNpVy{3jOyUg%*-6po;3QW`GJ24F01*}mn
z^}Nh{Wo(TV{L620WatNTU;UfY*L`9y^&BfD`8Jje#9kT+e@QM7%S?M|WuwTO#(K{W
z42#cWC$qSG|7R9^d&{n<Gx*Nn^0Zr(_+I>M6M2R@R;Bnh-LivfM%U}Xb>SCO!bS5k
z7N3UwJU?v-*#~~9;CyTsc-b-LD-ey`$-p1gZQ8=zbP&H+qwwG}6=!wpbZIj|zr7?V
z=MGvE1Dvk@UAR$QKUdSZF<&QAcJ?5WdwbnrUK!LjG6)yuo-2ytu%ccIWF#J2Yd6Ih
z5x;|oa`5WHn33;|11eLc1V`IBKHCB^FmVZ*ZGI=pe|0%e%sA`MaA2ux<P!D}2#)3Z
zhU-z~ykMhG#b$iqj1(=yd-zQ71?Kc51cJvW_ge$O;pE+A9Td*V*ZD!?$m{HT>nO}D
zF!)476K$7?E5MI{sIqfEjl6}dgNN*^wCxbVb%x8;gsy3eQz>@ZXW^y<RS98AJ{#pO
zsxW;#YS)~;b5k^>g(Y1JozDu9fqJIq<M2z&AGDJa?Bds$kKjoH2vjB{@?Njrp!jAY
zkWBivdE|w@401$`1ih_li_j14ANZhYQxe^r7p;#|s>y;q{8bxbkU_2|g$bIMS&2mq
znV#u;CyCL?Cr!vbukai5=wrSeQurLMLQUM{vDi`WfMtO!Urk0zZZz55>e~`C`}>?9
zq)TgJ6&MjnTKD9ETTc;H?1l5!o=QfYTI~;%+ARbmmZ{BRmM2_Q0MN@L@Why@$Flpm
zP-K8hs>H+f@KhqJPx94Fa)rn6pH>X4=vJtf&-1I4-RZr^pbkFNreF?38^&rbNi|X<
zj|uHx&t-ClN`oiGWoY=6f`8ezy^aKcO>r_0@v;S~LR*F|P1OO!{^aGzb)h8O?#f@G
z+i9ZpSsV^R=SG&>C}9{GF}{n<4u6fK2}M>7XEtA=v}H<l`Tn=8sCI6GtukW7sKl|&
z3_sb+L1Y-2mZE7<(o$qSPO`m)dg-}=09=C=$q8VfXNpV8XVuo@4=xwHmKyh1Qq`bo
z6SIwbcX9M)$Y07$e~84*Twt&uJYsP=Bx=G}JcttquZdWZJQLS_>(PO#S|vJ>HA0A%
zHsO1Okgs=|A>WU9w3vJnm@CCffx49$ZfA~|9TSM}p>5TFQ!*T%HY)(Yu(XLMl-pAH
z$&FADm-gBShv~(I@xl=;(!$}`rE2NAJ(!4$jTcB9Zbg}U9}-c}7AAiS`0P>}u4lRS
zv91(KaseECpdo+lCT<L;{$b-@`SBR-U}W~~(gWNSJmhIwQd%9k<eH3aL;lBk&2`?T
z`;jTRwY`dU<eSMFLwlaVS};*w<AF2R)QOjnbA14A52*Fy3#&h_@qE!bof&wu6KKB(
z@H(^E>7C-XNO8A^UeG-ty~lgtn4`5!7W6c?Uw%Kx8zI<E9T=Xk-R5D4T%&??61{**
zYNR$dnYssXM|n%hLSi*t6!zXk%kSGfo`4PYkLob+0$pCnNkOgO)OTI_I^+_G)~XB@
zUXU0%tPGZa+~O7u1jqp2nza>!VzDQ8N*?<gWx+C3nw6A2)gI>~3QY_#Li8w0v?|Za
zohS+VIOjfe4-vn}(BF>&T_4n)U6&~@d??iJCo`0-#`WAkgxoqoRh;T=NL~a45C+!B
z9X_To$2A9*O@_$m_?qkS%cR}(I*++d%|r;7!|kA3LFIO`X`s)WJ*uOJZ3W3<V<mlE
zPmSRTByK6@vRX7QthOilI-aI89)@1)Crj*&#!Z*I>=g}XfdQYK6G*7|pmR3ep!$-E
zyM%QA1<u3ZTbgQtu?9I($~DN(G13KF38?Quiw!f_Z=1)fw}EJ34eU1yhj)VOXTpa!
zyEuLZ%~_S;mIb6wwA{~9I>A*Yi~Qi#*cI=G#o*0;7p4{UV}ZT}b2j0e3t##<)L}T1
zo0-muVV3OB&xY+PO~;pp&P{G_pnqyj;lG!%JZJ6ZCD8pBB!^obn%2FXPI2pI-Hz(5
zb@loV?#nV4B80<lZL@zJnR<>S@sqFgn=bSXLYMu24L4tRr62B_8kx;4g7+&%NDlmt
zO@7_VF+ZjO>8B~8^Wt8MTQs!*s><%<tllQ8yV=t5Fm$N-Bs$gBIN_Pz)F3~k?<f8g
zTkP(DQMa)$`$r0cgyoV6yryIvGBt(i+O<yU<N|Fve+L^+7@pSgJu>i(@w802`#I-(
zJGhW4N5XMV?*YN0^*LS)U-&$=;9zJnd>}CIFD|&}Q1j7{Gq$J}tzgKSosaUo_?G4(
zj@5qZ7hh^$;o?1j*N-INJE~LS&N}Y^d7qI(`78BhE>6F5K5=p2alodzYZ1SkZg);I
zfms}e>+^Y#n)+9zi}_T8rxJ^(&Gj9BO*BW_CS$odpAW0nK9{}xi7F77u0o07i4AJ5
zT)HP~NG{tHoUuwMpaoIJK5`%Z2i^{oHkqtZx9|T1N?fleIGt~g4Cz!f0ba6%$+SB-
z2o^?avfVFkI5^iv52EZC4m|^ijxJ|EqCisWrl+y~dJb`Ia3E7V2>^nUA;w0Jt%Hn;
z*miiTMhT^`tJS^sLz9CTmiXTC>bKVk8LYjNi)JMUK-M%-XkY(!@=rikp;OzS*x5i}
z32V(a8k!<%pDq%UYtF$e?^R6b(;uANu1cK066XSM5B6#Wu|R3s7G5KAq!Y79eInLu
zLerH2xtw~(DQ~gyB(9d<i(bd`1^~S7wZsg7MD}h!I$$a;3VgSsh1u-9GIOx#Qc2_e
zPQrEi+g9>c{~<k)?|LaQ8D6`&j%n3TKUF417~V+|aa}gU3QKez&4P#rOactgyc<2F
zKCf8@`W52eBPwQ_T)zP1pG#Wh%W0B#EBJ?5eupbRH_1YBOfcujG|<0}dL2kG_xwEf
zTH{=6M?TN&v8^mztm+KL(@*uZ0+?UU{UzvWAA1bsyGrA==LZpd5(j&wMLLt8{5{nt
zWb#)w6P9-YZGCk=lxX~qrI$=|Tlh+fhe+OD^D53exFo0Cs*Dt_oLP@CT>f~MRT&l+
zb6{fjN;3N9!)1tntAK5^>7S3p!hLubsB?xhANZEgX38;eW50%-&60HB3KAL7;ip;$
zD$sn(Vuoti&I5AAlL-AJG|ffHc_u8qqa&<3SgpK}t%_CKZdDpt-FLY%F8GsrTriqP
z16JtSqgWgFvpgUdHxDu_+9l-JR_^KZNCX>drhG=9`>6r33wd9QF)~AM4_?pW{Cc?&
z=2q+f6Ezt^Q-aDOA}x=5L0!$WR{PV`;XC!NMAyr;vS7QdKp?xDg2a0D#QU>*j_(Xh
z+&UaQslEWJ7n-*V_RJ2ihpu?8BLz|=P%qY|t1MfZWqj2Ic?GoAMF!`Jb`SW5gxbTC
zxWk&|p<x0kB1Gg-_t=Kd19y~6PaUPzucFH1ht_FNHQs7QWkprQKU~@DTU;|i7I&K?
zn~iiN_L3k917ZFaq7>H<<c(gE*&9z}pAk=KSLpSiv%lZ;o*}ZAfrJHG(fY>pu;;Mm
zg?RC0)Y(dzhQw*$h^{mfQ0ZAat24OgVgRAz_a~1W2MV6aV#ribS^6}vC|Al4$zFbo
zvES_LNxKdtgX7PpHRnrqdJ|tDa4q30F-|2OK+i1}vTdh_juKBI&a?~Rl|rS~r`mUc
zk#6fcT|_%dZ5TeJ4H1xIpzgBSoxh#W_hxn=sqCnD5?*v`z7VO9Wj`TM>L7Ny!gAZB
zNU6iG+F^e9F*>#Q;mN9cnDuc852#CfwJbQTZqsXjJA9PZakDY0LL>H93b~~eDQgc#
zbbmhhsp#h1G7*)p_PNaIn^XO)SFO%O9|6?%?lTrPJ4t3E^s$WfjH`fK2ZnRxzk+T0
zL)?~n_T#PIL|Y{G+`Wn0Apar-wV})G`b(RT7r(rO&A-b?P?d2Y-CqxboxHPPGrU4V
zL^!-e^nA5fgU;3j&d3yy^g4{5-Sqb$iM-F$RLQ>_jw@G$OAy)^?jy82+wnv>ESoG>
z6-7D$%r=K2-wv~g3gWK)wDQrm5+T`3pm+3`DjC@_g)O!t@X`ZsjUiU7dV_Xuqig2_
zY9nN^sJcw8&^k(}z)UdCx*tt>uM}Lp8*N|3Uj3>8(65fzt21Jk?)+lwXqct__h-YY
zaq;@$1($UIKRUI+Tsy?MCNjueBotOj<J$Ag<Ae?(@RrJHDu(F?zaH*}zNW<w@%Gon
zjZdH=MO)s1bjDzPl;(?-z3QjB)G32}OFa&Q9+N!~pWiey&1bQFItBq@1BolVntlBw
zM&9=h=O50jyE9#_l(0i2MfXA-a^*;Rl&?XC{zaYDgLSo>Qv&zREs`SX&gWt?NS2$!
zlXF&?Xp#c+7I26CFHzRmpxT^a${PIag@;XpySm-4?6$EjqyF}DqS^I4hEAQ|;iA~l
zX&^pEXtZNOMa#iM^>BHUya4P+L$&sl5i6fuy+aIPjcvB^5#LNg#nM*0+XZKeO@Fi@
zo1s+|&&;v#R!{j&e@El{H;|)u;OygQ0jP>k#bMy0#zI3%rqc;|!e<_uY+9&J?ZS}h
z0HpGOY350P^q~BpD6ip~TZxUHWAk#P>C|;XB<jk{1*PcZ2_?_e@qP{i#2D7%U%$3A
z3_W{a3+I}%Is<W6sJ43u+!)nvqzGJ?3(ARQcRd7jxy&fhiZ3qP_W3_;b5D1qZs{o`
z?TbKPT5_oa@%sykrAro6V?@kfZl-^p!y(m2k(IZ^{$APXXJm>J6$HK80P*cT_ucC*
z2uhe)R<lXqN@I<&>d^oQW#2o4g$7&)@4e7``Ol{cwVbtxdJhoXT3Z40%ozl!1yT6P
z+J4ROQ03-JzX8Fb{3>3@$RLyZjxRMTivn=>ByYO9S+mNzN^$eH&;h0txZoqZU{fHK
z{H)M2fO2M%WZ^|9B6iJW8qAU4fE$i9o1Z6{{=>mr67pxzyJO||*Gkrqs~1&aZ)Si>
zW5$eGLoBnQUd=Cj9rCXmpV>|#+SNDj9z&CjqL;)V#$F$LMaDDBu9#}eC=^~Ry3u6E
zh{x*M6(Kk#$?FusI-$8I(oKi}e8*Ig1(Y~Fl6!nc65aXI*DrBS9sowY{4AX_u3nYq
znX~5P%@}+cUGi8NV9sc0Jt;X(tfOBr`4?bp-IfiunML;ex>Ihkh_d6|t;&xP>2t$$
zaG>j|;r$+oR?(Qo{L<`|>gt=ugw69RW%afc9}b!JUJ{r$(>Gk7t?R99{EAx!l53Su
z-?~E1LtaCMO@YEh(9^(Ejsf9r-hJRL5-q=R5|~fadcnX-cJ7g5R1~i!cZqmA=D`Qf
zw<jJFKADYQc;7`vr?F05zmOuaE@^LWzY$sr3&!QP(R<B&=rAEShe$h01ZGt#q84Qh
zJ<=>+{tXkHHbWlnBK{LV@jmuN9&WzL_j}HFg8c;w5xS9L(k)WZ5t6Uh_1s-nif!Zv
ztah(xc&sardm9-nRk?FGMkRA`*S+pzuWF*?zzt@Ee@I%dc{oQyYvWwF`!GGIFVG#&
z5(K&?20Usq3a#{O4`UnDoMy~}ZZgGUm8HqdUtcrC{<I*!o#<^~6|hcZ*d@Qe%3{5H
z>T8buHo#;}c=kmX-&wh$SOo1x@SU>ovZ%nPteNLf&|FW}x}()3K<;00)>dI9di@k0
ze?5-f^L$;V>$J_;^G539fHQ9N-GEy#>Pk+=W+rR$xJ%EG?-wv69a=Q?nz@g~QIO-u
zLs9s015hlx@<c9>D_358qwZXzgzZ6ZCnZ#M)iFf;Q}7!#oOGSNR-f7Fr*-t+sflQL
zw49Wt7;FXeg7biHtu(W~_|#SZx>WauCip2y(2uQcO0`XyxnLZbnZk;`X~6>{kTb4P
zfj<fwm1a+W6*x1aaMw@kfVA7Z90-R{p`&qnv!_17{!I?#xJO=ms-fKHees@=B9Npt
z^4$lU^3sggk~nFIK0zYcf#`0iKHbd(r_OQu+fg0xN-bBF8Gufnfcpd<+`l1@WrJ4|
zELLB$HD|B(_)An1z&Y2S-|Y0L^!T1Ll2v@ed%@V+Hmv=xn0I5o5Tk%o_ZaDDT!e}<
zbnV)UH$M>^I`u}d-hiIrQI%BU+^bE|VRsuM^A*tC6rXq7<wMRgk3w~KV-|*VS-WTZ
zV|a3=5E%42MGO%W$?1zPj+PC6HZcq<N?><<rPY5rLR1b-R3ppDm+At$6j;zFw?<wC
z3fZQcp(vp#rJ1q0X0k=R&_uC9Dxy~KetFea6-`eUVqa^~&<Zka&?m)Vdp=e_<iP*+
zXn70WN#qmQG&0Dap@UU|M8%-8#qADfOA9Sj5U6n^-SjcELOo-YU-E9x><;VvoD9qC
zjnf%+tB7R1c+*dEqTK34*H1z%A!1@f$R)ZGN^}>>X{h%|<frNP_6&%~+vu3@)+g8&
zvmjk2IRpFy6WidajZEr}EFT@!eX3bxKN1%9$#W~w>ttIXp0S?3qN)W-D?Wv<KBWx>
z>}yOUT>%6!X}CGv(5(_7fpvu_O?qM(w0CwUns0DBX$(*FGG~9N+j)S-+g$PT_nBrj
z<^7Fl1&ZKQ4<UJ<o)05?V7^*7?@C(~SE{k^fzRUbZFUoX(^gt~4Qs5?Q4&UBn&(%C
z&WSGmX7E%(Ld_gaLn~B)=HUcPy8?FOyl2P&KP-=YgvofDc*ElNFBO=tz0|zhD>R2P
z;yA6$Uy`4kpQ#Sr-*l+V7^^NOs$9*e&)80yaQfX%_%(xvn=f*}umPnv7e~g|TTy$E
zv-vsviMz#H(I(rmH-e4}a00X#=+&TV3-1P6dRe-$$63ezVIh+(KIFY2TI}4tr_5!3
z0yu6^pc}kfWNPZJm?>GLM-X$HU3VmhqaNts)wB)vztXGiuJi02h%~P~qnWc6Oru{H
zSAFQb#e{yEx??i*`TVrmbz_G#!-y0elF4S;X*Aq&PU_*oXl|cGyXvLMiX_0kH!d@t
zbkLZC(PKu)OLXwHEV@rGv2~STUA$Jgu-#zY8%>Th(Mzqv@-($JLh_L3IJo)=k!x*n
zaZl9qsnm08jf~4k&O^H>yD_!Jyt5c11;fq`&-fAw{KQ*_q<T<|G7>NA=et3AN~RLC
zJ}X9!sL}A_^a(avof{k+GNQvgNy8HzVpMS1+zAGvWfh?vWX2}_`X1+0z$fy>vn(6(
zhFXGdj@KWj-gOtB{P58_4VBhfm0EBl;o(EeIz@)F(7(@ojc|i(Q&c7$AB9T7@|w;o
zpIB-ss+pj#``467dYU$~q?$c@&fd!?4YK>WNGDpnnqw0$T1Jt>1ciYwMmEXqyN8AM
z=9724iE8Se&jTW4Ua5}8g=ts5+hjq1CfSa?@rr4G$1NvF3QQny72BMC6y+hRN&p&7
z=+yegJfI$tAAP2fDHHG-QpiPr^d8D`6#5faOigXY=TwV5ymU}hS%=%kY;y9>%aYNV
z<WQ|^IXs){dhlX8mrD9;D@_>1bb|bN+WJ@so=6YO@<;4xfD2Q@Cs5=>Xb2|ijm-E%
zqT4B2g4sA+S>!j}cWgLk&FaRS1=@*2XE}b8KK7|~Fn8|!tNBlqk>rW-d7eKUd<hD*
zd_KJR4t=vIO^ynAzaHyefG=-~0`XCpp)>t+8NbPr_p$jY?FnCYvJbTQ4i6}VIdWfH
zO?vu^y(E9eIq<G%*o(y?KK8VHZbR&8Kbr^G>}6(8D79jXjmJO=XxGU5OF51hgLBTe
zN!<m)I|mq9ADIWg2g(w=23D8kDN7XEW2x(b;~-$axGSVGv^K_7nMM+L=c~E7JGHoM
zdpPTbwe)CAbxken_-1>bM;AzKw%~K_n;^!cG}MU!jtJiv8^<bX!nhlL7J1j~Q@x)#
zA@1f+K2{O$ss_UH@~Nc-WLQtv6I3s)iAlye-}A5!nW%qpG2ESOW>2q^$tu=x<~vj*
z>y72e%H=G;NkyGA5S0&nj-MB&43MG%QD3g#65n&>WAwmQeM!<jo;y!zC#77!`B%K@
z2)<KT`h+Q5BX6L)l$N9q@*9z;e*0M@2Y9otKN+74MOWJd?><o&&F9CvrTlkLp-^$T
zYmC4!%fiSY?e&f0!D{Ol%}aw~gl&J=71lR^vQ`gz%>b{`n1tO^;e^Ws3CaU8OPd~o
z6Mr)^;+t*Y^{8xca-hyUzS6U)<1#f98Om`daCOS={qNEN9&NEJ&!j-IBi&xz=IlOo
z)0v^?X$Pk%BcAm5&kE_f>7mtcV+9gj`ssClK}rl@g}dP%qBOydK+g3}?4?jKP03Gl
z?K2efqk`zvyoV=e6>Yp%n%^+qk{kuhDCh{36hNF`&VXEjz*E#UTN2Il2loBI$|W#f
znGkkFCIiq1V7<^$n8Hg+Bb(@^KVwDvt8uz0&&8SFWNb4(w+o88Cwpb7@$2oMJWbSi
zo5&yxq=R~`K2Gy$VcMl;x!;hs@z-~^8Yt^?*T=_mwHa?Q&_c?tZOQ4P*eAy$6>)=E
z>eUh@uHPqLf9B7wF4sf_K@uVtq>?$&e97L**B6xL{f>HqltAmV(I1BfFT&9K70q)@
z*+x!V6qUN|blEMtJYJ3w!5Er~x%K{&vARA-{)^Aqnd5FQWN53+P-4W8YU1R@WxoMc
zh^b}_HcSocHoIi#1g0v!6^iy9l$GwkL<@(vz~(spPFkS+Vj+3UXz3Lxj?EiaAN^KD
z>m+?cU~BuI3i$(=#O9`4+KLo>pz=d}y0v|_Is@fRYmxi<nwQu|gfztG5r*|$C+{@p
zLu5<{xdS)<(9%q65faGt#-9yj1~|cotzhDQO1XIyOA>e8OBYMM>_{JvzF868bs(8H
zMFxF4E3Fy)0-y4N_0imPAK>adnD`$NYLdX+YR^w?SUXB-7VWXJS42;$-a#VaFRS9Y
z!$izgG}k$*<;kE#Lw?$X1gOGH26j^x;TLqoNqTYy>y<2R&#Bm}A~`sD2;LdW-P989
z;T18L$fT+nJrmcvNY<TRB}c=ma+eb|Uf1YgaPLw7Ace@HGqjSfr){P2q=d1I>OwMM
z)Y=Ht5#k?e!Ky=$h5^zLdnO(udMK1_BNrLOx~l%s^bXG_$+gwWhmYOB{U$BKrjp;(
z^BB9IkpoNm8daE8QeVpCyJS}$&8ghtE?pqX3|9q4Ps?%)KobV)qF50`zmCrH<|Pc)
zFlFDvepX7QjwgfpS(4Z)Z5iwSi|Mvu6$^)FOB}P>{B!ZNX;zIM#%%6NlG%?0g+erP
zTbXRMl{YD`S%@KDg?B~QNAsGanun#Hdf@J#Stt4{-l;W8kKi*N(JOo<(jDcl@Rpgc
zdeZ~<4iBOS_nEY}Au5vU_2ZPX){RTD40U2fjLU^~*52)(xn+J+C>Co)A@1AN6Uz&X
zrrrtR7Z|0oZnBBFdeY6N15dwR7`gzW<rM~2+O=3PEzX^WDbH&;NN<w~%T{O+TYs^K
z<S8=tkQ#?gT)K$B(GP}R5C`}6IFkNig$;D)HeZial<8L`6GaNBE3hY9`FA&$hnSi?
zbF5~UI>yT)kd4fYxe?PidT1CXDY4`aBJ%UtBNy5*E#hnSpNx1uQ9)uM`8Z$d)ZaU9
zR^lmSYK<og`q}1G)CdK;6a#g<(%kZA9N)-s;sYXy5lc1?v@QfTLYjj#7<gqM)YvCE
z4{hJbTz30CG+=^fB}9x2eQMf1Y45+q4L2^p%=Xj#Ke@gv<(9dNd+o6IuEOl~oOB~Y
z(|KNrfyAi`8v|$y&T=Ly_mA7tzV7LUCh&e$G<l;gFC*5N#M}mJ3XnDFGT?;2yH_KS
zeq%us5s2x=!x1NC*W9C>#MhqDiyJS;KTEexZ&f`_NOJ?bzGMw7h&CdE(q^aFwKp2W
zx>qV(O&DqX>)uJ0zJnA$fn6D&Qo(!k1fE={$XT8rs9Oy-X<ZmkG-kh7e@HnPyHpe_
zt~D=OjaS{`g_hI;&Q2E86H#;n($@bH7;IaK@oGU;G8sacoEb%8_|g3+$zR)}RD6OA
z*#jI;Z?@bSUzH5_trIsHV~HsJjx7z87Zf;pJl>{OJkecTch5TVdPqf4GTuv4^*i7W
z>?U~{v27%qu`-4VQ;irLc6dB_fDj{+H{-8F#L5wo$W69;GjFCUrj-%VloBh%*<UDp
zf7*>NH}(jcXXxIFn(r0RF9ZIKf<{M!&s<eMj6I`n&PfzxR`W}4AoJ{A3#HVctGscy
z_~Y>t8|7u*oQ3_08CkHY&-XF^FOB|BlXmhDomX2g{0%Hn**%MP74QD5Du<kl(c=kG
zZT57|eMw4kvOOR8ps>W*{Lo<!vpTB^+kcr~Q*#g7@=)}WCf~!;q-wfuwZ~rCN=xi>
zmL??Fkhz23qF-QvpDABkYRml&iM@IAaie|ju)f(&?3!tkWM-EP`{OnDu_$BIGMG6s
zDBbawdF^_2lx=LYzpeSzhwK#i&Yi>bi}q?F@JV~ymKsXwmx?wop}e1%Q?yiG?*&%I
z^IJSRuL4WXOtev~j$c%KBi<$AFdOOGppFFhROxl)$;=J#T=o8DVJ|XZ<$2);&D>*|
zQAf|uQ88Nmv~nzgrCS;eO#KN4^=Y1?NxLAv__-R0J#<N~yBayI<q3Z6Kz@wP2S^e&
zyD0hdDC1)9j=*Gbe4+=vKrYur+)^C=bcf&7)N>*vte-pYb6-V$kRCy*Bu;QSsnf_{
zc%Z+{)gnyvo1r23zU)aA@|;D@qPJ+GcCEF-q?d@rU4xVaw>-Qu?=JKS1+S&a%5$*<
zNly_M8F<%Zn}6YtL%C~;e(lHigzq))1q}@c0d9EFS9%5c;elQ;)IUFsQJaZUk7bMl
zR~@C0OG>y}k_x%{{8vwGm5fbyr2c%t2wpssk5a;k-%4}B`Ku&mvd|7i852bM&`%<v
z$Js!!X+@*Gs_)zD5L-qr6gjC{53Gxq@q+8@Bxvim>Ma^Wm~Pt1+IxYss(pCLRR$__
z38s~mAMc2^VDyL9<>t5)!XRaqoaUivBE|WWAR(-vbyW`JF2L)wg$gSMd{bu^ct-Ve
zjD_r3KEK=eMDz|pxQ(S*fQp`(3lZ1Pu!BkXVB47KlwGzOlI_dSHw{Voi#DatkBa~C
zEeXcw?2+p?LIcXa`O7}{*IV5w*u$blx7gy9QJE~{0IrH^Qv-Nxn<ZpMmA-8gzXR<O
z`^eRgqr(<aoSXf2?82qn3&D*#zx>zRXNsCLz|j38;(*v<`070i?eX@v=z=dPYwj$D
z@*%VfeMBp_3lj98vF=s_UU1KIJ#_gMN;!-hiRf$MURl<4Wh+F;a9;9q+@RfqelsEL
z(=<h@Kt&Qe!_<qj3T4X4!A7XQSf3G4TskfTl6vB6w$S|D_y$ocnpsVf(#?Lv(>vfV
zB(SZ2|Knk3(X?;OTb{%gJL(mCYYl2%!}<X*bt+r&kALAA=H>8_JhnPvhQC>-z@bsw
zJgG_A`|hu8{pem~-pT!b(^YTyeQey_nJm6aI=Ni5hGtbcxR9Gl=Y0k)$M`}GJU5f|
zXu`NgTG`fcWZbx>3^T(qOvh`mcEY9hKz4zjvi|I7_EK*M?M)^w*~9|A<iKn)A8PjT
z*a7e!{SX@OJE$;^oRn<{mB&;YehuA<1EcemzUWt^biUw*%chx6WoRA(Gn!s=rAERv
zak@?NUuIEs4!neT!bEJ5Pws8`7HgB@Y9{zrk|bg?JC=U>W|oZ?ow=nLd%A38xWzzi
za21~+Q@VT9wrDT-)mF4Hl=|4R$7bTj-Siioy}CD)gF5~h0i4=M*(^Wb!I1-9-DUcn
zm5%5;Xw;ihhBXKKk9mdxCicmb&3an3^;usQq!Ky%O7v-53wm}38WFLEWZP9zxm&k<
z(s4X$BNx-?E1)#@e82-`wg4q}&CV8bm}h?~CeaX6>0VnG{;Jia6Ggdh%O8&BE5k%p
z4HvAHf<>MvS)gP;yPjKvxY1~TS1FBZ%MRZypOeb2nK=Jt4>(9Mz(JOvngzXC>^1Rn
zZqS{*pZ$+A^TEVflQWDHHjpsfXz$$Ayd$`o?&9M02On~9DY>7HJFG_HT)!#r^;;0n
zjkftAxTef`VLpG;^K4MEkF)kF&uM2P`Td@Ab8R~C${*Gor<nilN#R1V<&=jF;U7oH
zMHlYU_IjIsVdfUz{UqVxC3In##K0lCCdS9GV##OEbfL@MX>Wi&=aHCx3UM2aed9(H
zBi0Lt!xbnGkFLVR=kxK!H<eeaq_aNJ0=&vV%hi086U|yeik~+ZW7P42=NXFQ+~`@^
zX6@}R=>iWkHH`^KY<B&d;l|ir{jVw0e)#@A5k*$f*NRh`*_=qfz0Y+StiXS%7(6Yi
z9rLI)&!3NWN^bV*4b;Sc7EW(iyXMfD%2&dMr#~`~M+&hZ3LdI$hB+GF1?g*h_UKbO
z)#ui5Hm(}tZrL<+^_wVAxrFld^W(xwOmgN;fhx-oS__pAB6%0FX9jMMh^E!aDa4Uz
z$6>iPcfK$(TqVd4ua<W>E$mAh%Lq+-kU)7_{=MMdi=jlHui_fsExN1bAp{WW@0^^)
znlG1GWzIpb@~;{Im%mix`XY?oHYzehMd;?7&J@+wFh%Gl*z|UGYX1M?>@B0B{@V6$
zrKORU25FE|KxzmP5hNv)jv<HcRslgkLRv~%N_qf6y1P>tDd`$I2A<9Dx~}`VpXdMP
zf8T4lSiWFpExvJ{d++l&j?X9d^y%o*?@iI!ibJs~Vw?@yo)3p+SMr3yV+M;$CnU#I
zsO{K|CcNG*VMC|kd?$X|ns<&?@g0=kbj&;psZ^x2#FZjAem0WNk#h98ARmUTS&#W?
z`|At)ckUi^xMLnP&0&sS{G$1YJnW*&EvAy;2=DfzD;jzdo*9jsB2DM75JUpp7;c7N
zJJ^j78tfx%EAycKFT?ixZ}lj9LcYS%YcM5Bg@7pvi3u+#dRF8C#vHP6*ZZoi`1DLH
z%HP$-60eljYlQ9bW<#E2mZVU}?CdeWwn^B#bhn-7Ew4TrnExjWK#_eThD&u=YUELp
zhNDbqYUL*ygueiV`~}6+P7Ijr_{_JmMD(!qjUih&Z(;Uzh`5~-w?fmG+Tjp5@9(SZ
z+g71Wp1`^3uldqNxNNVF-VC=S^_BFcgh{N$AkR>hch#~FJRWYwGYThlv3N6J5P&-N
z_YFLmD`pVCKpxpDdkwCMU-ZK7RG*}cA29wH_kj;oIS9>dXeKjPmag@qeQ+qXdEb{7
z<Rr6Xnjq)madS9E@l|w_t`c{hQ1s8cpO*a^#ibrG^H!ujBkFr8$f`skDpI>uW$WP@
z6Q^sXm>SKAxx5$K$|MVYecen0v_p(8!;elUL^gAF9Zn=QM-QBtymI*d*@5SSCF{)T
z8EP+JPq}WYdJ_^{y9>1QQa8+rDNi~a>)0bXdljUZ(|Y?r<-S%KIsHD%r=lf=#24et
z2IEpP=;QWmJuhb8ra6TbD3St|_a|DNDp3R~t<Z!y)9^y-1*_#!6Q`l*iXp8T=GlPA
z2Dd{K>cl_7%p}DD8gb|RTk=1RrRe``EU!Uh$?YYX8KAKCNyE@PTkPWHpj?sHYgjwv
zGRr=g);lvSy!c@=uj*NN)aK1fuQgW&p7IU7Ro!ZWqA%>;^3Rng0t$M+xZ+DtpO(uD
z`d<xGIEo5|sZF1BSR<6q3ww2a3I#qkZu(DGi>E&wU#%3)m`IzGhN;%tzsWYOes?d#
zzn)Bc!u0tp&_|`z9v3|zpm-o(4M}nQuD8?sGUQBZEwF`|aS{)iuz@4aeJ<{l*51Sy
z{!aMMu1?2auh}32x#(&8M>n)Z<xAug1-uDQFtS!KwPU+X_H0*SJwxM$TUD?FQ(SlS
zH}GjLF^7?{IHpawj45|olK2|RnxxEz+6?NcPCai^RAOVDg3JNSSLUep#l3cB#-g}G
zYsEBT@i!R0S;P1B*e4NXIUiTy<l1k343`>P+x^_g?`!nvzxyh~>pCsuS3KM(^>hr@
zw6_zxvtP{z-Gc4Uz(&gPED!rd1of<|oo1eB`p+Gq;<Wca$T#OBYMVArWgC{vw8Gyl
zM=Z`RPi?NJ9Gtd!y`*RvazAXJS5-aHzE%<4NrEqxcdqFSXzSU=w_0kh6cJx@bs1LU
zOxw(<^U#~fJlu<h7El5Oi&K*yzaEarW!or;-e35v2Jc=#e$$&po7lF6P3(kISVgyk
z^^lE|_9wH6UaT;hhF9cKWrBQ3>kenp23D4MQx)7_eOT#XZJiX|P_QhGL-}9-D$|G|
zpDY(*z2h+P2#!yW=z7VheLXu97I!+`X?1JI`L(~sS?v7Sb9NDPNp@P-E{v5APVtbh
zEr^#Bcsx`ca&y0@b6k+2*tJXHTi&dy#&ixTcJK|MY29yG_OL0g*VK4~x>|C(=03%>
zd|~mlJVIYT*o^IIT2}rtth^9w9E4Bg>FU%Ti&NZ9v7=#N94QUu4GG@jj4bYd#3fYq
zL{~PHF5*S(IK^Gf)lZGuEMp2RSKAyT{aWSe!~H`lKd{5!>HgX5n^{6*U}&Y)FO73)
z)eMG){R{w7<RRslh959&mGf-QHLrwi*O=Mzc%g)Qa*bxRfpo+$nLvh`$E?OTJx<G=
z=XMiFgen~o7zsK-`|}M#a;;M0_70simkzZ3zrMj}4>Tqx=7#7_;St)B!|uVtDN-iu
z7iaA0V^j+i%CIy$bGw1W!j1<GE*=@tKf+~z9xiC-<S9ZcUM{SyW%xIxV5QRjL=tv4
zVknUaa^=`OePo4abXEDwiiBVOyj!6`)9dMt{|tijf?n&g1C$H(D{;<j&Xv{Ipg83I
zS8>qWUY;|aHMvgG%%yL+Ph=O!JIRf2!SyuO>(1!iy7wm;bk*y4EDqDMA~$wIgv(o8
zy&Bv)&wp6|HA)sPPH_vU^o*-B6sBL)0}ILmq!y0SL^kiAK6~I{kD%kFsx+B1y*N4I
zo-Ujv)5L>LkR8*Rl)gP(R*3i4Ep5-Xp-JGP-!ER$_|M`K%#k>Cza+e8KOE2N{(HFs
zZ?4JJ+KI;YW;F`lkE63yebPCsGp{N-M>j;fiG1^7?8>Pd%<M)P2pk(wFf|-=$B(i-
zcV3-UYOzuWqFR{#1o<pZ+dypL6of+^T^6J|i{gt`2^bF+EJxYhO&ldpG7d!}B{HID
zhWJTHut>}xDQO=6?0e6#NFB}{*Ot3nl!3~-Qj7g}VI4TudlJQ+j!!P%{GLlMPN&yV
z3bInu_`phDx<YY6Q2R(~Zs7@yqr&+<m=xQ3H)&1M@r1p@f2l#VHYRS=lv~U#Ua<C7
zU^mJKz5IzI*ouDpe%zlzWD#|J_+z_;(xGqH|AKF(gvcN<&FUqT<?n-Z-lkE(nYNqS
zY&C11T%=wN*<FzZXJ2@`pc&pW<yn=Bau2g_sR7tDy-&B!qptKLx_+*AKvmg|HsdZH
zP`>|$6OCdDYX|>GL=&T(R{WT`R~ptbI>->)i`q&0zACPKe`F?)P=Ok6l?p3l<}H_d
z!4iqs%De`~cD(2GrcS9Snb7WvgqDQZxkZs>TRSB1(xg7aGQ)3jjU7uTh0`aqwyj-y
zWYSw6VP>4LqkQ-yO>j2ygY^PJ#lKSba|a*H@7dFU^VLK@1R4F2I<@0-nH+F(>BFj8
z#*g*%sdfYRk`=d|-~2R>!wP~^$SIHkHup(GrMUZZ#gi~`e%rs(KgpA;)5OFjG|4DA
z(JbB+N{mb~t{7`zm^U}_g)mbi<s>>MxK~rlbduk_R6iA>==fIoW@#HQm%$;B75bz)
z`*$`@zjg%=+;~nL$)(pe)|d=cKs?eQBLAF`8>x_d@6h7qiI+BC<5`x@!k&@|kFxI7
zQ9h)^%V}5!`&_tU{_#6*yf1A=>ZrGe5{sX9%$NJ<oH&*>R(8lpt%K;i=yQoWv)(?v
zxNC9FW_G#Rs6PW2J*gpO%`QX=B3`j(#A{`}P<Qai>Tft*@VS)uoycs=p|?ntuO5(=
z&%tnB=4PuQ@UQbG@WLLhc(6ks2REb^MtrPw<yH;a^SGQ=HY9w`6s^!L&ats-5O#qJ
zvFqaH<1;o6#C+auY~eP|t$Vv4Z!}WAVPGZbBzapW==@6PrnkyuzW6ewBfI~&0<17?
z5=vW9+F<s3&gIM95d)8Mde8~$_io%{`9}L<M%o=0&v(fo_xs*{sm**8!++ZtHdlrz
zjeOMVAw@B<BrL{c-F*4LYC;71cDCxE?BiMbBkk`+Z|e+ubeJNbU9*~>1&*{%NmV$}
zK8|yFY*Wii#|;rrhh4*LM?OyHY9MjB<<u7XR3Gv;3qyO!`1n6M&Int$$--hjgxaxd
z;qe8xz#sOhX}3NB^WD9DcUo><Io;M6LeFhoC(_-Ht_H^z!<7S+V`M8m+RFujkxtwN
z&kxzLXW+^B%RU3^`-{bz!E8%pab3@}jiXVn6t7?k(M%<(VmAJ_8c7Yh1%w$VbWPC#
zZ31v&%eiQC__3#4`i5%wJ4bLfrDlPDa8DubPk)2x!qn3TusI9XO1r0mb7J3j4aFh<
z9^(FB4u8iXDM&&r&en(??%4h_F4;Ab;jn154duwA_mAio)AWx2(NE?-8*6ECX#pLj
zc=0W2&wMr!r<M?@)lk^%M&5u5<YuLHO@pgR3%L7DN9BV}9zH@ynnD<&z8O<8oI<ws
zW{!kZGUl(RbMp4OhD5|w!Hb>^Ui5T^8f3+c;BrE5y3b`x!a?l`{N|8UlKtkQD|jeZ
z8{?`gBnqlY-}vZq4s!GM>=*IWm6J^c&I|Tf!hE~f1;!ylX|30%pMZG}Qn~YH^G63H
zdW-ryRB-<T-lI^5^Fik@&mclCP-irA6(&551(Ibv=SH(5UFsnk9oh4H)ra~?XG9cj
zO{c&1&Krgpq1={wQ&8QPo3_$eUZD+Mz!v)0lag(DedvkH5z4a(LL1l(cgwHI(;SBo
zBA$_C{p<CghJWy9T)*93gu1tpcqKK8I~&*c(m~nm&-hX1WRIzn+FPuF)O4UsjlT2l
z)#u*<U9>I{Q3^4^<G-_c174Ew_i2FEjm4y>hkWf1V^#Ua70Xr*fer<NxeQNFtVsEA
zxQ3nnbMM?#(wjn_qDfS@P>zS8VNX7R*{6LtZ&QQDn`{3tL~*SG34M5=Zk^;|zmVV0
z8SV6i3LFZ?dt1PPF~46oe!wAQGPr_sY;F|(%{t(d0#Fl}w4(jxun}wtGyP=uMn5rw
z#`qa!nGJm9i_+vZG8`{kAvXQK`-;$LRdLbR8cpRK5_fQhnk~!Jn4IlvOqb|tr=VB~
z)1wV&`lm`bzdSE4zAqo<eR=E2F4Zf```w$l9!PFDS{cvGNYPq`PuokC5FHlx;jLfM
z_OHTKmYaEdPQToBAJ=+FvL&^@hsN`w-agJgE1ZBriT78>-7HaO9kDHYEsii&zOhb|
z@{C~Ih{d&kN{2dsKE3H8-sC+ypXW`wAN<LuV%U?y%?Zo-{^{qI|G$1-f+QcR+6>9<
z)nVd+M)CYGRjHUkyBCKZ^QNa1{>rh<vT)L)r47sju3vA29D}3kUQf_IMsvYjk4BqS
zH79|j&+$)MAu91T&po9-UZ4#ej%|&polsXoBh$qSlDC3=<B9Z=%THdL;g6%LrUqzH
z*qt+TGm8)7eJ6Q%0*}2x+(AVcI#bE|9uWOH{0UDUxT=)ba~~e*hbVO7KH}hJN)?Pw
z@OV6UKoP4;9%>LHh(B8Z&UaK!`i!ne9hBZDIH?IXY!((hXq7utv#{wZM-ydWz(NPs
zeS64s$K?H_h5<<1-jy&d$7S%K&y|GsU6b2<2&l3kQ!}ys71WtR$)N>U5a3ziO>JI=
zJw!~Oa3%c-uR(0{kXmNvP*D5`lRmBHYRvQy*&2EY&b0N+%MSZ*y_3MTsFp}6+bGR+
zN)&J==f3yr)B^Y}!x^jyPFrIHWx535Wr>1_*SuypZ`xUcL}=eU2{LYBya8U^WgRU|
z3r<R9j4HEo#*9R9pdEQilE5r4%_drQM9rqAj++xpCK|N77{Xv!WG2_3F&Pi0f{T~v
zZrg71%|kzGp2IIJb$Sdh)(`S<-$Ed3&oc60bg4SyOQ)zI#O7W$zpiUjtSyxM?v3-!
z6DfQM<g~q&Kh>+J;8rkc1ENmm>2%D61|Q484=6h=Z&f0)kkBrmR!pRG9`~sD4A*bk
zPmbe3t5G4lt18$bWuA&ocvITCzqm$PIv-n7@%QBnEsMp_^~K67P?BC__yWa9#j&uV
zxSrJrW@Wh&Z$eBblf55ZKAXJsChex|?E6MI1X|`bnxp7uXVfbe-tpv|SMcy!Xyheb
z<f!pNZm7a~v<^5}x1|<~k%=4%sh-#~48<<W+E^}oqQaxJ*7o_gBQSYf#`m22a4yB1
zR`%}kA4E+^vAuf1K;d2$#(*$mes@tc@$5;UtnH7wyuBuJJ17}aOK4TdoU{1xaUwHg
zY)KpqJkk2^dXtIC4^X4d?fF*j@qEtmm4=*+D)HGhzn8_29g@+Gh);b$ANeRQxpvX*
z3Skx{=}Q-}ItH8|-FoQHn2O@F83etLteW0o)-}bO)t89x)@JXMfR&3s+7tQoZ>keE
zAtBnNYD`WqxPdBcjv>TDgFcX%QLu$IQGayYcJqE*OxR`|n6zOu{4;6WK}$GO+lJ27
zSrFD?EgYD(TMQ-qQqi|E5}q+9ZPI<JtMxpg^<w?jSRW-%k)hd@-5>N+@Lul>?DJ&t
zK;BNSVgA)oE)Gr?zljqt$y+KP2^gNQXGg(5Q$aI_RU$wjglaz=<s6J1!%cl^t{>uF
zA8!_sST)dBrFG~&@u`aVFPH@;^y^xh%lK_~`$QZ^3v*Ztb}l4x&rx(;yJwld)-qJ#
z%y6sRd(YEGwL$CBuWosV@o(=_=;xoI6=R@KD-(Vcbe}c}iQm**Z9)w0-IlP=_OH0|
z)_pJ_NVgZXUD^}_MF~-7;s;*r(5k|Q)QX6YqB-599DNmOjDlEZ^Hn8*9mlMEEZM~%
zx}?!4Y*ce0H(Dr$HJVx&s1PB4TTDf1I~mupvuI!VO-91Bpa1CT;=Fn^_R+0!muOMN
zsrWuoBSNQfvxp&q*+^@(`2*dG_-wO9WP`Z-wr(kYQb3-KALSL$7c8s(w@nF?K`*Ul
z_3e?k4Rtv__CmpIfMUm&GO`16Wg>acPRyh+`#j>`c;{iF?LYc&Pb{!Hg+|_UeDyJz
zN}==kJ;Emu%WRx#nwI8aKoZ^R7|4lytZA6t7RpT5`_7~+R561Ytht-mR8n>IP(9L5
zgCDtywUGq3eAB0ji?8i!C*ZCpp>i_6HUcRR{-dFn$~l+>h<C|$1Rt6XNQ(4bajnkw
zer`Rw5q;^V+Z>*Ox2B(uKAhNQYUs&M`$?q87^HTb!}^m1{P+2OhezQz@XNFaleGvq
zefYF}_H<Ak>)Bl*c^U0cA(<ABVeQEHyxGr?KpkJm4P6qp6q9m}L-BT-$Fhto8`pA)
zO!=2ka<G8wXDM&u9K?P}Ql>n^>|ZcV@Oq~8$Uc=pmYxL>?Z8Phj?NJu?6dgd{9#l`
zPt<Fts6((P>U?IZPR+W_<d6RrR5g(_d|%Q*!Sfu5C^;wE7nK~kSmdol#HfQ3i&pM|
z;Jz|_cYT~2wEkFM3LpKT9`rz5%hPF!1`mj2-ARb^B%#q9{3OLEs$smAm2a*Yx%&}l
z8hGLD%-81)Xmf{q65FL_1ABVdyD^#3@j4BreF$ocB&j*Oj8Y2^sq3|$mkj~HJz5b=
zX~3A`cLtkDQ~j;$0XaK9J!Kg9OCzcJ&)84*Yj~TCC|?6I0+v7Dl<Mb?dV+@txcGc&
zdmy>8j`EXq+>#3@G{E!4Q9ZM`KHJtKvZx%8vW2%6_F0}vdV~iMu66<qUWVUg^sh><
zzcd?LOuEgyrDtcY^-<=Rehv(=EEfI3Gd6+SLmoGoyLzgq%f{_2L|SqQe3C0qT14R_
zm##_80}_<ag*PhJE<ziQ#5gskX{3qd;IfCk40FNHkeoZ;nuDH7xw!05FI~-TJ?fr3
zyRw<#sr2I#Km3w0kxTBM1NycEEgye)?3f<?p<63|B5&>f`p3P_p{}(8^1I2?SwfW8
zW!f*h_^t-^9(~@zAxo3{)sbW2uYpNT6`<y6dO!ZKVN`^)iQ^Ab!03VbFq=F6*G$@G
zbLKEvN<$C>E<W@0SkUf%te;mAgEHahBf;s<6I1LNz7>o0TFWx*F^XD9jF&O;q)QTQ
z!rtAuWjvd(`}_-Q)I6OlB~UZyUOTC3zwX1=IaVWJaIfkD415<-_z7XU62#&cY)v<)
zy~^szw|(4A<sZcN3OiZ$jKUBmOAK|`Pc!NIHsdJam`sQfhOkS*9EhGlnH|$)nA$oX
zYEZPHTe;ilNU}8uYY*wiL#pvNMlnX~?P}5Vkjh|zS+s5PbnWq^o^9+#5cy05zZ(2p
z>fj;1JW6R3Yxf|e_0R1L?~?9^=JT&$UqcE*1$`j=O)c8`MOaYsJjtJ@S5Wy%jgpKL
ztz!5gxHB7DNQqwG#Rpy55@_tbl*gEE3{YI#m6c@eNJj&ZBVkA#?m=bS9TM*yr%1rF
zsn*mP>7BM;P8=#luy)>{^sCPeiAicM@5OvlIp(tB*cc`IYT=8B7kO?*Eqz{pGga`r
zS~{f|dT=GJG4p}mZ|+o5G{j*$6niyoU$aJoh^(3+PV2%NC?jAW<3~ha?Z@_bQfQSl
zU0;-+#t#M3-oJQsrg|>ATf4l8WIm8qNdo=v>O-BU-XG#nL(Vm=H`*naibiZl(Z95H
zO)wz~RQc=`bYAZdAUyBjnmQPrY4jk@P$uJ)DltbA&nt?bf8-Ta0qLO|4nmQHsPir-
z<<*lS?z%)B47`J!h_*0C!A>Pf8d?*JR<hi?6vSh~0waL|ChEqX!crga;dOfU3{_b=
zT>xas5NA_&QJm#<^l{kqDnjWv#fheueMgMq)$+whN3~N;OQFs_Pa-$P?ZTB?jmKi9
z0qzd3^*iP#NSrWYY<f@Q&Lv+YyI<C~h6i0N^7By<^hgWQtfcNq{Y4Ma&Qv+k!rs1u
z6`|I#fktdZ*9Fn)CrlA4k<dPdV;VnovRW%64w02UUxs@Dj=B{1t9a*+5lyHeP1IpP
zJd#8f^fbcj*H5)~e4;JD$(dFk_Ce{}=LMui-F*eUR+|zs^^~(r7tvMNJ3x|w)Qb0r
zp)OF3N!%cHF?`H6Btm6r{0M8c%$I>FKJ`=aJN;({fvu{JJlAIBc>3X;Hl9BU^#d6=
zXZf!-+~-xPG-@jjUQ7@v#g`VA;ow!QrwDbXB!*Kf<S471ByaWB7ar4H=YGFD$bShb
z)>%@oCQu?F($#k2oGl~sV$O}w{iqOjt%RF!R79;xh}81W#O!?+p(vQq1YA64QU?C`
zKPZt}>JMDSCeuwL-aD2t1_xi3lbe`(MKF#e=IqIcBVZCj)xO&87G*|Tg<#8ZMm3x#
z_d+&=(Db@nnj}pPt2(L549Q=Le}BP*G59Vvn~N^_{U2QlL26fWEo|D_vZ4<)%k9X}
zZ8;_d{<PNPKVuU8i+xD)Zdohmz#}P1Rv-rVuVTu&Ib*{M>PU~;$ajt)9jc~Z6W^8m
z&Cgb%)+mSx)Pt5{mjoN`|EUI-NeUk(_?;S(0kI^;G<gJh@muAJdfRM2jDH0Mtq6)F
zG{K2fAwu8Ak(cobcQrwab!Fl?Ypciqb_k=w=pcgIDJCn()@smU%igxBES?<mu9Iq}
zUze4z7$d6J_=sMH<Y8zEJmYGGJ$OjXpZc{bwZ~?{AE4jvl}I4y89T4-Xxa$M{sS)0
z5AIl1jj0x<VxR}(z>Q*nEKHp$H>_>>X`=)V49M0%pAtW&J@ysR=az3C8-j4&T6o+7
z#+1ch318AyP8jI9Znw^$^C=f_x@K}5r^%fj4D}0c<u`mGzuvCA*>xMDb>Eh`rWWak
zwyn{o^VW=35a?C@sNQ`rGtEl*#MTba9501o%l@biR-e|CWLLMt1|QgVHhgdYKKL)+
z>j&ZF@X)^g%KkWBtD&rjngK#R^9;nF%3DLw@yib^Ysuo5aL(C|J>VL3pRS-VDn_W=
zQwVQ4c?s9=R?973wq&rd(VFt)^(6=fxAXa0SHf~9boQ_NpEwV~5P_$?C%-+W@F~uG
zjy{DvIXkXNyxHX4_?%?xDcl)1p?%)}a%R$K<~KLL|7;wuy8FuBAC}WTde25FNm(k<
z1n2=c*F3ZHJ3M&QODa*hB@*WK`*{4-%<kp(O1uc=`BBn#WkRQ(pVuMmH&-7tn=*AO
ze?wK}#^fW?oYsC&6O?O2%I>dO7oy|ED~gW|%obO^zUgmC_F;2hf`Yjgqr}30a$i0F
z!Smo@uFwFSxCQM=FM2T{LAdsJy|R-J5RTK~JZF_(86IaH7zBb0{-$r!;x&~g8C^6>
zndd4b^^7L-k^VK8DGo^9S>D#3!UcnVLB24ts=DsY?)8DVk~5VV&*5{q^gIWdbYoNc
zbDB2-qFC5<SS;yG5#2cQk(O^12P<>#B^)O;JWprySN0s`9N?bDZ1QHJ&k;gj3?ffX
zDREj?ie@*K_g~)oo*ikUMqlzEYTtj;Ea@zt=iB(yZqU8SRJGA8+F)e)y!+faru7On
zk}Wp<{MwNAdN~TRKs1~6r>z|w7jFfV%rOSap_dsU+@}M~An8W#Gf;oov_|sPd#v1K
z>^SICXKW~z3&j;Xw)LK_mQ3jvzw5dfhFVyz2o;dx-1;eX6<+np0ni^p2eLCCpPq}2
z^Kv|lrpOUNnf7|c6nD#35+fY((Umi=ai<n#0y5*}D4sDzdV|!L&WE#3E0$zAJ(}Yq
zIqcepih83-{L#->p0JIZh6vJ~QllFSzIP#OFb#3Wo9am%hn`E2K;|}{>N&(@oJqwx
zrgqh1t4)nkn$fb4w;jzXl5Rd+BBP+rdlALu^}!c|9ScSsutb}i^`~2IMF{yE2Xer=
z6GrQ&)EQ{Cm&5@jPHM#C`R-z!LB+1PF&+QGQIi6PWiAOj`CZjF0yBX-c#=K&RhH+U
zW<(`^sSVy!{YNLnwok>)v}4@o$+Sv(j+GFnI+#4YrK)gE=tideAzL|EK$a}yYhB?1
zBT;PA8Qx;|RQzvSXiwA$j#r?7WT&}$b3SkGkL+tR89bCvFM%8O^$xHz4`i*B3Yo~D
zp3kh#yVF%XSq*~pItTQ047AFnIS2BFDcqw?U*Vf2UV&~}FGZr!d4+$<yLu*gozOGK
zxn<WXts9g3r>oT)%+U7R@6`o8&Xt5|PrWX=w3g$H-4DBocKAbSwA}E<zmSVhCw|EI
z)0Q@38Dccl(BL}uO62^Spn}(0QSW`0PzA9wh6{QQCp4rH6D8$t+9Tep#bLJc+4Hy5
z4J_d5k*1Was76m;OA}XDg&)YkS>jq{(V{M4=4)%Rq27s-aJmisI^m8Zy~sv4P!StQ
zi5SsL9R16D>7w$6Z+7EuhwUcOtV?>2-b+;`%XHUK#m4kZ0{KOD(F;2hm6)I3x^ts#
zfh3VhfBwfMXtMNuoh(oYk9=cc>-%d;FCuEn8H%U(W+DaD&Be?uFHCg*$UiOW)+*T3
zm_!AY{HufI5(LV`d+z(=>f{nA#j~&1=X@a%hkmKJ(FkT;e7+Yd%-b)%LVlO}6!(yz
zD!wXK-puh=bSZK}Iw@&y-MskMCKb8~W7O!-w>=rjx(*7x@xR>sj9rm@{pIAJEl)yx
z6V?QuT+u&&sQhO;tR4j2r4d-U9@)|GIOn~dLedEp?ZEg4XKXSPhGrPQ3t<x7+D_|1
zw4G?5)gb((i-qFw%*!X1pMxqpXJ)=NRK6#$L~k0z3Z}WLx(l&PK%eKW`Mm^lY<u(3
zmF=AvO^}OrH%eT%d8gk7CoGX*uZt4)ZZd$oF#*C31IZh()o7u`&Nr;vo>34Q*MmjY
zVqvN!nDgkHjulwh%$tto;^s5YS!2HeJsLGvVS%0E`>vDsoPhc-x{~*4>OIy$EU~nb
zb=SOiYp{ypjy+7Gi?Ak6da=2mq7sIlRV|Pqp5MTjH1OqHWjtz-tu3u2^w^2jt^{r0
zq(a7xfuSh1wMdHY%z}GLk-q$1n6{ytZ&ZWn%gE}}KX&Ns)d)_N-Z<XTq=x9QG=g2R
z6yL4W3QRs*9$)Gl<>_Ag3=dACI5}HU--Rbkzh|^r8wblLrUtQg=&WtgCkC;;rj?k#
zq5#cP^!6BQGTAL?Mr>dZYdV#a_Uvb0+|yKohNme!N2_tpDiU*UP^99fn5big*79Jn
z);MOGSl0t@%a_xUWwSw5s(L9Lj#nVsz+#ZhQeXetj0l`*`AxXU+sV-;T#E&;`^^7Z
zHRL?Hjt^+ud~}n65Z8?6Xkd8cZ<C-}x8puDTSK%sE9voj%f9}+6iB9i)NUdN<ECt;
z()~7{o_2AU7FFBk8Lz|VowzNv4{U~hyIs!PUu;<JTf45d7yWh_lG-!$U1SxRKN_p^
z`b)i$pzB>c=h=5)56r@K$rkAaCx+K6X*CEZ54|LRpM$;MV<!frg7XrGk<Pa+T0Qd-
z0sHtTI_`&2=~KyH46e02mX8CRqJH04wH;ptuj;&=oDuFk@2f`K_@f9;^C8O4ItKoq
zeKv9f-p&ct2<1xO`iSdRb@V$W%@)53o)g|&txXz@@z(V(ReQAcY?U_d(*L|65!v|K
zv)jX-E=+ObIHp&xJgsBshgk#0{2qDu>-|fHp1F&<3Bd#$KW+UzezDBYI&uH_nVNgr
zOAs|G6<f3W6GiXQ_o}9jo!+na7qWEfcelDrT?emZ&9t|fh(!!$8fT1fLke(F(Fg2b
zWAFYpGF-oQ?5MT}p^5I|I|t-7ixyiQQI91vK#r<Nc1rFBO2uLNrH5r~%&R_H=Olt_
z-Q=XkL0e5RXJN<9L$_KJTsr$H0}nUh+D$$tQl`s})#QsL;YaQ={uL9$hW@fDIgcI(
zqjw!v`E{#xdkD)=HhtYrJ&6<55SjZxWm(kjCqD3`Rc0=t-z4p?Y~9y)ai6hwO6=w~
zjUVWzvpQFn^dBQ-Jv-x;Hdlrn(^wWoFz@iie(CIPt&$Y3+ki2B4j@YRW#0n=lp~Mm
z+*dMPBk4>cBkYXyRc*pAHZF0yr{#aZB*Ly<x_z#AmdxL`WY;&sM8Bd=(~PVjhlcxp
z`fWu@-k3^u>>*ogw}ozr2jXJvU4$LCqC3{PzaE=f;_5dN$=_g3aT#IWh)K={g(ad7
zWaaN+BWd15@289wLPl$n1fgN4vurJCoaj@to_e1)VJ%@oxz!07Y><==Z$to(H-Fi-
zg@IfxPRJ`!(TmwQpA}=xMr6QIj~#M+MphgV(%;t|hknsi73EjwD>=IDRoSq&g}Xb)
zWej;VyghR!=KQ&Y>%@;_G|HcEF*nK-Jwz%h8B^I(+dGw$deRX3d+FARjQSgNEns%H
zWu!De3WHl;N;PP$m)2x33Gd1&XogM9H+e|}8Yhh7e=9k&F%e-!%fPk{f!%L!O5~F4
z{d=#|`8$oNf{)@1;cisGRko0P_WoFnO9ZJp(Fz^x1;ZJ?zx;z(|0=%-fki$)5zWy4
zrneU!hJ@uE&W847?_Dg>bNrP*I!@=r2?euQ;))s4Tb;E0v-kYy`MTgzj`|5yAHklh
z&onh0`Sn8tpNpg2B%fVp?by~hB1g!&&r``Ub2oEa6Z^4?){;pqy<vjYUys^BDh{ur
z<ZL?Th6v2LXTM4z$}`;|DF;fkk9!FY+DG7{jZa+S_<G5ff9Cb?Dc>!g$#$DE3BCGU
ziTTC<3cmaP?wO2UKNS+>h^}^dh2ovlJ;k6Ni^E9ceTZ~ZZoHA+?_Jr6I~F!E*6(Dq
z_x+ljYu&|A@#)pWbANn4YGfhjhkegLaAY|4aSkIjvFU*XUv0Dl%d$dYcb^szEYZ%d
z`?MqSn|f-w^r`z4*krWxV4BZ+QY_|25NnbbWJh`%TUZKupL`>X;v81&SEAm#LVFc(
zjJUj>p6RUW!uMi+8tKD*b9k~d3CtF<NWG}9wfd|AR_=|4x`+F-qA34{QLXJlMXX03
zA`DmCy3YuAJQiY9EUw!66#5MqxiKkwt>UjA9+SF}FYEUoV=Bypl)AMuuSa^*njADa
z|LWllwDrJ}U{c_0$VSh3X~GG1vS*r8{JYOn6z3F-(tKs<!glYXWG+m)JyJznpj-9h
zZUGbjs5W3U6r=etiPlT4D6QGDI4|a4un(2*y;dSm2udsGyC@90XW3bFb@lGX<VGNB
zNFQf8uH^|$;FfzM?qYZpv~yNYqBk;L&0)^|?rwz8;69Sh&-@5&*uhjkP2lmaljKkZ
zWTi5xYutH3OI}(K2eXt^0Xr5qKR)}ghtZWP+Wv%RXps_~4=3x4j*a9yV={;NH#$rr
zw$3mt!~K+qmu9Wf*vTXTRNKbL_lNuV$rXHEB5z+w1)oyuX9~UO_^|AArA~athW;Wb
zB@)A_8PSNvI`BJBm1s2hOz0g&YoLa4<=yJWB_hfz+f^LP5ezPqMZd_a1K!qWvm|+*
z%n#3oT>MsxRFE%5L|SWlX>YfhbXezQZ5WLO(~A%S!+^vRVxnG3=m)_U=lK$9z^<kA
zG&iCx<gz{4`Q?mN$eRgYO?U)}0+y;LBq;W0c!hLbrav<*?dKmI_C58coZ>*ab1n^x
zXEF&g=u32?pQ~7ZToITX%;-o?-D`^*f&<T)xWljWg!rvVqg0EpCWO%DX}`0Ukc6r0
zvBsb+kjWBQ$*bF@n!%3n--;Ffo&TBBv?H=X_I!y*{^u#e2a#FZ;cz8;J_P^$*~r>E
z(x|lQ!eOsVL7V<M_ND}1w^j;@=on(|5<%2BoU9@bxV^9EJr>+<{0Ejyj`0y0))ty9
zl@6+wYwcT-mlTm(nW&*lIf(_|w%HDk8`4K-Z68FV2?v%)F2@i&zO)u!Te`Tf?R{11
zuTN4O^rXBHQ}^Cu<7N&lMT2Iu_=h&#LHGKB#c^Bc(=|t-!J|)Ko?XwJdDwvA&^>dC
zNO{6X^2gdEKJr5vM4kRVe<u^jpN_J3t5$_eDEWc&9LvhbT*#s^Tc<jr3a-V)l&UwG
zcJ^He!2`yHa#K|jrCCTVhC<TU_&4b?yB`Ch1wyd{%2uWmw}L&RAGHV8lG1O&TIu8R
zW_w7XOCyvif~V?)#u_ofl-&~p7aRjuZ1gfe-r=nVN=QQr=?J@fyVEjZ2ajkEu$$y=
zic5Vbczpc~@}t6(`#fX=$m<d=Fe#j<URbtL=&SqPY8JAbNfAVAKyWMTiBw?or^?BG
z)GMwVO*HGYgJgsRH<E3PsQBX|h!!&=kRdDv1vO78CZ4Rc?8Ej$)|M>`hzckJ`_?`=
z{@NBWTKj@H!-{NCM_Q#DSq*iUj%U{r2i#;w-5{@z4?O30EWDT>i7W<7^krGT%@sOl
ziqjf>FZQ9T#Yoic3q&^QE#2X&cn+o;rnSu51ALQjjVwmh%ifu(`NoP8h2y9)yd4f@
zMToB8!tkR$ch4GcwzvY{Mm#qesUeDUEn)h0>W_YP?yb_E=t`TR!;!5dKbX>ei}-FL
zr7o;2_1eVA?x}@2bRE26Bcs9-F2XkCB*~k1UQwDQq|($0`WYigJGC~@I%U0}qUAuE
zu1msC$Z%eqCKkC?2=CbW`>IueKKqgV-|x1v3qiK*NWpWi@$=MJk8^sjBEEAT2Kv@J
zl)W!;(7iGtm-O-`j7H3s%@J4Kr)^2b(H<|>e9y~Eex<%yJTa^&OYFFrcTP(nYE^8+
zx3eg6r0n=LaR}XO`KgRxF`5i-Rl4{x^ucx_fKyLOBVk0%l6aR9U2{;@%T~)$n==~~
z&(5)Rq{*??uS%R~=pM&DDzx{V=KC6w@Nd;dEo0)}j22a-O?UXXNDh0z*<2y-z3W4j
zbOVmI^9xSjrg>jQ(XXpT(F}nZC6V&0d^f&)<EV4~`15&di7=#Ay3dGEj$P%II;&Cq
zCl_EY4ByxE7)v0f*i=PwdVOk?tCS$WBt`1v9~%UWqhuO6$MSb_J)=by6~AJfhyDE2
zWZ}h`qd4OkbXs)ld@VS{pzZ4og+5(bCa&Sk+)FC8<P6BsTh|MA&@uVM;G`Qi+eDnc
zFNu7x$b0L7bxJCXs3wh4Yy6I`7_Np`om)FgSxh}O@>(U&C>33!GanuP7*f?=x$<T-
z(2P5Ah^8bx?ZKWK2q?6{C=~r9ZG=(QZaRlPMPIp-f?I+!PCxt!2mdY_I2^OOu6!W<
z<LP_g76UmM;vn!4%opTeV>~l9#<|)Lfw*?j`xMu}gM$^Hl5jA;>+s>ZVN_KS+#+{o
z2K-!rsnVppTYOQ$e}`L|2j8mtJ4@u#Nv3jGXV6^=O`mx*P1+MfafdvdY1_ruo>&QY
zCR;f4fZIB<O$;aYpYYm$g)IK-=PyNRvz=ebk2qyn<(tXKRzq-HrQeY=>Wbn;PMpaX
z<+b|J$s2PJa{vih5i49V)>Z{D#s+5ABXN%8!T-Qg%8z7S60Mlnyr>+6oi;nHPVMFd
z>~%b2+T=qk#)8_e+(xvNvvTP@Ed}Z5sN<*S;y5@Qs`wwCX%NeQ!???H*OY`948;T8
z=@LwvknyFpAcnrx!DJya{)0P;-eyRV#PNHsr_TY@gNUM~l9i+P-E9ExfB*NNzoFpv
zZ+QK1HR@jx|JG=?Fb8pf0N7fvl%|huZ^3Y!e}&q=Wqn03qOIPBN=k_T;zUE2|8vWM
zxaDvR^h5M?B$3@L_U)l0xA}7Za{};-IGUdrgcAGG9w5$t{7$sOwD5-#1~%oSKkx-_
z+omU~J9pERk1^ud^DZ80V}vl?g;TX65lz>}rEj;5JQp#Wf{9-$@Gv+0`GEJkJBG#f
zG$oC{lOJG7MEtKjMclWiEuuEONzq{W)g#Yo>JkmTdn5r-s_`?OkVQm*cV;k!Go^uR
zOsC`5G8(SL(#8v$CP}J<yEF9aD*s;;#y1fhtKwfGqbA>EnSkPwxn%EO=qR%L#*ALa
z6FMFsyxnr%kInHm?)`MfApqPEe_vT}5c&R+-)AotPmJ^~iB(FoYnz4h*w2kwe?ubR
z;>MS_Tl^^%*v@vrDccEyB~8<xXj1sB4R;O5g%42@Ba}gdUHSmE%9N?pZ5Dh7rO|H(
zyoe87{X#?f_4%=xGSXhARymcXPpfc-uPae|azv7^!#$(#MPpA}FAvKRxmDfud%Do3
z!7&%FA~f*WPx8m*Q?&hmpfIRYgpuBNnFoJvlYma=_xyGAUHU)Hs*(ZI0BayYb)1gx
zf2aI;r@iZ(^^pu7{iquD&f2S2<R3Q6dqSKtt?nN7jO`!wyJB6&CJVO_?)Ct;;BuSR
zJm-5ja9DFZ2R%^**Yg|xdy`!VsWx<B;`YTvmwVTN24Yy>+GRbx(JIEWRcexm&pf6%
zKLCmI-{l|V5*82fpYvcwJxE}3%aRbt`Jxv?CyP;ba<N;#N70#d;Sbi4JPrBb8}}Y7
zg#*aWAOVAv1PXgK`#;=?|6*%ML(v}CmPR<LdEFw_;`MQhZ+}JE0OLmk@}<M^9=H|`
z4+PlGW+v@2{7$SdBJBTXV6bMjTUJ|U>P1-n?H9Wy-{I~7#VlS*PGAmJSqD;nj?Pp6
zhwQ=knJ+*QkQ107xBQ=BaQ@Ard0+4mAe!>O=DdFPza#&BWWJ?6cD&X}{qN7g|NVF6
zJpkO1RGTjvBqaYUdLC8SEq#wn)4+-1fA#q!8yvthDO7wZ{~xT4cX%@hpVUu)+5=(L
z&f8{hCVYeN1}PrEGs*cco{7h8ma3lTVsOao--j_;^BL@dwE^4?M7u|t>)=9&v3wIQ
z7BcjH`|tYT{rTMm->(E_O$E=m(!5C5*F4|~VG+GITu&+5Wso?}rIh26Z#GmW4_5s5
zC-TtH!g=9Imq7dS`!-;^diqF~>L<lH1_NM)1Z)G^%%WI^@Vu)g$o8F^Fc^zkV0B#n
zEb>!mNTQOWc{x&i5QI_sRwDv8R&E(@q-@IjxSc^hH>gvO<<8GE%IfP=W$E6V1?fXF
z*mNtmp}_tq?NXl~KbxjBzgO;aI(T#zC8kR)g|gT<{{ey`5d3)d#}GHYl^IP^@mrd3
zgcQZ<#%iL`iIcA1@x(kxz}MuIl6$k8@pWbOW%8e!Cfc=Nd5SmfbY3eloDJ~|<oDV)
z?l)*4B%+7aF}%ZXFGoB*R5q3&0N4L&Q>V!pYy!;Ao__1HU*)cbpnvVZNWwWHdk<8|
zLjr?W7h9#5jwbtN;CN=IfWPy-qJdHMm`GF;f?;jG@bZ%75z^%rTI%>Gc`N+3G40<>
zixbuAeLReU*o3+tFZfV0+qjZE$!ecivLRv3@*t6XbYC`<VqP96>lr(aeN~ZsUF}=q
zUl=5&H$xsthX@aZ<eKE#cIqV@;VZRQ?*s22^v&PxcgO$_DHRz86_v@C&2YX(E3w8Q
ziNlE7b|vh(KAg8_?P?-!Vzq)W7WY{C#6AmR_qD@Mz-hcKDD*{ien%tKX0sb}81&xj
znfzY2Ot4u}RSA;bDNq5YYS0bXUClZp?gbJldqd9t1iG)r>&yZY5=u;msId*-oz`_&
z#(gD5VHPk&1kdiQnEqCo&-jVz!BnUxzsKRAgE8H?bA<3eOhHTo8VO=`7n4G_8T5&w
z*>Y8>^Y(Rae$iBqTUZp6a+cIjASv7iV4fOqAl?51jNh-IvjC37!8EK)^Krs0=-tuI
z`ql`87zpE6^0mDlHr_oY%DHwv<TSkLO`^Pg2B(WF2W<e8!vhuWW>HKMx^bW3yT{>+
ztQ&h0GP|-CEMd@#ju0<6R&p`C4^i^xg^v~AN>O!Xcp_R*hZ#3jAe{b@Z_vSjgMTIG
zjYgEfk#?iT<=AWIsOoTywW4wgN6GI-_s^~lhi<EFIe_cTf!5vTISl&9Z_DK`KO1fS
z&Nqzldl&7hlJHIFaB^rL%E>m{T)6^Z)xT=GG##2UMF604fj(I}QI~bO2-?orEvV^o
z%jjknWlp0odryugin^@kR_=as%OIg>VolI(g(LiZ02s}4C&~Rn_&qq?gq)nrIn05h
zbW;!*n|1({cv%oJEC%Ao$oq8{n?-W~jA6O;quxFs;-x}s#A+8hkp%yhdLp0<{Bw_}
z`>!;+47|jjjHILFrfUG5Xd1kpw2vbDK9>TFg!`x>`-Ft&TABw8WVgdb7Cc#K&3e8l
zvL!#%Wrda{9V(|g4R|hpo?2c0@RfZQJcR4A-)4j(h*;LwOh{mh!IW^2n^zSMDY>TV
zM_s=kH?n^+-}fWR%LIHS{FWti(u__NXOL%RmMpEDc&7U+Md3cm<#MyAbQbWatif*?
zeh0DHJY<{R7~WgAH_I@r%3mT9;s=%$ixyI~qDRAWJ7D<%w+Mb$&_(6VZB&4=mb`IA
zMvk8s#+t+<P#Jd^z8x&|P>tYn7ZZ0bszOjlQZ(h`$6A$pf738V_1pr&L<<^U%9MX)
zxR0j2oL~Q6{bG|C)KSWayRjgZFn^GIo=8aj3Y5KBAXc3GfA7)bnp;LEd<Q=F5@(u7
z^9tNjf=R54N@9M~F0p&03-{5UGaK2x$-A4mGOnp=r}^rM)>LfZwvM=4z~+f%G}E4y
zYkZA(Sk1T#mQT5N`4zHnuBP$0#F#D`FZrlfucXD9aY}w`NZHCvAR78GKiT5zSA;IN
znINwNwANkYb}3?jQ#JdMw|ZPdc(375n*z=KUeFCy#6Oq&QJ$rEQ@~L+12sj=Lkw7m
z;o2T`1hne#b=LwZpsk$qYk!Qt{0z`D+mgDf54it*ja_;!>k+6yfHSE{2>SPBB`+vM
zao(q{Z_)(s0FAd~354`k0s{j!w2DR^3?5|3Tchd=dqZC!2KUR|bzpx~SUQ`Ta6*6Y
z?ie!8;lEbp<T&45bnh<c)>d!!jp^cK(W{3vOMjh?BQ8^6-g0C`vkBl@92-E7YYyZ7
zVFikiQyCXccIY>v0|ts~T|q&yVl!od4l?&%&U`BE9HfeDvjo9Txb$1J`)GN~Pl4&2
z^ZN^&qz|TRSHtyQYtxYg8LZO}XyULDLUE;HiGTfMaHi{~tN&KcW=wkKHoi5}Dfrae
zWN{DwIg@cE%+=uO-3gsl{m#GY_wIJF*-0w>yXoD(58yvEAC4Q=Srbk{a~nIWg!`e6
zr#`ddm+gK?G5&DxgO*6~^TniE6!4ik0L4{6n**CWcz5^sTl(sNsHh(Zhaw&6`iAx~
z&6BG+6TZ~~z@W`gOC>0Eh}$yoM`yob1pzQDuta@l#o9eUYZGa`1ofz4<z*#UQci=S
z+&Jn&3lqWr-Pb+@u1w1hJAfk;27!85_8<rpH-Xq9-OEDHA{h$HPjkm5onLGfyD5us
z0qc8gNO@NC!wW!9u+sx0?^~c)<*hEv9Y_X4>J>vw+CB!aO&t8#Y2)eg0B`!o)dU96
zAB}tcz#<i2Th5bBsk0T<Mj!&)q~5_l@dX;;7TF$9w9ImzoZ~Mxdo&)>&4HkXpSl!W
zco>Uj1?zyK#R0_XN$e>wK^A&oI((zc?4Ih>$yxt|Ht_Ivud~*4IluRy6^mB6+b3Hg
zicoueAF!D&4R>a*RN_5zaOI>VN_N-Sg+V#C&%S14FkKA13JaD*-gc*jl7<TX2Z8D3
zk$+}vOMk)D8CK|4Z{l?f3+Ue3*R8_hwX_RRD%F~7Kr(&M6&T!nCFCh=@_kMBqc7OK
zF`)`PzE4q&H#4^Z_<#*qli2^#Yddg9wvVYo>~sv=v}!Ya>%mP6T62uJ3YCAoxd~GB
z`FY5Lp(PH!%QS9nGN}HkPhN>b%je7ID9~!A27z_RxY`X2-tgS$y)bMIbj?}ut2i6V
zD#@%=Bl%N(cI%j~bXm<ef&ko1Y}})ANUA=~bLms{RonfKpkaHe8p2&0=YK+=mtz|5
zd}_}?GpV{|WMwX78+`Re_I!3$njIw^wa_$DXHR^(shB{`64eBrn@j>`)e)>)?3vr4
zd#`)6qt+*ddawF!bnfp*@6!QFuGzFZnyFK;rphODav|UbrItay7v40-xpuJdzAX0@
zBSZN^A6Joi*Vr8ZF?de*{PMwN5dUpyP%5eZ8~koc-{sZ1&S+Cs#*3y%XsC6dxARV(
zLT|&o3RSlyK*Z&l$O!C})z*S2@F_xdK286_$cr!DmFu&1Z5qdy;fvNMfRL3zcag2n
zuzd0-km#eofY*nUAoox1L8o8kDGjPJr>b-s@{QF$`D|Q{?y9X3(1|@hqNMTu!(KE=
zdS=uS<vv$NRoXxt?fe3fAw|DRkpS2{p-k&pzVIdQYy{%vnV6GZ{IxT_YN3CCxZ5jy
zbiIq6l|Zn;7dhJLwmjPK%Fm601CZ6!j>5BmFj?td8aHZG_qrz^;HT|;HDnW5nA(KD
z%=lngJuA{%LsJzTp{X`Cd@_=U+O~w?`GtbCfF8gx*c~>?j|x7%IwOqvi7+x37VhKN
z<7j|9R|p8JRe3z}Y?1i*)aV+S=qc8q=V22O7!sXm_H{Vp=4edS_L%NGM)>_HRqLCf
z>bc0jhtI{ApGQC%0LA$PB<KctJiPj2(6DX+UJqjI&b0@13t<WZ%3w@Ue-OI(B8EN7
zXvcoeFF^6YK3&YH`mH6rk0-#W5mv3Yrvay0&ivIXkpAVy@e`a}NxNKC#qZb6b=Nv0
z?SHZWF4qFpJ(}Jy8E*=FbVq6#!2<5+Ri{q9N~KVFy;0fKRYe+AxGLe+Xj7B%<_Ud_
z$#3nn8Y>a{gL9ZqJ?i_YrD_^||9rcX^M!WtS+fuNGKo`v!x%99=2%Pl>Fw~<9NMR=
zV*e$lQnmb@aE{k0ai*5wL~zjy4lAsaLjKU9K!Yz9Hylx!S8X^nDG*yA3U~jza`DnH
zeh53z%Sg7l<v;t*Q0XQ%$95703?^*8S3s;`K=r#8phN?Q`4|MSpw#P@jBALSYY*MR
z&+nrqwDoxPPXMtRYY~sm_iE`uyM(LuDInMJ$Eac_+`W%=s>@6mLlb!I6{2Pfm~uFL
zYP4N^Y%~XWO!|^t?*xZT7_nN=PRkm!lFgdw7L_HZ?5XzmqU8whWhsdD<7130UQab$
zlVjM1VuoVN(y_6T>@+v@gV)K@L@?+R8kLje_Buur&_ui^>02#_(fcb4_B=dUgWSvj
zj%A+U-mKKe8OPDU9)>FmjW|mZ_9cPUI8Ce;N<L?0!^g{S{k)io1nxE)(+&Z~mb5TW
za%xZ;U@`{ke&0UdH~nx0Ah`IlRBUdFcx>JWDXh387GK<te}ny!4WQ-7i=rm=959)-
z?qbstr-wV<>uTpfhpx2T(gbD5$W?q%r8W{8K}-zk?kCIbY}S6juMpzff{eru6MPXy
z^`#-|1u}?K1xGgdj24Mc`d*zwsZqXNu>l#Wm<S{sd_1b*(e8l+!ewCR!xEtQ#cwpb
z_A!4S2Ft^i0~nhhG50mVCGMUmFZlrdAvOs|U*a;Q6Z7@i0AoMbBW4KzO56(bpBo?r
zd{QN11x*MuzR8_Gp!?y$yks(<32$V*qq}-;#%lnv)YpIs<Cw|qmFtSrX3uR}Iz87s
zGnuhkYN9t1A`4|B>Lv<x=D-`LgB4_uyqj`S_Phb6(kGj=S6=}krA=z+TLcTSys^?e
zp=CPWP27)v3TP~sR8=X!1F8~RGhb(4=lQV>37EXN?EdrQn6ed_#Ue>OrfI+C*h1rl
zTfPS<?UYDbw@rY2oy(0-fM`?)p6NWH-}$>kMF|gO+TaG8B%e<6O<6o%G9=Tnr1p+g
z6vxf^r=2U*PQqaW1IjN0scP^7!H-LG%T^nXB(8w~_{B*Z!!`f2*o)5<k^Ct#0}XPI
zTkhPhYU#0_6|jHj`yNR0>kCDWR&m$iX(x&cgmpO7&|pd7X3pni;3+Njg)klsaqfxs
za>k}tRaYro<smxW9o(W*LYb;az@YFo>iX@wx6_%g-lKnTS$Y9^?|cq0al)at-x2F;
z>Qm|}BuU%gio?NP3d^EHVPq{2C3E)idz?hD<JXf|e*!;)GiL3A(~%bw-XHk2j<e^5
zd61#9(inqPH%Dfk9@%`)^zPu*BAq5x{gJVrdQo%b^Lu*&(q@YvxKOK94P>Zm_pPr~
zSnw6h3)~g#3-=>n?~5Q9ek7<UKQC*Pna+)S_aL0;{yNJNhgpLWa-m<`Azyz&N?zQL
zWi-ZL%#D1nykST~VFXx*GfUmT-XB}$3M?Neq-a-Y_u%EcO7nYsNV*hg+~YrVfgV(}
z%IoOuS8ZoFcDyi09F^p4^OV^bwyMMfbgo&F1p&_h=Ctp7o<qz5hblU*S_sUd6Hcqk
z?};1XroTS_{CbMaGE4n<@XOn@b0JbeszIyQYQmE^ve1i}A(88Hr?Dx=Q$#hsG!%3X
z3v&B|1+wEu44eYbUj(GUBb~CYx-uiISsHUCbEK9fQ;4WjoCa|3YDNiE#g<;Yy^PID
z6^VT=X`op>1mftl`n-C5wQ_Ik==tM9F$Z;4dI4I$%zvtAGJU@$3It{1EE61UZKw)#
zD&K@5DD@rRmjE;yoxuZ9`LCT-pXanM`*s>q#{M79-aD-6tb6+vL<oWs1Sv{Ks&qtp
zkt))pgOpH=Nbgl3AiaYiMd^r0FVaDJ2SGsTO?nj)dW&ag#(Cc7ch37>XU_S@Yer^5
zlJDMYuf5iNe^zOZt!ZRhdAcG?+JgZ&?_8)ua=B4go2QxyR|J}O$aH2KWI;D~HF+By
zr&VW%zT~5De9qj~Fmd&^{>`&uLF4~b3crl?Bhq<}u82MD&{<7<z?boSz+1Y?K-t@b
z(HTo>m89n9vzjZ3r{e-uTi#dhr5;C{t)`&|L={rpFZeU4db~{pF%xbB<|Q1HPoF!i
zrnL;<=~y3kQ`5@kp9a7JZ!~*Li5Y9uyT<9RQm;n7^4$xdeJ-I+d~|Ab;mm8EPm45Z
zFkSG?VC`?rxzn~C|2f6LviG7eX+RX2;_k~&j)HTs_swLRRfE{mt>89!hGVh%I%R-`
zo<GB;-*B7kS2q#u@>k3R%>g;;6oOlA&O&NaCm)`WBMQ)T_C=t8k)n3;+%A_+Y&r?q
zKeMSH8lqNjVO>-%zZ!p*(mWt=DMWtwT!M1?`{6QqM)H978@hePPXjP2{){U`h}8Yw
z*8|Pr*TrV)pXO&s2&JDcJFH6g#wwUSJGp{bz4|*Pf`~7}ug{QF(@PRhu^~Cen@9Za
zLmRzSvKMVWnEAN;8H5DN7uPsW)|BqJA@9qH6$ED~`AuD5$@Jx@C4g_@&1t_y4>`|D
z4BtBT!~pC-X3n8ien;`c)r~bBDt_|A7j{PX5|XB_!FCs7)G~v;g1y^zRfT99Z#W_!
z&AvNEcbctYw>+($nbLafj50`ZCe{jN*--O&JjononE9^m3gfaPIWqXHCx~LU@e=?N
zy)xaw?uFu!TLw98E^fCZJN3~X0!{2yvJ`|*$|;>&je~=qP}jxRMASEBN+uuA;0Pvu
z@@LNzcFd!NnFIa3un|T)RDG<bnd$7#hD&Av19;rhm-u&o<wj~C>dPLSJTh8>Nd=c_
zjDK_k%r~~_3{uDXy*2`ww8xh_*;6fX2Xtz&_J$pIQC#pzmb?b>H)oLtdtsXcEGyQY
z;T4b3dtv>NM_-G@605(mO+CWa7N9v9wwqfurm|ErP=a;INmseWI^87L^rFoqIGoaT
zce#Y+rn(>IBYwXoxg7AW%gtD2`2LONMAm@wwcQ5icWbzh>~AZveXOBa<^!gD#A9>x
zOHIR;x2GmA*Ym!M4eV&C#E^P6JfT>(Cb5{}>?<W$laV;^KYYfq(LPEOC&|`cE~)JE
z&K-l-xkP&>zndNE9>(jGL8OTEaw5m3mL`HvLI|=XrRTpkf80HSoXv1O!S}7L<*c0F
zTk;+0W0B&od41<hjLT)$O%ktQj3oc!WzCI<I47kWS$27Y&JkUdm)<Cq=Z_?(B{?8I
z*U5i%Onc2l?|_+iSa%4>wKllB7qoYdb@Q#wC?@JD7@UWtyF_pLzM(Zkyq3W6@g<*H
zwcn^75{*Z2Xo#vJz6WV%(K!3tuD+iACfVC8_?=6&QnN~J*lr#;z=Le)OAg4_Sz?e#
zvQ4MCuyyD5Z^a`)^>@j*R{LhWoH<wX&o2dHn&&9r!@sj^@o(N*>}d#4!?mbn8Fow;
zNqIQf`8!2znVEJebSRs8`OA=`4rJ*{dDD{5TZte8ikm-8Z1M)QZ52R-WsRvkJW%`D
z>4DcAfmp7pa37|~1M!tm;RoL{0h+)ZtMpLE{QBJVuX#%o-k1P8Rm@*lv4kc)wA<EH
z7u0OROikl@6tT4QfyBHTO|O`=ALYAuGa$p<Ir9A;CcJ8#%TQ?~IA7#1Y8;#wcI0eQ
z7#5Tl7F-S#AVV$nT<5Z_#LR@wq|eoJhtk=_Kh0mX2MkDZ5bWM;)ZL$xnjh~(ryx=(
zx`<Oq_K1)SO&_MUET0i--M@tYirRT>GxR=XK5|LiPM9|`gLfQkF`gl*jFjqNhi?im
zY{;7iTAdo_QuRb#C4Z{DWhofwJ$!q1oyN)Th79ySO@uRUEM_l?P=|=8Rgb@^yNr%J
zdZ{70E=suEAXgn65T;DFkYQj2VqNdI)4Z!x)YQVb&r-a%dIdxuzd7!1j>hLE2N=yG
zgBvoW6!eMG9~ANgcPSR+rSQ7>x<4>tGsq^vv*no8JbD9p5nmlRUa6{eQv_PHrS(dT
z6s?DnZ-hQA)0CEi-C<LHa;A}!f`~F8l<vBcY7mIeaD$m;>&8c7Ut_cI=-%hd)eAjJ
z+Wa>kk&ryRpx=C+&Giv0U^{{O_noo`UEnkqw`Czj&aqDmm50$07{jG+)U?E7io&L5
z*CgcDbZ)=)AoYJH9x@SuW#92I&8a+9o}O-n3Jag)r(H_UgXWI1rxpzCnI-g>5%Kqz
z@Aq+WFDpCIn{n0yU(H9nc)bor?!o9G#wZVoMRVr*caFt39lmir4Su|9mY!)d1ImX#
z8@Y?0oq5!tZwFq3VDHEK8!C&pw}c^1DK7T7FO)9VJ6d(NeA{B{FD(U(Lpd#9&6+)f
z4wLOOqt$8ae~h$zX2l5<gkLEaCJ{QV77Y$MyFtwE87^kLYIbT>LtS(yrh><@=tpDJ
zr=-R&pq!AXD%*aG_r1XotIzXrC*w^=NztB6ZYpAxC8!9Go+l`IJh#&qISBbA=$PQR
z`Fo-yH~{ur1`-<k`a*X*DR$fv2Bp#vOspf_P#3VzgIv9&{2eawepd1g>(*Uj_bE2m
zEeYqX#@VMIUbdssKW-ndUE03LI16tK+Dk#}mJdewXv;o^%&ga}Qt#aIv6C<}VPIQr
z_qahjNq5~*$%sB9@lXjw5wZJ0w7dh3(^hwJmHLPl?+LYX?Y_+LETx~Pug`5BkF?I6
zQ&E?Ahp(Ae8g}^^<YnFz(;LTmW{UNr%^ZdC8;sxZWI5)0R)YML2Zwi;0%}0Jy&yuW
zF^*PyP<uyHYv(Tgi=1?6o|Ih{O-bie&fQ9(O}d&c+B2t4p>FRV?=R7$r{Q9*padJt
zi-WV}eL%7l{q&7U0Fh{ZNVuz0oZ{R0Wu-IH#{S#Xr_}z`yN=!*1}^zi!v#Lq+vL#9
zw5NPsohC;o4Hz~NC7@lgduA(4+~#_jOTEI|5g<1Ft3lXTt5;WhOF13(Be%f)^wZ5m
zIeHTb??@k4X-G0co~4W8S+6cgpA@wR_YJ2fOQ$T<ZpV<WG(Ks1GED1-_~D_IR`ZxL
z#_gPRXSu!NmB!)Zw5=qRL8FvMBKV~A<W~y?ZWjfG`&!y(P3ab}_n*%^Nz-pknlEmA
z-ym?v!@N2+YE6~0YgUwW>@c;yeq!?eK|7BWe+hVD&YVED&QwF#r_y&nMPp0jpbm6-
zC5y~|i3@lZ@V})jTT|ikWP`%*=<9FzyjR+)*1actBX3$VlC$7!T#O2Hs`d%g*Lak?
zjdp6Rd9ucH7B^pgCs-)6AIJSm>lQ%xvnqdE%^*2WiR_qqHks+t7_5jR^4sMof?F5)
zj@{rwVSTxO<vTO;=4lA)a8hrvd;K)>1xzG)=7m!tw~c33@mdco6q~}kQ1CRPIjNQx
zzY~Y~7)2#SR(!Iu+q=t8*}Z&Qhif0sh*0ruCPBUwAUxgEnvA*X2l_0&M#D^KMjg52
zFoEvuoF>yZlvtK2K_W$DbQK5i6OlJH<{oCk@R2}^IBw#7I+6RTzTK}DkM`P$hGt|D
zJqk#@$;1<N)iC?H+&&Usac}*e;5NDR5I@}e_FrpX^uIia@W+BVHQ0HYIK@BW-3-qT
zAda_bO0}4+e(m+(nIx$6J}%p`7izoX@5~qyB*ihdiWgg(eqQ?<$O4Mq(-?-Z`ju*x
z%j=H~cA|dRiRyvFq<o<2T}}Q{7-JhbjHXG8BJA;}foz)sCu6c0v|D6pvZck+J=e+)
zJy*)Pt!bw0QG_?|7lxMfQ5{l{dG*N-8OB_BzxFMJvtDaMoyk7$`c=pF*PINDMfOR;
zXKe>S^mE$?knq=GU<&u}Zn`g{an5gD9o+q6hINy+)})jn+M7D5<?;qcKUL7u5|91e
zQ?y&z4`sdOuS;gb9H%$Wl}NG8<jT`?niRwoMg3QnUl@KqwCh6+E7yv6s%&RyZ*Z@*
zk6Sf2YjSMR@=wanQcE^v6Gh1l%18S$m>Cuo#SJBQw4LcrzLH`@Oo{7Osi{}llU`#B
zUEGiMN_|*3K^~Eqv_D=!yRU73c&@p6Uv?~%9vX*Q>rc5BF<NO5Dd5`2;GMn!7?%`N
z*NSfKqjk53_)LAk(CUQfpbu6Gg$HbAhCPy}5;|gY1J(bvAt~o}j7M!m?_8Fm>}W1O
zy(ujiOW`HduFl^y!pCZU5;SLrfXIQbnF;1-|3l(cfyW2dwVlDV58FgqVG2JyBNCA7
z);kJsC0;d{Hx$_f@rDVNXzGk5H=A{`opGi?u|F!0nw=PcNlC{wC_7bHvbImSGYQ2{
zlSYnt3%D^s`qx3rsyi|4@X8*)`JKVu!=SSQ4vBz`2_e}r{`aZ>X_rYdEhL@~hz6`G
zIR#Ukym`6`vwE-gE`>V?E)#N7fHqbDwmN=M=<00k>T148QdSKbUhm%LKdf0Kz4-LX
zk%-0HY4`%(=3UWx5tJQL*en9d-v;jy@aJ}HCHY&g>OcM9ImwJa)qt;()emqcBzWz@
z8m3J{s9((G@4BT|I?yMa@=4#AJVjNm#yhTZYplk{&tBv^wczazz(lrz>!AubE<R@I
z&V><~?kQ#lk9WpPq`;d96YAt@JwCZ8p`Dc;t?Dph4~;}LNvx9<Yy(XZf%2@hc8!t`
zbbjV_T3sjaK6jkefZ+hV$tL|2m9*Nwk&IB8*qK|zBDZ{Jz{UeJ{+utC!bii)Q5;M^
zr|YV8xnL%|ALg*H7#}=WcUaf*{aVt)-h1&bbz|2AKo$CZkXRWgT(}D>%t}Ag@HO(7
zSuyOREl;f%8Z=}j9vQ*&`g%?BMnARb-pW8E-y)f)7so5RtUhngcd@%OBATJn1!Bab
z?Zx%i(xXmlAw#~%2w1X#GGCtw<G2lFl<x^$P!uJ9*UXfz*mv8lF(v6!G0+!oNZnr{
z7)TP}nK(y97!9r<YCXnh!VBu4-@GeW5e4Sw#bcK^_gv3Mt(I9cxI*y}T+#Wj8i=TZ
z9?o{66PpL|Rt8uOWzu)@Rr8ATfP6JO_Tg#BKQZ96Z21_B?<90SS$6rf>pdb;`Sp9V
zgL7Qz=?F7DU^bD?Zobx95u-(;>vtInpYC3)3O3_kDxAqTr54v6YQF+~0NXI8M2PJs
z(BIe{NLs!(M?9dl*t~wG!qwMDTe5Th;Y%DDe8;ok;M1EL%^-isD7RaykBQvuR$I{>
zyv-$yddYjkBtY2bw6Sn1c;(Ve65iar?ylB+8E|z!Mozj}awyY-cKoT^hUh4w{WP=g
z6qS%Q7p%#saJrV(#pxB&<cOXEapQm`h<3QN*0$;fAgjGS7M$AliuRS{xNK(<1^xOh
zBOTE>n=kJxyuZ`TDT$UNbgG)E0_LrnC7G1w$|(pOi2drWd}AdL>braCLf-zK;4BjD
zE3P0poTMvI(?@w0A%SmaXIYbf{K?C=+NsehBim{|)HlW8@xz)k+Ra!Bw>e^yC0dKX
zdR7{XNI#U4=RiYpsaN5A1|N)=d(-6i(-v>Dp^F^){ayoSWYH-eHht;x(p;F&Cg1Oc
zZ-6<^?u=!&z7(fGxr<WV`$n5pnzYN(?eHha=v^?1EcuK{R+oXEA}fuZ)ilOv9r-a_
z;jP#(>E1(bt7__~IbBb|n7@zy0g7PuX{|%9UG~VKdO4|mrfH>s?=Nj?wu5!RtmNS^
z-e4V+wtbBx&tvUDWMO#WKN-NR_uC4|r=Z}n#;MVj!(B<6f4037y>o`QXR16Pd9cp+
zP@Tix4k>(4We9m`*ID9GKeC$ulK;o|U0)!LeQPw1`C)CZ$er~)yA2s_iFcES)meC1
z)}>o`5O-q6r>ikB=|M34-x+|JH|~}Ws8hc~h5y~bZ}FpVq5EXe3%NS%@{sc2pj4SK
z4ws}7eeKoBIjHbV<d`t3LimymNUu~~F9Hd`LBs5zX=8J(tpK4mp35Zxt@GBsV7-Dk
z8NZ(}(jT(6Q6;vAGl%0ad%I!YMwb(mbMf5V-WfRUXP&&aE$<|I^r?Hno3T40#sG4i
zq!E-!?30vy6iLl_-Y71-)p^$Ijh{{D1!ssLhvUdAcO>BB)q5Nq5-~496=)C_NO)cY
zwbJdF#%>Tf5cY2`g|4h8g6TXow(CDYXU4V(M+b_}>zVG}l%SVv=vbKaXfI^%_AzKb
zhMWc0hqx$lI(!YGPt?TeS|(Twl=9=deeKoh*8chTnB}`QR&OK?>(va2x6b(RtIl||
z(?d5NdAe@d<POE)nTz8(*L2(>O0s*;nS)$qzj{^qMg7V!9d7nq;TmfpWb__Rk-~c6
z5~MbSkn$Gu`@1lGUoyx0r}(We)y1?LL09e<#B67=<KZ|60VU8fE3t6XsAdoXW)!A>
z_WK9(DL1mpTztHlnXxpcBJ<@@@;;Zg@X+}r?3g%RV`!9_?h10ngm5^6n}AZ5sj@!-
zfl4>PtpMZ)QNjeWB?CD~GP^j2iC`LPID_(|_E_*=O`OF3>@icU7$kB1wbxW(HrVF3
zFO?>DxzA=t?K`X?Z)QgH)RJ(KQ(brM4c|LtQZb%n`aEaIW~l+~uaKigA-W--oiUPA
zi;sH%vSD-l{xTWx1j1$*Oy=I(ogdT@`|-wY=*-rqanro4o>MWB2>(g(zS;bDkrhHX
zxbBB9f;};T`0QoVJ(rhTI);v&bb8K8gFnASG-0=GzX<eew62pI-y_CuWACEgv$v@&
z@Ch&Q$<cb9@b?C0hpmjJ9^acY_4ZtFsGU?CXr$FXI;9vW#tJ4P@gR~*=<O~&K%SKO
z#r%G09kKfX<m&7`;S{^<uZx=iz8?NSTg+t24m`r)Fx$^NFTLH~y6>+mfJqotgOR46
zDpa(0pFmEu*$>>#Wmq*5_t`xoUKJipYIZWp3ER8>x(7#OJ`r0d(oO+6IT1?EzM7Wp
ze0%3QmS?HF(2a5VXrv?Tl5L>y0~z*5)lqM;XSNah#})H^4X09|b^k7Yi{Mb3R|Rj@
zo)+>3h)w^N0j_uBa5Z0`%9h^}CF3}c%%$M>SJk?fl;M0dsJ!0aY_=-Obxn)Aey;Au
zi+tW*qaRgfBBBz7M$W@X^kVf7o(?8s!z=0s1xN4U$v(F}9Emo625tNocDW@H`MVdV
zwYww^{`?<*Xg1VFW~fblI1=G#-8B0`a|u@@M$t1L>3fx@NG6_t6-^X0uVW|of_L<?
z6t%r+;aFRMZ3pEp>?N($LU|C@vp$GZ0;S%uFPRPua$?twUla#tc_12^m)_9Xu~jH=
z&%BdccxFGzjY?E2c#hW4+mG;Wb^7)}G0%u2&n|_JL-G)``Nh7w+C9p5!T;p#p|=^8
z8}+_79{J8&SmiB?nOAKxGi~FD5BWl1ZcVrnL4wiDpavGl=~#~nPb-eTzK}xHo!cmU
zC0ju!G~A+ARq3i=c0$lf@#X#DE^+~(Ei!HXkv`5mU>_{h<RDU<)z>jL$6SnE`swuF
zdVDAFF7y=LCGw%2q9Dh9A)%it4usJuM25i+gL7~CEC?bz^<LCj-5u*}_eBTzMn#nI
z)woe5E}AR0tQ!WVsepOgSw?P3&rRzVJh+<SmL@f%sgmE}8StL1s<r&hS$0oRZ3?pw
zz0L=YhIEufK`R9_+3F@nrA?(%!9Qq@4pskT-I_-|qFl_D9&GSLUw2<NIEp}Nl-8jI
z>&9*r%PhOCUN{SyYr6WGnoA$ih3(<xwM349x%h~gbwPopcg60SNknb(ls&%%Naje#
zJJ9;!XuhA8Rr49)xd!9}mrms8!PRq|b-^p%o-x^op_3M_hV5I4^!6Q5Ou4L*2jnva
z8mrPH1CqD;3@?7HS3h4XIASX}iZa~3J;nUnAzLYx0+IFA38|m7WY^l@Qn!*^+IKc;
zB(_JJwtk$~r^{RPovG3NG$e;|Nswdo7g<sR=TDl_-S3Gauf<f*FD9)~Z03;VyHiOO
zysx!(UcQx(TULkTkN>mn7)IStvW+z)(=VonPi!`4+~&XL&;lOH4-x+0tiFfZUkY>m
z|H;NTK{k%)e1*B{hbx3T&Hb8D>_bbv{c{&iX7MV#d^~AkUvp_P*Lg!xmejT8;8hC2
zddEmh+T&4*IGyQ557<mK%I+8Tq(-8hso4IHl<Xwae!kX5H5Vt#1-E_KKH<8ir!ZG6
zNeE(sF7k6W5Sad@s=pe3h2*&lRZ5uy|7Z#~_nXsO)+q)PknY)f-PQgoINDV$2B|7c
zKk-CC7`N!ZPduOA3c8NHT$_kT`Il}=Q7MRAnh|Dho&uw%jv-h#YQht30Wkfx&!+FI
z&)jOtH&0N0DF#{F>ZFsGY%praLLtg*Y@EA?Ld4@<ZT@@+gS1E`;utfbXE$V!Xg5SR
zmQ+VURH}i&fVaLM-Amh!H0z7PRyq;0#h}KY{o&&(h4Mo-hlXIMDYDJ<Eyr+ck}>B%
zb7SoeQ@Ml4`&bXnMY%BNR%;-;tA}{ENl${NYDyhbzgLor@Epi>1+^&~bJ8AFzP^%`
z?vdL10LCDepox#gc!=V3rE#q>aAa-H49P)PAr@xhdE#4{H?jJ-H#A+!y{qoyfXb7Z
z$KtNeezgHdpNYm$9n%!gLU7Bc9u93`Zi)9IKZ^OUshc4UVdx#5=lnG%Z|G^93O{7U
zP_}Qu?C>1ve4b93`zF<~gX+cK7gw3*GQutqb6Jv)#bfs2W^FAZiyK}gR(Ev{1pnls
zmZCFWmcb4m(*s5U<o+&7N4<7iXO9miUuXJ0rn%|D33>}9q2;2z8Mdzb&-ImnBfme?
znsKfFC2nG19pyO4KI^^rT8ZMB>@3AEI00=jg<Eb)Irct`<lKVotRLqiII6BELUN!l
zc@-55bcc|{F$Q*}?jlrq8IRFz$mj(ffq?bzoGmYEfLu{%t{0(wPO5z_pO3@F3p><l
zmXvD>-xxR-G=I2U4%(}t-p_yP*zm(p5+MI;2(4ZcXRCqed+rvpFyz_yH?WzTH7PsP
zDCGTpkIgKARG$GvX5ug@^3yvI99_(&!3{RUXk}LH3>bp6i}}qd3T!lgG^j%+9@j?i
zmCq^6!5v>#A%bc1yKBMp74IL@J48QK)CZAy1^?TOvNNY<I_A62k((SWa(~bCcQUW}
z9Wws8Jj(rzKq!y)yF|Inw<4dn(K?%Ge+HZnh$Ji367KK1@C}jVXhn!nvW9`<?Y~Yi
ziddXvij>)6siL~ocda7wl@qJXyY2U`FG|72I0=)FaOhS7P3}{#P?TUOI$#=eSTkwK
z4<0S*Z=H+Kc%~2CK+!bGdA(^s>oJLO>PaC><))iC0iH>B1RkvQ{Ega;-RFGWNe8`m
z@Fh?*l_*r;dI&yt5to1ykI6^_>%A$@wKSQgVnJkg2N@a3g5FUTVv-+hpZ1snBx`U^
z8V~8^%QbhZEB91DG_0sMKB?$^6xmT77omgYv#}aUE*qEugjsejN)^uVm4ls{`J3_T
z9YbFS$J3~!UDlKgaMcNpA2{h{fwh9yVDyGTt>d8O_4FO<qBx98Wj4J%Q{bv_=+22*
zHRbpd%Fbr?0QZ@glMtbkH{VPdV$Fwn<A(g+aJ%2_vxJdjf3!m|guJK;^)GW+QA!aY
z6*EEZhB8J&##B+h{Gp*t63lBbuZ`S`e6Dl7s~=K%!k(@6_30w(<r5tXyK+v7P8vXg
z_fQyCUzSB7{!E^Oq!ZLCc(-lXi1ZMQKFP@+n}5Klq&Z-7e2|Wio77WG6MEG-H}!l^
zS7d~n!Bx2-HAb=kf`w$1^V0e{&DI9>EIVYnA(OJ!e&b98DBkr>UgE{MiYH~?=s!=Z
zn9n$_Nx5Xjk=>3C`f1jfKU#L8@cT8<jqo-&Gegg-sKCIJM{~=RO@_~FOtgzx_IQG5
zJN>H8(AY|6?Rw58X2k=h#Bb~I3VSdB)k&t==qiofj~uk>q0spxjG#T()w9M=R|tA;
zk$={|1G*m-_lix&KXDI<GI(ZD(wq2d#5sj(>q#)TXwdCer@8$127A%mY}Le&wh?gl
z#v+SWarW+fu17}5TW)&lGXlNr?Sys_#wDsKiK9%FG+_G)peUn=24~U_iEBR>jp8e&
zikOnU4gYJ0p$s*5Yeu3`+XcI;Cx34+Phah%g1IPo<($Hlo?GV|%e)}JGz#;VOt8bC
zBNjQ~(s!k`%~l!W+F}|B`|PLea)>MG+KX{!b09tPYPei~o{{!|UF0kaK9+CU&#PD7
z%l!U3b;PpKyE?AF%0{=wNG+d5@;4jg^Y{O<UzzpJgkWj@DGwRpjq>7S;b~gkFZpvU
zC0?Oc!NGCq*=SMdzQRcsTvn^Db`fwTg8Ui!r0s@eSsPhQr>)~_E{On6GVDcHM}r~m
zM23PjXcSOYdQ9?mQ;{Tk^uj{kejrg5qBqx{t*F(yarLK}M&Joa^rA*pV30MCST{rE
z7xkI-o;%_2;x1AfN5+8JggnIU<UE*>oz*~yhlhnb1j0kC14GZ^sq%<PZY+0U_EL(T
z_9LEfZEWV_evVKRIOj+B3}0LypA^!4X7w^=$e)&hkh!Idti|v7oFh?y281{zv>pJE
zQF;8#9=!z1jK<M9OA+F@=w3WMYCjC9mZnWBkw#1~IKG!lxx9Pcahg?4CkpRp$(>+|
zW4CJM31YRx?*X?H(|YbUpZzGU4_yJh;tq`<y^H{7#%@`D9W2GGzx=JmhhH!vo(U_Q
zHGzvumD4lQGCoQ<VROm@;ILy@cJ2`LR6sWad78|Zo)J8~+Z>59eRx-)df8z*Pf@;r
zDC#Jc=sj5RJ~RNsBP~lka_pu~!iZ6`44+C+re^!zuV;rRpcI<)?7tjVoUUa^Zj+D3
zh-6q(P~|nTR@@V-q)nBviDrUw81kmbiHBXIMd2~rAHn&U=AgZWrufLdqjr$fM}|ZL
z3kd0y69@6_&aBW3F!}zhYyiiV*He6#m3H|y?p~svS=ctv39oA~6Z<YmlFgQl=;`d{
zzQil+3J}unosPqYCvoc$LvM42#nljRY5sy1M6;cugyn=|Oih&<<UBrwY{U;skoOy+
z&zdULU~i6+>obK?<S!NV^GX$GjDdlK>ruBE(P0mmVNZplzfFElSSC-}eYvd8aY{P9
zU$7C!V6S^ahMWf-a4-+N5;B*HUwT_Eg2;{a8jJdONkP`?i%62HIt}2vbKY^ln4BuQ
z*K)PJq81}h>d~#_3_XwN2i^o6HV(zdl&kH~GrZwAM43)7&&@4d<)g&6H3f$_Lxr0P
zZy!dADW;g|t8%Olm#(#!T{M{3+=Z@xV_7rG(`yP<_~9i8Kb-)!4rw2$v=&imw76fH
zTU+inw=QTb!xar*n)i?#&m#~jvPzUK?3^ovCEqs^Xi-ngV?rnbmkvnwrKKj8B(bGr
z>y;rRc`H*D-r54iAO|)>+vnNKQj>J!mQkhJH!O~{=ur%fb5<sa-G|Y(X4afv24-0`
zEi#f6U6M$i>Btd`9TYDFNhQ>OR9;XKDyZ*ybDUII+Eo8OG}VFk=i>}(uy59nYDsrq
zhD%(-%Hcrz-*FXM;yrDY^<6SbI+mH6k0|-OK201`;F;Z1Jv*;GHLSI}8u;Z&Q-KDM
zK5`;peO^iKP}z1pfwfs|g$s|hOzP*#k?wp`Z50oTp8V#Z%_}e`)JEO@8M=~F)0GC1
z33rN?)HVY3>mME;IMy(EHChUcYD=s6bcYC~$w9jVa{bj_QlBHzjYmG+LWDkIiR>>_
zc|=g+=K7JxD2d+2>&~&?-_InC<USrPaDDN%KBJsxE={k_51z0eeash_*ug6@b#Ko7
zbUo^bd$*v2Q7-+l0E<GZ*XO1_e(HhqTi7>Ozw6HYZ7jeg^q&XC$*ld*D^abr>nHD=
zW*ob3tZnK;q<Q$k)Rqv*rZ-t6eajwVWHnJuC#Kl^#^g1J7mHgqZQ85rQ1jMUa)Od}
zzFXY%kZ1@ANszJ-`L$ao+DwGX6MKu@A0S8t)<-O9kFwVxce^2X$DtUhIB$B1t*7OJ
z1czxk75`Ob<u^Q&VQ`WWEP@*+tpwUEROmUdgCFroi^Ot!P(DiTS->;3aJo$(PB_8u
zkl`NMsAq?XTnso1O(&vury8FI$~k&5H>l)tF(M@`Mh7#VyK%Q+3k)u=wJTYnm!{I!
zUqFm|2r$Nz<Qdrqoj$3I(GK^b19OvUlzovi6sm1Qfw~veH77&l?T3bcslC6H9lBg#
z3gbv7+P3TWoq1Jyh7x!?k{#fg2`C+J)8b>oB5Xd7@ScdUPE}FnL&YII)d-;GHR8w!
z3_D<IUV~NQGoIchhQ@O(GvPjXIOnmnr!hz%dK{-lu`K1Gy;`u0e-Uw2l_R(F9>d{%
z1_*NiE6Il&I@x7YDW#+4dIgE<xOYNX;%k)?%447pcA?LmQy&CdQ!16@ZMVgH%htU>
z#VDG0+~tsYtmmnk7wD(GiT3s`td#Ub6l}M@QG7ln!i=<(fnHa)7?F>YJdwFjZqRo5
zzQPk@=CbLC`~K4w+tRZ2QAB*rNqhx$5h%ELit=TBp_lAJgWh|W++bxu!v4g=aT3T+
zwI5D2?QKWE89)RN^GcT7<i$mw1(-u)dHB_J#w21bgkvaTTell?eDY|rR9~2+_yjHu
z55wp);#EP`k(bW>xnRc5>%+kGPB48}O>O6E23L+t3JOv@zlibu1~4cldmctsq7yUx
z(L_J|47-2uyEbqQzBKXAjEYe_Icwq=>a-ZnlaH02B|A?!QM@ZlnUk8~*hOh1;T?Kh
zp`xy)g;5L=wnBCl64?4^u&K{FLmfPJTW&GLKPX#}xz|UQouYcc#xSb25>WX;Vj-rH
z_u*!r*5wM2*A#GI6r4fs5BqM}(yT<Fxef`tJuIY6IvR*QKTUKm1>i)FB7l5Q(SGid
z!aNh5&Hc5i^SCUTyTA@~u{_)wd>AEaLwBh<h<`e|s6!s>M-=ipL?1z!mwy6>A;0n5
z)+L%E)@}c`E2m>PZTiRyB+c93-(Iim^umxic$=w<3io-bw5uEY-yB1|z)~^7h!@wl
zh;ipE%S{ZAh6#pBMxsA%YeLCnKkCsjantK%N6CLvfu&1GE@>zaCe{6QXRmxBwgB?o
z1f=Z>h2tka-$c}Gssot4y)%|lyjS3E3nG*SA+Ym)SNyJf5JoYd6>ZY^x?Fz_I2ik<
z%G}N^I%~Bl%|E>>g(%s_h{|Dvg9mwi)m9m47P?^yeVB$DG3aYwoq5lWsd0nh=;!A^
z=Te2yx%6m8^7G-fpWnOW*i@NgT(?Ks4c1Ct-j=A$nJW?`$C-9BdfS+iJ!n(&%fc|D
ziEmf+AZfDsboIS81;tXx-OC1{aR9?TNU_XVe4?+t6vv?UAvr!GZujdMd$4JleK(dL
zQAtnOKjz~Db8HESHCH%q#=0QPtAvas;~rne1P^&$idy44qNDM_PHOV;2@qn45o5Dw
zbcLSX%?FJ{q3@a-likgfhSrGA7RqkieK^o)o+@cLeOwwo5Jb3&O|j2yG9Ws<N}{(K
zPqwNPJs|pZmE={hq^B+u;hy7t!Btq)SR+U9C`{uk%eR4M=3=9z#YkxSlz*xFy0NFi
zcq0kz#X+mpKGxjc+Uq;}SZve4*>FQrcpQH%9=RHyca+q8(P?P&9OGIv^=F})z4Pvr
z&t)~6(fR%Zqt{X*Z!%^MUL`FZ`krJ~oyz*wj5_T3n>zR3TTQ(66}CQ5?3GOP@_n;j
z@Wp1l61}*Obk_7`88ekcxury`y~HQlU-TDtMy!FeEc#ZbCCN1CMKebENFKchP9Lhc
zZLPSpSiclK5z`3DdSIeL^?|U?Watxt1J~g7bMzZ434Kf8)#0Pv7C@8>l1Bvdsh(PO
z#pQX^{V*(hAu(+Htogx`+UN17=`-(O#i?ia;Gf7BFD2@1?XGbqeDN3vDqLOEbjDH=
zBdzY3EOwm$DeOx%@xui55DhY)r|sTS?cQY*-uU9qKx9v{ZS-peG+tUS(Q>(`fo2}n
zM9^ddlLx=8xW7}Ls9dfG;K=)^6N(<|a^^HGm)BjnR%%tE7tQuNFn(`gvwWFKEPp@r
z>LKeP;Yx2OB|X=6LlDb8lWm<?exb%O`PC7-b4{>l?NQja6-0MZ2#z98;_zPL`8f1=
zOAUIP@sx4CQXQD)G7}&d3h_!6Nk5;xW%8+zZC7Grc;)E(j!6T!zSo$4n5lU3aypdd
z5ZjVKN$(r(a8f4W*}djEDbBeuIPfW~i3e~WelVcF;e%D9W6tF;1RJq9?hRRH=@YoD
zF>CH@UM2FmAyf~q=0etjyHsRlkLZ*SIopq$z@(+J8qilY{~;x!PWmnk=(7GcuGmqa
zkUhI34{NYrN=!@_6J<1))YA5QivE?W@rvzS94TJgC*qZE_(i4`B{>BJZwj?WLY&1}
zw1!Y7LQMcivhA|mSz3Z4t&QOCTKWUQ-Uc|AoVfZAU{&E+RGmbI1wsYpur}~U^Rj&M
z((>kS$ZU#|_azhbT^(nL&^okJ23z|wFrwa1g5pw#JUG}T_xG@6QbBWS!946O6TDY4
zmA|Hz28GqUmTVsVl<Oj=p^T|Cl1i8_lk*5^MzeTt_u6tsV-DgcZTt#R(1n8WR-ko_
z8gcurtpS}g`2^dkvULx4%t*#|wZG<PK#z=Kt<hO#6Bx^BI|Z)fMNZYz?(+HROW!s>
zYeCqsPMwrw<X7*N_@MegiAPlNrZfIXx-P-&92#TuD1pVHdj=>zpXXg4KOOfLe$QDZ
zCO-~z7o4-CTw4$?NssW%{^9ljp7FX2IVbD$Rg`OYEHJ`hRB}c9J^I$AZ^k{Ys3)*6
zg#x}QE3Yd|+Ucy<W|mtEc3y6}?`xeFfI^C*)(vhY{aqAYjosZ%tGQ+hwddRM61XdU
zoyec!DUKF~jvVE)X#KsSz0z6U;DXa|lY_TMA!U^YK$LJ@w`z!{{hOVbM$$eA`pB0x
zk}Y4@mek=GabRsT=nMvCJ;@M<d+pK4z!D{z?(RXBMjK;zb(CRA?)Yc-8wG!UQt&K&
zyiu_HmDKe`gRA5YmrZg_3-61rbCRW7Lcc><2SSGJ=YSQ`dY{malY2m)mt~c<Jn$TF
z)V0$;iLj6}pbwkp3QAF?6#tJz4-40D!-cQah;)KCOJ1RV?jFS=CGLY)PNAlYQufGa
z7cptGi*<|k%pO}oOve4s>Ci}x$7?BTsF36jnx=jvX*=tEZM_5tW38?a?kSI1&eZKA
zSmViYa6gLdEc8S$Z75$r1KNm9&BPM}^C!JqtCs7slVg{L<C+)lo;q@#hDo+zZ+S)U
z?T63?rdhP6ojWYQT@T-Q)D(FtX{=ezskBwGH5I0?$&(ONI9$HCujt-Kx}lUtd>Tuh
zjKeWT8+9T7eMJd|m&HONZr;Aq&*quQYR9a4iqE?EB(>kLsHP|B;%BJj=^NV|6!E1s
zi<*PIuhdv^qTKGYDS!a_Zq7T1%<2?LKf|6I7BHhMH6<ALPI<BC*UJH$n)TjTxX4vJ
zG8cKcq|Xokl6I7rVo~FC0~^xz=uTX>nJC4z70)I(y6xVan#4~p6u;S~JX{Rcv{w^J
zZVI}{Hi7`<?zY_e8DA1nSv5S=^ZE#z=4-86zvr<1NMKoPdZ_o^Tsh?2+{F#0GvdQd
zKaVDk=9ZgBp1MWWeI;2Lp#b8WcJMf@W*T2c-%yCGU7j*X<uliqPeQL=hTv#$XT`Ts
z=kSg1qld-mLj+j+$7Iv77j<6eAMkwSgkotoEHt_{#=rL~oV?9W|Fo%~-0kJqPs^QH
z?)?lUA)RX3VE#LLA1TDa3ewxTQm^rjrRC46hv#adI}e=i$NW?W<C8&=Is9ZI+D;L#
zkHNsO;Sw0P53!Au`@P1s^R3s37zgQXC-ao!-M*I|WFX%=*P30X@3M5ccY9MTO~|12
zkm4K&u1*O!=aDT$)b!)rw$^;;+VGK_LVC=O8``j`WPtx`bKZelvm_W1Gu^9OMmINn
zd*4CdXw0gC?DChTQd5wqY4{ROVAO-~OQUM>I*L`D>EQ`Gh2IezLeUw;2TA7oxi@2w
zbkWQ97xv3%8kmynL}JZIslN&s9vAAYwg$NhFQnA9QP_92Ym|i5I28SGckYw@;eC5F
zzBZ1?{50Ivl+e1sw#lw@i2cxxkmL|&k&I%A*7o*F_@#H6kxjc<T{!x|RN@b;&Mi|#
zG&5EC<!csOxtz0tCru25KE;<|h6cs%5n|J)Drkn{lR^=RZ>nM<>Vwl?b>m>7m;S09
z0Tqw*nVL1nJDXVGRqCgoZpC<T4=E4lg`%rItNgHt0m+<iY*YC!7Q^zv4|93j%iS{(
zLt#T|<+d#Ls2OmT<Cy;1k)?M`ejc(se}bm2yS+=lyFZY;1_mv6Fv2OByf5Nt0^_`;
z(qFz%<?3RxdzI4^K18!^04h+gZa%&Ek5#;j-Cq~xBjBwPq#Z4~+vr>*VL(LUuZ2@q
z+n(gFM<)&_OH;QxK3-Qs9I$ly*&xDdxbbZis9m+nw(I)X3lEY^?x+4;&Z55eR1GrG
zs?fLPD|JPw_#keb+a!!tmG2-R^L{A9)1(%Y&ExT)Op?F<Yx6X}sf_YyLn6|@Tc-XJ
za#GSQ5^t%lNB*OpYLQ}UNqWT5>;i)S-TsP-)NjD_D)Ka^(Bp;!wjMSZwc2l8));PR
zje~pU#{qj_cJ^Bg_71%quXa-@?lUPi>}KCT*<hew-&HtH*_#2U<-g2bR}+nL%o)DT
zLkJIi5eOxsAYF`K8kYd8N-}^c_qN)fE-otz9leP#IrZN~aZG2|{|OGT($O|xvQl4@
zcXyqqlCdO#L)s=M>Y<e-e(gJjDPsR}<ty<Adgxa`foi4ZyHhp34US#`W*dtXI2p!~
zVmoD-SDZGkDvXs_C%T$!-o__h+K39r;6<`%GL41SGXrcse|-qXUc}J)08Eh0`nRkq
zc#8gDz>0AX<8oR~jKtJxs`H=(4Z$_Fzl`3Q@i=<Mn8B(Q;5A0&V`!+Snt_&zQ_Kmx
z+Zw>Z34|+&QsSP@*5G4g%cjEY99Ju55>g=t7T8y-=I1j;WlaIA0_xu?CABkT@)+K!
z{>MP>#SFfJp-7jYf1ZjTq*B=rjwl6<fhiwQcr=Axa(p=$!8rK)_;jKo-Ofa?iW<r<
zb^&8Wn}EIB0W$G1*({vSj4`HPfO!vqPbS&&Qq5rANVy$pibV`{sa;wF;gaEv3QKcn
z!+*5!_@tMXe}G3uPvV`B3V<HammYN2vJL#bZY&|+9$o;gN*O>~F;tSA+?ke|3IitD
zp)P9onFmDi+?Q{z&}QuM`g1^z`<FE{32<<Mei5p2`S&_oH^=#oJ}Adgcle`)&rE_m
zhD<0e9D-`wpOxsZ;ulsFBXRbLOp*CvxlL<FBD1iyUgfA7#N_KO0(*G$h1d@*&9`YI
z<=9w7^uV-wE`TMAK6oClGFDnV4=rtMaeR%>To%Us?h~w=EkgoB0RgZ5Nj1|LVFuJ(
zHS{vUS_nVyzt)q4v<cnu`8mYwn#|u%G7#*GTdX4Y{`4*bV*LXB0(gdR!~c1If4|Fr
z`KQ0X4A@rw@rmJK|LXz%e}F8t(A}e>U!YTVk<<UPApYYa_*eg6vH*>RIxKw1HDmbm
zf&PAs|MpKF2>xS;5QM}?{r?|~h(ZYG9uAd|Wq$JihY$VF6%za}gwlY|aI;NTH{m}K
z^Iv^#sXyNq?oBKZaaEtOGymI9{69Hx|J%p!?=QCiggljCUw>zC1we@j5c7=Bi>d$J
z67ly}!K{h2rtq}?S`#gFbPfOpGzr|iJ3!J^lb7K;)fGu;4*blsbfI<GtpD!$T>+o>
zEz$q@jxsYyPJo$w2lU~p@0T`?9DxDPuK-!e=4>NxcdcAw?_Ykk@0dU+`uxG=zu&bV
z<iHUdV@9+E0rM925z<9t?|=OAe;&mD{V(YRz#G*hd5icT&*hsEST5U4dtQHm%Kl%M
z%mE*GkypZ42OfXE`A^ad0FzJw9*kZhy!Cx?^;F-BG}p0rNwhu(acRK$dj#MW+*Ehn
zXB=&Td9*amVhtP(Ri6h%`kO9LN`N*2hM02JR01@_7%=%ie6?NP{dyh@ri{UmFRDkL
z+;^=LL@%6Sg5Sk58lNTg+o0rc@4Et42^d-He`7Q%*ghsR%;~tF1NnQ1M-+|tlriGc
z^}SU=zD-HpY6Xxv_UHg$1S5@j9T?a&f<CHgo&*>>(zb`E%kk4d8*Dp&`ODoFJf~`4
zUO&c2b0;fDRSh<^fA9c$&+l?T{FnhJ-%SiBFDlpIXG0G|#00o*kS96>ZN$>cJjobp
zE%0pwET1FbNEhJo04G`7NO`=<!<4VWYpL|VFy-|qVAakY`hBt#b2SJ2mgb)q;CQmY
zKn}}SX1l1P{$T;kz>|!~7Q4{5lsHTqcB%~?=dDkICZHu4#|^u%1V>F3sMT<QAqaaw
z?OKUJ_aq`P@WW}}^PMOj9oOCF-%V+NhGT@1SP5v<b3t4z(eBi(r592ggAq-(;2Gs#
zMvEr!kHoLIWB7*WU`N?t&Q8G3_!fMV`c7?^b;LC2rp?MnQ|XH;O8}AE?e<*-8qW@{
zy5(Dp&Fx{{r(bq`O5JRILr`9Tz7yPd33@LR%KTqR`^bDbff#Ho61tb4mwtU^AIKMh
zobMIqa;xxaa{I#>hxVrurWp3%8f~G^zud$Us+fE!UFv%JpH2LL!LJPBI13h-J^22P
zw}DCR^Gc#Yi5r2LP@pFuKIua_>E&PTe-*S|n|wqVD0O{!0<fU$=fR^7V8ws*6SRTI
z1GPX$*?!^|&_6ud`_Xyzw_^`|2#zTQ15;pdXa~SQA|?ScC5?_lx-4+Q(*GJ+Sz$?i
z9~eGZpH2Q#P`WUHNJ2=A)^xFrfSanMa~|_G2eaA&s!2KK8D(Ye0V{bmpb@Zy-X;D`
zpUCj*1PMSA6QF!<9hx;Ler#v}8gKQ1OmY;EU0@Wp1IO@7XdP$eH)(pB&~B(bL$It(
z(3Pl@ggUUa0L7phv;n6>XnkLX(h(OT4co4S*aKV_{cfxq{AL_Yn^G;k2V+cJ^o)`O
zojpAiC<)O3QN$VWiWC_q-yT9ll>I0$3$iC9+0XI$JB>Tok|8{eKx0oIgTXdFx`JtG
zj+I!t<6*HcHnQ%$2RnEN^!CeC9!`FEpVYA1)G+JR&zWTd)sc1mpfaZ(?0YtvF;!eM
zX9dVkzO45hg7~qUK)S=XTnAcQ)mnz0V8q$fZCyj+#(PZ=79ybqzm`z%(3}g5axJR}
za1ffW!v#}M&gQ{%URu;`#~g144yNW?3L_yE{KK*&0)}@^Fs%lvPuu{yA<t;fu)18A
zfk1mgy)=H=0n`sDkZ}LIK>e*7oU9YwB#&GFh!*|jaiEr_!vtHdl4rv)7_j7ZvfV{-
zy@YeH+bMAz^;*sWU-NE<1X9cZN7XQZ2jAKPAV%^*erx<lHYn&}pnT|A*=pD4GE{Ia
z*DjRY%#-N8mEHrHZ!RE;1#dxB=oe&gf*w0y2wL%K62gdAwzNnRgycyp5dfB)r_j$y
zCkIJwfF%Uc83^sRDl8@|sxVC}S?8)>&)%a7v<#hrVlCpeN#oc;QZd4QoZn^gA%|6s
zO2oJqLfd8SHlRnx6KERxi#gfkwaL$2jPS!zrbAJ0Q{T97vzr{|O<eTIY~uzoou%Tl
z*tt+B;<O(&m1b8(6vu-7HJkA(6y18m+STxPh~!0Uw8-WshiyR3x=FTP#u|F%_RkrC
zaYId#mkiJE`Ra*-))%$RJ37Mv8J7=uWVF2`%d{w329>sFKZt!hia`K!fxdb}4$Vx|
z|8S3xhkhuQ7GngbtB5f5!?Fx}tJDz$!6U}%YK($HvuGLm!57IEdbs$-QHfGIA?z>I
z5dc|0u*Cv{Fuo^NXT2ubdk}M!5S&j0Cq%DenG$kmT+jT*d6!@laUV;fB;vSXK8Wv;
zm!^DOIx61XP+>BVd+b&@VVFbgiF(ous43E?D{)q7k#7ZgN>={cHKKrjvfT7vb{3X}
zHrg@VK|_e1AT!rYpJA=m@deoYcP>sqWJs%HE$DY`s&sGzheq%|SMa727uuoF;b)@(
zMZY)Y)IjGeq}D`zEVnaKV|6iyt9ywE)+OjnueCac&ri+^=XxHI*EZJqBqd^<kNg9i
z%eZeQ{8`feg+u?#ln9{mM?G{V))SPh?dsG^iP)CbGlKMFy#vo)EFd?6R7T7y9r5L=
z5;-DB2xIB!90jvYF6c_QaEnhV-Ay*y8d_SC-b96L*%>`2D+ww#NF_Y34=r^QA97R*
zp>*sQt$C1Cr`<D9JPx<>=Ry=z?~<|clmCF*!ONyEt0LI9F-oWu1NrU60Pf=V7|}I-
za|cO`B-&#&yKA5`8u`>#DcxNuB!xhqpZv><y?#-0>K9z!`pFDz%43_2q*5ejZReA6
z1`T5=(5&{$$34;ByRWAQf)b2=ITOUE5Io^0*L}gYzU36XSL??!4bV*BWr8c)hO+lp
z8rtgsv5KE@Ku$rVK4@S-F1%CojPqY_a}zn)|4?;7F7}uS$EDwq^7R+KHL~{CwO=_e
z-=+;_L-)Xia)UV`To40AOfle%kZiCalDya(C?<Z<bk8<vsgoAcH&dVwO};%;ms<&y
zm$d`pMZf+;FlCfWnx{k!Y(imV2T8!Efh5~OX#MRV^U(cY;D{yscxZTC66MSIE3lY5
ziVEjVQ-TdND9x#l?X>^NaSjw3M`6~V{rL02$5NvD+?IhPqQPiM^vMVyA{ae<-V?hG
zn3!DRu)|CP1R+BcYpCoMz$?HbPB)k<^o+Rz<t^s+F%%_p_dRK+eE|qdkD=LJ=J092
zNaF1HQLK}gRzeLNuqQ}>+$k}bvm!sg$)F_bx12{219a3ekB5t0$8KY5RQp)2V$S=a
z8EK0|a%Xu-=!3J0Nn~F2|40x2XKw!AKe`O`OT_(^lX&Mg$Di~2Wl#$SZ)=xdJU!Cl
zDH#n;`Z7b#{_RBJL^2vcK^6yk3V2>SZ{h^JcN)aJ+Gzo<OfZ$yScroJpdq&0M2Gz4
zP4(b!wcAi-qT6Kwbq+-bXpdO<%O#CMj+i^t{l)d6Y!~G=*5Yezjb%0*{kUtTU^k@q
z@_pt2RC;Y)%Kk4ER6Ag}!b``LIWXlyfMfU_!i~3x?ZTlvtFBY>6(C=>RaDx!a+SeP
z*FZ&M2&1MeGknm`Bj^M`Z0|5&t*%?O90XsTTE2=T{$^_8&^`yO7+3vw<3VAj!p4c9
z!Wl^7rF{G%Bc(Zd#ZKS%6hnOUIiH@d#^g#fPfYY~ju-2`Gw3Bio^Gfi?j(-EPq!{m
zogC$20A?A;6pa9I&E+q@s)w99_Dm;@YxE{}yv15>b#frS-kRjNI~(5sNJ{OfE{;~i
zQX4r*APsHUv+*EL!aX`B!Bk<sUJEj<vqLeNW>J+r-^-5RvchDm?IW$?Z(C2qZ2#re
z;tUrg^5^m*tfYS)>eH|mDJU?qOH*LRx*zpmJ5pv>$1y5Q-HTV&Xz<<0O?p_vWr3+B
z?daVX1){r*wXsQHSh!=kH1=zt7>bLecT5GCv;gQpF<OAMSsU4&lv)L%d@ta-@*7y@
zzO|Y2T&uGDn8FNc@t&M)x$|hn9&v79tpi!t!D-lwcHFVt8e7AZf%t2#Q{cQoo`I~X
z8_?P~UIB((z+*ldwK}Gf&{-&wSPdu2u4?3_(887dm5Q&$mbm^4AWbI5K{VmGC3T<k
zHPhSJ-5!P#94&iOF2)KHn~pgLT7attzSyQwyreI9ACzXj;JPOQIYO<4pWBs>&Y)gE
zjmd)i43&K8R1>H>8X&l*Doyu%d?+nM0VOG_;vUPtj6MiVbp<i<g#J$n&A%Q8_cau1
z?F)K|6&Rej>yp0MPo-2h;t^B$4iH=s>CA<XKin6Uqd*O{zN5;P<LE?&K(;b|Usyo>
zW`QnrRcZ8ywavPlNcQDjyIRw-Z=n>>2sb^|4s3dzpL_cNKvfAMwRI5$D@R+ZNXYE=
zx1`P+*^fKiK_Hfca=7(jQ;*S-gMZlzc`LCPxgy%(*CboE12}1|ZA^=e7|f_%<3L^(
zcc-<cs&MWO;4209l(2;S#$1PXK*C)5%Cb<p1&sA>*D<iIqI2k+GGVL%%)N<?ybNPq
z2*oawO6XlO>E8DjtHL2vPNC{0P}#FvT}Y>Yy%>csi_rs{&EwBv{6^>x1e>xUhc$fe
zpcabvQy*q@heCaZSb@qE0ruKdwY(e1D{<qf69=Kst2Q+P{h55i@WxilE_7W&_sR}J
zR)i8WnTq6ee-H^_y8EHO%5e0UD`ZZB1Uk!2iDLm3-%sz6tX9a6dhnT<LM8Q8YvQ9<
zK7w3aM-(W&Y(Zu5x(+1d#$B=^K46=O|0#$q+F_r{Es{veJj-f9qVhl;>9TVS!Wc#3
zIvPWg1l!+d$HOsCn|=Y5l}U>3FrWhT#gBnBWLiUs6AmE1%K6Ye5R`8!@t4-jdF2|L
z5JR^D0B+C%%EvIW-AiBT4Ni~ORf>tNiT|<H{3+4>ABtb=)Sy^%bkkj2`%m@d%`GYA
zYCw3I2Ddm9aI<g~Y}UsnL*B-&Y+}cG?l}UkM}q)+#(drnLz{sMiffHW$5oEc=YVQ?
zZCF#R>j>x;t8eS*(TCPyh~;<f8<D6y>vM3KS1*$eaoBoiU6M}?4n7<U@S}42`T~|(
zGyNqDa#3&XuDF-W&Z%Qw0WG8pb48Sq(&@U=3T}Nnw{aZIl^_(w*iGQ3@yMZOG5r1%
zCW#6Rg_}llXxNmayDORiq4Y(WP%mGtUISy4UN$*$9@Jq-`xH`=qt1c&b<Knq`p7n=
zy=@V^S6fUY9fM46Eij&RrLxwBgihFjG6LZW=BGEv@cJ+0ubdUaF4;rL^iegn4=}ed
znxfx;#%=U)P-rQ^lXw%FZ0Is9_cJAz+{b4RJjb#unUG+nvT~t15OFCzF&z2-Qz_-Y
z|2~r(a)NUBFN9O7Uqlfffae+j2{b><>tRP%r_3aPNWMRH%9{jdQ*zxN%YLruXE3Z>
zc*)Qu!}rWmXf?aa27S6lbZ&Rak$%$vCjEeF4efJ0E9e~IYBYHMrgRcqrV>4MnfKv}
z{N$Og!b3zQllIM*eB?f^1XtILsFIApZ%ZKv=Of>TbA2P|79yV|;BJhqn_x3A@xNBV
zlw`<9Q)270F+0Hd6>e9>xu5I!S%KE4(k+^r{K|`Q7{X_7KIzq04C$F`O*8Q45CZMI
zaHSs<EJOl~=1xaCCoy<2n>Kuw0mIak|BJP^j;b<RyTv7xE|HE+C?d5%kd_XWMnF16
zx?39QR1r2vZ$UbxL`u55LAtwp!(H2R&i&4J@A-}U{qY-vF=T8UyWV%L^*nRVXHHS|
z>Eg{0oqC+)7^K(JU885dbD#t733@xpl3(!o!tpnKGKV#*0s$FL@X|8bHN6e^lp9lA
z(MNA04R(wk330sWoe?xzWdPMDQSh~vqt>A$#}~YX#|Gzz{T0`@VC?tZYk4l|)p#my
zvLV-QJ1#noL!2`_il+~!K^MYJ7TNatcQIIS&z<CwCjkr{uLs28TBYEa$9ZG5t!teC
z%pMdg<~}_#%}jO5_oVMdk_X%1%kS_Kn48L)SwpS02IhaAy4z%nm%#P#Ek{j(UNgRF
zuvZlj!oB;FRm(`DC!F;2QE#MsGzB-4Z;f+WZJ_Tm-X;uGHQ#u4v-q~s{|!p~H~jSf
zgwVuPQOhtVa2DmBgK(q2H;$nAJV1Y{B=H4WK*KMhW1bPT4KTE)1;s{)Xwz4E`(y>h
zrEfVjewAQ`EYXt9Cq{$?RFr6$rHtxInLSRnmtapoW{Vh+^tMW_q6#|wjzPBM!6!Tg
zM};+;dBLrGY@0?wxK=(K0o?Iftv}~kG`X5zQCzXzO(Kx=BJ?Q{-U1Wzq(#lprihcL
zWK0xttHvvR(ypfw2G>nv;C0-sY`Q%PX29;>z)kipF@jFgy)^h7tRTv<eS3~JZZemk
zqlzT&Y)kTz%6kNCv!!qIaCRlU8ZLV6f>*Rl@0<e>c1Q`JH#&a{>fT&zzd1pZvjLX4
z9-3Watw!`p)%I-5Sc?sS7fOnFD#<UfXIyQ&;eWRVR2#wR3lO7di%$eljZ@w0L{T+2
zNM(t`r>~{D&@-1QN;?1F*i7+EVQ3n6mOfX_=HqMZ$wZPH0rO+Mj-?aSJ>$BPnWtyX
zk&XeTqGYbYWuPi*xL$#NknUUjJYerUXi;(}>{j#PLp#z!{jxyrmEc1iXU)4v<L=&|
z>V-sLh(osVa*}TD2lcZtrqUi+3R4ObGNvg|WKq_Fqi<Bg;}KN(JAn04KZ?F2>*Cem
z)texTAD=b<7WG~}3BYen$nHnG)=;AE!`y%E3IAR<|KGoxzyRT40;@{>hi3^LDh;Eb
z=2kG3EvDQbhE2)LawN<MhJ5YjF2||@&eu~tcd`1-!j$9hx0MQWpnj`$>JBhUoOQ=t
zxG(paA-Rd3T9SE{6!nAq$46mA#L^)N<`X4G;-M{EEiW;XXS9uKTMmtKXpU*B<Vn$k
zFj2MMk@(+7%>@YEbIi?9;_gj`Kx2WSr)%+hxSxiQ4&zd<F_SU3*nadF1I#G$=Lv5T
zfE9b$)8hW<Kf%T1@F>zT_yK?+p^}x0z_1l_EIzq=jOoS=Jz4u`V*p=I`GXX!=B$!3
zexS1lWBnyyQbu4o21o#h2@@8=Hx@K@QJc&5S7L~iq+&qPog$swcKFkFvUGBq#UNyY
zz=%!GjPYGm^h)LvGB!Dd`r@^YD@-gRuFpj(17VsZ3$#rFRbxn8#)1z@SaNv%zT2^#
zikT1T_l7rUo+xQeKI}4*qmp60?=@sNX$pHbMEaX`Bbu@QHU2l>KyN@?VM^4o7F9P9
z@_ptL2o@c|vgP6mrH(LeE8C}IcHRoi=ySk{5!XewDDirECja*Z{Ev^&AZf%9ly1*|
zc1Pq%=!75X1tshQt1-9;P-#wycFu=>((_ZxJ`7yaZgKMnX~b7ibq;M6xjA1t=f*tR
z=M(Ic7N=-0!PIylbr4xg$N~voSDIm{X&Rz9Ybi1M6KP-C8^S*OdWPX^6D7kLv0F>v
zjBNGAR;Q&_9A{%<$!FS@4{OA<L;jp}U0YA^pIpyR9$aY7exRB8#)&I3NHPrmx|^!Y
zRm@XEOHVttQL^j~R{x2)L(wOL!c@teaJo3JM5U{Q+oUKAdE?ng3v8j5QV-vw1xBcf
z%mlBa>B{%*_d`*&(z8rTpVD5w?&2|8q%K;Atns%&nidk;u9-yd6CA!{jV_rdTfxD}
zSmqM=A0oY_N&x9C?blWMCshurAAj+e^B_?gyL;R?=&5!l^%uGa7XPznLm_%051y_?
zY~|l+6frkc<UEgN-g^)SR1Sgyv)_FxyNVP1pObb2ETCxBdGvkj-#f1N5A6XGJ_TqB
zWgUFjBlk2At(pWHe;cr8RO-3?83hqM*|dnkyZ`Z?`Hv4iC;muSVkT1~@o)7=x-@FP
z=xuEZc(zjj76c3ZV9USY1cl86*fz`o<>Uxb2mVLp%ij}e3d#CkZ98ZG+mQx>4{opK
zZnr^8U#x}Brq8>P89OCKte_@bF!vwd=YM?A29Tj8c;+?I|Muqpy?Xk8{eFp!><nJL
zVy6C&PyhdZZ~)ooPWye=`JaCq|LuRK7b8DGab{o>{r{{1{J#*B|KGoY0S5ouN#C8L
zZBh2wtZvVUM)W0l>!nHbx)b;3e^nj-?F0OOeZz)#z<p(Z=W!AIpH^8VSl5rQ21H8F
zFkByEC~q^BKG~sKlIQg^&dfp9G>!9*{z@4DOmFAZ`wZ3#rm;Sd1sZc8zZOJjBa)la
zik}36m8lqs!gSZqeuph@oJWf{-g3mwF>u;r=Jfq{RvH%<bu$ew(s^5o4KR@C>K9M`
z`30N!H)c;_oR-!Jk)~$KX3}xyD%cQ5aeH7F3$C_<iWF(A90cG&C#39t8dQJpB!h8F
z!I>-|pU?-?AT*w*h9{tFgj6Dl9wSxCdl)p%AA3b_uRoB4EV82_f*S!TW!X;Smb?Sb
zEu@+0S(5$Pn%@n8a(;mOeQx&sRv}>&@IP(?4*aki>PJC(AlHqJH~}<kQ8mkeaI?PZ
z+=Ny|6*2N@ARQ|iOj`iT@nfJ0;IIS%RiPT7ag0AC%JRuIcpWZ{0d@#%^h@m&EzDG(
zaIrLcWXjG|C=sLOf&lQEwm$5<I!&-=FPYkaq4U(wKw}H%?tOLJ)Uo~93eFskzsxc6
zm^yj^Oz*wVHQxh9u@s2>v*+51iBb06QmEe+5I^DNy#7LHZ<zVb_#byF5EAg*M-@3*
zHoqFq8Xs$sKG~P&IgqOC)}N-D_TT2<cM2@5_)W=+z#Tr-=6Q{Sl31`Ho&UHBv;Y*o
z&wnE0Uyu<dNcya~dm7Zw6o>tw$Tz!o*aaNHZYNcW-KR#R3qJ;}yCGpEXapEt!T+5G
z*&PFLXpJt2;i5oA+Rpzg$c4wrquDAN4CB$wAX~Q`VBu}yUa0t;-j4(t=%*xn5cnJ<
zgq(t2dJa&$*F33ba40ewN9C4)MzEYrGD9}|;m5L~mlPZqrl(PMUW}M1$QrWJKBr-Z
zv@-h77l!9~t6?LbuKtg_7n}UT$}%MVa$-_*F>c!+W24FUMUW2l87f+-=)m}952f`^
zs2GINkN@5I{=zB938Gr^0dn+Gkn~j)6QufkBLaPDN5{hJC-nPjBQEjY5GV!mk)qw-
zAJOkU{GQ^fy(IY^@Q2e#MqBcj_v2Qr>!o`N+c$83_mssmXaLuYbY{_iE=$lbg6k4p
zyM`j<oMy5e(SPgvc`FX4f6`fuWm7|O<2!G3_=M>PPLW{q1isqYWxm}+o?M|>cB*G>
zVX5LCJ6gOdW+*wT(QJ;&H_J|B!#)M@1`~jrB_XFIE0&OXXb0mR2PET2P)vUo@H)&o
zj-8Ke;9*SI5@LM_B$o|Vz!cKOUS@V(mkRv`?DY2z9>L@B_+~i0ULr%<s_@dBBQQh7
z7*REG;kMuvj;sM^Y~gJ2dJVWPCetlScl-hdjl3=~9C?a&to+JfW-b5@)Ea3f@1o>6
ziE&8qArl!J<0<g)y%#oW?$*}_MxP<KN}wMl<-F_xd2Ca+5Nk&z#0rymR93fY7Sebq
zEaxy)Rwu+%Oax!s?T^b2P*Lo)e5*O1{CnBRA}{HYm#yL-m(Bj~W!ncR)IISYP*qa2
z;hcY@lGAcz7$Ej{z0#mG*bOV_%Vw|#uu5>83Lx~UkU4S9?PEx;@_arz=4(d;4o7vm
zgEd`^>(xK6SW9v|lOnic3(N-pykdq0#O4ptua9Guc6#~hwFRN~tWUCh4Y%yd<98h9
zZ=v7#fGS|J=SQFUQZ%1!t#zY~f!EC=o%gfqJ2R|`KTdcreTrM5S%e>7aQcUlGt_|2
ze82|~arfk2f@ur~VQT<EnV#p<Fi-xjxi8U(gxyY;Ah#2TXHEjds!!MSQ8^aq*}k#?
z@bx|_(_5BeDLSQjfE-qVI=r1cbPjYMhuKq8YjSVcau`0R0>Q)7p(J00l3{LEd>H@`
z-NZ0a0N`YW#8Hvxo)si?cbgksb^Mt(lQG&uI%w^_M2WJP=Rn?T-=FHnDRm<1@6|eq
zQ=&2^b)9-mtpFCLNrYNCU93ViwxLURd&TPSKLuIuc6sO9%Zh#0$YlTfE*ZHo1<?Jt
z3T8*@jn>Dt!Crq4zGh)8LHe6<6(HM#9s!cOzG&XGxt5S(VpBE5kFw-EFepW+eDSX%
z>hg)dj6`D@bFb(h)yj(!g>6_f0h`cs!x%g@dPA<?ga;tF$M$#L93SX37dL_%kYJq-
zIE!}tF=$HwmOPDDJlfeuxZWaXXePm8Af4`eKSxk`Sk&aXy(ByRe51nj4K4TRGsN?|
zTOkIBl}qdQOFf1!VoIN&Tl3-ND{53->Hc|)>Yd_uiRS%0=z$0kxe!uIC6tUY^g8df
zkf&bwqod?pWH&PI26^NFsG!L1UP*IC-c<W*iXD5@&MOY2nkGQ<f3Njh!u2nN`*m(}
zRXF`MY?|E(E|rPaOYFXeeEOdGf1IYhPnVaJe!B2iDRQ1!S8j}nmak}Nf2S&m=&K)@
z_1UVk_xjXIhS_}({eGY8*VXCOc9?8hIZWSu^qs5kT8b~n)$i|HQvnfubG<RF(E2ug
zkBj_AzuqNzzneCXz^l=9x>9hPG$9bx5~^(Knk&tM832Qw6j8lHGtY}G-zm?%TB~ft
zCH3?KLep?zxBshA+lDX)l`8;4BV>ahx9-~(zqTqpy~9>4wmr8B-^D8OS}cSgrscAu
zlBNyi5$^v}`h#xe3)}~Idunb0t-V!wy$3WPZ$ZDTKI|2~kau<k1P|rllr46p%-3!X
z#`^)v;}X|xaE#)PrJjGr4at&-mrAuR+m+J-?jqAhol5a<TC^F)je~u_@mm3;2Q6(h
zRli2NLHf#sK0tcr&$fT`+alrZiI`j4nvL%zx+d3kNMQtMlOeuLe^w1p!QN#13k3IU
z|HgpeE|~l@E4T|*X*4KfPmZXVR*}-1Y2au<#U^KzN1Oi%nR&+!RGt2R0CF_VKEpR-
zFM0kM>Kl-ip`|Jq{wnSL=96DQ*>eP&Kqr<r1Q4N3N$A*0;4K|Zhrx0oK+v9R%9}hT
zma_)(5KD{q4-h$rN!34knB+J_fttzaCFx-=I77i9QVpJEyu#4~s0i+Le--vc!j>L~
z;LIYBbBF<0)4nc<1wu^&VkWo_=j*i{TLSZgJJ+S;uY3Y<#<nyW&V=SJ6g6$}BLp4x
z@5y}O>QdS~RC%WK5e99zcKijrxFmiW=F)VvF+{K-gU0kqj@$j7d{FUl>wE+wKcSJ!
zP6I3ry@Hl%<+|o&^3{=CK3Vb4?XvF3Vz#5`tXc|mt1*={A`CX-#c?2$G9D2fc>}B)
z@jS|ViUCJEj*QP4;3c*@Wb!xx%MoXV#<$d7np#8KBI_x%hWeX)Y%x&YOe&i-*#HI*
zQ|?P_@tp6iw~psd#jlyzL)o~eulNro#1{SN0xx3@X^&cCe2y8Dd~bGld_-t>GG*%#
zEzqeBfI?NQ?f``HK>=T>^`q!hvww)kLMlUQJC8cYzYGI@WKFLqOcYFyngDF+0*t;)
zd-=d^nf9*1rL98ha4JD{Fb(cKYh=(OVg>4pwZrKmU>cQGk}Lb|o9@tUIwCCE>qFc^
z?KcHa5R_*bVm_>)Wg0iN_dp07=f9$;1<F>{>#apVwc^irklr9JwEqvq(~LzG&C1Oe
z7rYzbVRG;-&nnxi#q9fwnDsYt>@Ij0G1XEbtpVVG%Qy0d6!*g1PGIciZ|Wuu6hDDc
z2&qvK<DNGTp1|{S<g)wNYCUbx%)l(g#nO9CQ5)U5%3yffMspeV^tt&jXx*_Yqe3WT
z2gODRVS0Ft%iTi=nYKr2wWg58?LNK)#1mwO@XhX+$C~$BSXq4Nn(f@Lw)`_^>ni}s
z&GS*)Sn>u6Tz{4AHv>>4NGs{xD{#n$FhAX}+NXaz`xJ=32;}}i%*~P+(^A#Xu2fSx
zJ&C=2y0E;3;#IN@!sG1QJ{wBAgZ0qNF<A7q_%KzS{hNB%U-R&AQS~LFW>r%m!pp>#
zv@@1Aot!I=IV0S-1BNzCa#3p;%mjw#Q#QVruP$;9Y2*7t5EWjT+d5g~2Ei72^(rFX
zQ<F+uuA{aWp$j(I?8pjQQYZ)#Svt^QrW=2@tny7J=s>KkNOQ;qXfpAZpa&kN)3b^(
z*W?G_%%^SpnZU<fliF5%>GUTpF9rsW)YwQ{8D^7CNjy`!@{R$hnm1J80mHxO>2NVH
z3fwO$)aE+-9_qO>I&Qstu~uovUXi0Gut=F?F_vhZ>akXG>wS@hK+Iz7bl;ku{_u7y
z+i|;cV6T;bzgOPczXOR8HNnJgXt>$ry_GncOw7Kf54BxQb>m0^WxQAV@^jDKjkO&r
zVg+-Zy^^4v%Drekrlr$~g*(1&lBd6~AQPXv*^J2aUEb|bY*UxF6S=PSx`mj4VEf7)
zmw}?nBYh#~pYOUM+vX-R4dy`SWZ-#l@y7a)K6d(;+nG!pFVQkIs(RNt$@qnvxDpuw
zqMh8}o%@f9kjbk(?|Fut1DjqKjf&Vlks^1kjlFs%^tWZnWR?9QS8TYM+=i`h+X0(9
z^ynJ5zyOr5e_EdDaab<3BfBA_G^}6dIz$sMV-i4psz*^|j7=SPpbN=>E^2y95&PDo
zU5*De99hwccLYFyn*DDNCf*a^w+9$<ytTu#oswpJqZ0hzIA|`?8Pl|$+eV=%4s0mz
zf#*y}*itPd_jE~ap294d2t6v4f~Z6uDBXk4BmHAQd?Z`oC#<ZeZos1AuJ~g#^H7-=
zO4VGA`yiApXr=Y$yj|3F>2Kapk^O3r7ufJ|jOB)&E2{oB0(%YC4SkLTbFk(*u!AZE
zX+wpi)Vm?X&~}6Odj9j+x8&5)h7uEmz8`N)8EU@DlDKL*2JzeijcD^S6)J0L(0Vv0
zo$W&d8pjU$ETskhSouJT1N^IHb|%z=kg}#0v=7gaF=L?yv){e|?=?H2t1nK0-hpW^
zUvyDyG>u09_jjhlQNUQxE(Q~RJilM1wgNl_=-7FI$poEdYeNvJHnaU>1){qjIlh=2
zx~b@v|Gr0sF21>z=1aR6HD74`v?{U277Q6-#l~>4Gjn6+Jn`)1?^;swy&UHBKk?m8
z!AKyH9t$SGzK4#EmY&XxZ7egMjJC>_z?tm=-WgssJb`QJtHyYszzrM@&nF*alTmx{
zfQSpNh)ACKByG=J@9K0XsmAKvZpy``$GaLkHy4*toznaUk-PNRoBW&HU2Nz~nq4#}
zxTw?Tj@yL4=o>gEKWrJ+k(}V3P@m8spi+$!_>?>y`jMm&!$2qv%{enCqw_;os=brI
zX%ujp_@h8n=1<UzYI@x<!Ek;LdOH1Mf6*z@(8-St%zN`;jwq%5$lbc-ZDudQoNr$O
zS2MZ93ml^)w}iLB%ESvxL62H!)+OQi)}NF*pyg-tF|dS}_@=m~&`dN)3(*P|bdvt5
zY*Lt4djxYi8MEDb_TW4Cj$a9-aU)QpKIGH5V};2<@9HP%N$|S`$Z!wgIHepdxAT4?
zkdyley%!;#vw-#^eYnZgf&y}RMA`7pCs6QRFjlB%`UTdgO$;CV+`=j$tt%N*UkPpA
z@hDiF4UZA{HV|GcG<TIr|CII(#<JxH^Wl8QE9Bi@oa*a8gfv7ity~jGQaaUAGl-Ub
z2`eyZG50_!1x!H;$$Q#XjpI!^#Sv+<Pu1iPUc5>ic27~L$YN8Z#|bwM;x6JBifQu7
zD;dRz=u!sEbTr=SPhTehF}ymyTP^IVkGP_nHK`EHEp+5)bY79a2K$#1@x4`r3kYp|
z;}6P|l*c&Bq7U}z=|rf}_{zu0t>__*giB13!0Mt|mUO}ZNNBb%CVZJSar%ui+R!>;
z%a5o<L?x}QvU*z%SJwLvYCqHj)r9|Kjg&sHj$UIP(Geat%5BQ&)-EE{2o?NV;X_BO
zwPhaa4F1t`qA)YozkBo`;Yag1A)J=8P50pxv?pqTaoSi8Eu<@k4WU#x62sPSR4+W{
zguoQ2#O7+CwzOV8V%p(tnNg3w^motRh~S#@^Y{eX)8;9gNIcFxS5S;h=t9~8$*)Yj
zLNBnucW{0S%CjH!Y<>Bpht2&&jJD+5OLaJHzZ~9J-sH_r-zb~Z61wnopZBb3jqwhL
zG}}ZG5B$DIgZ1&w-yboW`{zEV<F=REXRJ3mh8c>xB_;$MTeq<grfS&SRq?cQKHi_1
zqn;i6KVgzFPQ~)?yXsCh$j`<eo6Hawi_FIxcxYA4H>J!(5T|tn3G9r&+&y{HUqd$M
z(Zv443p!pTDKu{JJK8oWrPBVJLG5^7)YbA`zS9q8GiXYG7I@dIqPMCWuXh7j`+9=>
zdFn0>j!4_Iw~ANCrr++=I`NAtF8O)R+x3k()e!gbe{bx!icKp$Vw9OfC#Kuoj!^2<
zTPkiPwLo}{wFGR%PRw#HVrMVn;t&{BZgCp4j^A)zCEa5;72L2n7=nEv`6K&yEJaaK
zfCR?xjb%DM3ktJSs`l$NMQZ`?>U`!NOW}>?qT%9qJN1_(Rt_rkvigKCa29jANrh4d
zT!1-Gjb!aNMQLF3eUINe)Ro21@$#?-nLVc1iWh&+TYP!sGUhdo7DwN`n~#bW1M_M<
zq<CYe>dBv$(&0320Lmsk$8r(7d%sO0$XAhlRRHm7V1q0}u~14NCBexo*&+Bfj@K@z
zQ8?bqc?)cgJdGU=5&w1XK$cQ{e_^C4X!zwlj4_AoCS18}ZE6R>i}{hfc***FI_5R?
z2H|phwR5wB7Se@9_mSNBp<Te7U-9c%P7S3K&ng<s^OJ=0r;ONN!rPFi4W4QdAz6#g
z<2Te@+>4=Qe*EiFL2(%X7Vq4MK+T3O!d3crwz|=?W9{RCss|B1o$K;jm%jss?7$Bt
z?>hh7x>n37+i3;{Lw}TsGDq<$`qrHt1&eLhGs|C#C%;}O$Wk}RBZrd6O|VOj#t@pK
z8s)fmY~YA>^DRH>;@cGJ%E$I#$Lpx7mPCj@f3rojiecJZWjAQ^$VRVRp|OHlmU%di
zKLyiSkbYI|d`&Ik-3f{*mJvfVXqQ|qe~;vT?8<E{rh5VWdUM9c3Q{-aEZhJGLxEOK
zj45jnAb=Q}$n_ha>I)D%_6x5|4!=#-_UC58B=(3B4Cl$P-IC+UZW_<*{y^QWElY$&
z5~wpBU7p^nhJ&lo{#=O!D&#F-&4vs_3~)E=1Y2+3BT-f$`Z!V3KfgQ_9B;@Y9L%^4
z6<S7{ddw;7nPc47;uv;xdE#+>*wf&GI@T5I6KA_bNs_`(az|T0`5>&7!u0R3hC93<
z&%8DIL!n!T+=%2CG7@zyZz(EDH}jBD$JjqOF83NFh02fSgG3`K)J7JLyWi($ug4BB
zG0?0t4icmnwChdzJXi>{w2~-KJOF-ZLUM@xGwe?gkpeay0|W2L0a&Dl-I7U@@mZ*T
z|AFIh3`}xamIwDsgk_i17#AFB=5G6Xq2@K$uZoxENdsgi1>d2pDcWx<%*e{~FH!s(
z*v@%(s=w!6aS|{{`K0fR#o#1|hbLnn7bu~TiKY(3_*h$KH8GOy+<H-8llE8a1L$m;
zUwGW8Ap9WF_wT&o4h5DIraMQ(uYYW~F-m_oTuzY#rw8P2eq>YnkvgM}6|b#$G(dh*
zj_l<DY&gc>>sXInI3$RY;BDes2MSinYaAP<RaWIHHS-mq=VqX4hTCRDXCPXf#2S1M
z$=FgfZRHNI3SD~Swq}LcKX`GayItH~tNgleoCoH>)CE}$=gf8lvPP|2f5)&j>YisC
zvM*0(3L%kNI>Dt5GeOL6w64;$b0J!(&BEga-kPD;`wqFTRF2khE<OSy6FJ)V9B%1X
zi^|4g%Dqc{9vCBtQ=c81$R%O4T$lN5e@tR+*Cbw&CMni|1PkC+P}8&<9^ulr<=sVj
z<8fyv$@jZPeh;bHtxyd+ukR@J6mkPDaCx8wGxb{iA!3!Z(Y#Pui8YUNSMWX^%HJP<
z!G{CN5@JxlX&+djP1$0k(c_;l4eC-Ke@VLR_T23YDXKbx>G}H13iu*k<#X7rUD^C7
z#Xi$l{ypRAaU*hdp+}5<s%P_P-1Jl5?lTgMODlxy68o6a1Jui-iAvjHK^4q(L~W*o
zv^<@G+xW`e1y9j9+fUK1J8M7o30DQFiWca63YIUmiBV-Du6=KgAacZku2CEN_emE~
z<k#~|^ICM4jP5s4VEXM3kh>(j2qH&^>h0S{hon~z7bhia(>xs{)n|GrfI0s;&`+g-
zaN$`nA))a8A!Dk>zW<9Cg?`fC;;Le>gfWj6$D4>J#0dRi%&p1V6;D6i3seQa6mu?U
zs`?nA(h`$dN_PdiIiNgYlHjEG^YWV~gicd-88vm_Ou^5@sOeC}5kV7K39oU8S7jR}
zoGq~)Dm@6Z8e#BPc%Reyv4yTIey2q&Ul{W@`JrS351Te05)jw*jOXEs(s`E)$>>7w
z;%KpW=^*71(m`N~jXGtFe}a8cBzH)9q<Z3jnZ-hay7{!0DL|by79aXUne-dJggIv3
zh{QL^p8=fEPXD_@C|s7ut&gt*g6#So8XjVZa}*)jzTL$|t8>{u;?KvSda)b?u54B`
z($fS+B2)76sc_Ly!SaxtU+$&F8uQz&_Qzh}ev5T;pf@}g0RVx*15_DAdT-$Rw<pQk
zN8B0IfsOThVpn%n1gkS<-9up7%tAG|k)mIw^PVYsUgVL+I<_wg)y$Oq$=95WzUn`u
zGS;kJZh>rKPvNJD94jO^x1ZDPmYsgXgv#j<=k-Qr=TFViZ^|C0S{yM_%H$v{MU%s3
z$E%8|p|ZKldwj?40`IGoFH~$2I|OA~B5Wfe1LKQ9U+t2-rZ|11DPGGJjP*auBpjF<
zw#%m&Kezf?Mj%ODw%YE8g0|fp0Q59HxhG6%1v8|JJ{7CR{mSQ!bT6eMT_K27AXKwN
z{dmqzh}v8&%RZ<rN=b$qY<VRa36l)`<Py2Iu!P1JDbz?RK`GoUop$B`i9}g)$stW`
z{us=*<ga{0ood=)V^WnEDkSn<6=xJ5az;0|gkP&n%31v7A$sJc((59LQI!@(D@2E?
ztmJpbuJS3w>eD_oIX?}K11{rhG%ew+{5XRoRzG`U8!aymMk$}SJHCq^<6lk~TbwkV
zav&YC1tzSbN}pVP1X?DT^(%+2`vl$MDtrPrBe>jxc<%(b_DHt0ht(4_dx?rTOZ$?#
zRD6aE@-HCJMT!dAo`KreqVH*2^NH3Lb5rOq%XYJB5xFFa22tN}4isz@D?U4N?td<E
z=9lSQfqr!`y{@|36RI$Ys+b0dT~lOiALSQ!S_v`DEzjMZ!LYSDcYM;T{inC#NgQnZ
zZ>%K5FzFzk9+^TH$xDBJ!cuniz0L@%LPegpRz5&wA~S>1`q4e<TwvAK>}iC}I)uPb
z2r(A6rq4e(k3gAL3{NoQ`xh6f&bt|$!)G6N*ZX3QnBoyHg_itA9cyB7NO<9Ajs_)1
z;IQiDb?G~k{Gx?w?5jgsAPIpHLbd20Oz-RMLVdb?(g0bm2!oXulMp|@_90}Dm_dtG
zjas0KZ+hcdLg?=p0P{)+d|7sHWogNP1&9_aq>oJY=YQJ*^57Kq!1~9Roq)zXVLzTF
zQ!KU@aJC0d`yrj*Gs7%=72+?6^5k)WbJ8W2-q2*dSk$;lEsrlKlc@E_NS(AETYhdT
zTn!Rn>U7o`EgSo@&p<iG_PHKihMH3aO)iG;8<PO1*soBiP%L$kqz<M(^;FV2j`|)X
zpv9P!yDshHG>^*&4$8raGgpXXK&Q+;xq#!_)GE-OOz*rWLoKSm+`PMkt`BcgeNkSN
zfnXbRjrZKZ?~Wy0$$*46l=OEeh>6k(H_M`zP^%Rbefg#S)#riHI}0m`io}&6ja{$M
zOQJg2(Cg4(n5k$l`p*Uf=7mIp;;%OGT+&XIP}#(wt{BU1aPBg+9g%;dPd3Kid03#I
z(VInBLY_4j4eAPE;zm`0f-|P)SLZ_CZkt0)OZIUhmK@x`u<NuYnoHrAqp@!TN>X|`
zm7*=~0#f9I<yjO<7h837ci~!`tS7a5#Rt=DHs1*tw|KkV>C}FuJi~iA&gYsG_B1eN
zny!ZBq#&ev@nQ{Rs>-UCJGBpe$yavaAq$&x8v<>IOL@i+k<y!=bG=5%SQ+@2jy*~b
zz5O;R=F!CT1i2mRAN^e|Uc!h<jZN99wUSk3(T`*qfqoXNFv+MWuf(ph-hW#wTyK90
z@OsTho|yn1*$H2VeMOc-wI)^ubAo!xr=PByoWSV(XA8I<?ez79u3o`gKR^X@b>y~l
z-#jR1qXqT9x{(y3^WWZ#+$-JTCUx>_qMfu?cpqQ)3uVBq@Mm`HFYFIxpI)|pl14zE
zy4eZrE;FvI6BDrR|Jbmc?o<**uqj1(zgSIhiVirsY3$$3R)USbJ@dZm)*?$=B$##`
zFFcLY`r-4*leX^Bp8K(EE`Q@3M(p>F%{pGv3QXqd-CEP*A+@VZRo!<!e8WQ{uJx60
z0t=F$VMWp}9@;KB9d?vTf<}y&#xIXeYu5wS)Wz4I%EG&E!?e=~D`eD1E3}^HV@f+~
z$qp8qNC&-xOZe-4WSPBRQA-C{;Ranm_rMnuW(2m^ah|HE_^+2K!<kNx`dWQ;0X7de
z_#7cp5T-rdB@|V@B^8@<=P`JTAwI3F*~yOK>yCMY&A(4aEt7Jx&{V|3)>T`>7B(|0
zb%B2;JfO1W$nwt%=a(vjWNL6!F%>I=RV8W8G63U7gnLR!yGs{`n31S=0retN&6k4p
zHz6KpC=Fm*On=tr%4V>5=csR?U53i$p`Z!_p!}7^^;@5BbyIZnAsF=4<jF)+E~>(t
zq}Xr5IO15UL+X_)_;Zz2oUs%KO*5t0A=-&AYAA+sy7@x2Up_!sa(70omlbF*DZ75w
zjW0`ip#Q9$)?=AS1tra|8BB@j9I95}nz>-PD6`u>IPX;zm59Uge#vn8S;N_bcaThc
zOmQ7n2A2yA3tue2Q*p>$U)C1CMxc-|gmEx(SQ05a3*h_m4{NltkBQ`xlz(1WkB}RD
z^x-WSV~q!uOWGp@e0H-Nb#SF{8N3+lem?2U+r#BfxSSnX@)5?XiZ|}pvKIp|(lMFz
zUUmP{98+}^$qIQhgnzhC_n3)q1J87UiyBOZsDhKFrnZQ#u9dnAe<(YiTpFgyXP0sK
z?X&AIT_JzW4BhlwX~@KcEr1?=+@M|%xtQM$kmb@nb0as!K$L{%*wlU!I=G5fG{$a{
z^yUZ79fyh#ml2gQU%IIE7|JR7ewR+u!7TE8O5&h(*&pT4GNyXiyg;X9PCc|L%M+=5
zMGYN4a~IIYmw4Wwro&VgO>`njZY4K*|D$1~eot;kW19dA`h$uV?H-+hrMPG-IG;_I
z2<|}#Ac}zwK)SH08#xoZbYot#c(o1zBav(36}G8v+J(#!^)nR>swOHujO!uVysP1t
zIHRW_<j@dI#x3&x=GG&jR5`A(=85Vd4>H}kFa1U$H(G6I8LI+bx)Qcjzj|A)1ANMj
z`=yZ)EuER^*NZ<JMrM@1I+El?GaPDd$;Jy(KBalY4R7fLXE8OHHqUwSbJ^=tJ4iXe
z(7cH1uL|slDL}-;e4FDnMPMFKYwVpguSt%Tsy{k<(E5+PMZ3)UN1vW)Atd9{xJPdT
z<1z2rYs+l{QFq%}Al<g6MccM~A`15iJ0nYoRhO$9Y?eMtlAGQmPB4uO;Vc_bfrOk_
zP90Db^$)#k+v}gJkn@vhoa&`&i$s6nhx~bUKL>-Jj@F)*!pW}FdH!u0hwtYd*V$Kn
z4U0X(4HeAjJrRZ^qRi-Q_}vj0BP{7*-nB}h*K}SV#83m$sz1!*65Cs9H*81vBXz6#
z$B|%Q9-Vivqzp3!zVLRk|B^6Ajy>Gx5wzyP_poCSWB#0&cEd3jPD&I-6d1h{=;w|Z
zNQhZW|LPd3OII1liB9}vT-QkfE{4g#5XyIA)n3ND?|3ivxJLME57<pj5pR~nVg5Wu
zR7@(o6NYFTqxrC$6<*@!Mvg=dfrO<QL4Ki5Q)PMz?_;ke)~msFdGm0$wObxR`KP5C
z{E@yp&_@AUqgJhukQ1b=lFxGbl;i?ykWxzY@|lV8B>J}k;%6_DS-p*-FT(_b!YFD7
zple{xTrwZx;6Pf`&4MmPo<kK%ya~}yU&=!F`H<z~U~LoXqhJaBGcQ_>qaw?dh&}jA
z8=Z~fxje0;doUS0-yi4L3%oc;T=_KnOH9q}ETPSO?Ie;4PUeoQ>YRqR_r42<VL8{b
zwN=AaLmY|VklvxOKx=MR1-BW4oJ%4+J_R4CVuGJ(?*eA#M8E9!guL0|-ARB8M8UrG
zNUX?KfEb{C+=F)aJJA09L4_wF4rSW;KPn?78W%ImmNAHPp{xJ$0ys3BZ}dO?L@2`Z
zWQ8-ek()%nj}{mzc<9rss^Xjmto874yq!*Ssf|20!(wb9@8>=aco3;DEDZf_!;L`v
zu>BE<AyF}{HT==p8x*4Y_CafKM@aRfRz4p#g`e2!vS7iMyYQ$9F3RU${LyVoBn)2q
zu!2R*G|}#yMV^!OYH;)+_NtD=vh3K+W7@s%d5{cH-KM@Tf{2KIRmCI+VBhd(mn73&
z7ApkiD4EGF_Fg;EWT1?He`cn^_jX2BtRfyFf-;JYg1o_p(~vhfOQIYb)7Xhas9ioy
z+QrszW#Y|y+busok&dBH|3LKaV2#lGw`o#V?^&{vvW-*eT~@jzjqL_<N54G&kmWz^
zX1*r!^}==01JXC~q<`Xm6HM3B*1thphjOA_pf_F)jjZ^s&r5MO^$pnmaEO#nn?~UP
ziJpw|UHKw~Bwj`KdJ=-`Gj2^<)sR(uIsG=h_$;L*qb$TuZxc-0Se0FSvW)-bnKF)9
zWqF+9j230KQW_W{g1Q^$ONh9E;Um&I5lW``*w)WggVi}f*=2J2U`E4Lk^Sxz;`h(b
zqTZbpbuO#kY{Yq_capL1`~kwXO|^%Xa;P6Zgj3sSX|&*&iU;{g@2$ZoUPt&zBXDU1
zN8;U>H^r4Q*{8Fb3YL?@&Df)u&sJtYx{|q%I&S>>GdTpN9#e@F?&19vs&0Ko=%+On
z<qA_VxfP4CEUM4*YBfBZ@ujLNwv$<T*Uv{RbJ@E$ePydODapG9V2<WwKup!6s=rMQ
zrl)&kW=$?!ygx0dSsS`wcEi>Kl_xp#PJrn)gU>p5mq7nb2V+c+$4Pq^JIyMTIUczP
z;>wA&U^fQ`@;}s#a))7gcVQerky1ILevBTEB{QvMspZn}AgzAAM=vh`Yb#CJfU4W*
zi8h<0ZWPXF46@*?N*v9cyTVf!^>Uy1-WjugLVq^QmbkPBv3#Np(XW^;Hwj<m(T{$L
z69*ZM2avi{q)P16AM>e_7}K}v{JmNJONPb`KbA4#y(AlgO-orkZ2cjTNsIVP$_PUi
zpX$ZAjOiP;B~!=-k5M4y(gpuIjzTwtEdth-Wy3u@!@BVm*^lkGs@9g#T5a<Z_aEnE
zlbZzFTXd<13R2eiWafUOANur+;A=Pk5l|LZIp(77gADA;@rJK+=U#H5*Bw*0EsS$&
zCyXOb2}(+JV+4@ff+MuJp4ShO4Dg@v9$H{kC~*Jp#%nY~5+fU;7~kSAAizd1>~+4M
z+@ReXV+p%w%VE>dpTK4!9QY|K`jWq;YBahZKAvsC{cYT-p%UwbjciOQ<3Au3l>5vw
z>9m1<mI^&ILd&73{%K@G3Rf<lr%f7ec%K}0Wj@Ba7RRjKZL=bwJ{$KHv4V$TMuZV9
z>}7>1X`a{c*bmx!xyM;&-yEh=aW8P4?UEI1kkAJCwZ1yEYp-(`gY}l9fsvAcVN+ci
z+M@8ks=ve2cpcj(tF6(JwRcKmwdd{KtHXpaQ=B~(&lhIwo*CH@EX843Ig+XWj9sB8
zY+|nXYzW03Rh1qTl8nq4;|NbZdra4)hVu{~m53ii8*WbyZ<J{%!q`t{u`6q~zhiNq
zu(k&{MY+sp#wt42$md{v4G+WR@*mNZ5X%%&nC$D?YKQnb*T{c(y%o)b&Q40pRHEYA
z$SA_KQ``yatS?xYS}qrh@Yr`b!-RJ?%Ni76Sf_&=6Yo-To?t2ysIj28R9upj=y<He
z6^c9xT8Y{C+%Ptqv!dY3Y?q{t=?*lCLggu~)Ki3*b(|<#MK<iDGPz3`o6JnFiDdU#
zR-6SaSm$22jRbWN%-M>9(Z9j;qf;)ANcVV0t;d&gL3+Eo=Je^l8__qU4BjIOBgA@B
z#W^_*y41tT#TxFUWq*$WbME2&a06nvk5qF3uPA&xc((tkIPEL<_(uvBZ4Y`aYN<`h
zSVZ&KiM0aFLPY~N$jy{eg$2GQ2Y5%Kaf-Z_Yr4BexheK+i4qd>eY=@3gzduLj9S=k
z7l>2v`y(u9+IY`HUlo4H#AcA1h8Wi-KZ{g05hmb@qH8tSs<N1p_q7?bi9Zm_aTT(n
z{9I%Q?sgNokz(Yi=MzD7h#Y_%Mg|E_7_DgVK1JvGKh_<fZc9W`x?v~C3HZPVHx8H*
z?<%fg{B!pKzjZ3=+j3l0_dN%>@qd*=h$+6M(Yf#lUvK|&P|LxDj{A%^d=A>XSK#yp
zlukMI<W6tgMT28@BLpHPN4s(=NCpX*tw62gP^^I&YLz58y+`quVu?PI?O%;ld_`~M
zTv~$_>`sthMG{{m#Xnh{FAtRoB*ZKp+1@D9%xp9(I;LH7(kt<w-HaFiVvM<sNGjIT
zahuQZ{MGN9n&3tYuibH*oMJB>F~$%0UH!uDdJXy$1+B`hTJ{P3e6^B(2}bS@s6ZB7
zq<g}OkJ9C)OI2q)RSoTPGN-zDy{;6`NehyYm-L36i^)Hd?{&-7)D~*R_}>)@-47R#
zC;xqkDrNclyZn>Y-rp>gEsBK1j|4ES;<YVvTgb#pGOK{#XiCzv5F3Slz9~XN50&eQ
zhpRm7F0Yk$93~2Vx)6{CM^Uf^Zm6Qa!`la#JS(M$;`~;t?VlO}J)hrfdvS-ja^6S{
z&l6~F--6W^5coBi|Mc&`vaph;PbX_rdtLeZk%;Rjjt`biJt%1R9@ZxWR!ELON*WEg
z=nVk|0hga=_4InK^Qt!3RRZS6%Vmfwx%;?_dPuA5t5%;=kkDu?XFHC!<|ga2ix6I<
z1CT?CpbJ=mcE)UF*eHCHZ;7+zLLOJT-2~4DSn3L|^_xuw=Y30%m-cBYyT7UVSu!!0
z|3UmK7p;9^DE(n`1c5Cuq^T`I!~k{ZFgvg;F%9xVHZO(!-3$`@X9ccQ+L)nNSFl|&
zJ?Br&%R*A5#E%sVcr6pJ->?efO#VF+$`cbC5k3w8gou)@xFD+IJ@%;}ULL#(xt}WC
zT;M9&`;UUD_HV(|RE4l8lxZ$&2aYgXoFc5J$sf4x6wJcY@?Gc+U!NK=6KEZobbO>g
z?Lmv^$f8b-LF!s-9&R~zyw>!Y1#M&a>XW8mu^XT&<&zs5^b9gb!gxP|i7X@6CNqgL
z&#e%=3P~Hqy*0g0X)#A_%@ibuPNM$2Gxmx+U`fT`%2w&xIQ(t9RaoVkHMPdG!Xib3
zry^Q5Hb}U^+}beY1Tcs&5hLKxRP{4k)fjxng>NO97x9D_pDJ#{*0f%a#3=V0^=A3d
zDV0SRmW4-3L`E%|vi{e)nZQBe>JNz{-d}JXSrE}IL_K0U_j_vFn67ciNV&!%(YQKr
z>UmGA%^ef0+>?V)TTnWg+Hik6TlKLaPA4p>JPP^O!Ql@_e}#o&^kff^n1XD-wr=S<
zsH$)MX6&ea&8E8VN>n0|yK&q49VXjc(|Dd+Oeo9e4>|qN4bSsZ3?a;A_*!X9mOtMg
zOUX0XH+=gP((QG?WZJkbi{5PZ)gAM=O>klS5-qF<-{Nf0rkVD9%x7e2Buv;g&mH5U
z>2qG|o05v-d&p;3Eks+RU%a<Ps+kk3kn<n{t;w&^{*$eqZW_t^UxU!uEx3iawq4yV
zW!0JguNpjFy~l1b?%LOTI=jE7DiZUf{t%<Kw8pQ;lck=)%JJCITgs#o_z!*0zdneS
zBecCRnIQc!51Ncw2)bl7De?E4ggNBZmF_lNC)O9fvv4oa6FKJ7sbi>0PQc-wI96o`
zr~^H>l{zo~eI&eBBnRVo*NYFYe;ATQML+PntY&sW8c&~dae>qrvhOeboDE7fWwcQY
zlvCroDEA!1N%g^wNEa!kT?)~|r-(%_sSnJe1_oL|GnptwA9C)Gk<bVJMt|ELO|x_9
zxY0ONhMX-@kTAS>su|B(6^X(m37S9GD`Utwu!$~fM14?i-SSW-O~jj;Nm4>co{erc
zhd?Fzd<akCuVb0Z&eW(T61r?-;HVcD_$_yDSwJ6~i5=yh=!biFWuEk+i$*QO>Nb<n
z25GB&`f8~_+!wz#A9aXkgpXcG7%N`>6XYl&hum9-C|5}84E}BuVt(PZjj-dK#|JOD
zH4T1CJz*241#SIc&f3nmRJvZ)f|3BOPZ9NA?Qbg0Z}A1!>8gMhepPSH*HFNC4g~)&
z9Rl+Ozknr7=&rI^DG$~rT}w_$?FLO3K>@eK$URnZ!jIBR7`~XqXMcb{TGZXsC7;H7
z?JDCD6>oCf41=)R)?sbc7)1~+<G`?dqW(_P0+NNVmU?4?n^z*sj8nBh8ae}<;ojlL
z*3Mqk8-Wom9)Frmr`qUuy#zBjHVpc&B(8D0-R!FgE;-77m2`zc8$LlgVD8l0sE82P
zdppP}R?}|%tECCJmjQsQ#$1s@h=Z>a&w$U039HiiQ!0e=h(Eg&%it-&fvld?ulqY6
z;NwPf4bHMU5m{3N0?Qu*$|xI{TI}T80=RGJ-nV-r>>Aj4@iG6kZP9B6=RmX$8Ko!*
z@3l3G>>(;A+<$6L&bwd4%yc}ZJ!kx4?WuLdXqAY`B_l<>?-juHJie*<N0OU48mGP<
zP-swJa+aGCHl|;bdVW6|*Z9>dSafh^l_!FGM(_!;zP9g=j^5o)?G!0FP2xiCYs7}_
zK$Mzo3rz9WEvDO#o*q6h7SqgVcLD^fWF}C`O@1CA7it<(8u{0z2Dk|fCU>7F<N8Zp
z*B*3T-1T!0`i(**;}7n%>C&$XEVAVx^Z-S<7)kJov#2e(9enA@W=1q882vo%r@x)V
z*ZNX$*n-X2-_UQGhhBcFhz@=MnpWmcU!VwSVtA11w*Pw#sO}TTmA?5lUE{3Y=W6gX
zW7aH_o!ip;TP)LYDCmhwJP&yWyYv_Cd|KHh*bsRYLQX}WT)?q4qdo>;bEu3U(hrqe
z(eHu?lbXu_ePIkKP~W537rz4HubuA-Ng95I8i#16-7$mv1p&_Ib3REAeACibhfW!v
zgPjn(x3L3?p^p{4QsmAj;^`#tTr{G8V<Z{(KU~SY>w<-q(mHIt5IFndyPA@8_`7|m
z=#N*~yhMs)yG}1x$jt3B&Ct)Srk^HeDbD^gp6zH)Z8pgckNSUPUKvMpXb4#|W$W1+
z_w`Gg8PHeZH%Kxy+-5j$kEoF9z<))$FUxF{i?t`GDZH6he9uuAm=mfbGxI)8_;!Rm
z@qG7f;^icd#k8ybMx{I7q(_yF>GX}x>K{=uzum^YQ>q=NIWI&?PKmNw;Nv|ja&9v(
zXbPE@rlN@F$L*TYo1uEis}}zAlE$UIkYqX>_HsYA2@iU$yf*}iA_&&R%MG=Q(=9tN
z1n3jzkFm}281MO*!RP7}&Rsn%-xYQj?6s=l2#zuJHlO{1^@tm(+rbYzZ^aX{)kD@Z
zMy@V7Spa-mzrL)>fr|)`clN+m*>d+d7%Wzk4iXjc^^tx2N*X3#o_49Ka4arW<F?09
zvmJhBtV4p`2|Un}+zK*)Qm96%cpW`4hfdb{P*ROr<<Rxg5K8d;_~SaP65b>_7-pc-
zZ3u_zN=dW92{Pc&At`B}Hl`<>nKT<}4$13iH90S2)H>(|>6ru<*w&JrqOCzX6+D1o
zu+wbtrN<zd(@Sx6Z+U52(#rZCLK4$Z?6`8D@`_p&D;v^Oj{YYsWB0w!$(Fef$vv_(
zo;aRK>Tf<pg~wvRp=DB!8Sff3H-raC!RTWTmfE|463fiOuR(RA@mY<0{Q7t=Y8#NF
z6O-Vt?7p$a@W?@Z6#+ezlur2hH0=70Nv5Q<9z1?f${>#fhh@|u(+!xd5t2w*dK<iW
zf3{n=GtmzxWn>q1vCsgZa2$6s21ZJH9MY+HH>Aw(X{d9Ka`$2}@iRCo59^!#K<A8`
zT>HM$(XVgy$zKvFy{?RH$~{fMrYjUdfFgyof`gGoW5_2xC5Ka4?ZqROFQvjVlj83`
z8UTra&s%n4<_{gj-Y;1NCpo~6Vf6a|{;ddlt1IRB<JONAC<i>}bC-u2YCPsZenKKh
z0lgd=?_%nz;}Ckp;<^u;Ahl4I^VSH4KW=;=tP&4bgbh~mK7NmTB`ClI#zRum6_+Tz
zGprA!^`t$cueMan;KY7#RR7xU!M5CivFi-^C2j9XHkSL{7f5fE!n!e-5ghuJc$zh(
z;KwPgQTpD{qqwLiL!n!Zs5a@^<Rm1_Uu-@&@fF)YmZKSkJ&GqVmmJV%=#SxF#8lz>
zo^Bv@6;?^KAqIpVXKj*?_1F~-nBR30KuG!N?FPnj-20{|zdSh>XO;_X2F4Y}JBH<j
z2jeCYymvivHKNdW@S4QuLU`Z?7~IEJmjN%4%JcQC+_S-cd}{sy{)`!as>_!o>e>nG
zz619*an!eB?0w~FdNY<}fR<3^Zaf^0XyMP{G*GKGUO5MMF}Fw<EgB6qFsRBYX<7`W
zQ{hurvvP=<MLm%9#BXyTD}yLZkI^Y=P4|Fk@|!-;fiBBVxyu$0!&u>{=j-(ON-vj8
zzu+PK#E(tB2;WjeZ~Ym!+~;FH4(*xt#;}<(ERz%M)W<|%?-eSwpWj<TnZYpe7WOp~
zMrk1b@qkJvGlRlX_k$;&G$i;cxQD?04GueRUlTWNo2KR9_~nbpl#|x1sKNcn3<wX&
zF?%RM8`JvU=-1CDqVB=E5@}QiY~eT5{c-l>A)6TVr;!#IK+IUNL%^lqM=2A}m1>p|
ze}2~+le$UHve*HCiKK@_ZLzU(N+9@EQ(=n|Xc0p_qF*g)cE#9b!+pY6<Z2;!TC!Yy
zvKK80I-<_yW*R+i^B$f39_a<^vdO0f#gIp<y;ex&36KXO&jbd@kME4yq-EQDAwaUS
z;?)2AF?mutH5}5;O+4k<<DFs9Sc{r!<bq|iMXbijaXFZ_9#R<${maT(X7~iU<69}7
zFz!rXuMj$oYpual_$4Ew%F2nVk=j9ULcpP81kgg8MslJh)GzACrt6Qnio*oRYJY9a
zH8fN(zMT0gC_3=&Y-_xK^$($FWX#~6NiCJjB~ih87eg5sWt1jZJ4H5}#If%p$?~G#
zQKVg*hnRDFYtD@EBFSwbUvi;;k=w7*!K_X<wEpuax9BpgEjY@0SmjYg+@I4wTAa#f
zyC<-@k7Q|{qqpTdEcwL`^9$=NcPC_7SiH_}+8v>~qh)Eh6Hc;;KTaTUb|`x7<*A%$
zn;W<3_Z!51bIc*4XU{a%V>SNgiQSFjk=(D-PUt5G?po(mexuJbUcEbqMYg_)ClPFa
zer2;Y9G1aJuQQXvAK2Be;wNUc$1ct@UN@hKc4#V$Kd%y{Ba~XqsFoY=D>FYUDtp`C
zYFgoP^Dx47i^cNi0CM-wYU-dWs#T}0wJWc`PJMGK;!ZH>m85VV(MAyb0F_}I=w_ux
z0;H~a?>5kD4nZrhqnES;Rr<tn(+x`Ek4U$#L4RsRY(j`HR-hIp=_65+9#1O&0tLlS
z`<&YdP5)~lm~A+Hnw?-syn-uy%xXj2y#do>Qu|<zYti0Km)z`t+0Lm#TF4#OQknpf
zzh0oHu-M0ENCv|_N?V=6u9&e?32cFy%|h~Cb^tTI$qrK(xeQcb5o5c1fTb@YzU}wa
zk3G=uruEgM4fm7LH`1{2Qcp2MmE7~eH4JqB>c>(733LOG`s5q$qEPCWo4|QEgnJ(`
zH{&xs9H~hX2A+IOUjzJz8E7gfEdrEk<?A-{@*RoEJ4~Q8f&8yv7v)%vkRW#-Lk2R5
z`!nso%*k&wy^_9Lw>Q5yqOy>nkf_he;xLKsWV#k-`%Wr;*MSv49H7tpg7w9;T%B-3
z1X!2yPU4qRN<|aa!KriOgo$4mHGGlu9)6U&BMoI*BZBIF5R8WH5?+bR-^+MC0_Gk~
zv2ElYDYCa{{UScYiSNJ(E%`_aH4eN(?DC*VXv8KIX9@qHEh0E4Vh-crwVcGydV$>3
zrCD>xAgx3}P#FI{aPL5K5oaf(jIAc;dcVhZWwFJ4{``8snQ5dAp;D}JHqhsXN1lO@
z{3#<ia+p=IQqLaME+FS~Ub3&c2S%WHP>5QR%?$?HGT1Wqq*Kpb%hiPZy&7={&x1Y^
zn%-FymdF0KBk`WIGyIe}P*TsYWCVwJK;{FC4DDOLc#_|hIP><jWv#QU0VX{wQ@(@V
zXGw%aN&xwN)HX->O(tiaz`ajDIZW&vSY2p_kdjT!E3&hYN+vbYua$)d4qZd<E;-9=
ze~iR{UijHfHLcAwaDh(Q+LH^S*xRZzh=x56n@jEfYFWE&^642;P(5HGH9{O&L7xZg
zjH?Qt2Xo%>7HVN5Z~vAelXfRW<gUpjA;(B{TYFaN73U1!Tc7h&#X_8J@XA&uSfU(l
z)woM|wF6S3)VJr&+>N9r!wySrlhq{9OXb=9p3oa=$oJe8Gb%HQ>%`@<as+)_H+|C!
zp|&I85s6B!<Aw;E16{uukTR`sLQQHZ{LwY_o2p3b^{8r6*ui5d;&aSnFloz~L2_=~
z6IuGS#rNFso{59-IFL9v#$`n_hyV~QR50BAtpY|<$EMe>q}89*8f}S&egcAN#U}01
z6+fZEN*PfS24I&F=U=@)Kk(SFh`{N8vG<l?RjzH<sDv=-iAZ+{(jn5)h)9DX9Rf;s
zN;imfD=i2pEiEl24bt5yjneR4w@cSr@AJm9zP*3!{o`H7JUoD<%zIvWp5q+n7@RW!
z9T1zE6^s&q3GzJCEwYHUGUPEiLD+mI@Y>PBzj&}l31Dm+s0-VhCKL4#UQpPM`u14D
zwe_53<^66DKZ|>ZdiPyNPyMNa4?#<*vM*cl4uoC2ysJCtR=+CmSc(kB|9KF)qYXbh
z-K{FnLA;4LVl~6@j7}60dMQPVGE!5qCG#7;6RwKG6xW-p;NsLDB5ZL2*^}yFWc{_&
zuKc|dKNr{Ci*CwuMK|0jwE$aXLVU#g5r+MGa*_DoSC8heHld{Kv1G6Z47Qr&Op#PH
zZH>nuoZ>=E9^r?Bq$StyelT3Qf1oJRRYc+?huxb^fU-JRHed5_5m|ZBdDfkw_#F^j
zK4BFeIJ-HC15$qpqIx;zRjQ>2HwwhM{ZS+Bu-Wg|T}w}%&0pQi#9(jKUtR|yMTHOf
zyi0w8Ooh_@F!u3mgJAiIFgg-F?T8|_Jt}_>Z6qh%NTd?v@VCkMAH8p46%Bg$hXi?9
z)JaTTcwBILd0mk%^QN*bG1BuzdVZsw1^$Zj211(8m}F2lFkKp{Up?G}ozvb6zO#?y
zWfL44fw%vTntVG<B#YSe=wzKiTUo6(bQE_9b%JjvdLPNOxeK<lU~s(X`6k=qj6>Wa
ziha-Bf43wS4nJyjgzC&(K3rj?e*0?&_{#^CPUn>`bWM|*9qUu6(maS@l#YVE?e@}S
zN~uh?x2<87i4uJEp2rU9m<FZ!V;9OZ?mGMPG`#~3wD)c|%-kBbpTaq6q?EJ});SI<
zhU1jI!P*)eQKr;u!SNtnFgLuh)fefMeAM2AhuK-peS(XHw|s*=D^j|{xdqJ_?F>e0
z>-%{x>vAk8zzIzoj=d0e-bgu7U?S{xrBS(8fo=PvmP1%n#McgYD*`SuWL);4{+|5H
zcF>B&R;(04h$=I36kYhm4%1frMKSC0!g|OV)_BA_OLO<+(y}DsSMW*h@3fCRjlvq9
ztIP1UcA|75w|9BLk-z3I`iO{s@MJfK;@mx_J9$%&?%bDj+ey=l<eBx-C-W)x(85&N
z6K;6%;^c|s+1K+s0S%pD<~CA>M+Q~T*Um-Vf;c=VGAU7=&olZYQe*8+&OB0Sgl5y;
zMzG%PJ)A$~ddYWeOL<I-Jr+-bL!~&+-UZQ+Oxs=bl}xK81!PNhg-R?pR*T__2zu`t
zQJ3MSMGmR7otYCJdKApnY<c%Eb)<1|V%cMDiyQBFQ^sU^;xpq>cf~!AeiwdvOp9eg
z`93A5@%sbWkr&}MrSFgo{B6Dq^}c#X>3DPb#xkM@YA5X}me5%SRwI8^)(hW7>U~<v
zD^=z{=zAY1hr<h$!uz+MZtAEc9rPW};B>OI5y`LtDj(z2!|)6_+RNiEz)`C3Ja+Ll
zov7zyLOz@~2<3>XFw<6DL7m{VVOeEz_3GE2@Z8OP=E?tvT#7HZJ+3N?r%{jaeb3>%
zt=1dl+sOGn?gsgP5C<Nlv-sG7Mr#D_;-IJEjVSe&Jt&l#F!hgUJ#bDIgIC#i^oPG_
znRLO)*l+o~rdH}Q{&a&{zsK8168w9!)NA>=;zJR}Aygaew><MtS~z^^l>Cs*T;-Jg
zs)%mgpz<M5@V0!0!G29n_!o@NP3;5ThyK+KYk5SveqWdpY|NMIjm)i|H7*lf=;?~i
zv*5)Jo=r8dc-1K&0=fir{IYG*x!F<+Xc5bGD;SaU-!`}8MvIcey>E%CBe&SGuZ^yQ
z=V`{x{mv;>70X2jg>fCwTRR0}`uyTZJfOM<qP(vk7vsReHjbtQOH84b_)S-)r&bCT
z7fsxsb?ovmJ}tZYa)u|;Z|pqz;HdQzaa!vOhnNpmb?OO+DMd0acfZRsB6s)s0Nr;P
zwyHVbsp#wKB%46=BuXJgK&H}adI_{6!~jh<x+|JX8Xo~FB^={3(3Ac~2vm^n<8=|a
zgD%Wc&}OT7K`QKC1zPO597SI%(;?h2F+)^;h#U_!BwsCO?t|?#F___TdKk4m{R#92
zPtky81)FUlXYg5>7?f}fUv{cWq(R4Ye=+jdGaeSbK!u;DHgs^OUYm)qENLa+vYg;f
z;<L?UU9X3FU<2;GIE(}>%kstkY(FV#+}vPPNMCb&X86V2MxlX*Vl{b6;9P-NWP3Wf
z@x$W}YWMZ$*)Y0hQ7$w9^CujIMrE=|1(uj55<@Q^4a^X<pY^3$GAKFKN+gW~c+L1T
z7VlwD)h(YIYUslcNK?iET#8s#??`C}=##SX1LfQ-aQ!)ceuE39-b4V=M5Oq9=Xh_?
zJ}S6^5eUmPU{VN;O~faoZiA+tM{?;DZ8z~50x4tw-6h=zt+>ZRzY#O__spiVAasY!
z(Qbm99lw@xb1*v;lY~3BwVsk0BZ2}9mhX*pILB=Tl>5I%CrmmW2)sVH=q8@+QWUv-
zbnPi<#_2pK7raSk&4>!vGJU4*<VT>tZIvSUpzoyzF)yf>x8NM`q74c8ytq9H?!LF(
ziJWh}Z?$+AcaE!EpS&*W0zDZELfI`7)o8Mb=<5XPRO>e*0Ds|;=+B3z^V0=S@*rIq
z1|I%qcP-kV-AcrWtt&E`l~z3wMP0><reN3DfNg-;9!>p={UG}GM($&3{kP(|Z{H}x
z)TNVJbEALrB7U_z{deDrrr?5NKrh#;fj@Q_6@sF-xTs>LFreFxNDFhu|GMY@xb6Rk
zKb0b&t=X<^m<&w;c!{E&GkshlJV|2N@E0{tl3j`AalD~*Kj!zG{PDN`hd<L(!RA+2
ztZuFP^<hL`dE*fZHm8+8OphyjKb*^{`%fQE6wy%uFrozdMDM5l<M;pjqoo@|H521i
z5mm+C8|Z%+i~sz;lvbh9IlweJ{53j%|K<PSYfUHt)pEML^ZnOHR^mm5h>Ryd3&0qt
z1H=RAj!LVAp5T8v?_fQMN<g0}1MQCcucVgBhYGm@Xn(kZviI^AhP2q!du}LbF+SWv
zJ`f@X)XJHAt83rrI1WGN0+LFr<ckr33^qU)yvrXAZ{NzlrW~9b%iyOI4Dg;V0LQ2(
z-1|N>5y5Tv;2#WHzdjfu3yw0(#NRVSMd$NaO+%_JoSYd*%7h<oPYW9LW8EO|w*qpW
zbq-q-<zmf638D;&sVtB{lRIepCs<Y#K9K}y@Vz8qkD4;e$qL=*0gDBGi!lyY;Im-2
z?g}W{KaXFa;JOliw+(w!+No2R-GJ2*9ECBE&a+$kUpK=4Ff-{^;D%7I)MyWm`D1U1
zcF`1r%ExP|M$oRS3)OmGE1Uy%0t*m%c&`mws1eQ0S;u=`-<1VGHT}oueLR?;QQi-e
zm`n0mZ!#z)<K28!RDb&LS$AU8C;QoYk7Xb>8hg+y$PS=WXLXTg_z%|;;1%=uok9u|
z)a;+=O^p8gjQH=iH#E&U;BYHu`xq8Q|LXvcm!1Pc(6K;gRVH5jDVpimck=Q`iQJ~q
zkX0Gi(-zn`SWV^vg<<5?Pz$#m(8p4Rjg8HPPFVxNx_F>^)u0^UolhynMY}u?gz6sa
zyT*t&#$I#%U5fu7&Kzz$kU;RnyfWf~&WN8cv!H7M3atmefXrs`dI2-QTVCwmw|k7p
zm|Za$f6}}D+g~YK2YozU`|!PA9}nS{4X6Qi9Ik&2YXu%N7G3>7YN=}<$f$OWl^8CA
zr%wRY3+>axby^_k9-3dZ@G@koSZ))thB*O+^sRHCy}<}oQ}jda0dp=0!U!rNvLy`$
zEP))+1I$ezEmC~TR?p-6;}cLnGzNOZaln8_ZA8<Ym6$`{yBnyTg^Zr~fns8z#^6Ci
z3?$2LIZ)Pk1zIl|Ah8I*t-6CK>-%8zX(%hbd45SCCK;bnjnmfan-()Az>G(i0%$0U
z34q3eyNk6zVf|r|Zk-X3JHgAslTC({NvbzNaYrqQ%lPFv&>@R4_#F5JD19=5d)9rs
zc2>hEN(y2J^VDCe7g#XJB~pRi8X=VfZfQF}M$=54C*rFG!tsE*Ak@0w4%!zIx`9R-
zzO+aI9DZU0gh0Nugy1cM1o`9s=*Q=ZAFHZ>nM?x9la4Fk5~2*=iQg5dlCjKsp6^Mo
zj}*Mg2Ym$<+-%6U%VdSsV8s@&rE3Eq3rDR%P2J!+|C~2c{uGU#pwl;X?p&7j+axA|
zc%}i~rh@}R-xt9%#{$0FM10=;ai3v@GB;52tVq9Tx$((_mh#0KwYiAWNMt3%J4)YX
zsD@k0t5gk>g<>EmixkytFp0gWGoP)miyNr0oII6qnmG^^@%gcFdA5@Pwdri2H6zf*
zu4D0xQ90|Bq?k=!f_~PCg%eQ6(2QIo?N3eDZJWNjn7#S}O124t-_AZwAwc28zkR_1
zyjhTcpTs%HRH@#%Sk3e>VMqi`!>W22%o!Ultcpq<h?Shk$D-Jdf+?p(rF1VQ11st}
zVi}I`aU=3$IFHyhUc)xb6Cjc%M|H=fpUM^R1?1p}87M!VXc*&H1F_lb3(>iL*Y*I#
zE;dg!XW~Zv^e1Ofz&U9}CWCue08<;t#Bj2D&2hj3hK<{nukkT6$db~R%^!(#94HOw
z+EAcUZTKkR_|{U~dy(jk>U%#6!1yy!Kx?2QUMwK7IPnb7rkB$Fu({^PAcuf?h=Ci=
z7CbNDxM74z9q0~vGwhqFaHuH6rws7nzQuF{)P=xP>yb>8I$Qt!8RB~{%E&wp2h^||
zNqjp2sQaWXjP!XX?-C=jI6TP{NL8y5KrLy}&D|z9-s2-K=mHLLrl=&3+YbvoHK*;V
z@lK(A662@9BQhEY#dmkQ9qM{s0eyxjpqbQr{$qQ3QZhsR8OhUe=yaq)_v<%7({ZXv
zlpvopjJRhf{$S{C3lf>qh|*{y#ypj<Eg=p09*4wvp&Bgl7C9&?D)CY~`FRssihpw2
zS^^hGIoEZAMMz~`ybLT=MXE$Usx45O<m8;AjS!+SLCR#4)WXAJ-}2Ok1q%!`czKyO
z!09KLTgZK_f}897)<~KKch!OHj#2rr+Q%0t1Kb;Er$oWeOQb}uA-HfRyxVHGYcpxD
zYb*k60Xr~uZ}f#c(`du_#QOnXghw8VZZ)jyzULRVUhMJy`=wT91P;uKrK#r=e?$%&
zudgBV!)+QLjw=Xm8Hrc&SWYC4VH4z2y*HVzDGwRculojwUw1a0fJ#g}84#*)H4U;s
z7Uzc*P*n;15*@Gu{YDcd;0Ejj0zHO^pMff0BQg=Kba*$OZj%Zb%cYoq@TR^gR{R27
zF&M+TcJF>QkYSf|Knm4T?s_!P(TGURQa@i)<(FLtiXxLN*kGa6?|r53@V$l-pBIS&
z!=d)}mmr~D_-w1L?9X8V;t6&b(&~>{FE2dxW~K#D=YMV>ij3w%$tR|cGLO;1wb&lf
z7l0r}w2d@#ar*s^@(Q38CJF91Q;GXNz@5~zhx>@n1s!H2iyj9FkqfyLTY%e|f79!>
zGjM2!2VbL3_fvWwdrK&-*DSyrste9p2a!8PHe763s})isi|dLOyMl}(@o^<25Va7~
za&zRG0VAFoBQVDk!@F@}hn7J&h{lPZpNnrFjI@6N)MHEcefC37QLt}Y2Tvw)8ZH4`
zG!{74y{Ae91RfZ7qYro|4d?_za>TQw-$%EEw^*C(G+skAQg@Uj4y4)!UY_Np+mm82
z=eJwe2cydi&>}quB(-5MF{V?0D1%c|gA8)a&8c=nHFx}pegN4)P9Mra90NMZM)+z|
zH1h^ba{4T<m}Eo;h`YGr$+j(~RBRoc_|zjde}l|2TFT+WZRDrOSr{~jfIA@%G+i<A
z=-dC<vNAzD^$iOD9crcbqOIL1Q0l1$Op3NV{m*&$^|d$iOgi&yTJsuP*GtPP(pIVW
zB?2@F;?dDkUT!C&XD6ed*a;8$#_JTRW!kmFg%{FC4w5xBx(;^k<w@Oj1{z5M_=^1a
zizlBSTRx6`YWoS?ICw>>)S0Ro3172WQ_v^S&vRQPatLmwXnNZ--L_>CVA*|3L!Cpz
z?hI&Af?0RBg=F!E`N|9NMJyYJo8wwl13+u@=KLHGyMJz~G?1dYm-F?O5=Lf0Gg*%B
z!!FfwA~hS~b#Kk>k2W!5?H3}T{-{vEAW*Kk@+etdsqk^t#v}jd%QDx~2ohX~jthJl
zLcPkmUwGY~tg)zwL-S%W#SmfQDsyS+qt({AKS7Pn#P=%ADPvF>ZsUz)M*TG}uK%St
zdR+LsLQ1A&d$U`8dF6tuUC5oZ6GHf?0xf-l_}H4Rk8wYW7)^p}O4pkhfU=pEL8_aD
z3fDsmLEEds^>J*=XZ%5%176yI1cdD<u+a$g6;C2ko(xzX#%=|fobgUt+{ww@sx7vz
zS*OqO{ZU&BaZpoUMyI^Ib8x<o`tQ$+$VO{q{dJ$e&#b6_&InKg$YD;p5gbRARRxS;
zLQfCGMa%n*XkJz2a;%F7U#E^%2{pU^v>?|j_zRQ+<fyHXNoo(*sDOsdD~G(NIRpiA
z$B$VovQT{`cYViLH>iAlcI6Vh9pPB<b~P;}XYC@FQJ}OMpb!%Eem|WMkph@NDHP1$
z*O5#OtE58)2cf4Ssk!%IxoRvanv8dBRk>ds*R$=IBNpBrQK$8MKga7Y3wO@DRhA4t
z1R_Cq5lnqnJA6QBqcN?#>7$VFi1l`Z{$cl>KDziffL%gomgMvS7cuamwCX4h4D<`6
zNN5BLz}$|2IC*WYlp8`+vL>wC6>DH*2@#`sF~HmGWDt2A&Mp2R=!J5~=kMUE$yddA
z6EZVEkm(`)hR~UM;u7q@@d>Nh-rQ_abHyR9^bq2Ud(_t?b=RW1FX5H{B-zA8)Fx4%
zr62g6nAhPQe@(ljQ(wWA<)&v1+xUtqk|O9_kxju~fBe*o&l<9a?1FmK0Nb(5n5u-u
zg5P;SiZJA5o&~WCOPyvXwL42s-cu?Kpvfug%F4$Jz#1(QHCDuj?xy-Iej&>SCGM+0
z$WG73^3Yq8p=>!oaUff_<WaG+Iz2(xqxW&2e0S?88(O*I%vihPUw>~ucN)x+$!zA+
zdhZ0nfFD7dW^N#D>H;eK>`-LN+N!~{&ji1Le=xQWF5U9j56?{JfE(t6NCrq4SKo9+
zW&7%K=PLg^55K*3S<x12pH#N-ebqp`cx!Z)O)fKtjjL*b>2GExW7FW*R<eE4u6Hva
zD9C)OtVN(zU_MdW!IT*dN+M(NuUqY=Yzdjj_iVsz4c<m+-DKm`C)>D$t`ME~0G>Im
z4w5FzJX0_6v^V+R4Dz6)WePjU#N7Rux&kyR$f~>*Qb-jVkN1Jv(RCe8z8~LPl*MT8
zTjY&jyzl0^5W=vMf<5p9kd!}HB=f}Ywv~3P$7>fLPm~@4uq^o%DQ|B)!xiLiV#yTG
z*j75_@K2`r1f;3mt4(^p!@mS%DrEm$H1U^q*y4?s7iXbtj!A377zo1~aZn@`;<<oI
zJwyV!!qA#J(NKY)9va>^U^uFI`hAlS=l#--ig+Ugx1nH@-L7*x;>l3U^QPKX-t?yr
z9x_(8*Ns9c_cuYz0CW^y=h}&scLV~DNoVP<bYU=hvsTV8vfGJ-N9+4_^h4b#561R9
z9T`VT#c%1nd+d;>svMetM!ifqc*|2hksqX$wRRBD7NY4A3p`S!9r&L^aY&Z=Xklj8
zu_m#1?t_GAHfH3OX{E~{lJr{O46_deV>_RxdC;5b^e^y%9J~`-@EesGf3av85e@lC
z9@fzZ5^DWi2X4oU4kl#Gff4^MF<jUQmA?*9+AI(ffAwBXuOR=4T^fscBivHNzh*7r
z@}$q}^#d~TLtx^nHni{_w6Z)(g>E<4kFCkkjY77^+A^hULBH>%-#Ox+$>N`1Xp%vQ
zh<5y)rava4C+eENcX2T)H370rE19J7J+~h>b()`mnBz*)+SK2>#wZt!M_*Y!VO1V;
zloE2s?3tj*FAgY!L2ec3e4@eSUljyTCgY|y24h?^AupO9$>;E4Od^i-rSfcrz?%Z=
zH#(mYcg^C&?{#XlA9#MWsDtl<^N6JwkpSlixS)ItS*NOyEe$68t_O?@3D5wqlNrDf
zM%=;CX8=|0SVS6NmdpX%??Q;Sq$Z3j&4H|?jw*Q)u>4tt0O3O3F;CrR4L&-LqsC*P
zvkjmfLiLwO?CQ0zYT)a^frNGqKS$d?pVCy_2p=)f#C)*j)7V}3ejP|~)ljMBroo_g
zKTT!JW+t^L7Apbf1|MMv5@i>;LCghZ(HBi5%%{2dPqYUBD4GAF1y@>~6uD#$2yE)d
zquEETV*iuNg%X*j3)w#5_xz+Qpyjr5=foc;v-N3mUuUYYVY;*zK(}FPX5LT?&|HxQ
z>Yv~7L^3itKS3c~NLuGv7~eOL!CC=#2b+)2dO8K4e2IV(8xIH5Im8{XSN*_(Oi{&6
z|LifySFW<U0<<}u=X_A#RSP+Xz2ss|1rc8dx$Gx<z}u`A_BdU-gq%Ig#2!1Qu)Q@m
z?a#JP(QIITQNT962@;qkhrE?c&rAm3-NuDMnBP?m9Cbgh-$++r)rmiN?Q#j!AZ^Wt
z2iK!B6<9^-Pn*Qb(bpZSXERqF$Mth$3k2}5(VvQGsBxyN4n%)7_Qg0il9v|qM!;bK
zo~!Yf4}CG?!Xv-Gjdp3j$ldM9YwbHW$m!~@fuQ{zzKI)UNuaA{+a<Vyo>1L8MaOBp
zH3P?6)%N|@TM;SfPgBr0lG#3L{lKo$O;}=G$N8=#Re0S~y!^t%9``x^mH_^`Ksjv9
zVcju@Kr;osZ5q(U%udoru3;=FeY>`?1+Z|{Bh=RBW!vBI-S0M_Up0%reVh9jQRM@!
zVeOy0hlyb>z#DF)1HzDPfKhdSBJ#bh^K8K1uIAfQ8C+cwYN}?CV=l|Pa2a>T*1Ac1
ziMF5^5J)~Y)dtnxW>PMr;DDkoiUdpwM7;8um4&C$!Jbq%Hu^LCzAitN4MZpjV@ukE
zA^=3Mnkfq(m)u6x0Mwz*jDYK$(V8t>O&S3S&-Pcq6I@%r(Yp=l#K9C<zbAQ{L`f0Y
zn%ydxb073E^Y?~RIc<*g6aaK#8{8`o(Ak_ZyN8)Hw|Sk0#NU;untQmmzZHq>t1+^X
zp{8n)xs3<Ehi`Aj*#CWFxU0q6Rz%|~E)M(B4){%=@xRnF%ODDvKe;`#{A1tPQd9cp
z0s28cT47yq;aE8zMT77cYtSE&TX6-vt8C5kz5*K;wL$<QDkRKTn5^AsS4gO<tRT=T
zpGdk{=W7TxeSXc(mzS&pxL?is*Brv*`M(YNf1ZXveqv$;0i9O-orizm3}p`x;fUg*
z2tL!k7;I}iKU*tq3j`Me`E&J304nk8pp0$-`B{D+shc;EZQ&<#09c_Cc*9{JzMUvS
zNJw<bL6z>O{xIQ|lgHT)M<9ir;K$MV3ABd{Vu2co<N9E$W&Oztwr@RmVKN*}69tU5
z`qu4own3f77EUHb&f+RIhk`)aq6Y~-AC_$?Ci4x5jiG!90BBFi=esjvqA>H9#Z;)W
zuJ9Fr5XxRUSYc3OBU?b!@05sAQ5QH41+c}$MvPOzR$BoHF7!4d2)IFQHsG4_QP`QO
zlV1wq+Z^mtNEmkny7)Pq$Vf}~SfX0V!%s>VD_c9Cs3eJMI)2gpI0(bA0(mIQ1Cf2C
zOcr(dL$v{Sp-7w#Y4sOw0DH7UR9^~WGqi9-UcQEy6;3XwsoijSYVPG~wW%Sh?f@_^
z0+^*^o>op5t(Go1V5vi*T_X<1Hl<e7f%_+Om2ezN(3O3?{q6G1Ebk)$@nIawBD0Db
zheC}czs4T`@#^|?b6GM0xexD+Yp6E_RGfj($o^8I>~3|V%C@L~kh}kIZIzS=M<D(^
zb(w(dd$9(`QZH&HD9$#OD@rIWKg{7+3WvW9PoEAQjZm_@0)T_%))YX079G$ww1ozo
z(|5#Y5ldXNw<u4A4h@g}f=iP5rtZwg|2|Rw`x|c~Nt9VqS1+>vPf&D-69K<GSd$j7
ze1(C@DkdlO#luHybW9Go<%M)inlXSjqjP$+!>;qmiS<F=%Niq4l5)~+y_GpKDwz;>
zQ^$(0Z|B(?OM0f<4#5?#bAV;;_e<b>kzq)oW14~E2Wuod$fzX=xt8VzFlc}qeHjFr
zmBkm1!xp%KFVUnDPzZEh^6aVL8;590-NcU!VLM`CI*gVg)Oj={kepkr@Ti-X$tenm
z_)9zzrDsa+#*hwRZNpi=CNZU0U>1U|jBMa0%)}if=7$o3J|vld*NcH95{bivjaOb3
zZ<ry@qHLR)o7lFH-eu6)pCN%!_#M?D&v+c%ao=*-1euwhF$WhNk-qpavLkt&f4_~?
zI{2lJ5s*~e2WSKd9zL-~&4{j><-qbgDb^n$=mZ#<>Ee&7fy(oKxSQwRes-77OY{rL
zgpr<uqOQG{Y^Y3-=0qIxQcwfIGCIrS*jr3SH*`p}9xWv~UjTH}(Do5$RIcdI{@^NY
zXH%3k1!IkSt~S_gnVDifBKSO-<9i4h!j>Sg$fQvnFt#kfW-T95QL9ucd?aM%h;pR1
zm%CiqFk=1j-RGbfr_i$^U~8Y5Ia^C5d_0GwT#~r_{LmlWsi5$Fgn<ZHm_rdRQ=J!T
z1BCz@?w>$h@;KU_W(_kLoqJ?L4;=QQ$`fw$nyBvO`j>_{Ap&8E2QM}6-o4V8B*CEj
zETV&=Gvj_BuZ&xe3s;M*&lLn-3CjQrFXPUw8lT`G6Fgme+e13RR~A#HEA?qaZDVlD
z4#=`E`SOF5St?;2AAjoQ%lif4Er6R;L1~FTVS0Ga<S0_nW-j7#9iZ8G`1zXJc2UeF
zZ`H7+uCMb`KGE|i&yD0idL#JnBj|oIP&gSQvyU^S`FU8H*do`$;hEU5o!y*~RHK#d
zq)dYNZ#XpJARuHYbp@qx<Ebj!kkw4o;f2>B8JX>wXYf~I5AK)^-G<OoWE$G?HX~nB
zSSS^l-xw_$$E!ePGfH247_lrG0yUXW;t~M&aCIoY`fI_>UB%KgEAh%xZ^7F{vl5o5
z9{L2|Xuc1dr{eRwhVIK)?qagg5T~a31zGQbUbLT=&ifI6o+=6pxHAlo+Q{#=YynFA
z%^!K!r3bP(SzUr{sr*e@R>3XyMeG`29_nc%`9@;CQZ7_VrK)&jZV)oyyS4|`B0(^7
z7e5T`Xwq5)gLiv0ff<H~OBozZ)vF@=98oM)Q<oco`nXLhDHpXSzXa_LK+UXSmqVv-
z^1;zoA(f~L!=R?}mmc7re+^MFl`cWu#uO@J$@Qbr!wDDa&En*h=!^Bbjb)@uIGA;$
z+0TA!0o)50Qj(2TV_>dESx^`nr326?ki}A&dPx5H-oX7AoOGRVWJ43ux$W8m1w6D&
z7V)`_t{W|Vy65O(cMJV621e9H8Lu+PGQ)f%+ycua$tUz-gl`oJB45wpw%f2a2iLgN
zzr~W)@-Az>HH<@cWK1JFiBm$^GH7ztujiW66&}jgYPF2J_n{?>TyVsH&un}Ic@2K=
z{cztywrt9O5VdGQWHZYV&V^sAY8TZf7rq<uwgK%i6Yg|VCLW7;oISjaPQx=)7qtoI
z9}Ewq(w?Fmby(}CR$judtC$jQ@qq7%!7G4laO`3Dp(rfGMYIK}iDDr_L|rD|9|Kte
z?n6xuSLp#Z6;7-yzHy7*9ZefVEl)3Bv>=R9W5eD)B6C2K!^0AftZE}MPfSM@Z)vkr
zrl!gd%eK9Zq3iH?`$h5XIAKbHeLOE4IIZukitK$P77t<lSv|iqYDHzWNNF3mt81vW
z)7+64U&N*oMI^np-}!gfm-{+Q*g(Z_eb*psng%Ow$lf34U8`gvA0-VrMB0mN%a9L7
zSB)Cg9Yf8a>Xfc!p=pTH)w^0e0+2?_=lmt!iMx7)(F=iPRPxHGUGH?4Na|+J#E2u}
z4T}-)?7SStNk)CkV%e}7@q*Zr)I(^|_Q3YxM5aX(7i*``#97tyXqLIlZ+VbE<HG;^
z#YPI~v;Cgx_e$hfdT<-!=d7Th%I@g`P~uquyBX*E>TJCACayGfPof2q;C;}H_2$0(
zGZ<hb+)w2FvpfC2i-&&w+=oC8l;{bPPviuCzLV$yb^la=H?Rf_fehO1z>4ogA+?|a
zH4*F7*4&kt7_Tg3*~D$o15^so&++K&;{K7uaPh{rtRA;NugHH>$`R!OOYGTV`-fj;
z`SdR^{c&lQ2*KZ^7_AyWiqUd^akxee*vDF5ven+OEErq4gZfuzvBBr=8x*z8G!*d-
zEnE@6a5n?830?TAL7!VAe9z+)XoePk%F=7_9Q^R4=)P<hx$_h?1k1B#Ly5Qyh>Hp=
z4Dx!L{7}2BK5dPd*v^Oj=Lz`le!@g6O0934ycYT^axe?OhHM9<xf1}%Wm(c5Zr%hU
zKdItZ+hq3L%JzF~gc&k1^j{#maH#fN^k5Z4o4275jx}zw+<f$zH5MFAy5@OrG9w6h
zEr;BbL3L3Zz;@u$Z&)hrO!xqJ++Fb*y(x45BvJfT6PJgB;w*isPt@;8c0dL|gu%%E
z!V|!Q(iCv|MkJz5*kG`}3u<z`A*8sbS*Ws$JU{lU91&P-=wgn~w92e!XY!FwXQ%!-
zwP6SaU<2l~ltdW+{*d`ESLFjKuoNGSx#0eTl=#0Y7yj=9{O<_-|6l}s?&Bk31Awg_
zG*e7G;5QWe2Q-}j%v};i43kU;IRuwcM<hzM)?0Lh1dV6JNdk^$esumm|9n^ehoe}u
z;D$Odvt0stt$F0yUEmJ`+~A^#?>ysvAWn4q|LL}4@&P=6PQA<h1e6e6@SxO)$@kox
z{@*>O$?*F)a0b<jWsc?o0;~)I(r>2R|K;KGuNT~Zb#VNLG5VLkAbJ7uwJr}L8UM>O
z>)-y+qzS?unv(jmeleE*<@f))Z-+#{sk$kGlZ5zx`N99dME>{V|A+AA-;MUau7v;Z
zXWxe&9}()X*@wtuS!3_yNi~2!+rv{CffL}7)(L7IF%C@579dWG2guBdC+}e|AYmE*
z_%<$O{jpyh;|t(bX&k!WWr%d*cwh&Dj5sL6OIhIjxtx3iXq03`NaTBgUI#O|fI~2a
z+2IAiuv8%WA4DQ%RDIV0v1q1i9Ag|uG^FD-o+aInP&Ayi0VyLF@ZfuIHCb){+;sT;
z<Hw`@_LjIVfaS0RDAE<Sb3l=f^+SH?LrW<R2@AlVdp*CgB>Oe|Dn`OQ@of*kB55Fz
z#NT(d2`1vvtXD%Vz}4+Q0*bNX{jnzg5VLRfdwzDdaFKOypOd?88f%Xw_B(A>iPWo2
zy{l+I@n~Mrerq4u%ilz)?LHe6>KvOWV#RU<$jV(HL4LfqwX5I^=rGp(u7fNWw^wR*
zs#hsaz&~XQX8uzQ^cosBJ~YsfW*e8?Qb)9Dd#GqFUx6B?ODAJa2apX@<FdePX?sPC
zXQYy~6y_;X4LTRmBuQJRz_hT)vQ-<Ee0@ybE_wwD%Zdwk!F;6wrZnFk8O|Rk^$?EF
z!_xs1Gb;Meq1WU#M71k^Oqh28=OY&2d}K1D<+l|<a`&MEC|e})SiaWJMD+^BJ3R#@
zV@0U>o|wlXQU1|gb-LGlbD)AM$5y=*{r)KeJ!BpQjrtZGfbpc@qM-Ioc$jF?Dt#M@
zNI?NH*NG;Hi~fP41k43;{c{CAFRE?t=h@ABT_<)AT~Pmar1%8VK+#}l+T|g`*&4*7
zhy$S~^Vb5+i914Rqj&HVYHF+aMJD%L>rQ%_Vm}yxmG?0j5UqH2P5}x8>pTJ(zs7cH
zG9R1*;)p?}@144Moq$>bLDKlBtE*9*l@)Wo7!zbiz`gUieJ8lx3iy(rQhfaJcCH?P
zxte-H4r@x21fK{a{yHl~C2p9gP|pF7kolMtke-A9iW0Kk(umF84GAH&9LS(HoNWsN
z95@2vOv&W*kPHGQ8aotYi{2cqaom*j%D`EWjSRtK=)60sZ8+(2j7KjA`@uD=I8+Oq
zzV;?NX0I-tt3l-rBmCfx147gTcUO-UGGp3b2xTV-6QY+q{jtSN>Avd|(lSt)x#M??
zE$$|_n+!^NSZ1sQwvu|6paP!sn7akR16NY=Sb(q)d^*4b3hKB{sIA1!`rUH@ICaMY
z)PHIpq$k^4oJK0IcO98^Cs-QJ_e(*h{Gc!OUQF$9OGyyLiw=4$58)BXEE3Ke%YzMo
z1=Lw(ySV=p8DSfElJ18OFGfIIH2>PYD)x%8ry$Q^Z#n@uB$MZa=MKos_CcQW1xWtT
z7QZHUG$G_*D||efd{CyAbmu>Cx0lf%OVCoFtv|*`rgp@+lR9GT8pwY1Dwan9%R|TB
zT^IG^iyqJrsKd6^b21*@x9Js|zQToiApLUxh)UPb;AX9J@?G<INK}?fzww&bn;*bZ
zq7(4L=!@;1bbkjNvK1<~!!-l@NSzu7#+AFrd%zP}n?vNKt0j?>Shc(zXq+g{uU>i0
zb9D^Z!{ZO{sO$9fSsPL~0z<=S;JnEBRd6K#;<lRUXZIsuOJqPf16<znN<lKpa4V$!
zuN*Di0BtQnOn$}lbe#ce5D@f0TN}~y@)*lfI9F{-{feO9x>pN!+%(9_qKF8#GmxI<
zu#=^rqcCIN@t4*?86ONiz6FdVTmq}Y0`>q|^%K}GjC;bMyy`@WBKb3b>vuqQ7ZGsX
zH_AagZsE~u2y~$IJkI#;2BpfMz!W9F+}%Haf30sFS=sMLsM!U}PAbH9_-gzu#WU}<
zGuhs-%!hY=6OGPl%)tr;NP_s1HDIK8*tZ5snjjWzPCOo~qFGWq7?U}m<FE}d<k`k0
zvpdIj>+LRb47V`$U4hd;K@aj{V&;^&ZrL~JYAK$Liw-IyZAYgWiZdrB#L{m573*7-
zvrY^t4$-R0cD+R}VIN+}EV^26yH2>!xib8HzxN|U*11!OnopW-fzg@vhE|rx&&d}M
z53@0_chf~JebF0{ub~Sj|KjK5*P2_f*>lJ^BJRjL5^LrV>?eEfZaG}%2`a@}o#gIr
zuqbXqslKEZ;PtB)mvT`^Saqz44p3D#7_=n$MzkH|s)CPP^BG%^g)=<3NR!v$K|9zD
z-)4aITn=TKpiVPiUqhkq;-Q2~j@pSzJFlb&di-hBrykVzxq~UduP$%#Gsa&uekxW3
zcQIwWZ9p*YBQ29cee2@$zH3G8rXnVpBbtAv0kt+}%j?0s>p3mQ0id7S1CGs7(sJYa
zQ6{{Y(LU`6VlDdwQrmz7k+AG&?Eo_Ut|b*mPNS=DUkYK<h+O3z!jZde@RV_)5^7!d
zeS%Tv+CH#qxR7_NN`;0|&SJRcb357T-NNNGHaF$X0W6A!hl3Hed7jECPui4moTeQ|
z=k`orUV9FB<+7xbwZKgEwfx*^mHz%+&7HSrm$Yf)AD{oM7Tw!zM6xQZUg5fe-yC^w
zzU`{{VTzdj?=v-s{7x=uVK++8p}P*yo8#+qJI2fUcH@L-@vDt{$P15F9j{p`0A+oI
zJi6a=1w(k#MJA_mk8d;oqn%{~8`3xZS9D+S-L78Zl$$Rj8I`vsj0Q=a-GN<yi|wfW
zB#v>nNDVQiurQjJ7r(A>$g{)qCS?QgMo%ITaX)^EQx7$A{d%Qyq7A)*?}E6ZL`)gr
z3Fs)+M7ym4n5|Uz0B8HBa%X~=944w>>s)}Ls(^bK0eQ>mML*CEw9SrFW2zzqT=S%}
z+Mg^;ZcHC)iNKp+3iJS>HqDo`tzh@W^FabO(=O>kBgF59^Xe3K_(HkVL9bX=@Rzap
zIbFa|sVzHmZzLF@FdwIm07^&KNbRZXS><8s@~76-%LJo<WvQ*p4z9^<fDjife=Fd#
zN_6?Y`~#3~*^1|MLV(Qx<_J0p7KzdPyhOaF6NpC4lLQxkcOpT^r&97Y;co|q_6@D}
zFtvrek!;t%q7~x2iDT}7s~4WSs6{LBl*6$;W`i@yX<Ns&mv7+3{5?%An@xT=!iics
z2=+DEYN7`_dLCDEl`avQ%nAuyY-{cALe%-8p8!~N>$Kg>7MRS0yRn}M8i~AoGynEU
zzEco)BNvsL&D+1=JD)Zhr8}Sp&aTaS4bYLf@kTtoS1UO-&jH3Azg!APEE;?q_*+F`
zR)UTjk9b}M{^(B>Zv=O7wKr0LH&)7L%%8nCBx#VPwr2m#+bh}9Yc^oR)13x97Hx2s
z?6^r6auY(?K3%8J16G~%kWO9=6j#*&rY^O$&R2uD*K_aVifls>gaD=8UXSc;ayd7(
z{@wNVC&wUGEQL6R)1d5S*|jqA;*y~2=y(qht(qsIA=|it=;wFEMg8xY4$=joY570t
zD&$`^Wwn1k_1A*(Mvg|ClPCpNr2>2WB4Nu{+<m*lLbSJXGAqJI4h{q!@Km1XzH@jK
z^JC@QqJ8!I3NI>N0V+g4(gt9j{g$w#Pj<4`6^X|aP4fd+!0GDFW}xV!qo!IJSzzuz
z-c%4+&r;z-?+w_NzQ6*gd$0EE5zs*K`f7Kw{f*Y|o#u1OpkyyDiuu5xt^;Q*a}rO1
zFN4muR;c9$%*8dhh&6Nc)FaWr%1jyv*{xpN`Pq@w6Cnyd981&sS?d}>7egb{6O*69
zvr~N-2)67j*U!P4t_Gn=C6T%J1LwMW^(s)v1p|1t{}mXRm=5BEBHvzZjW-c<&6VZ_
z)AwN#2^CxJm1jBlrG&Lpc3u4Hup-A%?SrzB>N}o)ukr*SgZUQ@T6Mj0CzuAo{>P2g
zNVBy0ui#902Zn8E^FZJ>Py2apnM0P0o4rTH;=zl9H+~@_NWryx)VC(w)W0+H&OVZZ
zjTPxNm=w#`T!H)d)(^u6hS6|%ln!WRxKY%Ex;k9df05h3yMoBs7r?E5OVFK4PWmr2
zBMgW+S!Mb~(mU74WvtohL)ENcPhx7bP7<2dS{F!fFtcm*AuzW)2tT6)R;8uD&YGCZ
zWmlgc=rd$HZjM|&vicw{I#6V6*AU|b-2?!Nc=d`>WajcPJL_j)1TI$1+D9gd4jMC@
z+V&7>t|N+8in2yhUZjiZLq3h*w_?8wU0{H`f4+!1SkcYZb(8D}6`q4jdp{UESSo;I
z@U1isq-4FOuj7EGzr$)h2?XW_vN89=^Q1(fpzk1j?+=2fs0SY3R@qm{R`2T}N}=AF
z@g4i&IF!QL*h0IFwT8zTUAE|_jcNs&_(0$N{dy?GpBgCQhbnxrV8`^L8fcZPGXq1^
zvaO@_uhJvxhGDe_$5HP@?o*1G4dwb4+lT`F1K?&N$>1u*;F6!FhL8!msUCx{gk70a
ztI2UVt8#9E3)KM)gH(&o^=pylK^rHaipAFg+guslZ2Q&T(&c!qm!4a!HAAIqDtJ%}
zsRp2uSZe!l^#c-Hi@B|mNLHlacGc%LAtqp5M9@Y^6zYT3_ROq6Sq(TOHmPp{srrv@
zQb<EGto)b9i}pcIGp^4EicIe;w*q@zLt`pJ6C+gD?5CV3fStucNt?6p1o(+=>!5{a
zreqCA8!~0oI+4^+GQS2y3Q9nTl9nJs<n}nmb7?e|Q8lx&Yl|J!UwLMF$b+}C7|*aY
z?mr#qx;O{b{DNs^gCdL4FZ9)mp3-`$_~dV&=I66B*NsjQ^dXrM4;`;!U)HNy)(1V?
z2SKD-Y)-}w>1w*HkVdr^=(PJ*jkuNSiD_)vWOmgz<Ju^zs(K+p6^#B+I`RqfO=LzU
z-`q#0gObSWgA}5oS!fSXakGL41$}q>{S)4+AZI<vV|tm+&8TuyX^(vk<3MD#);iw8
z>6x<=uiMUyo7ai^cFOp4lH9&LJwNzQ!`-$L^Z^UB3E4gnX>VB9M*d|p4s5XXE$~EV
zyNN1UsO$kf{KV?w`zZTa_4+5P_TB05s7Q^kO3vLWZux2NXfK3D+BRR!eqxLO=>N<#
zD5jWJT&&jA#E3<V-_<S3`O3&Dd1g1(zL_PsI_63fHaalPxdYDEzLSMn5Ev#CpPn4N
z_cFM=dy!HtHF=-+qGq+PZtY|9=YGZ4&Xc<(>n4HTy%nFVPqzyyXZ8i;Ca2%6y9p<%
z5<@3Yg}dH&sY1^(7SF@d7zV{3A{nDK;FRuS`|3<Xh4||gnb7z^opi+YzIxQtn4Xz?
zTvpMnIr3ilHYHzH2iL|l`{}Dgo$EGrJheE5US=H~RHsObE^y2j7k}Alrmx3%!^hUm
zQKY|2x9VOA(mDeAGJJZY_;-Hnd!A4IOdCi7Si@#ddYuW#o+lX-G@O3My!-I$vq0}`
z+X*_m5>VE=3CImkIKfFk$TPpQ*mO1zSbf@KMs~Kq1f6h+A8Uey?^b3X^sF4hXpg?L
z8fzh0g9tPG0`8&zcrBZqYq6#obKYY%`NtQF=|<~wNb*KLVk<}Fa(c&e^{fl-?B3Ev
z#B8)@)%bG2G$&=`Xn-t=dUuReoc#prXyl&LqJg)iag@Zuxxala^Td)22oPmmG-%`R
zJp<=dGXdAlcScO|BOqs7T&7r}-M+Kf)ck6$w$=cH<D6gXJ`m$=Z3Rtdiw%lh(S02P
z)c44)4s%5CFX97;{qMFCw|h<5wx66ATf*#=&b8dFFm69Q2qSwQV{qN>hPGu?$g2MR
zyn0=#qa*_hgHQ&N{l}Fj&p&MPvF@D!IdS#Zg`OLi=^GDB#jkc`g&2ybm4-VHtnMM+
zV4aey*6J*rwCn51vPxR<w9-a-6=2~Ly^6#zGY+~x2Puh3ITZB2HM`JXelSB?sB&cc
zJ>Sqs{(#Mig_Mvj61I&PFP4Mzq)@G`hWxX$;Qg0)16zroX7`NGzX_}wTGlG0)-Xos
zoi7l2Y$~AKeKzWHv2~KAAduqZ=n5^`dw(oiMZS6eW&&H@i8Bx^6zM$PZ3s9SEsN+_
z#x17wP<!BBNr3<iV8`jPZD%Yp9D*wrIiHJY{@}3h9K6lEQe-)AGj4pU@Ynt-y+Kq{
z;KPYbr*A=sVKtu|QpiOsgn~tc{u-4;#9*?3I=z5|vHpCSo!8PaDC3$qRZMpUO-l|c
z!kdsOfYMYPgBJa16(U8UJxPPqtJ*iKk<Y@3ZgPqB@nFeSv^NACa7i>m7f~_3$8{7u
z?V$Oz0;~PDaL8kVmf{s=5|NxACW0!>DGM}ai!q|)5~8Py2j^<~EH_S5ZKh1#hSiEG
z5~t{%(Zu!xcx)f<#ytOg@E&=?9^{OBmtxY|%Yu0h<G^1ly#u`uUHWtm@DH=Oue1#*
z&^-iR^jOp4-BqkfQZJ6%z5epFl$E_(wM36f^io`Rk#;}I729h3zTYv&DMqKn5<2}8
z(2zK=0D7|a0;Q-aHn8qH#u5fP&!5>yM&WnbgI1f}$+;g|1QvuIpvksEDD?SjR1?MV
zT5N(2MWSy3Wi9OhH2IXx?SccH^48agDwM}T>tk*A^O<xEOiBqdBF%u)am!0fV<&?d
zUfIgE383nrIXG-RO({ukDr1zOclZ>R_;JeqPM+4g>NL{gk3IN{k-NQI!W=kD*Dd1<
z=8jNeGq&Hg1N~G6$+vG|!QQ`LjeIovx?LB?#vtus>Yj8pWWnq5>j+!nuQx3G^mDbL
zqpcZtPwyOa34PkO%?}&>@`QJJ4-`F*3IxCKJ7Fwm?!EYEh&i*!l%k+NvT0lP*D=|!
z8^F$8{=Mu;N;4G}!rAcifHG%un29d-8$N%}h}M{j81diBteDigwuCXIo0a)*kCu*x
z36@dy3pUWe)vU$0V7mfvO3Q`jYq&^ah1K9TzObGqL9_ympYjx;>Xxl@)X0<A>746r
z_8ghT1=yc~@IDhWCeC1p%Vc~ZN5YtAbZ>>J$qR)J!uBC}q<aClt^+J>r{;=5c^tn&
zC*6yJLVZ{N=0YO4E->s*W@71y5qakOQ6bZ7qDtiVm<X24r-(>!q4q(_)<7eonBB7w
zhE(<RMplfAL#f+`y5NdOjBv~-LYJPrKXlc$b30w4V}LT|**;k+J|*Y3W+XD*j|)5e
zv*3ogc<SeQHWQhk&~8Of5FC0PMr84B{k{T~-t5J;bKR@M@*<CdWfG$p&>B$5C^HnK
zuhc?weF=RGSN+S`$Fz%l>W})c4(Hee%c7|pBB2DwVIq_9Rk(^G#$p1I=DQ;aEBr)=
zc33_kS%)OT3B-s`wTaq8RzFi6M9qLqF^7?jz#jhMOM~UzZ*`T~rp$McyOodZM{(2k
zgfG8yl{7Be#k*P2J9cp&*?XOI>Ai?Zv(+hO^~Ru=L?db~1T;u_#{~1F?mBRpykxK_
ziT3@yoWQzjJf`t}ZbjhCIU`Zj{amQIoy(AK7SQMfp4Crsy0aV7wQ1LdSD%sVIU&P!
zIKa(IS2wxe(R62ZEKJCx*3(@1Ne)txXEVpEQ5mn^k^@PiqJ!RSj9Ne4bR7?(ENXph
z4+7I7b6AJaH;^z+mwEFaI<B1+)@;0Yy8YU-YdV5Q>e7ockraD<F?dE<?5_h0MPA<o
z3*n2)w>>t3*u>n3&p;1kp)yNi0JnmEvkkrrncL&0CaD{8kRk46sE2%QQmZpqNZeE(
z_sMwbj`cDTGDwzOWY}y*k2x0TRh1B}*hNph_PHGRO>5AyDeQU^c`j;HC53}?$&f^V
z5CIqNvnQM035P3Mx<oeHTTJBW!;eyS+;dz2`pMk1v=$2GHuv@LPg!0eeU1DeLO%mh
zB$kteNRdt<jpU>8yr3DV=f*nYjon7Yj{L#n@70qb*B4IOc<TqEi$I!K1Enb!56uWL
z^oL}sQAg&hK$>|sdLLrje{mxkT}0!+zI{W(yYYeuF|g4|nY!0T+qM<|b02HFZWoQD
ze}=A$0^hqp@5gp}X0TwnwomYd`QYe#*q|rh?CZ`D>Z^S5HJ&Sr&-fw{2M%4^$&MmV
z$k9%=R>wA55c|3{wyTqd7*+Qp52}ey-+jOB!nv%0j9K4RXy7Qk6Z|iW!e+0P(2Z7<
zU?L!3agV^i#G)hf=>2tX#<&{8V;enQ!z@on%!_b#iPwVU?HevB?vAPy{4OgbpG<2n
zUwTDst~m${&k8i)bh)IIt&j;u9PJ6%XV^~m77{o%r(+`6Z+x5}Ml79sLp~7Bw-Ag-
zz64N@@fHMnzEFd3o^IUHd9GYfZqFZjSEr-z%1YBo?}h#C`7om!l}HgwRWdDTNQ-z&
z2Ab*LSS(y;q;C+<eGPv$D1v2yN{G>yyaR~&>&C1UVK!a$vz}*n3n%scurwD=hH``-
z7mMG*283tTE=Qp|cabqCaobJZ0A{tAOM-Ve{~G8tWUo8_>v6Jo9K=v1fpf1cy&<&n
zp46Uq)Lb(L`?tEOHk4h2N8$Ck+jQcEmx*P-{=6sKV+)*qTUbxkZC@qKfTdrTUNi%G
zPtGfF`S(f390h)Gl@x;xjq;^UL&`Jt(CE#Wv~!~_N0)t!`Ffy0=z=5}*m#GEY)qFM
zI}3C-?_y@Wk-5ITUe3~d@N}nLY(@qPA@9?C<#8ejVyTYWx!@}3f6{e{6?vx>x{KYq
zbHDNg_>POr?spHtIi64VJ(hl*et4Xb77~!Woccy4h3mRT?<!aIn~~{aOUK(38+I>x
zz!gT|Iimd67L?Ra#)VBvfTpDCs5`Y~z?|)n)sEUwxJ|%Oe45ie_`@4cl?KCY&-MiB
zo|JGy$6AGY5ZRCk*@sPY>w0}sa~fQ)^)TLf{+Uac0hPbKchRFZHT+EDn<fKhC!8Zk
z_(KXcRp|JD!JURK-!=6kpP5C_u+5bEVuNwVpcFg8rYHS$04^epiMM!YcL@@t_Jp;+
zWelZwL}n7AuTB6QeL+BtCWcl-gzPUkb?@VQTq$XoBNl=NtHwJr8HSNyXHd?ul)*?v
zLL)ls;rNWn+OS9<hy}Q=Pr4kJ7<rKCgfS^p5%!!L&QhEYB0hLsg3{23cjjaJoY-8*
zuT1}V(>8V=Ch&I4dzU{z3P%9&c6T360Teykt7iYk{ExjG_|_#&s)hBrz0e}yi?xa4
zdV`-uogNGh@aN|hrTM+xakFr*9K*tp>o;3Wj~56V#vZlB^5U^k7pKz{A|)?s5M38W
z@dJ%I1-zbbwNr~(r55AfPs=gw*6McBWW^X3%3SZPv}onnkt9?1h<IXFAxK{ot4dbW
z<Hc~xVY^1}o#;+>c#G@ycX!Y%XLo1)(L0|fy{qwG4%a2E#^cu<8Lo@JcDxDBx!*4$
zQsjQzkNpuPLRzPRjftt(ZPr~P`apxGUQFeIjpB>^JAn~9d4l&lIajuw-HCHPO`oWf
z7sce%EH{;dy-xJeyjH=qmRw-nVWz(H=%WE~QH)=*;p(2SuA(#IgH@ytW-U};t8f0?
z>Tj`Lbd}bJ6*vvJI+LP|N3@wTzWGoOGVn!Nhrb2V<7k-W)Kt|4`>DL3LF+W`Ce)`#
zcsVx0%A<Q5uMmDiHNu<dMm$;sfVkZ;EGmX%i*5<0hNg#5OPvY^5}W24!j8Xv?2B)c
zt`|8Sh~T+YS(NafhnW9+9i=j#!3kcz_etT8w6hHrOljlL0nx+PkOIrlB3s~hAa>%3
z^x)5Fq^p0JbOBi80kWFm)%%k-Y+DKalGTs*6B?#yVp;;dhx~wy%%^3NC8Ao+w;g%_
zE<vw2ejvdcgFZDWtI^We&6y|a4iy1>cvl~($=0#bjzZl{o+W7e<Xpyo&D|59nY{lz
zcM7k$^Rl<0D&M^FB~<^d(Ee;x^tXc$+}^as03&%Z9wD_>qL7>nz^1zF7siKeP*qZz
z$gyqLmd>&L0e{;RII9;ODg`M&EIRH#suZf?LG2?mbP5|~46FPsNm3NE`NzybrMWk4
z7qwoF&40~ZdUa<<3b}Qz#B`w8kGHm(Yj0I`mH@B*re3D~U(2aOL4OSk0qn-_Pr=<`
zjT@_^oeV15twF+v)FOFZ1kSd(<}h_*ljF~nmmJ}}X&<p{n2I?wx<$)7t9C&1hG|I7
z#>5yc)qRrnkH>PJPjp?EI~#`8wB!T5SzEpbKrzcJFGR4hshcXQz}0JDD&kk-nkd;Z
zOgNkb{-g|wE>w92w5k^8qvH(=&nqOg_gYO9SR%8aE>63<tX%EfU*#5;!+|3}HX|z7
zSvH33#<(IJ?OBL>21DxfTG<o`i_a5MoZnn2JY1V{6v;uXW4W4kXOL(dDX8tLy)%-u
zWDt5(c4<{Ma9_SPN&#l$@k0J9*6E*#Ka>d4qd^5REbCQBjzzD(J+pq-89?x7)AlMS
zQGIK1f&l`C$9&;6hFN4k&raK#$&`rUJptGES`^xE{GNw%ernr{ZJW&KCEDBi;3h~K
zzbBB_m)E|Q<%d>gCu~pGzPU~?@y%VCVTlXbXq4EAiFRq^dHEgW^U2vuwexjFFD9E<
zm66JlPZIzi(M1htc-DVmk{k&VG+=;!w_bAfZ$L##V?QZk^dh*33LaIt0pxcl>{^1B
z6~FZm{A->$%mq|i;@ZNR>9F!5$&q^*EV^nT`5*37fZ<Yyy&E*)-~npC8y{pB&zf*N
zm5NWZHzHWE4g2oy1DQKLJt#S7tfK1=qo%s{{-!9=#K+8-$PeuA?}UKx=?7Y+*W*r$
z({Jx1nHwJO<;+I>XhEF3fS3Z;Q>pyqRGpGR_r6eH<fRooepOCFDXecH(NM+5<-Jik
z23wVPnt71b!K@5g%BStsN6$tBRcz;r<ye-U4d%%qg(vq(t=ECp9iF-mAXjs1QX4_O
zxd*AqeciFZZ#VkzANLTAcaz?(Q?J$Cj(~Ma!?V&oa*8=hu<I&BH(t#z!ZyOA>|Q}3
zaND|-v#?tG=2D)OtBoagDu3HIk|I?SYgA&yn#@rt2XMi8f{c-;e5>rjQ9O>q1(app
z+q*525X65@jMPdS{sM>l3DiymUb1g3t^k<J_hdl80Gge^(&`OBcA)#$*-7L;Hge-Z
zo_JJ@k)~lA6Ua6Ha`S6}4=9a(8Fb+kmisJx_D%K07O1lyAx8L9-`m>*_oU+6JY!hL
zNPw--{CO~KxJgLSG$9wVg|u`-$j^YEy1#&WE`U&1eWI^_cFJWjj6&|u125Kuz4M#)
z36|-a1KC!{^}!&_&*-|b-H-94hR}qtgZg{25RVR32<A1&?B$r?iugMq9*X_@RB#J_
zz{*|_Cq`U3fWe-=Dg(kVarY;~_T4~Dloe`UeF}OD*x!L$zRemyqxPJx`%4R;fjaj#
zxd-dt0Tneg2&dNFHngw<f!UMG9<DAHtP_s4c>ZmmH)#1$_c<n|w373ySi_&`u7dNY
zgZF#1=fH&LDt6|z8_Ng?J^hDSaSD5y?zAV?y{v58EZoEPT9?(!>?jR@E~|{>9R&t*
zxoe3BNiF;;i;kG{3qRc3h@WF>n7MTuz&gYw2*^CQ?b8nWIt%(Xj9?uS8bh@J+2P@Y
z3H3+iLwQxW%KLQ88p^P=u0m((uz45I!O`}wQ3_t5cmCnn|6%W~qoQite{l&383{oc
zT0rR-Lg`MWRRp931eB0w=nff5TBM{*N+e{YOF$aw92$wC8R~b>^Stl-Jl{Cy_dVyo
zv(~#7x@OIineE>9zV~%~>H;9j<Ir27g=GHQA(xqi26RTq<AWsIg5)oev;!-oL`?Z8
zFhv^`XMip9jQS-Ahj}L0O7)=Pw1?Wn?%kK$+x6%BC51nb@x>scL^7s$!-f{OvNexc
z8H=4xlCS(C-$n@c*sgwAdA`yWtlD7gr<-{<)cDMqv(GeXtY(LPpl{WnV5N|OK)S4f
z8v4m!+27*h3f8?_4YsRj@U2B@&py0=yw(PcQ!_gG?!xacg@D52X)3tqj&!Ko!tcvh
zK7JB)YrfrGYU|ELW>R0xy1%2fhw2RnerQTm*rbL;vYp(klNv-+7ObSu$i7S&#aN&P
z9)6Sl{Z!KfG=^3bvQM!}8IsE1y8kLdu+6#|e*EtCnke81sIeI$Xn_1bJWu|ShX^qc
z*#W*`{owbY6mcs4|1CupC|=mZ@$(l{o+16g7mMuSR4lJ-1M|?&Djw}tLsG?tufHC5
z$(}qnJ9!;&&(ZG@fT5TW({$2mee12Nkn7rj(7JNv;$&ZKFV7iwr3fYB(sQT%%3m8M
zObu7jhR9$My@|C<O#>9Rl1RgAm4`$xJW#kBrr5RUHK2A`5W+nGvm3}yS-%5Xi~!nT
z|MU@sP?X4!ApQcAv0>3F>zT%M>Scm%knN&1xFv+o%T}<3tx3#JDZqs!C%bRQ+`J4T
z6@h-APu1+9=QS490ET;&`qmz?r{N-Q60u7yxGDi1e4wH5U`ubjv27!JVi*kx4&r>t
zqcE1=bcN?rUnGY#=oS>Dcrj}0&$x?NA__*f#e101$;kpcw_Xl?%bhqK5v1!fJnUjW
z@t(S0#Gm+K-GBe1S#k=A^#kkQbn~76>=zSGYC!Gp*7)`YwcuQ1qQ@z)K661iq=8`Q
zoJ_H?9-xL_*>Q27Gh)}`KY!XtagfI>+FztCIZ$M6Z>)_th3a@FTG@5pNrsb+cnYg<
z(<i#;ZD9+`fyq@d?5_GT0}CvSjDG&)Xp58)iwgveLs8G4-6IBU-hFRvEL6X0<(etV
zQ72hisa{ahk1r0{oXwF1*#4mJMG1Higph7ydx(vhYHB|Yk_R7~z=cGA5yX%j&x(Pz
z%zY@T^>F=fO@ucKOe8$e-ZGGU@3WKNM-gTXY9aJ>5{*F(VNEoM?MOC_D>L?W5VP4%
z3kSvlWNQZVl{gXKgD=?*QNkW`NWh*~fjbnks<aH|CeFApu3Gxci7|MLTJVkJzB^tL
zt*13x>O~&~d6G*7oq~qU>9h`b_9|O~Y!om5Q91>omCXSp_S|l3)kx=35l%=+oZoND
zf)Suc*aYtWmFP}(iaIPZCmlgngRz|daQ~?#?;!s|hg0nlYGyzWDgS8hC@$cXt**j3
zOI_R>Y_rEZ?SW*MA2W=W&jT)#e|k6TP-Km$Q-<1*yp#R9%;_AN=-UQ0r<mKnwzCM1
z%SXzIDCWxqo;HgM=`(z&WD^IGZ1Y_&Lr5jw-Qu5D)I9}~wc93LgpiUn^X5_RP{&=f
z!q8~|X)3s#r+wL6{=lY5{2Q$}JWrGM<)>b4{2##PgTe*AWmjwqN|TKlDncI89QB9i
z{nl=$-AG##{i|ps4=U5VB;fD0tH^|<<0PLF7?x>d|2*@0@Zr^u_1HN@+;6mO`;vb&
zApd!tXrlJ3o-e9-;WK~|sisHVgJ}!p?&ivj>5F*nW(1h`iHYQ*`A?rV3Iq%Qio&z$
zb@F_z6DXDXz1WTLBjO)q4?F5d4Mp?q=C1teP{gz|thsdYEqf9WuR_`r)>NB!#oj(Z
z45qD#pq~F7?5`n$fQn`_pCUZP)O)oPX)#MpqG4}F!(K3$@nVa?jJ}5g*IAH}P-^uv
z@Vse`pax`OGoFvSu(UKggIF-`G+r_sCx+GApT0>Qj9NC|YP~8tIa?(hiy~fz61Cwt
zF=>2?m}y4)ZTV!+?5qR-F_%G&JE;Dtp%1rz1J>kAZcAgIC3sNT*93h0T=|kY@6Rec
z$XPB&qhGzg43M=I3s^ri*VOrSYgrz3RG*WXlJTByE*VL?w(mS8+VT_*kcy6C;6+cd
z+4&8q&Ai1LG!1|z+DA6u41Ww#_4Wm9*FC2xpeQujv?zG~toty%*SB!-+vt%H6t}#9
z>ZA8yL#QtDdoKW&XR7lx<hu^?fg9+N^V<6&?=d`DzWn!KWsxaO-#gCPQ~K46M{uPV
zBo!hT={6f91xTSMetUV$4mBxZtlP(}0{(L`Ftoqgw<ynVtA)E0>=9nyeqMmsxv?Fj
zTjSlHZ`mJXLr)J}kpRMcyu;Lajf{Rnp$ALVXr;fZdeggj)<j}H+g@7>l$|sc%ZSU{
zSNwr;mm0J|@12&LsJAwt>;G}pDVu`@vJ2n`r}pWD@x@@}L-}w-Tt`X-gNxegXZQB)
zxh2OowX;8<&u|e>iC9pgAFmZkxomge_80`5&Ev7VMY{*d-?OiRLW5y}K~g&ujRC0S
z<WO-Q%?efL_+?BeNTqwX^Qh95V>8Zofh}ga-?QaRTg#7HGO6Y4p>I#{qDU!cbvGN=
zVeYbJzb~E22T)1)?EK&mU5xh7|Bqnk&{dptTWfP^AUpE5h`jq5I1OsOfSA{)QdDjm
zAw604*@O<}Ukem`(V%Dge7E{#n98Gvg2zyLj71BUM!d5rpA&O+w{ACcO9Xz^s({|~
z$-JB|NK4dbi{0)99=D%lPk|X$vi>olwdN6@jL=+cRh`#-ERNyt1P<dk!C8RNt?D~n
zlMUuuZTGtqRG|TIp&{uzZDBxUw}s08$F^#YxON0Yz-w1aNAGd$QJ^Wb9ldNVlSP^E
z>d-MxR9+urO6GhMHScg`0ru+tmG1}2O5yTG+Xm#alqGebPj^-~%g&Q+%P<}Opi>tB
zYDM*3+PdEy^(?U-s!3L&(_WSZXu3DB{16lFk4b#wa!;cSSqTeBC?hFD8qLk25S-(4
zJmw?0pk*ZZx6K?GdS5{bs&|kxb3I#z3D^Z<crDyo4p*5?qY{?mbwO{96W~{GEr0V9
zzgD(#7TP?>SZu`D^){_q>&TOqDj?yc0@CRKwzE`Ym%H!n!+67Z!*g#KU1PC9Gaw(K
z>?A@Rcj6LQ?IH(T#JlnG$em)pX*1NUsj2A;sIHMa^c2aSt)?Iesm+c%^^sDhi~aoi
zHOo)$S82G)dJbxeI#~RG1|7U9yR^l;CH8R-&hd^Arjk(n9~#O^;v7mFo;D!;u4AEm
zVn81Epwp!T%A5}Cm|aak)4O;Yv?w!7#m?0fWe8pyIh0uH5v*G{-e@vlE=jEVwwMrY
zQ(Q)GH{$hu+tn!3O`>$|He}E%QwqCclh6_p#W`Q{OxmGU+&OsGjpo1m1vW|g!UPYa
zL}e;VtvRkso%b)7azZl|rLv>ycsGR=Xb*&|Udmh(_+&lJp!hg_Yw>kX?LrMNX`yV?
zAWiXdWbvEJiN>O@&TJo3Tkt#?Xwf1=hWx#Pvc=7NM%2-H9bYxH&%i+OedEIQ#M<{d
zElRg>2U(+e=fkR?<O<w82HV$~^WAy(T9={9fWMZKi7ma1OSv%_XQO~Ta+X%$pP(oK
z#oX0iZw;a^3Q^A=1f^D(#|P2-6h~+On2m`tXaC)axLiZsn6yDh{!Q`5g+35!2m(y$
zPAg6OPx0bn-Ju&p*AAvU&16~9Oxu^=7ryRXzQC-Pkj3@?iGi9!u7OE^crU@0Qd0h%
z%<s~)g&|9n#h}m3$+Sv<5GWC@8#&e#h>}}P{P@uK+^mIL9WzlKM`ky#cA;$+0AglE
znWmfqPFpCokNAYU5G7vhGnOk~LCJk5D?MxP^^8bXd9T-u5pW&&!dvdurjabmKawwQ
zv;_y*vs-@2&&!)H+XCK1%;DAB+#zp0a$`n^iSx%3MY&GG^ko$DMRh8ilM)gg86x2V
z!X)lE6g9GvQt4aw%1t+NQlN?%tZ@`O^SYZa^>S<^t%7>h(A=1=C~8Jcw$0W{moz9U
z6X>Oj8+fdJ{k(g8niuGfV-%-Ov3>9KjgOzs2^CcaC=}}5Wjz^*SJw{I>fr6RuH(bb
zR!`R0j3%>yu#+$Dhwf&_WJh1kbjCMpiOBL%kc2q#Yoy|QCml_`0gB#U>RF7!Up1kU
zxh?Uhih+AE%6syg&5>j^8Lj?yUH~jQYZ69^>BYoEMU-8MP8zic?CXMNoq)c(I;F2G
zO>*RDC*Xn3d4LpR<cDGAPDA=7%d!s`T~*I(oJ-K!Ff3pqdQ|gyJFC6jpontR(}$rD
zEn%JI`nwy~hPu^#pmD*M9do-$EpEg)9IFTyal_|*wkWNE=iLOmpj(ncNnG#g4=5W=
z5jS+>e32i=HMp8DX}sO-zh`JKsbLh#FFd>+w&$!=-+0?N8f{Co-~;@!j-`=Xb6~9l
zwwfDk+*enxs4m6RfVM{k^WtUR-|k>GdGjkwPjLI+g#zouOnTp=V|&W+DbTp1?Dg)U
z=q6+u(u-tln2z5#giIV|QM%sUh?T+EBIq;fN-%P#cjQ+zBKJ~XFrz0rS!_O<zM=gk
zzN@j)UEPzOko9F_?hWl_j7`Rk-_`ZJqulg1Djw&($Q>~wzf}$Ne79GBdd_eACqwO)
zlzp+X{=zK3(sA;I=f=mZ{Mj^HoIV=Ax!N<O?Y~25qY~UxiRW=A$MeqJEo?>+^D^im
zX6@hhrps+5T2Y?%RCe>Io1!EqNXOEH<WXP#387L4NjUhuVZ#ssIrWlw=S6>9T6z1S
zv$Glf3}uhcYP1H+Qa7GM<l8+pvu9T*W(2}XFspGt*@90%V&sf2HLNuOOUg8{8O45I
zsJiv)l>msFrM8*qO*y@PcD+1;>-Y`V-H!t0W>}e(OO3L_KeGVLQe@B3aTBIh0DbZi
ze*(%N7gYx5P{qO;;CEmU>H?>F%bMlEX$`<RMup3juk=7DGv<8(I`am8)eX1{iiFwj
z+Yoq~7Qz-^3uqVj>QPIjxuutPp8=_fd%!7*`MbRe?T!r-zwhI_MYO^a6AJBJk2;i}
z+Annzbuag`3(>&?m%+XSE@FkSK>sGl7W1eqh!l1THzidp4zAVwRC82EE}wZi0+)SY
z_$m9Fw&B7jw#rPc4o7{}0lCNDh=+cZFYg;|E?G~cAO5VoySnrp8-wud$4F%#r;p(z
zNLO|QCHjj>Kc9)yq%$=mZRzEAe_K{q$q56<gfZPRNpacxiHMqS@Hzl{Q<g9aakSp)
zYGl6B4ei^P2<MoWu*?_3nToA(#bUrM%P4E0L6Jx%++`GZ;|MaVw-nvGX7T$o+S?{m
zYwSooNx2pU1HiCr+xwb5Xyx`;dTN))nCFUlTbCE()M$AZ|GK+N<>*jvG&Z|g^SB-r
zD>(@m%-zB*%vs>(Rq1<l-orMzHsT<jb#P_e9|XH^WYQkpF18>&1kk*rua<V2qP0x&
zF{(WN3$rFlHLhmbd;M74bk}ReeuSXpqejV~&0iftho<?%t2nd`8SQD6@TN-wcd8Y0
zTKZy%m&_YzSc&Sxdfj8~PA7N4&Mj$lcRub7r2v7q2D`AlJ<kIPm7jhPDf{agy=;f^
z)7w%Lk4Be+e7ETdppB6wye>xK^MqV2&mBK~XN?eXW_Es63Fh7q{M-nJZ`cWy+k1ZY
zr~HE|fy&^zk7LA!_lh_pR%BtZ^CcO54rULFvX_tdtoQEPtrW->^N=*$;7lJ}4)2{>
z5}h2`gksPhBl-3H?TG@I3y+}n_*=8n{4W?O6`<SchlFNi)r9Gn$AXA)tSGIhtRMyj
zL3=q$d8Djex~cK5Fe1o4w~gVp?*ey{hX$0wiz$FNe>Q!IrKu?Wn-b=(C$AzV<5d7>
z;M~2M_`_{~9ntvXBV+2dh5DWo!&}k&_Yc>HK0B@vL41gVY(&9inh(qS4*>VOJw23A
zVkhZRS5i7BT?>p&Zyek=psH^4_O<+n1p<EWCnns!|6LkDPHxT+iZG=9&^EZzOItNW
zY~efE3$5#ca%rlIT5e9Fx<vSVi-rBwwk&xS$6Y!v&SFRIS{Cg<@q7K-9J%ymo+C|F
zT94s5AJDGHj7|ZP8YuVsddry~KLSIS;&(@yzD7hUTH>2z+cB0(A3>+n$hPAuPzOc?
zo<mH`dIFh9{VeoAPZQz15GRxym&*X02tNq7yhT}tM|2DIRQy|_d*YuT@sNn0^o#lL
ziK+8_P5QzY^ZopKo(8qa-^D3v<WN1uhJ=osESA54jDOrtmgx;(X6_=X_yV~Ten-nJ
z=$SXA9~HtCe#GVGIP`BwXGF|^Q2yTl%Ng9(RsCya{)J%$LGV?cBV#R=fxu+tVA54P
zI~r&Q$1~jq?j}b+7PivUb(5b3B6ByOhbl44@8;+D{X+liY<0A)`}@C6_z>$2#fHae
zZV7+A^{2cBA%!FLyj1hHsT47EkI}VTu)vhcaIWWrYo#)1Knv(UUNFNKC`2z#cjJ8L
z_m%#eBZL;=tN0*eMZaV93BbtW?>_=T;EWv@5!mDz&4K7;ktz@!=9SqEhM6qaVTMaE
zJG4^|S6Baqd{IUPLU1_l+wqkA>#aXuSiWok@a6cznG-gEGdyrZ2n|Sau}Xp`JSfoa
zO-_OE{TPSZiK(KQe_=y^f3ZdZ4M$5%%$tHS45TJF6FAthcvytj70@)+-!)b~*yE3`
zfeG?t2U)b@vS3e{0@muaLNJZSzE?--ucOh407tm4GjRK_xBlEMVOQ{V7J7ke3+%30
zyV_Xq#Z*=OnM2j)xAkx0OKEvr%3%J$D{SjH1w3BsK%wMu_(R)&pp0)@frM|!EmP*t
z%kZE6-cvO<Xj`|Z#PTk`{0?hp+(E~YCz}maTPIy_HaA~40O%9-2n3&34c%9-0p_O}
zta5u3h$uZr&?^LgU8a>(aOfwu-ktm%K>Gs^48$+4KliQz<{mNRRdNv1O)c6C5w6B|
zYs_QUj({7%8EX^(UEIhd*!xYZ0Oui5dHIQv1soS@KF9g&uP^z>Y7HV*#~!v<j-}@%
zYQkbrLOTMBPG+gK&|}&F(7I0oigXd+8g`-}Jn4D#Qa?A5wEjO!xP~9(I?4Y;`2KYj
z!_zJqoyveA0Gl9H+6`R33|X5KX2f}6`cpbl(64?=fBO2z1z&jw>n2E#t;hgovMGR7
zbFwwC@ax~w{O5=N>vMq=7+Q&zn8umeFZQhADolG~AHo>&{v!M0nE5dK;#Af2QoH#W
z*N=Z(XxKDS2l)$R^1B_j|9ES|_=&M|0XD&}6j{Cv8b0^Jy{U0~_+gA}WoG{`VO051
zGS`1SKP<O$(3p5?#H(9>eHvvvcJ-U*@^t;@#r5|G^8flZ3cEJjwGFvB{_lSO-{1fL
z`uT5&q{GDyq{3>->16kkDF4gC`==`7ucyj?_s%<cb6}iy`||USQ4CLQy_M2coYcq9
zxFQ5|{!VEEn(u%7qj3{g_@_u>EqF8$zFMTXjc*L}*VaU0@lPJ+|KF|FW({+IR(g;D
zdxzXBvA5xYAT>7M2k1%&Ur{J?|8-aWAANfBy;;UBuZRqgw@CO32@aD%jg;zd56}Pk
ze;E8r4Rc%}{5K6@C7+`I$LHzqUx4rfzt3=D{f^=9@BYtU|NHy@eg6Nw{{DUc{g?d*
zLKp<%20B5=sHGzq#N-b2<cBYAKEo|iSQHE*F*C%<l>awBL4x<>!UHWUW(m@NgAkMH
zDWK+JWEKRCOARpnmpM;?faq0he{u*oz2mJEfN3A@mjrzOmn1Gcu_1_%X(h{d=Wcd2
zmWZt5W9~`f*28YqFn{7VFPexqfu!Vx&+oE`2~ak+IjOX(^{9N4YEv}x1u68^Adosw
zw<&_zzdFM*R-7)|ofWrPV;1N;wl}tB&JRShyRk@tV>DkSbm=CK!~gCP(5GMyNDpj0
zq$f~d*8;i4$6L<6SaPdcz;}+Mpgv3h+m>Oy9Kf`Nbr^mFoy{~>;i);Ze2TT=R`V2B
zzol3N*uWOGiCcf$^4<b~FI>Sqj_eCMONjwdV3IoN)q}%9^}YeH;2Dt*LB6JzF}Ej6
z0?wXSb+AT1Y`XwX_*!gyx-_V8W1hc!yWPe{CcPKmJA>^*dH{~swk1LF$3=lcg0M98
z8K8Yvxp%O$S@^)b%=0W@Tpq>tK{BM65}p!d`NwKE^eoXE0HHd9JO&q_P&%1qdYcBS
z|M1`jFg{@k=*5^-Gmt>Ed{y~1y9+Bg!7c}_aZ)>xJ|H`MGMwH~?5%PK^bm9&l|BGJ
zWwcZnfMMG|TRH+XY`--mN5a?n91P=d29s=RPk`WVcC+rp@o>1rf0z+myXI7#0=55A
zcmJyZm%|??hq@~^fD2aT2*k7Q7h4x+vf14rK)A^Q35I%nIqv<`bPgn6ee5JXp!YD{
z1T#zbN3aYo))(#$&Q#om)bPUxH6RwYYi`a~2PY&7Fjf;5Luv-Voi+vXDIZ{}vE$er
z!RQ-Tpr%D;xO8$&MaZ67IPuaT1m}_DI8s0*^z2afVpkS)YQ?Tt9{C>X=kek_p3e(;
zC;f5$4{0|*AGr2i^)lK0IRzw69jNvl0cGS#!dDT0f{qJd@E`aGmM@22Iq9PA3FcFX
zV(PZj04t+ir=1B}!Dju~yIP29uvma;Npuce_m5i<b`vw`SkG##Nh@aEUC0}slf#N&
z0gBNBoP7{l^OE*61lQ37(Cf5&K5duV{PXe*WuI)-&rAWa3EREb&a)u4i$>Nl6U?FW
z0aU;gE(tH}tzE|%xH>gaq)xuP_Db17HVp+Sk2|}-a6;dmZFU2$Kk=sj{tgeihMVU|
z6!~f&bg{0SCdAM-zjdRTowguf|6ZTz!G4k+Y`_I}Ac0E_;B!%vz;CbygF!5M#FyQ$
zdWNI4vq&2x*2U_>-!e<yOu}voT>Et?Heh}h+!2&#hglTofUx)Jg*#*Bk*AxU^{&`6
z@ZBaJjFKAgL($K?P-S#~sQ#v=zQK}(AW5p07vQ89FPKkf38%p5B~%XxG%8m|HnJY4
zvY2x*`Eg-uJwsw_d_~>N3AUxC5=cz*8kPpM3f+ul!*^gpmKeu?^;3(M=J0Z9)zJwL
z)HG1pXa+vbC7GRsxN0DZYdNgROCgsw0P+|mqEg0NSwpbtEW5&Vm4$R~wcA1c+;I$k
zg5hWN0JT7-9<DKvZl;2)oucWC4KEn}Lo=yrZ=r5MZ2xZ>p29K<OMNxK{=cq4byTpM
z&Y+Se5MOO#(NyeIu7T`leljaLA*W}Ju+m?ewERzA;juM<QFSN&BxZ;Mu{gZ8*zA+Z
z<w+nB`TfWgZtZk4qCNCODHt45iv>rVd=RRaEB&x1Yh{$(D;{SHk)$~6nl^eokXpC!
zUAxT0!mz<L7lPB(u_B4x;5)PfKVOWNpQYBHT0dB5w3VFxh42qOpd|~gSaj~N@Fj0A
z+C3-gILeXr|7L`}kK0&-82rj0IJ?P6%)Bpe`+W^ipQ;xWRJ{CGecRA+4IAu^w50S9
ze(>iKgZkrYOYbKs);Ajj0C0rq9G@o)k0VDy@I{Mv137s-i}^z$>}aR&6ql#Z{K3t7
z@|kjMnkddsowqD>;@QW~lR)@RH-$d1gW<O{4}d_PSw}EkD?+s%gjp3hQ_KjKxdIgd
zhR?<yh<-7j6X|G_IP5Q<f?IdUi5S?gXuR{}+1<`h>MwkTzT-N4-~Y7<Vu22_$Yu`f
z4wc{gpNQPS_XV!iDG)ugH6{Z>b+CaRf27+O0iCfQ(@dSZ&7aSZ@8m~d>2IDSmjDry
zW)(0|l`8{vj?)(SBfI6mU&nGs2Fym}@ijXDSdzCu8DI$N4{d9~Y`eCqq%73+)Vgi5
zfTNJ+6TYm^P^-DS)VIcfL(tIMtVdwAUf|mbYc`?O{;RFN3v=VRRB6mG?bEc}owDx>
zc$OD~+OkvN*p1HN8TRE-&4Ek;(H+#hR~ovQdrd(WjC!^o$jOo<V6~?_4q-lRplHX!
zN-i5fi86<>kX-h#sMtv;BzF3V9a(c2ehT#F;#Ndb@&aH|vLjI7x?`ifn`^C;KMkwP
z42bOQ&|cmG_v~2pbtebFH--5uOU{fPk!ojn$&)e_8)^hi9!($CRXJ;lSpffrbRskI
ziR%V7j9+Fqy|=o=caf)Hi;$_{eS0XG--@Cm;Ic^jkJ%h9p~Ff+GQx2y$DHdXn~`<-
z;tCi|wH&|EK?%>a$%6aR&bucm-?9#Bl}NhnA{vCL1?Ht=wtALBR$<rkC~e;+pO)gC
zSkcGstUWX;qgeSp^6)nDxa=x7DjdjG22-zZz3AXE^2M?%aZZQFw_?S%wXj1v(fWML
zfSyebczW7gJH5}prN<bzhTp~k^=r2Pkd%i#2eT7x@0QnTA?XaD80(4?o?_y)12BVi
zbw9!L*I6XSbzF4X2G`;GRx1dG*jSOw0&{6(W8cnT8ObTEps`wD<nHMyHo2>R^TEZ+
zpvR=M>LD<4d;?WV8&>2ptuD1i3+cqLu9*oeQY;tENc2B9)1a5Xke5`tP`#)HF5LFJ
z&)*C#4U~{^>+thJ^GFAJ9+(+^!qypLZ8dv#LnNK{<bDRRzKfgDK6WQ4U!CHLm=1v<
zLF%o3=DvW^8=(iD<SY))<1K(Q=zEH)?u^%=qO8I2uf0TU*lyz#INLEc?GB50WKy&1
zvgI5b^gAseieaGoj&$y4l5%dLd0^35ns6Lny-_u<CwV}WX2hg6nE$7f`!zl`BORDp
zMrr^n-6{>LfmjcQ<MXqvfck7+W&aPawp4|uEYHTkbi5o(-n$Kr4Pf%+YVW=_*jKBG
zVQslnkgcs3MIOHSvFEE;DXG;G&=PyDC|}0{eL18-K=a+m=-fgvO*{+$<A=*BDbH^T
zRNbH*7QpbuF}Sme20kyYX<PrGfCSzQWVRwnwmVgjBXI{gjJ0hxBT9<BE)8n26a6_l
zy~7)fQ$teVg?qWC8blK+%CTde;VQB<+C}96h7ipP9Kc@`91-Fg?I9;`7H1FQ2lHWN
zqjs<!WS<w9-!`%nLb)Se4McAm!cP-P5jY={<QmE29XRGX{WR#zUo;-HbFQRAPO)`Z
z-NDzA7`e`UBJ92jt;dIXD{g0k$}2<2@F`ITzWu28D~DN>T)m<(6m9-4WPmx(4KWUz
zeNuidIddE<N`%-VS#H6m17V^JHZ9%;fHdgctQUf@MFIvOor_W7;08oNK?2s+Tpj2@
zahLK6V4GXS2|T3eo5?Y51Bv1OJ3(kkf^_J;Mn_Pn9ZGSMSdnx_T_$9x-~a3cXDiw}
zhHCd|mum|146jIbAzq}t!<u4vVrW$*nVsukDhljxO><yp4TLhYF~aX@pi=V%*N%zA
zA$T9ez`7C3GVOg|;`Ol&c9o5TIx;@{0#kDFObI?>;t5wQ0_vw9U9r6QjSZKYgT>5S
z0xd|($#<xn&8b3)wAOAC_yVhQ+67--oH7&{VjP0jcsBVJL~Ds5aEOsJ$cPSu5F!o8
zD=n)B2ysD$xy<>qw^D)%%_}69sPe$K40+}d0hkpW3q6AkaARvRv%#Dyoq~<8NA~!K
zHs-WVB{|Q8L!pkdccXHmr!0(jOVWXTxdLw}2y`_g^lx=MZ9{D7IWxq3!IR~%O?pmp
zjlAu7NACu%gP=QNeNO%q_EwXqtzw9HTCz<TX#>r?|A><Uj9?8@*-HX>Mip^TY6~4D
zAMG0RmAj^1gae+5M4UE=Wn0hm>V-F$7#x;fR#J!?EZ(Y-YPFkoJ0}*04M-stf)4PP
zhV-UIr?n9gpexXIdeh;wU<2wOr@Qw`7AaUebve8@q6J6i8Q%spcBJ_>z2((u?^w@K
z01vsNYb7mvK6{pXVMF>Du}(;cK5}J{snDWgX4lvcgnNA5Ak*-QuzTJ|aig|>y6^mc
zyG7J_kS8PF$Okk$^5X=X!&MRn+Y)#Hm*O}rL}@;-2OrN1!elpMCklHTJ@MLqxA7#F
zG+1t;ogac7WE+s3z)Vzn3~u7B89a7EK^?7TEOp@t@HNagwFxC7W#2tZ-z+fH!syxi
zg2$BvJzQk@b_0LVrz_!?gj|qB@7GMNWZUo=CN8(JVMAS<W)hg8TQ{^>M+S<8!!2>N
z9L0-GUbA;PVduG;xwY}jb$r+>!zy-;iDAXo#+|sY87yAg6{2_8H;S4%$jw=}U8E!g
zsM+pkg&0rX5vp<71QR-Uck8XaXA<s(WF^MQZC0|Iby87c4uHg~oyA9Q_0xpzE97Ks
zPsc?{o)qRYWk|<um93RehW62p(aHyOAy<R<@Lvv6{Z_mV@v}%qE{mkY3`X&-v=1ln
zb)CWBGivsS)OYS0j^f`#0|uG8&{#mD@qSux^(ZhZ-$PXrR|Ho~<f~e!P=#s6Q{U+~
zNo$%kb4r|$F}b5_l!jYS8dcu(=tF8-Yifo=rJ=>k)TX4BnCJH8g)p|bCZx1T9sMfh
z?)m-E7JG|MbY)y+lvB2>QSyVd!=fhk5|WIxT@8!pl!;-Qzo`oixVuNo9E>r4wiP*|
zxKFJfj$SDm7tL6voYC+1n6R?3P_eC>DwGw`mI>1=#?<}5?;jA;^g8gn<(CFOL!TvU
z-lus=bU*D?A<!2~UlD%DWv@Qp6uLhk=Hh!-qTBo<EF?1cTy#(NPUHHnMfk(I-aC0w
z1MT~XWLB4|!N|^MVen;cdsZY&yWw(gI6beXp<bFkY=-bPj)6Me3dhg*ccsJu**$=i
z?+I9{J&IedK#7eX6VXZnvH2C@Qyy%7PLtI7@lK#}d|9_z%Lx4f=fY<Zu$hPyo62P1
z=*cswF}vKbeyZ|WQ+F&h{lXt?*8JUaUtnxQ;Seal2wPMWWb%q|H&?eo=Lj)0t^b{X
z{sRTCu^&>()GaiG#d*Jfp6FVrTt{o4zwy4l7ET*Qg%wqH-L9#F;SC2%90R9oOIJsJ
z_TigDgfG^x&fm@1TJK}sE22o=`*M%9s(Bb9bY++Bahjlx#WWcS{erjQ8xNxNambB~
zw{#m4@=v_txU_d&O!`tn%kUWDQ-3!6s9)~*xfC{>8b{l4dohH}d8`rNqaryon5*9J
zO~pjT?TY>i;v{XNrN!RV1ln_W4+HJUwjWXc1obx6QziX0mY9;%{3xK1K}lzmO<OD+
zio|p5Xp0L~9#1aIEblU}T}Gwk#Y%jD<>eM*^7h2_vf<~T(cQ>)YKyM>UU1&fgrJbt
zZhNuB(tCKfP>D;UadW1iT+D}$Y`#+uyPf<L8%Y(|kM#e=u1vR~V9VUp&%WPPp1mPC
zPPi6IJ5N(sx`XoR4E1hg-j7GatiKH0<2gpVVz&GJdtLWi&nnJfg3Ucm5K$6v1?_P%
zQ;WPxnk1i1XI$CkTo=!vX{NHA$+mmm4Xc6SHO~VV;Qod%1|gE>T;(*;@A-ny#eDMD
zLI<Ir=DX&-uTL9{(>xU}ve<l@=<ckzu0-4J4#?F%A>qAZRJu}v^2Qj9UllGFpaz4t
zgZ)PlPp=5q*?*HxmFlB<{|6H)M~2_sH{B(^qSws4!HPQCPS4}>ex8um#T3CS1F)v+
zPay_r?GI<N2FV6I*(B3#PS|O`@RItC(BoJMVOZOwtWn76C&MOSi{ZktdT50-#yR73
zC~oSYFh>UYr^`f_v16vom^#RI^Pb|%{QB~C?}9mK(jlf{iOoDV{q!U}wmkI)yoTmk
zNG*pde9wr3AVJdpC6pdJwz;lEk*x3r?isoH^*x=-q1FAsBlqN*_3e9WBHKH;R!?7*
z^8(xEsUy!Y3430e82gP&qxW)~9^oJLZa{E;N(Spc&D)N{qx9VtbP7g~XfAV3{#-ZT
zkll3~bqgpWt8~xEwzpEqF=-%=C%(i!n?@ePajS;xSE>0xLJ;LW_kT$Gk@((%`1xDd
z`NSlUEAxSk>F~Z?2T96PyN20}AB4MJi!s6X@ouxn=(>QBN&Qy{I6|)MjV$D8=oH4a
zH^hp-ax=q?k8fPv8-wsaU}GOJcdR?xs*Evx9ol9C>4|Kv#Ig6)KeRH~k!R{we{5mO
z{vECQSfGQ<{U`F^2nq55YIbx}jfUN!ODuzF9QkkWc{GXPC|-r=JTSZ_PBcbj$ql*Q
zq>yj~?+8HpKsC@=NtYp^5}M@B7a;iKE{%0~|8qL8UcrW(X^+PuP2YJN<{qy_!ZQvI
z+w8qaU97rY4c0;(#8<DYK?7Ju+JbfpWlA|CPt$2W9xYccL&-)2@gWJPpi0}_5Nb=;
zYm_>^zDH>7J|dOkgC{dlhqS}FA4y^@nOKhqO4HGbd05kf960iQvKQu?-Fdd8*+%b^
zUerkL@XHb!VeyW>s%puurzo23_=b+^&*s7sSNCiEn(&1Wl*r}p2dymHd#zBBm`6n6
zffhK1#{4Xrh>N<DsV(p3vxgLtyj$XS8$|*={ReCurIEO2IF{Q(L}blw)bwdIxxt=@
zANi9(J2`x{V0<FIu4QC3P*0!ue?%8#-nfX5t*GzKBxCP}-CT^rgmlwju4HLqR`gK~
zP1@6T<4TB&ru{iO%x%cHAs7|Kb{l`;oBv|aHt!!Mm>g!@3T5bOYXC31d|Om~bBDM?
zTvrx0&d+Cx_e_Jk-Pv#aFsU_n3&W;k^JZj*(yfX}!ydP$UA5vXuj73v9M|HLI1|tW
z&_0iqf=ic?g5|H7X|wj)8^2FI_v6Mewfv5YbPt=mUhK|Fl@>&(WTmgKIr*m(=}yf{
zpHm9Zxhto?J!&w*xdL*ftS2^(qRPvSn((QZaEK?Ht<oHbSIiz#^@wQ5;|zil8x>jl
zQpTS;5uw4Fw;=GKDQBh90eW`xHs!GU`8KzX=(dJb_Us0`-s>V~d_}oO?;l;B&l9wK
zyZ(}=_}w>+nLjzSJ+&MfMLFbRE0giOs%1ySCGUZ|9Ia+2yTC}&;X~2jDpwoGa)n?P
zs&|#!9HuS%Y$6eZV>)gl)E`|GVh+7l#HKHkMpVTIZW;4V+}^^AW|XepiYki;?{O*L
zWj`4~(&seW<dGbLI}8!V$C0St`wWT(h{v%YYIlIsK3ZYRI#SQb?#4w@mBVp={N=&t
zN8Po5r~~spS4{8tWD>tkfZ^ZdV}aQ&eOFBgeiJSFd^}1>TVpQk1tdHWYOQl1uSHs)
z${3NW)fa1_p>bAxPk1rHf`ga9jwo_&fkOsOJX2gq5VZ}2tG$n`?Q%ZBVC8(M#|T=>
zgM|d-tV4M>xfB#P3ncLsSW?ZGZ-}10eJ3_;yllY`E?Ocs`s+tlz<4qKRtEa3-ZF48
zyZ$1c-Zed$kB{qrBopx?UBDk59aH9}X5<@6Pf+MpfM|vxmszCG_FBcq{iP=6;1!89
zstEXMo|gDxQv9Ah&}~()K^?;bc_S&92sX<;2?O0a{JUy3KZENs9*8Q~LbKA#%1V{|
zR+Ig?Nl{^_ha=sNjn$fn6;hPjTvug!=UrU)NA8I~@ldj1XO`teuFuwnb^U5>0t;A{
zqC#B#50uP(uR%A^{YUu{5;a#WHED5UNdu(6O5=RJy>X0KG`l50roS3<xD(GZ`(n7N
z?9FQ~HKeA_Rn5{UpZbfh@TPYfiSXAio+{~JYBlMOPBaZoFp_5D)b~Lyy7<Jk!c{uq
zeAmj%SYeBSbex<o!~5r+8P)EEeIAtq(_>a)U!r;nPmauki-enHZ9pm;yD-=XHxOT<
zZrd24Bz9#MMkgldI4&kkeKr=mH{J<{)m<=Lpu96GT!HBS+n3K*QQ2>}Ch48WsO2>4
z(+9@WM;Zq9Zks^;U|UQwQ~mnMMU@iM=@xmysrfxgE6F415Y43{<XGps6sZ=`@rAOP
znMnxS6i~G=T;%`5V<SVHCSPp?j4p2+zCY$E)YCT^PG>=W`LQst{xh&=OsQ13_dyTx
z*Bukf2~zQlZI7j}hA}0#L8n#147q*3GWf~PnKQ<xK6}+@CBQ*l-Ny7vrshM*iSp*L
z%Df)Lu5Uep!9s<PFRnA0O!(Y$DX1~MXgb9vRA2?=P%B*p`q1!_*RhpNqv<$x!f}Ch
z3>p5vw=E2I%D`krI4!aGdxU34IIX6*QQ%y6PR{B#<71-U)n!`({kqN3d%OZgx+MxQ
zMKPN;UPr*zxGnj8T4aRf;o6Yg^Zj16@u1nEK-EfmT0mIFc}eg6>T9hnP8!b-`kE96
zM~x7>O!%qm($!ux6I;m6^hC;&o=P_N%6(@$$po2L<T$kzDZC%KEZ&Bb*#sLRo?so(
zrLG+Qs=kO8I|I;2A7c9r?bS9}y{ru9=n~Z>mhU^S_Xvey`wTy~f2K#&7iPa##|{$`
zwiH**CMVow-Ry487uNRZp{8t={rW~&i<N1dch_KSPtw`79?s!bdD&~wX)eSeVRNe>
z?gwL&Sa+qGd*x4iHg62w_!D6fk4nZDD4!j3m3IZG(l+G_&-+zUF2AC*w&1mOV)+5U
zd*P0lf<A~<G1t&IaZ<7CLF`PhHzMC*4uS;_Ktr=$!We!ib@N+xUG+@m4B~~~RDl$2
zNEOKnzpW~q(B8tLR^12l6Im%CEQH38vdY=W(<{`*kXkoi72P`%E2loo*@*1(>bbFF
zVWcHFfjiMv{d97~>rpMGQRw^$_wI|8KJ>%OyGrXF5x_hSRCzUv<^BsnDII%%1RYqh
z>gYcujuTdVW<!#8syZ%VbzCmv1HhgA9lvUUb_zoRL*A9CC|5Ah)3vg?fD40dsnd%l
z+C2A^JBaMepEV<YVVs4(`Hb651hvJTt!57hUOycC#U>eHN0CwO^eWjQC}j%|x=cPB
zvgxhEl5$~RhU|@lzdE$EY|{s<K+TshRtM%@FASgv4!miXw9;Hlm1Sivz$D%vVS%#(
z)MpR`(AgE^yimFFV<e)loz-dLZN||x8<M`$>RXI6O+8X)l5L0~!OnDNqO)xfWIuT`
zRa7|*tysbf-GwoS_y^bIU>UM5nU#+CkM9LmY~z30XN;v#Ef-Jlk*&;;v?)|tCVycR
zM1Bg7?Pd@=+A;rq0FcpRX9nE|X4vq4n`!wkb<E;x`1V+I_QFx8#NPR{0GkmE#jg2!
zVJ4vSR@N3uV&P$DjE_YbfbUZ8{~r3KTV$K4kRg9_rMgeLGLdI6!a1O7<GM4f?(HX{
zA@x0(r#EM9Ir^kut3IDm>5Fau#vP!zQ$mw+m&z~T_derdqs6DYwJk>xcM#&bdnE|T
zKFvN;Y#Ib(SGdPD>2{#;+v?7%9)ahHs`Q397HHIG=3V5NUL&(KH|pN(s?gvF;m`1%
z(q#PjK+M^w{S>xAXX)uC$$W9tK83}q?IUnJ0e^LJz>vwf^%@@PwDy3O>@5vR(78Ea
zOl@D2sR*cKlTR})UAMKdb#ehbV0B|U;gL#G7w;DYb}0Rsjaw>=gE>DzhC<qaZl&a%
zI*3z<aN2!egcLxUVPSi1(Va(hE|s@iGV^|F432g!n@*Eug*XuGb9KOn%qHymMlAgm
zZ8Ri1#LJm03HCHYh@?oyV5H7dk@W(SsYof~9T!k@6lpE|j<$<P=<U-MoVL(H)J1!A
zy|F%m(<}rm<=S%;l&`(tQp#b%Sr3fhHux~%c(wZxpI1RMz)MS`xavD=e^+g1Y7Tgp
zP|XReu74uojWm>x(#Ne_v~HC&4U44gh1+knRz_H}c)cwR&p``k>DmwcaPqXuR3D^%
zrpJUcrbR-=_DdWyy#J{*?!rEJxJ1k5wvS;$VlFYq*Ng3KUD48?468ycUr~Hs_!R^o
z^TbQN(egUaEvi37i#yj=TG`BN9aPazRAo&nFBfnX`32Z?zIr>zoaLa|sQc}zH9zjt
zJ+s+l>Me)8Rr`RuVAi(-YfdTg+WoD{nw=n3IV5fsNv2fQJ}S)CU<U_J_?~S<?uJv%
zPX7Um$te_hiG{hXk&13wIDyL?RYJd9xUl0+R#Wr<wwN5y+x_nS;($aJtHkif6t&vZ
zU?j?Y905D^Y5f)N8{w&5S#fpZN$OY1)x2O;IjO<ntIawb!9_JQ0MB$t{Amz)=Sw%e
z*>KEvFW1J8bQ{v;p2qE?G7U*@E?Y7qjSCv7a%KEr8$*t$TvDgyr#WdVgQIk54lh6A
zqPV55kg|q48(?}yt>Su_nN(3=dMx2XZ+Kh@3h4-6_(YjGz3buUtaf2rqCjR!XD?{1
zc7L&--N;Y8HH$uCPMVB(vq)gGYC~1kB2hbx);4sZLV7&3V#KcE`@42#Y@$*<qXY(T
z4`+q9FCi^g*@6Hr^9VeOx(3)oUS|nK`i|mjG{#<?!XJ1>z{2Rn66YoNX@^J+|9dAM
zBXcpyA3(DzuDdm$`}~FgV>)|by9-qe?L||-r9I_T3VIdVdzgEh2R-QGEskTJL)%j~
zo`sN3B1n<E_9~OpArdHxz3s`6)N`SdVeHXZw2@vlsn^7aN>4)^Pq(C)ovQXELm(Xi
z;NkaXxk2iZE9IWyZ~m5){rpCR9(Q6OGWt|0OMF1o3BbJmzL#X%u2f#C-}9FeOw?|>
znk&^kn`{1Yg&c5DbX&PzU<&c>NIZKRf!EDgJX6n_)G}7p0jD-G8Ae#X)O=0i5h(u7
zO3h8f-7?Uc7Ai%izjCCYB<X@$7U77?`v~-lbi&g&?vjLLYvSKT#|d}LNy4JBTqq2W
zS$4*X#N~>x<F56~EQm=at^RfK8Gx||JAb9!pt?h`xlV1ZIHfS7^!-V3ywq!pyOfnc
zS1_=gjfvr~Tly*{aI!x1E1jj_u!VVZF>@iaQ88eu@41iiX=UK(a-!!RkZ^3*hNGG0
z-81g=LHzv0%LfTCeINRGx;`9vAJ=7jCM}1Ru}he_eMjbXD-UdrJrSAd(b4a&{_<|1
zpMq?ICTrxqBJ!}bB~N;T%Hfj3i?K6tu#G61JgKA2OrR}I`K8`T>8HKfaPThob6&s=
zz86*;Vr#c+k8%c4q8KV8{G}O|0m}6)TS7ZBMcJ650AX9Zd;FG)<J4&`jA-}q+l5@g
zubD@WS&-Qajnsjd%)k|tZB+=OAuZmw|DXWJgyJ~&{Dx}Z(u25Q6A;mFnAtJ*Yr$4{
z(06~QX5=u$rjbZId?)8~{^ZY<5}2ilS!z7RJM|Tq4alDjuHiJYnpPK*oVHZ?9OEtT
z{f>UrXXxsy)ChIc9$luhp(udldaI0^ThUA>_XV&R?e>)N#7W`)@>6#=;lMSRC(~Q8
zBJu*TxA)w;XOgKB;el>K5bF5`xya9Dqw{UNEzwxqoy5>yIHWY}(iza#OnE*=`^`yd
z$hC$Tvd@!;lgpvDYKlPAU@PH0cY;xfY3KQ<5UEAWBZAQOxDE>W2zg6JlzgaD=m)K!
zDBhjtAIl_6KFs_n4*zzup~0gl)8m7mzHA#Iry!JfFyhx4kFdu3%uza;Ck)!MTG>6g
zxC3EtL1OqV^Yq;N#lDD2mf6w;ykDGIK3BmGwZ1!eqGOyxuGmIwMDcb(AtA8QPA^kM
z#@gj`>=QP~-n$aJmAzP(GXMs_13GvpJu9DlE4@DG4}{~T%B7#F_qPm_8(+N6Hih7T
zW@+?;k1UggdZ$O@Bh{020tX^BIvs=8?Ym7<@uIIn`<W6M_3k~YEBXXm_%z`t>>N+f
zX5Av3fbD-kF?MVqgfG(TX6J5qj0n!cP-?$`H2W}J8;@f#uXhN5@=Q45vVh)MV^(NY
zHZstHPbEr&G!K>2ZNFG<X3+RVdGA(!;%CoQ9uzFn7iKZ?nZ?{h@h&^ZkLQw#4W^4h
zz*5=q0d0n<mBIF?awfPSfK*7PjXY3xw{4!rDGY2Z$1<KdQ;G*x#$gn1;WJ#fjRm!(
zOtT%QZO($(WZHwuXe?oE2T<ZC$OyE!CDpFn!&0@-R(=h!FPGR!p<8|J5n*g)ke}{*
zCa0t%yGOT6FY~czb_DF@Lhp@|Du%&t9vt5g*e+wdij=?M_s}qq3v(4>)s51{;Xw?*
zrrY1+67XwO4#t7p2icvwbbbB1Qd!z>SYU+mNO^+Uj%6=L+$Crb?2!sVKH_cwCmA)w
z+&8K$3O-1xYa@HmjH}4E<H*D<O)|~1YrnaR>$CGDtH4(#UilrHbBA6PO3X*v#y41=
zvMq!6aT6@$FzcH<TsS>uj?-IlK>}gH^#HoyE$cE&OwIV35Q*EYyawBUU7i!GYnEQZ
zO14B>OiY8J*-`ROw`#u>wOd}V<zQldZBu&tq7C+bav~{kMNxL+djE~*iLyIUTH(i<
zyO+kjM(j>vCDu4g$etAuZYEnNCD&=6z*XTuiqT>oe8BjfaEs49L$>oxjs7lT<>bzq
z8=aRd?8f35Y)hw(zjcdH#ajL_ImP)M$Y^Co(huFD?$Ir*n8|f)Ro&6E;lK0Ktf<!s
zi4M1ElM?phCAhzG)aNw5$z0c)cP8xgHH%|Ex{NPc5_Ld-77wr3fnx~*-wa#R+1E^D
zvg7v<5J93m?O=;dxonzIGy}w{|L5SLph{u)4?=GkMhR0UlBU{-D}8H3N2Nk|h-57Z
zetXaTrEejNzER>wP%e#ZEc=6fPWkML5Y>LTgg-WvVaacR-s-J-fi`P>hZJw9T#;0y
z9BsIFV*;ju_28TQ&d!S;UE-a2*%AqP0xDVgWU2&T82w*kn5`Jz1PU_2v{%ljavdI!
z2l*2r#1-XSRe$pYBomTKP!sfQ;D+Gw9xIyf3{<l0yqxyBMDI_gDnS$)S*`=9OZOH{
zRMHhu>+nD#cgXMbqns~Foct%2vxp~jEbjTZ3LJR+Nca#j6qgnHWO0TczXZpc2R8mM
z8;%4e^e(`1QY`~RuQ`g~h2wjDzI!qi&yLZ<SX0SxDi6l<Fy}cCiSZuG!NZ-exhM&+
z=*xJ+J%*(lN#Y#_R|Fc_m{iDDZ_4?~k|^l-x(P8@QV{d(8NR<~qKTI%{Hf*hm5x^(
zW?s%rN;y&qgg`4ZqfaOOc7fz&x0c;XEBOoP8vaWwg2j$n^s^PQ`Xpvk5_7?8;enj?
zhC@`;l7jH4S1}^_FGu6GigP#L;>So6htI4tSrZu?5q~jue<DSTJCOC2WJnWnDJ?1b
z>U(T)vRde{EL{f#9ZmXQ;{a&FK9nx9?B_rO^&C9E*p863_cZLS7qr7AWIsU11~Moc
zMNYOKN~N9?TU#4;WX6H*%+Oa&@<+a6t(<Ac14#gqY$-EEW_MW;Ql>S!UT)(KNn44|
zeh+{EY*MU|VnZ7K2HDQ6T0~!L3j7VJUE_1Y#LD*(EdDk)*Eb|2JP>2|;nd=IiY!tT
z7QANEgjp|bcONJ=Z2QKB?AbGBg`BqIGD!vIN4Kxgn}?MV5Ww_;PC6l`v)8CES<LEQ
zHyCOb?xfZT_xWY)s_7&~Eig+|KbKC@Op7E+j!B5Ow)ZiSRcSoA`^=bf{)hAL)XNx#
zwaejp@11dlct{zJG9?VCF`p-_ojpz%+YsHBbKgH{$UZe1IA#>-APlgk{CQT#T(gG1
z*m$CwqSA?H$&D~TkLzD|@zH4ay6ot_?i_o-ji^mvMw_OuA<`nOR7uBcKg5Sv>rGDi
z&h$#sL2%NU3pCYdbO(Zh=o~~bJkuz#hN~u|DnFo7df%BV+E;4Qo%l(>x3rA)8hqi^
zobr!2kCdTSoP{wJX`w>H;cEuHQ3<!)W%eKS?ju&6Np*gS#<N38VdL#-qEbZ@6&|kA
zmdX(M<GfQl$tKs;2ZH!-uWI1A?u-(KK1a>G=RnG?jE#r+kH0o{cv`#Xj^P_Qp~#}#
zs~h5SyWGKsmIR}`ojqxH$yIyu78c9+1lrwMMve+5B>J7q5JmR(X|PT0jfD6yXWL-D
zYiIh&zSORLHAHT}geS)qdZU4)^&m(uNZ$U1tFeb>GK=}xS6}_7M3INGNr#z`Q>aEF
z^&@-4L!uD+NBX~ri3VZT!+yhoM4M2H_*9|Bk3Ft#F@Kma<ZcHgh~QfVr#W!Fcs^x!
ziw|$p<I1-+2KVzYy$ep7``So!CIn-+|Kp|yoy#KL;-xINi7>6KiCHDH$CdZWOpvE{
zc#q8k;}lYcM>g22Xpd0Wfeg&KBRNj2$G<n^70DuZr#_X2SBzt-ELFLeijqZWa2|IK
z!&B!gZJ(@zyjxSKR&Uablyf_P&<%8=Xb2!gSV@&1>zUFvluM*|gC{K40v~5+C;Q8$
z+pEq7Kz9Q}Hh<#P;<%}!Wci}$fWH(<tSFcCw6y5l{9V)tA-npxCUsLvg|qgDZp|7>
zStio1PHs+l_HtD@uf-3P;t(A4OT~$YzaHgVOGfruxP*TjcC_z0qPIl5F~|Tw-X60W
z?OJgLE$II>yZ?oIo8b|aa8&aLj6DAM8CiM+vy^yWNrc{>Mhc8J%K65bqTWTaB=dK$
zjU~>)UlON$@Q3&>Bv#fYvX~3($Qj>h)Jd{IN7!Wbd)uJguoETB?XRtPT;u7nE{>`Y
z>ycQ_W&A9}WG+bSN5W*@#U3eWYGu@}g&}mqKoM_%<dpA6r$tBqH9Io18y9ocH!{f~
zx(TY7D;`20D4}tEMHUdn><CEj*@hd5MmLJ_8j1{aC}Q5CD*O0OC5t~o2%Tl%;S?EN
zVQ$=fs#@fHuexZ)Qh*ogDQpvj(4&Suv-%X~`L(>yX^KDkOU71*aAJ2y-1{;E2C|Xd
zs<)2$mXC|qm(My3IkwH?64pTc%KFup_u<5Nqd;2qd7~-aAL07s>6}z|vhFW;g+Fe{
zu1ZNVyHRYC@DaDKR-Cv&q4U8>&4i~2;<j_)K&-5a9hC9Q4K7jyu~z`P<>77iS?_Vn
zCA0p|MeeaPCve&7Y#4yMSI~nV!D07SzRmkKfB3*m+}68VeIDRy=tZW#(hgmYQ!@Y5
zF?lJv?Bh6EYc-I#El!M*XdeQ{4>|P69jRou$r$7J%k=Ns_YYmh|1Lm6{BCPwy5>di
zjks5-A6H{9!^O4yiqYy-?r*c}(!^O7gcs7wTP0ZncwbyGYES)PGbh2Q3=~DWqWDT_
z=E_V8mjZ|=-QLZ)dRql2J%17x=(aQ`LDfs-GUqy$An)*-B3_HJmxr^|@YZ#FW{F$u
zUg$m3u;W+%%mN6$A1wcj+<;sMJStzSymCA1V80B1I#y|D95>srrI9b>#l98P2!O23
z?4UA}F>d8en^|qkAi2h~Pp2jKuf6aqZg|Fe^bk%(+{-evK&A{C`I0T3c$tWtCEg5j
zTj#}%q(**f?q+w1KgE=8s)6S`n}^$46+ljTr|zTBH<h_Azpiv15B68$+S0v!bzg$x
z>PTC!of%Gg<fXN1%1~l}G5H|~pV6=2TsMq~8-Rx45}-ne<XJ*@Fd#n3iQ+@LMuS|i
z-ffT>s;+(w^OJCLw*6A7$crpe7mB@OrkdcWJ%hqXN+tLc<Gb1S0+`#IU18tGfNU-L
zy1ubs-Op??b(u8$oH}!nKpW1=z(_oyZr}}FU55}rIz*jUM46q9#ldC~{R0PL;w3sB
zcH`tmElY$caC&Jj{-`D9c%(Mpcr~xCyy4|jh{8PajRhU4Yp_dCw>u@ri%pYTjSeLt
z;_+Y5zpj%h>6Sj}D+L+0J|hSDb2%ZTJNr_O9A2cId~31+EzMt{#bH}bWzXcKc-@12
z8`6K=P3qF{7A$MKV)Oqo_SRuhMq9(agoKig64K!SB9hV#k^_hcDBU3--7QF`gM>&U
zNQX#wcXxM+$S`#H?eU!Rp7*@J>-&B0UtTg?I`ho)?7i07Yu$G*I_V#`dNV(L-Zxlx
zHO@a=1u-17WXr>~8dc^@eyLy_kZybjgRW<;Z)2a6SZY}&>kXODo&b5(Y~Om_eq}PB
zM_9>PO+^87GdbbcxqMa4HhvuA)^bjwspdx!Bcp=EVkL7&Zj?bE`8MxY?zR}enfKwM
zBk43eJOce`K6}zpjr{Au<$iqHNwFpT!-PH{A~`*<5@KjLORe3KATXRGU9`DiGKmG^
z(fYinFIB$TGLMjB<fdW-ZHO=}{o@>FmtO|4O~-K(d5qca3rf9b9j>rOwXL;Uy{h?L
zz2GjK3&&nw>ZRbKo}~1md^G`tzp9x9>!Do$g2#)XEB^*AoY|LNvT>~sO5I*d5QK6{
z7EP9jtct|ddV<^Bj(F`S?A9ax@{PMs&a=EDFDK3fQ~qWyROk!JMjiH>pXI{W3O%W3
zRZZYLrKMPJ>a6J=Tyr}}&m@XQ26M6mQCdS1gD8V2Ex$B)4VDqDgHFTaa`YfwH*GWY
z*xNBk8&v)Fz*dlL@{Wv09OxIR`VCcHN(`bHs6lhMb#9<D{wvxc{Xl5AQiJN%cEa<h
z4=D|k`;@ThT1Y2Zm-C%^DIINSF!IO;)QTb=pjztL3BOl=6dX8<yPXu#lBAfWWI^ij
zY%nkEJB3uU&b~T_=(ahU9znc1tQJ*Q;Mw->i>MppN&haPGVodi)8*8laK0zx%O0mu
zYYe5g%fkzIF-e^^1Dp^7G5r2LuE{VAW0l!3t^kC=Rlkm1{+dr~eHnMW>0&D9lpN;f
zmyrZ)jK<m_X%&f^#GxQG0?ae1rXg2kfw~Y@kU4U8a^0F+O3OoF(?hKFdjK8n9P8Hj
z?Vz(d5e2<sC@;zS2LEwQC;P%=sLzGbh3)K!%e7lKgM$hP1&RR#XcsjE0aT|>IZl}^
zt)B*Yz<LW~d~t3J8W!W*oag%v!1_eI;cShA^G%VxVVsqeNaOXR*2nf(5AdFzMNM$w
z5aNWkhoVU(Znt)$cd!jT^1d4M6VkP@1G+6fMA+}PTS@f?pyG2Af=E56PCG~v55<6J
z3nwhqq-~XHrA(ae$Bo%jHhkeuBC6D~a21<c!TOA(Gx$D=i-r{~kPTV87q`RA_@X(z
zv_&p!L7~NEXw3E>-OvM=95T811ioaS%g~>CnJH3geE*7eTWHFlv~`_%>(DRFU+~x5
z4rIsEqG>00)CNIja6}bkWnCsNrr@iXrm&<?kJN{i8eKgLu>*~9o{#XUE#)z;p$a;9
z_%9UT*ooG_i~Ffl<9Z##_USt0wm@gl7vrBq*m&G$siHJBkK$QY_xQU?Z%c7ssA^b(
zp72h+b3a#s2uw~&<Ck?$@DxQ~6dN*%HyEzmcE;t~CJRZ2(a~HP@^>=&KD_9WeW+PH
zl#QDd%+=136}56Z$ObqYl1}kL>bo_fK~AW+F0+U~IA@cEM+0?y@z$P=H6250owc%k
zrQzUJiN}VYWHeh4hXw<cRbg3BNi=9Qid>4hwqPQYQyYrSSQDq(^t`DKvT|o{b6uGp
zgp8Gr<L>g>P{WDH_|=}qVHjides&rqi6cixP&m`B^HGvpajcMw1~Y_*2;PR0aY*Jw
zN|Bg5WGaiF_cPu(4ysJq7dd>|Bc#yCJck<0_Y{cuv@tHbt51eKLY?s>65p7d{ZLm}
zMsZI%g-1g-;NvslG^b9z=iq+aXM)_K8>aqXPo{SZr{-^xSilYyvonvtaa~leFE=`q
zuIspz@r)RK@@#N((Wey!s@Qu+_I->gxc0dSEmb{x1^KwzC=!hN)F5Nq5UKELpNmyW
z8!;>*va?Z@K-V~2?^%jnvHGOy=&EU!53^edusg$M3<2%NZZ5J0?<1CaBc8PyU-n?%
z<?ZpbfLOAw5Q9&8Ox0nnzPup5m`)?5;$I&Ug)Z=k7qiWRBw+MiItJaM%Rc*Yg|<6w
zcH~uagcu%okf9M{wh};N_?OA1&&e0^2iHJHSNZo+gI#!@QKt{37s-e|=B8{Wi)_q=
zYwpDm%wBRC2ZMELh5VY5e6Fap_0A_%4kn_D@bd7Ol33iB8FJ&*aleD#uY)Ngl(QGE
z>69Y^hiArUI<N1^7<EU(6yiT&hkpp6A&<To|3&Cc<X~9GmLcT<o^*rQHlot0+6Beb
zG0)f8itgjF$DG~kv-6-SW)Cx5goC<mZ5nu>w)X^%BW}BOJCX6kPYg8YXs&&+^Y!@I
zNyMS)IT3v*@#V(nO{e{_!{Pxd(;4w~zJoeZE?L?ZmYfEI4_*zSbJf4KJxqRmKE=p|
zT6T~4Mu6^zY0NeAUOs~DcJfPOBL0_X3zUOFwYc@U-MHOq{pQ1irDVxhr6N4ax5+3n
zjV(z%Y^)rrlJ-ZZe^u!m$hygJU`9mRx7lXd$JNiLUX=Fr+3z|=>em)Qz~<AxFq*qe
zl%ZyDQ>q}k=TzcD>Yvzr->bBj=WUDNE%N|~xx!xYr~zDU5G(yA6*c8ELlU67V5^{M
z4kwEAv=-mDwp%Z3bB6i}h|Zb(`0k(gUH-nW7iu&K%)RX&l;H}>KC~{Mn0f2w3M|`v
ztE+)aMPsfx>F>N$zVOm*^oy4sQZe!NY{KlU4GrK0gMs0lKkk3pr}$<4z8buebLPt1
z?aG@mf%G0EjEhy87g7t+lu^8}?nwIHBjy{4f8)KlOwx-}nbU30>%MILV&57uoMdQb
z`UOP(e+&PjXh#=o*30|-%y~#CH`39JXsbW#_1FPYV$;5OM<`jx#jJVZEb3$dDnpUB
z<7j<vHYu|P^`2Jt`?Vvv2~DF;mNb+E)rK!`0z?NQFjL%B9dimD&AyE~*+vl}!8+?%
z#JwhHt{}PeC$TJy`$Hx7b3fl=@?L)KFyxl{(9dTy-3&-~dkzfIrQo0g>u_pwJ@eNQ
zD@|dhVRlOV^fbgw*QG2<!ur!mAD5CigV`ClGp)L%R~qdo<Gq@fG%PR6C89E5F&Wk$
zTV|yFzWe7e(D3Hbd|Y$_|FTl7$xU-zpw4MC1|dIpeEtJZaPZqINWm|-i+;dB*$+uC
zVKNJO!7`?5Ze6<F&z<Jm)c?D$@1@y`P1C2lx0?>V>paw1%N~#>@x3r-o%%w4C7?zt
zq_!<!B6{Ss-Cyv{7SdncoAh#`%kNA0Jn4Ium)^QK0Ge6f#*3zYeDtmA8J%HV9(nv(
z8OA{Y`HjG=)-D&<)2$por^;jJ>JFL}4B1`rl$c5e_0#EA2J=*!tD%VJF!?vmmTR87
zpU6v}b@FuY@e;51mk!Njhq3ElV0KdG&|1!$-;JGL?|~<Ga+uxP@9&e9YOLxZGr#Lg
zc?=D#P!2rs6c}|K3H~7NF(5wa%SaQEf@Ah=4JN-nWGrJ=zuRxuJvF7XF=7Sk_oQ%D
zH4`Cjx#MHBbH7@7ZDYv(jrT@FHjL%F4&5iIe&)$O#?4P7NpuM)QOk9mqL8e9zmW37
zd`Sy#S;ZWjo!$x3)3KAIsFzEzHrCG?e9hfuM>G>9zWQpitVv**2rfcquRy!Tk3;5l
zWo~rm3f#(=Y<cX_>TYF8?J-5sexik$_^n3Y@HN;_-cV)aCtJ_4rB?IROezlBoWTe(
zDnXW`^j!;cOQvBAI#<;+TyQm<C0A~c)8WE?ut*r5P^wbRsO?|qka!6u+1}^ac<#0Q
z;w<qcMPKXi^}0sBF8oclg*}pf)^zS@2~mY3=zjkkd}@MW=_?vJ+I2bkAAW8ZMc;Sl
z{nO;vjr%;~ND2h*c@rRxKc4eQ4skSlGEqkOHmtgj5LpnhvYp(Ee*WWm@*rMU3ON&x
z?)#s%%J;4k_fRH<Z7nj}vD4A1M033oG3L4^7OqQiy=eYb9P=P{e5je%Mt}7%-595`
zBF1$-`GgR6qTMDL?R5>3`&uuhOkro>f&QkEVK`>}&7%o;NA&YezfWvfkxlOe55B(R
zdp^8|J=Zyt!I0j_K>Fiu9O>E*j+^k3{jF;<>6G-R3<(U52=lvYSN!k`chF@~BSj#P
zy>X3mZm82bPh;(X$^FiL^U+vRi@A0U+<fD0NsKmm8(lQF?Q4?VO3mxiC!X&X4jk&U
zw$*IV(zyfflz3X4T5~TuRN{2+1qL+64w)@)9P?NH&Ry8oTk<a);%D9u2SH=2)JQa!
zip^cvb>dSmv6r&Qi|x1X30Q;^JY>@3#VFCQaWA}&61Vnz&O|BQ`D|44rXYya3|1a<
zOubWsyYH#4A*UeeEq?EiW%G@qKi2=?bo<XXJdXmIl}&z4WkcW`Y)>!P3^0${+#$T*
zA4l#t4uRy?d^(DEkmT6~XO(~;j(+s4D#F1N+4hvZD{ttT&rYQ>P%v^Q7~S>t(l)NM
z>zuI^sCqszNFw~H*PM-X?=G>0Y_s5RRl?T+-_>D!zK3^c_5$l_6Sfi!MnBp@{H`bE
z-4u;B=vXs;d$pcoYBU7%bdvmZTf$iiW+-UvMPCZGYK=YXBMaQ>Je+RM*tjXLZ$Dk<
z-y-Kc%I9B0esHp<(f-o{DYj<nA#rnFE%1o?+wFx$Z1Jf1t<3D)a8q_ye$o2c4ck1#
z(_=;6bn16~dGl%e5g5Vat+ZaFRH?D7H$6SQr1B>A_X*h=ElDnmJn*%CS>^K3qenIg
z_Hi@mh8^5FefLe>Wd~_#tg+?9!bqRiE32len-X<#!_L$HFkQ!C7qG?RV<kl=N)0;p
zVjP&eNi)NzW<|mg%en>+c=4Du2M0oaEZ9UYRi`1gM1L`{9o7xFr+7?+lm<gm{Ylw(
z84gHyLQR*debju{i2A|~op;<X7TfIukMwq3s$Gd>M<LkL%kEEgiK&Y7`p|hR*qkMW
zpURJo?f4`HyYf^#a~sNx%nxIwYHN4q6ppT?O0d;pYISa)T)RR<oQZ__hVMZ@qut@9
zPzXY<=0^d<zE<*L)u}D_=Vv52r|fFRz*k`@?SR|_p2f3NpFw#=ivTwlvV)fR+6S|2
zvp3a+vm$W*(D2GAUiJ!4hDU0F6{=|$`{9?Cm-5&fokBfsg7sN(=5>NZqiJj1my&Pd
zK2h07%v|c9%<CayC)q|-Pc-@PEAspwIo~-as^Kh+IOg47OaP}>%j7Li&Y$L363YG|
z7k%+OUE7w#Ubsg1ZM(gTqjGY%t2%f4iH*=T<#`LmyO>A4t8K}w)Rlb~U(Y<aHxYVc
zHZ8BNeu!0lAX}Wn(u*i~PyX@jvcPBP&|8ce?G(&KHc_B0KF@b+NfgZ;n`~ac?T%kk
z?biROnKXK|(yqt>*80p%(%dzDIN>5$_DugNiGVQr$78)cm!|2MhJos;(hjWK<wokw
zbL2uV0EBL6Y<TE0wm><?r}94j5zO*o?1&}z?+WYisIfGPG|%JFi4a<WolXYYs!fZM
zi$h82hYn?o#d=avt~_p-CJEIpXzpo^w^hE{&shWoHV&>GH|VCbt&t;@#=rA`lryl}
z`Y=;Pv35=Q%zIdN_MFSM#?2<u^$tDN@AM+(xrhAO+-|qq;lgMK6-fMbN5%m16Q>R0
z7<)k*Iw5!kh03a->kDc$#~(@84ujy?bqch%&8>;HyG}<Ht#G6NvuJe=;ePtmnoi=&
zr+CLjQkUhtk3(AThDqs%gPOnD;?F~EF694Xem*Fw8MF!NBvw_m)>9Jb&c>y$(D{8#
zmZ&Q_qezuzd6cCf?rikxaBqb^CPlmi;phJJ___^6&Nh)t6Y&pw0qT~mM5wwY7Tk)S
zE$aK+k%6!Qb-<nVM|jx%81LpRIDT>;>~^4GQ@(Jgf?rmS`7fCzc34>Az}+`D1h0F1
zc?o~}?71D?d3E7g%znv<X>!-|s93W`96u|_rgEn9RLB_*jWcn@BZ8jeDNl6$hkdtK
z23})Ne>VNOxk*)?dj@H8x819zEnoPa@6*>opFl^u^YiOF&5e@egCT(}%0E7(>J{!(
zRy7m#y$?2z`=-M4LpOc?_9U>qnbEkv7MI^_^xz2bD-x}s*ic&~BEW$>-^B`OE*mm`
z=A&yTY5dTvd95K*H+U+rIAVY!N8t>{e>FtoZL)2c9yX;oc@bZuJJVp8lW3~nKKiyZ
z$AhQZC)sp>ZC`Ft0=L(l`R?eQFh|5eCYF5K-YT4a={i`irGBDPgNW@ME25(f8jizo
z&Uu{Jwb<%nxMP7ct)dOj8YoooT{v)TpjmO}!o*`lP<qC)IGl)Ofun%?sKA;^>O*By
zLL(Wu8>(y1q0XHl`umvE;OmgadIeFd+SfGF=@@9_=q|Ui)Bc0ZMoZKjKvb}26P_}v
z-U4cwOi(l^Nuo)+V~CBW#`H<<r^UNaXhvv))kJ8h=4S6<JMD0^aP#iF`*nQ1FT9Li
zk!-xg<bpJBKXU8Amt7x}heMxvM^}yI`?((2mDu>}P4lWp4y||sTiPp}P7IX5h7%AK
z{`j5@VlGzYUHu_^kQs<Y{&5~BNi7Z*iic0HD`ocbzUjC5IC^8<M<3q=2qe7fY=0&z
z`H@ri4*tWhnHGP<4z~ViFP*riU8kHZZlxXwY`rO&67cjrEj`gnJ1#UYIxZaYi1(!L
z^?dhK5j)lJhlokeK6OfU5L4a#ZQqrB>lEin`vVuE3$VuYkufsZ5FY4txTU!QTrk?H
zNqaRR;nKFX-HIOE{{V6TXbpRT&j%0h#1U5TWe4{(TaM+p3m)wlyeG}1UPsnz^40rF
zq`r3|D6TJI`*YD7T>0Xd|FActidneN8POd-svx*@F<B*eXbj8<j_(y~yJEd0mS=#6
zGX-W3*ce;3d=7e~cSq&OIfAl^HtU)B+~(E<x%^$3UEsSU`Q0v6XBQ2@eB?ITKaY)p
z5Mh9JO}t{_q=sC&3($Z2H-3Q$@L&8RI#oVAbdXTwb9aloYj{^jXdfBcDQqUbSQh-g
zDynqE$sdQ<r0k`pAMu4sX~R8rSs3D1-b8T?{ZU=|rWRv*G&QrI@nn2Lv;8h{sO?A6
zQZ}sAk;>gvmC?8<i16I~_>KbtM4d8K2P@6FAOMX}1>^HhH`_P@F{Mee4?2YtkxdWC
zMEP4;q1EDH<1TwMVA%Jg<=ulyk0>TjC$>7jW!xWXNs~o1%QC6o!dVMTC4=~i>l-C#
zBzR_kPKFCH>ts-c?nJftcn7kML5)U7ytRQhA0vjmgP3Pw%)Cb}NM|+bGZ1wvk89$2
z{mIG*$F*{ovZ=hl@_DS6Ro_;8Q3AqmIK*5^&TrfMg0g)|{&Z^OnbLND?Q1?&F$>>&
zn<-9@PQx$i-_3>}j1emD$aR(9aGi#Em)zGxCj1e*r?g!cXqjTXfcVA=QL5;0ll!Vq
zV0e)YSAKg`sdR;$qGifF9jU3nU+8pl{vF}UgpKYPqlBv#C{9a+gNg3{??0Gn(?21h
zSsdZ!S5bJ<P^p+6Hv_KeB&aM5E2mZa(6OA>?}^2;G)wg~^EWcwaLzsx;Yz@inXaH;
z;;Bx1Vka&7LA+Z!M1q!slh~qM%no2al7)GaaGpR_9zh_b-Rj^!jChDn|01b=`A0{|
z2WBX{M^0$gq_d5PDT3GVv-MluHEBrd$w^hy!gkfYilUsuLOI(Tq3f~0DikE75>SS7
z6|+3<leev>bW0DGqVnsC6yu?T*aQ^dp`s&eOBZr{5~f+-D*cmFDJKbfsC3uK8YI%E
zZhiD|*tGC1{6DF3x1fZoIk<dNV|Bl33oIR`*X4)SMS6Jz%)!*N?His?#i7+E^d5;&
zUhn!Epb7Dq`Qo-3N=;+R=#a9tq1kltDRg>uZ5Rxq^;S-Uuu^8x(;7@Pa>kNUV%S0K
zdyX=C+{R9Npbjx8@q~~5B8KNC#(c#<N#7746&sJ`s+`)}mcZU$f-wMNl=>$$#4rxq
z{hq|g25g;}9Wu%~B%7M1XHVeJ^pvC<%~IB-IR<HE^`Z(p+ldlPeuJ^;T=ETnCGmA)
zNM83*aw-dTLAMKM)Z?b-SkIJbDG-5C>BU)0(s(??(eqLMQPY5V(-0x;_3$vM_Bg8;
z79(ns;E~?rIAyMu11keBPK`Yo%E&1?fnll%0pwA`O>}qQQx4JGCmTIo8&S1SjnYH$
zHVzns5RWZB$5nmO(8^=G<Nk*`mU*uDP<ZJ%d&rad+Tf=RfUBp(_-*os(@S;z44P(f
zX*|fv3^S9He@JYq*(#H-RY^moajO8?gJX5I*Ev_O1Sa#c5X0v54+IDdGT9cC+LcPa
zS2)JeoCvOCrbr1f*_trSijPL>fzy~iozno)eXN5qv9xDa@u4`jsG4F5z#!>$rXaMc
ze2Ca>YP$=I)6@hl0;H|EV+VB1M@bB|-#$(!tz2-??!xtM0`JaKk8j>i>QA6Bw6>59
z*w9@@o|6jjPD{tyny6^oCx_%gGnYab>2kYSmxw|~q$7@Mc~@Jo*a3}!EoZ+;+eI6h
zQdvzi(i!8ki+|ii*3DEK!~2r@_Ci$fjigzN1rULk0n*ddApaWP)TOXy&gI_r!_ilo
z<|Iez8?Fl%Wk7{OZYsH*%~}&&Pe+qQ93$l($a^7i{$f&pvr(=|4;wKg<0}vy^agTK
zEYjKHGr<;J%Qf;!8@6-<06Sc4(8G(=S$308C$fPCQ>Xph7IzJ&_+~%#R($KJ;4tSH
zUfmHF$H64#X#bxN_O{<G8L|jp%giUl=<bvO?42g=cH6&QbW*o*4CMTndl~Pdof{Ah
zE$grrEXbv$ktJ@sN#FD{K_#J*_t5=AgT>+3Xi&{<O>vjdze`4QI*z!gAy-J~&O3ix
z4|gdp9Pjg6EtZ`T<jO96`s}Gv9}SICdVb9iCtkgOg|ULhMX+@6VB0eQJ1}6cq-pUU
z3r;2;<T_^KbsWZdo|?hO3&6p0Mi5|fgpkE#CyK{cNi)4zuxrD=sjYV;!wk)79pGE~
zX*tw|efZHf*A5)uok_xx$z>hmd1`YwDn#|P_j|_*;nV9I!=aG~mkncegcHzrRCQKd
zudjiI?y5N3^;<^zB{98gKm?Ju09<Qqk^~zP0^%<jePfi5vM2No!NW#Cf`X>13~n|t
z*~Xp#YC(-N?md5Dwn`kPaY8x54v$F>2_|(Dp+7z;j=Mt6fgNuV%G-B|Xh#mQV`D!M
zY~zNgnMybReZ<7u8vs7Q2FwK<2@b9cV6t>B#W<6B_7x<-Px%i>Xo<y7B#P;@GDvB~
zC0sI#UsS~^xszIl2Oa1F981x~n9KLy0j{pImQ2Q@jb!&o0(Mz@U(9h{@N;*(xgrQe
z!%^<`z0={b$QZvvp&q5a$7~9<bu6kRe9yPP49J$d(%XQuGqypA;N(JcuaL7F%1sbN
zD?<R0i~W0MXYILxI+*yK%teSMpQrM|g@i{bONJ*cT^6exX*nKo5^wB@-}7#w-=FVz
z!4DpNypJj1e#d6#)&<!%<~p8(j#f3M<=yf}7s<EF>NQD*S=!O^XjTs`uyNqCor`0W
z``YWWb#>J@tfI2lF?>2qW|Gj2(~tBXV+npHeZ=$z^wrK|*XqvrmI@w5Pv%rdRdap1
z8NQ>a8`5LX(x>urvxv#8#@I9UP9b47A_vmGHB>cFq^@T~z%@`aI-6$ZRVM0;xqPjR
zsSNLUglzc+nMF)rG7^o(8lro{E=Z6I^qA-Y^hDCl62#$K82@d1!@-2&qm#Yjd~6uO
za+fT+FQVRbsPji?sJPbG6o=qLvLw0AQ74foIJSRGXph=t>{w5LI1(GG@<vt}!~f9l
z7W_1&lZoXTHk{zTw;~FeZwLt#2BrfF-1p|6?||9rkQJNAo&zphk%h*#b;?8zd!@$h
zXsJrG%*Hfi0QNN>4#AFY>DrfSIy*?ncZ!t)?T3B8HrA9+$w32{T*l9&atFFUSBJuk
z3Y`qud`b^jcOHN6L7+t~OEO>Tg%$jidq|t2N{y?A=%Crz$%XZ<?w@L@Kapd#91J#c
z=ZJ+ZzTDOr8s&C25$YIeCr(jqr+)v2kDhkPCevp;7I&9sLtOA<M6QRjuK(HR67Gf4
zOll&an3*}G_U$&!<!_F3y}(^o^xh$9et(8FqS;MZo*+5n`qEo*C~o2iS_WInb71^)
zg`Zi%jBO`pLnHW^i&2GWZd9a~cws3zC8|0Gx0Hd}Zo}IoKkc-~6LO5<9yG7;klB2e
zdena&K4jET^xpk2B@^~Y*quO_{PO-r1Q!PPY~pz3{k0FJJ*}_nol-#`h<EHoxINxQ
z9WzDBBKv0U8Kt}OMzg>f1NOKjBBK7KOrTzEQ&@3NZf?|If0x=;{oCtX45Le|_vfbc
zh6p{HmYj&#H((8{Si5}3-!;i9(#BaE-<|FuS(p<f=F+^hJ`cN|pJ9oV{JoLB@%1?=
zh}qcp@_D=$J|sW{Q_ECwO?IEQ^)<p*LyT#M@A}iCpGY$Q!v60;(tn0b|NU=kILl>t
ztIc4@=4ZP<w8tfqiI*M^6HDA++O&>_e)-l+q%61{4L&jekC<0W)sOZ*LpZe)lR#nY
zwRkfz*@@`$`)Xk2#-l-(z@%*C3cY{(fAjxBsL_PcR#yVqxM$+|BxyNRiO5`L2<Vrk
zV=KC4X@Q;K3ILf-c)dVw1?H)Qic|?}jnnGnK#+C<(DZ3WKZ(QcqoP1!@L@@aLzn*J
zf%(6_fnjEJ>%I?LO(=p}ksFQgm3@GCDcGlHLDlLVYJjW(=;&-0qT%8O!n7QECYM<>
z*WYvjOM`%A&Oj6>>%gIP-wdE3S)dMd8h$yKP!QV5K15xg2;wsk0NUtZj3#K!J^Q|K
zf*R+kH35LS?+qujeIP)I=2SNI5+x!_!-+@`?KqR>dr5@@qsDxX!|&cJ=<x8re<%NQ
zC;ZoI5xD4Y;?GZ-z^Jk9-lYyhKZwOnJ|@(+=O`tc0W_ZGp3fId^ffg9L;5)s6G_*h
z=MjNr<1f`M0^#4{C|OJ=#(Zoih@HudF{wSa>4a7SAs~ytF3Dxwf%puk+Jdo!N&w7@
zvVXjG<llTIZBUP7>13*n-4!=L(1O+J?+Wwp*!y2^>^}nXzdtnSCvpS$rJuk&^z`}$
z=$|YBSMSA6Lp2!=o)G$)L}i|5_IEPcU&{|-1j*?B?$3HS!l(0&CsR!~KQ@R<&#uDt
zko3ub9Tk?a<_fL|GPK!kTMe1?$?XaF^-dh{%|{XcKAidAyqo{q1x3B@bX@QqOk;)-
z%l`lU*8k=;4jB-~*hTn74`Kb+E!cb}z6|s^Wx>=Q-SGgntPQ@BF4;_mPyf7QA?W@|
z+~yY=LLkJr3G7xeq5mK0&Hv>R%MhYIKXoOs<G<fo1R6THGfJp{*hBeCx9`n5T%{5K
z|Ac(=N{&!O&ov>y&a*yfIQ_Mk;`U7p^J&wS1B&dSMMwiCy?6ojsmiBIo^}}5pCO9B
z6{rjvE9U?zKTjz|@W;N^6}ad^=&0ZuME`@-K4&Y3sh?)zZvddO`ok2e6DjR#oymNz
zYE9~Y^E5WUM(uWG(qfHN|Ne-P3>;RIPXa1V0qApV4duiA2{3kFMGo5`yufTka3z8x
zJC>vL9f)Jc2?V#({g-I^NPji`@u>sI5^Eq(rS3Cvx;g=HMRQpi?}}^iNIJhHH9ZBn
zRp#1}Jp+*(4lGS{!3WsmzNWc=6NVWu{c8$HMz8m@zTk<WqhM((&Og-LX8QEZ0L`cn
z#3dfQL&8T+9@dij!OaB3zfp`Zp|rN*%gX_|39n&2V1^tx0p8eo4K2$@C?ed&wTUoa
zFHR}*8i4gzH6T8fP;Rh4NB%#r2}~Ka76|l-IQ-WIfxX5=DQTx^Zhm850%YKxI|@r>
znRec9{u9jo$v&cA1{_s06qV}_iV#SnSsl9wbmR-gV9m*ZodWGE`y~JeT5y~7bBtcp
zEnN?i=((+Az^oYGRui4|>E9RuhTkhz)q=wEuzgT@QGQ~h8v)OdU1p#;4}tUNU2X?V
z)o(s4;>l_o*#K^WTsDA}Q|{CQu5ztYraoNhZa|+LrS`o#Z2T_#v$km|;?+PG04Or)
z<COw{-W;$;*7J_qH7!MLeU@Dwk);8#rS<?y=0--X!|?xIHU51s|MNATXmKL6<Srf~
zynp{Po#Zmg!79O<r|{QXy=fAc(1`q6IV^b>Uv<!f>Gr+$A)gcOI$)^s5EBkk@dBvJ
zaw^w(&JaK2*H~-UW9)z%V*`v6V87vKm8<ucyfqIl<Ikl(H&rQsI#fhy^2Q8EODlAN
zR>B3WvDLE82SBKJQaja9eB5use-2Qn8|^1k01xJE?pnWnqeq^f9F3|eyrzsfrJmhO
zw)aI6xdD11o<Pu56DS)+EC$nW{y!HgC@0uMzqEg#c|-WGPYcCBcMkPiBhRX344e6M
zN{c@DG@I#{E4Wb_`zj%Zce!X&X>9BEd6@_rz+2?5ghUNf`5$Ip)@`*Gz7rMfQ7<{@
zydIID4e`qguxkIi$ut8Xr_uH_3LXj=zCE`X|1LN$G0g_QSVojg0tfr(r9J0=tU?AS
z^hDR+XTncSF13izn6yJ>ef}AuLV3_pp3Vi&dvh&|!2fZA5|(L%gN~*OPaPu)jY_jQ
z@O&uRH#55pJe#VnBB>`LUMkSPehWBr8>k2N-CSiX4vHKq#1g<pndH_y)(1Q&7GNWY
zE&@o{6X0i^*3hwkj!LJ0&N~kM0Hmp5!121hWX9KYdmY6owS1(!XxzsiLZhkg{R4%{
zmC=fSZG>wa-VTUIy=&_;K9UdUm%-1smioJzfUtN@8&n9dzNT(9oprGP1d&e#iuvgR
z$i8zx7;FocIL_5skUmot<f-L8WXv&Y7MesQD0613Bhe0RLX!ZcXJ+fQM&V5{UVe&V
zTh?i^^`d0Me8TVcLCLja`<l<vpl{&a3V3Mi5*C1?)1Y0cR_{1pb0&Ea37rFE!TnDq
zUZz!3*VGP(eOuqTva<nOALH<+wcm=$Iv+-DX)hv3DA++d09XrTCg;GKYpL7Y2XTI&
zVlhrAtU?c=u&Gepo2g}j5u}0#^Bt0@n9yq(LobL$uNUP|tOfoMN1xE5*C?J^&Ro$N
z;NN*C7(cZJ*&;81J&<eLs>zxUeBLORulWVcC8#hW`&cg*E_j?pW)7ImwfC%>Q)}N(
zsW_1@<|opkg)3*grBXVK{0htF!Rsp{HN0C`HDgABBD1b~d{e6}*e*czv5Dm7@LP6S
zn1`XAi!Y8~1d`S--~RRCqo88*-jjVA%SWY;_JPuAOZL}D@o=NVpvI(IKHCvjc(E8L
zQfn{K6@zA^^GZM}@X)z>dbjP{W#8z`o)-MPyKe#rVSb)vl{CXE2~{e#3-vO>bdJLu
zhc7GsRP1yUk*&zXeTcpDbzNpUE^UaINM9THN2zK~xH>zF?NjU%Y4fUpg?}y@L>I>l
zqcwoe%zqBA9b=177Y(z_p9@kSkmaO06#9GHqxMSF9MZqGO9Jt~+d1>|TiW-HZ%}+{
zq>m0*kkGz^=%jN_%_W%5fOJ+~h)fy>QwHLX+7Q1cC+Pc}zfAsV;Ul)1s)I93dWJF$
z1GHYs!C;{G5s;RavDN=>ZTuGUd>Tl3y4KOZ%DyXG@&XcNo0+${O39PcQO`!i_RxIh
zi$8!h(5i)${l6mmwt!nQVW2KE5Nb?RMEe`m8N5y|m;Nlu`vbXsIvj_VS_cqL8SX`u
ztx3^b#6WP97|k!AhgdZMxXl;FwoAYoZi3K^8r(wQ1=5Dxh0Ql^KW8sTE#-;JUjtyW
zB%>o>t+oMk?<J5Av>lxG-T>p}aY5!K&K&R)%kU>^W8>WZ>Nwy^%PINT2>l%RhL85?
zfOX83uOIpO2^6j#;s<04xhj=RF|@Bx+wC)~&Fi28pVPW1Qbu{ZbXa0_Fr&O)_Y?p-
zTY<^Spa!58r6gusTBXjuhjTaPfSExK4~6&oQix1B@an?009fE3z&@L1hj9SDpI_OK
zMs*SZ7u~-v$BM-c{T<8HooqmmbZ~QdP?gZ|(^Rk0bVtR0H3c1=P?{5Q%N~t>04)28
zr^w?S72nxDr|BxJ;+JCguqed<*K`_1pG}Bik`qU9qAA6Clf+Y`I`bf)+MYed)}(qw
zgDKtZ?xrE-0tD<d)5&0(TB9(A{9D&LF@oLU0)VG8oW)WoorWcp%`#IZ?*~o?`uN{h
zpnCq4_L<~#1E|LSd{E-{+{3xO?A<l93_mk55xMjU?Y9@QBX6=3>gi6$>i_|EDDm~a
z^xHx$S+&`&m(Z6FgEkVGteq7gL&KZtYQ2)`-p<{0!TcFiGqPC2YCtI(3b{rsNt~ki
z)>s*-wShMQ;+K_G4FfdBlKAK>$HCQZL4p#s!#(5NazQbQ0{UP#i}dg-W23QPu<WDr
zAr85VptWxRX<fhAtc$31Z4;Y4#+3yZO*bC@J3w7l9&P^?n)-E6YBW~_OwsrXe#m#2
zU}=0jHKJINC3ZP8RDYcv9avZ{_x^;O>KHs%6WfsVs<4dW#)5*&ohen<nVh7HyKl;@
ziP5grcJ-LJ)?RNhL|disf{h8W6LR!-I}~3A<HQ`$ECLba3n}wc3T4rA6B^HiopqnJ
zkS$Qe3u~4xqyc=2s=0?K=WD2zb!>@vIxl*}n^JFXpO*L_rAv@m6;atra|jSMYeebZ
zH@_2rd0_{hi0{{6S3%W%d%8LT>S;hseoZe<V^Gp?;y~Uu8PbMpz1z=fbkMHX`9+T}
z*{5mgVdqx<r0S*>Dpztfk)$%$a|DF-w~XN<Vpf=)A=wagE=dsK|5s>lALS<}1e!K&
z;zOHbyf;|&dw51Zqv-od!epihO1NdKkmu#xQfjDQOlUSYeHA^mDg6gi`WsUe^P1be
z<mgnIu}}J4cZCA8Hb~yM@zJ@6&9x4~`PY{|VmHCEMWVF>KEMJ$!1NZmi0^R6KH^g9
zH0R26`+~wnUtNIa{gp!tzm+MJbrWVnsS+Uk#FYM(3F2m4AsAZ6O-v?2^ZCi(x_jQI
zsCP{Ch~-9>vE`gOW1G`%z8!gAdI>W`>F^1xiU#465PG&{n(6N)Gs|+f7n9ONx|hfP
zVCQ3Q`{*|lyIyIde(5`hT(n9%-is^V#PqWDtn@j)0ToSKA_qOv^wt{;8%$Ksnkf!_
zYQ<}sR`6}RrEm31zv`3b@0OGLh)dkS<!+1g$|;8(<S@9$cQwJ_ib;4o)Y$Gs_B;?=
zE(rNkICeVNRk|dz$B?@(s0t>dy`bINtyJd^TDag-Wq32QfNRec8=*7Fasb%Q4KpCn
zNbqf#tO=1$3Q0(+mlqh+I-DX&3sLk*L-y<?P7PIJ-y8(X`6kH;Qe!*Em4j&5szJDi
zrgjrFIQ3L~<bd;iEN8ybf`GkFz{4y?FdtW~M9LryiRUivn(-$@5tFS%itFQjv8|?p
z$&vdruY08x+Da)=Tz3=tac9^d$ZAtVr}grW-14FtGq3OGjm>emXAVef4O%3G(M`Zn
z_Jf-z4h$;xMO_wzdFJB`vjVeJqilrm8Boql0_UM`sQE0v>S_Vd0C#Ma)<G`(l|H<Y
z*5<mxur2#iO)sA^xfgp$rnmy9WsaO9*EVjJ(|vwG`cA(I5_wYdq{w+a|4C#1)UYO^
z`E%%tT^lCvI|zw7nCSY%uqoy>@cHf5)z;Eb#%`fVD`R;dsEbZGdIV&9bK#AT^o>)Z
z18o*i&*4kKRkFW#w50grF}KpPyO^-<o^$q4pAfylsRs|JA;E#`RZDF9(#ZF35j};t
znWXs)`;vs$g)Q;m%5EMP^A3&yyA~*(+|l&>(=Hm_Z^*Q3MzKewCQOXuBzsHfE2X;$
zJ%-urb<cj{v~_7$$j|s5RavN{)2UqN;X=Oi{bcrQDYiS#bsF~Xz7jBwh!q6r)z95U
z(}<J(k7w-`qbRhF9c-Bq@<^j}jG=Z;nFucW^hzP8(eQdN(-`_NMjZDp1g?3JLf`s>
zL7~QT+j`)Zc9jHK-a?$jXL|8^_$gJIMKEP7IwOZ1J|a8r>5nMj{`|ZYFy77RYA?yZ
zVT|JcV?f<^ALJ9Eh1(NIa?zKpYg31Op;hn4uve2ooNc=AOJkb&q4hAXB&k58iRDa)
zV70f4bOtJ(IabLEF2X6!@XDv6!Hx$@Es_z&W(jJcvy3q>c$xWCn5;%eQn?#q7J^3(
zHy38ok+2ho(#qf)f72EVb2l2rr)Tmd+Uj<FY`%VbCIS1~s?`r)DTUix?*b;kWJ@6f
zaSUcRJgJK=`c;y~NF)b=#qA8Z^9aN=Bkw{r!ZSwj4(CHI=rP96aOe9o?Ks#;R;lir
z_Vm_=o>%WOqjKE36riK0T^BfpirOWfmr@j%p=1t)^%k|o0fAAR585o=d@Cr4!QBn2
z)U|o8A~ODV=qKtdR(B0^A>4~}1Nw)E@!e_`o8LBG3u!|Ny3MV9&HU3HkC`$B3f{@u
zrYAJ&#S=Z)OH#$_$3uOk(|nm%>k>oqDL{98ssPv2ELLDBvLjAnpfR;!<iczr(7he|
zgY5GtDmvSd{G-Nh)}N>fGaFiJ`xsqo^IGkE8~3#^Di(v^K_ej%{}w-W-gu-d%XE4j
z>&8JumWu-$zi43;yQ}s$hxEna_(e!>;QP8m4Zt3u)?^Zxc$N>9FcaGX+=BLhMEutL
z0-+qJh<HBlyk{Zi$QrQ8dSM<wPclUK9z<5y60E8C|60|9h|uu<#Awm#N(}Cj{Fcnz
zmy;EKkIiK>wYQ!?%;!i-`(Z`BbUQLyqMc&ttdkf6M=scHzL$_b(aW5^x_^yaTD%@-
zm91FYV3dL|5JZP&Rb5wl5AHqZ9F>rhnKc_``bp>Gbhk&FT4qzv-=7>)p?>GlP|$3C
zYpBP<Vv%6A|5-2}Pkoz&%kF(<MLDjxH?eYPJ+~e5E3?5JW}M6HwVYQ3*M;`V5m4!W
ziR9azj5c~n=e^<$Av7e~GSl2PFP&v4{tZf|Rs~b%!WLHLuy^}M30hlOs$wBtF@hJM
zs=BBtR=!%MJ<Qwtj&fwOYk~U6Q+>}O%&pLb-rKh@ny9j3-zLYV|5`<RyOvZ$SYx?N
zX^vTlTx2Oen};u>WVpbK=3E8T4VCT6pDyYi9lx4OHm`7BkfcX{JLkkC=FEWa5+3ZS
zK3iHge%*lJ?!L}Hms=7q{vEH$n3;P?<g*j~VzOg6Q0Ku~;f3b~a%XWXL*&D@Y9V$-
zS>gKAJq+Hj2lt3e^M5fNI726bAJI6^mF6H+)#Fzi{zeS<(4Cze!Wg3bo28%A?siGD
z&2($ha*&~_&v+$nf=9Qz)JfL^Xm}%N(SsNxq3E-Nldh!?2W;^i=biT>uKSDd9B&wG
z;Dr7H-Q^KW#Q(LMiyM5Qr)^CN4KzoEqx-;4tLsQNn%?6X(KDvKCqR53rd>4)j^%8D
zn<GJcs!w^Hd{}yx%ZB^slHw37MQ}p;CAX4vuA<MlMpGr&Z<0d-iuRmDq@g^-XiC~`
zmBo&kMHWi!+#qoBF+=!_xv>8UFkh$Zt!T)%OkTJuBHp4mIdhdpauAQn%pYvMmlLc6
z@%J2v!xxSYT|wjyrO)~kN;`GvuC$)q?G61H`bmT-+AhZ?Q3+G1i{vNnGCh5mrJ^`Q
zwP*c2V~vkltu$_{F={AoAyL3Fr{}i;NEl;PgBH+tt~&V|4QCri*CmRMzf~r2#Txv?
z7CZCrX*1sQnCW?1_j47&NT1#lpfST>>{1kZp$Zc&FR)40bt&jI;T-hw0l+CSUUw#L
zcTA+mo0CBxFaBl9Vx&D|QwIuzI?&orR#p6;^JzNT@n6bLY1@iO+5Q5<sm0-8_8Y!^
z#)+2NWN<!CNDZfN-{)^xEUjf9Ly^~IqHizV5{kT%)7ey%^xi3%h-Xp8=n^-}GXB8z
zexBXVE}7?znz^$r1kDLD^Ij0$dJr+!4NmelodEl6Pen?5NJaKt`uL>E84maP-@M<S
zEbPqabv?G3e!Y96cTT?UGvSqOORKrf#bzXzFdIcboP?8}Q}k9T_0!z1XOA>(Vl}QL
zmXzAP>Oiy(yDJ7;G-=SI4M(wl=nAw(bY~MR6qY6?u1$0cz9=#3!)3+iKotN>()cw(
zwF=+rF%5=4DIrMEnggYmQ401$15w9_x(!1CU(0p@?}u#D2+r(~O>nF~_bKXh+joS(
zl>#0e;^lW2S%tFe_y+KFOV#(>t?OZ^82J3GuY#yudXq19xdF6%|G@bJWxsqE9&>78
zJYSeX1Pgb!Mu~N_WvfnWn1%*&I)Yq4v&bAMaTy|XZ0Y?HN%E=(B*L-lX{v#j)HNc^
z71f1z@W*l_t!f5Li`h`yJmZad=kL>VzLjQ+cww_Hh3QLkYSog%sh4w|4I@G2&LalF
zw`AEUIozvG%y-yh-dzwPS$Jnm;86$0Y1QEV_AX>RIFM=lAwFEzyqTp-sBpO^9nx^S
zPGu3Ku|AhIc=m)zk8Hn`l=nV<mzAL}MSFmBAFI=0ND)pT)WUf?8(-ucdLv|GrIxXC
zf#x%<5QE<v;KSryg_1q2yVvHz#s^3@aLvwl#H>}2tezA5&^_k!;mmt*bCU$N-9{3b
z3x3>&AE3Xz_9v}z#NJUxl+MQ5O1(1Gb6)bax1{&0O8nUz!i!s}QFXv5InTpOun(qA
z^~k>YoUbF5^KEGRku`jK7&OLAJY9-N9!X1S1yE+2DlnF~=hC<@`R64exvX)O475}+
zN~1X_1<5O@89vq^%m*OF@dTuteHB?sU5q@c)arY(@Yj<)vTDa!?<i#GCVFJMB!Afl
z@KV%#)f<i)!$!?pD=bMap-QJ;PpcEF3{jFqy`gQ2j5_ydcbW^?21$r_mfq5~a}%z2
z7hnElbX=pvoRLlRUM1;bS9}$CsysX+MX;zkgzuUvc}5Kqm01Ti@OukUb=3;5hcsWt
zXJdQIq{WH4a3VhrNtE?>EoC?McFhgB!<~7=s#+;ouAl4=uAFh{%Pu*AA{U=c(>$l8
z^M(1jT=H<z`Pr(M5vmj)`ldLx;GZ>0`9N`c9mD3~{8WF(D7pu096_2k07HBY3hPrn
zf$pRFaA|S8k7Uyxw?nw`UeuDCgk;?b%{<!&n;aP$=O7B0@kD0q)V5ceni_vi2u=pu
z7M~oS^zB?GX5eZM8lguw!Z9AjWRT(-bX5=z&>yXU%Ozn`+KKOR;w{E7k&8>c*d|%e
z8(n`s7!->$Eq%?rNj1~q_P1=}X7$Q<3%h_!yHAqu9?j6yqW~UNvEES;IjZN>8~DTc
zyr%XWC5FR6!jTys;5f@_;cb3Y4$eQrpO7+0Ap|!X`TG*^XYDW^uhYr-!=B!kf?S00
zClR&z1&ojw2|3V5SEgb?=}_g8i!|ZD&l-Lgn0lOSJFfKTy#+eS^VydHlm!)r>bOoy
zh5~XowIbTjXjlg8iA8k#XgFcuw+JXmW-B|Gjz`wWIAhf!g15O4w`wRQWn)c-0~;~E
z*Vd--?-HT?YjZ<OMQuX{4h+rNjEoWYX^SP^n%iqs-O3<;e_Pre$+f1@nbgrtMkFM~
zH~Qu<MZ$*$RcXL8HXL9QJC_$TKg3@b!LdUANV1+Fzz`)@)USCm43M`U=H<kQ5x6Iq
zi{?)T=m|-V%EXJ*Vz$inX)#-ZpL?`RHKieAf<WnGcnuQ{j-Py_c_M;QN25ZxejPqf
zvKboIBasp@euYr>Au9Pu;l495rKi9cjG)HGq(H>Wc|D1uD=u$7(@wQokhv8L=ciTW
zq|U8x`B0$?Xo99Cm#-<c3Gb%UT5`4wxXNRk8=z6ZJL7cUq^qEh&~V{jyIr)zVI25?
zn$GV~uL~qnJ#idWqiC?jA0aB-u&;Yga9ivrG2+hR2@6gvvTcV{g8F&t4=IT`Zf75!
zH)4sugK#7lORLy`8nu&N%ALYV68hR0du7q$L&aI7D7(Ld1eBTdB6>km%{k7NBWg5?
zMCRIH^ovK(7g~xZ)*tNpOnAeSMUGE1|HT3bcm5Hj^Y!YPs_aNncJxI=3Z4+v-8i~!
zgM<L2;`L9BROdZ+$hlnNxWGqS>#mqwI;N4P=jH}OS>6)nFBhjV`bgSUOW3QI(N)lG
zNW;;7I7aTh<$w|~)!x<a2WNmg+0IQP%VH#4Kufoyyq!@bu@p<~MBRjvdDKKUK!{88
z5Z~0qJ~x+#dVyPdx6)8)JuBaBq<D0Yu4H;|<oouIa&i4gQJq{re2f(QLuv<+@Qpme
zZI$lvij9in1xkr@ZP)x6zTLP+pOUEw*Bv3}U2OYTmf=Z5EuwPCJ?M8xk}1L2;G(Sm
z_H)mu`V7`#Po#P2K_A3K!DY^#tT*7p?`}eWpGVikvuj4%&@}Z@u#^OXrG&2=wUi7L
z4gZn0Ro%+V8doRb9!Yy+`Snx%fz%urzCjMX^m2AK3uB6c-$G+4#QFDO7e^HF@Y^bw
zZahpT7Z~T!o<dp*o%)u`aZGhi_kE5|r=i4Cs~xm!-)-@en~GZsa;v~F!7iiuL+cHF
za{FR=yQL@%nPlSAFw%VMg4YqoF*dTM!h<FT>@u9gw&cU!R(h4J<IP-f&d0{dFDrA`
znT;KwZT_~n3v8?TzXqSt9&#B8Nj5GGrQucE$dB9_SsT7pyYDd*4RlRyH-%Wkrd8Tg
zJpL<lD!Imcoy@7=x`*{ZEZK~$wq~d<?DdxpCYupB)12oEs}`9)ao%nqhAwt(V1=(V
zvezh=vj^zSnJ4FGv5fvRigB5}3EY44qkOjNvy6*L;ltQ*DDpw<+YOhTQ2C|LsVAW>
z>OtkO)IVOVmdIi#vsa%Zxr=NzgRjK8h;JHl3jZHrv-451Z_``Rmy{{%j~8T{)CLha
zi6B{3>QQHH#ZtDo+{PYeB6>67oQQ)<?i|pl_C^%6h%o9xPNLM+7gQ*nD`zO;y~b_Y
zKKEj~{jFH_MEXHfyz1O4Beqo9!(rj&9`EU2MvAec%NF3n?AN2|apO<wJbcoJu(vfW
zUk;AZmZ;2t6Nq?N>NZ#GJa=bc#A!6Jo&UVe?j}(Q&X^NA74#~y#G$KtP((;o_;M@F
z>3Yp0ZTF2&+_6zyY)@Gtr2@H@s3Xf-G#e40BD}^r!|V<$)7if*5Y<OB8_8Kx(mR^c
z$5cnIgB2{aNtRl=vUmNwzwfz86)gKYD)C-$h)q%m<|63?W6@SXSqO=|8RBN`wf~zp
zMDFa6b9M+hOO~~os2s+I!ji_iKTNu>eCqn)HQvR&*;ragQm#5l_H)b*Sh%6iAWzGi
zhr!(vi<3tp#MZ(M(^hiuZq=mv5tGlf$K@Hmtn@lLxF?fUJ*+zriGibti2E3bj1N*t
z-Or;~o09RACJ1BOxMHcfi6yw)C7ouZi{;qQLQE3qPJx?TdD2QpmC1K{V%zs?En9h1
zf}+@xb!ODZ6I<WQIX&zjiqb(M=<b<5{@4yVRHF4a!%U<zM$#^O-Rfb5p@R7z{K<`l
zJ8`;kJ)j%@PFWc1Xw2T_$k$b=e6i7^f%taBbs-`2q9<p|*kap)1l}pgrzc|W_5ISD
z$M#-t#<#l|;q<S~>7STt+x+;nN6OaT_f4!FcXB(0YtH@+Q1VD&hi<3#Xy^L|AE>5o
zGZRO1Rl0oj$&8lvcoJC~iU3uIzbnmn5sgv>YNe@2zpR$K7%}gDX4BB!%0{kIo*ISt
zHu+4(nv8zYGdb$ZXliaF{xuqL6qh@_vxpXnDu`Oj-LM;E+{bsOem8&?!nfK_%(3-S
z={?+&Syt2LVm3lDCsMQ6nJL(KdrG(7wxkbD-M4=jkDzUk$NDimb6L6LdNVCAT40pa
zm4)m5DcUyX2)bwSS)2Hf9o}^STfEgUkZCLBPUEqiA?kCEQa5pig1AL`HgND~Vs$m=
z3Xxw8<v*H7{su0coakhS6{e;+Ma+-4&@lvM9H1R_xAEq*4wF(mcJpkapUGpkf14}%
zju-oaRneOxsH|G^n`9<)rvOe5PS(?n7Qa#jgOB>uIP=r(%iq-uGO@GXRK55l?N8bL
zp@w3ixlY00H%^{RfoEMKfhv|1R**i{hZJ-;CXr&2-wNTH5jo4Zu}kOTsUn&V+vo>7
ztoRy9CVTf2D>9;7?>P6+I`7HmI)S8UES-`;>(j6amC}-}0+ge_?^0faaWm%tZ?BBP
zFq~I;E=<6X%E5_~C!-_;vL@yc5^Cb!kL(dDYCj2;D<1o0xIV&5*ue<q&-6n5Km>vw
z16IdM!yU(k^)T~pc2jMcbPAFuw0}^pw1aB#_wt?*wJ_d3N0nZWz1NycUqg0xeOB|N
zFu~~Y-$g?9aRUF}j+Gc9vTXK#A&_8Sm@OE;)h1gT@%^gQhkY?0^ZRwkYm$l|`S84T
z7BS#qlhF17hgkx|N{d<wxSB*-iFOHFmT9k8eS&!;blx++mca~U?oRw7de?Yn^(?N8
zXU(Yd^^?Zzv^ONQE#l*>6og|<!Sh|0Fy8pvIKfEEZSHR4nGo4Tj=0=s`obKmX1~yf
zM*gs2A!147yA?GKPePXR*!vk-yv-K}*iY7lmbG-OJ61KUabSh6cb?D(p55uOuaHc+
z@b9-M5NB%Qv`-y!RVWdDWJ-ibTT4&zofYpbMS-?#3o&_&IV+O{$sYE(49WW8z(!RB
zPGVP1n9ehGSvp3eG}hiHJ6REVZ83Uv9|Vu<^SqoC^~Fvrj;q<V;+dxK=$%LOLT*}=
zS#9Nt>pUmae$vJ;p9fq;62Qn9*oo4=u`6N8Syd!$OLwa7c<xYai+2yj6qTwZ5ThkC
zN9XLxK6sMcja@xa^pB!OtV#M{87>9ZHQFo=(pdBblLb?(qTHovMrfldc!Ko`vJTEP
z7A{7lwedQUMYKHSwB^=OK{a<STq?GnY%&KKvP65;CaU<GEZAD0_pMgtZx<L*s{2kA
zWm&%a(81bmS%%*dgxy(C(&(`yPg&usp=o}SQtD<WiR$FSRgH)Z{b#t}6LBkuG{ns<
zpao}<%|d9&8q`dW(O@?MI0SfFwZy+U+P}Y~$zo;IL1b8h-ppp4S3Ht%EMYE*mvf*!
zUIs?8J{C8Ky${PhMRZ9(H8>oMvkxj))2hjb`%?bk7z+onV|{jDO;eGa6WHiCk!Gi-
zWpyIgc@XJn6YG3Nm$0uaXVRfv)$C`!*`LG!)vs<u<@OSPM}f$0!*_{>x*ccQWj$g)
z{Sk)WR9wIO>h<*uO8^YFbl9pwjaP^{$Pq$Hi_66L(d1e3n1!I8`Z@<xbyacn$(-Mm
z%~<|04kK-Iv~Mly5&S=ly#-X1YZvz|B_K$MbPqUmBOo0^2m=TRinM@$G9W13AuU})
zD=E_AAPv%>v~)>GiwvE9*La?Dp7Xx%_r_YRrHh!k=f1Dnd;kBxoz^rG3^(^57K6xQ
zVX=mV$M4?oBuAxMnPXEohE_JwY;5o>wiTT|^+=oqQf`w_j&ezwkI0-5hHhLuijqdf
zv^?(g#GeNayS_<c_Pj2BLKiN+;~??5ee1z&n1923uraxIrvHQMr*72&nb=j4U9{c)
z9rwVbJJvnZ>0)C!O%!X0co1XJqY4u6@4WMz!!7@aUI+sePCqxawNpxZ06{ue#H24A
z(Fas1V@}56h$&Gss|RofAKr#7yG=0;mYX|;^VM<i0+A*zt+WxhC(@;S9$x%m?H{hY
zhCMz8Wv?An_*<X6V}^u4Uf<2-I8yC%_|~&clcNb;hGt9Z)?}smd|q>oR?F)?hKO22
zb=jbdk)Cuj!;A%`ribRZ(dKiGQdy^;$q=!(=XN+Mt`J16?FAVMRZPTRN?5Khwm;jQ
zlx*G{B|W_sD*-BY%Vj4a_xusj&OFI(nbmGyZi7zBnUd;~6>b4|sV|A#hg!!7e$FaV
z)`Cdv#|R!5_D*z^MuMBN)eQk~I?wc!r8`+m9@9Evi75dv51DPw30~HNr_V;TbWU8h
zSfX5h(L^w%GX`|@lC{L`XklFhraXLUN=sf5j<&7yz26Fs%QuBebTJ3!+_;#>O{W(o
zDOSPPaRged^F)UIuT#|W8nI8q=8Hg;+_NmYiR~xYs=`qc<XBD(IvLr$tIRW8HE(5}
zr1r99&nbWv)g1X-c2$30Unsh`d2?(ChMU(s`|h`jRFKV#Z(GjAhdRj*^DjGIP~ZRw
zo>r7|l<|ME$8cxd=L!P`g7&qa$LKt8gsu`tHO10lv?x1ai<1J(EcunQv_@<dnaaN3
zpqY<Q9@a*<QQNVv&>_ae<}~iRX(3qPm$WNRxpp!mj%lhDm$}OvELZlS>Gq2-q9|KO
zuj%P3(BM%pS1HQkb?dc|c}&mh2iZTGJMO&KMp}LS{oUu)4j^*)AjbPpX{;ounZD+M
zJMHa26k*+dAl=`f;V=3*7|%)`g*3lqCo7oR4D@Q=Jl2cWSK>)&G&xl(FpZCAAX0&M
zvTn5Cv`Hk;CQ`4rshLr+>{__Xod0T*F*~sleCxNaVb%-qP%`snP`{WLy<4S@KJQ8U
ze|2`?`pU+-UG1PLze_spMuihHgE{{--AE-&ZE(Z>W#eTDEq%zUMa!D9U`9m)+a>n%
zE~mGv*t(D$-x!VAH~FRE+!0KXAieG;=^Ed2Q+1Y&=bKQKqQ8a}_XgYa6_CCogXYBY
z4c})NjF%;@OT?;5Dj(-UNVjpTEtl~|*Bu1yjF#rE!5XeF)F$wK6#$_jImO*IxSG8C
zrM$q3lAcSo9nzn(U0uI7)EO4D?BXIz+UDGk%j{fLhPtZ6e`lw<gW5Bp2e;dz164ri
zg9nYB8k#jv&8`S!k7@Adnf6M;i}KFPZr0i+IjW;2OSf%Pn|`=?DePTr{63BJCF4W3
z2XJm@Q~UBb+Lz4Mn(42#csCjfmW$QrMx=U<i9P%g#Sv&FBdyGwpC4Yk2#tPoYab8(
z$pHk_?DkMnyPdt$aCCsxCqVG?$o5~|1I;-w*+mGmhw<W{dPqQUi#rtCfE8&5D#mjH
zmW!0a<~4Gt%)CaqhtqF=;6(bYO=FrZcboCc6b?KBG`FU!kvHvv>hlA<qAcvnBN2C)
z*?nDQ%)MhC2vM~N_CIML4qzH!uFlQG_DKaTh-bq6qt<EK%$qGNm`<Kl#W!(p->Z3k
zaWGylLm&XPM@U%r-ZZn9_bkxL+>l82(B<z$5*SUqi@12d_@QIIhld3f@fefwnqdkS
zZyh3b3G>I~%YAr_p#mMEK@>~$omn5)h$V+I!u<6&I8Qvr>*rS5!yKymUUc8Gp4zPC
zcK&vs(b`zv#A=8JB}u@1KBT%z=aN6JIQwTi_ILXxe_~$9NB`pHQr-Y@59P%)pZwJ~
zzq1LFX0s3*Os@E01}BwjhBIVu?4)|b{52C*M-FnYC2KC@Y=L6QpV|^<LXblFw_CK|
z48Kgk=4}VhT|Jw?s?h(}sxbN3)8qZE#xM2e8aCuFuC?7~;?iWmzlALpV10R(@nGmd
zheYvZnUN|8*V9Lc8h#vZP%Mwg+5Rpm>;K{>)z2Xde2U60w!qHucM#knp(4*HuZKUn
zZ1NRGC4y+Sm{ROAJ^oCc1`XHbQHF=8>DkP~hK1tH;CUe^hG%bQs_`Q2ZWY);TCIx2
zN=<i<ZWG!I$ETwBgyMclo0^Wn$`F8Q2h~lUPkGZ-&HdGiVDFRJ@6+*gwO3H=x1d+O
zGBoBkuc9EWl;&0TgzsNqyuyfAn*)*XwSn#y8%VN8R(FgVa{*;KQ%uw(3Zzie2%cPO
zU1m%67wD&-Xp}be<v0j~ymZ`Ea({T%hezWuCc(I?a252}Z7r)S!nDVXDd(?2J6z!6
zYwoO#Vu9k+#*xNxXm+|&aCM2Yh{hQ|+I@zegEj1*ok+B(Q(!@dbM2llTO)hpoMW+O
zs?+8%X0w=-)#cmWqTSu9mQ#vtFcneS=ZucDyqdUt+Bj%`<9krZIuiBM)+>eCTQ7@N
zGsrZC9vRyuA>bGXh2!24aP!a+j4L>&KT(VA10v-fnBu`FPLU@3rVmfPI%kkbZ3!lP
zABXMlHG~CQLei%|x-T`RcRoR34&;RgR{K6|f7dzI$@k5(vfU&fksq<ne%4Lltiaff
zZ>0+(&CO8Z=^n}3B-GqGcC&gePQD4p)gR$jrJ!xIPC0!W&M}gg#PLe92%+;@QZZAd
zE7o^d19>A>+(Yxl$%1j`FO<YN>&1JJ%PG!n&+QQ&X-_RoKWAqQu#Yg_Tu-!^hhd9o
zw&yRORXQ*-UPlh%z8PN<ZOCoXZTj#bt2$t`i*)MRFO?>M3?TIIw0j1sC-o>AQM--t
zL6=18y1SjNV<IjGl;lIc-Ama+uCqW;SJ3k)hS`S)9bZTew;PEF*w-x!9{mAD_R48w
zDL?Nsdzv|pXw!vRmY(TJZZN(7p_HIkK&X+zvw^dnjG_|t$os|4`V3MB6pfk<b^H`U
z^BDabrou55KO1cPQCVVv^wf3DbXbF4O)A16UmG^BL9sR0tfXqk8g?`tkzJI#lQw#h
z7um$<!+eU|7qXKBmc}U8Mh)grR9KUO5TLegR9h^+(~^xyZQuD%Auo!SR!%r;hhwlw
zh$=<3g&bifw<Es$1t=GneE6rfzKTtkh^lLrc>0;(CZf)OP(6B@m1g8-R||r41K$U$
zQxcMQ{7H0n96~V?sFF}9QxlW78N@)#BZwOL{3fgX&2QSHVKew)%(=$IwQ9B2YMMbB
zui>kfQY^k(l2cAC%NAV-u9KtT+AwihEjHhZ{Op!bua9W`M8tB+xo*Hc^S9OJ)0zZ#
z!(eL&(d~~^#mDa9?y%v#%D1Td4qi3KIO(yj&wuSCuY5fAz7&@jI%2(;nwmsFX7CgA
z4GcS6CoJ%p@daI(O79i=hK*WvQ~9KuE(ylr!XwRQla<U|s&Njvwm2as9%fbJM;oI$
ziz;I<hlBHjQv<o9_ucvwE{UGOUx}vKLFdOiKF;<J)~I^v>I=dI9F*3QQ;H87!a;EI
zK<=mos~+mi?rc2p0M_rE85+HPlZJE)lrKXvpSt3K26cU}UGt3O%`9u!d^&c$C{~7E
zUK9mdQ=Hfq`;OYRrFE-{8`3N@y?{CsENqXn?Bv^g><b-QRVYlv+j_iDSP@A*y$C&3
zv7RsSEG=$Y@&Z9D1)`G5C_Uqed%9^gA+^xmZ5|)$t)yTJo|wcZ`^<aysZHH2>P8PH
zaFaVo@Bg!4t``Qeq9lLF_XakEO>-4y41ds$ozp~*IR1`0+F1LtKP#a{@<##Rd&imf
zFX@LMA(?qjSmkk5AO9W}$MUtPBC-j=RK?4nEoB>{=_@hvxHxx~1-pG75zeI<Ke>V<
z6yD?1{i;0$dspd(M*E$pN{|}Se>RBoCQ^aS>KYV!<AjXu+ImL?aisvQ69sv!giz5g
zzEF6ypo-XKYDpK39+hB=P^?9K^8*%5_*F8SYYkm6T$P``*f@HUzpb&@SlTl%tupRD
z&(nP0MqHI=kX`4*I7}2W^q6IUh9p(cW-)u7<~=9=Rayqt0`_4-n7kTiyBRUf7gjzC
zZ_BDoG~0DD!&o8$0pcqZ-*5f)jVJWztfF^p7#fMc`mr0ELI6#*u%nl39dOr2Eoi|L
zNYo|Qv5}9fa36EqwMex1V0+{kv)sw^Z>79J;-o+ZlkM5tA|{Mf*icnx&J!lS(m`1A
zQjU=e2P%g4&^}6&RJ>QWKdh07OGJ3p=L(9<d}Y1WT8{SV%DbVJAV(3bJib>Zg?ZIC
zH_1YD!vYfBuAN8>-v-7^)dZKDICLr+1OcNQ1PW^q#Jd$3{qq3Ms5rUxvc%9jmzSbe
zfx>~KM6F0d@ouq9sJ@<T=c&YtA76BFxH`pNnu1otWLU`iL1ZMYJg@GnZm#iIT`>uE
zXuFb*U8g$txPhrVy^x}Kr153qs`2>Ax?3SzT})4I)xDbQeO-6P>g2MMC0F&)Zz${L
zQhNMSP}gKqK4a8w2_Ea>!nd@=*}`6GcI%(2B(c?n_~JS%P2<-x-BActJe8MIc%0+c
z<xzM~fBtsq*}8Zy4$%Cp7sm4sf6e1=H|z~-(uw2DM&IT=DVhrT_EptWF8a?deUm)Z
z9wtv^Z!YCeQ&`F!`8@IR$$2Wrb&RBau1V2hg#T*(&VHR*c9PIfz=E6J92)cKr0cmH
zsyp`M*uP`PW9N5b86Q(g$xeIv)eLOyD)4jar3IfUjn1t0R!0IGUb>a(D4*CAZ*u@{
zSG328#Tq#dt-RyII@|$~pV|;cE6DQqzb8YXow|eG9038-Aocq57`#&6jTWsyzQ>|U
zj6mvCw2yb4mtoDSM%~WF`+859P`B-BxFd<=kZ}qDfd?gTW#n!0;y(+A6NlwJs_Q~3
zdB*SPM=iG~Mn!rVrJyTwJon$Nf>hswI?q|va`PDLp(K>X&cUZmqOBB6haf3gDJrX2
zUxEmBnrV)u8(ZAU5c7X67$DgCvtX#QGX7FAiPvESKlRDQ0=$`B?$c<s&9NBh)FOzM
zcJfV1>wZwmV0k2*M2d=wM6D=5ul_kfkvDLk7SpJHn|Ik<t02SDyL=BOz5@dXAvWC?
z<$uY&2!s8*U;K<P#-*M&4MI?@=y%nWY;&Xwi`5(X(PmuuXiA(J+!lVG)Wm-y$8ops
z@ZEVh@u+9?HsdFiG<3kc@v5(^?c3>^j2^S4hLM`SERFRw7v2J{j$2CxM!t0&Y(Zw8
zrTd&q^_HX|5bD#zU}5MTrbcvLYn5Yw`bnkcwVg^l!EwSwJ#~8PLB3xMb}UTGfrv}d
zaG6**ixL+}wD9xOL&ZjP+^{m;F~8k{5BcyTqh^1-UJb+AZu4y;=OV8;C!_xT%pqBl
zM{;VUG`Eo2C%jV0D?Xy=T!qI*h7h8=$$QSqwaZGCWhwa&@{$Y;<GUU+oE(GVFQ{`{
zD@iJ=9Gr3u$7Qo+n-B4gzRS(MMX~%!GWxQBH$-N6h0ly5K8vRydvm?k+bF>=TH^i2
z`<s5WuHAmfFNKarJSnrw+&z}PmNk|&)p?@3zQfa&`Dc2MHW=`~sv?akodmqDo=f3n
zm2*GHT--X6tWuVD%YDB_Jy@8N^+DfL$T4$K6Meg=iT?VmHl3~<C9U<Dv{juEZJzga
zJ=u6T@l+R^fc{?Gb6_{ch_pO#t2ucRs1MKc%T`@bvfFgKSHw5%rPR$diDrc3uADuU
z>WzQJrx@uY?D3Ybes=sLuy2WuK>D^o2)Vs^QWZ)Q*~9j+X3zSFR%GbMhwJaCxRPYz
zG3*Q^v_xaaDNZaIp?UjGrX7A5`9L`P+BDTXCdNI?I1gIra5sjvoUaV;iTdtX7jpU-
zQlE?gi$wvPD5zma`!deSgLW&mNy|yU*0j*~9?8P}kQcIi`#7+seQ8&Ek=6NkUb1Ur
zFI+Y_?Uc)kEFm~!nsv=)%4Hj4I$PL}tX8MnT<QJZ_|+qDKwf_Qt125_jvB0`-@w#k
zB@1<xFUQ5_1#mgKUR0~~Oakfo`~kASv8v-!cgAFi;fFdE&8c^N)<!tuI!v15G29Qw
zzW${WFc4=p!Lw*dcd214{c!vFpGjnN>Tcqrn)hdWb(*4k7=3fb)Hu!1Wc2YeoP8bM
zeu+{I$(NqGk<T3DaD4`Bt)l<kT0<?KEr6s1x9jfQG0b1z<X@^LVD-F)D@l$O3UP`H
z6H=0&_N&8kXgnCBS72iN0P6_-kB{e{U<Dk_Xs~g?Y*qXmvT)^&eTkw;{E&_89xcaO
zx1QcOZ-qN-Zr_{S=u#o<AuzX`eE<>?xdiGIjDQK9nqp`437(ZZ{j)MmjVV`n`{qL4
zAu9b9vpIkHr&sR{q1u*NAr3HEs`tE&`@>pS8OQ_pZM`PpOojFFkd>sqzKou>$HzcC
zkkMoL_}P%ft}Jd5NN3E>XEVa&n*_99Bc;o%(f8JZuY_fj${w2w+2sOA&V@e>+du@e
zf#Vd?twoq8&628sooQm>N`b|yYKtEY&&rhSrSh;D8ek1&i0^s--~lI4cW#d9P%XdZ
zSxNd}IW@?Pc9=~f`0ELGwUfHzI5RA-0uR{-#`!#w?!Fod%ZI$cDuPFgTRS>djFfJ5
zP1FEl-eS{3fKGt+=B}UW;Puu0IZ+|nHh$!$N0E4HXMGp$4U|<$x)(~&eYA1-x;U@)
zSPML2@4p@~1pDpY+~r}74O-3k%Ts8~AksCvyLv<yO6t${lT^v$CdF+|@(M!UNV}r9
z<q=NzFHCBpNpQ$3%KA$pOd7;x$q66}$GkOaE`NLG1X(fP3g#V~LVC__ny$VN|Cq0{
z{RtBIpCs?U|LNqxb`P>1miisa5f73>`P+0yeGLo=j?%FT=0+B;D21iPR~m)(NP>Y|
zXqs=@Jm|U#9#pzRDJ#GnCyvx{ucJj|ibTw7ZZB)~gj=h4k7QXKN`ZmG9R+5@1bGXF
z69tgQ_XDeBX7lzcSzl^H$KDJu8BK>!<0>L1H?pX9ujV4YFueC>m(JampWU2%^sF?=
z1ju68q{qndDO!Q`(mtdymu46yVmXI#jToYFGqT#!=@ODLHs=DqJ0mM7wq0lD7!Pj&
z<}gPL!vl`-H>)?AEBwT=Y2a4aQo|~jCA(=DTeuzA*ifx(Z0E6XA<)CVdk^JK10Nr*
zOl{sTT@>=bTdo)IS=aagg!{RgT9E_dCtyU$Yq2WGi$1bW(icX&m`QeMAk<D5<1+%7
ziw~H%*G<z_c$<q~`jX?zNTpuwz43@Q;gS*KPtn(C$#z1OZ{XpMXz<sv=l;xT`!%db
z?6Okqz^F_$NDA^R-UZuNjHH-yL_XqKxh@3Y*ZH<iLl1~bCVO&eHpceqPB;q+HyqW{
z$;5himd&NT3;nmud=9=%N^X`cF&*{G{XQ3))E-c%))Wq>R^D(u&5X7dM$VOc<vFmV
zj1KXkCO#kBkD$mH@&Ll`r>IW4XA-p(pC#7S>urD8gRLBPQ50!C>n*B}$(Fk8#q~Ca
zv-!|XjR0=_Oykk&Ql89s^3hyDmfM`iiP29=nma~7p`NMM=TS+j+PoTo&-Ck%lD5<6
zr|tghsT&60RZV=lIR4uA6G>0iFr6fia@XjA)k<c>V0RMO+*VJsFIl&JO;y<^($?M=
zeA-|Dwb6OhKJV#uh~aR}dUx4%>o<hF;WcGV+!{tn&j1!=9$WQvW8cZJB7R5YMRpUG
z`%hqEp6h(BRgWZfIkRcke_-=tgkqI~8Av$a28)4e#(z%^mVy3Pw+bj73$Q|a$Q-96
z+pTPms%`NKGJ!{~;M=@qJnDGuIm;i$Q&c-ND7UZcc^<be`F~&FpQOzeSS#DprTLKE
zLI~!hBM7rdpNRRi3;F9UAbqv1PxcL?v4N!0?%3WKS$67&35#luBExtKdU!`rwIK8h
z&3{05=jn~Fz)$JJCIE>`CD4JsO|g6Yb)-cVmRyB7M2fclSjBh^HH#;Ke`-4Us%G&t
z2Cf8ILK579e$M8rTYGz%1?|>MaLkIqaQI-IdppZ)gUbab;Gq=ot-<q{wELc!?$J>w
zw1*j)-h)q%r#BR)s)a*84~Ok?o|rdy0@|(Dqn?<C)=7pJUo0@G5#HP6_K;I#gDo;F
z{Dl4Zs8uOXWhA*8x|kj}4?L3NdjYS5MSo>=N0=HQ$k#%lFRvTlu3ek|Cia<$)Q&=E
zu<t%hA>n}t#cl^o*ETZHJ1{Oxi=Q;@*Wo87k7e?d+X%dk6OPR8$OimqPEyUVX}+47
z#a-yk4U&@>1OUGm+?_bjk6wigA=M;InJ^7T<H{x}H%3XxX@J9%|3$50vfjn#f&jSr
z*^Qp=O(@N4mDnS>1n~}3kyKKov_P6#NkM7h#R)hiRS5{hLmhn!wjXM?8&OpD3%TzP
zsT$Q0o3KAd&JS?Z5!MzPsbbo@PJ_>7@>5w_>?LfNbnPzU_A`u4J`*(C&Jy1URbwqn
z&@~+t0pgxn@4V_!nTQE<jCtJd?AbW}nu!$y9JcTx^Q?Bjryn)%>s4NVvh#)wcHDmF
zwOPToBEKyT8Vi}HNroljS6lgF!@8Iaa@%t{;O8yUZn*<AP-2Tr4c-rp+_s}w0nixH
zSD4xNN8Cf80{(Z)xH~dz09xL~Ki(~f;6IH<G04FEHi7iM3PY?qFeJawj1)A!EMnfb
zUVhcDCA|yt3%R^@fxAJoE$O0y`;>z9>-2#EOj03wbN@BkN&k+c;`}bil?EMcqB<aW
zXVUFpuvkB8H^&ew!o$`tl;g}zem9<cbv!&&$a?eI*T9fw6D@DUrqN9na4962d?>*T
zvlaJvE#@Kd#q4me?x9$Ixe9y9=U2rpiOw?sLS#Z!H1O!&ijT|79BBnFbz5p$QkMEr
zm#KJQ(41LDHY5#?9#-?O<;`CjdKMC9A{Qf;+Esx8nW8z0E%cx1b&sv(N=oY}W_K69
zY9Y#D4(nRzzjEvIva&xG2aW;M>R$=ee|9cGT5@xEJ4|Rf+!vc70j`^j2mSg{zcgWu
zg+r)^kQ`eVpW*=-338eHhG_31-7=lkM@OGcBU!ubIZO0Q(DzW-gJe&Gmns<dvz7HV
zwAMtLt;Tr8Ey2n%zGaLJ_$7pKSz(JH)`cZn=uMB6O&mLN(Oj5%5x<=e>eY&E|1gR(
z#zE`SvWn?#Bq~&+{3By4i7s2PJb?my*s{&sf-O(h!}Mi5<#E+7A?D@XuCfNSdV(8i
z{pIQAHb7xHFJ?9D(&n+vd8|s!uLooG_XRdcdQMg3%4b;)Q!8?;;}JnQ4ezT?y3*+d
z-LASNaIPgfnA{W1Ow+9txB45<92_*+sr!_&T4oCLlAF$NqY063vgHj~BFS`FHE$9|
zU9{^Q09YDZPq6gd9oAq;4G%$9;x}QQghOnTO~nNtM!XV#zoIss;4ar1aB6(`JFO(Q
zJ%-nEN2wpF#(^LsX~9;^lkTo93y9Sx23jNTm|~avbHvx3Iju}Kw&EcoioXIXzQlQE
zhk?+IRWQEmW?Io|yFLdq7oSVa-JQ2?gNETHQDs;}-JNUmRd}<pUktZi&uvr1_At6m
zwBDtf&3gYLrrMTAJHhP{Mk=0>ShpC#L$?iGcI>5zzk}-$*kJ47vD!Qd^SBR0DUS^~
zn7dsu<!^eG+R+}>CgaVlk*i81b>BnijeU!V)}%6+1`CFvBLY1XUDdNi0ZOuies2Vl
z|D+1otM}6nEjLEr&B+!%W6)zP9T5jIwlxpE)94+8Qf98NccI1A9pMCFzBU1PcC3!P
zomiHqC1gm$u=w$#_33qqu@!}M{((V0ljLE2>EW)xB29V&stxnWJg`ko=ZqJR>YTnJ
zyVc72>hbSzq40+9lB}{aN?ykpZP>-W?7eT7h7oyOVDLvSwX0wTf4w)G+dqT<lf1ve
z{U)}|8(-y%ztE)q?D`+=_Wx{Xsk{UgY7uuEeX$b~GrONP#EA1i#YR5!WLfJHRZRh7
zz4%l1TZpqMm3xnc>(%I5Wn(!a<;^!mhpsWTBj;F|mjmRaHxHChpP{X|j-LYwb>{Fw
z&#inoW{_)|uCJ}8ulNnOR2ngOC#(+Rik!T$*Y%L^-8_^Xh*!P*#T{=IJl}xt+CIb+
z>w9@Dd%`i>KtzHoJ(YitH)j9q%jS_dZ`VGJ%SOQHZjT0da1KvJ!L->o4j1y!7GjSp
z+4q?e@g}05sG%nY#MK^DUfwoQZ56f%@K#7vCMhzTeW{t?oFw>0`f6!zY(hfWvh=rS
z&vVOSIfm6)9GBnGd@?J{Dm>m;W&>ELIBAZ|lzmE1tQ&;F`c!ICRUMT^oC!S+0}mQi
z=@XXYoo`(vG9Dy|%AYStzDfA`^sRL9XiPz6NkR-L2%ZOt`?*XB_JY*2d-FSgV;#BJ
z%DkOHkLp5egt=*Gp6E^MvMt(|WS^2iJUlJV9a)q8OmkhTNk89rc_a*;=c!fK5-Gg)
zjcPUAA9E~~?Nwj3G-TQrXsvv(2GwpbqSTn9ht6u7f)8TtA)g#^!LcDTP#|w6JHNWb
zG_X(6ySUm2b$KGU9%j=ij<KoCuTa1H2k`Lk!0UfKg~*#LviLusY<N+5{ZGm?4p)*@
zm>&7BUwosOV+UEdXU%_u0(>{&DyAfz_BJ}|zXY0pR?mMOll~rw@+^W41Pg>@G5+>`
zonyWKL)@5Zf}z>CvNFySf?C!P48xEJcWD0yPV(OZ*Z<}n1hKJD8*O{@7`Vb%V;8gE
z42JVSKfQ}l?7EH_+Drlf33}{80`vNRzebK7Zokxyv=pc{`$!#Ql2WKI!hOeM3-%+D
zKnIa7>GE_WJL>q)AJE*!<cpsUs<i*}=l@Kdf8C<LU;59_tO=|Pe@y?L1IWYDX=66Q
z-AfEViDP~>?OECXdMTJA4Zw9SZGlh*p{!{Z{40ReyJhZIh<J4wAmw20Xj?S!kcBn|
zLwx>UF9I%)o8`H|<#oDVsPRG!lr^RliTLh&3h@8?nGN2=9*{Z8z;poqSk1!9yc215
z{`iO_wV7)5&-)OIV209<;qdDJC*Ico>Dpw-2q5O9{)?{v*PHi$dKK_9!7rE*ld8cn
zO!YtA{-_(+G1BNTnf*Z})Tpio#sqsu-E5t4`wLRUr|;FlxLNa2+b-<S1uW47SPa#-
zU9niZwBUTDru2WgD!3?S5Dck{qA?TTuY1Fa4Uj!ijG#d|003Ql)x-a819pG9&xvCi
z!x-RzDx*tcLRy&JywA}`(CVz#G;i0HMNpl90f21y%SXS0-|<Rp8kU;k(kqw%?sFSb
zx%1~cp-bW91b<Mbn7=cb2S{NA*kKUki(S7U&u3l6Coy^*bqwywE|RxOZS%-oGTgPo
z%kFyQ+;&#)<O}hazqeZNG~M|7I{!87|NS8oFa$TPGIaFJKLf%tu*L6kk9pef_r*3q
zg;xW5N{1{MixVKImV(;RevRn|rXj)@ZRy=$5V0e~EIWy(!C6gpFM-^R<sK~-fz`jU
zNli<&zc0_W0YJC{a82rV-0EZ`04_x8FS^m;!<kM0mi_gSZB@Bboz0)XKWhaTXenS9
zw<vw)Bs@cKzF#Q!alq&2i$j1uXvO%K%FRIR_vJZ9ozliz#;q$~ew&mPyVZ9|%I+8p
zU1-+GEm73|H&gNdyAWZ%nF%r&<~H_=TK~NrqQUaX)>rS44EO`)P%?;uo*aTH(fDQ}
zl!J~ShU`vi(HzK5{4ubty147oa1r8OhF)Rn#+dkjp8==ZWi!1@zxyEusEI+jb|qZQ
zngTh@b`M8(HCU4xnyDAxiQxi`Li2>4w#>IWR4@1YiL&~iPeBau5<`U&$HuULE`Ttm
z8gQ6GQD76GF+C`vUJ`;yO*D=cZ_NIzj2@`hf9W@}J#Y-;2UxfN&4c}C`q+RWJ3*WN
z>7Nmci=FG=qN~AYz2n{`1!X4}Y;^&%e*$n^PFb*_@{1CQ1Qq(??ce_9935nKVx&JH
zLzO)V&iCJrq-L4U0hP!aIG+8sKbHF-YPuFpvk4ji&#orZZSGE(zt7@rcnrovSA#JV
zE~OEq{)%!HWF7J?DR!H-MA9@DuU!^QXNLz^E&tyUWWQN~g7CKmxy?TxiA9heEF%l{
zSMGq!R1^eTp`AwgIenO9nE!Qu2^F3HaM7aaao=9hiYrqR{&C#qP+{|O9s`<)jic{g
z`ZfRzT9kfTT=^ds?0??DybANCJqC+$^Z)q;!pIf?P~I?d0UM0mL8;y3lV2FFO5=~)
z<&}Wkb87GrLskhp{|#KPC_Sw^9Wu@sfG);_M{5<#e6gnbb2k#<vpd?cyb1CvB8wM`
zujm~<we|zI9HvjqHn{yMfFbns5I81>#AaXW9Z8DO!Q_X%fQ!mY7mP)EDF98|23v;^
zx_-Blei}1j&cIW9YIf55VC!_VyauE$rsn{(pmx0E&T7VoRJ`|~BT&&7ieW%FdvQdT
zn)UtDoag4rvdZP7%X17ZC=+l~t`-1FP5CwkNY%KZUtT5NxUPxXMomxgfY&`=dYw6t
zac9H<n41i|-J_<#s$*YP`{SvJ7ih>NgXqQ&Af!XFDgTHFu;GlM6?sC#<bMh7z|_C7
zORlthO?%yJF`pRnkLAEZ;yyR>!o2Np7_;Y?y=C|t8YFkQODYB6Ym8F-vO69F9{|&9
zAZ7+&TKA+886WaDRAtd|0S!wZWk<g+V`#WL9_wN*cED7xE@#v=53-wwfNC_N>udHa
zkouE%pv;?h-mzxAEX-r3UI;IM<w|Q7U1T_sI@|gn<1;ESTt6Ut7UdpI1eUk_|Jdc!
zE+jEzAdi(P07CPg26_KR<KtN_@WAS>A#rOk#Mmca)2t);P^lPj5~_>IYeK^m9j`TX
zd;nxfV@<%rRoAp+V&x9J{eR4bb0__7Q()P9ior$685!W$smQiyZnS~+J}c9F^Pu9_
z^jgx>Ydt^Z&@+@D{{x`wG?YJl4^qY(X_As3xF)Y=NS@@BIM%KtQ+1wrD%7(4u71{y
zjo473M~r%+-X%zrKkrTC3b~Xg@k1wM)1V^?4VTkhNlcPS?FG)k+$_!C{8ZBeP5)%t
zC#O1lhPgv^?YL41osl>1Tw@X4dfC{GK}Wi$<VN<DuP`ufgRzj+JzZreAo9s#?5Wap
zr{oyAj$bHvUnZTQZ`Bt2PPW-=4_;o=KX6F6_+T;UrtjjHu&rGv9I^SI3DoxL3JRxl
z0Mp2-=BzELPEC^o_h;jgsnkYM?o_9Tv=?9xyKVEhL<p3PE3x$EGWoXcACjvwSVZX%
zOk9&asvS7-I4Am_RyIk@l+@~OKlppfnlrGrJUzkK1hc%+chmhYPkfx|ZV(4ELzh4`
z+mxdd1Ib}?J^-Dpj@iX1s>{7*H!}zTZ22>-6;qf2Fjpop%k741OXxg>vHy0%vK$4d
zjh`X&)+r`q;S!kbtuSaF!}qtxpjo`1Eev@75YbY%9A*flWfxWxC4;wI{X87wcW%ky
zQBpmuu0^vrhd~*EL!70N>xMZJ!p#Q41bjJLA0qTHHWBXv`)kYSJJ7~4SN7;kSQDGT
zGNC1NJ1GIf(Y#+X;0Ouq?JGTeDv3E6Fp1N|8FcHD6;PRZ?~WC=P*hQ_oP7})aQ9?+
zCLhs_|I=qpiP`xrg_VF6F)f~r+HDMsX}kINrIu$Og+A*t*>AAHSpR_<!3ZsGy)rvr
zq}F<L%Qx(Ptk&cY6b%cnBF%T%0di)VKmjydI(CmG;d#?W@q16xK7!XBpUzHvF#MmP
zWXI-3%JN^HE<LwWXkg9W{@qvOZVXe~<$1G8Hf<vu_qu@XCzwhN$#u_{%o>zQ7GA#F
zCtxLP7{rf}yZkiHxAD-BMYd^#OWt4(bOIn=yn$sOd%HJ3rHk|>O&;BbWV?1hrsKNQ
z9jZ)K1~QjH@mSZE23{6%fMAQXR509=7QXTwcaq&F5liy51n7(_3m6YgHUM2_Yxpv$
zyi${MUa{+jxJ#yE90ywAB4<v?H?Vz1n<kOq;e+wfZ%EA^*IDnuXyfOtcqBpx*t)+8
zWe>Q<UtwJLW~zQdSV}Sy=x#O>`~yAcg=|8Yf8qf+<ZL=)eNR86S)`Sum<;`NhOvBO
z;B^M0-CqFs>XM3!srHqiRudYlHgxTE6Ri0I<?daM%tXJuIM6m|5rUWpn4TX_B*P@}
z^V)u=BU&w1o6i>^+RNFWSUZ)>^BOl`r={E!a_8SkGgG4+9+P7QdqY1OK5q#moKhTC
zWhOc)oIPHoZnotq`}CO070fa{Fw2JiGt2C-Iy(L{0p2M2qmp*Xm*yOZqC#$c&r4{M
zikyMIFn9U}y00uONj-P}n`t>sTya-9wUK~W!$Tg|`isybB10R2V#~dc>m!9{j}e{X
zQP<BtUXRjzy7c~^E%48=NGZUdc;a7?nt2A>uk7FF9A_I3n$yCI;=?&(!$(u>oFys3
zc@tLCwg4E=5vq8%fU=d%nOj)va8QX9(1{Yi0;)f<TVN8{_p|H*r2r-xo|g?**|rC$
zuYQ<dk5xA#IX<;0ZFcxfa?dmUf}S~ea2h;h9^TmnXRm@&EF-0_J%OBBK|n80orAXo
zWT01uN%FfFN#UYP>#Oz48gqNnJCJKuXsaLVH{#e5$b}QS3dyref~+Ip!NS^+YNYGB
z(0F6$urdn3*sQ3{yqyXNeY;Qx@;_nisaw|_-Ibua_n_DjDR?U_%tx^2Xf&A<XFbRm
zIXjBII0Z;_t}}4D-}aaQ5bnk=%5}Dp#_fE~O&~k;1j1FD5iUA(7}YI|RI@;a`Aq`A
zK`Wg7T9@34l2s&pZd^D{N;pn?dhQm#)obBg<8shb0d)pBJn?y+$WaStZ$q7}hZ2yL
zuEI9ZpI`=Gmvy))!l6>Y^JT2kN=ZxgMvP*Yt<m(1vq}>hZvu_C1ZVD=lN#7lT5`wU
zfiBrYAwL7A7GaO7r!wRBWm}lO3^RKwB-}3Dc_~TWGvPwC(}-awiL#DIuyrIwLquF7
zI?D^IGzWr6t)t)v*G6696wf_8p7a3=UX2y{=J2q`8e`lu5puo{AT+5Lm0Z?k7f8{t
zQl)Ual=Y;@bONZ!C!>5D-0O?7RS!#W^5aOKlTVT<a3`?JCDiEt7L%lC)R4urQr!Zq
zK`&_FW9Zw*^Fq08XIy#^d&NJqvfGv7B<Cw!I9!W@9E=7nM>$Lt1D-Hs%3tkR{jQ7A
z&3N8=cqgd(d-n{0z3zDGSh=JSStZC#xsjWTOo><p9ZKQ#2Dqk+#kd+|;fic%2$97Q
zgqGRFQuxNG6it)#ZoXQj*cu8QCG1tRGIP{Wt5<6%i&@@wfKhF9<JM7^D97%j$#w@e
z-*rz$3mqiN?c-U)=GC%dMB>I&^MZ)Lq@(yV=@#zkABj(1#zy~bGi{6Ur&4nHC_`*S
zLeplknYtCQd&5WL6mc$8w&9i`K3Pfs`W#eL2i*$GOFCk6<2vqA6jCX9WPbL|F(HO(
zHnHXYk2CBWx%n-xe>ECOy+R<Ulg5FU$EzS2_{ks3fazp-g<O?@R2nC|KrQ$&<OJN1
zqqC`y7wz@vxgWTS9)){7z=4E6jK`{zwUXgIC`umdXlGqgNV@n)_5z)@Jbp(J1@RAW
zcy3K3&`|N4|MGpLb=(!K)~JR)BU7lhXZ}i}b`XlMpXH97*AMvIcXZ8phW8BER<)ia
z)C4yX+-HNVJ7MWk1>aCp&+NV)`-J7BN6*-R_r+UP1D2=q#sT~H*;_19Lq0(SK+T*<
zvNboA*P{%L=DC5jH-L%b7{mQfR;v*OUCOn6^*Q5_(jIZnq0&alDc;gSSxRef9ZW!-
zwe$gN2m^qNX^GmIj1nG@@%D!Ct=B^Lgb3_-eaxoKB`d`*%DnP?iG>vsH>j<+)I7;@
z&<du98SHq`)>zow&ON~<w07hkp~0}U<7bbUbkkURdPP)RLGz|3-KIjZ0?Tod5|3al
z1ur#CGLdLgeN=j{sya|l#%>?yY6kl!FN0o=7uLcB>#<?}v2u3hdkDi*Dj4D=Q@iu9
z0r5OI$Yn@hQz+HLNCviWRiQsrc=9~c2zG<2gVx~PscUo<0vk$GH#Yc}9jFB5Qw2gm
zSs@AkVM2`b9Vwp};l;Dq7(1df!k~>=$v-2@`V(>SS$OXHVIi-H6W?E9s`(aq4(XSN
zAJ$(<<R*Qv&F+mW9Ue>Ry7?eo-ki5L2+t+zoz0ijAs7|Zharjxq$1WkSg!m#0?ZM>
zM0?cm-FtsWwSrA#=8x$dCCIL(%7?;3c1zRo*@)EvYQ*?zcmc1xY}+&wF7k0pJykFj
z^im2>?WTJpgb^5<+Rw~o7s6v4sbpxMM(S`(uVnK{l?Pfw<juE6Q8ileqXwL^NF8=Q
z%I3R<3c8>5XtSy$@p>s8mEAuFl?c+YlB-v*SlI0+N|~}4Zktj4O8fZv8c$+;DSAN3
zJ^0vpF`VOvz}bl{NbZ+FNR^#?!eYc<V(H#us$<#716@lvBzg!2UYtl4A~-}4J4hxT
z8EDmgiW8c|SjE7!tTnoqijZ&R0M?Gctrf)*LrhiMgTk$r+;lXV-mW5HYmr{g+%x)z
z<BZ`~#?~jf$HST5j)+a$k3cxxBJeZc^F0MItW)E>Fe9Idp0Z5={C?c_P&Yo1Q1-hM
ze;pr4>>rYZE>70Z9h8;fI(8Sww(K$(5jU#MxK{=(s7;H~Rb>t*Y}h5a{=x(bhi|DH
zyU&EUnyT0&?rF1@t}5vX$Mq^rqug%&+G})cSYkw?kOLgKOCHTGqRRIQ{lth*`DK$r
z^!{c6tmL{g+&|j$9og0v()<q>6zuUuf%4VbD@FQtn|+oM?=@KjsaD8=1VsI(pA#<m
zFTaQFhJQcgzebN6?vDi&wCi7};}!|kg%ElNK@T!v#L_c#zg60Tc-(t<!g>KhdK-{q
zL=epOXhr=ry|@_ru@b`qzGL;B7doa+@M-rcu<5O%Y(xon=HscEDGV#NO)o<Y4%1PV
zw=s;Ntrf>_vg4>OD5(W>N4$WK!e}@A%adizX+#<d$H9by8IqaWaP4DRenQ>DUa2>Y
zAjA}p!-i-5g0tVHc=NUTTe)52#6EI(2_)q-jaVl^JhSYSD7)`Ry%N`pVK%!XyG3+s
zRC#6-YNU+-U{o`hD*<1T!Xu9|CaeGoiNIVw)G^ZZvNRoRbN#z>K$yM_a=gx$V_`2m
zK)(*X0{7aU54zJn9`{5&b^t&~6tObnDsVAqFcux4mb|XJuCpwKu@u6MWKWMgla%e)
zDFx#3<K*B*&Yr{z>V^qd5s~&rj*TDf=x>!XFK{`!FgA}`$->;x!ih%}(epfWIq;S5
zBi3)-%X=L+-l*k6u@lCVOr7#hZ;_f5lsc(1??5RM@v2I;+v~?v5!augpH2?v=Q9_(
z>zF>$GzzCfHvy2i>ucq9N$sG9U{@dh*{vFQ`si%{H&FMYoOBRo=3<0Iq@h|+mN7CY
z0VI6qo6K8g7@c_BTSSb^;Tm0L!+=LR|L0~5lNB*w3J$qqE*tlAO?PVMeGx~4eAV=8
zFUey7a_EHZ+X(meg6C$b%_zCYPsSnC%@H;Fa=zHcc{X(Uyn{Ma*CPW&dC@kz(<k$g
zttlJ1GsA=$wOkTcq6+{eRL#7c6kMxMj9GNLH#=|!a%>%<9e*k7LudmqsjMvbMMCYx
zUY>G$48$Fi7IXzM@H9vV*`cd{BI>2@rguY7;6Ro9Bve-1=X$3(Zs*QEwS+TieSkFb
z>;!ZaTfBJ5XU_FH(he_+a(BLY;5bgGoQmf3ScR!hV_t)L<zQf5v$-$&(fO#5fV4nN
zql%HH-+iJ{X)|;YVuBNsXc42AH#|3Ho5i#vz!()Aw6Sc4n}8~r-V)pXYScF|z)v^0
z=^g3BK$8KYMF&TWdFav69tS1|Fl~h%4_Rs8s^#v}PC{k5bK>)@?K?@iGAuj3%TKPt
zs9m-xCTi>OdN-z?l^<M!>ZR9-U1fd>m(McjmzJ|B5&P-m`&TJEXidXs7A4~Xd@rjp
zfZ&y6-*74wSGfAD_sghHi?1gQnMSW~`9<j$XlFZslbp#rpJ#E(RJ)fZ;!=;3wjT3H
ztt9Hfvw)9|CL!mDCO1cJg?0Aeni$qTz8Iv}pkC}#MxxILLK8iz9hUKkCG7nHkX^K2
z7RR$<r0pO<Dl(CF`X}OBGV=8vII-^flM9z3`bVBI#^}HNou2|4k>=gxn$IoezPcgp
zO4FneyXsP7c<_vrtCPC<-}1cF5=?r%D%pGaFHA8g4zcEJV`<JE!kLz5Y=c_qhRHj9
zh-G6@sl3d55SC?GXu8mtD^V<kkh+35tpQN$K7?Z9hKNVm!np+lawsQ~!>(mM$5w*!
zvN*b?J-@s2n5K>9?hJ;*Q#Z3FM6M2G$jkW^lr;@^Yb_`hX;#dLr<{k&E$u^&CwWT1
zky*?2F*XOYN)Q3(jwV}}$cbZg=bZ_RZc|oaA7ELL4G$-Q5y$+2vC}J|R;YIJC5J$S
z79HkjAq8cN%TMd>$_c)C<C`eC<sf1B{pR8uSV)GB?ihPFqS%6WAHwg?Zci%gV47HT
z=0VX-V~?I)KlL16J>wezCk6^h5$mD|d-0^0#H-O{SayYjULQo*7ike!DH3O~6HuUg
zL48kO2u1qdfGfmFBs{!H;pm=2BUES4iNq5?!;@UeF4UT>0g!(4#1O5{*a&=3k31W>
zkZLOmKCkSD_}FExI+;!nxOt6+2ga_Q-7=@<IPZn*RNu3t**7uttW6f&OaZkjF!9=H
z|6&k;ZaWA@OIX->)Q+J<3q+#xh?n>A`G6v|nm9-pW@93*Ic@W(hwKxBor*<)6ELaX
z)WAx5ggDXl<!|GJtosxaHoMf8l%2aL{@VdgOczuh?~Zm~OLLz=g#XGW2tCRBx><dN
zD4ue;whGT=Liiz>l2_~7cUAIfv_wi0f&K&@G3W8j1m+VkTgP=bw`=Ijtw>Ubx^-wm
zjcZJ(G8M0=dEBiN7Ol=7<3O0iL#M{KpgdVY=6@LF%t;i*a|j6QRYsPcrFX%U>&`*Q
z9A2l&RD=YzbS$c$u5?RVzlK+s2May&jeC4Kz28y6S0O>=5uDijfy_EQsELcnp;`%c
z`4RP^iDO_}>k&kA^!Bg<LpR+nYx=~5nzoQAu;*rIo@ZsQ1Vk&6-m6)On;}d%c~#3L
zH_CH5HA|4RKB*AxO<zTycW*<0C@+rC>RKR_NN1&I<N!ywa^i919jBX8t)OBvN%tVn
zQS8k#U1OK%4ks)Z?6*Rvk^W7JehO<(7-F&Wo}r1DqCQqPR*Z|NjZ82N;kCDtH+Cw1
z)p}F>Q*Jp{X(mLiUdd;1I#C9bK$+l=+0t6b_fd?olzY<p@?R%^%sCE=Y_Ns>Um~+8
zOI)j{Lp}#(i9V~rUaR0kzJSU8SU$Rso-zro@%f2J=rze7)!BSdtDK%DKv#|m3v!J7
zBE0Ci8))EqN>!0td%Hp!8|7p0eotn{TZ(zP)~0=|MbiR_f{o@WL6wXOdalp^xcA=K
z%K?X5KrPF(H0{ticgypl5^vuYPt*1o6`A(X4iWb*oU%)cf{w8!*^0+bQ~sba{~Z(|
z9H*yZWhsifnk?kh?xu2v?F_oC!LxXLA{{nzrEQ;}`&X*oghRX3Ro2Rk0^_x4f0?MH
zb%#b3BcTNNy}lzxX4mVKb&Aa(*sgH#X~7Q&SHoD27<lK;=;>>PZldp&B>HHw_;0(y
z9<br0Nk8T7zp2w~Jr!JLr*2bF{UB}|ijHf_j)v%+j@7OoZAliC+OBX_t#Yh<8?z4E
zcIv~fdeu+<z1$_euI#S2quwKG)pd5vHppVc_Uv(y&2)+*{oX+}W)A9%?QngD>OO&L
z-$p{EB5PF6f7V17P;c%Jgec7Cd*6%++VGuoth?YU;PU;$MHeg~2)mgSKA0W;B`&->
zK73($Nr?PZ*kk|5YXS?V+q;(CqvQ(>)P#~gWUFGoq1x%1*?y}WJ(n#3+G{p4j$oH>
zm3-*$t=5x=el`y&qk7l7eeER7){T}ZKAIF@zb7$LC3Tz~A?Wz$0=QKSJg@BhJz_P=
z1~#!<9h-G-Mu*><;503@jz90wMn4xTq?Nw|YraB!yfmjwZIalt_W<f_R?zWa{Ua*3
z;m8Ip`168<ujC8tVK0!xw7FiHiyCi<mrYW;jo68}KdKIP=d)u-_T~8<F{A~tX0N{C
z<yFKc;0aRn^lxL|EmPj=tB_-{d?!FOHx-bxl`u!$_ETAoWb!kU62s_qzQ<pxfALco
z_I9o0ldnP62O*L0U{1eiw>TavjOzaD*h9<f<>HW_WR)S9tU6nnhm&hEzDukm%i9V|
zW#sZEOtP@wg}YC3Qh9*9hHB3W0`a=ACXVK{<Djge9ORhC=)km!cGi^a36y<?g3I0|
zq@34-nP}N+RYSHeNkKld#}<=(GA`^J)P=FHyGo+VOl}$%47qjJgbkF5&=QTPB)*cY
zWA^EzNk+$t8#&qxDs=q%Jt0NmK3j15+m9<;w}x{smR`9)z2J)#r+Ua2B&9mCwmVU%
zH&KzA6=L~Wxwe4#mXe9`+HfAH($FuB(%(lH7d2U+B=%i;;KNbvQN|0o%>zlB0dO)f
zo&R{~d3d_DvuFZ2S~a0v9i}?QA6yx!=EH`t{QV_}i)GBq6Xs9w*H_eI?8{p(8uFH3
zX*;nrq~2;bZsWp*t%ZODK#)VL42`m+N-<dlf3o2fr|pm9S4pxTabkKu^fZ6b{8^dv
zo+dWM^zx%~Lyp6dZVCPTcgbYDTe2OFGoK4J!bM#@@1%~jwh#viEWa~a=wK2su8eW+
z#*vr(sC{zJQmVNEv9(qq9M)Z=bl+#^M~ARWtO9rTqQr?P=dN&a0PbOeTQ<K$t<6`~
zr7{ubi)*@S#<Pl1zWD6U7j*Xt)<znku&=~dUS4ZB&juCKxAu}B-)EL>TgxG;yzn(I
z^SuNLk6g6Mtskix={-nCib#YZ#^YkVe>Y^zDayAS%?5h;5(J0^j%ghtm1l_8MDBgf
z5kq|+$#rWen3g${N#0y%8ED01%{Mr=mssBLr*|vLXUp=mU&$N}{1uC3eFT|YPb|1-
z92@f1@n{j)5#X2dDH6IdRsUEra`OUpVQh4?SD4CF8VPQ7i!JK{(~I1g2><K5Du%Ei
zPR8%?Uf$}m+fPc|dX0X1)rwCaGkzBN+Iai6Rw1c^@ycb0_B2iNZ&J|xC-_tDd3{pT
zwYTrAlMSJ}t<;O@zBtV4yHf8G$esgn6sb_2HUc^2!=mihBEuQdGSo<ix|p~o#K!y>
z)7oXojJr_i!!pjRDzXaqpAX$-zu7%NYRV0G!cxK>ScE?>w;!LN2~_tz+iPWQelLZ$
zh2jlfG+yjxj(T~<r#9c$*sUztdz8b^;=p!24B7jp-#c-stWHLzH^@z8)G}MP;VBxw
zZcb)RS#Gv*_V$#JFUQc2TAK!FEB#xY;f=Y~w4eH_Mk6;ZUcFx2!N!dM%}`7Us!pDV
zi@hjnIarR1={SDgKvS^|Iy`qcO&z<wc{;6qv2dBdq3r-0coF3m1rExtmdrePNvo>0
ze72Hz>Hj;FS+E%i)d(iYUz7Z2IJc~^RDjZ0`-728)97C->{(CUCxwlfuso0vwjtD3
znZvCb+)ZN~jBNS@52^cj{W-5Cx!W}Yj^`_y!9vtCI&&_gDUc~<T1_&OPaarOo&}YI
zG1rA$Z-=C9Ue52gGPElb$$E}QRT%pu=ZT!?7jYW|T(7y883PIZ;tH>rE(y}Qkov%V
zs;y6nW5==M7%l&xKH%-5M_Z*<W6P!7{u|Pc9EHQrT`3!~-|mIs?E3nfzbv|EBYR4j
z7c17_YmT#m)%<37cHLbL=SF)R;_bucUMIIbf-ro|t96_)t~fV{x9&l!_0N;=W&4ik
ztms{d)fbdswe~r*7Lt-^U!6r8H8f=lmrjk1etYs`pY-W8@GtFTr;Hd8gxSs%zUb|-
zGdxY~U}d+S<g+3dXL2AzK!m$QdaDVxH|)v>6oN-&WgeWi;LDQXG|dk8wdm9<Oi0>8
zh_F1Ozh5U15MJ?DlOf{!B|NOLRv3N5%&DbZ?>sZupY2WKX5tUuX@FZgI~n0ZObK8X
z8#d&Xi-*2v3w8}gIogrp`H=!mpO)-ucUJUPLs$w%98a8a-SzjlL_@u`Mg}_4O{E5c
zS0wtTXM_FEvzC<FLB3mj*Q;2MMJ`NZq2om(4OuIT^d6Ua==t;Oq>i9J&huS#!9gVi
zx1_I3IKwyV0LWeL@9R%DQgrWDw&T!Ymr)u#eDkS(Npc0Au^(Q6Zy!4E4$3boUC&Z8
z1L~Ia2tujr(L1KFkaHC!`$r^gNik@jtCLwwD?&HGB@}RaZL3W5McwSE>X7Y4>8Kis
zs2RG{L4Wfv1aW2I5zqn2_1hxq%rdT@tDf2X^+MCCyEK~<qL7DE4VXbZP_I14xYq;?
z{m?JTGiSGF8gmqM=a&%G4*Vue9`~QGt86T+3+6K~lm1?gX|Tge-F1TS8JjZt1E})=
zeaZw*Aq0dz?L>U0>cTCi-jf}XOozMZZ0h2Y6#OghLp{Ex#BAo=es_AQe^Wr=Vbqdv
z?O?ah3}i#GBu$@fu2Gtz_Jl)^Kb1wdP|p(!{~C^d7Zo93wKv+*z{&H2$cauMCt{AU
zxGI=uccnV<l~foD4SUOG3rJc~nBM5+{obmv0gA@A+YQ3-$t7>%oLI61*=haK8bs9P
zFS_WDeaHz;2?@KO+G|}TU+WEeplB|2z9tfS8h5({<G^aK9wuJAr`3V(aaxMQuDmJW
zxNEDt@aX5jgEsftg<CSHa7{`?`U?-IKHtkCwh4km<k5<?R1B7}nN7>jT?TnNvVkA|
ze2a|H-Y;+b9*rZF*86e(<CbYjYL%qps&ysF%F7?Ou=S@H3gTJ&ADh=96SQUTOl7)y
zyy;){TDC59wmVXmI&;2&FpefUFEA81L0-?~+EZi<q^!Z;hg|#;&Y+$RD=x|=;;oQ}
zBZD{-s^~~qS?uIvJ1%x7LdMG=FMftoQ%)I~`iqp(Wr_5?d5SbZJS#>amkUXz%hE`P
z6{OM;@vAp+{5+%trYG6Q4XzWzok$B-+I|4(UuMN?3N48@cSW^iJ)mEJGQtl3NOs)r
z(LzQR?_vRI%kAZNCQ|7G<(7AS-h=Mqm(3U3j;i&?t)#w!0Kjfc`i?p5f~MV)tb}2s
zBAgeuy??i<d$bvZq|Y8dKL$Oh-~gF`kPdp64G)+o)7Hz69)_Az>z~>@rzPAr`G%59
zaZFm(j#m2r*n11NsMfX*R6;<hK_#RIL_lExgA#@Yl?DZr22oN<a)y!yDM3;~K)|9z
zy1PS=7`l7tuJbIvcgOdB`+d*(1I~5LwXd}Sbusg-=eh6S{R?>t7!C!^D!wzeCyZe!
zu5*pfU2YYOSQx&)_x0SCAKr-@*1ccl>wdnMpNtZ=*i+Pa)4j^eo%$rzkr9WzOn}q<
z9yFkb6G+1b7sF7OyC{%sCVcB~9NU86Drd&!?UH@&an3iCa06Y(cS1X?BHG3wcO{7e
zbU|%Zp6tY|Fgm;_Aku5R#mB{D(d|zs=8k@9nb$Ct*arfiuI_fPvOZ}|Fw<}elQ;`=
zdQy&>G`^4TqoA%s6+J655En-OLbq*Em626~sCg~WZlX@iosJUEirGp=^%``xm&^Tf
z7}H%xhF-nyJnqFV?!L7{n;)J#e2|Fb{rqM11;*KZd?q}TKSLA(WQ+x>*jx0I-d#)n
z^6_8K9CUbyui-zRdbBoi<FCR<4fnJm{XlcY5kv{bqvH5Hdd!d$=1AYLlImKtxrV*!
zn}dQh+mVXUxT@+q9)Yw%jyw``c-T1}%(Sq_CvznChmzP#iF%n^{I*llTV77~li?*c
z4e0J+9j+e1c_1hh@W?a8x-=jAexe8qCU5tZ*ac(u*edZ5LG;qaX&oj(f&P=_E?Ii-
z>|%dW`y7!up5XQ>mb)I476|%)1)b{Wi0WH=r<84xO2VAGN8+9(=UR?G+N^I-H)y>3
zSat2ro`x}Ug2K!eo^lYS7-D&FF8Dw`r@1N^r$a_u)+OqR9nJ{4G?ar*ur9K^WUyBU
z_5zWAo9Q?(--*xXz(8RdS(i8Ux;z(7%^F7xZzr7Y&~GPoxk=4O)2^Fxbyp6_f!g%%
zh#*WJYvPcieqI>Tb^fVeqS5eaEx+Y(1LfQpq9aq|F137YRWmu=O$>iTiAo)mwipnz
zxr6gwuuK0I|7e@&j!aU#V}H5CsEutLxWkfpj;)&DVsF$I?_QSS>6FmsTy0MDJ<Sfn
z+4&^uG(`?s>=;FRg^$e8l|82~34Qt)qP=}eN4YoyV&preb>Hvg1+l1c-ej_`G*)RL
z9N3jd2(0L}rMzI}=~oyE4Bw`t?q1IL^eM;&QydA4yq}<4LBHeUu;es!9o&%-oy%U0
zvVDZjS9Qi2U;bXv(YRps4qxXv+-t1d4uq{ofJ;a;?5La&!32Dfc03OwO1htS`yHM0
zD={dDcvIS1@K9Td_KS!|v%=%yPVvt(-u$(MHV5|+sdfB1DI-6#jR*(n9MxW|b4E|O
zNXHq?i*)JSr&|^4QeC>tzl>*QIE+n1SjHGj`GN^NOOjhb4y6AqNR!f4pLN#llF-`k
zd(`+nzEr_2lSyJ&6hN-_6cr_dUB`F?_Iy7MCM@OizYO9>9LSN(898p5o$nxw@9>SE
zV|fKkR~WZT(85kzUPktLu4#&07d}0eTV6^I@!3V}k>93^E(vLyjSqV_d|IAt0yYi4
zKkC-&>!*8FVuxefhW;0iu^7wm9m1!blSTU=QIfKXLPzJ;qgS@?-38b17Uaq@&)_+>
zhL5&0N4+uO^Hj>~oc_}XYm%Z{<jg_-?F}q*O#L!d8T(49i|J=CyQ`+1(ui@>tdqM+
zfOMpzfkU=*;f3af<7kJjnw}R~m%6sg3zf=R52|I6AiOG;_}!a5rcqos;ll~Ut5wIR
z^l4!Y)^`rqP0D5&E>@Mfuun5F$cn`y>DHPT<ZQ^tC^k;3GySm6JS$=DOlw2VnG#Hi
z<7GO^Rfon*N3jP@aPF$h%dK)vK92=?K{0z;9Ngb7x{=^<M+d&+t-Dm)<?NyH(!-g(
z4u!X;feDB$d=1nXICU?DR6sdfEJ1p+e7N5IgmpPr?u@yAWcka1f4yi!=gu)maL~)F
zMp3wXGXO(R#mv56yIY9(JL;LvMGq2f*WUcj;{>sL)Kx_X9jjx6t4Szqpi`FOg&x1Z
z??=k`*YH;^L@Hgvfi`cLSA?Qx40}K)ji;Q>to~_}yscj9`cxv0Cs6#c`^e7JSCIPB
z1Uj5u8iRPM8gyPbRs==%6DI6<Q%|mWIg@S>@FWI#dAZ;#Ve`OAyOrGJSqrFa40tmE
zPiT0>Uj+mNPv~njwG|oBdw<=!6ih8G#GEk%`d{PAll36j54MbSQ<!GEwAjvrCop1D
z=NF!JQQjwJCXBilzg|s7m6|7w!h~6``NL&-E!L;@UAA*$Xj)uYJ#2<tmHxF{y-)dI
z5|tMYO}%QsgArnN<)Ze3lsCt1Xep8hmCb}kw0Z=QO!cOTxJEdwFkANB5K-3mQiFb^
zhRG#d!G(%wtV8#4@QISD!0zpd?&Y@LMj5wk=JbzAYL0Z4xf0Xn_2DEoe1enuAB!lp
zG&%@VZ(~Wn=t-84a_H9?Gy}tXhLKNQQOT$yB2?z+WP@;FTtn!A>t5+|fY(1x<8D0~
z<}QiGxwn@}<3BmAE-kP`eXnhvm$_cvc*&N`B%%2zPjQK5CtA2o-h<%nzOgt3&awm{
z!yQXW@t<De{?ro!W6s7T^qXo)1=W{BNglFf)X$x?9lzh!>VKS35=w^Ye`I#M<v!RF
zRKS+-g4$aoO0LMWg6}?GpZN14K@3A08DEN&wzvUroCCrBRWuq99V7$%Xx`&kU~VjF
z^57&$n);g3+P>t$#p+G1zEVo;lBjng5rh*-^D5Fk6qOkaYmS*eCuxWuAroP#q)nN?
z9Lj}1Ge%-i>J$r0ij<qmu#)6`h*HIp+q(S{e){6<&9+k4ZuY}r@-`!y?F02UQp-?R
z#T$Qtd9am<@YP{>(D4SzNAhaaIqBuqT4qV{q%Ji5LelUXasvO0={Pl{Ovvrar5A7x
zxXB59v;t&YFC1eLK?DbiZPwz41_{=cE<CEQL^2U@W;eo9yOi;|@d6?T%ws^esJ)G>
zOzCPP<f|Y!^pQc9U!Gtfsgj|<n?tgk+&Pro&N)cAltBNG=svAr^7!*(ve~dbq2t|2
zsRmH$bd1nx2gl}yqPOGE+`AHhc<E5K43-;Om@#g*<cO3+y0fl#QX1Z1`o59mNgfTW
zB%}OHgtKgZWM*H4G5@lk+n1ReCR~3{dAE#xH}~ouK8-`vc#Ff?E@MzcDAcC^&#PJs
z4vKOu&IhwnzYBE(KlW;H!k4E!OG(BtCm`6Azg+pc%j?9Yn|2b9=T(pw&Q7YQe@qX$
zArX&9yG>tj-nhQ0BH*$r+1`}SPj7Z6z4ZKKl}hE%v~e~b(~XNUcs#ZF&RNc1ovV&n
zVk`_EU6CVceOo*cr9T!kJ?qSOX6a<yV*B&K`*j~~7gnZY3Xh@h3<D%`6X)qumC6wi
zbyl(5wS?)w^X5T{LNj4uji&;SOA4A|`5D1bc0i_9w9N2z-SV&V`zjwy>7)|xE!eIk
zZSr7Vga+SwZj&0xoDbeXT0*nb)9}uhUam@UYS#z(tOz^?P4Z4S8YLJ@GGaV+qg<{m
zQ`;8YV;^{j4$D`Z0qIo~CM+v50m36V*9Qenq>m&}_;Q#|C9VhRi6&1!euJMJQ^<*R
zCA)8STQD$OV;VHc_8x65O{{RZS2bWfd!&x-<7|!LXU!zlPU&dxD$<)015x9~;fx<%
zj3$xX8?Dk({(>a5PmrjX3ll%hBa9whl2aOI=CAo712@Q-+Cz7he^rw+)3#5tN6)%;
zE1z*S^}G}sS=B`}Iqo6WmOmzuZXR=v%MJft$xp#bmaO;F+)oP?{%pl>>U2Of0g_9}
z;i61>KZw1Fa030wf9@+b5fYCcvJq+QSHp6&(=(I16^WX+Hms~s36sAVl#NC19g=S6
z!3bE=CD3YYeQfyBS}p#{LiBdc@C}NGV~m|9iH`4jhr-rLR2fb*@Fmbu{P$Co*7>N%
z-fv_ndX&<0I2~d@gIsRU%m8o+%kmr!{jgEv1hh@tnhDaA+fR5?^-WIaj+riFx?2G;
z&a$b+5|@+@ns>H{p^9?f=1o|^0jbixMv{xbR}{{?(ciPnsy6R!mZVHXbHMvH%;{4R
zx?B+%tBSun8`ghsHpI1Uh0~due#WJnQ*V{}e~u35^iWl-)+gckv$p+R@oM5CL+mys
zl4f`oN|Q3ZS$L_4_S+U`7_Yr0Uw`Af!a;tU1BcqipP}qU#|)m#Z4Td>?<au79#fJU
ziNdEBd8=yk5mTnBa{pl}^x#|}DyA-*k@TO@J9w+?9k0WsDj5IPqy8-!U}nxR^<7Ke
zEwW&$_LbQ9x<xJ=KSg6+=S$lnqq0b>zc1OO{=gq)6ZQ(Z)2GhRwfw4Dv_=ktD4uv-
zm|47Vvn1uF+p2}W1xjRXJ1p-l%36L_q%;;;!g2k;@?uMc#cb~|VzgIhlYHteO3NCB
z216)ign~l-6SiU+z%V_l6MfCThty;%t<3(`o4pwVdg=D&Fpq0*og8#Zo5t?|ZaX_Q
z&h=!8lxY`w8|AtdnGEiEz+Yr+5sQC-=(iARBs&Xd2E6o_5!b143q|Ka({dqUom>*b
z_=0_-tedx_X(8HvsaZ}w{qAZk*p%`)Z>|hw4u9jMK9P}PaN;@QramB4YZ3D?!ziCv
z#q@u+QtVZ{@e*`ER4on8AzJ$(^HW1J-*c`}quKXz2L$_Henh*K*_#gQc4M$ApVL+|
zPm2Y*W+buBP0rOgg!L82c%P+9iV26ZngFF-dH7D0lOAwGY&q{4;gL+rvO%XM@fW}#
zU0p?X1^;&q@yBvlkgd+Vtbimn+2rIaBfI7hEY&1!a0h1K8SW*Xtii7L@TlNN8hq$s
z8c_k8&4_K2FT)uyCAUE7+$+0Gvn(n8!)dflUrq*~7zXK=#B+ZDF5-3ROP~;W@e~MX
zB2l680qEUvk263_zUdoKThYh1r_=wI6ys6<@OUGSx;6&Vo#*i|-6SRnra0LC+>$B&
z8#6GZOx>i1^>+Jbh>14Q+ouB`iSjhwoBF*kg`!~L<5`f`1JOE=xKzRDDW`3%hXY>-
z;Z}lV4_;_^+EABwx<4J(w5B6{2gsnFPQ}y1V5fjB`Y_<p2<kD>mOJnXpRZGtuMfaR
z-0a!A5v($T(iA?JiR=Z+hBy!im7lR3L>F*6(rX(<Xd<bTZur#qv*RpdthQxVq?>Bt
zthnAN9m;O42|iK^l7X~0m0g%y?UXwLx1;e`*jl}gN}X~WCxI&*9KMo9vOh>%iB-J)
zQ=7_q?v$xE)grqjL&j9zh2=Q?E82CXQ4|bg51MTsKMD2OWIiGK1#?FH=u-<kRKH@T
zHVzwr;!ZE%XpYuob}B*{$dU%Y>m)u^&Sp;iH9IDPJ|DIY7_N#1wnBY9{WHY`x-4Ha
ztgeg6fK8%?28Vgd8sBk{cb>|MuK&Q*GGWK)35?tddE~Ym3o@tM7|RkQb*;p36_Ix`
z5T#CajiRuaUIJ#vwSGN*#5!e?OMM*^OXG+7$p(ubh(G2A%(%vJo*4wYjH$2xI4B`t
z`;2Ix`p29%CEsIi)Ja1s6bw42$KI|#6_1MM&7gP$^XgHmCX`<V4C<H9@U-`8dSaF4
z0CZ6R^j3E=n62KYP*Tu*z>j_H=uOJc%amoF-Yse~xYZ?Qtzc1BjZ0jAMP4B-vpdm~
zTsV%rGJ%|yPF)jHR`!tbcJ%pNxNHdsWh;y|M53Q}U;0qJdCLafuiGVqi<%jtCh^MD
zMMEGkilXXQte)-e(2FomYETK>90qMnyNoh--tCYXfHP7b>FO9^-8!g%sc7`#l{m+N
z^OW2J^9%@@Kr^klT3aAHxkWu3hsp~!yQ#ZCsZPk`G855QQ1ZQlU%T?_yh+g~FZ*i>
z5sP{h9p|6%b?}(QT2~$z1>L9aj|<jdVE{}X2ZfI-wJm5fE?thx$AB!5YXB+H=z2LK
z^`<(}I;KWYxP8fsyuvEd1#7GPmg~DfwL-IJ6QQc`Yf&<NLS7_l2oh%k`I&+`?Ir<z
zTt@6f(wq@y*3NaSq%QBPxh6vT2cX0J;53WVAF7PaAMkZAtiL3GC$X!bBJ#RoK5M8T
z%VBPmT%{ur^EGZKZ#%@D_MVrTW9iCRE@3k1i25~n%}tQHeyCSjjG<~H8W2c0<T8|J
zi&5kSz^bx$7YqhvMf%u7%bNQD<t8voYPu;bthAvo4Z#?4YjzC5NSa`+b3Fd;^8p7)
zQrW%F5%&HihLc*Ez#c=i^qTm2$oP$@PsBfsmTWSNvu_qJbOT{gf+=x8^0=geuTq(;
za~Ept@(DU4tKx9!3#%fMh`#rM`dXIvsh5EQYsbOkVcMyzOUToWumq10lY(!@g{&S8
zo(#Q};e=-a0+`%XtKP;QB|A}v*M)ESgZJEpc<&z62VOQ7b{RV<5yy;3JumonKp*W;
za8f{tcphd42%m)QzCUb<_v&9FXzya4FLBS_eq_Vr@7O@8c}M9U_;@-RK89F*`|HrG
ze)CXVvij<enRnKI?WYUko-yysug%WVdBp?@G5bvTeo&5!cQWI|O^fzazD(t<B7|4&
z`MkP0Gd-xrQ-XloUgv30b-^|9dmSl+-t0m%g$a%cUP{uIF5zs+X(l&u2olkU`7}}O
za)$Q}Jd?aOduaN`LK@Z7<Dm^V<sGo^gD<UrQkw3n5RFr=GRGQnp2BR41uFZ4-NI1I
z-vRrqKvZL7k#Rth<nf56PG+)K08FA4(4mH4_u*rbL`{Tu1C~bQA8+I*4N!b#ezyI|
z%gUd@(G;bt<c*5DmCShCM?2t0eoB~o1*+?!AxUkHqZhdsOmz%2gS)-Sr+u*Z@EMj-
zh{qXl23vz0*jcNurUtkL7{jg{+^L1zO}W;~5j_bYoTR`_RC43PkAz6?^}aMOfHVf{
zoHHq?@+%wW0F9<XoQOt$5_y9Ml+W@9Itj%_>_zoj{O||%Ja7FI>|#4?27XCWJiw}W
zK6i80?_6I+`_N2Fq}29|x>(cNy;e^rny6_2s`lQ`M5PAjw@|3Y?5mI*HD!Zfbu8@b
zh8Gc2pN;-rE7j{xN6BG&<zMI-X9C^OD6rr5u_w0*`Iz?c3|5oz(Ml>OONt-`0mHUc
z)Qt6@Im~naN)3$_UXAEGzxy!l5g<LIE%e!s<M-OdmAE@Vv1Dq&l8SR?=+AqtJ2X7T
zNZjX*O&(29VaJbpRg&^8Ef5exSBAdmWZxARvgdK^E-i><#HE|hef#8*+0mzgK6Q*K
z@_8Fh&vh%Lkqi>)NS=J4(liU~;;z&>?^#*^>?alO`pWtTL)X9`XEwk9VpWOm6A7k8
z6>M_;wI_>ST3@uBYO~Z^|3OS*&fA>MjGLBtPOpzfyu*w7goFcruAS#(i;r8zApNpI
zO0d@tM;w$pkfdR-;k)mMUZDtBKAQ?G=mZ%_zx-T57txR^9SEVJ<M=_ZzfCn9JGRyC
zOr35Dci~A>Nu5t}N|d%}un4@lCjql%j!xqV2}qCGzA}9yn-wcn@6T)r477O|&7ZK7
z62TV<;ZtzZ^klK}YrAjBHa&?*kZ73yIbVx$AS>Ul_mbWyd+mH5&y8n)Lr!E=>*t>M
z6<=msUrG%kiLMmfRw_YJY?is^Ad|})n+oITX3TV|i2Zo<9(fdd+e@wzvBe4UvK0Cf
zf`oK>qK0vg6&~WmTpX;g=!|o|mA!Snuhd7cqemTtwj{Kh6{r&T;x_L))#Q98+4_B|
zG_`TU{qhW}Y&;G2rk6LHzOBq74+M~s5=**5Yf%C#b|*ZIO|A@9ib9=wg=Xq@x~4Jw
zo03UEA54cEmbZj*c`$rm4WjYB8uW95kJH{Y&wH~>=BN18R~D>6X0?UutwC0*P0LrJ
zsMgXP{@N#KtN@@xZ^f3e9rS0T`A-+^Hu%C9fTI;r#*`?aY)H}j``6{suYYaWt_#p#
zZ{7hD>}T;WoMiL9n2~7+T>0Dj_ze>+ii0g`@tkhazklc7f6$%ijYAEUdKpwXIs#5K
zX6Z}LOMr@Y<T?{C`ybdOw_VAAD5zFIPQhnlHYC+=GoHNPRH~~;>{sglw+i6*FY)FI
z?j_vNsJAM@e@?dk`65N!)9gwd;~qH7egIvvNJg)Y=?CAbH;`cq0=VTRWvX&X-y78g
z5y+Ad9kKQ@V_*#{W3T?Lj%iwpuZEdPuAgVoN_+KW#<&1D6M~d7mOzyLqk&4>l4h-s
zen;!ampL7C`M*`bzsGIQMZg_)I-6wE|JCUK)BFDh7nf)T@@7c5W%5w!jzF)%ZJAbJ
z<MPs&b~Fw|-CmhV7YbBH9;8F**1M1*i7XuCPQLGrero<H_J$LPBfh#dcCJP6z#T}8
zEJ4kYoG~k?vHE$awcvMc0=1`qylyk%TmUw1w=_ec&HX>WWG{Hd=(`J*&%TR1!VFaQ
zH-^w@j)~qV+_=9t70kn5A95}JnD;JrJC@Puk@y#(2;-h?bJ_HTfBzc|`Y6I-Hhq+F
zm$?4Ahd*!T|I5W0G_E|;GFANJZVgB-6@%8Owyqj@+ubFw&*Xyk4^S0~fW_N*#RBN%
zvbgh`S<ddI-2LlefR^e%zP$L9!TROXuWwEH*Ai<V#BU5^vQ`k<?o(Ip7-W`xf(1$N
zIQPI`PuDB07|UA&sPaxB`8pgWEWB11-UC|Z@8t}HXa>Oj!l@|%(z=2XOnu`9n8nr1
zfs?S&vFauyiWda{gkp@_ejQUG@bAhu8`m1iN|-f$u?<`(SS)!zUU@PEIdB5W7sKih
z?apFy#PUKGkBG^C{-D4+e#p!WH(<(r_4$uyq<Vps7f7o7VCxVN=@!WJkrPxg9+p6(
zYNfuU4?a)~SZ|iZ+<<d=2JXWMaHvtn3Gbjq1*+*V*h@x$H$a@lA#ZY|DPUZGcU$*n
zvJ}>Ys1P0aup{PpQ4smQl3Pb+fz%p2Y*BVMl4Ep97X{rRze#IUmN6K;J5Rnj)le3i
zy>bMmi`~TwU=mXZ0^{34LB-x!nOCYN(ALvw25LjN^vBgo$TtB{stW+U*r4ErMPH_e
zY<5~&<$BaVerkVwalzZu$J4F)=chlHQd(Gb+4#xJBS4QgssgHgIsGX)v!I-+Q`aM4
zl41=sipiE<=c#g^JdbOD;X;0Vka5OcJBQPQ`QfWC-ct-m^?+yWg|<;3+sfxJ+4)74
zJKx}HlFz;aWigFjkp{57Zv63-0*Ats%`-*}zu*`mCw*f-cIpD6?y#)n09%TMsfLh3
zyYBT`VxB=L`k@#NG|k^3g0zzGV)~!I`y!CRYZi<GdVB0QFtgtUg2FDK`wn}vYZ-F5
z-PP>ys*gJ9)*nyqpTFVXFWg|^uv~I2;m3bI9orQ9zx@h;FnuL)4L!dgf6SrC_c{gK
z<S=NFJrYD*?COXiWXjsvw0#nIixtc#dm#BiyyEM1Mn7h=9vN!VNoGPw!VKlkvEjFO
z7hLf|;Ol|xNtof}lgoUoDwCjB{0@b&zw`B+Jt_y{*zbvs7&AW@?p!%JbsPPesWxOu
zQ?yq;FA8?)?$-$v%L_!@7uo*LCi(Xb!<-GQ<8OkLaeu61@LZq6uLK|Bus?De4&}|p
zXKb64Hf!p+uw|S}0!%5|T!8i%NRjM=Ip(0k?T)@d9=HLXuZDDrBW@j=UIK-TY+XbC
zu4J+-#X};&5OH$ngsXX>WC0IjpiSdHy{fZVz^0RR1Hq>8ulKd|^8WDnDHZNl&^1*v
zqSqy)UhRZBmDhnIfNvN)?Z!8nAyq!#FY~fkwTF{6&{Do5vPdxwmH>V4kC%WAO`%z~
z3<DGYTWn-d%EWHzKt*RIu#3ZhD(snhrqW%J8sO0owEIcft#u68^ArWAn(4nohgX3|
zX|P$IOhFz@GLJC8mN35DqM0pevAK>1Xr5y=oJWdnA*m<+y-iPnGB7#GH5TzWS=L_#
zckD<s=?*T3&rZhAhF%jROmBY8eQgE0NPd7xy()L)qBnSYKZCXc#MM{=p47`{ft)Nt
zki1|Bn5_&Al1D=BLJ*80u$!lL1vivQ4B5I1{CEXo5e40nliW7I$Mm}Z(6A^Dyb~bc
zCKsUsM`X$5<F`kK<AX<l-idd#sO&-O=vUMN`D$nY%gs$I?ZQt9F?z0MGn757dc*6U
z#fupU7_ga}03Nq}@7=*6un!rZfgxbRE|hW%nQh+LjgZYSLij;)btDiIQ#&=dR+(|g
z+JKwJfPQ*lu;En)gw|O0&Z&$=u`(_eg1li+P3+LlYv>*Tn+Xal+KriVL*n$nLO=fC
zTeaB2K_g;qu%`XX+1g<#wLXh=FA}T@P1iJ@nz0Q73oz~0G<tY%t)BbAY|l%hNJzQO
zN~+-15mi?3agbnDE>q>lw-R8j$g<Y3A3UozqOZpjG?1kaMpfRx07ERubsy9>esThs
zmrFKctMkk(X%fq(Ldv}sWn|F->5vaa)7EQrHM;6#H<+!7!=n6ae&=>u@<ICxVORIB
zYF#dRhJ@fOa4N7D(adv5nq}IrCN#plXf4(@Wye1TVdqqgxWNP8O6KBwLaFzDRh)wU
zeGsH?N*RB4IlN)*+M7Uv%1p8ZJ#WqX-OuVthpTjy$#ey4BCH&mvE7cV0B-C-j)E4e
zM>(Y<y~E(4^`WH?z{&cuD-V~n<mjtF2_lQCM`AGkOzwh7V)9&hK46Dh3w~kVGyPI$
zaCV&8A!>mj?`HX}NDq<dx$5tez$3-O((c$o|0w=smdB&Te|qTr&GY}^;QgO1-0Kvy
z7=uhaiGL7AS)<S4K@tcHl94WO@W3Zffb?qsCGHt_?{izb$Ol}2F^&S`N$!LFd9gDy
z$OKF3et->Dcg<CkB>HTU+=IxFeBk!4o4uk`qJ<PahW-pxxaSF~AY!A|d}Y8Gtm9-(
zpepbVbGISv2$Jo#oot1EJ}^jOA`74*3b;t27)~IKtibn?fKbB<u#58pUk(hgR=R4-
z%X^mNadK1e)+{_V&e5yJ-6sZ^*u09u!#g4pzK_RJ3_gyV8SNvcX(1S0<n`Se2HUZX
zCMG6+iHuyf`lvz7tv0U9l78=pAm($;AqY=tE;EYr^;Ed4P{Q0_%MadjEG&*Cl_eXX
zch^^bEKZ%he8fralM+BhQEM|k8>5e`e;FVQ`r8fI0z6tG%L#xAe!Q8ww6M4IVEAjx
zol3ak?<$@McO$Zf)kC!j@cZ*A2)NcFppGz>3s?abJuz8fBg8dSlHB-+`X+uBavvK=
z@k!fq^$uwW5qfsLj=V@mmMDaRWwK)fGRy?fn3$69Wry?E+27kp2CxU)lH@Dt*odPa
zFvfL8ySf0sEjg6WU|iuG<h?R-1o;TTVWJbBc(0$7p;(xYN9K{#R@WR$0Jw%<#-8NN
z0+hDgdHSJ0h|}?~Pr{{KIn?n3`%buMGDtXv0hLW#>OGMU%zqZJy8!F>2LvRoEfu#<
z-A%dx;jAKJFizv~L1J9C&A`DeVBd&sPDOjW%g8=nGCdgk{eUTp?mP;3O%e6ePi@}i
zWwrnAc?r7ITljTftM)3^LJD&hbbo{bTYHscvl!F)TtbW$evt?@MUmk^;u3+0czl~W
zp2r$${5w$e!h-_H31H1ES!ck>NW`+C@w-<g@hSq0lH}Ie+zO}*7)0cnt;=3)GP|)Z
zk*3-v(G?7d5L1-Y;K3OdCrTvCgg{N?U;Va{*r*yzDTfxyM!_h40|(F1lleV4F)$vj
zH=U;t>D5#1FAxyX&hv##lSMD2c#T>!d^6JAZh~oy1E(z2&TV$Hdr(&FV1@|HXp?Ie
z31ZLX3k*eWQ~v!7{lC|T*Cu$L{`E7mLSBI1%3qTCW5#=D5mdN}ptX2*2#r}rrnT&N
z9=C{2rUOB=K#4PR2~7@L5hDKqg8l)PjeQkh94+T%Y!idS-V$GaOMahyFZ`X(%-fM^
z&9)uHp*$i;evI@1LhT{;8yBQVYay(OBv+BsL3SsjZ3H0p7ld#=-jWbOnkMNqN9t5l
zdLT>#UX{@Bj%O_YBwUURK6^{d@K)J+*kIA<O~Bfl+nx83xyd>+gZzxnB{2yzNmn9?
ziQkg{WEXrMjqp*x&R#E}frHq1MXPBV;(NNUB6QYcb<+Eq$qKzg&>;T$J_7ebSu>$-
zUpUkhl@Do^+AwF^PU_o*xwME@H#iLPf7QJYbKC*`NFQt!Z9kg36nnxgs~c{#G?+wW
z0>L7p_1ztnY-}QDYo9NrA*$hM-hn!bjoR;{vioLaO9i7WV;0>Y<xa<Wi<SnFMZogV
zvV-eaa+MIVVb^fuCnogfYRPn0ehCux_t{&H$#A#$-pj)p6*LNcC5iVlx|S|9z)elb
zPF;R|uQEnwkiy0OgGzdX4EN$R)0uN>WgkV5%PT&`g2x4Su@CZUBJ-KM5az+Vq1z=O
zm9&UG5s`IK4{O5R8p$8YAADR=EQGnWw;3z^gJHcsR5#_F&iyNU*Gc<a_A;VJBK4xK
zS{avtBdf*Mu1n`S7bo$`r<ir{sSEsEU}-cR*NwfhOF($4(;cNV!cf?D+vl)bJ_WlM
z@AuiEAZA-3*nY|2fm>G>&t5F8i9Ik{ZP;p-aaz<KKh&_ZwJC5z&?+G`6cLv!TJ;!q
zMk?Gha@=>m1iC657#R1Fuhg73GbBe8Se074)cFQQxMQ;EN9fG^t|sL{9mycKExBef
zu-}<_B7YX};=(H4s$*wxus$n~NKCV)5D7-0b+r)nO!0J5db>>L9L?$pXx(Fvjf0(|
zEpuxKYNB=%HlKT7sf+#F1?O1yCz@}+MJAy-EebSeyiXZ-qUXa<77B<yt`}*mj$#f`
zM4tyKR|Qs1SVNkLtuDB9;&c`aV{LDh5ZzjN@BZ=clZO9W9*Upg?@vAbbJk@Lby>WA
zUeDwwV5`1^#YA}nyk0_!AY%9ej@!i%7#zMY{-!C#imzBRg~r*j%c7-XxWT}2LvY!p
zn@J)NRUtp}oo1w0Kfu1YQavj^%yEMy+913tGBy)Un#$EF7bRQShEs{X%TXeizxiR%
zF-HK-y!pvE5@l~L)Fe87ZO@#lmTYgL*S_>A3+?%P=%rRz)?9iiVfsu`_$5n&m9g$A
z*c1ys4D!s{{1jqBh09Vd&FpQELIV?JR&!8xYZwjLDQ(5eqcfB3eo8eKp7EO!zoaI_
zTrU$PdrI+>(%1dHyxSJ1xfIdsL7)W`s&1~hp5d!&fa%N@Og`H03vUrEddCxH{J2*o
z^vdLgDcREr$uYZwa5;P=Rxum>WY#K~;I<Ykc9PK~sidOrI*k6^`3QP;vif2oTnh3S
zyhT}0Z|AFCN@bpk%oneZ&z>XnsgGix`UM>Vy$Ge<9oF@hsk_osL{Y5}C+99V18OA?
zK9Tw!b>SraL$)wRzSM8)ed!aH`m_!S{EE%ioBkX*re8#Yb60ido3$CNNH>_DNd;)U
zNjSm(>AA`py+&o1%naD&RIJ;n!0BlqN#^4ywr7>5i+B*~!QkfHMVEDmrS_iEz6YP;
zm)#Ip1#7#5vJ+|aw#@w)l#<n@P}cH~<rIgMQiV6An{v~2TD@G`?7ep{8rt!ENEUt?
zqYKB{^W?XPUc0_$p0Gep0gK(Fmm674y597vDk~<xN9gbU+rM9{_=0}+@xtQMe-%wM
zIQTT@WGAM3zuqNye2<Md41o-J(Pk*(bP)9lE^98NHR$C@Tyy^(kG4y<$5eeFw`47S
z1DH`sW<?bRw?VUho}9g#_|mxn^9=Z!$n)mf5irK8^>Z3zL<Q?hg`lDE3r|IhShL<2
zf9A6vB5|5W3zI+GE{G7V*F?^mZh~wnv%)82wU=@5CHw7w9HljtOQmC;g{xZIM5ylx
zBauEx;SBTvOadZWn6GM6>tt41=*3CAq#vc{r9%Rwgk2Kq6R38`$qiR=C8dg4=D6EV
z*qHzD0`M>Wsq%&z&a5ZBq1GQLzj0jJPjN}r3G=IFN)$FV8L|wZCIpsDg3`nRSjQCv
z5CyZd6xXCtuS(n-hM+o@Q)Ra$^~s^HXtJpeP&0lpF4Ot=jWgqW`B+#rd+-~6FhR(k
zqO-g_d237wb)h2TL+76><~f|X)?V6b)yH!7m_$xAb5%CoZ?A+aYwwezRKep)58Hct
z7l6-nYyH6>eiweZ#-({1e_)CyjCod0>qzWOR3FUu25_^-iNNh2wjRuSlYnk;n4~pk
z-e4%$yzr)v{iukYS`4tMKUshEv{SBZRd1PNpx;c>`FjK7`f-ft%+HK1ksA4Dsw)`8
zD2DyN!eiZu)e7{x)SVaqTrX)+p5%53<aZKK-g%%(`Cx(2v3!A@Y#Ld9I4#12=BG9t
z0%L1yY{a*`s?i`+^X<Kb=g4o`BRS5UP?5I9d9OpXKNhuggVFr!QPgM0dJ4H%@*siT
zSujm&-iV4pWbjf4#SzoAS6q)KFS6{tTY4~;(9s|&>pnUR)MD);sMwPGZ90R`k%`G5
zPtI~K^=rnmb=fMvHW<iT0Tw<nl_H@$9oDvegjH|dCd~L_nz8X^ETZ<-N@r&L9ct~t
zy8}AL2;p1FKYxw1g_jgxi-zXaCG&UeKd7qDjoL?*2ym5wBiItO&KAPa0y-n+pg7Jh
zna?l38SN15Pb}P?Iddnuzi5%Mfpl;1%B%~Zuu(bDxzts<J*}roFca<(7@>PxGH}ga
z`SIM?Nj#Zx9is8pUdEM08$Ru6L9;E+%#KX5P}ddQ;zmA=rH4{{YD>P#CYH#G=gJOW
zoUPtDSk{Y14}ky&OBsS$K_UM;huaMijp&g~WdZ6`{COko*f8pt7l{thC7EjW5#c2Q
zy;C1@EvQf8mG`U=)o%)GMPc#EljdO5m)t6hA))5x*NqYQTyR$?)KQ+gZ&2yp>NBLS
zX<41(5pcy6wzrKgDZYm(&+z_8x_9R}aFuuMyG^|j70PeD8}BE)Y-)co<3wYWE3X2=
z&_=YMia517l<JZ#f1VdfM&EF<r+8Y>BSL(kugI6`X|qnb6@tvl%s$wx)eSfK^{~!+
zdMh8|vJRQuetEqhThjbc-7+n~Q{i6{Qx1GdCEa{U8x31a3v^ZhsH*8YU!(XWC8vJt
zjcW|vDmcohab=U)*17;QfjDHp^gMOyFm>hnd3)tgvJ5L!uHVw?RbzB#x7WXBRJAZI
z+ISgnbDG^A4K_zA{+0p!Gwur7KQA0;ymao4F3Z0bKWkt$aICe)@-@?#?tYRwQo=y?
zm9T8)SbJ+LDZMHH1o*_}isU8|o)A*vYIMkD`hd;t8Y8~6jYPR@C3>lA&@t@goSsvA
zrWXI+9U9#BDL24<g7A*Q{i$GoUci7W?@Xye(Gy4f1aPeiQsjCe1q%ZJJK<Ck;`RKs
zV)gH^6g>d>RNmse+>!Z~jN+zIM85NfAXskaW)=%TgMwCz2h!kezbfD|C$+nnz;7eO
zy1Wiu4+;S%g?|dn+3Qy1^$*`36!gN87WDgT4-?IrZCUfB(Cu!SCgMZ^%!|OwvM0QV
zEd-zqaR;vwLhLD?{TySCzWOlyj?3l@Y;O$o2XN1q(wpOymi?mduR0zA<4ep)nN~m$
zhjo1|5gX_Cd{e=u+osx)Io3rhaubm<vRKg`^u=F}J$HAZklI`T;BZSv5Y?9X>h-F;
zU|5Y#GPC{qGWrQo^X*Js?QVp!L!W(;8`e+|W0K?m!6s-fPx0UvR6C6oczay<(w<dd
zB2LF&m=<j3;utN{xw@(5z{GU_4V^$|R++bt@S&{3MI5v5BT{AqUuRCq+^CY>i<e6(
z`c+vz)n&H^!o|>6#PtJ^y#VuJbksm9|7OVjXCU(g3<+|@^n|tlg|^AkdjYi-FTSkv
zB?yT*QesNdf*iXBLBsD-m(Z?a2!WK!{E&-vg3`Ae_+R70GU~7JZ&o|!Fyau$(57<j
zPt?G|?pv~wNWSvbOapbz-3tSZU;@8x>c8HUd4)^%mo0-T{aAgYQ8P<&^2y2zP+FNi
z^S(ND(?@Oat@1dH<T>fr&SYFD=Py9-^<4z8IyV${00G9g`@?`;V<jfA3H`KTo!`HC
z76cr;Sy*RWP*?P8k|~#^tav(m^efJfet!E3?$ihJJYu6VB^)?8|D5_cfB$lHlCr%S
z#LOdl<aT4Nyliz7K!hZDtc=~Qj5u58oypRyQbH!kZJAzc;=gq+KLX}>G<xT^4pyU2
z3h%B?H6I;P>_6L}Wi+8Mb0Bs`_%b9-z_c8Qf1hpfz25!;fc(n%<BR-(vu<1I6Ebk~
zX-TZ_Wkj67#LaI{dx^;EBFSV>h%sToT@#+KGSMZ<;=`fhNr${ZGLptUTOu-&G!B~t
z+2fHAkZ+Xls+W^y{U{R_?wGz}Rro#*kTOK2;{%q5MG1rE_nyj?WRR8J7)x-ZP~}@?
zbYok=h2J`9{ZL*&HQiJCpAlA2T8q=+o}MrKv)-=exITefG&AP#6Y!#L*&s&hG&ap-
zdVvSwUbgLiK~@W4OeD(BTxbmdlMlvz-N#bo&T5TKP9+p0t*K)iX3}W%^_<qEA`nw+
zZ2ON2Th22uC_9ARZ^^_4qtRgeY%Wahj<mAa^je*H9hhPKSU?9IiH@(sOn-~demMXo
zwA3)7YGkW`v9ExV)5fFCMT$sN!b9u_?vkbtjd@c+%4}C{_xq|FMirL{8D6Q_@tO1@
z`IorDfLRI)ch&JG_Z|l}7U*T)2Hy=TGdhVi4t+pxPwvRwbU5mzd+@5D<-AQxCS7Qo
zn`_GqOXAE_P-qWWCFsl4Z?7dA30NzA|A^91KwMWuo`2qis%^<^A|1A+(C>BWv&MY3
zJFyaO5eP7jROEkv{AF#&XVYuAG>tr3iPUNSSs-R@QyGjCsT>@Sj-MzmGkfy!n^{m+
z0UwyHvhCe^iWzqA&2%TFn)$A6v!VGEq)@#@VILwZ&_(LgS~+G}-$Uo553X#t_~wO(
zQB%)K)s=ot_!vB!YB|Dc(6yNTv`ZhLq%&iLktjkXk!9iYD_?XIv_-`BE{veeRvanp
z`>swo^(=;;gc#{~e=Rc*OS*?3pr6?+&M+TumI-SSy<0Tm0W38Yk`Mbb<9B<<x)D89
z9DbIXNH%-BoU{fh#bpul<FwEjy{8GgPMC!sp>}SM%Wh;$f@F_}h|fxhpGw##!iLnR
z8`5@N5oVyYkVmW|lCu9c>Hdck#tV62!b|3_PVLHNgZ0>&=X-_GXu!%8GVK198Erwa
z4DXZ7>QmJ67B_hJSR>Hu`QBW=U9!l6SQg-TqYJLJI&I&?*8vrc)O%T1;qyLhB!jFo
zoc^FD5b?Q!q%=EF>E$<~_0g^f5J3QOCajgu8LMY+G*Ql(Pf@zD-K$K<Uhlr{3gUAT
z)DA(=4Fcbas0?<lfcUYGXT7GT%OeV0ma$l1&l8E-kcUQ77?baLU74c_z$DefxNI~?
zStem5K_kF};t>MCJ!W9OtAwrjJi3^dgS`V=+y)(;tAx(eP64Z5z`dM4bp+G~Uu5c=
zY^Ae^RC9Zjyj(*~FVyG@6*m=_+Ls0TaVCJFHCMgo?D9}HroSLK((Qx@o!=aQhMc<-
zQx&is+|rM^{?%K>_xd^q@A`@DJ-zz69c?KX<M5qcnn>$M5^L>i?IOF3<Vke`a9D8d
zUNAj_D3V#4;6HA1a4UvcAKp;MpQ3ozmy9we7#~NXG8C~p*|Slf-O_}C{I91USZi-X
zFk^ae<}cC@hrr${VWjPO+lJ%4!%u0CS8t3b@BXY9mGvMnL{n+?UhmaRV*vAk9~n;)
zI~ISc{W0tMr=e)$B{1J^QZu{3^FO8XZ7wfp=+AfV&HQIzq4uq%m+bXz5u?c(!mwLJ
ztm`Y9BqTn!iLkw**ORlk52E5j>WHQw5orN90NImScy$5%&+WYg;m=Y&0iNe@-YZiN
z79(?PIzeVvEP82R-Mm-HRP!#A^EadF(tUq+x}Uy8--?sB3y1@L>g|;d00y^-vJp`z
z)Jwo%HRfR6$*mgh<YOI>4ijNsxj`Ue0H=`%4~(4x05d1!%d$$qVZH0a9f0F^>1tuf
zGvSqOPPICFma%gi`!R$k_G)PJ;b!Wljf+}jV1%PkIu~y-MW1|F+Mo`bB5QpGe5Oex
z;LC${e2=)vTZ5eiFs5nEc*`cL7aVKc%T3g}&_UBimukJ^mD8?D-YA!vgnwCiOsXvf
z^JTwnnX5f<3Q-3)%|6zluJC(l6=+YmJEY^FuIC(2Qjp|m0LAIOU*BFXF3;1T`GF<}
zws3IgEj;TUaDI`z|Jq-2iP#cG3k`v}ReN`tXX?I8fm$Xe8;rFwZn=R0YcFU`e}b_x
z1MV-!gO*gD6<Rv(V2TaNx!zZ@+EcyRMd@A)5bdryL1Qi{hUglY&dx`l4BC<P$#jYV
z5I)r%&O3M24p`>h3H+WVR$y$?*h`ZHYcuS=GvoZZzlWY&zMS``d+*$qzf-lH?#(3q
zCf7tR3cq=~mF<-&=b<q><mZ0-O-V7_Fb5<zDy<`<V`K68<()Txx605Z{!{JcT9*Dc
z97du>AOmM!lqM+#m6R``VDc@&r0=tWcJ80KTuxQODDb8>^0`v#rYAwU%EWk-bvykN
z{palvcaXr=Xpd)jqbuLYa}DrM$NDs1)0LwS$1~=&URuD|rXnum-0qSuUpUzAFav&A
zA-6sN^{uQ}7CpH6&yK$HkCMcIzk)tmK{AOW*(gP<;dc1^kolKyRJ$4?|8G0q7*Tt@
z*=g;Qzi)c$IM22|sdr@7-e&qLK%E<l3cF2R8cVF!Uhyp!a9fF=wKq7=8C|pAc&r}k
zCO!_v)7Q2ZZ*_<^#FFnOkaIQdAbj|!aR;H~1w?kB^Hu}0+6r6KfKC}euZv+<Qz`6i
z-;@Y^73}c6JBL>`H|F<dXJfi4A-JeLQ?R3_KHnTUlJ+Lib?K9$?OgfMM9IOc)DUCj
zyU!)eZD>qqBT{2TJ|=>lNg-^@&0IMFkX=|j5dF8PCF6({+eNPrfg`{iVQE>KL>gO`
zd27qd+CJ}5?l#VPa~1L9_J;E{y$w2Zc;f^)FKa4#;)37NW<thpbCRBoyZ7~MH_Qfo
ziEoB%auQG*#iCkoM`I&6wPvPyM?kJY`+1(Z*Fl0^nPdfDaw;(+3EJ=dzA@U?Q_M-G
z=GwM81BL+aUO{YHpvk+<6eZQrJ{-%}u9K;k?t!|9Zch-<zO4Y-;^$VJQm=F07||S%
z-^H#fc3m<4%Hl^d6}j(<rRQpOh%O)47{-;nP_qAET3N9AQ<P0sw9#>dE(?n5n3EhW
zoS#d4+-n0uFXDnt_QN?+Uv;I}cIBU_6=j_v99xRA9TscjO|?gi<4eN%F-VN)5;y<r
z4lB?QLqt|!72#x0{UMfEHLW9E#B|v-Cs(Gm?z|6VJcj{7aLU<7q29aEj$JZUyMX;%
z?`AqsEVlYSI6kf8Bfai>hK-o%EKI*j687US9g+SkcW0;6qvY=CiA?ATu!^nNTl>#R
zem5PMm}6{c7sd7a);m^HD`OZTusNO7Ofp7))9;2EM<dXGQ%!$Rtd_k{Fp79im+!`{
zC)Vt$%T-boT>8dUb^$zegV5;SUPAB-_Xl&)%#a<k9~}(^bFX(7<W-RA{_dVAySvF5
zV^IF0>9SzX;Fw~P$dC@cMIYZGJ_5fW3_Zgd@=8r$Ay{QCiw7Ydf})%@eGhEUUep2q
za>hdWqvi_J<WJ)wns%e`ic<<{dY*B+@X81eOGAX+o5FL?MPf|lIRtRKcn7;RY7&5W
z>*4SNRM2W^_|Flkq!MZ%6ur9w+#k(E$aTvtu$26Vt{{NQ0!zW)+FsEqdzK#SU%F&U
zS(ZA!XtXmSksOHbJre>H8P-oC^NTS*t5Pd$`3&Amxn(eD{6i7)2TKKfYatN;dllb*
z#{CPy;Z423%lsw<&z2ak5$GX{Wnfc!IY!~XGGU|=7rnpERx{r*k|4+hdtRqs`1zVk
zuNkGS1Be52Xkc=>AT852r=01-2+#(bVguP7O9bLz1n}2wHKHna@pWC8+HM>ezzzU;
z<rcoLl&>A_*1-%DGJt?$9FEJZw_EcX2Td?b;R7$Qq`oDe9N+B@=t%tacb=py>bY4S
z%*K%S;CSy#otM@Nuo?McND{=VWN)n?rOVF}%Gy_(4O8R>tT{!q*O!|pEWUzyr=Q!L
z=jvf!8ywwy)M_GIj{R_}p7Vg-^3`;RKxabUS2aeYV$LWS(kpJj3&ouy@M>??(~kIi
z@)Agdp*mVK>o4WuO_Ae!5t+frSdP~}U6P_N1C$(JUdp`3PqK!uL|Ch=!8E|}dDBl6
z9>2o&=mDNlgFH{goh8#El&Vtl@eI4vOz&#BM3@Pj8CTH<bzD+ZW+0JJ$Mm}WgHJ@9
zFY5g5WXuSM+>%f#gS34VAMRMHr(kB>(J1IG=k!ieE9zxU*+-1^0x~(I_E@g+>r!>i
z3VtrUTVNmX^}Z9&2h$=fydBghX;|iU+1#042S4XfenR9-GMd?0TKX~BV}N3QdC0)s
zC<VAuL;?1rw~aQ|bH5HWysa<U(*>-)*6_u8EJvgge0Z1HinMTbY%A4DcBYkoO6U-W
z=<vx;cas{Uk_ssuxzTh|cINl8`|t2@C6c}s90^5DHy<r)sj(YAbd%}6FZ$6cBZ>8v
zy>%1(&f2y@N8yn2ctCj_Gcl>zJjHe|A@ll(<hrbO;KHq<y?}e)i^=(J4}kZior35s
zQ>#z$1G-fPD+VTpsAQp$FWJHmb-Xc@lB>YOGLVDTYxf<0pg8g+NRs48M&r6<B8}<(
zLL2{t9shn&aT#>&P8!1Me{}8VaR~gcUVEOKh+|NmGvPqkb&EupoDxPuL5NpLBn4|^
zb~ffMOl0=khq2n`&<o~p(Og3k6x%a*I(qjX>@yz#G3v7Gh1H+U4l{Aa$CEf0LU|48
z+|%jEEi3S|y%js}aa~=X54#2L)WbEaMO`wREX^k@)SHUcn`aIXC&GLx-?FVIS#R;m
z_hD$GpLkq&v$f2b%oe6u#(Uqz;CqxlpNe}!sn2R;KRt|BWgO#cnN}Y+%~nz0l22n@
zwj6w9S20bSy>^!*o*?^X%!4vi;mu6PdD7Ax4t%agi9^?K>IcHv68G;vQK>3sOzPWE
za*WEEMHjzNCEI;B+)gi7wD~pdTg$4tl6n!_L5sszo#?@vj`HZf7M;%Y<iaYtV2R#t
z#5eCgP^GMr;9B8t04|QC^P`(Z)CVKNF}>a<C8XvtOh4Zx&w@@rSK3B)C_ibK-T7x`
z%6mas(}A(s!oX>f@2yGp_GMkRl4XLYXA!Qa>H~J7?5iW>!;RR3<&UEirR}!|wqClW
zvEbQ7EYmCtoy`{B66><73b79cUQ)tIx^Di$%L~kozN8WJMjVB5E7mcxE8k-BIUB$H
z*M=&#2lyTdrfGX_{&w2=a~-4wh_k~dI9gx76(+%7A8d^JPV^3@OqrR+;5W?qkRtXv
z{t^sVDbi=@`k#N~A1_#6gZ+^;6@(f5Z@P$=9>fnJ{&;nbKEr4fn7SI9+hpji&NbLH
zrlrw`X+R|4{<rF-|BcORa0pm(HitF?y8rf5uGgyj-Y(}<RYE1}E+pl8^bsOO>OiEA
zVc;N&xn=b#Om(?{PjUj}&3p$m`C8Ds=z{!^hQ|BZQ6X2~RRc!SbI8M8q31YZ{J0T{
zurS>NtlV0k3x&Y>vNOvx>TlYOe>fxnwGQa+apnB~ZFHX@hLeF31qXm72tDE3xydOB
zlZpl9%_@laC;+@*p~4;O#%rD-(8kCQ?!S%dxx!@{^%8eL$5;iU^IRx~B_d~J4me6V
z`n`Dc-+4GV6dcgwtRVgGP#tS%HJ~tBo`J-tr$2Z8MFk$n*S`ndu0g@2YT_k`paw%y
zZ-I<mdmpFjDdcYhvj00h{qrm5z;7JI@E_@SfBxJ*UI3s7kn=bYB5b<<tr7gUck}Q6
z3;uD^|NkHVui)JOedqu2ZI`9<f<*Fa4-0{ZY)E1H%?j42Y9Qybg6Qf{%reAFeAE5t
z#uMnF=XE4{<KyUbE!}V>D$DK<(ElH{16~biFre{J?clHH#Y<J(;CO#y7eqM?gOtBd
zs^L%B@)*3P!^FlI-L`u&q@WBjE1(gw#0Y%j?*y$el&`eviAR<Tm?x;Gqy6q1!TR-X
z6opiE5Wf%%do65Z`oWIS$m|91;UVa8Zyo{+U_*#IM?1`F_7Urda{W;zlfeonZP+&b
zuggsq7vK(ZlAIi*fBrtCxDc~fbbrFT@Rx@6-CZ-26__})?{89eVn&?g`<<RJ_r5vs
zD4V&C=Q)3eg0uiDjOk9kBM=l11G^Iw2q!ikSGnmBO^6$F;W7}-gBWWdof=dr%3j^?
z`=l;~-UmT^+*8MZ$?*b8@d3g)duTTpaAcC{yXPgm`gyNxsU7T7K=NeBp7`pmC`f&R
zhh```gI*QMV%yjAlKM7(1!goN+Rw=K@QJ2=5z|v)&Hwylr7(dlcUMY&{!dW}$vrQ(
z-JvFcuH%$}94r45z0#Tj_I0dN)$s51H}HuanK^F#POYlktLOqUYuq~fM9I6jKyvZA
z;&`(y<USBM3k`#<R1(%ZIArn>yBRBJn+nZkHsp@@zSRP@Bo7L3cvi*E4wOeW!A6_G
z;J!b(3s|;bp?tge!6vzM^)u{;hD-{#7sUGB1vPyO6ce1bdQiu7Y5*8iPQ&(@d!3(G
z9feMMzl?Hc7+*{O5S!%Q|7&2?Kf06hZ?*bgi|+Rq7yQA|bkhB^<j+sGPRZbCwWtX2
zA_v_8Xh(Kb@ygpwc|x=5@G(&)6oaMoJwXg~ij5~Q3(L~QjNX;X+pQsEyM;wVVCvWR
z9hQsTbOt$ClHmeSN+n>?^e_6IfEcmQZgaPjBVabd@5DzP?<VGd7!H-h1^V6(tAJ_G
z`jkkk5o2AZTj_qSpbV@@6GX+!FFC_EITQcO^thV`EX*UejIBS<KEHRK1akhAYa~}o
z@FnZCS57WIY&3TP?_;?m|DzS-U)g@GMWM<bxVKZmp{dR^%BiC7^eQG<?SCKqf4&*u
zf;A30yh~n1@&D`aUWlybao->|c5MYgq%nLuOU1U+mcaNsPmzD99wIjngEPYvM8X+1
zX`LMoRAIm(wu1QGD+5)hKLA(xB$*o$<AWH5K`>l425VLs*w=$GDWs919|o4|s}4CR
z-Sp`X{*3t`k`4p$*&!7h-~%KFy^3O;)Qi!v3;Y<h0un@u0Itvo1#P`k4qgSr>UQXm
zE^ONDTzm(abE0P=<W@l5;)jzNcCiac=;^lvI}M1Z`*QVQAKcwb$TA(CZca_mA;*{f
znP2sdMYu!8ayb|&rde+7<N&V#D=4{gpnUGGi{`M&<FeX}B=<wJBM_Kinfcx>$S?t9
zyrG_A?}F2u)_YUxBjM3|K*M*Ys{cL-9t#0&(9u~1vPI{DOuW{yjZmk8dR+&(h#Fuh
z%dq2pr~OW^LSYR7-5_wPAIchl1W6D9mLEtwFaTs?WITKhG(g3oB#@j*_NpZ~?^Sc`
zF}dHNJA*dNdyclFz&$34oV;WO_)xxZ(E)I;?A0rHJLbP0YjR`FpPT`Pv?0I_byoDj
zw~Rk89IaK20~}N%t%9W!8Ysusx`HUUT(C)&vV|($BcQ~+MJYVH1!^X_VSmj!;KX~U
zTr>Me|I_HUyk58E!7~sQmS(73(4^pU(f?!g(^f|S<>k685p_i=8HS@pfCYjhNU`vX
zQNhYFG|YK`ea}ZzYl(hLq+}29rpI>s9Wayvv3ds_FS3`mjKk4adR6=tuT0ZDOPp_l
z#)@6tU~}Zkm`nsJfYfiZLnWQU%hK9fp=Dqk)%XBSt@}z<q|0L~VSEsDdDL#Sn1sg~
zPZ~0O0kD56&Zk^aINV@CVCJ;!?Dtqvs<JXRTH^4t5aVGZ7|w>R0I5o-<HdU)E4v3y
zfH$Q9Q0mJMyX%g+>Hbvzku)y?w6=d=|NrY6oO0k<a%@!-0vZvpmBiLufFijZ!4O~=
zgwhxgB|rgNv8kINKJ5p@(}&wh;ycul3KX#IfCV$+2>cIVO7NGGeLlQF64nG0P=apk
zp?z8izj?_i*avPC1YG5k6R$sH0!0}N#v*<#6q=US--QakivNeRw+yRt?Yck(Q9-3a
z6r>vkMFHs&1XP*@C|yb|LZrL9q(MTun?<LDbazU3cb@xk$G7*l-~E2)oa@>@xVDSM
z^V~V-m}86y{QQi4Tv>=gt&?2n6fm+$S5!f`8r{`Ogc=V?da6bXzs8-=(E#8$x8UMv
znh%8usA@6Dasg$y{|6~WHK6-Wk1^+b3H%z6$re19Q7D8`4z^}98OIJ;oGmCpG|+&i
zUx-|3g_0~t145qLaJh5O!z&{;694BQKt!iu^ZS5g($9ObA~}9kdu9Lf2h_j29n0bp
zu(Q2vs}>lIy>~zOC=1z`yF;~<#&XbOi8_(j@Rr8}QT;l&uMNNr7!P$SB(GW+0S(9S
zWMN-00>~Te0X&gpe+>0Sdz4r8iP9T2082PXr0o_LxPm?swRpZ?fl0}HwC=e{=GB2@
zN(%oDI_(8pmORGhq|p5^W@`0?WULhgC_HU<bBh%QB*!|zfpY+iHY!H}AK+=SXlTp<
zpi9kL05XXO>M3;Hc**Gzi4h%;7PNu^(*sAz2PGu)H#cuh)L&o$F$b5P3)%`lo_Dnl
z_K9W9UaY9(2^(zJS(jLYL6C5D031K0+*k>e$Eh7y<N5Os>Id2vq#{(rx9srtU<)VG
z)TA8fFgQhvtd7_jBy<>o4oqDGY-KIq5$Xdi@KS*m{!2m}0~!h(5^=->n7sOROxW2C
z$KJGYUTk5UZ<Uz8%3dGtCH-;I+O9gjVaMApFgD2<rvd8^%F7!|`?0hA_1o4fB5^Kv
zmF;t(w-Tfe$zN<xlg{G3jJ2QST(xF~zyBZCmVaD=_g|u%-}oQ<g*rOgnd=L!O^&U%
zNJH*~eZf>to4(kzY`tOCdaJP?#7@D=0}&HNYU5uaW0e(15Af-m^lm+{bJ=LUVZGDB
zOt3|L6c?u45e~fbQ^VPn_ap3_o?8>pcY>I#m1mynXf{N7BG~RJUpYbOKF|<wbOOe_
z9`HjnDv#T*e331>2AY5s%wNwckQi%sgi2=bbGx}xEqOn2PMAsMpZ*Zx{Q}OFxOvU!
zc~y+0w?CGElsa6w2dir0x>ikgL3T7IMW#Nkf1{kTsm8lc^+HJx?XU6dZfKT}Yvt-I
zm^3=1<Bp`>sUcjU8^b?X0Q$L&ND=~aYJw<gg3A=B_+5U~p0BrxdQ=9oJ5Z|03039}
zo8WraVs(?_nk$iU6?2~?F)wIs{pF5KVt{PCiqn0>d>jj<IFtGtCX2V5#CfKb$~V7W
z_XM32lT7{a2`*XNe(<o;5OA^)lj?sLZfI@#^fLBd_9`8w5z3k7g9eRsma-A1`+#<x
zn8dA@2fC&9?*o2tnRM!a!clG{R7;zu=$((_^6Vg@EmP0?IKFreqGH~*&5-zW?YYb1
zh~n{)`zNMbAHCTMV(f7g!uUBSd<{N*Q~6~2saRoqfn!EB?~AROzg=&0=^fbaPu1dv
z@JU(tCuUefc)sbTc#IgQvJgo(DFKsvY<Rzv0#E$Z$&ZuF)AMw2M0osmM0`fFEFOiH
zLXu<98Sa22K~5^T0OooxHHsFbJXNr1*+d;Z*bcs$b)i)z8w<ZL%ey0+w6)N%noc<|
zUGb9m!r4M^wPcb>l}ofzspWaLKH=^G><o{CgkYGKz=8Ivhdj=}q@HPcx;HJ!afxEr
z&PEG_su8Gb=}Z1@dcyx6uByMqo-wfaGxv3tmbli7iX0D43)*teFX0nUkC#=KzSv6W
z1J9lK*DNS@yu@0c@-!Hi$nZWrc~8z_2V?*-W8P&GS&GT^)ne2~WRdq9pLhWI61^Mi
zre1zorOssaOX~Q5A|1ykT(1Oqv8p>k4Bndd_-+9|7ZT=zDJWK<<XH&T0W#(wjm=M<
z9~>7xDXaO9+T2b}!C{K=Wq!UN?n%~OtwM7Y8i5sxf887dSar|88F?X?%js(C>a$Pz
zczdOqNIKqbx3wUi+B_7POI+6)QJ-tP$|ZXZr$u=jX9?VcFk-NtPN(Tx&s+JlI|ws^
z2S0PUp{XMDaKYyFnfuHSo&onB&x3D*e25KM56>S#hZs9jCzqj=iA$7$+|N)xp%!~S
zI>!4!j%Q4(DTv}5bT2qg^4r80w~Ac>!hCU*IbotDk7SL$WHHmP6(jU=1zEvgFn6WE
z8TOGEiv&{>H@59#&;|D+?1QlD2iLLhvCZ}GhGV`!WQ~Q{BEQ|Ve#-lKJHwi2r%;e^
z0#^^!0=w@^xJ6Yo6QOz%xREQz563^qdne_7R19O8H&|PiEjy}p<LO}wyWU{C9)M|9
z;m%+`S+Qzey1x?n@2P9uGSCptEHiSYDF41D2vM*`8T0FqfA($xwT<Xqq44X1DHd^N
zSeHM|Z%eA*HL|#k3hM$?3M{mr-_joL&a46t4<$F5lL?S-C4)K&t1&_agOi<8X7^|R
zbI<gf!GfhhJm}=-bRHGE(oHlzD;c)cqy>0Kx!-d$QQbtZ$E88KY^)Cjlk{ZCP|p_I
zL$YD|xZTUgiQ6lJF&HX<O+n5}KOQ9Mc)wvPHN^g<VBHnA0IJv0cMJIJRo82Bm%WU*
zejv<Lo|v?#CLQMmjb!_XrjMJqX$IK}uQvDfXhI~T#(QH7N)2pzEZXSPsCJk8MLZ*f
zQ6!q2u)tNlsqZE3+p^krtKmWXDQmA$$sJS`?9`t(vSia`x<y{|X7ci_wu#`!EX(o2
zjwrF3xZ)%yi5uR<g?VjlTTOteebiX%buEKnqtWXm&g*+iASvHt!Mcjm>l`^v3;S5e
z^7bD1{qe(Rog&2cK6b|8Ud=mL-XeA9!5u&vIDPm)xyq=ghRJjQ*6q9U-&jC(BjitU
z-|Y}A{}0lf6nQOh57Nol)s0PXMV%CDFZPE<OD-5NQ7|9ne+1r(&cEoUf+?wAr$8Ej
z0KQ1$P_HLml@I*|yUmb|4}r3YGci=li_470s8poSy@KL}O`WV2#i?z2(z*Af?46bG
z!vR~N`lQ%#kpxvKz<it;%&#gzXOr;Gl3-GRm_>pA$&6{J^iI{US>~jv@VL$!8+Y>A
zXYu;tB84xxR`&N?yHz#YC9Y|`P3#^!!V&VVcii<i@)!FB2`x^-^6XNq(sg1hmRzv*
z=yf`I%1R!Re!g`$AIH|2Mix+J#tmt%)^X#H-OAHh`XY~H`F^gGEbQhgP5}-1qO4-H
zB)TE5xk-OPRqq?(&ZWf^$x(f-jpC5l&wXOn3d5i-xr}F8tGAue)9K#bJ5*MeEn94g
z0bhY?LcA%KEqm-ef$%&~o>|n?Fx&EHgnv4r2`!S;)7vEOR2Eg?Rkj|#A-^bSvPVDD
znI;&p<SEH&?EARTtDL<XU^zKf0eIsXcKsE*&cu?joP%~gix17)`>J!B_T~Ty4$jdW
zz+k~DvnzdYBCWuiIU+RP@W3SPj#MYlWt1c)@!=N9NMcQy*AU3GOhNG6-jQbaBEF(}
zgE(^piV%%A#YQU*^mfIe0>AFwBxORmNJ~}aZecDs3EcMo*GbSrPy_U-9eh5_1lL~G
z#vEIRKL+B{MbfEA-?C2OtP=!dyTDn6kemXKu~?|Ze?j*A&>*<jTT_nacQ~+4Fn?C|
z3nU=@c=z)&vQC6;5AzX&7Svz3yoKQmg#JBl?bSY`kTbdKsuzF7jYb__;`d&f80<4n
zsj_SLm8ly9Zi#>`HN#~zQHBL)MOrmY%C$4$a+7c!5egI|UM_ZMl$l7tsLZ6mav^o0
z>jN>dqs<TQJkV3yL-7XE8<lTN*zRslSC6pWoV?gTI8AzC{Lz!u%?YB1LN0V`Ce2sV
zlYwx`o>aI+x0}k`e1feNdQ2kR5eOrn&Y!q`ol)&vkElIQ(V&D>r7x9zq@q|T$gW0O
zct31X=L#1qf*C?Zd~(Pd*bDv=nAsgzv5V9%0H=r$8BCGHg@b%jpSuM}bIl=KyQ8|+
zefc)|J%qQYMVXtE-M&-Z3CL8?$qh@-%hFz(1-A5VWy*pi(lOzy0~*q;GhKIHXUPXp
zlQzZuEZ}ZBh&6k*ztYQhn6cPdeH}9kFlTL45qIGK#RS980h*V~XL1<7&r5>$wM6Xr
z=C17uPtRuvTFE#=UX^sls_h0gy*?_fYwVto3>&M7Y<HVpYHu_54L3%(u&X&LqHHBW
zunt^HD9=%4Qr|}Pg=wU@nlB{XcdTWTu0m`cKyv{SL@*bLOw}L@DFA88p2vt*dOnaJ
zd6ZT|bmNz?V|%7(NS?0-^U~efy3@X(|HKe1l|bsfrDFqB2yS%g{nqPSEr^|SsN?S_
z$VI0Z+fucPr*#5z#cAax>dvU)H@eX;v2f-#Rv{|z&!~0a?x5+rR8q0aT<z{eNlgli
zCpBN7jOim|3(m_a6cVil-U4nB8jVg?I!#10s`Mvh&Yyphhg5JV3-X_eTfJej)r3dn
zfw<ziTZ1{ULIfA4gkvrILrrekI%_~hoJN!F(t&D!P&bJ9`SAk8zXH3gpL(_d9E+%x
z$?)d>=kPSiNdpLh^X#NlpFH3$>isV)Ds?}sZGx4*oXKD7z-{yu+$$9U<t9%YR(=Wd
zyyAc{f-!6C5y<d(PFQSPg`HY&F~<ZP6zu>dGSwCdb^NQmxnB@w^0zrO2X!ENf5RJQ
zS}Hq6LPURJQm^x=Uhvb-yYbH$B}KV*as}Ss`V@irSIS+koeXDNAHXa9RIf16pLTO|
zC2k=^nI#-@2ooCpe1|tjZ{!FPog1Yc*mt<!33ebLWl^yOsbPylJYkvM6=0{kvjAj`
zZ9{7g@;csAu^amIcx#)!)-yZlWTNWx6O~pTx5ZBK2v~d4B>JhitL^NY=bbFFF0eFy
z+hvk^mng}Dk(J`+O9#U<h{y>^Ek6ai`LCt5_(-ZnfvhZk9=Q%tYAIm-iCKL&qa;KU
zcjA+T$_Z$t#n=aOzG)ms(_2-_eu@)=#i~uYuYS98e~7?hbQwaCUd+tM3eX-kM@)Rz
zA6K;U=vgu$S0A5S&(>7nEwTAz3e;NbT+Q7`M`d3pUMeLwxjU(JZyc$sfGGKBqqH*{
zM?|GNpF!`(u>6W_6>ykutuCZF!uYzZV0oJx^Fk|1!c*%i!Cr+aH!03i{42(vNbNt5
zMn8RIJL$YXnb}X<(8`9R=les<@psET3-71DB9q4eFsIdBqtk(ig_q*vcYjoVOuv6@
z#5qF*yQ$!7`b!kbQ}wwMsCQ;hMSrwF%E)=yM(U&W&<O)+W^3`a%}zREfY7({RI+S*
zObW1*EAb8&8Z1r%K~pJoiDuU1+c)OBqPW?61yy)AW1(gP(SK|8Et>ekPeJ3Kt5TV;
zg+Ew>-nNKpf`$elLopt4cLEcqGJA$!qD@Y=upk>mb96p}$GBj<U4P3{)F!X`@HOhJ
z$L)&|gk){ZV!Y$!ylg!UipV5#)h1{gZ<Kb|?yl)d%+!8ruHdIJ=J;?{K#3$83=3p%
z$@!Z4=fx(%n>i-L6v)sH3`;1>mwpCxO9?50RXm=58);<=5~c)$+qNu8%L0pWH~2<k
zV}-lZL@^Z1HXi-;3fnU?2?~^_Q)?ToPF`GoAI{5Y@*W^Uo_NUeD{Dj2-AXvQ#D9a3
z{+y3rKWKRXbX6;rE)4q@y6XHO>RRHxswFZ-dfQyV85dfXc)=(BL@$`cV>wH13<B?x
zB`h3@IuPAl;JEl`);;R9tUa6D2gnQiP4*|{^Iv9v?ziw=GBY2yUZd9(oPBhwi(|X)
zdOu_=(Ma}zcFR?{n**w7+JHPdI<dZe3iMq3Is|;0J_G4EE3uXq)%^Fg{<om<2P$)d
zBs`J?An~c6Db&k=r+Pos4;;(?Cw}U~%thB@Fn%23u7+oL*3v_vmY3f{&@Eij(y9Wf
z52Z28NAW3`gcRN|mdG_2s7WVy!zocmfbh7Vb1hp|LeP+MOekOMLcngezlp8udjPO<
z_>+`Lp8BYHn%%Sxe=~ON-W0d5Dq4{Ue^nd)mbZFuS|$87ZE4FIQR0)vk5<-{L6;Ds
zvJ&(XLu{{$%h;;*<$z#RaaPhmt7dsYK%~)9tfkSY+FOH(_1de<a1mm?1P4Pvg7}G!
zx#f-eQNt<H`0C<F@=;E)462h|=I;J_QTM$N(lr>IC^%S>_`%dmy*s&pb$hS-WATEQ
z(-cU7MeL2lgwVNkhcg1LSSkU`_8uj;xExj9s^)lqgwYZp5QZp3fR^b`fu2nOk0RQ=
ztHo#Got6Vc{RPy->*<;bzuge9J*1XQHi3t*>$3)@NZe0+W;v_yXD=doog@h!BG&rz
zJ?!zuvntkaKM9#wBHpqR;RJOb+ymZQagw~$CW2BeMXolDdnpsftxK2{*-3m-A~77g
z56yYM-xPoOL9IkmfT##m=aQ)-l%Gw4m=oQZ9X<q_uiop%5^^er?Gxyha+WB-PAkj@
z_Bw?Ai;~B=Kt!FUaQ;coUqt52NCX9ul9>9uqHx~HloOVKMc3iQ22wL-@ACX^<sTZG
zO^pD^kljN%44<f&`9ASHyqvws?)CG%`nP6V>Wi-1i;3iW^~!to5>Rm8*+joM8|CGC
z1_8{bPY6UuAr5JavYh6oFhiF$AeKU;INvfwQa*>OlWP?lz$t}sWv1-Taz`~nh?{qY
za0S4kjs}|??Gv&JjccyEh>mCN#mam*yFafVE%xJ&L8Tir)5!9=HoNT+Whrg2!g2gQ
zSJpKvE8c5qfAD~Xlp_wbC2C<^?h}SttY)OKsYski2Qa-wSL(cFfBTggbLh!_QbCqc
z<)@v@`f*{R!^6^A9>X8PH>6+KCGZ;BMyOaY$JcPuX6UV4Kegx+qmA<2p}<(ZwIe0L
z#N@SR|6axHXJk2Q-pUvQQR9;A38A^~=`y(cXlKv6ENislj~i(X!4Vihy+ZjAN^Ov%
z(?2rRpcDCCzYJun<53(1Qc0tM93UGTo0&m|QC=}6ppB>4wV8(=Hdd3ct-oklA;IcQ
zBfw+=^ss{=rEUP1XV2y~pC&9wD-haMEafAu(&?E7sx1lz=!!rcu#|vHT1T}rU>d^j
z!SyJAUa4egj&m^WKGOqz;hv1~jCCnLQO<{tP{X}__dA+m7X{bt^x$uH#m_cKbUKA4
zgB8X;-tRf&16)aIwelkX!`)+3tFaEVQ}^XIOMl4?_~OM^OvgqOi=c>5>Wxd=T+|_)
z)*D_iTwx`_Dsj5imSEAe@cUNSMe^h{bg{}riNoDzRW`)mxV?dYC}7*P*Po{UVvNl&
zf^D+;7F&se`DmE$oecorvk@e`Qj)CoaX{Bg2#}HIey$v;!|QG`l4^^>MxZwi8WE`6
z5a=&~on9Y#IcQP&`k^_`_snFZNb>`F;D4gOM?d!+z-SN5f^|o3vb`ZaF^fO@dI7XJ
z`H&gLI?_1s(t>TQ$>%JI-%LgSJ7wjsAgVL7Fyt-<``YIB3$y@+o^{ndk&-He4Ve&|
z1dEaUQj2u<3uKhENYW{&-7A*dSc-&NY^N`H29E6eiCAuA%H905*LS>vuu7P><F32z
z@eb;G(e3gI{elpg$t2W+9%_)eG(4Pel}LUB#4yS6^+O<5dnEMQ%kQ3|x--_&<I;&w
zrofJZci|~WbK`s=N{Iy^8n=)!FZ@*1q#%V#XXUpz%$}Gps%o{p^V$&D3#QEaPV-6G
zm8o?P$lG<i=sj-h#AJ-;-e);lfxZ0f=JdVV&wUx~CpM5w;e?9ZD+}+BX}_KeGP;j1
z$%HTXr9A7xZCrAO9+XXkz_s7&3os+!X)f?LSp!|Vy$(RDtojLGm25t6_efignB-ZK
zD|=tR#4JD_CaVy#bC2u=blJ=4v#BG(MBY{6Z^|O{B1duemm1Mto||>MA^_{9l*gcF
zH~kU%J5(=JbZja|FETY?SfyL4TZt7*J`X3z9I;^LUB?Qhb>VNML@mW`qSzD)5*Z~6
z5O!h#wX=Iqa&%(3`(KrNHs%`$NzVHp-U+^QWNvIvUh$z?_C5>gfY|1{yE>KiG?RB8
z#;wn>xUqsd0lYY^X-$!IAommZZ)?+a@jAH6gg$UCf4M~7=md75Njaa4Q+qMLVdeus
zUuYcCF0;fvX*$f5s1%$C>)!=wgD`Q*WLj6x;ujL3%jB>+v$#u4&9MqWhs*tOzPU^D
z;wac``q~3|YIKwIEJ=Y26UlCJvZV(dhZL9JZe3jCwbBq%asVYq({Zf37W#Ra`{v%2
zuX&bi{=_-|x(0vy*||Urti*zMxoy8S{;IqC%E?)u8)?_GrzMTR(CU_uy8CgUz`iQ7
zYsyt%b&*qP68qH=(FB?l4w<OWP5RAyF8GgBi}`~Rn|yCrMk6=*q?Do1$&=w(K0!Gl
zM+>_72roj=xBKd%tJbXI<5ry67RN{9&Fa?Y@sbs`@*hCm1X%W*Rc}3F_TznOLs(2l
z6M9YfYF|UC`v-waUSM&?a&#3L!(c?Nbj*Hzm3;715oybTn`pp3;zE|F7@qgp(?M5x
z0P{d)L+m0FB9veGC@!82zU7k>iDC;jmKAY)Qx%j*im;Rn?hJ)1Xi_-)FM6)S1mBH*
zF{}*Uk&0(Y&6G6%5l$_!`ojv*Lhh*A9n?dH&+`P$uxM$(ljVu_^VKAu#W%dj>8rBE
z<t(*>4hL%HAGa8zlUhrY4+q9GV>657^(?sCbYIG&M#AF_5tMTynee6fD8ttlDHIlj
zvhe=04_owRYTO5pN8QKBi9``O9}PB^MTu~5556T`1fi!I$i^2UvbCbKc5h_-2y9a|
z2>Y<XFgP50+S_*8`<4&Kwa}1aaClw*Fw#)Q;q9J)1HJcpBW+lA?dsjiz9JS8?Yv!s
zLV=Q(Tg>m&l#9bt*bo))L;Hlg@uP;s@7Puy9~meX!9^$9ZPI#3OOq|pHfYsyj#NkW
zEQsOVYOT52bDcG7h-cln1PCmrYba#bkWl%rA^+{ix?|yM<TJ;tar-y__7A$A{EX!4
zT&1R<#mc|@>YraY8GcQ$&WMYJ<?i1;yL=ZJ6^YB0!Cdf<znN)v&FT*?fPX&9Z{PQs
z8p+K`MJ)c~%RheW*C*%PoMhO4dCY-+MRkBHQxe%_3O1TwcO1}>8y;`}gv&$+hIs5?
z{o@t?e9k{!B$WRexjGtF2J+t@cM3%i)nccK;1^e6Okrg(OJU-0V|*BLoQu&-^qw`o
zy!E%=_>bel;|~TSj!m<e;BTMR^{E5?ntH$;z&ys5HcA_3ZL-obX71sC{Ej~!^w(tZ
zOOiWb<)``KZMgm8P5$?6{^J99Sp2Ap*W6$J?Hm8`ssHpf&T)cZlq4U}kD>kBRQ;!4
z`|Y!IWZ;pn-S+%XEAIdJ>pbqz_&t*;#{Jv){omsRzE=vGSR_0Qfxj)C|5xvBa~mu>
z9-U<1wf6t{p%a|IBO}v2bor03w)_wolY86A9m}vMK_GVTn!fvg{226tPmyr=QQ46&
zI8{9U=QH<fQ2|4Mn9c5Z`*j*9To?iZ_X<Pp=!zXO-}$yHM6~|}OhL8l7yuT-$Mn)l
zYV6P6UtM0n!JS{KT5V(Q@RBB<_94j14}p5QQS}MHQ!Rdwd%jgiS!P19p>YOfi<bEp
z{pTA|)d_?DK429DV3Z~exOLJ0I!c5zWILlh-(EtNPesy6uSQ1-v~ncldC`ZTf1B6_
zq1Pzze!xwV9$ecCm6g!D_0YsgPbtq1<XH&d2Q=!1O}P$4IfV`vr=!5u(GaBX<={z0
z4nb9O>_;%=gAg067;ek+otAO%e~hk<9IrkC(^uVxt~6P*Y^eMR7|y*(BICgD(g1jM
z!eX;H9rSlwX#)lMPb6>u^*R67#D=~EtA*oX#*M!&G?iB~&_LpSoizE5jsPa5Enc5}
zghc3RwjsMcpaIkc*RaZh3$Uu|r+OO&^uG)(sldCT65tf#<2DOG)K>tp<19urZb7Mx
z7IDbiXJUBe%MU;g*c^!Dl&Tf_3C#|?%`a*rpdxU^0Q7DHlmQrzkOpodp$Z|tSEAB&
z0MN!8@@MLG*UmtwzYZ$Vj48Yf&i}J6{;?yTyE!Q${q=pDago89V~v+Y&&nuRH|EOl
zGxEMpJD3kRgf#<Jp+?wuTyd7*PQWshR+r~c%r62p?lGvOg2{KD-}&_WT{wy(IZ7Wg
zfytx29*{uby2qj-wI{z;!8#^I@`f)gsRAa`6|tel7GNFO`h{}>!V^O3PU)W~{y&cn
ze(17zjDAS|m&>Tf6^#tV&|y2Y%@P0+e)Qo)IXa+JikPal1){mhq2(N9H5{<=sGXAu
zQ3uGR5Lo%mTI_fK`Q?9(4kI}CMv%FUe&1`KS&>o-AM)Cj(T|0;K~`1ws~Jjw%fe{{
z@anZ@=1B>luNWlveE3D8WrH@}PV>75y07%R#jj=(zo=(`m*u~LkAEO+{^-3g@3p*&
z)lAq?Drv<ffJy!M%&)XV#6N(9FX|T=(_&q5Z8$F;*zA6ScqSz=nzCVZ2VYMQ#tiaC
zLEioGc^7*UR5M0EHkb(lu9|#(MFwynx~8@Tt9}FUL%re99ybIeuL3|~qJDM33FuSe
z(p6NK^b9$OpMb%R09C&(fNeS00z=HV8A?r-5ZexTvbKgcz}`}mWWwI(%z=0Frax~~
z-(T)!HeWC9moVT1)d;QWVMQ(fXz4a!`}hc#GKB>u-6ozQGbq0V;Tj_^HlBdq7l(jy
zNfRKu+I}6ATRo8P(&|(X;H4qxB;4#{uQ-pvAp-fi$QDe~r+w`MRIiV+L`saH${q!Y
zm0*ax6LO`F2(Xd54^P+f6fm9u^G(E*%^J)hjkxoMe_FFdupb`G2z0&hp;v<%XZ>JR
ziJ6`P_Lqmi-cKH*9fO)g)*2#!0&ZBqA(KR5d&#hB<tMm>$ld`c>v|yo@zfaVv2kxH
zeD%F)6Mhb*U>#%=+tm$NQLL%jq3DNO)6<Y45IbL!5wP#><h5`aF_uHM0t~pOaz^&u
zqTRe*)00FxSFqk4FHPRmE)ZP;(!2a(8bg1i9@B^RFa=e;!Vo!mMU5NaTp8@^ZbLj6
zXVSXJU`cR`&BodWZme&<X1#X}4S07RhIqFoRo&;;-EkHzSGzb@=X|}tyR6FrOla8j
zSjK`uu|@j}2KSO7%o7daREUlT?ZJ%KYAP@dF~0pOP?KpR0ajP{FT|C#&35CVE*$&)
zB|}~E)AVhKjb}m2JVFjfKmx~B(8JXVSyQ%X0W18#t@~mq*#~ddMgSwMXZP$KVnb&{
z=y{FWVcLNhuizGoW&yB#`Vb>OYTP~S9c%Yx^|kQc9^gppzF5_-P*v(of^=0ij~H_D
zr-(ysAhuO$=Zop8MJ~;#t<0R7c%8HFynj7X{&s{qKLgwLaR`n1@AqvpGZLCF7553i
z!#4ou%D*?AYF|`XlO1iZ2Aqy`@4ws^Crv97)+9qq-gmZuB$p{I10rhP7*uh`peJ_X
zeg1J^Kh1u-rtrw%%8N3<tKliyH7b@+(RDyu3nXm?V8<z)Q1FOeof8t&<M3TPul+8p
zar%_q4*Y)0(iH#}5J1V%VgTC$8it2WSU3fMfx=;a!yU4;z5o!7|LiO8E2?mrX;=%8
zi5r6JF@DAM7(im0P?tZLQk$;pSS5f<TJUT#jla3y^SKDJdOQ1oIoL`FxtKyafN(&c
z__Zwqb{hciG;jn~Da+%s*u1{W$$}QcB3cS^RDB`aG80r*A^u`g$+kMT3K`Las2%z)
zao#-En2i&%BTV}C((_esg4;KopyhLbEvi~i;7W31k8gtkctvypjJ0vxad8xj-&662
zu|GjO52ZOOn+x!zaJHKSXkQ?7+nXAt=im(L85Rng=>pnZDbR?}-u7itXoP*GR)ldX
zj((+RJ6IijWNR?NxBG)~1MI-Sv(-Ge5Kj^tU2Q5ZIqBG>SN=X`o8U6`HhU3wymjUG
z;5p4$*lFqsy++3R8~cG{-OabmO?c+A*WYlHW9$+@<oRN8Z;pcru$-8Dm|3=rfQ{{X
z60O(8D|aV|Es3XnJ7Dnu*cgyDHCro!W3F@qu(D3RGhT!{PJ$u{1G3?X99pCH$5D#j
z0+jOIv%(GFT-*b-;993P-DK4o*U9FCq3i8<x;f!2iOxL6E?TTw3?0JpZlKl--*|(i
zC>MPY6&<MgMBvNfw>V0Hyhin1G=@U-95kGc_R_V|^$oz5eMh85N&2X_*zmQW7FsDR
z?K}@<ayMi*=cFVnU9yve3%Ie*oze*;OkJNhp1)3=9l90tRE<{#dDc$v0N5dnca`Qu
zO&6N^r;Mc>q2=fl(w}HmOD<nP9@(rpl<&1lZmtq9L-*s!#p9CFyD_>$M+S^N(Rr5q
zqh&*Ps$b%V9oofS(g`@o?#T&n-OH3jFfyx&_y`@~Q0Q}OJAyRLu8%YQh@txEE8zCH
zBi;>F>St|zBwnE%{`X*7H`>XOJNb`)?|F%KEHW@d%I7MgCv6o5^=N_}Gzu&4R5?as
zl9*Ck&4*>~o&F@%c6IE*AQ5LM^P(&JLG~2{hV?hJ>M3hcuC|Gk8wt2SNt^r>%zmQg
z?HGi9S-_5{zkyXxN$w6YOUuf@e-u_iQ6@lI<O$fAEXOUo=Nn<umpg37p8NR_=^yGN
zy~8`XMe6N(hYr-liSf=UZASpe?c#WxGZ8Vllg?A`?pU9+k3vv~LLl1+c}icQGL77Z
z2$gmoR@RLv>p7;ier)L&1d}LNRwT+M_s7WsG*_j=T|`oRvKzDI^-3@Z%-zA#do%%D
z1q!D19yiXNEX8BS7#50|?7TNrlO|H$mo9(^!88b{O92McHMGke9xImO8RW<rm-?B+
z+8NSmS4X!zxdmkJgaWK3aN`dbFcE~wbs4nW00s+y_<+ydAY&L|SmBp=*q#f^TNg7U
z8?RY^&ceo2j0!q^TUW{Z6*hsfxq$)RM*}oV_eaph4B4;N`l7muP~2u}zRi@_IrV-t
zf;bEPG>xsihxQy8$NFg{9G2J7j8)#Do*nrL0>UuclBDu<YgTZuqXd4w4;N$0F%9!C
z&wpB$|D+6y{n_x|Zj<o09YX(acL*s-2fv-9Sgb@bvEL>!f9$x1vZGWZ$(x!ayxtdS
zLsgRxY7FuYX>(&X`<Vgqqd(VJY8NC~<e3t@9fTd&)wD@TZagYuXFU|JElR<MZmd>B
z%HRc5D5}5F)!V!f!g9qbav<q2EbLq_tZd4J&y)2oKFZEt#Ld^NyD6d<LwQ@t+V4)w
z?0Gm(nHNJ@8Ix+r5$i53RJt)iQa5onc{n$4E7pwMUfH*#8KS0#owiSz?kH`0g$P`V
zcEgDJ2s6!P*NV?=q@jKFc$`UJDXC>XLPdzQr6UQih4smaihqi4UXr8*^ZSRkfeH7A
zm6+)a96~j4bB-M*vn**36t+}bEe^P@@_5uDB9qUQO$f_ZR1zqK=YBdmVYPg33Wu#F
ztbYEc*29%9k{7T=aKhPcrvN`7elpGG@OR|kuh|TpuT!Fz)E}4F@=p{ht4BD<A1cj(
zS<=1RNd2rh)EO9Phr2HH-ok*=8DKRfdzq?~i6FseiaT1itWUak!@i^YUSI&O`!kYB
zI@_|sE$o;?{<fqwp&fPAWm<A~-e+$@Y3dV&a84+slSQ0wlVCI4PTn<Nwy_^L)HNu`
zu_jl40PlI;DC3ez!5psm+AW3GDq28ARk1DorF|F~xmu$PGNylN+Lz9Fqdg@TER{zH
zoamLQY<0=*-~?g&_Ivgd5mK#5VmAUQ!gye#!!5RG>r~Ml(`x1E?!?}AsbxCuq2XZy
zTQUaeHOK^BxFEDK;DjP3;@V)6%FV#+dyP+&s7ylX+LHutjF!5YANZh@;PfrWFUG7>
z9jLH9>h%|Sn3>s-<SiT=(wG!SQMmebjIuXa$PkBCH`~WhQuBT6xZ~BuOV!Yf+pGa3
zI*+Xe$}&%BLrMK_BNZuAp@fxLC=&#>-Q#BXxQ%McN?HwA`$F}4!7AYz&<W2<G+l>=
zhO}VGGpkX&u$~jGph+!|+}K<4`Vp^ejYUeB{B5&xuWrig_jm;j{7_mTuF6kqNdrRn
zw?NcotU-)(R10o+>7Wg1#XC(B?)WdPq-I~8M9oJOmbM=m*ndeVQ-zisY&>@u5fIl;
zDk@sG(QA-3RQ0*`d>>LKEh`wDOv*2bJYbU!-x@xP5-E5)X>9$dKDW$V2sPdEb>O{+
zS3NIX>ejhL^}U*3?x{I`PPA-F>|3qOTzx!uk(~~I==x`x@z=El<zrGeL6)K;m5lKl
z2*po;t9$UWC^`=_#>bAbUR65yngGQe;mk)2D|4$rLNV})E-aVynYJjNK=Sk`<j_u(
z8ucOizCE2SckL5_C<-ooVaC1Uzyg1q%)v+W713_4qcX3E<fftG$_?5$1jnn6);ld!
z_d&Bb5tqmz2AzT69Hm(a;d?U4yxy8ra=o{X?dHqZdGz?ukf?ERizu9RD%fhpH8H%Y
zeS9cJ4>l*cUbQoDPOlau1#mL>9DgB`L1q`m`+~aVhAWXS_XH`yitd?ujp$4Xs2mm(
zjf&3_dlzyu>ecvUPU9fIK2J~gDOprHyy^WIf936A=xH!d{2{f>{y<m(sjwhF&zs;Z
zD!E;jCU?QxXoUD0CW4Q7q)|uWM&zY;wAbPi5mEy%4|qhuWvGTR4l=`-zA9bJYIu;c
z**PfpiF>XZ#q31+5Ym!LQPB;3r%L{k|JICEZ9asWU4CB`ZATyfM9n)E?`LUTdcn|z
zKS{*3H^Ef+Xiv}(+LAt1@_EIl61|!5VFM+RF1_;w%My?It9)|X+FYmyfWhgc{G!M^
zXbEh+CQq1J+sDZCDm$1GV7VFwhB#$I3)9ssMmaTEnE{2<-Ey>lg9!dM>HP0dY%(aW
z$OQk$%oA?mZtJox29I^fULR-3%9a_yl{_+t^XNQxwfK1g%)=U*1BeaczLaiPVb~NG
zD0-%SH!$<RlMYq;*dzmW<HAcjt%=izJ%EO=^ZaGrMqxYCqM%R1S-${5L{ys9^S~)>
z5*2!4wm|I<OqxDd%2d)Eann12-xnKh?(O@8SLgb%Ex%*B`;@>eGR>5Qz&`ShX-7XI
zI&FZJKnY4WJDU7p-Cd;xew@mNx+FW@+$Cn(3%#)vN(9Q$X-Cf`b3VrLj&N_lFLz+I
zD(qSSX^pV9;WzshWi6MJxp&}TH4rzVMvxjMQt0!z?||toOQ60E#EP(X%eH9YnH>Hx
z)c-MB#Qd2gx<1oT{@G?_H}~s=)|=cmDb~r!Bk=&pm;|l^H}`by7JJ<o0!`|jtvOXe
zqP(#VH8WVS>J9L{OGzq;Z2$yaV!hNmk!e_Lk_EbPsbk8zGJ6%DH5<LK7b!X_%pR$%
z>50u1HH}lUr+zoh8omTIuy$oBUHz<e8F(*u5RXkVtrGC-?2zrWXj7D9O(nWCasE&&
z9GE+7vNrF?+F}gqQyyFJgPR2AV?dCor+)<S9Yq-AW&RQBSMk*r>Nq-BUT9e9caWB;
zA$}ROp>J;{V5EW0l<1F8>0T;XeXde<&pz5Ndq{?W2{x8^U=CWYKJS!1`BFt{53KqP
z4+a%VLF&qtn-WHJ$AWwJ-9v?(6_S(e8J{7H3<pUwBPJFWz;bl_MmR`18hpp2Q}|@D
zVK*96{~%Tn|AoOgE74#b+R1Bzh-kwGN}i$W{-o?NNV6*JP#Sh39WEZaa8`oB3*dp?
z_<<XJ^&yucHy0i6sz<-r$Z?-tLmoKZT%s7LknG{|oiaGbB<M)mmsc4FXBopIaF%6C
zpZ^0z`V&5vqJR*WhGd4{&oTvU-G!HC`es?A8%h^t4CG(jG~!MfSu+;pXP<eG#o6tO
z#eg!+qhUzJv)H=;X{vG`8fd5au_Z5AH`4AEm!=k21GJ(e3G3C)w^O=rVQ%Uqu+-d{
z%rH*w1w%X<D{Xou(6SE>lEvqNR`+fY<5;H?<H!x+HYL|kktc}IX3oZTm6n3jEAaOH
z8(&tyl}gHHsEBt@lxp;WuzlEwcmomIbt>Kc^e{-y6P?!17C3}=!v$(hgm-Rzdp+}m
z73Uz`OsKkmX8oa08hHSMen&JX@epcjL_<PMkMl|FF^B#@I^9T?WuVA9OJLPII?-PH
z!dnMoW+G<t4fq$~Ji2x~{zcXmulkjvdtOW8BM;$>w{88%g&Hb_@cp=x-CnryY$*Pj
zVGfL3{T;8Y!=!BBI7CRH6O^#$n)RZb{Te~`-ax#@ibY@?RN50om5_IBgd3Ow?LN5P
z!eo;~Q{C$G=x$}=#xM6L;o&@W1t<i^=_|>Pr1A6x#f{-}f~X_e@D~#+E<{wJ)4(Ze
z_H<rv`cCfgG`Tl3Pol}UX}FF<s2;A)@ntR^@@k+El~2UQf^hK_k@r=tC4gK-q;sZj
zP<M2P%TL2(RdTB0Q#eQd9j;r(L%zgzM*baX<Hw}!dA(XbV4kugYg1{{M419hk?mBC
z>C3Ses$2h=X-}Zq^w|S;>*{Qdp!QhU1Y1T*zc1j@j9DE9Bf)-ZCD)Lt<$q3>gg&}N
zADbYvYY)<P=A6XS1K<sJt3)<KL&Ytky2_+BigW*N3j8yY`hAX1+S`~*%+&c~i7Xzk
zgEJ7v?K+-gK{D4LiEBKkPO*}E{k}w?sfnh*pQ-E7+jnCfW<Hz2{y%7xM{xRAZT2e@
z@Dg7GSEX<hfTp&vS!=h>Xa=ys=<JJf&3>xKm<iwFK9w^03G2;hV@|m#uiqJPFHrM-
zOPetI##AAQTpA5jW)zOz`+N5Zt#K0g-5_}20Z2tp0=#`q@5n|B<(P1@kK)Q+r(LJN
zQR+L|_iYBbAW~o6tU5hZ2a0xtHw^Y87fVD2vaDlHmD-oWR-T6)DqIe`dxs%@Y*{z{
z$OmkT!MD$|2GaE7`vNRJqYpe}s)Q88y+)ye?>2f)7ZPiiX-f`tt01waAAP_YPvxMc
zkDTRmMC>M9XDxVZ)(QfXHsp(Tnu&(@D|0#myYju|?Hm<SZw{SiTS^6@Om2d{Pb?8W
zc4}E$nWLUB_8h{i<L&?4=6``<fBP|3ppG+1X7!I^lllaXC}}jQ-FAI6=8%Y-q^r`N
z-ET*L4ZyHl0;}R0N<I*cbN$4Qa6wVgYWCCH>TB53DHae!lf=CbdU*eQzW;MXmeB*G
z2`{XoXBrIN=e&H#?Qe!>xzO}w#Tn=N08Ue~K_3D}p0Fwg_I3wB)`8c%`u>kMB_|_D
z`t`;EjL9MDMn@hiB76s|A|FhiNuwZ>Er{s(n43^tg9y@J%+T|=x(OZlq#^s8txGzd
z%2u}vCJFMRy%;o-{H9o{ipI**rzkGa2DZ}!$nP)O+{&%M1R#tpyvpFvVa;og3$nMK
z&!=PPAldRh<9UY~*@z%{_x!fbyw(6|pwl}AU8Ub+l_xm>1K*E|v*5PUKb{Fzb9GZ@
z04=o~zJ8BeE%~6Zpf->DVpKaMfwLk}G=Evz!c~M1efL(0k5DQ}N&e7^RTVUY)<s|j
zjV;zPl9R0ZWm84SwR3E!2JSu2NfSn1|7&bA#Kdf54DXu=ekxldwWV|d0#OJuI62%V
z`fJ_d??K5Z3KlZB8%}@EbX54s>_@7A@cvG`1Aszr2n()t3Gxe51DslIJ2aVDQ$<7M
z(pCz46R4P1*djzfe_MvfS-gNMH1Mvp>ppa5paRwO!)_%K0VY=Sx$s#QZJEqCH_RCC
z`iJ=VL#3RT(mz2J<TUy&)%G!H2Sd!4X-UPP@j$SSYB6YAqJi1_Cmt+wik+Gy$B4Qk
zWCJL>D>O1QQZt1rL)bMz@`V8|2GxlnpWKu)eD7u8DW5rtho&p9Y}y2h5YJI0i<I^_
zV|C?F=?Y@g(*3kf0qQZBt)xkFP!vuk+}p?9`B5cU+9#QcZ3AC8LT;c4U{72=?{#Q-
z%J;R&6Ut}{Ho&Ect$hb43}D;_=AOxSe<8W!KQQ79g;5m|-TTGcri@=u-|=?IwtE1s
z3i&JQoZ(nVDCN_J7h9wnEE$xZKT4NV0s>U=%kTZ+!xdCk?`Q5;TPmX=4F-KMto(3`
zkR1Tg3PRJh0PQ5Wbatj7do-IJIk;Py(-rv4F!AwpEhMWw$a?%OGwoa8#J)q1YA@lV
zq4&m9K(ZViVSpa;RMs&_TCUHEg{Zn>@Te$}&Ghdj^}pWM{87*(i-lDDG08&wy-QTQ
zAzl)}33B`2<TX)*Rb^i{fa;%ca8RSsXkeKSdy<w<K&r0hShsHx-UE>q>Sr=kWN1!>
z!Wqca7(JdQX!m_kU=^?Q6sfD)N(IctMoc<cha0^dAU5PNXt38u^ChgWY9;B|3F*2^
zU&LZNfK_ol?mob|ao-0uX~1EW0%-zN^ITRV)Zg^N3sd~qAmk?4RS|kYxEraDe0t!D
z&j`3I&vZ!DMvY4_ylorbVPh0w`{1>{LT&;fFC!r8JuZj^XRqN;D^c}C(>}hzqypIK
z$mhp!9U@{zv5s$|UZ+w|&K?phfbK{E8xf*%dQl-4j-cWvt(^8l1?Rr*xv!5~QbnHS
z)|51VMsagSI+L)Q4B;Y3n%G1?$Henh$j=TozD14(UXMAuT*Si<Ghlce(pog+3!u|G
zRt<o++EVCYX>NuF*#)oGvl4L`Qa&PdyMAoqQ8UFqlb*8z;%%W|_frmw^^x@)_p?Ik
zLE7;kM`@JQ#2SLBiKbhFU5z~O1zgY}FD-M<+GF%XEl6I&3?e}5gvfBV!7#*YT&~>L
zAaalhjh1H{nz9Pj7|+S()5-7MYqFiGbC&kYy0F`q$PwFpQ3%Vyeoq@hDn-<`9W`3x
zRoNH(CMbt41_$Sj+>oj?7xhq>;xTXt@^hfH4LnUb5SU#6J%cEf?GFY=`x&<$xkNAS
z2;O1DgBohT7GgaRK{f)2X1NVPP(0S!`5FI%ExeNe7On9U&ghdI^KEbu2igzXdN!`%
zjK!HWpb21O9<<ykto+ppLEfTTyxYJWf@yS49IB^RCTBsrDE*?E<;g|RSjc~i0BY~D
z>Y$Uw`sk#oIOo7fy7-=c&1<)`0~^4rz#LX{v=4Lax5+J^EH$3a2OF^kZz#<3%2MP2
zahf}}diJhH{QNJK@4v2gD7j{HR?Y0;`x}14?Gk8>M;8MebvyZHV?^FE0~)IDbVm|S
zl;hVIftMdh#GB5w5Z0>=BU+Pl$lQCF$%~}=oqW47zRMka%ko|j47<1@*V*)PQS*V>
zH>Z<Z@5}^3%I`PdL?_g~EErcN)49YRQg+j1a(NzTL$J1uS&V37$cl8Nda=<B+SkG0
zqL?zoadCs^Nj6my<{8yY&z=#`MBTvv>ZKUp^^T;XC|~lvoUoj(tay^x4nRs|a`D?R
z7YL{Zal|6&;amw*^YcI-f^G$)!&v$88SHTiOpv$%KTTndf?5)9ro(B<&rxYM{%}6v
zi2)VWM_XDh<kz-;a4b^bnUQ;zvp*hlYnEsGc4-@}WG4N=Fs-XnF{RVW5NbIHYH7`f
z>AWQz{=$fQ;HL{nw`8{Ii!~!{^2*5{AlSaDTlfZR-jg*qU!@L8&1PP!_rJnY%|RP<
zc|K1}QX{5ozHF*%QB2DDpl}hrN$|WA*FV;mbauk@o1gV59Ln1|+;77My^?$-4sNmU
zl!@H+Lu6QgVT`+c(M8lbZD)$@2I%x!i-Hr!Z%4xxBbsH+URpSUBR|^rL8mOLNQ~vr
zr1Gb|+$B2MT}gsXCk#qGUw8KhdU;uyqQ{x!SqKiJ>{=mT&<+1i$V7x>CA;qlbFhEL
z=BBDLKfb*vB9@RKFp;3;1@UmCF0YRz;s*&JuZ#DWrI+P*e9k!+N0h0zaaSZlE{3Du
zV-Z`k_s2m&A`u;EK#NQ_OONv94Y>v<*H1$>jdj*s^!Jukdl9THEW@m9aRD#nvzWi|
z4H43E4OjJZ9)eDzOk?Ft4BtO4p3Tr}bP6x+SThZjf<_fgU1#U!C)2z1P$LKBh>~V~
z)eOn^d^#VQrdUphPOJ!;TTb`Pi+9h!s}w@*$aC*7iLwxoio{dCve^WkwhZ#Bc70Sn
zwNm$X>t2X%UAHmH+QKUtoAQGXXJ!S@F(m{Vgs+2|o)#J3&jx-a?O@@K8>PJ>UQx}W
zTZ|woIM*GdY1_T9aqK_)CNk)2kyi40&21`YKY)y=w_;xNpo~qM!)l+;3|r?=stW8_
z!J2`q`;F&;H#2jO_zv?SC0TFtz`dM$o46^n5{y+IW2lEI@X>iulc#OWw5C!5E+oU4
z5zl`Gzj_{k3(4Xp?2pj#GZK=$Q#CkE=UQaZNcNrONHcg;GOUX;kCD!R!K^pqUb@C2
zcF$~0+K4}Ll3@r>VS+W`_cn6-^TTB|{Q5xN^Edp|!M;MmabMVPFZZG2;DYk=2$hgb
zFuPa-zdgSs^|;69w@QI;@7-|6L3wRR_S1v+b~h+XOG~BkuFU1TaX|R};ZR!*@5Gzr
z&j7j?Msh`w&|d>(HxfP^+kLh@QQk#te}*G%%lCS&C=oWfA~OY0^TWcyy+Imph5@6%
zn)`fTmYRe!&~*mLhj3aZ0Q4z+jgs6)D3L_0h1*5CI1pHT<Wp$31$v5<``(p}(T5L8
zdK+@S2jmJ*I*AZbF+m?w7}a6OQ+YuiZzq`t9G{Z)RDVqA9F_Q>yux`b+L&%C?jqsM
zOXU${BH{7?O=wpbS)NDDrO`%s?*lL$dg1-b*GGt5PnnvbZ=z?9^(H86+C5Z4BWKxU
zD75bUE*C$x8QhvEC7DQwR;L}qnQAkS%e||CN3zi5C%n-+5i#@91wC+E7iT!P_VQFX
z!lF`(@QpWBQ?aUat#*-S6ys{8h?E~&7*E`Egy<^@sgn*<X)2Fkb%j}De{ow)iINQb
zaS3~Uiwm)%(!#JA2}CR-j;j8~n+4ENm?cm*!l9WXACN{~@kL^!mJeokzdcCKF>Fk_
z7A;cbTQl6Rn<i^ss>jVeSPgYTmX?Y``LN5pJGUAyl!ikcp;CNl#iA0TD6eU8vAH$t
zc*22pS~58H)oph42&p6m#L4e3^qLU{gM-Va{wa1IQ|K&lHb+5fgM2x(gdb15StL$Q
zm10SA#)P73)Yd#26W_lmN!9P{uN(wr-v`5FIh2LsIUh$8OBMlPA-l><yL;hiZ!2hY
zZ{?yRiAzKvaZ0kpBLkfT?P=};PS@g%6~gb&reN(>eC>yx|0P4x!3Cr`HtE&hap2G7
zNcIV#%c3rn_V>dgMFj6{tWl;XW~47$;UJdZ+bee7v!CyBXoLw-+n-AaH*gQe_;jWG
zP~=R-%(oT@jSI#X)EN(26;ra`mrs5^88SvA{cS7-j)1Skoy#MW3VZintX7U?pA8sY
zv^!FDO;5ijYJ3MDv`QznsToT$l_H{QGswf#8Tj@wZRrgbmBnnFo5pO&YX|OqyOPT4
zFQs8N!*5s;mOuN<vm1`&V|a(I4i5j|5Uvg(gIPoopJ`taZ+J#hjlb1mu)z7Qz1E2k
zM<jD3!mDLijD2#4%UTiI<AwC2Ma^<%@8NwMVRrVMX<Dbipp;Mxz#*BWg`KZ_#u6)7
zYVoQcdpi<kBxU6evLjy5SgS?Av)Fz0Y^jaSHdLnDPC##;F1&ge#WOSnA${kF1;SxR
z_wz;T)bx7I7^6~G__-ZZMXIjy`<<xKrL>D;HhqT-aeus$cK|$1V^uK+C5p;L_(%Mc
z*WSZTurr&~MMR_?W9x25cksMy5xoCmlr^u~(r3L|uNJIW^U#8@?jJ=Fb=h%PQr(PN
z4lD+^I%f_i_MQ0t<~F5$$8AhSb6*!2jVjV8h3#|<N_<vjh*(}|C+VU0pTdmMU+*yw
zD(!z|neJmfv7{s(ey-C`D0?6_N73wGHJ5Ta`tOJdT@2ti6?)5%{ts<}pWGSr1t(9<
zgEgk(1s84$Vt`*q$CYY)h6cyCmJl4LIy5SyFmmF5LD@42uLMjMBahI<KMKD8NYJsI
z>a1P^(ykR}{^vV?=767Rk%3=?*jrdNkaLfM;)0uK1{XR8{J|4cI-G>WRC@GunRHpQ
zr!`p1n+Sg_6T1SIQ>1@oP$evZH9oS}5=8RnV>la$%mC_5zN{oq$rl?KJ&H&m3H*7-
zQ`dvF;cgJiJ%OlN6)Kf_SSU8gNGYju^R<9DRMs~ESUtm3WyDX_ojSplYXSH-5Y11S
zqZ<Q!&^b<X9AXPVwt3hYeRQ=TE<((>+-YXr1QlJxjeCj-v@@yxe@k!wegi#?!v>_u
zICQ$nKOp#P$dk@>k<41_po~yL>Uhq~M(ZY}C!eQUZ3rxi5zg0d$p;i%M?(Gl)oC%L
z@2Un!3&LXh9`>I{gY*~vBZ!<0r0=k<qU+9a{0V6zhk&c4Mfu9lOSH&L(rgoeg++jv
zc7!rfjQ!{6|2ay3j%f39z<zEiLQ3KK^Vt))umIpi3a%gCEFqMRQoU%s)B^g{ND9Zu
zX)GuHTWEHGe{%@r@&u|py+XL7Kz6HW72pac48_I;Zm^jVdqP2;OEImFA@fHG^M8fN
zzpa(IXs`rL#lJ57|KP#CrUJ}ep-8zKf4<0n{ZoIw8(`et9|re-fWa&`=6@NMfBClm
z1vdYEhX2D3_{#_C%4nRhOq6k#tpEBj|8Zdc@d0NXh{1$S1N;9mQq8OYAD@e6@BV8!
zu+B<w1`w!<nYEfn!0LlvoE?!$+6cdRPwpgs9ykTD>Bv!tSaf@?FbZ$|A4AOP4W>~7
z`Ax}hj|~>%4CH<XsITdL1CV*-hgkEWHe(BDNCzlQ02zK{%i)NNYdOKy5K!?lEcH6g
z{$L2xIR0|zqKW-169dxH9OLO4Y2`A2Qz(I5p+D8_{Ryp@lLT<b%6Xps>cR#H^<nzm
zf3l6a3;=Iqk$J}sY#`1x?14M~$Fnnlr64br4>lRJGa|Q9xk373(ZSpT5!KWI!!QxD
z;AI1Kiz6#6$nX#jG?;9F9I<~~(}Pq5_`D+H69C<=>+M8^ax^yWA6Q&KS+)UaVY+4J
z(~eiw8=$PhrrC&k1}K8rYPAk*;4h2=9L|W;en0pEP{G(f&5N$}r%8S{H+wmU(sn;k
z9-g=(U+U;7l={*WqXeYn-LL!!b+lRnh-ukafqkP0K#JI86u%RGYx;f6|LrZQz~071
z`(0Gwj22V}Qd{CCHlOAa&~s!IIF)H$zj=4LmkPLD5G#S0eglwhO8~7bPvpqG`N|O)
zxa~ROxk6_i*~$kLJn4OKsQ6y7Nce}pVJfG3g);$~SVR5U17;`1Xricm(m;a1bp|Mp
z@cJ7BEW}Bld7Ym%C&6Nt{ZubOk1TJ4NkFDjFO(IIlWT14VtOxGdpvu`7LdcZ&gq8o
zF_LBl{~hmzR@WymU5ifJzf<m?lH-(tht)!<-Vs1$|Il$i29RXMT>`~lOnw%8*tL_}
z4UYO)9x!=_pqG`OrTy9pvCKTLFkY-)YLsE~gO7$~T);w6=ixZynE4@6!-LJG77JO=
znJ9Ja99_pxB#W)rZf(4F4-f@kcuTZ|VQv4S59?>JW+b_@W+x?=Vf@!4#F-mxJ!^_T
zu!6Tv;PEA#T&G+s{IK$Gq0769s{mnG4gi3FV@xTl6>Ws&P3IkZ?&cD#{3vEnLAlGi
z`9GJ%KTZWO>-^vwHE;dC@pjyt`p_86S3$n7l_U}vpoax@rG^^a>-+2$1I^cfUv?Ni
zau@h_3lOr8gBw!E5!kJ+gJ~`Wy=zB6aj`tI!VDz6xo_41l;WdjEfVnXK|i?!$3iG`
z$r2a`W8i^PF3!>`1mAQ7tVB52f2qrA47I0&d4|B&Azk?h^dWBtL3dRoliV#>3m_}Y
zaOA~8%`D`)Au1zyq-Z!u`U@su1Sf&O$7qo1u|Xc`$=TQ$qX*b<npJP<B_K!5N<eC-
z+%q3~*K!CL6<`;3uY-%`sW4v**k_c1AU7@WY5jxtyI|^5%x1N%4mek>3@w07iw+5#
z8LAqsfR38Y-}2o8#N3B8lBpkHv9so&<k_6lr2>k^1&(9Db+c*XMeT{Wh7fy-MHnN5
z128AXuglSAoE@xvggCB-M<5iKNF4=gZ~2F^qi%{dU|T|MN)hD}w}1=Fa|qhG)1Bg4
zNX~2vNDgn!vs4NTfN|^))Dk1|3!BP#)cA)*&oS8{tI*yT{?!z*I_;q9u89+Do(uu|
zrguN*ES4<p0i0rJaVWsgASPyvgKYc^u+bC?3{+}1E7o2Sbxqj-y7?H8(5rA1Lms5y
z<dHv#wa|`_Uh?5*paDI}YT*!Bnb$Zg2jO%1LL~;Fb6hE>@L(Kl!*?;q0B8ZRwFa~P
z6qI;eAnaCv6oTl&{g@#-&h8L2W*9TFgiYfsC~E}rp6WiV1<%fyt5Vnr{!cHb5-Qyl
z&A6CB6k&Gp5D-W3R~kqt<Z0*|jLu1WOb>qje>nRNa4h#f?n4SGD^g^Sgr2CV?EN5;
z5SeAqBAaYwlaQ4W*?X_-nGrIwva`wF<Ne;9)A_G+&ih{Pb#-;gczWFT@BaP1-_Q3m
z$^c%_vflxXZXf@_O5N>ja73sq;c7|t*+<$MRKXPh<(=p==Trnjj8wMDfU1FnO_I!1
znJC0htB4VWU4w)*2be@pmIq}mjaE6EY(rEP^(euVua8vzf<f8>Y%lr1d%u2IkWqDB
zeqh8nyW%{3`Mje!Y_eHuszn_c3T^oRDro=h^!@!qOGG^r?G`Ha2djVbt;?57_?K>X
z0+k#O;j+i$h*7^~l(vHc#}sgXAK@qnd5_O|B8kK7ItYySi(;yX=LU(CWM>>fJu!}i
zv^j{b8w0<Jk<Mx7*Q@*hs{2Ph#2rz%7WP^`r>Tb&;5j|TwY_kn&_htuX*bak$ch}?
z;U2eA5sM5tnl(6wMc@1L*RoO?79kS<+-Q()L|=)2!=IN<up7UT2sK^Geht5~EZT`0
z7&LMQO!tT}B^T-6yJ4x3Q=8kt^E%PAgYa~KvO&>tmXaGZh0-0j!D&Rg$uhRynYay!
zlwf+BnnJbuhDKN#tT19lX%P)b1j^8kHTV4kR^LL83*;2I2|BWH@Z}gvGjs-d-QB;-
zmMdzaui<A4{O}=Q9}~Y+B3_Vuk<29kcFyJ*aKoP>X0pSLObp_^7T~ww$s9>Yu*$|l
z{0K2S6ic4qD14+P<?r1nKB0_D(M#>E*C5+=tN*ckq*aJ{{J`#e{V+}{-3^fpMyI=C
z)+rpF$~GUsmI)=7l4#7yL`-ql<=wR<f8(RQfP=Ts1`&R4{5O8bJyF_}>1Ra}!}LN~
z2Y%D<4_W~be_g>axC+C3DlsWy2gF&e>)5EWRiCAI^_la&YH)wCQnp5@L+gr7Xp2<;
z6>Z+gLR{~7hJ&}S?;gC>u%Ba=dK3G)s`DVsOIdaI1a5#a>8_)9&UaO-Di0Nvs)D0t
zXT;f=tARw`(#YsB?rMYl#1+!rd%LJza@?YY&@7F*o}%kb31_O(Z+1^yp?IAlf8+p5
zBjCpc;vB<(PhrODU-$1;hIF0EoXcZPOb+9_3fd;7s`uNhyV(pA-P{iu&OeL-qud9;
zQJI{GOcoHoPnqWnDy4{#M?KhwH81-r;=e3%N6x=-E(b%4N!eZF_5=|EzMEMNDzS<;
zk9(-v3!&IzthAW3jo_Exl>3_s7AaEWf^m18+me1Y1D$oaJDyW!eUJ2C!U4co@*87G
z3SG$s#vSzh7<G0XQ<MqufxF#`*UguFy^cpML<f_WD+I$eVsbTNq`6eRiQIgcnkxh&
zPhbdh2KC%PJ`k_m4X@74x6lpkdHN`b7opU8S(TK~l;CjKT9%K=MY}nJIAyefxTb-;
zhJLx;niqnH6lc`^&Ud6!1!j<_S;|{!2&gfWT8_|TH*nXBRIMNPTqC0@EUoH2ag@t@
z0HZtqj;+AGalrLYqa?g9@y)AEU2~dGK$Dz4s^KfS&ta<*lZJK>ts5iC)Z#-xL+vU)
z^83A<6PGmrV18(ZA$m!hgDt;AIlJ}%oOHaGO2IYR9dBD0)M)N#BHzPvHe}aMk=dOJ
zwCzvAMcJRGa=lt2$P~~877i3k!d%~p_TptyV|ORw`wVQztNghJnM;^c+Le=XH@)uf
z4zX7CR)V`QONk3Z$r$Zm;$WOJw+%=EULrYnCCv=O2pM(1WyDa|DdnoJKl<V!ae2Ll
z>h4>2eD0Jw#nx*d_W2@pHm=OsSgSr$<2lF~ez_L9CKDNqYCBDpdEH=n+V>y^?}Izu
zj@IY~kE$BmIkl7vYPOWtFg6|Q5}f>a&)N#gD(pZv^21d0P|FB;%ZiO6Y0*xC->N|$
zQ^DKj%MJ_$V{{c`@x!w`oZ?g`Z+@01d(m)bK1d`%mSsLGYtzfQD;beZDJ}%|@saVL
zBzt_4EF7(ww3&NDihM|?2{q=^(VA^1*0DU#DQ(JP%-gb3j;cKiq8}$SIsFxhvaWkN
z>Ey<l@2NI9_VCFPJTXq3DiBfuT43t&s<0V{nHjCo3(qeOp^NHVEXU(rZzeC4du7W|
zSPB|Pn8+F@kI3r$nD_J*NucJ1!{2@6;#9CvsNUBuh1f;Dfe#}W)h<vw?Y3%)k0Q`5
zcBZUWnw82AUn6-I-Q6|?YSn)zBxlty_tLHZl-x}M{%e9Fo<(Z$)yXUmr<n7(ImwbZ
zDFpLnNfMomu7luddqKbp=Eif1o3kRh=mcLa7RUQ2;zT=z=rN+gT3ChT3~a0{RIaQ}
z_dD<~^1r5?6%($u^v8NWpPoD~?b$fcmq5^<f4Bdw?yMZiq`4`hkwRCA?T3U;7%*`y
z1sjOQ|AtG$E36f@&K(!HM@%dEjz;c*{(0QpdIi(a*Wy~Pix*nA7y_7$Jh4n-B6C;X
z&A+}Ngy{{`mRB#4l-NyFs@>;qIW$IEGWl65(u{U1@<3=;Xp3qR$cu<li%Q>npU`a&
z+?ajXNZ%_5RH#gG50?S`yKI-oKb4@UDp;FGB{$U4!L5@@j+wsZUZt&1$5_|31VI%y
z#`YPLAX_D2+Q6fWyqLNqhdLd`RcD;Tvh#aKVdW-p(F)K|CloR+@bpsehQjVeA49es
zKS4C)hORtIto><J$G5RDAx?}UH12N492@Q9t<u3T36+ioFGBWjL!rjXMk47*0TJ5y
zP}?5bHgOJhBh|r{Y@=^o+>aQr*b>g#zg3n<QPR1QC*!@(poN^D^4)1g#JH`ewxGd!
zO}4opoEb~y&t)sxoVK4Q(4vf%cO>qa4<z1@U6(nOGA=%n5T-*{WPiV#&VNIby6WF0
zegD<7$64UeU@czC>N)*;OEvodllTEMc|g6xUJffp2*0R)!bcs7a1WQNorY`kr_VD;
z1uIJt_2r4xvO6*>F(lb{Z83BzLzU>3Txv@HPR>ymN4xBSXpo0XOv^3b%V((hdDoPN
z@)*S=+o<mD&n4_5RA0*!qjVgln4{a`tEt>wxB_ohq_s|Nz8|^p)#0NDg-=G3frR9R
z$yE7-``IIo?WZwxw~39f_<ZmnDfHQ+4@$#Iip0|m%bP|UDLtADnm4u@N||Nd|Cr~Y
zKfukCH;T|$%Nas~9GQZoi!{8+$R6ngb!zko<KAbyjo6*%7dIVVKTwa7QsuqhXM5A=
z9I+}Q|E!U(Ilm?7rk=^+^ZD&HVs}aQ>A5d>4J&i@xLkZEtu`9O;<GH(BZgwwuuQ;w
zV5QvNQ*MeImp9n9vah6>iMQRfckcS&YZl(6f^IFH`vYNEJ9b^!N54(NNF{aP-ax^K
z9Z6_k_xeNc+3p(*<XNP-chrN6+Orboz`bu1jpZ}1YFwFUoImu?QLhN~A2Q7UzGv$=
z8xML>XX5_jJTq7~aGEwGX{T=s_f={;$;{P$vpYOlkyG}oaSZA!Z_B-?82i)G>jpQc
zwMf?kznOm4C$PmTQx5b!V~QNx$-JhqrH>nV8;m~2iv02dNV3}-UHopk1std0{!?Tj
z5$KjQ?qdkC3rD}Z5r|_DbG>+8ZEyUFO?&s%Nar)Mo@qs>F%DTd6Dp!k#2-1$BKojS
zU@9tEiAxsGC`Pz+?Kauwz<E`ZCox|-c`41lm+XzlqbxY&zAF!|8rX3ebkbPV_}JBR
zTgW}&)<vz|B-0s0d^D+a*#j)T>s($mXhQ(ab<4_WVe)m$R7xiw^!e+u{zj;r8)Pur
z5cSzEOKC#np%xe(rPVz;vV|?sY3j#Ja6eD6lIdDS><WJxEJ7n>3dCMi*@x%SOGsUs
zU}{(2(z1Hpx+Ri2^=hUcIxV3s6Yp;PReQ^dSLwYkwhJPr<*LXli+d^5LbQ<}g4A#D
zYL6AIptIC0z;5RoFf1xDyZKpr;5(9lvY5)XpJ&s4arcC*jJj3j8D}NMC%6JF<bmR*
z!bj$d{kb4|JIvI(#v7^mE*5ci>*a*9m!0eig4XKANEOYi9W^1{LX(hFFa5h|*{~zk
zqi<qE({wBrxmup{KUz51ICDZ*nd^R(*$%#aq!)4fA*c&Pj2@2niMg*jUfy4p6Q}r6
zr%>hlslB^7%J<x)3R`S%m9w)&u8vwVnet-{3tGFfM5FZM;pZ@dYp`M6J(2m{ac_e|
z7X7TOBjMX1=!&&t(mqbBXf$$1k|*jvE9erphL{xVex~k<_PsVsfqy)7ULY<Lu^h60
zAbytI7_vX~h$-xPZv2Z)+Dsy*wCBP$9$s&-gp0w~5S<BjoK{2oNNNc0W~OBGWmjLH
z<PuJA4&Fqya=Kh-+y#8{*R<~R%D-YyVihCNJ&n;o?`XN5;v(l2r_DmFj!(hG<}=A4
zKD+qFL!U}FmF{S<`37HNnw{un3O@S+yFGDLfFKfFTDqAd(Vl)=6GYwIGB~qA7ZBa2
zx3Dtq?Ripf&Yq+ickHO-(o7q}x*1pzb2aS&=(4Is(}ei!#r0O=7~O@%S*NSHOF&B3
zh3R*RY1QEt(<qCORZ(*m=MRQ8#L6^3`Sdn6{-p~qrHGC7?_NAYscK%_|9OIR`K(A|
zYIw(^{mDsRDsJAk-MCGGs120qu`*=Re%@3j^W^C^a5Yybf6w_jr0bPQFO|#}VAZ`q
z2RXk^8BG{!v|+{FVaQl-b~Sw=mL~E;Zw+a@KG;o-8947{+T6%YSE1&!*C7!|a`;k|
zVRf_jGHqOasC~9U!Oevm6pr{`|6YybB8Z1YLiGAic4-GA!ar&1Nkjd1h8`Vo2G!_$
z#TBZvxy_KW2Zl2TR<tFAIXR$a&p~If@W%4Z$}MWCOPzb?gT9X!e2?33MZ#n5e)|%j
z1LuDzMXe?JDMU!Ar&rpGR=?PcKH^C{WgVp5n~&I9_A`+YC{PN3(eiv!gG}m9$OHDo
z6$?4Rk4^ME=abkn?FA(T9<Zn+s+xST6>SWU{QA<k>l9$Abu{I2ngyFUZ+@p^(k%22
znHW7zoC>|lT$r$!Nq-!2uYB=A-+{!Eku~p^L)~W-GAE)OogSqc<Xds&#Xc}AAIp9o
zg(sBo?ffpY`UizXl%8pDK*0l{r%m*+r^raB1KtLou8>K*O2ur<OJ0bZcnUhG#DyXa
z<2KyO?>eU=flhzz^ZLh8Q;YZ1{b1fUQTX8$Dt^668+MY5PD($nuYHS*Z4ZG1!a3`~
zn$gl1?xc;aH3IwNHG>w8-lv({=_Y(C-Gz<KM{4&ksC^_y0Hu;7#3fK@E`2=_xql0E
z-@?~;51(2N`E{6VTvQ|YI}j9(h4ynf)6c%&`&AQ`op6P=p3b(Ven)7a8-t8Xk~)@$
z4r|zU(jV`KEH+=Jx-yVe-IK|)h~L{8jVm96(Nrkv>M#JifkGWC1}Z|-z3Ys)7zo?+
zAhqS0cN*_W&F?l&`Y(tv)1J*Glqb{ptMdU}+b!7ve}1Nfr5&e(A$+B$tA;-5WMF|(
zek!rLT|s5!j8*(2+Bhs#twQiw#Tw913PnzF=Npqj%S(R9vt>fjf;?L$Is#x~5X_+S
zwi<o7s?e{<kJ6lZaQ7-H938pLr`uv%kG2$wyEN?>dKbq_G+$L2%NzP)|J+>TNyj7V
zm#GRlY)_ArRfN4Bkk6&V%D6js9BZPBoZ69L2>GBN+7X`4!J;tePkUKv_P{QinwMB(
z$a9CNqN!QG4X%DoHx`+O1;9_IGBFB$`kG)k#Ixspz@recU|{c*E?lkWsAtdrk2|0m
zvIB-)EBxb0<BoQryCfrUxG@gf*tHjL`L8>$jN*d;Xt)ecHN1r`Az4N;fzQTKSND{W
z;5$!*Ecp>;0R!@i`#z^q^k?hr<?JA?NQ>th<lM)13k7WgRlTSTD5IhwvvSmZO;%TU
z=;8)r`O8?-!jt>Y?s;W1+ZQ&L-`E@WTahLTyoqn8`l+a+8xvdByFZH%g7|Fw51)#+
z*#Lqk9wpzCXfH-L2&Wp&P4dkgjp#YGN9**xU)yXyIPhpAMb^-(pG6kMH(4!J*=#C?
zjJLKMfcB@W`g=I~+)qx4DfDS{1ghm37(a@vki?mySnEa$QwthS{JHH}*dR#V!WeY&
z3c)Uo59yxoCN5th%3=MQgOwe+dY>|LPkq;3R`wG8l3{d;60f5fQeD=<q4?11QM&1U
zU7zcBpt9&CT{FBI^Q>|={z6U~#_?e;#DseShqyHB0-u=FTLX607B#edMdoP?McEtt
zH~3<&{u4@uLQt~Su)#kH<>-5YZXwVE-M-{}u&Ow5N-R%~B=Q#rQtrSEx`sweOUe6*
z23%6IHo%~3T1Oa(;eg%Nf$zctQVewJB0ld*b$I1d2nq9&d!9!r@3{Av|9go1E3l^E
zRdYVvR1_|R{hHy-#p?_kOGCFZfd9(DXtH%wkevc@=t`hl)-&^PXO(FJV=zv_)(9C0
z#fOV5-8+=nHcsb<;qEoIyk<I~u97^rJ7TzRlpllB1@63YgjiQSgYxTUQ8~wr421_q
zV@OO<8Xqx{cf^;&K2~^G>F^~bu24m*GkBfTJ`6TZ)s)f>4@g`B_)fi_#NRzW13_xW
z;rCAwb?LE{=uBA+!MR`i4rZ%O_Mt6hDH8RUDFo&|1Z$}@E^-2~92XG-7Oc||rJE?x
zo$;pDsUseM9!N4Zmytov0bKm%7v~;qwT4q?2B5D{u1bZRzv1sVC^cIIGnv9WG0`Um
zW?03n%WNQc35U4l!>FNZ2;B`;bi}@-39G=JZs{8@5kja@%z3`M>*`m8bjyiV^k;G!
zRp*RCvmAlS>5n;`K`pfzI{xE<dZvjgW#B^-VXk<NQHYn;D&Sw5dGx7iw&cZ2I<ND)
zxMC8$JCf{<Fs;E7%a&nGqzSt%=cHLFZR|d%uGe)V_%O-z>S!ugFN-=5Q1}p_!bwR>
zEpt}|4M}!wX`G4^vWe57d~J}D@*YTk=F<7ntp@IWrv`lKBoQ(#ZPke3VbFbbxrl8*
ztrsU*_oa5z*L|)>%HGP_Lqc75&?IY&Fxtu-PL9s&lgo7Y#6;?<#E&`zgH5hSz0Aj5
z;8pRp=Gm{U`!7(f*-Mxe3Yo`Erhj(yd?7CA=)BD0kwG;RU6ds|WOSh<*aO@2QB>Du
zH4PDj^DOyIW{}FdP7{4!1fL1^Kw)i^0z27<a9^P|`L#&j)I~_u;UH5CN0)LO6Plv(
z*aM}t!NDuZ?q;4XDsTay?OX&_nwAS|d+9t)Jw16y71Vh!SyY<SgoMWN06^MKk3Ib?
z`(O*_c`p8DotRE8HL9PzchV5E{pD$EK-(I)9|*AAI<{$ov~cPfZD)BN1yAQQx5$FR
z@np2rFUQzw9rYzSKZDq;pjbS&wuJM8h$oISfF4?osHq!)PjxKXIhBU-0q>;}0NN7$
zG1wzr4hX{YIUSq5+O<<^+n`o&fbzRR)<~+M+E;^E)PVl!lXA34TlX{ZX;giNeLW&A
zACk9PsS*JAn`sCfoMgCWgn2E3z*4(bA-3@`k{||-B@L+3DPoIU^vBO96Q_k=rp@cf
z#ylA0b&D*$&pP;W?-)+#dh1mFfiyf96R;u3EK?HLXhRZ>1`Fq{Ycwy>4-_O@-6XSW
z*fOB9BT3x>E?0qVT<@-2qisR|%h}7kwdxzJ17*B}(RL&iw@34u)Hf(*9Ja!J@5NJ_
z=c@6t9mVwZs;vZ9w5e_AQc<#gPDnc!y;0ZDYW}p0{c8+J*Jsf~4k3H-rHi$r{Wb%k
z2mXugVw_z`yrHF^Zl#RGKX_HenpT|9q(Vusn*@)<!plC2)h}g?uOb&v%bAwrFdHd?
znRT<8s4j}zzmo39WgS^0D6y2CZB2F|tWPjg0CoDtt^?WYh-AV5E5vtVn$C28pu_2*
z-9uI$0NVrX6h5V~813F0_}GU{W}~DY>txwLo`gW$Y3Yhp=^UFlfw?j$s$0h*ZY8e_
zP>EKe3vbZ5ehV{D*y_aGQx41uc{mg|W4vU=VfTXIjk;k)j8R(Z1W%AJ=_L7-@N?*N
z-W?-E<ffOOR}ZDOGB>-sahhzH+bN-)zN^M;uyzuL7=}@W6}?sp3p-#S#CnKfU=nmt
zh!--HVUZ_E#?!dTIqDfd+P>2kZ+{z1oPPDp|J76fhWQ;VzLZh2M!(2oxES7sykDZf
zA<V-2<8qtvuNxB92f#mOvS|*ug`ID98WZRulGixBCm9|DGv5_+BVBTsDr07Q0h53>
z`aaRF5VVO8b=|sKw27xxriU5AREfXpXtuQxfdoTFi3%aP!x^mCm|FX2$Lo<SH8Emh
zPYLD8uEY5L`UKC2`3(h^)wex?i-&_>vA2N0%6(dz0aLDg14EPUoHrq(+G>icDZj_e
zg5gxLs6FBa^!>c|$j^AqQQ`+_vMjA2S|ULiI8QYPg;r)jeJ$6PG%40V&mrFBBPs=>
zaOScPJ=#=VZ5l8Uv|iXSKzE^hP$35EVCQhlTcuvGYQKC;7PA8}J9^59CG94Hdy3IE
z7>7-hK^f*M+SkfMSEQ}Kcg0Ctqa5CvO5G2-Ml2}7z6P_yei)~!5EU^!@difW(|IRv
zL!R`4>jKEMf1mM~Rv_U{f6zn_-{iXv4RE4FSNmv(BUHyu`3cg2`V!`~b8BZ|8ghS-
z=)GwkYx^NGuztfVVa-SpmV$N?EK*4i8UJC_`z>e^4wZC(<gQ|97ELXQZ%ucZvI-}u
zI0+joso)h_OX?)%z7Bk&EbS!hW;u}KbI>+c<DQVhx__Xz%{W^I;w;t$lM4H(<P8&V
zB3~(`j3-9uh{>R7G^>-6vsawJgHGy;5O;tl*K>s{Ql4BEOlpq$9Nb1Y*AjcV4BE>;
zI1+!>Frf7z=VvPOJ3d338t-lB3x)Q@#EyhayBz}SEmT?+JOe)Q7h2m{cI`MlV7Tzq
zin5nl071r%<j`{aUpNM)fIu)hzT4^7A>E<%PYjM|G&rNp<R<F<zKpAHGsZQef%E4n
z%0_%)HX)DhYYKC`1`9We$b<ilgHN$fb#3?O##uEVjikV(xp(}8cVEceEqFcO`_O9s
z5iwsXmE%F*A@ODFDW~4|1fzSlSBYHs-=<d>MQ!y8^rpYrBXXWPLdO$2W_&FqPM_5%
z6d`hla1RcxtTg<?p7ow(g^iCrql<U=G?(sw+7g>oxch=(JSzU-mHEz`Cv`>|VKM5B
z4~Kgyx1@>$#J;j%Dv@I^E!dj}y%srazkXko&R-^zuKbOEE6OtS_IzwVV^ZOAuQ}gc
zuB$`RMTuLpvGdGBOg@8O77w85@)_)D&J8FMu`+tcNw6liE@}~9K}T_sOgrBClFhJr
zQnLl|Rmu9V5<%*pP`XO<VS8+&46y@2TVdlr$PJ$z4L)p2AdTweB`G%fv`@18tbIs<
zd*X7><=*t%*Rov)eUV|C4dhYwbnP1-1D<jg>ta)0p|mR+jC?UuqFmRr{P{xQ^@FSu
zBWYb@t@ni@+G)8<CZpN<Elis)3Plbj*VQ`u%4I6JW>x%)k0f`ZrhT0j_;*kD%4HYi
z1x$~w8}TL6b464Xlg375{u6<fkU0#fMbGUXQ2be~GDlZKhe_D2!lDu!j}1dE*EJkw
zl@A_8V4Qo8WbnM$2eu!EJy^()OaEN8{?_7Z!scHQCxGqaQoSRCc#85>Zv3KgxKiC4
zgMP?6HR!8~@zrt&-#Fbpy|se+@&d>53WgY_f+cI>iURrb-z$wCCYO$k_E$Il?)87e
zDnl4<SyUJ?|7uRJV`Y_0oaTWTI@Y<!&@^)}J>vrDm~xfVp$X&z<UwtB)dIU5cy?%9
z=Bvqc$BrOh3{@QgH^1O9YjBi8Hp?Nz`50&%7T<Y)()O4XFn8K;t)}K~{duAOi&)kv
z4-B%!RC;vzFZBDuHC(V$YWxi0n)|Cw)Y`|wIH68k^S!NLMy5@1+wAf83-H>4O}Ga9
zXDPVaGyo=QkueM~bV6D-gy-~=YcVr`5LDRT6IbisOQDzjTWsAm2B@HqWhw6rp8ey{
zskW6IgKaDa$Xgl2MC_xVfRxji6t3lS@C0#2p=4MA-P9mbh1+h5o|Dl5J38(0I|wC}
zFq{x~?8fDKx#W(&yYxW12Yz63z@??X*XjJvk@xGxZbSPM{vQs7e^GV**Y!nS3@jCZ
zjOvaZnDh@=(*JN0{y2ew$8+M>8~EEp`G3B_l}Z{eD*+X6B>q2~<^S_zzkjbFj@Q}$
z_y6k8*SBYf>%(WeV!-`(i|uc(Ecl+%aQ*tKU@_wgP(%iS0c!#^LebITzRi&J7xq7E
z6886mV}~Zf7Y+CR*4lOBCSVE<gI--4e;ff=qGLd9$weantUHCU2Y|_Y17q`DmT(*g
zChUuzXQ}x=KJSK{6jPYSs(-)5L&zLe4d$Io8Tp^{2!%83fmT~ojfod>cwydM@MSe1
zAKI`6D=J{-z6Gg^1Bg8#uxg7C7f^7iFaeYQY_-;aNBW=TMmlo-_2z&5mVe&wZ^5t^
zEZ;#Doc^6N*2(@KcA$0J)?X>^@$c#cmrMc7RE+RW%DIBAh2mOqk?}Z}<?MYRAa9%W
zq{wk0-i^Q*3@i3MwQ=TW5c4lyqGd@<UVa@uPa8tPDcKr^;EEzhgS6UlLU{<%V}ZO5
zcL>0qB`~nd0mrN!Bqv&&Xda3wqrGGcr01Yl+)6PnAG-AYFasSy;QzNSuuL>?=`Y9C
zB&GjcUhGci`k^+|-Ux&C%IxFZEbviCUNR9!r8twoqY+*WiAL1>vk8HVh|Rjicd&U%
z-dq|g-q8vWrV$a8>x1-J$#R%Hn1brL^<8Qe%n?S#<@yjZ8>Cm}O8S+7pslCMJD<?N
ztr~nk9|O*8>xXFzIfkMR(ON!N;tB81R37aaM<MJ!Xs=A;x0apn+Pq8n7;EKR{{H3P
zJOwV*9+Gejc-&du_}xRgQ3Ufpc!Rep_94YNp;MH4NB^C9<7$$oB^GA1*ch<kiou+;
z2E;_?$5*pQ5H9V<_0JSyACMxvj#A-65eRfKXdl;2mAp^DLTXB8EQAYT0IhI^3{dJ6
zi&^-r0LMrR@Rt=Rn}kbl@*<CY?-$Vg=hFH0&i!P8?iV1NAgTFtLwfAS#Lj;j<MJ#<
z;yPRuSYAwVGGoP-=%6kveg;bc^cZN*b0L}n;KAam0cakhd3cVFbpis8wWSGuodf^7
zk%fTO_Txtkseiagcb>!oOU$d~04ZtPe?$yD!2_0W2<Y;7s(HW=8-S~$X^Wy%P>4q6
z5#q1*_D7GO{^FwL0X3%+GOJk;9%3YFsQ}@_8rxeJF@KwanDlpiJ$lD*<mNu9do=(h
zy8&34BqRJlYoL`5BJ8vThO}os_KY=C2t953=m29=eZoD69zBQ{=~6rGJXt&jpg|v8
zDFO_YL$hSBaOD3Y`;CTh$?F0;st9O;>`$%k9iG!(V(lGA(zGJG@Q;B-f;j$a2lB3Y
z8D=QXT(W3;0xd#@X?{Bw!tvFi{-&q7B%fvgh#a52rO7%x`xN`Ri1#eHaYolu62?Te
z5YyClBlicLK0oMu*|Z=CPriVZV+rxho=HBDxkcaG-BJYJW){+_bMDKb(VC#V-hzh!
z;qW2<O%!6kYVMR#GU3xlL>p&f4}3}hKeNSR-y1}MsIB-4`7pA}#CtOt6-Rt~;#4pg
zxO`neWeHvGt~iOANePf&5>g=`8eXy3QYg)fOI79`ayET$^^xMD8v~R^8@l&IzQc{~
zF3n~oW|c7@Xc|Z9MH>1%&AzmN?<DLGC%Y0ty2=Mj(q1SuzReEpZU3NlgddT8hM1*h
zLaP*NP4;szXdckA%@PTaql~^*LZP&AZXeF>7%<~XXSYg{Zk3%shTH735W=DcnQdB;
z`u=tBdiqF$x0(_s6|`76h|9=Ue(Yh!%D~)Xr9e*;c_*n#)yjFqkr?iGxy+vCZnfs&
zyb|Y9{^k9nWiE7-qrhFRL{pFe5A6MeuJ1QjBiNDB?N<H}n{+1N41-0EHCSp4!G1w_
z5|5ZoGUcJd28JWNA2x2%Ku2J(7Qv_jy*k7YEUksdmLo`VjkzI(zqSM9{98k#M}0`@
zwRM9wC-oQCb4CeRgsBoP$wk22?wY$l{xJTmHwMG9+z=_J1a%kVyWrm*_#x(~C8RfS
zSByv#KMui&XJV81qQsdyQ11=F(?Kofszod3hot~-$V`zFoPESe6eQU%=k|aD^2L*?
z0AxSE#EL7*AQ93ph0yOjelqu5HHv2UYEZXBE(mvo;mCJ{?A(p7j6ImTjGPKX0PN)z
zD{gT<J%Np102ErffQ0&Y#JLhWsb$HLMl8q#brJaJDZ{~|L9;*U+E<bplawJxB?B<K
zT>L^-Z!N^KehBEKZ>#u@o)%KS4CS)TQ5-@UOH-IwN41iYN=<TrirnBU@uq>A2Eerj
z`cS)-$N&h!DSu)?MLofewkAVVlU&w6zfn?H-hUDh!E;STGAE84;?lvz>~I!;J;BeZ
zhdd$R(kNI~Cp<d_uAX!eBgeC#5@n~SV3x$$P$`LH3s7V~q`8xSrQ00?cW`ml(441$
zC;2<wb1RAeLV4gZn7Bm^C;1kb%!=1UyZ}@PA<5nTgl#-xu#~Ad)lvA+C_wTcnNxJZ
zb&t6#vCn`bC?~Vo`PO81L8auSBcOuU?`GsG4t2mlw*)l9YhUdv_vi#@&8NX0Qh_;c
z0e^B8Q6zIp@!mY8vR@EmeD(8N`>la9ttQ{pm-frMFOi1njdl=sUmv;hK;7i;_ff&P
zQy9m;iTz@5$9WQ)1u2;YX_yhI{dAzj8=0tdnmI;1#5o=2=-kG%%E##Zb6|%(75JEh
zcTaW2K%z##!xJcHJZv!Gs1XyX5fkNHxH9Px6Jc8JyX@vmI+!__(1t|@W{o{0YEy)2
z%w+et=RS<p7LNrHYisi6X|6~})+Df>B$FlW+pWQ-_5j(AK*<;l9Ap=BKSLBUbuK=~
zs)Wh!<h?08sz<3Zlqg;=t8{{<OBT{AN;zf#=wk<#oaaqbD49=7cf%{CQzBD4Nw#n@
zkX=zRF#K9!&x4wiE;S8wxdan$8Ye+G>~m=5Dp>*wsq|$T0kr~RH7w$%pwcy<N4-9S
zaypF~!|M<@e$fwh)NcJ{G>qTA=sAEbf+q~Y*h<W<Ya~1Lp5kx>7KR%--uKW#PCClL
zD$dV|<l%62&2{j&;xLW-^fIjF72kxW+53kju;)!)Xhv~nk&BQOoeCU2&%38`>72h+
zr2(({8ImF}aTvS8HF@c)6)MNuLSwz{WNKF!Vi~eKH2GthU7%$aZ4*wdyOCh_n%nmN
zy1Y{heXMv?P*wBZNWrx#l~$^a6yUA%j{K;t8Y5zKN5A(iVvOp&5dZ-~?11{e0!O;E
z9Nlj<vMEngWH~*j(i=n%`#_{c=!=ZVN(k9)Mf+-<X5-KcXVzL(MC!%(<u54&_8~0&
z8=r+T@u-#xW%%0nSji~Gr;U>FNGDbJ0pmXzxb`SwVeOkypzqg7@@dSnu^`-`g7@Wq
zo_KSt#;JYxIlArNE?ze^9?;Kc%NHyg+_nk36B*w8irB`KXV2;F^0Uooy573;|DaDW
zI(MIh==V+*x&UCo1->HV?bVd$TfONe_nRfy?qlIPGO#qC4<tGN5>3^6#a5?`*{?a~
zN1G$=9ycA9wCPHSryM(CzFJJwFL}WXJkjgd=Pl%dXuc$Te~CnnpB7pBDj~?g@zTQ$
z_;8Dx$a-VNyog)CE2Hzegq9RnRv~OK^eTYqcCWqay2-HvnZx-=xB1ExR+?qyawdpQ
zx})zJS@`3pIN>s^i%q5%Nhpt}>Z~WTRy91R4WpBb7-Y-!6AdFiA_>jglT)xk2Zovs
zwFelYhg2811F<eFJ&3<6JMhRVqOqG^ztN6vJgJdWK#w>_YKLCny@lPAu=&@@$`gwI
zsaPpuiLg&SI=nwx_S`9CATZwCXewN^D=&iLY03MbA(pR9FsXf>#`>d!w@icNjYCLZ
z3*SKW^4R4bD%`P#C$OFfRMywTfIxBJhBj=C^{MK$@uubbZcj2hJrj<4K$-mRxzvk$
zL4-MHoC_f~wsDV9Jj9H3G@#&>Auq#=9`cZ@-*kENOr;izL>DQq>gKfZr>DQD8tm+~
zxY<fuKH9ebh!qY%{rrDH3IE%bHb8xN-eR2j_r@%)iO4_<CuP<Q@^d9fFnbqDJPdt2
zlt{!<MjSKW8BdGkC)b>fsVrE@CGMINDWL0~@}*msWMk|K%}Q6YJO&Y$%8xg%Jx&k@
z&Wbo`yq2@f_i*-Uv989z#uB4XQ&b2`o;VZJct>0^%}Sj+tMW?O>*QB{OPv-B_HrTV
zD&L`6!mF^~JRKu^TO6YWPn7Zc3z(~(eq{Afz^JgWZ#B=fJc1*$w<v%r>^^!z!CxdO
z=SD9*QFbz&|70W?CUTNK0p+1L*XyingHi?G4b|}yVyrxO<-`Si_V}`9(E6JRDbJQD
zS?}stg2#9<QBblzt5zai%k*8F)pJH+eSXm9@ri3GYUaf+^*j|5$s`r)l|S87C`Pwn
z&nUJ{vi;&pohI-+t=>Q6>At}@_13uxB$nOpnR$PJkKURKlT9QM=G&U<L&I4shjY8?
zchK?_K1KeOv4v~*<ThRe{DT-zP!+SnvRMFA{?|PM4_72YVcp=)bkGF(yf<D;TepAY
z@^KyY(fw_Wz&ARmn?T$$rsBw0F4T?QxLH{QI^;nJtKd=iju6CPM8+xWsIqP!7^#P*
zb}T^1=!#!UT<Q?2Q$Ny&ya$GqC!S{;y=DY@xivQJfFDLsrT{5jm&LtUS~?s5Jge_B
zqL%?BOj#z3SEsK~8LOWUL>y_XVT#^ZY0FT|^|ef*gLfj<PD*E8LLs*THFcBBBO*Dd
z2$U9myp$<+A;nzV^DcX{Zldc7_0yP=>jin!L?7DZFGcxYZSQu8^lfYtyKK6AEuY$y
z<;R%o=bR<I#sc^DOgkh#qpjfsB6NyC@21&B-$R`KyPVW<45*?ro-aY|p_lgql2jXX
zPHeV7cc||G&s0LYG^i&+@0Qnr9UFD3i25CYmPPbD_$<k08o>dRZ6_+Yu8?<DY5gY)
zkGBxK!<!#I(haTMSHEcr@d!7_=#-*=;5*TAN*ovO0H@|U*>GeZ_bw0FwaDZ>f2Ys%
z%Rh--K%LhezqBUdIz3UolihqdX{ID@kT=mL<CcT^qobDYmMGum$$b~!k#6lF6D$9{
zTOc-C4{2C^VM88h;tp{VpPMy)+@^}S?T}_x#rAffrMz}KdB*1ZB^guZL9RWgz7`&_
zuJt?rG;uKngwCYZgZ}q!&y`9t?h)-6^a+$N)v1;mtziylz0YqpR?W;D?4_p`A}(;V
z7&c)y)=g8eqctjZki0!6mu&*a=JI{VI!}UP)1m>U)0+&H=}HW0t2cAyn#bpY(@~o}
z&DJ1&6BTNPe#!_jmbcr8VKxhL=}b5giKV&e;gZ&w&^7>z#E5wRBgpHfyE5Xv#J2Ae
z=68uxz^yZ+_q&o#tr%^zg4XZ1zM(z%IDbeDm783jgSfT3ec&R~!MFV&1q3dy?=M}5
zO^n)jRg|aaC)5Ky&d=;XtM^E^7#f+7E`rUnpF1uyRZU}nbz*Z0&yF|SRqx5ETRQtl
z9sCZxmsv_LuzY2F2_PMBx79_*!7#6nh*}P)?K84gvo?>8q`*7vw^%TAw2oz^9RvD@
z9tgfPJ{Fbxqvr(^K;f>OG-EtOS}DL+#X4hPB+^yXwSA+vYw`c<Zp0;%7hcp0uYvi5
zK(JRz#dUbXno?weXz<7(?CtZ)E`oOOTqC_BD|`iaifi=6`rDyVKeNX$ZepVKthb68
z%n>o&abAqZ4!P9OTXYRGIR7{A$5=R`0ZjTqH!kD^HD5f(uYDaSr#YNPI`7PzPj@1i
zbG+%?KhXpyd?FrjslnN+l0D42!&kwVf6m@U-)59szc;n4Y(6zj_3Yv23fn9x-%~`_
zpMte#O#+RUATR&Kz}X-2Uf8Q|Jv;^X^7v?JNNEHn*gxv5dMZytRzWc^4%$~}tN=nO
zSY=Tb<uNjP5h_i|QA9D<&7znw_XT=4vX7#4?6nm7g#`fAV#&!q#PKT5Lo{AUVy#he
zm^uafW5I~(e9~z&(UQT-T{%&n7VdO!_bif&g>KQ!BAyj>P5?y_)J&VV@$v$b&ChQ9
zy^(+6m1sh-y63|s=5yVgKge{ml?8Z1Ob9vX`m#a@TXRqOk#n-(_<#DEb|{xkn8W<m
z&|3Y;{^WZ3^MFR~Dboc<0bkvlYxjtL^ghJ;Bb!_BuRe&Q#603qmi+alIBddxXbH~|
zsQ2H$t9JV}8xYDq0WcMtw(U^zt-e9$6Vjcg?fQJ2)mS9nwpOwTLX?efppPpuIzou{
z-o9T4r+|a>jF_yA5nu|{C}aP3X4ibZt9%%jCDW#d=1vfNiz>lkUnP9QIU4qPRx7l=
zj{5Sl5NdU$VeW1s5Iw%B;V5&5jqCz!5_|&xHjO!k^q!^R=*jEez8L2Su0+~~kA2xK
zkr(CBdem;>OEu)bVR&P6W&~l!teRDRg7>df$lMs=5Q%p<4-5PwRHrX$OKh$^(Vbbi
z^ZOl8dE!LU6=1)YrJ_OR|B_?!y=%H|y6qB@m%i;_Fy>>258T-Lh8EMTVxQ<ZEhK&C
zIp8ftsoK2bBTB)5!CcO8RZ=q-<5eHLo}~;mydt;Kk9$O0Ee(s_p6_Lyj`v5EUho?G
zI$S(^xCK|^?ONdj=Ma|fJgARntTsEV=eiZQeIG|S2`cp;?HpviOUT8y8?qnih;&SD
zcBuU9?ea5q@^AhwV02Ns!1yb?(v?c^G4$)w!#sCICO9CB*&_`XyAn(SbpW=wH|+3u
z;t4U2zcL?nWU9gqVZnN$WjXc;kZJ@&-bAty#kX|jJlMDxfkmO(@WTHzb^!n0Y`qBX
zWvZ-PtpPQ`sTr!>L7cN7=loG$fyK0Jb$Qzf>nhY9Xu^jI#F%B<rPc2v*<Y>kO_>p%
znC43vu$~Hi^rJMs$vVMN<yl4d?%WKm$SV*2Or|1XFOU0VuKEvgW}id{L;rhYbtO?q
zIUFTWehi%#1KdEJEz6v%>U;t|258o{K)S)MBzJONA?F)<gr%Hspx9XZ_%P>Ok9KBZ
zpVUK0tWPseai~3;L3mcL8??Hi(f}EHJDkcea42o6$Yp|ZTWvyvHeU=Y$T8p8oCGuP
zp##oheZ^!bpJyTR$Pu`$!vlOvs3-bf<`?QYZcaL^Mcg8R>$CHB$QOdX^{qQStmhHi
z+T}j~zp;$}_8Eh_g2L5B(w5?XJVj-U7^2z;kM%jk#WF(NGPCgsDxZ5W7199(m^A25
zBlC<e(MSu1uuuBkIOQ7dd4}z~&?OrHubJRjvA+)B5hLQjrWhnL*Lcc0<rZed2DR2J
zb(?R6i<iExs)gLUWZ@Jn!-cES)!%V?oKVz^Dwjm#6lleLEs35S1Acf{4`3-BQ2I+E
zaj|(bG|jaR#M>~+8vX=XJ8#DrIigou8gJ~>Es$YS@UAylm`9Q7e`wtSK+B!x<dYd_
z^h0PcB`y!zw_YvAqMhh~Jpe%?r8V)jh#N6GWbtZOn%#2(&Bw;wM7{-pAozZ9#<oyP
z^0IAsITDBDC3!*KOLmoa_DN!E3^hTdDVi-g!jwq6+>75d>FlsCnt9IzMD;SC1&$Bu
zMZnec05JV?1i0^LT9konFH*&U>b7>tMZ~Vyfx77#c~Z~eC@x+-|EXUb3+QHej9%Vr
z50pA!jkpS?z%fa*)VvTY{+Sj_grN+=+)p~ZEsE+Vum23#fWrmVfi;Bm`<Ob1l<Gl6
z+sNy%yc|0}Sps>At%%^cp3~NDd+gJ>`d*V?G}cO9RWxeHdww~)Fa>1ohPk{c+U{Tg
zUr%zxdFsjLjVus84uq7ICr7zi;|;f>d~%cuw82O443+x+))EC)0Q!=-=yMp?Fd;C1
z-h>b9DY;eHO<wy^3-GJH`~bnV&kD>dJ_4TTrmoG_Xkm`G{ZRp{^<mCoL1y|4vz5X0
zq2Io*ZF1bDKcDb_TWf@We(XAr#WbGhTp~yO>r=DSSjaTTV%7B!G^aO@cb-9C<5k+<
zfWXn~o$=EeE^QUX5x!m~z??JaRyj_2ig5kxnnf&xf=a8~)MoWiJJ-p`Bk;HvLGs5z
zMi7q2JvcnNh<iNyY}vf?pw8iu+_GaQufo$5`Mp{QMbig%H(T`cX9m;4$@i7WBwy~;
z)t!8u$O{tWJatG)pQa<QpnX-^-!XjG->$jk0To$&goFBx76;gw5UYpjiS$lA<!YiB
zR8kHQJbTo7o>1ZESso)A41}I?JHSqzSBokwH^eg<1~p&RK0dNWIacQnN68zU+e{Fc
z4%f!FcoJ(Tp<xjuG8>FI+xZp?$2RTTp=PLIi>fw-sJ>Id;CgU*Ws|-0{W5^y_+P&{
z#FgG}Lim=-qK~VswqNU_gl~|&N4PqOv<2VFOh|cbOoOQXWbl)Wkf-;SZgvMKaL6oi
z@f7*ack7YYlpuXtn^?+|dta#A^rwA&-?TOGSR+<=cx`PWIG(vnBcIw{mQT~x)g?q|
zFF$I11qOWmc#YPVqC&=N`|k(%3ba?alRHMqWU>puqmkYgs5ca}6r~rSE^UHZJiGN<
zc}M^A@xW4?JAQbJRi4rW+HCxur4==&D8hZGYPl!1Y9CWS1!z>&vP(>atM6PR#pJnV
zzhc-}T8D!T*_cdJYwnex1udRv=Y;kf6F5*UQXSSuV&IXT1{6G657a2S%9^d@<Jbk=
z6&aBZ_?%>)qkLQH)GTY%ACr8&rChGmXU3;0p1;*1GV$st0e^#a;iB<H(>kYeuW!S3
zDM)Ddd&hwpi{aCUu9@0wL9-|3JNhNFi`7)M{W)vPK09%7w&szVIGe~TeU{q+r($u6
z3!Y*h<wdCmeg5W5^ZNPusdKLh*<*b_|M6Xd#SGQxXZ-IR`Uhzy1O=NsE%*-*I8ErC
zD?0{<2-oV+Z&**4K`QoQMaWn9TD`E>X(E0auY|S$2^_!Sp+MS$bVfTbc~!&2P^m|J
zA%=ufq4##~aS`a2qj`DhJWX)Mb~=AhVRjcsIE>=PyMs2bE(UxGXbL7oZku}2PYBe9
z;OO!CX%AE^^K~`)Qz%)ZepFX*oeWYR;r!fiv|<*#1;+?B2Nhj4v$W`NF+9|#XmM-X
zO@!HSIuMG#7d_u_t~tRb`nJ$F5O3(3T`s*``o`8%dmBp+R%NoBY~u;Kn;z`Tik-vq
zPY7R^Y&TS6$f+dunDDTKs&?!udx|$%%Q>pZMz62YHToRxfy=`wbK*%?e&KMbQM%jl
zqmq6Lh_AAjvzDD#(eR``Ayy^K2;cjG2)rM0ckS%1s2u=E_tM@i))H^sg4zRvBFZ3b
z*g5rvi1t4WT-`cL;S)|S$GoYfK`$u&(Gv&fJvH5WcuAAU)T@uHmyGeW@rhWxIZtx1
zXy0P$3j*(un^dn%-zE4Dmgm)y^H%de?jXWI5bs8NpZ3X1>u1ap96q0wPV*km4a&gi
zm&LLQsZrCps`2C{X<Nm@skh6nw~CX3#1ghxPGij7-4?!ZlrsaJfXpn}-YHsM;RdNI
zeeYTK>-gqQy$YSb(9wM@L(}d`QG(ptOs#!GRi)2e04=9D%l#k;No2sJqxt(k_aj_9
zTtWwCef}t=Ebn5z*lAiI)S|@V$AP2o-2Ir}*oP!|o6H4-3F|4mXbAaeB3tbXTStZ#
zb3J$PM#7D*U~`D5c%U6pEx2iQq>3W~2fS#t+&XS{OP?bcG8wOKGM;a!uvPY;-~P}l
zGpcM%k!N)`ENCQIAmxT>f`37T!)^kqch6&Q9euHfQg4HCJLm!{hdDp1*N3aT*r#&e
z-DWZnn5dPyq*ZD@B+a9BvHd9p<&fC6Vg6Fw=<K;htV!2ZiP$gB8w6h^{QP;xQ4<&T
zg7+uiJu&J0YGS-3-!j{>+hxL&;(X~V^!>1r<WF81>F5`#Kfgq&?VDR&^4L$;^z3FM
zSFZn+XKgjNqdlKr)VE7zUBX&QJ~|k$VE^n8Fz(`CKi&I=+1IY+AYR7&>ML!k){oaI
z3Y83LAD5}^xfJyTm*okpJoVYhtRQJJD170c2boAJ18Uj<Jw~^NSknhLp?WRMi#uw2
zHf_fxCk}UJAkT{J@9zuTvXx_837nLmzq)yRE`p3e*f5BU{Hw42$Nlca$H2uH!uZho
zAD{jG$2uu-ByjpE?Io>9ejShhIy(RTGp+%&G_;SK%oJ1q>Q%<Ea1szVfuWt%Nn|dI
zH?9Pp4$CXUpw;e2L@yLfAM03?uz%mZ{^4&+7I?2bq{UQBe{o?Nh$P<Bqdu+(Q`~y`
zE>|{E=Z$1-5gd<5`hf~$<JIOK5&ipn{;Q8?OR*TR7E2w|-~Mm!uz`sQdBufL>)M#A
zaDyir;z-s%{qf#Eyy4%!w84u>jJfrZU(xrUn#jNGEPsB|o(O(z2RZ%pzq-2r*BeZO
z$rlLEFkiN8FWs+h=dXVEzrM5Zovs?9P?~G>3>c;TQ$O{$mqD<D9u`N(f_$OGuMYX|
zf7idbLUCk7uL*O=tcoI8{^GX(#e(~<OVYWHtm5@Og{yxSd4K!6{?&z<6+;$<SafF6
zuLISFX$9=~1_2?}0W<J7GMv2%-%78fmH%;hHdrtLNG)WLG+7y|sYv@~l<|MLZk9~Q
zx(+UiVf^(A0&NG(9-6@!;UwZvgrtBNmt6YRUy211I7S|jw7C7oa5#m;3zNu82k|E!
z-lH)Ch;;}IQPs8>ru)DxNZ$G$7#7fc3quf0N{cp*OwGBc59{aydA~p{Tk#;1Ox;qB
z2S`76oA|7f=@+-4bX$TpzW^M`%jI*Up|jlryfcJPenIvkKBXnGEAV#)A#-{0%RWLi
zHGv`O;)mn~j(@lX6)$1wRU{_&-uq+eL3s^6D25J5nkguUa)sPa6}~VgBsT#EqrNKV
z<G0zy=!YvHs>l<;zf+I6kFSJ7j^ZesjaE_YPuyhqHhU;{4ML+u9>Xx#8d~;{EMCZU
zL|l>(2L)yBuNQeDsoUiP9KDNs0N|b38gnP$(oB(Jrq6+nX8;H#`pBf>cE_Lv$ZFqy
zpYz71d{Rv0M343Ewz*a^9R--u0zJ}nJi<Ri00(zfd{_h23x$4FG4E`T?cqmEfV{<5
z8bLbL=+FwzQi`kgBPT~N5ao#(F3}MY`V!HI^BCd?H3-&AZ)~Fe5r+`0gT-ZXA^Pc`
zJM649Ccxm<D>ysA;2@8Asd8sr$Oxj9bAgG}zf#l})s4`SCYu6_!7Ad);}V$iEW&pd
z0Y^hzxEgigLM)OP4Wo+0%Se@nr0x!+Cda67f9-+`(?A&DSPC%25ZvhD#)U^5vUA@O
zX>!yc8E6qA-tcIcVqZK*k`MwtzYtkd$bAR4hAt=&g_75PCNGa3AKBRIRV)y^_sb1D
zz@cRtT!9h!AmZ+aH0!*k(03~Butv|i@*v69<GuJDZ5139!0IVNtfUZL8xo(1i!r%?
z@NwV-P=>u(YQ}E6M%9?a@OZoG`0Ni6S-#B>QvevZz{~xo{Zvaf<v;CR%x<vE#O^Cx
z{BtQ%-Wx*#jU55q=jz~H@9%Kh%Ex!sxIqBw%UcMYt{5h6HAUdaujeywzI_CIlG7vR
za4Y<?#M*LzpJSOucz->izY^i^nA)1Pz7n{)b)_QzriV-aihYI4=1mo`{v~kyzkyTx
zh@zZ@^9#O*ukA~}IbgWpJt=AOr_h&|;Z)=`g`i;h->(ONVeWt(SZo0#w_V*=ILaWb
zH%Ykx|02K~w{3!hk%IS-_RYH!i#)-<ya1lOx;SAda(fqaFCn;+r8}T<9RL{z1MOq)
z5bM9#D}QaQJ`Aws2HjMA`>&nnHU@4q3xT`1$A!?l-JyC=JM^>FWL0SjR|hPdz{NNg
z7>dy<_BB8g=~@nZE%m=~C?|Z~S&0}r`(M0~)HbvZ6syMo1eAjaMqgK}6NqyoN4p))
zugqHJVuA?SWkmM>y!!w9ff)28IR^18vVR-_+(Fd<f|9`_Eo?i1B>3_pBbZ=ofNxF_
zFxZaKzM!Ptg=lKc0oZ9{RCzw;LxFB5-hZ8Vd>Nm6HZJG;AVR8w!&nn%MvA!$Hi`~C
zfNF;Ga0108)OZudB(VeUytu6nR47vb9Z`5uZ<T&~|8?bq(pA$ukbe!;I(ir)Y_lbp
zKwjs?{er@xgw69K!U{t+BBKURV<e;;$(Vgl2}?+NOxmj$99XL%;CY7dQ~+VJhmG$X
zz&U3|G;t_ygZCmlX-kMI$v^Dw5DW#s?Zn04vnrSMH}Jfetpw47$W^8Y9pWK?wS{BN
z@$qd>(5cZmO6_R#TgoB~m#<}VL?<WbU|^C5xK%qstz<s})U$Fd(-;mC#U|3j<3Q(w
zIx2AI8vEqsR9Dvx6AMGYmeg<#!|~UZQaOg*XQ-z!C+CsS_Ha$#t3}~UAeMMC{3RRV
zH7C;5+8Py5lH!?8KpIT7_+CQ#3^tuV*h^w(>-bUf20z&b`k+{y6R{Q<g&hEYSo<}~
z3C{v=Pk0fIa1dC4uq1of{<H2&m$nW#c}~hnvFSrP51zK=PcMtw&1QEiOb083QU~k&
z+%)d@YOkM<e%Nw)fqeGTWgXT{Adyjv1bOt-$FN_^pMy7-kMLeCxwC8%&ruf0rxUXC
z75O|KXf7#(JL>Kjz&y*-cW6@UL`)ZU#5n$hEIZC|f2&>ZD&;M$;@o@$Ql7g*f;iL;
zKw4S<@?g{L`bjEpb4dxkd1)_=kDAxTak`{#j}%}RnTdotlN+LoO6*$<k|@P88f+7^
zE4&@s545D4>Te?kV`^O3NLHln9g{~3Ip0ER_L_DEfNa{jR_~r!G}r<m2qXmQPwkq@
zfSc!MF1%wpa#fXY6~$rF&&~r!P%D;msE5nS1RsKmUNcrv!MP1{G<I2A8&A%y6VF@`
zv=>K6&071j5WTVaJr3?aPy7EWSm2+X?nuqz&MhzVd(AqljzuU-V?-}XLFduVHN@t@
z#sdWXvlwJ<ocyopr6%3iAinq_KO5EaW?EDJTJS{}61x8~$2%||saY$aix#4>fI|3G
zAd}Z=LZh*7IxveZ0tV?P!mtnLFh%(ZL3G<h>+wU)qFwjg<$Gr^@kOaYUi&H<lzihV
zdzBmfbgV6iU+<UXI)3UWS3C(;MP8jTw1__EWluQ2itz$8147mAV*y(qVY-d*8a>!k
zyl`0)Va9gd)<)<y7Z80so+E*dov3;<kKaB>*S{v<-+7#RVehw9LaQcL8n+zYF52yJ
zxunR)843oMe;}-4p_H22E}U*Hi07Es^&LT(SUQ}s$t|e7NE&Mag?XqL9`RI|kM8vH
z<f9jsXs;E}0-k*G9f?A3<aKNhR|f6k5*$vM%s3Yz@LnCD9@=YkCp`!~BE%KeKyXD7
z&+Z|<ZreX6<$VWeZk=xk->kUwoWS?voF@Os?762S518oNNA>0vpgkY#>Rop)^-jSV
z$6dDp5U4P_a*xIDN<9hAb;6c+FdlsM7(cfQKy=c27X+Aia^g!Wj%qL}yX!E6@bw*6
z8ti?sS`9fHtZ!bm+QODH4&%9C?mP%b2F3Knb$7d3a*>NBkHBbCNy@+92}`|ym3alu
z;#hE@G#6rzar}&U%++!E0|@F>i$M46U-Nc-__Gvolo?vZn@AT4t+mQ?WqP8mPQBOU
zezJ{>!D;mLGxcRe&lG_Xc4B_$(pTf+(uh*`u!0K@!|Ye<F#9GfCBEfhAO0AFoT1Z`
zq~*Ts{<*CVC$*eRfF5%GMhsDE_lU{^dbkRzvKQM661)m`;29>+lH7|ZT!!ibk1oHT
zX=EnHJHvB>;D<|FuBGtfmV+kzl}Y@)*W0ZhH=K{RoyVI_*nQ2L!G{}O4rp%eKJkw?
zNDBGY+<M{xldCpuJe)twKI%6qNaEnIj$r5iu$RM;@Ot8O`-GASI~QTO1df*SN(@3T
zZE6ZJje6wUdism&zs|P*?meL87&|^8_pgZ@1+kf&!o-LOF&PcRvK4AI@TIrIh9q+Q
zPeIX$^Yvg-Yl1GZzQM@bK~3^tQg?zXXRx480P`}c5l0uPol9_~ptr28Kl#AhH-Y$A
zIclfrQ~gPIZ@esGvJFO*Jqv47!%1)7ljn{F-|e9H;rl)T5i=G{*g_Vl>qQQS^A#qj
z(XGoe)H~P<^(T48sRrL3IRGb5<l8}=;ojRX{G0m?*PMb%eq%QuS4uf?W@j=I?-cPi
zLug6KwOyb(Ym{B9n53==_P;f9+nh-3ITWHys9C%vvun3q=jsRqE-9BWUpFTb-s@XA
zGrtgnCBmm3^624l|M?b6SDk^1%;#fNZkW!;f|P?}wN=%%FX<Bb6EzLp@sfzi-Y_mJ
zUNa98l973Qz+Rc=?`kQ8x+J5v6NFTRiRMd~_4O}lydC_c)cPY!yl4i12+exkibHf_
z!oU9ImG^-c2nt-yr-hsDS2~PJ3KX^TZ<=K*c;nF#A4^bAlYVFa{B7v*#7L@ex!TRH
zX^nd@GaPn{;+><5tIp}uiZsNw$YGtqwhk~aM<!^f37qw>CtaSfse{*tw*%|lv0OlG
zJ#t+g8>pW18(;Iu;ViuV>{_gY<6X6SrItoDOAez8fy;AO&1r270`c;s?E1$qP(g9`
zWI}kga|?>nD}IKHl#@dZ4{P+UcRZqBdyr5hu!X%qydISsq^>lvp5lu)MZC>VzLu9|
zC0=_}ij(^dP1=wbVcszKJ${D0=v(>!;q1$!q5k*(BW0wTK_bi8vSg=F*>@$RvM*V(
zWGS)@6OqAKqRmo52q}B^r7RO=uk2fvB5P!q<@bDl?&s6J_j~WBd(Q8V&pGp%(>Z3|
z^ID$I$Mdlm8ou(@dB?+m4|=^h_}!1;$B)<kgSjzNMg0%I))suk@A1_fbST;q-q6fT
z`!Z2wi;N7B!S%dcz?R*$L?A(ihO;<=*sfTukLLTG{=aUZeDuI^GmCX7a!IxXXBO(E
zD7?e|;vB(1wo=2>J?I~Kt>qdgSF3`j=jc!1@o%k%-hTVM*6>rsOKUDcy09+jrGx~>
z>!$K~5!a}Cs>NurJA-fcBJK&6P6#hP*U<d9qw8BG55>Q4yPv9Ks++>@|Fx48q<Dup
zjO)}jD4b9{1)RrMHE0Q%L{%9<=bqe-D-;)R#(mwL7Q#zCdQ6U|fuwG!!Z^pulX#Xd
z=`?$u@}c{!Y{wS@4!Is-o1^JIaewW(_8>p_3oW@|4&dyYC=@Lv7e4gM^6d>s#Y;5P
zr<SS5ld1x^Z*5IquA`iHI&zO&sU`~}I<6|JEGtWLmFb)<#ukVak}bsJX&ULn*!vV+
z&LnZ(eaYR)m7?Gts`0KWhMOo~FvfI`x^{Oxof6N~{a6eO#|w?9-ngjzu)F&{I2Ng3
zS2-D`?zO0w5#=*?vn{>KeC~OrNoceb%VD8A`)z7{1iM7LdOx<bq~c?oLR?sjTkbiV
z+fsz)W*b7@<RwD(nb_`TKVuY;cm)I_Ft0<R3u<hOV@s1HIkOt~Eim>Y-o~{a=$5|p
z=bTPEP|+}xzE0MH$!q~hSsVn8P(fn|j>HaWflhvkh~=xn+%_kqbK+8$z9PJbmxqR^
zEZ%R#Yj<?k4VL7;oRo2WgRNlw{{GsuP+e7LP;u#MC#h;Zcs0WP(Oj7n@yAu=?98Fl
zU!qN}A89Fu!}#RBwgNB#{HZOoKr-d9IE{A957E&{a=?9-B@f=VsVpci@9qq4nwnx}
zrYA2CX|fnA_|8KtgR%Qt@c^&#JF3}lC&i0I{sJEUODW=plp<~$7li&SMS%V{GE%fp
z$EK6kid)9g%y0cG{VP?Dry;Ee*!W(JzN}<kD69_|fC^EMo}RtFmYHEROh!l5+a{w;
zkB*#kv+g*HxiiL3?`$EZB1Pp|b)q)W0+Z;C>Y?~xQACi6fx}EjO4EQk;f8?T#Ctxu
zdCdJ9H^a345s=+XQ(nIt{o`3%H&@C@@K2^ge`b>T@~r9I+~NloG6uHJeoa&?#%qK|
z)Q4N;u77u?!&N{DEazsRxvF8l;JG<UGXv#*BdZkahe!vH><MF8rRo&qTHnpVZZ3F)
zdXkXG(B?k*`mH(pC77+doiz&2lSr%G*AuFwAak+cFuT2qhQZ5N?YYa~SxHCG^QJN(
z`ZkQ)T%;*x%Qfhk`<CCX%uI>D&3m2{A;z)mnaKS99-~Ye4vSR5a7Kfk6X=R~_NlO|
zz}2alKBlzg&W3-!N+>(8`J&;P0e|k0c#3~(T~18Pd-ru<l~J2>ewuhO+ien3SuiD(
z+Ut2J|8PkP#-)}(Rf9h<-3VpDy`Q`9%WP-ka&$R{fpdM?)d|nHLQ*!LkSuv#bh5-H
z;TNp%pODwL6$+l@J==<^|8h;z&AXJ2k4gdbGts<<nchSRyOU=R>a<s2u5VPfZhFE3
zO37UBo=YIzd|KurDs_3|SxCDC!^KC`%guZACN-T`j(x70dlVkfEW2J<W<E;@kvqXq
z7Kf)OOz`Z>6rAqKw{`~Nf%s~$4VJ%b#(1T=iY{0j8STw!<+%5%3XPL<xV7DIZWOWE
z_#7F&x23A|OL2Fd^W%^+lB+=CdvOZySaR|;FvbvJ%W-sqW53fYt<KXG3hzvtt>^B1
zlp)x6Rtlex*r|3FHWXW}G?d5<k&@WB0-d51`mL$Ncc>~E>XYzUC(vxhF|pf7-bKqE
z`f+jyb=ECyHFXt>@Wx$s`|uEPfV5dnLw~!@@aS<!tRzGD<M4;gRgRBCS|u5_<3d)C
zGjKhoKKH34XbV<Q`Y^e6`(><LmyDowMH+o)WKJSmgk60AwNz7I#GoMz`<m-AVp|uJ
z&PK>GkO#Xq1#voQO_J@IDUEBAiGUjnTRUia)annz>qP;%$3M&3lu7j$or~v=bmYaV
zW4*8R7I8Yg2{DsI>o^HtIk9T&2q;c&S*(JaVtSd>$wkbtCF@4Zbw4^Me$V=<yD^=q
zsRsuZ9rH$_yP9U);w_CjN^j`A9w)zC7yKYk>tTy=Br6M$=gJsQ_vAp)JSd~$QhdZY
zayq}?`P%m9P@g838au<LL9XnwlCc3R*@qsCvBFa=Rf)1j*I5fz-Aw;He%8Rt^LTv_
zo?^eGaIRTV9^%zsi%-Yvw+f`pxe1FO{GL-uZDC#Rcq88dr!!(!SJhJ7A7Q*vcgx-i
z+~Zq)MUFWLT`Ngx{Lh2u*E{V(2v{Lhp%wing(TnxNN{K*IG)*ya8Mw@2+hi7<F<E{
z6jrsb0IOZ=TH?$Zf-z!}qm;Ko2*ehMqE^7#To}o}E>P8z&EcV_1d)Lgl*dH?K<-J_
zFS?CcE%mBpC(3-et9;hzY9L-|KY21ZFr>M^9!khLe7e)*bLFP)&0Tb>qj9lm?!x=H
zWNu!^IK&8lnITmc!Kdl+etywS#O@S#bkYyZ8^ySFpkl<{LG)|g19hq1_GgqkUcuiq
z2It?Fy0u_2|L5<N&y)yIe7PF&22n(<5qA(zd!*}Uf1VsU7t{pQJ~y|v0B4nO)xy-u
zqPMJ_Toi&YSb^B{n4bKwrDNE%;&GB4%F@8lGDs9llUm;0E!ySit$lnHR*L-Sk*+lP
z?S&}CqmdX^%L7V`t4?-vErKYDSjdMJ^^(2hD%8S%RL5=~SvZpcDrWN54Bc&m@oaqg
z35r<JN?@PVeii`}nCdR&yTwf}uoY&Rc|J|{`U<fIQ9je>RsM1LlpY>2GC+|nu6Z`N
zB?wUnIta~0X6f1eJyaLKgc_zf|IY&lPsh&k1^6<buh~m5hy>j&B^YKs+*I~Pz0ruP
zHGY8a)ErR0I6BnRSB3b9Q+=&qDE3^mJ${M$z&YaCTgzuqBWHVyHp?hW3)E$xIY{2m
z(D^VSclGJJ_KZ0XQGzIU#Iai}UTR?@s87kK1Z2IyYDEKqR3wx27k_lq>AQkHSId-m
z@#5wB?baWzQ4t(7d?RA-l}{=urPFdg5VCdv3J?`X^5RsytU8;E$XAScoH%00Mylcs
zK7|7$%<lRb;912hQJ>b;u^&$9)BUn(PZ3D{nU-OrxU93Pjho=7&09-hZSXNU0Pd1W
zyAu^U_Lo<4nCesTn`cq$Q$wZ?pi_90o?>l3;_xW{yLa(ZRDatOkLwW}N2Nyss4O*0
zU6pVZ7E-4Gs;_(Erjta~Hr?EzVQ;bOU|19uZFpEBCkKeuq$)!m5pecqEQ$CGs;QH%
zZMQf&PD@7Rt;_{wAHr5S1Z`Q#dRAO8T8^Pno+?+bKC2pxc_(}`8z+lT9y)#H`$XxG
zIE+7O99;#1_qJO1o3%+WF)`xJjsfl4lK!)1m~ps(tjUDxUl`+m!jI)rWYa^P2D88Z
z3A;b0U3P&Rn^?q&p%B877n;$SbHO$>q-(5#HjrN~iMqoQU@8rnCk|RE`A(Ld1b?iR
zFp*`TNf)QRGGrV%sA+b)=L*p3UUK8#%Bb#37|_%y031Jh8*{-^Eo^kRiYiJ&fSV+M
z-6yhez9qT<etvZoR{%3812XaDf&!^syD9K<o#G33Q$q3nS-5z7P}W*aLOf=_&-2bO
zB@<ErkzAI=ebn-Hzf4B2XW_BDYPw|<S-6Axy~R|TLtumpzuKLSijT6Wb;I2n*p<@@
zcN^X#q=1v%bu;?q+3v58KrLC!7O$Bwc+m-yUlx+h?&FEhR{Vm4J10b<zKSenYH;Y_
z;$bnCY3<a^<3awacLZi6M&;42Nx0|J7O@yZEA>-R_V4TV@M8IkZ3UNmX3qw70ds`B
zywaz`B!sgO*aiSt>7op^SR&pP%j$jwZg7$y#0hvCGMzRcu{eLL+4H%h7iyWV(c5Cr
zPyS{D9!4g$;qHVN9DkO4Ef}Jb0)-KXj@4?_cifdQL6lFCaM*>sa6MJr9TeFywkT~}
z5FQ2^AESi()k(pk)Vg|HnU67de;_Q!^RGi>bBtxM&mk5d{z?^=lTq;~!Alj*U{bW7
zB#9N5dcUauWyCgcm~&P&%fV7l*r#XW{9*Eg=%gbQc6tR2p2xz_^X6YUo;-k|aE?|l
zao@b31b|z-xQkg(>XQ`O#i0=6L10c8a3x(5!j2;xM<Xm+kEI8SimNsgpD34(n^Pj0
zJWG=WzN9k|R4O{rA4#Ef4*;)aCtfS?hIL(B$r&|GLw>k3l5;|xOAyY`AI-MTT3pLH
z?mGJKLPY7&E~uP^LquYZapO=@Y{`9wiWDb#B_<UOq?w$*G+|k)EFnX39<}@)m~9ik
zJo?zcpcRw^HlZPH`)4o1+#`^(?X369gzx7#*q8aNOfADPdWgMdO)967-A5Q=B@Hjr
zx(<U)-}&y-rcr|j(M{_au;u96zg|kdcHK|)A<z)Q48yHi!Jz+RB@Z|59*NEluT%|S
z>hT4aA6lV~)_cMpr{-7Y40sk}@}3u!3Ot(Z{ZFj@*E_80Fu0AhetG}T4f*sF_)e=R
z1XSlh;?aAUYDKfSl4wR!SAjVC1W^Z%)a@WFJCy*{@*AkpwY+z1iE5aHDxkrpxh)v$
zTQD{$D1;b-TU7d4Bxd}}#&oU@hKsZWXg4<$teP9k)3I44Lp~fVj+Odt#Xtq;yaxk*
ziJQ;mAcu*TLVc3Zz%BQJYgl>H1DfP?k&rEbx`j~vm=8~&z@%Cw5&Z9-&qE1-=1cTD
zATz{RO%uXNR1ac7=_1>5`v(S%#Dh|iM>|l7xWNR;`wWWcRqgbx-(gkCO(+b#RxNAa
z{*?X4vsQvR;{11iwzywdltAVi#arMsYRV$*YI61Fhmu|u6Tt4KhSTny&jB3DEq>(_
z8_{n5F=JKA+Z*2~H)s}`z)w`4w)!I*B$1wt<BHf#LL(UZ^J#|5H)*bmICD`GR3QQA
zB6IRAXwdGl-+h{kdh<xDrI^daJhg#vJaP+DLJaW&r+mqYPh)a&xz9ZhNyuAOP4&`V
zJ1pXcazeP21tv#$fgkDdsAqqFuq-miY*I>Muykz14pozJ%52<Qxl;F`mXVBo^ty^=
zgyoO)s<96Z3x4&IK;lu2&Rc(C7iqCtu$ao%zXM~Ey{!ato@UQ1UPsCliLS$>@uEKY
zE}4xx@+uLtp>+ZyxquT=XilB)C{>!-r`Z?P-4!>_6wz%WgSreb>4JN2u~CJVWlKzr
zkH1Fz#wRat^7B71l1u^io=&k6d>-d>8Sj~ROMY#5bJo#thzW@NqZ^-^jW=Nsv{y_z
z#b-!Bqd6Z+8Hubw;`$M~=0yj=Now4paIMdM?W{l~3AO;za`ahDCO1`EKDa+hC`-aJ
zATEZCFE=eXBOjbC4)X|EMb)oBg&sQ}8022N_hLOvo9Sk-Kf#Yt@3dxKU`r))G=2K+
z-<)q)bqiYKFFSM>{#j%&)4Aw|mr04cT?ekDBpy<EnuD~**;o*sq9j9h?_CaMC-BEC
z0BewvtC`{DGYGypVP<MI=i4$+k+dE%QcyW(`EFC$rMC<>JQO?w?$~)C$ZRyNv=qir
zd;GYYwbEsB{Sh_BIf!0A4_pI+3&wnxosQMLC3u3aH9jPzrK(e2Egr9WVn@Sa?rV$b
zH=DW1`c3PX>y$2w=ojN`#?If`lgG)$-A1k$o!WMlnNMXE{o*~R*JMhxddRxIrtaiM
zbGuBT#nez&e=p@yV0c}#Yz|~yhq$@9rj*6;bkx*YBIEC#@gZETvQT`;D0K+dv&*!(
zEB&?D^$#Yies;p|@*xjw$aMb|Bw?(~o?X-xLSAA+Voso?oP~u&8ET3$Z-)!dfQ^`-
zG~2pOg8>X?n}htulRcV#RQT7(usjD_6PXBs4X&Z4(%Wgb?n*I;KMFT(R!y{7{S0H(
z<Vay2@zpDbHpExB#w=j)Durfq0FJ@EGW+&-6SeiCo>q#@yy&#6!r~HWN9OT%;2YEp
zSecgLPBb4$oE`np{&``j)VY{7|A|_O%q=({BpJ41=ojFwiw}RRbSqt?)IzGcbLrFy
zT+TF$TqY$ek!FRCFm-dz+p|;^Ua7>K?|78qaW4!hnP<4i_Qp5f;un+5r)ySZ;C+Ob
zIItdHB$0fVXL~{P8xqwO7k;NXC{OL@$M-=;Jz*joB3atb+aub_iW?5#V3s^6{eDZ>
zYG3h3<&yYt@3U|J2*<YcY1vaUp7KFd9d;e6KT4}c`_+oY>Nj56O{1io@+p3|8i(ha
znK$%22bz-0*sC5|gY(k3()dT$Z0RgEgfbe}-8pwGjJplFAiTCh$SluQlyu$qt?w#R
z7&M$1z+Y#a?)9X%EQvoBxO#X-s5p3?*kZR`hU0HOGu0#=jF;T+_Q-d})yK)M|4@{-
zdf)XoTQgA5QmsfQ<`xFgwOK}zE?qq^#yU-)(<b)T{xqPO<+Bn>zc6=1sxql8@3ZL=
z%SMtH?a&7tgR9uf1Q_O8`eu`8yLwQoqKq4a8>Wo*!g&YX7$keNW?E@A1$C|Nm0av#
zX%VPMZOiw|ORNnf;kgC94X#wN&jLI=l5mt|^E=!4%3fwC%uuGqxI81(*xlve<q*Qw
zm~Y4`6Yj_BI^(bAS#`^6I=YUof;uL@*}RF$aA75WiggJ61Cdk^;lTv?6Hl)_Z(?Kf
zSnruN52$71jD+SyIQApBI5>{g-)X9fk(<PJ-)7|Z6|Rc1h(NC*eIHvG&5<&x^m!UO
zOPX3S64ofdvOL1!=gkQ9cZOOhq6>vsC|*$<9d(3YLM%#o8WlA>C)<RnXaCrwNxXpJ
zVs}zOg$zeT<)G3B#w;;A3-_4wxn`R*zpS1wQnatz%1fR@X|Clpsb7&Jw+kXp77PSJ
zi`VS3Pwb#lT<clgoBl2oWF4^Gu=R(O#2Jt0FzOJ62E)%^E5gWvo#(9sDGvtrXFfnt
z!IWs?{Ix0fo;zdv6xo**ACek8SKSX&u%k>cm|7`;#w*y;*hKbb6;gxPZF!3QD4CuM
zwufptDO7L|uhFd{0WbAyx4?U176K2bZfnP}3x6q_ixxt$_2fQ$d{sS7+)*~_>04P{
zCbxnZLg|C`$Fb0@z~;%l*4esn&gse_W$|E4ss*zK7JD2O&KQm|9WzwgWO<uQ^i2V4
z3zLi4uV>AyS*)xzdXD$q*Cu*kgwr43IfSUROJl`99_#-e!Qe#Z%>A_m-P$1bA!1g2
zCM0feMS~ALosuN_Ew-L(`6V|JZm7ukJ<&OE7i)4Nek<#1SAk3s6xLUaNTs!E;-8Ja
zW3X=t{-MisBNOKMS$Y-I1>=Fkf9!#BZ&ZKYc>QUcZpzi-!-N3hPwTT$vpC_3Yy;H0
ztscL6U3};=GRzdV5qN<V$(Q!&_9Cv16swh*3){7Q?<-KtwmHY!G&&8-LryD~s+o@q
zd{;$vuGLU|S4Z{bse8*GL<Q}9Pf8iLFK=$`vy1#8R~xm9KWzrb&gO{u?%vZBS705#
zYS6_w?r;vZAMX55IKw`*ivD*Mx|x{S|2um67iRX-Nrd-TDy$dyk5<WtE1(e4Nc`mV
zyEx11Qt8;T0j{*}`&rs_oQCs9!2KKEa;xV+;8Mn!yOQ|mbz~`;hK)@Y=HpvSX_7w+
zDLIZVE06*$qBa&oVoAoO?2JkjYSNesBkRHV1ymQ7lAjb}O0rt;o2vHh2oQhKNtV2u
z^AJTSl97#Bny;yPb;0<ie@VU?tu;qjPG1Gv;m$sWrWE;gE)F#UcfZ3HSyK;srU8$`
z`ZgLSM`;i%QyNNKMsMx7B>aRCPbmJ|=bns+?+zYB8;MVmhH9xMedn8GGH%A{)0W++
zB6^-UF8^Gaiz>d_G&1ycWUZVEQ$*B^;j=eauRqy=gI%b$i&QYxYt=n!`tBvObF$VX
z!R`~^4Xw{?i=y?P#u)WQ!!ti_4xf&AoWFHXeo@?+*;@9MN4Q60zj*O{*W0)TTLU`N
zi9xnD1V;CWSj}(gD#6AKd~Plc^KE0X70#gBI&)dCpv3iB#`ny^qyz!6%dc#%Ok5pu
z8|}j;J;L@cn0odTF?aol)5^t_ORhtU2-(HaHH&we!QA~wZp4A9hyz#T2U{zPN#Z@?
zZTaC31GTDc=Nn8l`=gd*d$(YL6c4nXzII)<#YFiLgX{KNpYc1hrQ;+S@5Phm;cnUY
zyiyH=i+@$$|M@~T93f)9l<#Ev;|vTiKl}7P_{WW~9}-EP#x$hKsx23Ke#F+^r1gu7
zszi~2N8XsWv+cdnK<(_(e~@1HK)f&U{66&6qm>CXl6sjxG3;5g@|^56XPI+zi20ao
zkC10YrP53Hr}+LJ(q|{nMh@DpN$O6OKW9TsJnf%=tG5GVi>4Xi66AvmnJj_g93HSq
zQ|iwefS)Lb1iQEpLYsU%roO$<yxIPNp!XU@RCr1Ehs}tcqRg{<s0VUBkp7cfJQd67
z0{UMHG1-iiZt2=S&y~lIUtBP&Fb<!tO>6$oC-cvD+NTk)Z#u+(*FScay8$_2n@eWy
zFYd4<3gHg7Klzh8tS7c~S)gPcNff2wQlrqpb<joO2r^vFV_=?S2`ypzThHPf(vP$I
z2~>15(65w0pV4sG&?3M#{N?&KcrDi>RwkIU(5rt&z`&sVp!pk+URVq?iw@HmafI`!
z0EADVOa{4p$Ui{%_QX2!)M#??p#QG#2MGmQyh9wuTMZP(;h$vFegFP$Lx{A-@YM_j
zj(w&32hB9#FqgkV<Xtbg#RNjfM?+@8?d^!A`_HAxPgX_O-kb*Cp5Qwyr#^zrFyrnc
z&z0H!Zln*w@R!`$*%sO%d$XlqoD}HUFdvlA8wKL<$OMp7O1h4cE^v>P_JM-)!1o7d
ze?3`f1|3}+^7Y>Tdgs5qXn%k_{Remh=YM?0ZwnO2)MZ#mq`w`EQ<`nT(Pd`xnrlQx
z9asRR9KH7JV_}G>-yrwH%O$(alJ9$OT6ZDWGmbK%;h!p%Lpt!M7tZOW{55C%mqYT`
z`MI<gIW2?D^_hRcNB@a1{pKAXL4+mk)cwNpzkIJ>Sn}V$s0u)+s>!jPYQOyv{~JRX
znXLq15dkB%zsMo(ci-~Y*ZsdQ6t1%aFmwHzO_A>3k8A%PdQ*AYKno1J2G4A!|3;4g
z-;bo785#32`xrU@+sG*3Ft;f90BYVR?L}h;QY2=~IxpN>%GuU;{GY#)527r<YhQlB
z{cnH0eHOJJvM~KN7u50KzE@#3h_?!oekBhxAoHnJ)x$<4L^gaaWC;jedDOzCPcMC!
z-yA_BhNQ8PMO@_B=Y|0<E%}2U48X*BEs9<kP`L>h7<v4;um0vIdaeXN5WkE2KO4hE
zDuk@Gf*2=54|sZK#0OTWLIxe!gC4t$aTz6rCTu&&<_QI~&P0tx$q!+Dmvr$EV1!8j
z^1z6%F#9~K9g><1c@U45zkhxcS@F1OqvN7ZuUG?%HA2yK9jgyob_D%g3w<BCm)D-A
z!}gxJFKZD7FX9gmu&v_?3<~<8$GHx}3H{u$r+dTy+wS;txhUu$@SV=&nIh#sSBVcQ
z0OZ}(iGCYb5vDj^BfMijtVRcl!^ZX;NE-AEm@qz%(ds@Os$jR$fLJghw8mM;Se681
zA%v;-2@GOD+-PF^$Yus$gkoTxKOea<RkZ+()GNT+NXYD@AG-2cu*?!?^qlC>6RbG?
zYR=fJwm>INv}S&PDVQeqK>Vr`qH?crbh~u@cYEmDGD4_dR#w>c=XQQtgknc|!_098
z7+Ke|!!jjd#jyS(I~+*P@5h1CeDK)ipf2^N0vpzW2P8#h9!PLeAo1I^jO8Q2%uav~
zX61g^^930U21?Z?f(xxf%pUiUFnj-M#J3Of0$&sS59ECk`nR?Ce@=sTeuRu2op@gM
z*EvHDYght<{O?mQM21kbVFDNDR8;GWJCI(XY7DZizyGQMk%;o)U(Ve2Ns4(D_xIef
zJs?qLs&#5TITrB;JsB}yf%uL<Uw{jW;S1?!b-&r!y8ahGwE!VWtN{1A+YcUaKa!9P
zZErR#Ir&v{8IC*5Qo^S2#;g+{r08X2EHR5v=wX8WN#D?NPd^aKm1GNo=CdG-`y)!%
z3sg9LFe9*sjdV-W<m3>H(41hAmUWN5&v8({={Ep>Wv3ASMNcBI+N$jk^QC~@lzoVO
zO5qsTkh}uhk9s6=9jDn_{K4}f{m1GW80{k3f1$zNlIxcd`%mRJ^$(yu^<)%SI&u{-
zL&SBs1&q7O<I)Pb+kpLGhnuSHXG~V0+|Ou%5RDHU5hslU&-cO*V(axC(FW1iQ%oS{
zTB3)?JbEy%=Yh7qKVxHQ(i$Q@3@c67V#?-!abmOYw?86K*f@!R+hOu-^v1o$0nj-h
z`zv<pEu$giAJ;8_9~NEx4m=OrJlk^!RGC&enUKRoH+lNplan8ia02V0<8LN^BB9Fq
zU`@~u_F?3_eVUET(s>nd+-Ul0zD6!WMEHuq#*9_E!x=_-J$cd*@pj1*0P>&5$vX$0
zU7K72_rYd>hhxvjE_ts|DYu>sl=ZX&+Lk0q6!(x>2A3z6Q!okcT}Zw=rARR1!)Uo^
z5g;Wk^GJMX;TrL|H|Vz8hK810H`Y6%?JPABaxRje?hGX9Vn{R@fDG42QT40+kXUPr
zD5mF39Sh|@9=Pgx?pcsO0&C_mWU+zxbqk3!>$<yid`!-fa^<#Y9k}A7w(`Mvy1prG
z7SxdD`xH%28x3psaj|<QE`+t<K!&?tt|QdL1z;(?hEsY|FTBl9G^sh=D%M0*oGY~(
z=O3xdm&#)8iHCJ|+AGqw>G^7$R~m3WqO(P^6nlO;mBXPi1C#`j1lxn-LFv~DSrL6=
z+~+LCqe}K{?(K?}O+`?B@ia56z(n#Gs3*2=vkK_(71dRk6?*Lev0RDIxl3e<+4*u8
z7yI-;()Louyc}crsO){aG&;#GDA$A%j6sv?1a<&|td8oxy)_|m5pK6rPnHk;xm&-v
zkl*B!hX*(VaotiBXhh__V7Ufn&<8mf)^QOSGFK&YC6(Jf)QWpJg-AvNbs>XDJ_VeS
zhY?%tz;<5zj;=Z@y__1?dBupgA(J4rK8ZrnuwU3~l@WB<_#GlGhI^s6G3==}_yyUo
zeQOQzo2o)wf4_!O0w;3E4z4sE8K+mXjh@d00vRVksB+MXLxltAh9L_h2^$=S$-cxG
z;qq@Ors*w(TxL;fe`Er-cQ5Tu(I%k_?Z5<qpMDiI0ei^0kn$F`GkR=yK#h2l;Kjze
zgWcdj`nd-l6sw9Xa*=WnH$f<Jd(8n@xci=99JIvWA5Fv4v~aL&L1A1M_ZqPB{g#U7
zh=Md0x^lHjJ&`~(H5a}qeRFNT-GC^D;o-RZwU-zj=Y}f45ZEhQG!c&|Gf=IL&^BWn
zS->nR;JCFv1@pMTXgzXgC;N`n?rcvsiqT9f#v@QTyNu$}`zLCGA#t&C2(iA-@Fb|u
zYh{waD0Bt_O{m7?V**5DZlca<EU{T1kPdj4aiR9LSp9HFHA-5EwLap=McPl>Ti<V?
z$3lI?+xXDT?yaXLdSH^F$N%^zIIO;7uRba$R($VAo~28A5>37+L#ccM#+YlLl7e~?
z$88mbw{aFu?YB^B(1(7T=A^5YtCwqQP}2cRRm68WNRmCKcIdqPj0E|aagk=4-2l%d
zs3kqSn7T#N#AkJmo#KiNk{P79)RBAnO?Av`#dBKHleZ`WdJDm#c8`gYy`@YqhkqMg
zHkZ3qmP-}ag+t_WYr}y^4oi@jf-ECLgw)2Jg`i!CM?|i@K>3YChWGbfKM-eiRu9~^
z@;g-hlcG3U0;S7cD^2qGi+@dTl-~5hjxFFg=Q7{CwT$>S25b8Xp#oQciDT_XozIz_
zg2T@)9-mckj}m<(anR)bV`BD|Cl;3tscCjwhxA?sCS{#Z0{vI(2O_{>6+N8a-y`94
zZ1iR%GA}l<iPq~Q><!)m`oZ+aDA~LRo1z2>p#(#QH8QekQ5Hr`z_I*Jhp+s;;5!}Y
zf9v7-UkW?Wd6`n&FI^LH6}vEc`z7fu<_|vXTEi$wRUkCnf|lXJ(YaoclC<sQQ|o^b
zUv)2xvIWyId<l}zDZ`1lib#cq_tG--6&|aTKg26MUMFEXimkp6vawTTQ0mYzyf`X+
z@T}?4hltH&D1Q9xW~z7m&&)UBN0rRr^jjpoJz<$tZnrb=Qak?VSZK~Q^{fUY4$siJ
za-lx-7Nd@->7&w;{C3{LPpX`uw+QIUf!)@X*@XZ>r`={6a3#}IV*b_!wj5lpKz6Nt
zUhvVhl)g0=MtY&tPWPb=6?VO+WQ1i<;uan_XSx<U1JPlj_&fWp1M*dw&1CMC8f1`y
z_JmSrHqNC9CQf5Hm1VJjEf-7+hoQQGBzZ>IwawC|J6z7i(rVQF9xa1bQ2D35<BVHp
z*ghuZy?7W$=_**)q+0SGB353Kg6_VDxkFs(Q_;NSO~gRRvNauOePW7c-*gy?g0=}|
z)Py<~yF_cZKTIR%@>L^Pb0+R2^pxBS2^UoMtG`tAh5SKvyR1I<9F?f+b;pr1;aviQ
z<B1(gNfwLGcbjYw7bm}rCb!$=c||z~$odjmt~^2X1fkSdX=m<HM|~{e5jLc5^T;(!
z#tPrpM%rczsYeqL2N>ii<0s8IcLx@bKbCO`(8ImPFPt@%i7zGcx{_HuuL$?9HGQfr
zoI~y0uw9Pgd(!sm^PLsS7Cwh_?HR(x`={f=MU;PzGO5T{42!!eTIZ!2-E2NkPwU9$
z>j#xg?6@(DwbPN*-q+1Boz3~w=5>o^1U9F~)M9nVAF*wU9W>)(s~-e|7P@-f6>l@y
zws5Q7h}o!fd7;8Cnx-d`)}p9uH*3dh?{6=$u@hDgtd>72w3D`aRCH~OrB2_P3ztZZ
zcDru(q?}LW5!C%k=3PJC9g0~y_DG*)jcbT-WmpPYm1R)pL5zw<Yv1Mxi+P!{^rq)L
z@gj2?9ZIfa3fA)HUM+hk?}|>LVlg|JP>MTlwng!?QJUj>l&PKEkJebNrSNdeK&@K8
z8(F_s+qbV%tf`j%D;?M0B`B&6VTCc*9M)2{pWu+DTW!DJe_{0oo!oQ0eG{pwrSN^=
zw?<|2V07D@GtX*j1wQ3iC_83X+S^lyko$~>$TslFBL2rG-hl^tYkUcE2Gf25g@Ndm
zIaxl}HvJK0T;e=o+a>CHfI%WAkLcOam~}ypK!0=1>$-j@zOtz$Sd{Eu5}teC?WayW
z=!>kROT^0<*D|l%0zo1*4Wl41_$^J@Ya4k84<lxQznkzKM58k<9PC#6Nmo#&1ymgL
zDRP9;qN5~B6y>@yd-yJ)m?oT};b>xAW0*sbSbz(i>%e!v)rzA!cbiTsTBzSzVCcWp
zb@$0`75*=|dd*`)9DzL)H3n&fD!1>JW!BxY&3hL3i}$Ydo@zXq=1aVWVd0Ey)m%EY
zoQ{6Q$<enB5wAj~c35|n8zN?k8dF1*npu{_Eya@|LSa<82R)Wu`W7y|b(Y=I?T7;`
z&2Uhb7_`i9c50(RHY-R@-O1qC2lKvf9Pb2dCfgTvQQs1KX(=A{iq+K77+L2ISQ|4+
z_3kUGmC4X<DRAmqS==5TvI)#%<Y1MQjL*9l!Qf|6Z8}RNw6j{8o#wc|D%x+eI32`b
zjvAv2Fh5N{c#!lh!gah9|L#D_>l4gFLKx0;XAw?MLQRhf>DHuU2fgKSB|kzjD*TmF
zbq{u&-I_VTG3|@x`#`NmlO}<|)WzTRFYM%R5fH#Ikzec2YN(vBi&I)f1h=#}pIXM&
zLSyO;K7q^1YaykS=*z?eer^)uP4|XH<3vWTePkjK$Sip~*Ki!fy|=rc;GUE?Uqbcp
z&(6;WD7N3WD{-J|v&3ni3-@?;`I7Vt*@`a5g}NP}ip1clheqbb30}#S1UqT8`eOkm
z%?Wk?u<Bvx6f~0#{>%mkL4*CSl(qaS^tzSfmQMrYBolH!$^q}>A?}9NTyd<i%{lgO
zBs|w>*;%RuI?fX$exsT*Y4|CcMBTl$u}H?f3~!q)hjG1IbVyFK?-yVl#^X*74l4>s
z)$M3=Iz1DLmdWI#Fa=hwzzPSs1Id}lCC$k}VG54GeZ-sv43Xhg;r!vBT1mqk`Bj&6
zZXJ@~!I0<V)$Vhak*Y7}Y<{*4E<6z#eNOT8xkbnFTIqCAr2u5b%qW@DL<mqU<3_YH
z^dC)hWE#b^^!at^jJH15P3=Q@d}_wnjU9QgnzV$|nYe0;=vRGl@=Aj0N|lkZN-6`q
zPh6BT?J4{tbYym?U;Ho&TpJZY5%ft<{5%Kj66*VTzS~FB7H<FsvETVQbf4cF$Z!J$
zYDdfL;x`w3-OYbr*1D2EaCy0-<M_+vDB#*T|F4_SJQYHj(zK~uwBBIoRuk#XciD|2
z+`C6=QS{#UD5G1Ao{?rlB&#C`n`5AY_bG;6AjYf-TS|$o00S&%Fo++=;u?j2QqH@)
z!B2~$Lrpw|u(-8mmt5-1(M-vHB$8l2`c7F~MoW<cuW=gdoW}=Kmm}MzCyeqNg^fiS
z9t6pqoq&-HJ)3C*C`WrC0jeeFa<V+5ftWYyjGI{aEXbZbk;d|ff$;e}n7uP^tvAST
ziw(t8*fbvmglNF}S=nk)fy^KAXcJs+QT|}?|1#PvoU$n5Wn9SzA?wd|4DsQ<&7c-Z
z)Qt((L-IW&8TuYm-)xrkEBc8VPY;3EA{uAmc6+H?FC(Zn0>lva;5dBb3&ZPZywe;_
zv+4eiwx(YOzl<Js%MUASf5wA-I6}^Wur_Bwhbg|AqOYG7l6uf{U;G!WP*aX;9xxal
zQ@c9>T7134PdHct9e8!_YLm>8osp3+={P$3D%QAk=H^DV?3+FDNWf>h4nx5kfRYLM
zYwgPD=;#nuyPVRzV)sM|^8W8i)z#(h9CjIr+i(}=$>j@|@}103vlS-on~uc?oY;^O
zDZ{)seP%O9(~d}$Dg{}6TI*7nR@mmU-ge%{>q_+9u8M=wS*Z(_=B4x_F{(X&3~7(3
z8*2xq`%A7%JRBcw7v(W!w3~A+br|vy@!@j@p78jHi!-#PMbH~}4`@Ek_rs&?-N4!8
za5+vmLV9)FVJ<muDK4d7(Ig*_wLP8nsB^|8i$MLTZomuNr>nvyUYV>OwZ}U2*oWf}
zwwNDm*jA>in=3U?kOq!V<B!se9t$sTFKew!zly>C<etMj-<SaJ^*rDutM}&;-^V+g
znBQEN|0pEjIE5&&<fK~O*Tmr4?Dev;kMMW-@sdi9k!(SbVO~K(iF%7>$gF5rtF2MK
z2Ty&uThL@C1lWydb$8X5M<LCEV;83tb!VS1d(VCiXWJeWcv{x|?t-TZ6OfJ%rf89N
z_mX<XWKuW@duF-OZx{*IM!FXTP-It|zznxV**wI%5XicF@1=&^Ij~5!KuMnNL-2<t
z;XzGdOb5Spt!&F9JnlrjN)h2awaKS@Kxvd=q-c#0Q%TXq1lprHSZ-&%+59R}rX)}L
z(4&f$HO@XV!enLG7hrbhb~f$-!QWQ(PH5s$ib|l<w<y;&Sl+L&ahiiuEf{xz=0v96
z_$iO8YA8d(tok!)+h)eP;`%bJsvmziW%@N&KMs$pd`H;DG9qc^Im5&$v+Co7_~d&N
z?cAb@6xqLF&(=L29;t4^UUrvG!QAyK&ywZGyG-q34z-9~&=azp7?3W^m(-&r*WYmU
z$M2D3Eaknhcz&|Xv!w5p$6r^+F9_`4UOZJG-&VZ*PY9KghW%zd{>Z)&X{z20z@>P%
zb)^VG9wVCKew%9rU9{z`mxb=bWacOVo%1#7$E#z(dp)E4do|OncU-Va)Un!MW==3M
zmFo44kUt{@X`B*zIeQ6yw%Mz*6JaB}sH9ozddl0mB6u0pF6}4NFucUj$SoXazr&GO
zCwnT$CB3&f8^hF76&^Hym%we~cP>tpgQJpHrl)*O5IAYRv<x3@Unb$t*;~}|c0Gnm
z2I3+Phf)XkAE=qcuHM#iwLGJxmBo9=xm=;vvpn`SDab$V(_H3@!S(^dH4mKtiQaO1
z^kSb9aBXSzeLROE@stA89FJo&KDZI~nj?+Vz+7S@?W%@>XJTGv&w7gTZUzY+)Pb4A
zt4&i+<f;?T-&#$Hl`|{;>~yjpjGj&slLH1{Uajz|Fc@sd&ZwWu7H`YcIW8EUw>9T_
z(sHhpx0RSN+Eo>`H>SBWg>EtN1LKFzzq|mJ2ZDyFMgj(#ZqyobpOBf3$9wLd-sMR=
zZHL?|$pKVcW{$>?TkAHfva$^bv#4us@|m>uYOy!?GY8g9=NWXkx9jfasde>Hx_F?U
zroEB*Y0Pgb{riH#gC?*1hQO=8Mj8s5w55Q8(V;aB^RLal6+tP;P&nU!@ZpDF$5OYz
z-64MSN9-FXgW-#w*5glK2Ki@yj0C0>Kk>mp>V@F*&LOHrsl^$cF#>*TUbz}0!AL-M
z#1Rr_=SLApP`m=j=sEj~MPHKj0u?%{t`xmL9OMtg$y@P`3lvsh{ic4D<TvMijK~$I
z9!;x6jHD$mZj75hU$6MY4JZKmwRdrREhOc;l$P4rqmgPs{-_IsfqJ*uBPlY#OjTq-
zbRsk)KF6>Duzb0~x;kI|=8H{xsK_2XdoK0?_9T((C3k&#h~t5jz9p8c(capF@igQ&
zZoMi_T0KHY>RqL8)aIqgnfJ`~QoI!NTEAl(5Z4%Xf>%_a;;soSS>Uce>?FxOS9cwm
zni|^GFH`W%^8^HbgN8gXVM%n{c`8<}W`lZhh2X!?KqjjGd>s<7d0#b)YmN@l3T%tA
zZ?UZCKvzYo3XR#uxbgp|2hqO{jSe9rT&t5Wej!>H$Z;2bxTNU?-TdHqFr~)nG=#MV
z_E5D2bke2-<y$xKcn{f!(UQX@nMj5Gv{rrjx<ioXa3__HBJhX2#bIt3rZv2K1F~6T
zhZRNRfhSVjY#TWia;tL18#mA{kE?#!C6^eIw<}j0ckSgvOkXxkPu`z9-trJv+9laL
zTKeLsTm=j?0%OzRhw&lKvPaH~zIL8Etb+q5K+Bv|g1<}#;JJwk3;bbx<48bLi-<D~
zY52}TF{)cv?_E50u~j=&NuhCd{zZ-F$a;tJ<-OPaakejG<{zF$D8SfyX|RaUd10d^
zbK>d`cr`u_z7teKcHxY4swlEafqF9`bF`au;rj<4xY%n)kRV;Ft39+b@4ogv)Q_9E
zRQZqMZI8?X$BZxVnfVL8RP7x?x~sP`E#^;!vP2N7uQ<Sav6cLb!V)ka;bC?z)a@Df
zJ}X?#3S4M!Ud1OM1T--)ePx8tp9bN_$dOwtE{GU+b1i`FK_z^DiBvT{rfw>96$^<V
z@i{Bv_Zv(PnF-TJ3IF9Sv<jSgYXru}8Wkk>qU=KKL(o$Q^Ys1%eK5KQMMX2r3h1dK
z#ocSC62ma3=4;~bJ_cmcX5$9xP8eetx`IfxC`Y7|ix%~{c~YHt@yxe|tuHm!`#B6h
zQHkIY8m7{>4>HNB0)=CM7wQBbU_MOp$MO5Xy7;K2G0!h$+Z=~^fnH!vzWXY(7u1ph
zqbL2=5WQRA{3u{#&LxJVNEGW4FI4gT@L{^GTZrQ-L5TE{`SxoW%MmaSibqH+iiJTF
z9JH%HZpzRhmO<|!`i7Rk)LE~0^IDZjEyL(KDxe>s*#0=JYDS}afpVNeiKg!iD{$Jw
zC8HD#>m;@@`)SK1unj_P&n`Gbi!C4?fdK~xXlfBl0>yUH$wc^lZyWB_Jy^`HgCa^5
z_6(K2Rc2eK0V7tYmkY!vahH3jD|KmjkgRYW#^ar|%jIDA5y%*hzXo<nk{)j(IU8#o
z=~pBY@<KY$Q5FD^TJ*F<n#~A%Gcw@9o0;h*#is{(PaJDgKd-FO27n&t#+iQ0@%<49
z&{*K|QguoFxb^MVw?6^ZfQF)GrcNrso^ml^!`dW&xA!!q8-2h8AoDT9HYWyoAD-#i
z03=q{2zcV3-bz)qxn-_GTgW)2vECeoV5K(J_dN4n!;D|57g{PM!-bhT6PrY-rUIjW
zot@DDG+F-zO`&Kc!`x<T9qzWE{C0pKhsg>Lo7}T8n_d8SAjS>)0QqSn7JLWT#@M1f
zR0`wlzK7XU%2!(4TrhaAf8#*>gV?ySW@ba$ajY~mokzXn2ol$o1#OU}b8Q{}SJ%sF
zhi<X7gZ+tUqEXw&#q5v1PwcjTyl%WX9_3Xf<JsBRF7fx{E@TSc2!5*fchH9a+K^Ps
zl0SyPYx#l4;VFpqj?|8VBSc*@6vl<$ZI-WmjAs|$!8i#&x%$bNp||&l-4jdUyJ8Id
zkHX`fB2SHrt`cpxN4hraUayCkGoN;UUZzlQ+Kg!&E<=1s+ioPsu7|9<%)(r$0Ac%o
zSwN2?eL_${4;!LUR6iv>U%k7(BXC;y>G4$xhDKLi&F8t)mz%4+9^reaz9LqY#UxtI
z!m>B?eIr%}5nqj#s@1sgfo4^XaVv3VdA~|X@ycVHnjJ#7?71=GQ_OJ(NZ#7I4H4WM
zy?Z3bt(e?uZ0Alk%C+s;;|Y}F$00b-7BbdH<9CVoO^n_VO_wRmMdqy#XgxDK66ZQ?
z9)e8U>Txb;-EmG`H{CSv9yPanI>WP8-C<+=CBsZQK>WE=3@$`*;d0Gy3&s;^l(&wA
zH^Z4?50LGB#Ks~y>HP#}jA#^&1h?5W;fG_CCr3i(e2+zhpA4~*Kqh3(%zbgO#+K(w
z72Z9Ux9(S-4)%4T_E-^ZF9zjceE8{>x7Wu<52k8Oei_<nuF^=;2|Cp>kiL_5S<E#z
ze5v_u#=@m|p-tJWGFf<l<D90<VpfDQzC5khGgS4jLVgsE)P*ohQdZC>?fgN0UXAdc
zQIdCyiW%qpc7Y~Bb>`?8bZ6Nl+cJ%^hYfq~k8`L5Zh6#PXJnWZ({1djKIrBG`nv-w
z4_7L7eNu~Lxe8$RHNF<rGr>YY@F}~KAI@q!Q}aNqVPm+~*-ghmU**<`<2-;9eSSo}
zRJ)<dYUBo}IxEcn7nMv*HEe`0ptFpcp3BSVfPr;hjLkxs#$evd9;TW7&JFr`dp~B(
z*V;aT!B(#cvz%wH`#5=a-rfSN52Kq~vJWFyR$T{~8upsh(vex~Z5JUC1mQ9N1K~NI
z^c}`*7C#@Q=F8{*Y}W0%$dDV9-!?G&T;;QvaK3Jqb+at(m<QMf9bZ*(0C3nb?z^`K
zD0XB5B?zR)59|cW>NoAoS~cgx-3Iqe*Vu|}Gde~Xty$*EWclQU<k_-STsT&C@{{Qa
z#;#c)Eb(NuMgK@|tGl~=mmqW_z&EaYQl^jLH&bRZ8v7?eUAgn>Q^KEDO}jV+CK!k6
zVow<v6J#5vk#jj1KwXPjy+S9qtxs#Rpc{{F#;Uar)LlGNSiOZx!dMCc?@O^T_#Gz?
zK@?joL1Z62MNmX^zh#sOF8dTQ<h&F?VrAXW&gbHu3|+WIA&8b;I*)P1^^~!DD)x*H
z1=RWyr*YbnMQ5eub~fD*3&uehos3vZZ;cGqC`n^PN_l16zU$SiuUB^1OaEx(pL&AN
z-^VE;w^5N%_&Ds&tVIMeC>bBwU3b#jlmn|A*SMAv1;Q>gTqt-<5XWn!XT#=inw9s(
z<Hh$`ov7cJ*;o4Fk{^D8hN*m_j6FiJxo$y5c;?wQ%Cwi0BEsqb)N@kbuG5kqC2^w8
zzTmE31oYe<>nAQ{a>eL7ON&re=!1^k_gHT0Q+S}7G0_VlM8=T`P_M=aP+Y-o0!deb
z!hAL@i#7A%4mg}0EWSOBI`>>z=)J7{o*I_h3LEI;BhCw$M~Yd^jLHO}MZI!aOxV)-
znw584Q4b?~7ewojM9k<`ju>|x6Yd2&&vOsOCG&Q>5s13y&m{~>Tjwnw3c=8kNo`{C
z>dU!kt@c~J3E6*yw$J#|trMqtfNJuAU0B=NBEj-n!xErMg9ATpW%f@GfUzIljKbuX
z9Zw78oodQ_M&ENg99=3iDrG%`StuzJZP^CYfkA5r%#j;xDYg_LC>xqn2mv=4OME{N
zZjF&F0OdWj9q~3t`hl+-<<?R4Etri`G#8hvxEXf-&MC_7LjO|QZrfJtH<NZ7hdCV1
zeONh<c2Y>|I43>eb`2N%{F0~4_e9OQp0I{qp=r#rD#eWX!N4sS^o!&nft*q+!%uNi
zg%3!5jT`#M#NNbRL^-Zostuvf{g?|%4$(Upi(%tT`5)K-UnDH(xA~XO{I#HMC@`$^
z98cs`F5;es4w2CWvTrZWzmVvR`YAJGf)QqC8cCH_qTR;uef4$Wy=B~^oS2mtjvl$n
zamXV~DYG(cWSKIVdX?QhnRQc_vUoO4-=1<yN!ng|Idqnok|`6@>nZUu(&>mQ3)ew~
z6aIFEFp}@y{2(tE@7bP|=TAKrCg!j3;_yM@cZ@hm?0uEXGnPi%<_CSu2KjQ!*JxgG
z&V4yGbDRxRF5XI`@N8N^{$fcSh9s)qqp#_`$Rh&aQ)p$mYipg#UXH`XeDM~TJx;xj
zy<^7SqKS-bM>kSq77iuRPuG=xiBfq}zuodmBBCZvDn@zR-F$|_7kkn39BJ{+wO5j{
z3-!k}P^FBiMt&|CRT15Ml6Piv)3Cii9LjjdGf(h*T;v&VDQIrGn4NS4?V9ru%EgV~
z8zYz6)`%W2ii0eD{2E?Y>$fv3FnLrDsRrCE5-;o8w^)gaDNH@mkN#X5S^IE1ko)_I
zq5`KV4}AC7lg7vjd+Gvf$=ph{e#Lmp9`hGz2Roz+NNOaHLy<0bX1^HCMuRs|!;wTC
znslz<T<(Q)DrGnhQ?P^x<fBRdC;9Nd*1&HOLe0q-{$FT<0)>xY0OGJg_(bYIjOB?|
zVUOCCoVyUAsxb%+%(%^mGIKQr>y%iJ(#QQ{RKv25+4ef+)$AP!A?x((xECDYrBR9I
z3?o@j0T=1X>#YAl$0P?%fZ;9``oENr@F8U%RRe5rSZM(f?y2#;F_zz0y~5*Qa*C`n
zk%nm<M(qbUbg90|;O2*aoI5?GM{$D3N6+hJeid<7;;sc0F)s=;Rm7u&MzfyokOZbU
zrnd5x%;_#;`M)@y|MeUXNK}AcF0nn_A@Hw8+aGmEV;D)x1Z{rRp~e`O+n_@WYXQ?_
zNoa-oKm*5I`{wYP27PA>iUst@p3u(3IJ7WwZ1Rq9!wq<-|3;ILvi<uDduNw$<V<w3
zufsd5%{BbnNx1EwFnrd`-T}Hn1RF{aasVW|P2GCw4NNT&YsN{Rv;-<To*v+CT!*;x
zNTIxH^5}xe%71>cf9*p5;-%3c%qjB3YxKXqU_heAFqjn~a;QEX`K?2b)fwN!VSEr<
z$KrY>N7BW<6(w@QRuy#SnnFoX(<3(X9D<_cQJLwFU$|>YR|Wdlt9<`vS5)9dW3k01
z>r)7_NX@k11~2j<O4~Tgz08tMart`-V@uIG$-uiQTAuurQ8iO8{5#?^XZtQ-bTRv<
zoBR{jddf?OSX#|O3C-tC(Om2S)?N}pOHCIh%^Gl_8*uoVIh;s|JeD5EiQ2_I&u2NQ
zC(T4x^`fG)$6RIC^<2nIGP@R@zlxxea}vL^+(2N@n+WVF=O+9ANApym*oSb+Ecx{A
z{`Lp?dk&$y96SK&q_V`{62$-OhW*EzLCCfheaQd+=jN+GaXJYO_k=c6oqr>m{#rNx
z^F=@{vTqsB(*J&z`hU26|MI^HI<VVX?Q(ek?79E`S^bxl4*yue5*EAQ{<M9+0Pb(0
z6u=Z=#UtLL&|ALp9)Fa<E2-=JkAGmV`!9aLLdME^n!g?DAY4N|gs#Ldo9y%J6C6?+
z2KHqEVuXj{{kSKRNum*nv0uu3{0AxM{18V7i`AR0+3Sww>W>__=3xrj0)n&`7si1U
z-jEJ7%8&V;Do<$xKs4KH4hM8W6F?jH5Em7j33qn#!Gvni<Nfmgt&mFlG$jwjAKKi%
zD7o7BV(kBV<Ys5!(dInMwfjY4J)}Jh11515JdpUD`F<jKP@rfT<T)2-fpFyp+mu~S
zAO?!tYh$zEK=<P3c24C8;tUL>aDDLpBOjwXpRdhOVsX{lvdfo#0tyiLJp466tl}T<
z-a|vAP>W!+1XooZnL=<$A+3NRsPqDfPIIF0(*}iF6xJDMmi-W6ZJ9t4Ey5KmJeKjK
z;6gFh7fU9sR2@pk+cnUtU$Rvs`Rs#dp1C}LW{5bj2c&P7pd`y&zSr1YL$Rh?kEDKW
z{apP%%$te(T@U;%0b#c$Y~|?x`3w}IxZzXF_aF)efM51GSN@A-Z#oCZFk-BV(D&A)
zhy8b#(@RYu2XsF?sn7S9w;?pf8cxU1L5Q<eGlP%Eg}C!-cIibs2$BD?`aLW3r`7Ks
z%d3;mJdG2pf&R~+@}${!rUr(vCNGXlvX~5I=&&)Lus@Hn&Mj!<Pn)i?q2&1Tm%nd|
z)-{j*%Z|!S&Im`C$!|Y2tkONt9%vi$Fy$=wHX6tTK-<^`L4Ze)3E=^B&i}gI5yw|t
zI{&Y8?r9Ke6Kr<HNj!J<kwMUG6T8}Ee@Si&ghvZ?6s)|wjT!}L@)fX+UT-y<#f>4F
z4$$|X0aN3dz`5;j;I-Tdd`bo?vI-;h(F2FbhGER<v;%CgS#V($EMh-H3)7VEGw@$A
z0A^e`ib^o*L|m!`zfK;I27hE*Off{lZc0d%RP=1GFhM?qWRJz}u?DiM-;-x$FLp~u
zR)dz?GQyg)gGe<SWHq>dt>a%C@t-e1o(M;=$lOf{vp+Yq&+Y)*ApgNfy019HS{*9j
zm{TpEkW*Kxo*OKQrZ6FrEo`m<Zn<})9a?5*0-gQ~?Wgcjrm{sHi&>-V_3&+-%fGYI
z|8<DnMgH>R;J-LoPvhy<fI=R)dbGAiZn62eq-8Au<KP;w=P(Kzi>f<oUo2GI!&#C>
z6o!WbezJQ+R31uMRPQnYRL>dgtZgirj|~GAl{#g=(#clfPSt#Bg7(c0(Rc_!dKrQk
z?8vXQk`Kka*!(P_an|ho)f|GmS1dsz)ZL3po^0~r{X>{TlLx8J9J_Z-<bZh^i8_S<
z=>+Wg+uv8WDvDoC%v80V!nc)WBrdN%Iey%+D+{OJlDOVl@rM^PaF<(hKxdB!6dyT2
z!)-jnH5C#;9eK~-ZRT|0l_vx?n8}tP29W8M5uk}FR2fxcc0T0F@1$c2+xTTpIsZq&
zdSq20g^$6~lqWKZuaX7N6vV{5=i7%=XFyd;fTq`Ss2jwp7FI>T<Bd}}XTv*f7$vu1
zFBpF;-0y{yQNv?E&zt}=jIue;Y_>WfzjG?fAyXLa0vsN^%?7&sD;8fC0krXrSJB96
zlsbpxrd%9o>;@$K3X-Lbn6*pJ7Je!&mubyRFa`2m%_#Vfy@D=sD8}^0`{A-?@unUN
z8uFFmTZd|%3q>(WycpMW(wTZ$KuF)_XnNp4hXhy&Ib}%0AANy17^iPGIYYFuZpbO~
z&IO1;zX}`Q8+Y|s7n1Cjl;eCIY<TS(en4cCkkf4{9#NsHY`N2b<J!l+=tjaWk(;8=
zqs8;|B-_`M{q*DxS%;YO4KCm7)VX6*pkD+ch=zi~uZ2HcD|}B}{+#5@Ibr$Kse;Za
zbQwAIU;u9zq1^IguoZ`x`v!mQV95}Sxte%Px(HpgIuKRQc55?FDN{~>_vI_5PtgXa
zgw9kzj2}sV;zX{!U+3$Xf|#^EHGNx&uFJ50i`)-&1G;4g<7Y2Y+L+ok_Q<U}KY9}`
zM}Orjeu?Dyt~plH>a_S4sJs+=Dc!AIF(uqOl-RwQd&<mNBjU=Li>s>T8Hl`T?6p?y
zdzdYrxKlp3kn<`3@?!irhfpt+lDrL@Vt@;IzMp^ud4^G)<plklH;uZj7c<-pE57ng
z1X`mYjQAfBm}LLwr7Cz!g`fB<0e-&~xpA#0um<*z#1Zsq=Ml~<BK8mBb7?%5v7aV@
zK#7)_SBSmu2Y<B~AgtiK=SAKZZKe<$kpqVap|O9|>LEZ;<jsl2soiF$VptmWgiy00
zwA@<t(rp8gliWTJzI}sLT8KC<aU8nOr!A6m3%I)~+ehW@KRst9MOEKr=4BQSSK>^D
z{8r{M)=)-n{*TmO;H;F32^j#u*a<9u9gl^*+vB8;DuwOzwrPbWgd!Rw1FN&JO54l_
zuU#)42_ELY_w3|z?WIw3&NI=fUlGuM25!4TSf3S|H_vKL9Ob`fp*Sktc~50O;QHpo
zXutKQ%YDF3G4;w;5km9=#BwC329v;-t)^{r!6^RJR}YiR1s6r;I}?u%U6R#oUPZ1?
z14uM{-mkXa=GsVpmqLa%EOS4tu!hDW!tfCEWd&NHZpLoktHYif>b6g30d}#49z+Y5
zDSL*(9Sj=w%6Gncvhk#^bb|zyIh(J5_Wxt=y#uN4`~UGsMH)KEkwj#aaioxul8lhe
zv7OB9y%I@Lc9}=ABJ0>PGn2itx0F3AM0UUDyX)%u+@Je%b$##8_uudDzwT~V$9cbA
z@7L@3e5|L#x;(L%?~0*@tmQN$I2A%M^oA<E=FnU{TVhS6Sqc)vXg0ml2JwD0`Yfx;
zWc!#cI@cK$Q#Vrwy~P4i3&?NH2h`5AOsR#?BK;)kJ$4e6Qz}EDBI52U1`*5Yd1)i*
zqnY`g(ugYM@(JF;AxZlWTd!p&t{e)pRLtq&kUy9xAZZ|dw(qlxW4EHDcf5`z&QfH&
zo3m(GAGLWRrtSlxieXAs!*#B+5-nb?+-N0TgR9Bwfy$1g?fHOBujM3ejh$PJH^?jt
zDq*1bi_N!SFTmbn@fwe6X5J1_)<zv=%Yqn(VvNPo?fAf*+Ok#S$|k$iR-bX0C4gdV
zHBElbQm3(c*Y_K5T-q-04<7r)!gS-Mg7Iuysg)VT+^~}JIk+3U8tgT7C$REM<&TQl
zSdnq;fTQ3CN5QrE3ORSRW}fsP(i#0TXNgDR2w!-QUs`h(@Y5E0{!WlnlAC_v#_wJX
ziW$)=deO08vxccBNVM}qAyqhYg7`6#0M6pk5w*M7SIJUzJ8anQ)-w8oX^*O!EJoxe
zrgK(F6HUjA_GU)!CV|02pk(z;ahtbcW9{KtfaPw=a2BX6qu)1rUn-c_;ioh$*;hG{
zQ0_Cd??K}}^2W?EW;7)WT87EvWZX8hP0ih!W0GD!5R>=PI{C&|4>+qs{cRbKTIroH
z5zm_FjWgT$iXUm*7od2=LnC^4S&5TA(@DGG>?M;%Z_3PvJu%89;c?s3j13dj@7`R<
zJTkngWA_Qy-3XI0Lc*MTyywJd!7!0O!S(qp&gy)wo(krJG)DgxCOSY+A$EE7jYMDR
znY`u#Oich!*!M4P(?X5nUIlg&%M+R8ns2n@gjK%yW8Rq95#6?XscHB2x)Pn;3!9f}
zcJqOXZQA&a!7|n7BV)K_^zJM&aF$<XKRSp7eVqmUj=5-0n06Qyhqgl$*J*_zr!7hF
zdTLE&t<4iRPhjgq27BS)iP1KswA5MMfgIY)Fe&sJyY7dM3Uc2Na;{LV(1UFVTdw1V
z(e_(uiqksnzpH`%Sa|?~X)&F?{N0kOMIc<lb*D6?<W7OfR|*p^jiiz%o~zFG+f@wG
z9=~P1IRgYR+WqFj3V-57lZDRplw$4p6Yeum9k8>a>lfELXf-?veWRav=I!(rRfKCR
z^zjIxwQa83>D_zBEsu7*sB60Gpm?W6&&rpt#6-YUm#uVI9%#{Q(sZZsh%vLbJoJRP
zG<wt6uChe8Qt!rzhl>sylX%JDru3;K03o<bT+*G~KQ!*}K1DZa$xr2Y@~lng<-g7(
z&`N79HeD0-xhv7)XdO1qT~z!++bw3e$*Sdj;PRkUsTn1%^gd1YFtNGab<>uZHz^7N
zJY4CtRy`L6ed?wqvbug=60|+Ba5c0wq<G8#4sI{hhh6QmC{r!jETw&J3HL+FWW&a7
z@_)Y-{<<sTj}f^m20hyI$MEyp&yGLyk>@H!MB%USGe*-`b<qHpm$%O{#B(`QC!mX#
z!<qcf`JWjU#Fp9U-uB!VqUBAVdy6{|iE$>81(k+Ph>5gZwNO#Tajrq;pawa{3%)eN
zAxdimOSFbzFh8qeqOKhtW*pdY6!z;c>EpQGmD(?o5;fbdNf^G$BMe)4o+DA3ym-#4
zCs@nPKacM?v9t<fu+)@BOQ;tkDp=}vXHyAp-Gt1gqJy2^hJkdCWLkre+R_lM!a0;&
z0(H?C^<A50>|yT+tGDL{lWNGs-v%%$pR#XuVJHr%x^5|2o$5Iw2<PQRcxNeIO0}H-
z6x|dvudyg@Im#&#&NaWTo}gljAt&-bMi0gnS?>o2rz2Vzj}+YuA7(RDisKM0x<#X}
z6L>sOTXiRq*4cV9)vE(#K=AN!FL_F^lxAY*?h`vMODWoK=PH<$@N~*$D_wTt+txky
zw-H?BbSqD%&iMK;2RX7wara7jRlpDtul{^RU7^$2-7MXg-3EFzzs)^4aN0pZ5I}D8
z`-2_*>1#Sxmup2(w(M71_n`!3KBn6W3AGYMRoYd)-wiBWY|+~hUc1aDDRe(3?8a=G
zPDxK?>bvFg;JNplGv28yaC51?TW}oAqLCPUoH7{Q)L{~&aw?=s&LG??!$_^irEj)B
z+s$-at+E()b@RZ&noE3GVwXS$!Onp^V9#G0<=>9csvPDA6gK@>;eAh#z2o^(RD{Z#
z$t-wAdMpDghrQ`0(22Qj=-GmwKfEM^58Zdvx7tuO$0AT(v<j?VW&Y=GCfv{%W;;l}
zx~#+X@RY`tPznVsi4kX3c$bW|#u;Q3rsh6>`$eB~O~J4=CYle`k>4Y`rdeXaDbE#N
zYLC*A5;<?~I*|90tBY2&9NSmOp789ed~dtqX*vBAV>!q3!2xp_9Iu`=h>xk@Lijd1
z16mVEKRxA&p|wm1Zg@HCZLlPkvXdCC>RVyxu1r<U6n#o7OPb*$ST=Ee+5SfSK;Lvh
z=#2U|!s2)F9|{xYvx!J<>9SGFJ3nnsgTqd4CnqT`ZgtilZtJ~Q0GBJrP7jOS=1zI|
z?Ph$rY(>D`?G4l$J#l%e8%DXUc*P^!uC|!QyIJ~WoyCN2%wDaZVc-_&+pLUIs;x9G
z>sX%Y0y_=Tndamt>uKR8IjUC#``T!G;|x5UCudtXoSK8X6iP!1%JL|L<M1EVzMmz`
z=cOm;gpXT%M|!ZJdvvWSdC_X04@4OtfsT*AKTr*H)*$Ww+@GE2%t%6eZgR%qNIHRb
zv5-0I*X#aoYr2Qz@l{V-6)}(Xz+XRc`x(mS?eh}S_sAr&A*h};BdWz8P4ybIaF+K}
zmx#P|o5c=+jdFqY+J;nXWk-<Kejc>tbFNFYvy&HISmgPZ+-%(zcUbew<XSaH=8=@x
z<V<C29Yr6HA}4CaKhE4Rug53wqGJc(vNy2WN&c+8B%(Jg8n&@am}*|2l6M?%zAo`<
zT=b~Do+f@pc*!O>2vn0!$Y4#UQ*PkF;zeJqizA<idi))|F*SD8KG>J{;55~z7Q7xG
zx__TmF8(+goe;@IW`a@4Lxvcb(h!t$9h%)`-|-0ALQNRO)ny9($(ytzS^|?R3I3Ri
zxQfo<hnk!-NF9WEm<wv6f#GTc_;PK~lhQ9XdOw(=%BR^kd%2QrQ&M2XthB<yJ|DcG
zh5~J|v_IC0DtU|KjIEAc*fdHsN2cJW%AgB*1uZ?+-_a+?<`ep`wkkc>nx4<g((8cV
z@7DMflnm4ciUL_<xmt%V;i$TXqasU+b6uJUGl>G#yJp7)c6L8(E-!Ty`q@9^x@%@N
z*{glXgm9oh2mQdg(d@){IHT$|=3D_dQO|d(-<=GXwPlwLtcCw+8UQYn>==hmsVNx}
z=SG<|8@=a?P`^*fyseKOI5#R_;XLFtm9tB`Rmv`J-BzR(6#aNGyzb%;ABKB9mO3JL
zFnf4$gx=viyO9#YpsLO*OS+j4t==0>FplYk$GV=CEZ?U0;21J3$t=C1#rEAjUqW4}
zT>bqMBlulj^hT5xNj~82m1;cur3D0I-VxX&vp*iA{!OPsL5(G$AU={lmz?wS)VV-O
zX81N-%ms7<7TNymaOFPNWs$MSfz#gU-i{x87PWqL!c4X_L%f$2>Vd;FQ}n>?M~7SR
z)K(|+PahK(sQ%y+z|=$wH2(<H0Fx%JZW)5M<WHw$vZJ`BEh-F+eR$nxuh}<COvzv$
z=e&fv7CbTDKigeoxW`PqY8~@}N-R^;Fqk>bt}`DAb3sOI9Q71k6nk{KNX67jylsC1
zMGll2n7qL(u~H<Twi(>KE@w!I(r4%c&#WdtmA<e7C3B?K8B0plm^iL!n(+}(+FOK3
zv82VM6P`|y{@3-Oe&<;1a^g5xM@QD+ky0F$qFe{wwp<3APMuAe=7qH~>K)cJ_sQ;&
z8&0sC(0l1kl?MVZhTUP&)DlNoEwi+wHxtbF3>o6J!qd$!Z@5__m&4-08G`=6i{*lY
zLskNI(J`b%2WZG~7l(#v;!SCPfG?VMQHB1v2}fw8;+S0#ZNK=icC8C*><;ID9T_Eo
zuf6{)5dO!BaYljV0Y%O&uRlA`sxxQ5LSTS*yf{@T->}U)-DzDa+vI4wrW`A#>sV(i
zkT@JR*N4nOTTYX3Q4yRVhD6UbccewiE71+gM*#*tb1J!A<kVkUg@ORTl;?U0B3SHr
zgs(<aDT-DDoA#E;@T-yQyUwQ^=bJ*}GxlyS+c`nF+M2I;6^t0Zm~ly3<xGc+d^87T
zqmrT%0;Q^-P!TY^o)-@Kj1kfHffx#f-VML3VCLGf4YoB{s7beP_!;v*7R@MD5QbWW
zi>Cl;IEbK25{kUvLifn-aFr4^O@gv32gT$nD~no5wDKm{(o<0<wk~GtG2~cq7Wyqo
zPxjHWrF&G65H^-q^cX%v?{@G9IKR2-95q_QSBsHBbV=R=Z`nvP{ESo=)2?|x*c<2_
zX?7$`1p9r@W~if~hCxKCHq{}IZ;AAdxfbfRE-Cq2TBQ66O&zneSXBWPR{nqM?*HX_
zK1^m}dYkr-I&Kf~Bm4t$<`6E8=MZ-Lq>%iR%yKf|vKMw>^oG^RdQhrGXuE}LQ!1jZ
z0OXgm3e0JfKLALZ{2(|A=eBhkBDRy6Hae>ucLi<8bsF&sw-8ZRX>>E)n-Ag)$DVLK
zeNq0x3V5(@r0jBwf;h~@l7dn?o%=A8N5Z|BI!E$TvnRX_9!y17B8CMG>}2jub`=;*
z)_Xe_j7tb1Xk-)!l}^Q&&89eAKy-;{B9<kq`W-+&<uTFH72S85n@9AqcH{zzk{A*T
znqIs_z#0$q)XY02yjc?oKFOcbuXLFHI5%A@Zf**U73aP08@XEC<ZYM-=1~z`UufOS
zH1?pS#Az#dY+j$4h0)cfG?H-NYa}4|IK8_A4GfkD)uOYc(vkVWwq+uhO5pn>Q^{Jy
z6y?_Vm3B40|4jqehTX*B6+-Sk2u+r<I>uX4l_e9&R>v+rhwV+UL?ncb_<Z>aFH<`P
zoBZRg8d2Zf$a*JaNSzyR47}Sna|~ZsB&Q=}ukI1F6BAM7;$Ax0V4H8Ib}+Fp@r|SH
zor+h5nFO64Mhp9T&L7G2+aC`!?t5_9;A_&;P=BJt^398Hb*dc%KTMJeo!&a})MMsY
z82}8lmQRcJCeg;U>8XY%)_)`={as~$NQjtkLHd>_!=9f%aU<Y?mnLdV7N(NT0vT+k
zA&6_coC(9B(>TV0tFBi^Vu`Q#;s6eRKGC~3luKL<FV`a~l4J|0kQZrU<g;ZPfH5B5
zw1kFB-j(P>jJc$kV$kc$jEStVE{r7XR=mkc3X2fUaZ~cNql(@91;<9IL#%oawbs*C
zxad-2-5Wnzc|SmcO+u<tPjim~LPGUW4Ch)B>40}9dpe9#WFyy|_|jVtqIK{V()8iy
z^PcDuWV8tvi%`UyAuo?feER0x!0;qrH7nZ6x-e$#shuX2ZcR&0&e7hqcx^kWckzJz
zJ4Qje&<nqH2gRVoD~~D9S}>C=<a<zdo&WqD&f>4U?+#sJ!i{lD<qS4$TpXSSFeaN^
z$q1ZeY8@<05suBSjhRU-WDl7nPH|)jLB;MLMFm4EslPM1vcbK^UdyVNt9Abtq-$yh
zeA3un#P*4LaoEP!UGmL)V-H8>!u#AxA--~fQ7M5iIp;+}IE6x=T1OV8JZoF#T-{F0
zRF$!ur8Brdk8<A@xRo-l6O;`&-;&9dp6!vD8_%NUUN4HAmxQiOb<(7c%;WbjJ6JpR
z{_*a6GKe5dnlvPT>qG%j*9GVq_RfI(Lk^9$&{RDb&CrOCzm>P3=736)dd8<;;{&pG
zA!eLwk&`B49vpB&nV^?!#HY%l$WL(Q?o4udNI{;2Q%Fz#<qs~2PBBh~$J+7iS+C`i
z*7Kgp%u>a05Z?!I>d4DlwTk8N916$w>j#*9Hw>@KVm~s~e4Qi?^<prZmYMWcWi2;p
z^~kG8_=boQqvF~SrSW9*ct74PgFGb%#bN~@pUyTj<2fN;2^gjHks<_1u$TovIE|c<
z-2-ATWYiT@%JN)P98i}DbhT}GyG-BHwb~ICH8&VVe5X}#O>rb5fze`a6#>SAFh3)-
zN!1}w-37qNq;Ra}d6+1rTC1GP<*)y9s2rJ5pbXbne>tP?%tANR3ESIa=*n}kLGBtq
zo<r<%7su<C5JcMhYO_&p+g<7-6Hy-mQaQeI(rCMLgsb@>L2~DOIX2;~o2=@P+@@sP
zqxDI?^D;qsVb7ng6l%O^v2}P8E`FZuG4q3FeYVwlxa7E@n4-{_P#IE+Z6a2hlxI}w
z@f^gSO}9Vz8Vui(rRJ1yB;7b9xR^+D++AZaXZXBKGCXvH5L}08OBm_RURGYr2_04U
z{5d{`m+d6SZ&9o!x_0v`(Wd(wQJ>q&b4qr~gWPe6((a0>FbTQw<ct7mT5PO2fjT=Z
zS!^-yN-7`bt-KSev3WVGn2(0*gq*&~B+jFC>U2zK_*>&FKY?4e-@o^dX1JBFS{KpM
z%-3ixO4DC3sl0+qZ#7v^)-iAEjZ~*wA(*L^b2@XEO=(OjM%W|3D`rvBz66F???S$z
zY8lU((CIIfQYQ>5$2$7YA6tWn8ua&<wM$Z~XFWHN>h?RyL7c;%$5vQkgSY$dceWAn
zc*8q-j`6>6xf+t~$%2v|Ui*2oDShr2;16r5RpSFr^hG5TU(u^ORvS&)H}7CNxzUx8
z{xoya3tUZ$2>1-$#;|s@5vK4W`f%OqE~<aqJ98uN<rHI?j~cqv=8t&<irDBlfU;Zh
zRy;+il(z5TgqlhawI@nIVj!ld%)dcQZR9)cnL8EV5=wF!+|Cb^e$lTc9q$cm@itOq
z39lEYUxPPBV9b`?z7PgGMUuCSMU7oU9~wkvlzt<Gk7OD?_vbg84m*I}^IWTquu<Yt
z1!G4rGpQS<^_*v-xWZ3q`>|pn5cr&Rhx`_X?gGxNUF8T`K8ie}MO<KNOx!+LhV#D1
zU29BPz)CHDO%whaA6hs{>!AOemeo|(e#lu_1fKB{uKMDbDt{4;A?5<RgaM3?%_^bp
z7!HyD>SX(}<idsJ__cVamx8vzv4{Z1v2RYq^o@~(eTPa-Cf_Hj@q`K;9*`;fyN<%F
zxfXHdU<qziK{Jw=qwPgwo7NT<yB>7%5(|IHD*p<4d`h(PoWH+KOKtL$3=KNbZ(duS
z`o4bp*V@et^5Fvq7Qa1>4utHK=j2xt7n|i5r4~iq<7c{AbY}DjDHV2BrWfmc1*Rg;
zb|Q%|t=^7P#HuTkwXL<(9Zo$s?Nn2QN--37Z^w)Ct{z)84diaBTuv7ZSH5STZrX7*
z{~C-)>8U;Dxs7ApNhDk)xZH;{p&^*%N4J5!yG;GEaCU64A;sUiMf~<ugo^!|!H_A4
zVfeVAj-09An2y;@pL4cY3l+R6%-^of@WBj6!==qcxzVHjx|T|-yY?gTo3CB!DVv@B
z`E7#p-Eal#o@Z>Ow_1A7<17q@jj1t$4{kiuAJ6zypCFLA5QghX*N(Nff5G-x^|{VR
zEZ_ysf=?!WHt#LG-<WZLs#}}l4u;L&DQP$Ut~o%+1t3de&wEukCF<N&9_WAHfWuy|
zBF3hgtPrB<`ix5boK(?zTt^+OQ*zZddT3Mcsu-ky_I3C&N&m2BC#LS)4+^HwLrqc+
z7eA;9tg++X6m^w9{d~#sY1+mO`)A~d2E#qVK~sz)r@}_yE@8EgZ0eR0Xu8G!_&+Y;
znLkEg$CAE7wQLN$mxT}ew>(ASLI{C5SL1=YH(Q&|wRbov%vA$p8(HIy&_#=rc_e5Q
z%j{%vjkh0{P%T%mI#rugLp}^Z^S*gQk7=gU-^}+slST8`4c{|ixN_2n9QnRRBZl~~
zunSTiD3%>)LwLw)zNO$-QHs2FuE9uS@2#HLv5^qBoDVTB643*=u|2*An7ZeD?VHUC
zK{$2!&|E~;t?KB}?WxNQ50yU*cScEZ0Jkd>rKNtT!(VH4-mbi9A%bMwyF_<;UR_kQ
z=aCn7P&zV1oE1eZ?Uk_JEJw`e8{wug<1&!7sZ7UYs95^!qqt}-hs$~0xFIjCx;%)v
zIVVO+kS`!B*yr1bWEB?cIu10<CRK<%8ruX>!;=}&zS8glie$E<sGD<T(M|EL!3_*T
zzSo(n6}3^M1b0G<L6;F#`7FJF2Pzz4K0Bsf;}SFHZUgU_#eOf*CT3kK-T@&YNwJL7
zfd};2i88p#X@G|jZ{y^Z*lMrPV!MP>;wL#I7{UIXflQG3kc#^qwSco*rbAXgQuCfl
ztFtI3l6cO^qex6SmM8UYn4HC+4$8QM)@NxymfJGH<&y#TRgU79(dNu;MI!QFUpZ3*
z^tW}3+GI<K_aOC3(=LIPjELDP?2I;X_JSS#qG6fRsdn{OfLUp@Dc1Mf!w6XMYc}Sw
z45?W$+h2mT+5)*N7jRd5`&kv$&h|lL67agRepg>taKJi+7Q7@khnp$8dbE>lF{RMU
z3SW(Gs5esvEJzB?pmSo8Ev2IkYY_PPY_}E9cPZ^`rvz>l`hzBJcIhn)#J%?~0UM>Z
z+Mm58fwL0+e43dUTM=eWhG&jeJuO9{P1Q7QcR0*0U+9UWW{{h(*6**8Ru~OnA+i6@
z5~}J9^BKihUD~Z>BL<()ef+)zrbDJoYiwv`F832pbPJ_jm^<P*$faF+8uG%#Tv5}8
zDGqr#@?37rM>-wFiC&KNHLpe&xmhZbM(*UR_Dh(e-%Z$PgsERuyNa1UABVooJ>wgC
zAvZYnB98q&>XKj57$v&LlX^B=^?8jfS5!FtqKwb0Z#9a;lyPm1=XGoQTFxIGV_Y(7
zd$j!0siTYUr7Ko|@<~(1_Kdu3KAU8lGL<9S5uNPqIZd)3HBy$?EXDGMrU__A>89N^
z<W?qy3@Weft`>TndumoWmSV}?m%5NyU35<_gITQEF>N<<Q=y!4D16(0tkVu>9X2{F
zVfUfQ*3iRpXpvrZO|eIc?n{`n!0PMaQO(t-a+@Z&0<zUhzQR54jTlsnPBvuqfAHJr
z?vJ<4TBy-p@rxeJXwq=<dp)wmR-av=V2t0gvBj__k90Yu74~H4Oahx;G;lw0SM7)p
zMcuAl?4ZDBO97gK^o!pm^i+!AEOzY7qN(jWb*+GN5@%3U$3DqC8kfeo5%NWLNQFCQ
zxPqs5)wpclKlj?`=*+I-j_jnu$)4;NJ=ZbKB5Y3vlhp}w%RAvs?fA(C+J4dhi9r0(
z;82)|?P0immTmp|pFh#gMG{3uw8eahN$@slgJ8ex<&mG`?MFaR!ElE5!t=9d|D}8W
z{U7)5A@X{#Hy+hM{HwSA5w}NFL{s!EF^m~d<H*inzRtYA=GVvm`P)4B`_ooL82+)<
zEBp`t{J(E=9B)?&7~SkGT7U$k7l<M>muI*k`0QCU%=_+J1e7aP3+uvh=jp`$J)X#4
z<<{hJ>@}Me`K#0a{cnzxP~FExNY&pzI{WKQsk`VI7NTSGb|0VXH<PZ4VMy#dJ_Iyu
z-5NAqr$GEX9_jHe6hTt-ac)cAvM`44-SIt1zrQ^G<@>fDhA%g=XN>39gL0SunE3Q+
z9z=;1%ZLKb5F1F%o#20s?w^0<e~~h5Kiwl*^&|K8(ceLazj@5?bT7=n2|(+hD)IQw
z4){O*Q26{GXCF{}_2B6L1%v-~)BVrSKO_i(ivv30!Fs<Q`L9a$fBx*heAAzQtoi}J
z^EN&C+JF9^{I$&<1;L}bu4ZKTTX^;Npa1n5{LVk>9}53p{L1n0sBGF*PoDhOEg~{a
zq6;G~?g0^M36Ng_v~(paq(cFpAa=MQECg}dTGmw={U=KzDl+)jcXXHg2y)L{WH>qk
zE{SQPwk^P@<R1IEU+yC!wnN1BVzSauKz=U-xXZ@k&ZBQ<+RqXrHqm^w)mA{`!3ptA
zJRZ?aA<mwAfE;czZZgnM14MiW)O2n-0*G=0$w{2s=BF+ufx)Nug??<Wn4i1?nodhl
zCZ(T#z4v#h%bzRiDhg+j{JCTM|J=8pL8NY=!YG8Yd{Bm~+%QohFc0X3PK1{a8l&|=
zKZW9&<SF3Hg~ISDPDr37xG=dXUQ5TGxs7^u8j-}qblo9<6bB6$A@=N{FlavtA<|`S
zJqk+%u`JaT#7MkU=0n>ovY(<@2rSGR96(BVr%`ujAqb{3yT5SAf>{2_lGkaNlLJJ>
zPAe!%Pw$w6Es?taIg9f@ug4*NWM^xKKV$e~n^d)*jW@c+929C)g~UsObb?a|AjBuy
z*O^_yp4|{lgL!kch&T1z{4@4*`)E)khMo)YkB<0NB3PI39LdexUs(W{&}smtLg5$v
zzKVRG=_qr|76|&0#6{~90rs+_Wl*!MK^$jK8wM2qg@~9FgeB@L26t{BUx!G{#V;N|
zN)epD5O&9ljAjDv<Nx6$60wGtZQ(Fy^{*?Xo!b-9WLB(ZS|65TxczwL$+hjbkl(56
z_24Aw5RsIKjH68PW_Y$+ZE}<ufo|Fs8M1=NdShYg1Rbu6?IVObvTi}-?82$hb&2%b
z(en(R+3t(xg+LB5kSs7{dr~eSzD{IhBg01#0_8ICKkXxGF@z3DXktwNV_iI7itJ(d
zsE{h&bRn8d1@QScj;(|RQlhPgbN#-@89w%Zy2bWSE9m)B8u{-tHJbMQ{MdIvRY>ld
zI$;H#q)(D*LCZ(j^~PY}_qWVzWXV-ogv+7JyBRPdb4Z7?QlFp^uzQ6V9$2`mb(}{|
zP*vt<WOF!5ZNCN@+o7alVF{LO5HVgLu1*AME=0uERDkwt!BzQgP8~-|6g~=OBuOmz
zzApWYg9URWceryPZ*LB);MN*M2n<2%YzA$`16lr0UPyZE<DLvy*dlEbbG%VdYOr|K
z=+0+nJ%$TUI-CIK$wQ=5Gpa>DpaUEPrM0R4CWssdkjU5mf7dYfdS}mlM9v^!a$f<G
zRc8su2j5$g<K5f3K%6za7<@7e*a2;GFrhU(raJX1Orp~aT`l^VR|{MW#LL~Y8Uhhs
zVqioz(!D00o?Hi<s`=c#BdmOyx|@)`d6cVXoag}UGx@&4AKTN{7BgPA9dugW?!;FZ
z0;Ml+;rJP%c>KytaWaJ6(H}dw?2Pdqlv$oV(gm!RxoNtaCI<2~Jt;uuFzEq^lsmbM
zB;a*it<}vU*rk>c9wlgj-W?H8vb#DCsBIyGwvTNn|4_OQWXF5VrwL@jbU&?qOUKvf
zD?ODM@Iq>8OaiVnJ^)|+^lHMmaaMmHFE&lqZwTx-mLLnyJ!u^I7Q~!VL+~bB_?=HS
zp^{zqfaApJ-QmkME-B_ckI`Bd+z7>F0cTSo0O;Mw;40Fg)T8+UWoXtO`p+OjQMcNF
zwevIT=Cy*xaiQZ2PlW$~>wIQkqWaJ=Vkf~~hl|G8p2B@8p?0ZRl7Td;2HFc~0=mGX
z)!d9cN*VCmRn2~alb>i6az69iPd$AVP`%_eHksFEqfyg55hr3I(7TYOs(WhQ;q%bw
zu2X<{t<BuMY*?jM8T)5bjQ@wK=n+(IRc~=m!v4I_Z-b~YITI#@t-;8!exT9mh}nH_
zGpz@oTWrSd+L0O@0mUsQA4#eWZIBiXf&}7AUKdbTZrZSLKlQ>QdKU{otA9{PJNiyk
z(t+yT3#qLbzCH&tzo^d&5XO>M9o&$FWH3Kz0ilBA{Q7h<nT1u|V!Y^U!gS%-DY6Iz
zPkr2AfasgwK5L5sSMSfb{u3$3BLos8M5};;E{TY!kwJ29gX=)>AHtPB#1>?!<Cc<_
zdJDB*YMp_o9ukLd<96oM&i+7gs4mI@HRGcd5T@pv&J-C15SO`>Z%Pf8FbV<KOv`79
zQoZ8DQ|u9YFKG<Zn$judQz!2DngLTMjig20_=cq<NIy$lM+ExFZ7M3JK`FZnQ@D?I
z5XSJ&q!W}m?X{dx99fmSo3^^2DC4O8vIk#}1@p^`iY_9loQk~?59{oPg$1t8r6!7+
zr02tAthpIAcvt+TW&n%MLmcER9r)$c&G7c67o*$8Bp^<Jw^56f-hPm9jm!uh?$?DL
z8}{oVE~ZypfN@T=PJ7C0=O;hQV~=7-Ow3@-kM2*+Ow(nmGoR#nEY*@)f%%lOHVQkA
zqTZ}UnHFVfT2=y}%A>o9>)lpJ?(*%IZ?G-}L}YY5eBsx6JnMLRh9Jp7AJm=M7H~mZ
zQ|b)e*NoLLH4RW5o|~7JF;WufzSjMvz~Qx6i?D6^iJzNT5ZS~+oKL2LVU<}Z(s4H_
z-GCjbQwK%}{kXC&8_}_DzD1Ot1`6_XuTu2#C=?vs!G<Lc5QElAu2?nArJgnBVPRmt
zEFf$nOFWbB^-|rbKCvaUB2+-Sws?vY*UO?hnZPfae(MlujM=Y8|BofG=UeG2+_}H5
zJP<Qp1#S7auDD&E5(VZGg)=3oTqXQxOMEzl;fkIu9o@0AyDo+3W#5E#*YeLou%eum
zz&Tl-4zb2su30b!%3>IAS*@~{%&?Z6VdjJ=y~cekvye?~fajn1G?qItRycu84<UCC
zAvfzp`jUx=!Vv$4&h3@=al;896_mHrQ-8-lxavS8_686f!zfIN3CnV}gDR#d7H#Al
ztY9Z&9L$BbV?wJj6|HwOH_E9`HE1hP-L+l9tx02GJnEa92&eG|o=YhCg-4d0B6{`S
z+@^p=$YKskV?=!4C<xEUPiudqT>Vu263MMbXJr_F2RWt`Uf0z$OfMzI$xqPP+{-bz
z)P{XqP0O-if-%YD4rYcxKAw;}k&rV{U6~K`ENjsq^|rkLl0H4r+ZRHP4e^(s&r;Pk
zGbdc?nFrqr&iRA4hpjH{NQkywu-?dq+sB95n*3--acvA%*6R7i5+4?1YIui4t*T}J
zzJkU~$*E?c4{!QeN*XXF4ctgY7u*Xcc?a_c?O4dMW3PU*FK42+)$9|2?Mc7C#%@<}
z4(!dIPSe71#&K#bhY{zBm6!>}Pnxs>S!Isg2Cq%3uk~3T;krsNC`kJpz4^u;b+$~)
zSf-id^{qEhQ<@pX)}2Q8v7j}RkqTGO!3>XC_Jj6(yn3)$cRaZ+0NOCYH0!i%Hk@E~
zq59{&htGcTP8qC203q#(@L^5USDD<(<5^8hkm{C%j5sV2;Vx0gt>-pB^hGL$t(vVa
zO0K?zZP4k&WP9}Up;vW0XnoeA4<}dZe`+{0POT<v$lA}nzrXgHu1`vSuh|i`7l^)F
zU)v$qHJ`jdcE|KS1dMjNB^SQj5EANnG-cU(H|owv$ZaK$^Fp=b58qMW>6UW{_~(Y}
zwh6^EBci{~_!|!%cnOLtieN&-Btea>{gG-T&l#Uda`fQYW3n2g%%5723}F5*%{ERY
zy_rSc585-kyw~MF`O9sDoax|f_Fj)SV~FpxGgk<{M%L%I;hZk@DLb<UYo|YkE$#$w
z>hbzXkC;3Taw6sn$^#>sx^^<Ihv(BG58w*<r6wktFR>OKQM+<ti>)5iq^I=xif;Fv
zB~r!Te6#fIoXkU%Z`d#+?p-^o_2$HM#F$b^o1^ETpOzN+CqFTHBcf<8Ra91Sc-;im
z(NHb>`H&#3EWb^5&>a#z1#CPS$wI>MT^ds=FLiu7gBD*(n9~^>kT5!5*EQr$_NKcC
z$vqlOLQNp)IgOmQJ=^(-_s>yP95RXVSc-45`zFB?vGS2h>_-2Kr$QqjPn4JUdLYgI
zoy#|gsWWsBVP?2V*XLDDZgA)x)kUR*W{Cvxg74!&LYU=OZmJ`6cCkz$j^>%O&_l?W
z-7`wDOA{s>T??nkA><}^abLohj}+>t3fNs`a>e({c|;v&Mzt)8xGV4Uajr~$47Yqu
zp)e-j+E-h`!OHluu<i<3b9hnS77!&T#}s`@UzB`t|8kGP#)4;xuk;s0_g9nmZzb&h
zs{ktOtqf{l_ybjl?;~NNd*;nVwx3D!S>&yU5m>4KCXpzJ*a#Kl35>3}lUK;<d268Y
zO#}%iY6h+f#fG`?<qo{4hz31}pvIVTl#$HiiW(E2ML)e$i$(7)l7}PXh9{aoJigh9
z#~m_PHtJ9kknIT<lg?9Nk<n<4o>OJKs8Jp2@`_8FDr|eb{+bna`J4{+j8S<;aA;2k
z7=E)~_7kXeA4-zpfHBBOa9;TCO`-5mwcrU>zZ7$Fw(17)WsR8-7uHh5$0a6_cSUW%
z9`ZSFRg~R1(rfdgR!KF?-~u@jV&=k(bsk!PGFlE=S-q<nLrn=$84A+FJuPXYTNiV<
zoeYW8U*{~Yv)|W<0UD&$jCj&`$)shK+3ZEk=dsZ)!Nv64<QVQt=!BTU5%lCU3I)bH
z8izR7-bcT$7&K=12)|yJ5;c1+LF{85t(Mb>Z?Vy;60u~ftb0dc^Twxt#tuAlC2!L*
zPjc1U{#v6$cEsTNh%ed4uwvxantADY<;TltwE?i%nN<X&ApGl8^MMyyFU{1I(DVb5
z+y*UpCQBFHSS!y4x0CWhFr0j@x_GkqW;18pbN_wwZ6}TnzX{*4%a{@sL0U=w6u}SQ
zjh?o0dOg?Lx`?}&OBLroaLZ_AlCKrNH)|uE79+@AM2cN$7K+JpnD^oXA@tzoilxK$
zGA>h$QufzhY$#Pu$<$04(pVwxkFBZ8m5A2QwWi}jyNB=yV>psLqsUsZV3{6o8DnpL
zw-BsDFUXTyI4Pq1CC!$K+q+?4CCG+iN(yRIe+){E+6L=<MaE0~)H^Pz;N6Q3nSkX~
zt9uC2kJOz%aGs0O`f{m=)TAQJex$_hqm0IZy=V>N?8jQ@(E-43tt!flU-jwe@C3Y)
zl@EWrPjgDfPU6rQ;SH9~O~VlBEL!n10lN&bR+(9*(y9^f_Q@+gu@cTOb)7s;P%ip<
z(lWs<E5=~;20A>7JTf8M0WJ9%SMjZIqw~$-M@44G3FHMvIJq(-k8eqOt(ls=b)ApN
z3T9`qnwH=yxLoiZe5T1p9GDFME5+gM5ucVDV|z`*4P;XcjN9%BS(Y~0O}G^+bi0v(
zr6j-0wVP$dF~*AbYO%?%Hl-!>5i-p)!zlBXS<HHo6wC8hm<8TGZ39H+nTGoUXQ1OW
zPoc?&_Q=oG<8SO~{~g`>?{D~pVlu$zhF2v&|B26CufiYTDJr~Ub-fK4^6>(i>KhJB
zp5Zn6C`&1dJ~dGh-z}yr+xdH(=3--FMNa5yyeBBB9vV@xN$pA9L`H1LCNScNdxoVo
zWF+*9;k|s~(#83>;86M(0b~xn&ppIcbh2<HR6}75OVpZ4*_&W%uH*zK>EkoER&h=h
z6NWy6Ov@o(Ukeoht^T-Q>S02{JUQs;DCQZX&nFL{xAdJMYSjA-w1PY7B?SUQP--qh
zkza&NNEi+=+AcX!OA&|x^Q!FYVC1z<Oi9SqH{M@L(T*SB1<9b<P-e7P*W7Cjl@T@z
z>-Pv2Jw<g|cF`UF*~P_4&UfGQ<<e<@z`tbgIcPeh8h&>Ax)!iA?cK$f`E;m=&Dzu<
zrcsl{9sJn?fF2oIjqrH2)>gHO4O^A-)}0CzCR}#sc<_I?MoEuR->W+yMfMAwsyG9Y
zNJnc>MHqe=WEiz8RNDqpw>j^mGSiHV>1;w7NhzcKh9bw5Dxc>oxbk|xmB|G_Q#vbo
zZBr3{L3*f`Lub30&MKTM;=y8blIIBHZJj(sjuWP1iXJ%%(pQYkUbWhbT!$BWBSW<A
z%D{1b$F4cR6soo_PA3(LD5*9OZZ+|Piu1^hDcMYg-QDL>MaJq{j9udQ-?n)x$>Jlr
zl{$~h@+dyV^TZfMXoUDHzoAUzirC+)lwdHFcG_&iph)|qjKkJ@-w7(7P_1dd!X>Cp
z_OWy3a!zGx>s~GJv-{?7azOt+l0C}<h#tCJaL_qdCx7E}Q_S(kon}sA<7|u;=erD}
z*QZKAs=`$g-kschaM~OTb2%E+a2#=jTun=nW^1)YSIrdF0AAI==g+UhTjOUp@olBI
zHZ0e{>MEVqP#I5-mfr%ORFT;!%bU$4m+kJFI-|FH<VtT}iHPlse)xY_oS9=eFV&c&
zN&bL_sC7w@9&|!xEg5kVr=61_KIJR7DNTCr<qNG%nTmWHnRReFEORS(KTyqW5luoI
zhc0CRg3TVdwtjti@>HnSev9+_!yd>bTrWD*m#L7(yl_z7__{CG(y3fJC?0|(i^O^^
zhKr+hc#txmvpJpAO<<0~GR-{<Hrxm31}BvZ3~*E>X*20m4bFqdPM*mAVmQhP!({MP
z^>}`{3JZr#K;XKd#-!P0E9F8*59b*dHL7R?#roCxJ*Qa!o#$s>vDS((*IaeSEvqzc
zgjG<y?9M>a9l6kZy$c3pDd4S691zazf;y8>*P{K|38^$2W~VJZp!mXCP7V)RYwG(l
zi?$DDxDx`;2WVNyg3A56oy;pK%XfH5YuD^{Hrcj+=3p*-MNXV`N3kP+o;aLjZeQ+=
z$|&FB7T<&>i{|l45{#6fr+yewfg)~V3>CYC<R^BytuNO|VJDex(x!@55!8(tZg=QE
z=eYD44m^ha-;z{AxkmbPDkkoN&{CWqceTWnI_BA)34mH;Rn;GxhbvHpHJFV;cm9ki
zWB@qzs9wh)nK3?%;IP$NW4Q=+JZSECV6-<&L2zC(#|z&=LsM&LMd`naimb{P{OG}I
zadI#>UBpLoO~*j57)x=0KreG)5EOVEJsfSwBkn^CKpp#$tic55p)G?U#&`}BOkM+{
zB0Y>Fx5T?`<z*<c(xiXuyJgZdS*`2hGlB%luz(NWp-;!TpsW*rR1R8MW40jHOw5dI
z3?MZsyv9?@`9sj=O3Gcc1`+DHj&<k@!e4h8&C5_dd8RUi#rPM?0}9R5kt{X^fS*#j
z4F7?D!ZROKodK<|4TTz=#!t@(5hQ-pU~!wB_)Y+JV?Kath=!6A?E+mNGpDrxTtJWG
zCn0|7R^BZv<=2L&knLVj8}UG1Ox=AAm2=HT2NRR{m~VH6yc1`g7lZj@Ha<kibkQ8S
z$oDd0iEbPKLzp5Ftje;ys-(9m<FeAL8wZ1Hcu<Q-qK_lh%yMT{UKy4HUt@~mGJ8I>
zAZT<x%a3b1iDNjJTA{Z)e|sLRnRGI{5gKe<C^i7>?HCF}Tojsx(J^zkWNu2+LXU7M
z2?s*5+AjCJxP8nZ#&Pzc1mdjLs!Opv1K<_Y8lPhf5}bKd>4?<MRKrvwR7Lp2JgX$;
z?W-Ckh-h(9&kw20Qr*nE#SZ~HNr<m@(742sJwq&Hxi4<hXV53(j&Ws~>Zvd0te@}^
zGFUpM*j719#okXULB?ZoY)uT!VI^gE*@==PGTQ3zrt|f-NW~Mlr;!%k@AV84jRE?9
z*4pB(RF1h7fxt3LESDyxWMKlog<tM4(Xj9yfI!!?Po)$2KTNLYbF9E#I<+X-oInp`
zpi~0e>U&5q<aS3Ce~QL9GNSOKWO?29jdEqN2mM=kt;{yCfkj9yi=<zux3z4%ACjBT
z??So+<;m0pOJ~g^lQ+gKtU{wwkl|5-wrkeiERL2bp8<Ic%*ViiLrss>FbX$01*5Rs
zVobCryQ9AX#?Fi+d87^GRS8_vsL8yyM<?+jFU=i$m%?AA=igTshpdRl&u#5KB>%M&
zf%*SP)dQ6>-|NxcwW6tr_@5z+=pkfenHeVM0xqgYo6?|S!6e!_-C(MB*a9lWAXI7i
znNp|)TlXEYjUjJt!87xqSGs+QZ`QQ|MP#q%S7OtkU91mJ=t+rs0@V~mVe%F&7+KwI
z&|Ww4tFJvURek1XdJ2$Vf&8HEa#U7F7e=nV2Si<ycJXtuw=kSFt=`5{y`PW_=S#SV
z%L(>GFG}__57TH_07X;t72vOTt@~~QQ2u%!eNmK2OkNqF@GY51&$aIQmladABVJX_
zT<vuO#BLfkTKiy%@r<&2n8qBjtq`0i305^4S5*1X_oBF#;54{|b`TpW<>jysmgJ17
z=(oMDo6mr&>vI~mm9#$=Q9_+D^h#UAK>sswtBD8OlR4c<<rm5)-Ar-Hr{7$7t>=bm
zzgVDLU<ufb<<LHbsa{Mg9?OeXc&=3%z=yRM^0V;gEX)DTmv|Wghc=MlN00O*O-5`c
z+;WUPL7!?}Tzk!)DC(ztO2$?4;Ctck(XEinF!iwD+zax`H?M|G1ynq&?sC_8=}%aB
zcMMCbcguEEs*bAs=%N19OOrT~Cf7{`WJM(j&K99tkSSk<kdGNpZD|>B&Xq464-CQO
z&0K%^v&D7C0u9*wH#d<6es&acMuKCB&s$e~v-@^@G2H_{<W*hrwMoUP2<$DWAordp
zO$ppxoFmWum0P`CrV_3w>k^y@U%(oEZLm;lFkrd3r*-DQ9Q||ux<i4Vz>AVmayvgF
zkT91Ynk#X4!EjemPtbDc<W1JecIr=z3hqqkjx@_+k&@<Of>ZojvXfF0k}qO#RVjuL
zV}8}1!7|`Slw?m_x3Rmxn#m9ZJ(%U|#LTzp071h2-PFw;oX7gGLulH5Z(K%*+zb53
zWfr|Ek@e?g=19VHD0JqQD35PCxR=Tna7aE3Bv~K&)B~vXanZ9nGrf;R177XGdp1mW
ze_d~}H(Cq#8X-@dI266KR?~f%m~=y3-f@!jR-4pk4tZBKaykvr?|ywkak>>c)8us~
z<KPDh2{(L;Cx)eO`#6ooKk`7O2roCBEudyf2I|alOugDy%etI*4X>{6+wkNFQ_V8-
z<#(f*9z9-SF*#vhl_o<YZ^-GVC8Wv_U-!K@Oi+&Qt2E|dz8P%N%31$hiz}T>gz3K0
z29satM@4;6GLibn_T(xIK_QWbFdJ-6F{Ff+kcedGFg;c&Q90skI{eTuOj1AJSWf?J
z&&!$^vr!aEnx9qg#o)8SJ6X;8Ls&N}$9CX2M>yu9D!ERWTY2X?4WornorE~T=U#P#
zKur|thAuRL3^k7aI*0c~#6;XkxRk_+6}Fw^j&}<riG>Tr-A2s=yHmnkZt{i3naN3I
zs71fttz1jzH5iey_^6D2NNDK*a6J^Ir5MnK;@?7b6znp;x1;u5W9nAA)|NHQl2<vH
zyqU0>xtSrRTft<$E#<~eG!Gn6`yEk{$s1!BvPy{F=DP=ci)BkBcQAU1QquPaQ?km7
z??#38y#`vUF;iCuKdW~&My5S<7%aRLCfD?`WFE93KAR0oyn}>tOORHxFgS<Hx!+81
z4BTgjk}$_Hs`_q2Op>X6dJ@Ba=lODOY?{YvO`5Ef4_8NiO^NU^o9|4z)qj7C{`%Vs
zUl9?7i=$}JpJb}#v1gIu9;oIS@yeP@d3(b4k>|xm#B9Izkr$;TJLD&aP4Oe9Ba?IU
zeXIG5BQotSr4wabr&2U(@*JnHc^JKhUg$@OUAsw#gB{<@)E<&)csI*Ee9l3~7R#Q@
zDu<eC^i@=(?ZC*?GRuuIs|BguuU(!mJ|PpMcJ)Mxi(h6_vfYgQ5}tEtHmbq~MY!2L
z+k9*Ya4@a<bk^B4+A*NV<W!Cn942G9r?kH(yGHQ&bt^T$6m@)UVWwMJkYk0An#HLR
zZ_DpxgXMFT{a5X^2?k`(qsSQ3&q>OUE5vqtguTbg?Pxm^1QmsPBVS%CXdYWv47dS=
zmawfE+K8@pH;JFdyV0FFMbYZ`KxJvB*mKmHw9jCl+^z09RY{s_JaoQ^b#2BR;VEqM
z$N{m@7dJN3y505S;<x;#g~~??nuiZ)^*AQ3^|HCizYp(HD^HhCyU%MpQI^JO($cb`
zB|mlz2$Ne_cjj*ygCTnIovep`scHWH<}*MfGP(6pVSmxxRq@PU6X|yvfYp!8Idk6J
znxFKL5wYf!I`$aUPI9Hdq(8A!JU>Lia%MgqWz$TYMl!sy_W^^pq2K<0s^6%45k7i;
zF6HZAze)T(Vt}8mIBH8FqIdwzaZe*(-81x}ennz<*RnfR(FsO|)a7MphPZ;xg6>rk
z!_d2!2OLCAjPVi1eRgyI;ATQT;xQjUe>Y0+pFjQABbed!1pP!eGOSCTfJK|n_MvrM
z-XgTs3nb<S+$oniU-&($y3Lw{Xf3Q^WS}N@cW*&WX2b*GPhQ6@ptF5yRBTSv?^(5n
zIo`<2XyXeo25FODL1}QAtOOArynq^P5oSLWA>4hz%EGUg;a`+a|MJHR`N%7xxhiG)
z*8+&tlR(*JBs3myX#pxcpJFBH0)FOan%3p3$cdb>kqbi&rA3-|kyzw{1^EkaN%+8V
zSd&);Ek!BP92M9vh}=GNyq!OD(jyLfWlg@7{!H6@K+NWV$ELr#?k@!H?|->NvLMxX
zVqC7o_LrikozD~S#o-T7>_v%J*@=b938o=*kCAl{18Aih{E!K%>doJQ1vy2vhj~x;
z)g#1@d>>?Jz+~h4!>=>euJouV^Imzv`qx_gt)A${BiN2F6`zI?|07@Zcl+$|d06gu
z`D5q#{`n*R+I|0}$~try`MwguN(cY?zW=gB|M*AhWMo%&`K9##&A<EiPxsf~Uf@BN
zo@Vvt`M;L_zhhj#|KQt^C=pWrTkNs=|9s8=&A(994w3pcCf#R~{&oTW>05Z}AqtQS
znJ#{Z|84>P@mu^qwy*)J3?Cs~!USop+s{b;?^k#au`ttxT1w^q!<UEK-?f*M^I^Ro
zK%?~2Rpl(e^p{BTpBC)DuM(tltPl287-v`Ry2gd+XB;G=zqi1>1(AAPtos<0{?}Xo
zL6r#e1t2GfARMrRb>LLGXcclqC`r33Q851A@%uM>@{etQNQD@Vydd{k8QfgxLZB+c
z?YOW9H~%l>m+vCs5Yf}#pgjG*K1B9kgG1y|<&pD5|NBG4069c#CzT!k*H4mq8#zRF
z1z(Z;=d1bW>2)ChDdg+I<vae@59;<;5cl%0&p$)`I+1?6H2?ksRTmH_hFa~a_@7YB
zzyHAR4>HfU@^FZJ%)TN0AAj0k|JeV(u>a9B`+ss_4~UcfROGI-Uj7+v2yuoM3KP$7
z{gP)QPwuwGF&9wXErML75YcKH$KIXF4d<BUVGwcWA$oNApWpet5pb*!Y4PqoYhYcr
z^cXxwI!i?#ACeCOtk{zt5C|Exz~6G*1C(_E5>E30Qs{%ES=S+x#6SxWZsm}%M(1EE
z02&wt7-&0c6cj$zh(={G^7PLmBQ>@PM2J=pZ7nT!<^7eFe(3QO+!<^BFllp=MT0rn
zshW(nAJYErAW2otInfLEhz&}wKCaeou`a2w#=mNJCPAdI)E++&g)<|2c3B=GQgp|k
z;r@N8Fte9<2Sefp(!2$b02UxQmwHPPxchgvAk(VntM=N@j8+KVX?fx}1k7Dc#rN_r
z13t&j&TXe(aaqW%UjRTjL%amEHbcbZ91231FUV*XBx=&@JFvMl8!up%>{zje21~4q
zKEm5JDXumy!IQ9tSv^U}Yy@7SQBVaM0&f_)Gr@Xo);6l1?p{U=U^NW8S`Zq!2%6Z(
zv!6Ei9=&qgV|$!l_Gm9x<+dHr6-#Eee$B@uY4woIJBx2K9cz#vF#!5SMWOdcey*qk
zvkmy>29T+$S*)7t2x7(uT1QDApd1G2?oI^z+*s*D^weY0a)GkM(B%YW0``ieRtB{i
zN|6&&%p4$w(8Fj%N1e>Q0V0>hC$!HDgRA<pp0Z7kK@Y)jUlP%F*D|hd>?-VV)@u(>
z%ia-T`U)wRNIJPWqG+9cRQF4?O%#0jBGG@po&T#E6^TzF+0-Fj2LbE{B;j?duX6Xq
zUAjxg0a9g`LAG6=TcZyK3%llJT9}`_2I4oCYDC}6!Y~9TWK=!sIl@b`kW0ExGDY&k
z6vA5x5+x96ID`Rz$m7Oe!dj+?#&E|nL@9{@=s$`&QJqTs3>}P5sep%m2MKxXhd7(T
z3m@2oxk?u<4uOu(vhL&^9x`q<VCE@^tU%-|e4hgJaoJC(?SLOHG6)i2$=K7hiIYCh
zu{RdlptaX=I~;RZ(bk2lyXyv9@P+7%Mqu+Ij?>o=9>5-F{71CdVNweul?T+K-BT5`
z6Ht=ax1`}Vn$+^}v&_>VGe&o*R{~(^{R14f$5*~$wJePg<iGl$#UioOX5fQ4sqcs0
zJk<acY>Shg=@1cVz-i(<zsym}&8@db?_Y32F^4CU)H%7U2%yt<uKcML=jFuVqN)8V
zCnNp05|R$J@ULc!L5^O~Q#2l8LW|{*?AQaw#*HU_doFmgAS-o_+!hr$`!YS+%l2m^
z<X%S@cIZ)}#m)~cTLj`xRI%BDtX~@J4AX@cE0{(!lSg8CRjn4=y9`j;ryT8bZ(n+$
zU6(F`!q?eXm+EUFEK81EijlJs@5{(aeGNd-uJt2Bog=Q_hs{6qV`uco00=odncE1$
z-%!$6rhKL^puVQ^$eSvH75T0X1YaYhvl;-cZQUK`#Vr0E<O7czY{1a(1d4<c^0m9*
zXriQgMnUX!{m}vVhMpJ_rtaRd@9DT5!slpR{U*X^gn5msjM#(B_Pf<<+6UlVRHYLF
zd{G=UlXB7iaohzg8t<Y$!|pm|Z}Vf?2Vw2qr9_$)o7f6#h7jfi$09x~e5M_@s#vSd
zq4xvHX+m6~N2N90pS}F16;x8ZxyDVcAMMzyF-}Lr@bPvk=kjk;DJaRll~UwHo?Lti
zQWzw}2?;(`yV{dD4rzrKv))%%(dnpMKSS2bM0^Frk@{pf-?e<414;-3vUHYX!yvt&
zYW~u+sdLF0=(bv=GR-yh-lhD7#E9;mi*MZ*;RadBIT+sc94t+)%p~28KR_*vWXLus
zIt<J)<4Mj#h3kg5zcD+hh_C*Hyu$GiiMshk-eWUV7{an0xi8a{@;<$piXneEy3yu5
zSFOfyHP!#wqae~VS7%7+(a0f89Y#Wk^gl+7Ra<ST?a8IR#mzuMd#gZP_je_nCnJ&9
z18_iONr^Gc-_MJUH+qm?7R<3`>icu@Y{gnZ_KB?ZQ7bCqrs_@*&>!plYS{4w7SfFK
z8OOwIZ5xczJSEc>4FOiUhpa#eDJetr9$?*ErjJ0%ok+M|-o61lc7O}(N#iSZaA0xu
zGj#JNuEiCgV;BUHM%UzK0tBov!hY^eeE}ykM}$;F(pzS{b|*qG2)@<F`+WV0D<2{q
zh9%+7wg7#(W0ED0lew=eXCo5q^0=U6Y@hhjTgdvOwjV&0<6!C1RnH3-*9t5j5zwgZ
z(O!p;>*pDxO?lyUZW<9fAgQ;6aT<#lzo7T#yA?35Eb&lYce-8k!9^q@V9ZrxpsP#}
zA}Nwh3zU!e?f@;Ijrb=R%OT&nnkS{PDd98cH^m1T3v7!gu<9@M1}oQVxfWn*8?VL;
zLMF)CntWgv()g@;#VEaW(q>0B$nL6P!7>uOW8MIxQ41Sc<F0UAhl^$zap(6#&0v&g
zHL#$Y9%C1_(%!aI{{4eFVp&~O(l8%__iFyCQ~Q&{$o#q1>Vm6-?e2*j?D&|g#2t3;
z*`8+Tvm@85CGag_mJot)c<Tv=bo%LLckrJV0i98X(F`O4|MLiLc09Dg?u}$`NMfD^
zCwnJbXp<mlHN{Rh&81%bY&!8RH)u}t`-q)c<sXg!x$os~?)2`8l5<j0<-X9UpU>t~
z<iLyd!@Z=Zjtk|jM2_0~moq@8rl>aet|UWq>>iQ>u1hzmN-LgnH5U{OC(7R}%zsd)
z8uhbE@MSreK9`q@Nf~8DxuZvboVV7Lt0*=sYn%U|ZMhCw_khCWE9OTufKpE=mxVjK
z*2HX!X0$tQk_S6`HXcB$ejeul@Fd-rH2u*3rzap)N3BPW+kFSqFkUejH<kbIehm?I
zU$tw;AHnFaL7dR4B8Pm6Wx_~_x`Ef!(bh|2@f*&tyd1`vh1uNB08OTzRx4MYv@Y45
zn^t{ayBM%j>#A68X3@RlKiFTnyP8dHz~eFVh$67k{ZkXzh4uIbO%gFuY2|5O*yHS4
zL5|vF3uxaEglU<XqU)PLUd}dH`B|8^V)nj1l!)2hoL@{bn!gj?NNO7fi|a4+^2dD9
zGEyxyIhHfG<!KlDuYP!HS`r6b>XxSP_#ZaTsw&oaXU@%b<u9p7h3OfmXV*VXVshL)
zD3au-9n@QzAK1BZ@mO8A*;*gn`=+z{<qfXD9O^%OMv>jY6zu)Lb(R}QxVq*YB(e%(
z{6{qpy3W3$Cp(yqf0*g@0N&&~Wo2)`q^0jW3>vyK9Q>ScyX)xA^q^+m#YpIfIE8i@
zvitkGdS>851L@kQ@sh53&Y&Up=2zxA6Ml53p>ywjU(RnKDiP>YJyN9`<AO*H`Y7;$
zn04NSAPgw%HY)Gv&*bLe5<c&V>yUDbFJ>AGR=5*onoOqnM$8grBb8$hxxclj@3r&G
z6=PSSsv8r}$O?9`1P&@_NpsAlHJd#%KO_dKR(D3y#*OyvGuM5@E;fa7tg+F%+Gd?{
z^HpRz)DfMv9j3!N`RoLa>>_hmNW2G>XXJDQ@)51n*l%w=3y~|e<;mnXkQ$6cP9wVX
zbx*cj{j<k{bX7({03}-+u%i>CU&1pwAA56Alo**$b+zWX85(t^NjmBCQ>N6A2ZR~H
z^OD@LPTzZUjWRW>d3>Y4dHOV60=S_!VSCZq<CJI$q-zO%KE2A|L$V)5MHwS@2sKp(
zDjwS}3I88gZygoY8@B&S2q*}HC?GWq4I-t|-3SgMC6W>XQj*dwozmSB($Z2QjnduS
z%}@gazh}PhdC&Q+BY){yx|TD0Kl|SIbzPr}A8{1O?pXBtRV%vO$FaU_3QYDnL~LQw
z>L#|pA+cxj5UaP)>j!in90wh_PiMTqI|zM717BRCD&3<~)>F>0d}*_X9txJT-3l*Q
zRm%Oz9#*=8I35}U+fltu@-Xa`8C`Vz8Mt4USZq|(rWUN|5)@DCfk2cTAEoX!kyz*v
zj`ax(OeI>g8%29&1jCb=4Xh0y7xt`OWJ=jee)%Pe!+h-p>5p*MMrRgJ?%~9!j;c>2
z+h33(Ii5^Ih1=VvVpmBX0HR$iQ|1~|r@oTo|NY@3^G3}k`OU)00;1Tf?{sfO6ovC`
z&%Ln-gFLEj?%~D7FGq>XVJdiV<+*EY8B6G5%I1|d{ve=GZr8C=N4-X=E@z=T2HS%L
z^FeA~&uMR5E+RjKcW7US{dYVvYMm@NOdVcfwmRCWHr{NfejUGa;HfvJ!``_UfvWc$
z4SYGi`NU^`?!~;>wL0a*fuV@k8~udV>Lfg&uCbYtP7063%%k@x&GY9JrVNOuf+EkO
z?*Nx<_zndhL-x~%mq2VoiFipDR45NTA{0Js-5hA}iXx#b;m`E^43^G;;`NEjdr%Nn
zbOMK6`($B-f>GS^i17CnMUMLRcrf9iO=T`(Z~95P&}$cDig)Kfq5Zl)AhLOrsQ&^S
zc5lyXUDu0Dr_~3zQM&6sQO*08r&sdufoBe4Xv8J(7K=;_*CkyYv<@6C4rHz8gZx&D
z24rq$2g@97Gxte{3?8_c^Hx!;(YWmn#V9uTFL~DEfAc)utP(9UBT>gf6Z)tJuIw9K
z1IjkM`0=V<w?fN_ST^n|+!%h}6TXz1TjivkX_(iC<Yj<3SWig`Lrbvo4JIQB{KeNd
zpG7Y5zpoMn0VLDjD3`UD&v)g8m==UQN4ES89p5F9?<&z>`o@;o2=~j-=~d3(F}LN)
zH+in0S_|z^WA++Ky@t6y3u`G>;5e4I-pg9&t8u!PaXjv+*jv4pF|q3->1ioaF0E6(
zc?hfTX1G*b46wb~)@-?-egk1>D|%h$%~fAcccaE^SRkkc?7Ds~e}WGE!M+Bpr4UZL
z<K!g*qtIN27UBq7P5Z2ySwrjNgdf~|^<xPEYPzkDRw~!IK)>TjTuUyT16#qGo9&XQ
zWR;WT8>d35&h;o)1e<Ui5t$_;nKZ+{77l`7o(ueWmO%VM!wB0U=|8K2tbzL&FZ9vo
z61a8R)63Aq)p+liVN$*d2Nfl|9|k=gf<1Q)bdNLLAY>02y_TSt=*A6-uH&aStZ5p6
zT1i!JfX;dV+%@n*0Cw1TC=wFt9=YawYsS|QIgA3J@V60mI(t;Vyx9r~dNcO1)J-`)
zm4x=*<y&e2k`*JF853F7_&$ccIGF+iSsf$UIs+M8CIx~uH&dXe`{wa^fHf-*Yjd_L
z|5w``xgMFvfG1Vv6gLwPDm@(OiOAvjP;~s+=hNGi(Wd16u2`>A`_t!fSO3LpR9+#P
zWX&5O4DVQ58f?fk$&A}kSuMs&)+nxe#`-$43I1HSPh{fbGGhx?l1VCJ8XgUCL%9_U
zw@wh;+eyaDF1Dco-ND4!`xc{-_^aE!S$>9d!NHT}#k>9Wi_e!w>q;tQjjB4a$CRC{
z*EqMkp%mDrR_dn51Mo|<q8$^u`3>JZTbE54IU+!IAR{musQWIF^M{o>Q=I;3!2+*!
zFi&sB%vaA^Kxbs_(P#RC(9VuGFqhj}0CP}ub10cnqxuywJ{diZGD`0=z6&>Vy*NvK
zxghc_YZ1R?t*C>oI{NuO*-@|3@<}e)RM%aPlI-{*s+ME?liGFfb<xGMi3L<+QGH!Z
zh5qAZBHkE$z8J9!MKlC%R=H4rO>{due3jH>h3}^}cX2VHy)=b|PHvRWbu<k?^O0wq
z3g_%1e(9a$q<>*4-p2|xdJVECIx#cl>+i|9m}}wWD4p|k!)Ls%={fcCMtW1`dPhX;
z!@!VBw)9BPzPR%N*;GDIB*n~WoR%Eyj!fp()aQssWqQ$zVIN++&Dv^Dp$lB5XSawH
zTCoJ{xUhZb%dSm<s!sXaOwLPO`yR1w1&%vJx3cFICCjTi19)|dqUqF|Q{mcqds8z0
zfZ*WzTD%48)U${!-!uY=3Z`_!MrgN-!NV3VU7hxz&mx-IuekDnq$iY4cK+d^v<T@v
zHIebKj$)$AT!g3W00*e)*sSU8McHXjB6qXdwsS^!hseH;2;FUxvR`?4lpeoe;347S
z+)-GTi5oj2KC3S&-jq1{=aWb-hmZ*WkkrTENs4+f)^DH`PcOJL$-3-ZX+?p^Id1(1
zc!uY)WrY8ZW8G$t;ENvFTKLKwG<b-7{_?|5<i=hx&~buKXtN|SmnxDon4BmY@leHN
zFjQ5fz-IlpkG@Ixl=Jvk<DGFwSBTeN?k@jHDNgeLksRMBXVx0kw&S`+<pn|>K6;df
z4Gi0ydi$TkID~4+$uU`KogZgsR*Jvo3}+OMI+V+Pd=HN7HE-4sK(y8EzUWdk$QRy(
zy`555CtQ;_Nn6ja3{r48EOT3K<jm3Rm)v{DIr=z|6YJf{&Q`xTYhdr92E#Y)B!b`X
z>U9f*KUv%`+Qe3~V;2`uiL8yYr~11*e7q&|?x!1bFbTtG)SYT~&O&HD5+nbLcJB6t
zD=hy?>{R65_fcgEfzu%nq%?e|y8x0Pd3k~{o7KCc1B|V2evP5Dzrdl3EXC77+D#D_
zp_Z``{*E%~Sb4AOWOg7YudeeJATs$#4Yxr#3APOF3bu6Xdd{QIAWkFj4Fod<I!%I~
zrwE9!Pq+^j>@_$hKn}O`nl3Qj0+T0={<yFw)|Sa<t=yu|*7e-Ig+$H>F1<BQ!P;-+
z(EYl+-`QVGt}eeM`ZshOjh6(-HCO$ELJPwizWy__7wHkV+0H5A9qQ3Z?aQry=llJ}
zF6}0sk_Dx6EtnMggZ#=dBQuf(aV?(RSc#|H=HOwuNXi+5OcZS{m@X835kCjqh}8v*
z(!yYaWH?{sf{fVrij6#W*LM^^aDSC<MvgSa?_VD4WL3q9{re%4Pb4;$|B~s|b;6$w
zuH*>@vjm8&VJAo~E(tb4yUi=mg3#bS41Ezs&y>x1_FDzo5+dYrMa{#L#|Pobd4-X7
zOn@|81v-nE^#w)NMei-2e)y_qwxd_>?lr;AaqT82J|NHVYNAVd4%NS-`_~@;BSNqd
z)^MnsS|Psl{X2K?LlmSX)YcGLRh>Gslk}Mj2xLi)fMUyQ#L~IJ7q&>xiMo!{-+GUZ
zZ}l<NBVT%#M0l4ClA}s=pJAN$jg84X=2lmMrqu+q^{PZj9mpVN7SUAdzmDPaHt}U3
z+foALT1!beB7Wl@Nur<=d`^Aq{KCgA*O;$JG1eTy^#U%xc^|!bPn3$zN>5`QX@JN9
z{jnyZV|R-&M~`j?=5=}3yKPINwt0F~aXbBHD|8^eZazez*Y?xlaHvkxqp>?BJQtNh
z3Zw)S_1DK{X^!KCM>db(oG(s4E5<8*{!bhF-y5hhUbAZPZI}vC;c59AAc3>EcHKnF
z(LCWKLbPSZjntLketf8Y8W&@rcT&KqOmPBt=^qW%Yg+7Xi!u}ixj{Yut8ax0n2EA*
zbH!qoP`Q6P%*%d0Zwa;d-v}@gTC5qv-o&@7t!IN&Z3q9%TmnP`%JM(qrLUYS<fJbU
zMkTvyp0@m8S^DYRca2L>V<CnS?{v=Y*Vv^G9k!ZY<w&2*shdY8?C5K|F$FV)mSOM2
zsud~btJtgtPJB7Lx@9NYs(ax^4@rJ_@b&%NR&Ik=Y%0Ps_bbTWn`Mz}_W4TNj)ZD!
z)qIM@9gD-3Z4PrF<2H(}?=7h8o;Ie3Ig?@{c|;z}lq?GZ)cQ(JfnlTi-p0a*%5?y7
zyr8{Vvt`s?U=OgWSx|);#93v>)bl48W#VDXpq}wTkS-smdUSL;COP|@Dut)h9?J#0
zyp-$N_8fx`@V<!Hq;K4H;AA0A$KLO7O9PdSo%WGb95z2V0`(FV+&Q0Cb|CCs!)ucB
zO+=)PG@ditE`r=Tg^?UREVFcH)iYD8;R3zq*?fQz-uH0p@vX7?d#KPORA*iF$j9@d
z4g{j{eajFdzYW8M1>#q(v&dB%$!5<%W4|g42af<;oVTLG@19y*)m13dw(wEimu!W5
zT10pp=T4HKFI*9Y7%ma|%#Dd*=FLk-p@^$Br-IYDm+GT(;Enm<Rl+yX?_<xc{JlLC
zT%6lQB0{#p&0G1sY;6EjeB9h7p0)PHha2zgC37!x$i$UYbil^i*7q0xN=}kEzq}#q
z{OYjYx0Se{l#GApUrTA^$<6ZkwNK4u!r3~f%U|Y@h8>4RzV!7rF?(3%W#3=^veOP!
zM>3kovOXTT3-AlI;B}w3pBikY*FPDZ;392xuJaFXhWq=$a>oAoN{8w?(X55-<wJ4D
zjq5A`W^4_({DJ?D|1*AI^ntJDA6+uiX|PeYX?EBDbJBFAHh1xd<s{EH$Nh%UBB24&
z1w@yLYng!hs2mf7wM*CR{a3Lw32^Ht@iFMzBY?zbK4Q+IxFX=v&+8<i=K*)UcmK7_
z=dgO~>KcwPZAvb$C99j`-+(DUSAXo&A|75pS>n;*t|PjzRk6_Cb-knbK3Q9q%S*-T
zCa?6eGG~3cF)_grx|AKyp#Yvax7-iy)3mquzR8{qQh8l01*ULr#Wro%+!zea`j*zq
zv?luPz@6;5-T;H0&bd6&qFspWk3L6|m-hz_mI8cDka-kLv()|k*s9CRyTc=STHT4m
z|M@M~@0uBFA^RO8U8HyD{l8)?!A=K#`RjcY){bTQH)}-L|4P`#zHR}r;WMDZa<uUh
z>Bw6EqTnn=dLeX&Mh!LpEI5%<?o%_u_K$6LYHIIxyxd=CW{UQ}a}?ob{{w%(CSFb6
zZcAvWIYv2i-+m%1k|JA53dIS{Pu~Ak$2cABNpw0|-}-{T;qc#VoBo{1Cq*({hv@ou
z{V$jwrD+rM)e0M0_w-*IWiJ<UTp`nPonL&c@91qf3DKVT<tTZwhQHmvhNW)}D;=X!
zt?q02bB#r>4c-SVhAc{l-t9^nove9V+_3qDzM;G!{Baj~09W40>E#{L6|;FswsCD!
zJ*Kf9|Ie_v*VD*;X3M7DXK>(o8GL;iu;9sTq-WGUkQ1!}iVtdCow97zJX*!rEy=%=
zDA!JZ;s^`hpbn86V1E%{Haq@27VNZNtlQ36nA12Y;#IFAv#HQksFx2XkKEfjJgTeS
z?_J2TaBJ3ol^S`T{;^-o;6>b@uij2TvU9pO-h&cP=srZRG2amTq|g6+@yZ%(F+(>h
z8xsDU68YSA%~P*PS7Lp*<J88A7ZenXCvb;Z+#4H@jt-NqAB&g{<`V`XruE0hokm`-
zsvCXGH+6M=A-wv{DJh_0j$>xcZVXJVD4UlyWACk~E<~u_onCI9f3`anVKsg2<SqZA
z5=r)c@d2rT1;?P}(*|k67g|51*GucRzjKy3-zi%6x^h1JMFoX2dlu9w6mEOCSyCmY
zVciPR(t1<f(VC4(OY3mdG{wM7Pf!0SE-qbl`;5jsapdO46O(6^fnJ(FjBZ9QWS5<_
zG`^=b$i>^6>SL;*r))r8yTD!Vpqsnc1k$e4)6>1;VG5xsKVMdU$qo9b9BsaKaFsw5
zwwqe<^I`t=3wq~5Q#~#P*7LFqs+-!{T2?v+o<p9>1^3HODw`NAD}uj4y)*O<+X`Fx
zF4xAj*(m?ue?Ny%-1WFQkIoJAMXXlA%!F>8nLbZKH8*vgq4@Z_ixHMMJ?Y5f<8O3Q
z+Bm*<l3nBsWcydTsV*Lrztj-Zcz}J)ikh6C?}sxA4nin?K&7VB)z<g-IrjGU^1AOQ
z{{FruB<_jHf5*lf_Y-DR_z%%a(J5S~&>v*a`G4TNEc|H`j_j}W@wWXciD@0YYtLFb
zu8@+F@`MW0Uvi_m^G-Ro`Kb4YgNZy$?A)1tHhj;v)S#!uC-qA!EA%qoORW={-8+d|
z2TZUqd@@!jCWX(jY2pN!E_$@EX%m`t{Hic;i3kaKoJq`FDT1l(nFIv{E=kMByuDe&
z1(=cJmsL%5VM|5f1EJ|_o^67hjXLcDKMQtEbISAYlw#(cs^~mTJbh}gSE1m&^6%g4
z@qFyp6gl7*Ha4Y}f~xWA-&?wq1->W*4-x~fo8^;Odfef|%;?<|S%=Z?_~e}oY-}~B
z0#UA#xwG0%AM}q5KNly2i>GMazLK{`&~-Ui+>+PXN7U${Zs)uP=HXb=uU{$WG1fw7
z)Nb-H9-VUW*9bLJ77Pd9iz!N8US2y0ZKf-=PM+{!Eh=+W&LCqS=ii{)QQqX_5wrDq
zVG7M&{g2zZg@uiq#+H^1T7m1@jiYBu^k;-O4>6LjDZdd-4j>ku#CtTQk~EwO3k$E>
zTb_TphOJ9s%wWw<?)<8%sfou*+!p6+%%L2Sn=}l2K99L7<mQ0o&~O@l>B7R9x{$ha
z`HbN!zp(IM35c)BW$s_H-^mVErc244=`)Tfsr0)ty88P06B83uRZFv7)~3SPwjQ3I
z>)srWDV(OJrokdse^0qPQ(LuxU3#uDM4R`RTp;q~RO_}gFXiITId(BQpUKhk^0FZv
z<&|-ofDF^)q^yUYF9na;o2!QljzQ;wNm^OhV;;^a4?68Pbj%F1@AUNtc-k8Fa7xxO
z$GoJRwbUyPR|a1eE!mATBwS$jaP_EQ8=iz$`a~tTH+dLI!S6e}ySeq937#2ZtYV7Z
z)f(r0v=>?7m6)&{{W-vDI??qY8M``V!UI*%cPjG5Bu$!oHQArtQpkN|V`F2g6LE2o
zxK_WYYq7Vzz1^42#--$K3A@5biN4ed3Li{=Gfqeo^L+JRA_9SkMG@rnr=_LsuAH+6
zJsCLtBQ#BNw|(5Wg(o3BBje{^KIIa!oluV%w~^XU6sqF|ZUjkMEh|wQupKF(hZsr9
zG6T=OI5{~_gwObXudHScSl%lq^x*A*9h8ODYzF9V_lNHvFnd@dj|YYmy2vAC*4q=^
zxCNBNND8(a=>-CSHI3E!U++C~3drZ-BFmPP1T3#RckbNYnlgq`{4b$vHx0Lu@#W>Y
zAB#FMySKM@_KPiFdHtdw&~nN_+jI|)1+NZ22tkLhk^4+hE~<#*wZCqBX{EQky5^(b
zAx#M`83!lHCj&icN%s8=rU&)sqL=RbX#%P|J)Fwv49cD^%OM!}{q@VBnXg!BjQY#0
z%Hdxk&qKdqQ_+*8fk*i4<>Ff+Ubeu!%P9(q2(B_*X8}S!WTJ8XWe3^-j-G5e&>Ud0
zmaDmg5Oi6AD_X<Vq5=uWjvc-8hMM#1NFchu9XSoRasQ*wb_2w0Qgn1H6j=95@ybqG
zK?qH^5hxEfB0I`(V8S&7=(4B4m|wFDlvfTEy8R{u5Q2*@Kpb{ef9J}!FfX1B9nLIX
zz!)hyMt8-Po_~j@O&_Izj76iUqKg(6HUWlIE9629+jVU2z-7QK<R9<M?k!zVKgc4r
z6gH#U$eJN@tRWy4SmTUd?pm?znelm{MazKpn3Fajgm7AQnU8n@nw3TJ$Ntp9y<?yf
zsUe<2foP*z0T$2`A2-y?T-9wsU!$hz$sg#=o1<nRAV1A0Y>56L{Psiwx=DTDET_&R
zq&E;383Ek}Qxo`PI)2TcmKr9R#H&m#S7(6cwds7cts{gj>ATk}7qs_e_aYBit=0O{
z1Rn*$9b-W)oH~^$XG@_NLLdEpv<zNWSOFNRV;OMdN(_&{$Ju=G&&D(7H0(xK3l-9G
zivIXd&@3*2X7x)qi19KS-{znXM8UYN)lS6L$Z0(QPL!V2AFfl*9|kOgs2>5RqvH)&
z@iim@eIyhZpYu<Jjf%h*{Oe-=ym_L5H2rg0WwNk~ou(jVQf}i)Yh9Rg^W{IA&f51a
zpd~(b86?L{cE0qJh!Othd?5vn7`-1`z<!c2UFxSk#b3Oab&J}p5w`1<&Ue+DAiWZH
zU!{k#YUnzu-`Yn&RlDr`-VBJCv0B*{?dKO)JaFGu%{cBU6lq?2X2Xic(@mQu&*zFZ
zEk7<}0YD?xYTGv(2O!zEyaK&-J_3u?1_MiF!HSMSjesoZrSl5q!aMilm<$53k)zZa
z3%>`vn-^4qevM#r2rY51G54a4h-&_HLAFr45a10nz;g%{uk|2>E&zCzYB9i4A^Cuw
zy+hu1OBjimun$CEu7c^Uo&C^-KQuIfKn_$qIQk0&Q@vfV@;;*PK=)%BqGh&n4|WjO
zzl&n&EJxR3H~$=d_4KjocO?}{kBehqA%9m4)YuLnzT}(N0Tlh8nGO<V;jD9o86e>5
z4(=nKNA17noszZ%IUdFkpdbHC<oe*-ua5>SCkKG_doIttDqVD+(|XbhuIM4g9*HFo
zM3co11%PV*YC#Nap&pqcyPAbT04{AA-9fhFvd)SQ`Yi*s=9j2cvIYRIEu#@$J5S)`
zWQY5}s!}AkuI~hb?Y<7)ur=a==qj10CAs^gz(^JL{pz)it17qkEdQRx8ycXiSXY2n
z(8u8a7~1=!hI8G0IZCx=r7q=K7D(lp&~B^~eZU@WiXiEpP?4in*8%9|_umop00Q)K
zZ)dF-Dq!r(y$MVxc_$ybNlpALKiEr^&Y#uPn$sjH;{T~KA9-;*YGmmh7A}P%P5(0N
zh$ha3z8rs-b1LfUBX|M~Juk0~_gqvxWO9LXKWsfl&=pu~9)TsahJ*;98;TVOeg-1a
zgq`txjHipb8R&a)e_fhCs?xmCl{}d*ipU~*+|A>hd0sA&ZJH>A1{a1-{WR5NB1V1H
zlL{ErHt6AxJs-l@T&z5|A0=w`yB@;fE`1LGQ`(h}dl}cJA_WFaGdybYcFdp`O#J9E
zmu+HFq+-i9lGg-S(#8a(T_vxmT_GNuGh1ky&?ou_5?%mOrB>_BIzX%+7tZEhHgN)a
zU~0a?9AgBGI<u;55k&4#3^$HT$@2~jZQ@G_&rP|^lW6({P4SJ`9%;C4jFf5T<)`Cc
z=BPz&18FgwdP&W<`(YU^Cus)->{(sWj7XeN_R64s8Al^dn{Sc8=~d21mWw;p3Lmj5
zzd1HIO|XxZ3sGAA6k73ocz;kVr}W}1wE3muI)FxTUSC5z6!8PZ?_+YTt$8*ocI8Ll
zIRhr%r;SZ;w0Lf9{DNWf9A;shV^l8q;&T)P%nRC4$Jo7iZ}MANyu(jxIlvu4biX&q
z8{7gdp};L5A(VOxo(j_hZCSP#z}}|z2LUc*)%e(EogDkv6ub-2BWt1Mxo4qI`R(IE
zmg(pB|Dol!{e`K!i+au*&iB&69Z8Ov^TbEb@yq?oJ2^bq3=UVHaR?YPF4~y`s(y*S
zuF9|KBfrnR-wx9heS6O#_vu%6E|$wy2$sNqWCQ|7%@+H$oF9>nmk<;w-hallW^UVM
zX-0`Rdg?wmFppVyJ2z?c;Q4Ryuf?4{f*}~CY=a5@mg;g`C9bt8JFy#}O<YwNNTzWp
zjxeFsYJKTeiKu%77v$XByn(WV5UJkCp8MsAB+}w|h3ub>L({mN(frD>7lZm7_`+B|
zxPE+<6EDb#%Lf%Ub$FtONw0XQ9?=<$RLs(j+DDqa(oU5+f1ZS?;)4WB^`e|x&W6qG
zhAxT)wA-l&8TQpxDP%BGk_^c$vXJM24zRpxi57w?O`g8O7TzuYqo&NmtwPp~-|#B}
zulwpX5^L!TY|v=)G{yt%)x)2*#sBYEdm)jjNk?mB7<A`a!g>nuBi^^593$8fJ=B}b
zKU=Z0F&Lz$zws>hf{=V>ZUrzqD3bLa3K2~DXocPVjr)yuRb=`p=~J|=ko6sdueC|r
zB|s?N*H?GLuxixjf5`EDNn%=e-|sgtU(VOweI}}cxTPH#$v`s~b5~4{e?u?0^zUrp
zV!qb8quQxIIf9PVU=}CxyWd=HgaR{+($$qxjnrl&mcQ7s0>wM;)^qZfz!jBs#UQd_
zTq_1S&%hHF1GpPekM9kp$?i$3$olim5pmvnKv`HZHoi*pz3V}Fd;^|PniZLPX|CIa
z$z<isVn09r8+IQg3%0sbZ!CmM&aT$Be%Iw>`$rka=Ixt<=m)(iJ(hSilMh%`yUSx)
z$+;rqay(bR1p_JY^&SL57X4v0xl+6{N^>3A{Rizk#G!6~cr5LiWNdm9bW+&@Z|FB#
zUJ%ifmvw)xcV{Htdr>Z=nU2DN;c_JfqD4Cdl7u<x6GI;8V`AE5XZ<kQ@QE?w<%_(C
zV_PvBc`*u9zYfnS>C}28kLmttb(u#=ja9b;RRJ{2v;g|If==-Rs`cz{KRZK6-(;WU
zt1F2)@pu4vgg;Y7)FzZ1$~PmqMGf|o>TLO?DEplKJDMls0BVxQ>SWLh*vzVzgWb6g
z`kkMTE<hB~M$dnLfUPLsmA*#I2a3f02D9I(QaI9a(Gh*^IkN7udqEhAUH)(6tm0h+
z4>L3hQsEO9T@Q8L%))%jHSk7u^T{e^5)wRLi+E&2SE3XrR)>>pk*=t;w_+Eh$7@$_
zx*xdifCr%lCF7a3<w}*pYJ%thQzgW=!d48^WOL1z{&=u&*<y!tV~c2&ZH8(snvO(Z
z+7KC2wS@SMPNCPF1&a1}iJoy&6B%gd?2{97B_P7)oWt-~Oq6A^B|U%>vLqIp{Ls@;
zYD!97y0y}@UPBh!c<f<XAskzbiAn*Vx?13|D`O1TmV{{~+aLxrDoXZEqwOKJD_*jr
z^P7(lg6&?8h#5~xOmItZX+SpBzvM>|kt-P`etfCOj78y^F;<1bBfm<*3ktqms^!Wy
z*!{gp5~y?8ErE^AjAS<H4CbFZI*{nW?Id9$;c6IvQpT0Mc-MFA*N5)7NUr`-`>=OP
z?<)*xA~BTiEWT{u@M0(Uz$n_z9yYG-Gp2{|p#2$tcQqugz$Et>P7ZL#{-8KQ%7Fg?
zRp)3?BM-5^3PJrI8S&>g#=bqzkQL;QRwM*)*zuT)w{b!RB(RMQF?OS&LCH#Ii$~s~
zj=Ho~jVEB(%G>8=^juyl54$jvZoMVa<w%k@5A}fYJc+Po8+m+_<Vk}bsooC!X32Zj
zpLGSx%d$9xzj=H?7$6-sO+Io(Jb2aD-7ZSPU$ryu91)$Nz~AjtMWnu`x$pi0hB?ga
z^VxSJs>q4px|_M-7YT8>f)_12rShXS?Q3qBn2sZ_QbHQ|KwG<|!2x5CsMu|%6f&^2
zcyG*+5QE%nuvr2ggk{7`+Gb!B_?L9&O*dYY(4k%DWC=T=@Z&5$BRYeN5>MBfvxTf%
zxdmz>+6qmoY4&p8VwH*7jV7Kp{JKG$Z_0Qy>KF&}?m5;}=iterA8ceoQ)H$O>*or5
z%lpw==MuQ9b`r@Cn7Qk}v+06oyy5RU%0^m^j#cbVU1#}KQJ#0CZ%ux&_v@8OeDo*q
zJFny!{s$(f;uF?YzZPeuzVo*`avVvS!Ua$h^ITzC)giHxpU>CzSnj&q1u(1?Dr9fu
zuPryD(EB|Nv1CS=@^hQNilVUkit60z)8ERW-o@N)*sQ9H_(q8LY15jt@@?dwJJ#&m
z{;>p0g+-p3kxv5EU5FG*pt<)dAK8r>GbeH>zaG$%e?^H;oY){O{{1Kj7b6XEd8g-F
zRc8&o7?WXhL$TGO2j08MZP*F&;KKRJ*0{KPGFGe?I=9rKn9@rQr6$MiteJd?`nlw>
z>W}o(t>wtwOx6+n$>EW<zg7gj5%OdvH@<I3f{HHh&%!T{N%*k8((O`pbKJep8CV%*
zaXtJ)DYvG>x%cA{(dx}-i3ct@1O5FUCuhuBijbx-LMLlH2T6m;U~fELEZZ>DHEgnY
z_pD`wKJ8=rJrT^|p3{L3&L0gHy%MOkw#Z$vb8Y>C;mpauqYtD<STj|_^T`Bxq7aFL
z^!??23Pv5Y2@bF06eRXXrr7bh;_D+}SoAMk&dGu4od)?pgDd_Ca}bf;-3Rp4_vcO5
zyo%X-n`io79#gM$hDbL*(IesxS~2|!Wxx-cY9ztY@E9~V^hOt|*S<V97&s}0+)<uQ
zKCw#27s6X~YbTZ~@WGfZ5=>h9vNsC>FMmEm5#AG7dYU<6Q_bzS9F(iR_mFX>?|eF>
zc{VCGo*+1|zBruZD#SR*Q0*lU5lp_L#uZ4hnm|Ldc~SlL<V&!$@?l7hqL?oq-iPu!
zfcp+0Z-W&|MP)w!=j3E7{|~p$2a01i+$ym5iJYt7KF#wk^;OYX?6JR2kH+Jw^AE!A
zxKb$KU;klcLN2!9_l+owr5oQ$G(Bq*%il+oK%0g3IeEsWg+J^i^n;GTK+}he$7fLm
z1(c3H4VVgX&6KD?bjZC#xG-hW{{uFsoe2$h5LAjOpc`n5wWk8eEG22KJ0A~UQJ2L0
z7!Q2_r?nd#=+xFU^UsIkWcfUEeK#aaG3kjVKy-?oVf;2|=bD&kKvL&q;=01}Vdy!o
z3{y8QztKBmI(%hVloPk*Thhf=4SIt8A3@#(`yQUCXxDP!izlk5*)c8;!a5B-764LF
zHN4`b<)l|c#s*)*oXwcb0)?L$`2~k-++E?YKtmh0?jk7e2M7lx))7J71^0V)lBIK3
z5u&aS#cYJO$4?PQ4~e&x@zg!3NIf~Lh2NUc$!YJNNNIWV(I|CmlAGZUfCxxQnInE_
z+m{BBb5zu5#cbxN{c+kb>Bb<co1ejD7ZT+mC4_Ob0HloVLQFbr9%!#??b`UBz0#wh
z^qgAm#Vhf&>|0!y%vay-R<Tg4j*KKTxKy9TcWPVo#yr1i;K8b@N4;t~bM}rfl5(h}
zGFH?BE_yhRI`y7i3sJf45&0if4pXvUO0@*+eY!|RpK#u=!3UU4aKo>*>&f!?E4%BB
zBre{f_#Rg4_XbC7y&JpV)sIL@Ba0xi{lH(rA7Y$65`{;p=Zz<SrE0Rm99c!S#9zLz
zI!R7jJHnT5BQ6$qzC-$Cn9QU&gxb&6or?BoC?G!EDk!5F&nUa{i-%6`UAqT0QG#!C
zyhOVy?vei0<{CHcFJ$Lh&nyf7Qe7FvHe+IYW*08}oG>p^R=<YrkV;#Zj?x4b^(b2v
zo|Gu;QhD}#>*|ABa97MEA=+10Q(=(0@yHHNMb_fRh`BzfneP6kX0BeD-a{9;rVhi6
zSGVPBnzx`nb<(ey$SYO9yQ_xSJ5r;->Es^O7v>%dPaEu-lz3+h?V~!%8zss}m!QS0
z%oH0ZncUMtP^?oH7fln3F%{X3x7BqG{J3g7Kfxo0EW>!S*+k~It_(0Qqc7b#O%Zam
zjT2wLH!$V*7c1)LyVY+8-bOHE@+kW@PK*iM!o?!T^-E>HFd!2CjBZf0AlOJ)t+;1t
zkfsRJqIHpOdi>Pu@goz|PYA9MAGRQjYgbyFV8VG-qM*sZ$l(;6QMovRR50SGV3S*_
z1G|M=4k5!v!eDQ22at>?u=PRJZc(*0=jY)4$*^$vmSmY9BcBQ(zm_Irz$d76Pe9m0
zHCJqzvCex%%N*RfVo$DAGmx<~Pq^&E&+GfKUy+j=AM5U)h%Oja3;sCR2^;x9HJ7Sy
z*TaZ>(7bX>o;l)j+#(P_-hMP}=KN(qIP@)!Diac>o2>)8H#|h-cU!}0s*}265@`3e
zAXpA2xVWh3%C)Fz6$<IJ*!XWy54(%~{r(e?(UFS%kBq{8ll+VIA?KSxhLdZlgoQ-p
zb6V9hDe^OaG83HG^eVbr6D`PBw6qk2Ub%(d<XbW7O>+{=MztV`&G@*Ee|lh<XO^@D
z;hR&f2)bmVzyd*%af*!*T4l8wv&Rdgv?Hu%B7i)vgQqB8X-uRWR(YGrx<%T41SZL$
zv&RQ^QP{AE51UW%!J)<Uj7yxJ4}lTIs39hCX8O&?{Seodoe36Q7V93n^lgR0yzMEi
z2_AP2r4$a9aE(;<?&e0hnQw!S&Tv_l^=G{el_&1+%zRWrO}X7&3Ah)3K1_FAP$*3{
z6mNMjfEhI=-*q*AE*d8{tn9HeKexO@AtIqqB#=);jbog+bA3cG@Rk$x(yRES{RIn;
zk7druW{M$8o(>CcCr<3kEorjY*YUPQPYy3n<vJ;poRJGR-E`S-3@OIWmkv~&UQEml
z_sCb=y(s7U&(fU}awEQR_QpT^V)nExSYws=@??b;L5w<$BJXn!wpb#679VUq{O*)M
zR^!%V_=)S|Q)zJobM9;CpqwXFz)?mNSm_zlX=*EZwx)5@)v0Pek>a+Nv{6|Y><Sw&
zwBtSa?w@K@FHyL$2;TL;c}0#7yBLOF2nu>$JCE^oi&@=|*-i^;VKwRdg-%IrTTWMe
z{+2OEyy=!TpkSdc--C<K+w*H8c=OfaVGD}brHf1ZaRDWQ&#db(Ueb%W4T@58z-u=n
z^~>ffAAs1T-N?Yfd38}04dq%PI#67aPVwrD`0r8ztE{q^cOM$-nTFf=e~Fy0Vvw@7
z`!o}ymMd$%dUB$YDmy%Eg{!n60+UQ!gaScSwyCc@pVy$Frfm8q@YY%~b)tW%0;*l#
z!{o>_h608-3=^^z2DdzxPZ>md83`c*<TK)6$zUW7*wa`%26NCg@%Ir4CF)m?-&V8+
zU`IIm+;w$KzewqIn&bP`)HR-FKp;@6nn`$MIVElX3f0X@0%-vIQ_j6B?s$psPyC(-
ze4PzZ4NvTTwDF5ofH|V;cK<70qN2XbF-EBG4@qXC?*iq;NGHw-)$Iai%=qYED=!Ri
zTX`{idT~2ngfU9a1hkMrVcgvPA3nHIz1<Fsw15P8|DEW>poH^twt953ZHDkj)<a3S
zBWVWC`_o0VcQ^>$d*x9_wf5JYyj?n?Tic9W8Vp{4E&MQ6Ufz_u*@inkrc~55E3&ZL
z{@Z|XY)JUwd2!%OM?3SBsLgo-+1G3wKqDvVC7|S3dZbu}e!Cy>%#qufw#|@3im0QW
z--b~?@Uey@h6YL}20W8z{u_)7W0u`XAEu2_h2vZ!BkDMumXW{m&+{<8@EDji(PMpL
zCV{~lb@;cu1w8wVt7C8$dKf}zOuVemL1b=A$S~-0t^|{ITKVdn$NAhuNGQp^V$MOX
z>Qn<fl?S%wPdnMr8LU((u*FaKTbxNV&4J_nz+1Itj?o@$+Z_sK!;3FG(Ywa2b52cL
zL|c>FUvfW`hSRmvdcMKohij||nJ>QN;GX7xV#cph6!IgAFO8frm3(nV?{VZ~I8HB|
zY4o%~DR;Zx`S`Qp^_S(+nI7X168feXF#H9${`1miIQ#S8%*7odUMyuq?!UR=57}rp
zm~sl`l^5=>)+NcwCpqvgu`rZ(;NhM|zD}X<G-h6n8h;5N99k~v3=ZUx`=yXVCriRi
zQZ)Vjp2A>H?<eN@%2aZr>G^+0el=&H$?~&gslDMlI|Q+Rakf<yIVR#)6;v{0^rXm!
z_a82bxcLbV>`(yKQq9rR>&~oa`YtSMVtZN@Dc7!`uMox@!a3Q~-a>*^5FZ`2`Q9UY
zT<+!z56fjH5pAc|GfE_P>^%>bq$pcS0!JR=gYH13?Jljsh1Vx46k6XyM;VchV549s
z8Nbb@kwFA&6MVuiXzYW;BV4}sX3bv|uAV`dL`)<L7d|mZWw$WmurmAe>?zZc&91nn
z={Uwsh6EbFERMnWK|&vQcIxl&-9zw*W~u#0!8SHbQ(z%7H{)q;w1#CRp1etxW7e|2
zppsY_n=^Z$;>3kRG$A+X_M{JzEyjpYH<Z0^nO-D8{SUmR`K)Y{*u9nb$tF{~Ywk9+
zmh-A{o1D#aQet|Ov7C~f?l|(_+C~(el03w#{*xR2EIDh6DZpaU+(=Xj>FLoBwlX4z
zi3=sojx?U6lDnokn>u!f6ICc(?7WVs(yjU+8l?y8-c~;TJsC_ZP0cJzCql&i@VF@E
z-rKkdowXq~CcByrV>Kr8G{{?ZnoTHe$|ePOztHy6({DU^f9<~3LB9#cQ^vZc|9A!;
zLXE1drZqn>vh-AytI)*jQQcuPnH><%h;VJ=_XK-JYmbnMUs2-Jf)Y$&)0VU)fvfF1
z$xN3~_ZXSA(=Dz@JD+6G4=guHI@-7o+%j<>yxH{}Bn=%VbKCujqz1B?;~xt_P#mjC
zv@!oaksKd%bupnZkfVs;-w`&drBySe{|P~6A$N>)lep|E|5*>HaB1Wz(}u3PdGej!
zoqng~pY5T1#6L0o6=AelmN{agEmqCyztFbs-5h`KBQt(bn|9ttnCfN%(zDVFPy6|0
zLW$2)$0qhwd&&1ZlQ@_BDOBc~PyP(?#@$6qSDKC7vz?{3>xN0uVB`#V?AybcY#l9M
z-^P0iS^ad*nBc}eJ1aW(@UM!lQZ&G4^3(|l&uGqJZx}ja;!Wool<o87+@l8_*@IAs
zvbNwy$Tv9o>-ZC@cinUt)h+E#m0Ugufhc?}0t!9rG`_e!4eWc&%UM%B%bICBZD;Ca
z#xttSp$zX=S`MdTk<r>r?LrYTcTEMw{5>IE5ICve4@*@(<E0qzXHpJd2Z&Z4uHOtk
z7r<<g>#l$Y$yTG_&2ftQLpFId!%)-ta(0KDvbPoNSJs>pZ+BSE)EuJO{ndru&DB;!
z_AKkB9h{yNOG50!`pOkusWl(tI&s3xg{GGN^wjiOMS@WgN_?iWUTES|7_oJID&W{&
z8d<_*O&R@sz5T1@)T{dqlyjWY0&%GE@Y}TRer7>sT0M-ksNdSP6@j{7-~RsAOnY70
zpwyR}!hDPOjk3v5L%)ZCb_Y!`iuxC*4?mWq4ELa<G*CyM(>s5Qz&V|cKBt42jhI6z
z9REIxAiQ4tAoSj=o(3s#z!_LL@X~Wjo$2kg;qJ$<iNu)?qWK&_6XwkQVy~`D9S^2N
zqzv#dhrNm13}kS$I}pcYr8KRSdYlKx#RQ!JtCThzhP&!E3dOv}jE|m@*TKDs_gY9w
z>7spHcN$vstEatpZk~8C-0%=wfBECz+v?SQW#V#gVe2&Y09iFgEk#h&EHPEyAS?K)
z!7z&c>C@h0S*qgut$Q`MrJSb6|7u)@VI$5glrT3nkMYv~SyeF+2RRl@<lO{$KEwZT
zwVtu|p1e1G$<4%Pr`?azLL>B~OCa5D#kp@@xk)xfBGY5_m$Ss9%J7Dh{Imn!2d|G+
z#+id#<Yv}w*Cz`_@-ieu&9JsI2IGfo-lnUYKn(fiFEgJpqhfLGB;4i)-2M4ih4Y0E
zJv+CHi$q_KowE>aW6qHIVyyML)S@Wn^RGQJf~x!-Q%<Z`i`YEobT%lRjb&*H^>dkb
zuL6$4uLH7*IVnXC23A~2I+q4gx@){@p>o3{IOW8Djv-}1H%ZnNcSsx*`CwED?OR(2
z(ZnmkZN~>FCkHikog;FlSeaW9<C%n~UGJE#?O#WH@C+6zmRo9ilg@v>b#KQ;te3vC
z1kTjvaEbW3g_BWfx@_am|By&9;TX|?!wIo%ToPj4$nDsnZB8>{mrHH`T1puq`n3zo
z<&{;as)N{8=4?TXrSj?y%bFH5mMBUfVjd0|e-7nr$Sl6Ox>~AG5H*}Sc4QEBO2Kg`
z>Uvn8?IqtbWb|a(YI|F8<i`6vj25S%cQ01z@9zJs7vOk^?SFJ(oC$K_T>X0eO&c#j
zL%QCdP?B`5um1BuE*&JTSRwkg>p*gd*~c4b8}*fZ+e%8_=BQ-+08Uf$o`@+m*BQJ0
zC+R%#kzJp)^!ypT#n0tI=ch&a6hGW)IL5)0^Wc4MV&6j7LBqb*%2vnat`ok6?e$0}
z{x98MWe6j^S4)B|H(IgF7Gk>qOK_T7&}n*E%ksIcd?6GSbq~4H{Sw8C{L}r!Btp}<
zae1QWiz=rsH`C1*P8tmRtsjp#rA6m%5`FU`b9%Vts84Ay@_S49)7p-fT^?P2{JZKG
zrH?P{$`i5qs6eVQvbY15=Xp^l8!9J${O97H@>XAvspB>)1|{}h3}jNf2BT?c!%K3P
zb+WTIp_TAHLdd`F`&xLqgmpK*+-{90b9jSu%678n=HCVfjk!K`G52fF*%|ZwKVl09
zIj0Fb#59N-yF{eUN`jsVUEZ^gfoju_U<rX*tRkkSAw@S@Tlh{xkOFao$y%IJd0>+K
zA94Tx(fGMQ?92y1wG^1sCw)78VQ;<yDjUrYKc||DasxbX?;igi9~f5y#B3T08OWO!
z4Wr~M4VM>d7_68j2OG>(>C6d*VlJ!y#$`7FFfJcG%55jW6J3=P?jC)JJ2|c^oKyP@
zoAs*H+3{&Rr3WlMD`56a8niNhB&gru<YMi=LJ_b^iz5@3(@G%+baBo;+%dt9(ySR`
zL6lLj_NVE3zU}Q!#mm2&?Q9<olv^DFKJ;(W!uc*f=15R&ErV^zHdB~b(W=nsPP#l-
z5q{@!SD$~CvsqE98N3i$Z;Xjas2!pq(B?fD&~I=BpYmYnH{AK6Y6pMnyO~QyM_UHo
z=~-8?hNJKmmp@V1jjAYP`?tR*dv}4w9bPhw!hD%d7U|1&g3i{??G*K0;8>B^kp{*h
zmsTiVON$?`_6%5ZDkk#VX4gDewTB1j1$#%K2EQ(UsB^NV41I*Ge)*XrEh)sl6uC6=
zL0UXg8d+hUK__<=Bejr6KS(2m9)XPtVrvpR1T6VDX}-=pCqeo0_r<TuVuyKGWQ0Sw
z=nUv5cNu{l<Ny7}2z)An?xf>$6|Jab|0U-q<*pFEg<?9#w5>3b1OO>h9;O|w34edR
zpzXw)Ow5%;{x8qMg|ZJX7lF&QEy>26eeo`yw(l1#ip)okxuUIX<G<a^>1+(Hv75Fo
zKStp$urB-v$1#NJr5ze*K2Dqkc;eG)eK=Hv(39r?R6>Q?Y#msXHPxVXcAR9qpf1Ri
zUh}Z3$h@72#oy9kZ^%bynzz`il2fH`nlL1r3-Qqui{}5`yss){h&HYC<5fqi&Hnu=
z!$C*eV@Wsjcxq?s7!XoQPgkU%$TzT=NJ1CQnuKS@SITjrMZxQ%$Y?31!(_i&v2Ip%
zjIyyx>Y8sDKhn_{VEk^0`78S|`#HUXIJ`tveHAD5tbRXp4)!0-bN^1M97jnohEysa
zIIWqQXumtP?&T(^!tWNvHf=1{yxWLz{2-{n1Xk?{VW!b3^s1b*o?=FNK4?3~p|s>$
z6?$6MXlL7sfm@}Uzwhws2Q%A}`Yym(4+#d6HGO6dsFA-y<N%eBMx{ra{pp8eW9>!B
z8iIX!Y^9->0uCP&*O@lZ%bFTu43UQ}5wDSt1ZFapz&@E}9js{t1CAS76ttC~Zf*Ye
zrfrBsYyvNrsfLBmvcYY8-H77F-bR1_)Pha}|JS#VA!VRM-%9%hm{&)}5XA-hz$M`L
zOH;$=_zDd{9=QlR{rg;?rjx>=Y8A`x810sb2}j45wn%x5sDBQnd&L9~>9xhTRD_84
z-?Wz0&c7Gu_-zz<_rGh*RmS_CW6evV+jriTV2N3)Yoz_F=n(iTs?*W1*HGXDDn6a*
z>O0%8XLCMo1>#ld44$O;rZkm-wp<N1diIE`3t$$4(y}k^h3(lLSqJt%3nk(-iV6d!
z3hGm~&Z|ZcA0H@UrYH6O2c&JmC%7E-^3tMd?itQ}Q^q8W>~;~FnOx$+%pyuS@Z=W?
zBTv5j-J3_dX<8Mn>YAR;x6mptd64S3-1?Ki2M@j;1TGwJSUyusbvE&5XsfI3rBCq+
zbg%}-j#t<G%R%(9i^~VL4EhjfK)0Y|RMi3g;H5qGjly?Pvc29NI#Fi=k!t}pDLYDo
z%hVxz2|JP`Q>CG0Gj7f?fJA9j<!SK8I}B^+4AiV~&<b(@D+${d&z;dXPS<=u0wGK~
z%;oG|0sV*+!(V(?@s4oH5}`oA_ZSYOf)&t>fPi<A^)o6_aVh|sjUSKWF?^D~#&U8P
ziIhz<+$Ga9y&9~HDk=*iqkFi(z-d5<n=SxAnlCbPSxYzS%fnh``VL|RdnI*_9Rca&
zDIvO2jbm5@DC;u_lF3ZJ$!|`PB4^~wk4bcBWzck=Pkr?uoISmLiPmZwoIs{oRk=Xj
zB>BBqYq+oqnI^EZU>;OwHA_GxUQ7W<KK^mC5CD$j7wsI9%4X0KKrmp$fp$37-B8y&
zYi$!{wKgec=~vTb0>rfH$cCV%cu)V^y-0Nv1%buOXm>HN*DjG+h{g4n6{eKLls*{Q
zplxQG%gDB{+x)9m3@jki&`6iK(pArvt~P|9hv0QZam(MPx`Q=apsU=<(FHc2Bz~t-
z?W0#`va9NFme12C{D1_$3Y(Lf*Vl*%TITHYq57%^jz$fSg!kqud|GvkRsDN>|8qF}
zx32*YA+!TY8ZSs?G7aeC@GzJw^R8fDZd6Q(WUA7vw9V8&g|9ccRH_mMf0CI+Pawf&
zno_&~cYz=`jnHa)+oVLRvEgoP^_qhvEb6D&A?ON|x_ZomAy6Y;HA{gRAWn+yFTU1}
zucxiEyw@L0Cc;c2SF%%i8cSkB647i<ZrnD4?v^Y$3b*{R<*XrZvt$HkYMcW!j3qSp
zy=Di6gc%bs&j(XA*p3%#!E5YyFIbQ5fb&~$YFXukgXGj2bb~o=S|%yk^8_EXtJVTc
z3dkl`;&os}kql5GMPL~GN02q+wtvCQBDH{F!k<LtRGC*wa-afd+L#6A0X@>!ITn&S
zc9MwCJQcw|ahT_Et1XQyk~+J{)w&&%6VTTqJm?vmsbUU5(U+ViKj3=EHlFiqd9V66
zW|DiSKoVX+d7c~?UxCE}Z7q;%wmpt$Cb^-DOf%O-*MlElrwdQrV=XtL`Hcd1t`oYh
zFAc&?wI-47Z@eD$92RMo6s33Mo_o1PZcap$Yv@B2<z$RYpdhy5(dj0d>NIU`-5bjq
zah%ESFTCupmV0?R(B8#VN!>NH%4ejOB<|ErBIwH_u8hY$c<-qDo{DUrGw4Pf<}uu*
z2n5GI8_6b@ae^yt%tq0G@lu+$fzRC+|CR+tdo<kOc;EK{=caM|mY?XF)Gi@D&-qxT
zKBp8P{7X+>Eyy{w|HzC{aAC%mn8}75c){gkttOn)53<SHjzk*}0F&<|>HEno9s2t8
zF%pA--1PpEmKDJ!CQSBIa&XGERn?Ge$h{4NM>{`s-W@l6hvh|t1d1}}(sf`c87qEa
zZhu|!v&ud;1Ja-d9~7vvUk<mFr<9)_vZ&;ea};wPvOW}b^*Y;~f(IR118kSNHm!-y
z<F`AwN5kLolhjXtD&ty3L1t2;{;8P`lLpsz6&WMkshi3Ht-GKx^C02<C)R&E@!Hmf
zwmq1dan<NKbq=6cLmpT?Iu76M(U^q?6~co&rJBQ2>B`)t<ip@Ww(lnbgcFb2q=-aT
zjjwF1<bp{(m_zRbRB`8KxJscSu`Z>_hGj$KB$yef_Y`S=8aMzRO+7fIdS9mrs-KDu
z#BJR?(v?bn>-aIXjHyL+r>peW{x%DVB)ErZjsx5g^t6G?#M47W_pvI8&lBofK+l6Y
zHC{&bja^WHy#PYjEvyOk>RF^(Qui3?NG-7UU8S)z4I%euIE~}Z<nVqM9dx*5#ys;^
zU-%0^>TuGRTRKuU1ydsd-uANCFIfM-83%1Z<-%rz7%YB0dd6@kAkt9WK>Xu}UAo2a
zpEMX#*-9pcz;kz+u*=^G0V9slUF#`o61;a_21<{9nhZn8iSfimKkx(vZ#A@HH#tka
zmDInW!p`3*YtiemUOK&?`bI#qqvSLY+`3BUkB>(bWqaY7-%9>Kn3n2qs=$cLTZlgs
z5Fz@wlLit?*tirt`kvLYxZq-vOB2YyRiX{8m1cZ23(!=^2Iq~>5C*%ifbT6nLeq(R
z^v&k#Gw6pAI+9=YV%svizBsb-XU5Zm+&AfSVXO#FkYeNw*|5{Cl1)b&I7lylrnn0O
z7<vB&rGWS31nb2y+B3QVyt~b$u)yO?Dq7ZWV)FSgfSPG+npOV23{u*q@cw`W5}Rmi
zfW@)U^_8YgC)GZ~*v8Gp4f?5`6>O>5F}CBlE#xRI%qegL-z3kG%w8d)%dvzNyJ9Y@
zJt5|*_RQ_YmYzjZ2TH2P-@orl5)P!j+e2JQNV2*7TcyX1drP46{zD97=qg3_V^<8!
zN5;(C_Z=yx9v%2vl^r*Mf{`qBuW4&pF(B9g-)1$j3%l!>W^9)8cMZ(cTb*laWFaIq
z$_u*enogp48_v}A-7%Jj#Kyj)hRBz1ZxLYXilsLJcuj4M>X@4c!{r52UEo$8T)a4y
z*DBeM_SIZxyq}Wz{p@)XO?d>t!iF#y&u~{CIS?@zJKWupjEB*UFb&fLO|-&tdFQPt
zE|?Zh7+CS_mRYOvur`e{Nur*2+s@bAHOH@HL~7U6su>z+QEa+FaC1F_x=@6Z`o@%Q
zueIlDy9dDMT#DrMj6lNXKgT8B0|_)p)0U!m<r*IGAj#~ugO@6}8yeHe`<-GPaz}}d
zx_8r^m+xg4KG<~JHW$M*eu|VC-LaOO9Y2XUO_+)9%Q~1g+TGLq<dWMfU0Rc<Gvahv
z6IB!1!HD=>KvKC@ai&0<OaBKOnTU9{`3o4WU1eH>yqoAg4*a*ZKpXe|@5kEz|95}9
z@5G!dkxtxYI_qmZJtQQVO$+iD##Q+qf&Ke=eH$o5-aM@7T>X+%o8EB1C4kGMq!Yv0
zP325#oZXy)@WYkH#3Q`XEni8kJBpB^?H6p*)TQX#|7v@61>vB5dhzB=Nvl4}>&BVp
zA!F#6d|Aee_K`TExm7Hx-(TOHi9j`{qF;y$<_HUaU+s$-Y+#M4bpF*e{s@2gRq`K5
z0xn6<>c7AS*YymAXe+2(1ic`ACynou+%z^i+J%2AVsDLok#Yi2gq$nQ>8hHv&Hsn7
z^Ny#2|NnkA5eHeB2N~JQ-iM5`C5dc_jO@L4#1XP7BT?2#M`p&sLD?BuImjMIws3#W
zb$#c3-@nIw|8+es*LB5pe8&6pe!pJNSB&>tN`W;zcm1(6)@`hYLfLbIX}{8(OD&@r
z4VUOq;|VRWUvLHX-j;>+oY1!)@|RJ=p`|@3e}031QpnQURu{bTq!%u);{5xNPs1xz
zirgojJnNDi2G3+x<86a~LDTX(cJ$4~`@_mIBL{HG#r&YIeI(g0tXz$@zq~)CJT#-r
zLp>{;$tBUv4C4oNMyL{70GG^8d!^=?J4^F<l6}$InY0qtb8#|*VEJb<i+*X(&D}dp
z)UvDRJPkbBde8eQ7PMsCKxq}X=X=@x^vCM4#kOw_C8i{|%-8*Oh>?<9w8g1b9_D>x
z!Eoto^8lC7_S(fkU+bXe{L)t%<=Uu#kJ_bPBlM?1MZ9R*ZH~|tYcB!Y7uD<*2dwO|
zU<#s33N(<kFrT^qD*ylUDfss>`B#@#ln}uc)-4`KfYY(jka)b@MDUJr-XaW8wi>i=
zWahM@n*5}31-u>-xt;;p{W|iMpUl@)72m`Tr%5TotKll!Mi0c8z@z>r)Jm<jO@Q`p
z%KuSk{qGABELPDq>Q-6~Hr)yFR2P3CAUq`sSIKbU!cU^t`WVEB-;vykD@9$7z(uzD
zRy*tE_sFfC<$gmeQVJ#f$k|fHkSIYS@W(*7z&B;2K6UwjY=M7l4^wI@Ghz9QGvG&X
z>BPaeWJw7HDnA$EvYw|woc8MMQI`~QYbqWpN&{i!M+Bpxv7bTHfv?y(^0=K`<a7G{
zhdn|dY-kJow>|<Ns@pZgriPz@Z{CsE3SOaGxJbk?pKs3fXTaR?PS_NTt7ia#YJ*bx
zXPC8aN$@O$F^e=YO|cWLHg4zwxPOF+q|!dUKRlfNfBpw3$nkF*tQ>p+NksS7PtCc$
zf``x+PRy8d%kd>Hf%M_wCJMqxMw)navIVg8W`MT)>Tvaz!7`6-?g~8s_M}{@D;WaI
zB8DlP{cJ^J3lP4%FV40@C9Yc(>_i>}fh))tMBMd=mJTbrktS*rs^2NFu`7I9CP~;u
ziLXw`B4_U+Bk{lWg8s3N{`)U6SHhKmwPOo_vjSHSEwRr{K94tVd6yh+6ygx<iG-7H
zw^I70m8aVTIJ^h{uFdlxM*ya{5OQ%kS0|6(HGoSh_5sRr%b-b|lu$W?G%-r?c7|z9
zH~3ywnc$WCQxwmw;Q#CA`R||ZzyHpI1FRwzzK1OTK1D^HsQrLomUW7?2}q2+L2<S5
z?pvw-*T1)v)ea<6F}R_!dJcpbl8Cs!;{2~p%YSXJ|NK^re7N7Yey4T3*uOr*@Oy+f
zbA>OE)Kg4m1466tGaLqSU?>!zrH7A_0I<y$0O=0?v^49#Xa!aNuS#HN7$d;>BW8d<
zX7t2`H%6HbA0z;02LiFyl&c#E0CL4u5aS!MP|;!P2S72;Vz%`MNQ#+OO|xuQL73fw
znIFyuy|q%5oNqgaqyjq__ageg?9~7GVn=)!HC`-j@G0`&|Eisa6{uAyhi5@9vdN26
z;DD(=ou>+2!TFOKaiKNK+bYu8V13E~v&1y$5>EX-?Ko}+WW|NY-?Eo@v@=C>d*1yy
z0Qkp+e)-?phoEV|UzUI{16UV<f;#4yFbylvFT(|(BQI6~?I^@|r;Vi1NLQxB`<pH?
zdpOWh56*jky3sU$#qAkxGc!xM=7JaRaXB0F_i<>oD&e2QFBdc3iX@K?eRERY9=~o5
z0p5UV@Wo)@$$ek%IP%u}R4cG@fV7&=7C4|h0TIrx;EUbR|MPMcm81qsa6-_x@6$g&
z5U_5JfMH@OWdFToJx-mU-Hpp|046W@Es@aUyIh<<)A)_1QSr3<+q6JhAJ_h3PZ1Py
zu3d>t1X){O&Nkh)q`n-#Xau$<R>w1tc0W&V=3Eo5{r<XBHRNb!J7oF%&$3b-%_q>Q
z`3%^n8u29+2|nUGq(K|D>_#-a-t56Uuj-mg(jR+CH~gWi@{2YDsaWZW%>W5lyOo?y
z8wbY`X=iZQJMUNj{7jC3QRRIC9)RIUNdeewT}n*j8B!sGJsg)ffh+d5Kt{E@-?LBu
zu^RrT1p{Zt<HG;dJ!*67-)kHX_7h~xBO;xL(@2kEmHw#o@ZsW$XK-dUyk67gV&w=N
znRglhU1yKyt%ES2UFZp*!LeYy)ljIPl+J;JxdC{9`r^X4K5$@(FXKG}C8QUiOexI(
zD&+TG6@#m|v!<~TckWF(6jZR`;mCT8IF#c4*9X$az~%J;bx_y5>GH_Iw|M1Tm+8(o
zT%hyj0~SaC)#jc~<D3#9&Z{J|$hdm}V)_QZP?>)HpZ+KR^*JP3<F<4R`5eOapSuSL
ziwxt#xR{FW(RAkaHd5m)UmY$wlmn;dq`legi3p0g<!+p1;10{lV65Wegzf}P0iWyk
z;rMcU{b<m=9^=yOA7|&dGwJ-NQw``Qgh6*ecSFu=!))&4e%%J5?;6KnKUW&3Xz+<w
z_mLAY!3Z_GF@S6OKW_P7g7{-fIUoI!KUFbNKl_mWd1?F~r`J8M9k51?n|W(kH$Iqt
zoCd#&r!!7ioLMQr+=lr+^6?bVuitfV<?{pB!S{9aO%Uoi@jk%?5Ds}=%DO!Y>B1cU
z6Wn+A4T;U9#j#<f8Q@bNz(Fcq0gliYylP(K4D~ahOSsko4)F#M=G9+;=)KYK&Gj?U
zB8XoscD$}TYuhIs$L>4hoe>E5a;%P~z`Dk2#=TWLqI->Zz=t*)f>%}l>)P34#tU_3
zmqp3xXB|6+-M`s^<!B4YJU!=tkZBrNx#sB~K7oR~x;`1q58y4Ldyie{mebUBZ<QOv
zJsGARV}_DFI^@Q{5?;l16K9A&Z8R60@0G3@Cm@=DEQXb>4*b5Wr321%u1$D+6vlzK
zh7q{z7*_YxO0XTjZ9Uo!$e04Kh-q-~gEc#=sw)h1{+$>csm=`av!_9L^n1U`FXtz&
z#Spx}3U;U{dKbK>M;os8(t6YBJsG8v*W;uGx$MU4lcTgeBvmaZP7)`y;~I*))IzH4
zBKCiP!o2egoS#E?ojU`-2XZ=9v49KsO}7=^0DrnGfHS%~H|wzGDFh;Wd6oh|d-?M?
zps7|Img^d>X|OmLadI@!IJf3EM2zcsG0m1|a^NMD57s7~9K{&%1YILunuxC^8UsDu
zkEi{<EkyzEZp`THeFtuf!hECP+*;e(o9@braNoU4^T1@89`L5PpJqq)?y1H@GG{HE
z-ghM>uvVb9D)(c$M7~pB)7QfFDq!0i^Nm}_iR0w}BvL{+BqLg7^vr&B!%N6P5mw&-
zf@$eO_vrkF59Z_P^S<-nN_nsAeh^S_P$0?5gwG}KI4n}A_X&K`PrqEjEP@itoXOiL
zIHUa3aqda%1i~L2{XasQ7Ff3#U-nBiuj<@6FWwrBp6}JjeLE`o!BG+8HR1NxqHozN
z5vd29-40fO$ld}99tq{YZeG6Im&~r(o)>`IujFvE5S-J}hq!Ltl#x#?)QaJUk&h@C
zS?$q=<5<!qIbdS5WSHZu^u^6RS%`3jJZRl)&I6o?$y@zV!M`P{{9l06xYNskI-wfG
z=Tz1Z-HDyH`Em@bt%Hc~pzq8-l_W8yw0@~&Mi&O&Xj(gyeg8Z{T7PncrPu&lr7?z4
zO@K6rT=?^ATugXbaqDe*Q#CC@2o*5B%5cv!$|;@<CErMaYk43oqdp%bSpI=&i8o#J
z=!F=@13vU6e_|K}Ma1KZYpFFVonw=e$Sh>hUUpE{JpW%=uHi)V`h}2UX`agM&kK|~
z?2%NY_V5{wL{K~UB&mqxJ#8U~r#7UfMe8Y@)w)vE&xM?pS-%~z6U%aj5#j(!UWl1U
z!Z*8cRZ4e?uZ>rvX1r1{L|QZ~e6&1UP_35?yMXJhS6^mS+5j%mGu6T+v+2lt_||y4
ztoDTQ^S)nB^IoRoFu$vpgo2n1`+zC2gm{+0GRQ!SH|V_}oOkyjou&Y>2+vLmFFMDH
zGHV-*M%>-}I+$yn`-Zm{P;x|sY;iV=co=-OY+NSW+Uk0yZPJi9>t028&L~wfUN`X4
z_5;GEcL14x1mRwpUho}kQ7+l2j9sp*JE(!4&I~ihUvO~1`a3>eL;MC#-4SpU1ynqM
z-qcU-;aap@-22O>*`g$KFkzeiRZSt9kF<vD44*$u_PM#TQ&crB(69lVQ=?yG7c(C!
zVMu|FmopO0Mo72DC!4VMQfe_jKngfCyHQSbWi^hh`;P3oSv!u5X||dZ<#%WY27kRa
z>V+8vdjstAjW?{58pMM}BX4z#3N4RVy|#hnb%A5lC}S`b@<eLN`DSn;nFML^C_4tj
zv_`lJ<24%KV1fxR1%XM2*C;`vi}mXl#X_-1w>gB-kthjZlZ{PA=22yDx!pCVQu6pk
zkcHO`ki6zv*^ClHA4mN)Pih^~ls5H@pWHAp#!A#DSroDcgX^{StT;L9141+6fQfY<
z7!7BKc$M8cp~iD1Yau7zqmPh|Z_i=(1<llyFVY%JZa-$(3IaprTI;f7(h=+Bzt(YB
zt7B>2ugpY{NtMbX+c@<M&?;9Z8nt<SnPU|kNE7c9s@tC749(xic5tPCEq%^XA4F3<
zOpF}q)+I7AT<wVSck3mMb22iMS_yOuKff7mfgk)Cmvk2l3V+}C70B$Qgh2nfd<nva
z+hK~W;QjPLJf?y=mUdz^%LFwW-<&AZESnQjH+K!SoIPgn>nV8Myi3x4xA66;Yjqp3
zDdNf5M?ukt+{UDF)Wd(jEdKi06Q2=A)fzQz2X6~Rwdx0Lc@_xR6@EGb{&R~sk|LOV
z_f_%vM>yWT*$LOeKUo0L5XfcdDn(BPB>6Hlnhy3y;9xHF0@gIX8VKVSh?>?%-*|%o
zWtlHDG7J`Xv*IvXx_FL;DkK_yaLbiR@G9&9AF9s`8r2R%(?n}`G1#aLNh<-H|H?k_
z>^@pZ!Hcm71q6F?=)#k0s18GD)N@#q(x+i)>P$3*QG#9E)I4sv<i6~hZWKE5^ev27
z$~*bf4A)$7_XQal0zyUZGBA1u+4hZwT-0jYaah8AdLzW~NW?hl#NDe_Y4qYt4C33$
zAT+knVu%pI@Z#Ju>nsCo5$+ouS2oA7$$r#Iav4gE+@1fukSyNHyxa0cxgO_n6^y#{
zfYhJ@(ypR%tx217=sg5*&-5%`oj(fY9>8F$;5{WUiq&f*{t(O)C^s60eirR_=|)uA
z+XtkpV9)4xzuo8hO6dRlGuIbbwIFK!LAdiKxU-|-iTB3%w{f(CP6oiY-q|nqgwoK#
zngmffX&6C)^B)*m%*$EcV2ro+y2BK~#VNG-3rth?8i{urB2UaV<`wU;>Ub`wd!s*a
z3!<2{Os@tAX`IDzX$<TcYS_2mTZc>pDyXszK3Y-2&_ui%gNnUG6^zX3tiMZkmbPy%
zwboC<p$RTSM~!)w!CfvKHDOWy+mHXY<xt4wGNUO`teP3AWBuz#&m*gNaU+m@(?xap
z&QEMb3dQ*b$xoJ__-fYqWeVbmte9Sbs?pC{0-M7iy~B{y|NVZ85_bRlI{EOJ4zr|Y
z!`yO7kC=(#k)+4<1W0Wjp67{r3oC7MV>BF}Tr9xp)cn=h)%qI=RjU$v>3&v1QTpXB
zyup`uRT{MwQSaaMlAhEH7Oyqlo1scV|HkFEt)@k@RU8cV$!PSJ-kEz@HQyiGE&1un
z-EXmapzVXJ{dP#>e9vG!JGJavRo=40%S)_KE8l?8EoOXM%J89wN-L)}p*6VD^*gB*
zxz(S@U~=Y0!OW_Vi7iHaYW#@=Nx(W-aat~zSWc1IQ1eFC^ovc>#waV`Pw;o9Xh7{+
zVLqXTay*jE8)7gY0hECTG-~BUQ+;RVt@qU&kuz1(tK~iwey-PvXx0l#UOR02z{2R3
z{#FD2)uXqH;<^-{+_Q!21Y6N4P<s8$lJ1H#P$x{=q)gxAv}n1kNi9YOYqDL9`uO3F
z07@X&C#b_M=%|DJK5XRdaL!UWn^U=oc3Vz%h*GAVMAHg>4VC)x%cyK6c%lSQ8FHmg
zW234Ex&f-Vf>?WZlppx&y4(+7;+$2BX7Uhzl^gv8?ndeutj_O;zz!IO*MnMMdP`PF
z$}LXH<E?VPk{%)9Im4BHh`JJlH@S{;>_hV_V&qBqW7?pOkb(OpYIm_##U5lbQ0}h~
zCUKs8Y}zxTHoUAbMu|8e20;1S3}4~o4TUVD>qHzDi6$AS^s!HSzCcSNJZG%-(~NY>
zxba_J!rHxckJBiJ-5<IlMv1-*yS`q?+Y75aq3)i!%Nosig(>v}@4%h)wxn3Q%CBtj
zluaXL4o#i|Gu~w1&vxyhp`Vi>NV|1!m(VAX9Mq~wbRya1>l|-vwL3*0(i`s6iL1=O
zT};e|jHv8CIk<y1MqQ<Q#MbQ=+j=tg2$W7xEfFLSdQ?~!{QAo4v)%n!OA$Fq$Br1Y
z!;Gb(o=<5(I|_?;n$8`Wx`MfS_2gIh%YLt0UIaMq8ZMRJ8YX5)kpVh4bxu{t#-K68
z6c|)-_s4Mbd8q|1N5x%S?wl<xs2FKozdJ-N8i;$5bX~<46b#cBvRH2H_kFIrb;5P<
z8Y&8-W~|tc)C!@0`w*;-@GXedw3}M%RyCsZTnS&snGuY39M^d9zs{@3cN{!aKT238
zJk5GFA1C)I$>(O>@Y9wLp<@Cz2SFmZekViihUq_FKUSBhTX(3+h=%DFKQx3f&`c(I
zp}G7xQa1zA|A;R>euihrXhCG_7DT~`BQ`(@NHt^qrZ1CFYq#8_HvfgC;N-9%#L}BE
zWbSnl%!8S%!C~g?Q&?df*JiXG_Y}oilmaTb;b-%bwcXZL#!QvhsGR&mXFfAyK7*>Q
znYWS5aiM;y9dayorRB<|u$OPX+HewaGoZ^5*Il4yH}BtmL!dV)YieJZxTp_=8%-&o
zyeNj9w4d^@wR7};D;xY~^~vn7Ho=cI?@W@(><^H-gZYv%f2B2&ltV)eJ?VMEM`!5g
z-3=U5eDB@WkcyRo+S?KpMT(gDmt+@b^NO&Emo9GPlU?Q)cI1!t3uC)4%OiUvd+(@I
zP0<wmVmZ0o+<TZ3%8rBFU0=t{NPH+oH2bUMm@vpw>mzkcc8ypatA*LNB*2<5I`r+;
z6FJQJ)e%Zi-)?51t#)__{WHw|=EDM9wGtzpbDod0Klwk=g9f1X!z5p&PumB#nxFL3
z;^nM`?B<J?De@g_x^#Ff6_p822L5#q`w<#?A(A_iVp^ZvZe8nsB4pZ^0Ucxp?Zu5N
zB7qHqMBO?Ji;?V-6kPb(-21e6ll}L7tJjvLh-J3!$=OPI3Pk3H-kq9IwGu`McDeoS
zy8o+j1y&TKJ&TqmKS$QPd<yU$oIIh$D!j9{xqbmp6?h4Wc+Ki?9P)2@p8YZv>-M}D
zra>~tz8=Yr6NQVfN7#it_-L-S+UebZw2F`!+?z8>dZIyhTooOq8yBXFjB$CitcEP~
zhRJ&Z{X{11wv@|ZJG(p=TVbwq)tX(-%+46x6Kc2O8?`bYl_|LxBE=9@ynypv_~Mvt
zK|#3g63lK?35qsW6rnxVMF;C@@>FifU`dzlm>=^J%5~Cjkp1LTyUJ^ZFAKdPgZ-m_
ze5LoQV5g_$(oK7*w|WIzsGXn4A5H=qOk+gD{vhyOda`9Wp5)V55#n1KV6&h8lKQfW
z>abl8nCzJi+^*Q=R#WC>tL<fDLOzvQ)5!I+GeNrCpLDsA>Yh9ECj_+8^89!`G1U6&
z_LFir*~zOC$W*UuBaR+qUs!3rjzJqMVNm;q8r?6Lmj`nI@w95KP`Yxb&X2+qA(Tk#
zW}}OKF)x%+lypKGdza_WVeBd2RQG%5XqO0Q|20%!1+Dn&A{O`~Xc(226D2^1phTh!
z4*`})^-B%V49H@B<*Z=HXnEfCzj6tg0o!rOPA(qmh-HX+@XIde3NM>f?Ki=@b)m?S
z8ZBnY+=eekAp!Msx7g2_(R2;k^kPSMT{NJBGSD#D$?~yyhb3fS{bIS$aE#JQBbnG7
zZt|^5A0mzAdB2ZuYVKH)&Vslh&~W#Zrhe~H`4;awqIZ(4XFOg*xvqm!ZZE3q^&;1&
zD}Cw7`K=#gaZ`_boX<cJzTzkPm1V&dTg_r|%|o|9o!)ts(J=0pE1>)fv2G7t6tY$_
zr&7f?#ggyU;U;~-k+<Sv{dg@V^8BKFJ~|BeARs8RtT{3=i>c;@chNiBsJ@mToT$KB
z+i5{LsI=+V1zdE}C-2eb0R`;zhF#5@KoBN74SF-VewDtkOYR8G+b)y{KjqVKP&EeI
z-)jm4eU_&edh}<GIYP6)KYc5z=pr^`SjjLC-29{~e@|`~N*Ky|*cYMG?U!|sDsBn{
zRsuzBlz2&qXRYzRPlI2b<FCHGP%gkUZwI?AolfvR{+jv5_B!@_3tcA&xhGhCjIJap
zIdkVk%|^`d)|#;ZGW`wJY1PL)%*!}p<&P&T;}Eo5u*HS&xF+vFv<iKM?BuHo@Az{$
zm#3BR#*i>}B#m9<?&~?NxlK?W*U=7)u-bD<Sf=r2MvsI}<gb|cluya5^Bu!)U)PoA
zdhcv-Xm4Gdlsaez=F3IBl;T~V6)E^V=(;-m?|_?-si0Sw&&iU&Y?1y^w1Q7T@B!1y
zgQL?w{RRU@y9Te`@{Ggz*KP)oO>xV5;XqHvkEavQKV1H+xdgiB59>OR<Y#vk>{wpR
zT+3<|o~#PHGPUmFRVk=DzB>(IR}W>)<4kpZYRj}?NxM%TDtXvsy-YUW2aBNyZn$e;
zie_~ke9Toh3NRlhmq0w)gH^MYM$iFk*<)i+W$ps6uMhi)Ez}J;hna_OLPz-*SD3v|
z)Jsog_E_h@9~X2I;F4e;X#S~rM-kZ+Nhzy2^>jmQh3}Us-_swSUNMlqn2Ch5x-<9B
zeX1<fPv&0ky&hOe9qvw2<vre@>k~cCJ-Lxsm-9Q<`&O=*by8Z$tBw9QN8*+mb6#K0
z4m=TO9`?hACl1!dKt0G<ke95BYf8%YefO~cyD2Fe1KEiZpkPFlmoDepYIaH|&Ko1;
zSB}m><yvUh4&Td2C5=;|Auubw<Q{`}Cr>3tIvdW1aXW;Gx0A-XpW_;jEIQ*~_-U~_
zE21qo<J+Pr?Q|wj2E_bOaTGOgEP*xDT07gyZl~+G9Ch>Px=!F_{@(y{<*X3X#=Of-
z@PzcTw~D;6ES7VbIE6H`2$V`1K4P;$05$Q#ZGYca`a$Q!L;5w%Ao+qj#61Q|y~Cgs
zy3N+f{Q8=6FXnQU^<a$ri#Gw_wdp+5Ypf9*)p4B`pqzqH6J%rgQ5pe9!C-Xt9*=Gp
zQmo9wfPt%8QuET<Fg_Fg0dv;SMHN-N4c9{FxHf2Apj6+b(!G`qXDIEwI8al@-WcbR
zIu8-)=5cvog#SobFtEpR=A+K<`2~s>E9VdWP@<1+6SCYAw40*O8`Vg*MXBkEu@&~A
z2^Kv_S;cxdVp|A4-kZ^5;Rmpq9t|hvR|4!m9ouj<k0ieK9a#Gg9_CuH552>8BGPXV
zL5|*{45m8q{zPEq+o13+{&<vU5dwluYwl3kT7Lt~u!?2DmhdJL>lZDOsTGKhFx{?U
z6JhdNy+l>0)UaMuP=rGGA=BND=*Z3>LH+laF~_D1q$)esDQwumZC-_9>y+E{L(*+_
zzrJRv76L;&;F<H0g>{Aq7IQT(zSvWNhkf9&$QRIhA>Vg2rzZ@;`+BdD5Hj!n(h3qx
zf>t70kIdx9%U@jL<aJjrw2{~Kk0f_&TzX~JYcXCUjbx)&RvEU5$k~2LyxWE^$@7B0
z5usc_Sg2%o?F73Q$xkw?0Az?>oCS&s!E`+P<K~_^n`WxF2%lm<&v3>@upf|OFV7zE
zlcwx-tTL#u-Ojg<jX*jWhAe(y*Y;B-gt|>ky#n&3ZcbIK<hv;C>MqLm<`AMUM=umb
z!aY=Y_!v%-y`d(oOU#Kyb{gawXI8_x0WrB8%!1C*&-&~n6LJ<BvN=r@LuJ<Z8q)U+
z3TjfVOKro&30XHh;2~uN-b+~z2Rs)$Y+rhC&rKvj^;C5WLlgR2^EFN*`iq-tNL~rW
z6<Jt<Gb5YuhcG=sZ@)zO`?gHw+!bMGvC@K&hL>ae>Rq9Nb|8T+A$?HQEV6}5VMRW~
z5**%6wpE<?ziuQ02M|FY%P|@V7pAINj-TAHsC{4Si555r*b@w>5F9iqqB5bYwtoE)
zI;&dgP$R#OcLqY)k-5m0erm!7?*UYc0Z>df=a+B)@ikl{alhr5kTuYzsF~A)f0;ff
z7sS;4&<J0-9hlVgMqn^-&!KudfmgF&1G&JFMnXQH!j6A-=Y*@a+x=bY0MCOM?DC)q
z&<*4+e<%uZjXviMNRZWIEvZCg#-D!$>0)n#*?)8);KNG4u_bHT?8&U%m+%X#1^+z+
zY$a7!zTZeUY|oEc8RcWJ2@{C2y59g0?@>19<C`HJfzGLMPqET*Skt<gD^-0zjYrHr
zNmP5iBxiDtuA-31(0VqLbU#s2b*qyh+xb0X_fQ|Q$VOfm)yk$ZG~y=qdVQ?P*@Tav
z50+>^G?j)>xUM0_!(bylDD-BX({*jjoS6Hey0ZrQu<@9dz}3vqF6^B<`clJYz$A{h
zrC*Sa=QX)GCXv-TxLJBu<6V@@iz{*Dp%rw_oARZqQd{edGuIx-<KO4`BAmpsU$oKD
zJUWEK4?;SEO*}U3Ehz$A3ci8Ur1&4F$%h~3QJbdr_McoOUTnnGIB6Pst_X0Zr%y#|
zYp;#%)#}y0x8V~&{Vs65{$e(r?DEX6IYp!DFtr6_H@`*i#7H$&(plO&)G4F`S21@R
zTsJvaJjwH*PYydbT+h5)`H=dt^qb8hcI5(Pba{D46kK0lYDF%6i7U2+YgOF3>Vd@f
zGgJ!V+w4Ee5UQP8+-u#I886FgcoQ=zp05P8xQgI=&&3BTA6kxX9daTFW|=q8j3_y)
zOk!Wz#XB(;=Pz}HROxjBF12MGZv#woi^vw%bqdHxKG<@T=5p!1LDjpq*F_LT%1~4@
z49)dJID$NMnsz{_6X2RGNLSy%JFl0xlfKce)39~ksUHZg<e00ifqfpoHOP=Q1F22s
zL&1loqrMuGhO4=7<8tHn6m~i6-oxYz9P3FZD<H!m*WaR`C?qV7@CQS>*28Zkw$!HF
zrg4YtotMA%M!*j*LoeuH>Nohz3aEya>xb*;c9o%OeTeb>YiATEGzcygs9pYjQ#d`8
zj~>COzNF0~eRl!k0Y~l!>5)oiTWm-aYoQII{fhYQBJTCYoL4~RFK1b~1?M!qRD`~B
z%uh*?=3qxWdk67I!SrtDLQT#f9uJ_!-LN>0Xg}D!NwrE%kt5QcO64sa+p%!uGw1ct
ziSH|+#|=VutU?*CEm=iLFAsqJ#C-kr291GLvU2`A%b(%7J(TPZLswGCCv7x+j23$}
zUKJ(vwzo!zvDp$X5-KeGR&E%=!{DJq5qp`qbEfE)wSqZHm{X0KH4uj^EMFKLx7fD1
zpiRz*hWmT7Skg<J4D}U5K9EBN=>Xv}b&bI^&O|>nj`qWloIP>@;TKpRLrGH@2%R<D
zmkUbXRns8DzO#C2vc`L&shM12u*{bXbgsDLMzexl<_9=#_=Hx3g<pp3F?r~iiP+7*
zN(?kT)WE*G&&(xB0iAuClaxE0f7){83=hgehl#(p|HM>H=0x2ddl2wqS95iIr+M^b
zs8)#Hf#hL6Y;Rc7C@$1Vm0sg&#Q{sO$|Tw*6+f_2ka6@lo3T`dpK5sBUY}m}>D(9q
zHX83vUhRV<vw$k4Dg#t0qXQMB*eXc!T(lp-sANLRm8IZeE#B3mC4dw4vZpaRaZt-v
zb-2&d@0ut}XeDg+Zv|&Kb>by;`AGx<Ucq;58J|E_-Gird9Y4m!6l-b@GyyKjOWgQ2
zqZR9ds@!SmC9;ei52itz#{h+E^R&Oa>o3aTOq03DKt<@}Y_nrZm$}mR<UT|8OFlS(
zyLF|y)Q&3+qdcrU<$Lp|D%G?$c=&vVinX!T%S$%=w4#5k%>}uLjGGHQ6hJd)j!k$t
z1N5E&T_&^kO_E9UJ#kH^vv+q)M7&3-dc_JQ$Yj>vU4#f{wTg<&>OIfQT>o<OUAJsI
zy-TjMz}qNq^pHFHO;sH2f>J<Syd5^PEY%l5W5&x_uiF@#qJi{}OBG!_)C7Exb`MgH
zD<ZdCI{qxLEk1WZi`bi~zh=HF$Z-*hp)Gi4&e2w^p#F#_r_(ceAYeVSd^<?3tf9cq
zFfJ_eW%8(`powBo{RB?dLxGp9ED~%zt2NYfo>QhVhZOB9y10`=Pkb^B-+2)CfZpz@
z@pYs4(m#dHSR$8WF!HKYyM-Myyz@C?usy2LFMmwy(~(>;8HPr!O`MZMO{Hsb&y=yR
zYr&{%I5C0Aw&v(%$YawmLIQ&-XL;=lHV^jgo0AXdIar2E-wbfD^~K1yN4@cTIMPX=
z>PpPumvrvNf{wN1hFTynQ>r1SEuyK#dwKp+<{Le}tr8Z_otXR1q}sqvLr)D0ie8+k
zf*UHqnI>=O_qb{2#vUK6(&=dt5fa2zqPuc0H~;7YhFED%0?gzcEx7v`#~eDsFUaUv
zhzmbv0*n>Sy|2PJ59%L0)?&BUW{uf<$Y8kUUl}x7Qk4^k*#9j2kY2BnNNsJxBJR9R
zS6~)?&)2uSbV7y0V3*{WnRP`>-GW2ez|w<Q^J#6?idvV!dEhzoiZXftUQw-=L5&A!
zj$0=~gkZB>ssp?3N6b=hY({vREhRJY8Qu@lJzK_iw6&?|>&0gXc#+#Zdbg@^Ls0wQ
z!R<^N(YJ0yZ&HX->^*ONH6E-hQg9H{U?DD%Z}$=_!0&080*I?4Go3wS7LAd)GTi2=
z_9K#cF$Hn<v%<}%ee+Njrw(PZAJ?NMBa85XydvqjO|AAn?&(B(LeVB=R0lJK`VMFb
z9n%A>#0yHR_GJ!f(I;hGKds{&N)**i?thMgN>RhztO%dK7%aw1)`pK#DjxlWrV?3a
z|M9QNbG)$d+pxDG;*wH7X`s76eG9XgKBf}uvOThS&F{L@!CjkY)UU1X#fguAWwaSP
z=RQsNfQkL=Yy#sHm|eps$X^{PrOFDLKZzov9es@x_})`*E0#sw+%=cP&Rg0oDGeMc
zO@@>iv;+k;vly*w{1NK4la|L1P!0LMc!h0fCv!-T)5_^Au4eB`U|A8QMRW~e^U>*3
zQ8Gh^r~Le@k4Y!nivS0yk9ZNMtE^0FQ)pKlr5h#0!<5;2d)xX$vTAt?r($AxUz|o3
zm}wh&e?BaaG{_-yLWgVXi?;pXKhi2kXszmQcOB-`T6dJ4F6@>5(d{9b+vGO!R#ToG
zfSRCEX?Dt0E|MgxZ*VmF<%Ux*#@-8FUiiJK@mxcBkKe_Qwu<e4*m2gF%XvY7*nG<O
z>u9J>2+(KCSjKgL4~b#2GrlM}IzwTyL5GYK8+?%?`Lf5WOA1>o6YS76g1GM1O_bt?
zY@-D8EnR697uuR&?vNxc75Xp6{)-*7O0=hl{mKC6`0nkA+to{hzWn49T00L^E$<_z
z1+N#V^q1}aGNI=^f{V~IhI3{5P9=d|rp9dd;;@lS@De`P$sa3v5_&UWEf-|fYw8+!
z`v>?N_vkp>&B5gE({YS`m`jpJLZ96AP4Ddk*-Fi_qGEzTh6iC<k6IHbk?PN?$aNNP
zn6O?MA>zm1PD;Gn?0egeMg)}MCHaS3?B0$Q4J7_{>Bje0pJ=o_CMz>RN>)6-Q)NqH
z{UNeje{Uj}RPcVxrKBJeqCIj8n<0ODptD;_X$-m5Ca$4qLy<EIj;{~w$8o>EZ*Jdv
zNi9jybSWdHUz0#mdlBKB;+)iCP<OnsXa2l}YYS|;A|dc#V{t?MHnKi%%kxOs6pOsa
zW*vY-EVYXNQC$A_4x+_dbu-zYd*ZN@IOVePDfCtO%|@LJ1u0paW^jd1qc8ez!B#7y
z#wCV5XQGZ*0?rlG*W$#itoEWutM`WQKGnE75Mmvc*5%{i{G!!|e!g5s_LKAPUL+~w
z6qGzV#+<^6(GQy)gC{}(=Y4Q?HK3GJk+=d<kNNKMAva_@aPGYtdE%2C7o;g*prRru
z#bP9g1sFqofx_xh@U@Tx&-(@9J=f0XgQK@`#w?fdK4dnO5;;vix6ERL0j)cQ#ele*
z+cX?6M6L0d;4VB5w5|~o{{}PUk&3kV3_MLn(QN%BtHYm>R7iA^RZ=P8POE6Y9r4ob
z6Vc%eBF!L13!F^!o5>M{`rgNW+4qb<7Uyb0JG+o8KJ~2Qiubb^DPf+S4&_;<tF}!%
z3eIy(Boy^P752H3@G(Bmp6;b{XSc7#bOI=Tu0@(ZefW`l`SZ!!-&2Tz`Mv&?q0%4U
zwt+r6T3zkkqq*}3*HaafM!5vOiy>fPzwK_UaZO#~Fj;}{yoL)6U!{7Y#ZLUF_tNC$
z^-p@t1J##l>ID+GK85V$jmkdPMNM7Gmzx~^1AH}X;we%6QG`*;RC|Vu9^Z$i#qkW{
z(#zARmg3mRyuCAHRb(6@sAMM9ki@FOGF3k?<jp;H_>0w_D|+5kwGtZ2H84|jkO^q+
zqRBranoUew(HbYm@11-<n>Z+c9g0ao&X1khAKj`SyT!gDTdiK1d2dnPwJL?li``x)
z=Y?*fw}o)M(@WorBq>EzHmjWS2O&@T%4kP}ku^!+bm3}Y0k`g89=tkBzPOocpLU6$
z=NZs*d{@L9h_%+=#a#+$jktur8D>AJ8GF=PdANTAPlnm4y>=6f_`|8~l-7^a1$(nm
z(dvoVmKhL9Ok*LeP&)7s85Y+<oN(1Ka(%c}p{u%f+}*67B);Gg73183aUzju;zuSC
z)z{s>Nj9Gtzozm8ZZCdnxEQM~<A#*rSFSnj`@<$BtmWi49s2JDLM*f=emnz|<dII|
zVh11ueQuk@yCkivH_@(3cUXmSSeTm?;6Of?W&x%yC#v=wD<##-)H}JJ{X;d+vYJqi
z#JpWni3-bVE8c~ZppQRlu=^2k<ADWO*T<i-zF1pMVO*`!<GVo@(dcdZ7uN41ciPy4
zy($e%WQ9g`%9iVQ@n&@hmLl_z&R?m!wF;PC3Y^plD-*xD*d|N&9{AM|JA+qfEKSd!
zz|48eMFTiXH2fDU*>$SFrchM_dSjW(gUX<^9u;tdYN_)lFnR912>9mq{Wt$fk~FwC
zl8_TqHnRTjuQz{fw7Tm+{gDcs0*X2;|JP$+J=fx=M4eFqhjwVy*~fc%Od(VK8cO({
zDYsV#AN*I<8Rm5<&r3Pa&%_f(4;pGQsX|`UKy|lm_j01=4mEF#CU0Ia@IUb`GVm71
zAX)m<>QjXt<bFVTiKqNT>2KQYKW`C9ZPiWW)<p`r!`esu{7rd0W`J%O6O~qS*g3dh
zXPix|n%Q-h<lEFiRtD0=hW_eK%a2u)w#C<s#N+kE&wnMWLHFGMe9{Cb+h-C3be&g@
z`RHW0V(y2DQ9o4Xrq8sy1~1{wVBSBxtPZ^IB|Gj`G$p9a#J~CCoqUTF4;{b!VN3!_
zB@MkIE%p*j{Wys^c!@(guQ%1;ELR~NPjeHfQSsr^DO6BBS+OoT=#n%OL>5|Ki^=E-
z<VD?WX79qgFHZe4YH!8vS+V_UUy28W;B)HtKdx(|W!FRo;(@s`GEwvzuh|o6D3`3*
z&l|Hg)n2f{;aeP3Xd-(~ZUtU5-`eIrsuuoyraX!O30FIqe3zp?xnnz-if-SdCO&DX
z&Z<<qCcF<M<${K|@j~I(Beo+PBFbGBWC-UnhU7ucZi&F!-C>g6?<*Ue7}HyLiT#_2
zk@D=X9Hu&gkA{jvj_lB5T)M>BH%`q}^Wo%c%8YO=Xn=HYnWc)nPI*2&p7kt=aXJsO
zo(s8`_rb#p)}=pL(t<{%rfLc3T=v#S-u51)Ub`h5vzKbon-<m-KYlBns=t$QmOfUI
z`vQpRDrN7;i<|zkv(X;W3291FmW(!kzk0Q!g_b5WHD0<PNa5|+PzX9o8{(vJ&;cMk
zOLDyHTFjb{{=BZZ<hDp24CRLEh`4+MHn<=3k=og75?-uJ(XKQLR_j^!f^VzUvSY9V
zLp};X_^ZqFR3c=r0-kZ}g%zmONrvy%(*)(L+usO!B~RWa)HEuq=h6|R)=_a9ENhso
z^BW3qRy=~nmG7u(%@Chn(<Wcd=C;#`+e&JAq1{w2Ol#932ZDm9ar)I)8!R5*>8+EM
zR;JzrYKgdTY=X9!B6YAxmeu$<tLz>6h$ml!&efi`^p%}zF$Bw<DDK@{wH*!_BqkGk
zxGx;c@`Lk)*qsr~a%-&@`LDB!@f=M3{m8gvtKOBIxZFrq>z3Hr$@LSCtjam`l>ZIf
z&Q1jhlBdDjftHSG;x*-G-q6>zGhaX$`gh8<Odd9ZIt4zZ6>`xrYS{CnyJx|sb{IRA
zB3ivbCwA<3pp5u^D`#HvSlO1CbBF1>p+i$j?HSc{Xig&D1Uyz-y&qQ2B&g&^8a^z~
zFxDC_{PTh5=5bQ8x?CaEB?YaYUmP-p-jsY@#m~)nac3%}MIzW5LrwDrIaE82(7)Y%
z7sbV`z$(!Ds@G+^OR1Wyhm!JS1_5fK*e`AIEmW%+6?xxz+Ji++ZC^-C;~ZWKm%94T
zF7xlJYU)r<7P^f%vhZ-@@g`V)f2C^qX%6r7s?SyEBS%Zb$f4mHpF7u6nCtgH-Y55^
z4s+1PNwRko+J8m6ybMtLlB+0#6!0RqfAPpDPQ*#|UEkf2fP7Fqt*-rIleQXHONwF{
z$I?L^%KBBi?-O+q;B?O&x)P=M{&1l_)=kYv%9gfuT*25Qg@2usxY0z(mA;f@S<{sz
zHcuW5M0u*pV)pkyuP}R(B9&~3pcO-rUf3gf*935S_O?v!Jv0jL#?g0^kPqn1<lM90
zxCA)UCW&D;1%eqi(xX%as0tqMvH!j&HT^8|%e}y_;r4&Rc}{L56|HWM%HC08rKtHX
zMq)^hO8DFejNftxWPchhLK-#;uolnz_vSt=Oeh1nqis>HP{u#>j<%QZh@Uei0F3&5
zRzMPKjk_FvC6a$p32otpv}gVajkDs2Wg)KDyB|@<lfl%gtZn1L;sM1TEB}F`<lQN4
zH*hJz*Y=cGY#@YL5%FYtil{hPfS91_#&30VJwk3>3a39R+J#(+DKC_G*2TKgh$x;>
z*kvwmU)7{d#Y3$UY>QEcr!?lYwx;!mnzoXXiY)`CnLZ7RW9oNnN<{GoMdJn8Vg`yL
zca8x@%1luQOu-!Edp;6kCKl*hzhN>pWlcU5I|;=v_$$#Cu}R(I6yi8RG*J`-aajHi
zP0O}DOy0I0j6?NCFeIPQ__QJzA?3MJ!9M)d6F70(KIr8e4-<)Z+L58|4Trghof7UU
z*DHJM;6(~>vtS}P;(}f-%P`5)Hv_z<qNY9LB@CHh=o~vg0gq&R(`jIR5+n`SHt4Un
z!6H=H!-NPQVE0}lwy%7`lo>2Vy4}R%HoN>@7wZ_m4v8_03s0+~vSvBIhErWRuy%&U
z4AH<i_Q_z722JFsVd0g}%2&J4+uj)is!pt%vD*6T@Bpav&3WR<w@!xNmu){MkCf_(
zNWg8JRhK)%Jt9ca#n+;qky<Do$z~T;!STGP8?F$1pju(W(iiP3H~HUq^JDI}1~&J-
z?A%4AY~g99{0(w&X)`WATR#^S;$PO;g$j6m5}^&rKwef@?%1rGt4#GjmlD68?-~PH
z^+2azsHbu@#Doxfs7{g8v&pE$!cLTxRBT>{c7ync@@<dxkdqgfaIMp%JtHlK)yIZ(
z7(sirIX<4=)1WJN6tj^)k+`!x^;gFuoH4uxu*utqkpyw>2xEH+ja9d<WT@jaV4MHZ
z0&`7$L~&<s)hk6&z;+rqxx4~uy$fn$>OyOU&$pu#d0nn=iiB$KiXkJoRlflJx*nT2
zmc6^}v^FDAAwF^r(*NLAZynPqhlsV+pKg4ZGXWg$>?pQICKx*0$*KHfO?+2{kzrYf
z)XbU%)24rFkD=3NSbdgER3W(5sv99xSOASC<ukjt7C}r;d2g~J=m(Rfs<;is@U}em
zn+0#S5dgoKw9zy^NE!qc)u=&NUZRTWEfiylM~SO`a*dw;0H5zhW?EqBU~&F3TBFp<
zyCK=4jjgXxU{bhwWT9N7+R_53=F<l#NtlY6BRI*r4thQlHr~-D{A$2fzpg<m=()({
zd-h})SDBW;5B#{&?b2X`Ux%dD#2Q)I-3qtkZhZ<_e3GC*4R6WVa5By@*&|%k`6cpa
z`YXpPlB^X#Mj_~9Z2H$pAj_wwTyVi7JM!gew^&3N#jpRI7VyE8(VlD1%}IqDnu(_b
zpPR7GAdorE&udI)P7>No=Q79~X3Etm%uA}s$@J@v>`(8VM^TIhByF;OR?lv4b*uDL
zq7WadFzb&Rm##YTne<d-%}a964l=iycIwl6ByY3xyDFWn^Qdwn@(rBaI}j$u!_Stt
z%N=PygP*=@n{##6mfWh6mXEhUviG&{4vQlP+hT;vxgAX-Z<caQ5DQCZYuOaJcW<m7
zgX7P};_$<0_}GsVqTC>h-rn03V-441y#@s8miFdW4{OFxR03MjeN6)nlQn|dXXs=K
znyFqI(;Wm$FWnMN%;(CC5)(~<3IunuJBxLY&bmgw?43!tF0{z$?CNkzK39R20cb!y
z`C=E*q~AR>m)7|@R)4l#c7ESHH>C1(xgE(9lq8+ACcqxb<bw{)`g-c-E0E#sv9vU_
zT(k8cgT8=g_Ansuv)32ByS+8zpd2<%44l0A(~{1wrOe#8b8^$$E+K47-DF7QwVsNC
z!}~pHH)CwT@tX16zUm=hS?b+ZpH!sEsuUn7s8Ru~s(6c2zr_+qMR`h}zkE3yYRdHX
zbeL#`rh9?;h<!&Fl)|+3q)%E)w%{{?D8dQjDT7yn&eVvBQixuyeO+UN4`|@hA<zgv
z5G2_c(6RyTvcaT0oH|CFCT2bPJsES4o<Ci|5EniZt^FWlc7fAHa4NfZ(F-c4OyRGq
z8PVW^9nn*oc|1VIPAJl~03)Oj$;dy&2J*_?UhVTpacf?Gw0*<Q8h&urqa-eDK`bz0
zLT8e?&c7e&aj8^1x1)BD3bI@BtH$*!OG|3N?E07|zjs<Kr^%=ol;ADTqd7NDlQQqC
zjbVa_kxlFMF3Xj<2A4}l>;6uPh^t|0oP--TKr*rY<SCq}w%kn^M1TCyBj>L_0;V|A
zQ|+xBXE9;Bej8wA`(%*nv*s~5)+*DB+c^>TPKqAiaa~pk_D>Om3*SavG>nkM{eSJC
zlH}4gIZmCwdVF6bs-+F}1Kk}TFBJ-^3rBH3BCYwdnWDm<j8yAMwx}cS54X>a`S(U5
zH6tip`}%fbGLVZF@4ojx<Fo#HlUQ*;-f`kns$m4FJ7U6|FUp8NyDxNIfhJR$6>fm9
z{L^jrFBcNUCH!U@JgBwmvoID(b7;Dq*R*>#zlw&u#P`SHom?0lJ%*ZSDj4cyCsxT^
zuyzm@?dMMBc14CE*+1Hu+$3&!1sN#FNt0|4y>az*@c}Yth)R^9>hsX|LWrJic8n}1
zdxG&Si189&dZtO}p#lI)s0A+}e#0gY$E=(FjIp@*8%T0|#qC&ZFsVKm#TT?F@y3M(
zS;|Y$W85nsRMB@x!;X%D-Lz9{9l&W$@Wj;R1?Vvf1`I0qufuo4tcc0HbfaZ-2%}h_
za?h?}Cd&+wXt{(CTs9f3Dd0P7D2g=mp@u0uq4H`~Zn~ptXq@l*L(nf&(EFxgj|ptT
zlcC)GU8;1p+ACE?fLSt&7ZZF{M&6-9tRigx^gB$x?SAj|ReGxOhX?=BgOXscU&GDp
zvy)|+Huv-WVg&fePZ)Jzm#N{~B?m9Ei)BK1jpwK>ozBoh^nsilvTg}NNIn#odoK7B
zVGM{<HK?2r9R}|i`b>o{S=s>n4m_=xO5en{JYRTn0(6R$_|&4t4w9Yyv&>mQrTEP5
zRZhu|ziMcpRb))Zed!+kQ;)azA6OGT(h)SfcD>u4`rq@{fBm9G4XIn11yg6|+4D6I
zKywKBbXw%Mhjt5AnK;}9wBPu5rmqIr{$eF{0~Lu|4RGBY67k4)s^*7V5u{ML{dWJ!
zG)alf1gIMcOKh?1wsMHAI5v(eNnZqQ&9WG&WlkCd_o{4IE8jups|wog(hqnR0B_yY
zPJ}X_8M3_ec2gT#^6_(HMzR64&+D~)XOQm2ha>kx+1BGC<9`B?WKHWVHHi{PVmSm@
zqZ>`URu?R@qnWk2z?OV@^5ux7s~nBUAvUx7uw6N5+%mM8W&~$QHWYaP4{T+sx&x65
zYJE>dc$<E)$EWR~1>&~3qyNvdfj(}mKGKhIZ~Tu~f5FB5pgjw`<1L^@J;8(#P(Dc=
zIXce$Vzfr$ahVLiaOm384+dz-ZkN@roE;D}@K=D{1ekMsjYs7=9=gwC=WMr<jhA4L
z5>%69pN~83US@)9$BM<{N2~&;^yM{5Sqc#acg^(ope-f%njM$g1w#@zZ~BGxrwMT+
z#MAPYhqO)c?2~_(?BSv;+n1MMcMr%ukEtcIPHlowc6HQIq0fOW#n`zC+qLGt4P?kJ
zslgX4YGxv-eg?&^r=U^BlX`wO%H>T^U!1g6-9cGi8kGApvEK{_$pVOdlb$?CmsZBy
z?MMeOa7a2ydHx^APs<-qmerBLd|($YTWeNIOg%71HjPW}10NEZTbYT~qI0T_qyeYU
zvO9ZH$iT+|#rK;6;)!+BA$u==(2V`mC~sLvE!*e9MF6O<J034(jE7teUg$~XWk47Q
z)qw1S^7GPA%|`Y$QYy2gmhxnQr6U}{Yftdx3TM`M-8s7QialW0qjJitdWXjrACAcO
z*0rXz$OfC!e~1cxyfq?qY2Ytu6Jx)lR0sTr5?3W?xV-KDN5h5s;eSXqfsvmaU);(U
zf%(=G))5b#?HPmWQfy|^n})9ff3p+s`!BKAT@HrYlz)P>VNV-wADd_Qju{j0ncnzM
zL-jv8IGLWLXi2QDypRmEyEgb<6T@nzsXPJl5kp~$958Q9RUWg$5<Ov}o=se-nY!}*
zGEgSjtam_)=`NnNWH)8*+sj{iWP)I^r}^g6u=+Nsi8LNX2JIyq>gz#ezYR|&HSNVw
zeRO#y_JTlP%~QPW7p2ml9po)^af2M-lI6j%xsW@#kn?GpyMk}FdkE`Od!)W~gLs0C
zsKnAaGA?+J4sIeMJsHtYR;Dqs3mm$gWGX1?IzJV6L~t|hHovZBye{co)vlt&NbDN6
zo4v~Mx)SC}KaO?g3T~z6CEA2?&=VPK;Zjs4Bag}#DmIvt7vg#aEhdNS3y6#i`$fP^
z=J-<qD&Pdb^&Coeu<`Ks+QMtT_RQ8VVr$97^!x6`Md&tV%RpPCq5QSCRTSGa<8*IO
ziO*NKD*#K;ji(5@cfSg}I7KiD&ouqfQJ({WaPF2xa!xn&&jgOcU`Hd_bkARyTb7rb
zMJU<y<71g&iBi~|GYEcU<NlbLaV6Z|@=LWN2kk^llwGkmZQ=<v>a>8??YWDANqT}G
z4k7G~*soPA_?X0bin%=;FQvpgRU-?Xm4%M+69a|@>~6-Jgs2B^4?EPB;b0R6;7nM!
z;J+n{xNb_;`F(-_Lu9r$=xe9p6u!S+oB&tLcP&Mcm8{W#{A{X%SEEG`=4s$2TeUlV
z`kiFuksXV6RBNg1;mv|$FbJrdsojd1Q`I1yr(M1kkcA|#e|?3NKwK+_gYfH)2hww`
zq>2mXG>RmcxyuF*U0VW9Z5EezRUPhu*+uf8D<mvv^gE>}1W4OS)j?H2#<0tYa^ST{
zHm*#u{$aky>LL@6c|LZ$T(Aw&_CAd{M6YbyQOESM??<*I`n~a7&hyPzR1J*|!S_z|
z$&=+GbmEP=<uYzY-9zS)J1lj1!O0_Ebq+g|cG7lQ=28QX@Mgn+mEenrYkLH;Cw1d8
ze^_3n`LOl{eAs$ZBv^{g6^Hcbd0{FqNkkL{^$L5Dtz&WrN0bk*MNyugG=<2z|BYAm
zd*9(v5}$z$)G!H$_~^yoH>JPlXvxaTm;YMgdTl}?3H;HTtV@sKU7;n^O@D!smQBY;
z@Ko1Rhx0rZ*^zGh?xAF0`_~9%GNb&b!8klzJq%AcvWz$1uh5>N{fTaCTvA&e-t*|1
zgi8dv#2>WL|Iu6hmnqpwkh3*2F5U)BJ449BlT|E8uASX?|I?kq$XpaRuNtW)tG2m{
zwB1Vqo+oLY-}@xcP9Pp3OZUg=VW#{H$Y9Cyqa3*_NFPPmZdXEu$*Yza`c-9p*UIJ*
z67C6n$^|s598@+MVIM$_4w(**obAiv<j5ZLICe-~<^~aYwY6BGpO?`td+I*qkzJtl
z#ltD;4GswyX%qy-tsqH3F9}<t9Z0x?z_+AVj8Tz4|BM42#rsl8dH&M&y*N7*aMx?X
z%4ljd{l-NjkrCu(5MkHeId&xB`zL8xlI9gN2xh*vGhnqhlXy$~)b4K6j=+#IQv3d&
zR%Q9Cj^`qP+cc);Va|pj({_=yUnQ?f(L+<r5ccAmM!_#Ta`K4!&}xi07_P@=kYW8w
zqabU7OKnl2QJibn)u`M~owEDO5BV<sVklV{54bB-h6430maA{#6z<70ti>EfP#p&0
zXIs2Cu8e-JDqO$2G*p}PTuPNh<|;zi+s71o6Sfl$TI>Pa@zsFWC;C!@-DA`fbEhX(
z@9x^r`lCHc*}l~DAqgwINyalX<ZI@aG6~H?jox@!(|~`UZf9_4-m5UoY347B)m_gI
z+SR&Vf-4x3d}w#2>?wL!BIH$?o;do)&ViRFcn;|b8tsNVmyRW~xG*lge3@KrZZzN+
z2_=~U_^TUAqeOeGx>Qe$98*^_qEKR!uA||qj4NLT8Gg?nOu3oRWMAl<ih&phmx42i
zq?V+&doS9L0u%;HwvSv#026h?zEI0-BR*`!m;M2qez(hpMfB~!<&>;wzBet0a|hjE
zPyF!0Qp^n7D?Y-eDAKFB{HR92BkXm9jb9-xswB#a_Q;`kA|k2WFXF}G{W^?cO`$Od
z&%g<>uk?v(v^Mk?S5L{!;2<BC8C;3xl~J|S|F5cznp8CIvfQxZ5JhosV)%K>-`1;t
z&l3Kz-otAmE+rIaj;`~FYMI`)SO{)16{}fgu)qqYK7@a#2lLmKlr+1kamfgp)JZW%
z!c%2U`ta8K5<{2SP^)uL_Qh}|`E_qqWIInAVO4honZYbk>@!5mWnZ3v=0_b<lC1@|
zo_$D&blc4>u;Nildm%!l>F>S$ThM=zs*2M>{peeul?F>5{_1k_c53MGx5l6Ui?p|Z
zit_#Tz9|6}7!Z(V2oaD5X&4$55k#aU1rd<$7*bM@5+p<#0g+}zx<k69ySo|M=Nf<c
z{NHoVbJlv-V!381&fIg~SM9yO-_Mq;&q#vUxt-XYovSPCYsbay)&TQ-e-Z%gDIWS8
z-p`hKeJOdj9?uO8YCs!vZS<JW)opW`DwD0iyJCYy%bh)vE4btb`q5YAk;NFci0e!J
z7Bm5$Hl3JOYW4l7oC3t6G%(y4RqSTH6vdHypBJ0jq81(`*Yq3!zX{q~tb}_vGik==
zwa7YNmTaG9A<c}db)Obwrj%rOy~hrQ>M}oDE1+Hi$^FGvUMA?5BfkqjL2p(tcw*j1
zfM)h?^%JTwzs|3zV#%4`z{R-xhCy95giel7S}l{@oeLubBk<Ofk1=#Ck4YF(Nmv83
z9I>Q%ZZObc2Qx@F-?))uczV6(m*;ey!(^#TZ$<Xz<L$?_2d9%&@UpHRF}{QC?Jk#@
zxr<7SqmW}Mn$BZ_#I{SsI;|FhUKoE#MAHdR0M}aGcj4SqAUEWDw3XACXu<HKbC#?q
zk*EpbwKrrY+@m$+ELa}Msu{g^CD}&_+O%Q@jdn$S1u#47LUv&rp)s%h0V}Hf4os&n
z7Z<WsE2Gq|v0Si&;s|4r>*{+B`bh$%(T=O}QI931>!b9%6l-vUef}RmUAM4ybKkgd
zX1z!hvNL)_w8|c(Z6b{&EBX7!P>U!>2UjSMH@g|UQ<nhN9js4QA<rpD(TNEE=O5p>
zM`$WOu2?V6__8!K{`qBJ3OR6b6mri;h|r116)|pM5HRVra9}Xd{l9*=ebCg7=1zA2
z-N(Ur2Cy$g;lPy^wssP$Wpup?RNJ>WY}_6h*J=ED?cf`vT4QMH+65%J&cEfMxd6Br
zW5D3CRKoAMpTE!+@x9o%?WgC%7G~5Z%gBSvqL=v^EceGZ{~!O%#m$T{I-6+&O{<F&
zt+24CXD*fn+h)&EB$8II{C{q`<P;Y0+gPIjtTp=`Y+*TWx9K+lnxbqy_p3meg&gW8
z2a2L!M-PHuXZ8GF+=H99TtRE*O%PX^n)EDxa!}R*5)3w^dB#7Fc}Q+{SkvGeu*&!E
z<6LiXT>%`MT>`DiEGN`6Bkuvvkh<q{1_E@8`{zRc_pj^eP;Y=0=^XmM|3Cl!&3fCr
zbo8^foHU;w|24{f-|N8OdeGr|tDQLop_(B-27(50SxpMDaS&Rx#oPq+#H(4H4bY(q
ze1!(Az|big0$h{xAQ@H`{Qu#$l~RIxYx?b$y4PO=n*0_Dtz)v#8di$x>K!G5B>@ao
z6WmJX-Vfa`_S&of51IZoLl$WHJc^3lt{8^L7`SDzdVt+Nrx4K%k`%VV1-!0^426g?
zO#{#2WUsTe2FOn$Ox#b70}lnP#t~Utz$5S_j7GG~Jkbl!00XdYzbAQLorYn79k0D?
z)F_nSD<YG03TGnjr)_c7mYT5famR5lc~51?fB!`4H&OFu>}&N0oj<2uJS)}}pl(g9
zz>valKnU0b*632O5V!XTExb#3ADdRas-_}N`lb!VO`&z2MXv&b&Jr{^m2rCrCY(62
zq@)1Y9Yq1hX)7lqd|fCR6|Elw>ctp9_>AlcZxMU~vAzuM39u+5M3v5;6AEo11~<Xp
zYO|TdYZ5|ZTaWJCc%AuvnYoE8@d+SH7@=A_i-TB3z-fKR%m4+v+EsQ89a@8pVsv>6
zB44{qw2AN{kB!bZYJa^`vRC}}-+y5rTm%>>h86<jjDM{RPl$bE-qVLS@y$NB$TtD9
zm;_JIjQA_-@f<P!8wG!Ayk|DRy6+4WR0{e*DmOmdB07Phi$&4^+tHp&pA^oeO&gsB
z^*K<xjDrB1|3_5P79i-%SOJKT!4@D+=$SIF=G?^qLByLHu%3ijea$Ng*#h`yXAs)`
z`7Hrk#oYUB5UAxRtV_j@N1JwY6{^VJN+BDLKmYtXWd+^%F*y;sRd{~I$J@?2{O8!D
z`9$#-3*f&Vt$iPT{C)Dr%p(4}A@zj7x&J!G3M2taGYPpD+yQ#Q5`^1vgxr|@_FA_L
zbFKZq-sC_;a)2WtO|~6~3(8P9nq%OXar98G!t@64s4;*^fWlWqXU4X<w>Bc{b*A@k
zPk@6M6V_(rh^@#beC9ma6VQri`lai;)Q}BI>271-JA|A3JQ@DuJ6J?M<NRkm{r@hC
zv)*7exPacw|9ds*LIZ1h5897Pe_T9z^y@r<gx2_~k8rgxn-fR^{C)VYSA!OBxT{xu
zMj?vVQTvE*E$*&V13+}(pb{8XfKao>#+k(a*F*o;zEDbqn$6#DX?p#!NWap?09YLx
zqud@8nT3a+V-T+gpk0qq;)vF88Y;En9-cO8%zY4;_D%p=Rukh0!1tJ=ykYe6<)9aF
z(*oJx+5DH#h7l)V;;lbj!UqtULg30bIc_Fa83j}n2)ISlrYORJCEz3B7aGgZW8JAy
z%T^A{v;?Lp=cNyt<G@yLzUTrb)-gc-jO%;s$%z6n)>blnyBLg}3BdkN-2KQ8x`)M<
z8rGC_5vjXLpN;i;-KNDj;9(yD@f7(ai#1!pFMv)e1%mNO!!wYZ8o%24b=8YXmrj7>
z(<?w0^5X#c=9VT4ofKe}y#zT}%jTfAa+DJl>^dX-Whez$l)yTUFFIuw<L90am~BsJ
zHT9f6%%QKMm{uI$hMtxIz7cA2lQ|(gdcbO%LA!L~GH#yz0}dk)P~y4-+WpqM9eW_9
z)LBK<8E4=%RtxY~IHCfz%lw7KG6ceRVmGViL)aLs^7uZlXTy%Z6jU$pa}F6TwNlyO
z?YOxKTA73MZ?^<h=M3PM9p~G4cG`OxoR5#lU;2xu6?jA%bfl7_dby8*w3goFeZ!<P
zB>38*0}AI1H7TcErr6YNfq=3yjzS}Jcv@z9>H+#;Sape2LB(<jSd)c10%r21i}Q`Y
z=0w}VK^0-^OLlTe3z@;BhJmNk1Xyaf0I0>`tC`U`z~>u{=)<i>MSDII0L0ABchHW$
zNnpmqp$f2pw?G{?*!b(X<iwAv(TDI85RO3yIV6CufhSDF84m1!Z=HiUg2q73VoWu!
z1{eMO8~^2KSLM2Z?O8jHi6@Y3@spEHTw9X9T&b1Un@z6lm#5+n&$sx@f<-Ovj(8!F
z6NIAbgq?+CYp)9;g)OytxNCL+F=jY0-*^UC=57JX0ksthTVrl%<dXzfok}Y>ryGo1
zF}<-#F#X6a9F=vn1dG}CT(~O_;J>{Ekhw9C$|3IGr7i|vq0MPjUay|<UfD-Dd9To9
zife%xsr$DoCs+zT4t(V6R^h?bVD3AF-XAA9SEAG%ALI(?W6ZXF-o)ml_?QhNLgK3Y
zaXfe0ns|h9FJL;A^(yAi4<29P=7^mw2!85;0YJ+;7ZME5YhYnl%G*f!Yjyg)p(7Z;
z`pm{V&iUt5Wei3$=qv;ua13_X;pxLH!Is(k4CGi6c&(^6r+<gy%VRID17?RDo)Mqq
zd4Wp}n?E-$E#W8nWhpa$la3fITF}6nK8BL8-CMAa!U;I_WZQNCc6gg`UJT^mq6oYU
zfPBB~=W~l;82>I<i%$tZ^O%L#;}KCXP-5LJk}3kL*(HlCEn7a~DQ@5>DknFJ641mX
zd1);N=VB3ptg+|9^WeDU6z?lSbC1(bgPv@Dj<?2wWHV7@`?99cMhF%)3OZ>u7eE$%
z$f%VoiXw#&3BG&4Hh&LG?Gbyi?84JsSxdRoD=<w%4(WbMlBnWso}<8s(#*~JLp<v&
zpNNLpi|R3sf$xP=)<!Pv-W%$9Z`8RhfJ`5ki0RD-YFf|3e!~1)64Li%vy~il2ykx`
zy(>yt2!;gieufc4=Ea)t<kcZ-cEYMPu2Xi}x$K{pFNEJr;~4P*EGf#Z^*}0XZhx;9
zpGbUqrgW}>2d4~<7hp-C4jTRd!Qu$!bx6G+>_B~po7&PXsaw@WICbY!JLy*at+^O`
zc1arO#rpseZ-N_#Ta;Fy%IW8JQ6EU3o0iXb@$wRot&D?>i`|-ipWwj#%y5>~ecekz
z(|4?!9Um9IVB#w0BxVB6OBPWZ)$}g*`L;`)g$;aSA~$`*jcU-}p2L#eZt!QtZ8rue
z0x~W>&n(-O<Zm6|b2<Le%j%V$Q`L-zEi~c2`BS<qJ-Qn(K>TV#CGi7_Y0g{F@U~__
z{hWYm=wX7RyV7lpr$0?U*~t&x#G09mAWDA0_vIYd1tm$<0R?L+RW(!C-dmOXvU@{1
zSPP3^$#amnHN|nGE?EpPFj_)`hDwy(=yVsSzk?-M?$gvHPWw)j0xboGQWgO-=f3RK
zTdEgg81mbi`LcC2Ki^JRS8@2akKTnbNJjRi3P7oM8?cbl18sb_k5bNC<7y9?bU<N}
zZS3f^%Z59Qiu1TYbHZTdGP}eFq58)h8zq{Z?l>pfso8>YQ*vHsP7<p+h?>}>J}G}a
zS#G!Ig%dOb7W894G2}hNTaA7iUyK#_$c+wbDakko_A;sXnQ6~3P^=(Mv=Wh2)A*uW
zRPn-JhSmQak_28u->Ls6o+6+7eF)43u?=6v6Q+tMbQL*Vr?ZasMB4Vf{A&x)U#|92
z5BIs=;4BWc2(5wBwbHgQOAT1nRXDvs=N(t0ee~_cX_xn*UhiFAX9Ik#a7FzcH-&8<
zq0w2tz4jKX_8JW)=;mY2oQKnUiPZ-@6~XMMXm0iqZ`uQ3?Ex=7*d12(Un2xAd~3xk
z+Yh?;iPu&t7dvZK<fiplA}6D@GdlyQ+ePW4;U1K>?O+oA(0y{-+^%0{I!5<SoUUUO
zicIMU5cN#ZSx8wJ<_$K^cMnmLJ$uzB_UMW+*Ouy`=en~bOLL8Bj(~(t(vxR;gbD#G
zfq#+_0un(kF%_Yl9~oTcG%SDpP=LMH%_KSMk|TcHR<3oof`z?emum}@G1}!w8YEVl
zT5h-36f%*y+#yR5NP46@#NYAGz6O-L5|68jxQ?6p#oL+4C<L~8C3MmHg?i-lh~h%k
zqv0`%oaqna^xe&eRLh?R={-r*J&97;fv*`QE;3XtGPE7mV?vt0=??MIO@B)1^*S!K
z%Y&mm+HTLa)73lm5%1cEIyc~0P>z^?n`c-k;`oFtj__nZ_0eSvr`FW3j4xKdUo0MH
z1*3(<@Ho9LlGJfgeL%gFJT`!8IJB57O(_7WT9_YSl<sr{&`}L}g&M~w&7nu@;gHFe
z)6vJ4tsudhVy1&oA87GfH>$vYCO_RGzaDrq^S#R=Gg5<cr1h1NOM4p%n1q-Mq`FJS
za+mC$Gc}(;bu=$iqlo3>I)T#gN<9Twx#_aV+s6*{&h|S=cE3ges_;ETUZKL)dm-5!
z9FUxJ?<Udy5dS2COh#5SZ>EP1XXrl>JYBfj#u#nBd{^0UDT>qKdoF+!+b-RH$lb5s
z;p?hvOYR)#-nf&7dpV6WwIR1id^{>Mj%BiFG*ccI=}fj4C~j}#f?3k1l2m`(IDYND
zDlO+lsPp>#rlFGP5(F+qv1hrM91$J*6DQLTzh-<&B)K;*w8l@Su&l1<9J2Q0kJJMB
z71o7uPMIg9i36v)segp03THabwQR8>Ok(=0h-Yu&)EK_pRgjF^?q*MEi8r1wiS{R}
z`NN2`^vC_)v!&!+0;d}C|DI>=%=K}bAoug<-c|~PXo{ifKA)Im6>!N`eVigZ+zx!0
z-r$5w$*SF!=85zGh~YM`lt732itQQ>EdT8!PV6?}TJi^EZ)B0|Wb=NN0Y18&b??Fs
z>2=w%W@_%XIL8L(<8s?^>6!z-_1l>x@5>VQn{<TnQ7*{XxXEaJ6JgT9*}>b)vdS=V
z2m2k{7QJF+7%c!35|&nc_YcQ<hRHzj5;L70izztf`CBT&Zn}L36}vCG%d#qZQaQf1
z*ZS48(G=S5H<BU;;wKkg;EHv$+xM!gdYRB0?v(?dNZMz@BG*qUi4=CK+uM5F<8rFi
z4=Naw^E)2#BwLRVIrf)rgRVZUUvOo@K{Fe|kM{-6Zfp_WR&R{;&`mmBmA@ZbxHD)N
zSS0vs+vVZ;>MI|X?_AAKgAWNMN&VnsH={_XzhYoKTW7*_R!^4$5{Ekj%?49P^x4;=
z^5CrxK<w<4zWo97G#xg89i)h>6_>v@@@lyn`TYAvhk`PWpL+1t$wG248pjCc_}#6z
zhWJXE+ZYU2i(Y!waZ48TL#@m7?95h6yGc79%*`3NL(c=iiJl899Ip#s{m0h*oAXr_
z5F<V!V<QVggR-E>M`l&QRi{4a4IaWns?7Fo^&dS*TE<S@V~9o+Ji2O%qWkivwpLCp
z%SWs3aZZhU&F7q7seRzIBL1D{ia*72#a`Ac-G<l&-ZQO~P><eP(l?-qF8e8E_$mem
zI<%+kIjpWZ92GDIBbdGk(qvm-ebN5OuShC@IB63%n&8$SX*1s_IUis1>Nz@cLo^y5
ziX@vYG4lR21UR2F$=+8--=mm5Cj1eDIsJ_g!M!_kk1oA_gLI8tk)CcEClp=9L$`FZ
zy)T;_zgTZ?H1|kf>lHdyQW9*-D)^4)y%_Q?meAOak<2_o*10cuD%$xNOCjGG#OCNV
z$U`EpF~OEzm>kB4>eHQ$*WH`tw+JBB$RW`ojZoMP-dzGm8h1N~JP07xHQYN&W9Ou>
z-TNp&m)k5uHlIY@BfRp7kYdiQV077IynS&HzT9*#PMcj_0ai(x8JSv%Q&~w9`D1GX
z#9L50&hGlw2x!VOh=VRl&EIm9a>d}jdwy-aMg^0qjPYjubDD`x=Xr%y1bA$MSRMKE
z4a;XF-}0RyqtoUI1RwI(T3#(H9X|aY!~n%bClX%ZL5#v2M1o!?1ek&?Lr%w6&`4c&
z)-{|Xs6lqpwOD5MM~ykly0fy1b#~1p4q)qy9<~bB@+2GEd#FDW>Qg4lCzH{0Vu%Y;
znV8n7?>$0s$`%K^$fEo0YX${AGOmJ$FEm>9DPq4wXQKDiX~w(5e8A?<WXo!R&$i0&
zkbA|5W?N>y6Jb2X5<R^k4HL3X=s)6L@XQg++&ER9wz^V(CAbXziOWG`*)5Knpt4}G
zoz(8(Go!0(0*kkXg=Oo3L&fqSP9i)lx9dx2#3l%6cvIdK+DlMPv4x-BPj&_r7Xg|D
zzg;^H<h(KGZOd`Zi#dbiM3uJJoLKi$j0(%Y?l`3!k50<OS0JelSc$3jD&o6;B&vKj
zhtb#?i=Nu{zpc<rYb~oD;Ch5qHUlAMbp=cPByn=Zg8|*XSFqv2Q<CCS-VdE1)t0Cd
z@RG_3q1vqK*5@r)sc;X)0-`XPPmUU=86SipV@*LOuC~IkV-hp1!Znj9;^yP-0G2Y;
z4hq+Kw{Np~NJ=(#Z3l$JW3~Gq3Q1h>wOmvn0`7;aEP}|qA;gS0%SZSQ9}?FL_Q7GW
z_M$1fgaDhfq$}}W9A|AQCV^1BM@L)vw4sSA#lr}ycE3=GuI$Bk<g<#}(by@s$3W6W
z^uRnO-LrnMX82q6)1*_|2DO+pIK;xOW}l#ANZdm{iDhSFNpP!A?(;H6mKmF0OVmUh
ze3dtaGJAj5+38J5D(AW5ywya83ckhmaLJ2DY+OSN^2mYLun#$^GXUE#QdV4E#5pjl
zFFipU=i>Z4%jx&IRT<9ZRMq3L=ve+_$m513N*^zh8+}^s6*r1Q7M2*AZUrJjE2<=T
zN+`VY=G7LyjjPOHarE8CB_PX8xpf9OsOIBAR&%w!k6q^M25UvU7oEpDJ#9fB6~6Ic
z_%@BL4QqB3yjsPVx>ua@Ds0MW_^mE22*KTU*$t{rCazT-2XJ{bOzLFzV^0JqPnd;g
zq^y+A8SdD_Iy$YbBvwW|f3apitV8W#hC;-EbY1%bhy`@-*nt#9*<-<}q2DP$z&zaq
zeVYW32wJ2w@{Sef?P8SdhF4U~;UKR;l-w8Tl1!#m65)l3d`Y_j@?VN#+2=UyAMAMY
z>WeuqA~p`Leux%&8<Ot`Zl_i4S6BDfT(wl3x(p9Uc)d;_3wMdAX|4IVqWVjcV}FVE
zTNo^a#(6aD3ncZPS6LK_t$ltB(r@zKH$FW)2QX;kYT`cA`l@kMlpXxWVviQb4}|T!
zXH+mix4eH$rdc-azOfq3YE8wt;WYd;_u|mLkeRllL1OhC()r9T$2&tltR%l|{N5S!
z>dI>}manMrHko<K&m$(q+mgbhgKr=7((5}Gk1Y4p2xZY+7c|*|w#uLHLujnP3K8ty
z>G0N@e2ka%ayi4KS$sP_YDsD$^;d+;he2i0ZeWyY>$u7XEKm02o$({a<M(=J9e@*R
zw*63SonYtAsQ9#MQOY*G_~s2yj8Vv0Fn|$!qhEhG%)e2@ufFG8nt!41b1A%2eemvL
zSyC@Md;pRl6NR(m6N9rOa;0irO?C_NE@u$d?FP$VDj}Qj22uriiKpQ)8+L{FaK^<h
zxpn<!lSeXoa|f>??LE%^RV689L#4t--fPzWnF?dYfQhG8d>40s*!P>u1o5XciXhi^
zr4TKKr%$e<6Crs<l}xRwe1xKuS^c@1)N!?NI@x_v0odw3HYvH}E>{u9P15IfZ|E^^
z2Mg-pmoGZ3k!SF|F9*P`IKh$1t~a-kH_Wy_7ZH7jM~8%~W4(y+7CwoIE>7xs<)7jg
z08~eU&B}PQBu!6n(Wl^pXmd(?A7zROsvdDx+`$g^I-i*N2rn)HNu3syAl=xfTU@&}
z1+I$MzY6qL{k#-^Zz4IB0pwEz4;GJyzY<0^sEV9^nb^WeHy^jWt^QrKY{C;|E#S#{
z=xJb)|E*HIW@|4g)?BYFiy$$9Z|Kof0UJ~ppPdN9KBaNH`$uGe=_DU{swEjY0#z~Y
zuCeqls&KVrBNFtBAEmN{ED4O#z2@n$9&r%eT7m_8E><i7)5%O~98)2pA1rBwFGe<U
zi7Uj+3WGVu11#GJd(ICJwg~Xk=orqFuZ=FD*X91u@yy|&-qcffAB#<Jx`(mysd7i6
z`OEKPBcm@GQe>KlX#c>gHwe=MtL<DIdVBAxH;6OXI_q)bD-^tbLD<)&xuiZ2&i>|}
z|Kmvox9r)eFt{L;vRre9(0}(+y$IBMub>g`!PAGVhtu|QY=4GL+e~KId&manyaJGB
z?rDmOm2YQgH={`BD{>)~tDUoc<lAH(_ti-s@a;6>h7-~;Of%Lk;-HgzGECaIZCjI0
zo;!GWaGP|HIM{3vL7>{ZY3NYx_mKI;sSB`Io6Gv;T+@%ihho@DWKic~?2rPJmG^K#
zLS=rByZVPH>z$)Ro85j^23{;(58ZYLVW;GhEalw>+#R!)2$Il|W$kyh*I?nFVGN2(
zTK*-?4W+Em$rC)HmcfrdE0W>Xkp5fd>G@Fv^Qi?yP^V~^z<?To(_#j-DNUW}OE<%e
zB20u8@&WCe_W7U+VF?6V^0$fW_YR7u*ee8#1ZC3@2C&F1<tN#MX8ZKH@O`{<$s{eH
zYmpIBlxRRD4qXd>jN~=_Tr5A+j-9z_MvNO)Y=E^%dVEt#a0gwOuJUf(?yO#CHd(lH
zpxMV-1xs|hqUM6Vv{zsIzHUiumoKD4qRt7n-dd;~<6efKzo>%0grIXFnKy+l+dR1Q
z{|X0CaY5fTbZ;Kjzbl)dDpk!Vl2z{1puJC!znIj=!^AyE!7G>HU6Wkh^j?KE8bX5T
zGG|s(a>aRkHcwV3L`P7tWfY4((MfgD*%Oc9%#w}m9O3gg?x^s9aYwg`90*-~zW8n!
zC1lGUN(C*_SheE_8hw=cb<d^GTypB>)69{&kKy(io7m^ZH&)Ybz*30@InU}9jYE{6
zTGyRnsr~k^A7;HjM$ZY)zJDZ@xHumJR2d;(d~5jZ>YW1JF^~Fi2H(RA-mzAiM{G;+
zRk#%7P{I00@~Vj|w_;K=2Ss-?!iS<DJI@}&BuCYcH<HhCCs;%5&+X|AJ01D(`37HU
zUPCTMRhL#RAsSDqp+nburYVvIhs&BZ)Nnj@bIODqWWnsDDa+zaOc<O7vy%~edgiuf
z)DbTt_$yhpQ;akp9zuQietTG8op5kL<J$UdfwpVjVN(bvXmbQy(RhI{|MxB#`6bHJ
z(5w$`us5w`nf-fbRPPlxvE=Mf+WvC))X>yz#Aha{tlz^(qSv&kVS33UCJr3odq5RN
zO*z^NRjI7EWtvUTS$uP~XcWFE9lMB;?Fw~}GO&-hDvt0%BV%Ps?wyqJviE&#57ELb
zS!4)rc1&ts(qCF|n+M%gg3sf<XFqU)s?Z`Iz;NQ-wfq*En{JRsPK4B(zV@&U6bGZT
zl#|9EV6N;bZ$kx`$l!O#0I}cd`HE-jHPNWs18NtYJHFfP<!hfV{5JA{=H)H&S#pvq
zrw_XKX4UdXi}&y=Mb7O<S9|KPeDk&G^JqAE#0tT$BHju^)>8$Yx3!qA2esIblE3Ia
z8+0rP4WTr5Q5o3tQdRP{OM)fLk+0`zs-JEg>+jIUMu%>u8_<ndkR3CfL34~xwQo<u
zFr|i>7PQ}aaV0P0=p42Y;#f6F7%5%N>Dd=QtB%u@pQU;j<oz;Amyop0j`Ju)#2ILn
zvOyVHd^{{viY+6D5&D4$Sz#w1i(?fjFe^Q0D3?GoA*-L={q0_-S4n6RqxG+N<_bdL
zaju=&A*PgvOlLP-??U-rDGd>iiylY9{7Y(bntmQe$IVwPAu*<}In)J&7DV4{V3gSI
zFL8Z7FDNE3_A+`kn{fvgDm`SrNpmz2(EEbVjcl1#mGrFma#Uyf1uF#kiV8_#`9!!c
z?#Iy@+k%#*uQ!?=OzHQb=daonHoZ@%$+bU^?S#HWO7(Rw!<7|9JVG$%TG;8ADM1h=
zE!cbwj~sDxyMbkZM>oL7skjRx7-&~HZBK@z*HZlm6t7II?IW&PwfH!?pN@3&OX&@>
z{tVxFveo(KR?55}=iK0DSxOxzf@Mc;?xxwBZm$8qJfrEunJR%B+K=pG)}7ZEJv3!u
z_#7JUL=@kGb2qvj*$1&HGWS`-qgDKFB^Be{DN3q7cWh`Ew;-T&0nxNZr1bpyZ6ilf
zx~K`wLTSmAZGQUKuh3G}QxCjBN$hQjxrwGu(nPznWeFVT@zrNIbC8RZQntzSp!_eL
zua@zSpMKx?p-}6k-(yKnH=pm^ATHrANhM|R^(;Yd5^t67Nq4G%Q&YBrjhk|ng>x0x
z_d)$_KOkW6K&WmSbEOrCL<RW3JT-#RpMSnKM~~UDEmty3|9C>wD&sVB-^XBDU$^Y+
zi7|+ej6^~=P;)CE%q^vepTEzWnS{0x6?^YOeDL?=wPpQSXy)F-JrrSEukyZk=y)C5
z+m!9mHi~!G@beR_P`oMc=%s>L8CU;;0_*s7JB8Dl52mEDYN(X`ydPJy;1o72W?uZu
z2SQs=;=Rt3sbZCo3yYk0s~2YptnaU()bO$lh>wC4qgoib>fg)eDP6=2C;_>Q%{*K0
z+){lLfvL$2B0UEE0H($o`<kq9`kck76bOM^p*3mFcN2P&wp$8UhM7}UbiMl9a-0LI
zwb)kZc2wq)UC#)dmNTeb>gr78-TF*rMUU5-&L4ZJtW@FDxGKBYxZsPn!fy%miZAeA
zpC$}@uYrV(IOc==nVP@n1xqa%-IGbPj4<iQ?x07bT@4s3(F(BJ3r%}Qsn#<fJyZ;(
zG9W$sc8P!#c!O%KK>7+y9=d?45RSIqFkiY-ExYy@XD#SoB^Jp?pp;rBn?dtu8j712
zck}y<*WR$>*B+wUPt9uhNgOguM3FtuWceH;zbWD*J@d#{5efWCSi~8tjiyR+uiqTi
zeaxS{FT2cw(vWoo0WzltcDfaqB_7l?PSMON1=J@j(23q1c9t~ao@_0};Flh+Gh$7W
z3IodV?J%avbgqeMg0vy@U61zq1L2pk`6BCkh@iM2jp><?y(Qm*T;KG0JDFnTs?;m*
zMMmgbDZkBt2(Dqz+%!U*`Af8hRl~F%4@b5K^j5BysRS`N7ub(nzFtn@Js?ylAZ|HU
z53fp6kQ6iH?UQqRhW~)ONG8GX>j@jgBern3(`J;1#g~i}n``1)3QS0h0?}!+cVXRu
zFwjwp>+V|H`DTqwjPSUu{xl{nDZ2nmQd92n`K+Q%E&VihOZv9m!W?%0T?PlSjIQ7(
zAgf*jVfb{sRxyAB+w~>iwTjcn4$sAYv5`Zpjr+qB&qmTUiWNt8UE9nzE=pNi&?@F6
z1k)>j-i8T_n&7xV{bSX)%Vss?Ww!eG9|g5QKJ;W`H)O-0YgIiT$8-#;J9BK%k||e;
zh#(eE{`68O(X4uRb|3X2tlNQe`Y>0FICg>kmo+oiqGvdS2m+`%&F9a)N_r8-)rX$D
z^U%GKa$Q!dCpep<q@70T!spi>Z<K}*78SUEm~(U}lmI;@7PP<Rs?fEy-GZ3mmZQ|c
z^UJXtF?@FWO0u4*+q>Z|AV>R6ENG0vOcWMi>sriKhh*Sgq`@yX7sgYEb&RYnA-h;2
z=IF2%lhQZ8QN7^edrHz7d!|}I<uv+Hzaw{8__W3cl2uMdXI2fe>lu0ok?iuM1JBCM
zR!eA$XP#O%QG?Mzu5?{lO0C>k_*!B69N;~?g;|{)twOteBZ^n~{y7hUgVAaj5H-};
z4Ql^2el2n7%F;8iAtD~aNtiU5sNOq9AA>=MjQb*p)tHw)HKSv43XqA_i)@(>NBd}K
zG)b1(#^mT~r=6;`Cr}A5564w*aiZz#4wAAmA}2ReSf8z0X7dG|2o;9clg;4q_L;Xm
zlW!py6XWNmF=_FjSI)r9T}?_gh_z*^{S;ebi$jFvhviB$`XY#?@sV@_lh|q|sDgcy
zu?%n}Oq1b>l)&bVU&h+O+RD<ompszzemkbmJorJt7hUX|F`Wt$XZ`n4Z8ONf{h-tU
zy|@&No||Lav1A*aE|%{-@YM?G(Gh!2Bow!CFwCvbxtf`6{WLqoDbaYX+(gtHL9~v0
z#Anmlc`pL}lYwdK^n$R@eY)gs-^ySYq5Xcv3pEAUu2RB5Vw`KSQOEj)C|9TzOEw=D
zQI3&AykekGsPEb<W$~k-B2jg>wC)v)MU$`gU(V8#tSUUC2*wqKZSe+$qT~xy3y7(E
z<5rcv%3r4OeG0dhRY&wP#cDBRbxFlM)W#1WjJ!|xWB$z5vdKa-N_eC>^u}>4fo+z0
zlfx_FL!GSOHM+Z}Uoybdf(-a3bvwO3C&BbSb<x)te%P}m$j#ltA0Hbb@*Ok+lbsR*
zUe`p~g~e!iBj_O)<LtEKke_v}XRV#VIvHIkNKAcf%)^7`(YQ_>QfmDd0Nv6$+krk{
zims&W-(|*ej3lu4$g$XhMAl_Np|5wZ_Qy~icp92Uwyr>gW!^^P@(#|u70pWcHQD^k
zJYy7VT@D=YY$?EeG?+-~{GM{I5-#%ywbPTYVU_RfQHzd>CxN<I27%aaeXMqmv}E$5
zugM8tl@}`lR&HcjQM1zxER0bWo=jmUSGjWL?&G><!=vY?%t6>q0489)hPQ5_YxprM
zYp5^edT_?Qt9iS;gRYh;kFVTcanUz&H51eX8%t5KuD{RFR}GVd?}IpvLpAx+a|yJG
z5id@){7lRHGZeB-Vzuv{Wug=)M`=?EB{717c>@s*2T)J#!<t>5I~dYgteiJH?UOgU
z6v9gjmj0ZRp6S#lyAduKM)lZ6Y#|K&U|-pzyet@Wk@BYIQ|g=XWtN;i6~WhjFuHxB
z0VN+cIZqHsMkZ4454Dai0I3zbzya*x)qdWcQ!S2vS29FEHZtf8v*68NJ4-z-)2xBg
zNADJEFrp*6YP#ZG96v(JKdQ_ba~5%@d>qy#F@|lKZMqZ!Gg$m`QcjdH^43e(to3w1
z?TTA=?kU25*(e#Ch^&1S*c5DFyVji4ykrIO5ckATCdxK3Fh`bYLYI?+x<r-2Ty;GR
zIq^zRa*$=39HJl<y%*FQm@BQ7-j`#+86%{~BGc^;ZPleeF4&$P5EwYjNy@HZ2sog}
zRv-(tlI=TKi>Gifx~@G~PHJpu6Aj5Xh@F0Dww>SVQO`IqFD-e%nn*dekAC|=x$s9_
zD0;V?q`X^<wmrtAM`bt*l~7Z3b;anQE}<!B2_N+0ZHWty|EZzUf{0A<{F~|=8BTd#
zGU<3oYobdRzB?-!4|$hfD#cNR$CKm52oVKZ4|iw=o?g5*MaoUYJ9ovYV}gmC3C;p;
zq_a8fTW*C`V>P41x>m6@al%c0l=?2zmDK=R|8}7|x_IsR_7|@GAw=;q6oF3`mjEe>
zhlle*tAp9f1tAU*l*=@|QG#S%_~rI>r>Pf{2S%g*MvL1t4U><wVjkPUtI8gpx5$^b
z73R3aTz6{79=Kj>d+dALVTWpY(3dpBYBYGhRby#>S|**Q>f}B_TexA{kDFFgbzTu0
z24D)=i4!nb!H(<^#6QhvmP8utrr%mEPT@#&%QY4&yG4>&X2C1@$g)4)+yi%3{`opV
z46r`T&^S$qTgiJ(ZoEQZjw>v`>8=kvM+>YlL`1!->_9+uWGK%oenlpDZ8&2OsKAPd
zU_PA+_Cz#JP1iawH}4(!oXLxS;X5IDWG9iX^krqTSN8$camqQ%)?1McOmo%mQ=)o9
zddK*;?>_!Wf`n~u$UwwNoE?iC&c+NzJPB{Z_&?YcW^oKoyM@;~OMvE3%M9)g#VMx%
zo?!I)pkWNXeLGv3Fu}Xokn@#bibkJD>X>ehIj6MXINMSKU!2N{HSCEcr-sew;ipKi
z2{XMv0>PE0u0WGAy9(Wd+LEo_3GHKtU#zVee<)$wu{89YhFy=BDV&Sy;Ul1p-_*W%
zkjoy(LeotX>FV-oIToj^*q&j!hC8eL@=IQ5%yCV2hP?Kp)e)=DrVc-^yLr?l;lr}c
zOBNdMtwq+$-vWD^<#%=Nw<0=CrAJ8Kdwd!XycP|*?~F1-S2n#cc1pH|g_0@G-kQEj
z)-kQ=I86=rSXWEoCA&v)3Ejfzbrkcf>iahJqBz2oeY)Q@kZVNe?1}46mP)IH$h#G@
z<gr!81>rXqeG0O_d{ywL_MZ<)B%iOEW~lg?I!?O?r>v|;cy1amLxb>KRr1)N9K=7{
z!V^?!#UiTAx+jV2P8FIwGi$%A>77tqnwp%;q~Kj*Z|5V30Jvb~2^ofbjfnyX*6n;L
zHC87wug(;HszNXQv>XZxm!Y>@NJLi-IOXc}ze?8O?iec2y1g8vcN+6LOMl2`6;k_n
z{2kmBp<`XqqM>c_!F6UiX`%))4)H!|T9vU<4EFk`66S;JC<~OV;oNI<e+Fsf(^#uw
zQjV=?V)OTY?MsNXtjVS=B%%9&c7R;|@{7$i_yOc^oqpY#20eTmUo&#2cF#H!I6Cvv
zCCl9k9wD^pTYH9w*TraSAE5^hNf9kRjUj9W<d}=W+T`4lj={m9vj!RpD>^*k`ajp2
z)BRX*n(w0uNEfm~c3|aityF;yuF5hbRq)vY>!UgD+p`yMci&(1^E{ShS0{1mrwC)g
z+><8A*J`Bf{|FD~LoN>z`p^Z^c?ZuEiZR^HG=|G#VH*-E(BHM=Qlio0UGd}H=Q_aW
zxpT?@4K`oJyDJgg|NL4rMhyMD^19#=#T<1bNOv6Z#hHMWBLz-#|Cdpn`cT*wo{;0X
z+qlq|PEvmaLaoCZ2nxeWiDY=|TW0;eKj^a-Z3;qzd&&j!M6R1qJ%I%^L|PP7+$|qW
zW|cr56CKf1$+U?$@rBYHoNhaHqhg~#j*f}1v7b6o+Ol_3yL+tZR9m58OUC4c>4Wjk
zpJXbO7q^BUk27q=3Ove|z<hYYlN#)E1jK(-y!3QylBqH{C!lUXf*Iu45q;aWF-q!T
zbIg$2FJ1{|@mQx1=GGZCpfX~4mtyU^wJFQxq92^hPb@~EO48Ff(x}L{%h~SPRt7##
zsvid(cYK!!<^>)MCHA40mBH)r;V%n09)6grOplPhA13sj7~O?=uS_JX=bUa{6#Ek)
z7DIe%;D9pGG2{Lc%;yQY)JxSniGx2?lIX(Ou@=Yi3@csY36rB0=352m$y@YeQi6o)
zW(1zE98jz!B^(z(Cm@|ht4CLA?_@SmXHV|;*Xc$(pg4%L$dC&=$P$H*7n<ybdKwu$
z9nus=;Mi-`isR>ZZNO(X5CW~dzUA*z-Pwdrf7>WZ8DnQkuAgoPQH{BeSJo{s;do1~
z(|G+B#{Oo#0f88DSCCOU>F8eecd9uGsKIVmCk{^%cJlR5k5%GSNT7>qv(G6-)XM$j
z0N2`OKgIP$Lutu@`n~GJJ=zZFrEvB6Zrc1}dvteAF+L&j&!`CHgPAQmzDltNGY`lx
ze!j4%<j8hzV_aOI@r7ynfAQ#&mq3TI(|gdcA2MR)bsif#d<UYXlcJ&Qq6h)m6y}L3
zV)zgR5{uj_2Dw-{Q-foidHLtr8+8{zMxD@BgZ1H(%pY>4GUodF``(UEc>8+M_%Sgn
z2K@_X9Sq^L2>BDqAY~rzRDF`&K8kM5N#VY+=U4k8JB+Xbxlh6N<+t13(~6z;MRhn!
zpNa^FTgN%}&69HCh%}bC=9Cw@)=Wlq3Ul8JVRScnsB7Digj{kRGw712n<#!Mv35t<
zwK29j3UU=(w>cK>5XjJn9+X2|_=uQU*>Zlk{pbgdF3?v>!*ZzbemK6z=^!gseigYv
zx<}GkqV@eDS<=xal$~Uw)v)cjGR0lxlx)K^Je2p9ZTbhgl5h`4N*wd_%g7#QUuWfn
zr+0^pEwR0aI@2#%TBT@(v!f+zSPodqxaM<2?tcH^5oVwCV**Eqp+j~wl+Y;$6EuH|
zAv(r{V<}jgvc+b8>EYdK`x)tV6`YZHt|I8RiU>P?4_P;{_S@>H%&q+<T07{uw4t_*
zVC^7Fj2Z+|%va0jI?Q`e5E}42i0CV|Rtd~6FtE%9!k5zSRJ{WQx?e*$v0LK`$_89c
zZyZ9U>y=w}PCvShsZh0)EsL$3bZn(2u9Q+M4BO{y>2bE@TEu;K8t8Ga_!U!_G4p#I
zKLz7hx;aqu1oWjT<E0wagJF~dl9J4W-XP82!+CF4igL%%qo{AY0;}N$lziSP^w5!b
z{5pDtMC}rX4NrCDg84hR!gpU&c7I=7bHbEEVlbE)@)n!Gh6Yp~C%8<%Xgf?$L%I|U
z>2#4~9U`nf)#qHk0u61tt`4mk0=<Jvk`)JeQ;N6`QZUzq=KP7%>(V`7N6D4#INz(e
z4$v)mhN2>jJe&%8SNG|*81-<qK#r;uk7&Ok#zl0R^u+dE1i}rcEu6<sx~eaNFw%9S
z=rVZd)bQguF@`+s-Qt4YI0w`w8uE(7*Yh;vNW(2-iWh!k`Wn0fiU`r1_~&eEC`-H>
zj{bM9@3*qT>`{HTsxs_HbhTf>mfbijMkxmR&L`z<DTY(D-n>Yfr56HBvl!bsZQI&>
zr;`1aifKqoZ(ss2DOg>kMTbYc#QMYv7sR;n@F59Gn~s5wG>X$*jhtjN*f)O?9m;eA
zuC^bcH2`X;$rpYf7YDD(Jh5$GJel0M@C(lp!akpq4qIf%IRv%Zw57DnCh>9wkrdz%
ziEaBfK9@YpXoYeLu|e$mHSMJ|&&Ee;H7X=*4kL=+!)+cdiB-nKDp>BGTHa`{u~=$i
zi`zEmJjJnV7thn&o&oIj_cw`&?T(3c-;ud3pEilVVFvjp6-T`!)KP*Aq@YU~C9a|#
zOb^+w(RvSpzI4gDSn(d%zOik^P2%{>WC;X9uYhL#(N({aeetx9r{yl>`T*cS(3pda
z{t^#p?5Q(!{TE#(P#(OU2?%T#mxl;tg?r!yBeL*OYm$J8oagbHvqguTOMhSMROgjX
z^abXZMQG!FtL#gS&)*Dm1_2_;+Y<O`+FM-1v|QU9_AK|uG~eH>jfRg}Y<_*tnquID
zS2lGmU%l!nO<BtE!n9cQ&5Dc&JD`rV<M`!sPnC09N??3X3Ww%dLg0QYAc+c9pmFc-
zluF&1_9gwd5#eG(QtI#83+jchGnR))!{XqRyD7Iyw<Mod2_aOl<T22(aVc;Q&7IQK
z#<TNjuD5GY%B8-qf6JUqeslu*?rJZ2Sy`$WYLH=cOn>9@-ui7wm{UY+_z4u+?+G8G
zFehtG&?{T}Uc2j2y})ZzfdEaj{ev58hM17L0(4ZP-pz^^d&OjpaX_BUU$*3F;`@S!
zAw{46lqAO-7}OvyNYnSJ2?uxax(^OP8ZfQ^^vj5C%=1tlw&$j&+sUygtZ~NOB@kYW
zB%~%?u(O$X3m3+y{SLZ#uc3O!*}3FN>@^ebC5HwtzvJt9@F^M-qY>4#^?FnYyzs`0
zX&~2KsID@9OWrST&y-x#!9aqX{(wpqAVBT>Gpd>STJ=0c)_Z$RsKbw!o^yaZL%Qy$
zL!29uBa^>QW}-u|I9yZAjP#;z%m+DeYso8Ch^+SZ7vZRBd&x{$VOu>U!Ozi4&{p@3
zzh?|O6y7jWV=nLBIp2M85mdAag#0vgh|gj~UQab(;?4_{ZlA+FP~|s*pnbYa++jl3
zf}+M7Z;#bT%m=0kN4CF>=wC<IoNCDKc?^tcZqkMu`YRvP6>OK)-Y{=IAkfBj1$vje
zmbwHWCq=DU_{e{F^wqu^3C1nxJ0diOD|tf&n&nZWsX!oeuk3a&W)H?m>~&5Nzq^?X
zA2W7P%jrC3>|2yE9#4l0TIyoj2@9YV!$sj^LJgYbD?||QU~(MdAhO=!%D<vvU~BBa
zy3b!EL4WVxht!J*A8%C)m==-RaLKUf?JZPK`ihb32g<q*y|-HLzq!G{J#hWR@XM6@
zMt+#(lMzqb`&rQm+%Mi(eH+%`xn{nAmLFWt%~Vf<Fm8Rv1I>Az*(Yv0deAFjqhwLl
zeCts&$s?tV?7n0lh)4pYHW5NerKZVIwOR54C;ruIfLj%C-W)w9!W*n9acM#r#;~K#
z%B%0u)XI6cSO+~S%eztCt6epX-CNvNk_5AKC#|ctEI0DM3y_K92Pc`|*#X4Z(e3$L
z4m%bCuZMEb2U&Ty3U1!R{*)qEZg?l_g~^RJAVwe6wKUys;mpc+Ow&}NL3Mf|A6)JM
zV&YRoc-&SNc25+yTozpJK3QbT{kqIKD;v0u646?th(o_(rZFJ&M0|<S6G~7C`+U>n
z%?xPKONvLP{gTL-(q!P#WCG9zZK8T8BiS`ht?Lhd0+EM#1nexw4xfyo<yjJ^?2X!H
znS><ZF)JHTWaIx#|I3GQkc4=0P*<r+SJ?*U%o2O~lZ`BWd@Va=ItI~Is9P6bG7{j0
za<TB!^df42*Qb;QXx#Vk9=R<#q`0rM8lR~oJ0)BVUXxj7=*>E)pf%}^86J9=JO@$$
z08ba;QjJnLcSNnJjqOZhT7e?hcVH?hCxUFfz2xNmW^lD1*B>Y!K}%y5`b9xcjMAjz
z8~D?y)^6Wh^Q>DzxK`f_614wgtyQ_gl{em)(d;H?JpZ&hiyB#G%g~%so35s9f7%u}
zLch2foZ>>CCEZp)6l>sh>e^!(GLAZ~4=noC^CaEh3Kv*~;<nx&pq?qrrw{jPY?HXM
zJBX;lo9L*bIc39M=Q$1#J5i{9(G=oDMHQ5^=rN8}b}Lb?&VWGau@_iX+8p-}FDoR+
zK{Kt<pB{LfB9;<mjV+t11R$}TT2UmxIhQ=zEte5!l{NJg=64NgZw91Duc-y>=4Bo1
z%oo1@%Oq>)%mfUXo6q+<zW*~P5Ok8v%t@J!wM45*UFh^!G=VUa*VCDwao=fSz#ePi
zNiX5`p?CX4sXQe0L$ASmzFdwV@uhf(Cfc2f<Hs!gaw2T%RVp!&4(-eKHZ$Q!G(TmS
zr#7DAat-$#ZpmKB(+7AZE{w%PF|U57?;g<IEF*L}d6vDmGM(I&ZNQH#BJmCunxvYk
zkM<0OZ<K;I(Hh%q){YSt^nv$)SL)!CZ!n=$hd6R_knjQhv@~uT4uA0a<RkSxMED3%
z9^K2sFTPfwv0A<uS9XKAVuDHpMI%S^bM3x1hZ=bAjTd;0g}L=XlEExCm1}oVZPM@>
zK(FM1nvQ#4{ZAl!v>11h0_vo4mNL|SsXwxb2%?Br7-7;sq{VO}2v-sAw`I6dZS!Px
zgY8feXY2@N2n`sjvU(^{YX&4ukMczkIGD@naX?q}+oe+j=pt91BQeP5*^!Q~aQ6wX
z)6->Im#-A<uLYtF)pU&kxg140bIO_U^bQN=88=HC2z?8ql##MzCee)P9D8ZxM=9>&
zz#l`Hx95us9Rf>o$L%*g_tLDwvu8XOUETEv;=p2yU{2`}_BTUtjn)(JP)e($NQ=RQ
z^3_f{_=}@F78N8tLJaXHByCkb)8ESEM6DlTmHsCbi9R)wPrBXwLB3ITXhC9~9`E`n
z(f(7s;7B6Q+cL%04xDx*n4Yv?^u98VR++Qj7B21o)ppL<Xon;m6$<>&l@6Uysme++
z$~cYZz>!TH^=J<9@Ytk<9TdcN20^`q{P(uUexmGIGKZ_vuXSh(|Eg!9Vv;IUOybOH
zzNGe-SW*&8<j(BiSaaN%<F5KOY<}{WPYHdal+n0e<E1W3Do9`|;-WcZdrxP7#>I2N
zz;om&(jd+01&z=RhdHg-h|F9dvd>icS9L>-w60=msA5~vWl`ot$q}rOH~I7J)m0{Q
z?sjmuq;qtJ&;|@-l}7rv`HJQbDN=-Xbg>KCltiG%2y3gluEJzcFhpt;INNk5`3;lq
z)A=t7;}Sv4=`Jtkn2MtvBXML6cjx?suvk8w2MSiFF%sMfLEf0B#}|V)L`ck=x4k!o
z$g1Mr(gcMX!o$ORsnlrMP55A#tV7*)?mVI0+K*dE7dt5BxP-(ocRI6Iz}99X+@*x=
z37p>B(~->l8r8ooqKZ>~`-r|RdP87DUROwqvRt|DlrCIfrq-Ex1>)ouiG4z!)wLBW
za6?9DuvNr7BWdxoE1m$o_hg52LG;WX(T#!a4!7*;7jYBbJg}cYmbN4$dK&h)-74*#
z)nb4cDR{nQy}fPqwOSD`u9Z?7UMtl|gHEIjJ`!g^a=gdxNKFc3$$DEEpsBHb+p1bw
z%6w1cIoTf)X4IK03gw!Q(lAmi{AcO-_b&S9>v~It)DMOh8kh~wssYODRq(>Dtq>cI
zm<7)z_dJ=Hy%q3C@`+WIKC1aRXXqsLXI|*<<6Zo9AfEnoz0c|4fBwtgzw*b6)xi2Z
z&QYV+BjDgTdRHmla3ilYTn|;@ndiu(T9)x-SBtj>N8mCb>1&SpYy#1i@${Te8QvIv
zXP9I0pBG+lOOJA@vtIxFN4)%(!r<S>&_7<^z>1fTTC)QU1+`I5Y$K6w0A=S5*LwZs
z^JYmW=Y}g~i!_4I^C&%(vjhkWImG6o$i?P)E|~_!fH|$w8FZW1=~+=y*(AI>3kt%e
z52Npn0f(1aCwplZq5c=)b?tBZ{3}oE{_~R!j~P)*x8`z3@&A1Pap`;{w}FnO2rQ%m
zvJMViJPMf}z;?is)sJdpsvxQSG5Q1;uf~DE+ZdSNyp3>9b<NTOEP*&Y_=`Q6o@2lE
zG!-2M^9)zrMM*bs^YJ05O^4lwbvq9q{WEd@xV~2sC<m~YNdfi$j0ku~h{08J_|&zN
zjoAfd?*zq!x@tGsi&n-d)bk@0d~dejG*r%K>^@0^p1&;H6LOUMN;X9KKR-{N8fYOE
zbfZ`s{>gQKrR9JBvx~amtB8P~|9OA^`zQW?{ib0v5m>~UNfU1U?ON``_eFsVp2+~Y
zw3lG^n%3`q&h9r2LIrl38Ai$95eulRL~&kBTL1Zl{;xOx;~G+3z$G5{D4T}-#Zq{O
zhVgF%0a4_0CLUr}(2I4ei?0`th<O5t?`?M6N2>O%QRSd}xF{Q=z>7vKD$v7$#&y=0
zRN>;+<G<Mim!Mf|2?u}^gw0G9{7o_#6qs^UqY`zt94yn`Q<!2`5Xk_V5wUecMb;8~
z@L3Z84Inm9%0E9SA^N?a$iTJZugm+-$@Kg09i_l>!D+bm<ga<`8>L{lU9mzyKZ}z6
z?gKxh+6C~ep)8w~Swp?R>{<qlhoiu?JvQnEoW*0nw?BN+-2uvtFTwHC09P<VXdV}w
zAXhfif@n*CmwFg@L-VtYkHC~oO+Yh(%9f%i9W+ius@<YD>%8`(Q$~R&+_)EZp7%IO
zgnxV#%ugMm^r~T$;1{(vv_1rVSBg<ebY+m+Xf4xr`fMG`m^6rDDM&xFL;?VWA<9-A
z20*}>rWVcF-p<w;jJ3Nlf3J%F)r-|PpcJlGKaj>ygYY%`iv{r45M_+02PRkRbKt&s
zjq+*A>(48gh1itOVV`fBN?4$H1D0=@Vv9y<kH3K?oT!ORs-cw@_4OuKbzhIjM&8L|
zic2tS%RraK9u)ZwMFo+`^#Ab+IOgtGTYIj`Kwp3QN(xqNU##8zAfMPez}PdJMgKLm
z4-_GM<IG07Se34JO&Z6Pz>0Yd4!o+A*@Yz?_sRe~Vw^Sm)4-^}*MgyL`%H&3WZf^C
zvU5~CWc4R)VFL6%75M)*3TA;Z`okr7(_)v}gCs_-J;wVjxNlI+rOzi<V5Tn9Zl+B)
zXF*QG+SROU4*0zhD=FSEFRz2d8N8hrxbNk|C`lWooG(@tBuc=jQIoF6d;EyVOlFqL
z`}9$8(Qe+`c;}*B(5JQ+25O$i99ZlID*zW0x`l#f1Q2qsc;;v=pD2v)7q`)tf=ac%
zbp4Cxwxz~YG+A<KFaL%c{1>BIG9PSCd8FA4e;9@S*ob`U!Cmq^2X;gw;QO!{jHy*3
zJKgW(Y9px!J~0E#pCEr^9gGg-@G}$C4qg4Dob-m60UENY&KNRG`iV4h@-i(Y$I>ZM
zM2fgVev(Mngt6kV<6y<?|LIEVgMlG?HuxSVlg6Jb$~f`K`OGg`j658uM<1|O)5|y>
zq2LIy+OH^3Bs-&fKT!E7p5fQ(O01uEXPtqCi6Z$h*yhXU*pRj7sN{0+*L5*0LKN^o
z55-NW*>)-{wej50Io@r8On^CI;`&C%^Jp05e)uK`XbkGA_pLUVp^H0U&hqTP#~d7>
z_1JAV{y0*7*M`#Zt*OZVZm7WpLl$W=Afi}-aYo?0h;HMYXj$^G(-oLw>&8$z0qJr;
z=Nviq{)O~D27fmN#4h|3K(%4nFGptOww+hxXHK8+`Vizt%B;Yg4P)n#yK@d0RLr8w
z@y3)BFZNOR22^6CRX`&|x*ine6~<t@^_bX$sExV$+I9_c`WRRwir1?5BlJ5lBLYDB
z8L#lYZp7JRPw<8lTfis$);;ihJIcudWHm|qCO8*aR{8)95v3^`;cC@{IS($%1m%Y5
z2Ak&Nl_$M`)F_5(shn8m4HQ!dpcD9B9Cn|V)PW}tq<qSNMfSE}i~DzT7I|BsZ5Ra{
zj*-!3$8J>q=c0A_$#e`O?HUjbSfKnj3!`3|wR>SU{*M`xKMAg5m^c13EM;L_p8^gT
z3y{;9PmS}Zz<3tV?Rg@0&_#S~9mfB&2z88-bc<np`Fs?$oNnHA+K2u+QONXVElRK_
z(@~tN_d?tN5)bhI8TR$x)oF3IY%Lw7yM`S*fkN2$C2|J3wF?5|>k-zY*5kH<$E9RZ
zs)bgFPuU4aRRpoYnRUQ7+;;iKqy7TeS>7_8vshH;mGm5Trko9BKX>=q%uaHsOQ%&z
z;APsRMyjuiAon32vzR<vd?u5e&#t3;gGRjpLG|-)AWxw`x;I}|IRf=OR6TEico{13
z_m*}fP9NUQs0DbW|HIf<fJK?^Z%Zgc$si)>z<|;zE#0Vew+PY=f}}JI(n^CM-6<k1
zsidTUbc1wve9!FJ-Q#|H_WaLv&9&F;FpSLmKJ~kQcL3zdAiUlpCdVc<?DY}uI{y)*
zlaOY#7^(1E2Syx|w|g;1Wvy<gxb1aI#dBLH>H+JJvT)v4prXffZ{@byPymZ&)fLmH
z35Uamqk7iOJjsY!d&4BJ`U<7=wJiPFn1u8BGER(5T%^|c{P`7FTDg$xZ{o|Z_1NnF
z7JmN6W5gub?@RghA8P+T&jOiIZfG6>B#gMqYkYj_4r&8W-7M}QDC*g7gb`^P_&^CL
z1!}-jdShW84D2{hunpnB;my@Afnhbf=4FuZIow|`xmUGB_DFMoR7yhnu9GU{Hu1U8
zdTEQ{yEQ5|an2*KtxJ$XSJ)62!=D<Q8V~jYU+p!C2$MsZp-@7qL`qzQ$3skrAMmx{
zbqC%BR%LB>J{;983*uv-?E&`1XFqBc7*%2(7S17cU9ym;TQnNl8RVf18dVhlf_SsL
zCjdBEAfI}_vgxk<)Faw=x1D@tmExc($|LI9=2_4O3*E&ZhqasB1~MPHsxW<kM&<?5
z6SZm&^+?{)x+js1$VIGyD-d=%_F~t|YhMKh5yfYw@}I3iihjh`19KgJ7{}rIzzj+J
z-GX3lW)R2x0q@8KU{@3Bh%cz<Wrk5fG7z+VL$v#cY3k&?SV@@>nKvFBiw_PUQIkNI
zK(Y~K2_F^94QLEFJ`QU~gHIrL3MmISIdN*#rzhvj!nW0+P*Q4%=A?PBVHpzXyF4Su
zq8-DMl@~f$fGXUXUon1HHRFB!M&Q9a=8awCDaqY`ltgC_>^gkK7pEhir=W;%G9H$A
zH9Z29SyLcEDvcWILb9huFLvWT<D{{QJy~Tv%SZj`?S5w=64rTJl*(EdJCGi&vyx%o
z(h@=71Dz`+r;9_5J?H&?>18?QIIX%bg0ld(MVw(!8H4ukb)Q=&66>z3c{!EOl^R4o
zQDwXuGdmgMSbKSw3E3!sUZ<vO3mjOgwr#{X^L>x2MB@8w7)g{1++cLmD9U)J^3tq#
z6wI#!l-jVF*fXj*?p3FPN>P%?rQM-Mq$l!5&04XNga@0q$Q{v(IAC%GXXM_*6{t0M
ziGQ`@f&O$LsJ_wJjspH!?Px(iCCrQm@uk|<;6pVLm4hFK*u#M)&+D~F%~D|F8&jfI
z?@v>11v~4&H0Dzq2F?%~`mBd0+p}i>(mhB@nMO3%VW2ys)9jO{r^?c(W(F_ErOxs+
z1R~&tm$P_Y#Eac~_bT^}=OZGfqcOx}H?wq?KbCk}kFsqh{2wz+@78v>`k$R0AAI>4
zFe>CW&6!UOJRbt1o58mVLJ3G&5zP^AEWmvn6nM>BaCu}|U@|OU$Rc%!e%vDpzjyTJ
z?IHNR1>bbu-Ztru0L%`I%(w3Ck69cZZSGh$p0J!gVR`-Jk{EafL{I2wfeTmr2G5t4
z;(d6P$9v(E8JmY-&%gkX>3$O4+>Yqpf>_D|YVon7sCZ^qoA=2B>o_6JRNZrAgUG{e
z{KgX^BMA@S^`^e-a{*idI$dTK^=;m`E6cJl$q31$&ctigKp#}cB`p~GBdrWnzpyh}
z0Xq|iCL@O^WorJRyQ}TAz|dlWDl(VIH7A600Q5nky|`mp$PtdZ$8i%z`VM$@_An>1
zVU~Lcw*zF4$zI+*YkJoXMSKW_sNfz^oMZ~p?3_*mMB>)G%m%omy2K=-zHmwHd!}kO
zrVVh|?Wl)I!OX~p7{`*Zhu?&_4ty0lEL#w3@X&U^f2=!wowKQ_zFSi8MVZDtvF6~n
zyge3+M=TS;PDB}8;m@*0IB!$Kd#bIgb9t+0xT=rp7n;M{lduD*n8Dk?X7iO`BP(37
zJ0rxrS^b&i#9Ko@T4T`^y`ejY;rrD@`JG*F<%v|UvmTY-Sq$!JHHi`?;t67QWR1v+
z+3==47`?9D5g=i2Js^_w($=Cw@TX+LGqM7}H>LU7!R44xn#q)_@XjAt|J6oG5wQn)
zzI>)U#KXJ2?3=wXx0zczA#R5E4t?xZj`_-`Fy}WZB=)j*nasCNE*D`c8CER7D65L1
zdWA&~c!zYfH9b4i<xJKsOw_$M99{o#tlIrT7TH!I*ofav3-1@o#bc2PIh_$3&t8l5
zs#0Vhh}h_Cs!x+jnuWOom8fL>{mnEiV6taFVd*L~r|;|8XtP{bb<DNf>>2D4BVN)g
z^~0uCdU%0CzSS`f`&!Es(z&KXxDw&s5zlVqOMjTOYvTTj#SSl_tXFcoDa!p`ugfN1
z$+MDnj9&K&olW7*BKB&<56NMj(~}iSR4Y09&ptofqA-`xFTKCOhrw&~X>$I#1OD?N
z#}vlDbjJQ(J=V}R5{>+O!ciw*1O6xsZ;>1k|Ef&#i4Z-U<_8Gf`{?`hcH;j2V8@2}
zRkqFO`xwsr5G=WWWJBclqwS(_S=0;t{^kLTeUG7?f*1VT?74c{w1;Y2h^DXTrMh9t
zE+3+t7p6pd!h&&JGxB}`*3nrz_sP8;6D3OFZqh{3ptsS)o}5F834GlSPv~xrrHBNz
zfuDjN3F`M`)g$ld2?Awb+|Wrf?R7C-|7&xMPfs2TRg`s$iTKAso<nhNqord=<ZtGc
zR8^PhYq(N6Q};l$yPC-cUi@agFs95$*(w<=b%{Sf81RbxiTfVCg*e1_6pCWb*VtZ9
zlbxk&psiNIo3eI16xkv8DMlf}^CCNqoW7Z$HSD}8pl)#&9s%)~NYXv2Xz)7YjBg^f
z#~Y#B>pl#B=rfbcgm^A&Px(a>ig2kSm*GJrZ|#D-$VYWTRbm6*mzVNhwg7&KdcZ4d
zIkjY~HL01mZI*~ue7cx5?w#t9iWk5p&)+-5Ve{ULw}CZdX0$&^JB?ONf}waxeh@d{
z&yAy`UsmkEnr^eqEj_hb%GK%zoPD&}Z7K<5(3~jBJZd9TG0UNm?(AZ$V=3-#8=9(P
zo1`0u_Mg2cfxs1-K3x88Mwhr|WMBr|#`U8gkbeGV-J=&H1|1(DKOK7W<eKgMKSlG1
zeKu@A!0KmbqK|WUKj|>ts1KI~f7758Ah#mUCPJ(n@Pa$)$7oW5AQ0K=2<lB|g|{E}
z9hGQ}GmQQ|uKxYXdkQ2xA+E6eJ3P_IM%j3c4V5C*q=-Ftj6uug#g4fE#`+h>IE6<<
zhM6e**kn*UIC0)?c?`xyukh75idZvw+wso4cy6>8Q5?_%5<k`7QimW6@O$|o#U5Bg
zkE5Wl&)e73Z$c`zSDgX$5EPkSCwwIp>D><5A@e)`U?IZvb8t}V5c>2aI-?^e^=5YU
zw{n6M2qm;U78pk;-!b;F5W4dWWs0FI?c1)%aXHd!VO3g9q|ci!lw18Kz!|I{6x)U0
zStu<e$Ok&d?@d*hdAUFeupqdzuj4<FuRa!M*Z22!!Ebw_#mLe&zosoMBQc(5;n!wK
z34(J+KxLskLGH`Yxe8n}$SA3pV2GI;5&ENk`87z7HOX^is)7qQBZZ&K9_YGTo5!7%
zakagO9jV1Xeje2id~o{JJ;Wy^szN^v%8TWg=Y9ZdSaF&_woR-S1hGx7q!SZal$@Go
zwmfPhVd0Y6<?EJe!6TuwxJKl)Z0ZZnu?+SCPQLD_`i#Mjxh?iz?-Fs7lmu>|!ccb%
zapoOXz^H&z>YsL=Ec)TpwcN43vDoEVx-Q8z!szd~A>c3&i*!WTexdzM`umprINkEF
zR>`)>6VJxIr-zXta?7uL<J!|lKQc^?(fn6%tuLjKeEsmI=`XwY<3^0?9*o4E`ljE^
zxVVW}Dmjj}jW(UK?33|?lCUeRkb;<__Tf?ek8`lyZo!%ag|#~)jIce>T}e>*VZK@R
zWkN;6Uc+cIM-_zm3XORv)|D}r>?cK#D*!<umSb9cwRh8mrJ?JQ_p|^hvMIYX2c!PH
zxzVlr++<U*TQ}2~#fl^KTkleWDa=txeeeW`Un-IO#7f}Xc0DBfT^7ZrefI>l{<d9&
zQC9tjFA7z)kgnX9Lf<p1N1Kc5549cpqfVlWm-GX>+wPa?f=IG%!JGu_u*3&jDH+}M
z&wc;~tu@7IUr^%VA_a@hM-cRo=+4rWWT<88TJ^rg<FWjnxo1+b()lbbDP15!V7GnF
zgxYq<Hua9Z$UF0geO^^O_Z&+eEv4jl*Pn51WZ$eM?wUwa&TMJ33cJYc$w}r;eaJgj
zhCM@irkTfx+a_a9BotG2vY&YOW6@moMWKrHgi*r9eva@*DnEy0a&c16(>WQVsjb1B
zBy1&L$NoD@7j273cy}{zNcZ0J&7O)X#?_A@uX=LMxQPq`i|p|HmsF|}L!^3uB}39|
zdRzA1Zp14qx$q2&pT4Bypjz$8!7dt`tk`e|u94NeUAhx@xl7DVe+Ideen8q!jU|M*
zJa$j=at2W^<7SRnCg&wJTFOT^d1tOU+M3%>Gk^M-Pd@zpTea|s>gEzvMQQ&8HfF^q
zFl`2O>#SLB-ZgLxG4Jd}7rOqbnwQVYAG@|k@^JW({%iY#a`}4aGQ0DS7JjX3h60Xh
zMicWPU%&5Iae^lK%(CxHHq&+?Fd%Y{Lp$^~#Ueue!trD(Wi7^?iF3}9NEc!A^thR*
zTQG~%x>OIaxNb@70eVxAJ6CVgdDKlJ-Y*>Yxd-cWms%>2l%IcotoQ&7Pf%o#{`Z=Y
zEg3QaLU5O0exW`lsShjGG6na#IpDgYY|B$)Hw3-FDFK)LnX3Um-suqoI$&=W-I^)_
zbSGe~a(##NWH(QE{*(I)j-whQp|<ksoF=Hjosu=!xK+?y=Tcz3#O*KI=DFx}<6?4N
z6A}F-j7(!n@X5Ip;zcXIw7(`g3jSyOOL*~%2t+0D0!&g!0FD|j5SsWQ(QVilK&`+O
z@yg#DB9q6zd0^J(;AAV)-;nFBX2^IXMkG7w_{O@tgC^`a{NXCy*+5*wbD@nFh+b0u
z%UWeqQiln|I+Vh3l7s|`N+nYRJ{_Z;$$SFLsh*s1EAkLZ0(U=66E@KSWZQCX3Tc~$
z-WMiwSu6I=3C~nY@ZzAv%X(9fNkTH3$MBi|n`kX9&gA)@--j3W3SuWA9_8)r^|Xl!
zY1)K(M&mpUJGkROoG*Smn6gf9g%j7Qvfue<0{+>mNkV}4gk!#ys9MUZ+498MM2p?1
z@um|JF7aUU771~`v+<=K?Ijbsg!gBFff74e@enOKY%zr-7wisfaSKJpBRzL3$DicR
z|HzexQ*w6+MpyPceg)JTZ8;P+?p8V0(K4&K8jG!w6hc@tS_so@q<h4w)(P*$oqrw$
zV1TTIgMl!)GtmFXKv;(e0OLqFTl3pd?s2O@r_*hq=|{yLMG}l!lDZ9^4dp7$b}Z>k
zpQ;b0U0|V{^+>B2VhQgyy5zu6uyaV$nT}_gpb;|XQpEl%SS3<`*x>lqZlei<0w=;5
z;^pzU$xxMf1ix)o&Pp9eYhmw0>IL}PLSa^H;vNBtuWJ?tc><x=q9{|)x?oeM_^88L
zFfasmWDfvH7u1(F=#c$NE%>X4cH1_0Q2QdVYJ0u14O>x>3;ZfsCq-DVy}0&0A&`8z
z`!?!a8eSQ;VXZm}9M7pc=h{s?*Aw2<r2!-q4$VDPopN<t+yz}MtTJj?UndM@57Xsl
z3%3FFvG_#L4580G*jB@F8*89OlZVImavdENLg}gDutwUDk!`vVnF75=NBZ?`^q|K4
z-*0p>OrUNj4s9bhpLy}ZiFwWBM+meu+&SFt&dvl#fNwrywnk4v6|YkMnujr(SWYxd
z)paiGZ7ofy0HPPp_i%MOJzaAJKf{|ncROgULAq?)bEi7}uOF1&iHWSs{1~_5;{dEU
z1q3Jlh{hzw>#r}zr1)~e-PG#1o$-~}og+-|uU;2>p2CQ+H=rFV{To&I4@my+xDO%v
zIkwd=o0@Fnb(@ZQFzJsbJ}m221r;p}zRm`aBZc++wz<+<JQms`9IlxvX$nuylME?`
z2ye8*MO>DK;Jv-AIY3ptJ>Rvq=}9*rdHxPPwUfL0ea2$iCO-YikY`lH6W02QpS;H>
z6-6Bwm2a_W5)Ti~TH1Iza`*}br(O+y-#iB`586?cr@0wO$_iOftqa4+KBA*u?*36<
z?8LmH(Np`joHwDo9u_lmOW6l4H~E>W<IBy-gjbm&0Pf3O5A0AW?84>(ksXDOHqth^
zvRA<Tn)a|7qZZ|?W>Tbn_ovxg?3p(TJmfiN`)@iVA9l16GG`QU;GR&i+<PR{W^>8c
z61%CzULDVQayy>61d}!u#<Ln8a-0~R;b~ORx$)^I<&`KoPy`t7=Dm6w;*;b&2*Z20
z@*arqvo}sKzSa`6omAv@<Y0bFN{YQ0P0U6-7DoC&F6c`=(JAmc9V%(^^{B*VxqMtu
zRH^OItJjeeXZzA@eeJg;BAG7LSE!n;a%Xuiz0Oi5&gsL8GTrDfcl34M$&1Q8q_uZP
zJs!=`&E%?)k61$`3`sfyLc+j>X<&4nYRQbFbt`>PG?fRrw`}@eRBcrem&OgwC%;M!
z)7_7sP}%cd`@Af;^$2-n8+2aBbq-z*0Vn6qoageJP8C%J=I_pYyoRDCo>&eZstZ4~
z&PblNt!K!d1r(=8Ecx;*1M-_%50zNY=Ytngs0XFUtfO^j^aOPOj}fMKs}U{a7nv4K
z;|#lB&!#o=$qP?w$L9A13d}qrg+YqgY0CW1a8Z4bhJNgQ>r)dqbZLLvV+yG$&~7b~
z)exe-I&Pu3X;1pooDdqzW)KmXLoNi`x<Qf6IaJJB7N=qHH)hposUnaBOMeHPl-iMG
zVqV+VR`Hkd02#CsTgNCtvKh1XLFSDtWDM#Izm|&ymXzR<;@uT#2*Sl^r;SKqvyJrv
zSRUthniwT1)FuFfA)hZ>`Gl#jxvP1KpjTA6y5?Ci10Y!{VZu-C`)pa!CMWb~H!yK?
z6V=uJ##6ZpmM1)=O%x1rHj&n^uOxCIO1T+Ds2k+{w|qZ4*hQE>a~uy2q)$TX=~}%P
z_f95Dit=8+wf=1nL3ro`<E)}|+07ni2{k5u%*o8OV(!Gj02gfcmQ?p598r@gp6IR`
ztwRvo_vsyVv#(CA!xCRh4$h_~MAL(&@$#-iJG{TxdCGM;R&4BHC+0J9=K(=9nlZzI
z85l5q%$D0cuxW%}2$6lKM_(YZM#4)V8cqQtb1>&SZrZ6~2y^U2*OK>z8d;5b#>H@f
zhv#~hdu`Y<iT{q4L73c<7A}9q<>^{)DR<IfR@Q%O(04yZ5*%V4tGI$t1hB!xTVnzg
zhaecmKtSW2{TC9s6lZAiEIu}#kV~{Ev5)CzLj%fWQM=Sl_LCRZ9XUf*>r&8{pr?=t
zY^-TdZu|TMZCP@*+)E??ErEbQpp7Rle8Tqju)dN5f8CCpSGXnIYGSpo_IySdZWM-i
zxs=285Ph%5A1SDbL?70S)U&`T8QgD)+r0n{p#X1)I<Q<F@}_I&Z|0Fu27bNHglMo*
zXeed7gzOHmZ?3JXb?^iY2s4Nd6b*;wL1E}~8moPJ<rYV(scSpOaTphl${rTu?L=og
zSen0CF%JS3DB0KJp%FBJ>+sF@vPaWiSJ2z#@_H<Rue$tpsit~(rUiBz-LzE2>J9j(
z9+;a7wb`>{&WiTr^y9z@r_#0~)eiHe^u(b7v>dAjwT%Zr7k{=e_&Df@ZWZ0I-`QvF
zd(@3p4%bDz2m_aoouBv0)y6I28f*@mOo7M;(<eoNn#8w;I-Hlv|DFq7W_{z0w0__c
z%1>!wg(oId$50G<pXLDRmNPyDW_CF{^Ivn2FgJR>+DS(`>nfL11(Q@Oc7nnnNo8w}
z9!2C0ZX^JVW~-gxu1^&?0qFTRta4bEMRXR90W*nmj0n|e6AoJJ4sUrw4*l->aio0N
z_Sd{wpAG*yXiL_!Q`(A6eMpwnOEHDgWnJXrad*wBT!J{ItL%Gy$i?Dm>|VbuKP{_(
z^i0{e9An#2E30q?o+LBViBF|Y6KIc~{Fv%*a0wJ)lOees;GIdpJ_B52Z^LFhI%leS
zW9QpXewub%1UZKiLnGphQSQR<Kx;X0dsI3JF1Og%`=Q?!^iiv9DP<S+)p52bK{q!s
zC1&OZYwG2vpF*+LxakrC4pat9uY!pLQycg;fD^G2jllS`5to|`K|pHv`2edoZ^CQp
zu<}#e*Lw@WyYx3wa;h*KgwOhdgl-=cRXzPb*1dQ5@qR6JEn5Hf%R?vFMtQ!Mr_z(t
z%t2AeLj=^55gbG{@z~1U^+oZ3FN^)WyUk_E@Sfw2no78*$h17);_bbk4|)XK;_(iX
z@R+`yu=ukRq5jtEXDJSL+3OH$wBBrL^LRhvu`OJ@k&f?ZA6m^*$lD#jwS-o?G!+<n
z0J?-pA)mT)czgO=%cni6q)CuI?n!gqK<4kb_fvxG%!5JAQ8f0X;$4_^VO1OpwSrJQ
z?CaT3O65v=h}5NSp)rg91ER!ucUhz-S*Pyk=nElXgUyuAZj6SWHsz)Pj7I7!7FBj4
zB6I~J#{@hNvbwSBM^dIN>K5b=kjzZD$o!F=zWt8*5H^uI>7oh_ODo$^v?<dhwMuFG
z*)v{~Zequ#km~nT;d^;#8(?8&KU%sIq@)2@wPucRr6K!H7eEtWJF1A!2sRd86irGe
zH%A-79193}ygsC&V68i!`BuIhkdUWyDOl*(i3Z}aAEbjIzlsPz#pWA$LiRpv>ETaQ
zh$CEfp(Vh0DTS{mApK;hGt2#PsI%;(z3qEs+cyc@i7}#w>`bl&6osZNv9sdohyg*r
zFI-&*ag~y1FUy`ZyeN7?H0*C}A7Q>EHdg96DmcdM<o^&T490%|ndM#UPpfk*V(die
zfjyZ5rB0)at5l1#sA9-Q@&?jTz=7ebPkq~F^~%UAV9@==snk-Jf`2)EskAyeBjf}1
zr33lJ7-{EVeMMr(*TnQ}U60qhrSolfH=Q_kTey#$%~&|GYj5xz(V1-Gv#mZnym^NO
zE2J)p;z+_|+vF+x{@Xx9fs2ZMLjfQur$ZUvFsu~au-rM<A6_%;_^3BzT)`n$;7)WV
zv>O!>kKL?w7kX;FJgS+hKCUZk$zm2^irE>ozfC@qqT7>`_X$ui*sSh;CES<<<&Zgx
zFV`4xFFFf)1<;BWM#+!e=%uRQ9}B@&oY!vSCB-<ZZmypNt*6kM5YGkeQPVM=qnHV2
znG!E_{IjmO^~E3U!j)`9e1Xi~BuDqHFc+m}U;y7EtTVSARkNZ!jILP|d+dnkx}?9~
zEj&~TRG9ZRgIK1xj<Cpj-71t)Mkc7Oq7&TTg)P4|gsoa+TnsU$@aq-A1YAt0rUkB&
z?V2&8R5$ko;&wd(>h0ML$i>~i++_R@tcG7`JTK${Y3G~PPnq9qW8z_SVyIM~PgMYl
zQ7aS91Qo=-M1#G3RRs}986Z0BU;rg8qeL482HXnb?Y;(>qIl^%tpTkXIEV~nVs4fC
zRJfuI7z!LmtporSmWB91n-*taxum|ubz#QBCkD_HkVC+Y9^e<?K)ScHKp}*<fj<pL
zrPQx%@<s8uv54;vuPDSy$Wo9q1OCW(q=t!1eJ@Ic4ndDoSONoQ4(|@s72a%L639*$
z&f8PZj_W*DIvSxE;uc_L(MG6Y^_U=eG(yt|>AbjVdUHrG)s4>|&sC?o8!;lN8pwN}
zLj3L|*xYw}H4jx_LUJW?2@kU9g<9`GSRF)M0yO>BpwKUrEA}N?Cy|;|&LZbNlu$Ts
zv#39XkeVUFNJH~k05qGbk^$Q(iE5Xkg5RGI$(d71g~~bKI(>v@*oM^VkD1k4D#-)B
z{}Y7A*xiw}22jtv54MXCK5`~vD&;$ex0A~fCumh^u<hmbggIosqaG9GlqR8`Va7r@
zCRR@Yt22wu@2u<hozVGbVS8c%k|Z@m9l3+XJG$W+Pj&p>i9eUD;m2%qTZ1)s*IRLq
z-9)uTe<Gx3idZh?+lF5)pfyob<vXm<;<i-*Z-q9Ob_{1fZ{86>>N_v-I^6V_9woa^
ze*|p7n~CzPcZ~-a7HTF#iqdRqkLpT$5V|K}F5%19&@Y(Eo@$k9R=z~6nI(<dp*EZQ
zKjou@E=;|i>`fz=p3?vNh8t65jL}zWjJ`6#IlR5Q{%%lT-FHE*ksh0ruy%f&@;)ek
z#QRj!Yp6%e=(eMeCE|WeQEzt73AJ`7#|3z^NmI(aq9d520VxhvcPZ{hZ}I6C6mvh;
ztsiFe9|O&OY$Z<AL&78pN(!DS7o%teA@91ulsvM3H#`1b$AMSg5rF=^qxwVT-~0Dc
z=qNXWNbZ@GzC$snE17Y^=zW0m5DX(?&^vE{h=behV#HSY7%J|8h^kV*Nv?4R=N1Lp
z1RvD&`+KLU=_f+OvP*UHVrry(5dt3~gas&@@M+EiDSm!I)1md*GkCjq^U00#lMGqJ
zhqoe$+bI&-yd~>CMQ?dchS?!hP~S?1tvU*JG!KXOOwYbPz;KCKBV5!I2;@Z+FzhSC
zhO|0Mi0M8X>ZEVJ9VUme^-blA6=$?2%-^qX?I<U<{eIGBT|H@kbNMByn;hD5OjqV=
zqS~UWlWxg4puQNAXO;|d$MnvV`};><-w4^8eMDxf7O`7Rj(5b=nfA&!DPi7BBL#?U
ztWq5k$Cx`5Y>XDmGaHm`gCn>Tx)Sz{&c@4DxNz0XH?qQ(TuM1075J?sm17pON<~(z
zyJb!+rZ>9Eaq~Ozes&)1+E@ek#bl+*Len#Jg%o-^+C%*Ob-hZt_2m?`!{nq5p)*W1
z`|~TogY%rCMQ8c3YM4`!Tklg^XTQA%y%XQY1<43WB10y=X!f44j(2(pclBD6>)b{?
z9on&4eK_<($~{kG2)?U}BkA6Hn9r#^{g6J%u<jQ^^50<_pC9i4So6UhSMl=S!a=~P
zG(g8U1KvKOT8hd+-q&E`@eX&=Zq4P*y{Hhk<CCOE<?t7|3v$ZdDC$Z4S9+#_CX&Fe
zdE4(6{I95ZA0}X>@~3g0-EC1V9Jl@K31Y{F3O4G9zrA`9Js<TLNu1Vpc>Igz;_u6F
zH#w3?OTJl?|JUk4BnUlT+W~NPDuF6+I0eL;#pk~Of`rfyaoQI2U{&e5Axk#-Z0MoY
z9r!k|5&QvyjIBWZQUNL{lRF(T<?OY$sz8p*I-p5~(kG_6E|KT}W>d>1yV@pb!dS15
zmVIC(UDWxn-~2r;_=4k&{VadQOnnj*ynpJ0ph{E_uWBKC@%S~g6?zA#g7^VMW(^#q
zL#I!9+;6=PuRGuE;s!i}e3tGMHG|mB17!SH;fHVK8v3qTQ6*M;NIq8!NB*A$@h^MQ
zua_zsQwE|*lm44P_80V~CA<;9D9=Hpl`uvo5HI}3FZOD5wnienK2eXavF5Ti92<%s
z>vn$y%on05l#ua?KcY<j<EjB(1ov9;;D2NV|Ltv3Y)GE6-7NjjfBN|U>vj11)ql7M
zzh034_&n4A45jdyW|e`zOoPAuk^k=L{r^7_6Z!u4>Gi+80=zLnW?WN1PUe4Sg#X9C
z{I?hVz4Ot2y*eDBjX*cgob?<T42M&j&BAF(0HBK+L10M)^lo&>d=S%(Zibi0Ut+_7
z;9%79lMk@rycT=(Ul4w%4-L3V+_Qanzl>?0w0mrB`({XH-v&^k!8*I>HsSZ|1|HG=
z0D<3$#ifAQH#S;klvLT1O9AF%B}mk10&u-41^`4tYYGKHz?!|De^HwqKpK94*{C;>
z2@FVlv?3cpm1_m!X|<?Le@h7bDQ9bO4wGke8sA7Tei<SgBu@JJXa6rsMdYPuyfWI7
zoheQn7QXa<{@PAWA}d-IVmwS)(|R8s%t=}s_?+QB$ET34p7^xLH?EyMqlgE$ZRlF$
zie207jAhrZ&Qm@+t%Fi`*|Yb@+J@)!X;Ck&d#Yez{qRdc%VW*YN3Js7=>LAaXMnzk
zmNHbJc8Q?g``059UTQZhmcV11vcLX=!!eAO5f25DNr|poY4roE=yt0X#@!c@LnTi^
zNbyVK^HD%3r0*5jlg+jOjovyOQcfA9S6cVSa(S^z40M*3K-{+>EHyH~c|m_T$8Yj!
zg;`WEXw1!GbnSso-y6VC*JwDN!v-8oCtuGXl9*HxnOcS`jgy7^#|B?QNgjZ^4Q{%E
zEO{7vn$2X5+AV2>J->EYC+W)qZIbjw`lV#cNkg+8ty`2R|M2iJgeXEZahQdk{P7hP
zXVGuq!I&JEzs39<%+WahbnHW@@Ln?H&MxTkbs!n#%|e%e!`lhic61<1kiezb=&|$+
z7>u!7pRa`8K1$-Ye%|PfLWAhgWwN_fBWP29pic7eMeaq*aasBR7_$r@f2f-2X~Tq&
zD<HCWA@Hf6{~XA=jar;5h47%Q(331*qQ|^<>lj3PECR>&Xkfzol<m+1C`FV@^y^zq
z8P5#I?D_RC!Pm@K%*8TB#OB017CVKcwV6Kp_5S_!`D?fc+<?8}m<coT+HVh$0D_gt
z+D~<dW0t$8YbC#b(-{ze6oHuocKHIr!yUk2R@fFqaMirrYG`h#&G#?^5K7dmi_;h#
z2yu;I!^NQv!2&X<t{k*d)kqHki%R|y(CL&Rnva;FW5o=4q~}&#_mHGlnN%Uq{0b;r
z_{ULu{>xWt#4iiPNal1q&yp-Bci+r8{yxI~9BRKjX}bl+p{Ain${&OCy%?sQV};mG
z@tXt{O4YI+{lfRpjji#pP#A~5b#{6he=I0p?Kz84Fi^XVsxEY}s)t(aRPgewZ2uqs
z5ycoxFb-mLf{dyEHB@}4FzY~0w(oYd?dOZi0R1ABoN!K?3C$X(H5xLjQXjBy@RWo8
z*mBPSAUUQYGmalv4l|tqi}M9w-5v$>6k-5hH2wL#AP!6j;hjm)4NzU`O|?BjcHXoA
zX|;&Fq~SE5NxMRZl+%n7Y~3N`UgtaN;Vv}C!=MGy%0~+dsR)o8N$UVOgI_2<Oat;{
zE~o93z)s3%O|66N>7%dU=qnBwl6<bf>OHg<+(uTpB<>hw=DA`7nQfgCFm{0WL^9CF
zXh)don_E*QYS1r3M)*MZqovN(B{1NI=cK->DFk$&OXWHCBE6>UJFNOlPdyJ}biqfW
zj$3`tSNrw}DFtcD_Rw1N@sxk}@VfwZ^M_eH;d4qLWqAXb$5L*G>-0#tUw+D*Rf2>Z
z1eFX6D+ZEsS+u`8+n`0tOAgiwN*c6Hz9rLtwD|r8G$pzP_5o00ap=kx55i}&al-+Q
zu{P3RRr&$gCW(Sy<EiyO0?C_JUbWput^wZgaU+1q`Eny@_6tDUg~N@0`5Zv*QD7r)
zE(W*g@D9T%ppVCbRal)(AKn{?ONJ$;N%87<wtS;<d}$$SODqr2iNvJO07ZkJ`>?6t
z8IZ+xyBp`ozia(u)UXv;bvdXYY;1)+&n}kYE50^dl;w4GX|lIM9j~&RrMz19yyI5&
z$x>3!6S{^VA*>`XKqu&CWI83UbDbOC82_Yj{z9g<8H1q!w$qvQPWW-2bOjFFiYoCb
z{c^l8G6~jkr8xkb8OL^G?4xyI+r*D*M^iHFdDo!Q0<)C%Ex;hJI5`JdbtP7S%3Yb=
z<njS@Ai7>dQ3EycB-T`??D<dS>E~?_d`#9P=V$%p(dFs~CkEE%Kp4@9)RxcRUcY^D
z-Ez`-W30m)1?|>|Q`$%kox<3w&9zqQKj!?uX8y1LZR`~H7Hf>jVb&t}ZHe?zOKz;P
zpI-o)|7fsQX;WHR{3y=~$ki=1@Hq$Li^jnhvI<rvg?k4}Nw$xt8$9KBuiFzy$U&u^
z4U&;<BLyo=`^!EUu=u^p?b(*=WUZr2LEeyMppZ+zguq24V0%kF9QHjxd`;)HHr&&R
zW(-s;A7l~x;8S&~lejEn4??d{W(NTedBJOmobdy!zZuzMb@)PJ8sB(*P8fngSuA=&
zDJy<KIRFGoh4<LyRPdgLS8&hnJ#zGiILtKqR6Yv#piI%Jd6D;{6)=<a*w!kXoj?p|
zLKW5_p6xRz`~6@ocwM4^tAl9NAlRFjXp`-H1pwrS3ZDlk6*JTs@Nshv!*F>f`o_rM
z4)6i&7^zU=XLGvy%==Ferw0562wQPBe)mJG1P;@1D}ZUx)V0-v^jfXl$I)m<l+72H
zXIs`DE+D$|Irw4MZG2y91eKz61>X8TxLdswtZSKdzHcHoeAut~%A`_mwv4hlEXa6)
z*_lR--UH#knL>y6r2|;DrVg9TyxzgOAedhQ@}~QMt5lfmx|Cr2T4JZnFu)sHja1Ro
z`06)!R0pfFbEjLaTtSmwtdA-I;>wr=ij(~HCw1OiqR(GPuG103zbA)EHc!<d&H)YS
z<8G_)+iV!LNIV`3j(e&)UTGbMfxZIu$mpT0Y{GX$7i0aZLRXu=?k2GoV{$12E^^?%
zE^_PfaGE&A$!6`ARbjvkX(gmq;z|>2(pI`H0{9B>_0HwsYP1V9uc2$T{sx4hKp>(=
zrqUn8uCDsrYzDf9@|41hX{%}@^=nzQJx?onY$eX+MTI3+D&7)F;Ut@sQL?X|!KHsZ
zxuqSXnq7Ym&pw@gL7PWHI3hM|`5%{X5mZKelnV-k?4R%Wp6qQC%};p6H|g<$LDiwg
zenaSVg*pdSAct(uW`Ct$UeZ%u@a+EZK9E1W1lenERnnpcd*pJ+Hsr6n9^xIT)-4aY
z$EvU=A(fpDyNdm(gShfJPwsu%`9j8j5IH{-5}{%rG}z;jqpvujrlGfJF<kf)z|qmk
zM@p{_0*#FYk7xYe*%XfPb;7QW!0NPAx5cXJEC^zV93#sf&Um9u00?nLRd#-;EA1oZ
zsw-dv@)B|qvmP|AJnbSM97ZF_`WP(xOgSaDv^&p$YB0RU$L2u2Ogn)u=Od$?|HXTS
zMsBt37<%cU;=ZS`%sR231Q^5|7#=t>oI8RoH(6yX&SY?f`uGb_>nu7?I%(w0(dSHc
z>@KNv;<Eh~hP)_E(E##p-(e{iMq~&VT<f=bBxtUkpxU^ky1p{RufOEjfV+*u3k=Dq
z9r0)x&bun2+$>%&yy&iZI*dn%$xql7a}ZyK)ek#{j!KV1$vYYRUT=irPpK$Xe5CzS
zusK;<3&PXr2eVg7R`vRkX>9V6>GFavW3_k$-uu=&jB}@a1~qHpMmXP6eD@K%zTre*
zSenr4MfWpVrst{cjB;XpLm>q!Y423*%fmEgLR>}a_4y8#EDoB9Q=<ol#Ouuxm<w{W
za|{_UbbQ0+$B8+VZW2a~pJz1R|1ju6B6hIXSsYZNF?K>NC+I;kFy`M&+oN=Sg4;xR
zx1G9DXnPHFUxBcw1_Ye8ei0tE5(p?|T)-r2b-Q7-3b?x6EM=8qTHAVr*<)i*0=^GR
z_j!ruK-YXSAH=+z2yf8otRgeK`9$Z&)vGbCfL`gg1Nney`K!LBB`!N3Jp<1muDio8
z8K{mTOvBTLYuec)gnMs?(*8L<|NexL5apbUpWu(>T7--CT0aC`$PWY0c{S}`4)>^s
zYKfz~AnoEr%XQrQ7N|nFUsr}KHNSB{vGRzR2}n1A{{+jU)-3(DLY!5(&O8j!Ah~n1
zQ;^#%r<sv*BO}i7+jsuxLAzwWkuJwHh9R`{Pi_aRe7IU>hCcUD@W$qWP>o0bkys=2
zK*o|_aaOOBqwRH;*eKx``o%n0(O9$E<y)+iD4*2pJv@kBrlbp@+2@1A4N<51d5X6a
zKFH&tV-)tkPCCW^_5g&-JR7l>ppX>%F@%=!!(gdtT7mMh+$~h8)Y^qNPZ(q&E1%z1
z3VI0j8<nNU-hDaIuPIW9jW1o7YIyEDBO^GJ(1)>XoMmXGBOaL44xPW-D!#LoaTo3;
z1o$6vlo&qFlJ{|ati*Qi;S)%dLI=b$bKudAuc7%cIy5T+-<{mBc|W?AXnK`(p;N<#
z=r6_jj|CNjIQOL6-9%h;U&}36rYaX375nG1rt}F@Hv9INX-*w)bZ=ESXQf<w=LV|~
ze$m%gm*8RV`yiKAnP6Q^CZzn{PN!R*KLvak?NKVTbGPhU>0CZ0Ig;~Q`0(x3i{2&}
z8+RLl|4}Lb`7`;nXp<tc{8S<aimj+$cig`|SAB{K&ZyIsMjgMO-6_SdQPqO#a6;e|
z6day6$)a}9)0E#VlS8g%oUl?FORE!DMROg15-LFk0&|<|uq_6XsUkix`m6QQXDzx#
z+8@t*TaEIPGSsqQq$xinxS;g6A5RTJ{VDoy=fljcsI*i1mfH|n+Z#~l7%58s2aS8G
z;ItzbA25A#%fA9ixu_OwU+kiNH5o$3scSY|gyb_G-?I@Xz3lU^*TP!^2kk3LR5+yo
z1x;A6m%ySddQMu(BtVZc;s|;}$~HjTvk{=wZcB0u>>s9CE#e>Lz#aG89GUlRT(|VV
zx$>sd(02clF+OUJnJ?6GxKN9v%zR%z*G+^sBkk$kFGZQ?X^j5n+rBZZSwEM-*Dzm>
zP&|+^A~+eN-Ee|P%-&THf0<-=$MJsa%Waor)~qr{gclrD;<L;x>Qh@W@+nZ9W8tF)
zYU3T3OGU)M(PjCAB)hhHLyDvkTme_y;n1)a@p}TOr^0XeK$z35<(U=ttYW7q>D*T^
zX1w12Qbei4P`MNdFDO=i*<|9cW$%F_uJVf(?IHdIGtCG6Y+31P9+Hyx(qV_s!O8S3
z=y>a>=0>Xx0%Ao6*m#-sYF^;u#w9qhUK(}0x&Odsf|rEPAu~58zi^yEz5Iy6eVtv2
zRY&XS91r)zY-{yP-sfn?bx^CBjS4H_`STrN>E$vUD3*kz&rzM_h-K7Cs+WbWER}s!
z+OPiYI=N!soe2yaB*4v~t9i!4P)^7373!Ro=4A|S?yUm|9qEnAZd)k|MU+A_O&K4(
zOQ5?(Om*@}PpgO`%w)8b@ND>wp44A+S5`<$ha;G6Q+t7El2OoYA4(XbO>*J1uVAMh
z;c)rpu4!)yGikFjCugbI%1ojArL^m5uGcUH*Ykq?wGjoN%w1$96v#G@@E{@5xq%zN
z!r)O^CUxE9r2L~PTkPS+xCUqD%c*6%4C3+4(XxzOP0b;9qi^x#SCBC+aqZV`(gcLc
zUu|<Pi%=Pr;8kW2mwv@|ZZET3SM!OgvRqz;hfb@{_heoq?^58Z;f?O3ONK~@bDkAC
zhOavHEJZl>3i2$&STf5B@|7bh_M~&&GRuNnztHcbV<}1sUIqsSai3}6Sq(o`m0;{i
z;)w&+3&DXNM7WbF#{rzqT&}y%l@vOa&E)2V$3A;`k?nn_exEBX?ZtP%KIBH3N7)vh
zV7?3C3d*AQ^WteuM75m7B%ncDDuw1=>l>bIr*M1(@D0hUZ7J1OW(Mtp)E>Urnqf()
zv{P64G#8&gO4+{_=}7R!Ad6B2KJ&Ogv=IIGeE@-}&Ur7dJAp&Gv_BZv{9GOHVY<HY
z^AP4r=z->aP)~F<gB3qhqeQ<l?<T%X8oGT#pg4z8H^`SxI50!>s6E$QT{2V4!#Ez!
zQ~4^PqvhRsLai#e%?Ek+xGdP=&d1EJukoMEqQTc0!bBkFlnAkE1;F}yur>isc|Mxg
zQDYV9qY=Jhjl$aKVj>7`kaON9%(BGi2i*oHPcfiigZC0#c~ujS2a1pxeQ={;il?>7
z57PKRAG(3G6-(%i+x2$idJBjeItJ7f9@{>_PvELjBydV$a6po*Z(T&(5&updX3#al
zBPU$Ka3I6)<)4ocA8SF-AUFWylC97M?P2bSkG9kLiCmU6s~0^I8c@R~ZupxbIymL!
z3JU7-b?iIw&`L$mA(gD$uSPmA18$=R>f*I_DZmZPy()aUd;&vb1F(t4YTQ6H8YjFV
zP~p=-Szsd3!A;#XvBt;?0UM%oxj!(VU$BSY_h`inBtCF49?A7Pl;p#N3Jo(O=Jy#9
zbfIh60`0QbfuKeRBkce?4mw$*P|rJl4a(O(?c(C^PrI&1Z1WNm;RZsjW;3#|PzoLk
zX2mNh{pke|??Y*Pd46)8WT}=UAJ%^!@-cX&Hmx@V+?4Tu-4sux#4Q>a706K@(lK(G
z546^8OoA%eSWtWO>^5p_SiYtb_7?%%`<BgjA(OlJ5kh_0NcS-%#fD|OdT?WwKLLXx
zHBg7@({&ZP6MvqE>3x97oS;pyva1_RH3~z5XE#=nVeyR|KY+$QdES*-RgggkqN%nT
z<$MWI!?S&yHR|uh(}ORz3ZS-xxMv8I*%-npw%M3lUXk7oP2x1CCX$#aWj4N_7OH56
zGu__37<fc;`{o9IDb+Aq!#S8Pm-Q=Ub*}!2E$P#T%Zto@UP2p*sOfg2tsokiXRku9
z*4g^QlhB^+{m?QM@uN<#{pme}p@<(j_6AD-1AYg<8lM<{@5_i6G=IFq$3or!xTtOR
zOH=gdU?0pa15>Y!7X*`lldY%KdjTYg@-6FpgeVUuTf1vCx+4ID((-e+-JH0-t^^r5
zuO3*3w-Sa_8kp0~i$&a<DgbM3Q<h4TugZ>=AzxPwF>)78FMN<)U7}ClBhSfcS6z<z
zOoQRL+}6Tu#=(>0GV=A832^4x>f;W06UfDN)&R^Iumsij1qoR+Pt-gj4>kdWVKl<U
zX*YZT(1>0nf~GaHJrC5lrORp_v5vh@`y!`}50$P8h4bTBa$sT?y$>SH$Ekm!wZ)1G
z?o<P4g}1abjt!PC&1RIH2HO*it=12<M(WIpO?ndL1#g*5m4p)DKK%)>%yrdk@yD)&
zpb|oRh+DM5DQowQBW>0Gcs8iE(i%V`6!q8i8Kgq>J0B_-Y5Z%q{q6Xt6{b9gXV<6L
zCtttAcTk!bkG%vf<Wq&jyHF4fj>ZJ4zhu~SAng>gy{t-u#l3Sg`%s<>EuHz$Hih#M
zCUa_2W@M5mAlCEDymMw+?nCx5xFYY|jGb0N2!q}`)@MRP!m%AM$BW}?hTFQ}r>;%S
zJ2ZbBeE)hzLIRc|A8EWl=bBRe0p#W`p$}Px4Q@Zb9^kQ>*3F#{hlb?fQ%j95Wd}Bd
zPIsP$;kq8+7?VwF@$5?{_Ix+f|E(axW4*q1%TvfrpKx^tFy(l0@t21_yj3ZY?4L7!
zSHMI463C|?9<0A{x+^R;L|r7eaGRkoz_3;M_&KC@K+)6W4B8U=hH~8xk{4-)gX#l@
z=w#lJNDiFw$%*XLOR-=&fiB>cK)s^_y<-D#96}a!*>TvT09N&(xd3l7R$S55p*<t@
zO0!;Iz4KLK!0!gG1GF_J=5?7h?&mcC1I}>XF7%tpRYWZTAYyeN_5EqTsuvdrBdm1y
zzVc>2(~`p>zET7m8NSuYgRDO`HiTQ|N6cxMc{m95PFLw|aa~qKl!k*8R4hVN!}`Gw
zrV5N-LrFMx)iKf~&!etO1g8afADbZ#M3iSk?s5z+;9J~$qH^P^I2Gf`Nd{pB>zG!3
z=IOdaIO&~HYak^z#zGk^1P)$%$-aB2|9poUAOD6VWz%cn|BfUjT$5nlPC;vWJmE-)
z&a&;F4d2&&_t0qq?9&d4U`5xUx#xe)wmlMl)AzNRuSO!^w6r>HjD6?krIn3`f+Y2`
zb&jAO1*=FgS;YAqW^Xd!INE?;dpX+Lz3d6ynoj_2z9CP*N7ee#*fj+Tj@gIK>np%z
zKH3RSIA2VVGdY`#+bTX7g{vYBSV4`HcVeQnfacL8%zgb?I=f3ia8AX6$SZ)0>Srsz
zvP1}2T2_3z!cgEYC2(=u9kKKGYy|*_F9JKgV7N$!-(@2Ni}Tqy6di}0x6f&Ff=?g)
z47Z;KJ!wT4nrc=x+U;065Aerfuun~yBSbLR^*FD({qCpCj5pXeC7BK}rX4{bB@X?z
zVx_sWHXk~Jdt27XB%VT3o4UwubzXt_QJW?nRD(?HtXmKr^5uTCnz8b>-?@da^~J^6
zQExSjxxjP%=ql4X0%|OBHR^SB(L)BSqZQ-Dz5(4svPErx@f{aRWFW&!&|`3|ICf4M
zu7p1r=!;qsylvFff+w6je}ZGgCn3VDT~T_Otb({k#8#Xk+???PSn@|xBU<CJo3NIM
za3kLBwN)H=hcyO*Kn)Q_v~L-ep?HQjTbxGk3s`#s=$mh`5uB0Ke%l;mbnxtCEeYDG
zt$mM4*RFLFDcRqxBWylnAHGJ)GHRvKQkv)~@wGtF`8w;{<8J`8{b>WN_kvz+5iHT-
zVy;u>woLfE@;|xXV6)R3=Y~@TgRq;#CQ7&S6%IXcIW!-+O>9KUW3#SZY4y{jH{~5Z
z!l+xj$}X)34n;96Y<5GLS8v#1=t1H`N3t&W{#YRX-ty5rfeeYn?0ac*{8^8w8lf^6
z;^X2m)I4G)-Bjib%~)Cbn7b{lDoe$%PC)2@aAS`G7)FRq>GF{Jb&B+qGKn_&ba;;h
z4~JyEU$9Mc$m>7}9!E(Y{zc$saGNj(@!FIL&*Kv?O*v#HlzCL|cCeoE`msd4%nyzy
zu}o0_lIz-Sb{))?ljP}J+#givrN`JT@RlF43|Y(|D^{NZsJh~iV{C2?OP2=EqA~E(
z?Q;FhFeJ{iDcM5U33f1AhMNS0n!z@(5Ukhe2YrgZ>NbR<2&}I4S9Mfb!!64M%`_hR
zSui!#`$$8=;WabiiVQ539`y@3+$K7Pz+1S&7I9+ZF|ReQ*Qr&&0RYQIrmW<1fIFGX
zGY(@j@HRWPc$!(5VYAYetCS(p<`Ro>im@xopI6m5HBwrW-psCquRmOC{)$o<cad>L
zXC;DE=sPdG7sz-tK+-Q9bOg$&JA*g@mZ^jicECY6of!55_X6+BwF86u;E+wUU(^;t
zgw2~PqNeCY*cZQ%?m6dqdp_7ruzR$mS4EKWbs!NB%Oz<zD3PK-Pff^;-cjGe-JiiP
z(4Cu1ghEnlXg)4CvgJ!ZSPP^{xvl6{ab<4`?kkc#;kbTvaWBmJtWGhGUL{-8{n<pz
z=+CnNtjg=o!DSMXZs~LKO7KcAu2KAhFULhNV@XK`b<i_0@GiRo^u4mn*S1eVQ^sU(
zCn6^g!wTQ=<|%ZEb)PZxisAU1#}bgwzeRdXEj=K7H>R1dzh;%vDX@R`D9<@y1RUoE
zzEPO3qTGaAp`K(uil;3dRxFDNV<avypdc<`^RmcwZW_HY-WFdxh4$eKm<d^PVy?i-
zQ}TK)_pp&(fD>+~x6td#=H#?yrhaI%Nzy95rfMcLkXKogqHAqsYvDD*2ExaGFE`K(
z0<~2Hz{xosKHsynA7LsXnQlT>s#O>TxZ0{nhn44m#bJV1+LO7%u%XRkE!_Kn?eg#Q
z;m=nz^FR}{N|}q6QT(?PaB2Q6MOM!Av?p?!P5`^blz0NYnk8+$51@rx2WBQzAs$rs
z&#kYaYm+7_BAZVH0U{9Jqq6?sFc(B5rA5n^T7}W%5ycFe^z9MKhwI~?N({uwF_RXc
zH^dZI-n#XbIh;)QQ+G!63~PNbNsz&$%0S{Bt$fyF6=g??H$7-m08}&N7}cFB{4~X5
z_tE%0Dfiqs;)1U%wd+)?qVhve-sO(OyAZCw%sbdl2q?u&)yR_`IsrWgD1yH8v3(Mz
zwZ<st$aX%>Ltk(4`p|?G%w}pt31zg1yNi!yU0s7<qtjA%0*TM<Qw($`=xBOsqq;ba
z#}+t#wLcHG1p*SUvT3LdvU~5`0M2J2Ze<_#svVcLr<?E&+)SGZYtycLG*PxhUSRGJ
zwuY|K_pg2iEiH(I=gLc&9ii9Gs$0Q_p@NEEyoUDPuLIkjV1<}5SX$={fa}vl(SyC9
zqAYqfLtmIf^`P%W+moUPMx8l@N8u7FP;@Z?)LCwcGU!XO$e>Ce-aDw7+8-?>pW-@%
zKQsze#tSwXfeNPC5Pzn0XnOfwCpEa(K7}<#8*l%S2S?(ZGg^`=tKIf<+Xzuj#~~%@
zA1<Vxc*|L1F}*qYL)07RN*02R@Mt-g>?~NBexnPA)2b?2!BA3Ok$CBimA*8;Fg;K+
zbab=!?=WQQiB2qb?E*TcHbq-igpeOQ(V|Z~g|TF_K*RxFYem|dRIk^1sk3MBFWsET
zGyOm`3b|MjeV_2q;0Mfhc8vL#Mi=6Eg=pItth|OysT`&^g)DuWJm%FBKeFt~5BBZ|
z{Rwma{j}HE`3TvS`k<%9`1@HR^cr{mEGG1da=(Aumt@NuswsLNB)7kN&ni|7#}8Ku
z*U!XI45GM=M>c*eB{1u1HE`;0I_FE5n(B4KxH-_Z0sYy{y3O}E&@b;{$3YKfG}ZI5
z4n%6-!*cmM=hw<MYu2zX@0W#+6hymGMnrPsv#PjyiPz24Q^X#~zv=O>WMuPcym5hc
zh?T}<g;}X5>;7KeOl2WRz@Q6C8<Qj0H3$71VJl)bL(s5%y|Uju*Kdoclw=4W^Wy#{
z%I%Y;@#L*(I^orbx{NKF10~_g<K5d2)05jdo-%wC6pMy4pohldaqFZw%UX`_*t#Y8
z>cM^Pj@_1?F7AIRmf}a-aJ^Q9&j4+za>u&+?jpUPsqe$+Pq0$atHKhMp;&udPeLOf
z1!C!EB`3;x&Wn2mj}K%xyrPxydMTsjj?W(pHpA-W49Ug48mks0ooD7%Uex77YMVdx
zNgI?9j~IYllwtP%GG23*J!4sSCMQ9tVK)<31RdFsa1Xh<1&P3sq9xY3eaO=%GZPWt
zHIJI*Gxp51<_Xuaxb;`HLCJok`(O9nYUJ|b@R8Q?U(3s%)wSXc9}ls{2m15Yf7EXO
zym^DgDz_Y+4!y8kuEqDy{rg{3232$v(E9gf*!wS=|DTIpt}GI8rSW`~^j8SRAOI=?
z`3l$%y}Q7TvJ3cJ8qJb})1%q6%}{QXY%(u9ZRQCUJ+ZLZKcfVHelrnN6!4viuTg#Y
z^WXi~liV@%WJd?!sMB$NvUh!1hRb58)Cims76AM0&3AR3o>zOqe{T!?hgTFQk;Iir
z?a;;hf4yjQ!Hqyt!zz5Sukx+dval?0C{KlU`627?tdGCXwO?NmaYC|0oXqKC-~RRL
z{2iwHhgWWaKP&!PK-|A+ME~%{|NMscHAv1;O~vi5NB`4h{`-}|1cOA@yZOoAfAT-Q
zqNs<wu6o+97yp;R^#9#m<p27hgYiWpGJF4j8q<IOfLs>jC}&}Ab9nRXo%!FNgRzH&
zLhye3)<2YLB5hElZ5r=#COD-M?x8F_{%=>hfD(L~Mw66j%&)MxItwbW%8UYy0TTu;
zz>w4Ox;WtY?5ezjj%02iEv}vee%3DF%suW&<jMk^1VCL9M54CjAox`|Ln`bwYdgbX
zKoqrfC!jakezXB31Efo70Wpet9C|*j4@gPJ`XV>&Rba}jY4P7KWv({zdTDPo4g6(<
zi}a&Y3XySHL?ih#<pxctT7X<ORHXB1#X~)Y2)JF~Rs()T51Crp!xC^RcL42v?C^kr
za^tS&eeXpd0Pt4}um?EJ`b0oFF~JF_&3@xzEq29qO;kJnpb4Oib=d-P=r_P@O!V?%
z%c~3cWSLqk8mfDQq079#<^#aS&N|n<OaPHzPhq$MXD5pUD0H6ozwCP89*Yoz(Gxh^
zuKSni@qPeu4sa_P-deyoGTxqPT8g@vvJ(FWvpbRNNgxgh)#uAo%OxM=$=aX(U~QyV
zcig}acLDgK-Xh1&Dl_`p32f1*JWuzds3Qahb~TN{xMUZ6Z2=3HwH%05AjRwf_qRdh
za|e>tQ}2Eh3!dwN=;4G?3IiK3ZKSR_Vx#u}7kxaj%wL#r<oAS4{`+c3DF&x3`s)Iq
zehU?(<OBuxLla}g8qdS9W{HU<RRH&35s4*(`qzX7$uTbnKF#rsOpffc3kd4)$Ps;a
zIV5`sV9n3MXL|X*mNsHei0O|3=8x>?Be;P~es>7^CFTF)?5v}rUfZ?}2r_`cAdDb6
zgouE&APobE(nu@a3J8cO-5r98N+TiN-5pAUG)m{tF?4+Q>}T)&#QVIyYkmK)_S!QJ
ztohw>#c`e&M$z5&D&T~69uaz{7qAnT(G^1gGh&tDaQpb%0}U<*;tdwU#1H?v&m`_K
zgVS&|JW!?n`~+13?P^;p22Or*poXyk%tF}?G{oqFF#Nr|C3vW=f0Kd(x~<Q`DdzXT
z2nXyZwZ{3JKTji-acpSk%emHIEL*jQ01+#7F4$c>0W%YC3;eY6_0MR%=TspCl<em=
z!W=rq(l6?d&G$M73yq`%k(sFrpvGCY9kPAmQ#+f7L)8c{q&}C?%f`LE<-Up}Cr`i=
zL;?O>Pv0G-fnNXU<qTkNazPN;c3mq9R)D$?p<I9g3xFua-i7l|JdEXg#_+4=&1Zml
z>4=_H?9ohvo#G5kb{7E4SyY?{7TRFxGqrJG#n8Pr8ZI!!vOR~hqx@LzX#lXWdj#G{
z8zbbH4SYXwefsleCaWE-R;L>p;C^h;2$3kxuRZgi2vG|4Fhp~t1T;@U$#?u^*WJ9$
z1mQQmYhSGa2#NIo@+`zJG7j=JgmF_bmhU?<V2|Hs;u~;J)Uvv5fS$GT9gR(Rb^?EP
zcK6eRdz0JbKt$?}-q5@UHGpUA0ss`YnL=Fs@e&pA>Q(r(k?PraV}2F%-adk!s(=F;
zy!@Q6S7YvRO1h6zdV7K>F>FR7AMiUjcFVwyZ66k=5=W2K;j!qA?=~BwZ#8)cu%!>6
z!xLTSOA53*pE+VIQg5#m0DIbuao|>ZYA|kH(vxdTUw|%KcQRtTfjlhN6A+S<H{O%*
zk5$^6td8Wxp_B_JLD?oXN$*x~C=6=~Wb`8yn7H%o?OI{>JJ_7QAV-ccP(o3ew1AgB
zuO@xFW)o0Pp*Y_WK|U6&4C(=BwrAk7n_9kT1G=Z2j)AF^f@lO@`(r1pOK?AvtDb%~
z?IP+dA9>8%J1CRHPfP-dXwH=5EE>RW@3xI!pt9n6V`420vFfIy2%L?137-BZ{W=#j
z|AO95u6|;-El~ZB0uH-VwGG|`)<p9bHLo|U<9h|(r0c2qm2W^F77{g8o`REuK-*BU
zc>b;M6IW`xM$Gc|Z3e?`gXj))cYbKC#By9}O!|(A_&V_(X*(qMS>qf~A%@?$>!3fP
zjr*G7L*!@Wl_bC(y2R*fo$uXGp$n%H=p6}xu!WUfK@@BOKfae>R_~%7V}Nc>x?YmZ
zT8qQ)ZUG27Ux?(3A^-Ez%QeJ;-jH<K1!PYbI1g;5GU5B%c*gf3dtiiy1?gSMuz;Rb
zOCJgd)*b79K<JM6Y)6mn`>$Zk1p_s4Pb<d#L{MO0J_@Oi*qdz$jN`h7g#jU?zYdl}
zpY;bfUqXH^di+);P)3Uc2D>WAkvyutW)BmGBLI1qZE+gZ>w@#%tdC(D_<8(jg+hyT
zLH!X2$DdeqKQbN^>X=`EVH3u8hG?AKKtXdbSBJML;0Tc9i;Y)^yIJLfc)rwvYG|5&
zzm&(_r{$$C5+YZ3u0>0TU|~La&w`617l+%Ni10%r$#~L{V@w@yRIl2h39Rnd{jVS4
zF^EDqn+Sab#wf65oXB5eO!`QW&E1aQ0(1!0rhaQ6Kj<Gs1z)AV=I@1l2>!yo?qyI}
zvoQmcpE<i9TmmPd7&|Pi&{8}s^`O1lMq;CvoVD8sQoPnZvBfC0g{+Lk;(G65gnD3k
zF!O#gsnUJPM30kaq&Jcayx8tb6^7Uq7798=<$&$9uOUpwZS40G?7G))PZ0N=feM6z
zag6@@E*dCARRO;F=SlvrqjT}Cy2HeA7&71j&*;uNfur>4m#ti2I*KE8MM?ydh=LQ%
zpq3~G3GVy?$NO+A#mbuq@e8N@A33egMTkc4E4Z<6QoH_w&v&1i8_(a*1qYgrlF(|}
z(V}_-cm2W203%~#do;c`#X5T|TqpoeJ}+i{`o)*%ecX%cJ10S?y2sYSt!E!JBBR9N
z*nanMu`Gh32bKpixIRme5s~m()A=-}sUxF<FCYg{rewM~Vdy^MS^zgDUb(j*rjK*j
z#u1_GBnSZ<+9|_|po%T&aJ~RLDre&}|1VVYOzk5_KR>Diemk66aBxW$f>lJ+=;X@{
zZI%n8fi9VZ`FY(>9L&Fin0Q@l{q-y$A@!~*NA7K|&rRuEv?ik7s2|{yZ%4~=br=)1
zQm%j_U9nDVZ;{Z5^y}O6XacTf6G8}8hnx33^~bDMxh-_J41Y?C%QR159XFNJFR>~6
zWYQv}OL5nUt?>76cl&B<8T~?Avfi37&n$aBfzokCez*9yfaBk<?1x`Wv8tWFN&gim
zswm>EFlYbhQO?kRGNa$t1K_Y6X4bjl(`xRmq4^NCJ~o}h9Vr3Rgcp{=DCR|V=>>Ju
z84lq0pW(xuk!TSA;!FqzI!d|=>XnAaZM*}7IvH_onpa^G_fdWephdZ<uDhbHJ3Evb
zR)fsW(FGA?!i!HG27PBEJx?>JRyU(f-r9rl0q^Sq4;?>@WqCCxfKWwAVEh+w%~^jn
zFR-V-FI{(VA9f*#8k|?hm{(UYLLIH7M#STn{{#-<Ces+4s7f;T;a;gIOy|s%_SK~-
zwWj2cnAA<`b?kyDIkVhGb_2P%z>w;nARkmd1f^Z=I`^ku9pKhm0@J#n8FTm83?AQ9
z-+xz>0BFwXPFIl83BtVY|8jeSB<#t;XwNe2&synQU8?z^c%-1o8fj-qU?+AkW^$Jk
z_R(~6UK$7EwL2)SIU<d|cjCkmy;JcRySM8vnoKO+MSgyD;(`B-e9gQk2kaCwAMoMv
z_|)<Emp_56n?*S!uH7}WAq0}(!tj!{{UvKkczvv3AWiBuP}Iok@1Oy|^T<_}Bz)^`
zFq_E8h8GlNyKC-C8b~)}8#L6bDF>R%7d|UxvsUo;SMqmP@^5UXo({bY^Pk|gl!Eb5
z(k*yj_`jTuW4-?p>BxS8WZj`Mp=*!cWiH&a$xb}_3GT!X#~TInnNvBPab{IXQkU&c
zKIM-w!$Gk#a31d<S-&toL@n>d5sl(A^oAn6D?hP^PO6n~dd=r4KJ;G>$nFr&=Z9b=
zgjD8o-g3_w;{FkF4@uH{k{H6-L!`bPcN~V_k?Yw)t1i8$u4~cP+Cdq1`wuu1^NX5y
zVOdeZmCdZksU_oXdu5RwY39rR+pHcANc%w}k!ElTiU%4`wTgt>i{AMP?yKUBv|^Ym
z;e2t(N?y;9d{K|H$4{B3*+&!Yrm$3<$sAmnINK(;`X$7O^jW({ZD<)OGPLY-h2GiQ
zOE#<AJ%uw@*8BHfXcI);P8QuKhUSW@TgTiZW>AKm2<A!-yO!l~&N=^aLCv>ywNL*p
z!Gwlj)CO_=UARzpz=GD^!oy4*t_CdJb$_pmn?dg*aj{+Po5)%j&|?TrDR)O3-aD!n
z*u6HWn-)YpStyL5Q$*5q`0*frvA762<kQ9}z`ss8Du_^!pGZFRRx}S9glqdfWJrXE
zsB@|)i6UYoSQBv*JJQQto|Y811BSlS%hu}_w#Fl4_Ch}hfnz_HuagM3|F<iS{&4!l
z#_L$br4<41!>4<X2_+$kI8Pxx+b`4$1GgmleX3<h`48#UAKN}ws(UD7Yt~!jOzce~
z{bA>jm-y>_?{83sHcKA~dr(BwNf|+V+cSUi8RJeg+=G6}u8_}29Hu^!A;o?RHOtas
z!95Kq%vbUGasxqmCgtgya8%u5PG7AWA*G{}DZouCC@GRn&rq_NECQvGY`toaHNpD*
z0G{4f*W`;^Q^`VlGlgcjD_X6<Ugx-gRXhZS^`J@uf^WYOqJ;d|lclAUK|8nWaRTIx
zd`-{Z>je~|!=$NZa{DU?32>h5^jiy_u%5uQC82osuj-@ByIZwu090N1fOSr-&nq~o
zk@`y04rnF5b24a&jg4Fd5Y40RD{~hhiui>+j`Y8$x*ip6|BjsgMeF_fj;||g6sO$c
zk}A3^V7;Z4Z6M6pwnWdovHd2Ib^KPnds`;)7r==B#r*SndjT8tLQ8J<$n6^n{aWCG
zWW=tS(WIq^k3=Ewj$QO$@o8M4Y$eCu@V<x;7ixOK4c2fDLmzalB&~Rd@_$L^|43N;
z05W6<3F-FzGxuPPk(dZKC!?c^+jhIsr2p6l1CxQkDO+C7!vTm|Y<o!Lj+X-osmvK5
zm%=W!Gd=`pi#Lm@z?9MXWoT4WiQ4;79>P;gGe4C)mEEafICWS$=s0?y<(YU&eAXq-
z-N6JMaEcfJ>tFfdc1N0apq75q&w}0w>C18z0kdxG5bcwh-dWa%K;BVA-@;=*FWE^Y
zcaUIz7d7)eA0!t`ciC1wCagg4nHYRj3%`za--*lzvIXLgUQ=RBxKksqp@`(nFi_j@
zh}KBfUQzuNa4X#69f}I@M|86q+y<VZ^%(DNi6k^7x{0P7o(oEcP*x8j-!WmgR?Jl4
zBW@9@#j~V&eIcQ}s@0}W<L_k_Y%>cZF3+cbPAa2E|HskPJOyt0is4F?k!3=9!ILA@
zsKFM%)jsi<H(OD+dof9Ji(RFM`I)Wc?D%ot^f7VT@`rL7G#?QZk&2w%xPEWRwP9=x
zYS{FF3DldLZmi=aj^ckUwF)UM*UPkCtO`8ns0<Zkx(Fm3INZ?%3GUw8nrA1nEOC%~
z2U_V+=$cTj<Sa==KPC2ggVkt0|7hE&zGX$JmBs>zi+g}b%<nE_IGH!)w}%4K(SN@w
zn;R3Q$%xES1T5Y<01kui4cIlL7>!8_l#*^8t_(=sxch{O^1Z;07)N#mcb8Pl?4n32
z7VER&?2ndC(gTIHf`|6WDmOv@oTYkz?Fudt8SF7r{T63-r2=1TT;?IO>XzI^aprAa
zRgw?qYVV|%T+XBg0%iOOQ)XeePIQBKW}dQMgTfqw0_@3P?m_tW(k<m70Yfm@Q#qip
zk!}mqnST43Tbh=YpA2<$n{r}YJL}DDlN=xUfo+eJWR!FZZfHPVxTn}!1-=Ek0+TX}
zDdTURqF+|LNq&;7@0wfBsz8b|HDf>9V^ttSJ--Dqa#B>JhI!5DVTj=KBL&q0rIchz
znKsH@5Tb{;;E|1axfGD0qD)bOU{37<fS1Wp?YEJNEPHq`>Y#x6k5hH7UE&uz5#6XX
zQrJyOIzkdQla9BFF0J{`AVvew&=jL=8GmNyq&)#<<wSPYX7;g)vl7ZrL-mAH$ff~<
za>h^R=y5)oUsO~y$8L4{cwo4yD=Qa+^dUGPHZQXL8><5BEE!9&L<W_w^=g&tA|*l`
zc#6w3#2w(k96L#5sp(k`Bk84zi^1*o!}MpR{m3C`Xhu77M%e^Ldluts&JmAsED69C
zn3ujVJSo{gnCMS%tF>A2ryGT}>1H>7RnXyTK+mo55eq3|5__u@$FFz*Mvj0SI<)6&
z3^}RN&|XX#+tMqP+ox^mYWZ&lE%BlFgfci@{Q7X&A>7w#K1Qml^_hhtTY(_>jI_@x
zPpD9l`PrzQHTT$CF*BpX$2R-3M%4*2=hm~Iia*2=<I<T$bbB)(3^Jd8Bne-v&GIyb
zpLAgOc&Lr&Lh4y1%F2>LR0upUt#0I6?Guwr*X^5&#=Xu9=|<%C72~(j{q}~81f>*A
zBc6YBd_cLM9qSo?tb5Fl7ZVx3bp4@t&0K`%OQBJR5c`|bHV#YO??8quu%`?1Q|Y>{
z??B@4u}ZW1Q4r73R#wODfx-%(a<m#xx(*t91a<W5$479XSk+YQJ~9&%3R*Q|qu6hO
zddX*iO9i`#@JixJ)3V@VfAp8;uIJE)WKIPaJDr}I(bPpt)nUa-42nl6o&%?xaX%dB
zBTZ?99{p?I?W@fg5XxbzufVrPFlDw_rtV&hL?O`MF(3Vaj_I`Z+hsG)_!(jvHx<7R
zpy!|Tb7C?;IG^mVGR~+bV$+kiz9+cB&ef!!{AS1`BXhY&S`vTCSM-b0ittJaG=D~5
z9djecC|;)J9{$N-_NVL|C|A<hjd`N=6A=HBgyGtIJF-j?abNlsmX#tJRZN)M9SKD_
zKn1TbIMV$W(A2(gr}eSyzPFP$h5Z$<q0EB;UbQMdh4P|FkAQ4TM#_iWrF6x$kbqEz
zgMgf)bjJ{ok)=hdE#s2{^q&I8Udli)<;ylw*b=s}{B}iM4@!?8wbb7h+3C0y#66s9
zH&+>#?&((=3^nltgwT#G1GKH>Vvb=#{K?#9U%FDl^OPsKIP~2S>EK99(uzgJ^!fgV
z;sehcid7xl_|GJx2QLjzbUEiOmj}dFizO?gG}kVBz{0)q9tuKH7w?$g%tl}l?YMJD
z=Lw90;zJ^-5g$GsCWT;@6hDseQmVUduE(|{8^j~s5NPKE`F;q>8rs@2Lsp>iOSC9$
zsSTTzp;H#&yhZGG#fVA+V`Zr~VeQrYPXtyfJH#Xi8p9IG!O^90m5!3a2w(n@SSur9
zFy-1qR{kQM?E9SDrqAw7KXSq4$BI)i3iTh>*m^@mFHM4c?0<I??PI``W$`G(_-#Bp
zwEkvamL1o!{vZxYY(d@W<101?!enofCndFy>L{K(&B{#f)&jo$(yp2~9ym!dac~8`
zM75@9wRg!ivjJ4Oit1r;6xF4M0c=e@ND9g^lSBGSjqW8*L0FJZrp`VpjjXd=s$2Zg
zBHIDwNtw$~*C{13&bFIC(41kNv)us3Y~JZ(uPpkY8t^@Qw$VC%Vlv0j*QK=%@Dsb0
zp*q($B~*SY4+Ob9LAKKVh<cc+KzKFn1AD7|vTcT*+EJ~1k<I0c*i@JLl7mW6%IfM)
ze_HXvx#Jh#c`Ac7L(iB&OtLq(xCvo);+HOK;?kQZj|ay-zs{pPce=AyvT`bN``?>W
z|JLsC4uQjA@9zAwYo}d=efD4vctPxWP{d_YUKQ9nFO55RCZNSxy}~fLFbvIi?$RD1
zL{%=pS6o27pwoqK$0LYGf+g7qE5U_F?(8c6+j~PPlh1SV;sfxM+?Z`>XTBK4EB=I$
z%byS<u0<-7%_r@2lp-V+=G&-|F@41v5jsjO7D9x-mc`YKIxw$Xy&nXnPg5(fR1(FC
zJyL1e5P}?{OB*K1Qy8ROg@C^mBwSsEGqFbTk!7NafadjUze&C}J=`yjr6NFuJ(|-m
z7Ws_50B-gdff^U+Lg9d()+Y2599Pasp;KP&S<76cum=TN9eYT2?0w^>sN83Wn_=9U
z5iX7FETx1-lk`hT(1Z66AQUgVhYIJFq*{bgXGr)0q8bR#@`$Z6M{70Oj(oUxz$hO3
zZ>c=ju|tt&qgjV)fje*mpKt1QH<oH_Mt|hraa;O2QSR9=lp64!e<wXaL4e-0-VHu<
zmA5<N5!O22Ib)QfALS{V8YHNSAgn7kuKx`jgu_87O%V`yf~s3vSIh5{npkjTCkOM0
z|5=*L#K!V4*#pA(hi4~%aGjsQ%VQ+~m~gJ7?^PwK_pu$-ts!R~B($6Q9><IEvhZrJ
zLw>@VHMDZgVcNh;HU>S^Ol0ra>V5#rl~(iZ1FeU4{>9HS^+~ThXxQrev4mM*2gKKs
z`d6V&)UNAA<yrc)1%C8?Ui#gbvowAt(3Ykg_2k#PB7?N_-M$+@kALwZ!ExlfQ;qL^
zV>CDj5aTivv8^XgbcsPgaa&s}<LzaB`0LFZ&{s(Q!Byj&)Dn&sRZcz&lC$~&O2-R5
zRL0Chyh(7JsX@MO3(0ry4%4vt&2Bk;_8Tn$pkZ_-?DgXWHOCuRJ~iZ%pw2!_RF0uK
z4nA6~@DXC{b<lB!A$^xKZ#a}|KRU~3)`r6ea?>O};FBY|l<R!!u6hE1R--eoO$id*
z3d4iKJ|Vx}Gnj%F8=dtRE*{g`0|vA9FdtCv_>V(umh&V?1U75;l&DULh1UyP#kst2
zO`YO#O5EmOOk9vOGb6&!(U@zGL9$;qWx!OWk-Lg!Gd00Tv@MWz#~<2v8$r$Ih<V<>
zw~csQr<I|Xbn9dA=-)BqfB6qM6KGdR4Vvoz)vkDiJ>yO^#~ZXVgNKhtRR(2j+@nNX
zAIVg0kw!pkWCZMI?XQl`&=5|w+R>6*H1QJUOPOLTOF%JXND(*)YOioR2*rDB4X3L{
z+Vr&{pD~>!$LzoZh?8vqylRW+*>;OyVlH%vFm4XiR_!1Ll*(X4Kna@S#N(PFxte^Q
zf`?TJ<|_&>5I5bZos?D!;sJi@jP020bim(|dKR}9X3R@NdkS0-OIMb8m!Iu<RpQ3a
z&T=7>A&J&-6+lL%=737b$2~Ag@1Zp@M4?~lDA29H974BNaLj0(TKhdza0ZWr3RTz#
zdpG5_r^qHQodW4v-|z%d9p-%BWe&)-{)<O=&8oFp8A>#yDh#<r%HIfRlc51spwCH^
zGU;^LR01t<F^Be}_53l@7eXfw>#_TSEqNh#YytUr2#TJjw5VhudegO^RVBl)DLj%v
z-q?N%!#CU4$%$ivUe|&)pQ5R^3&)dEmXqgmg4TG)P5fN-jaSqj{~|TWUkGK%Sl*1F
zAU!$UN^%tK<$AIXB2lKjbt2()L!)Qz`WHc5MQp6jMR(V+lb;!xn3c&wCY5+E{S1A#
zeBXe6fI?X4Uge=u_ah2h!<=YFF69$@3^PDoK3K!rgvb&}^4D4;9$r(`(ZO5v(<V2b
zrI!jAt0Zy9MKzyc?CgSZ1+!qnDzq-)gqn7!KjP`1VCGI{z==d-I10Nyv<#yiukdXF
zE+Y5N9PCb8`Mh(~F4Z1-z^RpETlJMWM~DsDfd@At@Ms=q?ANz<j6LNkdwG}of30He
zX(JnZFd}ZLn^SvI7r4Ea*GKa`dtczO6eQRc5qc4AYx9neJf#8Q7kFF^LN4@3<1Wf+
zuB24k(HyNl$JN0$gtmp)#VOvdVvVL^jacaC;D&$gVE^Y@hx37qO6A`&DnU%t3?mXp
zE>^&nqdJ%q$t|Pa%E0PDijrfh7UV>pD-OklteBTut{`%_r&KW$SFUEQJjh#NZpaK$
zdL<<wwE&nwAi-i~rM(X83c|;c7BH2;QA{zWSRKmxuwq`3DIxcYR8!rLfz^i;U}erL
zSM$)BTT0x5&mL%hn}2Nbgg9F|X$~klx)M1j*Rt|lhcO4^6G;oS$;iA;nHraFgI)}G
zlN_olqm-CgyRX9}pEzqoG6K_s<dyb%(HvtD)dHv2;(Nu3X?J~Hta)i!m!L4);E<QO
z^~m5}WC3lpv{c=b@q(%)$}DiL@9(`%<R?bOGqNH`0Qwrz(#_moUWhkj(DE}RgeNb)
zvcULb{K0+DcQoyd7u2L(2;?~}54zz5>X<MXt~BSODDKm-kdPr$AizIJ48h+OqeBLS
zJe3wTenP{S`(+<F+*RDpLxGCFZ=aHh(QPd<Euc{e2Q0w#T4M8++m^yGaqVf<<S@%4
zn^hoo%T%Ffqq0*UxO0Kgk)P8cQKGp2!E$iBVZ^rIbN-PGbQid@hJw0JTp`;&XX)*e
ztLqnF@E{3{HzWYcAEz~&n#rZgSNhc~Lq9pUq4`FC2A@dT`0WWC_or<kDa~R{+Jhkc
zxszi9JUKX#tL$o=T;AdITiJ=~5@se3OXDrcQ7IN1YXzx_{d7&jv4+H;^%L<$FH&8b
z+3Ci-B43uF+)zzQsm&z6OI35P1~ns)$8k}`18W@Ad9m21_Hy^Buf5_I=X|m-^-z~X
zp2-JvQU<b<miP6$ZrJ$kP&iPpQ=b#<jiI$KB@QBF#@%6~l2XkfA-x-}tV5pxy*qs3
zepYp48+HdhbF`-`+GL<W<#u;cAur4pznv;<hc&F?Y~DHT_#9+8at$8m*7_=vDVaxO
z?bKY&Z~hep<?l7T*j~eQ6WEAlFc=4oLpNubgC5R4zeQ6b*Ml#FTuPZ-Do-k2+-0pS
z<HqXlf1q8_sy1=)ZUlXwXg3(SXX>N6H@UQo5*sJ*t+2-+OSfP-lRu%0U#2dB)Bk|{
zf!p8&mPytHchCwQp{Zzr=Q=%@;n>jKu5zQh>4<iuDjK(_cM=RXSO5+Z6)3s=MHQ>z
zQqUxfaZ>uLsqioC)ZbYmOcHeULcVm0{JC?#T!$E5IRHKp6+LPYfWL((J}1;-Mdo$#
z7h3ehn%z<>FiN5AU)Pmdk8NamVv1sCp={@Jz>Awkp0hC?HMNUlt3#_K)YIsL6(yDp
zyrMI0r|YrkuU~I{1ifl}<yVH~&9i&d!@Mzg`1BCJF^pzHYkYy{L*K9+GfDXa>ASNu
zay0jcy9N0-=^wiQtVFR9!<)3T8kmAiG8ehnlA;zEkpnDyO8ZQTH{q47Z-%c#bjJ(+
z7@Ip~A(e))RJa(FD(+TRe)~)>O>0PK-wnttMT{%-_|hrJN3uh{@5VcT<eUeQ?gVI8
zVUv}vsI4FugsIxpP=+!cX@+DPHho)&9Q0I@f*_ezTH+6dSwE-YR^h1|9u_0y2HaS>
z#zpeBG#yk*;y!NEHEf0*Y)rO#VoBp<KMyeXxdS4d3VMUo)A>+XvhB>Nr7wD<?!*lQ
zq*VC2HSKx-a$s?e1rI4wQvC`hg*EADByl@7%FVJlwKAGJOJwB~>LyW`OiPP1Hvwkk
z!dC}a-w@j1y<>{LA(u&ReS<v}&rypw#9VD)ZO0+V6pqyvUBz&{l?%F#I14YoBi|hi
z6e~6c8IPCgE<IuyvELMKSo&;VcSuqOb$q`}S>ekcCyk&gVl@AyVdYda%Ot(8gU$#X
zz$!y=3E_2=4<@Hso%T19lQ$Q^J;EzlMILPw)^z{viR3GM)t4)N$Fg;+_^rjz5<s59
zE=}TnztG8c_vo|G>Jfio2P%SNq#B5zx^l+WEak&)YEyj?Ejnnd4DpOs?(e0z?Izr|
zr#vY>I3fKFlsK-orbG*HXBhXkOk^9i1n}b;(R$k$i-j_G-y;Ts2=gA2I&$%d0{Ya$
z^Am+S;W#+Snp(FJi?0h-X%%(zcqN?1hQc_H(%Ah*-f#8958-_1_=?ob<FEKC7`sps
zZ-4|_c8tP58W(XR-h`BFN<Gr<7s1nQc7DxNyttnOin>nf%}W)^&?S0<B)dM>_U3!m
zjqNz9M-S8?u)DtCYb#nMRJL&H_)M6Bh$85)96uA@VZ;{OzV3Ud`X=h;R`Smy@@b~G
z4N2+OX2Ob7L{*60=oTe~0Uf*eQ6muQJTrN53dQ!OsdP1>I65M;MZe6!)@Jjye-BF7
zC`=<F2m+&?#p`!qodwaZ+b+10tm`}z2U8baPl>s9%2y}J*5LucOH_BMtnu8jI}mrW
zrkO+|Ui;cKl^dxQT9n>7;p_BseYkUGM?AVs+)Osz4!Wn(qK0v8Oi?(_$(~KFG31vw
zKFFGhhQ4OcQJgc;E6ts`mTva7vgaepE%p>JSILdOY|^GsYuE?5q7BQ78P)YOQul@-
zhbXFYYQ>$4_<e=PHEB0_Q>PD?Jzhv(B4R=wvG7jGiXS=%?KX<<zuBJ*k)vRv*T>b0
zpO0IaGz_R>>IAQ!MN{6Tnu_RkHp{z{vURe@SH7_av2Tqgu^{%=p{Iw}Q>A!DJts}>
z@CZ9yVY9SqC`X(&bEIxiFA~0R$`HYGGqO6qzKgioFJPGJLDX`lVM(-mKCm3OPcuDz
z#=BESl0P2@`fqoTey=g-jYpisJZXO)mHhki_~Vca#^wOA6M+Figz6Nd*_?0fLo6m0
zK1?Te-~Jnk%kWZ^@m31&908L`8${i4k|P3I7(#Dm|DOTRzqfcb9|P_zKaH^K`QI(7
zNC^#wLSTACpi#F~f7%HI{c`UfYdoj^WcC$~q3pFC8^HAilH4ne#YW(r{bCaQbM{Z|
zF8>DM{VPm`hTBlLL7M_)D>d>jzdW1T8=WnIxwat+D;j=KT_EkT1|H>uUnH;G<g<)K
z$Iu%2KiQ#p>bO`CMZ7*w;7+BHUP<S9RwQ9JnIc0lj$VskZ1{Uc{&6OUq}xm?G!S+2
zNih2JgBhr{1&|p<iJxstjRSLY>$f27nluGyPCikk&V%0(B|uji2C`c=9p7T1zb_;v
z6&R!zn4-M|lDw|n#jd>jy;TZD&xbPUe^onvAfgd6&EN3@_}>|Je-iZK5PWBywF{>d
zC!@g08IHEQ1MO$tfBuPRlOeO+h3^4;6NmlF0Yo&cj$o3>|My>Z1}Mc{Af+%B!D{a+
z1+Ea^fnyM?p&OmUFEVYTSr}#U^QfbJP#|GOj~bp6)o1d~R{A0|#(zTFc~coW*SvT4
z1pa$O5d$yR61`X~=gw$<*OU6qO2B3U>p&v9JLI>h+`lZ6|5&`5=u=aR&)y^A|MLT|
zfTJ82T)|xCchSc`pZfoZP1w+d54PC*4YL2AAAkwu<P=cmlE3w=|5h3MP0{QhFL}ch
zL1Y9|jGz9~uI~T*7yt1FZ!bX2UW0tl`R8^A8?-T7+HhyCZ5MEIzzjh#&3?W;=kc)4
zs0g+tsi#-W-_)?D?!wLg`_ZdI!Bb*`u$TNvsCy$N(5{Uy!R|8;aXDW6+y%xF#UA(k
zVLZknJp+uoJ~o_h*uDhLYG`2f@ybB$;a@y4_WS^##aQoIjYLh9S;+%rUP0Fh5Mqyj
z;%v9RkID!St%~F)DSv?Op8r867t@Hkl;#3rF=iU(h^7`X`v4i=kRsOP_PaCpAIlDo
z7PDT%PyBNgH)hMu00QzIpgjx#;pPw{?pfa*C!h>+L;Vd&Ag+BtB(wt0GHCE9D<EKl
z7X!Mnq$g3-^|~yK(GH+C-+?vD0vZ<vFd4aCwOQAv?_VT+O~KwTezltKQxWl(7r^6Z
z{a?JT!Gx~ejf&|`AOgPA^_lZC5Fg2b6DG7bj+0CoF729D7Ca}fcR&S=c}sau4xDtP
zIsm5vR)UuPW9k4kR)9eA+n9Qd+WO~8ezT1K@lqmK%Zb<#f4;8}|N0T&wiT?e*y;k&
z>1W{MA`d=SHSdudP>iF2zfRx}mj*hP!_^$@XP2ZDB2I<DO#s_bV$ul15D5XN;Lt^K
zLWm;hrrlHa`HMPk{X>ipoZ)ef(gM%kz%<ULoJ7|o+X;()0tl_urSF1+UN&xRv*FUi
zJE3`xNkmEWRkIt~1PAT9sk^CcPsAp`Dh4jQ|FIqa?f*a-!QpZZ;`t{&-<bVy24J%H
zh+Wu#Uj^_n`2c`mIr!1aRNZOzSUj6vg6rYtUBH0H)`E#GY<j=4UHxaOh-qbjFM!Fo
zWomdxelI#IwGnn$QWytA-yE<#dj$h<IAWi#a{5_Dl=7>(taf>ztDevxC9NpvHo#pX
zoXc9{)u&nob8FL%sQ<66A`uUsfIY+C{u6&O_XQ6$(xHWv_cY(h`~qesUqRn|7feNj
z9W&DvK(N?cU2`qz4e!`UgC9djME}=s!2rUBhr)lJGr%+fqZyCk6%tH~XFzv<xPr>8
zMnB!H`1yEdF$<WVI*VG#_<NzvB+*|!aQu*Q2IUyj@nW;VYz;QBqV?EcEdzC!e&iTf
zq|c0;mwOYeW}$%7wjtQ40<IsOMu9ZUXfvVS(@=8GPe3YwU+YOv<Gw$DM!QWD0Tl5k
z<4n*dvtSQ~HCt@Xn>PF5`M3c$O17%qPPWC&)VIYiEkx7A(eR&lVQ#KV>onV%07Y+-
z_%y9l&=25zCyVYhbstXJiVK4Kei5|j-l3f)XU=m0Kh>TJHl&UMD5fx@0mdbV?b4A7
zTLZNDChzFfU&wH6qZ9CJ^*JE=_=!W06+HG-e{-3{nR}++pu%%gz&|_J<G$^W^zbt<
z+)i;H64;+#_!SMvikvhlk-V=4we7)tJz=&(<~Zy3YU{*eM~xF?{FtAT+;)TGU65IV
zeY4Gg#AgL$uX61&XLCTKHE=xgsa53|8v`|Hc5svxN0wfUhlqR{6jw=9*&s=IVd?55
z6q=5<@q{z5DWiu|=MrwSpWHjy1IcgF-gD=Pd&^&WI2sQ?*i#**H@^e!_G#<;CrDtz
zv<OTjZI=dL)NC-*2*2(=JGfn_yzk%y$N2yXLy~~}G&RxhOBA&^L-n1Q1nu-L!{#Vb
zMCHnYZ^?i9hPEZv-ve!E%X$9xL5oD^_I848e*w|3anLK}EUEs(s4@yb@VyhZB|ZEl
zRRrZ|>{?!i0F%4E>Ghdz02X5|3Y;JMOYv&s4iue~)uM>7niVw=c$E4FCV}z@FpnW0
zDF8k$<_?VQ>ULWIoz0H?fQZ7J4@mmK?Fk>^TQ38;<U!!Mvvygi8Wg*C%Sji_Q7?iM
zvAm2V(FVv!KDq(~w-ANdo&u;$5#*i=SU+Z<Ds_<|-C@P$zflKdn@i)cwKgEMG)*KF
zH8<y3!u#zGk?=<=j}V1CYJWzBpc@5UDwGN&`os7a6fSHT$gJ6=qIZGsWhbz#G(p6H
z!~FJ*ZURhtm71kkD}_>x1`l^qImv!6LSqOvJRR+`ZoMaThtZl}Di;N&k&p9&^;w07
z2P$TIlM4@Sast!W_`@PeLi&ERW($C=WzM+1+wB=(5w>76pex^aHidW*c4-5zfP1Bm
zAduN;dCT6iE>0!d@!S3i2AOhH8qs-Xh&0FocS;1WoevaZOb3C%XAq}eUF+i|HoP7n
z;{5<b8{>bvQ(lIH_PwSZ;g;kXZqrZD7GyG9CEdm3gfKKND0`lKz`FzF8M2oTk&LYY
zki2UnP~$H)@Gu`y&_~hn;rEunW#IS&B-ZS$q_AK3_KxTEx;Cl4WG?OwP3IX}F8Wm9
z*`V|$IB2QkyeMGP$Ka_c@L|%<#6-c1ML=`${fqH?Q8$g{O7SMH&S^uaROy1un~1wX
zUfl@HVJcOhUp7`qmB|LVc-?EY6>B0IxN5n!n}!B_RS5c06yf{<7=nqN&qt5;7LUV=
zvScG@Zw>%Q8b?J)$OV-m^iU!))c)!@biNDtD3H0gt{?f4DsjEL-l`Ae+hySmD_a!2
zB*>#M-d<>syA_c3ISaHpeg4kg_$cUrWtbTr;<HVK;MlgBEYAym&!uEoD{&bZL?Ii7
z3#DF`r#Pf<ZXXLdq#tTN^FPL&9|e^NHmkfcP5*$a0Z=*-`b0$YSa1=62_g2R|6{X0
zBbKM%4j|l{4BdQA%~#>b)}uP0G@YQlg2lhj9kv)fTI3O&0X1ukC#pd>xs0iGeY%S?
ze|1EQqM<g3cu)JiHia3mK8V+Kn)T;jo{b=h#9cRT^1<=BM}YJauWYRS93__^>#k2R
z4XilDPJ?$S&sPA90NA`e-s>lHoxb(2!BsmMOE4<+Hs)JI@Nc8#FFtrHm)+)nN2I`E
z<rU-jIS|pj0wZgK)G$NTRQ=VsLw+be&xjk{ME4%6jW&lE6^#{l5;+9pUnmLxnt9C{
z@{%>X&@GaDSDn~RNuc%@AV%la7a~7I;Mc_Cx5S5(#k&N)tbVkKw&LuPYQs^I2~w(?
z6ht8xUR@DcbBD>_hgsc6Rhle<TG)dTcUbDU`$B5s$M}#J#hdei_+P0vbC%&<S46gj
zFNjf(XH|PopK-33SDlVFT$oLdD7wD^*OU+}jY1YBWO3F+?5G}rP!jl=fDcP>S^9ob
z>caO5IncQ)2w2>_DI_7w+Rg1N1;D&{Gz+77g5sJKGi=Q)8klDsYbFBueNUEM-@O+<
z$Fj(|bLc8_oCdPK{mANQiD$@M`JRQ#pMiDLDivVpW9wM_UujRe(^dMM9;@bz(@boE
zn(S*<{LW|c)8$AK)(e*<fle^nSy&Jjc^@Vct(O9LT~}~be3=9RZbFE{Pf}%(cg~#q
zk76&pi~XKQKv7u2eYCP+cfk>CURBawIWw(g`<_4jXnMdT`Ssny4r5~D+}1Hav8-nH
zj44){lNTaJKh?ZT)i9HW2^&%$wecwVhbZ}vZi0zj4oJeGld_7Oj6|2tEOK=!<>5)q
zeGPKs+OAHn#R%wzYPO;wM^kieVt;(detgJFn{{cbEOM2#4B%)amYpbNqac^fk*E>j
z$)EoACNHzqMN@4?^X&{Q_6#fssxB^ZP3c{mwGC|?-GjQL_|N+rorx_f>EIeO{;5Hd
zFAQ8x<3ofJTn6t6d*`L(aR+9P%2Eldhi!tO@|`cA^v%euiya>|?m{*C?7ZZ{e4o0O
z2izW+6{c1uu${#yOmaZMA8@@fcE@s+6;-ExK0mWEkV-Rhjp74vWnNZIqFetp$ux3}
zHS7%#!BmF->Y#QAYdOPCkIpF4myU76zh{1O_B(Yl2#7MZv5s<H|I<L~qZbB$+y`tI
z5P4Vh=t0i;*o}=9>!467q!aU_ejM)!Fu2&5Y7Ug;H8jP>STt>4F{d{NynN51Sckli
z%KXCy7S1j&hxG2^ADTxru9UDyy@&RTU-70!W+w}se}w^u0I35T$Yjd3;m{Pl4w94y
zOr@bNLN8IEM{<Wwnh}a4fsssm-C=VciyuXg@VSqfk`|%>-SxiH39Ncsu?F}9HYBOv
zsS`oQ`1nlbL>Dl*v9PcEzjwLjfHsXtmffBDVK>AqF|3lRROhoO5_sSQDi(Ryki%TE
zX|a0<JaPnz_;QtRV|EYX?B|q2L2tM}3GI?}Do**}vh|W;3+~eU8I7foc1jeIDpSt=
z$U2lKeJEAvcR{TP-9oUWO^GMCeBm2V-WThY3VX97^5cSNn6_;KyB#x(JNjvg$!(;s
zy;*0}o!4|fKfHIbc0eAFNQU|EFvs@uBXN4H$$U86GE)4!2Dav?8J@Wnh#({`Uzqh8
z)n5ZHSFM!<$#*CAfu92xcLHaexUp_qt*<lBBM1f9HaD~-GC!O07Agjv!j0X*@#a2I
z{1%GWauw}#HJ9Qc5_GIqHz0(9#Gv%|WrLiDmT@KTUa(+Z_D`#DiocnUV1mDFb-9~3
z_%>$e?C3v4;Qvu_`0~ee0B+qFwm(A$byiHsof$&hIZ^9%1crb3O6KR!mY8+WZBEZ)
zdC4X$>1R;+rj{psk`w{qFP)587i@^J!hz0Ah|nR9F<#Px-jDm^os?~a1mPl8+dQ*U
zyycl{x#h0@_0-X`s_9pNo4(xSG^3fFm207kBTveuf^D}D^uaYHXz<JIll!D9G5elx
z2Q2t=Qwk0Hx%f}9X+R(~f!YN6O!R5rKC93Lsnc`G7(f!h(>2If0Z&)nm0fcmcCODd
ztHal~8~^!Mx&J5!#LO^BpNX`Ge69`H8r2ln-avTaH_ZYyABFob^y3taNw0-wi-wfp
z1?2%3E2@lSLsBb@CXOWvCyUVAK}~RqA9kz3#!i<a*m0CT6YX<Ms-HbM7K(%jFy%oO
zKF=}2wXKWmok8BWY(+li7f{txO_Op>tOX<+gCf4o_3*}sP>&s4iW2jl&ZkOad?O;;
zayhfkCDQvBR8Yd@y|HzG>443BDdy*O#E9jD>KUCrD;Y4E;O31b|D8W4s1vcil^`_D
zYsUEJc7CoTA&??dcN`p`N`~TMWYs5uefb7>=i8K;8hpA1mX*4yoWg?lhN00ZQ<l<g
ztSNE71P@cmhpr?nrZPzBy!qw0@hM<~6xJOSFh)jrN#!oGjCMk=;K}@=Fr=nAaz0*S
z5$Y08iZ<5d5m-ySOs0P$({eZDMZl@Njx20DD1hY#cn!!Z`>M4D#9uKQ@YNPuYO4%D
zJ7fB?GL_N3Ih2DceAISJLdfJ44pnZ|rC4!QTjxQb+*fT1$?OnPRej%Zi*=U_W{Pg~
zN!ML_B5d(Ovk`B?`VUmrO~#)uW%hB`ISj?lL`kWkZg`s&gA5g9x|N;zhe-~r7Zeta
zzkbI6iFOeUvYL<Fib-1oHbfsloCa*tR~bOjK+c89fpC!;#2e$ETynd0iS4gvdma_1
zJtIz%{g#R^-+g1aMt@*DrNL?p(6$Eu09hBm{Rha}P1}EHm@$ynth&}w*o7X7Jn&R-
zx$w>Y%`jZbSfMqKpdAjj77TGjnQRfi_m2HQip=hyJm+Gbzsp|91zKuqF3OAc-h5Xq
z0xs#4WYdz(HqZU7>FEi~`D~Z_;nPGHRP!|<<uWqrDaLIt7Y$+@qB#v?(P37$(-Tk}
zvalHj4Kst>*3^)zHg%+Wu6``OmlV5N+*Oto&Z_n)J#7>vAgW?#{Z#4<mO;f0*!xqh
z;<rcBchNj=ULF^%AxFW}ihRraE={h;4NXv06(3MpbY9+t?qzb4)w=_6YZoZws=bKY
z-})qEesbvid#`{@Cr~fd6fyDm%k`5r-Yeh@X;ApkT1@F*o|f;vOKXiu1lV0OE-m)+
zqMTC_db6opSPKT7g<7sW^^PRQB4;ABXM>tpKCI%AF$UWGd-j^T?163whzYBjG|6KK
zgx2GS9)MM$n4^*K>RQm2RHVk`M#zBSp=zA+H+L+Y>$B)JWJHC&gFyr`EdrguyXP4e
zD`6@Nl4fvSMl@X`8mc17EeD2-C_o4>Ws|Xsu67QAe0hMY!k_l;iGJqAthnxT*P}xz
zx=@(AOaq;q(~X{z+U4Wca<r)@aNT_XY`iI&#_%N_`I_<vw&1%$fW1zsxZWrHYQ+$7
zE2u2>POB4n4`-`$<^jtH^B|-TKI6~g0zz8mT-H{}x(;-mOu<l*-q7Rmvk7n!thFVq
z@B-3H_QGeWiZtcgK<;o{TBd-+(^Nw@uS|ow$%zSkEHTL1kb_~b_UvBc#I^bMctHng
zQQWxmS%DGrXf1B>?8Aqo;wNv;xQ9TZ=nI9;$>p+zD1Ln#)8-M__iuvl|L9j>;HAMa
zBE0^^>-TdbaACb$nvIX~3%CFzgzT_0aUtsIYiSW$4D0b6<nh3Hfek5jC8s01^>iQ<
z+19&@h&D_b@{P^WDM?pT_PcmIH4hS^7h9<cJ7f&%%!tvHd?X21qPzd)L9(~TLNqE5
z6g4_J)8@Q$*MK@gzp1cNjEoL}e**Pj#V>p_RAOTMZ7`P@`(##XfZ?Yx_ajlBCzkrK
z{1`$HkXIAE_N|`Dq-FXAh9{WP!wD9jKDRXVTkyXNFJ)ZM6S;&c3BST07I&+Xvmw6p
zSTQLwGEAK&hK2CQT5vCmVOQW^@DyZ0ht~O@g*8h-G_s9H;_cF6t{LA(9uH+@a=pH%
zz_K5&pr#M`<E}?ssj!(WH(cfJ&;?1oV4ny;Bu4t?=U?>ty37IrPN~IB%KO2+bfbf%
zU@#B4)y=S9D@k9|pk_gy<n+$mZ;WWAoLs|*wLP;0%m~WjY~5i!xDG-~6k|#F_z59e
zzrgebLx&SkSX+VC3ne0YQpF5lt53c;T-J=4+%YjPnx5JFx60rDs0)52jT|&sjQjgd
z8s3dyBt=4iLMfpcA#S`jleO4qS;!c0g=jrImm#!TMo(<XNDe5GwxoPAw%Y*eE@!F1
z9l-QK*iEYLnV@xc#|LsOr6e&?54Jt36js8f&&M?4Ix&7;nQ_yvMxkk%!@$5mi12V-
zr4RN|jwDf%D>*qe!9~FV`Ha(6mA?A1t-JHq`~xtF1XA9JM5j=Nb)9m_CxM)vcQ$6H
z^MZgCdV%Fl1YTc?^9_g1wsMLOaMOTBT-1Qdn2w3ANMFM1iSIT5`Dextn-0AgQK{h9
z&$>OlVkS(?J+){WyNmdHU%l2U2v?HYzL6{H*G(Dn5z*)qb#ZC_$1#~=Fp{m&MM%N*
zh5BocI}-@O*|69C@R-NrJY;miS)&NM`P8QD;}5Q;;Jz?A@hKysZ=bou-_nXsv<Dwi
zvMz`q8<I33+Q{hDhQ8m-7D2cC{+Qike9ZIaFGSQ1Yum20=_*wvsDa<|Ac*Ux?5xlQ
z-$w=7wlSKR<2j4qWPAsAY@yCCP-q-Ydqi5{v@^Gom+RnkrB$G?C@~6c-odsE4CCLu
znrteW0&P7_6$m6E2r6o9;x3cXPm8JQ(XdKXU`%N|YD7w>zrS1dMrNprZ1$!IZ8FVV
z`SrU=0XVNIuNNuM&!9DxC=Qu!0pnIkiJ+4nYZN>s0rWZ8M%IjyMb7I~&RdqX<Xn&5
zu+3@buCP4`P-!l)Ol(VZEy@efdIQyz5;*}iJhJYq&0<cOzFK{5cpHF3tEoW`4tWOY
zkMBVIOEeYC*UY>B3v>l0MD7Elty=8<a_Bn~-Rw?UOg0Ciz3stL+}(%5o5s1L+pSB_
zi1(zE(`J43M)UQ4fuYi$P}&nba9*bvT~Gufz(3|97qA-XZuQc^IXdqE4V#SvAU*dn
zezkSyW=fMKzF@?O+YfxqcglB6(l5pSD(?S}9rBN@0h32_Hj7<9|Ke<#@gyRPf$pOQ
zjOSvJ|B0c@zm51&WFo7DE^vwK+BM<ac-u>mJ=1e_b`=yF`D>=ZG@A-A4XqlmjUruF
zh!Eq1H+fo{M^URkI%*fB*#gcAjp~6jLk)0ps03dSr+j8P=q(Fak8uHMg6Qa1d2#|g
zD);H6mNg>&Uz6Yx*MDtC;|x~%BvInFN;xQ7Mn_*+JV0%4%?AnHv+Qpyd+}?ciAqbI
z3zt(8N`Q9cOwOk#tqqne#CV5SdouG>?6|)X4H;@W!x_L#$J4t}XX(NsM3N8jXkiKP
zperV4zBbAF5R;XH<9T#jRGX~yN^NeL_4puY+z8KOUf;l!WSk&nkbpRVrmV>^LzLrV
zpCDp>4^UuuqTRm2i&m`HfD&JK?+5AJLcJg^xQ7>^lqspA<!!(AfJp4c7%n5k{_Ds+
z=sBZ}D|S|pxldtq{l+;7qT=HGbXhmzz<cG8gkiFw2AId)hnSt&dMe<*yV=9IP(?m@
zLFami$t?4+-u(zhngS2OVuq@>QiHir+_J9-t(v`z9NX@NhG3GWu;gQ^HpO#UbQ{57
z+$kwRnL*JQGOwBn15T!v-JYHoD3MlMnY);KIO<LO8rB$g+;l*yONqy)dK3c~dH@7(
z?;I0uoUrR{#a>sp(UM$SxR-k}@~SL0zYY{(=^NwM1Lx+y4otUour3$<JYmUD%nZ-$
z1+Ev%Tu+h`m!D)s47i|Rl#-<?;>jeBH9YYZ41m+Yy9Wi~pPleu`jZZC<#7I}lUz<%
zO6d!mv^~?#-5L^vJ*7m_rtdw<5gS~218W}^H8^~SUN~!s+1nZEOB${#0F4@|cRy9h
zc=Us>SMu;`8F@GpTFvvI>QMNXCz<+aW;Xzxb$91^`|d3kOpUUnR{aQHa{?r>!I3N(
zQa)WF=KZ*`@?doYvYx!+QFvpH7>5UqxaKf(iBhLQ)S$%@t?wuV?uCK%-+nkUazzlW
zc~<rK{P}vgdZTPYF}h47Vr@qIiYuvnR3y?+j3>$2>6_b$+#0Yr0jWD%?N%yoYy^u1
zdYaHN3*V9?&YnbwGexVLoc<oy$6wCiTS&&=Q+?6x8N^Sl6CI&og04v(Tbj<+>E@$t
z4QM^j?DteZcl&xSiFXH{yGtYX0PE%}S`M@z6X5FVktOQ>CAv2u8%fZiA}At#qjT1q
z@Q1xyti;|IA3cD{&QjP5@(5pU1qW7jd&UzlhWANzY}2*$NHHx*ey~Gx_=}rQcGuAS
z){!&(o7)fq`~|3W(4WBX*n=k>uRIt5y~gePrr#n9Jpjkq<mnI7gg7pEr}!iloT#2V
zJGF_lTGEE`61T&C^5=a7=1mR2xr2jR$)-DF`q~n0YxhYI7zxH4uRPLuS?4t4jhhP~
zyLA3#Ns+zCx6XFWuSABMfO(D;>20M_FrjQfWSyjZPa?}%g?`QMxw6>T(qqW;CCR*g
z4)qJLdl}+Czs)E|V#1$-o=wTYFoFBh*U$Qt6J>8Vyc<+xpZ^fcOLfj<i;u`>q?2KF
zdoU&0sjIq)diTIRO2@BX`3UkR+29dvmuw9$oeZb0?ROOI3E`;g%Y;BBg!r6MRpFBh
zvAi7TncGz{&MRGRuX62Z%<5ESIYnxz^A(AgCV<ld&B*{#OwMeeY`tk(<#ibwO5nLI
zZ8n}wX9kY|icsi9@~!ZT{C<0@>inN4D%0vpasB6rLn`Oh`dd<ZYqXm^WJbWR={2an
z#06${OJyV%eA~D^l&J5#2-`E%-j3KZsK!NJ%qMfHd2V6y?A`He;RZ&x&u9nB?~Q0P
z?&&Ow+yw0VaprrSK$ej=**$9)Gqany8}sCZ_QXFdGBoIfG^@_@kwdah*OnJ%?e{z%
zLO1EFKh2#6ZOV&<m{oed0KMd;A9!o2yz*7cU;0MtMq3tUT_YzxZjOSS4}UsXB$r}S
zl9ow;m@QyUd~f#Bbxv}5`&@40>7jDcgZplQgJr4wd0V$=$@ow>HCy$rQ+0JePct{?
zL1TA79D~pz?PB6er}Lp;*&|}%fK#g%H}UH@zV@-dq8_3DU$z${jZFf|rKkKh;E(Mk
z;R|6=m}mSo4w~z9STM0{eMs@qirYwuMc?-PJ{XV4PBeX#`EW_?IJVgyKkH7^R~c`Z
z=;6CHM4=s!*(@=R`1MCcjFEX<>myueBFMFY`B90wyn@)Q;#wq${rCi?Y^t$jf<D2l
zc9l`_0ykMadMp>>Az~JaW99a$0m{EK(4x?>pM}xFVK4apbgjz6&=XA9MKns&LlZuG
zG-`d5rkBRiavN%_rE7MGP>UPIUk#@b<}V3F*n$p@yi*?hYiF7n=tx1Da6~;pO37n3
z4%(lHXvH;xyJFW(AbBilKx}o0h)b*PmE|h|HbUyI3UnJRrJ)Ea*3x4TLPR6(^7OAQ
zif(6kErW>W1}zLb>`Ea1HP-EPsP<+`<9jTUY}c}<IDJF;ZhS0i!dkAHlj>g>-l$?A
zyx*ZsX3&xvg~EG8pIBaH&!it{MVIG)dYK==hfMwXGV4-BJk_d@IW%JZX8!R)IeyC#
zL_3-2!<s`9+e+4U$b_^zJM#9B0E>05#4|;oa{HRWGI8^6icl%Z=UUI_@lly{Do^j0
zrwvTQgkLL;e11x|SdNGKfxmt0^Uv3*2TQ4w<~|B-CwRYV=3J9E_3(AFIEG~Ju%~+<
zykC_W-36o8#g1m6NI4IWtUlOEHBdM39*$?fP4$pV%HuLl{z|`TBox87##k@bx%vi0
zOtK$AwjXhA3Y@o4WK?+mo#BjTPkhfcSD$*T1&+l;-%TAf-fSsV-9D_T(I+<d5>m0|
z>R(T_O1)p(pK>9&oW3QB-?yrtZep1*k+G>KR&9CY@tB26chg<{^2lf?Q+1ZkkxZz;
zDPdv(R2jAB*OH^ecrVy#NhB~Zv7!Iv2U5ukZuo2brRhIjMQTVe(yfZ)2$LDBHcE)I
zS}joCrBvnAFV&}qoIl)Dt2M*AA+}*nkszP?d9~x<Nc3DHcUz{YAK|7eii|#C_fLF9
zeSVa84pYkgejya#aWoP!ijTR{@iL*HSFuk+j7Cdd_w6>rvM0X(N{5S0PGQKwu(M?=
zoF*rm%eXiB`TpaijJ>zy?)th@3$7<_34xzXb<~yHnq~3ppAE5w3jzP-Zz+>)NeS(#
zKVNwa-~GNzHlfLA*M|DCJfY8l|1z`o(UtIrY1MbiGxf5<bM!2q3L?eF+tS+(hF#o0
zO_WZU|15tccB0(zutt$g==0-(h*D%~zu>UBo+7wV)J4~XzUjS6Sc%Pip-Pn%AvE9|
z53<(ucgti?9tRchhfaK)0H(8@UwLwDA1CR2*i%kz$muy=dhzsyd(;o+S2Ue>s}>IK
zAR8QiwaY7>;C+15p`n+lJOFZE{bEkdU+S>IF!6%)i0blTqUXg!3nh;E_R~6ahg=D7
z^oC?9&+~OPGM-x%_t!}|W>@E})f*&b@n`joI%kUPIOg|_yot;1m7L73M<w|*?4VLc
zCzRd&->MEj?sexobk#5F%X$>2u=!LeDMo~-?|*&ZgD^NGI{&t;)L#qYG5LMqY}Hcw
zi8Aj|DsB<nOTq<<^6q2FNs!KqVcQUdKa82muxA46<BL5(zXkv1$^80%|MV}<2%y;3
z6DaxBKUdn9cLc!5k4+)G7MSsk0UH7XyNRC<^re;PysyIgA9_*#l2dbfPWkc%dhvPB
zVeQAX**M|;_k{|zw8ZrBBGkL$CiYio6A7&ZrVkrH9PwRYG&hU#wqh)Lc1-l$l5U<6
zFq%|&E{O5^|6}jX!>L@`{_%)tm8cd<GNb`f5<*(0P%_U$rlia>8P+1QsDu;>v9L1F
zb7Uwg^E^)>V`QE}_?>s{=Xu_}pWXf*?{WP8eE%y8Yq{_1zOM89oS(sc9!)}K&KYAS
z#J@NQZcQ(G#(+n`OK$!4-kf#axwmOZ+DHL-mE8F}u@`=G^G))}i;)%$s@8w|ywI4Y
zk&=-dvuLpU{6~S+e3BgDDILRL-)w=wYAjY6ZBh%E!BRIg5O+XcT%R@qzqtLTSXUe9
zfvmt1FFfjtg<Mw`(5B3H@7bS4PN&7Pgi%DU!8oTFq35PWv_0G39x-CZaQC{kECQFL
zf89Q>(;V}_W9f#8_(e9-aXR~8DY|Ts9gb&=^TT^883J`+6)I44e=&^83&0199vc)2
zpIpoV9n)kPn6D82n|qm!f_s_9U*-DupXBfBOG=g@xH@Kg{;u}?{nz+z!JQ3wvnc5F
z_m}){wymEFjFPl<&mSx8UyJ$w@n6#q$j!4Gvzz+;r}?|unTG5&)!%*n_g`y%NKQ#A
z@TGR7{nzz1ogxKfGarE155>+i=5R}}s^`y*{rf58)W1J>m)1b)70bU?_n#L4pDJ9V
zhP*Ev0>54l5I5}3GIDN_qVK6WL$HM5pq#OwN#jAUJqKMN)X@@-g7Bs?;2Na^)hWD7
zM4ZLS3m9*8Ku1RT$FVcI0W+uoUQXP)z)0=&<Q=^M)(H5&nYbSdRRRFeS_~BD=Uer4
zhH2k9It@|3&@*x8ax2zPEi#O49_8Nun};+>4nAh6_W4(|zdmL@9inb4g>JW?3A5Sm
zHYhuQl#L0tHPAEMz5B{eIzIaJ1YYdHz7klQ0>Dh#gbBn9;-!dq-EXeW9Q#os==F(}
z2XKsO!1$d&=lJPNnzM7SSz-@hmX)Owh-(`#q+fs9%w;9bKj;E_*cq_g%m<ZgH%O^$
zz;S;sYddIWKQ9wLALBDsvO3U{RFIx$5i)l-AZ6V)=kM1-^Aj?-gT&f|q@S<L>kffa
zDK)XYWWHjbd_oM^=HMtJ*e&B0kuB563r6cbNNhAzlA9Iz2wPbT1c@k{dL6Puqzooj
zkkXS6mTWc7V0ZQ5vbrcZtL_|=(6%XCPXOk&bpp>6g7>T_l=zG7B?ZQyUpla)jL`3(
zb`;G60-LTcOU&m@;1{Mrdt1=1awF7sxppxAP^NMjLEYo+@c!RC$F9h89KE8D{PXpQ
zn0F);jD~Y1OM#_;R>&tfA`EiMh)Ig=kMss5H6i}qEJ~WB2snQnm^sokSs6Z&pY23p
zgVTjBO!A>4bfTF~B>XD6K}=UDNjr4{kDGJy7*+Vtkd=U^**CEH%fr*@uv=)#D_((3
z@OHX|?0L=wSTx$I=0+jpf4^=bp6wEUm!KB;>-8AzeS9BT40AJ-P(g4p$VhI8S&Y9`
zyH%YOuDSPP(?J*D(OLq_G8?fj0@t6?SN9^fZ)CJ27*vJO{QVoyT)P2vj?Od2U#E-L
z9-8m4G-d#TrED7+fO>2dPNzO--*o5YU^qh!xq-Ux*S>@@s87J2PR%BXLt=9wM_C)_
zfL5XE8rk6KTLtr_LpC~hgJ6>O;%n}6|AfeX$2R%rsf*D~5&U)1fsAIrw@3u)d(G;g
zr>Fa9u*^3gOHmO|rdTqp$a>YlA`!9K>8Is{d}ZiX!;t4h01QBswnH^^YLMEJhblE;
zx-H~ojmd0(ku%tyzD#In=Y`wX`{+#c5vG;s+B3#iy43u@VAZju<-iV*TDl?h!3=Sc
zOU`hrhkT|+fo{Xy(C=-f0dKV?sL?%RsZFgVqSygqr-xsGJo0#LHaz)#8oE~MGwiYv
z_R3Z_Cej?5(Rqt{Ap4#HFn6z&R006v4Vyn)q>9EK2B|h03t3%m5av$Y->W(E18zY+
za3y;XZ6^fR<XM5rVR)n4Wh5Z|w76LQSFlU$fp8j0LTrlwn3|X~P-Or%)(L4j=8Cz8
zGJ@>ulsUR&ipG)(avod8xK{y2G6RoME<nz{7>E20jk-Raw)80mO#=<n>w%PSDo|&;
ztqKS3*pw_tPo8hrG|hVl8%95!_U&Maw|6_$B}8MNhl3V#0Q3{9K#}bM%^~eZ0X#dH
z<rM(_k^l$m1G{FwdRKb4MKI9uvrxljO6OVt7qUirjd#G`DM*RA1Zo1h-YflW7VhWE
zfZSAplcl%`l&A{6rW8MfDEKLenRzxfB?xj<`t;prOu&;|`h8QNrhp?Z$ts1QGMHXA
z1x~JcNY=4y+fEAgH!vD*S?zhbxh?xv`o@+AT_MMngoUURt|HODJ!Ee~Nmai(^v8A@
zAUgzWG#5~|K9GqVB;AMN;z-gkZK0l&IqswuTUb$yfGvR_9H;RBM&7z5q^BPkCYQgE
z!aPi`Yk2|jz(8P&4&SQB9@nToHw$s@I(Qy6^tGSj0OOHN&hJUhz7@MrYm@!_=1)DY
zYT7E0Tsv9{$HcJm9n=u5F5lPyw10F3n$==P)sQOD4eV$p39lR_c4?B73&x*rA0LM#
z+<ji0<m2eDh>Ld>C@qnpVr&MQMwvy~=<F2(u=4x{Ziam1%g_7^M7yCAl$>0mt!Tf?
z`6FIw@X!q|BT3)=*|3C#igd?|rqaQ)c6B63)Ynbzr79~Op3CVQ&|iBH-ZFA{+#&Rz
zK}8{p#i-T{&!GXt6$`L{5%g_5d*CC$##ah}k~=!fr2j%U(3>KCtk%BIS;tAIuC1}}
zCQ%|818CZz-h7EGDKCKRB$$^A5BLC}6xxpFhE}Ooj(Y4osm(8@R9Dj6RGQUR1TDL>
zlnPzX%$L(r;c%~&17+V}TG?f?g;Q!az;wy4Yzu+_FXnUOUH9SrJDLKRt`wVgUphbv
zizD9|a!v8>!5c%sBKIjsF&v1n$T+$oME(WE9onooS>m`2ODY4mv$e)*n}AVV>j1x-
z=F!&}t*(IC(vNg26UCZS465H1*JuW))ui#~oPxq0d0XK(=2SM$4h*6Xc*4G1xsq1g
zQDlIiWgh8GYM$v$n%eD(?5|uw(n-I+SzQA61Q4x$;%KaJbE)E*!+`(u8|pK;=(^GD
zW5sisuPxNh#e0bQ%eQ@3ILPcT^Sffq>jdfcZuGuC>J&Z|5^7f|rdEFuGWr=H^afLW
zu5c)A+?s<ehg(OM`ruWj0)LU$2e;tB3&N5St7Sa>v3#$qJt>30faELi&;=fg@S$BF
zCQaJ4x=2Fva*ujIib*4BwHZ?@H)<4CEo1&zWR%8L>0B}Rxt*XclftOnVuy~Oe*N0@
z;Ghb$t4qqzg}e1H4}#m2LNLXFIH>Bm7?+^L+F$kbQP2lEE3m;0rdU70tj#fZgh_<5
zO5<oFu=lFVy`{j5T5^;vLkg3k<_OlwGEAeW45l}aMbhZkDOI(rg4YeU7hhEGz8~x-
zC2@50P*>G%Y7v^+VUp*QP<7S!Uh)`@4F!Yj7TA9ndmmDqWTO_9VjA)X6gg@#({sz>
zq%QUA4{t}W9AFX=)G_oTKS8bFJCm2$D4qbse4SCY7W|?s`)M14^lL;$a_XD@p=NnA
zd{cRJi8pOk^NpAn&M(Gj@ZI|GP>I6)MT%xeEQKx`a}oDqOP54cQicP4N1~rs9DF36
zMGk>V{)`zB!^0V_dUu(wQQ>N!;E8_g?xMH1ONL=A$!D4^z#g6UnjxcoCE24^218{f
zy(N`17eGig;O=BvR))iYYSVrG{Nlk=Y|Qv`dEEKhxKHqB&^G{Ew?+GIl=^%<XDp%l
z(N~WYi)*-iIW=oN<6_KhJJ@2RnaG}CuXcKmHOn%mUsNeibANcV{cf#zRZHmx7%EeD
z#mw%*^jdtjlby*FaXeD*6wRcze{B(5C8RL5cgJ-!2h*1p=H1@-x!ymw7`%6>F;-=&
z^1*Vrd-7~7Z`&4^^@|1L>{aa)lS8j#Ve>4KHa?sm+pQPG4}0jG!R)=q-vvcZ{JCEK
zPi2m#Z)QKi^+%1@Oh$&~E2&Q0IGcPD17zl1QTDYQ!QL1Y?MZrO4lc4FtXg8VOv1kR
z4@Wnh)^B>1MUat<ILvcUOBxOVnWJG}H^-5Am;n2%sgFzelLljS^c~@u^)tLSD}bWm
zkc*8mjAb#j-%QOY<~;mVS*$Mc(G#@wLwa&WN3JMM`ySYm*zkf_osnu8v|Jf&^yZ{W
zaL^Al=~+yi;VALcm=%#$xQz5!k^hwDM4TaB@l@od^LRcNEtUh#*(P+Bi`X3REs77z
zt5u%t=GS}S+;=Wt!NRTVW5H7JF3GoQldXy0>U-r@zOvo*oxb{BzQe*gisN0%P|(KS
z(ig|RwO&BeX}G96rR*|~jgJNFqc%2&e{?1Iju5It)vA*Noi5-26z+~Yh4lq?G)g2w
z$IljuF14&kJnfe;YxwpYtFm{?r%{H)MF<fu2G6<{`S<z<4biW&TNT;hyP%ADiWzgG
zoE+CT?!8v^1`6fM2{ei{p~WB_=={UqQEZ%VLfx4buE?##4iRU?YvNU=*jyDBaJsCk
z5$DXbPK&0-Zg@mVN*WNa`;9KClo3LgqE4*$h{*h|zhmbTqshwJXYogq<w!b4M#3-L
z4|HuNZ!vY+oA3Dod^Anbu_>!(CrVX3Ojpfp*$8EMLsgtt6Pe@qdF5y0g4fYuDwDE(
z>iMQS;I)>YA|EFx7&HISqq>ReHo8~)y=XwKbH&~>R~Z75y5o%3)m{ZFjxb7J=Pb}G
z(Kl(El)&)c7No`nN}OjHOX=oyoNP9Qgn>s`)G!Z9$y!W}wGx8jlVjBNLW$i)bh-q%
zXYmUDQM&S7XOoDwDO0gZSr;*Z5+xW~v!t$mmLZ{uSBJTCs!zmHedwciQ!}kgbmA7A
z(mr*Q*(`*pR5ZQ0k(6qW=G1kXZbcm(fP6An*2V9+{I}wzLmKor;|a~KU!$`jdfqo!
zOFiti>1B*ffUk}vTAiRG*sqJu$uNH;vlik$`%Kz1o&}!EkBt|%<;aUV$_K%Uc_D>H
zB-4LWaeTKdRz;SEhpKM$tH9MkW={`Eed2-;+AVEnE!VzG6}^G|JqSL!ub!yB&BpWZ
z6T;pL#xn1)d5lZJU(D4HQHO*pUNZ09L45B+H#gX?kBhS{CEGQ~>`%4t%lo`+hE0bO
z1$x1!496F3JU1id!+Aeqr`qEBHj~T*Y=^`yZ!Ndw7A)-YLtOh~Eh&yWDs{RSofi?j
zjadv`uf`SGnCMe8WBNB%rte}--L_q7mKra;rHF)C^g&9W88AuH*U>>A1@=uN?h2Pc
zg9GQuXjoyo-*UaokBfvuJ=l{ewq;DCeAGVSifmR$eZTG)GfpfSn@OeRBvKq|NIW4x
zeJw)qtOdq9WF1{KZggYGOsqlDcs%xb$yOVe%;)zOZ{p7NI&(PbRJL&!_W-xIq4M=A
zm^DWPEvh?SF14#>_48Ss78hG;sp0LjhB(TERUFi$;k{z&bx4XTl1gE9=RG-ny`|32
zJ^bOm_NQGk5n~VQRHulF>E>rM-o&jxyg#m5)49QuS+<d~{xESBj7EErM$*=kTD}jl
zpLtNNJ*O!ldXVsRTk8i{)rV79%dZKq++)(`BTQXdcEEheKzJ7g%@X@)OMJ1f#g^vv
zzGo^Cj(7%Aq%<ThAE=VIexczx)>a$oS6#VALS0(L9J!WcrB?QU8q<xf{W9-yszYYc
z2UI~dhM9AM7+$-@bnm9N966>E){X7&OUD)&0-9o*jjhU&r#Y^9BKN>=dNeT>18CC9
z_FDWxaZE!<<M)sx%16?9>tT8Wj=PQqcZ8KfLCC6<KqHUi%HN7Ddc|VM_(r7%qD-FW
z_|sERVB62?f~LU2u>@&40NJDWc0&!S=d-{0()PiZ#ZQs&DnD5<!14vmPAGjWAxEL0
zRrYLn(C5RX=TI~ollK$eX{w8h9dDKi&ke$CRHIkdp-*g($wjf!&18zzgDlA$Q>&=Y
zb2{DrHf>7S`KdUo7b=PQA~Piow5w`IR-q8s@AbNb>d;y2YuRIvJ2<w7gtl8PYTeZ6
zQTjg7mEh(IevEbhn>1(LpEp%0`PjzBwNysDMV7P0G2Yv1d$+Et+!z*IIcB~I+%xev
zfy+8wfg5&&L!W)Aj?s~h)NC!^WL<=O+g>PEqc$}H^b<P_I4F~LA?}qLGZUPYZ@~4p
z8-h-b1IDrOaGz2iG`?XA6Vus$y&YeV#Cs&NFR!Ql8-@BSd{#{lV6jaF!>>}1Azmz5
zT#R7e4n|2AAPsgzjx}y73|PRv$^^*o7@rc5AKU8PZoa8Nqr^7B!5)9VADDRBtLrZg
zF}lF;NKX6!ej00D$q@BGIz!fX5AE%KXyYm#dZTume~=0Ime7)jW@<OR&cLeVq^(Dz
zbS7qo_COoE_J*QS`1$f5t%y578>I?J)I2^XdBbifjnBMAEH>?nZ#Sb&)FKu}L%Yo*
z`<hgmAs{ye5!s<)qx)H6q|Y7`^(ZNY<4AobdPN^P_*~NHP}~1h@ERte3vsNH(ol(`
z<5WEpw@09m--nCVpB5Gwd2&&%3$LhdQGDqXJwDDv2(XUPDF#u+)biG?%EV_>tkU>=
ze{NLWS0=!y9@6l2I)~~o=Tng;n1tt9JS-F4{^469J>tN=tkpMncj1EUy28P^I<~(@
z`2b1cC*3yw&(=xeg~4rH#W)#VKeaF<me;9cAqC}_L+IwF<J(}rnm8bGWz^2JV%uc0
z{jqs(8JO*%u6&GOBAkVY*n^EabIj`Ez1jZi-3~1<43AjT=QxWE`B+{4SPJt@E#RK!
z@$c=W!G62g2ogD;Hcwqz&XOH%W0t{q-yChDKQmY>qN(j<%|uq8b_25xoiY)6G0fsF
zIeVw^B8uB5sCP?a;!Q`}$nJJlDMBlRO|xfevqtWR&_pH<A-Pm1E(HXFB!-hZ-g%s@
z0q?fLV%S-L5CC60&k=o0(lZ<Uf3;D+4mbfP&5i;lhJ99_68B`7=f=$;^T=;=s=Mh2
z#zRyUV5s|Q%_DyKuy8i);@oZksVp{L&>tpdN`r^9$s3B#@XRkv)e~V1%*@nVo0dBB
zoNwbSbuJXVILA)yyR=&_1G?Cd79|^qbuqj1RyHN5{`&f1YJFdam1M8ysj1j#Uz5e%
z3~_ToqUEfaE#DbYS31WxwTlKhP&mx=UmuWvL(_}oe4P9Led3XLfmU%YIH*xY#YOHt
z2Je%CIsguE)GdGqFs@H}1uH(+vbmMN9;<yu9UhipN@isqb6_nmgT=3gwluRkS1%B_
zh)8VCq;vcpibj#8`s&0(f!N|taaEniQ0watv998k2njhj=+Ke%q|4-MnlWSW4%F`-
z-x3#N#+?&XqkYn{fU6y**~7IgJ{Cr7zF_h8+)RO<c~s|W$@9L{=rzEiloHkP>FBE}
zN}421@T4eCGa|#2U_}E4Z+y+j1ZD>+sW=ervE6f;)MbF3ipcou!6`v;+To{kt)y8C
zRl@AO?{AdIPM&dO0^>-NJfKTyxKghK+dc@Dcu-UaK*UjCAMq{*llZTe_EyOl2Q<O~
zS1p6DzjBqb=Y{+Ezz6q>H8;`S!9js9R)mUQ5I|kZYWq3Up!{0w?5GOIO0Y<xiuZl_
z*tsAoh7CAQ*w(TI2Af*HD+F(@m;5mb`R_yJue;ZM31d)C+Bm|QdiCPn#(K^N((H&!
zeH?c|_X$Lcn)A7BANgUltSOjUQ$Ab|@d~#>2f_ewA`U_7%t4EXz@M5&?W@0ORUD2g
zB;~EyLEjM^eJYvL`4ty%Gvwy7&h$?T5obj7#GYaF!)7;pJYVe)jAx$vQ8t}XjArD{
zZ&zu7rifV&=Gxi>qF9);Vrv;BFFo!pw1E=u!)Tmx&+^d?MWP3XAu)LUi_?O!wIG}8
z6Tp?vk*T1+i7b3L%!c3f%)fR6?zbPy7l;|6AAsdjQz-fJs+Bl_u;0h*Diir;`vGk6
zkgfH^D+dL^!u}QtCv4PPX>hf9Pq9>S(p(iqFdr2sXb4X)jnI#Uv0)U1f_^ohuXs%j
z0x08!r(5H1I6}CzG)Br{{3(<pL0H+LYq3v|+^P1+UXvVVr9MQCTKUi<(lY1zGGOjw
z0d&a)=`>1fvjc4SM=0)}Fv$UOA6Pkg4UR_t@&bTWgr7<baYA`>oGb=5p5HcCk4;LZ
za`!p}x$Lt#yY?MmET>84cFpf0Q`%br1s&P#6Un#@zWm{$jGED3U5J(bR(OS7jq>7h
z^oB*i%SqL*zMG>V9IAQ377HA0MDRTI;fYafU>Xb5PuvALDMU6s&*Dro^k6>_TV%(e
z)H+`AFk_}Se;-K|2-vtm8?5Endi{(PVjx0E%M95+E#QA(ugIDbve9gsx3IMc`zqB|
zMDMMVa~>h?v8Sl&WG6S(3w+C@oeqp)MccN`u!+rfTEE>RQs`dF*lbc?`gb#0O%5i_
zCQO(t5XlbysPfcx>QHZ}Jhuu^{7Xe{a}4U~`lQu3>j{J-v?Mz%n$ich+VzPdeAMW0
ziu(Eo*5h91mtullV=4j9uw0tI?;68qvtzIMJ!pHJ?LfrNe6$lgGp-q%(r>F!bE#m!
zSr?|1E_XtMeD5$hq9&$TlWn&b=Xc${6A2LneSm=5FfH;>-+Jk{RWFl#fsn(S8$j`;
zUPkZYaT@@OJ7Xm&mgd#jbGb&@6{%IVcDA<~B#)1`iQfTqduI#O%XY&~CFy=L^@-;7
zmc%oCmbn40E*CBI7V(%m83&X}>v}RKOyA9H$&j!fQZml!r*sBp5SgwnXWLy%$}yVL
zV1Uyx1jzfP=QV%<Kgd+veY~+0pyFX1^Oa_iBx9hTNnjru#;oCdxRz)lTqzqgw2a6t
zI#SdHDX4wODVS+nM0k$&%HEmy@Z7#OSWr$|vCgW`tiwzQuk;oRKt;V|(r|<tZqluD
z<xA|zrTIWA2EJ+$!#fzft!W=dINa5?3tKxs-%|Vudz<^1plwWYxl&(XUqayU`)*rC
zYR`zE1Li4x{qg?;w60E%wAXik*)7t2Y-9JzrnxTDj|<eSe$KJmgNbs~R$SYRdVK(e
zmAYO#*MC<;?#PB81(VA=0NeAk!e`Hz+n5vVHQjX~ysSZnH-P#u)7e<U{6Uh22LZ8}
zYMwJ~(S9C^E3e&sxkT7aVMdpYuyBXBBh)E9W9>KCQ^(o-A}A&rB*%tzmfz{|Jf5@3
zSyk4q+$^eMFDzf=u#?>S%BYrEOaRJAPV&ww&D(WKVgb~LK8s@>W%`;DxTH)~-PSna
z=JnK>lXHb`nNVI2mt{;8-6#qRZC<M~F|W2z+lIxhdbIrjf(yFOKm6V3S4j)1cO&}F
zUr0<dJNYuP&_i&|%tG>^B#H2rZ3go;xG2ofG%E3k8~OH2?FfjGTw=an57sHhcFi}8
zSn()YtP+btRhqvP2^;fF75YbN%@@^IIxuPCjoeU6U}X>09#*t3Y{|x56$yCAD4@Da
zL3DYO(yL~mUS@D_>;ssD_x~i<<h4={bG~Jy-Bx|6;&qhp>4Jp;^OY9x_I#S3(Y%j2
zpTp-Hsx;Bb4)@*h`6_J2emv0<_w9I;hs%IaaGvZsy1!lzO&2nO5VtCVF$39V%u(gu
zhlSS=Puu?d@7L_Z=L7XOuNkf)9aW?1j5sSFxe-=Pd?^?udH;iq!TtlNcv{_BZkg<n
z&O12lf%t-%ys12aiJPXn`0Ouk$3;D`lxLE!Bn6`7)CV|iU%Pe1IU`89-@GDL1Ydyt
z&_qKCc9ORQrTyu+)4;ZZb@D5*qK7jz?a<VvOSXrZi9eU%!OH^6E6UbFUxYc<l(!-`
zl<>*4s1Db!#TvV`hp*u8qnA~yMfCH66oZcy1ay7<94C-SlR`J3ik4RI3!op6fo1pN
zyNSz_%odf+w`mVTze{XA*l@t_0tT>t+m;$Gy7Q~B>YEmsMyk{5_~8zjDW;fhRWIS=
zl3j7L9}#Qe^~vJ*tujO0E(zIE$y@4-WNLaF>^{SyD--aUsaE2Om1*?FKi;~5JwT<+
zjCp3w!{r>f;MU<kYk&`X!x*olAv>4}5EQ@8?ApaBPn|&yN3Ne_kpDi+e-cE<Nc{We
z5BxC!R}mBaVbRWtcS6rA^GNQPos5wndDkytJ}Nkq$?~w`7)qRdIi~sMY?5=)X4qzL
z;`e>T3mv0^sEa2hT{trvg^Jj#L%4E1>NSp>&>zi(?x5=Y64O`=bEcDiuO=H;t@7r{
zpveC01e5A3g`*dqH322@Zm7J8T6liuc3Y8}`Wa!*twiI=ev^dab`PC)jW1Hu4!eUn
z{V_6o7VN$V#4et_sa4O4vHqc@AoRsJWK+w$PvNaX9k<_<yi=S&>}t$vlCb(#vd3~X
zU4DSXdp*3d8lhLT>`uFYT9;dzS=7`KDpy>H?tv)J{)}Ub8|;OQV+DyWp83b|-Cwp9
ztZrC|&q!Et6=t@$rrNX`m!QoZconRPBPmg|It)1krFExmC?e=Tb(QMZiQ`^xltgd|
zhh?bh8E~p%&xaYa)=5-x<(JFuA$eY<emsdQUz;HZ!&jVKVbj5FcP!oftqS*ufKo^H
zN_0kW#A;**Ftpa7l}}H`HyU&mW-RqfRb89DEH#_$8t}UfVSWcV5Uq#)dLEjRl756@
zF2)p1@^`oSkFTX1WTDieqOtkwWUToJGM4w%`NH`318)Q=OgaWE1gDW&<-dMQk^_l6
z@$G<fUk{w@mS9q3w9x_PGx(+%@GAzxrIaDQVJDAU*x&s{2*%u=3bSgrvGK~k{$1A;
zSstH2%<V0~>?L7B2d31~^WAW0A}s2B)0Pt@<DP#{HTVf#H#;JXk%8j)+kg3{2C;v!
zct%4Vq0P>P=6^i>F8S|TNcfC=SKtE3XDV<U{D*J)4@CaQzr-#c0(Q|8&d9jmg6aSD
zGynHy|8+_JdKsHvlEWl__a@k@qybqMLE8Ao<7{XDa-0A5zppugP^_+S;1{8I7pc>!
zEHJ9-1=RO082@-XLHU}2@f}Iv;<5X3_3dPwKiA{9G`6547b@sL%#!9*lH~4%)OKJY
z@Nz}GxBy~Q=8&&qy>9C{gO@Zi2dfn|Yy2H;;#Yqg(+JO8=TUBof3eQ^$f1=!j-V<?
zjwys0g^kBfk7}WcS7rzLAZkbJOqC56MHH{e<0yay5#9t|bsNsv8GsK8!Y|&{N!3Va
zxN#w(-W4&`iUux)oGZc{pYF;eKuN})n}Lb8#f4-tcEwBI(6cTYv#fu4VT?J1?Y}Pe
zpPp1c9a#1O1yh1{zW}R<7f7K+D^13D**Pses{%!<_Z_(R6`guum=9g595J1LVhpgJ
zT!bqGc*)TtfMeacyi|YD3iuX46^k*#!gxmj2}?Z#$V#vCWUOB=6ydznQyYNa30K(Z
zZI3YTWPNQ*?3>Okk8bo5LjB_-Cq_iWGsd18@ch@cj9K4{-Cmzqg&*#P3uFToho{tT
zB-^<F6ga)WOu372)1?R&I6}X=rk)940(|1{Hhj-7y52Zf^m+yCi}WAdzX>7%;)f2;
z&Avf7Hk(6aTnC>H{B?i&H&y{Jc%420gOf*UCqMr2Rj=BEER$*_=O{_(3e6tggPq97
zt(b8Af3VsAbfy1uANVwoYo0$YX!Glu-yDKYIv4Ko9)C!=JS!);37AJW%)RbH)N3K=
zk=liwrf!2~ybvyDT_;2~ds{~&tiU)tduMyCe0U3Hj$MdDLmTn)Tj<HZA^e*r{?H_8
zM_mwO?qiyUYMM&*3sR}A0?OGVv5)(9YFf@?tGpKJ0vMv}K*fZbrO*5USbN1aw}okm
zAn@UIXvs0%-1#vf>I~RZFR)Z@GhA01(?OONVz>lIRTHvqC$qGtVJdyb9X4;dF&7vK
zKX7<)PR}wvR3it$yVVrr?TP_;GFg2K&!Hvgr&htot*1Fw(hAl|Aq3rXFM0g9og5#Q
z^HYlQX_|9v0Wje_DEe7mAux_H!qiB*nEt#HK9#XD8ZX?jTm_PYY;yXWS)URO%xRQX
zc+A_`BSl<uaG<V;-GYal|H+%~RLxAzzR;;1a8l0&tm`9E=qlMj`mH|#*%W}sn_sdW
z?ZZ^-77Wl_@4caG&wxnS30qt`*p4|P(+1~MH(>EtwMd4l2F6ivj|8YZxE~)vgh8No
z?wuYO7F6<em|tedSKayy*{_CEqWm8-Hmaeh7FlLCw@B%OF1f*RhF+(2(gJn7P+-~~
zU>gkL&%|j{l4m_2oYw-}jE|`mo7sM|hrxy2b@ZS+*jH6L#f?;7vvON%xYYDb^V@Z+
z&22Ci@LT!;n?VJulZ}#R+%)e!FdTCK2487R3!J+bPAek*yOH0|1<T`opO5qpR2p(J
z52Nxqx?jJOnfqL7)uFDL52KL^^pJ|774RdeV)^iI5$OqHqk?Qlt|WQDMdTt&L0hWq
zvIc@INcZw+o+5G1rNH24rHPC2jn2$%6kY{gRz$_A8Yz+wpi}HbxGo9vdQdaZK)+5+
zasZEUTm5DR`qBe)r`B#@G>4R`z5{9cS!$&Nd9foSz6SnyLdjabL@X2e3c4CH5gaJz
z*_)FV<zUkc)4KJDhbpDQ=VKVym%WFCgDe_GUAM3Cm3DOcF2hRa@WTMJXB#H*j3VaQ
ze#*z$MP55}CEZ8)(4vl;m$wjQG>;ysW6X;th`>AynNa#TvsjRGBAGIgAkd>MO;(_#
zQK?=A{B5t*Fbsw=cwdHdNmxC&zmbLzUC<{sVYiXY%7Liq*oM@Mj?)F*xAs9Ig_h}D
zP19onHAoMwOBOz5Sru9jNU*E5$(`le1V6S&2_qB(_gtF-y{=;b!o3M+d7Z%@<;U3G
z>6l(~qpD?n^qg4;+uBRBk_;=FtXKAY%PX-Ksz#T{T{_Ni!DRC%p+_Lza0GQWjmFy)
z2tOBqFLccv7Kr4Ld7yLhu8`E63Mf16TB1Uj*hAtZ53&}f*qzo39R^?fX(u(<MfB0d
z8uJQF;Jc9nUbJq*!m6e^U5{ojpqi$X*GS0U`E$oFZ$&g}ZMiVcPdCRthyoOkl_z>r
z2yxL#wu!i4?8}|?j<f&{0bdL5dzbb9T4w%<P0*kXovlL#t~L<a1e=q{(bdZLC5a(h
zV4O!KUI;sssLu?acc=s{{0Rn>@n*|8keqeqiS<2z8oLoGqPjD=3_3)$S=K8R2vU94
zZ-)0=Mp*ZV$zBfv3X}T6xXF~#Py}E!?e}K`(1{0k&=Tz)3}*6dAHRywo+|+=-#0*G
z55yh|coWJP&gsg_;HAV?W>(%W#Wu47IlxZ@UBF9itn4_cHtL<}ZO%~f_mbo+`AJVu
z3udL#%4$585a!x&+4u8wU!e`9mbuyoSj(pvC0VbL{}_F3818)@RVVjK@PfmB@``1l
zMX=dTmt=Q<EwHJCSyY+wI;+$bjovt#BvKDfLhmDQ&y+U6*t%hA{Nm`A;tUd0aS^|Z
zHbzw`B_(+Ni+!g6B~0SzBewe>A83ZK^W&~*7m76b9J(8EaHvcI(L|AnAVDI5fl7(Z
z#9XZw=pc9&<^FTD%PK;&39anrAL*hVOl^VD)4mzW8O^Rp#`e#S)+3(+asj}oNBDas
zl|aHOd61A5S0Mh*zP6>g%P&KIB|PQqia4q~x%{~oX*ENKvXz$C8(1j3c`I+2)od^I
z7$k9xt@3lI83<mrrcv@z8#s(ihW!fMz?;>lOq!&8bTSH63Fzz#)}h43vqzF?vMO8|
zYSg<PN(@lAXXy~%2Et6j=afU!+x9Zi4C;Zi=z0u{i6q`=#P;?}t&xS^U|k0Lkv=4d
zheob(An^BsOep%qa_Q!oUH*j{f@I=5I$gFw3SwSVN4)~7WrpiGgZ&atP-k3avVDZr
z>75s`c<^B#wK?9{^(oI{v)-t9a`gjjI5(CoJ*zzEP^fm|$TTB%xujY~yRH}Wxmrfe
zKLd{1>PppMzlARy`EJWfvVw&`MNS^_V7mH5WRiBhJuH}_cd=mMAT=Aa_=UxTLno+1
zt}+SN_kEI-2=*(912Lqiinhax?p8_u;Gm4qO=p)Gwww5+<(QlA*N#=z!p<rf@?P>@
zu%9R!fs}tP@5;-=t}(pr$w5cL);wZqG4=+6)~uWTGX`)~t7R0QVB2%5D<8Se^gc)p
z_M4Y*Uwn60@I}AzWIjc7#X10oB~iTvijl{Aua;yq6AKoe8VLsVuQ}Z$NDH?LxeW2F
z&9JB2j~*Aq(_$p<j1ql1C$E`>!Z}7w-6(~T)gL7eNZCayxvLQ-hTmPApg+&VxPPWY
zvVI|D{-w0RbEkK_`asaM>H0!^E6t?!q&Ls(uq(@D5cCYhZ8`0{S#G_-rq@;@g+XF?
zT?Qo!YYHot_%^l?_2*)<b7wYCbvvtf&uBX>Ks7DJBpWF5Dm<V2{6E!OBxKv=_odwo
z|2WmWq|e3BL<eotuTMIvhg~%~Ma}_YkpfUQ!R#iM!-vOS5|4LQ(^6vPU{>ZnpBCD$
zZ<GVZ02daO$4}tjn7>t{B~zPkO5_a0UQMO>!Rc%nD&uw4mE&93O|-Z2e0}2m05n6I
z`2t5E5uvmUou)eO$mybzfAc5>3H6ob=4zSJv&Dt|xYzGYKhkP)70WLM^pcU_&~?=^
z9Q%e7MlaYCHkZp--{pg!!Dw0Nn$G)E__x@Td2%$%ub(TRNLot*4xa^EZ-$KH%vXSX
zroPmh`9vQMFYlRza{#>N=R|j&rhKU(1i-%%O|wj_e>>eXfA3PLf($hVQp%OCHOf%d
zTSma$5Yj>^eqo6T!X72<LGf{3OPr27?Z0YZ5~_ZJ*(i5td}?1ud1_OgUhCv$6PQyr
z->ces5`;doahUlikI&JOCbeBkPQ$9aRjDpl|FL4YM#;mUwG;8-m-1Q6nErd6!S=0)
zab;CLx(EjTbA$pjV2-|=T7W7%Nb~NM1Ah=mz>4xfo0z#y_f#103hPDWXcL0l+f&uK
zh07AgmwBE-B2RpZ{L<%0aP@o1&?!Tf*PragPnD41*RdRh*u;J=tD(gR8@<$biBVih
zN%kI*YWrXwAq&9+bKe&P7KjPS;2c|_mb*~=eF4JVFHnENC!0eBx8~wRbi=}cGwt5o
zXQXQNO%_ZQcBQ5a_j9dSj*YmJpAs*iy4wa}jtC2V%c_)9y{MUwkX~cA*4YEK+X&{&
z9fb+^S8R=6qSRHNnYZL4zDt|SZHf(&stL)7{@f`x@{t?B*njCKRhid~DZcZPu18X>
z6Dm!fj<kf4M!i-}nOVCw7rn@2uQpB@Ndx8x9SzklZnhWwm9{=1YR-g+z`=RZOZbAs
zWN$dYCbNX$*v5D;1Vu1tsj8%2Q3&TfCAS;MVpa4}9#_^bIwaCH;tVqRj~8y#K@D<c
zdIlb>GXyNJTw*#$2|ssk!l^Z7(#pJLAkd}AzvN++Q}lpK;EPdjLtE%r{}vnk2^x7>
zLMzh4R3h<nA;c)7*eR2go_nJ^xKX6DP(h5V>e2P{9U$FE(kuYDkG}u(y?AdO$gn#g
zo;BtDph<KGOZn{z6jKGzTTC#e)OM+plgI=xn7|oGKm;&k`6<FrE3UxZq`bP3mlQ4!
zF^YRHkoh3=llqNbjM-$2s#Q~A^D2<%PGcS$$IA0gf3YPf4j%K42)~39c~rQPtd&Z|
zxz3k9t{N2)<f1)vttS=xc#x(Z+^KGAyhR*3)+vJo%vw(_eSKvGoZ);pYa$d!j`iM`
zkivvI=8KErB6<lydrT3+4-V-ZVQdo_bnZx_T`Zn12MIpGj=n2tzQr(Bea^}))BG@*
zTq+~X;X;m)$q_TJXYOf0kE+r*S5jTtKcj4<#=OFxw-SL>5E6gKx8XtV<YWN(Hruok
zUr=!{gIgJA%jg9^Z%a_!u(G@Xp9?>Wu0+m^>)G9V{sFtAF?9yjMx<V-sv%ONdnK^P
z01KmAmF9-m>s$U3%AjE3CscH8tj_4#=E7hXV`55aX?m5P5M4NAHxdf!lN(2@6p2~!
zLMMiR%3xxzL^F-BX@A(MV>a5KYsbEap8c{<BRT~kRHpV9rd(><FT9SfHM*VleTk1(
z--$T#u`g`t2n7k;)E_0v@9cv%*UERU_<ThdoaH;AZ@&rK9sh}8a21{b8v8Y-xY~)R
z%l8@=l_%9>+dv{vT>x>n%^0UZ-jH-@3boCSYCrV=$|IghinP%Cn2T;DDhu!z_rsvh
zl7T}(pJ9Qw2uhLBg;uP7B+zZft}Sh_e~jyAF=W8Wb4=)2GLx@^yGwvA73a&|L`4G6
zt!Tw~3n<K8vj=#m>-GXga&21pCXD$iJsASj*OCy8yblR~&eVk}#={#dcydLk2Y7gm
z#z$TMW!><%coECIC+Q<IZ`a3vbnA$}gPcG1WLb6MU2YvxYTc_$+x|av#YS06;`&^b
z&RUMJk<tyEUZbK8z~hj}-TuNH{b@!h>IzMv015&&MPb^S`Qk0en;3S$nh%vU0IOue
z6l8{vM97YOb&%D@h7_AZnfv4boGU8MP2Wah>usZ9Om1*TFBx2XhrE-(hdg`-Aqc5Z
ztEX;Z83vU05QHx+w#+@AomO3w2Fa$@s474G!0Gi*;F48gXQDS?4-9H447uqjo1R9Y
z+50t5uxPKA$$x^CZK;;I&-EriV<YKlu;QBTGL9f1xR|WLM!k!!U)rjt5~$i?CkUqn
zZ6fXKg73XUYjS0OI28MXU+DJFyOIRLGwYG}h1D`Y>h+GV^X2_?3dz^UJ=%~XFbL*i
zVE=N+LPsNkWx=x~IwG$a)<dd-DQ>Y*w@YCftn&#mTR7RzoP}*pldN4OR;^Q;9TQWp
zuH+qay>@BwAPF16?aJao9X9GGGN68vOQK+=;+bQ7oZjqo+3w4SXLA%E1MA#=HLibO
zFT|*jr>Q7DS^Bk!?edD)_d#5YJm*~La_c0BY-)P7z1RtNE|863dp~kYy2O)_IQ>9i
z2Qo!443Ajpsi8+@bG(dKQs=mV5ZUJHFE;J3S~Cku6)7P-jd9ZrIp(Z3DZ?5IP488V
zo@Hp2Qun9C`T!>IFE5$F7Wb6QD}9#czd1}~^fl&qGRk%f|4y6g9=AoVAaKl{Dh@Lu
zBY)SbrQi*T`=)?%JjRCPm8|VcPm%}1@CTDH?bAB2ZxBImRM0%kJAHeW7-ynC{V3?^
z(Qnwivb2>N`joo_-Z!-@F%{R%52LQo>SCb^0el5z^se6uCT6qn5b6EtA&SA|_Ln7u
zES)P*w0^4iz8x-u5Lw0ujb1h`A9HU<qK|GLX+p=E0njvqgW8YI0Es{u0LBvp-^;Ij
z73~<;^VVWBJBUHO+LOzRodeL<1S^^xm+*lZtU4datxBLn_zk-C6X)XNhjT?fU7CX$
zf%e35(Ms^AqR792B_F+2(O6ipV(bwPN2h;r2S(e5q$wC}#ZDUr`>3!eurkADQH~bg
z1h&S-=`U;cyL}ffpkq|EY7<WdcEUh(5~|1m^pnRg0ELkPJ@7+e$4MG1C<lW{aq(Oo
z<Rr9*dO%jjMwsOjMz|m3fzSM<7ggTLlVI@Wh!5^)-|#x<tJui<_{@hv<-}x0a8bUm
zMC33h6AB1PH2e(UB2#&ks~U|ANdZ7+S4s$Z%?oHI5hxl?kgKKW@YVK{?&Es8m(<)*
zEJ_l`g>b*4z<L`0G_kQKQ9lqHF<=Vt7PT+0(tyRuoPMe4F9nB1=`0NJy*{;5s;j=r
z4E)}GV7VQl>CxdQq0PX5fr*wIL*4OUN|~*+6c`@iyV9mA6E7U-df4kldJRx0stIBD
z)o(6q^~RR5hAjZCy`;}uNw#JqFbTDH0GZ`Q&v_3O7;KJ&Y$oxj-K1V1R#f^i+F$#;
zar<qM=%sY~*G{+(6Y7yNu#-IDO}sw?TEM9M)ZRymd!uOuK)iy0B18~LUC7139z1S%
z3ZqP^xRRv2WG}P)JX^Y@jXjycOwmfKFGXY_PB?H)v#)dMwn6+%`-DYl#=#Z$qAnR7
z+z#MiG#f>rzLdt0JB^Dt%{7<aSfj%?SoH3EjZWNrxvlyzyV%ynvf{UdzMqJ}OY-rS
znoa7H|7LzXBo}GFg<%L^H}x1F7H|iLyM_k}<Kpy7Eur<27Km21o5}r~5#n=yHrcP(
z61WvcrG7W+WehG7G+MF6!)aE9;9t5VCe>N|{bOeNwmArx-bC4bIyyc~wRB%g@C@Nu
z2dJBFVJe;@ay$X-_j%iIF6F%<Q%9Ii^AQ-jv7Ei$Z{67IuxeMAc7?AyW{!)WN8qOh
z8NZ9vols-eSDq^X&QSzKb^U|q+EojNH|AA^ZYMP@Dd*QeU{5fv!Z7h=gAy&0qP;%x
zaHFuVeiD-i;~mTwolg}Hw8n?8F7e0<QV<Tb?V(}_DYPl6NYp6+GNaE6%uJf=OYS-B
zh*g`%a?4!(ZlP6h`Ht}O`sRDgIt!{u_LnZz>>;TzD;aN-j|}R*%cNC*nVdSV7s46B
z@-eG*RtA{kNv_RJSo49j8Jw{G(pNDn@brn*OU8V`b!6QB8q2lB@_fC$dz`f?ZHpPq
zdTaxkPV-Vf^euietvoXXSrhyOOgn&E(o4-TW|yf2d<zaUU+Yo@hsb)#gz>gsUkg*j
zAE+y=H`Gt2lAqYLI&YdZ=Q8*`yWjpp6~~G)p&TNMBmF#b4=e!47w-2_0(W|F{+;z3
zjpx(LBCtH|aHG+>PMKHoEZ_(=nB1qf2JM3-J8^@bn~5*itSte*tuiBxOW96(U6AY{
z(G?TKVundE>n`grJI(m|$<8|8Br>cC_cLvHkd?4_>IE@SxE5*TmG)3y^25kYD+mtz
zWQA`Ct~~D@lzbbTxt!eh(+Yv{4hXEr+m<6JtQ#bW!wfa0U`Dmsk(fm2ES1@cp1SV;
z!^snD$3k2U`Cr`_A5C%jLS&4fXuYWWgi)wfr23U>!{L8Cpg0gk@#B2i@Sg`*M3fjz
z{gWi$U*4J^Ug-ghvq44+wZrkD1;#?&0ua6a3hMaSA@`b4SQ?a)r@BrN&hjL%Gyo%O
z2GGPy_$OGt1e!O>S=ZiQz+ZvgvBp?!<)#-c`O#DqDW_l@PYu*bW|%8?D6KitXel<Z
zj+d8Oa^>D1!7rn@QSt0olMRX9wy0MeyW-!41i2ik@U+M%x<iRg#^UL$=2Sa;V**xR
z1T^Zsze?VNu3c2#ow|Q@oicD-)!4W4>=Po39F80IJ60j`m@+_rn-kC{pJ#*5$McFX
zjE>g(HkHW*;JseLpH0<8pA`8%^q|ttp)jB?l+;U@8UGC3xES0h=Ox1IdV!2`Eb>@w
zyVFn`gP&o71lBG69U*8Nnxq14m-Y)IENR~NQB0@mUgM(OrZ`kLnJ?aXf+#Cln>h~S
z_pKCRI3ttgCtb7SU)iHC-;wW1J%Bw~Yb)P*n;C!EsgzC!*tHB@GO-S+=1~|`+-*i*
zwPPxVa<~U!0Ki`MA?u){wGHTdR?%elUERxkCJ}*88IwNjBp*)+YKhyHGr4|l>H+Od
zxpNql8KsDM)N03tXE&@%Q>ir#Vsgwm9C4<&{?oNzH)&T@wRE<kXEhhTv;{^geq`z;
zHfEl#gF;!7=d(-!6^SFB3D@2f@vT<i?+eghz^ND!F{&_44Ltlyx-+yx_DuZUj_&5p
z15W<E17=sx1W>9E#2L!P1`j^l<lu>8iSuR^&cGOKg4t?<VfEP}8>c4PS=EP%URfZI
z&!}xsXCh~HVnQvjXC*Fmc`}^JK79_gpMxyHp8Xm>=%Iaa{3o=W4hRbc2q$D|m6SJ%
zZ+>NTe3H`mKJnfDSaI@0m%0rf&`c`RWS52ArcojWgytl1e~3+;T74m<KIR`3ztQfg
zgO(17XG%?>*8T*3G7Fduk&fiv$5Yq*CG-~_+8BM@*-l)`C-gq3eT`Eqw~I=_X*wNP
z$gbs(!9G!8R`-p#IU!so=a0X+;=wS+KC#g(6q1vqcx{iWdY4_6qPTc|P=7(vhKB3N
zG=%lFO<LFoX+Q6p>iS|4+gy~)#i?rFQk`H|&0pkM642|CbG%OW&VtIfjpj{W&cwN(
zEhoLxo1eamI5cr%5+#Dnbs54nfB}h{t<J=+8Xf<qnPYP|LM&Ki82#~!4q@wc3&i%m
z#AKID2=d_3F|hxTl%SyyD$|BF-!Q^kplkcg5~HT_Uz(ATOt1K?^Ze6p@T=86iWt9q
z3RP_R`3@0*Vye`n2!QGgtp8pqdcE852p3&}hew-Dnm;PyNBJ{TR2$5%x)1^qk4dt$
zX=)Rm#Smw{r2S8;>SvKA<_f)K-IDPi^F{_au{MmkTR$?vg(9djFCMtGsEIc4Z9K+r
zt`@4AF3|l*bLl-|X+U*cnD2>CqCEeJ=!Eiiz92AX$wNS$FZLP^Nzz-Z;3qKydu~3A
z6nZsu%;Me7)<HvULVwd%{H5$YzzKnySlC>^r#^9?WVa=dq%sN}h9CTW9segV%KS7;
zTkj}X;(z@GANR7e+=C9h+h^}#-dG`gejcdCK&oa9qf5@&lGMwwh);+wHM2D|TX(zF
z#w@!sqL4A3QA$68J+HUyQnw9thY>aSou;~_S0$?^_Cku68RWvdC!~CF2A^mG50%q6
zRrG%bwZHwBnhy|K9w+TlApM1+w33b2haZ-%s~d3t7uzJm8lB1CI*ho?3`<Al)e@8K
zrrq88rAfjTuhT6CiN*msV&wSlJAcss^Y-l`4P}Lk!8_%Z`#%TL|AZ#|`o5TxP=%k$
z&OZE4YyD4So1dTM|K9Dt*606S$^Yv1LRpnE#ou{sZS=qzKT8~a^-t`ozj=~YPBtgV
zAA1X(&h%*86hCEmzJ(@2_?ddm!U+CT|8)!gv^)IiRWDguE&G~NRzNLqvUgiyAj2-V
z$jpG|DdV}ft<wMIH<6HWsv80_lLy`eT{T}jENJGTb54FP@xT5xdNJT>>}3BKK26&5
zL;u?!kI07o<<8A@^zr}Ag@{=|&XgZc*Y^JD^8CMk#s9t8pRdsWUB<sGpZ{AR|GzE}
ze(ntbS&Rm;O9A<2ZJ%@st@YXl77HTzEyD86J8FS<6TZfD_J7<1)5iZYLUTw+dYGfi
zOsh=witHZ__jU<jVQ^<f^d%r10BG_)0Z4p&v&nPYm|G_=3+!`yVL);IGB7BtVB1Rz
zJNY=y_6hLi+E0V-Y85OjoYEjR5wRQ|pV>i#(l9z9XnO)lp;zSwcjzB|uh%m9o&$`g
z1l+2&;CxDIhI!&Ed)W=<ii1CE+Z6N>u&vc|Y>{#Xj$R+Y^QvP7CE9DGD|*cG!}|4v
zv8gh>#l62TgVatn4~f5nCI=y0&A@nq%Ox6m#Bfl;I+G9jYkmef7h+oU@q8{5#DN=a
zR*7yKA?8augmHhw@QTT8GPa&|XSJ}r2S{__u2Gwa)n^u1rCv<aza1=9vCdsQ88_^d
zox_#aEa42kRWQnG)HeWcb7xR<`OX`Fp28BwzIljO5F);S8<>p*Gy(qa42Amo<LC#{
zhY;bbM3&<FL*zRi--juF61gyye$<7?7~I0M=oOGQ1&lxK2da@b;#-B3#%WeAKY`Sz
z+_qhM_j-V(`l2)^LKE>)PS2_3G)e{Nxam{i@neCpn_bZVr}jh^PTw&6@bF;g+j}<C
z-Py)C+)2WNU8Gbed0gfV5&3koV&^^6FkjlFxs7W}`XXknujeO{;H?w&4@OQA#@^6(
zSzP?`UsUXWSC+Ljrt>UA@BwUO80~xj5(KCgVstOy*ATUx{#jcAsJuD-U8a%~dtp*e
zTY!jMtyaJFbjII!F|Q;3)3ObHH8oYr>k&eKLz}nfz#q{E9fHpM2=MFf>H_LqsrHt^
z$OZtTEVz6^hd*5llcIQ91;TL~D9GO;6UEgR<v-$w?AmUw1OAF<89`DPAMBw*5qFs{
ziw4Xl29wYZy#+KtvXuR{wq*<ZeS{H+WFnxs+Vd(<6y~+Lh=;;B=w?+P@(K;igC6=8
zjHs;Gr&;|NZy}!3U_(W}N!j_6$n-`GX4x~~vsB-|c`F1KnMl7Xc+b;k=D^Q1lzsoQ
z58&lKiJi1dh^fCVbKvH<U{+Mmqcc<|V%%Eq?ZE~000w((l<_-M97&=i0gux~RVT}V
zo>#Vwd8x?Bt~9t3JYK3bGXWPO2T83r?3pq*QT>t2xb~&koSTvU&zsDD%AJ^VWDFvd
zNy_7B5!Grre{Va8;-wxIE42m@S3Yq_j(`~$Z;VQ9n0b6QgOaU>!|=%YI<8!~fQ0OT
z!pxZwh(P%Ujm};+dQRJ<A-so+wpo1W;(>VOWr!(`;i(;fC*-*F9=T|P94;ifWhSTY
zoYC#4{^!fOL4|Dk7PuWtf#Itt{8iA3yTSmV!@HE!y~YHBOA~?_E(0?aa8#GY^t1_v
znK{Uf%32ty(gnfz#W{|4nBQIM4VGa#nm~mCG4I8>kBUHX4CWy{B>qmz2eC_vdGik1
zN2}twEAYDZF)MJW9{!A|j!il(P(z@btmPA)0is8kcIFvivEnFEhsf%0nPt(V8ZEAV
zhC9*Z40HcIm#HyFNXVvS#Mtgys9E9*LHD(=v95%PPZE>tkYQN6MZP%*JD)oY+5E#Y
z&kdT&dWMbY-+aN1H~uY+?KL;D?evp2Tx=3SY!)v=sI0?aO3O4tRQtwEvD+d~6?Kdw
zD}iPRLf{T1%&VI!Ha+*R2m~whY>sP*Py4}>>!@`34eRGv4<N_(mT#|Q^2}sQSGDB@
zMTBX#k#(&sLh@}FotB|j95azQ34MAUCLa5&iAy()1NbE-7l6Db&|zV}4oR<)&9;CJ
zP`6Q?8EjihSRPTf?r}k6B(J$+l6Np$%c|g*)s=IgwBNoCzpDpE8qwOB+uGtz0W5ET
z5`3p~`Z~k#gS6d?aa<%76ebJV7vb0d!CC7YdKh<SP8HE7-mMA#pN9^L6d@N^zB?V!
z@Z^7Z5PHc18~bGvEq)Lk*F5RLa7Darmzf(|KzE{<J@90dbfFSI?EYi04TfI_!=wWk
zXr<fR9(vs1%D#GsEg&b6*_ls%7C{Yw^L&<ptUbvUh-UUL56y#51S(d1D+b2XYkc36
z#3Gp9K(I78e=EKPuf2<7H!!Y<OQ+v)XPzt8u=Cdx@-{(gqM<wR%jMdHCeGEs!2A&w
zvGX1<<uvLs&8t9?HxRNh&qv$Y?VqCOi0AKsFmiPEY((IdBJkNUf9xhd?eXd8U7wcm
zXMJZXDW(WcgRMy!2r2PMjRsq9%3Qj2OVXQSdWC3_9=?rq>4H|AB(eAF`5Gh$dw{jp
z!(eR=!4pHo*;Z2e8ETxtOY;(bn!h%BHWkhw?hWnu5J{t*U^bavge{M;L8d~(&H*_|
zb95UrBn)GUgi;{)(f{}c_7IgYkYq@(WD<5v`g){zZtUC*?tC>th1u%TqCS2#A^Rpt
zqDI+|#X2LtQSc4diWIP_xYo>I;u3Uam6Awi;aGQTFkhBKW>zR<QOEQBW#@0*^%)H5
z@`Fatb_^wU1SNLfu`vzn?3hXp9s?J{C>R(a(c->eqt%&Uzp~QQHlghUdofG61zavT
z)2n9r#@q}ht}x{Gr>1A>pqO-7h?_Irs5;}6KY2MPfe{DS$qF$=+)S(Oh1?F+<L2!x
z{hnlp2>#pIjgX6Fyzx0*YGw`WpY{j2DGs`KtWB;%w)=7E59b7g^(FHkz2S0Gqts>7
zvsmU281DjK^!W*2#LBKcG-LQpTErLk^!!fQS*F@l<CFAyhk=4;X0QI)0e7kM?&%kZ
z!>4iwe_S!$*80`hxlvxAhx&aTeXjax8iy;_JcgumNa(f+O*7>HDa&WIGuyao)QGof
zB}%U=mu^M9_#84}x1D>*Ya;vIA!PO#a^NAAK=@Xx+p{0ots<7$wT^8^#mx!P9W$_?
zF976=>&FNYc;`Gf?*a#y+a*1DD^!KKW-+Aiv}`GTYEbi*<knM0$DJ!+iTOM+%ryg3
z2v<71%S`n8Qj<NEQ&b}E{JHo!#ZLbzbsVv*JSpJk2^-Y1CAq_Lb?f{eehBd>;QuhM
z!r|Ko8?h-;D}KF9EE~KBpx1KHo0W->TZ3~qw&YeXg5=ilyWeU~^RC@cb)jHf)=WvZ
z>>j!yzokcWV~cWf+Yu=7cQM~8_MPY3DrHCB01TV?MiTaPv_3hlv-~#FBljoC4zU!@
z_7R<;9^7D66vck^JGk{dK2VL765oNor9`9XN%HJ14A!?Mv%7Ek<C)hG)->Xd$QwCz
zUVhm)r=TOq8`wVt*YWWAkq($sFeg-ZP`WX^M{;>xMJG)#F-*~>SIGjLlbAhMD!+h*
ztM0t&%I-|>(<okc7iH7nCcp?Qf~t6LoLdNzfmECe0k`9A6Fly#MNP`k;BTM{nuA_5
zxMnLwoeZjSKn@R%eeBB#<@a@fHJkfb8@NjtzGycUprgjkdj-bR1Z~)D#g|Kb1*-eI
z_;-!Kmp>-$E<Z~z+LaBeie;Ls%@|Z>zRe(JxeHY_y<H5>hu{!gIdzie5W#Pj8s)=&
zlIYNCkTfl_Tv;@G14U#%66aprc;+5QA|o?yIhf;Va!klK#5!aswz&D;zK-X!W^UgG
z%RS#Dj(d0jv%V5Hi})K3+@satLRTmYnlDebNL4<e(hIy@v>YWq_^AEdGJOI_D;z2B
zrs!eebQ-=KSjPO-{Uxbd4U9ET2B}Ut<ZA|Kq^qXyy!@9JfMJA$!Un|&&L%jiBZWp&
zGs_sf6*l~sHfEDNDdTv)!Oj}lBdrKVGDS4?;TW9T_t`k(NoGOQ*9!U9$-hH5=Bu^R
zal|*EE2cm>iStsw6FBUKSKQZe#DRSa#I#jR=j|ji?{SLWuukDglwdDTaKt5JXeCr9
zZy1e8_;1JEj5WkN(B1~kN+6gebAL;<knmE*o6sKVT!L)qGa;9brdUo29DH$O1&(_e
z-*VfQl?;rk=X#t6j2+>|?;8y5aBinawow{5iHzwuecZn|nM<NFzc!sx=brNK_>@;F
z;)p$(toX1p&-`%lZ2M`asQ-_#uMCT--TRfH1q1}?7>1T^grSF&MnI%P0RfRxhAsu9
zhEC~_4rxgxhVB|#8bxF%>9ctD^Xzx;59dAS(}KA!Vy*SR@B1IW;B+r(ZU!<wrvaXZ
zK{~VJ!gdbbn@`!rAjRE%eZIS1CEU$8u9X_}vd3k=GtNz~UKek@T)s=P``JD67W{<8
za!g$4o@viZBWH@C<l8oE`g}Ai)uy7p1e{SZ;<DCrTO8QmW#5%bR9-z<D40ACi-}iC
z=QuVid;O_`Y0;cq(BRQ?9XlObJ)<8lRo8&!?EA6EgD=UY*w^1gruqHU%=SLs(W1sx
z)VX<`%T?1;|D@@<iU2OAsHahX<vKfew<TTCj7$%>irX%BMzFlfUXEoc1DacuZOXeB
z!PhN$`|;;yrQk2ughdv9sO7a1iK0#Ptui9z^;qf)jPbYnzR9+7cYu29XzSTk)03M3
zwkXA$4t4)ux;K5gGNAL&PqtaUZ~jQ~U0K7XYyqNG{W)ji9N?Srb-(@H^-@FIaNvll
z7I?3XtyO4EMMFonoqlIO=;&1b+*|9a)y?%qmGFT2yry2fHIsG>=$i89gF4SK<{Qhx
zbvHC$aO}+;qw(t{9i>~l+jaAY%r~)uXh!W^AhQ-kYYL;=Ey6E*%eH|WhR@US27_n3
z=^}<9Hk1Q$6DW*kY-YX*6mr~`c@=tyW_-zfLy&;<=Kz3K(LufA{eLMkwtE6UKQ<_O
zZLP7+6oXXuMwXSvuwOG<ujeVjq-)IAEs;09OQ`{KdQWWICt=}fBOei;WQ8B{f<Jt~
zK`3g6goWvd#?pg0cpL8og;`)>!K}4qDU4u5NQi$uSQ793op3DeJUgp}*BX&JWaax;
zAD@bOT)a7LZTRwHb9k_LD|M%>^@(AtM+wkMY|vDs_RGbC=#Vu=W%-|eS*8~Of7lJ9
z$r^*@SmKCIy#^2n^HQ;5KfRG5ND`f{yBB%t9SFfW0Cz%k_>%5=p7b>qUEoG^fQpg}
zaOP)G;h^B>+U1`jAu>i^9-AYxpqok>P~&A5K7~4$4;?mpg1#xx#%KNos=FKzm4sNf
zm+y4Rby#0M{?1`-<V45H8+VViJ{ONPR$8fxWFR<9Es2=4xGaX$KKI1U7ae#{)}OVq
zQD><s7f)CxGk|#+N31HL`e8rU)5&!;atucnqHQqbzM{q-V+ondRpNQ)iVc6CBBI`R
z!^-`$K%xIY@iO9;UXWlb#YJ4DB9VmiiA`)<XWMvcl$~*^vH6lB@L?S8cqoHPYWSoC
zc_Q23&+}V2tKw)RB<G%c$5A##5{z^qH|!B9JznY^On2IT{g8(fL>ruf_stnRqixqY
zc|UX{F4wrNKkR$20y#i3$I!tGC$ZfnZSI>A`9>b;jTGCiU+7nQ>63GI@0u-&DcT8?
zsqlG8Hf93S<%QNSEtL~azj4$g?R>9Q*q(&Q-czg_6_m63)SFP$5~+lW6Ml0CmnS9s
zS((+bNl4G9ZUF;n2$E<cm=Qh~h(9}K5KRE_d6rQ-Un^2OR*T>Ci-*tOKnZe9Jxs?b
zR8b_`JOexij{BzMq%-Sr!Uw%lv8UXOla{0e6Ar=5aL7#pvo$R|xAu~~$zs73IG&Z^
ziK!(rG(HxC_{;5pdf?9s2;VdBh|7R?$gub9YuIq)XX%|=ppsyfv1v&}vjd#OkTZ1~
zMH!Ldun|EwHu6%}W|TMogo$3H*gW~lYDO(`gQ7Vzc{k#s_KG}=jVdoJU8k$P5m10g
zl~!z3^LVn9o(t_qr5KqJMkMAauj7e>q63)uzu?)8)$Tlc#!oiEXid^{up_3-<3&4`
z_rvua@AdCc8R!Q>sttOSyT`fQ5vMylyq<Af;K%ph4glMcJ<r^BP)}_52SsW72Jq|j
zu~Oo{P3u0iU@RCq=J&_rrwsQ^l_t1QSj_l-Sx`8bj3lnNjO`eJZC7)Pg9_%;utyN=
z7KNk%4nd~mrn;IOxfF<DBQMkXEy8Cpv!p3{SiivndKj#)a>4@C)ddCKSGUrDk6JfH
zDyvodccuKcV6uJ=0}DtSa<X158r3itD~<VmQAjrQ*=+s;oEs>AC$2P_EybKcBGi+q
zGMMb9Z(?=y!b7I^V-9sp%ANHKbuYo3F$$oN1*cCsd$VcKSZSBM7lwDO?*L44BZo#S
zt83qV@2ps<6pN#kzn@p6EF}c&I|JBak~5L7V|-18ZY#pm1VGw;F>Y`|VNikr12yL-
zHEnB)pjg`P#Q6{K*xep%SN7qq=fBL33AIqBxg{R#?pd9IiH=f_wDj+j**5}7`CeQ7
zD4nW8^PSa5@meB_X5oYCC-9>oTsR`RmV+-|iK?_8n_7Olwjgw)P7RAYF(y3F!IQ*_
zAs>P>rO`}=MH@y#qFvg|vd#_16_#g_c=C)R&i~$3Of#<K2QOVfRxT<X{Qatl1r*2I
zPdM%%qCH}y9lVg+9!0?2v1()kGppQNq@!)atmf@?Gq902s~)l-IExT~F>pAq2P(>~
zYnlB(&zEl@LfM%>g(|#F>-Qa!nQTR|;pm7XT#p*Am<^>Bh2;@Sa6uz~eO~EjsxbVG
zhpLHN>xn(+J`h8X#3Sn13pATG!P|9vL{_%~O{m=D(7Z96sKMPuogf9F1D_R6O=n@6
zNKocX&CYzg7onHA9WetybdTJYVtjjDKkJTQRA`x-<nAeJS&(9%VB&QkUo(IU6}Xcr
zb5!W91^a@T|9VLs<4SjY3FFMJd#&!fXI!FupV8@vU6<q@qGsBrDXzvw0AcCO@)N14
z)2p}sSZvf=$2T8(SMa#avD9K>>>P}2wk)qnRjh{!WU+n<O=tM8Uq!@&wVTUR*VC^f
zneTvRgE!;wwrloWH}SYNu`=*xwO>(5|F#LlO%h&Zr7VpyyBx<NXTZR3`DT++k0lHN
z#j8*m#9<Uh;o&_3<znZzLHYZTEP_hc63m2TNYwi=w&K)Jnd`)rpfwO#jrv!?oW2jV
zZS6fY4zLztVeGH~RDab@y{XocBq}VR(zBZj=g*8{PY7FloTuQye}&}|+SOSH9!YWr
zuFC*SDbAit0%il>h9d4{VxmBMqFlC2`a}8w{8}<4d#dI8B`2vQDZkwjVN7rdwGB+r
z$9i=UkUE3E#MV5C)Njv8@^h+?7p#n^wn;SpqcWZqCK4^H^+&`7=-A%5V>$>o0lD<=
zK;lWIgZE-Bo$ibVtNNH{MJ;t3y|R>L(%+nSmjF7YWqx_df~$WVe(o6<6qiH}Hq+TR
z$_~sVNR500+IL%%hk)dt<o%4#TsT?cMq=$z*;uWA;T%d}>>6%j1LYRoQoyGx4DiEO
z!ZOSER$lgnB|N}lcA^%Oe=#@^0F9$;(Xeu_d=JBd9#85y^m0J`0Zo_r+)uk-SpAAz
zQtAg8KHRI^Trqbtj5{6538nQ@2i6WJ+2mn`uS<WAJni=Qy82A@0{7%$$n&hWM%a8z
zvOY8>`tUnmYO?g*kt+r|4CM~sC{AJo#qV-^Aj<&m<7R8?uT1lF#@o;t(o(zRitZSP
zS>*I8KPH=c*QZb8AVK=t^7m<FJ@bl_hNQIL+g~KFAu9`(KLjK}5!8XW2u}^fp!#T+
zFG9HktVI=NZ^h>ti}!NhVZ39BlC<(fnsVcz2vX9VM8|2Hnsg_qega>YEf_CAuJi|x
zdA<4uw8&{{$EEG!*zI!Nwx#@34hSLF*rXW2Y(l^$B1Dk(lmO@<xgbt+UIVooF16v$
zC7@qTC4GGmgG+t1_x3twOT?a1^U+2^b~(|ZvD|uuoCykKx1HHifCVQi(^wu2o+E~z
zR=L|IKhx^e=TZx<zgu5xFZqF#E?hR#F{tGjTSF5^LuIB1e;VjISdWU5F>l~U--3Hb
zVM-*ic_&xyI+<^STC$(E)p@u!|5X0hBa{L`LPBqw*>*ps!mDNhuYlmN-*WMNbt4h7
zlrWBJQA0C_(9E%V1v#=MIWxQ<w%pKP`u<_s&wZ;~e2x4~T9;27!c?bkUOpbk=7|fc
ze@BR_VYZte1G{D(4{Vb{0I+(=6KU9W!?kYyb*ytf!O;QG?s@7t^JOWKmx#XZt2mOt
z8KEr8)ti%U?Kh|<y*)4%`=|7q<a34m;%saxxuww1Os>S3qmR<c?lkVg@HAm5q9!#t
zSY!4(d`qCePK1^g)y2Jq0cK;fvYn!&MOqEKVKaGt;Xe0;!x+6@n9y&my10svTlO+S
z#&O`61nBpapL(~mC%v6GDQfBMN2h!1_F%_u>-xqB?DBkKbS5PgZY05HFc}IWrcHs`
zm3VS|0KGzDKj(-EkjwA9G+1k@e<$$B3My<*(mDB>{D24fZF$rIIZkbxK?+MEIKqb`
zC=bBI@p73zgDA*08kB~q*-43=D7^d_!qQJSKJwcU{JxxCF)_}c|8gWy0PMQ<vAVNB
zl8HMuF*8aLVMkX>|9eFOPJ&<ehECLTyMaAENm(_O%5Y@<{CmPR&=BXeD}CiqI$n1M
zJ)s{Ml9$r*Jl=`8sK0`&rv22-WV~F#1(;4Tv%Gf7#hA~YKD~%EcgOl`(L8O&Q!n|<
zf;9~2D`#V?pB&~fUVYj@dz={lY~Ma*#cU>Qep>T|>=a;yp|iVr6tW%xrq999#d{Tf
z!Je|ppKuCAE1EWGXvM6**e=wgUY5?y`U58?Ew$c38$}DQ`<5_IZukY0sS!afsF8n|
ziGO-5*H1Dl!oTVIEx<xI@>^G3>gic&2c`$OJmmIQ!fQJ~D|7g#;NVC8t!zb^q#IxY
zB#M0OzvuZ_@+4&`MY%#z>qU(JwgThJ<gtdjZ8ra|pI4ForEA7j@kKWsDWueS8PuV6
z0(Q4bcJGhSId}xu+aIv488)q}wD>8zTClx%L<KirA}Ae&w_l)MykY4sc<#N9Nzcnq
zF&?ciCux;3giV`Dr|}fd|K?U{!I4gR5;*`saT(U>YwfN$n8++HQb&Axc5z$_y6X&-
zfSe5Ag@~2jBs`X0?sXfK(RtNEK;i5)k^WMLZf2TEuU55q2z=ksWbjJht+@TlmHj-|
zccpurTp-u*Mdi$In0r`m9x;GsWqDLK{fLwPiTs8+B{!GO(e+$iG<sEdI^WdlQh~N7
zgX6@oM{AbNjt?kwFRGs_NQQr01H9^qm#<G6hBl)XsU(T(6D2O*gkNJeu`q0=)}<`=
z*ABA<^UyYBD)q2&X0()B`=z{zf<JYX9PvB({0$H}A*=l+i(sa77^$AqnOVPl{z<F5
zFvOG_`#kuT&~K+n`-RF^zC?16Yjk*YDS;4h4Dq-%7sJ(gRn>`O32a;q%d@<l89oT5
zbJ$OdDOvw_+_&~3)0@mila#wNfbWXWW3h7skpxGFurIK5BMs~1Wz${U<v@i1D616l
znbb-Erh0T!=upZVgqynzPg%oRI<LK>SCLJ+Q7%-iE70uSuMcz+xZ}yp$*vq0e{_=T
zs=AJ5*h~sr&7Ret`588)iSAvu04c>KA1ERk51CESoe{_r-#x{TPm1WoP$P%ck@ItB
z(+O(7c>N)+BZL&woW`D94##(b1ViVO_KJ9}kKiM0jMz~aR>F*wUVRG7gs43yJeWDY
z9KB@uF%Mc!SW!fluxOL7GJ}dHd|If_LQN>-`BepE9~WInlY+2W3B)Lr2>+Og?GIna
zL-BIkakaY7MXTn7%7*9U^2mOgG_Vl0j7INw-qwm3$}9e)DOwP4mjuo!p?nWE;o*@!
z;po<fj;Pf*>0>~KTVPEHM+WD2K55=JF7q=rq53CL5(Vh$QCNpjA1<aYVfD7GseG}I
z``MdCNr3&BNgcWVV3e?Jhg!u?ol3SP@bxny5*x#^r<UjI;x)8ABuFEDT{4$VGT4-J
z{<50}*5sQR4A+R3k-7C=y-aihn3(0oiriK~v|NW(tWqL>#&}9}I9}K*Nqb{{0>MDi
zc$aU<ZAYv^?+Y)V#H<9yx>SS*AhHC;A>Q0jHNl?m>^%??*RkZ%J$f7#wzU)!8~IN>
z7PZgSLv$LNEe65^Itk|sY^+SAdPx>x{aVkFHl5rg7+<0{UgTu5b6=2P0DIId$EO(t
zYi;*qHQhBA0RAU@br-W5_pj3ksIA|1)WT%XKLK6gKG~qi0I8<o!Vl8z00N%^n~Xm3
zz!g$g)`IRYjL#L?!nlya!t6PTbQ!hZD3I{)>R~yG-KLvu9Al~y`0N=GuIrTQNKgYV
zW2H~Ut5URRctICbo`uarLZpZ2bzpwMzuD>68N#)bQW|g_8XZcM(dV{_g(ML5?8l#%
zB8UKEO=F`Y7&=-o&O(tOYP>JdW{D7KbOe1HO<1x^C(sLZh7t8+2H~vDz&tI+2&~P1
z2Q&ups$C9h@W0UrbQUm^Ed*N&vdf~z>{Hw<7mz_k>oBHPL-Z^zfuqXcTMJ2$|K5?!
zI;o>WWNMYQ9LZ{!H3&%n=ir=&x0sH^y5WiCN1PL-VyDWt7tr;UzVW+zJ3&rl-?%3i
z+?1w!oEM_bU-6=>`_Fq9&S7zc-a4(30tGdhbttm$D@hjzdw)celnVPS-|pAOdR`u%
z!p!@2lwrhkh0^NxKgpP)xV{(u(iOG9;7oPve-OlT#!ip;SQ}wzp(Fff@?FK#>F%lH
zXIT>cViPS)_tc*L(fnw`knHurXY?dqYxAwqrvc9rURk<es=Lw9JX4P{J{uigG%UTu
z)I@2CZ_Ix*D^_zepg3k$<FGq|vvt<qF2XIyp?PKt*}#-AK0T)N`BU;W<^^fK3W@-Q
zi*h(8M%~-JW7BavH?nB|MtW~Vn{vAdes5g-X<>kA=1GY3dkI+?*({;HOSuo$am*4}
zqQ;=*T;CV>GLH)0xz_#=FUZjl-~E(F7z}+y5ryW)5=Hsgd=OS>R5T2*Q<MyDS<kUC
z*2^6loR~gtakyIXo_*({9{3Ap9A2SFtGHwrKUNLSvH}oq?~aigs;+j$qMu>#gs@=j
zIaK)=N*o18`L1X9A(pwPd-3({Ar`80N}}h;2m9W)4WWB@Dv#2Nv$<g&!%_Bext7Mz
zYHWQ|?O>{cjLAhB6YC$3sOfOpAL;bk7+c%R3zi&2zJ%pTyqzD9$s_;D@l81aEplXb
zp3}21&Z;PET2o#<OQG?>bg=SJc4E52QJ}}b8bGYUCt<Jm7>#@Sz@v>GYhVmFPY{FC
zb5_j2#pGO6%3aR*`l%Y6Y~~Rb%4OXarC<7dDsrElbu|CUm9L2dd5tj#!Y=$tp9~8J
zd_FP2%Ig0}HFvFrY4by2N16R>qQjddGmg?myV@^9+BDOTM?;X@sGWPwGh@J~N~9{X
zr+kjO;hjU)9^T_b<z=62%s#~A2Zi||-7~w39^q4)g&L=Fkz#?_6Qdr&30oWSXbr(A
zKQ;J3mFS7vvni+JNQoyz@s89YniVB3nuqD<%vQz;r@<U+76}52^;pv|xrCg_`U;lC
z?J$qy()03K;7`g=8Ar9-%CXsC)k6E@)vP6&a;C;kG6>_!M>fR1y!vtS>uvjd{n7){
z*Ch1T52-zv=i{fbEA$K;i=<W4hlJAjX~I9aP#@TVJl4&cv$19C>t3ImqEmV-AtUz`
zOW!?C;w9}2tq6YL3RO+1g{t*;HF_;gQx-u5hVMTTzREbYUjC~J!FmQ%A<pw!M%J{W
zSW0S|KXl^v4gk>;-ycGL6aU)gE$kuu#lvx~BY&RevtPuEC2_>m1TXJMD8u(uD#;}U
zHI*yL16e=Fl5P0*{`@sX9F}cy;!gS}mm(MDw-s1TY0ckr0!(`i$*cMM-NK`Egz@Re
zh&HA;Kah}Z;B)wkytI-S?`8KDn8|@7q*E>|z%k}-g}sHaoEw2Gre18Ft`8hmVH?2*
z6o0~6NEEBW^jk$+i3=m!M>}D^`O9GS;Sg2D*rM;HX2&HQx;6>8<wyN196@;-@xs|}
ze@nG#WpLWbA-V^kxLz{KZQqW`JA6KROovA%R&VL`<Y&qJ3eO<luz%QJtzffcQ6bOB
z4MW$o3@8*>%C*MXx(yJ_##6?Yb2CCz(<0Z;I9TW58KxuDSz#nJiarsLMHoCv=nK^`
zUeok2xlUJJ*djvkL27?LIicEAjIJP)paFRw#|K#}!AKK53M3W!Aue4KE-A(Pm9N)9
z8FA)3IMI-k*+=cA{%fe_5LK`Js^wv2OXf6|Bm2s*tR``^vXMa2b3eUkT^x^(^;!|4
zL%&$8Y^~#(t>&mKLuq|?IkFyGYmmy9`}Y2lOMmBhsGKTv`Y<>0*a_(_qAdf1)tocw
z7WCb}vMOPOkP<nXPYG()j27KwzXCsoR2!#``?1gU$E-hwUA;tm-%vCS@mpu6N*ll+
zpARgT$#anBP?SvUB$3v7M0GjdsluMv2@{;I4vN%?s0)XT5fu<$Zfl9nI|`*67iodM
zjR#8eS@zPksqB!Zi~2HJ+w^DnSm5Sq<ujJz#6p?hsAc9v+SwuqY}5#XGzyWnA_*@Y
z)ap{Fjx;X5XkD!sBMrxM5VR<s*xB5$1Ra+CFrGx|@XOr8Fn0&h4f~0?6QA&ZZC#j=
zFq~*$j8)gi<{D`dl!KS>;^LcTj3!SFoR(W}nTvJar?JvR<D;klZKgtqLDePnV4ECh
zX1FRl(uYD{PZp*}Kc@qds=$`J_m*tYx@Ib6v4gl(!MJ|jL1UIQ5O5`_Sd^=NQ!z6r
zcT%9r_IvQ~{es3{F@c|F0rw=298fTKzZw_#WqiPRnAN95*1A)j`_#2|5v`~T)tn+X
z2Lg^Hs%R>th(=pO>;CD#s*`@aL3YyLVa(C|!DLt8<Z)D1o#6w<O6Q9SZ0Gvt&k3ce
z6iaiSNT1)|R-p`jasJGS$V6jvcKLH0ailH!y(3Oxb$A6l`-$|+++RNUi?~wi^TN;Q
zP&CXhkINDb_42v#a=3d*d!Nz;{tC9lGAhSmvrThn&s0o?-AJI3%%znSXBu#;fSJ1}
zrB(uFM)*fA$O_N^d2hqEQycV1G?t1}#!1BL=^!pC$)a(oEpEiel$I`}5>3hbSr$TQ
z@J}nOa^sZvOxm~Z5Q|9G=T@Ag|KvVR`xBJOb{DDIw&NM+D2mz#Np&~q>$#JZWw*Zz
zvs^s&$5?}w0o<X|;>%1?rKnAeTB#h3?|DBAFbO`5`d3#gyeQrMyxl?aZvB8;KB+Q7
zfdN;;yz)NAq4Us2P|^|okqpIe<k~9oyFibp1-22UH(bx6rat;Nh}Ry21S_VnBbT9V
zOf)*4U4}V&il=8rH0{NYnP{~-YZiBEAQQ1Bhu?(jyK3k)H=FE7V4u+S)RL4w{hfrn
z$@qGu!`G&Y_b73%S-hwfCp8w%4PPs%i|Mm|HmQl5M2rT77t)+h1B+P>t}U}Wt^$`W
z4C*S`izzjty5w!uzH5cksPV9u6;&$y>-`;e;&f(F=1nFhkBKFbclV!)c1sN%OM3_p
z(OM<vX_QXujCU?DJBgie?Yb~jO<4>CCJ+h~NH`CCo+L(DQ=kNwO(@T&iN6DFz_P~^
zwKK1iFm>)hUZ)u=7P`oPgNp1t#)xAPycvt}1I|+B5h0lfw}9rhz)<kE94d4$y-kQ*
zb(N8%jwuMOd^YhWe}BA4{&!wvY2YLHAB>&3OKDg6tIpwRit}2L0@PsrQPda162JWY
zeC!ou0+cUZI&S6>3WQwrNGTF;X1J(NW$n*%rbFnB9gMZ7jFfy2S;B2f>V|B;E(i~J
zbmD9CrPgF)2ki!=w|2@M;Q?Q974NXV@6^(>Ac+tZ;vuc#Y?FYjh@kHv*Z0JO7|EjZ
z-#wspBTcEYs8iL;c3Yk2_{yQ-f${D`I3uATm1zk^9*ADgL-B<dFXs`7@ci%WK`>Qq
zqUnc;7+`Dwv`fXXHskS-FIjNtQFq^z&D=SU7R5Bpa<!V0>3LVez68}g=P*!Iz$tsL
zd)6U7I>QVW&&irB*DWjja_((_?hL`iF)}a@_SUmli7~W+;%x$p4K03=$<9pKqaRsG
zmGYZrF-eB7##5=wA|x=6(}GUvG;r`4C4$RoOhB-&`*YbF8w{NDDHBqR)Ua=e=>$W<
z7iH0kujbhhx=V3#pTgCXGY4WmL&%D<{F-;ZmT^~Ts#t_gQwY^tk#TPZ<An*}{fuP{
zOY`)#sL6M#Au&;<-G36*^k_4KOorj!kN$_$GfAJ7W%(bnlCg#bcy{MFm1u^=51--v
zs}<c@@>eT5bhfvVRx3#(aDU}Nh$@{y#%-2%+cJFWhsUi1v7j*&iN&qb5IPx6s%0k2
zO5k5tbkY-_z7Eh}{2ErjKu;QZ9~K>}B55Lf^UXyrJ{Z)JC{SU+o~H_0BXtKM`6Cxc
zS9|9uQc`*)S=Nb>BvW_7iJgT!tZBd7-n+9O?y+Cks6wz#vnkUPr8h3=8)I5!u`u*i
z$GxoiPH}8C#nRNSy!l*3gj<@!I<+$Oc8q$P7<T5jSMMC3^wzda6!|42;6rDvw?<EL
zl{iYGT~bP$0hX;X<?s;`|KzF|5Hv9F#Gb{XOs|(~;`@L)Uwu<ev#?t^R+5Ix+g52O
zdJSuhjAHGXG*G^<N-uX=Kjs^?-F?8RZ^3T<I~--bDO+7`_&oZ;cNce`>_s467=1fR
z6KEtG8)iDA_6z-3VzbgW?dmit=X$BOMwB}!zAoH}s7^;6xa`~^L&hK~<1%t$BIiwW
z9gOvAhUk(T>Msg*mCu!GF63<#mePOr?VQE+N;|y>M{|(kdW2qQv<PGWtOHc*gQiA|
zSV2iirEJ`n%|Nt9LaJ6^vI+PG>p57zLj~nR)BFr$)DdX)U!;g(s8nwk=LeH>)3-L-
z>E2lRdcB8tOYNVsCXzEsu(Ej`90lnF$vxzK8cv2Sg3VSb@DNckA;O9D#X)&sbO&Cj
z=L>qW?c)anz@xdgP6Hf|Qu6SaKp8S52YJg#{|y_n+~O&F)B9cwy!(w|?B(zP(_xDg
z)hhtEP%=#tpJ`iz&=;fauX9dAfGa(PyN;pr`eW3a7(dV0DuOx>S^41MaFbQzGc}Cg
zfAusDKn;acZ14rYU$+yf4qat%T~2M@<!bfIqK|KC-27(B-(ZrV41Z0S3B?Y=rw$4C
zc)9-Ar3-M6WdYVmIP{Hup&6}!)?7EBx?-0f6F?O`MimSy<=cX=1e5eQEaVv)b_u24
zkqf%s8FKbJRJnh?iMh}bgTZUR(jtLm)i6=6?3jODp|_5o3uQ7CY?AZO>izZ>ym%`a
zIBmGO?;_20W^w-xT+Z1d@QxRZVbsz9=otuN24Y;X5&Is-_Ufad)J-XL0WifyJA%mv
zDka;mwsU}n!;&^!SFAJqL7eC7;7LQur(=XnELnHtiMyFGGNCDV@}ukf^i5mh&4Lj@
znNbZDKDl?K2+?~=v`#h>4EEx{phIK$@J-Nu_65_;B`NWUyvgI4@#<%9^=NDr*^s#6
zLXW~-QdRj>zLdQ#<R0|^82^q!vy&Q0v5fURT4x%{yPd#bPSmv(#l{XrGapBw=WZ>0
zHFF%c&fTsZC5oJG<$0on-@h_7UH|sh-cTSwh@ph{yPN@lRTQ*bqzbd`V2v6p5sV!l
z-$;r1#*f|<FDUo={Up61OFgW><>~cM+s2q7o7zz~r%VF2=jTM~vwqC0TYFiD<2t8K
z{p;cUEmh(j$)xl|FU_`#qi$A|pAB6c^bH=et?6UHA3%TT<$T4*zPC2g$;YY#Ls|bE
z=a;uT4fs=$%Y$ZBk*X>j(=V@>9$X2<A9z@wzl27rGPzc1ScoKmU7xJ|M1Sd$M*uqV
zo$lCSrav>jfdqeEMC$)#RPi>-xGYh{E8(#SKXsItOApV~>!=J`*+~dvWCZG(1Ggt2
z)ZSsL>8@`zM|IWbQpe<ukn-eCD%XWZS^DpDK0aAXJ9;u|{a7dOZ#9vQEGY~j6Omd*
zh(4-}CSg=a{b*vw1o@UQtV4HCA5wQ@6krG%U{fz_S}KZW`{GRB^d$|)QFONW@R7=!
zQ;Ra_D@nWWL0wp}EyH7~S?96ij|gD0j(?^)ppUxdq5LQL<)4hpGLa*IY}Ia9)P=N@
zdH>TSxOdOlAt@MS#ixbUrrI7Z6>(V|B|x<z8zA%Z!X(e!$(1~zZLF@GbM8f9t^1>-
z62_hRa6IHmKdVz9=XJex*5CPTE-!&%&m4TkTwBIA5%1}<3RM5%{v%+q@q-97vASMs
z*s8JjGe$l5Xn$@hCI50IH3JXX<?0pvwkgYRhCVfNI6OcWJvZVv9ZA5!Rd&DgIalyq
z97+@GI>LTTHHeG6g$kJ|f;)w3?Mr^l1z}@pDbcMGker#09~C(M6ace~MpNxK^<0?P
zf>?>w9S=-@9ZE0XVyD03&^nGbZWvx)=WnTR=FH~1S_qOsweK>iTbTs0%)(4mmMuZ2
zkwU{&wQ-w(h0eml&G$F7HLHM<T`U7V9X!)k@6oMli|8k29u9L)a;t3IwzAz>0EzMa
zZQJ3;&I`Zqe{CwRN_PY$ob1kn?UamD{XVnaE1fIBz?2-R`ikd@wRIMd6)5p*eur$w
zo4;{FOTVG)dBBiJm_zOHQQlF*5z|q5*W;V+J2>bW6RdNhNMSy|g9sj?qjV}4rYP$K
zHe!n!KoDLeuKavG_&V0d3LNT)e6*ehl^tA>@RhzY2DX+FmwVONxztxu!R_>KW;@UA
zS_b6qhlt|@mhl2}8~zHp7P(iW%!mR2e!7GgNu`Ix^(W~qVPQc}u`_x!CNwIg&7KDd
zhNK9*k{uQsi+D;XH<z=OO`s!&czi^&i?iiqmf9KGM;G}FI&gP8go#YrF<haQk=@W#
zK=#x+f(iLD-x@%P3By^gCgHy<x3fRWn!|?2at;Fs-9EO%O!CbN9d>44w4^*pm94iw
z9MxbF*W*pVw8rdDb04ReB*kb7CCtApWVR+^gHG(b+mMbmc3@$$*r3EC7u7Y@@AFFw
zg$FH+)_=YdUSc&uD710R0~q_c7P`$I*LBjf$7^5|^U-kz)qrQXLsyguBCAAQJ+7zF
zmcOQOKYn<j)D8j~_Y=fGfcA7~LRYZ^v`e7pNLX=O%4LA0HSBd!h2VRu1=trw-#n=o
zk%p|chp)1<UAe`$j_IK1f9)%HBqTmN?7H;N<+Dz3&Wf}wzx&Z0G-%D8fB7r769;{c
z>ab!3jW6eAJ?X<d@{N2n8nYXQi5R0gjX8YzQH|(AmVPb=E0)<eZzQ&QkuBYEoC@0w
zv^}l?lRk2A2f4rLNf8%Wp<(zj8eAP-+nOa!sERkvLhufa@2z6B9fPRe?GNsj?;2?&
zq5r%`>`k&3T+frE=haO2*g_`=Z&ttl%X-uKy$E;C7*%&)yi}-h8n6u6o|@pm31h6C
zeb#6-hGT^&R;rCLLwY3U)MwpUxAIR*d+oyl3Q(oe`AN_#-|E4&UDu_CXf_D=q|#*-
z@G=uP>)DL`3d~zxxyCoR?3f)R7etPO-?dI@=iWSh2$QW+UAOHQBcAQ5){(=@s0?tr
zXR9!Qwae)RIwwD*t`*|$VUuwrj?%;UX*By3B|lSN=$<~8)Kl;MS55)}ZQ3ljJ6FW`
zq6Y78=s@oMOdYh0A{czkUo`XLF(YMf<Av$-bkD~FQ;oIfzCeojM+S?bqc|ZIn_%}&
z5DvSX(vRZt?7re`j8Cm@05TfaNkDw+LNFUw14Xnn79e5wntW0mm(wC}Ij!s-O$v@i
z?(6{o(!@u_FD{K?nSzPZ9En^VXkIn%6v~-oyf!DY9<^Cok#Kn_|1b<J-q;h)HWx(R
zOG+r*M3Zkqm|5`YR^cie|9hv1rp+H>nsr<aJ0jElbA|XYIm{q%Qk$<<Pcm3R7IEJu
zCCrmrxQbX18F>|g)0!m4xhc0k1g^0j4N(L;cg3zL*r1=1<_qI^m{2sZ>F$|iLW5dF
z-UC{!LUw|zcDa0Zxk5;O%QMDw693y-I$hxN0NFaC@)t(1cCfq&Lb*D}_(%>dF6JHu
zh&N*3&eZ0LBOm`-5%Y~^%LyhN3U<gCRT?nXK!n!FwioNi%i)}72Syy*L-29nIs}5#
z*rfXJmv|0>+fMqqDhVMo*wpx#5<$p^(=mM{7Rv=UB%I^J>1}f&jN8mULd1~Y*s;qv
z$+c=+NN<`5fuGMR^avTw_wdh2U2?hl?5zF-eT=-6v$@B`9J`Gpp>uo%3`e{|4D5~P
zvU|1Jwzbo2@Y3+`(o*o(6<9)FQv7^7<1?vbunqCRZ`CBA4MOUYb*G`3AdUz*2$*0}
z`89%FHlEBqgYDzJ)Y#J&@Q9B=`{*33>n<cARXW_{b<7VZR}md6T822W(d-A?av?Gk
zl+Q<nK5Ay(3vT)FEFJ8skI<R2k{x+)=0cKASj<KcOhCjwX?fIAh^R98RL@or;U4Dx
z+!}YcG|WG+p2BGQJF{dmhM?V#A4ebPJ8|h^16Aeg=<U6f31WhvGP6!S5J6p0DT&5t
zlkg5vstI5;3DYL2=Cjl7?oXBu6u?~zB=<WmHGc=tR^03enL+M|Wa0tI5qm*4q@;uF
zE8S21u;!%=m1RZ7B+$+SxoN*R;vP*b4Srk=@;P{PEV5J%_9%WvcVkSSP3@%H=E?h?
zVEK|h%yWzyRpY$y6ItH^Gu^@jqhjqTg>zt7Q}8!pZBl2Qe(}AFM0;`GEEf!2+pI?Y
z?@}rgkAM1sDp#0^h{&Bt{W1Jn6~EoLk6XabJ1>k8h8^G-If{Z7Rb05Ypt01ssYo`n
zMKZ-1d)QBdRw;CPwa|(0p0Kbxv9Zt@;9mP<?|O7-`Fd2o4P28DRlWNwxv{YOkfNPp
zYeq@gPo$F;hD8WrCzJ~Q6;sZzsA>F-$dM<qVeGKFdf;2t9P#cF;|Fya<PU%^NDuR&
zwt%4ZNt*aAvTQR9h_+{*T~fnR@KB&SuP-NJYg4K{59|2EOt{~5JD6WQ``rFy&*sc_
zET>Z~Lhf)tVawFUn<FS%rZ(#H93JAW)BYR{!A7?yPt*-M^kOom34w1z-J0RT;g3GC
zUsGQ{1c(nr$^idK<x1tu(&Xy`<7@&RqyY=wM>yy!aP3S0Y<vKdWPbP2n!)Du?$`W1
zeMasdZIU~6_dUqHODOK(@VpnBSD0^5T)F$q-BEPa#(BK~&+J97Wkn5-#eT<0sM8k|
z<#93NPWr5JYSA=Ek9MNq$Dv18MNB!{9Af<Kj@Gu&PN-hCOlKQk44my^$4f$2tnSm~
z&yYzW$9Jwu7WKG1c=gxer(4YYhH_0~qUtp3$c}TVbD47+n_#~LVe9t1A-OkIBi5LX
zcTMNhn(Tzq$7kQPKDkm^9*q{2i=e@Tl01@7N#5F+EgIt1VPO5<F&u4L_e{C;&UmO$
zCFJYNQAq0U9;E$CzbGRgkD;RAdaO8T02{+&g@C~#`18h)u>aXzMvt$0tpjiKQzF-m
z?^Zc|3~697V5qr&cZA)9y(&>w&3f3P_d=;TGrH`&gup~=B%k?U#uCw%-kTTr4SC@K
z?_1vSx>w;@*D;UF1rh&ln#*hIw0uWw$(ok=dxxHtncvx$vM$T?jt~v6XrIqc3@!+Y
zekGfjOMRKQx0S!e9J98Qf}U|&w1#ot)|IhWk^}sc_xt@R><Y3GKoRq*^CT%_;@WVn
z+`D~W<eU(6#k^c#PC3_8FGQhg`CZt(Px+&!?vRkL@z+%A3cbk<dZ}QaQQX|D(d2O%
zy-bbgh+d(`XW4x0R|(3yBU^%ukM6OKa1ys<t&&<kYYjtZ^H$xZCFC$1W}kBK%tC8P
zklOeC5Z#JL>N7(B;4q}dD(_P6^m#0sw8bt%CwFSj4tH9fIitnL;goZl`@`L_N9!1l
zzoCYvq#rX=$qD##oKjiV*WzA!^|TUi7DpG><Irb{mT*?=Qn4j>A4%CrCUW(bsON7r
z(;Z&ke`^+l_HKf@WRAzrEH`n)JbdLO7+b_uz^5hlXHrV<N-3K8V02SAv$(u-;<3om
z<wxnQeAYTSpD$fcqs7A`>Z3*DWrnvN7pZuHwR{@1%omFG*y$5!J#^zmFnt5ij`4vB
zelilbuvMV{<b*|0sw}Ja&IQX1qYk`$?sNVK9W1tx|9CQy;mf@hOUap=-AuC7ftEst
z&Pk*bPZz(l)=%zHau`eXX*#g%mwK63ZmFDKNbNeGGMibHHB~6Z1i=|bJ!b1vY{b71
zws|H@f352gz#mz`H_flXo_efisq>W?yyB6#*8*+M5UL<;5(6MAX{4sroGW{g^c)DH
z&1WS!T2~ybWcgV<is_r}qht0$zthSUd#AVw{Z0>Tgu_@QQ+t@fm5Fwh=bjA^qJk=x
zii#W<y#@ou9BT!Sg+h*Vr=I;%a+r#pV9d;EeFs&9^Qah;?bK+IUw&g!oU$xBp#@Y!
z#DTG2dqlMECjG6J0fV||g5tYI`EX(4cy#Du6B_Sn$&kAJo)@>i%=jr~sdC3(`3TfH
zFMd(1Cyz~n*Y~W<3yYh_P_^xH8%X%qNT2>L5A0<h%lf-}_3z6I+(71NaY+BXaYFL9
zIn{(Bn@y4qWE9i;y*Mdx5O_s!)%9-#;_byPu)~TC6r>CxM&^Coc}#-4j?wVoA)C~O
zsZ`s=;{Ge2=sw>)cZ+$#&s60!_p6o6vPv3$Hb6U+;f`7Mkd<51$cAa9OCU}Dd%&-B
zqK8$(m|!b4l>gfy0s^MIWu)jxp2Lu0POE_RW!>8zry2E2^pfZsIh*;0v|ou|bIC~&
z?r&oE3NJO>uF5_S*PM46jG-k&(IqYYitsDJCLnP>DQ9H03F0xDGa$b&r)S}k+|oiz
zUA049e;h*h<R!*XEvzGM@4@pK^4b0;W`i%ClfTx8qVf5XKIMhusn8zpX!$oE3^+2E
zf3$Oy+__{r$F^Y%5+c}43xd9UmX1SO$9q)H*wvRSwVT!QbIR9hxKi%V3(19-78fUg
zLcRC4pogu=GqN+IC9=FX4{9VD0JwqZiIxH*#~<rw5HQRR{eDmw8~!s2_3k-0X_eOP
zAlXsvCV%GfRLoIpw;C49)n8p;RMdF=WdYv^r69Y{quk|+LY|P=RRG+361z9<nz*$*
zLXV-GO^?A2s$W}s3w-@o-u3q-5Ci0CZROEgKU-}20O~oamBb37QWp5S@(4DcsOY4)
zwB2f91ga$z+7qlMBv}0X^R^_sU2=m=AayJ@zNVe9r6$0Vg~+s*)=#hF%UiG0M<Z~}
z1)F#coc0M1=<2Yc%6g!_ff1|pra+7UNPZ&3R*j4yxK7Gu@=c7A$+vjZyW8fTlHLaq
z=8l<9z(-Tvp~(U&N`S7w;ZP~n{^2jfHuQr-o32J;;$;$tj(6Uyn9J--w{%Rrh>SHk
zHsVMbRbsLhKq|7TKu8N0i#g1T+hCxzdG(3_Jqh%K$U75H>qO>ZXQLD)6zpcyOyMg=
z6@ARcYFs-ew>_LcE*%)SRGqAgYbtHv-;RKuw^HGsb#|H5WnR}MN`i3}YX0}$sEN!M
zTl!6%^9&<kGS5&evEuqs#rpa0=|>Ei^L0+uQ!%%a!SX$UcBOl6%rIpqCOg-+05UAq
zalIzKZLGmB0TB*6&xO_k8`KR{)-@y}yF4wg{_cn~(MJXCf=s&~CiA4+?W%OBw1r(Y
z`f9=v*97>rvTvTIT+&eG=XnLw?Rhg&J{U`zs#!-pji=IliLtb*-je2TM^5h2<&T?h
zvm`zkN~1kc_C?h9_~3ny2Swjb>L?S{X;~=H!pv`3^D&;?Sd=QUs8>?AI)*yNp2Ww*
zKjUPL2YA7r@=gyY+Kw?7kYpYQMk8JZUEWShOQQ^O=L>-K3IxtWKU*G9f6qB_8w(zQ
zhj+k{Ec9C8+RhNMRfs^2cy5Q3sosb%V(30EiSXPDMyB``S?d|;-t+WhSw^AZ7C#E0
zVBzm`B=^C4(h|4_B6U~3sI>b|FU-pxPZFPJg~0WU`>tMkFNy7)U9jHPOJid9&KXh6
zs8RqR+H;<fphaPjy_M5j`Q=x$lJ2_&g8KMh597b5t<NggKSoZm*dN*69=7CxxUW^|
z8~=v%{70C70%s>4$thV<jLgzghXoPJ$;CU#;e5&Niv?rqZSdSQnUKg*08FwkV9dzp
zMm<qrYXiiA_^my?92Yv+K!J}^9GWHH@%(G|f2^F%`xvC3f?8IHeB}vrf?Bq*;W%_A
z{wtV_R24vG3BEtOLc^X0CcLGY$~v>!84q=QhP2M@A8C;-tpYyN_eQ*z4#x`~NqvYU
z6%M2po}<3ilFi<u>-CFip>rD)kTZrpTN2S>Gyt5rz*btLEU!md{6;|^s{kPcqZSX+
zeoSq|%!`Xml{<0_xj`xka+ycB)hq+72M?W3xi0lEO#Q*q`E<jy;>1o`*kd_Z4EX;0
zD6tLS(f&<q#*370W$JJfTlg6uIL-ibtajPH&0>X_97m7?YO{dwbTE?OJw~HFXcnE*
z_a}QVJ{T!y2~hCsFVGC)qKqqrPJSAGerE&b?*S*K;hxqtLHu8USCP-3$s*|our9m=
zT=@OI!FBa@QZBM_F1|-eSuQS`ZtZ*&NDO=BHywy02M6uXxcTlafKxZ|?MNXTT7@45
zy$^d>b6IFx(#u<88kINT94i-$RK$SiMrBt_XogWy!JvYYI8I#v!TaTNmhTftTVVRv
zCMb7~Fx?HTr>r~q*{9{_>UoRWv`^X^{T<AB2@bQMkv_^HHo9W$3*qILF5=$k9$JrX
z=C!gQq3&N&ezUI6p8+@zTLSE0nh)|zH^8%0nDNZqALk9@hBhi2dja@S?AUCanQ_&b
zlMg2tgfJ)p1d7`qUN76F@>u8~zMV*Pwu66_F0y1#pOah9ubJ**QU|_G0C_-t&mxy+
zF*q#Uz;<VgL*}Ok65r|GRYpr})KOr5`kPyMgwOSiE!$$AS#NFtEeHb0S6otuui@;O
zcFz|QX4;N}J$EJ)U(}sEDhQTV>8c0Uq}HV#Db6OTVWZyN_QGE0w)%f1%d~(mI({Dd
zNrq0btEt~DDSc1C+t^Xn$@amA*&FpiGAxA#c0{Xs-}OTankbek7^z?|qr|}Nxq9oU
zjJo!=Xeel%cvHgcOSn_Cq<vEw=rc|+uH2uOJ$GQ&jxgdMrzl^&Z7$pn&+0CqhBGIC
zt*naPvgqfuO5XO1>6lw@3~jcM!umLGBrRG;BH^Wws$n>Kn-*6v@tlcNM?)-Uh}jIS
zCKF-?`CfC3UCW#?;g?A0`}|9#N#bBc?*<+>JHVZ2%ut&@-2LZ(`4=PU0f?}S8d?-y
z1H)svPuNBkztL$~B7U`65PU~TGb~rVh6F)OD2PrYt*t>t9-Vb{jQ2Hd<>O-{hqKxI
zHSGEOxJr<{gARu-%yj`iX6mB@w7LPK$+&o$fU;Lx>>~Qdl$(m6dYRCi6e9*2J3f_!
z0lfD6$^~PQ77C_Yc>RrtDK-e)&p$?N6s`COk(CCDpG-FMDvo^%4iE^utEy&v7Z+|M
z0Ga|cE9@!<xby<RIuCtT3Cyc%w+`f5*5^4*T)H3BkFs37aI6NTKkkucf<5IpS~k-h
zBD>i5LX{j78n|dn<B9-50XTo*M;1l^d+V)orgRT{5=Br91tzVHfL9Vu2wi`$Ye2rr
zHTdIZ{*=oBu%L)30Q~~vMu7GoD$VX3yQ;|NHMT;7MFwA?Cei%Fh+rr=cKjWr*^GR1
z?U&`YE9nu^{SvK7RjXeTyA2+fv$0ua-oxrDmVj6+p92F4@ONOlNnC9j@vWG0>BkKs
zUvoxEM7C~B4I;fW8nDAW5xxkAFoRkGl5DV~j^U?yWMK=zJ;iUQ(id|h_jX4{hx&<^
z76GN;uWDT7Pnr(f>w|cK1nSe?K-8UNNV#9=yKs^W#0`1LaSKb*4y|dl=?{+QY|d`M
zZNx!hSmn3PVWp5Z;Iz^hW@2O}LnxPGokCkD@xv0i49fU}I6RE!htkz$b90u+4AZ5G
zoUg!Av&44ZV5SKJ1g{M>M9>#<g%N>ZT1mX?6TdG<!7hu4FB+Aj4TzRY56F;<&RYb_
zqJkV%FNT7iE|n70a&}=9Gk!lX!NrS5^(>7~pYWjb7EQ~D%A}RA5Mn0GFXY)aPIn<*
zd5>jlHt<OKNf%J4>k20#``!5$28UC}Z?n#sVg^@%{fOqm9_6uXzoT6;#9|7hWiUFV
zxED3n!e3=L?;Bh91j@GG@7PKw?NO?rb1qk7iKAGj@p0t}wXN<oH>@NAkS3N1FYDgf
z--!9Yv+fic48jt^M(V~H?8sgZrKM)uE7Uj)Tt2SZ^vyF;oyFR|`B(2oaKJzpn$uyV
z7)Z?i3iv7#Jbw7Hu(axqWl-C&G4W~Y%6nM>UjAu&X1Pdj|FnyJfXjxtG!2fOg8#6O
zh)^{LgoI*!a$?qu5aH*l<HQWx&Lni>Uo3hbxvX`E$YwZ&RMrbmFAl6}o>D8w8mQ#_
zN^N!pW@`sf7PTJce9Ets7+Qv$>T$%cu_)yiPjyW^l*jMvM50wv;`pvrMp&NfQts!^
zG+W6%Q*~n~+2{nxp_Al|r5=wQ#qN<Shci1VJ@CeR;RKUyGtTupipAe$+>n)si!Ace
zG!6WdHj^q^_6fI3&PeY~OE07A964^|!7vMkPcfep26vcq+R5G2#}RtrUw%mzw-q?g
zOdZBOE!=m16z6|kXkW>J`&z}9(MkBZN+!6zz&nB2>4xQv#CLhNj%d%0#M9Vf)g|_Q
z)?HjQVN;^8_VAHZK4uep)|A&g_TjFieyW-$n$FXTpuC^@InIkCfnw+49s*MQp!;SL
zS^#S%T;*y?yzt62F{<#i^sHD~IaBY^>wDKo439J@b7calrOEtqL3!WAMA6XT+K$^&
zonHXy4kKmB)sM!lq9*gi{pOmi{2}#apvB?4mc6XD_W-R*tJSKg)MsAOkPqjr-s74#
z-FTSzjuA-QEhK<lQk>rGy7r&<{2%q!-_82}ypYS@;WHw^6<zjmW4@QbGXFp&Q+XB7
zUCCnlx%YjD#n#}(U5fp(f3yH1fFG^;Ng@s)lALI*Tj-LGvcLQcI3kq%SHdq2!Um{I
z78oQ_cmJz`_>V;E-<9HDsE3q$7*LGwB2hekb4fG7%vDcFQat9~Kvr~Pg-1e}tA68M
zZY#^wNc(&q5+%6J0Cb3(39=-$0{^+Usz0|>sqb7uTGju2^uOJ1)h7~*0deTf6~I%a
z%LlADuipg7RsqDh5O(M){%VHXFn)~p!Wz!$RTt?~LOgDYIlnd9Wh!9lN(3i3@wXqB
zJHIQEx!PIPiSc?OzE?Z(bt3bmw);*)(HPLDrl1N0JPEnzynj{)@1@N(9|262nmX27
zLd7>ATcW&G3&++h@L#&^Y2QAQr)_xRq_5ACTDKPd`RKnN&wqcHLWyM+S%0+;l)6HH
z7-sN@Jv%VJGBX^6y}tD&$Cdfh8GJjB<Ho_Fh;TIbn9{fG{_p_D-t6>~8!o-Ow&yxC
zV-wZNi7?O*=>ReRuB-Irxn=#WinX6b6|mH?UuMfJ1J(=Q0Fn2`j^d+{f4|fLN}w?0
zU9LIJ`1eQu8A5*_pZ~rbzVEmtb^MMnzsx=HCiYpu{02~Ywo%@GW27k_SQNpIdZ#M3
zBC-peTwssUe~~cPNd5q*ReC>kd=?=TWRT_m=Nlu#t-|l_^P9c@IqnZb0?077@a=9Q
z|9@Wh{}u}Tl()~GSNmA}zpLFB_X7bCLwevISLRVZ`#V4k%$IFqJKXi}H|B-sdGq<G
zf70!5I`sejPjr+qtpH33E>f+S|M_1d!M}l^5Nx^Hb0!AjNNaED)>`mZV>x!~^6R+5
z3gC0+-w4$Gp@Q{*JV*$@IC`^<%9oKw$Hs)@#Ym&p?QhBbM<V{O2tj7by?53dj7d36
zM~r1Q#pnaE<vT92UUe`6&U>r0%YWXtUPjtNnOhD&?*F8S{flw^zrD8T4nbJ}1~!n5
zkh)#wf8AM!mGVma%|-jS`K6Qk-&2)-A`3u-ta?;>oBkOMm9Cxrys?vYClJW*d3~-A
zvlyXpU=S4wh|Te}0H+77N%|(%*D*llj4oRId)RBh;#B=`H%P`C5atU7#I_!m0h)6J
znMp4}Ezt&~2g?%xYc^Bs(8<4^4s@&O*YU$t&-14>BkipTZPl0(?Yl&SXT>|!tlq5-
zH$nqWH^T3xL>P}|%>R6){0eYG3aX}`IlOzoWuSg&<UjrYT2g_}-vPmttncFe;-3%*
z><*WW6j<=Txx%Ql29&YQW&y3gZ^|;~W8JCLTE|cDF8}n(+>-(F7@YkwphC6;Bvgn2
z!^*)Glra14{v^Jg9sF~j2SA1jSMqLuUIvP*uWnfv3$FpqCoCvr_}Om>2?y4i#&Nyo
zOumW*zK@e_t3=Lth`M&5-Sz(Fyz=I*|7EW~6uEF4guj7kkbwf6Swbz}UA?GC4ga<b
z`Q&Tc>G^Bbfl4iDKVXr(2Yeg1ECD3XU1G1%;P(``>gT@x!~f5F-w3+B_s6Q<(*M49
z)}W4fAm)G9B)1NR2RzP)77g{RF`PD50nD>EY@nJ}VVi_r=%&Cb0AD<+%`&grYCox~
zd?#}jFM|){$bPq|019sUc%RuH+bkSsLQZ%`<y%fJ7p+YSE0W7+(WH0Y0&)Jk)coq9
z53tPXd<biP-zIaf0YBJxfN<Rs6l4bY2d~;|d81ZR3;f0lZj98O{Eg2Nk-dOWFT-!?
zbZ`7u7lr@8QvZ8%{Lk+!09Ptl$Hvp$Oxu5sAq!IG6_z|j6Hn~8HWPe#RuyvWl`M|C
zm3d^H>%?{7Dw7{@0+xtf12RhjK`?`vGl1-27QQ*lx`A6rS`rW8aTnINKcZwxKq|+Z
z)cMZ*AGY2)D$2GE`=z^4x=WBox{+=HQKY*Eq!~iGhi;Gt5s`*LLOO<$ZUyP??tRbm
zKJUA~z1O!Ef3RFGaShjf-De!f?|_Rt{+Vz|A$g-(t6oW6he%<J`fa}2l0tlx|L2FM
zAtEi`ZwjgR_dV7E)M4%rD(+kkcqH&9^Ak{kYhC|bzWdpm{896N1sO1-W`nml#nF?m
z*8Z*0J`=#x4!fo9y8&q<q_KLobyENk@1x*L<X!i?Kq5X4U4HnpED&$+EjNUGRR#DT
z&x9|5oAeLV46gs^n#_AVuLjf;xS*5t_WqlI$E|?yc>jHiO~lA$$3LFHTx%}T7x+7+
zoKC;y7g7(9vs-Ka035v)Af^u2<YD*yNFBjqf#lQ8Xyd^faA};SJHUNgutZ*FNe~hz
zlDO`~e_Y1CeTsGVWAxkA$)ori$Q#aS9q@D&5O6&FJ?DN>>OlwImvk*ZJuF86!x{Gb
z_NS|-uH?M?b-=er8T!8cb~^k9n5<CJjZB5X?V-@reL$u`(+1p;*@)4n@E(bb&a_(I
zg67kC5iL1e2q#r1d1j#@uek)gp#V&vntJeDE})~Z*3(uP^_i)2%>Gl4x(>`wDS>z^
zq?k)Yd~fMFAhGX2%n+Umt!AAlFd;MFcIZ)sjSq&&o?6!DdeLLh2f*BEI^Z7YQSMlY
zBOMyY81FwGvmZXdXBc=XxV7j`DJ!bRr%4FL(}9YdIIWJc-T=uc3}|g@JYOH+KON(v
zFS5K_g9SYG2i!-bQeW(wJZ_kL|0Q`6rHRAs-HF`=ga{M7`(TxNf3k4K<xwwKh}tF$
zdQckRoL?#tE)`+GuTLyfFU8XeXM5A<nDe-2^0dIEHM)HB%j?Uc=kVSwoObGZ0kmGf
z#6b^hQjbNw>JK*TzewA!du5LVAGUR*zPWgoZ2?`g1GTzcJ8<s00sdpwCw;M;1+oRn
zT|N-Y1F^?r3Ag{w-}w?xzPReJ#tz`qmc0M7{E*WitA0~@q;iYUtTuso91WLt#-E$@
z=%@(2)E#ZPl&YJ%6iU@uyhyr%ZR{4taPj>-1rk(o+Rwz=Ef&>*jE7%lz~=>0jr+65
z@ZN&)0{uT#&T9-qAZlRd=?llhw8=witnB2Gi(>23YUe|0m7mo$Oi_~W(4~DD8S%Cr
zA9(MFxO4(<jMne1oL7OrOb#LuEWQBE9{2s3)KlF&VvU_R@GiuG7gg#~0l<-4;N2{_
zTL2uk)|X1I*T4(U!qRa`V5`0g;L5st12rDa<#NDN<@mL)TVMx37$5}R6Wp1&1dHDy
zdV@NETFS!X<@|#7Hvs6rUba88BO!Gu+c8u``>F2y@($o7cK8tnffKG;w<J6dpWrjZ
zLqfPa-(uh7e^<^XO3+o?)lY?m+ndn<8eke{#Tfiuucj;V(+zqiQ`j-u4yvX&{Hcp&
z0)Ug*I5c+~+yUf|tK5%W(0z;7sTH?$IsO3hM_{$9i00z!F|_gL2+pmCC9{&J+Hl=%
z-uZD(z{6~S0MfkJ(n%bG42m{p=(AsWXP>u1vAJ76hKBY_|E-_YcqMyD;qaXhC<qlp
z0YBcdlHtAN+3>4w;EgR*E5L(RZ4oEZhm)A_#Pgcyi8%K-REEYG2*;`h$_)d*by6xi
zGHHeEUg_ITDs7tecB$JUDLy~?4H+!oL7iM8vS)cf?E1>+<Nl~ZEdtwyy0eeY_0m<F
zO?gP?D*Rd*HoXHLY9>b@%P1&lZg+mrOz9=T2+!f`O+R=o_xP~<^e56rjV}RDNzwEH
zC?2Lhh8JUyAHy{m_`Y-9UCWOL%i)b}7Zoh%e(0yneaev#f;c5~h3AoA7-=o=7WC`D
z2xrXv+B6nMheA%6<bVa60;-(8KpD@8u~Ko^3-1u-3oIh0T3oc07kUiYKo}KF$rUhy
zvzLcBMnozyL-su^B{2A=0uF^x+fZdl7|5AS(3VqPc|3Pz@CdH0NG^S5Gb@_4l$UCe
z4hRU1zx`A_QTGjX3~wF#^$36QryIH|o0fpF>T9Mf7DN^-7ObG%zq;Yehw`B*kA=G}
zi6gMW+mJV~bzG({05m%vSHF_bxcEBnR*Uqg{-U+Y@v5wd?ID83=c{`3Nwy%`FrN6a
zbh%FbBuaJPHEJ6WJfD)9{729H$+hd9&Wv}sWQ1o0#Wv&8MWt7!+64NlI)(%~61x>{
z;>V&`K|WVRkR6T6jzY;UcorxM79OeA?0%_kB4?WU+C`pmXm?CL^S~lE8%KE{-@vB_
zJQ9iJ&DZx?|5SFW&x3%>v88iTrR!}@po+=6=uX}wt-S=seW2n1jFmbXGV~hTkluJ~
zayp7J=b6}xtO@xWD)%v&FTP3Wed}|o;_&qyf}nt7421K8DMSMCA)1#h;J5Gc-I5S#
zUT&0d%aj(;ryfGZxh|_nGlz<Xw0V$<C5A7}y2)x)9dqh+c>g#q;{NBk)g2QOWD(DI
zx1CP5>Au+h=;oZB{9Wx2i|v`X)OUEE2KZ+&9p6(FXQ2U~#<LEpR_6W!X71Mcvud$r
zVC+_xW2%|MSe-(UP<!~Vu^B#sC!t5;Ym+C;EQ|zeE$++XCKR&BD!MWdDNWh`x!x3N
zpU+K2&t)A|4%w1Zie!+NCrnudlHQ)n&qbd;BiWGQvS$#Fo_h}ehDdTSLCTw3ig2Z;
zjowxHnHH<c`)G!9`3gS>Lsl<9XYHR*^kX=XPb@4{G6z5NG-CSlRXK#!HM2w}#K~8a
zDVbFhz+03Rir>g5l0+;$=jF-;pmmpj;2X;chQf}qSEq*-vw#h&j&<T~ics&@&WEvA
zb*CY%k=LX^Cd$xPP-<bMffg|@p}5@VNZQ-6h(Z{CnhD{7LK0_)&a)G~Xb!TCSk~x|
z)<@Cj#B7Y@m_@DH<hg&7kTzJg6{dD|lb<JY7hwcy$<?qDSwB<ekLDl|Aa~x8)?jt6
zGy#T=B)GNm6q=$tOsaX20a_ugtoT+T?g{v9j1wQ8t-Kgqvq)JHW-hvbU0G$=`G4^t
z{=l-lquDc4wz_RV!*C)GwvPn(w>4(6B9Y#f8_`s4ySi1;#mYUu>?bPllAORZJmOI=
zHW;OXOCxg`k4mwpe!`i3(FkPnDz&u;>2F6jYIq0tiUy2c4xKxRYJ0WJj6dt=plX?k
zPq`#IFp4f8O7x5oe0Y_20~rWqafm2X21%aMYgbN>7NDQYT2hpUa0lEBtpUSC%Zpp?
zr&DfqE?b8LY;~3_wKs}4#yZ;uQoE};UvkqE4ge=lNuFrVE|Wh*sZq<=r@{#i$&H6m
z1FuC&591||iE`D$YoGw;u&<G^XdA<qql!PD^^i%{pXm>x6vh4##O+Tgisc{J8ZQT?
zps>68@6(6gq>yV}a8Siwq+FFch`uaa$&X09fjH}~aQkyA-XB<8LKsjLI8k{0%eegg
z*S~Q|wr|2Kybf)71f%4(D_{wczms*|If(*}Z{*D-sZ={#2$qP+K@kI-X6{t(PMrH$
zFUhyrTVsBrP^x(t^5|MVRx^$TBS6tO5n>shGX3!vd;M!=%pJNO8I!VitT?qeb`5~R
zY359!qk~RO2sAeF3+nb7?bDA2N?-?bUte{5<_1C75dRg&*RP!M^7g|vda24WmBg?(
z(xtvT-J@J>Lz_Tj_UL%2!~zA1b%FS8{e!><eQ0|9*V)}?XNLNbZH_L-5HTd|AcS&H
z;4=7M%g7)vP4mX_%}ZyS5Go$FZ@McAJrsshn4u^TPu<!CC@vJ!14K=pDP5I>N$zd-
zdvZ|MH&5n`!S!DIql0R)c!M0cjBi3#<45v)enx2-X*C&|N=PJwrlo@3(9sDFz{k>2
zJAUo=od^*e9@GRC5}5cvTn170z=%Kn_`H0XF75*cfm4SDjq@-95X?P+5q`>ICHy`r
zkU)QZV+_NHK+|<!=aDQ>W@_@K6XR*g-p(}~^8{l*#SrYk^iw7!J`iEUV8l$uBX~GP
z*ooac9V6SQag6lK$_>H6VvBWu!;&0ohw)ndM4ec1`Bcd+e>h>yKEF){2Z<d9_<G!r
zhi|s>md};;JV2<Rb<K*7m{TU9Q;v0B5T?VrQH3$|Hbv@fVcg~k=M*^VHnH@ls&9Jx
z@)lJR58t@p%8PiM2Dy%Nqg&Kf)6;yFUteOUW<Gl6eJbQi`37+DO3RA;t;GBR`Mdhb
z$DVEc57|=HR!9;TBK14xll9%I0@8C6P}mop{P)saB~lZD=3io}s{=8;AbZn+#>k1<
zOE$QT_Owdb1C|s_?n>X!>8Ibk78+dR_%K=uOq#35Kml>Gd1q}me-~k?t&f?Hul+}m
zT~CDq&V;@&76!bP#ccZ4x6}Sl%c!@Dk^fZiLPHhB@_Xlx?8g9J6n8ulYA(y|MNJJp
zE<M=&O(Urk62Ze*_j+p6OyAoEMy^LH{H=UvYG$ZQOi6gV@FVG(*MdHYf^P;Vir%pL
z{9DqmqwiimaSPOq;Xj(#TW>*n$F^kvWL7NShsaab@yk&(pSnaW$1(UH)gHg)poO<f
zeuE2c|89h;Q3y?Wq9ep%G;Hw6mV#tQ(QnOw+evp3dxIH_&T@)lwVpv^mYGE)6b8_N
zLwGxMYa&#nSQ1TbTKT@&t0Alq3?5HQ^JqMvV2UQz5bi`<@sN|IF(=C42XHbA-q(0E
zBfBv425ycHd60t*+B=vTC@3mb>S0ofnqlKa{Kzanb?^4w{>kr%Eh+I!-B>6L)A*K7
zeFG6c$lG=>uhfqh^7f|zGebT!qKvtt#5<F<i!K_yTe+*ADH`xEMin`Or~GL{IE|Bd
z=7MR-rpz+h7yGZJ@4ZL6#HfrJvykk9=%OR~UnUm5r3T$z`WbO+NuAbv+8v!<-F?>N
zTM@z}d}E-CfAJigMHe}d`&YS~Ygqo_uUPWTI8m)pdKH%ke>^$^U9ns1%I3qz$6;;B
zQgub4PIk_K`Xg`jzGTYKPoN@4Y@E|JrWF?;(N__q>I|}Me&-c>jL(~citxYV@*?%%
z{2|QgPDk{X=5C$QpR!3Gun9Pjk{sjsaQL(Nd=yAYD7XLoFP5?-PWnB=U-ND&5)1<W
zy8-1s)R+oG@t<{J2x4LRHY*jrBH?Ww&xTcik%WbackUKi5O5uBvDiGA8)2+|Xn)3i
z&`1g}Sn3~CTi$*d${|K9?H<bWL{XtTMqo%J`Y+GqpVvMc(rGg(e+`x7RnzDPJV7AS
z`6JM=qJe*5FANGf0^>9p<HcZ>m2SXBL%Dt$kf1{t(DkzXI}yNqEX}V7K%3pus88Y*
zzZuZPgYV^q0bcSRO^riVATB&5a_o}sQhbMuls}PH?l1P~9Fd%;ou6U(OHNt_Bs6e7
zC0JwW2<aw$Hkj_nT4pH&Z8XyR`T7nAoRJ%jyqw)z4FmzBg6hlz7%T-?v{10v-xEs8
zK7NKJt$>_;4dC@OY5jxq+THK`@AaJSV;zDpgZ01jaPE-p-O@*3w~kloayo1y`N|P!
zo@L5A&}X%nJGJFH=xXQwik>PNuh;Mpm>3WYu(ks8B)rq4=W_XWzEg)f;S2A6Tr@$m
z{VXPgZ_^Z2(Deq_OvXIRWpW|Gkrm<D7~Csmn*R1<B)?=9AN9e-UZ+J+X&eUNb%cTg
z6ejAQ9QH*Hf}kvzK;x)dfwO3wkY4-GObaANv+m+>Y+EKO8u>9uLuD#ne8*^^<^Dl(
z3#eUG^B>X`rOVSjT<ls**i{pK_MFfYKZQWuv9iX>evZn6FY1Ni#ciy(&bLo*8FTC7
z(h?5N586EVd%Lx0j)I{TxC?y}w(OJlGqu9AmnSQ5^};T3y@$i~zKs;692yS8K{n8{
zp;LnDwFu0F{;`xPKoJ$h&pqnP;H`Z)%|Q_*u6!{q^WIHyZO%VeoQmSIMV?==GvoVu
zEB%Zl0QI%uAru*?znGquR9=98i~rC3hJ*mI?+So;<y>R}z^U{Q*q97<fTLRfoa)lX
z2+dfU-1#Fg?H}s<dwNSt_CER}_4>4sZD(ezk>?WFhSFf@l^^Ks94eWw5Nh8V5FBtc
z2I@QC%Jt>?g}L91fw2;$>9_iqoKvr^MskfA$wK|WnCFLyM8U%o@Y^M>SA9Je+@2rC
zArnIvLLwFp3~)vO;F4H9-N)mXqh{>%xUJ>#o-05PHu?9{r_<f{?F)SlHx1fb#fakd
z(i~J?zg^7wslC2aR~1MlMz9ztOLBEsyIs!PL;wnwkz)kx^FsI%f#;Hn6dozUPP;Gi
zmqBx6MOo7)DORSYVK{;s)i*Nos|-!Uu!8h_1()v(yfF}@;atFk7h48(!Jutm7J;=M
z!FvtI&0NUFyzSy@**EtH_mTVTfjBSyg)G1!0wV2KUIEn~V|ZCtvvHYx-UhXAI64NV
zHlIAlx>VUC0G6~UObps93$_ii8ng6Hcrkcxe~{Q5C=up@#TTO<*W5Md2S@N57$3xh
zacz1$2j{n0NWG>xlU|gOa7r5G())7V(9yrIxt`M@geUwLLCh6j3~QqYPJIWh;afuA
zvg#@-YgP^!19>oMPU}d%G=lUC1`B_o(w`Fh#Z8Nf{^atz+r4@b+j7tZaBGUy#=Oj*
z3UfMW2K0hOAG=7~cx(X%#sM|x8A#4xjNxXc*|;#*%kId!gx0a!Dap5F{oA2|K7lO=
zi%%%ls8mx*#USE9PV9WH746}f$2;toAtd|goKRW^p%>iDyWl*(sg?Pw1-~@K0cg8w
z+xR$qKiKgbUzF?&ak$siMz(zpAS!=1Y61YSYMS}+3P;m;X0m5uH;t<G0$@L*U*&?f
z=g1<p80)~XdbNE7&KWw9j(=@py@a_)a2C*mtgt<};LjvK?Y%#7X##`=%e188uCDuT
zyG{olLhw|xFcwgUHuL*N#&y?DmgA)1^F}=^<@o;p4vh8aP}VnwaDK98B{Z(nzkoRy
z011@r!vvl{1`pcf`VF^j$q#aVZEfCbi#Jw0s7+3Wb3hLk+wqn;a^t83(+zBo1#jk0
z4Pt?_xW-S61PU5JKtmye?M4Q+mRF$1db3moB06dZQuWhemE;_a*qfTh7Z|WtiZK(O
z_C!BtACA;jtc5Z5nqtX7oMn#H3YZ8J>P-=xvF&(9nY1UFbbJEziR68_r8MB>*WZ=}
z3VvznWFbk)1il*tFex&A$Oolw1p1rN7)n3Q3lY{%g^Prdbs+%S3%+!r?RZxkhsR(P
zm`rrRsOg!0$raE(3YJ`nIBDk^769DK4_{TUiaqK>9nT(!Jorb^0pJtcZNUav#=1El
z1LqqA<WEKEJOgo^INEf{Wz&Ipve{ob56c@BJ`un034(IbjtkvxD@^03e&0$S^i+|h
zMKUS_MhO#Jb1!IjT5i_un$u`A_W@NYaMRrT{_DtnP2Z6b<su^I4W<*EKnfLu(W?>g
zxj^AtAXz+-V&16wAs~z7CnrXbUQaQ_8UjfqZ=4Mx4BqhEs!fu`4k*K<629*utqZbL
zmS62OD$_0}Ml63EU{25Ar0Xp(LUIg_uOarA=xI9&c)HuU980Wuc4G?Byv;j-*vtjv
zR0sB;a3%DSImlpX1`V`&wXRWrQIvi(P0HidHwwId%lv9aVs?Wf#y+DpY+EDK6y3DI
zT7W0q!Z0V9{LE(JhxF2)*s0gK+jnFJb=&CXfAk$HtUT#0zi<W8Zv0vovKnHLcGu@o
zs3dC%E)YM(;r#JErhHUDN#s=2sTOes!dK#V`18uR8<T;|t_UYPSlo57kSOw5!qk{Z
zje1QW3fOO|h{j5e@8eG(7fPW}=CV3ohAFfdHSMp0QGTm8?Hy4JcJc^nP71FO@r0?u
ze3B=S!W!a7A3ueC`uZXG^FT&RoHY6-WFXOE0{<jot9nJ|u{XD4RY^&~?YXuQD*j-g
z^Q7_H13%E|JnE3R6a-rjLs?APSGq+;v?D@P&9`$7Dfs0U$7Anf)SqfO_=SSq%yIit
z4unBHrD+*+n?@qfcn4w3MgkFfwmzf#F`-bC;FdJjF!Gt{5Ne|WXHOu?q}Q+=AYF1D
zq!H<HXi6!y%Sa6sS{(|Y4h9SFk{C}|6X9QKkZjL}5djP+Um^AfQ);ecV|KRp9?NRb
zUFE1i=3A{@zG8=UbV@D|*8yv>nLCS;*)~7Ob-pn-9i~W~>t8iu15R<gYgwl>LawEh
zdvp7V#5Mn_=t+9aavWV$*ayK!>Q)fjo*o%4-(h@kkAi+tmHc+}HYyM%g6%Xmi?RKF
zn=E1J1Aih?vBN1g&3X2M@!YxYF8W0%G*DQ$ka9ilr8d>hT*E<WfVDB95z0e`b5XYJ
z`4F<l^DZt+S$C3iflZ#wfHPo^fq`T{1?av1-=0uGntcu=W=QS@@_onlOiQ7dbu<9_
zGZ#W}(t(U$U*RR-O+CoAC^?8=cNO8pf1hwN4h;wW0Hq7?#1zX6IxSkrUe)Wrs03)L
zva`FNBKK{<paqHyFBqc%>Y$$JT#{CJ*Vj}bEwtLNGBQR}jI)-onhKXmm-Qf#=KN0U
zsg6pw{nJ$v0gNaLy0}<I&{c6%Qi07p#g9sHOITc1J-N?21y%N=FBL)Y;-|2Jt-Vh$
z1RAShviayAhaYhB3n3dGjVx_~>Qw$ivc%XdzOWx?0BMO2%FMGEHppsgKprgRp^%Ic
z3*N4yMqFGWYHwAW^Ze}R{F`=>WUIiEn%NjghBcsiQIQVLYW|%5mG;J~rQYl2&i~ho
zc_hcAiXba#J9V@JWgUq_1F0-YaXV^G`o)r6+g#GbpM@;wXR1THtpKd?97&x^5quN_
z!&^*z-_34(63rU2FemuZxqEjZUb0zBK>hzx<_HbwHpgo$PQk02bI$+RA*vHF^2J^`
z{kUF@DPqAYV(Bf!P|*r4V0~>kh1Bsisy@;`pNr>9T7Nrh*M}5Rygj#u{IS&)YV}-s
z!s<ej-487)ztw|Wpe$SVPjX{1Szl?uy2Q4LpMu)5JPnpAB615Pwj@uY=V;}7D4O%0
z2$UHIeLbd%^EOCv$easaz>4BPY@<SQ-Y}E88os#`T(C=AmfZId<ST(hp>wAJT88>L
zFL`0gb2AU`XePaKp_2Mym!L>_E#tewRzwA`Z*+>lGO$o)99E#?zTZhrWYmlPYJEuq
z4cPD>%;jr?qcU;0TvJzcOS@y`=2nmdaV+r{5i$~-Enf|g3!<w>X8$Z@^GQ@#11N`f
zx5XrDxT^@*r=b;UF?1u9#mbwBwqxuylu_5DS3q3dkk}FZd(#1urc)9U0Zd09P&7gd
z#;De(IX7=w$`mC>XpU1fGP%{g424{N3`1xME1F^(P#NxJsH=t&B-^y*O#s1eq;g=C
zXa*4l&U*6J_qR;bx~J|y!;uM@IDBhWuPoVOp1CE9k|E>Up_hBBrk_B#kPeT1ir%-k
z>V_sVIA`WC#}|!)%S2Tu#;oM$c*tK={7oB7y;NBbXXT5SUi`v;<`*Xgdf>mTsRUW;
z_^|vmt_1|;n1lj_7{yvBe^|Q)-bi7zF-+j^d+Un%<l6K#mT?aZg`G`t#r*!<&L_;R
zHF{U?Ro}>VfIW|ohp_WeTf1^sQtIGcJ#g_5Rx@6rI~u_-JaRuJ)Fwsf?UmLPG={ZZ
zw8}{?Ow3*^{aep`vfJ8bT-T$w=arE&G1EZwS8Bn{bb^4CQQVSo-EnS}+&8}&zbUNP
zNs4Ip`euL}^P8dp)^qF0SsAI%YIZ|t_i5YXQ%FRs%%V68c(%*Sp=JE&>kIvDvu46e
zy*@&Z0?kEDUtd5o{{ftDQ{+2_rYlR)&%>=cU!XIz^7?`oP>KvwB7IK-ZlM3&Juzg6
zDgF>MGK`JqgAj677+KN}E}<K1oW2ATQXU?2IXNv5HftpgYh)FtdnV5;)?lO-Ide1T
zF$i!5psf2>#J%dSgo;78*D3xeC8{F_?F!M`%J^91_tJ(hHSu}AZ;$Q7OhQ5a%q2sU
zfy0PX2ny<qhIaC<u_Y^mS(Q;zyf9xyZk=pxkqPk_+fTcFYC7=)>K7xq$)52>aqqkV
zHnchq%jXjjW`s?-><n}s2Ic4=?P{k4{RpiK{T<Q4K#6JYxCHGj#+0wLV9|dsjbyVA
zi(LW=$v86V_-|u*JVqM`5dn>D^N|Z^LR9}XRCRh>oVz44kgPW&;Yi>4jDKVcOlcN#
zsj&#oJkiL#R7gr?`8IBL8`{>a^>~nU;io;Xv8?D^5HOaevoq1tPjq3i8XpU;)|Awq
z0=f3SS_e)GC-ZwY&QJd+4^0aQO_AC~l@+qa$R}Q7vS95Yp-8}36V|>F>$Gy?>nL!T
z;mEcuD4=|gkusWuS^*O+cd4CPT?W(UQI{lnPJ-^;3D_JqA2~K{X0-#>DGyPDyu;)H
zxM>MMU+C4+CH?rers1yQNTH4$NY^i;f{b1LeL*q-8zT|I*+INGX^4Um0sC#RbehJ>
z&C8f4eYk-*N|FNkeSd(e<OgeP)tu#u>$yZEj}LnRku48$VS}=P6sa@MD|T0ku?x$@
zITT0+`PfWwPaQ!wfeEm6_vfQ2XoJYBj5`A%P$mJQBxq~7k!j2?#?cUk@MLuHx)<w5
zqrE^Lnu!xQ*iqb$HOT6*qBvRBySGi*0R00m>N+w<`DxzamB?j9ksqZ|+NDaan4%>y
zd1oA^9YXC!vDrcm>e^EqXtUVP!Kc#s0k(F)Ks+Tifsja|t;8kE<^+CQ=h{W`x@DHR
znl<n8dlZIVVeS>^NW0CtfL6_J$A^RMJsa*M!Klhb5kbe;$nFBKWvVC+VKe7>*;dKv
zcv@TK70EvX5~H&r-{Dan8wO`D<jjsQ3Q`#0&*vo_U33dn;5^mzeZ9g8^62XR_}^%_
zGAF<qxQz3&>4^wv^GC<sx>idwB;m0P&}ElpTL9BG9jWU_AXMv>(n}wjv$*pEFkQuV
zlrA+-K;I~J&<$KGbQV+_(EZ=mAww4B4eR^ltCkVL#bHRTod9cyBFr%X`NMoHZv&Tf
zW%rM9s3_Hul{c>GaZuqBNfNy^1wfNIuFo;B{5VlecV=`N)g=K3=VCAlV>Lq6qZ0V)
z!xSujPv0}u5-hK38mvRe0?Ut5ZgEM-W<Mm(hoUZb!Ppwt85KSbMnB&sL;HrdAs&kp
zz3%k$9hV9bVj}0emRWKLJrDDm5c?tv<|Kva-s)$TxVDW>6tat|G7jG3>MUXr2Sy~m
z>h;WTTfNQq(?;-c(u&q{w4mKOoX>EYd?IGVgvT=^*Ka1?I;LTJexdNr!R9}mLlabz
z4=0-id{P_i(lQdG;uLD9qDo0v?fhVSyUFiSU*(wotq5$aLV!3by2tTqe>~pqjx{@F
zw4HZoYYoFd?Hixn0E%P7q3qvo;&5mV&0bK1XN{eCdzDklnn7)@CgWnUS;68XRfOc0
z|5;~a|9~#=kEa-3bapIz{9_)iBrqkUw6GHQnSA2KbEXjDkql)DOTQ9jFjHalZ-7$Q
zdG9KD?leKy{0I<FyzyMZ@BapssVp$W3oUtD>=pKZpp!n&vDJ|405<H-C?RhZynQTx
z>yqM0-G4L?7W_4XT#yiD<H+eoNLFs=t5J8CkS-dG$JEN=oIZt$5u0E%j=g|H6F(<X
zL~;DJtm2|?w;<uXo*Go4D22fW>CBqQY{S%gej@>BG$xtL+X)Cy!lTry*xt-$x^Ie2
z67*-SQhG_KSI*9RWG!(9S<!e+4A6&Y+hjR%+=!L@x&v@)L0o`lWG{}OT8Dyual&(!
zGfidmZTw^$r%HY1G`|`gJoxofd5oHAcOXub;%`Cnu6GeaTnoka%i8LSe}Hfgx(D}V
zALGKGsyn%1pLj}2_G*d^OZe4f#j|5RJzFJO=W=8jb|aVrBRSWB3N$71V*Kh$JhFVT
z0Xy-+qur^Vf7%VXqXq0WjUMb6@WFig3M!N+-+f|U$hvRwA5F&13TvPGfDt<<InMn8
zv-Rn`RLZ{^S?_|{U24J0z11H67~yBOTNY?zozD;7nFc7vt0VT4rR1S20=Fn2BzT=3
z7_j``M~F2LA)0xe_t~dsi_hq;R*dl07#694Puj|H{0fjsOkgPAL|a3s;p}TmgEx?8
z1gg6@%s3nwmK*q0?8V6VwmbTwg6e|R-);f)mY`Kug^srnYNRs386B=Ee|NR&r>Huo
zRBs<{FsfxB!b)sZkTSA@>uCqATv%l?ZSp7X06YtBy*Ihu`{fk+yb2Y7gaLXOmJ!v-
zf}cXZIL*WJ0pNf#qMf%6>&>g&<3RPmx<)QxiHO}V<v4$rZy}nqS&sNE>b0S3BrqqI
zJmXPtqb_7K;FM8=Tf`iF7XSgHDBQvn&Y4VD+*rwRqL66$f_mWWKOCijv1K9P*_3%C
z57p*&6){hUF<e8+12uJEg<#)&f5peQ9YNJyjItzh#>adFtm30qu+6@5y|D)N*lmgC
zCw-`FjQBk=nP%H^oAYh2@R|2m6~3>Vh5D^gS35O0JU(b|`R}3cY%eaQ4q_8qIEsX6
zu_l(mXhGEFgeP1wi=*prbSr_4sMc)e@S1o!+k}$uh{sZdX<R`Qzz@V<-7`?U8}tM!
zupA`r$L-RL1daB|nuVozC&K}24Pm#Pw3fpzOiTBF!bD4r7^^7lRk_xa+-_5vrKS4T
z*$gT-{Bc2u221%Z)$$MWq(=M`l01kwh|ktu^mTt3Qxtj$)L(;wH~QPqcFaY)s;dEX
zkWmslNy$g!gEKpH+q<va+rsrxU=w}q$}b}=Qb+$R4-M%$W(ql2%(b3da4Pvay1+8P
zi=f7a%W9`YLJPD7mAEw|1NRe0$#C*3jZXoRk%8MZ#l!|^YwD+?-r@3r9Qj?{h90N<
znv7d@(t+Dz<DVk~K}*k1K1RgplaBp4rW+NsCKy|IBxC1GbErL->Xq6NxOKv?DYrvp
zwIFwwaMS{RnvuBLVvX6Nw4wurSgmyEiDoi-r8qTA!(-9L5;YiA7=jH6p~HaNm>A;=
z^J<0bMBBbeVuVAs;HsQX&eq`sbR`SWMu$Y3IyGu`6W$~dqN-Wu7JFjb9K);Gvs0VR
z-Bw^@WaLlMz`x<3&tM8LMA<l#jgJ7S)ABY0@jg{ZY^DBFxyxDhHF+K1<7dHoLPXW6
zv$g}H+&o~5aCYu4!p-)6=k7mmPl&^~$;Zn;++5D@H{DMC0_Q)!w27CrXi9LW#;2qu
z^@p*>6M+I~ajNaiNXyYIk+Usmwn@h^UD1)?PdugnDE%p~A`k$Iw}#1+X!s&Zj1Y4%
z7o<OI{OzFsT@0s0SbYO;^CA%a=a8Ohz#L##T6_MTmO3a*`Y)p2+t`r*!+5CIgYo%e
z7p4r&Q!pe}CU-g*r!I}MLK=?&4;2?J_}vqFg+kd#35#wwkYG(*pW~Q7&j$NJDL00-
zkShrDCbaCpS-_*;2V_4oD0PX#a=TbSZ(MFPyPHlIqLM%_sf0=6^DbnDj<0*H;anD_
zAz#Lbj+<}YfI=>z7BH2Yp5*xKmdUI)bkr5cyH&I*YvE7}xKP3OC3F*S2@prx*qQyd
zGr!Pg$zf9L&x6$SQ3#P0PbbcuqV_1R(c9RIbO;b%!7ymvnfPDOjpE1@N7Jn}`YWJB
z=VQ{f<ORG}*!c*kJyAhmk&bBbC^<4<0xlV&Sgg_==c<DE9Z)Do<PXrWyOBc=$OmHs
z{p@%wT-|{-31w9-HN}hrW=g}KfUQqX%l{Rdp$U~xEd=@<lkZ~$U~J^HO_GGkS|n)A
zt*b3Ip{V~xAyh!)%bs|BW7idsFzaq)j1|nnw*TNBYmvhNV6uPQrJ!HP&|r3PHvTrZ
zn~^gjDEqa`xbYR#)D5Vq@RD_RbKfPRsvDw`U*4R*ArRqyvnd{NAJQJlB7;&)mz0(|
z4_`Xe^MSU1o`>9WT;0epWz%&hPV-ji)t!>qF+{@bToQ~E9lxthwos`~Yb}tIQ*hc_
zqB$ny-C!B*DXN%6U2^qKUtPlmOuHhk%(4js&-rAG6iMXLYM`M$xkJxE$ga|NapDu3
zG2T>hK$NFUQhhMr*yi9iwOmJopi;3>a!#kLD4doyJuX>mR2Vn$L_cX~t1c$LXq4)(
zFbIgi#s**ZS05UpRpt-;>a5Vw5b3?G?F=IqFR03oFBnxgA^^|dYMi6*ssqRQN4Q#4
zh7)KKm>*#IqBM<JT8tHcX%PFZDT;bBxkT^97mm*9jgfUcS_C9Fj>q24oN3ppTaz>&
z&>}3H6+1}slhexS@irC4jXMuae!QORm$YVkJtODdcnP_T_EF3(kw=$_QAI1`VtyZj
z0XQiDog5%dNAS;;#LIcd#70TXBPm@fYodrhUMM_?*Z}(b=B0c>?MXVx(g!I~_qtPL
zqc`jJl<Z@qnv`P_<jE6Wv#5Ot5OjHC?1kR^!^C`4*)RA!pB%9uiR3m!w4S)<1bxGQ
z0}o#3UWT-;&1E!j;W<+2``N6`G~b=M;2Nd5i~}2asVOh@+XLn5l&h9N$tXNs-0io2
znHl6SW?;^@I)o>~kKI4o-K8sQMDk65gy%mj0rxEakUa86Q=zW#H@Nz?MEoL@>!^IT
ztF1o1`_644e^A9Bo&{ugtr*1NsLMkkB)Vvhny22$CY2I4hEKpo#>e!<Bvb9YCD}5s
zwKMj%JmgLe14sYI^rh{Q?|VZRd3c^pMWwC5dd*2{7vB>B6d<|*FCkMdOYWz47w|hz
z^0<hz+xi_T8SS{W8TNPRACFNt_1x_W_5NHjyX{45joX|rTenm3_MeR`k?wdvsGjxT
zMw%(wmH5q))dg#K+1KLjkc20_CXNmPX{Kyd+Se`r#TIcHV7tD#v*T*4T{&^JgPnm-
zpv4Buc5OEF<rt5ja59hYb<!~DCrbwtieObkX);S52*#_R+@r)Z4e&{0Zd}sVnGsF#
zPr1cqbY#v$%(=aCAGEVq+rI`_x`Za~A(4SBjfU*wGX$Dtbmm2|S{cc&-_LfP(N)1%
z_LBWVuE^i>1WZ56Anr|Ljd(5_pJIQPSw%kr5n`p>g|xmxJ7*GE=gH>wGy%)w>3=1b
zqoPuYc>0L4NeOsRehIBUO8igKjYnX*nyjo12_s9_4(x~+4^l>~c*bX8*_=*G&LO=d
zo0#Sml>Z!hNm+!tPQ;?!AW=YAeR?&~)QW~y0o9bEj3-MYhY6yxT+#FKRrhYk6nKpY
z%<?mGN54||0AnXG!q7ii7Pjl~hj5;nll^18o?bVkpd|Y!n~NgWhD(!Bw<*R`jXnt^
z15BXh;#`x{$^5eD=qbi$!d6Q5KxAZ4K!M0Eyq?0@(tmDsB%5V15jJrC`#4OC){l6c
zU;@iBh5rOOEdho6t<2x#ewI7_^<Q#_llK1EmdbL{u&-+N3A9SeJ+ZB3ut3%jmux>>
zWa2v*KiZQr5;5XB8$R=@(#ZuIK3e!XDkDP=d&O;I^B_Xv05arhA+0C!m7k{Z1H?VE
zEhjD|{@sH<Iqz`^gcr+<E|>|~g(&lDQvAeSS?-X7EnJ-q*;fH;w@o=O;Dev@nzJQG
zc>siwUKlJt6JcD9hGtJvpmer*Dn2D&8_;7_jS><qmZTd~tP6}-Vt*)ncuitUTP8JN
znlkj<)gADE;QKf0M|h|EKz<Y0qSGr*PtR6f6uqSeTx$M%tQSbg28&DrWD}m|bQCr~
zsRm%bCRyk+zm|?fw9Zd0o!xVUI#m6HdJ<7<3qN=Xj7dqO6w{g`rAV@cy;Y<~Ii>Q!
z^G_1*L8%B95=rkjaX{8FNm)|w2d1Auov#8N7~-gZz?_uque|B-))zRRct8HDByjxs
z11G8_R*?wEoJIwvOOu!PkCTaK86$tT#ybR48Wspa-9{fD9g86AuKf{FBSW2ONuiA_
zC)o$KYvlnMZ^d9JHN%7+a6IuJ>5>bUYi!XT<Y*Krq6Gd^I{i+^h^!1h1OxhKb;IFp
z%WZ47=d7BypmT@8lw}yVDDSd@Vkk-~`o84cXSy!?7a{jEa$A1!y(k>712M$7-i&5M
zKJ~ni<UMB%eMXX|?X^oX@g<^WmH-`!u+NUNZ`O(rPE}q9D8KbC#huElcD~j?Da6Kc
z#_vJzzj1UWa%7-*bNQVBQCT%lF!|~(RPv(e_74|dL5{HF2}iN5V-jd(_|+(X3*VZG
zqxEE{AW%p7EH(ZdI@A6}Yl$;#f^|;lHm_NH^btO3Sh9<>0mgog40DQlV+y`Oeuc|H
z^gkv@8QSL#_LLsj{_M;F7%rIN?WdhJzjX8+?(xg3&H`>=eP%HE;$hm0{?shqdI=xk
znSIMF?*gNo8LqviWiTy+(o!ckac?dF2+o(3COrLJ9tXk}e2kibf_N5>u0(f7Km6|e
z74u1v_=vp!C$6Rnl8*~S^MZ0Zc$D+kGOy}dip?tr<7`3?$|0LtG$^3EH-zGSPA&9|
zdH#77W{Z!J0fu88o(X8sfaD1(t8?mN#PpRNO3=x`w)x_6tuDc}8U=rK=!JQJ`XZYq
z#rx$g%k#L?UDFLZ?Kn`x3ya#+QV4=Y(H1#jnB$#EbxOhfuf269`S4aIKatJFL#F-m
z(;nyCkv~nP6+!ZUm9#YdKfkTK7If7iRq07wJ<y>sEuJ=iSH(Zq08(;obsfIZ(D_SK
ztN!7O=wKR4k5rTz0rv&VfNr#`0+RzfOH`32bqQDP2(iaLD&P?40qGFa_*Z(B(fT=S
zXWBdD-4?gpavc;!b|HiFM>H;i_?hMhH=^Mzt#^~Ee>AUzCj{%mDlaUw#g%0T-veqK
zm?wbw;3)Cr>HqHY>s2z^v85RO!l5#hGmC$z$&V|&QT#~V|Ko&_x8RY3HbrNuZ$0>L
z2kq9mgDzP8XubbgfKi^1hEUd9uEhynJF0b;&yJ(77xOF@FIrV4ch|o~`~G>Q{_w%1
zX*bI&VDz6?I}t`Ciq&GReStJ+c|eHFBHtkfL-<tL9ZQ{65&VO^DcvMLFC_vyCgQ{<
zoo8pJgCS2q;VJj3^@TlA>!qY<m8U$9fy^J|bCEfLdR9*R6X`#Uo*)va+<wc2;WDt0
z0S5iBclM{Ivyqd2zsJmLzJ@8q=Dkk09ufmmCpQ&yWiRC+v8ND8SLX?AhGaSLLA=MU
z1x*AAo1Q5DPPPfu4f=JeFqCz4n(=A^^_<XAcS<)AGGTScwa6cXB7AcF>G+f?SFJ^W
zN0}|iD!0V4oKlEuWSPwaKtU$Og)LG1c$+;>IthQE4!HPaC}n?^JIJ11mlZKADRSCU
zw)FZv8jIR+jC$1zOBqp6*_i<K5+GYuc~kc3jj<m{)eGverOC8&9wj1`mQtQFX9q1j
zSd2GcGaY&hE&??q58gTK&2;$hftUT5_BU6B^EMJEObQP)T|7jhuV)+_{t|)I)7^!B
zcF(j;=^_d8aDI%||4rDy%3#my1VqK|`9smTrY|khi$Orf_+$}D$eb|pBCRO@>^OFV
z?1z-PXqo2zoHx#yQVz9dvA)YkVn`s2YW;t0m;Wy4RPhkes!l~oH*}r9wc0tn*7+^8
zU|U3QzW6rry1NF2aLmlm2ZM*l`ZO_$5UgzILw-*6uL0_Rb9T2-gmqEylfB_KnwI^j
z#8v%?BQ+tx>d#g^J*tJIM&48cJbDA(0)h8((%^~e>kXGOLF!Wl+6~Shi5RR^B3ijK
z22mCqSl)owz&zQ#fp)wXn7y$N*8L9FXsh{({++(b1Jge$P|bd@&Mzh6xCZ-;Qh95W
zyYl9oS3RoD6dqB96=Ds_cRJj6$i!dOvKUEE1$BN(oM#VgaYE<JE!755xQbzdM6{u}
zV3xR38nO@7A8?}*y8J;KAnbut8*dW>0ZKw8gFkTz5Il+VH;kb$Mw`b4n8?Kk8t#6s
zIUonVJB&-NgGg^v`QY{K;PR0ha6V2C!dQ^f>|fpeg|+zaDLy^`2jn67y%A$>H4l_+
zY({!G_~Q~Y(2zz|IZJSlFO~SE?xNE_O-P=-7~NOs(_GHjvrukQvC9efo@u0&pG6V^
z^J(PKicD15DQ3I_cndyDLbQxgT;67qLyXKVa*KJZyRQ4HDC^n&i&<PdPClh7>lWj2
ztvlo^`o`~z#P<b$+Rv6B%}PM<UTFShY*cA@fN;`rJ3;5I1&=U9(mPh)9(=Q2m+rGq
zzNEIv@krxrV|ZE^=%*@#FfDNlBT2yB+xW-#y8o@QItVk=QI4I$XL&)L!sc(O)6K<c
zd3IcYItk)Cb<{`)*z8C?rk!wt^6dLKF&fueIXEmb#P|NY_x-Pn6a$ICf#y6PU7qCU
z28x8Z^531{UdIKlxpNRG$ZE;Ge2Mwksq2-Ra1H=jku|oX=dx8@hW8aSEJYb9eP`sx
zx23pi;z+C06kR0iuf(qimG%JFMXL9h4EnrxZm%q4Br~IOlQC9^yxqimC5+LM9kgZ5
z)DE{qYZ?3$WTCBnp0PUcHA{0<_;Styced#_XJFKuNN>1T49A&;T*q{{2k9Hod36sm
z8EGoE{z?uQR9Q^I9Y7r=AidVAT=+8c>qOx*I)>GP{aEV`V5#ys0sK!CC&E#)qAbZL
z0Aj6a-6ci%>1jU){kNg~MLRx5c;Up`LQ^_Zf`o5MsV_~xYL}=Y5F(=P5J_`{Vh}M0
z=VQvClql=&C?zr|MJRO;AgkPDDiI=L|9!~Yy3e^d^gjan&O?8PwzHeod4sKA;%?TP
z)>|!G3ET%^!dyqp@vSFNS70X*;cWwGbADO>`uyP*wli58aqzd-VYMg9<7ulncYUbv
z0ElCM{WaKsi`Y?|-@NY?!SrgB6tg2iyZ}ub*I7a=x%sr8LCN0Nez-R(rXz+Esu$it
zn61$BaXnGd8Ysm)BV8n@Y`j}Eu1YjrJK|jb%K~T{eN7N0eE}q*U>qIoQ02{`T*fC5
zYQe%tjWeW-<^}zqEw03e**oGGsP#|SjO)0)lzz@}@|d36{X>!bI5I1)M03ALNauw=
zzY83nA}g(u8M7QVwh8f{a^9b-;bkG|7w;>5yL=us=MJ=F2d=FdQjXs!UWsk|wtL#1
zV)+}b7k%=vVfx(=AUvyW{-kz}vtSCOq>lnDgR|p_WsRH`Vx)KDqhi!Y+CRT#pl9}z
zx7V0GX6TT+-M#d%fA9T5_2H$zAcdRlqWP{njTOXf>9vwOaCJ@1@MT!!`kfAB$dVm_
zUzmR?gjBVjs??BCU!7^}jh9G+Xts|om3j{?_cg)wcgp+)z=05L-8-#J*PI*B5IcBA
z*f8PJCwRN$(oanaSPf5J_GP8KZ^?^1uKuuF2jB*#tdK2G0=<>xP5TXd8*W4rL^%J9
z`bsVKUU4rW3v1A>NL0;9eWSV~v|aT>S;7ZQLKTwtktm~6_fI9$PeR8U8GY(<kUtL?
z#g{ywNa+}q%~eq$K4wA!#DBlY>QU}hmCv}hZ-Qi}I-iKm9%-%4xO<1&-v~W1%mYHX
zz_i`1c6Y(6V$;Wss+%ES)vN8oC&J^W?TmmGi#9T=$C4&sUgFw0{zPo~Xq5$@-#xYH
z8Msj)KAkSy$GlqvB0)!fx(&NrybqI?Mv?p9SH2NaTY@!^yP18%P}8Kx{aCdTQ?h9s
zJKRgSNw0peT@){ful#)L4`Wyp!Z5IBI9_1%ryo9SyK_2|=W{E_ZerzG8fseZ)!oDn
z7j*h9eAa7vM$om1J-HcUv|0GV0MM>D+_*P(y1_|bu)6Vv!<)X}ccxR_qyj}W_W?mq
z@xI>g&%&hKM#DGhyPK!FB`6dmjCicLoFG0Nz&U|-AYo_3FpZ!RZ5AYq^lEPg^SB+!
zmGpibj{sjQy5Way7IGF*apCZr1jgQ({dzJm<hLvTUaMk9qf6z^o8I*Pi>=V$;48D0
zCEH=rS0kk2A()_Kq5~nHbKA4tQC99{|2I$YhVb{AuHohi_dyS_Cd9LZtJ=_YfO$k>
zpu}-?vR9?gi{22Mn6$H};a-ZSB_C%loVz27+Z}DT8RzAkOXues^G35t@!4)s#TIQr
z(qmL?(kwF1x^6r%IkD%oG5!{mbI+GCweU%QX2pE!S~3eXy6$o~e@-gtd9&0{nshLw
zqbV>qK5FPkc!}&x8YpMrJ!kJmZ?bN{5#rk$ed*?GrO={HGx++CEiEa1mhWag9L}`+
zzBSK3cVh`<1T_n~T^OT4{#(|14vOtGEbpm>*N04M8-9?sF*(57hv66{NBQQHW9hX-
zCiZW~!4$TV_}!EZ8sB|0K(pq&p0j{Q*(Z5AAaRz#ox6pv+xAC;wuFF7PLT8snp9g|
zG4+k)dW6)ySqfpc6;(I1+O3biZAg40MW^dm6UO6V7+>GbkDd0XABrk&CvtOkk3X>I
z_5nEKAT5@drPH?i@b{tm*c^>J#pi-6tJ4{IWG~2$CNhS8cY6yt%y1Y0PP1V8%NY#@
z{61QITu0xdpI&w(IhSWI`^2!`^@_v1cj&{5av^_OOHGITQ$-utN$nSRO7o20|5mVw
zQxxGP9bUv@|Nh3yzx0fCts(*<wDiXigT{?i@U@6Hx-hUE9O;f&0PLq01`Eq0KYqzq
z@G2Guzddo9vXjvMAbLZEagaE;d^neP=)5%<P;3gYeSC*+>LuK`Ekxa@qrAjt488L@
zoV;x;KXnT3c%92!(b*kN^H(0Ze~(fVy|w{V=8gR{z8sq(Pcl(j?S89IIIT81=wADT
zNAGT)+IilcN%n{jh%p>lwYJoeWK3M2tgRV^f>T6NB(?Wxbz>hO&ks+y7ghiZE&r0b
z^10@dRy?oZS-`7>(w22&Jc}GABO&mq<r(c$v5|kYbUz9c%05{E-~VooZD*wWi0~(Z
zlR|ZsU}@-0aKaPS@rEUP)YG#Be3NJF7oQQH#!nIyPMGDJ#r&0r(qH?D6!C8n<vX_l
zu;dzqF4AWz40K5U=T`->G)m~#9*MdB;0)hsZ2BBu!RV(m2HOmNJMDv~J{s$9ELf_V
zW?H-(XUE<65!Wc4!M;<NLC?VS=6^+)t}4we`XEtW*C~o#d545Np;nPEDVRwb;k#MA
zH#RQuwRM5vf2}U2-wVlc(R2woanB$p{skyoWL9l}*ZcMFNZUBt;+nv?j&X_jQ1AEy
zYs?xV0GNe+Lfu6{V)UUS1QolIL+wjgn>Dz>H!jQzzljn%he!6eV$+9xxkx19&`)49
z)GP1E-#-K;()lLdMG7fLE)lp)Y_icq-0%<V8ZQT4eb7@NfSI$zl-=izNYs-ts4k$b
zAaA^k?zGQzukP%s_LYq~`gnapcJMme73Z#)`nr=vEz94e8()^Nf?m%5d|Xj>4w-!+
z^-}IW;w!N)<B|F5)Ax@>;&bnaQnk@S3nK~4ls;LpVQTBuE9mvmUJgEm*pWW!6?;|%
zTQ&CAyhXd>x1bT^?y1BuWJ4a^L{C4|eX&neWA4m!)k^CBy`|NT^w#Es%7txuiLYsW
znNDe%m5~>u20O^W`HSZ*>&D60A0s*ZdTM{s5&-@}-&1IpcQkfQ$SJ6@DyGtB(bw}X
z$ydJt@a>fzPQ2_NZie{&n6!KPI><8N4Ii3?5>P=Q=#U2twv`CmSN?hP-S&O=!+Kz2
zVtUhE+MM$<jsBLmRT^6|q2t4fOrf2JWqekDUgYI0Ja-#;{@zf_*USFh_NsVxyWqU8
z*cq$;d7>f>CZdScM^jo};<I13dsVb~!z0R6u>=HW97ncny3?a#Wr{CVU*sVdNNC?I
zQr}HJ9Qd88Ezw^m;zfD!qFA&dCwNz_<scZyiDxYh5cIraIX3RV$D6o1)P)F~{1tp!
zL04*BOY+Y9!gbRV??S+}sq^?V&v$?^FZzti;)kzdRRw0+1^%HX^v}TD=pC;c-;wz}
z-3#WW2J$mP_SJ+?S`v@+>i{w-s+AmUMA1UUtK0cMAW?24PT~ZDYOA-+Z~psEG|eaK
zFhFd7vn3^H6~%2kNxEXS<?&q<X=EXoyLmD|qujU972rhv?*W8=6C@IWfE{vQGuFdE
z{4^_17S_m1)K!!soNL0Q(SM_c#7xplApKV)RQyM7F9%7AG6p?GRPJhUQUE~~{g>A$
zlN={FkKChO%vSlWDIuWUe*6n1tBb!>Pwh<cJt3}c+%ZLwOgE0qox$R)=-8#d`)S4X
zM~YIn_*=qj87<YMG}JEg<9OJxw67nLT|PYeCe%U;jc*)ajpJ90hW!_FC;?2BB{+VU
zEZDo6H#f3OSm=&{{<$%@V_zy*c7s4nx#!Oph|?;jLTp};3{<ir#|)*EX$@j)pU!qb
z=lyRSCIzkXW9?$56m78|zp84Usop?O2fV~037b``i^WH@TR4(&`G>OI#M@H^54iFU
ze_hqbC((M8(%ih|aGn={G~HZ+Qc*_>swPHbqQt#DpIcb-9=*E~JD{y_W#!{yr}a;&
z$$OM<JMN1kv!RGhw7N;5mi-;^msn;FPH(DJweSD7y_FN^_3X<Za6%1IkRZ}NT<C%#
zA>nWHwbRzENT)@`&4fL^Un6Updt>^SIM&cPO^2W-FbNjltiIRlY{5GF!26$5SfeKK
zM|0F^qixAC{`}sv@6p}4C2#m)U<DOCHlsY({~O}}Z10yaLd$uZjvw$}Rp%XjcL@$q
zK1#8&D&p$BioN#@czir>T_wY3&nTJt8t{=uAqVnM)@pS$x%0Dio-A<?_5vFA{@pF-
zk;e_~{d(0T6r)~M`gMorz>ehq_uTydjuHGMQ6P$AuM}X%ONZ*|5(SAwNi)g}e(sU>
zcpjvHvYGPe0$e?cw!S?s-U!r#nDKd`28xCuij@GI=i6wxSm~T@UdS}^e+oP_0Ypqh
zbdR&L<Nxz=|G#`c){b&~DRW&}<$vJd3Dnbb>Oy~>nd#j%zV&IwIOjyf)V@t?%oY5B
zfphVA0&qs=QArI=0p3qE2rxBy0yLOuuqTya#k9U1rdcy!SdoW;U#EHQs@F7>&X@>H
z9en$L7<&(>rnW6=SP(Ucm_&qtKp-GZDFRZI-c?kZv`g<OARrKW6$wZe6p-EpmEL;?
zA%Y-EM~XyxFVeq#?z`Uaz4zVU_mAOlgbXw%=j^@qT64`cXI<Ga#lOC)?^oC@*x)5L
zJ(mACasJ<nau6kw)(S<74MP$$m&l}@kr#qLnE-i*oh2|>e777UV<luILvoszsHp{c
z<nZ;KwxH+UX2Cv^Szb2u)gO;IszFNhI^YFei@Ij7T#62{OsRK*_tCDPQIr24vreK@
zm_RoCG*Hg6gcL9{?NfAaJ^kr7RBSN<$&!FcvYgOM0=Yep_&@>Pb7CHNJwO*MljP&?
zBU2ZoF7-d2<C}mJez@CFtV0w^qAFdh^eZk3fhS;OHGOxk4jpXt7>Hz(BdI${X5nDM
zDDPh4%~IC0udVVb(>jBB>MD?|j?0xl(4DXcLP*P7hHtjRY52_twJj3OfUJvm+H$`G
zP&5wS4Ogyg#=#lZ$4}%4ZxIwukZ~4BCWkr5e960ny`1=0eDXzK0Pzji;8+O@^!eQ%
z{0EYwqn&!%BqBBGpVzGi3lJxm&OU%vU<0Trsm2?gJ0)2pxmttTl^gxF4u)4D2{1lx
zU>*a>vns`LU&^Ml^z^oZ%jk{Hlp~#?R7*_<W0BnMf8DoQ8L<4vCv$cGqxAT@2Kk>a
zQk0-&!xp=R_#gNEKce!#?XFpR;tRyv=QU@p{P((Zpz{S^tIZrAYt+SA9Kr=WQw}wb
z!7ApG-n6b+Ydh39EvEJBZ214a(7hL6Wq97!>G}7g;Y~;bm{;c!0cO&Rw%+?X--9MB
z@UgrvXC##QUJKg1PJGhIfq9g67sudL@xE$4_hhEa6&$P|DGaHA=+8ll7`MhM>qfPK
z;MfvSeJO+1Di4Fm<s@}F2h~j0d#_EK1IQJt_vji(3JxK)5c8fS`b=GrQZJ@FJ%6cz
zMWgrsR(k(*@wtqQ7(>d7FLA$?hNKw@sHAVZ&!*z6!7$x>P9?MJ*GX%D7AY3M3&Wdp
zpO=6tUMZx7z<NJH`1=(R2Lc`;$~0SXe6+u`-o;bio2~$_bjy{NdD;MST^oQ5Y`?el
z><W`O`uXbWvkq&B@iN(rEB8<;CPTi%1OT;@^4EG<4lE>nu^T{Ltz`)8iS8u8c0j!3
zDrg{Pai0Ao$sA#(4I|3FpZEwWuJJSyVBYieAH3@5->wJW=cKn_TD;0;Y`>oRCmLT+
zmEqAIt5VgVeJj#+%Ng!U45LT1Ptw@cZhb8VfwZ$5=+t90ZArq5%|G(V7A5cXXD<TT
zBlUi#4)q6=moAWG-HkH_IluJ*Z&LY-B+qT_eW<4n9PW~AfIiFn(Y_^tE}r_DS|I1M
z4R*V;`+M#;Y!Ph<K>16;o|tHelr6z9zy7G|HcYGM5#tK41dw7o4iVo1LS6mbbw8(k
zb#-6f`|X-bAqMMN<AjR#-<$e@f+5JTi-3Vh3mW97^L*;}Ws(~Bx;RZBwJ2N$;+nuX
z5bb&#c;hn3UD?eW2vS5Y?c}_1Fqb*x6R0Scz-toV36x*X+s2427RV^>I6C(-9RF01
z>5#9xlZXV$QC@}ydwnsW{y1k+KX$kl?ma3WdWi*NSYPUc=KOVs|Fe|%*Y@{R0NcOL
z<UL!^uMkus3ErV%oO{6y)iZvwXJidOA!f-j6Dc_djO=gxeV(R7zP_j;-N0(+c;|O_
zB7hnkayYH0xi|myQ4WH>4?kzoU=EIeWTdQc3jntUU+KF&1RHwEI`tV<k*2<DD*%{-
z10tYshii|8`kCMkNelt+4%>a8b7Kkuan=%$ZWi?Tx!UIr#wQJi`*YN$N1=jw#p8!a
zIvrf%px|S|jnA*Nj7gyacG2hL`M@U^`6sne`Yx%+`K>-ho>K6#5kS+QrpyZKL18lt
zr0F!>xak-3-7<b;_drJ(7VNMg;0A~A7~e?R83dS(6-16x31<LTUj|^Z!F@1c?5H?q
zQN8u1E8{8s@Ue6Ke&QJ5J4~iVt|t1^WQa)sttv>{d!S&Er*`wjy(G~aqEAApd7BEz
z7QYYFA31%O)0xg8oBRPe%h_-gegIFqBTFgy@i9=yVG3b4V+>)}Z+DY{#g;FCu~jbX
z4KO)c3QRA$<ipW6&?IfTQX)_i!h-qK!2+XZYsiGsb9<<r0OX^&o%|Ru`Gsei4uC9&
zcPY5IF~KL_I|@!V9ldZ)e?UO4g0XEfxbLpJ;IJjL4nVNiar0%QwTW7N0~aJj3vEDv
z%{0sJO8J-w4Z3;(^{21PZ6>3!ye7sqx6VKcM$}IW;C)~dF<?(;8~ZV&n&!YK4JxtE
zC^ZU72lPdTQ{Ry%xf0^uxOaMk)Fp(Md*)s;bIm3gz@x`?Ovjpp-Ojd3_P>>R+^2AB
z#id<j5#XPJO5J(@-U)bLP&@qS?K7cLjoRri=sweF$bzRQXYRsYAJEB4z*;4j@Hj~4
zLPA-z+obNQ>RTywE)!jBZIwzu_;Oc9uZ(8rVH6Nk>1Dkei<l<S=Lw!W$?#J$gwpPD
z^sE6S7-{7+Qw1)s9YW45W8`=2MZXP<#5J$GHcz%o9NCw^laQn;c&*Z{uR0^Did>Xi
zcd^YsOM%`u!8YG|A5_H!K$qyoUb_rj?4nUWjn(Qoh5Gd4hLD}7%c|AV4flT(+;79@
zTGI+9_Cv!lg#bP&4-o#wV~N^3v2n@s#f~p<39P}|sMES)7*}wEryN8!5E~oNEMoHE
zqQw$3!|tsZ1->7jk3N`J`t00lEA>cQIRt4&9=-8-V{X-dKhFN&!*x~%I##)C#ePLA
zx6l)Qv5*wvHaH4s(}F;;$SaFh5?%AJ1xE<RU((RmCxa8iCr#(WezK6*k4P59xmcui
z0G9c{eX*^61pmbqa8OLk{Pg=t)_X23qW)lp2g4}qa|>vjGY3|zcV1nBBscY_qUQiK
zd47%D4^6d-yc*5Yq6ulH4&?}cP2TE2fg6}t#2hZl=WiEyEDmB<fKTW;#x2zgCWH)O
zL~vx?pI5=Z9XkZ+8E+@<#ECVQv`^lL3kCrV6e2PnJ0Kj3wq&(hFbYbVK8rYM-w5e+
z`JGv3<nh;lo!iY=A{7~8_XNLR?$9G9L?S+j5EO9VUPPb^L5lff1RQCau9g!(kOCw7
za8SV0uTHAkC>dz%9ixoD0Fx6oSu_Mu2I}8XkQf3_Y$3OK3zYGx(18)ac(L))mgIF=
z9UgG_VWKm61n%f+o+G~-JF&@0QW5BleCPqB$Ch4@8y7$_p2NNoIp5_BGQ7Uu6JZAM
zE1qBRRJ9)8;v~`fGx8AjtA46`A6_VOYj=?*!2%u{Sclr5>78m9-!2>$ux8AY;=~IV
zmjow%m<Et|5m3au(h^sImhkdGjmSpG9Dx(%kHKpP>uI-Js$q>Tg3?!HvjZ!ilQuFn
z-R^dv-{tLYM7pyww79}Zhx9$SzNF`w`L!93zXyKW%q25JC78o`kk<10d+o{^$Q1<7
z2J<yhq)r1j!3#0!>hLyG|8S08uaGYxn2&vsq97AHfxU*<xZcdqK{mQSd?WM%@d#uZ
zg3%yc8Z@*jA!F(_&-r|Vb0$;|=ujG5Z(`<O<^tzs;YSykdy6WpmL0X!MtwW^-OV)*
zt!r(3JJ-CpDkl7T4Pt{)+NP1Q-zJ|2R#x|iAtLHNEsLC*nFH8lXPL<RGaDTs>!%8w
zL*bnu4Mh`9(<3&ct2J+V8K969#>;;H!KTs$Vy3+eCBCox;=Re#GDs#q2F&Y2FmbND
z^Prpg!d;2&hR3@`u&vH)5T1qn$3tR5RyLP_j;_M4VhPvpjz^|XC=w4T<=el`Il~AF
zJ9`zvR6YKgcAUH02VLbzkZ(D}kumb27)>n_?Bxk*b$GG;mtXr)czy%+Ze7lZ-6luE
zR0jT(>stE_)QZ`1DZ7J|6t<aM$nirn>N!7Z!|Z7JMWtrEH6Xlv9No|+FcsiY8~i&z
zV=Aw4I7HsT0<Bla%H;KZX~*a-sKO#h7Y=4Ewo0H%94jU0XII=Zp6G(PxW-37AfR*f
z<R1j$BD|P`T)c8m5Cxlg-7LBW%3K_EmWn^S_4Cc!HOr~RkxBBQ22G<$;~VncoSexz
z4@B;~=hQ}nl?PLU`-(t7;Bi3S#X{n;HjREYx)B$!GiW3~%I+m$da4lLdNz)XI*o4d
zf|;FCR9tXJ5wOs_;SlZ@N9M_gnusR!E=w)$0iljyY}&E1SCZm!8c$^`D|H2`tH2)h
z#FX2(b97oQMxp8x$38gelMkm*^)Rm-TSHnfGJ2~FWPb`?+~=2`*xsD*D`F209_UNO
z%RPw;=1w~X6xlDm28KGXbg5A@hoe8lW5$3OrHc6;+q8eMXb~`WzT%MjhdhI$ik>zV
zJ-q{wwfr~3goekK4ycU&nK|Md;kqI7cr$L*vx3~Sxz%m0HgUmPd?+(>?wjZ1n^+Ay
zT_)4=sBr~k>s@T?Q<Wt&Ge3b7bCg74WdTojRCus4-wwN`v-nblasf7xOr4nr^|p<A
zv8e@UXzXhbo;^HGi#0MPptV0Q(R)`7vN}j95zv>5!QGWenwje6-p$(cn7L{YWcAQ_
z=dnh0Uap&=!*1d0I$trTn2BajyiUm(&&$b5!&yB}7Ew7ipT$-&J!$08FTvId5vA!V
zP~6F*+AzArKR1oXEz|S1iJ)-iXV5N&{ZT(O5(~%)Xz=s$`;u+(hKjDAbNY{p0bXz&
z^k6a`&#evGz2*E6QRTdTUkaTV7N<UlrQ9+MxhMD`mN3<UuXS#7>CwWw$X8$M;e)T%
zX*fN>IE4h&uvHBV6c8=G@KU6A&37n#(x&UtN~zf|mV}*1jJOf*`c;h7+zi|Gb0JB<
zBq4RBIREnR_tOr^>btbRf9F~UDdUt;<WWKM`30tTsgGbwqYJtO^dEPu<|pr(L=HBW
znJ)|Eb_oRDhB9YFWr3D1>$_V_EIYEXm1i2YT6!$Lqn}HAOeCG&z*SpUgIwC#=|*pp
zudgy(R(|=^m5popg$R9C{Zz5h1*H=0I7Gp7w-TSXBZ>%B#!F9l<n;379G}J#tvxJm
z5g25Z{*;Dz?g7&OJ>iu}#cI>12qY1vEcnw~t@AAbfwQB<x`FUqDq>Z}ozs*vG8M0Q
zjvr<xQnl}9ct;?ND^Yot$HK}#aoB8pYl~-~<iz)J^5AT%{hV+X4C55td{{@b*xXA(
z;$!)Pr?6*7&#q-!#zce3sx(3Wc$l9%GcT5rT`J?dK8qrQuVq^;tNF5azG6J{oz%L(
z7syW+<@`)%(dYShWGS0W=G9&yW}B+nN$>_Dk4Gy4q#F*L0?j@Id##M0O6S^`NgG3s
zU0f1MRuCB};plNv@+ciaZbM?~D#yc8-%=R(wy8$Myzjl-O8HjCL~^&sVl;z0#=y(<
zX8)RnQma*tcc}=jGP@q3ih-Ff;@K9ds<A5tDoQ(9uO#&5%G{0dqs^AEssf-2!8yq6
zSy1lpqyEtdtf`#krD3f4)~y*yOPo)V1VOfK>N@u)p}G1H19i5xyRyw8wIyPL5lVws
zS8Nn_!=*+=i~=mHV@LOG0~N;JTCF{>Veq}mWY{wE-+hmNRS<vweF%XTN0WKk&;M1m
zNM@2ChFmT2S}j`$$IK~e(W+=yTA80X)U_G|%2u*n>7rYq6{q3aE&^AxE~mxvncab<
zJ^wMpAcD0P1b9l3TB*s>S)wy{#6(R2eMNTg3+H4aX-^Q$a6Kp2?fb?%W7^3RXfMQf
z7R(!*_<5p-#Ro`6SL)Gov|NgTv#Vve@+#wX#B+>gW7#DZ*^4f<(z5;W(vmfB0!7kK
z7xTiy7eZABrDh|6XV+T%C;aN<s-q0}ZIyGvF|Yi@jIXxT!a|-mg-!9-uZ7y0uDA_U
zPsO!5?0SN;TM7}fsd2~e1ytK)hNq0J5eLP*Gfp!w;)!Ct_vp$anTD#FS2}GKS_F=K
zA%7p3Mj_7JxZ2qB=1?}nR4Q4c@g0!ZYXJ4nJB0mFz+T{kuf%a*J-ln2b7h`4_L<KT
z-e{ASq$K1xbeKWJ^n7OW&MQQO{i((kP?Kp;u*V3Nfmxg_L}}>#0s57oH$~-y0&E-p
zK@iruT4>%Efr#Ws%WNF-zbq~nk3xKEmK5?J*$!e%UZY$oYVA)9q|Hx+(Z$8Ts3?rZ
zZT}u%a!YQ|6>UN>NB@nA001#-gVwgD7ZRh{P>v@A7jA>o;+yU{WA1rgc=Drkg>XSs
zXe{pJySH;d*U3EIx+thph<P9Q9Q@&&S23ORXc3eL#UPRF9sR5^_HZYGYNxuKZ==B9
z=?5_D(+G9fD~Ueuv^xq=G<<DMoIB&<le{5cUsX8DJdTWex<@XGJa{p9nb1^anSS7;
z{w00#sk@!ucHL?d*~e$e1mcDQuE)-)K9vO>y1I|JeRxsD0{Abj3W5%VyBiT1<oRo#
z%8XE;IMW>y;Fc$Icj~cN!&gOB((l-rqQk!E!YqM64R-nNkNA>IcUUn%Hb00am~+XX
zifNo2q_SOwCx-3!0)p@3YZtzRWrjl4@aS}MW&1S;B~5~H(3@nkO?<OaJg!mz9{Omx
z_v)DHPTXEBjzIu^{BU`zOZSuca_fiD5w)?>*2GZ3;SOC{i{-+CE^6hPS5>aF?d<7f
z&cWFt<Ic9Ogh`H%wMKFh7BzQ1v_=y26>O3{w|RHU1~cDr?_Ec|mfaOS=UVlC^Rt|N
zV_&N)VSIHy&mR7qFuso$BYus0k|z>_{evdI;?)m2kJUy3rud^+ozn;v$MF^MoTtId
zc$p{iSUa<Gjp*5GKsB%m;opS2xCKrGgugFBbGtkjbX+kaPmYn8WMXR0)BFAi0y&Nz
zjEro`pkBTvi%rXSj8U040BaTfhX6rq52^*nR#gUt%4Ve{9eNF$7ly^87$DBC(Q@x+
z$QmMRsG@+oEoFHR8Mo<g_R>Ks+r*ssQ<#mWcKCzy<6WYgRreBv3t8gt?l|qu8ND{S
zZyaJ=XA@mO+UbJlZ6Ab`bGc)7;rFa#^su~pTSx^u73ZYFXFD?EV=fZP|D*6?fC?|T
zvzYo{N#hl^iSYw`b734#O@M?v(0Wk8Wg^K2-+bb=`xa@W<k95v5FQIyIS8gow$y_h
z?h)AevCpqUY53bj7fr@?@5^4?4H>2+J|FGI*kosGj-l%sc>853cfNOi<B}e3sX~)u
z!_L=81gL<s$*dCe^P5hK`lq_G&*@VVGwDmul0=!4N+r6No}28Fq$GBuz9rKlFEM8L
zl-#=ijS6N$x*~Yuh1c^F`IHz&V##&L+9;i;FTa{%5d$O8EYsxDbPD5?;aMRpz~#!Y
ze4yMx>Oypdrlf;yMkyy^7Xmd}8cMu*F{Y5e(z*SG`v<ca2bu4>7wunI*usctEZl=N
zBV~qhV~~0vB7^B`W}*ZTl%r`X0RnCxN2AORY{0k%=8_Gos^XtvDVvl?ti?SfCR}Vg
z_I*j;ogCLHkG=uq_!bCFb(3g%a-uO%#SP5tE}>_G5Cyq6-aY5#|58prl9!9lIp@Jt
zB$M)?VjzYT!?+RPPGY^MHNJ*e!_d`F_>ooVj@)Wl8rLzYF<ML=doCe7_agHA-#Fbr
zzSu+)9T#c--vR)JWb~k&l-Gk%rwSwkVMKhZ3bN_ts@?MKcX?(0!<#~()=mDTG?Pzt
zcA1Zj?-(hL#jApIEgA93pX%|eVRCX&<$0+$8awX>-#x$AqplvF1%Ei1;Xd;`qL(`b
z79T6-j+OtZ*hI5Fhn_Badj2MRld@n}*rx`ocZKrBibbwJ5`{$OGNP(<r;W?;t(<yc
zC^*7==-HXtOP2;)O`XC4NUoUvZFSLO57Odf(dC<@);ow&Nd}IM*Bp>Kn<`otmiubY
zoHp4@B_rAx`eKtUP8g46y1S8zLtmQo0{X?@9yj|u1}-si!M-~4-yWWu{kR&&fzNQ~
zfA_4zaqXeqlkh-t8O18Lz5>+F8?E*%YuvS%YnkG${j{mRq%6Y&q#`Zbr2F@^9?mbI
zuk-J8Q<h0dupGP%_IZS<rG44ks&DU2xkZgn^#x!H+Uw_}p+d)i`G;D3t5YrIrP#Go
zw2TMXmpN>yzHeij0Y{DEDL8DNL%^Jp^ht`F<@ppVULmDpnLxFi0Xg-$GE=3gjM_nW
zRkOOqoBF~EnVE^r5|hn8KFhz~+$1Ghh{~SD{GCwL-f0k%HbY-3n&cl1HVBkOOj15O
z%X#2t*@qXCH$&%@8KK7#Z!u+J!Dg%(`ExVUc^|CBxm$FJ_M6rH!DXwd9{!3!>j{Qf
zV8@!O>85qIit*aF+FOD*>Dxt+#7y#bO5Sles>gdFkQhk|5BgHYoYPW*qf$(yGn=%e
zbyG91dEgV;pan!*=@pV1fHtIG3FL=^NN9O=r*G)NxmhE$j3hu84)f^G)jm<W5bd$z
z{w??bY?49hV9{vX<{zNq&k+uzv^C^}5LLdNE3vNZ>r-=_q(ru0gK_`pBa+39k;YV(
z(MPJd$!8}o2VHe>VIRcH&{?4Mh4z0)9LM!`+ojpMR7~K-tg!oUt}O6^y=Zg<rtKJI
z?7rN(6-56SxHR^1*Rqh9(it5Mf&aJ4W+h4Y6W}ILI%uhznCF^^l<CQd^=*8%@IFgK
z(*<~!bq}KZdO#<NZ!yZbQ=TKJf<^~vTFurMa<n|zOWSQ5y5491`&Pj`)G82Dn_>T}
zvhcib2!^sgh&o!SVGP#8)XN%z*@-4mNhjl3Wbe9Mzd~{I0Sn0iDE}i0wZ_BP?HT(|
z@<^k_h4z6SU=-*i=#bkN`5b|=j=%Y-A+NehxN(x$P<#AkCMLMS;Hn^;j1En3Q-i^}
zpjo#Okm6q|+O#X5oC{n({{S=Ldj=iLRq{!ie8Fe%32=~s6W+U!;Dc?}SY3cZx9ss&
z!EhYzF{cebiHRa-p)4WwZSzT^Ik{u8!1g9k2i6a<yi8E^dlAX3V2<{?xhJO28oxy=
zo9i;FYaVA-1R7phMmMNes6Qj_yvn?;p9oMZvuu9RB~Re4NVh62sl_#uR3`|8B&nSd
zKVFEEIF$iWxd36rh*b;5U>Dn>y1TzIB1TDU?67H6b4lZ+8<YNY+y~b6!?#+m%=K-M
zCR1bm&FKLO;|H}aN3w&9M$_p@(DNKW03~dEEzn*6K%E}X0%u5o8YbGsds+MLokgE-
zyvya~e_5|LvRs0m+np<xc;>UoQ`5({{ECHeNU5QPAvnQ}^su*=v@Y1YSL~nbWtCu!
z6ZdpmsuBD<q5H=dd*Prvci674MD<r2x)V0>3>I$UV#*O<-mKOOltR!=G_o+W@h`Bg
z;D`vhN3wE(eCwn=!oaA3NBX*?OD~w&yyktgsfFpLVJZ!;k@l>i^aH3|{8Qo|ragLV
zw*64o0bm&xE}~s*HH(QrpmYd$)kZ@|^R8N<7-HV{gazn^VLBU`hX6FGs(D}5U_jU5
zo_~#=X`~#VC?l0x_7*!~;IpTw=H${DOfLX^)HdD(%)%O}l12(nik>r43&bO;zEcP`
zuR#pfeIPM#VMNXsRrW#P`%|Bs*mlsuSTPKuj!<rO=~pCGxiUTpjr&>y8_csk{xHLi
z@crCdfdWx_)4pLhGoz(XPP=j$@JOFj({CEh^4Ao!zLf6~$fe5E!spbG1X|1J!yiDf
zU%nsnO=6I>{=us>yED9uSlOQgAbyLr`7kNE=s^ZBTp@N{E|i>ZY^IBnKZ_35yKol8
z54y(>?)qIIVTbKNotW#MbeyZ*uxW4_a{*Fk*Zfar!_;F4yPNb3wa@0n#L99$yLKU5
z<MKLTcMwqcz%id!aL@Y6H?(dr_3P<_-n<acL5rKO#0%HP?+=PinRNQN2(nlsSM2)J
zt!utlU72qbPMnVyrx}51CvJbXsDo!d0fgYE%1)UtHO}Un?TWXG4TARD@$RBp^4^1^
zvSC3_aWQ63!ACKC@e^7g@#{7!Fr|}m`B32qbaZhAk*>Bthc-++43X?LtRpSy!`RW}
zMbpf7KS4|Aaqo}0AJGmifyg1qae`Za>ZaLknwyEuH|hB~WGKtV9*VJ_SyB7IitJ{v
zYal8;bgaR%R+fL3-E}>E_<T1(P~mz416yBgr)Z}L<6(;MIAvF)>?mHxzc%~+uDDK(
z{{G)*#y`IBR50X5&jic=-vah&tdBcM3Dj}&Ao`}eX@o}idH8<ST{NznwXH+ga9_N>
zvue)$h7ZN!l=D7#aS;mfm0BGhg~PhDX3NMf-3n9gXkL_D9M9%3i0kIu#@7rELPcHj
za7iH+m0BNkeJFnXIRz&0=z;F^6x~L_%ShA+=uo|@{K4apDjV~}Zv2!ZVLXnB!JsB=
z9W+pj34Ux2;f9ebz<h{V$Lytg)u;qwelmE4kTiy`Ma1FC1W;zN!F<+$x*;#7n}V3Z
zuHXns6H^|AOiN$WW0k7+(M+F!1S*4rN^qrc!>Cof3hPgkGs|P0S@+`2mrq@tZ*$`Q
zbarGiZkdJM`opR7MlpjFOY_p*e?+#0GuWzS-6f@UnpFNC)$^3B@XU_9*|KYgt|K)z
z$VQFIx<{;84vl`opXX#0T<G~#+u}>o=j8zRL4G*B6#(pwLIu8FLw%CNavSoa4CCBQ
zf_v1I#00{iT#P`R0r15Wa{EmoUejrJ9F!iG0nz_c6g!rvvI5X|xHQlgE)4JTIIe&W
z0Ui6uhtaKaDrI2hf61L`VsJ=R>Hj7tZ=}?^Jz?DT6f$*SXV#s2yJ~{B^DU@!`Bb2X
zdi**VU}ev~B=zdhMhCgMcieGEjj4pab@b?zU^ow&hI)h4TPYuDtaMFjw^EWT->_Kw
z><Qcn!3}aOH44kPSG?g`U;C6xATF70{~l=zUmpQ_9*dx434Qm_pe?I~?^qNyCY!k4
zo;#jpThT7zpcJ$X1W4MndiZCPxliz+wq(b2N&07c6Td#))`4`QP5yB6ai71wbAf6*
zA#82voo!Wv=wgC+#BlVs;Zc(yjk6_cr!%I9uk15*q(tu<J*YW2+s-W>+VwV}c~&BX
z=Ao`hGOwQn?~bbX9{th`+GwNoy><LHbU;#+Ej2B<pr^}&Ev!t0w>x7SE2f(a9U3i{
zb`uNUnN86q;vv{4I$dq#zSCR=It0nxM`(|Uq(Ag?_S-y)PV&9C-rXYnK>bOt#pXTh
zh@JCo8$gG%MK^`k#E;C@2Iy*r?8fYfwuvla!L-DiP|U~9DW=8P{eri^*@L~U*L+|h
z;TQkmSA+WREDm}h`3(U`G`7ME9Di-cPj5*Oy)L(!5Ss<U3ur@W(MDd`GcZ|E5}Uv`
zS|wbHoDZ1K+bQw>+1~{Vhq6g*SwYrE>on#fz~L@&V8l09Ut{?0v?J)lKm|6cs7_}r
z^;+&R$uo{hi)?ny;T~^QFMmsP)iwVcusd1>@*Pw94O8NsjSkNrrt~-0)((Mw^b+7o
zoRcCkrrPm9ZAoT)$LxhXb7`?L%NcWv1{646S)%2SpAbI`$rgE*jOG2$2HxD2wTy02
zhV>xyl@(s#UX_Bhsd~A0+4Cb<|B1uAvdCp*KI;p1LL|2=iGRZfG0Gu87(l0`1KfS6
z1I|k}Bi-=aGVuF&MOw(*5msVOvmPk4T9to>gK?$!LEHz>kEI;}&2#hUr}UjcV=oX7
zF7tAhh5Nu&7^ZIR$fG>Zx{pJHBR{V{?8`t{+S%55Gty;x1?Js?UD-(E_P5`3Ct=_a
zFrGutOWzG-jM$c84}qO3yg;nyZxZ+lp+jhu;`lcHa};#UK8kS$#J1ET%Qgcaz<2U!
z(u}Nw*23(G3n{SyE)<a_>HdR(g0iEx;~B-%k6<GOnDx6habP1<10XBSTc}fn-r4p&
zVO-<$<9v_4fI%vg&Dq{`s~$;V=W*8`&u06@urJSYg2qr9e-imy4ORPd9p9dM@;^3Q
zfT=w+SwtXt0b}?XZ!vosV66{a&8tZJ>b<^8RUW*d_QipECX0v9T$PQ$LsAv8g%7f3
zaQz-x%Sw__w`|1g-v={i0BOTSTxV14)1PpRn4nVqhwS@WgpE1J^9!c<47$W&Y_1~P
zmf><0ps=Nhsd||u4`$$^&yTQ+eCy`z^aM?vSoa&DBQy-u?bJJQ@pD#qlfYG1g;7wp
z$&e(Ck5sv{dPaK%3CA#)M;|t1REs+5|7tD$69-5(#SJ2p^gOun3+POykX-ehcqY-p
z1$+9!S&}Lid5jT##A2U}^CUx_HYS!OhWY)JX;5Z;H|a)#;jky4!g)y3NqKRoWL>_b
z*uI|ZvLk(>RSfp>%^B}#-=5_@$v0<~q{ZHd%-r5uMZ9S<py@KM9nMkU4ae>PO(n~g
zq`|h+2+~hwoO^15-H~r-o$dr-)5mvP4tL9%%?lEvpSxyEd%Q3(zO7xrkfh)`5!mr;
zI%CRV%8>T>lUJatY8ubwNZBLZBUe|}!AJUZzG55g0(5LV^NmiI=VMe0y}$2k@fs|D
z7)U-e=R^i7pC4u`MVj4SinnE6pa1CZLZQMop3AC~q}%?s`iE+`Da9ZLD>+27peo`N
zb7m=6>&$+dXOv!HqHUM4dj@Y=^}$A2?^x+lQv345;~LMbngUv3%Jyl;Isp5>KGTz-
z$8M1>TYtyo{Pf9zzAQUyL37sl$I-S&E<*>=D+ancSaQ!(BDZraDdt>`qMk%*-U%E*
z2u{(I;+Y>;<>e~v-4MJ_+Zbh!17ZWAd#>4>!DiVEjY4bb_Z})8u`4Difefe(ud!70
zMSS?am|@(WVd3N5Lq7MeWLsyE9-kd9Ybp`LniSmxrFLXEzir9uRm&hv120k4Otxl?
z<otzy%#i<nvlFU=y5$u93rKZ~Yq-spcrjIf?fx^(D=ZLs@xJgRNi$*XToo{SB%jI6
zeJm4uVmuvbH|=|o$hnxo>73EI+}FR0tLJWkb$7nQTqX7IPlsL@T2i>oWyUaFTmc4!
z=HMKUj5W0s+M&}CscAwkABn<0i`q}Mz`7LKt!s)of*K4j`>;uh5|4ux9d7`<?dBW&
zKPF9oy?N~c_=3U@bTqd917Y(ek_FHWcVNUDSo$r0ZBLNM+eFGy6|%h2YC13e(@<o>
z=j3VSAoYvL*&9UGZV)x@*%h{nSB>nEbyHF{qZRfL@g>RPorr7-0}LTfUF``*+pLrz
zK0tF4)b`K;b-)s6;KkJfr_HyC0|CFgHGz`HNO0sF4=#N8k1unr9Rav80f~UwA8_q?
zG>NZu4<JI90piXO;Iek)y#SIW#I`kS@-Te^;NkY>Bvu@dm|&TE5L=7Nz+<goM@mPE
z(j$z?=kUUj^Ki+-Qn&2Fczbt<8dYf0J`9W>yN6x>7qY<sNLTyewMqZU`f@nu3p~z8
zAi3h3M22<iP?NzL)bd>nePyt7cZdgfb!=<ZukJ&1nldra5*g;T32ZRLtOBU@nPykf
zdN-(epi!n*iG94kuhzXF-lBx$0P}w>)BjjV{}}B4^=8TyDCEhh?*7L)|Br|NKfWm`
z0c8XXDGQPR@p%4;{PxekpiBvF;ajF0kN>|{{9j)NEHc1|g@4+*_1hcuZxq=7V(+|p
z4oYJ6$Oq#8e0~4(n|y*p%)K+SZW+H_$NzZu{(SR+9B8lQ-nzo@>kIf2-JSp}kYdg*
zYhav#j+j|^D=h~2EP^I&=X3yUVGV-9^4s?L8@0D2%f^E3UY5cF8+@2jZdgkS;LbAt
zwj+E^6592WwDPTg?RsAg(hvrT@g$JYS9pByh=e>iYrp9H2wZrjht+lmQ(a47$SjF@
ze&o#y2rhxBy4Zddg&_=%u5Xr)o<WwxbmZxq!Y6W`z|46Gl!0Eb=o_NgMJ=c*e^D(R
z5ALfR@%wJ0`rVJV(9B)w2$U|3oDpC39rjWZ0!0j!1DgbUM>a$g2gJH{T)?Q{%6;HY
z8}czpOk}<cc;gO9ouR+}eSYZCv*%@Ype!8$w5H+trK<bCJJ0w0Au6(;HXQjE*fGw0
z&8UfFAACcbaEYS|&siCUB9Y~a^c}Py%)vZ0sdW*|M_VCY)m1SuQ_I<H{!Q2AmtQPc
zDjOMIoov>rQ$TH#6AfX%*>y|I0CHq8<OSi~L5Q<JkATQ)1B&6&X`wO4s4+lBEcFxw
z)5*Lds|_b*qCJ4iM?eJuK)*#$qYcmtYpbkE!Shm9O8IhQ5kOsm46^vw0$o{v-n&@{
zGPFGLV)P>5SMrl=La}Ug3mQ6r^5{cn-&0h;Vd&|KYD+kbXc9I6*v2Ce8Lv?@^7)+v
zYa*lXki7*IZ?EOByPN%XRlJ1u@P&|0*}v}QUU&nb?OH>mC)iy;e5|LIrp!D>YJoty
z1bn4SfoblM_n99JT%wmT;!`VtPGA+(P~0PK#Q^xfe|7=E2yRiPyQ9iCWQ5HQ{laDj
z$cEd%;Cud{!SS9ZbOP9b{1VVyA;Gz;?@Lsx@Lm969-*;FK<<Y~@YMZGdtcCW77NG^
zgJ53$QD8LUE98L_aHe=`z@Q$R58+%@7c8C5Lr5ykAq6itv0hCDuSQw?IFP5B&4&gA
zn8^I?*b?mgYQcP;VE;3p2h_c^cU{l?1RcvqfU9U<&3@+4^8*z`^WN{z7B>TiuKCu0
z)pKccOZC)m!chPfM77m$&jli#&}MVE<_p<|2b5m`KzHZ_7_m$*^f{h{dsiH_(3S0!
z*B^4VoO5B33GBb~n%6|$u6f}juv7ii$yJpFC+#TPM{IjDJe%DEZq1SOAVbCi9{AUq
z6WoO@7o3cw4P0j^-NA;R;`T3@ZVM%o>H}0}iYBT1U@3@|0k!U1u%8BHdk9~161SVU
zL-4T9*_I`qB?>m%`$beHP-6^pk0F*Nt-zyxIxdpCoU09??5ffYva-p)iFzl4!2UUz
zBk+LgugEm(1Kyx-`20Q-ggnmHw5JZjzK~la07&_(fe`D0^xo7BG9IHqn{S89&I9b8
zTW<}>^I!AStongjJKfUh3&dZQEoDfquz^FX^cH}P8I(wgj@zYmes>iRWO*|G-UFt;
zuEH`{QUw-k7I*LU^&G;>=Xa6Wh%5Oj!qpJwH@S4mw+ryR+4gtpk8L2!L2K~IB><)y
z3xG3>+YCX3_p&BV{a;!D&~%LT-sW7u>Zuh6m&T^_QBX}<gIntyBieoAM*>WuJxE<s
zPeA(9nD_A^wu`HLuo*64?IKRgYb@)(GXhEto0*rI%P$meN(zra40HKIwbvj}E`o{B
za>$WlPI->4$rns6wpBtp4dcKP_ma^Gj+5h5YkQ?oXd^m9XGq3XqWlm&$7seXItw7t
z3rRm=_kIGWou)DnkaQm9WI}DO#+U`tN`>0D_aQsag03Hmu)bgPTX2h|-9ZRqO5m9B
zKQ(Uq9HQU)0NhT0><QQSK=%;{I7YC7f_$jZUo-=8elpb(Okg0-z9?hsf27@nUY!VV
zmKhE>usqXL3WRI(<n_%+^B+gCDV@)MG04bS5Ct~wz&4<o1|&|}TsTi_T_B}R<<x#N
zM}e&@k%t6bRIng%FbPNUxr4!&GN7_`bI@qOrFaDJ>etI1haP7qJAW{ZKr?b)K8*7!
z0zH#2kE8-`buWPc@aK_(O(=|7E0}#Rz(b-7raVkWje%2b>(^B(Xr5xE8kjBpp%2>X
z55se9s<X`+x?|JFL42)x^d7Xh%YeVe*RbAWSZjjP!=w%dlj^7&A-e>C^Sn2UW4KD=
zU}-yyO#K1W7L37ibzK_1F9gmz(rQ4WWu|ROgZQ%emvqxZc;*nJ5KVup_O<|W+OM01
z*P3Zrmev91dIaKK??8jQWkAUBWq0fDInp!Gu57Cax_1Q331QWjNd#ArA(<iWAq=6a
z*{&rbIF+OtM8uI3WIMg$wNpKdv#s6Iz(2(Q*qmG*Ctu&P8ele5+E+M=J82}sLt?sY
zC?3FOf}W<tHoDI$h_6TXO)<SV0#QWEeZ_YdG{upNHxzXE6^j3)Y7ex^j|YpGeb99S
zC$xeKwo}Pq@-TMi#Gas$_eR>1@v}~UAJDDRa)IFuszF`SDx4&dW%Db;_@``W)&}aN
z3mVPv{A&Ejrm&O9wv<DZeXxP(S+oE;Zw;g_Ls~UtIa^zrSpBY$7{W5N5N4^{fSu~V
zxe2_4S4ts((BHb%eMlk04sw~+@>8yXwT}auKMoOD!vPOY1KYzqG4MPRCQT!&4pReJ
zyN(X9^ZYY;;B2|@{Jec182_>iWC8B!b%=I4r+$^p;2kjV;^PShM1)9Sm;E(4Io06)
zh>v}1?F<Ya?Cp@&=m<2SAPu`s({WPe^Z~+BxxUCh<$JKfK^n3%LJm|R2BKx!qBv%0
zIoZq6!J-JKhP-+ISpR2R(W_e9z}R@`zUW$u<D1PWLCMv1@X?p_iI8s5L6X`T^#Fx`
zeOBrY!>8t_b{Y2yXW(9|PmPiX`D|p|4MzsA%v6G#RQ3HR-}~_D@X1NyH09y%z*09H
zSO*^(^hfQJZdLV&?>lNAWPh||@IN_P4x}{vLzyG=9#_stx5!o&OhXF8Eo6FUSHGQ1
z{F+WNa^FmvdJ|xRpY^?V&15=m8#JGv(<u?6o2w2q?^LZmtl`XWbPM4<Wj_vPi%b$y
z(1RC@?^xTsRM#M#)X4n?TK1@SJfOkg$PTDN5z5UaWXA8W>obbV7u?p@)N@iFv>N+(
zJT{jhmRq!h?UJOY+0PNe#1x9jV!?Vc*i^_3G$c!<BDQHW+x`YDqMuPiEdy8j7l3CE
z*$b-D@@iZ-qO=Xey9ZX?E`fVKJ33mj^%n7Obb4m#@FOe!@!7(gWzQtnr-FIPfT$<S
zd}zlb2!Fgmh2$?Q0}a;^=emu=fHW6E3SjhKZ@ablV&7;7=s(CE!MF+yx?i$rHH(oV
zXxL)kx`?m9ZeP1gdo4dE^UAuV^Z9hq+pvy=YoPXxm56Ppt}fIj2zLfvI|7ojRvl?t
z<w<9WMwS>_e{An~0jhG@4|+f6o3g7l4A)T26x9<7D1)C(sH%%QAgv8!@MmOi?Vj7H
zQR}sh69Fbe-xRzKwES(X6AF1&^h(dZ+GJQf1}RUdJ*7R@Oj{WTnMCwtJGurg_=}4U
zyxK(`PbyrtlTtn&Tk$Ga5xmL$y9~TmrlJ0>`rjotH8IJknm&Ok^N)MFvr)Umo&;M&
z99aS%ipR80IlfXd4|{KT_0%vP<;ssbVb1N+;URV1!CwBrzFW>V&fTz6*Z<`hfD-4*
zs+3TarggEeJ(nGlyGA$E321*l1G|qNsX`7&1wvCT9_3iv<3Zq$BfIcur^@nLo|DQQ
z0@>R*vJyVjQ`2Rsf>&@e`!y*7+1?O%dao-dt~&Jam5Q#@)-@zl4q})kjyxJ~QdbOe
zcx%s<@t9k1`#fLZc<P?jIcjwmaH!vqIjzf$S%akR<M|ukBl<*jV_)q~xe&(fnRtF?
zu1CcLpZCk;!CFA{+33h0+wxPDK_`>BU4pT*%i7U#3wpvRPTgqj7^Tk7a(+%WLwa>f
z3CRLh8ib^;d?;H}Fz@Hq#-9vE2Jp4_0>(z22K0<5RvmrKV-zY3*&NnP244nKG*LZN
z7*tk5O_(lYt~x|2643Qkz(sR1t`4*YU;+*p3y5N^z>s?GyG;CN3Eh;7Wdke}5}zK5
zbre&K5WG<0rrbweqcRpHHpaIZGVk>WPlw=Jwc^}Q%&hnIKbGM6XdQtz#W$VT&3N#Z
zsH|CQ!EX^S$7mEkml0f01jgw&m%fU71D<@F8282QQJqJG@e|YHi)Tj;zLwWfntU9n
z+b$gzYYn|WoPK5Garrs!*CBN?#IGkN^a)Mp_53y8ntOTKskjiPMk)e!8Ivcz%+a-Z
zylfLB0HmYM6#q6P9&fcwBRJfCs&Z(1#FpWZ#US8**;tejzK8oPIo04flkp*oWO@^F
z6SViHHj1@PYZ`6NPuMk1OTbI5KH;${-H&SrJhlYmFeiAHcDie%U~8CO%F9Yd0fIr>
z&2jQkGF?K_Civ}`P*uu5kFYF*gDI4XJc_sN=Xs%@jlL4&INeo-6|Nf1XuvM{^3fZQ
z*cCL^tjQibQo(1}?~GBH4U#*+rYRkG4zGR8tf8%P;j!L$el>4aRd>(bO~@fUr*Rcz
zuA#dfr*eRs-gYs*wUKwbaj?}bF4%ZsJ;yJTY@C1mVFBElvG$t^3RNfnEvv_5C-emw
zW%@yg9!DkH;w=AaCB)dsZ<PO=DD+>Tz@I<vy*^PNp!x4RZCzRtVEgvmElO;Oj;dc1
zKjE86+KBUUA|FQ;6;YR#8(AE}Jqr!Hq;2hq5>6TUQ!L-$B5`m{Wr)KNRzF$M1lI@V
zAZTB)P`yTeW;`$^DvgDcRd9Y~axpC|Wk1c)Rh%8X{m>V+=7S6AaxoXL)GS9NRkpVr
zOb!c<BK<MvO(sQ>JIF8rIF1mvsx!F<bNb)pYy26DzqF)@Elpw=b!s!$7Z7WexTJGN
zMABu(cLaScv+lxXF?KRO&nHguqRV2HmcOvKa4oxXrQbugzGT@$aU%SvT1f-R|Kw5k
zY;>xE);YM4vQWmDn~ZnT$6HAZ;ChLb3LR59gZa71G3MeO=iO#Isk2oJY4&FaLQ=0`
z;okOV?IYb&@zC*|sc4ma8y2|wqAq}bhJuMlHrLhH`N%_ksW-?*Q{^;u<v@!y2P!S*
zqeOq)4;S`Pyjb}RvcF6uvH20rq4QH-IwKM9O;=gJ9M9O8*yPc8-L<l?ohTP_B$|nt
z#ILpGXEWV%M*38Jab7fEIMc8KBG9EGmR-nCX4&FCx8fj4n&G_OT|3Fk-MV2er7F2h
z;l|@)am^PLWv4*Z=@nxiUo-aE4(KaYAnba(-)AI+RI6`>YjpOb1`IXP+(Q0SHSUz(
zLS4R1CI)#eCh<SDNq<tQ*-FiZlyZk{p0l?Z*N6*tk)L<?uEwIc&O_m~?j@;B4d{aM
z9JOq$7)C)*bt%ezb5P39bQS7UZ2ujsmhuR~&h9<f{cD<2QUYn%OWz0^chf2`^W6d{
zD*G07G!Ih~ndp|Qa5;BVDd3GWAmEITjC{qC>md3u`U<~j<@az@IkK!ka<pTbdWQAb
zJV~pA0!d8Phc+(%8OXgard{%bP|_e`))~#6Rv5iISZE>>{bKtw+m;O7WU6L$AoXVf
z&g*RXU@nLC%1km$9}(*o*m;ikQL%qEBYb*j=7Om6WFis+e)b`2yv3{(waEj=#9~Oe
z6r<sj=VsWmu7DY^sKcaTL9_PH)Y9Z|-#^%DZtYI5y;V7Wo12_%K~7L=c>zd8PaY6S
zh2ZEuiqg4UyC>Ln+lA>0$410Uf1eULm6X?mE-u5X=C8IllGRH=QJ|+zvzjG$+$^J|
zEp17v*;3Q`pp{MoPk~_^f4HXl^-T|ICC6AIXh2l-i5f4`37R~mPdZ-jk!xuYhWcUw
z4u}EfYoMddWg^O6TuVy9+CsO0QI)CDuQ`qUlIT1><fxm+6z(xL_*el}0jygbch{uF
z4G1<~6Fv$bG*p$^#&Y)*uFlLaS-n6Ydhnqauzu8N+Lo*O>Cu3@-Gh`ud2gy<4QvZd
zy$J36fzljjJcD^@;gzAPZc5)3E`R}I(?J#oVu?-mvzKXSQm21YqcVNcAt5lC9H*)M
z9g)E7(rQFB!DtUJHZz2e1A<%kK8ktAxvN?IQ4EcE3J#f4I4EHWsjl(*3rYTXF}<N6
zwj|5t+i$S%ag2f}#z83Ia^O{1av5ravTI1;3$2Y86aAt$MYcxJH^Vm<$MIYqNbFj%
z3&lMVfR9;Hu-(Te&lfF`brXy)Iq(FG6-^L;XS%3xMK;Qz$6s%K+6Nave!6Z#$9Y_4
zdI)s-KAsXMB-PA!&p3PZ4vp?7R3bj<UXPXd@^*y#F)y6L2F+BM`NM@;ZUYfT2p%J(
zzSgPA?hG#AfO|h&KJ5IH{HwihTvxfnxDnA^y5OPmO*h31cx}W~DwGYc6_M{mSS<H`
z6qPj<&37R`4ECBa$`8(dw7=YXb#V&#X}qO3Ne5F>X0Jx(a%?t+y#d_$I@mpAud2Lw
zR$*BTubX1UZB4ch(Xtfx?bQ!OHJ?dUe~6uuHf>&ptQR+~O+m67alv6}`4u}?Hk9+)
z*hWVkUh~$?M7f9Y?F5+p2|6Bb4fG^~y0PL5=xteZtLD%v(*9bb)X91wX58%I;;$Hv
z=bX`QBchIWn<gk<Hj*`MuwG+QSFY8&Q18bYRK?2#?_i6Ly9|!{bd^m3fK&*H%iSw}
z(p%^{Xz8%x-!^!EeRYn|lo4yPVPd2$59w4^W~(}2H1G<$Rs~ka_C9)`B|sp?Q(tu{
zUq$2*{46&<74Vu@*hzTnjA1EV`^{!W{)~1g=k{1l{MGuEB+8G?t@08&VB%~eyUKdU
zbo70$w<CwF1fAi*6RBEPVLg*Pt7uE-y>RzQ0l2Nc8%>!rz+n~i@zt&3dRVoWoO&&f
zme+bl9UN+c*Owcrfwnr1K?t5hzb$#pt~e(axx<WNYp-Tm@#HSD&EYn4osaz;j{V<Y
z^{fSP;0?qOAXfdcy;$6HiEI<5E)Fya1i9kejf)_96qSVbs_iSp1wgk(Ot=OP4QB=h
zBFZb<V}}dnY0*bqBszf>&&ifKuQQ*p&kqQFoKJ6&>I!K5B>+&Y#=f&oXA9mQO{Q`g
z0ldy}yA>y(gBc#@o_h7_eL5NCpj$H14FQ2uEdX9JS?2A~D%tsz`jLnG#)6TTOobnL
zMS0natMhQ1ipc<@c+vdWn@B~cKf;Z*`@EEF;`b4~P0x;S@7-s62D6W=*e-7%x(Mie
zX`i^HDDk)RKRD_e4}xN$&m*5i5HoEfk1%<T;mKG?yc2vg8xXu@-v!FO(7O7iS`B0P
zuOm|ZbZo_$O2hf27Hhz<xD~tjb2_{n8Ym%5&dGRah^&1Jy+SB3PQvp9hPysC$hxc#
z_%7j|_ZSlQBs>SjPPCT>u2rJ!4V|ZQ0X{6?MPm2)p3t)sRX?v;e-%rh!R`Sepbz>$
zE7Yaj@SP-Qi1G_TE7qfY5C|)Wk;GMJj36lUWlvVuHbAdTph3exCS;a?Vvze)3GO#V
z-q)|H86->}o>O@Tv9*QADHKAejbJ;YSDofcp|VQN1I}672-j%slaa(3zN)l)N-kKS
zk}e<T{`#D~=!*HE&`Mv(!Wb#hq!fCPTH>zJGUri4ri!Pd?(d*@w_pI`)yiq?{_3nV
zP)Oo^2RPvNmx8y@cX|nl&iWh5!G+30Z-n8X%jlA8EcLDVjbPQgWntr(8cPIR9+Db>
zAa$n!fssasJ1UY5(){B=2tmf{meB}4u3|hNVG`clte3tZJRT+Ip0^?k3uJ4CSnr|<
zrh!*K<a0I#$vq(@>Y6=Rp&oZb#1ygD12E@sMK)LmUDrpYZHnCQ)T}dh>6zy?Qgboi
z_!OYWy(G)|<)K*BMMC`<w?`wp9*FNvXYyaUoH1oELZn^YBLFb^n_Pv4x<pTmNx@^b
zHJUU75?QpoU@WFi`j#Kp<PwF-IVPzZ)0x^)m6iL?e11ip30^QIy()-9eFBz`VKS5N
zt$-jZJUkfozDx9zIxnuZ`N{ESIbt;<>V8q0MX&Cf@dlZ}Pig#4;5`bIC{Z|2dM3$*
zP^IAV{-tRod0{#(e|ZXdO)pSWz3p@x><FQRagk#TVw3M2y2Yt(hUUn}`un&<%VmIJ
zt4V4snsty0EF$xm?b*b%&LgQ*(apyPnZ)bo)jh)R{K!4}et51YBgFQj>TBmU(O_I>
z5ttP&IQuH_pd6fCt9y5cR;fem!jGD@LTjhrzh0{b;|wMu5*euycjNe4sTJd5egNdA
z=+VM&W635t(Q(tiotQA-X7wYHA9YXeo=2nj#;G3sDX};nOLNGNz4;OZvvr8l{Xt-A
zz@y@5^#>OHao%0Jj<J<&_2w~7?qTcI5v&__IP(0=5cIg<hEOK{)H(8eMYd5HMcsvQ
z_rBDuWC$XjcH4msuHHobVOnQ5I#a{rV}Tau9c~mGbG*?*gwi)d2i!=Sly2y=a$87o
zul+C@WzscN&CcIBm+Grim)%MBkl!Gq2b5p+WyMZw(gY(p{>2*Y^@#f3EHJ*Y$+tHE
zM&wn1yhR4Q;u<R`2;aU@;3^)KSF#xq^EY3%=Zuj#%yQqycZ1?`V#4_AUR0f?x&)8;
z;#9mf=t6_WRO`W>@wZxIxo-{n1u8py>hAA?Z}Y*6lPZP2;>AkLmOtcTx$7rWFEhj8
z<OveG<x7(S{3yKvdPR|6tLqJgNi{D0qb^C3tMJmBES%Fv-P!2MSx)upE(^dM=T$=`
zn1STdrD9;m8;p<p->Xn4#n-M1c%VIWzb8~i<S$>^6LK)oKK^I`vqH(5q34AN83#A^
zrj2m)uPRSm^R6iSNJlVkz&EFvqggAkqV+0r=-q=ptP-K1<%OG|VhzuGAe<40^9M)2
zb+O{KE=b0%?4N;4?0%2pN40%D(k&UcaeZ6mv%^2`9(xrKpxNjJJ3KzKS88$RJ={!s
zfHk1x)SFFm)tX*}vZBGS_<>Cfiue^Jb6Dn%?2i7WeQx70@#%4LepDiMzpnS1$uFI!
ze}Xm<zGq>pT^4`K%=%&^Q`iZHg-3bS%g>)3{Yeh5dS}Fi1Zs<OMp#;4=}h$jX?hW5
zi{#`41wowqlg=aHp(gmTS%%7dxw2sM^<@&-37>+~Mtb6)m+9SA2U<q~c58q;xXfhO
zVGv*DUS!!L_RzPS9>cg(y_1TN^ZFB)DaPO)eW>*_12DWL0gW?Gbxk27dU%zp1&P~I
zkbXCX#`5iyfux};OHT6VeCqXL<7!O5bjrY8-RF7`(v5j%Z>x~m!?B%T6nqKv&EF^t
zZGgZ8X7O$Zer8un_b>1l<e{*C@TF&?2iRp}qD?~9R}*Rco;OB7;vf_LgK&MA;(Cx*
zSV|}_`J}D(`~#Sly1S$Ci26&++M04nocrgqYdn6P>KfWCkW5Jx{?y#w=FlA7tM%K-
zhg>A4oI^lBmw$&Z{`rG_$M7P|9eKy)afDU-kRhjxe<%DzL9cv}S9e1AM1HL31G|vQ
zkFw1rp|}U+yoPID+iyG4IlqBG)V~3g7&yAP!zzW!5gY46cjiA-=r|oq4tBdVC%eBt
zCuM%sEN|*k8(-2ws$H?X#Jh>aGxl$Au|1;r$Ib0!%p~vpW}wM{N`-}U`(P?`a;({^
zznBtZc;xxi`5SYXFw@=_efOm=%C~g3b?N_-T>IyhaRs9Ju-u#ZrNee8>GRB+>K14O
zFL0x~5GR>Vv`jOM6Rkxwm&#!xaPDNjh^aGFd`O=e3Uo{bDwUF~2C$8djvQemV%8>{
zS^3~!(&(!`PjXrv$7!9k735@(VF|MZLS>lyTGoZFeEy1%Vp3a^$avgWaE)3mW|2;S
z`m0yePz>MdI-ZncC3RYFKL(9FAVXjR>#jhnT-JRaWECd9H;ZrfcvU2OIcPwPDjs<0
zZLQipiwJXPug9g4MSdYB<3xx)-_<M)Pq|`XP=?fk#zFMLgw|i-4jET<A7;L6B2)HS
zOK{gBX`iMc&Ktk;Ia>+N6YP@rTG-w=F`M4{T(zGNx&$GAv6j*`wjM8YHsI%rQ(o|^
zdRMO;p0JIB(TawcV?oygi-ajF4ffWrF1BHlPR3!hW`xIIu#@@q7|)lW6Nd$7QoSe1
zNs%_a@vKe_8=&><pUZF2&8d-E<dxJDD=y_d4PR4pHvWE4dsw-IQBAL1I~}O^Npt;!
z%!quVuo44D7PzIDYQQ0$x&-$ele-=gYNQ5rX<%oE+Fost?yrC}xT#EpOQufTT5SbV
z>uFPJyJt;YXz2{i?jOYMOyC*xsl`A%m7wQ|7pl6dqtwmDgN4PeDJyulcmb7~2#Clq
z1eA=B4%bZRjYZm?c^KcF7G`<?S4-9d4JfB+%BgpN2xJaA5iYO?UGq_$K6)nUa#BYj
zH2i#UMLLrpoL-Ma9T>nt1IqlS-0%W~YaO*7_*<@zd^^G>3X{c8dg7czYsZS2^v?z7
ze6}y|Jo-P(y>}qheftL<DMzYvw2sK(P&8zQ?5rdqm5gkWJu?p?Gh0eVs1${)tYc+M
z_DHtutYbz9zw6Ue_j7;0-{<N6{r>wsf2p3^8K3vKuGjS%p4f{ubLX^&8Nz-Lgv+Vo
ziz(9zh6A`iBTn~x8F3=w-AD@k;{{wD;l_e!gkxeyC|$`Dn^I$|H|{E#FLf43HA_q4
z9nWTIN-c9*FD``aHb5U{7;F6OWifznBj{_RE(y}?1lPRyqjQcpp7Z0~*{hy6*xY{A
zUH_4Va={7{6?kPcQ1Rafq{QVisr4QpOB<Cwv*O6+#Y%qqRONGgC6RFxweWsTzkqn5
z?EE;Mb~=5j+RcvBb-dz$9gR2fOEY3I6T=x9!+9Y#@M?n$-8l0;JBfs~Vdwr$+{Kpk
z_ggN)3snajt9Oq}?zE%va(=C2ryRbf|9b6yHnE_rPdJwNI+Nt?(AOJGe1f;L;vUzR
zS_+F^8I4ei2{dZJ-fXPa&V!PvCq$TJ%<+2K_C59P+2wYn3}KR_+){P7hI*SETEK)u
z?cA#NK|iM&WS9#M+b)SeI$C`&%B&nOw1H^u&ZS3yjJLCfa-#ia5eiVUXj&is=72X6
zG4o?^;9A8TSt}A?-53{mA02R`o6X)D?Kn<7XYX8?FKnm$nyC7!aOBYh7(1s3M>-f&
zV%@4lImD=raI|n?8>M6Q58?_xi)h*wzG_p_aY75vhq>!fcRePAQVZ7@UEe5t)}U!_
z!XeFv)%o<eyn2_z)SKiOzxp5--QEtLjY)xyp=q7%LSosu#m2z~yUBx^6EDKPVsWNz
zV9%b@V>nV9nvt6uc(zd{a9scCSJ}xIVC??DBAKTklYjldw`V#o#?4=4yQm#5#}>9!
z_3=jPC38ew%7-%K>qKsYYh_7mbl_uCzn^`uYsxHobYBfi%^vJcy4eUNlKMwYX*1Wh
z+M~SJD@cgz6^Bz6h^G0|4P2MPV#WvhhIz~%XFbZIpufY8sCgPivy;q+-$ksv2J@Ix
zQd)*vMI-SC5*c~73I(nbfTq*vQ)u|4lVSSy{vt5gZrmlpt@QAe!HH(}Cc6gu1}YB!
z6oP|_Fvmq6EJrU)!01+n^!e>LA8@Y_%g)wtu7(+dd?J)cIq)^3`cBcMPn!OwIh;#3
z=<>5<#`UqSvDA6`4Ylb351gM6TS~NI{QN?>?-TdakGYk}G;$~N_!XTOUd#|eo2T>(
zHk2^LP?zX)+|j<n;SlP<9kMh~CcK;d4gEKp=D>v}<!{0=`qxypb+43Nlnd$ylNlxK
z#+Bx@srG8e*8M?JK`J?GKzP#M@*+1-Gzc=#)Zg-W6&Ps1vCMe)Zk7Tzi*7UM(fI6g
zNEByA;NHevLu*pGvx@4Ange=zwS=%h3j=M(x#)$`95~Zz2ox*L&f5~Q{lWvFUZR_m
zTIQ~C$CCHlyk(&R7O-B9H0unTK8}`)Cri;CmMwqeM=p~C7llBUl6~kG$EozVHK<Bs
z{d$6Bv%mUhvL=Xhyd|za{DHSa3-XoA33*<{l=p$vnlA2ILEfc!7^Fs}XVuVUWvU-~
z_(}wmmw7&b$r=jLtn)$E<TS@5$yZMv)IWcuFHryN;w8xo-m3mbz8mOuSGGj^1qO+G
z-wc+<qoUK{3{!iiXlT0p;EM^+3c}ewGld9k|5%a|2_v1-u6%Kx8RH#(>T%ZSF?KbP
z<R9$XEM*3IT_ta;dAl7r{YscAZhOY>GOc2Ndz|8gzx>AohP9=}2c?g91g=P3sbv<^
z-&eImru>S!_AE{+$Me}v8*1{SAH$;xKc`~$Iip%JFAFHMI6~#}f-kMW!SI?&J=20h
z6uhpnZ5OMh`zH<7ZjA1_Lm@J`%E$hL#l;$Y*m@;h+bl(evSx3frqGPi9{zKfOAk3~
zbv;r0XbxBe>J}(Z<s?ij$&C8Uf0ml|Jx{f8!0k5Iap~-!X#a~hf!SHeY(V|=&fYHL
z#NfRfSEzcwwbZvhY0DZ`lPV=W=aEjCq%*%8Ymp~cl-umkn|K+S_vBt4<);nor`8SH
zXb|>iy(RZ1A)nU|f8#wuZx7`s9GM-}8#G|Ia?mQ8c!&vGJDBx^a5_+sLRsLO0q*ve
zRGQ8)oyF!Wvq*{w?c9za<z+kRTd4O(=V;b>j~qH^*9k6Z#w3YzHS#s$HB=vm2L|+i
zIHKR^wHpo#E>52dRUGH?o+25(>vY68v8-&tv=U3sRdek7)q7K(U&0-evu{)v$d|qm
z?Qs)eMo*O&NDRyF_Fr%j*0U4nvzGE#c0RHh?ueV=Eg>+!9xVx4^IBDMpYe=45<wWu
z1%AsBv}1QI$vEQTo9mIq?-$&?U!y(6^$n(~i>iQ<OFKscCK@Rvvg3mbU;T=yxqTzP
zEAO~J{|JNP$>$*@cYJkYiu;`WN9;)fAxb_tE&iL1WI*o!QjPY+P&u`yx)@$tJ}`3f
zJafWDW^!p-)ZB-NVj474&q!!-z4y*jI%4YQjj|qXTqmj>8h00?xpwZdGn9*Ti!&sm
zW<q@2`}|h^wZlF-9rR?9mkmE3=a`H!VsGcGUVMiM?%I=KE7Ke_8Lud=9+q_WmRt_?
zY~M-ziwhKz@$*9(J>mRX0MFerjlJl+I(pv0r|kG+mTqA~V<Xkcqw|;Y<l=zzUn}Xx
zvhi^>;}QB>lQ^2SOYyS5&Z+B(9g3ntxfUNK<KD2{bGmq$ze4Pr=dKt2H?N&m65bR^
z5I3wK95B>#om7x8*&?c#^QQBZ^3&~XThrgZ&#+Q)=W>pOjYx&!;%D8AQI*Jlkz?X`
z9)0pLFkx35xso{jF$7ka@``@!qcN>cMZ?dz1a?k}s!A)i2SWIy<j}VF*Tj=U1+o*D
zMbhs)H7Ks$uNPBurN>)j(LlG5bGzFF*EiU7?54SVjegDT#YCkfhom1tn~iG6W1Bvw
zwmLZGg)f@kX}JA#D2@!j213Q7;-r%y*x-+!CKp_V=LbGG)joZqW-G?v$Cg%qO)MdM
zX;RI`U7>a<AujVcqq$Hw-yBWXj&GOenwMiA1&k2rV$U~!lW8WJ^eL@6u8M`FxBq!P
z{2$_M2-Ud*<PJ|4cGP=tJx6AIPA|)EMLqnrsBWC1IMn!DYKK+2_fwf2e!-)at^4CN
zQds{4>@$(8AmqqB0UxD)<6F-TkV~nfK9Gw&S8EoY6N!c5+56pzWwrsMzuZ(LFLYlx
zN><TI0vaz3-P2WHBkjHQq|g7r<ree<kjzf;R%+6(cf`d+g;^xECv4?3BCJrQ1l$k`
z;5r&ZEntMr055Xt0FT{;n=h}vW=n8wa{xy(6vM|;l9MzC!nj_8UrV0NNKN;La29Gu
z6V;#ZURROD8dw-zb!-X8@Lh-tYcI@Kuy*(%quLM#GenU3)K~gjBMlfM<RMjX9%L-`
zyjD)My_0k5Ppk%b!{2}tm^SfT^f$-IQ;gLkxfqyqpMlOwr#I&$GECV9bP!8fMz@?s
zAh>6VM{PL+;*UfDQ|nQVCB=h_b3UT6ns>d_e@&F~ddZEOKYyF2DU}HDpB-wB-swV^
z^Y!N&(Le>b({MERU{BNtYJO`Toc4R60pdD2=r>-l`}+TkcKKh^;b)A{^L2&Je>48;
zy>2Ug0j?aim9q>EUMTcz@q9&frI*n*yqh-%9U8z4hNjI5TtB^%_Md+Jb2(j@J5a{U
zv9#-chQ13J1uzVd1=i;NrcV7~MajU-@W=yWfC>NCyZV3k$K5n=-%~0Zym|eH*Z9*v
zT|Exm%W(gH@CW*@Z%!8jatvz>Uu3}lF_8Y#tNx8e|Nql3U_Gb|aWztp?1rA9?@n`v
zm#-LE_oo&sgHL}YBK)^U^bB!B|0a_2m%B0;tzC)bZ&VZ<dxSYMmdaz&cATEyTx}~X
z59Z0ccX)!yVkgiSEePwA$&?pNZg@MOev|MEZy|6eV!l2V@@-7<Uf1LCgi>=oQTcyb
zFwZW*?Z`|s8~D4&+iOY=U3U(c^nHYI?BLh6uov$riGF(otoq&~YPy~hTKR&x^GafC
zFTtEw3TdmF-k^N%me71<rf>xk@_J5ucIuS1QWRIYJ6L>v<|*DF70m+1g`6v&LhSQB
z)d;^#tQ{Rq<R>;N)6Z?^bewZ@QU2ucDcR?9<$qm9|N17M`61E?JeSsAj47_npBCHv
zkcYDYL(hY|HtUq!^?#GNK_oSkWTsa5cjsz)?Zks#Ps+e$)heTOJw+6XPB#szZWa3i
z(ET8z$j_!n)AM>v^9J2Fy5A(nYY2Cmj{r?3H`Hy$zybJ5N8>3{FKz0n#@tIWCQdDW
z-B*;7e)r>FbKL*(%u^5w^S3LH6Mq*hWjmO`GvY^C>GCU1In;TNo|nlo-C%V!1BoD6
z$dQ+vKeaA#0bz$5pd1mZ?5c^0p=ycWhT9w=VHSp6XQ|1+6kRW88%!815mTlUyQVpd
zK4P_(l;STATCMtOeRDgzEdBe;`j2oH8I5rqti`wQ+8+P49lgRlyum!Ub&pJ>Zt4!{
zB4GP#soc*x7vaf7!G2bak+I`Pdu<tYXapt`{`If_>_7hSzkL4@Q+$M3EHLP5=dJ99
zc%`q*bb|ow5-`mlA*M<QY)bch5$?gFYwWPOkcLM)dOk_oUH)wDBZT9Rth<ZaYxFuU
zlP6+DoNb+!G(h80k=TyD_g#wcW1-dpCcL@Ov9v(hZ3r0X^bA5Oguu97#Jm>))3mfF
zp4ursdR$BlZ3$Z)y!XBzZnYI4VAC5O<<|G6nX&;(@(_`wG2^`cbd1;49IVz|<wt8^
ziJd8f)#`Z7GLQoK@dmMQrxb}qT32;!BPR8;eZT>=1n2KsgM=l-JIq@`0ee+-s_&Kw
z^hXz+Mt}01UF?7Joy~RsWvjh*53=+Mj>~-7YQ8_pzK1&S<q4`wSHl`Om@<!_kZV%E
z%7dc5w4agu;R*k$6LP*XYR4(bXuQP*3p`Ml4|ywH&D@%B%o(4K?&Vz_h}L!>Ep{%k
z_iMrbzNKDQN!%Q;+cHAziP}HIFymfm7Jz&kGl(h!{9zMTax>f-OA&buA{q3eM02?1
z8Pf;#BcDWIz$p<Kpbd4SZ&mK3)lUOY;w)K);%V^c=KF|DromeiY^p^+hG&)UrblAF
zRb4pmgk-pnu1**v+vW7b4|X#htO549c1?75p^&KR8lQL%3%(nu))wIWX47<9R9^~C
zl-qrkNvsfxm}%Ad08!J0;ib&qyo$_5Xd9VxH8*m#zBx9Un=1)_t|oq}D-<^EcpcTA
z<7pqirg;U?;LMy5M6TOBC`iWZBWp|dkVTI`lwCjxs0<v2AF$O3Wwl6Be|t@cR5yB%
z`{J=%j+ttI!S)UAX2OM;J2gOhIA+HAo+?TJG56MIG(r!8nZvLn`LqDx@;jjN3B^sQ
z^c;4h*V!~4T?f`_HzJlnrb7k{Pr<Lm;%m$Eyox<%s>D9bg+f==6?M<>P23<~@VB73
z&BQFx!e90&|78!pMh*v2wAL?PP?^UGkMHJyEM)@wvCIfGix`J1e4lQtfGhbY%&sP`
z%Dwn(i57>slGT;36Hv<OLm?)QqM{jV5a0e$BRl@n``LG74e`^h5ZzCA?qs}dmme)9
z*>3V+IMRfqEm4Bszcf%zL1HL#g%on_;M<0F@=M+a;vnrTfGEw99+F+E6dZ^y!vZMW
zIY83><b-CvNFj=nL5%xjAvMqfnYx1=Y=&wpCU_<TB8!?qc6EsPoHt?FFe%z7AP;XX
zjS_JEZY>OftprpsM$sGw>nnhE@_#QghdFgF;xwI3b8S96o_km{;tJ0LaQzhzSe5Y%
z*tUb0?E)IR9(>Gr{7cViI1~>Z+XgY#F7b6oNZ)=^8okBLB=g$qDoXj($PZ-B4dw}g
z$WZoxLD5Q&*{Tz62yYN-p(~GA{cy=h`x}Sa`0cEPXD41EfIahEVbpFLCbFRp2GoXB
zzKHi-_uCKj37aSRQ`!Z4<iu@-g>oNgDzV9S?&v%{PEBQV+~I+i0(bSHv#Da=zj)08
zPtkwvJIoo+H>r<clCl-AQcg3x7=lxrf7~*w;`D4v9N)XHNQdTBHAiOw1CQ4zZm!`v
zIKR@<<>2f{#_v9kxO-<Q;yY*Lr&2Lt-{QA2aYH5hl+~jPqTI|tTYpyO&TN-T{;YWB
zTmAS|!yt{vdYndc%*TZZhY6G$OfOtlyx5tyKV`Td-q2Imid#->ch-KT72NE8h{z+p
zwy$RHfcvcl;n2ZDIiZ8HGok*=P2$_GW_=ODYg*UMr37Xet5?4MaS|{6%-dr-q2k|;
zDLbEf1fFX{KxfU{j}^t%Mn~@lECcsQkciD4jdy(*DB6ZlwMVsQ9#Y?SS?;XH)K%Z8
z^SK93O%9hyHfQ4P$Zop8^tv6&r$|T`m>iY#7>#!@Z7#yF4vcBfJ`7xWVTs+riXC|g
zj{7Hx&$x(c?BL7KOzdKr5}C=YWFFkbsvLT+Bg7ROasvBGhTiC@P0eWhY8x`&))3Zo
z4bv>6-LzJ|WZNu|>yRU4Q?9+3aG}oU;@f~vBnGfqt&`br0<x`?lELb*R<?qI9o!1M
zjprr_gB#lGfDM~9+eqpdCsUgb$8q`HQ?yP(lXBUKC?2A0`Y3afYf;+>xV438{&>=7
zpk~V%&KIj|uItugl6(UD?F9Df<I*QKj!M3DyFBWAcGc}-q9T0YA7V}&jo%v)NOKKW
zuP?)~7~oi#Up_kAnCS)4Bdk;99B*oO-BR`o^hIP0YOW%!?&iLE6+P8nHcj>;;;VPn
zSIrmnT~}H1U%|385jU1i<`3U!n{4&D$35VD3=P$oShMnYVBp#dk!f$jYk80rI@*?-
z=*RO6Dhf@$9A)m!fl6%E*i^;g6Dg)UQNrT(jV#il#tN#YtmMHfXj+J}@==M-L2C7(
zLQ}7ChB1bjq<1%0A~Vef%IT_wX36J!gqMm0vy=JK^Mmd44OIK{kh7DXm}A!{_H}?i
zA}}i=uwGxM64ynEn8uA(w9;~p>#I-YFvc}4rOcSZn}1jGeK^n{W6Uvr`?*9A<;F?7
zD7VE&*A*WOo^8OdmOf=nWq?H)ab_}dlO<nP_K#(Yr!DyppZCA}bpNs)Y##p2qNG=#
z8f2{!3+}hEl3y62eXbjgX^{d6WuvJ12?}|CmVmmFFFreTL7py0I?N$$Q@ufNk)v!o
z=U69Ne1d}KxVIy3vxuFom9;%HCxx(U+6snPeR}vu?BEZhz7Yd~7hV^KWJ!L6C)DBe
zFNy}P-2+-6E6veDwj#`7zEVaG|0WDCv!xW1Y9xmH2Y5GKA>p|Z_r$uA6=`m<E|%BQ
zWux(r*4Y)=39?qs{zrz2KWh_C)MRikC=ZtGKC$|=!|6<V|2w^$cB!=_guWtsbbVC&
zn3Qn=d~*V`fqu3j4opq9%?9@0bk87Fe_!qH*b_CCQdb0u`gz{pT5p{^_Vxv1T#vMT
z3q#*b^RY2L9FEt#rsUCy)hp3z*nm1wPG8-q+!U3ms&pYO-dnk;8LKH8j>u&28eTVx
z0d=O<2gCJ#C!TBi6I2Ls?}WrwKBw2LZb|v2MolnyE$vaurwL$5x44qlixTMGiTHEW
z^xA02h^ItN#Pz5MEPIK~-!er+CFOP_a|b-$B``Yn*>L;O0h47jeX?^iv17;MKLP_f
zy3p^sds6CXVSiqhQCujdbjjjzy&tCc-x-Nko@oI2Tg%7_2v(kmXj=LAx%9E$X4+Gk
zs+%YvMsZ4>gLoS)vHyjX$n;?!AJ?jj^PaMWHQtz=&DWinOdn&wAZ{K;$tEi7q;Il&
zZ4tvNCp71Uw-5b#_IVocTDgd9fXwy;dITH4o$V;G?zdV5^B=LH5>rjqs;gkTc*|)S
znKND<^m9z+=|pYMyL{EVaB!eD=0wdRT$>@>MW8La9Ed-;28C?Av68yvRsSy(z2=cU
zht2Kfp^ivY`IdMv)|j7PWpwpU$YLowMhIduraB|_Z(R&I77Z5v+4?s_mmFpT0^ku5
z&Lss0rF3e6>T%~i*Ew;O#wGqMZ)#cC;+^zovH)hVb24Tp4IQc(nTE?=bv-sQ#T_;#
zcp5s$i-$m<d>4!8vVEw{X`?TA4EoJ7F)C$UtNrbUY*vQQzrNuArEdT0kC#j^m>&kk
zf7@dxzA*YV%78>;(j1yS%Wi7y@o2&l_v$6pGZXq8^de4~h$r33yM)cONF9ga_%U;+
zipKF)BW}b7WUhvL(WFTm_HOv=bN9wd#TaTzv$Y?~*4)p5L)vGieKz4BTE>h+<Q>#v
zeVWH^I%YUY97SI@2QR%~=S#)IZ*dcdfq!wW=%E;b!i>%PC~hb$ypsJ^g87fk@x>R(
zLK5!1@pz|(U9!RXBkNE(QYBg2lRYCkF&7|4rQ-vSE+@P>eXt2aHo393v?Jp~Azsmb
zi+rnX#w5LFk=yT(n_YArv8agl+XHdSw;E|HZ%zVpFBi;J%S4J(iOE$s8LhP22DKr4
z=%v09?WTFmqlF((^gMbu71G7ZKZHX~FA|ErM!Me>l1-hGW#ez9pH?Ix4>}$sDMB-g
zpq1fS#c8aYqNN!4#|Fsd=OVh(wN$c<_$_XTTi(beW$!Pm^BGVW9Oyk};TrbM5^0#h
zO{jF?=47tlRxT2W(#P`5S?zlp&t~IHhF0+0W^#)2fR3W!*xK7KlGaO@`SE+km<>O_
zn0#RLwQF`+%;?RqaP-YBIH_#D){l48SM6^@ulAU6bcNnCEMD4Cpa#ap*;%#NdX;ZQ
zNw-!*`|dIPiN=0`VhaquFR0)D)ynR*CP%!U(F=6U#fS!duw)G&$U}^)Gzw6TbCigQ
zBcW&qkI39pp}L0AXWHQk+m2vCDOt9UzpG57%DO&#lg5k5+}g@xJpQ>xXPv_8SL`vI
z%U67?Y08_J$UClJv&2x$)e8;A9pq|5amI>W9ok0`5yWXBePc)uHxeUOn9s6wwRL=;
zOY)M2>nxxl(9h(l=U{0rrAoj58CZ1E4vyk!<-4wX$ckRUwbXigqcvS&nYL>W?>eun
zv&CU@{FI3*O07DkZbMi><!#Xc@~u0^;qbdYCNkHN=d=ZRFgc(p7(gV!1V~|R6m9Zp
z3EJFAwg!SXQM+bOQSVXM+yTzOGR(ZSUe4%pMSOGF0eOIYX)0UklQo-(XHU9Z`V%n2
ziX4B6WnQW2|LO%?xltlOD8*}5O7+7~q-#!b<dvoGQKL-2L)%3}%rcrGyQgnxY_5Wu
zGwt!)%VGC1m6bg{y8E6>yz|ERIXneF(;@p5b-~3A7M_ZwCL|vTGo73A&dDl7aQrLY
z`@`zO$s7<P`Wm-LJ&5t+$ps=$wXW*&$Ef<yRF)SqHAgdaoG3@vA=W3y3oX`VybV(Q
z132oX4aAITzRCZCm&59bA)CxA=gy1yc=Wf0Doz+b-uz(DtrPp9IXKYZ-uTJ63`r(w
zQLu1NRpWKOm;5HhURAPeJE-dWz`dGk_-<?F-Rm^UOUuQ*b)B{$`sbeMjFl{d{Uf9o
zNFgZvYt=GS%tX2~6})6bx<!VGQtca?MZT*J?OUy?`%nJ`r2P#y{0ooJp(Q6(%lrk<
z=<WcbeSV!(dlCf0mCkAx>|j0%hEbD9WLLeANU_;snB64y{dXBHK`i0K>P59DwbdZa
zh!Qzxp;hXaA-^Sc1|n(-T{-#=>d>7tzPb*GQyUxwU!2i`N9})rUra-{LMUm0<735}
z{JW6-mL$g-eViy;F%YogvJJP3f?Xq@s%^Qx?c9S&fHy94q{-vr&pUtUA01e8d7BtP
zj>xqO7i#EZZw_hy(BDhj40PI0k1O{lLAKi{a~Z{)VjHN{H*-LFl9NV0&T-O2!wBIH
z<31KlZW;*`{#@ND>#YwI&KYRRxFb)10IUIhMS}5{qzv)(uZae4O*P7_pa<^go{Z^E
z(Q+$M{Zv>4A}KcXx1Qx(j@{IeZ^w;CDs|RAsy-y3EE+6bpa!J9j^5XobLF1Z^3S5+
z`X`Y&&g7a!=+Cslh^FLWU<sX7buRVCh(NLkKU+m$`pHJ8tSiGKu1beD1?SD66UXkE
z@<Ps|zp4Swy*DoYX*-l&qCj0w{&yaQ2NhF?B!*`;r#%S>l_iEsn3o7~@Qv~X2)p9G
zd4_BN2+Rp!MAAJvurtDMcN#%S<^Ch~sxJ)4aL2a!mi^rGeU!(nFZg#L`_h}~d42~Y
zw?C9>QoeiTZ~>ep(9WJsbq4`JJG7?PG<K846ME!v4>_fMh;cayD^4|M`3rx00h}Dd
z3@|0ps+xeHUl9y2T;=fz1lGL^`xw8|x5M=R!0d8MeQ^F%x;S*yqw&ul`q8D=H)B;$
zWRB1-EJM?lYpAohK^NNJ5_)>v`0(b}4pzGEoA&()f**xbF(_pvHiu7>zmKDa*5^`w
zkxii6`j%k=?JwGMrQi^Gc0B(<|NQzl?;ZF`R~gsUt3|AysJb^vwbaZ#xG8B{^&9{Y
zD}S;jyJ~~OKTm&CRd_WdL(K%VjY_1WT;(O|>wwX=L8~3!w^LUuJjMO<-RUjm&+1x#
z_V_tYMP^9xYnO46VSr+eZDW7e@NgUsox2}TC3iTlNWJ!%5Gm$u1Y1iJ00b^Ot(;*M
zV_I5p0vuZj_mSWEYk!Z0^cAQPWi>$1-F~k9N<@0`;>W_!R>Vw11K`+7sZKx)l?fpS
zMaW$WhL_<W>R5b(F0VyyT_-DUe3hC(Fkc0IxL=cktCJXFoFbAB5LXgnDd3Fi8O4jX
z1GiY+^<0zaUA^H%#|6d#Yo|fLPj5X5EqldXJ_S%PdQRBsd@<O59CWUxkgp;`5<#7~
z*as@T4m39lO{d|bilMA~=y#3%h#Xx0!4ovv=>SDe0&Oa_Naxcoc356*-n!YXcFg4R
zB346h!?=?bf*){8UGm>y1!nznsM7&dR&vW|y-7sWc^5cwN38tz@$B!~eg9Guz0BSm
zFW2G&`U!{_bZY7Y)}ICnyy3gxcJF<s`e`!_ryM}ql5}FOPKDqn9FRX;{Il4f&0#Eq
zHRDod!IIc6__0`KzTs&NM}V2U<$jQ1-`mbSIupp3%x<r*8kZJrXfbJ}{dk&V;GlYF
z6O6|+z_QYBZzvyj&VAmJUv|M^&&05JRUd2%EiK%ty1>T+`X<s{;NELJqg0)H{|+Vt
zF>a}oN_`?4FO+-cG{fKq+5+l{>q=}7?Bf=U`bz<7bwMqv+tIXUzn_7x=2Zste^9Y1
z&Zf|>ms|M*Z`POA8m5j4;HKlmU{|~mf)f=g=HZ>2zpHU2ZY#ac-oA1xy!XXlAo=gn
z#Q#``7wpKr3dXi&|K9(xev!l$LHuQd)Q4#5dUJ5L;77}wUBANZvnEThQ@T5jBQ^Nk
z9Vb6(7ar?;ey9*vbg41~-oz4D)_4Wl&gKcxm@9rYbaA|5FCgia4Khy3oamxwEr?Ue
z5?8*>J)qn1Vk5f1Mu^Uy6a1yeUr5@Kz0iUM3c~5{EEKWEFCH}bxQ{>I)hHu6j$3?D
z?AIH_0~M<;F_V2=`ktn)G47hcRQK%&1#?6pQSL=q%Uiu&KzULaIL?9Uxur#48%1WU
zXK@zR&l8-DD>ZOz9OJy$vM#XyF@^PPnc2`>sN1BOX2h|d3o92zxOxs-@hRSK7HkTR
ziD(uane7LnZKDsra*dBw`GAbQa!s{>^6qhf&2PMoKdq=KeY8)L>ENa>U&Ge~>Yc&-
z1!0x$>+K0WFbTEF$>-J+CPlYu_8B9=g_W$DqqG)|R=(r<&uJakzYc$(ZP%pC`(W3#
z-=ZvDvr3nvr(R|TbS}AYx^VW4kngs`rxFEHoK)Y2C2;K*nZ~^sX5iooUMZwKJwvxq
zlLx_|=C{BNA|SNJSztLa00Gx_DElZZd-OI3DUP3<_VCNxM{lI17iYj=lXWdfT%r7?
zrY=4IM<Su{ot9(&Cqy}hJ3HHv-lGmZ(F^E_K5G^1r{`zy<i~DQ70!d-Ml50N95$-g
zrk3Fj%f%qBRiR#kknH$^<gSyqW&#Mod}vM_L5_Q{@<lC9xuJd{ZH<E~@J<&Ys-1%;
zC^x|ViKsC+8DzXoUK0$7>rzQire9&~X`2*j-ZWYLU_~_^O<mq7HC<C$E~1Y%%5YS+
zODmdj+w!Aa5Pm+RHM^2H1!bRBYlh`l%J+g0$aiBTKYUx_VuSpO{I(a$fbn+v!zl5s
z+G952JX;$Hite9BT?+NKKUlOX|LOjdekKUx+t_!l48PzI+N%_OOA4s>$b79A0CYY7
z&hOs?X$^obOif0|e2U@-s)g<#xoFjW|0^JP8|bNZui{5#U;^<RFH4`kX3SFELIoEz
zRbKJE;KxG}<?`N%YC=r#V!pSC+8mvgvG&glow%g#FA+x?W8ESOoZJ8hmj<bkYH<n!
zR}`nbe{VW*oKEWQr&})+pfgkG=vF&Yj+11e@N5=Yx`O&-oanc)PcNMCb`Kf8qVUDD
z1FgeQ;%KP$BYCPg)ANhEP{a%PWinx3XjeTZO>*5)vAuol=$)!iFqwFa*SL>%4m!oG
zuu6WET$Ccw@Kmq2A8QWW>Egf|lgMtl3#NGQz*&%4nli|c+qvsD0Sw;*)XT|mg7_ng
z^AY&{C(yfbz8!s1x*XSp+uW;O;s%7OB)aew*QJxX1U4mdstWj%EGK1E>~i83Jme1Z
zw|>Zb|2YmanO`9@o;0(ylHjGT5*;5d%yUykk_^WTiw-tarAy^rXTxw&uV{u#zoC0U
z?Tsbt+N=S+WK<mTUwJVwYm>qU&&#Tt_1{Y1ezNgUjXKcm!otz@>caB%=B;X>2D<tU
z8UvQ|cQZWg?~nX&Wpj8u{WWqkIpW*t)mITL1QICr_bQ)!ze}xlXakmdG{HZZ+eRq^
zU&wKAbiy&eH#B9!nax2)2g<1VdvI|Zm}A&@SZ5y3c>Tb85-dX2hP2(6FdYoNn+SsD
z3kaIbKOku1FiJIVy6E_8<f?n&z=cP(ey(UX?EUj1B|~+T83aS-H#LfjH1R^;je{sz
zy$w*+7>`4uQ)3gD$-_R5#cb3Y-*<#^`Ar0nOHNIUxLDpwPd!=n!TH#!TPq~}Xibj5
z1N3uz+5&lR7R2Tnj_FEB2Kmp28XT2hznl>GAX{eW#0dF_{(S|Vbye4-e7l+up*6qS
z;vd@SCc*{*ei^05d+N_ul5t};J$G(rzRk|N&S6(S_slh%-a)EFP*0#+_;%A79E~3j
zYlnjR)G(vP$<AA@NlpW2H=m)QyW0w(JeHF4fcf**(+THKPwr%<4i}d522U%fN1G?F
zF?6{^7n~c+s7PPbXnn%R`#wLyfW=R-Q>&-E;0vCnX!NN)wPm#UCd+&Ckn!ACoSP<^
zX!-<C#{0>|_p&;<^3{dE8gy$!^eWMBJp5?xJ5gw|{akc5O;toy=Dv7C7G*(Oq<L=;
zB|%oFQ?sW{QJikJ$h={lx4AbrmAT$>l5+a6Te7G#<88~F7)Bza^X8cIR(MoI5MS+0
zgFR9;p7-Lml0MA-)7kLLeU|Spu6B;xvuI3W<J9tR8P)OwkCg;5y=(YCqNxD1d2+DA
z2nvf1iDCUk+tE2ken8)~^`hqHVdURQOwTrxg=-|<{sj>Iw;zlXV4w_Ph7o@?V23eG
zPf1g$8J1S1Ye^*#&qVXW68y<SV<w^u_9Gl3)1gymoZ?b`{YFoCDOunD)XDwz8<8Kl
z@CdU+l1|~9%&#MKH<Rb1hlrzD4nl7_&$bT+Um<|#zMFl&y~W=ueXu%r>$k6+3Qfz_
z^}}+Xe*gJ@{OZ5IcFR75bKt9dmHyXG`9iJu1uVUC5Jx~tA5OPPd48B~t$tAb&AYuJ
zabZL0@FA7IqQU=T6$5{HK>(qO(O;CX_;26kKm5HXKSE4mJumwEq5aoS@%K;kx1alw
zG-*rPUAYo}yoLS?rvbi@9)ukf`{l0%_dg1?|Mt26_V;&p!u?jHcc^3cKdsQ8Xz>62
zK8zz^wOQP6H~9bh(SQB_|Ig$7@1ONQlT)540AiCzr5O5GM7RG19V*0dCh}VL<C_|9
zGD2@?lVHS>^+emH4(#16V*!&nlc96?uWkE}Us{kjyhe!lip#rSkejO>3g)kDKRGIJ
zJ2JXm#N;E61{T|hhZg}h*K|)lcu@5K?yC9Z^XEWl9eaS1V8r?Uhpi6do8mq&GqT)X
z65l>Jk_n?BRqgku|3wHlR{VmbAtM0=H@^VU?j4Nn@*Lhq*eH3peMd}2VQyKASUh$e
zkGP`WeuX!<5qb%H*dUk+S+qQp7%9cOc>U(M{Ov*i{WAu!ARmV%Gv(&*Rrh5Z^#f!K
z*??hx2ASzD0>TUdNFx)hpvJcM-0lA9bpRf@gtsb)trvJ}S~@JZ%6?v4x8s4D+6Ob=
zt|UI&Z16Zzrtdt^NrV?auprE%^c<3nRyRUqf-fe@BD;=tWL+iYCz-B>imd7mfX&4v
z0JY%HLE2ziBlMpmqgn?gUm>(7EwHZ~Qpi`imlUPHrx(1_gAG!vw(7;iT$YWIvdH{H
z7=8tsc)qO*LjWb^A#PBJ;hpIf4T+AwXpR4I-Fl`Y6aUZXaJt{AIXe?}=-Ge`*V#(m
zFF;1Q1KH@2o4~i6>j{A}j<7Q1$Jp7E5Ao|ioqU)>G>pU*fs;kH0UvS(X4rGJ*FPN*
zg-OCp*-muOJ?KkRb03F}^cZuJ2Lt1SaO}O@Cxq!$R)C)eN>L@|sTO3l*2MVA3W>Cy
zdVH_|S`2d}R&FbdiA}PPC5=(U$+{^U=$uU;dMrHO$s@0bjE*bmNdRG4KpLJ&$HFU4
zw^P#V#UlyZKa!D9XMx4P7_Gmb?El(My3Y>4q5Ul6Y0ueT_sq^mWuSwy08>;KRcbQ&
z)?q$lXhAJ*p@4|7#a+Cw5f1QRm3x+-y$#)=bhH3YXAUPAw<YuvQk`01t(qVQu_%ay
zRBj{-A$wG}n5F<2@{&QUc-?9I_rdMWL5%zn`LiI*%mbc^_cPH6rE{&)KdCRQ&}KhX
zap?oTn*k#czGw>oSHI%FkD!knI#O^bj_1!n_-+gwUz9DnZ{GYmPPp)x@63D1*t~7%
z6WrjFLNr<t{E8IhnA|ugOfv=>a1psLdLjC?!+q!SKbZjEMH+x2SC0;nUkBvZGoCOt
zJS&-ufZF-c@^VOG6k&YL4~aNO<E?M18$P~fB>+z|Jr1_&P!@92<^y!yX9<R|MF4MA
z?xfeYoxklzYX)YiGr5A^OUM;G1LP(yNjwCQ59XYBqS@A|Y1cR}z3Pl21E}bIFG4ZP
z8AFWC!tOckU0=z6Z-L<JT59Fm!lS#^&xq-=a}Y-lcxfqm(<3nhw_MKN--u8WI~s~9
zod0+xBL_xRyfui|Eg}i~um!)M8%V|GPJGPL!wB1dV9!E8M-=En_EggAaU(o0U#1r-
zDQ&E%{O>@_>IFU*W`!aao^_Y+-lGQ`Bt95^=Vz^%1uoJ)Uu$@L0_L#8joFWUcJhyb
zI^*)(SVM{h^XJ~)@T-vxe&*D-%(G;iGCnr9a#Y;D2j59}_4zGWVomuYYSm7MECL0^
zAh)dtz6IlZ-zyo-A#aEqhfvOD!F0RCx9g>{w`tST)2#CP<8}mq2lL@*ulqQYm44lQ
z;Wr;BO}%Rq@syMD+s|kIh+A|z?VbMY>35+iJ@XB`%@lqxJP?y~WH}TA7k~x8Ap7<=
z<nwi^C1y?LnwbkWUpJ5B3$?$C(8J(-<ESXhcDQDxZtB?IWZw^6?MjOjdaPde$;2wi
z?2g#^=h7-Yo7o9G2938f&>|cD%#`^22ofbMD)?>30gpUpW|B8}Dk+M&;^n*iyD_>1
zFevTl7F+Vlea;PG-S4vzPm;(I*ef;h*XqB|QO(|NJpO~jj8S`RJjx&hwN_Zi8e8Sw
z_5DgI$=fj9i)HB>FzNR9Zozox_PZvbRPKD*CkQol9l-gH^nlsuu%?Y7Y$zSyY4@FF
z9`=#;5BSo#zZs%UpuRV&nGfP^@*iX*vYXX`$!8?z$Oo%kGslw+Bh_obYc-zJtQLuq
z+f44<5%>4o;{X1<7X|mnet*4dzfH`a@sUaH1vH=mVdY@jUEwG~(@;zL!^o{C>8b6<
z@81!x>DAXLvc1F96sMM&#oyDopvjI@2v(_}9${>3`tEIS&iaY}&<ho^tpU-x5yGI7
zTFxmWr<SG%W$#<eZR_A2`Y_(4Pk!z>$*F%t%!6fz1WNA11+g5KupbEJ2=5O2a<*?q
zJ9P6fBa?X#{2Iymi(~%Epu<fx-YFSlG#MtfpQ?4A<bfVVMli4#aEj!<h+2gM+;{#C
za1yA<u-|YHTSsG>oizYYluw#lFd^&06(_h1$J5#j9`lzM!4doeBc-d<^$AMc1LQey
zVsTa(_`70s4!6T5KepoAkGL0??>J)kK&<t%k%?xP^gBnsah5hor$x9!=r*hA4@-tV
z#Z=-Oyr>-RhXrApz-P&cYn8@V{}yVW;5J;tbi-Y6?sJ$Q9HL!X24p`s+9>8tmv@h#
z{v#yvu+ty8>GFV2C!;>)!&{*w&5qdZ$x94D(Q^l?2`f|><X%L+nCn;owAQ<yrLEF|
zZo*W>0wU$9XZ%@0{B5G)sUy~)dxeo=Wfn1cPNQE&-(r_vK3dM9muhmbxiZ-P36Wa0
z%CgHbIx{)KUO3;6Zv^nS0Qqr&=_OzRV9dy8E~QN`KOo*kY(q=Jn`yHgGJTo$%6QZk
zGyrnf7gFWowRuf||Ks|D9fv#}AujLqa54LQEDTW%L@)vZNif^(2clnI6n)NNWd_Z_
z2P=q0&k7JeSSw1~tg2pFU)(EqUd=p&Yt9g_E36mYy8s7{k3x$;r;zmXuHA08_vw<Z
z00-)<)J{iJRpQx$!&&bM+#6i_1_$gGp(NRtWy2CXuo70Y=teYh>j&|d>9Bj}?kx<N
z@EKN$N|deoYSt{mTXKM}0UKD=XI7h++I4S=?gjN93t~?a+ZRUGf8Vyd&y!)_zC@f<
z5qHPBUFbR)x4LR{owzF-_RUMkQ{p6|n2DK5vP%7C3bqT9wn}_;yY|%i;3i?C%W)XV
zktWX&QD=x3P7*1TMj3}9sOelfBzyxeKO~e~Cdr*4VmOG7Y5$^mF)<Shj=PP}4_+cE
zbHP8XQO>pr2{qoGQ<=OqxPH@?M$4Aw%ChImQYeALI>5c)>q#UL)XkmY3gUtx_QX*_
z)!LR!bvPk#$a-ItrgeMve(e62h+VL9XLTU={&@q}zF57W=YfF_2~5<b93(m5r7Dsn
z-sWhp$Mu%c(0`Q|PsygD8<X@N)ef&<rQO-%t)-hx%H?iNU~ufqlHJhipffaHg{f@G
zD04+&>7k8p@XD)BVt0bh#D4g;>d=r^7Nxfn5pDLfaJZ*-2M~g;kOZ?yVpx?R%Dl1r
zdB$D@cX?7Ydp{JUxw-e(T-xN=Ui}>7eNB<2x2*xzE4=}dO!W#ONb32e@p0rRkUnwF
z)R$COM-jiWir;j+Yi064umB93PhiWcgsL?nW`_~BlVce*y~!1o9I_4D1sz*X-#Q-y
z?pi1D&9Xd1uv>NP_5h!JPNk1<-LAH*yzBw}!L&HiE?y$G;LE<%?2fv^xM?r9cxuvP
zZX!y^4fbIuv5A&gTT9E*d3%Ly{c!{}$3X+KJL61dMbv^zv*)vo!rud5g>~wVczIIl
zC5}F7QiuGlNk3{f`BdN3K<NmAL~wQ{2P;;P6>E0_yY&j<_fDD9#EuJg?!eOg9HJ0u
z&wyO^yYX!C!N?o;_wi<^MfA7pb&>H`yEhC>GdE@mV#z7(22>-3a#OQ~*d7&ck*epL
zmnuaUULih4xzkUZvp(V4IE2Rq%*f18?+I!L)%@rWt_<OBMFd_q*c>g40$t_ll?44;
zcmE>x|E)>=cW2kKNil!**}unuWS9$G_jO-$6F;Y8^zbXzp{~LCW}p3KPkPb5Rt_rn
z`+Ccgc!ESY-`c~wMv;YO{Mv7)>hF9~=SmJW*#pAUfYAPt%c!sMFAp%Yq0OH5ZCuNu
zvI+S*1r8#pg=V<w22!~=Ra<lxB-}X|Y2+&M1n^8_G)1@d6m5<{hKu)yiA&;_g5DBD
zkR6gv*igE+0W;UpJf@X}`TRGntOcQ$JRO;8Odt0d7^38EJvMQ$((IOgx5`&9J6}`4
z51#8|p>R0>^@^i?(y3sz9F|?WJc)r9f8$gV(9*@dc}OzvqO^n^n>Fye;l$$R+q0P|
zOTkAvdQQ69--iC+3!a%zQjh&Eq-5>?_%4z-QD|(Ih{+7#pWiPX7*vZmT1fqf>@9N{
zA0QU6D+E#Mm-9$qS3=4#Hs6r(=DJ99TzYSH0F5b2+8E>X`Pq+4`<wiS`+7AO!qp-$
zr-`eDqqsBEq%msmYp5BF$oTqV$>|XPz5R(6S}LeR(212{@Z<E`bK9HmyTtpZ&6`pf
zx}RX473SBE#(&bcNb6rdND<<n-4|h6QA*>QcX>AR?p#=1$(-+(NWP4yD0-3#x3{*x
zY*q<R(NOg1nUwD&s>xzlAX#%L37SgxfjUuDR<s;E8KZX9?Jtz#s0GVtt($tiDl_5=
z^cnR3UK#&sxo9DBzjV6&e7{G`DR>W$y_}bvTqSqhKe>NMqLu!}NGyiY>C%ul?Q3?t
znAGz9{1-Le$t)BYRB+4*isNv{+U6K&qR1v$Br(M)crmlcFGG5(lGZ+y)uk^Lt0+4L
zLogW5fa=kq^Ii;+Df)|H8F+X0aX#TWD{-N$2ig2gB9*-~+A-ND+obGQdbS*VJVx4B
zKdyg1ov{ahF?l51#{-lG+UVyuEH^-RS2?Xw%7Fvs-~7369oK7nt~FSR$X*@SQ<pE9
zbUrztCOd!i^J&95)8G|t8G0wq>8E~W`{thGH1a#@87i`PI8%uI2r&%RE^^n>>8R9<
z$`#bJIq3dktn-~RW{TtNi`3A=d=!Mli@~OaA=vtfCH5QZz_*A2_EOrM6QR0tpx}}?
zxUuQaHc2{26F}OggrY+zp*?44c4UX*BQ39>iV-6yZP6n90og~8rgMkyrm|zdPNdVx
z+vutzw&GK#f6VP-mO=Y4Y#Ostr1hkK?BTw;v|i+Qctfz%+hfN5ZB?Gr51Hb6Z8g>$
zFy(;TE`Z(sANG9-#qQQ~+s|J7d+V1`6@00Jyme<<9XrnJ<BX3QccI5C7?v_qkPL_@
zKQuo&L9M4930AKGr6P{>ZdojvRQExo;Vr}6772U9A1sc=^(FSvfxD(jhdDLb^`IzE
z1}d66s3Ndd_O9+g1M7y)1qpbFK@w%pPTOUuIY7-Kt<y(Pug~tgtL$$9wZk~h&E@Q^
zn`KfXF@YuxyU51%!?T%lMX<PNf;lhR&;(bL89ArVU=XCqWQcRgj%;?i`ktVlI>skF
z`E!x<?N%e4RJ3b-McmP%IeIWZGXm5`{hH>69LIREZ^va_jv-V}yU>T@1IGw`W{<f!
zEM9yf)RRP^)*Z)7H2#cu=QyK}dnjz)wkzV*sxSi@d$qc+0BX2~%EidiL=}3duPr&g
z@UBW8T`I}Bzip(Sf!od7$&8-RI5UM?*-Jl>;3ttjCJ<lnx=Hh#DNeE=UGuQT7p)&k
z&p;3mMV(*4s^xEh<L4ikV7|C?b(~+n3bhm#udp#yA!;|2<@~{BX+G~F1J3e%=XtEk
zPnk`}r<4RZTVyPA1}lwY{ce)%<4%c9#0w_A<+u_kcv0NeR)R=e?NHn9qc^#Jgy<?P
zi*C2X#SDntO+gxU2apy-5vb0Yd*rs12*mg)1iOs@g3P~aZj7raTsaMXhcO?77JhLJ
zE7nIOi`8!R8yKpGf0*X1hisErLRijCyEE_2;R*2$?6X=8iGF8R3rEb&ZVHFCLD94>
zw@#h$39tdY-ms<Fq-AanECZV#zFmNlO%1^Cnou)Gvm&rkrJwTg=Ng@DBmNQAy08k!
zMkb;5;@MyJL7rj<=efeh2UZ$Hhq^&2U<|X;4C2I7OaW7sWP>Ny9Zf1Sn$?By{*umI
zbKYU;)t)<`Rka_Jyf9D#O`OkH8fC>)Z`21VfgY%34PZE_llsCY<KP0>VOM1z58QZ*
zh-Q*At;SMIVjdD5t6N+7YWF<~QPOj>r&MIMuN0O|&+Xflg3%AM!ZW?#?|f(l&HcWT
zY3Bz8L2D$vN{f^?Nhycb5L=FtdY_iu$nMaH(>-B2wJr~W+b&8!M7=9W3MDB(&S;4<
zorPs&UuCTGS4{)gegkckJq_12N@huKXVTEy8y8fo3$@I#!+f1LS7%fkNbSx?PA{+z
z*F5c$k47*Ee3fAwBldx4KpxAm)az1Ec1083=Iq>mzDz_jk0|gCgW|~R($8{tidru;
z;xft!wm%Z?Y*kl@@v~sAFppfXKcFY5zW~pLzbc_|>cn8lR19Ij{4~YOD{KzcO9xG-
zZgb)iB+HESZ*}gSae?j1)46lYDKQG1>XD+N=%0#;YKnQ(>wljt)>H_9L8I|$m&atQ
zETaNPdO%p2cj@ttmxKU$?}Ft7mO38~1>wtG%BwXU%VGk3)Ynh@nl|`2uL5rOgfkXL
zwJN7pV?#q1bv1E0JTqW9pK}`(*Y~7D?!4rb72CN?0!5lx4*<ePH@)<@X>Qs9qQ!2+
zC;IcsprXry_D%LfGoW6)H6sW;%saT6U$C&YAP00KtXNBa3tj;vryfoLLoa#Dn=~<5
z?F%2Cd-j{D%TaGfj8&wfVzqASf@nyN<LgX&-^Zm4?v7UkiYuh7cWLWlV=Aq_HCUR$
zh+L1WZ@u5!bF12hzT(c%1O$Jg=QPbCU2IJ$6+vN8{mJB+=Pvb~{VAQZ`bC<@4-d%%
zI=hKB56Q#OCPYI`4~yTm6Fc$Pmwnc(zvOY{#Y`&fK3O%bRq|*gIT^)5dNGs(NgT4V
zeeZ7DqSHlFSj<JoG|zX9Xsg!%w)9xjNyQ;G^0Q7PshX>EqQ<`HrryvfrTBM=N7Vxw
zuN-DlRNM|$lx&gos>MPxTFQv~1S#^Vw`^}Nic9;8KIBUTlR~5nFr2*}{}20;hKW7$
zZ!~6BYrq27(B1UIk1w_=kd82i+MJ^3159PuAkXheQwF=})-d+Hze5l;o{6zRbg!)j
z<ZwLnd&KT`uP8dsAMG@ORYv)9eH1PC`PyE>$uAR%(TA-J8}8if)d+fZe;e>q5jYht
zIXbF$J9dC^P_hhaEuN|}Tt%)4J)x-F#pl^Hkdvm~8c%MX?zvF?o9-t+D8b540A`ra
zN!yOs%Az0fxBC%V9btXSCUmc~ra$>a^qu%oz59emZ%q3+ZuaRcXe_8Vy}Mc`3Yo9~
zka2Z(Ac#(u@z`f*wA>`U{loTW6%Z))E<E?^RY0vG;9k|h(7+D)!A*#&I)l(29|#Vf
zh8E_1Ah#i>>0{Vy<)X7InEAIU3M3OAqH2ATWEs$RX2yPSSZAc+9nKbshb;<q2iE3e
zxB@Nu35E!DOyBL9IEn1ttCqYyz3KYfM=v?N$otc9qYN7jLmQpZR@K3O!K<t&awlwz
zW#MES00|mBj1*@fWozX8Ve(<;#!B6!eFA-%Cc+{&j8LMO<00@^;>1V8q$FK^1u||@
z=~V$#a`;8i&9TK2q#|38IYd4@8<4efHSwrHXrkY){b;)qs-KKIZ?0qp&xwNU2e`P&
z6D))gB+KWr{O|&el5syfMb)hVP#r$>`?yaj@fE|G6>Gd4qHn;-sGXS~7{}YirXLpZ
zzB&%@1uDa(`f`f#&Tnsls#YhZlArSi2{|{XZPhi9%=FEFe^Xzw^ESF2r(aexk*bj&
z_-#eC_@-+lR9lixe%3+j^IhHMRwk~c5oX@bU0dM)7>=M+G1!Y;N{}dG0(LDy2E3f<
zBB8Y!A}{&n{Ue?Lqw!qjUS2TNy7}fkd-rv4gO2D5oP&ERNrh#{tAw;*LmZiKZfXs7
z<ggQ<tR~>-wO=B|Psx&<=UKY&&j*HSQ8y>R)QHX8Qyejb#cG5|9M03K-fGVdFZ`n@
zgSmwd8G^<<zbf@>2=+{r-1Ydylb*?GFEV#DOZwkH5WL-J+_4HTn2m}^>Kxgn;&fF~
z(`({{$YGm-lRv(Ltt|^9*-l&^279O-(!`t3LF^-D=Yjo^Eir=luxAX2Z<+@9R8(yA
zLi+omZws1_slrjiIO$`+o>cOd<LFe<u>$!Z<}qe?8BmzB10i-5go<b1(L*1A%&GdM
zc9UoB`tqCa>=QxzD$oUo=4j~hqcGz5kVJOv4*}mMTFRUZ&V-;Vu+mZ93+zdX>p(#I
zLbI<~7TEo|E^sJRO3}?G&iV(<^ViVoUI+BX&Dta<wU5y|uIoFJ7c-=4CnhUWA7@=b
zr8l9fX(O|6^Qo%dLDAC|EIXTRP_>xWp;IAMBCe2J=~ay|A$RYiJ0y}Bo-%PVFVccK
z-thkEj~L&>oxv8*N*cxhpcN5Wr`huZ2v~DNbpe<s{+w7)*-NYS*DM1zUZvsrZ?c>g
zjdwke>_e`OGzeEnVy_>zcqZ24%cfFn!{<A8y9A-`N{RSbL^=x@GU&x;8)-xKG)-Gx
zm*UI;vNLmCFaQe#(qOy2u&VzTo{Uo{QzWjp$G*u*^UmTs!&>$m6@hr1XjXGGizo~y
z$`2MXy|m5S2C}_Cm3^9xH`?e&{Q)z!W^QfNVBj>~o~G~#W(9_S3i@Zyravc^Dyk0)
zhL;?{k3s1g7qCGOQpW@M-qWdR2i&TK@zZy??M<ObIDcoZdUuveC$@7myHr5z^S;kH
zyuH?9k+apodjl;HGjCJ-d3p*m>+~~>1nNz#w2by8_PlCEMBf*cQ4%Z*0r7QP?IsD|
zivM`zmrcm)VltM{JO3tHeB?n8%tmG|NMX&B398FxO%Gi}BJ4~nxn%udd>D5N91n*}
z3Fvd!<6XePVGwo5I0t`i_gqqBVwl~|GlW=>a&Z(`Z+z%a#vD=H+ZZbjR<2de+Z-Zq
z9@~`r5+{W1lmy20_a@@DyDT{TxwmPpNZ}`^)GTs81!cpCCc?%Yh>4w3uZO)?fiEi8
z%GM(*+}9pMyws-YVYsE-|D%$?Wmo=it@eA5*`@)JSo`|R&HCkot2jdvk?s58rTU(P
zqlfCbFXDuFXbWf)9qOgzVglbZ$OwezgYY88FN6=9B^~=jP5(hbvjX;O6!po*$~!d+
zu=PUk>=#|WYWMv4J_ez4HQh2?ohE>xJ%H<_6G>C59M^nm!VyMA@SrsKf`2s&_e>U`
z<l^Q5w!scEY*|?Iwi3E(Ct^Vh@DCAYuJtRi)nX%#(kz)H>c<NXZMQ#AI5a)E+*iUv
zXC9rn>?t_Ke8g@c*4g-m9a&PG%hVe^HkBsjamFRaqOv?AuonIDl|1ju#rL|!h-zA7
z=osO29)2}WNJJ!6&+$<r$2h;k?0rEy?quSJ7EPu*9M||(M<W^EG7y!-`-P*4T=kk1
zM@Q!FKc&6idG;ao<X*zKJGV;fZY)K0VU>m)IOkfWsdh99O}yZ;BYVtk#9_Iga3-(3
zq;%JHEH0M1Y*;;tC{U#PRlBr;e{%a&^3wy;S_^B6pDm_b0BmpYdH6$5^+u!6uIAyG
znth!|Yv^L9&$iw(ua+^*122iFz>9?Cl6;{a;#|d$+L9WZ7RAqL*bS*celRvb#%T@J
zU0HtTQU`ssY6?Fr&G@8k&hR$eTM_7dig;0|-O*B0Q&j-VT5DruoJ~{{8>ri=E%X-U
z70BEGW4j*B24kG$dGi6E(D4fCjT(Vcn%T*Y^eLe+EFynMd(eGRx1#WRpC4VqWTY8(
zs+R7Wxl*@w5JX%LRC|j8>Ndz-<5RVz)Lrwfs`~4AJX`SQ%F-ny$Z2M%>kBeK+K{lC
zo2|;T>z^fwuQ~sN;Ai)4WX@T0{>^Wfh?a*t7z?ZlCs~mgKdJdI@wLMvi!jwOs`TAQ
z&0Ph;qw5<?B;13Bk~1=KxKiFK>S}DkXC5St)KuNo)v;NUubk@s*{Y+i7dTm^F8~bo
znmEC2hgQOST=;Uq4HwDjjfTK4@$fz=CI8R>`@1e7U&y#0_%Y@FBm)Zh@1SAFOG|lI
zQ{JoE;V%<FRleicq?OJ!5XMPdxC<9(hgyCL<?~jgbMlwC<ZtB;%<4|u^AecJDQdFt
z7s3HB@&mEmoF640fPzhDl%Up#iND(unqp3_7fW#ErC`WOm=@HM4q&2cJu2(%;ctDZ
z5s0KHJmyH7>I|jcxcR$ABgAuFZ=#FWU{8BZ2k0$2Kh^OKbJk!M*eZ@oUq;B&{z}o*
zU0S&M(-SB41<-QGB+L4?F7`@2^;QeYJ)W#jZ;O*%jl$lSr0@)4tPrH&_rF0?*~(V+
z7G;M>0f!TF?N-(3m|sC3?rj|4T<KB~RXB0gZq8y{&tK|>CgHVpn+biC8SXMag)+)6
zle6an&TQApC~jQxO6N(s#G?>7=Gc56u?nc6`?$1B-Fn8&kuw!vVS4D~nC0DU%j=ZW
zXUPSn@4h`?Py%?i&!sFCqN^3?`}W(=F`rnSiF#MN7GBDgPcwx--+U^=W=1ZJ*mx~|
zBIvBa!|JFYHnV;A6+3TqhSc0REMMj8tho)k*Q&lRn_W}{CbPr$RacH4;BUrxcs#n{
zqa`D=y@|GC^gL%4Lim5Ed+VsGw)KBl5p^RfB_SY<ARr+U(h`y?-5@Oz(g<uq5F`a8
zqy%ZDLt0R2=}@{Gq@_jR_biXzINy7{@4xRDXN=>(jce_-=6vQ8pKv;BvkV+^6_)Pe
zxuq#TI{iVWE;fsj^&OPkqmG2)F|xL=BFe#uFAnB5r2Itj?$REbSU7oDWe^OtJCwdV
znx1p`C1Cg;@fFAMNJt=KAt#wiRsEHTisWIg@zn_nxo_xSw7H1!f`$F#qWYcbxsX{u
zDcqh51e5W}iD&d<e2cOOZ*JB$?i7l<RR;3m5d)8>o1zkT!Y05~aSGl1)tm?Cd{K;J
z=VeTx$`B9s2eWuhVo|4R-&aNnDOho=R|}!9GtWlFe}0<Gc(VG0Xz0r~{DGK(I|0M<
z43?L>@>rPWQz=yjF%6&b25ej38CoDXNYAO?E-hsK?yFP%Vq1ZTY=-iHb&%P(Me!ZC
zf~wj(=ODJ^6pL_YclKThA{F3yi*5%=43SxeR0^o^G7t`}y?S}HZ&hNKFhS{W(>65}
zwf-@EL-Xd?Ms%U>THXM8Th4syxyrcjd%hx<JBR^X=_V;Z!nA(&z*uf$P_`<F-eJwv
zirk4+c<7ZwoMV?qL!@Dr_^W_LxvWCeI<D1J6XL&=lf#gcknvS_l+2Xbvi(Y)WW591
z7&E_tEPFlch7)@Nsw!ZM=v7WEz2h^T&QnXr(I)W*V`;aY<JrjKpy|@jGwAh<^TS$L
z+)ji+%tOKZ<mk7ns{5y2PK;=o6_t-RKe9c7#G1X5Wux(o+j_T#s~Ufajw=Pl!H7pd
z$KwAGDA-3(a0iB|M1L<jKJqeHgY~e)u4bi|NnXL!k2Dugl<Q5@_zZ5+vL7C;sA=30
z)f9HOn}gDS!K?fQUm3Rp*X;4RV;XsX2cUUAyu1do$+y(!!1S{Rrk$0sef<v~z8$=b
zal-#t!m)Lf)UKK68|Ht4!~P>CN~8x?Yj*6LuYZ@G_GD$S1|z~DEQXiz*i|y9QD{uH
zn?&oMBD-efhoGQJApSmbW&?Q0{2dH$r-qF&xCbTkabPV!L_LIj@;Kl$1}vbSM4(E=
zj`tY;+%%`@yaJ(~pZ?9<-;pnRkMR(9`6Yld6d?oPu5S^FIj5hTI#&o|-##$FRBid}
z51_^|EcRF6N3M<7i<;rg+-UfLRCX3%A<7L1i{6#R->$*`8BZsI$wv=hnb(;AH_1E8
zjsXaz7(_rl%Nf(c;t+rrmH;wwHY~}7yZtWp4_s!^rZ~ZEBZSHzyL#o%?-nqO$o1Y$
zZ@&CvQu!a>+s_}`A4ecMdbnRO|2VS!FM$eHlxY-#m<{`E_dj0Ee;RTiztVXUsm{D^
z$`kyzNBTd0o~VOh7-jEt{Eub&-=2X8F#=@8K6Uf<-}lFV|I+{VNl!;)3$xx3{3DU`
z506I&0kX1^U~>H{kaavzV*KyG%Ma@DYF5b7)_aU!1|2n)G$XK+?mo8c8T%WN_Mb1q
zPECNojVW}tZvTuGdR`1`0QMUMJsncYw1p)Y-&jK7_iB7U=X}6N5e93ZxD6fs=s4;J
z38njp=#t24`(x811cfPlxE`_$y}f{K*5WUUh@jPZzmlxyaN&`gdO(_Q(EJ9Zw9=78
zK$NuQR(xebLkw7kys!{hbPeeMwt(#~g*uCvW|S=UKKajg{Cm0Fvw*kjUeI{2>MzDW
z$v7sW;PcaCy#E{2GXuarut2aa<5h|q5{dj+2y1v7<X-!rq93+70l=z$+!YGMY8tBK
zN7}EL=Hrj)R^nm+@F9E~9Nb55fhP>2YQKu(=<JV#DD@*A=uog)WyJQWVW{MCxw*Y=
z-g@>C*iS^FAMXku6j|Y1k3ec}OXzEtf*jcA%P~gyaiHHZ8FB-PFTp9w9YnzX>1+Ex
zcP6MX+-PC@j`3zh{Mx?u^qz>a)jG#Tg%<E7Hit2P2!bcm8Bpbt!8~>CxB#ykf^tJR
z{Rqk}_?<1(!ajb))T_wA)W$6{(F$|z)GipsHj=Qy;KvcfYIc^ZSy_D!#xHF{v|pK2
z&3xcSFTbVvxo;;^dkn>r{}E%m`ZQ%fMBUNS(`VAw=*j@9=|lVx#Ppc9BcPfW@P0gO
zXc=qMaM2PZsICa4DgBRNG0zWCWxrj9{#P7MG{+j(hxGZybKHD$NvU&W^O&HWLm!7&
zN5f>K1Tdf<rdBGU<w`v%vAGWRwB=LZm2V!QTo<**3c)Akkryro*vG!uhH?<04s{8m
zm_{NoGj+`?3d2v7DR_XN`Z5qhA1$3eSUxo>g@=&<cZ+}=1}-ScDX5*)l|6lqR5|MP
zpOGoY@#b-mI~S8R{(a}N$mUKvkz#V29h0s-M(^o0J{szAZ>|IaMXRD-IjIzlv8b^h
zvpgjLR6wkc`LzuE>AzZimHFeLEP}}dr?TEjTos>RZ$8KhLxOs|Gf8Y2Bt8njkF8GE
zZA&H2ZF`_rYa9`P2Yigf=X87%&>x^I0c|ZR0a?6+P<96({}v#iN5E5~_W>9;IJE~m
z>qeylp%w1ehw$i1phUb@_{w?qBBCc3AYl<EeYpVii@0mXdq&o<YqWJi`*a&P`28?N
z!lU>MFz=PMx{_`g<^ec}zU)e+^=-zp$9ZeJLtkzQ0ShA=835X>LjYYwTzIc+4Q{~F
z8ibA>*{Ihz@SNg?5P4<<^UPj9grhsNHeL+e&le?ofX6Ky0|aF;+9?Q;ZL<JyO3q9b
z%z!4Eg1$>R5gNFf8unlvB9`_21lnu5l77RWPd8KoSd@zt!EhL-J6&CUeW~a^WCi*j
z!_d#a0ksTwmiZvWB+{vuoW8)t0|dCm`z0HYZ_A$nE5Gg)z(Jb!A(1PZ4C<PE`1Xzy
z3n4Z!?%x*{N8KPZ7=(`*00KEU-f7aGBi%c~QqJ3UdJoVPW$UWT-(dvCUMWpTkW?83
zN{a@nEp`nrg~nq-D-L%&j(BzvMqBA^?g2Ur;IuD=7SQNN^?&|1IA%u_v<IV0K{iZB
zgV4db^+5vIyv@;yUmivn$0&vN4TL^x@r<kY4cN>EI?kCs6jMM+QR2?$A99Svyz8%r
zzM>JdH9Xu?H^DhX?{2o0^Wj6F&LL3X9MvOa@$c_-cs%qzLtij@0O-)%kvZ?;OGByW
zmtvhOKsoSBWD_wsnR^1%&ULRa%+Mys(9%SQABxH)0MwTt;(%cEOxe%#e*N4&D3x6@
zCp5nyhvn187pugR^9Xp(zKC>Y`2*FLaS&D2*U~Aon|^^fErcFCqMX@d>LdNSXy29^
zd`F3YgZ9JQfb~xf4z@-)6Fe#jd0(~wa@~ZGAzv)Ix!<>WZzv3)()Wl^!dJY*N60k3
z*Gm;GmHReNa$B$x#~|uP0qay`e)RkpVl}TxM3L@`L#g|YHDL`s526OBeHwvl`-1x&
z&`-4_dswEwu=+OB1Lwj}irpW>lP5ww>fPv!{I&DWN?~ZAoB+bscv@4aoklMP5O!kI
zPfUNp8aYM*6stAZi1;=O$!}DXbfIb@pp8N2E&5a1ZF?6oQ$mAR1}7!SZN&XF5xnKD
zyb!b(eW)x!0*z~eDmf2;SUg|F<m-J6>qJ3FEEZ|K7k}ODbO&=aA1bd-G%EKEP8Svi
z!C*t<0<>-g6Y>PrCz>DyBo>Q6Vlzqh?#J6&csgCD>Yn+LgJ;+{g_VSSov*9+iM7Tl
zoXPX*H5c@8L}i1a!SLY#s@Nrk({B8SyHm)yd!~R$R8Bks(f`R|nYvA;*G}R#!@Lk{
zXZKqR;GRE+NTQmt=!GZwtKjOLMa`aeH3*v`i$wB!fX?#!{grAIHI+f>Nh0-+j{W+l
z-$9%f!+rWOTLD!8=OwA*O&0*S8xG^CUQcJTz}O!EF0WX=j2qIL?k6-1kfO!z%U`{<
zUi;|7{s?aBA)qt^+-aRabjR9|Q;f>Ljpqzj;7Mru_TjG22sGf_MA_xB`4@}=I#;yl
zDeg(K+FJ!0n3p+Q*~MBpF-x9J&=;sRoFejT=Dn|tXW#81PURO*@(vNM+r4g1K!ZCg
zxQKxv<z$ctHssmWSk%~d#Lw5tq~cq<h7(vYwc$cfgw3%HxEpX-*SxX0%fT3Jrtxd=
za`5SmfW3VPucySki%?d!(D{2Unk{{R62!~7CK-TLl1-gWpS`oj@Wn=>rXjzIiNF;%
zqOg}^)UQ>8^Wq~+(`@2woNVLlnCwfaEW8uJpa_~CYnEwx2X>Dyxvig$Lyt-q>+I#f
z0HAInl@bQ+`g7L8FfQ`CIlQ~-F*qXWT$x#%9st67uN^yv;wkssRDvXPv%+Z|Til<%
z39fr-^uEC4n^t;h#2mlCH@hE~KiN?UhkSZpDo7W#y!gEM>`gl&&kgh$1tE(-EDiyW
z#~=BgNRewz?-&`1zVsiDcENMscsC78tQZ0;$F_?*KcXoo=+2<Jg--mRQjd((<CJIq
z4YzspIMBRoy5_?)#YPzVgxm@TQ+d>z>Cvff5Q2>vK5_4-zDUvQ!venYx#(%is;P^_
zQK`rPrRgN)0;Gv|Sp%e3RPq})4kviakFn;w|NgZ}@c8AO%t&p<7uvYW%!?Gc$rQwg
z68tw5+4kzEDMP0OwJeKOzI8fxohCk@I~q^;A?7$}_2D2au&w(iXp-_(V2G;Yop->3
z`C?ytjz^qKXFkmI33fv_C=n$9|D$T!D{FePnPSyGNn~?Ga0L^acs>o$vMV2FX~0Nl
zEs<H_U@m_I7wbTOnZS5AO|k1mP<dsFCauDj^|GHU4o9@M0PeD_ji^q{9kyX5ZCgAe
zTN_YfGj6JC@Dcp@WZkja#P+7PEwxBd&&l9w>{E0WzP2kW%;o)WCGGU8-44s&T)pJ*
z1%{-dFvXGg4I~T0er+&JB0JhqrkBU693)g$21w0l;i4|V28oLnw59Ky>Sobzn+#q#
z6Y<xi?}uqSMQC?MYVSsAM-y#$-nbKAv<~I8?8@^i<(;(UyG4?btiw{9Ar3WyvGhCd
zqlXubo40T4ATlD^1guL->#ptA1!@R{Q{S3BX)K7#E<%W-REOk+wq7)4#*h9c^0sF-
zT^81fY}qlan$K<|I9Et<U46eh$#FLi6ia@nFZVY&v8%=Otu(PUOhLMk(bM9OeKaW~
zpyNlfk&^rP9{c(1qzHXM`y`tT9osQMCoZAU_2R%-gB&f;j|mUF6OJtvj(x25JS$9b
zO*t3Gpr4T(-`@wYbE3WcwhXOa{=r&49Ja%P+zN)?O^|*Sp1p#`H>Ww!3o)Yn*h{AL
zZl7^Wxwuyuw5PVS5E69Ak6fk(+P<D9j*dPHGS=-N=N6&iC}r2fs1HAk1zJ`EweJaS
z8{;gr><1XDGY`zEhv*cM>|a9PG&G>J&MV-E8VE0YDA)fXFl~y*6T2xzxA*Wz+{gOa
zg7k{gsV|?>GsjLmv+OSZW0@^-7sF{q;h!el<CulGp&?Z$xn0yfxs`HeanY5e+-RLg
z!TBfjb*J85m=xpoLe&Ykg`tZBLO!9=V$(pEo!{HA=rt&V+;pALO%)C8V$8I?Yfo)@
zQ_gF=Q}8+f)&{s};#f?XRb|L<J*m<~m%3T11FoU61jNmBICe(?$6+%jmeo7bLhLGj
zpJl{cp2<++M%@X&8r;&zXBxR#BeVv$8fvR>8g_YlrxIqm?|OXyJb&6vqDbKduHOr+
z<AKI_8SuZXLac@pHkuk_`o5oV;zUJ;SUB?LU;9p5s-}+Zq0+uHVb-bQR#)(R@ZKfT
zV!2$Gj1n3aGyK+>Mg56sbZ%I%O@eax$-Qs5Wex^9D?nl0U*I3xyCcxypdlKH?OU}y
znB^-H`K|TRY&klD30+xOd1sDMR0PwamD4*(V&d8}BSG)yzQuP|PO?&~GpEC=%^P|q
z`f(DC6jq4d-&yl?--)XxZV%^HELps5QJx!ob>Dp$FFkCHdx?8ipPyoz_kMi=S?{}w
zQ+p>xE$dc)7+$)+{OQ2ZN|e2sv_G0}L^2~Uyf{6R>utfkgQ{Eu_X}eKD^3>{-+Ude
zQh$1hCt&2eqQ~g|UW!)-XL;$+G<8Ebdfwu@CeGX?yNe2s3=Q0j+BVnfsr_bn^bHM!
z%809R!#DVISRB5g6SIY8{@jh&L^xG<FE``;)no8HE=h$gBap3+<0-!q<a^_+mHu8u
z&-0M?e%x!5VZx4R)3LSf?xh<-XW0O4M13f7wksOQU;b-SQ`8HfTeJm-R)RpKk&^_|
z)wMA>ZyS^x?wTtF7Byu;EG`E1v32VNY7EAWQ;|bY*3MeFpt73rDL&w)Pn2h9G*hnN
z4Moy7e6Edt>YG`(A3xo7A7}0&1s&N*#d?=3zGe1+56LR@hoXx^{2oOx7iPpW8b0w?
z$k@TjfO$L5Y1E(hwff!lN?}nEflz1l9^I;Ui1IS^E-$2(0eJx%ESLh2S;yuQ&`Ji&
zh2Q4Z2)S=}iu)O{6sB!9kFggnroS--r`;tL%AuGP$iyJ8;<h~R=#}VS?gC_`zQ`;C
zfo~Opoj#9$3M5j~Q(&wotGzVOew9sVY(BTEz54ak_6K5A2d*ZtcQP0FbhouR^5Wg#
zh}1C8)Z`32FrBLm*i&$|BAiml&I{l0WW-#$Z4t;LDxUYFy~V4KhW7B7m%|$5;SUz}
zUI<^VmSseX+pNr0&|$XH+5f3k?`b2chpV!G`r<#$#E#2-tST8m6eCy}IdJMV2g4Dt
z7}8U;yQ#I{wlix$q4-6>C>{Wl!?o%W8zCRK0U{6=hz8k~+pk{dd(a`07Ht9%*@%y?
z!|Hl|I)dBmhcnx9Wg&Nkf*6b=c#irlK<ed5`CKf;PT)mWQVtxe`zwU4+C|EhDoy2q
zHLZe&F}~~19_{+Bn}%Z(wuxEC`qp#qxSDQeu#`u!kO5!R*=ee9==ofdS)16|Sl|Au
zSEn$TH$u~96W+8+M9fkyn9pcOn^Kf~K}M|n6W#<aGb-r(5C%km@JP?fS{=%9O`(~D
zi6yWr3wEp^&ZjUIy#A(~szE!k7Zk-z8<LsgtdpXh{ZtS+<F+dDQ?fW%=0Aa|dYruP
z%DxDQ+dNbZNx`OXt)+=YV-_JzfW`yIYudy3hV#0ynnEiWOdIU2B`_Tx>~P<H`@%S;
z6k(_BEZtw>dYj8)X9W`E*LFr90Nqq|q{?+7DZ7^S;@iTyBm0j++R7KrF&!5s{MrO1
zE}k9XjOUo8cv@)tj-}-F=NsPvK-3~&l<VxIQ+pUoZ{d4b&Twl3cTpibbA7(0Vr0mv
zlWM6(FsPodE1PnP2^sDOA9$@ZUS0~S`gAg**uuiMc*N)Ntmv$!5W|#RRhIcD4>l4z
zoN|lTFKq+L87s1-9_SM8n=+W+RxvkuF!w!Lqvr?X``nRr#$JV<y^Hr>;%zTZR=5wG
zTHVR-FA%zRZv<_wa{H^2AnP#qEb~4hIi<AI3J2@O4;{NX)vHwvbrh>K)w?4zNi$?t
zeL7Ig$E**1I+-&&`e1Pdi8#45|3mQVI?^asx?Vy3o4V>Nau1U}$NMqGnRf0Gd!-He
zXzFe)U=52oO;0Gc306k?-aJdKW+f^j&WZBGRi~EO?2J2(*F{l-i$iSoVYi-5=A<Z?
zTbGr;&eg4^CB6A_2RMEM%1XC;%=Zp20D6?#!Y=P=36%YVP<@-PlwTFU>nC7JD_UoB
z6*IMMueJ((@+`I1%4y1JJO`S~j~#ljlBQs6h`+USS#=Hf@vQEf2GAmuiCeR-n4O4t
zR{jk-MT5|y&|lfQ)<DojnoPMvu5K2$wk-bgvxO@hNJj40+EMb*%!o4S>G=kI0HTc)
zg_kNmxS>((Jf;L2A$)jQwr#`KAu$C{%lcUC?%EV{t2isxlS=@#xJ3}7#Da8^%bNq0
z%?hawqgId8rt!#OGV1to{`bA_uALBZDgiDMnYq<H0!M{<p}{6irJ2})K6joMS!V7R
z7A@}!1azo53(5ODZmTO<@?B{g7=H4)pyb5?^(l`;4wudO-K)_aeO!hv(32TNz}e7*
z&_Z<w&d_19ba(y{pbMn)LmSzDssKJFn3wgF-Z70FCewUyccfM$)+F{p7}{3adkgWh
zz^8V?I?>$a<N5$_;FjNA3%XzZm}bD!EFxLy7VH#Pzfse|C5sG`ZSQ)nyAO6oc^FeK
z#rBLvxz>1Bl*UbCeB6ZY@MU6=TT*(wxb9yB{UqM~xUXSLt@oJ;>zv%DVburFLy`uW
zL**<QycaRJpS<<k^q=Y&;aS>Sd8bkT&{XYjjALaF!cFj&SiXvQxdJ;%FHM3U>!f<|
zZQM+P;C>!`NV*z^PiEqsD(Z=L#Ls<uA?W@#*LC)!RGo-8?be~}%nH{XC~GwquN~0%
zo&q1cXQgI+prdYY+u>KdsLJ3SM`y%sxM~o#3EH*kK;xd%f8{;uux^~yI8>q5L!Rc=
znl-9u@&Rdm1k6_6X-r{TbiCJKh^P&+e#|Xi-4L)@ece>`blYhL(MUwq>&SVZF<xfD
zJ||aB{(f^qcQwGWb<R7PUe3%zQRU}$?lCv<`2MOzl}p&xjeO{0$mE7f0!|$uJdb`M
zJdsjla))wX|LwHbQJodhPR($()YVsG=UO)kzw_-0%A-B+{CS|EiJ$-C(LZN;dLp><
zfr+UxAW$EH?vY)2z*4^RA^l*+(1w#*f1TRf%<l}IUnuU$R*=$s0J^WT6eUIKMJvO#
z4;y@EeJS0`SdTU;c>!)8?u&KqrV{6KYJ<oQS(Mb+hA$cii;VF+@L;B?0|*SB+e8XM
z$6z|iPE6Yyw@9M`gf(j%$Gd=JL>(Nb@?0Ru2-VRn{#>OWb9ra2zNv~jvgEi8Rk8@q
zew=`u^@EGo`0?hhXBFj%iX=S3aeUAvLHum*OSaHAL2`#R`;Iul02OM?%IPrKZp33C
zW#wfxsaGd9#{^=y<Y?TOISc9EASzF_saCqt_4OD+?3{K0Lgp%~DrC&ZUi0YBdedO0
zPE8qBWt})23p_m2kvS}h9P#R$El!&}dB_oe-=i^Q*P>_cga#_QzGOw8Tik8QZAlDp
z^ja;#EP8GCU#(i+mRxF~JS{JbHa8SS^{=dYHdNUEd3+*&2L?2@tj&m+4E*PLTmrJS
z0Nq_u{!Te42m)~`DKHAwpAT+9bChEAJ1~mHMM@D}QjxPb`T&2_a5{lV@R@Bz#*Zr?
z$_q`k{!izs3a{P3x$71p@J;G{_R=o+sZasN%bxlUbJm6D1aA8RK9Bcm1$BW{VCQtJ
zhMyLr!X`%GtQm?E4C-p9lOSmZt2T<28|p>)7?YQ5VDy6JDPmajc^dpL5(K{JWjv~5
zvZ$#gKro$?T<R{UdE67+B^>t*?Ka=W@VwQ|6>dC49mm_Ac#ahk*M@J93jG%DHt=U?
zs@>dXRP_P*aiPmNmvs3O_Iw6tl~bHp$5yA+dmvhVa+*8{zoGg<oea-+GUH3FS07=9
ztpHNXRa9)$!_+DU@qN(sc?_gLrEF|tJC;7K++Y<K%+zO7=2hm?pYKg*ME=}Qgytr+
zfsVOZ(UkQ&+m9fCs=5h<`aI76WH1D?(Hqp+vpCoY-19|rzgGY1UP`c<u5;Q0jb8V|
ze9Uz>jjz)?A{uVZe;+*m``~U~8s1)lT*Pm+oD%_vjmvXH83MO>&9k(NlZ=L;?zmqg
zNM!`@vORrNYmPmtg_?mqgVS4CwnkpVDbPvnUCrDB`uda|1eZ%yXo{BV2=0>WnxM#}
z`<qQD25bdUC`26r#oWr-chAEO2Zzf$mLGt=-Qd#|iu+AD9+$22VI>pcjY*aB)B|f1
zwV%hUpFBbGb_$k&NwP}H8W(Wyxq~!#U-F#UUMrEj=rn0u!1H30!=p?xqE2!f>6L!X
z-FdSL7*#pMA2{enWdWu438$4QL!sn0{o1E#NR!g(f-@mAEf|<_*W8)&K0wT29zcKl
zyp=eruvuMMp8zGF@oHT!;p2yB+#58hH3ZX;W@uEZJZcw6>Oe3*PS2TxnWuu;ZNbAs
zDl5BCX2fpK%epQIy-w*>1ylQ6){y-#ebIWiP=>Qq2Fe0fv)<<><M9;8MtSi<35O(a
zh)1DZ^l$F!ihk$5+`I&(AIE7^$*vMRjoC}tYg9cX-|<R7wB5l^!gQ{`z;MGb@)Cc1
z%4Xbqj^J!(fFz+cp-ra5VR7T&mPo}HMOmMZ=N~>};k$(MIR7%jGNpEJ;3W3nVgI1<
zXua8A>P2z#fYaL9siq1CP@EL+H~(f^^ZqUn9M#_;Mx8Hpn%1GH4lF*uYPBR*kEtbI
zC&8a$)la!7V}rva)VQ~+zUst=A;7zV)5V|kY{Nn5nq;%$KNhFP(f2sD`XBQ~0^f&5
zO=i{&UN#CrQ||Gl4d3%OLJT>F^b+Q5QA^G*4ru0PrV!<(?2D1J_Y1CmFh+o%`5O%f
z$?HDOx2Jig5!nD3hh|m?o=Vz;f$L|$B9M>1iYu6hMQj-NKSPin?!ckPt$%_s&Wd2L
ze#^(By%mjZ-70Z$R+K&tOrF~y7eHJ2Q#dp7*cH4er>Y&!3M`i-MsO5Mu;MefqQ3!$
zC8wN6z^;RTc62VN;>s$epC$D3^I;g*fT*O{b?HUeB?XnMC#y=Xd*6qSnV=nx&n#u>
zGOYRQ#ODbEBNFyc+u3=l7xf9V#Ca$OgH&RJwz3gxY@mDDo0ehDSVY%0ZL26w+5-Zb
zlHJYJ@=|UXO6|4!9Gj)EZuPO5P10!dQJhU$o)oGCh$L`U9#wW%Ek*}kY!misv|~-Y
zjuU>LwJw8e;g?WjhqkyeD<*bAj`}w$ZKv1199+b4G1wlhKElO(jXnO@R8#R_;QpfA
z_l8|TSosc>oql3@XtOpWfnm^c87_N6R>4(P+-t;V8~V%rTD8SZSnbawu-X?wkY=^$
zKdF$$!dHxCf$O+Z-!3qnUcOsBET-Q|Fu0TuiVVEYYw^uQC|Q6Atl5(2@^_dwa1ozt
z=vQj5Ld9bLXg$f?CA<U`Zz&Hk|CfXbZ=tf+@8|8-PsvUA=MaYbc5xHmA$mJGe3I4?
z44%mBV($aety;T&*6D+9STQJ4cWdV>pc6PAA78dUt{)VREz;`qyzBItYh$YF`aUF*
z?fS|^%i1w$%<Eqbb=lbjQ*GJP9$JtR1r?x4ijktpmVenOxI&7l!LvAYD{RXUp>Plz
z#$xLj92i-qQuaT^zGj$<_}+HCZl$6ex)kz5K$5rMVSaC!6KgN2MRQfi)2m%i5&nH|
zb>VvA`G!D;20<axDH*|RoK3u;rx#0HU^+5H6Ub6Sdx@bp8N`}cc=aDy*P7@kUvzkT
zq-V}r6m;OXvmwmS5Nn!jO6xd<A#0~^7gC}p`)bDpsS9F!)q}aasb<py%~EpePj$P`
zo(m#Vf^GKB{G6rN*zT|-XVjfpNy4e;LtRZ>RYx({<BOSkLL7O!SL;cm2}&kBKI*ad
zcWBtyKS#G7XC)!)y*t|i19?@oV7IWbQ?Dd#sb7`zbo4WT$9aLehYOnR!wfQ%YjmlP
z4c`&zWpaOonAOZqE~uZ@Q+mO7ASkc;`rDrqv9B=sRqqf}`vp*7PeiRL4AfdZ#bxIq
z$7@HxnFxHxXuWz+(A!6fG3w@gX*=KIbPH}}8L_aK>tppZ>rF3Ncg2<-^ZL<>iU_rb
z<9>9OQnS0OQp`tsm%?pR0RdLfka<dQcIi#o`H|v<QW=fM``PY(6lx^5^XA1mPD(k5
zXV2v>WtE*3Z(jFsC8rUYWa6q(%ZsH>=~|0+HZ-n6Ma|$X_GOh~+$>oZN6i8-xgsOx
zf?o=*zHPl$EP#1=MJN9<?*g_g1u~m5B;PX)!Cl>uLA}qXZm1W;I%!vz{zL53JVkJ_
zWg7rRJ%;VV${9*Y?(R4mqi*1-)STxa0}auJWP-SlFB*d`8v5ZvgJzV3vd+wp9Iwhi
za;2IBQ}Jb~Oe;?VNdZI6>A`x=&Epd%420{-{0aiFo?qmy@86;48kBQC^&;pb4a@g1
zi6E3oObTi>ddLqu(4S*-?Jc!v)84a1Oqv^*cltf<NDQ?*R}KjZJUw=b1IvKYDNK?d
z_i_1Ljh^BK+S$1*-V%PviAPZ80Gq;*J#T?_@$ANn0^@_)+y!6s`?FeY<_`>qV3ivP
z`UO0MSu`%typ7!SpVht~y63u{A5$6p;+bS_ppA&%Ee&t){$mET-cH#%o2MtPPU%V*
zu(IReN`<SZO!%A&?YC1WaH15Qs8ppjFv+Dg8i5rzS?-iLAq>mf_tS7~zvK{Utl}eS
z5AfW}Qa}3oaQ7sX{K%@l{ZNU)+$1;UvmwJqO2GPj(Sbn6pd*6q-QHJq;Lgn`XW6Gj
z%Uk*1XeK`rz~3iE6V*6i*0>$idReFeS>pTz2b$De*!&!Zqx+&0?*^ZzQGsy3`0Ed+
z1GX0a0Vy0mdi#iI&GEwAk^2*|(8S}S1tQkW@-K0R&s=zfff+@_cKnP)AQ3aZ_7kB0
zS(jo^iv(zS5udqw?P!Pp!%@P@z`MC&TJ+4Yz{cM2%F5Cb;Nw<XUlxv4vSCskJB|`L
zhJpL|*zZ4DVo_mQw0M*M^<$A1+)haq|F`yXVFF)_d_?fgQ(NY1EmenUs6*=%zI{)&
z!{SlC<*-0oP$ypTnyyPSf#G%PIlgAu2yZ-@Oxooe3?{03l*KBRy^{A2u4G)5jtjtl
zt8C^qnj^fW;nw$#-YjhQwlsxxSn3aP+$)~4-J{d*2VXwS6W1qwOV?AQSVMu6M=ltR
zm=SY%6x|(F_@Xcsc1aSem2NPD4)e>qt(O9*CJOt6HMynh1s!HAu?1KIISn^g?>4Ax
zXE-MoJn!ceyw&5hP;Qi9$`a0}{9#u7#a+dyK`L}6y?@BpoCxKR@6=kd->LZ{LJPb1
zjqS{=!jxGjUQK5R4OoYDm&8vmxVfl$;1|#zbQhC+y|}(KpL)n(sL?@wdoXMJ!DwdH
zB+W!5$%1<rTS8&ewj%A+^IRpzD-+@U=EFUg2g2G_N(WOTcqkTJidg;qzbd@rqWJ!~
zLM6-X^`iNzfVq;HU5%_jafs2f-12Ab-e~&82it-=1)~|azlH?&D3%Bh6iTe5Jj>;h
zj*Hdc4Cf2r<-R5yKpJ8)^ol(@EkUM)I<C{zl<h&V8RbHdJ*dsaRT7}3Hnmq@^bAs6
z)-*4=XKE?a>z=be_-;_v<^90eiRG#%`~UK`|Mtg*`xvO>jdNYESAYLS5vC_V@wuPI
zMK)e@KKJ!b1`)O1BQrKj&tsSsJf16+qYhTKMdvY=2}0F0U{ymBHnfR@9p>?=^DsQ(
zpR4NkKlhKd=BY~c7&o8rmOr=XujR)+iXS>h?sG;5w9<bX0XtGI<6#7srb{<c%kz4!
zydNbT)o1b+VIT-)5OKy-j(&7dLUr@s*6zPuMWllUu5kYuLlxZX&vl+ykEIO~?v5b*
z9wDmC)H*lcm0}5G#$xw_?*jm=GDZr>sBK$7a29|Ncmp@19Q3VR^$@9bizyJXwtxZn
zy{M9qD6wN0IX4ptd*Ag^Ni$P-zIZIU)4VYTxw~$~9uRWtJ1Y?ZCpbYZ9Kq>v092*a
zD>|@#Z?S*7&A;xgi2FELTnf4uQT*DAo)owAz;)gc^sxG&<iJN0i85~GtL5pIbloBw
z_~>~;)>+5KcFb*O0ETMKgRe_&Uu&Xs1ZIflD<QXvWQ|G!3|vzVjGX+@ViLtzY)|~-
z2k!E_V_?$|IkO;&-Ub;#Luj);LqEzm#_%6*==Y22?8d~!DA`==a{3)7Tu41%c^nSF
z3^yxG@}tEO@B(Fk2GCSE5btsTRZBnMfre(|@rgd!O<)qpb<R>;ME?CkOilf&mXa8A
zR1yC1<(|Q}8)4gD5e=tRBIiF{KN-3H{Kqbf-`D@YA?v^W`ePIoX5IP!@1Ori47@6L
zj9L#HhvK&HKYsMzuJM090GB-V|IbU>(|P(*xO@^tahrA~2D{!a&D%$eg@D%4kH{ZF
z26DgDitU^??EI(4%uV{N=W_n%Q#q16MosnN*C8{XgbA3I6NvrP62Q0$AunnN=mgW^
z=hq=yi+yYX`=`#5$QREY?C+E!OAIiN@tHGxi?aZhq~xDZe6j@K#o)XM5u3H~>Pv^;
zp^O~518mgY6@cE9){Zq|Qyuf{rRyB&ZS75!%4$4~0U(g5bCg1rUwS|bt_3kC>E`;=
zYKI%Wehc2c;jq)0zcylL4we8wND&PC5Y(}jVANRxWe>B<&OQJeZhKS!V5(7Nxi&TY
z4mgN`nT}GAZM?$<h;|ks{;s308@CZ-6AKWGEaq(IO{u--QG><Mk01a6)K&odrE+%X
z9O%Mv_v@WNRKLvKpL*ev+(T{F?RCKyS+9RFe}56r3Gjf+aE(GlT~OEKzyS?~h$T|0
zR0Ve${Ap{5@X$gSj5~*F`xOSS%Zlj!Kysw{lWR!Xg{WE~TwMw3#FYckG5jh02-Hx;
zEfpk)xk1d^hM1;uwNHbGFx_dm%;s{(W9IC9FL;G6jsO;P++Ad=n5W?f)Yud_5-+QL
z-k*M^Fw9o9si*}nz})4j=+7_o<<}QVK)%rK;r=S(5d0RJG+jk`?m+tOEwvi!T|`}d
z`hWJ+?_dP4$DR{TsCNU8JAWrAZ!vYj=9Li>VM>_*r@ZPmKz@x?WaRs@)ba`|D+7Oe
z$O4nKA4FioAH9>^Km=~JV4B1yfqElf_br=oeHx%lYDym<?<FXgp3Voc)xOQMK2+@Q
zapyBOQuy;?rUEUyI(liW0cJaNo_`4*Xz|l$y}wVc|C2-EJ{IA*3?CxHIe+`fU*G1s
z=r|y(-h!55El>+DY5|LwG$VOP_;ZZt$6I=v%1KJJEKa8|G%#92SnzVG+1>F)P`vKJ
z{8fK=8lJ((tDWjs<otEMeMTX6*4(bUYb1v|h!+%$QtP0od&NEB0d=$L5Ri~9Vfk5n
z=Jr_xbQ-harKUFj%dwvjCywE7(ibm5U)wR9K{O2-2@MF``KRtCoP0${Q2)NQc)x|^
zZ4g50;07VuI@_MZIanIc4ixTP5S7<=1o_|v5)(DuJw8p86EwPFE75765l3U@+%f~c
zthPN+=d>gZ><~U02LFLCpA2BMXQp&VKc*DA6^`X4&iM^NG>ucRBBOKwL6Xkl3`Gd^
zmJafMYC--4;KFcb2zn~oN?8bU7V5x8#K$AB<Qm8kX&oIN+|~wIHvuqAHh!w}!W539
z^gYs5Ga{h$_+hiz&d49ySby`$sR1~gLie@9GD7=deiDyp)f7+`W+BQpARnYZ1ggRY
zelpLb;XDnK93+p+5wk-WBG0QKYcMoz%}9@J26Sw0{|>;9bCOHjA)2vX!}p+Xn}AU;
z6@|lH?yUC<+DX-*f4CZUTDDk=LNxsT)(^01A!u0;kd8fC{cz-N`=I4&>C2qLW~Q`7
zle+;4lxMDHt_>BQ^I?zL`HUd(z$i##2$ttivA@aw=lI@Mgx7KzuS~rXM9MjWDwfP3
zN*04vhoqJgnrF~7rSKt-p_(z_o&=|za>*X_GKT;V^$=2s!i=0kt>sIa`v|W(z+9~2
zQd_%1Pmi2%x8;Lb?F2N}s~Htl^z(*C;JW3vT?O;l2Cfvv1BCnu?@B3cXrW5BB+2<r
zm5T1e;#u%mh)pY$c;yPPe?-oxuK1pLp!<{g(#PA~pu{?AhKE7KLGf%wSkTfv2-G$Y
zxzR9Ros7*mcnB#?(4v=n9Ekg>*;u_XIV}++S^T~H^xrFye>q&*@I+Ak<iZKRHiJDj
z#`k%)=Of70X2k41{m8%AWoJdFSIB=JVE)gU9m7P-q5rPnn&e9-&?#x%6M2qJTQ%bz
zIrx4m_@#y%0!~j#;0PX0in<(Sl!`AV3JG;1VEn)hm`>v?Mjc8(ZaKntlf0SSytoJE
z@Dd!}x%A}Fsv+dNetiuY+Dec}r);`oF+*$r6CLwT83^}^d)7mMHvQ(39<t;tT5dKK
zs$~-&Bk`;eV2eZ&;`lr=-Zc8pF-=MDVhtb^DSn_ESpfPWZ?Fb*;Odbn4T+iNq_F=@
zx;mu;{M$0wbe>+A*#i}HemB7LCD}<LDU&`s;^0&7W+kRLepWmH#|b6Uncg7~cv-fp
z3(IDJA_#Jj(o1<ykKqT$<@@BdoB=)vi3DW4)`>_kzl@iPY3`CYnCLDCW`Z{7&6mD)
z!*(H%2$@VT--A9_gH7jKNPF|8=aV5P7ys(&Pyd45CaP23&IW#+&EKK^E=@aaFmLjb
z3ck1ys2X!zD5<L~oI@UAwj3w1L_3auygMw@#`n^nu7L%|0I(^HK%qqS!dt&Bpo!s1
zu(lPI5565SZd2a1%^AV(dhqZ;uGw;5nfVB4`#qSd?5WuJL>m&}ka6*y3=@+<u$g7g
zk3Lrlc0G~d4@Iw}<qlzq-C8f5ece!@XyzD9)D<FQnZ<DLU=$L8lzb(FAJ7G2%e|;6
zMmPT1$xIN@rnU2_jqQV(ikNHKHmkoEVxrwptQi!kJcsbmK(qAi0T2@Wlee5wJ|1;Z
zym3%FA`$r@Nmof^*rt$kPqU8!lH7nK&p@cYwqGS&7p98^j8e*!Yq#*NPN6qrPK759
zOu9yM54Ki4j?lS;7q}?OBg3)&u&C?t){YWoy?&i`P@I&pGwwESjqB=q1<rdNDWZ_c
zaiyy6w>iq$CxOPc=(t?58J%dn#dVze@|n*uth`u<>%1=>v58d+9$@ac5)Ef5S+9#)
z-?g0pG&45hUEUA!2%ke<#J`JXhAv=PU*<SX-0m>U3rV!4z&g`i9Sm=3U`S1_9ZjOm
zsIYDEoOneiyH1Wp+X?SH-+ZFuy@_Pc-IwUu02aiLsnEsjqOapUMKMpB1~E4c&}DiO
z7~mBs>4n|vRW^TjN4xVP5r`O%*Q8iOcT&wF68$8=7{OwW)4Uz8c{fgR;MLG0adiUf
zArz}$VCC?JUYv_|E8!{}JKNQ@I+Agrx64U5%7~kJSy>&i5wy3&km@B?3z*wtv+@R6
z$_syC*byZf7E^KkU{R)MR&t(L6La9nY})tu)^qFDo(8)>2_lAKx<!`tJX4zuaZAIt
zS?)Wvy79?3)=-JnJ0}-(3D{C=Td&?JV5-t`b6^X$S$We@_}oG<ZaB_pz9&m<I&+am
zNnxY5UcXu*r21rPwIFrF`=*31x@{c9N~dzfG?~pbK9DXOK2qmw48BBmy;R$j@D3z?
ziCnrY*BgsgDu~<jp*A{A?01S-v2?+V(=k$6wT|3@18tobX)_WPZ|Z*8+`IQ=4jvq?
zvfrAA4d2=o`MFPXX9vp>&z>ZXKTVA7i)icS+K@IHjUdwMPtcWTS!d6!46v1&<b*Wo
zPEB@<Q&86ev4W_NI431mI8zCo-5U=%E^}#u$u(n-@0Z3!ciF+yvhYSGS9`sGb-ihI
zRd7SGoMn4DO+(7S!RyzpIK+vPql0Xl`*a!%MIp_qMkiP?gV5rR<Y9|64YV17w0|g9
z?a7ZHWVz!1{)C|4M?~R9I9pcxi(FW=lvj<W&-E1_7Fe5{)m8N;{X@E&s#GOz2B0Hl
z??<)MW7^=@RSVWg9QVV-TbQP%Ownr=lem?nD0$}i=hdF_L6tyUcIP|Eas?at!P2U)
zDdR5T;N+0jD}&&qhxO`h%fZ!z1ux&VAkphPLQ&=nlbw(`j#FEuNHO2_X7cnxrv_9O
zii-G0(;2N>_bSN8x{V5Ap2EQ)qYJX@P*lk8Pftpq#tRZZ?`%LY5aW&KatMch8(LtN
zP{H>*6My^VEwN<sfgp5c>*zzftJM+B0Xg2+T~w2c=oSR=NlfNK(aM2d>^qau7cA`%
zRi<P;DU9dAGHtCmIm{GH0s*r0rZPMH5Ax7Yu~(Yd&x9D0#Ygg?+mt`hk=j{>O@P1Z
z8PXjKe6|5EZ!c%jkM?`FmNMVmV5*MvrM5DBYxb%^ed8`ox>-id4NYa?%;LNkXk@w^
znm-`8c*<X(m^(EbEf<tr0aT9brfim02`+E87?UGH6%AZ-Jew*Kx0Er&+M<%f(HKFF
zS8R$KxZ+{cD3k0cV-*#@Fz%cExd0eOOyWzw{RlPHeYpc!GZk6$<(y03muGaf-j+@}
z;7vW(;)wZ|1NiTMl;?$hg%53-=wGFWNFV;Z8TB|qF1b}1ihGgI-ZUE2L&94MQv@)j
zJ86z<rO_RKQmMP}gj@8ZOvE9ZFdhc?H46Mg>7YUYjh@D`XQIcFrYrll7TNV#r{twC
zbr6OCR!?SH9Xi6BY~iP<uiTB9h^`a*e8T9JB(W#H71{G`^T4isHmQK~>QWV@aXlCQ
zoFGKxo0APP8Gd}-4lP=*1i+?jdcH}%)wr+NX-)?73jRYs4OOZb_q*S<lXaa>Mli~?
z7E0?T(~*43PoXc6FOkFRYeOA3mc@5yZc#6Gmvx2xym$)dqPC47Q!?8EA1NW+x8#=Y
zK&W>izK+UUoybTYHXNL+&jA^D@0iSj)5+-9lPmd~Y4vEtqi#317#4qNC13V)S2j0K
zLx1Zmg``c=(=P2=bCigdE=32NyJq!X-jBbrGKM7KPd&O7wx*0O?;R4Du<JgBbxBa!
zH5WJEZ{5`KNO~}aT+7+rbg7uAeS*SW(850Vv8JvevvcZI0|ESCa_aw3g#12P46!y9
zw+_3WWBxjfL~y9b^))RaLFe8<nXYiXnb6&nw6e2FD95e+k>dVts}i7mh<Nq@Xcg;I
zWL_(Ui58_5u+>t_J5$=+h#6zB$yx<FqOsiuhwom~HqxUw)1!?t1(wLeu#qH5Vx@kB
z%C;h%lfF}y*eqb5Q7$Qhu*l3^(&t6nf)JXGQC4n6CNe;@k;^J>tCAi7xiQklo~pb^
zLWfn+*&46iIfLp#^a3_9*qV8?)}Ga>7zVK(1J)mf)M8J?hopqV1s_>Gnw@yqCP)?&
zc)R7{JH2M@FQydj{o1_7td5*GvKN<4Jx>zlL#iKxjny)$qw0C@0yx;8YZE*ifDt(E
z&_YdbKN~Us;N#;bT5E@6L6CFj+Y2|hfi2JmF$8;xC}%%?JL+f7jAMoSJ*?W828cj^
z(F~;biP`grO)c+>m4z-7XcrcuC3LTI3TvY*SkD9+F4yE|w&61r*iR&t?-XSk2bK*C
zn5WM+APgpoiXkA^_-5*a9yE_;a+ufU2E?XlN(y|*ah|uZ11sph!hXu-lZ6t*S7HLy
z+I-#;?hmw*U(KW>BvY$BM_dx?+g%LxSNkD&cdNRIx?bNAT(ToSlJODr&Ufy&-@sks
zTAqshn!(@!5?ix0GnpSwhJA;e>UF<6@~$VYH<kQ#0DOr)^5|~f5Ibsu*S5kze7~VQ
zO;ElhNl*<v1~Wi>51u=aHI}0-Ar{Ouas3YUe-8vmNq!b$!KK<r&0lL!=T@SM&tv=@
z=V&D|u!B@_#D7~>TEPgS2I?18I7LOCRYOW}sr~_eyNWD6N!bn;;>+#4J6)KJe`Rao
zb0cC3SNKDdjZWS=&v`aUZ^*5#*7sIYxNmgHXcfBj%IfIrB>)%YBXfADtpx4ZycJyE
ze5ZF0Rc{asm<gZ?YmUvRt+kC6w^-5dQq^@fj)Rh?P3(@O!c%d1GvB(PE!6RmK)r|e
z!nbEH)l1h%Vp-AchEs`4%RyVjW(Zmsw=E+XOH&l#X8lF!uYU_oUp_-c>AF_U2)@D6
z>^)r>Z+#y7kIP_uCR9~VT3DeMeeJT8>38Gngk@V5X{>Jq&yMN?q~^U)=Ulc@R$`hx
zazgi=sK{|sdn8viIW0n7gwq^MXAkW(a$fdaXVN1XX#+NYaiE+{%SfE*h+@~S*#m@$
zU_fhT)ki^<SoE3TMFmZidY#ktcSlA;BVRYf&><Fb=u~$~Q=HD?ZHhMMHo+Y;()#*=
z_rx3&1AOIw>>E%+-@vd;k@>H;n%IIW%EZklQh1X1b;<#Z`z)X!uBX)_WH!Hr+WoQ7
zF$84Qi#|3KnT5?-hmT~v3bRe+C;9x}Bb}Ec!t5zIig;bubqTT?wnlAto#$%<Xt<q+
z!*K767Hu-<dsl`?)6uf|0YJ~<-RLf9hL~tPXx=iqs6~*->wcInmLJ8TPgRcpX2c6T
z=tpI^@>B$Qj}eG1)FSTBmw^~(L1A*;1%KIe)fs}bpm-XDC7UkJmzdQ(VU=t$nDkb*
zmqr+hQ{bh1W4ZD?8#5L0e&M>nHDWG4uUtot4D^!KcLCQBq}y~oH<k=PrCICEY-Ym&
z)bANz@z@uPXpt0*Zq;gYUf&)W32-^jMu_BArEVFdMXG|{{=000u3TJ<UfPc%AXGLb
zl(6;tT_kNs>?Og*x!lnjt5wk(eF<vJ^n(pRVO(>*dFM5xQaY4#|6$ks?rC*;V*Z$-
z#rj9vU;Lx+<f)?tT!HJ>N>#QU>;)7!=d_gs^WJ|qV_8k|WPb7J_+4vH$&~}3g6>qG
zq(Xds+N)j}A^53v!4-}xO<&hi&=0X!mUCx!oEW?NZ^7?xa7qwcm<YUwvvz#K>`CCU
zC)hTm4`Z>hy|72Uus@JyPC+vg3#k|o+gmahT^wiLu5p9XaRc9BKJBx3+Gj}wBJ9@K
ztILgJbHvD@GZZ#XQRsy(YIXCSjhW3sW42bkYPYY}U|H0C**(DZYEg$GfWK^~U>u{a
z-(#T6mn^sceJpcck{;*A6bRjixFL$~neU|?7bjc#C7;>cPi@!b<O+_x2PdK(^$3K;
z*!Yxt(C4(qNw~RKjwZ_K$I1D?dOg?a!@!L}3t=Yi>~=ENkLLm9soqJdeN_;2pl#TC
z+|{p^Co(%RE}Kc5`LF|7-AQ7brw~rwnT=?E82+sNvV#?qg<0$8E{fp##HQ)&t_^GW
zruqG})N+KIDiPPLr4>jLBC~p#@?WQ-e|v!e^i*tgbidoKF1Vhmgcw4wdaK2XPVHC_
z4yP!s_&K4V;4x3OPUOcvE*N2Kks_4+NX6_H>J2xgO&y*>;?zV`mgCJD?wxevZ{iO~
zm2g%s4s$BibEwSVr(6Q#^6IMdkyS0VJ*RdKNd)Q)L(5V%@i#!4q2x9m4mKIm%g&@h
z4aE6=eP-~Ho6zajTL<WcQB(_FRyCr+YhZo1gm0P7>HYX?3i^4V!KEE}61wQ;Yf!+C
zwJy3(I770E)kvginlwJzSpx_Ktpdvx`m2L^;)#%}YM~elrT||^u{<O;r@EYuPCE+%
z5D(v6{_;?K{>u`1<S{u6^_^yof`hMdzE^XYFfAN|xragmp5<*k^>)J-J_5n+%-z)4
zkI*8g97gMPCa7O^@WGWw*MJ$)#lA`euxhYTv@%b5%Jj!u(T{&+q(f_}36zi)n5~5l
zsrVdda&i&jpEFT{^Cu_V{P(l|52KZN5@-qYKV>fd)sN+id|Vm7AeM88Aj5V!)VxCI
z*lxHmei?Rjb}HjGo;iMt(r2cg#aXuzQF*w1sX1I-Y=)kE9|E0Si_>lwJJ28+z85CU
zet8O>WVblR!J&+NEKZq5-?MUn1(culLE^yQ3l~f83x%MX5h>3@Z4HmD^H{9NG0VLd
zS;^gWA_mg}LD_}Qr5EKZbeLKohCr%7X{3qM&CK#!?<A4orJ`;^975<uWf-=mpraZ%
zN#k#y3uCp-WhYkj=V+O;jH?zJY?1gMpOJ9TWP?1!bA<k<rRv)-QtNpu?>>_qpd(7w
z)#vo@rI^|P&m4#1a`BS0&YNe2vd#$=(7^NZ6=z|lweq1b;d5Eq2z_b16BcN)7+NOj
zn0F(RsPa=_%#sTb`>`lkc}qTvCC;|vU|kJ&R)6sDt9yxuG-Y999dqI+Kd3GG0$C-e
z_0HRwz{-av?xrK5^Rmek&=wYOik>QT^ngBD*X;N0WimvpVt6al@4dY*m(oF~lN|_2
z_$T0W@J<C?yK5UO9838vod<|8r*R0#`cmwV;)=6O=H6J`Ot?a<ZC$+v9FwPY=#<`8
zP{;N~V2%OD=t*EYbob>65bRZ{z+SKXx%&sNzo|a!pG@zkdXN6R`_aLUELYPurPH`)
zKJ>IDSIgn}(urASI9aU9VD3}NX!ekyuZhEo@@@j#BxBwyO)UYsN4!TrHd?kHR1Jw?
zoOY(}p?I^Uv7fp<N=L{+(Y7LNCVMxAa*%8$QdRd}@#WO;?X%8oFUm9>cN~M&pnjq~
zxYPcpgnJ6Lsn~N!IVSx(Y!tB{CwA$Fy+@87PZrq+c?)Q;n{eWgY4m1!o;W=Sg_H7&
z*VsEo#w!=IS;nf)$Ds8&k9aMrT(+1-HbOmSy;wHbGdb?Al3q6rU8rK!WG5bv#tv+i
zP@C25HMmGT*N=4BS-q~s@2rlEX`}RjA?}f(8$YkxO>{(M$;!@2Ob{JdZ1Cy}QSsPG
zp@M8|t;sz^0Z;XA3h~z`Xw7R!yZ2Om&^p(4qD@$}8C%r&nW}~Wz=!d6H)!w*5vx5r
z@q=h=jaDDlS!%(yTKcG-fhCFOQ`9R6(Y>6cLt7rd(GcS2v#I$B`!)&Y0!moyk>#hW
zc~=Iu!K8Y)CBB@8extKvHksaTBIMa6{$9ig3E#ZHc0#uDRj!!fq%P|+^!$PhS@&f?
zBp&;D+USI+upOJ`tZqklNF3Dk+Y8CJ>}+I;7zV~b?E`OrLNwEK&ZXT)IlZ7c-DI&-
z8=sb#PXbNCuJ^b+v~@T|XX%441{ZNtgJW|81fjUS0vO2m0|1UNP8F8;EQs3bvo^Fj
zQtu2(*w1=t+9|{%&Y$Uy4ytA*?qwel6xJQP5Q8)ak&PCgw99d!mgRc9?Y9SHBvXyz
zW%_g;X!F^&+1scW*%Vk$Xlr}uxDtZxF3*F{kT8&0ww^3D-WSDl>{%l387&3~;;na&
zCJHhRq=$;{xb!=bahXNvKnKlcx$}oLhpBcr{vpzj|7;-~R&pI}Uyy*vZ_ZL{PPrtx
z$uEXQm-(^gmID_f0#UWFdHH$uGwFnv*@5AQ=$tCq#oX;_`>@+)A&$pQ_}=B)4a^`-
z|L>N`j^>noaQRGV18L#1`{VxB0zkjQZfzAqHQG2-mz9@w5os7b?+^Y*XJ`jS;vx26
ztZTI#29gZH{v_|%*$-K5k`l}B-`Vi+(`VJTGQ@_ieVk0wpCBGJ03~7}ze4ysF28z=
zUPjCrUyl%@rEc5d2v{~tzjOq#j2NNO(x}Y>Cm8FdhiuDy&v%7En<SpZH;a-Wp>0`v
zs&)0SFXxLH;^@NA5a-$i&w}Ms6_L*g{8LU&>fzDGcZi=$M_*Qm4?|-G(bQT9KGV)R
zTFL34OL#+{aT_Kq+7A`nuJ6bf7v$gzmG#B%PUs|`_YbE-%#JNFNy{U9o@-1yNy*W0
zyzx~pf<%?&%mUqP1j${4OIVFBK+=P8v?YM<^KHkDN88XX6(@^i6yKhuc%Q7U;&pY$
znovXPM@&@R9U3Jp40=?U^-WPC9D+*{X)DegXO!r<n*#2Q*y+3Gp~VZKgr*Bn+H^0L
zPNo%n3o`0M$$_k-9UG(f#4i=P4fH<25hAG<pvdL@W|!x9*)!K$2*fIx+p3p-Sl`uq
zuCKe_?fmt%ct|nbK$?Yum-5bFB=g~$w1ivD*ZGC<<E+7eK|g=$<yL5BS2|s1Pi2HD
z*Hwdusqv8s3wO-Z(OrQwl>W4!Y%;@Ggn!RJ%J?4sQPX>4LxYGr5(6sND4w+|>=2pZ
z`^scz?WH^D(r{sYa`;kt*E=5{Lnbkei%wdi_2<9{S)g*gIy+&+ec9sK*$!$fbYHr^
zXRWtgUVeqp76-(OQw3dMKpo;WZ#ENK&VYBnaHVQ@GSFs3m9ps81GXqjnW8KQ1?j1b
zF(Uj23>nbjFgaSdvLRNITAsHhk74b9yZGsqyN1bQ1}py80uJcU=0yZRQsxkj^QV3~
zt_Vu~tfbkmIk!Ei#4XePFlAeq@=#?=yH6+Z$)!3=IZBA`{z(*tNfDi1x1F=`=J6X>
z?{qfK7KPFt2t|fJViQbxpJA~&T9vXZ?~s+(tfuJkB`-Ow@|m=pnrs-!HRfz-VFqe%
zaa5)fjacOE1`(Kj^3f0qQcNcCg>xti*XelRTYl-rDUfPV-je>rKr|_j9awj(=hVR!
z{||$ShFoO09+BbdQjWPhS%OVYQW0v)&Q)1{e48uN2j9w49a~BWCIWZquJ8++70YhD
zt$I(l$_q`cWp9Ezexe=0-_E|M1hM3$AECGDxaoxo#kO>&u9RGg!AED$TryLW%!@Je
zwUFgzCDODqGAz8>mM6olkg7v3M;eo&*%4l>IM-B3#Iq%4r)sIvJ(wh99KOQC-=T7!
z^WalSMvVEEh9OfzlS`VzP0>#>=DCBp#hQ&~S2`|dMOw`!%)VJtJm`$gDhSsxJ5u{#
ze>X!hE`S85ia~?<>wt$#^AGu>dNq%4__3R^eVSxOYMMOe(hwuAt|p!RL-XKoaErgX
zSXX655-4UK*g=P?1mr~$lNn2m`K4<AaOD2RkR`yNkT>BR?}*R8fCwH79B-yg6ure%
zbnSPY^@nAIYibGkq!#*zso=lfcP73(iInppk|oN{z~qN^AKaHLAR3i|6UI*kv^dbB
zlHIxSYx(`I@BU+7PBZYB=*M1CHcjk*xe?C_lsB%yUl^?fr3A%1-71UN2k8ZfF)ldS
zD*hQk_WOFBnh0R3*QrbMUsL8Ux5uH&TTJP`OTAP6|2TW=xG49o?OPEPK}8WnMTQWN
zMiG!0TBN%XkPxLqau@+o5d{fBT9A<LZcz|W>F(}EQptCnZuh>g{akx{KhOJ*A5muJ
zob&wETI+ip>o|Sb2S`Gf4q)&824VQ?24;u=5o7-HYzB}&|8V^N%?@xFe*!RuFdt9Z
z|Gbtzuk;`9%wHcMY!1`8=i>NH=KuY%|JOT}z&)5Uqy`4=w^#QMSJHA3xqY=fAc*^i
z|NXBI`rG>pvmJi=s6_Mg|Bw5We-vJtaZMgXUVnSb|C>7=AOp{b?HtE{-lvwl<3CZ5
zdkyC!VaK(En)^a;lx5BkSTdMzg4tUSD2DDMWV=s6^wj@43heRV@^nR_oPPag=@}1b
z7}16RT0L=iWI8RUcqam^I`|^m+VOW`h9G9XeZV$d@|1_9LXI3)rz%4CFx)~=h76z>
z1v2{OHj?6}W*|nlgoaRYNXMf>Kq$C7Z=u*iJ|M1mNQo99E|ikBEEYSulyGnPUL>7X
z#gAPeg<UaE>Pl2Ti@%TGl@HGLXrrOuJ(@vex-D^`@{j&;cU!J{K!ni`5t=@f-Hmk}
zF4;9)+l2hk&DJllA)8&k^)h)LLm(ETjsu0gAXtX<T5AE{;h*o$dfN!W%q9pj`VpH%
z5QF&$Kq!G=1D)E&GQsm!cI&WKXTzk8=z4_UT?v|->$PkUHg3X#5n*xf!=uB4@5_Aq
z9E+AUqfS1AZ8AQ46OuVj4_D#W^8u==uk|IruOfR*JgPvcTnlH=7#&90=ttWO)akzs
z4&e+Kwq4TR5B~TC4=Z_nD5?6;Fku2xueic0Mc-Nl@X!6T@;tGF&`ERyEw6>&Av*$z
zy%*8bh}~1aSz3k#_@NdeJkSK)3PP|UeRT_=!RV**2xiOJ$AgxBdiJg22xVF;RN&j;
zik}WdywV%iSB?PQlqpm*<-b$v&s5Z{zh_PX<^vWaTxn;abQMkk2@m?w>!Kf;D7%(8
zP!0Y~l-MEs|5*C|{#X5nbdR7fF$HG(<j?;Q@O$+1A>d?zh=JV64Ww}4XUOvO(aF*A
z*o*iU;EvXM3;JN3iG%>WAAg@jKJSVb-jbD_YV>(J?FcY`vjji6Gy`15HZ5EvVZ77q
zzJIpbnE9cjlR{&kqto)8_(iZ~+rI|_r5k`_9X_|Kqv3mOEq=t~9p7)3fn-UTc;cJ`
zzt))ia~}7EI4|Q`WPG!#=|YkZT}cX*WpMJ>&iNr%6wcenDatgHh#OI`Xw{dI7}=>-
zm6u39B*r|G`c2<$UGYiW?SFf`fLv0;3>>skH2GC@g`v56(E$jWNOTI+UL7`}LemfA
z!OlP(!atD%DzPMBoap%o##0)?HX)<$R+goU{HRW-*1d4=<|{YAoDKsS`eH;glOwRo
z$qgB-en@s$Adb1PNl-EN8Pp+L#;PLarx^$-i~bXiT7cht85=++o%Wi9?FY;bpV5#z
z&={TkhSeKTK<ID-g4GA|$DfoRn3;Xx20CegiId)?a%cmd!1ZT$qQQ^JaHb!`Q5;>t
zt}8I9iV#!7d-d0c5Woh*s9+gTC()U<Qc}6Uk!6W6lrZZ#Z_D*MX!zHGoiS;hurD86
zwUIjv?xI}oxj2Cxl4Ss4IQ{BCLw*-o;3l%c?7k27JMGp?lE5`k*7z8BZJ<0W1xFqo
z8B^NIge8$jo)3V?n)t0unu>9UBHq;fuc&tDwV=-Nf^%L9deM0sC=5G~pKNzFVQYlA
z*;Y^%nk8%k&ejV&mx@C0Vh9pCj-ypS)||WD&;|L2pG{*uj-j5L2X7}}1YwOpj1cen
zu;fbC@NG0e7^+tK_Fha$qJx@wSUKI-ph-wZYcWL%n%<30dP^phai+5k+HTw_V0S5e
zj@d_C5unt3uUTO0O$B5*xRLIA*`EyKf23e&&jDF#WMTEHP&ea%!Zk~>>}z3x(r7Q-
zH1;xJ)D{p%Z$7Ci@c`+l{3k_Mmp#|(l#o?$!}k&|hu<VnZR2jRPOH?oN*E=#v@c@r
zuo_ZY7E;a%LU)w#@z&0@ap-@N4d*ecXDMCQ*+^*EG>SC(&U_0Po`jCZPya=8?d5fH
zsnM~G+eykyh8ZcUH-7y4sHSQ`!?13={_f}>(@Z$@%y%$rwL<WtJ=jN-US7Ty^wbPv
z3N^MDRk`KZuR*63Dfj742aLp955I?225U9`pNYLBg&jcrAR(f!M?w7Q5-ev*5V8aE
zDN4wC7wKv-Ym_gDmie6hl_{+U*>Q*o98}wO5IIkvi4{UjEe1%=0PO6$93nXjcin&S
zi@BF*;;9Q9pP=2bZ6q`il*4%s)u(>w{&NuzLtxH~B7Vu(V9Cw+ZW_evD9@*7<v5il
z7?c{3rpGnnF8k2qJ|AiID*MtF*$e@9r4)Z+7XOqAX@5vEDLKLkWE3#B+i|5Knuf>u
z0W!7|7lpO{lhS59?LI;hlaLvQtK})YD5SL^b$`+LNo1oO=tm_RZkrKd#H(ej!lD8u
znLJ6X^a&>0X{eT%BTQkwR3pWUx3mEP!fMloE)e~O-MlXTZi!Q(lk$ZOouXVIZl<C8
z#48DUtwj{w=ML=^fT{Wb_%)0$(#F!>-Gh~6e4u4lW7tPpdz|)>@{C>Mx0BpGRPVQd
zyL@RUIUn5>3<xK>&?xr{eIB&^sX<Voo@fhxKb|psgzh#hJkRiOp7P+k1Om7%Ae+n&
z?Sqah6+7(iKYT2~qtLjMldriw*o5V7|5%8e_Vt2OlI67DZ1kASqU)y$=uH=whAZ85
zwm9$SJNl%2m)>RMCZ=a1HSNc^&c7^=TV(2=y7lzFp<qwVls0?8mgMV%y<77kH(e3)
z!2I^RmvWaohGMwB!D8~lQ2uxw*NuGZ<Rxe`UPW6gxVuuF=O2>k(<2eND3NjvFWP>}
z*?wa;AX~va<YsK{Tf|4T?`hY)7e6?gH~f$1Pko^W@RiInp7_WR)PBC*P`K!-WG+|w
z`xM|2{O|59bbrcITH<hb#oSl?D?V08dqS_MtHsmroipbb-eOhL`sxe1w3)xxjf_-9
zj~G^5z1I*n68(N`7_5LObV$zbirLy^Jt^5JKFwU*LuCQG`Y5zMIlgrAN--66u`_kC
zQ_9HWHzQ-IJ@RC>lU+(T6Z@`xp$)<%!{Cu+;*zO+5-i52Nv2%ty!u@sb%b)(?#>Oz
zg`R9s3F2+NCqcla&!ceec)6$kR;S|}(im-|DhHZn^mU9@1&&muj#O1<L-N1&{XB1u
zy^^*~&K(<e$B*KHz!L=vEDr^ZIU^pYIlgSpK8rRAE+UH>{Od_M-~6{e3fOr<QJ=4v
zDlHwQYajYe5qtKwy+8Jq?t>^n^<0H5=?gl^WIE4D1|oG|Jn<nntYEu$e5rsn9l&JZ
zl`MIRHJOX`uZ9TBCCAfr!_vc3?DcLWu4)8DSPiiZMcpmOWt@nSu{7=t&WP5ispM@c
zepx#wC9^k?1)B<ctrw}?magS-3UM~8Eo<G6BGs`c9<bBOQ0r1{!ZWj#YvtGZOgtT0
zJ1F2A(nY=@w>DT)(%PhxOs>=M=5}RQ;?pNzOIDb`?~4khU47RnqG*YNeHZCMwkNI8
z(STOy9<S<XyBM8+g%?2}TlI9E!`<F3lt0I~0S${whOTo^SSAt6Sc@p@yVC?m#^OTr
zL3`T=vwPxwjwx;Of{cBcxsBZsG*%dMmBRGQ0tuFqiNKF|k3Ncr+E&PH8VuSkXMe0P
zsQp$7NcUPQJ{!|Xwt$le5_m}{BRJQ|Hw_!?yUD`ey|2m0Rp$Q_fi$PDyj~ZdHtn@3
zxTkK@obs_ksCJl{?a<50FKlp99XXJsE>7nBE`;;KF>XU1`)+m3BJFuWVtCHD*6)rv
zzK_bk2K*VA%^b!Z$dg%73_H1XQ&jl4Gp%U*u?71;n`VKT8vifEWOsu;*%G%WpCrC?
z*VB#ow&YBLf5TD=Gr^@?2NDO?YyR7@j_@n|#Ox_2QOcZHem$8_S41X|wnxvN5J=vo
z7V-QEMM24xg#bkaVFi^^>~^&L(j>Dt|2bXz){*GwF^M1Ru2DWYpDq(~4guOJg7eWL
zP*YCy75Mn_c%$04%G#aZ_0t`GcT=r^K4<gvXQg!-gAnYx`C2kI)0><5RBo2pmkJT_
z%%w#R7U*26G!YD<y{w$zDEpgpUwO4tgoqzYzPoK6n5<D)0*VH_-Ur+-VzEqwOF8V=
zv~2U<r+ri(GjA&QZqSFv%di%xlCYY|rqLZYqciLp25kIXx8J9uszpQ8{4z;2Q<&2C
z<#t<-F6?MYLCqK)DyhEqQQ+gz<F&agoov1-*qbjcr3eqVTUvD$wmUd<Vp<&&+&Y-i
zYgeb0dLBY;<JgOvd0#3xk8XKi@GM@c$dCJ=Xf%a#%k!_>lN?iTq<ry_DdFB$YuvU{
zimUIGQ>U~2<Zcg;%<&(31ulhdavxXWTHYW+=O3k&GJj+tcJQ4l%3IUCZNv7|-%s6t
z3oa-?x1(Q|lW_0XzMMpCvMH7api)Ta6+RUS+KKjKV~EL@T;=}mw9qjkezJ-S1tsBM
z1E%?0o!R<sYtE65B5>AYaU$3A0&d2h&cqUE1kQTwZ4p>3|M(UW>l>_)uHKL>`n3rk
zbF|FNRO|*#7WWO63&bv;)AXb3a&O`ngXm`3n>WymmZY`#i;*6mL((_NqBK3~7+hvz
zhg0O-EE+MdJ6I!wyDwuD8~mJ<Z`IwzXNZmXM1+Y;vi}j+T=u|EeNLuARG27|g16o%
z+^egR^!`_@F_*PbaG$dId)(Wr+MPrEwA;Z~(!m)z*Q}Em=j)TCa<NqRf_L!bu*;?j
z$TQnI-E8Y;nyDrOMu!lBguZyzHW^v4Xs@g5pqH<kLb-w8oF83pp6TU9v^u5EXgV0E
zf5(nWA?&TI{7wXV9rJv;cxqOluYR&3Puc7q<zlj%s+qaK4_z*lF0T+R?(1p&<L}y~
z4gy4m6xy+g7R=4u$P-$gy7aH{-XUB*KVf#Y#e<yCzo;&svprU2iw)KL;aEbm<SQst
z)2>)fi7@LHhODI*N^4fA2Q35UdRpOX>RL5Al7?U@evPasc(7it_`ZmphfhfzO(fJD
z@`@tgJGMS438+FpjVnoR9;gq+FsQ6u%|;@@o*0oQ4K>fFXuP?Xl___hl$0b1^n;%#
zaA5am1eUlO&}dV}exW!-qG55-2RXJU?3SP1O1uZ6t{X(xFAayX*+Iw~iB-M6_1+i(
zs<zc5?Z{~R!2J1&U6%3G<3#65eqV{wk4v8HexBIdviCL6TP+QaBOb?YVWsSvupjVJ
zTiJ9+?k|~M%X*&Dfao?+OPsk*ufVI$@9P3Ge-(1W;po+h?yygwG5)b#c?v=V|B{xj
zyVVhQP*Z*Ae2X_4OTiuaAto6n)~pO^#eRfDQD$tfz2~C{#hC?DILsow_PB_wZ|A)E
z>TI&1SzF<u&JsGnU>1F4LyfOgLdh;mcM5gbt~;J)s-)`znQCoiIoCDQNFUFJ3eOtb
zCg$#R%PY&XUyWcczc2pY18&&Xc0?9k(Uyp{_D>IaIP9CE_Dh)vwiGWcjzQszbL(ke
zV!F@ZpLE)m_M%#ZKU+`H;Uh=GytJAJ>w~-_dL?`oB-*0m{Qej^>*Q6{4<jvTqV5_&
zhls!*Lbn)<&0RyIiX}PwEFZ)_FQETlFkTioeN)tj_Wep7Mcs<&C29o5hA!RCUu+5$
z+_#wJW$24jniTEw6%-bORzGeLND8k(<kthCl_?x-qf2h2F&HY5(<uhd)Cr&p&M-l>
zx@zFI&aVD@=$qEDH-~hICpbmN-2-fXzpLBG2ai|cn@3z}a8<Db?Yf;V#0F2iME&BZ
zNtdj~U)jgN?0%YHNcGAH8&?5gz*TP(ga>2DlpP8eq!$(5((E)S40XG+S<XCdQpozn
z0cN+r6^|$jUAym)rn%l5oyk@CsSDU>?2va{--LiTimoT{&<c2;TNP=)*afBA*)_#b
z()!ewPb-n;=`|d)<l>EM_EDJWMX*H;n<f_A0tk)MIzkF^1J{??=6d56r5B~UrnR3w
zg6c*Q$lmw}HfRmALTGy)C?qRQ%*9sJd%HNB2R+wbv!4vsD1@w7_rThHk|k;3WDJvt
zTXeFJu2TGtiPD~AQ9|-s=6Rxm3g&5vGXOqxMa0zxxcs@d{zI}dh7TT5=MNjQUzsZo
z+=#uFftRNh^%DL<uFBE*Cgjw5jYQpwih5<Ve6PDX_M}Ltvv`PiJ{z?H6w+Ut7h8Ts
zs)6CvCnOH8zWvqbv-uc^1kByPS5wMHZ}-W-$<}H8vmLnV1(V7>ePlKclUm)89h(DZ
zzgz5n;8nne;a%_>k9F*3#$C60>Ikw&=tRXy6du0l(pyk|N2+1cz_fn@|M}?^_nC`l
zaOf4x-M&msN~zk(b`#%4Aym{4S18N}m#fk@li0r2mx-pN6Mmgtq#UTPVAmifBa!75
z{I!!Am&Ghv!%QJ?I2y=T44cwAbpBWXvneUP{+68Hy(^d$G|?Qp)T(@!QZ!wV6Kx$l
zosF)`XrwSGXBi+lD5h|wG1cX@L{q+rV%+La(u@R1GnCmRMnmrs9jfC!gLQcX89eNb
zv0Iz1G}}rA?Q#QPmn<2f-LWe<mo}~W6h68jF`8TD+;0}(2&xfET**IU+x0^@pkcFI
z$$oQk>uyFt7Xd0Hdu$D4b*?LKFMX^zV)+%xF&cn;^wd&pT4P01!*_sVs)02m(+POJ
zqfN$wM|R{H5>q>ejc^+en5)>d-t2*q)IX5MRqc7n0J2<!V`@J^r;jxkMxl{|g~|eg
zjif^!a&rWAvL8@f^lMw2Q3IR9I)SAA5K4Z%I7x;yNwD#d;Vv61Af&nMg<!Xa7Y1!c
z6>YxBtgJMP80j#U;L~<Z1D7q%$5{!pD#5QLDSHG}!!&3W<PPKcsIN@Igx=T3UcoO5
z2ZxH)QsXNglFc98I(-36>sS+)dz=V!sjj~!mb8BMnGf|7ub512dTicvOmxO|>>_oC
z(T)J8&8B)en_+aB$@NSO+tCJBiHvZVT&vMgr50=e?M;FZnbegczFke&O+PMOxb7Fb
z8L}C;d4gz=yj!x@`m)qGFT^xY5hBu=XJmk>BSacof--TWWR0UNImu3ymsfY$^nuda
z<&>=GpWxC!wQZ!tIbcdgsE6uTjh%8&W{l3Ne$o!g^tb!XQFn$1XZ3{eUy__-r)`>T
zlv>)7AH`gtj`?J^x4l7#y2tSlB$aBWE@9m(U`eRX+CC#kWPRZ$kR?jDWkZ5iH6%H5
zZ}_s~wn-)*Vv4-5m}|aOPJ*_uE{0^_*HyZOAcous$ypE{oq{aILy~Wpu-e>dhgbNA
z)tVvEg8~_40*rwjLrtXFEQIF714mSP_aJp@?7QmY;`wL{dcF|iZRwFBf+q`y1oIc3
z=mvWdj}Q(!`BH8Co?tu9V-gLi8o+U_KvRqJHKD<C9Y}$@RbS%W&1||e>0TlymtwOP
zxN;+rei=%tuy$hANWJDh-g^2-99#@<P3z=e!DTUwmXSq+I2dp1#QA(~b?GH>wo3?`
zCh%nPoM~5KkThNP>&2O70j^u4jt{cp{l)u(`)>QM$WHq6j6E+7!7!NXs>>4pD7oJs
zK+3g@@M#se>X&4WJ2=5g^ZABCM-~}(nSZL^y`5M-J{pHjXIfVVGdJ3ePCFJ+IC*KS
zC`2;;XN~OtJhJ+pg(49TfAp`Ci#;<A@hMrb1rtF9LlRx1iyMPQOr5HbKCeikFBiKS
zo#bDB<k+c~W1La<D|#^|T`RDi-Td|eB-!yimsxpUGmOIK{31GG0y71mvgNq1q+SW_
zseqy@$JzL3=x~s}jO+tYKC#3V9g87aT@9NU#o9%#055mTHmbZn3d(@SEFB6PVyKRZ
z3<XhA%%;qh<M(JoW&va((Q$TeYqPwDh2zXm;T5{WvKS-=5AevPglM@hLfV$3>f{9u
zTIbK7ZD6r@2AmN6E~M?emqSget8+t!VuN59nAJmi{?ek1W5Fk4nVVTbd<q8`d{t7*
zpieS;Yb#j>0Avv%TDJ>X?C-zRd`9j3Vza15Rk@`Yg34v|d}*%+7Nr9icdzDduZwr_
z(pTQpO><qz?Z``XiHP>KZ{mLUqjv`pWf4NeJ;nlyFGhj6KEJ58c>w(86@O9}RdV2c
zq8N7HK!QD9jsm)pD%sODyAi*UJ~27UBSzOBNX2U~5~~(}KYfCacdA#V?G+rUl{FHK
zu(0t>m3Ybe+esP<V>%nW{pc={Uo*V9&Om?r)&0Ekj%3TE`M@IeUUeGW4!uM{M|HEV
z%=}YV3WF{&XYdAF$Y3*`77{0+Y(5+oag30U2<@Y!741gZKsG$dT7oEVn<8e*WoDsG
z0P`d0io)K}6`PN*a=i5X7v+{GU8s89;TVOaen#cq*n7dMf9Dgd`l-|30W^LJDQF={
z{j>_6sEIQ+K&G6$m`3~M9OnUuzO!Sv*rdeMxukUeFT@s<*#BX_cn*_^hZFKHK(?5R
ztS-!bf{P3{%pIS5tL^?0UH!n>=qN~_6hfo#5SRJKJOF-7QNGw9!nHtJrEH1b2On8I
zmT}5wjh@G+s_^Va71I0hsHYxtB7zE;jpCRV7(Q(LtDi5+S=!o$lw7&7b~<|MG1U$%
zBn?75kv@bC6_eqF7o?ldIjn-bzaLKUhS*`3c~-lRPd0!S13yFpJEf;3pV@O;%5*30
zvrU+<nQdrnI6oIB7jUs(r~{_YnC;eeGU8i)W3z^r`JtH52u)u{pVrUWo_YQgQ;Jn;
z6vzxG<Q^+qFXt(pX61*RO@pU&&QPw`UKJksl%XgVh@Qq8HAGSyNqEf&i(a=(w}Vzs
z*NI%smONW^&_NkhrPa>QpIAuA2J6fiot3@nQ9@RoyXo@$uHCZj6zWv{s>7%#9n#Eo
z-1&1R=e=AbG6wOo2eF}u0~X?wauZvxJ`yK|*ew_qPCbH}TfKcRBDmK{f7Qga3kSyJ
zg^v}A5rmiPw<l5Hq4(-mr1Y?jf$`mlruGsn?n`q|AB(Yp&8z*k(J5{jL~J_Vo6g`%
zP`4psuGsoToVE4nKy`e{X{IKPMzsush<UcQpT2HC8=V&}>=+ZVnfS~{Exqs{v>WdW
zVPn4SAY!W{)BfWNCT|U$lGLVDnBmYNb|lj4povxe6S})}VQ5qqT_dvA?%9&KWqu%O
zQ7hgtH8j!JaEB{hsbAhndOo36T*`3+g15p8K}bX!S-`x-eIJt0E?Rj^8N9iyWDSvC
z$?VKogo|gFX4`ahtl5-(cf2-@7Y($<bYn=j0x}BwzUt`Qsg+z-dEaBYY)f5YnprB*
z$>WDA6kNL9cg0~CFRu_JAV`M#j+w-n9h1C02Cl>xK}%OrQfoe8lv26{s69%?9z?XU
zc&m~Zsu+dEiRM`<53I|k4iu>~L!r}<osP-|xng0;a6rV9r3<|S{Z=e@K53t29&`WQ
z*;Whyu?59))nA#2866zr#F;0VUKw0l`=oe++uxZDUV_$+(v5CalCTb}@Q=428@CUn
zl8K9@9fX}v>1l|L%s>&8T=LJfN#!RJ)f40Tf!Lhj$+{8-k+DB6hxSZ71eCNb_1@Eg
zrIEa)MYLt`Wg7(ed!c}B2x4t#68e%Gw01-?N+c6QwsEJGEt6l<76v6t;&8tBMtiqI
z!sdA5Cnb<%Gtx<~tyEa1w$|AnZ~CORD2e-=ctbLV@o>T?&s27HI#J<k9d*tHcU*(0
zBpk}NG8uYaN#3L>l}<^k6T>0dI!$Joqf8#mo)d;EOq?*0@Q!jRoP|J_Zh+XSw-DEy
zSX`Rna#;iY)&~>l_$|#{)>vS1>v6ZAye^jnUG3`;jRi9);xKP>?k|Zodg+5=@sOlC
zU{u&-wS93jow!mO-G7GhjD-{fok;RNrw~-*%6M*oMp->1nU!7YHTvOqn|5?oNWQ!k
zB#@PK!Zb2_G0__L`4r*|GE_p<&)}p!<PeI`aw@1bi8k4LeOk^awLsiXR`m9kp&4p_
zaG>3&^o8-XZ;xXs&hdCUe04B^^=S}LOSE?Sj>&VRyN4J+N`2SK_?Vl4coAS5Lu3}d
zR&)0wB`e|r8QkO_39e#ef{U$v^>kD9^e^zZx_|0v&|8XI-!3fbF8#Q%bQ@va3hcn$
z*k_G9B`Mi8zt@d+(R95>xAwBBdmbF_lDgvE@=SbXpO$a40zVrm&aX_d^opl1*h1+b
znd*ley~x8=qEocVU3OAXdXkzO`}Z32Ash-ZB*kTup#E3sDS+V?rP~Iz2Bpo$D<GL(
zCm3-{rPF1Z&%Pt7n$5)+(;C*F)teo!9Lgf9-Bt;_D&a5}mU&~ZDXyuLtb+MWQ_AA%
zv&*r8L##?SGeD}W>Oc27K>9J4?Q5?$+H);sXU|EI)A|LDu=?v(QL*B#D}I%76syZb
z5q(n`?w1K(V8qDp1z6T6a-zmUUnVW2Dl2RGieVYe*A@Z4a?@yR)z(V3HcaIWlDjCA
z)BU#GpU#1PoO-A0(){Pw=PVM2FOw_r_sK9_X^rvg44g})?UEDfo`FR5D<?MUFx)%D
z-ytl~#)wN}=45iaI7^qzEP{9w1?hIYy7>1xs|xk{<@rWpCE}e`aTbsoVe45f)`g&H
zAD(jt8W*2vE3+0RTnmUYLl?X?7c`4%d%GEH2GQV*f`iF$+rzGb4m;_KNwkbY%naS-
zY-W^}iIh_0<h=BDaUu8lUC@EtoP0uo)AIsf-H83mUDVfj&TL9CBx6Ds%NA}e1P=4}
zI??MGhl11D{~}!)15YOwKRT{ID=UXQMDa-&aarDpsN#57-HZv;Pb25b2}opEFBaXD
zzI27iOMv%F`@5<PnY-^cz=NP`etpT0-LZ34A484%o3-f>P84DUloUH9{%43M9K#J@
znECk4ly5=83S<_H`SYY2ItEBR@5C1@)Y()w_Pfre3scJ$^_em6KnqxMu@LpO)n28&
z#aYKrrGeyMLHxg<DjrM7h}^nBsEzaMnESz5egsI@0tl0a-tVr;U?ClB3H-=4biC-+
zNeaFWXcOY|?ak>y$~(syo+ea{`vF%CmP3s^?1(DyIh5C@vPY}Ex=Ct_%D$CB>T(g1
zvaFtrArffM{#^uMgLI4|NPcd&H7|8-()3@q8sI@HcouM$DuU><=yPgXSEtHis##LW
z@4O7a@x~vZjqJKptQ5Z-Z-iqyf7;PF!Z`+UfKW9nz4=HQg5-=ZMx@7a#xEZ}J^>c|
zhFAzOrQ=HOTCFekcaz+5dVHMTgqIT<P~<>@xej2}&>qJPJGjOoWG#R8H41`cZ21jg
z2c5lrD7{*EGbYY-*e;|vc0M@kar8Hg46sy+_>fKU)PG0(R}2URXiO5w^|Pc!lk9y}
z<haz<ABQxJU&0wb>jAC=+%EO?o0zy4^tpnr+jsMBkMUVIr3n%E_LGj<-p0}X=)3U3
zQSB;C>Yv~FZ%Oq3893rFYAs%Q#QjcE{cocJ4TEkiIpBBpD1Uv>-#^~u3<5IVjWMDB
z{Yd|(ms9xzGKFTE_-mg1pOM!;z37jR*M15wJ$Hd2;`amQAJMkY;AnK!pWt_naQ}9n
z&LN^k@uS}N{#Tmqzrns$cs>pBqNxA5Pr_cOoq=;=d8!XE%v`{B3JH$yX=iY?umo4W
zpP&phlIfKsMIZTJZ;S^UGBaeptv&d))2!jj^NiwR!S}^T9Bht4kuG748qz0?aUsmO
zx&wyf(ilS4`~V72oF>diyKtCwFz`c>G2VuKunru+d%*}SqAB%9x%<KXCb+KjgJ@0#
zBz`1lkk%|hb(k?NHQepBT8B6p^y0~f(rNMfCW3XH;u$$~8>}>CFajI26-#ukNS8dz
zpPdrs`MwI?od?75fJIP)ISBO2jzZ#g7&<Zv^@4eja%v*HBhx4@B*cUd<09k`KS=Im
zDvvxnMPUZ2vO9;lRPg@i_Wjqxwa0~-M<Ei-_h+hEas-y50)UE!A%oAg1N{|^5s&;U
z?_GDDI|2Q$7TR4AlhWtX|8NLVAYmU)Bi>>RFSoZU4=8rF!BV{sj75ycYkl*eupbw@
zFCz?%>tLPfXj$_Ww1;@02WkLHnnD*XC0;cHrC|K?l#56+Ji|d(;~kf49HN!@x!b<J
zG|++FbO2FIAfj(XTq>omD{-8V&W3j?%xD3ErmXqidj#h4h#++namX>7TdYgVdT-XP
zV<#u|P9=pEvf(9*km^Y^QjjO!vdaE=lpAi()`XS%B00B`-~ES0FF>)-1*sPI*J7sA
zW^jhi0edH^kTxMsUc@-<{Q=UN6}meZfwh__KbH;pZ&Ht?-Y~;5X_~RWZYJItB$*ih
z{4QefjfmQOyYn9FUDwmN>@f*Tn-$VdW8x}g&@_kNyJFEnV1(6L>ql{WRh@pc_l(3P
z1Mrd5tp>3(Q{|5-@*U_3!WTRY?(N244OcnF?=vHj#Qnai9<2Xqkgm=o7=(Vx={Rw{
zp=G>AZ_XrliO%GwbQ*RH#HCI;caG0H;S<|-(ABKGK-?VdqO<;wHa5o*4oQ!SOC#vr
z)jHyV#2TAif5i5Pm|yDa=DYavLPXSF2=rFRgqrO`xt>o4w@{G~jnz?%4}{ju*t<Zz
zA>C<t2yv(j%`NbI-V1qQQxL)FH5>r?M(%zL4TYYc2AM!KSU^ccdG><*Z#MZDEaEFw
zQTwFp*OG9NJ{z(-k?3OM>bKlF)I=fplK9172iFfaVoG1hDHWinZDlVogoCe7uRt%7
z`y>L5*PGG0&$6Gp*=vofu**Mv1R$BG95~pg@Q)AAg44};Ja%7#!<8WRU~qiJa1T(#
zc@DT4VncjF!2IO?(O>%uedJ9gqX$N?`8`J&U3p_uMVnSfffzAsYgm<kt1Z^sDn8ZK
zLv1nU9+Q3CXS8jSZWWThJCx^zf;1`-fW+mMoE5Nb{0P+4izv;+-KF09fuQ^y;B^o1
z54S7w{JWcF%Sj{>d-g;$=-1uyMc`}vg1%@G;8@<cct6TaT0cTPaxi6Gfm3=Q&|A^u
zJtJ|(!JeI|px|h$hOvClr|-<CkTwKzIJ*#@mNp2lx^*MC5`qEVB6^H?`DkIDJ~HMh
z{)Z~ZuYIcpF&|zi6celdbr~CHJ`)@k!{@f>xf)cIO^a~Eh0w<{*wdkN4*}6!Oe4Vo
z%MzJQfP@S}vegYqG{|jyu5UwdS19T}P*$rMhZOlavFiwgcWrkbDTRWi>mye1-zfwK
z(g?JQaXSLwVxfbAbg#=WM7Zju^77Cp#Yja{qd>|O5^;WdAKGFU19QbuYl30Zop(58
zJaa36t&@#Kk=80uZ+C?J$<>8=?#uSIzhWUj(?Vx(H0YWNVwX<S572Ff!1CSPYUpi2
z=P2ix63}q*0BO(ZQn>(BF_Of92U^xix|;vWjrL(#dk%_Pzm<in61w49fvJ9?soqzg
zAu!^k5Xd=sku$jwtzmL6LjscLb~lq9^Z|e5@-^Vrb>G{d7m*G`Y?vIyD!6IvO9A+f
z7pBMf`w?JDpiNU^Uf&xsf#!C0lXsM}(++y(6pfqXu=eDG$5~DH6(gk0P9IoTo}Nzr
zB+;ira!P0Z2lzn*TBah6O>HgPePL0&oM}EQ3f@vhU%CQ0fks0*N(I$Bm%~9yZZrV7
zCVt=($5DuhNN{j#4U)p%%GvCSnh;rZ?>k0nh=I-T#^T$I2)=rHu(RBlk&}PpvN2~2
ztonT&!U=;VIO{#%Dc{7|qgOnCKjScgQIz~enGDu1(XkwTTY>Lfzrmegy)A=q=M#=-
z$JcV7`{U5_rgv_GtDIbHc^S6tQX87I&i?C;?%Lkga6dT48-VQCd3P|B?d>_7&E;XC
zXGC2OyRFYQ+F=L>pec$4*aFK6bxoDO8s@TOqCWDQ^_K}$4Vq_8-uN|2GsLNwabKIr
zj#9(I`mN`T=c&cPvLdNn8iQMbLy)EW?x9IpI2~y%sG=$5db3+%3`)A6{1;QjaH2u*
z#q;c$Ty!If?$8ftfNA~KZ94@aOu;(%r*9gGxBxMZ)F)eTZTy6tz3&go6NXzI5}V=o
zA{gIe!Q(?|CY0^U;hrY`WeZ|**ddF!t500NDe@sTEi-~m{X1sp*jjlg$(L0~@aG|+
zxrYxf2OK&yNSZ3Z4S*XOR`ob}iUbSOMezk1L-!fTjx>Q<*$~-&EfQ<L!)f69h8Rt&
znUo=JZyK%)vCE83AszJu7zm#QDe=usme>~-Ng*AQLS`vNdG#HaaamyL=Y_T+nmU(Z
zSdaycNDpw`%GJu?hc+YLh-01cn5-vj=>622Bi%*TcY-pcFTIh2ghkdM@YmN=vp&5g
zo|3wx$=b<eFWpW3stxXu=GQ<?B%;ZIq)qPYP$%UHbaU6Rq|n8T_;5E9p@8#RhP=Le
zN69Vp5xEc&At)W9&oYdNC!UtN^wZq*6*OU{P^CZA)rXj#hHoBw;bINx$K0jeQ6l%N
z$HQO4en@VqKr?3FRz@d@FpzS3hj-RzgO~$DG}U&zmx@AFn;{mn1o@otfEOgg1cFv_
z)KoGYJ73B@_npe9m6$KDRTbStIw%s9(w%F!#>t>2e!=u^gII>e0I4*GORlFRMSG7~
zEq)twucL6f`AVxw&%lcA!~w#i4`2QkVR1Q%uZmvv0I=cnHEFqA$HfrWyxYLsw`VOm
zPWNHC?riL^N5RT!V)04&pjh<rp1t(bN8oPhwp)jygkA%?jmZT%^E=ROiH5Lo6h4k`
z&t$f|b(kxGfaf>DB#+JG-`yAf2Uk3TUtE_kLi1i(VfToOjfkJlv?m;kt`(oZT&zj+
zU1C)ELexy3IFSML|2fXqh0m5r2=$1ED?}R~1zhnCddPLa17Tdn=9-8NwL<UO7f+w{
z$JkXuwnN9+dc2$6!OQj|-m{N2cpp6_KMJuO+hRJegCBxE=f6M7Tl}5A_*m=+>fU2_
z)l+O7Ow93w2P2fj%VPnWn?Jcn-9)*Pxa&IQr*+<RSQUupJcRss72;2#bLc!9AatLV
z$76&B?OJbN@`v&SSFvPbx-P|+wi?~m=-hFQu15`?grZ|%UyiOP*@nn<DSP?sKgROB
zoHXWNI?nPENa()TK9Y~IGz1`Q*0yO5;PO>y%fGe#3f&*&$0qA-q4T`cD$m<+R&;0t
zCo6kzv;SD5kgrE=m1oq)iW^7CT=2*~;}S7n(81L)3-xgv;~O_j|DJB~(Ni>GXZCDw
z3E{2BSOOp8K{3s3u+Hsbttg9rw?*ri67Q2>o*JKwiSF)#56ejE;)|H|>bmb+QCxDK
zvv}o7<}&!?+J0gCq+zrLun6u}^f}H=q%xpoS0IsmYU|T^ofch31+v*|4-AxfHhYZv
zU$;gi^Fe(4o`+KBr6zY~X_C;QooS?niW1W}Kmgv@mt9PIDeeV=k*SIOZycY=CFk||
zZ}3;w7`}?!lqY+Gw>6@%Ef;gGTeCPoU_Vh4N|I09vS=B5uOls$H0$7m^=81kZo=tN
zZHKs*h~q0c3WKvQUT(+exMSnDTT)ZHhPk8;koKiLPGtLdTiX}i(@amcEZ_Vl2#4x1
zz1<|CJx0vCg=7xj{Y=KR=Y)TS<o_MuB})q*(7gQ%bRwb(IEOR$WFE?N7itKJ<t3p?
z)V8*r&1#^g9y304w+b<%TgIAnajTU$y%{msfBe*x7<W$04bpkEwvTS!Mq1i!I@)9@
z=D2OOsFhwh^EuWFb@45;@%D?pIn*kjT$s}Gv9)`esc*p(!`lqaVr-KoIh$E$cbt0Q
ztj3CB`bPn#f-{{CB=t17_8VQJLvQ1Bi;IAoJ^jRjHR?Tb)Z+*g=!lrJe1dCDB(Sd7
z=x(ay(>D=4756Boi;uQEgRb*3uBvNe?rUA(CaCG-zq2#H4CHC3aH##ks+4mxy*I!1
zqB-u?3GEP9#9S2<eeS}+IOV}}(R3u5OSeey(e{)Nj@EFiaNy)`Uh7k~CihN`?{za-
z>>e7`TZ^+{Hz$k8BgU0ZUK+)WV+se6?wOgVuj<NUC8AGXWxTwj#E+aSY12mIH&Ms6
z)!y#k&BKhXKfBZNg{aG1qp|O)SF`WHK>U3hf47pyu6s=_7h`Q5tg9`b(^loB!yBzY
zExVo+hu+vb|9Iyt>aYy=D0N~r1TaD6sgr&i`<YU;)dH)xY%AT!5`#7?V1q4OpnUgk
z`NmEey)N1B9B~xGA-ef$QT!{iHetlswqcL0L3n6!%xWKvhn)Ir)f5#MH*ojDCsv=N
zd(pf;dK>jI_e*tUEf$i<X-rK)MKKQDx;JKy*sH5j6WOAexM!-}0wE}TGvg|YQ*B>{
z{-nR6_*p9Flu*~s$MOpyU`6+cMnPgVPMuN8;<7<PIklAiD|w2VT)o4(rE*^|6%Mpx
zTtrX$Ywrc}d7YxTSW!pUy+|={`UE$9)tNp^y<mS4G;79kD=nd1LTOIUnu`p5%xSdK
zaZ1$s@hs4_P{<1b%U#^Rn#wcEnEQEMAR>NwH>h0v%lk=R`7dI}B^2UTIl|6iqz*A0
z4nZI33Y;V5o9xuGrf6(fE8k+y8seVlRDqv1#YBZO#84MGvv?EGn1!J3ZiHNX9s0)p
z26fC$`Mz#+y5)-ESZlp*TbDZ8u~3dy^h1=>Pzh)ETCr(mdry6&4L4mCPU+*pKfM6*
zmH<IazK!xPIEM)ON)BeI`~8<(beESS?xL-rJ*_wP8-zR*Xw`?2Y-Hcc66F%^{&x!H
zuP{rP%|q|&hQGyM_Bh&a9ubZ=R3}GZbvTGJ%=csapn@zEY3R*pHWwb<ZMU?SZiv}h
zGD7urB4yTG4kUgkNxEd${A7w+r+&$pU1#7vWH$Uul2CT+R>no4x2<^XK0Zd;4__Ou
zZVtyPT_^MJLA<3QnoPhX8`#CZj)E>(Q`C>^qnDEj$?k`HmFFYw>xmki)yjJtr^BaE
z6peT$YQxhsTxIKRGUH*zp*6X=6d&90u1!&HH+nwf>Q~c>-;EpM)4@N-Tdcuo=Wo<I
zH&)=4>vv*cO>=3_;+}w|jYvRAaTgL*Sa6YM4XVVK9II|~CF;I3t(GBFrmJd<>H>O4
zFU7C?EZKT*0JWH`<A{Ofn)mBTu6kw52ezsYr@@0-QQc{C4AGp-e}qDs5o)Z+zF$l$
z(u>+(R<{`G^Yf@ARM^hI=I+#}=lDa2)BzDJ3&-BOzI9&R<$c==FV-7G9#*K-Jmpu8
zYx8g*!sw-3+bE>YiguYdDWpU#wd>?tAnLgiO>~yjH3efG+PhDV5n1JhAV5Uok9j>s
zX2CnA2U4gNo4l#Vxh6NXvm)hkrO2}C1UHrj`TLCX92`5>qAbt$Ro?)UdPGr#SZgQu
zSsv8kP2r64bayt<3P3EI=-L47a%1Cz7IKZS=I*WBv7tLc-p@BPht+exsLR2ckYbG9
z2UTYuA{o6`$uj38mzJA2`0uGizF^@@biiNgVQs%h<$OQ<pJz94)IWu$9iB)i0`AL~
zxwgcb>*MtcR2ezr%l8|=s+P3k)4O^5Q3Z=>pa&zH{HED-PO#xV*TW-~Rvs0f)RlAX
zwW&U76prZ41lgB4qk`u)-n5^PAw$>hpFDq+>RGQ!a)q?}MRU&vhi@w6BorTWQ3@`W
z7uOH2&t+WxhM&9=7_5vPrOs>a>_6(s9ek`z`ye+~X=7jety+m$G*!E+X~$)l`Nly6
z8(Lpb?LU0jN=??^g`KoqeEfArU>%Xt==K727k@*7?7T$Q1A^-Ge8}dDD;Ru>dv=i0
zJ+7q}ywTO#fNsWCuFhGC=?TD37D9j}K$O~KE**)^SY}(b2pS$l;}n|e``ovOsg61<
z@$!5tTghF4eTZ_uex*mvlm<zB$UR#4UA&^w-0XtC*ePDxXa3D`jBxJ@^)i1YkpAfj
z#IPO~JyK*Wdg_m?UJTuF$%awXW|h#vkLDPnX@a-?mF@>uA546X*lvG${!>LO-V5Bp
z9gVW(QqbGQwa6_tBmz6gR-BY`-f%TNugQB7>Do0CePAf5yv1{VU|iNl0WC6W&ClHV
za|I@lU=*2#6YIq~<u`phLB&jwe2<N3dQ6_VRyf!nqt*|xbShc0pdPVy-|0U;RxE&(
z9oBf>Y2Q{L%4!WYL#0b8*~7#hB}vT;yNKS==)BDq&H4OH*oBdW6pGjXy^wx@k1HL#
zngV$xJrXR_f*m(CZUZsnW@(-!%^94;YFCZ9R>Vkaz{f0LgfGhMOmx8&9H-`LTOo(#
zxY<-*C!9aVG!uZM+q%^sev>R0-a&Nb&CiMdxYZsZR8-fza4$$7K+snm?=N^!V0QLM
zS@}vqXvCQ(&gVaT&)I_rE_M4NQNejcI5`sQ+W&C9`w6kPB(BpJx=>sqyz}S=BN9RY
zn8MgcgS-e&q1dwLux&R$=uywW(ose(x8Zix`dNY(#ml(5;k_j%E{tY+N0OhuR@QU&
zEX$2956HBS9ZWhDo?Zd>w-=A^G}he<L_cpD)f1c~Rjs;1WP6GBv<?L4@nJvZoojnY
zYr!8}8w63SOqoS~_ogsIJHZ;(<g0Tj@Y7DLYh88GF{0&&?l)mc8{J|z>TH7*IU}J>
zQ=cB=m-<AnhEttSjm3T5-wgts^b|&PUV3pM^9L*}FHS2SP>Eob5@<cbyts&96A+7}
zd9jk-Cv-1BQ`1q2k<{|6i^7>oMy7(I>ud9Sx<M}`xQKj%0ae?}oRt+UC;Zo7T{sP5
z^`6wr5@o-J`UM=VYhLB^pChAhI<1m0*vb{Wb`{LBNND+h*v+l4(zV8lZmUbL9_=Y}
zT8_ysz(^Sq9e>)TY2-UYu@Lj*ly?mYlNz-R-KjUwjG0OBZN?e^cKApJeBl5j)aL*@
z3uRSqcc`Fjh%`t!ib$VdwL^!3+Xb$~sc1;`Qj@!>5s@}3J9IIRwO{X5wWh#7oQHeh
zG}sGr8W$2r9RqzK*zFFuXr~uoU=TD5?pQQMo$j!Dff&?9%?#Qt^yK6WK<e{C_0cRi
zi+|JfKa{2ytD;vuml5BK(YyJ_QIHnit`upd$uP9<zBqU+j6`jFbCfLayd;fC4gTY>
znwRF9*S)<xV(`Q4G;0psdzsD~^H~4fS*EmPpW|oRFDj9owi6XOd_8Oq`*v+FEoCVs
zdNzFaf=k$NNv3UST3VX?phKz)+s>%K7q{6pkzgHAdiTyia4mPzBaJ9kKk{6UI$aNB
zlCC~ISx<;#*1#y4X_nh*X*bpI!g0`{TQMaQ8W6RD8cuOgj;&>@d$Y&y$)oK^0|<>@
z(#yZmocd0>GT)J+(vhl`HUk+=mNl1$%asU0S)PcO>XPOiH;H%HhnGIT*>F!T){=Eb
zy|>al^(<~j&bjgB2t3Q${WHRc4&$EwuRlCWsf15wv%R6)etrlCpX!k?FAjkgLrlDx
z_O`M>FUe8WLyCt_;4Q3AmUqcLV^vL|bE_1!b4fp~mD?D_e}ko)Y>8l+k@l5EUvw&#
zwPL-k^%=R<d1TGe>JQ^GSLf9K@F*D?vb91^<`QGit1Jot)u?fkg}qYoYUDZ-Idzkn
z6hLh1`fcoiyESmV)aT<Y83PoRve|}f3HGYW&xg!C!BLFjck+w4VU0pgGOwymd`aA!
z(P6q;020=lkmXiboopDN(@af$S1Y<IK{|Su*ZAw}^vNGh_Z^C27X_jCVDOgD$@JsZ
zjQH8RWT2aBNt)<s5}rUOTb#c(y-2r~&mr=`P?O0SNvc1aw}C9W-z|R>1=9~Y5!FRG
z5C6@@{`*X?>F;!pBLM%K_Jj~^rn$ZX^C`D?J5L|KFZDOs)?Z}LU~GxJ6bkpIkJ{^!
z^1YJG;C%HcOOkm$g?SC*BezR6$P8bdd(75l%K*Jrl5mC^Be;AJ|JxbZ|4w`yMY>V8
zxoS=qKvrMr(sqM@qujSt>^dBkIM_{7F>0P2JCku(g}!Np-J`BUjLD_Iy4}jE*wpU2
zAK<qEIH?bEVw8+s@s3CxKs*)<-flN2#RzWM<-}RCBzMag$bJ$L9*l?UCc_EN?S8$5
zIBDZM&gGdySargke<qKnXb%ow2479_{3{`SxstoI3}lMP`8x!s=LTk$iFJ>dZtWZT
z9e0S)bXS=`nOmYm#aH#Hg>-|yaSXbz6#&&KffTNaQ7AB=n}PGh7i}(-9kh0oEp|7L
zEOiUY&~)#nE&5{g4<qsC==8|N!983bk&XMuPkeMSDUNDVQU;X+R57O}pcr$i-FlAD
zPOZ7Mbelua3fR?#@;e`oRb;Mc<?4cJ-JqnfHQ%)T^vai$#*8fAp<Ly4$p5|mkp7dC
zagd;3Z+q=efWrxeee^!OSzsjF+IQGm2<~t*<G+)BWUJyV+5VjAgtA`eM`zc)uKfbH
z-Rs~SKT(e-#KwuQ$s*$*t8#Hp#sOpQeOE3rB~_Ckhv9C5UCQYiklTe7@!fV{QB#vK
zqW*E#%U1QrwUTnn{zB@3U_$LEONQNK{nRQNK6ipwsZ8%gschTNcW_)VrPyHPEgau?
z;ND$CrGJgp9WwSN9mz`6775pip}Uzl<N~dWyO6Y<%_;i~$bC<>f(YQjY_*-I+gQ%)
zH}X0n(a{3fXLc!yceLC09jvANgap)2s~Y4DOqL!@7S374hjnADhMCB5U*B1}mZUXv
zQ7!CBueZY`0vyNXA%Oue<m7wx93=57Z2Y*m(nT9frygv*UGS+*-cDZ06TH3&*=3tO
z=ANWxF1$X{?_almoVAugrSheXl;PZk;voU)y$wv9<*Z<wqKo(<%;l5RdEfKIwqo;R
zw8Z<hPG^Nyg*|`r2oYz$OZnu_H<x_+bSH1ibjVVF1kJ;$r^mJ4u9fJ)Dg2a&tS3LI
zMYpT?TcL|rQaMtZh4ysg%&CSeDm9HnU75RvNlNOp$eR&zB^~RTnFLyG!91j8h(7=!
z^&Gmb!!+CEH{FlTy0q|i6!4k0y=y2<ol3Qf!MZ;bECk!ehWT$#40m<UTKf#cQulMc
z`>#J{aZcv<9C)#v{Iy!nC4Yb_++9Ju`!f~CRvMBt8~BLd>NS_AYIkuE1%)5GT*XnL
z(J#qOq}T(7p5a>SpstC~kZcp8!mmf$7@?%EAJFjT`JtT8x4i#8M!0q;z_ilktp<kU
z0H?Q{SidLH%d4T1p~>aW>I)5$>sbSLQ7F#05!coNHQwYv4Wto$^DbgGj_ILW(|&&p
zFpR<<ul=S-7aES>A~Z?3yHb?(MtQveuHj&*L!NnN3H>skOKez3?&hpUV;W6_!Z(Y3
zyiyD)YK_Y0hidc3x+5Hkc1xqN9d2rMZBKYNf3~W<qg)iFGjmvZzP~67<5@IenAXH?
zAMGkFxt6LJZaKA7$6On(!N`$yE?DadZ{c-9JCup+C8!Ngd0aK`$u?+8bC{lm4&RMw
z7U3EXII^CH7LC^wzRa@A>beKiF}Lq^qS*+qtcGf;saT4&S=9{V`j9C#w8{?Gg9*ts
zCWk4~l-%tlm8TJ2PpT?Jw_*npza{THSz`6fAHaqWsSH|@tnoDq93|q>0z>A8BZ?*L
z34+}MRY{UugOv5=D8D=eNEyuYakzv!#6@&8T*D_HY6pzF&Uo)_TxX+WD1Dp30hG;}
z%%QT~wU7C;=e5bs`i3VTreYP^xiYZXdhM$Z-wO2=FOl)FZA?P_`uJO^i(W33$39!6
zWfXD{afEBIHT08J#F$S3+0LEiVw%f8_WW|XOeovr`y^F87SC(3m1O08_jy85G<m&w
z_akecDi~4j`?Z9%IIrm6?+R<+TKBHo1g}f4wGlPBMzN=(`*CYC)Dt$8EqkJ@e~j=Q
z1{mRe>XtVKtt%q#3}HP52_5`fhAqC(3lVL5nltJ}?`XZ@o?3VEKISrB^mMr9dl|j?
z6ST-b?&~mm87{-le&RUcADe#iZ5)DI3}Nw!Hab?>QU)n^!=&ebbVP=;4+kxGo0}0V
z&-e@#GxEf@s1A^9EAh}|%=+kgigaAbnjsT)?+ek0W;{zGgLNOc<9Z>AoJEB)-!;oE
zY_p4_p2SZ{QhF~yMtsztgZz6Xp6AiSK9(ti%mH!ZAxVYoWKooRGhtr|)zs^hn+PmP
z+B&@Q?UV;CXNrSFKIQLyVzU(rGUC?@*(rOEci3SUhD~$E?nZp#yxO>OWzd%YieZzx
zy+Ygfj$$dl(eD?(Rve=n&Yaik7N?eT{c^H!#<jEe)<E6FzTM7kNCZ6K4?c-AS7BSs
zR1!5%XS%guNFl+2A2yIxE4eyAe@6zJzTr@*t}pBFuw*Y#w-xf8r`?aPL||}>6y46-
zg=zWG?cBlJHmx4(+BO~b*re|*j`^H|u4#@MW>3zJr7KMG=;4aTmYBih6)s<ByghYS
z@PL!v9_66tK-q#(XI>t<lYUJp9S-N{nTwht-Kx7cRQsmd1@9E-_3gBEHB#6?p*u5t
zZdW&`%PEBVUzQ;J1~Mvqp5ez5e=btOsgE8%B0wdN1w^DWZs%8|%;#ov;z4_-GpWp}
zPJ)X`h~Ww~$t&_Vtc^oSTTGew$!0J4PN~g6hGATewAjQ!>_bvL*IO5Xv(BA7oxCjh
zH2RE}jNen~Gp_pzvYhuYJbIX{;Y_y1zirEhapw{biF_!<49##Mr$ksaJK{bdW*<yy
zDR5Pf)sa+V4JtcrjsP){HgGYwypl<iz(}{I#*(_UZ#KS%pcsf8g!*0ToSqrts?!w3
z-(9=ZD{%%_5YG5l<qA?wQ)Y`2Wdy;_!#jJUuNPP3L!WJ?43-*SGqmZPtN6bSbj>ha
zd>l5}`$0y(=7Pt?pAbkjXX3Y7u1Vebu8i@d-OM965f4;VT<LE|a$6Ea({uiKuZ&5b
zb-3U3CcKkEj`enE+NFCGvI5R=DVl(OM9PW_Q{*NpS+~8DlA^3gbQnVFqO~tCa9j4}
zUtP@1yix$BJDs=uE`^7;6}=;R+C-{`w6!B7W#i&ngJQP9VqXEFk(kPtQ|GUKD0N!a
z>)6Y=_U||N^yhO__+g75umAm5JRac2F>uHkp-tOp+l{z9D>}6EO$-pH*%~#bYSDca
zc3jobRY<qyL8bY+UHP|S9%v3J+j-ga&%yIwI~+chq8*&F%$q`3FaK=>5+*!KycW&x
zWEE}%CG(nesDc}Z<mr!+DJFbI#(Gukm&<xqC6l5KA0{|Ze}C=YK26vb4<F}TcTmQ^
zj_`lneHGJ?%L<N+`hR?x&f&+8YBuhg|Lq$7;cNfx%i?#W!id(Mym$Zq?PVkykErx*
z1$r<3ng{>*P5%B#$gdPI!D6gl(tF|m?PWEjRQQ@Y3Wd-9*RS)R*CCuia2hvV&C2m_
zFO%|s2dYIPUI+)_D`acVhmYF`o;!^$``3Ae1dFMRaNQA)$|-&whkkA*schMh`0Rru
zfidi}rkyvc_<iUVBd5pVsMz>s3mk&#5HpoH3JFU~lily%kS>f3OKi{t)4VfPZHkxZ
zG=s4RW^GCftkM>NM8F?Zr78mtql5wD?Y~{}e?A&TKX_gbY@Cn(x)QTH`g0vAn{aT|
z@<$1nxlb>gIgD$js}BjG6CxjAHsp0}HJ+*`=$QI(9-N{PzqCzQvHBpwu?C`+u|HOt
z1)B3DJ1q8fY8P5)y>L>q?z}PNTn(WrI`W#)Gr0&(#)obN=G~bt<g6Y}?D+M>JPR(}
zo<jTItD$A1Dl8)r6%ZylLOHz`Oukf>FY)OAefIu!)AS9_P~jJ6zmf7g^7G$3Ja3L5
zN*2g+w<Df0o)gb`PQaF}5aYVDxe54PALu)Ehs?^=WpD~Gg3zo0y90Wx42TU8byOdK
zRw|S6@Ris>lXnAwWf~nJ9nNH(lxzKl-zgI^<4&6Z74||XYKR=cE_6*6#<0xiYUwVP
zB^tlwwM$Il7=e3dsatBF1NGFLOw_5pTmQ1c&Q-vzcM-xp^hf$2CLfmsKb*%lt^ydV
zgwAU&PY&U5(X1e@YasHqM9OQv7H@sk<@H70_(T1%Q#CxDPvise8VoOEq3g0%k<}oR
zQ;WFf4nS}xJMqb$QH?w<WT~3fKdW{cl1p3`5ShZFkH<gs_+Q5TYj3zBW!|&vf85T)
zxQ`Es2p`3P$)#8cn?Xp|4|;q}?dkYJ7qQdXh3|Ta{$<+zA1k5aJWlFOdjLoJL2iFK
zvDadtI0r;SW`)pr>^4}RSv0-jMQ--wVmIt(ci~^N<if6ggtCX-4v;mLU?7heYd0YH
z4(shgS=THkYGfFdBbE()06(1B{ssk~en21$5tjl>i02%^e(IUqgw@R*gg22@=|DUQ
zxm~yI5zN(2gRmJM<sBG=x^w2=s_+~xecK-&M$85f6X8veYvR6zIA{pm=H(DmHQ72_
zkm_lHKS2geuy40>-843Ma=Dc~uOM`2=p8rSwWfQG5n2F0R%3k0K(Cl&kUsM!1<BnY
zW`=!$t&wE~{>X3u*EwF51U5DDcGVa}DeW-R@(~Cb17JN}?hStbh*&-XQhUNkv>x%i
zx_WNB+c<uf@Brplw#h5;b2(6?6mTn<sdQl71tKvI8lp0)rge0LM-ZzTVG%uHUGy?K
z^*ZGA(vfKt?q+(zerbRg?GE{us_$=Mc74${GK9i@Q+njX?3mh{xkNBik_4c}kvOl{
zUDV{7Bp>y&v8ljj&t;xj=gziTzXiyOWHyv^x<FMXY;XR~0YQ8997Y=pU7N%TS|-d9
zi<;srPJrlGx&#zmWeUN$e*p2c+78sotc!T2GcLPXy3A?W*Fo}3NxTFA*CEc#2mR?<
zd5{M%i1jvrSueYV)=G|uVhT4aPb-ZTZA8pD9}Md)PJh`0XS&cOI&dC)ZwcJ1BN(yu
zyUb_YS3rNeS$xaE0*U~c2$7T0t*zE9u_~O(Pai1q0de@>@e~x|WV5ea24U>?6H-gp
z-O`&+4YG4Ku7IxMOp`>NAH{g+^~&1?|6hAw9uDQ+{~x6^Dta<V);h_MNK~Y;RV4e&
z*s~O6Cm~~}CT%K7*_UE4mQf)z_GC$xtl3KTeczXU@4HUtdA`qi`kwRr{{D7dU6-0M
zbKm!8doQoWIwZ{h_^XUt%#YA?E-@``O(%O0t%s!~jb*M;NIi0UjLsWqwh&*4q&Q8~
zp4=bqs^4&O#Gq-mJMZ%#q~#Gv8Dx(0L+2Tbm=T8r(B{14RWNu&d10kv9(3ATVLJ}h
zPqh)VvmU1Klmp!PAL$NWo0F|jV>orG{j1G3$aKsxo^>nTek>CTkcT?T#ko~v>OQ+p
zt&IaFm<eY!GIMS%aKbAWO*bcX4c>O6a7&Nc(1h}s@F!=DZ@{z=znk6RbPm?hn0PJm
z&u8WDyT0ZLBpfIY%l~n-Y4*@BKU?)@I+TGVSoF*zsju!2-R;KU$Tu+TbA2!T8Bqt~
zB}vMx^V+cr{34`RvpilePPh!?4Z)|>7Ub0V2ZLAE5nn*h-D$peO7xa5ECWz-JDmap
zOV)N$RNHF@2WHvTj5H9d)Q`}NLCVFG32zDpbqm|G?IN<&S9>9Udks7i0dIr;1%H|Q
z=9gK|NCUP{QBOJUVRM>cM1h;`H&(s+DNLp_%wcl4&cN|UYe`i-KyJrI+C*(3Nu#~%
zrVvrDun&lEA-}KW)qd58V-Cg~yeJM^z#xRV%*s&_4E#J8vIb~=M_Hjxc|%E^-?J(H
zZ19)(SRII+#U}`s>c#b73@&wV2sGNOSr6WdP<VnewOT>3Rk=cxI`#haH!m=xzIio4
z?56)&RE70tl<-hEj_;fw3xh*L8>0LNLX@U%X93I|32}L`{zW;pqj|<?dm1!>OQ%Te
z>l3m}PO5L$>JIZdRq*q>J*!Jhw<7hi_BTnL1OwI*z2i{RDRC))24pMZJ}b7~V<*Uw
zGRG>W)v(l{*WjhVALFUEa4hif62>%s<<>GH=cXDu4<5!292&!k=|}7|7GhyZ)F4&R
zYVzDV*gxFdKql{R&ev*hIH_7AU5k1#Xi&EjtR`<m@@I{0kis!)(+LG`4cs0j%W!7S
za6vL^<ib&g6^Z#QpU&q<J#z(4q0+;G2NPBTmlIdA=pW^ITcnG90lRXwf!{t++_7Kn
zSTCwV1$Jr@(cQce4dL`oTNis)a`TnpRbiIU_B_`y&hh=>26`c*tiBpdX~Ki?FP^p7
zJS&qgVmEdxM$CT2Ha9W{G6f!sIu<C(;0tSHnznF<ff&W!pxyIKXkooKS1VTUo1mPJ
z#$XqvmX9Qars4D)5Xnw3q5h3(;ba&>O!a-GFnpy*qerVpqqwxg6>&mG{1~u4v&7_?
zlv;oDl$tAo&Q#Nv*dAwU<K<`LA)E*D3I|_lnwfS6fHQ00QB+JzZk%UlF!VWhzI!&_
zo<|)rrZ^Tve+HCzHo%RahH>i6Z{c#3>*w_t#3uW2Jgdz#Zy4}jr}M)G^u@k#hG}&`
zxBKuf&PHE3JI2MQ|B(1Sq)F<b=b^tXv&v57z)Tkrdd+SP!clN%_5IfhMd6?T)4j~v
zr;I;|&MB|m?Y9h8J9=AY0|!wMGc9^^A9F8>H^09-zapjM6wjr*GM2bYd)7~1={3WK
z+`LPp*UJRU-qRqX=-DeDQc)@&6ma7i06D~`tBitttc?5Dh6@9JQ!rTo$YDjPME8s3
z=#?IVOBGGlI3xDKzVUK^bnH&ea~}G~5_?Zjfbc%*QMQ`f_Zt-V2Kj8|rP1gwu4Y$H
z4D%>m0;|+7bQI<C*cl?=?v!|L-jd-6$&zHYC-uQ%$Mw=hSblwbJ1BtlW%CptzPI>8
z#dV*0-~{X>HdTnRNn|_U<B7gSqu_dr{VT5H^(UILGfk*0yO$4puP&yxoMhc;+_E^S
z;uNI2<Gp)+tA<P{2ayn8p@@r{kluX$qI8UOfOO2usupMNw-@z|dGZ1gtMQ8K59@(#
z&qa(CYqeZ~L|7x|QgFLL9aKE)mu58$kH?5F_&VmibBP*>K4MuHD>m*T`mwcs#OmS!
z?CTc+0S8N$IdOfG%%PhUsd{MQ{-TPMG}4;;hJ!Olsqyd}#)-q%y+^pLh%bq5f{{l)
zd6Z?FQm@8)kl$HB^u3Sy<EcDe;-uhGbHzp*vqEFE@NB8#$9gywxyB>%Y9x=!EOL#9
z8SfY8Ft>{h*N`zq&rc9&dzov`@O(u^czN6`+EFA)g9l)b=_}L#-;(Lfm1~6BfE~}F
zXgkoTc(?m%b?MXY#jmw|QJ0yG&hS`nTc|g}p5McHZh5ppst_RAM_LSI-pv|blvpt8
zw@-me-z4Y!1V7k3%<x@mO-yHAv*dEu#NK;@z%K7S?^qKV9HHqwY06zF7sxx{UL-PI
z=h^!$htJ3)Z=gzhpy3;x!lFlMiv1aM=ZUX&Y|AGK-Z@j7Pb#d;d@&ntmsCIbm|X<G
zk!I(;x(*n+JW%wi_riPCxuhN*_bZrOhz{8I9FT%1roUWRH2Rp;(_JC%)#SunbjFxl
z+Vkd6G~c=WTE+DsU6#P!!Q!Ec(um4&rM$vn2Zgn@q!yQx!wv-<<e9H#t&SQW*KoDJ
z&cd@qV|{k2EB=VM{iF}Yez{Y?1(tWg?>A*TTcX;Q9!CY~TAn?rO?BE;pt+DN0@E6I
ziJO$}S@nLh?3HfB*|m?I!u93~ET|3pOQqVrz}dxuS^W~S`Ms@dZjrD)Ni-q&J<7`n
zRURign(waJ8<JMS-fdO9D~g%A$VU4JIrH)R3LmD^5=}`3iwo8|LAtba=hccgkydF!
zT8x|e(t8{50b(H~qNpoG7iubj^$@<u6sqmaJqrA(9b~?Qcdqs<V!2(2VNZ{|9O*Et
z0Iq2NYm9P&W{iIhNrF_$%GLOQ{v|55ti8!iE90$jfQ21}_1z8s2vj*Dm7rz7r`vOi
z?nIR6+S~HSE_@gbR_*#_(oG^w+MF>$a598C<UP<-s$HJ08_Lv8=(I79@d&hz9A0G2
z)B2*>J(x+pnhsOD|K<(nQ6P7iyPa>5SCgQAMEl1Zdx>2XQS9_#jcmvlv`6(NQL$03
zG{K39s>lZy0=F{Ip>-Be7rJMJK1oIKQZv|Ua!Xv7!td*r58@sU7$%%~S}rd~t!BeK
z3_Br^d6ksvQaF9b%B-W;HHkI->RfW(lcxNIk%=wTLwo$oP#=eQBHEoFG`iKxiq&Nq
z>{64z#xU|ojh`5E>BPvgI|NLni+DNFPj|eyP|t1djc_iYa`_vdX4ZdZL+urq$Va^p
z=1BE;R^n6H(x}T2f$r6<(<%Pi!Ds3zlebwi8}0a^#nWKgQA1_rVC@3WlV28I{;f2@
zbV`o@1j!g*Uz^CQw)WA;PdMgIw(2+5Ra}k|x`5TS3aGGkPsb^=Cv5MaK?8bTqPrIF
zyq1ENt)}Dw@96nA6eP#m5rY-mWPtyaMk805Q6fXPjaaThRh77MXf|u{?l+Lx{+#Tf
z<#L>`TiYcrlU(JB<+if6JeXTRpM>J*0FC>~vhFL%o9KeMaXE^`LL{NDli?a0xl1xD
z=cb`krAenkaQc{Ks3jb7I8H8+a!g3(DHXFC&k|lDl<ODlIgzY7soJhNX#s)cww1GQ
zE@LuNs|Z*Zi{V3XVyspqHf%>{YU?f*l2Av7^`;5h1eOfwr4wu8=o?Y5F<kmE97Q0h
znW;KxfTkyu9o$W8S~DCth5m=Ib%+#i8Rsq(y*;F17DC6~c@e~LYrMefq9wtBakHh$
zHscY~SH3Th3Wep0#@i|MsQ9@Py4A9^b8DIshYfTVx~R1p=TE@*iR@}r3&%vJRfR6E
z9FXb?9>EVx0m5?EI#ydxU9H=NfWhBd0cuILtpS}}pax;HYW4+QC9X~QB#i;VV{aon
zuC40DG*GgW3+tyPJK<nm9(LTe<nN1K>)Al)czhSj#FW1<Fv9%)7kzxF)EuglbC~;P
zrS`|3axnJ6%!`ZVZ9z&rR_mv;mY%-o_v4;P_WW@DV1GkFn8q$Ep8JQHjV!o_a<25)
z0!twSPSQ`(uA`xI2IIS^MHfL~H6jXVDE(-+m&Ws=MWrkG22YnB=jc<*G(4B9Cykl4
z%kEaKjswGk=)yT8rV-XsWvD4s-c4Ztu<>bkxneN!&1f9MjF|p$rOG+zITEiK-Da0@
zw1BufpExNT$sCxHv&0HcJ#2`Zz<HoZSiB3_!kP;GA@U3nF)4RAx0ZGof!T%Y-jkpf
zJjtqo$l!wUQUz^qEFzwaTncKp?o=LOE$zib1A9b(SM^5UZs~?>nbZqVsXffx7QV%^
zTG*SA!*o6Z+n2h;THj9>h-wjq={6I4coLirLUr#~wL8Sjx6<~85;@|lbCf-T0}QK6
zh5F@#{be$@Gc$gI_PPxuyAsrAPqOr5?n`$hI%`bdd7@7pszkU|JasQhcbrHqStoKd
zdvdDIs2KL+d@k9&zwXwLvkBJ_iOiB&dMjK=^5`n^+ZcTJ?sSgF!noN%^-wW>8;`sP
zBL-8Zc9(Lo`<y2h_)IfuXb-1j#RrfBGzQF>Cp8^a9#qTRLSqYQsQ95(vLy3fkejO(
zrZ<2>%152#dJC)wWoTXWnit#C7TO)kke+j)-lZ^$5=w$oRLYU27YE<ve^#{W_j~C$
zCTY{0Dx)&vXvwjQ=iU+Oaij6P5U={gk4IzZJ3OV00Amt-5`vI^0=-(yXDmY|QL^7r
zbav8+<C^h@CYplTrDJ8-9o#A79@r1gi`54wuw2(lu2*-rEvO{HZVD}`cHynz8j`ls
zY*eoB6(_gxZ*^~-KTlD7U4c?$p4Jm`0gz-qZ-$Np{j$)PVTzRVIj+s$`Cj~yt7o8z
zKUs3N?QbXwqZdOt4CT@r8e0i&x<LCFOk_w-^%7QYNYNFA5=oS#U7N&6R>uKUup0vp
zt}YQs!TO@2gM6mRTmhc>D$z@i!}PHvET;JiYhK}<=DUT93l!ge!oDLZ!tsY8!Dng1
zyYaoY8nwz*IE`Y%H+n9Yv2eZLE*YZ0qFMg&)n$zt47JpyG56{Ki_@&t#nQRh$pCuQ
zkgfFdAd_frpbUFhcSwJ=uuS;7Q$#g+kxm<HamK-rEs90@7ghP|kFz|li}rU!nIq)D
zWO-TAa~G4@WTp2f?oEc5WJvGm)a@zKSrzd2y#(756OIK@YCIaliRBSUE|fe92fQQB
zO^x3&3)z024sW8ndu2jb`5KV0$sRVIb}1%SV*svV`W7l5lZy&we}&nQ{w2F)G>?k8
zF+g}Q2K-C0n6IB`7wi}r-ANxO!@e~5T^=iFE#J9?ZtmFm@ijr4^4kaUl!P);rU=Fj
zUNtF-5!i@n2D%y8gQzH5@ky2Veq2Ig*R}WU$Ed%o(X(n571kyG{&JDRZ?Qw|v?|Te
z0vs+|JxqN6w5UsIW&)4rZAb=#Tr$J+YQAOOLJT658@8}dKOY;&cC$96GKQ{b_viXt
z;UTlt98@N1HYNL=>pL#v%SvRy^i`k?@u@w6whm}2T4PI+Rbr2o?a9HN?}E4Nn^vbw
zveG&!$WAnU_BCpe)3_3&dO7@!6GnCWXj56q?8ERU_?GhOup|4_CNRuv^OO3{)rZSZ
z@x*@^^B%scwmo%d7n5CrxB#e>PE>3xMXMJNp%+R{mR;Y^z<t4j{7lVWv%%Fpq@d!u
zZ&;aeGNmLrSJEX#7j<yZnEf+z!~N5^jUFKx)ddYaZY~+Am{SeS$;0Am+Y$2TQ{o$4
zp}Vxb=bN3>A}A}KFA_H&Y7VVyEHVx|no?VNATK8AoAGJU+~XrxToPY>XwySSzlBdD
zG@w3@5nM{A2G@7KMg~4+SdXozYXTjx@B9hl+ei8<SK3n^Y(95BrOZBWe2zFSKdIaR
z7tyDDk$@{5nfV}m;$|&fXF%gXht6ovt;EgE5+6y~KC3;v$qTnT7Bph8C#NZKS_R2|
zJ#+ag)_S!YeWspOd^1yCymxSxtD;HR_mcO=acWrL%rA<=EON2h^g78|`n!LxcB^QA
z0dV9UPlFPg*Nl)i|Aw!)x-$Ury4LKz{Lve<?gBwGFb1^|tM5=vd>MP)H^GVPa+He>
z;{oF71g^M}Xtt5#sCGj$n=+GC4?_IpO(=FBxwo5iy2>WP_T|~k;>~+bNt36;HPSfg
zpBe8@skXktGPyJmBRXS1lQUxq91FKZtq}VXSnk)kJf>n8wQvx|?B0b9xxzRAN}b$G
z9D8F_j!a`J6vYzj9F+*yeNpl53@OTf$0eoGUcBQB(zwdA<lI19IS%QTBNw*wa+eE=
z=*MTuindWbs0xwpzV|kq5!mxKj7<X6AY@DyU@^|E4|lE1E!^{u$@QUHwpWsUQ)v0a
z=NUuxn+P}eAP%0|EM3AoZf=Ow-L#Ev5vf^jYR#uNA578Rc<|MGhz1D$zKDQIj>fV?
zzs2%Xqu5%LFkRM7_r6{6twuMl0(<=#cx>}eTi+&rGm~?YbqBx`yOOb*#LRQC13&R9
ze}eeM9)~j8-JS}aAFt@e(o^Dl8$zD!>AjoMFDMb(UIXn{7FX@QqUmr2Ca4nM{Ya^2
z6=rmMyHn{DC6p|>DBP~neH`C(HpCy2HR>RQUDY0p%k*ndi@SxF7kiEFIqvjYKWbo#
z{TroiprPmH%yX1mA9u`rqNyK6I@!svfE?Qij<0A*Xktze8N5$rdywMU{R#vRWbsy%
zWWHF97u+Q~v4IX@1!YX0246JI(#ih5Z5o(+TM6_8f)?olVt2&r-T#{YFy&05*lq7e
z{E5{n7-kCRrp9?M_a?^)?LJ)0vN(Pt-XeQCTqq$u^P@_TMh`dA2kA@WTxyK3hay~b
z-VXU9L41tJyHLS1ESdu<J?RfA0m%n4!<l;S_@Z`vhFMB{P4>7gRFTk$Q|Xr0El;u0
z@AN3dvg2=S6n8<X`XDu7Jw0dvT%#hdMGY6kUaTK@4|Pm7NvJ5}2>L$fL_iLz9_fsG
zClXW+&0oRqXG|(Q%UaU~(4A@TOCzsz7v>cFeWzhmi-4xVJ;W<J$Q!I0aWu`;n&EU$
z?baA~sT*6FypIU){Z(c3u`u-HjChPseP5BW_KbqJ0Pc70#+`$?xK~!vCrGZdBzG9d
z=b03-WqLz-1%^p=slLZ7^vXQBLTF5HtSC(~p&y|qc}l4UQp6q?v=Tza#+a_C#Zatf
zXUlT^k1zRqyB9WVm2s#QhuczKUISQ~Qn<T>h@}>){IEk8%@UM%__rmgoZfXnU6n^6
zp5g>8wXL{0%qltg;Z?}0_g(WSbK~)u`V0kT78>9|nT0x1OYWnjtvfIo2`Kw34NE?-
z4eV1hOrqGMW)m2g09@T-3E#sUCqm%SZt%$RAqdz)yRZ>N`dAtT9gNCW(ncNBOp>aU
zP&r8TG>n29PEiONzMj67;k0w)aA~B})vD1m#hm$MUWUz*E;g$&>uEN>nZThu1x4qZ
zzV|!OH{SF<I;RkOw#N;Wc6ONUCOJ8aDl^ejI1M~$Qx1f|hGo=D+56cyG7nnMwmcnY
zr<_T!a%ur=obp97jqHuEa{2EiEZpNcnWt*v!b=c5M@Vt@Q9Ihn+$2lbWC)rWr9B-l
zQdZKN>vJ@-Dl!iTbzHLGwXxP7Gx6h88w#`KOhK*vHc^^)$y~<qOHt~O&fP0m$`JwQ
z^xn*|?Dm^-89eWgeRAf?>uU1pG<4xrURkZnqWBI#Q)5qZmF0A#Wkh)PdPeDvyX#A}
zJ_5%~#>(5M%O@|qy;Y}VXVOv*CKKV3Y%W~Oo_5Wt`i+SiF=A<M<2C2{q^ao^E)#3g
zg_NSi4eft`bO)}&+sqDTTGsEyZxHD<4zf8vIP)Fks>`rCM`U|&at`U(#XUfYkUJ;t
zT;W;{VCOVhuDnKVw8<NYmptE-XK$WtHSQ7<7_pp6A?>lt)XD_#L_n<oL9c0kw!l+l
zy_^nOVcUB2V=h;Yh<skjG-{!)8#m3R@~)*ly`<Z9)a{pB7hj>=#{AxI8=w4)6<zd-
zgz3Wr<f{vFB$-ctGO$UrSZ7EhR+PxpUEom=|Ms<sMjy1qUff?CCayHXf8q^HlMfrk
z`{nn~I0k3gcec^?hA*+d?l&2Y@Ts>s=`k<CToD?uyX0u*XXx}`N8evJ?l-9pXyE||
zEgPu%>y;OZy+&k5ZO=VG^i5{&IL%Xi)>2euXY;lURtp!C#@v$j0~JBEzv)f2@aT;(
zL~#nVJ4T)oeK1@B0aBx|V@guHF7h}XVJ3wJR3Zoxa?hmbykcSgavV|a5G~q+k8tvS
z#~ABK?z=}dueR7$mrC^eeQ3P$_>C+xYb&@$eMoeZfF%5$Fm;2uG?v-n#^FBPiUIg?
zez=axbzrWQw9g-Y2z}}r2_BllW1@1S4BiTB_ncU9z|Esoaw12i)6tI!ysONRB#Ss$
zDs#}rc=zF1lZ#p{H#G7sp*@@pGh8zAa@qj@6Gv_<cE~;{L{0T?8xD-tr-+iS>@-ol
zatwTNcQ&{@zo=@HASNKgXFgo`E?XmQ56?)~Gwg~)9n$NH2$23z>d@c4qv7!W2;FZ-
zsCz?<SktY}w@drv`IM<m%sq@IE=m~V`(Lkm8C=~X>rEpvwm|HMn$a{3n{#5flv%+?
z0~dH8d|WEep={+Fw8*Qkdyq$3uY32SMLsCjg{fH{9L6S3rFf)|hvZ!*Dg(4~%WoeN
zS+*~jFt6Uc62!GD44pD5*J28ektJ*R@(DUskZHL!1_Iw^Hasfgop;+atuAuD$z!$|
zt;Ahak74HNfI{h|vg5o>yLTOka(+aV6ZP~?ez08q31WpYpa-MMzsd{EJao@!z8tnJ
zT8p@Gcs}WyCMtJ2n>6L0S=i0}W*oh|tx+oE>G*Ddmurn1rmS|Pd<@%?S#miR%k0H@
zbtcxGW8X5D4zGD{!`RKM<erAP=Pe%%Fh&Ezx-KFq9uqBdtB{D^@F+~LU-TR<pxWSL
z5_yc7E}FF1t4T(k2t3#Md{&BVqoLL_2TeK&>M830E+^%ju7+jJ&Yn8#?Cxje+Q>|g
z1mlJ|YHeSpK;fdnLXmsUWvf-IRIBPLjRoNtos%8R7HQ>q)Lt8&8VTm@H*3eh4^7w;
zJ79P$*XZW5489@1V&i;wo_c!eukqOwCP0?taZmhy_Ly&C-8O~MQ@nM?q-L1Oje$2)
zm5!Z3k@iS!ar&4^tBQ&B2-BsQGc~jloIqymg79KIiw<;c6)e^G@8)0_dm6HDvYeu+
z7RMHNt%U2=ZN1CNs9ki{uOx@q6kkVOBu>9xrfztDW9u%3TcwFnOlIs1hBRqSx0n)_
zHVu2>ke<+!9@)@fuS66J3T#`JCWlSCkFy$2s<`zxrKp-zC1z20l-jcIWL2l=ksdxR
zm|ak@Vz4`k-4R%iG;=(ci^OUq(ph<MVuYS^1IJBpi`~#bVvZ9dQCYPoDJK0mRD`ZL
z%1j~<oi+WAP%x&OX(X>mm0|ZkjG7r-vgR4Ze?vPqIWW-mWZVRdp2MJL3dT1+;}-q?
zKFPoVkZWl5T&`?vy70cr_<8c4ly?eqiu_Ux2IwSQk!qayn&7F_F9VR}MMEItGKpYu
z>5b7@9V^NObKvLo!I4@wf(lJyCP`3Z(ezhQn}%00mgI|57;!~CB%?gVaLH4srCPv?
zVu}d1h}X^)CGO5fJmg-fnw_O)hHUlezkj>>4#AD@D#Okmxfjc$O}k6;jU#X@I6C_!
zOcCe8d4B#t`OZR**}<Lnv-Ad@Q!ai5Ji(6cufJW#UUs%C2PLO9Dsk?^qSa#L2)KUb
zFq^Zjf8O|MaJ53o>~Pe@C|Qyu+V!KW%jXZ<yZh}E#cR<Lvv-pD&65@|jj2U({x4Bd
z`OtHY<3$A%(d8_*jRU#`IzlZ(7m=Qxq?&x$f;0Jd>=(=F^3rFo1(r}zl%eoo8*#bF
z+AQh!F5jAo&A9{?N&BgMrM0?~d?wHobL;Dve0{3GuKazzlNE>I)J(_cWiD+WH2u6#
zK=*n#8n0OR8rXxAVdRlB<;mn)9j}5uo4f*NHH==0PiW9Qv+Bn#VZ|67DO&twoWkRu
z{jDF^mVZfd{Jw5w*>o^^zoEkS|1|a6D(D<5S~B;%?y*qa+a@~nX065AL|u_2ANLrI
z6*hk+RN&ybYYb;+pZjXJv5y<b?jEg|&MiLVwKIB~LbApcO`)`ll;!8Kce~*4eMMR5
z%jh;jWbJJ1%q8tQcL00;xJ1aMFb^Yw)cp=S)}GuFI@I<g@wN=JQq=lK*{;|36$fx^
zPW3q-IW*-cEL%tutme1M<GX~F?3CJ@Lv>Rujmp-NM;%D|=<!DHS-qC){^^_pD7oKO
zVrNnd%GCTg&hJ}3Z+z~J{l{jgeg!oj6PdBvtH^S0;aP{k_2WfPRt&~x6i!M+OJEzF
zv8de|Qw3kwEschi)CMH6g;ookfyHi}ZYK<6bJNEIqzA6N=iW6nO<AwWAF!rb@mixC
zt?u<NSDr6Vnc@sk>**1V=gn;w?mamrz4Y$Xbl&Zj4_T^)iL{y2DHf9yY34ko!4Ate
zPL%4!TPyf3hus3yiAAG9<G8Ck3(43)Z}u&{@7u}7!lV(P07>5;?pW#D<8!|XKcMks
zIfAVD{!Xf}`^f^bQ~7yM(F*QYR_!UnR|Z2Z^t6QInTH*0!AfXVN=Pa9U3sz>Au}J8
zP>S$N_+n>LH<MLcqCzf3AqO;5Nt(}8f8C)e+XdJ=nRD4UeqxXgX41o^VHT>6Ys$u(
zJ-R6u?n^Gvu{fxfe8RpXCzGJ3Q@e>{*WBoht2s>hz7(Yq-p!*BxOMYxe!TEN`Vj|Z
zNAmD>)M?5E>!*j#KN6kc_yZpHFS8E$zgR{jW%I2N_?gW4k@QWGKuxGC%C)UhBE%%l
zWLdOYfXgu3mlADd-3!8ZvZ}W!$&moSSQ!m*)1e|4GVFXcV@_EvWCLNRW-(SR&e*b{
zu@r)yObGptsG1oYqPdyIct?I!$zi$<7%NiuL8om$VR&*o8Nk<29TCqwg8O7GU7U<d
zGzb<@vjum^mmqI@RP~#|+w(7zr3D;8c)AUm?nm;QIqGL)!i}dvZ0?nPXC5he{}JIy
z(>Qih)q}H%p&L^J=)=SHLcRgQZKb*%$pOede08AAW(Y9iG03B``#(fnHm+ST?XM1-
zhcu$?{)ur*pw)bM&H2C7!ucUX;&h{-e?DP74v<hD+=qk$YXH>ho|FhQ>0q60gLodX
z(hLMDKM%^X?XzLUZtD#j_`jV!C65PK&gJ`_xA#BEa2;zSS^1$S5#f&`)Tvl+Mlpfm
za$>*W$WveJw}9<?5SBv?m)q(wzjYp<?*svBx@dN{rdGoW9+L?wh1nH*@8UD1slRO9
z;N<~Timr2G`q_)m5Mx>ez~50oC##+=>bQuA_61_roCw1$R|&TPwG|`2B~7y&5qqKc
z-6SJ^aBi6H;xpqY`yJUlTa(&GI&!b#9_THskkul{1@9Do0ZZKA*9K5?y_J3hJNa|6
z{f(deo*`Q!=l(rEw#I*(zQ6ou>^$Itn+v2oenbzyeBZyl5NnAnhclHo|BEYxn@I8k
zTyT~{4bLxICjasC1jAw4Kht{2|KB_iK%Sbw2#!U(Qu_VZk`(Ddq$vyxB9yX0zoZrT
zusQ+m>big9VutU+6)ZcErPP0X*6{73-pb%HWCaP{d3oP$ur4ke(%4HP!o8I+M|iVa
z3VnPK6HFA5X@pg95~Aj|ilBW*Y)nAXG8-1_<>+4*@BaC$W5wWkk)&R*{XPd9cltqB
z^(|`~f;?437=gkoK&Z2Wh1d*dk_vCx{-;J4Hg4XQA%A`IC!pgJh89FCj;TgTH-V#~
z$VmP3a{tjYU0x4(WxF;r^VkBi?KPmaLTfsKB^-Lq%lJ<x?w@C#GEC3V@e=Ky5A``c
zsEP%A2t8c74jS!_n_$0tgUD#jgLeoX!+0FDy%8<WQmCiK7$xExKmP>SJ{nzs0?Z}A
z$VLTnN>(+t0F~|?V1Kk8c_rX}?r%Rg`hUez|L8$rAnZNZ_e$~e?^7nWjUGHDt|R)G
zgdOzj@L2fuT2u9p*tBI@Wv$>M_6$N#OfB+wJ)$Dg)?t^h-~v=$y%?+)BfXBn{bPUn
zhChAvH<(X+`{;guBy84x!<*?w#G!tb=Q%%k;^#K;s}d2PW)94uk~F<t-#@_b5&I$x
z;uvfjI~s}$5Gg_co*9BZTpDbph#k4u_7MPwTETvJZTQ70EpzbJH43H70TTJBodY!D
zvUYGw>jBQXB>Ry=q@>N0?cg}4TKWc=*+>=+n&-iYhFUfhXIlZ)YX|tZ=#v0o{qBAa
zj7(c#Id5+mrI=iK15_*plR2Y_b|}t`!BxEmBx!S(_lh0P1fJf-vG&)mVT+cvJ>u<f
zx-jJ*n`mH#UV;s7?*e#h>eYdo%F}49Hy~g~0i=?N0Ja6x!p&j)X=wn>BHJOFt;H1R
zuQk(%Gve*b@L6Nfq>cfA>$$jRF*WV5HQFHu7Bqz$cYq05;{sv|?h1DtTGI#bc0eX7
zV%1(fpC#vI`8)C!A(TbF*}6+0G&{{Z?PZk^BD$^*Stin~oT3nYY$7{gtc#)2RFDB+
zz;<AY@yH<R*U$Hw8cY9;$&UI5Y{e|Q02PQ#2F+#q@RE6`^e`14W=LcGVh9brQRtgW
zlISBOtS%rD8lheDAPp#ZRc@yL!>i}_rG++aWjC}vj8vf%9mjxuyOj=`{|%9r1$_{G
zfDM+cphr(k7Hw4Hm$dVMLVOk?ehU3wj*?rbe8hGBD6rKKLLcJ`3~S<d)STisC}vPu
z1RG|vQ1=k1Ja7)V-#{I>)SLB!EgW@Tpu>ICqKefJ?0FD**w^c8D*;|lq#i~x5Vq3u
zfYB2_%Wel1zbcqYT{RGSxM`6J!J$0p%N*310w!Z>Dq&%=DLj&9<8DE=IRK?+L2>+f
z^?asiKKV;ji971^5Eq6+%46OZH#rS@a2Ws>6B)?|!-5<LhpYrzsD<m&s<%_U5+u2U
z?@OBCVsUgIky`D&1Egav(dhmNX%|XbMhU9BgFGkO)sMJgBViEiZohF|r;kEmz>zE9
zyLvFRXG^6ltiR~p`!z%K=WZd_2d9}%*S%}M<5QjJQ4~yUqF5)08(l!e8yR&)URLa#
z2RXPKU!M3|P%;uV!{{w5$r)|n+-(k9mq*g-HNgGdV2P#oLp{SxUW1BgfDjSXW?dLv
zht}NH_~jU?@N*j7@`UF&(g^~aOz%TQgqG8sj5`2o2P3vXcF5eh?@TMq152pDzJLb4
z)Qys_*4Gdq+29CFMtgwS+9h$dK0D2XlXd2zC|`rg9`LXm7LP3uOU<Xnw0U)*MlIxy
zN(M&$2<uCH?2Cu4;7;}lb_;tb=mmkYoqimB7Xn`wIzXv<R^mPodG$VqmaKA{f(mWN
zZ`UO3gp7*`TFYTWYp6Q-iyLlb27%>0($1F6hJ<2`ZjH2*jQE;42oj$Fk~ZF|fo#@3
zHPh=r`i4I|2`ly$PzgjMnMq<z&Ryt3O-_L)8Q5G7U5!wPZ@VcZZ45%q*#I`ac{sKp
zIY#;zK=Dy#?_k6Flvz-7?3eY!e>vNnAcPPeC~NyMc#q?K{XGEia*UylHQN9vO)r<B
zf#h>@*fG3Khd_Vzx^H;+9Ef@`HFdN8hEI5%GaPd?o+Y}(JzQIlE`6$}yuC`88u=-}
z-&O;NAdCwbiNp=h=xrzliF~mNK9n6vh#fj<0TH~6qc^*>Z`~9Ckz4r-51r4ufeFwx
zx>A-(@tL|JXha?ZFtD&$)tBC*PRKBXzS+*O2c3EFH^wBC8x-YSlH8L_6Ua3LbiR3T
zH)sKjLKnAU&nA4Ux-2DLu+|)BAy25o@oBESZvj(6J1EC*qpSBE=`~EpgrfT!XG^3L
zIgeG6t?r;%nnE?p&Zu#x76?R;ckrBa&qQPz%%sf;_wGfM_)_i(HX!U_E@Am1y@0<n
z`=Lv(Q*$@#M?fPeMMnt1eX$1&?$)OHuM|5gZe+t%MCT!p*O5LX^w(?woXND#t%f~#
zEk~fRq4=}SC9w>2l5-<Kv1X+#l<k?Qi@Of>)=P+my@fw1{G|?h9nY`XU22YF(Ijk#
z<NLe~P#2~XQp;;)W>g2iV?W6PHSUK8&f0CE+p?XLmBL1O_8E8)kqr?`8a>uTm?->}
zOVmrU+8eY9>#Q~S3~j#?gcaO=EMfG03v`fHq27^7NYODvT?#}K_Nm%PKSjkN`aB3X
znxyi0%AL}cG(*J-4klVZv{FMefQImU7R%olLj~T7|7DOEdjX;<^qv=+e;nESMPezA
z5%zgjjHF~0BcdG;O30H>$l-wz3YTE|6g2~>5G|^{mWFacr6mu?Dp+2DJ4Y|$(T5OZ
z4u?>2CqSyg$|#!>QP%(;1Y187Vbm#Q(BkM9=FIyP2GFWFcBA?cwGW#<#%BtLmtxsz
z!g>6c!zK;R)s&a5c`hT3bHD&Wll9jTtxa1Cs=!m=G~O6;y<8j3O&RWM|K1Y4F%D;s
zS)NH$>2&yVa%>_wx*1=51&6Lc8kef)Qc(~3B-QxSrcSYB$m}=%9dq>O0Pc?%ZyF+~
zYKKzc9}BQ?tKV_Q1eR!We46bF>{7O%T6Bb=|30V#pgtEMoe(iI@mbw!zF2>I3RzbR
zo&j!?&7JOP0~lp=6}}0p7Gl&yzKUQnzVGgEAd@js=w^JDlA17*IYm(RHse_wdn<y-
zR$lE{HkZ{Rj*2XZv5w+nSgY_MB{>+bq@l-&s$myEjNFxWeC{=62QW1uh^Ope`pY2Q
z?+&u}+rH)eeLu-8g2jp^$9e(J2k|9OaW)yKiL!yx<cnZ+=}M=>ZR+izWPT)s83h7R
zEzPK!?+Z{ZIOxq_>N{Ur$z^7JjFSZcdCrncxzqzOEv_t%Ant)`=uij#(5FC^lR8CO
z@}N8ZrWyoebV1y$PoYVUppD-zPuL;@vTI|2SV#-H<NIA6IlNyuJ*Na|v->Z1fdFc1
z-4~G8VyB>nkK^TE(p&-aa6|6$X*3-nU;I;5N=~TJgYuup+MgoPpDz@Z9Fk*kG?Ojm
z;zj0ik%H(X^i*$dXa+S|b66ASLhirDC90DjHUUKIAC#KD_6m629?_w*qI^ag;I9l0
zL$m#8Gz9GYgHJxmfAtfpv-%vaf)L)aVaGf<W&J52i+dCPGa?1m4*g-?k~<b2`1MB+
zs2o;hPqp&OiUJz$(Yfsi19m$vmUae;t^r3Kx;h{bIIu_3`mv;C9bXo6pU5iY2c@$g
zP)k;vgPvn>RBVk?B*-2VWn2Wc&Q_!w?f<at3FQz)$byJd9%w)C>uSQ!G|`Ww=F9h`
z#Py6653=n2+z-T>8bM4v3du$$4CZmX+53P=n9Xg-;flKKGW2o(t8jV?h&M73)(>%`
z8!8Frw*67!v^fw0AgumCD{KWo0gW{YC;QaMsj}L;aSK6&1~FF}ATwOSc_cuE7>3s7
zX>?`LY?_rgJ%ea)DL60rTZgObuIc7D%?{RwMf(6{&5YBOv;mHz#_H+b>kv{ORV_<l
ziK4%@gy8>nzq^sG|4HmKKoT#4=yn1Ewd+&gorWTb(kA^es(C=-Fwc9|w}r<o775$u
zK^^6$(4#Q+UChqWAei@&<&+3KI#4wFgn%K50}v?L)y2w(VAHENyZtf?`QwFyr{Mf$
ze->OB_dQr@G@(;VTzTNLy6;B6Iv{ruH3jjIHP(t{P~0wQOo6U_OmbU0D4i%c3bWw^
zVbd*t!0P}Lq7btw8&1rCr)UYTWA>2{?-2PnB-NP$5>7m%6;P8YEUXb&uRKg@(@Tr@
zZm*hOTvRdybjc`$<29;#QpT+A7&7D{>{PKayVcEVO6?FB!I6^RiI66c>;>Lvk8tz6
zpvV$OAL)PV{2SNp4b2qm^Dhx8<T#3vQy5uHr`5LVF$T$ltIwf05d2cuz|?ifRB902
zX(*Pbd0XBw+WlL^d|{*Zx&7=G<T1i6^@e3@)Mwb7WM9<)ih?T=U4d_sV-N=(O-=O(
zrMflSuDz}W;<+2v#Bllr2y0Qe9iQPWM8dH!8bM?d?)L>k_mhxxu*dbiL-<#l`gZ^@
z3?Vs*rzC|x^M5XV8Rdug0fx2AB1J>8)qO(p1XRwlfxuN}00c69r5@#mb!%R@TKmv<
zIk4@Awak41eJzBNA3-K!d+r@6k*I-0#!c@s^ec<|VJ$HaEP>|m3$Vug0E;Ip)%)EZ
zL|s5}1HM*Q-y$R<{T=FW#9>AE`1qdx2+RCPa&l&}q{SNPkyMrrZ8;TOzmRW&*b_2=
zP=xq!h^b*x?zmK{c<fyS3k*9TaJj;?2;h5CqJyd$TB8Ui`iA>ELWIyxz&WPqQjPgN
z6I;+cTmat<)z+a3GcLj^)C0Pn!k{YM>*I__dTt2ZR8_i+oJW!Vz6MH!RcJLvJE}tH
zJxgYZ$$mZ@fXMCWe9>Id#wExRKAp>6Y;6HTYU=P6)IQ2TOhD{ZJjEv5LQOU$!}ulU
zMpE5*aplAon1ch3v`UvbJ;E}grsK#321rJXb(k;Y$#&Rz!afzyN1z7xZ*+Apq<2S*
zQ@gRG?JkFUW&{TiiER9(#@ULtbOm2j9{?Ef4z&`bCB!BhlJN!AE|djDA#OQ#if<25
zI4NwK@Vh;AShOV#xTmGrk^%q$<*s>6A4{oAS{K}ql#wtWWXt&!k+=juL)lpqeyuT$
zf*8-~vyO;OnL>q@-)Zi<{voI>>CpVER3j~5CMg1jX(pe90`&LvZvD%;LIw*|{)}#^
z3j}<p=4uu);J0h|NvV}lgsz|rFs<Yt;DQn#ln17RUDx8=9Y~2BLrc8Ehg~4M@6J^q
zp&llqV(G2gj5{*DNN>M>i7auU2y*CJ<6_S><RijW{H<&O%nr1vGdLe~DlS7~0(bYi
zZyTAyI;_;RB*BnQe_<K4ea`Bji*rPmp<-;(4u^|5DwfdLm<WBuON3ddyYg!k7b>(t
z`_>u`Ad6wjtIM_7XM+VkK~Y+fQQZq?Axqj%=+i`V5r#Ay_y&%0&%C|8OOZpn``Afc
zW%U6w@KM-WNZ36E$I^BO0*hwDr1eU!8l46M(*e~`(<7H_whd@_X(nm+D43}+?Cc9_
z|9TkAF&sk4Avz{shdd<N5S=ZdbHi)7{FQEoD3p+dU<;&t?fBc8Jr{wqy`vaECr@`a
zKb?HBj(Aqb0ng!#lt|(f88%5`8-)1Yt@IxI%hAyM3_*DF>mxN&-@!r$XX#OWnB#C9
z(pYZc7><)Dco$KT(GaF-*f|4CW60iwTj%2B2^;W}LTU673htF(qULvPSz%>ue4OoW
zA}*JN6*6aA5E)k#XXFrVa7c^Dp6HdHa=#*Ycp<A}AlK`T$<v5!C@YiznIzP+P^DpZ
zM&CNZsb1<~kayJyfn1Dg()!>Fg~x3%Tz%xRAxTL)-S+n_jazaBJ$En2uhr!UI;#%K
zj2zG0&i!inIj`MwTF<Q&)&S|O?fE8aO&;jg;NW+u25m&GkV>SRKuaG=yP9}-hv+2l
zUx(Q|NMpMEK#Zc&_w6ONaT6d{&!xr8Cb}{JrOz7J=Sg+|Xg><LQx?9|PM(ZofBm7E
z5am6YUy8f-BL->|hFBijD(s#aP^DsS5S9h@C;v9R$vygiT|ZAc07u|R<?Vj4-!mox
zotnoR2<Tex?LJ})Yr`z>6B37yr{@h*j)InGsuk(n-0?x(4_jq`7G!qEg0Q*K`Ry;;
zmH&*r{cPbfKk>-?Bmx+)-Fj95YB6Tesu4e_V*}^*VkaDXz<F!11*Z1a+9i!O*3e@P
z(MWG38`8EGjScrX|M`Kx_$f&aqGbj-+P&XHMoH00#DNLwYwt%bLCfzIqW`F0L;si0
z85yxqQ3x&z7#{o#wtn%Qf8-NC-j{?ZnT&K`o|pcNQhyn~{*M>_U18^cy!gxFY@-xB
zSeNm~^=5zTmj1Zbewdq)4#oTayWbj)gc7dwwW>j}zuX7@h_jGOpMgnR^ZE-Ip8fJh
z{=*9{>M)p5CoG(Q*_HffHRWG^@>_xd7%-e<{Cnf|`y=s_J^&56nkaGUpL6H$U2Bmi
z48V+t+aE32|8OIJ{v^35q?0?KzMK130vG>z(wP>>rER7QKny@2!CK9q`7dwMov$v#
zO!35cPLFl3vW+Ao96d@a5MEmM%C!pqccYUa1m#QNR!{)7iTB9j++^FCm!%ei%|El_
zM&K`TqW|&XXj14uL>zm)`^cq6<ppa=M#R~z4TJ-1{(mm2E=g<O8|WtdxTsagY{x+R
zAU$ZK+4DaS!~eF%9#e|ISm%O>L1t3Wc#}0t#+m!R|Lu)EfO+5C6I>hj%P#AGdLWHR
zm(b-Km(!;I>Wb&|V7Z*cE^$f!Hv=ixiL_f6YWbD^$BO-{Z`S<3Tl=@6`@d29_i6Wk
j6Y}5om;e8@w$7VAJ*XXtx%k@#_;dEO`l)mUQ~&=5as&at

literal 0
HcmV?d00001

diff --git a/_images/IT_sample2.png b/_images/IT_sample2.png
new file mode 100644
index 0000000000000000000000000000000000000000..27b0bda7f9bfefed7dec5d4df4460ac977a67420
GIT binary patch
literal 282807
zcmeFYWn3J=wl4|<LU0J~7A(jh!3PO8!Gb%1;1Jvy7(4`b3&9<N+u#t~A-KD{yS>TY
z`<%VcdtdI&eIIZCx~FTZx~f)Jty*jSSFH|K{0PE8B|(LQgTs*iAfXHghl&XYhp_bO
zB`n8H%1s59QDh-5t|%=oPNnEzYieO-0tfdYI6e_sQALZ;=hy>~`ZWPNB<^=IRVtja
zn0clwGAT6*woH(3h9QVR!;Up5$cJX<XON#Iym3j8p@CBQG)83*<Gx|rxBh1z>bm=p
z=jZeEB(80b?NpDWde<XQM+!JRlTytTc;**TO#RUkHy&JTYq&%=@IG+d83-TXdNxf$
zwN_W#;NqF?UBTYSqU%?AlWwKn-p~4y0f62uc({P<l+KYYlPn0?Bq6Zk4KkcD?;fvi
z3N6L9OPs~)YSFmkaL4{PSmBOih(^C8<Wvz!O-SGnITM+s;e7^0AH#I@P=5p-1XP9x
z0q`<eq-JTkxI!b(aN>r36HS09Mn>fB*cAGg9q9WU9qO4J(sT#DVne>puAv`|Jv$0M
z)f$_A$lqdrAD{9TB}la7ZG+kMSe4%fiwr78vLUrMjhCNFF8rv<;D9NE97p;-?i<Qv
z91>txpO{DlPyv(SO9;|zOSB40Ur5wQJc~h9-goAtG>sn=tk?Odd}j~fLZu3o6y`9L
z)U|_)-i%!$=mH>Gs&}6-=%G1(`!cpni0i)b5|#g~g1=A+{;b9^A6BF}&R?~@QFDqU
z+%nEFdgf*yQ43+Sm#PEmMXds^&9<fbd$ux=9o((zAXwvdzN;$eGgN5C8E@gyvBQLK
z;i70YW4KzdT@u_1CBDb}==hw;EB9+9$Zd*?*oDXC3NCDUx^Mj?LD(LP`fk!<S^G8I
zQC3Aoh01tK#t0nQG_uc|<g27BxQw8*`vDy{W$B|F+ch?WQCpwa8SwHMXz0Gg<p?gn
zQ0m|TD`*JsiGmPLKf|y1szwE?w-B))TZ&RMzD#c8QH1~5Mqvj>_7e1-iw|koUv`D0
z4&KH;Q1s<pn=Cu}lE2a`zY}bvEU_O^6vpUUVlLRYXFib3&x{Bmfv+?1J|IeGZvNIT
zAbRWjHiM)9^A1S>B{i^Abn&-m0U^oPCDC*PMEZ_*n#er9W^Hock&xSzP9S_pdWhR?
zNGDQtWTgREZSp7J8xja<LK|Qe2ayW(75ONr7@NLbsyzzv4RJITb2Lp3dZ8pqlzPsK
z`rzIu>J8kNftiLxM9g8N^l-T0s|ExH>V`a@-?AgK%YSznl$QgT5}jknwtX>Ttti#R
z<iSk#`xM67nP6~jpzs-BXlry|acs{=Ms613*p0NZW^YkvQioBOS|_!@ughDC#p~Cz
z&V8ET<_aN73ms`ATytH)S{+&QIC;E@L`3qVXp2|@eV__@b^DdG{e3&J7?yz#yNoJ5
z3GSz00|{Y!CVO0aKt;S;>Wl&zU15|QY+e|ZNh~l5Qzk?kKDbM`7iDC8&REG}2fZ&S
z*Ur>k4Qe+D&<jv7R;xrxl1~FA#PG&k$BslF$1W=~;^Br6()q@S&weP(*3AHBLR53!
zYBM|F-G&t5G2<uWdEn_Y;lzi<Ys#R-oyGOXMKFERAW^1&o6J}hU*FFh&m6Dtc3F)=
zDf8#d&##5XKdlt`3mhaj>)+~4N=$0RmuY7-fG+*MP~vBjD9@@=II}tfE@^7gzhx(l
z@m0Um<~U?83FRry`?-}tkPVe6Q9i8x5f<@VeuvEcr6=wq<0Ix{RYQb;?5B(tz2^7L
zphy31bPNE7uT->DcDHr-$PaXR{UN6z)*nMZB5T;S3A7!n2`mrC8Vb1I`SWyGzqFMa
z4a?@DFz=Wnn8TSHsJ=FL{Svvevv;!}<m|Ge`<r+yci(+CZ#Q8tZIpZ>dk?&Waw&5;
zyx07>fI~A3U5Y<Dm7}<<gfas#SvGn+M#;_29mLbiYyQs7>I0`S2|Z~736vO<beA}Q
zh?Vyzulr}YccvuyB(}Uq>`$K!zr41r9L|~Ds4x7+)bnnndg&sXxNC6SDq5>#menzZ
ze;l|7$-l)wn`s}VnamnnYoN2*HD;<_u$UFX6&yFoGdDdPY*aU9PEadVYgQ}cUKE%S
zSZ<iTn`Ft>LaBld?tQ-;>K>)<FfAJr*HN1norlb;&-1zvxBwev8g;JlJVfr*u2`?i
z$XUn{F&oK__?5}U$)@?C{9euuE-`%Ed>8_x{0ssb{L3a|-&Y4Q1_#(Q32hVA*z$FS
z>jmn!>L{Ea2Zs7t7Y{S$&u8<hkgBeSj~sdJwi~AF27p8q!4!=`#%XtHx+B^n`Fh{<
zAZ4C<FIXJaN%pb#<7>lfJMX_<BSV=zknU5k*4B?zHpVxL_DBzPR_{AUegV45mTYI1
zN4=At>*1_K6@Ml8G?I~HMym7>V>)5#;>=J;V1}UoeC1AJg~NhkgW)TVAP)ToMCHcB
z58w&(W&b!cRo=6jw;I~+o$8;AE1d5zT=i-`XcLD#dZITV8ku$(1O{1$xD#0uu!h@&
zDupBwst}%2PccB>s1xuq+ej|dxRtmWKx7ewsp$>$+k17F>*u$wMjTT#_>PAm<F~R~
z`dS$^R7P#P0WxUyXpwY=xJI~G^j@pq>@DkQrBiZ#<`_^U*QQ>_9CraWu}~L~==^+*
zuQy^gwwaEYbW~od2&oJx2Y$6O2sm-p7S(p13%n@5c-%E2t{@Ii`a0M-7)zt3)KDNK
z%WEmEUOS_#%b39G@TuB*wEWG7N&snpTKZy)xHxI{$HJrwWDDC;kd4VvPZvjMB&(7Z
zjrLgmgyHU8?hCLNSOn0)vd(I$Q`!m{I$Odo#^+|vjF)DfS0^rgt2x;^=Y8qRox{Cr
z=`~Q>2k7&z<#DmL^!?bj=hs5{;9USvWr@RE`m_x#y#ozKwo^7v&6h1#Jn7)H?tH0`
zx7RvN#Aw3M@UDC;e8{f|?;~yG;kTZvA+FV`z8xkQVMUfvl{u5SgMPEJerpm(5Z^Kn
zwVRLU46NKc%sCu8tUOyeOCXmL<T<`kj@Gts9BZtskZJtlIT@JxkSf&J<if4GuS%%l
zy~MiUS=QZ}S2&dewZgJdI7&0~FuHPi7<{O=NuO$2^VGi%y&AxjATuv>EG-f&@U&AA
z(gl=I0^c3~4(|{S)FCM*`9iWA-Ua%Fvl~s=`sT@D+x+(^)M(HMqbEF4BQ+z9@r&P&
z1Rf0De4YXAHJt{9muXg(&Xz)_MVDpw3cKk)fh8{P<HTc0eq2Uo=fe>xxap(<=?*Pr
z)&}z(X!e9QH6s?s4Ma|-XLPNygaVr6W`IV0Vdt*P{=TBYqM_-OQf_Afm$aLq(^}k5
zA3w=Yr<T<&*wv5D2{qo!MVgJ=CJXZd7S|RUTm&y%u2XjeuaAZsnHzgruOGa+AE&V#
zD8qzxA79-U>{OhsS2D~geDUUeZdTfQBe=LfK7ae}Y_~&LZ+SqZ^7&>=|0L|J*h<tk
z<EVb|T*cqZpI2|tbM#sH!NIBEL?K?JNH3{b;3;uhKefZY<FSmctY^u}t?1U}SuoJm
zns${|Vacq8@=@_Z>3rQ+VNiiQ-ADv0jP1$qS@~)5!E(Pg!#1*;v!}aL7sWLTUsN$5
zHWuEkiV{u+2sc4${><)LnVyDien$D(@0a&lUz&$l3Y@v<keg1YF!C=PReAO7dL?q(
z;H)M%D_g*H;gV~X0wp2YT~Fn_S;~=ZIIXXDuQAs1r^V;$&%mM=H~Ei*j~Xw18m165
zVAgAhiKeuvygb}nSo#$lB0LEk5-bG|yM*CM|2r)O&j|P8ANdGyaDf(Zi2r)#Bkca?
zh=N^z>iqNmBIX+$GVI@L*yWai@Lx}(VrIPfuQb9EtPJkGinz2i?5<+uU}9q9Xm0Cd
zP#`}H%RsUFpydb$M?m-If|pi)a{_CB+Co*+NmE`<z{u8`&A`~!(1gv++U`#~a6mT!
zSkl_W$$-ku+RDaJz)gteA5RFt(tol6G*th1#K}^KMpIsqO5E1Lgo=laosFGF7?p~O
z3g}>LDxfSO^{?u%e?m0oPEK|L0D!BjE1N49o2`QxfP<f(AHdED;N)b5J;CY-wsA6W
zW3_Rl{bwWp)sBRTqmhG!os)&F4b`7^4Ge9aorGv;{&e);$3N*bakKb$Pd1MKdM(%s
z0{)Z$IM~<$|J^pMD)3LPfTD$)iItXwg*9x<V0{Sla=ioo<N5zj$-jI2kD8kQuF1*G
z`#-AwN74Vgs+yyTgSf3VtWPK5f6v#y`u)$se^mqm{=E8s48=dk`5(Ekffhyu0{(l_
zgi#|4kbGeCNM<3SpbERgxa`jXZwvcj{O28(hTq*HK_T#igA;|5mUyq~27i!>+@vcx
z8_0nDiaH=lQ5-*~_BTQ>ws;u=DiI=H4zpT_DvCV(F6yhXWT#y!Y-$<aYQ$HPRMo0t
z^3-v5E!RmSTEl5VjjpaEvmVz=_vsHd!?P+|gTV2nB=B}BTQold0|E`zU!A|1UO7dt
zea`=2CaH`lA^JBb5S|J1Ep3p`tG_u#8EB}w+CIk3g#7%g`aZ8d*einnif-833gEzK
zcggBhSC#zBCPMuCU;ORR2g7JD$EAa2z5T1Hqe4(7Q>DO$$N8Jn_oIYph<_?$1-{x}
z)rZH4TFW^5yE*rH<wOIc{XinPq#)!ko4`YkNB%o@mB_$_(f*l1#!6D{Z`d8yYYtxl
z@b8uz6+&u|4<53hk|FcYzhoP#uJ@aNH|Ov;I}$M3XM$hrDB=HQ6R@R@^*8GrHqHO%
zOZ{EI+KSNi!n*0ukKBpB;d<Ch`+k(n(5bbtX*`5lPV(`>T8yF=UkIf1n)!({-Sf9B
zRf2MHI6p0P+KF325|*b(cCg&qn#^Z!SR|!OiTig$?F*$ZU-3h0U{cAe{=)u4%-(L^
zy4G1G|D*L}(7w@MGSBZR#I1C1uj^kgu>RHP#m?waQCZ8E|HDRqShDec{^g^NA*I5I
z3H=gw$?=zBWvBBWEQ+hIMEF<ZwgdT7|8`85gW&J#|KI-qh=~6?{(sMZKa}lHLIEvy
zTxP??Ox$-@hx-jqyUOE&?sec-z~_haqx$VM$7Qek5>ig%ARUX`_#X^*<cx3L>X0*X
zkP_i<#=dNipZ?!l;J*<O|MYsdU*<Zk>nJbp79n`VS34?2I(ZR6VS|a_y6BuJ_;@kC
z7{{cdizjJw+C}KPo#s)h11m_T^&w}(`~RW$K8AFs>jlS1kKNqF1#v8)jh3goeO1+F
z_e+O)o5m6yoBGWK1@AkR+WxmiIl|NEgp?i!9A$Wc+|9!~Z@auZFnFCJMV{Q?(GqB3
z%_z8T#%mCi?!KW!c&l@3cd*mq)Cn0Qu~08evkX7pC;i_WAVpT+JXJE|<Us|I#|h7i
zQP5G-`7m>ls7pUnew{(bxBA<y<dMtW@gC>>sggOj+C4|3?l60g!<s}L&&$by?p(DO
z6aAD$c+Xl@rnlAiX9Ivxv6^9l{W_&=Dg004kq3yX&`PcYbYZ9Je-p{jic@cUxO}i-
z@(2?~1%AKCN{jXlpEv$jwM+&ruDMKC8{f(OQR9dXewaSmwr$I8o{r0`NykVY@MAL1
z&>?i*X31}jKG|wbDvW3}Zq-UPLp$X8V76*8MwV@tUhcah-dA#$nYKmy9&*WPfjcY~
z%d8K?x{(w@3uKOiI9^*`s<Zv%7BDAN^LFgihq#2j`l(mqXQcWK#5U>|;Im4vP@h+-
z?>{fGoj2Am0M&qnp#Np^A77ihceN4fLq<#Lw|*3Utq9>B0S*aPkV-}iKHaP@-fpKK
zzp^w_AuHy2BGk3>6<qy_yKu2P?kNC(&KMs3uq-`oQpRoZ*MqG1B1800MjIc_1{Pn-
zviv96LVw7P`mzrrTlRreUm`teZ>8r{Rsg0x3Svad*wE^2N<r}o`~S-d$vhqLho1e(
zW}?1JQ3+K0?I)IM=fgSfOFb^#rnL~}g^C`Eec^`_amtk8JU&h9hy0JJdlWEEzMiuz
zt4|a1Xi(GmG>Jj-uAnepj5NBHt?_`f7Pi{)o4hV3H5NmdOEqs&uY>#5mm~!9ybt-)
ztEECzbw+BWG}o44c4^ADVaW_0n0Z_ro|yK(sKQ_r!(?`bQ{(D4(uH7#w4Eid`@?oZ
zf!MXrC8y6@$+in->7H9#FdaSJ=z5G$HS9#w)wZYHIMt+<TIT+f3<iP{XxDYQwk2T+
z1>QxAq+dBGv5Q}?85$%B^(R;xX4+X>f48N<$lHIj^A!bm;r_Vwm_Q$5A6R3-kJIb_
zt3WB6Y{~~2lk9t->zVk!&1wN|_`ZWwfm?hqjyX5NPfKl#8sRtp1ML(d3WTk#B-^&^
z>VY@oCq7*5Pmfrax4k@yRS>qfLcSl91=Pc&!pxA!^TUzH@_pL8K_K};J6d8ok@s=)
z^@7OLDOMR`uZNKPg`F&GP292=acfz_-rHK(nz>)r_qxq(JZVQ;%uUqWd%9nKCI~Qt
z&4O!wii6DRoN2Q4R?FQU^L3f`W04)J*I2WMYb{I=G}^X4P6?g$(%NkFMlZRqBA~IX
zcwJ;u)?ZHRFAKpWxYro>lOEr;@af&d`N(2+q%a9vA{p<Jo6&TUMp?}MD#Eb{jK3pT
zWh>J79Jk~^M0eN6$3Y!1sX<BMT+>PELg{&-w7PiH2V0|BE}>u2?pOS=7IsFnJ%41R
z@OxZYT|XXsFSFJzIN1F1fwvizBn<c19wAKv$q1|W1emImS>ZKGC6vIxe=Q?^&<hjm
zwa#GS<)QB!9Yeg!5Me;9KF^|)N_u{%UL|FCh{uCqO<cu|$0}s-JIB##Af?N3%Y7-p
zKTdv+k;S#Nyd8D5q9J1PE#}tHF6<4j)jy0Q$UjVda{UBqDoifKQWVZx(lkvnS@>fq
z(}c1pUwlb(HetOab2jD#5Fi!57lCmfRe5UXd~#{a(5KVH{l&EV<2{%8{ruE7Q_GSn
zA^yu1<{Pt<8v2K9EjNa)zu=MUny==HqSN=-y^O-|8um-iL+mBKJU?E$$^x350)zD_
zaEO6!<;Q|TFr@`{V-S(Z=y;vs?Y-=IjlbslaN6Us{~Uk@gYSQ=Omkd>FpO@(Ox#!9
ze9*ZsML*j+8WZ)s!0+|Y0jZF)J{y?A#S|$LmN6|OQEP-Ko{2WfK{1KVC%MTvweXq#
z!%+2gf%U_Px#To1e^fpGk(ro?nso}MPhDXOqcQfQKwl=Jf8|`UR+Vn8&0>>tW4wmm
z?BkJZ>v@P<Uv>5iA;^s#ugyZ8pCk|uS?NZ{t_|tYr9p1(6O6NudPJTo8n5Oo{|Msk
z?nuh?i=P>GL^i;S-{P2?xiA$3er~_CGZx)WGE1{JO9SUMK3>f?r|Uyb1H?0<q{X1U
zFpVX+2%BIsJthGP@J94eV(UXL*8LxqT6Usw@$1<u#c`P5W`MPpr9Pl00=7n%ncug-
zD(P8ZOEQz8ua<YdmaVg{U7R{C;onZF%Dc%f;5O_GiaB;LxIe6Q^}gw#lsU?*4vGhi
zdGyCI=`qFFm0NT713m99PG$U&DcmwNN)N_zW!oa9Hly2-In&-%t$n7Q;cl=(+LzR<
z%(2N}QDW<xMq{mx8WBnNx+?@IFuFu1^3Ivlx=tv8@U3GPgX#i{l@*B5jvXXW&L&h8
zoB>TIei~HWBK>gl-r4A3ev88Q2kS2L*1Y<{>&NOIdL?V0mr_dWw@6&4w9Im+cK3^`
zBIaQWcLB!y+ls=e!k!Hpuc?MLStNsRz8giHqOMb?F<Nh{X-sL41U3=1kCF7qg!r#7
zL8im&ymLYi^y#Kc<`--dg<PB#w|3QcD|nh+r&+J-F?c98?pS}<%6E*SQ;x)T-R3`c
zJwDx3EpX(M9Yyf1W|#))6@8<|>UzINZ(QwAYB^onvZ!TUGub4kaa11xf$^!rH7D2M
z3<^VvqO+=P{U$+A79l}JwV8$jQ1gxRK}F9Ul44o&)sIG);q^e{do@UXTt41v^d1Kd
zb>j<zXu9peOoh#UNp0!lCcS%hc4v0<2d(dJSF>jK!oo)6VHWHy)YGRZ6XQ=#khdY3
zuijz7cl)e|wrX__iIH|!7umDy8S93g8j%_0U&xnV47oX!l*0V|Q^z~EP--f0hIO55
z9F%Xwt)SE*xv8M+e>h%5seT*aAy-sOnh_$sFJ9{+CnzXqxtTGF%ze9)dBWqgqbSlg
zUu8DZUZnYxZzD>4p=v~UKVGAJ&sU6eemCplVq4^SR$<{30hNhy`|;4$`&`DWspFU2
zLguTW81e||5?QwT&deF9I$|RCy~6ypNDoO$dKEz33IZ?qY$MC#;WP6_s_WA1qT70S
zp^S?G4M8qRd#E8+Kc93@jsC+Lp=6?xsK9y{7X<QM;xhu~Ko~M5q+DbfhT@yf=m421
zop4hv_eVM^4_HDMie-EsfaO1lT)#39tH%i`DM}<o_ffUqRHZ)-dM92hz5BWevwT+7
zwJqps$3F_*2Ak7w+lTkkp0b+>*_w8>HO<yY&5HS=ORDsrEI&WF+S=nWLo|8C;1eD$
zrdyvTIeKx(2n7CEsAc9eI}jCHVl)F2G~|A4f@{WgMsz)=QAw{x&>muOnsO^k<3Z(7
zr1wL%h85=VltZz)J^L}R9g#kyixJLnm;0!kAoPr3S=YV`{}Cs&vMt(0(=9im`wTn6
zZQ~ytT-$ZEACHIpeBN1gG_gL+nHw?-pOg$Vh}jpo*3I*oF0(6{h6pWg@;=R8rj<2G
z&b4<~+nud$6x!TpHr8>m33Or1-aL=jWB85Yr4J@|aVI_vrJ9ELxAI?!UhmUhS%mOG
zO{vZl$vXRYVT$2s-;91=(F!vZ5QK&cW^ot24G&<Ddr3p;LJtz`G9-P9;9Cp&nH2$i
z=%9T1UAxJM1=Y~71!~%TZKqPNy<YOLy*~DSp0}=B{dR~iF5WKRid7%vz@Q-rUyL&v
zDalkL)EMzPQYsI=lB=Iill<#+j&$tXqxRpu<-NyKw2HR(<L=ZGi7@+;=r}CWvI-Yt
zy(Da3b$!R|tG#Yo>*bWrW0u@Tue*zQ0m!tBKE*wc|J4Ty^GBy0Mr-?}Q^il?cVH!b
zy#K`gzI?uz*?b>zh{)bYACxv8h4hRnbaE1hUJ9|OEqY!rdab;9!j|!5#LgrlT7X$`
zN;8;YeR}l1N}>|PfqjfE{xswXM}BKm(-qx;2u2R=CM7jf4%7PH&DRtrlEsFZ8{fuA
zQ?vD-@1|GBpL$(q#PIjO0b^CuQ2L$G&(;b3MePakkUO}c=2A)YyJL9BDBHH)_d2$t
zVcSY$x3E>StNzJZjWQv3XpOo^&0hxwI3C}ql$<hrWW0J)>x^lS_T(tRe+siQt`jJ4
z!#Oeg6Ewp*IluxFc%n+Z?Kq}*fr>lmY{$UY4@2uSgCYK%00oy4HK^3w)sofz5F%%U
zL;F|3O{~DpG}!2s6&z%AzSoOPj0dhCiC8^P>U6uM10(!soe5B*_=-_0$%laFgRn8L
zl;`m0<6HZXQa-FzRHSZ??R~x_k$QEW{=D!<ebKd<>KGQks>J6yHKr00t{X<1*k%k@
z9$7p4$!uA&-G1#!rM=q<6T9h<heD6T&FK=h=i#-L5$~qfdGm9OMZMO!X6*L*t-ZY|
z624!~HAY+wuFGB{CFhTsM?dG;T;`v0CcfBkeXCb8=uhtYibY@l+;Ww7zf>t(PuE&2
zk?prw{j`<I7bK~(!S+<{2Z8B#ceEMmF-D9mxCnfcxm<rtf!={nt6pY&l|w*#P#gJe
z14+-6B`}iiDE(oBVX|o~Zp1ho8I9e1JC{%MM-zJ{#JaMNK9|=yR}$f(t=#IJ>EL&Q
zbn7i-0U4mSKSx6KZnCs}Ik$Q3BX1>C#G3L&0BrMs627>NGwl;hL&adD$?aNB6!B?B
z4yK{|d-00E_q`}dlp@b%9Tud3nY{q4J=y;OSfgZq;VJ%@oBhiC1EbKJhe`FdOL7V;
zc&b=Il2Fghz?cAzR;4mnPK%LKvMpwNuu`o}M8zlA$}aCDtfW6|5J7!am;s9P#W*D7
ziCBe<rQwgcOSpuwT8mPgWws_VSgGE3-egAbn`VBGjgv0BH;yngt%CI%njTAKFOB52
z+cX|B+BH81QfG{r+b|w-Xc}!A%x=&ft!g}Kur%%u4cHEn4I%{Hta#hibS|^x;@t*#
z^MG-h&Oph(&REu#-<cF-u_qyz8Bbw_K4=?-bJM7tF`Y4&0hKo_FhvIMJ}J#u2iRe@
zBwS}rRR$=4%Pog0O?09NIq1^ozB<5%XpM!*ofhv9pMA$DCP+hJ0Cx)A#Mgu+wObia
zDTTIMtDpE?x9H9gI0oFC(s5TFPM}>|@)vTaZ}bJ*{lNbz1{Ax^|Cm2q<7-QaKZl^=
z#0JKc$^&ZSEduxZwfE|?M625)la{?mJZ}!?)t)aueSX^cMSERxcdPwiGf|cObYGz0
zs6u9?v1)^eRZ$>ZrqSW8{!XY?S;kX-5#QFsZ@W6Mmp5;PVm4uGRL8a%-)2|yn~zq{
z8BpH7d=U3=-C@d2wO`}V-$!h=7%|G)3hmMqLvM0*kMXoXpq-e)n8nAX9}k)9CQD-;
z&A0ea!6P`ooe6<F*oW07+-iTdA-mI!?-D{JQaDfBJd=q6lVZ&qhcGDy3I^@xHnsdu
zj5H*lA+AMM?Tg*&yq*4Z!SE3cliA3s9uFE+^~RoudaWcJpQ1AMlgn(Etw|quv6X>+
zn3=t1*dKiJ546e(xe2{mh5@Qw4X>w7fVcfA>v(w#elglNvkFMtRI&x;K9N;(f|I6V
zCk<5s*9q3ZJI!|wZCG9_a!mcKEjMdV@pK8A?G>ENM==D*!LmO#nIC%IirZN!$pP0f
zWlgPu3CO&vy6o&;yhOJf5xUlGd)8jVss&MczH{4*%;VcBg`uqWnhs3CLEBl~y3Y(&
z<U=ka_O)FEI(2~*?ge%z0}jRgCXK&VW|p`kVX|ysf93$zkr~B-c9m5$sISOA5MM@J
z2f4QGC21W)nj6i-MbGlX7VnaP>VqMLJ=T*6t5F=5`imZNXR-tl|7=J(zN4MyG8f&(
zN!6YaD*^K9JB5!opihWpb=fT)GHQ_I)aQq7k)=w*2mav-RRPG&N&DtMIY$gmdp9Sh
ztF4-<xcv9rk(tNQ#>YD&`^9v~)%i$XeWPvja&0&opdkttRu#Uu9x1$;=3!dxv42_H
zl<>SJd*!P0V2;cCIaL|0R{D8~qsSL);kMl4ITdSRqLA{Bd!SB|P^S}P*0_A%V0%1Y
z-gV!WuY9K_@YthM{s()PeCk>BeaG>zy~jaok&aP`T4>rTs1@nHfwgwN`ZpUL<@d`j
z7T2Dk>jvAFr|#t+FoS=gT*&2uK)RlH%{Th!_BqU@%5>OaprpEZ-hTQJZJtNr2m5Sz
zoo!Cn^|J}$TH|jTGy-7`qi#9Z#9JYctJ)_Lg0HU2W}2&BbA5HZ%g#n=-m}fZSMt|a
z$;aOVejk-SzkXaKu`JvBW&`!Tbxof)F3K!bSiVIq+wMZZ3O<&odPn36@wmi;o|>&~
z_Y#kYoM(~pui5jiJ_eiOT*}r4ZN}yfIjrXCU|m{}dJCs)r<VD(+BQEyJ+6`nb(g))
z8{5REdGpS{>EK6#HObH^3%WVUII%0S=QC)p;~kB9A})1gD&_Ad3RRkQL&@r9O_Oi$
zA|V8><9YJu2^JOF;p@WcZq`aN4;<c4FdQwQ^I4@Xf_s`;{2QK^$oU|bBtdU^7Wo_*
zRLrPIwtX^RfwFc|T?Zc;3~)PXjr{{_B=ou{ABr|kDAB?-Awn=O)Wg)U#qsB86WA^h
zv(8Zm?-vu|r2+RrW{OjYq_)8(8FN1_!$+q)s^b!!KR{9P&{%t>0x152s^gsP5o0Mo
zy&C42yDO<4Um7@{gF*6j7+%9!CL|K^YO2MH3~;<N(0twVh2a(pq#sezZ2ew>`jsYm
z_YcSd^)oFq659Bw$0|@uxZU~g-F|7K&CM>%(B<hNu(3~Mp<LAyy6q>(0?Xb^q-*{N
z6m&+g_j<gX_Vj3+P?o)|4BdQ?;v0Kf@;JWK(}&rWEn(dz`8T|Uwh24~=kdDDw;?xU
z444tDOBF2r!*G1YtvwVTEDLOXSm2)QlS~1H-sEhl@H3NGYlc2hZ)tB9%xjyg<NbK)
z*_^T^1pY;D^vHUK2&Uktco~PwvacdlUW}D569JSW2CoJXLin+cl>=dD?MI8kd$2`y
z*Zbm}Djm4ZuP*+n$Z{4-+xdT7k>|d|M)VdDoMQE#fdQ#l<xV!CFPrhTb=Wr_WjrD+
zwdUDdx-fAkqU+qqJB0R>?r30+bSLJL>X<6Y#DEHhXR(7;_W{r#nyr0{dIGpc<kZfX
ze?<9uJDwtk!5ZCUJ)C#x#KdP&n31+xT=1wadw;&xR_WckGY=<``pGPk&za1;51Co0
zo#!WKn8(3mHfR}!e)Xv%y9ao|gv2b>j?jXs<2RzDlHTc8Ndi$NUkb5eD=B=77${CZ
zn6OgXxBgfqL(`l{G-%Y{`>;o{N1&s*^BHY>#Vu`-*pC*2m^ZgHvlP|(lyiV(-Bs{?
z0_LHz8w0o)MDgJc=XsHU0C_)t7R419X=dF)v~5a=GUn^@f_byIFf4uiCA$sTp6MN1
zT!*QnR=;@~%e)Ll(7@RHZ>A~HF0q_aB(yJLp<%m#)`v4X>7bO5lt9|Wr1W6HZYL(!
zHfk=~_0CUCPs8F{rJ&NjIf8M@?#Pvb$el){z%bt?!_bXKy16jVt67ExHvTE*BGYgX
zkV<RsN1aIT?8M{7i*EFp@Id!@beS7S{FK*LD#O}p;dRr}6)n~*BLGYlI+z9mBHU|l
z`ktgk{M(rMorL0q+Vx%_)QXGD)rph1H?*<NFkB1qoILEt^JO4W258iLJJt0H^{bUi
zSo`qXqT)5!7UWa=dd=IoI2AI9Bo?PL;pYcO5i05JcTPV$kk6IBy{|JXPkx_NXGFzX
zO4K*iZhHFL^@5mPQj1ZzS6}u)N14o6F$sQ_{R4Kp^PEL~{d%N`@WTZN$}HK(arAv9
zgc&!FiDxnvw6<)%n_^v(v{AT0v?FRk#TU0UMt|Mm8G$Z_+v=`cf@%Pv#d0apf>L-t
z*w^mnCEp!(qKNo+$9eSI*~!T~@9RIKiabVJcgzmc9i0UU{?Ri{3N0UO2146KAvB36
z-2(_FuW}=_Kb73R0eqxRu?X>JF0IR={ER+gXPgxQ?;K%IsPSssUC)`$Yf^dS6k^#R
z#e^B*vH!_+M!2VrDK@XpBvHQ`gEO~f{@EpQ-v(6+jdfyVD5AyLHtex%f6+|1_PNcY
zf!uYz#j5(a^_==T@x0XTINW38a?cu9AkXdw=2dHyUT3HWTk9OwZyoI=R9ZU#npgd3
zM6Tv96?up8_FGaeb3HCMc^7CN2`gKbB@1rlwGKj(pAR0{&Q`odx|0p(pBl;c9^WfP
zwf5+_|872?@;GX*0yqQ0NQ`9MJt0Y{@4PBW-x+5J?yrrrSS>Dwjzqv<fXmN&!!1Ia
zIkWV}0h_xj9mhp()b#fThbztA`z_ELh{2QB<)+4wh%Fv-FChgq*8BM!#r4X<+htrN
zef5s*ch6;|bV*rjMTwPs^l`GSX``Bk{wOm?8hDiHC@|Se!+W#KuRP@>S!Iz+0jTqe
zKu7zyV!8BaW%)?IPJQ=lwc!2++Cpz2C2zoph0W?ysGIt@+^o}HT%pZrF2VVe_4zuh
z4873#payl=`f~D$S91@JP1a^Ucx%_Rzc{&J-P_x^?$hbaT!_CMDp!45Ct*LSG^p9T
z4T;cglLqf)VUPRu(P5rK`VoWn%r^=x)PROM1QciC`MJP`<>L4oNlLjDx>POe%UB4@
z$M;L*Fn=YNY)wGV(@CHR9yHu+ECR(Y?m&st&;aelrQBKu<im{O+{b&&<KH~Uf%=RM
z=Z`WG(z^7U(Y_XmHw50&(`EW1BEAYlW4g^O;}1W#we?P4h4^m}(~l|Ai}6q`$;Gjw
zNRZGBuuG?H4YubB-tA;JE90Vqt25=|Vv$e$Su^V+QzjS#{a#iA9R$vVaqmslW;lYy
zJzA#qJiwA-1lSo(@1I&=TpqGv1f4W_|H_k-EEeN|nvlalby$gQ<|?x8j1|vU5n{qI
z*@jGcpie6Ye4!7eG<zLk_nR^-F_Psr#Obp=fsBq^9Ov$k7)L+1MaUWe_rur2C~?eF
zRJzSI)nn&4+7HMrU5IGTX!dfBrdo5;%of*m33U3=iG^s4+kAThnqLDP7<((Jg50q|
z<iCuazOxT;Q9QjMexePFa~+F!kko{2QXJuDbRhEYdcj=9IbNG_Dj!ds*-YAs<?6j3
z)+r;Nc+qGIvzhWIY8mVGAg5h){qk)pX_-rs{2>iKE^u@b?LCNZL-$t}&AZ;?Ih$t_
zol}u*n60%rpON(=Y>417OK<07m-5|i#Cm>e6FkWY`xVF;=D=!=64GUX+W}KRc=$>$
zzp51&nTPfso25D@Xkh2sWUyt7DDayaFd2k3vTV}cr~n3<=F+hvSbB<`z4dn6{KWQ%
zjEVMi#-gBKowA-l(J5x;$$sm41yxMrVen&VC+O>Bj2l_MJ(j@Rw@qSt+vIqYiijKJ
zdmMO~y{%!i%01`SX9QM^iDCN6`KthM$W<65EXI0l5;wArQfmOTPzDtSS!^sgj$B&!
z%5@UGk~8=K&RNQA365;6a9gcgk2rr|(q}1_YB5Pkyy&3p=UE5zqOq*y>3vkO>8HU`
zHJm00@xSap=%g*<K?f7wBGM_&xo@cT(>t&k>Bn&qphvS^!2fZ*_GF#^L;dLeQ9q(C
zQfc{V;t8f~-`vk&yD^t%^dWO`vSZk}e0zr7vjt}B->{-CmA)A2ef8pUc3t4G`H&hc
zE>)Q-OU7xr(6xVD%;~CA(={O*R^J}VgGkqQ<HxY&up-|KN)}80_V9hHviB#r9}lL7
zumNW;O#<qaZB-~0DOEJ0kdt^bt;1`#o084aUKtc#f3JD$73b*}r=U7EiJm<VLbgzh
z1S;Cj?F#+dg}?IwQJ`dvs>t~mHXd2kVa%9g^@i;%^0T3Bs4ow#i*!Zo`}cBliqtZ5
zQr4(q`z#35VwMMFiNCe6zg!~@3l@CIwqA6(<3>F=j$(9@{?T>e4#awDD{UXu_JTfa
zN1}^7c1JFwi=6!Z%gD6|c>0G|v9BiI^3=rrQ`bzE8?3b~RLPBT{K6sI#Wo`;?TJK>
zp(|nNKGqN@Q=>+2J^MWU!z8o~PaTkz^Go95m+y;5QT!BQ&0na!uzf?dtgnfZ_T{;d
z*qSg7eyOyN3CQ$U>$n?o-&LgiX#pJIz8neG(kVy`h$1+!OuOs-Qs&?r{IJ4&VmnN>
zvWC~sUkYLF7oGhX#;!KvakIM7k>-P8GOsj`T}c&V_I+0!+t&K|D+ZmqTPvSoRinkm
zyjN~#qh{I^k5h|Jzc2Nu*@=&nxLWLV(C;dk7UdmVg?8D?sAlpNY*iBOyFIV^wvU3n
zg${>B)&WajR>c(vYbQs9FPqjMrEwV=UtIHzJ#N+MH)4Fll<{hJPAw|Zas92G=fkJb
zC&aeWbN&n-^|Vxnr>wg-aJ>zfZYZ*8?)~J@rj}REJ9_xIuV3fuKYeE1({UZ?6e)C<
z)41DG)HK<qAo4WW40JE{7{Vir6uK~W9ci_BTN%c8c)pvbdlOUT*|B)YyL4#Na&&An
zrm_11inVmE;QH<KKu8~OJ)nQMm7wXsVQXy<t+<(qyMB5cVOUg4)IF}(by-he!0c#I
zx$6%_0_qVz{=DDmh~&St8A<it^r|=>(XJNj>zDv-Kdu_O&Zd{Sh_E9K7??IYv?1LS
zdn6;4s~y|^pxM+vq(F`982Y)nHE;FsOnJNf4BjlN%g0)NE|@<A-Q(@`BN6KTYQAVw
z!Do#MxYkjzkGcxt2}Id|VX&JnQfji*UYgf0+baeGm2oeL`tcdrf7561`DY;!Y<>TZ
zsy<cD_iF>;jMW+umyJ;mnJ#5T1k@t`7_94iNs8B9{tcfjR;X8*ucd?XjPgaLTrpR@
z`25#}uu78UFr90Rr!WekhJv9k$~f!_%ujs=TEREiQbg>*A5vl>pf<lV-g_au1xGGl
zR?3|hrCEZzXc;lH?0#*$Is@#=w0x#P+Z0%HzuV3Cm}MnaAA3zlI`*QA4v*wJk(pn_
zb@aV72#?Ilqp_0UP3+UuS+e{3b?b9;qEV8BaU~%!5;!%O{mz0>Haf5O9U~#lKr{eF
z!%j}jkTs+Q1#{p_+>0|bU_bxS_=p%2Hi2e>9}?;Zmcce0DPu``CasjO{>)$zJ>kj?
z<_+)mHb3e(Dgz(GS7&c5LdA)Z0tDL6PdczXBwu;c=08Va8H7DET5twSrHqlG5?(})
z>rO!=Z>4PVpQtH0!IE=`Fn>k<fj%sd!wUyH#GKJQ;2}x_IJTy~&%I|xP*0s~_z8J4
z%Vx3R)@{_AKC(Tj3#=(FEwHD%hem+SdGzb%1#S_`32cNhOujy`S2{o#d{zi?+gtA@
z%bGU1=90pB=6bh3;a!+}C>mxRWI)>5-`;Q-q}PhSfB3T{g?q=c)Aj@V62m09g@mlY
ztQ;tzTOo|+x*>ap;Hh^v!PuEC;hzP+`CZ!KXtYNn7+?|bs<jeo00}%%8u=>l4SM~@
z|BL#-$dFMi|A-`u7tI%A(*{E#H?aXGdWQ-@3eBJ~F3ZJqkUO^59Deb4&fx%be+K~z
zbhAApya#hab6I~hXC8Ge;*M<LH_Mq}J|^s?wE0R2A1fvCZorGq)3->TkjSe9;74~V
zK_$#y5f-`bx5Od)6W$LM9j`bstZ71u!c00@<pbDZ%IJixr+Ab0NJA$r7GR#d8Zd@%
zh=0gvl*|}dsTo%-u^Uv8u@aT5R7x-ZdDyeG!p^6nznl@-TgeV&fFemAv8s2%916sK
z_#g-kSo_?}DY!b4jUg_ajq!#x0(kbp)`b1STM;vB2}Dd$AM}K{iK6wr<zrPni^)5z
z1ZyTN#qDv*+y|;q#gEb;W2`-mJ=qPCxCPA*K0qI!m=-yYl_oV)v_{mYC~y?@nz|S`
zsD%ZL6nN>er*GU{hj!_zp47-XZ`dL~Yn+mBi#?ZFhopALaw0X=c-l?cTujK~(5+>c
zCaFH8lGm#|jg7S-+ArcjV9j{R<;hK?cuiYoFbGc^bNYN!2i61o*%>=)%|`-_vy0C}
zDL6eSP75?BdVhR#*Z+?Q2t;_?J(!~x4s*`1(0l5Z38m3FMJBDr_gBoe6}NkzBTt{!
zMM}w1q}5r?Mm;tK>6;KMDnKQ^p5*bV(e<J$06fBS77rw3o1ucabRj(g!>iFDAW|wO
z@bXK6(4~rxDt%m#4)qCww5q;}2{S9vniDy02AI#5o<pX^LLCb=okJEvca2hO-J$J9
zpE4t}8e=9cPAd8;QUbgc?U&D*l9Q{@3(|^!_Ll_!5?eo6&bH?_&D5CAW~G%+)<k3c
z2JzI9&i;ZJ_rHwBV3dNQa#W(E#Q=?7Xw$8ng?LPHn#QWrt<qu5b{ZSws+&Apu@@GQ
z#YvP8256s4>!pa<S!GqcY03HYqbhhd#>~1sV`sFwQd8kmaDL5%qol8S03}!J<GZ_)
z`P!nV{l;GpqrRXcoy+t2!=@TD&z(_ZiR@*M%AMw5{broU%?8KE#{t_r$I)i&ynOxp
ze1J#mj9KDNGqI6u8rY=WJBt$ilgOOYez?pLl8DsfwbFBVkxgUW0oUqmLWG%~TjjN7
zKzu7@iNNR0ojE~3_|yF;8j^R<#|MRrNCw-;Af1IWZy-3?dj#eZFYQU^KdPq9=5HKM
za)0h}rHEA5HK{j0yhN6W1iC7`^O~}x8)@cTO@euruTGaay4qsyA<PN{)CV;i!IsB$
zZ?xS;k+K)BDF}qPT_oP!Ci8~2-!5HfWZhFV(rLQ{97|djS7{@ToTT1xShj@RBjHwI
z?$#@TO6!mIM&o8}bG>+<D7LGzcI)GnNmZ{6%7d4B=>;FswUR@uL`51lVq|Hm`U#V=
z6eWa=Ww7X$xsghP+y|l5A#xQAmG}#iBQuDu#xjJ_OuLeK8Q4h(hNGTdiqFYeQlri)
zkf1>tS>sR=)M@MFmR`Ibnz<dBVR`^<GF9ZDn1nH^u&^!<)4o16glCo_DNBV8fO>7!
z4&W?e4OnJP{XY4A<os&h)^>L^<{n(ma#}r%rnT0_Al_10GDo9BhC_ER(JKTk@5}ep
z#+FZF^lN6~l7MDpZr~&@gvwa}D*KvxS>8$TBx#(<O`-K(=M~^2(U>D*-7(pN$Og4P
zy#GjBrm=3-KDv#Q%wDi{+Dkt}a~D$njxKM)wO8=sUQVfG!%OeBKTZl#7wuuS3L16&
zpkH;cdz^nM7)mjG^J4ZT!DSurEM9&E3?{EnXh5qc20djQo5`d`GdbMs_W^OS9249+
zANw3^LZXDd=^m*7$ywr|yV9u&_a-6II7Uc)4(1_qUMz;N5X6sO?aw?#vck(hkj`R+
z;SLJUhu+E?ha^1-@+hV6DNKdt{aT-Am?KD@U_BkCFQvWXoj|tWY+)(@LFdJcQQx$P
zBo?DF?oaq%ypUUE_`*a1gBL14ubNhDLe8V*r9eE<`ay)t`4Jzd2EcKV#VjbsQ&;)e
zqXG(NrT6>tSbhpi)Xgy#T|z26L2|cmI88ozCGb_fB}zJz`3ii!#Jfo*(&DplwGpd8
zuw+2yS7{{!A7IJDe_WQ-uA-D`8xr^=aE5R2OG%IT`u&xRqt;sV1;Fz5!`<|cR(Wve
z0D9g|`dpd{0~e+-nz6EffR&s;5@)^wv&tE4M}>F^?|@bX7;H5vJNnqP;Y*|Ne4VWh
z=HCUCR4Sc$dQFpjE4Kd5%6Pij7gtWJlEaGn`Y@A8PBIU&H9m+Z9D2k%A6abjj(NZ|
z8hMb>PXVB^a`xC|+{)AY0;&cv4$W3Jd3jge>l;sIGDBb+=lP~+a1q_weg{n%v!7pO
zEYfOJt$zN=U0tgNhf|W(8RHP(+`zOC3akC<!bP!b&%<UA%1*1I8E&=u@j-nRXqm?7
zrn)V?bO<7tWt7OI08`l}H7~gt$0H-WApiFVJ{erEe^rXAG(uIi4d7<f3-Di#6HCb4
z8~F&zeL?*GPp?#tze@NYR3P16&9~&fUowf9eIoY{9KbGD{H6n6!wsDhIvH5FT3f=}
zPFv5^iVrMa>tvCZfQD4AQIg-ekqw22b#0%H6kn*GUS5E{u<G2zBr0oKCa}JIJ(%@P
zV>qLj^Av3x7UIQOX}v}BGIwo(-!^CM^AxB0ktcnXk_6MW*_4%Xyz-c1kE@2|5CeEK
z8rx4B4_SKISUZak#}zyof1@QUM2IDBnNs_0ZDnkLg}C{lZM-l<)7D!+;D^yLf7YrQ
zMXdXC4bzCEDLk9DNqBsg&a=JzMcbf;Y?{YOVJpZrlBFyvF~3q*k&Nu2<1)+R+0*ge
z>3|_pTaKaDO{$vx$%LaqP{ZITUBjktk?a~}OWP=vKSEu{adW34@hSRFH;tv;)l!LU
z!J_ss+ay7!ACCx>_YF%#Y3_c1ne7CPQ;`K@Ihaz3Y%r#5%`0;Qe0k^{?B#s3G`iYD
zxq(l(e$AZ9D`t;v@=cz1apIZ7!b08X!I%)25}IldtPCn(_#S5*mE5)=gUKJ*#}(HU
zGt3fy+t-ru{X=8#bs^S}w%V$F+<u|B&K-e&FEm7i6HO2&gjq|M754>%VPHdof~BgH
z`^hH3ZAIZbix>)Zp#*MAM<;(jk942|*@6<0m~>H!5$8y;r<>M9v3s0H@zK61N+<Ks
z56PrTuMA)i!pqxAU0^I=nv4G<u>>Y>D4z83gDpX7#`n+|o3YLKvhU?tt%{{`nS*_X
z1vyz_oIow_=_0#Cvlc_Y#{?alV5?B+R;rhhw0m#Q5Y7ahRq1ly6HSmt7C-biNBX_B
zuq5F0)T0;bo8FjqU}Hp8oPN1^uu;W0Jtp;r9MzpAiqlf4us?Ja9EZFrr>U)6hju*S
z@MejTDj8pt^4j4zCaqd6t|mtuj2NJ_H{=j;JyaP1pk<s@chH(IXAcfudlAYU8-iuA
zJe&>LXp1Q~vGyKY(-^2}+JAB%qh)RCm5PIp>+*{!)w;bnG?_bR#2gzzANgix#SjtX
zamp!6kr~#7PXG&lxaB$WW09f+9%)j^CfW2&l0@cs!(0+$ty2#}Vz1&r9}@or<h-$d
z6{=-*6*6P7iW=h2=%@@zbcFOknIqI;ws#ijECx9Q__!RYy}6k~dD$x^E%F6JJb@?d
z4zo)aZ>fK>qBo$oFs{T@gDkA7I6utPV3y?!4qcH<VyQIefMCv%z%8E@Z$Q?*Tra6n
zYi{i~VR2_ANmrZO(i@wa{-5DdOkWBu8A%6Iznkn;IwRTVSo1P(dU1$qQ)ZRi8-Qym
z!M#$Ky)VE62bbNnML=u-;hz}ce|<*eJ1ixtgFnwGZg*M|(!os%yqMoBryJ{yrbV0o
z9B{wdJ4J1!hND}eqp58+k}QvP9GK07>5luf>RL~`^=N%*Icg16H$cl@VR5`??QK7{
zdxr#UJtS@Acut3MS2`P}XnjQ~t9C1<d$M{yEwJ8y?dow*5$Mk!Va+&$%D+44?#BzF
zY_)muQ&OkztH<|zta;DP=J+o+Qn|FJMx;J6!+7lvE~Kl(RX?K}Qa{&?L_6_z)!yX(
z9I;&VBZxdVo+dFFc!~R%V7XxrZC>FWrk^DPhaHyBbo*Ibu0^oEEA}U8jni4EO5lLX
zhuD%*Ab^MYw{J>NJC2(oYPMp5=dbJwOp4a3-G3K5-LiHeL{^*a6*8ZeCAhOeT5BT0
z=YH~i^*%O0uKyFL5l-anTBquuTRnevRDg@+97yT?xlf3+*s*rzs@eNyu^!55L|NNw
zh}P(ZcHxXRa|M-UYp&CAxjnAG<k9T0<+Ac0pgVS4O|4yW-e;?g<_h(Ps+}bij;MN^
z_J@`^f+*iD2%U6V+mu|s%z?&RcnMu!oZAev3C3Hgk$ESuLPmhQO%o*{-u35Yi=o-w
zd{w`88ivMMpR`B$ou_xqdibio?QB6yATewT1Z~Io1lseDgHSTzJ}5!y#>|uDp$U&u
zvUPKX;C;-phWo_Ocj56ds{(7=X05#{t#h@r^{<B=eJYPvd5;7KI?fF|hDC<Z_uO5=
zloBdbc04$1wJL_mdeAzff!uc>QWF`j7B+Mp?0lgfZWXVydAO<o938%wJySqai+Nj1
z<AlzKt3xXbL`rD0(Yy{+Bj?FfP-sIM8ufO2a4ciFVW=2%YppgQq%q`78g-@9Ms23P
zJ~I{rCH1n})D9i+H(^~ay0C&tFx#1*-=AJsByG(c;cF#XH?Kvw&ML<q+O5wYcBi9k
zZy-TcO?0G<4}j!nBA&R=i#N~lj#k!R)UTS*d(`_``8Rl>eLS00A-8--k@9Re)+KTr
zL$;m-&k^z&x7O{m?WAd^9yie%Eu50LfTGG~O@2^4h0XkD`dU5aYlzTvwDk;U80+#E
zd+;{%o%}`kOxkpu3B)t_Vc@ED!icrreSI{hwPt}A3Dn9>G9UBfb;$BJyrgd>th2B0
zVUerno+3~+Sw}0vZkG)!N!uxlaf(xualDGA<x^zXmkX<M(F;ikdCLggUI4F;#KvU)
zheNx`o6Tw!Od|X$YxdQIR??b%Nm=DwNI3dEf!pT!jEPnvR_g8@n-jG2L!-iqwQBA>
zmZVa;Gul%u{q=vMopL^YK+~nzq-7q!E`^0)oa8SVEjfD&?c%h5tT*1*pQDCigM>Sj
zrsvjBTf>{QOxY|z;BTz&EGP%(=tR&WIAt5jM*I!PBCg`DP(aO})tMi8ilF5=CoQZq
z&eh0Pi^EraZA#bUEft&QV<y3%u`(De-qp8!%JU8;8z3ssg4xKRWP}jfkG(5lOg6BX
zc45!)yTpka*mr!Xl!3;@1u?Wj!@zU7R&M&`_Dwph9M9T+7*|4-N@J1Q9Z|U$Mv!3w
z?OS`ph)uG2;$$_&)~;$TJmzc7Wc?9ssj6}-0<e#S=AW1`o`bxD%~}gUBD>Y3O+tRE
z)}C{pa26}oDW`%%(OR88Yu|vD@sX)TXwfd(56;wM-tU!P$P`Aw|7BVY{fUn^;nb0s
zU3=)8HA_G2QT=X~9Uj@p0;V}qXx@fNPc7qY>_FJ&pH%O^<R{S%<G&;MQ)7HE47qEJ
za!css_9&SzW7q(e8pWusu$aI{RL|jAmqXZx4<`w_n$(HtvCdW@ECv%_f`4kUU{<!X
zuzG1C6?aX(eolb>{RFQr95cAC?jSaUDe`a<vps|G`~RWrEQ8|gwr!o@?%uc)91<KF
z3Bdz|1PSgCoZ#;6F2Nx{aCdhP?li8!-S5ly?S1w>=hVHmZxuhNN|0{WJJ(usj`0jW
zwzcS`;%!`k;XS{@x^)`^nbZ%BUx)B?2Xfqiiz;Pg`QeuzjElFGJ+juSp+S(H=4d?l
zs)P#L<Hc^5piiZJz)h6YnVQu5D-)XDYsTaRa|oehB~MbTyAAWqR{2zbVg2jcEpwV8
z3(cK<A`;hqOv4Pq>sy&|h=c}Nnp5xiIP29i&B9au;(IYJ!asFLT85XnkK9j!+uoJO
zJEn%Ls>E)N3qyu&iFmWcc<^#k<Fy7_Wk0Qc`cnb*=T(DH(?3}N&)?3whc+@ekip;I
z=WDYy|8gqkG#N1|w-6&~#RhLjm#sfdyRIOF`l?#~JX}LR@G@@_cQwFHjz)FrD=N(X
z&Qz$Uo5akoey1-(r<8J{e%vG2F#(e>7{zI=YF~+n3z?~yCb|?D#UZC$=uRN!IF6Pe
zouAPEvRu}a{fgA6;#(B_&@aj_Jz{F>-q+Kr$vUZG2CKQ8uoMvG%11pJXQWU*^Vksa
z#AcE`-N6iI7a9JKixGy&-!xj3?aP;zT3oNO80l5LbcQ%<KgoS}-eP3Z)Jd+6F*7}=
zkp|S-Wm+=s;r%M+awgKGu=AsM1D`?RTy;E7(_d6>`KEDT9v$*kU*#S3_{;<P1d?Ri
z)W}&t&y-(jgM=4A5KF%^s21D70p9Axdlm0mW1V)m=!9ox=G+8GG}`Y4Z8!THo6!zc
zXg=CR!lM#x)=`FQ%cKKCblKN^eT6YSGn?G!^kiyk0c^lA#|L~#dJbHmq|iG*u5?n;
z;Xzdp2gVbt3oDkp4OkgeS3AUA%MG-vr)>xv&2_o}I)F&+GrUg>`lfjB^O{sg<+X!V
z=46D9fY?ayS5HUrX^^My`QOj?^UvjW=}Lyn+htzKd^xW9ZY#!*e3LV94cvsaIyU8>
z(m*SSl^+Dsfa6Xl_j(_2+PP`^5qP3EUZ<sEMo?L;n>AW3j)LnubhR3nXjIFSBtbaR
zS*f=98Vf7bmwBL(qN6Z0#-8E&tdeK$6VKql(#Qm!$|OD$9Ti-!bN@fiIkF8?b?Va>
z*cI<(+@on@Ywf{I$5%O8^Yojj5jBI*;ZgHd-BduEZ71{lKaV(o{o5R-`b&j9(Js%4
zF&;ef-dHWCx$fOd)7Cm%1P=#cel$d+`-D`YN=UIfO+00tcyT=byBaS1NQknc69V>P
zdpF7U+=edN4hl6)AE&NNxpf8=67Lq=TxZj%{$IBv4A11@h`_+_w7fBv8SYmM&MdqP
z$#r^494MDhM@%8UfHYNs$w<A(ptWLpO<ZaUn7yhEsWx2CM_}X)zwN`5M)6B*%DDPi
zKcj^MZ!uj^78tunf@3J^1}IDa>o8JM@x?~Drz3Of5>m+eeyrv4c=t4sER59U?%3mC
zJ?EnUbxNCdhRH^W2Z0S(00iDHWSUhD@l7><H-$r^A|0+RP|BRPGeGH9^tGqMQTDan
zk>z&@iKnBeU+Wcr&SWIOD~5|x#{8A*I<_DRytjZPmeXTMXOpE$Bkrr9dF$=AN0Zu0
z*%fQ6DwO=Koa@by_S1xwTBI!nlD;`+=qJiE9O$Bc7kab0&2J>{OE<NF%g9Zvp)DnP
zLj3xtp%$joRsLyNym%#I>aua=Y<d-obPhADn)hYDj7KVcRAz)r3pbO<shg)|5!U2e
zVmG?Si(@5D{R!fb<D?Owk&`Dhv~GWu&=rd=?SCO2=GX2>E#WPs8g{8J>r%L53ss;f
zz>U1@&B@OHvht#gZSG`_op4;T2_yrv4efZ`*Ih$AA+!2M481!C(nvOHIIwTzB4~5G
zD{%~qF++T-zu&k6at&Ir`n1&``5}(9(lpPYSu;w0OKc5lkuer$<XLpQp-yzoBjVGP
z>u}Ri8J!>k{~#RGM;a@3p^n;RqsxgzB|ImEr{R5#Nk%k4WWqsGm4Lor*Q?v9Eo0$!
zcJKViWj2I)$+9OopY_pzIOs5az;{W$F_XtcxiF?q+9Ne@E(kW*W}ckYIdfCrA?vSA
zv87URDdVqA5&3c%@UY;aiw-zP1r{k1qd0RA%LthL_7@m5!AlsSMEAl6Od3g>c*jx~
z^yK4LN^NwS>WpXs&1z*D*z2)1%xKPF-d>D{9aSkCHdeKFW&-PLs5>=N?~MGHSs#8G
zvG+lquDDraIpAXKESAxSnjQ;OxI(@cejzpSdy&BGN5iiZ+p+xH-WM(mywHhc-E!VT
zR*M<~EoohpvvavP#8_WVgf2U*iV3XFo(78&rGBs}5=}8V653kgYZy3Y>n&i7r>>kx
z_vBbCe@2i&!1h(MF;*{(6X>c!q3lVs*hsk}CSh27VtFb68^?G%nwyn$wtQXB_{nn<
zs{N3?iceSUS7k9A(rLK(23N2v@+}S9qQ}qOhUPQL=zQJ>6N{b>V3$-J_SH0tXI`}J
zP*+9Tq9Crg!X!3y0DNWQw;?-3rZ!Pl#?1Pr$~OvHwI%eTQ%DkUTTY}hpX1@;kyxoe
zegj$$o6$}f6YGP*F&2yjo=lY`BQg`w-e|NY8!TgumB0;wG+>6u%AuzNbkr))cpeS<
z6yNuA*n`o<exhJE2tx7)ccQdqV`6IO>~q4<FCM4sevndE5Y~#7JReg^)xnR#YjnPE
zZ6+(}%Q^TEcgjl-g>|#NB)ZU(l5e>1Qf2YVnnGWVe}5gnVbv-ki0tcqpsHj}&f$Er
z`?UYvnGp@Eb_e3Y<Z=)@UI;<9ks}4^+OU;8iMMvYL7FjF)$kFY&lPJJszog|J^e6&
z?xD|X?2oG)WG*9UrA8Fm+@IO6!xUF>80Aj#b%a{7#JAsc4RQ8cGQnyZQ!KV~1&Z#N
z?JyR4tNMt6B5#k#BFo4QgTbKz8J*0JZ|>J&oTDe$g0=Au^C|<0^yd_*@4nh#JXgT<
zU7O4BkkQg6mCvX#gaHG=+VpayV1=?y&GcKO`OAWwh_etZfRLJBmtwq#C!so!3-nG0
z#jJ-&55gtX(9P<-1tRi@?yj@SGThR71U3%<)ey(tM@R#Wp>-$booKiM@dMEEYxY;a
zgTLh2r=K8}@v(tTd)ub^r#ne@>6iM#cc25{Ld*WwjtGaRwwSRh-_rNjz2Em#z@RMS
z@sBYst`Q-(iBYJpb2n9TgxV0ia&;Zh?CR8hXKVOUus>8hj7yNoxu-RD*XCS>WYbgU
z`7_d^l>#2S6-vWj5)PPz-aSF8oMBBsZ?HaNtr>VwoADzK<!qvw<+QrL8Z`8jRhPx-
zE=(lg!0+i>0Afmiw_qP<+pI4-@jL8g94Wx48`$^ivw8aK4Qb{BD??Bu8?laa#cZOU
z%y=alD?=wt^V5`%Fs+=cwL*P?eK4?<ywZr*@oa$370>kgJmw3CLTTy~Buhh9t8hw%
z>S>el4lb0rR`r*_{LEu{^J1inL5_jY-;b`Al|bbnv=rXMPZHfMiqp7%aXGvrWHa}0
z;eBH;NxU~92W4$qF<dW_z=7|@Fmgx^rxuYuZGaBau%qonQ<)-P?jxIZxPRFR3*PDa
z5kS&lT0$Nxf5e6UY!0rBvrw-d;abF@1!jQlC+0krDJ4WNE~9uiba4vhPneEz1qwbW
zPrZ?<nZf*g9#WwpfGL3}y+;&EIu&80c+{wfM)^nx@C3qIeG3L-zdk*&{+Q2vq)PGr
zt~;Yk@1t9Jpwv&mH{G^NNI*xoem+Qbda`A+0_>v{k$uSl=hc<s#OFHrt&6*rm;F?c
zI9vW0b?;_WyBL3CO3+SD)Ygf2_*fXP`^Eu+`_0&5!?5Xdz?)?t^#C|j$=C_S8gl>W
zv&TzW!)XPek$kP(Qbdr?)0Zr3XZ4UBqI0&T)BqH2vg<i+gtYpKl(Iz<3h5tb9-U|6
zfmu#G0r$FA6(gEu!{g8KZSVq*Xa!?oHLi!Y+aa$|q_*7%Z{b+;#go$0lNx?TiVOHg
zX{166(C4htV|zX>-WKZC-+CA2;3}(?7O$@^Q(19r^*3Z^9CI?ITwHo8J(52he{`kc
zJIGebY4K=r-RE?!L}>o?g?RHec8rf@U&!iu7kwuqiAX5DgJVsZ-yH99f0@(8_K#TM
z*j6DA&rwz41Kyl(07@9wimjyG9_pSc&N>KOo3Ku9NCxY$Z&8Z|s1VO(enHnJYnWvD
zzjPsS<Cg$l!l0$hfFA{Ro@{xRs`on*V&&>vu0v9L##kBVhOh9^jJGO7-LIa?r#THH
ztY`Pn^wbU<C)3BU&>xUv#Z8e4*Ia=zZ;Ef)rZc8&9Zx(~DQ=}f`s>1t`MTei-Y4rf
z*|8W5I7r`3L?H(k{+1;)lODaF;DWt;IQY|-N><Fye)clA=^1QgijS&R`H`t{8J&-|
zea&aMcw%V$f7l>(2GJxOoM_4Dz%&Uh#>%q!wAWVTbQUNh;_^k!15mkIEiSm5Q=4WN
zIf}NKEJbgf)5R6N9zxI1HpowV+y}~kIIcMR_e22|g<4x@IaCPAurWrz*~bDgR=r>M
z5gFD6XJv>tj*}pEce%;&V!x;?!F5=BY$0HpGdNk|w%?cL72t@$T~oH<JZ@}DzYz9D
z>G`@@W*hM^==7D75p5eY(%|7`{~uw~dexLLsDE~^=E1oV?EiI8nirr?%-)`80ykKi
zNSGdxO|J|<mhT}zbNd@GlsC(w(JkcG56O@Q-4l+Q#V2;TuFb6g7OHEq=NpN(b{JT#
zTIvY?O^)KiaS6({np_?pz|5NMzgs;Se<t;yt~Pr9CWl0X)Qc*_cdBJLu>9nDh_#KW
zQ%2nbI2PU6<&CA(>ge>5IEJweMkRO*gEH;<xry?4=9nyO<_LfO5E%9WEwIJWPikmY
zexou7`MS?<pt2nR_@b9XHVB4$)&LjvQEC8HEQjf@z;{~1z&SyJOB$}>-{uZ6Oi}0x
zUqAHfYP}sPkUk@M!9e`8;dmvTX^IQ42&JEm^nF=(mBIPF2yCdEE~V*Gy$xoOSnL@4
ziLaj4t}0~|9qzDG;HX*o3BF2#Zu~u9(bg`fmz=X~xl{%Rb-3Qh=W`zczUuWD>ytl&
zjAeiL8|WF?oxJtWUcn|B)ZdQmKS;-ZSxeP@#dHLpfaY=HOXlP+f&)S{e(>PYc3Sg)
z99-1Vg#MVQVJoGJMIq1^Z9PdL793ffwGr9o$7%Vxc;SXq#A`zL0rZ0v@H5i*hMAvM
z4RZ|Pq;ECl%2=ZWunS{6P%di$-^zNC?0lF?oOI_Trv?Z9fI<B|E9;9NO`$Y&Coqb&
zpL<h1E|8E%(lr2a66pB9Z3KW0?t|O7*HLnn3G6F<DIb>hO@qj=zjsGg%DYeINT6$5
z(t~U2UCyLdejr$-tdrZ&91_S10B+kya&#ni@7EIRr2ExQ&F968MI}4y5UY}Nz;{Mp
z@Yh^+>r@{9>Fy$JjIpeztmRg553ggxxMdr%ye^UY{>35T2`)V0@|Zbz;#0mtyFGlu
zMAERyV;WV_Hj6Ki7<10i|Hvyqbp8}8e(n4`178+ALsb7^s!AntH&|sNez@GSu3hvo
z+Z$Ajoz>0_%y*HWFh-k5TgGe01bG(h$I_5yH`bgyd*?(SBA*|lg`NfOxBEM5+I>#C
zA)y&<k9!1Ukz|&w-gPnkf^ihu1lOu|`weZO%3AK?ZNEtI*yP_xXJgdwJT-l2xxWax
z5<1x#pDc@@zngKL)DBi)c^DY;z+tYK(i?Z+w^XwO`_V8WPw;hf6HlJOT<d^ZK@r^+
zzTt3c1o!vMp)VuuHNy2_yea;p#<a+tJ$w{ya7glYe3_^d&&t!u<#@AT<Go#8x;8K-
zdh^D(7cv6u!Y(P^e_W4SW&>Y0VCFTyHJhh@pvbQ<WVgk<w6dlj%2f54=q}goqKsQV
z<v$`Q-yz)Oz!9Gye)`0%Cl`Lj%+%aY^bUzdL!PBS=w6Hf5$xG-L4V-~Sxc%tR9`}e
zQ|+o+>(AwYzFBkRp{0xQ0n3uJOG$!8{iJrGtuq`Icyt##HnRzic6zYdDjS(jg;&sS
zk&@r=CQC0=-mGQ%qMhBs{Tw(<jf#%KEf*c=X4(Te`b9PeV4>TBeYc$X&)AWP=RTy0
zlhB7!AZr6Ad`#&`RVdx@F4EdB)Xv=LGZaj@3^u5@0aSQzok;zwZY~qf=saQq%O!n@
zCCPlnwVBmehf>OJXl;b&U;FcP5)T>&(T!QB4S?r9PjaIi&$zsg4ilnQ&WVYoG~aNJ
zSM#QmBXwH!#wx4YW?T>Y#Q12(Gz5@V8h1gPLjdApULBd@SnK=9JDS+hHGig41dl0K
z>alhV*d3=^>gDnPe^I1M!P~fCiwUq=S*&(f1k8nm5U|u^R!>lDIiw+`9_6nvot;ls
zJJLbU?|4=OFQ=5;d1kad-O2FoMh4!Zj~Vl3d-ZHmK}68#NPMYQs&!@GPN03dDt}Al
z*o}H1Dl@N&dg-ds5q%v@ckAyERvUE|V+YKm<MPGavGxgyBmB(PlCBOh5{kE=Ju54_
zBbg4-kB#k}rykWNq2W|XvdtC7{3=EGIY)bO6g9FA^_RTvf-b^mn>ITDzso2Xn+)0)
z)m>q;#wnR*H12w>_6y4)yJwc0Qhv`6OlX)hUIX(AUexv8Y~Q)(jFVsW@K+%pFuha{
zzF~GdcBGh7siMRtQLFs&hZZcyI5|896k?v=LoKEoFd{Zl*Yj7zu&;3@Os)$86?Hb;
zd*w|%V?y5mM>>jTwJpO=kru*znux5GU&>sAL;(T|Y-d|qfcMl{X$9!94(|qlEKd$$
zLzzt^#`q4Hz_c?l>pJB+67ydTqQxI(HoYCGBAw&L1ys301%wInG$<q{jyHpv1N^Nf
zV(bh%p!d>hC6lYOS>6jpG7c-|S30GLcZLPVjGCTtT)r}-`fG5_o)A`ey!)(ly>4Qr
zFdE2!Y*BtB389etFf69LmAZNAa=+c2QBz7WI)%K6cxOHQ*t1B&Q^6Bko6H4nkh1~w
zMD+$X&eddUhE$yyUr77KjK-)))waA1zJ40*T*7o_7Q#<<Q*QgWNp=hNt|&~g()8z1
z+UOss!#p4p$Df-!=1}k6xp4*U$np-)lULlYp?mI{XZEQD&LyvEpEcX3{IDlZv;Vkk
z<#7%&CJ~II4Nb2?i{~u*G77hvrsib>emaeObjL<}bglh5H2*56s)ThJWYxS@+_g6R
zErowCH};3U6UiFlDBQ0f470rIgaj<O5wr$;pbKj*#eH*(_GGBsszc#4`zxd|KkNf9
z-m303SxI0><l$7WVcSjCsk3;Nuv9gg%?V3o%sBW1_2+%Pe_&8*$*|*0u&q(=ibd_b
z(Nlt2jZH1((7Zu*q_LBb^QNZLf>#75vWfrVbT?TqSM-srNHsOHd1cRh=8vil=nPsX
za}ow)^*_9+el|uCD{cDFE~CniJO@m7_3Cu<n2;CW?`Rx*8mcs4VfRHn9bk-{+ibq(
zE_;3f^km@sc-HbQ&PxxTa3<;t8Xo)>L$~K5hyw8s{*sHiy?I+{9*u6k3a~9qPs~@B
zll8neB%NjX2Ki0(pa~m#(i&+tQ-Q-YhrG)|du>^4%o=)|#YES6tpyv2y~exJbhj8s
z5@H?_VCFYkEP?}@>M|Y|nkHAkqr`V=ic@xNA>2QYWT*XP!BMmO;$Z!@&%5uu51DZQ
za7GP;M;lXPLUzaCW$n?ni2$XxrpJqooF!3Y*3|^HBUk^>pMSQWT)J*ckD=U=XIUz)
zNvp55U2jpj0gqqZRr_np4d#G)U(Y7bwPZ7v$JD0tg!?6IstgAicPpE^iG9Wf<TAb?
zfF7W-mVC&y&?D#MnSC&0VOh7@iQH(|+7{<->~Xz?uic0)l(gI1a$9<TlM}g2A#r%@
zQSn?lN_&KK+$+z*`~1eS_vHK1U{UEVo~9dSU3bjma(;%}(NEdrecH$jbLS20x?3(#
z6DO+x%~J}P%wzL-ISIe_?mT#duCZ>#sSovO&7~DJ7%X7Y<$Y8~M?%#0nMWdmXDf+w
z85dvk01jOf0OE9VSWet9SPpKYFW+3#)>EXSIB1XFO)R@~X&%9YN{#~_ALw4#9!$++
z0|3QmhlTH+!x55(-gwO1RIP44AifNEU>q_$E(%lMnuetEZBcO>P6S#ml3DF}v>vmJ
z#$kr8po8}fPxGdaT~YbxMt%r!0v1+NXCIJJ$JnET6GP-U;C;c*aLmiQSF*f~Zv`+&
z*=+}tb&;UwCz$J}rj2D7nOVVW$jG7V38`i4Eif!o*1COjw70vXWt)n&)EYI0zt<K8
zI)74vO(uUbKWkP=^w;2?rE+fg3HeF9exL;^zT!w%V?MXdW)SC5H(!tXzK%vXL(BgI
z)z&3>I4&Ql&NR^6dSX5sjgfQt2XnS{t$0c#@+HPuB?GmSGr@cxlLhVyDjnCW19QsW
z@v9_J1(6kU^u-W!<HZIZrDkic6H+Y(;=VA#4MPP3X5vbM(@fF`{(be%q~=l1e$3ju
z)T-dL8^#sZR4@CV0?Kx2Ts^dNp{4$0TsOz&6aCk(gJKzCZ=5J3QYmw-u+7)o=Y3*M
zq}P(*hkT3)<@bXF87!TX7jdxMS}EIpqfvq;`#O?8WmO?k8A~&333=toW#0<mpjR^N
z@Tm73^K@nih~Iyc#!$YNwdFcFzNh_ZvVV`jB?9skO#FUEW7{A%$gkeK6@+i-dood$
ziTZ8}$r!tJ6o*fG-}sEtkY**H3jMeI3rR0@=+%$dYX)%q54GKsN+MZ=3^lXwo8!_f
zGOd(k3QxP|DS8#h&%XoT!uog8A=WV@@+gJb=R!Yz-N(Gk0uBoVkM!&B)3eK9`3Lbm
zFMRi0V;yD3i#E?Mw;9h;Ehx6n-(Au|Xi$UrAFY0<^<UVdJR0p|bxNpCZr&81(g&-e
z;`jU3*7sCVhIRXTu>PUf4kP>x&koxuI$u@N<*uDRc7npRDtwzMFY286LkY50Z>-LI
z*K8T`5wK`!ETN}Y1S=Wy(>d{`oZ=|e+5sG->ap32WWph~q0Ld8$wwZ<vnqy|`%|c!
zoh^hw2C)Xf`rDWMj)DwrAGE-kM4p<;5sD5nOlK$P1iB1jO510K?KtLw==(w|bS%~W
z@Wj?$$IpFz1}9XJTz`Ppz{Mll1F}7eJ>%m@ea?!mdBKmOIyac-Go4`V%W5zMiZgK)
zaI?>aw?}sW-m6}q-yb^#>i--kqQqnq$K2a(Ug1+_Rp>bIJuFhec=GuzjiV!g(&HIK
zk2sTll)5($9d=933#Tk2C$k}$@4>sgJqvg1C!ii%Cm<8k>I4I_#zk-`6*hYS2SArG
z4&`nh?!R=|z+8wF7=&MIXAjVsX@B6g$2;V<|7gZL7bP^sEK>V+%M_eGY`;$kykZq~
z0>z28TxKqR8toYX3#AF!hDi>pki5H?3J+taFy@Gevjcid4})S_qi;ZlsA5_c1OmGd
zwg{OO(GR&t<Z@Eli@kGoHPxYEQc7q;yy}_ji%>^O5~<m`QjeNP%f+)*apTP|$ANMH
z+T_|0*oZO@!Kt%YG73pFEuc!));WT{4QPBjk}R5)0i7u@u{~i$p0SGV4yzutd5=CN
zBIV4>n%qRVptpd#z;@R(<z8*rPj|6qGPK*Km_R5d8H-sp`0KQ~P~~U;*I7Td-*jw-
zmaJ_tth)!}SHH`$S&S|HAKD*cnDrQl&<7di{mKn^jV!>GgCCp+pw$c<ZVbcWJrrmw
z`=M_z9sB*lfSzL#zOCPne^IpU-b2Y1@|ZCCHCM{JVHIGx#YHN($tX>Pw)K&}#n!W1
zV&sL}C(wX=9}|3h&)_>zOXGhHQ6{>V!iKv~g3)P5cpZZnS5Ngvws--L;HU>*`@%O8
zWt82yy}R6EegmR=vf+pQ@g&6R^m=2bJM#4hxGUfrvRX{1=~t#iP+~XRCEd_sE6|UR
z>l&1YdSml_M5$+672gRNfVqL`%S6HEM9E}--3OCfqHJILYb1I46}ZAbJ4=m=(IxV_
zC>Z>d-L5vk?(DU0V0|+r+|h*&{sVtO@SxA;{~8KMMuD0p!VW9Q*o_Usl<(x<wW_jE
zks#2?eP&mZ#iJ@eUwMf9=mkBP60cG7hhR>B9bSCyLjrT>x0#?HYGtQ}J_8@mSfH=s
z5aEcDPu?e^(QwjkCpqbPdB3Kmy-?6H^r0)OSE}nimrLOI@IXMGm986;YAeEPNQqfG
zr5Gm<7e+5O{RVa2O7NEYn2*O=T8TLKsOHOn2Ud&-;y&#@I~@LP74sm8Z_MeLe6!C`
zQEBndbeURI3Y2gm<xj-#6%TuhtgbJO@~np^-ANsPN8dMEp4%}BT$KfqIYy?6>MWrW
zH`gwjr-maX_6j`5H0@l$Be%I;9&MrV{7JMtvV<@^3)(@bJlc=ci<%^sz!PyW;EDmS
zG0jBY#yJBUc_ki=PDbj|ikV+D(nhe6hkn;jx9yVGCy$Yh1K9aan*FzT!#j7$onN16
z@a&5g?-@%Po7KN1ALA@`-8(M1MS}RZFHfrej5ConmefLu|2!g+$=W)uxRtpM<7^wr
zrmneM?5`y}rK5f>YpA^UX26c%h8>U~eW>&FJhupFw1+QKTE`Eu9}{>gehjz8rjxC;
zeP~c|N^R37s~h~r*ibV|wyONq&{^rtAMG>CdIi?ddC0KCtfogSy75WTgiV?V^b+2p
z<<~QWE!`zg$m#8dxDq8b>7*QCh4Gj(fVN~k;+p?fpl-JO?uS}1lQVaj8}7B!b{OlD
zbH4T_splqxgcet5pyhQ-U=JCCYkJ^1xikusk8U>m(eR)@XXsa&OKj*xdjHHh3I2iX
znED(`ZZid~CtXU=Jp66c0O!8z!n>-*!>!ERPVtjKZ<`f;3Mqx|wRDS`?WpJi$1bRt
z6Eq1N7Xo6#nA{ls^?-qtYL`eyAjZt{U4)dK!JoH%xn*kn=h(r1sD0ccSlJmj5=I|L
znexdK;D@TMr(H_YoiH3GFllmES4aI!6`*v^{(Mui6y`!hn}<v0JKCiqm*A|3q!i*1
zdL*Vj+$D=b-(&p^Cijen5UmyP`v}NC<sZ^&pG(vyZ|a_^0BsR>@SV`jc#Mqgn`?ne
z+>k_Bqwz!?UFCPBD{kT$aQ^M}V!AMDv^*2=u;r_gRD(o=n9lsFl?4D@;OOj8*x-u+
zFaznwmzAZY%lYHesWvg4TDESla}xW+;0(GY3Cso|VTuGNn@zdOve9YII8i?qU2>b^
z;k}E=*AZ_)2IHNRbEm=<h_2AI_^x9PD^^mW!%2Yg@p1(t_`F<_pX97<z@6_NG7syY
zpYGkYy@Ty1VmRc8Cgd|UGl|!BS)=M=1|2)vTq3)JJqYc&A8}iQCvc?;r`V?!>2*5j
zn6;zR>G746qaKq_0vMg44Zrp(t1{t|4I$``0Ok>!M8xR?Vt0wS)1d}{46xKz%ye3m
zenm$bATTO5>2mNSFyU*%Qzg7J#J~Ys%UY%YvwvtOz_&ngM#JCNJR>>#s=6e**w3S>
zP%2TFauvrh<RlR_pzt7uw+@PH%g=Z!1^YK+_1=ks<E^sGC0BaKm9e#m2c6>xHCj#)
z4kbvpy5PH?^s&+^@QE$1F(rt{+2%VusZYs}jB?@6pzKL+1^RJwp?|2^p)^(Vu>c+`
z15p<auI~FAv8T+|ERs#qLnU67G6647=l&rz!<=H>{wh>oL;=EM3T6c%x<jAg$UmWB
zqFRjkJd%79wZFI7BD$TNo5{MliS{aiORE}}uK)%Pyc&X!56Cl@(NuO(UQp&X`$zou
z>YNKrLiAj1DQDvj%u1N=uic=1Ute^HwUF#L2^bz3nS`1)nV=6$#)T`4u|LVQkqU(!
zu1^?rGgW>QaNmz-sGNCs%Jw^uMyNaS&PV*!1p^^SqSZ(Ug=K(!UEut^fWT$CvFX1y
zrTUjm3D8@gP#qO~-*sdk6$K|<1;EnT{u?d;6Y*`{I=K%=+QC<1=<lae)zyxBY-~Iu
z!uB+M!Z5x_C}@)vnXv=p{Am=l?{>W=DD-2nMA$8?LC=^8$xC@%o$ZuQY~>wZUv}#!
z_DYPfa2fqgq1OhSexLq4@AsmG<mU`?cm49<gdD?vB`7yo4K1hR|7Zg~^&N%Y6KTZo
zq+LNkPjlwmefapDxrdD9Mr8o=Bb-l6#7Bxf$)uYNb(9r$uWoXMB%P49K7JJG&$cz2
zm~-2EU&U<7(<@=IZ?=2TjAXIiHkzN3bjsak%PPPP6y<^p6PsmPy!CM@v8{AFf4IHp
zHC$~NzYSoUQSU|I_hnvlA7rq{H}vKZZlGhSX?$sC+FR3W1H%-Q2ae3KTv^pD*@YOD
zTd|4;8rZ<*ed&p^JeOAGCuad|z<SE2-xHxv6WG(?!EWX&aNjemX1TC48KQMy8sbq)
zW4=b@TSw-8lU5@}n^i$e;HbKm3Y-m@(ze-T>|&Sfl<EPSK8ji7w%zhb+3j7@`d~J4
zO;W^7ur^sUf5%9<L*!LMQ;nIR%UlM-$q2;C<2mi7uIu(<k5jt%kH70_jzyyRDdU%j
zAn^G-UB<1`@LR^P<(<dp6KvyAMV(Zk!{?G>%(8k)qb2lpcEgF!&wo&mL1)sD>4F;C
zN~C|9bRZ!X_9yw4cp38C$Y|OyR&PjwoosL8%TX5LPo|skK(dB0aH^3h=TN@H@vVz%
zXjHCq$`GSA&13JUjrUb{Jo#H9aT18G?^po%VyoCAj*9gf(V6yw0nN3lVFZ&5N%~JW
ziR8SJ?1-NhofMI9YOHcvXXuZ|ZL8oj1C6U3_ip(IsaVrsXbu$_dpHl}U!I7Zu}stT
zwz`PN9Jk_&;mtA)bW7c2q}8r*8PY6_L((3zwq-L;7?u;1>1C^=$5baM&aCNLCy0mN
zINbWzZwM#g2{Vj!rBaE;7zZ5BpGZ21t-36R46Vd{PI}7DEqE|yziGrM<vFg3sN<b|
z&NJS)M(`{#jN`x@{C<Lbb$TYNilAg&U33vmv=_yEOS`t=+8^&1e8*Ul5sxAxcpK6R
zXKQXf@K&ZM^s97E@hRYa-jQ6P2Dd};CJXUC(5vtmjIS1)Z-mfOpa<M<K5o9}yL3>{
zT5rfDThom2^X2kt{y><YPK`cJ+aYjqeuH)?RaxBgIB1K84H)ye-3jVp<^|JthTMJm
z$hf-G?qVBJOX+_n7F|!C$)!AR*4bJRK(@+g5}BH|(#!JQs2>LhD`&3gZXX3aXTQE+
zR{AyivEo8A4MaUC$)dP7Ai=XuVfxXYzBx=bWcUUp2FZCSSn>}vklsFihFCXziAFzH
z8#VeBWd)-Xx_Xz5j(g`g^oOi8(fM>Kdz?naUdb6Xfmum-{{7c;oKIqc(A<Mum&o?i
zcvVBwRyI8%y&xZ%p^P=H27SRZ;giq@$@(IjQvs~>F~AH?6wtg_?zAL1wo0!O_MGl<
zw+DOhV@-=kZ$5&=)>`$dThG|63GfbDy?6(^ap5;ih10|H%^tKCcrQozpOyY>X%2Fo
z;o7rr@lVyAx_~^_FHKIk=PhU7e6O=t50e#l57ZD@ghJ-A9(}hjqQOhZ9$HL9U|amx
zslB^PxFytF4iwT>4S%6la4iM`d8VI#*bLs}{`RP8b&r{-_Qstvd8exiIJ0+vJC%xt
zN1N0(yvfwIUOxo!Kje&NJ&W8*FX~b~HIM#VTTk#W?DrMQXI*@67sSkhLFI{#3Rq?m
z4PMOMVHkiSs3sRSP1tY-^9JBGd&)Qhz(n$>#Pv)Q0P1N#)!t2pmq?IlH6ZO=%erUr
z*TBi!h$6GGP3=?mowbf4j*JKWvWA6n<UC(XOM`8Q?;$g38bh<{j-k3t@qFdb#}p8k
zc@m@$s;F8(Wvp}It*3p70~9dDG`p6v`rqLnZ+(D76OhOkj}T!JnDVRkTA2=A=F%VU
z`cA?kD1VP5<}#pFj_}P;Yz|@r0BL8gMk*-gmjo6WKRaz0BAHi=-*O}(9bWj~O)r4k
znpv#_2LO@K>+!i2K#v1IubScY^rJ-xc(Tt5z#u_BqSqX<1NMG2tK~Pbffgr~IPi<F
zTu3syiSTFqqS5DL5=tx>w;En3+${irKH|`3ep|it@Fi^LaPDB}Ut7`N5iSa-0RCE8
zOls`!S5}Sqx|Y6qH#;<>bwpXrys+a6WV;7rB%aYkc+_<0Gnp{~1m<|Zg`!7W_+V>V
zXD0{qKJzsUK!Rtw?#7nKaWj(PuI0)u0l!dE0Sfm{-A5JyAaUkp>gVksFKqAGB;P3t
z(;E)jVMevA=)toDy7~SGhifp5cAkI*{*iJ8Aj&uVB`dTVNXYLeQ@5!3;u`cJ5S40C
zjMSzZb0vi)`n^rmp~^fSD6bt`USJ%Rn5W$Qp~7omChz`FRiylECHnpcF$5idJK$b?
zxmX`b%sY|a!FN!+(Kg-Y>FEd{Wz96oC^okW|8-gX`^8Uv1VVvoI4A!B*Dn!XU9@Vm
z7YjgP_(Z!}g|hxKL$L$^Z|43A)noF*4v_l!qk}6#C;zW73rWH%N}wNIBlEpkzjwql
z&oSCRp(N=0bgf?j(1=4g6oOS;3?K^uWV!dZZ1}ebLc!57Iil%!CM2R0yx8t=xKN`h
z{s&+x+n77w1DxX<Yt9+3jiAbTIFR|O=RsR7h1Ue+WUZ^^OdH|#t$D*%!gr8meU>`#
zOD(^!>@DMo6*=BJm&j)b(OlbQox&(wbaQOZRp_@&&}pL>qV#kq$qUvjPZNgc|7882
z>sWXZ16Y%b*0U;qKZtKIoB)-{<BT8olC^KDkh{x3EDFx^t_1kZ(>$2^zw)+z3n8)<
z@|#ZNimvYb`2G>#t6PBw7FTdEkfMJF|Mm6?i$ndoXIvj2QKmcrMz>9XB6^GwObc!R
z*4Uu5CYChg^M}*0iw+`tYhb{B&7n7zF4#pwlA81iuNw^fJ!3E&4rWG}8#ZHTX6;4(
zWi0*&g+J5cWno*?9~Aw)uyGNVUa+Q%F>muaPKb(nJ?|zn8llm6$TcPu|JN7)zeBGw
zkzO$LP%)(cz-((m#nQ<;95<hT3?<WinHxI<Fbfs@f5$K(0kC>i;jF)(iT|31|Hq%~
z-n_u>PbXE`{~PG_KmK|QZz$k_$5Q~Zkp9ykga7_E@d>?zhCRGhM)`ZC@DX}jEt}&v
zSn;bBFIw5{SbuM1v5dF^z4q7mkAFhu@=HZ*lHq_={m0k5!M@>0Su<)FdjrmGk`vQO
zXK~ly=C+#jlX(62pW`1og8%31{-5vFt^rJIkkEI|rM@(#nM3YIK}|5XeB#O>HxyFC
zf3F3?sn`VXzQu)R_rFw^k<JMeU6ZQ*gRcKsJ^!wuztIqZMMy%M;=}Y0S!Ddns>Dj$
zZki><fv1yC|3m6_A_6S^;|A8&o6gJK@vh7L=^^IklTHskTOcpQQm9-GC%dfeX(7#S
zLh5MI>e&<dv{bX{4gwryjejb<WG5W`sTy;5f&M)e13ER!KoW=kO=Hfs2f&Kl1SmU}
z<b(RUoPme73CLIB_bA*xAMLF)5_S=RN%_$ZL>%n{cil$O$8z8X%3mWll4kPDF_qsv
zh1V`T4nF7_K!_b3H66D?`2e4-Tmk_aCryfU0Q1^GUhp!`6{wlZ0DzF>jFQ8#&Es)N
zFO+$k3*z>DHcuuoB!diKKP7Sl#Phw7v3F-iW&tAho&dk~{LUSK;x(Rb^trwoVreQ~
z1Y#KYSW$YK84xvq43CIfwxa+P!u1zs{Y_ru=wg%G44%G9xNKx2Rz<w_H&%Ni*RjQt
zAH{b-S5Fwp`)caiq)5mDsWw3KoBZB58^W0?j4dq?pZCLikpCaL5)(5hfEV2djA=kN
z&yx7_ZPKOcbq&ibu76$}QY91mhfgb-bxS*3b52nQb;1y?P??nw-<KM9znc^JH4HsL
z5|}4H;$?sYU8?!Flk}Hfq8aEY?|}dVI_7);h6RLgZC%ajqrWim>vrXhJyroZnxndT
zg}6wTVX8%~jf%FreQ0bOGrRukd3hJwc`V#ZlnU_P7pYL2@3pubwE^3_<v(HPJC_<O
z6zHV(Q;J&7xaj$^$?NfI5YK1#S-@)`3o(&PNQU~AD4ywyT~JpF6nd4v<2Rme^#6+6
z03?)sFHsvKCextEW@tBnVOtO2|K^y>+o7CY)8^Yom5;ZXGYmg^Is6f!+In3s0Hiu-
zrQZSYkmY)t4PtC8WR`jzv0(S}5$+d|<GDRP<Z!XB>^YV9`_EK%{Qwn8Vs<@070P0q
zD1Z;bi-U1mR-$1RkTt*mRI68d{1f=Km`R1)>Rq>!jh9}gq%*4esE&ZBDA!Y1rHEFz
z#T~E~)ZGYN!fzl0I&%5YF^6sV7t_x*@OI;2G~cX;9(h5dorGc|;~xl^j-9Eghiks$
z5ZssTRdz3FnMyw7wVI8|eC}IuZkLfOws?rAZFz$*krzasO-bn>9;f$OTm%**o;O#n
zBPK4=;e&el6$80ot|x$8^_AjO=$x<%=APUy=~m}$k1C_8dBLWEKl6gF>1&RXT!tLy
z87awh$HGj0?2;QN9oIl60M!wC;&@xneWi_f2u}298vg<Mmtl5GDe_cAn8OYKa3v9!
z>;7(j3)R-b744`As_FVu7p#7q*GTZ3;^|{Fud&F15AU~_lGMZxE_ddHF6WZ(k}u>p
z_li89#X{ogOS`{+q3j#{LeDrB<_Q3F>G63J%=pbz%3?X?cWJMMJQ9f=Kc$9=kNvd9
zkcri8cRE^Xh`teCtuvjZEd-QoQ^MnhiJG)GGMhRS=stDM=i37SK`GC4^5>^kH)m$A
z#Tv83^!{R%vUFm%laAnKhKn7xD3ntgVp<WQ+f8g<U_pz17je<&0AUZ7zjf**H0}pZ
zP?!taV04{nbUf=N-%yX)ymNwLP0H3JjKEYjU7s*w-cNmPc6$?9q-hyUVz6=CvBDA9
z2w!J2N&R#%umemMU+9Irz$3x=fzd?39F71_B?5>sA_N{FJoJT&#Q<&p!j+n3a8*Ye
z`F-^0)2n*(rFunAHs~T>`XSGU)4<xww2wwF!9Fh>cw3-=Aq%m2`->o${D$-f@+JKV
z0GRaMZ!ZsK@h=`vZNz_Scmz-^?mx-0?|gqmgs}+4XVEACzi+>1U10~Nt?SRPNaa%6
zff$`jcv4Hn^W(PX(vsdLuw{WlmQVB9AfpenrW+K7tgwM4@j~EAi|bGjCe?6O`=ZYR
zcxZF(3*RJ_wdnRnIBec1Q-L%8bgmmv`?gN|$Yy7XY>t8O71|pwMCIb}mwIsl&#3wZ
zZl?HPs*nmG??(}P+^lV50n4T6@8$COKVL3O^}4YScN+qRyhqxhzJh$$@z{$bjEPz{
zy43G`Lh)m}zSczPeM>E}+{-qcsO&)G*uET}C>$~+4E-s$Mo#HnHbQ~(_UNz<k^l$|
zRDpb+u7ZINOlv`H?~j1=ok=~B<|wf!)e5~HdPZjgry;X}>i19h3E#`F449zWeL8`Y
zVIG~YMK4)wHHPy~8J@W+WBIf?oq0BOVmL9N_*seb>_#Xxsx@ZxxC1NP1WNaSh-nN_
z#y1%cEg)RtZM;rfxgSY_svAHb&6P=H=@1ymZ$c7lF{|=SkRSM@LmFH9)Q-&BoXM=c
z6=;339?F_9CUh?=1QaqsXXNqBMViH(56jm8S>E(@_Wjet(%5@~%Dkg7`B69J6`0Q2
zp8Pb26c%|LAftt^j8O;ku8+)fijg46>~syZqD_&)RqAxQ?h}t$o+p$3s(4+?c7R6N
z&-W-FFZW4}D_k>=QE<sLHvAs8aFj_B_@h>9HQE(wG3Tn3Y2E3^02|anq<jeOy`vkM
zySnPnw*6m4)p;W3soYj9@o&CJU)nUM!*Ga_o0GAcXChBDNqDqQ25b>sJo&_tifr`X
zOOm%CDDWg{=iU<kRRCUN_0vHG`}ul^m}p#D?|Gc^+V%x?c;tV-nv3z5*$N7v+x}tw
zJ5@)QR*AyFch_QjI9ZC>?2kK_XXw))zHbbW&p0hJDwoQ|%#24gDR?pu=m0GSLOVsa
zzX9*P9TtN6tih0%?wr${4$1);Q2#Tnf5Q#Gy}|tHnGmpwW5PPw!vUsK(;=z;T=)qJ
zna!>cyFZ~3d>M5P%#08g@$jiJW%+o24jp&CDV2sFv#XH3Ch(QRn}6#NUwTnKz)0eO
zo6A>4zI8}$>i>oBq~i>@lNPCf8#W^Nvzxu&9RT~6eAyc)<VEJM)j@n?WYkfIFr8PY
zoxNq`Ev}n8KzS?`IcAmI!+aaj90_w4`pVz;Ry<{l;vePFjWUX5Z8c#;eQOz!Y}fEx
z(U-C}hkR+&+8@&xAfqW%02Tsgj;&a&Dd308-M~b);d#<r>-Ak=rRj!|d$Y*Ko$ZuO
zN!(@E?b-6|bG&mKQ`8s6X%3zHFY$br-v*@eN$D;@(?Al)M7O-*<*#SvmKyEwU$)YH
zvDu{^pN7vpv@m!bOon{(=CkCe+A7Wm?lT6-+RCEs&1Qzm7fGW+@6uTViD#Jhi5?0n
ztqge&!)ia}N=7Gr_khP6Ri~~pc4v(*`W(YWSFJoGOkHR9%Occwr96)>IfTZS%^K!<
zQXXt-=3lq$kUI723*+FyuP;S|d-)mZZ!-!~u^(=LII3zx{_lth4EswB-{Ype;~7;3
z0eXNA%lORwD3JbAkjFhNm9!l=-2-4^)mRq;Xja%Ho%$&PO_1R|@Br@e5BehIk_oM;
z0!1-aLgR-nJQ?N=@W$6#bA}g~_$^@#Z~=1M8s;3E>VrciqJD{`n2H%!#HYJC1?t1M
zqUakJ0$(?|l(*wh-N2vpBkNXNgtt7~D`Xxf$4b2(Btk$;U_~#_yA>zEth(MyzPENS
zY|l2`Z1FGvBO`0Tq<YfUSSS{VwvhcBhyt>UD*v$$i?5N{lSlAd<VHG<KC2BWFrKYV
zAOsW*uc#bNH&ddXe=ahT@($M2W;0r%w`I&4p|M8uLsvJna(wyQd%%#zdcHF@J0zvY
zD={Sx{7395WZ+$i0DWO{yag23-^@7b3rxnd7aDPDdEjMAfCE~MIiyuHvEh6S3}%+e
z_YV&YCLg3vXzD&YEB~XH5EDoaYd;VCBUv!xegx2LNL8m&Pc?kIZc56}bwG6OMX?^Z
zH9elNMhMs(GD?facz1|fE_9l8VCB$LPlNDl4I*;1gw-SRq^$svu?0JU%x7m3+IJP*
z?plw;AZJS17SLP4@@_^Z^Hq-x1;~r$aXWv7GO~R?z$S@@e?5q5c&ysOw^p|h4uOoN
zJsTn6#8)iU=SjWh+q7P@PYpklni26J#y(P+?<Ie=E^w_(*cwVaxWlqXVbaJB@8W3E
z+Bwo*C+_s(FeOeP6m<^^{>bBpo$x8fMnuhWrCIH_yco#mEVLGSMp;4YcbUvJb!bx?
zFA%@9oY4bQHcPXwuPbHS-Ep{Nk$hk82wCRdn<|VwGRqkYGlY!KatVwAka`A{(jV7X
z{l{#&K~C8YIyAUr+%iJz$^9~_tqtb$m5muVR){Vj`!;?ik?RhqN9t<+AozWGCi~E9
zdW1J$>nzg}o|5^QEiO2_dZ90U@AawT7^)uD05mh!UR!61`~B5n9#F#A`+a8~xfb>6
z5VnB_WBQVG5Pncq)3AXI?_9xnbIWHc`D8M!RCmxsYi(%i_5n*JFNXwWg4=fXR=1C!
z*%U9%FzG3o8%4wuUe`2z>qy-|@6VuLR}*RssmOt5>MfE?a89Y%vaiRhVzP}iqRS;9
zekc>tG(r#n;Ylmu!*&;K4y5wT{wU$<CA0xRU6~4ymxoea=Tjl`QFU^v95p>o?Jp;)
zU8=0e!=p(+topUI?a8G9(6(1kt5lnvYD94jYjUkw{r2#z4Vxw461n@OJG2XUWJ>R2
z5qzDwePOl3l(OXg*)#X2iw-KtR8K9Y3ZfA$4a~YCgN(n?Nv5-12kucV{Mkr8fmvQ~
zxkmVU-PDz~=eUsR?%+t=c@!uWS#CH5asX{q>h=z_tgTsLNyY;evk+_Jl3Ez*<s)hf
zW`c%TqnPPJbNPT)?YxhOH!~XRiDXfFh2T0O?h{l4P-FVRqom2t=T@Iu^<ek-SM_-B
zeV&8!ypUtPpjlf)wi&%wx8>9~<Iw0NzthZs?lVp+562EG%NpBMgV!td@6psA@4wC0
zEwtLM_-k%H3^jg`Gl9%`2H_D|E?0gD#m;au{rveG3W<QIPzHx4*<!s%j)t>^RYSFQ
zWP#b;56^fet@8H{hf6eV^lOwbLHE-^X#yp=FkWM=cLi~bWS-f>PR+LS$F^JIv!<u|
z12L|Qrv~Z`Tj?U(cAXZri15Z^=_QFpUN>;88LDTnzSt{>Vkk4ZCxvdeXNVh#`%-SW
z_kd_;SIQ(PpY0Wg<q}WJI}cq^G#OJJa+I~=PkA9>gq$DHzonR!_jq*!7KOYGuV4m^
zz}INaNmTNBlz7<HD~Og%o}sRF=JS*ou&sdb-qOktXg<1B62k-NVb)nq$Fz%sS=N1`
z^9{g7Ff{}}TdJwfDgyY8K9kS_nKR)xl*NXFapP_$AgF#ITud9e$DH*Giby*EG0N_N
zBOt#A3eLvI2#P2CBEu`JhRi+8u`zN{Z`lfBKDz=uyDZsv77L$4pU6$Rgz3-N&ej|u
z?)m4q69mwg7!7D3E?If$MnZCQ{o7MSY+92Ja@rh4v}hrMOyMW5JOzY+=B=4S`QPs9
zhEpE|NP+k~O(FI*7)+ETQ6?A}v!olZR|vCCL-oP@FsvBRqG6%c?PPGeHqQXS<#H4J
z@Ln~VmcYEvc2&oG(mx!&)5n`5P6Yv%t-@!3m`&I{&}AnpkaU8QJ&0TpU$oc-Xf{ZQ
zcR3J-p9yrJr`m@TviV73L3<^wN~{PIdPkuNkShplzM=@pKu#g$+E~LcHHfr7UeQF3
zk1mVOVCM}7mx&2^0bywTmQ?b7C62y7`xa`<U=t9?d0q<@{gVaoCP2)sqnB}5lt0`9
zcHuiyco5>ii_<-H=<b-@LN!WQSR>+!P@U%jV%hfSKq6n}`EpH9XM#h%=ul8j2d=&a
zL%)rdfgCgr(+O(kP0<z6w4f_g!{U!=d=qt6+K=Ob>Nl2mi3Psa%y&_i-n1}diCQj&
zme`8;IV=ETg%ID51||lGiw#eQk<fg_BI${6ifOrt9hVi#(kO{6^XQi_;Kw}v?q^%D
z8QW=KD+BFr3yUHdvW!j_*@+LQ1tAR1t^~F1F4apg6Q3ZPGm2TD4w_7!q~|8IwJ4vi
z4{tgKSe8KchMWMni@_V7uJ<&46m27PMzvr)!|#M<7g#}j_z~U+8H0R|pg7Kq@WvqY
zF^@x4jqcuTNS^9}c>*Fh(?jZ{G|j3qOsPnB+bFF%*4*=e_HJYInpiIIVx;7~e9_PF
z7sypbMVAj8L=HSsE!f|)(W_&OJptiL<CH{Gu)FwrqUh_Hsg{ilXVFRg4@O{<wBUP-
zctz|4^F=sJ71&_kLkBh!NdMF=S1&;`u}?Y5BKSOjqW?IPX{1UP0=312t*=5$ykhZd
z5A)|rO?+89#&OIpWQ_#tcuaKxPh;Fsj0{y&)D0o@=~XYG7SN<9Kvy)MN_IkT1DKLE
zugH$8bb`5X-LWn>Pot4W#rg@x48>L=@M|B0B7}$*R4xvG&vw?+FR&7K)lUK4>W!;o
zjG0V}B!*Po`^R|>EoeNE`(L_)^ed<eS5Zc(Tow~#H5>XUVFS7k=w!E<fF^6#Dm&8W
zTKj@~q|ck{LHHpYIfZWcQTHy0>)wf0K%|aqMX+<~k#=eR<Rhqh%yL~vA4DtI8K-D9
zhZ(`PQOO0AUETHnTiIQ@+Tlt*Ka8}6L<Jfgd3q*<#4}0;axO~cv*b&|zPrh6<&Dh?
zO&mz$Ude}>1iZbmS}~9Du%bM6Zac$IBDLmEN~On+MdpxNG3j^~nwtHYIGH;w^848Z
zdFelzpB#P#chZC=3~)xak()iMGS;CriH!0cL!|r~AkWmGS*w^V?#QUht)W5n*c@0*
z@bgJgEp?D_`r?{igKqz{?IH8HPw6~o+vM1@{`9v!$8qw8?QO=bB<zSzX2nv~Q$5>m
z#Xu<=8k6B6_p^@A4XYPgeChQLkZ<?>OSqh|l}w2iog%VF!^xZy@>+W3@RSX78E&7@
ze>d>z*y3Uj&Mnws6m;m}(Xz2i`3}5OmG`Yd+rapM;?;3SQXH-ewGZ5a6QtwcB+FFy
zFry^soEQMcYL^LY?5^^9=xEH=dN;sK50g+&&SxiI+IfZoM~mqwYb2M<L9@>4pd$k9
zn_yiY<2}IAOmN5RmKM$yLBKAmWAj@u_{g#4nxL45f|y{pYlsQXx`P;641sO}PFqWi
z9GU{-y<9CZtq}WZ*jeanJ(Tqqmq_c1J_o*U{V6fbYmrH+^o&9zSdIZKpI1z`sBn_t
zpjV9!hrim=C7%3fP#!uOcN8T^^5AzVOw6W#ed&fOSo3<@{+^Pl?&8n9l6r;6tylZ&
zTP$Y~UbW5_RA`>yR#I+{pOrOVK3H@^V}o8oDQi(+D~M^?*x0`9Bgp3Fpn22Kf#J>I
zZa#so+x4)Wydi*v?}8dUfp|$g$XxuR6AHREsk?lyvln`uM|E5DE+<_iP3H{eJvoNH
z7|uD69lh)3?S^m=XE|iHN1rpmQK>eY^jb{TPe%`Cova&IK%tkd!@oID;Pvaqe=ECL
zr6Mc6__qaAzlCn)O0!r*-F*k(z{XVaa|J-zh3_V%htT)Y!`~phnsGVZalo+XdV|<)
znUv%m^?OV-=bjXFUs$Uvk`dhO^+a4uMn#s%>RlOWyJo)&A&0bM(G(}>>fzx{go$~+
ziWMwS`t9jMj)~paFXn{yf2e!wxT?49Yghs4*fheXr5luP5TryDVIv4^LZn-|TS^28
zML<G8I;A8w4bn)1bSmBXuI=&sp8K48KhHVOecylH&xdn9Dz*1_tu@!2V~#QAE^e=%
zM9V0(*o)hr*WGVksD$c(@wcp?WT~zyXqUO?XP2HD>={77JT=0=NOSTdvx_M9wzi&q
zkd4%<8Q?0CPDZpazIrt`zG^Jv6e&t;Dn^Zv8Tu|C*NAoj3+4Y6X=6jTF@91cn|HsW
z*{m-cRX=aLb!#1`oOabL-8vwV2u$<2rg~5IJSdv9pFC)GxGx9`^;`}X(?@q-`kc^U
z_&WqR)?WfQpEXe`+C6EL`4y|-3-8O14!9S^0O{7nLEjl59VCVh97>l*Nur+)x)%9n
z*}dz1CN-oW{_68KK6=O{>*Y30u#$`Cu%9>PsWl`x{XE@Uc6#xDLTye0op`&4`DCTF
zb4@yTAd`JvQTkTc?YmOl8wilvc5B@{2<FiB&So#frnGtY(=f3<CU?&|g3nbavujY{
z>MK>SrcVgSY~QSva1i5f34Ythe6%&up`DWlG@3NiA#&gG7L*|v^VAg90;ed)RoAxg
zDQIOca5d;BY%muiadgenr=@+M7kBHLGg3o$gfY|l<J0@11X24u!Nkpdj~X{1CXt&F
zgm(sWorXZ_Gkih_0=Mngg2?rA&gLlO<iE40;0y&PmENrU9(^iC)n<v<f7~3Wk$L@;
zo;AePOh&jL(15VJFeMoj42Ws*;mP<*C>L}Q&W3Cj!c^Xg;zZ2q6jF$NLhHjFRabu-
z>pG9X+%v*WE4OUC2{VGOnXP(WaBW1?G#eS~jkF0@i>Xer_(`K_nyM@Q%59P9tsGkA
z*7#!nbQ$YfL)V$6!zteS`~3qmr=Px$!`|oF)J-LoSt$&A<-O>0Npang@w!aXrim0a
zLkyo6e>^?cX#ec+(d){u%wqIBXG3fAxHaka2G>gCv&iH&H7Zf5!gKa@@pF-#VdMA4
z_eI&4tK85lm5*0&Y22SqS=OIef5LB1993-p()QCJ-Rt;qjVjwu+A*lRyrWr<6i)1}
zOM6O>d5@RHtaOM99&hG3^p0@O)?-u`w(nHS=m)zStaITR`Gvw34CAADOe*ohU6da(
zIAOvTMttj(cLsjsVA(3F?TT7-Dk!RnD!gweFdb;onCDQ$s{!A_vn-C!isKzsI<@T4
z3)KwBj(U6Rwv=&Y7j}?iCsTs3g)if>o<7DfT75=*lHIU;Lx#QybxZfp;n|pscsu*J
z#D0x{nkY(Pht$fNB6M=+w`e=VTJEl?@0DkI(c>+ARTEh5xUJ{#WHSzEpuoKJux>NB
zJq{y>npP-eTQAC1WT1Cb-}$halcf*6^L#6($ZI;A`VQ}p93^u_wR%zIYqP9`Hp6bZ
zQiymB-5u#K!KF)FRG}<gUx*IDCu20rzugMFAA%e<s1Z96MMNfM(J1D?lahf(R^Rth
zkHn6}N_==RDucO#n#s<7I8#}Zb(pR+15GrFQg3AE(fn}E-f>JJKjmK9E&RY-%UOv*
z%%8jyuRsOK&-`VWd_1ajcs(a9r|JXNJDP1_EAe@B?jU02;WQnf@up&%I%rU_Uc}Np
zuo>Hb5NjGm^gwK^%@NF+D$8JIcK$-SMLF62$h_G~oXU;ng7Wht1_4DKH^;a;A3NZt
z9|kmt%cDwp$wi*Mh+9=W`V!awRYR6(dO6W?jG^GcX58~G4dR%3FWN^68a%Cz7P}J~
z#DhEc+o9@jvPvu0SxcfpFQZRb*^YXNcEW)70mDN@othz2)_b(vT|>9Lb2?qJ9anml
z^pcKWYwCcID*DLle2&#GB|xzGXIybN5Itj98<%o_<7M3m-E2SHoSxL6ZEF!%2-9&k
zpzu0#?On{{b@mZ&cq9^4eEOv}&F5pw+1rZJ%A}VGN(Po0spd4#1Zb3{GtSDo*rNjK
zxp$M@T;!y*yl_8$@BUiWBWN{E-~%wz9DxCERvAlGui)Z5f=Z{E$HX6*$24@QoS%ty
zJ=dfYIF<|JqGqc9`WX^vjbXhX3DKc4#?^nGTU`DPnE}4}UJ0v7P0B6daH06>&AtZT
z#M0dg?$jR2R(FX@1TPsX;_4%9(D8HoGE-~3|A6`w_xaGd-WYhd!*jmV;XL)Ltle{b
z(T&ICgF0bZ&<gi^s;Nv=yDwDJ%4V;%-R^Ifrwrgb{oGaK{?(&YC`WLDcSx#iV5@J@
zH8Qf^B9Z@Pxdv_Jns?@@tIE;)I%7ZOz-88$EE+{^S`AU<&<(Z0e($jk;!WurKN%6x
z#a`kr!u~(r)lnZ#GpbIfa&46hp8k*-Z1jqb+{e^aXpZWhpk#4zoRh>tqrcWFLXGw6
z+OH33e8LWbIm7AGiubhIE!d}Yg}UYDQI9g86Pk5+37ufC&_Ow$<mc?om=&F-gwR5!
zB_uWpC3u%uesl}Y(V@lmHS7dBwpJvFjLr;xce@H7op0;mkj}L0Z`g+W)5f-~TXKBI
zq4RFK)lj-j?npz-*>xJ30o-yi^QOo-HfW=sxNG8s*9Bx$asN<ZJa44qKsN{wHjQl5
zh=&}=+ti)$f$@QqBdGVr*`eicX!Ox@RaMPURRlid)%!=kd|B$BHPrVmf!@TA;_J`2
zC1tJ`t-i7h7;vsYBZ>#=2{?}wh6oAbZiI?GD2&{e@zN<`p?wWEL{H~~<mFQ^=Zn4G
z;aR<t+lGpAV>pQ7_C7w%J6=NvMK9;Vy`)q--lwL+;!<y{$F3U+7=~6d&Sa9cut8E-
z79Oe*ESQbCQurXOnlUk~7sH%}Yh?x*c|HhCm}73!2MZX*^G?Nn5qcxYsUt=FG&?_H
z_4I)b>TrTzR;k4XYj;3YDJyyPw{j)#pSbAFNV-+ALBJ)+gh;4$+}+jChG*?@8dSug
zZGD<BBO*NOyi&v*$2PyWtawg*_qT&*a0os-PK^DYR13;+KeJAPvU{U?DC%|zf&4Tr
zm}HfS=OU%96LW8jR`8@tYkJH3MZZ^hA4H!$JfcRzfA+dR&bCrW*+Abuj`o<V;X{}X
zmD8*CM#fN_herz{@JXXO90r4Q+qD%4ZD#Hh{{F3U<BVQ;j|j6d&Ww7$p2jA+`A@t}
zxRdx3t5o#ozlx@Z@T4dJQ;EA>RJIOYB>Qtio!Xi1eW7AD$R)iEHHG(UjJ%#|3Y}2&
zxk<~r%ml@KVm9A=ebLyn>H{Z@qQv9{>Q=35(POO<sOisdUR|BRb=0J3oTq*SF+24E
z_5pwzbb>ALrKw^dIO#Ge(%Xl*-O}ohdm1k=fPzuIWw0(}uGyg>8l(9X&Zu6sr;{jY
zG$xrM29-Hfo2fk6j8UFWE!ZlTIs2az@kk-&1-#vKi0}U9vlcI+d+`8s>U=%#{2Oz6
zAdR^Q8a1z)-!YN2F_j)-%o$z;I~#H0x!-UO7wU8FisSmZe^e5j2+D^*>iH(x_&z&t
z{ez0K)2$m=Cshl|b7?+J{KOZ>?gTuBHlG2kEHMx%bGI(e{r!`AZFN{}L5*7xUjrXj
zKofa`Wpmwa%ByCpZ@b+SyGcol(JSQnmxkAbbb|sa2b-y`G?_n^`DG>b2esb(49qe{
zolfUe9({hKg3^YG@oiUXLYYXu1Yq6<2q{dAojY#&C+!zsH4i}=qr1~+A}-Hu(J7Yw
zDEt0|ENgxs!Y9-hA56)Xz<RA(_f>iy;70S<y{~szI_|*d`K%{yzxVlg|NYzTfsZNM
z#P}ccBA9U>B*(%re0zN1>DQfAp~eoJDLAn-ZGFW%yqAIF!G!n|9mJLEPpQ`6fxX4N
zE6$0WK?fo*wodf~WvXY4dc_bwrCY@6aise`{XV|F{hrbv;=ei%6Xi{~r=3-=+RB*Z
z#J_T+J)90%dW6|@Ua0ml;^WYKTFGf}rBDoNG-$h<O9sgZ9|Q78YwPvyi4Km=8qGaH
z`I$bctGZr;o_9i{BmN%ys~-xen%IR~fKX~~bVUfR%xtjq_=u_jVfC$a1^?9cY3++|
z#CsMutu6=^6S$v3J-_<yevIG4-wUlenpicX?6Yn|jYUA#Kbba09w}K{4`0A1=j&!}
z&^v`Ce5_noWrUXND8wrJdRDQLmb{Cc{e5Lb11pxoPRqxrdJ8XpT^x>5Qs2CEIj8%$
zRuN6b2R4bGL6z8fdWrRE@A5oookiB`TxrdgNQ32_nJ4V`AHi?(6xl~zM@3(?OKVtt
z0Z`fLxz3MXp3oa`rolcjnkDl4y4EoMi!zh5izH}K$$)lGo+lkO(YFi8#T;s(D{c5&
zr<P+8$+7LSRjM`uT{lD^@no8p{1jDXG=>zS%4GKi{}#ZCFRk3OD+bI6V#%sT72E^a
zG6dU9ts`SYOgqn%d_iaJkKPpM-0`8D&_k^Tc4q?-Rds;@Zqv4b&W{fZ7WIa+J-(Ke
zY6w3^B(KLKnvQSutL`HSY8vEt)S{XMfrjQBQXxx1yw?bArH8x-m>Kz8gA>GyaP|gQ
z4>&q(JqIF9t@j)t!A<<|N^A`4tdImbcH*J&#=*A0O+O5poQA^ohq`VTaJSc>Yts@c
zRR?W8$A5w>I1R>(c2T<B<evK$r54q!vttA}UYs5bHPEN1-hw!`Pvgq7l-kYK&y0AF
zxgtB<HTVKXX|8kr{lw1ie+BW|`Y<AD0<mk!Uw!H0l_L^(5FbOrbpIuOV^JLx7VL+z
zI0aw9Hh+Nb?0r^c>{$GrO|>k%@D)Y&c0%o=*F8BL@b@Q3Y}5<W_}~!0u}+3=-!{Hp
zbiPa}uH$0<@P#{x|LqU7^c^wMcrdo~oAX?>)a7%V=-MwpXR}`cHnFh%d`lH{Fo|aQ
z1LPgT0n&ovb-Hhl>_)6C#;T9d_NPULK>8Ln9HTwnsn#oN9@So8*GN74N`bMI4Pd<|
z=a?IntY#40dDM`yUQv~mY?JN>pXQ1O`dS36+flbn0?_f4*+eVXFN>}1^|$<s?IP>n
z7`&bMBoJxW|Lmq38{eR-J-d+*RTHnd{}FkW=A;aj7o#{(PX$>Lg(=$2Y&?XUXZS38
zd4FoKE^(6jOoxMEJzNXlC7tP^x9jUZ*!pfT>bBE1zKboFE;>{<BnXhlK6`vjGy>t-
zYu0Z4)XP>KlbhQnll5z;EXEh3!wKF$-x<{HW_(Z0-JvA@UhMU)t_ve2HfJ+rwaBzP
zo>>^{^Qy;*S0m}KtxLU4=Oypx0|NIB%j!s5tzu%$_qY|+h9L%nF=#@M33A_l#KRHb
zwlq!_XO)^!M6p33!LKp5OGUfqi~Ml#m6n0vYmI*{Ei*df;q#aNyt>)`#z2JWkb5)2
z^+#di1}jumGUFqE()z)jp@5=i>9oYb6q$a_i(Jf$?uS2ay1c#uw(a%+(RpJRMoc_v
zztFbUq~2hvgZLvtFW}X?U9^<mgrJ(_+4X~3W8sflLy+02Bp){(7Ide=<Z7^b|5-o3
z1tU$O-6uf0sQeL-6;zUcEB*mOS~%XKO@8Ap`BM^Yy5%1AB+5{Hs%V|EXKSKA06FwX
zfo=Ws&9VJoaaZ96T_-JwyK3UZI#AVHG@uG#ehXiKpJSEbfnXHFVl`Y_Ifx34>^}j7
zw{XLt11;FI*4?KkKOcg$KM>FQdcLL}-~mh!aRn|x(lV|BmS*T)%NgL1vD?(FeN0rG
zhZY8amkX7jE*@KP?u;97!v03&#XFD3(XtA;p&{(5{)7v^rn;GZI{j|A6IIk9-r60L
z4v4j1HE3?x7mP5)7J)Fd$%{4~QUM92;QeqmqH8O1T!pnGor<Q_qd{w2z2wSJM%TAL
zW1Nwi6y=JRL{j<qD8xE^&z)TLmVe1(vk4M>#9InZ^rVKm{BXCu#Gl`DL#M68#jM1h
zXV?b1F1{(UQ)Ae{?Dcrc&S|~hSJ$V0@R=)qbLE-)k1pzX&B10ZWRCH{t2I$q-x(#v
z?*%77QD7G#-vL3;a&Q5PL(c7F>{3c^t9XO8+Z)NYdRyX>h0|a-VbT10EtP3<t8dsf
zGC{rnSGvtBy&ob;dP)-p4b*crL**&LSqLc*pTmFjN;17#%8It{nAqu^>dgK9O~T5}
zzIc}>x?YEAYlfRRpgv<!y;;GB+?P|^v+-yMNSYlluEdPW?;2m{UK$-HMmP!EJg$%H
zy$HH~OpMX<vi>!poCKQT#LI=u#0@&A_-V@sUwhAm&-uZ-N98aLv+e$kdHuZ}+eZ~w
z#`?6>N0bW&{JMcBwRa7KafEu6I;EaKRrNx@%1hO^1W*B*TFPR($_sIf>v-tY-$7fz
zHDPsuf}+*1`SNUQ>&H~J!19y2+eYK@uK2M8uo^s-s;p)iOjwpODTaywi^D>D^yp5H
z)?8Y_{x<);yG}89oYtzVMs!~Q^$DwB(^Vwv%;gn7r63JOd0=bR=5#F!iBse2d*a;=
z`rJtrp_POL$9~tfVb2h|>fZRoErvnv1M0$wh5YN}@pVx_)YtF@44*SfsGcL*nvaQ}
zQ7H&N|J56@{)iz=`C*Sf1ez|z3H#z+f!clR9!Ffv`HS3gcqIb8J3o?2=mlt3sZ9I0
z9nXb0PYsNkkK`u_c%3>uRW}q-ERsvb8Z~@4Qq;FtPZK|9h;_63tNVMl3}8Q-bfmd>
z^xkDAN$>+D^H4EKN~$nF(Y1<z)AGCFWN$Ho0@g_Jjj%>?A7$6<C?3XaJ<y_uwcg4x
zm}zNcHn>3=sYpX%hrwWi>Gi5;;Tua>{Bzvnv7AY(eSsmuXib(N>yck60yBgd3wYHg
z2?hlqsiH$4<U%??$Rl}EGI-2LYX7FdFwp#0bs6mx!Iz2Px4QbO@MO$r4PQlu6o2BA
zOY>qL@L1CJaeepv9M`MmB`L?cw}xn70ASyYEI^N?DGtma359i$SqAhEaVZPw?3LOj
zazW8i0srXp3<TD&9m-gILC}`%VntjZpk01Qq~j-F0}nHbI7_>lb%vi8zhRkzD_2CX
z)(e3>{8RgtSM2A2dXK<@v$DDOBhOML?on24Au$FH)tfiMi$Nc@y>nW;N*x6bSUnWM
zG+Tf2`Z#YrfCL7VCJ0xo58KgD2@L8~JL<>v+jPcP_x5AhI3MOd8DAT`z%^TnGRWwa
zYZ+bpF_essG2%iuviI_Ix#;YE7{14bQj<KQsk+;2Jz*^-Bc$zk?m(dEJ2ItqKiph~
zI_q()WDZ3+rntLl4thTnPOO3?g?X#aaO>BN<r*~9>-#$$ZBE<UwxIEFu0eqw0$=~*
zf{1Fo;VWvDu#@VtKFLYHT-S!{vYw_duisRa%vKQ&RnmmrZ{hdf$!}|V-3c0fpZ56C
zm4}wx57x7|rcX*bHj+S=eU_{#PyQ<#;g0{1^Q7X_Q0^lZvxT-u{aYFhqxm`ux5}*U
z4y#@FMq|r>F8~sD7Y(z>ya+B7UcGEZhClHV1MOQ52VI=Jdlfaa=r*a;uA2CqVNF+#
z7_@UUB-e5~f)_Oy#OpasDuQFe5gDhb+dlJQEDsO66jCsCw<c5xP#ZDd1N_)+1)SGW
zT86JF5|hcjqt@)!i3PrMNe}~>d$@4g{4$_%GaNU<WH)RQ*8h(6l}<lomN%t2O(fog
z%b>n)4=|xy^w2fvcbO#Jw;dCY6{jN!1YSGm7R&o!Ba+_Uu@ns(yhL_8BOqFFAx^b`
zog(&L%mx1<Q4=-`7rd*)yqz072jCs2`%)K9-!Ch7hE!iPY$I@%9}EksPSY-FG}5l-
zC2(kS7e2QGxbKsy=<~Se2%*`918c$4q_t4RPq7HW4UmEGF+<cv*D|X*U*8iBeFr*X
zg59k7Ul|DI2`-EqL`7{(hU^$XDqYYX@;B#$V;`eLp117{v)bFmwogX`ZD3sUPRuDI
zi1iTyfETGA-3RY%m&YGP4npUn?a5-aZ8+PkeWL1rN~~2YWHt7R^|_8$?|#<n65-I$
zvFFWJhi3w8(DDIoJWvJFKrZ08xE$NFPhWtJw=mS+6~hDcD-Ws&W)99J_pvuZBE<=@
ziYJu*<pf<6Z_&U(PVJgqF>Xfi8hz6P@e$)3`xnX*OwHo5H!F!1NH0BSNa0?W@H;mt
zeI?B>N5{Eyv46&i(S7Y*18AfPOSr!$KZ#tx8S$ZI^wh21)AriR%})Js+seVWp60Up
zt~2ENyf^HO%2mz)!(a5s?T3(5e(uoe7-O=HM6K`v<?z7d2ZEVp@!XadE)f(zk_RdR
z9%U?h_vV_FJ3qS+Wm@(=wx6~}O@5tq_~KGvbFSlrN!Q1T+57UXW83)8_ZJrD>U4xW
zxZ9}yEnc6r?4M8%4W7)0L??gq;R!tss;U=ot`oQ?v2q^X7Y03QYd!Z~h)qZyG*w>n
zyIt2IB#{}mU}{`=U}KGgA!I#~$>Pbk7B1!48r*puM#JCE`+%g65}jH4tMjv)28oSA
z?Oz3B?D}4m@Dg8SCp&<oxs49a5aQR8VBoETV@_8rdpwO0^Xw6~7-n4o?rshf<>na)
zYkPYHBWiSQ_UblQfN72&e^&x1ZIUVPSUj(Ud0zu2<@m7?5L6>0v)M+!UL&UD+xYe^
zjjYU1=s&f*E}($Qp0=pyI@-SCV(y14-H2ze3&?`ekde`saQ*;HQHq5xk^9oguWZ#W
zn@4zT7lh^C>nE7UcL*JKQJU_6+1inmEy#*{e+MvPxY<(~uHaZ|pAD}=dFOEF+Q22d
z;tA$s@)7@p;LK5SI$pDv_pZ|Hehy$22uZxGhWE<XB;wA!#1?6hCSSI|W(Y#_lcv%N
zRRcfvh7_}U(m9$X%5vhkJBFz>Uj->as65Y?{fF64DlB?$nkCOo0fs~keoLbb*q0TL
zD`FTGEeIo0TV<$Bhsk1jo$=zWQ-$r*xz%CpgB<DdV!rN~K3A7Dxb$8cf{|r!XtsG&
z#Z_GqHiWd83eh6V6xLb85}hI95}?MnZpDc;)CX-cz)r9RwNNUdsIddd2zp7JjrV~W
z;;>H@im)2!O{4a2O{}z@RG5Hz23{K=j#_BOaY^mG5RapUv9tt|PO!S11&RkSx%Wz4
zdHy7rpl_%l_TuFXi4eq&#j~p-j~!tZg0T~f+@^!J?Ai7HL>ywwk|VHUZcISRJGX~~
z_vwMl?59197SGwBto$!tw2oehT<e@ey@m_~=0@J{3Jg-)#05BUj;&9b`^5Jm0>syy
ztI$Vo(zB{*yVGrU5YKK_ltzl95y}BZpc74yaPJ0U<U;?-|2)`PZ9kSC;rv-*)g!H8
z1`tT!XTM6_Zi=Dprt7<NG`4OA0`@5rGeNVvXgr@lLyw^TOr_2dhIM$)**20xHTfWL
zTsQSY9jBJp{Y<uXxnZ-JQ+BKO<{1%h#r?NGB-7U&H)GiCth+l#_Y|xJB<;78ZB-rn
z_IYtqI~R!2z$aufZ3QFvd}H3}h{p?W_Ix%Ztn^u4zUlOxOy!Xl;izil(;zhhfZivT
zaAdY4MHaWM#`^tCH-S=RqUr1cxU#Gm=I4PL34dKK;wbIXuHg|KL823;t{L}}@_XSK
zhSXFjxaCYXA&1u&XhA~$ZR-?rQ2kVNSMW5CV&QHbpWrbN-c{2k9h_|*5iH(mK|M&z
zm8&yfi+V67i0d8n^T^D-u5BB0QaeZrqz;>k6m*7nMOw}%D_2+2Hk)AtZC=}OlD2^5
zo_8U-MtU$QtHm&0FL{?h|HGXrZ?x&pJDc!~&AxrHf*wha{_;jI!9mo^K&}jcsWnH%
zGU7YkG7M<c-<Ef#SpB0mLzZWOjre9|R<U^XB}upKUyAdsi`>!N-dZcFKUesqHL18R
zlus^_6X>&>nxLq(<i)NSC6T*%2AjUxh~6`uto-TVnn&g^nA_gzOLeMHc=K#n*zK(1
z$n>-v%zF9qb}-Y8SVQYZmS>PdjOBeBx0+$w2ii8(9<S=epZTBfH}jOf=y4yfaSP80
zlV#JcPfoV$DjBTSWf3Vh&*<>le1Bt)Olfq;T^W$Gx;5G>UmLZCcfM}zHf?^3w|gYz
z)4inqlwJGiCTE4k=r&%!&PJz(@gmT!bjbAJ$F1Kv8XC2osJc6nufvk3T^Nr?C8%5*
z#nR~&J#A{UGu?agqOjcB`B`8d5qB3sx7kZd!`&}4Ud8#EF|R$|xX@zPMccf8i$;Dh
zNU|fRhTeUCvbO{#SQFn_%Z_H$v*48}tXcom;rvVyuRrYciSCtVKw7%INIE+9T{YHs
zD*IYok6F52J{~S{x_A)FoX-5|t{T}(q90+Uf(*)#xNUn26t%V}Y7UF}Crh!f(>Y0%
zr)!)a&rCA|raGx9T*hLwkjujN?r=-7?UURlGL3lV{fpUbX0*F_&pw@(C=6eFmvjBG
zMlAIN?VSOx^*nVAl={Fq<%7((uZWoZUjw>UD5W&%9UY$aJXH-N2@Mn$4dM+^d<!LV
z4W6WQ&RZ`ji%Aa<@76Pj^rKgt*9Pyt3ByZA&&>o%S5c{4CDx|zyGz}7vGC=+Zg48n
zxS2#4z=!12216;U+7#6Z8!mPQ?=z%w_+{MR>u9xL0owh=WY(tK96xFeS#_Bcq|gGN
zX7CUUMki%5gqUV2*>O$1qL-6b<KEcUBrX))S`Z$T#)Eaos!OtUImAVo6+EeMqR}GW
zOlQt_mb~kZOTjBWoKRK8Rlt4kCLR?*W(kqNs(6unAPRa%cBs_4G}EVV=7pT$FNyB2
zm9BAR0FZ^~j$Pq+;<?lLPWwh>2OjRyRg~DK9c^i)Eh7yk?)pbS=`|S6jLyMA(FpjF
z9>|y85||#(N5;fCa?pNfRFBbRnsoAG5LZT}FIs`6T`H0KdR1Q5kGps;!{5FlW18=G
zCPQS)2c@GYI!}j*;+Pkp&DMw~ZBHskr|t006<0iFaGR#dVS08~?bfPzTK=1LchKZp
zdVuxguFh<jTPDa?DgjZ6X0t;A@Ium=q0W<qEwSlnN}EIG<DUnl$_ROfaXK(MPz@)u
z)oFinA+kBmL!Shz#BHnhFSj{%DW`5G_`WAHT$h9Zr<gdRE366ngMk1WpEx_QHY#x(
zXQ5r_DaYz(Iv#ey^up#huJ?y;1#%~>o_+AiU~Xf$99tz|y&jJ2Ql{sRynOp(<bq?L
zunEKd7(g6LrQO`bOoQZi;DwWB`vC*l@>lJ_kBDtWZC2G0A7*5wh%b~rH<th+ob-d>
zFWk5|<*^oFJQ~C`HMZ&b;_Kz|@5Abup}lioP__tJyVsNn7Eu>z_s#pCvCaA0G`tcN
zE|l~JOslxZPoKV+-nv$~?#(l)>BC5n@&2ysfVu)Sj#cLh-7@Y5(;yiW-SA4k+0OD$
z&D6E}87(C10q9pSs)66vT`UbJ$hX1OhhN;CMP_|br({MhocB$?wA;rN)JQ$o&6i}3
z>5OIBI5Hg8S*^|Mbl2$4*$_??rQVWO_RL%GbqJ_Y^I9b%?Mrk1Sio<A5^(vUG&9`?
z&=}?vz9bm*?l(2|3NPY4TJGVin`9>pEJXdv_inQIXIf;vMj{`2d9<soyzo?SG*fRz
zQMik85e}fFosutp#*0s)sV$A65nO+|O%PjTC)I`C#y*VkVHfDM*Id~5Wwr|gMNt>@
zbA<Y5(zg6dbQi2fk}Q-=Bt8amO*|}5SeIkH@Q$u$I91Q_7%-53pD&5V-_)-==|*LY
zozoB@R~vK`>OZTh8+3?<KgIj;r9I1j^=$vu5f%9e%QqjHNnS@O1mxrQRJ?^dp(=^>
zpWuX*5j5bT)u<XqK$b|14=CO#_)srt=y;GbIeqhT+26QG#*;!UVo`a?6Bng<i97oA
z7ZkV<P&xYu)(uUpmzjx5Y@ZjKl6VJFC*X<QkK*l*k%j?GFPNzV;L^V`XS#08TSWNW
zPs>%?q^^Q#R+UJ8yv|ZqzLNQ3YRmLhV6#lUK91)NUH{J~6(V}~*IbPj`w71Hyg8aV
z7T%>|XzVVkmj^^M28XYh7^&XirR}vp(aF0Mw3*`66|{ZxFiRyjVY>8Ttv=VdcGyi&
zO3Hgmn%30uZ9dhhHhS|z16BL%TMjeE?N2=IDPvFYMa~V~Oj;i89vS#p(x(gIwtdIj
z5bBB}$nIpJRekd<Z{)dzY0yj3Oy#AQ4>Po!&Anj)=7ae$!?^~NQ}eac_0}Qcx6kcQ
zc88KsUYMo&1naquyVZ}n6HJid%W{VJ)z?7-Z&|EmC3f?!=I6JyLXT)37%*(eSZ?h&
zU9`yyGK=%V)>u^cJPJ#9Kirk+nolEpDg7YuI;pCiaWO8jw1r4~j*R+esSdM8A5{lp
z$rCYKuCbxtr=M+JWJ7<-3&U%oK{uznbsN>r#s{%!Y$&V45pz@E+4BL$l#nJ}Ty%Px
zSwAdST*@Xk70V~SWQWz8XMD5SoDyAPTXh*MpZSXO^XJyW9;hB0&eO(DC&HTLYg0ip
z$RdebBc3(`NPhYn`Ec#zpi3<CUrAJO1@p3q8DxTAk*1>wS3v!us1E7vEWg)Fyl~dn
z)_Soo$$Ux|$d4Zo(1r5a6A0hj`Dp{1AZT)|=o*;J1q71EhFrI|UwTK<R(_W0dY<Si
z4aJb`8uTUmL>J9)nJ6xuBz(G>apMGio=mub{}ZPg7CH~;Fp$9?NKp4Gh^tmL68js4
z`VT+6Q{&05BIS;_v2@+^ux6uhj}#vf&1*UwNdH6d`O|<Zxmy??g6(H&t_8UEark{E
zT%Q$HeQV6{Rah5jcxafRm6edmG_{Z%5@KsR2<badKldB{w}zFxdqL?A=t{~2E>m>=
zvSNk#{Lk&*>IN>>_byBB%8gwn6>O&^r}aGOX2g;oZTrxWkzchM##gl6su#dD@9)jL
zuB8fv52D6ib-jL5=ra)ff!1GX#ua~FcDP}6GObWnEJ!?5avbZ5hXL@Q3PDSurpWlb
z^c!uXiPXdZcs$bTL}>&!?8zt8_f2`}@$RpBG|GJS_fAQr7!2fk!zF3rcaPw}nIR;a
zet6L9iQWP-p6)Nu-9y7zctF<gBkO?`a!NwHnD|_fn|jUf6w8ZZxaS_hu<mTS6zkbl
zogabQ0|U2{$xml$=LoAEdX&>!5dAdOG^q&wU`hU&>bNXNzbX_{%-(MPuVvPUNFjtZ
z-Yy=q0I^8<0PCaYv30##)3oJW3Mnlf36mde!w9x!XL(MEAr&>$l7`#Z^Z_GDm0dYl
zw1M1R{)v}i1S|ct^3~luE29{D9+jo5Opuqe%?(b~#Xk+le#ON&4pVxq@Yk_(0ny*K
z{X$c5hNSlPcsfdoIRzwMm~u!UMrsXdnGKT4<-CI2@h)hG@E0y9eO%3Cr~L@n>ku;U
zoSEnScnzlr^x{m#FQQ)7W3|1f8*GE!I^A6@JNafTxF>kp{(R^`2)qu5c2?#q3<4b#
zC^{kt4W}0jU-7Sm+gu^`&&uF+==NAc+IKhsFNpE<k&4>}K_Ok{U6H$hymCK|Fe2zf
za*LzjIc6-2)jpxx6eOzHc(|=dvr_Q81YqDLd1qIAGgD{6n;M-^MQR@cap2ej+~H?s
z)^*=oA}0|tzKAA7Om76!AM^}*hF1VTA<qN)EJbF{dAMI|6FawMK$T7V;rmkO?vWhb
zu6Pa(3osp=50|dMpq74@D%R@FQCOwmJQ=<Mx2D)?T5ZD<=hVH)tCduVQqN9@@qGD^
zDsumZ-8XTo>JkJ2ia>xnw6l*BFxl`_%%$^XRztVoohDQnEw%&1(~l9S9NF?3kJ;HQ
zF|WLpHtPFo2?!V2nT$9;&EtN6i!k}BwvChE&$}h&APn^JV{dz-?i%`H)z(4LSWlSl
zA;`5Mn}!VJWS9?=78O(4eX;eQ@$p7t!BlOUcY{)kzf##Qn!{3?d~tVgR>22}0<t^I
z5>;CTUiH~=KE;n!bS0bj3)2gw-4PGOn3H}>|8o{1b)MA>Ps2d%#3Y!HSfRKM=kLRP
zOJiso7~|H8b8wHLNtPLA#EW5W24!wly>v`%HcTbdXGrcSv%6<&TaTf#h=t3Zxbc8_
zE`%^FX(h{psxksS5~B-+GnC=7<BEB+xC={v^R&MxqNb0<2(7SVWz7K&NdDZm@37k-
zytuTI;zr*Fx9ZbOhx>|n+aDHKLTrZBLO&zp2M`4}jR*^L*mxUT3P?PR3=xA?9%{6w
zCDS!zB-8bN#0&B5hcQMS0a6waQGqXkD8eEGD&owQkAtxe&K){OgmzBNDiL9)l&S;w
zYtE*NH-isbrtz8FndbxeLp)x!zbfiKl&(wBPVsM8)rK<=qKHm~yJi>!ctHcwAidqq
z&wQ?3X4s$+moq+pr@;)t_ry$(cO9N<abC6byIENYrI@`L22BYO?0K)Xd|J%;%6oN%
zGQ`Y{g({}LgQ_iyQR0=)E3<9SMXXGTb=;Pl&ZH?}-nFzI@0=ULj1##gC>Gfk;rntA
zn73zqy`j(<rHH-K+wB2X(_Ttbw3VqhnQ1x_I&?QHM|kevGg!24aB4jOSfJ*lkqfm6
zVzYqiV%^F1i^XQE;5}q1JwML6QXG#xha<^}2vM9VjoKJodqrF4#oMrQaxi*Ph~xNG
zL5!l(mgHH)^`E5c9VW>_-`%&cz^Oka-FKmgALspb>W6ECy%If5&v|tph171WUS6D8
z(o@d=kW7gZCH79ozwJ3tNH4YU1#11tprIp>T36SN>L*7V>Hw|mRX+#5i<AJ>(ZvM2
zi}2nj#;+AJ6TFT*l`>}+Zb9G08K)QSz)wFhM~nrtW!Ma)PO9(M?y?gsyHj~nH24Gd
z9Y|}?tsebcsNTXWjjkbdCZ0Ef9@n;~*++iv3vkhk6DlgxIBM=wtsk2BQRZ>7crf__
z*~IjQAH?@*wwc;<P0*AF*d<YJ9fgxeF9`x6y`z}dO|d_8^s@KpV(fUS>{^KNpUI8!
z6QWNdnp9ipr~3l$NT}`_ce26=GZ;b3o*RWTWBd(`_Xj^yHkh5?K{XgS)U!oI!|0dH
z;;WPg5MImhFU7YK@Hv`92x16t{p$6zK0sUUeV@nm!ib+Fo~T4s@E0L0Q}!APdVbWk
zon#NDAkFjH3P6Wx!fGMD$ePc<lX~bzQv{j;q0Ei;<Iic%-@Hd-g?{aIg6I6M@acHL
z8$iwI?uaD8XKsrs%H!>fHmxtdh01b5yf?2y)L{!=>4}>7$;#Y@uBYDxaPHVJ{Z``C
zqig#mt3Zfp3#aIo-xNKs`~LGU;*9xR@ZH#`KRdhc8-eJI8?ZRu5D!1x{(-UVZXw`(
z%P`?uK7Jn{?}IE9esTW<fy1+U+Oqj*l?v;Xv_6_u=%p2yX)ElOf_9JY?X`B(47Y9V
zLt@aJI~xk6|KmZ>`Nd(HWY|FrJ{wGW6k9#$##T=id<6JCTwG6R(ia|oaerLBclquc
zQPUN-Tk^v|oa0s!d<B+SK%wRU&K6;uwn0qvhUUe9q%>IP<|qQD2rYqHG_pRd0X&x7
z$ZOrt>p$E-k%1|!fz%_MQzGm-JLgGg)*8U#nSnOFR{v1KG};Xni$od)#t$quBsIEQ
zu8tR?OI>%2%ts5~>(_X+gc8Ou$>SK3+%)-9VI6yILIOr94n{yAWrlA0eK_E^Ve|L_
zl&}kk{~53Kr?!<Q!|po*ShU7Aa*1dSe11^<nj;%<)<D7&9p&ajnLy|`SZb<rF$?h2
zj*Fx3JTMfYn6f!VhATJl<p1Z9H8BD#yGcbLIPv##5D|4K%+a;rL`S(QFJF~PCrfxZ
z9Urkjl()@eo12~eFscls_C#$Q+UgI5Dc+YZTRy|Jp$Iug{XDAgp#&N@)n7keZ|1@>
z@dZx@J&on01H6Y$P_(B9K79O1GbR|1s;i`(h697z=@rSzRi%&;kK0Ww5*UhyrjO(C
z_8h;@<@r76)J0g=8;));fh1Ei7jZJDjE3#<$4+#3du>7v)2E+>GvAn7<b7iV6bp9i
zT3%+cmaQJpA!Pwl(O~*r94>K%t0d(6^ue5+CZ(;pIN$4wiQeGj#>beO?f)RRXEkbl
z93xe-^ynU)5T>1;pwM6mZ@WLZo|8`_#QwOR@Bexd(nY#u`js}90>N%OU&2*-ChizT
zk;kKgjgo!4{<(gom5#}STc`Q=fY>5SHVAvB`CJVUztz!T3VT>izU_t&C=OT#JA^~i
z*|qX>8tx0VS26h^WDS8yPN``(+4oghalfz!0w?nxgaohupl$vpZ_?9WvpfLp5XD=S
zpsD!=04W<sO4!kiuqru}6NPIdl0vpK(7n1ZM>D!bUx^(NE53MAo^?y~U>nh*Kobu9
z5OOzO0_xih_G${)pL^?_G+^lL;k(V%A>V7#X2hL8$)f-IyZ_UlO<seam6X|M7W@14
zQDXjxr!MUdtN4)+tT>>8gQmRs`2I^3B^=2CoZr6==^65OANQh=^gR3d$A8lE3^9dE
z6)(=R8v{QVW0Z%Iy!n!O58mE*{QLL*r`PtI@A~Jb78yaX+9MRCz56E-{|~k4KmLCb
zGdL6<ANlsN{N1bn&u`Dq^A`94l8EJuKQ`_E_5bOHknz_Ed;8?~Q~CAMb!28i7cr(l
z_p@^AOVTn1aa^>2x_X&NtAaHK6`wy#Z$&I=y*rM}s!0+amHE1#?}O$vGthv>us{CP
z0lJhtz_Km&%i#nE{0DH7-~!y!5>DMp)fYd;K5X#-d9`MqmKtbgFgb6GsJTk1zm^uo
zoI>gqjoWam*FL^m1de&C?To7||Mc~l+yF~f#&DAY^Y@Nnr9iv9JhQC^-M%HT#bYZh
zN0pbl-W+^C{5c5#W9ujK!fv}JTnrK986u9$&*3BfH*idV!P>(N1w5|X;IvDbuJpcl
z7`RK`32cKXwhkj`3`$MH&cSB}K_5RNIDkN%zz^`zIuJoPO90~oAu2d?g?|6d?>P9k
zYgXzD9;wbNOk9-TZ<2%r<zKyCRNAAcD}tJk=M!8YS$BPEUA9KZFjV5fFe4_+8-P)u
znIvSJ0w&=+cq?el4rKD=?e^ug5ZaMEZFPK5A%8y;@Pd%78JZI!lxn6JALMCq>691+
zO}TFr0BimJsc<@D`G2~>uhan*@Y#(&kE35EmY(NM>n>74WEzY#d4b^7P^0;V#j{-k
zCxvnt1f>qNoAp4<qHy-oKdMcA6Lygz;nC%Db!Kof7edRw<0>i8gGOf(bhnd`QK19i
zf2qsa10H|=g1<XA9Z;~|0_`d4%zu2PABy5ExEK>2W+`_n(t44w{s64jH?j9cm7et`
zvw<Ctzo^J;yFQQ+Vwh_RzB37X&EpUV-Ixr#4&ETI%0Y0w?n{J9{?jNS9gGExP3F*M
z7ydjE(`Z^?r+@d{i#*>SQZF^>BpR=>&xqrG`lOh115B&@c*8R+AMJj!u!dH_ldJ%w
zCxNx;_Jn0oB-SBYb}|Dv3Xvk<=rnYL|B9j+aS5c(i>F>pTA!Y5hJ!8^dv>aK4GG)B
z7B687+L77W(ZMVgQ-Ib#{2+THSEuab>^Kr3k2^r*P(76*<Al)xkH`0woESKvpjGq@
zE&_~iDx`|)Pd}C0e%EK$F3jJOooc$a36}7i$~gW_e)+4zOY&&HqnhFdc#=KnR`7!Z
zqEqi#4Gg*DEQWLFk%J<@zLDXf|9OOqDMaW~1&D`A;L=<|n(F}TiN{)ndTjTQeW7l^
zJL3SyZEufZnD1zhuouuBVeP0XG+@q+L(>iN4v={+0j1R|+lC7*q%lT9{U&%jQ{@#g
zEUKBs5r>&du_ouoJA(zfLXL^L|9Dr30Dy2=?=`>d@36Pcb3JsU#$}5e>83=sD5HlS
z?g!Z`fR)I8(k|joc={qqU-0;QZTb;%$R(l6#cnsxE+7xzD|^<H&hxP0^k9Pp)-|^A
ztS^O2`>FGrA2VM7GC0Et*|-CA_6muWz^MmQ-UA%HBX@d8myDe-9D&9~5x7b?dI37p
zqBKZVd^|3&${3;WnN9)X&by$k$*x;zMJP@%U)6AVs&cjiIs)BDlGB1HNN#~%uArWD
zz77l(y?gj$oH3k?Yp^e{s<*Gzc-F<b-*`4OD}~-Jpab9h<#E7IkSC(4+&lU?sV(ZV
z31zIT1#HsAg3mL9qeg7{uqb)=+t2%@!}mvkh`&3YE0!@zShle-lqDeen+NS!!^!fy
z)01a9hT^CQ#TFcY%}~)MPvA*5YSi|cNu>+qPK)_vy)SorwNmUnfJ$B44R9zbgh>$!
zqW^e39wXPoqKjYu_j;sZ0=C(s&H6JNF75zW5jd?AMs#Rq#q!5ME!zbm=?(uH@X*ir
z&%i;r;m=>?hl*h#MW?q{11LG>;Vv@^KoQa3eI{#%NjNv#6tD!;p%K+&BU?y9K*6lr
zg)8W}-$)mN-2gUy*_9B?ys%K(76*t2h@u`2y&@~C!lYbJhBtyFeSp7-B}b69zFXEX
za5K=Q0+wBHv;d&Ll+`uy1a@+Zb$h)--qKE3Dq)aM(}0KdmK?FLGS9bNE7-voc%7jJ
zoHqGltJ{HZ!4*?Pc40kPsRnj_b|1L;R1~71Y(M}S;f|ZX2TxiB{$%DQbew|732%d$
zO2qGZ=l~lX!l0H(>^klQNtr)^<S~4qFAszf%=cVJ6<fF+{HGTH`g8hQyUd&qjm8Z_
z>A}|YG)Jm(EwZqpBgNt8fb_Qkj#@0RPLz!Tjr{*NX9c2m4Iu=}+~bzO!EpzUh2|!1
z`kd<T5&{x&w6NiKxI>!<%V9Kby~!f(N>e81zIHpnb#X)spMJ`6tQa0w1=(pN`ik`W
zLXs)@0W;a!y6>k@{%Vsyl%nx9P5w0erru}NY-Imtv*F6y9tUH3bboa~F608RE$nD}
zKVN`(ql$%9^VnRNb(=LC$R?nMikL01vuVUZ5pE$>;nW6K7`rEgfZ-B0Lct&sx@-8Y
zPZ*XFc#Sw+0+ZPq*su_YuF_rLX5Nf{8(a$<e9D<xf%TpSV6`&m!b;M~ER9a)!9&#3
z4Z004iD8K_*yQ4*VNlIoR6n|y`lrd7pT8`?;rw4e#Qifpfze&Fv^f@#m4zh+e5-D!
z@le(E>E0krg5V<@Xyi=|yic;A?yDc{1FD?X9*WEb5|<Z4x{qt7fe;!^wiXH3-Uqk&
zQEtw~4IDB|v^WTztlo+<Fc3pd&fEoeV>gnJVq7v1NUDWZvM5jp_Ex^M>&goSVa&o}
z!Bw;O_Wfxx6Q>OM6cgazguTyQ*Fc1AIYW|^i-B36kOi9+C-KGbBlR_=Pd%-IyMa61
zU+&De0{=7>aD_}~15V$+W{UrZKa1Q2Z$4T1zC!TtM8^or6DUkjRGcAJO)sJGL0nq-
ziIw-Xv+q+|!|_zU8|9AG=`jZe0Q1CzwV~{dEXNCZn@wKh+&Pevt!ITylih&JYyq0X
zfXYct!AA8~YrW3>YEZGbujy}{sA6N3Wb(($w(T~<)<t3h=*ndR&>co}%v$v<0|Q_a
z)!>@1p8-X+#FPB6I-|*(bF)CD&j1c%q^lG%7*FU`SiJcGis_gI;8q+1%+bV^kpxqD
zn%N1z3K<x6%iuTI$h$wmgul(&9FPt))o)89$bV;TtSDeZ)saQPZ|ZXrgqy<YL;dNj
zK(sCOzHoN~L5AsFPyle0B82%izpil;!K_Cna|v&`p4K=m?t(>9$$gk*vOb&}V=VK6
zpA8zARtVCAC#y*j9c7^MeR94USw@I&bkjH#axn#BGMfJvD?kqIAokt~uG$CgStH##
za(XKNBN_g?8l_qpET5j5PFwh|<xA@)_RW&vti}027w-{NQ3I+(OTKZCsg*Z^O4u1i
z<EbDHMoBqud{OIeZKr7H^3({HnA0j9WCm$p+37y#@sv3zpqQZw+=Lw<QDy~aINSDk
z3WG|>h7<U(CK@Uke(t_8mli2L_sr<f<;NI3WIR2;KjzO`>o29(U;ir2jf7+p9wpcV
zI13a@tix!*1hl}mP%TAVhc0ukEDikw3Fn?^nca5fq)jYv3wr=;NST0B7Eatp$YdmN
zEuwoH@U1?LI~F0+Zl4K!A>G5wVT4m4pf4yafb9^6ycLo0`r_;;x`vw^xa9B3laO5g
z=gREg^HM)q1@J^=<7Xqdf9(kAIr{!011YziR`MI&N2g)vpKZT$gWIPFms<38UQmG4
z#eQruaI$sJHEPK=YJF)>Y!7aX(q}!C=HL<=zJv4$D7PHF4>SjC>f$@{ad%{of<VC%
zORn!Kv;7qWm#<#RTL380odIU3I@JzYKY$4x-^!5JEYQll0`1<XblB9wc4=D+hDi{V
zc6F%>&2TOP@$JT(4F{QDTVLkniWX@rc9IQ_*+YOpG^n^n{L`|<!5<s|k(h{xKWoPp
zBH!(wAdUT&-Uw`Yfj;8#{$1eH_Xe5T0WTVn6OdFPEnXZzl;}n}N`VV3oUjEhvVhEg
zzE1)&ZLVH*DcNXsD&~fZ@aD>k%_%mpeTK6?s=Z@gN%g7s1Y`;aK)h6Z8fEGlPR>K)
zj4OpK?zj$Xk+mYowiD{T&bk0O3qpDy^JK<zPb~@1-^i3?r+~2o8b0Oid%%2!v>H%V
zD@Y|dmn@<3alpSf9z=?V&%P#pC^%3{<ZVihvQmtsWI|f$kdgYc#rp}`oa`>~n|6`N
zKx>^hn6rV;qKK(=xT#Z<?Lhk7U!@_iB+nhfCGfqA0F|2p5@|e{4Io$qe!_$KAgv{I
zla$XaR&WKczgwVBoDt9yqcn3K&^yqGRiVdY?;yp0-^B@?<*=n*;dgkrO2(&10{t{&
zJaH41iluRwHB9TaiW&jP&FK>Ls_fvm3PI!+8rxg$i3KSI)-Fm5QcZL~I_18#VEa##
zGS*Ol4RD6$w(<Un#1Ns;ecYJ&QY+m*k|u*{K34nzndn{xn@Wbxa91x9S-;0$DGub}
z+?9FtYzi1^?U(pmo+#4#uj0&0V%ARKx6^Iz>MFYeC~|i|k#?<{UiqDXSyZ{z_(PuR
zjl+eUjHuU0J6<o=(&#gN_cbQ29|8ADyf02g>xxe??5SEJAwx}+bH3M6%dc@d>|Jfm
zo+I%_kTu6c?HaE?tbJiU?sIiv3Y1;`u84>9fo>pn62i##+X!cnphBvmYhyV`7M{#j
zx*u3H(ZU#ULAg-{bz6*?p@W^^`7y@Z1WT7F#WIW1K&<D-K?OY^&fReC(9^_aCI43L
zcD38yGwU#j*3G7CWE7y7&1epGgh(KEY0$DjK4_$xf=L_(AZC3qNgL<F6Lui?))>oN
zm=W4iI$`$SIezwuFeV1QRS<0dB}ZYuks|bS49jdZxOyXFfrqg46(=BH-?uEv8Px0M
z`lomXi4(}2$@WQ1j{ht#uV07RdVnKT@c>6<EFM&q;Y)@Zc->btPW>|cR%5)@1^6>g
zKgKlZrugy6=IJ;E$WVLix5!53if{9ClDSjoQvx<d$V@X|1A)jQ#M+*Fx#f!1H|HlP
zHWSn(PvpZV=mG=mAo19iBl)>Op2bF(J4V$GOP*hJfFWpQX(+D@3`+{0OE5w~PhvG)
zQ=z|1;V&u`VhIkhB|vUm%9q4p*&L5PMv%c7HfOPY=w703I0*;NAw5<RGNG7+1esj%
zBL%>K7g>Mms^$auXLx~7D^v`AeNY(2ShxJv5=~Y_7*hrAOvrF=q0S8fB`6KL+mR$-
zreG6vjU=8MNI2|4QRse^XdE@L(-@8PNd?UB%xIwK7<U+hhpc&>1Li74A7FcH`#n~{
z<OOfcgiFgw*YftL{hx}P|3hx`$ET{LkyqrsyKR2IV<Hj6mq`M%i4G+Yf~_@d7}m?l
zh0Ed5*7QWb^EFeC`j?>Rc7TVg(OxG-H#-TKxmovO7-k;1CmF*Iy1Qk3$<?>bHiMqy
z+xr<J6N6LUO}}lt9e-Mo46H!RILC3DsO6&}(=y@|Jox%j@DC8g-OwF?r+wTKLXkaw
z1FS_9DH>7i{T~qG(xu<-K6h{S+lV~4mn>J~)|#?y3t*xc-HToMMU$6em@6P=GPr;k
z%wDQldIlXF!sKZh%e=Fjul*G~89Hg$B}F`03{N9ugcXVU|Dz8&Z6wr@$JXuoPjBtN
zT<*RE0Icy!+lKtp-T#*?V38eszvCU#n>_zC#{K2;-UfvGGy)PDhJU}l|H;4j=`jH~
zw(uk8!hZ`0OM;NQ4*>PQ-FyG}3tr_R;oNU=rgi^l!Tx<+8$-aQE&0GT^q;Nxzx{xJ
z%X<H^LVx?$V<3C$$G|4|FL#i1ArVLewI9Xr{Fh4zS{fwyryw``FLzMKUy+Qz%d`Kw
zkh5|CK&53B|Hp{yUw*(}^7emQp}*Z2o<Op-iW(I1w_o_5zu^CaAmEn|!S7AvlLH0N
z60!JyeHf>h9;C}-i2#1r-W`7(phyRe(EO*hf(DGeA^+|6`iP}jsHX!w(q9)fbzsjK
zezuu50qMpGrM=`oPK^KZ-1ybfKz7d$M+AD}DQID0fRxGX1Y{Arcji4Q{>wmU@>Uu&
zPcwn3{Rw3Rc$&+W5huL=<&I`;L8gH%VSPFO6!QMxqdJKsGDE4qZ5Hu6L;1hllGuD8
zsN>kr3#|WTr~kvh21SFA7R1wY)A+x70DiKn$g(Z@I70NlS`6tqWaE2I#j)+r#`j;=
z|8M@91_6a_F@!}j<NiNIUjDZS^WYeqXV~#zGT-kG$UnKr{^e`_|G)gwTG$wXQK3E#
zplQS~D?pu)rPbUWZx)xW87@$P9*B{OXn{}x?40!$U20qet;Xo07FxViLGv>Um}cAe
zWjmnQW1vcDmzpTCfz0(g8_jQ&K^pC~x#IM|I6w)q01;wS?xel}6bUE5chm%Y_by19
zaVF+1zI{>x0DUH?#ZG8a&b?=&sKrcxyEB_arNdHJDIhdtIT7&ma-$$SF`fVmDmCk)
z5!tL-0NoAdaneS}2b~6Quh7~HTfB*@qa!W}MMn{}A-Rb9e{JM1V)y|Wau+D$Z+&q;
zAd0vi;vBR_owBt7cT{PrlS=sh^SD06!<g6Y4S(hdm|U_5h%O`%sI~2Md_v~SEInkG
z)zR9YG&_|i0o$eL1X*)eT8$fdFITq;^S4zFiQK3BNZkrMTI`I?0uRmot5_fA4$A*J
zQypd~I&9E7AaZkJVJ8fqZw>E0-dT7=Mj#0WrtAXi6{g)MF3F(#bKtpw>9RScL$@2c
zf~@;<ALnV+4Y5%-Ae$1v@@o6b3IL4bk@I+tpuf2aBv6#j@{G+f>0~f=z`ThL(()X#
z@!p|}?O}z*<0Dh^$t&<JCcw}A$?ed?Uz$?hXx^Vd$}7qV1|od1NgW~!#q~G`a;8q;
zkNOC^6$}UCMAmG|N89s1=(l^twmQIEFjIgBx${;6WM~(Z=Ure9Ym9^+E(H-T_h-vd
z?%C-b+q_5FQR1NV-r?!kPSU$SgqyMSXJhYIT$6<y10=%f%CpJ;hJVQpDCPwMZ5e`o
zA{XLAsfu5n$p2ahQvR<YlAxu?Z#yHNB<WR?Lfbm^X|no8fOq>Aur3r?&&im|9;}h_
zIJCMc?rJyK$NDfcG^H;{hJwPViQ))3NCTie`82xsCo+V|gRK=oBbp=8r&X~TX~}yV
zi1d16UH>KO@XOJ01O)*5`#5khbOP5N6EHTR-lU%pD)1}<xONU5)qR)2B<$Bomh!3p
zZhmnWTi!F0^3H28B8~og{W%hyK;knfCG(&KXN))mmy|LnUX7zLM79BTgClxUyuRqX
zT5N=P`v**#W=Je8CJKO&ix9ssSA1&*&g25U>W4@vz8@=XJO<rJfb<He(P98roAr&7
z3xTCUhd2R!t_>bJ+qxYEYtv$Y>Wi)xi-!tgw>y6O&==q*ZkKr5{Pk@26RD8v#mK&5
zb1)u~EY-8_$LyKo>!*nOl2jo<`<%};%)ty={}R;V;Wua_NSs2hKZarK*8q=k6@a7N
zSmDF`u_F572P?AUw@KCf|FHMgQBk(x-sr0!AxNr#4$`Fv11Jm)QYs}FfJn!P5;Jro
zf`q^T5`vT>f`qhm#|)h+T@FYy^w50w_`Yu4@$G%~KL4C`)?UlyavAtA&wXF<i|e{v
zI>HQw(ARkB;;+Y#cF$bv<kXQ#_@fMFg`B8zDYhATkhj`j<K;oXZ%%qJ{$Q5Lb4jc7
zy}c1uzEeYGe(JikHK&UqvU|owvL^3XkaVVxQLReYw-`N&^*P#Lsjg_ZDIMbmHA57j
zSZ{+ZBMR}?pT^d`G?daQwKWh2ZkrFvVHbwAFL}?t)-N;<K0=(uradl#S}x75p6~>S
z&ejeV2zuDF2kWxfWr9=&0sD)vfpV$sG?u*2Pj&^-qaB?4Y}rAK?he5M%E0wcup;z2
zzM(WNx2ITh&X|<@fonHU#T<*+Z~#Pj@lk`&kcB6YS{y6bAvX{e<9#H5X8=S@!YQDi
zzxH(5Tju*q81@Q_BV%<}IsyS+{9ZP@<4!fIL0*vdh>%=-pWph+rVTWf#6;ZE?wb(+
zG;CHRUWO0{FdEly?Yz10z^aoW{V@>$Cyd0w61Zs)6uzD(H<Kzr{Pp}0^#79I`%jQU
z-je2lI5@_8r3@Ie2aM2r!IA7@43o_KUjEk_``7?;bm!3;V;`9nfzTc^s+ZFhtzhxq
zkrx2QS83PR6gSH5G0mPk29ETFO|Z>JCwkzjen~XI-MANY<qv#a%U{d-U03p4@@u`#
z&+K=?n9S&KG*R4v*pu~wVzvK~&|Vv2r;OC71^&K$>sqJs4@&w!62F{1u{(Z+xnMNx
z2&Q;f*NQcWVvj0iKvFv*e|kW=e7lCq_LR7EL%Ihz0$V?iK!TsW1?7(k#=}a|Ue+Iy
z1y33)oED4gX+n7#=u$HK30i>1EC(`6O?|a_)!T0f{Lh}BeW=2Z**Jdx=VW5&TLlkE
zDN}dK39>vuW9Deuj8)u+CQ}x0wiO^5_W}I=*>IVJ#gr5&I@}8==JxmDGFcce6W$+L
z*|~iOe*S^@-)+}3r<1o?BfhSHoe<9V(!7qp_@})mk$83b7(n@nKnBq|z=Ayn3xATM
zjpi$EUfr0yD8mIQ-n;qL%6S_s)ry1jO}+g(+avNb?)!Go$&I*x1!ZWkR}3E=<NL9W
zkHQGGcM{AA`CR2E;=Vce@u|(>9iuP_@Lb{U=?~p?Jn~4)Q&Cr*26X{}jg7QK_Mz7E
zu8(C2-lt2RSY_r=NuJE8Np<={dK$R@Rs|LtpGrP!Z8AFZOkyE>HqKFIF{j9yl!xPp
zPspCr_IBD5t7bO!b9NIg`s<M!D+gg-`MwH=pMmO8#fu5|_5hnmcV6Y7r6lQ*X1_JI
zbM^P<@E19!%3s*Ah4*zIkMMQAL{XJP4)=RpcFe~krNY~WQ%)UvF&jJZNo#lVwl+x*
z5R&H+?vYt;X#GwM(dTOg?;jv~ooe^UhO}7B_Y8jUu06cBZ;XBnJ$mleWQ|zxbC`g#
z=R7%8`jItPx1fPQvUO}R>PXB)--S#c-%ILrzV~Sl8z+tfbhS7L1W&T+r?!6AZ4u{B
z=JQ8?)$meJm2_*%aa4m-Gb?;f)7PHTu`Epj(O2?`p0be-z|cSN@J6lkt26M?Pt}+d
zU4<yFCR~4deuMusX@7M7W-K9o3E!Lq1Brq&alPJ`7mbA8RRF%zGf(~X&LUuiP;SW&
z)9(egduVu_g3z9ACYHib5y*X^M;&!ZdzWF^W)uYjfIw&7%%=I_IFn@kBI5P|fWK#W
zpRz<2qg9x}=8djzA3vNGu#gs^%LBYh&rr38%a_AeGSbd7af9<;vtMsh`(+I+*^8?e
z#Y;9S|53%#q&zFMJv*O}fNuj~S#rk}dVkhSGtkGax-nxN75&6j80jVNZA6;)NHzT4
zy4>YoQ58&uQ8j~(+phxcge0S^Jo6`h5{b6LxA<2rb|S}5hDnQTz>>e|H)^<#F?^=O
zLi8H4^3y=k>XoN+o<2%c5sND7bUu4@W9cJ5%K{Wkplb5vP5Fb_6IN$keBsBT9NNRo
z`+3yBoNn0hbipUEA>n<jWP6lZii}sWY`D~VkX?pW{9IRmxT8@-N<+N@l{jYr5m3Kr
znC$&zNq03_lNnc2Lj+!ucleXjn+}r{f6kYPqpBYU8Zgu30YGRl3s64E<a)(?5b#xz
z(*4}|z!MxUMwsejP_$TFC;3c#JNkM^12;sC>*(n~t`mW<fCCPlY7a<!_N*SZ>2V8(
zKD@0wI|X#m-CC;cO~h>HrdVx{a6ZS2A7}cLfEoj{6QfQ_`0Rs48gKmZ*+)Q5c-9g1
z2JI)&?a%2ekVR%utW%}5Nor?JwKi>?+Vx_tvcD>=-Aa_%B>gWX%ZmM9CUXF~XU@bM
z0el`mALk`<ykNsqXeDKLbg&z_<JvgS18HI`dH7@qf9bQQ1ErrLsdef5A9}iDm|{%6
z=ltR1w?~*i(Cujk-;*2=f7id5dF?PXwMc<TeV`;tZbA=4Lr*QwssL>ey<4Q<EP%Bb
zIy}h;k@)Q8lVEra_G<av;>W$k{F*|>b<g5Nhsa1HZ!&f?y_F&f>RA8qf^F^ef+8^q
z$KpU9OHBIEwVt?@DtgW%n&&rkXT^HrS60*eeaL)zC%}5RBAl5v7~}KTXHbS{fC|v-
zust16x9LMoaVGz4T>9A%0R^dy#!H^$I+dAB0fqPlNzVwEv}x1W^6b+K`w~GjrtX{x
z^v|25b$*rFEB32EYbKeW6DY)#Q$xaL!p@)x8wAr36ia6o%K?S27<Rp4)&(1u(X3zJ
zg;Us5j4T*Ky@8wX)3nQT2<Y<J0TIZLv!#Co2HKMzQ<+%~uRhyOw?x`Eo)P;gxqjki
zU<EvJO(?A?o6)iY&_Wy+xLiWKI%~lqK*L2!p2~_yC1sS0<`;j-`gVGfjaB|^>vvjC
zRT^>?I$t7Gl<RG}v&2);oLYCDKS!u&+@_(SEvUt3c0YOIUIBMc_i%>09B=I}YiMc)
zW@S`lFPz!>CT$D$71|E6)IdN1WeXCvQ{{3=g5+8g{n-A*@zLttJu8%77vyXo&g~-T
zNs5c|2UoxfIJ^&N5OPI0(u?x@Y{7tL#~*QVE)N@J96S+()v%(IoA{!j;I0O(%In{B
z8<B^gk*b+;tr^sAMhTzi5~-;GMd!F)&{I4R{#OEFQuo#3o*T9=&OP9j3O~?}h3PfP
z!`oY%D~qrGTq6o?B8NB+nsg<SCfqi8Ne4*h+3QQPpc`b~FKSJ6@?;TNF7CJVxn1kh
z$Dp?f^{EDIOgUlELp90863zl(eDXVxak$|C+7j1BoacPHtf*4_{<59Vh~rh~!NLmL
zB&To466-Kr%6x{6b9n-wug1Wx;fN727wz%Ugp$GCIwI=n*o3=k$Yr$?amhAqB6v_Z
zrcYqcpc$`xDt_fbSV)=Yy>$gnLWKuKv}8!(l65N`oL+bo=!!0;^yMFdv3?a#^Qymv
zyTQVd=dlzax>IG_K+Lnqe<5xEONPwMPfbWv%0ZJfz`D*1=$Zy|jRNUp@`^r~JCY>s
zdp|1#^rr(uX%U-m)=03tZRYJGqSE?n$k0kl@cgbijiR#?_U%_u@EG`1%JD&l&m~4P
ztpKI-dYsb91v)%488yj(IN3aKZDe0?t68gYtxSN|zV5!6Sn4`*w$v8Xb1pw%b16kT
zVkK_p-9*>v%q$!p*zV1bfZ0Fq`6u?imA>&_EFNF9v<E9<VzT-xa0#c@_7?N@XOF;h
zI;Z<PfAQmc%WO|&e|C8hG;-sJhua>F;d|`=tb!%1jD@C=TZ0p3`@Jc++(J&)yZM|>
z8Q2;kh{wCaGlV@egz3z_Szo^5FmQDdz7csS^JUs;8n3ZnHe~VI3A}cA$>5QP-q@`?
z3+L9w8Aplon8%X)28KKnPHkNrw+vNW;JqZ((zK{Ki=OI|+x16pq|p%4>Q14`+7aB?
zTv0<cP$UZ31)iC}Y9`MPSteW^pIOGtN%TaT(>t5CJre!;HT=hz9atXf`1+B3@UazG
z;^!=TyxYA&)i3kUVpXtimyzbALK_kR^n}Rl+=?uPolttT5{Yhtb>Xe55j(TF=9}nA
zx_1ok^{f%pJjM2q=?g!Xnu8YR=Pc;3UdC>raSqoJI@GuojUP7^Ny(@}nRR1^lA3)9
zBuVUs$ej|sB-;Qc%T=^gA_r7a1kst!j;RN2b{8`kbhSe5v*RcB7JKKGA@r;%C8n$z
zlmZntB}3`0e_g@U^ILc^fMvkGQ@7sIW9pfm2kRyMEN)HK`G7h2mM;6au}!7(AL=CM
z^v)XR<%bGNweuIj5;JbXh{9N8We&@4`>3&0HJ{f96OA^XACKgdQLX^F^Qwj7Erl9u
zJqp|%7?{eF5;!^iMcu6k{}f+MuRv@T#St!5Xr+A$UnjR(jq~c3MA@rz--K?*flav8
z+Ghf5K`PrXM@7OqhlKChwng*RprDt9w(IBZlHGo+VHpymfLITb7bFQR_EKOm(eomc
zK|7JhV?ZaQmni_XJ=OfM%y!2d@)=~veNWH?{1~4SJUH(ooM}GhG!W*pYlGIVJ(v-!
zrk@7Od8_-45joFUX7r4GPbDg$H_b%RZ(*a(+TQx8ri*iLY;7e)haYpuKIk}lFf)Yv
zD)Oq+?0|Vzc4sYz*n15(2BTU5`oWrUrbWxU8C!;8QAE^I`S+5(c}>I-;hOc}Ni#`>
z$n(mVOtk5jnKXJ^Bt)^}z2nW~nDMG$)it5v^M9;jW^^ZxBMVRqDjFax3ijSen7DS%
z6vTT+phUFjKUzX*SnOz6q=5=Zw%Q^?*u|jG`;D*YMFn+3xUde-77{Ebb%k@7A6s9}
zmXt@&z332dtFF@E(RW$Hd%qrNeVSCbar4!H+d%I!-KxKT7ZtN!_Lrxa`pbJc5jYKI
z(t@c}YI%lN>d*@!mTJx}D|m_h-l~glpRxc!*)jRp5NKk5@W5x8v%LtH^Pe5g4kx)0
zTTZ9n6<+=RwtE(6m2Z^Qr^`B2s9LiQ-aGH`*Xq0StFP`f2*bG<uP3!{b;Nyh6p-vd
zvI+&(HL-7ST0YyDwf+3{0XbWiUD;GX6Z^Lvuk$bXb~iI*lj0(!*+RWU<KG*biooda
ze!Dh00imOFu-tU(5Wq3Cfm}1xzRUGmY|OF03BrG{)e-b2&cV!7>lK|t{^ov*V0(oD
z*q5w6TLe)m^WG^A@>9oH1^!%EUDv~H+)I)$>@n;W!XI}&^r)Y`kIStcmIDy5V5MR8
z@G!9sHm><}L1C5BCQEH@OUUE4i-(-GYw#}J`Q@x_j3|e+5avx&>YC_S)Q7mv0@;F_
zPU^Mw3n0#K@6MG}S+U)=bu8SB6o+%Alc{ycXY=7gujpPbdu(;NtrI6oRLTYiy{|`M
zhj4ou_7eKs54G32j%Ko5CLL$dRF;&hSn7p#h)n<Cl*<b9LeyGRC$UD>w^N?3q7BmD
zH52Ykwcm($O?T_UPBd<VD)T~H2g5i~G7;J$e@Zu`4eTlD6l5}ulJ0vtG#xq(hO%uA
z7$1DSwR`oUN`qwGpwKjr=djw3bkF-<+a7D7*DMMfna@k7WiI)iFx#rS2d^IJOtU7O
z;JRRV55%`0{u1BC{*0S29Md#X{;nGv6eTOJbL)>a<;FX-`K{XC9btz10B-AK_R)_{
z_eVNce#xYq$P7KZG7;8caKDBUFR(8TVszWrTV8ia?>q50wb{Ni<jfNrX?*D@;j34g
z;&(MK3MXMP8_AiM@E)O{Z<k&C*^Bb|;q`Cy`34y6=KEWEY#Zl&3*S9L8up4<((gs=
z#N~&5iWbea&MIQ*tjzed-WQ>nP?q8nw;lZK*KYE<XFxNlZj=E<oI&}CE<$%#17$~{
zZ%rrRw8pQN<QRSF0_D9^0Ky~)&p`RXP7Yg#Z+i3Ivkoqgj=$;=ci~$(wFU!YFIG{;
z`n+n?6k(~jCTzm;6(>;rJT#?!*0=9p&!3`)k<Se@5f?SqzbB8IZt+5{1|_k8CaioK
zNZ*2rqFa#pCjx7Nz*E)bbKRMSm9<^Y<^;m=`_GADt{%n@zN)ig@yCiE`d?*Q3M1Q;
z9J=>Gu#hRCRe@SMdb<~y5@^O!yB3%VO_SI&OF$;*7MB($PSr#+s6h4K3~F+c?Kkpx
z*~Y8~2uB|-HPO^xx7oHeja+!s)?D@+*$y&U<ly?fR?7sNZ;q5y%8aMJE4nqBlcVKC
zi^P%=*8nc;8dwgsH0fxRS!hoc5HmR5NICX=nBueBZbY~6?A`|fSjrl!S?wK_DG?Zz
zJ!R(2siavL8An27qG6Pf0W>wzDCAn)5&AIp;S)ZeaN9P!b{3xZD&e+emj*f<AUW8f
zynUv%Q+ca-9}SJcci*zSkb|#?kNyM-gLaW8s&FwWRoIcK9G9NIrmP!Z6~|P+<!7E2
zoq|VL%^um`*_6hYO|y=@GlM?WI=-ukU3<2TCG^ZycbAcM=gW;Q931x}%d^{F+ni7&
zrM&Kzl;iS{#c4I?uJe1!R|D;fj+6Jfw*0CzjuxYnRGwF%4ymgxBX)N?G|qN;2!K;2
z6TU0~Azkc&N6r?a{5CnjS`3FMl>ho$1x1DyAq(xX!^h*&S50jdyu8+<QaU8n_7&xR
z-2!7dGD>lbJc)CO)Q!exmnomPjM$WtJO+x7QWhFhH4kuT4(*a)RKgeeoa?YMu&HK0
zRsF0%$&3C{tHDJtUNmZKn6|TsD(x97Giu4*Xeh%__UuTQNG`%|a=xXpf<hbs<`ST~
zdq6t|J+N%2*n6xvf1-p9Qy&*lq6D78c!!SH9@6wK^k#WSN%72bzl>iEkrKTISGdgH
zW41Vny;OD4p|q)*(UI6^HWhG#Lm>w4hln{Atmd8vWX^l9KSflHagxz8?wi(l`tx&3
zb6;rcE!NRA{pw-{_6wOA+>CB_T&Gy08N4DYsDg*;kedzhmT&mtg)MIH4EYEryq@#T
z59U$zTz+cw)KpcmG4+d$f&b!+I%lf{6`hSl?reUN6C^EEPo)NU5GqVMRP-8lJsScR
z1HV}9g7h{Mu!vM==R{|A!E1BjoLxqxBV!lX%Z!o;DOK3%Hq$~CE|@2Ia1(+&gfTmG
z2&rQR?bTZWVUWn$&inhWKO>_QJgF#jgQW_rD)i8c=sP!0BD2>U&JFt-eapgP2k}Y6
zJS|Ju<^HblbotoqotObWkU6zTyDQAaK#}Mqv79_1T#FB>(l7*vB(Smybw*h2q;jod
zPvjs<?;gI|9+Go6<!c*~#$bi8Xf70#K?N3X=((X7h;cy^ohoOEVu8`A4?zmfK6!h+
zJ2IQq(}D*=w=P!ay!sW7z%!EgvVq%^2B;!gF$tE2IbbPEvFChF=a<_^CCNWF5R7G|
zb#Hik)wA%Ck?4z$0aJ{LCy~NrRzOQ-TS3)zVTda%uX3d#9b0Dk4H@zH+LYlGJx?pl
z0(Xlu71oM`roXcx+!@9i;+LUBn?!&zur34@;ogicmbb*ZXo;)+Y*Ct4a%v{3=3`Fp
zyWFF=hunR3bCOt1$Q0rvhXjM_Lz>9Qnm0$AnHkCJk(1xCLo3C&tNYPTW<i4hs7+{l
zarbc?tiC!^1z4UjAk4*EhIwF4i^leIy<+W+#5PTna$Dwy+TBcIWP@dlRFL5t>te2|
zf`=7`hfjHYL#_f-hpog4sQjr)*HeIm-L+9jRiNDbf_x;A#8uhIj3T#o394_qO9NY>
z*W~8cRnI9X{&N^k7fRq%C6l~R-m88r*S=8piWg#X%D%rmgP9BY{TT<<t+M5VZgGE&
zg}(I$`zuf9KHzDamNlOg&di}?$^9A}dhb5w4oFLGws`S-Ge6VPm--@sA@%jWW7!xB
z$A5Pgyn${PG~d9ep%cHL;XVV?&5+5C!A`<uHkk4n33~h}fW8;_o~T%+*MQY)lJGpY
z;K|fEe4I$S9{u&}od;aM0~sUbNrgTnO`|tNOOP`6j$<Mqwvw>`#b11XicX8X`Q}Xv
zW>Vxr$m?thJPfdDjUW2DnO#N;te%yENmKh1v~%mZMWDdVm<<dAy*FNcV5+77%Q_Ik
zapHw1tr8hLaP@U(|8oifMq8}x3~cb3f)OED84nV|#3-mQU)3HT;$$u`dy%2I^fH7K
z(dYH83y68n?xjR5EQGW%GA}HID~yLkPXHO?16WhHTs1gAp5~K47Crj98dW1!EuRop
zK}nmkKRWwnC0fk;B@e@$acBuyiXz`5_xTUiQG8T>Ut~p$^^W|4EZCe({U;M`7uZKb
zx+_9yqT#?iX>-|S>1^wv?hSu(PDFG3D>6#tLh-?J^fxEjeYewi+vLcyjxk$&2wSQx
z_wvHE)vS#gSbl)Zb<fnD@Vm)aR;>pyzvHhQaRAGS%}Ed$RD^+)mN58H?V((3iV`&q
zI&^JTTlMN!8g*%Dz9KE<fN)K|D;!_nnSH!$gx<CNEKPX%_ex}7B~-~yElaNkGQ;rN
zl1W`2D$ccE1=l6GKHhb{w6eYjfcVj#8nY<=DEBFtZYkl_vu5|dt3~Y>=XbWTtC8TY
zZ56VFZJf-A7+Af9VEXrO`_9io{_V^9b5uzFa;LsqJi8OYtp28)X&~S5KKKswj+0=-
zR2d?3(*Y0zPueJYscEgr#gfDj(mRuW{Tq$SOqhBg_GS|^%KlXHun<^b(tRlS^W_wh
zb(2q$Mn+7-7;nzqB_yQ)F<;OITAPbN>UD?N?D+KeiouXbiBtD=jP9Z5?&$bHgpJY;
z&|3MtU+<5nF2eEnKAN?!dETE*JnMG2ogwD8u%OYXoo7TP03d*#^0uO*WoxAMLK9hi
z`PdgF&u)oOnn$>11|v1yZ{nd_4YQyPx;P<VLMh%Hd(E<H8-O6!n7&C|^^yGGG}i5{
z_rjwRo_6B)Yoonp8ZEIXXQvB@uWl3h)~>F$yP@wtIdnU8iD%63T8h^E5h%eW=0JSW
zQrRtFStqZUhmE;n!^+()Hvg(U+RDQ!IFz=xtgZ!HSfojquVk%TB;p>7DBL&9P_tO{
zGh3Y03j`fd(y*&w7^hp3_5hzGwTs&!7!ms<V!wAe=2kTVlA}+E<*NXZ&=4K5E($0l
zJtVubV6MZ>*O>PD@Uf@Q(l@reDZu@9hf?>J{0=UZjJ|A(upOXm=kk9?MU=G~>^BPJ
z>py%N*Kp|xwojhvfk+o5G4*B*K3=!{jEL&7MC{B<T5F_YP&f66e@j=oFv^brQLJo%
z&E_@J6h=EYrY69Wh@}ns6+TS8+>Ml<07w;iv=hYtq}J1o*y=RgLz5ThwBIWjcxf?#
zpAV5AXV_=NBDY^^8_Zm-@OC~nzQEqnPFR`_MUk7uF7dSMuSpSR`Jd6lfwvY+cx#P6
z-x|=Mw_UwLdrtCJQfzHfU&RW%R^x>S&!crz4Ayv~@4H(sOFQF>px8HFDn+QNK#2X#
zexXPD|El75r8OREq*3?6RpYnm|B5GOJOW%?pZ{04Laa)S9oQ6LV-g)Xg$;?0iCl;v
z)MUsSeH-fgZb`y1pQurv{NT*OmD5g7u+{gJl|6BDD<ykMhCFzu1)<@2<CC8xaN-f`
zdw{mTIK1r(&f*OTcTj%q<vc+74F&_M>S*--t+|o>1WTQ><W5QWp+)a&H9TF1Lq|KI
zB3(iM!K)&B2Vjp}E8}LlkPqjBAw`77%kQgPaSnEpU@!Jq2IDCGRX?PFjO?>g0E|QG
zxxk@5I??c|ZC=UxeJIwrcPmcI5x`yyM^kk-Inn?6rfEP(+3E|M^gf)LKz8_S55c8#
zD)v%^l>;mGuy@*(+U(d{&wZ$9(|0^vI;s~{%s*HKevG&dpeiWuzTqmPi@w0+{dcNN
z0IBkO5fumV0AJPwkb<l^%!3MPgzwX$_x9pF@7Dp^2;*8t_65tnTn!==NQ9MD3t8Es
zRoN{2gBZ^*4l9A*y@3?PZYN0(4<rHVEhrg6dHCx`0J$$fp)QA|cg02A!VlF}-6x4y
z3xaA+r$5LHrDRDldZ2vyRjWj=i%yIn*0Xf@Qr5bJ#btdKh(2kzg!y(VTw%RJ)uPJv
z9Qi`-lLZF1B>OLaUn>B5Yo?G7J3a{Ki&}UIYss<(Q%=@O2^Xvj8PVbe8=d*YN=k7u
zR|+7p76{Jj6&0-7v=XGL-sq2WMdDW%;SBKv_2w&pc2=6mK|+T->4bO^+9aq5nv(JX
zX|Ow{=(XBQ2&oa+nL7>8rj}|dqIK)bSlCXW>;W6zos1-jQ=H2AhAQSIq1o!;>dC}T
zo?>_uG1A?W7eHq1w54PbO69bLc}G&I<vnnJJR;CJ#m^>))u$V5R1pTdR?IuE4_GsK
zAr?QCYJ~I$(>{-?$PC%EEYS+2eZE-0iInoD9C0)IymK_PtkFK9VP$|XGRkO>*5B3u
zmE%9z3IVcJ#~q9>gOaPObG#gwY62W3_Jv@^4wpz5%xrH2;?NPfY7GE~MK;O3vwxyM
z+{6f0Q?;^sl8df^7tu*>J}d{5r2T6qj0)?mSrVVLnV4i`YI8#G1rnO-(2cPt9v|Bl
zqcg3&5Z266ixERyd3P`3xRfA{w@Q2W2qkQz-e_{S_-rCh8g{j00Rfqo23r9lC^Bum
z9!O&3S6Vses_;bIb2ovXyp&j_+BtIoguY{+!$qSrPRfQnn#mBB)SvYvC@Be5CG@1i
zFp@@r37rpS>y^l^M}|*#vGXm$THc1m$4<FVIgFMe)ngj_sHgC&t#yeWATX$wix>n=
zN2yytNRQXTebPi!8mo6k0RUa}`suC#%J1shoXXFyElMom1d#$tfj4{C_LwYfWB|UM
ztol3T_1H#FIK_xfzZ;tc;ZFCJpG6L@$Ou7RAnbjxV-ALHcQN~bcrBIr{S!S0F(r6&
zaR$uFWVt%`YI$S+40!O*8fDYW00Z5Ny&NDYOl#kyQ?njZe2CVI;9cU^yasL$7wBL$
zO2(Y^Wj&7UeD?JXKR-_SodkmFXt-Cw!ssDEg$`K6eXC?JjIg10xnx9uSq!~51CAWj
z>iPqIjz$bwd)rMVVLhI`|2Rn*Zd%_o)M=7W?vGf9pbH}zN9K(iz{=w!%IGR@Anp(=
zUQ~QNGrp|NjPiS@ITdve)E=Q&R(vx#GxN>Sycaf0pBW4sX#>xt`_8Fg6KeMsven(Z
z0D}8n;U7-qCs;?MY}^8)BG~N9?*!%7uLqm_#V8Xm|4IxhU=J}@Rucs)+3nQ_q`@$N
zqChUml`^w_Byh+U$fEXVl$1st`_&V2@1j8kU+_hd7rxV^MMvag2nI%!Rz*UOz&mtH
zE7ts3d9e5bG?7|zDL-8%peCS(w3#|H_h+AhI)Z(%7Hh**v|U{9S#f}#p;e4A=gV#v
zsO0bDR!#aYw}WLlhA~}#a<6ei2(UEV1a*Y0Xg65_`ohz+fmU0`y9{gB&+8id22`yk
zu2*4#D=5r9>^L;|lh2zdZAk)jn)8d&5D1v<yq^|5w-+CHzn*#JG72n1Df)t6dcio#
zD3{w6dAaVh&z(g0?q*b%397Dm?vG4poHukeVPSIwWQ95yOMWm7hAydkgB!rZPv2A-
zwgF@ve2;Ai=*_Pmlzb-@Da@q~X9RtQG%f97cEL2N>n^F1^dyn{FSYfUWD@_3r3%R-
zUQ{6a+5gIZh5EfkY;M&qD8EDYybZ`Vz9K?#5)V(d>M80;#>RaUxbFlx*zNXNd4Hzw
zm|zN=UtahEf7c%5<i6(IbIab$x*#Lj{fX@{T0CX%ONQsQrfX60VT1#qPuS~xX6E7J
zcQ3TZJUkPB;sbvt0Ap6N&1&pqq{^jvQcZr+#McN@-KU>EUm(%#NUyI;dJXl)5Vhw^
zH&TjRWA+BSbVbhyW4U%4$U)9{(tka#{KuPB2VmchOP`=kEDDf}m8aSbmW}75&nPg%
zh%yIG9RWYrOX`gg*cp2%S`}}6<-~_tLC_Yf4-5i2t8d~n+uszO6d43qlhl``y(mv(
zty7I20-)9*97D^c;k2pbrX1)rU?5f#Hj$ex`@>YJ4M(G1F#qFtL`%gjBsXb{5T~cV
zo!COdxCOs{P^$)={$X`|eAsgt4SpMUKoW6=#y7xTECUu{;wWE=qX~37v4-D^`!!Tu
zQ4KqUZj`~??M+1Z>TVLwljmSbw&-O)0`js6Mo<r`4}2f&6dmahbfIHu7ezfxRNxGc
z0A*})-<u!=GHYWfZ;>f@V_<n3bYqe27k@0r_IxS=JePL5hRK6juVCA*eF;!Xx}~S=
zH1WIw^32;qqsxt%8~>7m6^+3hn%k_sCz&+CM(k!A)n_%XfEDdvj(RWR4LBtGJgo--
z)I2Wi9QR`gg%ex%vP@uBl}5QlX1IjwH?GQChCaAVetJEe*v0Q{zh(hczpor2<m1u5
z%UtHMQO}Lefu3A?#}5V)AckJ-B6lwWL!^twGRZa42bmki0d9WRT_Xt;Ra7H_RjW7L
zHARn(@U<Syw_5HV1dGgo!Zf*u)Usc0>;0Nnkrq3l<i^N8)&E^gkVMN(NK(;B`_(h6
zPM`arDp-AuInjPeVlgM`LP6226lht_U;=zO{LqBe)l&$Pco@{*Y^RA4(-9NFXioPD
zDq{)eM&R4WUzn``>eJtX{kz#Xz6QU0lXZ%bxKIi%(^dZtssS8(u33~X^MZ{L9YUC$
zAa?(++z)%iYSzxMtRVn-ph}RG`Y=&$^Iv65X%}smx$-w41fU>bdG)&Mp_!1vK#t<I
zF1V{Z$1|rhuk-!hXTk3ejrRIpzgaH<*$Gp+HUcVwN1<eo-aT65IYeaNQ*qNbkUhEz
z!!_?xj1XYZ?+X3PU@cI2<|BTJXs086s}C`6P$5az8>`{KW9F~dIMn2SCRN7&0O|)z
zgIZG5uC?xv!WGa@Iub2HH&-fgdv_~7g&<^8#V~l&i!ZpkFVEBf?=MBniz)`B=}?sQ
zHKkmYvqUJa%hxm#_@fl3MPvOfc6CzAUl1TNU6U8-2iT8{84@Xn70Hn>pE29xCAjR<
z%N`%wt@Wd-Rem95ZVOd{xCaogB$wCZI*O2K<{D5RzA`%a&LFeE8Wl-kb@{b_l77Ex
z4#w}qacU^rgxVZ>C9U3dh&^MJNQo4PAr6{Xt~3TlNk&FywT<4NX^)t7PLa2yT30<7
z>Ry+aSxA#vnACf{#O-!iol%&iF<Z@1IYBssbf6{-Cw61jHy-8w{F3!zas=oNE2k0>
zGFM{HaXI?UK^_3a@7n6=Z4GYGfO`N$`6$hv2Xlba@0*TU@Y1J%^zC2|*4$WXuF30>
zq_TwBNrB`}vR&nI`#1Fj)g|8h1`rg2#d-1dsHt2a?UFye#!a$`#Oq*6Y1@@nAj5e>
zRR~;wUz+IQ<zTo3tJt0qBWwi>6PLN?w8ZEN>YWd6ldx2#BfvMQ=cSzS>Rk68BJ&0I
zWar!ng=7LmvTv~kt)jmUo#r|sj5VrOFotew9o^Ls=Nc#{nyv2!3eQ&H?VMAv6QQvj
z5^j@EwFX`O2O~6p4e7`zDOcFv?QC5SPo4c)nDvxYtPO5g1_N>6RD5Q(%_q_tyebJq
zbLnP!KZ1B`hH6T08xb3@A9R8wq>fK;`leUyV6^S+w26Uv(?$w}lPc4rFDYd{0g><`
z4laBWY|Qx5(ay1d=S_nEYyx={&3D(6C0RO9h@$Ot-7DlB=R%^{$sPuF0kkp=`dX;l
z4Mq8N4C58GfXO`oCK^5@4pkkJT21;O7bF!#KYm7RbTT|on_(&dNQSK|Sd<0Oa*D=g
z)=HN))~|u0j2no(5bHe()VuBVbD-5~Vb+htS;Xq0HM!r}GQyDt$?>{;!?$*kPny=C
z38mpa2iVp|YFWt3IoU<m?L|-J3&I0a49RT6iAzn1agAcNyXvx+s3ZA`N<U^3VrYrR
zVvSug+Y5Ne_5n~gq`9Qe7%0!zf(Xx}s5b~EIYfc^>G0l#%HMeyRAC&?Hqc->-WOK<
z#$n3w0lfXy<fAeW!>_e+m^|ug+MPJ~A(qMC+AI+>C7wagl=-s5Akg!cxBiEB0(o~!
z3b(Wv{)T^#YcD7=C23^C%f{>uIrE|h2vih%9>=UEh?Q0&HDk^r=N?E(^Z-4IjGlR<
zQXf`<rH;p0X_yj{TA*53cmgtSrhAnrNx{0?T-Y`zdm0T0t2vHf(Kv=>(dn44<1p?y
zfpS679-uh+pb<7TxO&UqkXyH}gwkC0_W1Xv(+D<AHGlMw_MaV5h@25X&@WwV<fq;F
zUeD7-t=PrUM)6!N5e55BCfj2AHl#=~k@1UdemzMmL(S?7N2WX|Yly5+y2w)sGY8#Q
zz9(h@0X6qJ%M4A9aiE4!$9-T(?T^s3?49zge<qSZs0YnU$DE%c2C?pnNBcd2GSgwJ
ziU%PZ^d-nQ=G#<6lM!S|UO@T7&F=<5weL6eg!RioYmz}pa%QR#>yQp~Y34EaAKW5~
z{KWZpdUqtBe58?P2r0QSajq0^axqXSmK#mG(8zLx7*6TIS%V?}<B^wFJPH*nmP5RE
z8d>(d!-hVx34+Kg>$L_*^O*6%Cq5lbegw5gy$OGf&S1LW(RYU0ID+zMUb^VwSge>w
zQWZ{-gu|Wa&>7J|;T9sxlnv!Je&k0#4i@k%@7wOJ?v&-9X$LlIR<=}j?Axl@6Cc<I
z2v*`(gUARoEn?vn(~hF4fQ#;=a=D!%4U}VGqmcO3uDhDj_hIO_YM@^$WU2kC#V{JZ
z$NJoa_vt}vnRR-+<?Zb;mkI7!`8K7ac$-Tqlxh3`Z`62*-f>?ORc*Lr-@>LInbjVf
zf~U_wS5=Uq!c7`TXL9zI1`b0bGHmH)&?XV0ffd2==n(?q#S)G!P84JMh;qkhy%DlA
zz^M;OEGro-fi)g4_|0<HM7w3%17?)YEvMT$q5eGelQgUn?o#o|LM#(ZE+RiVxfAJv
z?7L1AD-{Y2$UQLO!xHp?MyVH>lIcPB!Im=CGV0X&nvL2mrBU86<B`Pz(M9F^W4DZT
z1DDR5|1FU5^~y*A=l<?25XcmpSvW$S#_10)SiPNbO7zCg<RDLW4{Nm!fYDrY%Xw;0
z{=UvWhPPs9!#3ik`#J_+9n!*E`}I>ZI?^xi@yE(53;M61Zv80G;25XclSVZUUE>^P
z*>1+5N7VAczzL$m%N#_#;Q^j6G_70IJgMAFg^T_G{wTS9t;)*JXzX1+oaT{5ibDB2
zqRv>3g?n#uz|l3Ej1E+e=}idBP)E@y?y>FwKBYt^*Q7J(3)>7{LS{vvcflr}d3MOg
zS{Q%5%W@Kya39MrRZ~PJ(q-8L3xQ#+Bj&AFmpU?g<L|!19#m}ty_JMWi=+sNq`vUb
z(2@q+&knBa2JbrE6}sPL*$oS!`pz#LUKB6b#ZD#>a(ZUNl*kQ^UfAsWF*;{kK64%(
z$>eo($!(ez4<JEtkunK~LdPH1zJGi8^eQ|+9<FiXNGf4|wnGS7tvnte4&8HD_x?vC
z<a@+(C;d~ns(kHjBQve6gB0=JU741|{v;*+5Ney@6`(=DxWKh26LQbt-U?nn^>XA%
z_NTtF`+x+#H%TQi_cqq-ga``ekm^`ctm59iq%1bmX~==A6yKH9*N!G<c9Rub=EytG
z)-cqe3s=><c6bQ~{qXL+29wd}+W>hi7kQD&qB5k-<t=5>J9QA(Y{|Zlq`*?3ucmuC
zq@UgFRxjgrgFiBDkJ;I?$T2@H1%G)P9(5&ATzL<>m_@PjRAdDRo1H;?j514X*6n7$
zy6^F(pd(%(jpe<Nj2WbPu#M8Z%Thy9KXxSEzNd)XBEJRF6?+XSx-lq}EBrpVDE{k_
zbbjk4@5I)fmvH)Y<YoOMVI16<AE<3g$)TTRI6_@2Z~Y*iTg+T4m%@gIbbcIiy#3X^
z^zn@M!KmZZV-+4;7}MnKu)K1+B-|Az`NQ=UcAVpz2ys_<By;rdsv98Umrm%nrXwI_
zU$j-Px0bX>5{A}+VHDagKDP#SaW_<67WQOf66#U|(wZayCWpIvz*-igGeSGskWO<N
zWmK72OXu4GKQOZB7yxMGq>a%pT$1nG<z>HKe&@#|unz8Hg85>*b=2K4Fgb6T|Aw)`
zT7iRUMvh3l*}D<Yg}f0M`Xoc)At?H=OE9X3t+#e)L3DnKz!4XbH_YRgNGfhP@+s{_
z1L4h5b+6elv5iVfp!)vn+Jg@Rt?=-Emp}L;-jiPQFrX3>bZ>OJLdNLLL>b`!j`2M+
zikdMK>+hG+DHH1Qc?_+`zW7&AFNBR~R_chc*whi-`}~BJ<m_qNHaEg-B9>6o#L^*Q
znc7im%$X3fY5q46$5=7Sk@|T-!T$5kBhvU$2P!UdpA!DUZhE(=H`7-&MElW09q(Or
z`U$&pmT@DSe$Ib4#gybHEr$2XST{3Z*#3;Qz`LHK7&PXGjmt1Y%=~OQwHK?Gwy0A-
zOCa*b_QN5Z+JzPge~a+>;;NyIjT#Y|w=6i$sij6;;u*{ACS<74FyCz^A1I0d3C!`(
ztg7A7;q76R)ZH|dwKd6UW^7DFUmuG-7^F19h_t{0pF#y1;|0}rJ{AeI{Dr-YcH<o>
zla1mp<8hl#20VTdi_RLqkPQ38kdWacnp|4+gD(^oDno5WzWr=b5<AJS+oEeb_Vuov
zU;HsNn<$YI=}M=1$uHGw+PBK_m8&xo_3VesfQqK4%(Flbv~W5NoI)1?y0U3V&Fy&f
zN5PZN)Epm`-0hr#yy<_H)c`4*yL+-lK2&H!J6S5Ms=+_IkbH<dBjM*U<4Z`<Fkh*d
zX}31qk$D6}F@3o*?)614mz0ySibq^|7{&1s<qQ$XwA1Z-867C^sM+-UH%LXm-*j{-
zc^)!i7A_63nFp_b#X9C%9s>oXq0gZcTuZ8P;Gs-!<2ynA+h44BO76GSD~@aQsmQF;
z=Jiv}-Z_FNyK#Hk>?y)qgh2E~i_IphwJ5XrQN1DWb&ImZI{b+jRde^eM)_3>6z2;5
zl6(EqeJLc>ozr<Vg$}zo3mP{KTuvq6t;@iimFwq1=F%9K?x<ntvw~<O2v7m`v%>F|
z@%Mdv>v7VLD^J8^4G_WG*(E|qF#ok+CBKnsBkE<PU|3g9NFDGJj=9uqFyo3{0U}%h
zCP~&Hp4%JXzc~HzWtdNhQ<ZR}P5g;+6aM&ZHyO<g0zl#Tc39-~*m7|oQ+9Z-27RA2
z=TDeK2YU3*13(ov7OwGQ|0!I4eS7~j8F#3C>thFki(MMXdCI^ZsG+>uC4dC20GTy!
z<}_q0`jL*6k}=S0+XD9lK9WXH{3L0^(rNckT?2N%`;?w5C+#6i=kyG8{_V%$t~d$g
z(Jlm_pwHs9?#sEfD2*p>VzKf1i2759h0`g!9q>^Dm$8}GAMM;;dH)F#|Bb|gpLx~a
zvP?TYxL7v*N_7+rEuEP;0>ni-0m(qkB*c)!{?*Qw0+fL$GC%%O!jzZFPbI@`ouI=a
zN}wH6|NTg($$&!XuTbZZUjWxXNS9nK9U$GrMlrHS$UqTRD<DI0R)HY>2}0w9=x)K<
zgThKM<#3FkPCi2#a<wE*5;$H)-#pD7%4@z`^ZQNxq97EFZ_>?yF3}<cAj?Y{lFQfd
z1Q2x$RQI+(tIhmMEzmyw7&K(%XG90<fElq3K&rO3pFV%H2kb89j=-yRfnVnl+5w&#
z5qJONOO9W^_4hl33X`cN!g`rKSL<!T05zE>b#uM%$4*l4Po6mecA#g6EAN2jr122Q
zx;Ll-_;srKZeJ-EGDDZ$f1uZ_4$>J1QS@!TfBiN822pRmHJaW6)E$Pv$gTujv_$|^
zu_ys3YDprV+n(|zz$7ccg=G)e){1@bZ7j>I`cjjH{;u6N&w(Smecy>!<o6-@`>F4s
z>OfAO=O72;yAEaS>nH@PCJFZ%4-~N@oiJ?ZiXY3t+XS~C@yr_jA^e3w`cE=HRGG0A
zjZOg6B&ISLMfXSa`Ij>?7bI8Px~RyK2zvi90E3Nai8~^kPJ%AC!~?*$XD#w2ZX3w|
zItp)9h^JwL9QhSFzaROZ08i`-e9N5#I3E@CT#<g#?u}JCvR_lSI>97>>42wwa`NWK
zzgvKRbCaM@V$M8jO8A)4um6Q`Lp=x=BRQ$-KRnp~^7Q}LU-+`A$f5ica@(R!e&@yi
z>)-wlKbYia5YkF`hRDAS9{+My|MOoMx6nplCnK_jEp-3*zW(!n^RNH3f`=7TzcMth
z_CNd%812E!f=y?sG5skd{%?E;zANP5-g`Jp{1grR2e<WKKkvW)3jq*tnyJb!-wpU5
zeu)2HI0kY?Fe+fs|5gpCld;z~-~{IIfAFJ@h7$_hiUhZvt(?O+1IiB&NY0?nI(T;e
zk=UP=!GGrs)pJ8Z6;=tR5KulA-D^M+>}B=@S^NLXFNzHnAs7KH<%)X2lM+U%slGCo
zqsUG8Q%?8a_!#Pp>EkTlSV6k{h~mLxB3LPMJpQL&>Lxe17m<(lFaL@3{6BvqbwM29
z-_z>MT+aWWy~zLl|Nr;%|KIcX|JVB`XFxpeg*U(hKU$^ukGS`_stXeTwif$06-<u<
zp;!_BKIPyC5XCxafMD5BT<cd8DJQs<=8=FY@EHTuJkHR2*P_-gm@5{cZ}a*7UZv>o
zlxh5t76xX2jDg}CTXYCy5TnZg=JV3?2QGnT10XB48mE$4l@kLrcG^T>BN`y*OWYcX
zbFVnpnt`XyU=+-R4p8U~Fkk{&(^D?M%v26Q%04hbUZ3F1bS4zG37|9`0|duN$vC2m
z(5Z|8v~u}qyR??DJEP0UEVulJJ&+sM>H}TpDB&AeD$A5R)SLvJmji%-6D|U`kG|K|
z*L~T>ws-$U@MQF@ae{aPo_Ms|d%Nh%UvZy4eTaS7_6dk_hsMrK<W;TT#)EAg*On-Z
zEEAE+r8BYm?h5w@b?`hI&SK-PX6{3R-fJPYKnto~3m3C3n<BvzxP`HygL5$!U|PEi
zqY5dzO<^ndPZyg$fZR{p;lW7PN?@KDAvhoBg@FGGtW>J8j;8h-2cS#RQSO<ar4-mp
zcmOO6j`ZAW2tZtDxUT`7*TmH1*$m+89IOVw$$ajM^AB5WihGskU2KBnWa;B1oU&Mq
za)8f;D7h0jyPL+480Hngl_OIQm*AiaGm@Ef<8@F#r^*6F`6!?}r@>UP?})+#&{el3
zN|>YEz#!T&Fr6O9lS!cNoojcR1xHU`$$^@$%W4a|gD%k{Ba6@=Q0*Fn{Jo*!4Mv^>
z><`o)&D4P3t;BiZJYK+#z;l-Yk6hn?VE0C%L)ZNA9FAZE$0-I$y`VRdO$Yz;mrp<v
zVGBO_6$}^!e2(1nJADcwTzWfBa<fN)`OxnVDD^+g;FPCMdM(0&v<R<!Ba-*8V1)~j
zOHi6fXoWHCNels3{DQ=>9Pb2Taqn3UKg!O4<6SEWYMX}=^`9%8<0rhgvW?stPiRD>
z0ViuV(5xcMGfesQftbK^GYo9B_@*+#>I15uZ4Ac3gj6ahU*0?hbVS2mz-iD`ZJQ+7
z%A(=RlZd!6WdSf#K_imXpr(02pWTpISuh^H6m+u^ko*6@ckU|BZmIz}iJwc>9=`T?
zS>|AhT5@>Kx@q8Q$zTKFB5i6g@aL5RqO$@Jb01uo`eT&dDxXwzWnX7RnGz)-XxQGY
zgLATi`B=N#X-6VK@nHoI)Rr*F=O78PJTSdyT*c*z(>jefS%wgV1gW-<7OR;e$QR1N
z6koTXX>W6LdKGzYEjIeqF4oEppcOTYm9KeACA(b^UJf*{RWvOG`KR-klTIlh7$hK5
z!O#}Y%Uvq5sd;A4`zS>^?zo24@A=*g*~}xX>UeENdWEIJ4oAkHUw6__S;bcNdk}}a
z3x|Q;f6+%g;p{0z`Q_CfRe)2!4{6tm)LnrCACf&#pL{;n;0%5cz(vJgVGxw6_~k>0
zKt+oIXw#28WApb$XJuy3Jn&YbK6>$?EGSvtp9jK#t5_6Jxt3uKwuEPo0^RBY6gIh3
zk0)@NR*Fq!{ES(8RN*)Mu$XWc0#+(U4A0ZU;{k1fk-TCHoVn$%(UU4^r$SAYP`-*U
zo0tmc@S_w@Gk!)5ISer(jrBJF@>PvRSbu(~X9@KlI7B#Rp`f#>*9L)Uv_sO>#Zw`P
z^`xKNGa+)UQlb&*l~2IcAX5mhm7pRKAdGhopA&^JlPWM-bq8JJ1-jTYADMYG#Q7$-
zt%Qk#^hPadL+!ZMJ8Sr=L_2VjS4wLSKN&5=)lg>C!(z&nJrCq;Wsi2ShL?USX++0e
zvA$KGSFXzuyVgVw`2gan0tenQk)zI-@_+HcXVK2-137g8l=Hq;gIU^l&%_ZDVOsIE
zyD59oF5h6=gJy?3T4dk-BNtVT_Fje8$#W{ja-rSwd9|ow<thNrUoFB#^^&o}gW16E
zqr{g+QrVsQ(Hs+~5qLjiKq}UAw)b)(sJVVAf1$)df2lU<$J%q=uyM{bO^ulHd=a$|
z0G_WKHRi^*KUIdFJ*g<n-IfR;9T_MCKZ*3wR&yYLU1#^E2MCXq`alL+SwR~W1tKrn
zkzH4ZfhjF2Zl{^}3TW4~Pf^y{ARnULc6nrXgGa!OUeZZa#f$$U+}|TKt8|nT0HncZ
z$R4(;*t$<gsvOYS@xR!5NA1<(^+O~DNia57_}g#W*CSWqvitGjR4u|!K>woYw=o9f
z6Z2=R_}N;kgLdvs!*&2!i<$BS@a%>gDGkcv791k0`{pUMA8?XWT|kUk3%O*yuDt|<
zIW61^T%#!g+lkjK8r}xb&wxz0t}*bp&DvLhtVboEQ;5pHxS!uill?Ld0Ol4G_{rj$
z$>$X34ViiHfDmdNLo}XQ0d^L9LZ@R_id1|AesS<BawrPOH}ocCUmRYv&i6GE7Wc%&
zAHPx+xb>a_o2a2UR@GE(HVaK%@5_jod&-BU*OZ!iuiG2$IYnXAQmJ-zKM3LkG)J(#
zg4c%bttzCJt!lB|iGUM_XhjvZr+qKp1u*&Uet%$5hh?H|_0|KHOx`7ZV36=KpI%M!
zK+`Vk+Pe>1Wo&K9@9G^zNjwBLl0*Mfx;~RcEMyt`nrL=@ebtAKy-H)3vva;*a-BSZ
zf%!ir$TL<rSaySq1fsc=81&3(eXM(HZIgo$SCd&5l9AN&AeE%J7|ulfjMTfHQ>%Ne
zn%;dXi39=2VFALQC%C13dOt#eYHgzEIIgE>QAi1bTt#2B+r*J$rP~^+!J^l^GOk!%
z_Jq#?IJJ{(m$&?=W#&?@ELZ^&bUpH8xeV&Ta%urU8=uOUUXN5ZY0?edU(cnn@L&Zd
zI$jl`II7wvT2eQ(Uc-g-RpWAJ7|qnhlA(KWkLJ&A0(uW$yF|5$fz1es6OW0&EVio6
z-B|<9+9Acw!k!Ygi>p+go9|h+=$!}_%93uyZTBSaJa!^vUL4<++@?PJ1f7YrhE#U3
zUa{5C+0rD!IID2sGH}n})WG;ry|^h>e5J)iiaI(B{+w)k(+{AFF8$YwBz38nhF-Y8
z>r6AW+Hdse@@SWoGoqo}A|>6^=MJZ}iF6T80HxHWQosxU5N^4oza_bwMomGx9Pg<-
zu{PB0{Y{W3Y2aH=joO=Qt$2}5dT1k4bdC6?tfFt)rTN+gjkQBJQpP-j0mg7;dhDGh
zXVxK@JGbNw1Vnple@!4K-=o&GnY$G|4fZ;Xj`s$<WBh>ZQuyfB$bV9RN^?I8J5~9W
zP}X53&lz1ZfQsI8j8k%N{a&;cLPmKBATY7UK5cv%JG}0UdOv`!1IO7y7omu%i?kyQ
z+<-*k*Uj#+e`gI)4TUoLcfqifEPT+>3q0hhn#L^>z~7)RK{XEKzurTZv1i7COSFof
zqn&-KSO}=05v?bVn|1<BSTM4o_HHCR!sj5Q!M09dGF&A=A48?U)OCh`CbZy;8N2dl
zpt`WGYkzm~r}P@Z`HbT{6$N1qIxa6r;N~!3y1dLG?9F|$=eQDQf=5H-AsMhlL(610
z%Rn|3I9cK;paW#U#W9v16G*YfOvtPts#(iA-XivvonKKsmE_W7nT%VSorQRNmEyG~
z!o0v^|00sEzhI+S*mi%)(HF_4NI`b~F7OyK!h)zXPCSWMHDFW3xYaH%gdyuOumy#}
zZ}<fG^#B@s5b-qe0Ga<buQBuF59BbwT&w9>_vbF+hu|!DfDB7Y69*O(QNijQQmMGr
z;wIav%wi8X1<4d)!VMna&Tp!6k<8YDCmkq4YfS^y!4@h6mPf-h4+Iyh8kW;Z4z$*c
zIGWCIuH218w+b)r8@@D&CBQ}<$U)`Q&7u(+`U&XQVK_xGpdi>2u0M~-0wotFS*R-;
zh?U;%gU(_R@saab3FQVI?TQs!wrD5_hr4Wais7-aSj`tKTPN($;$X_Z(?y$NEn86p
zS>yZ*E`O6w2eo_G7uaA9+SH+8FX_6aaYRu<2*YOb1ws{<64W3W{Qg{y8{XsdJtXrC
z?`5N*!+Io5h7_a=F_|XdD*HH_5&r$2CkRSz8f#9A)_E<Y-Mn4Ru#BnIz>c~@q%dLR
zi&sMXz9`U0?+P6P`CYtaAe@v(67qSnZ7@!FHD^Pu=ZNk@qO%uR$D-by?eYF4rJ~`<
zyTcTFe+|1NRu#JaaQoo%HgxFhu*NorPH(Ib@=5}@wZ&H^_Q&y6^n6bBlC71>`t1(B
zL{hbRCBsjK+S4bjT&6mO2dm7AD~scoGMH1xGSnOR&Xo&tjxQ66>Z8N=Vc5eTf(fov
z2>*7o2<ehGm%VqOi5eX^WieSJWha%8-#qkYoBhf_LxgkCwx9P4(RlArc%(~cVgER?
z6EsR=vi<kh<E6!FWS2Q;SL6BRWa6JG`zT-Rs8sT~Ap;}j@lH$^=h1MIqF>t@RCaEb
zXMt;(JNz<wF;WpTvOtl~k<3_aZnnN-k$|@>mUJFDM2u7xD}lm^?hBcQ7&$}6?DOjR
zkfkc&uH)Wt?`cfa^ov$?wZx0L!AMzo2%{{9+>5DDKIU4V?gNt#4gghMCx0tauM8RH
zb=;bJ8@>1X+3?ZPr-XMABGz7^>filqP;W2X?*~z&kA{X=T5PY!_vRcgu!l<6%_?95
zmwW}74mrA>Ee~IJJf7q0?WTUvAc_Rdm4^q_cDEt-=fmnW$@Sc;AhSE6-V^sy1Zh8C
z3nw|<Opc-evicNBP?;XIYn~0Ye#uNKq--+XhzKHL?PY%a#cB@9-QR*e+jIFlD_SHp
zh|%)-J5mm4<T<EvWNrOTS42OfnX$FSAR0-dM6WDK`IPaNP;vc9EZ<B6o1bzLXcfLa
z^wWZj{cTrcJ^_kxSP+(6;|$|4QjX_x7g*(*aM727jQ1j9SaNqdiaD|n(Quk-LUFNv
zH+%w-*f4bkm8K%wDqOg!38R}t(5ifI45lI?wmAY~u<5XUwvz@WC+AoYY3#H@p|bwx
zAiIc3%sXEh8Xl3v=RgH2vZxmQy2TDw_m(mJxIx_4JE`?lEntM0%}zjwkX=FaT9XLv
zdU+a51k9ZkL#s|pdY-Of>|Nlc1?KPrs(p+d?1f1&n<)d{cmhl&!eP$W(*{;eL0>44
zYP|7Wzqm#A)cg2bY0V5T#Ub?h`9o3EEm##>sE}y=9bZ4P8McQOiCI$gQxc;oHv>a6
zaw`@tPt0NwNMH(d!|*0PW7IMBHx2c}a;Ai`v@rCxw_t=Y5NwhHqrwc=0&k_1%Q*7>
z_7sdvA8i54g*w+~zBrhIZjRx+2L?_!B{O`b46GsakdpLM_nOPUoT9o#))pFwu%Gfw
zeQrUI%>uOI*YNw9BOq#x)n2U9k#c}_t{!L&K$n08?5n+b``1qVB&;Rv!niu*ZQKy1
zh^m{=IER<I<szS;?iwtg>w3w!?JA4x;!b782U4RPieF2S5(cl@VIEhuU<sOQ%fMNt
zSpjsXTFKNQ<-lY`)s<pwv7Z!+V(F4n7CTX7N?kNQ(Zzo|oc!L93muQnxQk2?Uvbh>
z2Fujp+r!TERF@=gA`5XZS{r6B8XG9ePl}$fEC(lo;D0~H!7`*MjyP(zikFFMv47-u
zz+SuRZ7*PRA?jnq@r~5Z^lFYaw*N!bdxyi>bz$EjdS^t5(W8$pjNVJ4a}z@JmPGGe
z#OR|%XY?RJL<<t5Gx{Ki7DN|qwCLiy+|Tp8@B97fa40jgueJAH>pXv_Pi!&&sA9_N
z+DR9q2-{}2MXsC2qYPSY$9b4@xwoNdjc?`Pn{iv+Cx2A9)Z~a`YwxH2yo<*ayj__|
z(7&5|o?;&M30Kbmv$Ej%zT*YnIFDx6*-`hf;vtcYU-UM^;oaCN<g=qeZ>duLriPM7
z5PCiy^^iIJu$tXJB;lO>GkVdgG~IYmTIGMNfWmT|WLY{)lV<<Jt#_+!av1QM`5HAU
zPfM@lb~EXae3bz~K+nvI&oia@rj}F+*I15?*E>ta6!W6Ok~A9M`)Z3~?4^(-g?g98
zBToAw+K0o04;9<g-ISEXZK;?q8+TRGb?8YR4e1=0tpc?JFaI8hxRzt{H)XtML^X9Q
zu@s57;mOh#J)WqKVdA&rdGZ@|{cN@8^GSKwFML{Hqd4Mpr9W_Leiwm=3Bw+E7i{%b
za5@Xae<x#&ofYl$aL(+&{?af)Yz0ozwc1c}Fbg3ZRjVVvN?28dz(Z4q8mdP8>yQIx
z8@50}L;1#=R><e-1zT)~Nifxb&?9-^=H7y?e4G|b|MQ#FQW|S4y(!4c!AWWf4vD2d
z2gNr|pD;*k71|R$w}}wDZ2?jpL`gU{wtlsz3W`#2Z1``sn?+=1b@jSA-e%Xz3$JR}
z=wOq|XS2>Obr(Mn-%V=Is#ml@RecZJ7^W-$PBd_hHn2Fs>wqfN2#sOA%q+^s!nhhX
zq#@mhKhC*+jC1^;f8@}t3E?&9#{3sOH71J{|28!%Rlz3eZx8Hq58^cy!+8^u<p!{?
zXJxT#*<dT#0CV5f-|Q}U1~z26v6A%{f9sa-UD+gp+6qK$!%oXjuE+svdJhTt?=xct
z&iChk$1<9<wj^BK5m_w1bcMsEg;_tke0~0+eM`dYn}(L&6JKDf?SXS>M`zn=*pRWt
z^3>&q5=##+KN9ZMJ<qusfOe9IeQnQ9&2t<O-Y`$gvNfIXph<xUM}O-fQT%Glb!mg*
z+yJkbc0zPcM&R_2ip(&z;Q_Y4vq{z!a-c)pk@1PmJ~lC~4V;{;)u%R3t4}CURx-)~
zt2LiP_Fji-r7+15Fid`wbmiC>o;<nk{=vp^xrstNBl7C2Pn9@b2@%BykM^<Et>?Q%
z<@6xzs8CrB%}!7h&TbD}J9flb)7dw_olGVuy&|n7Q6{o6%7b&y;51vfM8_tosRt25
z<VIUP?h<5AHX90YQ=9JrPdwbW#VdpD<bcXQ6MJF1p63Z}^k-y&%xyH=E*tr;LXWhv
zN}k&!j>k%V=WGOfPCVY8(^thC+*k(mb${%|arI8izc2l68pdinIRrcDe2L|m19`D&
zYRyi03miQN5G}9oDc<9l)E7T=?VfXp<D5NBwpGr3iQb8oeB+>dwxqqhKIEpq;T4AY
zqqB&6$>~+4Nk`1m?EqTE@-q1&QZM=5A?$RCzan2Sn&*&8G+7+nEepY}xo!Q>`U+YU
zFSvoo%bIY&?mGz0J#Glw^mlXiv<NgpK~8*5(mg6xvtvHfHu7&*wp*q<%9v~ZYvu(S
z%44KhsCKy5$P<>E`YEh;<0b0lOBW-Ya_QfjzAv$Pnc>(k^_6@s{k%S5oO|eLwm9=K
z7rBE_!-g}a4A9>;PB4q*OV%tUON{p=Kl8qDsM<Hht0|hPpptC~%_c6vuNx5i7}Cp0
zczMX|X*ea845{2O+643*CmHtwrB8NeLZqL}7`uFTu&F#gZ~T~UQJK2q8|2)yzI7U^
zy7@=>RzIyNW%AU}*i5m)S}X1^O$b7gNPJhl@6%-dy+Gs)rkit|3$DGbG2_sD3Z#;*
z%%@t+fs{hW1I!fgZnb9YmT?xOjuN!Tw4LkTWy(4iKZPO>KsPL>-JM$E9&)vW&tfGl
zPh}Wxew>~mJH9FE5>ilFX}s_B2E(ze3k$=0^qAAw=?H~1)=R5aS)xeelYcSDHwuGS
z{GC#Au77_5<MR{2BxG{S{klInQ42NB*TD$$+{CRZ%z)N8C|>5D%Tbxa`b?f{w5)KK
z=k2d%eQzN&5%fiai060784=WCXI*+q>lV}8f8<Qbx^F^<*sM6jL37f0COGrg_e)x#
zomNf?9xu|i@F6yZe)(N5S#m3+R_s3abui;=;1I+MW!-#rT=J0W{QXvLw5zmy_04${
z%^{?=KjN@DYVRBHAAUDcy0u8_=ew!nU!M0S2ocuF`9!Tzr1QJJK2ZQTH+q!4U1dpx
z6TCH5|9C#L>)Q?`JJMQ{^mu2PuaBy2gm{ZzeO>ZkSp#$G3UL<aOD?hHJZ4M+fj~MC
zUc2g5Ybe&HV*KC7R&V9b_TGb83ykEI%_2%Dc8{nM+o5eVrS1vF81<AAQG3)!LPvD*
zW>)ZQm9iymKw^YenoS*u4EBjC)hC(V-IpN-pfjb7Ojtq5Y-W_zx;<zDK62|nn~WXN
za4{G*Eaw*H&vDhfFYQvMI<E<N{%YR^Z{C9<LLhYt^45L1{t4=3pR*}Qo+nYn9+VhC
zbJ0+PDQ)da`8Eh^-<4g(K*_!@;z`N?s2VP=Ynq|rNj?|HT^!RsQOf=78<N3Ep0Z}6
z|B6+@J6WtK&5@A6RLRpwc~1E;^(aiR%^SXlm8grqm(%V0eYskNLps+R&B$7%;I_s}
z5O<0EhtSNx>NhEkfI<y08q=RVi7++j--1i*xP#%J7hY!^w%$qE#=>y2AW!GtP4#7r
zyrge6o_%d3S^Y|~;-|D+6Z2IPG|W2aQNvARg+_PQq;bUIs1(RxrMBXPclqPiE{M?y
zFf}ftfeFHaj~E5uJQpO$2;Zmb>#eFvxnVlWy6F**E?KtX<TL}3bQOD&n}}bQ?lIS3
z@2JVL3RPwBHllk38Jr?2%6i@zJ9Hdq@`&A>w==@}?E^m4{IrCkNmxSp{gmo+gnI=C
zI^4iR-)$iCC&0L5Eoz1%xKi@ZMa`HvAKLPrJrjZ9t2l=CV6f&%WGN$I=is}i5a~>9
zfYDxLne!W9H=Vrx*4OSOE9!p9K{Ap>b-SiTzm_S+qT(}aM@j9KIX);|P%{ky5w7~O
z+L(GKlztob-^er{YGJy9My-DuzBh3!RIM5?7J4^1p-aRuQR)Tv0ZL>Ki9cK(*@T-e
zN|l%r6N@?{k#drD6?RQpha}RW=<m9{L%67ETS>dQU@x6mzfD)O$@}f*5vpNFnjGs)
zlM#8Gi1Si%LP0c@HYCLfJDXj8qhod0qf=0~|NM=8?h)0{oPk<K9{*DWIdMkAFlR=#
zm5e)}0J~T(t!wXF`ijT8W15k29i?J9OqB-{mavBGx?Zl)b#>O02-tEPya1XgqT4B@
zvHXFNK#yqsnyUB^^Qlq$Vl|V+-t>LvQ@}3?Howk_=6GewBqmsE&Z`L%akcw7AlfC`
zlQmfkYX(|*<ANT!vIV7`!yd!6C()88YcJ+7qC8!K47~=eoLg7xGQ{N72g*(BAKZtc
z42K#X!;R=))AA?iZ#-1pdoYw<Px6VU<#}QWy|m~Lhgpw|M~NWK^d-O(8yb-9t4U!d
zrrB<XV;ek|@Xhc6J$EjxSEDbI4<czzHZohF8R8hr=ymCR^1`0VXW)DHiRNAgv{vI&
z=&Yj`(I3qxu$_7pnYssi)I%W+Pa%d;KYKlUXL~$RyAr+amWvWsL7&@mSL@syO^fj_
zz8Kl?P-RFcHKLi~uLljhA+j&hp8v5{-Ar3>)A^1DSLR8|h#|U;*5HCGhxmI;ng;cV
zCe^KzzUf0Pv7gXOCFzl)Eo2hp<Npyt-YC4`=yhJt`6qwpk*;_x&}n%EQjn6=?)#xp
zm{(=G`ia+akTVnN<?Y?i-KufJa^*N=pUQFrodDyT%}Lg6ez{tFow2w7`A(OKm#JWM
zqe$QSLm{o^mE`uIt^6r-fIUO3%N-&~rj6a&GTkGLvH+M$F3lGPzSJc+Q!7AI?QPtf
zHcnzdk8cLy01sA}p!(dyIGm`2)xwd`6BO8tW`N<KK3yNUT`IGoZh@{Qmn<W`cs%#k
zFlcf1T^L$vGms1GhlV<LgOku=v*S3oCuuTJ2I<m;=H{KX9fwQ<R*O}2j?N$eHmg6v
ziwWgO!afVb>dh!WR%3!^1CI`=rlc+GnA?)i@vS?!fGf}Yo#P$W#Gl>i+RQiC(^ji^
z*)822-NAYvcL0z-&Ts48aQ-V99Am}8(I>vzbB?{raQu${{5=PzA<f5RP?{2-^pYI!
z!h=`TiQls)_A{vQcP0O2czsNS_4J#N6_%IQbFC9NL=-42D9fSxlD+1WW}xXhGb_^m
zHGsF|+-=>P5{wV9!?(S^EAA&aJ5scLR5w!+m4&y_ImQ8_6TXFK4Op+u=A?X3A9k{1
zFtlKN>(@iSQSx6F_VL$sTcqrfj{0(5jJMn@I)%^{mN%laIO>cD9d{;@RZTMynX_va
zq$u7H12gJg5@0X7bv}JM_Sqm2T42<8xA?#Cvsv#q=LKbcRnrfO1lIxcTq&?=3Vj^a
z9<ELw@a96PS8D8sys7i@_xluMh2_xxG^IidmVIhQ!0M0gbO)3C!7}u3dfKprWB_{d
zC9Cv=gvyd=#2p8iZnQ-^;*-$7S7u0Ht04PcSfw51QtP<H=V4%z3~t<(2bT1weY1_G
zio^(DD^<OK5xNkEJRe^RFZTLk`QHIfplYGveKt?dRRyZl$rOwSe3X+@P&l9(-P8QT
zSj1_}QdIcfLI5H;mcA(x*7nnQme6uSmy_}*FzZbmFxDzmDM$A}%?SGeOj)1>2ElaQ
zPRhA(H=O^y59m<~6*EvY9sXU#F;}(_rFLfdGOkG^<xmceSfPxTlFeJn>ui<@R|7Id
z3c7M35BG{$TEOY}Kv24HXZcdArU<-86%K;1C6njBfe+C!WcL*tNyj_?4qQIysP7kj
zQdcU$m>QDBlASpymzu_WF=rfm)JVN6yOuYg1eMFMrrc-t7RZw#Bg4}?@y79FIlN{s
zIFw28A9c1Tl<%ZwynkV<)sB|h@ek3X1koL}QRqdm(8=I$iGCpFjHJoibbH<{@6D-o
z(&g%?9~a!xp(1c%vb82YYnK}Jq~Y;TkyaecBtZAd3@zWayxD%nu={dm$2q(Agl^98
zD6RZ#g?{Dct?a0*y>e2Yw3^U6-y`SrGel#GZ$i*JC(E&~M+0ta#Ihe$-OeyKKg`_r
zcBJ@C5*UmIDeDxaJMENrd$+W*D9F{<0xs`)KP#$gu-x*f|A_Ww3)(YY3Mkl$t~nNY
znKuxx^>>u>(w8%xcHp$uBK~OH?K}Q~^PzdjzE^=)<2!9LS%aoO`p+cm*7w(f7H-rf
zY~!3mD;sW%=32hZS2};K2*xxr?``DcG!(udi0x24xtg`BgkE?k%_N0JpDtuzk+obq
zHjS#t{3z2VQY`K2&DPSnefS*!SMDv%9<)w5mam3Z_rCXdYc`&Fv^W+m()fO_=o3L>
zw8boFrB-5Z5(v^kwjn%TA^K@okC47ejd3D9Jo`f{H?w>G!75?#mL#hxsQHuCaQw?L
z=ds$|FI$4YBUWIH0-qU^l46_a`(UA&msZ0gj{?mo(Mbc4xFWaP<0FG-^URy-^Xy3-
zH6Gwcp5#h|q_2#N6uxA5fK@b<niZc=QmN6fU7PlshlQ9Gk4aN}-x@O&{lG;;9`PC3
z&jCDq-rQ@^iE*X{bgjy~KuwcOT8hbf^D86sj+VQ`%_=nIA~W1#In$J*YTW7Z7E43g
zQTgE$_^&%aAXN4YWKW4rsT3D^Gb*bl{U0gkH+F?db};^Gzf8chu9Wyjy1u~6Priiq
zIQN&Enb#5iV^1w}GvmM>M~U_wawdrptn%Ci6jge~ow1?GF##v0(Ce9ZBVUOIO%9o>
z({G&rEncw%4|KizEg~?${JJFx3iqVLwbwZ5rPImkfAbE>&59_rk_JN1!K<wQZt$Zi
zyHE|79F8s+H&%$y&20O}G;N|5JX){Ac6cD(WM;6F%VD^YzhEa4V?l6fG?X&(x@ie0
zRjuA&mF&cD9DIHfY*n5$(NPbI(g(Hy)Mxe}o%w!FpRx)OJ#oQ^WoW+&$4sGn-<A=l
zum@qcHE>ebe|V<}vG(gLh}eS?+@JHLE)0mOXjV8`ri;*fXP>~Wyx>uj=*Z^qz)h5?
zFr<|6+5gV5p{!#cu~N8OJF^M$#K}AZQdeT=+5yw()N?Hq&sM+3JvIO=W9|x&yIy&k
z+acn}9?HrDI>Cbhq*}1Dh0wR6ZUqaXloR^<tlg{0ivOGe_}X5Ex3-&{#k{unI~1dA
z=~Mf3V(AcmGuHUb_Bn0L0iW^`TUc!wLlwhvPBQfW#{%e(rHc?g1bt<i$7#fMW|<T`
z(jJFCFr|uu>kBp56gn6%s68jirDJP9OmJ&Rs7kPA?iLQDKn>NrzYgwOCcDA$)LWp$
z_qbHiQ4{Cfeta3jramAt`e;;~a_OTe_?|z?Dz-D*kr6;r*QM2JZKIa0YfMB^C>(k7
zUvn}ZUDt@<{^`F``kJz1t#cuqQ6i}HEIQ&9!Yax-4Dd==ScZEr%i{g%_Ub)W>|wD>
zmg)8l{=_>!Ga%6X3mZc+6r>p7Z<*W*7-`FG5_!xD^420__P3mVMUy4@5i+n9u#W4D
z=O5&EWNjwgqn{N|zb^LcwZ_I*S$w_tYz{idhse8NWH7g=xXdqMR{7g48GdxoxAxCb
zFU(~S%WOMu6WdXmExs|8@Azo$mF1MDyNfq}13f@|`a|Y<ux9#;v2Qkq`w;qb9A(H1
z9WLgJH(c0L3EuOaZ<yY4|46+3Mm)FOnGI7V(-oAN`4&yc^#qZYyceJj&37bX+Is4g
zq@<9$>r*kQ{$jm<<6|nu=ahLNk!R)F*yXCS!Oy4hnMd~a%zhd3-f0_P0*5mX9fTf|
z3Delw=kNCDde^}pd$5#!D!eHfHS_3wFk$vAfV0blL7otS{ksov2(!qvKf1))m4|x_
zzSrrynTaAmsAj(;uDLJLxN+kcJUnyHVrDteTDMTkdY9A?#SI@HQ+4PgUrX=^uxN>9
zmp7-spR8o5);>3Q+FJ`J2HYE-VQyv{v%=vB87ukHuyV3rFqt=`@tZ~5ogx;jk$;&3
zY};>6`J_s%<=3MnU7$zAt9;IC4YM*yGW933@PT9FA0JEh-1?K-;9A(?zZj?nUU3QI
zw{{_O@Ljuve{`%fd2JO%a{_?lE3=uG*seBeZ3mBQX+>+*SZc#{Y!Ym~$+uVww#o~%
z6ks@E7s1&@?aPh~c3c~sd($tI!h~2^YIXP)(`SB(3$usWU_Z0LHnB;t*e@W(0f|23
z6fY@RzP1RhRpYHC<B$Y+;7=D&;HOlEp42}Kn+aaGoAss57<gJ|{+EIEip}b2c@N$v
zm(LD^vpv+$WBJLmvy#V-{4XsTVz@Z_#Tgx=0+(p#MDn%UBN@vkb+z<}EzLr6DAIO;
za>bwYyEo}%U7VQb?0Zd}VzA9PUWmM-4(HdtY&Ka_Kr-hQX3G^K2TqhZ>i=o$e46I?
zvfdtKc)H2dt%G9bG=Jn^<s=o=Hj|bd%OiQ$;Nwmr)ck9~X&1Th1h~uuD$WaWgtwF1
zB6KFE+fKjW&_rZ^Fhn`2#(FT=PRcBn!F07)bp0iqm7eZ%Xg$&Qmm9se${|X#p3mUx
zNotsLwbSr$B5GCors{d34(6G6u+wX2Pn~gALR%NqnJSNYN$x4pZ20L*payAK<!63y
zv?I<Z^a49e^}zWo9C+Tn%+cBEOq8-KbZ*o-Tp4m6r>vJx|LgiRrN`kSLvE2;Ab%b3
z;q1&;f9F!e&#~bYR@JvjoEcw4OK4wUhTJqC_=S3sI%bJhwzu=gE*`yVIBckNFdy1G
zb9r*oL!Hb2SuxjmGWv^5ZBNq4%d#pLsqe*K#1md~6Jtze1U{~2JAM25Un60R0y9X6
zJvVb_&|w@LP32v8ZL0FM2Tc4OV=Z3*ciGn20@L-eFBJiN37C1VHcc&;Fqf1de)fb}
z*BMk8&kAA#A_JiWwgy~8G;Gv*dy1PMi61TwsilpAc~=ZUqGX&jv!g?<aP~S+Tw6k7
zi(lDU_pM(ts&#8ioxiP$&9h2TrFoJJ;?&JY2<4YS;27sFC7VR=SjiU~77t7(5l+Zq
z7d5BRL4-b)9g{arKmUeSYF32a{>84C<A)xS^uH#eRe>2ZVpcCfT&-w=3ls@5R<(~8
zsUsmjV~2tT9}h<&QKQz#h5c-?K6?2?Shn_uC$4Q54sN;27t2g*3y3zVcoxi_fc=r?
zMSu>ZMkTu&Kb-^fPSxe)!XD|0HfSRYDD&U+C-eUS@mIA;G>ny8;A~)~$*cM3j0~d*
zW*3e&y3IcdJv7fgl&>WSBE;Z%9s<z6Z69-O+7B<IwRMg$xGuP(C8~eVd<Laop?hK_
zqZ_Bvy>lNd(tA}L%cST#Nz{#E?GXl;0+<sjFDuP4o*a;3H-9Z<&T;+p(#<b?`2!a+
zjD&rWR=)_%BaYe+x_^VZ2X3zn{>ZA+s{airaQ;GX$vH=RG~%TwIS{X%af<v+>wM&G
zVa&T%`JBwl_YBjlYX5Xte{hpFV?!eaaBB$*mnZcYgAVq@F_P=+2<V^UyqM2Rt_-NP
zA%s$}>{w{up83p1>tp=vt|8pk3`xQuP9!^WdMmpb|B}>;?F<t%uKI7?<sLN4zVZ3P
zY=oj_E$}%hxjPxoRqEwS$xH94SN_aBn@0>?pQjv>tlrBN8}oiLC$TFpu83kDx{4U#
zdS5ZBLif7FY^a_~2u+L$<YUn9*S(${XfCKNQDxE|{s`?9=*!APj%OYtvu%CeRwsDe
zv(2-Z0rdyb1sOWT`|e}xJ9P~3!g2TVC23K;+D+pNQK9fvWy_4c%n492>8{0jk)E0t
z%}E=ShT;*U=tQ)Yrh<65VWmm|2;f|V$e(_x8$cX^yavx!xlx8cb2r1k2|IeoZ*Ub{
zXJl?WT^(0fhOo!q6IG2;S+S>c>Qj`u53k*bwdjwbO~Eor!KwocJzH;q10l(_d*ka(
z6?J8|qB4+K*T=1+FPJ7%gEmD~1xTv9aiM_6nB75k=5VfTZOlJy5GW__f9x8QnuS0E
z*6&O9>f&LSgYg3gY%Wl4IAR7~z*$PnRg_h@*Tk*%b-JSn*4D7Zh_4^_T;n8zjGMcJ
zo9}5pQsp-KJ2|UpE=3vzCtVP=<O`fWMX)L4q4L|YTy1fN89pZc$>R6JoR>EH;YS5e
z@J;POe`47fm5K4*R82Sm(ilWJZ(>Aa)rV*Mn}=}B_|P8krFg_RgGn3$kJ8;kU^RQt
zOQvU>8jGn^_HBrMo#~WC$GTK2#YtsG3HMUoJ~_7Mj}BVP@;qRWc}f(3iwgn8(@mfT
z98x*y?)x`=h?nV8BoVkyJiOPdUTG)V+ZtXrCA@HL98HqeiUHc8-vwr>%-s&^OnYV_
zkApW&*0QU*xvqo-W*}j1RxVJUV_aMYoeS>zx&bz~b18K~F}m0@CR{qyJ7b+qx@3K^
zeZXS(br!t+?<MuCm2<f&%Tb);rQZK!0$a4$d2|%@0sG$H)lRQn+gH<idi>?$sX#O1
z@;#|PyK3Q%?)Q|9dw3qbrdC7MIkCj5iXMX4j@C*<JaJbq>!jexO*--^kFCRtL~?cE
z<#7I#J&z;ar%+v%XgiR(&?~k~6P(Q5(ZAAYO(_5K_oA5O^mH?i_m8sfr|R^pFC<2I
zlvQXwe6-X*Lf7Z<Xe71%F5K8;lK%58kaW}&(C_NEaxg{?iO3^T51-kv8Ir%+yk{y|
zWBn5@7xOIvBtnkP+(P-^)~N;O^CA8L8X#7#`edsUuNO96Y_#^vW&VVDvh*#p{H%l_
zlO{Mo)yIp%`>#_l!9fkjXz5KU{T1^k(g^~>qgYvfB&IpBoivmH^2~T$Ht7QA28?FY
zlCxhi-Q6W{o9mO=IQ03pP;EelFn15_En?$KwT@QNg|XYE2$VhBQ;~pF=*W<S(2a84
zyl8BL1RC3n+q9B48vU)9#Gn53JkdB{XRV3%Mtw?H^@U5=)M<6%X$5G*K56egGk?&C
zai29;5$P{RLc#@J-{f7Bm`%KppvHpAOD)({V?CF<c{#{Sn*K~Q{LK^Q^Sllg9XJ*Q
zyV=jGJB1YJ^*IFxznJ&gN*vHEAAPV&KNvHTrhsj_^Z;>9Sf#Wy^8}zHHCL<N{#gw@
ziJw$NIQ5OYtn$H>tt~1l_eNV9PZPj*3haJOMP-<EguKP`;*T)QtS<59c2aDKmyiNU
zpDFa2;BYi1pkVlCEIXMEw>s2@hSZ|f`c33Lv0m5{DG*$c4#RK$q5T=;k&s!mS`>0s
zp+h#_Vb!W|)emka){`bH)FJ&)6{r3p?x7=$70q?KhD%1m8JOo@Tqr#9aJKHO^ygYq
zCvV?smDP}#1h+r8V7)%~eKeWBk0V3huFszr|F%>LWw8+P6sIqvFG@%qeC^LICM7)V
z(d<f0n-toD`_WD%9<GjhLs}N#+&Ud6^ZTl_<1bx4_IpRsY!)kY7sb*asdSWlUJVU!
zk;m5R7mQ;`d%YZV-g8);Nf<3XXswBS1P$+r^(?7svX5&WOQ)p|U$u@g)ZbMZ@DDe2
z<#zMXvHtm{OPt9N7KGcUc=QIFSv?4m<i4gp)@>RqQ%F%DWUb_St0f5??HUeJr;_`K
z@1P)(O4mn$V{6vdEv*C#iXN59Cq$7JEyftDXX^_@hb)U?0*Eetr3SOBh~`3#9gY~{
zgqro_;PgIRFpZ;F%E;^)XUF5AmdCv)cv{joM<lKM65XSIdAx#|0r<D}0SAc{aGXK@
zu?Oc!ki0)Wr3(G$;x&R*qroBR&iby$Lf0O7$e3xP*-8DxiU5THogRw8{^9W~u4%G8
z7Ce&TWJAjUeSylxXAMRwCchZh7^ML9Fldb$>>AMYfp}5Wfp<fa%pk0xbS-5-VK@t3
zKtHH-)h*ocF@TZ<TK2Sp_kLthv>(oo2Ej?vdgBYGTl;kP<zTsw)9w(L>F2LMrH7JV
znSHrwd~eznKo{k)J<Qw9%?I~5SFSEe7~!NA?B_%((|OxCzS@quEDVjcNJ`0ZBAM-x
zgu=cXvy}JZInR+INMEuiCk3YTbB(gzkn4P$ypK`QkU3MKRU}k1d*}0%E2%-AjzGAr
zQ%<g6$DQQr_fY(0jeEPR9DUe<j=rvG$-|3{kk~%av}{K*WvA<Epv`P|OfI;c;g}$-
z|C7uyY3hF{3px$#yoXgQNYac|E9q5nPo~R2@csH#%A=%X0?P8|s)e?QH(H5?OqZL^
zL{Uw~pm%rxdw-K`tse<vO+6$#a?ge#1|;FxbV(w3Yr!N{0RtIB$DW#%PP_1+n5cfA
z%PC-JC3MQ*P5#M_96B#j6^a}Z{Vw2RleLeZ(di-dLtP&w)AI1jV@&myhRcrE&In9k
z#&Hq?eYw0VP7%9F5fCm#f){2cGPd(HFdz(!NC^WPY*jz!S&moN!1Ivq@U32_l5ysq
zqo?Px!AHYXk%W%n$brt~=7ZVaK}_gXAT9&Y_EcEt&Tfaqj_&mOa;^U|fvyr5ra8_b
zDMom8X(uXuiy6d?826Q!QC{>rOe;tsQ;Hh^WBDWm_S%?Vy;)~!6xafd`lwr-u3B6$
z6%Egfb_%zybuPz?UwhXv&tqMh6n~F`Veot@5FNJVwy1Yog>@gdMJcN#vJxY<_P3VC
zOAcLWIS}|jw3MPeD&FO-YTvU3`5EAwJI8j!gON%4Ek@e)>zIGYx}&;Pz$aPok~&-c
z?KokE)idN|e4{)@ZcYM8sHJ?URkPD#N68F+{WW?w`*1S5Xe7HMCHL9ugFMNzi!U>u
z!q56@_FG0EC-bN8B>xUK=LX41^NY2LTBA(4_fP0R$UI4iR6N<57SHlDwWF+;p$$~j
z1S*=}4A@(t$0B|X5O?v^LDLha+ktQEWmfJEV?G<y%T=tWJKbhZX6+Wi&4dAVH$H=}
z)Vv%4r0(dQt7jGhb&N1}R1<K#@9UWOAXN4W5MV+P`_4kUkVoI@)w;7;dt?F0A{tq+
z08Gnvk8Qd%1D*32emF^r#z}d}c{m6OAR9<^@_ZB$rEFRQ2dzbaAb1`iNOeN;p1)UV
z7ng>7<a*7LM?aJ&sZV3^O#VNXnlyGRv=|y(^<8etq4Yd-5Ptg?TYSss#P7r_AhvNH
zv7RR>4n+%xcbdL`vE(}8FyWT{b|$-|MRfB|ndVsgcxH~!)|JHNKZeHAASgJbLh>!t
ztHQmk#&6^Omb2*-N3_LnA3#_!4+Oa}-jEGAXkPy929HnJ&IpyZ*XwNvw6rf55?RtI
z&r~@(4oWTEX5ag*b<e#l?c_?Eb!9lbcl<UN6ers2J+a`7N*3)L%TC_bV%_V*Jy|$8
zKFKUw^DILi!s(@9!m7>5{4Oh6PCjMt|J<~Wh#jp`n&hl>SWbh9@+7$>y~qak)H8ux
z7xl#v%8_i_JZ5>1Fp$(Wv^&iO4SpAV>zcyS{fR-YQV!HO*&ukIcHLA+QH*i3N8;@`
zpq6`xc{Q_fVXf5zAX{Ea4x+uR25oNDL$VDkMc2zk%k!hWDno@!BI-F3WaRTbaufE3
zStm@z2*O^T3ahaRIyfDUmP-D*^b0@ajI9&pWGh%J2(&iu^Fw7LbvHED`c$qR>W6a|
z$w5wxwwm$Ddj+aq$w#()hDuk$>5>glv#z1T{9Mr_(Q+0?QQ+m|2Uy6lKe(OJPvkZI
zB`IXzB6>fIc{KMp(X9mdyI$*aMT~Y;q?V|XjSoBXdFP8p^L*94(j#<>#eC;7`m_xX
z;JGfb{Fa+C#18%LG*fJCtxFpLWwP&OOIoe9bQf}uI+~jLS${@2=K4;<%e1FId;_Gf
z)cY=K9F9*58npUfHBv{10l~|`Y)x6qj~CaQM)XJIF%1ci>>hhH{P9d*I&2Ty6)i0w
z7O}c!nBVB%oWCd69b~5y{BnLE#9#M!ynk4V<Z6$!gQn7XP!`qrVWK|sB<U#Cp4{Fo
zcdFsbb{EO@G7(Q%uVJqZ@YeCmr%(Dsty0^sTLYp$_|3Uo4q7}CRg?7MKEud`P>enZ
zXdG~8lUL*{4U<sK7FEZ(MMeS%mzT`(F~5@}@O_$qfPu@u0^n{@Tl7$9)xB<{90WZ7
zo9k==;Llv&j3|!Ft0Scw93Nfso*(W5Ah_T8{1}lsQCsX&Ltih%N#C;&&TG};unoE$
zt$^LLvFPudMYMYV>z8(NTBR~0WjdRyCVJ>dTlsD0!gV%we?cm~GLy3p^~4@Nw3WZ~
z$}MENU88FfCQ~d&n|w2}afXpNRC0+i#-m|JCDZDtA<g~>hnZtf|3!??t}QYW;f7ZF
znyE}1j)-=SHc!qRV>Fo*a${HA*bJN@N+zUFD(2&rVm8CKRdr?vxJ+=Dx6|=f-HJ@^
z3{0s3vFH{uZ;J`*^y-LBojikgn{nh)thpG>aL+Nu=VRrP!9PSBe`GchS!^PdK(lEF
z&T_5!0Zq-tMwmG)f%8$m+hEK!{X5meV+@n8w71|Vs!{AfQq=EH`31()PJb9iMLnCe
zagIeEcuK?u>HGC3%?u<@2c|cR2WnBdP+^QT`u+Qjwx<2@>cyqz=-24ySQ1~SkG0r@
zDZ7eCqsx>lk(-CT$)*0!AkER!MXb7{cH52Mt)FAFHp(0I+>wogVEkNf;nr7NaYx@@
z$p`XXa(Rc;GB!_19=h3I3Y|oVW3{edjk|=vkpKK=Uva%kl0s4Ye7W&y@H3jLtnktE
z4&}OW8h<P`o0Lz`LJ-JyDMuy8v;}{Op%X01R924(g$r|ue@fo==j{42mv9J8{0ncJ
z$>1k1|8fnH-tWr~cfo$aCH}q`M=%$>QzLI3r1W`V$SxJpTAgypB-QI`SU#GrU3Czs
zdXhgGW>I|GtJVws<_sX_mjkz`YDq=*)xr{VDW|QpB7!4a__j*AB@Z2e!+Wr{OOnjp
ziCLq2*ow4Uf$Jq8PS#}*q-G!>$a@4Q6N9l@ziA*)LKo}$D>IUEM71Bhs|I|ii?wcb
zl7`7b)CV@U8g(@@7FGxH*O&rTUQfCqVfRQ_CxJa4Q9P9OAEB)gk(wA4bTNBbw2c;n
zd?=C@_C2pL+!LQe^E2Lgjl$PADW8nk(H?+SKT*Q{oqU=UR8}atX`Q-t<*g1Eof^Mq
zoW#c94ApemzrbYxYXQwdIO^<mUa$pmfC=1(`8<D=F@~LttG9~YBeUgh->To5*do^l
zKgVH-W>?c!8i-WJa*<D@<(_bVUMR2pG2$4v3yY8=%pLES^q}VD6Ii`*Y+Je<onwy0
zBkFh1{f-nlrtn-jM5uTgez^je`UuQD0755m0|NcRVRPb54zRwSI2+PtQjs9GkGe<(
z(D+9R?&=4p;vXr=>T?lzT&uNil>UO!!Q?Kjs>7+U?Iq6EuZ_E`rX1fo7QwriH>u-Y
z5zk24lY)h2aXJCgV4O|9?Rjb#;vf2wUz5{1DVF}JZ<+m>^vY0#Sc$@9vB<;4Wj$&x
zAYW$FnA!s_N=X#(xt~i1xfYV}_{Tw92;9ItDJ?{PCVKUgD%Hs-2l^xQZPz_jc6-OF
zsIEmeV)y1}Cdbv(vgLmZ$?i%tnOim!x<1|(q?5@y&h#sb7L^*XXXSoEv<VyK-7{*&
z3s8D8*rJ+?{~9M0C-|tx>s9@O+_b5~i%A|eM!4}L^B%b;0nz>K4B^LU&sq;Qgyo<5
zyK0<}+N63#Vo5B|9~UR_k3D(Ez35uPs83XQd~gO;7n(f*46EmZ(T^ile?ZkvJ5c-x
zo`6EeC_Y4eY-4Y+%K6c+Ev*oi5)U_Y$wD8`yR*g*pNjIL@Llk49>)IVm7z6{xP0%n
zYh+VQPqO6}ASaCd9S8${6H40+z40^-U6Uj^a_wj6n_!CU@ey{0rA<diE=k?M7)P;D
z9#lRs5LU<o!cvw$aR%v!TK-&mH6p&hT066t;6A|2`!DMDDjJK(UgSm!-3i?|>w-R9
zGW?%)NH0pE9Ix^`C*n6$W2s;h-c3i89sieRvums=I`~+6)#YSLHPL(7M2>#6FiB09
zvRovpDbyXTLS|>9lCsKtv;`yxa~?D%{J9N_8LlTNef@MGd8=6INfg|kn|C0Kgfgz5
z<0<{ScgZ4UY=%l~%KEV)b{fAel!@p|I=!cFuR<-_zuW?0u4nY8b5|s1$b#t|#Fx{D
z`GER+XZGr%3X{2HaE?y0f@3Iy1((n8HX_|obV@B(*_lUIieH;8?EhKV45C}|@CAnP
z!aQKR5FBMncc=VK<DpKu9fj-a?w^^NBX~sJrddqHkQP?33d`V`VQ3{igoSQR8if7E
zcPRxgzGNZbQfle!0zK?z87$#|)(zVTdbH$7d6d3~d?B*{wIIqf9a=lf7dgmdk)c@2
zEVyW!1)slwXKGtNUBo8}4lBAj`fg>^(<SS374Ue$0NU?nxEIP^ZLu_RBel+LdL(UI
zUx>_6N6DBpQIVa#LVPHbpS=0@wxvj>+ey9QMOFim&vps?v#aEw|F*0212SJ*o$-WP
zK&$hjeAQ?{%E|B6J-=jcg>K*DQfh~fU}(HZ2=Dmkri9i;KC*vAbF9kmaePY4IOpJX
zVwt;86~v*05&3=|?WB^e3zs>h5*sb_ZEwk#{;`nQ=uUz4@{b?-Re)6XnctBt2_?H1
z!-kHKqg*#;<vLU|c*~I@wCQs18-s1wm*IU~Eo@t3sD@7s%Y@F_SoOE8^K|)#H8mZ@
z3+~vKVmY|@!58bSU9<_FtObJI-Uc){SsKUcEPI~CqqHR(kpeWrrr=j(T4OosgOFdz
zw+?t<H;Wi6#dm_LQysLWt^x$lRt3&@n2AL>j*tA08wWqw4YQG1vd|w`nK5nR=XWle
zyiQyirAmH-^Jpyp-dlE<O6kFei>IMbMvaqa7NaKU`fNv#yxws)^QdxYFw<c3i0FMq
zoeOdCQM3H)o^IiMdz{2$E$;jB$tkWa*O}A$WceeXIMM5meHzoUJ*ni^S=wWsXk;=^
z_DdAsbbjCb`nThj@Gm#R*&#9t_w1-Nz0P=(+P<x_OA&|Ho_m%gLwx#KWW@T7^T3dm
zeDSl-j_niT`gycseWo|VU|Vi^<D=d@aix$BQY?<M&h#4#*sEOrRdsH|8%3m%<nPxb
zh~EX1(jFpxC6+0E8^Jhj+!;k*!_xX@sqnq{HjE%F!kh132y<1THD7YA*>dldKD@QM
zTHx)o`Yl#aeq(ssuq@KiELH8HeGJujq3`%f30NJK%{f;ze_&9tpKI^$a%89YFX#d`
zo}<jS(-aC=MjKL!P7;C*qi>~d$o}E%^UL*>SIV5ODFv5~RR$tLw!<5x^$Hn9PbJPO
zH-fe-8$c!^#$)PMPv%a(bnbJ>%~1r{TyM6J+@uFR3FH3L=>@~2!$}W{NHf;U+sW2V
z=gg~@L(5*I%gc4X6{F`02pz9}j_aT9QX{h@BX{f^;YR;lwtLVk&-nsv5)%HV_R}@l
zV6%EqDgQrV{qJw}7VGkJCic{FS#*vf3)6f%5g+n!H#bghAXtPN3fNRzkFhqmO@x@;
zXz-T}1YDzEx0>T63WuRq@Z(#MEK}etTVZu8#ZL`5OTpTIye_acF5VgixN;M_<w007
z4#|C_XNr#$x)d$8JCrGx_-g;@CZ-(@xp3`218vL<f3ETVp#z=KF_}e#t#leN{%Gne
z?o~uGEvrU+-L~@box@h+@E3-i3hU7&{<4PD>_)j$c;E{D!#er!RpIW!-~5P!2wKnD
zr-@UK+U_4nv;97hjSrApWoj=XU6=l?H}rj5YDL|-%a5`rF1TqWh=9$pr=0GZyS?cj
z40+~TQ>4~a&zpWd>?P3(!5p8Gf@c$q{L5Ago(D?8NJ;`X08u~1_GebG@6@=f(;@*B
zA|lj7?t|-RFNt55Bc1GC;~J9ITV<9glK*X9Co*Co%h20)7qj6%{oCKDiEM5`9<k}v
z0*wl~24E0C&)EW5o+PBJR)f-{qT~E^F^VKume?V^CbNukZHN`A?%eP<?{hP4t${)j
znU92I+yRg}FL<}(yMOsocUH;7vzKnyOhXw<@3M8r<PNC~17&&ULG#~i!yXHqaS4T=
zvo5gw=M%eoQiaF}VD?hO=m&=pZhc=#D~)@j<xYwcF=uLo8ih>Z?qTUy9wr5qv?+d?
zz-N5-6L&vsbq|}hm~TZj9jn?Yj2Fkg7aw5YozkA1YsHMm)qss)rPXXS!*LWV)U2U=
zckzrgtR%0p0|{2*YJQo+zVxuE`0#(`A&)x$NKm9txi3GYoZvzbcoZxPF_uVtjDZQ&
z5(w)ZFIR_>r5IBhZ_T6V)j*zdw&5*Hk6E1|U2hZ>>E{mPdixvCK-SGn0xRnoY!--J
z+OR(gF?*ge=1Y|dtrJXAgshx{7sQGkZ<;1THMn|ABO006q<+6o^+d87FbG-bB;saR
z_i}DZRK=bRxHa<*_bIcCZ~B|i4gRr|rCX_Bvk2wgk}r9%VxPxmvB(R|MqAh{VqWDY
z+_6wDzBST%AF2=_j{V(%Smreo<pvX&#utK7Iv2xOS{%_KAZoj)aL<S8+xTeZ@2D8A
zCS26d6cV2y#ax1LHOe1O#HcUE;U3B2-EI}E;9b8bpHZ=)grv&~{#yMPB=LWho4?(1
ztcyZa^lx<kE~QY+H0?3_?38FV{CC?43H7$s(f8ZM^!iT^{Qp;xjK$^fP5f6)lLwG4
z4i8GfL7HXO3IZ&DN~WARia!sQ6H1n5ekO?PZ)8eZ#Gr?gZW<5Yz6z~|{h15hoMWWh
zM#{_huGK}oX@2p~1$6Z#@<VC>=?Rii=$$o!L^Db9Q%$1;yyenT>Zh*{a75qq?q0FF
z?f;9>KsJl*2HfiHL_Bheu)G+2MznGI347^v7nQZ^XVx?~pC`b->qh?(p9_q&*dcLO
zGXA@q;P<hK@cR@wvQ-HiSUQq4mp58}MW4GTm?5tgFalkkFXH)RAN}7Y%M!j%L5F^8
z<P}%K4}K)SzT(@g1@XVYV?_g!zB$B{T)|f|YeN(D-(JYe@W^w(H(5tO9L#Sys7xK%
z3{zb<4v!ph-xJ@(?J@Qs^{z^E7lKefux9=C$%&$tVTFEh_Nl){TbHUOrSUi76&tJ;
zqhHcw@jE>en#)$rPW)X;zU`{l#sRP7P&Q|PVcQG|I0^+#Aa1l>E?$ZSghR1=0>8b2
zp~Mrm;r0@+h~*zO1Vev&0r-a0*4b>$4-5B@ru%F*J&cf=CSpkbggPL|R^g6B6F|ze
zbB$p}9c)!t53C=wx*T1r)c`Qk4e@9_`@)Vd)<&NTxT_g%kU9HKg!nD`WIZ|XX9Q{=
zlU^w#hi^qR&I9GK(xW&hsc4`52WNqGr_YC`ldsj){_h$7C`N5(l;{?{3jT0>{Xx3?
zgY@Z{4RXg0SrH!AVHSN1W)vz0KHJx>y#Q|3A{YqW96s)lJ)L1+8u{rCShIP$9yl}c
z4X%#pPrHEZgRL3A5V#reZ$7qm5okZ}uCJ!;Qrz%P6P0lwfqs?)v|UG<;&62${2)H?
zHIO%$KEDnGIGBm2gqm&dCAbCK=G&e~)8lU^-h5g)ndxCKba!8w`pS{<ztV!B1a5VF
zM_qXG-pBR{O-@ZSHu1r5uL#^W6WH&YVBO~8VYe|a`LJM0CLm=fqI5ZcP~+Dv$g1`Y
zbJ1XEBajHXDe;7o)D@r#^#acP>svuz-T$-HfK~UH@F0&o-kP9fP+mnerf_i7xd8C(
zHu6OBO{2b;R9J@H{9z~<f|?A2Kk5m%-r;1>Kk)8+ZRFfNtd>F@Z;6jxA!SCiQj|+V
z<mAs9RGagM-b&IW=iblg=vr{1+*htIRYrSZfc;&sCgvNEQ$c$dcR@q;j%o87eT_J_
zn}vAbbjxD;S?=)#5M)I?u_BG%POhiZgF6*$^@K7$f<U)|?IDgB0ezQ`t;66A<Y^Nx
zemjT-?>*xk^;NR94)I=91^|+pHmY@FNKB5|mi8dvj5FW5Na=M+fIoHhWjUHD@B@?`
z(i(ET$}ITY)Q9WyU504=2ZBzs=105iu=j0uhH6^Asv0FqSB`l1`_qE)X&=9aS*H_+
znoRLt;m;d~*fp5iD3a}@u1WkPIjj9(@*GTZ_CBO}teRrS_=*Bdk;8`WO~obtIpvEF
z?NKsUzE#W`+j}#`Sfym}%!Gq?exC0$^4+{sb19l_-2Ko5R=IC3(QsuEZLkHdP8DT_
zg+@dI={Yp;FUvG73g=NXbq|3l24J>7Gv~D6td4R$@q#=}-)nN7Shi{OsFInM!81D6
z_iJZZf3pBUW&eTAZ@*o#>>6>}KZ^+lD2KT+i;+Qnkhd@GJ!%+MRbG&--%LugsiVSR
zyA69HdgOq0HH7KxBCkpE0NnSACgkRhtt|T2%SH0pU3QVDbZKZVzm{)_oJOUNWOKV&
z5&F1fA5pwNvs8IloqAZo@Tc=<DcEo(!&c{yB+;>_IYuq#(~1?`ib+HM<MYf#z(HD<
zl`Oev>X<jR{iAwVG&ai-Vzare$$O%TV+Y`Sd}vZq<f|eY1EYisKG)f`Paz9dKcU<<
zx{;91`m_MWbjjl^z|kv`l#A3*k*o2Q81swCsEKl~K^(su1qQ2v<-IbWzB=#0&evMd
z*YLWG<B~fxYhi*oes!nRYNf@Wii~6X>d$7(rL`2OkSa^@=~=)WkC-Pp7U>xqDFQAD
zUWm~AEF6d8nepGG|Es>hBVn7Skfo)2r&5e#_V*-PO{hnSw?W>Y8wgvWv-f+Gl6e+|
zHcL}bt-C#DzC3LtW$yK2R^a@!H#ku`ETK*s0Ji0c^ALq=+jk)FE%S#(&>nTX0Z(WQ
zukP3M|321t)n0)go0VFl7P@%erfzqK2P2|X&5y`PIc^oC*)V|6WJDi{(<~S?4{NdN
z^%r#aIuI2Z6Q!6J>_Aa|6YhjRH~l}$1#e0(Jz>pxvTO{EG-1=umkAhq-^FW9bg0)R
zrxMNcbF9=vOS^U}L+LtGX(mmHBJ*@H2t(af9C~<|CrLwKXwvnBby3rrk~7l2l^VN(
zn(Yhq+d=C1i{AjmMVr-N_@`3C55asL@*btYZvg1YAg0(eB<ex^H6z<k_D${q%NTCM
z7+gb$Z&^)psuFkhF}S!inDl~_)l$IAWPZqnh33WariuCRI@$OWA}njUGr-=|W-kep
z_6$=C0Q$_+qb-@|PM9MJYv#U9q_9B*yFnHmjD_NH4|U>~j^vvA;!SyI96;-Y@q9cq
z?xOdhdQGg@8rH)6k;F}*9z8a}M3f3die|3BOAye`5#x#cz<$Rm{tdj!kv&*l4_g3l
zm=_N_(X#a=5VSO19@YspXDS0yX1F#>kr}~}9l|MUkKvp{?xy)l@MI<KWGCHav~(Vj
zY$q8sdB<VSRgy{WqH<h$j=kil{7=r60@6K4sXsZ%5WnH(E)?eu0P+>lz??PSr1T*N
zc!PH$<WO$L3uFhSN#7-uRkAb+#Fge6Fcb}({?QKxu7gYo%nVHgw4H+<D48Vp-4FGP
zn7ZS*DIuXwo&bm0(6!0=qV-K3%n6biAsk(sEsk#PrlSy;XYlYXDPVT*z~WIUkwn*N
z-{oQ7gnll}V+JVK2@Ne@ClWl9ndk1XZTSM<wu-OKv|`wf>k~JvkZ<z(R8WHy_@#IB
zxtlmWKg1TTb;5mC2lOiU=v)m^M&IJW^OSjP<W@8*3_bc$g#J<?hRbU<AS#;svHX>i
zLVlliHQy!?d<mXA`M8~UqI_uV(<Mur&Y=EAiP`+O(WGx)alnCE5Y?j7S1ILVfPO(&
zGNhL%Ikm|>`i|nXR5$E$V%`45QdIApH-L=W4D|p~{>mwA*Tn!IQ-(}5%V*%^WSRfE
z;VIG;xKFK+4}A~NQM$;s6pe}`kQ(g;zPm&%c)RwM(Cctp6fVg<>Vm3z<@6Y{DRYBx
z418RJqTU|^L6nf0oBqS`bQ0NoeVWvF;o&-m#R^1+Hu>x@jvbjK8?qmtIs%PMj?m-6
z)xu{l00|@LrP2N4j2f_+Y}(n*@TgP#@}NvF@uPrH>Wny1H<VWX101yQrbC;$rR!r1
zMlIxk(GkE8YFdju61H?C8#v3P$lUlkV)^$d)0_*w6HIRV*D%U8;l@7Ll}j8eG6K}{
zm%sHtcLW9O)i?L#JB~S67T@e;&bv>ic0E*TUmd|eUm3{E+vp%~4W<+Ziq8UjN0Qcw
zY1?jQxZoAUfwZ(e<E;7rEgKzr1T{1L#Wp<Gul>lq3}9>ohZR^@4~XpvKzKwk#a*oq
zG>^!<Uo@d>^`fo#A!_#4-Fh(SaIdMAEoEZw3z{-hb9<)5EeF8tk0$nt@tVWM(X$Tc
zh%{=Y-rqUG=8X4W&X6GfNMR0BO$!b(p&li!*jstH?*S9M+as@jM$sikdSkDbRCg=s
z7QCi_v#>$#+Z5JVy`xp;k&kbWsWNf4^LUq;O~FIOT=YWn+~Pj|5v)IPiR+v+9M`zV
zVU09}qg|51<V5Aoj}Zl??zd*&0KhirEgemx>6ZF=EO(4ZZA~I~qGuzKbuAb1`wtSp
zxn$Tsn^>P`{THs_cGDEG1#+0)IQ(JN^Il6P+e2n+1HP0wr&<*VwZQHmWCQml4i?Ov
z3o(R}6D<8MPY`1O@ZAbFZ`?m-scKv(U#~Wm*2tw*1Fj-uGI~YYh8$-g20`Ih-SGXX
zsnt*9xg|>9afmo@E#Xw#vNR-F{3iGG9PQU8s_zH~ey?Ueh4sPJGeKn&(UIcEGhe7`
z8lUDaU2F~gR4yRp(~NC=n*VY>hSu2BK3#^a>j8R7uc<&3y#OG`rnt@a+2_cTi1<f+
zR)O1cT0o)8_x6AcVP8?T11he!T9=PjQ2<SG$pESW_CdvpkT1-z?Gh1Jm=eVLr*&YS
z*qc%$vG7(ArSV8BwQghdMppGJzhc|4;yj%Kp2w~C;HhA5+xGqAh^rkm&o=y~^&v)z
z{`2-EKA7^L+q34)SZ2V}&q^joTFY#5NtH-3Lp&TSq(X<xCIF#y;rqL<mTR?3KIC~}
z&$N*?yhM2@vZ;yyP&AeH>=bLI7xYnEf>o>bhi=(t;f*o-6b~iv2Z!oqkf!#m?8i{x
zQF%OEPEmCn?D$vt*E)NC=_iYCFB5}46i`Zw{2-<ACu7TSCiB^F9Oj^0?kpk+9<?IF
z_X6N3f=^j~2`V24e`WX+?nH%-;k?V9+t?HdDY%e6<~~wx`DHphW0^<n?X@d7wB@@(
zr3Zz2>@2?ffA8P__wMxXHL3&3)xpL|9O{fu1i5@{#sK-O?nGH#JD9}Z1o_9>qdL90
zVx9CcxeY5Bu@k`?lXI|+Rc!*UF3qF~wTfPpJB~F7yT21#hJu5&V$DK@p~kkrHpYMe
z>_%flMsudHS?Q5EU2A3XphsW)Zv_^somcA^=l{pvTZcvcZtdb<5oHjC8EMG@QIMgf
zLmCAE1%aVUKtvFvr33~TNl`itO6hKdp_G>HlrBkW&+~D=`~0@<@A2K|{Q0}yzvRNX
zm`^;<de*wvecubca#@7}tbd{!>?{m@oxNUi*2pf>1>JBT5qW4_CtgI@fJE7l!XNA|
znX%zt?`7M6!|_UmjgeL{Zkwd#wlEy_g}5w?l2@1TSWy7=EgJV}6J5#S$o+!V@jANu
zgw38B4v(JAPRcCd+azSh+ZjU*`>|O2oM;$maoUhNTh_h5T{xmUjFi^w=JX>^Y2A}G
z$n?1X+Addj-{ed38IJ}Ew4?PLKU!q_?y$hDx45IMM=jY&yGP|E!$ePm*?S>{uF3A(
z6)MXS?}R_Pr%h;<e=$I)GI(r%xzHMyy|(O-@PJ$^(R5pRSjbV9W}aG-?hYjk7N2bT
zHC{lrytPY=-tI!eP?WcerB(YI-5s9MC)elS=_W5r+{g$Ju@4k6^(t+yteDYAcjWEe
zFn_&&Gu={a-S_Hwo#NvH667d7s@IwndnSLa+5MHLQhDq<o{uIrjWKeD);s6flgWK&
z!?*e@+%lJM5zse$E)3Z46JsoycRRJu{Nn1I;G;@%s*xBOVv3gh-2IA*Exx77gKly&
zs~opEVYC@aCUbWv<baWQDX@aku#IR^HSvyTl40LEo|$Kb_-^z@MPdv-8L>*+)6(95
z$pQcEM{k~6&-@fen@om?*=egwC<Lw`W7q)nqX(~Dy24~J8f?1J#k~tN(moedi7p*}
z@Ghju;B0$1zC8BXOdrf!fEq#tXM&&M>g+U)&9(lu1pEEFxS2_Xfd5aS(yh@q)6#uE
z5hm2d+Dix~3qA9h6SeTI!S^gE9%IWjVB7cE=7ybx$3`QiMXQ17)#y1{JZ6R3jfY>+
z@h`!_<Gmp-IW_5$$+N}&FNndv{TL$!n-s7Cj*`nWDm#EUt=N;MtOV=<rJOn)!3>J)
zdMh1_jU`kUU(^a+{ssosT}}ieHZC~lmYFnmPlAhxg$5iJ0r%Z>Yw$;CqaGbuP?!E3
zE?&QL87DQ`qM9he5831{0^8kI&eUeFTO8i_{_@~LA0$*M{(KlX7T4Pe=cIn_;D5=I
zfBVrp{0g_n!L|cvE2iK0vELbz(rOkIYaYi&boS%XELS>YZ)`VC2R-a}Qj@@L>?WAS
zX(E##Pyb=24>-Q1vyXns06;&OJbMe|>5I(LhG1^&|4!74aOc=l!2h{MJh3RaNaczb
zs>R$nO2z`-L9iAC%UHF?KK_aC{uBHAJ@%ZJoCHQrO2F#wqfXTQC~b2v)LdB+3;S7M
z$iMoFLXF|+!W(~do-h}fXmK+R7~98@9J=De7$;L~GT!kr3MG4;?Dv5DTc7?3yEmZ>
zct*H;#b=-t;7&YPKEtq(-@fy=NAz!R_b0IF31g+Dz2Ru40^Q4h^N@f4KrtaJZB|GZ
z>?hdyfA{4x0pOB1)V1dB{b|Mj{U!bX_jbLhBWiD5O^g5|)WY>DfO+PXRe+uGJ-Efu
z@51s8n$oMBC{nDtI{>QZqKzh+@523`1Hf!*9(~|)8lp=g{Po3k;4vyFF)&sRETVeA
zu=7`eZ$I5#tvOt?1pExgcZuJ=_t2}{INm-xaPbpDfi~O#2wbBXECyFyEVa05_Dn>;
z%(!mg5*fX00-Dt~AxDApx?B{uLFYokdxLlU?xsN}TZQTX6$^I-h>xM1Uh2_*TxdpT
znSe<+Ns8NA`PFJ*l*pj@9^zR8oyXiZ7<eO`UC?q38sJ3Ve{3+CTno*ph6oEMSK~q<
zgW*Lm_sS&>$lBvO_VN`cGYyX)Rsc56r2reC$Jz#7p1gEWp<|l3rUxu!wFkF>YcHR2
zyZ{ICj!V0hL64=w@ftLkr>Bm9At4DsSYQ;*F&mH+nppzlwI%070)K^P5>Aue1Ze>J
z`MBQ%>U%M9;2ym=6~LeZP1^a!L=&JD7VYF`YYR}9cH9Io>EyEp-(LPPMCR*h^L-F+
zQ~?*V5J+dMd&_KIDGS`fZe&gc+&D6Rn;kWE;jxR%Fz^9u3A170P9t;>>CDp2fSzOr
zc#=y6+n$8M{7?wd35j9&^A2p{IEAtv1GWm<G1n8Mk9Kms+TW%@ivM=-q8BNF&9<%0
z_hAZrn~KFW6onfY`b1*^-IRsW1296zc4iRFLXL*N%;PV*`V4)_*pOOa(8#xFIj1ts
zG*mcEtaoE~F+EuY+m$DBli?|dq&;B5#;~1xyWq*;#E%Z2Iv`bNUIhb+7a?O=9cb_o
z*VYy{AtROS*L`3k{Z*8!Y|_U83|znB!WY`UU|dzij<Qx${Yt6i3+DN_9su`s@p_F;
z;5QKtSv{e~6;S)b12qC*qLrs5@KIh??<etCkRd$9A~px2T@iTCX^%a(;yrRUeDRyE
zd8^FFI=1o|G>NthEep%(M~!oN=R@ZhIuo(Q{CjZv<@BsAV*w|O=w(zsQgPw(BOq)Y
z(s_xiQ(4%lCRly~EabJ~Bs`wZ;J=LpqhGmKL8J~b1vc_labm7dOsuFx0OQ>gSc@+N
z+UDKfSt;!7=x3<ppheC60QQ*i92c(9qMkcAA7(%HU|$6&E~&f_x|>FI_2FV_bcg9=
zFe|M)7#GzI&cG<YN$A|6AoK#Zxi`^dS!b7mENY=TapmEm^5duKPnUHkbgCOX2!plp
z9Kr1p2_!XD{_k}{4y_YK<rtFxW}S$8hk2}+5@wo%kDF`kO0NhAC=DO#?f97GQtmgu
zntbz}fAE+L>|PJ0D82&QyhmTkT2E>TBwpC+9|8B!qU1~yR6G+v_m81BJ=PmNIT}jz
zkLwXmN_tnFq~2?`czltH9f|;vaoh~QJnY}Xfj2Qs3^2PFfrITWfLl7_B3akUZ4vw6
zB$P0&wt;kbeC^lO33y1{TfxBdX#JokePF$vIP_^~rGj-MJb@wV3UmL;ya>!Clu*OR
zU`2PX8QWPO)RB1RnRzZnoOX_dtmq3;C3wGm(5k*hgg5wP4J^ppA3<lN06W?K*wcR4
z^KezSB6amT39uHuQ6S`91pQ|kWOQiR62a!aLJ3XWP6em(j;H~Bh7N5NF!bf?xpBOx
zT7G*pps@-#pqE!wtwuT70xQ}3(IL<@ZVGuuDp1gA6coRAvVN<R8*fu#S%Om#`#J^e
z@^3-sxQZ7TuB;54WmQMEo0py&V8?VK$%C9ATw|se%8<$hiH|0TbG)W>{k6?wS%?g5
zleI6DT}t)(>9qi2P{-<1ki~WSP#sKMR6^aYuF4O4;J<ro@<3T&D8(+6Zd7;zgzFGf
z;41%A$FWa?CYfM(5i<)Q@OaP7SIPG?jlO0+Qhn?;GV-LN+Td&MIr=C=&;1p(g#1@k
zll>6|!2P-Zbj6SdSgYNT?W&$Xls^q+^V-~HYXuXXW1v#62slXZ$~3#Uob1nk&wut=
z4lUw#I5II_b^^^qqi?1mFbSy20XwEo#x=P@27hPLm*mQ5(X`E>W3HRNNrJX%2)|Bf
z@iEilw6_Yz7s$289*vaoSyY)FnI=F(!=C8#_Sb+H(y^m2F<}(Un*H!i5zT7g!&-9I
zPWGhOqFck4R-XU|LB-%^4a()!4<T;PeFZjv@h;a!+l=Qh80QMZNey;!+1|RX4<(_`
z^CN4w;8A(dkS<bF`^v7*Wrx5sUp!sEsT_ddvm&S)>dkFhir>e@f}FuniA@L7k!)9G
zW?Y;o&_K6$e&{w6@AOI%g+>uTNmt^$Tl?C?^6uU4EFHt9*=^8~NJ#cA8=s?AqWh8t
zjU~rl*7|tR=b@-Wr`l>qrqs|z_}q6Niu+MfIuG9;1$)lWObpJS%#kF5?sTY#V~fjQ
zs?uYw=CR2trDLEXr#sBtr(YF5<GLs>{K4#rA;?ofXnSx+Q29@rU6K;UNM?*AFHJ(F
zF`bVCar`3Dz^usvaFR}t*#E5Rge<PPOAFG8MG(ssQNz`<9FIRwGdJ*}f|=fexq1?>
zPQC@)_<F{GA{t@ZCULkjR6#Ob`Y=P67;_JM3UTX&gcs`k4{*{bH`BX4qR@OPNa6mO
z0Po!B^$wun4|&x4`Lsz67!q-r$OpaP30B8Q;2|^+vk1(766m%bmlDEO%3)lAHNOr;
z9tz-(!s}QJ+=|YjOUe=q$$L;!yg}{V3LRZ5{)kh<(nK5)>lSvT`FL-cmeL~!gAX}t
z_NMi6fe*0J0kEBpn$ZR)_m<BQ7(*DFH;3t31S4y0*j_xmfB*cN1C!Ho3Ki5-EqD+m
zK<K3&l7#)I7l57_bf1r~S}j6|*B(pi$t}`6S*Bl9?DF>khErH|`FlEZs1Vcf(fp0h
z>GKD|*3mTqF#j^B&w_ekNr9Obh}!*@>vhbIJa+`*&jdC)5jfHaeZRZ=egvc}H@?UR
zT=Z{!=~Ou3d1!{9kS3MAgUwxV17$EFSJtzgVT}z?J;cgU!AuqnL6hRK`7H+5X3_C*
zcx`yI<`gjTH=n}_MZ3M{J3!4`z`Krlrq?t25&mRqI9B9c4nm;kM!tIq$mr%%%95=<
z4S-JBt3&n;W+UJdTr=>A*pzzps}B+9rz3cv`q?J>CePt!kR?PDgFFU7zDKMjgti!6
z)f|~~ZQN7-;d5ih|5NaZocJmb=(*CSsj5@L(iU2KJ*X9t%85>Lhx=egLbM^C<fOun
z>2j5iDeb7jHr{<P=_aRBC)me!Ju1Qv2*gPCgC87$u`R7WYNN%S*qp#jJyyZhAEFcc
zDr;fE>=@t|&6o(%t#T4*L|32g!AcLO2(>d5ZMGLxy+lgshT%+gAGoRVX{-bHJ~NvV
zaK4I{5uD?1w{tq6wmXER-`A~&i^Mt@OD_LtOMhUB-GPq-=X_U>ml-1ph%HY5-gPFx
z4Uz$BuYH}sv`M8&tVtSwZ(AcNzD-wggJFvf&F5?)CMOhjGx5M8OLj$hygb^K&*PGn
zXIuQDTl*Wpb7}&cy;8Unwbip&LNlYEnE$cuI~wp#-qD)5jw%{m2L%qT@<AHMpOT9V
zslUS8!9;?{GrnjwZ6oh91+@Us#OKCp!I_Qn&jg{y&Z8fK{;20GN;KjJqrfom6)zxX
zkmdgB{gDgU=R<jD3)A@q58(c@W*QP8*2h6W;hsld&&1Q;#ww_34$VZdKwNN@Igb%q
z1QGdDRO%p82dhk?7~>P%Lu#TytddbE`i+K>9o?y$mCtNn$k{cRV<8jTW4+u44=`?J
zUd|4~-fPLetd6~i-Zn=hJeACQ_SeIM?M3Z`vy@pVXxzlmIZ~63heqCsPaE+rnn+P0
zcX1ENHOtVC6wufQgkOe?1L!oXJ=)euv{w2py)yz{sfe%<b-@SKw2JD?dGK&A0r7)_
zhzTxt=f*l{+!nm{E_&~*X9<P9PmcZz5xk5}ug6{y&%=`$Us{*2-W-L6Ww6UqXe=u?
zgeh6e-oij)6)2b2mW?Dizxh}QrL8vs==I&uU=>t^d#BrY&x7c%B>AMSiDa7mi}BSI
z#7s`|E<R`kkq6q+tV$X#KLHKHDcnmZ$HO;|IK7NP`@XxuyUX}<01)ZhhYiCGcQ5o(
z_D{mkFl{&RA_tI1dZ519Ozjn<xQ}=W(#@QGs{J=4rs5^%X=T1vLKx>15+~JCD<tpH
z9m-j`Qt(JTd;=aneG@Zvj$k>sa_qs#x0CMZnutTV2Z2&K_Oz>Jmmu{rTy-nhG&p??
zL9RxL-EO^e&Qio$D|n!YaCEfiO5Uivms!<oeI{H0hb3<gchvO&hNDS>@?<_6=0Rzp
zn>7fzx}^wpmCsQG`8x8#77kO4?(m1kb}!~Bj!#X|nEe2OETQJ0FSn!^rDHK_M*EuH
z)zugbZ@7g{&eoZJ_joIKKK6DZ&#v)tp0I<dK@Y9P)#T3vtBEJiUc3}OZ+ewUx1mRH
z)K-Rm8J-rcDctI?w0pOSw_&I4R*cfiv(0cV=3w&`mKTD<{5Y36u}8uR;<&gi*ftx{
z%;iq%1L~P8TbJ+?s<x#jribbWLyB7-X7pSEB+p(;!TDG-Pm!^Hy7il-9fO_6re-R4
zOKBq-8NGR=7|#m({<Gup_u1l1gk%{Py9u(wa82funjYxF$f4xPj5pFI_HAfk;o6}$
z^xP=&B`2`FquX;c#^Ud>ezWdk6rY3gLeZF*H+K4t>AJHpJ2@56(SCB{{Y+T69)f5;
zB*kQ-7%R8HM6pEE+>&VXbqWmfd4RA%wz39PUr@RuhjkA`zvlX6#g8eDXKJ9A!|DXW
z^X{q6dbE()SOw+9sbk`s&K@rLArn}oZSOTqyCQtvF1>H^gx%+@hYL|4&#{`^t+oqI
z%;UX>p(=W|;X;I*WXCam!QK~@eHzQWntdMmloj_x{iDRKO2SjRQgm7}=%?3kAtngv
z4yJ~RGshK|q_KBKfyuHQEFcK3v2DdSTU?Y0*NI@zwUN1h4+AfsE!QLnsDA>^A`a<K
zcN2iT<jnaG-uXzq6AEMM53ps8SZ(`}TmwUk<O-wPDPg#5S6Ud6yd2t`U3#HMfI8;Y
z$2qrh?+s4;7&(~r$atp<-P`(v0Jvju6(`0dRttT5**VD`InIXTu#hsew&5!ZhjURP
z>Di_CbTPKYhP2fpa9n&FrM%1q$L28+Yf(K%O5_O&w`7XIt@xCnFBApj;bNQhhL!Dq
zu;*rv<VfGs?MeoAO9@O_>5*RqN|5f?q-)WbuJ^E@{Ntvt@V_XH&9Y%u7&&b5L9y~0
z$@j+n{njmJS@n=Ii*33jKV%h$#9kc<1@alY^r-eXT3S{I(@~BQSs1~{74o-zOqdMo
z&giR4ZB{KpB4hqS;-x>}yp-8Z><3({q2&sdchbWyT2#SbHj^jaMaFSXH^zA0@=NVw
zG8ko@Hp5QGn1(JoOQ0^y2~DRyXTl#pZfb`=Bj3PNM6Au((-87JP8O>8a^(Y}|9wxz
zR@*}fks+=Tr^I>I#eV9;=9YxFHo_uP>*5y|&YX(qKl`WL>>0lKUwAMzxTBv^3CdOy
z^=$AIj{=u88}0c_VUM>YR$Rhkd1O8FE@cVqxR(aWrr8p`YzpO6t<$#XGs~qyqt~i6
zn?>4X#?I(d3?ev-b*Qc;_i{O2%Go_Iqe0w7zh%csk3E~cvSvGw(5T!lP0hsN=WAS-
z1Iv^??w<O~<WfVx499te70uE|Fiivxt=G{;NCvLr9tDg^1*T*YG6Py3u+rF&&RG?7
z^EIJ>jQ9Gli5GogDxF~3c~o>CxlK`^Lk0?A)U{skqplQgPc1lIXX_n1;+*T}i&Y(d
zUJL)k*iWk{DR$FfQ9ZR*0Oeqg7?k|Z{Fgj8S=1?A0}4&G!Ud93M)s0qG7~bALU&vM
z$`Y8D73pVwT7RJ3J`qNcx#-Nh)^9mtY+MK<@|8n-XPLk2l$e&n#K|Ctow{@2reC7r
z9sXuj5|&yw7~vO`9>}3@vcp-x9wKWuy&%2r<XsX7k9>0b@)YAS?<gsyaW>EQc-!?c
z;=400f=y#;dVR`(f$FOK?pghy)TfqKSep_gLv0e$3SmNZ%|@U9(EQ=h@ql$B0&r^8
zj15LOUkuo4>0zwCdqy+T-fyGJkZ7@Bn;r)ENuDb0*aOl&*m9f3atB}L{=)r_k46%M
zomabQItFh<&5^awQmY8Zt?v}#%fEV-6KLvm|Nj66|D!+#z;PLO!wxFP*HntXvZ_9k
z$!6Hr+Z7DizAqSDmc8o4*q$d6jVmyAdav8WK(sZ~Zi6QMoFmpc0|4sJYZP`o{j?Pt
zp&%0$u2cpof>&Zc+BC#7$2h(x+jC8xUIGl|G86PGd$xLc@`DgQ{jpw4l--mXiI!Vg
zx^+$uoOG@rz=*ZdkYnAlSR9~!1-h+yRlVyvqvSJBNC!fpS~hc}&`ou8_~pO`ng5JI
zsZF`bDelUbjr{x`K5)(H0wp=U7u}DiXx<i*`IN7m-F)p!{P1!R`^{15X=!r7cc;(j
z*-gK7Xyqt<1mr5Kfn|1Hw;9C0zK}Vq_`%L51C?7o7e}Vq-Mv4boIDmQ$n;>?n{utq
zj*`}-ip(T3jeW@rrH>M*B*|`~jZ#xb%eQeRWj42ZG!5mMMCm7~JD2&e)RL4omoa++
z@&E{H#Yo%Foc!+%M-{a6(E&n2rSW-_kz04go2$zrh?QK+e-O|4P&pT+_#NKc99V{%
z#H?JQj7|tvmm&Na)@Yp5iz)%dNuctoO|+rji2v;P%FBv!i87EPP8;Zv^f+g(W$4bb
zWPM=KQ=Kso#_WYC**+$=*J~uc*ID+tX;sWd=c-TU^hG#lkc?2&4_i8hDCJi5<C3vU
z&Oq>IE7KVxI%r?iQI%$RD!`So!GR5BoCZ%c7aT9EC4OV21IGFlv`*wX#6IvU5}ixR
zmQIvVbZ?WBMkjZOZMAKK@orTb7eh?HHC*6d`)DU>aCWp}W>?f8_!DPU>rA_Dj<)+o
zqdNMOGP1iPy|VAh2g|<y9l#hAN9b3C6)O@$k4-IA?xxcw=6$Cp+ztd7yg@j>!JmSd
zWEf!{Ht;np`~seYEISb@PtpKGoJpx&c5$swC$Oy|yx0ctEVV<Q@K7N8B&VOvZ921)
zI8r(g=o*nYKV?H*Ezasu1qBxyDrR=KOL0_<!@f*J#rO=XK41{CWl?FUrm6|;0nq7s
zQ2#-~(MJ*)a<QTc=j&k?G*<ev#ni51>ESPtyfQ_nzuUfwAKyy)o{|TT!JV<B@CCCQ
z9NF|F1YXL!(B#}I+$+_-JY1Gh;Je3~70xnkN$gw9o};k$g=oqMt=(o42Kj;#sl{(;
z9=n5aU1<Ogd~^>4GQs+nYD!ZKju$D;sF(RF8!PkrWj3MSdW+(+^LFFpJZATFIdr#9
zC~Lm?k@d5xqc;?%E|uGfpghbsu1wd0No%ViGqc3ju)6iHpI^1gSeL@i8iLtXD0BJ`
zFLJH#%Lwu4O&NhuBw|eyB_>*5G59R?-48$hMjc$8)9N`XCY<c152MxVXV{>5k1|eG
zjqnYCBPTYdrAn6T(;63Z`yDPp#k30R6r-3?(aFJbdXI`|HeHf?mF~>iodyjVk5DI(
zT<W~n^+KQBuZ`}d0S+C%A9=4>hd5{!x(25xbACM9bR8vDs^Boq?ooK2M@CG1{VE$N
zXY=bhN-a1sFINRHFl|v9u|dfBTP^#;KI&S(jWbFV$rkO*ntS)^_)XEO(~N$sL1Mla
zs-0~q;@$lzO*~Vd+6Qr>*RB7wcAnvjFvB&o1M}`{W&v77BbX4VHwtjL#@_eph3q4F
zIj{8No2Ct%l&Ja_MU?Kps8rNyG`7p@j5O9prPBmudxQosW+Z1$+eE)}_#yD3wOJxI
zkYub)MxXMDZ~1opstJmlJbURns?p3jScdHoa4Di^EN9;qI%v=aZMSD!O0J*Qa5@|i
z>M|!7@GAR+`mrY#XPmL-SQ)9HM><1~h%U4M<2cPjE1TgbWJnW~wEJBQnyZr91B`_^
z1cA(S^kw&B|BJXybp&2YR5WpaPMIYb<iimvvXl|^y;MZ#UQ|T101Z>0=9g{ZU8t%x
z3_6@Jz;riGORin3S5tE`9JzL!CmpE|Q@>g!-CWI^^T-YvYGrhwf^K-|=Ebiv+;Ftj
zIg#FxVZRi9a4y_-XS3<#EX|`A`fTd{Z%&Hsg~zc)N}~048l%~Er79!D%OOMAi$*gS
zExY_j%1d*4a&RIxp)`Uxne0w;%6>3PMicRNM#>7Il~Z;-`DRX_5^Gzu6UZAkO#uwq
zH1?+G`5OrZ(b$0^4x<oKvD_F7-VQ&<)|HBwdCHuHMWuwt<V)Lkr`h$%15&?ge)%p;
z!8E~S@ODwqbNWM!h0UT`O{a1sNm<=&InfrcL#A@XMt-(>I>4y`!=&IwM2UEpebL(B
z4b3?+$JQ@|ZtLNaxiBNop}%kA{qwiALd@`}hXq~p{xmk9-ezeIkx<5IzvxoD-@kM)
zMZXM~og=dg_jyX<T{9|Qiw@!)O5*DB1Y(v&^(KkYk=Xm6EXrtCO3OMz#>j~M55F?$
z<p`YM@aCSmF7JGP6F628NrEwhg@Q>G4EK%Y$}V%_F6krIETQJ`dsu0e9NP4nroLm}
zTkWcb`Mk+{jOQL+Wn$mj2e>9r5J`QArE5lKloL4x@gZ%drzG0HHXrb~F;o%!D7eO}
zmC{a^9WdL5f9p~ZPO5=<>Y!i{41^5y-uw>0DyN$`DcJ|0TVzi`!W(rFzRk39DT9QW
zn`qhljH5(9&KBbfj)vu;)m-;kcI`*v@b$c-nB@#DT%D~);<9}JozsHg+4IGo2t20o
zb6QQO$8kck!v5E2@1Ktou+~9DT_*^37E7T0V4J5qW5bqX657v6YuA8~xj5;}n`;?8
zXB!q!{E;bZu%zmOrtvMopb^AKv0Kf+;=9Xw=LqU?R&`q``bM-D*sB`>p>bsWBE}bN
zzF8&8*~W-@(6MXtz_{L_uIRhcvN!lkE}V{cRAKr99jo5Up>TDjS^=+`>9uX1*8Pw_
zvKP8!@TTyJV?YmmIPh$}wK*b(dkb#?O+oBbNt=jja^fRFztnw{-zY-ytQSGdWUd2%
zz98GjB9qf1L5HvfFjI6OilZnVJe<0?5k?boS*@w@3bQz7N81!X->EAlXmI=m*}9?I
zzym)+u?)j@!3!8aoG-<O%(T<y<za5F08jZx5`BBO^*~e8Ec7GYM0UyW2bZ4s@!kS`
z6P9!N_|r88FaiXNno5?WUuLU(O}sOhnTA1mrevSL+-6ntSSWcJMi2i~3(ME8lstOu
z+N$e+nf5ZV3npm61fVQjld@h?V$aKp^kLDR<?W9^7f!A)vSrSV9;Dtr%L>=dD+mq#
z+{aCd_sd`%xG~!aVA!xjoij>fisnKQGbs)OD1R#^_=O9QGEDK7NZ!}I0IS6Ld7SET
z>SxIhd@kvlfs-1Z1Zzw)3!v@Vqrn-DVgYZ?5wMvGHtA!k^W?LJu3>@>#55TP=}`{6
zxzwnRH%Su!g$@La)U4UUcfm7|ddjR3XOjulNrpgQnsr{VT8>@m^OR4S(d~v0jJTsY
z03vN1LXAtrni)}2dta^nww2+b4!tfSKrm<rs4_n+{(sp;?iMXe_5a(7KvUEgFF<D~
z5N9PlNNF&+!GuSl1xz;HSH%QT!Ak3}#1-VW0YNNu6ir^xAVlr^9Qx!xrbxUNm`!S{
z=k@H{`xMA}L!L0ct;HksLjQ$SEqj6riyWErn|>)LxOu0^Z}0ijy6$mxRdEga4qd&t
zB*fZ7JF};=ky^XL0W4CR&nrJWS1;?keiZ(Q_K2hCPiAF6A$T>GBfyNfABWnKI6RTH
z(GN@)QMw+-%Z6N<0O_BmTX#`tZFH8sZ&sZIig6AOWxEUgGkL)RcPg2@z#DY#;Bu8u
z&}>&__)C0GQs--KmRAXu5^-BIjDJAAw6Fk(F9cCh8riB)<5PPjfIJ1f%S^L^uMA4w
zSh|d@53fTHn)JVdX<ErM3XG$p<}EkNDUI)HGyrFuu{B^Ro!A)Ci&#6O`@G-io>_k=
zOSA5*w>#5t${3HqWJSgKqV%5xc`d7)a&sFlz`W7@c%p7Fu}C49T{e-HQ0AN)8sDH@
z5)EcnvIp_ADfO)iN?*)!Q1gep$nTzQxWBC=dll=1D22xI#%i%g^&Qm*)&vgj0bsw~
zvA5&leOHE+3@iLphJ&tvK#($*bPClvqLXGWjb36NUXf#DXg|)R^;-Y<<3wWBPsE1!
zXPC1-=mG=D?x9LYpEQCQo@$VnXd@Vrtm(e)Hqq)~ywj)EY07IVu$mSLq=1z^L~Di!
z0S=J6G=hm=>Wq6?XNT!`f;07psiZ$VoRmA21>>!Olpf{cQS)rf_IM43Shac3FMP!R
zYFB?XV0LGAH_j^ObEj2H$mBT*W?|3h{RIczU5l2_45q9(qph8%wePMpm9=g~%`<qs
zut0iONM^N&8H1J4LGcX^=}<nrS*uFBq?PC~%ra{fY*X^!^Pb<N)8^OD`*Km;PafJo
z1%;zsBo4T7u>&8fDyn+-K-Y7c2A}IeuO%)ywH>E3{hD@E!Eo*jQAzJ<pFx0TmJCAZ
zZ&~8}4lo#R=J1df9F{FzD(Z1HPsVfHdjl|N!KRaZ?(_wDVUet}k6%d+5Vpw%Z~P(z
z`x}-HcPI?A^O{WaZZ{-lFhN}R`-X&ud+*z$R#M?PDXL&FHXA->U$5-3($qPgMiWad
zMQWxzvbXQZf}GhJU~HrB)1X4K*`w|ZWr{6%w6xJH+(n&f9b5#Px*C>B9Xq3jHc&#M
z6~|bcPZ7|*kWPu|VXUzwiN`74J13<np?N~odTa4pJqgzhfs(OLGSu4G)9UDGC1g>{
z&c&cnHhd>G%<28GgE&B9Qb)(%vdYSrchx;tr+n^>{5h$M@(jE^=&veME!>9n`Ec>C
zLelP0xTk98kh14a89oKW5C^bVF)D1b57=a7EO&=2Q~}k(A?~`R+BxG}PeJvkj!sdU
z;uLI@c0Cua3CLnI2E)o_wMs~rPdk^QdNzxZD(#kb$oWqM!RbI_Al=(vqVLCe)i3o8
z?G4M9k2iVMkg_Ih6oY)7yzLO3#;=U=`}0jSOi&F=pavUhk<lOX_pA8HQQx)w$x-i|
z+pCvWvpL;s4I{s3ovH!LbYPRUTs&CLRW+vL7`O$)JTW9eD4yd{IQQkwFt9GN|M-}k
zU?YIPb@a*t0n$bpsi}y(qlolZM8+s0GurF|UbFbgrIF#iqt(DhJ@HDj){tS3N=RS-
zGLVUs6@#Xm+|GlAKRV}q&xQz@ax(&nx|wP_5-)u|TDmHxIW#XLZp2QMvu}@JB;62l
zUSw;KUIYzKk8_AyK1l2MjaPR{qTzHFB@xV7a%u2w{<kDjwrlpr)jWP!Zl0Wr&R96<
zF%W;28)7s%$?zuI@>&WIcg?P%!cyViX9WmJmf{6O?ijfv4bu~zD4wHyQ=a^a@_=vE
z`{>TcVE&?@ygU4N9Ox2>!k?V;)=xgs2>n>Z(^SoyM<Uwvj@Q9zG`pwdGf|sss2_~&
zc$UlSvo+>KYK+h33l~+TOXwG^51o-V#q661&hRaJZ;np;CBCfiIVW<TDKcED*a?{P
zeXw5+=Q?Yv%gjyBr}FJ6)oa5Ywro}I(Va#;PxU>P0As}c#ov`8AR`l@;27X}k*QJs
z80TA};bDF+m%-}L%!@M>4)yx&z3!i9;qLFcve=w0J!Ti};3^S7-qA)qUgWy;(n&^~
zx1>8vS6pGa!%8Re{VUTV#2iO3-)!M43ziw7W|_+g(o_jRP9QxPCoMF_6<gtPeUv2~
zzCG(ozZH|$Hnc!lWHu)vb`_G<V8>jVH2DfZ4`J4`kv5{HLoaBa!T!>L0~%iix!JZw
zGbaJAK;TQZNb0czZ|5V8v?<!<ewTree6ON;GTUW&VgLFGE0!)L!1-{yHBk0PexCT{
zXJ}4T{{1hf-_ldywDB#xpZ?S%Xmb*ExA<)Vs=#HUCl%o=6tngE+RWGRkuWdV)FnPY
zv%;z{m!FjXLo63gM0s4{j=xL5ol;ukIuyiB7tZlM@so=qv-z%Lpe9ISmXb%T^>Psu
zsO@+xvTU$M_>+PKl`+OP=|{IbBBwT6xl|!BKx>$5P~MBiCwLo7k|yCbm*6r;rY!%w
zCsDEfjdd6!d<24M29^r9XS#VlI?@B5dsNi`uqpOX=BzU^`0%KW&9@@W@N_0kX_&S@
zsF4v;==GLpB%07Ui}{*}l+YQz28q*tc@5F%E(Q0lAUG#`Vb>`hlDCPxjF-HLZ~-zT
z;zg;GK34DD@RdP6$C-y#_99`wL%0DyCh{g>dgp}9(oVIVh&l5TD{kZi$BH3!r0yZ2
zI-VH7?=W7LWtBGamdFw@3%^;pm)VR!FpobVju1o<rm^RgF%hvKth>12YhM`P-HbKt
z&BPd$ufA4GXB&{l)_kQCyqpcjMXPFzLm#Rc9pr>G&td|Cje{$(85uOP*LzPN?*I;;
z#W}5s@|?2tSTkx>M<snGTR)QN=r@6tY@i?mRWeEwL-&?*lho!7-$10M4jV*|B}AB)
zo}Nv~G2%u)FxNMj2!V=0&MhfwDqUl$wdAcqn8L|})7$F!+3mPGXd_7pe>&ZHs|;gS
z)qqx^nI7_nb7!5*N+=~;vP{YhlJ=k3D#Z6DHzk@@z}yKwF1m|(v@MP|b9VYf^Hl!+
z7X^mI<_k@qAkIyQg?r|5^KicAaXIk%IP7$v<KvJs*Bc12i{|m!Cv?1+V{3ooeon(y
zDcRSJge%u?C{*&~ent!&HIHIS<VP%4<S(Ol5l!SRzO0G+dX@PVHVgozuG*_bA4+1|
zKc9AeljK6ttspe8CTwty071tz{xwW!p}wtg!D3WZZ1o=-goPoojc+8<Dx<klc-+x+
zg`r(o?sFE+Eb9zl!hX37q_Ni*<4gY3U3z`x6#ie>GW1u4uYNH9MqODGd<uouXFKnm
zf<Dah%ZwEm1z^+8!-y15sqD}O<)=y#8()%!;q7PYq`g-s#t0r=pg@1+%&so3C{dx0
zo2eHeV<Kw_{wvULRZa4fr~T^_IK@4gNV;a-NYZ?CAD*V`4WiawaYuZ6lxMh2nXGlM
zx~sLLQc|Y0DE{eoqD2(FsddKr>|t^{Mjz=aZx=#4b$hYd<LY7DGZ8ajUf~Ic?ObCl
z0di3`m5j90k7dbajAf4P#g7(E8x)ddE^8LAX@-2PP)&^RA?i*Ur($On-DwlT%hVKm
zAX7B^1~cb0;!d(y>nlVpNTKuKO#%TXaO!|JNVqIO>`Tp1|AmOm`}$3CAKwRZ=oUw7
zI?8)6wJ`mF6|!>eJaZImDtdUsy0PH)#|n-^I@1PQS<kQ3NkxR4GLt8~s;5Jo`?f`g
z(l31Z=zlRzTWfu>sZ2NBT5Ag@ahtHuiE=HokVT71E6ll2e^z#ep%ZWTwB?ui1?JwX
zJ1zEweJ+OMtx;|dNWq%ZSbG!JGx1r}gngL9+d?{X{;mY|4;Oz%&v$gpaAt-XmZGK3
zzABR5E13g*bk2zMx))E)#NN<nknCo59?#7`aHM@dm+tD7=c>H5Q-9o6=s!<n9q@!(
z^-V}!*oL1wUv9>3kfvqt;yaBgMlQpMHyVZ<o&1k8adk_=P5E^P_jaD{idvJ7<@8K-
z@Wr?CjCPLEuUGKws{4P5QNK6jm`U3wG!|WZlN9Jmx*L>#(MJDG|FUfoKd&I*3{wLh
z`Z!Z(IKD2(W&_adsj+{(B}2_K?S9Z1O7sqVaQ8A`fEJMw6JPCF#rW1gueXB2dRqqI
z(RF0-Tb2JeOzK)z#2&zb6CfS$O6Ue)&30%$muF+Y0<rjw2J!PgJZ+|PP5}f<9{neu
zSr#FtqUmrR^Csw1y`})dv*K^Ag%JH;3L-X;g;V}bD)~=X<{wRp^8vL*mOTZjbr3JI
zdWLSDsw3Or=eIQw(z>nWF9OB>!<3+SL%p*z{!gGLG~jWL<c;1GWO||iIK$Xlz?+K&
zLgb?DPt0K{pz>y^{$9P_o>OcWblVH+dJ+Uv1~mh0)(VhC(z}6=W%4?tz}pfBb9P-I
zHgZAh!YSsB6zH-EB>^Q0KjVH7>z@c2pz9+4vUYp1r*OgijUR;~V6U{^yke~by*7X4
zzgjcU6sUSoKF3{Aj*A#S=r1+j*aGYsbkY-WW5IuUgSteN4wUSRKr`9#9!YgMZ_F%O
zzkq1N{dN1tx2)Zu9C@_{xQGoBZ-P%xO#-Ul;!(;4Q4%0MA`JqN`g4tmD9!(j)_x5i
z3=J>eUU(NK5L&gQk*f%CgPaO;_R75*=?H_VyrT`CvO#R*=ah)JdWcNM0TW<&d=lK)
zWs&SrvTqv)xO3frM>w=zMgT3I+p-{CxN4HEYyM|o_HPziOdh0n8T?Q>`8VtP&kKbc
z0oGFjw%8{2k6Fs!{aqs*q}(#j9QXNya_b*!@%L+}m4!k-Q$HZe|NRI4&%gC1R01mr
zll_#4^S@b1|Nfvz!=ZPKUXIfK&))IB{DmthNc>r{#8LLA!|ESO3K8aSVEd+l9vOQ(
zZ!0QMZc_i3=U1BnJ-82h%6fmg3~qNkKpT530^Iu4?vm}#0mZC15JX$<AzT$Y34ii(
zKV;@8l8A5LIRJo1yq_k-Rn>9$BG&`j83fs-S_ZBx$8&-oZ-F^rR!;#;Q*e_gzaQf1
zMWIce7uB7mNr8~m`C0B$g!c%6)W+);Acku_t*04(p(4I>Z^Vc^^zdIVoa!n>8h#bX
z8~?Ar{k>D_{jwHZY9bgyJ-OYhu9^z*a2E?&Ih7!c+jV%A3TX>PI$vTD)-kUi$CiM~
zWC_t^@iOO=y(A#)MhPOb_RLFsFE1(g095!Mh@!RpJ^9Cxg$|IIZ%shz&u#EQ(I6Pq
zi9|6yqWR34B=qBGs|O;_Q5d*P-EH%`_1&iy?DhDc-BjetOxTU2Ie?%UTsBSk<@zN_
zz!q3R>b?8pTHN;e>n#tyFRaH240zmbH9X<X1|UrJ%@!fi%g^rE)?X034N-%K!RGGP
z9p8+)zFxIu&K}NH^-Y@Mgrm4u4UD1Cd>D@dKuQGQREKQ?{-2;V@bDPIO+G4Yk1AuB
zngVe%hJ&~Ra5=4KsIp1XI(nc_*7Nylazdl;yGW9r)eGM1p5H4iqxGFWOQVNpKqMTF
z#{_HqvWym}!AcdEv1<6`CV$&d-s2bd{z-se;h0x3X^vz!se^cw9)Nj#8w@h^VsOqW
zi=}UC{01afF*apW{wqsV(RtXjnZ~SkpfdY>47@GUAfbys=$S400wNR?shK4K-<b0z
zV3b*xT*tont$`4H1ERqjCdQinDaJ`&0l{HK00ZO=oVvq@5dW?pz=ff_fsHjl4aR)D
zVsT@6;p9uzBA|6zL1;5Mai}@VHuq{~PQ^87_LXIRBZU$G0VmvIUf%sRI{jyew<(^+
z)Fdd1k|E}p19-9NUchK%u{hVm8e(eqKuBJ;55L7Mkfd1Fgw8-Go|B($^&B`<6@iXN
zb*5`6GdF3*iuS?q{>a2hlx=p9)p&Z+(`%77E{Kna9}uV1;OSzj`vB0Ny!!Omp0OHC
zs1*BYbpQa>n_KOIhEK<q4WfT?+xY=M;0}>T`-bCYxe7h}PPiA|4&;iJ2C)zq0fE3H
zISW<riOI6CT|fykV@aNI&**Q3%7N_jo^yQ3i_Ubn!I{~+?&A7dMYu)CHX!(0c-dO@
z_i?>kB~-qwiZkf--&VbG?0{q8$%$x{+@JckHxrf)3bO0Ey6z(*<C;k}j|{is22z0K
zOcjf7lbrrAGC5lB$t_NTeO&Pt)cJ%I(km(zk+aQU69^b<kYt8pX9UI^dN5SmbMcQm
ztP^I?-YN8pG-Jdmju5x;0p!_Yk@+e`xt3#JL1Hs}8^DyQ02ma7lme;{eXwxkVU}pg
zHInbf)Zc(EplyIpzKenSs%^CLKG^2QoHS~$^{YT~^XZtXbn#bCkI@kVgxQm&HriiT
zz;FA4*IP{p%Suv5jh6)W=Jj^agwQVY%OJArEUnqehD2H4-;@j(xS1h(p$Zx0egLtE
z=Z_#&QVxhd?kiw}-acIa=moU^4)xSM85sg-pWJ*3C^H3c!nz6YUK8ZZu)x0vbR;ew
z0+rJ%D78mD9R?ib){#OXqHsEC4iH*aS?j^ZONL*D2--tW8KjF`=VH_C3<aIxK7H-=
z!2Bsr0tC}^N<()Zq*%Q^58mfFG3G9ypYV)m9n|wLFc1nwuY;`Z7N~bkw#>>oZOuk$
zEAA3c%SN1iJZjYAWLq386f#qWl&2B_l1r4LT%L|UNcuR5#^H41K*6DXJ<zsDxQn>s
z#*d?@^KrS^Th#eu5XD=MGQ!)zSa#wW2i|Q=EQ~4hRM2y%J@!*;L^HA1gc*#Kkk(BJ
zyX%A>*-+joarjk0zxw0a2}rSG{mP1%^$cG1gTN32>5=Ak<&q#O<W~vBaRxweM`4dP
zqdz9V9R&zlw#qv_IEi!?lVO60obFPD`m|=;78La2A6xHJ8>3b?LD+IT7+d!o*t|rd
z_=|MW6cSVB>))7D`6VF!c0hGSNwU$(qA|}1c@Pu_(20Qhb^zKEF*RF>>z7dj$OoDJ
z5Fs|m9}Z*cK5XRjyepo#2DDQs*6GP6n#T#mH;%rch*#|Ix=Uu2wQ^Q|wIUIkeD<h1
zcpDLlOgL;3tO@6cg%aVR1JIe?+)ERzevcq`l^rbxh*B!BO+yk9v<kubkAa${qn|84
zX%9p=zG{#|J{%p%20X`8N<q$}b_x_JNG8lGTYADvIqg1o$g^L&>vM^4#1765-x8v8
zh)iRd<G5OJ;td8sqCG3L@Y=V4b+w-1#I~eGfd=^zdjs(%(V_d6Hn+d12|gHcc#$q@
zHSd9fJ`<}oi#&5qala0G5TezsvbynrW%U9P7J!}36y-=C+3f&w>EXWAXoI-CKv&@Y
zc7IAgmOwTWy(zkqSKkeKA4zg<u+6*$1w-cBK(ObrTbSG~Ej#=jq_f)N25v;^=$Bq-
z<nS1Tm`6&s_&j@c1EqsQcuzr%5ipf%H1DalxzFZ05I2@vLt2+M3ZnO_7~cl(zYumj
zHu>+h8k&6l0vJ4nqA9IjrPjLSxruT8lI{ib;B9C|O8x0YGkV+IWW>nHp+w219u}1W
z(PfaW_;5NIlrbUJ&_TA=!}BgX^`Li+&H^b?pUm;DOxZnJ$KJcgq4tb=X%xnd5Q4)V
z2uJReU&(q$Zk!JodcMNR$R|toNhC;}Yq+QS2u{PGJQGb!agD?PMx%@g^H2^w&5l#k
zL9D$vOjvIUF!(xUapO>%(;h}B+A$5<8rtrYrx~a=kH@Ylz{vTQ0d`*w`=b4ETZT*r
z6IsqQI!b~3Yt=dF^pv?^^bY_YKXR=ZAXp@tnl6PcKa(p{<&<U}<oXOhAxEM)q*HV-
zjKveK#DM$|&5VhV!_FS)$y@@iYBmxK6qgSZmYV|dcw=UW>mDT+r339|Sb$%xe7W>_
zZA@`9Qx?gt($g^UT~=I+nH?Z~lKlW0io-;bFUC@C7nG9a&_<|*Qr|Oc#w*k+hqNe_
z&R^pwvyx<xuj%bimQMeYC;$78GAws=Xd7BaK_wZ%f#sj51e?oDunA3PMJ@nD^^Y2}
zhNg{nbo6@t(eINRrLrQ8#a|SKfph*5b@o$+L$vb@ZU6VYP=B+!he9B6D&((+OQdON
z=@&o!(0Kxs95EaNzya^cB2c8dCfhsS*k&Tp(s$JG;h=zl40txQ0LxO|NTG|W%%W4o
zA7dqO3_PQ>8Ej=RI$UxtKOr(t5$h<G)yq$ib<re*8{<2qXt=d+;MG*jwMwWIi^erP
zG|YS^r-a(~d(Bz7k!YC~d!SbWVy1bkFAxpo7L3eFT%9?cQ)7n=*64&RPxl?NJJED{
z#(w18B!IernQi7x0NRHk{!J*Ym9WXmUvmNVvbpobY;l=D)5U3UG$v(xO+PHxfFbUS
zob1i;<-$){3`}O{u!l{}#oA4of~O$Dr<Y3e%ft2FPymI{UM8RYfBBRC7U%v84fuqw
z2v}4*t!oOp#)O&o7q@#<J8hopCgw8JR!#ZS6vV6p?tJTr=7#-w=O#$9se6vV&ey$1
zS<uXQ%7_ySLGKpcl$_oO4o7zIY~KMhM|olRJHN;lvi`HAHr52^jKA>u`F%pcAJy0t
z`T4>YY}XF~3E{q-mHo+w<S@{g96hlW4~Caqf)lF%*MO;~75WdZW?4HQ-wBHbRsxJr
zlOSXS4A1nDt+6Gb(`2A&Mtq-svIjMmX-$J`cerh$>V3>+Trv}(s~IBSi>}#D-QMr?
zs_A3#2t{5S1EY_7z^4c2%NTxpMvxL~FXO5`+b_}Mw|aCz!H5+nm4!17{|3(T73UiS
zOE7|ZajO!H6R;dgvyC3*^(21jRHkSRPL4ZRBP-BIv|i^@Wu$`Cp#|h0ZMk&mbE;S6
z=;dXE21A};T8A-zZ9p9*kiyzDKT^;Y+Iz3+aSe+2D$ulAC8dNuALBY+J*QF>AD;N|
zJXpbM)v28)fK(CX5m6$=K+9@GM`i$$Po<J8!d=A)hal&TKT|(-@oAot_-~<53QDSt
zQe#;DlvD$gq2LnRZ_XbN^>u)xuRziYn&KKi%bT37N6tKmYoWtSe{&X<alt&%Hp9+N
zpKFu*5<sLQ-evVDah6}U1{rmI-QfK27Ybr}V0W+xkqS!g40VV<jy2Xh1-#*Qhj`~e
z#%xI6F|xr)djw_e9K<+N-wOf@>LpGLKWk1P=_?f-6FLFiMhmTe2r-$75&b0}tKCC_
z<o~EQQutpTl76dRP7!V+@0{B_`$R*GEXcPk9u#6$hYk9kE}#S44NC7M=!kasgbauX
z*xCgg)lR&3AS)97DIE_Lo?8IX8=4zB41_Tkr?WJ1QlZFi6`-}>10*_w>nGttT|r{O
zfW1UB>oON>(GB&4$H)fkq(p~tb%0LHZ4Ej!P%A|l(ngP%-y@T=NiG|?(h0I}-DIog
zO51OSKMu5X040Vu2xQ|xBinP^qh+4pm-l1z9~$3uU6^8={s19cV<5jgmX#|zd(xs+
z^AeP0We$K2xl2myvknSMWyqusI>DXvE4rkK_%kFx>beN&)1;s6f?U3ojQHg)R37V-
zC);-jTLQ)3<I#z#El73WycJbimAZc~yR-UJ{R0@}zKL;&e^hYGi6CST()m7W-33e|
zB~ZXT90nAYgK~aP@YUVR2Hz{TK|&@P1Rs&_en$!p6ZB8(M+&r&H;P_XY~UIrs)2tU
za?f=Dy78UL`&C2)?^O@3P>8Dnr`qi|Bpw-t)UBuW+?KOT%}+gMN7%a3jX;%Kc}Ukp
zeD-zCTzkV`4tc6d2!t2|0|kFsH@~eA?-KkZm+9w^QXhjrrVJ8hJ4p1du2B?_Dv5D2
zJm`+@2Abh&2T+Av)e97;w3gcr*bB3g04|vG+we2Ub28t)FB<}zYJf+mCLfN@7ueb+
zxy~)Vy4qd)VG43&8>9FL^2Iive5f!K_JV~=?7UYs`vQ`F`k0!%My}m*h9C^K6>AEW
zWZ><&J^fxeF{aRfg{?w*P6!q00qVS$Jxipw+DXP*>LI!RHi*viqk4c}J9Nx}hXYNd
z_o~E{0M57$ObT{|H9cP&7=ohlD7-=+03kiVW==n;LpN*Mu1eS;lNP3y-(3n2(-fcr
z;IjtbdqK$^+xmb6^7iaGL#g)y+PEtpE`9&f_YEW#DJ;=?ywzZWa@$Vjf0?8R8~XSu
z=`EPdUUV)ks<t86d>*MJ>kTPs3td937pE^NxkJ253L+*=77~q9w63y3&~ZLnvlV1`
z#KZwA(2&E=Tmxc&zFM9Dl|-#4Uo6WPe1Y0`y^xXg3xmx&B`d70Lf0O!SU?)D7NST!
z_QBfhj3Ujyr0s9nlV%p=+fWVyLpq)FC{Ohqe=eKq)uljB6_JMLvx*1PlBbs66<KG6
z)4PAG(|#n<wUhMfmp9);55N!8T$Q%)Kf!cHl6cr+$j1PqtKA?639JiVoIf=VK&tq4
zx%<u#j~{`U&H@FIAO{e!Le`0^q~D>r@&co_^{BseED{|HURy8qz8@-Jn63x^A-D=o
zQEghkn}p8kQ0bm5R7MZBt;FUwK&`GKLuINkXrds8$1e3W-GD|rpel4z7FP6W0Ltzg
z??RM_r+}lB-015CwAmL;M7kv>{dTT57>yB+eX560Yv9P7Je!4>0FNbH)6iq57geGM
z0JVbGh2`h59Cpj+o$(xlYT;{rwLJJH*?OYR?`UE0PEf`roYiDT^xkEF4Lnf7&cffS
z08hoz3t2n-(cwp);5$_TSrOfFwV%N9G>kp}ES4&wiYc_=q5&6PPEjYQIkthLEpzGP
zVRBb_$?rfLo(4jUV5t?R5UrV2vB9!lHc<l(xy>CQfmETfF1lOtL3@teDhRs@f6fU|
zuks|Fov_V=sYi&RlO?T>Ez=`VRsCGRe)IFyc~-?KMw51M_VL<roSK@1U_QT6(3()7
zAetlHSckBb2>|kMIss3?)}80@i&N|G!X)Azxb?Pz!!<cnXd0`Rv^lP+JKO*s^SpbY
z$Q=EzQZ?=x$Rx*b$IRxB@b)6uN$qb@sHDhKJk+=<wcda(VuU)I^h|}?g2s5g6Tm15
ztsa70z<|QfP8?4LBem|@>DK))Gdl|2V0?xo$PY%S)N0Y^@~H<z{sGi^nMeLXXcQxp
z5VI;;q~V%wVPhRH@mh1sRG*e%V%`h?*Mvw)JXu)_7*qae06Hh~W_ECTM~Mx+P1Qz|
z131Nq<NP!<wOMq7t!m&>G}E2k4~@6>H~A>$^6=5c*R+yMbxUa*y!W=kypBvcN$_}<
zPPaOwD33^H1+s2JW)yKC!k4ESN*cjPWK&^1Fzk(1Cn?(At3EXbXHY~YawN+xMXItH
zXo&_BW}GO>5mMG%u)>;2JP3f{WqL2$&FGz-ZcMbNKD)#Kby=B?8n2mB8)1qi@l52i
z*e=>OgA1jZKc=xML;h4=AL?2cYvjB;LY{o-M7TY<u;29X@}p0DtFXIQS6^17=4!}A
zn9=reaJJ<;C>qLU3B6ESUh-J`wV*JCoQ3k$DS_kFA7VTAWxO(Dxlevn8RxNY@)$MG
z=C~ehG?R|-GJcb0?WKhiNdVIt?L7GeY$f~zIV>ZIVd0&wQ|4|yOVn_;B6U`0&CujI
zOE<vA#_~e@uMu}o>R=vY<6y8?pX}Kr9yuBchV^|uJ+0XA#P^JB#A-tr_AFxI0$B#B
zUb`l>prK(pZC}{~nXiOAtWhvSV8LVYg>F%>vN~s6Yi*{6df5~G!<WQ54|M|%*WGU8
z>cLf#SNnm!pW9L^MHn2&>7W$JLKa!8P)^++I@@aBt6ki-O><yO?LT~@Wtm_Y7jhJ3
z404nQUyGlH*4#ZOnCU7{bxWV0pyB=3*aAE8jf(_kGfCdV_xL%djoKvhjsU=4o*Gm3
zwE7pQ{zdH|<U766_gwhpF9GtG^ceIn@8H{Jy49W{xW8;)=Enb9UJO+f|Cv`%+Y9`>
zO@FsxC{?T!F?jeswZKX}i~7=UUkCktDt(ZyDnx&L|Le>D?Twqm@X9!#{s5^?Ajg}(
z{$ew@B`bArO_vZxVE0(majCZmROXNp1O-{-`m<*G{d@Tn><lHwZWrRbeo4H3!Hj?b
z8WrO=;0cie#`j$=F+=~IcUZaD?B8?+z3)OzEV8NiqCcKB^2ld?b-=Eg6=L;;Ftvkd
zPB_p2aZP{EegE;If+fHa_Rf%C?zijw(`)?ad*e2Nk|U~zq<?(r|NQi?{o?=m-aoC9
zU!VT}Jm)_}=-=0sHxH6lY%azWGA9!}xZdXdr~UHxtKJBounmCXi$JG%sg6wj5-F~Z
zTuYA!c=($#>)4Y1ngsq$(D|F-AmHssX&^KP(i!Hs5a<fQ;EZm7zu$O!Mv&(}Ukd-c
z+1dxBG2qbc0nIzUw?kCm0@IT{MU;~zJY<P%On!@Q|G^b{f4Ra98oKFcKq)0#$a4kb
z%dBJa|8k?ZjX)H<YCEgz@PB<%F)kpSs8orR{;`+%x5x^&H^eKbjePv&kBQgcJr|>C
zsBg2eG<xxmC&b@9!&-c(b%>``Bll-#yT4gv-p?U8ts&E1>5p*Qzb_;ZO-W-w;zFpm
zPoe*xZ;Ss5!dIYzg*3N=Bme(;Q~b}1{WU26b0_}x+y8m7f49m0eFyw=$KId_l<+wC
zVY)6>n|h^0NP>wXFault?7}@b(Y0&4<8t|LP1fo1-<Yfqx;35+d}~x6nygRergiV$
zy_G<8?Mr9R`RdQD=7ny~g+p%N3oL5Z$Ca;qzCV}DX3=pppWp?!As+L?CDzNsMV9JB
z+67`%c>d3jMzCH&Jb=j+5c$+Xwvvu&ChX>>j7B>jc<z)q|JO(UvwtL0VN`Z`YWU0B
zuAe-CFXr15$Hnf(o#laC)a7yaEoDLXJ=^cVl)Do66b?bi9OP$d53~dM%RT8~@u1~u
z-%TTElK@#y)|Z_eA65bqiKtlVX_;C$H~#rkGOzXUK?Uwe;GMY9CT$7R0|uUJpvT)A
z55Y{Op!FHCYeY9@w(}uWqJ*5E9}hV1+WQiaI02M9Gwn;Eh3Y`1^Lp>%knhj?{*P{I
zmibO$Lmj}$heN1%q&ZHpkY_4{(dTI88r+J;m4PE{zynAtrd+pQYh-HL0L!i?5EOY`
zq-O@mltx0R`L5h1RinLJDLu8(%-{^#ZaD^_^SO6`ZJilK>5r2`N=4ot7m8i^F#1&h
zHl0xz0$hQw;IYuKT!4l;uzXw|w#s#s@jq_|$lmXkN6QO!0qJ85h=Ip|W_cDo0m@Q9
zY%B-I=S1<NA>P~eD}(uvmcbqXvA)nK(m7+`y>~u$#2dV7aLbaMc;NAu)9^pMXLJ8+
zutd}eWaZ|VDB>8X_dZ#ssSAirnva1!7YpT69EMOtlTVX>{zVDx)nxYq)7ub!JO(Cl
z%*%(;4kZ|8DE=?T-ZH4lsP7x4L%O?>+@wg?ra>B&kS;+&K#<&YZ@L?#5kx?`rAu1r
zkS;+=8qVUrpJ(PhGiTm8pZuUQ>}&0{uK)Vw^3-(sXUoEL7B6r|Z$AI^X3TA8{8M_$
zYR6oT@sQS+SOn+4;Y=}3xjrWm7P49^?XYAey)PH~&2=IEb+Pl;v?)~oySo3g$p0UI
z1YFDKSw{pNwO!<hfhdGUI=|abCU4hzA}ZbXe^tV6{&|fCu2H<r(mET?2G}Em^Ls!O
z#B_v^8Yty{?*b-BqmrJw{D_xM$DmW%1(-1+gHV(oO+||Uh)YOY0}o^u#Cjs5PZY?*
zNmbV9{c@<uxUew&(F^emA&<A(O1qiLh+S(6OvFu)-!*mAdGq>>!*d~<ary?pLlZ-E
zZ|^{e*%!%%P9&^GL}au^aM>KGvzt+mpo+r>-ySb_5D?M`T5BdBvp|2Z`-mw$0h*6m
zvAxm)>T`>-qLe59=L!K$`{!=aec&UahVjT<0gObVHRAXj{2*=6p)v)`M62}TKXpv8
zPcD-T4uMbary1(w&f+<b&uHaw5RtsUw0L??_i@&Y15X^Zmp|8UFCKZB;q^1=>PwM3
z6ILa_?K#jH_l0J7+mZ@0E}w-#a~=S>JUzJzKG)_WCh@L)a<Y;Dr2;lR5LRXX|7))V
zO8+24uXelu2odL}D4sfy4b=<~I^8c?PLeZ*AIg%MW3W&wE}$v&iy&0>5HQm203of)
z8RPB7b_^kTtvQ*Y@bWhuoH17XBzyt<PnyG9`i{2Sli^B>{us~q0L#5ue9WExmD~Ps
z@7;d|NCR>Mxu}#7d2d5xbKAdwP<BE_$92U=_)H^tccxQ!78b&>_~i5vnr#G0qV<+*
z^m56cWW(OgRk@%BJfo}<t(j1uDkCW<wKMm_lfxk;jTYV^s5be}S2BU16GuG7@JbAn
zpe%ujwRAnFgKBbB*92As<9A$V*rx+YxA!haw=xRoVL*f1v2pTme@0`I2x9qsThmrV
zaL$sbcQiccs*YZ*ZAha?(jpEyK0O{Wo+r~T&3I@+?^3qDm%oJooeJx?7#3R>Tim;{
z?D)7xA5Z1~+aGcSrTS~80*TbJXFv*ax(E;znvR!~B0Zux{h#G6&@de1_SHpqncacn
zc8}*5aseWo3Ph8X-2fDE`Z<+j1Op!FlO*I9p?Zmds+dp$q(^11!A`vD+8`?m90;UD
zRPhMSM0%jKx8NZ7siKvpbYKc}b}u=j8)akAnQ;3Esz3ba8AUSv1%m3&MY;p3kH*Q-
zA9^iT2WC;`s=sM5s7aqNRRN5;3H0&E_Ahn0|GZCD88++rKL617gD=cJMjxO-84}J;
zUhIk%KN*;C1xY9Cj%n@a>W*n3WdGNP(0LY&fY!SDtpdET)Luid>w%A%Y!B5981d|D
zd}iS|0WJi*Yo?gt(3EcTDPXoNK6GIP2P-Jc*#f*0?aY@qLNg^Ax#J-&;vyQ<)O_Z_
z2IxJXr@#hXd(@@;{RK3{%9uflw4Q~8r%p3(!i6s{R{UCMCA@?U@7D6uysHTx<2pYV
zwewCWh$_iaLsFiHX9G03c7M68C56ZLeUk7%7e6!Ed#!zNx=i$1=PUuO;bPRYQF6b<
zQ5;0@^h(80A_>qX$TcLd3S(ruLve<kiH^RWzIeC=9TL0b|G-yj-y>vGzoOp}P=I?U
z!d_DSXA}5DFg=iZGW|J83W%X#7unz>Pm*%udQ<P(sF4G*EV~-3E&8|O&Z~xoJ(Nr<
ztf+4RFLy>UaQPT^dRZamZxbuOVq;p-$?sIn7j(kEMObwv=Da574sc<`M9f%*4)?uj
zc5w|+r64P9(ak<9{TMc9r;#+in4_R5<d~!m_oHRe6rrU4;x5Jvp7w1(c@(&vg9K<-
zICpdDFFV66#`i#Zre?6>X$?~PT#TAsBP1)SgeRxVJ#Of=1m7Nj6ZE$=KIB917g3fv
zyc{+7(?5Q<m&eSX94PF@YZcqKQ%!u!PPl7+;))&M8+853e0HmNV<i(^<Rmz|Dw-`w
ziwCZ=4F)-m6Z!JD+M^;V&9Gd5)#%KB;J_c4ko5j{P)3#DLVLM_c}KF+i*DHZ4chlU
z12GZU==)55x$c&?qW-BECTM5dM1;`}B>;&)zgima_d7rVHG868Wec|MeJ*D)KL(>k
zBM{lYNQQv7YsW9FJ<LDUQJU?hP#BQxxpID~)edU1IPAWJ(&3oV=;u}k{p)G4g{;w^
zQzwX@uLHc?#bKfTmEBR))9L8n-C_IImH+-Mu%=+%+E6VMif7gj+<-S}i3JtaQUVg`
zjB=*PQwGsPCmj1Uyqcyb_SgUZlsC#v4Kd>68qfdxox3fBlT!Cfl4JMg+cmB2yiM(x
z^(Z_Hd7dQzhY_&V@NVyJfamA)Q{9n~-t7*T7EJt8>j<}h<k9zptb7)LGi?c|gvww`
zI}-(UOD^r;h1Dc8G%Bz;jy>E_QFXRE_w|29gnEoTI~d^C7aN_|U+VraECku$dzj(Q
z0er=A*hvoRdz^LP*%)q{CsD!=JZL}nc3uyxeh+%}kKS*jZX8whYrvI=9p6Y;K4K7w
z?<jdR&vxT-6$G8CWd_V~DzXT7mq1ZR`A^HELRQi4(?^1;_KuGoY99vXz?*tIPAwnN
zh||s@ITUCaU%Tx~W)rCkL(fa^TQfR#(#G9jfIj{F%=|DQ#uw;;U-6sD-cjxcf>G+L
zg;8x8(!Mg#o)lF&AAn5hS3g3}?_Py891;qF!yxw-&`dhIF%Z%TzUnjqx4oBPbtvIJ
z;FGkj=P=7Rv*OBl$i@d|E`g0=z=e9P`|PU!DHZ%3P#}I$m&PJM2OHVbZNUeW{p{r<
zF!xm&wQy+ReyXaLP3eK61%QN^A5g>Zb!U%CD_v?0YLqa*6U&mhziDv%!d*gvL9Thp
z(wN<|X+0@4_+pw7%(I>NX)y$BA;JTf$XCb#q=ZJ|yvsCzg;9|V*g*6|UmFFl$2>VY
z;yFVTz<asSN&rQ=>u<{Z0iek(k|VlyB&EtnM7o5c(-anxh7(171FF!7{h#H!9t$As
zeiO`SW!rHrdYz$8c=1PoMV@-g;K`R#Yd15R&6wiJAfV8_|Fxsi(!0+I3I6jI`kCFj
z?&5JwRgEEkn--;(&R;~&j@vK}extDB+%q1o6lL3Fs#GfxdYS+xOMc5#FvZGSF%$2^
zX_qnqWqH;DAi1%uzp+>UP0xq@bzpb}4KDI!Ql!`5#cbQjhrj$-V5}X+VEGuhaqoPm
z_U5Y%-GJZpTPG%1xGH|VsOwIitt=(8A6tTxl)8jfe$@W4%}puF3rn7PQcFUJH<)O<
zRX&>sp~2~DKg|CGqSOkaWtqay)S5%$0i{R;5$#CusamW1HN+_T#__&HB;iSSe})<X
z7mkZbYX;0uHhLgQH7N@t<`@A=eiiS7@8Jt(RTa?<YLQ3!7f<LiPEinD%N;{MS(d2$
zFWn}q)CkGW*Fn4T*I$zIFiG+MV+(j?+G8RQTcaHNc(|yK>AQ;HI});kWeXlOp4b7e
z1KF{_?G-@CKCL61Q#WM^?6`}BIrYmhVMYD?_)7K7+5$k$^%uxi=osnr#<vEsHaL?+
zK|4smnD6EA^Z6&ivO{V{h3yl%Re89_Relfm5g+hArpcfc8echwdEt%%rG9&Y7@#!J
z^~s*nIc9zz3WH3R*1|4Z3wvu)<`s)zexxf*4pc?7vL8#v6!4_q6y9q#kaL!X;agIi
z+J2a3@59|-BRwhjlPMX16CLN(bOD<FUQG&Gh^^DvYOi>rX){{N<wS(mZvM5k3@FkV
z_Wke6${i)HgaIZ*oYum3#3P8#f@4|uY*bj*%2e=%Z~1TkVz}Cik$sByad3zw9D#}?
zUJ9(1Y_d)pH}gWn%N)uY^izv7W)E4JE&2AZ3Z|nmi`{(@#ZPMawnhe&Pz2lgbQN+U
zhLtwAs`;uRLhnCzfTKu25^1M>+{E+FlwCx+0|vOJ^7fGEuRB#>My=DP{W33MI~4C`
z{m$%`?Pamo{>=ne{#!>jQ}&o2qok2c{;4knQ?ENIQ8SPA$;t&x*^ts>L=c)88@Vek
zUW%>m!k<62<USkQ`S$x)-h$%SA?i@K0rP+++-8F<el-$c1+W)#T9y7O-`^Hzon2yZ
zaPy73plHV9@rsL50or`DmYh2PdB4*|kEP}+3h|gAC}=HtR?HM0#^#JKgW5%#P0}Mo
zETCFJfugS$Mx(B%J+ems3soQ6D*u42Vuwu&yBcAWjC=j6^p|H*5A+15oh@54^e=h}
z5MO%a6yXf5Mg*;f_a<h4oL7!u@I|C>VJ#$Hl=y~#`KfS{#_wm>=;ZVNT#EVdZNh~I
zEO91j;p1#x)gOs*^bjq=7-huno?-Q$HQ{Sj*>tZ7^x!8I9z@0(To>fB#OxQw)&%W4
z+nGUpEq;<s3j5(Z1D!igo$J<VSEHga1_(aNzK7OyT@#`BDse{zK;lj%`(E{p*!X+&
zdXsl!>O}-yj^(2>pUEJeU?T|^VhtEW@}60<qE`Hb7OFl;Hcd}5ZgSR74!V$`?%6}B
z!9YO{7mUN}9N?UaPo+;|W^c-%#2WG^jQ&Bzc7u*ho<}-fimocdfNlgd78w^HHd}R{
z$_;y8^#xZQacmfaLt8$qs?*;MsUms_koe8`VcM~n+ySl7iryibi{G!QUxe$~b)P){
zqz$osqep;<H4qrb%CI>}7Kg{O1$hY_kVO;$cQx4ilm`!~KltToch~1SHI`Z7BR5qC
z=#n9pV)IWoxFG}axGtX`icOVQ0wKzDG7GV9C^zr){A3ZK4`>si_hUqjos^8M%y!po
zEcbD(w~Sh^HwTd3?D1i>vFTMYD%@;%YjY1cCgrwA2~xUijR}Qmjy70Vb|dixGeJ^3
zUsSP0)9X)e*)r*I!O`~w+(NzN{jm8W-iI%0n2cC+?C}SyDK*7yekQowN(yQI;h|_u
zA(fWMmy3+N{AZg$?=A{$p5ssMk>AdzFph|nZ7tJd?DNPbbcYWEaSWdDztax$!WOY~
z?jBoUB!9ie82q6RM8S?kYmOej@Pz?^h{^)GXBo<{fi0c>*`2;mY^?+F9g?0ltF6W~
zo#5i*Sv}czWzm|~A=hV4K$ngDnhtgo8s{JK%90eLqE>@=v}ej)hEQQ{L~j&L-j#S{
zAs)Ga!9m1_Qr~oUs^-6!rahVfpn+Al`ntnNae`Iz!BsspD&SCn!;aakpZXq%6Q_U7
z0qs^(49P&EJ$=E$+iEI!5Do!;;$(f}8U}p{Q%Q|-*cJB4J1(>>qyd(w?GZBGG)=Db
zoGV^YjE#xnC-`XWUuW2>6@sYcui(2$DueGW%=J}#4`{*L)J7C#d@j$#)PiY0uL$j$
zajRwLZ#obE3VduuMU4}FGnv8!cUPP$4LN4Y$lY<4IKN9uu6ug6D4CZl{dII}t28F%
zdN?;^4MR0fjRgP3N1A&DU}le={_O(ohx{YUXmklehj6#2XL3KWcl3X|<%P+kS$~7<
z$4%30(VaOj>%6S?$=fb=tR@|Q`5P#Xwc7EratmTaKYyMpt!wZ298%X)9_ugAXib0;
zAuZZ6?MYJPPEXYMa_mI`R-q2I97XO$t7}d`8VtwM`S}qgK@B<l&$Tp5otF{uVF24D
zGK1)j$Q2=~I=y4hnIS0$vVQQ)`lXcCHG2Ly-eZD%){A^`Q0>~An{~Mpf!!AX(M!H%
z{Uw|qOD3LzdL}LUH*kiknI3K)0BJ-kL{^{N4bx+m;DkGP<sQ8Ic}@RX=ERX`Q-vOD
zIRLVQTH}%GzVzQQrIZ!L(pmpTjo<{C0dtQIR>Ee8%;11xp3H-<twhg!DUI@1+{W?~
zeVlx#G-)()diaLXTYEw*s{jZ;y16quzWlpU3EwD2I?3>H+VI|5X+xoFw!24M#!Wm9
zH9YI`GUC8-@>nxm2XVbP?ixv-dSb!jbvSoGoWWY`W;RxtR<&6c&+|A%hUO$M?@?H?
zd3vH=2y1duknS)ucN`{+o=7n{K7Z~v)qpvao;gig32oxVZ)Yx32DzunO>y$UmIPbT
z$a~dnvq;sVrEYQ<<RUZh?4&ned(4Z?*)3}W2T$58U1%}*1dnncRfRYVVVD8Tk5b{2
zdA=hOB$X<FA8fQq4P$62*+}*fTFebTiO}Cva?UM>$+XhN_hutdXkpj~KoZbu@G~xP
zwIC=p$b$G?M)Xg9X>yZn%KKMvaf=~qExr?XR{oUW!d#Rps2&~%O;E)nmHDSrv3}=t
zW+bC9>1|o^mY#WL=V>;XmV23yvMmOI1ioEJ0KPPM>RUew$i;#S%U~j~5H`FIlVf{w
zkTTSKn<thfF48p0l-r%k1KkFs`f_8Q&rhYQ-{t>_bxvwB0TU&L*{f~Pvl?YGDne%7
z(jHdNTzDak5f%V>5_ZA6ZP}`t;eMStSr{*WQ%d|I_o3L4PX7W}tHSyj9jMMl(HNlH
zwQ3A^ws$guCvM5UDisp%-kct7>MA`1u$dxXzcE?q@yd&Vj|^HBSa`HR;_RAbByraB
z(QStb)PmMks-jA)w*QMWGa-#!#BYXR(B3m#Ol~W#_nhnDrzR)<O^DtTrAz2-kq!^^
zY#>euId!ZJpT^)lvOb)kMmmwLv-BG3060R?uilqf9WLAVMOy{)dMO!rEo0z1k%5Nj
zs=H$_xzjgLZGt8JbuQ`Cq>E+pg#oox!q96xnAG+6QdJJL;M&Rpx~RT_x8F1+kt%*i
zSZ$opK~93nug2d8x7W-=v0t$JDF~jFR|-hZgy)hC;hi#6M;h<A6!a7*Dqe8+byIG_
zS_@xa-z#udwUu&UbWS^vB0|f`!h^K@H9t>5ha4lEv`7*tX<W4bs=zd;<tWDTEA+N7
zg{`5KYoAPdI25Rfhl!gpcwV!)822Kv!Zp@vxRv&V_0x!E+5#HWIIlDyr5mjU<H}$b
zNHTD>#>eX6h~z-%WNFqUfHkn<f+o4<NRu;zGrdI=wsf73TXF?|W*=ekfM(fFtf;j}
zy?)wjy>M!CPlbN^r17{``qSPZy^zUchp%n_o{-x;K_#W>*>j}-g^m;+IDx<*1Hnfj
zGjxKIb5$ND(a(MXvZgN5VCuO~w9C3X2y65kPIGhFN`)L}hT@XZ2|`@Ixv?PYS6?^8
z9Yk)$^gHoqayam=8h3c)o>>ab8IeT)S~uuK;wK5lbS{7CG=c3xa-QMxGErEGihwh8
zo+Xj=Z<+wSPI#%5AoaASYmx}vRwpNz>cOc-jQgA=%fldKl#_v&Ps213AEnG3`qARu
z3i?EoroSrW@Fh~UDp~l`M(K(M6mw^-hzN5BofU!y(urCQc13|~NNm|@?TfiVD$Ry;
zI?aY%bS%{HpqdxoOr>5>;)f@0*_hjE-uNax4?#NR*rMJjFn~v`IX=hdY*2CHNy1y}
z9kQYSwTWE4n)89mAxO=b?MuRZNk@7UL*}Ph)z=k{9_GX!h^v_m1k`dr-U_Kv>et4q
zRG)3y4E@uJZkPIT@mEjIZ`iiLxS#&DM1eBfZu-BT;_o(bG^3uGL=5q~G5tb5(wCbC
zu8_RLL`T^ky=8rMk$le$VNS^U?tIgK?%U^4(oeC%2LwlMAObW<3e>k!|C0B3?{l?3
zD%!fL6t%X|4lot1nqH72okvO+eX#Bz9iQ&`t}-0|^WtmoF;kiS`Rs+R9kp6l1MM0|
z{0-Lcsk9zwBhOCGPB$beR{G1?-RAaUPU%yWVou%OE*aftMU&MVsH$Z{fe{(}KUyw3
zFczP$wIjNVPk=YWyBLxNtu);fd4sl*Hkm9PU1Rfr@5eskrtQW`uy+b36{5RVi=n@N
z=@Fl(4-oCuJW0hv@o6A)c@@><H}V3H3ItlG7vl{|O|ECSVylxP_|)D<=9?Ie{9enO
z_V;dPCUZXC=5%{St)>y46j`J<>MT?GGS%exeSjE}6aI@=dkQ?)KDXqlHjS-4)L+~g
z$U6G}3znrV6a1}=ygAz~fN^eGQ70O~=cNt-#j<CB`fcU7^IM}v&t{aKUF72-#~HeJ
zeFB6<FP|;m=VM6h2@>miG&(bc%k#RaMGTuSit$VGx35Vi9A{Kl>2=C;R#H06`D6X4
zg_4&3gt2V3zHsAYY9ao)iR_p?jHz#`KYByiEClA9=0*<ENeyVKa^qh_oyEH~Q9jCd
zh*%8I%CeYRDBwe4jhE6aU7JkYPA@QR_el6TDqlLc;nYv~iy1N;2v@6r;kP<7c&+yY
zQ~t*T(_2r^=hw6`whN74Hd=H!2OKTk>|Q(szgct%wC{0VmUuJWCa+D|IM$VnT*G(b
zquqIzp{Mi8lPGm@AW;UQbJ5W_!$k@<K(*^|8w<0_AB*%@-rU~{5kDQX_gxQWwf0|W
zQ=K1~RWgb({d}4h3x5dA)f3Sn8=bIgbSZvMBlgiA8r8JYz_*zEvil2#W0|d3V-wq!
zg2q6SfqFF(dGixizs8hK63C%c+64&~Akki5_>U-KEe$uyd(|ipogd6i@!_UW{qJkT
zWVG0x?&r?Uln-94{Jh|v8+IdRDh`9BNb)zE26M!5U|sbw6L(0ej<~AFYuq5^uO=$j
zkyP<>Mhz0<G>o|MhbS8L_KJKs@KAaa;~|=CN})0Q--YKzxrh7PH-A`bDz;3u!V~}N
zAHyUbT}d(1b9Zww7NjHiJNP=J70%A&1Z63_%(_CeNU*ONQjQma+wbA6$E{d?^YJ*u
zBkzcl?_CQgDyNGP`K(9czQfX!U)yZ_@8a6TP;5Qn2f8cS1Ru%O>k5ak%6Gjk^$P%d
zy8V;>ps>AE@j{b!<L#DMww!xRWwH0gq5KNG64g^ru%}cNzAv)1w{xaUXN0$D!`6B-
z#HkKJ68R;a#>uB)ZOn|K<K`XxMwmg=nq!J-N|O~6G)M%=9t9gF603YkOv?2X`e3@{
zF+wmPw$HxNK@7RQnHstCtY@#1yI^3Z{ULh(08E?{*K#O-R4ZJ>j(hU#UTwsB(sJXa
zr;>G{VB7LcV{WJ}=03w)d&4`!ewJm;800td-r19#v?9yCaoRV)jn_4_Mr-a9`WLyk
zdJuRm`|(G@Wtlm^7*et3^mPeaIvhI@<V@uzln6S^2Rq;Uw}<>g{WtABq_84OC9B9}
ziSK8(_WtRgtoC@{ANyWACOvBv|DG8#nXC<}x((Lew8L%$clIRt5}(Sw*2c{hzYF$A
z)1*NVrDL#Qk7>NO!p473XQsc~(d@drMu+3~?SQpQ?uBrk1Rw5%&?hW8jT#^&+>#ia
z9i_BofFU)@98>1+_a?<UC-QZU5luMob-Y(1%lYSR`6{M0FqI(w{4!plIQ(rxLVHKT
zJ@@`ShjFFK=u}LB?J}kx(o!B?g)HA;Kk~Kt1`6GH!DPo65d4Af*wn9x(h9ZQjV1<3
zXV-<Ouae<>_&*hIPR%<uhwchP+V!vnclA_S%&_I_3y(g1_3!RV^FJ`c(e2fm%k)(d
zJlq$>c6lcd$Fp%#=o{M}$uB4C3W4|a#}=FrtaGgkuh4mBIxnsVRC0~#MwIk=43ZP$
zNO+wo9rPDm$i>E6+%^thNN@EeH;@}@rDIFb-}z}9*X-RCTve}M%dScxARs(E95iE?
z{cr2iK{n<sO!n?Uu|+#xkjq9(o1K8`mr`)JoVLjeCHb1VIG!>Hfp-4Ji<C&l&tH|a
zYPCOQpVZYJd0*9yJe>PH{BQuQPnP$Q$hPmV3vt@qF_8<%2W#VJAO*a2`}vVR2S>0z
z{$stC6DS=F1z`}S0(Fl~a9yA#a_?_-EDRzJ%po)$TY$E$x_AKngE)&X$d)hrgyvh&
zshy;oiG7dZCc#3jzj1y8fBnbCG1wtd&h>OjD7w_;_us%Z?@nJSb$=ZD8$>m@T%P7W
zPy-Y3%K7pOR`B<DT|l}``smmQ0ciCkHvpmkm;sv$_M-!`8^mf*R)THxIB?Il3AOF2
z{eA@I9=&He#2;G$z1$5_&(Hn>@imh;XzDs_>5t6?p#;BYI_}*(PBT}imq9fZ?qUb<
zF?2tA#sl_WkqxX!0L}R4g|_`z-?M^lfP+#2a;$K<PLTD{`O#fgl0(TBGzN73W(CoL
zdvIK8le2uXxi+%aRaDqTq|`F_S79@)Th9>>UK`jJtvNjW!4O7@>1>r=eY9eE0Dk}t
zu+SliCD>a20r7!%e;+#=Z^7qjXtw~gFLh^-q#J-ou{FFV6Dy9WJm_2g0{gwtqxBm5
ztJlwRB$z&JT<^dUm~pvofmFZdGVV9vOQQ#`fUK>5;~Ll$9bs#L+)V3`bW)e<F#R6*
zx6-}MZgpAsKLvatGi`@%zj9ZAuZCv#r*7*S39uJBVgTB0U1_W&thtN7cjn)$0Nrj2
zXpKwXC3#frPZhN@iH-s%70qKjd@v9(lpHOqb}|+PxX3k>KiT=X1A7nKf789N{IC9V
z|8=EMneLABPUvHnSvMff3CF!LeI6*rJQVbRyQljxLBJ{2d$Hi?4LhB~0^13~-3eGI
z*IAIha}CapJQHriMkff2zfElKm{saZ*#lU2=p>ILG}3%U_PoW>(O#rL_$or%E`Uj0
zGLLF1Wq<zuQP1yhDD#>i21p;O0Kkt2m#iJ^A0JTkSp{!*fo#gA5xotqSg)6QppdBI
zcCv93)G1rqQmN&7?7lF0mu3upwcW|iV_=m^{Rsz!YlTP4kB-jrsWS<;-w*dk4_(M~
zmz`CB!AEhusO<iha}G(Ak7~$3x&8{iVe7v-mi{=pmq)Jwa~<B9!=twqbbsQg<VtKN
z>2!g$pO;>0Ub#Eey&kE8tF7(i!)kl^H|<0h6cf^ohVEH55lJ29ZEE;@4nO2#J$kwV
zf4NG`a|uQP3(oJh{Q<}8;>Etlip?|}hQWq2Uf6elpT9j}GU~RJyORP=9<}!>Ga~ha
zinMs!YSoaw^$r?D#ZCezl>4wwS=y7hDq#P}P`mZhEgvJN^8=2O2K1fxAfV+TxI}!D
z9hoO@EU&|j0vJnIQ2NS*mRu9%TL70HmfCh;g?ue^O>xa`@zj21c41=b)IVm>5aD{x
z1e9w24ObNJa{c;fld_iTS8oOMN62FkL0dK1kqJme`TZDraS7Im)`N5@fY0E`h|~et
zQvsLN6VHkpAH95kMq{H|uF&pofcXqDGuC6ao0fDZlcHOzeb%IY*=v9D{nSMeg9XyH
zi<4<q49(7Ck_U(gCh0nCIdc5KQADulMj)~F=%zQNJ|9q&oUCGgxb>Z%${~Gp-c<oV
zej4{;k-A?mi7rYX#cU>6ebT8<fR{g?ER4}+EnA4%Xa5G$rvx?oI|;=|-0uG4R~0%D
zR<Ys#m&*+S@`3>;KyMkeDXbLSi|;`3YP?io)>;4QBJ~%!>R0$~NdD0uq(l|t;qqE=
zhgZ|OMO}SD7hgquEji8ECjx4<PP>4Vgz^yh<u2M5Ht}4i9u>C0?d6A2i*+LFZPpsF
zl-MMYLs2K*ue<;Pd#x>O=MgQ;=O+P}^tPWE&*h1yj2NQD-a@`S9L<tr*UI?Bd*vo4
zI@)%oR@ajH6b5$F`oVoDN&RH~pIh*`F+dzx;*VDeboY;ykCZX9MuX_H+K1C{KohU@
z9jf(z!^~Z)c90CmqY~y1x{~gbouw?rsy%+RUuHRVVG1^cN0o>hg9Q!)-Qr()*o}rG
z+6TgJrHz3%TB^P#4?Aw|mIT=r7?)1JERh@?i~T_5ece5J0oGxB9^+4vC@Srk=jaFf
zsI;vj_;D?JY~_iT>XdLt%YjtJ9<is_!qC0oN#zUA<n1!}Da)PB)DbPHye}T+eG8?M
zA-mmCloB<ydjQ3O7OhCvq?V{Bv64q1a>DUw%3)7#vaTA$oBcBQ6?@M33zHgek0-zv
zp%O)Ol+Dbrjjr;^E!jyF#1t20DdXSRB1$u&2OnM-x3bBCn^Ln-TE2Hdfdn{9sK@D)
zsJuj6D#$B=;^y0{PssS%CJG&q+&X0ASUk-;j5mU}Odjlo0auvU?n4M!<anmv@+DuQ
zR*g8=pFKngJq9Oz*1fJw^y5_wedY`8=t_wlDVEfJyZdcouzG@2maL|OyyPLr4+HWv
z<h?pRdU~U^Q-2;hzcF7Fz`3PI3n_`dv71Io?<x+m`l2ueZXLRJw5YX0X@kcx7?2RP
zu?I_YiO>VG&V0c6#J`TC5KsI$JP4HA=2IYP;^V)z`VI5}RZuCJc1Ob6gQen`W)|E8
zT$B}{bFqo^(Q?-4fhw`E{c{Tl;6g^l)^O#ZrG7jR(qb`~L*gaFe7RAh@>hT_C3-p}
zt7jJNJ1^d&P3A|OfAymo^T+#FITmq2ya{0-=LPH$@;RNAN9|X7{9`DdATk=tHy|JA
zV`<@A{S`8&M1AmNSk$nuNhrJLd%AzGQ2Q}X=Bz({?b64ES2C&dV}%hxJE7RzQear1
z@FeX<3>ajHLRR%4A=7c{;Dl@=IpcScit;K1YaxefyI$A$S0k<?TKU`ej1Ms{h%Fz-
zESu-mTr>)1_pr5(1>+^c$;rk;=uXr6)67CbpsLOBLQ&G>6hi}TJy082YLqwaLQz$U
zqP#mtpl6tf0T(oj@YMQ-$eXGm-L15&+&NNdU}ut>VGjmlxuG)Y>hA=S|74Ego>zPW
zA)=j{lVoAI$xKpQ5X`X0o>HT>&y0P^I$4dGvb*GJY0ir$-v#p>8~e}cnudoFB^D4q
znsl%LWQbpP6KAX7h<AkiFH#X{lG`(m@rXXovgzG+n$Yg}Rax;8phw^>NWi~no7#t{
zkxQiZQsYy9NAwo`=yO20xb7WieiEXi`nS3fqYLedXI?99!zz#{n{p{OqQ#kXuS6im
zD_D5oB5hSc=<Rm|QAy8W!!8uo9>|vbko}74zZz5GHOy=Fqx^zr6obsnr`S8zttPeW
zg#rm0EY<1KRy9<ac!@)*#X2TV2%YBQ1MIVVEzG$5%0CKuuB^IpYaja?RF(I|aLW^u
z#F&+p5+RP3NNLJ0LkWX=_Q<m(rO)D8Le&ILkDW2`rXrqRGyOy6&AYXEyYRlF?Wf7#
zJoUd}e}ayA5@(_u^m;DS$9#(M?1ucS`Qdbrg`x93f*-V|g_M*{2dPJ27@rW9TfDM3
zY*(VX?NoOG!Q^2eJE^-bsTS2fyVgPpCGM%SfXlR&;`8vfdy?=9d_LQtTD5j$Pr=;8
z7rUkec=b=VzR~B5<Lcx|G7iUy^#m-DYNPZ(hmbuLCAvwEA4@y!Yr%TiFO0^+2aKyM
z7&u)5_$?&~8QI)IjwP5Adf{YYkEu)_gOg^gY%=lsenYHFaA`q-7E7vjTcHN@kcpK-
zR!hhhYXo)s!(d_@(YmP|Rgt(JhQdNuj*bF(5zdr)`Bp?C7f}MQ;Ky*b4HEHE-MAu~
z{#u-^wqcpflN}B^t6j=-1Km4Cs%*xhI;^=Bg=M#=U$uRk#bXv?>#qd=P>Bo)#jgw?
zTT-0};+^^J2m#_g<&FVw1{Ck)bbqdQW89tJc|CBGuY*R~%-D|sv3_K=)krsND@5rO
zkNRm_K=j-jsYvc0t+JDFa;&`dgx5+l1NX`V=!u7Fbit!o+&d|~&S);NCu@4x!k3I+
zz?lG}I1xBXg&!SfiV#?m-k~>h!>B$8L`jXO>tEtj<DanEKBmVE<yYICZH{~^mNp6c
zMk*{<fO4Yt&?{rI98#$CpplK@=jyeGz6o7o3+1--<5Pc-z$1u@%Oruy6a8%Dm;%IH
zs*Hb>cGrvCc&BJW{Gy9VdG-sf#Qxp4NrWUnc?(`E$vq!hK3|KN?fPeUEZXtZ6=FmQ
zIbZs<m&FDpeM)>&Y3HVu#FuxJ5w)U)TwR)Mpe4f*$z^oy4is4mI6e7jA7DosCt<{(
zrBkIOcf50yH!;A5-Wb!Tr2&ldc#Fe41E5N5Lf-9N{0F=kf2LK85nYtR%Z$aO@7@SV
z6iNkXtiQr0`UfzUFQSsP_r<I#!o?Q^1}yxGpvaqSMS_vkjw@>m{NnNykbLYlt-kj7
z_5!G865*17NZW;gUahMw{zQx5LGZnz!m(K_&f?LN;ec=|W1EcFGm&^rRUYLIxS}h`
z89&8VWW$<jClXDpS1qt&Dyn&47sg}l;O+Qt$o8F1z8&Hg1SvW}qip8d)BcnB$?>D}
z!PvvqpJP6T6tP%9ax)IthZBs5%(x@{wLClER2jE5t^@0<vl9`kRwtj|L3q`Jr%lul
z<S63vSttuwa0^~jD^TF%i&yrT?b;>AI#24KSForQV6%g3*sXaHt!mrXcLnUmYRvNS
zdQ)SQOe^uf%~x|+B<xiSB0>YauTZa1C%>O2Vnj_@UPc<w%(Hn>U~s`Q<SJmfl7vm4
zBc_!WzC7XAH1LVq%?M?2cAb!%H%~ti%l>0QSks~GS?f?R*q5jL6{VbJ)2(YSl_lf4
za=1#w)M6avZJ{4KnkQRyM(aU~>Roa6JHqNCuLYko`C>Y?s*!c+*yNcqwuJJ#LA#GV
zDQn-s<VqcI6@_pO@p4GalV{PQy~=2v-=pDL<hv2<|NIvuAv}Fo^PE2h{Lr5<C-)@J
z5y|CKA~WO;yo0NgTLa_>c2#G%zFp1CX=z*P@JX64Mt?Ec_z}>A>1iFoxSOsf9~_ue
z5y0dHo%q!)yBf$1rTf~ei+tn^O-ww-)bNlYcJAHcixm9WS1>Y|hitKso}{i}Fh9$y
zBy=4>Ei>C15gsTvN0tK7CzKEU#%HCqz3a_;!#6tX$6}i9s-&i%DB`%YY5BV=<TYMe
zs>*EMgBji@`%`+xH^j=%9P56*{&im(=9y<4u}UG&AD!V7%lLMqK}@2AYswdXo)8;{
zcMu!(o0UYanfHcXFtA3j@3GnUb)w>|=g=9r>yI4fq0VSo1_cyO_EpIbcq%685M>fk
zdnl33*Pe>E7sftgiR<qh5EjnuKzYAmL9yj2t~R!?A=_g`bsDZ|cG~PoS#@km_O=o~
zUDtX&Gbw9eif_+zf$zg!ca_Q13du0mO?Yds0>LZ0#OY%3@J0v8cSJ&cKLqa--(=@=
z&aJ#&c4PS3=kK$j*uSf&7*o3i-O7(ctc@p@n}i$iisxZzy<e3gzG(ty7|eZk@yXb$
zIYS-l-gRNGjx0fWg2RH3ol&N7=&lG`r?t}kTO6GgswSf=yK7nZG4CV%XE>Uj_g`g@
z5$I3Y@19Y3`j0Wyd@OuV%^Iac+f$B8ReS;85am^jq<)M0%Ce6w(A1W{IpMA-{-)P6
z2!XTST3%XD^|`Yq=aKz*If+9u`2>EQi-rA~>8oI?9%xg)pqW3yS1fGqjcTUIXu}J=
zl8e5lxW*tb5QaQuZeBb8YzQksP1WwH-0?@%9w-WTXvAmwSP4Pp0p@h{|J<pzo*^ll
zL6~p@v4ZHXMZRc98K}*plLVoBfjnm-jkGmd)&HW&_0)`Ep{`5fot^iyLkn^y-hfz{
z7(`uT6pLo#3z|bUR_C}ubQXqzBjHb+kp3&{rEhAd@oUtAjbE5>M~NZhoOSOwWkSFE
zC!j~QtD<P}OhAXDBCk%bt{Q`;&bX1KU4>n<o@Uhg_k;yVbKXueA@^5SsUmOevf#?_
zBfn?q?8Aw>VdGI^vaKoLz{O@1HzB2WH4`^uz~jf1Weqs_-dXsi+WrKw>&+{h#MsZm
z{Y{n0`PR4Wlwms5<kV$TrbY^fYM}8CcW5fFj%QL^5ZWcg+bT`YN|Tsj87v;5B~OdP
zIR!OfbN3YcCK<cJZc08CImH&sthG!(xf#F|Dm%};Qif&AX#F`N$f!29UXR7FxCQ3Q
z1x71dFj}#|N-~9K1?kGv3oVEsQ!UhW*|Br)<JfQG+nvwr#3iNNQNX-v%1zIE82l<{
zdzTRxH*+!cFG6|*pF?^Mrl)1{tJk}`$DTQpB6rBw;)UX#h%ZvT`^`QQ(`Io1MhbU}
zZ3yXF0G3nF0BMnMigxBc8h8dLKxcS*ENL-YLqEo$m2+WFSa%>fL;k_kk2Obm_spjR
z=Q!w?TabrTf*2E_CmTuFj_NhlQFOj+)U$uNn0hQUHNq*qp2(Y^zY-nVu(OI}M7d3k
zkFk#wETUu(w^YoiiAT|*6>8gMLK|csB{+B_>tr2J)Y_2j8|o~?yv<lIK4LCuDbN$u
ze9YkXI_I~&AzaN|kGvs!`}ir$Fh{7MRpeKF>TR2HR%{|(4(eNbkyj4X+ML4LGsE^i
zi3vn%Rkkl0U%<%EYm{;{cE5iNfO${zs(Tg+FKjXOPW_uLey}OmQI+{8l<e-X@oIs;
z)XOi+x0WNuxl%S*;j0J;t@8eA<{q|<fSJ|wnRxu^BX(7k+h(yK8qY{6*wFPMkrCH?
zrO@v!ZD73-2|eB#0!otzU3k!moUv#<0-Xbe!?gy<Ypa=lny^8gnz>ricTjY%ms(^+
z72|W`E5lvPmVR^YxATh%rhSCywjh`fh1?H=#bJHnCcCo4Vh#bFA@^{qRBLVOsD_D1
zx2Yf6DsLJdccWt=Trwz>#Kq!j_UTzfPp1xdAJ?JbK>pun-(}`^eJ)v$Pis2!LT~T|
z{tC)26QQ|~y|?X`HA{A~4otdGma;YD?J6`)MYO;!S9Lki<R*jYjk|O8=Uu|Y4USnG
zDoJ&HO%8hs3@sMk2@G-m{_!m9)S(GoP@K=0&l!G_i54V|-i~d`kZ%w2f}-2t_^JQh
z<kruS%J2d(<fL4IQar7Kd!LGaV%076RX6#AOVg8xHw*W(?|e4woKXp3MfGL8M*?RN
z_^rc8$XpJwrnaBnt0KzM_R%zj2vOu#lEwN6ol8ZGC;FBYES8d)?BA*1DS4$ZDOgl}
zkBc|?3f~SUQm=wig<kpm<f1c2vH)V`Upy)C3oTtWsb!?BYNya81Ws|>0kw{7S?nLk
zFg`EB*|kj!niAphbWA<X65jDOH~;IhFEtK--4>nu<uP<>(b1yFL4<zfMpL<0^53A<
z;SCZZ9{hKF+S5N3m+jzBk6B^K@*Gyp9AkN5iGpX<=AOl84HAqR^bcAM<5nM#S<j<T
z*`~`LJ0Mqel2Dn2S@?K;=u;zigR6R;&3i5V<x-GYQXo@IwN5s14q@8urb1sw7Ej7c
zCEb^D1~?R-nwsxA1M)f{%19P5JO=kuti0&-`+2ryB)FAIPGcw;rnrFr3uo<=w~eEh
zks%uLce(AM5wfif>)F)laY&moqTzHrXdd~UnW)EFF-B;87KK6g=wHdAgifSI%N~6E
zKu~X_2bi91t+Fr&$+h<Yi1RB9wU1Ki0(47^=tbo=rr|Tbzb*1b=b_#;5KZ}6l$28c
zzY`1p(w_n_k=KjQ*G??<`HLfs5hwIRU7K2eFbX9`2fVFjA3yrkKq8T;nHvAe(?nWK
z+8B~_oN!b!&O*l?Q?FO-A-tvqcRlG3REb*rpPTg&xD1*luRk~HflkJ;Gz^nYd@IMN
z)&9GtgS{iJdQaN6O`5wdbcz;*I-H|M^!^nVucsfl($7wmBi*sx884aBJM7zAu=O=Y
z1<ig!=O@DwtL~p}i*+|YI~SEmuKhh1?_tg1tof#6PEYWnMmL0A5TQ@FKwJ#>w;7l@
z!h;-0bF1E904WEKU{bSGc+gH>_nyWpee9N)_8~pBH>2}<TtCF(6~cQ*&iQDwEtTZn
zinoyT_sr78Db8jO^)(Gph`X)L)!`~0@<}OC$sQ(!2>ua|<sm@tI7$9`HfM42mc?^f
zfNnOj6J1Q!NU7%YNS_e9h*oOOXKx2g_kuGsx0anE*XLZZXTA$Qg&y;?nS|&q<Luzg
zY&6g!c=FkS;$@qZx|9J)zwk`8!?@5!y_7v{0B@^!ax{KcbrmKpI4Vo##@PD^B5XKQ
zQfV6@@O*SZ$mMWUS-M$Y&UB0oD!Uz?jJ~73bIRg96n;b!pJgMUL#V<#V=G66QDqpg
zvL9`Nl2oRATt<WU=B@YMz6@2wbFT_=ouS5X`;Na!;_NUeks%a0X0rC(iDZ>uBMU`P
z+V6i+Ecby)P%$zwp}!>&hm#jgZ1QY2tJJgKBvA>F16fr5!!`&glW&lUhm+$^56&BJ
zX8e5ufN0DUezTj!Zx2oN*!8gSA|F%=@<4GkwD`rqaNPQb=6a!TZEI6PN(C%6^v-B%
zjHoyqkZjPxCqG`_4J|ja*YL)S%p2v5<H3wN^BU*K$omo+h{g&{?1y$e9dLN(abQ7b
zXB*GEFR`!{EdI`=b25Exg1+mdXmHAP@Fa(K$EbB*u8Yft@R^NU!5^)!gTLtu+Rc=&
z{GX=1>k_vy{pl?Mqgu2guaVgNSw>NR@LZ#9d?UP#CzcXwOdq5~@bN%?jrG6RBPNs`
zp`vlLc*+XD#A=~%Y$CbK`gUkxf~0`lB^IRsYCK91OLR9d&P}vDfDm$#0S{u~$N@_j
z!^xY(AJCvs`aorT5r{pA%7(t#(eqpKzN#ehd|vBol#>})Ux>#~JD>g7{Tp6u1A{L*
zp*sZW`25_s2R<}90ljMPdF$^d3<(}9-x!~xQg~JP?Y4ZP<7Y9*NihBnyJl45K>xyq
z*P4m6efDrx(Cd#YO&7HcHCEyoOZ2<9G{j{rB{$wECvWboA%B~=qISx#?i8m;H`Uac
z;8pa6t71_ZeMTTp03z2{U4u{2@5=64AwGfv8P6QAtA(k@5wO^-J+;v!69^!#jdsRQ
zXN&VyD@Ad5lf=3*+VWYK-JC!=uzsrAE5bMo*9*ke6)q(A7wb)iT1h_>322YJFH8<D
zE{}hY#2?IG^d+yvHlN+*<|HqKDS6lSgDi6eeFA*<bSJP-7R$0ieZtz-n|}|5A{9-t
zTf$Xs!@qSzpGv_!m2dgX?o4SYOOT`6d)!in2X}*9kC&Eytkz+JrpBKS0h*)B<(Na@
z*=LDxA{}~b3}O(}B(<s4Vprvs<Y^g72GG}0eRa0ZP#I0o2r#_)C3tiuxh<5XOGvA^
zD}aFyn#dDSt@dP<aZ)=jVA)dAB0K+H12!R3PQT`5S|P1JWKj`aM1IPvFwv9sla*q{
zzO`2``s9dTbBhsW;WEHCNK`#N65sAFHQe%7s#c1D+xzwU+nc01rF<w+4OGs&TsKOh
zGt`;PLS&7ZC&RSiI;F!8gsh<vG#zN~xlcB7`^B^?5&ga@TJ|J!uC=9OXK<)iJyv~t
zLyzNN<+pHRGo>IOgLQImwr+ZD+IM%7Vmm(hs|qW;hw`zCqw^aI!M!SLA<Wi%avtf=
zo5V)wLj0SDJ~5PQ#Jwli^v9Eqy9?E_<GzjR-uUUi&PoMqLXYZt6NI(n6j`PyuGHR4
z2Im!tEnO4<AV3)DlmO9{=S-3diCXODjBgf?oKwwBTEk)yYFfdRpO9Vf?b)bxMRD&&
zvBrFK%e^z*9bG2x^?jA}Lw_UUjWTU5YT9}A>^a3z3Tj{9S@lkJ;**`#9v&WxqFmEk
zrQE0-gA|i9pr@U6s5<3sH-|`PMG1YU$n^esGb;xY($%K6Ul2LX$oEl>L>vH@iC{{F
zcm1wJANR-|FAuL{9GQ9M5hA*4MsuD+ZV=0<LsrnJF<Dh2UrBePG}i1MZ5>~atHoer
zh!=gA|J;#wU?Q&ajad#0>Yk9NC3QB^DR0y%ITJ3{PSQ?h(gLqfXz#054TI{bfr-%Q
z3Ax2*Z*c#lpLDhd9Vnf*UizkX0DBl;-vA@#zx+NH_!k5!7O!`Fyg7ZpkcwwN?IHcs
z+~X0%K9KXYSr|QS9Dhjnw#3dlsK2^X&S@A_>>|99HY$BGAYNsirCEEdrlS6`sOV7i
zAECX-2-Egyr`yrSg~D!ksp7N|FFTgQ5L54rART*7eSh(tBAKW>*_}=5No85<yEkaB
zj$)@`U-$AIy3LCUY8-fuY|;nqUc~?0$(c=-H+^;^e$DVPiMY9xh`rKGly)ql)uwri
z?E|v7)aPX$MP2I|4Z|nPU!E|2NzYD^48KZHM&tS=c=OB)`Quix*`ilb2#GaTB611>
zPYFa|;zp4x_SU-)bqNi-lNuK4N#M%2mRJyxhkA=#|8{Lh=HIsM@#$4WXVg<%#wx9P
z3d1O(jfs5LR$hDFY9aMbO>R7<HeH(}Y;uN|(mUNYF)o-91hOHovB|4rm7fK+vY<Bg
z2o`j&^}-$e&4muH9}T5x&f={0<2w{R1##QZas=eWLwRiakKrUM-Y=K+0_(TxVk+IW
zSjpXYP{MG+Pu=%M30`6|f6wFIASceXkfHsKPLa^6HF{!fDW`%~9H&n1!B3Hj{zbhS
zGg6Kc(ZZ?QsmtFVk+YAGC6Hp;E<}IV9}+wDKM?gW#`s%Xw|jpADt$OX>8y~cUeIi`
zy;Xt-V4DxV2OY@cGBq$r4TvRBti)YllY7uIk@w(ZQ^mPPBAQDTQso5XB_P#M)u4vj
zDXn9NM+QJC?wC=Io26G&9;8h)vCqWjW2|iG`n-FB5DB~8`Nq*}8h<}h1T;auI2M$B
z?p3Cr5Rs`^PKU7zj7;@abso=W#K9POWKU_jgl)zaxmNPjQXbDv4+CZ|aPOs$FdNA~
z9NrI3ehQaJbIQlnY(EJWH+{+MXY{k2nltKB`D;=?+6zVBj)PYAlxY)(F;mf(k*R`8
zcO@{QG?e_iPv43<=|$sJC+jL>$hkJFr!cZ>VMq?5$(V&Cq?pMB-kvEZmi{;~J7M^n
zamqVKs!0q5yv#(38U7n>hG-|52+G)y@fFMymbc87f7JsTdH<2gF8C0yQ(XwvWE^7!
zYxa>3Fb%j2u$@`5MDcntw@LC$wc-{ZX)iY&2Ve4;muMC}$ZGeEhc0%qg;j)E_NT#x
z<LTn08IA}x5Px83urF7}_)sy?`^D9lvTkI1g^Yy$O%4BMwPM*7A#TwlV{BUWi7pu(
zYZ#qqC~7Pc#_7x#!8e4$YC-R*%5&qx>WuF^^)H_FGgnH_^QAvQ0=4mXMoG3)f{_e5
z<>7;YsFt5vD`rI$Ok8OK#Tzm{WqvplvZq8D*gfNm`i?oExZK|k?d*)qQ8ekyk7i~~
z7goTNO_5Pmf)izKbxRie8JQ{@HoPNXJi%GOw6xt&XM~l!znb_j8W%2|h*=&y<w}UI
zESs^Iasg$Ma#71M2^4osSMN`2As_eNEl<<5&|0tZhXo$X?@3La7O1v9PnT9bj`gAz
zEHhN3E{wH?BGx})$=cnYm}7X&m-A^%u0$TMk`ebc0VjNymDXMj()GPkAYwe}5^7NO
zygDzU`Zq5>v0FeBL!a+-Y%}j@&zWjx4HmfKKRNK;aVkJN$>I4$bKcgxopve;49<Zc
zzw#^6&#seqoYM4VMIAAy%HozWD*Wt~8kJZJvkx@J(I>b;{72N+Jfg$*KGk(w+}PEt
z>XR=&p0J8^#}5YvquSHkmMM9_``{hau13p}{~1VmTXrNq_>X^WwQxhu1WDm`BE>27
zL|@VOD-&)#S8URc`YWa5S_#2FQq5rAjk_}%YAOCd=iSB`kT+q^<hkA%DA<KWWe;<r
zOFR1XpvV(=8Eqt*`*ts1z3>co>e$0kyXU(zsWR_OF%5Hc&sNX(V^ZwrG-00#re2r5
zA6athA%#mq?<~1ry|FI&JS|qAeAgC!YiZ2HYY`EY^dv|`$$1?m3R#(;pL{g9Q#b%o
z0VpB4iHZeOSfa>1cn7Q*(-mS@iziMLR~HHcq}QzpMe*VY((-`h8NqowAcAeuq{n~X
zM+i^V^%?yTawtL+vJtA2Y0+#|oQ38&#Hu?OrST~fdLkjVZ=@#^F3Z0(P1VHAvMf)!
zb(G5hBIIXH&NOV@pItBA{tJ`wBMtn>T9q;0zeckZuW7H~*)Rf10_$tpq}Q+Or&t*~
zb~A@Ch*}kuU>F-D%Y(?y38SCCqhENioFiVfT#2ldyys@(CRAiB-Z)CiCc>j7*+)WF
zoJ7Gr^FEV`kWN!la64-FUc<fYd=#CDHyJ$|k3RH_Re8%x%L`Y$e6L^idtj}$#)+%>
zfxM7eAcv=;il@q$yZ+a3qPFl5PLz2$eOP_w(}xUGWr<NYwRT;5_)bBDW!Kw@<anaQ
zC#L>~YDO|r#l92manE-qx>;!ANRnuF6^scygm(YFN|fLEvvb5wYz0k+sO+j7ZS>b)
z4XnR&QC~tq-$T8AzE0MMHeiRuNv1O}Ox%>>^y1<>Pk*$i?2_^)^#@kTNoZR!&+5(2
ziBIBnN^DSEZo$C#*^+Zi61t~Cs8&O%q7PhDekH!)uEwKG*S3!JvhYH`;+2AchI7(j
zk1LN9Dc|Uk8-ZfvSKAw<qYw6QtH=U1v9YPCe?EqDE3Z5SYzs_tuQ5)dxl{H`e0h#C
zPX*O(3huYye3P>omkHc@2Us@U(=?XK2@K*IHVLs057t%n7&+ZCMmMWVAlKqtyp{KN
z6QGyy;FCvUNZ^XV%L6PkW~jn|5tUO6jtYuFqd?ClO`w8b#^3DZ_DY-d=Hp-EC~eG2
zeMSoQCjG*D6*|)>UKk-XaN!Zcjpb3^h4o@0$K;>=lwCJV^6pK!I4*x^$mLiJB0kyB
zU-PJbz4Az5CBGp#(NUB=j4KFXFBTuxiZ!Ei9IA$}xj%j!sYFUL=6P61<unZjkK1|J
zML)yi6yHLRE7KSlJcv|RyRS=;lGqx0IDllW6y-t2HP18Iz<~cPk5Zq1H!ggX`nWa1
zFrXEi9o5A}Uz+tTM0v>)@ld!vm)L~tk9cpdb2GR@pF*?!H6w1rAd!X;v_7fY_XFRp
zdcG?UMwp8#TrGiZe_d7F^SypE{se%R{~2X^zH_@q*4OwEo>WHszegE|i+3#}^jLpe
z!*mc_Wj^+HRTIp_56((EsU{C^aqr2g@QBh1PX4DHMtg;zg|q&myDn2^h88^?Pv&bu
zZeg1)>Y1FY(nq{*+W&{Kw~mVXYuCmlBm|^OY8a62?oKHIDG5o*5m9RB?k*_-0YOD6
zVUTVZKqVCz6a)qdL0~8W!QURg&-0vf);jO+egB$;EMfS}-uK@3bzgD!(~^0o^3S8w
zZ66A2$xsmz=;(Kw{5q9%r(W4MyhmLOsL@Nukjvl_C`5Jr@9^Xw_Rrs$|3mz1L~g;R
zUj($_J^WwxOaC@}GPsXvah_bft#e*89y#-*R4J#U``ASqnDozNn*tXauPqqiLDQ}i
zWNiIKI;f@Q2^uDnv_~fy$+TJ8oN?0*AVbZQ2gS&T#*4;$rPFAeFTf?GJmJ_l`yt-=
zNJZfhf1v{FyWy)a2l}-)s44R^6>WOWlq3wD9b*xfui#VYC7A*Txu!AxD39Z6kD^dQ
z0;6Wh%aBr6jk09uH;oml%{V<g)y0kq1#fEwfd78k-HJlEr55s}-rW=R&neta!zi##
zQlpV&m{8aPv%5H^*z#0AOub}UaLHa)0IZE>lAPxjtl!VGf`JZv!H<&otQPBYVx0W?
ztTxexvROm|(1G%dNp{bE(uTYF=%P<+qmj(U8fk%tliC(qg5F`DwVJ;@#}o*8mxX&s
zNAVcFFqUW2$eb(hz|qW9<5O0*{hTIZzOK>rCM1nXfxj-um?Y|~<VX3_b;_gxCqGr^
z=f5!n<L$_J1^vPfIp7UKID`jv*MiqHw<V{N&s&fJP<L=Q1!LV#WF?!Bpg6jUH-7ui
zNOeLq-pSNe$~Z@}^Nn(aUxolrX+kI>HkDCZG%%MxoM>BW_$LJEt3|eIwl_&mUw%OC
zHD%e7Py}BNRWjuTulcI)<k2+Kz8*>P?PLfC@a{@>z|0F<>RU%QC>IUFtLvr_C!4v?
zhK-NHeC04NpTa$kJTea|biBBXGBUHTZ_FZv1TEc<Ui$pVMOJ{Y?>*yF+cSzG&ea&>
zv1l3KIO9?C1e#?Y7WF8&kGjNOPHSErv~gC&VX#Z~o>xBqOk4r)5j%MO`vSoiL5iJ5
z_TH(PTLpYWnHk8>9@&4s|Na5K&GGn8j2@Z{W+c@vzVh-f^j~%rSD#DucO*mZCFH6~
z)U9sW(RQyj3$$_D$?FbhO}O3Wg`tl~fvg$3!2{(@U85wFhcpN?Ol#I)p1Mkax#Q*}
zn_zUR!F|!G@LUt7$+UJMr}oYXx0M0isxT_<H*@?fQ#TlQ7f{UJWa{35Hz#3Y+^7>m
z(t~XYh$6pzu=E9zv*rUwxwS|cAxhOmVsnhH+<|X#2xbH<0nG$%*_W)iWEuo;dog`@
zsqSWDh)O~<sg5ExEodj0w=^U<5P|aG=g=F$dz3S)s^>{-CEB@wCi;n92Zl>bFZzKV
zA=swL&Df^iUUjovc`67nac{gN>MK$}9a6IM?qV}0Yr&5&F7YczL^A0>;d}^Tm<8+?
zmV))uliLO6+}037#6@11X9p8!UVGE{WLSfF=D0HS-R(NakMA%dW;f}K#{qK=@ksmt
zyWIquMvC!?FbJ0~i%z@3O)D!^A)O<G#e&yqqa5X#8f(1rqpMxHWJh6B?37QjtWPE-
zii@~dY@j66vsutT<`==d7&)3Y<pkv^JT{FrLt09+ZF-DM(Cw<06EyPa^xU*FCOEoT
zxt1=3!SM$%mDV}huX)DHjTchxw#TF20xDTiM>e+(M{f=#VS5I(pu<L|VXOLt=B2uG
z^@C~^SsO5d<2q>`^l1AcKb5W?ri!=+a`HkkrSVy{5TkZ;Bi<#baZ@tpn>7s6du|*1
z&ouJQc4<zBC$z4OhurrGr>P8ALH0?7eF-{<en}u1k1>)}M;@T@%Vgex(V4`P``CmY
z-`3qZgkdFe!^x^RQ`qF0pc*8b`z}XjrjkyVODMxxNZQSHmmuZYF|_oV$bi%#>tPi0
z<3k4cC#lvEHR#knHrz<%&n>w7V|_W{O$Bi4+_v6uAM7xr+}90Yx=88Zy3nyp;_z!?
z2-I3=*WOVueI<MlvVbWICv-($f(Vq;NU`?SG<p4>6;D*-g#OK;P^QeEw5F}e3<R_9
z|53iV;v%@h`+$?v#6qowI_|>g9Riz$AZ@(UdJw5#w0+Ln@XtQ$nc?`>7``Q1il(2x
zl^|H`qVjake?42jd-yL4S%sb`fyaKqU#vtID>!X{4Pvvvd`A$kKMmMSI6Lg#;4bui
z3uB(Yk=k?Y1@)N!obJC|8*U>H21nIiMJ9+DYWE?;4waQ-WkW#+70UmWg7B1srOzO3
zG?!^K*H>J*hy(+~o1eg3lUhSy^sVc<{R`qgNcv4xbSETTL^XNLt4Jzy#*1Ix;Cnol
zu8c6ff3BDLnP9~|NR@gMF(_h<pG^lBB>^b|C9u$2#WvK5Q;10?c^5dc;f=NKTOGY)
ztEsW$UbNNXF|J>P;Qp^SBbs9^I^2<h%&jMYHHaGR&QHIom&cIjWhL+^J2YOfzBdxa
zGml#Uv+7mJZi%A>)-*rG_7vfS*9!BSe<N3KP2kr$K&C3!^f<39C<Q+C`^@;OiEA}l
z=L$kl+`6aX%_gvp@UpgF0$x&x3#Ox(j{2MaU6SmXiGd6sWmZZ1l{BvpVfnH!SA)AI
z-)!);&$ZfyeB>H()+ZriQ7}>cP)Ghphyjk<2+X3_0_7;R>o==DHJG2E$;)OO#DcS)
zEv}2oSLLG_Q<%JKedT2YBTP@@Q-oqHrY!YYsUJ(vlQY8w=XT95BbMcb8E?wFWZv*h
z(!v;WD1zWnGcEfXlrCQ(!&SRKNl--7j`>U$;pp0KnD$aIN#gz~h5NL#xL>iL#Qdct
z=ayYD=xspwjYxuohfkGUC&-qZliKej-;9AR%!@sydX^lLA++_?Rq3DrOL;}jsZ{Fx
z_%Ubq7`T{76}-zrw9?7))PD^TZOWw{{Dt~WS1^y}v5p$!&mXaapGp)E+BK<DK6_Jo
zYv~zTf`U7|n%uR>>r4reQOyS7w8c3!2v@NhJ1=bBq%pEsQw=e|WKK<fYJmNjG_Q8!
zKj&uOEgY?pa>??gclN!e`yF!6METY|sADN?zOdxS8BcoiGDs_V+f0I3rez#c>h&Mo
zpr}03RzAywx!WA?ApD|a67Q8fH|W0k2b^CbnPspvN04_ya`(5BoWMf;dMJY&&xMp2
zKRMXhC+_Qqn{bLmnR-om%lez5a;!m=kEm^}l6kPaV%DbHJqU^WV~+4fDaC6nuWKMv
zAsN-G$=6%sezt$U>p9<o@h)pY{<s%5@^N9qb!PBsz@n@hPU5Ea-3*UF4)INXqA%*}
zfG;fQ=`$74^+gjdazFFQK4f>Us;<KN*^?w2k2W%EFH<h0-;l$GGOzE=bJLKp;w2&(
zpN9aETVKSs+y5#6k?_<j8h8Zb%%&6m3is<>9%<PT5|CDH$|}qK>-aRp9-ki{)!b@E
z(2A|fQP2Eq992goGLNiugkG90_u$iLP$Es?85YbO^eXx<Nc6Wof(<x0KP%3DQsmmo
zf1fkV!y%QTHsH|rgjkAeIsc1`W8sJyoKj-y16EK#{W#<CD_ndYxIHoEl`F|uzJLpr
z(E{d1=C!_9h4Q%{V4xA3S^H|`#);g(le&;K{?Izn%nurJk-v>UV<x6gNMk5<Vt7>u
zvIwb*NN@hpzf@~<Y|I?d9cs&<Nld?;OU}jF8!4gM7!5%>b5Xz6=ggMU$<z0z&OeNa
z`0N5n#eAa^>iXr#(Ja>RK|^`KNJkZ&nCPeDgMKT3bV{eJ<dN19&v=hELncU~qxY59
z#+n@`r2JLwB7-bxch|cMe#htE(K0C#5@08BCuz46dpSmdON26<!wkuC%A*KzOtwH)
z$b)s1$s||JPW+{4rqHgs>H_P6hFtb;h9ZPh{ZUk9<(uGCd;Nt?xw5frsV!+(?pbdO
z`zK>IZ0rcZC@G`L1OkmiHPpEq>V2;GG8}W?qf)Y-gQWr=-u4#pUcSi+Zgp+OY)b!Y
zT*qEhwc-af<XzNbRutkf`b(<xV!pN&@Fhe6tE09<J9>IB&wn1-rh8r`?35Cb(lbQZ
znj@9?#eV-Gqfdqm3YnM_CzM9fF4TEw!l9<N%`Ng-7apmA`!WMjj0=x{L8z+`Tdu%2
zO$#AWy*;XKu{O!c2A6A-@m>hu3j_R})PJ3rs2oBWwQq!D4mjn)L8FyX3UC4#rT3%U
zSaM#KD^55#K_3Bc{jToi&vHY^`I?nbeLmq+stXn>VL4%R6YzN!S9!%DIH*KARobyS
zAI!Vsr>zq}k(wO&d+Ld>UB_HP%7zb8|Ln#gFVT;~-3?`Yy5xx`d4g@cLyBA8$TVb|
zLu435qqAt!<oLu*_T@<RYboLr%RR*r3NFv^@h@qT&K}XP;kZ*3vm1C6!GAn)Ki}Bh
zFs_UbQoWb5J?<AN<Pttw!L}}pYsg&O?XoNT_u7b+JYb0<iyS@nY<PwbES<#2qM51n
z{@D}{@xQa;3Z1Xp6$GXI-C_B=L=!Ed%fk&qBAXV6Qu!tEZqIh!PXT5ewOL>>KZiwb
z*94{<p&&`_MiG(!i<w(wekg{U+C%@E8c(zgZUZkI@iLiQ0diCI720+v!G2hF7V@0p
zogJs+=Rx^xJJx`pJmUUh3hl*TV|W@!$JX22`P-a^T<Rmr)Lntkpx-71V%tJQT5l=n
zJBs;#53T>kvW1VP;A$aGso<8(i&IfA$YR|RThhhvGw#wohm51~q9~3cPRxM>!le9O
zUmWqJMvDIItF#<d7G6I&aC~I1)6~N`9i762>s%|TQ;2c`GMh;9W{6Q1K|%%(xTDO^
z7CF;q)DvEZ$W!wF62c@leTP*V{`AY0;<_l9i1)&wtW21#-<{BUxqv3z=H(d0)9>w`
zLmd19yfWr$g^w$;n>u!_NhfV*U%A*^+{?axe`kD;-G~k3@hOfMy6j5kQoHyxw`wg^
zrkQHZ_?%rCjcANevD_Zkk@ianslJ_~t$_ED*%V^8byw8ajCJ*UMSfasSj&zwPrQ_z
zCe!><bB8B?U18qH3rv+EoBP?$Z^b?Q=*G5E6ve=f!!y{^2!VtE<wFCM;Hj3jdbe7O
z+WkrMv9tOa3sOCflY_3#ZFm<R>wxZ3bO<|H#ttKWh7zv3TKDptnF`{W`gIcBW7<;B
zno{G%({g@{xj^tCbB9>fZFA*z9`cp4*E}Rsv36H3bBIr?|BMdf@+#VnUDDLCt#f17
zSn1l&R{#GL=Kpa)`**ee=O6EL;_BaNgb9@F?DdJddpn3z;Tz>(DX7=*_u59J#w~1o
zV&MwKw239Ii34wGSOfMSG=nMEhqqsX{aGUauz$mBg<JkA_}88855>ckZ-46d#Q$Py
z|MNHfFJI-~$2XK+qggP?+POWi_9^=BPZ6iY>7Z17bbrhX(e0ac@SgXkg6OjQPauJ+
zgq@H0f2Ozf&4eEp%Dz0zk-;^Mj~J_Y*^lZ*|CO$JsrMvCzkM^H(D`5O+`qcI|K+8J
zSv3N-iuD)i+PTvW)F*M^Iu>If{dEShE(d(zgRQqeW_krnfp>0C2Ym?aW@E5$6C&7!
z-&}#L79=@306I815U`kHg;{bWkEPDD-y=$~YC`WHx(U%<{Ncb+@2_fU8R((j3O;uY
zH&i4{>v1D`*Kvc4JACw?3ZP=0rRS4$J?5XM_@74fe|gLQe9tts2ODVh5U|c;vHq7Q
zn-Z{w852E5hrOq}A+DBy(;R$qG4K}&-&FaA70q!!^)*Q^*1GWse$@D+^L<O-)3m5b
zX>5NM7fV8aoH+r!2~X=s`D$W-7*Y5&Xy%q51hg+mJpZWtgGgTySE#Ov_3@wbzwhuZ
zEe`kfkDs#+eE*_t|HrraZ|vquBv!7m3}V3^yBq%WNe_;}ig#3?hkyTPk1sSLeg;d;
z*QOJwS_AoAZh;`r(z0W_yZQ<aQE?8LOAc<#0ao_}(yhccv)EW7;nCn@hnvon82?>p
zcR9Wd^nRuY!vBBy)c^k;Z_l<Wb^%qtR(SqjKT0H;;ula_v0^*+y%*Ywd<vM%^kP-s
zgG#Jg(nhVv<%12M!Fclj>!&A9h<$v;jeB7I{`<G6kl}*3YeKBk;nc|hFq&nPS_(Pn
z@-$NpJON5kwz*qqAX_CmX`AhQWmnS?2B_t6riQXNJ3wIMb!_{(_BhPRANmj|PGhiI
z#?Aq@M@aUrH@f6x$_2OZZ9ePYL|qDMmoei00e?!L1qhma2tb3_rOV@%$Q_zp5aJgL
z?iGu4^j>Y7Rb9<16mGyf8}kqZ2OXE7oU;x3Ku+zK^RMU_SOHe7w)dp__n;lty{}Jd
z!E2h6FMRmFT?GHvdJ`#|7(q?A?%WVm_3u&;*&5#iw9M1e&q6MaW|`b~feSMm?^~?M
zwjY@P(SX<RxdW`;wI7)HKCr_B?xmZVD@ZmuU@zQ<%|w3)GUMW~k<#63>yNKQV)02W
zsbApkO^szcV|5}i@d{}6#}HR>c@p3;7ZwO)q4ccaQIBuOT-UXIa{ACp9N(*ZhbScx
z8|}`7mGqK@dhRsOULOzo{5DSM1e*%q0ff5br9cae8#)xtdI&i`$La*Zv+HI|+yg2m
z5vt`0<oA6%@MN5>aQ9(#x%GkjEA>M+Je$<;SH1x;uq%lXHj1ep*eO~qE%KR!&;Q>a
zSPb@o-F65v|E~v@(gW^Y9@H4#^uXPDr*y?WAR8(Y`ozjks&?7OPq7RVk%@LeV4j!B
zACN}py#s<#jpTPEIIgR)DtwE;8)e=1e9I7t&2PhL<h3K*zjcZYo>$m*ATJt}Ej-oz
z^QkV37V8;(fF_1NM8wq|EH~aqe8x9A(bs?sxg}r6W*Q%6od997<Ztltkn+E-S9=F)
z=VCjE-fbtm1my(e8E%6d@b4_%$$|(G@@Q->UO(tN7%#!v1A~SQffFE*i_?_!-c_%+
zfv@F$u;zexf&OP1*!5mBpz}vVOcpo72GXm0={q+xmqB7)_+>)r|FzhN)KP-tr$Z=g
zyyM?{Q6!kM=X?Qp%*cuP)aOs1fGh8^UtO@R)5<ra@U!VXAOaLo040ee9sb@aY4Utm
zHmV1!Vy(CD#pS+ks0wd<DzmdmUHVI74r|c-EjjyCV-~w+3LarB!F4WBAdV-$4=8^}
z9!}|U$6}3g)4vdArgl3xJhAmBzV~~Dz8_RBI;_C@in|DT#w=E${yVFbp5jV(suYSR
z;uHAm{zS|!0g+(M-ktW+^w4dmeiFMq@aqxLlS}`XLp)M84!cGyBZK_^vqny%dVImZ
zk$gN*ECpAAwi|ok?BrAco2m3@ptm(E{UtyAPZozrCsC8FPDJ$`g;i|Y^3%nX_bYX)
zodvLY;y~{h<cR;EEdlp+|HP`{wSREY|6`qhd=*>Ja<dLY{!`E-iN4{Gq|3Rt-|J|_
zmdhsCkwI|My?=WF<N=3GYzeuEjY;lg2*iqeLv~Nb?2a2BwzPj0v;jB`Vr>EF+Bpuc
z^>c1xwe{dgs(Dx!{I%vJWPSw5(L)!Gj{vLg?VAnOUp<6z+KOAo%V2$lS%RW8bUY-^
zGN9T)xqQJ-jx}%&91<_dymv$94^VGATXU@H#=oHe5x-;D0vA^YG{)=>iBF&pnFsr?
zyW4weEMij{y;=Wmd6ZOLdf)&@AFz*R1eGY<O78Q$@Qb|xZY3Xx!E|WkVS^dE1t|F+
z0x4pFwld;z$c8%GjS@^wbqqEL9V^4-GvJzQNEN{mV5h&`y?^EOZRy4P6-$uk-LH4=
z%;do4&W#aUdfp-sbu}=B6~C092)VR@+7ScPQP^Tgh*0Oct&}qjs`S@Ake=6%VsO#7
z?Jc^=npLtF!~G?9<qz-$-!vke(l7fm4;3fJ>Yed+gsyJU!AI4fSW*cuJPJFix!1hg
zsK|LlZ52H$gh=EZ_gZTQI-irQUjGRI^NsgopTk37!iSP7cY!5IjZ0-C%N6#7d@5Ds
zgMRh1m4p>qOk$U3Cv0^8_t7hh&Q}55fof-1)Ac>DvBsUb-0#aD0F-0W0fYsRVC(0t
zHM4~f^rjo1Cq_nwd#8zP3+Gp(8Q-IGccb9KPc>q~Znk4Jj)g-PALuH+?Sr~Ky(Qlm
zjnVu*d5Jv_Ewo1Hj&ee?;=R<9360afaM+=~0E>Z>y-IFH-faL*M|Sa&BR8>83jv93
ziC?&H5Az0`HDm7zUVvP=EIST2+@1f8e`;iNSOF5y<3mgRlb?K`2P(_YhM4t(10hW2
z3X=t!&hCdbC+Zb@p>Xz&WtWE_!;>w7n1%3k)<s2WSyJ@s);9Q5R+P$FLYDzoW||{U
z>Fg7zd0_xYu1qZ6xchUe3S<y^$HpseJ1k>AQ}@y*Q`UYOS`$_7>4%z4?r8iAeg;{e
z`fIY?KmXZHBGVLsyTMDIt9PejYb~Yy3*6mL4&RgGD*-PFf+1SC%DSV%j1(}$l?+`y
z2O57)`<87!5W9Z9{q{qZZZWU`uvk9}vT%QDuz#D)Q2FAJwU&2B8(7fZ4)zK1mtu3(
z>+Jg(1PIpi3ReQ1(ud9e8MQ?kQ|1d|(thjTt!AJV5B~_h!a_NSoSf?Ee;-J030EpJ
zv1`p?B<tof4tbyo7B|I<!{y2?PytS!vYWu`|3_$~;L6@5t`&~SGBD+(RRVh1SRma0
zh@>3ABVLj2K24<*Dgl7QUn>sZ@+a$i%#Q&(oxy<uL$><|Y$Qo4R(I@i>>9&Plv)wD
z3Ax?{knegN?OPm`JSi!JY#|TZ?gS0s&12>BXc;1VhC6@<WYb@&ivdfZTWdQqun{=t
z$ALPs_I&^0KGw9N=ixPC8VqSRa6vffPEENgyA0UsB9CY_^0Bv@6%<Gn=36bQWx}RD
zI9zeSA%zeFv*o8auaV<e1CAc?jE_mnFz;aNcfi*d7oU5J#jXU<bnhRSYlmKf51jr1
zIHvaAyDEtnKN{uD#n&i#rq3?om=G6%PrMJRz#GbbNTiC=Mo`TP*7(Dr<*WZ&G1qo|
zqVSAN5LYhYNuR+zLg+#yix*<IpJZXlGOGL7tC@^)6I_npQX0NAg=5`5yf|;xMveds
zox<)MHTyk1rna8b#QExgvJTrf-B2zzo8g3%;%Bw$edjaPN}u!y2b>ei?8)F%Vzl9d
zDxpe1>245)KFQF-ZfG*$Eqo2pQKhRyh;ON1)T@lgUkJNIHgYS1rzV_QrX#n<sU^!P
z#tl0KC;w5;57<Gb_S~J%XrJ5CJohlStRmPj433<B6l&Cd9dko%ITrAs!s6^aAe&Cr
zTG9B-e&UWd<C>76U->8;IOz+dvPQ8wt8CG^jI;Tr1TG!b1~979%;O0iCL+IMJw!s$
zuL8gog3J5Y6|r#eaZ0qRmIi{Pcm~<*peTiB-ts^-nmUosozONMGlE6Bb|Tj|&tKG`
z@m6F)iP3dpVy^V35JiK-NCmdXM;*jIA|v-idfjdnMIG9aB5e?2KgA|2I@T3;8_rTv
zVx7~j4~t$!#v(Ph1dh^=yWJ+sBPFDAsclDXQ6GURi^r+eyN3=kKtxJFw8DGOOna1~
zJRDjx-Q~9XJC5fAnjzb<%gl#1D(ve&d&8cJNaVgN4tcW15b<IB;iN!)1)J0Mk7q2n
zDbfd;v}O$r3^Du4B((~6I3B(i9iDTXJH$`E%fKN{=fPC3UEc2CJ&=}I`$49@rUzHm
z>*hwHfYPAUJ_zD&ui03KDKv?0Ki>F;Kh(N$n&Q&%^8JyB^%kv@y4S$mh9xAiGv-)o
z7Weqhcd;p`2_}ArmdS)w2^iemwKbTn|A1A5Y*vHbh%r^gRPimW>V9$uM``wFUX<e|
zU`2?NVbAC$s{O<NoY5@Mh-$75Cso6UrYk;#xWoejMAp~~3(QkSA=4q>5C<f#^C`r4
zg2E<gLN)9E)z5<C^kx*>{)!)AAC(46+NNDnaxOB%Vvp+p#79qcfIQGBq-t@XUH|Qq
ztS21<366nLH*V?^`d|YrL{?pHvAM@-JOb`KReI00b+Q8c9K@KT`7~(FeL(;;!ko1u
zMwJE^ik}CJcOGl)2>@;hwKv{3f56Qx<bE1+6Aekg&sYj9ZBl*MSHnS#5f@Y`Ydzbl
z`GCiYud_uqK$6m8kBzS`T$Y8|=2w@F&~I^B);9~rKoiUZ>H8+g7ximZEC(T^KL@X<
zrtDDY_IuhL6xM^F490mfN2W?nBBDU`n3NHGJPk3T5L^6FhN-JN@!r#hyxLu{4J?lY
zPkX*4iIS&Nhp-8N$$Y#Wl@6Qx&2);qN5an(Vkk4()a=)kwL)~N;TyPmD-~ZB538gZ
zjX{unfW#vL_o#%#-s|Obr#<aVE^+(HS4XKx_ev5{AgL}IA}f7m<i=d&Aw>vccG4BW
zfx$1)d;_=ks=SoJ)yZ=H`%u4v5PAx0O2i?wLi`Bu0Nyr6{3cqyF2)bX@$9fzdj?a&
zyek|nq#<<HMJdM&1KXQnw79dOh{`cX07xuyJTSRdz*EZb<Cfj?CJDtXf26OyYI~Xf
zOH&F$Uh<A?RXRtis5Z0jo9RKCPt_6Mta30MyIB7QN>z5heO6~hlq|u`G-M8q8%Jeq
zgv}bc0Q<7)LjqT4h5_}d=uAcvt7}aG$athHqA6BG>yM@)S7;9H38YEwzCh6Rn<YXR
zRc5ebbJANwywsTWF(5cv9(nEXwWG=|pP?a5DdUOs1+HCuL4N;Rp&dLE<_n!`FNI?R
zBa{e7ay3~i#bc!&mFu<Y+3m%>0|><&zHjFHV(jr+J54^-GI1QJe59B&4eX9Q&Zy4#
zNW>w8!T%XbRN><I;tjE%mcKOa8Vii>qGO4^?3DS)bn|+66&p^+euc0-i5?CM+f#)T
z`;RhLpt?$|I^*a2;Y2Ft5pz%aKw$6<fCQyKLo^(Qe9heF<CIWnpN?y58r-uT(wzIu
z;3}5Kk0Rfi`hN0?vL`H;?#@0<@-X94RDhD<%QOf7r2@gDYv=fD8bJ+uZ)=8anAsc-
z3`h{{dX8^<AH0uRQm}U#^uNaX{)s*K8r@5oPevs1Nqpfu41tu#$hto*7tFit;t*M-
z$SD8z->5R*d2kD9(x4g|jC<usdAwE{WA3Cf>So-qZ>;nhO$jD;<y`g^L=naV1(ECM
zSbnTwiSa--53cP}D5ZQk*M|zk_0+qmG?JFP?ZRT(ImQYe<<g@fOp@>~0o_02Q`(KQ
zzx$=nEK*MnV5)izcOLsm4;%CY|9|f#nqEGSPq2G4P%%w|?S)Xd{*U%2VQzi$qn@{E
z8g3@ZWD6}AYqjx2yV9J#U!hojyL+jBj`LIOiZ;fU^u~-P<=66rh(YVlr5Wz6Dk4T>
z{5hS=0|=E7&4_>p|IY76yEgn@704uG>z5jtKZ0guDO>qQ+tqV2c}G1nL*ttap~&g7
zLE+UqG!Bo-QWKDLnjuLTXWG)=Xl`wT2c(?ptU2W!{?~2Bg(u=P6UKy(J}x-wYKWOz
zGUde&Pf1jTAT8-Nop(azl_${=4!~L^1lp~DX-x<|R@y1!>sV#kO*nZstoZ7|f*-mZ
z8HYCxsnZy3_P*0>Z0Avw9!dVAPo^<WDpsF<QqrUG=2S_Ns>T|RT0jnt@+i;U4p{fX
zXw8()@H-yIx;5(s9ZiE<WF5RKu$-1jo^n{u$p8-H?GRP82*12@6j8YGz{m^Y`9-HD
zpRL;k`nSjv_vkfmC1I}JSlgU9lQd95!Oo;nB6HtW@%wp<9vW+`^WcS6vaaD+y?7&D
zp76?;Gb}s#^kM}*9n$oPFljSiuX|w&59+K1PcAmmk)Wh%CaXIlbHOAHxT(!<mC(ry
z+%YnRwr5e>B5hG*LFFA>Rp``^8BU`dZr5oSO)6%$ZX_}6p4Sdts3qrE=ISxwzf75!
zBN0SM&yJ&6gH=YC;#rBGr^h99Ukbi{1H<N&O>(9jI)}rs$1I&4ZY_o0mXag8GOM4A
zSD&<qFm)S@(upUr%6r_sXjR2Wr8N*Z6|-2}H|^}RQ??k^o`dj`z?Y?Ep;`5*>B-Fr
zuJz;&=bxgQkoi;593s{fZZ)om59?l80snP~UB&4@{Wr!zq-@d^XA&d_YLUy%f8OQ!
zz%^K?C&7#u5tYwv);wr-omu2AGTXogbXw=Ekvc@8S8*L|?H@7qF`?6q)vf!T>9t<_
zOkbhYwi|_3A{ws@VcyB8^q{R0a4-m3_6{?~M)pc1|5HZI`*J=a<I4C1DZLb?)Ht{K
zRZ}H$l%JGxHx`+TkW)aX9-|6px9*)iP0QGY*n7CR^QsCc62fm=2OG^;lKD_Tty$*D
zpd17OI^-^uZN^+VTI|=QdZ}K&D9UI-^9p+WSi@iH$qYf(+|PN^5*cjcruU9%&|dVK
zkBAX7yrMHp7)1}G>Epvx@HKtVw(3|#-ay4ExN5*#YjS8qqK~WIz*Ecg=4sTK;ASbF
zUzRAoj)<de;VQ2~SgaD>XUBi!F%rlf_u-)Y3E3Q-b`z8`qf}+iY3bFmewnp87{DGf
zmR<O1cct@CzS#$@jE`xXkYIxyMYc(NI1UD{;9?$iinYA@p=>#4OgDiy!G!jC_|4zq
zQ8`JgUbk?110D&KfXeofYDu)$@--5oSGaORU>fJq7+h<ZVz&0R?qA4_zA!F`b$Uxp
zSlvS8(XI56p+*YYE&Exm#dx{!kn;~u(Dn=WqjdsGH~FC(DL(!-1qMeRLezw&uvXAW
zL;Wr2M4W2)3E<DuQM5dc(T2wxE8>?6l=&}-ny@9)Q);AL_XiEE;qZhdvLGtL+e~3K
zL?tNdKe)(G3JOMcq_ycp%u87M*fC5jjR3B94Z3v|?oJFEms;336f@NmKW`EbBl)Os
z@E%}5QIY4Tv9YM@yZKIO4T3E5Nl$u?U%2b#J6Kb^m5fRtL)<j-;h*{~zZoKj;4@q@
z9QK*+$&2z5Mfz)+MRr8H_~=<7a_^s;zlmyb`gHEq+LzbqCHjGGQd}}2!o_;LUXsVU
zr}3#^<nOngV%G9wJx<p9d25k7>T8kXh)@(r(HUbUh9=G*cR<Mwzjgyt$tayy%fAoC
z$)VauU&q3A9I(TOY?v7sqPbert}oe>cmpwXVn4$C{qEYj!U5Lu3}GSkAhi?BY_4J~
zTVl=3I#-xr_2^JUFO94$a-j@XFw*?$;rUJrVP>b7-pGo$*T|rGHdGrWwK#a*AC?Tn
zWsI{h$$`9vraH%`$kqEKHi&(fbGj|waVXooVK9)kUgIu)7zGm?qw8@27m?8wX6?_#
z{wViI!N7T?B*Yzh02s?GGy*3bf)W@<-By?EsN!grV9dN+{Epv`sE$RHxmeRlOUIgE
zZn}wb?;8KT(<IoB(Y)bRf-^(-NmE<NS!<ij+_ls<9`j-OJ0?GHtu;*lj6BA=5KNl0
zz0f=N>gE=!g87X3^BW*09ZCG)xlKFK)l?*iS4(acsoU8bcabk>G`yM-{RkgqOnKUD
zZt=#gpH;r;!5Z<;zjOgqr|9xFV3VcJeTEA?7&OcE{@bM=kmC%QVN-Ha1t+e~mn3%?
zB!}QS5ghUfYHcwy4817KXbD_9l><%1$bBI!l7#?Eu7EuJSHILpCoeU&X0UAYs;Ui?
zB=*_REFIOmPVqf?y_vxn4JP#D*P+ssBF5cZnBIZ}8ODxPLqkLW+e2UVH5+1ci4^7a
zFL;%ae!w6-y);>!Nx!J>J+t!R)ok$WOF~Q(Y*h)qK#mfGx~W#AE;kXqX(A1j6}A}q
zayRAveg=UrnXgmV_&fGl^&qyAW$x9K29$H`{>zT=lk$m`fh#`;BNS5}-NLyruzLeL
zfA5>h;eTxl*Pm?1+2QDhb)>%&QIqm^m`h}k?gi(bf>Vw{nUN!tj_Tq>{|(RAnPHz<
zi|j`%t&?-36+*1qOhTzfhP#qG%61G#>9T(`U@jjZZf=X&4<M@0w#hF%hY9Dd)yHQL
z3Mw==IqpJ)QIb$E&=ucL7L1Bl7+@V_5G2Qh(3PCNcRdyg1@4K`vWx9y)KjmnMx5T1
zuil<pWR$<L&;*^WCxp0u20+88zY>M9Hhp1AL*eLGp&wFCfoMStxoov&l6f9d&Z{v;
zdMEwLP?NE<(>2=}{uRd*_3&6&j-0M-h27|n5EWLZ1va|CCo(UnF+Hvb$HnT)4F5A6
z=Sdqjwn?G6to+8ifamD#r5{d|XP-<)F>ga#h_n^F+agu4lJg{#zyGjEOzYkx)TKQp
ztm-M>)Xri9e`giF!x;5<$&H!`@y(WvKbrwkz@f4%7IA-7>tsxQL26X`agnsOq0@8)
z+kn3wn;xTab@Q0c6;Iy}3TTYPh_Rr(#M2C!iP`!e_Wc#MBCh82g)@mX_lm@JOvd#$
z-GjDoElTEIz}u>=t0986XpZr0fQ3&-0^qkA&+kvwf3WUYA{DJ3k?U4uYRL+&jEu>|
zt|@m&^FY{`h<FT5b}rz2?j>mruDl)L$w~40GE70v?gHi<LP|<|*4A2apfrnrdDu_Z
z9M^j2NZs!o^u+0ww`utmpOhj3(yed3&W2_$pn(qhbD*P%My!de8-F^_;U}mp&}Xhp
zbCaxQv?yrmd8k<^iZo`OF;4l6IhM+iv9yq`(QzEPk{Wph<6tX;L2|qa)gQ&4iB8(K
zPjxo?f4xWF&AgLi#Pi%WdH7Y(;NwJ@IefjvS^owtsxA;$7S++}^U@d`h+|}QnjB`g
zZks*TIbqk07G6tjAXdi)jk(4R++h>V=mb4Kv3Rb`ufo|5E_(_N@_wWbm}#~$4vdc0
zrgq#4lv3n^YqN>HhZrhr*@@PBF#Q@e;_71wEk`_i8B+U`0G2Vr4`+TI{5pW#YFI<j
zdHU7aRInv?#?8!VeUm$}Qm1Ye<aeIC3fXwqkn7X?P;9TkL?*to<BqDk*2y$Ub_#x<
z@DJ9yw(<4dk4k>o$=EM~Rj}eEebR|^Y~X2gooM73dQPZ)TWp<s$Rl#E2#o$jswpFQ
z`ZPSIaM%e1I}fK2^}+^m9yMLhNJ0xE2J$!iNQ9d-rBmt{?$sls5{;eruo7g;;(mPL
z?6C|{2Ey_p^)Rr&LR2cFNQrc#b`OT%r8f}BF|P_u4aEp=MyTMpLubi>d`6*>h=L_I
z@xfso-y^o<yBkHXa90()I5*ZX5?(>&{yFtc(W1vhhf7#a58pM8clBj^fJ-HXTd1(R
zaIXJA%{=0c#%WIGhrr$PWTD1KiW_CoUSG5o1pFOB5_ckF9Vnho7;9TunifpBzZ0!h
ztahhLS2}a!_+Y}_TRP~*cCn!yjLxMIj^Xr`NGK6_jX$1kD97s-kbaoZT+NXt0~JZr
zOD1{QuAi?S#r_k($zw!fnpt~4>`vrr$WuJoua*S=Z0!t`&E$6VtSB}`pn7}h?Dm5#
z8R+%4$fVdYq?0kCY-oNI>t>Z<)a)G!!Ui0Bl1K`@^Q_?q{aI3dzfu}--33Unpx?4i
z!M>1-R*~kSib^yQQhf0EYNjVJF>Btv{W&8N%xvcO0G81x{Qw<W!frc<bgLUkSCCyz
z5f2(s%`-Ndi0@6`=Z32sL(X5afsi59(>qil!z@nCneDcpU}yYj=^c{aywXG+O~Rvz
z`*f>*C<zywk|vi?rC13i57W9A%HAQXOfbqTP7Ek|1Pkis4s({si8M`S17CzP!pLg}
z9UY_+j`XIxY)qh0+YEtR&+^_<c5+2=Jc@k?fDn|-4XcFdm{%&DZqR^&_(}5-^MDuX
zbFGgYMVzI#X!f(UoC!m{M17yl(yV%&F$Hg6Ed+}eR6eR8I4?F#XGDGR5&}SSp4}2R
z3So@w5TGI8K2w6HlL&mg;^fTju1croT={0oyS#-H!-k=LJ^3`F`n`{;$#7cK>w(YO
zC^&r|H|*}h6~e>sV(lK&a+xabrd76E1-yy|=~(B4M&;+hTpKi?NUaU0r}Z>JAr4#r
zy>G?A6%)~5tmRmkcK54<*}l%*MYz-`;A#D&S4X|FB#2#k6S)M{)X4NIPYLb}#y}~W
z?~WlHT=+j5Upg0WcHRqs_B+l%>VIaUf^EHuL?7IkJ2NFfp?u9?ym|$??kQ0GPhF*7
z2GSeN8qCdxe%DE@T=)J7W|782bv%8*)nPQR=LUNIbLrg%H@`b^%nL&0TW@Ea_A(+{
zwlXc71qB!U4{DJnxw}<NVwkzNg%NsB<P;lS;!+~iTL$cdT^M;_?VlWRoi~!j@act5
zE^1#{D9jM(EgJcm5^#n|U5B;&GVo)QHKl#Eg+J~o>7Rr18TEn-xdOaQy~Ruk>Uk>C
zV4l#5ginN&gf&W&L1Sx;3B#F^V!ql2RI?BEi@MmIO4hWqDK|}a=d|pOpzeXtx`;d>
z1U<|1rEiRkv!2~t{jGz%753Pc+`UFjXjXA5z=*)Wvjy2^9F<aNjK7c4ZbxcUhn7hh
zzx{pRI@G=m!AIcF$PFXWFn}B*&YpKqRr<^fR2=QQ3{K}3yYUB%O-X%}%#^C&dLP$S
zP8<Vz(hX}`^4L)}fT2WMr9^O<a$s0&RkM?-78+?48gD4tme`kf%#eL5C{!s+oO?lR
zGu_M(unjZ<#w5L79nUo#k8LTdRuZ{O+?)tA#k@RC;o3~UY!*Vwez-@;h5^su<T;)x
zhP!qif-F$-yQm52Nl8so;D+{7PrFFMZ8>3i{Dk^@@so>OSW!NAj1TL(XB=7PLv?rE
z+*fI~alOP7ud66q&`s=1!XJ#gmoE)1Vp8t=Ma!vmCv*~AWGm3)tqy?bx33~JRw7BS
z*~Vqpi-BK>7boTx4Y5Sh9pjMDp+BsyDWAB+bE0*qjbq*+We&D#>e8U1q#C-4Gq~kE
zVlNf>jeZ-Wn-?ELE3^c!CqiX3hCDFkw9PtmTX;JL&W$|fcV|{=Oym=h?dIeSy1Db@
zFRWjdBnh}IZH0y)M`SEb?yZdoC*?aF#WgUyQ{xP5=Lm<Nj#bd@YohUX0!$G3mBUHf
zUOV_54X}jw44LFuOU~4qOlvco1u9r}056{4VQP~od?*RzkA~1ZyP5PweoyYvt@b31
zOml&BjK15P6O&#fhD$MrxUoq2qiz7J2A`DmYodX?>mZhlO@Ut6j#)ELw20dOo$<~a
z_-}*pe-)`2*d876VtqMTl%y1NrOMWXEn{$oe(UB1LuhJ(RwzaD8;iM8^i|=cD;`q7
zC=9{7MdznKAidyk!792Fur$1T7Wp)O9}-F-7B8pTi4GmrTokFGyP-kJw*KK4QIU9P
zelfQOiV;TN8te5L_gIW!=Ppxg^4D0osEZ6ab*t~H&T84TGm<FYM=b)NO63g7|CF3k
z{rc>|)Y62|3yZrM4m&t3cR1U5)+zCfTVNH3u^WO-#`TqK)b65Ub+H0BFK5g(gVHe{
z8Q1(ol6bt`pE0`=+IIq2P(!dbaYGW=Q73+Q3-NS!ZMsLP%i&MpD?k}dh?%Ma5(-p~
z`;(yiw~VHptl`4}3&^m=QeHXDg^&z-^0Xd(l`C|Jf83P*c8vKIL`bJHGW2NUuu6O+
zeGCDSu1jU92J^rixz9^J5#yd7WB%oH*4~J!H<|DEJ_l4pv#m04jVVdjQq~Bh`Akp5
zCE(#)ow^oRTJFI2oQw)G#6~dL5AI+kqFN3qYw06VFGZkP<Zs?L`+jDOAO{w`iIjvd
z#~60TK*WSz+P9Qt7KI;Eh27#mL0&N_zn_K#ijUK3suyh_D)OvyhcEd_a5wL)Tz|Or
zN}2hr_M7(->K?llWQJ|P7E9vQHGMZT##Qt_r`Y7kdDHVom8WIQTcb6gJL{59d+k7#
zDdAC*+xQ}A_Jsww_dTB3I#B8h2r2{jBp(SyN0sNh5x+F#+KFr_egeC>EPv7(2A0Pr
zp;lq}WsXz1i4p%o2pT0O(^clSU!{z8iPFxZw(zEPneo}UNHRT9{v+R}{Ev-mE@0(j
zLOH*G+%n`;_c3j?&T489QVoo_&F;<Dyc>$<U=wZK80fv5{~LRaeUF2mudkn73HOKa
zidrS5@uW+izWMmEftW%$2156>vZqpLmOWqe0i;Y`<bfu+Ac^f7$EHC>2E*(#wy_11
z-RzY8jj>VRKSEOxwW)&7-R+AB_n(UD`zY{vOB<I^!E8t(pEDq-S3{6i6z{2yPagwQ
zoJ0(tTTnpm^}?Yff%@`pVJcGg*0LW_`(7m2w&^<=lyN8GEH6qSUX$qQpp<T>gRziH
zF`<_~h2ok!NQ@f9!LXz0;YCu8q?JjIPj!8oZ3FXHM$%X9i5nRDs4DM>An_V;OvpKJ
z99oIgjA~!nWgk=yPighO;UR?F7!kLgc|edm87;fX-!(?kBunS6j(&X&=}BZ?`E2wf
zAtOOYeH|U%38Y0O<JlX!AEb??`!_($<?g7smuOPr#RIV+5B`9K8JLb=MrysP{WNcy
z&r5QOtooZK!ehwph<Sva>RzSdQ!o#BUir}LChDl)YY)R956Na{P<!a?Lt9zv_!S)r
zV807tc!AL>$C(<OJQX!^Q!FkX|CX{PeOpHROGBE(=+l57bh7K?&6A#(z{27&_3}@D
z6p!aB?$}L+y$sl3HG5{jbn~NxQS6R6E9W`#XtQXJ08Kly?K!8Fgd?v2vy}hNU{%i}
z-SAaaY#>jYh4*8;pD{N^qM3IZ5p=&KjcfhogO*M0(+5oR%BDb}dg>O{-Mvbrj|+xY
zhPY<w{~1l^vlG>?(BPR&e#_)0cf!Zxf;nW|LmWszqG>qkcN*UF)AEl&%2Z$UFccD+
zw>I7=Ja>uP`fWuTr=aY?Ff3w4=Ss~1l>A_O<|Xt;&>J~sZ0(8V5R#@b+~_I7Fl#Y8
zKerPL-0wgPynMrb+JY)pH(zTfU8%vvGa^{0Agxg=>~UKGfF4U6Y0CcVzS?UjyBSil
zZ{oa8k_04s_LB7ou|!m1xv(sSV1a&_O@R$R2U~+&=qED^K|y8ncWQEIra*2U5?DQ_
z5}Jg-$K1=r-w&rBfiUDdI!^E*d|hzGlL}(DuY{l37~zu}WTwenNoJoRe8b$-M*rGG
z?Jq4BpL&?m)Cp!40MTuuqgJp;G5c?$s{?bI9Sj$vYSwTwy7d>cFHBKwAEqo2ebVXi
zuA*CJ22GW&*`jH=od*5y22{JEhwXGSP)e^bZ<`*&rs<nIlMHapm0mFY#Z!2yh%c^G
z2vfVG)M~t`x_)iz|M_apmj7VI9wSG@?wH7>)D-GHV5&EHrR0e^zyRNT`7k|$W)>#g
z&_wRO8YHK<k)NUIKVF^Kz!i8VkL_|=DC|QLNC%P`@d=T*NVT(0-53pKfvuIm2g~ii
zF>@mCvDW3sy4<J0rDduppLuA^+IGqEc*pmR`wf)gM!Bb|&1mPx=Mp}ZA-!~UE`vP`
z<nq^(E4DBOUr@F4NCOxD3JLQ8Qzx@~Bn1Sa0f<<iAxqc+fkUpQTHrH1942NdVkB`2
zp-B%@7UF}xa<KNH{LfGDSCyph(G-UJ*RAMKRMG&nqSYf}5I6&KJ(gR#A9)`jqXV<c
z!;Ao?K}4Qm^h(D*C70bZjjEm_!yI)7I_DvOpK1DYXZi&mljf(N(3E8Mi(L5Ju$In3
z|3?%0xp}<wSc(r-d`&D)l9}zWt>fi*wg*y3vG#yIQVc}HV9Ukqsk>=iz57HS<m2ag
z!Hi997QY->93Iqi^V++S!nRo38MA~JXeXuAbhjtlkltUJ{uMi&rjzEZ?bt2ptIO<F
zGkfR$yYB=gUoBH2?4pK;IM*>2dm1(tC#WXluI%qGEi0q!tZ8-lS)c5W+K|pIXSVvS
zWuBRMZ3gCiz?zq(*QM5?%*WqPPofn*?bMzbPl=>$m+H!CQg-`bj6-I_0my)*Y*tf;
z{z3c)WN^S|AZ;kTZt+voTYw_RvLKBFjuc}!OJ*lS2A0w$Q{Y2Be)|;nsh46durzWN
z05*~;UwRmDdgE|p<5X=9=~E{wlU?UNzw<D=p=C~hcgj{NmJ)a>`7wEbAEO~>*_or9
zjKjh~tTp>PXg=@5z$6!0W#X>-igUB*690{_@VFnOcKbs;S1K4F7LF>o-lQYL%#TxA
zC*t#QM+oug2trAxz#%&l&B9RsN@|Jf`LC}+;p!p%Z}F|w8fdS(sP}9TEaPM`$8dAW
z=2<_G?XOI@>Jq~q)nrPZY^_JhHvY|+B3gEi{s+qfg?$DG%j-R(O`Tnn^<HW<q9upt
z4}Lz4T5lt2GDdyEg?%8MPv(fxM9@7G;I-hQ=J7TU$h+DR=lE-sds9SAOg2vRsu~-d
zr^W*KloX24sl37!Mj~Z<DYxC%XQur=TMd)$zWy#muowv$k#f`*_#S78BD8*-X`yq=
z1*d`%bKNS7Yxb9v)+3QY`<MW2l%>_vo-u+NbI~vB12$W1Xd2}7hz=?7TY>9SpQPAe
z&SXp6<N6aO+z^=4#Ml~{Ov6_yJJlC|H2VWBFiA~K#DMOmR~cHLOG*vZ^rrcyUy4gf
zTc8?lM#`|59bzO_O8mrEkY(ht*|+vhS+3T-D;Z(cPG;Sz#WU0jQMH_pM2@6Of0e(^
z(EN)yJ89mt%*zwP0Q{9_MP0r>A<64mt(k#7-BJ;8_oleXypO&g$1EXFOUFH)DkgQ`
zBs{PA6|M`z;kF>-l)9Z=v|q-Eimwo|$?w^lVjZ8B+;-lQDs7^#`u@+T8$;)^-aFyL
zHT)}rRbM&x(c&ttnOYzWrdM|V%CJSKDeSE^56Olo4nJ)`4#@&%BYV`kZCkEX8L(3M
zP>jdHWdH+ar~U8^9P)I-+~rwKNactng5|?7oy(Gd-47@kjN3QfpuU2`Ja-d2HFjG5
za_%IWzA3$GIHcn;&b0zrtf|3u;#e$&IoxTqrax5^b^z{>2Q&;d#$hSO)LkjG^|30~
zM)BH6{Beeq`f)imytK^5uh&=%jK6#5vFtE^qCF@ESJD)kxw_|QLAp0GY~;6_5-d!~
zx`pKE_wTc}>75Hc`5MsDAZ@L308*=p3S4zYJ07SCxZm#>v7|Jz)s&58iESJpFzFa>
z4<QT|T4mgh_6w&V++VSjTI8&)RR?4>>(+&;2h;aV2xdClB49;9R)9e<TV=N|<g|^=
zHEDLymE8r22zl4(GS+b#X`D(O)>39oFruXi(ckb-so}9{ceFpE=WEeFdHfu0(xVLy
zZ(wG)Zm(PI8QzNt&Ap1A)3s8FP^4YpSUqTK@Gp|Og_>5N-DIGf5TAFbaeLAmhDY4}
z!1$_<6Uorf``^0PY(FK^unkMKc~~oo8);`B7;hcdhvn=_<z9&ERs-i=uDNshw4O=|
z%cO+okpD5pONdBeRUKL?i!{k-;B~X7kx;pQm-hawq%NyO#5I$Z=#(T?f>9+J=MGY+
zz|}W)WB@?6&RDN58^J*OZ71w`6Uyi`ct{KLiNt~k7@wUjuE*<$sK!J=`kSAtmpqI@
zNCN*3e8%1=A~{w3%zEHtM!={9Y-2Z*)zKMcf*H{ShoQ60xjMo|r>SI8a}WWWcOu<1
z&PWp1Vjlt$uAtdwIpqWa`ldFe1e?$8A6w1038xCw79<iXse4RVX`=}yRh`Emc%{oF
zUA@wp_9Rg4tC`tR1k0*;%Vmi_4-;5=Z!PfS!$hf^D@h9l>cyUiG^hGZR0d{Din=-S
zxGg;E3OHc9%g)dg8*4i?ih`6oWc12~J3FqU%__Sf!f;hiwHk7<MN8YVF!wBF0jow%
zB*$Hf#P=dPmJjdYtwI{rD3-a9(KZve4k3+C)`~P98)@C=zqgW!BjVPc2{Ew~HiDX5
zeRG2p6XLd`imrsHzgp%kx1uQEL;hmY=pllBxk~vuinLLc8fLy8Oxdhp1!#wFY)5(6
zG>~LaQv${`LGqk<5NXTUB%q9L7D%|w;2hF$tP`4-xrCzGN7hpVf6N4DPnor&SNsCH
z)PbYF*wCu%ZspJj?j7d&Xi~M05bvJ+CM1PNAl-qwb*2wD@`qZ6=;-L3#aqz{5s^~K
z3aX!xM7ANHP|`<9!APYo`dvufi+j*xe3TgG#cRVoft~onqyhIesCw1~0Y9ZxH0iNw
zoM^lDWzrGf%~$S9sL{g1Q1o-;7xR0W(@61p?Muj*3n${NOt%lxSUs|O_}~eKZ*cyQ
zC(1p8Qt)BrAKb^8dchYNTxD+_+(w}KJ~+Ex3clfLxlH!d{XVTt<HE8-AZtmAL|&XM
zDZ62M9!%~cS2ZTgVWH_$-V?*Lb=C3cd5HuX7{(Uu({D$uW8RV>eHbe?XU(Rg$RCF|
zg(#skm_p0)xzJ4tUB^P%;fDzYtP^+}Y;M!Mm2H_n@HTihl(7W8X{Sz(qclW%$W0{D
z{Sp0XA)&LV7$sb4Ga^N^HnE9Pf}I83dtarRzv@{kD3y_}Osz#L?vE==t3yYeVvZCy
zRG^JTU+9fx5|9)bemRV7{GoU|kzGUosHq+k;|%E?4sUebg*diJ1YK=3$}X)+7w>XZ
z=zrJ#BO+`hKr+!aC?JXL^nGB-r76khDHo=<73&V<NcAVDG9V8ss>829@Y#Fw%y4Pt
zT|TlDCWfWwJ^`1h#vdCPZ--%c%+6$-2nE0K2%5rI$uRc!kDk3ar{kE8GGQ`uOjfb@
zh}U*?_4Z+xAYu;M{YjqQePlGd>0?zySI;lQ<eVzb`5a<@_93SfZ$^5NG}n|!)?pbl
zJ4`f^K{c68|DG;!#f1pnRMfjnl<+ah(8_=*Y5x}OxD(Srtv_(Lq`4=2p~#oF@nX)U
zKGkeqdWx_qfBI0;`NnZ=ZM1^ZRoGh=4QA52@2zeq2GuI?7W@`aQYo+xn1-6Z9dxVF
zdix<#e)Yk&Uy4)qZ^5q8KSQj;?^Fkhx|+)RO$Y`m1|OaB`x~;T(GLyD2-xdU7QA<+
z@~_n;M7JY!qy)51?dTu17<?trI}_QC@r%3!p7u_?V)LjfI|u6fuW#M0|2lisi>8Up
zYhz}brvAd-wQ~0j$y{5#^e!+`yl?a-*uVBG9*39C-4KhS=6{OtOjy%PJJX^`(Fcy|
z6VSnk)Z_-KN7t?K!)J88dPl0P>0qD;hq#x#GUV1>G7x+5B7928tsSMxuXq<eaQR|T
zy6WD+Z3{o>ANwLPSdHuVO}9r%x4zc?&_2Bn8mnyIBrwE?>cloy*RYM%HLLX5e{+8G
zafnnzoG?t$Ls&CF2qAGi4izU4!59NyGip>vbda+ehcWv~#ymu3PYF+rE4djnN*eej
zy4Zvk^2kBoi3+EJj{3NgM4bo?(UD`A&$r>Wh5vtyy#-VhY`6C<C8g9zcL*Zg-7N@;
zO2+_F(k0y>;DCT24Bep8At23=(j^Vj-6`pJ&3)g`^PcxP=RNCNYZeQ^1;fmB?fCD#
ze>)(Csnl7~qq|f)&FpuuPCRDuNBpDCaV4t0{q=RV6SM1HY;U_!=_Ve;I*OdVy+>GT
z66I&QNSWV=4V`-nKklSe1mt#v#54%N>62XfShI=!KY_g|u@Nqj?_!*|ndKHzz6O{C
z{=hDNz>m1+7EvcH@lmNRlei*~R~5dhBOTwK*~6rQL-vi8FP#y3$86-DnGX*1{+cHp
z-qu4qv?FSHv9esrH=!luA3qn4QKPP)_2foQ#&wAecrxJ0_H=A?(XKurH#2ysv7l-B
zs7tu*^~Z(>dSY3Py}g;_w(x)jO0{>~v3#uTKlscUH3ApG(m8<<=E>(>bgxXrWe1AD
zM2FV-*Lz*vB=bsdO=9S!AFH3H2**IWODgW5;F__yC0EC;q-tHxwBxq3MjnP#h{}lN
z8p%HCpi>mJrmQ0M#N(FZy(_V0`-;lIAvwCbqLavNLW=oOH~xi!5jQJrC#+0SAL@*5
z@V;}eWc5{gzSH2B83OfHld&TP4HLmbS&96o(gZ>1Pj9m4MoL!#>$rR5IUpR<JwpCJ
z2fhS%6ZPbA4mr?Kel9c*ymUV97fMlUYr%OQb*Vu6<jTT1CW0>VBeI|$URf;pOpTNm
z)0VDM4j%k@;Psjsemq95UXPhhK%_=J*Ii~*`a;ISZPW_<3fZ*l5^w)m?;`_i7>*U!
zNnTD!zB5B~J;)N5-+1APP<`hb#1$mh@bAh-Wd>d53$5t1H{yY3`Ma$tegnT;{u+qx
zqz@Rt6p41iVS=3Co;>XQxbb3RWh024Ea^vkYsvBOYu@T!{;ISgYQa_`C8{yOc*P=1
zHNi;!A0DI<K^-m|KRq`-3Ww7qkQg1@cg@SbP51X6S<m@6SN9}L{1|hW&ZHW%`%Mf9
zd7;d~_uosf`~K4f4G~j1!nFF_Z(A%7*@`4;_i?SS(`?Bvj*<38{k^b`l4k#Cx5#oK
zs9gQ9Wr`arm?5H}3o;Eh$%+-PPacRijQ5J-?17_*-Fy3L2hA;zhxjdJ5A!E-^8tYs
zk<*+$!Egyu%6rQl&(Z35tW#Z;bEzFaSNzH4cYTP|Z*~LBmQS`zo13&a5ARR!G69bL
zytx_aiV$wN682=F^A*{usv#@m{q@5vi1eDJ376T-#e*_x8LIa5=cIKX>vOu3r$A?=
zhm3rueUurFBVu4sCzVI&Od$MVk_LuUXG-`r@~G`>=X1#XZ}r89LSUSA-L~()U(1)6
zQg*(3CX^T|L2%K_ps*s5RVW%;P#IB@Jnv9wneUNIqx5CQ4RFLN{PQ;+*c}f)@b@1i
zgkzynlzINU&5{bVVg=lFT=OziEPRA8dJxF)<Vy|5&ZM*AO6as5*+*QDYVPh}X4*Hu
zdNZD`H{{49+QEL1cwG~6Sr4_aNoaYELC5A-Z%q7opWJV!4;~mg*KDwcjhZsqjYubJ
zUT^C3Rfe<r4tBQ~tI2LU93@~94%(_RRG$^$95rEW|G4FsF7~p`o)GXfpJ9Z_1AX@y
zt%Br+T|w)aD0U$ON%HUgxn=SrgL}MsClYe{pK%%n$Mvf2S8Fg+Cx?`GiR8N;P@UHH
zwvb@kx@q3)aORljy3M~W>(AFxn;tK)wYjIx+L)W>qeh`RqCjbSr9XnG#;_GBlRqlZ
zS%-NlH?y*V5>PNNy+gaIWEci{ncDcF-TWsd%o)z$b7i^5gI`5k-D5tn?)vDzyPdZj
zy`m~yT3F1#uwu0FifGslIkMCrc-hPfS>fExd90>s^vc&|(yb>-RNhmQlXp$&X^D<+
zi4EI#svo_@9!9T(BE-I>Es0$?Y^lAmar2~?$W198d~x$2*4wJl9<T-FeP6RB<X=kj
z7`H1K%G!@8TD%}59;`cyO8v1dJtg8+P~!JCf}HmW4c9W5X`9$qQi^JhNPW~fr?p42
zY;#08rEv9eczi3b;fd17r|Ep=E-!OM>Vv;$bq!>tAp_OlK6fvF(r^E1XETEVHXugw
ze8bA)s&lG-yeCS_P@<Fk)wQuHWd1Y<T+%sN^<27{vnuv9137IJm)QQhG_f@@^(=_t
zmd0z;8>Y<%7r8=r2ln~Y*C~(o;*9siXx(56yW#{21DiKL&@7C_U);#aH#BizvE2B&
zLdqWS^i_L|F<{Q%JjZpm_15HIt}3>init&UVtsl&j#~FQv?6acK7IKSgYE-zMD>Th
z(gK&QPb6gd8E8tKv$hfv5ykJxQ{J;vC@_)zw)aEv?08$a5UMT_(c%&Ld%OBE$@5Cr
zLK9b&YP#_c7~eK2&0&g1o$$cj4+<pT%Ecz+pYYEZ8B&t8vcHeFG#?P-V-1&n{X1dG
z3bwiPWUu<m%ybk}Pq0l<n|?ZNapY&ASIZ58r#SiqtE?KEXuO%>x<4JNX~q%yl{Q)u
zZEh<B7i4aQrmu&Hre1H^s<{11Q8_0Q(zaVoFQ{bup0b55s`rY3es;zF=Z$M)gi+kQ
zo%aK4E+a58nKv9j{I<MG!tvFg`TVG(MShKk>kx}=_NS4Y*>P`7Y{s&a1C`3!Y*5_3
z?8VbCOW8hGqm;pag1p*G%ai$PGh0cXqRFWRb#uA2tKN3zf7FIqr`K|@&u%|3a<ifz
zBrN`ysx4KZCi+M~kunv#?<Cg%7JAjzz}@PjX{{uhy94+9R-IUQWA{uh2|hF6sT;q)
z?)_(`x$=HsoQC8SbNIKPbGE8=1^<K^nP})M*bjR~-#)Y^SbA5DsN#W=|0oJ(nq=ZI
z>l6Gx=G_YSM=b(}jh{K7iXKn4`lI$k%45EAJ!1H5m^4f2i@azcfyH;Z6b6q$l(k3g
zER)|X(ZN{0@kx(Whdrl}S)TXaw}Z>c26v-=(WL42o`7Kl6=Q0~HXIN;occi$8y@Rl
zzxrJ<Bu{gyy8RILoK97GF9qF+K$GEOY=E))Sy|!%BBJ`W;9No-FC-$E;B^aDh3+d*
zx7eS^Yj3M?*AZsbh<;ulHi_V!#$8eI#NzUvEwS#kEaaPZd)PziwdyhpHDMjpctsW_
zJo#zX?nRC6Pyt5}W4obeIYmIk(U{hr1i2vg5?wBnXOU3Sz1UjL19sda`l&S(Q&Mzg
z?qWTUUR1v$l`D!+VP$H7D3DkNQ{fIP7RtqyB9Tw1m_WJRuzMDIU$>Io3g#6@Oi&#b
z%Iwu&Calz0YACv-rEm9Y+C5izMR(%pE5~cF7S{~Kj`G&&IJ&G3-qjlo`)HS~LDsLc
z8*r4uFFFFKX)cPX)v~J-sdYS5q#knxGwaibu>h<@I)FdEhnf95pqx(-p42)Ue2|uH
zCz%Z0-xo%WlLa&LakAsfZvO&KtldGimew+NOhwa8h3K|GLl(nYnbCgT7n&^_w+cTb
z3Pv}=5(nL&U^><4`eysD1Jm~w{ot}?g)6RIA?M6?Fs0zvKO;B%=yWl^z%Jj&Y5Ot<
z6SBCQ7%aHK&$l*$Nx0-#8s8)7>v23z+T!pa%hSA-rBY95Fb}a(;>Zr3<Z!rNDa#nH
zpb=rni`*XEkxlkHipw}D$@7k>?U5z5Rz8&V%QiTu3L~8rw}Lg)f<C@l(DJEca1zEM
zGxDR>?{BJ}*XEzZ?kmUzT5mR{ZO=J)D+ExnAsf?rmH7VR`e`0Z5l&!57HoRD!u5!n
zNcZssC_XoWkbiJNLR~*&e4K5h22iziyj1vBsv0C$Ce@Zw9iH$1q4jn^;ZkzD{gaX1
z@lY8XBPe94HTVH54b!t+Wf%q<NrM$Ofp{wpI%`K({6a}9s}a9&Ao`;YoIKeq?mz1{
z*G-mxDqSqk=jW0d-5riZN+%lY#hlkt&xezalhz|v2U(UB(Bn4krko#(aey`VB|7Y?
z`Wd2N1l(XJj5rsp24+96$fZ>c6hF@D8x0cex%$D9q1Lxr)fFu*tZ!+Wn+hv?#9^9s
zUkdbW{Qy$w=8|F$mW{V(wg6dP(ofSLXAS%@>hdXLVwbq=*<I4U6p}u*=h|y}bKAJ%
z)4iOe;yh4=IG<#;-miXfRqdA*fGS+86W5T-k;uE=`Gm!@<*{EMnknz}!)%Drn8K_n
z=SrG3O)uO*(nLXD!PeGZvN3WadJCgORhRKmM_dP-$4vBO*JEsf%*A&*yWi_jpg!95
z^^Eq8f4O=_30J(6VEoU0k}oq=S(Ns*^qaq${du7+0qq9nETl#rCP7^!(+IHbc;A=T
zSMZcsLs+NEs1*ZuOPqz@eHlCbHGxk)np?DUrzu#7yMDIW0z(Z)+5FY_s&JyKHbb13
zMfy7Qkz0O}g>zw2yhc&m;pEA>@A=&NvcQF+BSyyFPV=a>^_sWsU^&f5m_3+%{p6)P
z_}-NKh7&AOr?rbIkCE1;(=tyl1B{OE7~<)pE)r*Xa>JfaMN1fdeMwvimQ410NWIZa
zFq@oW>K}^%^Y_eBrhM2`tFAynV&t)i9w@-o$(>=E9`s3j1iH8D%ifP;6A|p=NMn~o
z%Kkf-ilF@i84tr07?U=b9q@<}%CLQuqaOb%zkh(Q((}!lJNb<vxv#%iA8Svy5}$dd
zi+m<Ax}S8e-~wBC0M<Vb>Cga)TD-FWCO8?#WQv{%5`pm6g=_~!rmW^(Tc%9#<E|~$
zb4XbjZtJ`(-dSc3*MRH5d^qk|04w||87j2YY9DXI+&BLj_YBH3nPP2JB7i^kA%P%s
z);5q~vTgrhAsF9_d{R!iIYE~W>+H<546~Wzdo=?w3Lyas;l>>Lj)`X(=5ho27I}F`
z4iqNRIy;A<?}}9orRuA2+#kr*<U(bjLuD8uK-rt=fBXc_Z6aHyFa$+}9rO$R`t6`U
ziOXz&jjU#Kqh#&#o!hk<(BIW(x7!f?h9^j<!#(`GtfL-k8Uo#W+4lK)&h^1?j?z`_
zYt$4d!+A@2h)LT!3<%F%NEb$B1BqJM${_9gwPo;>@t8GFY+5>I+wN8p5v#?l&MZk)
zrat%}vIhTMd8Es%3c<Wv$+>>+8xP}f#)M2_K=6Lku9}yq1UKFxyNgGqqY{j5?_(<@
z^D+9I`<ucZwB+6m+1eduuLEH<ACf*VO>TO5ZWLo30H|d<gb%U8u{hvXkKo(WWi6<<
zA$~WP7?6Yzz&7TZ#LbrUFAF__GI03MLp1W|;kXUZAWSZ8?~9>m0jICVu;oc5QK-26
zY%MgqI{CB*U8I%$zUN7wuCkEzW(ky`9E!X1=>jq#pxWQsjxo<{DymjXs3uOJHc2A%
z>C&2H3ju0Wd%GtW>lygW>8HOv_Y|o!ZJ&N#aC^9qu=vCYf7GAB2zTS>f<Ul5_18sW
zBSeBC-XQ?>VS87S3vWrtCc%is3X|FX)KZ1{*^#9hGyG_s*vbCKF(ZV`^JSk{lBH(N
z2z~U!D{Jph!kSk9VgXd0x^fe0)ZSihqp_<oQL53cy0wNp2o!GsyvWwsvr9fgh%Lq?
zUghVlQ@G$u{vpn=d*n-l?dhPV4Q2RvM{Y}SH=bY{7V#mC<6=QK3j6YLq7XQ;BHn>4
z@Nn*-Qlc8qQ>-dlh^1#ecVT)fI@rJ2UqSSI3#LnhYY{uV%_RQFVauwq#pU6M8$j=d
zfr`NNjEME(1;=UN$w5&}(>t$LmzIj)i1=Xaov+it;(ReX@#Z<8etV+C^S9B-#aN3p
zo3&{KOTvwmtZ3sUz|gA4iD9If_=IXSoB2<+5fjc>UxS6X!_*^wn_Qz(HEt@k+LfC>
zfbY4y9MnkGV*N?lUrnm@x9XLt5BIi>k<^#b6PJ^#J~TCMvd?Xi9pb73xK$7ct5y#w
z%^1g)jssS?MAy|i7|}z_F#0|UTrV9{nRYUIG`;U^*lAgb*i2n~npVL6MJS%pmu*+r
zfr>-4^hfUtzM0RAefk@gOOrozRHO#pLa;F*cQGOR7?A3_5Ow!jwPC?Gdmn?bFMZOx
z#5>ttw5BoKlD^|DheZ4d!Y*esvlXQWK%>+ewzbSEw>7;;|2}K0Q42z0c=zbKP@nIS
z?6Bx_tt3tV<Rr_r)i1Gj=BMuFCpQuj1p)fx7giz7ABY3Bh+-VeKDiRt{9|`cG*Q<+
zF#k0#@I#1R^4>yRn40*B=7tDr_^rq99?N^amnVmm#v~8=-bnvMEAmjgmVezX$4Nu=
z*i{OJhGHtEt#!VoNZbQv7m4Y3ww5c^y?q<M`5tL0=umgn12AXSvP+JR?rcWE=pr9t
zaRRnU@Y}(6BP@RzlV`jr?atx7gc5XHCiJNajwGKa^lUhqmE>bcx*`FeqZ>oSoGDnr
z(?we=0!Ic5K^yeF^Pa_2U-K~)eN#yIZ$#1P@3-jY;qF)p$0j6w2pCt>hSSfhF4)_(
zw_N8j+1BB)h-m|rcPS1{B6JPX@j=h-(X+tz;_qkshXihtcKtO@{(CvP%jz2PoE)}_
zzN8S)&fqIb_fA$Gj%Tsjmz3Pm8URHnB*b0D$|HDvc->;4_142rTF;X?1Wtcf{*DFt
zAnDJkcf!QwM6%svZ+^1Lqc@~JkuF2PVh(A9(`R}#LSAsgxQxf^5_Fev**6#^trWNu
zKXJNB_XS(A+!b!aSZsT)ME!}ac0s3yMrJhl43z<aK~iYF->?Dm2tNyL!Hj)g3>_~D
z`tINw$YhESKcEWYr`MBt*?NpK5)^<+PcKV|I*8T(H4Y~KrXyqiLpa%?qF{`1B@-DS
zCgG7jVVKS|q?^r=pv#r<A-%#2QO^xFdw;i`-okEl+aVIyiOfw5hv|4rRSHtaKm`k2
zcRDdSsSp4Bh*F+E9&HzJcD3(|>*p0==_WMs97k8kFW}I&a~Tr5$FX=M7kZGgAwHZb
zc=d5qxfnYCczB~=CJ#5qJW|nZ&8&<ur;g^z`Ng4GV<y^-=F4`k|EBUh%5f!HR9Z}7
zD47^9;Z><J<^|pT$T&wHRnNetRMGayK!Ty7ThEld{*}`_EbqR$%;<+$MOK61w1l$5
z=P#59D~3U*G-deJF{<e?QVhDff7y-XZG7U9-2eS{CSf9d)H=6W>YxS9MZtnfTd`S{
zw)HxMcQHwPH|9IhvcrQF9(72?vK9F;MRe_g)Ee`cz?P)15JT&vqNyK29$!Y`Pi<1W
zoae&bF#MBm`^l53{4pdw5#=$0h$g*dj|X@RqVf$tXq|@MCBcRVgfl$egl4;C7x6x%
z?(gHXSuHG_>D}4y*|_p{RwqpQ6dO&JD5;Rif~_n(8F%7lhxE1JVf6x4{MiCMnvm-h
z)SmBk9w%;Iioc7Q^2I@><XxJ1I`kE0-K-nz_8Rg^*4039{No4m(&)$sUGyC@DVIBI
zsGfnLOSKzGHO4c4e}~x!b=Hh-$uyQWmUI~fRukTi#r~nQrO~Wd;ire1wv=v}YV%lU
zqcs(5<Vq(p3RNV;8jBt)m&-0GUs;l1{$N5E8Y`yf-78kDOi^)YLXAU+<EeP3L&Ez3
z!P;E&3qLH}5&HfQyPg-X@nWh2hzBhPq^Djea(69}POMI;vln>SUdHhkgucbPTY<aD
zPGe^G=v{-C0#$e>fdie9>}-4#OZu;`8f4G%zpZH;*2UwaW@%xpIq8JMmO@V47W(;W
zOUkmYMQCoTc$RY52_y1~>+_a1{0r2w7lLdw?N5kT(wy^Y17FeLZxOhCXQS75s9blX
z_8?%nxy#l5@oqW{zO72?;)rLR?RTupA5+>b+Nn^l0^<PZ+9>IHdU%Dh{{uZuJRSEO
z4b_@a@r1+6&B4jN4MqyWi&zT}O%`SRz++alk(Z^-s>aPp`(yOe+S;-SZkMUNmvanm
zVre8zZSgTB?rTlc9N!Vbjg5U?tEAujBQeG>OB7OaqMoJ4vn^W~`O^-r#<_fkoZO7P
z8p!e;o>vhwW!vHi<e_&EKT11`=0QKCJSA(T+}z-xFO?`5H7Zrlk7k?RXHXBOUwz^)
zEAouSM<e=b%87q2I(IbgW`3%H+1n?lncRlV__NqbuqTe&J--vK+!RlRI>Qyi9A(O<
zxR{<74DTAMDbCHMMiH$il9%NJ(=RVQnuPROdN@ZF*OR7v_%G{sZc8GYGQ>~XQyKb8
zihOdH$J_}w>`{sE-U>MxYW#Jy{yt#;kB`jGs5(<^sD!2{MYnkzyH8hiwa>u8V`e}N
z*DOngO|sdUyEF7;S1_R3J<UvTEQ^(&Ai@(Q3j2TzF9Rb!1%w1cY<@g8{@XA7Zy$%`
z&;c*1uI#6l{&z7zJ!=>)COUnl>8ktJy@B7+dS&df80rXSkwO7`kJZ5MVs*YifGSoU
zT&qX&(uh{s->wx%=jQqZ{6n2^^Ugna?(duTKYt8V#EPU88sNBJ(tLYq0`|pPue_vY
zL%OIrb!dAy;Ud6%6&jEVHP_zjyuUh>CD#U^DDyFtn@{8~89+?@AP_meV*sAmVlciW
zmWI@gs}4ZK%zYEIkF|~tzI*>GB^ODw`;1-90f_Iy09$2f&=5THo75)Xt9P-CN_D*(
zb+J|j$mgwwiq1rpijJeJ``|BI@jpM~zrPi@$8-QHr1cy58X1x%?^*LV22{88kT35L
zz#Zk7>T<thElA}(*`DBH2Lmpge(9j`Wtr3WJ$QN00=11iA0j>fBA^*&7~gND3mgPH
zJbxO7U%jLF^N|U=Al>T?nb=#~1u^>0Ku5>l2aCxWW6*+r@blwHP_QvsijTDH75*P5
z<{yLgU%sn}-UM(F2g6FVud3A~FIG%C$t_)lwuz*)%Qj$Gu9Sg*4kH*zD<yfrAS>7h
zpt1Hc-1QPyl&~zoMl^rx0{7bb(gx5sGbZ(d$fT(yR~fv+D&=FJEdPJJ#sB@a3{*(~
z$1=`#m~HugJCJ`xGXLX$h+{y$2pyW*>Hp)4Fb!B{w0>mFYH$KSFj*G{_;&S5BV%~h
z8?lkw+5h8D{O60@*pL`UM<PxibHjhGW*w{raLTR@2E_JX8rmbVm&ijkJ>D1X&~(8Z
zNhQMT1HR~Y*%FA>23Bi~tGyIeLYL3P#7YJ5bmW;lVB@M?2r>bqHoJi@H$FxYAoo@v
zv3^D(tKE#z_ke2RyIRu}F3o}3hrwC5ksZUN@{_)c#@iPD=IfWQ)dH6}EbC~$e9sMk
z16UYUy60Nb!Dd0mJJx(SgBpOEXMqFAiEaQkk0bB__b*wu?qoz2ndl*qMRxAj-7(2s
z-<*})3uJu#*;HRR!Tz22{`XI$C$NCK6i^p^^3PxI#SsX0B?v`QiwWA+F20jU2t&db
z#et9iv@Hdijhcl00Pf{wwZIidRtp%?+DEjL8vF1d2OT+?Yy~CLK(skT6=-pl4ARU{
zFo(noP=J&sEdeNtz{1#aqC8CdX)F1pu0d%O&-!y>29H^%hZItg32|EP4AnOF73c$?
z`cw7IZ&%1A5RQN;;+lL5G8NGs0Hs_6#5$9LF7+PFmg5*XJZvQ5k*W<$4p4PQ)9#O-
z0pTK9*2?RR<N*Nia8dbRA1;pTJKV2|0Nr^%|J(x9Sr~YUj5!$^08Yt3UWY9hNedNT
z3TH$dEVdz|!&lp73HJMY{8jqDereXeH~8P|^>jQq)OLE@PZIu#@Z%Nkuu1B)n8?%@
zpfZq=jGDC^4FKQ^iMqt;SnZchjUp0yJrgGy)nFh~;UQ;}Mh`dTfze*W;-8sl0Gk*V
zj|Zjj<-+jp19qZa?VsDL6jf;zOM#T=kDU9%RdN(ttLxxwdQ0DV_fux%5?t?4F!ND4
z&)WF*7a$~(;7MH{1PHU=HK;SCb^t3HVJQeTzWxU#-`6Pq6xog?;-M4V<Fzo$istq@
z+2A&9$D(bs{-YqP(T`~l-s`gUbnkcHCXB>vtA-WrO$g{Wk?7R}FwujF9@Xc)74Za|
zYgAG^1{!fN755%$bve7IE?0kg^E7xg;(xniLoj4a2E)}*A^pdemJlGjPf<y3%Q`1G
zQF>n@wG387t7|%m4m7X~B*?15@ORn!3`0-R-=7Ik+Lp=u222ib17RJ0TJr__Fi=aR
zMUFh`Po=;^gTwWJ)mlYbXYCj|P0Jz~c9ORE7wY6%0It_kAUlDb3Wn-TGJMZi6@=kU
zZ`=nkfh8_D7?Q&et>14F>l6CrgE9;irL!tp&xf1+^8a&lFl|$cG5c*0ocPVr`dx30
zyez}ZTYI!(+=%~vhyMK)X-Yzn@kQatsSy72cWMz#0w87^;6&0bLfIA`tAjTiCotRh
zX??iT`?Va@+?K?*%5vHtrC6~rkA<tAT&KGiNBGXP(@`0iMG2tVC+yd4mqawy>V0f}
zyLqMmQAFYKU&-8GhB6LNK%1|5ktPA@kKpLeh3vrF{3CLKPg>fQfYpvDRr|#FkJZLP
zp8htwnz=Qm+k-Hv+J`CNmacx(1)7Xpv)_&Lc}J8^ujkPZmNp>zks{5hqrha1gK<q{
z<>EWaI{hXOTW!e{VncDjW?KXX^4@;qd3yaHeXtOu?V~Cuxt6i78yG6B4yC8@TT+a@
zMpAe2jt9gXYJtQtr?9R8bm=lMEB@~G<|<5y#*+z9+o;ny{lIP96G&RnOcIYF?ZL*i
z0bspoaas_OIDOjgEh*dC;;B^wb{(t`fUP5my(X~=BoR_Kki>Bve3;2*ULqb6&UvkS
z!JdqNbOYKGLXpg`MSyTWuvr>#0vZ;UDto#7q$!L80mz$r-;K_2PkHPVT&Bzu?<4bk
z+IN&{2A?%GEP%D;wi?Pr@_}FHJX&=h_KCqV$oxRg$BeXrNb~M!^NL&N&6hj`Sk|$b
z+RcUH{u#*>6w?u4tB<<us5>a_Yhvc03B>g5M4ecP)7bNUpfwNG<hlm?NpO+jiU%Mu
zqH>NneSuw#0^80)@aG9zalFk%VV}>Ks&S)lu`=34^JabuVzS%}Y1?*6G&fl=?~yH#
z9E(;4Dbjj?C+s2i(6>sXC*Z6aDDEKxA;J9#nC9Vmvt&nG5_X|}LU)TAh8@bFKevW<
z$i}VeKs3Nc>yrIVuE!?sso)Pc+^BE$hs#~soULRXlkoffW7unuc&Wg0TK9&{05Qug
zoqN4p%?%ULzw76L#L~K#j{o{N#=;qZKRug$!EmMpKJMw7P9H|uQ6gTXSyibQy@XH_
zpiiQ@V0PkP_DvZ3&#klTNamJ2BRQ3=wC?V_0EflHSg!dmOHXbAZ(WzEbx8NS+(dIr
z5>YaGl(Rnws~NWIKhusfQtCo_Ea@rq+Vk%K1kvz^We+)lF?w}XF`D_iPBEaId0@&T
zAJOoM2YB3d*WRCL5{VZeN<?G;65T-`-(U!Mna33L+J_4?XldOi^c@;!z0*qa)RiKn
zUC`%!j-S@h4O|}sfb27BTfCeBqW>DO1kyYNcQtxxqT`=PBU1pK;5W7BO6;CqY{#rW
z0C);u2zl=}y0`Qj6$ik#0W;}7sFg%>T{BQDEb`+{U_JK`?p{JS%^#3|bV$)3&PhN@
zr`c+^0_rKd)(a+L&4jJUDB>QKaM?@k4iJgGodg(1gRyr4SpXK=(P3raKU?e6RQm^z
z7#UKXI>|Zn<#Rhl*gwQ+4X9wAQG0D<alVnoLJRxBoWkyNHf1?LL>juI#IE%gN>9sX
z5q^Tj&PnIfheHkS92cokEVF?<gR<~$)#%rcFBByQfO~ZWO2hXuSSXQvP&j1-B|UYJ
z!iPXNBjLk#NE^5_8}*;j!Ugb!7XkvFodNdEiN}{hENct!RFJ4Q52K@{#}&m^l2*o5
zuck>an425&G_EcI@(zrOkS_(rBWt1L<z=pb(EJiROmA4Rz9+Y@>FyzxTY#V&*kIz-
z^o25<f$-t55g6DnD+4EejXAo5Z;KzHJx#o-K?%5c_=c&$E$_~*w~hbQwOyF6<;blV
z&OUofK|n|i*p`8zRF}-t;_SL*uZIa`AiK?4lsUUY`2@ciu9NJlry4W@_<SYS{QziL
zz>t`f6`v#($vXCI7XA711xC*v7=b(+42L)^0@J*1cE>K@8Y!o6iht;;+Nw<ADN3rM
z^Rw_-i<2g&0yL@bzEci<$CRYe!|rYO%1=5-jSbwUEVj{0LPXAK|NIWa9p0!PT?9}0
zoD_Gm^AulnBeP{Tn=Olc5vtLLS>YL>Hu(V>EEu{VM!D5rZe@%d7hiqQ$-DL<-n_9z
zn0%cXc!x4vc~;1p#Yqy@(@j3>b7Ta7G+;K06Sfp1gbeDEHz%XMM(rNGsw7n^?Txt+
zm*H3D(~dv|Jzwvb+Nfr@d6hFUa)@-o&jeA7;3n)AREv3(M(FyWg5KK{xx}V?hl^mg
zw*er}CAC~i5vj?R0#54vl>vHz-W|))QuDou-rjo|7t86fiI=ruD}F&DPGWbBS`2K8
zzv-Q2zZP5b)Kq~C8heblH$+>8G}IRSN31?nj|?R1dsM&tUF9GM<Rp|9ZbL<7AVgPU
z=S4y)2PG0a4A5;n1CI%<=oeauX;JT@L)cM|9VvPRcjR{XOxwfpw8D>;UhC&%<Fc=o
zlT{4K;$Ecmf!VxohcQ$zH$b&+(%)%ACBR)l22aJRLRD#DzjXu<ddqG2Qah>LEE8RI
z_%r3NJ7pyh{PKo6-bl8k0aikFNw5dD%pKgjSc#KesA8xr?1$pu%A)X~%N76=Qs9c~
z89)3&Lt@nwFGN24UfE=~v1kn5k(wT1F+G!-zSLtjTCCNwvGAX~?3@U^MGRac9STZN
z1$04td|@R0Kt>M>Eab}jLvJtJ60kw=9OIC|Ab`!3FNDTCjNu~FH_Z`&J^!w3L1E7P
zFbsn!Bp8d)$ydd&K*EKCB%~~vRN+hUn;gT{ms-Ycrv%YZiQo>-_TcTJ$`3HNAB?!R
zVN}lH6LDAj%~I~I5g0-wqdFmN(-gACMMJ|)1ikb}xm+tgMSFR2Nf*jpGu`LBqtd_B
zC^MsWKx|_#5Pue!i3M{3$%tVs5bYa{m2A%e3cBp^mvAev4?x2q=LALSYHH9xy!!&H
zjX-hq{EW}$o6Zg6?n&_Gvs`h4hF^e`VYD`eaj6!vxT)}YQhNxW(f)Po8(>3Doiz$_
zh<{O1w<SMCTV}Tly2X_d@g>tTEe}Y!uLJFPys*$3jFu3f>ycj?o~4|YS;XqzY0}9t
z{Y=!+jszQ9$HQD)o?!0N&}YZ8f;F_M26<*%rBo&VHdT_TXAU1ad;&uMkdgHgv}S;O
z)ak{ZWoil?gL%X2*%pRvKyBDFl~G?>y^9`xk^$;QrLc=51{=ZqJe*sPP9u3>*Kx+^
zuyD!^8ya9RTZBKX%C#4GW?B8pAhpbyd4B>Njq1M~%?6C{-bex8pXDY2Us}Dh(+9@O
z=y17GAW+<1hj_Mu#mRdvk2|gRMp@-LpEa<K;q*l%WtjCRdMl-?IF`0uM4E-4l?%hu
zCMMpOiq*JqbspzF&r$sDQ6^SeZ1XMs4wu6oh~sXlf-k#|aWVey3_8$AvBmxw@xOC;
zW>ZvvtS=?<zBBg{?`g;<cwRv}4Jxz6a~2*G?N{Pw0b)EhA`ua;MEVcZ)x<TwHIL(0
zCdb3EK9BEx-DOr*?#os;&5md%*@C#-*Z=m5q!8lrEpamVxE)<P0$c6tS>j<^2wOzN
z!;i=wJZE3XI1{tulN}WQBcRP|bE2`N;4w3g71=JsBvv%KYp`)45S*90%a|A3u{x%*
z?d|M=4=tXSxAB<Ua;w@$eD{0$yNLLM@DkQwX~a)aDorq3CpMO7>**{5p=4r~R8bq7
zT)gz8RBK;eb+CU4f%G!v&dB5XedfqyVlGlvlaF<OvuLMqWU?m#fe0}JcOYkqsBI^S
z|2kqu@{6c)cJ(w_^pl-G)$CV6y-zWmw~t099I^G)j_a(K;_!U2SFM7emL{k@$*jDK
zFHN<EMR&r|A}ky1ZZ|djx5}g%v#caXoCK~Kl_{)VFjJ|jbtl4m?Hx3xT!(1-C`?Rc
z8`{x7P0jrPd29w8F2qCVY8q7b^=9lvr9m>hOqi#TXrM~#mvOH6h8cN%-+O)gq>qnW
zU|VMK5yg+~njV*PTn=}WX(<xf9{!q966JKEPlJD|X)hww5`4kQtf*o8yt*Qa%X|!b
zDJv>DkF06@UiC9LJfLRnb_WmA<4v8-waZ@B=0nF7#hj|zx8*!v@~&DyuN1_$Y?hYe
zu#eaf=9h*rULhX4#LI@0h^YKl^e^*V`v3(V(byOJp{IEiliNQ><0~=Hz->C_y3c?c
zV4N5(N4kNb$Ea(F@he{lUiB8y`x+J0zF|1ylxk8$<Mqg56PGtF+&u~SQl9d}w?v&r
z5S+Kdxol3Z!e3GZCwujcggn!;xep1sZ9W<`&FVk?KPkmu>BWgUYC~Yjza0QKA><K*
zG!L$o3dpbx$6n$R;(zW(jluQP$=!Qm_1a{bo$BG)xR0rp%^Ssj)C-Vgui~@mKcwP!
z@1}0|!J;4}R*a><mR|_zX5;8I;Ykx`tm09X5jbxBZjr>o(~A<J>-yExAeK5U!ik`e
z9wE=3JU}{O`%3bWREUyAL?&W#mPNF(WQBM!Q}=~xAH{1^lDqZJPgFlx^V3><V&hSL
z9-)i%C5fg7&FKxtBD8SYsvrI5)brd@dcAhWD!wflqfFaHsJ-p+TldL#0WmI3zn5@H
z4H*VjeL|j`$`1RC!RiaI-shq$7U_nbhN@h~>!A!3mNwPlx{l&TS>z2V41@U6?P)8G
zQ^lP(oGNM7;FnxZ_s3(RT2@v}$%IocgOv*Y+YFYK!el~jy_Sq-I^@{Ay-xGA$r$+2
zADS#Wh_%)Fj-bv!XMM$KlJL4wywBOzoYMIHcm$cQ1}^4|i6nt;u9UZ?u+{hrn-AV6
zo(+$nB=<&t6kjzC*Y(2)!s9oAt-spqAxm%S1{*4$yW?rU{gu)5w(?iXW@&O`!T<zB
zxPcYGL+g<II9ev>vxpoVew1T@cKo)pZ=TFU<--*>Hm0}Prdzs&Q9z!L=cbO#|1ErE
zx)a3|n${OS|3@MXj7Vw$8Tr0F%<<u*a!Wtep0i_cFx!F`N{weNVdpC&%Pt}fMtd5t
zc_5>1o5o;vk>R^miv+~p-N{-vp>WBA12R<Gt6y0{CUn8V`OXpT(?I1@?716m`{y|}
z2KxN&6Kw}7^&g7f@BHy=06NH?CNXKX#Rb=X3}VJnKkDlFNR(KiYs=3C>w+|rJ_;jD
zZ58viHuU?gMOhg7@=ih(wQ0=HI%^l(4J&j<oX_E#C#$|fs=<3lSQARMwmSpQm}o7&
z_s%IJ?wq%OLoW7!^gT(hckcY=cCXkr=a3@GM{s7M-DaU=1pPVnY-Z<;Md+yE&m|h_
z^LkGd?U~vIf!>Tg(YW7^`;lxrz^`&4<Mcg>y77y4U4B~6%{Q+4vHb9|5yOuu0!;>q
z)tgly<R6(EOvXU;l!JLVWA4I&R(2)yN~;9PhAlDQInbX<3Ur!e1MLnfgDP>yXjsZ?
zR9~W1GBo?{O%10Xis`KDdpJ>%Cg!;jqIM*^KI~ng)V@4s_`U(yST{nx-5{7H?fYs6
zrBbPP<J4m@r4hBsp;Dri?3-1pLA-Jj>%(M#ovedsp2{|BY|f^jM(DC}*eew5I`LW_
zsME5l5~p|?9Py!XV5*l=@^(pZxIHY;X{v|;!@HH?nSE}K7Ugb{kz3w=>8C%6)ug+n
z6yFp6t~h8BK>b?aA%&*LSOoR}%?#zjN3Z-$(bwatW07Qz01Z{T0!rbVwIJ=k7WO;e
zotel$Y4`Qv#is;%w>gA!BcfCb${A#1Sowjp4`I}CpKI?;#2=8upL$BIr_c?IrbNNE
zotnEPvKp^RERwC6GfH_isD&M&b5{kn2K~33AMbjTca08yW5x8IVw_avY8Ait@?jW{
z3&?a2EwNsS?I_h5>|n^bKlS-5b3OhJMq!G<0pL-erB68+-bOjs10^NNiB*v>&-({g
zUIwgP(ig3*>~PFA9PvV3fZJT&>tQ*LZVkp<=lx9ea)flmTI))JHWznd;EX}BERk+D
z&2`>a<or5EDf-OYea#0cTzG(AhC2B9^YSdIQ>lu`h+q$%L2r8^O|h;M-dV*v<COce
zkjdVVMPN#BkW!YElF59B#0Us$j4IRM-}a&DV%Hv*SbHg(EVUDK{(b9p6rV{8iz?pN
zn5Fl!{-;Hj1ofq}Cf*nC|CO}TqxR!G^YG}FvkMVCl^5jU5r4$5>Un!dz`pQwsvE(N
z@019Jg8iTAvq@H`m~$AbO@f;=tu7qCfqO`b=p3MFslaj=uMg6_#$EF(GfGIA*H(>`
zORB_bYx8Rlol>~1IBIq4%Z74KjNWFZhmAUBfxg2T_V$aVc-`yG3dbP>A&g4lN&Yws
zb)d;?wE^XYr0+;}u6Ojd?}-s6P{U@d=FopmPT3kma<<jP5vD)+N7Y`u_}zudWU4cb
zZ$Owiz}w5-ghjMX1d~4v3Cqp1k~H;u??a-3mc4+a?~CMy=6i1P)C@LBpMmbByClxo
z?83w8MnDP-5Uw*Kll1WrzTC)h2@N*?r1U_P(T}gN_|pnpY}T=dwE!J`j{U^V((3g+
zHb%qpV#fDEiDj(@6=8gX;-g&=ptnI-9Hd;W2BP&bA3HR9_^6S_SliMrewl*!YZ{Of
zIV%eGHDJ`s0;3tb+_cICEPneci*(t>J<Z6gl>B9;LHua!g~s=QVFSWTvH|g6sWtv^
z?G2XCIi6XSUO#pJ4nupkN+}vkJ|Jb$KwfP5MLJ)KAU<?)3u3`O;pIrqSzen_&OMVI
z0h`@diEE)k^3VE`2qlw<m%rXvoC+N{V+Y>eWCr6Mk-qwhciBvv8%&Dj5rIKZ6eMNz
zcHuI-&!Z_oIg+&N@?>*4y8y_e%|yfu?DB81z7gCzQVSC2o~KhzwXfT}&f^KO+Dv@%
zm@xhzFOr{bXFU%u{RN%4H<*<1RqM-7x`OUi{xB14cQz3-Z3Lc%(_MTHC?IHqG8I{o
zJ_Ib;{2q)?^7i@wN>p;)OpJ@?0KsY|=^}z&$YTq5=bRp9?CbNW!;k!@q$L*e^2F~>
znZd5}KhT0)-iTx+a2pX`hpjLF1&fbRJ|FFv(x`StGA@yUpngR0Gpvf|=!0;aa2Ou-
zmYb-r-|+sd3w?VYjgx=pHxHAKiEN*ZbCXEvta5VGNQ-J&xF;q#xB84St{tTUGe4dE
zxQ@P3oE9|g!5}plIxhwGb?SwSKLr1{fmlnYbv@)erghGL5loS?lRl^il@-I3B9J~4
z`MxZnClhBLK-0q}#h<m~7+L}d+&qm?O-CB_bWBHU2@a9&OQ;FzbA@_csz?Xuaao#B
zb3KvA0!%sk75#gsK4Pm$uy)~wSBIx%gkCl^FVU#3Ew8r+lOX}z{Ap5658=VKw=}jX
zstTl@^mIHE@0&JS+O`K!wJ5w~k#e^V?n2frC<_>m65a7sdKvf%g@kWJwoI#fcmTe>
z>rh=9;Y3UrUG^;Klp`dF{bhwm2LfXHlEk^MqmWD7G|S$U6mMiO6)0`32H}^cS}Q5%
zD*jPFrAZPt+ps-N5e+U=oeW&K_3YWhAIu(Aif^EmiH4uz7tNLF{8*x4ux$STAfL3N
z;i+b#dZ$y7wO^I>tgf)4%4gpRCt%1Gr42<hCg%<?c3M#4_MNP5KC4o1uSI&EiO<i9
z2^XRNxSd>1sz~-S@%A7=*kq(e1Ns3-HtG+kd*PriYF_ecVPNZ3A-Jb!+OeUMNDG$9
zf(0(aF*>?hc!WpnT^>?Yt|ECpcW8P*CI_`g7EFW`Rc3AI{FlGKmwSshN)Q7AbZsKQ
zf~`0kL%7I8t`?`AR!rKR@!ncVUlD}J+M>8wEl3gPF6m9#M{2tnU@uFQC<xsTj~0Lg
zwSZ_P!#q3+y0^q%#A3Rd;oV&!M{W%@@Bv!YH4Ppiy=_t#eb19HP~IKENb|9fV<)X<
zR9-e>*`K<1E;;2PQc&b`Aj_D|Z}g7e{fCi~GS`xZyj8<Wrz*Nfmh=J{+8};y&H6*K
zo{u6n>10v0a1EYl+gB~Tbf@=m!yDjX-BDLKd&K3k%8iH~p>xti1Lp{-jBslcqhKX;
zqmYGp>R}o!lJX5DeZ~XcuD`S>2F+$5fXsV%Eo#6KA5YqFvK}QYmJ=x(S^c^;87BJB
z*HznPb{qP$%a!hr_fC1(#8Y-@Q_xm3=yG=nYSez6N_zH4d+k?RD9A88{5(+va3aOT
zl&=2q9sjB$Pe%qczRLTrVg7HhIO7vsq2`^775+1bw~cT)W^{zaMydBLSO*mkKVU1z
zJa>zRiP@FYiLAuPEe+OXoN@D&3$;QjIiH;iLt2;w`=1Cm?LdQaUjzAIm!MczKy@d}
z=I7hm*bn@;-i|4grb?#xy^ymSIqs~y30z1T(;1yO552l6E*|!22aWVPb9toCWZOnN
zO%;oUFC9<$^F6moEBv{e+GNb!xA1V5^3X413FH18jI^Nk=xmC=0@2U>oCJkDd00)h
z=+@IN_p0k#MU<A|XP8v-u8rwQHP$Ei)!OgMwt|S~slgn=ZznCs9}wxws~}Z_<kJa~
z>Nn{bAIkonN7@ee9_yAG#d2=pPw>P#zoK=3sYFDa5b4turTBOxw@j9Pvg<WZz>MGy
zbTK7)2GMUn=1&5tBra_P$HK+_-t7d)7`3MRFmz3q2HHdB1`y~eMiWn8Zkh@BVwgqi
z{@ii=k?qEys7$%)^({yja#WQh(5)BdI%$vlG2LGO<z5+AeHJy)O85_F(0@dH|2Q!7
zbEsNz`x6Hliv{u5J%J5UO4ew_Q79s+eRh2{KTz`B_fU=s7Urda;*?aK-2K5u$6(6d
zy;6#X2E4!g&IlH~th?*w-c1tM>-|koN6m2Dx0KCFp4chUSZnt(!#yV6o7)G)*O03`
zS16|g$u?1?=*A|0><h0Nf1%-zzv7a6C)$hPKo=<9BHIg7`<yTdBZ|;{qS}X}|MQmD
zuZrP5s4pe1Plbd$Who!NP}IHP&dRQ&5qR?w;VUb6P}D6S9MRaG1i!;yKjWX0VmR2W
zJ3Tn}Yb&Y>!`qsxoHUIHGV+l50eEFVjrbw6&gr_~rlrUxT9P-<uQkaZ?arHq^Yc3o
zv41&`FcqGCI&3St=QYa=FRnW)GimdwZ#>ZdLF&StUgMaNcptcKON7rp*kq@?qfoiC
z7h)+nrd@fGG^v`7y<1(`BH{=<*X-1{D&zl1fCSJzaG(4;D}pjGLmjSJEJ8oCwoZ4)
zGTyYZqK4`PVxV}>h?dvHL%bn%<0h|WD0Sf$G89gqblp`69<E?iPA4E2JKmO6P=^MK
z*M6%(+@KWedUf~8np1WiEl@vKFO++*Yyg&8Vw*=1bQRH$HGsiPY59Oh*uw_JE!l~>
zJytf0Q1T1zbHfN-!s9v{t$8r$5UeYH;ME`eyT~yHqW4o#j42<N^8JZ`{I4llbBRT|
zo8a#^TV>J|aZ!&VI#h($u<@6jd`SA0L`y81NmbAoBrDHi;lxq19gT2ibc>S?i#!4Q
zBaKU4lJUMv)w$c2*}m^ks(jr=Xws@ocJ>f?){?OP7@tiMHty(D$I92psskmE)9U_^
zhKBtH@n^{82`zeMSqj{ZWZi_s-MP2HDlOa-S$gx4ftS-!_C1MhCXoI!AO0@~QE{>6
zo`ZYg=~Pxc?DZ25(F*pG^2^M|YO|P1>caCgxP>;HUi`F-ZEAj;&FmjPXDpC(FO3;l
z2Mq-wQ>JcGZ==HkYTuQI;4?PYaTXUNj>XLHB4?JU7O{0}R8d#tl^z?*vgwX7iE<u`
zc*FXQE#g=x42b*+Iku4@2%?quNG6GaZU*qPi7J^4AOgpCjINiHl#Dj47wH&YuJtys
zKij_z^!9O)+d1FxbfH``nhiU+X62&GG?T=;s*rntcZ9GNh4;R8&}isO!2_GALSP=F
zGajAoJ}%uc^tVS=UdESVkyZt>-tg1mlakr}S*hd>M0<nu%HD+Hf2e_HETH;yl^Cn}
z=$}GwC>Gt`!s0CmU}E6)sb6c$L>(h^9hXpk0SBP_Et+;sv$Gf%j0O1E^DuMZEy`?x
z@=FpN_d!kgJF>3Xz_;PSHb)j-m-sLjcD>x1Ly3JKbg6I@k*?Were!sLA~b7h<(6!L
zAfh5Xy03I4MWLY{>Ny!Jo(Bu9oQNMyu~8kZNL20?y{NTmV5(4a`3h+psMg29*(__w
zI)Y!jR6cOOU~$-rCg<zqAY=FqHn!-0J*!z#daz7q)Ha>@o#yKmT^j#`e$0!Nes>qo
zIH(hS0tBWUU;JIiZCrd(NL$S)fnDb{+O_FT>sJhifv#VN5PH=@S@Mg_66nJm|3A0(
zH^B;m?Wv&6#3`ub4!cSc8;e86*cWkd5$vki75mQwuuxg~j^32$+;~@WL9^apqF@$j
zmw!fUCM&~{kqQfc??sK~7AA?9edJBO?j!%yQ8j}$dcATA^yWRf1lnLXDXubBi4Z_p
zx6c}AN0jIei5~ayK`yzbv%be~#~56`CHXWN-4QvdE)w%rz~{H)dIS_Fe@(IYjSp+3
zxam^%P!fhINj#yByAqT>@{;I^Sn{~DZ0EP$+i4x?FaZ+f3DAD>B&otN!YW}GOYs=}
z3}}K!ax*zbd8F**>olOYh`!C@dv$ioKjOc7Iy08Tu1ezggA}*Ny3_N<U1Jb)#JtCc
zE<dq6jW?os+0ux{luBVu@g&;u^<LZW{?m%p0Up~fP0Qk=M<3t8@sI57ZyPka_1-bx
z4`1qknm+NIq+v8GvPp{Cxbm0&u(3c239?Iv4pjjeWNF>J2^}b^BPGNGV8&p{e7^G%
zQje*aaw!dAtG#XCKByluUmClS16{ffQ&;c2CiZ%ov~sgpLW0d#+6-P^srt{FxsiY&
z#K!j7=j$}%qX-!n!rh>zqE9Ui$YCACx`t;L0~J5%v}eJq>Yqg8T6k7Rst@pzDObLd
zfsJ~44jO&;M*}^@231A^C4&-u%I;0-Uh$H4)9hojq{e*6UUN5MTJuZ``j0?;pXmVY
zqn*$;4pQ#?>j7a!fndG^cR_p-)rtGLj1FiL3~t1rR`Xc)Ka=K%FVOwf2(r);Nc{+o
zQ3o|Gf#RaHl6*#Xx>2g}6ABN`pWs(Nqf(nB-Dx%D&eHyQ!CZo~wk8BG%Y0M+T^(nQ
znHdiK`sND2;R)3owZBEtGLs$gN76`$!V5d0vYpU9Ycx+jZG~2n&R~g!`)fGDaBg}y
zEEUR-3&mFzYya8(4y5`bNP=vjhJFkd^9s&d6Ehs%nZr})G4PXQAo%;<p)1d~ikjj!
zP@N{5^aK3~7!bYRhigwpM2=PFcXw~N0R;>lLWd4{bk}=w+uRg21j-p06mdbJV*jEm
zS>Zi&FCkdB?1;A-<5qo{H5ngLe%EDxdG$4e?7FmW@<OjJoD9)B%btw^`G^4txm)>u
zBZFyxgc}ivNkgV~Z`nFeGHLamWxBff=`0DhsI3O=ixux-A4fayZ7H}i8$3X0QdX5a
z(4eg29Tb(y!G^i){+gc<+-4t&kSW!l;wQDvnW}w%ZC_nl9@^>%n%KbaInsQ=P3Xu?
z`ey#FGX9L(*p?j{NfpQCovoONZt)+?={U?3XMPcWAI{7Zh^~cpVNKQ`y2WFrE2@&X
z#DzR{?{==d0A;nn`wFAd8LGd`g6th;EXWxrm)>wMzA95_gyL2kHE#$#y(YCq`SSQ1
zvn<Zq6Lz@y5Oj%V-(4i*@=3FUX<Hc>1Vo1%U_c&vW>_r=G}D-=C9M)dEEtNQ_Rhf^
z$rz9?7?nmOYEUZonw?{yvEwqYoOqFOeH41fgJ%f#s<fgf9+qiJnK)?C1%KDVJ&^#K
zYkil-6?dUSt3l^o$Pp$aYR#y|DvfRg^gl?zGh4Rfb<GZ&dfHc7LuTitl75Q}S@LKg
zTSvZL+6}PSF-Mk|su+4MM+u=G4qJA$xF4y^Jq=4w#?KvX(RunuCPR%oacUq)ya3Jj
z0x5GQ=$J;FtUYtJe~mEaa#&-ovhK8lV4_2UUzVVWL2fQ<2MieWX%t2YxY`Y17n!=b
zU;4JK&`v0Ag|ehGcRvcQQEUq8EB>M#eZEpjeDf;yWWxaIe5?{$fM!gn1%|D`dzhiR
z-qCiQbR~7U*3jaCxT(@5@15D5+35XliJXrt{9rO`GBD7X{9UzVI`Ovk8#$Yr@W%Q0
zrmwgIv}nIaf^|@+R%x9kNW@w<i`cr1)YGSkxp?YAR(tg6l=GksSGwC~Jns=0Qv_W$
zbQu~4Mfp=Le3W~`wap7tFWoUKIjN8<Z2i{|mC0VMMqUROd%<VdY$=xMcF(0UKo*<)
zeC!RafX2oW-rldBj#4P1wdFXQ%tKMJ4xW)*zfn6i^Jwt5M<<b-ZO&NKz@DzhID+qS
zWyt>d9M1>6jJ{JiY4Yl&UYJ$=e^?VrWZEBvtEBR;w4WD??lE_-+9D9-&N}ScG0!ZT
zaIFdU=_{5O1x(W#F!xfUGb1nyu;O)~3IR|O3xmnd;MbKg>!`K1UW6G2pd+fuBXutE
zJSA+=f}YJG5r*FHCF5~@*k$+|5p;J6z1<t@f<Pj<SB{jM0L7Y@wCB5%n;RpH;_PiB
zSAYxN5=A@j^P#R_oyv`6+(oP@f&l^@W%_;bNQGw+nISE0+9bn-Qc0(eq$Y}ZF}%fQ
zNM{M<?vru1D2R&d48Hdi|M>9_{iJlHwZmW-TWLeZo9E5rdGFWIoT%_u`oCu5YJ4#(
z3u0TX1$zhXy0EOcec*4wl)*uuqvCfEq9baJq^nvzvCQ-d->)&|M_j+)BB-UuG37y{
zxuzeNoedQ(1byeqlu_1?mg8LcM<-^1_&i@SYMn^~&(ruN-;kK>_&{+J%9-$0KC6m#
z>w&D=<)cf!kS`}rpq(IrtBUR=y=2h$*{|UNqlbbi(XLbq&55yhFG=8K&wiLv!xYCj
zLVlejh4b}yc$biUI$>nzRSjAMyHy-p$(Fd$;U`ZcN!fLtTlQb3S|$RsL4vUYB);T_
z2w@V{O}XQcy-l;;%MaWiE=d(kzbz@y`5m-k)_*P;=TWF(BnuXSWsL2T>C6hjzBTEP
zc2qeTUshVBW$v%*VKEVgwxY-KnPFwCSGExKWtZUMRon<}W*QeXwF|=4_0Pm^&cUwA
zXM~Xj=Zr<Wdl%A|m*J=XAA9c|Pj%nNk4NN4I484gGAdcwTSk;Zg(D*)Bs*KkcI-+S
zDWp&tQ9|}OMs}3FXOg{Vzt^YxTK9Ed_x=6;et-Y|xh{`Ge9rrGKJWMI^?W@?JVX19
zZZaH8i+OVN!pwMBYa~{{;&LoSzSVX<{jGbvt@ZcMM1wwRspH;R-RFGmx&roL<MdAK
zfyuZ_9uL^p$!DArI@1)-df(gLwWuget+*|EM)H?X;0b3toTl>!ztv-I+2+;KioC~k
zCU7&}Y!-;i<nwuZOu>T0%y@e;pm{#{UI9C~3>zaan6RpQJddW+enKvDhEW>Q!*`Xh
zO?{Ro%cP6Xywg?{ex=#5^LZf|Ny}KuFiEu;0QMIxv!~qVCEoEbPcSu?5%7BnO<(&P
z6aJew5aoY(o2OI%MkbcON`%&5N}(QQC8P|zrzPL8(MouZu~kegj*my-QB8mXFYDL4
zn$enm)~=k3_XRad>DxG&c{$m8s-Msk(&M#JHP|AHs^bc5e$p<SHw5ReC*y282t{(}
zf_$+OtV{+B7>gq>F(yRzi7|CH^3DgE8V*&FdRRVqHZFCSOqSusgJJB6nkP)`M1hHs
zf$Uqp46$v-@(B;_eW!{&F{<J{9(};RsNUC|*ymZO#)zs{HnlPBr}NDk+h?avKAQqP
zWqo#i)u}QWZ|31d+ZVm=Vf%LZZ=W;6F0ZQ{*RoDLvyG<2U$-pCX0>~<hv8nsNd{p#
z3ly`s@bVRYX)V{E!&opZqhUSODn<5z<+l2FN^g-p_O1)YosSP_Y)iIgycLSm9b(2T
zn@)0uWxPs!q*fditH=D3VD>DI7L%h@K8P_^l{M|pa%r)V)1ED2?;nqk$=<=bnc1`!
zB$is!WK*H=Ln4_mkEFZ>=WwDbESGxgm-wSe>TBMe_A9s~YaNsn59h?W{ntO&2&L%*
z-tAFOw<kS$#^A_2opEp_Wc8f*BwqZOqLBx+nqmZ<pWxnMk6J<zZ;2U>wwEE>3io2>
z>Bc^qukQ)AvL^%Xrrhz|=@;;(N2qmQ_sbaThFIzS`*id-;^kPJngyA7DTHpEH^yHK
z#h#1O+2~wyxDol)2(RZ+txzyJ=S%1Jj`nWnQvUMY-#e-*8KO3cMFB^cX^KRaWdsG3
zs{7U6F-ni*&zLk1_7GZ&p}VH)<mD*H+}|G*(7YEPSLILlVtL9^@a^{Bv>5UEFt-%9
zdR%ztmrCScr5+6-ipAk-*jE)=`*)Oer%AqFjK;_Z29#JRNnlZ$Xg{^b2WZVP660OR
zGhhB$Ai}@>Dd#5y1e~y3w6yDw9*R_UBV+NciI^Jb;cI{bE`=gdBaQ9IE06x_Z4kFy
zt#$<01bTiOx)ipqRw{}+w!A?r?;2=y4Ya!NNE3RDjT#j+w}%O@20G+Mc(J~$NfqRa
z#SAa~NYSahvIRXGx&S776ns8Q*Q>}fur?}zJ%emRS6AMVjVa)rJp8<iruKls&dgW3
zSWOYK>N?6gsn(A-7u?W6X38Pbb>xiU^%$a35B3pv#^ll?=a`<CS!N5TM|ZmYy>CQV
zi#xHASAI3d|5~IooC3GMymDw)AR!)LX1{#(77HO3Jw7TG%j7`rvNai3A~mzF_WTMv
z;o$iL)}FlA#$307)B2#{<4*hRg@64D_-~o-0iET%jzXaEhrnP<5aHiyr<UCcU5CJ7
z6w;Ns&t?}6TygBXex_GxHDoi|45;S}Z4#-hN+ttwBc}U6&N>pp?D>j}B+osO0=I?G
z{(Mj09iUx#N8&G-bt^_TP!%LRap8wN*d_pIx|lp4yN`*()N>`;97QrIbElqI>x9x{
zo(}>SX%uQRr>0ed%w}J0yikX)_Ry^ahS~k13T64CiH8?;IOFK>>5y6c#&(Hh^0>??
z%>t~}bh!YN`xgIoT@8DF;BsAow=?2T77TG2{R?X2WFfewSCq|d=cAdK5FJ6z8%LIB
z6j-Yyw;_)jNT&H=L#J<(7p82%Yq0FksKe*zc5(2u)cwmwkCdnpVsnXmRNFU+gBQIJ
z=r)`<FIN3{2cdWe&cPKXy&DBi5VBoj{=;2%W|Cgo_`XI`9wWjg)k=*hZ9fF8bUnJ8
z?AtZA&b$YUIBt2(f1DoZvs9<D-Gb0>r>0P=1^V&2NT=`s{70x@1E0C}eL8cq-ua47
zzEh6kQ<Rrg_!^-{nw?qXDsT|p@~q0nKm1K(ma>OX^L)%xssALe%g+$G-){-98z8S;
z1h)thX2!RW>SY3doC1*|&8I3#8zpCRR1yw?u0ZGGfk~R|TSu5AzrZC3=t^L3C>gVJ
zH~<fMbBrn6i-^`JPq^V6<NN{;y8P7t0LehK=0b@3fUrV0JaFN$hOT-Q8H&3FK<krW
zR&oW)(co|#+A=$@?)Y>2;qp-!PG?nbfcZ-ZXby&n<$tj(`XIHMsqjb0CX!W73DYEf
z^UJ{{ognmheTpLLd~--q^j!^u9*!u*(m*U(xHHVGZUA%jpd^YU6y--dBMG7vL&l(;
zaiCh)*9I(9{PdAmuNOhPo5+w0Vtb!Q2s_w&13k{80~Z}v9SceV3e$Iwb&VugzcnAi
z#}>5&N%_4~Hf}HwXX*)-=WYAo1c;okJ9tKHi{u^;O^I5TFH$i)#~XtFaHGP<NG!K2
z&TZIpD#0@NMYL>RXK4H*N#ER~*1k9ewSjY8+`oGPoE}r<9sR&Qa$yHN63HNee4*6m
z>RZ*bgvE#jLMf=XjNX3apAJgPXSmT3wPTx<>>y1mDAr?qu<*^X&ae;4cvqYf5F&l}
zGfLMn(e@Km59T<crN_qlVZtP-Y3WY-2o1^z5wU)ovxV)gu7Lr-Brhei2il_s9of5Q
zl};hxw9_rOv$3!+DFHp@=1K@qQyYTHA2oqipDIdb9oyFSQ|jzbIhQenNB``R;55(Q
z$Ig7n*g0qU%I!Z#2Pr2BK|$dyKJzl>6s2?MiLg8FJW-GRp~c9%Dkgl%(16?J!ht4M
zENz4zF)OVQmfje{+7_lHt&_|ykcYu_iH@H;xpi?vmsk8f_OB$<7)b)&vm7>KwJ+$4
z;^gD$l)X`NXu-5;odIiBLUoe)V|nSWaD12WM2utT*^g}Zw)xN^zSsA9kDPWt+baHo
zjP%9iduQpFGoZwBHK8{>Yk(iCqpmt;Ro%g+)BBE=#EkK2^kYTd@s(5t{%ekHVFxgu
zkNK<eZm`8RQN6Bphtcz;8n~y&=K(Ib93udxD+6?<=vQp}5jten%7H-=yaegljcL(m
zo=}dR+uZ^NgNF|qr^Sp-AMg7@j(>w=&!dn98g@GWjjgCkG&Al3M)@fEUY#dIRC4>5
z>k#vjnJQDxx_{A7dIj8tgkfu^w#c+r^H{xfQ*y!d08bpN0FyYgSgRnKfBvc2t8Upb
zLBl<z&2a5HdgH(k@`R=;StpM!vI|0I^>ZIlw*c5(Uttf0qmjyf&Hm?B%-uUjX6kOA
zuu(m`hSK6*d-yIbJ5BkKVALu%U$aM&g1%A~wk^$NTiW|Kq0YZXpwJ_-z>O}@(Bt9!
z<5*yCzd~3l(t2>TRG^mnYLRC4`;cmn>D&5_guJhoT<;y>tzH~Ss-<pJQ6S>tB;@rn
zli>~{eh)~HiUF!^zSI8C^DF)lO5n!1h9G*LLwL>RIMb2%Xu18O+jc>f;X9Ew^1Pcs
z?JqWt6#vMguo-eI`mXZ)$<XM#)CDfIxPVoB<rcPNI!%=K_1D7&0cM!;+Pm05wU-iM
zl-AGmyVlvQ<PAcxIq!XO?_xJGpW!s~^b<;e0x*-_GMfKkw6nn%whnyR=~z_8liR20
z1|9dZ`e#=?mZWs>dGjG_a+{hYX8Rblidf0grjA#&|BhXj$(6T7Q>&dCEO030JlUW_
zahowkSfw_1c%%=uQ<#0by@RXo-U-z2bHXEx1~qF>7F&+PVx3k~oJx(S7|<CHEtcz+
z@D~{;BAq(sMDN;di#fJivSN*(Wjt;7sI&gt-+R-$NN-x!bpG{kz3K0F<!?ng+RsSi
zKw&W`UUT)6V~A&rZai)L#n=0{r?yf_9uWnlpdbH@i25JD^_CbR)}uyd4&eBSoG4_k
zakC8*QT)e$=9h%jA8pIsU;o?*exD=$L2X68FivIl?e7SnkmMODf9&gk-Y8foUAYX%
z24^+M*_aT?Rw(=1#=w7X&;La`bVYWX^`qCe{||p7g&gIzR7+>llWXGFMi1-<s_l&h
z7VruCkPBmoAVfAX-|<TT{BJ9RO#dkn&kmB&*^WOy=f9(i|MrSN9=fu6WOHe}29Wow
zpE}@9Q#o8hI3v0RGvWVm@i_T5<StG($)EiDZTpXf|K}g@v*UmzE_~h<^luCLkE{HT
zFQ!Z*GQ1V@6zadf?f?9g{NpR~g^0ebUvu`{A8S$epb}2$<2glRVfRmja^H)VSx4<{
zugym4X+7b-LWta3goA&|Mc85m5z1V>B%;0B`(+wdjY=ho@fadZlDh|ptO<BYS)a;D
z5}&cZx1(PKDN!}Z>LPL*fkEfaAkYl4HG|mE1lW=`YWjBO9;-w(DsOuex0}zicnc#4
z#A99rNjKS^$#}clMThJ!-y$mEpHKm?eNfJ(ikR*QsNJnCGgce+hh&pBP{M}S{(Z>z
zmu2H3fDhYc!l0}_C3U-Ugb<Nxiv<j$+U9*rm-&gerOOw@L9S8*4+paml&}s^2?s;*
z@Udxnw$h7G*U&|K^T+$PWw-PK_WA2pX^*eyuZmn8fr-jtgw;<r-m|e(&r+gkR~<-d
z^293r6Z25Jec&PB`rng3vH`sFU|`L)1_hEUcPNrnELM|^=GIV<Ep7o}S#h}By9Uw!
zF1Fp<KkEfz7ZcFRB6aqynb+3l9*>fGXq9?kc@TxI_dW>Gs*4=Q2~+EPEtAWLuX*w{
zBI?djr1}{4*|FusFxcj6$&Wmu=r(YQ++cF-_gZP86NRPUyqe73-bFM7ia-HqnQ}o0
z=z3Q&gq(pk$!q!3BB*);2@W0#gw`Ohzq9=@+4|H{>FKRb!9PAD|M^7=@;Q>Qlf3x*
z@~6aMDwlf@SY0LZ_~cks4%g2cKeyw^4q@2l;0i~<uZbcvy8!C<Zbu2C+@Vbn21%BJ
z?D=sS?jsO(!v)}~gFr=G`i^O8_v9BYUrre!4M3ltu?YW5>WY&egwol{I=LrE<`In}
z6<zEh5=v%9kfoLKJ?(uT!b)v$Z*Q#zNb$_U6imlyZ+I;UKo(-Lb{!&JNlFGU0Xg1}
z6A#n4^B+%Yd7A2@P3*s)MZQ;(OPv{j{kHq<MfAXNbF$o3>>~gwt>}er;=<pcDIgJ2
z?l1|XgM1HAJTp7;L4i{u9>1Ij<*0WlqNB`=`b>$rrAwGUY?<UycVqtByl^>4f%or_
znvt8$18Zw|Q0-gCn&+(hP+Id0>!0W5-?s_BbmaLKmVEa6=#afV5oIh$PIeLSXzw1l
zpA46`p$;33J;cdqLvJ`w3i{SkGUD-8#NFX3*cJSebjt0HQjS}=ah~pOr(rbQz5_i$
zFdUc})Ie+YC*#fa4PWvMa6p{Moby;sk1D-EJ<k=TKS&t1x=>U`lbn4jTIc15X7RgM
zlRc0glejSy4BIu=Ar<aoWh$#T@3|kt`|E`S;r=u79&;en_bUjPT~VghJAv3H%qXrJ
z67|JaLW_uw*EK#?m{EO*!Db{@re9k;eySs}ZpFEH?RTaAKkS1kUhvq)wp|naLkOGC
zM!;Lmo8Nv83K)~#`!ChB4Zenr-J)q4Iw6d3GzZ>&Lu7-@ZZ3mwmTxH!o;-V*l%(mU
zo^%cfBtwvI5p>rgDGVgKwOB(XuOPAoq?Wx<X}J46Oo8b7MPo>mro;vVB5M8gZm0Xu
z2gVtm|8bKnkekFS@vjLSJQ}taMnG$Mr07&_YFG5V-EjffGlQ+)mURsZ?oV*bIc@~*
z?QBp+OE0TJXDn3>V>Y_YGT0t}c^oL_z(+;j?PPduDvU2{3?;_q9sVwM?HesE8ihwA
z>v{uZewx=nknD7cqhA=-PuLh^)1IP$m~ez<=y;Fa-UOQg3S?7rhsO0FAD1JkSex_;
zQC4YIe#eE6w0PW4q?Kp6x9cblz~aC*y;VanY$t<*xtB$IZ}uAQOP%re8;9NPyVVvv
zq2F^@;y}@u3nO9I@5f{B_<j0ARcze{s$x1v#A4v1xjbRVWSG=2bQ$8TQS5vJc}Mwa
zU&1ep$=zLUnx1<8Ow2W|2vmFJ%j>?U^$B&RjYX%QI=NOl!BvvrMFlnI;d`E-{<TS0
ze;xCsg<ka_pORUn;kV@d)r_zmla7zT7p)s~kxMk+g59{d_I%$sJQ7T_hR#}UlNXaL
zVdrgUIu~GpRiU>^U!}t8*1S9Jtt8p*R#l>-ufd{Uw_(xvqIaLK#W$ZWTNh~|;kp@0
zqqiWST+w+UqTg*h8yHYu&e%Af`A%wkBKY05dLLn|>g(jy!U(l|ngR7r%2ptmFFh2I
zn~}5sI7wSr3U1zhX|dN&3DCB?X*AQ4bZUNca!`x#T?8F9=~Z>Xpal!BKu#O*oUlk1
zd)apJo_#?iyH22tU-A3TDhz2a$l<7sFS#1UaIBsu(T2zR@`K32gyH6(SBV-jYmcTQ
zePWgpBxqtIL7{DwtlKP5bL8>2@B*zY?u;{iqj#Q^M@SGmy%A?gsW7t1t1YEOtS3%&
zPdB-A8%{g%l-E1XO~&0=_d=7*C3&?EIc}B1;p22`u=44VrYf`gsI@KH*Z{Q}UxI_u
zQP*jW5)KM>X>Fi@dhjt^ckAYjiSUuff>zM1uPourUTb_gUCpw;NTRU;e4Jgk^<fZ0
zF+h<JH)!yEA-}c?PN>b@AK*N&BJDWWBHG`)^URaZwqT*Hu9srmL-f-2ncoJy|9O(I
zw<iHIZPRu~>&<VwL5dY2wyTDQjgKw^Fam)l-uFcUlv*Qj_8IUj{jR7j=Lt8Cc7lpf
zgigcZvgA|P-$vhL<}52);I(l?j@}Ita5nxNc+6Hm5lRtC(xePe*NgL^i$s27fd7w#
zmjZ7wtv(!gz^^OfY%4uHa`bcpFFO&Nj^dFA(s}Z&OD|)HH;XDSB{`}+A?JNl0oy)b
zpsC4Ir>&-7S?}$llnCL8ARd1Oo|<za;Z=(x?n-pL>+GINCxoGfoL0u5kBvyKK^QP8
zsz<fnx%1L=s{EMzW+AwQ6usgoX-MMJrCq|zX!1SPkVIGK8Ebe@Fi8sUF2_|gUs5cw
zvr#els&0g?g>Lim5SKX1=g&}F4?b_rTM(<*iVQz12j`COqW^Lk9G$(|q%E;CKzybG
zL5E!7e$Xq@QT*Z4S(>4AOZE5J6iZuKlTunn&*g)rHTG$Mdzk12{d6NoLm!S^dWkzX
z>4AxV(U3;0Nk}MEJ(MJU@f>vw?VJOzImN<tTbVlDF3T5_S}!ZYkDz<0vvpTJK=uEU
z>;vXI8*^G)##=?A8hRo3MdP<MtI4YpNZmuMf6SX}Z|Hdl^LKXJoao9#L0?g|<j@_D
zb%ehAc-Epqpg>>>Is1piuI__5iMj9Dy&o{W{x*IvcAZN_x7La|-->7z3_unRj*EJ5
zF?ualjlwj2Q7}j-n<gw@YaoMTd#7G*e{{^X?yWOYzZve<i8gKkllm}huWsAupGI|0
zQx#ucvPtpQa(gu$F^zhZVE<Vb{aG++<<5b<Q~aWM&HPAqRVgqIDZMvU3}MO%Z5g<@
zrDPv+$7vuQj@dc^zmXhjJXPgg*rkTIdaAuffmZH7z)g6)@h2QB8R55&EqqgqHxsO=
ze8S?Cb=}muaSgWk71=o9c9#2xunsaxcXJyPxvNovtf?2jbKh(JdTWP6C1R0ICEo4m
zFtm9iPRZb{&{ofbf4HjEd=Gu?p$}1&hhH(iXSmVV;eOr|iS-~lBl>gz^rp}Ejb8fc
z0<IfD2)jz`nDX|%bKsmnQby;Cg@1Uges6l_E<v-n&c~8=`L|N-I9Fx$$yk-#;$?sv
zOLF~db!lz#H0xbyi}a4DybO6Y1<X1WouTssvOY9A{UEzo7M1B1_!QT_M@OW5gDQ%$
zCWO+hYDkUy*(2HP<!7~5Xp>2?6XTR7OtDt6MxItTBDjob1NDB~RG{%wkjzc-f{|%S
zI|Um$e1*5+-L2MYVL}vbLSA)1P@;3C_DR~zT=f73721RqB`%a246)7WaA|GP!!J-4
z&;oAMdeBzoeZzfK>~I7*dJ+9W3RLwwzaTRrN(6Ev>?+1qT3`~0Jv%})BfNPfuNvuD
ztLGVDK2|*oa?a{D{)!9oB&^oN+Zm8uJdaKMapt*}qxSnPe-T2m@5k1Biwl5C?5<}q
zj-O;(JH$rb#fsrcBW6`@e6${*uS|3lC1aI2QxH-V?97haMC>cZVs1S+nb`VcK}GMV
zTq=`+M@>Mv|GKXO(Zv_P#LaGs0^#CEKCARTM>WOAHayxVb)5*!=is|YvK)+h)HaWb
zj6MHgpHGPA`dhS$+2~P6=2W4qozLA5M2QkswNZ=TEWdSPXg;%a#G3N-Sqj;)zD$3{
zdN{+IHSAoeQt}Qs5NyT!;da5P=+g(CaI|Hc%*Lvq&D*l|!-xjhYxlV!+hxTs%1q}`
zp~OR>hT^KdtY50fSj`T{Ivo?c)!lFN{t}XubNhPBnLd@#BTjvslko*l&KG~Pe{D~c
zMR``vq_j0yXo|X5JaOyN@VWj<e-2w;!GvO5%6(7#(;#WBw+$Lcl>2!i&{aZ(RjH?9
zkM$P7N4VtfI?p36rqy1QFeYu~lWgOI{Reybq;*PtqPD)Sb-VYsAQ=tso_}V!W+1rK
z;-C{<pWK`al=o}ag7q&V*;4SuY&yIN8^)7IF9yz3uI2TpRzzxRPB}ew0w5L=`%wGe
z-eR_@kB**5Mun=;b(w$ago^0Lio8;u2f=J;ZyxIc%;3-z)wH80#;TCSy6Mc9(R5g}
z17wbQZ@1a-bZpcY^_fUVTYdgV;NV~VvOGVVF{R*Nw7lmC3vMO!jCtLG_KcfLe3|hH
zfPlm;cv{UE1de@Z?Q|)w=L=^Zad%_EW-YA?mHyLgJf?8PD!e-M432wBmapvNlp9qe
zC)b4w&6jSPvS;CGy{O(BPu3@LzQ7YXhKADLX%i#RH8p)hv15}bA-EKlF`R{S>0uG;
zNzG<C7;WKyrekM|D@BbNtPCimo8M@g#`YeiDx6TUKlcf5@RO6PPHp5;aVJ45-nzO0
zA9$cux=hYz80?j$nJhkV2@<Fd{78QEUeEEzBjvAZw#zkm(j+@^=kqB~n~BZ7rm5%O
zN2CRzdYoA{VjBaYq>mqh@L9W{mf!r5RhBLnW5j3IP3h4<s-sRP`%La;lSk$Xn_M)d
z<)xz1;^O)-qxtrd!Nd4WRr^A&0abPEmDzVe1#wkE&wQa#Z<>t>ZS6?BJ8_o-cskhp
ziqqHMs;f?!1(7-*rJAX|xs^d0cJzBDY2?$q(6~42^(=zqE$azfn-|p$V&ZdU1ZS-T
z`x?1jV(y4OziLX`a<t<yxdnls;PcC-)th*{wsy&-o=5HFE$=K%X+>rDBV1z63B=Wg
zhL#+qO5)}UzO0VE;q)U?y7Bgl)e`9Vbs|JJI+e;oNf+*Hq>0fcTzxriHC3Yb>0<LU
zJJafqfyIv~B5K!#6if+0Q=C6z_;?}DTeyGYtdXF70F9H_-O$vdZo8(m2~<>HIJvqm
z{&Ix+xNH@BNQQ!HZEH8&11lMS|BRs4jd%E?7u}fz9gRH;qv^QbX#7xq6ewByRDdDA
zGeM*pj}HwEC466!>66&d*J;KVOLaHzgkTj&nVT8~^EHZUC2Y+@ytfdZetY-o%@n(<
z+|BQtLnIh&E;%O)MD>wLh{Wi<T(FN@Dz%u(SNy&I{tNK&dl_|xbkVio56)eslInp`
z)^i0}@q`c1JE_LeS?^Om5jtCj7xQ+#b7&%TiRd<vFD`m<8l^XJKFgSP?dW@fp2t`3
zL@UTQY)x6T5xb*QRpkhuU4PCy%1XAX_xVQY6gXo|ja6<+?S-N<faXa$^NI$|690kz
zhP%pvcsex26Ybh3yT(=3E=j7a{N=8_U5NUmuSMuZxBmOZZ=W>_bpb5IXQgvmK6BCG
zy#2mo!lfxI@wX43){?6F#qIZ)Sc|<TWll<&RZYpTHw(Ff8|kgjwSn`CO{W@!$@5I9
zaOW`1Yh44%&l0I6S@(I;!)4of?|VG^rY_#2o@~{{SLV(B{cMx_(mN(uHI^Cq6my-@
zH|rdC&Or!f#~tPAC1ZSm#j~#~La$LI$0}N`8=2BB8?=i6=zx1MWvOSI58C2mPGL_H
za^BZuohd=t04Vthm}C74(yCIXyT<CRWXgR!OUl4wmdf%Vk5Y8Do)4pP8f|KO1L&xh
zr_7S+(>PZ5rgHbHCKd}32|?>dk)6YQ4No0m*s&3({*JxL*+10*&GtgCIETS<Te`e3
z+~K~5TM{ltSjgTy=eFJRtzN5J`M+*?XT<-qX!I*!2Y^wZCgdo`Z}s*hMxL?3)KS#<
zQ|*A9j;#e-VXOn%kvV*LDvYn!eLQmT3cAa`nKMzp?JQkg=*MrFVQ;%A0|Rj4&4wn~
zinr~=Hkv{|&RjybLQCiw`thmr>8n`vbET@=E#4L_D@{i!y7Z0DN5s|>j?(z&^KO*l
zO$W(lWj;i4{)Ew@!O-i+wuT?Nu1P#6FBfKaJ<ahE+jb7+7sO1+PC-Qdg@Y(iYrxnp
zkjg3Y+08R-vD)d%wHzM2$4%5&{LC|WQkuh~Z9ku~It@J4JoSt+puB{yyD5MF3KmY^
z<|+;PGak5ZGS4bmR##bUt9NMZ2*(wNer(I$v7FbEDZ@^whromzY``wr&o4eJu(R`&
ziBeM*M-gRBn0|h+l%FIXW7cHz!kmmG@!8e;Z`lc&^1rl!!z34KFg{$sj)O;O)1I%4
zig*3%VYt7JDx)JXm%BcD^zmHtGjzAbv=NHL<N5Pgd2tnAUvj_nkA0DeyPX3fJK=}*
zX6v;{_J%2q=uINB-E62SOfe;0WXC-m-)9>J2CQXSvE$}#a<0Qe9I{lZv-Ni-+h37e
z`5gq#_HI%+`8%ua98n{h!Za^b*&~bEH2YAvB7he+{N?SSOI#l=%50`;n&e~#QgGC9
zau{Z-2|0{Z-Ynmma_Vltu5*y14q=4tV7UK`ug*0Du=^$t-Vq#4rDt`Wa15ZHy@vJa
zg|!PP5=V0be$bjnxBl#>Ynp%w`YQPsZeDJfnHt&oEx<x7&wY_nV(Ue>@n1*oTCej`
z31*blt|<@uRZkv^D{0a}ku;emv?|)uQpA2n9?&AJ^1xB$4^}xd3xKY@X3<p=$7Z%g
zSYcsmGPoy|Q)|7)7OUq_MLf7{Ea{CD<QWUl!Ctm+-HSS{y)1o_l!QQG$5MKZ_ffA(
z6D@`4*^F<z6^4V+wK7WnsP6@Uf!>6no0<;@*15uY-{HD-vrJSlo}#VnwvUQZ9TtIg
z(4;ar>c&Xr5RgcLB9Thez>g<XnBTZn`AodOgnhXLK$)^gHuW5uREONNY}P|tWeVa@
zi+Wa9r9vV>#I%R`3)}xUrS=mp&0vT7%A?T6emb)^DoPO-g*{n4csyqdD?ALoX->6w
zJV@<T#BpmhN^HD#!|*_U0-<0sj3=d%4)J!gqP)gJxxe$(J&a9)Pmwa(S`NE6b3y)9
zC~;P)X1W9V^p%;4(pi`80v?-&>+9BE*Brk_rlo<dpf`K&3?OEFtH_wUGz)!H%MPn2
z`6IVR^2CcQp);qLXYLF1x1*Xo*}pmu6qP9L_8lHI*5mD*nai%OdbeG|zV;o;tJVEg
zX!I{~%q_|oy`ZU9;EiI-tx^`HB`aCpfL2pCE${TXG<QP(O*L?U?s-BJpv*PXCe;58
ze}XX(qeGlPRR%_YpTKaM&MhxI9-Y|$QJryQWSzX@8$CYX^(OFPTsqs!=cp;7z)j2E
z?W<_k9n&>&YXbNrtq2j>t`SyiO$nfOI!3{C)7h=t5VMWYc!%of&8&me^u)0$!@~eC
zWWW3B`aj^h^}q9Kf>6{n6ztTMo<s17(C<Ec^~^@8M=0C|h)3b(!nWW2_B_O3B9_ix
zcc5bJr2{%#5pkZYPPKC#l}%*f9W?2k0}QHletFoOl*Ca6H#-AiG3`Imr`XeHU$^6W
zva^Iu6;`<Aw$zpZQ7#rA2fw;y$>5QIWzQdZ^#<R3^~Y^^GSy$W2W}glW(0Mg2<&>(
z!qrFFjFV-(AA%-Qj{i+#f!zL5bZaUN+zzbHQR^F?nGU&G`#Gv*sS6ozkx2#Bc2}wI
zNPHeao+0dzi0i8mfCaxCST2Rzddys~SNsuJGA=!8zb(FuE1AzZO!E3I5K%57XuGp;
z+45ao5NsA`1qQ6<eIOi1n|=frsq!wi$8#;sCr@$HZB}n0K7!rU1A?s_>`#G7GkE>Z
zP#Uz~W8u|RoN)$1&vUvD1Cn$JD)gj42@_FP#-}##K7GHgnEoU0mFmuj`2%NpgXVR_
zPrmG9joVly`v~+ehHezXC&@^fzU}mzNH1t^_Wj<e{=<{izK&o+JM32oe;XL4NDwLq
z1>D=|A2x8V8gtLCtVU)mLs!BBilUyG1RuK{b|_CX#3875*3-@mqvF^M%3a;lQZ9w{
z%KM&Oez!jko-oZKWD8Vys&ksPr)q8)z@K=8tMM`9F()J9DO&1fQqA&IZ(!lKTY2YF
z77v^6xvWr{ON1^&g|#TMymz`6A8Dy1uu})gRhO3f9*7dvY5IO`NFp*9@h2ECrJlTT
zQ9Mi}8|MCXoxO)gSv`0H%%3RRgNOUqVVZ1R;OD&QCJP)bI&qmq=_>YIqFIzg>E--$
z;;J|DfjhztgEW2r{xMHKGsHECg8lsWv(>kFvt>ZiBJu)M<!4LxIw_=_wAF{e=yl^J
zlo1+8QUKBE+hr2Y+TyCrO}AeF-6^PKxrM<V%;L_{y=L3Au^eTkjo(?HON=;nHr-<B
z>d#!}QpoL?V7{6ksWQrHMq~GIg?CE^z^i@#1B6Y0<Wwk1LPO;Vr^iLUvehev<%zO5
ziBEu#&D%b66xta(I7e(<hS)Z?I6mA&4|m2&IfWZ4i5CG@TO3qh>iM|!>gftMi%aj5
zGRK6r_g#;*>3_ohFEiUm8h}+SC!G3SQuPg}2vZcy(mixej)G4p!p{_7-vJD#t9Q}#
zIn8H0V37~M-H9SL&8o?I*Ldm^*sy70<^iF0`y~Fp0B<T3UdwyE4e=nj74eE4JTW%S
zhC?<FH(3<PP2Jt5%#+Xdos{9FV4u1`hApBEf5{t<*o7wf1Rg?6Hl2taCP<z)X@CJr
zAo1O0Obs$^n*dR=vu<@!PRk+sAs7ewkPmQXbXMh|h>GT6rN_tQmuV&Hih80SbJBiu
z)O$+%s*+_R1+1(Mh2ZX$9C#?4cCaGU3^}I|?0Pzm#?$dgx5~|6oZwq&)2E;P(Ubcx
zSoRtQHS4_xmqI_}I5SoDQCL+8-5j{=lV{WQW)VREL`=$CC1<Y%I-w@fneUjilwzIK
zH9iXgKR0uf!UG`$qi2u4qQuc-gwK5Va>Wi4CN$}(r$yw|!<?{`Q2%hwWA4c48B_VJ
zaKIeD^cQfC{p?U+c+qO|fV^-LcOIAd<Ic$d&eVtow&Gg6q3r;zTm?9!#MgGng~y)P
z+cW&^sazogF2y&&ym2*e0-Rdaa*Ds7qKjrTN^fHlGz}!1UxuP>HKFL``JO;+&A^D4
zCz$}(up7M{mxp507Hxmajx}Z6=}g|grfqM907@IbCKN+-#+>tT0=B7K-HIIXES^bU
zF~b-DD?Y{Ir>4Z~$Z~w%sRH6Uj;xlt;k7KD*v~2oRhwch!kyPb(~j|D#%)2o#@ArR
z`${C3=GoOpT8vux^<IPuMy{nv0w(^`F~$BVXZWdYUkH+aF(C?rdE}D{CC5*kaVG%p
z((?c22BYa!Q7@3}xS+cpa`n+o&jFq%+qLwmNb#rIPI@*+skv<E=_wmy>C_u3(s{<}
z_Y#@;KaIowmmsk4l6)$ZA#d~8><z{e2Px*l+BV*Co9n7#PPJrwMJb95%yK?ngmlYp
zAdRumII}AiB?Eq&>1~2)PNHS`*^Z)akdS_eexY;R1M-hbBBq1<P7(f(bIp!(@nDVx
zCI4T*ilc;JCwr#{3@Hhd3mlFiR5%ClXC>CpUt@f-_RhE(XG~US^5vCQRp7x=RdlnV
zp}?$S`5<jVz!@SEeP8maDM8+M4eV7)fxq&Me95_<l5rg&<>EP9fIi;*L_L>y)wnv<
zSjT!dCL95tM=K9f9w`w2-YpuJYXaUk%d?xoKX)v32CD%k95#_ExrC=t&!V<rla0rn
zUR@ZLi(jn27juZ}%ps~BVk-2uY{#wnGXZDN8T_Newggf={<Z>Q`Bk(vGKhWmaKM?7
zZI=<#@Zl=n0_|4L3a%8=f^Q6vrK2}(#-3?>PDO)~<jE_6o>Z6IT04zyWcJ*AT&rlg
z@$~7X#o8<*Vk(0}R3{Hn{UmlWqsN7pCTQ};YvYaCb8=yu_bl(=j#&qe?$@TTv>^ec
zg#LP^hq)+TF{*eZh}VsanP>=n*3<U5Mv>3Y<-W9R(ht@IRN!Fi!s@M_S$=g`J!{N`
z$0*>A4~T#K^_CBRA@L`5@+YM&m8~Cd+!e0!?>_kbz>nsK@#i(RVFrJ7+Sarc*8fS|
znJei^jUR5AeUYGIa@P+CBE49aosc3JFW)%(kW@qJi*1K1z@B%d1s{0u_LAr>ThZo@
zJ>OSjNH>Z?RBEbC{Xu%KmS?YKT*{|cR~qXUHF=uxAipEF@rbhClOo>uGzE>Lw!K7c
zLq8Kx`KJ=5grwtZOW)X<@9|t`hKL*GSsH2eB040?1(>*E?my4p6~q_`2$~3p)eZ5V
z^<i|VB#^CB3N~L)nBtR;r+HV&?Jc)gLRwk2kP;am-?C%h7y5CPV)gEfF?+fLfMO)?
zpE9TqjIWgdchax4efr7$laN`@SCv(Di0aLjxSF@%?1w5mcIMTtl$mcR_361tpCq5e
z@;73E;yPM#C=KbQ!07IX^F-q9=LLqQgHmlDiC^-iZ5GlOIQ@2LII*W`_6aE=r2NzX
zV}hkw=vj1huV7Y$gtAfj@xeN9WV>$dVQ4ISF1gXYwKR0=GZY&$F5w<sAh(+Gwpc$7
zP~{3i-Fd3Y5QLQ(Z{B*LBqceR-=znCPxh@K#!Qsd1on&-S95+tBC0NrWHpKSk+$iB
z7JK6HuLkR_r=mQnl#=}<IBbKhtNlgmq=TFj)^1wX(3=ttQrJ#{Yt6be+KsU!UQv+y
z63Nt|X6@U=+s|OKJ}_O_O!G;~jCn<BKi;yp>A+AjVi&6;e92a<%1l5jZ{>x-rCTiO
z_h-D1V%|kcP|mnR+3A&YarjeD1$Yf^i!^=T={J|QKm79Yzq#A~JJ;wN2BzifDAzxD
zFxSslX7PM{MgD+@K;d9Dj^~wpph=gezusYItsB%|<3CjSHV~SiszvbUaFf6bNHUJK
zHDIh7<K77k59-#|H54g@=ZEjw_;p^P_uuU2Z*0w^QV{v!_4!#DN2F<8RtU+)Ush$E
zC#HRk5o5KHZ*LeY(|b?KJMd7pT7|*(8s<X!ctE6HqNCsue=5Rm;fpvvRg(uSS|-=A
zwS<zXGI#>P`$l*n6elUlhEJsGL}~~<R{Z?#eTtWq3IwSKiRgK$V_4pcPpgxVAD=i!
z6e!s10FAAVc)d4H;gj_dhjS;W6bq!sT#3%+Dq_Rv>$4iaQitK48g~N^o;!$jrbY8;
z)ny-kbcXPn6BV|$GE0^Cy~UJic~Ne1Trj)&@o`1>z>%0c0^RrhMAevtH!6L+i6|{T
zeWGmZvh7R$D3SAJkKS^~vS27ZTEV$$4VjbE-DQ<lKm}j}Jyq_WAHcl|^mT6>xSjEW
zXSa;!V&JFzg;v5;J*UZ`UXx36{4b;%zb}M+$&B~dmczfmS-(z1eK5PH_A`Wj24+)l
zfY{J$$u22&`Xyh1ADLO4-j%hyLj)XS42@J(r%j(r6R<$q>Tgj**;*l=Na{Hj6`WOB
z<fYgJ)&-dz)kRr@C4rVpxOb3U(8+Rs+<Lvpb?=#&PFiQ!B3a(Oq=ty`umY{I6!lR{
zojHT%+M>_{F+xg=k@4gh0dGy$BW449W|tKt!-xVueW5dIcHbNgF$`?;3SD}-pXeMz
zN3`F3a``)FH-Cr93kS3bvy)JrOW(;A#>P+&V%tFPN7hETY6Ybbl;ai$MW4V5?{d`^
z{wjYvmsN7a)m1tiH)emylxUoUU9#?d7X5wkV>{?TS6buHlnNczwyAg013+Nee@HTB
z{9SM^e)a6M0`5X0x4roxqOOQ%GnMh7xiRv6@f;m-X$Se!7OO8GF<;w*`BuI3%+>#a
za_%8)jD7Q$)W7jX<ZFnq`Io}i&^v6wthpnRsHmVHtEK|6n%Ds)%?4BZBaGo+>1l(<
z&&m2@cpp0Ql1=VniVk&$5%?Vd|Cnq#tU6cu_3&AUX$cT!9~dcq=|?;ouR~jECVvEL
z{hSd)y2aKq5kh>EHcs!yc{Ek_U~Q4Alm||Ss`s*I=a_61oA-=Yo$>~~#&(P~6?gc`
z%Mi0qlsf45-O<x|sLRJ*9<;hlCXF^m+aKv&#`Wj%;{8PkGGnG>CFj2vv^kf(cKhsU
zuD4K{Gao+Zx2zt196fYJ7R6o6E_5qhz0*o5%MTa0-LlpoW*fR-qnNMUu$*F~B6vfx
zymX=0Gf^g3QiVbBfMQ98?Sn%(YW+(0&n2?4UKo8#^IR|XWu&kxt?AvLKYOI8iu%*H
zfAWj1NLwTs-cbw6X`mLmnyz`z;4G&Qw~INC(9Ks3n!(oJpIlAHmyw2s-B&FNK6Ik7
zje7m(&VH9<7f?o4o;t0|ZX9d1EHi3z0t;(O>#~hL>)oscx0LbXt+F@MTG}4nV6k~Y
z%Ie}UX8ACoU7+*ScB$(H{zmm7f=1Cs`Q?+TGDDZ(|5nG}URS!qfREdFX6+z5DYM1!
zT#EfR7Sk~Mh@|B^$M@qy+KswI<#@BAz|`VXkvsC{O40@WZo)|#hBQHT6Q>*LzdI$7
z40!0^MQ(&R@@}zaTZ5~YMu(tgBD+j^qU|!KElFDWX~T;Nj%8Z&BMT~0F6RpuZk=LO
zqNY>4*~$B}ZhTW6kmUW{lBE4Eo5Y=EJI#BQ!39TLh;-7ehK0%I3W%5Io>_D&8Z@q?
zJZ1ffCP^aM%G7nyxqiOheX^fId8WBgO1$vRyL+#`Og{+uWq|abfc&plg83i@C1n1+
zR`Z|Srxb1?1icb)98)Sh9ug!<B>YS8)J{rZkUM7aNQkI#26^FXwEWpoV{XA)T#P>^
zj!WwN_lE-hlCe9GRUO>r<Z6xn{5eT3-&s6Dwj4=vNVF}O+4On_&3Ah;QSPZ=<+r~m
zOP|K-R)H1m81A^Iru3TDuZoRzbY#S#+wgxec8y;G9nj&Wj8+=tsQmuM53vH4M^_|O
zg;>8^=Ik88`TNikVCWAKIdI<vJtv7l(@`|VnCMO@cM=M9zdo9Ur7ns|>d2K;p0LUW
zR^OmDFm~ofT%Je4nt&WJyX_t?;u?a>tOi+8htTuVCZNY>lwgyDtsW+qfDm?Dg6vGb
zCZY!Gzmqqx%(boc=jHryU4MRI+=GV8?vXyF{1+DK8#doXlBG%*u<ErCyU>7r7)EUb
za7c>tOfHYeBFzXLZDnCLkPC%J5YS!%VOmz3ym}Ir(TtE%Zr7pm(q_MX#hlwY0o9S1
z+5<?5f6soMa%lnHCjGC-2DErm6L)nL$Z|y0rRMTmE_eI<_c9-jdq0c{w8fZp`NxcZ
z|G8KG2iRX^B>PArw%{L{pb=8%lSE>RtatL+2-yW(P`lG|KbPE=fxuu}gm5T&VrfQU
zP%Vob^W@WQKq@i;b%AxyTc&@2U;cv3aKu4U3dct8Df9pRbN}_b?Fec8(1|{{tiSx)
z|NF21{|oppE70Cegruw^JgGpadJEo4@cgoo|39wseZ;6H30MA`AlMgm?cB(N^UUBV
zrVjaqTWjm^*yX}3^V%I-!<W|yk(R6mDqBV*^kES+UM20fVUj=w?fN|kzWU+YHV42n
zDL_(gMbKbXBk_{#XYkhn_s~FmJZgd&?C6nrpW7VSnW2>jh=AHB7X&<6eqKN%M*&jR
zVP^D=L-vTj^J4$wm2dn=+1tIM^Sf-x_zearW@<3!88JrSzO)GJAVGu=Yy#+5^6ic8
z$UY88|DX&65^K{aFt#b-678_Jo(CBiuK_VJ80cXfNZ*3w0N8uxfxm|oBbXcRdm#a8
z0<a0~48z7lyT9;rzfCz1hGCiEs6?Zi=zH<vj;2^7$q51Nae3|8cNInelB9qqJBG|8
z`QHAP0Ia~qm}qhOKZy$dI_&xN^^uG4)+n0l(^Y@FFO}`wzPFB4u~z59LoBm9MpuD?
zX?9n8v^z^J`uPY{-P8y*AQwD#L%<V}4M|_J<6K7Ay`#fOq>Jok6{Tp}9axAEz{bR~
zkkmGS7@f!h&Nd4uj(7!zz~#s4Ga?4KiAH9*K9wC~h=8nBdPSX%l0rw?d3Xp7zrzO&
zN={0;>W3^DKsO#5P}Th{DBwSC&%fP2KNBQ*%w)5j+Wfa0OMjWjXD#!_$20{v3Wu)N
zJU(q&1+pwj_z1|F0Hl=VG#!>W<7aRrnn8?DJxRLPkZJrV?PufTeIHdkcfSny08U>G
z;w&Zro!;ca&Zp6$KWlW=uM$+f7}=GCwxTqJz~+Aw4gZgQW&zHwMH$+@Kf0`M1aL&(
zJ@6s*;t*&O?5q|*y+pyX(^Fque2R7l{A<%4IxmU2xa=Dzt-|gqff$0pEt^vOHd+Q`
z&#7EQXA8<<0my&2H#7Ro+3NrNfrb^M_W`V%(Wgya@V6(xFA6mhwT7I1$hqSF{dGn)
z#MEdMR(=3IM)9jdRXy@Llaj?I^CQ>9iw5pA1;G(HG``$ju`j|4iOi!MEL*AwWnJd-
z$E$`vRsd|vFSrXB>Y~#aMHcQCs~woiQ48!$MO?PDNJg=*wV}aEIS}*Mibs`ysXT)K
zx?=rxhIE2*Vou&`Wt8&XrXuU*tLf^b{N<ab@qk@dvxoB@mEHQJ5ezwm*%`c{WJ|~a
zU$X$A9gCnD39119?0Je!2=BUwb*!bSLEM_BfEwwQZ#|6BR8F4=g+w(OuQ}@*WD_C6
z4l?f1$3;#J&h^W|?<8!Cc0dSZkE=j()KHP}(-`4bye#=%;Nb%gt=VRaz0Y_zhwTKZ
z0SB`BCZIOWmMZ^7Hh&)@yKDf063o3EB%ot>16hG-w;=+=9(a3n5vac*Z{dSzXX^-g
zK2dVrveSSbAY5l!p;t`TJ|9|`gmLHW469L*s}T|x6e7FI=EiNf8dVO5Cqbqs)<ubt
z<Sae~c8BO`$8|H}htr51hYoe;24BojvYRoStYlQ&myaN7xC$w$`64rV8Bgb5EN~bZ
zjVv4)hLV}z&1!Gx@lgL8b`5;y1O6_o8F@6BUFV;$-k$}_@1@mWUl{^T(a2rie9qs_
zRgS0w)uMPQToc-tEviP6G4-U9GcM8<c4fNDYp0fg$<ISsb&N0xkwYL`1>eBAFF{4q
zcZYY3;-PSuO?8BsLeN19aaJVMn&81v_Nr)~)fXSk&<EyWUdH$IE`UBYuu#VwD)gyw
z^>AX$jH%(RQ2WT!Qn`mId<H`czN6zt$i}}pFfB`0zM97Wl`jZI_JHZ!nIiUl4)V9<
z+i%`Kh55*UC(F6~UW>$RSEH{C!&|Lx)i78Z^A{IqKYC&i+r}XStmBY_^{h#!4?maz
zv~E&;G)P{E#Zu|+B3I;AE>z7Q#UuA{YiK+6>&+qjUZZ3Lx^>Eao6J6&qkQa>daNBB
z(fh#LTlDfHGx50rE{5TkH@p$ihv8-jgUvEUPArs9+m${8XpSSu<iPMzn6b#AAyT|8
zj)KMW8{0j8cd=zZFJE*HPGP@CXc)821a1mRZfXZNI4d5AnS>vpGO^fCsJ_1uw{aKA
zqQV(|Vwg}~K4f)X^OFeTZ`#&rK9(RSRwg&uXyDW`IVL{_^{K`oe0<U2sll<m6U<NF
zBei5WUvWcI>gp5iiO^L~l-Z}aN5?Am*WPv}=l>7x^k2?+Z)A2--bP6Huh#?~lCnlf
zLa4C;eUSJOleQygn@6foDXF2F(X+eoK)cOalo?}!0$_aOPZolfH+g0QW4v{DBqDSa
zW(+P(E;yo2U+qc~X7|ckLsT*v$0ivz3Ys2Wc~pq}cQ`7Y1h3@N(?^d|<*~=2@+Mpx
z*g}ewrRFJF0-X?HxPBoJ5`Ct+C^2T3dP=X6gCCmE=kbPMfPDe+(=`;Qe_Q2(NEHgx
zacfFN<qO^mna}hm>&_GojL{B2R?C5-RV-sd*g;5h8--oOg^efnK=`?vjS{&CYquKW
ziuQap`N?4~enuv_)l8Db#J<R}XwaYw9gg$90NwoL+a?8+F5rIl_X>b69W8ei?|~x0
z?Cq@=R=BX8G|fjgDCX$hTDAH(2<!PJ{bF+Msmfk<Bp(akjPYvuh^liqA5kfIi1)Ke
zFGXgk^hKgHWkThg4=^z-p%+ZjU*jIrU1m@(->l^?ZXm~;@Hl=ta3T{1V0^|jE3m=1
zT?`bY!0F_5iTmuj-&@D2=DjO|18b+G&Jnc<5*YzH_J<XDhq2nY4LQW=WpF|-*||fF
z&jG*8wu9+mH{IRtu9)|50d?6Z6c}U^<Ew@_>>3lDYY>v<?wKd5Cad*tr|o|~i&ata
zB2<4!Km9Q3m3XZL(jUjK+^WuN4pd4IArpGf+r4;KTT1=q<c7e;`(C70Y7X>viGzYK
z<E&HtlLXu=yL;Q5Mm2!}GxNoB_Fm%*y}(b`FXZ<Tob^l4b8KohylqjPU(_$xa`bwg
zOUw6^`p$v%uDw<|s(0oVWqfDj-X`b?HZ@<rF({ZT{QSB_?55EBq__mBiQ?hRT~q(U
zX5MBIhY4mb;U+{qVgvFc``K8C$BW76#V6P@-EhADvA7lND>74)pUn3|t$FtSt@s2S
z?_F^EJsx#-_Te$dv!C@Wy4Hn~v}EHoKkFrfN-H6>by?1m>u`&@aC}0M&{=|8H48rk
zG6RzKGjodHEN&PrOc?e6L;oyQm-&=zZ}N@(B(B{i_1q^3sv>6<Ztq+EEC&w|4@q^^
zgamBPJpH^QAoM#H=a2dQ?gg+==GoTr#@HgY<w(@ho@nON{nn+S0Z?Lm5Kr*!95+2T
z&?hVUrh&?ZE}wsbqp|SvisioKHsAkME5z}@mzRQz{etap09U0dx37rUzJqyKnj~-V
zC>L#$P0`~yWgIO@UQdX=Nmk=z4V=`Ib`g1!I@Dt#%OFfKtJrF>FAh3O#g|qB%qYr<
zA?_siRQ;EkQkCV#gOS6z<(Ab7g}m|10W>1K=?RO1#xrmA;)Tq^^m7<wJoELhS-To&
z?>JtHn5?i$PpBK5X@x2Dd<Si&9ajUxq9{l~oq7;5+43;t;0cc`DU%pKpp5J0isjw@
zXu`s0o#$Byr)qK9`DIhB5s9QbjVD_NiIV3JNiv^juJh+~tP|n)DB{764`v@VHM}pK
zM2Y3ZAAe3dG;ra@{)W-L`bdGT(9$5zX7<@AJVAIY7saY&*p6Y6+i?eyh{-BAS8KAv
z(<|~K%`azROoNs|IN$QqV0#BBDlEBA8}43UjvzRlQPhf5HrBM_9yvuZb*H%b-jc#z
z>)y6XmHA}AI^&t^VenUS{^J4p?++64y!mZ_BgZB)neA_e9F@2ezWure&&J+-<#iwq
zl(I-@3krY^$X$Hx%T>e5pu{gmfv!sTAmX;Dgg<iLer4&_^zoL!fEaEKS3#DXt6*O-
zxSM7@>}Y8dyL~t)z@nCa05(ZJUd9IECGeh{B;+@kVZZ)Ha`8xuq#=^TKscZ7*mQBk
z2rRGevl-je2}QOU10yZgIx8=@<co-Pkd^cCf?wVw@d+ZI79YR$N?YFn<{W+V1#<!G
zfce;u>xxyudih5z*gA;)w~|Tp#oYnMD>2@HsMcl6xab?(dvH#z?}p5pBsm>NQ^-QL
z+`OmcV6t6|oC8yY{?wa;)m0EU+8Xp><^#rUEQ(&YN04<&N&4wwlFSCGrR^;Zr8MB1
zkih}#xzGJ~LxO+awfxGtpeel}=fn9g*vj}NRb?Y>DwC5fLtVtNasL;fAZe-%z_KYC
zxB|0@q8MTdvJUYc6{RY9jG=gPw+qy@Qh~#8z*h}$ig`~A^bbta9wxz1v-cc_u!_!=
zaZ1#oVI)6r%yJ%5Auh<a%sGogXNVN5QiVTMJaH><RHAd=Y=EgsNu{7{zn`5XeRb8Q
zwDE-jePUjPhi)6t1r;*VM~jHn5eQDdREw!bx+OQaU5TFRDv<K7AT}tg!nAs0>Qv4^
zgA9xm?RzEAf+G2Zgk0HsUKm{0q@~b#!2SJbqL&14vJ3=l_@2j<_5IB@>7Q4K0NNEz
zkKZdn6&?a!a(13`Jh;bsP&suB_Oc5p-VZ98gOs}?SZ^sHi&Pt5XZ(@oun2vD3)c3-
zSA&!fpwJTb?$dgT1^><}TAg{c_y>mEH)8i)s$JjGL@VA3?(aqlK(FPP<Rj4DSCFAT
zQpTRZP>$W%g1+DWRCPr6n3F(Y0PMeo`i#5Zbt{gG4mi{ZOI|9ZGjboz{c`2vSD)xM
z>d&(ovTLvV_AiSa(icYsfG)h>?)kffrm6UJu%UtE!}rPL*g8D1!|~t}hukpf2g|C3
zuwfr_Gnaw$I-}&n|F_cT3MrBsJpW`Bl7n5I)6R#kjDj5vCAOA<uUcV%0>l!1TQ@g1
zpmX6u=vRCx7b)GLJ2rGK=(uqDV~QPs6c_HTgUN`G)DJjx2^@7EINbL}l_K-9T*$K&
zeJUs7dY{7JB8MryOgObFRp5UA^y<?p;AN6KA|myEyAs<<_9;N)VpLvd+wz`%xR2;>
zACdJ{(NY+>eAR9x&>e%oIrGl^@u9fg&=T9Rh)pr_8bfica+0dYE&cxqTmF9Um5hxM
zsGwQ({BLdmmANMtgZRDJ)WNfKM6ys-coW#KK~$H;t7=^p4aGNMU;LP866jsT)Jj6+
z0P-UTB-ffDFo5wc<br<09nB0Y;(G=f=G$?bNEl@K%Cm){19Q+R@KaOBg38JSs3lIi
zwF1uonYEst7K3{oROzCFH3{0)NGo@0A*RK$vZ9n3vDCRwaQZh4j*B_%JWUd#hoO;!
zG)a~ilv~5(fR1_}Pb;>KLUJe!vJwu<_Eta#@KNd(-VNJgL+db$Bw~g@v{hbJ_7Yg!
zOwX6JNzU@pv5A(byj=%T*gZ-FRn;d}md2&L<5^^Aa{(sDVGxvFu3`Jv8s73XmCCK$
z0>Ec;C#kRl>s-42r*qPOKrVlpDagwqy7uIxf{uTs^tS5~q0O0`+ZF6CIe->2GjtN&
zxHlzML5fJMYmoYA5%$PnkoEHYfcPf6+H3|ghnC8*!^cj0LyP251P&wB*R8X)!%3zQ
zd}uy>jbV5*xPBTCC^xr8kHTy)!81HFe-ATX25*cK7YX2+RmD6kISZOD3cBEtC)C!?
zR+8wIQHre$xxLpVXZ9hHcivx+$Q4iNc-+o6j~5vnAIhRjCkPM3H5Jv->f2y`L3c1k
z?ixIZ>HDgCg9%c5lZ!NC?rI-%I0GW>izQzb#)K@I=_@GhGB`*AAEVMnLV6IdH*$o0
zW+)6!%8`gILFdM$C&HfwhuK+oM@ZzbZUgWx7}5d$DfV6>YoqCIAWE8pZh;92LEn6T
z&y*&5PuWj8q*fu&aH0gs0|(yFTUQ>dGAM}+3~)`hNn`_^zN$x0dkZL>xX2tGyV<}1
zt9YXt5O!wppg2731iiTDzS`R=Z+Ihlxq!)SuLPa+z%9bVOl<HCAzdY9xZij~fse5a
z=0O+;ZVH^qpKC?7l1d}7=)6;os>BX>jL9MtK{W1eK43qwQ-W!Wo{?9&Ac^T7E09j+
zZhT3$#IZ4LSPU9q!Y!^t`1cd10~4XI3t8Z55s*a+al`NhPZ|&QAu*qaI?L<7B%+X!
zN0(SPj9~qL>AU{h*L#<tAM<G=xKQy6_N&Bk`rao$ZH=0V1wl2Wci*B@Tw-DygqD!T
zb<m|d7xKcl?4b#KRiZcuVX~)8{No|z*_Ev5vGOZ~(^7K($Y+KnHr97FbA0o<q20)t
zLwo(!a|ftJyK4#ZFfd%dk6M>q^rIx|*eI;nYf&~qVz=ed^&ZKac<#fUaDHnP0sEiD
zv3iW1om(jw_rW;m^vFR<>0AYI)tl&Bj__#x+~L#I2e5;%8O;Oez<UBgrVaU#AY~c6
zu{H@2)`4{ob@HR!(~n|p)Q(JqS;)-jOfOqkHIoCiEhZUs^@0OXNi&Od1hyj?wYKP(
zY;Xu6P_UF`Mc3H!r2?Kq>&KRp(pgy!%Fndd*3;nMaMo+#?l76xLoqcnf6b06r%72{
zhkF6i#$h4XwX=x?{}f=9)w`_0%9>#OMfvdmVec&is{FRKZ|UyNMJQbh5RgVnlu`si
zx)z|Mba%I)f~0_e(v5U?gLFujboV=#|FieA&vVY%&pzkN`{jJYrLMKO?>Wak<~6SI
z8=t_`%xrBqat5eKo8y;Z$o3$CdrdCkRo7_`CQBcP`V_x+kLxrD5PDxX5$=Pv1r(qm
zlm_(Ku2*do&sXDF)mq@Jtqh5Ln2{Egq>b<lE}qGjQA@x989>0RRqrQ%b%;GqLlD{$
zvmQT4bTFwbL9k5=>@i;6G6^e;GF9jv80ct=oweTJ;4D1DLM^l3?KMJMpzyJu=;Az%
znb!R`bfKS*3G#oBdip=%=HEx6Yj}yJ2jkxs6ptT<LBS4?u}idnN~iO~-Gjfd7byyz
zONvi{<TYAw=8JGv^!!+jv=`uE3GoFCNjOp4Owf-qVsO6DFl4hlUan?24qI<i0f*jn
zxZlS|6&rB1@}fOf7`d2aID#CZkINkJnPtMzTdPr*6lO^og&37JC0g0E=a&p3kk*$k
zPt7E*60wf)f6%Wy_6ga*mby<m4(I!z((Bd9C8{7tSIxc)mr!$|Ygw8{Z-L`^d>+Vx
ze1wn@t}HD+)4zPtPYAgx=7l4YOQ(1}Hl!h#)lL0q$ZtX<GJm#Q?*UBHUNLeAPZ^5=
zD}mv{N_QO61u>Sd<|N}lvVTaN5Z$fcTegJo3F+$#;I(mc!w)$S#V@wA8IVp0N{GBx
zp6A9y$6oXee9{(%+?EE_x5Y_ib}e}2>(J_R{p?l-JxpDkL*P){=v$lzZ2^6UOAOv}
zS=5zK>NNyuqCLePTh^7~c6^ihcNZ3z6<U+l!`I?xLhQ<bz4}90;^x@;h96U~L?C28
zt}eVwpyovpgG!b~6N$)2<saM{>e@fbl?O*F|0Pxb7zt>~SBtFdk$>G@Po-ZV#yf9Y
zpf}*})5S>hkj-@&&2ge4t7(emG0?q{{sYqrP|^Gl?rR(ZNrTeuKc0O_45lY0Fmh7M
zdk{cM`o-IP0-Q#q>R!;3lCBjmIm1;_te+WIWzp5|I2axy>mvs!v;mBkr&;tZeOntL
z%~^-8Z5?-^mumyM%mpJ|m!Ary5L)wuBm#jcvD<3&>La?Lw<P$n1Ud0`!v~^*hZvhu
z=piSM09~s)6DeizVYl>i>@YeU+~nSyO!nCf_fAjEya3kOU(#M-oAQzgHMf!LFK-kY
zmULyPF|l?-^EQD9Ud&Y9zd)o1#p_u3^)dTV4N`O?*=p#ywqRNawIMCi1t%s#iNF)D
z2i#eRFXka5RB39Wf?gmut|q+zf_Lh>@C3lt@JR6yn17b`UxnI#_xex~kkNlnFc&%e
zy{<Jz&{VKXA56V8zHWQ={a=XPe}6;95AdqzNjIDB-(Nx^Y}%sOwSXs)wxAAC6Rxh4
zWK^R156HxSRA>LwpXJ!VVvyg8Q?L2^UsIz0syqfO9Jku$0UdQ){Q-!i1n{lK<8^%w
zhxP;={j*-bzo&$M`$3Kn{t+XiFJJ%tBVH4_06C=<4DTd4k1$MiUBdvw?r`Cak>q?Z
zMRG99<yzriyTt#dFfst=jSvGQkMM8OGlQzkJuH7Px!nQI6MZ8w2om7k&n_UVDA*Yd
z`j_^}U$HU51SGamD>-@P|5DcW@53nkEhMABeS1WA`BxVA-?`&|-Dk!`_<g_FWcz3B
zK7U^l|H%(AaMBlUNW<SDHvd)7|Ic4<dIFBYzn*y2JWcv<T(*CFuN9zABp_uJ{U_h;
zU%4DWJ_a}UeOlqjKhS9ZTkrJ<+`=`VjlBQi_y6A|@Pvs63>t@Xi$E?eD93B|Rq0<l
z7XB>;{C($BqAP-jnV~<9fcfM?FOBqLq3U_J?LYwjE%~%LR`JiAJ^$go{{E&i23-UR
z;2hQhp4mBX>Vg7nZD*_P7C~#^=kFL3{r~6R{gb!&kAG!QjL`?2)`mduR(nVZzUet=
zp$Y;garf^`I)eWXdg6a2+4(jz@qpH-AfQ27f0r|Uf{vE+Cd`rVU#iOfM-gVj7_98m
zgVt9g|Fa8s84Y5~%~{_EqyNzt@D+zEWDp6D2mb`3^&c<n-(N|;gR_La2X$Eg6{E%f
z{`&vd75Lw~|G$g>|M$IrU8>*>YSpdaP>T;saU6$xzZ^V&0HoD}K9F<40h)W4l)`^=
zKL#=S0xFWAi54l`U`d@>B=q;xm>l8fdJpP9f3Ua4f`HX}@D1l(GTZ|e3<1`WMF6@p
z{*zCF(>@ugrh#uUV44M5RZeAdBG55l6hbXVvi5=C5I|v;3<VvBOn<~)_JZ;k(wKXI
zo&XmIex5>rv}OoQ8+o1+f8~dgdl<mU!np`aH$`*i;E9C}&gEV^wF1J-#T}Gykps>f
zFMk#ZOK6#4zeR*2^%6i|7nu3d1#B`j@bNl^DNM3@$lvMGoxDki9V5Rfr6*lp1BL3O
z!Ip9Y*f~T%hbq%%VUX43An3!=E4U>WwARtI{;^6KHH!QH2*~6iO)(_6peXN+1D|?o
z#Uq*)qpqR#z~g!uGPqZG<pp${KvuQvv7K=JV%sWsB?1VWoV(wzj{pn{6aFaQsX^a<
zjkB@fwp{=>()dHtqV>zl;cPeYVVzoqXhC3qO18bdSP%ZB=f=uW`@7U|Se6V2&X98f
zmKvmwNT3)xY76ArK)Bb>LJ|K79Nz+W&*1`nv53zVR7F|<a&ph82?)RRJO}TCfJv5_
z8i-qqkh^nv`23nyU?iiFi>6JX&9M6OEBhiM#wQBS+DbQbpVa260LCC0*}I9m!+F6i
zr7bqZ47<XsSRfnJBbEPzB2+hm4_XB3`0n{^tZhqMC{Zi;ZPs^u7n4)1M-G60E|@V&
zsAc9QM@Kbi|852DI`L%aOTsg2O^tWU1TL|l{k7+cw0tmFO|{lrNQ|@nA>l^0_}zF_
zgL)O6|Ls1I4u~4SRqpI1^%|)ZIUbU(zfI@*0hGWN&{W;sVqEc0`T_nt=Ux+#m$pdd
z7sJPN!KMGRb2u)|5a=|xKu@mnHhX^oOgL{pRPF=JN(IyG6{wAnoNJrJ%3f0=<K^$m
zUytY6f~4y*U4r2WT-R9j`;l=~4`d;PUM2YX2Ar1SwUg<+o>PCy4m?`)6~gQFgqxLG
z;M6sTsi$XRfLSm8r5<qd1Ep~v-RA&Zn1(>rc5ZG3wJGN+H(R(H4-nh<euFQfv={lI
z8L&-R|L7wCuV#_}Y=eKuxjsXO;8f@YT>{Lnm+PRD2o#^91NDIm&<p!w83}SOIqJTF
z_Sa3kBdi8@MmISgbMzKq<f<Jt+h(BfY-$mdvhX}e1MjXH?~ZCVL22kXtqRmAY(dey
zChsC+xuQr6aqKHF<-AFc+NqM9F$6K?zzJ!mW|6W+32l$quNk;N=->+Xe1{e|F$S>*
zJzbjP?0sr2j>!EE)DY0}&*-1hC`;|!E*_T?Pdovej@oVy9-g=f;FY0bP$vw6LpLjn
z%>AK+m#v_wHq8Tie;EPT(6xrYO#hi?_fNtQCkD3G;xFL%u&I&^CZTO%7C*p%$aH?-
z{Bw|n0CC#0c1j}utUc7F4JW@%K1MD981-hK61G(7p)g6n_fOTYa=7qqdjNOT!ArBP
zoB-Ad2?i?bxU78$?>XGnHNnA<ZE<mwc1I2OI#%(69ePdYvC6KGp@@Ad%p(PF7ImpJ
zh(71Vsa7;9MUpbu(hw30vKk4keTE#61&ktOsQ^;TbzTSBBU|AGXz%Zyya%AgO#4zM
zG{FBM$22^l@Xbap#t-MAdRqkb09V54>=2+QY-U$YFDFZksd`Rbnm|@v&m~tJRiVeR
z1-CgKn0Y!W+>;Vry6|$iP^|SmC9nztOqQBwe*WICryPD|Q$F)w3G^bv{b>5={D+}=
zy8#Gqr?+EpMfVFyiGXpYlXp_O!(<Gds;s9tsDK87inF8SqJbihF){~Tye$vmKjUJ-
z&(t}{LHV{kY4`zN`E{KSR0I(WY1WPBOy8h#Ap8mNaQj|0!xGW9uZ3hl31D;w00nn!
z?qu<!&})aM(cv>$NN2KeE1<4osAMcgGyKUwJ5r{igK2zGqSYizNJ>iZjTnW4tq`zE
zcSl^{o)^zOcl|lT<nWni$bqtj4<Gt!20zn>pDoha5mS^E?TKWz;25q2yk>f?^&`Z?
zLWaz^%a8Z407(RvXvj^?k%WtZw&0G^Q|htzblUG8X+w!Q>tgz$KRPxZVoK#8a9q+3
zs+as%mDde11Z<Ztwarg@3{XTH;0^e@D2`y)(B_+OG%yT7fygo=8T23>)Y5pacbyyU
zoqxdD1TDa+MluhI(80iw3I(KTfrn&07rnBJzW`%}q8K;iS<S-rnjpMo_B+ib;H*xg
z_=6pmu1zBh9GRan(d-`n=wZu4o<%X%@-epCyavw2v4WeQT*By7A-rG!p(`0GVO7RH
zSf{hPyn<z1GKNuKr2u;QZS1Dwi$yvkBo2{%%kx&@2Ui(9Q0}LSMVWsB{e~I*qP2$A
zpfwBcuClrV$ja15bba6+w*{TgE{vW)I>&%_*~3lEf1VU@)GXHye5?wvlHrntsU-3~
z*!P6v=ISjK;X!o~gm<SB@o{oEY^KT}qE}~Y9C${Xwv7m@fPc_k=Lq8lm}Duyn8<1T
z0tCj{PFZ?VXTfQ6#;P~T63kxmzoXR``BnAdkZ2sE%VY9NwQTVJ4dAfjR`8Ia1w~g{
z6+A~4jmRp1en%*f8cA(>bAMd|m@WbD0k$uI<G2AJCEF)#6AykhxPeOGcSV+`Nr3}I
zC!*o3Am*kCW3_pyUCE2L_w46B&s-{O8RQ+z-1oEE&1a+eW*g(|UpWAvxWBQWAdNxL
zSM{w})4&o8;oT_)K<wd}M2@!fo1P2JP^O}+!tNPGqgTO(4`a?4YksvoX5Hi0cnH{4
z<38#Kb*Am#H9OEhbrY>zy)^GMBJGht+;Lla>iQ5ACNEpk9*%1=&R6c*7pI~PidVL7
zx~s0yIF3Dcv<uhQ+%E%`*0!R+w0L%&AHW{w*@T3(68}LH3q0$U++XW`vN~%;phaFv
zc3yP@+uH3BEm0JSE-6pD&qD`}@guHw$@zY5G3L}`Jmd5IU<qpR%i-vbq7PmpXscp5
zcTc9_)jUa0$(zvC@wA)YW_s3#^AE978X#jtV{^!<k4R69Z|+48dhXia>gfX-kKQd@
zFPP_kAkJ~ij_H5k^)e-j_r#Te=+tpqoShJNsUe}kfwC)lrS}VU&sr$k^h0%>%rIvM
z776BdpizJmCeOGV$WMboP#f(04k|O#e<-fZbkpZ9sTLXwXOM!e7uR!pUK3ZFI<ZcI
zslh=;>GLAM7g^8w<oZt?Avipia$}p9Od7(9L!O_Gqr-Ge)ytdjPzVAM6rm%ZXPmG%
z7fk<gZ-S;Rv0LnoF{GDzzlo(yJbMII29i}AKTlfVm+z&Ez}fdT+P8BXKCwr<5Yl*d
z$me#{hEK9QbSy@`ZlXsB{UvEU74dV~GxE`=W=!Di1dcKBy$bMrJmIHgyK)b!+T-F{
zG@y#U7FoE6s@*VsALwTRRwb_DeM6GKW>eI~N7+v{z^E2Q7HCWhGQ4u`Sw&24qqGj8
zr}1k6rSeXZ^nMQ9d0+?IWjhR(xc;It!lMxA&cI}6PK>LOeota8!MFtoCzQ5W^}uLG
zA{RJt0<UYusbG5$2ZG}A04E#6Mpy0oD54PCYCu%ht!t4H?tD(5`ek1+Az2()NHEhB
zhQ<CG8*~{enL{)he}MkLJeN_DLc;6X$B1jI3*>xr?ZkZj{Bc-o1T(6y#KG0(BEIS~
z6DI;FyO)hz4AcWs0Y2wG_TO<GNH<pK%pa=;(nSgT*ZAY2*?mdvsM!@g>RUef{R!B4
zYs~3BrZ*THr3%?oBk}6^E06>Tq0u;~j$W>VHf`r-i5+xA#S3W`ftb}_@vJRM!hZ0<
zO3ta~X{vY(40L*MxJYLs*Wuk&OFA+&$i7KcBnAU8=mJ8sQU<Uex4=CMGolWCx%NNg
zN#IocW2bw5WE{kd8;jf!gYZfI1=1(c>{p^AZyjrN)rB)`V{jilW+qK+`s~11c;9At
z@n-(4VWQk$c^(dj6@$bUxporh{4oN*9a>v*mibbTjgAU`7F~%bUlU-*E#T*jber*_
zlyL2|^S^9={}vSS;32y8;e^iea3*J1LSnmE$UYb7RFU-dxqkZN31y?TMiT)n+mI&>
z4N+p=rUwZ%TF>)DUH@fV2e-BNB9=ZnFh7OsQ;EAeDf&IvpgX@mCo12AdC2tG6g7i@
z^vtNG+OjsSU411OP8;FM&aP90x34VvJvlungd=Yen4n;&7?yR#K}jd=k#4-gih9pE
zzXvF2<K@FvDK(cKNPr<(tef;7EL&6D)nNOgaHL3)HY;P-Cs90&vKPT#zl|a6b!LKl
zo!}j|{oHp|8spr8Zh0KZTmytkihb(jJzK7nJv(Y&^4p1O?#ch$WKl~a6Q`b;vbf5B
zBIE+BwJ3-8eXpN9YEOq+efmQAkGnlroUt9?{|#L5B{%d}>K{PdFWoqFsB-TV^b?`4
zr%lSkbMG!P2Sa`(iQhkwu(}`k27_n!ac!g%3l_2QUDkAMd?V!vvoX`Lh^EI0a0-tH
z2CNp%x}bE}nP2gs@oM@yW7}w!`(qr3Zjg<ja*G5#e5i#>Dlzv8L>N!Rh~PrdghRuL
z2(wJI!|f;~*Uq{Hm;dKXqXk~nZzdXbd1IC_NdL%~jj6=)yNDDVV*KETR1=gNvtE}p
z+@Cn!NBY{aVe|87=n5Fv`nA&^BP8@nZ?`0<7a^tEH-OHKt62i$nFv|FwZ3x!dxc}e
zgk^J*jcpzkd3{(hV(QHPuMSO<c`S>cc9dbJ-ajqIPO@p?*?>X62?c3vv83;i4~7FK
z4W$8bTirQv$8eyrTlY($E)&j3F--OV8|<N>l5E23u^e#{)nK)l@=~ELWOCBv4Ho$l
zM3e_e(uxg4jo(PBnJvkAjOhF&gy2{Pt`3$kG$PW4^A14|#ORh<mLZpO@4LObi<9Ei
z*dQTaba@?-J>Y*6f7BU(?yJDphEwg{u|*VDXZq^;>)FJgLS+OpJOeQPg1220-|Zk5
z7I5gw#lJ?-#`h~S@NvL1<`jqbX}u38qc;$IiBl-HG?4ele8)Il!%Z%>oi9@mhypIL
zf78C}qPGsjS$sr;E8_!qdgFB%6ku#yhtQV9C%pccF>fd_e)z``Ez8Q0I~@Auwz-Dq
z1aBM*PS<G(O#u1J6esy;6rF)*Ogoxl9k$7HE?(Jcg)4srDj5n*$HyE`(iwwoX`zf3
zQq><&Pu1|^^zv;CS;Mx65R6-rFfe4$LU1-5_Bj&pGs^Uw=tMh^!b!5o-f96SL0Ncm
z7_wEHFG3;g#D!##vEy6P;P(;Wb4P#Y5*1?G<-UT}CRP5{oN;6zEAj*0ihP^4L%v1u
z!R`>(W1+4KbT@pO!|P(5;l9Uuw>2qV>xu7Gy5vLp@zLMYPS@8#fL-&H-{5DYLNN#u
zrD38{R8-l>J&3n1g9NgrnjySCrvghX)gwVUp0uq75_qH}2fK%bjJL8e(zmRbqM0x(
zS0TtEnJJ1XC8G2e?!0{J!~6>z0)A8X`Oo^pH^OtTD(9jrTu|>*2(70y9lBB`%w{Rm
z(dB1wJ$9f#vS=-Z7(L#?xMb0Mwa$%GXFTRy7mQv+U}gBNbgOBJZ;Ly>ERLJC*0soW
z-DOwtcO28Dxolo|t*6As;9=DJGuBUw%?wZu=nJwJ1N&zcMV|e=mP@T!P){8W2N1><
zqS;2JKKG2Rv&K2pj-F}(&s!+P<w$(viUld<FII%`7HMdFKaqO*+waw=mc0#`(gznT
z%HYh_8h&p)-r_7P&LqC9eBs6RG6#P~bhBArlTZ^lF>Z&KS!_dbn>4ws?q3coIvLy9
z(jge-xo=b%A(k&DIvYsQ-L!a!RG#O5Yy2b}FF<;O)WM_~;tHyUC-)hSp(P#)5W`|b
zKdBL_PWyz+FWfShXq1B#ECXtBtCDuZaP+nRGoN<fh*pDwFxMd`BvXRrg$TS6Z!^(`
zt%DP%<rcfS?94d#a{J$^xe}k-%hGzhMtZ6r&LbHc6#AEth+1UPB6l}lvz9aYu#g(+
zW0&^=3k#_1cIfJQ^L`xk!{|712_|>W@5z-ZdjjnYD)c{iHr>?sfhYCZ{Fgf^3bOj(
zhqb6z2iS7+Va5#wQ9D0v%+fq7w>^ihi&g7C8RmYtsaMP^6eB^O#1iE2`_h$Pye(WP
zV4kXlR+x9R^R+i$wa@Zh)}jZyC|^=Sw<Y4YMM0-Y1M$jjzU{HgJ*JQ@%a#Xa-ST(J
zb(Z&-`PfNr=i)~aF%UgP?gp6*d8x9x;r^kTG|T;_J;QO#?b96neBJ@erMw-9vcA|<
zy=tqt7l~RTJ(llg_H0m`IB5czCzYw1dh*k4PRsTL-->vNC7r(!8bBt!O$sF!l}Na%
z2r{;`eZbetlc;nEW12NT*A@A5<XXU(%#woTWzBRLFjVx-F2G<9@^|c)d|~QCy`E+H
z+IjN*-wP)k$EcXHqJv=q_qI*zLUrC;eV^v>O<str=|#qM{8Z7r(1)tVTPD15CU_t{
zx99N%$AEo?Z=23^+^{G2?x<PDJrQH2zTVKN>3T45L8)u?EURGHnx{D^%{Z@!S9?m&
zneu=?&;0fZ-j!6daBdW*kQkcEOZ3}x{GvBsTEjV_;NxIEUtD6$P>~_hl4Ikmi@>W<
zIR+aCO1BeH;YhU3PYP)3Npzi$5}qyg&E6ZEC-P$4SM2MNf%$vh&Bb2X#JN)Jm1EfO
zQE_zLd^RnIC}Dq2R2Ae4PoTJ{Cu;^we;Z35hBDa)I8{7f%wdSny`w3;t{0+sIoxBJ
z72B4rY>4Je%7>L=K26EwcaAeX<?E4~zv9RKur8IXC>kGIRD*G>ZuyIguest=|3#?}
zZ}|0_-3mS;e>eSaNmu2<Ip3WcafD3;Cg-;_<|X(Rbr+$>C6@R2-u|2@^`$G??qaOR
zWa>C_*)Lq}wLWabxW}8+H>emfiyv6JC;xl&vq4GW{3BpttGiwTjZqACe7zC3yDN;m
zlY8>&n_W5Kdvjy!7uycg;*5>Yqer}%hLp?NqY`ln-R7QNNQ_BOi9VOD%?^9p^rKHi
zcq}b$2}m<MqUe);p?##=CY7p1r)En{5#p{j(E`*~Q@(zjFVDnzt?-C21*lVPqWP)Q
zX})5Vk4ABN#)WY;okw9Hvemp^HU*kUPT`e6<f}ZG3dv4mM#AxD)4vi{gAIHGud<ng
z7cU4pDJZsS`&0#J)Z0B-#v!s6+rgy^r03%m3Rb>#>nHH}MAqJ3`4u6bPJv|Tp!n?o
zGt#}UHA{k#pSK<tmfX{6W<~D~!!oDr5TDs`AA<1aZPZ6u_ipo<naPRq^3jeTR<n6i
zvzdB-uo$Vx7DtG$X=NBvrQ}#Nzzk6!!+fI6`8>%*HcQS|hj?j4h8UOOBTp~w6c}F<
zQxf?EVe!kiuQ$gUEmCDBD@3I*hRR)y?BoGpOONveI+6{`lSeTaNcM9{p2wLGtE)l9
zobkEL4J@P=#4@l*7UPV_P=%2XVUmaI^wj*@Xcu#i3l{hH>$EY@Bo0DVgc@sz;<yz{
z8Q01`JnGi_w_Xp)^igHx$BbofX6UNl1~bJGhol}FJt*^8CL(_x&{~mDmws*Y%3)&Q
z<`F7XG{e#U%cz4N^79BShodVp479{#`3;rMuRrf&?I6jVj7jY{z-lPW<%G=Hb9l>K
zG(vpbAhFM$*M_2C!M?JUk>m!hm&OUlH)E#C>T!%6XK{J4FcNP6JlbHO_<(^%&?m4Q
zNsdSGN}O(J*;9bsjAEx5<}3Q_R_<k+UmP&&hoirBU4S=W205p?Gni)_&2wN^i2B;b
z1hT+nGd8b2W}C8z;Vr2H+UewUexKW|@@Rs!S_gcfjov(+W}&dmqQR7tw+K-o$L}{x
z)LDaa5@v85+vWW<<sL;+XZljb?ec3o)5fF^?t54a<7MjYIF8tkDO$3}4f?lwroDZW
zIjgQYzE{s*6+hv`L1#S*UkqxwdM+zG;*dQ9*8Jv$rSr6SIJkZ2tE<FT?DL-#;+RM<
zMGw!Ye(!-!kUz&*q>rZa!_8i#HNdsm)b(J<V($I>bp2-{PYw6c!K$XIoBX-(W5O?(
z*E8udLF23->{S1`rCWdLMk-vw<Z(ccr=Vgak(L#G0ameC>=#XM;;#je-dnw&nX(>N
zhWd$4-hAR}YC<8Wfl#>~p1=VN><^`lM+w^zUt|nqey|CU)FiQ1_<f}}cuI1VpYR*l
zp-ls?P-h=xH#7a3qh=SMgBDpL&EE2LE<MGacnP|2?Othax>_1H%*%LM%Xx#@sb^_+
zjDfX}h|~h$EroN$a%Xzy);4S3$?fkFJ_$TSGz8Mvw_1{kWkGgwKc&g>8(3WXt{XlX
z+77CYEy)iYOQ!~<+(&3-<oJr%n}w7|)Q={ZYG$n{2N5QMWt%zf|G+Yu{L01fNT7z}
z3N+HuC>CouxLHB0m-AoP+n1~Vc$KfKLj{*|c_J_zwlIqnU)X&Md8ShKM%9@@F>;v;
zp~@PniPSxpw^SqOcthd6cS=O|ZOj5@`o(lENCa(5VwUssYE#XOp&s>FM1oJwt__K@
z_=Slxfo!cwy0M+()7Rg~W*n`gubMU3e<Y>C4$>nZrzF+Pt!V1#G$jh)?1m-<f-XLK
zRY+1yf0{zsa?kJ1n-M^PB<fpAuhrgNd0p>Do6<&4tIb5C!&t+XO*7FK7{bYMa7u6W
zli!xBFxU<&FzJ^@b3T8qwxy1`F;nA~0RykIpmn&)Y#>`<pJh{vuE~f%n=X8O<fhdQ
z%&O^6?6=+P7V&o{pKZY<4jMV4E3EuJtvmLsp)$_LU4*LG;jGnyLe~LDM`E8g4h>DX
za^qU#>ie5Bo4SLJU5%;_;Z-nGSOjD@)lN#0@@V(Ph8et13gOI$i6<iG<*fx`y8)t$
z^NCdkz{rU@LzKi?k6wdr4nHkYM>4;Qq)(m_Mrq?3y{OtO8nvRkQ@L@rV$oQ8<hkCt
z3mc3bJjC<SY$F7tad*G7#Rv5>M^6(m=;s`<PVOxeR9iQzpRncSkCfGjK-PH|K=aOD
zNgj1iU8bqfmJYnUU@n_JTGO6C=~h4Rkm7B}!viTanlW3fGLzlqMjVOwNhvW{IC0$^
zDYx`puDK<-vjx=SycL35sVVMN;5o8a4+;Q&TGSY@6UIrSj$%+3LD3$9^d<S!>J`#l
zH{$4sEWhVOJ*rck==icgd(}-Xoi<Gx-PX!}r#Mq`yr_V;Oj>=E87)iY@%4ko8&~}`
z-oDZ2#VBavslMBfI%v%W-+5lf7IdI4wI|*cx8KgMHK{)|W-j+sGT5>%vcoca-1$_O
zWVGYH3L%)c2SL6pFXsz?T<iM&ILYxuQUsICsd}1ew=W0(&PlWF%6aCN3j1S3aAHVj
zrt)o7yH{%3WHWfKbo#lj@Bd(1lD|X$6bvL@M>Qs%70)j;PirA(>&Hqiq{m9ah7^fT
zhZd;fv#VdViZNkdyXIJU@#G}tNj?7*M+>|>n9m|Q50w+q{LX)lHh_kC6zv#|nNCd>
zt0Qt2l~6S=z32s2ZI?y^l?G<3<06;N1T_=Ouj(vEscHf*3hh)7SHrcEPgJ}Z_13!k
z)@5Ax`g=}&3U{^JD}LITjt~?*dt``}H~aI!(U;>7iy_<kTRw35<VTGj?xPHRl>$!<
zE#)z;=?>IAq7!^mBICE!j=Dhqr5ebxHY;=iXh@vK5`V98pz7`(wP!__kZk!(+6c0r
zG}mEZ3*pY~9Obgn1ctt~H~V=aFK89!@v=Am#<ZvQdvBgcO^?T3CGE~yYDDcaqF_)v
zsdnG?&AFJ@(&Z<o$Jh~uY7I$JS=GG~wf#gE+DW{uZ={#nDYUb^BAe^#kmskI9_Q?M
z6^7onlUk!~y_l!kQl0)7FE&etgXM^5wWn9jraUtruLhdQY6km<m6ebVGIaG`?mpLu
z4dTyo)T)taOsR~{wfWZRdRt!gt&<V2Lp+?Hd-Eih`<wqr)49&`N2#=N)}v9OCtQR6
z0)mBg6lmE>9n~DG-p_Tuvt5mzo7di<N02Ylp@`$KQvRsxx)_SGjZx4VP`ZgcGGb88
z9Li!;V2leA#_^;)2AIuK6~e^8;L1RMY!)@m{T9A+sb-ULMbRG6)TffD>g$V7xgElJ
z=!+PE`c>-~3x>mbSEj<y*5Tb6jUw_ye|nfzD9ey6E<!Q8c~EBwrOau_7MrLD{r-mP
z8rH1~528vcO5E~=!nBZY*9o9YVFH!;qso4>HE{deIOcl6olw4@0-?m5p)Dy5?E5qK
z=&6?|whmyvZo$G0PlQ4Y+Bojqn7`%vzt<b(IB=tRXfbjhlJL~K*T#8NDm&AUB{s^t
zY)BEgl@5YMiN9Te2`?)87hF=6wlwRBMQbR2zR{5T-SKc#S(k5-Zbv%cnd_oK-;C&l
zFY0@O)8TG2qwwt$208{~$qR7G!?OL7qL_TKe2c9a1(BFXI|$#^4>ILGM7}{pmG4_J
zc<w-Tic_&ZTwoh>i??&ErXu!Gl@yy8pL5tbT}90__xoI$a<xFXei_KpRA{3-sZbA=
z?&0Jo5c6&C-bM)!wRddC-<a&jcx{4qg<pGod*pGHiet~K_%NPGut_MqI=#t#Y%BOD
z$PNFda8gTKFZyY56>;4(c`xn`Tjs?nT(p>vqF2E?QI$>2&CpeAf5@<Hy)f+3@Q2ow
z{su8+oHJsIL77q2rZff{L0P;w*BThwF6O>A@I8NaMS_6uD@zDRIQZMfI^lR4PMbyO
z@^yDIMDg=%cIe#mv~N>Y>|+==d>4OU-w_IL^h-lmo>3b6x+XA)!b|;3_d^(AqBqam
zJ>P9Ux#-CQz!g?=+kRSQ(fJOq;~pscQ`^|{S@fDbbi^I})G!BcoNKunkqe-|k?er}
zYFct*vPhgXVR$bK?fX`aS#F;oBp1+%2-SS=)xcP7d~n}t={uMfF0O^Gp#Q!-?pQDh
z<H1aK+=>-x=wToJy_ime?f^blnr5oUS|gLn8`(F4njHPSrQzSYCdwh*5`DG%)ZK$w
zS5=KS1z1)J681`cfzZJK$%i;R#ztwfGAq!_sI}GBG`x$j>*f~h0F{!3_*DI$0^w2p
zVBUGK3-=SeJ*;KSJ~im44(N{fpz>0COFo#{-Fh*+yQbSRiIxUcReqkjpW2jg?F8%%
zg7M$lK8aKNVYGd!o2V;_w~K!jl5*jxtBtQq+|#g*o=1Sh`%1y<*I?a%kbKAECuo+g
zyDM02PJOi`7<A!F6T1zio4>>(ik`!GAR+sNsVgY+c5UMkc68NEJlKOs%^cxJBdU&%
zV)fk9t$N-fH;7`TBA5zzUr5d}D<^O4>D~&00eNyV??dA8oA#sjQ{U(OY8Jv>eV+Vj
zKtHDs9^@{B91rVzE;DBYWGNR)O}!Hj3wsXPl+;K-E}1zue-=QbykCjEN4AG=)eJ1^
zb<It&u4A^k2*y_NB=O#3u4@M+PTRb$43!nb@pPD3w9k+z@ea9t-R{++SSA07b>(au
z`cL>w5$v(b3taCkAfp04uebyKOLlP7rdi5|6C}E-AR$yc?|Mo+B)jllP2UU9{PetN
z<f8?NIhh_H>3i)-<B5$`aJKo%9**saA_)yc$h(T~ai}e`j@=y#0=hu4#`D}8xl2Q3
z3gx@N)1PqsgsH$qtmv~8rKGwp9~lx;Z%*Fn4SmG%UsR;csj6Rk>&ac&pUgSN;k;?n
zbxf;HD?vD^@x0)Mp*-L-l$x_HOaG*W#8;*;@H<Lmw`;$|IxwJ?y$sbHhN1!s%&xu3
z6FKnK&nH^TR<~NpJ1ui`;?t&Cw+{G?QnX{%DvW?*oib??`P!M=%pW=N;W_4=Exqsd
zV<P@{J?Y4k`}rX@&x?b;-$mJ+!3;q|w4-y!c&?4m{Thz9oiv}d#tz&Ug;oCY+5Zwl
z93L9Ur@x+$qP3p7=M1CC3ld3i;VcWXpYz(1-@@O<p4j2OYDe|Fu3@@<>-x=npteiB
z1V_|VplN=D#<H#&hxGEs(tv6A&E-_2f|m}7c4z74!;5BN*FbAQL4Y3UL!GA5(OKh$
zA>F11!3yn<I-QT|BYML?X#AFfyGwSIQs80f<~|)PpgO#NMaVgZ{5KW=sH?9IcB8KF
z4cM#?qJE_+R6D`lerYC9vb<w5FQG#=dRpO|7$MrVCu^9yia}-Dtr(z2{IkVVddJma
zJE|R(xo$z~E+@suear*$J>>{hIQzSk#A>3FPDPzA2#;c+UI`dTWpFM^#awk!6-HcN
zn@5Fb{d}>M$X@t#TgAdBaW+pzov~?|4>s$|s+D~ap3@$X0Vei3Y+eM@InHj8x;+bW
z%sqLM%_pPr*3N2aIqPL}AXt{5Pb&$aCnlcw^|g^wtdb7Spei1HvbmTerGRbf_xiJ6
zmW2JAIFR+E##MgwXi7N^W#%tdA*ZkMHYpLV>DbwIVzDq-EIaPAgUmap-;q|CZz!~X
zf10|l4-22;-P-j?AIQ{&5q(Y!ov*`Hj5gG1uDda*L}$s|v-}v%t34Y(m1P9OlC0x2
z^J7qfERR{Fr}-uE6Cu0OB<*ER1?a&v@>4=F(Ce)dQuR*+b~LA7mRD^aO)a(ck;b^Z
zCSTxr_Ike2M&Q(nC}*ClIP@ao2hn6AdHYp5OzA;cJ9lEJMX83TOoyOl#z=HUsJfZL
z4B5wWZ$(W&d$?bQG2e$A@_`{eQlYNf9$IK2u2_ZE6BjYaBY?tpWU_K3nuWwP>t8+k
zF<T$03ciaQn@zfFO*E*)^heL2CGir{hs1l-7d~}x!Nc@;QORN3N=-Xq%*XF+kOK?G
zq<L}{e-`dx8h4kdHQDPTXdcH~q`}`fYTL}^%rnsBE~O>qRo>n*FR?`P5Lz<GomgDi
zn^z?s>D(dXI1p#2(i3@|OudR>XE+dhA-jgAB%|tVNVH_i<Tap39RvL&n<%X%`}zsd
zawg2>K3|^7ZT;@bohR#6Y7*VV^5B&T7Mhq+Ja=1?hm>$AmX))M8uuh(WkzoN^eX|H
zKCGjtB=zjRfETp-nxA7Bl6VXs!!qz7(HSa(?#gaQ_TBX?%)-8i#!KzV&>PW9M~g>f
zRjxMLCud7SnV|tG7<=e@6`}sgy`TcPXFOZC*Z(*XKtP2*w-=<`sQ-kRP7^~@)&I&C
zd-4)e*(5{?<a=k6&Su(7=e@GTS~Xt}*P*Q@aYhVS`bI!{uZ)dmrVD!4LMO*z);Z?+
z@168KCRmHiM=Pm>EzU-r_<J)GL&NyAEtgO5>@vBS8^O9WzKwFaw;Id&*g}w!hK6ir
zK5HsJ<*L3<Qaw`2MCymxQcKI*q})P{XgiPm-oul3!`%^iiDascMWHBse3z@s-%I)x
zX<F*AJ&A-WFvlK28$4c|IrE~u&G3L2XVb{Khq=c=%&B;8dy;7f{T9L=g8(}0SW(o7
zUi`vD6Il<nOlE?=5LPp)PfjTz`-NsPy&$5uPse4)fvG<s>+8|XcU6=026Ns|Snb`R
zpc>>QjN+htloH6WNJqng$<9VVdB{H<c8gE^`9d9nzewv(z(Ve-k^P>#PArm`trmua
z@v;ZTFeySOOBO;Dq8Lz#iCxI&Dz`3&v{Ue8QxK$tUiXI-m@_!izR*!;iJJ|*_l0S*
zY9!p>!rXG-Dh40p54(n|#U!Z^=MmSes8W)C3L*$nptW$9X@Gxa%>Y{P>#zO@Qrez-
zo19ycc=6dz-rRpwAalXpO*_+(W=EqOE(z8T^O=dDJit!c0dJ5;3X>Y$J~2wF<U(SX
z^*!^GB5jj_C2|?R|IQL`I2A5B{LR)DXg}vp?9?l+Pz3WrD^!lo4R@%^i;7=K=dXV5
znzelJk}8wb&r4|*(ze@<djnH%zcN{|V4_`|-M$SfkQrcU+(aTp*`m7qc$~R0x+Xn!
z#&~K$asE^qfHc4Cd%cW};=kE+5<GEr6Q2RL9#9aLvOx@|)*5><)qdYXWCwqM(Hm;m
z?Ze}(n04~*{U>)qKx6rH%9WdPK5mq`z49(jm8V3r0?yb7YIrVa(bNB47FIYlqN`>~
z^_*~d#zO#0|Dxv=Z<){%)`!E+p$W-*u#e-q#P9b;c|*;|C3(wWOIU|t9oI9Pne>NW
zaHe`CMMy~%U#=~8udotDptBITKEw)P&ec9d;40rCU+2N%x=6bbS$ZH74%_UT`FtV3
z?fA%^H^Lh~8n%2BNVaEdR^>@$EnR1^%=0KJCLUwW!v)M^;VTnfzS(xV5=be(3qtIj
zB$Z<^AbtA=J*rgJYXzx;hJorbAR_j_a?~f+M?B^S*6|s)M5s!G(Jh)CD$&{+uYg$n
zh}FF5e7Hx$3R9MUJ3V^3JBiU<OoRQfNQWm>^s#F>Aia@P77@I9q0DH>PI&JqDsi2Y
zjP$c?U-nHi@1AQ9S}fERp#&ZAZ3ph?#WNQ*aZ;#C(ddp-q66JKsq3(GKf7Q2L(#Cl
z$RCeKUh(hF<h^5vl&_sJ?dYhM&u_&?QCnC|BkRy7pWrEC=FEJ#=4!iGx@k`F1J{c~
z%{$&o?WOwZC(ga^EJebS^QT`@#xkw+%iDcJey}3B$Q$vnCpV<iEs!!S(7hj=us1!D
zsk2cJPhLJb|D81l@K>asbf5jDD9w?08UlPG+edhMtE5Ml^QN;Kmi0nbdjgb?)Ty=k
zuo|a;!;Wg}^|5F~teT(`p|C5h&P4v5TWwn4lBhqwYE)97V%jSyRZU`WI1Dc=@Vt=*
z{oE^B>%ixMI=pF0_e1NKtjVfe4j4W*0X#9KlmQAFnEM*O1A&RZ2KqQn@>*+dab%sR
zzxC;dsB>vPG8jT;=acm{29sKl_HvUpFN37<B|ft{LC($3hn?CmWH5D5nLeXv3?*9V
z>ZAPpO};}0KjV1;4KlL=I|pp&`?nn8O-6J^-om{W20qbJ-W-JdRFWC?ZQmI+azAUQ
z9Us?N2}keG0gLPQso>R!g?!L<JOM7Rd3O_Q-iL4|GV=*AiAZ<=f%HP*BW!a6Cb4sG
zx7kEK=gX_X%uyFxk3<Jsuis;ijMP~~7Q2|DgTt<tBtoRm?)A5o6%1Id3+ChSxW?>F
zX0P*-5AY3Y&qj6r<MvMA)#|w~n`cHiwJy;=m!R@FQ6%bGrD6}eBWZ>$t-QZztcd2N
zQ6@-cuv&{pRNH(%6y)D#4NeTO=`j^=ZV#FH5+Ir9ckDPCJShLca!69%zZ*Cd9;(YR
zv<~0y%Df*5aB&0AGQ%8JraIt7jDy*8;8Wnhway-Y-}g1FC}RC}x$EaprDd^;h<PfO
zv<p76n83YfZptJxf0K7Je~WxRq|RTD?U0%Q5+!4>WRByr&xg$bqrv|p@)rV>$ZC$^
zi^viIhcdmUIccsxSM6`!v}E?fH>1iBt2aOfch34~J;D!-RhWx(EO0_Jh9tMPcwSzH
zycbGyQhd(i?Aq`oL}3i(8>*NeW0&%UjXJB}WOt1>KX4u8o<+^C2zllX#js1<%@RxJ
zky|QE4rA?vx-?X5$^Gj?)JOOmYXu>El^sj6$N>e2JM1#AL*}yO=pZBOixeGTdNTHL
z$$PzdiBmi<^>wyohE7#_CE2XEIig=xj<K&v^GTUxYC9&;k$;FDDrQQ<`sN9UgLYEW
zRR!~(j_dR+o!{d|1@b*kfZ1jDg^%Zl1^CyqG1nNXY1oo&^BkI%+39u^P;kEybT|}U
zdf1!IsF(BGi~^|g0<zpax7NM2ysPR{W<$TPmA8gr6B{NHkIhLXj6VpVcH|PEGa#zS
zc9uWruW%L3Kh+XYX~V+FwvGA0x#e7bziy@<aY>4lWmFE>iyg@=!5|{1@JrIH867$I
zs~jTKl+)HIu~SV*8cYgpC)Yze-b^+r^3?tLhU}ES17wG43T`}&A@lPKL5z^VWYXyG
z+`Aae0E@vxH0sl5h1B!<jW~tlu_Ze8V~UWbYXM9%HWS3i0%j``O4tqo{?Gxx#In~;
zdCb>lmCn!AY8SSHFj(E@*uF?yqoQQ&Y1h57?`<^fg#DOi8qe*-516?e<ffsm(+Z~@
zOZ47~bfwexaS|@Xe)%AIJfO&D%sWn~j?6`jJJ;kjDE5<lO+lAqtQ?s9s@h&`*A;j}
zg;h|CczT~M3p%0x!1dT{%D^^5OlrRW2P_2)nEXzx5~FT-B_c@3n+*;q7V34mWg7<1
zU#-0wD196}pzxae1(~YZh>-O|^Kqr?+*jlg#(|IQE`d9K6~`(7f_|V0Ra*BMFdn`~
zXqQ>wdybzcnqa9SLdZX9cOY3~E~v`ACY|5=Vi#K{4u+~WaucF;h;Se_NN7iT@}M)v
z?w3oHp~Ma<DPz_Sp%*hv{&{wxDJZT#M8&nO?><QA?)-d@G(6;083I!w7}GNV=3ZS9
zO>=J6KS9GHEmo}p7!QHj?_ul1w!>V`@d{{rD1LaF#!xoyH<5eXc_$Rf>r-nBIHYg$
z0bkl0>Mx+$fFpZto8<r2c3!qkvmPCsRjC}_9feD~q@t^LM$v5NQ6v_#V47-bd|>`X
zwjhi8y!ak|W9MMT8DF+pt|;D;J?H{=OG*lD^Po_nx&TDIGv|79D6T1v#wDp>l}TN&
z1q6AV!_w%%{c#1m)egQwwzkqz5MvH<ek(I~xkM+RH&3JbiOx4?@baZ)3&ko?DAr4P
zXHT`s+|KeYWQM#yeBYfRYD4NAt9%)&n6WOMc*eLhT8fx-#vf#E7ZciAE*@s5aD=mt
zy>|87RM?0%(iAZ*Wo*6pSBm1GS#7Ro$L;menrn+d)WfV6sY0bBze+??a!>o76%Tix
zWg|(^UXdmU76Aoy>gz9-sXnq(PD&&bELTz4b!bPfSqDSFl>=bsLz@BWpSM65A`BM#
zp<5n8Tl84-NpqNrxT-wg#j^#o^rkw5WhB|oj?gNN@Y?K;w#8kMx{6dSj9WI-%l(Lj
z<K{`w-4Z}EaifAjCjcoWg+Bj{qqwz1iSS38@RhxG28Keumx2D(b*=A=+5OR>N)cGl
z)RBRb6m(jUM6ghEl#!c*hKJL$=t}P8HwCx&xFZr2EO#=eDPFv(T|zWmy%ag&_OIGv
z#R=>^`y)IZ+=^hGowxKK=MHzQrhmoDN65YqSVdaJ#+oetJsrl_C|eaCFCzP;&plO-
z8}4ULiW>gfdHk++#(~G{35A0@5jQz>pd_D-@@z-_5y9SEx0zP6u85;1JGUFOa85kh
z#cRr;4QPJ9H2@PSDdT%-W7uJ#qO1BT1-}R|v6W9f6rNI9NUQNJX(G>+V3FW4Cl-_E
zo%h>CDzT#kx!r-AKPz<vX%5BcTasVx)yy7;%kBjO+ox54z|^s>PwY*aG%Ei4zQZ(L
z65!lbVUXwObs}Gv|J=;snP>r4Ld+cF(u-=4iL4jp?cK;4P`-c5eG`KPBY2QN2{Fv<
z?5Ff>W-{}06vAA7gS6M&2L6=l&)RxJPPB;f7W=aDn>~`Z0EM25xJUs#asD!O-v_lF
z@+5OCCQWSESB8WT9u7dvX=a-CQ!j01a`(sS3I;>xc@xOFCS{a)h<-P7sGjFMZ1BV3
zK~B=dVtR)qU<*;?A%k}U7!c)ExZ}$w;h&a&W<4`V!HiyqC0UX7d3Y87<Z5Pu`jZeh
zvlU6!<TvG}*u<$fa|oksCnHn2;AvDN^CeOCO$T(gW0p;qm<r?@V8TdP5+=D5c(^A`
zyo*Co*6(QxHMMF`pl%aCtp9-1hrAttO!QWWjEEZ<OkH}YDbJfg$Cto*QFFBT)*3On
z76%(8Ne55k+av9y>LU-sICa|KU|Khp1d=X;xUaWb&h(xH@FYf@(IBo{kGA77ekCc%
z&j4w~_fQaZE9@C}J-?>Ez%JW~f|jM<VH2}yEyjr74^IrH??>A&l+~P*CRvh!8@IWQ
zRS)mQ;PcMn1sTTXHrlUCv8jqQ3iWS@`kIAjn^C94>WjhnqUPCR7g5Xo+2zlvy_KQ3
zqRIj+5JiKH6OJ81qAmBiXP(x1#eO*aXh|j*0{1kXF{p;HnTGNztj=SFe==?O7rem}
zh|v{EwVI|ITZJa@KsC`i65pv18KO|(<CF^;{h}svLZ;$Baw6J^p5zMcjIo|(Vxq(x
zU$iAHSIJjfIuLwr!2^9<9J+0Q6gYnxvaWHX@Y#~Eb;i1!rVFi-oMF-r(@dJS1lOb3
zHy}v+lP4_>8AnNGlEYAHIfBW$3RS$QjsNugIPdmH+NAT_5%Ka+EE}xDWb?SM&9l$k
zXTntG-%jV1kDBMb?i5cz?@RprbKStYP}C-C{Z)w)Rg%y&+xhE@$F4?Lq8^Pn<fufm
zu3i<&HpK`DEi+E7yn4Bosp*lfLsDocROAQGUT92XT5`uNiJh<9luNA!-5ex**%uAL
za*Ys2+9Po7$scx|b!|Lfc1_$LJc`oNiykiW{QN-7=960L(0sNBWAr;(EALAVfk)()
zgp89dU#DG71;1OLC4~nYAidMGRvUjY{~R*Zm=VE;;Fvzk=qYOo6BnzrWsM2`hGrnG
zHM`y9Y01Ema4tA>mT67bFcy3KjI~#(cy~3VAo;Z6t+?_Wu&1Lil^(mbGDX#=Dby=f
z1M*rO<msICICK5HK_llqbb;J8&(}l?s8sQ$Rl&_}65kc&r!HqvFO#VT@8ihs@Ohb0
z-^+%!l@LxNj;)W8M9{P~x%gms*W2`oJZo6f9jroII;aE<F}DPY5d@YD1v!r1bvD-=
zNzC8PC*tenNcPNEpp(EuC=JJAF#7vw4dGVevri&sMiI45@>&^Q4yuU^*`#wZs;VTY
z4pnVV3|W7@Q%>TLSnJ<A4kRbajuci+$C17ZbPeWDz~Zqv;wiK#q#jLsGsBd8csa!s
z`}XSle0t;HiPhqh#jh4>L$oFl87%c<H^<r=MAHoNkfNNqfrp9mK73Ytu=UY39n%l$
zb^Wi%eeZU%JbRM6s_mBRBxbfIt-blU<;Bm4W;jxGI`dhtJn*Q=?cb1CHYc6Y<jz?5
zS77$p@Z?A*^^smyfN_6&{{8Qv?1#P>sEEc$&^7_FJv@6c=+N!`$YVjHbp0-BRS{3R
z%P$dVyk>&<X)ZTT5~1wI&M@=VD^sS~qXV&dvs5%=uXq}0T9x~pbyOw?Lo}4OlUxP+
zI9Ty`-MF}X@6fkIgvi82-#xuVvxB9pktyW(N6+zV(}kns*T+W*?pbL@8H2|;sWzV!
z<V_SVB=yWmS&j^pyn~@z{KwyupX?3q3K;uDEb>Gr4<J>!dx<!b+t07)CJ+bm4l?Q$
zSqd*X_INpId%qmc&ds1LD6QMBtYf>v{+zO)AU6G*GG%vN*UT%>j;KTK+XOkEiD*zI
zv+*vIfqf+6j&w(@aV{Wjzr`8rSQhFS_!bVjNV{BK_YOME+`u;25{A*!jE$lS)p>?L
z56-EgL4(J84T6L{3d5a{mNn;sl#fWJiG@}=$49Q5M|W+`T(i#Txyr2!?%7ZC^?ev7
zU3LR-G>hWEJ)EN`&=gTB5gA^26$kGdePJ*<nQcF;FzJ1F^EQwSiAu4}iI9c5&rNQJ
zSnsKDM5W&>b0@yuP_;~3mt<f8xve22MZqtz?X;U&m=!c<`a7#$2jAUD^SA)e@_B&z
z?BFvMW^oBSgS>m5OX7n*p=4_9oui&Dy*fv3vae<=sIYfcF>uG|l?N><`VE8wNgvB0
zqlhvx`702mJQYrMpd@&XP0X^w60q4MW=a&TSVupQRca#1B$3!>!iV@?D7F_FGD|M>
zMOrfWLo*?h%*GgBiN+W`6^q{GAM$`4B|;m4!zqD$IVQG%n#6P3R|MoPl9tM1Qj#)e
zc|LFb9#HPXWIKD^1BC4ud<O6z+itx^Z8#!AFkOayL_k6!8>0^yr{I^L99%&}n!}Dr
zbW9CG$a2v~rRgtyk<Qram^+wjmY85R;wu~>T$8)|^_D;Um19e@%iJ|{3n{8ps2Vo*
zZpW$IY^Ru7$vrz`1J<B&UBa^i1^1vD5<?1(5bnN<Oj`NtC(KzBEmDV}S|avR%qcdA
zj;<a|A)A``?=e|!((i8Ak3>hd<L7pnXjWup-YJ&0#dz<^?r{&I#i~)7Lk>hO(qhqE
zstj6>a@3;f;+%%z9Z~+#uFv1P6<AjhtiIDd#I`a>Xsc_#>gO||3o@I2YBG+~%RgXx
zi0m38@W@m(ugN}x^Qj&uru3pKKKHymvK**wq@0DC_o{U?r+G-tB>9rJ8ORbnN{*(B
ze1}Fr|Jdx3z5Heqt7Hg`o{HfooAl2euT(To)V6~V+U`trTUQ7<u?b>yw*!-E(uYKg
zOMG|xLM^KUgm#a{2V#}qt(fL=CGZ#de3kBkz1Rl*&%#~eNH2}z6Z_{6tmpAAEi2zt
zd?cPX-LgjvM+zbHm36Q?ATuTKH22eLJ3E_ioNWACC{5<=;9zm}iqLuO$uKJ^1CjR1
z4^X&tfTHhEO&g5yLHT=Uj@wzW$nAeoVn>u2yn7SGb|^!#^O<)dcZ-GwbI|yWIC$&=
zyZ#w+O=59a38ms9on1)W49s(=J^D9aFLyJC^~DO8)uczdl}y^l1`e!jRLt_tjkLB?
zMbU<o&?;*t{N<CG8*1|xKR(#Tc+m%;nqToPDqf|W-m@ni1A#Katazg&(Y83Tm>1$*
z*gAXf=H)b9J^CD`L<p7~F3-!R^G-=4mP15)X^tLx?{dr(MMXT6kLu`Xo1@0nNMAH+
zeamyOOlQQ3eu7EC!;Qwxyg`y13%se_o*NVNw7|tG6aWp<LoHO%j}Rxf*FFAZQN-um
zB}#JSBx{^>Tw)-7YeHL`A}Jz91Ib$wQ`r;I1ml`F4Q7$%$pgOJO>wl9)&tl^-n$Gf
zo!2|u?A76$mJhsa9#C^!i5gA!{B^>vI*})_3S;ZJW$UfA?#(M5OrN>{npc0}@a8R;
z#ZujsB->HP$Wd&Vibf!|oHKS;9**zDr<@#h129=e9%yft?R1I){8KpI8(h!Ssn&!j
zwKQkeMG~YfN~X=7wwh-i6~12}?_DSQ!oGuX$b96q##_61YTA1ng!M+=o3OZfn4Zd(
z1cqWS?>!F9{o1LQ8#S=1C(x`VB4IuGY-`uevW25{EJumBx)yRuWNFag1ljiK7@ajS
z7rz~kExbv!E|&1-nBX?T-%CohiIc=xpZVqJ_QZ6?wgRL6Hz)=S=F`0s91tj(E17cL
zZZaKr>D#S8(`!Uso3SGN%9ARPzL;&xn0cOTXp_+#Vdme<pQ<#>$5EH-1cxU{Lh!pj
zrIk?kduN}#QrE18xut13*?YddzO<=)m%Z!cu}<x^_y3W0)?rn3-J(Yt=@3CWBy`g$
z-Q9|Sbc1wvhitk#B?Y9U8>B;0X^?J^5Tx(i@B6*qIrlv0-gBSl{)La?hPBpQbB;OY
z7{8wj>xs+~Q)2@rd-L#3GliqRLHNE1{*;%XjG!4pm5XhPQ}G{-qm)yNI-8(K%#_^a
zksM>P56$Z((o@Fs9u);2Iyh$^8@usmtmKft{$3B2U#fGmHlGWBVQ1c%tJOl;=GHJq
z0|J7DtfC<IguZi+%^d#z9ye<8584N20k^*j>LLteGLq__*>IkU3U%<bNHyHZc0ehY
zkgy$C4HG&jxc&DJIq<W%Holl$?j@Su^DS+m5p8-%AMMhQLA@{tkilNiA-^@;gF0EG
z5mcltDgER<jr;6=?%i9busHeSX71bXv;K`UpK|`8+8ajsosL-FYYc9YoxAcsr0Y(0
z#Ur(kINY=XZ;Cz<ueB)OSs+Lf+XjUpj3Ld`OyeE~E3IHJ_j5jE7*xAvoHv}`Tl%O^
zyKTb1zPXcGm-8Gd$hY=Uy7tIsoO!#zsQvAE!8)R3q%r?Xy#}piZux;m_g}&p47bO<
zD7${@9)SYnD~G9xN()>|e|P-}#tbOIBBZYw{?HwEom@A3<SBKw6Vr0i%ba?+p#{v^
zF2&pARY%r(?6cjExAl?)X;5<Q=F>F+d|^{M&Y_X7iudfxTZQ1@3kS2KmyAx~VAfPi
zpkYZF-?HiEyIKEKJ5@G!*OEPIg^5j9LUA53OW&W2tT7}9LF<TXZ}Ngw<UuU_g)5iP
z(lR4_e2-fNoY<abhlE@4dv=E({#PHq@H<rV5@@oA_MS1Xm(KR6r!n!JT^7gvoSqZ!
z7$-b7+u|xH-0yMKseb!peMfIe!0XR*#o8U)^xZF9=(S{qO`xrwr&c#3D`7>y7i?&!
zBm}ahEmD+6L8<Zv8oq^zSrm7^&dbM%K^w%&0n{43Agva@#y&38#~fD7gh8=0-n~MM
zz2S{7`Q-ka68<cY%ckyyM^8qwqI9d$+?cKI6OD-?2JMT>kFEPf|9c=~ej9liR3E|Q
zJ35l_p(VWmssBjm9<=k1DNJuOM-3ua!eW{Ii9U`AkK6fQgS<u1ZyY)s^(6B)mmEsg
zMwcUqJxnZ$9`DgzGM}y&wP6Jvmd+BA@+f{}(%EZVS~)uVS!Cg@#Jd_CS?{A-)hi}+
zFYn8h=RHAHcQ4;y>bELc_vIMX04nyM@OIbPorCl_2V_)V{AXq$JE-aMk&#g9zq#oE
zi(bV#tRrL&P>|}{3DD!LPfWN(>4+nEaSc8ro+ap<p=OZb-qLVb7Q}-7$=xr3`PE!w
z&|OeUGd>useqRy9L`o~<;JDDrU~)b)`Yw}&e5{}A?wdoHgXX~mgcbK;SJ{t`1KqnF
zmDP{*V>j=o%!jCB1?%*gkJ0ov5f$Ix>l3f$YB3Cru)DmDdybBeGZy=edX?&=8HlDh
z&XnRDKAS~cYf_r&J-y{?Bfv_J_%YJ@(&c$o_;21rGP>amB&UVUlW(b;J!T)K5052&
zu@Y#$1qIs?>uFk<V<XQl{QF2Wce+p1Q6zd$$x{vN2yv87qN!OkpHV%|!qpV8L}wg*
zVPN*d-`m0T%MFc)p#!=HR-Rm0XEkJKBx0@9eQfFm3#4ztohg45w2HB&^DGBaB21uo
zq_9{1wNg-TE*x<qzYQzavbVWRY+fG%Fm-owi4tqRJv{X_`LyeN>x_@qb1M<26&a_c
zi1*3Gu4*y$ttu$OHNPfXV@<Y@7Z(CR-)&`%;)4BdVam6WC*>(m9Q^JlVJ9WoTGFE<
zH{ymZp&_}D+q{bX5c^nevRxfOzIaz(rVRQ(^-aV*B#7ts$H%(=0_&e`!uSXT9Ai;z
zmh(iCdDb8$t9ZCC3}axeaA<`8n}KyN?cc?N+)PZ$;Z`vR8nXF+{@j)c>4&2wi-b8f
z_(K4}mGqPB-^@eMy+8d#v&c|HS`<+VK7hfqv%60Q{qOTQdL8N0vz_3eXkeGxiH1v%
z)l8Qpdh&oXN*Nfwa%H`lhm2l#qUp4nVN=zGTzBm7F#QV)2d!uD5Fmt?KVwd6^bUEH
z!685z$2%y1aeGOE{?4ylIpgL0h|XA(LB)V0OLFJPl7-ok_>!E!<+@I}EeEt2fWJ-i
zrJ@{HDKFO>ULgFC`Z@Zi2Ugy@aOG|!cqNO^z?LS@clLIox*!a|jp;vyAwR+tLv~u)
zQFxUjyZh0Zl34;nMra<y_IJK)Sp{kP)<FJa8)X*#YRleAucsv~<g2?aPmVpkAZQM9
z6({dYzrHftwE-RYgW-gMcr)_MO+Av5sLLbI4YP3Kw|!vc1IB%kgCm~{v@gH)$8m11
zu#`J?VVAu-x%>HLhm6=dSu^%)Al(TU&sUd-T%sGpjwbrg>tgL<<r(2fKmErE9F72f
zB8ROjYwRH`gg5jO#Zc_~qcVQYkPR(EOyMn_@<3tB<KU*kZ#dBOi2e~BmzytCE0=r{
zd0(28T!m2YkRNMa6qj=~%pTn#R+~PPc^L^SYEzy9<e&9(^M=%S7lOrYBxiObQ|@RH
zH*4nY7^M+1gDmQa8dbn|zr^lTm<H3ar!^7(z9f$W+bf&x&8#N3^Ly_&F#a7s(uU6O
z3b;u`CqAIgy~d^%_yEfBJ(}#8-&$?=VD34To1G}>gDNMNP2v8Y`flJB+P+HM&S+rO
z>OK3Z2Tr1|)I->ROM}CSxtb1*O<kT%^(SLN_41I{e<Dql-o<naB=4&)e%X1Jh+B)`
zp{%w)2_q~-pjoc#0&2Nrv8Q&fHGcy7>waq2zEeYt)`&<RTY&gxrl#@LL}4oZYB5#}
znPYx``K6=ewn#A2KC3y=RdguRbp+N*yGe%*4MSO0diR^k=o#N#q>bL12YA`8+boJV
zr+$>*DYp*Sj>%YAQvf5Lhs(08(K}w`{Ydv0>v~tN*!@X9<kQFYMT`B{ruS(us6@88
zMG95uw+|o)d>C6!>)hx5{_r4wOY4-#u$U+lYSJjlAjcdM)@b6YnD&8vBb{AO(pgcm
zD1P^uSg1-UYKRJhak{*0Le)dg<>iX!{Fv{>S>~C5SB_J*(>J#^z3qd$1H;2Qq5YPk
z<LmO{1|suY@*LC=iTdY}xbOiYlsKq}e|;f3LzS7ZyE$wCBYMrll&FBrl+HA3urSVY
z(hO*OhEn?jFysu)z$KZ}RHpl<(Gk;`W>~6r7Ad3|)u@a&QqcePVUhHR7WkGZg|pTs
z+Jqbm5voO!tW!owgQWqoY%{QL>a-9SA<We2D|F|?rIpe?&6w2gZ!C)DO~?Db8B_g+
zT-QvH92@*f&v0;5J8OBP<NT$pfs5}0RKidmQZ2EbzL=dxxTO#g5@nTXK{c|${)J5=
zIYJSis#hZ3$}`#u*2OxOlVOW75d(~J3liyUHp{GW&yl(GI{<QCVS~y~uRuIOn}y<w
z<;3VXS{Nv+?3n|&W<zSix&l&+IuVPSU}NgzG)Q+UH)%TR8@l*(sEXygkp>t<y4L-m
z5nB5LPr34GZLBR&pL~H5Y;H2XG^+<rpo~QtufzgnS~+~J@!DC-npk6MInQ*-cc<nu
zm5YwgJd&j!r9A3uxHVf=ZkOa^!$rMY2JSSqbnL!%1K0pcA5gp}YeIA(m0)JyfZpTJ
zO{{%3Ml#!ZyNB_Ud0(U`L~AStQjtn%PljMmsZaeVmHRqXr?JlaUBn3;o<DSatA^H!
ziQBLn5&zRyz%d~H%fRF)rdKOtTXQ-^@FDDjE2uQ({LaFfJS?tl9K|KTw2L(PJTU(x
z%9}_QV%jtyA;pS|gy9j-Z<itD(@Y;zll0vf-=ot6I^~|s>azC+|I_roKZGSM48u2z
zUF2encO3&DeC7ml2C!3A8zkPikdbVPk{M@exxwl0F}Q$Ux)o`uKisEGVr63sCX$o2
zsCLx#&(<N);!h+Xzb-_Q$;irpVq!%?kMx(*&StE3bnW*I$pW@96dNoX4tAv$aT3au
z!zkYUl5H3?0Lr=NK$ixqepKv9>vQ83{RZ54zly9}?n-DV3XqnL0OHF^L&qUx%t0dj
zygk}i#Wu1{vEVbotAS{c=Skksu%UlPU2ZRGVrC~jp+5W~wV{or^;h69yt^v4WblRW
z3!hLv%F}}FEFqnOH^(;;l$Wjv+mo6B7a6VzHOarZDyJB;n}>q_cn(U-N$d8P-bsIF
zll)%q$1vZbw@~)NG`^3yc7m~hn|>u`G9_%*-0QGT!?XM%AM#@!bWCjgUJF1WXN{}A
zB>|IZWek<fKG49LMRE;^si#pYt)-Y+QY6Is_07~8mJe6i)`4%z;C}cs8(i()j%|z~
z*P)n^gl)R+d-Qs2cj`=()+mAB838@+(mS%hG0Zig`V=7hofZr~kSUNLDQrkm&{Qe)
z@RpJ>U;RPxiq=z4M!s{H=jv;RVc3S#3Dya<qHUn{Lw_)|5??w^A>&a{Ske@)D`v>*
zi~b}xw(awp-mG!(BirmEH3!B?b{lWjzWqq5w2m;HJ>v|Z5b7QdcFpo2TF)Wadw=uV
z+ktci716=oJH>B9=-+dS2C4|0;m1V!z!A{>U6Cn{TS>OXG*n&vIMy}$ncOr%YA(B|
z1T?ancYqZwjTad{pwnGKgkR<mX8hVz84?~6hYwenmmzdfU6VjXlzItooh`yR+I?h!
z?GkCV7nc7x+W3<|=}(+v@LxsjP^)GKJ{rx_&M*`9G2=*5u5BV9z91Z;q;q~8`*A~W
z3Jsc-UA9k-?t=5fgl!0l8AP?lYGxb|CCN+ZrUeehpTr9kWPpX}s+)8p)AbR2BBUy{
zeDBIkoKRk?CqY|}UQM(yR+7Na2Y3fAUpcolHn(A&Y*w@!z4ydqCq>^yJeAfO7E_+$
zz>%WFErKJKAyn{YOT7K^2YoQSSBj8+!jCmu(Rk7I%0Lv4R0f}xA2kqSR?``N<X<aV
zV^W>b`H>#6mJh))5s3#TG{z9aKwp5Ii~9!?NMvhjO6Q0Q4F^Ilctyh<gE$ceF&qij
zJx>Fa!cLj+io*7RijPeT>AitEp<i>TEs~KMv8+u&RWO7W8=n!%Jd6X6Y8_2&hbZGZ
zjSXKg4WrwPZK_Sk%G`@TXv(DgrfS(O%J8Pj>UG<a!HK7OCHfZDoR?)FKtQ54u^Y#|
zk<lFP{mW*L=d=+BxIEG(ka0GW7iEcX$Ud+Sq<>2I(OD7M`R>AID{I<xIq%!hP&`#x
z0r6aDYZXA2c=4lF{LI8w-+y(u2+EcdV`|Pb6@QS}HX81>@iD(0m6k)Ok-aNoF}7AD
z30QtAM&}am8icuN+9SGFGO1L+IqeWU`y9PgDL7a_&fr_}DZB@wnqhXX!ZOvKD4%TJ
zEO}DYlW`<*m7x#KL-{G*8&j(N9^dbSF3X}BOKb1Mzc2ru)Kh1~wh!*aPDVfqYma;#
z`eGoycu=ynq0HJ_<`Fy*2`bSxspn3f&ESg50HRF?WcbRAye~0wh3bTTD<rcX96wx}
znNF%5e5Je_<jYR*dp*w&1P-E{%*dzm`Y`t2tGGA^QX4uS%ymZ8{qtkX$WAj{)F^@1
z8vX||+B_<`6u}TJ4D~+q$mdk60pgEGIEcOmyb8ETjKahD^O_+#WuF0CCgCTB<td8r
zpMcBZX`jC1$b1=WM^Sx+g&Yd>>A8zyfj);@YMh>~6YYz*vwv1@gGH*&p6EeOJ!p3~
z+EB`bt!42-w?shS(I$Z=oDm*8p#mb!Un@N8GOh?)Pq(O6Zk9w>w#JV{!KX9etpNy+
z5@Vg$(xhJ#7XHZn6%KqxWt}nYiZL$kBr3A7VHL!l5UZ%39Gj#urht4)iYt@TFLB2m
zrqdJ&g6ruv=LzWLCKS+8F;>M9BksWsQylR_dHnMcBbwUtV|EQ>{%=iE9il%y<HBC4
zEP~|xy3!b%#u)w@O#K0eTcgQ5fN1<~%>?$I{LJ8fyyXI)b=tM|;-o4JDim_r73da=
zHb4cj#UKBSk>ra6k7+9B8Ee{}fEpZhMfW>MAOXAYYT(f^Q|-7|p+rY83;HdTD6OK7
zUky1+<`Yf8^Hb&Q*CN=-VNq9J9<OM!ESV)dp31fuKowLvCY>xD_;G6|FGJVDNX1Mx
z7oeJ5Zk9PRR_)Qw?=Sb{NFwebJL2`7rB=9=0qx>slJ$ewv#+HRQBni_s%^K|%hbls
zcsJJ}*$L4mNoDSp6)Wnes>#TEHsm}TPfj^W5ED#CYc4%)=!e!y9iy6|k*}o&czxhZ
zMYKdHAs5C4B*fG!j06ib*^kVl8F5%@-eR~UNjF?E6y<|Yy1^dG8%q5sx1jIlrBT<m
zSEqoyu<ZOS=KT8b*6Sx4vT04u9!}3;09ZQwAz?1UeHqi*r=vtsCr^xD&O`RzYf&U$
zhX+g!+<C9)T?ni^!d@#l&hw$dj6F<y8Y={EY2^L2x<X{3kH|4McD{YhNxk1Z(am@2
ziO0)i!FAGd_97L!9G65C2bH&pNV`TlaQBI5SH%xQV@PVAd$u>PTOCMq{+jNSy*Bv%
zneP7;N&dHAcH|K?OL}Y|$%j?lY|}rwuknIq-hj~-`53XP8I2f|XI)bua@-HqK<jZt
zi)O<+z^A)m35feBVvvdxcJV&dKYHx%Q9Oi$y7rUdWtEKWzeCsm_6Pp`m#oRSsUVbT
zN$eqe!Tjd$7hyaG)gTHro`d;|gC8;N_C4cs2%Dkh5%6|7H{>*a$!0SS)xOxvc_SqU
zAFwJ;IbXK=m9L!u_IV(7r3?$o|J*~}_Mal>|KpGEBoc*dZ~$$(mX5{O0FAi-I-gAU
z%fCdoZ1b%K<3D3|?U$*iApsx7`WUwnM369>j!mOi?~|9w$WH;z2Uwyutxq=wfoVG<
zsyj34o_77h3CKLG8nc6W!?dqJ)xacZ=p_ct9$yy=&v(WwHz;w4NZ}{+FtLi=U;O=C
zM3@J;?1$jP{+TfTr`r}v5&&vIM`1--UVs-$lA|<C2qQ<~BgeejBFX?RE0W7~+P^+K
zPUL473Pml<<szy?gL!rF9`NH(H5zGDS+BSB?@ym*@VUIlNAm$aKMl{5HM`&H`?zQW
zLAMQnagygv_nIRFtV+XfaKAu1(cd>%5_W^@HmH~W-)^wTPq&VntTw<~EPRy)l(N=<
z8;A^L75nQVX`pa~YQTawhpbhVn#=<{h}>fWX9O($$A%(s@bsqQ#D#FH-ar{#2e_*o
zy5Yu@thTVhwn*S&!ui()4uLUV#U6|+`aj*Y|Gbm>@5TeGCv4d6FqX+!{{McI3UL4>
z*4t24>i_z4u2Ep5-x=L4uv>d^mrx=z7)JdI#uo%Qu~%LT3(ID9KKt6zoJ=Pt|F2^K
zKHy#iJdit8thn%h{*QqU^)HZwT&T62Isj6)jBUTQ9%wxP0Dlkd1rl9z7@?4+9sFr#
z`MijS%eP;a172L0d{p1+9%=3{{AGoO4ov=Q4e|taZmsKNQC7ykHvq!69*pK1MlYDc
z5NM}9?amwcm4k{Gz!9vR1N<_<S&jD|N8lkw0ivAc{UL#2l6SDeG(A2HiDZ3!zFQBh
za<j0kN4WVwjLUFOEIseP1JVC_tN!cb>f{vzm$_b2MN{_gao0fwr)qQDcIi4f4Pq7(
zn5>c7P<OW#G&3Iz(8SaV-|Z@2MCjjk`+{EQc|3WDs>Zo7b9NqcjFLYr%fO5d>kJzO
zswQr*%U@Xi0J`M6-Y*0BmmPptSs`S@<hp2j8yzXE3)Wxe$|~F6!-Tjjm6x|zgoh<#
zFx;|Wbm<jpT8T0p>}}-#^E2qMBtN<a{5rMU>_->9rR)v>5bSXyIPPXi4?|93pU;+b
z_yJ7mEbxw*q5~6n$(Vd4aV|>-@PQ$FXSv#g{aQ5u-u^}q;!meQO(PmDd};~XZvXRi
z|J%`Nuz?Fk??8%Z!N32Rr~%bP%S;&TF*7ZK3%fA6XFFliocUkwph@)*Tx0``%UTZ{
z%~uV{@7{l!U!$<X#dPs-?%oGOviT5@Z(RXlYz~H%i-Ec3YGJalDJ$c(WuODM@dNkt
znFnxRx>y5(U=2XR<vD!LNC5QOMLo~m=QluBwK8gzlye8JWzz+MOIWr=WAafbjThYN
zEM8QP*n0a$u_tM{g56?kpxJUt$}w&qSbkaRLG$_CzTRbX-)~rA@(;=IDr^Gz49t?Y
zukcQPx?_R_M=?zLrr2Yl<M6QZaHo%-7w@~V7QteBPH&7`84In}ER8Gvzuqs{T?%~)
zUQ;ZV<&(Pq_*RP`IFsTSp9AAMJd~+o004bP|6EKBW6eOU@XUc>IR9W6N_zo-nI1Z^
zNW-i$Gxo(WB;aoa(qM+0oQK=~A3$qHuo*@QyVy0q-nUPN_3!pHF8(5c`oVy95Bfb{
zy?@vURlP$~Weovps7a+)n*=R@#GC;cR7_=sUGO8ftw!4p5S812nm_Vkt;1BF>oAy`
zbnrPJ2{w`N0m$D*xmD{#JeP|$<_cDb{2m|Wjr2c^5K2~WSUB*fDw^d#eJcuLXnYA_
zPZw5aLlalxrs~BYCG-wE$#ziv!Rt%Mx6;2ZV^izN@Ix<Wjk)^S2_Q0m7Xo9dIcc7!
zbrq8)BhB#s8kqLvzdqh~r^ia}(o?S(S|i1r5u=k5XOw(~@P8k-lsG$fh=IJbE1USA
z(El0aFKGN<w!kJzm>vG>bC?{=0Ed}S%LG>&1RaN9WOa-JjyluI6&R@R0g3z5Dl=X~
zEZ}GNAA;h_IzZI!hv~gK+5wVu4#pzZwk%9~x6Go~_;7!-BsUGn&c=<E-avO~1E!?%
z&((d;j^ykapzl4_D0|Q$HLLgW9UA&R-v^0+A2btid_5J9Sk62sSFi=oRn{kOmM?|0
z9eSVK@g!pa*6KSNJ8%M32{ZA%)phya_vI>~1K0wi5!?Zxu)|^4-p*rKn-rZ@&maKl
zG^QcbST@@EVS`@SfoB9ZB4CN}df3ueKf*pdOP6771QR!FFTd0y8a!nv3Rzi-?TP&g
zYyT$4Of^(}EN`Ou#tW?1WavXq0A2e9yclH%PEl5EiTx}07d$N+@$%0ID@%l@gJ6Cq
zTsyudP<VO(K&y(vHelXyPP5Ztc(t31C{ES^ONkkEN9}VjSWmspp8D!yl(weOZ+jNE
zEZ-j&kJkPG>>k{E=P)q2UH<Y#eZ!oz@U?0NpwG%@%-OMvl2;9VDD3ji9sPK?M*xD(
zvz5{p_)9n_Wn1fj7khRee_revQP=__(jUc>SnrZl{|u{ZackSxy(+)Ha%(>$v)jI^
z1`x#s;4NJ_$FOyat0UNv*8s-K^Qo9~rpC42-+us`eeWn^cnHUu2lQrSfP%w}WxLBD
z-%MW{Sk!{;vm?&kPwG_bReM~C{kEcX2#~VcFWZi}eu1aYM0yNFk8`HfQBg!=5TI6E
zX89@hih@CYh!g1=s}GKGoZbfzfiJ4jW_I*l7ajVY7h<UJ743Us{+SaDMyTvTw*RU%
zInvOfnU|2ri8vY~iVyLd(2cdRbtd-C#dwH1CdhwEPq|;kiepuSNaf4>+NjHk6<zx-
zq>8DHwQu~#*7$vmzptgXKKw%T_%lD#{-cx2XD^H<x@s+xVTnY(e}Y<LC#G02!)&Jt
zb*G`~Ck54C(M=aM>$`dWQ;uol@Gc|5F(mkWBrs=1C&!07!}2MA&GmwmaH}C9HUp-y
zXq3z_v73^ZOu+}(+zCbP-KjSDD!3ei6M*;YV+6CK<0HgRu+R|=8)x%k(Mfp$l*>e@
za{7v2@B$(_=F=cp{2>?(B)K{FUO-3XR){ndp69CW4QgWMz|?-wBRY@(W3+E0tbGTQ
z*II=xw!kkDWn-F?2f<f>lWGlkOZO4p3Tf=UfbB~`-wIo2l~^C?%NX@|{aO9MKgUji
zo$7n!l;ecA4OSN8STcqb*gCEB*ihB~CYE<T1ccz<habT+Y{JR@0yOppTpka1;2*RS
zK=^jpEt4Xx)!a-4B?GuvJ;GQKEZ(1RH-PMHr7Um~j4z@p6G*>V1wi2o=etu2c#K|5
z*r*=?D)uwJ<RH4q7(o7)7q&Ki5D@)Fqga3`whNxs*U$BaRkW!DkB)SzyF*EFO0;i9
z34H<bGl%@8;Exbft&OFve4>7t0YnK)omH{)qPt)NMIT>~4bBWkX?hP5Z;tR9*@^RZ
z?2nRbl^JBza+i!$5z`I`_N;FrRzCgxK5QU}U4g*o*oU+tp6XWg2R~Q@bqOiozs`9$
ziScvylheLU8}shF84+-Z@kD!d3~se^prDV15Gh#akar_RNUFg)J$KY{obA_hV;?q&
zb40~FIrd8f0|$<Zlco{@X}%v!Q}a&?ENLi7eNQ6vNq6en+k`cnN~!Tp?txgR8n~rr
z7c>apfx#$1fv^bN!0ncB2bSWXQCYarx*SN`m!&sl3k6!h__*@9W|av%S7#IvlDN<u
z&m$|=E%qBm>2smN_mEf<_-kl_!MCG~INLZ#hs(|2ieIFZwP!p4VBd@5DYSwTJPtiy
zMp0Q^n?Y2Q&XRna4wBH`2+Ja8JMkL*-wW4cgqMGlGEVK#!ej<BL4{Ur<it+m)<z0<
z)324UkjeUPT#(LbN4};~e7Q=D>ZAD6iLw{&yVxgACNA7h8rs>%vy}z9Y%=%r`uSM%
zB`Z?RLD9;x(ygh}Eey@x9#wZLs`1r3YxI&acSiI$8@n8rvW(Gy((o%LkAm<z5$ZV?
zd|Na}nc4f=TDJ|Sl70g9>c~rk0AJYl@$M~kT=qZjf36fF<ug*B^9d#PPNPYHy*6hM
z)A|((LFtMNafa*|P|;IR4jCk2i4n+&puRs4TLds?pMXqr)iqHu!@OG{W37>Vs>?|c
zlH<3pz2aC%ntV^GDq`_fj5NW$d@IYLkJ#1*dsrbB8;A?elX*+^WC&5HZmSqNK(xg_
z_!1XFxta%Q1j)wCR#d}JjWmQI!pVWCIoH$0w=ue7n$1DwD1&%qqMEHii-8{lZi_lw
zVqM9-e~D_8Q~<N5(l37#F^^l|q`kPLbTu-}&FbxjWfp;O(zQXZ_mYxD)KPR-_7#kE
zP5uDR&N76SoDV*nTF8z2`8kW@<wIbxbMq$B#$!A>TE=os>XiZJh?h8>DyTGHdsPYr
zs!)-xg4jbhWM8oWi9)Dkvylky2UFRwR}dmmjNT#hVpdH&g>VETd>6D=FrsL)A|Hz5
z_B)hhMzUUGkV<J6US#jOtj>f>!yER-DWk*+q_SZgk|-aa$R0=bXG<nSyQ9M|fYyih
zCKJeT1aHHv8M-e7e|*14k<?SCYo@|_JBZ@*%!Slw2AmK6D3RCvXdsHwl>6CG6><1+
z?Og1Er5kL1;i*w-3&xS*m$1@=u>8XU@JncT+b>>=Eh`)4f`z)RIX!?f&$kibikP+`
zbg0BOURm=~{h%d$>?A|-b$MO9GJKEpbp`F1uLHpyNx6=C9NN#^@~S`bY2E-QY|qXx
zQ@>!#PftM4&gS?v75O7Wfk^Z&xeu2MJ|05`bCv&yJ0+4#(Exb{$4=#(bDaLc&!$*t
z@LSS<he-EQaPr#c`y&+1Xgd}}9IS5*nAJ$dUGX&pZXR$uCwYt=xwVV@mHexnmiG~w
zk(|XzX=<h6YQ%gin!fSv&nodBa{ko$;uHDM66A}8_v~%`C-IRlsg=`DQnf{iOcbKY
zD|yK)_xabE$hZ15^W;1kxs9{k*Pu~N=bHV-wF6K1;D(^*AJ?TdUJ-cyiDS0ohA$q?
zjeS0mlh0#@uP2n-A<BY4P9&uXj(2otQvB$?f%io+cgDf&`)y?MMYK28@6?Ykl23do
z$K6mZa&i7%!v1=X`dHzqNle~_*$(C<zDgVLtwLtkYPD?2VYY7GExI280riyvzvPsu
z@Ow7?gu1qR2aD6J!AT{$cyT&-$-c9z+}_^`N1PmSXdF%J!x94nD;0-25-*h3Azwz~
zp_~bW^G5QEYTwM&s7<I5%)s`~cKGfVDULv3w1zJUikfU$tWJkiD@+|CK?ko-r)GMe
zWDORy(Sq1-8+gwX0_%3kg*(NP&)9L**qg*h3!JSyLebMRP1Q2%4F&}jL)qh$*>QLN
z%;8-Z3cSNBLU*)WGc7=huUTN4=_H-|ZaZ;sv-04Ju1DQvLOo&P_;l`+`RN`5cAOx)
z)}VDbh1ShvN_*lUyzY84MAC#h%&f+Dw(-X<;;RXzL!}t!R(iYy<Dha3I)JQQ;wjxS
zZZs9!SW>Vx=V-$E&8MjQ4P{<o9t~HLgNfD@9Gywt*XLZ8Da1C~4(H1)tiX@>u{dXh
z8VM{7aUW}pet=+l`ozP}EnIxm344=1=;J`AaWdz%e3q?%9@}f;#ZR^7IsE$;3HN~=
z_wFnu{ac;;)-H@$o>6gm8q{baIJKTn_N0jkU-ffJ$@N;6?Ph7ILIE3+DP4_2Q4|pq
zFY7m6pc-J;Dm!tIK6w=PK<dZzhR}}7clfbMUt^&zYMR9d1)YCySFaHsX%&1+1_u?{
z_+NFLa@kDA;$pr$txN0wjeLq6TPVEc47R&k>pU~2)q*dhFZ+q#2%0u5ocu8mqP8*=
z(vy1NA~0lKKgs)Z<PQCk(sZNQX~sGD<hn9%@@tH<*2m&Jh{@r~0&ZL|2X3}@d@lz(
z^3Q8ro5i!bh~y7zVwB^>sBg&3-;VyUcz+z~si68~*j*TB=P6<p`~NmN(cp1Vsoo5V
zX(6K_*E}nDM8J0lkA6@iwR4;#C{9|V_NJM&f^4bX0;{@A*Ixl9^!DTnl{1VZyn>0w
z&F5WanHkd-EqHga0%eNf9B=~jT*vBrpFgyhzvQ!&ZXx5)5*S)}&nGR)6rr95(K4-h
zG45P>fHIpW5mFl)78*;TI-eZk%NSEr`&$k#9|O<w3sL2wuUXBwN^9nKb@H0}UCrU7
z0>7YBP0(V%u0=j#a#(eutCit+6`IPOs-nKT{5g;t7sacq=Cv;vK)j}b-Xbrg^pIcB
zCq~r>Y$L81-QTFw$(T26sumuyti}7fWrUvV@wvWSXw~4Nz(Jw1bD?$@Z5AKIc+?&_
z%aba(vPz3J`Bv1sFsk6eG{Bkj!@tHsNV)qpIIwiJ#*1xPvj*iORvsp-pvQ(5E1E0K
z;*kLJKU+9>8vjlp3V71fh1=;<m2=>V@oLt+f89wuU{i(V6x&Qdh<v!FLg13!t#W^X
zhIq^(zN}h-RDr#-$kv(f6fdFrq=IOJkuu@Ka~|5*^Sxg4H8;xf%D*4x%zt^9-Rjk|
zc>hr;Q5^}#Z12z&@e)4nc5yRH2Iq*|4ix)*0h`p}i=;~LUZ9SNQ6s8FSkPr#Gb+~Y
zY#W~b^jcd70z=O^J3RnDfwJ@^v=Oaq6v9mTp<7bSQ5_<0f0Ul9E7TahpruH+_?*f%
z2psfbHfxJ%#X3FxW;(V}xdifQN1jAHI0#Ljl9H(1A}G84$p9w)%HlSYy~h-6Bp26?
ziX5&Kaxv_aBD^dar4|K?RtE0&Q%(n-e{z_Q(tfAc-CI%a)wDpbW1Eb`+Vi^@6<4ug
z{{{BmcOX`hb#tJR$CGMp%<Q>~Lk+&ymY|-((nu(j&9`>o;!+_hUl?^gabEwH#9L}G
z$Bd~{nI5y6h&CBfHoMXNo-*Zg?dwLf?hwsoF6)jV2^QEu*!D8bD^+lQxloYzUdUcM
zir)j*8q=-w5Pl2-!1i=jMNG`iKXl!6?LP1|rCH6y6dkYDV9gv*)6Msh*kv+bdKYBj
z?Jg2wt1#Du&pYYR+!-}Es;kpYrmm5!{0X%IdsDJ6W3xMd)?;{dRxL$8o<Q!9Dhm0V
zr1-%`!P{E>;Hw%f6+RjicB%|pC1??M7C<&nkD}qH9Kof|h_K1`Wuh3?r>4GYZn1^j
zZcK8mz+0QWWd9X26wAN!)&ktq^ONbT?VRH?^*+HyZc~NV$=NTiI(3T#UT|*tNAB-d
zzH(G1$5#09G0KCJFKjS@KBc+vkPGwlj{!Bc`*PYCD~gGL3gL!jXD!vIgJk{T7^<{~
zAjW3=({8Vwr?sMahc~;(j`l!#f8BZv#1U<S)HRk)B}^bUQ-sM}XxiCw<!JhCR5R6G
zCWyP>nI~@h*z9|dG?L39;p(^Tk8LxHJXr=61V7w;Ko-ztk>u&USNAQR_iKa9==MjK
zX;--+k|`mu4eG{|p3QWRuvRwry9DW&+w~}?4ZJJbU$?M*abrK4lc6&LV>$5d)B}Z8
zK@M3=n*r2Tp2SgMSyn(-He2HK<akm~#ckhJJ(E42B8H`DBfdi0=o^YuBiHled4_4#
z4_|rjiiPh!ZHs2p34ET%IdBmt+_gzi<hD#lD6r#jYXw;Y1(NZ}=fXp$X@k$a*XM!c
ze(TVEJy1zz|Cv>JT*t<CB3ZN8c|fCO?-Dnk*nAokpT7R{HWTi5=e(3AugXK)teR1Q
zw2S167*{2dmZ+M7&j~zaEXUIw0c<XumGmXz6%6n?0h9W9>ymVFTpM^xG>6p#%?QZ^
z$9O30ROsHz+b2S&;`6&GI|!j-*un;Dp(l^_keH#2FFlAH$xw5Og%RbeBvYgp7Maf{
zk3J&&Twok%fu<N<D-v_)$sV;4W5ZIGyRPe+GmS;KhP?@}vzMC}eQD>~=MU%nR+Vk_
zg9Cvvmqa(4;@|m(c+Rdlb(HeIexxvIix5g?7c1KI80h4yEPKv$b-`Je4+>YRV$h&L
zC-Y+HOeWORo^7!xpS{8U$tw;@Ae+Sg*(oz^a%#W2;-X@vUO49{j{J3OGT&+3*y~Ib
zc$xqa929Bz-+#>h+M<Ku&8mP}1-g-m%^Od{q4;w7Ai5u;@2wHq*<dF~K)^-Wbe(xh
zIsd$cY&8Ilq#y$HUe+#^`I7ZMlZyb=dag1AgqCso75?&1YOi|oSm(gsMWV6x4AuCH
z;Ug76YLk|1n?k7#0AJSa`KZVc8NmkzDOJJ&LFd;x3RtG9<4N%hH>!VdD8eBGST(e}
zL(6aK6<PMBI})YYbZ?K6XhTo)N`5JwG|=cyYmq~`sW|u#YZC@->ax5~2lZTmA}4x=
zm>?1|uBr#q<jKA3H)fH$^bTHnVS(>Z^%Qcp!j!Es+HXIEXlcG0TaPulE%K;47ox~#
z$*GK77=4ki5SwE=9E8-Vw{Y+J$K^lb6%-$$$FhX0D38xEA^uaxTjIVGJbr?H%M}!g
zW7CRw@+Z+D#{ZgCKEBRKvsYx7U+5H5e(Zr{rTQs5+j8bZ;mv=ZL;T?+a0%7Y<k?+8
z*Nzk~Va1eDiO%mUv$kejegNH8_z~%oPo2?{sxOjiZkX6~57>xS))poKDZQoxL?baf
z`;;AEba;yt1Y)4d1|FaFceCEnow@Zn7epcHPd*9=yjldVz6lwrB+l4=ppNT76NNYr
zuZw6qu7J(?5npWI2XuN+gX6^N`U;#QdM|hU3`F;w?@DFE;S0j#S*-fC_24UJ|LSyr
z<`7DsZ4W;WdghOhzciw0JyOHnd@aSj-bTz}=H5gOj1r+jA?~vy-LvXPeu{b|nLC`b
zib)Uha|1ap{GkV2y}-MYDH<qF*F_;X1xveARu;HZ<>o~M?x0mad6h(ukbty*9yU`9
z)BV}Do=DM~tBl3z5{^)H;aq|&zyV?!HDiim%>cqRwM@Wi<ak&LzDLO~zM)s9FZVY>
zBeCe5Bc3}3*+vM6bBb0XU4=JGH^<5TdWN+|>G`_vH;&OKeOM;QSe8u7<-2+mXvKn{
z19xEQx=h*$v#C^8VOxMdGH$GS_-}OT8CZGgN?th?f;i3+&=HbkG<)Qj%RFrs+Ix8-
zZXmvr2ya8|7`Xi8c3e6ZWuAQAcSD?cOfh-uL(<QTEzzA9XM72XM{Ct`E!rv8Q~DFW
zuRA((9D_RL?>=U3NcJrl?s1wvG7N84m@AIm@=)J+d?ng}8uvu^)$a)3EKga%540n9
z*wBGlW`E&BtY*+$Njgq(N;3HeoRDRz@R5fFZIE2SbXjdelNC4@D}H{k1Wg;39VlTw
z%spsY>V-of8H^({LDhB6Cc>D*OcOhjp|L^nv1XfQt6=P#+qe4>%R1y=5$lRv^L*D_
zAS9BF+u!|i)Xw+IF24FcgNa_`OMm<xD-5Ar4rJqbrAaYu)HvMMT^`~F+Ea={dQA@3
zkAkLTkK*|@m9?SFF}0u*!fda05sn0W(PN9<*1IuUD^)9UiROs7+PiQgJJe087cOj|
z=p%61pxLC44(JS{&h6v;`&9L>)ihERvGxmV0y|ZL7;8cekJrxt6n4gnrMe0_2HTJN
z2)7CdaSA<M^{785V|e+H82{Ad#|SlV;%ZF|77SQ4>9yuUvs(Dc-)r<f)pIvInPTd2
zz(l1h_@o;<NtY0qw{)1i4|4G-{ZM(CMO~~d<?xTm?k0mT5+)g7N8L3Gb{&FK(PnOy
z<!GLCAB{tYwHQSfN5zJRo5w-ykzB90G%44mO_-%`wY^ncaT-g_rfmd$&d%t>!RXG;
zlxsrmZ=!b4++xX8Yocb+sBOixQ>$ox5}LS`MnKv-9CW|*S)OPEIqeu}SP5GCSEO@F
zRYsrYnu1m8Tf`TMu_4|8=Awgo0Z?7!p>VprR7CZ_ybaqUy%|NZwXrxW%%yFv7afwU
zyi^-+5SX&72-n8r9b$2QsiAI4-Spyz(Xju(VEUJq%9(ESsV_keyOR~z^0=$-q4g~z
zg07x)(f`W6uA@BadqFK=5-C|*lA4E&W5}F8HX&w5C(oKzK5c4xHd4x9NyR+{UvVkF
zgM0zph6+E1*=;f%HQKy2f<Gx_l(XZeu*b1`Tff*3X3D8Z$7X^ve`oN`7VL{7eN)@j
z-kPy%tz6zl2<%uID-GQ0^yDf;3@GlF#3S1f+TULJc4wZ|bg(x-uYy@a5qX%rz20@5
zffIk{B`lBb(E=F>8<(de@^$V<W0rX<E~qs}TQvcZxqgQC`HInn#=2WjOQk*?4U+fl
z&M#!^$LXQ1@leEc%SVDa=c*3NheS0q3aVJtil6|u$rejDS_Xq@g6cz!W3r|AXW?NY
z<&C;Nr+Je-Paa%S?`RtNw`t|(%>pVb5_H27Z-#5BDx7UuPAi6~t_5)ipPdxpo4g$S
zG@w?<*M4q&GK-nKGjB4;g3tDKGqj1z68{lxLfxbSp&W;e_csfQk`8L|r@|V;$DHI(
z=jQg&+^2+~BdHTY2Tj}AX#+FMwskj@yM<({x<fy~`M^+#aZO}Wmh+@+8Y+`k8grQI
z9&ATWHyOUInv6!~<?PL<vRv?iJr8p;#%VGOoBGL(w>YAaG77us0(Le?1?bY%M2?wH
z&=dZCdgZ<+bnJShSGe?9yuM4D*1bQR5A{c_OOIlk1i3{-+|!MC-^g|;iF~@9(xE5U
zx>{Ys5^$rb*lNP#+ceq(sbM8Yt{Q}FcWzVt4>#fkpA~40j)OF@Md)JXmy303yIxI6
z1u67MAEOMd5n0VhKu(1elDBkHl$HvYKH3mA)!Dd~)wWQ3Hb+)%1&*jBUs1B>H9Q%3
z_B+wF0Ro$BA{rK~3J~T*xp(bKo(?2&o4ETJV7!ReEgGF=Qq}mJe;yL4U6OAgD@-?O
zL>KXGIT_Edw2ZpVe7-hqaL<75V{37UV*$B8m#25!k?w~FAdc<n538-4=Gd#0`Q7f@
zO(fs*Gr9E*2y(RkBglF0)$dkTd)K<cVjNz2MacqoTfQQ2gs#E=WJy6$OAxBXrElmY
zxoP6aQOXt$5(<7r9yQj-FsX^px)p^h%1=9ff41}B3r**31X%4t<z(`%-LDk)RWCQ@
z4f`UN@K98dEk4*3Eqx`%AJKzH<UF)+*;JzM-31l7v&p=?Z#KImGLcI3&XCD1Ma5e5
zQIM3M)M%4!#^FKLNP4Hp+!j~m$Pdpiu0u)lR7BaN$&9OQtMW(+y|gkxlu~A(b#dqX
z`Mpr+YdRT@$9{W^=t$~f!b6BxyI_upFB@V#;KuCbZBDXTu1v=c3t|xF|D90l6YKM7
zf4bPnU?PP(PuJj)d2Vl0v#e3U)u-a@_gV~gDSf`}%b#`Blb{l$iQiW+x-DAZpzwvw
z<%o3x@dV|O6Z1%fKNpzDe<=+}@NcYn{uK<aKYc`qKSgaxjwB-oNoIc=*PC(sdH&n(
zWf+AR4#|_(2C6ChI3p%W!z>(F0kxDbWp=BJjCHU4B?cR)s6<QRUv@);S4{L2b9A^b
zD`_xyD2Q0jpAm@<;iV;%h4aB_wG5gtPN13LXod#}h_(i1B6UFwbNr&KDKnN*5k#dJ
z(Jd<mir(}wK_a&hXrFR31fE!+FGW+8ZD3Oq*MzkeQVtsu2Y)lllCa%iqRo#ydz1w)
zq3lsSu<^D!#2fVrsSWWOP3DPYnykb3>Q9WlNzqFNXw^jBcxOJWAbpBF;1p|N(l*3q
zl@fF7)s|U&!s|k?7=8#RL8E1MHkx<1hT=4-;hnPm$ti(O(UID`<<hc}%0DiY`Vmtb
z0x@#yxCQ~jC1rz1Y+ePSX=pu#F4+bHH`Oq^HCs5<&C>8<+xr%6$@+;UQEoB2VF7fl
z5tNNbVt9xKdyd3cDcfa3PY{*&rohRkiri!A9LD`hp|Ql?*<^0wKq^FO?i2C77ejQ%
z!#>rQ9UG=4=f6A-t~WE=Rep$bct8g>i2G!7&x5kW$PR7eq0|nn4@&{um__DgkydJ?
zZRDm@6Px<vZgLAR_gEY?>KSQ$zQ~b1k1mKXY?CLX%>W>9l5)|(7bSM@LBk&Y<Ypf(
z8HJ1cDA$a0jY0L2$rL@UxbrnmNCaOFpUKBKqO3QJQ>E)8-h7^~lxuVgDK=Zw=kI-q
zU%wfZX8F_4AJC~n2`WkRM7Y*4{>iFb!#!%^b=Xu>m5q%L4ctf9{}ZU3%!f-|3aR>X
z6G_%wP`LE%Ex{<mP--n^Qa80CL=~|3ET2}z`{U&)P^I`4Yp+p0rf+h0*i@dgt@B<K
z7)Gos;WTk0lOv<+IE2UNrmMrj817a{M44qj>7OJ|zC1-|GqtpEv*DFBa!HgPLK#wN
z3cm<pfiM|d#dT6gFOvST^&s>L4-Y-}ovQ{f+uRL!*`|!f-5YoZD!j{is#;pC4VhxN
z&`AaK39nUQIVeYUkc0YKn8Hc0zWcV`qvO4h%8SX18e{#V(8Xp8A>G(!tzY7<ns6-d
zlGn8$GB7tYQv~*+@l(D6@pe_SrWA8j2H{<LOYKr*!V=Ki5<keLPxsWFm=I0MiO6<N
z$@uZSdsC&u%P*3*vo6Zo-R=P)YexRL3LZ5MA0bsl5`N22ULtnYP#v9r6&%Q*j<0ac
z^o0{6q71;F3q}@)Qq!&(RNH-ZL|<XD{T4U;YQ=<C<z49p4O3qYZ)8%#+S)IX6{Slr
zMe*-Qx?6KNDW`9@)zW#@+6^MU*BBW@_iKh=7G|9)E=HnNt7*^<;h3ktH2PGXrG&zc
zx$$*^u$fH!?I49G-Z_$Hp1i27L>49LtAJOMk2S^P7e>y%IG6-IyX4a<Ex`B`v_m=X
zu~;IvXhF;1#;ysOh^T7%18lR-jDxQv(L8Xh)CMk9@KLC8YH0#QV*|=xrTQNw)v6Wa
zb-g`G^&wy0_1SWZysD;tM>zgm?E2*wLUhK^W|V-c*CNUxy<%ZzU!QoUwQNuQkcYif
zV8?4eQ)on0tLra-s*2xaP0s<9nTpWKLn&ew&FrPtY7xosJ*e()X0G9#NgQ^tv*NQ+
z5q~@3l@J5cu?_M!GJ3^&Yl2}ekN2V$y|7a=`R12s@m=sx%A&21N8ASyG+m83;MegW
z+(oX;4-Cm+{DXQ)d_!*W%%O@eBCQsXCc}G8%h{hI@CnT?Ds9GHtDQSbVq?1uNC+va
z_m!&mr(HRpJ7$}-q+C=iT{c_Z+*m58jwthhLnM+8gn&2+aCAol_4Vfe=X{b#3~`q2
z8x|x{1!j^a(Kq_}I3Tk3sNk1>dFN4f>qnxQG9BaX7jh=Z>ynaeWA<5SPYDsj_$esq
zKL+~`glc~%cI3bML%2)n>-vz78sseMeK-(()rjJxB(Sjxz^YCmsQi!9IN9?CmF=|@
zL|6NmJr})lF<u~ti4{Z%6J#C0D|Srb!ppSyXIRh`AKY(g_hx;MRkU}i2K6H5;P|{o
zIrPlIfg%E<QLQ^fp&QloNu!v~svsfCR)(FB@`8OL%8_oGOuoI?6{r%*8k3*%;OP;y
zm%{Dh@u=K+7yicsIfa#BB%)^Yu5?qB6p^5|0O2WTBd?50uq3H#h!a_Z((d(x-@f02
zmWa%0LiF@gg6Um1KDCu_szG<pP?`ww0w_|*?Qiz0a)e={^vUBcX7YmVi2l`Nw5ni<
zDEPP#j6LIaMOq&m7kA_)yM_qY0@Fa}w{c~B^S+ZRgw084d#-w+!zLNmS@U0SdG?1;
z4!L`FY&MXyn@3KD_3vvE4mCwWQo|&12rp*?bP4TQly6e?Wz4Dkk*o=_8)fspCI8N6
z6uO2RfHuw<uL^dki0V1$_PbHqV-wh)#6#T2ee?g=%t&Dp%-FBcXC->*F%Qj;u;Y2@
zvd1NUZyfAOM>y@0iUhmRnHyTj!&){<A>Vq;niL$X_~57%(tNxGa&DWUD)yC8d0?S@
zpN6RqkwdV8zt}%fivWD1x`{yNpI{;6h<+T&stSaFkQ1q5#h|OkW<lj1Q3%P_*XJoZ
zYb%Aq!zh!3t<8|>q?lTwh=qyU*-r-<y;d#-d3%9JBIpN^Z1YLGK$yMiwA%W+--vw5
zuCF%o)ljx#PnT6i6P@6)qeM>6O+J$|KhLLl;XV)RRMUeT_zVRG;XZFP{V?+#XG)|^
zgfp@&(wzw8$FVrXAL_kaDVwN$ouOIWTu-K0H6aO`ujfOQ9<|!%VG4;Wk{(Eq?`}*4
zopb1ikhX!khCfAAIwEKNTM&r2+SVcWMbT|OiH<_Xbk(%Fb@8~cO7h4__@#T!aju(D
zwl^pmcDoJY9UzT;HDMpD7Z#+xu=Juv(NC6`z(qu|eEbPjis@$8m%rZr6XvrsW#7{T
zvf=m|ST0k5OtW+tWk79b7UhL`d2)Qri3BMk{rO$3Mr0gc9EV4Jg{vDY`^4woF1<dk
zjZbnSTXPEX`oZn+5vAo!)gJy_3<JUC2uZu!lWt|xs%tXMM3ZbP2AP9W4<DIH$tY-V
zl`>w0WpF)l=C(dp%wtZIIdLdwfBTv5+emSK%30UIg~%Pl%3<B%aWR`@lJ|L_Pb)qu
z#uIoB+-hXh|2PB?F^Ht1T#JW>nXBEW4`3C^##yRubkv;<NicIT<|rfxbkMtb(xqU~
zsRh$G0BmnYVhUZ^yCP_gCGKKDB0J_E5*bVLmch^A&?*CJs&u;OqQML(;;g$=0XP1q
z(b|MDiF6Le9tA>og#?$WXsv|6xq^}lJ1cmznjsUlAE|WiQM{glaaZnB8{Z`|Ih(Ai
zKvE}_Akh`nETj&^()Ff6gAEGeyoZ`&EW4ItuIx^aV3{a2$}7_*Hnu}PZd}Z>m^7L9
zbVxqKzt3ZJ^i%1A`{GXFoNz@WPh;7qYg3J=aZ*t&=$?i|TY9F@xJO`Y6G0_*j)Q2f
zXt3f~1z9$<H}F_K<S}EnGs?R}nZ)}smXLIycu`x$(Djx=4_0gDjs#CXL93z8tD7=u
zmh{a~%Ke&Nuf>G;xmx+9lUVpF+4|Ru>cc_qwVK9i>8X%ukE?<36zMS>ZQ3`L;-J=+
zn=xU^oJ2ZKNP!umI8@rlr4-A;_(50dl;?+zlOn9H{64<l3-mKq5tle*H7hThzUnq*
zUlI|D%Y$XBSiVC|uchu~1dF3^Rk{5<!0w3(e;{X|QZJ^^z_GlfSZz5y@j`D#f{zPV
zOfmdj%A8TC)H}M;A$46FKisFjZZ!A<oWF@)Dds8EGm^wzSc6JDxtObJ5On|)ncmYn
zdo$|Lw6KMuUNSrL+5vM|u1a?Iz_L*A1v)iBIza*5OBy|vIvn!>+!L9r_l}<psnNZ(
z-~Nax;vK~I$uEz0ZxQ#CTM+Mf#63_+NU0dXxtlbZpXf;U(IbBe$6;cV1OlphSH_jM
z7g}q$6RJZXjOm0(&$LSA>4-?EoT$+jU{!`Sw2M3a#QiGYz>kz6(IbDCYgIs%1?V77
zL*Z=L#uP)b)o^!w>u`JQhpH_?k&R+>cpn}974I1=_v=KqOV}Fwly_$4w97{16otGt
z&SqH4jWOPrQlLeAs31IPrI0_#M=|*JRY${?a>_JkhLy*RIRo2e%NP_~(^=^Xt-RQe
zIa?XcnMp!+W{M=6pg+eO?@;lrVNk#3Ge)veVq|*l;lrxM>f~WCGS4{hy*fsHVxM(;
z1O-=2SGb$V8g$Z)t-Q~SsNL9<eJA~T$IMnux}5>PFD{aTxFbKoD7sTK`GkC3NGppw
zj!hxM+9xu#UH;BcZ-_V5V@E`kL@DYDazgM`#Ig5`n#D5>CVJ_ek@v#}9eY~NS;mtc
zNm3Up5%UG>P89Q;=QJ)TT7B$IbKJPC3M2t?`R1(^QT@xlgcP|75(C^#LBCHrbhxe;
z5f&%%!r$yPcTC?Cu8ckWprcLTlEXhKVr1HpV?=P)7+DE_L@bsAZFkM>)KN8Iw@awi
za?I_2xl-(t5@FJB#JW<Jzx0`U!9aF}ao>pSOzF-T3v@V?+kB?h*qNCKh)uo~sE3H|
z6n`SwQH`4-7{AQ>ntOamph+R()^S+B&@5Cp^~Ly*oQQ2j2&8%W%s&4$>#@!H*3LL6
z_D8xp;NR7?($0Z1!EGzPTBk^sD{(rlOIg0M8VlI0C*Pc^WE_9KG>!bEKW7?yucB3=
zKze~(>&wIz35EWd3RvJC48ZpL-?N@;v2UW?+rLAe=M-%Q6vv40;vwLW;lty(FrI$`
zo*e{4&FLSu5PD6vCUGr<)kviGhk-7Is0mZGa52<GNHmW^gu~apf9mOJDcpJ#K1O_u
zio-B;8X##;kUWpOEvH5>6x&9rg|poQ#n(N7hX(qKS?5P_4@pl<H9;!U%I`Q){R%MS
zk;2^)FBxW4>c!#3M6u{<TBlE$=t3dgBJ2kQltI5R*IuG4e3VTn-+Z+yLYUx<i?XN^
zEC#=NqfkrbMJP+EFrd2kx+pf4fhSQmmb*)ZUtW>$(X&l%(#=Q71_)@#-NT?$a`*9J
z&8=HAw`M$10feA5QTs7No$747KKx_W$zXP&kXGwrEA?TTUIYgOM||{0)0G6BQg8Ql
zR$1gJb{*Hl1O`j96Ff+s@@M8whY}=WCpMI)a_!mQ>z$BZAr?y<e8wMMYxwiO7<&t-
zDBExSUmBzZ2M~~ykOl<=X^=(`B_t&Tq`SL&XhuS5P^2WKkr+xOq?@4yBm{>1-}Am-
z-}9dT@2s=VS}a|PE}nVr+V|eq_1R*RUBOlE4T87$+ua$mb1#Yr^JCxJlthm`RLC7u
zg<#R!1MyIKbx1;T<Oio07Bfbeh19V3zIm5^paE^vaP<XHd)G8yep@K!po<j7eMHdp
z*#C$Lj^}%Shsf5UsEc#@&NN1Z4HT(s7Emg$H^j)W5MZplFj+RoV$Wt;-g@QL2+R`(
zuX-PzQR{B+yJ?FS8E}(5N<TLb4CU12+ka3w$*Il3!a-#oe?0f_><+n;@gzHxr-9)>
zuIWhsuPBIwne7}j-I_KJI*_uf;h0+u^oHWjV4XQa!YS!``D5x>le0LJJX!-wD?Aj~
zw!hZj<wv{u=bTJWlld0RhT9t=9>Iro=+w;sm80M2MfYd#EhWO-rY9QDiZ_IqZmWQM
z&Rig5QWTV_!bXG~_|vE&9`yPo4%#^lZ?23sJswk}G0RM*`&wD~srV4??_E692+J5t
zNCmCU+7Llwlg}}AV^!T^e$$`ogHl19M5m!H*<q!hVR?vnJhGTG`J81maOCUb;082v
zJpzp0+e5^+9V^HbYtq$Rtcx<8qJ~;U25X}B2#?)gbYPs(t-0HPl7Y+}s&l4^f$el>
zz8=%vw)AFA<G8Px&?-gqCot#OV;UsemTn~Ra!Kj%d1kKW1=4%NSaC%?Ty`2QJTM~I
zwWjffncm%dK1>s!W!3jegJtTmnH+keKCiTGxK-;^rf_-y5x>NBkxxPj|A>RQcsC{#
zIn@S#@^hd3`=%6)to<ze*pFSJ!cSCOo$VnMvSuzO2WyB70iAN=L7Zn?vach4Rn@+|
zYfl5#JEP*4bgR)q7$>dZGM8w=I)@N_yNF85Wxj#jM90e)tiFWSQq`@LU*9*gj@=_e
z9v$;*sqSghS`>PCi4i@i1@HZ1?d2tZVe#8|!~OUZ^gsqFNwQkp|D>S7Z@gioB%b%l
z6u-^gGd`m}DR}nsp;Kujw6T39-F<;$Rc*z~>j_zYw|8d5`;nuV<Pp`ALy3r|eZ1~F
zA0p*gqAhpTH1Vmf(vu;v^$wq&$<x~<JifE9<hee4s})xY<_0&f5>H?9lb40ozDt%a
zX3q|h-^U$im#}3TC3HZX*VA;DHj-<`5uWaUQbThmGnSwBOsigLVQ$Eww`Vu_s&N-i
zz|GC3Y4&mAJCHf(H{?Hd%8!YbQi?9LIH)r4sa5hMq4*u?Lf)^|`M5RiXC3~t>IVe^
z@5+}wbszhDKGknrfLrj5#hQ!zaoId*aDgvS4<Hr#Fn?3h)Oe&?34OX|Sg*nrd@o9N
zAeac$gdgEA=4>T46?61p@H&ef#M^E2-;HHHi`|&+vR&$>dtK*U3rA_y3DZA*UJ}{G
zmiU(pDfqr9a?23z<#2+X-8ds%F7P{yZO3+C;$rOm(nN6FoZweBSzebE%}=wP7>T$W
zF33Qqpm%@h;<wuRvwER!%j5#i@Ackv^EviPNRFg^@;#5V$Pk&orG1n`xR&ix2GMVV
zU+Pgf4yabwf1fCe+Wr<E!I8b5dp9qsbPN{ufg6GCwAn-X-T>A&{eFRLsaVs^_ouXo
zoOrLStc@qc7h?hH`YzRbv%}bcv+vB8#Yt`DDTG&(oXP2@o87aA2^)@W&pLVC_lpOs
zcW-J?$IObA2;Ohl)}5V!^dw)5ch{(7;3@aBzY`?w<xgOkbv3*45|J;f_GG`bdBk7T
z_WMY^{{Hxy?$E~wOs<7Nqxa1nbN4+`HoXsqj=jVwDWVzQq<*>cIy(yA#M5&#y`{b0
zlKFCg%h8>)$}()mJxC#P*RDyE7Shg1#?F&BaLyGWiCEPx?73f9Q|49|WkUxP@-V@7
z=2^c|SLJz_5LH>?@1yi{dU77G5*8N_vYCwU3YCS|RDS>6xs$ylvO|KTEd6LQ{k|x_
z=gEY~bZ`?+jq7Umke&fZ&Ea)h?gTXvc{a}M`hA)2uV-o2`^`3D=myjwsZTJpHu4j_
z7u_zsL2e;DiTg=LgTPsmTcF!>EiVDv4_2O-wC^~H+-QiC==8}o<tcFr)YDY*PldJq
zm5rP0edyijm%R7F_WayjMdRT=2ks5g<-Z-Oe@<LrTVslnWQ^p?N&e9h(}jaePWprx
z9r|8J|NX_O%u-}-!2}8aFo)CSq2U}q->pa1@-#G&)ty4;7=ihv+_!MfnOy(z?d?Yd
zaORj*h0LWsz;%=<<Ml;*4jBx?WQ17h8mELV6rKWE$E{W(Dseg;{jgamvFmzXTGyn=
zC_P2u6woM)383ZQbR?-tDZOugu=d^2F{{fL*98b5d7`>KvpS_3f?ahb)#wdkk96MD
zM+Vy*C;!-2u_G4H#T*1ZXS92$ZrzcED5N+=Pu9<~g<NODI}(Q~%|pvQKJI7KBbvXi
z*oW@{cS>R}zG8`f6<XY*CQd0LJV%mejL77B{tLK26g-w>xnWrs5v4~=Km$iWDJ5u4
zLH``yuf)Dz@sn0Ctarqeh2v!lL(Cbv=d-f7{<|mEPRO79Ohe?fLNbTE&d%9%W#0)M
zEB=}={ZT$F6zx=<2D+7R0wWKSi{3t=79{VKQc!H%YiX?OC5aTp$~I5pV4m>_n8ZT6
zr7~m7X(q#o%=aS<6OJHIgYn-(5mYi~2})cagty<q={2R@;qEl72rJcvAadttZsN{h
zgn~ONM%Sasa}?>|IP$4aZuOpmBfP3?{ROw#XhETAiinb|B87||yR~;qN$n=aSv$7w
zw5;y!Z}e+_<*DpFeftoDxvwG^#N;LJa|F-VN)z$eeJGY;8L9i%lqy%EDV;t(=}*sn
zBkt8MK97ePb`xbjyIn<U|3@M6*OY2|Pg)No^ZFV=jB4C@0u6Tu5o^ZZ_j-bv;%`K0
zB?~0e`Tx8jmos9G3){HeaOGKU2cwM4#By|CR^QvRXEAqe?3%n%^Ude}dx9EsGCBZ4
z|CcVQ2I_$d#5?rpODL|6eoEs*qXF|gVrFp#99JyPvsa4Yq5HroI-?q08Y#>sWT?uF
z9TDu5Gb4NR0}~yTtwOSc4kD>AijvTE*SPTL{^T)Hm%sBY6Z-CTmg_iyvbPS^H=G7|
z|2NS1!UiZhd&^p{&Mvs>cvVFtj&g|2QBFg#ovtI3-<#~Z{*3;{cnR!Ag_%Y!!*Xi$
zui{uH-m0FOPHiEQo_<RO?m?9@E@g@ZhJ*qqb4_hn0eg14SEl|uw`7hHAs|KZ@=_E)
zRJ-M;vhcn+Yp;WzKfj`3;E_%rFyctnRDEYJ`?(#P`T=L^>ZhkW05GI>{PpID=X%zJ
zhq%A=StU8{R3QHl*%q69lNEfmx`$Y*o=t$Adp=YrnYB_q+#Z;>e`>xvOHQI@XA*Qx
zO&D)ui`Nz9Y%om49>KFBYshO>$d*!|J8&7(koWYyH?hUdqoYvXs9&KLcU$$~5rBdf
zx+S?Vdm?{YNNk&_9wgT3-%P;d$U>lfTw*7N_7<l>)#!Jm?Go@kdp;5<x1^x!XlOG~
z_YW4Ok?v>|wYjrf3)-68y9S;Ot2a0v6vX!)GfmW~P5&e}DnC(>uk<T8rr199^vsEg
z&iQMhXoi=83$IP&y;OLrksnf}tm-ppBf}vI-c%DeN;5XkGWwcH$h@=tgA*3F$CeZh
zgZm0h-#~OpAfBR1{?Q)DWOADhs2&E!qC?TF77!nPtn$sV#>e2%`%5WhS!8?8zF}ES
zG7}cdkeCPlhk2tUdtB{Bfy^CuaV}8cIm<*q1zmvh$R2L4yO7$VDc`8<uY4^}o6p0n
z4%z6ZWaX@zb+Q9|1Nwy(L-FUd6Nr@$PI3B)#M743DL(HngB&=7x>^$M>8cG%eDwso
z_BmK_cjkYWxto)ePEmgKh;qECWvCde-{Wocl6m|V4sua!i=UcP_!$w23Gn@N+5nrX
ztCQ&E!))oO93oipM-^{c3&`PE|DMQnqX;OfP->1P#2fa}qt+dIAS~iJt>2#<;GY7f
zaVH6saS{Ra8t2cFVWsL145H~SqXYEKVQv)}<Z-dGN-(=;eGmQIop13E6TZAe+JOCB
zzhR6qz>0Jj+73KMgZVTMxyVn~&F^)}Mm*&DMxgDqEb1E+0Qs#NEzbh_tr+}d32N*Q
zgFID%P}+>-EBYpmC_MRl^|yJ*1hgn>X`ss60$f3FBZ!ZIf&1K^>nALNqZT+!VYnfF
zPj_cj%1BhB3u+|q-Aa#iF?-o0kU*-k``i{>_3xe(q#3?(x^2NYna-I*HFOIYa^5iV
zy@fln^jouZB%V~hIIINj#|(w2D?*0^Z=KP#2V@NW<pus#nP$LS>TxzDJRVs94`kVP
z5--fxuZG()PQoHEuH-#?AC~k^$EaosS0-SS+NsDAk}=@r5yzE48dS>&E~&~Qj}{#x
zd|p>&Bb>QOPk-joU0O2TCCyS{en4ggM|>HF82LDYZgoMCIwn=S_S2ZJoYJgcvI!+r
z$MC7%gvaVH!f=TV%c?ThhN3d3(-$1$T>|z|90takniLLFCzaH&$6hUpUljWtuo5~|
z@PEcxoUMU!?<+<Mk1T}=*;(EM9F@*mmy+>3&atV#C{9OzQJgHq{JK-&H_SP5gZh=G
z!p&2@$toQ_iYm?XghkLJTn(7X$zWe)?^KB$Pk!C@)urj=u}<YC8^cJ@Q#HGnBqw5e
zm$@T(D3pzr^rZ>}1|Ti*O8NK7hu~V`$*+2Sx9+-%(aA1X0t@+UTM&}bi_T0+_9heY
zMku0oVqFj9I%XK6KXCVq<$MVAY5;i447jx0_2o&10yB0PT<~`?6mYamds2OPiOF-y
z_dQ92VntZDW5ecy-BRl1<^{Ln;kt%AW<N${Z@3X0>IgFSWx)^9dl~cknLX6!;g&^h
znJF$P`w%d1e)be&<r=d<1RklkR6m4=az1yBDl774WY55LQ~k|R(+a2DF}#@>OuM6T
z^RJ6fMs>iW%X9^W<4&@t^)c!aY~EWKBIwiHXM9^qEk-5EWeVZ?jNYwn30Ij1$q%2U
zagc_ocHXnUmN-a=_h@||i`F!bTisJlcGNn?<Kk4+vUU)uGWu<JSAa;;Kb&BcB6{pc
z1e3qC7%6G3M$2NL%_`{idwA)Vz2%%MJqwZ|hICdkcpGj|#VtqO;_2N3gXx>POD>u1
z)UYDX1nRzo+7vJS@pvuTiR9IiG;28S3S=5K=02i566b8d{ECckQg&W|R8O^YGj-_s
zr#z=L8^$A~aJ*v-r?cP|R@Hz`xSupaf{e6rxMz*53Xp-knoS6l3@l3Ju6xqMM7W~c
zSQ3lk?J3WbLqG6qXu}vKnqvg|55@*-zWlXe04-;L9xZYM``Sa0gi;9@l!cQwp3MH&
zed}N0U*1hL^2X^o`{{dg>T#?E!b(m%c5de9#eL72Yx&WgepsMzcRA_I%&<i`It_l%
zpSw4DV1o3Gs!YyG;Z3w>#P~~`L7yeLtQT4Oov<0WxH3U<;7ko(!qm*!*KeT9bRU<2
z#2?p&Gtpc?qNrR&)*DVyrM5D`kSsON@Eo{4(x(7&0O9C*#v1*ArOvw`{<*456nW{Y
z(g^5Da-Wq=N-2;HJ0VHVy${<2V%C3Am<;r=<7U1Rax3x*=A5Zy5ftdRqK}X_sBYeR
zB@XX{ncj~M#(sr)ixhu_q=4R8S|Pc=j|H7zFG;Ig3`+GZ0F%Doqow%ULwE@^g<iVw
zyIL7NX$4yFjaPEJI@G2mN4x5G8Po!)%-*R<rJlxM`NDBzND);dkwj2W`knk6-xksL
zm%hsobKTSb#c58Qvngn67gXP7^6+<7tQs`D*oef3o~j1F>b1Gk!Um=y{H}2aP51+;
zmx0gQ^&_e!-v=n*YJ{{2>u6<VGLLXyUAQWALWQCNFZvSS_x1;n-D9v-=b_)#prap=
zY*Dtwk!2le7kGhzAkm$1L8>}R5y{k&NmZ#B(hEt`D5EVi*i!CrGvDvSzD2ky>1$wH
z^CJd{VJ^pJvPpq|&87^{SQtq23WVJOX13GqJu|rq(>Q4_1Z3JVMIKl0HO!pg*6X=I
zpJ!Ueyy|QwS=r@(<rdj#5rMV035-0Ju<WVZ6C}(6yl&>yDA1IG)rEtkCp=18WK{Ni
zq8CRn>@JC!{6nJr^BC*_paC!9R?3fKHS+`-%lU>~;F3nw(ou!NpUg3Rx0BrE!G{jU
z1Js1QkR=m1R{NjO86Eqy8gP6ZLOvLo-vhC60kc3h$Zi(I8v97Y;J3>rWV)4(S|g$I
z5~^&+d4S6Hcsc5tGK1b4E^aHt-2PnNQX5X3@6zqOGrje}^#Li))TyN2c{O}3Kd@R|
z?I?7Nbhl=^hs=>=gaM&?6%i~lpcCWOecn0B{X8Y|OJB0f5&t)vwDNYt^3Tm5^LFTZ
z`Q$oE<9zb#mnm^$n9-$ZXcUiDX8y;u;~OqH?WnIp8U31VmMKO8Q^)2N8n(GRAG{!)
zqa=D_#r;og%$teJ>lfJB#VEndqrl5KFMW%tLijEEc+oW1he8W%Z?kryTdFrn@Nc29
z`p0)o%0h$Fu}Om@&kj<pJ|Hlb3H>p!APq@1byXH0^K&Ta#&hU%pcc65oHnfJ4C`Hz
zNGVQu_+vMHo_<!d$|kJRBXRuxWHEEhHJyv)#u%e#j|waf<wr7Er==*_xbr`I#MI=F
z_aUN$>7}IR1DY@!N+_Rt`4`6-OTwIflC#m_O3U1)*E3<@P9Qs2K(p_L1-FMCtb6_{
ziX){n$5Z6ApW~wQzslrS48d<vY#7#py7$4{FiZP>NrpjJ+e30^Aqz{Qi&B}@)E=4i
zgaZwTvyon!CYiTT`?wbvNF|-3`eGr^Dsxy<athTd9((i=Ywo?(09T>uY9BGKul&9e
z{wlxY1aqkq>8!Ayngtz^c)a7MSIwGkyg2cGBp+666pSZVdbl?>ysMS4<MUD&M^oBx
z7tu1!)ez)(<(1{MO7}zJ3>^_@ud!atl+3!S)T%o^U*!JifC%X2;!6*Ren+3FO*7~*
z)M1{5R9(UCx9EIv5kd&cAj(ZjuAEZw_0f4nmcyYJw(bl?fefok-qMfM?`E!_Uz$j~
zwU<BGWX~pkP{uC4uw#FA=ZwNE+=o|9s=F=zIMhX;3_cNEW&gY+51<$_c%HF+{J3ej
zpUz6Owl^F{_`K*~$a(&w%9ugn97+khaqK>~V?b*w;Yk_ECwI5;3pkzxfGct736KJO
zjrH*1`PWAMKhD5Ef5{hjL+iLg5>xlaM}JRcZ_~%y3C$ohq?EyxRBCtPHY;yRnd3|D
zf-BK~pTMnVB63UGaO-ThTaEyLB-FqAh+=D$rrIDl{;xlgRC*CG5<aU6&5h0ReM{&2
z`sr4z`m^QFe?5^)><z6Kbil<cx|*!{6L7DpQ4r3V>N2J%3SgFbe+w+DQcR;J8Wc9C
zEdceyB_o#qe3(Bm?Vr{2KRl8hi3hkB&W<ehh#&LmfuwAs=fiA5l+MU_l2mxVlWYeC
zE%2cbYON>eCR_!4A}c`2+U$#UVagT81OG0Yq#iI!c0k9<s`YpN^Qe{qdLS7-S0iYb
zP!bQMU0akqka+iLE7o%WKd}L>fsEEhfN@5O0-8sKYKGvT!4vlwX1l+<KfFXJ_L49g
z0&eoZJoA71SYIRCzzMdNp8@PY3t$UQUAl1b`TOOUk+$*QPsP$qcOBQY1+=N-`IDB3
z-XRoKWAxKs2uoi2Uw{uZsmPiI2KFGzpRj{Ia0RqX-{9`Nl+N(iO`XDH%`}b|KPz@r
zQ*e}|c?_%rCO#wBjrFJ(zJiZpFq9v}&Hlq+_`kk+1j*^l!qsaO_Jdj*Fm5KM08ebk
zZany6HsGEbdf+RWMPP-E7XwoC?k6E8I+i)1vq8rDUi(52nd`HORtOIW_6A=^bAPey
z|6TvrxB?z0nCvT+V@yN;*QfX+Cj+DX@7KKoL4B^7-1r!rKi^rQ{tH+&I9d*XRV&mH
zaNke3A^JaVTF63BGw;9s_y6rrG9jo^74Bq&;D61<GwQ$>+6F&VXBR*~c$xKGTxrg#
zDC_l*#C22A2JN0lILglBoBZWUK?(lCKUpR}x0TcbFez-OtF6ND={rD;$`%x-w)SM@
zxBjn>LTQRxDlxY$%~=0D$c;ci{2ng@4H0XN0f2?G7S0$w309^x8H<YL`K(S0-@n*T
zeeXXbk4B%)0ImERkZqiQzp&YZ<2~2|41_yijZ~ue2jCL!nZalJMB=r*?C4mueynBC
z`OlR!eWyskig;PuhK_%v+xJEJn@+F1d&Av}Q4$;!xZ4|u5O#o+xeA!nL=zl-@pUHU
zQ3V#$<ux2J?)-1B_CMb<#X}8V!AMNgKjU&Ox7Z_bRk;IVg5wcoo~b?mYwrvoN3j9u
zVXMVMz@i-moR4eaN9#%NF7kK4XU1N#dK?Oh^G$(6#)rP0r1FzsXe=i5od=DLHv8AT
z275pTNcpb?4H!E`%}a##6??vW-oCKb1eNDJP7f0|cZ%$<0gvPJ!wLP-!l&<c5$%3T
zdO}uI9LNopp*iskbJ!(FjC@7`a_p=ef(ZpH08FZR#&Q{{5g|Diuc2d^fcQU5E52wj
zp?YkFpLG9c=(`2bL2}CIC(7h@y#vG)#|1pTy;^`IgUt-Z_6IzV8GvCh1FLIj=kvSE
zQd*D2z*r^6)%rgyfb*3l1Pbh##0mfyJ5m<zV0>#)05C2y!m|Q#pNco)C^KXacSP`~
z(3xocnr?paa${d2ib;S|Yp`wF@6LmSR3^qR-M;~^r8}SO0hb2-fcRhTxqdMYXr1f-
zxcvXOPe%jx^?^XrqCW%PrX#Zmd_}dyK8qOGFst7`wC(%m0UM0R8|d(UqV%7yKp;c!
zAZr=lW+8q^<^8K2M_$YkJS$)-89rH71KlBp53=@r){9w*;(%PnFmbM#s<VCf*PWfG
z@|To5_b502F>C+lO}-^PMPZek2l&DV!Fm`JTGsyeF7el+0VX8)c=w~&Z&Cf(2adud
zS(lBY_5py&weI}_Kp)ux)SyCIMqGVT6Y8h{t!C;_e{E;G%VqBw&DHp`>KMRJW)1EN
zHNU4LA!T4Gtk}qn{_>*aNR;*(P)R0Jv<+c?Uvf!LdVyZfeq|aY7e{5sy%)~sEjY^R
za~FE%R#7}@Kn#p&6|fp%MxA5Sdxd1POoqO{L90kQBh(HH?#HGyGgt@b;5GEvqh>-K
zipeGDx-~Y^BXKqyKOKTcrnFZO-E0h-|KwCEK<Bx<2;90R$;7D~?Fkq5)x%F}L8MKX
zKAX_N_>Q;l5D@Ylbh`)-!B*~;X05-*4eYYVvGzV+SF_y5bxaBMoO<samg|)}y(p!V
zJ@dal@jnJ_F{TZ{4t_3FG4P^#D}u2_N5Z3Co3rG)^|XuU0+5^mN$tlr(<&19!DccR
zE?)EW4+_o<7-XLZKfKPnGC{jtw*`7P7B{LD#Zi3Cm_K6qTCg)9Fr?q~<LG=?p3b%!
z4<&M(Q$3%VKNigMQ_B<{e*M;}uyB2pc$%Z{9+rNSDH7_$5YW3*w~%e;-IR=#PP+9?
zenmzsn?3_YyA}hUv}4rf1`g;&?rsWz+sY$Cty7+^Ah~zIkT~YR+63Tip49P~>O<`j
z+hBOpeV57JRitRWv}zk0w*T(`{fUf@Kfr43xW=440OKP3%2$|^h;bc?NQg%Psn;@{
z)iChSZkerscUL>?-?SfV&pAkTI;_1fUZo@!dW1p|crT2rm0cl+{Jmy%&h8=3X?Ft$
z<Nc19&cSZBG^vvPc*?PtH8)Q9nNRb0{E>K%Va3n7GvH?~|M=LeBo8U*-TZs}Gf3#S
zZ?mj+U?+f~9JxBKYn?(Fg-r#g%RyMsGsAE8z^K)Vf-1uR43=Txk$ntCnvMycJ?Nmj
zD^gx}kqz38HUq=LssL9`noqvMRtJ9<MbBSrx0?xlAe^Uu5uGkC;Tm)hm|8bmA&vGM
z67Xg5Jpl=4>!U!DPwCE_^@~BZefK{S=-8FFoPD?iGrAgd{uoMM=M`d(LxnBpR%aJ(
zrXzkmEF(f*q{eok6l7}=Am92LzPQ2Fr{MYDmR*}LDyUtl|HZWZXUnGxYzN+bSIMgY
zdaTGY$_k>Rh9Tc?y{yp3^m~%CNT3r&An^2~ovvl*&eTUgqWpq9`ZDBF39+~KQMA0N
zz!>FKkX#&8<RW3E2d1Pxc7Or;_;49z(Ef|$MD0X@KM8`7cqv39z;s<2C>XLWqWBu3
z7r(w{{TzvqWUZ%s0z-nMWrioOc%Ar}D&w2gcTd+*;C)&k^+B;AsbtGgj5aQIqPp)O
zPipH`vpUI<%0>FyA?KTqqzJk^+ze0zQ^9n%X_-XJ!F_g<{{W2syx^?E*IIU3Mb2lr
z#dOJB`S&9#;3p{Plvx`dqog0$?cH+kI7Jf3z$XDLZgku4=Zmspx_X3Mc6;n81HaVd
zO8{JZY^M;MzXuR&g0t~VMVy>VUXP6=J{SXhl)Tqmkcv|q8qOr(3mUM#c!ebH!26x)
zRr3cQmH6SU<mLg6?Cl7J>f0zgFqvS<P}hwoB>KfG=4dolqHWO72MVtqD`$l=cjit5
zp5rP))`1fnx1-*Ml@@cpCrcD@rvn^>rt-9|$vuPiBOX)cX$?Prfm$?)_7BM`#UW60
z(X%gC+y-#XT};)A0Frtj<rG%@e9DE0^lmyYb8qg)`@DF=kMqq4<k=8|`yR*-8}dXs
z<}aYOux2{RZw$1gI6K^}46~?^x;GOjL-V;c+<00}b0>KE=osmA)Yld9I~B|>nCOrS
zN`5PFv8|iYCJ1U>*?FA{b2N(OC2&^iDV2KdHmnoy(>tY{(IRLvK~`Os)qP}-l+Mfm
zHwEZRzL=)mQkSQBu`REpT&{Dhd09MtJfU&7d0YcQStV+F3<x84%cJ4MRdPmwG}6$5
zp`vdLnn$!TdGx>6w}oswK;2*+A(`z9E*%6Tg1(}cbdHRBO&IK?{ET}DnR29p<5-}1
z6vyALx)^QD^r5?ed?nsul80}q)kGw_!Grlrzl{dwNxWj!y$-!d|H!K*OA%~XU|>Ap
zbu=vRjjTc@UcHl8`(VQuEUR!mBwRy>aVWfIVpTR1Zx!?4ci0V)M@=tzCrpEXP|yDg
zn`exw`+tH)Py+zhkO(O{I-);8RM_1>AQpiJ(eG5^UV?m8LrtE~hYIwtF7)JCaDYYg
z(8zQK3P;FMLy8!}Dxf}q6!%c~>?{B6iTbA(Y!Q{1I@OTw<r5&FB@+dOmJ4FWZy?B!
zs&h#lL01f|CK3N#@)SwO%uy=Q43KrMiTIs75IFK~$U}u7<c<BPJ;?(AP5K3xSg!ZH
zS-V&eR1i~o1Ic3(v`a8$zRAxC_q?Q)?W<D9>aRNe(yXN|e#+e`RZ3IAhSq)~mk@!$
zYr=quey5WLm&F_(gDDov(^$h6>;o=c$KW{LgM5c#Ky2x8d{;S)_Ny0MMpe1(NlNIU
zT)=bDG>_Zdcl$R+2M7acmd3kVz-EZ-AddDS;lpE_V;DDCjazr%h`T4t>1%=qRllvg
z1<~SYL$9JoC@M7Qgfh``nn_KELVJP`px?pT!ly&J1&d1jkLh~$%{58)qGiom!p$Xq
zDMD^UTfH}w-t7pY*_XUfHouu#&Bd8N5~h9<o%#eRKE9w+k>?w1DRbU~j*C{x;3QpD
z7~3Jj$XqU)U&2Ax<@5W1`qArG30R7y5gHQgPY<xBVlPHhHRPkYx7eL=SCb!h>Ohzo
zXkPj9=>~-wWA-MBJf&@aqsvD>3MX)q^CsYn<TrCnsy&Sn_~xJ+R)NP(Q8lZM*_YWw
zKXc@spK=x4rHvb7+*Vh~hTA~<@f}jqpx{dL3WQ(bNtT}Pu#D!V%1{~B@(k3jh)+s8
z)tyW4`*h3QV;vMT-If_coX?&>;wpcI-PVrF7YY?wE`BMYpnRM2yufd4_&-aiR4MQt
zzoM_SpWvK}CM$?B#vHFUcr1CSOr`X@qDZyqG%vfI_#s+UpCye8{7eR++tpdnkf{(U
zqqy52%YApO-2f-_OqjR#UXS5n!g4W2Ke_gQq$lOL-&03g`7Db3-}{FDqRsIqF`k7#
zq3TYtX$riEHw>+OKz+dW)ePBU@L{+F#TBguJ2r~OxaTXh;yqs^@frgSf+K1hWkYeI
zZe(%Q{y0@t1d78H!-YBdJz6{vNpO(Pc@Y0e|B2oIe>w5R1+WzVSGF4{vKilZt*Xet
zRHD0J3PD$__{uN^BbWj;Oh4&0W2jk`agt*DL{Jd<C@jJV28k8ed1K#Dj<u-*>Eq|(
zx2UR5?iEoT6<{<wJyDdu$@t5xDm}HA%j^rH=Tq(UeKJ0dq>jc-ul?}SDf5?$-()lU
zWPkO_ZUJ-VevgN?EYNtQE?aZ8%Dp+NR#aSYDxbp|VH?qEoS`k(?y53H7|fR@5{nne
z#jE6hR57bn3F#~>*TiOAGJAH%f|2UMoU+6j8a`<XKPjC6sSAGs=d+QE2pa2luB0DQ
zX{4mQZ+G3#hCk_Bxwn|I^S_+2Wami`<|j4P%PzrC-Xhrv!DXHT{E*93uK{;C$zD{A
zJ9TfuQz~EcX*egJ8|^2B0T^V4zhUUNV9sP*M0H9eTUumWY9utpYU$Gv=rU?nRhAa%
zkY?5XFzKU_zgn~}tkn9o(-h3AvcKH<6X)|@d0n4oLo!nSq|oNniFRo-iJmW$MC)`A
zs@``A=E*9|6BnOU#?Sc8C>q$KWn_<3#%~IcN(v-kULt)kk7yaKy!6!c7)-pb+-%pK
z;xnyY2`1^S5m2#jueT{8UJXS@1gkEvymP|Y{vGf4T(-UN{NjBp>8poiiVw*cKCk*+
z1x4&!(QYLw?7?uuVesf!dCRApJV?97@5(f;WAPt!3NK<@E{hv?_6i=~f`C(lw8bD!
zHpuQn2ID2twfl*pmQ|72&hYljGj92WD*;jn|MU#AebwEKOGevG1NXTf9{Of-@!E*d
z?wSk5awNQ`-SakQv}4!apMcD2cjz&Tv_HcyK%22()HI(?N0_!QKh8aE<osQwuyLrF
zKrBFtrFWM~g@%iVa1?9p)c~~@Y@JZ^){4Mw66ixk9-}pxe@)i^gkHujjBVfko6znF
zsX{c4lU!!+{mg4A1ay^3TSO)x5P=0Nr!>3Wrx*;3@4ncuLW6+!&ftpndU_{YihKYE
zmvnkR8|6=q?RRor$7~05?ZU_kECvg30k(WRv5UF<zEsS-3aY7;|K`MS->VLjXMQim
zK^rHv>qr|pXDvP=#6Z+nW5Lpow)t$7G?xk=0~IHZy80bqK%gb`D)1c#z1;hV`-rzb
z1os@FB8(hihOE(MLcdHB6vvnObqnE9(NeCjU=hJ%C9)l>Gsp_SlYGQ?-~ZF@;r)<Y
zK-Fs>#gY1;>2bDM)VveCG0QE<@_Jt;b(3E=sDBqI1XLgtS?O%n8ETRosdD0F_rVxw
zu0P^`+MFCZKrj_nA8s>BREtl7V;);!E2>OG64Bm5X=e3(#xJ6H2i@$RuI-Tv!EHzN
zT`d9K2G=0(Mi&PIttX#vYw0i~;V%{T6jN6@Y(_Vk{}vEa(PP_Af^oXmj*T7>^vK2}
zcBQDK%FfhEAl;+Zn>#h+FN>qh`+5|{<8v&}z1ibK<gO-FX;nL=bbk-XFXdZo)o}h0
zG>`x&?ktk@R$ND@UsL@OMhDNyfiGIs(<Tg3)5z8g(1GNur-2S+C3fNO3oQla$gBiF
zKWZ;mvF-S>n~zx=m-asPc@UOBa~=Dn>51OkDU|^o$DB~e9b3itqyH!K*XB=(tKIeX
z2WW!xCnIqmz4ygFisxIEy*5QY1+N~RnR^d(NnHkGODsi{9uxpFRh(Q?xV%%iP-U`F
z#)G;FzJ(N1pFEg%sLQGEXgnHGzWl-&L{-M_s2k%guPf#<0jrPir2SZ`_i&y^v|Wy4
z*Ve%~7=O?(7*agv%VHPkU?GzJ3xyIjkX#yX$WX^go%6AOZ{a;CA4Q?v#dujCJ><gW
zPfps-*EPpeMs}D(S3v^nm3{A87X$9d`hH7Q6lwAGZL}&Ia%Hkxj@lpU=dove+Gxvg
zMOpgYSvbs*dRi2z^&H&>5IcpEc=Nz7u`^12zQyQe&W_f~=OKEtb{+YONh+8U?y-6<
zQu;ACb=6hxx9r}`G>CQ2r9mWTZ3Va+99=n0V2#&ZFI<Y$@6lx~?;K|<B+qv*Pd{gt
z;Azx-4eK(9=dPVJ;np^Hec+2Dgm@~B%x}E+bRtUrS5k_A`%-3WIETgbI4TMC$=WLW
zsVhKPFgU)5(R;H+WVa;JU`qxYIb`PuAk<#Ue~Peh$KvQHX}rJwvXR$z`azn6F1^h;
z=Ov`f+<b75MJ;pV(zx+-+F<WqoXD2^3r<bHCslp~ax)9cp3IzHpYq16&M;;}tTh#)
zzV+yi>FM66u4`K5irpU5IAhTzx6%K4uMXbn2BNA5WscHD0p7~`kwtAfaeUO=-yB5^
z37P>u_UV7$tp*4kG~+)m18><#&+#<o@CF5C|H!+vL0{OrQGXIM&l!WIH9oV<!Au4$
zTa}|q$1{V+xpZwc+@`#?$#3eHx(vAX`=`B34H={}i|QzG7LvLwl75)(eKmVH+oLmm
z1F^z3n@i+coET;VQV!XEBMOf!nG%OMRzKyiPi+W&Pxw9ok!jJdE*b180XI_neLjq&
z{5`9N68skj3CpH_%*%N_T{<C7DXh5H+#*c6qHfJZWyzZFFWLed$PZt3UjQ8M`N&SU
zMuv=PhJrGm<>Tk0c$;SPge>{dkAei1VS>-=MK}2yOt?eEqxllxj^Y+AzsL1_%YtK^
zvyit}73PXXe{m)7h`rR6sY6}VTL-baZL5lH4|0vC^Q})7{9A`w9rOk>zi>)4_2{cY
z7B4-w#1DPC;|E-a@!*HJ@r3v3DK?0#+|w82{#kGT#2>@qDEN{QOfKr5oSPIA6hf@-
zR23l!y-lT^co*PY%Q{oc*@7i?2%IFW0D-H4tO)zroxZcw-mY9!;+d<n6?HSF;8H*B
zzn|1oly%UebGbz9KZ)2K!4n~&R*!IFbV_(azU?%+64GPxwh=jp-~Lg%!@xktEvZ#)
z==g0s`MPAM6y)}6&`^?t>E3S)2s<H(q-a7LVt(fK8n~#1G*tp=!ZfJv*Z|nLt4zbt
ziLOpUsCtq)d+<>TF*_ZU8feEMs`Swnyfq+$wk|O&oTkUc8u6RMk%WNWCh&)3pd5EY
zsp~C=sSYbIqg1U}H+E8MvqwMrAX_7co5G2m#mU$ZQ--VW+%mk0)E-HzN=77!a=bFN
zI=mbo^Vo@S6^w80ml;^f)0piO!qh_c-S+wuVU@T|z~X|Ec8-YJlKLw11YT`}<ghbM
zKg9XavF~9ckKjDu`E=)c=qhxwnOTD#nrz-?ZPx$n&68kn9(z4Lb$s&duCWHdKLJLR
zl6ay@zG=Bm;R%56f-SPmL!N%{(kg>$LS|7HaI#}ngGlnzqh)qq-Q~o;Qgx6bu=t9z
zA{+4()n#SW*99AuQ7|&4VYf!KP~~psTwl@ChSNAjRPLiQ-ZB+)A@QMeqRz;rkF3l|
zAQXXSo8h&P-e%@haEoDLXqKsFB93`S&&(fY!o(n3jUuM+8)mV(=A$mZpPP@T`ziON
zA18%cvufQQ9e5nar!@8Y+jDM2?A5zL?t|Iyi7x~Td9%60L=L1fp(Ab7&+b<VfZLqN
zqu6bm2VQ_5ZnqJVu2)IL70OjR*@0)jfnzo`qnJk`AC2}1+I8}bAZkbp85{GN0a`^*
z>_ZU)RjmjO==C9%HWTOdn*;YFA2s^z0qm(1ks!`zJ#bOyH_ORIH_|pQ&R-10i)^v6
zE;Gsx%pf5dmQxjxJ_)Y{oz&$OMbEHs69$$TTp<`p<m|fmdWhB*Yp3d($e8W)O~ETC
zfBHH7C<+r!r%3S?6t?{f&EM&(f7S_zohZTbadaTRt;PKCPd#J<5?JjK5R{e$XMay_
z&dNW&SkV+ZB0!2iDquCid(*^?gP7%geh6<OF@?TEVDDU|9yL8J=oFH*2Yt<OBRZJV
zQqu6z)#=JyA6nnI+N%e*aJD<D$hhZbsTG%vC5!|4sjZDjf-1VFxbx2h#=rvMK!L)~
zL2+vrv&gJq5mXcpupZ$mLT#dRxH)^!9ZC4BW}VM+^+=1{T6KkOEwV^V7^0Pu(QKJ^
z$T1&q^3;CrOK%ZK>|`5h%O%a1hzR1Gu7dVR=g;<@=R!P9Ng<4r>ZCaBse6R7li*g$
zwpTr>CL>ML$%q4a0DP_Q@|QQa6s}KD<%09xzz9F6H|x{vpH)FMu~q6gd>_(;w~Far
zRH}Hym|apv)wZS`6u*=|4;9vVNT<j$Qc1x%+PI0!caJVA<ou#p_^)KTaVZ0IjhiGo
zT~mg|<Pg`48`qG+$fq(3^tlOJ6~VPdw&~!FkYGx6$AVP$$w3|Ah*$_CI<<9$>^I@h
zl#~LWVNeUO3HmSOcbzcANsXKQ#M2=9*q|d35+ipf4YcZb@;pDmQ8+`r7IZ0LB`Oq-
zNP9$scrf2{$F6bThpFSP+Li)FMhc1o&6+R8h^Z%dEZNr=j><6OWzj@4q~K35<ny)k
zD9iBi&hvEl3%IUgl#@V5CUyGiZ&5KcLnWwYP%T{4D8G6=UP&fTFE^Qr+ZQ&E97s+W
zj);DwEo%LXz2&|<CkCUvj-)Ixb}!*A=6s1#5=QNfqRAUBAeAJj%Mo53bp=pRM4wSb
z6h&{6gNlu`BW;^{kzZ{XSrHLFD;jxaHwujv&)RS1ZU9X7Rz?RL|96JO5o8g!9qNgf
zu~l)FPK@wh&vJL0>O^PulI|m0rg!9R3DzH&O>;6g6jJv9Kk0F+PKFBRO5l9t`JbxE
zKNXhm7Pyu<{QrhxX<+8-4?H-6%T8MN=8)&ZAej6+&%afhN+h`k7Fy)n4CKEb%3mDx
zsGUIFB)sp)et=iQp^AZ93(GT38HrYCOi5UcDrHI7ziD^EzE$H6TRYv^Q^ecaOk`|y
zw0)n1x{4-+ehIv5-C#rka-1g~ITJqlLspe-<}8=ocWVykDt~mS?`w?{>nJc}Gsr))
zv3U8@Qjv+mOqVAq6muw?{8zzHvcXTevIu!Zvr1>a%D9BsORh4{EFGXqs&eChNs(Gb
zvWQGJPpslZJ|tb^C*_@<FJB3p<?$3je5l{vko|RVtb}MUTi|L5xIj_lV33Cby1Q^Q
zTBf9#WfiR%e%)tqY$c-<AM(+WetX?ic=h|_UapV*j=D5(F<%*7TGb(IzWZd64=*{z
z4rq1DZ5xW@id7(w`E^6B8q!)NTO~z2{MdpZd;DCuuZKVvgr~#SXv3pLa-DH|qw*tV
zQIFqgxX)m5b$I%`^szaC2TXWJ=Ea=>_HJ@zLMffhFzQ$UOted^-%p}P@U$fQp2ZtY
zEc=YQayfbX4j7R)2<Q2;W9zaFR;SOpP$Ad687&?d0Gw$d{3s8g1|50hZdZzu_5Qe)
zwS&tt4@lfv|2u@Y0wMf6<Oj~zbuk!J#J#lEEV=4>G}v#~_e~o2=}yeVzMp4;^gegq
z!XQrf<i<TS*YD>kRgY2?l&|BPPj1xZeYk2(dqBu2C!4PziTf<y6l(awOFn8ul#Pp?
zQ6V6pT0v7pshE-0YDDU-0@-ttPE2W1_9zoMv-#h;*9+o%nU9>mHJ=R3&L-~pWt&ZB
zi+TIIWq2REkN4?grZx0W#I$P6zI_QeRuTaN`o*aY_gV(H3e0zj8aBOXaV`^Zu2PM;
zT^U|Y=pM!URe@}H>BPyz3nGg9*wz%M7ThA%p5l8gSB_8C=rGK=Z`d|`iwyL+UO`^i
z4xI<;;SA_2q&28RM^#?|Wibq2YsJ2AS@dZXd_jIJ51(CtqQ79Ya+PXVm8q-|s~0q6
zrO7{$pcoJo^Zq$mxvx@3GRa1<XLpg>zQEE|0jfbg3k_F+_G@&c6rdN^!)ooiBUl4g
zcZ}bR=At`Z{ZWDMadKf<bO3I|zrIv-(S5KA=+PPf{^C`^%kK*2U!s>=3qM6++Yfmf
zKaZ-3eu_kw5Ijp8*w15JZcz)YztGWV>2TWCf;Hy#UReJeXf^wGnv>Wj9przTPGa%P
z?e=b@+R)Op<V~D8FKIzgjQ*{(n9d__z<Fq;HMLhuvk4R9VspJkL<E5VspYdZqhX7P
zSF>BX_g~tO($Od98604EOyz2g5;g?eg$8pzumR<v;RT&)$<ChKSy1fGRBraH<4z5h
z6OMor%%j?WJ4<;xhSbNONk4d2OOpJwgDWM3lGmSmhiSt}+p2Q0SEq}SBJfP=CkmPx
z7C_Rj*Qj1VvorFu-_yN3qXEA<wQD&UTl)F*gVJ+l;f+-6JpY6=w+>qOoZHJf2g61k
zQk_Y5NM7gYFxx{kimj|J^|dA+2kjx4o?7oLqvEEtRn)fKyS%-FOtqX3yiwJSig|HY
z8Y$+>{NFG0b}L&m&WfkyM?3Z<@o_$;GSx&@?}V9uN&BQ%A4&X-Qu9>)DxLMVtR$wa
z@f!x{)v-H=GeZRjhtS^r3$YpQqa*jbCpBb52SVijdxzWx<QE0&3>VJP;kMV^CfRE%
za<cYWL5)&GXp9|&Uu66dujoU9U6X@F<Roi>slp5dJ!*Pm19cY$20m$OLIl+g&{lWk
zV08T5=^E98E$(OmcNrT0y_9AddKwpEYAXMpWGqY<J2Zh{-#Lcn>)Y2vu6qo_@1p3w
zqq?rxAss>%woiq>i3N2oNbDToGM%ye%kK+0JPm9;nKgM;3p6>$<S!=vX=0E1RFat3
z>}Ys6I96S3djFYe>Y+CzoESYMeL9y*8nK9mWnioPvJZA^AGZ8l$2q&8=mNd%2T&05
zfjRE(_;;OyI4#_*;NtUDgs>y(-PKVMvOo4RR%l+>nSwikq7#R`-g{g}e^I_0=xo@s
z?4}O%TkO{9v`R^i@>i-|CzQBO3dkb-D%z5kt5)SrCx0=yUHKHbsEPd^(_woA@X-S9
zV#$f|To9Vs0@Vv0fV->{G+24hoyvvh?wC8{^j*++bvLx?g_iEUW74oS4E&o(`lr}>
zr8N61-uv<=?QKDVUBqiTwC{bteEb<hH`fzS_pXvYpddQwsraLFxV-p?mavX?qtXS*
zY1Q`tS6)uWm^%;eZ;#z=D8H{fQuOqh5s1vBHiJkUD?UihZ^(C@sUXY0m_L@fke@MG
z3J&cex3oY>PO916-Lut%YR9k`g&}uHjrqB6A4VES3h#P}oeJ7rwbD)tMjPy+b2?9`
zaY3ZubmL4Fvu)|{^cObQ%r6RPx*oea$Al6<Tu6=A<SIEY7&e@>SBG0loJ+AF+sXWn
zM9Mrmx*^IM7xFaUOha#NL-MnT2OdVCRa$P_{V-8cm?)t*kU9f(WfvXZc1G+?jd)jV
z%s@mi5~u&cu6Q}S=Yit+I+95~ZlGc@ff$kEh?c;YJBDiJ5{VTZ9I;zDvS!!xa)gx&
zvcX1uJ<=3hkm<s0+duuRPx;4#tqr;tx`kB)F4x7TKg(azy*Ok04ehq+X7VV6auOYI
zr|o?n=+dcPZHBg`dd8pB<`z&*^P5^4(nv}eH&JdwY(4yLzUWIBDemJh_zF?3L?rp^
zh;=O21fh036;;echl)A*;0npvu5;Z+bva>pqlhW*o+ns42C*LIbQVjI0mQ~07R!5|
z)vhb_;TPkZv>!CZrwz8_9}|Q)RKz*eM{kj=o~(UEl}6c}-Ny=aa9#>;aObYrPvxex
zRSg_a-|$~a?pk$6ch50!(_7IOePzoeohGU%th4%+v>Ug{&zP!0b7A>yqhy?4A5Ka$
zUPKtO@u?JQf&AB(vE{8(lBr{ft-m&ezrGVjigsN&uTNsa{iXsXSJKaX#T~|1$88Ms
zPQ2HApG&H_PX#Dc(xWj4DPng}7l)NI&^yM`4y$|e3$*tl+S%DL9%=F!x%|1Ity7=o
zHL6`^12wgC`-_$N>J=@*9T4*g)Xkc;p8Y5v@L!BqM(OApEE9;SQ0|t`p%jL~b0d>y
z@ju+YOX0wH^m`t;H<GV9Sck?6|Ew?Wy})?|Y_V_gC301e6L+tbW_>iKE}{?W(>yRo
z9ACXhrXjBPWOw~3aNg>vJ)j_bq6_EQX9vBjd(52!m44;wcd6q5meCxIyB?8I8(-aq
z>8=U;R*c(fSy}FUFOJXLL*ojES2AOv(T)Rh7Hi}5Z>vNob?P_1&Av!zh~{uk2x$mt
zzrgo#=F4fIHg)DpNbskqf7{6QrcVOLwdP}T1-|6KoTq%F3xk#est7*13YvcCj)c!O
zzlkA>e|!+Z)pEMLR@OO45tcjq7|ItSQc$h}I)kB@D9$IVYvY*HYejun6t7^>`U3{&
zGa$7}jdZ{d8ZO-Ud?ZHU0Mm1`@g$g-6?-0oVMohw#=DZj6PF&1Jt*h$Bx_XOm!Wg^
z;58FrH=ho#;j`Cgf-AJ9f%sTV*?NYt)sZX9FiH$cVxLz932M_N&*GAw)aeNiyHUkc
zI<trC#YUKJWj3^RrLCcNbf;&RYaaN4y3t#`5&??wsg2>hL5jPB6mO|@gQBBT2Ph&;
z2i%oo`0DEW>z~j}_p=vensDn2QM`O&@TE(~^ulo|6B(3=?9i`frKp893JOr%lfM+Y
zRq`ra#9gWar0iY!ScbbxG$}f}FIJ_i()){*>pD!6@6i=oQ$)UvRvQfM94_<85t%$i
zFF`K5A5r%$b0cz_SNBkf+rdzM$NPh#pyA?|24itLuC?5}IFS#V`Xa?lvG*^TR`c%n
zCx4SXCO%Mi{d;j2N3Z+@1x=xDo3+K;+WNz@FJHeeC~v?waVGs}3s9<0^n>uPmb0K&
ztN2~R6dMcN)x7TVcMqvgNYZgEIhu$Qxt-~qMR!t8IKGY2>8n`3=p!XQy#7$7vRUDJ
zlF=>76xf0YN>25IQ2c&}+5reO(NW1-6GMk3<BM1=EtM;7=jf3AyU*RS+{NenXZv)f
zR)yD7XH4waud+HC8T^HKA&rfQ+LJ?~?gjmr(XQ6Yi@cQ6jNp@@Z1lt1&Q$&xXFW1v
za(lmjU%`X@eQxcIB5F>{re<RJ)(^dj{2Py1#x#g19(OMkH60#aP1`Qcy=zqlr1`g}
zVg)lUUz&$Lk<9$iHlp3T=yLyM^0scludyLO@ah|TD(r;ws(Z=3)Gf#0+LEHRZXKBs
zys)J&OxfC8a6WQQ=Rb57PPXZCq~X`IV!ufKjIO};JB}!ppa05<_|@yzvm5Nc>KOOl
zUZI&i)o?ZV*WLnl6JwHFPzMJRt*9#I|3rN+@aPjAw_ok&`HOS;DY!(x<ianBkRicm
z`2H%v1Nas!XS<2Ru0rb>Fl8<l#}f!NlKeGj&^eYr>qPU0&;^rjk_YP%3&;v}O1)1+
zmuByTaH`d(JY10<H(A6{#|*R+EAlAWq~yzQl)I^+K8hvbF!5lG+rVVM(c_bciU@Ak
z0eU)lc$c$#uZ_^GgA8sA$y<gX*@)mtf+c)>G|sg&F2k{Q{?&rM+8HPMDsRxp<L%v8
z$q_D!u16zTiyhCWmBs7LTOm^G#;*buz~MIC(l`|N#pPov1GIopy%m@{t#rq7eZWTW
zm;-{Ylg?0IPz{%-aXBs5ORZi&Ng<y|s04~DoEQ0%kz@{ElXMp@mVJ;;PmJ51t%7bq
zL}k7!F0j%cDZ2()VrjgBRixFAt%GBo3jaRm7Jag*ipeWX21ZkId|3X{xJvy!X+Hv$
zb&|+-lS0_*C{4Rh+?~0<)B>w@=()l^tg;nOOqh1R!CZ=nCTZcj%aCds12iZ#Cc}6P
zyuPW>@=+gZl_9z3vYF%x??PM2EqP*EaRm?266mGfx(Q_gjsy3Egf)R>1&2|+(;Krc
z01{N>bI0}Pp^fbNh0IS`(VU;v9NjjU{))nS$B)+@_YLIotYG0gaEg5C%4Wks-$E6!
z%<b-h^EWZQgmcFw4(mWwaEh5$&~{rP7_)JQ4mt4LT=6BwoCI7ZrLr~dY@;^*%RMV+
zYd&%Ybkijuw7PR1f#x~EP`^5^vtjY>L<rK37cmMmoegrJ5fFFsKide#dOd%bj(=*&
z6q^)5>`83ODd!RUEA6}OqvhAv<wU9ZtK1x>J~EO)I&U3WBU?s4SKme>B?(+oGkzZ$
ze33QKNQ?dnq~vnBc9Vy|O(clRPt@3N!*($%F&?>f^D-ekkc4PlBkp<8gpt4R%*oii
z1D)j(ee1)0DKdZR!g!sT)eLv`I&u4pot`~yB=1;gaFfi4(c+TU>7kr;<Iwr@#@6OT
zq5a`It?%*|NDn5edH&pf{&NR9#ZwR#KHXM$L=K({^=M_d^qp{geFaPn9LBn;Hj49z
z5R!iSF?Q^bhZo<Y0?^w@$6udnK;_Ik$+B8zSqw(xo1R@TBs}P4?P#oOd$`OTsH{u#
zZQrL<HwBMEkqJ+8dhXq~BqGRdQ7lYV1RfD*HWQ~3dIvNAmW%)G)U1)6?V6U<U>~Nk
zew-}<o<!OF&Q_lQi^gL<9UZ9Aw4(UCh}F}a3y|xu*RktM`?%{SHyv?W;ivHaX>!ea
zz0QG=5T{ko?ydTSJ3}1Da+rdOX410VINgy~H+dy9WksbepPc9V^NrIVZCSm7Rt9}x
zS^(;u-GWane3rWNr16X*iaElN<XAa&qs{D#+Bc65tXw}2rQ_gCV=>bAeLEfUP;Cp6
zy@?bt*H@k$aXnq7n6eb8&LXG8nbG}3FLof+_?mkW*|`d<7Lqj^6A0Y@EM#iwWoJ}7
zhGYJkAkTI<&I4i_)xflH(mzsQ;3%;g@bn8^KTm}%MPGMkRlQs{QrI;=VPw13KmUA6
z%N#aRJTi2MEcvo9QQ2B{wow@0AwEc&oZ((Mc6Bgu;OyM;boAP?6&cUB5wD1wK93dc
zjOLqO`&B2g8`{~N;gs*aZQ{8UN^ysdouneU9z@Ya@|GN^G1nvvWQKoy;|M*eLGcNa
zsrLddXdROE+_+xSn5oDccgj=gl)sL<kG)1G&Mcexf`tY_gU%!Fw<Pl}5ckhSB$|J8
zIs+x1M)m0}`)CU0u+OiMa6<7Ls&rzhx5M0Nr$Mth==!p}hKzwQT$J9wu`P&L{QN*q
zhKeVP?C%-&=T}N`OlBGyao3W>f8Wu^6mD4i)N6F2-?eKb|2s|$lR%5TlqruxwLESF
z+PAN7*qsG~jImJjHGLX)jDbd0J><o2P;dVKID6~3sJ6BZSP2;!g&_y&5S1EI=`Ixk
z1qlU|5J6%D>24TFgOmmd1u2Okq(dnMksKOEK)Ty+ZBIPsJ@0wW_r2dA$KQw?nAv-;
zweI`6ugF%u9f&;)q|zu_n{FGy+E3pN3|MVVJfUk<tP%F3_t3SojR7P!?kM^O&(+)z
z`}n>%GJ+GYhN^tM8PIyW(6d=VLO-HFcvF@6l!BPPsyL$8#EB8vfmtc)9Yq;$9RV#w
z83?v0Z+Rz3TUUG>83$LwXiKb~B<hdf!;r4APE~>xV7P$>Ve8INCb-hMg+}>>m$gro
zHS=pXj{8uE9gtuLKlwNTCg#J!^0o&AI4Sc)?70pj|F|m98azosR|8#}8mG*pr+>vS
zOb8C;A~{uXg3zRXvUUy7VSZpOF9*HPZTs!}`fkXX`E<$U*getDf(UbqfC)S|$Lv6A
zp%w*Qem^di37`rf<?#Yjt@O2uj<<OQ!{>+9`ZJ;=m&FcxqP!pQ*Q_e&Pj?$cUwDr>
zM*O6(0B#9S4>?E`sDoQ70m1i@;V_H)Tsj+Lv*{BVR{TMKMXfm)eK=tG?zzpzC<?TW
z>*-S{VM+n;7HGK&<&By}ieK>Y&~H{YRH0|EEq%TOs_gPsdTWuVbQI}EK$8QdI0v;D
zrs!DIbDJ``)l&-4cAbVAKy00|9JZNKO~f48Q3q)FUSBI6<vv{K!1eLz`Tz>yl%sgL
zehbfCLcD^}tlYEEm|Ebg{AGL8<^G6O-aGo(T~{CXrLqCj;|~jO^K8~<K25%*cfs!Y
z>g+)2CbailCylE?2J7W~FSAKXWR-qp^C+Kf;j)}e@sp>Q?7;s?w#GQ};1jc&r#`<5
zFGk04{TZ^Jlbjx~t5i3WtCy>IYHq(i_{48t^jbr_E&QDJuxu3{(rXoDSCh+M$fyIS
z&FfnvcB)@?fs8q@>~r+ReIb@6V{yTiTgCtdGqHar0_sq`5g;pR+;?wha8j22F8=yy
zeO&Dzm$6rLdBo*jK)2*=&Q6~C@#=p-S(y3>4%W$ui9D*=tA8nZj7ce=@a*Z3qnUy@
zW#W938%bh<wFR<(Io&kXLfsz8%<*gg7a)dN0)NkEITIXq6(N$!L?@4Cmpw^oRih}O
zr$mNAE<dQ+xmCVd8N?6Uz&;`B_)5PhJf~%^2D<mlN=lgbx?muf2*%q>K)7M+PRqUs
zS#PkNxC=mFg)gn+%KF>WD_MkOV)f2IJ%do4d=bCq5_{+2qw4{?m_5YK^BgF`HCT*n
zjnb5lw>N@Oyq8RBwVrm8^|Z;*GtXp)ZgIBA#-lZ$mY7G7*Me0a93N@Hy4vVYZ{{<a
z9x1g-G5pzk`~64yd6Vb1Ky*`)G5Dl~Mo)w&xy5&ovZ8W+<@N%t9~*AN&jHrlZYVi$
zi;M`PzcqXAFgZLea$m0Ks(}J;2*L>@^hTBYmDQS_`?Lb7dk$^YWr}64GRWyay)4(9
zHbY~7K6Z7VW8(S&h^H+P?iS-17F`OJ;z;uV0jp4GafEY(9c@Rbn`rF})6NJ;Nq)>O
zLo2v;)mNvBx~Lp8NcMb=0M3ipy4^3Uby#adF*N4$^N6k1$bCL5;)+B69Sdd3YhoCE
z2e1Y%Pf;SsKrVDKTg@mS3@h@|ArI?FSUkUUgCN?$NPwX%Y(PpFwnHyqdN{3keZd@<
zn-l}Bk3{>+WxkQ#3n@OwdsrnVhp1mLz<>7D`Y-I@Y5DS&e^&P|PfMB#0q?`LOo<U{
zyW%GDIsJu?28ve)%Ag7bY(c|BBHQFk`#gANHo`_yrSqf^m?cp5OMF3}!ONSH<r{y-
z<R0Y~Td87mw321BnnIx?`3SkpZhcX<Gq@!om%?{pXDCiV$>@+iAxOtLf1(R~6t?q9
za=K?`PODkKn&>j^DLTgWVcj38V#$izbxXzKdndi~8OQ9~n1X{uh}2<~2v&9pMa0ut
z86`_#xe%=pWBW}CJS&8dozixEP*&_+Zr)ZJ4t7h$!-RZV+QJ+&-cFO#0*`f2MrH<O
zjIaH6H$cR@Q#6Sm-IkxZM2h-|Tz=@KC(o5>i-9BVbA!7C1d6V#+{)C~jK?&V{6%s#
ziM)<zig+<J()rUpG8E|=!d)V4rOK{t<3Bri-`{AWHw8w*2zg@}1dC`l*E7^|lJ;Dm
ztX9$(n&Q|rMv$1!gzuSeJcV^A5l*lWoLa>`U=0ibSp)NdflaY;-z~;npG}A6lxDVZ
z{fNM`ZrMz13$Kl|DmKe%FhW_+$EO+@)p9nKn?sWg?|xtTTKo5v&jzjhH}F}dzuZe>
zQmXn!R$Ym8xGjfDI7a!*9llqMG3md~g?eiCv|=Apzkajd4%GeR6AttWsAf4X*5F2e
zebVLzE7lQegex00_o-%oeY|p!$}3DY;+Sn%Q7({Es5b$GeM}jy9Zv~9dX`F!^#Ss%
zVi1s9Yr@25)bZHuls!c)n-x3j4ffC1JmaIrSq^m8eX*3qhMFfS%q7=2oP6a9wbL{J
zWk2#lYs)S`tu(`~L^r4DvnQDZpHpyb+XzD!ZtP@v;Y7_w$$$8)rk~AT=U}~ocGf;T
z9&9AYbN5tIsGdqV9F+y&`sKj#mhbFIH*5urGP~E!v&X-XYtPyKyt*c7I~n`)%ky!b
z7nj6*B5I!pChkd1x4)_apXtp^LJ38x<F$Yid(GN$+LbPrm#555`OgBuRq3RPIE(7R
z??CNuMKLQuK%#G8Kmc&qlKfQ^|5^~goy>pxlWv{J&J7YZg=~rAK5+kM#r@^jKEn&a
z!LfBIsQ+1e|MAQI{M}72PUw!=X?lRg9g;UqOsmgonAqNOSd@>Zx&D`1{TqH}jG+Ln
zEk-B4<X3t3=MVYcF9M#QK&ye%$qf6Lla<kPXVk@Zs`=?QB*lo7TEEI5?Y?Q;S+)L0
zhy2$cpv?r{dQKI$p5Naf?*AZozrC9@Zvp_lM@BG7_77$Y071q`?yXFlZ#sHqEwl!m
zpgDZyc+KFSuBG3}^WbSd(KwL#|MDA*v55M{36Pz<1ag$EsUJ0CD1xo0zz=?7uOa!@
z_x$H;LQT2G2ERhW+27vc|NYNwI_T5WSxNu(i2c9(m{Td>F(SNA((d=i60S#_Uj>Sj
zAz;d412kiWUBH7^v52U>1cm`|3Ted;FZI1xzeK=(6X~bQ;?-woFYdlMSM}ny#Gjw)
zA79wQu8Iww3fK5^;B^K?094PlkS7k8cHX=4SD@;1SN}fDWVZrJBt%%xo*4M~rCkC|
zXZEx|45tG38*r4|DF@QRAyoBZZ=MJBqd5eSAd)Nst9SH+-PI=YieDEypNS$~i%6vZ
z23N0pBy2Oe9_(vqD3Wrz71AyB18rh!vGcx$aaN3O33M1hVc(la4@cSl_38g|k${qH
z`*a-pe?9`WAC;z5@gw_Vy}ErJ%r@Y@<=!)&2k>kWlxudpRp_I9C_3iabU`zPQI}9P
zoUSKNWc_)h?6K<sP*d4Xw?@1N$ztq2*AAL|>qlM2ZKvC!DnP0F^@94V2cSI7?-Fcd
zY)X3&Pz}?&=avPDUm=GzQ#*^&5Zt#rWD%F8>-)_q<~aO75!5;6eXzv5(l|AWtX&O?
zTmvOOLp8XA9Nj>X_81slgz>%a6k=NGY3U}C0hy~Kpv{o=0CqZ(s*J&U_dVB?J@5$@
zkXbQHaF+J-zptOaJXNPSpzXr!-6hRGZ#5s0{te~kn8vu;P0kUZcwKt^Qsa7#1e)Ru
z`?0CuD?6~xhCW?PwfPBRv~56y-=gc|p%&mLEgBerA%W^rTIKJ-0)BOev;4t>)Oi8=
zS)SmH*aGWo#zs#1!#h8L66*l~Gu`#qm88UOD~`^7(HFjL;J+?ScY1;Vsn!4#orMfT
z2OzTu4J~sJ8De;4sr2iW?iJ92wAMNM?Sw#ju`w`8!jypUoFPccI_eqzW%-=;^xkJr
zys%30s$W|lt?RfeBn*Y}Ifp|F)<gfYP5y1~JcEEP1FVMm&+YAI#EJfKsIR&QCkfMp
z0??L#4%B_1u>st^7H^>H8}f&vEaALGa)NG5<WvL@rv;+QQIcu+?KdZj9nHV8E-3B-
zj{vDuu@Cbt7jcpPO-*U8Z*SsG(1uZvX4`?QGBTf&iUx}%&ufN*<ywv+HAbMw2)j~y
zU;{q2NHYdf&Yn>#0}I%*dQM_ojo_DJ7m^{~o?mwhVR9B2{tjA|QMJ3F7yj*u{|1(<
zenjdr?H`8=9V-DLA0uO$>5pRVc(8A1lkKt`uaX!5iG6U5ogFsXto>U}%6S4pwX_lP
z(?w$8Job$su8V0(5l6#=k*gpoANTg~0Q#l#jEl5s%YW;(|9WoC34=>}5Pn1S_e);?
zg~OMamv`a3OPzje0_dX~Uv=CN?s@<|biuo-g|`v@6h8qNX2tlpda3+Mm&n&aASyh!
z?41hg|DvJ*$x3LCzcOsekqFLn`n3*cP#`4O9qfbFAjqH&V$0%_UG5!BA?qx8@{9bv
z<`h}DtKThE0&L4P0`}wnUJ<AchX8{wQTAKOPf)8}Hsmmif$3qOCUq1%$J@XRLElvT
z$;`)zK?ZB(nMtsPMtJ-@{4N{P(!?37&yK5sjQy^dGzJ#+E2@B8-^&L;#69vN$4|!Z
zGMk~Y%-+n)Z6I6?QLY1h<!70_PnyHuWp>0zOvCsVr&)fwKDuHO!eBzcOJVh;w)n)}
zLSCT?J%kHD!lTnXH?y7|j$d2AgKp%}%yg8`q1p7k1xMiNP<%{D21$8?B_9xMDM}yV
zMeB<1AFs#twB?!&llJpMD{iC_&VT45e*9BZbky<Xa|?p4S}>_T{E75h0Xb+Q`$nXt
zeC^TXP9^6@eSRK{Z-bq-&B6Yx#=(R92CD(UJEAvljadPoc=Y3`<+S^tU$#?o1DdBm
zhMp1`<<xQ4=?F7kE8L<aPK};RfZ1z7JN0^qiOwbz$=jqZZq^0G&SUT4*nFTl*Adtn
zBLrW%Hzw~vpT+?6pG!cdtuXJZQ^vUG=_uJh^VK-D=BubB3HGS@75((KmKq<;(pwwc
zJC7P}{F}lyY(P*OsQ#Fjl8XcO%{w!H`T&WoW#W0%!W)lewuxX`c!o`-O4Vj1)Zvu_
z!SpcTVp?>n@70SXR~##kJiuN2()z{)Uk1^=fEZia;89l0dWGorxu4g<+%|y{GV=S^
z$HgU;fKjR{`$!43*mo;3`oaBil3uSSw~=>t&zx7GcACK>e<mj4s~dW+w&mb6(C=A*
zZNR+KqLZ-3I)eUk`*Ph+@9NoDgA614nl@$+M;n$si%lo6Baf4-J?$j!Oa-ee#Ek4A
znGtZ|G)>7KrJQ2`SjU~t%9ihr!u_vB=o}J4HrU4XHok`)ICYyi6<@*^=a}K`@eYmT
zKD$k>pBW{cZRTxjgoRVLso(agYCif=^~X8?zxTEJRj379eXALx@_Q|J7sg624(wBG
z02FWq>}G8WmkIc2F_7Xrc!J!%7hH)kEYih8zgIrrH+S`sA%FZIUI3sM4Q=TJpNhAc
z;&lLQ3hZfT8d@NtE})!VozD<ovpeYuPK6+2#M;>%ur~qIp<C-k{dcXQn5l>lX9?LD
zp%!Wh_|RECpt#is`KLhIcZTon&P8(NfQ1BlbCZ+o#o*}U-Tf^7J!u;}tbs)&lb^sj
zXaR{kJ6=3>fM!gK;DIHVr<yuPyMAZ!;bd~l41^Ih>%dF^v3UFH4_`tu%fvA`qut}3
zI+io~xZsoQ4baj(#nb*C=m!|=5y>=7;3PH!F?Y#qjcKF5{1+gY&g320$sa(kVhh3?
z%7GiihBIiF%V=0BLnM^zF}}1XDcPIBPR$c&Lc-*D6E?ZvtaAu9;?K@N#hUAYGiTpb
z42QUWDs^&F*Ti!bmF!m=NGWV#Cm)kyhgH6;Rn#5v;mmro*C)Y582+5{yRp#2PP!47
zaj%VA`KZ|Q?fUYvDsqIH!@{!@Ql^}(=Y}QNYZxBa6h9|#7J|uBFtMj7-Ds&ZO_Sy*
zF**E5(wYSxE?L)}3y9ELJ6x>YL88;lX7o$p+gh}_9_t%Gv_@jRE@;|}$qhM=f6WX@
zk7GvwLnhU@!c1M}?C_~AwL5R%yMoNZx?fIFq<v`(^UCjD<4#RQZ-43l)*JNf8m$to
zUi&AO&as`(kSyS^<A%dF5rn<vxgN(otygz|qL0k)k{P*SJk_~GP~j}nC2G5mA1%9>
z)l9|<0<P3nK>c^HE1$K$uoSo!SndWaLW9x-M?c4wy8w;q2-Kn;`R#UkmFl(o*(FsH
z-O^2ne;@heSrmGV1p62pf({Jm+hGsmrL-k|J^)PcheTxJ1Uq*4T4RPDCh|4=ZOV`J
z#FHRugrmOd&R5`?LYOdvl{wvV)iYlH&Yx=3WWaV-Wf|8g8=eR|tqD@d;#I+@8j#FQ
zHQK*tjuc<W^0U5&7jAv1u#ktT*eJ-<K0GWh5YI=^?$vBLvj_@q)gOK|{B$7L*i}-`
zSPRHnF-BHNd3RYN0!x2Gw*FUB1r<>OItb7!z0G|{qxEa;8C!AsW?jfPW7Ci;^|cEL
zvZTdO*nkSrYw{tv+5=vZY~N=m1!-cNYB!Qs0xvyv0LoYg%;AgrD97R#{>G=%$wnOc
zvLNUv#Fe;13gn3@>MtpX1wIQ9DXs~|K(XIv(&=;`x>Erd8FL`k{DSM5h0=M5{3W4N
zWllk7eFp~DXI0Jh-zozuS85?35pkAPPO1J0&}anKzmKP*-71PcRRLHHV!UY1SWG~-
z*mO9e&R*6{aUxKMqxozAAM5S<x0JN22)SM~LYS@7uLq#r3P~|IIs8V{mdNl0UOU^c
zG#DE>Z4*Mgp2398U#Ho@j{g*qDK_>zIV6~eP#HF7vq|P#l<a5o+1Oi~*QpeQh&-K$
z7U+r@jzTQHYQ9lT31U5K&o?V4<%gMoWWiD_v_Kx~!~vg)s?ky*Qgh=;S8BXYZyV&x
z37{Gb)p|M@d8vNz`e86mnmz^e@>KncAE^{>5M9FDzeLhMnk&p(mL+G@cp^_NoMNW<
z22ABHFv>6ubJ9*xA6G&IuhLGMj`?#pUU${5kUd_D%0#~;u8IyRH8+&l9yC+LVu%F-
zrTd7dREb<LcevpB#ke?fosQ?4dA}YV{TIhktD7L(T_yZ8M%<)+@H5t<>8xmAOfO{V
zbS+7Hz%^#QcPX}pBRhcj$X_B$osIFeOwnfW1YD5DHm|dnWhwMZ+n~R)@Iq8cD&QBh
z8L9?0d11EZzGLu1LfNn7SlIZ9T6K;aC@Gi;EX0I^*`&_PCfbx-1FrQ)n=Z{E;h&)@
z8si;bhxqnFb|KUnlZ_Cj<qLVC!8FSn0Tz*#c<@F!!z8P&PjPsdvRwWc4KdcdMr&Nb
z@}pSLYK+A9d%Y7<)3Od{r!Q|n@*7i#DTO&$pT2u%-%DJzT51_Pow3w|-!=Agq1q>K
zpmMmB0kl*^o}EJ&(npGJC(UAtFU%rON^hJyihj1wZ5vxztetB{mKUa+<LO-0dO#fh
ziG*DpPdAp_L9nt`n&_M!xo;U)-u-9W^|vw{{1tEsj;CM$i73_!a8AH63H$^WV2<R0
zpGuC^s=9g6!A}E^rZgz9Z$v&8_91TfoD|zo?(7T_=3B#&>fzzZ2?jx*4CMwKM}wFW
z_@6vnmJr!c?&;*K5;59Qp6Wzi6|N9@S$UIVBWO4KM=}{s!Vuq&mr$N}?FzbZ>&pH3
zWxB$zEwy+P(94EeRMm4kz_rw`RX^i$_@QGz%9nlFNR6bnD7EriQD+ce0^Fi^ez58z
zuLGL`e(L?zCgN1}0g{oEIIm)<OkB%EV|Hf{Istw%ERMdd^x|+ff0_l#x9gIR3>I_I
z;rShWLFq;g!vQ)-(BYN}-aovv#VKNA5sT$^*!h%TV9<`^)|U#Y*;ciKYd^1rhYBHY
zly6cXOUzN8?Km%o^4+O03zV_dr{HR#Zk&}x$KK)T{UZ3(dJg@8&Yc&};4d@DtApBl
zrLv!3!N*{;o*vp`s*s@Lg~$?z^Y>P&?f92-SW4lj=ud`+y~|IMEDRlsP1G1$d62WM
z!oyp)&~NWX#*!R7tqs4t$<^wJWHDb>5zUmP34q54SM<eHUb^TYXtef(QIDzcI@MCs
zHRQHRrIA=nrE&GF22#`<Wz%kG8B@8{94naS8W7f$u}nv0xEfb!C*;uch=blM>B{XZ
z7qYIRb~_AbVk&jMbn-E^*5DNmv7Lu1EWx%i*m?nnis=M+vM{pqVQHqPF|VUqc>-KV
z7`fJljjy|h1C&>tagxwII*qo`8uFV1H6z$EyiBHDo_|7^v2bfu-<V-1^M>Tg4RYzt
zn;h!HXB?PTjGa!on5POdSL;i*^5Q>u+J%KC7`%yaxtem{rlR?sb^a&WxTD&B)o<eY
zWV9_pjvj|)g<gCPqKihOWB0dubRIgs{Fl<}8C$(x*Dw6~*YVI4eZmyp%%R3tb{mwY
z6)~hdxGh^)VLRgdWSW6~JK|9(T&iu9#$x&IJ1$ZMg+MMCOfm>GlY&H#F|zIq`|Pi|
z)ZQ1uznVYVyi5+W$6db@ocoTqevl7U4ik79_Uv@}N&kwB6T(?7jkN|@@aGN2NV06H
z_vXb6J!Fl2=TA7o{n_Z-+d5V*ru!OLx4*Ev=#%WFL|$5FD(4w;;_JJe7}91f&*lJN
zS$%(XQFD@2p<KgMl!yt<#ZSa&o>5g57%*BCNMlR27HjaX9rxsR<3LFkBU(q;f#)&b
zfY>b^JH0qNfmHE!6`w4_P5%8$?DVZs&goX-?Y&DK`ikRr*3JnAmDZBFt7rBTtMO<3
z(6>aIx45TicJxleM!AHij4FEjuW+?0v`STwCREaVEMLPu13RwcXs@N}l_g3AvAO<J
z!u!{&?Q2~5{U29cM=n*n@-gMebniy_)@aZDy8jZhk8<z-z<c>tw!cd%$LxAE5lphK
zm_Mpj+u;RvMCZUTo8C4_FekZQbaZ6koq<RvuF)7D8$L1?>WW}9lbkd*GE)~M^m-S?
zy+SBe{o7qPR2CZqx7+rP<1d!e|K6*huhf5GWz}mr=E454Wu4I@K&ik)T~rzEm9S||
zOetZ4o$ojk46=pqzb=PZu7($;{0r#p*nmeC!w4{2vH|BX?>)2x^AFUv2Jr8;o5xg3
z5-i^ueW8ZcTKYu12If9YXkUBDP<UHPqa2nS1sWqaJ9#;5N{g*kwcL=&9Y?M(gGacE
zvFYVAL*;O+aH}~#Dtk0TzuJ<}dyH5g-~XlY@+Fk9y@guZAy%)oYy<F2sS8{?gu*-&
zD(aEw0zeOHD!DWn+(Ud4-`Je}!)44Zr}FAI-~)Hv!T$EwvO~YR?h&Z7(lfMIepe8>
z@wq-o6s_mgAZw_AFPz<ZDhxom5Rcgm>qVylBhhw!N1dd)I-_cK4VqY7lz{`XA>M%D
zOk7ElHPR4ciV8gK($XS^wP=ng4&!Kwa9MNLY(x2z>9t}{*naP&iYwkMG)?@N69G5G
z?M`VVIA1qxRmBBIg1H?;E4EvC3ZE#sS;lDoXxls=q8Oh=)}TH{*4~c2Cn4&6xJ#Eh
zdi|anj(b|?dff=0(^I&KVbwH!nYLYC#iT!7Mcz0tKpIiW@@i~(S2;w=laD|}ji{g<
z{3^e`^H4F++%QbcwI3!n4z|8!*pLE-gUK1eVNS!c=F8L-BTsI7vCPHqv3d8j#82L<
z5-qU*afuuC-N@An{iuXX952#Z;CUoWKAFNGI)YBF<noRo92R^uox-rwyhKp&eYDXE
zOL2i^@8*F1OU;-6W0R=?uGk?S+mOA?=WCmKUYp_^RevL}{sp*GGQJ8y@>g}@g#N&+
zye>}kzv?tfrZ>Gz#D^B=Y1Rk!{pD=u?RGHEz~JXqwSw(6snaBn0Rs>rQ>yea`6~Rw
zGX5*>$jDH>KA6>fbFQZ{ezqER>L>8QYxaL*@RTPiT5<%*roLR>if(^QlP-6E!xurq
z3|f0j4wjZ=X99lO<vaBjf<vQ&G>@$rwF@zdB=7B>U74aKUIGxIh>JbbEZb-F<<4C4
z;Y<G772Expc17AFHWw+Y!NH{n`_Z4F;5q24S0aqYVdq%T_Fr)B7T3b9au8(ft#f@B
zdZ`e6wzYeB^$f~_Z=8&a!JMC6T#DH!YXPp(^{&$XZQf0={e;*do{MBKEsPiq2ggHn
z2>6<@Rn)S(^G!I(4p7ro?=R$Ko>5NMEH*{g9XUIM#pvp~=(OfdIqbV9wMya<t-%_8
zZsGe+^y=~Gi4hfbe@VQOi~QJ?dpYn;d8JLXeUWx)>*ZF7iq><cU!4`csyrB%=lvvx
zk*hJv)rtf=#?rblWB0(C%oikCfV|k%ano?c+)=JC=uraWH+Dq@#o4=u;5eZqbCwMG
z){vnejE>5KksRd7RPlxV8+1m;P81w$qBQ$k9|3ND8uv+Lv^m7zw=;)es=yzn6*V*{
zgHa*l)-BVk1)-zdI%`@T#r!A>G#Bascx;SfA?H6X$mp=~EKh;H#|HmOM&a|DpCZGq
zl>lV>_007ZjjeAq_wKy)>LM(V1@3)&C9mnfXQz7aFAkbT*)rxZEp*=J7e*3gX=jbh
z6qJb;gxXUi{nUG9ESrP_gX8xAa^rMpu)>~c;S@ko*dNX7_GUy>iCx59D);bwqCThK
z=X^`3LN0dl6qe3z<g6M}jfTIBai#9(F-3)}5l!`pvKDi5K@PAlG3CLzL*q)S3~5}N
zUUa@LpZBkdt=XDY(m~d}S#o)DfGbdcrdjbMfe(0z^{5sCO2i_B)`usQc7XRSZrC+7
zoLuezAn8)RyWgHfbK>HPTi?Yz1S871xhYZUX%He1wH{r`^G@hq)@Jo+l}`P~ID?L*
zZ#DAC)vKk`({qx2Mi&R#n5b66Hot*>2%BRW@dM2zvd?E`^q8Q)%7}GL(28kx4)2v_
zmL0s?A~8m??_aN6IRvl}fXT!6N&qhSe!d27mO0wL^frE{78KP&97h^{dqLg-dE2JB
z*7;o7BLn-n>oWE4e;D85Rjg>!F$5=}m!kpg=*9nz;0lPqCQ?-X+zpK5Dec1i5)4q>
zd`w|2%j{R04PY<Ym1M7p&CxCk1*$Fi8+d%ez13<pb(LY=xBKCjP%YGzT3}wsukfmb
zWZTK4iZ`MbpB{=JIZ2uk16%_o9RRUr=rqLypl>sT9ZH`D>Y}@Nl=u6sf%~$@Y`XLB
zEZ=WVegMD*uEjz{M)K)eE#Mw(C)zzLUo1E$T%pk~IA8~CT(SZza%!A!ybcM2s2q6R
zNeR-Pxjf7Rkn8g7(y2fu@EAT^bDWNMSC+eT;VP$=PEQMjV0QAEKRj5fxIdjRx_Fxg
z$2pl+y@fv51zdxw+c(<JqSJ*Gu^mS+s}2HN(;_$t1@fh;ONF6sV5|5r4LgmC$@uwQ
z4T`tndR!6cie+{Qd?b_eV7jV__T$uBArl3Kjq-HWm&a`YV}sW`>9gd<YXjphcS?+t
zLFC&N!o8YoFFXZNVzk4h5a@<#9XGHtN`c671!LT!B)g8fA)vvfpT3sxYaxFuDBQZ`
zM3xtzTzqDfS&xEj`~S<m1FT1pg=>}nVLjTEe7Z2C!eHBHf9vN9HPlmO6PNMf_3-nI
zO4KvUUrU7YZZCsP+!DqO(r*f~MQ}7W_yVSBiKz^YNubR6h#4RGqStdzaI_Y61G(}Y
zTj-X{-7G%;`lV?)%dJt+G=GccBx2Op4iL>qPYDbFr?~~@JElX{#eSeTSwuR&@uaDa
zk_T^J1sqln@+j84smR!0D+)Xfy5TZ8ieeaglaoTPz`VPeV~1ddBqum^G=8}%aXsM6
zdbGC2Q+H*Xgo?QYc64zI?GONHwC?)o(<(ZC6y8@8!&t(?)B%d_qB8143K^&A_i?}9
z-UTZ}$v_G~0g{|O+w_e*3k$%$(Ebsl%Nxd}x1FS%qF+d4wu__6XwFtKEY0ajWxi}_
z0G!?q&-<<P!wVqKDcTDFyepS>Rd_Y3pjL2*<Ne!doktnqnM@o8U<!v1RGPF^V-BFG
z<#%ZkqquhysL{{m;O;LsnDP_sgE>iZQ?HyXWbE}G3?9TifYe%T`t{4v*vy3kId|{j
zb#GrwHvnXIL35u)fB`yGiOiHc)sSyyeF6JA#_7IGPa-sVOQ`03{7J^FvN|P7Fh;QO
zhM4|<zP6|9{U4cc?TUUY9v4vqO<FFZrnuv`QRt40gMXR91)l;Fk`NVXuDIW4a9?On
z<|ly3$BiL?vj{cw?)l0+ZJFxHR1MKTmFL&;EHD5U1y?2_fb~LNn~Pu$^$}5I|1-&<
zJHZHK2=O(Z&39%ik{2lkHWL=$+PH*apMu=UidIZzJApq@e)H^4lQOafwPVHu<w=_P
z=TQ*)p>AJ{YBd$p?OuNrU1@V6g4>N%Q;LP+B794#&~h|NEJ?M`5irCb!cK;rw4ViI
zTgn<R51n2F=y)giTL~hYb^Q*o7%YKwS<Ps$g}fw|r1Jy=f_6Kk6emHWC~wj`kiRno
z0NZbfM{L*QMILEkW)M<ry2-ZXL;lyE6FUX!h^=c#dIF8}-4YE(UQ!LorP794EG){y
zNJ)EU8X7Wiu$*uHQ0nIDJqf>$`7Q-UEl2F(=~2Cfc3t6vVSN6|h99-!`gjOb7Y+dP
z&zkGvMZ17d>){$>GM9WvMP3d-`rtHjUGt^4WU(_RpP$a)c4brPhBKOI%BsqhHIniK
zULsJ_ZK;=r-PgNg9#f?C3i<YnEP_ExSLwR>Ngk(gmvBzl?bA{Ww_rEyJDxhJ?$aF=
z4_*MkPK$hCn41Y|iM-<-Cg%}Vky7BD*i)bL??!T)sdEcY6W9{uWnl5$QKEJdV4Nw0
z%ji80^IxrM<yqI_gs5|wKAzj}Q@arA!pF0ONeS9^OYQLuo;qjrF*l|?=r@X=#dT-#
znWK@e=ulCxk}R-yHQOGk$yy;TFX|I9%}*H*KfV`ugW&=1Fc`u#mM}P&+_|@KN%h-~
zpXuJ51?<H<d1)Ca{7dP~zc341ysk1!99B8@@^!bia4Wsukn7XxFuVx7KZc^Iy<WIv
z9VV$T_^k~VoAFc*q1~UcJX;3mL?W1Va2r^?$tnSCJ66cUuY~1jPE}A344>AA=&kc9
z94J43j%wrL>&+GFf2Q$m=e-2JNJ9Vo1)YjcolZr_7Oq3G4G=4)j}mtRA@{r4TTXqd
zY8ev$GO-s}Jnf00+$V4NOIQ8Omw$g4A~r0C4jSStYUONDu_ch5t^FACQy0>xi`W-G
z1$QXcr6P8*wXChTMRRt71B%n3LlX}l#ly$(VIS~lnl9oX{L}>pzfiBuOcvX!R6@|H
zz)2UuQhMqM2Fa;{F4i_EF##l10^G5am>s{ox(Fb@;L5U?Y!O37Q*?5hA?%yuF*MnT
z=JGKCmc*j>jX$Rf-5<Pi|IGJ*r>%Ry%;cLPuLVSlsKn;mr+D2jy1bsC@J$MPwas{*
zmZs}nY(f~H?Wohh+2#3jQeTN}@Kcu@>|a!V{DdP&hCGM4uG^S6f%ZI+e>X7<-bu`P
zE2popTJTE6bqmzZVkoZ@c#gzpdu))PsEa`RAfdtJ)OQQ7py+T*N9KCnlC~~#E5YEC
zFp@&WW0iQ#5?;!W)E@@EM4fyY31J`o<xLx$TxKm$-a<%RyupQ1h^}h5NHcrktD(a{
z$E~{8(-}+-aXI*~r3j(QXYBwwqIc*B>0vQUdH{U*eO4XQ6Ds4|q0~Hwp1jJ5ZbM^p
zbVRq|AeC+H^P4TnV8NmGv*10Q#Fco{>kjE!(SG}O@CSH;<V)n{d~nnyk%~i_qvzX;
z83A+JucuAP9po{UAFqhbj_4Tiexi~Bp7hs&F+{BeTDoiEnykg5$r;lmlNvukw;|yE
z?r%{nwMm22OQ?3ajJ%<eTvev!z?RlyaZZ!Y8pp1));~91%2Ob?-BXJVSqt4R>z4RI
z=asOQo3D<DJ67~IYCgBM4hGvTg87CYtx$JGkcigH?RqwPT-wmFp#L)Ev+KO+vSnfT
z*Kms>i$fD$mptblX2T>a)OPrq*r>sG*O=Z#=N|;x4segw*gZHNg|h8bL@kGw%a2wu
zEYL}Ae3k$jF0<n)L;8&bj!-kEqhR|YIyw$=i8)pWc@0C;u>?>R*)$L%Z9>~?8W{DY
zQrRobwH-Rr1#7Ear3R~arH5&!&pJ#3t_zI=>ry`5+t`53IVznol2A9(q}uuHMEV>q
z*%gWa=~d?XL%CE9wz7!BgA*&4YJM()BAe^(EZ@qB*KPg88ThH1_?(3Mu<*5Td(?U<
zl#A9XyRPPuWow_}y`<39?#6a<j3r(*!NVKOhT9~GH$n`eK;W_d61&$1`B_sG*ER#R
zuI+qcz`Fhp*7ePlw|<pY;Mt6Fu9s?-yfO`UVEsAQRKqWfszTJw+Uxwt<djeTD;?l(
zuj#fS?)kP(KEZG7j2m4<i1^~&2-i7~_;z}YjGd9zheDq1g1lX`sbhjZ+i4Pt1lEU&
z6ZJ<JU;R$!ANSbCrs=1molmmG8Q8p+T*h(`necAE;8zOA1n4swnLg!I;8m(!KPmsh
zASSG(_i)eo0Ga{7!qx{yBa$mGmtUAPK6dv`Rd!Hvacj8uP3k6^$Zya0rZHqim!y#N
z^cf#9kr}KOagT;jYSc;>sDt?yy7^3ka)UiNCK!bu-Y}SC-NNLOP7;_hR%Fu>)Pk<^
z2VFy^LA6K)g4X}sFr}@ZbF%_e0NM22F^#oon39<iA=2eyb<OI=sLCw{VS<#msg0AH
z&&@MF`huEgz^ovK8(wT<!MHNyOL+P95g@Co%6(Z<`=WR8#}#k^X4&ngS-ip}R7$TV
zO!y0w8p-v|pc}j1QLD(N*Kr)m)r81JT3}ns5eM5ywRM$IY_+k2+EupUKqlKWJ%ny1
zN%EHzM>1{9{fdJQ&Q&9VC=l0PZ>!T|NH-eCJSAX7h>70pd4@+x4lYS&C!-A}!VKDn
ziH@E*KaD?i^3<LZ?t03=U28svFk^iGCwjAMcj(WL*^0l7yzEAc<`PtLh>lw$cz(r9
z*8lRpzd@A<mbyjGDN1ioYRf6LJYs@+jnZkle=__@Pvkrm<~4=<*V?&?PM$qEgR2Dd
zhkH{prf8R011TokXT^c;^VIR8bcK)d%>4w(OlZyM*%bn<O8N0O_baORuT;n(Gd`9a
z<i!Sv<kgZ>Z?wb{GBhv3UQoQ{SY%xEtCI6n_1dIHOn+VZK^hLn;PzYt<UEwqG#FS}
zPi003=dkIkuPAz+k!N_2GD8T*XH*eL&6$M{%pX1bzWUf3$-se9k@{St#mJ+@T*YLK
zt!qEHle|JWlSM}RTATA3iOUBxK+OWAfw_-h;krhf2Z<T81I1B<OSJQbWz2r=3zyM9
zfSEt3RsVS*AOcCDM1H7O{!Yp9yGJ2M7i1aI@`&g%?2*2u`C44iRV6ZgP3}-Wttq!j
z*030^qM~b-F84#@#-ETYponhqT2bm#tDbn#jNaSb^O}0OGDg(PgUV)?k*DrYJi)+w
zmqsy0>e<e1d^F1Zvw`J<TPZ;Y)QD3>cH`P<HhV3ebR3Qn;N3S}zd3ZIk9-o>W`8<Q
zW!e2b@7=l-MO?_NFxL2kP#K|%3q!_frOSsL)OYGn-{bm17S<>+e23Bk5mi2>fg=^p
zmwJxm6=%~5kj_vKe2n3ONrja6#5uEwI`8uHGj^)-^jja_9g>kk1isK)&Z>c(y+pE?
z6~5VU{P5b?ob^n=+NokVJ6*F?pKWbdTugvNxIyXJh17C+AH2-|8K<WBey&xUR>}My
z_WZmP^)lZ-BGZi$S%0pwqueY!{W;B9S(k658_^p=gMijo(S@a1-JkePje3KaOfR<{
z)rd*7Sd64j>l}-F4dtNbcAlO+iWUsGBL88gt=_Y0&;GnmE%rWR)K{l{%FpT!_ayG)
z);)(^G|dS2*rZ!84wKlvaF>%Ce`BP_N3!>BU*;yS@Vp@YChb0O{RwM1IMrnT*v)xi
zd^DEu<l}o7((3a>5138SubF1Qmx_4j6qk-z>^R=H`o8CU2pELYfw7w(4);v&j)q@;
zEzgp-?yb9XAjhpoL(3JxZ4kH-9kCX_-GJ?q`e@kft(iWuKgVTc^f$52U-wTsZ)o%M
zc6v+syKqO&mT34iiM59><=f;F{kwM7PrqHkvBd=eKBfe&-E=b46;n)oDp*c(a-x1l
zMe{fAJ;Lv&>WPLa?!0`;#xGFz;oDYrT5YGTJisCOFNlOdKJ@hF=jjFgPW<w-0`DZh
z_3`!52=jCKUk!bZwkBWDR)heSg^~Vko({r-d-h#n9ymyS$Q#<N&Ib6G8f;(O!+*_D
z<O*TN?{+}bSps@A=`kKu?XD@Lsd0B&u?ss-+w@LFhxwoAyuXpz{MbSJ6sF|J`^TF_
z=+1$-9w`q{j4gpIfHDZ*FwC6HMgzrd5on1tqfb3MvCIajieK#p!t4+`c>qxpQSwt|
zc6%McEL~+wcJW3$qCm|BjBD;y)>tP#Lpds+sAz=RPOK_~J!X<Y*`imi0o|h<h`&PM
zUkfM))PKKRUq}cprh1l0gHu03(k?DM(a;bEvL3wG#zG`l-mqa>`fhe`Xa4sAk{rb2
zb))}}N()8+CFGoqs{ojpHzNn#`66YT;2#n}ejKada}I(MUGXsPiC^{PtxT06ps})o
zq6pN1j9^}?gZ4Sw6vTZhn|-CRDSI+71<K!(2K=8^8MZCKzyOH13Mz<#pTZdze;aVz
z*di*r8QwBQ4In`GL?H371=^Ex5NhyA26TxzYE9lev`w|4I&%MVM+m?jfy+4pYVmJR
zknuKP*bD&btPPan1djy6M>XArySX=ps4!i*Ecpak0)VK3Am|1??|gK5K>Uf=q+{TL
zRz9=Z#By8&ktq4TJ1P85W!euo!9oRvXhgiuNp`-Q2>5@~$);9R#bL^s=PXr!|E_;%
zxQ)3XF~e=G9`5Y_e)<15nd&cJk&|nbFC+Z#pZpJc<zH9npReOX0Si-%MyEWK`;QCr
z-xk<^`8p-p+q3^XhWg8U{O50|zpVra>ThyObZ-CueEZMrZrc8J0P%l$oy`s{c;%RP
z8UI4U{pUCS`zH$wup<2Xek2fnx(4u?pm12oIl>w|)}IAXZ~mPM`^PFZfGFUx(Nn)~
zo%Od>AkvB72QVzSwE!vmplR)!aNkfKf>|P<lM<}Pzp|!BLS=g*5E;_94O93vs2GDp
zEr=*K1gdj$fP$zMmWbL<DOFGS!+ASOvUmUBy{mr@C4o5d_Od6_rvT}6%aocatiFg7
z(BjAk!Sju~G(H)^f(DCLdm7swiBGNV0oeME&Ef9aHi*1wjexkUkavTZno{w<i5HCN
zsK8i=POsq4tAKz+dP@NCDh6O~ou*uCB$G8{%_05saHG6WOKk{D>;`fULd<5+cgR#}
zoucPKty&8B=^Zj_Pc{3V<EYgFyzM&$+Xf+a98hffO9Q8mD^`8ET-%;Nt+0Q60tlun
zpup@aXq=ej6o$jyNz(QJQe#35ImJ%^Sr&mHr;yo33Jw*=8`A(XDg|ubHb9A#9sKy5
z14>t0q`Qw)HWs760w*HH{B58EfI3u2^Y%{NuwXCeF(2{d#7PoHgaRJ4xWfRauFg%Y
z&B_@3`$O^9r6b@8ol~AabUFVxr&xWRF7fu^0FBj)-G5>lV0`ZZ56%0^bhz*^<gERl
z>!8!I2{@~tfb?*y$UoTC?otr63$x`OfiMm<us%3_3q)h_3Pp#e)o7R#L5yH9v|@hU
zwS9q=PHA-$Py_ih?t=&(XAosI0s(CCR1#T!AL)DgfWzfO=)ypU(-O7OfReF3#V$ZC
zDKnnrldX@f0_xrf03hzDoxA*V^DGCg_PI8?iXss#G%p6%W3)nv+Tg!$a&kn$f)Ji)
zKl|45_iJ;)58&Yq-dlP1&&BFiod$f%f+H{+im@v?6+61hQm`}8N{t-E9*zQDYsdpZ
zifMKQxoEIcKFe=>bKxtL2?tnxbzw3UAg#+$XBntSgbm4dRK(($S7Qo2Ie81NiN#|q
zB`;H8!Bh_lRLIv6i{E{8eC@n@-BG87j2VRSgYliTP|NQS_&@EF@ua`D$J-^LznvQb
zPE><{{<#F|0^lzc{P1Rs`Pn;kSFRcNJORGbFYN3F`*KNzaL^sbS?kZ{K`JUIs&}yf
z@cy<tkN)qwoZsX%uoX_;Dddj%bqv(s)|db^>&#IA49ic3$R4Jf*jfQ2xUAc?fN?qu
znB4c>BVM|N1pp|u0?Gw4sk8#g77+(_N9Dltto(zlkt>i}oT>2Q|FHQ2_^BCO0m{8W
zXh)Lr38K+0A_pXNE~`4*agP=0YVglXadn4T4Rhd3toS8v086RJ-36lK5m!ceEt*A0
z^?iDwWS>aV1Q52b_q_Mao0#T+9Goe(z8^R<Bm&Yr{b$rNVg~|5^?hgPoqR8IlpCs<
zAAFbn`3Tr_<pSNfb3<i3SsS4aOQ|2w*>nM@_AInCZ3;hZmzc*Hm2-xHOe@vC7I5T#
zeZuEM5cdpneTWVbfl`g>rhI=X9pPIR)$InJ3PewTOhN9+YvelxNtust7WUnesXsx>
zWdqWnw%=%aaWe%QC@)A<KFwkTtbK!N2}ynXrdO7vn{S}lCTN)G*bb=cIuJ#gz766+
zLvdI|Ajc>(!og|T_}Yv8s|8l`YNhj_SQ`OJO&2^M<ru)?Z4W-yZj<9CrWE21=rqB$
zAoEL=uRdoXdvFJ$BxL%G1C48g-&)9VEyTQDwAQZj*;Ey$Bd0sqFsYo4<ym^XC$_Zx
z13{*x*@v7}Hvt2}@GIBGnL5+064OuxAGYs;&FB~2|K4QJ9wf)Z;+8s3LV<VTOAu_=
z;~kbBDe`u^$28^C6rJ8qxDT`wnB|)=r!jf$e@%^aiRJg4glDZ#?|HZ4>L!C3plmKG
z2ca#614i*DuLYU;0Xiy88?osPzIzxlI9Z`iJ)F5{)@h*u1?JjZ2MP&^pRfJbP%kP}
z)N9Xi!1R07S8M=4lLtwu%0W_OmVrUJ&OV9my9CL}T}WIIq%t-zu?KW@sMZkh^iK1A
z=1p-uTx*+_#@MIqNR9=}nQdOgtAf(F`9G!cUpL1;UOxm?1<mhgZ@s{I-%Cq?V_gA>
zsSu|R0h#+T&oabPwgt78mD<V42|$fzX_}pJ`<`q8ivkf=-TKBB>2!Vr6}4M@BM)Ax
zU;k>%b_{e%<i#+5dD?C8u;M>bvBQXN05=w0%*S>A553LYNe}#frUj6d+&`Q`vfX26
zdEa#<kxfp{(O3Tj8w>mhL#CDck`q1&9iO9M6L?P2%pRRu!{*LG4(^o&rmmgn@eH{j
zD|`w@ytk7$e%M+x<^nd_kZ1laPVpRcX2pjf5e7~9^sJ*7>x?__sK_`o!*uG9Cm3v?
z(nQ9kG0rD@e(}H|EzMaI!kwEk3)Brv8MhZ|Uedic8vY=j8W;d5W6zB2TWEmiLybbp
zyvNKm#OLq9i2~tuwx<*ziI#&5c?NdP1(xR%VCn6(no$%l0%?g#VtUGaK(sImI2H^g
z!Pzpke@3_y!9EB8YpHTB<|+D=fSXTCjUB-o3#Q{}d1D=Byb303rw2*}fDxncqp(tk
zb&P(fRBRA>pTyP&v{7f%KVc|1GBi2p)4MOKzoKRrZ!LhR?_xa~Sduk%uVjyPleB%+
z(b=>cz+pB`8Wa{0Q$45X?pub921*15zyfQ50~!5Ns&9NqxqXi^GRU<m(Q|7@+Ntz3
zo(cbes0<d!C2B@!)_jP<<#<hPC$0I)kY)myKkbdB)9YJhQCG|@eaT#-jG5hlaOtdx
z_y)-lo#2%fY8b&eDnu4iu{M<tOrr!<?*T@9&ndS;6CD0Su)S&B(RP^orR?XCz(<(T
z>r4Yf#+;_-6cyX>1)t9GPU+5XM@w(9iF+4gCacGf3LvX2`;WW6X|L=9$Td(OHhWz~
z{rSz!Z^(ksQ3acXWeWq<pAfnKaUbD|DfmyY)2$DA^Z@iISEd|ab(~A3`^rVKMyD4q
zDUMh42hCZVD)$qeq(bSXe<7V>z|rw%c`3kqx+pD@lbO~L;*9HiLUoc_N9Z#P0MYV(
zXuSqxZc+$g)_TxYIVmJ54#Von#AgtSt1jR!wKTd3vZaJ!;XGF$qBe|#La#V2Qhymp
ze6A4b8=x!AdCwsE-qU>jvAqjcPB9=E3<BHYAFj0Uub8t4+lC*2b8)sqQs1JfM1$0+
zI1m}O3Ht!mAP8&rDy17+V5M{9-?Wctd7|XYU{fUDu3)AnmCwl@+SE<~pI-&@vabpP
zelY4VISlP35?b~lcCBkBX+NexaT&R8`sw+$Z~ViE%l%h}tZj;4RRAvhsnpg_II>2G
zW;Tp+Ey$k>oFiL1)kxk<Vpj<^GXKj8mB&R7W5M%t|I=rPVx19s-uXNWjPw(UDRDDh
zpXf<JLR2$_Ci~5XyC!3l;}ErK&L#RgG~I=<BaCJVGmLrTdr0OJVdh0+*M462`)(YA
z2+M>SA_DS3%4koBc9KMSsX9@`=EdgXa30KKgjfA9A9A4jk;!*u+vN>LACP2N=HIv_
zLnjp5<AaDaz81cL38ctmbvMbc0tYZO00lCO6md8D++(1#y98hQ0ZQrzq6B8VjcG46
zB)F0nzLt|svMv$wltE4(u-6nq8bBX)1;|}<6J^XTNtFwRFF=~Upt75INjxiWT9!CM
zP+(dnLzs6=OEuhD|FhwP=gRb)`5X~%y-J@aoE)1R_1mMDXV5*jcfyyPzUW~n;-wo!
z!?~67`tcnheV>8W`6C_DjXX^n4fsof_b8AJf>U=}&o2<2bQEMIgNQR{)w2Jj#xp;t
z0lQp<Sxu`ubH+`)@fZhPz`_|Fd}=hg(u>Y?qG9Hr@v0Q@?ERU1?2gKZ>Oa2|J}(3{
z<ox!a?h?!S#y0Ar=Ib1v&64oa7rX9*rao*hLi{zafGV@|poz(B+@av{2ALls&VAx2
z$~%M;tyJlGGyRVP;GZk#RUe{568m4iH&6%bXL4i0`u@RPJUrVd?hCPILfD=VlF+<%
zcQiWE0M_Ao@`^(`_{Tc+0WsI9rx%5g2=nD_eg`%)l|#?fpvg|;uJHZ2E<;>wC5+pr
zvfrx$9L08idibyd^+(mLJPzRHR6yrDiOWVEEVoVO;cE#73hlU28$9ZSaC-_lYT+YJ
zLLaYy=`=cDUew)YK8{=y_DSxb1s(^`&qn5$*M1zss2EjV<<GR?I+ML`*YjxRCy-!X
z8~U*oD^z(f-Wf!fP_5Nz7~lV-Fichmng0;=<&)vn-X|X-sZe9#LfBO<2SHtYf5F`s
zSzgt|Ih`$Z32-`t0&U<K95!~hAd3oAXojuA?f!8644-Prw$7GdbHHqRLTme3*(?FR
zcYXS{^tC!uMkpp$7-?O4{<{JgdS%>t%!fQu`gT0o_^FdP@!S$uncE<CLcEo=qWzmI
z?-4m3<@zYe%tBx@N}|2zb}r5@?5hwm0pGo8y922p?9B2PYzF}vW-7%|8x~ap)R5x)
z@50Q@G%iK9d{B3}-jbK8j=jk>oDTmegsix1KFXJ&JnrImr|78LrS_|%G*jPmm2Wby
z)`jTG!!MdoGi|&LJMzymu(wUr8Rj(eHBjruy?CrG)k==kce&B@PW~YZ-C^jPRLNR8
z?>$^vJ<^3U1x%h5_T;H*?XPEznEM2O&af=82t7byJ8;r=&VUr4^7F<9@d_Fx_c>l~
zBR%q1-aVX4;JTU1@wv{GV?k9sIt1dlFI$N_{CF-{hrL?Fc^1iIv0N_Q-pV*jaM1c;
znBt#0f<KXy)xtUA|B#<tsdCw!1I{$cyKDg4=v1wn7<qkPmx?!h0~Fe9I8*ydZMnff
zyGEl9=TP1GR|ff~O};`H*;O6td{zZJJ-2gVol6Qdu`pSc2>KYdcx&Xxmj>OZ{Wuzf
zG{Agm)ZGLh>wZ@$!F-1-dq0<*Q7LdC_CTm&Vz3xjddFpe(Lcb{D{gK6N9B2(owfEk
z_A=*&Q7IMX+&t4<ZpQP{E^N$=jTs1(G+dA}L!v?~UgKM>Q-?Df%q$-Zf+U29ZngMc
z8sH<R)oio>cI-D1X@N`)OAEBVcA}{X(ID`UhBzxA-3O=;wd%c}um%N`N)ydsasZR2
zIb24NTH)K3h8l%&MN3`k?woZFX9qQ_o(mnI@p;|w9#p8&;|=9O<ST4oFs}hyPiUGS
zhZ&6!2L_-c1eSTi6zM$$<d2X626J$>Tmya-4S)77hGg`v$xN1PPyn>s=#_&b3tQdN
zUv9X*HdHt)8#-?v?uxkAIsQmW)${YN4lS<N7*88C(#^-!1?k-v6dD_ZLM8r<x^|gr
z7ZJ3>(*GCJRW3i!bjuj(R_VkRT*u^@-v4;E!pHpdZX+6N)I7?LGTS&AR|hIkW9+jZ
zpNIturtR{cT35|<UryEi=7oG4A0~Q6^bKkPKG1<BZRI?V*0Em;Rns<PE{BhnG#l}J
zu0Ik$ZAcIsa-o2c(4<VUHV%~19d_!Av3=|-UvCJj=RBAmHhxQPcuh;A0)a%cdvtoD
z{KIZ%E~3OD_B)KMhL~i%JxgUPqWjEnv56jH&kdDRKprT*F*vf=KGhW}pFGfp4{|0<
ztL#1*0<v?#`<9k6_`OUP{O24mL$fLFmvTN$U1p@%_~xKhNJ$^SfHpbczzPT(o~RaU
zvitco?<%gaYU{1tz&wGBfq9Hh#$)8hP1T?E#Ny^odCBT62A*pzO~j71{A=IIkBH;4
z-6|4)doC`Vs3dHatk@N8F`}6kigizWC#*V=l5|A~dN};T&=sZU(C;A{IJ<M06~0m4
zLG6ut&jRou%-ihAK6BQe@FyLcVSV-17rUr#Nz!RjvvCN50ilX#kzZO|UPL<~=jaQ@
zSb#H-<6H9%57;zn=`srlz0%l-Vq65!kO-0$&wqJIxEq@_ZKt=Qfc19JF*HB^WD#HN
z4r~pC#`xSvWb?M8Ha(G6uMEh`0c<XbQ=c}Cmt$Iu43^x7EPiovIUc>SZp5Ovnh_^=
zmm6N4K>?sjn|SmuBQsV44S>F0{sVob!xL4}RCDqIq7fuHWl+ckmWB#2B7;oTD2-2L
zqwI8t0NtTtSFO+*ymBuF4&@?O@>IK?g72ZzKjO}&RF;V;-kx55+-0<xQ2n0A>qZ6d
z=sQykC53P)mCy_e7BpQ!zH30M`Y9DXtl20Z7`#kYJb$-3py4))PGA6Zb{hdTwhGhs
z`FKa<FkV-92@nY`SL_ausq1}zX}1b>s8X5Q>@c%WqNb{Wd|bdh?mS<`6}ov2GDl<r
zmzyN$AiDT!g3`MA)-QMOw2|4a0c5}&W1o9}ADN+Y%y##QDDF4vwhr3elj$gvtYOr6
zj>RcD^E7(dI4v+>^Ba_!q$OP)bi)9Y-45amHPt#Jfz=&lfKt_pSmrNornH4$XOXuw
zxWEXCm(2<Efn}ACfDJU^An<Jc3B>fgf##t5GJxfc?i+`TF7MZDCZC^W0rH6p5YJoU
zTZ!#O#+u)Ool;1)%Z9P76m)6iV`_^ZNNsH3-q1wlv8`ttqU{BIs;0fa9m`LJ!(l6K
zq7EH?J(_cH&t^Z?XKqCDo^MA?gK<w@g9zd$&Y-oY;lIpO-J+QYoHK1ea|uA6r{QG(
z0NeioG-o?QAo3EQ9FUwr=nmixKu>@GxEZi=oN^YM=bWva$iPdd#<6T2T5@}K_OnFP
z`ei0zoQV+P5Ufm3+cj!yHT15ZPc5ZEK`f&gBAALS9bO~Sg<0cl?T;P5dW$wT{C0o!
z%bMCLX}LWp)f*9cntgnRKh6MpEjnwYJYiER|CDKEsMJ?ltz0#+tBI`(tQmSrw98L~
zctuUFQ8k=js<xcxpG{xpV>jl#Y7!A$9<@r?5f4;oO3kb72BmK^efrq23}=oqBcGJ!
zefFay)7t4g3E^x^Am%mr3kk&*fV9U^d3@iC$M7gGhVKa+!>4N)o?8vR8NPt%?Ioem
zgt(Ws4$-ZCLX*Z!kEmI212T7Oe9{&O)dGN%9;r}@>8AlB`H+ui?1eTYSqO~H%NEUo
zZ%1-BHq>Hzil2CbWV?`$Uci*&MpU7>LEGWW&oEk&18&CrVOduh|9#M?;<f>>QZxg*
zNEEnf&wGuX*)i^;id(|>&~#CjyLRMDQ`)Kf(NMNLB^fQ-g8qL;B#jp?Ovvu@{0Z-$
zF(bHtFgzes;n?SN+RH@DVaLgK#l`+6{;_rfe{V&zJ*Wly35Aj5<%WqbF5<M*of}VI
zd@xhSi~ZKz`EK+9q*TJQN0nT%FEEHjS$@Q|-z^87%tNU>ZQmIFR~E}EW!k0xPkYxH
z)>O8xl@?||ilYP}nh-`hp-LB|1PfKiK`?ZrP3Vy(B3;mcpdwPmh9E_nKmsVBln{b+
zX`yQ<N^jE7%G{&pj5D6&{de#4OnxOh``PPTYk#YK@B6*$BftXKXHex+yi!4)1)UIo
zdJq7)Gn1{9US~O7!74oS0_<;8>LXRXX>PcHcY`ty6re#MKBKGdr<*xE8c9J1_JTM|
zdSN;zIPKWu@yN`H#mtly3zuZ#ME<0CZwST<?+3N?voZH6MAKf8DFbQ===|kWyozjD
zoLREa@*D^si{h|hNtl~tDxc{ul`89zY#~M97<Zzm1qT&rRtnf~3Tfm1(Pnk(3cTyb
z559L2bD61T&gm71gxyl@D6@cGVNrxGEUoMMzE73}d*K5Vi>x$aK=(u}YNe4{;*{Uv
zsh6CD33uHTFMVJs+}RNIYkw^mZ&QfYau}f9@MzOM)=|F)M9ISP*dUmPoi|yonM^*a
z)8BJ8kg!WN<;wb`Tz%81Y4XAeRqU$at<hGL1tFP3uS~5$vb5A3JfT|_ns**~Z+O=(
z-tm(<nr9t`*58($fA-!8=yIz2d*Z{^^h!Q<gB78imMR_bT+F;g&3_a`h*Ikd64s@U
z7ILF^n0=3?HiHE5B(~d4D8)Bx$C*w;YiA9tDan8WN_(|BsB;eX?QkvV=tWE}B%1kw
zFHrcz{t$2+nIt`{L5t9q!~f&WY*xs?9V?oNOVJ)hsx<Cybwqp$Cta-~`qLIS5NFtS
z0U^FOGr(I*8^N!QUZ9haN<5NV>X}^t19N?p$iTXB{}OkU3&2>CFvjG2;CLdCiPjq9
z;Q_p7&G9B&GooVdeSy*LkebhQ<Q6W^=WIZ3SdSR-z~{7Bn$fF0_dJ14T&Y1*-j==F
zZfWfcr`cK2Bf(h@BnbRu7l6ZnV=P-G#Uf7d7XBCt=_|stP^f4-v-XmE^vGFJj6z@5
zm=dn>@Le6VfiB0UVla)WF!}z^yw5|}@wEkGzJCbMsH?ANax(}HgFM$Jt~V<}xj;Cj
z&1XWKHR=){Jm9v2`)DT+2AI3SD}&VD)PXlcbf)Q>7s~DUJbk4Tn=0+#%#6jgLs=j<
z;9_qBU|yF#)YQq74!7@ZV1LRE2GmM%gl)zi4KP88HYVg@x}sxuk@SR5n>WS0P&*Aq
zi=eY$pv>N4)Ko_5tABKY@(Lz0_5*HGZby|9f2V<&9{}Hr!N4xmhjs_kRMXLw9zhI{
zYm_Z3Pa~phi!OK@etW<G5C$_H!ayI%=K2iR-P?ulHhoj06;~nba$U)P;oV2AGDF(T
z<oR$TRy8jByuy0EUOI(gV}_+*@`^+u!r0aZhP9Pou**2iEWuFd7u%)jTXW#mi%suQ
z0fgWiuqMkejW$|+OEh5n7(WISXUNyz`vD<A!5vdaZCDTLV$-dqZ6I-UwTpOD@OXXk
zf)+!fbzqallml2xV&_C?H{}bgb<*ZIoAf+<?^ea&CeQc2plX;-g;iSnbDtLV4)uF8
z5a*@Butryb3F`nDS1LGOVy+wn2(=`FMV0}Ch26_eNTGfNM{u!Wai$75#k!b)4?Q7*
z%l=@6i9U!3KJV3dxEPQus~WQpZ!y-ifJjH8K;MI^{fz$Tvp`pShf8D#FZD99ZSaI|
zvh$N4Mi4C$E9thE8mES#KrUm#cZT+yH^b`2K#5Ph9hKlInUJ1k<U5-LXbvhm8emDJ
zuXiYWiF3UjXe7b3&iFD9#4GSX4g@TiJhDD9bO1<>B$V|igM(I{dF=`=RF8o;<L6ko
zT~!SGi{_Cf^ZO!Q`EUh1$r`gW+9`>+VAf=%sFRDYG@k*X{EDf9tJfmVF%6{kDQtzR
zYo*O|=3JX*(t2#=nDxj2aeWAY5VF2CeA7z+4vd~>E0kbR@GVP8yS&gSAd3~IuU4`H
z<vkCQ)~ip6%*wI8Sw!lGTe6aX@X*HaN!6P3)`b>rZ$Xy2IiS2his9+s>*-Ws>VI^p
zRa>qd#X`e|6DU$MS1V4EP6S!Q9$O~ylh}A-@UvER@&N&<>kbDOIj6;z?Q>V?5^phM
z7hmQEWmbY<N+raaoA-v%(dDyYg0qabHwS)CdGHLpdcc81J?i}~&lDbo=4O<ANyiO6
zt;le@qtl`5y2cs+^9lFXChHFNM}VFwS1&xpVKWBwY>w2`r~3G}`~RmAZbFIyj<ng)
zJo~R*E=chyyhTrbqv7AuO!_E!Q@y}XurEKscg*D-Q6_R_){dxACu5JBQEQ`5*KAR6
zzC@|=WXu6s-xNMir7{4#=<v-bw3@^0QP<^sha6Goe|!8IXYsS`cq;A&G35+S+-YUf
zD5&qU%v*DLr<b4H0ShKt?eZA^kTi7W$V?u#0qsPaNobbrOu|^<)d(0{ocNLXO@Lah
z(ne)*HW5JIeW)1CWA=EdZJ<d18qpiM7~vtPfjTd}FpMA2JxDN?@cYC-sHuTAcDTmm
zRd7ESzUPO$F%Wv=cjy@a?I~<ce4ag?;>tci8{|lu9wh96`WZ_==SC?V@f~w<dm-8v
zi+K{QfL<fzurW4UJbr@V!$2=5u?bVqnY3qVj|>O)7Z9VWe<yb1FVu81E7L|}oZMql
z#AgVHwQ;-ko~O~#V$i!)aL`<u?X9eH3nRAsa3f(vv=9z^l7_7A;ei`yJZv%6g-+`B
zgAE{3uCHI^L#`d_S(N&_nH#QjkuB&t&uwQ|1?hFv_^4XHSCeFO-bNB0#n-)ojjZgr
zF*$~cD+1)3BsMGsZR<H$=b$@Cok}Kt7;abd?$8*n022<x5xSX(@B9dKDP^D!bxzwD
zsXla12O@TD_11>zHk47d>8QeBY2&R&Z@gM9sa_A-jxUVRY2}>;+iG@Ec#*|SN(6D`
z5qwLee*2<8jB>OTML@J*_SMov*d!IqYy5$8Kw`#q={3Mi<Snk}VTR<Ptx0U*6tuDo
zM=7=+fPi_fVA1Faw-Z|26X<d(LZy!hM+JdWm-sq#E)6+10bknhYv%%%NPRR=GVgR*
zuFJ0-S&w3G<%i^Q_91e4v3}_`sWyh=J?Pgm?&UpH^JBc-e%+)paT?6G7dFO>i@+x}
zOOopd@rfM?7VNx>&+$8yx_%&)#e{1!yS{g918BL~fz0xH0$oqHP3Y-4n%AH(ojSf4
zD@3V_lsQDqs7=iGgXHX;N)Ae_;GmPFJuNOo&e!Dq0uugP;2>>4vAzjp{{~g3D95d$
zpxpHv0yy~W)y}8*ZCg|Zb^C#4r*@%wBG$jm3;RUZVh%OMv(JuWIaZv4zF%Ll%e96%
zY+hoWaTabr0;H2;SJAW|LovBGR^<jU{|Zt3#Wuv6Ku@?ObY+o0i@z{4NL()OlBoa-
z$>=&`p3;FLONjzOFfb)Vc!q~acx48YMu3vesmMaK=|sW{+w_ZOm1wSbPB*is_94A_
z*-02`A)T_!6Kuxru$#w*J5+{gm!3qS4lBvjJYTh-VEUQ0_Sq$njY~C^K3*A8aCMxB
z9p=IaAdl6)b@L=(5}vA2V^468AdsRuT<>b~oDVQW0DM`BCud-<Re6RpJ~5<nb)QHi
zhJfXFz5hYbo?RfR;?o|@7);3ZjwsWCh0{?GF7MOuD1)X@%%L9zGB7at+TE;sxWWbM
zb(F5Qb`AVg$|~N}KGme$r)FFHGeK!Mn2{Zjn~S5@HEV<Y@I?r+*7NI1*%rr+=?#?I
z|BTafZ4(le^Ea}ElTY*j5;fmLg3l!JJ(+72Q9#Cr^w;H&sf{sju?X~lUEp4%W}-2C
zNrpV;)}oztPU!k!&VeP)zBmk=ZJxt94#K3mB+L-%REFBGh1QXEzN{${Nw_V;`T+In
z);^YpL7iqCWN!e<geQvKdD`Iv()-v$+|)TjHR|=J9%l~o471l@EKMT_(Q8+O{n#`w
zs!OxT!ry6El*Y7nN;>PFt<|8PV4<oz-jRc=x}l)S7yZi;+q&`=BF4MG&hm*1t5tu5
zm~&^_@y70B-H=iLG9*mPm1Z=E;YyuHy*B+|q8q~rKhxh_Cf)~gm4YXVN8M(a6#P*?
zWR<brPp56Inztoj&lp`&&e83K)KvWHfx%3!z{O|U8c6B-IKzs0sdS&rt#TTiX3i86
z4NYU%O1UhTUcv@o-dS6>n+j9$#JEd#<__$R+vkDWdCa^>7p7wUm<tI;4iJZPBDi#%
zqWw};gd~~PJAVq{JgEQ(4DCn(F*+sL&hYrz?XLVQM`(B2Nh+fMiEpNz3eYd@N`^au
z|D2&FBzN;%Gpp5AA~7h~mi&ykP{nJigr>%EO7zg`A|(12rQlpI_Dc?Xrqx;J!iuV+
zhw*Vb0Vn0Tg*PTO?w#D`=6)5jVDlgvZKH%xBi+sWMPZ#>l*mINF-ma3=gfH#4=I=h
zv5I_B1#hGs;mi|508k0Ok69=t84*n<BANX&*`t0_nMdp)hj^>kLONyUom)%|$!y>d
z>#3dv{aT324^gm*U*dPqBazZf7vC}Jq0^Y4q8d@xxC~lRHq}QTH(hxt<#mV+Kg}@B
zKO*w#4%p3Bb{1Jx1mvV<9<qt`i1m{8k$x-9W1RjjMa5Z|+^kWNzr{%NpV!OgY!8R<
zH2R^mCin5;vovyLkn)l&&kQE#v%$Cw9Ue+ChIJDmr(O2;Yz9paw4D0Yf)_IE{py&2
zSQJdjgQimpQ14=zR><Jh#bn5y^3W&9(9tv<=zV$a>j)1n-ihwnYEhX*qqNzqjX<7w
zTH%Y9S^w_A4R*`7<sW&U&i6Wu=gKE-VCwzrpPu{;cPUk|40q5zqBOVkE^R^nO4_rr
z$(@~`APtH#_fSVmBc((wtUY(hPMju=CeFxQ^^6%zxOBm-UShSLl28565vNE<Sk&i@
z?pPG-t{q-%7+x-JX}UmJ^q2N>Z#nwjNN`pVr`*syeod`xVZ&LT;uuZsm-FHzdwbmp
z7*2n4C*AQ7$JpBKwcDx6gJ{7;`J7dYm6tp3*GXH5bVl36mfb|*j3>(OQz?#G+ZW8w
z#&#+$lg2|}a*F~bBp7L<rC7P4g|xU<+46D=8nk&wIo{Fi&E5LdO5w0wyBU~Rxe&*K
z|NV=_zE-84x@@w}=c~qN5aMDP^CJDxju}E>)fPOT=0>!G?vUki_7VLqNBZjKY0W7K
zc@xD9*wQy!DKgP^=01aB%LGczhd)34_bd0s8TcR-XHF`ADf|@Mg^03<h=}ckG=6^m
zD%U(SzT=-%MGHzwy6$NN4JO>Z+rP51a>r!0Qe3lG3uBT$(A(>rZ&d{%hnEJTUn|vn
zSC*T{eo*>k@fS6smkH82W5O(+K8AfUDTB$!*SEjIclL_gk)*s6#qGmOOBYkp)7`OH
z>~tL-nxP*)olreEXc?4I<JfYqq^_<`^`Ki&P*q-j{%V<wQJnmiz22<LYc)m1Zp)XK
zdxso0AI1So&<#}|OHf^7MY)#3e{9(=Mek>l=P=1CivDsT?I*vT?ZAMImFX7U$4vTi
zXLR$su{k+8K`*AAcM#i1kJk(h4Gq(a<1N#G!{_Sa^6~0Wb5+{*hwl98uAFm{RuDm=
zgM&j~d;6J`v^3Y%nF0RFi!<FeurVs5)qJEycEf(bjU(S`8fVZnX{daUFXmY6S5i^=
z4X$xzvsc1TVf@{@3jsivq8GlQxSK%{5%*-!JfG+Wv9q;(Shckf2H|AUD4+WveZzC)
zhsqP5z2@tx$M0=bGR>yG+WVEeeJS*-FUP!rRb7t3n_pf3cUS$xU&g17jafhSV$=JU
z1z$UQ7c+PZ>AM$x`d*7ffkhc)Nr}JeHGeSWpLIVA1yv4YAsKzIMH#>^62!_1{_0_0
zTTj2o*aB2}`2a%WdoB9E_!8XEW5XjOHmd6C&n(&2e)y|;e(g)(p8JW%Ko{YJtHYe9
zFl|bN^76iym)B(5hYfYVx3!tv&JP1_iYpF>d+a`DBF|xHU{Dek7Iu@xSN0oS<<EwA
zbTAmDtGlMF2l7q!w<;-}aT6B&^IhM#QpH48;Ba$yFLNJyxpiyq;P<{pE)j&Zy857r
ziHUCVl1Z-8r>%`uA1|-}ZSC)Oe<;~losC^`rCzVExhdrH{na)$1)euDiZ}e;#$E=l
z{WgbK``xbncLD#+2mgKt|FysT-9-N{n`oO+%E?UO*6OdyyTIop+EC}2mUZYq07Vp)
A{{R30

literal 0
HcmV?d00001

diff --git a/_images/IT_sample3.png b/_images/IT_sample3.png
new file mode 100644
index 0000000000000000000000000000000000000000..b8c7b8dbaa0f51df0beff2e18510530baae14aa8
GIT binary patch
literal 396642
zcmeFZbyOTr(*_EG1PH<12?SqQ5`t?2f#4Q`d$1sjv*-qg;I_DXaCZo90fM``!(xly
z&HH}xJ6F#6zWjI3y?bVMW@oy4rn{=2s;8<USV=(&>pA&zBqSs(>5mdBNJtoWNJuDI
z=+6*WfZzS#NJ!6%EX2i?q{YQ)mFz*L7FH%mNFRga;?a~;HA%i5y8~1Gpt#>-e<#zX
zBB{JL&y+=@c!_~46Xcy?Bt@)l%N`W;jc(^hkdGzumy#eOL*?>mtjZwfeWUh(fyZwz
z>u*OMAJ5VgxwhT6Q{9gmT#h^)sF4UwN;OiD*-#=`2BIY3?p$kY1f+1}Z%Eu3Pd*~`
zZkmK>uCBHt#j)JFxI)mx)-UrW!KDz$qy7hfVBZ!pl7Dte*T|MhmIvh|iBQEWG^8)Q
zd%QX+^witVu@?B%VzI|z4g;@n!W_m>jeknWeL|%$AxA=e6VE1%{Oy<6eW;G!^Q6E7
z|H`l+AYmpza+aQpD<lGj7d!NubV7=HWJK<ULvdi)p0VG-zJbL)O=r*#*TZji?Zwg9
zql4f>-51l3`CFXt;!;>Kg2YN#8_ll9s(dy8GS9J+jb1|NynIx1kw;a34Vp5^y-wdJ
zctw+pM=sRe|6Vj)ND-Uq>-VR#tx>9g{_l|^aR9@ryacwyH1#BE_N)Boe6Ty$Lgfn8
z6t+-|)U|{2zKmVc+67>gWZynX&|ORZ_C-vYFxPF<#dCgG1%IJ3GOWfSA8|`#oWE*)
zqvrIfNbC6PQ5e`Ryv~EePO@G|FLD)lWwtFj(7Tm^X76TI?}0O3@4c$}Vulv;O9m_Q
z3*1nVYothejcBe`TxTe_P$D5ZsnZ~nSMKLZ5O|7<%$dje5-D_fx_|uyDq@H8@@CRv
zSqmTOD668PLUp_~V+4tE8tvPw<jcfMq>P}n+d*xxiu6$qXpO^g6!Z;016e);^MyBA
z`4i`#81=}&70f5^NQ0i78X&KDe~JwL+)4^SvlM&D{4BYhM+x~yJGCtm<uj>wTzpT5
zePvh3>yd4I1I3=*w99h7Sn^dy_c_6Rn)N;@lKRUF&G*i@1h8)&nFh>Hz6auG5`ILL
z&fNU1RY1z>&6+`8fPM2+03$W9RBZ9LX8{Sh-;!9mAu40%TMaZGZ?kr}gr{ij$|oLt
zPxVl@+n=6D)>D@H<Fw14xWdUjD4^}YRXkMM=jc?UQpLE89g-c9sISPPXxXCZa$XdE
zAdmc<gVGS(7x{97;8|d%5h*EKC<P-DLD;GxvEgSU9s^cRG*0;h=U?)2KvU8)EZO$2
z#_ScP8rVG8$v(QF>|Id9D?>#Cpb^OUw&K{1kCMvlyF<^@l{GtydXsvr`qX;K1%4gg
zQXF2N-gWL%DA>h=G%aMLon*~r1!r|+&Hdya9)bGQhq^s{Md~AM5c;*>n~rxKWbbhd
zzj4ZZVk9Tf4K|bzv173#ume`afm3G`DH#eQ<q&dVRw1{*DomLWYy9XY;Z~H9VUV$s
z#VPf!pj<0cXEmt9#9z-}@rzpJ(?t0+DQGlr^i|AA)N#zR3Ns<WcM=BgXz|&Ph1oh8
zLYW?)a#*$4><O>G7ZI`%B@?<6>a*a*g~n;fV8+5?2V%onbk)gK7+I5<tKu34*y7mY
z6j_(msFgE+%>3{x{PM#}iNC=9!)61k-lW8&dR&=SMx)e)&sQ3vOmdakPt;EAPQVMg
zn)HC|#4*0=w_2|cIZHx#it~PKWe{iAN|dM^Rwspq|CZmObbIDWaL;^?eP7iWE+DI$
z(W=+-u0`tJx90^G5X)OKN;130I&38Ag}nZd;}Cn&P*Ow<rxvl6Lp8DG;aFn<_gi0{
zPU~kN$<fekE^716IpR6IxxwlybC<6XJ3D*u{U9gj9i87~W4ZfoyLr3Ny|huPiR?Ys
z9gGW^i{ZT%gM!x@p)Vx)vr}IemzB_D04K{vkH=`ZIk|&)`gqOXf~`Kj`9jV}0VS^`
z!=~6J^CxBJ{lV*IAotdkJf9rId&K!*VDuH=vT`_QcB7#%fTj2CNcGZr6j}GLajPiJ
zl38|#6#j9cMUVVzEX<jXQM$>jv9(49tKBaw)e9E0!UTfjCVA$jhrgOWe_?~Fm8!L<
zm2oc$%m^$u&fZJ{vNbWPYJcq|TnzP$GIpAl4T)>3&5O-@%zvKebtZNeYLaQvz9e)P
zy;ZwpzbvByP@!TsQ6BNDP>EAc^VjlwIoUf$^KtWG2~hAe38?cgn~Wu_{=)h-$e}?3
zidW;v*AZzDXxOT!cDf%N8em^M%$Pr$&8vD^bv1nCz-znRI9)#|L`oe@-6Z@a?IulU
zL~A5pFF?<u%u^2q;P9DzA9p{lF08KW*6#|fmd*X?Z3@oX`myT9_-4@_#i91<ZP&<8
zV0YOPXl8j7lK9wwWF@Bb6Z)-*k_tOQwU-Rr5nBguhFSvq`->mwZsb;Y01O)}Z}BJM
zwE;rUxv`1-c>=vT6=tT&dsp*TLpmU-zR3h4`S!zA=<`9FcvMjneg08s^vhDNQr6$y
zNUe$4!)!v7zbBHYlAOJqVyb=hnV6T&=EFh_xCCtIA^Sw+C8ME!N1x7e!~E9eh(n4x
z-|?`=__geozGg-Zt#SLVzYJyrW(0!~fiVFNqt|MHon-^PbV|;T97F2ly40)a<8I(4
z&hv$*3_jjpt~R1Kwpor?v{j#}3abvP1o~MS`ky#yiD@~_1)i6m-|rfeRgi@x`u*zq
z6+@?{+*lwi%WEn9xo$>9hZ)LluUl<BTK?)|C6HnuEqyUsT%00Xp)m0r%>q;^Wn*&G
z+x<Euf?ZjYPHU`T!f5v<7sb`fRTS6>SZBA?E^YG|f-Mmh6LGU;#!0iye<mws)tGFX
zgIsuX=Wy>@dJWd~1N$L$JkHjZ-U{t|KCLu&kOJT*KrG>s?sjeIjnuD*4w|N^`LgAT
z2W`Bzov#)0cABU0Sj|}KkjneQyZnl<eu^d@e(Sj!vO3M`>tW&%b~Kq!GBBB&+5jtS
zR+CubxYqev+xfURft7oQIfrA1m9P~UluA;N=NPUMrDfMN)>K&`)AZGIGBEWnRk*3y
znfufJClYnY68nN@Sx;MD;Z#no6^@PKQJR^%@ul<Kue%1D^r_Z0PyMTq%Ry`jO7k*@
z(jvhEPg_-C9bgHK(A(qRVVxp@+T_LLU&(jFx}|>N?M9Kby?U_UHvc_ZYy8U?t2Zn{
zJvAeZ`KwP7ln0A9pJz~OO}kO?S(=rllcn%!(M8#<;%@p6p%Q1eak8;QA1-6Fv*GX*
zf^-Ujbo<saYs2|YOgj>rnh}fRMpDO97(<&ZiGT)`8L&xT#HssYpugx>(a>~CDYuh=
za~gc;w2nYmL05h{wXANzwqbNmxan3d!ffO^S%eq3xVF&fEO_pGmAWH%bu`?>*3{c}
zb?4P{KaFEg6Dp!}kA7RQQvqAAWSUd_3gLZhQQmqbxVS$)fBhD=+bN>AJSbZE2p`iw
z3569~iGebX8Wzt~eZ72n^?rGdKC0Z=I~JTM#)%f`CAJ7W#82y|cG`8`mob#}E?I$#
zuALtR16{1?SLqd(%vx#gmClvV*1Z*fDN>~yi{gZGJor4SJWSqM?$>32B6{BR_H^lB
zxMUHDDf!35AcL!DkYt3ACMe7wIbAB#({RmUGzLCDA#43<?(b8O%*BSl+Fc@OKk+`v
zf6i`DrUC_LH6vMpfYXIbE?J5+B$PM3mGfpPN1!lzZ%E%4oJZZo$7%ziA{2Q3J;}ZL
zvu}-4Pt+0K>vt0kX;XQ5Bv!;ZIua@}Inq<a88TuQL8kbxb4g@oB$R($e}aS*Xn}<K
zA7vB}$3LG)#QvwwzmF)<0Z3?wU-*a}oblv8OJm$+p#0}KN*3ZC(mPdgX=%h!)!5#|
z#KyrK<hY^;Z$@0eu>Gj%fP_TM@MlMsR(W-TXn)$`lZK;)yqthB$eP3O3&_ZX18i;k
zryV3AumIxJ+QiY27Hn;0;~)SQru$b30mS*A%RoBXe-&}G6sFUVSE3aM*_+VvaBy;P
z(uq8$rKJ_J|6(ejA|d%7)e*mh>C7D+Z3Tcp7Z(=}7cLHvy&3Q|KR-W^^9}IL8+Jqq
zb_Z7*M?)~XjRXC^8~M+6BupHP?JaB_EkHK3f7&%P0y#Mf)6xBT(0_gY{hlUZi~oL-
zjl+L*3(-N~pF6<U9Gt-aY8z2i=+9LFB@3{Lm8OJ+HDb&V&k*6~;uQK<`TxtE|9;{>
zYHIv<%{ScKJpZivk6Zues%j1<_TnIG#4{a5{##!EQTd-Y|D&Q1@K4wOF%<tE=YL&A
z47A8|A>e-rP2~Aja{3E|94Rd%6h9%32rc{bK^{eXG5`CBI7bnOu(SjnAR&n%NlU!@
z1V%ncMOz^3zV4+d&qQa|pd&tL@uqT*(5+x}9-V1=UE8(9{q+D#_MqZwiO8KqkQEQ_
z6YhtfPj1tXH$;1vmX=yv>fC`>u!n$=!6WO$)N6%VmNw5k@r{)J;BV+bf3>~k_QuA)
z<951Xwxy$e@;BQG#>bQV(7&ro`^5U&H&lo<Z4{vd9xn3VY>_L9u41SXf3?jaA!9tu
ziP5nH{Y}UJbh$kp_E*D>5LPwv6a2dxZwVFL4}bOY5@J50n}0QAh=DdoL45_cL<!EJ
z{r;Elj|v$;`O8uJ{S5u%$@DXP>34*G`TiIWv1orMa)}Hq+{l$OtXI)K%zvfsv<RuO
z;QZZW3lgKCrES-z`1tCn@?Wa#E4V84-%S);Zyqr*pEjPzkI%FI5-boCUFiRx=>J>U
z|9^U-f5E@gb=%G5G5wXmgDd~nY7e95?0msB^Kg5do*t+MkCi@|ZgK%pyf%^`iWp`1
zOB~|d#N8Rq4s1Ca7Azspl=Epz6Ly~~H|*5n{^^$Rms3Q}=1JACfMpTIufbRFdH3TM
z>xly8nB?R%;9sImiA;v&bgA>CntIJ17Z=9EGaPCw0!BIWNu+(lzwLJeh8*|gC!WW<
zv%0UGZ!jLTxO4ux!_~%r%-DLD`o!Ovu{rENHdsN6CFk#YbC`f|AP)IXiJbn%d%*Rs
z6-R(#ZOrHAg?}Y+F+oNIq6;c`e9-+XM9(<<V~!4U;@-3Vm4f^ZN4Q_p;RO21B!6W~
z%?1BB0)iH~k$>k1{J&rPwCe54GZn^RZCW}J)WXH$f!H+*S(eE`KP}5zOv_qte#ocr
zjZW#gTWfoqb|*7Cswyabs?AHOw)!F$t4J=b?iKRC@}PchFhv%v9M4&n)gYrhn;<M&
zj}Tp&(sfx(IvVXBdbbiTcvL>FAUM=|dn6c=)#rA3V6mMJX)S4n4GiJ^WML*`)mBw9
z;^AZ+V8ZfaEBoI~x(p2T4OY1O^=AB`HyU+$-Lls_wcF0i+rzqLmF$^ZS(17CL3Zm^
zf0A0S`y1{Wwfg*Yuen~slA0-*u)$cba+SN=%R@^PG;9FvPnGpy!DFjwT^HTe1ZUXG
zjLJBiUY5mpme=Xkl1XYmwzbx;-2(o1bLx}qA7Q{x*R9WYMtQzYFbf_xpB;Q>EB%Po
zyPKbGGjG#W;<P_C)aZQp6){HEdrmR8v>n~L&a>T6#FsORQ4t|z%q=g`v{K*eI+r;S
zs_*66)&&}Uci$%`X2QnR;k?=&3suyjA7!0%C^@~`;9{IP<{7>J!=EM;3|NzU1m-@~
zEEfp=Po2vQEWAFo6OssC;s>53@H#xR<6))U*Sht3i|x*)CpF#croVRHM$jF;8h9Vw
z86o-r=6Y6TX!w=o0fFJ=dM$3Vzca-{I?5+MYtN>-ENMDg$vk>94BD;t5M}YNhjMB4
zhErj-oh>QP4kYiy?&@ogFqK5HvLjv|FJObqE4XKd@AjpuMjJv6e$aCIUX)u1%ISB<
zHA?(X!Dx=dI{@Qh$!(u~f3eAhm8?rSRlrF`c=`oeHghF?gwPmA;|^V?(NyDpspX;c
ziMsV2uoZ43xL1&E^(*1E$m~{<B}db00EN{M_q0{O7pof11<2#Q^HIY#aawP5&)F|7
z#Z{LjK~`;7g48$nmV?(h+$uS9Zl&BE_>i7@^Xhvm`t`tSgmF}2@aEu2mY>dah*RHx
z^FPIM^JG+VMmWOEURf39LS=G&K2#UxrwLr>a8Wkj!3Ga!j3a7}+a8zD@>?(-euO>t
zMbTTY{PezTN24MAeE1OrJx{f1+)a8(rp^9BUC%Z5eiH~`5yGJk;b=7fe|+x%REQn<
z9EtZFg2t#*0Cm~_XUP4V2LaXbAM5tZ+Lj(pBOo*J`d*rR<7udnnx4n2Wk?+j<Zcw9
zyX5<0xw77x?sHKqWsMW1+#t2>qEc2VgqgDc@+*lSD08ANvgnv+@}x5WyDn2aKx?GI
z9kD<BhGD7B5aHj+8)D31XK^|1C{eum5;P=ywGfbN#gB*Uy+xPtMu0`Q%LpIyFj$Np
zz4?4hc4<CNU9SOeX;ATBL+(wATw}@D{`=PKPnmxLo0#@1&e^AsbPrf?%k5E9;ep%v
z&T4#McQoZ-UMs?bTH4dQI$E@TxZbWSsau+RI0=Li%mrOigIADm&&L(9L<L-qtj>mb
z(=T?%drecFK8qLBAm;P+a@!p&1flw_4>O3F6Vl<k*+`(kmuY(t0lCYzyH`~dR^uaH
zPV~CjwayOXsSGAynlp|NHdPe4&Iu!Z-685Q$X-|7>UoO*)RrwK_}Z2Ky_&l2q-naR
zGlj5QweV>d!K!V)bIZk4RjF<hloAK6{&ChUy@q!Q+$jn<n5htX;V{gfj~DoXSo4+6
z-ny#womrv@F`|pd&#!u1+%EP^5Kqv#QyP7MJl^UX7b=hWwP6r2&b-HD-{^ebcz!Tj
zwd3aX_Dw{rP;ofQ5CCpT*dz%pAu~<LWF`tiG1)=L`*Jg#A;&Dqf-yb5A4p+U?j4mM
zQ>ttiA({_;L)->DXuH3hzx*wZ<Ldo-Fpz6j6ajnWPGX<gq9&g|z_!j;18L?m-ub>K
z45`f_At3Ea7D&B88s)UpGw^<Dg#hPD{;h=Uoiw5S0j>S(FM5FI8opi#uYk7vuxp+I
zZNXGs*RkcW;#i^fOi!>nMnmy?9x5b8!@SHImz2}sn=W|OzjoQ{c2>{*aFXXqiPCp5
z!GCE=MfZPdN^$+#cQ22Fi)nq3-xaLt9qJO~(X$CL_n)_p=NUI^DWJ}wc<V#%Cly5R
z9d8j(?cu1dX+k8bK2E$=>O7S5X$+xIA+?PzySedmX6c@v2|apdco!YVvy5+E{$uuV
zn&2lK__QU1T4xo6PP;xewzyx_W`_$5Agm%Ekd}=5Yb;4##TS5VADza!9;%~s;O{UK
z*NP<XrvgXsn!Q(0+70p*Qkx?~y%S4n=LU9mPE!m!0~lIhSB9?id=lWG@Y-2Z3S-x`
zoX$Bax;TV#O&&F$DI7ByQKtoqPV2dsEP7roIwJ)B3Utx-cr)Fm_4sg)wfb^ItRQ~t
z;d%rz*Fyo)v1k2G2h3}LuZ6TkqF1c{;u<+9yS<zTd3xY3Ss}1L?f8?byqfAA9=XT^
ze4Sxr6UnECYnU&4gVL`4sB!-wFWClW^zIo7F54SEPMZb3{i%{<5`JpO-|v<pu&D$H
z@610@vk1|yD{x^xkhhnPny>|3;f_;5?oNB+1C~9`!lIaPJLuAKInYg#nwy6O_Gd<=
zXsm;*eJv#xe3vg`D{O(<T_GCyAMOT3!r#a&6wKqksggVAXgV;89dXmC@o8)3SI$p!
z-AXKcXWD-(&okRKEPSPNyH`}^u^7Oa#Ap06*4u9Hkb6oa+g8+Qzj^PEVnvc*dRzFX
z%J{Ukr+XMiEL{0TJ+d|^q~=vo-Ga=<8?37u1wpOTQyL)dWRt0*A@Z)znCfYOVbOaP
z!Hb!y!-5#>s?G?(jLEg%0Gt8w6Djx29|&NpJ_cV|A;j{q?gthEHJ?g8+C1-tXXNO&
zd9CKx(kT%kGxmk^Z;=XKZ^U?t=KM%@j6x3*E5a+N3muD)LhZOUwr^<eR04+d6G4R4
zQCac&nmj(aeXiLiU0(6k5D6-p;uaT5dii1!_QO%P<#feZswb=U@xfixW=Lu~C~ZT6
z7WgD|_BD3Is*kBd=q`=hPNvs`&*_lp<MoJFbJ=W1B5|7AJoiu}^~IFB1ts6-k=y(f
z{lio6(ylU2CxYx1^>Jd*;2bm^x0-D0?7IdW;r+uCH`jX-qk9s#gMy4lCDFxU*So>;
zq`?~njhInyeH*W%xOVQV-a>%Et9lZBk5O#C6(nHZJJO2Bhg)~$<?3;T{d9jNY&^=N
zDQ!^AVeR}drVQ?O3v7Vx!{D_&78OmG9iN0d`fdV}7Xo(=V^tL@Nc6z%>6njr9DXty
z2Vw^Cj_8;!3wBc3G2k(2CI&?k#(P9jvp*hS%f};xyg3@Mk*Rz#>gikkFbt`nO{AP%
z_-ii%Dc7zl4g%r0f#C19W5{Ak2vj`9Y^U+n3UXQtKfx`nUw1{nrgq)H1C54o9wA<|
zc0}YBy!!k_xgWxpmJ-bj@5754nNU?cY_X7_4h{tFdm(fgW;#u7g1JiKg9+PgxmvOm
zy*u$8+6Vk>CyQdvS{&qP`n7!;>HNauu=#9Q&wOgfLls1ZX=pNwzV~VX*Mu^2c{5)B
zsO0t~Eye|=U0$bxYeBCX8+=eOg0`~fX4f!~>T`jaIK1LlkiQZspmxUSvYSMoP*;48
zy%oOZJo<rT7SZ7&P^XcBo|6gt1?4U$Ka{2H72JK(fBK*+d!!PhLOR1xyzNx>$H*J)
zh;Orvj)z9-%WhFlhf^to#S!o=?Rm$r!-E4^g<9^^oiapvQd0-gJ>(2~Z8n=Jbqu<t
zYBg&TPYztc!R$e(Lz8mg8*lG|6du+X5<97RJj>knJ%GNVDyT15j2*SKe(m{kn&i0)
z!Z)0@FDfNML}i~oxqvE*9IT@JM5FO2Cwiaw5cDtAKL5R2fG(%^#5F!b$jG<beJ-vu
z5T|Ue$uvlc%P;e*0TD+rFLkJHg&T%ZK7vY2M(_t#gg%cLXTdA@jzl9GlU}~$zf(3#
zxt{$?n6tKD3i}Rks&-}1V*0JD+l01$oNT<?lIZ5$*NEOTg$Os&N*@IeZjKl9<vvJf
z&+Y%o%T-0Z`#In!uYEWfRgMupA>A7KIquA;Sa&nub__JFKEPOG7Zdw6nGdf(nnt+s
zduIg5oYD!Cw*eS}XI`u-;ojL%78z<xH*Kf6`>1;j;@6-n*anT^(p7#q2p~#8D8)QN
zdtW|Fv96tM{u$1Bl54i`RGIQg@9)Cz>55v{KLvh9%W{ylbsgtpJd{%hsdaIF3@SKA
zSe(m2Uwplea9Sq5V&=*XP|2y3U$=S@!lkl1832hdtfArYJ?!g0vL?Wp<Z(Z9G@o=l
zypi~V&THr|j@xwsxfxHt13e&AiBM?RFhB7r#I?$@q_P{oHOPwWZ3lb(ici1^n?{o{
z#GmN%3Ch~TlPemiU`1=gLFHh8I95OJ1@!fk57bfzA@$H3G^xaQCsMk-D*XW0L0bIq
z#y}Va$=h23wLA5a6`V2{hyDvUf-=FQ`s`OZfLqTwi~MVWV7ITlb;WKf2Sd>)6P*)P
z_5Z@$hLgMKk8e-J7}}S2l(P?Ga75WA;!uqP<uA*fc(uka1q`O&7vL&zOqvDzm^a#H
zNjtYv3UxvCZP}Z)E@iF8y&zW6O7dw28$(9$X?m-ZeEk`rvtbj5;c*9;#Md_>+&Jwt
zbZsunp3`-=<&rWKvB)}5oe06!gdY5TDH_iWQOBQqj#}F6UJ;icXr+<Z+T`%*Vfkfr
zOi$>3>!9wjXIJ~;K@gKwFoylTYhb1xaSYZsYQPgB^bK7RR-qqzsvbL`PHkFUMnkv}
zrs(36k!(5%+|<Q#?W8O;)zi@dL_RsHbSv!JuR%X&w5qgY%(}y|co)it&>s-GY0zIn
zHb!sh83)k1Xof_4Rc|l5=1WlnUWfh+EXKdve5dO5-=SDUWPJsd8OzS5lJMg5Ub^<P
zx{bu)HMRa4g=*XeU@4s$7OVYQLZAE($>BM@ekaK7{POb${!Hl&RAYb3g*UBH`;kP$
zRqN20uoT;V{dH8xhwYA2@C<_9E@oM-iKfJrc2uoh)idzZ*TstaJr4g+8|B_kb<t}N
z`IqlS>BZNszbYhX18dkIFLm%~%i0ghCpah4mGwyZU1pnlOIN-l<Z&WeueEmktm69}
zo!n8#)Joj4z-z+|0kRL%4e!dcI8Qz598E{LZN`M`jWX!kU<33Y(zZ@~g~%0eaYW#>
zX!?Wh%9gVo`9d=_rnB~gjl{2=WG^HY;Ge&(0n{b`P?k~d9Pg?hn)Qqa8+cbM>QisF
zioGRM1zk9v*AWn)9eXy|$}SNPH*L1lGcmKLl$qpDc|<cYb?gF&h}Q51(3~0x1SK~Z
zD370*u~csekn)e_w=v$uo9SxG3X`NQYt2^>K!RFR`Re!Dj;el99zweG)b#a3mctq^
zCl%9R7DtFw9_vEb%1h$O?TfGQh;%;2x~#q8bV(aBn0cl@@%M#ZEQSXGW%~_4wd9d_
zR9|qO?C}1-S%%mxgP67hg2W{iBO$DH4AD1g4um4fv213Ot&P4gEQp*?OCKP|VM;jl
z)M0`b-6S?Dp=!~sh`fMZ4<JIAjgFP+W+EMA>Iny{=3kJK7=$aQ)Wg0^P~Y41oEzB0
zR<I%{(0>_b@5^@?EPy-y*FjVsC+$l=C~n&uq(681$Y^{XWe8?(IUn;%tg*$KM+8P%
z4X$yD0E2)kx%Mac#7R6e1}L2a3kc6et3Ip@SOM@aysnwnYqCu-c?YUr4d?(lr>u+s
z+cX_d9w8?e2p_pr)BZV|-n5QG*yDQ=OuuP;NK4&$Z+x0xcd+`MFhKKHbdp_s^@&vT
zI6$6({VvsQuW;$OULWPzixd5GKjVJO@{?YB@%BfpwFp;R=uxO36Pyc>@ziC-8|`M3
zzY|oep)QvMH!C!R-eF<O+GrpP!usetGZ$bV+dEU7lNLK@Mt@~{AXLVvaCgIgTEp{|
z=+i*8z76|1KvQ4zqFubyf-j^W)_JNmkU)XT2|WxG+^ZMyydlI^7Ac35P|6yWSKF18
z5@SN(sZ30F#{I_#U)Fcgr_&8@iY)b8m2Kz^nq`=@6gA(fAHnm$vtl;O!tLkC=DDdd
z#1CUW<$YFzv+2$bgBj^gj&r*m6{Dhe<}S<jZStXvb2M(lHWbzst-`lmn3sDYpcy6_
zL+;;JM|uu(J=&!p&(A4e7^+s@8Eu~_wpHIAd&~rQJm}v#ce&V*t&&-X^KX~eOz97`
zR0+;Uh=y0FO=~smlm;GqE@=3`n?O=?noKDm!Q%#B_DT2CEK7JhtSi~7#&ESIw%K4|
z#?HA{{wj$e^hUKA=zP$YT8A{y&`S*-)1!jth0*XX5$jpKocEe8S%CHE-q-vg+fqmF
z&A%mQz=K#;m;9z|LTlTvjZ>Gez?dzR2Q!)3hmDG-Dy9A*j}WsHCjA-hvKC{S=Bo;<
zhYg41%(E9f(?Skf$3)Odr-%By#odm^hrIonUyEj}3rQrHLLTQ2OjbLPK%zU-{E_Ff
z^KS-$x{k$TICErWSRS2e;0;X*fep(Mn{fU|gERJ-L7oLG-F^P)$2owsC~R209Rkp7
zy;<9z@|-%Vy_xeYISb|TQzfc%)qsbHesk_eIFN_&j5k+pkGkba!A@ggJac{8;dx%p
zH0)g<wpki}`<%J~v6j$o6+oj=193<-1*o~=ipiDWvqFWyw@yg$QqQ8ph_LzuwVp+8
z-1jAhQ_HiNUf97Pu<al{*aT&5qic4FHI?9p_HwJI>Uw)s#m)E4LvZ5Za$6g)D#d}O
zkM#4?o-m5-c=vqvVD6o_Xq^a(Jp#nQk%fNWz84e3FJjf0wW&64NO`>57Tpp#Ik9#x
z7WyrksVk>tcrV!H#9lYSB|Fng1?@SdKLX<jcw7A8<cN^kPKkn1iQtjBXyj%4==M(|
z*M(D$jwnYB<p73b!z;VE<0?ceG&SAC+Tf7+es_2n_ol&W^8P|Eh6tq77u*n$)=Z-T
z!ttzs3xV-HGXpe?Lp$cS5P6{HrzRa@N#*A}#2IJK>7JK~!Z$nFOWH8XL%|!B2{PYj
zSTI*#x435S%@Ovxg|TuO{ozRp9KfV1vE9p5i!#-;a$H5XasdBoMrk{~1%o;$?WbZ|
z@Q0?AzL88%G=jZqL@2EzUK7C2J9mROI2ct|$kkV-F_=dslTZLGh(VXqdML;b#Ru(K
zEn(yc-y7b*&u_;m4S&MC_z`XOalP;>(IY+uxBK3`yEe#0-gDcNS@J!1h$eRIY#I7A
z?Z;`di0^V!7OVKc+}4%eEavWelYDh}vtO2eQ{bjqDS&c}c9gG>aTEY#WB<*&&IdIu
zEYi@V-!11;q@r<|u$VZ@)DNtg>!o%Fcm;!at*eH3wnQSC1=4a$`^WMSG>h=Vk8qlt
zi$#}Z-JSTAr}s&}C0VykxVp0O3=8sGbi$Qude1_o7CLB(5+iGyV_ctYhV#s03mzC8
zW$%A)A5;-xb<^4GZ(iEgTz{7we-rJ3c1U*TyN@pmSnXXY6eQEm4hzNf%xu0@LKK)O
zpwZl1KN}IfDmryMoSU4!pSY-w3l^^|QE>OUyhg;bEknZ$ozhNkbBvu{einLeeA9on
z>Oy@AzL-?^Z2ERdS>zW<TwRAD=vg^cdBV#-2X@VqYPsJN+Anfj7JVt|Z{2d?Tvc<r
z$CB-ugLWLoyX?WWa=sblOmM%yvZMH$I=>{q7+^7PU1x2_kl_Zn+f-Kcv$bUVNBrV<
zBWjY?2?;bgjNZ^7Ex5hj_-AxGAtlww#tOf0l$;MO+&EFA7BJg`ibp_mD&4K+PrHc<
zQP=mZIjrcV7Lgmy%D^F!1&@(M0B7UkgZu+T?K0OXTmWI@ByZom+XTG0dFRq$*Y`4z
z5QEq<7$WN<3Z*y1Wdr~zJN(V}5{$z!MDL*{r?|u2sa291Z_QH}KmK5f=t>_Q6Y+f`
z{mv9d{j7^r3@5z4BG#UvSB!{2_@Z9RUq^@@$GJqaHo=1y;%@>X!~Ivd`iqxAPsSM6
zh!BC-&7s%rp&5dq+KpBSt%V0_4w7D}K4^aCIZ8M%^kLxZ2+^SB+Y5rfSy%<j&TE}X
z>!rM}A47!<vl{n*!6VQItPgF-FLe7aP>vLpc9m)E@t1<+r<vPnp0xPh&{)t1gq2^D
zBrbp6YieF}Jrh<anC`|vGOG$EV4Tai5^?PKL1vxq>r!{<ysOSLceQvLsq1kRr$EXx
z(Ks|QwB<b5`v4m((_@~pxKZE<(^jdUesVT5S*Q1Dqc=^Eawk6d-sui_IhvtpT`T&_
zs7mnEaLnoWbrn|qp%K&Aezedgg#P3F{s>ybyDj}@PU~iZ>lVHpyX~uZ=Z9UAdBL(a
zGvO5T$WzXa`mwKr^VjNWtNvh6>rE|{=ZQ4W(RRh<`cC?S14rf3Ywz`Z8mw9r=f@%t
z{mO9}hhCZIrPWtHKKO+32><EYQ=X}7-u(k%9rvuF`#n@<eNX<0<))eFkeT_iHcK;o
z`|l++?R!rlFeC54UP}Aumh-!!r)?0pSL|WqRk-ts@-1<=+T8WU^zuRV`f!CAwfj;}
zBW%W@G>wBs4wNl*Z{8V*b!F5he7V@qGoSY<0czIz+|X}P-&trz*xgT_cd@#5HE)0J
zlpS5Gv&9wjA)(K}XA?oFDlj?fGk%w5A4&+~RAA91$%OMfk$plRy^lzbzdjj2CGL1N
zmsRZquYbt`Uhxe0X@khm=w+-Gw-XEZXx4Ffgm$=hd$w4kQQC@)a=YIDg!rMOA85Qu
zxulPYtB%n4p5`*2=6<AkdEMM;4(W^$NWY<8MMBy4Sn$@s6a_2Vv-@~0`JzAfU8j@<
zEIZFz>*2P$#{@IX&G@xzB3&H@5GV>&WYTfXF_kkN$t}?t(Nd9}1cp?-J$MRu0X?%>
zN6F~K76eFig8D_#d3<jX%LMLwIi;waKJk63WC(9+z=<<gUFY>dN*jW`Yi_fvZo-qL
zTM}Dn2X)Dz!%Lmkvg@<kRsg3rjw|G-_Hu-83O_FSl~vUqB=#H|)AUqFxIVi+LZl4d
z-jFn}Cw}{h0}~{euRiR*3bQ}AfGdUAzjsq=?KA6&;K`{fH;vbI)|9>;Ag8`j&mp-^
zl-5FR_b7g+$Jm;K<qGY83u89=bc42u>T=Q5so8Yt`s6j0h4%?D-s{uc&~{(@*QUQn
z)Ajo!HgIlqEMht|VW>ii{F{VL`=w1nTCnpa_v6De5uK0pd`csu=@}0zozyocVYWfx
z1x+k$=lun2%=xj|kw9D&7&1A{Ge%uu6GBli$F3aJNRG%szpurx<`hS>>e%pf4ATkI
z2<M7^3!t6^%|#*t{R<0l1Gy_6eZIgFK6D%2Ka^w8l(?7l*v&Zq7>m3_Gx2zC%@Px;
zsti$n=pe;A7TKH38SIF@VmDUh%%?UIdNInjD&`^W>BAXZ{c^~)u4@s&%54y^+(go#
zT8HOcY7#7cX=k&Enf{2uv^#i<^w>YxJncKEHT}gx6~PzZr|d1#MK2-9s3(7W#Ba$p
zVeB^hQnIxu7t%dzV7&3C!&9MkJXDAW?)RP0=HaG<Q`0$=Y9hScQJ$>3oQC2}`k@i6
zh|N=x5ptZla*w<IK$cj{J*lql=4r!zkjQziyf0~_TTny=x_g>uLGkG~;foFa?UeOf
zzG^9R<8k6l4tj<25;&)?h{D2aYRN_UCTK0SwG&KrZ1iYDJO|z^2}=@Md#6vDjnZ7T
zqDY;WD&9mhx_rHz4q_{7w(pET&;6wk&ZDQd8<ne{YUZhSe9M$pS2;01G|^L37rGTi
zeLl}DIN?x@aSacDooDb0b*Zcw_MHk5^=-wHqzHMVs0o<vMYs4=v;vH9qZ88B{wH6V
zV*0&>3c0_zV3T7+)v;@*3DXxCw4h^uMu~?4n|y;}X`$ac$%@Lt1YgIbV${{LcMaA^
z@XI|LF|b}}U~mdq>qnJ<%S5(+(Rlq6SBL!+Bz@BB5JcnKfuJ4}J^YMRG0@qB3_~S>
z7Z2)eAHT9hbajp)jK;VewwnD5f$ow*PFw=BBxj)OWmd@dZcdBoT&g0qK7Sh5a?7F0
z4`%Ve2i)^h6AF}UxxUv4T^Ud4f(*F1hM7ZF%5ptoYC`QqUFkKc%fEN|S=nk77;^IQ
zt&L=uCOwU|m7l9ckWeW^j4)fPo#sl(MR{sH+%3rwo1?4||M6?#cE_2v<vhYx_*B0;
z_At-qGn}3;YuWd6)3qKv<gFRHxgTw8^o)Of*rs59an-4K8l&JM>=iS#33y<jdMK{z
z*y6D{01IAD&=*g8IJob-z8Hw0zRwnZoEUT-?kAb49By352Hu0#MB(iYEb|)Iw}a_8
z6>Zdz@{)>S)>TJC+53iHFj*qcNiDuzXYgecsCAoZ;<#-euj%g2^5BOyN44y5(_Aan
zL&>4%3AMuSncg-*YR?l3_qkx0O)KOZ39eDpQTCN+EzZ(?U--=NvGX!l-ONCdV&#`N
zTlwjS_jRJSe2jC}kFE1IjhCXALuK9XZ*ZZ3M_v;Ox4oh*A3B6_X1ZIB-2{{k%^s;A
z3@31o8h|rIBzl&jjb?hTXKg~qwje#g$uZ&~70UEdPs<KRl8;TA4r;_hKGQUK-|QR^
zJer_zv@A_&xTwwBMI09Q0v3gDPdQ3N?4EqA4|md0;JYQgUEkU#O6ye6yJkJwMKon0
zXkDS6S(;)}dvv+<JHF*7i<j=4*OHHD+B@xMMHtQ7*DI(m38tFi#!rMaYNau8vUnu4
zUL4ThbgU6s1Lut7pqv$*r~(AX=mBe9%7JIU=)?Is6bnuiLIoMUa3@#HBT`owyK&9A
zJ6@0ab$|9FTNO{FeYMu|>x+n8_a4{bK#sVFL{kEo9(~R2S*69*O9OEoTmUMay%_0}
zeRvm92PNZa!pAv8zQ$8yrQ46p@Ee2gMGg_lHu@UOnR9zKhsp@}Y&J2H>4jV$9!;_1
zPh?CE^fd6XafQ&4k6OO%3rux6jWnkz%~hp-;sB2VKXjzgXV86Ia5UqpIMV=E46j`4
z%Y%-W=f>|jY@`!}!Y*jnd^a}`Igw6%U-VEO_AKtXy*tT+r0TAKnzu_M=<mTiF~#VQ
zeH0s*cVf7G8;nL%Io(^)eI|Tj4L@%%Pko)3^<$}Elt$WJ1&AD&b3lIVl1w5adm7}y
z6*Nk!%rMe*^PFypGb(~Kvj1yw>-h1j<#8JMMz^(hZZS-_cz9AUhW7#@tncL{muQ8V
z+a|Q-7+<p}wD_x))(<8zorHMciNHGOSPs7qpb;yZk4N?H&X4vmKu+=W=_Lj{P~iyj
zw-^%Czg$hzPjVh5P0zRu_9_SWpxTU0s!E>v)`}d0DKCf!%;frV`8&FEApGizX+O(^
zI=?|KDp&Mi%mj^yNt7VgBJTNaIpC7ExPZ66bAm)#wiZD)aMJ<Un0%owLWh7&Wf#~g
z%`xU&b2+*RVLL7qCuZurvNA`{3lT8$5JawE%OY4(ja9uoO@ECu$9jI<4{33`V3kO8
zq4ebLN}hPE*4;@SD>d{XC)xY3_L1z>;q^%@4#=08Goq1EHks}iri}Dy8~cLiP<T#?
zZz*+RGDn?(%@z@26zYu3ZP8)6>9BU~p+|o($9Y*Twi1#7y=1lK5?R7HoZaZ9MP-hF
zDF<l;q!u5a6`y8^tkK(vWMxK(H^t-lzW0>&S{24t-{V>WR`cSu0}y$GR=Lg5Kt$Xr
z`dg)(uyjkeltOj#?(!Yz;7j!}*KdWbB=`);6+Z70j-M}1{?2|o{q9P~y<LdGsU<83
z?f#R9d4@>-asX*?V>)>BidMMq>U<SG!_ttYC?uqL$z(<WEIaQTD;utA{c&|qFb1@2
zSp--&15DmVoM#+<XwB169^UO7Eengl95u@<`Kf{9DcU2PKpz|t=jRv<vc&sCLb}pR
zJpXAc`o(@a<x#9HkaQF@t2{2?b2E%M>KDc8>@!NrlsXW&GEDQIej0T;3hD2YKabNP
z=~!np%-V!%bJ%Y^h8(R2CjVq_7{Q{A)gi&=d#BQWA=HaC7a+@Ov-m}>|K3J=Im>+k
z$)KNGs8gPnu+$;%Lbc3JKa!-h&@>+kZ$e&(HfzM_=`db+rP1%&A%F8Kb-zAVN0z)1
z`_t-}(AmfKaboAOF{X`dfcA8h<!b+?;<Q%n2@CKK11VtA$>*|<c^5@+^}z<8+ur9}
zpL~kcu)WNNUj~C~55|Mc&rPDMVv}E3D{+4DLL1M#>6rlRx7z?R;yB-EbSS@)rAU8>
zDtb79cx>0S<|9UaNUe=+`GyuRsg|R>mS%B&Y1g>cUPG2xbw|YiT`;rlx|YIyDv-v-
z;Cx9PNVK@^XB|b5cYUV5;PGy>0XNT^M(Aj8$;SPW|K@lGN2c_8C0RXta{7>>q^*CB
zVcUnM%nKF^>EW)^SGi2GwU(Xd0t$@Y*8J=^`yhLzN!PgPjXY{bzsv!@T|Jr3V$y5k
zCBNM7^*4rJpM=!i-lnXS?SSuGdsFPeZzm^IbJHMaXWLo+#EGI>nV|HY**co<t<;ao
z{Z8e4J0SPl?VsF7E;A!BA5N;ELhs1y8Di<j&f+4{zXv=%SSa&7-kxRbjK&{-C)RR&
zFz+`ZiCB8P*~~gyP1HP7I|)48PZuT_;#@yZqj36?9tRDRl!F5<8cuI#XhiMg#AOP#
zWmvNgcl<P?l9hr}ezV#8RuQFl3;tpsC*;Rd*<ci<<5kkOZ>c)K<v*3*c;-mY<io~4
z;*AL4bK1uUA`JIC-gW|hJ8AfAWC;OGLLpXpt&U(+LSbcVd{9%6gFjtprTFB=GL@Od
z-gh>N0RF*7rPP+}P=tvuWX<!bm!K?ul=fFnd%B`5hlP)%DU;=WSg}$HVtD*)Bu^h<
z3w9kDee#Tw3Y>%u!Si%wi+(yAA|HdmE>WP=namxJTp9&*JpOi_T-oQ_A5m^NZQgU)
z6A=2ZpHhLN^cTAf?raUyF9}CpwWOF=+X@}<MQLqxQK#`+d-5idC=k^upl+;7C;!4Y
zqT_W{<LeE-ReEXveyU)>y=`ahg(y0OYma;i6#hQZA*(VVpgnk22yGyPKwVP7Lkl09
z$u=a#X`nCon2vW=+dd5i5xmf(TaMDRdCzR`lbW?1DRM;5sqZqnmE}q0;<^!B53%uK
zv5#pNNgwUaUI=)bWkn#l*VasRfpm(XC0+7S1nW>CpGH8na#3N`Ne1){IrT^F7?%$%
z9F}P-4Vs;i#Fttce0a_q66B*Iw+96Lr`csi>t6U$W;I>$%2%0RHc(AY$GzG#_mZN&
zO@hC<_z>LQ*x|wq#-h|&vSX90-MDa|LgVbn>R>5`edGZ~6Q9N$ompF-+`iy;db4%_
znSTd}>D9S)diD9QrqaTEdAuicK`^|#QkQ)Z^ozT}mgVs~>j3Z<3~{}~&9^)?)Pwjq
zX&(e5Ve2AKf=F+J>rD8MMPwpK*YiAMw_qGJ79#~08p8j=U)^01+#pged3P5j5radK
zH?U^cUW7CM$3dr+tA>!)y$U@iiVnX*KEoQ8ae#eVts~I61;eNddQHLt3C=&*6zVP0
z<laMnO`prB_^XQ|2!oa)pLS!u<{Vga+!wqxS~jeu&)_-@d+0A(vI*ePqc$OgYL0b!
z60R*%C@4|zzchVgaR)NvUziRV21{ou#<>|D7_eJk7GFplRLmi=vj3GeNRUH?C)K^Q
zvHvcgu)>d6*>V4c1chVsVPSgQi<S}lo1`1;#tjwnwx`Gr;e9XI=QFG_QXdVbfAvwW
z&x6A2UcrsuzwM7+w#zgEmsdT5y0NF;tiXt0J%zAYlSvr)l#Zp}W>qpD9Le^6pI(oz
zvRB@%AR3uOpd;X4=twGIORP+1Haik>m(a}k$Vs{deHOA%iT7cFfBIv){?wb#M_!B@
zWD;ywy?!S4Yqdttt{-D+wqlxLAJMfF<ie(rEeJGpMo!^Q2VAElF0PYt@IQ$i9C6Eo
zx_}i>>r?j+z5y-KtQj{R=Yq<;g#wa~eCwdXZPA;{us-%~3iig0j!||T?4VW|y!)5s
z?mPbTs>f2Xaru4h=ntQ@&Ej^Vyj>(Zy;z7Z8J-%hx_oY5mfof$aN3!9*|>fRi4S8M
zaT^Mrh8|UYIqz_f$kU5w3c;<@LeQFrYRzf{MoNPCkd5m_8$MX^u85M{BEn5I+c|oH
zD?r)S@1g{&yYg4ML{#97rO{;tzOLN;)WVU_Wr9lz2Dq_LkLGSrO3jZxO`YHc^%?fh
zXcO(#XepDm(ZNyhA#^`nXZmmJC=oe>0-AB>z=&n^j)ZE7g20(@Z;L%^#^+l0hW6BW
z*;Rfjlea^fB<Bl@B!@&TE>nZg2~>!H#p|!^B;<Ag98fGSz(<hATbUE7-(`p;Q9-r&
zRmAb+1dK$f-HbpVIlZ5+Sh7CMWyYaCt!@v(wEjGICQZU<j{5`u_-$Yi-fK%s{*KV!
zk)nPQ_2M|f6CpL^L;<vL@fS&qL0BVB&@wtR>DOBJ{S2L%mGf?E-YXG8?WkaWsr$8J
z`|IHCdS;$7Av_4j6Xdx=n`_>5>QjfX1u2bSQjGSm484J_ob)v6>t7Q&F&|74;h0{2
z*21nn?q@H=(Ph~IxRxh2Ay?bJqJ|VO2AZ4PFub?Mx&$41hfo(r8ZbX*fC-cRDJJo-
zkIDu>6$(y8lNI`K(o0Hu@g<4AIHx<&EG^M_4$}N>g>uzZ>oGRiVM%nNrJqNhelN*w
zj_{>X7&A@#v0GgYI!6ar(o1U1j!1QP{*%7SAp{UmphD1%25Z7nF3f;I@&Q$NoAlSA
zZeG(jA|v$F!vf|TfFQ|XAB|bs&S}l4GwUNs^aq6N<eh)t+pAywh*+PT`Dv7fX0I8s
z+{$D<>4O1ON@xdVJWVP{b9hw2P$x!2so|X21ys|UoEsL;AUw9F1!@yOq>W$5vLB~6
z?;V<&uWu?#8YXfJm$u$~U4F1<BYG9n^JqF8iX)Zc|1fc8fmUQ)wXLj}Bdp*1&xOBQ
z@jnUca?17%+yX7Ep#*kerfsi+DR9yAzt;lR*INxZD2tp5d+lX!1lskxP24(A&b0FI
zxo#|E37gur+M3^i^$@|(wGsO#iVY@%YvBxC8X_;Y1v(f86YLf=gU@6YD`+SOQOdhM
zjq!VHw#&IL$%Zqg$6Y3ybZGo2i8h{XfZKNg<~ISOpBlS?hI<kUL}e^4_^L+$`uME{
zG1)!)4%bS{u*wgvPE+%Dl`RSNE(Fg+8UZ2isf}4)K;IknFV|acM7JWMwRu<L&}&l+
zHYj*telkfB_KUtN7s}w7nlJ)X!#Dd`sdu&Dpt!1}=fprgr7}YF<hP7qnAR9CYTM%_
z+DyD};LLj25(Rdg)pml7;qAUCc5P?)V%r9#$?0(NaZOBw$Y7&m>Zs{$XR#N?jHX$7
z_f{f^THAc|y7-u!ZT9WBLW~y4>MBXPrv)3Jx|V<6zP3<Xg)fA{w!iHVghmO9Xz_?=
zZ&^_B!Wj5^=YF5ddl2DS$D%pL>vim!uWMzNX6v;V-P#0dDQmg!alaf3RGg_EzX!2Y
z!CMR=9e%+!!7J}-cknUu{~*fAuMei$!w=Q-JnS61gDUS^wsnVq<L;LX`F``Nb;c0t
z?J+C)dE2UsgY@|H&-U_D9&zW9<(`j+AJYR(k6x=z?#=}<c#6+x1Vjs@FvMsKl#Z!b
zE_A9$0+{**LpPgQ`R8z(mo>&B8aa-@)2B9Ot}XWA3~}M25AXo?-9ZP<o}k>Vg`dq0
z!5B1*&`A3%>PH3V7*sEQrZt`ZifVe<NxmY5EmRb>T=o|W5()Y|P!m%&{<#Fe>7a_9
zZU5kCzAa57={)zePE~OoK(Q<OGr&yaP^H%9%_NXW$3l7;j-ok`=<!kSrH~z?LeOqJ
z@*35ipZACD6O{1bALaT24u~ZxLGId>ifK$O8cA;(blwgk11q}ly&ANjrt%Je-Yv#Q
z4?StY^xoGu7yz<;Ku7{XIXrjdB;I6bMSLEM(eQd-D2?pOchk*qsImkwuGwGF%-~5?
zUrA`XU=oIA61@3@B|Ub~KHO3xU`W1)7|K7)h9DvRW9|TJxd+3B-&6M5pq3i_=%ieu
zRE?Sg<6X1g985UsjT~SlOR`Ooxg;T9lTZ70nr9N!h95CqY5?Gfs#nF@!R_0=Z`(Rv
z^N#MsSL6b8y{<P`%a?g%byW!q$RUwv=Ayqj3<|Gg!WkyP+r9Db!yybKCj8(z^5*fu
z2Iw52g5zqub2B1PYPBB5%K-~`ppg|6Kn`K(6(Coho*2(VM;_La^c8irN&PkobW%t(
zFH9v%n}2cIhab>9H|Q+|wT6OE7Rqyg#$82at!`hH%#Zb*Ce;`Yqa|Y_CXgn;<x^L#
z)TbFOhj}0FG^(md)p@$AZ%L)>p@k~(kwasF1N79WDHQ}zduN)?B!c|}g2(I4wA82K
z%aE)zM;br4h-|~<Jc9afEk_M^r0WN?03u@vToGa5oby9VC?!Xo6=y`=;Rg?MW0le~
z=b^El!Dwo4Y3lpE;tMr&*_o+-`Y-_q@1_ECPh{>}v+P1(z)K4cedlf>^|v<*<t{X)
zgu`lcg|$OzQo_nClmdw5I{)=&o4;jH$#a1zI#apuH%~1wJsA8y5X%ie-Y!3e>zFNT
zi`Z)1tD6}Pk^nuZNGbI<{TvMEi2-+pYEK5SU;Z-OXJinc_%Szy{|E)#RlJq2vssks
za6~LVjQ86t@5E+n2rm;@w}`2GS{W@i;ff^7=FK}Vqn|Z(9h5cym0f!w1+8fcpzYDp
zkFR7`YV-T8k!JNu-~8#>0q<DHza5z3j<+3uAgYi46g*+HTIR4l%wUmNZSxWnvO9P;
zWi*7vX0BnR?R=ffswJ1~mQ<LXttirbz^#9GuIKnzRVT@m?ygha+bZ}$BUszGrFC>o
z7D&!s3HKSLsnY<uZIz%6E~V$IH5sfP%aY_7Y?dThf5wEgsOvClIUUAXx|wR#!@l6%
zmGLMx-%g5wYK}u8sY+@>8m-_sZgrU!o(K=3j&zfGW=u$?jd8<gs9<4_;o`R|jb<v<
zBN6D`Ce@Zm7IQ-m8Q`uVSQ>;ih*8iSrwffXOdO^MSl$Qx=l=qT5PH+0(*If8db9JG
z@mjVcT?n`M^i_P9ydKHpA!9^yE<e+!5Ura|pJ#)=#!UOi{Y|hd8nyck8NO9p<@k%H
z9iyf#t$lejQZDg!aL|bT)sR29{ELz-K$DmVs-Yp*Xb`oE$bu-QvEDiRh%WQ8Q1-b7
zZZh9q>~EI{RC=V{(!R2%QGz$`iWmhgbcN^}B|b=G>y)3U2mc?Oy>(EP?br8BhajnR
z2?8n&N~e^FfCxypbT^w$>5`Ie0YSREyFt3UJ2tR^=is`Y`@Vnoyfg32`_A@{?ToYC
z=Xvg99qU-@`~56YXUCG{D2+|_2xpx?;Ts^AwLI;PpR=~mf>1-d<BW8kmRF`EVV?Dq
zY~9V9GJWevOX?hxq=5L8fb^fA`Q*-26$h2E^HEfOGS`2ou#_F9{#H~|fpbJDRcFU#
z5#vy_^3Bn!CPtPxXs~~C-0qXb7qZ)$WP#HxH)|?|v&LCr!G%!W(x3N<_|Jc-aaxLL
zD##^Ki{*f{0jhb#*+6e<(*9PIzs&DN@uh9!z}l5T5b8Y=hNtW?;Z>C}rJ0GzKkGFU
zb?Q3n-}M?+2Z|e2M0#CG`Nch=Cd|^5Zz7!iSVy-Yz@|cd<sDC1+@V)+`>L&2#ivt|
zJ~ZNuD%-H0JadHe$iv3M=#TbYhRNgp97*Y`YWt?gbH$i1_vYXKb*_lx)S6^}($7DJ
z$ly7!$JYJK=@60EiYJBbUQMIY;*x(iNUbf$>_5#~B%(UO>SOWVigYMz($7l^@yXFi
zhV8!+TR-(=ylrRKY#sf8z10?H(A~$AN>~(PmEzxQTm6jO==7}z<?Y+ewIO#KMcb!S
zo$-w;iJ%A3yw2<QmPtmoD@{onrj7?&IeJr2_yvzH&iF$eaDbGr*8xT4ukEr1WGV4f
zBj~6WlII_#s(C_-d<c&!vzq#Gd}U~C{gV5F`q!Uc#xtsNuUwnHR9P_i2wnLXYI`uL
zV)HV{h>!Mz{_TGkP$pHzxInW4txS-7Coc1T^?Ex8J2f!X?XGG2FxN6TTyUqPNA+9C
zq=x6;?X{!D56=TOUfU*V_%vT|YVoP*C5S7m5keKq2cyNEwJLpt{F26HxGe0UHmF~p
zM_d!`Zgi%s@SIq<;=-6pS`ftFCwx?Dv*{+^q8Ip$TeaZ47S{B1oslNSIl-XpcSkud
zR>RG1K~(wMXKcNzj3l?073FH%@rd8X(k_`-*>{g6(C?wfxkkgO9_Tu8@I{-$s$#!s
zsy3{=5p6=RSjLJRwR>*&NSL7dK?5M$tu;mn$Hq|kjf;yCoo<5V*x}2m^p33d`?+kL
z;m`Dpqn0s=A?k_<j01E(Hb$l1)1jDLpn?!^&=Vj|bMZuNT>=)9g-pVsG6OMnx&%&0
z2VyIw^t*A(!u92LV4UL>TPbTe6>Al(Rwxjr^010?H?m-S4@;;naBO3|pJ&APgsh{t
z{h-A>sp+^XFleohplvPqgMmv(<MW}8e~;uw@0_EZvRS>IC#0%rF-J~0q95#qL!!s1
zT!NF|l6zI%-iY<8&Yj`^-S7(w%QFKO&rduoFHK|lKn(!wbxJ6es|Kp8-rcH5o*;F9
zPNTljtG2ZDxD}NK<Fr*ON}hS>@HZ#7F#fYvUgGj$Xf??Uj%_{a^VbXQZ#fwcs(~9s
z$aXR>x*EEqiew1s$V#RXB32PwLc0TN<@83)lS`VEMcQS3=AV^vJw*-eFIXHZ4~q|#
z6KLLJ%Hn5;I41(SqoZzajCI^Q7c$D=QO|5S>Dx*IWs4+%2$#$_>m9ZKIez1Kci|ZS
zTpmt%b%mHDg^SuePA7DRXU7)V$3gV+0~XG)1|0+NE8e`Hh9ZQqjQV6nrDB$=Co&|5
zGXwc<p6A28|93~8)+PEhw=XrhW8v!GtJfj_w5*vdU;qALhDJkPR?|NAk$oT5b^393
z8J--qh69ZpG|S}hVYM5DK(Fk2cQGYtN&G$2`T*>9;#0@%O!}YgpRRapX9NTn^)o~~
zgCw>`OcKDrp*>KKVqdMm99IY>_y^>=*z?ZpMdhpVPusM`CLn-Qq*jLN;=U#}M0veX
z=ILMKPwv;J@;l>Rm$l@pGALrwfCk)8!<C5EPt`B&VP(cGUhuS^N`gDH?dLaCl|ho;
z**-#YJID;Jzc1>~;-h$Q`YmkU;WwNaA<;l^chDh9Q=M&Pe?T4&PTAmm4jiYcxmL1I
zY4oaS<hWlZn(pR*8K!(ZN}s3jiCgiu>T5v1?s=p#wv`t=hec_r80?pim5nN|3FZ++
z7%h6dvXv&3JjNOhvN|Jx;Z{uvxpKZ&B=MA$z74B7Rt{MnW~yJ`tBRw8niM>`WG=tj
zbVX2Ve;H&|6?ZL27M<>jUydz2iK(wnT!V&OgaVhsi#PO{Ywn%(f8>3HkZ39v7*w0{
z=$KWDGO%0gO@OEA4MUp|4h0)B<L~F7?rMs<wzkr3yRLnIN@yiTYPQMEm^kk7`<n$8
z#<pdGu+ZzMefrKjD-%y!xe-=$pXI149{kV^vV*F((!3i*#TU*vFgCUn^XiT}>ah8-
zMftJcc3LOGz=DFe$F@PX;|`wR4UD*2B2_sxP^kK=w=isx@m59ly<ZI7m%&vl0fU%^
z+KDgHe__)Nr%2dET$r&4noJm{b7Rsh<bNv0KFZJ5Gu>kf%8HQ%Iy3B?7MdNW#gevB
z9_>t8TS5$3DrCY>6mRz%R8AZCkvyTkI{sx{HALZW=^9IW<dUcc-(?m7vme_AKhxbC
z^8PPV>wmS1o;8^cgRO12hZ<@ocAdw$7+X|u;hrkV4-NEjY6DshMNZ>$DI(WjgPO8B
zaXHwrlKf~e5QNAdEpPcJ>6W=q`&FnY_2(g5Dfv?dxIZ^=vAa}Eq|ex7XbA;%7~&~n
z5C3dxlvkXTWlN_1lVkpXFac}O?g9Utb1zi!mM?<6ul)35gw6(w8=U8(O#Kqb2pg3w
zVxCMX%^417rM}{STk7lmybtptCHX(BzQX$GZ$BMN8ohgMj^0d6^@%s`1IAW+Tu{`v
zENcKVV5-IHr)))*{}RuvnF&0AOk~XFrN6ymBabl|M>@$T3EOH*#Gg3>SF$9~V3nsu
zE|`}OK9Q~vfE~BfPT5gv_b9Hlb?7(~s53spB6-`>O+Jbh$Gf`Kes5WS@bx@!I3+gL
z_UpalyiGS+0SzPedtXv`T<xAw8>82?%cemOeGJ7Kc+9K#PHrzMd+Cy%X-X2UMR%!W
z{WSTN5M1PK1I>I5jaQ~Xzu%~|uxQo(V-VuJ*;-WDHyJG_s2W4~-oq$CFs>-;UKp+W
zr6@OH)w&%cR%d)`FU{eL;%`j)pM)=W{}#NQXAJ3B#zxSi(=Day_+lR;#~6x5DaenM
zv1{??`lXyK61&WC0FJl8Wm|j@rnTO?H`&_L>@QLZZ~qn0{zge1_Xq2HPG{U96Nvp8
z-bX6Qx;gYZY~8#yNOF8C-TPzGNZakp?sbK0A9;cD@zp%vgR7Ia$J%wiZq4<5K^8J$
zLUZBrk@uFR9YT#srqDs1!}*|9TK0{|PgPUsIYxMd<D%W|gc(S8iGYn}?I!wRr*K4O
z={4_8vc5ul_q((QsAa_>otdO&-5}9Z*z(E5-sv-#yWkD^u*0>|jab2-@{jht7dzLl
zI=GrF=vpaP4H*eud@Xv3q}P&MmzrZUKL02?^<V1cjWSaejN^^PW*`B@x*w<vrWuQ}
z-=nT9@{U0DZmlu!od(?@^tAEEn!9R<!QMv?+Tj<!GUZzM-n?elPbb5e{1p17`JLLa
z<n*H3nT*g27a)OyI=&;|#_^z?8$NN4jKG0LZ&?EQH8lT+dr^)MzLkQ12;CLZ)16OE
zLcoM98u_s*P;o~AcS@obi10GwuoG9`*ByUi)QNJ!4W(n!-gu5ZMoz()+xtaVios_+
z2>81CL!?c#yN0)7>EFA}OuoShTuL%{0IAlMfsdeVLvg}6K0M3CbBiTYjY@n&o95%B
zOLw&e@@D4RH09KmC71R{Oa+(|6C-fR^Rr9}f(px;*0Y0XE@FmlVD~#(d!Ax11OmnC
z+6Onk^OJ5$ddjKDrFD2gi46gXM#WyXOW)^&H|;j@sM4^0>0`<3Rej`L<-&qRpu)Xh
z^ZJO8A~Esq-|hS4ZGllUL96(2Gf>OWVnQn85hEDq7*9ZN;GgYXRmIQM%sp)j7u(MO
zEzqQ$4Qp5uIMKTIAE4brZg-+d_5Zce;~4JVub*z6)RYWMo_6?JBxX_eLJIRHcR@i;
zTJdb<IVDM~nYwPa*V}>^ipuW-5cmHTKCJ#NxcxuEM?}PTfGycqRui5l<5K)$G5`DD
zdYgX`&r3lKgTu($#J8d^O|EwSle!}8j$}&+d$FjO?#q(9v!8SKD3SE=dS^dx(xF$z
z3t=J<rvTdhHenuoXDMBxE8xQ9_lKXO`nbxChy<*h78$w&Q(e{bwjpVSYy8?z>FDc$
z0kZ6ewe>+82FreYZ=y{XF~(%oPq?i1uIqJ5A3VATHCxPJx+B`fKpi3?9fW|TULb{9
z?g!?|1JpJMR+v2<WR(E3-90c1;09Eern>-Z*al|66X%>&KMUm`3Nr{>_ME2P%o-<f
zQo;EaRBHWnm{!xL-BfS>36-us&sa@AM?}Znpnh{DlIKMX(`r0n{p1_<G`<y6ejz`%
z_HS=ewE9q@F-)ddqkJkqoZ6(?T&31Nl*O?f8WC4kuz#HHz3qs&LBZgUNvc5}!U_L4
z=W;5g1$jwPL<NVwLs*qf?pw^*S`DPm(Bg!4o+>oH@wm!?@}k13)R6An%1Z|Hdz`pN
zuJfF20s5a9o*WKG1P2qI^=($+o~H{*w3&s)UGCdQ3)bZhS^fkmOWJAAem^CDh!P4{
zsz;1~N&6kYVg9A(xg#ho#$lf-$h&%kDsF`h$=K$<TDn<&cb`aHPmSeTxb1cvvS*m$
z9%p>5<RYmOPI`Zv>%WJ#r71U0MD8}xG|I_mDbG!DXu$Vv;MN~X;kt`^zu=S8(DKq8
zA`5GR-fv|sh4Bt^`x3@HCAkc4#=>1A^-T+t<=)#iUg;z}pZuW%f0o^1)ALhA{^8Ck
zyspOiuxtuzG{iS}?PlZWRc3u!1KX_mo{)sykbq=GXBBDgcUlIFG2RX?^YK%pEge#$
zmV4Bt29iIei@)W8a~_9$^=I?LMZdpU;8{D%B85fMV3*du(<Lpvr1ry?fw#qO&#|Wt
zBD5%FgxVh%*%XFZKqut6c<hNQ$cEVRy#tfI-S2VCX?D?@OJ@~^<d>FQ_Wm?AE$R3^
zZTPzieTjaR3Owdo0|Da6ws9Gc6o@6uP8W4+w3j$o@&Q9^-DFLeIv44Qjz-5>vrN%D
zDmcKNh7lrW>BjHV`GheJ`32tvZ$qH#{HPYz0Q6%@T`Egsdud5nNaX4j4d=HXAx=ik
z9(S2Qs9m^D{0WP?Co8PKOeM@3jWJ<A^KP(PB9O7T{;+O4lw;`F1U>+0sFwoDlb0hn
zodT_<Pqn_G4<GZHtCt0LwmfVmSsjYMtSG>x>54aH>h_Aj1`LgBw2;``R2S1;No4*j
z5UQ>+@sSFEwU&?jcm-!C@B&`&P>Q!GgWPLGh)M2caq$Hg1lfjkvK>bdMdXdU^z7Tf
zx$Ucwt@LJuaZpL4g3@4*IMVQUB;6BPJnhI}%?^{~`_8F04`T35$hv!>MJW=Obe7(W
zfL{GRG6!VJ2HlI4vUq#dGo<?$iWm?=w&(cAD^AltSV(sA%t|u9e@G}xe5PZY8WCU|
zP{uzA^rOmn$qjw4?$mc>cZ;00lTh1|%0zi-nQS*>>78X8b(gg?z7nhlT}-ME$18O)
zs%TPiFXD$1zYrGPmt2TV9snywH~$FQtrs&6<>T=ykXMg85k0S;x9t}PHXfO%zoII~
zSbTTq<9E~8@8tz2LhB0|^QZDduJeAdvG;vrzh4&!kG3i>QJ~Vw4yU)keWxEQc;Jo`
zAxYsNGs!8ACN~M1r}<9(t_+J^ega3|7XMVBi#GOFeOchCT@!(o|6AJF!D(iZ=7-{}
z0~E~JPIo|jFSh9WG#?wiF35+ebyN~aOoX%B?!qwWqmW>qA{8|12qaWJr#2QFuBhEH
zW0jR9jl6t%VZZ}AgMx~a&Mj$|#77iJ!BQBIiQ0~6Gvn%PVu=>)qAc>DQiM&hHvT7D
zd2_gd3fMGxQg*6qt^7}}<7$aiXil-hD>8+8_`EKer$(w?cm`BKYB=|*{;7k|M0hwX
z|IBIXySeYpdOX_^{+uWu%nCi7XWWUxbSgQm2L7j;D`aV}>h-d!nes_|la(xNoJ|X&
z+*kCbvh~XUtKVWuKjN)EpZ<k;NHvF@t5)|g<e6QHl3mKMqNgVP$h=p2ZL6l;WWj(t
zu6c3ueO;}{;*UK|`lWfVLLP?Kj*oVzy&ocuzCSim<OZ4IO^VLUEZN&{*GS?xfoJ~O
zs1_dn5r|PMMP-L*d_QC9)iNL7E8*;8hd8>)Td;{_wDK9$H%SI%=n_e-S-HH{sx2XG
z&Du5F4^H#(Z#O%nHtCQCndoErHK7M6;ugRf_>PNPk(<DAe(cjGN#|x^_0L=Ora){P
z-!#&acSiCZ^WVyv2W6;|)T^cglgjak8s5y<9^p!ot*k9Gz9tw_#`omvH5+ottVQQU
z%JO%edBLP139crzws`xFhpWb5sLsEZrE5UO<JL8qb!6VJur@rChaqSaMJN<nC*nz|
zeT5u}E18`1iGwRDPMyLAaba<(;oc6PbCWLi$zp9UZ5p#bBG3MWLe&RtuF{{ReDtdq
z*TLIw%gm#&lf7D!TUcY0Nm)6rE|N<fd{=3wJ|AkJ&h+$$^>BffkwGf|0m_|QiXxu(
zgU~Wvm^(BD2VBv|+Wz41sc5;=B45#KTi<X&*a=*n_eVQw3A<>mO}gc$ah5HG*1e_$
ze+-AzB5<$1C4+RfAb%@wE9F%KPSxZrF=1F$n#H%>HHx||wtDD1!@}E<?D68})3}SD
zIrW2l5B&bV<hi@XQ}Z4{LKIk9ZbPw`BaIO=^pqZkYhB^xIj$@wdzKZy$Xy2RSMW%!
zPnr*g)DO#*wcHjq6OFLcorC-jPDzKVEjb*RG|D??J<E!lNemU_I4T{E1j`4TKeOyD
z>Kon~z*veXC=IneE_3eA*NFJex9`lm@8^=#XREEP+_rW@imIa{1o{cozQNzNCzsFh
zH*S{6(|DHPwl|mKF1eWr-5+OJG;E+a&LVerTb<o@leuL!o^{=NctT57E<2YW63bDH
z6l&{e4sVKTe~{btt(=JZyDw_yk~<Y?I<~^8mmb*)k}bL2g*|=P<Rd+-+Pk-5is$mI
zR4=M;|El4@b30x%R|7Ryu1QNNZXz>{hn;<_zpX;G0M@uf{L2|K0V5}ibjtlCT^jTC
zY8s36t1R<=3$D_XtD9&ZT;0*T70E`%t_V`A`t!N&u5e<~hSRwU{?;H{{zcDQM8b%I
z{d~zZhaH|Y>DyFJyQ_r#X3?gqjKib3m8IEK)O}Oe{pxaHd(YWcg=`em<Q_Ghr2Z*y
z+w^yny>~Y13GwF?fz7Ua5GD=&+8g14)Eq8V$fdk09+}MevrG5QE$Nxuf?rClL2nqR
zVWJ&A{$?E4sfFb3TY9v!gD>IrB1(3X{sTcpFvx6a#sdCTe~_J<YtU^zO^R3VyjV!B
zh~|=|ZU4fp<4X&ktCLVqtp-iftAbSbUFq3Ae!E4;);q3~&AF_Kxq0D6`QkHK$M+wy
zEVjhN0cxo}0iWVIg(M$=BR4FI>5%S^8x(`EHl=b>tv@-C*(0xH__ocxLSNQ-`^Y!v
zcDb8_hVj@dAAc>O+<7bEndOj8fKg2^&zie5OiG%5l@&Tpuw<BI;Z~D8(m#)<DK7U*
zMuM_am5_n=(C<#78Q#N2IY_c*D2UR%+5bK~sR8gtMt-a%l-RG*KWNZ<oSN(~a{k2c
z`kAzUags$TDug)6Jv1(1x5{G|jqNWOhhK9IYkhZ)pUgVGMSP=mj`{=QvscDD>r&zH
zT6CRnRrqEEdn2NPYKeD`wMkQCDd6;{(RKR6rdWKF+&UfCV9*V}(4`diF4|2Qt1oSH
zG$>hr_8d|EwXoNbCTNiy*<s8#%9RDQ(Dh09HL(X-6G`-e5$cA0$r~S*ir2V2<eBq0
zR>NIF8{VssZuB6YPwm)!k&9$`l{SGpA#}bn6M*bMxPz~eYA~T%f3o3KhJ<>v-;u`p
zIs5}<Un4%!>=TAU9G~TCP^{$d15Gp))~{{c-__+?a`_yB@cnMeKHkAmb~(>W#(0?1
zc=gm5Wl-oO!%H?fgSxxhBHFOodE5!kO_glu{u$-n3l-K${M|ppLJ#vCv%m!~sG7`n
zm0==ZvTG1TUtwT8&ML9hLF&=)op%KoonG^|qvH<zG1iqe-)^7$lH^k=*tR2h`IcUT
zo!*{{{uv`w*^*vb(>D!L@%orv`g-}ka7f0DbnU~*NrkP&%U!Hp<Wre3nL}rs)I!~!
ztj6T^FT+PJYa+yslXmX6-RNa7ZdQw@V~h<)@6%R8SelM6gbs8TWf}@rna!>PZ<U5U
zwEodAd1MUPQEuOIq8wSwDxLE|_04}=)!eE~R{N$HH@vKaoM)7E`Bw2?2j3^%wsdcy
zF5Y!SI8hxa&Q45po~O)94$+=0!|Cs6w1e&TyCbrN=IlOfR4y}OMNBgP$|wv~*B~^_
zk@@cVV%9lV&uv4!@sjOg(4E_S&(%;_Jdd}bsq?PDG*FDXjQjEmnl<y56jV2@-t#XC
zsqP*$%sG2Y3CuP08ONP}_?~o6x)A`8x`~wC-`|T|cUW=K`bXr1CO2J&Jp`MVgt`MN
z2;pFg|Jv+-vH;gb0hEmru{Lnf)CSXo<5;D4^h377fR_~QR@3Aaz7^U#BJ*s~gSgDH
zt3}sgC$Akv*4NVX3BANu20EiH>>eD;-?_t3-r=jxw*%K(^4p39!Z-I^(W-|bt(ow3
z6(u}-cv_0_j?82<_n1X>ENgB7#JFV9%KXvZD%E1~wJfe~HMrxWwMLPMt@yqj8W?HI
z#`_%HK6yQ=ee5W_+io=m;dkp8zS5sQQvVKG>c0Eik@ohxJgp=N&eig9d@JVab|K=8
zlqCusjxlw%ST?eHhfD|RsIv7?@CVfsWgyJrUuA6ZpR=8*`o3J8TGqMFr}x{otrq9n
zbVlCsXNigZ47EleUCM@7nl{n17f0+ZG6}-3b{WT^=%4)l4DtF=e$y*U;@@HnOuYR#
zEcbQ&xzlc(b5rk^F5s@9iFskvIx*!r0YCBjlUHj(9r}yZa;P%3+uFTtQL5VyCvr0G
z9Q6ASLLF@|wx3?mM3>OhBgWfnDVkL>mWPv?O(o^83d8b8a#+ez#{n(-lY2)MOP*<;
z*^I}BM|v*>r_l-cWbhaTA0|RB+s2EZEmXf>6092f9=$8-r;R;-VZ74#y~b=9(eX9j
zfxvYShoF0AP?UhnL$u{8%#~$=X0WbyZ%aKro)C|@>0FO?wFKJkNyIao>H*_%U2mc{
zpPp)O`~U~98=-3G*jAfb1&k0gZ6QqC{`Np6+I$oqU_Qb}NsAp1J8WDXt7@g&ivfJC
z?|kx$jSrBKRLCka*T{Xe!_I8yrqhbHVJ)pi>|f=IjEB3~pykx7_~Po5#H#s^jqQK@
znO3`}P?}5-)7(3oD0l}#?s!#DA0a|j9qb`|FZ8U`okx)^u;|qowss9#Z_nb)tg4H@
zV=`<xjGk;LYfa`1s)=_)MIc1%&Vh~FtH)goR&$>TBiMxQ%`MTjlbo6$=I-6rJBuOe
zPV|g2bq%LV^Gi1R-E;m%nwG=-_YDrV5J-JWRmFkR46ldO%v}Sr=chNUR6}C!i5Qdy
z*+I^27oU7MuDiRRR&*OHf6+4b9NiS@ki1haTFjCOWNlT2o>_>ErCfFBg)nWR`oyam
z5<gfP;)S~Ot5(0^Q+?K^xEI%II2+Sol@OD;C?O_F<g6@TO`}Iu$UR$B!RBk%PvDBO
z;#A8N?PzJzZJXPe4i^_VkMnd)(2KN<x=obZIc;P}?g?$5i&{7OSWGXIO383Gv(Fp4
zW#i(Ot>boV7F<}AV_@mew*_4%Cm09bAHy$^+5~24JC;<h^6ViiWG3Qd^QS6JnzBOP
zz3*373ngxvP%Pfi?Qc6FvAr4AWskD3{v=C%NqxvIxcnYX`UWX}0Dr|fj_H67DP#yC
z!{}pj55_r-PB-QIXGTxQg?@o<ED+dNv%`O!#bc&3MZ><1M3?sy*bt=dp`j2t4tN#E
ziy}|DBSnz|2u2C)U!*YAtWx2aGnsBgCq!e(@D~H?Nfx@Og%q@F4Yu7Ve@fYqozQEK
z6aKoBP6(^#?|*glRWMw1y)}Rt%{`ygc#khcOw0H~5~2KalV?CPiw^65+;4DDcFo3o
zs)S5Ln_gmnHq1LCUiuI&wXQiubtq;YEQ%A%%0FRtufqV7OwMD#2^-JNn5Vi)$D$P)
zYtF%zZcG}$M=I8eo0rz?{rPd8t*axTsFlq_Y$;)D`FB3Y+Wjf~x3W*j&1-Xv4#33!
z_gBvs!PS#|r!*VhU32G(^*laY-tSlDu$x|T?6x10L}gX+YoB%(in%BzOwhbmDcOG*
zR)je}td8A03)Mc@6#NlX4_s=8$^y5L#^JL-)SXs7U&p7{zBfM{O^pk_U!h&7BrZp=
zemf8~2<ryy#nsz4&^5-Jo0<{2jEDF9#^qy^xdNZ_Ejk2-ZmNzVPFmsa*_K4=zh9Cc
zVc!^_os4NrIxo{xmLD_c9tl(Wz=m9tn+Oly%w}dI?)b0iqlm3Q>VX)}Kq{i}Znvd;
z@h158%;jN1-D+#03**!7%%+k=dA6il&CdAvVn%~}6GLso>^@`y|HOI>HCI-LIr4rA
zoj&=q5+B=Pz~z7TvVT7!O(VRq?!gXW{Y&*S<W@xTXR+G#@8Tb8yu2xx+3b+uC6y(e
zF@KhXgc5`F6cIucnsJmbCu!eZN(k^!bZY~KI6PL3(Z)wDd)bJS$y#LV17z9o^*;eC
zKgyXi9RY(gU7Pr-WNs=q&ZTX1J2Zk4bCOk8m_8KBu`T9I^h_rcJ_OsOC<i)T+u8gv
z-1?V{7e2vQlv|tNupUBa&=)$QAoVdD@{_!E(_R{g90Jh0f>@7?58dp<x_673-!AL4
zL*v>AVQ<xegT$V;P(tDxLiI}mq-5wuKPaE#Zp5G`OKg$0saO%y?A!OX-(}-QXU8D-
zJ-xuU$^ZNJ|DK0F;=|Yy#P33<<LeUMQd{&Ebi_-I#L+FC{$`K@N{4H;>n$61+Ez+k
zitB|}+JbKcLasIj1n5+YHq;7R!>)I>GUs@Qw-;$;KR+jf*iKvt`K2*i{3=UG&Z(|x
z|Di3p<q2@0$+Qi)hbiwS)iKxIJ6zGDeySH$ciMNzF>XJ#^QkQria5JYyIuHFGzsb2
zxwGqUT!V2e2r%(fe=EK!9&->1zwhnTuZa+)r@Ut#s?+4!d+Sc`WZrPczv(%EBP7eW
z@UZwG^$^&7MbmUO6;94Kq)w9DbcV@{wa}2$V0#jHtH6apDA=YNmQ)&Y`=O<?Aen|-
zf+xgJ#W<U3W4wGXl+$&1rmu?Yv`@L0)qTsEy5}X!@J1m$-leGqFpq(P&hxpqlCxjO
zyVg8*i$*SAXjM4j(4T0T^^R_0zNfydrLREG=YU|JV(u!;YKK`{FY}x@vFk2l6Vf?N
z8@0V*w#>BX*Om^yl2n^x^PoI=N#;#^vL$%GJT`V`(|kHjFR*Hme8p?wRQOlVoq5N&
zxMJcv*3}4MI$71av9f|E{?##-<xZ+~YKgVbk^WfR7IJE~1;6D=>T$y{-ygC_!Sf6s
z>CGT({=K*={3F|%GRjl<1N2$0lZGiEc_)O8cA+y*5HnSpv|XsKfxjaQ&Cs10!8B^`
zTuzaOGdz2$e$W&vj!zb>OIB_<N654Ivv-&0JLS_{OZU$~$%~3NdH%-8w+Y`7@5=VQ
zUP=Fyec{?}SH2mvj6DZ>2J{$xdPIn57MQ!{yZyE%QOcU4cRd06=lZz4G>Lhe3*<&W
zw*;@fn<(9g(=v-Pn{vz*&#(#tXs;t&1P}9A)`orY2Yd^D4P~e1Hg&UgDO#j-UtCPQ
zn<p%uy4D<`T~j`+bLp4ldH5|L>n^30*Cj8Z)>j_QKyh+x^i*{6(w3i5_0lX1b>-K7
z$C7vN+clCgz6xx#v?H_HDy_tVYPalbrT)?eAQk9BbnN9tX;tNEr97b%1bOam7^=g|
zJHb=LwWJAcXs6X!S`{4S8U0#poM8hW@k0mpEZRKf?0xL)JDcX+Fv(+IC#mfKwup8j
zis4OnS+bw5FU+}TxnpN}T8CY#21g#XZxhHDmj%%crhkfSgodxMH-51CH9;!Z$RL#`
zKln}WQjCZ<W-hWN*$2+^ajst^Ze+>L1~Jw0&+V)31>t<{L%U!(klPJ{EJ*tOFNj$f
z4G34oHmxzk)FTZ@B~-nHjjbumO=|CB=;#p+5`+_%u?wVCDk2%=(tj)$!!&tm5UAn2
zK~9kQ2B8wKW$Z;<WGR8yJB!-aNy4DIX4YK?d^?rx8CyZZ9C$74j0Q*|r2V30$*KSL
zM1|+*N5JPCbbn?8Rsv%yq>^8r6Y`Rfy#bZNc>MqH7vOlCJ`!c0(rVHG^3kz@zS|5t
z<8veo`{$N`<n}WYJsbUAWh{Kde7z4JK*RY-Q*gcG9!F{5y-2Y_f@dVvsq{5)8WOxU
zZ>~6eqzT)Y6iN7rdQ=Gl#JXBqZR#uV^`XO__xgZgk@5#rge4Y2X{rC^B6(NIJ%Z5F
zz6X1w1-86Icx3badBlDu93@%S<-ghUh(4T5{m&+^>m&KxRo1=pAM!bgn9jr9Rkbv8
zZFY}S+__GWFZ5JioNznf4lk_KhSNk<^9A4{=B2h^ZpNMLkDTJ={k_d&5KoV30od1r
zbYHB~$xQB$<p=|NoC*Mvohvz?&y`CX1J<PA7uMjDZ#l6jVGW>UTLol3w+8(xL*{^*
zTK>p1Sh49;n<=V4+>OU|1MF(p*`ovhVIG*f$L}wSJ@Xs?9xIL@&LY6a;n&~a>^1&X
zqLbhFSLi)U=N%z<(}M-B_B|;K2!BrcH6OLIRp8&v0ifU{C(W<V0PtNQjo`H45U&Z@
zi1puCR7Ht*u_&tDECYPOdB;VFk?JqYcplLxg8zP)-m`%9tIWr-@Gsr;vv*5J5WYp_
z->>D5>_^Q`F>YVtyW1-k-Xh9qv`-c2oTDCT_QI|}rm2L#Z9J|bc?BB8BvLwnv*5TI
zL{Fkl=F_Zdl=4^oQXME#@2d1AH`*nBwXR>@wqpxbc;26-0sLt5b^1ut3brI^uGLbL
zM|0Fe)Dc=c;KqhJD)VF74X_0erpgLjOoU<jzzDoKMZ8YgY}Y>|fu{E-gG`y12?(zr
zWuo~=5@948bIU?^=g~rxRu6@Zm$Ov>ww<$>>-(r89SKERbidMkh$3(x4!<Y&<nF=;
z@L1P})?lIZhyPU1i=`ugpWU05(bxY-cmOS#9}FDe)M<gM2PDgV05xZ>O)yXf*lhhd
zP%bo2ahQC~Q~X~uWkv>&i2Nvd?TtnD#Z~wmcGu&BI)40_kCB7RpoCC;>gmwupEECq
zn7N+lhs{0!STExzm+mySgNkvIulMW!^F4Gvz7Ve819IN7B~O++;WG7V%Siy0eNFp2
zH~?To`KUhU%SoNP0+w+%(JcHE@&BC5LPY~EfE{7t;9vD@G}=4>O+TWoe+A#QDPeX2
zwL#S*_tH=Fe;t_r2TJ#!{N(@rUQq-rwl`Eec>ms;T{ImFLoivCNo(LV05o`=LZ)^Q
z{zLin-@mK>mp9&0faIUl@&55g^#AA2NX6qMfZ1&MmjAhX|DOgE>EMCeMOQlVA3>r*
z9Rcw4FQjeD=Pxzv2LD>}yjrCmDMV6~NMbOEp%{#%coOzcND;oJIR;vq>kj~l&=<{g
zYKr}d^W4Q611T}XtpZ0Z<#A<s|9X+mTW<2>J$wQJdR`+xGTraYb@W$O3^Z>s!2g^`
zOZyD#-`nXZr{i;HiJ!SND-ckHpz|ARd|Hvyp#bIT?rN#ptf=<a@c;lW4{xS<-nShw
z@&VFX2?k5U$gy2<<7MA!?BOGd6AnIhbT&iwwSytenmpg%Uf1qN(u219MJ#rQ4+gXG
z?SA?2`qRi%=6`NB>kQf>BbigT4e3!MrjFSTlxU^^@Lek|zh3+Z!D0eDQl`bJF=W%U
zO0hhfEtbafFOQ!ALxk!aM2cdW%(u^`4_C%Sa1u5CdTN`sn`lY{14@-?i|U21HLxC<
zh>97*q(8^%eAKKtIDyTG?~^Ch%$Vlv$7R}w6RUo6n6>4Eh^NybK&s=@Ae+R1KYIH_
z_Yuiv-bHN02%8K?t03@hICC`fS07F|1=|MKMozxo%YVN9hy*(9F-%|jx?lDZUiP@l
z{k*ec&NMgwwfFYV!L%pe-h@eA+0-X=`oo%QEp@xbzvgkO^;}B!zp4|=>#$~cqaOz)
z0G+AbC1?-M0mPlfBUJlQw+R5Z1;lZ0Qax@=)Byt!W<xDxJVM0>NLmNCN6lBa*!;&{
zf6Dsa)LQd^hS~ur?Ue$V%?toO*J!|E=m+I=pvNnYK+sP>CxmVT^4~uI3_Mw&T%3tE
z54`{WQkB+v0J&WL$b06s-^Rf*nFCYBDnmh)IOP&NviQ$C6Zz2x%@6mq-Dmu)ITHzC
zp^~W$W*2+l4wZwTx$p?TC*?H%>98}_nWr$RYDE4(kv%j6+K3?y22@voIa`G-1RIG1
zhx#>=+DmD;l_7^I&8<0c?=}k8eO(|su!HVYvosw0AQzCYAWiMptc7zf*ef{l@pe>K
zRy|H4SOB_J@q=J!Irvq$EH$OW`ZDd42V!cMw%I>kiyvQsJ5h~>qdN_nL%KAQ;T4ZW
zPRrU2f;rGZR%<BE!I)LM4fQZ`^NE3u(#@IVrWArWCRhgsF;oC5Q|YVS5)hLY{s67K
zcsKo4#G`m6)n)hf?RvDFIUul>0?DJTElTcAs;2Zc8t^bXmz-qug1R;VR^`qCOEXMQ
zS<@jlj8dpckDo|kXRg|+)9kmvOkUlu*)19<-%fUPs3Ye_iX(^qM}^=l0m=HJAEM_X
zq;jhfA7ki-Yi;!r{S3bo8y@dWR+>%yMj%{-U(xG$xIL^dPH|ohTKOz9=dwGHNHK!N
z{A;KY_%mRxer)M%`duo+sRHgvo`CCqN~tz)-Tt&ol8{8yd=ax1)&U^$rug)-jOgot
z>~7(HbN-rVam>46uD{~pF~b3)fBZ%rueY-z%)t}0cui!n>6Uf-*#f2G0kvYDs2xB_
z_qJH_<Q_M(?(f^ADAm9|M(Q9(qqzvFkJ61@Qs4wJ<JbPV(*FO=1aQ1<MvZT0BxV^{
z8Vu<v+~Z^dWi&8JZ^C3a2H$e_n+rT)+a0S6(RAMvpR81mZ!3LmVbsC3H2cfU+z6nX
z$_Ve?DZmUXC?@ko9T4SvtvS$aXAI)6!#@zgtrVD{WbWksm;_r^7#mTS@CHaNHXDif
z91Bl6xmN=pk-e$zm#XUS7k{G{r!H>6$ctmW_hOy_lpfBS9^1dFB;iIeeKYM-d*eUF
zt=wN$H_ybWTdF?G04V%Px$G$H1Yu#u()AzGF0IRt=yCf$9_w=Hf8s_Xtq`JKh<=Ci
zhPR4}^-D8VDsso(I)2ROASI5N&rsfZ5uE`PuwI14G>%5paNQj(R?T;YA&g~#jUYg3
zK`Dl8?jMWah>-aZ;gulzc~m^xgE+X09zb(4p>#X;b_YcNO8~fcDTn}c{H2he$za;M
zmPVgQoQCdu8?aSg9<iu4zJ4Ej-3@?(x&ag}d0>^xbx9(BnjLyO{t4OB<cb|_DN7Q2
z4j{%Xz#NCmBfmD}*@R5Aw{@YX8P6cQ57T_uHPWWPA5g_yG$X^Qs6P}_H`Y*-i$%p#
zGs*}f6Ko)h;Nm>XA47Ug9k~2O$wevV*BG)G<TVBwbvlML;lq!IYl=Zyr^A8y+UWVM
zA$QK1;c|t;Pjoy|va;73eod218Ak9P@MPap9RMjV=`o|6Gt=~><h%|IeAQ#<dEKu)
z5ruI~z(cX{tTt6FJGgvUP<<FO1n&?;!fE))j$lQe(;n}fAlZ%am-LRnlXG^#dG*ID
ztp$+d0l)^`vwV7S@>f~=9T;a)&dg}@+?xfLiG=m%9+0@?yM|7%Xem^h4>V85sPKQB
zkp7k(;c=d&q-_PW12aNcbCYd9s`F?=HlB;B^Qm_M)OG=%e#h+7tONZW>Oh1h3!T&T
z-tR(FmeV2$t|N#AFS$@_j8xxtS<=G6?`8Aymk&8f?D2u|63vf}<)`WT-yQ(>UGl;U
z3_uWVRXNr;XLiQBdGF{wX`Iu~IvIkT%Q)nq1M$Hc2@f5W^Ei!ul6_RERk~r|j9NuQ
zP|E75kC8qc7B^u<ZaTfK_)TBuo+_{vJjpC!eHbb84VNR@{!>PhL!;n@>RZ*xn5*qy
z!U(Xi!%LY;>1QAQc@a@R;ci)4kEc)SzK4mae+#wXTKF5f;H-J(-`HJg>c?8Y|53W&
z^IRQIpKPkfDRFt9-d=(o3*UnIWkMK<Qaw%!WqMs2Y@Oclq?1sY74-H8b4T&470Y>_
zrc=s?d{((yaMHE|S##Cj6s5t!_Yvq&ePl|V_g^J>eAVY9$2;?~;bgyLa=RV5E-A1g
z71(Tn879C+*?f*K`z6`*aSRAx`sxpW^FtPHmC8cG*R7kF-Qnw!aQP{#^e6l%CG1_I
zRe#$>ZV#+H>xGccY{{AU%PB98d*15h4miLCgEi!6H?13x0JaI`tXtiQ4&7l>+Z$A>
zZ#QyUia!8)^MkiH`Z1T@;A9VW>FFI9@9@={I4XCWWkB<h4yy`6znVeU_8y#wx@B#S
zL=;j?)oL)qFEmf4H=R7KE8)?BXrMt`8o+Zgp)i*z47ZI8mH8$yi^1xnK8=@)^g5nS
zK7C0)=7YW1)t&+TvpYL5b|D-EN(z3_XaQm(Adj$!kwa*rg1@wrkA;&zmTW8SFXtYK
zytcXIWRe$5ZbCnvT4Z<V>Xzj@$PU2(3%&Jn;QRTLdtps7V{cQc0K5!i81!Vex)v3!
z@l8+QRzdc3!>>NBJ0my|^U_X&EUaTmMeDBmwCULk5XwGZN8C=+MIt6Wi<)3Lig7!f
zLxGFOKNu7@%2`msRN-Q$8EqUv-9@<HJsx;n%XC+Vm}$t2V*ZJ=#WiqIZbA&o;%G^I
zJHC0mJc=@OYQjb`l#((n{;}N>b*U3u<ePwbPovM?4J`{u=dMll$t>J|FY!*4T#Qqb
zBJy`#zf?K42FLCEW*>hnC=>VQ48Hzkv$i{*A+ds%0tzzTvQ?k)mSVq{_rI?f>0XGi
zjpq(QwaT1x2Mv`xuxZ~j4~ojGRF_qgv(kt{q2F+x<}3Y<`9uk-{@+jGoyTdbSCf_c
z#j;y*+yP|2Htm92O_<Um_9G!;PRk(IQsG>pt{2Ud_o@ht-dQ-gBoYlVc*wuMEQ7sk
zlId6Ayo4e%wFUg-i!bUz<HjoqSveGiL!By^>Ev*3jt)Xth+gl3)FkDUq~&BL2F_D8
zIC~F_Y*Ecpgg+GRLf}p$bdL-CVg5WN_75KN+L!cDZ7}LYz3w#iKq&zc?xAe?WzMLK
z_pi`V5nt=Xu|DqR<yf(tSJFK<aM?4|75T5>?^i1aK(>{K8z+eY$YbnYa)D|=jRt^s
z+UxqVuTkVIar6na-z?b|i}wRfJFfqmoRthX^&^BK8e5gnXHTFZ^50*xVVh5HA2OOg
z++Q_Yan6}@XnJjoBOK#lphwcHEDO%IsCsBh&hg+X#%tJ0UFhpzEP~Bm$?;{PA^w;#
zGVHp-A{c8U-sJO^I46oT<~?R+C|)CR$x^q_{fPv#BFVFZF(*l($47NuZSz|<Q}~}(
zOXO!Nlx3?^hT=qR4f{&SRFm0eiYG<+NgU}<bx%k@?d{O*fPMttyOE#%1X?ZTr#nvv
zJ{A;>h$LuhxK%<RV+<~yC$vK5lNweIXFZ}3iLM^5X^``X2cuNiO9;o2+u2WlA(%)x
zXlo^P#;H`$L~NzGw$5As9sGmlpX7F<V3>NnX(@2kwP%&s@9E-ixz#zCpPM)vI%jq}
zd6cZ-xiR?NBJ|;Y?4{lQ3Vi~()6yhk{puHsVe3|c?j*~~j08Y>=JV*U7g_J9KWdl?
zy>1~Yzq~Q~nP8a8c2xTDYF7e!JCQ4xQ?OzQY&T)d^5<+-D&qOKWARNBD>^67$-HpH
z5MTayQV=|S6ZuyLh#&q~sw&3_UCs3MiCb9@yYK5$augJn1=knae=}pOx^Fe<57zv;
zjz=5W0A~O@&lOV2ikM^X9)2b|eB}Ay>K^|$T`Wb1mh-55aFWEVaP*xEnL`a266A2f
zE1^}jsw2*&?y@}nu!8NG7nKe@1R}@Wt|!fy$ysD&nOEOQEk}AdxU(QmAOdiJ5iYg5
z^##9!ENsCr8(H60TuHTs=Y2}egljyZg%x}Th6Tl$f$kM7!z<atbC-cV0$#O;jaOR8
zyCn!|uQUNtzPJh<tG2--@(<!s$pV9Wibt6K_a+%n)I5>%Um)54{XEF;YNjgiKtw|g
zcnt^XlFV0`)g=kVqdDo}YYfb|DD9tp!KMxup~pjEqt4I(W@0tYn>Q6I;LRW?mMlGI
zB{mB+)kT3(;&KN5>YS$gxDGb#0(Gk|l2EZ~#ay`>{-h3$q|exr@gn3iztMBy;GjT8
z!ca{lw$l<guxkVT!g=l042ok?*Ia9qUQ+>r=P$^|L(9vau@Cje2ikBm*3owH!+yVN
z^|a|@k3jG-D8$JLKW!Hctea)>UvHhiYC*)V5QL8~x~E$y{KYi=G2u^w_3}BMR6*<x
zS~3;K=%v^pwJDRlae`f0KIt-xB1Im})bfYg?aUBG0^I`LO_r)cedgBIQ{cGKqQPbx
zZsjn+-W2`Tp|u{($aDiaI9tdRY<H77XWbog<ePcME3tQZFnz$5=kjfbXM2xGKjy@L
zaqS!TF4EExIAz+xH$fbR5LZefJH?4?kCtD)&+U`W5)lkRhGImS=?!M7wou+7QuH}m
zF=uFJio%`uNxx`lW-jh!K)rd(4(~>h;$IsNfv*kd#1rQwf&3+>+*a$^99agb`1pY(
zHqUxw!>S3-PqP3XC3u+ssl_(F?^GoAC9b{H4T`&u|Ci}kUsZIG|2l-+_+dlWAerp&
z>tf3Eruh6`{3Q<SOp(iT?LU5lO(Oj4=Z$;dz~L;7-%%ZZqBYsu<znXHfw7`PqOh6b
zG6Rxcvvd5+tMZZ*WF=-bYc1mGL&8XUb!)r?h9npAtWpEVK73w@vu_>d0=0*KhmzyF
zKg^5DJy!b$$erdhM$=25>!h{BDZG8CvE3NWR8B)IBjX#@VQ6jo^i5>{<+^Y3;?rY7
z_B1Yo(5{gTjU>!lw&5<XyHGD;qhy=!vpy=>er^iu446Ck;T~)^oRCrsEKYH1<4nR<
zq0%Gsl2(*m7OJur9)+3D+KTmdS;RM_uO#CTKt51*F01`tNz#XBud;<PTb4Sy(>QMr
z9n&Ev*Mx(a2wk^Ol7gRUGBs@j?0bv|aLIEGw?|HcG}sTSUX~n5ja^52RWmyre@=DI
z=ZxYJ2hu1Z0}XFlMBv=}i%Md6t_Cm$=d7A6Hlwa`pZmkYrj}TTF^;VUH1r3m_}nM=
zE4m#m-k;w&J+zYX-=m#Pl3FDy(>EJ>*!Nxy22<1=pGc;?b?tXTTYI=nvPy8ezYee6
z3Q||!SpIfpam6Mlcpn4VI`B-Cq=d{52pnmg-a%$lSM&yw?LOS+Ca4<iQ5fvc<fd`B
z?z1j7P*o&Y)|m``PYQb%fGqZ@5D}`+r`W;_N^I3H#+0H?MHW<OrR+)I*u~(_qZO=G
z8vSS}TF7<sK=vZ@^p5Dj<_rAqSBOcaVI&{xq`fvSXU$lATPsWvj9cSEUlG4r<fb)N
zhYKO1p80{;nQZYDj_(HD#qF{@O{x7|I)gC>7_Rp_+WQKm=$2s!D2>5*Exc8PRMgMj
z!c}x$Wx!vT#*%%1))yd;AS&*-$DP`p508eQPTBX|Uf)a#^NGfqBI>ir!E}*Yc-s>S
z6vie^C+Vn6-mATPkjxbsi0?2U^l@h~4k38O&U9RkoA>+3f4_vw^HuN{+`U2k=u9jO
zpG6+k%!)E*2}cQs{$Ak*^JuK8kO~E7cg(5zR&#+mk~8~{iKkb(2a-%GGMThcH|s6X
zW}7aa<5N0=z-H{%CM)nbFUanoa?122O&1$hV!W1%R5foacPEhnMt<fq{ZN7pfw8Ng
zWhAtC3$ki3hLHRUoB(4^k%e0g@-5S>A)Nk3oH-QX%)Y#O%cw1W92C|rT5ToB)<o%_
z9Wl@N!>yd3gd<lybVgwhw@S4T88LH1mwg?=d;8N|U#ulxfx&tn4wj$z4k~KHpE_gO
z@pMUS?lvG<Jx&08j)R$oNET=gm*%yFdl7m-K+Grp`Ym?V->yK}_gK@Q#_5piS9EMw
zCE<M}NV|-<X^fc%hnp0~>bKVLOB0+;^batIcFMha*;};p_&+Nby?FCtPJ4870Kex;
z&$|JABJYG~_zw7Q9l|o&84K{9EkY*ozi7LitU&hjKJpcu^37x~eOlp}M-^6veud>y
z_yUFRZ^^h+-==Uj33&SQN=E%yzkvY8ZoQ=6qdsTC++s1}oG(j1Bc?nNd{6mr+=GW=
z#G2_1WL3x!9HXX*P1>u&#!|N(x&7e<4HRA-75AILY^j=;3md?}TM^aZ-iTQTOQKn<
zSd2h;_*HO;mXllhFuo0O0&)P2`gKP8Rt^g|OB#Y?Zt&KGWGz+hzgPgqhyMn|KwRIc
z^!W0gaQ~?7(Cx_z{8NNGf)N;))cLbOov($G>ghVqrv42Uds#oCYNChuZT!xbjMe-b
zujWV_3h8F-*cp?6&rNgLRrXc1=7MZ^YgzG3?t7=7nhRg(zx~>L+RqVax8FLw3P)Q`
zmCwkAHTwE2wSYrHzki>tYx|#6Kd+oOFAL*_8ogeHII>$nXqzS^ue3RH?u8i#J+Anl
z0n7b9vNtR=56jp(xUbl81rf|ey&R&FR06{XxGd*u+VY}aoIw=2j4{dkDtk*P%Q;cb
zh24mIc%aG0*t)JPXA6Gk=#tGNO)`Rvn>vgFz?e1L9Ca7fUt{Qv=T6>?Q75AHWJ8(l
zJifs;#}zT#i?s`+%K!5s5%%U<0t%t?!{}O1{msO7k8987$lIicq48n?^Wyq7_WI46
zf|&%PgkI*75x1Mg3nwkhBe!)r*DXpbP$_dQirsd1jKXI5c9iC<&M-{)aOc;!(Nu7t
z)Lh%+;d~;#kGh%A?{O@OF?+$K(XD0OIdd83T3l8w<d$`}A)u@VKQp5=r7^<{ZM{Aq
zCQP30^DW~S+<|u;w1}TxlN0)JteTjn-4zr~Sq8>N1ue$+xL>-DHxwVZr#4;}F`QqO
zd$|1Av}w^@yGSg%_MEsVesHT;bANC>gtakv8jCMa!CnD_K}4zN?Ua7(q5}llUA>qx
zZ&!@lUGAZs8Li}^jH&(A>nHheu->0*M$o7FD*W+lXzkak4;6R|J;D0zPtpg2ZFM?L
zMox-6MKRgGVmAK~viYt8|LQs-CKW@tg}Hoy{cC7PoE`k732~3Im)}=R_^&axOa+Bc
z_Isq0?Z1YKeifEeLs7M`?9|P@4qO`yE;FRf^nPyS)LlQ1le1J*xBHub_d^A%dxRCn
zhmI#IV^6k6gK_8kf*lxyh3>wK^_%odhc2k0uwkjer>fXEtJnyto_fb+zEy)aRJ93K
zg*UHgOmY%Yhu(d{#QXX+X+vPP#uaa&BbcK%*e6+?^CpYCx3F-LsW967sd?wCOn1DS
zpe#`nDbGlY=nA1gc{(mnCxt5bQUt9MJoo{2Lu0vFwP&sti15S4X?RpAuHq4y`)cxa
zvG;i7BOlj$=1*vJ#2CV{8Af%^RcAl7yt9D#PayIJi^})E!euQi<gd(3SD$LmkAQQI
znOuTxa?Ld*HKLJc&A?C~Ojs@q={gKvfU>MFhiUiKi{UF~V2*gf9KFUbnUw^Y{D7?f
z^l-iO%20iHHdW_FA``oi_mr<cmc@oX56F3r`16zu-jg9sXW&B&Nj*r~)Zk+rLoI~B
zYLyi%(w2JZJITMhsn|!=!k!Y6AbI)&u8#3SClz0(=9Ie+d&$r93Afr_3%_&|<h2CF
zF{85Orn$|!I()6V4NMJQQSBSSra1J?#&a)wEqN4=h*~va8?|sSL+ul!(CeD7kGADA
zp9c&Ye#^G*U1uBS#lYyUl8La_;5X1phPr^fp6ieyR}2cpvtN7M2Ao#yOYxMlvY2IR
zR=OnRmvTb!{N=Ykek*SHkEC&Z)IRd+efKK^!gSwQWF)54PHy31QHei^k*^_=)EpIB
zB67ItY2ubnF6hJ0b4IDsUW`k>T*!P=b%^nKh@iKibUm@|ERP{u?64fqrc2XQ2$#OD
zs+@7U46jk%K>>caGS!Bj&Dm2^Ypt*41QZFY2jm9ox9N^O&-KnxozYyP;F|V9qfaX@
z0Hl_9^4afxjWmyHCu>9sJfRw4x%`~Uq1=u*A`|lE)e_y(XDwg;z(2m%X{Fzp8AH@6
z<VKhgd`ikbcl05AWWYKK|4kM0MNXAJ%UxR7<~L%0n?OG$1Z`yI$t&><dK-49bv8z>
z676&af}9Z>3eo0>7v&1gpPSUhvP9+yBfrnO>$xX2&4&n-GY{KuJSER!yNPA!r-~@U
zkvnLERPFuV_{}Gw-?v}d_3DPbDtsT-TN$lhnqg!Ssbuo*$FOb)VS3HokW<t64WkF@
zepgvdnM3jROEInQf@M9QEdl|!K4w;Hp6nH_*&jHDzl2`%Z5jb#uUY&Dwq%I2*zEK8
zXEE0Vmb$zwh--;F!x=Lx5#Wjxb0P&-1aZcydl46DTf`VJ`2QaHMfmU|2U;3T(e}e$
zVR$QH*)N?(utAkD?>bo&cq#djh2Q7Kyo&tq*9hE`4mcQ(@GL>voY?c!JM5WmLD>G$
ztgwZr>lT~qrJn)_VRXs!|A)G_4vI2<`~7!m=`JY+3F+=mK}rNfy1S)>g;hWVmQrb?
z1*BO@dg(5u8|m(5;oN-V`Tm~Y%$YgQ`R~lmurlN5eShxz`owj;->>UD)jy~%K4D^U
z(kRHw|3+STYUBPY$az@uVj_bK^lJ>R_GWrwQtEWqbvBSCW9Ojp<q0Tr<9Lv+<!%q@
z2@cw?YoKpXJQ2Bo|G_zN8}=4oxan}+h4Bw(US;8yo*t2LFm$^dkReSl4>xF1;U%_O
zF`n4?W{+gl4a(6zwkTj!0Yy9;&Bd)w8RjBB-MR!UM+crEe!>+AL^*EL2`Gp(+V(h&
zn=(@CZ~sDPMa&ASGH>6P44zea78>$Z?bU)WWV+jP?NfMW9FDndzFcwZuZi027>l-$
z!|&4ixBVk6VXnuRn}jRlJWSU_U2Ne6iL=b`(3FoX5KQ|TlM%HV>OL<8bFKJ2hWLc(
zi(AL=o~*^zU*CSvTECOQzYY|tlM06K5<_r#q7-C3YEQASNHe2~nOqu%740O`f-y%h
ztm9OOofi?5otFx&2hYy<l>B>Z7l(hJ!xevi<UqQDUz?|aoSv&afqV=|MewsJ>+Hx*
zD~W-hel1ZY8>s>OodupsHKzK_rk$5~WhettM-=HREL{RnJj5hiz2RPG)^Ag!TZTCf
zeLQQ&cEC50)z5(@>_VkV?tWuJR0gL@;3NmNttRjr$pY#rjEInda&0h=((lsGg2TG_
z6&JJ@ibrEddlyOcN{{BIOkjOZID}E3ff2-qVYQe46k*EY$DmLe$EusqNg=ZJ-H9dc
zX|<%=B`TU%I6Cg(@I(1&-eGL?$T-b#x`Wc%90STY6D9hTJ2YBbdRPo1uIG<iPbWo-
z;YSCY!}{<KKe!X3Tw2C`QsCFI6}05m<nFzx338=etE7asXw@;lqw{iJLcJRaeiq@@
zf7LAvzYg(_vE*1dI4*8?Ov+E%F?ikCxO3UkPl6<!f1C-5_6axJ4}2~;K}A0jeZ-TZ
zLxwQLZQeVT%@@CDbKFFQe@{iZwxakr=SS@KGTUM6FLAiupHVCPLlz<nj+PD1@-Le&
zo(D4yKMf;r+{U<!<Q+2hpBO)|?A^@@sD0rqVTUkHAa6Px01nUxVdq6ov>5K3+CNDH
z$d30$c&X)WVg!|de#W-dH&$EV&zLv#HScAR?Xl(7p#65eePT#P=-FnT-}$2pqlF3z
zKZB>(Mb6*rbh4k;Sce&5{)+4*<RDk3U`4qN9okT*%t4N1&0vWsPz-e;vfr)-ty)|c
z>}rg{4ZVsq>@s@#TN0ZGZ+dQ~*pC#SB-)S&{iv|)o&ic20ykexy8r5PorqmW`U}&h
z%6~w+bks1+{lbL(&(F(c($9F8HzO7}v&!i$>=YwH?#IF2+ZXE21d`W|buFlujK>RJ
zC^PIRjo^fv(I5Fgq3?&DEgtxm+C!zG(b7%ZVcr?27geU>0n+~Mu0a#p>)G||^JTy>
zZcQq}F7Bjdo6rqA+G%SRtmWwk5;E)b;r>a19=iFVSO^8Cl>kCyrDZeusEwerL^=H^
zI(tg|__XaXLH+iVg>uy|2?Y7@3QenwES(}$8aEIAJw!|!zUVedNdig+bHzDR%<auF
zb`ij?DZhum>sGxrSA~z~)f73snALVdyjRdEo(alASE9H}d$P%HdKG~}A;x#%QvRN_
zQ@@5BPVy{5Vd6y?Be}A^i6{t<@JOZYZ>@&}kf#zT1a}E8hVNvP)zEQ>{L~?^C2gzw
zf*w~tTz;LVlGc=K(q+eeb$(+2rLAI9(vV)xWO<+Lq&7V|bm#dHt1179Rr4{FDt~}L
z@iQ1C^MDB{(Xh!*B1$g7y~JE1u=E?{hgL2l=MlN33tnBKULdT!%g27a3(LmPnj}nk
zm;Ht0wG`9~HfAkD*0+YcH_d9td+fI6u8a$TPFP2d%jMMh*VrpB5^79;jzAkm)54}T
zevWe6nSgm|!tH@Kyi#o8E<3RqnI=bS5hRtRae9=eWb`f|u?`<qA_(YoK(Gs>JMy~x
zXx0^$(Fl{iZ)7P7jZ@_W(s^BbOV3ymo*i!S^qh?Frs$uWq@C<TSaP21pip<4hp&7=
zr}o21n)A2rHvcwk0lJf0ONsRxb0xsM3_c*S$668_)wZ5>yRVo#nQCmoFB1g{{p059
zSN+717+Ylu7=>T9VKHczJGoBA&9rTfpDhFc$xb0|&qXA$2hS)+90}a>C+h7o&yeAp
zzsFW-JoF@w+}`L2BTQlL`!ki%f%G~y1ZxfI4l6+7u9GsSN>Qc7LDi6$mP*`I1)DY}
zi_d?6x6dfu*iGrR?sS4Kg}ClGL~fI6#h4rk^yu=ymI5eA(8KZ=aT)*d1Ks$<<ix2#
zbAUABTDfe~g1Q-lzO&f@ymI&7#ZLT4y6mLiv1V|!zuR}h<)G`@KsUp0K$*ORz#-L!
z^V_H4jS@xY&KF$Xa(yw9X~vrHbNck)m~tEiq0Uu(u7m%Iyq~NQJRSe-AcOJpL+!5B
z$*1IRh7r%!O{Se{PbSt6EeVE@lK!5x_tUcy$W{LJ^@g&6yRrIo*ZR&ua9FgkDY7fk
z@M5laN>@nOwejv#9A`jOF_w3&cr@@^p27P&-N9dywZe;H`#a7pi#H8714@og&P;6?
z2F`oBvw@xzEOcHeqU(oWVbG(wlA0AAO$leI^SHApg-iZah)qt-;<nM}Hix^T-%KTc
zXgr5p(8~2#9z$$x`16wk3l03=n`W0&g8x#1tEr~9%kq7YK;Fg%jlr@*%dCTE=YC9#
zA}2picSZ@;P~tzYp@0wua0`({ZnCw?QBJg$3RgVDf~`Fcg>4(1qRt&l=4d@}H)&3&
zP94iO3gJS3DTTA0>~FA^5IuTPN;Na1Es*o+{JqD-R0CKt9d1T|{BAbzgqsN^nkY7f
zsB@Ucy@TjZ^V1ze7npGtjS_5GgG^{54DI`DIe_Jc`20ug#&cHiD&oj^=Ls}AGjJl|
z2+K%?=?SAYEzBv>+6rf7g2LW*I%NV2@w-QFu;gZ%1UyIO&PDks&b4F7LcxAq*(z0l
zarCsRrDuEOcOY9AEh7;J%xT(&NiOHZr&INXe9e&ydtvvFR?@>3%$yCyd_r*5H$jV5
zU)5o(BQyB$v#OdW){HTLj{i4}@6rd_tB?u#8lO=qZbJ_$u&Vd!Fdpy3V_px(jA01F
zh7Gr+z{c5!`HvV?`2YbCj#67fbNCpHLd`oz37Dji>s)fzXnq18Y+ps^O$5X7p0|2l
zk`UGofr(qYR9;K%k-$^N)4XTw>pa6mPQ7-qHYI8|59zL{JypE#+_*!cJ&(<YKDN{`
z%@R|CW0QcQ)La120aXY^%uTwQw$sh_K}xiyd^iDTA~JXv{Zu-<e)7G55?#r_o&Qt(
z$BYOHWV+KqDza{L@~;A~#jYSRuXca<?GpxUXM$Dk@a@8jxE3@T3p@5ii#}AZHR`^n
z+K=Kj5!}cS+|WiA(8PrEy7=lF@GaH!M449$8%y*7pDW{^I;PA3M(CT$2Zd1AFA*MW
zuP92(F-Wb(V)Qs&aA{ahYi7(7%c_Fi0ix5O)zjht?bGP5qgF*P8HwPp>0`0pDm~4g
zOqdSSmI^k_R%@m2|H4tu_qXR@B*&NKb?K9s`V-cJdZy?=#!O$vNdMQ;iZQVz>oAeX
zdB(c#q{c-^CVrW0p1B4E^xdQsD-le0_#3)MU41M&X1CaiWjC$Q>v?0uqd+}*&*<px
zxu5~w?$y4S*9uR09~-<I_o3nqnZ%MYS<??kOpd!SE$IdCDU-h~Y~H2DktCf(KOVX~
zTAAOQTLH}bj#j)s;|iwZ{p?+oAiu(QgVkloZ%Eg<8dNE}l+yRuCq#Vq43t}9*ebi1
z8bU*8mg%iN6wlBlw?}?~KK(X1DCX!h>$qt&hcgW0Rr=|v((*ieT4gb@^R-P;NzHFj
zmK$f^Na%QFZM<}Qenl@ZEB<2^e7@vZt%l8@v4cn7J;X)SG^WIDI6}THMH@oBglG3L
zjn#028f;Y)k7g3Ni$4E-fBsyiRdJj*!U%sjG4x{FpS4G0ivx!Yw!5BvZX3S8cw|>V
zHxa@+LqJ;?tR?T!K22F(`GNtPQGBC7>LbM1+xC?KwO>6)=y`Z!<3Faw0E@PeM;KS3
zSL%qM?TbtT#*@rHJ#u#6@8N;dNw<CR`O$ZO7u37gwHUy6DSya+GKHi3^P?kzZh?B2
zo_tB+`N#MG><wSt!?<h@e{<g4e~@!h>?Abn!035Pwdh`ZI;DUxH3Xw_Z-eeY%u6bm
z1JEy~!lLN;xQF^SrySQV+}IILx{g*m6emTd$U@eJ(hKWIOLQ_&^w;=cr5Zjy%PyF9
zH^CCYJ}Slls*_3w596o{?hqEpLGh7%NO4sk`YkwnDV4IdNTgqz3G5BxqZxcrqoHM%
z;KXmv6cM6FW|pGxmk<7>&Wl?Ob@NgAI*_d65y1qz%KE3>z}Mg(--Eos`@Q^KH&@^>
zG*(qArdRqqv}uH=L?A1WDU(CmeM<VD1Y$3>@4rMKTHpNAk3P1+csBvblWrzhhJXKx
zO-V9MTSc#N+voOnUNn;-{9P_`etyS$*Ba*C@0c5y$Qre+kXRt9l&$$LAVH`tmf(;$
z$5k79i$r}Xxzt3xIE>wpBw8eZG8kySk{;J1fNj&(lcqio3Yt%=<L&aRd3i`qQ%dYN
zq7ot@IA)#UKIwx0HNFSi!TaJQni4gcz-QK3cyLjYgpnXIQQTtlv9L?pvA>V&n`i@h
zk+p;pt#6p9AVrf_XbR?Kj&w&XrrauZ-{bA#r!jd<-9B;4u}4N0aaafr*vx5dHmN^s
zVr!OnCg;%^Yxa@KA^}y7qV-;<=)tiYA2pH?QLT`w%Y%y(d`$lJ*F|4+J8MQK=vMYC
zVI8yDACq<Md*ierAo=Q?4QvPATy~-4!AT=8ZWasL9b?gB6B{@|OX4DIaX&|E>pvLn
zz+%iXZ^6V+A3)mO9x788_k~E_caWmxQn5OCB%^`z75Z+3^A@0@;@v5!H7ODoLzt4y
zl4OcCzmT?CHu_2&DIQmX45gkvj-PI}i1M6F_d30q;`4vmocd!;#Y7)``yeTN4xAQ#
z4<rb_<Ju3k?V`;2CY@_)y(3-V=+q9{1sE0A(ND{Hwt%FC%Y0aOCL<z$^vuAF+!FQ)
z$PIKC6!Ca#zPkj<w(5J-9BJH!JO>MoZA7zN&Q(hvaU8_0`#U7mZYq4~M{MI?d%;xE
zWUF7q)+%Ij4yGu^aU6D+CZXSUj<l!nDBWbq#}ik#N{c;4ZL|jR%g()RRU}}cH@8ru
z*JG|0r{NOEulHu}8m6R<{1A)}J<;0%IAP$WBuFl&C@bT-hIHWov!yVn<<jfL23{73
z=<b#3zr;EZk{wD3bPLz5hPLPZ2`&cWgE#pVnDli)`}|H@;@xSUQ?`q@I{ODMk+WeL
z$P&nX)j@6d_du&|PxaZ~lkuVx6+;QlA-m*=B$o?{-z!E~o;4_}s`E91z5Zp1gsa4Y
zgAqE<xKmXx6tXHTUe*vBx0`ql3f9Mm76**++-5@!c+u=HJ%Ej~NTm4jDkRX-aT3nN
zo^e{&$O%CYvFV(8^=sZ$$W*(A!%!V|-XWCg)Y5nUib-3pL}9Z6?l2kg+u#inz~asK
z^$D<RpS)wYtkI7p+nmXBl6|I)Qc~<xjWWeP(0>2Hd3-A2YrqtrB9KjVtuZE%k0z6e
zz4zTQ%I?`ZC(h<EbU(MhsU%}Iw1tN<uV`hHZhP1N6>x0Y(6AV5cfQf8=|rDB>QR-v
zm{#pJ=_<&0Vkl9e`{8EgeJ=5A5dV-Jw?S*t4a*SQ8s=#={B^A+u$W(;Imk(3OtVZW
zRTvub6O=oP&;D9{R&&ZZ`#YMGH)5UO7&v<(J;H+xYB!0@e1CB6F57<7M2z*#{@Oq9
zrrF8|j_g<NT%T0g<BTe6*-Z)Hrq;Ff?x6{G)%D}zhW%fTYfht{5QOvzev8^G*p#2{
z(DS!IxHRCEY{^?2UgN_h3lHaOu&uc)Vn%^uxsty?7M+wm<}{03SAi+!edAEV4-F;D
z8?PGUSwKyzD@Drlfs7$G1G07k?ER;t(m02KB^U83`TBAthTH#Ttz0Q%<xTRrSj_bS
z{70EcE~gkw;j#v0w+2OK#PPmv*AI6gW3Upx%>1#%dY+31^xi*NW4-USZ9vD|@5wFN
z{ms7dVa6Kpx{~`Pc_0R|%;^r+Sqnt%lQ5cWI1Z%BupYo7RZDHo1bP@=->tGpAGvO%
z<Cw{v$Y^q@2B%0f^;N`oN-K8$z9Amhx$<3*ujs0C92UEVNL!uPU*~Ev3JcAT-a9i5
zjRvFrGP-692XbRsu3P;3-SwUo{w);gRF<&g`Pjg<zP;#R>KqiiKYsezMGe_i(I_Sr
z@OI@LO>ZA9llFkU^cY?xdF%5)nD#DhPZII;9Cp%}-tq=#8}gzfO5JDRZlTO^ReJRc
zAYv7p3=A_Spe1eK^(sm!jMlc2Tek<peFooVt87=IT<-V^bp_CQ!n{e#|3(FFYLQ8Q
zznRMP#S3*8!Z2uD)E3&kn@aDu`a}uIUTOEpaabJk*M|m@8)hhgs{x%`Mmz(_HqiXJ
zq7~!u8!&!t(*!}Fu5!>3j$n_}a-yZGG|R5N>c7|3hi|-0nLnT)_?Ia<_cmcMXaGLX
z#ov~Ua{s;B$me=KQXV?pp7@78on<WVZr6`R4iAyLH+g=h%LhKv-jVCvhS%g$E60T%
zxdRtc?nRJ*->9uu<KSEK5zy3AZL8gkyVLyACj9a-ppHIzl&0f`@#qp&i<Ri^HQN@e
zS?FWRc(C-VmZxEzfI8&oUIyl6`<tiivER}a5k#=3B2XUERq&mbiq(&AVj=aSmrP$W
zpGWJ$s8%Dtf`2$ulE_en^O*CNwgB1?a<^zF<t)TN_H`pr_tn!odz+>}c@{xDG{%RZ
z|Mct{!@5U}vIczLJ3+6P?Y$~#0lpqxCw|tUT7?QpIL;EVjBR5>d84BD;_XbIX2td>
zca9d|xk7sHPF0jhr)4q5QO**_wW9pCO5+<kzH?I;?RI-<tpk%9MN6NqdXHPEX;%Ej
zc>)LAP>M|E1{3x__w^~-qeQuk6dW@CV1=WPj%eq%ejJD2+}7mug>!ej6@7T%IBU03
zagz7AXPwqz^YQDJY&ykHZoUNX@>~u-N+9T+37<a%>e!Pp14X((YAgTO%MF1B$df6{
zZ;d;rdn4y|s*(Y~u|hKhrrm;n=f^$dcai>eG!U@<>~LCnr;Ur~GJVnF99b;ywS8@r
zz5HgHPt_WtoVrV7Op+`%{F7Y1)P1ptKc?PCH2VN-7Lu4FKT-n%joL0LDzaf546hr9
zO=jlzkym2N46ajn7z06xa{DA_DfivAa&4&Ple5#s5|m4Mt{hXf4xf8n$(UXsGV$uE
zt=|O{=49SC$<o`n<02&;huxqpE;!^B$tt9hYnY1OR8qnPS2@^AF`We*OoM9wj7Vt9
zn}?-TynuJRR+|h#zqI!hGBO-_0)%0I{{KG=ZNGaZa9MQ@O>iOsUZLK?dT^x$Y*Z^*
za)Q$9EUfIQPCfdl!7NWrAh}}V57}t#^;ZFM#80Fr1G@uIed&-?3u=$`Ug&nSMyiz0
z*8@Fn{BkyQFxJ&tG>-y`uO!wfISZ9*Z-6wLU1fb>r0fy(=4%^E8BCdIbSt$6Yn@`L
zxWUAoLda|rdjqk-H2Q#X4Xyp-+vmjd!>Xr#XT6sM9qan5#z4&V3;ZkQG@&hQO_;gY
zEdNTi34B6PN4-t{3Hu(@>bLj8mfktT|7KJOI@h634-cdDjHTtDr14Irt=+z)J)CzY
zFey`=Or}U5J1<3DSO0ZQ(=bIGnSgRf_XX=Ct4sHHZv_`RuCnLdK$)>>HtMF>*!nDf
z_ie_T`SifMN3wns;M>t@w=|IN^?<EZ&S-yx*izHQLXp{zMl2<3?CAcoMWQ<wK%t6`
zsIaIQ;KAh~KrB!+AEN6ENCs}HnZ8=V{1n?50%)l!?QS-}cfK3`i#O?|0r!3S&A-8X
z^58F=a<A>P2-d)NJu4j~ny)c&w0k0mRhRQpT?$nh*-QI~7eko=!&SXfMj<@AnWIG(
z$!xgJ%5Vb67iuU+tPPFS<|Dp1r50dNO@FE0BtrN3qDi$Bb0m8z)ol1}(-kgQ9KZL;
zlOCL_Vig1AY+f;j^pba8(?H@_qZCGEsC+JvFD=G%OqH@sqOeL`6krxKhhg%DHDC|v
zvoox+Hkk+6c|ss2;nagLnEi7lLD%NPW>~$bJM$W?%lZ1r;&5d7)=Wl$6I8}}$ML@+
zmyM7Jr=Y(LMXt+i<0;v%HWV3*(-c}y%A+??_|i<z?G64K=pFHKUjx;+z`(RH>T8Vg
zGFJb%CuxgIy#t(wLB31Ck*{t(L2m5>rH6V!Wnqb&2XD`LL7PwIn%5+6cR_!p#v;CR
zOx{@gD0GG4Mq!ikY-f4M-+idT=h&($oP@<;Ro|EIz2ayaNuu((h|@9ES|X=}?gR_9
zw~0mEAfVdKyVgv{CzLN;Zt&XBZW#dG-G<6k(MMN>lD}0s*x=0cJP11uOpFnqSQg|!
zvx=6hNfxmG(*OJd%hdeI+jwB1&8+NC=!yS+p}5A-hduN@KZxOkq4rc<6hZG{Ws;4n
z;3F#?$N5K$F8m+SZ7uJr_({Q=OBEWuWH`-=kT?JZ-I`#$5(f(RUQafF6r$cKRg-kV
z6)J)rs(^6u3s)(L_SrxVHBqielu1@DvPUe@tJ%bK`mkMOb`@w={xfMe@-)}<s%kYr
z%aYWMoR6+XYFdxqu()Oim<0(!qW>A$Ke$%U<_`k(uZYEL7mQG2@np;ls8=bUp_~Hk
zs;A9^dJ-rM2Z17?tHCerXFwjW;>rbZ0XZ-Z7)oq4dp+b`O;HROaex^)qnjqDbgqfc
zaz`ucpYc60*d-HW&hlNN+zyED$w1wk8h;8}$A_jN@ln=}j`2C7*KEu!9#<2irgIH6
zz<%h~<K@-{pwRkp0eyY{Z>4aT76N>CcV5=L+dhmOf*Y6Kq(TOq&QBVg|J1MFSq8PV
zpq8KS8U{`H2+su+BYTu7#y5J0F+0EuE0Cu<d!9>l<P2%|;u}s|4PEc)YsGi3MQ>Ua
z$*VG^T|E6SD==sO1O|LWcKP6!KX$GA@ZIOGZZFRe(zn2<eF;DClypCq_s&bUeeE{S
zIK22{n9%&7nZ<Ryk=)dQ*O*sHw1B(C`XsBEKFO8~yj5ZsbP~3+i*tW)G{HZ1w$ai!
z5-q(M>Agc+<1|0KzFf-Pa1<ck4&yH<haFwz^mDjWR=E%BiB6#3E=z!oK52i^Jt*d2
z`SM+@_!da&^$R=&&r$)FuC09*NqFBzVNvPt_k@0&kJa%ylS&A|{1uWI7#1MMX*9&A
z#arpV5b#_}x1V>lh#y#D3J`daRdwtAoc<hot8od@TlwqM3R+>&yUz@%z=N7%;?_Z#
zKZ}8(D4pLqEq;=yMrBNSd~(f6TIyh_$H!o*$wuGpya7#qPSQfVlwW4NrY!nPAa%JU
zM991-pWG4Nfn7m=yM*~en}_wW64j<g-Ii@XN99X4hTmnHUxxryvw`z0;K?wQ^(^LK
zS37mI>%ray>{q6>(QAGP?qX`I9sH-RgyV54(9<G`jV#ZX#ejn?!I`FKQgc}11Xe>L
zkhd2}yf^Idc4isijX6qo>BR#Uu*&0f2_75R`w{MN=AK%>6uLX75+i|{NhYIXvY3a+
zvK-T;Zl@R(VIv3%sKw3QS-=tH^X9olftsgDA0kklE&GQbcQYeD?8(^xtzo?^t`T`7
zQlq5l)8F39+{m6jnp{65zkU*sTp|R=Fj|+-MqQc<&NnkVBhd)B#u@eCeZDrR4SKXJ
z9)P)^2|->lc8TP_T*YBXks2}@6r%aoec#@yDmrB%mFi0eMKR1<VX%@qCh_E6vaONp
zUdHi#l4`Nq>SLS?ZU^LKCMAwOEl!cb`gyWv?CVsQoA7?FA~?u;NKDW7li%mSZi#Q0
zfHVqIZjE{~%raPNxnFSvP#VSM+-?${+Fz3!j&1b%1v?OQ-+wQD-}@}F>Wsy7-3!9v
z4--_9zxIZm+sQp_WePPrsB<`oEsDjPJBSu~ML+`Ui^D4hrLJOBoH)>A*31^opzA-+
z^HDUO9;>Doiwztb)JPM?hoTFB0s|m1^*L(Uk=dqeKeXyTNo;+6a>Cje<L%S<hnwZ>
z9|~EFF73X66c$flMr@{PbM(z<j-qL$>CBcnePQadOf?hI>Cz)TMF?km7x3#=L+WZZ
z!ARcpf`kvV9*<}RzS12O?Fq$&so%XBYe)GUr?N={B9CJ<_`E&|z9LNjFlY(9N?UJo
zWv?s__b3}bNpLk^TI`W16u0cx@64PEZ@1~qDhw)_hGEvPUbg|Kr~Fe>%P7Iu(TIfQ
z;AC1!asw-(>S>;Yh~Js(1bx9q5hj}7dUp2-Ipn-*Xl9yOL!(b9uOYa}k{F9#0@P!a
zUID3hN>n2`w^!o+(fHpembu??d9QTbo0Rb>&hu{1FetTn_~CiLUSaIcz;t$fIi}A2
z%GeprWOfG!aK24yJ^{9oX*7h{@<4iUY%9%kXw)y}Dy1Y6DYZTXcupY3O9y`#{G~qg
z_FZ<Qs~Hj_Q;yf26?eM`B$8C1n-kMQfXV?)7BQh&W#jF-oXW-PJ@TRAiobg_e+qm2
zz=wDJB*}6tzA4d?gZBB&AYn^qrf<~O61Wh?qW4Ql`wO7WvT`t}e3N|P?<<&;tVtFm
zRboWF{Li15%26pc77zMi9`X|u&5kAgtZSSz2E0Pe+l`&ApbQip_b0f#KmI-CK7509
zmN><y8_k)McS?Wtrg^tb-}A=Y)xNw~ID;m8mh#%fDN0X0UmH6}o$1l5eZM#_v3FTH
z=%v44MR#*TvR)#*$k$);3r5v&RKLz{dJp&SxZSMENXjsN7hktC<Ujb4K5dwz`SkDl
z0FDM<<)D<~PI5(&lMA`A6OrfUs3||mP{W*r@4fuGYY?ROW#APVBS@0}hNe=jc_+D~
zW_p0*d{oi!;ELrlr82F@N70p4`efb7H`4)D;RI@QLHyjedk5?2qS5nqk|)DK&6Zr(
zM{73Yf`Hj1YIbI~d{!us{C(ii#jvn);L00F&;t58FTIhk?8$Mi{y-%6?R?{9ZQ?*t
zS)+r{)oK2KHVgd(AoW_wx}g)wxSS&Atp73@JUnNQ_H9jC?<(M*m75DYy(>jOt#7D|
zZ4E>ll|<-6J?9=uHLv-IlLsUK3y~6(PYGnR{*WOhVn`(-gxPSTG))7R)hOs+Sn;m#
z-?@p!j!ZKv^j}hpMblA~L4x$|4wT^nwoEh0M3%3MK3i-GTJQ*F|Ek@0e~Q3iEcyZy
z37pxF%xe2wnZHLz@lBbAAf=XdMhObJ=G<1M?U>5;B4eqJvjZ#9Oq57m8rNj-YDyf1
z5HVK9eRh5ur2P)YI<AyWvnl>yX-L4Nez7XSs1&A0SBg9GO7CRHKYJG_K<W2C>2hae
zc*<zV`Z#>T9Ia~NytE-!cUeg$3Ah_o<#dXX;0aBTjnX#;Mh!6I(lcj@3zG$H7_L&<
z8Asu0MwX;k;q35nGcWOlLbA|E`W@2})}HHhU>qkVQVZmzgkYb^-y%3A0>9g$cb4_X
z$sEVx(XwYlr0d-?OS`vGhdcaiiL7|hjSf-Mi*mS$;BWujt`d}g@L=?w21dkEE!)V!
zKjLT+OGfJ;a<u1jtIBLx<M9H)&oJmq7*sb(<TD;bHjPI~MolR@-%dK85N7JW&c?ZE
zit=`Qa~9AJEe0>3yj>fUQ3K0tzZriFxz>cjm1<mt0AfW!&Eo(S+z(1Ln<el=dtNJ6
z?vEy?WZ>Yf%MbJs+QXkw845H7!fijD?!7s48<hMkwcLys6qXTLv51AGaf7}R>;E@x
zG7v#Yp;btPs3Bs29y44>1<;sq%E-`mqZA4OCf!mZ%qS_81M7g++O4nwZdj{pz{+E>
zjQRk%uDX7Cnx9H(u`@B&k#1```@%jUF}$Xcyhh#i4F}N!4JZM^D6x}u_|3hynA}9|
z$rGq4PjFPm-hLmRYuKm!f_WR~{%Dvbj>41m-N{$z*Z3GwgZBHu?*xUW#zDxlnY%H?
zr{|R6N|{LP_u>4VWSZ81K*NmbZ8lMU$a&dB%)R8JxQ9IR^=GIfjMB4O`jO0nN(Wjc
zP{gwDMf;W6__~IJ52a@0Jgcl--@XViaGUt$rSACJYwbOW6oM%@jrzsD*Y;-Ve#O=I
zd`U$^O=pu556DZ;TE?FB5c_IDo=#B4Z(_{ce}}0fDSc25^&nwKOT-d5GdCXTU)9mx
z74oHk5?~64-em>*9RVMmlrj`umpA(h+qp8TpWiH!Oczelc=S9eEK;$e`yMv(BX+Wr
z*R}bwllys#l}S$?VO`f6nOIEUUwTfKW9G&~AV2JFu%k`x`O^f}yyvT(E=ARLBkvn!
z`UY%D1!gULDPyW+qs*chrjV%6qC`i!Q9BJ-3<rL>uiQ?mZ(h13KZz6<#ZLLnO{obJ
zv?AHXnZiNKED($L=l1W=I0EQ!JI)4;w#@(wWhX5j!gVfaNt&!m4k{e_JSQiaL?6e^
zMzUg*jY769pA{gEQTBlgrg%vp#_cSVA*1|P3L;B0UFdTn?om)FtTHAY?>^E>)%}-7
z-#<zrOeYmbtO>*dgQn^3%dE?+E4gw$<l$HeZdI~7i)fykX~gejq1G_3GIwFr(Civc
zaYnqFf6yxD8$IC@KLE;ItMdO`-1rc)mm~=M#JWexxD9lWksp+v?zrg&KCo9YNyvy>
z3{j;ksrQZ0yLT>7VP2)=vhuJmBZJGLH0T}nR!1dqp}($EZ_0)!qu45SqDOdD=?-7n
zfm+&xQWDFp&4kI9bQgx-yAlG1mvCB=1~yN8Z&EA7BNFDMt-RDyUA=YYaMBW?$6!GR
zL(4zi7_3WbB$hA>0*kvgiVsbMZ~20F_tS}69(jrhz3+QAI`^76OY_mMCcz9qkXO-d
z7wIdR*6}+Fo`HL0h#N6UI!$S%9X=y_dv&iYeH}OR*xLN-OjxXJMSdJm!#(KUoL#L<
z2YO=HvDI=YXfL843pSD(lTL>PYEB;on{V{x5eAZ}1!cVq-X<}4pX+Ezyu^L>@jnaT
zr-xHVI^<f^^XN*Nq;qTW$Fo%PuC2lJ6(8O|>i6lty>2*tQp5ux`m_G@hL+Dx!to~U
z=WkZ%@2pC%uA(?)eiv5`T6iz|vz*-7FY-r=A{M79zB}CA{gob&Hc7E!Zs9Ws8Jrx%
z<$vjFlk8<viI6)<6_w`6i+5e<#Tjm~D{VY|dy;Cp@Z_MG`C!n39C_Gga}}S+TU6Fm
z=(o=uBv?D=+t7U7t3Dyf9u&|S?PQWHFnxQyK0AP?kNhohKS?phCTTN3gYqs-zQ6qT
zd4Ec0qR{|H)nO3sBU-B4i!MXJprRpp{_-R>)m8Wp&R)&cZT{f>?&)-Pcf$e!_%=>-
zh#5vgY4S`gow}aqw*_HqD-(PC8P-Vevk=I1it_i?eY!sI(0hZkoLZhci$*z7??;w6
zLwJd9@6%Xh(7%0t{zA{5`jKGz4J$UblJj<}=qGF9E?QVO2;xR6A|I2{3pD4ovWkZ{
zYtCDFeWhM-pFq8CGXuHxgNZ<aAkuzj5E*4mPq1g2C$XYN3tJqUfT;nk0oG-^`{&A7
zu}mIqf@73-;j&;`1|_y{WiH{~KL#jW+&DlZ-&pnLO&nJbs@O3wW{AV+qd&pXV(~B|
z6|#OeD>j(ZwDHK(|0QTAA5tVfjc1K>j$j|+3dhLp;+{c$_DxE$;di(FB%tpq<i(Uv
z@pQ84@AT;jYi7}AeJhR1^L0;dtQX9{qiK)-6SJ>Hp-YPqdc(SC2K!C%0_4vtvn1ll
zHd|{Mt&OMw;lB_GaU&)l9gJr={fzp_)uQ<DYw?gUNY5xcM&-4J;8dMI4eyK@pv<89
z&64s@6Wj`L2)S^kfmMk{KVHv#?D9)CMkIE*6Xv;&-{nX589yD(^_b)Y#qzTP9c3o?
zv|b^|7QbV8=T(o0j~|Ez)y?FYd2YLVhB^$q&JFqiCyO4EJ|$b{?E+hWq8l>*F6=FT
zjMzef!o1nkdZ<B*%d?91TF3~|$H`q~sP8QH2T{g@&WcR8XhNL+!kPmiK#~PeH7o4@
z@kV@*{dh5I>s}Shbe-^#!_AH-Lj;owpKv=WpiKf#W#)%$HKyldPa02S6O`N?nN7cz
z2LCKefn3*arx@9Z_R-(8nBy!KQQ9)bpzI~Td0?71EBo5yE5)A5kaGi-A(v|<SZ?LA
zwjMW695!m2Bf0}KY^FGt@*Z+`;oXv4(bBOQ4ivV}NFLD3uTm#!Zc0QjCe?jKGfVG8
z3$$<<6#G-jCp;W>ct>cmk{It<s{-ZANo~_iCa@#e4s*c;foOXMQMqD9j+T!<kQ2`<
zQ$WC(Rz{gQm6Z60DBjpsexPr}aU)F=-nqLHVGD`|V78-um1oz2=VUX0ama(*GEt1d
z)WVk45JA{iYk5em%s$By8w3^x$z*Es9H?1I8&e`0`y>+!xXq}yU&;mMXjGxr3tnaB
zESr5I=ZMa1>U%hW=reawfk?@5Mklj)EkWYJ3g`Ob!7ZDuO&jz((I;=NC9NODP5Vh>
zEI!KsGOQV{ItkCi?|QRU6D<ZhD|~If^QuZ(oU<Q4KBfGD+j9q+FTZFFPHrz>;}u#K
z-mKYq)D!BCLvVgYelQ6Lm;nPGp_o!HDnfK`mQyzDso_y(m)*jp0)}~9jw?ae1Pa#3
zAb)8o2X%Kky(R1mizk^31d)5nMy~KXF_x3*I{BPsU6Eo5WSk^QV2I6TB^4v7R4pTU
zN1+zh7bw)>%FAk^V`X{h++CX9gT=uUlAk$A=!8O*Li(G<S6Xke3m=;xWTPA_DiF72
zl@3X92c^ePZs@_F_=g=2n0Bj_Rrg3$pe^3-bBq2n@qW+-CSF1n79`S|>}5Z8FkG-Z
z?Oxkr&eOH!ti}&neJ_GD-vn59dqMc`9xb1)1vPH;B-v1hPYLHq($gPqEUbLyN!IN{
zE}jgTe!opoIQ5rh#Mj_fS~fs46X{Nx9ympzK??!g#P@96h<5w@9is}lbOn@vyXlR=
zRM*BI(dw{%xMW$f{W?|Eu5Le?FA?F>Dnl*MNB&_x9g{frc^vfHBcd5<RCH|2Al9cM
zd4&7g7M0-}Kxb+3UkdXO2I^r#yw5pmBmF+U?({2?swVLr4T>0$h3=#nkH5#cFGVm4
zP=l=Z<PEgDwDPov!2Vi6`I%qK)L2FD3gvtjrEP9lRdM5TtkmeEq&B2eCA&xQA>c>a
zEZZ;rd|Qg~rvq5|fu0+<wvdGL>5bI)czBzMpa~=YckJx=<a+jhC`7sm3Gm#@V+jPj
zp%^pbWf>=22j(MrlD_GUpqAgvtf$t<I6e!fRY-dO@5QJmP?#cHF5T+E<7T+!H`m|Y
z0X#+~i{2|g3t)H1Y+hASHyi0^>}KJGOtZ<=M>x_kQyT_ivC>fWw8Ox@BHNnb@jVjt
zm1OmTZibr>M;~R(OTk-u<wrqGT^zR`{h~y9Q<G|a7d5Bv4o;{pXs)cUe0xNIGzF)=
z%f<I>OOwTmj2|X2s$)jeO%VMm!Cfk6!+oiE*;h<wsQG9jyHVg>+`%6%Kj${j;9Y&c
zl{7xymP?Sxeq9~7o1}_U_4LV1PV|FqEsOTHlX_%UR4<lnt8Drz#C6pBQ1?1XE!c^m
zZ>evsZ-qXN;Hv10T$i2QV)Qg!aB*1936L5t=2cDoU0ayD`^Np?68@->%&%RuCtB)_
zpr)QRKK6JP#xz|ZWM1@L=r9katoSA4wWcsADs_0LZ#V0N6)q5v@@*}6&T?0qTY9Z`
zZvR58QTp5DFV_af74+%P??>qd8Z_u+@}E0=rCe!`|G;?7fo$8_44_!RL)IU;W#@|=
zt^u&QA2OuU2XSIypQ@yhB-dIlr#;0kN&4x8&FX=?zESKHkdNnX5R_?XJ*X{H(~4yH
zk2lQ=4Z7?g{#wzx7@<%)KR{0|6l(f}QC5G3WFgoXOKPF0ByULMVwPlPb&$78Qy950
zS{*&Gzks*au*%=VyM5xUpYNhPL%pz)KT;>cJiO}#tAF3Itq-^!N=;zsyPsi#cd<X_
zXylmu^A~?@RCb$m+r3PlkAG$bYYW;koqsj4z9ubQz?&j0b<+0&x!K~9?rweIOb2A&
z)_RHRuohyBLX4CAqNMxQPyN&7ZTe>7<knB#Sbl?#zZgm6v%2}3I*F8^rKNPV_Qt=^
zparrRe(DQTeLEQ=QBo_f5=;P%uY~2Pa~E|Zq2I^O;X}zw{K+Yl<ddaD#dBeO;b^Z*
zGDu%<aG=SwXWyUK;$}6)TVjfrY7hqPe}6K0Nr3_xGid?V;baJ0@CPn0C`kQhza3ZD
z>`eT)j>0!pHf=N2Y>>7>_?J)S>i}tNBtA;#6;j7ooGtcRfkplIcQ@?ZKr)T?B+=aS
z&o2f?<Mg*l1|7P`=ms5PUZ?ybWr$#(^(UJDULYaGAi1CT%7sz(Lr0m?$3DiJC;=tq
zgazc?Tk0HS-eJ#1M%QM~9|=f>I686JmPQWSUicVG%J^TlHq*)(#(3s`U9=hY1HuJQ
ze}x5M>Ny)ZZd4pwe|y|N2qXydvu1+;ORZNiwu=kit6pOCzgkOx)>-g@bL3*$Nx4bh
z`eaC7hf6(AW}8Ia<@++H;kN#uL&HBVmiZg=SvYf5o4>Wn-Vq(=@9E7P&@2*O!k8&t
zW2kQbi!JtY+^yEsWa|&*U{2rRf|}(!-`UcFReKn@HuF3^6L8T~OlPTAdk+`Q1t3d`
z&sBcB_?K=a=_4DD9fF=H^L*hdXpm`o|DsfUpO=3~AlSer^hdQKz<#4U`L2V{IVWeM
z>QwU}jxUnFTK#IX2)S@}eyCRx?r2IQs|?&b$p!&WuZ!MW`ycD1BJoVxW+AMHK(nzV
zppBAgj^TqLj`4K-G?`h40l1yop~D2uvm%>nV}}@<YGo4Duh$ABZ<m|+a?gEZ$hVwM
zwU>bIV`C*305)=?boM^Fc_yZr`dOgh^^8E&=;(N!JuJHPJXZ+#Y8Il!6I5-q5va15
z6U&t%N=(nHdxkJ&MgMd2pNf7Q>%(m_{G<;!2wNXjrV~#4nwR46Tyltiac&*a-u-}g
zR4V%u)Wrlmy!WlEy<Z1wo6k7g;}q_Z2u!KWIOGJ-(Ma-v3pvQ<_WVUQ0!w=dkjn^u
zWme|{Ku7+67a0{!@`8Xzl~+6BmcadRZ-)Q7dxM7ha4>%dIt?qCuH_;Aaa+`q|M#~w
zo^*QgV*PA#VY6WLCLhw0?E1k^d|$Iyn0Tp7w1fb0000~K@2@9Y-iD?C#R#vP-^n&9
zKfSoW|J&s7Kl{c7;$X>j%8(~s`rohqm{#T^Xp6S|08xPofH4Ba{bBi!9rb_qry15*
zP>}Pg3dMgOUU1GsU`3yMHK&1syb2N){{8m;6O{CS{u3(;_=iHZxSLY{%gaR=2OtF~
z;*uC+b^o8c@O!>{fqS^C(g3gZxF+^X@o|_9HNUa6blkDV|3{Yge?B}9f79+^tgGV~
zjsNpb;6|fN93ITKk}*K)Sg>DO=LzZbw){7p{{QR=D@8m|9A16XSp2_vj+B(>08mtj
z|IbVQ<aoWeKlP}XHPf9+o39;C>V_dh4>tSiV-@ZwF6>+>-ujtJw(1wiLN?C0zPNup
z^ARKLhY+7-j_3HV#pa)jjv00q;vn!sykrZ&tO|vWiRgN&W+4CvzAb<Tq^cZ|!%0S`
zFmhA<>0M}>E%OYUX~8u7<yO|ecfLKOD#r#AAKy9-0wg#)pr04lN%KWhnRgd}>WoM>
z|3ozaASm_4V@@p|s62w$FH;11oq$f9Hb6a9NSVf6p*oRI^SehM+xnP}rXXO9%D)08
zp4hAE<izlci&|rhQI@;K`<oyPnJJ(wF}>J#CVKGzv|R^?Y2$V83V(j`5&j-n)y;6N
zn~~b^>bzI)Rrt$z`Up<jW?&V<_Qn*E{=!y8_s|i{5BM(!<~)a;Q`#Jt9``)!AqKEa
zl6T01pwtKE6;p3U^SIf=Fa<wmA>;h7VKO5m1spMfX6e3vA0Y*nU;z5gm9SG~k%kW&
zFa)>o0&qm#qA`HQ_bVAdrx{MFi}~Fibq@lxV0*Wr%dLW(=)y+6S$Y<(n1_yXPBqgu
z`|$h~peR}xF#||1X8-taE#GS~c>Ci%Hm-!cububi2QbN2YYw+ZF+l<w=|4Hvng7iK
z@ORm{;)ZOg*^DYuI2(th-EJyNzf#8X3tYxA#p-p42)F>4paQiE?;VTw9{`JW_jfmV
zL~KUX(t(XA&*#0w@Cnd~59rftwryX?Z<O!SEcMMq8m%n6=vMgfoRpHS?tZU!Q7DYY
zL#0TC9=WMJh%dft)_gVSTC@M;fZ-x{>@oV60KJ!G4+s43E&wOm_(=Y^KjZ$^))PQ0
zrvQU6bqkEAFyJGP>|5+<I30if2m+9VkUy{1xP$D|1CC$t1BAcSABiS24<JmNP%^!O
zFQKh30CEID3IO$Np9839g&RX6D=_0lBrpI@bniE_N{8e0sTj5=Tkfi56J;jKwsUu{
zN1wc3%KfUr2NY9_6MP?_LjFK;k}y87kMlWva>4(=Iq)db^uf%~Xe$SLc|FxFmA8~@
zirm(hMD<d3pc~ZVIJ)Cd{A+wM!vcyxC0`>jumQ->TR(Xo7WQNROOeH{z?Lp*$0W(~
zsrJTu<iP=3*&71>3}E?#q|7I_+yVu30!V@5#DKB%Rj=k40eI7OAc@yzEnZ*GWCl;A
z|5pCnU^W_75Foh~RO6^n7-eKuqO)2kJKwA}CDVI{6-4Q6_`>*V7ij)nbhU8NtpBr9
z3c0133Sfka4yP$S@-jpjUOIi7qF*xC(Uhz<aQOLS0BANkuTQUKCkeVV4mj!*Tev??
zY$@2q)ryAaC?_0!esc`hR34(5JsiNK^F=(AZ`OekczsCcu?i?~QXocUTRK2Af4BnR
zbdCU@l=lQkFF!xH?_iNKm%6_i)<3sjvq^WsmwZ_3HdA?Y|GhIc6VP6=mA*+%7IC+5
ziRY>MyOab}?L32GyZ{Ya43O3Hh{TS5D6Qjl6#<iEy*5<J^Fe3P7FPN9p+jo_b6Zks
zXsAKCNlk!=n%0bn4GDDO9=UigC_?li2ymN>iL^V|W<P)Z!X@pOEqEIOJUoHL>xtUA
zlJsZ=4>N1>zFD5!NUr)GnX<-^oQKItdxZ`o1=zqaA5;`vSkx+ywMEI;lxkK!B@E`?
zJNc;zDLhW3_gf|MJZ!~G4Z2;S|8xQvgS(2GueakDx&03t7Llb_!1ll;TH?4zC?7H{
zzEyj{>IOJDDDRA}f!ElTtL4^1KEvnH1{Z!;<7*~;#ADtzef0<K;sapN@x*H#J+0M?
z#zD7Vgm}2Kw7dN1E=fQN;8+i;aW{<#s`(NP6*1?%Sz&*N@98P?*+?g`nKCcvvYkF9
z2)SdLg3bq8xn-okQZY<`TEx~GwSg~_hXcBd@sUJJ1fKb7Pze0f`x(r)%LAu>rYbDi
zQT=Z(F#M!mc}RO4-JT3Deu>%QhYFCtG5avlG#q$dxefq>>6k>XubpM?ftS>0=kXXM
z)wxC+GqoXn*9VyW>ebNGmrvPHA<wRg6-#Tr0*%&Xa{^o2xtdOY*DT)RI5CCMdMEka
zU2WhI&48aFR@_3eF~O>X$Dn)2kUMa;XnTY`l>|)EWSe2~#{lJ%_kkO}Jz=e3l->X&
z%W>gPhSb(yL&HQOY=Hv!?uJ-t?TiDv7hl2$lJMVn2Oc<b;(T$BVK1)#0QeQB9#2&@
zj(53uNti$vQ9tz9@gK6dqWrII$_&rV-^aM01L1GvEDf_p)Hme;z~Js(U_-a<y9l6n
z{UOHArHtm%{p9}&V*M==kVLsHiU%qOz8S`XB!C;~1JSF(n&K6M?k|GwCsEQ-5LbUP
z?s5%ZKD=$EPO~sYFo<1Y5DnB}^nTuJa!Z80pS}UZ`|0VU^;ca_Jt<Ef%h0SR!b1x<
zT*Lr2CQggI!DW*ZuM<q-dYVr6T9+#725Feig@rZfD!h9EjS3%FCqBjt_{D)&RKoOH
zN<NR_ngd5B0)>r1KARp?g}R!WD7W@fTJv)3qU9cfe#0Av;>#kp;8@m_xAGl*RiS=g
z8M_42Y_2Tutf=0N+U)dcBu6%p_K`g(O}-Yz%)rVs;u-f%RUN>dAKc2%D8=zEUpo$y
z#+Hhh7aZ2>zx^=0wIO|9Hvzu-UJ;5%FKH(ENKSTaIWGEFPd4KzQQ}BY-#Z*~J>L5`
zR`zBW-=H|>sEzkW;S_6sity47@QuIV$G!R&JMSL^J$Y7!Oe}WQ(z|duhtAf~dJ|xN
zEukd3F?q$i{njAPR(Q$`wW05M(J9h@3pFPgCHt3QxR|x|nY8&yv53U=tBMDDa)f!N
zAQvS!T(^MCQf^ifEN;;EE%5=Vd=m!1^KL{TNPzACn=kTr%7yQ~q_Bo>hO~R(6e_+5
z!^QM-VYU|5eqB1}4t@)x*Y9;)GBii&4x=8Z`VsBz5zDXa=MUt%U19<gg}v`eHBG&z
zt!EOg<2^MqG7tEft=6)exrf%$7r6c3WZS<r_~G0+^7j1AbqN&3Vrg9HLs>-YOoZ$p
zwpp&1re4WT)5pmYZhampsxpA}U4E?2J3tC#RZY{T^>DTHU?R>J{&j@iwPZQg++_mC
z(^F@ZLJRao+8uU|&rn{?HgaM(?`2p>xHNFfq`%3CO4o!PDQ3xS@CajyB<Ds0aW7(V
zTT47Y6|rnIOF$)qfY44u_B}Inra^Q}_!8`KnPTZw>yL?PJAk$UDc*H!2|8jNfRait
z4aZscV`3>_6`5%P+D0Yo1to53IS=EtcdGOv53yGR)@ANutLs>XQYM;E2_)PoRrDHQ
zhcfOdc0uO1pGn3eJgYgS-P7KHm21;mU1IlwoVj}xzYN1kuZF8R;97oaHN*WuGW&ff
z82yx{z*+V|-s6bwX8x=#l@7fptECkOjY&TVQ+f^%OU~FBI#q;{ZIqL-WW;wmpFf~G
z(<A1XXT8YQ!O~ZUVbV^c{&AXCN6FDR2B6oM(f7g7C{9S{Ag!PBsfz*HcFTl2mkj23
z=vMFp^RN3L2l<I(N|!n!XG%3BK6Cq1Uox-Nc_ub)Z$Gj6Kwq{3J1F$+Go}5uiB=JA
zxTsgZZ|?f2lPK}Q@GGVvuW_jlO&HCCw-tci+i=|%-1{c{uW21ak1Tvjb`c=yo^VJc
ziPHPypMYv8!1}sF+!MjX@qK+YZF*3CF$(hrPPI$nSl6T-roIv~MD~O{g8glvQ}HGN
zoTr}Wu|bssI-#yejCsz!%?93JPR5=#uw)b4p#{oNkU}-pFvtiqngFcROf$@hAuh6n
zBX4A*oFZsLe9_*Y`M~0h^Ruku<lXfwORVo=nnEjdCzSQt({Bjk`HfRYfC5%ib$lnM
z7;`U0bMS5?F4{pHy7T_gX7|W@EESEbXUcY%Wm&spvNGK>ETb!x*YQOv9DdVlEJW91
zbRSK+(_SWzfLXsnYsQ6E)6Zf6aav1FYgQOt;Q`{}Y6baFhS!tv)FaQjEQ9rk))NXb
zY`2>xf5%M6h*=VS7jQ~A^I>&%LN4HMpcKGOV3hkM6@Z3j_upLrSs=OS;?qCU;Tks9
zf;u~I+inxZ&Mfu8onqiS3mvc1jSY}I66JOoVaD;xy-mBV+|WRG-dj_z!ysHpCwvL6
zs!q1%@+lKO!1Fs4bXTDBGFI2X!1DGJ5&le0L9e%tb5(aaOOYSvC;ogyuR|qeMoF5o
zo&W1Qt#-N&aQJAxJ|d&&Q?bj~h2^=TfuOy0e|BlKt^TKXPWey;?cYLK#v7@wQs`nT
zOc4{3xX&n)BfoI#i`;P(DUSDsY8R-snGas@ei~P&^mUIW@Q*N#F<(LQ&T2k=RXs=x
z-+p^Tg-;(8jfp!sMrs+bHqQADKk~`sYCQV^T^9uUoHBEn1SER09*}Ne|0BM*_-~Df
zxc&X;Hzv8#zI<qCX0U?TE|0h6l1=)LprAcR(?1&^2QO4~!DI~=ldZ%#&3RzEXZ(Zg
zog{#e`$n@<T&b}`7yV^4jZzan|23L6oW?c|%yevS81LTkVvPC6BPiqb8{sDAUR!6T
zF`P}-QDxnQ=R1GkgH`b=B;=W{dW#)DIkm-$72?W?n~F{t7ibZwaDVykzpR(fajZd1
zYv2N7@}3)xE@;LplP;8%+~-RKnOGX#*=SR-==Shx_ff|88F!zZ!uPNdE8kL>;JQ2j
zDsl++q%!9G+%llz7n-fdc*U0=kIcxY`NmaGD|R&{O=CJFwke(;1!hnm4Hca~LQwUz
z_5>4;>NwwWSZtyncGJX2;1zuAga$qh!esrnS$wAnk?dYT9U;6ysbN~rc7aNhH$I$5
z9oYLmUMB@v>64eR$XsTJ3=)}{sbK#ro<e~hqQ?0+0L?gIW`8pTlz6LflZ}INsGw)2
z%^vxz{)c&Wd`O=y@smIMwTppz*+LXwu&78NB+xh7*zx?mi8ZKcFeCQS_Ud`~O7-SV
z#eUdCFi`G1Dpq*C$4ZQ(c>C1z#K;6#GrtO(Y4{3=(XNR1x1Tf*ORffh#(Tm~W1Rl_
zgon63v+!M$sPr3a`t-l3`U=0O<7VBZyTc`yl<v+2Bo#qGO1e?Hr9m2(ZfO>z8|f5;
zW$98<T43pJyz6`Jx%Zs^fY0)q`OZA^M9pr+NuXR;Dj9>lX%SswzQLP#z660I`?ddD
zE*gV~Im|r$D~r|)Kyd%$%W}wnV>LP~7zSyfjJ|J>ez}-Rx9d+eWM4;eSn3_-(T6_T
zQ(4Yf7oTytJDLL+z5^-=4c$Il`LoQVF4Dh7e~7d0NEmcM@&4}m9hCG!6?nEHqgSN&
zIt0Z!8(U-=CB1hN9Z+IqiNU>!hg8zg(h`L9Olc=LJIO|`s*o1{Ot_n)GxN2<$?eCK
ze?=?BP5vF0z0=ERndYcf{+U^M8YN3*HRyN88JsK4j0v=UgPxO?F)q2!9e9M~y^P=c
z+8@P*FZUq%Jk5Wk7n=3nHzF;fnvNnA^XVuR$lMf~cV;ALvn25@u*hLdv`jbfE7N}M
z>cO;gB(;BbH%Pp)QTPfEuOM`n#IP>kACrFQQ&|ISQ1Og`6w6&HZ<E}Fx|D9}K9+P%
zp(lqI)l-+4?4cBbCGgwzxtZsbl9H1BiV7A=YacI3P}Vylbrd!wKQ}&IWS_Kx9dxDo
z0P~SfL_vjt-<&;724(J^jUxp4m30>(1JdulaqqSHk?)ph_JSId-uPoY93s(3+%KH5
zKNI*+<S-aIlT=528!$?gt5DzIhM+CUi}kJ|orj^D?5~LQZ8J2p2$;yLhFTQPclG@*
zrfX@(Fnz60`$%7L7n*-qsA+_<D+VbZF$%Vz-FQk1zjQ`^hp>p<pT1{>%p@e4F@kvJ
zJbXVAg@&(@r`{=VH2X-1z%){eYmlj}$c8|x{bYyIJMyZ2%Vtoxfhmd9W@+Jj^e11N
zUQBgi=^_rw5+qb&uz&h*|JY6HUSVTL7q)(q2f;oI!ybl=`!+ZMfdI76+T8gc)oDz$
z46{9hIdPc@{`Yj&&(2Wzb*z4f-oe@{XgfJZXzT$QW;D0+%)lJ!c+lU}#l@utVK=Sz
z-vIKHIzhb-^P|t4td$~3N<)7uTW9iHe9Dscb>9~-)U`(k+M~-&w(sOe7cQz6i#&d<
zG8MKA5bb1M`VPdobw=+rABaBanReq<Xd}VskM6DTBD4CG6s@eR(7WGagVD8s1dbq8
z@y8$a^}2v0MN&t0F3|!e=dj-{(_Yv-tncDUb}IE$^j4?@o`3>hfQC4>1M=2oJ&tOY
z|Du7IaX-HQ_CDA!$;x^=g+wkRNmuQzrYiYow<mH9$#&Zuq=JvQFefP*3U1uYoNDj*
zm*#qZSwPany{d3v%^H+ne+kNg1VkHezFx{Ke3^-!wi4G<E8?csx*A2ID-PBeDa}}$
z5m)3tR{0};EEuy^E>1W$O({pZPY#`tKD^S)qC(+$A3Wk8=|3s2CqWbA37x>j9`5ZQ
zNOVR9!+S6*PZsHW9A7We*TdgLnZ;<$stQB&%o685D>JoyL$Fq6$5o{ZSsqOJpQtbJ
zmRxoOGreq1auSHEV%Mb~ak(Hsp}k4SV){*#pLW;^se<d#zXDP8=bo9~@SaB~&Iqpz
z;;crs+~CkNniK!k8)|cOvK}ZeYg+R^;kru)`o}rm7G2vcsiMW{nRv}1W9>Ex1~aMx
zgdsR?sMIJ^)B+6DhQl@2NN)!==J=u--Dk`hN!4MVPZV;oNjz0epea4bX6Jy%g=Wl0
zD<qb~&z#gDoN&-c`lz(?X2HGnkZp_P%sgPE!hl)%6|<JxVxHTxRC#>xk>E8p*rJO6
z?;BpkfPuwRQ7yJA=XKX)5Je-sz(;Ue)rgQ49Ib7Du+`9qm$JVR?gy)a1;WBcsZb@~
zWc|bDME>cf<%ZjP-q{7DZNIZ4JU_2|^$MIIxih-n@FR14>4ejeqjmx5yB4681<LDv
zFik~V9A=)|YqeGUj#4QM?lWRNi#-!r`1953cPh{F=)W+PfzW~nX}(}j(K9lYsHC*>
zyBtLgy}U8wmDf>B6O+<k7_LXs&8pkNPI=9}nAgOZjm$U~CIvdiK)x4;+osX<&&_4h
zHZv#Gx;rOacH3aRqvxb<FE82uJUSK;71S=fkMR6KW}@hI#n~mwOo8gqu-8Ul=|&Ol
zigdU5^sxkTIl@j#3eOeuJaKO0o(|X$GcHg2!F*e^hA4&<kCZDfbACoF2&}3Wr-=H>
z+@Bz!I7jQL6p~)0E4Hrm!(q=ng<9pBvF9HP6@ThT041tkCPz!yB4wKT*`EPbm7Usm
zkPzV8L2C%bumLaHA9**+iZRT4h&bztrq1Zpug@6KRh!iQcvkfEKjsx}k&ayMwq>Kd
zNhomu>2LO<JiD~8%hCPajM*?*oWE&2d*3EvUx*o{HSZgoQFSJrqJk(3a{pmSi`_FL
z7y;@zW0A6f$XSKTGt)Gs5gaSDhz;o2-xv5ZCE1W<RwvoH8Vvp^ab<px2zx&I{zU6Q
z1~m~WpKeeqN2XEV{`01e{bYiY_qr%S5Q~Ygkl6(ZY(_|9)BHeUdliB2ii8~U2fd$9
zx#`N<(0iSWoIg7^Xqk1$hy96ij2eTqb-g4lNEm;dSFJm=CEHAMW5Nr8D2l+JXdmA-
zlZ%W#Ysma@P}ffgzpuN*i&h?ToWk-|DWa=bf$iZxHSShJ^bs-3*%{~oE-cfwQ4dMi
zj>wPz>AN}lt_!X^$7m<<$ST}mrJX{e$tPGmM1}N9^#Oep1mm2)JJU8XQk$`vn?Iwv
z^7@#fh$(_<4`(hJ&z^sQT+bI%y0M7{iNvPML1*sGgbje+fNd@_t=k@$%9p^Qo*GU*
zJk=!_WutPlsn~AIXs9k(_nQa5lPl>UHQ8~(zK<zRC6%^x+Y(~O{6kg_g9!{q1TDA4
zyDv^WQ64LPj|6bxvAOq{4YB2ov8TuJex?1wV|C#`0@V4Rf_i%f3`CVJaiLbr2-625
zqa0f@u5!hXb6}K}1tbnAa<95CNwc0n@p5CJv2zHN<fpjTth9c)KpKUu5bpS_WIPsS
zC^l&(*0pkJE4se%;+{NQSGbDFG7VTsZ>Yw#28v@~Dj*t({i^iL<ZTPj2a07Wl6CD7
zIKoKZ<*zFJtP-uqZ=*3Of)duim^8DRu^3T?*e^e`b<>Kl?G>?6G}3FdDbc=AY-ZIA
zlB`A7sFSXJrHG20&LI9BW>B~st}5Pz%%HhH{#Czr2W8)1#4J*MdlXWYxD_}k#npwq
zIxf!KPU!S{iTD(nU;dH>P0yKRu9Q6ns?Ge=OSYTgs5F>atN}3*XhOV{(hJjl-PNoD
z5uDxD-*BAl256S7%*QWkqIx=9QrvJ<uE%PdhJyv#-*;~zeSRm%owhk~TWjEC&}y4k
zgaXfZaiu?kA2M8oT0D)C{@GoM=5^F5=2+O6v<w7nEapD>Ba<KY9^|d1{;=GMKhiv*
z)n;jvdK3Urt3r<6=|&PRhxO>IInwpL489SLabhsgGHiZLbB^{H5!BlfExh?LJ(%Uo
zXK}bxth_alDd?12I^_8N_;$x}UpqY-IQT%MNnBTYhTVKsV8(>iD>lH9n5|^r6Kj|N
z>;#9K{Yb`g$D#Y!sPR^^c|f6tyfpMm=T9$^z{i;{cT||rCv}&LfYde!UP?yUoL9UU
zaisUENwSy}i9Xm*c>f3eh=j3#d+u;aU}vh~O!xM++%q_z;WqNlS?YkEE*)6O{T*~h
zXoMHTY~)MNwLVpt*p`Tvm@&X%@uN-Gpcv>YtaS7&s&G1$r#B~ORJ3&T)iQMTLMQ@Z
zp=d1^9}1H2&5sUL{3HL>Qr3=llj_3cs6Nfal+-R3uGGm~ARHL(R7_>&C}ebzH&PiN
zpbVCyEIoiF!#>UaNQh^Rx8=ecx{&n{oFG}($muwmi35{Hyai}Tj+bewlD_%cjPp>(
zj<c(O@4ArwThWWW&Y%X8mf=!M@^p6QdR_N8`E#^V*Q-uPLqchvC0;XQjNB_+MM$q8
z3~49eOJPWn8|xN2hTg`SZ3)XN6qFJaMpAaA2KkcufHS)cC(kbg?@+BkPt17n+>#%(
zVma~!2s7{NR<x9F%vTg;V!V0dx851@Q=-q8yO(l{?tW2$_%_uzZau3ZJicB{aMNfK
z-J%*qJyZ?;lnxUf<vWQJT$)Q#p)Y?-kd1aikwQt@MrSF#ol=2;Oi1WLRO1&0epb#W
z?!BrpgANIr)+dp6!N<G;cEZUtV760kD-QX)Fga)b>uaRz0e6m;l&`7ja#--p!~$u{
zYK)0Yfy^Ptn%*O=Qyk<c&v%fZ(31Ukp~iE3-z2sr)AWVB=y4Tcjz6wU44dGP<LMm?
z&!WI$8Up0xs^WjKSgYXFUvI6u<%wflxYRR^saO_9WJzXuV1>u=%FBuMb%w4L+-`DW
zTc2{6Cp{;IJNor%oL0PZRmL})s#VK5=2~e~>_*Zig>93Ej9t5~uclizj`JtBW<i#t
z{239?tuYR^9aFgQ)D#-`Zv!1cCxIw*?@*#EG+50Kngb^B>hXk3*MB|6pFA5;ex^g{
z#H<S9HONi*UpOKf#~Z8@6>P7V(c(d+oYe(na^vdli+W{yj+chLpz#3wOl(*2-1#Eh
z^>Wo)WqXXA5ZCc*Y^k_6TuHtM$ZJsP(hTh7;B?}eIQPO&119q_X3!9(mt47g;ZTyC
zFQ93Q70a}baUF3OkJKxSk~H(Ru6^qKMdQJjnohd@Dh+5K(GVFL_h!51bpaQEA;*@K
zhq>jqcn80pm1|!srqiCwR;Ugdln>LQnQlX>L6A=@KWN$UlU3(0LH<1ckhZ7R)u4Nh
z6_!uBHl)KZ#YFU=Z|Vb|;igxm|8$F)QrFYDT7mr{IKjdu4C9I)?F_F6$TChrX8VSl
z9f7`OSYDQnl0&Fg-5(GoGL~tDygi?kFECOq1_=YdYl>g2e_PJLc7S(0u0Eb4^{q>;
zC{4-Om^?wF9=^k;>VtxVu6LBZ?`gT3hV*;xbhoosAgfLJ`%AO}lC-Y9D8BiQ9Z^tf
zGyw+d=6mz_H5s@P)$Zmf2)Kd5XUU#PN-F1Y7YWw40PN5gL~+Jopq#kGSQjMB91FD8
zZCfYFM~jv|($1r{4Y=t0Xk=UL!k^1U98vwj`~&jImC?u%7r6J^|8l6lCUSv=4}$?N
z1n$smq>-=&`9VZVazG<Tli807tZm%w$HE8BNa@59m#TeJ`43H8gwJVbAOtxdFYscX
z3gbWAajg*miXJt*2K5cI{IdGb&i2yMeUTgY&_|{5f%TWcVRlxdKlXHN6^W2Ci=C<V
zOYdW(<k*Aqfie(($enUl2lEDv9Xpx=Ds#^7P9jMzDAoY-BN0whHKkNhGDYGEYVP26
z=5sm*vJ8N)dqM=$qeh`0u%RpE7UhGsbF;A}p6MN?ALpfoEL!-d(HyLz(mV*+Ao)1y
z9c5=iH!1^tsej8`6<O%R=s7oL;Q`=trUR>j-o8gFo^R6q=;Z?n(7j;bXBdHO(r(PV
z9%RxDds!;?n-Hvx`71iV1@$Z_fFo7nmbQ+<^c|}Nmg^i(;K-Tc&+n$5f3ob-Wqw}K
z+~Ed>cgwIox7HneE{SWE<rX&nqt(i}K(j6y^9fP<pFvI|%N|T=KNT5Lmskrhl;|L8
z23hkAVoPF}S4)=$JKjQ06|<H>-tp}*N)fvQq;%zal#{C7+*3Sq>;5(zgVNMYG`BXD
z{r8X8FD9lI5{IbJU}Y!I>|R;_4*A<q2T_VtYG4XTw7cye|At)m-^CPHnM-W30x)5(
zxOpwOZJ!zOQ*xkr6NZ?>C?_L);bX%9OccXcid>I`51mGc5>4{$IlMI@ImzZ+B>%mM
zU8*Z#6y3gM6~|nvnf@mH)rst&gIF)9fzU>A2<wC+iT#o;hP1|=puw4vf1`UnKID>x
zvm6l~QbDx|c(Qk}CRxDw8n)B!vm(Hf5cEx@_K|Z|lvYPC;(DrPJ_~sK_Z<8e+APy-
z%ez#RepTkc^}fA|T1PfaV;*b0ROunv4%zu(DX<Dsin*6>^)4~$0f1G8JAj;y{ptic
zls@P-QR`|oIC!pf<rBUq-&UDq`7>WCV$mm(n@Ghs&Hs|t5FUUDR_e;}I4^ieRY>qN
z8cEqMrhB>E&dTn7X`<C3Z<M$xj501iVH`ObX_R}60CN^VaJ{3a!1iRtMb*7A>cd=#
zu^-?6qX`sDkjs@}9O$ump{*yfDSxU60q6KMR7ZVsCx!?#dkVRp5<oJ1L*&9cjL=yk
z`LTf|>?FR}VR)Y`uKclKRkWDJ>2cB9XV4trR>K-4{ys7(uN1dG1Pjh&-P(y6h+o1g
z9P%3&#6N4BiwC%P<L;Wb!<tG|q*8aMrr_Knqp-9F8$eOZ-<CpB-IR-Jo_Xb*vtP5X
z_`+!;Ng)8G!-a2h6d?9Zkqewui92+S@15aOSkMkft;#8dMob1oG+C-2;G}wEbW5r$
zgw-;QG1W|>j!yZYh{fWsKk9>kX$&nkF|LXZJ@~3;8y=E^)s35fN#=nvbr}oDd1^#D
zjvV&+1IE{xfm{t!Wb9<;$}PA3f1m2|M}A)5MqlRYTo@GrZm(kzIN-!;7cZNdzoW>z
zw?heYNL#}IEq>Rn?MgVlyjfKWpnW>Z-UrLz;@pCe%@QkKt+z-iasv3sSxg_4wc=ab
z3C8TfO~W!?umbs)V94RpN{@oZ6L&Lk+G6a3fhG4@3Pbv)3x3;K3(|fiHdl|64*2sh
z2rs9YRX&WwhC_MP+rt1R{Z4O?v*$JhoEJ4w^!X-<q}2@lrvDV5uisD&G%?-}Qwv!!
zWdEcE!crIg0sGfqG|*_7y3b{gP16YlYF}LP>r*Z`8|~<X#B$qPOi_pEW+d*v?{Vzt
zSd}R1{}*dnfriG7+9}u6i~RCT!|-E-=z6u2JQ5m&uIjDujZkSrs?dmwofWQ8_`fSW
zr{vt;v*&dIU^ZZMb&SfO8W*bTE{T18krkaZZ&(4#syTs>!SQx^Dreg4l+Q`JfJ+pg
znW;vCEG_WyERu{enLlfn<xp`pc1YkGts~XM8F;A{g~XqGc0scu*_I=AA?>^R5AtSk
z1Gd7Y@A(G3>EC$t#)wQ%qrMS+<n<FnHjfvga4s{xPVn)g>5)J_iY;JLMfla$AIDD<
z08twO(Q~5~^aQ1r5_)dquM-kMBiDd48{cOI%m8a{ex=A0GCy=d7J$H=;S`hoUpk-P
zuku}wtrNNcPthBSR49w7tN>g9`+ZjCdHptH%H8L`$ebQ_nxvaOPza{sYz!Ul0w#TU
z-1m&6GlW8ZQ69JFdPkE?FWI0}2<acgg$A+Xxu0E$|5@&?Dl?ncfeoD9z=M_$iK(!7
zWIU$ALj4@E1X~xP>2=z_Y3IfY$QP0Y*T$Ea+536&I%E#1_B)6&9chj5K}kF+)#Oo@
z36ShNuIIDNPvr4R%ReK?kGh@kkvdi~Rn}k$!EMJ(vN2>BHo#)|EE`pA(rkWoHQc<K
zWS2>lX*FDGFW|HEDzNngCD>Y%Zhdn+5<eG^&>#nyb?G%jqIKL7Z@GT&{6#VlC(tz(
z{@db(1$VYYP6rUIm|}j(i%Ag<<Gv~SfRvE}J{Qzk0c=J1BXP-Fi_lv!pzOGJ$S80r
zd=GXH@%GqMJape?_mkYe0of045{da6*S-}s2=4{ay_3WOV>U6NNhdY&H`tIl#jn$E
z7vl4Wr&w7mn1*@34(Z{h7^(Fav9|W;ufrL;^?1x>X9CLKXT32%*(7=Fdz>1I-s%_S
z2*}s%@nxWcW2h3A+8Gv$k{k4P0dFduHy1yXNa;~2+@=YA4F^yxi(@fFqDAWFi_@dr
zT0#!eji~_kl&3ho(?BvmwXP<tTg!{y8?6DAwa^Q!=K5?>(#s#fBU-9iiDOd|p6GYI
z#xI-b!0UBb`0dDecAwGOCy&DM7iSgwy~Pl5J3#6)6nq$n9y<iI!;CrHNe(ret5z}f
zZT-b{oX}TGn_tFEL_Wtn_-Oh*^Z}Evh=}ZzljtIf8Fxn$th0Lq=Q6657RpwsZ5r(c
z`Mow`u(DFby2x*-es@J9aMzsOuYf!ow)u5=o8z=>GpunJr4z>uV)>GzJjKi2*_Tk=
z=^I)uN&UI$j6N##71lb!GESOXI6F$4;{NI4b-{n_w4+$E5{<3(t1Wyy0*xZ~bG;5r
zW7{z$o&UB7vZ;0VU7wKx9Pj+HjPtKe$&N_s@%@;DeyJ&bV)`PmhQ)NgJ*F1aE&i-f
z*U-`#s?JPs^WyqDNjuV*s%1FO#%5b=>#enD(Z1z%)-<V~!1$sgI#>mZvSupJ2G2=h
zd-P4#%9uq66n-FkV%6O90(0am61eE3?-eUxl8ase$Ss_;IyB!6U>dM~$o3xslgbdD
z5Kry4pDr7m+v6|5YQv<C)2=(N5wtN|Q7Ir9g1~Ba0cSw6ISm36hJJ$eG377-Y4G{v
zQzz4?`f2y2(mAiwODZa;E=q9M1x}Cj!ntG!xCkwSwTKI4a!<dA$1n#~RZwc@8tKTc
zzUcrch+ZK!2R5L42ugo5n&r76UlX)$bVR*D<+C#7HTp=N9EJAq!!ZE~9DbmQXRH*A
zL@5@?OQflU3oM9ML1#p@^x|-C`mr!gP%6?(kd!AQ`@=yJX;M0`qMIFY`sW>&1^XU7
zW=LrLp`mYoo$7?ckzA2;pfr>S>KK1^?z+<j9|?%c^<2x|{lxuF`FY>ZSE=uqHW-8T
zF%C}yDb4Hrse0|ssNTX+pH5InT*G~HZ*wu?&`Vr{VuH|~EEjhzhYOjfM3KBq(zhC`
z|DQ5iCLhWf0uKM$=_LFHa!xB3Bv7Gv(~sSS^rGFX!0$EhawZCMk9XBD2n?4lQmF1Y
zYNvLxL2|=&SxX#W5#s--3hFDwV?bF?Onh)O1Ds&>&EtFyT6s9HlE=al!(?g>DdSYU
zza>COr>2khX?{E`QXEi8J$3vpAESV?TftrE2*BN1;EZf(==8hgnKd^YicaKNWq^m&
zXS;=*ewp)PkAEOpkL5`hIIBnP%EvJct4~X7JxWN+f%rgB9qCNaNtLr~Bg|XUVp8*U
z=TA`Nz`ELVSJ8%%Z-OY$%GL%Yv|biG^EAgW5i1RczE2Z0k3Rfnrt3tcPZ(mTn8fjF
z<xgFMPifO?v}<G--cy3s@Lwx@G<aW~ajHUY)Rxh&(3nlSI;JHjewDPxwzLeTsn1py
zmF-B|PA9%|P41xW8$~+bf--u6qiY4cZY(n^saHAHia82LHoxV&6@Q|!qo_9N_dC^k
z;UJG!tNUqow&%P5q7tl(g|3-hTslt&OUzBWDJos`v-;bW-~`njXs^V3LAUGVC^<t4
z*jNQf|NFiax0jnI+?zfRgG@9z+=eFX!U<nT9t)k`?mWJc84a9~(26lN(G~SLZf8a=
zQO^M6;t<y}Ym~>M1?M4CUjjM$W=2fxbRICn+d*k9a6^w5mkL>skCtqiN-d6x=-o2_
z#-9?$i?Ys#I=jh88;{v=9J05ySGQ`WrbnjSQGyy$Sd(tP`Z60;1C(ztJ8QoDs*}Mw
z=4tG!XD3=rUFiAm+SR38u&A{zeC=&v2Hp~4`c~kO!Slqi&!_nKy?OWu?FqG_!Ka~z
zgao?-YZGb|E08dl9Crp^<T{=hfg)s>YU|(B;$j2WP_JN)p4|s6nGS##dKkgK&sJA#
zqT7wr0ipcLm8egCdt^4NhZrX+J%G1k&FeNX_FB72k1ywM4L$-wsM-cJkp=Cz{y6{=
zN0d!*<y!<r9k)VR<*ckKdcrOm`uTcZ#(qYytHtQIjqe3i266F!-`QqXCXu2_^OpXE
zl*=(ss8r9dsrT9Jm4nF>1vu`g&A+7+xq3RtUyOSueVux}zP)gpS3hj|j7h@QC&_&q
zTV}9KMT-j9b3<|J!^8}bJ7T@$<U@0eOrPJMq6-L&X#mJ6K7C#Fc_odU6=YnIcE#Gn
zRxLhvxBiqTLoL_}a&JhA<OE$%K6r9quGPZW#>Fg`b``eY=pP4x$ku9C$u(~O#|v95
zhqQ`Y(%MoIN7y<RtFH8tiVwOqp#ARqNRU~&gplm1U>A1O4PQ$rz$lA4qy&5>W{jPS
zT*1%VikiZ(LhY4r_CS@!c1=uU0(|+$2qZ)cGb8+)Gj|2lnYV;cQ1>TWPEJeQ2vm@l
zFtM>?nk~jkx3eyW*&+WDnv%uf<GI98=3IgBP)NOm^RoDe_Yy=v?<)!_i^99w-l{8Y
zubxl8_{RnE+QGEa*UKIWoZ~hx!Ct*L`z}*@S4y<ofXK??3gv@&8|-y`fe>8h95g_P
zo;aUYaie(%6SW=(+U@0HZAEhw*RyjX%-;>OPJ~?kJ}>X$+6bYfk<3M|)_?73K8kvH
zXyJO%C%c$neKpb@r34es8heJTFwhXMaPs}w)%V&@(x*SV{C;nB$~DA>fT1Z}M!WSq
zd)Io5Qqg+Yyd;b&5~7(CtSExv3cw~Vuy0X)&s!@w1l;60?&vxjMM9AFS!2wDBIYbP
zdMnyh-6l4KT*+P?Y4wO>Lhf}mPuFm!k^U$*+mu*9aMc5roYuh3!~KcJM^j1w$rqKG
z?WYPSlKnQIUnbXQk0L#k@HU?Yl^iXY1z_Jby741sZYzx(Q#R2rua9FEq+@8eq=1P0
z`FN5A2&|$%vyOy;=+>&`r3}EuGQ0OSwz}}c(uVktpV~j2VM3|ph@w|_ANVGo%@N5)
zSj9UD`oBI2g!e8>my|e*iWkW$WLp3p(uSjJOmeH_x4t`4HwC2g5NX}IO8!d2nL1tX
zm^XL=``e-VSctGa%?RRFlg5ZDHLD_x3JREBQB7mpa4Suz3YC1()+!1fPM&@qGi{qi
zt~~fwRc*{VOI2b2EsOT|bMUH=hw}uI*>xOv)HTtueEpa=u@jiTBRgW)y(7KzFn;_n
zE`i@b<l)o=bsARoa-%Cfsy#0q7SVICiul%G@uKJ2pgz{6c(MC!{?J;&+})S<<$T&^
zI51Kne05;WWmZsty#@+=VALtfw?p*@M`>{lmhWQN!KZ2?DabC8E=Pj-e1~66%XnTN
zpiWO5A{o}@J$fdLi92d&gUU@xu+h!-&c;b3mMKy5EO#Xa@nA0n9^}xUrEo$-vZd(?
z;6lOUK5%qtJxLW%+PxgqTG83AA~4q?XjIJb8_z91AuJpChQPxKmQ^(g_ZhD-^IEDW
z!<vqiUK#)f8RYCAwIsz$29Nx@i8L@w7?y&zYHi!-t%vC@oK(}&-;Qv$^h3T;*;znR
zn?opYhmwActT(%v+MPrdf-eDs>p+S0dsQdk!zB$_J_nr%O^)p%5y6w~m%67{a;T(<
zA?SlFlLKoB@*v4!T`~0q*m*yMb+;G#u@&h#7K<c^vW-z&@Q1GU(?*7lKB*KyQ^wZ>
z8!z}#t$zN#bSuHbMB5d|*Z%m&m0wRj33jFADL#)HV>7y^$h0P!6%lrZjmA!X_8R|}
z`}{UDyz9*D_P2U&%MbTtl&7z;V!*958j%~CYDGu|%-mbj2ySjZeBIk3&qi@nr5;)o
zZI}wBwvD6QXL4#3uxAaEf0|mbdeaf3KoJa_9n6HrI8+Se#fU?U%)VKaCH8nU^VrFu
zF~4vA`90Gi@zln4k%f@@!RFnJn<CB+A7!cGmjYhFRWP;e70o2c49wtJ7Z@%~F-Pnc
zd2pX-qIvkHYqsrYIca{RPNOszC<AQ{vjH2-g#@T94C9gj3ua`s|JI{Adx`qpuBeSK
z-E<?qyDUb|b&Y&eGq^hU$l0$t=w->bsHtvPDJIC?YcR1pQMu!QUQEo7N2a0QQ9TG0
zd_vJrY3Bv&m{i+qga33agOOQizCG1J2Jth+VPykR$FUSQRHFrEu@*Ry+%Q;wbJEx;
zY{(5kjrj)|k-q&&R76UzrZ|5rHt#iC7Qd?Lk^Z?q?+Tvj$tM;U{0}shr_gGPZ-m{3
z&)Kh1pcTibQng>cU;mt79O`e6DZ~$I5-AwPjaR>;^%9}it9$WB(K(waEa9<reD=~p
zvh}?nGMXkzR+xMS$zIQUeoc7yXXm9Xmr7oHGKwAMqzcc>0thm~zBJqS+L-3h-l###
zBw(oge*fXU`(X$gp{nvmv0Y}r{g*p5lb;M?6@8%F$e!IqthP=;6SBwIG8@Y`t;i}{
zCpsh~Te*|CI0UBAdd+G*(r4XBro>GRHCDS^Cb;2#6bemwGRTFF4H0`z64K9RrRsX-
zZ%X28MB^=i6ZRu-r)+~DIyH+w2{f51nQA9R`BAdwK<`1B^C8_aX|Ovg$;**!xBDii
z;flu%5yWNlmeBJ%b9AXHUgjWVg#kxNe-}_F2~7T>p|LHK;cR>epxJ{35l9Lq#hPd-
zyFuY2v)RBN!Vg4QtrKQo6|=Ck-{mw};UT6LvaOO~V1(}?U=ermy1}6z`BrA$sPFB<
zPxEVcs2|3D<p`Mmt0$JaT^wP7n|?q76_ajMgWKdNIcDTMXW?qI&rK95NQn=Xn(h!u
zu;ilxOlR51IVJX+;o~mE)=FZht<p4?;Lzbc?1{O`50Pww)1yJnq~w)x^|*^ENOJrq
z!<#u2j2Z$D?u_vQfiRSZdTZ~VCW3Q^)O)8omQFRy8-Zp;ivhu%Fj~5V&bS23;4sz8
z-!w`|eFfBE;40xCq<<cl#f@sO&MQ|hEzoEle{V{zGHyrsIc5q_H`Z%4Km$H(eBrEL
zihV(IsVn*x*u+B<o8Io!B9!uwt7EnaT$6JPox+1hSmb!nVmkGU#9;<+32_5G->Juo
zBK9<c{j}@efK&c@1taMs@7_6)y-1T3Ll#<77dxc<7&fG~{Qs&fQLF!7m2GdbA2J*_
zZsQnn&q)|e$u^APR9567Q97@L*PhkY$CbRKub1lDecD7hM}nTz{kqEg%#3WLS}anU
zpebM5^7}wf(|iiUe<Q^@8LK~jUsL^NVwvchtQ(DOrwz7!dBm1qTFv8J2~zG}drqZT
zqG8p0p_SnG_Lp~gQ_97ZflT3Q@7>?X-}k{e?U8-Pn7dTuR@v&t>CBklxaKA;lJ4hd
zv&_Xa;!!?yt6|1`5!b7h?+Gcybi!|s)1qBP6$}JhVd-i2xnMSaX&1#zrOGVjwkQ^=
zu<}v_YEQ&jJ{S;01yMAxc?}A;Sv+na3(P9b`Z?K^N5DPMWwu&Di0o{E#pV+oaf5S%
znIEQ!oggb0XW?HO>t%y0Xq<Nw&q5h(ZGf!CO8)A$_{*R-EjoPGkBSTVXA*mXNRnhW
zqRT?^vlhV<e@$;IM6}6lB^*UEP-#PQbj>A4P)oz%GT?0(Snt{jqbDdbsY+2r^67B(
zshaxjq04*%Xa1@5-Ay~VxtZ+&c@ivXm8Ay#l=x2Dtj*}F;QyUdW3wGe{*wAplAi~S
z>FIy4=lb`i2^G@hDe5T^%IlB8@3$$|4a7gGY3_ctU(Ig+ubLb1$T^07=@#7*da(Yc
zxaFe9O!sFoQD{AyHnxwF7C{?n{JO;n6lFuRSN7TGlA0^Ifx=HmEWYz$t~N+ynE9EB
zr<N1{iA7;p5ZUwFue)L=afLgeZKDc2*PMs4^C$-|%wS_X$3*WUV~}eXL0C__z-iX>
zR2<WFVt1Q6vIvo){WN3=GN9n#GxEpufs{kAE4)x8=+M%yhC)8eLxO4~`Qty4ZqpfE
zSeg$N=z>*t;^&5T3w}wYawD}}E{KoUM(adjWaNObRB~*0ZlAH{uaC?BOc9$ZK#`&N
z_%jw9Qs~x_x&5|3_)~Cg*?DiQSUV)UZF-i(UBXyTGp$2F<BnmYfEk8ORohQHhL6+_
zaq0I^c0p%kO72_A{pR!W!n44JlBEFC^@LXXz>BH(=Khmw)GxW4F;wcOJikS2rOrDI
z6vo}fYVdoFAk{$&IdLW`wxZg#Vgff@M6E^EHb41n)=iH6^v#$*ppFcZ;tj7Qgu3df
z`i^_#>C{oPhE0F|k(3XVDs%*G89YMna(R3av}i%8ubXFpzv{<`!fc?;+0vT+uWoHl
z%o8^akx3U~i8^TxhzIKCyVKUaGQ7_Y(Ah`nRpmgD7->0LTX-CNql7Q;{kzv6-elb&
zm1ntd-l1j_qg8#APUZLgh~J<zn_Q5SvBQTG6$7W#L(#=4wHX=M8qo_BOrAYvfik(S
zcOp0a=8;cCk<9(gr;7AdOL@-70#!K5|JmG%eE)d$HwlZTOKpu5@$3KjLgwjK*q0hE
zBzb#hYAMCz4Y}=q`6q)apBTa=OMRwCy=V`O!?nKp8n1hOi+N*Z39sw^s1rOC$h$*u
z_`pB31}4MVb70K>_4-XGqM{O$nn)Yu_-#(_{k8z+ZRHVC;YRJt)sL)LoD6A;?^|Vr
zCM3s}Z-vU7=(qT`Gh!;MEQiy*guW>p7l=drYmuLs+V-fL&DC<`RIoW)QaewLz%oNM
z@XZ|6d<#$hYwhQgY8II=N@za>cDF4RISUl3E3(u1r6;u*>G<~9j@y3F1GNYksWff2
z?hP6DN+jWk486EUK2@nR0?2prxM7iQX$neIuJJ`l*}{_bNjr)_=!T|LJ$2Je1~j*p
zkk&N$8mdz(dDo{Iut7FsHrHEp$K7M6AUndvYU)1B>Lu@8!`M<cBG*%WIQ6-{9muu&
zQvzRkzANOodYmzZA)i72S}5^De?s$_RgSv|S^GLS5y88Ns(7)$O{Cte)O@ZH5s$Y;
z;u}Hx`P5U#*DYDriKgL$(W(l~8C4+4U^g)PJ(JZpR=c7w=GE@AwAu@6DK~|5n`F73
zWBwjt-VwE#K3*#b#uZIIOHk8s6Bt_cTr$2BAB|E^{(1R~VS^;U)dw#t)L9F4>VbKp
zKvch{3aIg9dOg}KznF8-#d%A9pIOh29emB@xu+)-E`#)}qWL9qlT84RWJJMHMb4@A
zsv#OgFNvQF;&A%oA3Bbrr;D8<{&O|t5Q0Y%+{7syQPh4^qMlbt3Jr-!)c2Mk^kk~w
z+9_m_Lf%%B>k_m<R?^`+AMC;incV&&1)hr~>K4lApG1X11$vnYJwNO}z~IWMUD#dN
zw>DAQYcj{an(uB)S4TgR2VhSRV7$PP%+-wt4iXckl7`XV_IW&500s8czsOsp9e?Xf
zjz^&bZUlYDVY|TRC8eJz;8(~p%m6v|K~NzS^m_rRQm;Dw*Fv$rN+Sg<;*RBFWVn8k
zNdvVx32jz--9PypH%tQa(;(48gM~I=RsG!l<1>$(Uq3?ky<)f3HhWt_v=bN9*#+E9
z%}3z}Ekr>lVdZ2o>$th{8(NR!`_m$Dk#3s@8@Z_J4vC3icCTzn^0%XuEHVKTqmdKw
zei7r;%96nQzzwqeU8=RM1+{Z%oOqL#S(3`=>>xW2&FJ^FyZNM{OiET<TunqHLJaOv
zH?r!xutubO)2GLkiUlbJ8dRm@kgu9Oon_3l^uA=(Rr>gJbw)|I<&#_Focyrn&*;<y
z#xEjO)^$_7-I1h4Eb>nqc>^696=+w*V^2$^w>e~j((}!*ciN=Q%K&Vcxg)E35^=!P
zUNWm5r1F}BD&cLKCclbS{1nR|t4+V}8HB1Lpa*~S%EJ+wvb6ExQGH2$X}~*tuHkd=
zvI>3Q$LPB+F3ktgXlK|<nas}w){MQK(ks=b1>Ub{etBM9yO@XtcVv^WBaUsz_={My
zM1(C5>h>!TD&0aJv_#SiMjP%(C(MoKW(_*2l~<0Ndur7X>Knh5+bUk|i2rhzN;>zf
zX5Y8sPYfoa{+QF`NVclBIOR5N6c}|;A1HZ9@JKgpB7$BdYt=xyoxD?loP*c`DfMps
zk-Rp;$w;j~@c6UwUJ*fHWrDs5O;8ihwWfp7vN9MVDiVx_lmi;TeeRNO-^psb%t^qp
z(ej$6E(((_E92+3fzeR#w2RWc+Wr^!OvleLQ=APbHuTK6TNhsmB{J@^T|&Tj1-ibq
z<z`MYQ5Ixjj`&~owhP>W0W<;qvo5<zH(%dZRs=V_hBbuIx{E!ETu3-lMQ{46T+0L1
zFZO<cwiiA~ebM)_y!>Wd6RrP^aD63+kSb?RslrRYXJ_OW6+0^!j2zN4_!2xddiXUP
zijBv@`<_%uKAiAZt-a2LFmmm{@@!O0PY=IiSuu5owg+S>jw5GJWOpq6=WdY*XSl=c
z)`U>Z=aE=(SUYtGdMbEr7dzcLCJTm|qCZkrA|RW)MCa9HQCj+^j!-(M7or~;kD}Jg
zZsjGwNR#sv<1h>CGvBuEouz`KqVE1T*ty2JVM;Z>_(_MbW&e&3$zQTwPtrhV1|}iH
z$EA6GALI>HR5WCDd>DfX1aGp7)e~zfs??XK_^e<_Gkx_Hl^jBa+zo=f%1_0N*JBoH
zfAhUdRtN4Hz&R}(`8Hh&H7dI)`upGAZvrxQdlG$aO&eas|A0k{lABn8ADIS!{NvmT
zlG;pbwi#Old?!R{+TU_B34Ha@>1Ikf2x3%IcD^+EZy8B8s}dc)ezou-!+>pO8k30)
zIlD+Hlp1^AgpF!H5h5<omO&TQc3>nVaB8=FZ1sFg2y!@8Iz=z$l>*NSe*s1}3|a9l
z>p5=;=gpiexQ#6Dk7{!>+&E_0Grza;>~#$O?8dvzB!ytx8nu(&q*8n8LEe4Ns|$zR
zwr%cS?`|iRHvG=-<xl+D4dvOADr=WMbx4r8VxRxoy?DGPW@IQt1(l{WzGVBG#{zBu
z-A22)(tj>^y+Kor5I<`y&bs$CF03jY-<(<od8y~P$Cx?o^?C$I#B*M!uI3(|vtsjm
zi+cWN!9h+}qiCpFC?u9c!arCD|DZn=q3Et={s_G$PJ7NcE_m$soRBG896(;=&NZ;z
z^5VcVLcE_@(MB^GlbO-^p6s6w3$a~Cq&PeA2I(>*ArlMOoz^mnXJEmb5jWW+Xeiz=
z`g=h1%ZLl!;~#S=3dS1k=f>m>3d`|~5iuYw3%a5=xzl7)wv3$EFLJtQHyF&LUtZQB
ze~jxJl87I~1ZD~c2I1@%5#l=Bimd9<v8y2<EfR87w(HFT{b;&JjhkB*+-XMKAB^6j
zFUZ$sjGXf~{@V+n<JMGmLyQ*cN>}tr)a*#YxvmeBk#l0^(!Lmov?%BqXxQmX=TnCZ
za;hlM(&B1akQb4dM(0KmQ=l?>lj=Mn<%m;+TEod_T-4gn3FoBa;3>pXhK!tp$!1e0
zOxq@J@x$lKReeb?_K7Jt%LA19dx@oj7zv+>`Io+*A$1u7X*~_tH2QS-Zt3o^nuJje
zAUmAW{afmx`7FGtx^OkV+!6kGcocD*Vfz_5c%!Pirhy`Dk#SrVzy8IHZ+=R<xT_53
z6PeI>W418i&5O<}q-I}E_T+IOP3#4&E3VdX`cu)V23!+?*_~w1|LZb$rxoH4iIm~!
zY}I~SfoHkXvq|BlZd}jz2J=hne7VNNP+v7MR;I!-f>Dvx`7HkonAXAQY(mU#Rt!B@
zgb_Vtvk-P72SueU2yO~7#aQ|81l)0Asdh&1=obf_QONACAB>zV3I}r60njOa!2zC}
z7jH$s9!pcdj%ej%(s~u|;`y1U8T+JoJbo~2FLOT#uJJacxIN@9@h*Oc-EIqqH`xD1
zDd5J`{zVZgcx7j`_*Q;NL3~P_t%Q}3J;J^z7qLXEN*h4r=It7-yjWiKcB$^cSXTR3
z$*O)_Of%=b(Wbd&kaZuOhi?I0Y1gQik?<A%R7i7*>7p`1Xb}r8N9(9jW$&`dU>c44
zW4v){o?#o!2^;_YQn?Fz{)1A9?HL#0Tkqc-`D<c__^U$4+TG<=<-zYu*>z?&rzT69
zj|$@BQm_@?PM5UjNfH@;^1%TGt2xm2;<)KKaMDOTh>py$?Ab$)*j{h`)OQ+2%i3pA
zL3G%!C7$u!>-?fT&mEL+o&CMNbCz_K_8>PHr_5VkQ-rUjsr`%bTDv*vSEAPD2a(&k
zn0Tikvy`7%EXCFCA|$DQ{Uut+3<qPwd>E;ly%v+~1L!so)@p>+crMEgxsP9p>BfT8
z-ItMR5vcAYff9>+)1PycJ96pVAYqy1ph4fakv@0kw^WE3d(~8gv^D1)7^WOR)MR$C
zQ`Wdy;1$hP#oBY$aVuU9avFJ@S)|llBw_|{!0$p`Q1taDN@z|&Ueu3O4bYy~Qe!Rh
zUCQ@rz6HlLaUUcT=shRp83~tJs1jv;jAfW6OG`lz2WZV6`c*Jn_%r?6mw^}Js@c=3
zS&kdK`PRPS7sg1(g>t@`t(7dlNa2F)=tx{OO!#SK<TC=GcAjF`54pcshF`}S`Is}#
zyRJMg*b=}`H?BA9QEYA4Su`u??yj`0xKrlL1M*)Nf0cuarfe!+oT0%K48h4Z;)pmC
z&h{@40hg+oTNc8}*pFLn2&r4KKuUy+Cko7QDui{FdQUZ5linH&1V0P#{^!wA0DASP
z**y@7bTf~Q#N}8t9uedhL|+xS%WR!<Og0*MpRG7P`lQa(2t6(|{-G984tg03kjQ@6
zrYSUmf?|Sjva(loe>De)I%gHo6%(oUuu^l(4t}G)M!0nMiL>Y5l}_=O`T4IMBlA|(
ztf5(D_d~-(o{_tB_8nzH<=1;I?sUG-Yfipwu^vHNs;_1?x4y}N(<O0L=*Z%zr(52R
zqdzE5WL-Yn_Rk@sv|*$UDGp*@-z2g@?~;@h-r_mSIc^_Et|TW!bgXyuh=_y9wwVlh
z+fnDtr+63g=NE-jQHg|}U+Q9&b=AZad+akEUdeLpQ#8cqiV9MPVjzkoRc?!qHT_>K
zg=v8IQKqaW$BQ(TGqiZfZd8`12uwb%M{p3S@!8<z${pbe9+9xcJeT!yv+pN*$Q1@Z
z=g@cBjbuyG<(P5%<PIm=`f1;k>N%NuevlYbq~(wtmZSr7>Grx)+>&s9e%FMgnhm_q
z`OizZ@!jY9M_Mb3S;N=MEbg`o!HDJ73gcvna+k8&G8g;&jr5=8;8<tpM4Bj@E0<(X
z27N8rDV>6h=*Vd`OlVg@UV*eb9UYV8MUZ<Sa$i*2vau#x<XDvQtTenzAWKNjQ3{MX
zqNkHvm}&_VJHFU=Ielcb)6VCQ6okD_2I&UB7uw{N-}9GOhG532sYIy{x0-wJs%~(0
zmp?)^3d}|`eT1K01=dN^kDWu~V_M3gBd23SZd-W|M~)!mqT;j>$@X813Q!JU!S0+d
zxEAx<$Z%b+JM1K8LwX#c`jQ}AHZ|7m{a@l1RUA$K#hK(#f|pX{IBt=ub_`9lSeUa~
zl&N%)Zk}J@gkkuQSUB%^-7EOJa9&K~y{b)+TW=x}bFcVjGqkDvE|i!y<4E(RB)SUu
zGvKoVX*r-u2&9oZ#R7opl8egVe3(Hlx&84{%8nnnt&t05Mzc!=hDW#&<%Tb~r_?FN
zC=4Z9dE@&*Y2O9iZqKk{Mgl{W=ErNkfQ;>k>7YZcpA-v(KAmO6)2jnL71}<_%Zmam
z3qm!NMW2raX}T+_MKJ*|5DZagR3#XgMK)0z(ekrZ+}x{`c~*)cpe(ZM+jM@G607-v
z9~7J^g?hIOYMUWu!1tCn-wVjx7-iUR-;k0%55|`q(3R1doICEM&37BuO14FtN8EcR
zQkk>JzNW<`=@#hrAjk29l?z1+VfJ~pQ^{567&Iq=2HcXxPl9MJDV|OUzKhoAr32sm
zD!fsoifl;L;6r=adL~9~ht5`lvdT`l+0{TIj7y4d5@YAREr{YmjyLPJp!HPF7L8VO
zwDdx&=M}BusPptgfx1plIvp+mGqv%>FiFPjG5hey9&92t0l#_r=cKb1UoWQ3rh@m}
zS>jA=sP2Mzh9BAs@>Cw1AqF0JdP8!|RfhX812-C8=c~trhM4AD6p=-r)B~@07knl0
zSW!$7_7oTf`^Wd1@2z5Ja0kTqJBIMScOVTXQru+SNiWg2v#ug;1V6@-;zv;J*kl!q
zxr?YtE|%S+2ssDNeU+X}k!Ln#Ai!r5>J=Mh6e_x?YH#IK<b<XYKys46?4_wjaEwF7
z$mvurL`<`B_Uf-<ujlVSR2qNlE3L_8zyGWrX{>9<(Ce|8e&`NFem^$gK^c{9MOj`c
zgd(N_JYMt~2`yVAX82uo(sX@AiPERN8JUxK;i3%`Uc8U&kn^Ozk=P4&@szu%ugl;u
zr&E?k{<E;qH_vtOJ@6CL*&I%!SJA@?@s({<Bt9-J{r<^n_v>;(z!a9Uuvq5cesddL
zh#q;<#FMJ>98JVa{lTgZZz$whWaYMZM>*IDiLi%@Vr5Cym>SctVOF}E`54^HdI72N
z-J6Vn1&L9>3T8A$JPY-RB($@Sxll)QcKIc-(X@N~u)H6dcWC?!givv7r|H;Jj^NEd
zY=Qdt(&nMY*GS)=mebxV_EV>GL&)n%-!!^*f!58n*m|<Ci-o=@S#m*#h0_C;Gr^$O
zOpSJ7;6=f^^<An-Df688P9^`agM{Dso8yxI)x>)ZQgOm|$@kFW5WAK9p(I(!&8=oe
zQm|FgMJ@P%@%~Hrds;DYgZlStg>2;&CI86yD9MZ%1aQQ?8o_U?oDurumk8}v{41_{
z#72L?k@Q2HEqIE@N-peo!qa<W!?K@iwQ+)>gamCN$L|wqzHY7sUNBvZou;^?9w)c6
z)kvp3#~?ZbuOiztv#gh2MEhwht*8ou%M)!UKRBqu#aw-{{O{gSA=ZIiNPkNxJsPvG
zr}f2)vRar5LhnKqCLacYE}n#Wswaa|(kn$h`;kpIKocUv=wLh4Vw@)~PhEMZdKPmm
z(!we<Rhp})$I7!A07@-y-WJ;U6a?y`JbY&&Gc4rgY%I}u@NAoAA^aE;L~=)b2W?7n
zW6M%bksh=#jPGaJlk<r-R=Y>GS$Yx)$Q70eiPh)Sg?3qA3BubPX?lIrL89qTRpAHG
zjxezhUWzU;CyTtMl>W_VHxII`e7kJeW6B?_b?YdIBKYgUP50D&>b-z#>0gy4BX4zQ
z=r~`3O2#pf<b;!Ywq*N<kDG8A-C*0z7R=yA`33%ea4!-hBETsiRygI4eZgO)W6zLi
zhe5XkbSfA%525d)w7_PhkWY44NZSlCk354OylE&<XP5duk=u0UW&4wa5Dg_fWm8Vt
z)yCr6IFV!o`%hxrEILiQxD%4^T9HC9Gdp17rd+G8yx?jj{W*Rr!YfF~ikC)KP;Q9f
zhhvAD!S|Zw+ws_M?$qo)TgWV-+oa>9(k&dvVLUidFp$(z8YW|WB<JQhqGfB*=a%?_
zBh?N!J%7_^?eTd{JlU}p?;fp)o2Pq)O;emWvFIu)rujgWM>j>c!1Qa=ldhAf%AlRS
z4fMa5fkvBuq})oXv?&Fo!)W#_F#6?j<B)g`#9~=|ZK(KCHy#UD3crHgihs>$>wY80
zAv~-y2)ccoBiY>kg<_R;^<O9KyB@et9XqJl?m1dw)<`{V$Yuc};1t{Kj<gCU%Epac
zrOalP^zR+`(+f;{Z;GgQ5a)R7K5f!l7qj0yyl#~ih7Uc1GMHCc;ey>{2->s!PR0aw
zXju2m0g<mfUCuV@Z#(mdvr)IqO&SIyEOaAJi=7YNFn$)5#jd|u46SYl3sIMLSEPSN
z_{6V5VQ3P(o3t_D;C|Xex!f~~tO%Dc!f)(0o^6M%rsBcLg~par<=u@dY1?Fvu63^o
z{<u~QWu<adSj3%o{A+MJ`oH1V=eOcEQavMX%ic`~L*hHzqLMwUwfT`n`NLd5jSDX5
ziQ=C>*~R0VZN{soCaFt<w5FHyksznx$C+hH&1oWscfXQaynNqX2Yp+fe6D}LGuQ4h
z-Tw7m^G)vlcW%5{8J(_3+M!}{%)P8|nDwW2HJQ0<+fA`U+(^<4Q@{6uC&QR2+jU71
zWmaEWzk1QT<E~QqMn7R>#Y!L9Z6)^Cz^`uoE)5Qa-X3QJr-*I|Z0&!IZ}_nVj&~br
z6UJV?SV4H4C`BpdVnu!9!gQP4+37!SZ|_+&{S`Qr>JL}VrBNB@z89()H5I+F5;;1s
zf93kPSouIZ)V$s321h@&ImvW~grZ-rml3{@_vYz~eor3ur`>hYUS%jPZm|YAU?z-)
z2NXKmwM9zC-$f4Nhn+>slDq2|>L)#NgKqOCBSbh#JEMZPU5NV?JT?Cr2p%FfoKj=j
zuiA4W1AZ$GCD}+X5-yS;{19ncUUW{wc3w8sF##dckeW!p{4*tRd_Xas<3=J^CM53?
z4fZ>g&0zMDasR=wKf&!BiriRYoLg(+DmLG;RVE)*#CP+qRtn;+1{>O~HrKEW1=AXe
zApdv&4o5#5g167LwuHLLy6YunOco~E&)v318_c$ZF8QUC{G*Y>d)X0h&%~`b_ejwH
zVehTuqFlE&U<D+kM3j&i6$PZF8x#Ri5KsxFr6mWXV?<KALrO(K8focn7(yBxy1R4u
z?nn1l_c>?3=bZPS@ArMb_n+-{oO$M+=egIq*SglVt|c29sQ$p(<#lDg*qmj{bPn~m
zx0~>(>0HDI&sDEI1r#aqLYeD^T^~*O@tmDz#H~XSK4VusL~2pGOOmUZs)pdk-kA-$
zjcUfM43lP*<Nh(+=7qi(eTY*Fx~<{MUtG)i=;s)1vbu{`l;w17gs`OIzn7RGU>Yj;
z{%}9-xH&^tm{qunI~F`CzP2$m`hL#ft}ZN0iL+{P_8lq<z)z2@F#&u=LT_3}BVFBC
zhuF80)Sq3k5j|cWZZ3rZx}F>rvv6wWzMO?#9MLtm&jc(Dcs+JAk>Z-E;So`)*NYZd
zNR_Ul&D~#4rxE%d{D8(U9X-MRV+H4ftIlDV8rK$DQyiC1T+F7}$S=D7V{yN&^E@Ld
zKTuGq0iEl?g-Ag_o05J?`C$nu8=vRc;xx&dn^9cNbW$>o)fwWv@F?+YBaNN9eVZ;V
z*U_28=prmK-)h*)L4(kU$u}97HKVR%a7u_&Ev%Jz-aj>|pnSHmcvAv)pfL14q40HY
zdRhmk;`NDDUR>3zp|_u4-&gUrI#K`EjU%{X3A!s&%%0l~>PLOP=q2%Lt?<t~`g76R
z)#tTK`=BbE@%D!RzHj!L$(s#S7P7<1nLx#9lhbvR^`S$nN#RSgl871S`?b%4h?&(u
zZ~JFswf6L^5++C=Qf5`swc>c$f*6u|Bp15cQz?#nbTlhQyy`qF)Wei_4bNSSpxgqe
zXs&?@mk6bEp5&JYKcBBG>1jfA@{J<@^gkD8ef!jzbHhZv?Hh4nxRjs%eEHX!eQ_c2
z2tpw2W;k@!dj4Zn!ux0{B|H{&3`p6qMp2sP;tH5&9)6eLICp&}_S;r+4p4hz8>x%U
zm6ROIDqT5?19P?<^Qj%P=v7jYV?Gqw{eqNhydI@)rFiq^=RdzlRt#F17_pq&KP}<c
zUH)UIULvjP42odA>6%s|Q&^vQFl~X?#&f!bUtff(=zOAnlm@REyV%MO(mI~LHtOfb
z{>u*g`RbqF|5qdU?bS!(;_j6<nyw`={>Y4eS;}v3`kVjl97A&c^Tz$RSBS>IxO60U
zi=I;Qd(r6MzW*<8z(z#=_X`cKx^S*@%;GF(GvmE~y4^p`^6zdM_g)^s-`>B!xiw(T
z$uD1GBOV3INGbl@pa1by{%U>y@y9sMU~%uozLMqi{rhG8?aw7dvta)BA9K5L`E51o
zf?lX%_}|^Xzx_r$qcuayzdYT4ZyeCYWKToJbUDob{eSlU-|wIHRrg9HP|+O+5{@DN
zo6jqfr~<XU+CZTd{&eGnrZI~!IbJ-VTsxd0HBj~UKlS%D{ACMT%V|d_DdR<Hx$b93
z>4Mb(=@y}*oubP4UR<i^|HTi?!egmC+8Ccp3QFiTlGR(=1>@_Oj(`Sj%>Vv^1z#WP
z1n~4lf4?AK`b{2=|K4<~uF`{@a<#)bjP1X+;2sI!rRJNNb(#L_kNQOeGxX!%lq>nK
z@1-m;w9go_{Zs$zkFq=iK8|&wjc@+HHYpAf$cPilIFtYDkLLIQlPdDzLVfsyHTs|L
z$Eq#JLppA>ME}<w^8bqu)=ilZ?Rs*E@&;;3cG?RdKbX(yzQ;`HoV+sQvO7w?QgOIa
zq84vL?AYmEbCx-yb6=!~0SXfK7PU@>_@X0JGzD0IRk*AMS-I!ezU>eJLfJ&bE^`u#
z<_>b-9e%Qpc$jT|TpV;@vjCCc>HB%E=wTF^7gDZ1TL~<(76eHYkwDur2Q=4s2Jj9s
z3h#Ulir?OAvwaByKKU_=3q?R>)Z==j>a%baXWIml;t5~o8X#YmFjWNvPxr@V<Q=<(
z{l@pYz5r?NM#L9n>}1P{OId74E53ldAy#z4Zx6HWtQ(||I9`cftgHcn%caiK@(W~-
zVBgROA3o{}qO_V^gbo)Gg!pneTuaP!NO_=Qmjf7(I9pX;O(YdR^z+YHvrTPoY<r6=
z>CBhwjOW_`iS1QlyHOhz-BpSn4FFRbK9(1h5%SGCG|@VII$)d<j{)zM`P1FG?MuB?
za<2puu*c)TCoNQQU%e3TvNK3WbxZ>oR^w%x9aM9C7^_e@N;w>ciFr^N3+7xdi__I1
zooBdo+Ijj-Cwe;#iG<gx)>co@s|Cnqpp&Bm%BP>oXMY3}(X<$HLtR$SIhY#9s4%Vl
zn>ccZ9Tziefz)u*hAx*ay4}K+i`ro-q`7mI7_leuP>_TH`-%di8Q>l1R37i#mwL%X
z_F7<P?@{8xV}3C9LfW_f77on?K;9+y-wGr~!Z2pztRL_LfgX%|D&e2NICsZAp#MqD
zJE#}7PuhG__qHP#bLn>zwQ@2vWRgt-$y)Csv;$XJri9~l^Rd}~IWUgXZqP0mw>a6c
zs0C^d<e~>%L0&+G+!X_LTh-LX{&11>VpOWL!<F6TrEGR#nzo#vC48;D+x$BNx{s*V
z;#Zv+)?#lq<|e6s{o;J=Ll=k(5%3pjV?s<&hu?N<Kvi*cz+$UOme%kJIhpUR#^Ve_
z5oS}MIygQ~jIh^59Ek863;9+b16g76gx#{)k1xzXCH45##xGwPO^T6u=?E$)G&3%P
z3FCNBDsA>EQr)WAa^f|E7hlDCK*8=JTkb>`kQm-t%S$}YSh0ONI;WKFsAOM5ns5%Z
z`(>~3AvXC6kOeO5KpEt2%??nAr-U@hx9m><u6Zp3<Od@CjbAJqC=ES!%U4~-fkdtj
za&gtMzZ43=d#y+DzdJ(?4J<aGaq;1$;mjCN+8X8CfObZY@3TrU!?6}|)QI?COFWG$
zWgFzsrdu-;YeBm+XqZT;al8u5*3o{q6h-3-)UdsQ?I4ZmXTbJ_^=a6RKA@@u1T-&^
z-4QNJmv<#V2#_TXf!@DS*(7F=+b3X<{{FHf_!vxczeS~E@#6Rf7G}7KUl-y16QC@9
zNg>~$@p%Af>0Yc&O-XpeF1Jh~$(X*bL?abol3AcUCfYJAxSJbQ&H|}y$?zInT*{S@
zO38}OHor1E@2qcL_M4aXB7Iwlyd!AqGR|-M{`fMCVIB;rq$50Q$F|B^hZYo0;Sn2R
z#INnH2tKsk=|3sLV!cKqTWRsC%;4FL$siDhsPRt)OVp?V`SAQ6tmUer%Hw6AT6a7m
zxR??4A~?u(E2$Q;y*{$*^W>WQf<jdw@4X=-e*oeGfqDPd-Aw_(B|^Rt4!IBI<6cx4
zg63p?<HQnTwn4W}$^rFKKHt6V!!n~#cJ&Eq;@%zmjj%-O9Uu$;I$)IxGwA1m_4XnL
zJXzX=?b6IezR?$t;2<@+2q(jSbpQFu#92m3ZXzS<m08i<FIPLZK5hGPhyv$H20atM
z;_iDSMFVC_s~?CPKBhi$oLq1(fcs%1!mzs>#LSBn?zp!MS-!8<niY`$e%G`e@j3$=
zGBJFO6dv8F{W2sdcB$(k>IA*($6s6NlSklZM4)*`UuE`5N7Azo17<~|zQqgQ{HuK$
zMB+J&{rZ6*?bS<%7horQjip(Q5#_!{SbVzc(I!FQ1+nB8(z?l4nNTI&nvw2AVn=IN
zAE+~;XA$TWA9o&Wn&padx>}L#RC(5KYhHX?E(TXn6p^kQLYMk?rmP>|X2WIK{pMt$
zGSsK48*nx2?vuJ^;d+P^b~P)q)233rVe&)x1kg!idEfoThgqH+ZeTk*Xf!8ivCpz4
z9A7jR@LTLxrGNpiah6UFOh07MsOpS55auw4xvIia<t;ZSHf}FwW7@pg)a|}~lVusd
z+`<rX?Do0A^SyiOAw313$k<CvU;g!6AE8ed;OpV9KHpN$>gmLRIt}A@-ettT#<s3k
z0d&EumpU)PIcG8!^}0@@fXh6in>0t(3TIB_O1>wE2AULkK4H)OjP9z`(TtQN1f?He
zWOwZH3)Qm8c%yETL?fvJv&*?Lfi-xQZSCQm$x8ZLhyFjKrs#_pZEvGVyh)xudC8ne
zSM0K=hI)R14ci%R^0~rwoE;n6dVYc>WN5Fz7<6OIeXWm(b*$+gvtYAE<v`F#zWmF=
z?8Jvt*V!FpR9;d9_figKgQHM#WFuv9u+X)x7sqcpQY|Z8#CMqB`YLK8vgn0IQ2GiP
zbcm|mLYg2HK1EE&ZwQc*&XgXGuDYt2=C$+QAcVJ$@6(R<P4-_qX4IS{m``vHR5K}{
zP5iQUE@@KT{3cr<S7X?F-o`VMrWINuGtI(Zwct{A@Rl1X7o1Ooy-#mC+u)1H&obLh
zUjNOjN1wvRPS41_7j;CVB6#WOVPfgz&Uoxr$N`qXDx16$g61v$4KXR!%W~Znosbd3
zIQG;S1*1z=U%vi2r`!3F!y!6XYxW?ZlJkmFg)BF^JF4PquUEwO-MN64Ku<lOcbJke
z+r?Z@bJw@HTXTP1YSyeUi4W*Ck`ysLSXZ-d=ke>27b{n!cE0;F@oeLEqvYC?ppbm&
zG(`s+r27CDF?Z=>G;3SYrpM2-ZNbnHo&9E-wV*;i4+f%?7Qju0$AWTz%(;6)4)<Pq
z>&0y76puE?fK!bm93&KGn@<-M7l;GpQIJX*V%0DJuq0uv5M)6Y`hdLshz+rBohdFi
zT}X^{LN3PBNcl8rgYgbSZx+u9CS<aPDK;YM!-+2JP&KNWIqc>{;q6TUC#uGPCzD^6
zo^hya#OJ-P)*>R%Ujkvun<KyB=Bd`rjpE?V7n&G+4}UrtaSRE(x@BsiT!+Xm*t(DO
z1qvGfg@*aspl<0)Jfha?N8G#p;{}}?y5XE0_lC%=P#x%l6A+aHkGU<h2j-VN(Uz5p
zuk4bd`lxX`4LOB!IN~WH-rpOU73CA-oR#)(w=PSQwv|sWah=S_$Y^}eoK(U)tXN~v
zd&f5CGHa02$Yl3|M^Z!c>`(Ykd>G51Z7j&lmc&_g9w(_RO}x>n4NQ?g?%XPym$SIH
zs$7KIh8{<%m%6^1lchKkSvHnm)Jd&IsYjjqj^%WUy6u7?G9ud|W~dIK)e+JGaDHhV
zE`bm<xX@|NL8k`abUkKx+}U~lzSbLO{rc*pK*2QodS*>U&<Vy?5BYmMiP(hxLEo#J
zpJ>y)9&NHJLD*pb0*PPym__&&3QQ-fhhji7*g>avH!rQJT`N-?esu_M%SsaM_NP~y
zX;k%NL3aD}0|wNd+s?KilAPtTo237>#TDeo0Lu4h{cDP7K?>rvsSB7J@wJkBY=Pfp
zPx0U`=Bot0qU0}?M0K8#u%rwAihrC|<OxK{M>Npjm>UVlONMp}$6Ou!NzSFS!a<eF
z+d-Q$iFQDQ-M0HtfMI+!tc0iVT9#<*DkQn%UHul{tSCt!!&F?}$)7FG@+ZUXt~MK9
zTX8iCLSiRY(JQOZJ@cUe6SJoyw*v%twSFK>5NX%UAl;(f$?+!K(&aUOzQ#utPACD4
zqiW(R2LMVNq;1Tm9^Pl=q4T;6;7DnC`@yjw_1iMrldzLwZK~n=+!EM)Lf>f*Xcmbo
zIMYiuUWnen)e#CakK59%dYxO2>SN-fM8#|!_NiIOO`jSp?+AFl<~<Si1ld-VxN81x
zX5M+${OGrH-Vx6kjKN?ZXvoP*JPA(`)}L}ke{$J_AKWcZjn-00E~q4`_l$<g-EZVK
z*7^hm=J@hBkI+P-8zDg#GW4X&k>-Xr*zoDuL&~G_HBWXHG2WC}(1n0}K76x#fZgTT
z)<1oC-9TYt$Xwc56-r#XtwXTMO1`Ys;BT`2I;gcD;IiEMq{wePTCXHuBL=jGyy=33
zIqQ_>Yn0W@JmESj{1CIU7=%YP05R^&THKeu@U&L+aEWvW%oCcDfBa%fEnb&<4<BAt
zuvxk?1&MW?WjGt{Rtd$nSBONtw)?fj>OOv!;AYHl%3OTPt%{4<Yv8M?JRYv}1YA9l
zI}&wkQw(APcTNkA*iQ!N06b;hF#!hmth^wF3^*x07nM&0=C<4dm@n1sxbpK8+D58i
z9)pCst&s3Ip8O{lvsywvf-)6b@QHIOkC?PJ)PLk(aaGkgFzWYRTJlU2XA$F2MBl&t
z=@cQHr##E#2%O41HS8b=%>iomJ0q>+QpSE+Wg9J1S|bZwb*!>^<nw#}hQf49AReE0
z22(QR%n<#U>y7-&aBf-x_<g@Lm3h0z86MxsbLOk(-(O`=2g1(dg~R6f0L^|r05X>v
z5FQR~eq)?pc5rC&1QS^3O+kEJATQ}@?c-b8Q4Rv;lq*F1piU>R{6s$@r=CpB<I|n9
zUC~^3=+dt(UfzWALBcKy=8VFj=hG@B9U;seVkbw}Hpg(Rlt&)&Gz;eljC+tibhs5L
za4?tDvuYGKIE^#U7*z=-&C!p5K`5lt^%qstNhb6Hvn)4=^%6gQ#aA4Un#y!8cf7OW
z%dDczZs)q`Ix<72pnL-#Ixd-rp%@X<SbV27aLYf`bpt;pyF=2wxr(n78SrSwFxP*}
zkKqW4>9>x{mh{;d9SE@DA43ur)y0(qr5Z$bzpV$_eaNLyIbQj~_q?L0eG0fGlkgNX
zZ?!Wdgwl_`oOA`bUYKA?y<*;bfp#O<8*2f>9tuWuK1L`nIGTawE*&fr|ALp?JSDV2
zquSX-;4(}|JzS5#wj+dhYb+mc)GHpYCGCU+KS|FrPApxnaOI)xMfLP)8MZwOKhie6
z%63md&h}vRu^nfRi|yfJ`wPK>7z*}|U9{l3dERPS^uglX%AVJ_x2`a^xpZvD;c@F^
zcJJX_duK@DbVHZnbWS!<x2U49j876)*>>`WB`p?%5ZFm`KLv4+Qe_W^nn#|Je?x&;
zKHkL`x+NFwit)7eYRDI8H{bjo`{X@aD*}n^N4LdQ#L1z?>t_mU)`^Ge_F!!Ds7*YK
zptJjPy@FfE42Y=fvfMU~YA(Br9aZ6CoWYZF#P!7ufft0VHHx8c;sp_~Rc^+w%3{XC
zDQA|<`qwK@4l6Oe$O(#3U`j<ZzXhny$gS$WZ9K*4^;)SPIkpNm;8`#ZNt^Tb6g7Df
z{R7ONR@Nf*OLYhkfnA!&h)8gIM2JocKV}8qU0dfoMg17DuGmm%niJCL8!4zg<t$8Z
zImH#unOQigaqdlwrfr-_;w0#YGp4pky$;K7%&*RN>`lkOwpMBs$KkxKev_ZjK7827
zpXN2@GClB)%C%HGPT{AqreGVz9U>qXCWiXA_(4pHfh|tp@pewjAVwd<3PZ}IOwF*<
z=kqW~$r1?`Ij&C7VC&e!V(|9QoSC?h5*K=OouF8WX0?h#h+gCfy<3^c$gc2G-&1vm
z52iR$4$juI6ttnJ@EE~HEbpkYsp}#|3TK~D?}Iry!g3$PpZJ=3*1q)J;;5bkiH^fF
zKLW>%W(I}^xt^eRCSuoVv6JCSiwX8}4|fdH{I~ApzHpPdNukgAS$}^zihXn+)dMF$
zM@QeT=%@S?@MhJFYh1`A6x!0*P8|ZX9>tPMoIVYmfL7$jbkD}0@kZm6^hjUw(bKfy
z#f6ty-ba75$3RdwhYePAm1JVDDDT0VU~K@%NO-pqUf0dPiXMG(xi;0eB(-c^)Z337
zCl-33hHyDtE`p^ZFBQcQl&)4B-YFraKW@L1IEr?(j_-JDC5$>;tSfJ?*nGB-os{IX
z8H<Xe=dLAjp(m}t`4Yps7kei$Vp0S&FOk~8wm(VR@arM?6F>ktZj<e#vHuXY$vfOt
zBEE5(XGFqo2aLrMs_V5|q}~2)^EzOBuH~U)W4}S8vUeX11E(X)YX4?qBpj}vkiZ`Y
z`jU+C7K^Rqt8KMS@l<zRsmoRzqVT)kX8SO|z5pDa0SV!Lcd6oK7Pc=_p%W8+>};A_
z@5w)!z<ZOf62+0jEz7jO?56p-o6QgMi+{W5s;-mb!)*S6U%9Bdb=@Bup|t9&1?E^v
z#%(L!`6#!+PBd{5j9i=bW4u82`B6d^399(K{Au?zzx3<2_3k+n;?K8lBEDmvRpF#Q
zzI`B86%=D~7QB&ktS5vrBIw>URgX{iIUCrKHys8aeqT#0R^}shVC>W}(l<>vOko*R
ziP(=q%wzExJ5dwj`xRd@CsDyzl+oq(zd<B06?`@8X}@)ly`*m8g&=s(u*{v>Fe3sb
zD&mQU5jTU`ABY`qO^^iH-$4pBor<-na6qxf-L+Fco%2#|P{VQlqa-!OLX$|-MY#UN
z)k~+zR-I6XXZ4cfw7^6!1CooMZ&WT!MH?{=G$X;Zw{7FaA(_orW(kuX-Vdm))cK)d
za5rvreai7^@Ra06r@I6$JEsM%zO*{0BF(1ljxd$+D<fBV&(gw4t>%6?Y(h#7L4BDC
zQVGKt{7h}vu~<kJlfQ(#Lz$XOwY~dE?O*~CyZ?<hEZJemwm3#;8Zs5<6j{lnimODq
zWDtJz@c0~|vs24cHm9#J!QUOS^yS5`lUUnV9}uNR;DnHdw0nQ~;fbyGCbzyYENji=
z#L}i@Rz4_^Y7XKLF?%*D6YEkG$G+3=)L45k|AI3Q9%MbL#qiB+C!YKfB7VJQq<@w|
z#gOJHft8w~YtSZd;a2?P#0WaEg~*EIhTutavwGH|k<fA)$2L>ImsrF&9O|#=i_n5{
zmBoK>wx)SPbZq!agJY{87Gs@Q@@Kn+dzw3|_@O$$CpF-2>3ptS-kl^nGym-qHiBZU
ze(+-a+C<t_$vA-lJnviUrnq<cDel)roD$j<xIzoY^64U93f<nctJsXcuv*NjS^bvZ
z_3;U4CE6p8a$0drIQ<OENL{@>$470;=IFIr?KZ)T$y$`czC09OzGCP^j#+?QlNrSF
zWqKE0={Uu$qm22C`Z@M|(<`sjQ4PhX#eQJ*9gZRE*K{)b48|nd)wG$xupLGw*|W~|
z;S~n#@OM083j}oP1P&Q?Z1wzc_+o3qVgBc^4@F@<UJOYDm&dLzBjO5Bq!#|J6TWq4
z&e9QV21|0zI#0#0&`r<PFEa_|5?T>kNwQ(WBTBD!P^>gv-fQLANKaF}MxYlQm&iER
zggJ7}q581N-9}0lguaHBjJDQKTN^<qeYd1Q6B6DbekO2_bDs`1*~IsXis9I91TM|l
z$#mH05MQlDrC+(s8G2r{x)P3k4mYda*&<e^N7zhgq`3D<HP(BpCyrnc@@UBMr0Ky`
z^PWDNM8lh(L}#LP1@)a@Eo~|Nwy_+f#sFj~=i*H8Y69pVk8E(rJ!Wt9-!E&%S_2zw
zaw~*>-dcXBlKT1~pcFh9l4=j)wJ#jBqa)nuk6nB*#`8Ls_}FFwWq@+Hxz&Hul}*0$
zW1u1D5=T^{mUV1!5YX2sc!#`-mLo(PNk`{^!+mjQPF$7W6EEmVrGWqa;bkKYd-!a4
zmOba`Cx0-7fzO`tt;(Q0x)Qt$X0U0zQXsoMxr3oGZ>WS%aQcY@aXq&A1$Qxns>dpa
zP-}0`^$WSrwMT5<GK<0or@1f3(77vWX})G}yy)2qDglu&8F#DWZY9uY+my2a<J!)3
zR~!j_N19Tsq~mr=XV7>}C3*Y{2p-%#@4p3&^9@V+K^Yh-JaO-rL9Bf2TdaRui^D8m
zG9w{!xZb&r<Kx26^Pu&phy^RvX*#bBh;7K}uLh>~lq5hAp&xIznw&}{Dv9Xwn7rm_
zQCu>Rs4wvuUcW%lQI~1o<KbD-eCBW|Po9B^0w+Sn@lK%c@TBdE^{ewCy@7c6?v+a$
z1y<D2_Fc$;Sn|Qj61>Yw?qXB<R?Zs^pyQQ$gLUeVT1Nf}iBE2xGmLjHarD!C<6Pkq
zY*r{r`CA=|x7Afm|5}P&=wvY_uE(GO2nRpNB!qhL)*d5HR!_FWlkCbSX|Q_W8H`U@
zP2lM}SIv)ymAr)5?XUjIHaU8+=7zgrBVd^7Bdy@NoHq_i^i#VF>DdVcHE*sJ8^c5o
zdG5iCyG3_D9Jf->>NFZJXNP4eZX^*Vm~D?PW%KpV6EIiGtstY&f^ahs_jMjFqK;SW
zgu?7nhX+5t&*)i*+U3$;>rWOn39g1Q4<vw^gz+mr8|u>32y!F0dM_$-)jG3=o-Dd!
z{>RzK$3jnuaCDcfO4~k|5@^``gd?=saNFKg!(QtaM3$U3Da_Ly++VT<dBZi12poj;
z&AnmWAdwO*Hjiw8lCV0V1e8@6YodeP3YHcM!-lop2^|Zk)7i%3Jj=m(&AdA0wBKaf
zp4)1S+#L-4yu`i7+_MJdHB~j&flmqQZ4MM%`1EYGeD}d{WMmQZblmcl!9rU14Z$;t
zRA!1j)vpSNbglLQ4y-Isbx%#kk>gv<SU@HRs6(~BDm(yZy%W4gj_>2)7rFLE#o=$s
zA5G&<V--d<t+z-QrtJZYtRFAZS?}TPio~F7hjtS|<l6yO=7s@VTQF7e26UKraK&-N
zoObU5b16SryM$L`YF^@Fw|WTy?~dIMBlqnCxo2NQ27&!j$i#-4jk8j)?KK4AIQy4N
z6??!DE>_9VmR7A5t6!PP8QLS%3=ci+!Mj;*MlZK<!Q9O2aDu$DIEA_#8a~>R(DOhj
zbC#0+=KOCdG#S0yNIx1(@9dHp134zDz)x=Fr1M1gu$G?fFcvkf%4f{}eP=wRsr^?w
zBh3YVxHu|wC&EUsw#6dsaPwp31NGr0A@(h_;K8ufXd4l|+=wjYQ2~g*^wt{F?B{a&
zpPFP&fPuJ{<d`qs!VU)-!63NeXL=GPBi{-}10y%FXkG{)`3-x;elgFY@iFqAuQk4O
z9_@R~$T;xkY~yp@42AHb_pJ(VE**^`PL4H(OZ2PL1z&nqntDdB&gyDzym6b^*s<If
zm3pbv$P(vV>C8fPG_K_^9e*?eK^hZK;c9jjRI;q2v`jKSOQ~31I<gsG^M>5UneT4n
zY&8ntEUD#s>5j_JYpYm`n5l2>Jhdp_K6~``xH&7ckdaxwMMoJ2A9xSzk;5CA@`Cdp
zj_z`XN>n?5ke8fq#8Tp(6P1|-!^<lSW`vigSPfgNBUDaxj12nn31Pc!Y&+|D-LS%J
z2qM>VhEG)Yk!3t?JF7Doz-eD4G&v<n<fp6VjuUR^PiO78`PsDXu010~j79~3+Aczq
zbBE|X76${~pmkbi3R_*{-#f?s*a8&<m-E~8R1*vg%)DA<x77_cNjzN!dx0pA;=Nz~
z_*%{yt~Sd{(YQ>SR?^9sj4coh6iVCP@w$`DlCAkI=w6tp^ZsbE%Lm>U1YtYut?RTe
zL^F?Vy1#OZ*%tz*ur&xA1Ql_yjvyEfM}1;ivDGhgE7oP!^P85TI`<+Le{PKdDIX~|
z!fY`QoAT&0tH{Bk)zGFuNd**&%5l69;R;(e61dDiT|6`CNv^5tT4(+96yq?#G61ge
z@$q6M+9Zg>L!mB%5c}Qcj`9n=*7R&{o;`_kEZLS}XR?Bvr|3i!c0nSYP^>*A?H#Tk
zvxwCJCK={5CemXgph3oFWdrZs9^6Ok3VBf*2AVE90U+;1&DXJP_Zm{@3-RBGi@8@W
z+<Rt9xdzgjFRN!uw2mgoh2n;u`@JC`%kQO~ZQ6bG^kfW2tSM-Vdib725D;Hzn>tkk
znj}%zVVCK`3DE-wMGtNut(h}&ZgU>#zQ<Ivemu(Pj#DP|Kr_5H!*7<K`6#3^qeDG1
zmOne0!~x9QJ<_jk36yORzo}xIl#GbZ{j!{kdaNj#;UC^3yt401DQ*<ooFm*`HbhWl
zem&fsSd<?pA}~Yw2`ehdWl6U*VIA72fB5&Rn#<yexPz;N7WNYTbs`U?VHpol?bovD
zrLg4=*7SYlfGXDakTP;oWX}2N;o^r>W7joPD&gU|ykRW{K3Yq^)XE2GKg{+3B#)@x
zRm4^j>{z^-(vtV*Gt>ZpQ9YpS3`Fc?oM_|30wHrF@r6Z;c0J(CNknu5O+;xCN9)tl
z!oLW0c~aVaglLEekPzW>&X>bf2=Fm|8AR7#%ebw>PS#->=YR!&w$UKVZD}NZ!2hmQ
z%=NLSaf0F49>9R^w@OB#kh7*qsy1>2XjM6eLq*Z6Rbs0S$i>6PK94#EAwo1K-gh95
zjMpt08=;`h>;~@bbis04Ucz=phxRQsGsQ5-k59(69*6fq<*`jVk;cOA;YoNRzO}$h
z@$g`q1%k2Mk<Y-XUqsqBaQQ;OKl7O+0@}3+eg{ROoPcV~Bi-wbafKTpy#&(mpN}i~
zz4HY&8|3q~{?apcqtE^1u&q_&^VQ7-3@wY6qKzKAcpa$xZg4rcIfwT(%j1-SDz-kd
z+Z@Nk8z_ypE7MGyEcB2%Su(JSC263%Mdh+0a~L&kPb`?&hZZz`{styIjE&?ySla1t
z4!}_l6P>`653=LcI#{f$kP~U(ppi}PHHgj|)ItHMs`TJE&uq>6K@2Z$e<5@fqd;X=
zNgdQP&IrZz7MQWLZ4_j?uo|v*PRbS~P2=Wh%kp$U(DptF@+Saal|d41<1Y5lt4m$!
z^2F}JVzw3wnDAQ<%F;`EWeUliU&p?ef-_N2Px@f#3pA*Er|b+uc2f_Ay(>7ag3V(T
zBnBQ@wAqs=0|LRUl7+Mi#EY*jYFtKZ{qCsmm1weSZi)oxQ<eid-o1rf6kkyQxqPko
z>xq2<Ps~(*7+DbU%@wVZjodzkF}%fzNmg4S7(=I}TWW}HkV~b%_R_AQ*^ip6kr$hI
zEkz6M3Z}cy+UoK>Kk&X0JIr=y#*|4VF~2$X)28@`2zDdY%x|D~EKUj+)u3(N!vr$K
z+)M7uN26AHkwVj&e|>JwU_APq?6;>-S%?APr6P!}<<?vsQA5dH7<$bc(~Wz+HcG^3
zNP3~4fl3l3VtkTIy?Sf5XGn`dPHxlPaCmG`DzRK^FsG7M5CwPW$KEi1$Oj(aha><y
z++NLAX}qVRW>z32`7ZM!8t@W=^1Er>3Nw+H4AvU&IWQ#HKU+Rqm=CA2D1*DLe0Uu@
zZ<|l;U2+kv+9vgDp);?l;wX-E?AHL3qO9jj^I2c#j99`h@T|c6bmL5*PkOY?wWnKk
z1TtpLOflSUrT)4x;-reWijDOqx}Xe4zBSNfg1K)0=zG?6g#(08>4g!)Q5z1LJVlv>
zpPGR+EyFuH;o*H|ANd+}2H;sZ=X_wpGjU*UeLX<Bdnq}XqCB0^Z~DuU!{QXNiS3Q!
z%I2^@OD%<Pl~|#5$NUXg5XhyiEG3ZAawi=6*2DUha+3QB6cV)2?3lP_T$GV1l=oxD
zIqBjApM#8VF<h5kw0uD+`(9WQ;0s3SjYYNYfO2nK7AQ_vBD>aq<J;zl<CX!llH(i)
zL=2TBFo@RW1gcip=ryu1SdADax?qPIb$hcrb>;_sl}dE#M(YcFbMHtgsC@9|%?u8*
zusSb5wDj1<YV!2#S;RXEJDVeLmNhD5s3$%`^1(sc;{vEo?8;A#KkBJUM}uR{i}Hws
zUiULJZ@H90k9O`Z6sTfgmg<-%4|iGZL4|91=TxV^2Dd<{2~!=&<ERBGcw8=m)U3^P
zV&8!y4T7eg7x6<pn7Hx<YQ?vD<2PX>pGYo)jAdsi<~jhev}TW;yMcgVray^-HUGXU
zttApoC<rPsK7qT?VZcI-k#y-O9&iyv%m~cd!8Ag}x7yVOy<|L;EJo=f9Rh-Z-6fe@
zGyDWX9n+`ocf1w)Qwsni-R<*o4}=6v1R;)ehsY-Ldmb<YDtYYFefQVoPjq-&#!eFe
zGvh3UtSnm#ZT`8QY<kQqA?VKlBv#5mLWjTsPz-kpx6bgmT)nU~Ef}G6l(1LFKxR5w
zYu9pi?qrSlBv^T;6cD1kr-FYc>?a3QL6G?~2r8rn@<%CnS0FZupDV(`u_yNV;qlhz
z@1y~X?$!;1J{eu=E+o$~-S%Qw!28vNe8eC-VecYjZZv;2x0SPq@SdeN6t{_bs`V;|
z`r_M^5#@q~dv9?>80XJgl^oCvbL_R+v{Hd;$_3fsgA5QfV96e<tE_fv>I6g6)RSRe
zFI<4ilNA=n!!|QxQ8Wm;*~KQH?DEIy8_k2kWxm#L5fQ<hU6%W4*jF1VvYT<y2-K85
z4@vfl%nBgqNs5U?_r?}<>tjJgnJ}xWgxPp`FD*T5$QpJ=rdP>xgJtR%RBz^m@8OJp
z4stOKLO_8{PozdX&V2@gfkaSx(%9wMV?Mp{7_v<adSIJCMF>n^8^!P>1d-^cdF@+N
zz&0m-*(Mv#xv|<p`DxZ`D<I|0FX)kBh<}`s*CD9E*#sy^PeuS)>W0jd>Zob*KvSL}
zWI!0XekDY+$SOWZJ6*KYR0<N*eZa*Ta>&p^i*CImX62AISpbL_slGKxILtoUJOZbm
zhRb$u!TxRZ2nfP$vR`BsM6B8m@P^Sl2$GI}19*Bp#G-kYS<rWn*{WiaX7w)f3s(TK
zvY?MqC*|5U%7rK?xUAOQUv3NvOmhTb8(XbFg6AD4Owfpadm61MYPAfNy0!S_4%Y_V
zL0szUlB>8i3PMxsIBy!ric9{q;tI9OTpfhV3;}xPJ|f;~z#!4(-EO}{$$WB_VEh5K
zd#;Z8Ly%DQH$(^8PDSem&SQbBH67*a<`H8Piq)pNwB88Q8m?Gh0+q|X_riTcua{2Z
z+oV8{vk4XER1z}FcZO(2o%RKOJ2FkrMc)LpJV3jWnbR2_5%gZK2GH5)b&8u#o*j1a
zy3=nm>KI|r4DN(RF^JBF>-`-UvxXqfphT!PH~S^t?%S7H(R8!NINVU<#E!5u^^=l{
zB`}HPBw?1qQVg(?T7o`WW6wjWcZ!k-#aio1s=UkdAQ(jQ+A#Z3k_4U*n(hOYzwv#)
zr6p-N&Tu6@IXnVdm*2MSL1H_qeVVyhNkeV<CVJ=5(<I?bQ$u%GFQ2sTW&u9Pknm;H
zfm!Wf<oR$s43AnSLdL$<z#Jr~nB_miLasG3Jp;WdiQT$3?l>7tku<(y%Arjg$YO_(
zV_m8!>LBM4kfCoZztEHPbIx|oAbNHTSTEMg)@|o2Cj*5_j;3(1l{rp8*J?kgY#l6y
zw_DiPQVF8r-3Ege>HaSyEy4Jq#E8u#=Z>>P?&{ycReFbeL*U2;@gTD%h%i!m)yMXV
z8%47AK)9Y7YOio*BjKkts%m4YMzCi;f1FU(%N`?3lpU5H!T|In8FrWLc$aHuEv4t_
z<c5GfAYvwkm-wNKhUcDZesx}!e?X66Q6*YD)igJB^jC!S<A6?z)ouHnp>>7b`wN72
zFUB9CAt_Rw>}2@;E!X|{jXmAigN$2VeNe`sB8&h#{6aRE{3ir&flvC7Ttl5R)9!5t
zg)X99Kwh;~5#BuedfM>!I9XD=fdKsty;H>S&M^3;1E`3+z>}iDpQ@f^dI?GZ$%}(5
zYgd+zK|(+0NIh-|62N80gx>8IX5GuwGn6cY*-_<7`o>PhOO#;Bdi?DQBs}$FB+~%z
z%~Gi8N9Nr_SNcR<Tvb~rDScW^he9^?DywByiH8#H1CKwMFNg%>zJpiGo=<cb@GA-O
z^9od4*<lb1DBOMhM=hA_l$(a_;K!7TlyA~qN;V9M&Q@-xoz2NNLTq^~W@Aw?`C$B*
z|6?nP{qHqjoGyXFUfRiVklYQB((B~aRA26wa>Uu!s;~ndsjnf#ekb<zJt{D`p3b)-
zt?L`8=8YuGwn#++@HF#n^hY%t*?d*tU4&dYboqq*Z(9%S``j9Xm>kY4)`|Bq6`Jlk
z;E%yKuX^t;s8p6P2rDz_;S+VYi_2<I(wRGx>)Rb~_xESLZA&{1NVt#O&IiiSwNXpC
z?T1E{U29g+0`RxsH~%+iIo`n!4CoK3OG7M&W)UG)M-3u|=QyK67Dy;<gH5<l!v@{g
z<Ym-i#}h!ju8|(Uxx50p3sMuS`P%r7J=vYF^tx=xn1kZ$E&KA{+!%)^Ry0<id2Hn^
zwT;NI9YBT)D&XAjzR#`!Ts!QB*pNW>qE_$r_$Gc5fMk~;NVW}5Ya0h*{`pJS#fHWP
zl~L{J1I&+B(GO<b7#hW5*^g}=cK(26)5`)yb_U*^ag!8JbEI_5)4nTa2VgAvcuZMi
zJNtz74uEr(Z{7oY(si3;{QY6NpmHk9B3tXcQ<DVkLx{TDtJ3|9DX%Lq5>s@&j<j^3
zr(*8U6D_z(oy2H#Ea<e7a)^!z6BZ|SwyNeE{lIkhe&N<r7Gxs}Sakj)1Ch%ReOL%N
zJ-fLMLyyCJ!QW0+IOlYyBLM=Ry-75Tg8)r+3zh~2Z&<!g5WljtvAN|>oHzcA$d7KA
z4tS%AGwJMRV3zKg^sR{1*-ZwKd70*5cH+(uG$j5fBIO~BRK}^(nGE22aQ`t~)%lY$
z48jKAwOgl<=vg{0eyFhgU`RLwYMF7Hzy9*s8Hsdhkp)HI__omiREjQqs$Ix1EDGtg
zhzu8Oj4*4i`(o(sKOYv7QN!vr$u2MCHHRv>)l6S{*M^LOZVsfG1j#V?u0rnBh0D2n
zYiGOog}V%v>?!R~3G>)=X}yaB`bh$shOrNZ))w-bShN6_RgfjbJuU2ly&<4h%eqde
z@a|0t=3eBKPq<T^R!u>!Q`hL?FCn%WgIs%R1!Y?D)b@d_T)~n}!@|RMp=C|uE&$qJ
z#tY)6!@POU)3n}VLg84=rdJ7HIS36H#?fVT2*rFQ2BS4vDFl^yu?}bGD4@WG#d%hO
z#6_il0kFVeo>L$y9wB9?kDTJ4eo?5nGt5FKpgBo>L*lCdJIG#~x)V0%Agn92ar0hq
zujNq7Zea{}drx7lCzxg(B)g+?Xr_FqTPT9!w827QVcz-b4(XF#%&h#G6oOtX|5H-r
zTg@%UuC6_(OH<I`^$CznngvnEOv~uW(DOwcPABb*;~~ODyRW&d@E%0D+ruq}0Nn7-
z*fqaCFS8={c>843<=yKC-p;iB?I|F;H3Cqw>cJuefVp+qX0;AFWF$)9_+HU)SmLT|
z41r7fSH{@46<_lzlD+yNK!==r==Fv@ePwIOy&6#5@)QlV3NkODF4_>7EEf*5iSlTr
zU~CASh7WHD$p#xdyJpjBT9{OqxzysdOJ$mK%Z*Lzmxa#wil<{E+V1IzUz8}y?QF{M
zm*X4dCY<O#eWSFNNQec`h2Xi)DOFqCB#@dZ4JDtet<JSn5pCQXA4Iy^R^Q`uXcfk$
zVh1ShwK{{-A=LvQ8sll_cnT;np@_Ooo@@uXr>B!M2%6Wmx5crFbKlXRGjm&*j0;4t
zqwRqI9x5iuo(1BI8eSd{wdj(EVQ@v9*Mh)&bueA_y{4sz4bI`|*}=0hPXr*kcwnLQ
zcYG}RHik6J<pl`8)vigTkJv^m9y)hKikW_X^>7?2o2H0VGy2C2WbE-76=Z_yEJWl=
zEoUuS7kEHzd~$Q3rdXs@W>}Y=KwHn_or*w)b4N>2hNFEQiS*QEr@+bNqQ;~*_!)Ke
z&&wzx_1&R+8&dYpOgc;F)Qq3j(X(nMr~GlC;UvH^D1|9;yU-+ynV1oCee=H7!NaI(
z81m|7FgXyO2He-M(<H8RvCe?<z4}mU6cip(S*CMZrKhMjNF|?AIh?+|;eHXbb$rUY
zg{(!!{~*ouE;)6GL&ZUFwyhOr{UzV~(fWvmzz@fr62z<nY{g4&k`q@+YtvmF)*%9Z
zbHd=NOsng#{<TJvuXOV>R7oAFTp4W{)|7}YKh1DuuMDpqsfwO{qs>y|Hz`~p<NyGT
znS475aR4iNOO%Au%f9n9{^+aVK;Y$E5+H49&4o2kHI2$Hq?ds3EXjVqRV!hu&8D<c
z4*?`GbUY#wEho3e9kPvKuMOd#PCLAyg|;AAIAQld3*ImZg)Oacq^?|hWSX=Zc?bZT
zSWu@uR1De#@LEFApj|Sp$$NKWF4--U;=z?{E`x(KmGZP6`cBcujf$_y9d9>irOvia
zMtmV`l5P-<dE8oe>jk4_9a6CTy4n3+x%Wr_qNJbQx;@7kBOW*TScLOiE0?feh>`7A
zvAWa)XLNhr-9r4tOk3_oW*DV}mvJC*Sn%)JCmzeKF~Cx*mmH_BQ`$H<xEMing{aB9
zI9GNyX5b5PS2Szgt?spw{_-sA_%Oy1PyQTl#b^E7u3~E^YqiA0a1Kt98YlR4=~{q=
zGhN-F_>~SL0yw>%50`Q`SC_)ox<JbTzzM#6!Crz|FKb({!vpX0<hl2^H146=`B2mo
zZBge`D)8VY#Xx1P7Ii#|`bakmx2ZiBKrKPNCp&J*dNzlqa?-T)oQez{d^_Qs;n0Ul
zqa}*>BsxA))K`c)y^Vy=P52_NoL!kO2N<15T?`M6*usNvfC%ShWT{}+YVxDK;SAOU
z%feSt?f|A`Wpy)_JD!LGzXrsW4qI6>1@%$Ww|tZNk__=L9{S|5;~d;qqYgL%=DyVF
zU>>>?y>pYuo*?3k(TNex;TL%KFfw~%+o+6uqNQQ_z65g8f9PD%03lqd!VTFFlXF9B
zT(s$md$4GCB~tXzqxhbxj~b14agFXCC=F2o>RRD7t-9BkSD4yE<46Kff|{XXP1hmF
ztazh^xo2}ky5n`&)zFN()xwQeFGCO$w$(CHern>LpE&O4szzoqN_6e<oSQW=BJ0z1
zT0W;DK@`6D>8&jC;YyxgI<H6ZwzN;n;5ii+B6tqno!W|Ph;@xS`szLnkykF=rBB`$
z36$$WO&w;mJPQjyE>3%vuCKS*&d1JBy6&K^Bj^mV$B%@G%>lY_h3TEsw-|2JB#?-;
z0J>AtrSJR|c5nnX!dE}FuSS%;G=4mX)6Jdc$9&4mh#>BOD5+v`^JmFU90N32jkyy)
za9k~k?mPi7jYf%i7YZ;y_M^@PaFk^29+D7Vtr`BdQ^)}@P8rx|YQWN5;e{~Ha3uL{
zMEn-_eBOpSRSD@)wcL=MZpQUpP+TDHpJxF^<q$}lmRZ52tKT)SkHFVUJi$+`3=|E7
z-+EyWxAV0;T5sH)6QG>Ev)Y@iYkFbyPH#Xu@^+&g7|g6xPPHd=Y3;g;t?L{Pkmm3a
zikeCkX*9Z=IHb@Jw5T4tY}AX`I8{z+rWqF;*I?h(zWA322oxYrrx#<iRpo2@yu2xH
zOR(1P<oT5>)e!WCP2j%f{!orJ!E&0H6F^<<7O3l{voL3I(<x}K`qXEVc8{%dARNeb
zVo!?+rowIS`Ld7egnNs3X!A?C5=b@n8sAoTNXagupn)8v@f!6RoU>FDbxs^<XJ6xH
z!<FT=PKv8fiLL%vkhUNmcj{v6jU<tP^tT^(FzFaC+@9twJTe{K{{+x&+XPD^KclcT
z^SGVjR2K1rJkBqd%+Vsx9XFla1DkM;0w#hl21^>9;0Pp0G*BduO$iX-PNoHp-;j5>
z$3!}%6ep9Y=Ux2NGE^?OAo%7+osZQ(owp27(PEMQGT}gf(=yP6i+zpwi?7TD#%A2a
zW)hFrrjpSVtf9Ou=W{QvXSLYxh0(a`4f^JKjnwho8uVel;|1jw*`|Hhj^!$luDJvJ
z`iXW!EObp-ll&9n>QeA2waPX6W)OnZW$IpP?*B?-Hhbn6&;N#^X%G3O-g?%r66&MQ
zU;CXk0ow6LL$o&SxwkM6|0lBx2R&OV(ZS!ocG=aa8ia~EhlxJq5lsTFctuNjvS!t@
zTZ@6Uc0-rm{z11LU5CE((Q<awYhg>#PbWuJkG2gVUraE(l$V71FhXE8r%iSbK{Ki`
z^7;A_C{nmtlaahuJGPxynOre2cD6bfrnvgfk6Hd~R)u`Rd*%<K({aL-+=VUt45$Of
zK)i~%)m%DxV}?V47hb}(TJa`%G^~g66>q((=YseXAzs(bm<_IkpM9imgY=zg)hJWx
zZ)-od+K=BSOXc=`0EwgFjG$DZ1?cUN81XG-prpWlKWCtC>2H<ir?&3_QJVHse#8NP
z-k2W;*wf1#D#tbi{9>R9C|RQ|daVS&kKV;RPKM+;WTDCAr04bkz(k1#HJ6)|dbEIh
zBE#^A_i<d6iPj$?@}Lx^OA5MI!IY@qP*n)z(&ol!YeDqoB}5c9PskODi$W|}rYNPv
z=v#qZFf}C$&@nY1XE>ng;!eo>Np!{HXC3d)0u?lQg`@v?Th08-^?95DvM7xzh~esS
zyo*AN1(NI1_x4L&{?WHqHY?(G<N#P*RKQ;pZ7-?vy|NSi)v)}O?rpV}3FG#!u8i+J
z;Z@og01VTxADN`>A51U;2pj<f%;Qwl**=8`bVC(X5|yp&mBMC8Tx-`YerxFbv!%Q<
zy2^PjJFi7-xSrN|t=c$UO*vv1fW{cZ)QbjDr-rc+OS-W82*6`@g8CLANs)69-FPs-
z=iRU7Oi|EvyolM*`g7a<csfJ$0M$LZk^vHyafsZmnO)HLbUavs2yfUlPtLxMj<)OK
z2mpU5*-Dpkzjp*eLo<ZuW#rE`(mI(A0359}Ve_|&-ycRHjuA!@&3?R%0=U6ANG%B)
zsjX#;XWp39fYw?{fXEj#pPwD3Bh|L^ggBCNLv?FbkOnCzxKA{p_g?*kO`>mrPV4py
zPmX_WhVMT?LE`1IHY5gX33BO_CkLQ-_rM-h25D&bo1~L7UHsnTLY=F{0NTOGYdQgH
zBG>CT08NWJVKZ*l1ER?^sIdRMJ-^XW@Z9#hwSXkZVJ!gasX+Uqg$KYE>RI&|n!nye
z>DM<x=!p(U&IlH>D_!k7-D$a#<tu)*PZ)TChIFo4C9U7`6Ud~#o-dwFaH;|Q3N%p3
z$r3Q3J)m7T0{W510WXRWY96bFN(cle%Bnh^A0E2`aEww}%Vj&K;gX5R$H=E?ee&%8
zVuNr{y6pp2H0RhxD?1G8+-yES18!|GfbByQ8kCNuq-KA#SAtcvod-Sd26*|^f9|e7
z41tlfswwE=5gPQxc|oN}H_D^?ET8YF0{U(@EU!_KC0(8E(dG}d<#z|ck%zUEjbzbs
z`Bhjw9~}qoKp59ooPS|$zkl|Z4gP$!${#ZH1`?{D+vp$m>(6)o^Bc6SAl;D;1=#=a
zo<CyhfBI&w2cSAYDQ$Y{KmWQvFXh$@n)3MsBwYUL_Wq+qwDChvgZlI7e|Q@I$))HS
zfZW0q!S=0RC()lv`n8+?>4$#*eH@dJ1OdDb`(OX_|M=?vdk>eC@(3FHH4cn8PMGZA
ze_Z9izJbwKF(5(@B)nw6(*UlQ`MKfJIA|kI_+LwBWB^mnGZ*6vsYcb4Yf(ec(U%#6
zz5m5FdUAaYs{NQ&?9b#~nz;j%uP!~p4EyJ8{_h_0|4CE)yY2sD=lq}M$iFs{e|Yu(
zG)I1hBL8q_{_24Ke;_84Q;g>fBoyRML1Gv6gf(C9YbTs9UT>hPWk>9{eh)$#g=cPV
za!2NovM#0%j+-+2g}t<CpJ=mt1T`4_OV1?B4vFmWtzK$8`ua$o@MK%-glB8tTX^aU
z6lKl}?}E&M$9W7Bh6_QBC5501(5t_E#U}nR;m-?ljAHSAv!7f8^r`F4fa#F9;z8kr
z$D)7%)p_=S+R7!UI^|tdHqDC99nl`D@6ziZCNHB+jnT(5sjx}&H!A~|x0Zfu<6aqS
zUUCB>ik#N{EZ8XCZV+pxY?p4m3^k&G-d>kkyM3&r{CL};8yPt@XQx{;6Hb6gfZ^Uj
zf91DzYXFxL$^#(`!OC<7AiSVb%hRtO#wG@e(sJu+kh1fttML;^Q3@Oln%mE%Tp`lL
zIH<5+ipEwZ`T}wxNE@X8HKxO!{K=O5xU|@TA}7TNBx(Z6;$Oe4zW%=d^5w^Xq7@!Q
zxG4s)Y1k|jo!JkzYLe0J7B||Qu?6~l-XNU(Xl)j6(}(xH!dEF0B2J3Y%91Awh6-*b
zo{UVv6G2JBAf{^E8%Fy*cAk6Z?S)Ci{}w(6N?fxwKz*bp2OxR8t|`kn!-4MGO2Yyl
z5Y{qpmsfSW^JT3yxQvHfwEi~WsNy~ksx!os90Y?Lw6!{Sy+}sD9Nth@D4-;U2CfLM
z?pmB2?-D}@eu)dni_W0?00V66&;mVU0S5QPo$CY5)UMg`&d0Ct*>IuGwkmrPNPWu_
zDg_pxLpwmD2<H)~s-3;hTJRWZ6b*aV09C`coCR3bU*Z^7MXwxfB8eePJ+snGQ!i%~
z>Sg8!4G9O1haGkmT=mO<@IB`^1BQA%f%^SwftJnCT7HTMYaCQX+dcl#&9!{K7-;U)
zLi7v0yAMXo0UyTOu6$7=m6pvrAE0s`KpI96>Qp0U&dh)K-0z@3X?j;UN<5~jWHBqY
zr8Dy+v7F1T)ST(?<ldz_-Itlr`b<|pZbaXI{k}=zp+U$M-JtN$n~!jB2y^PUDSX0Z
zPtaq$&i%TK^NxE%(YJ=(%<U7G?5DFDhKc3Ke4sh_Wp(yxBGP3wFQuUWAanjH_|GCx
z_1J7h50hA8(PO!RW@|cX`!P4EgIEzKTdmYr1RdDvfLPsr7vWO=bxk7auKTVEA$h1Y
zA1{_vm*|3lLCFq}&F1o|5Juojx-ne-zx{(lgTyFq4QG7}WQy;e_?-?1DIzn3NwBKq
zAYFIzY9Rroi`kIM#_1Z&)BT(fWx(&0&qunuR`{&>3jO>GBhYBMi{{1S*mNq|ig;0l
zz}Qnl_I=mEK+TBL-Sq&!pQv&QKC`(s?cgR{1nxZ56|0UofpJL2at|aWn*O=#*6IgP
zd#r&&4$B5aRL}u(jzr(vm*?L5wb0^0ME16la8mTW@pZrNzlq!H)HSh@9NqnCpi9F+
z9vIvirTXl3p~cv%ZI@LiEy$8EkA?5-f#FA~%amepEXOO~5oGPr?5OQ&&GkJO=_(Qm
zLJU*!XGGP>@AFTm^+&k^O%$vqIMfK*V6bm<!FUawrqX<PJpZbSaY%Z!OuoM&*}UQy
z`lKZm_36L=BbgZ^fE4X3F5Y<vSnfJmI)1wc0%9P8Uvjx?{D=~8j$ob;V{{yp8F>6W
zin|2?$zmw<oGt;_+-_g0T_&ke6qiq~zkfFDej+LWc6bf2Yjd&~XtK-N-egeh8HSO=
zKu&0r7R`M=r9@__<zK4VbRA-3w9vw9v~DoDc0r4D>$*}!;pX`yW8@i0;avAFR51a}
zM;eStyewL{&0O`Lzr=p68mAUC{0b@8psszWHhD$Ik&4?EzH{@@#?932x}NF(VHbVh
zVd8vP2HfMT2*8-~SNFbnjpgz&U@Z|N4u?@SpnH&3S<^v3uWdJORygHCBA83h*&r|*
zr`s0^T6wa9#eyR6My6>2fn*#UNPb^7GOr%#h!_HSy_>Dn!Q4L9z)|2sb<Y~LZyS-_
z+_Js(fDY4|A)#YD=74n4oHw+_M-Yo_DM0g}3Dmw-k6vT*2OOZHHy^9IagW~rFoLQo
zTy=mP=s;)$L~G<fIofQ0${aH0Y6CT*2QN@Qf_fa48VCjA*L{mEnUVJOo58nyii`w;
zeJ)eEg1QV*+5MOHd#}jLs$4gTO)6A9jzN2`PXApT-Jwsn<JaT3Lr%bS&6J3;{)(Lf
z9Uz`lunBa^_|!pZbsRiG-u@;Fr{Q6y%+b&Lr_^%}@TL5h$6mD_LgEh*x<-q&ghyks
z=eD7)>NOzB#GPY&DrN@KW76KNhSVi20U>WV>heJ)AT{C2JG4A{s?`v1nd-vBL95AT
zc~>iYfMJ<5_PKk)K(_(RRJRK9&xJw|Td?+ddS+jV*Yz=|&YWrOsy*lcn1g1e4b1?Z
zz^o-*O0=KAt3@<lkK+Df1QY{v4m6x!%Mm!cZ_y!M3aJLq0*8vs(!~o96f)_{m{<%*
zbf2ZM8_{vNMPI|T1$?^E2CY@s8<VUjZ-PqjK!Ni9l3HEZT0JMD4bA=GuY_yWEyGe&
zW_o1er7qUd&f;6eK%$@Z>A_2=;k^y0p!D^3W(9w|!O;D3Td`#dVi*<a5hyV{^pGpq
zyr30NcMOJ1=~u(HUsc!wmPxK?JTJx!gwxC$M#B`m9Nr?|uP%^L5YNTt2zE$)gn*qH
zSiM~WXjt)X=N)cqNw;T@Cw|0WUo_86(4+NH_ygS@S4tA{UG_Va9>E9H2Nwl|pHBC|
zn`Z}JA_y={&XfQXpQTHe2i6+$m3`)e|FXaPejMKknVF+P%&Xn`lzmz^*I4*#&UcA#
zv$(_zSzMsT>xBI_!y)T=6k|i8;G{!*MRM~_$Xcp(oJU3SgM(v@KV614(>d`9si+$D
z0$Bv-MdLM#BN&Dz?TNSNiPfKurNOO=uX11Ra0&EFi3TEv`7~%LsJ75L;1bc|-%YKK
zRG!<eN>%z<mvxsUIU!K7(!#{2Id@`N?gT=Hs0V$d<3XM-WY<&)4%!Swd!&r2a?js8
z1Zk{L9Oxw0z=Jv!(pl>!ZP!85P!TJy&aHF!Zc64aQr=gHzpo;^=2y>n)Jj%%SNiIa
z85pi<01uJmATv<qnB-%7c{zCLb0e)zdS>Ap-w%g;-<@IfIxdI7ts(yHua9bhlngca
z5{A}IaF|ea01c~C^WN*JOk=crZ~q_0-a0JGwQc*B0R%+4rG^0s9a6eu=n@cQM9CqP
zk`C$a25FF1kZwVc970+eQ91<%P#EgF#&zG%{k-q?J>U2I#agbl$;=h!c^=2H@4r2m
zSU{H;hcaan`b$AL5NA~EO1P-H?dAp`XEI{)lsbb%Ew+Gx&a_f)5NwoI<q{y(<k@S4
z?g@T)Y%R&5iP??10XBl!zy+W#puh-V4+BYjz{44j`qx5au8-owaC_=75nfiYH1hIH
z35-^wx%O)~5u`DJ30dIjfctE97~QAv^yY!kIept4;?_@+30yWJq;~=wY^E!ta)W^f
z=K{y3XK)k1{hKzOLL3nfT)0oEj5B34DJ+N<2kLNUuQjrVF36H8!p|3b)-n=yz#3(i
z=li>6i-i?4;G(9f#o9E*ja&szn+?`y%JWW7HRsEsP4UDMoqtFemWQ-Oq;5ivuxCov
zD25YFrk~kzCl^v1dzvLHfZxgc6av;MmE|%}mLzyh9-Qfaeg09h^9H!+bw08mx<Ej~
z@z<OUIdYZwm4CucC!IM<&wa$RLGlcmyYDHALccPk5b|`suj}QrlXd_^h=+ivV|Ir$
zgh6GXo=bTC@l6?Xick;#AT>D$bsbled6^!E?sLuhW6GyFtX@|@Of@A&X(8K=;~x=s
zQhdTXz|Gq;hLMDw=uN%&t{vz{s4L{@L1?&B63gfQSZ1rVjHFlge5^`l7e4ynd+o8=
zE)K6q>g|<#jU6jXlwfN{puHW)ALa9iIiy*GSw1gQ*pZ6Hk>h=ps@MBBh&|bod=eA8
zn%;_KTx|4`ph(!*NL!lNEB?bhJ^|dLu{?i(I9D&d8_ax}|IJsW8)&Mx#Y)8#r;xFs
zFvk=FgY65=IGG-}hh){{jNqI!xwKo{>Mz@!qNQcGf4@*AcLFhqyBKO(iG-O#Z1{Kl
z>TLoGJg$6?XTVLKeED(pNm@VWu?}Gff1A&QDo2K2pmv?ttCTFi{sgI;qJQlnxA3b6
zn63oHe;8w3-9wwZcHqPicPf|Bm*_H;Wf}M{|J`+USosPG{0+%FTW=HMK|O1#`#>bo
zLkTB_mu4)#flk+Mrx?d&`%fZAeBh9i%WY%c8;a{h%}F~Y4A7wHD-1g&fuz2NFVFzT
zAThXCgt~)n(H(H^*4YJzhAukmDw`{L23}SGTcDB^f~EO^?3smJR;UN{IxyPsH_Eq`
zqpQ8;$p~2f8x2o@*y#D^I8S@mXBhuEv7nU$medT*&G)(*cnMbwCsqys;4CpAv|9Eh
zBIXgaT7epI*8;feIOGz*i`lLXoQjls#IDa)yV<ksCZh1j$)YqP^b)#115)WGczf^|
z3gbEK6Ht-nbY@A2e=tQis!CxH3jRQb<ggI25G(TSbMBKtk|fN?0LZlrx5DfqMY;fH
z3sD0nlRvF9bnot>D0>Llo7(!tJgWey@b8c8H#)fh!@aDRgpLeCAMyaS9$3tV!X}Ha
zy#fqlR&0WO2iZuxZTuEi_>O|GoRhjVxLhAM0+AxtBa82WDV50PzyqcYF|zkqwqyl^
zO_c(+@Xm?hP+TM%Df=vL1Qf<FWr5F;APoj3G4DvR38PX#+dWY%YibbyP}@#0U*lZ!
z9ynZfD<oTf$qRV)vW*>2@ZawMy4{)+1@mCZ)Cl)A?1BfV@x>OGuY)i2m^tE8aM$A)
z)H8J54UI5(Qpar;#-jM^AXIM<^%d@PU6k3ZkL;a7>FW3Eo!BIT<{Vd)(WVJyoE{*A
ztnk>Fw(0)80s&aF<kWi#uH4pr$zHup{Bvq^zWsRNs9r&YHU7JkPwypX35Fm9r-e|q
z@_=ooX~XSOz|*8%q+YB%kF)%4)I6BX_x9SHIatb&aQLJD#^_xmj)=ndf}}ImS5yb&
zoixr+w^Zps$<lMM)8q`sRj<Tg1s&?2RNb+?2Ec7MH7RF+5*#WEvtsC$Co;^>FJ6+r
zjbO02?2mfTQ!hDoaQC4eDan4a7sq}OKeiXn3bQZA6_w{BCYE?vI{>rSZItKeeEHJ+
zUvJ^FTi?$3e|;`dnz+*cyYrPufrhnt+YaTkI*Nw*@({*)*&Tdwh2bOEt6ay$KC8xl
z%l;zaw~jeau7Qw*5PLrxo{jOo5WCQ~x#qGmoo!AjSW5&hFkzk}`WMvnotgN0aHhgx
zWeUT)^mxat8**EYv-rK)snVe>th=xiF^mXxp}^W9QZk-6Kr=^j)$F@sz&T5prp+o>
z{zDD#LD$%ZIRpXp#VTD)6d%dS7z1cae!N&$*b>VZ4~4tm)cmt%&#ObYos8CvzWAy@
z+q}-dD&Q_*-mM%iTg|QPLE((wQKg>#57q-$nywcAzWI>~vWzz>!2%2oqCzJkf*BFd
zdKh+mxrt$kQw6tk;{G(GldV^B4gz>FDIOd7a0F$8?8`q5Ds&B;)b|DFS}N#a@ZAXK
zW82NzH8M+~E<c)^BW^Nu0TN{2R8MQOJZA2`aC!-gp4X9gTK0d|HfVBip5<fnI?h>h
z>=e#DrOcIEfFnOtN4>s~I|I?23UW+I?chYgrn+<%e1R(H??z3_gw?Vd3mJtaN*CZ0
z>QKKbWJkSVMYK(H`At<WaM}v%Bv&_X(y?}&esr7@<=i#ZQTyC@k&P5ciq(T$x-F{H
z#7Ux%oG2A>#(|m<Ph*XQ3W>RNXssOOOQ;%rE{Ov6EC$l;7842KQ_P(S5md|2j}eF(
zqf~@O*Dg{usDhKemjqY%!oAc2sf+95A$twJ@@b2Z+6Zc&is7jBrQfgJAxWR5Mn%7C
zlMX<mV6>LSq6Y1(prUV25zXd5m_!HK%gsnOyueE&GMBGYzP;ltc1fYCy>#DgN*PUa
zATun$XZW~#3Dz3CBNn&H8BDzcNvFg<PnYocg`5oVQIdY|G4&#mVIc{!k~Uy}LI;_4
zHBDuzOeP`I<R(H6Ks=Qa8TINv79#6)3>Y;xQg8>S-TVagIZ!I$FE`IGy01&5qlRfM
z-$8?3@`Gg_d6*hM7Q*BD+%0w$kJPbjQmcCTry^A~vnfS-YSlbPTy9FU=v{h`9Qq*|
zt3@Mj{)B4XTU>zo;q46>T~Z~k&xQD{y$kNrT)vtwSI})Ld1@DLqmPh_wW9H?_Xdrq
z*&be0up{my4neO0R3gu-pBxi1>vGewg%ynkjj^q(B+IY7C7Bzm=oBxIH4MN1h@vyi
zCU*|RFhxJ#64_W^NHb7lDJ9w!Kn0ZT!)j&VB?rh%q?Nnldg{|g>!;71dO4l_Y?&60
zf=p~EvEap#^KHxx{O`s+l^Nk5MBY2-jKY3L%%7PF{N(J2uawQ2D!`Xxzj2%7BFc<=
z4~^H^uo&|ZIhAxS6E4XxR$?3Gtit^r?ts-EC>a1-J1xk?m7;WkZ%{&R3#T%~TW3N7
zM%3v&?d_6gi0QNHUzW3qoMgQxDL61Q{y<bvj82Msd(|b2pKmmI`?M^!(ST_yoBTJ!
zQ@{bmS^i3==sRb510%XSP<n$zTM|_0P*^)}yGwa>>s{;sJ4J+zgFv7p(eLADqt?KF
zyp=)g|F={R<Y&FL>R)7*9nv9b0Yz6wbecSiSO(g4c39m0JXbvA*M>i=Yl8*CLN%z%
zS?-qLCOd@4zy)e>&j6!@n4+UHdo2Dljmjv(Ka!nYPsk@or_kpX?@E?CMKX{B<T~2-
z$&#g2aG*N|xFPs0*@7$mz<k;0G~LAGU+)_!7Q0f+qxndCHqZT{eY)TZ2}Ss&0P4xN
z?ig&1Mugl~%9`Fy_Rzz(rM6>7lKyX(p(6^ph&FllEh^e_mXo1YgZp1%MOA3%8NlO1
zy1G?1c%S`|Ea8UYU`GR7%36))#CF=`3xZ%$8(kmkY^Si;Q>2I`R?_>5{nTkGcu$>p
z464H@K~a22JrzPn9tSPbPh(-z*lDOB@wm_^{By?Mqb@8JAd|(%Nyx{fFDo7VTHWa5
zo%cDc{rY5_#~yfJffAr*%TUQr!)YigV4wl{*~{jG)|wOX{K&CK_!W_Zolhc_&cllj
zo}47q%LvPlJBe|-tL#7TKI1@ka^h<SpFEuxvdgO5db^!h4muX!@Q8VvqRFpeZlb4Q
zQRRYcv|9{H|IUpBSr%yrra!%_iKL7?#A~U26%#aj2dfS^Qur(J=aMYU#1!~GF%w3f
zI{gs}F8Y@D_iSg{${MgZeA^MO)#QW*x~~HAlJu5Qa@33aOyXD@G}bw81}WcWb?1uD
zD)VNP1H6RAuZ**JcKzSa(dxFF6)Hl$57g#;WuChfrFzGlGm*bdtLzm6VGiE-6uAlS
zpMTv=WbbkxR^bSAG8)N}-S~C}fEvYKoEuFCZ{WFF1Kd?s9G9t(M~*KLzlN;@c8FYF
z?%buPk8YLTkdo|Xb|LDMoY!+Yo@opHZsQzHOm3I)NRKH24LsER(k?(hN6`yQU`50m
zMDmCxP^W1e1J)x}+t3z&SAy@qnt*5J5POsjyv&`h%BGO#=#}qkgi}ch;@FkJuTAdb
z`fh&pH~Dh6rDET$8IQtjWN~2$_`q_}=#7=Ar>{*y=~;)<Y|R?Kp17j%dGZnMea{mj
zXRY7zlSQBVEk49#gZ>0c_LnV{RO6Ks#k>crzxIGqe(UvH<?ls;_~d;-P-yY!f=;xa
zTyS1StNo|(b(|)ZbmIe#1m(|>MpDloxDhtv;3yDm+yAPJIzWntWQ6otQt~8A{j29F
z7<}uO)pnheYQ%1=!>_0OKxW<o)=X7%`aU(Si4%8ZxzT>it2dNQgv~FwT2H<*CpA=p
z1?E*=pr`3ZkLywN9%0S&fVsEx_npkn`Q|&)7(U-jKNstY<?~cf`)*zEcS@Gt=eJi2
z8nYJIpEp7ZBQ(rL|DZ>#rH|=czN$-Y8I+<tU(d-&!P1ht%JwCY`j8ky-=w7X`%F$h
zRwy+RC0@h7aU<0-Th4ly*qn`-UUS7<M!f0y&?zzXwPXl#!?fJ5pOiwq(Mml><ng|_
zuJ?W%zS;&m1x;olKGM4R@y3Nu(}Tm9lqxyI@asjVemB4o!rZ)D?E7N`*CS^M<Sc~}
zv{y6`*c&g|fn6`jEqGo3s_t+S(qxW(vjT{>8*Bx2)!=#Wa}HnRsf#E+5co4QOphjL
z9?Ca)t1q0@baJe9lp*dG)SRiJz>j&6Qn2@ln}i|e)P=7k`rF&7$3AMmGa2eGIuY{9
zynV%8&bZ;t<Z-aH$j<NZgXg(H@O-YVT1n^qdrp(qMdxn(ytVfj5R8qwC1sDUW6b}x
zZ*ZE64WDE7jfw4pzc-FAnN-#~)~F6sFad2#;um#&m$^sJ*%)2E%$=HbY~v&FZIbm9
zPPvm$F)hhX9~kut_Ag&$I`lYKEb2nDS4t1XLg-tvN>MYq_=IMus$g{7YUU8=6n{hC
zBWn={Bd+7VV=txPDc`~wQ-9h0ar@SqFT}>BI)`T5%(#-s?PpQ4)`Rb;<J)9g(<TG?
z0jU(f^y#n->wZ1u{bouo^?^&tRCnHHg5ZmnC!i@6@8vw({HF&ohaZ!3$nn%$uF}E2
z1|wMCmHourwY)c1hkbzQFMj+3S3xA=>E=D}ngeCKC(gps6typVS;uTP6_!I}*t=_M
zX%H-9)DJOLbM=oP!g-N${J*Wm1p38%#wYLWl5ir4yC5&$Qe_&TCrMhY%AX&qjIJqD
zJGFA3ZH(47_XbH|bHlzkV7&`yhI4LH*SmD$`_J!Ft=(=}DD)fsr{QWUiv#XgFbG!p
z#?5zoko+T;4hG8U5I--~GWJM7a_ys}s~&mLrjq7axkPTMYYA)gF|aWBa&2=-B{S<h
z!9efj97uwoSL;pUevfVsK1&%8N>I+c>ysCYPwLr<i(po<DrCDn^2Mh^a&rc;u$FYn
zgF^gwh<iHB!OTddc4TrJ-Z;fM0lZR0DI`Qe66&rWszY&&fOq5PrnD^D!n1;V!cg?*
zs7Q`neCGin=Pu@+0S8ewpRx<PXJS_;lau!)rZM}%X<@Z$2zLu9iF3oNWk+CF$hcRZ
zvw1*#k=0lYP($XFnBZGX!kH6)N-{(AnzRxkbx1RHueQVU+kl4R!evtHbo6s*c}3%x
zw8yW-#_Jd#4S5p<@tqJr&G~hsV)#Wkf#v|^6{p8VB3BsWf3|_UdX2pvC9PFle#@hO
z3)^%WGTBPAM^wIi^j+n`9s~1fQvDF$x!blIVA4P96QdF0NZH}ygqDnQ@MIB@$kOYL
z9AM5fzu|(sRog|mc4yjko>XXU{5~ej%0BDYIBWHLNnBxO2n`MPF{`lpfOpW6yg&D!
zcPt!?-v;n-*j__q;}G34$JP_<gs3oFac|*)8iO%zzNVi!l6>LC6(vTd$G3OswjvNY
zODCsxIIQsrP2QnqJFuunGh8)15JxJxv}Sv3+S5BZSm?ABADQ;NIL(xPwl&su>*vbC
z`tn<!faBf+A|>R!%`DIjv(sa_(&Y#GTN^sGHTkh}XXcZv_L=U}+*2<%d!oM(pp%RP
zuj8*jJQ%wv7&Wb<-}S4f5>`v{Lz7OEzX=t03SdN6Qf`?fRmQYXSRMh=`UNXXVT}Rx
zr{1mwaTWBszwLgRx0CsTEVTGfAea2Y0RzDBl7`9mMCZ}>E(tp;4B^%OYYnZnAJEOi
zG=5>i!uWO9L-X9e%xjRA!e6u&;}RGQLX%VESN|9jy8<hy+|vf2^*Y<iwaRcsx-MtZ
z2nY*LG_-RY-(TkZ`K0d_c9e9Ai{f!)WE?18PzwLJp!v4H88wczmCh?TuJqC*=BH{H
ztNO{sYYEZc6oanvb_lMW4`XJjAGr==+1CR7CU!>uS+4d;X&WNtvCMgXefV6M6SMKa
zfFgK8xoAve#<5uB##e`Wzkog5KHjG3*>fNdp%9A7h{oaiR*2)_aut>(Iy`>>^kj4f
z;~M`=MBZk8<O_W#^FrV$K@W!+r!nyV1GPlwM-b0V6xJlN`HutuTP+Ofosa<{SNDsy
zb{<sII#O&lqf+lDaq60lZV7hwS*F2DHeJ$J)4+_1eB0+0GXI*v@zH`#x@#G}9p5Kg
zu-{8;_?A10y2Z2WH1qUsX8wZV>{r;QilcP(^p#JVIWvr-(c1}K0<RK_JX=uDqsY5&
zd4()rUB-5Oi%-!bds>#hc0i3_?3yrI=czgS1?CYr_9$M9dX^s9{7D0==}GfkuVPHf
z_O|na&frhZUOy92_6i^+B08Aj^FvFU{?p4KYR`fS?v1ziE@F4-5l!$qD@E>z3_^{;
z7k3iMLTfk$%%{w=QhQ6KPv7@_Hj2SNzk5#6w7diosK!swF&Ld^vV|}Y6jwp<8hc;3
z2N7~{=`8Fh>$yR0zUS{t`g_bSW9BavZHg!c;i$nsB@<t*DBOZ#V!AB1!B^if$sF{>
z$T$CH0a(_s2{8S>GrXH#Cm;h2W52kEcI3)7@^42Un?DPanhAO{Rq(C4iau=EdUufW
zs@ZW{xrEmJlCIZN+Ro%r)Ng)QENfmZOOs^ZX@Bv*mCq<ExO{D6!nh}14VvVa&0hqK
zK#(Q>qV}y4M#B)dHqa0W*hcbJ_a|JqLHvTlC;O?A!{#1xb(sC#7zli}ExLsS?O`u3
zNE7HK3E5ka@Cg^R1GWVAgSF8^`akiI=RH80%lC;8t!F)jjr}`o*fB7N`er9fWZRRT
z7*5-@4JF;Q(gGxd4o-gA*FW~l+_`(eZkk4e>SjI|(Dgd1dFe#lHyq2`x&=|MLh1^K
zLnpDQ9tqmzs2s87BtZ?8$jS<Y`R7}sppT8N$W@Wku4DII3EhS)bW!8n@p?|$-%IVj
zci+?3@8m8+(iGdLd%WHrdPeO{OxC8sA)i(1v&{^HMxxYosoGyuO^!s$rrr6NQ|PZ|
zFDNr%GR{2yyBx$5y}cZP-A+83M1qz=9rcNP=8vM(KW%+cMUS@|!?kq6e^OlWF~<ir
zpwlC>`w$+^7JFh3S#Z@HZFWe)OYGC2>1#`_e`R5(>^z1$kv0SphtaXucs;xlKGS25
zv0lp)_W<U&20O6xTw&vTR$E$hH!$f~VAXAHxySw@hOoCYJ@fOIH!&_#%qx)g7}kp2
zBo4A$AN_15-+uE*l!_kL?D5+1__U3({5pvS?L32*{b!NT3ELK}5*k+ml&}uA+V4jT
zq{bXiCzLnOc_s-4q6<!EPUb4!kab9&u2nT$0d}!^#SM<XgX3W^jwvEu5Ln{|?u8N1
z%rEI@#ubfmwx1tFoBJrUM{)6~BFLUe@Tr9kon%++5W&t?rKp`%%iZaVZ1xP5&MX3-
zNo?6cfFXFN@o$go^G)TNAFPBWdQ|o5eGy*Juu}?J8#t$u*d(`2X(^d#LvrbY#eePk
z_&DhMmvRtUx-VzQ=1K0cs=5T;Ayvls`}YH(M6N(%1rh0i2<c$&J5CvKWpH)627qC~
z_L8zC=?|Qh)NK7y9%{2z+0eN7wd{)wW?M_`!r7n3W*#bE&eJ(8C)4ArD<@gkFH~~B
zI41~6#HBU_529gyUO$#KlV0DxZZ-~3K^6=EqHO9R{Ky`z=rIH8LiWo6Z#XPcpQ@}0
zm@VcN7dL+?iS%LXVvZIeB`gP>!amh$l4Z+GKFL9@GW=^}Z^5cXaSFF<+$SiF6vqlH
z1Pc<$pCt>?lpnX_r`MCfy=oNqYmaINK8HEE1A`;mtHeHFmaI3IAOxjr=gKJ1XfYc0
zmkH@G>(A4P+3fx;RA|Gj2n`p@8~kjQesV7H&wYA=Gwa%!oz4FVny(jjc#E$1)vqsM
z-WwZs&#kfX=;snUSl;k6!x4#!-VQzr(@DibS(|dMX4n*~B1K;C%v#OWt#f$x*KUBz
zMfzvS!(<Y#j!um_&3w*>5<+&&0S~URR<%u})a79Q;RzgJGw$Li-tD@JL$;BN;zOL9
zk+tN--ZkyN6u;sh4T}+A42GI0dnrYAoPTwMo+I5+-lnz?JA+;7u#Xg6BSm^ausexi
z0^t?CZ7JSpU>MbTf=#QjMZJ^OU$j<e#8kS%wEK+Io<*jHz2B;+R+jPo9nxz<4+R<3
zJXXiY)7p=fwr+!m-qCY5?n{0a-_Q2cZBMeDHK5nDPZiAfkUhZpQUjHBhcMeI!Df`0
zmC@TG@86=LOuYhbDYaR?ixl0e({0@@^Z)`TIES2sMMCSpiqo^vqxC0@lnfuLFFY|(
zM?gLk&LvAX+I_8wI#l)zJ{wuQQ^-@x@bbJwgOg-1(NWga8v1}RAud9}07^7Dbui>z
z_{~jNS?~LBH>&ft+zwyJYxrz~BVS*jgtstCJ}2y!3Vht3|9ef#V}l!18T*FED`tJb
z3|s3y<Viej`Obl!vC#n-TCNT%)0M2Y88pQvxx40~(r0(-Z1ASu$wgFIinG8o1h2$l
z^HY9al@F#;@jk4a?*6^`8Xh=SY(L|3Ug6s~3q6HPbJEW0h(?5V3?f|%V-CY#_jcty
zUvPi(jeSvo#^IAztMu%TuWVB+C%-93@6dM6ee4Cf!%ZPgfVh6PGQ)xJv*iI==%3kZ
z;CK8J^)uEI3T?@n(iKX1Mu{sO<HB3KxAA|ST)pJ*Jd#%9plovP)oc5P^rBex;%5ow
z87cIf2d4O%#$~$HC60&Cs4hA7Puokz$`$<8Z|*@jqUOwjqbG;JdaubGNd1U$1dnBm
za#<NG3>B>^O{i^XD{H*fD8bw-9ZkgfaNW3&)^b5?-K1eWUzJ>ILHR;bjv)H}_5)lo
zw;$ci*VU=LWd#1nHp3sHzzk>sx|v8d2*MR+Bd&0x<mMGljE9{mY$%Euu~8)vQ(tkq
zP>L#oob=p+4vP3}B(x8{iA`LuTlMdV?AI8{UiVFircs~?f|Fh+bZ_HnNu5sl6*#^r
zi4>O$ahn)R%ilvJT&wY{YQRS~${mi$RXm3A$}2X7=VsP?1=GZ8@cF5*ilM%zfbZn7
zDmj>ub7s40t*e|B4m4;i2bWgD(JYroBYdl)oPEb;7nwRhFE_0_Tiep)kp}vb@$Ut<
zdH@>cN=U6q(G`B%^?`!E_mh2v{L|=5JtcfZBDENmlxI^wugGU?mLR7`Bt`sWVhQ`v
zX1^9&*!?%0`HsalAd0?SfP2NhCt4u>k=t?N8c_sGQE~G->@ErcI4?BnoZdAC`dOyZ
zWiVF=PbJMOpsdRW%UrMNaj6!mBP3UrukbtN=4e7m!VFikTvz@&|NC=Vwoy$f!pOTr
z8VRWC{M(6CFF^Lq^s2nRd7q>FK(?+YVqW>03bV4xEKL>X9~+n57Q_$`VL>WYmU@p3
zH|IHfUlMx1LC`Vd4k;`d7V<T3RZ3;lFgpm}ppkkS#&9l@@pQ~Um8J*bTK`_AbV=6v
zOF%zcShhwFZCHcMiC;qB@ngj^3$3_HmC8O=98`fc#pO|%YzL8<KF{v4jZp`k%V6K*
zl|sMCO`Q!(Ii3hjkvPAHr+QmZE)slyQW8B=Chg}ymrtoxP^W+DAdT$WrVz-ijkJg4
zFVeCO$}b6L_KRsOIed`p%ZX<Qce<`zP#T($j_~7Y&qLI+jn=BJQdN0|dI=Wnr#38~
zIDB(@?3Mq3spA($;RAG;@}0Vv?_h=Sq|B-&64CMG(8Ld<csl%ykCT!TdGp(jmAvS<
zmX)G=5!fO!pg!hv5xoIyMJ$ULmJoj+^H0(t=$;CY6$$DzSI#{11E?8w7wLYygq(Lg
znQ1Kdg4hlSu&IZ^azB#ZsgWwaUh~z^s7nqH-Pa+-je6OS5F3m)X7?Bn*|zwx?v<i#
zC^k40eOa5(!$yWh{H~^q_QAjFz1KM2tO3s9j58<Qm*aQ}WvZL{6fWjU<AJ+)eV9eR
z+Mb~xUJ2_~B33Q4gY;0!eoQQ&SW9TpHEm-({H5UhH0nZ&GQ^O^qJlL?^UsxAzl(bH
zb)aI`-6UMKBG=rwrRQi<DuO>Xq&zIFq$W&{AJp2Cf=GZI+C{qn?oU6R?6H0!8B^%*
zuvav_lkH!}jq;(?R>Y;-R%9C7rSRgAk&M^nHv3XT&%|{wMc3y?>DAn98g4soY3D7>
zsEW<T=5ADHk+D?N>y6!#m{Z@pXXt-*I$7?HV~Yv^XhHGy?DuWF6TG|KcJ2$<@T?YF
z?=aO&DPHbXkEE^`ew5{p2}=Hbr?I(~R`CK%!fD?b>D<4MPe@m<nZnNbW3DVvm#Rs>
zH)ptQ8758s^8horL`Pl{%_S;gQzw+=)f4UzMEfev)j8yVnL}FlFHLHuF5<~)6@?sP
zW)G8j`E*C$FEoP+7pVb8!h)*tC4`XJ%V<p`L~t>SrfTzO^JwgLXI8A}FgNNtzyo|w
zfTw!)*<H;kf|ZKo+uy4eOaet*$kF%Qj}=lzF7EV=uExuJSJ`bod`B{@FzHKV!5OW4
zLU1{*ggRMUc0h}`s6(~ob^8Qt6H|tKsb$l(ElzK(d-G4Fa?$EdTxKeE*0u<Z1`rZP
zw4;a3`j}BuJ)YlTCVEZ2ltb3($3Qc<yJ&9t?{LeMo%J;Uf0<uw9WIbSND0rMN>s_k
z$^ZVrOQxSN0H@riPN$cj0(@lB<jfP<9irL_CjNkVIJE?OF6tbN(!VoImJ!Y8I7H)=
z+|YIM{F_s{Pf;&I_~Dws>uVx>Eo4M+$aZUfjHE^ewa5#W8?*M9bNQNVm*+dcS6A(!
zc#bwm#lzv=Pznx_15WD4{3^Lw-f8EeYw9Pr0XcCF+5GJD!Fdvc;p*YEUD2rJ(8K=A
z&pUL}o#_WBkmM3iV7|15)o!%291)FHTn>$+kb0b|72O(doZv;)Q4ZP;svhopUqDsQ
zzQXDER!Hcis@YHzJ+tyC**iBmQiXN;)TVHY@C$$FgDz%osA~>pJmy`$sv?8Co!tW2
zZD<Fr-VZU~$eAK?b(>OGQL_vb<J;|3^rhOCC&~L`vKHCQ{%ys4$AfNnV>rXYf9n;A
z*cz!3o@p#iyFf^%!<B#6%XHz)@9Ux(OH7JhF|5pD#H*Jtnqt$6+lhaYz-e-X5YMJ7
zXY&K&5b1nnl6=D=|JnsgutD08MDoY?2+28vp0CTZ)r)?E<L!qLJd@<S3Ja+OyUfYq
z)aFN)NU;VkR5}%e@?Ru>OP~1DQWjc~ADRt*S9KZ#oec%7n=&>|(kkio2kKv#Vbygq
zWP4ozrqt%Z#!SFQfVAj20b~i6OD+f}7fkV0u1lMgi!XwKwtkF>&9mvVT&oaQLz^0y
zD02*F;=Q8H8-q`=LF_bg$(vw#SVCw%*jaJcm&ja>uv!je1GXoZ55~upbHEFth6Ph6
z8I1~`D+E1H$ftrW5@y1lC~?b8PvBXRPX)UR;=;oTG6ivoiVP~}y%6gP*A4h|^|SPB
zMR<wK4_8m={~l>XL02r`r?_%=j;GT<EG5bj{+3I2i4OQ|5H$(bHF+6wvg8~~UfFT&
z{U-d#Fp8VrrT&qi&A2#*DJX~IMvoVr$d{%*N3+J>-6z7e^RenJipz#y+78Ipkt>P-
zTdI%6?SXgY<a!Ft1sg_oV>W#Zi{8-t(_P@tQ{96h2_VXIUTBVD9a6dXq4ur0acvxy
zTBKYM8~Inxq4+{_5}DYZ0rxGgG4e`q6T7fdiAR>W+7Y6bV0+48IltFCSv#R7fiXc9
zCV5wP2dD2fB3?2AhZ<II^Nr&d%O{&hFm(gf(O|q;TYmNz52VzN=!y;-4%EHK%L<XS
zOF`7ZLGIt~<reAmh<2+JlCkS0@UB5t5>Gq99tt0~Y)I+}QM~I82?b(y0GUTFYH9ir
z1R5I6`&ksp9p3wJ-A%pER_e(yUh@D^*+A0NeJ-qEe9kn#9V|BmiO<25!3HJ?ipt~C
zFle24^kp%wP&sqG=9=xA3c~O(u@*Y!7yT~M2h6OVNMAqDhsHw~&f_?<xCQy$@p{Il
zV#{nPfEm)}X1>}~^tmlX6ZjArYB$yu(4aj}a#cw!^HGac&-Q3)P@}RTh}v;sgSy^n
z<MzDI)bE#QW`rX;S2ALcZY@ZpV+zf^p4C9eO^nwfSSuZWpeKA8D7O4y1UIvcf{U!X
zB=AqxR#iQzg}n*Rifo>eiGs<18=7{1xTyv5=cW3I9GBqI44<~<vFrrPAMe!T)s!&y
z__en_uf}Iyn=a-Xk1gwNJgcr-KA}2!Q9ox+YfxgEU3B$m)6!2f2j*Y~Exci~!D5fV
zaqxYBQk4<t<pT!8A0Fk3%*{-_WL_)OO8@T?L@Njg`4Y02Zk&Fub!RJWJ#*T15s%sD
zb?c%NtZ`C_A&UgEJ3d1(X8q{bqp)`#R(n-<kO#vZ_3sxD?5=pj1D?!R;%NIdTKMl@
z5uhb3ar-`ZxNYslutNQI*G4fCIZUTQ?_{Hw@{yU6^+Ohk>*?@KdLh{MEpB2**Sm10
zS1zfw7nU=fa`cxPBfOtv-p_Ofpi;hs62FiFyzp4qwqlUh6svbyIbeA&*=p}@@o3?>
z6u)4%+iP78ZgpoXe~)pT8@3Ndw_VN@h?TXh?UfDAE3}XlDza6OS|jD8F)#kn8KwTX
zH_!7A%tm;JewH{w-;_B>aA-})-lT!N)yY)sUc`wB*cw=3NhL0-8~MYQ=ehHZ#>4cp
zT!waxcgNXRY!rVVFmG`w`qUiMNQ>pjPpH4!WX+F<DJy!}Bs-5ODSR5aOcvv}Oa+Nk
z_TktYG_;lOS}JcGZ5h(4!w|ckuw2TZ2$)=tWBKG1*@=0@W&xh*wZ1-YM%9G&n8`-q
zQOIX?oat%HJ_*;1t%!!2N;h9!zaqc4P7)?VO1!=I&H&{nlg%n5P`^?}WL$}Yse5X_
zWYA*qT!l8~!mu>fpg4y=2(A4II(eMd{uWUWbntLbr0e~sD-b>T7DR=6SGBgmgV>f%
zXaCebRv?Tj-4JS;bQ{Yw0jw_6cxzNbV~}j@Ho;34{`PJ}sG`^Y@0u^cp?)ERIE*3p
zu1T-cXUR$3Y4aaHtM+_M`3rGJ64S)8ry#(&jr#B+YP-%o#0Lp$j0mh?$7xxyl~T_H
ztMa`Y#HJuuiSG?XUuKARkg1Wqwj@BNfc2e@817)fLu!=42zJFl_aHhh3OU115Y5>Q
zrszke>;ICX=Szd2V+!7;peNMn6_14VH7mCISzP$pYdT|tm=&2Un0m<A&qzzaNl$Pi
zZq4^P65_QH{v^dEOjb-`B0bK?CVRZ4P~a$lQK8|%Bvl+Zy<2f9Llc<4ltB5c6sF{)
zWtFhh*KScd#VkPW^4Y>;1`e`Cb*Y!DXE^oNZmU?D23OpAAKgn!ZC<aw5OHo9B10Yh
zxbr^BqDe3S&#inWeWZu!K8@N76-U*?8E%JPED|+l*cs8{YErDc$X9E@&4!0Tv`G)k
z67~`IKX7{s8_9}mVCmi=R5MjXTDF~NP$&zRSC{pH2<#Q6KSB4M<1GZJy-dmO|CEaQ
zo%qlm{0$tg$E!+$qT8pE)CqVJ+Ip@6Xc36&{BVnj+$bhkyh<<6q<flbaf4GT0hQ%}
z$hQ%_XFCip2~EGOu=hNf=*tceyIrJCEw58C@$|8hSN>A&!xeVg-xTSivTZzs(C@Oh
zO5$oC%fXSlrncj0u8*JURMIFci6^d<Cx?}w%b!%OwGM6w)>EMG^#Pz~>Nn}Dl$$R!
z{|F^)rHhb*zt^Noe{h2QG?5f=7&l>|ILtz^j7igV81|IL)spH_hoy4aP>`yNmB_v^
zef0dE1z4jlFt7$3YV5mFy-H$lHha`I0BE5vRA`CbWT$+#nmxIA^Y4DoSZraqB|u8t
z!s3q=V?x1O4TfcEnOxv1zA+zTc9G&+gtIDL>D`jlbnP_nqln0cOI5H}1rAWe0+h9E
zttJt?md<y*Pm99CXTzU=IYu(oGFW9TRJCl}n^ee;-+acDiBx{3k@iXSLW{{CO+=uW
z7U#W#KXJ@Zx*8;aB97joWiWMkb~lC08~3gV1E8MlUHEt`?gk;#F2mLb?t!Opj%HQs
zoLYk4T|K7c^ms|_)$3S0O?rg%U@Mt76a3bmiUDsIvx~__i+;}s_IWRU9Xs(5(I~%U
zFTAzN51qF;g{=f=l;Wqk`5a~YyZWqFw1L)*?(^*ko|vzdCx0fn*Sz1SxNiW;vJC9O
z?g0BJZq`d3<}oTc&|bJ?|2!*tu)sKhXei+k7w@!5WHLYD3qA{G3tI2458iW>U!d#$
z*+4y*G~>g4&zE5rueNOdSvaq{^`yOPdg~I*OG93I@Q|^;LEOInLAx8i|C9p$%<H>`
zZ$tszp1Mf#6JLoA=@`j`j)RVW=4o8`f?8+TwCpyViFgF~rUM0x6xh@cfPaMcDthCH
zN8odcoP7;qCCtM=h6?Z{{*3QzQpFC8<O?!oLR|pv8D>Su?dTgQ4;8;k8tC|sDn0cp
zl<d4xdvf??4POr5Xh+`72)`O_GB35$fW*1ROpS5y>0Oe4ot3_VJ6FW-%)*{O=*v_g
z?A`5kK|A+Z^%?Xv!*B|gd6F2S4xC5QzrZq%qCDsv=^mYbZ9v3E?ZyyjhMV=fOaunS
z9Yt9*-6BTfSrXEmzgeS!(%n3M9*Bj)^^aM)`jOH9bkpMljaMd|Er?AGzBzMFO#3(w
zO>i~My`tvKOyT}|OC-TB#i@u&Qq{4G-0Ncu^3H~27T))PJVh-QO|B;cuebejSUr>Q
zCVD+oL{7m2PMlPFKrP0pH-9KK=$+7N$iXc3leY<CXdpvDt=|vK{eYy3{k8|b!O!vZ
z&{}BajXSEJOY;VHZTUxPTjY*1Tt|X<d>`xQcZARO(#pyV*1gY}2=^#wb=!gE=9*28
zA^a8F3%E-@hn?R!22+mK&()40Oce5Pbgs=38{n>?5p%INCTY1Pl{!qOi92*|x1ePt
z?K?2FZax9UdmOq9mKJx{lO5A{OGh=1O1yJc4mL%HJ+<2HYOClPeNU1*j7N*a$9Bbj
z#n)DCZ(dx-{FOT&-vQZ8SSa+5D>K`#k@r2I-aWUVFisswjl~7<aPQ&XfT!;?_Ac&R
z{AMCnB%ZChA@T60s(!P4k9FIs)HUtJc$I|?!3xV-_2Qj3QFx<b&#hN)BVqvR$egfe
zqI<=)6C(%|LX8ZGIrXxEFefVkyZP4{JkI15NvZz)fn_^^FXYne2F0a6nFJmvk+pu+
z0s5_xo&p3RGSxXDJ~4G#vYzI3h^E>S``RON5`+3}^D^}_+k%|Um*x~)#_ET)0YWVT
z>g(snq#9e+EPBar{XkhhED6P9Fb1wlSB5akRG^v}@zyVXYkthq!;Qaf7Y4j^{Gan#
z13m9<mp%<>8{SY<$>pl%q8|!0#m~snoZTYW{*&IpziF<)nxm(9MsMwPGI_whHyEec
zG;X%zP7%_fid2p}5Z#642xo4x+`I@s47Zb~1MD@CR+j|2Li-(R1_K>Nmr~H5j@j!v
z!@+rfZF*z)pq*LCf5~S4^+Ur2rzkzP7ENkd{4#U1CwZ9{R7*)y%(ngKC{dE*K2Hf{
zVa0qbn_v7CS^6)0q?#5bi|8vSkyqn(t=9dM2mi5bvC7NR6Bj^pVJ)OF5O&2D_85Q|
zg&D-4qUY~YHGGkD0Kn%u^obAhC^8^XizZ$s^!f#hrb-i!sbT36kPAh4cpStiMT&f#
zrEdclVh;1_0NUqyNPliI6+~fK=q4w*k`aUEx<ZeTdNA~|p*JrNSzI+>b>3jy(Ew)!
z@F&cBr-uE$+!6g4Oy?=bibemU!sdJ;Wu57fg6APhWUmT@^;52M0GoqBfBGbq0)?CW
zbn!>P!1qvTS^VvKeWhF7sp)Tv1XOyxPfWR1#2&rkT6FD?w=8q#dN|+8G<9@xKZrZ-
z+_5(_E>t^kN|eUA3W(5W92Eq%#~z5cqJ3Gdv|h+|oL{JIZ@#9IPKfgEqv)&0M{^&J
zfnvIkdqNEZla79SA<g`>9XN7%bD^2~*ilL*`I?(L({1x87+u5^(>iw_x~wrPl){T3
z5`v?M%EasAfnP-f&JUxagNKBWs;ATHxyt}4v#<isz3l<$B?frCP?>DC(_CTN_+nJM
z56*OT^2zJj@v>%<&I(hB!wh%4>VR_G{(n6#-0IcS#5S8(VQhtwJ`PpLRc*>S-l!9L
z>(Sh$Iug!{5im;j`dV~|RXhB~9iuPI=fau2`uSf*zQ4{y(@aR`+P&Us0D!?{Iq-!#
ztmLJE<Qz8SadABzU*@xTiJw)}KG!J8fB@*PbVy3CF*U=#6<dEFy~m*ei}&t1$x#Ou
zE-om;$Kn#yKLjnZ-^<W4`166OrUi23%6ch>ObV?)dX$4;ci>NJHLWa1Wpz_<OV0ow
zkJO;JUCOD~8w~@?Rp4sSSZY43<;$wvbP`OBuTPic3X#EQBUJ!g3q}r>$9D01A|WJj
z+W5_E7p)mbfG>1fdlu)>R=C>2z5|o#Nd|N&?A*+&A1LBwuA9Fpd<Nw8H=hgJCUM~)
zA@Z{4CZ+23{o)}VCG%ng-hgo{CLfpJypP^7hLq~~78c!vZOvD?X-otEqf83n>1Ar%
zvgAYCmr9vm9fir1nV37mSrpfuV#<vZ!yJISj_9LC`=dJ{KjZ&cu0I^4my-l?T1B~u
zNSoInqJsi<ZNS8FJ~uH`nf3i(f&Id%YtYD(%I=TXZ^?ALDtW#@eIGLR|GmjvIlUIi
z_JX}q8cUT%Ovy>{Az1xR(~+(Af}_f4U{QEGWP@=mrYPF%CjSA%J%h0v+n;(@t3x`R
z(v<m0{QmZ1WhwAQe^wDtN)i7vzWB$&QG$;JCC`Vd`|nj+J*|5l)03<%EmnYk%;!R8
z`pzD(lr*BOS?rH`X&k@dx+@XaeySx_b6%$hFwEU&0R3^0vJ~K?#1#kIE*@i)2{-aT
z5|x=E0m(fqWtSto=ii=CjDNHeUlPcMP<k%gs5JdHx~_86!pkx8ej4Ad3I;)^5rT*0
zHPlL^S4^I%@e8>xt0H5G(^Vg-T*<QvBs(!jijc%hqirM)JF&kCOasY;pDzs-9Fcy8
zc)>s10nk$YFS8ak9_m$qHAPeN01$qztpd=11~|pq0wy(f^=~+UHDpxwveS75<Mbhp
zHf_5_K%9+|K_)>}HfPRa*V{fE0v=-sPxFtUFx;0bpY*3oKdjK6h{M>RHwpn?x=y72
zELv<@9wB>+*YRqXA<<Csn%vaU`@7QwP%Qyowgba*=(Btr!ogc0PU@M?I*qqFO_J>Q
znNz-q?!X?rK~oQc%)bHg4m8oz2h{;#Le9{biBj3ALF}{(&O(gAKjNX~DKmuSig@{a
zwqM%nyjZJb3?{+B^E!l7lTik6!?7N7PbA;Sak4OCNM2~4UF5OhgBJjQ34#_v%6a9#
zQ1JT>WP8U7dIE~<ji!AM3a7)R6f5LwOk~5Z<creL9jZS4Dq3rs?kXTKT!RLdiP=pW
z*HCv?Nl)XP%3==jhz(hbzw?ibg%`_S1$MH(ZCCmY0H`O2K@P2>Wa=-baG}4}ukL2;
zgM(M7u+2G71HHUz4)A1JsvE@<JG2)K&w_o@;yH>U|Aq!kNZUSNPek8L9n>E2ePfNQ
zKQeyzUiew5gNP1OazZ@AuG6>3l6lVB`}^!w+9x^n*vOgl(^~0=-!!4p$6COtu{0cp
zUI6H;np&X=SAdbV{)xfIha^P;bWgpC4le$2ChLF?*&zBHyZ_h9bq@?)*G%ZDN!8{0
z+66>=Q`Nr_8_pDZr~+SZ-ITFn0!947pw+tp95#KCAb~T*EC018({z}&!o8wB2Cu$`
z^7#pt-gw`K+Tju5W&j_V!Ry{7HZa+c4RHT}tG}C7T41k*ry&{pucIdL<8j<A=Wc*p
zMuo6^$Tvzs@DP(cvrckzu>-fxp@NkG3p-AW#$9!u76d6>?-~h$l&lrgcckU%xI}@Q
zBIT%)f+D2skCnK)ZriifT|i#UUFzaQf(cP_(M@uMt#Z_Ld(tk6fQqY8P!r;g1Ggir
z-J*dB6GEKlT$ayAJc1UfEId0`9=z$$l7<NoK-_Q=>AC@RDAii4V1y+32w(?01PJ0i
z6JV5bg1%3;JWd~Ymk021yW<bP&?SA(g;A}^84v)XyeI}Lgknh{6y|rwnPcv^oj8_b
z;*&i}sA52Ut?D0TCrW;Hq^}FcAOCR?le_><vK#|>Kjij#iohS@@YHz<yG;m@g|`gd
zulvfe3$(QtfoC5(T`yb_zZR`Trze6xbp*8Yr6qx~K@Juq`^=I>005KSchB?xN4{oh
z!{7$-&2rbc0k6CwB^cSk-A}NEG{dFxW1}PJ5HAHI9$Wi9_BL5t1q-|;B^#SRQhWH3
z(ijNn!b_CiBbyiASBK@vN);`Q-<Cm-$KSR$&>RmD-<MQ`H1)`>iI(>}lQ~>HX?+`a
zVlmZj<k<dg-W%v=*1sg{=>ptXA74y54wApf=Ov@-L19@HT><J<!g<pyV|&Kk9=RKp
zPer2~q%#0A<I=!CRT|=4dus81sc84y>G|mnkm%Kb;^3|2ya!O)aLb?mI1LkNdzT7J
z-0A|XZ{!GW{)j{C({i`tWtq<xO9!{%XLaIA#$SxsRLb9;%uJO7e!Pbf!*&*i*Pro&
zegc%`_|yF<33vffttdV0Uaen(^AbkLXJ;u{JW~2OPaHDgB-^3bQAYGpZkoFr`0e(2
zq*&7{2@5~Lkd{;uw;aDz$Ct-$`J);ZjJ&bsk}wBBU+|UqIb^@ZFTFmBAw%IQ`jZ3w
zN4=&jXuQ)#A^u+T4ouJfr0sa54z0bU9q7&x)D*i^A5OVd*D3w>ILQt&@qWMta)}Gy
z)2K>k==78~DjkWkvF(3uU@y<4!p!i0yq-ToI&|`Poc_wG{w-jG;28SInx%m#Y~kwF
z1AG*=B#)S}C`UP~qp|KkjAqj#1EXKklyY4}FDvM9+fw%q{j-hz9~25HQd6@O#b+&F
z2mwp&n=PUK;3v2*Lp9ByY4Tt>rtn*w(Z0X&R5yXMOCzts*@DZcRBV1?5@hJP+;l+9
z$$gqKrRiWOGFyTABB%m83^-4Ys=P_qp54=<J|jLthQ9U$NH<H~IFQgq;oBReq@au@
zN`)MsYEr1}A0<-tBNP$7P;H9MTW~-&5EmJ!xI+mRk_*z1n|=*KKwj^01XMw@-%8*4
zg44x<m9z%fmyIHgUg9V`hiP|;u)JRXOAjD5(9Ms-Bm&ea{&tOJrAH#Zj3c3LJ2)W}
zEf<eKno-Q4hwE~_5ZOB+uv=fCxWWSRJ*w{_QRyK*MVB#!T1AeNzd4VIgl9W<{s3Eu
zwK6g-L6nmQsa1b0K+fn(?yJXSBT>%+a*`bhV8{kt;j9zzKL`b0f?h1Ay>gSQSK|BL
zA!9~L(}hIH4=H4zgSKNW)^^>e;H)uXl}Dbk(yJcuyr&$*_;#eoiQ9u8G{zjjt>cbQ
zL!G`ojOaRf>&4WmO3fyL92Z$~O^%f#ed7YM8KgMHLy@kMtSPUQFtaQtoPhCa2IyAY
z;Aa43f*J93a)<NaU3^dv0RfqS1e+}Wb_yNK=~eBb^*VUB#GUc#fg5#_uo*!1P*U-m
zpUkzoqG^2!f;plzy=hC2V|`a2kn{=niT1VD-cl7w92@#oYU{ROlhe9{@QQiFOfcli
z@HEHfLu_s258&nWDZeW6NoCxw1OH~9i?)ucC7=g6%Y%0rPQp*Va8X)*1Js|Cf0_qb
zeV60^F5@vEERM@qkbeB+e>=SVkAx1|TUbaZ=o1CQp^wL|(V!Kq=@>)7b&(<*d`UwU
z2ov&d<k&5KIi6P%1*<Q<TYS-XnB4PP=^D&i<iAG{2DBB%1O9mNnCz`Hu7ucOCDuHc
z8o)cM180MJVj*HU5EZR33pqN7GE*nkL%}N?lwlabwkiLG8N90dL(*cfdr@)TS-4k8
z1h$jxEP_m$#PS?E-2$}o)^eVF-_Ghq{h$-Bb<%iO@ja#Zk(7y~K`I~$w#asa{7iYB
z&nVv+FqS1u9wc>zYhMzxsFd~NN|N$u<RasQ;YjDVR~ex6!wYg$^I&qEJE(2VlTrR$
zR^FrB_Cr<%$$aI=zkmK|`wySf<J$=zD6qJD^(9fhgT)mZp6=PFA1ZQ^+56sR-|BA=
z6gDLwu`36;MWFJlte{1bv_sW{Gv&}N?XsB^*5ys9LB7~RLq`pkgA1<kY|p#jBW*~q
zV$p+5$44QKI_@XsQ#F)8rBlRG{IHlqMm}iI!@?m)_*s*K2%KqkSh>qD+pXZeuv&(0
z6Plp6G3on&CB-}&N(pf4yVSnQsF)9t<QqXDk9Ys+MA_T}?yXR4S^wsL3flkd-Tzvd
z{@agZ22k_;9<_UfLjbKNuq7*dQ}*y~6Uz^`0wQ9Wm-tv!Pl>SR9t#>X5n+i17aWQ6
z!$Y^%`Qht7-vcA$zzx#ZT_!Gu-Wva3zv4gtjV~VZ8W<pl0`Q`V#X>*8D=et4cg#td
zyYJ8X!>y~li<9)wTj<V{@c79g3(q~Poh)Mx_HsnKBzw7}q!94;0H#x|Gh1Ty_<{e;
zllvNHLk9;CWskoGG;5*20s;okMz&0fU+QvzUk6d{_7@2am?<wwuw!T9jkkK8ty<%m
zlU^}#f=aJI9ZmtoovaI{q7Z{hvH$mq;;(t)Z|V4dvu)Cq$<8)qZs_~~kV}{%HJoi2
z0QDSm-pSo(z^bAUyLbJG3#ZLkC=~JN$kR(OcvT~+cee+TTJp9Zks4mw2Yjhag-!zh
zuV&l)e=Ujs?MD6QD*iA3kQMVcu?A1G{w;j}?@#*w`tkpVKl~p*)3GE76TnIo4A5MC
zbNg0+Sf;-vj3IEu=?22h__Zy5W_aMRQ(+xcu{R(jU;Hn=#1iI_5^=rE{Rbg{*$iWd
z0+Y4GxiGcW?<?R?AiM`84Oys1`_Gc4zrt|<NRfYYK~LutNTdXz^<(lMk5#Yv0r1Kp
zK+9qUe9g-lYlt{WdjJ;fF1ntg*TwW^j|`Jmm*_t#GbRbTInTSH1BUn%<}vn(%x+UU
zKQTa+1tA$aMo?i2UR%e#7cE}H|BKIFh`Hp2-P0ETb;)0`946;ovRwjn;ky`$&PUI_
zD^&G)zz<g-Xe0zMb?~;KQmgJh1^7?Rbt}txX83<?W}1%PdWNp!iX1%FE}J7nm{1a+
zhT#C&Zvok#c}UkZtUmy$>npzh%|*wJk^85U`RXg!ON0=QFq|U}pg$ib#`~}_IQ5YI
z1OOr*5<htT`NlqEK}*RDkR_^t(`+~<y6rcb7kC8^fn-6P3K5>_PEa>YY_#Ke`Q~c*
zClJn3U2_$P$Aqu#J(1icFa)JAMo+vA<8GXVNS+U96Jv;Qvp~^eWEiW<L|fAF>i)3*
z|JjfGAO6ik9=Jvppzkuf|6ZU@<S}r1Vhr~VLq_(N5+DB){<R;7abf`C?v_J=lJt=z
zz_}t30qA$I%}EN8cG7@m0DiC=@(<T5ISqY-uDZDcbO;jQ=Y>Rx-+|~2yNM)tCKFO3
z!5f1jO8d_HNz_h8hkB-ZCbN<PqX-iP>g?DPK&7>3mMk>-dB)Z2E;bvi0|QJ^GF`t?
z=WQ2ca2e+RiQ%RJp`*F{8Y#yo63akc-UT!OUe<oa6vxH@{%Zd08t^stMWg`na?XYT
z3Z(1p=ac(&q^*D<>Yg$10EKN=ZEVY*4FCTU#(v;y-~bC;R)qN>?7uEw4n78!@)Lu-
zgC=NiJ{Dni^cphiy15*^5&h{X{xpa{LxOr8?2U=Q#0M<6=Qt&YCaU7VDF>5(EAtWG
zA7HnMfTpj&Pkz(y{HD0_D_wBWoQcHr46xo-^Lh=acimk0-&8G>UKtaBG~X%@aNPcU
z0wnA8scdPT(bJ08Z7ECmpNuPa171$XX`s?N1oAzGplCVL>t7xn^wPk+3D3JeDa8VT
zL;O_&y=X%v`851fO9%Y_7vESEv?>cc5v#IGfK%>YU-~4l6L`aI34cuH96l9cE;Rgc
zfbHjx2xAQ>9|7L~X##%=-4VBoNcLL(7dtQtuN-104RbHxYBXmaV_tb6<3ji+Fc2mN
z-uTAED<J!HvqAB8B&zb$uRA~y)R1^x5_LF}hW~Gm|KD}b6bIi2NCm4m1lF=svYuZ$
z|5zjHNDrl5tWrxit!ha<fiMGm)(s$>k1h=Ws12xP|6Q|z{lJtHC;L!@((XFx-^&)?
zJ)CO%(1)LB{~yxcI;!ffd;68{ZmCU!fOLxVrb9qt(@0B5cXxMpBO%gVf`ot~2uLek
z0!m2ZS=;-5o^yWV9pin^IOk6oZa1*kcdfauIj_%UE(|;{gCKK_6jyY&XgX0U%;05J
zxzcfY?EBF3oP)ZqPt2}>D`_$iYkWq}AnE{cRt&lG0@;aqKQG|<yBUSFcf0adBvH42
zK~`|2p;iWaBAg>X@$^w00vtuvMD>0%<MJ!Gq5dpmd0KU{(kTRwL5u{c<)a)VgkQlP
zwD}Vr>6Wk@iYz6fAr5koLtt)$CmOKVxp*f5KuXEkQZTf7A!Q4A-2B+zentpo-HZVu
z<OlKiCqwY_+VHV4;Bp1KFg=R|Tuh>y{CjxFQE+p4{o>1-a1gsUyVty-CqxTpm|+VM
z2bW=e&25kL+fz1qxC$=q5g^P-5PtDC0~2SCH-d*)Tf=bDmJ_gM!@=TAwIF%NVbbXY
z{Mg@C3&AoLJ=!AM-$we{15Eef9qqSQlYx$ozEzXJ)mptKq)mGN+7~V*UjkT<O&~ym
zy?PT^6z>2`j8+b@*NuRewX@Ln&9G8!>RVaT*I#aLyzCnbA&%2PJMz?f`d*aEy9n3Z
z==jYme>**p8mqkIo$9{MZgLfqoPN@+neo>D(C$?EX<({Ht-5BPO;fu-WDz_J<`I18
zru^#i?l_QE9Pq*nL_U+kv75P;GeDMHsjblv*P~np8{=RttjVXM_y}p^WvSl<U@^Wr
zZ`<5Be;eozkE~hEVA|(!c#Lu#Y#duZNO%uv@(15-^-%j0iZVgg{T*fQSW^vpVxaUz
zMZ_4ijRbXol6tfT?&nR{h8aRp!SFRP2<%2|2c@aX-_GobPWtQbw3TE;=l1v-^y8p|
zaGu8+I4#<I6ncz+bs5Pqp2D*bK;O&M(`l+?MrFZnT6!*WM$oNH3wN_j4SOw$2RqZi
z_TaG-@`i(F0@TU7h+aSj#hJr{<M62u-~6EN+y3Vv<#xpv29g<Kt^hg*%`v!}^0Bz$
zY$7jEibOVi8=RpFYB?8i-EQ8l+BN;$rlNg0*Vaq&7zjmkJGjE%AAI3!etd~1ZyEtf
zN<?5&)^iJ7T3XFeLlgBW6u~#VuV>)KN3<Uh`}eFi6a5uz{Mh>vTyh2dbvxbRw9ar+
zeb)Tzzv38p`Y=J!@Qpa=N8x`fm}_Xz&=NeOa}B5$)?VNxh>k}fT}(Hv_a0o)Bz}PS
z)bfk=OAx-aG8)3s+=L&yo{vGM;|dXR`m~t<{a*3GQFS{MC4h9mxhtlT6!C!t=_va*
zkZL;!4_S5s()mNSDs1e6CqVI5nNnJ%>#o0q^X+&5?So;>R^2y;GeJh$3m)u93O%%I
z=Y}(79m@E}07a(^O8i-#+C7c!o-Qj{rvRRBJeX}=VC$_)`-+|xA6pzCx{ybo`QfEt
zSpESio72+~D9q@_v><jB2W!*?{}be)hAu=)>RD*N8xXE5%NBkCKx7i<Z`Tu$5xBTM
z4`hI)g0-?}KS}!mCPqxRD1P`+A5SIKv%NIqFlvz$0P)^hGdlLkT*H<63AS~C5R~&O
zayZ2+V*GK}grOSgvGb~|`0GGZ<PQm=irMfwRxS&24~7cADJYkE>?s=*-9j8Rw`XGE
zo;RI2Jh7jW<J6oE$G{=K7lib61K)es6}>6}9(kSqo~FRAv9M^Fj#)YZNk$<69^xIx
zkMqeCo-<3i>h}asbrxgY!kLIlt6yLb#PX1e->%_RDnCt33d9q?G_|Xo8HAD<%?u;&
z3z7HjslLMGAk~;Wl?)un>UZVxT^L6!ql_9(V`DTTrP$-6KvQ5j(v~L~UXY_ZeZhoF
zPZE5s@<-Ws49Pu1JggIm0F&?*7<IEGrhoa8570W;{Qj{pw<EJ2GJ}$w<)*0<r<k-1
zvHdvHpLGIq#OBazhkOhs_{<@Ug4aFtYn^ps{cRD4Z%^Fj4*9CA;kTyqhB%34=V8Wu
z%oALfZ+n%6t5d^T4|maak#=P$*msX1Fqu^u$1>WYKQg^Vc*P%^=2ce42kZS+Y2>f{
zH<RV{3r5kdq&)QCm|O>me<h?e$WOp?ZFdMxlZQC*u_)*E?bZ+lKY_lxbq70`eJ*>2
z?}li=I#D4ys1REk;+)wWAe6d3%!gfR6pg}RcL7MQ9^28QuleX8AXH0IXP0z74wg6Q
z>D#;XRN?gmLQ--TT7$?-5D2pSTdDLS%Z@7Ga!v4CQBSEl4Ad|W?}mjL%ukteQM*t!
zpKV)FJyjt^+?{ybu}SzvAgvsGdIezWv^;7y^dBNk<BWc^ZPM)&9CgkTcTKw73Af&J
zT`-&wPMIUi^5v!d2DXs++lf44dbcljh^L6XTInN>hiR%eD}sx<<-Y(?tF@()Awjrg
zdQ^6{%kpdTACO@${AGl`t0-26<@>*yx9L9oA)hfjG;#S4rOb!^0gNe`SR7F7atCff
zJf-2DuU=&+>oZcl2NC@q`&E^H6))zWNi+}v$>B9hxELOP<^hgzPdz*KSl1*w)veH1
zjs1xw#3jhcBph)K52R_b6{HQ#%)+QGHyg<3Z80<S9AB6rZw(`2ahS&L<3H9z9!Vjc
z`>tndJ=MMU;N39*8Vf=nVv>>GXw+(;CLNB-G>W4ZW2P|c{F#=<+pEgOLyrKY$(xDJ
zm4KT=vig3sU@RB5O{(d^HjVcs*ndD^s11=@Es8UavN=ApMwTls9Rps{4n&9mQdeJ!
zN2ts*^AYWrRA=yy{uksHoTDx?2nyuiqqoErs5jA|Iy6T>&ay+&TL?jx6&;TfOd75q
z)3G9mz657y3IQ^I+UCuVCUr>-q`EXTEJDyVOyES9v?>!fO@9f_%JHV@(O11-5+Bfi
zb6U6-Dc|d!sGCSpKG`cv<;cgEsFELgfRT1BwTLl8y&4>cu~+tpe?h(8$!&sXMDHfz
zhuo5sBKB3dhgvSfA!|tn2+wV06wqdgBDQ>#Rnd<~eTa+VB@xVj1P-oNR%DS((X2%_
z;V&jeaGch5=w~I93`DOD!~zfb0vj?AhLWCKex+K7ah2>vhdmWhWm1a1VPDGHYWoM|
zved-8<9k*S4Gjrgi;&Vo<Td1yc5BdoW0Xo^1fKD+e%_<K0ID!N9PQJhh^*jP?6LXB
z1lQq%0J7(rJ{9B1e&(~d)6x>b9?F4w*zQ{h>y)j1O18EolVyhu!Z9v1gvKv7;u=OV
z@dGGmpXHDt9cDSlUWd3VY|;yg<C7yLTi5HyaD>*%e^U$mo1Y#ZWK{W<ts>f$6d7K9
zv^i?LtvaRoR~x7%Fujji{tdc0Gn+QsRcGHO6akfo$(MkgBWW8$Eb>R>ZG_7F_`}*B
ze&`<s%2P@~0-k~`^8sQevl5BHTcoRqj+cE1-Vd^S?FY5{oK>b>|1wGo1|u?wvn1AM
z5j(hMt+$Uid!qZP2ui4PAI<N^2zYSxscTCLnv2JaMSMcLvSW6LG9_Z=HKF;;sWFEi
zh+)qt+tzS4t-O<Xb&MeyB=X4hyX=7F2FWR*AU45gjfUXYg-OFfXB?t}-L_{nXo|<>
z8{Ga@K3@~xZKv&CPnDNqNoS7AJ#)jKAOO!dv2vjsKv%bHrzWgy(mk`U++AUlcbm0G
zjZl!V5#EU7n{YxzW%-Qu6=zoulD<A0^2_)6^+D&}*~LF2Q&Z>)W<UNP{VZmDgyig!
zQ8bxy6|7#3)C<;eO^vaSuU9r5enkb53j9jc6a+&pC24gI7rA0psd&@BwuSmcUam1H
zO%X%!Un=J)uxnsJG`a|Y5u<HoL)IdO6>6S_HO^{ht`Qd6YTdsDn6e3)y>iKUB~f$^
zpu0mqfF7V;YB(v!`CFdtXDKOEzmPt$?1{P^(G`9NXACjE079bXKx^*^9@}DRny&_y
zS_ISH7v!G9Y&U6d1@BBX<wQ*}*<_9<M*_)s^h{hc7%={(<{er@A%c>UQDcoKyQc;Y
zecqXV(>ka)s+%$~v#d-8X(VN@hZ<TLpV2hR;X|ck=qF&JHDjfRLv(85zZ(*vZ1=U)
zwg}Y>sL7Qj#6cQI@^AWfy^1ymDkbi0m7CI@F?o)2%^tEurJvbply4w4pLV$J=;v6>
z7H<S}0v}hIaEyF%QfX~230G!;mM7FEf&KwBS?tLo8`qRt3#HJ@2BDbudJhu~^$m>K
z3<y3g7OYX4O0FD7K$nzd^5VMUQ#QfPdi)wA*_GWDvhGAI!ml)Cbfpqt{<-rFG(my-
zfgPLG>FXbDq=O}+CW30VQcS#plkbagFg4zfEW&CS@Sgkoedculk5wMUf#p^u11xUR
z#b0~j+gQl!51#bZa%{6th9x<3<dZq|&Am15H$qj<1-#cSQRc5nu1SZjlW|X%{<6H`
z(;CPK7D%w;So-Usc)>;17;MKtePM^E@jo)_82)@fbexGO9u@N<L;^IeT9p^m!#T`j
zSD*m8mb9G*eYnUuG@P?yHXrh<P;a;DNvbbYs+8aLzfLXjQ#Z4XZxR#S=OlE#y-bQt
zO1{o_JtNse7GoUOdgG$JHP%hDqAyx`6#GLIq3WEGm_X8Vz$;}eM)XmQ-y2tg_?Wk6
z{^C(IWgOFH*AtCi+e3ErCUP*Pc={`-cMi9ROfGfl=%fav>;{g|vZ);k{f2%Y|0AP~
z?x;J~6GQvT6LjLOcMLJ_X`r9p%&d4SWBEush(<5sFCR7pAAc|n|5_ESaZ2`gC$T|A
z;CSlw->Q}>7(=!QcMj+Gz<*c(2&j!HVZ5XE+6bBi?^#K$veV+>3cKx#;6MO`j&7XR
z&XFCBVXd&hDBkGh2%s=is<VFrZdT8tY(|{b=?(5b)%OtrqN`zFETEtxR3U3(y_Y*&
zeWN0ZX`oQ?9aer;C39Jmo0#U*fy>Vu>cNmfNb@X8a+UF$b}?>8jidOpI>}x?K=aj7
zdZug}amIp0V>Hh2^vMK)-!o!|#t+OqCVCvhP!oi?#<}O++CJi}!4V2Tg%}D;5{lwV
zvZd=zlGqopt-%DNd)j%SteqMIC2q5iF9io)@i<-&z()qIy~ZlY^=#4whQ|c41<!8a
zfm^?3hd+(1$bE?7Xc*G%JxO;CC(&Vj^Nm%APIrvZl<ikFcAV6{V3;m(J-%mGvI*?A
zY=Vw$9BQ{<U3n)U2pMg(IP{xIMynPVSIABPYxS1m)s=FXI&4j##0HiFck<Ne+Hb$c
zjN=*QKl8Q(^D@^tw>OQ(MCu}2#HoCrfF{)%kA3C(-<3*)uqh~rafWG68sa7X;T_O$
z)78G5mTqg#8LBk0zB-AieJ?;}tEi%O$&O$tn>p`AZx_!Hvr<aUgg8N+l_^MoIu*+Q
z>uKVK2|D+Mc`3CI6U;ew*DDr*ny@nogmd~?bPlfz-E~tSe|}Xm_@aqg4$r13?X(P;
z<yKWm(MuGrK^V*S8(qJAe)eC1ScM*Oaf({@IUiYkXqh^x`NT)Ox>cvEIZjhl8limp
z+c0oNbMtlE^&Q-1@W&?$(8oV^qRTzn@E~q0wmjPK(tRQm%3+^9{{BpDfk0QW{r)`)
zfup<}2`kaE)pvaaGT(NWA0(j8@YHDE=#PhtI0iQ-%O&F}8IAsagtv|=kxVFRcn!ce
zN^PUq+k`v7nu-JB3eq~LU_jeMa;$T{D_*IcL+wm29d*;Wa1LaJ>c(`wHDB@jUZVS+
z#Pwq*(D9xJXr&Ul8u$Tfvv!93*0}$2{~i0|Txa#Tn=g(C6Sk4sj_(0=&E0pb%l!sp
zy?=}3D9pKr{Y64KnQlfqdvQV3yw;-n&&R+ZajA=)|EDUmo$z=h9cL>DOJ(^eCTf8K
z=7Hk*y<KGxN_;Mk2M3}Wp(R#7o0Y36pgp^8U$Rl+a{TG^Rk?7O2Tz|H4}@(rcGSQ%
z#e#3uw@(#TtbR-#*-@J48oAGo33t#zyUPhKPTPOy+OC1<@!}43<;NJ2A5Gr=s>~I8
zcK%Unthq{@NY}c3CEXq6U`bI3p-rsW+E8<271#6W5Tx<#2b{Q_Mn}W$6;ZNcvF}!@
zvEtM&a$5eX2j+k7gV|}9N!DMq=RUbu_x-0?+~1wh4|Eu7SAN4<cdW)75?M^VNZJ6e
zJ@dv1>bl<Rm#Gd(oK9ax$!p_-q-49D4qLuPzzN^W*XhupF0EzW-<WV+y8q4zCW-5>
z;JA{aY_iQ4Cm%<^_MJE7?0Z2w$InkfApWgr=toej;;^EH`5Ea-uc~49h;U9wElvbu
z<QnNQAjx4a65Bq;wbT%*1?R`!RKf<Fl<ulNa@sgycg(HEkjVY*hafIrW-mxq<!SyL
zqf6-+5=Fk<+Tq^tGt@;!?)y`MT~!+4-J(!T^}B;KH|~+_Pr<+_!1`rt^9oTNc8r&X
zaZ2m(gXe2$d;Aa-e$QNJ5o5gpZ0m-P<y?-OuK;l5md^~pp0G{^K6`LVtaa_;DWJKp
z1E_E3tn&6u<1{QAb(yNfb<m&rTrUGhbuzLS4;C*TB>ei@OzIw$g~r5kQs##S@7@T9
zozebwTjruu*?<a_{)~HNi>xA569ib6c$0x7t<nPbE35mM$JzxrKm6B8ZQm>%c)WG&
zTRKJ3x_sdO&<E?R(tQx(gWoG)UkFp#sds4p;G?bb<$1w|xFL<B%7{1MC2J>Klwsua
zZ}|f`4AnvWhr)J|Lz+mP80GxskY3ftal!j9#w{3M+;ab3Z*65)P3@uL1m15nD<B%~
z!|TkOIk=|7UI=~wOMU>U$ulcY*TEbWKGQlud?s>VwwYlT>?ccghicIQlD-D6RVYmA
z1(2={*3EvTq2)r9XVS0j+{WU38-Lr?l}Cq)rv>-|2@rQQi9Rlq*}9^6YI@=d`nTov
z5H8N*1np}4SC1~0G}S}IR(e5PsB#39&B+2q1G_gZDvHBpzQlmrTh@H0|NXdvQLd^#
zL`L(ZQg}Rpl_krjQIdNsd|&p`TqAENAwYV?&k$gw|0)n$@gKa}9%v*&%QW!s_qQ3;
zg_IbGQQ!k}l8Qku@XC2(Jc*GxzDwp+=|@$`998EgmLUgc&BfXvBcQ*~)b`%Y!oI!%
zZT@Ob6-jjel>=j;-GlT=Jn16E<e5X3&E8pbtt1Z(;wdyuyl1S%X(h%*S^KS7&q1l=
zN1O-_oGtMY^w^mlV_8#J2o^k2f|FG0!I+G5u|QvDKw{3%5l<i_Xo=G~$+%R|jF!B#
zn~HRHnX84hnzA3%BfPMYFiYz62T1^7Z(0|}(v}*}z6$spHVO@h#P;KhSnHee<PWJR
zB)Lmddl&6@g{IjO`7^uYz2Is@jZNY>Dk-ymN8S<CfX(~lPG>=yZ3Axu6)SfI@mtWQ
zq1>fi_1nGQ{(b+paz9HTssfYI_z{@qY7N4R4iERRFxwmh_0HGUtWV@_0IxzRQqAEG
zyt77tvI=#fqY2??4=cINUS}UG9yL7?|4jd2Y`@!K*i0PxR~7kYwXmd<c_^{0L{zx_
zs#FC7SDp-R5_{LQH;pTcMF8RkYi^E(<-DMfvC}&gsf_QOxKXGRSJ@ZRKRzHaMiQwB
z({V>`ciDUC1FXmoBf)LSxTKbQClHv?VljT}AHesCZ#9J$(L0&x8^p6kaC>PeC2;hz
zv6dzYXUZHxIf@wgA@!Xg!O>qKu3~)1u`@3XZJ#=KGi#EriEpy>$2^sk_h^dSy!TFi
z*M9F9df=!#nDNW?N1Y~jcZFwA%oI01$UXzTBf$Z>#;2!ZW6?f}B7fx}G%BZ{*|+OA
z0F0>^>ce{RNyl@ZF?n5*4A=s0YtUPUyjXs9(+-&Ad#H7pr_%b<yl)1)_F=&zPB5c3
z*ZLVv5;NRyz^b&`JFC_*a-qqZQX{FCZ~(*AX|+<t_F^}cHMRP)Y@Y$ig>YSB>Q5p2
z!t9ObmGOewZ%@wRHiWL1OB2Nx6<m_`I#CwQ=vPPzDJ$Yj)XkgN4<kD`H<<7E{ad?j
z$&*QY_;YoWacn%BWzf&;T^{eR<&wwdCfUWz$DLBW5NV|9<m9$e5d}DLhb9>=qdu{F
zavtt-T8e0%B=N4oey{O|L_YRo84Z~{6h;D;EugVy(BZ*wn-6-0f~X}2rZSittS%4E
zgLv@0(e4Zv$c42QZ6x$=`2MysvBQHz&HrN|0pH;cM#nxFfGU}};>^G8qUGe4HGGix
z2pg1s1;&pp%EjKgAu+f}&M;0J?yQTq6-pfLs}|CH(AY#&=GvNh1=<LQdabX<I;AJo
zUnUP1%$D_>Ef{aTQ`XFqXB&I6`ZhQ5MVKbKEL&;`u`&T!cp6uoU%7oENEgvd(+*o_
z2j-z8gM)1uEWL;ZkKb+6TIVuYo<`%P`#k}5cJgnB-as1yXpV2Pr3@`96b{aa8X<(`
zNno4+3KNKKw@=|6Gqg~mAZ^hjZ3&$@S^Q}9JYO{VL__T5JN}n+C0-q8hOxOCN?r5n
z--?XpGiYHIyHPc?x-YdIni(2%$cAU-PAy`wAC^biySlr)Cy<w2UjOjI;NE8PTj-3#
zHG}Hs4KOje_;r5pbY>}=+)0Do#b74H;mwcphk>0RkhE!O%2aZ9`C1^<WB`GtKwUOL
z(PyFY{+b=e6y0pgnfAO>I+Y~%^~$YA?4UNR@qixL;ErhR?M4TvDifmaLjLI)^*=+8
zy3|=J=i1079WUj0C=Zk3Ss=^KYLL@8EJ5EEd>lokl`S_6D<Gn%+EB+D`rSLqr4g1j
z2E>ck=sX&mlD@i(J*=@WH8N~UqCBacxm%-a@mn5Hwb;_f+>*9*u#a`B%dV1XJc@Zw
zG-Z-h+1M|fGyN82k5xLaqxZ{tLkCw{x=d`U2-4LMwLgzdXu%%7lP3^$o^q8K&sa6F
z9&|@ED99%B^eA~s_CmkEDrZ__;kEkZ>aYXSdAJOYxiiK?Ob*1h*r~HOR4zUz{ut()
z!Afq!#u^JgksP6$Hb!QFrN&HbRU<kebQqE*=qHY9rtERb>e~!`bhVTUlghP&cL<VO
z<+`N*Nb0lVB||1YmPAEsy#JKIYSA*)`&N}(r%kp#ivG`nVb~SAbCefB_LCTVqHGz}
zub;=uU#)nkQIeU>|19&jaeYNB@^vi4=UumnY)_r2MsiZ^oz^QSGiC4T$!DL=1nBc*
z9qM_qa~rELRnNM#of9`i9gxLx&_j2nE?4XQt(5G5HYwYqWJ|F_V8rA))F@)wxA(9$
zzGP$?70e)JYHVsy)+CFNk3dhF1QltkKMv3BfM^Y)<OTxjENpx?#<$T8?HC;u3})t?
z{p68qTcFEsUoM%Uit`nY+i-ugkLGWfLMt4zx!P7o*rA@wpfVXSqahDXwlgq?_&5--
zm(~g`dZ|cgwiE<mGp!eV&^c>qHkb}KjH)B;P<{VynqwkOP(MecZx`N=kwil8@BRJX
zx?dQPD5}36nVfm2(TQX$f$?SK^RIcuhBLV5=ok*q;fM6u4D5U487mo2#jWzM_#0vD
zwYT6Z*IhK4Au#ef-z^=V+t&J5$r5t@iDa-AiQZX|p_UxD2IBQ9UPn_71;Qjn5>AYC
zI7<wP(w@Hfx}VeqR~zXu>y2f(NyBW$SH?<tL;aT}s~#!8YL%QC+3=U(MGt0%>Q@#P
ziA$QAPnEr#iVr7q%Ci_K<nVw!^d;cl+t?~$ZFGXojgXP)6AfQErvYz}O{urBB@gGz
zwkEb-n4#R&Yol$ACyAvV89Rx;;k9>$fvtg8rK6UWD$07wYz%8GpDI-3#7nO})3>p}
zp9X_2kTqnlFs3*`>{%RpKAvWU`;E$eAxDSozN}DVSnc3*0J9D=*O1Ln)g%>7Pz+Wt
z)NP&r^4vu0mfNAQq2<H1@omIEjw5a&0-KPZBdFrTi;V%~-a8;;xXrt7xkN|09i*V8
zIU=Xc_OTwooWZ5LA^c_x-HjnrhHjR7Tk3l%6DLE9CnW=UVOEp0UrI}60uQz~&B;^S
z;H=TSYJ=}xE|a#5kI*HPf~O4;RP-z|#=p0+La^vq#+XW;nxv6X=voqdP{8EqZd#}<
z<c{hF-cvpaC1;iE2HF8y?;v!T9AUZGc8q<{1XA$ZN5mL`_`?>8quXdnzsA<poHfuu
zl$<}qS%dRi6}fVn57ila1s3mEdlp5R0rrCMsEO?j`TS*QfTS^G9kife-A^4p`2uCc
zW~WMw?oSC4>!;}K<Kml=x2#xQKN@4tv1TY>9s6Ek-c;{xuMyo$EE&~|UzU73yu2jm
z`kXKfy=1rYO%iX4796`cn5%g~Gwl3HPH$A#9gud$m|aFK&U|gLp)5!k!~?XlZ;3Za
z2Ck3ciET!W=33^?L@ufZ1H)d@OBQ#iSF9*MEeD^p>zr%tEv!4^YaRh3OTZHM_&jX+
zeYD{Svz%aF{jacqGC=5hX!)j%DR#^Dg?)Nu!B{}9wY4<m$I~DGPQ($yC*q<jM*f@1
z#UJ%RiM1o0{N)-k6TPfZM@f|mfs!i65I2GR)zj){EVN}+pAaRwR_f2>s94E6QPM_3
z^?mnAO=f${&;x<%Vl!1;aO(jU>rgn(bWBS)fKDti@2oKI{?NDHF<y9qL-YsC%8rcM
z1KEK21R*(?HymG+L~k_t22;!D`O{D+)5(XrDJhv1zog!Hq1AcPBE9BnOwiQHONX_c
zy!L4;JV>L%B>5?;vC3&Gg;1k4lWJ^+zhxg&qb5PuW^Bd7&PJq_w}<Xb$d37aKK`dJ
zC}m%ffKqnqnO}p=yBCjat1wwjx9%Cops+oLD`~quq)uTakW|(@^`bX5U4nEn;puNZ
z0|oQSt0GXls@@Dpvsi{ha~y-M8#t!M_pxHtS>J5^W`1EN?|9V-eZq5*Bm~i$rm5#Q
z5DZAtBR$yEpCmPT+kG;YaP~DBO_cuTB3x>`S*<WS4`Cr!RF7mg7M54Gnub#Q0Hm15
z$f1r@tMdF5$I3hNyiEs7Yv1nVmCSQ1DYx@k-9#{s;l8(Sd*A(!KdYbMQUW;#a%=W0
zoJa0Razm!^R)@-tB9^X_PWd@XZF<9K0owKmJmHSsR(6<uI9Fo)m@$%&EF%K4Y_?FD
zYoEIwBsNHeD{Z9x{U(WwPH3#fY1Z6>dVrQRKO0S#Q=E(eo`GopF6Yb>D{h3No+0s^
z2sK$q(@*VES)rM>YA{K0;Wi|1FapVfhPw_knz26uqn1d+BZ)C7*9Z2+oR2Pd6JqfE
zqtMmkT#<S!Me4_`sj`>gw>a+{Q)njQ7&pUk>WnZ3pZuctBlWWa%f`p^QU$Us`q&d~
z^P89=<x9)4jcUl0f?Vn0%-1W8UL)@D9{ki8fvU|K+;6qs4r^Nt9ZTnobOz)<Ym1k@
z3{5jxE>ry$b-vpDar)D1PpDm&YW$GusDOLD{l1ags6A?=pD|S=WQ~q!i`4$+%CWCg
zX_!fnvvIGTQiNujmhzZwGQi^T$vf9urbc@AOgHWD?9XyFC)@ud+`NtrBD&wF-~GSS
z4;iE%4Dmk+qGKfS%x}CQ%Vbu0KhEsV@u;-_JtA?%%IV56D{aGO1TRhFH`~g5QqU0Z
zOJpURBK|SBiL~{ANI9*7!}zqT-8GZiR8RDUbo8tMb4-UzItJrq@$MiKUQFk$c(xpD
zk{<b+*@;+85PkC5X=F(D*$aO<Bju4tmaarsMYs*@iQ$RB+jRa56!!`6o)z10$0uA*
z1;!r3Q5t$rQ9ZJ;XlAej-UdYnlDd!5#;mS~XcAp1jJYVF{HW)8ftsBxT4<=Roz8MB
z<;F0{VX>vevQP0|a6P#F8SqVo4uLj289zLJGj|$wopCF)(E@m1_{E$(HGZJQ>^<Dh
zTWMiMfsK}mZ#u-aO5gPBBOEnBAR8qNc&5SOA<rX4N-isK>reUnjF&NTMzkvXLS>a{
zLsPiha)#FNo}4<zB4?Cg2<4Qu&Eq9y03!JXTcKFrZA5cu*W2}TD(h9IqzYh=0TrU5
zmejVQjzYV7peN``Qm$n{i~Ov)E}~PrrI^FH-wT~yct?o2P<M|lvQEz6q<)r*^;Xci
zf5vNG73{HMw`^D9$fsj&>y~Yq(ctlJJ`ZgS`-Im*KdQizvF+A-xW3sg2kO6kPCPCg
z7?2X6W%3Sj-~saTvOkY^KXe>)E*<Z9&!4bV2+`pML5VE&D<{kD=rYZMDF)c~=(z78
z<MsW+>*!&IY5UEy>%vMB!M;-ayjTNq-$tGFppsGv0?MOk#~?!T&E)uq;7!*3?4g~e
z*cCx`s;HtwY+V09DB3@Bv%&ZH;Pga8z|8CzJ(zd}dCg$LT^U3kNkTnb><9uUGT9-k
zQrlJmkyy+8j_$EY)JTsPvTAq=q&oLnekEK+-N2=^4~Q%4J2Z0#A6UZj9fLbauT!LI
z>6<zpjQiV5nj#srNo<piTJ|_G#h=oz;ZYOkB8+c8ty;cr`5AdTwCtCvDrmbGY0Q<6
zN~z~u8GAeBV6ID#0S1D{_i@7zx4&AtXet?qmZ$@B1LXn9ywkWCd#1TJ%g01D(Y-Qb
zjrYK>GA_vC&QN+DXCaFRaa||vGT4P*(eo=$WSQ$Qn~law$RXbJO(k9XbG;oS!_xbt
zedE$ZcaE^`jZQgZaHvAg&xo#>h_r_U6bct)DVeC<_qDafE3QR11kPJIKDU|nv|dK`
zau6Km+LYqoe_kU)a{anqp}k!`(?!4bw(4&|nU(0A_vMw8n0#AQ0^gs(auIUPgwU{n
zrCqo);#bo`i9rzu9ZE_DfLfi9^X|dN=nlBZslVgFh{n^B9@YqopB#b>pv!Fc;%|fD
zb!dO87c9U2-wz7z0?6V<E)&C~uR<X}4w*PXBM=TDj+X-9r2`JB_X$N9`bjPfQT#K>
zK8F@qn!>2bB_XwPD9KbI9*wA+jc9BTPGeQDYavLz?@PGI?F5l(WfpDCcG4rL;gA<H
zNt!}YDi(?x`UCRFa)Sz3ds0U}A+;FIDO@QkNi_)&6zME5`V~Ta8WEo6CuIT1B?)FF
zD8cn$`Uw?A_p)bc29K>X6y2DVdaE-0t|**bS^XX-88<X7`l@{HnE6C1pzn9f+i)zd
z?$0N0K6v0$@OJM)xp)oB#L!(Fh(X9+Wf6uCK3(1U-I<K+5adPKEL3$Ol+-08Er68v
z@}<TIH=`PiuJM)a6iW}y{&whD>;%4S7CQaIi1G~mvFXPlQuxIjQO^k#7%rfj*!q%o
z{#}<cO79`;<Wi(qI7o}xq-7k(n}yAeaEqBm^LkCoY4K1q1f?at4V_vltO%lvBo%W9
zvCQ2HZe`2FiC}XuBz+W;_9J680U5_T*lm<oE9u48G5qdb_WsmfL@^qlUAhAq*<rx`
zYIZA|jGiHo6>k;HsnecyzUG4Ivva1J-C*>7xG2+W^iiOQK_`;kZv3s9Uqmw=V*1(f
z#h_knw3N<t2s#hjhc_Gb?$WW^I=amd7Gh#)8^7D!@muCbR=28MeG#wf`VI|*Qtyav
z6%XmxurHG-@_kM$hwz%VgBauxj#kveBpmvraL|Pt-S_g5+Z=^p*-5=uW#4HO%qgJ$
zEzIW7oOKUqH%~r~XtcfR*diT#d#Mtz`h2AVm-~KnC2O{iDbsLzpK(L@4%=MpJk9w4
zKFa3u;Y;$Lxd97&+q&7+UF*HCr#?v%5{i8kO;DGn`w$<cp1Ba&&mldt4$1DSg2^R|
zi<N0TZ~O$9@c1?0rTle2d>(0$$DiPw?!4u!jowbrggLx3-h9;h4C6{zDNx2JX-LWL
zc$6n@=silEz5)h)%Hn)6GyxYD>Z1X=_cglu{5bj(jw?BQ)@t|!rS3-(f1e5`WYEB`
zs~ZCTagTGEP6m2?{EjEwk26p<W!)acwT+I%vBwdNQt3+yh!k|VNVp&!)3Gz~(T$9b
zb$wJG>&-R!A{}UoO@y#A-7h@p6@EL-S+rz|OE3K#Ke%S8)I;3@k@6<#Nll2Lz(V(P
zoi8;sd+oggnC$b*l`4IjepwwQqXB;he(CQ?UbW>jgr;&0e>0w}AZvj*ZdHA>-J5O}
zEs2z_lj#eZ&?l8e<@XO8^+8y94wLkd-U+Xb@>6(PEFPOh*!*sMU+8qf(Rmpdx+`)Y
z_>Sq+y+IOvxR|0qY+2}FWRA-*9;8z^<>`jCm`Bv?JiZyQ?aRh}Oa|kL5;nW|a^AQB
zrkQ9inO&W>TT_WYkMN9?<?EmT5#GB{yo?+`z^cSru0*2FV7GDkcBKxVt+Z=cu8<Z@
zt$h#{MX#qErI1N$JkP@Ec{9WpE&2?ugW0-A=Yx#V^NnGWlMhInBYBO??s2cRlRt6x
z-um?Qq3!X3y|1iBt!H1zq%;8c>1)9R6av>L;zN3mPuvbf{o}f^^(5kH;;yWS9MS?z
zD8CN8X<J>qCYAX7)YjajC{Y=K8sYAd?fA&OM%>Wgm;l?G4KZGEO}Sd{O`qU2=3!V}
zQ$R>UapC7zl+eflQ4HtU<^_S5b9@$Y#cof7JE7u9*-6$dQ&Vo`yE~st>6<jI?<IV9
z5SUdp{O_a*keMjB9|V>_l(^|36n-RWRI(pL*wI5a%Mj_&OXyiC7|oFp(sdbMBV&{l
zDG(2#qfRLKUwBH`)$Sb@Wal5PWG_6u@b>7clyRS}@;jL6IXketL!&LgbZ=$|>SRL>
z3&MXL9E2Vy(Ep`IS`r-{fewPqOc>lO{AoZq)rDD-%j+GMAzS!*UNkADI`hY*)|i1C
z{PBtNgj6-ij3md4U*snkF+6GRNvJHE*Q#ikm{EQmMcqODA^4ADzvFkO6cvSxYCVsW
z#+K1YKqOVsdmvTSPyYaG6F+QXAfKoTqAkF=9rC5`8>iutvvF1Z<_hyP0s;CEu4QmE
zzhxwEuqOIe2khwdgCEHnUO5YvgczD40xyCR8Zj@&^M-b$DPk0{wUTDGM^gu7q2VM;
z?{;J5(j3xISW(ekk2-dKyJgi->2Atw9yAvX`6kgr*oyoh4jeQ^mqmv4+|W*lQMzKj
z8$Jyd6{}TU$99gXV~7#kz(sT$YzB*iuwF7!^I-$^ejS8O>*pns|9z;~z{SUsyi|Dz
zgh1#7r-VlspO%y%)*&9Y<s!{F?@oly#!*|e%%;?LD2q9=-0melZS}0xmi>dw30?@R
zJJptZ49g+<p`yk$C(%qO7?}Kup*+Mr9*{emS`Q+?GFD#Bx-LqN)}$dUvM;_QE(1Vn
zNiwoLNE!**(~r^?^(#|vj_c7ULdd-swB4J}rD<6PpUa=_e$jn~3kfJ`E)(-esTdQq
zOb{g7D(@J^JsUEWeH!soE-<_?^?Jy=QZV*+u8WGva0nCDyH=u3E#8PD8cJM9UI>bX
z^FGxfjjRCaaTVndbgg0-Np;IMUobTwv-9N~RZmZmA2(_6nd;(|Otzzr>(k@veO7mJ
z7hwyHfwHGXwAO1Wt#~d-Sg8QQgsAR6@3oZEC=K<aD@(0R+bK;DeW@az(Nk&n&+@|e
zQ~M+q=N*{N-CN$r-}ciCgwntDnGovv+{SD59WR6;7YXu$&v^UJk#gB<l4JBbi(ndi
zlCSj$3c2N>vA!SA8mS$*F(lw6bzONxS8!$qPpW4s<_XN?p>`#1N-HpNd~zK5bL5h$
z2O%uOINYmVqcSxvvO8tKHs$9?M7Uv=u3EF?Eq`ocU7%T>E_0k2IT6_hnmyQXY?zd>
z^y-V+xZVg==eS~sq2}rW`YhI3;>Um_jR}(4SG}R#$a{%DtC3F-DQn1G-D}X&<eM02
z$Xzw%QUQ=%l-PAP?{Oy0!gn9gkhr?2Mz@VaqS#<6zID%$Dcjkn$qo%^f!S1JV`}p5
zA-^r**Pc88375XSt5&r+uZSNH3;g+*7cbbUG;gohwn8mS%~=u2I}`oWT4%`bl<TsT
zWuLJpCAxZmTs?O|mYRvieN3B&I<&gf-pRsoB(Li=SBY9lq(X>slz677l*I6<>dX`~
z&<-w!+=m4)-b&4C3<vHCOdU2{m8gsw3>zdNE(ov2p6SciME$V*CizX-#IUuXpV{8+
z3O|O2zW;*mW-72DK$0e#-I^9F_NQ({T^$B|cMxTO447E^x#|A1S%mfX69oNuw;7LO
zMcl-3QfgyC(4ISxJj!Qg-3Xw-;jCfQQ5rw4A3b`&%-gV`ik8|_(!voXEwUEX{3{cy
zEUVG2*vU*+Sh#*Po)^oKC-D2!f!fWjK~F*EP}2{i^=VET)r!#in3%O@-?#Q51?zDx
z)F_0QKA+y<L7&Dx3q$)9rs}<=pFN|9wuM;xOh=>j9T^J37~&Wj=0bN~n!kimsaidG
z`oQ<6CqX}>ePJ-AW0%_01Y%sTWYRGv+wWbsQtdL{d1)@D%2$uKVVIRi$sQ46C?;tz
zzd}FD;^Jz<lt0~HV}mS(LVkHIMc;P4N8->j$ObQ!M1O*DXgxci)@O;hY}4kG{K7!T
zFVY3-r{x<?827ti=<&2Rb6GBLnsS4xnHj5YfqQkSxA4YO+6vK}^-lu@9f+=kew8RY
zXMSHSa~lzUZxVPz4P@?r%2OQ9&6Rm+Jchi{Vumne64-oc8uST!Hc>);h|~V8RT1q_
z2G(Q$QtY!^!S63tDJ*BYNnEd4^0ZK2e{F7Q?z+k<&J<Ph4c_!elEt}pT}-;|l*qa}
zsC@Im_*MS2Bk$uPzb}EDFR)Gs!G{U9|6LpeK?YL-@|0(@5y*mupriOw4SgpI%i(Dl
zTU5{lQ_O%LBKvc_eXC%Bwx3#l86uL6pNC2~a@(f$a;(2FvkCg81KN`y%rrWq7hxTN
z)F`X40Udj)LS;7*62~VVB5N#2qi`%`_+E*U^z9Q)`wN*%K47Fi8rUbuWzS9dm83f@
zluLg*B_u;=N)^Vv-F|yA-jzbEC3gMvxAbKx2k&ERB*O(O*3nmb{r9af7mFs+jc+B~
zbyjtYSZjD8u^mlkABDp{qQ)IM46^QMTjHLuT$X#repByb8%(;IPaCaxBi1bo-t!B!
zV)aa)`y6a!<oQf#<Jrkm4w~(<=lck&T+I4Dbhix+Sv<UqC4Zkw{HE7<_C=^%!(;iw
zABHY>`f}|((X{1H&zx{c^U(vFcV@H}O~<5J_R>|JlK>aP1oJ-Yuh&|2F?;0CsfTV*
z4-v`?yG`9vcvoH#xi@1BfZP-36=xiB+{T3I%n-KRS-QJ)jBAqNh&Tj_zO(7NbA5Rj
zrRg0y7fLgpcE~4GbvgSS);gA=iy(g0KS9QIN?a2W&|2vo(eYbf)wwA#!+b3Gvsi(Q
zI2Wx9K16)P{~X?7K@6rbbXiDK%)8RE@JE!J37<RMj34Xe)N;9;fnNckttNe#BnhG%
zdc4&Kk(_~^eF7u0zb_A_lkQ($f-+n~4K_LANzBO@4xmf*gz*-<+Y}7sP}jHQ4m@5H
zJ#OzvyGvnIr}uP5>@@NCq=`?C+HKw++UV$)IPkKH6UHNx$Lk4i`*fPl&6mJ1l_W}&
z8%mO^OJIfP2W`tP$_K{3-UDtC!F>rVmTUFpT`Tm$25UGWyd%@zi0eti&<`7|8Oydu
zZ?k7zN2WO}<d^cET~P^5Jo0G{KTMb`%e8VJ*_oHySI_LE`S?(c|Ai)_$$k`UC%X0W
z_?vaXE1jU1kqgRxanE<Z@sFpMC#F7bk|P1VOk3Cn6GNRW1FPH;MB^c<*F&k<KJ&1?
zY3p{IxzsxfSA->>VGgI83$|mc^iSh253D?-pK>cNPj#Eix@u*w*h&c0WEgvb8NasQ
zl)uNtFfT8zbZ8>iL#?9u7Ui;R=CEVi76Wp%?eP+Cn{gPX(0d#%SoR~m$)dOCgcd9E
zr`59ryhk30)s#j17-tj?9aDhU4QH=ga^1`zoU9LEaZ$NS^hr_(oTzZrw5hgq`MrAP
z&ackClg7euM$FX~kSmWKIE)_n#!j0>Zui%H$$miF!G<kg?(5R`m1-T2Dy*Nmr+F87
z??hPz^E?!(CI?`A?aptV?lVz;lF!J7x`~Hg+$d0&E-zGsw~plQ$71vywMRtZG$!Oc
zxwIj@S6V8j;%^$IivLOWk=HSx<D~nm=-8$6-TBwVR%eFA?X<C<6N8fcC0#q#&QF7X
z5pLi&@fux;Z^(>sI=H6u!{lpvix1$7Fjxcnpoa0^&E-Gq@;mx-Hh<(q_&POxPGE|F
zzg5|hLOsjKx+-17j>sWb!YUNYWW4%u5&G5h?BC^#O@lxx_i-#c^%vwet2{%eY=YjN
zzvj&p$J#JRT^W3HaHU93lzraq1^*crmQ6|ZN?{@iRw$gHl&ZU`ao7RLk{MjPbq&K0
zr)phAIxs@C?%RObbC`n9s<~$bFUm`7q{K<>a)V~CV>3lLZxopT;RRT*R5GIGQuDRB
z&?cJQU*YXZVNhRJ5-K`kX+M6UlPM0eilsLUW@VSsVT=A3D+YaE+?&Mbe0seQ8zwmT
zM1CdRr$GaH^Rc{N%%Ly{XB%|hW)u>k?kRuUK)*uQ423z^b*-{Jt}O7WCd!Dy%hh%9
z{?RZ}0&|J6;;CZ@j`M`roH*R;)e9ZCdOsub2mr|_Ps|8X$YXE4A5Mhym=6r}!h+(g
z!n^rtthUB^g0Np?EE`Np3iDl*Ox4mro4gKfp8(JDNE-V|f{O(y;$1fV!Vv+(UG`9)
z@Hf9x5rUL%&XkuJ8`H{M=e$x2k!Q#V^cc20nSIVx<ry(>pWBXeXXe;I$4<e)H=zR+
zf@!QRGG6<Xut=Cff|cs$afMD@v1wyTOcig;HJ(KEUsue5T7?i(M_?fo-=666J^d)u
zw$J5l!sLD+;?BIq;{F|eojYYKKp<;M60T7B$$cK_5rbnkqq?4k<v@Bk1F|c=%c?7k
zLv5>9n3knxr=Ck_8bGsou?!p>o_CHAgfLEm2Gh*pt;naUC48UQuc!B%W;y$c_dme4
zE9%ZaR2=P+KJd_jnhDp6XZC>29Ptd|y_CC-PLXs=Cow(>yZb#9mOhN&o%_+#_6CBq
z?Vi!ifNM21IYPh|pn5hGrfqGQx<25H@1Ev73yh{)-nK_OJ&&`?yT$Ya3ik0F>b>KC
z;!qBWy)lpn2G>uc%$QAeR3F{_DKQE=W$vfQzKkRnT5WZ_S#EF+Z^(!3pYsz~hxPse
zliwaC4KVNDAGyv(no)*&+dqiz(|(nJpSu0TGLnz|-=L-Zy>I=qf18RjIDu|#i(%dO
z3JjuKFyEBXwZ*sP$E_^&g@UHs2s~K?(hu7Hj@pcq2x<snaUVS#C$$;`CJ}r!Fe3Wu
zn;a`FrW&6-H0W?0uiZ;UdO^rCpdL!>c>vPWo^*D7cQcWPzk3iqxiPYnxi6?vJH|bm
zoYqgd7+fMjp!?_w_EEzK&AZwol;s&x6wk_u6({eQr4HbO;-4R=>GEho(e!MHjnz2X
zK=lNWF;>*IzWDmhx<F@7TiL8>>xoV!?h+Mldc!4u20|Y(XXQA9i;^W%2CB{)LxW`;
zxd*O7@7gt6pg&rjCZzSau>#*k#lEYp?!Xc@T&0v68F7)Da3*`#&cVNKveYHVi)qzS
zv(-r|^x-?y&*h!!qsJTQeJXK&BymN%jwuemmK)xaP(MD2!{7Gj{Pm7{6_^zdHa>|_
zEi<m`OwA?fXU4mc+G{MWtPofcuD=Kin-=p1e)i=<R$FGAG^_#^jwYL}qjcxk$O8pK
zD%}r_<ButUIxXJkP}cs}Au5KeeLUNfP_vTFXB`0#NHAIs5!&m&d%Tq_oZZrHsuLUn
z3*xw=vK``NIuG*vsAOr{EhGM{^<$$Yw{jFh+x#wv)01|8ow_*2hq!>S7E~j(tU|dq
zX@psk`1E?Wlm?Aq$S0d$I))ZxjFkxL&dosUGy+GF_Q}rU4$@Ezt+cG${{WX{!<O1q
z?fq5vU@g?=uD|7NGTO=^mo42VIX@?|!3kX_yN_bwoq9|kw0SS0(ahgHx~(4Ay%4=}
z`XsXBbyu9<-y$ST-Y~Lb5GPaBCx>3^rv9nz=FLu;uc5)ZTIMY>%eth0g&#x?_1s6s
z@Plp>|C|o<E!~a$O%S`=80*!8h%}a7GmC7=p3=CU4eDd3w&`Ys4*uCA!tJ7cn`+vs
zH;wT!tJ^y1EZgMmGZRMFnv)-18G39Z_BN|Sjp$FE8Q$(BcG_)qI#~6&r$zC@>d;>$
zxu+ddR!y&zeG&RBs3E`i_H!sC&4J;`%@jvw1lHqr(}}!>LeUtS1wmFX15;$KSCd(9
z>ws7H-sn1De+G=6?)WphXEe$OzP|!J@#DvndkkP6j>l|<;_J_V6sZ}FVeMvDxz^Lx
zg4P6{k3(c8%iFvKy&pKBXbh%Z#)aJ#fyKvsCzduP350@oWbT>2Gw%~RReMTl%hjz7
zBiMbLkPppM$a;T<zZ*S#b061$>lo~v*_-1Lai4mhEO9!#=Ku6O^4BBt?in&6BYCaf
zeOCBGm8FCp^;0O_Ma!1pZ#jhkf7T>v(X&X3oIu<0CYRMPWjg%BHO~muMog!8$SnKc
z<5E$+-y`}agcs|R;)%yc5A3dzX^#8yB~i@aWZ4JhT`;h|R+go%LvXQ#tk|j@iaXDf
z)c)NuSP6p8kdi}0C@e5ZH|eq(+OgK&8&0623tA48TmsYjwyc;d9|feh!3*<hvep@8
zSR6Q_Te<ohFIC$u5qV<6*Bl3490I+5={LE5N=Ndk9mjgkR#EGXmA4hPq7#S)JB6;?
zN)6lNZz!OJ5h%Rkc--`Je7&*b6z9wpUoi&JDoL)tG4-=aj&=u6*t=Wm`q1tN4at|1
zVxei00ZHPdU+5kLvBWNRaHEEEk7hOJ;*&+?S-o?fC^a>k^AG8G?~ce6_OR&(DMV6}
z6jCXb7bI&|mo(BniL(C5HHj;Vow2;y<@rO&kFlHK9ah+z63B!K%t_*kNS1^pFG+s!
z`sZAm&dnfhGL9jVC!uYxfBwe8HGyM`b=++zB^!#3ih#OcmFuuLnr^C;U5v|ATzd${
z7~dJ<R4NCkSgr%-mb(iQGV&O*<rK$_NR=gMXxTDor5KO$mr)ciODQnWzOGDSjLL<{
zoHl2ns7`VPwy)D7%TI?%a)-I;Yge*|ymrs_K}8tis6-nRCD%f;Arl1FWmBD@yqSeP
zhPye}MV2PNg*mlkb$ov<>JWUN-wW2h`Kr#N<FcJ{cZ`9oxy?q+0%@UO+E6HqNreR>
z)vqEuOFs;%ztGy!q3h)OJ&&*%s6o`T*2$+AhAg@szKJ(!iJ<M$b~shBI1#YVN?_Y)
zNU&OT!0KE?Bl+8$55a-kpL_#x)U8=4?Uz8DpJGpW9GT+xyBQ7i;^xYve*A&3ARHbF
z9;{)t=kk*(`KtabjJ(&Fi<*K$S?rF11%YhpH+!+(eV^JRE|S4v>}PoX$DFWxHM?A!
zrJ&NzN@QL!sH#x$E8V~sb@1q7r1oN#^um(Pcu{zA8sE<k)C&1}mzadjBqSxCc{f*~
z`_m%ZoslIq*oNB5vxYWf-8NK(@y2fYJAZMXkthoM`vkB>00B~qLnfW-pVR4Y(HKXD
zkjDC>nFCWvHEr+5i_E_#r5p+ZDOZdRs^+roq|MiuSi!yC1W-`o7sYAKrPjY57{E|+
zKVQNip$>=Fkfq@dTzN@9W*3%wP%E>3SXb44xl6Q*1SvMPj79Mz{k-t~0Y%)<-6Kss
z%+yyG(wd>BsiaO{x;T(!hNU45QoIwO_^UJ2o3LE#Q;Cw;gQ$$Yd{2EsuSXn>vFE>z
zTU>1L(2I5e*46+CL%W8pO9uQDBR`cltUd`l&iwdBOYD@vnT>&Kd$K-A&5OgUDd>|Y
zT~9FWh*}6VU$z$z1Nri!ruVL*JJBLGCANtCqJ%7YlbbvJGtW}VFpz#$Vpq7J%PGce
z#kizJ;ft9to)EB%@+kI`6Ut;iW4_dRL=@MifK`UF{+zgg<%o)4fXQ!6$WkC(_k>t%
zY+!Vdky>xpZ-(QM8P53fCor0Rb)%YZeu!cCE}=Ej2fH<Y-+H0#rfSDKfY1s?O&>`3
zA=Ns@GQ>`?{$3$##(9j)LA83Ptg5<a?T&eIMe6GmeE~+M+Q{>SFMQu0B=n%W_s)|E
zELbJ9B~%ELIq$X^;wV$A?v*^2Y%Zz$m3cv@*Zf1!n(7sGxoWAug0x9)uM};+UQraM
z1EXb;{SQguO<M_np5uyokz+>7jhwSvaF~eKDLJ#nvdn7)Xi20NSW=-@D6EL)3~VYG
ze+EJa?{L}E?}_bGnJ#!{6YnpXY3-8#=<PCz=T-h9-U_?W64|<YVY2+3n#n%;?=|BQ
zJ!Ih9^9k*8#eapW|9MOPA6I7ltPFH}!1OsXCL9;h+hQZ+sbpeh$g`<cTjwR$oC+s1
z-aSf?7Z!=>q$?5?ZHWoq-<=s#*DMtkJwJKQMhM^TRiMbSm$@+=)%<@|vVV5ye+Bjb
z^@}fr2+MctbH4!d#>unh3t<>r+U_`4pO`)H#~$Hc0dK?pO9CfMCjT*JsJq)s%Qyf3
z-!K>s=2$s!=)3!EwDLzyIY^FU3h;WpgaIvTFph!`^nz-+Uo_~tjpIL~=9IB?ds~%|
z;JB(iWa!?EN%0?vz)V^a<U)?&3_<^2{8jN01@OJLv#po!4#0BPVj%7j<v<q$-r%*W
z$^uN_!+(_p<d{e%{TO*1<md8sK|*0~s}pd|DrVwR_H9%joOk~7DXCy0(%-6LPFzR*
z&rj_?x68jZ?f>bEU>|Ti>*7iDaKaDk|I@$zdwBn^|IAbqAAjwYrs<sg|M&kx5`9Jn
z{1JUF_Z<Ga-<jsn{e|R;fw^n5;&4(-Pz8PGSHAnnq<h8V=Uqbn&!XxF+#+D>E@1Hj
z$P_BQ0Rb9v^0Vvu|Lt#sBnjROA*fCr^S{4SX>9+3bs@DH0MOR&6!=B}7K)`C5pOil
zpQ*$EVo*2c&5^xx1WVEGuFjP=`-&%d?{(-q$m&x#90LS1t`!=;h-MmqSIbd0ySea9
z9D$A-8NIfC0CP>^&y{*R_-wm|x2+{^U~pHiblg<o3Ygp(0JqMS^BzY$m2Iyg14B2H
z|NR=`o`M%tQ|Rn||M@#Th5Q6wOHo5$<$5!uZtVTZwSWkaVN&aV4q&x~|19V7XA?X7
zEaUK%)W~hB{?{;NByfk9A9e=tC-wsf3mLA+W|9-QZ%2Un$=WMCm1j}FLiK+U_SRuh
zw(s7r5;90ji*$ENw{(Y;bc52}Idn@)w+hnIEhP++DvdM)f}}LQ*Z4fYz1BMRUi<xX
z4xJfh=DyGCtncTm;{$jIgHQy#MgrmHWLYiXy9%fpu7Vdo!aw~2ysMr|02?FMePLfF
zC(ix%>##$K7e~Ha`V~ZsST}P4ReBT<Y4O0Gem4r`xBl?)+r!GLdZog#IRH*xke1ke
zTha%pAGZMigabv`5*OI}OG=azBH?;}5&0!rv(CH<nrN1YJ%lNBiDi)CQX3#C3;5Vn
zerS3KM%?~wrIyHlLC>ortThf~dK<l8KpH9DRd_e@!Z8sxBU-Q{taKZ3cn<#DcGi%1
za~#1RA%`>C>JgN8hf%A59)!+o1KYVWX+~pk7oA||Xyg3<zYDVL5aG-YO4#-9#T80{
z8^YHBbdV@SJAaOnf}scu*hw>jFNCm>{}|zdK%|g4wNT#x;0zY7CzUc=1B^$F1VQ0k
z_+5qg3Ba-^eB-G?@b$0m;1kNMLR{@u#oqOd3*T{n6CaQp<%Kl=tRF2mOrT`0)ISZ?
zXj<}Gh(JJP%)jdBIc|ugW>ey4C(8Bg(BeiDgYj}RJ~y$xS?J-2Uf0yK&?i?l!MChz
zA7`}w&aqbjA;<9C=LfJxz*3jEZS2OdKQ&yr(xo*0&Z>=ie?A4690(zp@~Z$Oq=P`{
z!F4HtA=_aSUB1VZq8i60&4B=dc<!Bqy&4FFF3$eC61&uZlz&6@6L6we$Aj<O&9r-g
z9Ep4jyq0d^xadhdexu3F8gZ0&_w|xFKn0zM7oCNul@lQ08_N@$1{_Id7gu?vf-B+I
zDneZCbJ&aa38bh*U5E`5@2-=)C6hhBV4fjlW7-n5|4WF$j|XWJK4xv$Ki%8=)Dsv;
zY_HciPQk9I7f(NN_r<_pCh-S{&NadsJB%<{t>_@r%MrakbjDm<fHGriu0;TJrHJn4
zliy@fA~x#tI+})Q;u@o~@lVQZo~dp|YEZ98b$kL2jb~IDK^LDWoqmVY7n#ptj0UMb
zx_gj9iJy}AsD<I?Q7Ic_04PS?1p$fX6Tn~{0QOg;<W@V7{&xmQN|)bs?CR7!#;@@1
zSSMG*Bv^IE<eT0_`t}JCLl`sn!Dfbu$xeGU*m$a4bQ}Fx8)EhbA_cCdBEuAD+V89H
z-6lkLFPNUXAA_&$b4-W6#sbKKmk7nyp>()@)IpZnDOa6fVJ8wN;0H3iDN!3UG^V;_
z<@UgPscpq)_(p#Wu$4>~NLHP3*F%|gRnq=@s`=lVF#jIda*+FYcXCV(1Ummjo~*0C
zoGy=M8wAg97BDqL7)a{-)q{>xU4SdIh+@V1CooLsK}nRjyKn)Vik40ODmQa`jC$;h
zc?PZWhni0BOOY|maJg2_yUts4c8jmSGUogA#YiKRRGl9T0K9OB00-}GP?Be#v*Rei
z8B;$eViD1;6n1TdU9l0-yAlSDFjuJf!x$cgsO^6RSGT}8<6`9$n&e#sj3av<f~idI
z_~uaHqr2)me5Aj5wG?%aQMT$M;Hs&i59Cy6gjj)phjv_!xMtXKRUmLpGmI(ioJvYs
z`}=J)X#L#=$Z|m3_zwMNV*d%c<hU1r2lQz9v|JzpQ0B^30lL4caW?8HMZRUqT{LYX
z2|i7Rzso03#dr&AZhvMh@X+h!1bDHf`6<A}!!tQ8CQk6J@wq>00x)NuX40{n1dFw+
z$M|bFGFVjt2>93m3s+~{p8BFNgMrU##3#SK51Bzti3l&}J{>({!dl^Q7(l-q%(F9+
ze^o>y8tOq0eG+p$58!uG>%RexRy1S)Gy*7>0VvOlUS-WG(c8@zaMgbLCYzE)oPY?v
z;-revJgFZe3U27dGX{HT$5+)WW_)H?u8%Y+fDCpK)QZc_6&slcz68{J4}k)BUcDv$
zB{wFVNqND_1Tc!kx1n8tO~#C1hLsiE9cOkw0jDN$v!qh{^E7}>OK=XLHQS82!s&el
zWW#5|s$zR|p1VS;7Tk$Vr8yeg_BHRJzt0h9U_>)67e(2)SBd<n0;_lTrNBG<2eYfc
zh0z^F_c4*UBRhK1d7KPZ&Z@niZV`zD*1V&&qB*7*HP-(y@Q^)FO|uQ}ZilRA19Z7h
zWOaw_I~w{QDMJCxzk4r?wK@V3Ttrg#LHiL~UF_^R*xxYukzwFU*#<<gNpQ<Q01F94
z?15g>6M4eBIl3+^eko^|_`z!dsJT7o)s@gZ^>KKU11LCuJ}$7hV%+myKF~EVE$KD_
z{dCd&mUH$4oWqa*VF4&D3C}yvISETVNKd5655HTIAukJ>jH=*%Vqd(3*w($7Ren4Q
z33jk*U5zB0Hwhj!TwYrbZ;JbKJK673$Yx~x&&7J5IvLlDcMCjWYh7$h_BA}B2lR~(
z$gC8eLe;A*l%bLWC==Voh6-+=M;d|X&A_)c$dB%YZ6^g7siPo3<dLA*|4`Agbw<}&
z3yh^(mT7J&#t|ZY2w<iK8&!zqpP&BxBVP!=A$e*PXZ=Qq<to<%6-Q-fB(?CD0m!_k
z#$U*%@4q;NTmN(EzeQwD56wIb{!LSQ3UMXzPV15aVKLY`GT8{dT|w*lvj3;GI8Jef
z_XQXIjD4U~Ai293k`PXLSd4Ae2|W8IVuAtQ%HiUtt;9k~VUM&(MmL~MHQkFrT|=gm
z8V+wEkefr-+zZ3ta9NXy?FVVfHU|>5EzoA@dtGj3iWG5*eZnjL4g%!?T|9!JNx>g`
zn(OnNk=zUgJ=!Ax97veojocGhUz*<T>b29iBM!d%k05ls(dq~S*U8Suypov+g7giL
zfpiYt0&15=8y}B6n;=M2UgXRF0;exh|LFIQt*1@~#~7v;(nB7nh74MTF+P38=o|DS
zv@+ekTiJ|~6osO7wdz40T9MzzlFSw5`>ZvF5H$t53ho=K7a8bIQq<Vb2*T0~2_V7A
zi~<1Ir}yaDTN-suBJCmSLkfTaHJ?@$VDFm0GOozSA4as=t7A0fz=gK>!DSj<hI5gi
z`@pm2#4WRWTYh8>aO8ej$<7N6hmM0h4{0C8kwSMI8yIrbh+O(#`Yv&~R3?acL-}pM
zG3!oT$+TiW{s*LR{_Pl_9Y6}}!J3DQ7o|#tzj~J)0N%^9lTDI|^$~*z-kpm;AZUF@
z?A3IOPh?K4OfdX-`R3+_fYNs$Qg^_##}VBrVRCPDPBXls=KAz<6M<3ODhR#{@tg+j
zqjoIusz+<Dvz*P22;As@9f~ViURgn0aY?C@G<Ao)*{r96juXsNXAZpF$`Uspa{)_l
z?amd>@**el{bmP&Jj!q-dAxC{KTgGo(rU}#wUImti}HCg`~ls~uZajKa6c8ZSw4d%
zJdVe)lV^Hnt>rDMK~(<F?gaz#{lO6sX7iISr_Q|=UF9i7DpsB#b{|&)GEFT5I=|`~
zMu5~_tXp(y+o8ow^fiWz+n`^(CN4i&J){5{6gK8Rz-`mEGJ|(JNFDX$5ki$gS-?+N
z-3L#+ca<wo;ahqtHMlerUVmu{#|Si$5fE_SQlUKIWn^+v+2Pgi<R8E*1VM8xMx0Xi
z*Km18Jyx2hIiiF#ww+{uvQ_Ee!-~IU<iH?r8()}X`9&p1bm)nDS|`uei}#H12f}n)
z4|lAu@va0wMCf~h%DneUtHT!b-45$BJ7EH{(!W`zE6Z$ig!%I*RR><x_bFOGHm34Z
zJhmN-Clw3M_hs=@Id#jmgoFyuK;vJ`Inj(geD=@X$v^>)Yz7L;K&k)Stu44cEC^oY
zwn&s~=L-U&=hWngj1<9{W%^i()EYK$f7k(6k7`xYCzS2J(1Xg5WO5BjozEb{fFoFY
z!qkv3<bE^=Zn9eKTTf6Uv^`~s_Fx-i2qoxk$3?5{2t%S!)=9Q?3srz(NDZ6bm^)&7
zW3~2msR{Ss?Wrv|wSQpMg%Gc;0F#W(IND?qO{(CVb;@&(wY4TCeSJyIWRyKE1Y|0e
zWBQ|}ko0TnB_?*N%Gz0!-iW244K$*$_oXX5DSgWQ_{aCGx24d?H;CFbXv9I|$Vo9S
z?=9QU;2+483N;BrfA7bkEU4-+>Ml1_Y~-FJ7E#woTGFDtq5(q(4s%w$jIlqU9d4mu
zElbs5k%oJPlOOmlefjiWdmNoHoz>{m)Yc|aE50o688HoWoV2V|%2=Z5Zol~TPki9I
z@su%h(q>Yvu<8z2h{GLcZN*QrwMP9^fqGIlp9&#gPh|pKCA<MHN9<UO=(>J1)?nGb
z7u=5F7kR@nOgeS%A8s{vvc2BYlD#PTS<7|r8=?)pjb$ArYpfv4mLb_v!dZO#pt87D
zUn<)sH0$?+Dco3{c~XU3-0oVL&(bPtZy39&r=A`3$5MVnfsfh|`x2FL9x)irGb=>1
z)k_xI;`xCFJExrWk8bVT9hr*~oN{clvQ)1^CeJ;)ZtHds@t+IF`e0vqF9|E%%xUyx
zh1SvcwS@6!iSDIeDCNY_YYfKfFUcUJptT?f2(E6@QeH-XH|4jg<t|d$O+*{%n|X>j
z6nd<vB(FaQ<Q_pb9ztet^Z(Wh;{~8z5Z~UWSy!FcuflC`jl#mzmd)bj^0sYwur3&_
z1M#7AzQ4mj;#_wx$;>gzl_aB#Q44s6s8AwHUs;AHTOH$?spLbv;>9kMt;1XIpNI@z
z;WPXAnepPaVVv57V&hsSNX=Xw4~3~RS!c4LD7Y-omSd$9AR$tzpyhP16y^$;0ktZl
zd9{3jHbY}8>^s#gb*8ab_ITM4Z?*avUBozpC0nty=$?F539q;MRZ37_7R>wo71!E5
zd%Rl#KXv}1;ysydiIgdiAQdYkMw*|`%ja@G4~7PG&5fSxdz(9kt2!Y>{yU}0aQcp$
zY)#e$-$hT3e~QRw@gZ24^#8O05ML55wV7`C+5$L;?1;{anO_a|!;Id_YA?#<)F!rO
za?Eb<y{Q*{V?k~Fl04xaege~@u_@!d7jrQN!6(0F=>+D(WEN=qCx;^^Aq0~ICywVA
z=KYYHQ#o4lri1KahBx($ZwO1jJynaxeB?MuGfyMG;2AWd8pV#{FAAGa8yaA^bzI>p
zGYr@>T$v|~Ys&l?m7-pZryrXcbtmBK+@EeY($&=RxNg9eXG&?|Y%NLH8rXJ!C<z{d
zRGK(3$+u?Ff4Be@$MOkX_ws%~$0c_ad?({eOp9}z&7_f`46U{dt(tEuqF?aKb2eJj
zSVGZI*5B&Dq+{%}X`$KzjS64A8DDeN<%gz|wzEZu*`&f)%2htb*%HAISFmMz+%%L{
zb-DkomASZ6jZBJt9?4?QPk2>~_MxnyQ7m8rls7BGO@#UJ<Q~SA;8}E5Yc)9*vA+3Q
zrR%TO$z7}(07N?{dXZa;3Yn7%YZR1MTwferX&<}RvnSf1^K^mq&$)dNhZbGVuqUx;
zhg?;P%yG47pL8THMWu6zKf;&qRe<C9G|p9Sy?EN8&0TX6&G=})5w1LV2Cd=Xj2fx^
z%xDZdhrHp=t4^CZ+U4pw3zxlluDnvO?`boZ#j@m2!*uv8Ea_HuXh%w&$j3hYtZ)!W
zrwmt<Py9qRzV~giz9o29yWrT*3?tmSS-GZ+`^Dq&$M&>uP<QRfUUVF!e_fdhsgGMH
z$FZu}eU6CMT<;)U-A$?w#N~}aA=HUwo$m#*g2sXHRzdfdcXsPL`0$0elli7hxx=WG
z9<c<QeKDRcWqX^h{C>8t*SQg`XzoV{#<GV|e!laroy@3RPG+$|$4eH4h-q3R*ol^b
zWK`%K^DV=lYf{$J3pjU@BWU(c^4BM}R%1vP;v?C7{<?79!M=BT<MWGju11IPEtwp{
z$*-br1<#0e9}l$K_`SHJY7vHtpv*DTxE0cj6`eqm*=g@=!V|k3G~;u}W!_@EV}_@@
zo+p-Cv<p9mVo%179xN$l@+{X*os{S{X?J)M{8N;z%H4~j_^<64vtis?^6SYeW!Z)-
z9tNA_GY4Y#Dw-Ms(~znVw-^bc_LSS-3**F?vEy6aWA(B4o(_!ujd|}Z6@r=c#&n$R
zMX9i|tl!&x%X}dPXx*)NVQ4a+jJPo-%}NsZUyA8@PN@q|Z7f#_B~T}n{;JhC6Jeho
z&rd9Q&lWm+|1uUe&Pcz|Hgw5|ApZ0AO)Iz&Twd4rS1Iajp`oxhC}jsX?OpnuSn5wt
zJ}tsO3^7n~KQ1%4j<+cqM|ZSw{j5lA1Lt^S_;ohs0&NxNwar)F<T|Q9UbLy@rDxjX
zipVlMGhKK@r5y7d3Whcb)NgbmPh*;Btyik#WgGH|MeuZ@iJ{g+*|3K%*3BDH?IfsV
zv(Lab)@H{($1>&Cicog6PrZvOd4$HpM8)@zU|ctY3;yo|@JViYqR-TM4L?R{lmv>p
zp&Aa~usa;G22eRO*=*9#<KS%k+D&fMCR=xHA8u(>(_GyZ3nYa3w%;w4)$pEA+}UP4
zUk5mKQiB#KF3xHuYC278HyhsV%%o^;(8zC@jUlS3ly$D*lQz#rwu4(X%yRC>mc|O}
z$@XJ?rJQPSDlh8CJv#Z9k0ieX&IO`9X7&iT&^QPboKMCqybvH}!yM7rg6hX20K&1k
z0%M2q%w*9?Hd!wAmE*eCkx{QImJ=fZ81HjFj7((ZMo2Jncu=W?=p?k8M({TNGb2qr
zT$rI|_tS)W5m`GRx*Ls$C%)vdZrhT7<6MqqvpygS*OH3~5MJin_5G(*$$AVKF8hUR
zX!?BSpRSSF6-~&sVuRaSwF{!J(;U*1Uc~+Yii|DTvn1W<M4%S$vh75LsIc4iAd6Hq
zEn>Z>xHj;Fn+dllZJ(v!H6BPaN)y@S(#8+1F{0huK$QtHA+{{SKCT6ah6LZGNJWZs
zU^dZjEPQ;g+Q&h!q<sMNy}EatND`ns=0A4vh{a^z<s@!!U<*8M>1M+>8q7WxoLvi^
z+n;*%BZO=*Xh}QmfiOC8Z^%stXx<LBEYh;K7(b1OY|mzjQ_JI#^66~J=+{X!$lBBA
zk8>q+4#IpT91DK~TWJ)!S@jzn*6bbD?X5cY&F=Y1<CN?g^Mb(WO3T|N-GHG{tcBte
z*aCAKEp-xOR9Ta}&OrJ8T8;|6Wo7W}WqRey1!FQ@18YL(Bs)V9^lA@RUS*$_Or<Fq
z66Pw8ep#|=olA;?8F@uF@1p*Nw1tC5(KkKxFWf4Pwx)=^1x{HJU8#yr?Y4wMIES|@
z{FO`-cM`%L?!aayt;3|8nYN{B5R0>V#euuD-e_$w-z)4l5AkBsd}4t2YEJd^%Ie)5
zB={NJLg-`r=|ZU(`a!X)XUAGBYnzA;Em>OI>DDE~uHBWBgmwG78xjxUM-Rfqzr8Eb
zaji(|84x^k>);4<HE`0EUtIbMYJN)C!`kof-aC7?iGvHWqSizGZ$LGGOmOh>o9DV7
ze`bsVa4c4mxj@+V=%AV<#p5FkYj=!6x0b1os&wMcLt>O|F)UI;A^v7Y<DmQbNqmU=
z)sKUR8NOqn)%?T(-{|XJ5czp<iwrEHHO#<gl+#mG;MlL_YPT{F8zzcG94<GH*%V4g
z;tN~CcLu#2bD+DZ#*E@C95r)_Y`5G(-Er?h4^*k};Hr`Tl?rXCxWx8*bgBYX&e}mI
zS;ua80?bmtDt^^uWwKnY{96eanOiHoJf|bK^p%Y{4;KRm{%)xw)Uox{8rubcr>}2>
z#MFgoon4m<2)1a_%p3eDCiX}#88ze_8cN#F=>lkPiA>2(+SOY7#NInLlHl7MA-NVi
zU^ko^j%VX!rF+L0`UF}n$8dxTbsBs<-`nAK)OItApI4nIC|t2RP|ROz%B*AaOsI!0
zxEh5$S}8Za${6@i!0pSO%g{&iE9*&s1AfJN!J1{xHBxo7&~KXCj2M%?gR|^<akk*h
z->-n?^Qd(kXv7LuUW&*I*yiJO1Qzex9ysM}8tcV5b!soH5KROJ#|udKBnV4Ogqb9_
zK5i46*IWzDckD$aX1jjA@N{u)YHPx4>TY7?XnA4zcavGZppkk5=xH=|hEqfaCH<nE
z889k{s<b_MGUFzNMYKg;%pQMA`vHBTV)_qlWA|sB0Ke+@`|523v^Tyhs-y!E{I?Hp
zg*+Gep<cOvzA-PTHtR8W3jZ?@>iIj3dY4(Dl=4rjUj+~4_xvkPMjHfNq_-;H=RH+<
zHa$myS^2Ah4W1*Yez;pEdZPjJ!={{ZLKKyF%_ou;lHAq@tZ_BXxBPk$)oFCtOf^24
z-Wr01@iN?)M`bE=J$;<1<^;$pE@Slt65}#+Y*Cjew*aShq$nW6TfI3YBRs-zekmXr
z1+~aXEEiqHLB)ubC+KJ6A&3y=VYO51K$v{SHXC;DrNtN-Xnia^EW!z7$|9Sswr(J#
z7J<wAK8=SDPxI}#6Qs;YHNfEgOv5TR_gz#P+nD(bW-WS4fcb)q7^mFBP0D`F22m22
zbUtSxQGsIj*lro?4(x;LP{gx*fj!$^Hqy4K?88((kbSX|Q_I$IN%~8?)S|=33r=fi
z(${=cVq{%c;xPH<*sMAg8LE1f>}0_)m$zf@^m&28y68zg-Q`En!1m2LHDGI;JfDoJ
z<I>jQ$f3#lDEm@ULaSZ*MK29Hib=Eq+m7b)`%<{@ygI02B_gTGbx72Ip9y<5vQ+<b
z<ZLF(Y7UcbZHXWpLCSIz-*@Gi{;~!B-!<-kD-kMOh5dSiG92LMP|fM~Owt^?KzVMK
zxm$wpOPo8#DG>~Z`d*f`#EQ&p4AoG;w<@|T{Uo4bV};w!{<PCY3WPO)A)Pn5ZT%4*
zkNzVMD%^+B%A9>Pw+ItZ7m(@;k>+Eu$IDK>!<C2N#H+&QQ}J=*%%wd90j9YREEf^M
zq&-7IP`JlM>3xJrhO+7Vf$wLYdUma1%n^t5&DiT8F7JXa%}A^esH`Z}Z7AL?=SGEj
zdJrTS<S7+NoFspEzl9WLVz>+Qf2P@pGw82oBa`i;JbBolvP^TYJ|%MODQ*W<KS9J5
z)<HjNF~pK+YXpGV3-|114`1mXiM+Pk&0szCOSYYDG}KL+IOpD$jmmNs<tVqF=-!)g
zhbVxFTFH0WliWvH>ctZGaOU;-E2Va*IqlI#JgAs&8<|1HM6WH8y*+VIXjp>Xpqb-F
z(`PDObrpT*mn8NZf@mh)QlCE9>g`p`N?vMH%;TA4|9P~?`P<z3*A$vmSV<Qk473A=
zcpv(n{kquM>pSFtS(1^+OB0AU^Ucwhl#4{)Yly=j4d(@sXV_h|&tcAQLZqNMzk4eY
zwRR@K0jeQaQWBBkZ=ND~KxYW~>8Hjw-M}=#|4)qC>)%tN7#~xdsedBo^@*s~P%i58
zOvN|OTTls1Ji0KdoJjk2QvXR?B6_$MOgW)Brty2@O2;#?3d$(9Xl#)cqwFYckd^}+
zmg`jOzxTBiio+rEf9<A4;@&sB@Hnp=q73dB`ohpY$?$2?YhJhQa#c#3LPlFxw&4x2
zquXOjAP4sLwM?C~Yo94mO)N2}aL<(Lq+4JZKky52Q90T)UmGEAekC^_EeB!28cfSR
zaGGRzJIU76I{M7Wd`%di_)U3oOnJYHcUyVO6MQ**_|8Cp#DTHuB-hq;9>?h(_I>Y|
z^21lRNB;HwM3{@@%$L@&2U8LY2BfO!(~}>WkM^5YvergMr5L>-<sQFU9T<LIfd-_)
z5Y;G(6M?;+a13*8F+0rwn5=Y`CF!)XDdOa~O`I$a%fYo4>SFxz(d-_6j?Qt>UX<Sr
z7AQ(+WH#aB`Pl4ubMQM#xz6Yfgcc{sU3*y0kc!FV;PRYmG=1MKQEuWfzMdhUo8GUM
z3|SS;#|ewmzjITjI`f3V5xG#UciThG&Jp6yjat`iX_Ep{TS88>4bniW&AH+6Nw!xW
zY?)1k+ty4tneIj_YhI$nwdI(eZg7#|o1L<yv1x5l=eES(+D$pb;Bx?TX{u3~JmEJL
z)Dpjj0`(u4tOk*PUukk#Uw>xU6WeqzA8qBtw|hNQYD<1RW;z_PY+`_N>~#qmS%-DW
z?2So^&U3T&uxGSE)CGpKTfKJuw3NZz<qZ+H)$c41%yTY_*}LSoa!)yJKC-?!`(PJ$
z;|kholWZHZ9t3&)Y<rj2(B1;}gCV}i{Zo@R`EENp;qQSYzSBebr~WW968uz-!K9}9
zJe^ZsjR~ycy8Ox-lWgR&v@d;y6_W5~UI-Ho3fNG$?=PJqcdEyBr?3|T+BF_uG2T;}
zRTE${LBTN_=3i0g@X$S3e(RN)Fojew>H`?Yh>>h6BRZ{zwA{s8%&wg&VsfG@{NqUl
z&Po2Kva~UBv|1UFQeC9>lP9T{v3Qy^oAKURtf}%#Tfv^AI&X+)x8CcvX<u~%D=Qh=
zj@BS-<7F=8Jijji(^Z!PU<^Nf_M{PQ1^~C>UgdK(5;jp!oYlxhvq$vtXUNhHx2etW
z<6U2;3GD8`vUyAJP)qP0c8(Rra}noZ!fjPd;`Q*&<GoeX0(ru`D~%q{P3`Fl1nR{4
z=Yf8l-CEYUx%DckImbLhhn6w&_%z3a$=riK3gQ#CrEapuPs8jRvHT~a_r=a)n6cMC
z<7~;EmcSOq{ETIi)G1~BWfx?xds<M9F>ejBHO3vLgZZX9C4#}&>Ec>0D6VOB<;;1D
zzIXaX0A)NAMnyqTTju?#yf|OFBd^pwz(Z1&4-5ivfFp3M&z7&)lrPNbRaK9>_3L?+
z-Fn{#N58MpBrZam3C+yu{G53woAng)>X)AT>ktb3XNBio_`P;jy8qU#p$X~h^P-k_
z-^X|KHiqU}R%7P&Xb>8w#erRZ(?u|hqkZ_6#hm-ASC-yVVpH6g^t}Ee6)L47kJ`N5
zNuy6`Vrxns1cpt9-_nglm2Y5=Qt?T_Ml+2bTaIjav@~^Q`IIJq>v;F4ORGnBFeEi2
za7`EnH?7x^<e^`V(S#U?xW8t2$V)UedI#$Z$>dJ!l(jEbp|e)hG*1nFRg50|DI-xT
z9lhcB$HN_#KBp1<h1CNT3xZhdDb*rHGHDSF1Q8>@!ly3TR+p`FjPKQB%Xz|TLR4#N
z4Ee_F0k5|g4qtyU^f5&);V`q{SZZv>3U0)G^UbyM@5yeQBuZm{k&4FD<Cy88*EbFt
z{CgN{rB8{8b~Sz!Kd*jq&$_qez0o6Ge3S-JwQyLH$^4RN^`-B$XIkww^8zUM^`}xS
zMTU7aD-yTthsUoCNK`EO8pJoGvaKWSoZ53wZL7)Jo9<tpPha{FHB5;-LCdD}74-m(
zvz}8KD(6gi!>nNMpfEqvh8Dlq$S?H>_bnn^*I6!zK@S_6kY?OiRA;O~1_!+{cTKjv
z%#*<owu>bB!G-QxYSbmH{=T&*;aDLtv!5Bpi<5sP)4?9)q~7?56HnboN0-JVYnY9n
zQzZ$`R_<32hCK_9${G9&#Yy>8#o>uzQ+R$&eNzVU&S^M)auQ{N9&E#1|Fp&&hSXZ0
zH4a~Z58r&MiqHI_P6f?6$K*VLU)EdZxykzUPcLF8&5=YoMXRc&RU`%ev`o3&uT@_1
z!wyYt3Q4fBk!oabok5?ER&$K-wTiKsWPKwmzXHoelQbwCvuNOLd(els$|?W$+E>Co
zuFbmGzp71Q^4*6b70Ib`m8^jl?oiDG*w+vESF^)P?HUC0_6=Tpbi56ZUvwB>`ilmV
z_`Mu<u|eg0BipV}kc#0*+V=FyZg|h;h_ict8+RYxjb7ql9NHl*;ABEI*pb6=T0FE<
zKCXT;2AaZMZ=QZ_f$@yWoG!<=3M<AU`^jDHSp=(?hE(Ta4RtDkg3HPO>#N1n+<(r?
zcn+|2mf*C^y!Fq$!i9xvw#^<Z?}c0g2)N=!2c-d)Q?Sx`qk7$9Wc08YU&cO4C>s&!
zg;RPV7613zu9MoIqzbV=ac#b{#y%skfQZb@qT5wt8?EM6Z=3U_C$4RX|B(~^RZgWa
zzBzQnq<nm?fmHZ&9fvh)OP9Y}%zT@P8Rn>r7+DJkAwMF)xc^gwQwiUihFST6;ssfg
z2j+W~!kO|X$G1GP79+u4KkM}es?T`fr4t*MYJeWTE3_hor}%=3$WpfCZJJdbmaI|T
zm!fz9Z^LB@uLi50hM%<@2RzO+m{ZBh5~Xi%Zl%&$*OED3Y`d=I@@IEN{4U_HwOdWH
z)iBD5%bc~(<cl>(`=oCum7R-`p<L@`_<XS3c38rFXlt`OK6=F66Ws=G2p9Z&!l@py
zQRGyWF#il1@mjhe$?%}V6vVDqnfXxSD?H^ZfkA9-*(eV~V#F|y*E|OvR#T|Or(^Z&
zZxhBxrV3vn4mb&v*nJ+NV!LtvS+nq_BbW#E%_N$cC&e0aslZf%CqEl<g5E@4EmhJ0
z)fSq=odvc`_frZ|=RaYI_d79l9pDRDoEv@mM8ziAPG~qdxYA5$c__cI^g~%1jPwGM
zj)*r(wbjKur=sNt1ZQr=*q{=}5v`bhsj2GXzL*A8nI1gqq4x$&>y19QM7e!a+)ZQc
zvvx|{gq^BZYjQGGz$#}0Vi*%OQ-gMeP#%jSKgT<Q;EO3_I}sW`B>Qj_HW9;x(u;A6
z;;8o3`^{h*qI$$ug>o<RH^2cenIYD7XItQk`*a`S3Al%I`@k<LUC(50Pmt_zvXdx6
z^=!wHdPx_~;}bP@t4}OTY5LPJ*1UWL93{~MP-E+**Ibtz{hqtSVB;cFd|G%a(JThn
zf}DtVR3SKdp{~h*zH^7W4Sq5mfRFrehY!wsq}$&HiNtIJ*F4HtaUypohQXW0-z>K?
zHNyKkHAhqK|6btph<&^H%Zk~*;!>belE1gf*B=2^Fup3fgE^kc5CO0GfGozh6t32S
zG!zw;y4<j}5OSYnh^9u-TNZ7ehH0#_8IYEc%o*tCLK-R$0ZYqEmakp65~ZH<h>^`G
zovv30{h_VOg{4k4lWHk4D)Kj!OMZ?(jbq{nTdJVK(w?uD+4U&0n1Y+ZY|^z<d8X?f
zsTjfp0YA$b#BYw48Y{a<YRZFBDZC7&LwaSg;^?f~IWL#NZ@CE%{TU@fT4lT0U*2Yo
zR7P+RtspXzQH-ioONB!fQr~kw-+E-a?UI`fe(yvo4ZSOFqA8gex6(yGyD?~p=AN!#
z9hXv0STP$hf!79JtXwp!^(G|^^*;8B`L-y0Y)>EmELRYZ18O%Scu?cAq7-l};dS~b
zi6??)-0~ekCY%YA8f{a$Gx8>$)k7hi?KMGJ4A+HIEU%TI*RzTh`0OYWtZ;<z7Sl6}
ze3~r`&$dc`qmTFWS(?aI5l<hbU?H?@adLdak!TaGjP;>9QJKpLQmZI2o2Mp`i*iOY
zFWaH?gMpW7FMNa$>b}O?AI-pR&Odv2Wt*(!G$J75oUuQqJ_4A~@xRizOqu^Pb6^72
zQye#$O?O6*M#ppBmw1kN{at7ob|rtFZTCKrr>V5M!YhW>k@4)_v1d?c6%a7-CQQB_
zh=5Mg%EgEL=oKWC=7&bs*S%ailDt|jJf;o&M?WT+ayZlhQMQD|9Ahz0M;yurhw0c=
zeW|MpZxszhIym*Kla9&}8%)M!m^QKogSqoDnwDpDH}6#`b;?{tt<d|-ss}nAKk89p
z{jl+EAxoLd1p1_xfo_n|8p#}rmAVrIZq_@JX)U$)eM+m(F2eg$lFX+&HHCm2-16s_
ztZt?y{&z<KoJ4mMsfD!b28%t#&)_;e`5h}@lP(vg<hz9vmjN59b6d98=2&^BU9BY=
z=gv5+kcML<&f7&s_zpCJ)YlwKQ=g~DAgmA<@F41WP;Smr$@wRPIjPjQ$uY4<FKObB
znzB{<i=48V28#+-8lUd8)j=}vCVHodi=q_VkLznEt&4KX^S$x;<%>^t%GBwqOisdm
zDJ&Dje$SQ|5(Mx{1RBk~xp$&A$hqS&Q6#CtcnjONbEmpr?5y^R$eL2ws$Gq2(4YQh
zl_9$&44K0>#*}kP`AM6v21`&y!Ie%$CrglN_g^7{wAo7<E{xChQWbv{-C)E9Bz)2+
zDpCYZpx0og+XNKlVJWp?`wDv`(|@~tMI0o)^s_Qz2xr>fVW!@+lsntlZ8H(oaOR3I
zJml~chA=$;FlJ7})p;6+TJ+Wx0i<UL^Ve|Z)K6fa>c%XtxtGWz+yHib-Bc+w`(^f9
z`(+HCNZ9@(pq%6+nPbc1nRi}L>oPlto%8u&p5AH7NG`{NUakUZl6>O8Lz(g7<vX+3
zNV967k;6C?%%Tx=0KOcF5zLBVDH#ZlP)b;y5i7df%fC(lT$RuduX0I*Sh9-ER7|oS
zF=k13?a!SecR0R2km#L)B<Bmj?_{2+3gM|-;=(-RwsY9|QD_YwXxBHc^?5sn_DrLj
zq!DU!q0&nFcI7B<nZ=fBRH)#VGdCe;X-6jDZfquqmOD_Tfwk3}^<J-(2ikmSYcGjt
zK9yyf?4DnKmkGjrCae|qK$Bsr`Q1<>j+_yeT-b@5l-R}_yv2?u8bgk~+*xIlW>8`0
z@Tu#jdxY_rPvRjvXR}@h4-6GB%3V(6-!9B>(q5sAP@g3V+i{eJrSxiX6{(4-hh`bB
znQaJi4wW~ad$BAj%g=L<ECX+ipPpjI>himdc>R5#0O>TDc6vkKCTVCf;;W99g?A2i
zi-fasKAgVX(`Uq-m*1s#6qnZDk}LaaQkIm3G`mpu+<EHkM)0khCoos=yGtcn$ff{G
zT=txLKddnfJ}}Ko8`YxDntgi|j(67)@X_h+yr0b$Mccxkl=j09M>+-0%*O9&l>iya
zNRmBwU3(Kq3wdCuf7F!cX4^g;&hCIpi{&e#&~*Srk;~0f1zKC$3+klYQ?9iG)|~3)
zxj0SuqO%sc!F`Zgib(chRKtMZH$_#`)yM4GyJ+*UneRtag~y+U6MBi{Iw|Rl`^A23
z$5f`!y6-3<=7Yl4?VoQ*tb}8Gu-~ij>Bx@}zRvt(lrGcB8`gZjG!sSFXkb>5bTT#D
zuTUW_$J=Mh6_*Dmq(qx2(D=^W(4??26i_wa??V;n0zzF^O$j1h3prw4CwHfwLkh3M
zcBI{30xmc|*0r|eV2q?W>N@6CV@4B;o$cb`9z0+#b}cSE<tzZ^th>vSs$O7O8q?#G
zdi2;*fmRchmhi$Ttxn7w=ro63r{5=ZBB5p>G3P8?+0lugAtsrgmxD=4#E1_L?}~R*
zscrRr-IlXKuVq&loRACh4pdx7cum>VJ2F(&T3xomU7+{ZMz<edovmH$(f2IvknR*2
zmi1Q$o}JZ&$e$zJWJ%2ejObyeiK52}<Eizw4%10Vd!)}&3*;z{d`bOMouspE4y46N
zvw69OB*FSl!m;X{S$=F6lxIruie1mzoX1(p)HY`x>Fi<W3X=jjqHap_NfoPjKI>SA
z)p4Dn8$;G-)WU)$_$AuZLVb5+Fiu3~m39@y8Og`Jb08%J{Jcr)KI;=Q*~8r{zl0pI
zZgKpM>0k-N7z>!c0;hg9jOROjBu}s+c5rWO=DZAezhCs{4mkUE;n~2lR<A0bs^BcC
z`i<etvs|eTlBSh7g;C^ve0W+VTZB3FM{{U`{V{y5u82}@e7FUzlP1tJyO2|w(*fFT
zCWkvh;cASv&%f`VeFN6zyu=_E4G+wne2vityx*-Xm1YXg+5{DldB19(ZXPsj@~u(x
zPw%Sy`%{DL&=IktI)7O?^RFy|Kr$pklgK{1u;iZEQDbT$9kkIx_iaEO0TOyY9welR
zZ0?VS%977|0tF^ig5mhmig=)IU({oYIQ11lDibyJSYc@``EPS@Fa1X6wQ4JVEj*C)
z3pcTq57Ja3YV1|kxW9X%DD-gfW&vI#3^y}JiJ%{%3qY2}dZ>$n6rH{4ZLQ?@KntI4
zn67D$Q<A}AvX%J$3PPrLstIAh#l}mMxISoirgtj4f9i@Si@yTxNgplGahflGUKBs;
z<Zfy;Ucf8xwAbsEaCVtmc8G#qJBiqe_61YKFwO`*5;N`qb8G<}Q!e!yo&DclF|$}W
z6PSrJ^$)wJLDQTU$`0fS0+I!XiHT_5Zd=m{&(qD98@&~op@wo%*-B!oNsCAYNyjT`
zRpeK2Y4puuubwun<4HS{8YBsa^c6&8fL7@z#hNt2VTKvxHB1U7dLhnNC%e$zVh)sg
z7Fla0g0sj0n)!Ve(&w#VOe+4_758&t76T-1d>G$PrI=haV~Hb<k$t1@(2i-%rxTN%
z6#Upi8n_0e4YIRDa?QSpcE%s)8|Zn4J;ivNJV$-M-b^8%`rW=fx~h0-gbcxcnt{T6
zbnDiUtX<k^qiK`>N=-xZv$hzVTRu8CUHAwN5e<BOXKw`MwG&NI=92pN-T#Uqk22tb
zZ+Rx4lUw88FHL(MtkpJ>*l2+3KRp{~8*7QrkFo)kB<gw>)!3^TYx$L75SjU>)TaCu
zkPRYG6(I9^%avh{Pi84G$et>DVfjQ(BnRd8+1#walJoI!OL85C<$w{%ICr^K)t8on
zi0fjmNLz>S17xIomR^G|;}>=sW;Y_$kUJQ|59MzZFjk-&UnnhA_=~v8tvX&FNaqQr
zRiFkU<xo%sow6hYkD8dCQtVadWNRAp2>`Frg5<!cqPM09R)ISnr~u&Z_*+N;arIW;
zN{+`9glwxqO<+|>58QS3+r);0Hb~teBay9czg3evljW|M^#&BANGwhW=o)#)MH^UM
zKU~}uj>u@Abzq0uh+44@_NL+S<Ha88vB`ez9(or0^1aIQq4IN)&c>%YMN*j7_<Url
zlcpYzhT9M(h|ULc&6BEr^f#08b(4>#_)1%x`kLXv(dy}36(u<G66%ISLS#Rv3ExCm
z7U)TbaVl|IiSXNkjEuK=Q#{{-yib(sLN0hpPqs_N6%PDQZ|EC6{@r+xLJ0AwCciiT
zlgmg#PDAd++7rE}uG;+OE8H17g)7SwSDst46WQ~rKhGtrpHUjQA6ozU>91poRQI$*
zQFVPLhB5MzR5L-ZM+7My<L;LhoWF1xfJY0Cjus<1w1D%u<Ra7SXBvDPR)R!nari+l
zlFXBF$%8y9(P(jVNpH@Cy`VY%6Wri0y9vr?UXvuj<u`zim$Bsy@O*d9mL$zR771*~
zq7)6ch_ifM@1{*Oej0G3pHBHHIZ-Mj2CCf$#AEl7r8$Ekv(qLTwj)FrZ}B?{l<|Ib
z<SCBhnYcE)<@u?#Fkx$VOnbjiw6>NA&hXIGMNe|+4fRTvm|wA2Jn^3P1+Fb<Y!BRi
z(gCy-^C<o@nN=%G<#vC}9X(@@EBhJN&&CG}EJc_pl#O3cBR0?iNp&yVHc|J{H{S$o
zTzAM*ln15dZ7Lo&6$A{&4E@L>(x5KXG`aLzE;wXAPV`FW$*@YklzO7op%rjU8wXh;
zdNg(7xofvnzho7J$wQ7l3O-8tE|ag)+^%e_Cw*sXlIq*kcmX%_p9&%=r4EQA)hzik
zWvcvdlr4-s9LjP-NLUhtlK;#$eyxm<7*vZ7pJmc~>Qp67f#EAOhpB~|(a!ixIEB6z
z_Gu7{P3RdW=fjLa>WDyK$o*S2!2TvfVyGqgXj|sdT@S_Rs(#KhZ!Ll=fvQIHRUBi5
z$-b=CThA#pnQ)d-`2nu>+iK+tGOKI_`@IN2y;o82(f>787OU_k-&|DEgxvaDU7dm{
z18)B;ED{ZUi{2q5JbrYGYXpCV1Ama!LD2HkBdgQ^>=cP{Ieq?DJaR0lZJR7O_(^0A
zP|GjwU?W)2oY9$)Db>1kUNl}R9efqO0MWf6wpHSz3%5E;<<G@(!u6587*$q3KRLp^
zd&7diWcl?rMlDdkxf$*a->#Ex2SO(u@h`bw7MHKL6xg;2J!h>se=A+kI`Fp@dz**b
zRMb*`m{zMT%5#<G@NtonP$)1?<6{)NJ)K^9hE4YQuu3~kriqpUsJdPpHT{Wf^afhA
zw#4$Ep;oJ$<|7QIBNz`J&CY)It>7yc^r+_%j0Js(kTY}-Jw*xK$bx@%fT2AQYO{p)
z|EdHK3n)l<TzcY@M3p7Xnk5EeIX#@rJDO6G))vp*UN)}1b9-uZS$0EGeM1sC`&I?>
z8OIGL_DIqn^CDo>i}D~U#Goa{=5q}nk`1@S8=bdybYXTmxB$+?63EeQhq2%aFxByL
zlCd`TnI4BSbuXPQ8lBc5kN&wRPnmA(nw~*+2L9_McqmHYyv0U*JkWJBqLCp{OK7+q
znP@TI&Cud{m?K?G6W1>|Pna&8SBE&ElyFSg@aullo*d?(XoG5y<M$IvxyD!<Qs&3h
zyxl*y&sT!GyK#t`j?7`VH(>@^iE<1hoay;wbF`0z-kq9o5~CR~Xx+re6DL$<2W3!<
zt#xFktf`J9_?AaE*%$e+&NDoy(c5_mdwHI94G&9kEHS7@P^jcfQsma8G1xc0gN^Bo
zZ=E+#Q5uJatt>zc+%`Um#d9oqzHoLUzXWhHJX02Q5@TV$e8fYO!-?<=$^=P*vDbdz
z{Ui{FjTq;aEg9s`t$9wX(kRNAHv&K;_dDJuq3fqkw+%;QNlHw|7<ZrPrn~Ag*ZAW*
zvHA8O&<+@!w#gW97j?CJuaY)k3HVB9=1YC`STU2LN<O#O1Drvcmn{aQXH9+lD9^>U
zJGk?2xGdcG`b<T^#5K0JF=*kzoE_bk9U0>~@bj2x%U6q^pcs4MXrGgl&yB-rqAQdU
zgZJ>Z%K(S1oxBFvusMI9{HTX+$u!7odOiRBMG^^|6Bze?Nn8c`rEN#wJ)713s<x}^
z#%>)8tNlbr@1xG6wEa`3yx%MlDT#6vNpfECa&o_qzM&EuB|E|YGA?kDXyfncKL4-j
z0b>6NKRTq<*_A%DRX)MygDX;ZX`wvo{qz{k&+jH<D@MM$Y-S!?wj5iMSd{Q^W40c4
z3o=REMMZo8MQm083f24#(9*;Mi9{9tmHrRgUJ2rSZVaui3+&5M;=UCi`&mb!C2Ade
zX;Ae-WxPA}3w2o*s#!t%u{es*FgrdwEJg9p<5yL9V`A?MGV!cnc{M|0RaY0hLhlCH
zq%Rpn8mfuM+RRx;pjdV2TJ&^`Q`SVGPp)~8NKh2lUjLEm52H#pU0>`#ZR|=_gFP+@
zM(*20tksM?Hfr0HAC3hAj@s1kUSJiX$^Ry#*6!Q1==fN@1>bp*r2_ue7#RG~4~-i2
zC!&XA;ErhpP&*v}49$lXxUf~T6VF>7sSc|2(v~*^GUF?!?LQ%!SZ=18u+5V15xKLH
zvo=C3PA#jEYKz52l8YgjDaxQdSCe}<#JkaLP;o4b>`MHa(NmaH4?HA~PsaNN1V2R7
z>20@%pK$zd`Qu)sTseNr5bYcgJp9nM8Ot?13KTr!uBGvy0b$7=fS-?R*|P@x4>!}b
zNUgHp8jgSFJ&wMCS}C^0IX>|n^o*9Vwnox`DOo@jUJ-qvftKdlsdVxv=ee4{a@IBR
zwH}&gdzP3JVEiy>wb)?5Y4llx{?<zPZ1!>5>DJL`_z&Up>4qnxeLwtk{56}$s>q_+
ztZ9ZyR!+bojQNY7-z&SM+FjvgnSKirmV@yM!I_z=$MZ&a!k@<$Gv$w4RCc$}0@2nX
z^Xyme5&ni}q-zj#Y^29l5)D(t)a%3%zmw}w&O+{3h<AzY*GUzwtf5l*736%*xN1tu
z>-W3<MC1LYn(o?qS`%ZVy1Zv!x3np==13IWDZT-f=wrZaDBf|bZnZrZBw8>Y#M`?2
z4ro?!j_VKmsGKhjz*utgR_IbSI7Jy8kI((nhhn%<H+<)Phj&KV8dMC0Mhm|FJ2j8T
zMskC{63hAT@fJTOg!wF-y%9xw9>uUHYgnE6Cmtf&%J97GFGfUL((?Z;+A>hyHws%*
zI{<PXukTyRoShy$gu>OzW74Lq(k)v{X5m;!)tm8$!~p1&_N}uXpOy>N=pLgG5K)KW
zjZTdx${Q8UOLs|FTBW-HF%4>i*|(!S2GBd`GutRJ`{omdSG1V^fnBD3RAnj^<@_zI
z!qHrut$vt>LZG4WSIgX5SNnA!5bECK8%(&LAIkB&=$+aya|6{ahdDy{rZ?2G(O*Vc
zt(yAdnD~2G03&h3+Ge%BYf5e$9>Eg5c@S1+Ukf*1m|pF`pW2<`xCgcU)>U{mi|lMp
zB&|uT{q!07G{H;bu8v7L&y0|@8GI<4Yt5wn;wM&pR?A&i0?)A!OXP+mY0R|TD70c1
zo@+^lrd*?_%|k6IPb;-9H=9ntsE<afb%35$y6ut|Dg6(}b=2N-4-VPPt{;y5!;*e_
z2G;3j8*P$8_vH_t?j<CyBKTJ(AizBsdG6+(wRV_VaMORp+GfTt9p7$NzIe>>i*WR-
zfo<NVwJ+Eo^VX}KVIFBWa5?Qpcv7xE>tH69E47b9ws$Nx$^)w@z0m~+uSxXQtT`%}
z8-DnJ87NxAn%;wkb&+VVu;G~x+j3bCTgaZi*6i7=PLT;uq51@bG*e2f2UeV27i(sv
z@-^NAJiXy^ztQC$9AByX%Pdcclc3KTTWf8(xW~3@j1X6=Cv*_FSibzmV7T($-9mZd
z0O=dV`h!`|P;}{mQdWJ!gwsUgT38Fy4zp{Dn!e&l-w%qjsQTPy1J%ZC&vlJvl~RxI
zLK$b~%%aD=RhyoksbeQ;l=GBV4-U)sD{xkY9UmP0V1Bh&ft}ji%9lb3wbh*v`y%(^
z{4VrvD0-a&ZA)G5p-1Lcn(KUJag}-CIzrO#_yKQA40AD?ER_9Fe4q0y(oddJ6zqBu
zh9C0y0=d*>R+QHYq^(!M^MGO@+N+lh;^%`<A~oyyw!bRvM+j#8$cM1n@$MEs_+3E`
z91vR1Bn-!<6tvELHAe}3=Y}!<-ss$AiupmzZuLv`TY7*(MmNW!T=brfBC+QU9Q=NN
zcc!L%zET(Z6Gb$>WBL6_e|FbL{fuL8#Jm)7(tJt+lIKqa^07tW!M{U`HAtbIe0Cg7
zuNQif&r5<^23(D7J0#f%O&Q6}_tC><!K@Mw;{I?28CzSlxXE5}8(kL7LPFt@_u^J@
zU0_HmtZ=p#gdk7tiM?@SDp>BM&U(sYs=<HWb@_pNTNLdb6xFlg2@#C&J*RY%Z|w#W
zR1jhYfg%lCrPe_LfI=?{QQND;!&nM9`?m$a4s)eMw-r%zn8EU^ft0;ua)X~yDrQ8H
zxdf2xM~^7zBpb43Y5Js2L(S4rrigrzsEdfXS%cJshYKRAY>BJ_rIr@&2^ZlT-;-f!
zupxlO8gh;PQM(Jd(2j7XG0A>P!FVY7GE@-B6{nBHJ`@T{nHaaw`1_x!I1MjvJ+O})
z3j%*L(>s*fKG0#f5Qn}8r9G`|Sp{uB)`%*{>3Oo=SYH9Y?y&6$_9l|sJ_4V$L39|_
zQ4e&C%@8KvCp~$`85sS7$4r?PtzmpT?prW3j*w0VJj!FdHEio>iQ(!Q<5Ci|>Zh93
zR9HFI>|uK>$u3lvX`jJ1&e1#V@y}?(sjZR>JbG0hl#}ydPyOO>wKFJ1F@0ma_Q1j)
z`c~{hj6!nG+f*AM$ny(}&xA<sAiZ;rQJKFoD_~4Mg|!u?_Ozi&U8D8BW5P9{Ce~(X
zN8d*+Jpp*`$}>BBN1G)<+-za%;`FS?j9?32N_sKePC4(Ftk%MK`Mv0pH0t>vh`Y_)
zxZ<cy`dAI&r&lx*kkooq*EDg=KEZXutLM8&{krbkdsYT(&l!N3eBMx!8|qvl-vrKU
z>=6k^N@(Y1hGDhkpJ&}@O*Ud*yVWTlt`gMK>PtQkPmA8z=OCje8P$--8}=8{B-D}!
zq}zY_x+6bT*S&LZW5=W57$+rPkSZC8x>EX16-fwx&HQy6Xm|c*89*=QU8lNYZ#dna
z<vK|FM|0j|2Vlx8cH0FO{}&ts$h)BTFn0AxSRb)d#S;j`WRHrPX4$wlDIBmc%q?yT
zy@O2MGiTST-PH{czl~(-2^0MGl1j3!$K+#H!Lt2sv<}UhA;GXKTm5!e8SNczm2I6u
z0fz3W{!c~~7b{naFe#e3jc>I+nOtt6ke)CfPNXC2M4>H`6DIRy^<?y6%QcK_%($1L
ze`57IicdQ&B-8MwkS^~9P@z5I2ri8~)?i3u9wr>9*9jmxM%B%EuMic^hTl(E|G=Iy
zDJ>m+84V*kH~E*K`evB0^i?lu2Q#mc^6R0{dWCFQL_pvbyy8UJgduJ-cAKg~gI`eH
z7jmcHJTmIZjm>=6w)y^je%axPrNPqrUE!!qtY!CE2PX0{O^tGCieFACi^Npm2)1XQ
zuo8L2&7pi(c~g!pvu)!$oip|bfqY97C*e+cU(SXd1><|8CUhxZV#mKZ^5#S%yY4ot
z5KK>x|EdT1KAdsIK5nN{YXQd|7fGa4>bO32W-Lk6VT9N-)Q797@Xy+{{&1P-bM$=%
z*Dcbj6+G-)mTvU$F5=4e&KcBshrS=TryOT-?$s$KrLzV2&bvI)e@q)`6!OY~7(Va$
z&RtaL-}y%J3~{5oj-c0hj(QoizF#oG(;FeO8nKe)Npe?;``eMaPtElLF@$|_SvLD$
zyq13=-QTg#f5P712X>hE7kw<AMxh=f%tw~i`H5PP9xI|eXOQDrVBni#!N-4CgY+Ld
z#J^8pmjj|^*JtI#e*^0O_R4?XMg<GSC8rtqVYjr;A;_SWghlUCUX~T(BeCG;O=Rbv
zKW8Yn?PC4<KVIZh93n*L2*hUpKOy@6{fGbaDMHa@?|D~j7X$}oaa+pin@12boEkn1
z-~$1iBq87pO#i)#3d8(W#{YbwokZ{kOWafV|Kn}{KOdCSA&rY)JNWFrtLOuT51CD#
z=QrkTIr^6Ykhd!FI5qIY#*u!HVe2x@e>J23s5#OQjGr_&Wmm#~&zS%11Kg+n-~J`k
z9`Q9y^Y5j*{ICD#e}DGB|3X3!+T3Jii&{qi>6}4AnZtX)o}-xhD44PfT2GZi{{|9<
zhxb2b<408W5h76|WEc<x*h$~6D54L@%9#}YZ$Gu!Fdq2C!@LTU{}JPJy@%umUBOp@
z1?lllgHH!QYb3tZVxU`Qt*Vv20OS*VL?wxiZW?i~vFJZ6fY1!u!-j%ewp`aiQtH7S
zoS(pE*Bn#;9Kk+F)ysgKyb99HR}ldqRfxRYF6cfpWRx~zq0*?bWmcuU-I4#~&@h$}
zl(_en-`&UsZcfd`f${}NP+fq7P6QB%^CGg;b-(7HtH3>+JK(I+)Nx4|LDm79^}cV;
z|A(-*4vRC3x;=rQ!QI`1ySoPqp*y%18Z5ZGy9NyqoFq5|*TOAm&`<<-E8OR;bob2t
zX68Qk&+r5&s?K}PK6~x;Te<&iseeAg|6lHwv;^>Wev8b95dHhHPO=~XR$t9!B4xA(
zfDjO{0F1iz0-&mm)w^TQE^<eJf7o~edj)2Zlpd(CB#fmhZ0O*i<W(&<a4y1z8zNv7
z9CRUo`bSg5=eEi;LG$nkWCk9KI0DYl4>vHJ5|D-Xn72y0jq-<@1H5(TYXGFjchIJ)
zVfwWh5P^MsqgE|MTR`=p8~zU$OX^__yBCPA!2>jwse^;>`Jtzt08%6jMl%V%dwM(x
zl!C!jzEU37AO;aO!qVeGZZ!V(Kngispqsocw@`KK{?}6={WC4V59oVc_~ig)t?J1R
zv9^S&`#p?y;|27ce#I;RljX@$OXuEF4ptMvnyDx-dvYY;%@v%?FddI?H#7}K0^0$-
zgPLKcoUpc&PO+p3C$C=eQZ%}vtg){?BVNwc9asBoT0oAVq3~vs92kJdk^?;7nojVL
ziTQ&kfLo1Neb^Q<h5(?H{4f|i<d$Ca|Ft3gf1cc6GGGVPc_H5M3h|$X``H$;Cdvhd
z7Xp;&oi)JU?%wo%!vEN{7XVBx)&L`2H#;?8Xv>8Ej|p!qXdl4RK77{;!;T10mUQ;g
zJ0@krFh!>qGn$dWuu;cJP3d}UELW!&j7<gxpg{+Dsh&UrZy2mrGISjmkm%wY!+ruX
z++PC!-QNrF&=^b=|B^cE!`DDo-SP&S0hq-dXu&~<Q&@ES{{!Enj<}*t0(`{zxeupD
zPJzpn#t;}`=x23+yXHv(SPozSIo|*->P4cTu#Y*oJ>n+TbnWPq>kl+oEb+4I9gi2{
zZXlk>Js?$sesgUtpXXhRpZLP`^eSBSON_UCA%BkoFj`5#)D`F~t2Yn#nrgRfoE^<d
z-}nx2+-`e&0{9V;Mbf5982q6ZaJ;Yk;~X2(KFW*$s}!>rz#&O+IFo&p-?>Z^mV*=r
zOa~hKU;Z>(w9KINgqf1ZJ^)Z6u{|dsmxKEHzMt4ej0PLDmH0$HeKKl%Rkseh`#%EI
z{DF?YNq+a<AItQg4pA`NyTO`}I^Zk`1FEYZ!-0M%btZLR5t9?7EPze<%=&r(gxxgM
z8*);EW@|R36V`S0spl>I+~YeV_n%kflmQMvh3*)<2k|{QI-TA9MfX_Ds0UBK4I9$$
z1<FcN2oMVBySPo(mu9T2D8Bs7bnj)zs0IGh+XosM`<fnPv#&rFniCO6-x}l|pxM+J
z5Yay-#o4Sz4Aa?;Wb6fmVW4Y~2O9>gg%dZT25pkd7Y6)t3Amfd#{xK0o*ZBkDTi&h
ziHWo)SO&Qt%-%W<56tSDz=kF|p4OHq@OaZfwIiy&4Zv*xWxvA03wP*%^|-!i*-@<N
zFtW`@P|pv*CsBEB2HuXEAcJ86=K2h)&M0r8U|=siIRmnp5h;xRpalQ-8`&(3NJ92|
z?_Vqt(0U7p<CLV8Fmo<R{lgyX_caW|`WAIMZ}<b)Oa);eEawzM*8(gZ<)r5t{pgR)
zfaY%vreg-LhRKLA{l@m+m2)&Ygt5soeOtNPQBAin@krj??!U$l>uPn*tUB2cr10!W
zdKf}sk6EV@$$$Y!)&S}9<%!RH8yJ9>v4c&pekA3CL{fQZKrf(JcJTlV&7`~IggXkT
z1Cr2VLjULS{XYsA90YA#0NYS%UiC`-pM8KEipsBWa}VIx){a+zf&LSuiYDn&pDi{M
z(sV<Y;hNj+oVNZ-Ai}iSQi2-=OQ#+Na0~cy0ec+0Y2se7;yu-{ge`0S>R{}dCO!+^
z{hO&h)V&Ng=VLU#miHkTi3nAdy;fc~@bPlx(s|L>aH9j*+1lLi)t!8_y<pIwE?R#x
z??V_i3^2%4r&4}pKbG<7c}33YuJ>k8bcN0MET#sQQGwo<(Z0!8>6Dghy0Wcma`%>o
zL*>JK`;3MdQo`S|o&LaUKD+(_a9ce+VW{nol1~Yq?Y%0LB|=J@=3HT2z`kRm;Rr+J
zt<W*O+hRFLQ(zy|+D;Ms{na7^_o=2T>sMCkH3XX8)O<O1(dO0nY6qqfBF4hDRE@gN
zhMfQ{1^uA1DbJvut+V&!7npit;*-cs^rs}j523HgrAg9GG`5P^56#-<*w(5|*t;$(
zomz6NrsKulvgf8S>$3Hdz1j-#xdRM8LeeS63;?m+48$1Qh_4#iP^4>Z1Tdp|0sEQY
zv*dC7ye=qSV~QlkH<an&a@mTg=b+4RE#eLU%bj^-NT#|dh_nIS$6lZ@{d>MPtPv1O
z0Pv1(!Od{No%$~TYqxz0^tX8`fN+tI&b_Vsz>IoN4PYS@lc_oEIlJl?Nod9eV$8!k
zo(E-D07@WbBgDt92@(_QT2VPo%qO`Bge9@vU$z3^Xa=@nOIx-1)_9I-*HCd@EN|d#
zJR!OzVgCJ!5<R{+`Ak}fwKwDpcca)iyq3Z;wief)$I;i;32g_{$x9B1Pb|uldmUbU
z19VZ_Mr{Yen(+d6#8ydxj_1}$K*Mlj*Zjr4$bUuKBTbMbvYvB4;k~!f$$!t~E;{_h
zJgmYpl}s>xJuQW>D>r#x_<W+HIZBZy!k=gVf49f8VYJNBq`!p!JLJgZKcj)E(T`35
ziX+&eX=R%zTh}HMu|I^JlRRrzjtS=W^^^Hd@(6H2Mu#zgN7#Us76-2%U;rKzGbp?M
z{g&ZDX3l5A7c%2=CI$fKnq!rT!f7Q5k?;@@n_xs6=?i*DbDAG;rIO%yh@hwephzD%
zUv4ZV^xGh!gMeMs;HDcz66b?|WZUasq<F~kp#bi2hGR=QX=tklg&G53^K}79-Z%(T
zu6gI5PJtt3ZUgRYV^R(q6F}0xd(vMPDV8(jN~H>GF#Q`1s?4k^z6wl)fPp61x~?3J
z4-EHNX=U*H`LTB1*f*Tg^H=TR+>jw4>0L)_`hm}}7||>~!72d_E5>}9r!|iYpk%aH
zvV2Fnb^@>+TT>a~(5Fc(U4cWk(k&4+>jD6beMmtVAzPQp{hW|I(fmsipeU+tB%=lb
zVIr|*TVkuX38_8XIZmDLNQ%RG6`r&emHr0eTz>Z8nUf)kEs3M50(c~KFQ9(-l|t$s
z>+>=h5jwmhdEN55UmyRHqVNyvQKwv%uGE2sTB}CZCJ%s;!kA7}oPgb;VB&`i{?nZU
zpR4IThfb29t?HN2X8b^aEkOoAsJOgLP!}oQ9s_|CawoA@E1s)CZI@;Qcj$L)Lsh!h
zMMeh<IBPWJ(^rHg`JI_Qu-tCpgjt1sU?fO<&4QCVg5(M)HN`&pL5G8D;UC#9Ka_lY
z=hD8H>S|k<kfI|5@hW25JFl7XcCqz`c<*3J@DS56`y)DA8uqXE+8Li`bAtvj2DwVU
z>?<$!*MaZ`lyQ)4>wS{}PUD!7`vzWidv4&<ZAW})-=O&rJPO6{+n7<aDjDQ4qza)G
zH|<0r2YLo6=qFy;g|Nt<vcG~eRJ0ue^%h#0V5Jv;f}?TdqgDos=iPciG+<z{Yoz;&
zxP+omIC6OssX8-tUI|>&!&t0Wt8g#zU?uzu(I(^KXl|d|ggjgXH27`x-KrD{_gO7T
z6pgMN(L$6lAwkf`02tLWk~+Za0N<Vj)Z1X;hx!RPmezplf@{Yu@TT*2W|ZB`kPw<W
zgtoXb%QyuBCCx39z;=y2L#cQOv*mYUvjA1<AgXBf?p?3NUyZ=-?(uV`_>>kMs1G{e
zC3x*5!gOY{imlJ-?+`kd)u*$S7hV1C!|*jw0a}gBR<``2?>~Pc4nk@ua8XMQt?#7!
zx7su{MCF&c0Xb9G;S#y&zOm*>NAg5<)XfzI$c8IW%1shwbp<nG(b(g%Pc9=Q=+-!P
zG}Syj5z?V_rLO_KR#{(H0bGd9ATi{Vgi~XMoXka>R=~a+u+5i`2E*R4c`m37p5tuS
z3?XX;@ypbit?Lo6qa_=M-8Ty&q7eQ4rk$_F3YQE@?`Yv{o5{slPmWiDhsa|%W=<RJ
zC4195jq}<UvLSR<QHP`4T5I}(79;pGVq@5pZO9J<$vt@*E?F<rI%B58g)me8k!fJa
z<eMamFwF-Us~$3(W?2MFGYp^y84N7HTD%xu@G39?N%kQJWA7LSyutUUfy#Vlc{#Vy
zxEkQgJRwWdPxm$k#T&EW0Jv^`D8wmVD!d#4tQ76iF~MOW3STUO6GV}FqPSV^am;;<
zP+11tC%b?S@V>bU<)AD9$$Dd&Ai_GeJcnV%+re_D2Hb{dOU}uh@*N;23obb$oO~U@
z`PI?y{uuhzZPInILtY!ONtc5a_`6069hX+I?B;+A>Elkif#cpI%Fcy4i~W~-oHd(N
znxEJ?Z+_G;$oYm5t^|~<b_x$BOH#jm2B5<GbIxC|VVa%6fx@Etv*PM&0~E7+80p<E
zl=X-DtQHsyr34||9TgS_km6B3$_R~i^j|aI3vDp3s|6zKrTB%Isyu_eUa<A$A{J6@
zG&5Yt#6^GDT2a^C{hV)EfeLR4sNok6MO;j36_{_I%8Bh4Sm+<7RyR^IAIdsPJBr66
z%3o@cyGUJT)*)!MvZyya_H$nv{4wRTAnM9*aiKY|kgGriv-(%+FkQw#dri~geBnLG
zS6L$3)gg=lf*iHUU#QJeF(1+WM_5#6_C5!O0^`=EqG+w?1B`K6qG)%dZk!a6S^!Px
zwvMGQLp`Gd)ipp+j{8Z5gq$beG|pg_@UQTBd6K`)Jln8nZ-HT3Cy5BEh+8ektr(wF
z-|xSvM7y*g?F+9@12t)$41wamax}@m<~5L61G@%8gr2tMrw%?}Q!tuQBp73MFx<vE
z)~!AjXt&yuFOKCbDPJ_gTxk=8GXjDy3j*_ELa?HFdzpkS()FyIJ8sQmIp$t-9<gjn
zxP_|#oQ01;Iw=}1znlkenbrZ^S!4WPi+j0yAFy+~fwRTqy2=G;q3bY_a&cGW*N_sd
zyxhva%JJs5LrZ!a>mbd;y^rZ?wOy?XR+@kd&S~NFA#<b6y@PjqOY*|heuznZdt_{o
z^tm8^pUKP8T+V^ng)S$}wr$sX1Q;ON_8DPNBqW*txBp=pg;Nt8^8e^~B)o`f$<z*I
zy7Rh=>s}4RmV2&^ava$q97_zm+0d>d(wuSydOU@k&k8x*jVr2B_OfL0-301p`JA^?
zOhHo>n+o(y3I;(@jeM0y0cVlw#znY>4|RGD(TdV>f-(tKJY3Lt^Iy<whh+Nf@C+Lw
z>q6GpBG%#a<*78fDI&>2b^RjNABF1O%g4llr$z>FiGnQ&iknfT6|tq_(82CLqM;&n
zmt|vmYA?tyF~$0T8IMHj;?*IlDfSt8gEwcw@!sW|;4;Ze)&|HFDlVhvC|7W<dd;Yl
zM)EhsO*g~1IIS!?=n6qXW&0>K|CSE_bZJKPgU{I~7tE*a=}9Nj;l5fUt1QdHWSOF5
z<=E<m0PS~U5_bHIp8Rxjn>h*v6j|mR+0#s9z(mDzSI_nM*W^_ph4D;ldw3c$=kbFA
z*5`p=)F07NDBa%h^S-uq%jr%D(heZg6D1?%pVrtif7xR5Zt2yi0x*?&)kjbdG1f|m
zHGkP}^ZLnVq<g)x;K8Box+poBJla6&$mpzh1Fh(bi^8*lL`nYl{^mxOGt4`rV^<1A
zk?UWs5&@0E+H5}V_vZ9ZO7yPYZ77aW$>4S;fpF)B{@!;FYG}vKFfN+GyfAQX;w9jI
zr{5?l%)ERF%$X}V@~?ZHT}?igu9}I1VmBO|`OHf;tc!;m4>}%9E9CNLVi(puEHGXJ
z9bim{YVNA~@DQ6#87H}N`i34$**4v|RbEJrh^y87-QP_|T(PPDjNu>$cf&MA2P%hB
z2Zr;;<TI(RTadBIoH3rx5?EkPmwrjM;hBd@a9y5%`Z6=1P<ts;wJ5SiVM`;?y?roq
zSP<XC&x*iNFc)h;Y*(N(CsSbECM@mUVsYFOQ$H%zmv$HagT4~ieY2iX_T#30p`TpS
z<~d<LF-+&ZKsSf^={JhhsnjyTj=Qh;-Df87E^8elHtE!eyJK;#6#b>*a`Dtvu>(^=
zqwT^8`S8Tf>aB3s(Hib!k1lx!7zV}Llb4EwFcB}kbqh^%xKRCV$&hbD9&kRWxc*55
z#g%@!d3Wpkdo>dwo-yij%`X!m4Y0t7R;yKx&_RHe4PY0@IEa4x$ctyvpziB%_B}pg
zN|0r6c9085(wY){)g+;P)mzcsBKbqGJAEauao#>PJh4f1FOSqjYj6s8va{fusZnTA
z^5P?v1H9|%s{#sgy`S#XyVklW1>>KkxaD5v2`O$?_^edV<SifzpRkeUpHf{S<i6$#
zp_&%VX}Zgw#RjH0-_Hqr4Rc$F=(UmeG|~L(qcQ)lU4k36ruX%K2f8#W1}TC%&Qs=%
zoTQA%EzlQ>qn)cYHEmmXh33lwcwQaJgX58O!JsuNJWc21CX5038xXHuy1QTSQUx@g
zRx_<*v{3wYoD#Kv!(pQdXd2BsSplcsPX|b)qj{*6b|8v)MVx;Q_og#DT*##1xHu}j
z8}1et!Lk~iVS~21SUCDLBk5)p_fSg|NH)h=bn1FE%z}264dgMV1Rc#KTE&kOQ;uba
zr}mE_9V`&nv2DMk&HgSiaV~E>?AWBCwN!hfNLu=78{rEM5@t*WFP*IqQA)BNiQAT1
ztKj=V!0N|TgP53>pTrj1_dAUEH3&5m$kNK2p^`;V_~LY203zL+U=JC90kTNK3mzm!
zfGY{$L?V84{s~47bGoBS>+fZ;a-Ul!txwne^N<1jPH3!qNaGZHXv0svrN5RGUBzZ&
zT(#ZP;7DWNV7r<XTmiV-qMh%0T%<F`I}=!=EqX3c9xC~M!|oWtsQx+gu2d-<4+z^@
zV!$x_nxZSAk%}a(ACV+YPE!8g?ktHYm(fTB^>YW7AugkBaT#BWHI_{XORN6a&X9wp
z=|pCZWp+z3g}(7Zuv{?lr2lUF(GFoBzY&j!=VD*ttm#GoJodMSjTwTO{4<W{98ph?
zFD=5p)W}rsxp6;P$MLKn@*>!l8J0@9sl+``J0;l59`RRG@jK%1zxA6E(b8Oa=-!qP
zOu3?e+zoxA_O(U`G|`%S9t<RFY)&d|*&w3JMt>bccT9$cnoSo)6Y`skCWa+e`OBAb
z8?^UzQOz_8HZ<8Xg;wu5ZL5l<)EQOl{p;<Y@~_X|AKM-K?R@e*?$~L2%G<fT3pnZ6
z0Sq4JSy|RrdADPHm%am0MG5r>P3^BsV9l2*0~R?!$vMjbMXX%@^&9TQ!b5```Zh={
zX%(oA)@1Ct9A&|fwkHkzX=j7_xp|({o;GY|s-UCzGxRq8Mq>WH;BlDFdjY5{DcDX&
zax&b&Bb#J>-z#4&&}{W{90BOSJQKkoKCV|tV#H<(rwK<{@d!8*R%ot5+i?Os?sKjN
zw~~$exkGCkm7SuMpi=)A<npWft1{iQUl28dQi;bjns>!Mlm_LM3pu`F!nvg(n9E5Y
zMO`faq5r~zb6VBlb1nwsgABhPF!5=uC~`>&70kiYC#N3`vUf>W>XC07`++}le<p#0
ziii*_ZTg6V3SPZ4qjtiA{gVdxswgf==xL-g9K94G0a3u2PiYY(B|*&HoFXDvuaAkT
zrbN0lnBVheEwU_KsgBqFDWM<{jke9eouJw`k(8I!l6M@8EH-b2mZ}Yn7Rr*OJyJ8(
zQ=_6&8{_$H=9V1$2$5nsQENsMk|*N%;phRT7Rz;ylc}bCvB4sIbyQdUz%5*d6Iv!g
z^!)}jUzFf}xiV)&wdJyCkrI^Pe%1P7DtIN{$fEXvB0tH7=5`jjnOE&8V1U&6MppVr
zgctuQ;t)EgHSI{VPoKMeU3poSpyy1lv+8?~8VE8SZgcgn8*Nmz$zdOv<$Whsc5EsS
zA{J#(r$o<o`{d&TfbkEEDQMx}Mtm=*(U?%Yl?XeHEO0~3vd8t;+2Gcs0R3dU<BBr%
zm+cb#>*Tt>p6t4gHnYD;UlN*}$Uylqjp}}3u<C}bBPC&qSnj45Uy-8$UmQ6DAuCw&
zGs8ptPepI+qXvGB`huX&7cSumJ4^gW_?iXmjfLl@yoOJEHgqOlne*CfZ-9<OdKzvH
z$6Z)m7=iLaRX9O==dR681EBhd%H;Wg&8csFuj=z9T*YhRj!#s5f1&^RFe`z)`k>BD
zCxDi6I$w|Tsrvif)H|{)W*#mp$Lve@BY~z%ig{tuplc4fo8`mvpK;^whLwIyDBe+s
zLKc*5IcHKjZa|u$7jSw-klD8pCwGz)zpMl{c4OMybDjCm(T5Z{3h9Lp0Z+ch%dpJ4
zkBrN%B%UDOkk2c47hFE><(PGZgNn_KL}5~gV=WV*Z(>Y)f6PZXRzXm`II@wK#)0nS
zCm5q!D!4Qua{T;LV`x)vw)nfq_|+9u)06HT-oBBlaRTx9yxu*T3{314Dsu9BadOav
zg0+WtX#a79`WMKMPHEovEZ*G37Z*w23D;$wFZO7b$nB`>HLAK3LdGe%0+;1{*+*$Q
zQDDotW(v0%3vKo@E5kp_9k#v&aS&Xlob-bPNh5u<OQJi6bF%il!mWf!Y1JY2YO$S6
zbsuecZI+VXi(>t>NaIb0Q^(hx>8rZ+^FE_I@rsOY1}%A8L6fYPFm)iu4^`m-GlexP
z8&ub){5J4leh~r6^W|F%A$b=5mQ*3!Lf{8>n6-RSz?orKY&fsttMI5;1T?(&OaAku
zrR}uXqNOW4OrY4|853{2eoysI^-bcLGoOT+7`UHEM}Aj+(Qp=`g7vM-Pplz18D$tq
zM{wj1U*k}j5YqN`EmG&Bt=?O5^LU%_;XbKFBomsI6B|AQXmsn@KvEsGSY)Qq(CC{z
zcS9wVEM;Zmc^&!FKw_ZN!?jq|cD%oE;6<3K1CLKSwVcv^?JluBuOKVnjeK#(POs&p
zODdzLM6d`rjV+;9LTGF$5z)U*^`p-3Q_17GmR9WBU6&gj8iCfTfNaGoBQBpI^Q1<b
z{2*ag0^IxbK@wdk#4vB^_B~w;e=413f(Ib@X2-c}Uk}>vG+19$Dzld=Xs2q8Um}R=
z4DiS`6;-<*4*o?ZxSk07#PQ9R*&*$WA?f!*01i(bd8djdz#n)3m4yF&0GBAY78!TZ
z_X+*?6!DYQ)MFi)u<@CuAOC{7ja5%ES%~KuDsfHCM|bb~Qwt63#!lTs*-dx1-%j?3
z>JLmh>v3y-)={|0xp)LMPCZabl44I;$3E<y%xH-*f&54<<O0excuS{CF;g@82yF>k
zer!6=E|;_?Ry41=%jD3h#lD9ew}C$x4c^@8zP1p<x2@+YpJ${LvnWP0(u!$+n>o}{
zML!iCcT18}ShUwWm6KuUqO#)^)SN!`XsN&M3tNN;2k~&Nn(~hLW&`qq^+2{v_q~n7
zr*k`fQvqB~D<<vO?;e2YlQRTmO01ULv8s|cnHSh0cJ9Bo-L8wXg}KwHSLeAvoE^_k
z_r6j_5Kx6Q#l;$w4zxsI360T4&hpC`e9rPf9LS4P*&H64=_k?e^E<h3z@69$NJWk1
zqU$2R!VdeefGls^080j7I<bhL40+PFaLZRBO1zurIhW}wo57<<eV}q6Id?|MmqWV8
z^frA`>l*r`lOqR!=18i)sb%z=+uRYM3Awdv4UYec5fR>?aj>c-hnz-~#=k4Y3^L9G
zqOSy2tMaKggxqm`NkByV4?vEo9jtKUK`C*Zpg$jTXkDkIPpE2cs~QF6Q#$wIR%w$I
zWe8%~3y(c2lsZdHTeHcSa4Ypt(Shp^MC|2fohXS9Ee*<oZxLH*xvbxBq#Q(0VEN*B
zsN&*3;vI!T^)QF8C`ODC?dI)7p|0T$*kVzFOl?MR{t&z+p)*_hlqOg3o|i$j_EP)G
z;tyBD*!{5w@%*?BQqLmKT1q}gB@=VDHbCc)AboBcxUdu-^_yoENbL;0;^aND;FM_}
zGl^U(AXZ1Y?G;qcI+b**X|MOr8T`Q@8({6d+lN~nxCr@c-|yJgHEZ7c&f9)$p>*$B
zwWmr=qL6yW2$r3~F;AZS`jPa`V_R#9#za!|GS^3?OXsEKqIINFBvf+x&C9K7`UvI>
z5^K|)SK$w7t)JVQraFsyyF+|9C!U!?EUoeOoOV4)L*1eUqp~&dcyX_?Wg0|Qc8;JJ
z`x_~uI3R-US-M~V^gY*<)5RKi!F{Qncb#Ll-4Yd_%fsZJ#{<eYU7p*vnxDA{rTyp7
zM<7Rp$4rv@w*-I{2BEawSUl~y7ih*GEu@vJ11`VahqMq&iGOpJ$8JLsIgTvy6Y2a*
z@e@Ol-^uweizZ@Ylr#1qp^x5(*eYJ_>zPFDNFUmh;*g%&xTxB%YSA^&`~7>EuH}of
z_5!G|0|Ca>!)BJAN%w2H#=ZoxPV8H9A2l$YSnD8^zx#(YX#rE<gE4vmB)BXNHMe?W
zh!~s($sOtDea}F8Jp&6O*plhkj(SPDjW;;<;TKk0P^8ZdsEJ8QoXTUtEMSvR%<?_~
z(0?cnx)T)lciCG2q@{#Z@izie#fMg+^O4c<6EEx+sh#u<qtMNJ{b*FN&ffYHe*mcv
z2U>o~hn(J*(;+@wv*;6VvXOb79@+KNX6G<Zf%wBFQXaj6{FUe^?wVK0q1b?kwQ0u)
ze^+2sq?EXw3&T9Sp8V=h=O6fkx|cCes)}c#tlC|`OEbXk$mu}ZqE8+P)9#wg2m0UR
zn~9xd)oHvGnjt{3z;&?ZB(PLYG+=nL!KfKq5_)lqnD%^MXY4>QQ5%;?(q*evdRvB+
z!4T5yOISIxa_L(>NI4L8p~8FRsUXmj5UCBGR>}ymmYR;tFcN1g?48OA!(XGbb?#3z
z-d>bU>>AyD;+6Qf>WdG8GxneK1tO348AQgMTLzl#qkdZ|GQ>v@@@!8pseNqF`+Dwt
zXUnnM7Wp0$9w{9#9~ft5Fz7nX#cfO{uGGmV_%(mLOC+R5^E8{OE^(P^m%b+YcRYz$
zALa1}l<Uv&I?L%n$*7wu7VP<|A>mPHR4CqkMO;ebu?-8KbmFtH+Yep9GCzbM*<RP;
zdB24P334!Ri9~uN%A)_;-ct4R!Zd$cPO+@D=@*&%Ua%1l_4gt~*3<~<mg6_xq`In!
zWL!eA#QG7ptYdJRr2a_sWy~a;*kRkVK0(I5#4fKz{lL2_9#=`{m;e0PU;XXD9$7n>
z3uLyw4U3YRZ~I17=wU-29h@_YXK@p*OQ@R;`Ob&1NZqZem>tR6VND;2-wPig;qkQu
z+mMzO=R;N{kE(PF-vq(@8#J7_aRb2$N$kKeY>}5~SMePmuYRy{cwsU;?z%XfYS5Xx
zTY*S(v*`ObM3})r#7eQyfOdzCy}U*Re-m`zyX<yDyLk^r>@_!*D)8UC1Q0!LULdr4
zgmfsoS6Q=Q?-M|NvnIYH-MQw;=#k^fw9Mry%=h)^wF>Dh9LI^c>iwRsp4@J?QL;=S
zLW?w!ore3!{@8=Vbu6B7Pc69@c&f3)+#cGfv}9{Ci;~Bo$!PWzi{e?8N(8PApoN;7
z%#!azU?zLWWBJaeH7RTsJ>S>xFeXv}N26>Hul<@wBgeYZ+!#C1=ww^;q^x33C$a+b
zjrEtVP2b%+rPccl<3HzDaJMjO4L#HUT33-Eus8FqnXmt31SXBYlgsHluF}4c#T!vf
z5yiq_B-zZon4sgci7X=*q`5LR{l*+*cz};lO#<hYaA7lFu8GC--d4$JrzmXD3ti*`
z5j=V*Ju(f^(Y=K+m^wnhdsP+ZXy5O;!+NMgA+8>T`_zs0!u!c!d!dxpd)zDn=c!zi
z<hJ3PfRy0v=Fm%`<ZgJ7VTSY)`Rkh^NxCGQt-WE)D~f|jot^Gj{%z~FX_?#}k_evS
z1TOL@YYr^apcG<>c@&ELNxXel^balKdVTvZmwf`V?Bq}bNfLBx?1o8n@lJS)6Kj`R
z+P!XSvn&1H5|`IQk25xLIfum**cLXIe9~<-`f$M1cY4NCQ2N~b*xfeKI`4IoZ^rQ|
zupMBOBqSiurBGuWZ-5dS!b4~u*XgAPUe$Hrx5ns*$(z_b^U=b?^PWfWNdR>$v9#uf
z3lkyIa!fZhJkg}qvxaqAm17tMS>s2Gm)^)anEWbEbf$Q;RKEemjf98{3V2@XuqU0_
z<2>EHN$`agzaMdeJ_JL2fs3HzFwurhm5fg&BgOEjR-nDhFx>=?q~`X`1!r(ryaR12
z2gwU)+f!)JT%D%0&xqO9;;$W!ThGT9^{`{kWNT^W?4kOPsgyJiqoZ#_-pG+HcEc^A
z(;=VDfsxs~3*6gDW=x%e`1`gwqokpm8-wR&LN*SsDA=fr(anp+$u&Qiy&%|9UKN%f
z;yY(XC5#L1K>(@=+7k+2Y@SqS;YUl3QFE<9q3sjkS%1(NLc;k#lRha(BKC%hrq&$2
zsMCcXH1EeZ$Tv{d9(7;3*zO(c3@xB*HT5;^1&i}e{3>OG@T>#3loL@zhA8_6MmKbj
zR#eNZieTF%T4wI6w|tEvsG1b$Xu;9ECAr!INzpV)GGnaaJYq-t74<)$SAQMA59}oI
zws2~ejE@Ic`i=TcTv52OU&2q*<>)QE`TI5Xnr2>AZl6y#JPL#NjVs0E-wo%mW`Rl`
z{EB3M>HM6Rw_dJ_Xsy-|{nvRr??6v-hdF_yEa6UkUeQ)c%6Ex-qwSpcl*@c4pAYLT
z?rHiae-a)3oK^R>ybw*h$C_rGxJ`h6#P@elojlPyk(*`Uf`;t5I`1R*i9qEf=51eU
zJlQ4hM(4Qnq(?ax<kl-nKPTjdUJ0q%Xf}jC3^EBRH~P!o?d?p6>`-Wmxq9Vs3E9sC
z*lrk><!@n2s6Dspg*~O#B@0cLq+l$jJ0m~xA2l*AP-dRzxs0$@E~nbfdXM;)$En<a
z{CWm#66~m7E#&nidEnyxX%f)X@W8qR11yos3jaL((lp%Ra0B*C;cDyp2xf^$@JvUQ
z3VuwhUMIcEvy3Eh+}YeBM|k4TH7{_%aAJFQy$BIlM0r`e9xCdY-BAcU$Z}{Fau|(9
zg+<bSW#kb1Rzy6KHqD+wyOiAg+>Fymo);(lBzhy^85AbBD!NP~a`88@B-T9o=JH!c
zqo|f6o1dgSimyMe@}7Lj9}nwV`jFF`%oUe~Q|M-))hT{!fs33HoWSzQ@LSiLd|e!p
z*EehuV22q5f|DIR3^$`Nz3B*<h*oT#VpZ+|+<7e}%I|c_f;g^SYI2k~MX?Du&b=Ui
zuIB+|EY23SFZH8y$rRTWngH$`kCwCvk=}No-aQb5J=&8LOsyL~^<!$-3vaJil1mo4
z$b6tzDq@tAbt?3(<LiBOocgVb38uGgdPN4rQaqWs(fY{L3HNDezRv%wF089|iN%|i
zvz`2N#;M=rzMU>cFwO$=!B53S78z=90NQ-TI6+KxCLgR5Q^4y6pLa$N$%yK5nn9sb
zE%&`pzg+*EAn6LW)XBAA@2A6<KwfjQK~ro7dN>P*s5P{R2pCj8_1yX8)my%o%v8aX
zZDVWopRMaNnu&PWBR5i8Bz!VFIbV(1zX4ULniOv!<Und_eN9t?Zo->7-vvNF1+|c!
zgMDTQA8|pv+p+$2YLfbAKlrZ8#umo<+o`KVIb>1Y!_F{_gGtK1Td=e1J9D`@?0V66
zy&%VSH4-do*xp${rJ**(*H<r~Ai1wmsUZ+=nQ_#o`ab&M^4fb}{?YzPLXkqG%3F$d
zl}Be<Yt>QCtmqI(BwS=D{peg|;uQgv*ZC=^xFoQly2L^rxoGnieZ!1<@OY8Y2=dql
zp751ecBOfsoqcgAIGZ?4HK}kDN)yf*%A!J9J=pg<WKduTjMGt5@{>2>$1HB)@fY-a
zdH+Jh+n{dLvm_|exsRGjnQhi;lCOw&(M92WYvY)hl|-y$&wwHNmh(r{BvWZ>0o8>=
zC0+Yt9Noo9Lq}c^)@#c0(T+IcbwLNXWfAw-@gljSaZ7Q(e5I3=z;-d&niC&t+reIP
z!J=`dq+Zq>z4N_JzbQ@<gW7>0|2Bo8E$=x;uX)&%K<Dl!VPOk;hXe~XFm)M-+k<^D
z?^ePne{5@5Xa1(>prx7W-$%~}6>-h!vL-(B#lMw$a5!#vREnpY+zl({Pp{k_me8rP
zSc{~1O$v#i<YvjcplAv)XPza~c3U*t?jks`js@?SZyrPHTh#K5Ox}kG6-e~(W_kow
z9jGbPaSS9J+yT;R7D{fmB=nMT9BPGUNs2zIAZjz73p0zb!$Hnf&cqiY<4Mch?Axoi
z^@8>vy=hmApMk-tB41S+^^zpw=?R`XoNgMvxe(mQhpp?gXn4FsZkP1=`{@vPfea<c
z-+8MFqKnZngUyl%q+2hK+jFFdX4~TMMUHDF7sDsONDMs6zLr$npF`0YFcO%*Vs;e=
z`=}Xp8xXd2*E`469Ew^dqlc0pQi#Kc*r`XAK;9(QnRO1l&ruE}G4iIuBRQ%hL=kh>
zGc_yYYa>&|7)DoesK?I++3qU{?IS4hgh&l~n<^JQrK6LOVtXKfZfLoT5O^3fm}7B_
z#1h)_OtwXj(75!x2raZK`+`@-TOK_q=@q@%V=Og<9GEhg9X|6Hr<;`@c(uRZ%jQ`U
zX^m^_KVDf9BXQ60ktI3JR?sl6<{kIRxl<&Y6e<b$+EsM^zIT)4&T&yVy;5h_eIM<S
zI97eLtzvnWZ6xi$)D1{`teC`aIX-L2^T_};vd{~{|1N+x_nFHTiBA)2#d%cCsj|%K
z%xvFrmOjcNTtOkNv9C{~yo>XYZ>@`)QjCr*LYzva(COnbb<f=vVnA^7&987LANMC7
zaNnm8{Z4`Z(Y9o47{#KC92f)&i2JKUSMJ$hCl?xdVQj)R`1w|X9a_^Jtn(t__SFG_
zXcn|BS1@*RP%F<Xgyr6ksAmWT-N&*cNE|Ec1s9R{L_MG_`)E5Q7TmiAtn>ZpqX8EP
zSFCN!t;{0#%8&2zdcM7quYJpmPeg|T`t1YJ9<0v`le%Nw-d|wam;9#2>^<72ICTBp
zB85D5vhtQ;jrNK!@gVh*;bF}P-9_rQVzf?;?~!?f6qPX!|2-*stSH1Xa6(Re&FJEK
z0q68^$nWbXzMTecPTR#|=p&D0R=uovh33NN&Xh)AvioV=W!LAL#G(VJ0A8;q9nAjp
zYoI4z9jP^*KVSG(UZ1qs8g@6v^JAAIA*Iri9`A{S3p8kc_!x}sXPUCf30?)p2VXV~
z*2&aq*09y6kYk6G#1-(7o`bHz^|~S_uU30Kht$0dfBbS3Kx(6M=uDgY4t;v>U-sM|
zOXvFc9sV}!n7WN;Jp=v|i_rEehrwlqunmeA;a2Wgj_{(DR+i4mTPJU3pK8VhjPe8j
zqvp$Bip>#5h-&wrurEqdG0zuX(r|A0xBO(@+NmWytAgro`c*&l4aApC6z~$QC5jH<
z?k!Y~*6vieB~e3ve$V4SxNWBXXK&z^M3|?2&AOBHr*Hq~?!I7(I+SFyWNXQLnu&_S
zM4vcZ4&rUhOz`4Se^7_ms-=z*vJ&6+Tj+ADlc`6Lr_NKQ1`kvLgO53J%%$OosWPi4
zc4b*Bdm5hlxRIzTr7G>VA@7$~b|&V6qYM;L@M{Mfv;+1VnF8`Oj$ZR8r>hMTsjUzP
zPsqpcl4__Sa;MgIH{{GT(0jc4gkj2r9mup^Xj|54OI%^g)7^06M8}yOh04^MZHxO$
z{vpTb9r`-cbh<6x!Tfjag)=&d65zFGbNfDQT|VUQG_W`<P3>)crc7s|54P1Q9sa##
z3pcL>1PhiHk&4IP>PdbdGN+@foicaPL%(F|{W#2RFB)RylG_lKG|7%nnxm!&7f9{>
z*%&9USveN<6_Z>-^k-ZBu>3x~faJAV`IHOQSu{#-^G$0G`3RiBHv{U`foaUQIA^ng
z2McJ+fPkZB94tD6C<w!GM+ez9wQq4qP`Jw=qCDgG^em&^7C&h-TZEn6DNtEb2vXkK
zEr~eN(PR>JszKnG7BhNAW~5gNHDZ$1^DQY5OF3NPujELB7?0u(ij@>;H0iHxrSvII
zJ#CcQhK1K{ODGR^iBfUL8^l&DZm#Z1!YP6FYECqP6&<&aS27B-ZAI6RG{`hE3P4}^
zOvRS&|3w)v2IT!5#a&(%g_ETAtU6C3q?MT{20C@oA!bKqj_Yv6FeYoAC}8lAf;z-|
zqWLHDz8FMd)I%JV*|uV;L@NsH(j9j6KkzNF+!8nsT%sLm4Fm>qoiOL7Vf$#Hd*^k|
zbzQ!isc2Hc_s0&TJ8VeVou7Q>^&mUq7zH-BIN|g!Iv46oUyj^KjA+e^qTx;~GtOD*
zJNI_QX4VT2{O)S?H8*PxzIdv<xI<O@ZfZqmd%Ifc=+0JM)`0Cs<W6j)xLTo*|LJ0Q
zht8kpQjVN}LTLbRmyeRr5*8UNVa#20LLk#~ue`+m$ly=6dg7eun>l0wQ9A9eb4U}O
z#Qh!ISx9XgO*Et~ABdm)*?qaX^3Ns+dc{2oLKzTl(PTFJ9oMNazv;rrb)jkcC7c`>
z5{eVIQrp7zo6?IDQk7}44OQ_TZlL-7NIdUNth?>}>X8McHed-QGNpTWlPPaZqqgb6
z^<scymhXbjQvJCsoO@nmGm$5~Jw3Bs-@E<dx&%U{eXVK99`Hv!t5Lx<6{iEYM*ni6
zn^oHiIcr9}lC+vDZTsT8h3&@c0l&>QCPJJnhb7WS!pDuAsiZ8m5$b=v(&S2&vjO@s
z>X#w83ryuYcOK@CRg;)-52joneeu1G6wJzM$0K#E;?(1(Hag>RoaG6T)Uh~TX1@5i
zAhbJ3T78M;s_r1}+$Ia)=Fww2uen?jYs&>yvr*$isRq+~GYxW@)X2;HFqA0b2uW5=
z@x3QQsWx5N21LFw&N3ewi{6^?3XY%9aV?yXRP#V6>CR`Pt8M^Yl<k;yo;3%6dH!6c
zYvdh>^9SGHwnZl-T?+mpPH&;OL(ToHKB+CE@EKamEl7a-cj@sK^_YmVfr;3J8S$<B
zV3MHs&AF&DMO^Tf!}LM9a^S!Mu^5f3eewX4AS>8hO#XyNgYzi(*SB!tnra-SMp|I&
zt*w>Z*Gh!8v<pW4tj>L~3Sh!(m|(04oW5%Ox_e`s+GGTh9HNN43pFlapAw?G&uUL>
zOE~Tgl8o`@K4&`&t{-V8B)+mg%xNIMlNbqVoV3&dY`@jx-#kf$xFd^A%=`<~q&!|Y
zSX|z5ng5>cc<79OU<G$A@7B1RiG5O%ce!#Wxzm|rZP5GGu-C-*awOrFyH%NdU9@*9
zamw=|erhDLHrB2vZ~3Fb5{J|HCFNtkFR5lhY{Up)5nA7npUZ8~bJ_*ra{ePZnG)gx
zcc)h&<XV3eDcBcBX@bmxRW^ah1}n8!w2d-kA-^M?QF}4a=Qxr*x!V92zh6iKE`}U@
z7z?ToiAolZc~l6&z`7?B{YPgr`5CNz=-VC7AvT<|*FA6%wTL~+-4+c5aV3}LqG2zU
z^-D|*Lr9b)xzkMzyH634WBV<K^pkjWi>r{vxD&>Cp=R7Y<;M8Pu6`uEw5G{cP<gU$
z64^{~q+4@=)J-+hZ)X<4dC|U1RNGz#`Nc?dW#YnB&6q?``yF`#)HL=vkYNy<ic!!f
zWakhCtT+-5Gk8i~0x&{w0;+9#;K3qKl@<{RVRuSD&4=&{6{_-75{&LzRhlwWwRC0$
zX02dW2ZXv{as4~$0TGO2;3l{JS}@L>h4DG528YP?HJB~)tu@OkyF10L4^z7_Fy^h{
z7!(vl8`m1OmZiXUbd}E=l)MU)OU~8AXT_jlxbYq@RYDqMNM4{59ka}H7C3yMPC;?V
z@@l=1)u9+QF3TPM7D8U}<-jyfCXKfzQ?q4Ee?cEnbsE!IA&={kZzr@a0x@+Nham+3
z5)DSzim0j1voHVP+1GyRdy$-W7X;ZNYef*>yzO4L$*7|B2kv(zkRQ6^cg0UKtz*0q
zo#gab)AcqTvV)W9uY+%tyK;~o!wc{`;@tMojw_oWjF=7Dci3)ReL735J}*{OAB(B`
z%M&H#pJ(foO~wid?$VIN!W!XhW5XrGkkE;^xI<>FrpQr|t?ypCvF4CN*N=eBu9@~z
zk&qSI5#hAF>ReI$oWG6+ajfgby<Rt}IWu}f3HLj*+GO6(6B3#^qU9)Zh?EJ(U&A|J
z*0m&*Lj!V=r1!D>abu;i=L)%eaNdf^icIflt>O$cLY0qVH{uRRCb*4%0c)}SITUdS
zLDXCJb%#Xh@%7M#$)CG^4h8%Ln>->>b$uR3eOTwV=sAP(EQg*ZnA#tS!RZcefC*5;
zrMsdq5$X(Bvo_y3?Z4{9phc);UQ4XS4#OWhyIA%Qkq93b;;8$nt9#KAnCyFwr|E?P
zQW>aSkj~`d8z{!&{t@&h_)&V$NFzR)NX1~DZ)|Nb3cQeYros^yyin6UAe(FPX4UbD
zDzrjKkz%}W+&NiYA(G0ceAvLS%yD>%=XlU@2wB;lzdkJ9XJK&Pdr^_`fyV@V++9jc
z+2R0m!!ZMVE2WP8rE&xFP=}b-$PSj}zXv9>S`K%K@9Z^-!*3Udnez|H^O|0*VdD|X
zOSqw*v+cifCEqStIGhw3d)}dau*`AL%aiK_DS-z<#)DnVJNqAw80o%gS8sn*xZpgI
z&9CL`Zs$Zlw;i%;klXglQO|TTpWzyAm4iw@y8`~cy9j}yfCbU}p@6<<4O6Ngtw@*O
z7{|s*^La{Tgs_rpQ3ET`q%T12-4p@k7AwJ+Cd^cr{_J3+e?*xIIs};;1BDOt`~OvY
zrWlCa6y}p*%|SgGNX~H=i<2{#SQzMG{lb06;&H_!uTG&Tq;V?N#H?|`dTLv_a)=0K
z9ivI%rbHibkAJrPP)5Ifg#<=Kf99Zl+zb`w@ksU<iZz|GM78FDlT$HfckCY($g?&p
zVA}t=c6I@56G(o_FG$V@5j(gj(>8Eu6NtQP{fOl7&ytaeCcmJQkabs2{0^dp=S7i+
z!<bVWv+nrjcapivsx*>+r*fO*xNpH^d)X1&+UuycFTC25-zrTidsSF<8X2UiZd`!T
zgHZ5`mvoYAm)bUjA-#l9sWTEj5=tXAi!N@FOspJL{Dlat?cqmP9Dk$2<t{L<GeRJ)
z@0{jrC5Jt~V1!<4BM}x(^Z`k1B-{g|(f_DkW2aqrdk2tr^uJ+hNoH-TWJ(&<+lVm-
zjO1U}XUb&Z5qC`yw&QQyuuIhES?cdQd?2Gs6Yw6D^)-IC3aAGf6Lgnfe?Avjj|31?
z%C3w~moN*aK*o>K1YbYO2fh^=CT_J!@1(5;^pMD_{#|}Z4w8hrn9S}AWw{lrFH*C?
za|^*cL=HvbIUC!)`IB6qd-VZ1r(w>$xKiM$(66UFO;oCq+wuk~_Q-o#tg1^OcRho`
zJ5bDdHYm2I+7kKsp2RgxGzt3qRy?l&E5vso?@2(lvdK<-->1*@eD~=g?*`+~)JHvS
zP*tcncsq=fOMTa1=n`NZ6g5~#pasJ6ul-VS5MKxiVc~ASzAWvnRdzg3xe5I%w8?g)
z!pRX^(d$Iw=67iPIIjS%13P;}TEDm%5pCY)4XO5<Rp7XsTN7g@l1>To0yC)&(16g3
zQQI5p^-4yK`y~$9lgdU!<aG20Xbxan%Uy)=;0EmxQpkQqeB>B?Cc1Q7*k=*d@)>*O
z!n}++EwsSmWYL%2q<~u4C^f}YmmVo4IP)8g(B<R}e5`mPnVj_%Zo~n9Eh}l^^1I8^
z!2!qiQV`HfS=N9`|9Hs1`q556&VR1oepxBFmUum7;+lfoijr7v4*~I!m0fZZa|qGR
zS^m81h=TNwvgaUBKb=Ol$-%E%Vhsx@&)I{k9KKX~W;DVbgM{=nX{EeM4wgyYlqF(r
z;cU<d&a6uRCNJEHwg#MXo^25cI8{ZLs~6<Ld_nT41s|d;d=cI2jO|FFepsQNhTEsY
zI_2|jE}3)84!~Ah)%(Ip`Cmx{Sy~#M=08RC-*(M1Ji-!XwuPaJzP+ZXkBivQ+1x~7
z!cCA!=4LSK@Iw71UpwU{_lTPvq08<{<==UmOiR0SY%@bz|Lne`!Po6O5-P<zZh$eF
z$pK>X6djhbS<7O(CK9bSIaVcJcH1tibSHR%-BM1ElA6##jpz0%F3Lb9b|-qLZCdD3
zdJsFgQXE@<=ma%6CE?;y#Mv5dMHi?eonvhF_VCueunldB-tcv$GJhHVK1cik@a2Z4
zw2S&=&6^E?3llsSSLvRaahZ{o=-;Z~i4>7iVKg|?M0+L4i>uIS3iNWgYFT__2vNmT
zVbBUNa!@mVNFCBufyAP8IOd5?5w4LqyY)+4h~AcV#+(YB<Bp(96l+UlyEn#3TQoD&
z{f+mfw*7F`-L0-mo8c4TH8dQU|8TJSQ#}@c8g7d6XQLvnWwz~A!;@i)M%`a)n6iEK
zDXK$-{5_s)zk&ruu1b~}`KO;$U#`>0Zwc3&tXx>Fz2!CC^Q<MC6W}*{xT8<gSj3k6
zRONX+FywFSR`OaEJCL|=H%qIqZwv5MQj`r<HXr0ya}^jS7*g5}=<g1^EsQ%<e=Rj<
z%D+=za{IW+xL}fTX=p-iP#WuV7o{XlH;n)3`KaspDfcFQqH^1l^82ABeaOn^`JrC3
z=?L!=yFE}j%K6Y*Bz2xVl{($F5H)_!>I{9zhx32~I`^=<?{S#Akf&5Z!s1j7(Jz-V
z5Id;k-bvYsk>aKZpH{{c>3<7ols3pS`G2|F!!G?taPXAphp8?=r>;UyPzpiRryp~X
zJuJ)~6R}XONl|3mFbp!TY@a(}Q)E!aJ}iCB{~rs0p83y$v`0dOnW7l512q))M-vn3
z)S-RZo5SU#YI$~kRps=UIoFYfX-#JgTqdqukh%r7L`4<IVv2agWoJwc^z`T0i7rEv
z!uiz4Q|~A}sAJBi#IOowno5Q=24Zzlg6LOLpcR~*%;oHCygm$~I?rTqvgvStJ&5kJ
zvgK+YY>1ro2GZzLl)kgZP@USzQ!MoRdF4}?hFd>yt{XXHcSvHdB?`mfc%oC4iT3n~
z+A35HpKLBZR{L2hy!~VE@*X{3jqx>X2G|2B31x;t;QaO;M#{GA%@)yueA?~VO`)g|
z=u`gY7Y8O=CCCaM;?P^*oX1%PT(;EJP0-W76{Rj8|E~7=n93E2^F-rC3Cs)ZJzaMN
z`s~V4XO~mw$k5Dd8AByk)i{d>iGg2foc>MaMsD@j(^ZouFz9Wa64~*k;f*o0ben-A
z%C#Z2fVUblezwoz&UFt;5E+ODc+H<S)!k3qwiKpSN@gjac6a7;etjIyyg0Al1Wari
zfRoU!b=2s1-A4gpsUu!k)To^=*|z`n8^w!#O^#W%0jGzLn?FPiAAK=^KMv6+%Pk)5
zz%#w;#CyW`u71-&%S^jhjtYvi+RmxjA1%rMuLf)Cj2Lp-B<yi!@t;NmyKCgUgvU^X
zb@0rsZHDHggOGSI9%BzFAY`ca9ftB*j+MNa9c|x7Qh&*lulQJ!iM;%Cr0~&on@h!>
zB@yIZHIy+9nN^CwBSyeD1pXIA)b5%=qS>{3C1@!x5ecHml2lKRQ<18zyl}(;KR!Ns
zy^+gymbK_G^0aA(*J$Q0hm*Ie&va^Dnm#1BxtgAB2AHB#x(1^Rmk>+`p-@md?OHVZ
z^rj|stBXnkD1UEF^%`1EaLsN4vmo~#9Tj_OVw})c^7H8{)J2&=9*rvJ#EV(kYzV+!
z5M$j>PV}<?x4qeOe)$%TDwu{}<|XY9{eiy11mC>XV)9RXSbnl^#n+?gORIjmm-wM<
z;<e-C0+QDr%<1CIre)m|$=wi2Bx1iGEE4br%S$>GwT~4oNwl?nVa+nSeGTajKx@}F
zQAbherB~wj@d0{;#Av`<6xS@-fI`PWz>S_~$mXu}$hjl{!8=onmSp8Yaz)h?)&Dsh
zV4ds!?hfEh{H2sk^Kp#_T~_Z(c(5L5>!>EpEPM)`*B+^?&aIB}`i%q)1e}E4ooAA0
zL;|>|{-yNM9mXM=m5D5x7YxQvSi--7P`?=QG+F=o<f^?M?0_0y6}FQze?W1hNXtp^
zRR@NZ{P7WAPk%&`kZo&QIdaUP@kvgc@Z08NbMwWTV7~1(U^6?T@@i^Zad;j0(M)LC
zQ`vMjL#TZX^H_ktEIxiZjfZNyJJ`=m+u<kMlAj|n&LLXtb)6qaU}4jHcu(=}f|)b^
zDCU-T!s5(@QeT7`^5+yCFKh#Jp6vIGj;E@x(2lUDUx8lii{~KQWXNp+3pH;8jtHl4
z=Tc4hLgG9u)IRq54Q0{tFX5z(LWju+R!>UelcRW->Sr5z9E3K+#31`tYra0fcP~Y1
zP@uGH!g8X=F#UGtt>YpaO&+l;Y16)iPIfcaV#YYq_v3xv2w&J9(H_6yfM?WRK8gLA
zzMOpa+fSN9uIxsye=gqRQQhv!4QcQHJ)8e6n!AaVq><%Qd6q1~1WD@~y&Rho4{fz3
zEZzuF*4O%5c!>>JrtQtW+MH}N_$EHA-Y`|EQ<_}#=bHuT5k%il$aivcyv`viR!1aj
zTO4<^g2GwokL6kH0f>tRz!SVI-^uInz?&?l7Q1j2Lfqcn@LLRRkl2Btj<{o2fr&S3
zW&d4MU9QK{@J=G#Pm(}=H$iVAUflK4yiuvn^W2lnAvUJBiBjJbKH#_fkl*L=uHxfR
z&JoAVisEDnQ3<ezvT|uNfy0OG;%s>!j;=Y&nNi43>?h2|$!`vUwbrmVq%jjxFQxj6
z&LWd2pAvr>n+h7Tu}6~>g`3w118bBP9Uj7Cdj$>Se~e}q@eX_fbiD=}2Re#%3)=sS
zvbPS4a_`>0B_uWtQc^<=A`Q~rASLLQk{(LBySs*z4hcawN_Xebjev9v9Yc3K7yI7c
z_x(P{^S;mT{Zl!PiEF;G*1690S@N4#r0ShAg|7(W!PD(>`mTGbZVeOg=LRyAT>UB#
znx5F>Pt#1T><O`!?F%x|r#C?Ja)#xY_wntFmNK3G@&I5g`A7#=Hj*TNiw<P*%b?g0
zUr~dB96Dk0by|(!h$>^c7w!TCe>PDvtMzAjDrxsq82CDWQzIB@iU|*1b74=(5_y+~
zVVPJb+keC-`TPaN_{-J5Qc^J9*}9SEt40&_FpMdA=ju<uT#_n!I+)Jl&oj&A<-@lo
zu1Kg%td7G?X>OO%n?)>p`0M*rD8*;N`?rE49?%Hvi7KZ!G%9J!Qm=(wWtmsfBjKm3
zLVr-v$HvLs;REyCv)sV;uoxzh6-+Q*`uP_QpFZdL%82ES{rC%0+L&1e$YZuS?FX>*
za8nQ`5n{ygy)M+A0j{_tRp#+`%3W+ds99`yR@9nTB1@z{(*&unLmqc!A-zpARj#H*
z#@1&$U>r3#FgXRqClY#p=1B5uT;}@Nm#!yKd^_>$gZ9H%Y+Q<l{CXM1cB0s8`2vb-
z2-BI@X*^w{$LzM#Rv4UM7+2&_qGvN1(4iymr6Hb(*gn#uYG<*FQOjk$Kd#jve<E7T
z5c>e_v@pN48M<ko6b6kruVW_Nd3v4WDR(0&v*_s4<z1bWytwYm!nW?@O&T!)j~I7J
z50A)}Ulf%&@V#ivP_U$aC!;t$VM-&w04?u~L+@AbaikDxnA}`%toqhg{%mF5xuE)@
zd;RV<%aP#Q$edNnd%urPp<`~JbKqQ~Js&6W2>Mh|RS{NOq`FeNgxlkzwA!QblSn%X
zHTx4VTwopweDj!z?`8jCpPH}ZraODs4{-wkk(7b)qj1ezMLu!57X4unXvt~9K1LC`
zpC_9R{aahb$6yVWz`rNyWXt~hj{5IQ?ssMMcV_(fw<$^_h37(%HapGu_fWCxt#2Wg
z9{x^*AcjB<t-~oU$8|=gi@vCAj*be>p*&6o7LrifB}w=)c{nqhkx_bxO}Ku}^<+^d
z!-%mU1l6OK%VY64#EGP7J=a3^S(BQ6gYqJVs9jtUsJqNaWicQ#JU1Vn$$-jCpQr#5
zl#qh+V8Q80X6)>PC@vX$DS(M#&_qVZRKPhuM>=?dxUE+f1BFYj(sERy5>hb4XTrFf
zL<G`B*k*|}KdfNg0Bs$wCH{ybec`o?LSu29%_|FnUzL}?n0SS7!?4Hy2_inj3}&{E
zOUWT_7ds5VbEj^9n@yW_+Xt2_`>ZeVCUE&Ft;wK@J^n9Hv=k-FI-s3rwz3$|>)|(p
ze1MB2q)9@k=PQv(mwKVy7@`VCfjjQD0RpgtxILD38}=cbz3h)3E}|)mO?)u3c(0kz
z@N|T;9x54K=$o_bi>97YO%WEnSk=ssC$cyG5VTzU?T?$jP62*Zb{?1+o`hNMK!1pu
z9=+b<s*kB*wX+_&&H^|EFF9!qCdr)zg-BP)3z!uK@%KM!zcWuhdecK|s^@#Pp3Tww
zbIWIu8Ig8wNH94nek`&RE6Eucr+8ERbajHo;OaF~#8?TQARM3R+s_rgRXZx;eIFW#
zK|i!53J}2Z#om&@*>=wx^d)Dd%|{4(y(f<}wl17+%k4IXrpFMB<T!*h5Xb8-)>En^
zS@6}w$N*7JVy;ta%gD_F!nZ{HrbFUBI9zQB8SVIZj?~*R_F6dyk2$@80k266Hx*T@
zhhvMn|Ct^wqC2r%CXE53xqdihT6gK&E>otx)NZmt{;-5|C`3Uws0qEEw1b+00~F_q
zDU_l@Sk<nrG2K^k*wu%I$7v|S_YOqMgGMhNr$Bqk{_-(b%@V-wOBOIVPDo)TUdr1J
z-~3lO@8x6lzc>j3r+*CQ{!i=tKmTbsK>tJ$MMH`?ZvG=78dfFictx9TKUDZ&OMjx@
z%G=z58W}h0*1aPjwa1tbLZkqSQ#ZrJnsAp51U(T9-lSEN*jjN)hS&+l8HESEI`qBo
zHDJMkmoxAZ7J{(@G!>WWMAKf3y`G@EdSZ7hCZl7?nv*}MfU=E-C9)Dg0|8;E>Qy|~
za&9h02yLTBH5~jLi7Va9{i{XZ!-3{IC|85y8_jK@{+m$urGEIx^_8jbv;d)I@q$Db
zt;&}q8;58iE;&lm>*Om{^22GyxHz5QMIjLff37o%=1!hTt6q*I1Z%q$lMC{mlb(e9
zQI^%OfnVQU#dbx{?zj)U@kBekJ_cm|u|hoVtV#97M}l)`Khs`&DFFT2DLD>YmML#O
z<&c2Z3+Y0-B4NGCXO~VLmf;x>v)%8<6T(`h+b5xeM2&uE^~%4MGYF7AXl%<7W4b7Y
zvnlD7npH>LQkv?fSSo35`0|VQFqukdLoBz4lg0X1c1Ln;cy~MAr|$@y*fmYBHsc0d
zMuv68(4Dg2?*m-2SV2MJ*w*ULiKAT7koN|vl=WHC=lQ<seiri5v+{ZedX|6fLNo<R
z68T|7zoZQfX1ww2#@zM02WY788=NI@mo6|{+?{2YKAX>U{o9Tu@F)dRHo9~F-xtOI
zeoW3B5VQ%ebdwrbyAyPIF+jFs7~G@RFCC!O0MN~eUt^pPq5~A9*VER5j}F2v27c$C
zF+srL%crU>^ZGvv?tg#n|JN0&681#C+wqjY02rM-XV>|KMlGC@KB`Bwi6Hcp-oIxP
zD1l869DmJ+>hw1nJY=le{_B<7=pI);!Q^V%e;@4s)AzSWWp6qJ=DA;%b>g1noAA@I
zKJwb`wh&oviyrh~JLxYG^bxI>-5j$&?0k0n2&O4s2E?cS{Z2JbkEv9arhl*R|M)xp
z`y-ZNBfF2@0F$k(`5VS!k|V&`IRDwXNRQv>i2}iY`DXv8-{=4NbhYdrKjI=IJ_7GQ
zum11P@c(OFpn>p%TaiGrzklk#U*$FG{ocs^px+|G_kx9^=i;B2{!<Ecdhim7Vyr|M
zS-p0;8G{b#|NBq^o}abQ<44_&WF@5h&v)33X#<Gcw&M3kvHKG__otprjEk+f7?}X+
z<@~JFZ(+$B0J_Wq#M4w41=%J?9@@ivn!8^8`{R_k*4u$r_X+^ncQXgHUDe@agj7H`
zWyCJq@oIWRHn^_L5t{~gKMj6mES`dum{mf&ehuiGiC-m(3%b~B#f~;g$Lo;W^7fmo
z>++p{dJeQ2?pxLXw0msTQ(zL6tT3<*m{xMG2F+&yGkaJc!n8$ZK_9Kzo$vqpQ1^)f
zK({tI>(urCG*iy_V@N~`Jlvk_BOdM%Oy7ioQPq#th|JrK0RURaHLz%SjMG8seb`<D
zaA@F48TTi82+tIsi^ip_{vNg4F0~n;$J9Lt67m5kg~8~&lXro?yN*ht{{rAx6W*48
zinWt>EIr(gJbZtBOJGS60OaUXkE!t-@O}?`C=%&20H-PA0@ifduh}wQ(y;TgaU<Po
zuco&7hn~m!Clfgu2OCLL&@GkMUio7SaVJ3G8L7_}0`TVV8H?<^=H~VuJvM`Nv{VNG
zOK`}(9=e$kmyvT6Z9ZdJl$FeJ<|`F0dVjgLFaB^ME;|2+!n~QPZCMw(EArhb^32(9
zsQ-0pTZ}j0(Kzf<V)EG0wSQjQ+MHWk0)*lly2Z!!&)3UeV(N7(@Sr^s`1S#KWT3iO
ztT&t<z4Zv7cqL@HuSST;t=&A<ORI9ZAeBG&xsE9v{s+sRH4Hekk~=xF|J~dwC;fzQ
zf<T_q$N*SLIF6{O@xWlWSyizs_%}BR!g^p3+1>Ntx}5v%oSM1$4L(<ZJ~umL>stA(
zEaM7Ye?sgiLTzU0ZedB#64jU(nHqrG)cX0gW9cs!v??cb&EY~w6`mC{yEUy*?HfFt
z8SL9VoZD%h@DjL%JYk437v0q+A{=}C!`#U2;K+=Mts@cMRa}@z*CD88_V|%c=zADM
zGlR*LY4~u6_h9%Kg1*1`MDPHXe<H<?I_7s%>sJYod3qiW^Z08HvjN4eA5%rpw&@QI
zH9MDLUGjo$R(FB?5lh9AwW~#Gn(H^h03-*8ZW%vd>!P~UXDDJUxD}``?-7@}hivAE
zSrPoQ_N*<&qesMbKV0QJ`~rzD0(`@&HK_2g?{<o>C4m;2Xyx;DuB~p07jIzxuD!4T
z@!L-Es>cmRmD%C<aN)<_geo;7dO9k-zf)XF9b?$?Yv<f>l+N(~esZ?`MGKrfe+Kc<
z{rmUXkbaT}Gq_(jxQ>ri^VxZu#p)2HJKuq6!$|OigE!3rz~J1F%V3MhFYM}H%xX~I
zgT$mX0x*sqpXs@9>xnS#3#>JFIjKX}%hMw9g_i4|rJi+n7fVZhL{Xk`#|r4TH^6YS
zgW#CzLoh!`a83nZA=%xztC8;5A6=xa87c~Nf1EG5H)L*x%lTPE_^G*f5LpA74RGHg
zckUmlhxf&LPUc(lV)!j9iX*474j}Fp5x)RO=kuF2@q3d-0LGWYA*@bWafyEgJWr2&
zgXga=i`QiQ050Z;e&k^-uo*OI(D+U-@VRJXn(y}QY9Gab#UrSja?|wGWz3yd^f8&R
ziWM<U*O8vvf{Sg3fhrAk^3fhYc*PD_`O&i8XuofaITqxHQ?&=UNrb^owLxXhZC}m@
zjL^rO)8erYFyt<zMtY=m@dCaQ>6R+<OF{GWZGttzU;X#SDf>U!OJ@KM+B^{vT&yYC
zA=+{~jafdr2OK^4m^&YTVY|56k{6%72QaI1eJ@Qn4i8}Q@qb6KhnGo3BHgwg1G>qQ
zt2ZR_{w7?sAkl}b5rdRWz?`gSxOB>)XVYPTGfD<8iIKQ0KB?CBbvrY1KkE+K4--43
zWny1ze#Q}QMQbt{QLV7~vB?Cm>%FJXAu#{XTKZ0n{3v7W1L(M*yX~C&Z(OR`n!W(Y
z@fz*3!PO^&@<)={iCBy6%CAfRdEoNWt$k(vleAPb>$1o1R{lb$1vqJ~KUxT$p{N}f
zH($9`OFnNDnJU)*!KYfey7`LI8fI+|6q)QqS={Gy0-i8ElW-m>DjviTJK;N0=ie4D
z-!1vXnt^96033@K`{sV<=2>N}@3*ths52SMuKbDYLoq1cl*=xh7V_Ja@Zv<!cl`6b
ztyv$h!Sa534Wi%Ipa3%dy+ASv$18U;PvE-y7#H>N7cltS-NOsmh30-oq<%&m9(_u;
z?k!@SDDn(WBAnT!Ek`kSbw&XZoIR*c1xwUkdt+9R<ty>aAaVOr<$7$8p(BG?RCrqF
zWRCmGCO{;|mcFW<*58$%kkYvAHn^2#vVF8wgr;;nwrQT_CoaKPd=OqaJ^*O(6!RLm
z0!;XgC11AhzDX$$Uc)#7J@p`Do){4w4@t>=xQu<U=kTD^cF(6+usB|fxilsC8(6h7
zlSnf=0w@o{pLkuxgB+3{cW@1RDHiVvUVb0IgHz=XUf=v?9_OeVYTh3xbPJGNXPe*M
z4=vrscvaqp7+gLTUrfC8JJ|GPm;ZUy$CcO$V0;maXh-*ockN%VfT~M}Es#pr)z&_8
z3Wqi^M=C3WU#au+SFp|wbN$r{T#Lu&DqdYqFBcEB+HQcmZgu*o?R=(3Y|0gwagl~^
z-^J4tGA<`5_e>O;^1k2y{CRMzk^D6@*NyZC5f7FJ^od?@E9KK35gGvc+X4#EY-4W+
zL2KnsiD24Iyo-Pz(y_FL(rM|-jqVZK>r3gLF*U%XMss*NZtz)N5ncXIJZ)ud8H1Q_
zwYtILO{fOI-TL;8%l7?vCyfukNf3wk3f5xjXYnnLM-j)sJ~OiHvRrvVj^E(}UtmeY
z;#tY&Sztq_`ux_8*Zdmw;|*Vb&!JCY@Q>-6j|_Rxas-Ib75@U573S9h4vw@f)(WVH
z1wKnRi_@h2<l?MSNgWgXi>2i9oT2IRzJN~u3&5&6UXTnq%{&3Zq@E$7?hd0gtQDJ_
zE34i*jKQ}|pda1<825sR<FV&$76_LL7t77FG+?JQp=6rIG$#7(nP+j*xJUJ}RK9G!
z4*=jM)!MfKP${kDsaz%pr==Y00XpK|U&SSk6iY6pUV4w(^JdLkk)4a7SkM7LV`COG
zKzI_IxSl$uL=&Jpl$v=AzAbGyt|u?e0sJ<q{e`zcD|7Eq>eHMPlf>27L}dX7qwq`s
zkS8XdMNAU7$J%jK+j7P>%`xb<eUpkX5#KDJ+Q>b@7|D10J5c$7E&Pga^YIGbKR)T=
zwVN?L30OD>wAfeHnD{h(!Y8(SoHI6!B6r3Y<|^jlhIWrXngGMsuy*z%;DDSiD>bCK
z_Ip4)`~u=C&FOr6Lft^0L~p~lQk%vcg|=gI2aYL?Rn!qy{{*=isE-j|LSQ81pF`N%
z07K%kv-fOvx@BtSECva~gtD>p^)G5nzeo7qeh7zmZn+7OG(XXvSRiGh7WT5TKi3nf
z<Ib|9EQEJY>gm#+8Gj``RG<UrfU#!X4V!Sp9ma#bCWPFFw1%3vejGk{oOUf-koQPO
z43#7aL=RQy{SeorKp#pJjYP)E^?6c?N&pDf_VS|GkMp6lAW7;~ErsKDDU=@J9M@UP
zAz5G7TCr_KHOC+`RdH0LXcP`09G<tpBWVgNBYW<9;pXRdh}$mIY<1*C&_RPH-kU>a
zsr5DBvt~aKge+6n*2&c`i^g@QCQk;7V|W{4uVA&ihvcr?*m-3T8Se;KEKzO(6YV3*
z8i@mEc&9(Cb+tAEX_hNcZM#UY06l32F5q+Sd@-PtD_J8n4*S~)Wspj<Wgoza)@G_O
z`uwq)UO|qm;<2>#ydWdy<v^9wY#UHIJ+N^AW|^Gl=W)pdq6;O<vSDW?t#kEInz$3~
zCjiqGCpgPA3b&qTue|u^iX?sANx?$R_5MU*NysKDp|Rh6s<U&c!qY8HNPWZS7orlw
zuS<;v^(gs#icrCoRiPY&UY~XRS#R8LltUB<IcrIqmY+5^<VCy13$=Tx3T-po0Be2A
z2joL?mJz}?-fnlg3&i1<!`;=gXFpRt8)YxO9&VNpSU85kN7M^LGrRo0Yy7WgZoR;_
z<>A4qG6wmr9%ytlcLI?1N7pcPBof2fD6K(-MorI1S@OXd0IX3ge&!nGp_ZI}qXwc*
z2yGd{|54Rl{ZE!vE$uJ14G<|`iM&0^!qJMWB_ZD6jm9KnVtr}SRle28Kj35y3y$s*
z!A*E*3S{#yM|R@H2x6@pp^xr^ik@pB5ae$qj(b$gNC2Z!Jwb3~+z~#R4%0o5C(Pe(
z#oqVslUtt}&i7=79WmAT({@k>ULz~wcl;qEzNj`2Ft_uY0i1bylruLxOS*$|y8T{;
zs`K2wk1^#`4CbBTn%eg!?xd{8Fe=xV+q|b{f-%?>+Yh)u=J`i!I*d4W=w)IEdH@Q<
zsOQiA8y&qy04vLM&a)VC_f(hy{)KhMI5o&j(){tzIbf_eX7L@1(ue#FuroTTq_`u}
zprmLL2b2>PzU#5!UvZ*I0x&tf>rgRd(sL>6*g^WnU=@ef``y-qj3MqmsyT932p=zx
zrmc;&ZFkScIX`egs=hO;1o?^FsMsW!&UaX%7c&lQ1UUn}OV*%Cqs`8ZKVZ>IWfz8<
zo##4OHRgMNY-w9Fu}PwTt`*ovqFZOp6}vsRd#Fjgvsg>Qsqvmdn;9LFOYqze7a+IG
z^xXMo+t?27(;~QedX96t4A{n+cDe?%J)&;@IV6-opJf;RE13p7`4+i+qCo0!`cBXL
zs9RC4v>@c1_*+uIOH1UI)$8{SAY*01_@T-IszXIsYk{Kcc%TB|wyoKFSzD#_Nidr?
z6M15S>jopiSps&tV$|*SW7Tt<3r{gLYPhbdEL-N)>1Kp<K~BMn<~Y3JB<(sB3|Ext
zEWq?8^x(17^iI$WE9i3}v}`+(&D3csr8!&T=4R;E&HkK#EukpGHiCXN79U4lN^+hU
zPOpeIOzaRhm5I)V?|$v_&UxHenotoi62!~MDRaMKAoyb(h5FE6riH)4KLvW4#S+0y
z;wjiK++Lvni2#^6!sW+3!$4l%KlgUY6#g2>C4IotZA$v{l)CZy3D_?csZkQKITAV7
zN&9iWGppH8OBsx-D*4Iv$S{)cCqjR))1+Oq(`hFVaP%!FRV^n$o}F%*y1y3M5o(4r
zawu{qfsbby=?FbMnIAF{x8D}>6w#&`q%(V#9=S6u%EIXKY|j{|a<U;Ju09|-6=Vu>
zp5_mu#IslFGTF-9ulXujjRAN9b6;iNZqZ|XRG+%4_%S8jwsmXNCA^wh%GDbj5_rEx
z*I=&i^JkGXS|=(vs-MnM>pg<~EfdL*`EGp$p49@W5qkvoj!>-&O|t~W6y1Ad`D^S`
zvWA?Rmr{=7##z%pXnp7qskI`I;B()1aw`QP`voC#h@er;E%sBqH4>Y5hxgw<#hRW1
zX9)$BnpYg@up;;R3QxpV3gksjG%?0>!gn57Tva+)<9*?ee3rJ=x;3a%*0+$!sWo4l
zw1yu~qk>Cl@s3fmAj@iW&T25eXf}PIgmP1-E_XBhsw?K7W8_TU|BC#so-e9SzsSJg
zolJ|Gj>LtSr|6p2=|aZ3AS~s$VArHpcC&wWdGkb>t)C;w24)(KPUgp{sXO5DbWI(m
zM7u`ES*6NpU@8`g6XlT2_od-rv@o|FvjZ(aPmzhLP9?hj2VdDG8Fgb^I4Hv5M&}RD
zH^D@u19YNH1pa0%`Vg@kO%puhlFuq-mOvV1OGIl|Ti4Ate=tQHk0A`rQpG*0k@5QL
zp(>(q*P&HRj<Q&rT6*_SJFw;z*j2|`XHxMVzHS|orMb9HbVe8UNS$tOx=8>_S&gIZ
z77Ug9wDZl2?mbi;0lA6(_M_5Q&s;-hLrs%86(+i8gc-ea{o&q>I*O`XY0JD!@8|%4
zdym{u&XU;>H>1?hwY!LlxlrjM@p<>hI?_^YQQ34{2v+rmy<O?KEw>hv30Io^o^w8H
z#ZmjyB!jlOAg5mZ+{fltsj52J5YKFpz0@;ny}dG+Jg}(O7gLwVm3(h(3lD~37;&HP
zJ*re~I5!&VSgQG5aI5QyU#QyN*nB6L5!vF-;`X;d#DKBf&G=tpAQ0#=uHaSR!^N=;
zS>eeY!5&_qTceptTLXU7zTym`>W#GUk14u|&R-U`6?^<H;ptBwa=AIQ1KroJ*tW^^
zm`IGBU&smvy#>2|L^t#pXnlJ?X2R(!`JTa>8IBT!A|r#L^my6!&#|-&G$w9p_Yepv
z24RX9B(ffdZn^yMRcmQg7vqW?RJt6n%#rP&r9_R(4Hm~{oi*JHC5?uW>FqcUBZhP0
zoZ1>>M<GbtT(KU}(^r%&_XX^u)LL+(qi-ZTqA)QkUA?9BTX2?;>r7u!V$^lsLW`pC
z29uogz8RaHT`yvw#tGYOl9@yeF6sO-w(6-f>V<9tasnfkVO%1u<WC!PQ79Kww*G_J
zTMn&m8ekBRkl?Q5W${Frf*4@LaCuI0wNPZW)dRAWMzbS#|0a(Q6jJ(%B@=^V+O?^%
zGa=y8+@J&jj^yLP&8Wt$=On1xPOl|XB&|)$(KN&gf&dJVrjORf#nAX1kfIz@qS<29
z$7MER%nIHsUz(KX+&9?no<3zAl}M*~d6FzYjDo5ZzPPokp_RIPmn8q&?gx_xU9b({
z-;=%NRQ)od{|**zGUjHNyb1cnGd0SUhQ*i`;@*30&5#~e@biqj_rVJ2uTGrgKC5D(
znL9c2*g>UuKjV(l9&<zuC!f6&aly<;5cws{RnV2?ECDs1pob?_6K)U>cK#CR0e|R?
z2Ev$VpER02q7MmSKfV<h2O-hy6~#F;1m6ZyQP7Zn;#$@EahzJD&zAncm7Q4BUob&D
z50*_Ux2&m&l?3d{N*+dbM)y2cC%osBeC<*ibIMYl<v2X-{`URzNi|$}D|ms_42|po
zqgW0U6fP<LT&_|cvj$w^;Fz=|tr6+nOHdO*a>bX2Oy22|9E0<VimSHa{zB|<=d-4q
zDx)xSRaIeI%<~s}r^j8h4g1sZKY!W;TZ8t9oEaJ3c$~S-@j#kq*(c6Ka2d)3@A?3>
zBGS9p%02;IE;gkq_;-@5`j{jO4zOuZ{VPc>>n3=C=?D4L#~6&|`d4z|;65;l_(0N?
zLfYGqPNjwGRH7nH6U1H)!fTJ|D~BRGbP`}4NSiKpd#v68>7`0e*RB@kl80tuJEynA
zq!mFie0GUxb23P*T`O5p-e-k-1rPg(q#M<b=MS`2K$OV;;2a%<pqu8RVCglMF8iWq
zQnkt9wcSBPzu%(7`P^tL^lKn`1IadC#S1JRy$&At;Xwog<$j=Nzv`c^!x!X54Xb*H
z-KZ)jlIMrfQsh;^0<`=0Q+9V#k#ozzOfuM!dQ#+TF7cj?$yiR0%Nj4^)1l^uNxYD{
zSnHOX!bkj4t+)X{{1edb?xm~Unh?~j+n5+(QvSNfrZJo4=3a=6HE>z)8KK)k?WEC_
zs9lyJmwz3Xg&QWq)FF^BP`xRc3%FK-#i^n)oHLs*(pX=PyK0kVFs3CTNmdTD;c7o$
z5P7%37OQ-jPwnyrDK>Mf;CHAJ6Dx6`n@p=AnPe1QkG*0SEx;!>8&wonP8RIBffY3z
zo9f0@qZ<l&w_*4L({H_T?MC1+i!7T<GV$T-9zO^JdMgJiC6i(ze{eCJ7{bcp4yIoB
zwn7_D`6L$sK*x+>54eGyBMlzofs*0b0Ku+UtZ?}LuauBer+9ZJZ^mL^{%PZD?H)xG
zS5;MTeU`|Nl|OA*uslllD^Gwq5-uqe4OUGv3JN!tHjp8soRiGWx}L}3M<-X~lT5B`
zy@<G%jlp7xuHpPb8;f6t%6HSsOJn0>wYOo({^*^_pQE(=EVkz;sqLcNr!!2WmXqjD
zFCK?QdP%86Yj5GGPpq5t_!z@8tmh|>I@eA9O-Ur8wWk{YD?;+W%tY28RA7J-{{%bN
z1nT_TFMYWYr=~iYd{j7=kF;xpl-d_Wiro;-MTu`T{LFM3X;6Dka>l<CLd%|V%aqR%
zAS1vCSZvAvbz5O^F=YqnYzk<vQn01-Nd?za$6`l;>Fa?Rs?NcG^^uKIsI@S6WNwqm
zPLya&u-ALnU%VAoRuR5Y?%BB(cDXI-J6y4xhRRP65Q9;YEs+dmMGQeeV<pm(v&?Sf
z&H7WS2qQYbRBVx<I*j7AfPL)|&*VP(MD=O2Zf#g=e9M4>Dc!9gyWaMAi53y2W48_G
z7wRqdb-<7Nw10K-tr{=6{PCA>YWpwv_sjU>J(RnIIF<ll`IsvJk^W|{Cl2o(fvfk!
zuZAEW_~Q{GU$M>gIb<w&$-|Tzr!v%g$t)hRSIv4BDJ^{h2$$E5=FQOf5_HimNqsyE
zem!K|CLXM7R6W2xb(4x+TDWF^`~&B%-deaw<QqrV1f!mdRFt2-@9`yXh$&U9$V*Qj
zf|U_Cc|l9_%U04cO&6)rKOd#y=H{Olp|{lZI8j~Y)>&i;wc19bymX1=o?=)HO7gB0
zHh;>mlby-SoO^p6D^leejI(>@w<3t*V^pvh^ux^0B2EmM@vs*89}%64Zl_-Y#MGmf
z3x_blj%)8mflWo`soQgSIV0J_kkLf{{AwaEtj}cWbIThB4@9YKKPT0vSD$)rgB(YZ
zt&0wZSB^JD*4udm@q~PP;Fx^!JZIlj&<aDy5SxXzU9tbfXE<RJziI(dg6?Cg!h`F6
z_4M_nL<Z7+8w#HFuyC3Z=LqD!r)eF)<G%7p*Nd0;I4K8|?ABk?Y!p2}l5I=0(!0bw
z;!>!Bk?l)B?sLyp5xN;vJ_-=`ZG;g`Cy}S0NUOGNbY}SMkx<xVZcYq?_NB8eL^uZh
z%pydvZ$#3|zpLVKZahT2_T{)^@wsd2&{F7nLQ~9zSEOtf(aQ!dwFMi;t@+<ccRlGz
zEoV6SK=9IYlQ=0R{4+QqsjsoAPdF}ix~dMb-d<Owl$x(WGq6r?b5QnJP=jsi^ygjS
zYj@hn4SK|L-_7TjsB5N0z4gNrJ$*9)t0mSp8SI^>xj}MyP6_aKCgolealv+<<7w~-
zxt6=>E7ZyEppKY&Jc^>nBnSI;KzmVR5Z~A=&|+sitB{eO=B+V2yS3&aQ}n&YFk>PD
z6-u<GV-1-!w@2D#&(-H%5(MDWD%8E{?PNtJ#huABgz<k_14F~?b^qcst_~b*s<Eal
z{*(G-ie)yGue?D8%PXX*n!=OphR}slID*f}ZB_E7-SyKtAiKo&C4PH=EII;n(;9H|
z^*G%uhjK?%X0(8rUN1SOgK&X4p-fa;6sv~5vW+|AqFb+L&6o><PERA6veTN1Ri3sV
zNI#d6I^Ct~D*c)`GovsPSHn+yl_~yp7oF1He~o)He`tAzv`TX1hR$ZsI+Or&X!nWh
z*7*BWl%Ny^TU(``nt4L!JSYY~6_WL!%2=8en+ZIhW9T?}>ke&=9j0;e$2rC|m52Q+
zkMcdAOcxN0wu&DnebQTFGe`L#Hpgy-5*;tIn#r_-LGjO(D)~rrL1PYAko&ExzoC0W
zbYI7$d03b)xPbr71UK_V8Q`1ZJT*xg48{ci#Lwcykj!i!on|DHq^&b5>(cs?@35d1
zqOP=fH@~GaU#xOT@8ejH2(23NJ+HKt>9oK}!&)OjdlhXEH}ElE80`so>zBpUC(ROR
zy3QA&?LVTTQ!)LPznVP1&YK&(FeA9vT5cD>_(f!_!fb*s&*|Nt|A|PJGtwBp8$)p5
zvx3mq39{K2O#9=fA;{4-WbH3VchKN(b}UF&V%<SA0OJL_0|Rp`u|YM<SRHxw@NxD}
z=}mABBff=pTZf@+V<3JbIV9Ru)nk0PtwN~*!;iH_{XnR&U{}2rx@hE}A@6X?K6oa_
zy*cF3C-Y;ApB}jVgvL6I7p=CQ&oTrzvBrHL8<~81%|Mc&69_<;LVVL?2hG%#skyx4
zgG0{6(u!0+7c0=q7+%k7BLgsMBoft++V2fwa%*|Cac{TZ(r#DuhGiG{j3gu^jp=+}
z_26TYXhBRYtb1)R<7E$c7^+7(7%72lq;Gqp=6-NGu>GsnY-j?CyCLz2b<ya2)Yee>
z73A@Cx1!b;dT)}`%g(!)#KtzC#I0`8q{de95l9ro^Iw7EEz<;MQhV!?-FWc67z93X
z0i-wz$jk8#fv_tainLUO#Mg24jG=NIhq6N)TBrjG#cC=O4@rWiXkrn4&l)hEkkjM!
z0h<J$vc$}?_IR5am@kTr4nG&fB4ub;GTPz(zG=oj6~Ut_u%7hR?CC{4SpoJ&P;~bT
zFZ)4P9GF=g^EE)qL2)O|NIwph?v*e^*2YApqhM~YW*q#JJeH6o;F)|q_lUmIF^v1l
z{|!<^3}3<oo_0t*&l)4&fn-ZTG$xj0u%M466YaxtmwSeN5_$ps#$Hn2qbg|PwTfci
zAL&86U#_+R<N+UmsGyg$z>sQbVg8xoExr?(G&j_SfM)j<>xr&B#!X|{x8BMmd0qhH
zyJ*zOdXB5W=W^HG5|*8N0PqKoh8WB$s2eco*o3gDP};FHxHGcgv}IS;Ep@+72M-)z
z+wC@>Ex_h;&ZP{3Z3rkww|R%j(fJ|*1*MwfYn=eOY|R^m>}cl|k>7a+9Xk)Y$>P(N
z>d%@WmlIXKR%usx^BiDsV@o~Bf_BZowFjS8rJx9wWxM{MlMnJ(q>zI9EdSjz8?roS
zaF^t8QrN5MkeT^Wzrpo(+^q6+ZtzNMJ?$9<1&>qT?0M#iq_#A~cW!PmnM1{g5bqkP
zA_p3f)sW(M$a-l~Iw<3}LyXv~@PnK8TvO81=vf*?2SAOiWkti11WO@Q!vb**`Z<P9
z1<GYS{<JQiYIq?h(gscQovxi^n8?+tuhNJ`4*uy<ww6Ji<oX?z1HpM4n)GL}aegMX
z#^;IHSP}WN_~a(HT5aBb)ndG9ZqtY>r|QLtNKLfXf`v;KTZd<*(K3o?KMS4)5MR`T
zOqyt(vZ{$;UR+QH6)D!!{Q4G>$l$29+WR(UECIuBNI-{lQ`EOKS{hc&fZ~-9|1;KV
zvTHH+c%AG?{P?)XW|k(}1%gh{U7N^tD<?XKWwnn~hW$`vKp=}_=<Zuc;5VUmL^VnO
z4}^jN`TFhGKgdrt8Dc!U<6%B@<B>FM<HVu#hNcp)>7p->wCR?w)#-1u8D}GMAgPMn
z`krYTH1Mp>U9fzzg}F<De07(@lyoq?@Ygx+gu+LSND+-9HsicqeTG|qI65?6R3I-h
z|1S%clYY;i5#_Uj>&)Z&b~TMVsPdHuK%uR9#PKbbvoHJ`b`DP`l-Q8m=hFm8`l4W|
zZ+hd3k{;)zi|c(>DpRbUQP@^`iMbjeHX}~Lwss!p#hwW@m?RGdF(Q&B#j=0_M3s4G
zH+1Ao&?g6pfbNXX+Bj>WswRe;LT@7g)nj-{BX9q`f&}DPHmt@bAUy2T!yfrcRLkuo
z&!a!+P!Lb2V8#4xl~xAaCC`I{{dc{B71ePm@97c4szgE&3(W{mn2|F5YA}t9JAlys
zt{`x+omfRae8c3CbF%b-<pKW1yqhaPu)=iJ@V2zu+b>0A_|iMs#B94#w(H1y)Q)R&
zDz9#YM<)4ow62a+&I>68LQJN<#B}kVlfi)U5PLYKQ`-PXYtz}~B(MEsUR~HGd3=M$
z5~at^%1)R9gQ83#gWUlOG~;`q>wOYBL@h|cG|T81hk(&LyH5mH80|zD%GG_4OGGoi
zaNt_Yz%l0AWH8+%g<2R)CBuM3W96SKB7c1u){q-R>FdpM6yB9$AbQ4(Y&H(AQq&{N
zg&jR3cX~Ph@m`CMrhs>IWyU1->~w(T%gCJDtR<89YmyI^=tWts>C@`RSs_B~8uG?s
z1w+BU!Hhj0j+r<X7sb+~Cw#w|4|aOw^o9sV0>dnKWLO9ZV7FOx)HSxHb)mkaXeo<a
z2oa}9Vvf|cz+Q-=3|L&h?z^8f^mLuK3TrQS#u=_Xzx%bxfN?`DDMVFB!N&%tk#ao*
z2YQ_Uqfa-(?_<8hIP5dSxS1U^ImV>A-(=A<WFfjCcDR6QbYql%OLVW`crZmKbFVA#
z%KX3%g9X`|(4p5QRpCv9V{Xpp!L7$SF8`~wydTZLI4HH_)Xnvmgr8E%!xe0$ZN%Pv
zU>Fjl@{Ww->kXCEV-Rr(_23W~tzlkbopESGxuH5kYjO|WEoPgclntg?=RHLkB(UVs
z3tY$%<n=67OTCDCw?rN+l&^AO%b#4NlH)d^Z)rvMhiB{J)gcBytY1Fu8H|^%orxea
zQK4rsj2mI*V@xmxeuBtXqMC_Br;0r=qx>?opa?%Mz#&nj5KZJ`XR={ei8>!}R9?P@
zmLs3Te`lXV5l@s4fByprkO9@)nczOPZQEk5fGQ;-B)|j1Pg$Zm%EK0=8v5%HvmQFD
zaJ&8Da|q6QRNH)QTWopYQ>5OvY-6f#`QI6S-{wBxDTf(WrU#3cF|p~%8DJ}tEv>hI
zRul-d$jyS;X4<VgEn1o2--pvrV~SPZxGX|gsj?k+>2Hl#8L(%jq?$hxC^`x3V1;t_
z|4;&jmv03dB~<(@G2{PeOj+pT8sTcxK#*87Zw?CfAD6$JIC|ZJKjn#MNoTvC?b;dF
z6kHvsQnI}@Q9#d@=vsRtML3c&F8dc-=XcF4vf?Zb01p3IMO$)eFR{sgyqa_RVlVGv
zgOR9uvGNxHs{a{5LUN+NyW45j9ilHAmaBUmd>&bz)$LDPPz}gN4kllxp(kk$m4|(F
zm<^vFGk!8p;ly?B!ejTfsM@&v9H3`(XwU>~TGlqfm=S*{FBtR`GWh%9R?7miRdn|+
zfi5~1=HuuTxDPM@e*NrN_{1BAJuujp9D_oEW@!k0dL*@2+RxxggXN&In3oyXX;!2&
z^9nR#snYngt`{pYmN(S8EYBb@1*}H2ZOkg|R~yc$X82t#GYVsyku8qj{EMXiS>}@`
zT*#EW*KHpN193BkK(A^xh(%DoNl|YQW;DD;)>RdUyzr=T6b@>7w?5FEa}rJKMgQF0
zY0n!Xi~)bQ!_qb8xQ7yt?D#Q4kb$N*FD?}bYisguEpnpoxa;37ZlbOKtIxCvC{Gq!
z3+)jpXjhTKfX4bNq~65k*Hj#)FIMn`se?JK`DVsaOAQAM7b9CBCTa${`I(@731m|u
zoZwdqQzBves6O5)jq~07JIUKDtsE3RACr3M&i-KicEgsz+<>A%sD({n{Zoi<azL^u
zMIhisk5Xe^(@l8lpy`<WATzDL)Z&LK_F=0|e*uotYsE!Z;Yv?nta5|CDba;aA0WtJ
zlpD@zqNVlz5%FEE%u9+4LAY@cX93joKiFm9TWu?q>%J+enh;?^M4F`Snylw5qrm?R
z&pmD3QsfgbRML}yt4}I^!^bdJq-N?Lunv?N(7}g@pT6)Bb&-+32T$2;5b$NMdHcH#
z6h_X!-vn$YzfXMZM@5aRpH_q5x9vok7GUMk<4676lj`23z?u|_wB7kNg$EuZTXca(
zsUe5C-}E1G<_^M?z*T#!dpw!zo3QIWdZn^({{SRP9W*Qj0kT(0K{jdZ7)e5l?MypB
zgJckyM4C}5f)G4KmJLwju|nmu(WA5*Yh6xHl!LVRobisH|5?QN0K=j@3lKjv*Id@Y
zLg*(Q$~U+7PX0AOrf;D+d7~ZNI!;enAg;i+_ij^2W+M5GT-T3$LY>;X=Of1N_JOVG
z{&S3fNCX~!go~u=?TRU!&a8&L?mqsNlv(SxlZ?MOdb(IO`sZ*8?d=$hc6~LbU_b$8
zjE_Z-ByVL@QtX}FEC%bCZ-N~Q4Ve@eWedRmB61M}q~z(ap~%<}QrF>$d!4A&D-V*3
zF=8PxW`mu(OmrnW<FSzIg=5wvgRWJc#=)b4(OGh_(3a=YcIF)h+f=dLx_r3)v{Al#
zaPQ;vn0?)d!=8>O*|<j<PwFLl7&eS!M*d0nc_n~-7l*y$;O;BxBLe@yuATf{zG(m!
zfP-?3bG1#lp&xgiina2skVItN9#5_5bmZ(q%n{lMj`k!BO~Mb49LrAGn&g8F2-)au
z8@;~f)nMesqdkLb{QK3jqd0w-o}QCpM;t#eUUi6N2DlVJ8~u7U_^U=iWCcij_`Fl|
zrKXEqkh|Na)}CScnjp@_T(QZ5$@kZ)qG#`F-8jt;2w~eZ4liPJT_8a3_SvE=hM&9P
zZeo+Aqk6)h2J=_cu3X~*KfYuO;{gLk>OA*MXm(|0i$vC$BOC(_(XO7}JcCSW>sNvr
zIps^|Fl1;&5>|dDp3g81bRn(S?Hy~?=m&&A6E0e#X-??(kfc#p@!=bfwfr@*V`VBd
z<@W4zR7`J+jK9{>rbChHk4oH{t!4o$T6#l(Ap81LYUxwctZZ4oSYYiA;U54t-4Wvb
z5@lU*|Mtw~@o^*~Zi_a@1@m0Fi9|Y5dehjd08b`GDzlSM&Qrk0oEP%lJIOK#gxikj
zq`5Dfb5eZJ_u60A*Bh{xZsAGJg-ZA5Fbb{Cc0O7>o%3G#2*V*TMen=^`_NIKad=5m
z`?DRbekE7^2%3gCMeA>04mIG!pK~`QWd5foX$DHr7S|3*QKSzR!wzG;Opzi3JCFr9
zREh^5_CVRNH<iJHgmtmEZ(E`e?oBI#_cP!6q}p^a^pASQ(JY{OE5e~gIu(+HYG%RA
z3o|C8regv0^sm)EcEywj^l>vV9mB@34Gk!4Io%zy>P~2|2t<N%=?vD0^0bufZ`9l|
zQj|BiM3RuT2}}&cTD~^*xQ!nI9js#mltF(m3<_nrlbE1`-HqVipMG%Ead=Skd@%qt
zc}`gtudVQ}$Klv_p;%V%P1a=Vo<H}1m1RuyH*&RYES)h4w_r{jABIT=*SsO5(30q|
z?^GX~Jfrk${wyP~*gZzaxDtC?PET%veOa~@!qr%!X&60*i$ad2@T*OE4rYbAPcafu
zWe4S1(Q|q3PI`1g4W0ceI&g$Li)N(t`6GdScu>VW;H7QOdeb97QU{9<;VxT#QO2%a
zDkSL%P|A|XO=UcclNcUce9+WN^3}dNxp)|2Z)y=@x5Zs?YMAFP=GlRKCFkj|L#<W^
zOmjM+M-jOVtewd%bgg=bP<(x{5x4L+_<1|KhMHJM?8Q7OF($u+)Hf~Tmwt$`u4!KE
z>K(mx6L5^ar#4}=Y51Oz1kAF(`^dy$QfJ2|!yd(G8B{PRfZQx=@Tcc1Pz>vybZW?+
z+myCBFR21hNgCMNpZBvLe183%;Woa<hC8{^B7E4?1VyOYb~Bn@U^Qa%=vE!d0({Bb
zgM6&55oWvVD2-$`JQ%qko*iJ9ENmKwU*L!OlH(~Bgf(4Yn*1@Y)I!GM2n@~{xz3Jx
zWH~?T0R!}Ln0`KxW9{tL6vo~7DATpirC|bc2GdNZG5AtC@-k8_?ezF?x$rG(GIwAg
z_RdMx$%yz;AF}jD@zouCq+a+p3!r|<bg%k_B^;0zPfr~O*XaV~Szr&*U&2mH0UvnS
z2)SAOSnAT4TP+_USE&i*lkCc19I8PF9f}vb=F{ni{H|g^*#vPWz1yiz=Tzq5I^`aD
zMYYbwu;{TXqlbHS0bIs=X!w#hvle31nJPuaw1XM#&+t6iZWaML_Vy;6017d4;&HOu
zVCh1ynK>`p9He7CTaWc%KcVS0i0c0?0Ar7p-6duKpRzzCts~;9F7p?q3xiXC))G{A
zp?A>O<ep|~FXo(Oxx&U)Cf|k|yK8jH)rJE_%QdROA0GVK>{dp;-I<ovvvPMWw_dCO
zt4IqRxLK#0Q^g@dC6$CA;(KHA7C|WTWN{gd-`-0Mbyb7^7?)2ek^1`6&$#mAu}^$^
zW;0hHFO#VDd!AVPRV<UN*^SXidN2K_YS-Pf5h&Mq3w;|W$uT9+#N*FCYg$hluegVw
z?9{Ef{Yz78y9U`)=@<7Y{CCOV32N2AkjyYA)tYho<t#8))i~WE13X*UH^UI31F3mz
zgQ;TqC3DF`UID()F_1Cjjvofs0=_){G!b<b-wPjtt5Jx+F^D@e&2`Qz{W_a`Eqi_}
zqgNhmCj##D@9Tx=g!8dslU?=YMSP3Ew@7Em!1gV~EU|P;FH$c+op3+`qkLYPKo0J+
zx7F$8XA92&*Md=6wzbeE7#83pOYn6VoJ3sSj!r(E<mWF2u6ns9I<qtq+QPCp^@YXj
zsoHet%hcfYD3VPT;pZx-923Z4efFk6Tlv0nFppZ8!t=tXB&{6di;;jGr3;rReo#jL
z)b02dFpKH-1m9Y8Al&=f%(&yypiC>}5pUF-MPeQfUW@a~lD*om$`+ZvrYpq;JBBUM
zQ3-9p9PC1<PSQy_m80|$E8?b6di=TVBBG<g`N<K7=5(COPm%1%d1AN|xh7;vPCSvJ
zRrQ<V4_DA!?C8RLqb60$T-e&@eSCyw>-C?dnvXvE!c`Jl805vN=Rl>1<}}L=ri*q@
zl6x~f-lUs{{i!&cvxjF}x+gu|(7WgEw5DHVDb*UW%fM=e;QYeAHWQt)qfJZFD(#(D
zLSK>}O$1d65FG&?^2%n_DWk8I+!((AGNpPVY3p^(gMO({`#=hCY<+^3a<ZkTk${v7
zM2IlQ9R|{VuEtp&b;@|23TaMbb_CRm<o*&~Ji?u3s=}G2Uw_FR-zTtlVfy5PJFjVU
zck5?4swg9eMWO7FV$zRuD7$Q|2D;K%iT=cDM6RaNSYGNh+bM*Z&4(*=_yZr*bXO*=
zN4B;0x{Q)kNG_@vOz++@dBt2&Q|oh$c1nB}DHgmt$+kb}D%#;Q%W10HU!k>8SLHN6
z!8+cs`}~^5W90DFn+sD&&8EnvAQ%x@QhMxk<%I219lbbwhi>zs2QwYR4;&jT=y5(|
z=Q~k#xwTK0#fHEwX(U%Wr1)NqVa5$_6oXiIxi|~jkW0pXRtr~>5vHLZ4GXUl^cdk@
zmu#HMi;Q*a4I<H2#*!%P)<;CSV{T8to=iU^VF@)J_<<v?S~<EHkFN{t7>A-+b{gJ^
zlm5PHVUqrWq9?C`IY&{jAqCl2hl0#bpOQJ3OmSb9wl3mTbqqEDntrr2+1jL2{*>3c
zI&8uNkUmT6hGs$?!uaH_1ahtjt_o=&!}Nro`x2+qPu~`VR2U6{@S?%S^O2{+4Q-d<
zN672ZDkjOCi-iU)K%0bTP~@QD?9BRI6W<5rke>trKsg9XV>-@6cb^fC?E3RhP4cyB
zF(Xw5wJ<SUQ`^c}ZdWmxBu<;b!+I#+FwSc+K0#PW)eo`UOxX_(6m6`7n?H#35MT5o
z6n4&0Y;?B&;Hy^^rG1FDYP5bK6w)@zU)4zKph`z`8RDvvK{(dEt9M7{s<a_8Xz7A_
z4vE$cl@_<)`O&;eT*|qlV6)ja)$<|xJnDx{yz9UXyl<%I*NimDr<T7@&mCeXlI%`2
zHtu-KwQnOg<$oFzeWo@<H-?V=d`O!4JXbE;0(ksha^T&wujZx*^I`5)b;;^|;;$MO
znj;phAtQjPP<(6*{9Xs8l7zd|Vn#rCx6Qx4or#d?J+HsYbNSefh5iBbFo=NjI!J6C
zfM&4DoQww`)V~;Kl4MfPeDSCz2(3>CL&|wvf5;n76wr(p%1&%d6OTdWDC&%NsdWMI
zz@R~9)r~3_Cn$U5(X8Xs&UDc$Tw~Y~WzS!~coxD-zK7?Cy`p7AQxZ~P>=VTM?D_}0
zQFN%j@nD$BAC1eS_hl@@|K?=hKX%sL4S+QM-KM3C@85)iguZ4&Lh_-ycpQ`A>UoZ)
zzQ-5KqI`;h)YCJ}6;3c$Zl{0#nnCo_1#RSSIBQ@C2pbuHnse|+h~a3TOy{u=t7~GR
z28cp(M@s7!wv|TuS3PtruG_a)O&P+K%83*jjE;$q66cz5eS+@;(lQO03^}4*8kyG3
ziGzWrj9#!jt)NEp!oFaI8!C6&hpyn$i-x`fU&o2J2^KX2MuRl>5cTYecCrvI^8Du>
zbR1+z;nMm<h15L~IF?q)64|VZ_&*NG2tZ^YvKSDV>zDeo+q`K|!?8Yw%?XB_4d=^{
z<KV9^@ce8KVTQ5`x%v_v<4`Al-%%2-!m_e>oWA2jh!CtjAH{7)mzhk4IbUPPGo(F9
zmU#c8OCaVpZ48+Yr~XM;ZfH%^q~&rtH0|hkQ@fXiHBi0uD-qdJxRU(+b%(|PNQRPC
z#K@n5j_d?<RRw9@kx?Ds-p_QDsQna3Tl?yD#a9pliAYMhSj1hQy16OlEjs8bS6I6$
zU$3I9UFt=H0MW6<39Am2k&Ks?vZ9VlY19ut^K~xOD7xgc=X8^AceI#Iwgv(JN}+1f
zmG-HZCXJjotFdI&k28(w#)<>;w@(WAckuQp139{$C=A4<>qbSH7%Db=)jbtHHNux5
zFC8r!C;7T|N<p@(_*Us(5n027f0JgyiS}sVyqaIu7wb*<C)p3P&Lr2~(Kn}S7*R#|
z*POTJ0hu1De1s6~LJ<(WVzkqo<eG6ju7}F{qkml6B}i%E0ls^c_!J!`d8*607=*Jn
z{BIFO(vXGVehZ77UV>zs5CeP~dukaSrX%GCYYG*M6z3C?kTYMSGZ9fxJVUE}FEXGX
zMA;rH3_BlO-z45ZnIR!_<3T|7IF&%VL?zidlOvK6?R=dg(O0s}OgY{ac2c0h?;+kB
zxN-Sp`G#s+iPhlSn5<YJAshVpb$UkQ+M0Th>K9q3_g}o=kvet_rsK-9&8Fa1Kx>`v
z*(*pKie(D`b%RNER1L&*MsVxVphB`;yO+4adu6Kxv(HWRHd3vl6VdO9jT}W~QY4f=
zV#<)r+R@aPpMCI%sR^ON(X0A;+RbI=!KzcLKvVO&EKgM;LE4YUpqnK0K153~OUtH%
zhL2Cdw*aX*_(zP|HKvF0g`ldzl19oHU9k8@lXYZ4Y5NA$=b2;=_hy1TzZU?Ueb@T&
zPrsl}vS9J?7`=e{+wU#m-bA4)Q*1^-6nsx+$>HV6{%aXaw`Mshu4mF|5V7J`t8Q`l
zEZJ4+88u8KiJ<0)opndRnJ{{e;ujePmt(SFl;Ohiqd&m|%?1r!+WMCd+-Gl~%4IJZ
zpSOJF`%{nBwB<Uof5{%4FM(8zf_ONb_4fgxg^Lu8MW(Jb$sUrmcZMq!z09i2BUs$0
zw_fRO9`GC)J<)4*Rx$f9RDlmfm>eMl`&>;J?Ojky0<N!i2z>hF9zQyiw;t1kG3u@h
z6UsU)9lna!2d8K`v9@$CV0HHmhak(butJdgBe?UU6*<seVZRqD8N>PHJxsWR;bGK)
zCq|YPks^NllUJLAYsU-y$ho`SiL8uP(QrB7Sy*mf)biLxwk#rm$}&}6XFqI{S$Fm+
zT)1V^Q_|i3srg~yuss`Au=Jt)*Qm2$w|-(o;Frm>NUva%mB@zniw~jhLxElD$=X?H
z*-TVq#^2n?8Zl!flb0qL7b8dqT0$a9T;*=7t7fT<rW#e@&g`?px@434{vyd}2k$-L
z&iDql`CFFu`S)*HUtsN^t;ZN%?s5BWTyEU4?y4+R78IFs7Kv%ZhGH<}>)6HQ^7MoU
zopR_fg~EN_r@n9QA9;f}n&D%^6E>8yMR)zz_Z!QbNh*b)B-51W5L0N*wX80PDw^$u
z-(qaoe2Z*5*D0|jNIYuGCY$XB@A^S`b%R>!K{;bcqvmAl^83LFrgYXC57s4wn|eB#
zEtVUvOS34zDoI5#+U`s|m?fm)sbZXG2x%YpkZIlJKZ6qhIQpX$1LJF2aAotHke51d
zXQ_e8X40LE`bn0o`C3K4rWq<E!O+4N40R`j`wZ%m$8Od?j1oc!a)t!0XN^z{MQzCl
zM--W|;02x}tH0j7l9~AqyP2`?3a$EI+`V;FRPFjc{t74v(jkJ>fFRP{ATV?Zf^<ns
zcY{cWbf=DlASK<YG$<h*(lOF8G{0xg8|Qq_=bZQa|66MoYh>Z<J$v8#dG7nVuj{&5
zd!Zsd?0`C0Chs9@(n$bKW4%R1!L&gxIH|5SEw1y;6h^$<E}9_@*Df8@r8HSL0pg16
zK5DJkd<NY*9nu|04d*uI0Qp7Nm&md3aiRL;ZCAXJR~b<F16ku(L0qv*2L|pR#^sUa
z;g$?jSiPfo@8>uX!Ixa7c9L^66#7nMsxoJKX8;3ned+KiTXkXo$x!BRFeQb_fj11#
znqpIGGG9!N^YHF{zh><BQIlrUTd3A{jMAoz9AjF$ve^ha3RK>yfM+>$+-=#fsS0&j
z&8P=vTi2<Qju9ANOk|%;&9LyQgymG}3#-I`dtZUE7*i*QO3p(=Rk4^V!1je>P2J|@
zr)?_V&9qwIswFZoF?KF)aPVkQQaRPAk`4N7%P$fGXBbz4>f|SDnHyf@W?6f&@aM<t
zao4X0xA2y9=lz`PE(Ppw+MP{SXj@z1368krRGYN?gC=25Da}4mD?cc|86vR3uo*YA
zK`Uj^@4uL;Lhr3#{TVR$p^9dDA8vT#g&^Tq(Jvk4m!2zKj#~I9`#zfUUl_G00rV)-
zR$WD|@%IbpXkT(E>dm&`eeaMYxC`lRi**@WJV`q1u_Xi|NhPhNmwyr>{#YZ}32%AA
z5CXML8*U-Qi;(cD5l_Q;9bud}h+yYH!)Kd^f_H}3!ytu^;^^XK37(e`Inss)B|x9Z
z9nd!}iJncE=h30Gk2$a6;#<b|pv$em4xA1fNj{7khf*X%Clw$>bYN0&2)yC38k7KE
zG`x6%I2UmagqzK|QV2SgMmgx^eEF)9!*_^}7t(92$6f)@@C7L+`}KlaiN&ESD(=8E
zDnS+_+-aM@lkL#!QM`s`Y$5!>?bj}mmG#znELCJ;_v9I#@$e`cY&4e6n-t${1UdWD
zqo3<KVtwTnQ!mqp?nqF+wGta21WUo8et^lFwLkQ}MsPYe+FltV@#sJ!5+g_stu}YC
z<+x<c9$!8oOxTVsF{OtNBF8l=d|z$>0BU>lG8IEUn+k^0_ZXI5##;>1v0V)E(y_gf
z`;tPpyE?dD|M(!PR`Ten0wBp2A81t>gx}i~)TMl`t9@iWX78!}Yumpsv``oFCMm_-
zZ^FDMh;r1thE8QwZe^#?;=8fn5BlB@R<I##;&U=jE^;y-0-R=Vppx^<mg<o|uuH8P
zo`8R$`~I5Eq#E~J9skRnrq4ncw&eOWY!Pz-5kdz9>#l)rYFOunE8;h!A@%1cD|~{6
z4EH>#CbzhtdJ2#<+z!{c>%btO?DTX|gpjBlsQbUiey&*_XYDNR+jy37K13=_w@(K#
z<WhvgfPoV)<l8$YJ3@wr=LIx8uGz{`2YyjC#UF|icneCyB)@#6gRDouOR!9SY`>WO
zi_XD;HmObg$4sT)BaHACeWaqE>o3bgKfT#&yzi_kd_-3FEjnV!2r!^oVS<0|RewVF
z12M0KP4bY|QlhQ#e3f5z)u<_Z65Qh_YUxxomFM5^YXYKX>?(EbX3qfQ2)-#7g;J_A
z@a02eEKmS_osOL7b%leE2`e<ry`xI;?p1Vbyqjih-rCc^a4B9Gg~3x!;efst@?aNe
zLMhOB$tBRJMe@owlquV8an3Ris~%I5#G_{XSg9|r$D5A`b&BoQWuWXmbY<!pu#P3C
zx!2b?JWy`-+q{ftKkfSjBkD}kfQP@`67TL7Op9Uvd5<RK_6>7TLJBhjvpCA@k1yiU
zX{>q(P2%12=u~tz)2&N;W0Xb1V;W>;&vyB4#Jufq@yfVf?@}@6dAwdye!R0&+Wx-I
z+R+_@@%uY*U1~I0htQviOXEw=sHJq?Rn^_!Vp+eL{kAK1k(HG-<?iN&SXDX@m`K{$
z`2|LPBjTnw5f{Q@H}weo3x*kw>J7c><%oIp%@}1pbeP=8i!Pewfc+F31%t|EmU^7&
zwue&OtF1A^d?dz0dHj*d#4PXtth^}iEj#*44GKaz-HqDZ=BAO*%)=j=h9$NI-z=R~
z7j<G&m{Clj5A({!oTIHx6%#jAq9q4R?1t||3a>h6ZI>h;Y3aKD(C}3>pXhMO!pI=E
zZ~9!uKtjx&ZwBdF1A1P3mY1!O-J8qthyfbY*`yaby)&7~$y#jql*R#LM_`09UFeTy
zHyF;7Ur>K_3e6}_Q_hv1`-VQBYeLvV;%Ex*8%OTH#HaI(qA@VTS~B8fWzk71)9qbf
zskT(qJ_!ty9nh&c9@kx$Pc06YRhHfDsZEZrF^;0mpkJ3an9BJa>!WPYk1-h*UA!>s
ztYx2ezp6S(&CHJuahY4vq%*x`88@?4N?S&iUX5LU@2;@QQJ)ab+9jVMT|ZN49jvC5
zLzB57Kjv;EL5jQOTumIU%=x5o2~<%Dv!hr_$)&VdTO{cT0(Z*>A;^qfQASu?J>{m8
z_|n373J2n!7}tW%*O(2&VADC3)Hbzhr2YdMrv5vdy@KEDB_e$`ttqz$NFqlz;161@
zn7tgTFkbty=&79NNL9Gj^LhKsuRk#nDx<U|yoHZ<3xDI`t-Pb^8yOe`1w#FgSFB1K
zq|P%!=%W8Lx*%&n>_g3oD@v{s&GBmy{=F&^-{$65hKReCiY_^yhkpMbzyD-+!>8U(
zmufiVen-D~J5qYp%0h0`3U}@5B}+wYLC+(W5|Py7ql(;Dw5!H7soeNT5uQq1w`lGv
z-H!qhQ-uG?yZPtw*}!;aO42%KrZA8wtm>j_R2l0eR;&q(CY?9;mZkU=gCVl|QK}J3
z*S1<i|H;D2N^bL`>Ku{pIJu6QKrSmXD9rdv7GVidq26X*$N?#Yaog)#z3=PGGW`ql
zk`K5me*H!`LbjII(f{IUgOD0kM^0i;35k}o-4HlcU1k6Nn^wqIdd73vq*u6GB5yfN
zy3pv#Vs_<#!&rnfgV{?#jSq6q0&cQVnFuwTa8fI>ykuj9L7(LUaipJ&b(7<Ny0^&V
z$9}rki5nRHR1{~_%DOPeai&<M=tQP*!VxUTKh!Ens&lNlw0V%?-zM9~^u4DZ^v&z=
zveq3&e5E%BE*0#LuvrROSkko~%Wux?q?UIFwA0nL`N(5#zDf{ifB4~h%q2odO?!=a
zqz;Fiml>(I6U9xd9N;VMM<Y-<w&awVe~&$*xP(o=HD2?)thcWEo7n>UklbiLPYY1$
z_@&CRw4B7f;43RPk5@}{(;a(a$VSSP+p|KoO)cVKP-Jr60n<GfSd{BiAF)W7kFRgn
zkvxAvIFG_SVj^Otz_9hM)a~NKoa5J~U|Q>dyHi6bcV>l7Ewl|Y3#2d(tM9WGGs#7`
zjHJb9Cw`XXc)%7ro^qkmP%h2xAMZ^L2ARnwD4jcv3r8E5m<)0TG`XFIGh~x0o*oFu
zzH(2%YxbA3P&W2c!(|=~u8GZ0<>A&(A3Qb{8`l{3>!Lke4kA>>$mO-mQ%|Fc>IY({
zTD!SNz9LqIn;&vbFKDK+5n#m8eCpBm;LgpIi4w|j%W7?Cc=CK%@y$_vc;X@e^qCb=
zv&O!j<}PeLA3@N=-}X9XH)Je4q(565`kaen@M&I*za|U*vzElxRGFRk!3@2)7nyU-
ze#XngK|aLMBE7j&10|NP5u~t;>r&b>k3ApB*lorA&Jcxl9wY!QItj!T<)5VTAJee9
z_0Y#p3vIBip-*ps>2io$d}P<&bD^MC<H5}4^05IL^eGzj-60y?rE7NKE24iql6V21
zLw>ty54yqV^jCa_e+y%eD)JQ9{}R1rMeIrBK%LT4lb5qQ{`$=rVXyyuJ%yt&OT>uD
z);w0I4vY(nWng|8WxpK2xs)VzRk$5!6!blT<<g%!mlo|I<@ZDKtk8Dh4H--Er<GwE
z*WU%bZ4T<Q1PK<uL3+>>zHZ@u@tpkT3K?-wstP<=AwH?@S)5l^u!gJM+7_?*sr7{-
zlAj3gH6D7!8=V?uN<G?N)Dc?hPg9s!<kj42L%v<|;SIIlEUaop5{;LusF363Fk{9W
z#z?8btq`-jEEUgR2TDme^~N~9se42ED()d_Q(Uf*()&Se0erVGRbVfk6KVV_vMC77
zB<|{s6h9Wp8M|hU$MRb~6yIL{twMG?d^mlM&q0U&rxSGI8fOX&ZuX5rt*SzHho5a7
zbFssCkXRL#(Eu(JMI9&?)db1lNbozzV%Z!=`x%Rwdr2!6`71-}aYz>s3=>11m|zJ+
zpM?v*w0iDqGn`}A2ZZulCb5FC<#%u?f5ls;nT%dgPa<7Lt`{g1wg>E?eUR`F$8c`b
zxV!orP`}){9{!^bdD<rkIo~PGHGz>&QchSH8YHUUzkS;;PPhzUaxO7!^+PUcLaOLV
z)uU9~(M(+xc_P@d;<D}zQm-UfD=j}=<8Hof$=_QDK`i*+klnANy`$^Gb<?upzmb88
z!~brlm4z#xMX9h%sn@Ys<>FUmOfAh+7rz;c^8NEr(R}uG>2y=&i+-X~>#O65T}RD2
zYS?o;S>2?{q5<5<@lDQGDsa0Rs;0QxH>ei5JV2@9IVC}k*L$iY-WY3QkzvbgsLTBF
z9Tv=N>!puw?u=1koxW2gcZ+B@vlDC~u<z1+)oq7rTMwO+5&mUk_=9gY_1jUB9oJ$d
zJygU?znMgq0fsJ;R{R6gCG%qk7L2H3@3N+Yh%(*GjEkH647>D3gOQUaF@jzp+}5PN
zuo8_fxB0UwrQc)U85#T0d9Mh5kNyAlGaw($1H;}510GZa?UKhar9z`MNO+U^obClp
zuy(p~%u>kf_7>qbDILjcrbCDNR}@x@Y*tKV!tYp}4Nj!qy)WtUFQ<8Ve?mM_tTY(?
zS5q<zhMn*`704{Hqrya}OWmB?%xwh|Qs&!E=bzMeH<?x#=<ZN%F&vl|2673}kW)iw
zlnACU+oSw)m4hCmMM0}*yl`(Sf=}841o765+Wllj2B)Q4IZpZ#f7P)F=F9ujxoR}y
zbm852J&6)plTV8u=&o{kRp_Tb6zuFJbnN;{k)4ecU4+<_Iyc??(F5Hm8q;DZ4r|19
zqUL?;+wsK95zFD<U*xfmr|#aJ$i?|GmOYM>%C4*}WGQqNmij|+Okr!Cc=7FWoCjXF
z7HzCAzCeGn@|bF#m>^H7mSmXv9m)C5*y>w%92CFNjPsleG9FP#+%Fr%ypr{us*fub
z5};%hZl<z*nVFBZz1$K@G_6dP%y{IOc7dUKo%$`&CZjL2pzpG$alFad?1IKV=ZLrK
zG2HJ#J*)3SlPGVz%LgBdSqAz!5hE@l5Br5bP)@fs-^(wlDww;F5t?Y%qSoOwi9$q)
ztle#u>%$zmi+lY#a6cW|t5}obo2QcMT%1-FDr#&;JooKE=#hV#+KGXInN5m7>k0lp
z0^p<a(+)#dH$k<3#0Xd-|I4e(3Pq}8A@xqMtQs_!md7y({XjU1f%5@=t}Is}=Lz9c
zG@5DBLZTe7L55(Br&>ORVN+1QbNq2DtVrUYAKHb2LU7&nMtSVfKNr-$R@Q%g(}oZ?
zvcbO&+kg_mrYv%rIlDl%(oMUuDmD8_uMp>#Gg?^b(TD<D@O#P75a5H(py$0^9UNcK
z1<irCe1?gWX_|m}i;#bQn|CbW$kQDKfl~kF3;yw^MzfHom;=C+Z|Zr!$k|FlHfO9$
z4bjM}Y+wVDH<}x*yW&!90tujz3<*L`VjFiEfyckV7e4}GUo4~nae-&?_B*~m61=~X
zw(n-(=neO@enkH3m;Ar)+a{r}6XPv%eBDSzyj)I!CtU|vj@NJM03SA!;~u3jzasx)
zkjdyHxK5Wb3b-T|q&%<G-?l*sm5A0rR#pxy!$QrHwe!_5H+0OrX#dPK|2z?54B699
zzbS?Oud83D_!H=#uoip{l~CXKU7dR-wuZ1(iKBHrLjjgowx>s<D%K@P&bkO}LH)TB
zF@37hn`yV^-$M%Ix_`U^<{{LJWwo+!;eWp%f1c3)`2F`gx9<AewvE33AAk4XKJL3N
zSo4latg>YO8G~wQeqaXmdBJM3E^D=h{U7o0t`W`UX7k|!smbX|f=UO-hIIv!W`L{T
z!rt+OVW?GpTD~*(bEUsO4-+16x+boOt)<_ismm0_eb!ON1B}~wmEr&hcp;!N;t&P~
zb=9`$6fVzFfxn=2&awFbP5A(4e-L2I`Y!_4aVlrBl4Q5UHl&XgyUk?zYxa0n$&h<@
zug;GZ?4S7f4@=z-Mp_lRE|k?wzcTK`<}04b$Ax^qivmobT0091vdLP`*2v53fU^g#
zuFYpDh9bL<%Q-if7X0AQ)-&rF_ATYW`!bYJng4x_?e*C@ZsTr?_jcucBJH~i6^;VK
zt1ZpNNVXP?8M{W?edUuukTFx1)~}`S0aLXQAdF*H1Riw(`*rKt-gmlQr5zNAUG?tX
z0$Y*%DS*Pr%+t)0%R4Zcf|^gJnH0H)p=L!6Q+wdNFfW%ckpl+`mrLuGBHBE?pCl-9
z%JFaJCOElEstaEojdpgC+dm{R@K~Jnty2BxP5bAA{MY0AP6)Z5g?@TP{Cm6w)ZE>}
zOY)cVJl`znWT~Ac7y?!_rq4Zq{eQ>nTd1#c68NZm_QdNJL&!Va7kpD}ikeGX&bN%i
zA0>bEwL4h@N;@3jIjl2j`QdH)_V`hw^4RbZTJzDW>xBOOaUJUtr03&DT8m5IdH=J6
z3U{D8?_gZlJ{YLmD#e}pIa7f#p#I8YHCwW-eKSuJ=*nX&eQ3Vet~KmDX~GWk3Z?S%
zcxnUlIU1I^b!!D^=K!p_h<k`*HQD902a1@98rs+$@YF3L-DqreYv-xy#6SX4y=9SV
z98x~WJ2NI#|MEJ9jHhUPfl!fMeA20(Mc<~7g8xpd_&7B4?DFuNwdQAV$AYh~E+#||
z*MJwuJvsI26|56+BNG<V=a?M}KwV0(F_B{6#b{GI=VrX|{k^`GsP(cU*C0_mj}yqb
zaWP5(2hQ1+ykc>0v9r<Xdbe?;6~yHtSZlF?d76I{BS;SoCS69AI`y0fHLS^&cMtm6
za1DKrHEd_(nM#O~_1&hX*PHm3x2_msq`YRpD*136WOeM{gb#x<>>3!aH0CaVfBEA~
zaJ?2gamDUIH&@+n!TFqhz>RZfbp6>D;M#2}*0kS69w0XD0ZytD9DExeC9VzmIELn2
zt}rYnK38UJ!lg!cqB7GL=cnuV35s+wH!!%s0d(0dRFW<{D+hTG1!fhMeRi7>m)D6&
zeF>NkGQ#|#dXX_5rPCj8{r%IQIsN}#h+ktw6@u`3TPRZC_WPP0b&Tc_mIGxaL#}#r
z<OE2wv&D&M{!Yp3UiysUfUh83(-)M7LQE4|fsr01#~mx|@OO$2xGkS~fa%#Cz%J{#
zIbzmVP*uk6Yja&3`WaJMQ$;`h9^grU4XhLBJjwCyb>+Qz=KvyPB8Z}tqW!I<^r#Y_
zKK56l=Zg4t;EJ#VmD705aULIXh}hF}S<IJc6R%CZK_;IP;r8VcdO|%y8w?lOt{U|z
zCD15TJUh&}`8Avu?>nJof*r%2L=u>=D|po>cmT|}M$8De44ir&p}qtyI%OT9AHWWr
z9C%Nf-hZIoZ4L^02XNXUdC59OHQ!P%L$1x0q(_Xio`zijyEL8q!O?KZh{m(Ej4QPV
zyGLxp+TnP~lnCFf+ixC>up$Vd4%ak`t{)s&yRb|$oxb49)S-#Z*i;6$UaqpWY=gqE
zm~a{Jln%t8Z)bp$rf>yY$`=oTU*(;8cNAFh7b6z#hM6DTZAFi50pP`?Y0B(`<MDL@
z7HxH$>tU)%^ghqlYg8Or-$Q@<GvDAz;l&`r5U{?jU(4Jdus~Cy6P={BfVAbU4%~dA
zsr&8i7-Hi?Oi1OBAQK)Vp?2ni3lgBZX}>cQ2|M2JrUL$9>OV#60mUK&7@iN$ynXEr
zA}LQr<1~ALvsiQRgrR~P!3#ZV>UcQke{-cT*bF>`8cDXfc4w#Q|9y-7*BxS65bOub
z=NR9Qe~+OfZIl?{16gV^Dl@R5yD&!BVu%nysAxp?j0AM?dcy`%!l1&W^kffklfv(G
zxkKM?7XAV|5#_Bl_tp!2l|~Z3g>XSkrrkWYF!NyQc1bswqATojjjz>l6Fbg9?}?$`
zSzdtzlt{Mk4!I59xS{`5ED3WgV;Hp8+K@TS3qybi#z`}UHx)zFJirIzs@92Jk7dAf
z@5A(;a>ld5l*{D{5*w#NyI-2c6B?#fti35pdq4yOAEg(DQ`?4ub)+4~)n8fgEy@I9
z6SqXN6NNPnzu0MHW3K3dRCk8QCiOID*i?j<EthKnXQA%Nk6Jy7P)T_59=h<zb6P1?
zy!R(S4_pq0%Z!iAaV^OD3-XefGOLwB2Il=PwplMY%)=Z~iGoIvyCjGO9KV=W07vO7
zGf3!Ha+c2fr*jyKF#O<v@G3_7(Z;EzKlZw$L4)agHA}WZ(HID5ucXm{9_M@xY^&W9
z$`FBq8wb?(=mpOnpQGmnXwOb)8lIR0xK->V45chE45GET=!PlE&xCWPYu4d%bWeZ<
z%>n41*S42t>d^64V~?Y)*JD)G|HIz%*CzB7tw9AO(1-`{dw<;cXW_r8z6~>lY8wX=
z3%iozdWL&I7sU^OKz(sH;|dqwBp$szj2?q5Mb9Z*=9abF!-Z|&r?a@4Vu-Vi;h**K
zLxG#yg|{h%8iTvqDcmiiJNMlCxf=Y=eb_;k%ZQw)X4r`|8XsNxi&6X-s}3q={OGVF
z8M?<S$;BIC!jjV(<M?!S!6Z5#$@RZ8Y|(YlRA9kxFVW&_AFkY|qgauFP(=6Uf!hq5
zz~4@Gy17Dx0-2alET`v}F#MM7iPr1t0v!ad$>vRYa_pT9j5OJ5FyGQ>E-4#8e%S`e
z>hnF8R-VOnGQO7bug_%Bq;mVfe!+!^Vf^^>bU6lr;hU?Z((fm?dUHO}s`py*k|iQ~
zz^iwcJy%3+$sEg-P-HIvof6|yafSSewU}68Y50>2BFNJk$mJuecLnLMigF_wG}lO4
z?}tw}09bg!H|4sGDw&-j{Fy@fqOz2@a*pKg-)Hb-cV$nlO<^p5+>Y-I7<h~!F8MdW
zXjRzjn}(+T>gI$8e?Ko|9shk^ffFd>z0vNg6+&iW<(;Rk3b13uu03{8w{b;JsezvW
z#r0Fqm7X|CsaW(`PqXIZaV&5hEIKu_KGNzl#+iG>DtZ~4KB06uWY1D}la10-G4R}L
zM;BQlS@=wa+Z}fz+Zz5%pz|J=2Ssjs1`yGa7rjPZzX0j8{zs%2vIs=eO-9Yhu4c-d
z)zq|AK{R2<dF-R=5YJS|y`b;$47VR$m>Ra8E@8Fap~Hg}&?M2&verz#43PEo#HaEs
z|He;E_8^YN_!c?rUi0-iyu}w-RPTL+LixA0)uc`(i>O^$3z;9-hneZk#Q0yDq!h;X
zZxF)hTr1g0!bd_f`#r>=kUTiporzPaN)kC#PnXY04FT59C~F5bQ|JS^=q_VsTzsmS
ztb$Zt-=w);1uq>wZFuq5=~yBNE<x6laJRn8cm55kc-LCE_!VKlxlNg+1eerI7cHur
z0p<nY#4)I0&V?=JWHyWy@e6tEvkO$zT?Z8>0;0a~=VK4Bl}0``IAYu0#9qYRWURlz
zlS`2OB%2pzO9U%hn+qJ77#{N8>Fg=`CU(;#Qwa>KaZ#;t5ja}kTfrD$*F{Fz!#-dg
zBJVZ0E;P;#t8-M58DP*c=Y1=n#<D~jHyL|_Urrzw8TRU7XLHtm7g?%i(Y*gP2+i{|
zOUVX2UsrItCALe~5=Y}Mi(aHmCKATjx?Y=ks8<Sx-#}s)JV>v@iI_VQOqffn{%xN#
zTsiH^FE8oEf#VkuewOV=7yIPFvLBJQ?ZSeuDw?OPKXn*ZHk?L>HC}d^MiXCk$!}3R
zajV{y+aL6OCiVB}>;*EN{THM^>FxmEf)DEhWS(Ta_p{vqS7H=oma92a^UuXmB*P@4
zmS~ROi3fIQTV$=q-|En-!TYg_)D-sU3r|X?_C`MKbj;nPt5B!`di`AG(xi_<!oSHB
zKdpL7Ceg9HC=8nh;jO9Qn)$GsN!7`&`S`{C_-Otmj1>B?mfM<I&(Luz<L-yxQc>oo
znWrBuQvSHpi#BC;v@z_rBmDl&a#&L$G0#X+GpND?FcfpGa;ZJ2eW+)LlR0m51YLg^
z%p_FB!AHdzQ!ve<jS=Pq^$oM@4@c}H!V4NsJPLxlw;$k~VLXT!i^IS~T9owXx)Qsb
zNq*{}a=L>bpY6H!K*e8SOrh(}gp7`KX%0sWT_pABphTh)6#YvWkYS9kGDERPcz6{8
zQz!B{G)|XwNX3eIj&8$rZmzFH`WD*)Q1zvQRUnM3`rbc2zfJ7Ydg^d-4lnbi%+J#p
zQQE2KXB8DoweQgx$gv}$qXG*6Up$QLF8*g;rhab1YkqT(8&k4i1w<TU+vVQ72Bz>U
zn<5YFi)}sQIVXM$rKdU2usrs?^W5~s_??~G#c5%7P)XXi7PZgMA@Lk#vetb~7A2+z
zG_9YpX^^O-T&2*pyrF^b@sFW^>YVC`i(63-6lTfgcrx{!6GHHkQcj!|E7qLZOq3`&
zO~d-lFqV&mE+B94Ih;r27?g_kG^1OAxv9?;LkWij<A`9~4S_1jX5&Zo<{S34RI%I-
z=Fv3FoV90xfk+1kmC&Oom!9S*(IqZ8M=RUuhX9WE)P*!*gyICfJZw&eQ*e)!I2k>k
z8BhkHZh9SJE(d%1fPs!{zT0mew=$f*m8xpz%uDE?dh?`}IF@%&RhoUxo%rfz+Pdrq
z?0Je+?H}0DKMVc8D*oRwC26oF^4JN={}J$B!M_cwc}zw)y#D!A{2k5oH((z?7^XN;
zOkIob`wOIIZs9S$o{wWl9X8&FHaNtJWPfy&<s^bI((qFItq@bP)8?8YTOrCYcl=wp
zQ_+@Oxiyl);-`|L)C9QrUrv31FM3Flra^rvqkYSnxd*5t^p&YGiI%XJsBkOdq&}k3
z;N2onjjk%*lz@uUVNVBC1-SG|p4*7yP&>Ue$jvr2{PF<i_9zZ`bdB7zO*M^|i=V?Z
zjDzB_;Z1J{4TTa(_>L(FHiNQx%1jhmi>%A58~D+!((q~P>Cs@eHB()qH~W-RXrBYU
z@X_NE<Hv@p<r_}(H5PF>qOYd`ze2BXg$yqs7KZm6&lq1Xl_5-Xj#-xC=#dFg@9kuy
z^Nxc3k$dIiXH#x(dpoF>V=+tyo_FE<wvkG5=Tm8gM#u}92;nQ@;o!fb0-A^h*Vi24
zLA>X|o&b`eo7PeFF0xYz5~JgRgjbkX&S!#+##Mut@S3$@17heNw@E4!CO%-c!f7hB
zCg4iTK$C9vA+V;=eB>EFjW6v5^>17uaa(xwX`03L#qV<`SIMlLYcZ{bttlXn-~zXM
z2@46&MCvtN;g`k1&~7PA8D5ofJ<|b@qz&H|hUt>XX|tP%t#yNka*gw`Dm(s()~BD5
ziBZqK3I{N%-W;qrj{DIG!BU*cATauF>L2)171~DwMF`;0e7We3I>L{S4X}8BsuVd>
zsWp;i6dI!a7>SHJ%}!gz$HltP(dF6^fE?)5_g)&jRQ3U2#XR3Yf{l3CO1A$1M|WLa
z#+B&Vz$yI}b?b3oqB5rcrJaQ6YIfCxfp?NX^G*A$!&RgQjX;BKA)zJ1pUBkz8?b-E
zRbD73R2v$9<$LFkZxM#W<hX_*H)A@M6*8SQ(<y0u7MYIH+@D#dB{!z|=o_JtX!ltt
zOsYc9s<%Du6+cFhv_IbcG%_<?*s1nfI^hU*mwTA<9be`sOi0SR?I8<_h8|c1oXfgl
zgV2j(LM760dOK)jqPk{Go?sNoKC6yPdc*c6F3H<o_-o&Sh~C=<1-%|3FXVExSKeE8
z$Z^xqrLT-R+%x$C4YqW8R`pRGd!cfXLF_C%XLQ6^nlNzdWbvUx2yt1n_J#@J2l<?l
zm3~Xu^ISv4+(MyfIQ#x^ce5Op>w@b!>Xi|vIyYl=>8r@Sm4`|_MKklGp)b0is`@1r
z%tZx#ZVB&vqO`wMSQ4mn7e3O-t+RZiVC<SjHbX-5JuPET(}Pt3mt1Lx<+5dtP0=m_
z>`c{~Cb0LdNPCG7a@MXIy5PSoy_n+_EvK?h>(w1b`-G9<@>w=>cmyt;@C7D?h$PYA
zzN&7ucOT?-?BB(v=anIRK|bB)DzevxI-L8}GG3?Zs0Z72W93dl4=j9W7&h~xXmIAm
zMMbB5olJULwopjg4Y^&N@u;^<g8;qC87WU36Xy%X@yLO+TnEaor!@-+yPiDl9O>m$
zN;?tsRMLbuFVnnb8*t`p8^*uy(pO1G8t8b4PTR#Ba&w4!H;CR$p{ebX6TV@p{c+va
z;jf$4MhS>0C$k)qtMt3)^@`^<UaSdMOC-8B!!9{TDfH+vk#6bp>SxV)!zbRB=d}QL
zNW0vb&^WsIP_TUbuT8`M3otyjMhkO${}&bDyUbncDjIlRa9WeC=ioZ2Lh}Yp>jiZ=
zw-c+P_EB4tr1i{n+;O~o(RjcYBP;6<QYnl?_xRB#lEdC6%kvA^Zu8(@dtkg}Ptc4V
z#9NK%L$F+GA5~c5*IWu#`li`cO8pl3siZ~+`nWg3;aBJ$pHbWEu$5FzQr$3@<Pi2<
zc&V+mhMMH81aHM-)8P?iK0PKUOirSS(ZQE-4NVj=qUV-#R-}zsELG3+z+f!$=@+#o
zYm%P`l-C??CZF^-2~JOSX-Nr-Uw+<*sE64#Tr0obZlS+)d`t3RFnIMX-+Z0&(i4Jf
zWua~E6gq)Y?a%6^JdAA3h798a$e|Gv5-S_94zi%J;D@uei!r=fm}{29+gs@gwIz0;
ztUf;*w1v)6dJaM#I)xz|(?%b4P;ur-wM3C9RSy`uUX+*W(kWQPG`)(`g}1O>yzkR-
z2IjzZ?EMWXnxt4Hwm<U5nszS^c)#d%IafL^6kn0}a_W#F!hYG{pTVT=u+EC$Uo&^u
zFN*qn1`hRIgM153o1&X+LSSfn*Qq3KWU3xwhRe6K389C3?`$ytuY2|%LFezjDDs=u
zHyS#BW_w-IC@MhFTK>^JdnnnZYEz;?K<!YQyXfh8FRTJgw(t@Z8)<*Rq5*$*m7B<=
z-8DMRBXk2YqMwYua^i~9(=IeWwtx-J2IC7BvghZYro9P!-rQvlE)<n}YXi$H7^y<v
zt6rY=9F;$Rp=KWM6kFApi)#Bqme^`UnVp6{c8N`Y!?`A5y!K;k!y#0mJAGBF%AcX8
zHhm<eDlfOgXDOTk#p;pLJ?yde;~{`iwL^te<OZr;D%rCwQ5kD1t<sbvu<23m5Dp6m
zyQR5RaneU@StpsPifGhkFfT2K)+6~B=!DNQUi>Khth8lxE|M|A=X(J9BOWt~v-0*k
zlNT=XhEI6f`BT5F7mg|=7v-6;$!|8!d%<syi;Vonr|?aFt}_%DLDZU?tkmVnFs>Ns
zwq-RSw>iZuB~a+X^5hkJ@25|UzV+j~m9d=XV1O6~$q)9_jddGKTpKNWS+k3j*2}NY
z>rTd$M}~F*inf?$k$pERDf^tiOmwIgp!|16I-C}`-izKN=PkA}T>bIftOV>kn@RO{
zkuP;7{77S4QZ9H*n)|9(t1D}3;&^9ZH+K@OMD!h*)*of@7bc&TaJXzdQz=v2T8Bw2
zXs?(VM&7|DQ?JVS#um9YdBmr4Zgd{Jz(AtG(_Fub$kkeJDqbv>GQ(p0B+z=dBw(!N
z$-iPk|M&geBhVyi^wDVi1CI2uN5lKV91$bdRP#{?d%N7_ZNN6kEjFz$g(16B&ossr
zG+$KXL_&y|vm8@FS4w|Fej?2`#?*9p`%T^p*D01OU&y64cPFw*f23KB5D9;}P`hV=
zpY<ls9f!*A)HntbEG2M{#OkqTjl^kXn8zY$u$Ab0pA4o_Uk;l`eq7<)0@(kCUJIx@
zQ|y>o%u_*A50#U01SW+tVDM5#w9^rEAh1bF7`32ZLj>ajnV)F$gTXF=Qi?XGqV|%+
zcNuiQfdy!oGLrfqnC$CR?W!8O?VYj7oyZQViBkZ%T{M{Bw{}S4iV!i!rz$j(*m7w>
zi_rpGE5<oWnUgH$PMUDTETcrw%Avg5U(&4#=k>Q503dv*tUD9JEdcOPG=*BBaZG;<
z*I5~38^%QaHx(G^#vqt+l3(zbuTfw6fx@2UuK&XsMT}44Lrsf5vgahXM%GjF&~RvF
zqFuDFzDwc`{+hcq`B8m`Bm4Ll$Pr2sJn?noCQ?Jg{7rPklO=VaZjz=fx^*c!iQR4X
zXr`@7yi>W|E!C}?_mVN^CxXG0CT4_sPsgR`GM|!LR*y&<ubVsoLGIvPoo;gj&pVQ=
zQe*FmoiflkVeR4l^$z?0-ur+34Fxg|KJaU#{SgO8*eJ<;0Q4t*p+#f!`725NTM21F
zus7;s&Vt0<8BzEIp2ze_2OU2@S=UY=<w2N_M*be2FTfH$&A8g&T`PlDp#zWw*FPOk
zc<cm#EiXC_h=Z9_)n~cW4+$)f@)h`!4Y>Bcsik`b(L)UIUGH+--Y+J<&Df!a>FvS-
z<5?x1-UsD-gL;Onw&%s8nId`q-u<SN+lw(_p#$D%KOUH;(#$o5AV5;!3o5WT;dr4R
zgN3D6x?vdp$tVJNO2qZGHZ)7=d5b4=LZ8hE7GpMCp$^tB>ioag0?2z^i|NwQHY}z`
z1<Z-6`Nm=@=0Pi5u}0l62f#r6``5A4^@cH|9FX#pUk&h5)`T4Zg)WxV&SQp|z)KtL
zDp@~Wly2!rCQgC*al*yb<2pkj^7y1vyvRm_>Idl`5asbK_QQb3?wc<+!$+ySuS6y!
z)f>$HNn>q^@bw;$hYiYlUW$vSxn_GMYnwmFu}GHS*Ddj1YYFg)goZuS%6E}J)iq1f
z=m2H-a@^w#sk4Ea@SZk?ESLD3_g$bXlyaq$^Y>tspwhbw?LgN%f2N%Q^Yjh?(U(#v
zyw3ttui|#CzhRb8*E5Zif{A!U(-W@n4v*aK3chc^k7%qQ|1@qq^0F#)J#IO2S~DM;
zq@8_wx4;tj0L#5Lv1K?Yuw(aTFI3EfqIT9vVNtZi&v^KLvvR_imvG{t6LvCZ*=CEw
zv|Z!&5_9Fil5h6h6f8oDFsLE}E;@sHZCQA^RpWXQC^Z;Cube3OR%-O%MEgtC>=nzF
zHun_Ukm$uP5d&+x+mCqDiwTq*!s)}I=Vg}B`hYAtyz<VnLo-~*#KugKNMjPSA0{|G
zH{{A?pmYlVT%`QgRw3)xI``dzy-ehoq4c4{4pap{<G4XsxG%Ohm$X*5VpmkvD2!5C
z;ns9{PH)VxTBLO3&+$F|H0&%bE~V>e5*jX@SEKm_O<*`!W^LY)ZG+*#qr+s^s;p|^
z!cTk#N*)nc7YKx3<;LgWf+~#cp-&E>WB2K&1h@!S)O7osk(flY$Pq4LCpUHFfszyA
z7^B5Z;YW2B<l9$dM^af}HE|e*cCi`nS~?}dJIC5pcsFqv(tg$ox@^o9`F=TH)mXe2
zQjIi=9*``u>jY`GPNcshSGs0))vFi<SCoVSz}%KE{VED~exUi=$K=JA&dmQpMffI2
zWNeQtyMvaKL6z8(Fwpz*iMcR$H`+Ft<J)Ar(AfM|j0;jqw<`73yqMwWe|`G@{XKgr
z0mkxHFudW92s5&wF9$22i~{Z^r&=YvO1)+nLh=<fU|HSyA{~fWBA)BUOaBn`*z51K
zULlMmO`d5nrM=JzDw~CG2Z80^=TiIx1XgNh#rDLDAS{LF<SbtC3KUg&AE{N^t-%=&
z&6$Bk19Y`MTUx1_aN(nPeu0roN5&@hzn^<6tDAhzA5F>uv(m`%zxJ&9nBdsYQfmWl
zj2V*&M1cN6F#oPUSO;XYd+13PIU~P_fk8KO!z0iA*gB}NjVNBu>Qs&J4|0#LVEP00
z4C1q_?eIwLnt@Eku$nuY9>Jep0Jmi#@1*)1O(++q4Y1`(=9WV4_N(PIFyriUc2L0%
zkTbnlURuwB1%mbvR(AhXM{v(lRRSyH)$kN7&wsmNuiJ|YjJNFqXn=rDP;<k{cNr3;
z!AC4#1fX9z^RlYDMn1|!z9IwYt`QY2i5x$qF<GQGFU!3qit5@i$Ksb?&Z9Vvv*wZK
z^`N!2jLf@yzcs`&<mi%ZOe#|PfH|^R?ZYxua}I)N#(u5)Y%K1)zfz~Z-vxv|DX$!&
zEBb-Vik;<`kt}Q7PPQ7DkdJv7F+c|Aqi6D#2LdUK_`8l{*i#PBF8XS7^!AUxigHBr
z_&L;h%9x)D*#Yy6@CBIJ(qr>zqmX@UWacJJKs}rfrU`!tfs26sMlV*}rC**b3?1ZC
zF~~OO=!WO^Si3-&EFa*_%E0z$jB~N#e&Dq-H^1+9NLFUQa;>UL7*>0JE@?TPexE1A
z*JDGOjyH8B93A>ra(q#oJ04p#I$nZ2EY|~BY7fNQsI<~IZN{unO$7cj+9b7}IgLIs
z#j<SxJ$1b=0Ecvi3<r;J;;8+GAApx0EAf#_-;#EsW>}iMKDW(tq`Q)BqQ?zFLFqcC
z8W*}d_c$}9hVG$<l-jIGjL24vS-0*832e0;+jHakVi)1Yu&^&51UY}R5i6^Z#e9US
z3J<He&5#=nd+dgxt4CgeWS*=p(^Cw7282V`_Q!*t!ye5%K%HZbYR_d^&+%_T3<Jjr
z!<po4u1&Wpp_+sS(lY3P1d-f0=awIs=duJ6#RvDbjBXjiH)>&?*<(Mpm?osNppsG>
zfv6CZhVq!a!(#zOqQNF$b$)kTC#VsW2Eq_&N9;|B6%mC^GEH4uo|yRL>uAJwX&Uqi
z?eqf^lC)hU(7sJn^>rj$ItU8UtH6beO(@JnV#cw@t*6#vtO3!l*>I_^xe%NN<p+Qv
z;fS%7^9>nD>~IQ(^3kKX-;85x%{kW-JOhkyQ6c+}X5ux6Qdlw*FdT|EGec#aJTHbR
z$5pzpt*M5IMuWX7tnHW(IWGE5wdz}OVo!c%^E<7`?v6&XHau?BU0etA#Rdf0P-q&G
zdV+$JYR+y~un^G3dw69Hz35L5Q50>IibZw~d|TRiFYJMg!LMwPGRMa}8-FoY>?7c4
zFuiJJ)RL)h{k~BALF&=2D5iqKz8+uAL6@m&md!dJioaeg0|&3r4!AT&-mJs)x*`^8
zY;X#MZP#u1R|zbZu|1(jQ<`P}zPkOtwG&4)I@(1{(LeKKlUuh4Gr{1bCcIzoUCPl%
z+Jl1G=Uhf_NB&0i%h+uj6Cf^ifa(+|!i|$mz^+x)YlQnbFpuHhNn34Aj+7H7eKs25
zcE>~K0et3kRdx>6)D<E{%eyDBLR!$zM)C4m0OGgq&ReL`?FQW|Le{#UZ;%WObZ=5B
z^DNhKBUnJr!WT;W3^}KEXoeA^P@ssz$>^x40>}Mdi<Mj)4~9IoUO`()+&DQp%D&|$
z8^N@+z!bAUsnoml3i)j93%bx;2f+Ecvz6w#me!1u(12q5Ge_hjE{QQ?>d#4s;=8Qv
zL6X<sVJ8&#l*TaV@$V?K>doL7Mph8sxkvhJri(lQ59P|dA_j5m%33wYr%_I{ow5#{
zkoM{VI1Y18q8?7~)1!vW*td!_fqT~u4K%7yo-_)?zVwREV(y5GNhrwn>U*GrMk|&h
zC(_GG;h9HK#W46r)coOat6V^!6!Xz`FRy{p&wKQ)w+S8J+$|mYjf9@yc2_`ebUPZs
z4dn~%=*<QlU;DTkpXk(}<#qY<A&C0z$ZV}oL%9#Bu%blHLj;VLifVsAzTZw5l4vyP
z2)ZX0dbhdy@PIz%5g22Lm0XloOTl*Z2p*Itdfeb?CR{<UMF(Bznq?nmK+JAOZRbi%
z8(BT>tY6P2XI@Gz)1ZHTHfx%t<9~G=C>PVdlAeDkTv(Pgf$T1)J|$NC?HFF&qbFLR
z9m8CqU-G0b^)WFsnHBruw;7#_gB$-&!IzM@GS6bHDBE%FwaQ+<b~+WR=fzVb#w@=8
z6KCp)&pmZ*u7n6JtZxGAnQdBT$|HA%y0v%%I8zgs7czdWD=Og~z=Z+dFX73S0pFBq
zf)UKlcBb6869rM!VxSd@lt3D4vS3#(cpJ4fd=pFXL`C34p02^)=GEJ3acx@cbSJ`i
za&`~a7~na*MSeG&!|WjK;`F2J@=%)xTcJ^W%EVxUjM(REUG-U`I6B{2w~6!>y-m%8
z$~WOTE%|BBHYR%N`-uom0`pK$^dE4|YK|h>I9*-5s7j49Y-{!%`|4E6);FD}sEAJ+
ztgY-jA(6~Wh@~S$ulISdH_Ku5$4M#+v>l3Km0A0<k}i9caePAlD}zp7jxu&$QuNJ>
z4ZtlTSIkE2r$0>KqfgkaD5F0*KmB~PT;FO^aZw;f1*mH6LE}c7K7o1kD*hVhBCY!M
zD)bXycA3YUj_lJ$lB{I@%wzu(d`gNA$c5zSg4!vM|G|l1-@&T`B&fi=m!BGqC?UGJ
z`3>l!vZ}JLWzjY}Q0+~Va`%2Fg^tMT2rEjnE{XF&0;KZqlJN9Xc^#Hd!>}F!T~UoL
zI_~>8MP{=1hU5oCYWE=NQdN%oLlBnRxd{S_@`{u=v{0Ib{dA%gTFP=2KuK%qg%AfL
zW#?w$rc_XV(+V5$y>iPptY~|$v)%R`D0*+*jYI!YE*afRg2q8d8NFGkWlEJ25YL;B
zka4KQ2)xyfQIA@PuZqIh274L|)J}|A>X~;uQM3tES@i*W&l`UL*iPsQ&D!3R(T~e-
zt}iBnN=#eb4DrqJp2aF<5a&WnA3$4N-tXEs?MtuQZp!FKzq2<oiDL<rn!zLy=r!Y7
z888*jg?~J!G{!O}a%9#A?Od!J^j_Zc2cPe4((03zg()9tH=4Yuw;rz6Q&-CQu+c09
zb=fT0xNxiC>W(vwy^QDsGw!xv=EDe$Znp-}pd!Gk>Ro57c(i<Lk?>Z;S%C0MRhp|<
zu%nfh{qi2CJ4dm4%orsEEw1{~=T3snEesL{)^iBsdQ#ABHDX{e@io<$#iB%fNID&m
z($zado)nd)ZAQdCN;52OEN+!8*27wtatSV!ay{e;TaFdtn|0`PA+AQ~v0XsAKfl1G
z6rRl6oW}7z!j|B-W{AH}K&Q8f`sLLc;SIA!C4wOZY-bSj&*ljA*t0u054CE^XIV<D
z=6utSMOZ#Ksz}pY=qq9(qLJ)-`m@@`ak}Gt!#I~X`TAB)N)`?0b9e_L%EFJ2J)5+3
zHszFM_&(+`MOtR{8_#t2T_j7k+_Et*y<cPiwo0<o@3GVpc`x!eEBrA1omB+7zej+n
zj{0QIXDNz7`T11l=^WQgF`L)7S~vSha;4XJY{mIy#y+Xx(Xlw08_fmIA;;5OTP8KQ
zMEzZ&7KMeRhG8`qTmLR2A+^8&to4Rmtax2sd2(EHPuYIdc^7d|PWF?0;e-|JJLGJb
z3P(v@E*o=}naMZ0(hWPs8e5{j0pyIEI?<q1n?<rwHQEbP+O;+&76(VAP0O%N!8oE=
z?^udhZ%b5WOVk5H19E=K`8}rR2h*vd(V~HmwnJ{m?QNh!iO>p~P}okC?oTVuB%l<y
za6``1thZd=UzPVnH~;diE2WdrI1!?|KOM6ok&$ry6em{&4<3OwW}{!?Hx$hn{Vjv9
zDvl$Ab+djloFmRX`si$AH|}-A*Pz4i=$A}S%!UljY%IQo<_5=Nd6%Y~4e+mt%os^q
z@qVjxJvmBq<6_+yby;j0_~0VWX+BF2mp*Si^4@~7E3F9};G2LJ&Ee6k5><+yq1m7!
z)whPYssJ~xp?dlu6>q#mg5Efuj~fPU4PbmI*_VCi)KrmtqEcMdM90pHATevgLjjSU
z!|{P5#9(QfSEX+sX-{dA=W@tC^oH(Hqnq2y61gqqR(fk=gN%`jyahs@xDD4t1=f1H
zUtm|+ENh)WT;e>f@;>Ng&=;g#vaexJ);8%&%-&fYiyLrWBRG8P8ouz@v%~*9GfE1q
z1S@ZQ&AT0J2bO;|{P;x*lhP$ydYUK(xm*p8AB<M<@3tpLv*=>?7ak@IJ^Jj$_P4w*
zG9>RyCkF2IAK-y^DR=R*Vu^a7GEJa*(}QfU(sPT<0S?A0qqm&3(917j`4?OLKjXZ@
z?JfaQ=ItT|i7onj!uAmkI9_w7w#B~_u)#VV;F$bd;Y+!zw@0;rr+6fQ19f_6!&nfW
zEyFF5w1yVOT?eX4Z^IqH{<D=rSo_iYGX33GgTw-qicJCfpMeiS`JX^GyPr7T_^U#}
z7;_^!yIjI-xAmr_zyT>YWO^uKM<F4<1bTBUNJ4;dQDHt=Evtd%-yU*-3w*l;XY`Gu
z|Bywp*AP;_nMd-Tk=)4W?UW+;?{drk)||!*C?$y9o6u+~()(LJ#=lU!{%Q#R{zgy|
zw6`tzi%9-%H~g=M2SU&%;QQ!*{nYm#g5$p*^`9s4jtj{Y3Hk8qKiuvAv@!qtsr>)&
z0+4p!MY8IR0a*fDf%!Aki~tbnKoi_3hwdH6FhiqRlzn2?57LF;hTs0zO$%^LeI3;5
zSa;~U`B%7Q(d+??F$h3@e8sKT=Zk<ZTcmEqu=UYpi5K@A3`&RdFar?Ic09MU60N0-
zV;@McgOMzOMKB;(yhhGsY#k7ogcUh^>HJts)H}J}ZoByV7%o3?)^5JuM^eF&h!Ii^
z&<T>p5%@aQjPfa%=*;H>P;BFLtNhKLJXI=_L!_x2Dab4Y)hyEFr<<d9TP|3UtZCzl
zUV7)v)&YUWW)<ROj;=jG{SLA!$l_IGZ%3Q6gKgda+qkey0N<GR)|SH`*Jgt*0QrPQ
z0l4SID!?7c0S;F>ZPC1$fd&`_Fl5>j3INvNx4hFI)o5o5m%*vJ^B<2r??fV{>jt1h
z*a@1Xs)ZCKlvV#KY_LCmw!MHF@S+P1)&=-mt2b)^OezOp?#VOntBD>;a8z3lz@O^`
zr40prw#QOAAiac8xX*GhgEZ+_2r<GieF)0J)2S2i7wiD&Iq_IyYrW*u1F+;G7|Gz~
zY1wg?;U36zy(eFND$(n+r6-`-xktwR_!2OonT2uT9B?O9Pr!cZ01K|hQFSCY5bfxu
zg0%??u^4H0BEz&?WJTLu&IUg9z@h#2eUF=Mu$!1@!EX$mKMforX@QwQ>zr5W4scv-
zrjjbPQQsRd+&$R1Gb9+4_Kz-)%lutEoq~y9U;s&yW~wDDJe@M*Dt<X_ovryu%>!_?
zJHhorAeXC<oa>_kV?cPHniD#&pM!e<!3J`+2BCBe<@OTogeYJR24V9~woPOJ5zcB&
z?dVlpX6}>z2}A$6vSkot*np<3oeJn4s0h5Z9-qU(6!kLAzm4z!Q6xEs3XaF~yClYF
zmcA`;*(6$k9xp8d0s0E=Pv?ttfGI~qInX?Ke#v~|N_|aD^CD@oDAQR{XwBaLLRSm0
z{6r!qLHA|_w_mq8#0~!RRA?nuQg2Kma}vO#9pB$e>e@=@U4^5B?#V!Dl7{dfbJ61o
z?|@Ql52$PIm|~lPd6?VapLw^bbfPx8Jj9p8wZxB>FOd$^8w;)e{9s^?cj9Y7_tl#+
z`V}ged9>K|>Ed#X7!G3IyX;$~`8v(?B<R`bCs4*6g0ayzK|nM$-n|Yw%tX<GTOT`~
zOu7T!8s?~qShGw(|BNG;mt#lv)ww5@X0f<(=!p%}L?;4+M=CGlO4VGCxcXCVjo8Vg
z$-G7HnR&LS`TdVZDMY4~7rU)uXB+Sx;8gV!pTe4)%RUws1Ni6;Ag2fgCnI`jyKptR
zBkT{JL6LLj!}P~4p1~a5!UBMRuQ%I26Kabr1Nb-I1$F@L(1TPgXy65~C&Bake{>8K
zs)yhq!@6(aVSGBb>vtbu06%ZSnEPhe-Lc~BYvmg!#2fKtH#2UNXi*RF&e$v(-khgI
zMqc^>&c$&DvVz8ouFOlgiW}2Udeo|oG}zb!&%zzp5rO_b5ZPivcg!e|6!`{L1Nw<~
z|06!eK^15HHsyrm_ay^t7Ss*cxZmPxV%_klUKn_;%CE1yXXz#w3=dt`z#J}(^#h%p
z>?ffygxyGH3#F1R(B&Ja=(`oV3~rB$bC=i7d-XqJSy8+m-fdQ0%l6*c14^5Q#2@7A
z<eEdO=8CgG$HH2!#POBL7}-`t+BRTyV&njT&cI3?dMp{D`>HTcxPyWK_keta!wHw*
zCn~P+_fMNo5TI{~?4|^RE;{Q&uZ{PMK-|6uw{}yNq4KT5g33P0oe05#5aL%Wi(R?g
zPygA_MP|M7B9K3_&1=#)G#H$QsC!OfetMBru6Ph9tmm`aIx*zlHDC(cJ;`^i6zhxn
z)Z~U(t5INU9=ok>`UF!I3L^SmGLdO(dmuVGtV&fYw2t@7Cm2vsq?}&9i`iJ_$g2|i
zmL_ybLQSS0XBjc6!mUDA1KW5vYY}@+>3gJ_^5V<{4H6LVbiNeHh65_8KzlYp(ZP%w
z>TW0Ijt2)w|FG>h593~eF4Ks_Ca=Cq5@Oo`9fdRg2B2AsaJI6r#9|(diA$2#U)lWj
zBVPxlvt(_><jxl{YcJGJnUh-}ABNdr1+{mFLPRw2o*iDUGq$yiBQ7Yqe+$BWPr;b<
z-+S>_O0i*r<_Cr-Lob@~l*kHaE0B6KgV0`-BslvmYF6M9krYEUgIg6QIt++N1X7pa
z>pao7W7v|-nFPS6LTKvz*bhXqbWMrl^pCE$?~_s}2<e5sf~P>B!a$3A$P%WQ33Q4}
zfDAr4lB7%a&C*G)iYUpodl)BD`aX-i26tiZPNSV4`}@b~mxM|uho8UdSjH=kKGz|e
zw|x1sYOHw}V)J9$8lqek<}eI-A*CRsgU~-mo#?M;=G63ZR@h~cu;ykg&oikIa{naO
z7yZe`+v@^cn#*;|H-_As{TBnUSgA;Aor!ydpI}nI4zD=_xO5WxEr+q-C%t|ifEZ6}
z66wiBpX9b)fA`h?&p_Ph<kITB3YG{9?s_Jj+?EbsBgGsod22)K5up+31lDD61t`-Z
z2m8Rfuyesl1)lt_kcM=Mnebh(Q~O^oQ|(7EmF^vneAx)0*AD972W?8|nxny{y8knG
zA!e37Gwyaa9kPQ|fEVU`kiw3lcaiQ2`5z5~1zNmwgM=~2pG$s#GKN11W4Z0(H`pk6
zj5#05yd@*PIEGIoQ;&rUjwt=IU1amMYu>#VP+@`+hs{B&uS0wt`y46q@6oFr8`vzm
zY?i`wjk5zNsMTpHIeBa_GsLQjI}KkP>wQ(!o(1DueJwJMmr%RPI$j0*Sr-vFR)-<q
z#fM*YeYYq3$yxfes*cKsVdqi`oev+tH{TFhs*W%Hs{BUDLTr2f>GO_n<NwFrTgFwn
zt$)ALC5?20gruZ&$Rq?rkPhjRmhSEj>5>MK?v#*j5Tr#K6p&_8XUw(NTF?J^p1t?~
zoPA!Mc(XpN>1W1$k8zK2jqCdTzSl`+^*vB*Ja+YAtWwvuENmK}jg;CFR~Gp0hmZ&P
z9F-iv?@AQeM9i3OL7Vc#(X}c%%yi_rJOvq8>}F-fE?jndolj>DOfZLA8(yBtw6fxs
zu|?<y{!FP$t9Juc@<>89-uZ(6W|0$>5@+`Qg_if%?LFm%hncnzc8_H-<e)P6yaAAs
z<C!GxQJ&YDM&XZzACc|O)ojAFCk#VwZv|8~VL?Gx-hqiL2F#%K&AN898&v-FFDQ07
zN_%`Ng8*m-86PPC5>UFxcvS_&E}KP-^jTSuwv%O9nx=)fAKkANO&%6t@o3T6Qxq63
zf#hvB*bXyP)2`p*Nj8DBCxWg;j!4+~y5aK(iAbsSIZ5>K>>-#BQ-KzpfjkIo{>!g_
zpE)V4Tre2Am6=4A6T(E(+i8vZwKi&doXwAyK)Bk~OT7(?p^));GoS8CZ-Y3_KoVm3
zLLd6x=kjZ%&m*vfwi`@FbdPY4|J=QiMY8VJtYtt^fpAlnGBW?+G6cCxgvxfcg=Q=j
z!UWoxAT&GMmRqlW%|Kjh;?9D(Y(Q2aV;{|n2?o2;oIgQ%l;xDp6cK}*+kz&(u~)eR
zwRbH{gM)8Gx+iB>!o@JvKQ6dmZU}6B99dH-s_n4af3Zznm@1fiBALLIsM9GurRLHa
zr!o_iDGpVe+MB{=DX)@@W%edX(G@C<VuRPR<(9J=N+0|-&La0hKk{h1H0tDh*OJVV
zQ?B?H7-`PytPU=g_hQxi9|^<chz7K#2GPbMjH6GO9Z|5zcf2KA#@nV!ug+j$GKjv#
zgn_YOrLv%`$X?8#y%{)T5q5pkG$2;kR*xOI^<=5!p?i6fPu4)7I`+D*=whuy{>fZ{
zF3445-exWxCV01;E$j)~&0-xO)Gg8mB1ficHhq-yAIrWrUlBB$gbNIB$sTe6O2Xk!
zp}Y>wo5>gjKN+G~t<3CFUh2x!@+JPD%)UAJ(Wv^d(Au|d_!zj)9qzcto*5hB1DBB8
zeY6y*`J=xc@qZf0`*%P<RJXc^T`&FTp*B;T_J*dDnzs$OXxN)*C}@nR+i55tQi--n
zXfAr}e+q<LIFeWcHGkKXF0@*!Zd!p`v1zv`gH(zEk)hxbA}%HB$<e|T)C~J#TzF;`
zh<o=sYIk^H8#`L&^~cJNcLIoTd$3I`j5^GR@bSPtgdx*q!BVfJMl}QTnC8)I6k2!{
zKstR$ILWNpGX6(IBCos?Kq_1TS{0_e+9OT8GO$EJm_qHJLPl%^mrGZx(Hxs>#$^pG
zx@$lcjunWE`NZ{=S*7~ExLJ*Az#5qSF2BuSXPuZZ%`mgD7Xt*4ZYoLq<1H|Vhg@Xu
zsGO-8yMo3CR8V;!T;pKd-l#!#Rt7EUG^B=B68IO|v4YU)`FbY{ZU>%>WWYT9piH=e
z^wNa?9D*akD(UtD-3F#)(z`SklVU2M>t)({cmm%X1x>ZGZ#aD~N|h*h{6>cYe_l}g
zk$)3=fq^{SL5)R&s3_eQHMUOKQpg#t-F?}j=3<<EygF_Lj@vN0Z@G}{lx(UjpO(W}
z!=mkpb4arXs3LFk#fzHi8L&wfm|_}7S2H=XKehuG7HnmxLSwZJ;%9FgNR*LHjp}fR
zZ8a9x!7~}<U3K3&u8O8DwuW!C)Js}|=2VMOrn6SFuS|Y}khI6Nt!;m&_vXbOFvRXj
zA3v{p-eNooQsRk$g{5a_{30BQ>j@g?)pf;#GE_au$4jvvZARMGNSi~o_S8DJIV-O(
z4ynNS6{n*PJ$sH=<h_)CIpS>zA~I)X(buN^Vhw`h8NA;?z4GL4!&OGxVuZ3Mv*vVg
zZHT3E#Ah=Hx8y3i`w(6pA)^($zxy5>m&?QN%6pdV(nlkghMw5|>~$u{O0VuA?Xmj^
zX3xUrex7$^L&~pM+uAzS&O5U3RYYLn^-y-K*Za3=neUzeDRAfsUpXD|4^gc+H+=Y!
z%!Ux{q}Ge)vE3nV0SZA6pA#_4yAxxHoKCcqfrwKeB{l3gRM*Jz{D#JR`Q<rWd?1s|
zysU^UCD^4y4qx65gvgq?BkdXh`&7v_`qeI2L|g`um=OgqwSp|X!X5})Z+D4eUxkwg
z7f2&A&|*Bs&7694cbwAcR6V2;gKQ#ATS${kGf54OH8dZVfw8MgWxNpE#!acN7$$xt
z%))X&qA6w)RDF(Aia8*5ir=>Os<2K4MUsvSH$E^<Gi=CsH%PBx9SYI}UPMfBS7A|o
zm$F3AO_&$F33woTz{<UgTNB5&>s!DNy#YI_O9d#mCp*5N%4sVW3x@{dyr`~=S_vIU
zJ#}o9XRJVwz+DTn5=RsQhh%tV=4PKh0<Z8Vp+VhE0YH?Wz~o!!?~Wu7#lc0T3F~fZ
z15327ps{Y(BZq)1>AHgKf&m}}r`%VEQ1>CK&?=8c1d-*@a1}KXZaFlf?=OMEn6MG!
zk0*uYNBteAkB)zUqvy62l4Ed(50EA3eoh0cu3k5DdL>|<`Xq0C%TJlwj&h%U5>!DK
z^Q;?*XDme{H6PaNY#buRlV^?xdGCz;G{vGA)oKJV>^|LDnCe#ZIbxQ7T?#b#{O$Wn
z4=<-aj!Af%Owz2irK?ZjzHtR^e7q+D?lg`iu1%V}l)sdPoq8BkhE#^B7`<K4m{}hG
zuWibZk^U<{NMK1JTQg*yzICdLbS7d516(&kT%u|s=IX`5ukI`5BAgPorh41_(r+Xz
zOuWUmQjKp!6B@A>HL1dK-9~xmHV(RP3`HPi(vq8v4+EI29`}PLi!q9YJxecizI`j*
zm(4i_9K46HgJ~r%FrXCOe?U}z5&lw(Vb8vWeI;cV`+T24-shrvi_X2a;+!Bj?Y!i_
z$>(yxd20R_ad<y!>V{;;jKyR?UZGkPp+Ip(Frg)L`VEeTuEUo<>;j)g_BpSZ+M5e6
z80I072Xj+Tce50|j`%cZDlE7pcs?-PI`dT1cYvAXgXNh^uBcLeosG@FCY=aJJ7*lA
zzI<U16TIXPGL4lU3}<39{F1`wYEwL_w~)qk-Xfo^3K#uAnaIXMp%naS#?VNvuEcyY
zk42#WryJ)WtM-!#i+9RAa%|d&g&}<0SPJ&1hd&K{>cT9XuTwYh*dmO0w4P|w12y{n
zUIz#CC{q8X+dchAqfTC?(G@DFz;t$=VU0&p$-pla>85E_i&izck<=n!f!mW=I4C}u
zXTTQ)w^cl7wL%nC%et?<na2MqDtzQaeJsOcNMxAS_0lvMzFVFgv!W~a{3MAM`e4y@
z{mJ7fk{kw|*-!X{kH^PtrryMU<qlyhQ|D>s(x#uVQ0B;o>_sxnU3pmIX*}02WGT#1
z002qycd|SJ&(aA<6|E!5&5b@*_5-R@NUCW0LHnS>WZpU?QG(9Ke3?6KXj{k_ewG1g
zmLl<PBKod`t3tsmx}0SmJ|9pXzFqgy9+-C8Uh10qn6j*i8YX<GJYgV~K6s?oPix#I
zRr-_potWb-1EDk!VMWgJ_px2T<NG=8@67h8Z@ViCv8EgIeBjoOSkBY)YK3T?it;nJ
z?D{ZVUtMx64CakXM^cl?L=mq!Is)F_B<(Y4e8G18D?MX8m&r97yTcjh0<w2NeDy_5
zUF*wXXZZy~u=MqFzf<<Y)s$b@Pks{0&o%<BB+CL-hEyacg<=Bhvd^!7RBZoeQO4id
zPudAj#4!8(zjf~u{iyWaIfbWeNx?8Xx5Z>My3UG5kk|%UMR5uBZ<=2;Wm=?dA<L{;
zCBC8biHsUFruLZ-CaY(Bt#Ha%@hHFhXe}X;6DiAuGv<+!8wo*>6Xtr~m6xApBs|%U
zRX{e2@uck7Mwe^~ty?1dDD+wQ%sG&u63)h-A_WveoLQGRZXtFUzF8eHQnHdoCc-9#
zGpF?ohe{yvVPWY(_rpY4;tgcV8ggj`1d3UWYPhIfBY9?oYTJu!9`E&K)CS8eH_sm%
z!!mfz$_zrHI=FXns--A!!+Gq|&|UCsN5X>AixdcMw5<OqdJRUmnR*$7phnhlGVlBf
zs;r8#9}kfn(Timr(ywDGWarb~8((_TG6@5Y&8nwTcLV)(?{h#1xy^p@{_4F3<1M`B
z7QU6ZRyd!I21_?pzczsckt`d$MZ=T(5$)ld!I;IRETN|AukLT0-X3IeLR4+k@s!7C
zkNFlyA46e&1beC=E&2EZ$1L89E)=S$B^fPw-0e80D>{U&7yK`ZE9jQ_wvYl=L9&OG
z?d(&dFX#mkZA!oCB+|yQNcD3OGx_uj2)fs;k|35*gtsZLOaGhoM*`rApa#Vya(_;O
zI8<Md_tK<!C~rl~_QhoF!W+1YxrT?h=cVZesdveJF4yKOb$MuLao(+i0y!0|MBD_z
z<vxi(<mRM4gtwGGQrPN{()<Q&)CzEEU1;1iTYe11+i8F5nHu&8s>wFtk(by#mR7%9
zPd5koZp!>;T@xy*f>!dLTIiN`=)TKMV616d4CjCQjY!N4S@#aIa`qO~bnAaKruD4Y
zaQsM%b26avoIfi+*=4#i(86_@iGRf{e^?i|)Ep3|Yr3h7jR6rPGTCT}E{G-!x9|Ym
zt79flfVY2F$Mdb4(Xe#>4A(MdKtsy=qc%<^F@7>KPh-T(3z@GDXJEIL-gqwWPsxT^
zEt6=vgiZd8Q@c#16Iq@v+Y@s)T5U@M>bAl?9(FVE?6C)Fyh8lPvLy9jG1`i!MX-!n
z$XS=lq$fe$vg>O5TI&-VtIbr{gT5^Fdvj$Wk#qc}W@d(ci0JDVmpl&IdcsGv542l;
zHi5*sTT<U?`mlg>H0kE8t~3LipiJ$rtd^KOM@@9y6yo@H67GTOPf0bm5?NVh;S;0Y
zqblj11*$$6|K0YY2OD17)e_Tx4DULpKTuaHul9kS6#(7&<ZEI<EZtkfH#A0sDmd^%
zpfk`9JoMeD?KE@BSf-$n=|r#h8g}n(2H2$*=|;`JPQ9g&WiJr<SxSuv%ROQOY!{!*
zOb6fw<Q6}<@kD&Sk<C@eAKYoHD?>|PPs@*^RjNhsF-V72y09>LqxGj1Yz)1Qo7Hf`
zW@@G8V8E7yC!}wbfLVyB4VVxjLk<P2{i~x9L%+3xis|g_B24Ne)LRQu?vJu#>gJ-n
zgF1xaE<vvMY(Q;5M^;P?;|IyRcV82v`i1t(Yu|m`aq?9q(E!++(&EN!P-W3OqFZK(
z$BQ4)%Ex8dtyB-{RNvAd92tnGyL}U?c>xec3!A1>)%;|&^MVQEpU{2DUGpFjCf4yo
z&-B8G1tz8N-G>fO&sW@5_k_o-@GT?aZyNv?tF3as!NC`TNEhv{OmoKKxSkt`^<?48
z^li-w&L~Ia1jC(dZW>RjtM0rF7)7&fV{aI9#LHS3i26reK3!M5zxyt(66zZ9xLc^R
zlo(s;+{(OTRxe5P-#@KR(V%=&@j>m&pAYM+tk2&->n~_sah-jbu9V#m8h2%hA2g<n
z@vUGoK2qv@0TrUPrJ<z`c4}<{hh54^G?EX+7M>8BQWY63IP6liY7Vls*w72H%P#C6
zXpEV!X3ATBg(XQ9v%L0reYxM~tO|zsUBiSV#u-yu0%<kibw4rAE3pHIA|O+>wKJ|w
z%9d>a67556f?AjWar^16TkXB<J=*hL!`_v=gs2{~Qt|ko9tmWTdrA%X^Z0~n7^M!)
z*+4GD0Uo+@nfDG56R`qW+N#N9SS2dKCsj7Q>hUKi&~qNnO^|g{{em%b)<Nc&sF@P9
z6OX3TtCc=;vKy4FpuLH&ND)<%pOY%2;4~12ozfkVrhmE%-8O%9&vy0Mkjp!FpkjL+
z#?;?lQo_lXst?1hO|r)~zYOX$lO(Be*IH{kl2Yv0>2>BQUFoES(w(%CL|RjPI+rQd
z6a}Y4O13zr&o_ohKR*1GKKke5BfiQA3>}9T@!Ee1qOkCYGlIl|0?e@6!<!rjXhJW0
zep?T{$}`LYNt9p}6g=wVRj}3^NWMz@(0zp?({Xb=%+)&g>v<9jtso3W=e($M^{Oxi
z)|rb=mj;)IMQTfeYXIF>_XH*eo5YcmUnq~8#TLN9tx%AY`dF4A&T=~$y<b9mljYDy
zLiu|i@T%CPqy!I`i!-NyNfFgotxsw)T<4L0Dh6PI8M3}uy8($M$evUii%x1|0YP!+
z5W&J{;jwGiv#%OLJ-Bqzy{-yOPo7Bi8UN&ugt~yb$8~*zOYI8@KEN6KA`I1gzb>$~
zm0@@8vI6J=&1z0HVr5Y*vx_FtbVfRFuC7<_59TIp*0BB4oVp8)UKTep-u}duAz!o1
z&5+A=x2dGPdFq=+HDazp|N4(K!N)^j{b-x!S)Sdy8TeQZMo^<44u$;toYcY(uHYRX
zsUx}{wr?a#OtJVP-(G+=c|-kkpe~i95nM@vx(%81sYOvUoDbh0XrQ~k0*6v3LRgLQ
z{A2ex7*0k_So#ZQW;)}|z9w;O%~B~8TM)>J`I9L2&r>Z9X|Vcm^M&)Te>`vA$>NNF
zr?U0?hyu6hEloQE^U6?!(EJH&buS*}ZxjK4p7?py3@JIuY{~9#;HHpu$Q)C9=yUr&
zU=Akma~$Q>I6uYA!G(mWH-Pc(=bhE;DU1%uLg&L-dkffQF$KiG<kDG6y7NucN)fQ6
z>Ldl36YRstLuOJKZzf?_mi~QWK_?646WUveQ^DS3P#kQx?>r>}G#NE${`8zO*s9xf
zK`k?jzgwz-Iw*lMzT>+-_gW?X`V;RvaKhW7k4Kd&&0<V?1L}H|NIw(g+2wMIj&4ED
zhfqV`+c|bXx@@BCo!+~EKV+q%zkBfqi<ah0MulPAwoXvs!~6+~vYA4mCHF8hKgqL3
z_{Ffc9%0sm5!qUwW#_rgS*5HI4%&k1qDOM-pj=V5j^6AR!Cb_o;2HG|gR#$<KUi$z
zMeio)Zovbq9%JIFHFpB1kB5)1(a3u?wIH!Q>F<(3429fQoVNqF(V{kmPH%nc1jlP_
zf;cWlV0Gxz+pMHMsiXY++F*yL_ctfkV)Scy-RZGB!`?hfc#Km_nZsf4)U+BrDPY?F
z*VU;4pcBVpv#x^E09WA)kafemO1&m#-V5JKBVtQb-5vGeE~xzU*0*4P!t>v!-j0D9
z<Gp#I$e(JCTn$v5uRGvC==32VjK%9EkqLqd%K`8uIwZ(3gok0fPr+el;mXG?hAku~
z$KcF{RXGA}y;MPFp5%gDo-Yli*%yi_VJs@qrRgVx5CO?BYRPF3{T^7-%5}k5c6|UF
z#)U<~MQ;o8Me#iEfHpKu264C{kV!M_ae<>0;`j8aV2d}p2;>p8Nre}Trj*wzj6dKl
zwIUH)P95506zKKXm7U9y%banXL+`TRmX+!T(0p=GiF=BR1LL&8bk*0Q_GY@&xb&N#
zmT_KJ8b-@WC-P3E1i>&O>HUCDf%=%dmV$Sb$llAl+na4M*KClQSGOTseT1esDB%8S
ziYluKODeJ8HtKmF0`6{t&7hS_+hdzhB1eKrCx0ZuqYi%(74pd^;i*BdhMI~GPM75O
zWZ!&@L5oqE4FC*#JWLyq8YpWe?2R)EwxTIp1X==ITp4CvI>4~c1f_cUGsgaSqm1S+
z&ajYQbZ&m-?G?yG-jF8n8UT8ZB1A_e2u4fm$c9aOXD(dhMAbGa#^=IgEC6D!OG{EC
zT3GmaRvt=7D64o#AVmtIRZNjlmdU{|_0%^yp!iPV`ppm}B@I-Q>xvT87@_%lO_F}5
zqoIb-l|W83x~a5rTU@nW{a%?_guWLobwZtC9TBo49{pEg9;s~Q;lOPXD<cZg_5kBW
zy3i%UBEs(`2J{ivngh0nazD&OOuQcNu9GZ|s!W(KW_VR_0ha7|BHs}dfPp;4h(<$C
zI|e;d|L_fJC8!H%&84!0<(1Qz(ocZ8f=7hO?2F!+0_x?27wKmq4qsoFb|(lfw#UE@
zd~;~Dq-Kk9Nk630ArD@z24;W);$3b>8LQmzs7Ie}2at=4*znIK-11vYP!qvt)QenI
zy!EB;-&gy$S+`tX`1R^mkt*f?Q6EG=RLH|1++|PI#Z{QXAdG&JdWyq3q+Z42y3!3(
zrhg0EGKe=05Xr=3I7Px&>&E0?ny^P=Pz(ZI558{*POIi1*@nt6h3GkOaQ{=nn{%E$
z_1m01E1AL_8YmOGa}GYSFNXx6F|;Eke(45r`j73#FuAUv&sD(P#XnQMks$BUs%VdP
zZXgI_*84=EQFIj~OmzC*8^rr`tLxZoI-9e=I41$&ue)d+krGLJ03em*C!90qJZ<VL
zLWK=9hPPK2$$ZZq5_3}PK$gwNI%ibmf1_Y`P}mZFWlhahZPY(U$b1%a%a*6ZY023t
zADgrIsb{0l02pl{qUeNag94hW4WCn=!VNC=UNBrEC2C~0LbNd$HfI*jVKU*!S$u_!
z<3oUbZr}vgCpU^2B{3cCE`x4K&fk?Ty2pJ|B?bxR(Nv2h7z`Vi&+1M`2P;Hh5;u-Q
z3*P$HfsPjd5z4vNwWvMg)4`*~s~G{j;f1T3JoHrK3?qn<A;b#o=nQQ^z<2Ytzf?}Y
zi}p3Lya=g66q+TUoXD@3a2?y`$C)5ba=!fy!V2WX^ls3hCo8k>dTL7KT3cl?)x)?(
zmzd956Q<qqrmUIeTy0`cYz~kU?lOs)!Tn;V2)OpUUtMzFBsy5-47WX<oSJ)#|EQg!
zJeaVZtGw-a$e;I9N!2P48yx&H`()J4dBvJay?<}f4~QpQ`!5@?pzzrY4oI_%(8-N&
z2DrCgK2@`vj0y#ggzmS4ZhVI9Fe7V+dKF*In`-r^ofiFI*(6s$M2FSX8qn&;f1S$+
ztw&rLOb)hvD!*btLP}ir%GwZAa08@rUQl!6>%dB@qF{admb1Bj_ReLC$T8Rx&yn1$
z4lbnpkRfhv%mHL0k?}*kv8F^LNw?QW^5CJOBT9SzrY)(&#5+W@6Hb<VbBVXla>Bog
zR)DeM4dQKrwU=;un_av(D6-mX*BSawpGAjt%J3No(r-m=^9o~)XMF+mL!k^)hSI&@
z#73QJJ<&N)bhA;e1}Kb)z7LV>12vN4+=x@VnkGYkqa(%5Y+m$|CrQSLX=D(&Lfo0n
zP0@GP#KIXY1_Q=oYp8a;SXm8^|H5`CWa6~`{i?Hj@M;=^EcBM`@cOjD1xlwEyYlb3
z!AW!=&?IO4_4rTaX1nBru3(fwR1P@_+8=POg<7YI!Y0yfH6>z$?CjLkdL|g}sHSk?
zvEET1V{zdmz)ND_WkAZ0LucoWucC<i10Cp}OVMtiQR2{r|MpwV%ed+>qP(H&Gm!_0
z4MY+mM0e^jgtrxmx6KR-%Jhm=haktj^Q*3|f&G!a;(#o9B&P*#s4kxr`HtMudKYgu
z)(_+<%6Em3=fixhYur<Kc;q(KOct1{Df1KIf~-%}{wt+E;u6g+=&^|cI$jZk;v;YD
zozz$^RxeS|ODKGNxCd*NT@{mma5sR&w&Cu`_bDp{J49Y`gV31h0Nh4U;6WtasLGCj
zWV~a5Tu)*Jp-M5)>(O)e(Vql5*i5qO9RLc&0tJ3$2eeH7t;iY5Z7WR72E<j5^dcfZ
z50po-_3~^no-9liLBHjR4JQF5$xJ`BGqoWmP`Ph5a`iJ@+lvrNjLb8DKHKfpVRpNU
z&W8e6mZ%>deZcpB?^3J&w1I83rn6Fr>CFG^7s%Z_Oiq}e6Pjg##;j{GvYl0r*VD!f
zx``(hfb4b@#?}L1kM=*~OTsPJ+=ex-SRcy6hl=#%`9Ptc)5qOH;Nbc2sP`?n4&oZ(
zYZ9l%ZUWXog#L@xRfX#0El_`zCeo7>m#^<i-cR%sKy3=-wqyfMBOvjZpjhTJtrsuD
zc5Eol5abW(w`uWRNm1L9Vcd3~DuSJ=G~eoUR&EY{9Q2vjfRd9Wt`gK?2oO4C-Zh?n
zhJ&k9F!a)?TLJ{4qIv+R@4$J&HdJ&kA+B_$YM3@NKju7fUye^$KTSeA_>RLko9@y9
zb15>78~AMRa_^q7T6(Tfg>TfStiF&P5_#_e%`a~G?}#XLKo%RVQ1=%&qk|uQ080v8
zUJ9MGyT^?d?{Tm<PL!J>`n#9$n!yAi--64<Psz_W;VKFnovtcJdVCPM#s&jzX-^!4
z@la`TD<O~BZVE5fphKA{#&Px*EXP#V3H#JTRCC=3E^mej0h|65u;m4na2pFmpV6^9
zETiIuyQO*QZ`I>F({F#k#)-!{3$qS32<eq;#i7LI+%ZS@PMj0t40`O$Df+$t_W<cA
zG}y<OJc)VULB~nq<86FJ@t(6i`5zqmK94^+^p>fAap;W{CdN-)m4d5>iFZrsjUxqd
zH}M3Tn4quUct<hPvlQsY$T)peNxhKe*-a?d*|l*$)M6nno{?DGm7#B?Rp!t^Nuca^
z^yz&%i%R>6_FXbF`Evp_;?D$W;a6V-$6C}HV6B6qSQh%vYR4v!`>=n7!re24tB!ac
zAm|#aEM(~eetK~6_N$d|#nF}>?#4`(#^>)v*&V42mkWG{W4`=!FB^$%vl^Wp-mW>x
z^u7v&yQIGzY^Vu-+V{;M%fM^mKI&KYLE$`9AlX4$q%O;o<aXqms}OzIsU0Yk;+@jA
z%kkUHe^$>PWfYWCXVn#F-Q~upYjwvsS12%ha@aMB4b?%e|M^k*FFwu<6_^9}{gQ`&
ztDOZEa%C^~k1SiXDO4WpYK7lPMGQPC#CKK|sEoO~L!}PDv56umfe}`oKj&c}jk9=;
zf@-1b{2;IFf0aG|@25F`Q5gK5>X4dO>Cfl~0K1I?k5KMGkl8;iw*?tySrh|G;-AfR
zs~0kU7D_M$C+x5V17CLoaHbAZ#9IX~-&-EO9If@Y)(3ie8UM~W@{fxTqFz(V|6Ayt
zn~x~#3NuQ14t)br&eZFn1iTbU?NN~G=yXb-iQ!XtA1+RW`Vd6?e}Kc&<!Ticl%`ZE
zUA`F_0=dKqPi#HK>0f34zwg_>`eT6_Y!#Spf~tSTd%5usm>&bA_fJlb)14!bq6`Lp
zfbsk4?I`XsV$hj`oFUV`yAPorg4ieiH*x;cCcYw%aw<4{_P}IM0H2V?`BnAx*N$nj
ze|uQ^FJoYl0*=uxXnSsdy}=(D9;Rl!1la@9hKv2V@%$TJuMe2CoFRCZK<?WE%EQ`L
zw?Jich<l53`!9&||K$t*r}6iHi33FcXc1-4|GMtK2jKtQA1w-Cs-(bu^!V%N{;O~L
z7q9Oif|;F_TO#<c2IjwcJ$D!;`X{L>`}b1r|LLOt>reTA$PM78X;Odi0KQUAO5*Ps
z`o9m*|3_}dzn-e#1}xy+fo$&kBOqxk9#&L6`!o6buWESz>h+06R9H(1!*~ec?pNvh
z$2XsV^m`2m==Z-O@0<LGAA01C4-;{_fsO|r^&~cEz6dB-Nk?-ZmkSfa|61z*{YLP=
z{Oc(aJgP5H&g)n56|$`zf(S_M#p;`1>{$x``X2r7)8_A2ayf9Q3lBlgYU!h*rTpdo
z*W|tb^inyo@GtKlJa}+_ziUtK@fXJ8XBOI$&-1JMex*`XZ5vSgluY1YyL>W;$DBZ~
z!z{(Dq8~J%q7r1XdM#~gS^d5cqJ{q6gjuJ~VpU_yx4km`+V{XQ$C1O_J^R|HBgefx
zds}orGi`iCm)mjFyuzx(f`$ndj)odB$PWhxUQ!(O|MLGBM}PLBy@__dVsiLT-u~Bn
z$l-ABRB`{~UjaTw;Q>7DjiNXWbNFAM`Ja56)a7f6|LjKm<6Al5P{WS+;ere3(f{w<
z)-@&P|6nZszVOe8)YRRphy)1<O#k`FN{R=Gef__45kQOJhw6*}O7=zGe>l1RxF9ew
zQ2vvl{Odjc|HSwo5BC4-CdLz*BTy*_fk}qf<di&A+<xHaXfk89)Bip16*%Bwg$zam
zz%9#O3EJ5QlZt@$2vFOCQ7GN0S(`T!V4Yi-yv*J-Gr99tCe<AaP@u}rDxmB^i~jw*
z%@qVoj#|Gh?JA7t)6>{_Yjdw-S@^yihNn~!?EvjgLSdCi7-eV;MssodX-q{1lF-#v
zz(E@X>JoyiRiEMYelLUGOlvSkkS>g*3<EMu3C#Kb{pFnAs3YmTEW)5<xaPTfy>j|k
zQlAiBGMB&~RH{YD0dBz%Y-!jqtL(1hNjK);NB0j&@B4Qp#cRx3;^tg~|NY|r6~`j;
z=mJ<8gfjsJ5Dvp0;vbw2ird9I>e~x|^h6A^^p3fo8vqyELpxyR=m9vEr1qtq-`GA!
zfk#07U5~#Fdd88e0}<Jn``ve7Pe26+iVXd}1&UqI<HVew+otaBz_>uQay90<mH&40
zM#&fY`f@6zg8&%B1eF*I#L7>8fj$ju>s0rb6RS5f#iEZiQK>t&#qO`x1jgN8UV^_!
z09t(ZXmRz9&$@0>dJS~%K#O%BF&1=?{KGWX4f|>Kb9%19faOX9vr%SU3;ts*_qRp7
zJ8$d^23A4d8&EntCj+W+eVEwZ!-R2TUmplCvr<yP+2aTh-8&pB5NJ3b!pC|pt%Lgl
z>vojde-;Hs<8{<$P-4V|Sq5Q%=@38~FxaUa&l2LduIfTI1n_7|XjT!GwIF7KhGI(G
zkBpay>9$nPCtQOyMYbzv9e)q#KizK)@NmI6gl_VOS;Kk^kYnq<k_hnZ<IY}#^%?bf
z2BhWJM)j~P6HNA+(-lGP8#j!qnIr`Kf1pKD`S)#`8!!ciyMYykn3@9*7#$-)A9O|n
zp9rt9Y!RWGpXOp1;4tV5gVtn~RD&Af^~rpjJ)H0WLeM~^*CvbMdo^|DsK&ua{mX8m
zZg8J`809K}>~Rg&N;#F={N*>7MVp@LeoCR4VxO~*`oTc{HGCqYyMgc8F<8%G3O{)Q
zPF2{zCZxwW21agt4M=9uV0F@OKnK+go`Cg^jlr?Ve~ijVKB{<JAM}aTvG47^{+iaf
zhq~Li=gK;l5C)pMT6GH)X6_I^D^)t{C@HTT>L@m*iGzH1ANS0jqk+03fuN-i+dolB
z6m*=AyMyV+KH#B}WMZl_X1nDj-i3ISu7xhh25zYJ1Q+ZSVfyne$SjX2l$Doe*tK)~
zy*m8UcqdgAk(BX9Lilnr)@NhZ=mFBZd0<AIrSTW}WvPVGl<<qF3K3<^-w)CZYi0WU
zLGKhUy}*E(4O024_b)4b$9G7xJ3be!tHFohzc1en`<{Tat&2v3JX?K|&S)TOs-iw0
zhuGDi(oz{^&_Lo#!9h7Un927<na>An@CZCg=;J=;wLN)}?0d<gObDzJnDUo_-@<@!
zvEwM%cA3NKODrPjI(fUMR_URp34-??2$9}QA`0t6ThOnI_U4Q_9|N{`f_YDUbqmmb
z4>f%R0IXv<n2!GRak6IKzJzX72k@z2gZzZ~GazlrM7<9nYRlax$wKv)K=s-dT1C^m
zAT#lHe<j=k?v}U^Y870#0AhyC6mp4#zELga#xuTKg_O-XzA0*V-jWSxUlkmSeD9gj
z>w=;5maEz8zhjYqy7u_XL2@0s^(j$NJ*aMXWO}YwE;?>LeHarv{eXRS4Vw?49v!8j
zRZ^jg3!pLUH#R&YHJAuyFG94Bm=9Fhj;3?&c-nqEszi+g1pDg!uKO%<;Kn^zA&gY8
z^ITkn5kWTyGwlx-&8V|jw~Vzog8}jA9ep>|;Vo#6hOOk*jjuu=(h1s4F!>#&VFNk?
z?qoCU0nHVIKgHEd{W8QxXn1lk0c*(sGY(H)<Liq2*n1`MXNdg3F%h&Ia?u8OhB)M}
zh%}rPs3S?|PhTHDtvHj~#rtE+_5UsYVEJLj?quOMzfp_~`t~Q)65*{+$Lh9VMJ@^T
zGcgmnIc@lJRq&S+H;~4^y92+jwe@*9wISRtQRgM(kA?4#Edw;WX))~p4bpalH<M6(
zHTytUGVY~?DUcc_uNgNsH44P!Lj``H%EYyoB4;xy-BdxXFmn%^qxcl{7%0f{dR-0c
zW0Bd^DI1b5v}pz(&2flmy7a!KcBl8fihF!h{lFcFl!K^lzN5Sm1?`2x@U^Nao94|j
znyof&t_6u;*EAA|3KscChJe}UH19ShupMNg_vqpy$f7&=sm6|c39!=gJ5}_B=rWbk
z*qsR15U1%BR7sQgEp6$+_x1qYMB53P$(_0ANm>lA{**+{ta{0GP9npx@BXJomxPNS
zM4ivd0p-*O?3j(AFa!szuya6q#arpFSXwW{N9Bj1?%NgCL7@u@bM;3Ha~*ubCu!q9
zz4AXHbX<Y0i1!Z%H%#qrF#W)o6#9yfYN8R=r|*1xea{`BtM_Pc)}`Yx(Mzq}?O_e;
z?7a7NuFpGtq0>1_dmBK^xA4B6sC@@w_^Ez&Pxl%unq)ASt?QV8ffuLe`OgZg%<m1P
z!d-VTm-r1y4kx?Ep!}O3XF~T=uxI!)zpNN&=T5;oZ~|t{+FW&zbXFFxmHORT-+0$a
z2$}6sn?W_&#|b5`ANbyFjld`qlG1Tn7<^Td5eF9M)lP^onm%iw%G|u0#PS2G@-`b#
zowtRnTjl?_fsoe^aD=;)v<Sce(t6~3f8on+WuXyO$jmt9ccwgK0&3zrDN!$+X(8WK
zg_M4Oy@HIr0c~*K5GJ9V!E3bwAz+h2akQU6hG_)+3BnL3S)A5Dg^&N;gZZHPcYIHN
z!FpNnjd#`pCtc87`+@VDH&>p!Z7;IEFBeU}g1jjdF3>+Re;GO=F-y(Aix8hthrMSf
z8kJE#SX{Yx1u}lWyoogp!YMyqX~q=z(0J}H?xNH>uKaAYn9h%3cB2v)y+9qX;mcEI
z7(m+-JaL3h0c8lEFjxrBx2HP!pgDrK?i5<nCfBl;GglNSTCW*zJv9D&-ae~R`(1p1
z8FYUFkF#h;c7)waD1RzK^?5w}H<63xamC_#pW+(fg%KGxoYQwUh*f#9C>7<lQL=!z
zb>#06xj;pgkJ|A&+Ro|IrVX%ARTxY4yMQg*MFk4_UNygJoNyOMF^>GKIy6#^B!Pwj
z+eXOPha~+gC~wC(CAb`klTH;9QN6(0bm5#&z}VNkaC~zG?3^Jm&%5`~T+7N)2>-KU
zRBUVJ+>})_uvC`%1b^0~;mk+#&AIGlT%zEdjn_QYbz2KYsIb)9@w*J&Uj}i|bg)UN
z)TJ>^IuaUTAPlw3E{Tx*{?)=N5T&gcp3~ic3A%b5?6$<-v8!lz{DpxP)rcl8_<bps
zBHo?-c4X_|xc7l$jBETpr+sfBM`TWug&9|u5f*zk4K_U&g9%bOGa>G)B2qeUCD^^>
zTt=NRmsq$2A=W1QH`w~GH1)>w*CUuj;lXa`T`B8#%1h7_&0Ylyk-&rn!FYgSKyOUB
zO%i<=kJ29DzQVPgrHc9`RfN}gLNWE^Gz#AR>fJg2;cFV9`bT@G+XcJ8m~P*{d;<*r
zi$^i8ZFFrH?T5~rl-^UYP^usP$2=_3opqz3H$gBC1)=5QAQ}!0HKVv~fF?q6+{iiT
z;c&QLd&uklaK)O-?<iCSF9F#^WPmk}vy`fU3Tcx&`}aeReV_{cJe@uT_uB!MyRKO$
zJqzFpVx1fR!RakxiZ!3)(dD|^qm0~fy?l7SLqm4}qDx956Oyqmbi97RYup__iml_;
zqP3ZUR0U?r3LH_Ct8r0&9&*gS^bhd&3b+Ch70#RUnv|<&Z$AVRzN2Ny0P(2uxFsV{
z^Z2Qmqc-D&QLn2|ejlfg_YyDb&N~6X(@6`w0XkE-QmjlGLP=dC1$iy5Eu`mrh?f3g
zFPgu@X9{LutDQIi1zae$*LQ?Bq0}4VFQg5}q{;TE&H`&fo=MM3{xFG>Y2JLEB^dwX
zZC+M!)80+O+y!WyxR$YcW0w0SKuk);g!vtP^JXwH2;))+eIi<hubHVEsnr5qJCFMK
zJ0_R!r(9o&#l4;h^~&{%+gEZQdVSn@)Aly&`)1LUxUSmfqr;*H0+H_M-UL@>qN!Vy
z12c8*EMgbLj9Q7uDi<joflraRykLH<W(?$kk39S#6EU=@0&hi+tltB!GxfL@r=*$T
zRZa`GrQ4TB8mAqh9Z6gPQa(e>#co=Khy3&RIOfZDZ87DRHQw}Idp9o|%S#ho+F4*R
zSE>_Yss?Z+y6c(X_wDVdrf9z{t@v3D7vRH9_D!}MIp(LZsHcAaD2+Yn^mi$YJdPd}
z%<!1=yajk-JQIf7?z0ZP%*W3dI5>&YKs;(d+4J-bODZ-WZqXl4KE9@c2D#(**qJ+l
z<3~Bnff)eU=$}I$MDW1aqOi{0-(GvGa-x%L>*Nm$cl9cJxPhplczijP`$LNHfvh-{
ze?>JD&tt2-S(dHDIH&m~(ZLV9UJ8XH<orvZvSB^9i2Z`-7N4x4|LJ$-*t47)WUr|f
zpoD3~GEH?_55}bvX3lLj^?sl?%4sNQ%FtrlL_{1!&3Rs!wcT#`v_d+$__Jnn*vGg7
zr3H@T^xvxJJJXu@1IVXc2Ej2LWN|0u6bdT29WbhC$3cLuiAc>S2b5lmK19SlB0)!}
z_^jYm3(?$eal~n1V$jDaK}X!w$sjRSG;vPjH9sT{!r2{SE!~iU{vX&F)u_>419*2k
z&!#^Uv9K#d?S7{gn=q_u8iY3!1AzfLa;E};){U8=BM?Zv3*jOptVvEhwzOPtV_(CD
zUj>oP7RIoqmAL=sAU}u=A^!;t?i~PtQ}o9mxe0F8#f|^K?(HA<y*r8F4Xl<>H#lCM
z_#=2SV4(G@L4$~T69sx8H)&4&h?Vn9qmA$hks~vafO4`nNeKJ*Vhu)09QTXeIWYe4
z0taG86_;-8w5SBc9|VbFvI_%;1k`okv$O3uk4(2+(wz#;({Db37y%2)OsI+)H@@4t
zJ52LyI^}s#z|CiRXkyMy!J+MNLLC#rQ9gEYr4HEdvR?HCJRIU+7;Nhoa5Jwyj^#VT
zE}tKix{1Txm7!6^Wx{5C98VSImIP7)Z}#cyfKoC)asXvF%t1zlfRoP<1^>#gT%9mg
z_w|+nJI#^}ICidVKTj}(SEP={aiUI_;G`2j4FuL@R-^%X9EMqNz@fsxMLA0yWGcW8
z^V<mZzz&ZVE)*B6HbQO2GeVb=a8Xw-B2Pq4s>2baDSqX8jA6!H(bB1q2W#TU7%6EK
zDEvOm1?=xn*5w#VG@GrS#K{Z=yfY`4Pg8LL6v8%+B|pYpYVQGzAtDUz0gQ6<&F5de
z#L@Fbt4f2vkcVXrFqaQW<h^J;phn=rl6q-?Mha2$Dc3B1S#UFBv(20>*coR>gFJm_
zk*}#>T#i9Q(3~-}4w_iGVmQvk=K0+`t{@=hrYn9mafVTc))6>j6o|`(fCLeZqjY&Z
zMMV%?<{Z$J{sO^4?Y%~u`~&@g>KQc?c=wkl5<jG8IWPR4Dd;EvQtFqRz2WCSyL(P>
z`5i@UN6G~HbSuCB&Og-!gCx!;>W+$~3}?CquKBgIf`o9L($-`h2J~Sc9nO7R)&R><
zk(iM%eD!VIbF5vjj={U)@IRLjrw3&YoXa&u(o<(;uu$3H2d)A6sICz1;dBuWUyK2X
z5LxPV)qIF(yaVQmIIBf%RazE{>FT6|BPy(Yk?5+##ki)giewo-bGw_D!A&gtgJrxB
zoCWk5HNJ^lDsld}Ip(id2dc?38CNBNT}~YsY4BM@j&57GiKM9~TovY`QU<nRKIU<@
z)_<%Wr)E21^G*<P_FOy3{_E_EwtnY0-B8W$r4JVPmiRSA;T~VKIZi&+M8DrK#*>`B
zN_~+g@bWUR$Dl{dii|l|26FGeo&WO^ElUjg)!|_|M0`rd3>S1;E#4WA@3O}WLE8x?
zd0ymKfphzdTEy7H2dpEzA@W@w$Xt=pqXWflF&Xvf8$|nf<+vl+B8$<_3WXJ*9E68)
zjWfoahOfEbkWG;X+|bB(AuV%0%hf6BFfQc9ont^kIc+$4srg!*J%NArZ~?WP=$HbY
z{zaJg&dJkLzlwwhli0KdK8ihddpS77D2Tw}!>dV|?@ao>(-1>#lJF@@yIo?9g}O$b
zR_d<%Dj|1X+Q(=<&0o#^Nc*yMmW|Q}7dG)++Ki+&a0}x$GWtA{+C2Q7KDOv-7VxvN
zC-;}hxX?UI@)p~9qg>w}!rImfb0uqKp@qEf6TS?HwUsH9i$?Z5`-+7ZD9cSCF_Di;
zndeNHjiS>O&X>#XJt~WOm&wK{t@9vOx@^aNHc5Aq&?k>g6hc;Vx-<t%1q6L=i|}h+
zkCx63mxhK(Ylap1g%6B?&~}J53bGhXVq<Q`^H_au{WKot=D7%L=wVF%Qde=EnQBU(
zvRst;t2u?892CNF=U2?YiDY})RLH+giLXYYPEMsRIC<<@7lHO9L+m8MnqK8R;<Mwn
zc<MaG^40?fqo5)S(Ax;V+#&;!cGpHQwVs1RiEa^+@l)#cl&g0)Ycr&ewqbE#m{os8
z58M1Dzz@0kt|TF)ecTM?D~*z-jWR2ImfUWPC|g+>wRY_9`Q0Q^Ge#uU9AUQuP`UE7
zC?fc(BlzNnRu0HBw49Y1aW(UBp9#(hP`@0Ca3JrrL8eQP<{U&hb^vF$x@=@R6)1m+
z>x|xOFQ>;}2)IPmA!fwYMJOLn?X$cJo>H!x(78#|2^Cogbw*u}lo|6FDb?w%W$lym
zs4>>Wt6Of*Uwou|O=a_gN>Pili-#&|W^c2rE!P?=!8g=#1KW7RFp#any<PSs*>H7O
zRHOJMjuO7*KvC$r^2VcgpLH2z7Ed3Io;jI_CSwnH&~;1GL9cG;eRED6;&~yKWap$@
z@6YXoX6YQ|_Y6f}t(ZyFsG9LUEviDIja_vylvzw7idUL97;3fH7@lfSZoFU9QM}kz
zGgTz}MHgxMb$vIES8clN;*ZRJV$6eEQs^6<%qZ#HFjx=Ni&zL$WpQHjBK7cT|5K~`
zionrvG)9U<2Su?HMI950l0cHi(_)XFreR?$_vNQke*1P4!3q#!>0+L@f<^hl3j_Tn
zSClE+qo^Tsg@TTw4ukT16!{~(wTGiY#wJK7W>BJ8p|nQa)Y^i2u~QN00_G9hKc4rq
z<i@ryXPd#{nYodQ+s1+>TTP#BOvI?fCb5sr$t2pd^Lv!$E=Ma7qU51+GNc?uryZbq
z@r8@!N_u9jA7$YSV&{XL`##kn1t(2=Z0qUZ6e<r2I?k=cjqzN@P<?|8Pr6O9M0hKe
zEn5+v4LT|Za!mxT=dbm&c=B>8x0Us$+Weg%CO->8ctSpD&!KFR=|)rUrFKK$jp02^
zxK}pkCdn>zs3=Y`mx<tC`#(s2rH%QJpg8g<Wt2El6czDqO0uo`vo{>uwXWb8{+-1s
z>q}ymB_b}Z@(|8cLIo)9-AF<>$LD<V0c64<8yPd6#4Ztx3*k_4e^v9@S0Q)CAIwr;
zr5u?H(X=5vR&wHGrLP2n`^oY)0v}N}<c#;Fk~LBu;c!g!@QcjJ?7M#)=s3seJ{x5D
z^p#Hupslc;J?!l*7lPDYZf=y;glbV6z?-x#=)z_C+7cITsp8z#?2^`tjdu(`jXa5C
zgxLqF@HY;($9F%{oMt$GQVAeCjAQQ;!bu-dQ4vjF*am9XxMGB3Gp$@~a4^c*$B%D7
z_4QhG35C+at!Lyt7)%PsqfhjzqL_XYI!q9}>&|en@x_I_NM5fq&K}lve2Ob5NUq&1
zic8JkPO{A;3h~am+kiX@ui%p(EkZY&O+9OYOAKvplz5uA>kN3Rvb7Ld=*5bfY%V^;
zT9I6vJ%8;?OSWC|OZeLMtFira65Cez_MgK)J_wHp%JQ&SnZ};#QmNt2@uXSI2udKQ
zzeI3|$$KSL-EV~Sj^s+0kh+AcGryBnn}<q9>veKU&3>E>4PNfd0nrG&sZ>z`QufbU
z#VBUb1Z7m;)SIAJoxsxc!^`_=Y2P<$vvrfA;8?#2bsA1ro+>l{BlFLV1h*9vH9c=3
z%EUPq)I0u7K>{@@Kzw1th;Y>Fr5WK4okvka52s*Z!}NsI4fkUzMT4gm>)%@3$xJJ>
zE_Bw<MIE_K5TPu0BYQ23@%!Y(`@-Bzs_%IfXIM^xxam-R!CM0NhMbmiPF-R;NZzs2
zbI9t}DLT=sb(c8a-|!b9$U7Yncl#B0xiL_E6E;?_^6qi+3ADVMmlIX;xYX!`etjv+
z&GJ2X{)6o+nG*g)@&$&QYJZwDM>FsCo-4R*6U`o0zjJHtY!ij|=V{@eHJeVqw7+ND
z)6Uq#Y{IGuCgp#<=y}np!Koi{J=eypNH@~#*;)xwv9<^J+(n2Z%Rl;1YmaB;nn;%J
z`%drg&Gn$9)#G;G{CHuouLC};#O%~iz7KmBQ5G6|Qz;h?t}XyYcHT5hXP0VJA8s6r
z?W~jeF0ouI<;}fV@E4gT2v@T2@6nO3E$rUh%FSAuRlSIuTZT{RGB_|sJ{S=`YJ3-{
z02*X<`Hef(j(rQ%IByjM^%{^plSln-JIngJSzb~X=iIULZL-aBG;@W%1Kc9VH~6O3
zWGU(OJW(=xA<Kf(jsuMmZ4FK??T_r<`={IUH}=-**vlItO$IC8ud4J!Rw@I=0n1_j
z!eJ=P%|XLTXzN;3XisruVqFk{sGe`0$(y3~SCq2CQ;nUsboVg#-S-2b5|rCL)pL<W
z#I>`yYQzfpT6Pg)4HCoNYp)DVmt{IE-OK>y3i8@R({-i(CW0>B3PlC)%#@0g$TXk6
z*rt6Zj3~^~48;`4xu+M^^ntVqTEo%SuLrKBO|@Ta3Yo>16dBlg7Ezn5`=rEHw_8p9
z%}2^~Il0W9?Um$vP<~8|<>{+MYiiy987}PT?;x^nxe)Oy;Y3LBYxsz{wC(o1>qpA<
zx1=P|R&+}trx?;>XDu~AqkrL9?l>+BIp_#*@8Rrnv2D@iFAN^n32|q<SI82S&{Zs;
z$teQ+&!%{5k0Rwsq7U4~xpbLpT7d3w2-tV#d7?d{Fwk|!Gd7W2b9Bx{Z;=m_U~bxq
z?R3)D-y=er;z=xhvXmliXi?258k`iAi|;M-#i*PW;*S@luVo$PuClN|E~@<>xhPJ2
z90N+T$P_q|D4!<liqL(z>_uSwTl%CXdbks}+y6Spj_@YXMg_ZplU!ZTSuIDb@VD#T
z@=t-o?+1_c9vHjmHmwPc^2s1BqT-IM%eyV_znLq%Fnt+DDH2?mto_6q<okB(P9n2Y
zniJWoFL}2(%s%w#5fq2+Z3H%RJ;R*{&b>DmR#>n%MIi*~%=A{l9z~5#Kidq~8aVF<
zrks}k)tMyt_9=D^ZX{8<dnb|r$YuSiM1@f$Q!5M4d4R97#S^kec!s{*1rUJTuZSk4
z+@qe0;;~QPM(RlQZ8XSD&*@J@KgG@y%|&2Mcs@h1RlQ>|QCo#2)jovM*FO*@0nNVg
z%+JnHjs<QP{-#J7r|0E4X=;p>+fD1-`&+?ROxI&fj-xft7_J15=UF#UPD#Oxz~uKo
zE_#!iQu_xEa`jd6_!CeSYnM#MLAi}^&pfJsb26fyDs}9iZTqTZN_0XeBNSH&r6yW5
zOf+1e#69`<{*Ld+qFgG-r7Gd2kICS3;VWp>@)>D;ek^RYzoTw_;>?LH)%9@bbR!T$
z@%&3LRf$9OTOoeQvRM6*^_QH7sG`WU_jM$F_&b4<i;WZ|#!Y6t`!-#Cudh>k!#um%
z;G^jeWw*hK<+V~~loo3~<G4pB$iDX-aQ>$!gI}|Q9C!7)pTbTtO1XAp^s60X<0*xk
z@<*l!n|?R4aA||jig)8qkLn!hd|Pw_KPKqU8R~s`sCn7)VxCm<k$cLx((*I=WlxY-
zxgy0rAzC`;dSkwX=>zkO0)`-kw&Sk+k{81KsP01&na`t*T51=k_7n%v0XW3wh(%dP
z{eJzr%O0Ux8@`^z-8pIR&ZHYbc<54P7S~IQEyW|KNBDi`W<*e>-&KHKI0y{`<?0nk
zDysBI^|2yCje?&Rd((Ecf==4VaHxz7?-i_1{0Gs3AXB#L7RQatj3PrZhJjjydrt3e
zS1c2b`dx`9GRRgQXdQx#FoxM+<^D-2e%|xGIrmNj_L#t@rORU1RTv*`e@u(~5qm30
zJa~75OlFNkt#XEO>Tec4D>n<;|HyR8+-5gVIGR3-fLaecF;~7M2yQ_0b-q6FN>>);
zisPUa+05BoD*x6D?*V#Yqnqtk?!=Gr)?#fAarW`X$)rE@T4ntX4S;mIkz9*&StTCz
zU(lFv+I|=xf5JPWoxvfl>>9yPzY$nHNqtPeDsI?!8D6dHGlh;8>So)r%RtHY&<6i5
z^;F_W3T}v&)FuYP(`OUKp`6NPc)P(Uwrf1qEj8PQx`qw$5F?*rEV-@(l_!Mta_f=|
zuu?sr^fPMLmqn*xIAU$WO)tZrKa5!zt36Z~DPpUt3dj4*>ZB1}bkwf=$iaC~%6T6#
ztt2CamRjYOBqVo4Hvq|WZz1;!Lv8u`3BF^`^5b#Yo@WE;w5-z2AqnwcIJPR(Bwrgx
zNL}tTvV@SlR{iC>urvN~C>@k3-`9+Ga(}I1xcAP&fCKk;OAT>n@!PkqZb89~ytvc#
zxThn6I^%R^5_A0N+}MU)=xJD5<E*{cnM1<0m%_Q${@JK$z8~h=f&A8cU8GlJi$AJ(
zLhpoq9iC8=X1d^W2qoXWII{S~Sla+=<i$?$D10scDD_sZ4&wpHuUA;Ddt`a};jW7v
zcOi2VitRcAHVi}`+h7IxX_)(y$lTkl3&5jI;m9o(u5tOi#e907NT_+wIehg-+N^$l
zyL5+eV%&F&J7+f|s|-{k(w$yd|1|%Q36;I|8hV}m1EX<<v9Bx%C1Yzya`~7z#Vcvk
zuB_tuJI?)8m;}Ne{=q(b+=CI*0=pz5z3>)EouNdd(L7(j(*RvV-Ept%nh}(RN568o
zY+0ydsl9R?{hvjRNSOB`tMYk0l{Ww2L+owwVjV0$r(+7&rX4|1599}?OTFSh;&hZ!
zS<?>NCCEQhjpudVf8^+Xl8ZCZpGRj>%W|uy%$c%f2vDKyEt_+Db=V8Y4%7OQ=n~^G
zUSF8dd#Ro3PW=Z2Hek5m)%P(BClbBRrdyqNXO#~ejsdvpv?J>-r>>SJL--xmB<hq;
zHuaK;{t}W30;Af302#{2R3X62rVJ=LeT}0Y74jdZ?lT?u+JGP^1672oyI}z#N^bF|
z)yryD#r(0`li#XcTvRge9ih~++dJ!{p0A4)UvE#E()LI2Wfj<cEnS>(58H~W*P8qM
zU6QYgFz7ck{4Ig;q%!n_weD?9B4u_*YQ6A@$?h5T{U@KGPQhdL!cXDPc53DLLmpb_
zpRS(nADM;aM87Prt%F?Vi*l~6d6qv)U$mQ}>2dX^9&3UZd2A{q+n8xTS&5U%hgd#Y
zN6KpKtp`ni;8dF`UB#jh$}F;Urn5GBuh(nW{tJ2E0}(%iFE_$0o<kdDM|8-+*&K83
zwsrA453JLk?Qz}ym;0>LDktu224SH_hZ89;>4q>EmUUYYW%m2>*Ee$zp{`)OgInuR
zxjL-eqjAEad`tpp3;IaE&czb4>15I~=lop{steDgMy<h}ma*{az1wW&V2^`M4sd?=
zofZ6Y!}jO1Vu@(qYLuLhY-E6rb?7Myzynxb$teE^V{aW5)!Vmy)188ZG$`H8&`3&y
zAV>|;QX<{mElB5(DpG?;cMP4<A>9qq`0Vi)_w`=yv)29m$672lGkfpzJim2(4(`yn
z#ulGS%z^>4`PuZ=A0m$PMUFs`I%)Jb)l7)soH?C;{dpD-iZ8_)VO;n%wC{zDkB%c*
zBI_#FH`s=`pqloax0h;qa<ieA;SZG70~B|@>$*mj_T;WZ%W1bs*TLKp?;YR=dF>~7
zp=KR(jQTql!(7YVG5)P!5-NJLD8V4YeV=_xQRf2vws)jy$)am2+AXr${PNK96$?`U
zn*HIY)7+bn@sjBEW^Zs}ytScTQy)_pPZN7Q1bGgl8n?TM%f^yBjV=m4ffDbR{?087
z$SzI=KUn=>ez8pP<A*$zyRhmcCa)F~u7y+#eT9k$l`7(s9ei}uO;?!bX884JVVSgF
zCgk&(kYdi(gx1AEa`qSuS;xd3Cw*j!O8aSQi~JRv>HxP{)aQG3ihEBT{R#P7W>nV@
z3T+l!dBqg!$A5xY4-~TC(P@YQCsyrj`xan$$V@U6>w$jXz)`JmJ`=MwYMa>|Yds}T
ze@18^OCgMJS!N*{Z#^Xi(ceC~#<>^rWp)_1<7<vjk5QcdsKVeRXJGAmFX7B!fD<cs
zrfEPMYx$9J4o$>ArRE7{1zS1j8bJX%x?)o!*I*r8HAeEYKwl!p@?G1R+0=6Q_{RV$
zuY_}}^=Iyc(;sIrMcT#F$@L@TU!N26p?wS0HX$-Soa1`=m561yP`l%?z@AWYy*^fS
z8M3Cjy{W(x(~fENaZ$Ne*8wxbQ_bY>pPDRDFsFNG<fXWx0nuakJeqw-t-`!ehp{Yv
zYN|jo=|SMd*Fpy61BP2$k|793-uDcSa~k=p2zkq6{vtsK-Q5EFuITOVQBm$qYQY%Z
zX-dIiC2u_U6U+LstvLPJnDP;f5x+&4anbB&dU%@wy89z>B3ZOr(trlqyS;N03OYb;
zMs@{Q6VvhXDHk_Z=D86#vP)x3+Cui7wzA(xvIHqI2GLKZ64TEKYq&NW%IYb|dFa_9
zl1_QK+z5@H+G57wp62(p2-CPPo@vgDY`o1xgCs_YFUOd+wF?$2%<u2nqRJyhZHLX{
z!T5GbKuN@<1~D{SgXu|}xc8#F3z9S^rFxso70wP%*=g3&p9xN{jn!^94nn~-xX+3-
zr@zk2&$(X9cv6V8Y>CFCCk^t9zqYk~q=c`2%~#97aja~Yr2H!5rvCgZ!_m0HZjZF2
z*_4KF%-cdxYAOJ_b)cv&AS_^5>#dpU&QP8}r20x&Q=rcnR~wl}ByN@mKqk2|o++8^
z?JSH25|Ejpfct8nq#?|3c8@#K#(0K>84`!l-OZ{2UM*gr3O<bPSGT}S(sY6FwwO7(
zrH&(fz=Qs!;+F%Q0eKz8tLMyVa#BnfkSv~Go+~s@>#nzQmzA815^Fui(N^LEX)J6S
zUwLWefb!{Olzy3_um>f#d-}ae_f_ltGd|-<7-)*=uL2%i)hqNbJlY{0lnGGBv$xV6
zF;fy5cv{O|S=#orICfOlKSzDKH2RFWN$(u2f>=KBm`qvPxGMWwkrG&|PX@);X$(nu
zELw+YyDOeI?I?erw(vwnk{3OMkoZPrX*YyXNmyqbF_A!qu0N;x1nUJ_Kg7O+roRX>
zM8aY+X)u4VWg|_VfUCOB5PLHCLN-KCkEE2T7Ai7_cUZ0fk4zL5?Ehp>d_n;)ZA6K_
z$ZOs#l}6Qp%pr`sDCD1#!gy)r6JOP0&+pyQA~7Cx8s;l(%FEqbBvps`O_k-jKPaxf
zM1dh&u5jr-D-aSWwvIY}^$nf{Rj}I5S<|Ra7JW(<P%V7r^k-hm5({Zt#^iqYl>(HT
zFdca-xL(?|*{Ri&#2KIDbJQkdjJC;H#`G1$RpM}81Y~eNajtz`&t-E5Nc@i)kTHC(
z+W<#3noR{3DKMBibDAhq6vXtLI5B8<lOa}!mOdfn?h3D20^6JYXZ$1i0|tf2#vv<~
zTVQx(K`CYSC{2!xiwpXY)4bqp2-F|-$62bbL%g1g_oI;<=!@p7$YU0y^ww>fC5uh_
zrr<^<=+fdn4;V^CHM3JxOmL_BnD()yLp^RlnB-dPDOX>w!0#l5&4+4kd^u8ubX$^U
zN$v7Q+d|=pr0W~$#h+e@3yepu7OF%SSj}JcFTcDbR{e&5j(#4xU4MGR!(F8IzSU&0
zEGfQvFrJ=oOgrr(2a~7Fc8RJ5@;yI=@s7eQhK4MX{THs`;)4E@(gR9;8xkC<Xh`f7
zjLt>OZro9nJ11yCA-zy4yF)n9Gh|oUMR~-^U*Qwc?BzJH4vQ}{ZUoeW#>&({blAZN
zHTTk>3dUW*9jifO*{gxguc{0vO8B`hcQHGo*H!t9qJCTqcLH=r84<_@2+4GruBSb@
zFmx38sT{r3=MJ^fS8JjK9yWWkZLU-0k9uakfvUt>R9no`yRorx!ptxAS_|pte9SzT
zh@m6I`d4f|$Nr?z>~}n$ELM&H5!d$RXP~*_lZm<BG9uvV{IY{aRCGwi$>z~uOchs(
z^bPDtW2I;kGtgZU^#=gpSkIo2u8o%cdtj=8kSyi6*I*ON>Zh7bvYE;YI{OXu+AhTQ
zY(!UM=@D_H#>-io36hnH$cMXDmiyno6^yd10^;QiZ@AoeeqoD6Umz<!sG^r`mjh~X
zwRLl-tOR$tw33?VN$M-udanAGE8{l7y{2Fufu(8R0_d7^gucl3hMMSE88)|w;48JA
z_bXf*_;lZx`IS0YqWnh%_|}e63eR4Qo>RNFEr@&S*Qj2Z4Gz>D!nx_ySqiV2v6E$(
z-!Gk}d_P}sq^Gfo=}k*rY?VtgC2?nVP7d=#assM+rXxpxAH{Q$*q<hk_GPRzeVF^J
z_#HTPqa7{{DRc<K-;j&6ISgn%CvpvPoKkwOpP`WWiroV`I)L=Xcq}KDCY0v1dC<!t
zshW|kVAj{p;X}2O&B2TzTYVXve1olou+zD3tYB>VgN`PHO#5k(e{20uZjuK@gO3j-
zu@6n3%WMjK*9Xe_uV40Ou<lHV{@J<GV-QHfXJOhk?a@4pXAZEvPLxe%_NbHI4zWO7
zDcH4NpkG>jXuU$r<&&M@7CwqJ11Tid7wG2rf~D#6jl*mnjqMGhH<$aAgwgB?acH5e
z*NSVnJ)9QKZ^}5UFzu7M-_nS@Sg@C{Ek9^dIdkUT{Nh320n9;-YY)$!PktimO2Y8p
z9RFS;Mi!pwJ+p6xP96W6dMzV9klg+_kRe`doOB4?zD4#D#j`MGa1@2-OS_%tOW0Fb
z!KTCH9{H4PIzFF2<=+RW=T3bCysxGc&OH6{R?1(B=9Rokez{E9(RT+UCK?M@2jX;0
zX)nO?g3EAC^@L9XOoZSm0t_{dZwxyc_&RMNKgO50-%3c>&fVr3>n2Tnpp`IPtQra!
zrs~A2@cQw^3s#p<#2oJDl>(`W)nbZTqIo<35#BW{VE<{({VRl^XERMYKeAy+s3-Qg
zcQXYrx1adx`8-4=KjwHkr0EM_RmT(KR)r`ks-S42-kR_*Lm-!H%=l^odh;0R41^R*
z^sv2edNBb`qNixj96oDh&zaDJk~~=d+^L6;zeJ+!G;ODbW|f6X!mQYR-podH2h@-a
z$o>YVXHP=a-DujX<x2^yboJZcD9r8j1cctIkVPTMgxyvq?&!h1I|5Mo6bZOPqTN6Q
zG|`Jq__O$zSnPH}s>EQafJncVqz78EH;)9e&oaAdbEg58_q6I!L0w<wtyR7t2cFgx
zrsL|*v@Ok3jo}2uN`zqL`h5h<Zh|C7l@YC!LWMuem%D-?A$1w&9o^*9+1Pz{?1ZA3
z4LSb^n@c-MS;^4V$8sp<lIHKnI8|>@-RAPPm4n|dHBTr1t~-*=fig>*gqLj@C*?c%
z=ZqtfVDc%F)Ik)tc7$4!n41T<M7w4<&{jUQL|s(}Mm|1Nr#<FoB)9yK?(3gsR@r@z
z`-1eH)q-?AkEhv57}Z%aQDgnKV>F47R&LhmAlLPZryGW4#bqZ$Q7Q-ua7RVkBmHue
z+?Fityaew*peZV{SN@bon<lvS3JxNVc7<AvxLL@1rJ7V0mz!#3hcb+-i1wlF;!b}`
z1$r~Fj@3u8?_-KMS0w>1YeSh*fu3a&itfI31=t%H_pTk6P~>vpZ;O0yg%f%jR2$iW
z?tGQjW0i<Ian7u9mLKT@1RqJhyG(12us8yODl+N&LL;e=V6iN!--$U5uJjw99Dt#`
z$?NWB%)%aE=71r`l)|r?M6Pse@&Hb=9l$LzP1GWA$C<5#@r3X%ux?aXt&J3C3L-4A
z|K#}&>qlM+v{q~laQ*0G4N)Vo%H}`zGx9`TVh7RMzWW{UUXYD#TQYsR(xYV<wC8g0
zVe?jX02qLDTc9VpDu^|o=oFgA8sw-Fs|R>%*M~D=EUCA8?O?pHy)S?gp@E6TwU_vu
z#l|mo?<=HUU?c#$e^=TpiuE2p{Vhb&QD$n=o<<z?<w^(8TEJ48V*9hrkjEmtPj?bM
z`(g(+qn@g*EL-Lp{SN80Ov1<Rs8ip-zYN;;*Jrj=ndOp7la?G3{^;~n_W5QVJm2jd
zvP8Y9gtW~>*JC<25eqJPn9*G3@hLj|9#yW?EREUM5Tc1RZ(FEm2ajdcFEHmHc3d$p
zzdfC;X(tj(E3y2oS80hiGN+_ps?8L^NBD3ouR(I-31;=Snmi?&2C51bv3tov)qtA}
zo|8~;O>niO3wfC{;5Nt@p|83gUoYhR+ME$YDER8Dm2pAo*`RNlb8PkX`-VO?RWNV1
z(<HpV@dYGXUp$BI<#sN8-$6TI;(Dt2Ytht#(Tl8yzRrr8Sns)2QvhnYk2P?>`&&)I
z<Lw6iE9|nbdC>|-aV?~s2zKmJBOwx~3v)Pjtd{#B#@zGq&OU&SqONJcTbn#N$8=9t
zU##GbK8bqv-nQawNS}u@K?&qjgQ{0=?6E?|;kR!f*v&E5qrmci_L?r)8=HvEd!#dL
z#XzylDaKpz3U3*(u{>CU*D<!J3QSZFCK8AUp)PTiZyyXrHxOgK3sYSW)}O$ZzryZG
zzO=GS_&F_H70UCQPg>_ng4CXlZAM0odu7yf6G+yvZ}*|<mEG6iVB+5`ELXvV`*$P6
zWbHHU4{;nSXbfT~^Qz=AE~Iv*yfd==3jnOS9drsVYy%@#fs@w2UE#w*ir)`p&YCiG
z8o@83r9k$CbGglc1}2!p5Jmd~inap^1qKSmx$Dw*CGASo88G%(V9o40A6Y(an&><j
z8`7SPpKY~AnYI!24uDy?^oP{hX%VZ8tz8w?LCFejccw`OgIrg0)H2Ue4F+)p^RPFq
z7ArK>iYGAviyeP%UQ;iQLdRfHk3c%7zzPm5^u3TE*V=YaA^1fZSXM>-3rCCVDv(&f
z=SYm@+LBk>F)*pF?HhZ7IA+kHuh$dOkT_@5cECg<G7)xL4#GFv1)cDb+oMq$ij|>_
zT77%mT5BphbPewD+RNZ;#q9xn;hzYp1LuzNl~{u5Sb}YMwPZAQGLK`33GGbj(%y5V
zUF3ng3Vis#4FgRapBcr=rQdTX?GvUB=uP;1?~`*&{N=eZ<TZA|ndK!WOMLHNachNK
z8pYVgXb=rXC9ff-oL|Eywkx~IXl@JLGj9YS$?>X+Ouoe&Sl3q}HQa7sl{a0TU=r*g
zmS;Zu06sFYEr^f<bP?$tV`RrCYgP6k2OYoid-kUtt(6MGUqPy;cem$TlbV@!q!O8&
zPsSM2Zn1ikP6*eqc8m3sj(KRV-x$h@S&DP-0|S#gI*#}?*r$$DtU-i*9gF~6c@Rr(
zi=2TywBC6OjX~S9$&$Q^uV+u}<r%_)UB`}@((WU(Ck%NWH=ocaaiJd`$K>r@gjQKo
z9CQ*#8e7HMk;b^}Tm;wlU1c~uHPbuhL5lyr>{FWwLpKxEu0}Utet+FYMnWD~mZwHr
zLZ+U1+Bc8Bb*%wJ`LDDO#B$3uM`GI^4~E)te~DU>u#ud!H&8qS4SA&TaPck>m$s8=
zHg1yCIBlX83m@F@pY|p@8K+hY*_krT(5wE^+Dda0JqY6YR<DIOsokPJdyKy6vo~WF
zf6&w^O#~Dw^#pL6mP49&Lvz3TG7zAfKdDC3?{hVe;dp&(7;#(k%v=NOyjivX$>t{;
zYDG}X^^jxAaJ=@DjJs#8>q(wu%O%Ts#<Y$s&YLcj3~UKTHM2gYpyIsT!h60g3qRFt
z^930b`R1o&`g2<!21m$qarlZ$$A2Gx1!kU;s*BZbX=z+LfC4dSXS(sP2reqXeiy5@
za)0gZc~J-;f)Jhwqn^$zvAw?Id$2NAm$V3>p8m?~W{%Te_kkx*|L)=;<)Os{>?4~y
zGLyrDbaaY$FXCn1Q?cy_T`MNeO|%QL_d9FJI$^v)&QU<Oqv$i5_5~(L(HO!|pu$_z
z<BfNMiTExfgTDvIWBr*Z{6V#`nk#_j;<WxngWz^*{ij44-z>@PS*QAurcOaLv?-dj
zicMsTu7!`TlnmfaNB7~qNzAi=pABfc0ru59UFqVLxHRP@WG{uFM=x38ody*I)$M62
zYQ|qkkQxEBLUg~Vdio2r^bfhMe8yHe0~z?X(i4QEI<xXxxt(O`vs()=rgQv3wuD!M
z1F^?j)lO;r5Cxeqo>zlgo@tj`8&(+!@D?;fllSlL*C(a<pMJIbJQ36L=D?O;IF*QK
zsbQY%Q_%D=q|mh-$1mu7!;$`ro0;3uYH>mE(p7v;*JR-$5er_Q<qhN^w0PU~a{4<&
zQhGQKS+><mb~k|iEgN^wbN>Kp)CU=f1)#{_w-}K!912_J0@2i}5jPKfW7SLP*4YA?
znellL#w<(;WfboU(wH8EKlEUv7A;dw4lA2buF8*h&RZ_UmFEhGO=0M-L1Q;pnvZ-d
z;=lf!tCUapDx?Z)Y;ejuvuF+n85H^tb_6WVFBCR%W?qoM$C{$xk}w6U@Kb7}HfF>k
zw5PmCNCSpOv8P^EbUe*nKXc@+3$4NVxmR=u<;}n59>Te$+ob~bduY+52UiiJ66~Z*
zPY>N&=h3-$KuI*GZ)0{``uyo=coT=rz`8^YAir@GC8TPwpj6s(0w~6kY+k3;qZAmP
zg_9)@OtKS#KxBqy`!i}hJk;$ST{HVHs74Bt8*iI9U<HunA(e%r#!P`0sPcsTX&_M&
z?>rr$-PVp2Vy=4T@LzKWj}b|>$r1Ph*|C_9cQX5vHF18?Igc?_wc%D}p}Mb&6VVeN
z5ESd-Fk{2Ko#)qNd6z=pMZCvPkMAJq5DJk&>Y`UXR}BN*i|hnMp!dZ_>a_+01xaAp
zMIrez`#xDi?qLS^2?t;j-P0DLD(<PAi*UY1V2N^{v`uDNLnA%|cj_T(k;<ywfJuS}
zKO51l#K&R~9H7ObD-wBVeBa)>MJ1*opd)9hi~y5#-iR;N_s(EGqcr2(Wxj}M;?`xL
z>H5~s`bC)MY2$1V4~AKUT|i8~@KaQhQjNo)$N9@wjLq5`sY9%mi=36Sp|z${jbaBX
zUgQ%A=n*EEM?fEk{!X);Ig>10n@cHPU-0Y~Ru1!fpy%OZ{umeb7zUbDCB|CMK3%(I
z#vMX+(dYFMGb&OeFx5V2;&)<L$m7U!smgA_4;3T*hXtUf^2|QQHQlUoyKj>k>bF`Q
z#M_S+(O~t&i3legbI^v?m=u-7iVi_rLDNo?B<4E{_Abg!Ub)Oee>f0*cE4i6Rqgxo
z)RwdL?tQNNs9J)+x%auL<#mrx_QsYG5GpFV3{qec9MST9|Dpc$?yU9P0LIpXLa8l*
zvWs$0nmJioK!6x6^MeP~9_bBBE6XVL3{L6|8O8)N(+;*4*;BqRza(oyKmWa*DU#{j
zQ<3Ux3*3ACB|Fe21P>2i#f5@4UAB$%r8eV5L@0nXW(@~=yh_nJ*pPd9u#M%B<H>ze
zDIGC{)0yyjJ9VeTBVO{xaPuB8G5bJ-PZw?>WKqOlm6)>^vz7);fA%EzDADbF+%P<#
zGwY*-q~xUoJ&k=L4jGpY-5rk~WLY(kH~ljYt(lr2f~JGT8TVnn2{^V;r9yV@%T%Nj
zfGtp#DE^|x?P%U&@6f63N_jPq#lz*EL2HS7_bwNGw6`h|Ymu<a=-hFevUDxxjE<GQ
zb@XGz@iaON%~62PHly^FYs7oiZ}Cb1Ky|smBn$&D(q2RAC^8GY<|tV6gVd^3bE6JV
ziDT#Wm@T)U%u&C*8Z2@9YIFeNYzsQ3a&g~oSs^7xox&KckirT%7Emb#l`A<-xAS(7
zb#d7OlTU-4D$Cm$pXYtdDKY&Eqd`$Ch%V#X!Q|1Tgbx`B?yhTKaSUqQhUr<TRnpyf
z+YRx(2!)R96<es1P>2+;b)XkB<5wrmn!YRMQz1FfBQ9W|c*b2i!iJ^sHZK7z#L$cL
zK*!2$ZJLNkH$oHWSb~=xiHQAiHlaDUc^HLWy$gJj+g&7v)~5yj<vZSkKu4iykm8ix
z{Aweu3AznpPTt_TrM3tLKO3SfkV1L|!%ILu4OC)B#`DAcl+{kPe`UrSrWGj6O!H?%
zGl1ZV3(1o_qxV@N$?5dj7R>O*@<#WjP8OwOS1)Hh!9UgcMizqs=C@STDwpFt_{}^v
zYR@8S9pVw*tou!p$?qm@a{Bz1T9{jctk8$46|5vV@{|T=8~wWAh#GCgZ-;Edc-H`G
zV~LFVRI?xCo$b*cB;lFHZMR5Y&VN|Ig&uB-yc3=$|GGWtoROUzpO`~9KxD)c`N%xi
z;!Z9@i_Fp|DTnHURICqqXH~Q?(<`CJDIr=5kTsfP`tOL|X-ySOc7v{gv3gqUB8``l
z6V27jtnk5Thb@a8iVaX2y=Y?m;N$TtYLTR`71zSin`qZo<09rP<{J7^OV9Tq_)>AF
zc5h2y$9}GbYT2;jlDAJXl3z`-&As^`!2Vi#)FQFe?r;of(lF+SnDablV`E;rAv}Oy
z3HYjb2FQK3nnbM3lfd~*#KNJ|L;NXM(TVtiCmH{2fH*8Mon`4|#YAK0o$k&^W{QB$
zd@uJk#ktC7<lvk&zszdOw4*)pKdt(vAcT9`6hJgz$r$_*6v1KlGXTScIAP1K%#<&;
zK9!BiOp4b0Oxi6cIX2!eE!Gg=8h_vOf?f)T9e2yWWB*m9XlUNdOem;#WkrOIA}N?}
zhMI%Bq{n4m-Y=YV2|fE!d^$$bm$%y?tP86~VIx^B>0o`brp{i^a~ZIm&SYP78X8ky
z;Q1EZNl6`DogWF}!2=-31HCRkFr9}EsEL@9Kq;kKqPd*d{^3+^Y2H5qU~OS*kcJ_{
zxa;<pL=UC5+u<`9=^q_u-2^7$G7|*A<Fd`4pI_O2_EiC`;4flkB92j+GMD7W<;FtJ
z`3NM=o<eJHRvQkiUVeCT&*W?>=(>fh$5RXf;KPe&Xd#kK*8Eh*B!Rss=ZXmrQ8G^?
zt99dJXyhMWQcV{g#Yb6qRqQMSO#sy&_T%^~@lJ3@UZXUsb$W}ykB7X<IiXy5o>eON
zBYPgA^uCjF?C(}_Sshm@Fg;%h;R!RADX$h0nPFN4iV$Wd%~o>bti->Uzr-SC`Z|={
z{dG?U+(QS(mHJ0h`bR1Y?jYo5!yF@y9%Gat${~8@uZLnsYQx_>QW^jKG>b#3`U}Gx
zUWe}Lt3ny@I3f3!f-#KtvE=eG_GJ|>w`4!~L;iD}JUNs>EtMMFA7dnL1&MbM`_f2x
zdX{^Le(nWNnEoQ0?<N4mKJ$oKFu94*M~Jf&=xjCkm0<rS(sHPGD@$%<X4Q&cx&7o@
zg|B@L>Zx_+*c}hUnH)5P#vYM(t~h-7!j4jPqm%u|^+Ffy!|ZlrTrBsx8MHhaSxxa0
zF6Tv5zgVj6)>X$pd&aOV$DsO4VzT2gbeGb;Q&S8OuHN)Gn1@<Z9IpX7&L3xnKylP1
z!F`n#TV6i)e%Er$yCAu@Zo{l$7v;=#uFg{-qL!r=V`^Anu7IQ5eHe<4FMgho`iYPf
znp(0y6CpPyXfHQEs(Yn6pG7>RT2nHH;SZFsjra%j9`KG{VajL3%G<?R771i*MJvu{
zkjfBMqh$fRx8*ZIPCzC<CTJ6j<EwqpsdzKSzI1Uf;oDSS%I;gNCecq%C1*Le8E75S
z>Ty1%kSa%N&)`-Qi#8^>(=B>fSFlwNyjwZ=u^y(AZVs805PjY~gAzN2-~l|pc?>vw
zmLJ>SDZvf|X!i>e2O;{>RIKC$laUY%h3;zRIL7jHg<k8z+I+}ch{G8jw1(`Jo@r=@
zmuco}h<@Q&m1nD)#LFOM>fMs?@2W*(3)fzfN{gmAmH?zb?XTTm&6Gmb2j?xxaN~HY
zIM0`#V-dYM_zZUzju(`CCe^C!Eu5B57p-%cymQ7->9~^!+fFlr{?;4ACrd)wLg?EH
zbLA+Z#6Od~p|srjBAV3-a#A6!nuHT)GT>c8Zc^{^EUyhV6Y`<mwMix^6JmG^_)iyz
z5e*SKydUiX6k6kwHV2E0musMv-@6_u%R*fq53z#7QOu^~R)JzE(a>9Yep_)hdVX6{
zwa?7-1jXIYVXT0~vBlFRJ2={DHA}=9qCXV1NzLy4Wsj_y#+;IbZ0Lv4R*wF0gk`#T
z#>eFj)H=q~cKR>+_ot4z0OVdAJL_qf6)keQPNlUm4i>2`-L8~G--EOwA8EJIbe*HF
zE68(@rjpwxC0ShjXS=)pPdh#-7gGk>LT16N1;Dn>&-h$b(PSO5I#*xXDd<&-hQ@cb
zv?8druC>_jZD!#Sw8}gj+AysOMeL^^d`b<d*FNB%3Q)PyuV6~!eV~@5c2U?=t0;NM
zEGe^|YJk*FM{iP#HP2lr8-xnO<h2z8VHw2M_?9Oy3)I5zf<gKpPaEt>lp1%5&&s}(
zH}nh1a(<iCGCqPEi*>l+kxVw*QG{cth&BWI?Z$EjzA(&9t#M51F2tQ;j?)otoAfBF
zjAVhr?~z`v`&GSOi+MR)n5mb5P_od(V<K~jI2vQcbspU{7lay!3XyPS<2veNm`)rb
z$d0~wQ(2mXzf{4-M|Y={OKvcFG*MD0RsZ}nWCmk)q-<Ue_Aop8(j+Sux?5E*eDHQe
z?jY+u8v{Xsa}kY@ttH`g8gB9ypa&!I40BQ^JmI;$EA2fGkJ*d3{p#rZWh)q$_&Q;I
z_It|(2aTDu-~ds<?o{M?ETe^`1Wjj7N!9Zg*r!ViGj9Fqyj<9>du~zF3w@1QBaix-
zxu8)ynZ`4U(k7!7fX-3={+Yj}-27U}2r@eK?fhfh(>Pp^jaUxT-H(s%K4p=@gGe6y
z%1LAFq4H3?r5|{S3^k(t)9eObet=QdCiQI2knq}*o+R){*dA%qPgL>u!naLvp=~P9
z^B$+ROyv>(wJHAn>b{n0b2~sa5ZmdBA@|H|YK(L<<c6C_Gt^HV$QA}iN+C=mi0)Sg
zr`8jSFH$ja3Sn4kb}ml+Y$-knh~`-!Oh4tCk<qG@Q&Pt+XLz?2*vJJBaAYb`-@BW9
zUkRfK(}!AOK7uc>zv&0b^x){7b`s7-+%Niw?@4v*!uLtxkQnz!3thDZ@hO0mPi|Mm
zMV3GCJ7#p_@R1XCIJSyGA-<(+-N28>hdfUDy}kxDI0NiZl}b55cO}bchjvBo6(-5+
z$mL#N8qUcmB!~KUG<GESK5+?~&+0C^FLF7`JPhTIru%w3@g?^;k7A?<?WDQWHa>#Q
zQQgg&fVo<4tyqHfLl-BizAKzpR&~&Cq33G4?@*&C_ziKag7s;f-21Mpx!)u=o(@Q4
z#<f$V4x&_=<EhWqkjF$2{sI_k?j<VqD`-QCDJw=tUay!qe3m1ya@3C@<z!?0X*#T2
z?uNV+bkE@3ph$03?q1w^)-PqN7qPs>4g~K$URA~W4jNUEuXIfG`~12)8@ar_vxSKD
z(9y6wpU`9)|9&k1k!m}5df51hbW+(ch%a1-H+qkZS=3Pc5RmPayTe*?<A$oWXB=Da
zo3b@6Yk1aU6IJfwk4EES>XMCZ_DSzM0h#<BZ|u>NPwhr*AN+Qk5($I#6`e(1cz0WX
zTe8<)FThMV<Fz^>rWMDt1(V9xc+<QLo<hg@?w-ufE>@J-0%?a9_j?ymwo|KWe+G)$
zqE{$or$s}eSCiV4lVKwX?b>rf7&-LqrWgB7fluj#M1lp3pTQT^P@}x%X1BfU+)Gqx
zxgB+Pq276MgYm#&<KwHtooLTmHHnOb^uqQgVif2xsU32_fZcd>w8uK~Jrmgx;0IfN
zYo?MYqwVVyRWZ!9h70Z5w8&j$dAkLZoZ~8&I}4P^IR%SZcKP=PfkLp2-hsuWC8Na5
z*-cYi&op03DEM*bnBIkyS*pNw{2o;O4X-W5E_c-ZL1cuSa6mn0YhrW^YD~bKqU2oI
zU>TOm6H$#8lw+PZ23B_Mb!PxN-n}WRMf013)!G{|yF!mtE-wG;fo4AqQoYQ4M}g?k
zCI4~Bly7HtJ_VF(f(z`{XbYAMRxqx$-_T*u7b4slI4V2`E$)od*RwLNNtxy|1-dKA
zWtS!9*gWBN-g4~_lY}6O)f8#z5jbk0cIf$`e5RA1#HcaZKyd@&LkzC<8WvEVl>&|-
zILAog-*p7;p=~qY23RVG>*(Fc*t{y|v`I6Ecf;aTgJLLYb=9z*^&C`b#&Vm}w~ZK`
zfWOr~HoyHAr5xi<rIGK$Cp(aa%Pd|~ZBpJ<Xqzmpl&A$F&g|gor~f<=Z|D#bR{`Yu
zrQAH=>M%q~d@GLsX&0EYKM|qF9*q&#Beq-_WdC(d6CG^n*;Qcb<Ok?cwqK`Njt!SR
zx|K<<mvXJ9UL%G1vlnK#wt=Uje5+#RYWaV|N}`Vl@MW=~g_HHT?SqD&t``txz+(`P
zY#*uj0ft?USU$**%zDya%67?SglVY1iD}KH-m3%-=~(4>7IUqa*xviKPcPg*mO$Fs
z23ph?-h-L68Xs45$)rQ0E#*d9G0EvDXvNDKc0Xq;PZE-ZA2lAe3=80;smt(!+Z)uf
z0}v2B!{4hTHluDx`_fBM2FkVtIbv-K&{`FZ{uuz*@+&S-D|9R|nd5(|0u5P15t>8Z
zh#d?_z|uqsVm!zSy?_MRaerS^dlxJ#=Np8BteGIJ_fCj3Sxt+@tM%|39Dea&Z~n}6
zi2dEQ`!!>i#VNvdnhW9S!5!SgV%4l+nM-`q0GjqAk-=hw-ehPlohV=?A<`JFx?Keq
zSY9ET!15{|WQT~H90CZO_p5@g$dHWs8(g@~*a%OijYm0k2Wr%+Gmq}ey}oKE#bn7@
zBbrM4Bz<oSkG!aqYjXv3Z7pM2%r6ySA%?I*HBMn4%l{n4<K9vp9Cnw2t9sErEBu8z
zm;vP%fTpws6T@PGaifZ@dcY~ZV}qeONOgqZ$)^&S%DHF^lni;a)iN?(oKUAO;Ba~2
z%&j>xYYAwcHui+t!2%Rt%vvF5DkMazPDTovTMB+Hpd^i@sy@mL)k&ST=~Vhf-8&i0
z>MBZse-WrkX}L1e%L~S#BWfD@=B;P{?MD`ObW2lNw$&QN2HAK^{ltM6stj`vO?BWe
zFTk_zR(Ofs^#(0x-#6|{f1{lkkmWad0Z^q-xPAOtcMoC@QH=DUmt~rLp~0Nd98Qqh
z<7%g|IZ|`m8^0c*WJwBjxQCS8dslW8OMZY*HnhE}v-!_Sgb|n?hA@SuQC<`PL6l=p
zqH?2Q8}vb$sUI!`(dbQPuom$~VLMZeYU)}}1U<s){`{Y-<IhD3=n#~!PASZ8A*f#=
zGh~@&-9h#AfWIM~uCxm+b(zJn0=O!syT<3I(NZvN=*LGnOq?X@bP7^vYz#_;b+kWD
zBu1V=j2aZO*gwc%v0}TPO5e?*I&SuP@*-j9>GM)zSLrK@6k=2A@l!G;^wO&+qLZ}K
zPF5Xbi?HUVn66fws!*e^22^4?u~0;vrdK%Y?4e-qerlQ!I@@UH8??Pv%`aUdF5(l<
z{msD^?*!|0nTixs1IMXF{t7Tgq-V94?2F}v$O=-g6F<#+%ET<X#DKSsy|HMScbI^V
zArIqUr#j{BQ^7!DS)UhqA8{FudF^HJ3(rHU4o&^EwML&NYbRexKdJI9#8~rJvg2op
zN#%3x$zap-mpXIk%l&)X*4VSPAvg$7;{Kaem)5uNt)2z9#&JBREYHq*t44s`D&^aW
zPc!q|=bh)>EV!<IN|OG;NS<%XlJH|lN1;jRu%WK*s{rT9JT+6D8eCq<Wwa7I*yY&5
z^PE7>TCNSR+T{uZra9mMsd_=1dtLQhFb8sz(wrOjW%A&hvlSdPXw@|Av>Tl8dlLPI
zX&h7vyl0e~X*UX<mdI;*yKf|<)GV}7*AA;eujQ6DS|?;Fe|Fy?io7sDQFtglvvFv5
z??YePslKBXnuSR<hGL}WMz2qlI|Jzh0l|tR<HzDt`hVk^|0_NVXs!V`PHZGS4Fj`S
zfvkWC0|CPX*1>s<QnkVPjf#aoL+$_bJ^}!wl?o2?#})sXnEV|#{rB~`M`(}hM|?T$
z07?EK+BE=Pbs69~Q`!7DCWG>l3Cna$?%%`W?@;e8J#B}HD#As>|HIX3QQn=o{hGB3
z2P{Uo0pgx=00yG_eF-Q(>VG1+zdn)F1c3RR=7&W6scruEZ~yDVr)?3ylgCat4eS`e
zt_X;2^2DcaomSvn(&vH3q9e=y`)g{@065fLvr@8u?-JsFf3N@jH#Knpee>({GW`F(
z`v3BA|2)2b{-!h@xX1<1OPqhQzW@FI{`05*bDh@;z)?O$H}zi_&$g!Ewu@<#qjV!c
zPq3A(Lb`sQA5M*hFLo@Y_rC90=Obly&EpVGOhh=dM$uT+9drAC!HOZ$X`ljk?sHzT
z`@f$vaEeeEaN_L%I%wO=`hYQ2Cx9HByAiqoF3b4`=Ywx@GoA|S6S?{s+?kYfUO0R$
zO!NQYEebP1FKV&@C*l>r(i0mvSiNU}jOg|69EnjeBsj;d0~mKBeiLJqNCWs+n!+he
z4aG^n2&|pOXgz-z*RxpuZ%@w}ANbgFow_{#tc}vpHhAXjcV0jfIwUw>rYkma{~ka-
zylVi`iw=M+MF{6#y*ue9Rd?4t2P7xJpoDWLT!KSl?6FccFe@-^NE8R!Ej>Q&#7CbS
z1(+)$DLd>)hfeX{Spa1Wpgj^VayIGes1eKSuDwZeU5#Lf!vLULCc@{V&pQEUGz|uw
zv%@c{DGHB?;cToh0HwkNARK=mPXocANf^*>2Q2PSFx5R|J~D4zyTd)KL!vnwV7I@3
z8zed?wCuA2u=<Sk02i*#6axqH&0PP7cR&Cf2H1{6f^{<20$EQ#Cvj8#|6iAXAKl+C
zeHtT*>{TY?&kOy3ZeuV3+(f0*z4dnS9niwW^KvOWdNtGw*x8Z7SzY0PjX8tZ!$~~;
zD<jt*dRy9&nd~UEhh;!`y3$3Y?lQhySM&hqan=D?nZ<q@`-pSShHc1N%3#O2)KcK&
z8DRhR`Pgr)`}2DUbNT1#jB+Bly(d6Gj7$1M8vNXsVYO`AAodvnbL<Fso!%p{9hm?Q
zQ7bTpcpNyrt`ktQ=soNAyvc^s)K-92`dQVGcJ9sOaha9)!0jbKO!}@?2uGE5!mS5e
z0LWZJ=qQuKNo3rsKX>MB4yUYz@!#yL|JqOfXJPdq0k2IDge)23pF7ixVF$3SRsefG
zT=oR@_kfTI@HAtFGcQS296zwNlND1%zb6m}FoIkm(|~@(EUn`=;|^fzg9qfhow9@v
zEoa$BC}p?OrIdn`ABj5sN|v2fry~J!(-vZCB@lqPcnmw|OdL%#yb6$-*w8$VOytT2
zD3fY3q$`f*+Y1nygFApfiE-pL-Po}|3+>d-IVJwSjV^r!$4Lr5T&;D&(S`_!1Rauv
zZFbgXVgOgnVxrLOpD*z5KGUWm{}awU0`pBp8=phcvs5=!|KolT0sL=&B7M6THGggF
zgV-2N8yt=&sK-Dv=Esl%1@_&U%)KV{EknSUVPLfa)|?OAs<Vp??;(*PYA>>_=_`^5
zdWBQMG5mkL9R?;6AN^sfeXQ_#mIbVIwO`(VglYCM`O|)&j3?3rbo<@LmjS!VhDZ5C
z|9;jy9u#V?Z91kKuD>a=*6d44F%d(!G2t`&8UlAK`2HAh(qP9L^<~{>&Fy9$nV$f4
z<#MtNXx5q~e;!Nfke6&C5?do5A-mt;@mhluvaSHMA#qv(d<B^$SC4%dm-oB_dbrpX
z0CKYv@T;N36Io9fd&J+a{srH*_Bi9PvvhBA2;j7OyqLFJ^x>wKx~u{cKNXvX+mDuy
z*)z$e&y|x7fnAgqsP|a|=a$eDW3ShL-%6=37+^xIS8hB#?*NOjFmxIpZpwFUh@s0N
z7~cV09PyGeX(ald03ZPiWt;!~aM>CTp!H_eT;>3F>TB&Lq4aq6l_U6a+1dh#K2-*&
z^uK|9Ne3R}YNaUqJ2oB&y2nS^w!z)3Inb-Fr0%#p`;=d2TmL?A^vBM<9<*A}qQiu9
zXFCm+Z*+feP(Q`!k37`}D7p&}gV{Gyi#`dL(OfMv915$vMA$C?&$?=}IyiFawjYi`
zbuvdk0$%n5PEz<Ikg9)F`W$#(^)w~^`&ZBGS;_!HsjvI*sUdVM{ddnb)f(P4M|#6H
zQcWb9R&X6LQce6k;R0MzlGoPx$xnY5`R3Q%UR3x~;eodEDhD_lLx5_3^B%dHM$7TT
z@B8K77YhT%rJ4B2YotaliH&u4-CTYv4&I+eY|HD5*MxLP9!k9d0Iks<aHv1?uwGh0
ztbP4Pii3Uh&FSY?&8(S#D?G>*%QH!)HV&BNSZ3TnaY~8<c(v_&bW<xZWLL61uS|yq
z8;>1jlh-rDnPo(O*tBKILu?k;a5$p3%}*Yi-kYZv(Sp$i&KRG$d{(eKfx#TDG<@Rt
z-1~j2W=d;pxb%IIcNfkh{$4JB4Yq$5%~6X1wH$hq7S@2T@`_@k5$NEnDJ*@6hwClG
z08;Bw|0CBF!BnL8;L#uhoemnTMK0eR<)z_7N#>Vu=i}7IWbK^0lyQD-zlx^Q7e`;*
zfrR8ZAKw?Y=d8J;hXJRNb^_%nrPQ&$B1f;<$P+0Y&yV8?etR&LG7__~V}M(13^ylf
zrq*&K1qjFi24**laG2$ec4+G)U5_lVRev0yG)Fn2XD~jAdgVc*KF>5d!l&y6Afo1p
zUOm>IAc{0Xh=+eS&Tta|%yv=p9pdit_nu%Wh?H;_50vC9fePwvmaXMht&scq55Qfe
zeB=xe${xZo(KmNdei^X@*zA>3N6PhVG0S5Nqsx4)`K?xm!<H;Jcc#c^m_4X>`HnL-
zCD~{#{e}l{9C6#0N5FOW0PrC4bEv+Dw*aEQTVe5{JplcOr|vbLG;U+tI@n8g%S)XV
zr%|M4KH>TM{xTe(y0Y8><W(04Wdri&JRF9H-J3P*J`XDXS{WE|JnY`$oY#`PH5$n6
zZVn}1bL#nNTUI4M1EiSQid}pbfr7q3WxRp&^M3l4#AQ7v^1z`95GEl~$>>W}R0B6M
z{IRZH{yt&rlBH*s$7-a*M$R>#*ShDk!4QeE7ov_@zl45-7lF`OQr+T&p1Ry_PmE<x
z7>`{8jJ{1|-%&saLt}m9!4LM4Y8cb35*)4Y71F2H_xTK-?W$wUM5FpE5Rr0*$GMa#
zQA+>p7xpN6O^ZOu&5#^!uZ-rV^0m63HS`41CUc@q0^AyZ5&eWvH<CA+ZTWRwZ}Fa@
z4{NqSF&578rWnuKktCyp7Rn(SAqWWRp_oGb-6a4~*K{~bMC}D@N6=QB#LYHvijsOP
z(KS_G-2=#Hkvw9vaIqKeiWusCr%BO}QeRxsFh{S6OhZCN{k-IWnux{`puD;8v58<%
zk!=LR4mU>>uh>~rT}KXThre?s>_5G2E>D5YlTDoh8=s&FXiZ&;kR)u-W=Zoy%B6~f
z^3Vcp4j1ZKY#wfb{cV}_Rs`7W{nng-LOZq5$h&b_30lFGN2kEfd!C&LNBFW?ewy~l
zY>3rD>E?K`W#_R*vAkXPoB-~CoDG)t?MHVjg}W}Wed`#Nl_>QGqx8jrl+xirzf%Xw
z$q`_Qq4xUid=Zy@;AwD;Qo1&mUI3&_KkLPzwT(AmEp0>1`l?oXU3!y1i>eh6<i3a$
z7n?53%3Gjggp8s*A8+~@gblKJlGXmp-qAaAZRk5R{}V_$3$DpsOIz&~VCnCL(*<|N
z+e~1s?39(I3gYIyywRAGj@(Ot-*pGD81^p2D9BGkBm%cI$YdI%$zP<)j9W-!rYlKG
zft)Jlbkl|fvC>J|oCrQ)Oj@mT9k5W8LSY2Ec(TkSTtJMXZcei!QZ92<fO`PUuXCR1
zO9f4$CZ@S)>7FYkENq8czRgn?0jShjZBaglq}#~)lk$ffnA5Jk$UdpvK-kBcx!J&*
z<b{TaK(74zIYo&&q?~)nbOBzYY$%B>;ipAUK!j=BV69U{2PRyS3DKh`_{VT1(A9YM
zNCBOaQdZk)r@}S3AM}~51PtFS{P8P8p2exZ6UZ1Iq~)s~p$=7{#YrW)HF?<9tX~?>
zQ?F3sG$B6s!d4b+1xlxoJwbTyWn4AJ;z;erm1KO3tFkJ(irSanSMW@o<ko5==fE6x
zw;5*FLjJp;R|;K*25$_1ybTCBjiv)B0Yaazl3QUQE)zk<I3s;eZcJ4e-6_vyFNAcN
zgP|=^es?%gfqF-_!tRD?;|eGrVuJg`K3tPJWAyOaqn~RGj?ikT=#&~TX@T0e7mxhD
zJCeu=b@@csGlMt<P5iDpY20j+cdkEOh28}}2Cb9ad<708_DaleCg5PIQ_Ooxiq-_%
z?L=!c0w>xA=@~X`yH!*6R=@)sO^}Eb^?1kXas0_muf(sa@tO}bhtDTDu<cXOdC@7W
zIQx5k*18m1cv84zS3h@up-(Vs|6FP~g`%4Dlo}L8JX(+UnI>s2{D=LPYTrYHuO91{
z^MT?+b>Cg>mMUB-0sl$mISCN&l;RwQD^Q4jB!0*7ePQ*?y9Z@(Y!;*6CyR&F_r0-u
z_UZUb&%&##uR#_>YZ-P=T-Z_mMD;`PY#i#rD)>(b^^HKPFHI4Pxliv-9B3qBt;yzh
z$yPDvjY_o-nqL9YSkV#Lb4>J4h`e~qV3|IMKz~8gw_1p>LjBZBi7Q6zy98d^7g8GS
z9b}?F@QtM(!M5*WWhLl2h@R{Mjzo&j-cP{Qz^W{b>&TRQN9?QTY!;X;cUFx2f|>D>
z8B5~^Ix})8(T`y9)KhV)m-b~<TF^y9*~iv~YpLsg2is(<BOr${tnWrbwvjoJ^STo3
z_aSlNY&w>C<zglhV9WkUiX@c3QzRtD4t#?2o(ORjOLUE{YW?}54*^Xf1q_d!S9&Ul
zU4bf3BA6-Vmew_ahJ>XWaf8Difsz(6ahM*0TgX#Z{a9a?zNO(hz#Ih;7dw=BpsF0B
zJQ=+(mTy?N6Eh`&{1d)jkxb%(%TI%Z#2K3xw~*%iu4li0%2J2|iLF>R^JJ;QHG)m(
zjr1JmB-N6%$butbrIH9@Uq~q9oRGQD99}eYf(cN15*Z&38a@M%b>-{;?kj@$lrgOs
zAu`7865yQk8|~gIKLM-+6qxe!i3{<H-LQsXU*M%zwawtsD0XGUCMQJoMh5koX#3jP
zp1yHP2EL?f!VUuaw3{FDz=jcJS90IFMs;-Dg15@)`UZ%4_WD`h<3&G7C}3tF?y^9+
zJzp&ikzB_Stxb#v8U$8)0I9J#8~X0K=ZSCGtdp95Gay?iR39oeIUS<M!8$NSIw0{(
z!v}Vau2(w7MK%0%d^X<fEv?&fT&JjalRZ0T<ZMz`MgHIu5gO~>bAw6Jvgh>QXRoC4
z@Xf`7e`xja##|&Ez?|7Xj;a@HC`=U!SGVeY>iZy9Iz9OZ9}m2dWd10$%2MoRJgEUH
zRuhVDS$#es58+i$15fzZFKT#TT$>Ns07S5bCmuI;P-wHLEsOkAG!ut5x3W$h+@fNl
zDv8NBXfS}ZhI-?<i5!DF>dKw-#<1^>@Dc5l>wsdgPq?a)-9vkjSZ3f1bNxZ>d>CL_
zTe}a<-dnzz`tD=rwdA?=@k4CdE!dubcZgdT8JHj%Lo|CXI4={2vQ@jec+EZ$Ms@~i
zP5TV=rxVnAAPG04t@+Ab254p~E-$txM%kQvKkp&i$9mE@n{JLxn(X@MgR@lT&2t8O
z9&dJU`=u4+7trCk#OsY(mFp6xX%{%>aX?qfm<c2^-wq(15-OJxR74}82sY^@@94~N
z;;)NY^lfsS6R(kseo5rq)Id33Peym;M_p%he&7G`)FKD-gna)tROz5|xAv^q+<#Gm
z-9YgCd4gFY-+3m1jXnc7XMi%zuQJ9ZGq$APY0(FGM6rBA@{)?VF@uGi5QPm8Z_tp>
ztkbI(w%Qwst6!qvJ47<t{a=1LQrH+XdH+Jzfd}*)C1#K~)<=WILa;sFa*Q>@;*)h@
zyR~R|<0=o*YyL+@HQAeE7)E1dIIfG1r-f<~is1+2v9<50?BCPG(oTNl*RGV)7L(Hs
zl5;j#aE&Ko(3Uj_i719eTiUZ{KuY$=-&HLJjA6KsDXNVThZtSX3-fef_^B7L96&gJ
zDKHet8LS`qM2USkw0oTi3s0PfBqBjrstZ_@K+9`w5wFL}Smm_iKq-YI#EV8Fg+%?G
z81Hv#j;D!>#uNdeSOyF|@dN2@$N~Ljjg=&dgM36(u}5qz5ic<3kmlaUWN$GE#0W8l
z#HwRt_H@h{AFA0flf&ziBjshHif1ldQa&n7M?Y4BV+`cAg*`>9yTfwGV#FQ~aj5ZM
z7I1QTQ8U-QyF2Oi&QJ~1dH4NF%Sc4JtZ+HmSDYp<+XNW0&SBJ^xm-{u-JOgiR(O~u
z*@Wl|4VZ9>l^o7NWNNw_t)oASA0($*m()&(Wzcg!iLBzL)^_kVW-(yMaux3sY71ys
z#*Z#fWzAA?r+C%}&|)(NFd4Q|HrpaP$6)zmF!iDlbH|w1--j$b3h@HIaU5F~oSGTv
z#TQHc^6OV@O>FUGMAL3jFfjj={SCAhSREh#x>yya(x2Y4#NKfJH2UE$w0o0)a+yR6
zALMOK_fC$)G0?k<sjXYy88};$l$c2_>tkFhLufRVt{s)6*x~jz_#@8-3V1yjY2EeK
zD6Q7mn`44xa?sVu358j-Cp5wITj|(OoW${-?D63VXld|&_(`_aa$U&d$QNY7PoFiS
zerocw#gprp$A4_$nRq;2uxEz;j(X}{>UBEdX9yt`gk!;daUdr^^z6<2J3@P+)R81R
z=a>A-*R#DZ;?oWCnfgX_pj<PXC^XJEqA9~Iq7J>aV^h7n_Ct+l*8-RQV~WPt17CBd
zzY;x-tBo%M@26I@b~(!S)ZOfUQJ7DdersHv$Y2$7w|2wMEvH7V#CWSXQf03Zv-zXV
z;-z1co$R3XI>`vr&1AKd!MZ`0jsO{^&=Z!*-Mz0V=+}I2bolpz#stN6)XXxuKFH7S
zli4#Qr-mlp)1T$r<@tD`-3xiioTC66?5kgi9c%x&SRZ4ZnXF^25zQmaGq{gC+usAZ
zvFnK7!)2LOg7GfbJ@dWD$m7uB{@}VVZ)PX!ikbZz82>)IY1_&v{~qK7N>E@N!qVpf
zfp&=bIJ^e>0WU+4(n9r<Gwnk`*R-na#|7NxiTTV@y0U$ijgj6CncZAWDjOfyuhI3_
zxl{3xBkis)KW8y@&|p8S?F49{l*B+<qJ>#VLj3EVxMH)VOR4y$e(`qs^Q$kJ@gYeE
zZNEZM=MxPSJ)}->Jd=D<LqY51z#;8~aXntfcCMo<eFM$jG;ST}FJSlYG(@LyrXd{a
zu;L#;E2^MnYnyY+Ns&lu=3xyCbKmQa+`klIbY%9Tn}Ga$=09mWu1f+ek7{8By1Vpl
z5fc|~+W@1GWuFCD2QXIK7*k?*irA5>Z^mxl2@YQP4O5W#;Qy`aY{#L%JVXu?YPpI{
zEO$X6C%pawiZTN&HA3rRb_IeB5^uW0Uj_u>K62^MD{y1OfQ0zi^lSiFm-p>f{;)^h
z{#rq@4Fzw~sY1(HDn_nl=yW7R&PHN8FSyD0!oa66-c%)caDWT^$V$7KF;gM}#j?Jw
zmmKH@TTaV|?K}Skl*0%-kqwZDz%{UA4HV9yG2=B6ybPBR;)yO_OONDHdTgcwom+T5
zIlwqlmggxME0M>X*K7Km-nkIclI4WFHUl&|Dz{x*<;gc)aSI3@t~k#=L5@n|GYkgn
z0J{m04V_Wu<nkxRJ5_Ww0B~d@&(zk`%|1YC+0Gs)FBl-xkr@@>H?ZI-@7%tZxk#(U
zi$*Pb#rJ(f!fN<?V)7=jdMhquP3p=PD#qDz^=4Xql8&EIZ2rN*&rJ8;_s~&y!b`$x
zNyo6bdWB2<<$<S12NysqR)sw<-f%|+*7Jf8oT1c1Ns3l2Vmfk@vJx}v;bnkUfYqGD
zyaxhwtqB~d;N>mi4b|ztYQ}jYDC^_XUfTb@rE3ug<(e_2o19M+lyX%W7pYLom)>1Z
zagkxt{VXO*ZGjJXv?{PIf60hHmo~hb?W*YwwQGo<*IY0lu8&_c(w;<xR_g4AF?0t<
zV;TCdYK^$v6c5Dy)?%!S>r&<6{MEk@Y%XaiL{n$bZoyuTadkkdNCA*aLS#s*gMl^F
zKB9;z4)FUn8H(m%IEQHD(`a=PZE_!20u}%h0H=3sX0r7y8`kEairczIZftCp-F&~j
zP2^O);+pr<WdMQ1Vn@gxcH|K4^hrmTx`+$k3Hh=DfT@Y|9ID@<a+gr0)Dw=hh<={e
zUi!+$6w&F{)gWqWs7%h&I}sg?Z&<6L9y6oO%;b>@XsGUl!j^smlQchZ*2Yucw`GU3
zOau|~cry$bwo?gnHpG}utTl9zaNU7Cl0&NnU;nx*oi1sNEk0j0#&K_k@%usB*}v6`
z@)_RFAZh;6-8li?@|Io_>1O}#4V6}~xve_G;nvXj+g1gp(<NxmMOh`_ZPCD13L+<)
z{QZ9@dkdhbyS|TG5EMiO1Q(D7=>?WXTImLnSm|0pV1bM7?lO>+lvD&{k?vHwq>=9K
zhWD)3b>Gi>-|sWe%saEgEX?@dzyCew{QC1fAjgmOa*5@(Q9#Uv<CIrd%u1HMM`9vd
z?4$z~U+WsVl>3}Il8=qF-#b{YIAO)BSWgyqK^9~Cr`pSa<$rCXH#eJfpQPCw*}k{E
zqW5<);ws|Cr5~Kh<ro>cZATjuA*8o75CXI`IN3j(%M3e^)I|QmE&J9yMY7jfi{`yI
z7Sy=;n8qbde#o%7b-o?KbKQMr4F<qO{o!p#To1f(8zG-X>(HVjGgmGEt+!`uX0S=&
zjTTLjxxewzp+OL*ywu#!Uf3({EpZhUm(@w5TyTmNf@^GfEd|e6!h40horM&GCE<{n
z+K`)s<xWM6^lj5&V!Y+$NJ2ZN*IIs)n-*gCm%_?^4lW<Bz=H2Z@+6$!p9i@9j|jKP
zU4aO{M(7)SchB){9u*ALCAl)oAt%ZTp&ak$BX*6E$-UnD$Z>|Vo0G5s1G~;pv8X49
z=*-`^%ukWsM}8gpbBwSO8$NyVqxfCZM~ZV*&ZDPFVpr<xSo`E=m=2U}g{MZtRHE!%
zwf*ehDFGxmr0tRw>Gam=hoYnOrQX?Iw2iN`o92)TnHBllzWp%od}dO3A3$Z&_Dorm
zi1MsFLYk8$e+nshFwTm<z;&7{{>;UayyA?}w(1-<VD($0=Dyn10ZWAjBh4$z2~O8W
zJb(BxUYBxblhS50n7=navGXDwn|?vvb^&q~Bljyd+79`-U05YGPFLa)3n4ZqcuVqa
z=b7HUkSX#U5!a&oamtZa;RA(Mf7b;1C_pa*Ur4se-{LwLw0+E&nz%56+d8C11EEdc
zY7)4?nOoylMU)2?+2@-3Z^ig6{L;d5;dIQ8_lj@Wfpk^X1A$iUqfShba7D=YgkGLX
zpg5!z-a^JW)E%-(eAm`<o?O(H`TIF`sCx*-Uvl}G9%^m`E@X*$NU@St1R?gb4Ro}7
z$0K%oiF@$j_EHZ+f}#{tPb?wb+M8-XFJWhtP$JJt2)Tj9l|S7i35i#a=B(b}Hcap(
zsM@MpkKTJd{YsX$FDrDd1DCkGlpRvj)e+b!v4(+{2Hh&AX~t9~<PyxzP)U)qU+@9u
zD5IrRztbS?Hlbhl13g#-RUJ*T;X>d}$Ry7?+v$R1Pv^MBvfr)yIjE?d1(Y4;&qVhd
z-+H0hC+ufGK?frCLLxlmo^*r=OZ3Jzv(p5$JP;kakvyYiIL3pjskwA-ekFQxE-D;$
z+|?09)T={UKy>R+{Icz)`7~v=Qu3R`AyEh^nUxwfoQ7X9z0S3<!nZCGTVVm>^-`?w
zy3e+Fbdq!D9^_>Nj*s%sB=6~wKQut#v{PX?Bt9C&YH|I*h3nw9^yy9g7!GFt9LQJc
zu1==58vHOk2zcX5bNaN3YyH<Os8AkDo6Y$?ZbK1B(QG~Jmfi++R-@reFN3>{_vwhq
zwLz*zZQAb858pF=r#4m>#Z290t|xR@XAP3=_iD5uKR}vni?qu|-$;_~G4qdkAkgL&
zBGsHDgF~Jb<0iZud6+nn>b8_<t{$`KB=;<=K=uiH-N~HjYL+J`2$M{|G1Kz!*mA(w
zyXoN4h9`~HlNq1%*b8=@X~5mq^MEI!PIzi0&vd0N!ABA^Q?ze5!E}(L%j-)o34kNh
zyrTp?%qqSL<oC&%C8@HMXk`-*i;c91@lc(kF<hj*51#90t7y}HQ@cWEqsu*(KCX~<
znOn}5pX*b6S9v48uQfZ(79Q_La!6Q`R{0s~p7qTtmfAsd90!Zj!B3QT7Jq<!L7RU-
zU7*CBUhYS!I*#BEb$$X!5a2tSjsf@_1rn6}d_`o%Er*vdlZ&<C<{vV(g-2*ED7_B)
z8%16SR=%n9ho|q~E9fGU3xU7b5J?la_##5{GebktS!4S=$bPkX<!1L>MQ;4Pud(b{
zj&f^Z1$a4^|LogK*drNqGMRRSR5JZITbYpg)wor?b3f<hr%5YTYUydS)Se~enSIAH
z=$msHQ$Mh-w`z{Py<Hx*2%AsxcI*3RRao4>NFHMTpK`BhsDvg-e3{V(2Y=*a-WP?S
zNL*>5kd=}8bx-Y=XO``~zi6$(p9}cdN~InsERX!a=y$fQTGyLgyt^M#dC4E}7Eh!$
zcCa`DDP~Pl?K2pgep5Wmiiq4KuIeV`3o6eWBpeliVDjzsM|+4amYxu~dZ#||^SX+?
z*6dCotoF!em+4wzxhDF2xat&!=M()6-n){OyU_q;JYbJ0IS#Hw3V#<x(l_1PKk4g@
zoTKBBekrRmqtyk)m!Y?lMj1V@p=BD+j2<!8NRXwkm)02eFo;-?Quw3X^5Z&sZr~V*
z=!BMy5{10I2UvHh%gdk*uFHpeZeXM!??A8DfZ6KP*sCPL)I$Cw<!Tm*PwXqwsI$IG
z>xkig2vR{`-HED;GV*!(TiG820o`AJ!MyN$)ofpxG)S`}SFO47w2uDStJG9~_G%f7
z%7%;k`E3@5Gfusoaewm`eL9=*0)H8LE@H;90ugDHO8@+8u4@>_Ri019jgF8UGi5P@
zzo~hF5`0)I_1297Nv)?RQMR7V`i+lpnYlxFN-C>uzPHZ&(D}|xD&I4uy~p9ju}2;M
ztzyPJGE!LP4A#~~ebyI}X>$@Glf)L@aZXZa8CY$Rm1+@4m!@i4Hii(}X7yxoAk+@e
zqQ4~DWlH*f*+o<vk~PhbErl9dw0tGjab6Lj+8C<23*_35))AG$dm)`_Z71)Ef85?T
zYm;8F>Ttx5q_zoA;Wd%8JUZ*X-C^*kHQN?Wv0C-kFO%?V&Sa~}*;R3USw7>V-HyYX
zGQLqVlMYjUHdK*a?NK^wg%X3KVO(%A(zl(<Nel7c=uHLnLLc0A`h{b<1yC#1*{*#{
zc2f!C!GsYtbsE=Suuw`X+k?MP*9CevNv@flntpf|u5)4P)HBly$)E_vj}sY@bEyx>
zBolLdp&0nZTO`c*1EJR6qqzH_)u+Pf!kW}?vV5MT#aeImie#0#psk-!Q5jSRS(M&V
z{gNBqa~X4_kVcEcQdgLaLAcCt<Pyf?q{ZHfR*0^vmt+qYXZn4TApXt(M^Y~SNAfi*
zgZd+F`S>^4IzChq+x+4etFCF~bLg+hW21NG);KR#A9tgAe{l*csm~Yd`Z2o2pIYS&
zWO<|}C}vEF#*^NL>qC|YM;6>PPUS40jg*6q2U&Vtk9w)Jri92{_%E0euyDFM_k|1d
zlf-M8{toq;X&X2g|LsKo3H20wjZ4fSg?f$Cs!x^Ty0fa^-=xj>{_r#KY;$`s#T6VU
z^w+~jiKiW}wP->8{k8&jtP}HtuGeZTSj-CQ)|sUjJ{u{wD;`V#3dVc!7x#^KpRnZL
zkzmEAE50kndOFAI_Vei*8b!4dtfy}%;vbWRVX`&RB{C?yiNwWzq4k1zO?>+_KO;An
z>3Ax!`ux}aq_8OE=P$p*2M-UQNaI&kzS%1-&s%S6%*38cqH8oPKb^_RN~>_o(;+4x
z-QAheJL4x7b)w$_1KOd)Zg~?Ob0jvE>Tbkb7EaXom2f!|zs#<_XkW&u@?>xSGhbhu
zjnQvTB;8#D@`FKdkZV#MvR5d|Mr!;FmURTC(wQrj$$np1Ud;CQnIh_Oi!zE0n=6Z9
zL%ql5!w+g~c{-o;LtUk)D9VY(^Cu8wbV*LBa#J~79d}w<F)QGlP__<?n0DxM8PxD#
zX)itPLa*L3@g2gdMEs#nGhy>_n>6`LNr|T^dF6JpLhzabrV}1EMUK1tR_TW&3wS@F
z-UDgG?A9NCjVusJOyp9tqw1EoGq%wR>#%PRVC&9r?dhmj+6fu@bCX+ZFJ_arSId`x
zVk>dCkRdNgh^E6W6x|A4rHlj3QaosDDG_IqC^gY?8uyl`V<<?egq_Onn8aC|iV_lI
ziW`g@5qcM6x5=Ue7n@*9do>9!IB78Fu>EBGNK*1hL<TMSVj0e68164ULVPRyrF*;~
zB1JnyhC;sB-@`R;>Ldx}Oek;Wxuys#^RHNlFuX8XXcb)wIQ3=jn0$1OVj5Rr%Sd@*
zq3%g98+~x_4zV-_<T*_HEtQYIYcRPEH(lnnM`1n_Xo<N=wu0x{7GyV2ps0!b^?4=+
z&%kq6np1bJ1m1r%9eLm;q6=f&(e!$?cq{SV*(+}ix>6aKh1QQu_~yZKOxLf|d*N?l
zgtR!-a*fP?-_bjXoVn|lU+EfyCea+<L|_;1?D?_cZz(5_0lh)p)RE&s(l)#oh@ZW!
z9}z$KlWxDSGF6}*?dl0LziXA7@~XrTHTHViaG11h+efODM9=OiFIy5v!J-?9DYL`-
z?0AE!)qPZwQt7u(1)f>DdlVA|$1<pk)%w*{)k@hie8uImy|LGB1S1l~)~^f(6}%mz
zanM5b63bYRPu4%C6cS?)5`dQ<217%P1Qn}uiCu@JOdxXUQ6}<{TY7XhDHYQAsZshf
z0g!8%g6`}WN0r-z-RzhB)Nu2&GgX{G_?P}~r8tA(4~vLZ$xa19QPiuyZIYI_F58Mj
zIJ45_vj}gG_=<|<tnp)ft?)GleDb4lD^3#BL)@#5AGd8=G;(mBtzk=rK>yk;GpqHL
zW}qtEG5S8$XCo#@Z5x$*&RAZizJ0ZX+1?rHILsE~K2lba#=T&p=1Lc5MO+MPnphlq
zkH~o68i72+g$U~qd?~m4>7P-Vxx0rnJ55g(Qjd`r#t()2wm!ndXO~y3_0_q}^`<hP
zMI?K~zuw#8lN;=3M_Q#>SUaVQ-a?}|9`|Oj^BVsq9$rf9!(sbJ*gdV!<!W0u9ogxg
z{eEo-nnf<UouO)XgBv^pT4Cci;}8*T_<Md0zSv~YQGEc@<b$sx#MFq!+J(p22a#t=
zOJt$7`2mNRUR-zqhgd#37-rvYatT5b_Waffk9m+g+K*4zz7+lZO`gHcdGjtA_haPu
zUt0Z>cnEweT`VaHEUEN+n32azJtM7*2lF_eR|m;ZNkIp?MC+bi>_YldGU-j8Vdr!A
z(k(<R;~>%HJS0n=Q11ObQplYM%zE=~1}je>0vB=rcvEaiOe8rGC<y9J2K42rOWcqU
zBgbCCh*cHtg>^jeGSW=*px0VkZm?xFD0?z3A#vk|#0`0=r!OQZo<Gi0w#=>DZ=!MQ
zxZYYae`fG*K0Cs_ULY(a{t0Lf<ny_&Zd;hV)&5G@p>M;<GMU7;%PSx6_l|YnYNy2J
zP}pNw^J|MwzE0{6@e?bth0*HwbsC#74eB^8WOr^0r2eE)TL>ce4(V&pdKLOrXDV~6
z&v1GDRzb0xjCi<RkwLlvo!Y1A-rG`*I(d?#I%c@u=^duYHf6iZEu5f0CMM7BwS7n`
zb1oUF<be>e%#G=KWoJUxJZ`z{c57_=>(Qlol<@_}(i1+L6t`}mPWliqxH6V7M|_Cr
zQ>`AS_vX}3ZI<l4-Rjf}KT;8I-M<fg^j<dg)AIQnN99kwTAcxc9f+~5njCd8N2R~N
z5)aG=7f!uuT<DzyUJbqk66vkYxLES1K^jy?GkYN++cok@_)ymQ;aKX!3yIcDojVuI
z&F{!F30k}>@>d*bX<EltP=lU3OXT<#ZWbMLd=qX~S~dFJ0V-4rMov*7sIibQ7e>L<
zbFx%|Cg#i@A01JyJ>(seAp+&M>1gI?E-z+ff)7lvnKl*EM2gw&m@fnfwM=aC^olr8
z2sv7`+{3e#-%C%t6nXqqYM?7<gV{kKXrjNMFEAvw;zg6@eSD>#)cAb+`0Slv6BZJQ
zXJSG~_qtrGQ-ylT&ua1wIqfL%|N6;;e*+bvGsF{U(-L4uOO4*JSsD-#y*eQN{F!Qt
z=3|zpjw4zXShT*Pxa0WdsiV_gO30{)L|f9WL)wWp3${}Hf`>-5SQy;+IVBk2Un)8!
z^rCfmQxQ~6=Ev`g+%pEf{ZDj8vx>|b>8<!8TGgp=_TEs$4Utb;y!-7bE+H~Cgj<;F
zS#oMl-hZD%Wd9uXu`*5~NoO*vzA2kH!WAxHc;JFo(8NQ6z8+1VAE(#aSJyssk8MNc
z-4fbGQ6mlzH#6~?@No<NI+gPCjHso-)Pf6IthOrZ&8o2qZJ8-VL0c_%9GwM8WcQdv
zBCUapQbZq0ZXr%IjSC8=ekvFY<1lO++dsEVzPbUasD9>~tY1opFLHt~4)#twOL3ys
zZ1cPUF^{zr@M7GIaU+XJ$5D~T_uCDryo>i!=TpOHoQT2FUl)Q3;`W2e0;eY2kT<0D
ziS4v+8me&Oq;&3%0SVgL-p*(nZ=UlJ8^iZ=KfSq^vs4Cpxn&!|qy7Iruh+3i4No@x
z$XgH2<EO1_(nI)=XCpG4g;~N3jno{T5pBMX2A@>ha0?K$f0?|ceEX>WPzZN3oZ+O_
zm)v9t^Ymc9I&fMw(*D#mCClcSezR-L8=UCo5d7mOswj_rU4kpiLN!M*PIWa-SF*3a
z)Z$idTD4yAclBz#RBjIx!f$*Si5@fyF3;+rzq2q|GZV8cuhS<v^b1O}w&0@WOr^V>
zgv9n~>D-na__dLp;cjW<BkbgVCV0B=zS?R>#(vMAxxVo>>kK0aEv428bG5aF=1#lA
z3j>11M8e*eCzRcaTo*F40wndiu1m4kmvOb)C>!r}F4-lxVy@pZ*^uJW^(l;C(J$y}
zOPffu#j*=R9PH|zx#MGzqr0Kgnis>!qY3d|ey=p0wqoaTw+@f~S6{q*;HIR#in(dt
z?k@HzK1jvO<x^Z!b2d&X;>Fytx{^J0N9utXSHfP-(kECyHiN-lckMKjAD>7WDNPng
z;lxC{I(<}rKLw{4qwb)8`SbFUX`q_KCd$Y5XdvFmaDidS7T=C)Z1HHMlEU_&dAKnv
zS$z9Vqj$xxtgA4lRGgy*t^{V*gkO0^bRkAtF9Spy?i}41n4kKS4{X1~c=4sOu7EOS
zso7h^eLT%=iOfI^htV3RY=uyXcUIoiaLyD{=9L*fR}*ar#RDJ#xT&zWNoxT+mThd2
zXrxlP<*_mkwKf(S&3p=b8_n~rHNXguys>~lvyk1Y`S&m$<Idg+v$*akJ|pk`f!Vfk
zgm5Et-rPmA3~c?=OCV(YP$q_I^R%qIAs{TmpIuCl#3PSn1H<qHw1SdvJavK1N-thn
zHV1|`2EH{Q*;MW{!te91{k5^%I6|c*C^^T=CgsiP;rf;Xe`+b<AUT0H!H6!Q4>n2{
zn>&Sds}U0FAEtn5xln=oBp^(Y^|;4&R5HV8OWj}ejq|Ni9->=rzkiBbewR)(GAX^t
z)?aR>O1<S_vA$@7$IwV6cgtgjV#C<0xO_+yzqFer_v1j~KwI}$1(<P2ZaL#q$}Jj7
zL?TN<UrQ%}!NJ3wtRmlb(6~A06&hym4=>+y%$i8dqfa;)ov;|!E{Y!XNzDG<#JyUY
zHj_Hp3$du7dLGQUcIQZ9WJ18n-`yI5)eHo%XCr}B#*vZMfImAV=dre!@K+*6xxFfr
zIAJ#XZ^jW{f2(Sgjj0uV)wn8EUE`=(DaxDQ$Q9Y}OCNa{J<;?uom#Q6#TPWsBXJPl
z_!98ujWA1!K~Bz{QgwNh&aSw%(Fl*$phwz<H)Ue=Refo{=UK1#^@Rkr>{K0ynNkd|
zK5iVBtnoYf?!M|u;4-*gO0wc8dLKD|z!SMUq|x^E55bWkIeMdvT?Sj%^K+@d8T{<f
zTJ%TiwCY6USlP+kCH7&gyA9^9QADdjxpKj}1CcJp!I33ogY}fvN=!{A(Rt$4!0oO2
z>#N_<s}Xtl`9<0FeDTu1<HwpxV@$hxSBOgHnJ@1-HA1!AJzh~Y>nQ2j@vWyd24;yl
zC(vXf3ZK@l<P@=d)@D4xXFU2!Ap{y~{MpmB>Z~OHHjrW%ZMVD!c_7?ZTqNc3GXCs&
zw><n>ea?9d1sV-!jy*m}h%ny5?~+$07+e{uYDf3P8x>M`Av-S;r`sdB5b|x}>jpY@
zzSE<OWjoW4utV@%603h34ci6OUOY(P$+ZJ5bGLLo@e!4<(M~x_7ek+<(dlKEE=jkz
z#*e|%soj|Mr_8cTIK`ZIU?uU)7spAp%xaDb-C}cdWD5P;6G`vaH)dRJ)$5PwIdmIo
z6|8^!MO~eEuI}TOW5%h#t3=#uyumP#v`lkxi~qRbD*62~Wt2Z|tAB09;{KKO6nj<h
zv^nF_;giJLJ`KkXtg#W56eF?Ti>bgLSJidQeqEx8<$JmlPz*@Qz@Zo_eDRWSf5~&I
zoCM)t$&JsLdgM5q95dcI*oY<N%~4;8@)JqW;T?@|Y?KPRg3@}_q}X`Swl8mpRIfeW
zDF~guc`-Z<8X-Stl7{XXMH(w*rkRVpW;g(W!#0bz>Ym8QBUODEt)>x>t*{bMf|}|G
zZ`M$Kaz48%X@iXmUUKeZppFdRmmW+;yfPFy<`j<o5F@i{ViWS}X~Fhr#B!9%SD|4Q
zH#=iER5)ei1%rVh+V7OWQedKKG+dQ?nDNH~`QpS%(y&LPU`Sg<Evj+)aZ|sKx8p>7
zpBDRBuu)e~87Ee1JND!HCmj^y?Be3Do~U=lboku`$6D9^?Lf~l-%vaZJU>@Bmh8+2
z%*@+?BB9<Tej!#OUGo@FB?6q?;$dj==}kwhXNcJz>N@(*tilO+D)l}nS3s)3=8Sh=
z{_qx`>JH6JCP%Ta)YL2n9liowkGs?BO06b!ruTQc^S8Up-)=NEs|Ch^;SdX$i+5uI
zUS(e_HtXCXZCR~7{6jW!&kv{f_lk2ro|DI9g?-(RryrMwf9*HU-NbwX(RhSw>`EHS
zjQJK5AO3?p!b)Ul-jB5D(cTB28y~p`mBm7vf1fz}G2qj7a|dv8*!c`}(?KGzLjAke
zlBWj>uutd=wNBm+40JQi5pMNL^0t`Zypbbc4DKVnpLKiIuC9pdi{6#dJ?Cyrf8v&+
zqhMrre##LM3!TIFg+;lvr&9>O_uN?*1fc_MgnT)U12KJR3zE4`{>}{nW-UWsu}}~K
zAd8#Qq@hupi7!>7K$%{jS;TL}gpt9HxpLu8<j>vWJZ1!;`$84DH<OHQ>Pocsz2BN*
z6M3vb-kp>3&ELJDi4XreKPKuK+$mW|HqB$&>loW1ijp+KSV|bt7`fX~P~~=RQmvjQ
zeXT5tfSMZjO|l08+4M4X{<AGo8>~haC_%*F(#kQx=S03~i3y1x3HQuFl8hA{=CtXL
ze}>cVT=}0h>IDWl?DbGb3?Srx5)N5?Dm8AI4V4aif8}fc@d=$*T4f+UqIIx+#nbTZ
zsPp(p);cOJD;}G$BlfH-^JN_cYWbhQXXY9gMwPVaMU5vD^=NS)(>rEw0N)e~?@Q!|
zniVcuET629$_Mo!U3V&DSUi)GJB>JaZ#MB6q+0Mf(#6dPF!r)^amQV6%^Tp5j<s(S
zw)I_XcV-^Y4D{aeR#g7cP!knTNZ87?EvqDRri*(aw%~S;({`^!UtL1(CyH(f^RVsg
z!pS^T?Bp_X$A3Bx2j}IzTILvMEY}N_?5M@G05xRj*O^HrDMei_#W}wxO!dN#f*P4m
z=*2})b1p-p-%RmP?H94jQ8k{Cqc${{oV`2MWh-Npi!Iw6sm43cI=}$h%lBAI{URFK
zs`CsVG8^^@#_^VS{7rNHNbudL<ivl=OEmo0=a;Hliu%MD&fOU>T_8d~dy3n8o^)xz
z$aTpZ%!O$BA)<YP7kltlWMb}^yBCvy5faw+IvwGB@YmhW$Dh`_ai@;s`GWh0yWiD{
zHy4du?@h+O@qsJ}S6^sri6t=vkZ8Eg#w<&H4R&sy<JNKJDYE1<6jJ^B5a&ThIc>_?
zVE*M#l!Ha@I~ow81%>UuNs{*VUMa!R+rDkfZ?9wDc!ozu3b{+LK@-_jqVxqn7`5wE
zHH8#Ap-n803sPfd6Qr}fFu$3Zkm<RrhW`cE&gd<<IyOGeX+XO~ou+<lk9E=@zolhd
zH23~fJvV!KcRaD{bm8o`s#ICD;}T{K{1*g8G~+a4f<yv`AH1Y9l0DTpjar)t*7!XM
zNlQ+oE+wSie`S(BRz0^M!m)c!p(jg5Aa2R|S#kBHSoKfY@Y@CpLGJQY6`Vd$Gvj;?
z(E~sBLt#S%ODa-)^%b+mNeNB!C5txJ<HI$E#a^ecSUUVfq~#N;X4{HFMu}$~FF4;z
zysOYizcvF@cgL4ZGcxtR6n-3=pIYT<X9*l?>zse8mnD^|3&%Xoxo<96r}{KOE?GW9
zom_d@ac@l3hBy0q;l1kpCDvZ)9cUykp81H)+N}*1`i}6r;Jw)xq49zu`PFO#nwP2&
znh31Y)k*J5GmI!{&NK1~wep|0FD2Ehp}=fACpr(jdd?WBt%Jx)DORRT*B=}i*hVbh
z{aUJihWE_bx`CBo{)v~IbA4LE>4?1H`^uUU`TCzVzGvMhe70My5glRU=v)s5VIH0D
zp*!^As>5Jh!lLmc6~Bu-Ap*a2CdU5WF0Bbei(ZYzvvgZ=prto?hZeFJj>UV*o4cJ7
z%0OE5?X%1yf`_~Q^qEf#HY8_)2@2sf-O1J^@{8uwZEz?tWY-3w&D5Saj~?#vR)wx{
z-Iku5el6h|V05?hT-yKf+%h&zn=Vwl{80yi7fqgVwBL`kN=3E|Z7@@PFV%zAaV9&_
z-LI07iqut}_nHE0{rM1sLDI7eO-@)Y_i^}@Za{ol_)_VZRVTp(SMU5WCuX(Ge(i?W
z{>0`mF7$X8laUG<E)|7yITJ-1M8c(uySWqqc%7BZG+0>_1YWdQtV!>mYjr~(?ui6G
zVfPo^8<MBFZy>ekVz778j}!^TCLG>-#|pW<pV76KZkw9|quQM_V#`QtgJF5OQZ%~H
z_sddm_ICZe2{|U4SScdCJ-!;@!57SMG^8n2S~L0~)hgR`)3a=%Ct?nnyUl5JSMne_
zo)C)M^LG<#2qDQK%r}Wb!B?`1$U?sfrm3|2?5cIesQ%Tb;zsbQ99o?MnoPb-?J<W-
zp8suy#TbnOGXD+&|K5kbID!}svw?rnj0YeI3DLQ`Fn?d%%}mkJm@=IQ$_c#c3-XBs
ze(qRG1dnWH1;(44YBfp?|K8H7IFS@jxRXfPlWx?&q#0j_F_!F4zhXRL6Je?zl#=o;
z4C_EuCnFtvP;(^8fz;jMhNbb*p9{IzH2U>=q*37yl}}zc74(x3?aLAw@d{<{jLyf1
zQz^q90>^dRA~Vrq?!5)1{S$8co!;JU-cDpcmRgVhbQ!1^Da$2@+ep+U+i@pJ=UZS#
zS#Ui5giHN;OY)`C3YjvmwnHRkmv%RwHSah1(s-=bJzU0$yoBn!ow9uGk2Yb7#yy$+
zPokpLRTE4TcJkO$YgzHv?pyNNYxDN2P8B4Yn-$qg_deY)Pqv67`vgyjEDdmP421V%
znb?DhSfPu;ud6{-@u_tto0|6$!n_ztyz71a1$Sh;w;dmhe<ecXk551@pNqXAK^1TN
z91WOU+07n<ln|YO;#|+K0uYu2r5(ItycK!Wq%dob8eXyKYKaPJ52L=d^fYTnP&lmF
zU@x38fkULdz`h-*s860(WD>1xubd-NoQCD<pA&bm;-_9GI&E2`5@gnX)O7sP<aNm%
zoR}M80h#}V`|A@t?5hIX<&BSI)4}-EzxM>6J+6J!Tu~mDk%>EXDUgCYo(b1CbCJ!M
za!#$6ul^I(-tBU)8R^Wd8<4X8?J_u=6r^&NgxohyHJbw|E=`wc*(7D7b<+^|sa0H1
z$F6f{0A95*ysakZ-ARmq5KD|F#Tmz+aCQe5qu?F$Kbfq5Rv6Kn8P$hRv80Sl=o>IA
z;k}X28!Vv>EegKU1gJpb?ssoSTu^d6-w>Lo5B{vb;2W5tsHI1$!keWQaaF9R+={Gq
zay5XG-wsOQ?buKtlP%JNH?nd~NVoe+E9ycI1=AXfkN)McIy>bu9TghW<*v5iEsY65
z`9iwCr*JAHw~CUeE)M$5FavxXov#U@u$O9DHx>iVeSuM<n<Q0wuCqa6g^BAYchV5#
zV3s(|L82jO6pxCC_3`Y&R`oU&7>_pQ<mq5Pe+5a-gl{34XDUFGP-f*&BtoAfN!ndi
ztcy@Ep*6!oD{3v=SCH&2-my~GX&|;q%Z8ZLjj!nO#J=@)K**U`WAHZM>}-a9sk9q_
zeyzNLmmzg0loSlH!fZ`(zT|PDRf0db-d{r43<T+h!fXLAfYlGu#P4vrzUVjlLoXy}
zg)ME8#}I-b>)viZ9ox%e+W0JWv>RMmQyBX4G{sblWIuPvlk<~GA4$TZ7|HeC#t-O^
zbFUfSlL(=SaKR@Gh3`GHtWaqP(a9%C3?j>~3+u`6jC7gk@Ul&a+2%xNNx&1O`zyd8
zrWEQ9HK8X2TS^3HK?`pF@KxGF!GU_(w(Jx-U(alb1I3SsN+?Igj6cT%ueJQ#&uhQH
zgb;c6^NfhZXfVi}<Sm|^U$6)3^zMhI%T$ySxr+v?8-Uvm@*WcPz9x=x0<dst99Z|<
zYqQAi5bsCLa^%k}QDDZ045J$E^z}3S=|S$cdFei}_Sr_(_O|m~#bd;MU*F$)f!ehR
z(JOKM9XfK?<50)(5y$uVDCX70GB9v1)lxYY&$7a_QXa4G3?~^#b{asy^C-5nL@FvY
z;>^3PSjmc`R66O&L&~K8ge79MH1TyuiYj1z{`npJWHiSFW1p_Yw{Jx^P{mfTk?YaT
zG%bME4hIsh_uSNr`d}&V5H#F70&rz*`yk~G`zZfv^+}<zCM7{IJyhdEE>ihiK{`F|
zM&aiY<VD_jnc7383cyvLnz}K^q4W1ArTgF}Xbu-I#_3jhywFpf6N<i~7};jLz(tJx
z)K@@$YlU1T-#yN%C~J)}y#xA=v|IHb?<N*EPQ9O#;U-VGeF<jty0-2x4&x$rnP+aj
zuY6z$^D`D)&3f+-ANul9`1Mep(OiI$3A1V%d=eLICD2c9B!0WRxh0_b%(~%2CS30G
zsVF=)Zo|pip&pEzvL*b=G0ME(OTYC{7grJbP0ICbHTUwxBq6ouPUDsq>Vs9a=xzR;
zmTsfkgQh9yb3VtxyNZ*iR1RHSjV-c`1wgyleCa#mvwbU@#v0^b_ShXCN8<cgmlJOE
zFE_yB<G@geDXFNtqv95ws_5Yh7gcLIzBCPOI%G+I=N@cV5!M(y;zcA-C9JA9lcYAR
zZ$9Te68L({pTs4-!pQu}3{!(svZSpNmZ?7jN#cK2YgP{@!UGw5|MpIMBHN*gkyh+I
zAbF~x2{{~R-JMSw3N*TxKZp51pom3H+u}VsP-%kN-I|o&+rWvRU^F=Neg&eoMk&p$
zo+RgKg_XzDr>@E~i<kI|5$6_qJm|69t+>gSuG7IwVXtAbS04%U21_+u=l4GBwS3Qk
z2kZ;^ep#k-`N_D0x096$J8m=o3|GfvWZoCFpf6?DTNqH;zvK!_7jON(55k_MO<cqf
z`<eHd=_F#mdKii!G~X0C+Jo#2sDYMG{*`)*sI&_2)cv?)8(n-LO9)bVC13v<v$%LJ
zoXveYejr>?k4WhDfPzPEjb&~{xGM7Jxy@br%^~~_hQnErb|a+P3u+C6Q|K1}P5IYu
z$@Nb9F?nv-_KOz+#`fzED=}<5SNbx7Y@n@;uX4RW@)-85Zlo7X@zN&nEgNaQt4&K!
z%Z<{Y9<zn`Gy~S)^1LE{_hFeZ3;x_%tT}u?$=X|pJ;s`19P-ESI|=l~@<}W1i~A>{
z2dBQ$JFE@%QM>(a0gvBq2`$w4kGrFu3-0}(_j-$XhxJJ*!WZYR5GK?&>BBw4Jo&KR
zBda6l3eJ;6hDb-9G&&eFZn3M)up+Caz!yl}i8aTv=eJHMpJue>KZzi)XUv&MwrFlh
z7W?;M?S{gG@6-ea;SluWI8gXk>N!4@>>!Ld$I(l|4kUmwzq;x0SPCC1l-sb~0Ib+|
z1tXI<M?K&8p4q{Ilpyq2dL3Ry5IlMpdt>9tVOjS*OPyCKPHU7$m!bJxd&Ux>7&jes
z$9FCq2^#0)>n|34ad7N?D?9h4j~qSI#R+#3G>hYq?>Rd!q!HhtG!1^@bJQ(TI0apl
z`4K{7=j+&5F+))=Y}Zmkn12=!2D9WOWd3IPaM>%euLNV0=6J*Q0H&$ZWqnX3A{P#S
z{uf3bJpXQ}6uQ<$ollLUOPKx~0qVFKNG*Y8Dvu3y27jez+%51@lAaI@I(F0sN!Nfb
z3{Nj8lL*TUt#mqAGP4xRLic+$H|_V}^X+MX;g~n!S;IEDYcMB8FvJ|I$&{xN`jU|1
z0h50@+7z8LIAl-|FqS_O`Y9`h955ZerQeV!DkE;BrHHEjTxMw)YUBn9KDlL#zlo80
zpxTvBxHN?KF=liSYV9F{$o?fg6zaZGze-P7J#3|WP#@Y)u$#1~LUJc*@d>9#%Lu}0
z;CvLZIBw%c1Cc&>y%iAvC;l2gRdw$1BslS~#i$UO;-NsZX@~0OS2V2EBb=mjR`5HL
zCSOiu0`75ftPJY+imCXV3+jNROW`9P$>6SgnmK^CU8Edb>P>f(#?^}NCLGF>bkOHy
zPg<DP5KBxercFT|2hrw0%I9EkGOZ+oif-I&Ln6T%qT#h+v3&X>cM;>0icZ{<BKgH3
zWw$KB1D1bzEx1==#u@B?s@H!Ocfw=bot>oe(0#uU0w_P9`;WwWnQ64iNFTw{u{OrY
zVeU)oFXHpysatuuN#y5ECqbE8nA?!a1F<AHvz%0U@%N<z0&di|t_%{Ef3u5fB{|!F
z{JZDQ4}z@6Z!s$+N!txk4Wxb+G}PgR?U>J<3j>2hr9{Uh9-;7tA{B9nU*d+c6BO~`
z=@d$bKCROI{xZUew;+<R>uZj#oV9ZrcMK{aTC-B?85>0&(a^mU`diI+#1{vx3piO(
znY`-Lp{NfMb<E2zICU#?Mi$>H(^C-Y?py>^Db}4zz6mn#evG4F5@}6NLxuISwAD;L
znp+}0B1m{#veuX8%Zi&~|Eau`jw7@h|0x?C-`+DSf>1s@^I=?_EO&XzY{O>_14?-b
zs-{e!O|9`b^D6g)cYsjTqJZE#$QeUHU4Prld!Yx+hG*huY@~fM?+z;vyLR{JUJTq&
zJabti5<z)ce{vOK$1jwekrVe#C|4InI$j$@8E2(=>?b}`uN*A1oM-8C-e1&s5i)Rk
z+N*VjEL=6a+xa`<D)>y^{oB>3!Z|A{CAZXZ;5gSxn6i1`aP_?_vsK%(H8#$_22{q7
z42~0EA;i2;l^5P6AQk>0Sw~g&E*{wta57;IKdK4lb=0M5$@sel1Dw|X{`&upZ2$YG
z-&468ohsUfUQP|+G>|DBHSHBjj1RKgqw76%eUUewWdU5x82h8SIQ{>ZI{};p!Z$Ov
zJ0B9CL;orH{JZw^-*4W~`gZD0TGO3P0Nx>wZ0Q&tCTom=+K8xhx)WwaYpVhSq@DQR
zjtY2liyGYeEyF~X*8jfLzsgJh{yQ1wC)pV1kpg;|$Tz+W&Dq+?;L>kLy#Drs{(VtK
zY>=o?zLQ-b68^6U@ju=9f1UOXQc6Y_C%dy?j{Pn$?(<R~I~(8geQ2LyVatB;Ki(Aj
z!v8;>!2jpA(f1F-eE^=|1E`}p{68=JFPH7#cmKb?`CmT2PZG2K`f|H@y{MiFltc<)
zpaJd(#!$tF^@8d%D2o62ruuX-0oBQu$>_%$T2aN&rJJIXMdZMS{qGON7w5niSVU3Z
zLfwBqy1NFL%V|E|f}5p-O+Vh!H-mm(L02Ab42`QUp58z_j*9fZWJkL}!REvEl(o{2
zJ<=bZj?L*LBdLHDJlicWeW1$;JZxj&%&tmvJF5QOT_Z+Q+;oR65MTTPCIXaxlr{FN
zr!P4^It{~V+g6s*U9|uxMq=Zxk0N4R@JxrlbkzI59^0)XaMw4Lb$0*xpGn|hT>;6<
zkp54y+7R(dGl2HP2O{$J31Eolvk4kv(hPt)FTSGzws@fEi~nAkAO6eRizTgrtDS=A
zzSzl(-H^I$kR-ln=>_>+wa5_z4;xp2P<m~W_WE;v6S8_!3sv7m*(-K#EePg@nn2q#
z<tVsVa<c;4*B2e^jAm|k&BPS^`i&p|ff50j31wRpImy*YyLd3LZ(@D91>f2p8Gq2!
z!aaBbdN<&j8)%Q+70ll?JM@Z@#JzmYBA&a3<&~I)Lo-<k=F4lMawC0?tEcawStEnm
z*A&9e)^c$T0~kGHufKe4iq%j_bY4YaVHVXMy{rK1;q%hNF^k+(_t$BLect9@`4tg!
zlT<7G3g^;0$6(zQu}CZxqm`RnK|-KOLsj%)VZ~fH%zCsqXM|7K_*u{Gq-^Q-|Jqsq
zu{Mp;apiUd`T`%;|FfD(C2sV00VdNUo%CxkzD%@~>Vd3ZYq*Jj7p<eDc^BQ-UjZN?
z&ER9H3D#)hbUb%k^<gLZs$QSh^w1A~>{Z(49G{dAnuc{MK$=dfQ7hj+A9uo&cB*s&
zFd}-sXv%e#WVX+9?Qbgq>DJYoWq4>U@aE573SV}3{rj;$eT>8DJniK|V$uV?P-8%r
zJJulx0g#F7d%)wi1Jn+tCbaw1!KLB~`hl7OIBR``?osV}VYN_2HAkWjPG1aKhG`Z}
za`}S=0!!j%`PoST^+h^sgg78v>Faf<OxmjObn|XeQ8u1KQC-iS3hNwYX8^q7kCvA!
z7<<6veAoqRzH6M5i1;ie<bB~vV&Me{VEAZ2`Ka4+ikDK}$MmSM@CJ$Uer$*ijaqJ=
zRd*A5TkinEULfGVa9LNa=FF1RpGfO~P&!F}_~0b)BR;Qkq{@FE?SG8^KY#SYV5DiB
zzHw9Q;9uv!9Sn%PJ&F5(JX0m<9N^F68J;weOm++dnvm8*?1(I!OJvP%o86>+dk{dJ
z^s-yCqq?|DcC@8m8`Ys{N_M;B0Mo2Xh`8UC(h<e?i`+!oGi2v}C#dR6>YO!GUlmW1
zLt28(IicqtSWecZ-R??>-LboO!XX962}dJQVIZD^Ij*RcZ-v0HxTewCUWYdcxeu2W
z4bPq{CxJS?4aqHq9qj<byK%q6V$h>7-fBL5+T&7F2Xh8QSR3P?4DkiU5=Kc_Pl3aq
zq5D=UwhXY8JDUuHo<AYK20KgdM+Elw6zB1u`u#krau`oJn{1&x1Lz|1b}=B)<TRgs
z{DE`Vk^B-(85zDDC)_vXa{+NGoAS;`#A$<+mU1lR|M*Y;{e1nNN=f+FKYgL2@cGYz
zZ)j1SLGxA15$N-gl*>e_`7V8&>(%LUGn)HMmlDjho{nRNL*cRU4UH#r?6lEYY#M-=
zP(r`oGfU73{EJ9YtoTOo4Y1#j9R+^59e)AmQ4U@Zvi4AG+X@XvN?N0BG`#mt-n+e|
zlY_jNKcd1GtX7v<i?@exsKOlU8G6?*B;I@MD|7h^>;q0trpYhdw`i`Uy9#0UL@V+9
zFUw|Z=Q#KlRsOHz^E24U+TD`V|7`E{T&%0pUY}qP*>iM>?03xqHOzdgZ}?UsAKU3J
zjYBB4-XoyLGr{{H^Xrt+13*2pE)UT_rv9(&UhBHkchb%$K~|=7yP;$R7yt@_yGBp2
z1*|JBq)r0VZmuhX1J^~{Z}laZ>Ko<?0vR!;0zidbwOZT<YKV8xM{UE}-C4lmDy??z
z43Dn*c^lF`BPBr*c0LYhPi=!o00@Jn0|gj&<4*Q+Yu1F8DT)!ZW-tS>ZT@CTfVAuy
z?DI-K_DLC2hCP6Tvl~fXAl%hL^~r<j9z+n+PzErWVn$x!RnoOhXG7=?2c*~_pLCje
ztgQILCuUoszotPPKJQ-#Bsc^L(3G9>IyC$bn7$Q=+-m$DE!)ZYBBu{o@t)MV><r(u
z=e}EVP72?hsZxizRuc*@d6YzQZvAJH6(3u$dw0N>Ux5=>=Kn!xK$Y76*DpjyvY=7O
z#5(cH{OT(l98+dIJ(l^nLo{>E68UZLvg1rjNovfr67S1*oecA?pCJf;@wQtDlb|mO
z8Ft_P(Fhc6QGrl7zC+Iw(8vFIy8_}3EQkJ&Bk&MHTz<PTDS><mGV$SY2I&6ytDE#3
zoMY^I_N@=aIJyC!&~JAgO<MkD28`@IUEA94)s={kpQtlMRyt*(DaLq3HN%p94NESO
zqU%<X@rKuDBcj_C$K4^w35#5(>W#e>O|d-@kIULfU`|Z#syYbnN!0_JJ#qeu%?<kC
zVwbKHq%rlKPQ%#>SM~!wr%?N_n2M<p7ajdmI{RP!4~cq^W-Xdllv^eGJAPTfp#V_F
z*^~JgmkdeKj4nc1irHlm&_IReJmwiY0;?VZ%sB7G<+hIt?Zao3E$*CkCz*KHaXKbp
zJE*kFf(!}2AZMr#z|BZv5aKo6mhun1ZO;-pikUIY5U)4a6c!r_8bI6hvFLgvXab5i
z0Qaa8?`+_SC*8p~AlK;y?p~d5drX4a`<EYf@{&Cpa0m%bPXNb=1X;NVDK{$xNyZ9>
zY7kj-Vi4lc@dT=(aeb(tU&h;h(Prim|Mot&Q9VbS+fu7OPm@N=D8Atx?lX|epd>Ir
zk2gAZzC&mEnms`k?Rl2Gt3S44i{g-Mpf@_#Bq1W<Wn!GRmw>EV9Wduhn~IaIU--mm
zO61zU(*H@nAut^`!(6?Aow><4@LA|z`xuSXb3da5eveLF)8%C<5&HGD5$;eq_B}NQ
z?#)lbm#4p#g_+*9Q~ONn6RC2w5^K)QJKhX5s9?G3(5{<1<k;->eok^cVJl9;Hqc9Z
z|0BR}+`Po{qSERG+&cnW;j#6SiK#X}U7N{Dq7_r;-)^s5zJ*WNob7`Hw)b3L3Nfg}
z6Q#Q<6*_8*!9}>0#xer@x%PET|B@P`lD9MXE`uRd?iL^mDU4}~JAcm$voMDRzmC3S
z!f{a9O3Gu$C^ECBzh2#zEW{}IR|hx;*Z~I*im{#ptcS-#v4RJ4?-D$Wc8Tp7M(jij
z`^erCw$PeMiOA?@On!@%zy2wD#-8O}{+7+Dpr+{nP3pE6Py1h%Y`CBsdU&l=6LT}R
zauOWABPHKw=o+aZ=85DFT-5_kF)neO@|YckshPty{wX^BhvV~am)!_P0GxVQ?7IQx
zKh8Exp0I&+3au!h)6#PITC6xde8mWWLh;e@RtCevmWQV8(N0|CO%J66aSR9)X`&lh
zX-q<hR0L_My%x_^0qqkM32#@>;P~Bax{XB;s#2WC_j-uATqfvEbn7TFfwUca69%-1
zX0PG-!?{Fj3AsSLr!9sT#x_Xc-fL#QLGEG576{t_uLx-q?nrZ?3(5k`z(%an$zb`R
zn?jd~NXieJKv(R1Q;Q?T`g3qfypP}*tZUp0lMD7l^+oGo8$&Qjpvme~I=o*<guW87
z)IAo-xJ!^h1xD?V->g8I{Qi7<jEmySKXmjBv}Io{N$N#*elpxsfBy})Ccrt;C`L%a
zFd(RiPMx})3DiQHhU`T%jn*F6Y?4o(1mp2m+<MA>NqTt;qPXv0L~Jau`BCtBZ4)09
z^S8CcWEeAcKlQVLuVv$`9;4-i!p0xM`GHI25%`Bc$MLE6*+%5dE?R%#*!a~)dzlRL
zTJE{@-b^JNa_eHh2xep1=?V~AxTf)dCafx0+=0;|C#TMSgvSMESXgmvh^A)3;4u}z
zNNKWsVLJJ=MKc<F2%@J|bIGZv7Xf)Vct$u|q!BCi#vkx;Ci4eFM^|hK?)7gE_R`Er
zodhv-aGAEf2wP&Y4Lk@H{x%oF_q2`sOu9$9MxiQ<1mgi+3mTb-eV|iA3EGauSVXT=
z$7SgosZC)vH`r1RZl3npkb{#sUZ1MALpJqQQN7EAjJ!B!EbNsLQr^hmDYa7KcUDjK
z2N(LFi`W<h2KL`hV!RLNZR=~^G)CY`XH1+*sDB)1`oF^P0DoWhJm!B#cn5?Ss7J&j
zFRi{>Kn^o;AH9B#;26rRRE}WG`SK*rIe{6LWd6y5gJrO=BNx{w6IVJDUiD>zZ5+04
z>rG9xBdy!_RVauzX%v>^OP7>axJGKIMCzCJ+aj~MgNvv{3Z+NdedoD3H96z0uq7d`
zqz?(7FFjZJsmlm9R$*2esSqN+7gNmz9j$18TN(aB-^@_S^MP{1yXmRpX%-KnLMbBW
z7uml8*?WE#h-q#1`mE-t5X4tH9h@f^{aU|w$V}e#PD{^yHM>@+<UHA?T$k!HDF$&$
z5jxNKJbdXhZ&CtXl3mC8AxH?W+l&UPE8EH&O8Ynh?6XJO`&wUq8@g%>lST_992N4p
ztlDOo@kRvW>*&&B>-5^>4>f|92ZJ8drgZZgvmIrX&?9~u%<mR)hpgv=w?-1xn3Fz~
zqq!j-^|8^}z1(wh_Pdt;I<{YiY>3jNQAwJsy_N!urC~+&SMEpe>c;r{&3WHe<c4rX
z+7PkI5UFTo_d}ny0pRGD*VKkek4d{R4SCrN`wNy^LEu}wMzIzVrc^u4|0SxwU{{k;
z)sWO7yiIM6e1DfzI*|n1;VFgkV^U>nG7WL_O!XqB{a$e&&y(2@w)n(^?(^@UR8?J4
zl*a$O-pao|$*f;S+32D2C@G6TTWBU&WZi9J@>3jNMBGzipdkiGu*@O`09`fLLm1X*
z*gxCkWf>i}gknsMLih=d<FWrf3Lm#MmP$<7?77Vy|61e!au;ts#DKd0zt$HShQoKV
z?y48O5iEQxk_E3I5hkL$3RjpdcW@?7Our=;gD|<mj@AcxR1Md%&*%Y&845$NydU*b
z<9%EDTn?iQz177Hz$%I#m_a$DDYW5chOsjtNk*d#xy_3wMZaa3PURwgBD(Wsc;9UW
z*!4oYvWwWg?hj*|2U3O>b&bE&6oaR~@|&7`J{;?C)Ph{Y+I>7527muqmkOt}i`D$7
zOu2l5J;u9Rb};gO11^HAyEPuM$&(Z$Mp6ox$R-?BFh!UC$<J*vP2yNSp}gxU0i<}l
zwme*x9WR+SKZ?)<s$7!W)2?N1KPxqld;F==Rk=LxENd(IFwX)aY)YSTXZ2Ze<d3m}
z^bb_Bnt?~y>ap;c2!wvm`PRj*_f_gcqPIQZ^lfS<$I7--%1E8Tf4cQ&EX>JX==%~8
zp-`qnOAg?gs(0^R+;3CFYDFv#uLr}K_T9te@|kI1l1g$@9k&L|w~D@;aqdY`3dA3b
zX`Ca3533`cmZ-+ByWYK~;CuaI62x+J{;pW{IGFo$SD8wV12U*x$LHm@Gez9r{?97-
z9W@hE$&qQ_Qi?DhFzkbA?@E_jRuxU10<^l*e>PV9`waQ}%>fHW<pCwJGwBDpztcGK
zQ8%`jx;w(En`(i7%sSToVHWt6)JH%b$)!{G$(spSV@kCjh#Q3ia4#qC(dEuO=r35U
zWHTTV$3h`KRPVhns>o}hTn6~)We$7$d<5R1_a!DClt76B(sjo;!6ckTta4?ZJ^9#m
z9Hc1>TvmT<0ZgX=T1ILMvJb*o2Z6|}C_%f!(DND36@KcU`V?EQ?S0_Hc4RJKtK8PT
z;1trUz`tM7$cLdeyX!rLd$VJIK#}vSeWk2{UG?j^H%8Cmo1l(B$h`RB$h(SDK0Jqe
zU|`996Zl$7^l3%7rzv6~;*-2df~*OntO=T8MMUdg%TudY+ZbMp%?!ks1$^ns^u^Pf
z6rS^kxKx7k*;(admfsaSww_{=@8cP7@{uoxdYWiRD>~V;dlg#uT^&FMuV#>lF#Px;
zmRLgBa5k@`AYqrU;kvq$`S>|+oJbU^ls#9soLKNt{@V6mKqYRZVBjpF7(8R4b(vFs
zzntpH*6bqOvDYeI3vXp7yYq1CYgII4?ch=5zd;g;GeIIBS=$mNdWm&u&YwU!tS^-8
z`@JN2_x^D&<H@=rNoW4F`6S~u%M8aO?RLrFySSzHM1Bve4wf#s$1SRCnVhN%_Ewz4
z9t0TpI1kFKXI}zbZ2pMPxOp{)mpvQ_QM|c%4T2#Y1or5}i&_u9@#!tkbag!;#{bJK
zT`Ga_*_(qhTlvr7V)Tx&cmi2n30vQ7r^ZJdw!mqWcog+!^K(-_Z*)pNi}f!?pv#+j
z@G93%Xet)uxIKn?JCjQ4E{?||+^f(Ly@$6@J;ZVW@U=aw%C<@DkY2AX2aZh16BDiZ
zy`MWlhN!B=hzgwZu9CDgTdhEU_<g*kae+xd{pR46uxP8ZW~Qf>-+8q)cOf=*GRRLy
zjB@1d$GhrM_{5o+{OhBU*5~I(Y-zD?;ZGfA7wy|Awa3z?S8`My&*C~6{nFI#mff62
zCu(jx+?*&yA~?6q{OYOq-*#&hl@r?Z-RkDf#rs6SlZf4rWX4Byk*GUm1mRu3^Q8k6
zaisdoX(;$Q0~(<Vl@)A0DrZP7&H%VI-qNrj&1)oa*BKn97~H65;&HJ$i8eafs#59`
zHsvYD$ClGZr(F?Ya^;mHhhn4{D<I6G>}s`TesG~q-R1O`L~$j^uL*RQRGPVD<W@2i
z2_1vY=g>&A_<iA>qWi~4c<2@BIcYON>IhM090Tl&!i$)?EbsW5COT}a8XQzt=Iivj
zUwG!_{VHqDWQ+ayr+MwJ&$hR7qJ|Z%1pLL<X#LmGpz4I=Z9A~=AazR5w||dargYsd
z8TqG6_216%f4;eO&rj^TNS@ihdM6*NHbBItF4ur(w+^)vu!OkJ?aVfRf<Mjgd<E1e
zLwUMjlDuEs`XOiYAjBi?e*U&P$a6kQAQO%oiAdo{I>|s%jNPozY-VTLP3Y^JZ~4H9
z98oGIR3?ZOtu3+9T9}-gL#sYmKqe*^X+bh1B@9nl&zj14)Obr}q+shsp(45Y0)6EH
zU!!4o*%HxlA>uF};q!?yr7E&nKNDAl)#q!4s&#C%bMz(1EIGSm#f|`YwZTdG`0Fpy
zsE*OT?<~g~r7DlXp%Gm`Vpl)@dP>RO@g>#en;3&(+)3tvBHlZU@4^@<qR}j&S<8w8
zeIUtkC*~VbeBMzrbJA$u&PVQXH+~b;{fG8M$Y;fJ=eCd)dik`l{94<24^%yff4;Jg
zbI;|4_thm>FeEXITx`YFAD0*`P6g#iC4YbEau{s|fA&_<kSN<j{17B2Y4nG8h+Il5
zZuA6$T=M8S7wQq@|6%W~qpI54w_!mV!A*lGEg_ARfOIG!f|Amm(%lWxg0KMv+<*!q
z-6@-tZlt8UOB&y~&N<IH&o3UocYJ?+e|%#&9H7A7tTor1_kG=007S{G5Hm3)B9;DV
zyrpjsKsL6ToDQ@KcRDe3X69W*C=HbESS=1#CNB-hH>UL{`_^vqPk_oHOXQ4}TC1Wy
zpJ=*lz=jHRepQF9mqG=Na#7@S2U4Xcx0Tq^qZ^jCo~&H2XNyg`|B>i-U@?spjY&oM
zpR-CZndp~8gLT4N6D&B;7Qljn5v+KxKg)1Lk*YU=;JQ7~yxTyVe3lk}o}0D_cX|A$
zPBt1~B)Deu4)F*kPkVLWt<C{BU9t+@r8|^DGaRj$FEVVt&>1vOg6nh$dLFKZYj&{p
zNRw)pzR&kQOq=avn{5HeN9n#B06<Ji9~w+zNrNDyNL~W>;yArSHvlfW->KbeqqH7W
zm3?O+<Id9$cB?MA!u4RS^(Y85)@^#3cWpUl1ib~KcTT7sjYrQG=DVg7nsDf3RR<o5
z9smk+-kQeu<eSS4f~hVnJNFMRpI%LQ&UA&-Q@b%CKH1!^&(i=l8}RUku;Pfi0WxCj
zE9UE^l4M}|-TEOg18BkrTLTvPD9LNP>zyEgU)SLRihGbi_XF2h{cM`8anp6&HnB=L
zcXAr%>g?`+y+c@6-4h?=j=&KtSDLwwfDoBNOX|3KzlrVVyJKDZk1WSN{)N^5x0C(P
zPs^mh4()2Ort=pF3jd1M{pCF_**4fjs(@lg)+v&bAbA*fnSY?wdHHqu4;GF<nsnJf
zF}5PZ`q=YP-0+TWz*ux^4*5u<%9|YR(2OVGjBB`z_}<qXNTi<9O=4kU$iLMmtEw(~
zrnDQ@$}AmsVYt*k)oJ#}4H5kSPJDM@L(N-<Nmn#(mc|Z{j}Lq+Z&aAH3DR$r0d#18
z;ImM`i1Iq^F=t6p%Drpq<FH$48yjVx*Gnmz^tHC$qSc?wN2ZfWcR&_=(x2dXSaAgP
z%agZJO97o==v*rQH9yYVNWR88wHBWFj+sg#mFN61dLyiZPn8v0xX(=5NYDJv#GPio
zZRgnofT?y#Bmea%E>;z=x~~@m>8WjymCSlsJiT1aBNFR8nSlNKV{=4*TwaTu8TR>>
z2A)Za8aZ%;zG%{KT%_loElqYoO!BAp$W3g^?k1NbFJ5WQw%nq38V0!@AGaf!)*4Rx
zp8Roxy{QfXGwZ)@od5o*HZ2gKuqOT`SqH<q@ULKyu!am#)6gA}agw9*FlkOS@UZKB
zu;RS&C|2yMsy1c`wS`mj$paAou%OYfPeMjugbf=|j`xS*+Jy&%-k>0Kh<{u+z6-Ju
z+Jo?gI@pcp5S4o3n_*st^EKc}Wc@go=Ll{>WP-A&M@ycLXkz?L9SmcoIIj$VJ????
zr)Q>Z8i)Quyg&G}T(Iz|KNkLzaWPVaR$TA4acw-V(9~lPw^)`=HQ3@qS#8$)F)Lwb
z5!0<3%fSAH$jd3gUwXu2!aH>frxKe206&t+H#ZGYng-0}CqDozl<t?Ww#@z%8OS4k
z?NuwEJKg%Xus@&%i3U#<CPL*i()Brj^xUZg0|I!ZZx}jnzP*uQqUhZnDl**cM1q%@
z1oFfqe!;<q!c&-JUls)>+rR-FK_6(begYAq9A@M_OZ>3pAGKc|&t$-!i$0)M$f;6n
zZxxu&w8_+^_iZB)LNq~eF~B0a!h_B~@Y07)0%-8w%#iO{x!HS-qgM{0{B>X|Rh1db
z+Xzhk)xIs=ZlhfXqRv-ne)eBYMUG+|LPgoo*KeXi^uF#^UkEvQ@l~n)v_+N71%<43
z+!tdbJzN5$PB>KxD)<Njnz_}P@5z_Ot1~^Ok})Cy4j+OH({`dkX`QN1oa?2YYE-$u
zlrOizT~|}fuvqAk)Uqv0tI?J9IFu4C$@fxY>a@xVx3t{I)t^sjNO1K(5vpzc)F<z9
zzh|H%P)xAi2+;_#NjlioLE<+!Q*l=Erh73}Qx5Y)1}~5P-lPAwIrcjkJSuXtDgW;_
zL7bYfNM@i2IH;hNmXIjkCQu9}=|<=^C#C}wx(UstsfD#khcUEzIh7SoeixqE^-Hmz
zL=SZH_F#~K>6Av^Evd?7sLU0waaZp}Fdi`C52Y4Bk%EJBD^RN!=a~|pT2bqy0d7qb
zo7w3eWbF@O-*2o-DH%6`oRA$Z(`aB+Y9PrWb^+#t7)~19k(g8(14q+~&C-lb(s!C#
zqUNB1+;R;s+?@f6M>GV+zel|A6M7pk@9IUOMKwt)7`B8HG);X|D=|QP;PO7%HBiT9
zJ^#+-@SKa4htJe-{-eP*!{=Tl{aIT&lIw-<RsZD$@C!$QNZfNgd*S+?B@%A1oz@JR
z^qr-OL3vhmA}x#;;)U5E=c(k#18mOqR}oEt*qL=xJ$IeB_7p)O30vWxv>`Hha))0c
zZcLyPUL%nilL01^b6{AE$H;({J;Ej(3gG*}*n`5T{naq<>>EdjC)r*@F>BO{Rksw;
zR*nK)J4A`9JRW97r@`*Sa|R3og?54lj~GSq88Pq~(GFefz-AcsUMdWO$mfKm%yuFf
z+XwY1&Q*?5WHC-EP|HnWAo%5k$E!s!Y`no3bKq&ifvnGvZy#j3BzJ+b8xAVce|}V+
zD8p_+!nuv>bcgIx@79Rx8yUg+FuNa}14NzxOnJAREtg8u1XrmolMheP=2K0RwC-K1
z)kH)j)Ke>LF#=rqXPzE3!qdB2Y-M{DDnq6_yg^c@uJ3asPpBkK#{b<Q$Y7K(2>Yte
zR}J5{4g`cNgM7{vm#8}Aq;#Af=UQQ|2AHDBadDT%8jf!$(U4b6IV$W-=DL|h@WWfj
z4G^?Dm@H!;oN1t4z3th$_xgD*J);o58DIZK0`_8%R^^S)`6aJrkZ46zVbyo=85xfE
z-xmxOI&wO(+*!Rpq}eV52=qh(p-`ehyTrcCMCpPEVAGqB>dv`V{_${`c^vXkdlc2@
z5t)_`Kp5YK-%l07tzNVrqOL<!s+oO(KCbD>%Z!XdM-Ni5uL89;6_Zay{P%(@p1k9G
z|HaR2CC{`Qa}du<jcV=c>@Sy*e;?WY{c*q(tVg?J*nGP5e|1wtao2FO*alN6>x#g9
z8iQ1s5?xv?T)i9T4QCP^jf7y}wSFiUlkOc@-R)TtyOmIsu{eFJbrLWiyY4uU45BfM
zsRAXW@0%5K!d<==0d_!*#ZB&MlLVzCjxT46qAkW8!OdVqwAIOtX0}|x4Sz#UhDL`E
z_S*d>$z9Nqt)mtba~_L7y$|5Ba}G2bXyR!QFN7|w6i)Mu%UTfN{gJ(2&`t_K&{N1q
zIxY$ON8$|C;QSa{0y&wO!UUuC%4j;{Mc)|H(9l>}p2`7*E1kZH^7#qxaYGZEU=2Jc
zO(-C)f9kT@pRqO8|IL1;C;<$ej!xVl303l0upmpK3RGGrX6Ow^INME7nTpzc=PI13
zobj#oqpOf^k*KC|MZ6=6SY1t6O)CkgkfNl6Vb0%ZX=3SDewvPzrMmjz-A6%K%rqz}
z*Q}eB(duQ49!~gcy>XP71rbE^2#s|dBccfu%TUbu`xTLAbuZ_|zlQopIa(512w+XG
zR&IoVA|V)gm^?1>^&8ux(eNzre7vL%g8ei@&ty>^rf!jn5(}q}Nb?Q$rXQ5LZ8A!#
z>;A6bo3ALafN9G2@Vv7mR9di(YAqceK2Cr__vm9cR_=~KmyaI4+4S%+y$+(6ZjcdT
zz;HZ*s5AAycofPTsNV~m%3}2jMn4h?(R+hwOlf@E7%1|x)1+5r+J_pzJWjx+wtIio
zRFtsmaQw57iaO=e-9V74i5T=%ZFnWRRto$1p_*)EafPuC7Ds}%{~G<0*2f;_wAzZ|
zAb{Adu#<jTKjUwL)&KHAKg28(djG-tuR_ECl~DY4yCh#=dA>`!>|=3ve?EFrrE%Fr
ze}7Ee$1Oo{TI^deMtI~z>S|TehNeAVrKPW^gW*!&6n70)XrC;~dz3LJJ=fx7>u{oC
zbjdbPAYOxs;UZaRro^M-@vtOHj08y&2^>wX-!76IlBbE40Wp-3<Ate5LizVny!uni
z$_V|KRu&(?FawoH=Str*XPgR+S-ummOXw7yzOPU9HVB<h^ci!WfXtfW%8J;=8)&AR
z>_X=tf7>DqiwkzRFX#X865aO0qbb?6*)yAtgWjL_*F@&KwEc|yz$jva;N*kzLA?rd
z-ud{!H!_zuC*frHhd~zuzkpFjJ-@9RA*QMSCn4;9Rh?7iZRKj>H`psbOY}&}od#P@
zUXy$+=8VSEL<ZT`i{AnB)Xx;SEH~SD^7r{zanFd>`Ib5^t7CW+7nnyWpLP7aOq5c0
zzp=wmCg(5fv-D$pFW5`P_3-m>F|nr>kox7Y<}jT{V7+csmlUuqo2=3OV-t#S;@ptd
zvQ6a4d09@lZ+lla;ayrkV~U;0b*J`x&2ozZ?Kbbs<fZtY!oLIp|6W4=^KqswV9Av$
zD6ag&1`|(1vsVaz!%LdPS%0YNCNOSBEaXE+pDDy0Phyc_B6pOhj5EJr@=4_EGw2Ta
z@`x^<;$&n?EbU?A+Wuy;dxHbK5?;Z+s@2~eA_i>$G%1|}$x}(SasVN>o*o>g=?;CZ
zLx(aFL1=g{4+mxBGtR4q?@v7rWw;O3=$<p3<p~CMW;J$7<w7E_|M17+9bkfeDIHAt
zXDRm+1dQqXJ=6hYBg5}ShK3;|>Z{E7y1WhqPv60-I!pB(vph#Q@9zCt1Fxk4YA7r4
zTJMQP9|v_V5Rpzmik|=Qhg=7kMp5c4<qZGBmXk!K_BmAfgH)w>zlT`#a!vUr@;37~
zQ=+HunA1MZ2l=n7`HftEBj*PeX11&gf?9qqf%lWhf93A}_g4&Pk^!w*lbW9Vm)|vT
z|7%wV{o_{#dGP&@%Vyu${>R_uzrW_c|I+{e>;Hf5`hS&{1Tj@qyKjL>mI>Xr_!3&E
zT(~YL?z8;0Sgo6G7sURF)&Z)cVL|mW4rtas`v1q}@dqE{@B!HA&+7aWMgCeM<oRg5
zfc~=!@N$oREfK(?s>ife$6%Od%`cKX0p79AS+Sx*G<&RrMyMbUat8_Y_#{SAzQ~`G
zcGRT0&I9uam<Bt5LrGi!!m;JMF>l+8!=c)6rkoa)=z4&+u!4sWV1G6fKU0ivs5r_M
zEoIjc(n6b{vj-%qu%Rcy?2tNh6EFsBbzeB|g9e@m0t7nWgECD-QDEBn&CXu}-v6|A
zS7X3-FugZ4^4E6o^j-i+>w9k%kl67t)he$D^lM?D4;<Km$Ty5r0Hs$ueg{t9W;c+=
z$OR~`>RUTY$w1R78bxuR5vryQx2sq1fm59uXTp%%@rQ00#6J0D{lJ#9@xsv^B>5ph
zD?PK%cj-M35?nIu!Iy{nfn-d)glf<ggaF!G_1uFmRMNpUzpl>zpz4T#0$~m?Ww4q<
z@vM*b3;LvX-2k{63h^2PeqMpFvrs4zPPcuMlfWPZ@<+r=2wF7;HrcOFvS@RD+^z%L
zXESINRNW6OKcuvE*@jY>F^i;6r%Y&BO+pExNbACS(jq#3@IvOwwjQ+!qNPh6Oz}j(
z)n(is0L$oYg!$DfNk9iBr)g8%iUj<s+Ynooc?H}pGeTM(!BVZlcRc!2Kn5h#Is88;
zvRQ&9WZb`m=Yoll03}&E+5#D8HzDh-pTxHRoGyPXvj4UOoh87w#4)?3`qz?UPP;~`
z0;MeOg7**{D|*&3od*JtGH^ijVILC8pRa7j>4L~vlW)V^jexVT!)HT?Zr1rkwS6z1
zUs?&+3L3Z{PrE@JyeS7AqJ<DBbdFri*ZTw{uTDAsgyJ?KMpiJuWpkFS&rVzXK)5Q@
z0nS0eq<2$>&p{_X)M}6`S_rDyfVyM?7!7TiR5)09X`%WT^JIoV(#T4`o-_|JJasO1
zK?q~#7d=Rzgcr$H@<?F}KQw`(@nU~BA>$(ZUM|R=n?W^=-c9xosRVRsnr^_AR$Ez9
z*umzyVsQ951F{{z>ffvE^IpC$`jZXJKb445CnykRsOs!MOZj~lP-%zIO_k`G1>b`k
zN0a*45sDAx<g5kk=GyNRm0zHfv>DX$%f>KCMQOiq9)T(QX8RTbImOp+F%&+OZYA>4
z|H$^?@%`a_&|?^JH_1Yn6*&61Ld<2ILr_AWm;3l!p`PKlG;5w3U&tN`1LnlC@{rqa
zkBR_%z#1H*71aIL(C50uwa><y5~rh;4J1iAj*rQHIZ(Iwe#;}26uQEKIAoVoK*Kb`
zzF#yuX%|pew949~@AAzcmte#9b3u*OQ2@qF#kA{-%CCpO9DrZYdiOTnuf@AsDD+D}
zzbB#Tb2QQnyq=a^(tNO<X%kR5y!?!!pn;{sST@U<bCaM2O5kPW2NbBDx)8xBuNfde
zc&Qb2LC=Hm+hB^!&edT&Qf?{2D#s4*0@cmnt7~zNQjdlsUi4;M9ey+fD`rQ2ckFS-
zgWP2E&R<~ML-s`M(%J8U0eDETff_8MQcNd}Ub4-RP2FQH7I;@Di>Ilr>Z6XkzB=Co
z@j|`q=W_+=$!t)9SjD>&F~FG(s9FYUidP+#TaR!!s=(gy(1-AX<m&*(@2=}=AlR{V
zzYgX6wPSHTMYG?FSjCQHjYb|0DN}3KwaZvEVgR>D))e3=&L{$!WscXZoTFE-u=l9d
zS&IoP0a@mT<~*oOxXSnlpuNX!!wA^rxb|>wfj6w_A%zyI*6>n_BCzr05Ld2V?4<qF
z5#4^)?IPw^UW+#FAVaYkgv(JjfH1z$JFed)0vp-hiq5n-GCGMbV<g@gy_Ero1pZz#
zzF348M^Y@zc>mN|{{u53q6dcxW1Fg4`k$32#>0RTG(Qk~={HqBdHtpctmsD|+zRls
zZ{o6-J^c<Q*i-YPU=qYSdcAbko)5CCH*+I->9-fDf1^c(a}nK7lNh<hwEaW%>mb>h
z#Ht!`6~@LBs12+wXnr_%+o_v?RL8!l2%LMh8w-ey@9i!NZPA}^>jA<Z5&+{nGXy?{
z9-9!v0+`zgk$2X$rlbx7)%bPROEP0zY9P=VVwP^B?g56h<Pwz(rDkO{_-WTY9;3MR
zK=$UdrFxlDfGjQBxDBPCyH!H0wz)O1$5uoN;oQW3{q76a)fG5>YpUiThHNBU^G#O4
zGqU2_a_<zj{WnZ&AMJ*@^ep!B>Q(}6WX)0c;lp1MYP}G*6`d52yptl?J|*j)c4{EO
zfO}(fbAN32q4iwsL3<R%g5(ov$|{MRn>6_AZdqq4#BPmS)@r+BEM0&>+cnLbU^f+S
zw3f)zq)jpUG?%Pw2)O3+(7b+0Tw}Y^68(nl><QbQ!C0erE&w9F0cG`FTvZ9^fymff
zce6NSj<k2jEHhHV#va!r@s2>W-`pjKG2_uIPgf{c69C`6$Hv&US<(DBPz7+6<a6|{
zYzCZ#bARy3&nL##Kj3ZT&7A)@M)tp#(xuUKHmWstNTPokypm#0Mg6PzX(bRcgaVfD
z#FxtX`P<f?qguO_0pnE5O=qVf+A+iAZZ&hPrG9C(j8I{6Mz>TF$YiYSXcs}zSkts%
zi0#Hq4WT3Lj)gOo--36kRKh8_Nf#=TuaexsF_qG<xGN_tm-y<@9vIZQCJ%LEJKSyf
z;koh0nY%5hZ8kGw<?E9EyZZSM7b<~r7!vXSdPF;5%u>7UlIH;9qJZ*-EBjj{h3S|{
zO&rF<uc32~Lo8C<yp+lFWl$q*8~q`Xd<qK>p9htZqY5FS+LM;^L@NIXMDEtk&%3oz
zm^+ijv*FNtcVbFoVNGp=(#F3()_Fm7A0X*Rb-Q;QzJ5r1sa?e!Wi3pag{?_G2}A<T
z0EnZs_X(?&pQ0edlnjCndG;lmxQ<XrAFv4)KL<@9Eh^$0Q-%m&qWbOe(p(4vd=Gic
zIzPTX-->d8wjlZ2&44-sHSE1d^a>&nU$q&qa5nP^VA0fKKc=nJWv=#;0Yb;_4HPEq
zLcA6~%jr&p0oAsC6LyQHEd`LYb-U#s&|WletwJf%HGu>MV0Z;F)yWj>Zx4{WI00ve
z+zl!M?lnLVU#3H!!cV}TK(mc<Y+L#XWtzidz=kC3n!mCHPOtLFCQ)#ptIyML4zqM-
z2nlIQv4Y)5lTJ~}KrSQ9U=29ZV}o7#^$VUX+kqzpuW3ZVneQ|u<4n-SGx0#OnMhe=
zAmm$k_!=d1J815v(R-j>ycwd7K(!`<;c%ou5BExF>wYOr_=`q3|C)kjR~6B(b&bYb
z93U5h8kZoAz7rDtd$Ub|y7v0^=s7o+9+w`iYikYlrFND-3_swWkG`lg59ttoBS_{J
zl$)Tlod<MwXKu16f>2u^_L<90c!JZ7N3L%f#`>k~^?{jUZz#_Mv#AT<^9Os<T`mru
z0lWKkFGCK=Cv8}Ropm?^r;rL4h-|o4l1(+rl%`utjB&c-VK)ZND+$5j#^*K?tzgmV
z3YIhWp+ooyEDZ9b(<`XDZANYM)y5x69hqNS7LxW&eBcsX6>;-(NY!;t9Mhvq83*f(
z->z(7Xl2(2CjwXvG<E@xUyD-&ijf`@P#Q{vp}UY8YuU@qhK}T_WXV=93Ab>Er|zSC
z_Ghcf0L3sLX)kP)Mo;Q6%-;-Dot^r6r|D!Dw1dcdgrq*OHQG_BV1K^*z|!n%AUAFe
z#L3hI`DfF&s3H&F#v0&9+F+q~5YeUK@qefBs<tzJ9DYG47-KCZXn=>=1@Y7qI3?+Q
zTcX)G@Jb`QG!L3_s8P9fx{YWfoas)Q2Vv^VRWUo+rRgE!`_br>XohHL@<n%J>OM6o
zpPW$Ro-8K#yAw2Bk35$_!Ij5~!f>*X*$q>p4tB(<f|VNJVHJdn6Xek$r5Hd~I@7R>
z3d2e5ekbeomk9*Zz;{*5@%>FnQ8eOH|D!4bdf6&>QWO@_cvT^0YcyTJ+AYTofsHJG
z(O_(%K0;TyK}XCce;3!ej2lh1MPgbG7J1^z6>0`nuGU~%F$`U5Drr!T=0fUI$cE|!
z`Bt=OplERix+lff+w-S4Vf$1d6e688_?hM$;)y|Pyq6RH+SYNub(8)_AG7%hrFsNg
zhWGwsGDqX;s68i;8@;<1WZTAw_PER^*#7Ww`1(2f(>^e<o4+JbcNWHb7t^X06LlJX
z26+)w#wd6}<=m7uXP{Y|K%4rR5ZWXdilx1vRJ7#CiY6pW+JO^sy<CyHl}s?eO*Kq}
zzFfF5;<l`uow9eMm|-dS>+8LcQw;TRO7vjQ?gdqY8%9rpS$E7(im4XZoLT`w!<rBL
zPW61FT3aH@CDy680?GKnfGg^_is8e>THzY!S}C~9!?IJ#!XFA~5_`jV&aS-Q{^hCi
zKgxnXD~6y@|A)ltfA?;y_fadDsX!2B`-j|;*1bQpR=XmlKM8wCEO=Rk21pG)_i`%;
z4FpRMkg8{~cc|Y{&{s4eRr*S7D<QbGulyo2+Uzc8Di_NH!k+gyQLX&|{5MQ#cD}a9
zN`V)}h*!a4fNfc3J6`$4PtVdMk{+Z2Z{&S20Rd5hrZ?q?Mx{#)$jP)!;w=K3EJopZ
zF)NMs`Y*1<R)kac(CPDTcn-7(`rcBs2@pxWTaeeqjM*!295awV^Zd>N?x{er9B&uz
z0BPMrB|avSO^8>zw?(N0Q?V~|C)@1y)ZJ@}Fe$`-@Gdffdwn_FQGnJYXS0MU^(jZH
z#n6VdtDkS0C!U{^<!YFs$-E8N*!L`q`Yq@NFYdnC876V+yQO49uG9gmHWz&w8}ncp
zZXwxdfe5)UY{Wbk_$=!?h4zte1)j;AYTq+C2dJ<woV<%R1$Vz*&x%mNNqB*wK=)+H
z(r69*%tvBx@EG|c%yPcGtY^-8H<Y2sKW_3vUnTaSNm_QGC2e5p`SQ)9rDLeaP9?8S
zEOQy0S+bjYa@%x-wT*LvWg#ECH}7@JLvk{~UURWfz>Mf0GAHGzmDj5@F2??q6;WL`
z6GYi0Z75pN3p$;>xiIR2>^^Co^yNPotFh5wi`;`h73LA^#4J5qxeTiAh-SRqM-5zR
zqJ+E6H$#H+%2J93Pg5*Pto&H@;Tv4bWrW=bdlUT5@0Rn8c?$5cl(>hf?)~Sm2`nFb
zl}vrMTXrcs(G>STSe~y}>~P8_{8nc2)qv*fi`joSug$#Cl%d7!v=>o*J$f0ZLRPaK
zbpwiDtqFXH#tKdspEv%|=$iA!b5sP%zVVe<#)KCSLDfCLwsC@5uC?@;K$%##3vuxW
zqjUP@fZ<F#l4o9~CEDz?VB6!i_-;G+D)PMAlef{=xYM=V4`eoEks@d}!l?<KrsDpv
zA$F;pBq+>&#Bid(AbJ}7P!jbDhDr!ZVrsJ2-j3>9>V(*Emu!pC?{o+vkoIQ8?bM|0
zC1Dm@y&q*}+t(!$p}`Zznr8N%xu9agN-yjHsV*H7P6P8N!G8RVAv~%ZX~3)CZ#kBE
zZUd~<gYJr}de9Yg;|39N1+!ss9lXzXec~66MLh7SIxI#8Ay$fYYX_V;-h^lS=LVz~
zZ`2CO8_8ogBHqBf@Bh3F9!?*vP&1d&bBfZW?`mf?cOnTcXgm8VvD@5A-)fGStvy?#
zglv!GQ@gCPq&swiUEz3UJ>1~TEzuiY)~h#bNPJkSU(tj5zHbk9&+K~sYEZYN)8r}S
zeq3deWS>?2ii3C~=aA;?KCYi$`q9Kwp=9L=BF<2Fxn3dB@QZcSgB&n2XlT@XJMpI-
z@qgM|0{nH(CWuVX|ElCx1yD^P2GcZvvF?DiChEo>VXaD3hd$LXe+$p8z5+Jcuk?w%
zuJ(MY1(U3-Dz9aWa{D;7I&wis=PVVA15P69=BJQqzYsP#EQ=xl&$ffeEgIF)z2)dH
z-AKZ_egLrG^LDNoI~V*@*FrP9aj60FZBggF&sD6mpdf9STdP8A%i+}N_B3dAdJx`a
zs)L#EQ<lT0lwj`VXbaA^6;G{Sk<DaaW9`m;hp5^uR(Us8z_xUYR9FW5h@ok;)bN%p
z<jDOa$f~nR`g&8avrOf#&$k@Tm+yLRWz_1nZq2GHurF<=Z0K2F0^m;3Z6`{h4{Q}#
z1G(Fd6KHl#Y&ISfgWte+*y6Z_M<b!47J5Yey;jsW<g-pXMM5`}rdA7pm5xi6E#QRR
zNSYFS!=&Q?;(P3eZ0IV)8<X%%%bkc^LV>53NZbsC8N6v)X=nwc2j};Oi^7s;tIYVj
zV#XN`mt3LuzVX<%!!E1;v0rj*;XA?3JdG&*GTo`OQQSqp;ZG{;N9RZL<ex6Vo(-4L
zb2IEZNyh8Ovrmz)?}0|vXBPpNrctesQr6%J@?ohV5NhI9@qMIz+(d<Nr?v?cqoyC%
zSEszovETJj@zXaAx1K0Uozi{^z=kI0xA~rFk#Xp>TL%=P)@|rI`>8Oli)DdEY{s%f
z?~47{D;&}uoiiTw!IcF^Um%alrR00NC4=s!+;^g^55HN)K1G;j&k3LHJ2yX*n-acX
z0sr1N&A9`p1g8p@UIIRThQSlRWl97=)-planTd*j6<m=}!4;ljEkYKa;#Fg?{W7S!
z--&=2N$XU+h9qyXTn;T^nT4{(X>@qp+yFK7RMKWBauW@W2=oitba-YcU1TFYJ$B7q
z1pO&#Q~oqKhB~DN+d{f;Ew(q;(_hURLSks9xAv0LP9>P#fls(Lq#zMJ9u^~*7o(XC
zu*1C`t_-j0#FW$PJ$ArzZK9?_38r2asxEk1kTM(V*@P5Bh&r)u8DW&l6Mq2cC>PKx
z7cPX;bn`q_2oyte4u8;7UXHsHBeLHSN~hUr1XNTm0jNhFfF`Bf`s^g)I|)ieHRBDl
ztwP5qgcs95t>|0lm}zjG`^GV^!aIm}!7iBPRtv)zVCsLe5h5Gd$fM~v!*Iq$*3+c3
zOv;E}%!ACYn=R~ijqy*u8r4a}aQaBqt@`E|_plo4<LYFtF-QnBF}!tdY0erb=G3!I
zmwp{zFw<~Qcx`8&pn^zF>*&>9q^q3ADDEZO$$f<c!)cc>m$|VxTB*De|G0SmJ<#(!
z|2f3TJxZ%%zDb|RP3p#a!de=w&SOm(j-)6BaNZC9ufPOGYrQ9ZFnMzL@$z%CXny#c
zmQ-H40;wp4T4{qBE3j3Wzx4Lu2zk`U*3Tc6USq;htot^D>;iy29#;r7C#Ey$W$$jK
zwJeI}HGx9Vm?xq60m~OWqJp5`ASHDDJ7^d|6p}eAe^I~~`B1AG?pwRyl}3NQuY;Wq
zs3CZL%@1?26X*&??43@4u)c~_e=2!m-Wq_0xRiQ*5=2(jpM{A=syDhnkeNLBdUyV<
zQ!j<mQ<6f!zp)@40BA$O>dgI*BQ~k-UAl`Ck`6t^h&zng5mk3=h;6uqp=KlcNv;qE
zYg(5~DbDiL$&byx{06mFl^q36-Lq6d?ukIprPv)Jx>?Cv>5RO&Uc#k6R#r@=ADAis
z0Bs*aizrg=j!R-P`q8c9z$#pbusaVJG1NVhD!c>YKTA@#7BOICBUSFVxs8Hr?f3(v
zc?S%8PRH*+5DIUAVRj5ZF0lCcc@bA&^N7O-a<JOmM%of7y5El0u`usS*up01meiO6
zR1(S`z4@L(lZb>)gU5BbB1(rU&XhA@X-=iL)ttO(CLVCGtl@TBX?g==Yy@_EeeXQo
z1rVRfN(uC`1k~Rfa`roWNYwJ7FeJkaWkk1xl&X7kGrO@%XFF}HMMb?{o@4F6@|&+f
z7b2d2g6$A7!?%(1!wqgx#pp52dDaKYv*P`BP22f83r@`Ey`08u{BtHIS8?Em2;c^*
z)y}mD#^1S)fry?zOY-#tU4IlKqiwEOq5LPGH8sl1y;#4^1Nvn<Gqtkan{NAzu-L`Y
zk^Vdd{pH}qc<k{sSzA)8OP#h=8!mz`AN%q*+rTC`er|uJ*n2<T(i_-NOdatYpMF#J
zeO+H>RSOQodvrd(nTf&zBxJ9-@&9|t<B#s<WJ=nz|71Gx@@%Z}r_6JwNp&ZDBXKmW
z<@K^<a$y(>-PQf%>H)SblJo@E)LI~lGND7JgZsq1+-kUKO$<|{F<2A=o=e&~*DTaW
z5Y=g)y;xS#TFJ;T`H2i30_dROqzB=X)n72%H`P@=+p04-vRFIdqy1ns#{F)qoDd|^
zJuF{P7W@Jh=NqwLI1ybL-AR^c@mD{f>K<lT0}o^5GL8_ki4Aq+wz4p6tNb4LZ$cPt
zY7P_Ghm?_$P=uI!e4Ddvjonp(P#sXKd;0W^%`u!<GpfowHBvvU6XU{lIDK*Aw$4IV
z^85pkoyQ*avPWe`LI7PiQd4?Zg&0~r)8V3>Rx41UR|ohtuS(V~F<nOU&t*!n$a+Fd
z=R7RUIlFjs-je`q*|zjm5=ZX&ICP}t|8b<5;#_$S?C^vcy#4t0)3YAbgBV@c@tpNC
zp0;_A{{Xl`iQH#uXQvEQa+zs(;T;5V0yy<nS%+I@YJB^H7)GLPNeHo)#X4f7<`A^A
zv@0-}Q3(0&c+FUo^Q{?EwY#T~U2x0EunD_vey>>=0c_x%F#uG<O7Z`mIsdnr3iwnJ
z0S41)(ztr#fBKxQ+M#ML7}6dBwJW<Kb6~xl??Kwjln)jJ)Z8?q=McRYxiS6?QCuAu
z38~Dcor*4k)Tq#|V`+(A>^A2)^_b>xillAK=3S0+0#+YFESMW~_{ouYTAN?Uo`{|<
zLODlr`QyX5;a|H$K9+t>Rflf;&G*+wEn%^`fbJKKaFw{J9<E*fVF=v*X?GAW6tzrP
zZM{cp+KDRSr+M4@x$oK!Q*Xwm8IX+pl>AOz@oX4Pr_Slf+XJA=9+$ybMieb;PDuD@
zo5ag#lU|szBkPSwL#o7<>Vdp$n56qPKbt&VjX1d~CyfwYK^;14uX3PRjQLPpESP5W
zz5>%kTSz2z^>nGW8qbv@S24G==q(IrrURv=+hkBSA|lKO_INZ2?F0fFOC}!p?1Q3a
zD6YO5dFa}1p4H#QqqKIb8~oS0n(u|YO-aTV?UN0)#x8H<)Ua&+5}mU(F=&V1&TMH;
zeF3y^etEs{4Y+a4p06b^WVv_S8dgG!Wkc*aBI}_N$PKv?SR{&6(g{u4l(s4fOqb1h
zj0b(exq^!o0M=*x31$7E1Ar^sr&$RvqHjMKL{C0iXK-S1!sKkyukbHp|DDc;0<bMi
z4J`g8;X4btmh`kJJcRsalk%t#xw<*nJS(C0bj*bw-XaauKMrnnSZJjv0)Cixoa0^;
zPFj9^R}yw_G8?5!r|PQ+;5<(~ygUD4lUnjFsUy@Ibh=nxr=?mcVSZal_sAF6#^8|g
zKyD!%iY*~8^s<A$;<EBrQ83hNzIhzrq;(=27ogAprh%pXbaW2cy!0=dJyp)qMJcQ0
zTU&3h=^F%~({V^da8h>QBITz#=&nh1!*8-#Bs4mUiALbMhTkhar_5ObB;PNI?Erq)
zb)g*e0O)nqiv?L!o=y@JEyA$kOwsrGP0^GBz37qQEwC-q1Aovs(%?^mL=mf&9FrtZ
zMcuF3BLJe+BPgViZk3z?lB1nb27*}wY*>MZ1lv)lXK<t%`J&dpsIi4|f^uH0uNFYY
zr!)6DP!`Ms&|14Su5Q-KYc7N=1`Mq7k{#~lKPR(q?YPv?^H>Uhp|EZ@N@qSFpdick
zDL&Z~#Zdm9G8#FS1Jm=&hlY1?KL>g!D1kAPe$J)pPe2=AO*DNK%b(Dhz#msR+UoIE
zO1FJpO-28itnf$9EL$*lG2|O<`cyki)El62-@&o4llN$HSYS~6SH@Px{^X&D2d-ak
zjCr4JW#a~a%*zduOQECSJzNxSk8j5bi7e-a&VU^-dPtP(yFcFLOLknBNWh5#<Y<OM
z-lYMKA33vs^}IM)sC(IFw4c6=;?b3V=o5}oYMWG&C6;w3zJR#XBCEv(lx)>c@scZV
zik=u4Uh07n9KDZM7T5bpi(9p@PL-9_XQfUE8*LV+bw;Dc6th$oBlHKqcu*JYDxyo2
zPU$`ScKUiHhI}C4t3!cU)B2B`q;=v;_9cPXdNw4voX7lxi14!U-fQpKng=g9VX+5+
z*hvM|>DIAA>z|OmwNVXXOLD*Er*VD9Qj`tcqRsn<lAJ-nrBOfEFLe`-Q4ohw5R35&
zx59uDA|3B;miC2$AtWE%C6%j-t#mkhs)f@Km@;57M&VQ#P<2}=7ljOQjXZyb{(=UN
zkq4I%XMX`oD%Z4cLbsmhl29!Q`DK6zEdqH@XXU2};gh7VgRs2<wK(&HY$PR>SD<1?
zv5c4s_*UdCFwWvKz5@QHIE>eDE12)dMv(*OJF5{7Gg>xyNTWU20?|tINFn~L0kYm8
z@>f%EWK%F!6Pi&IX;TpPue^eC^gRQ_7X#d-h8ea2LC(PhJVsO;Go^urzHQ}69rMUR
zLnsdg-|QV1+?@_il51dup?qBj2#2<y>96~(p>pq3zCr^@2EDRRn-w8v26W{`5D5Vc
z3hGKS%HTBbixi=ma1oHboL_caKMFjEh%AGP--U;y8=&?WMDD&3ETx3Z3%KI5cMwme
z08$=sp%`uu$Zzfr3M<dcERk>RyhDl$Gtfv7xcl@Qs0XZnz~+aHrMOVc;DBIoDY71j
zefy#h4gZI12h&l85Vz-{MWO!UfHa~>=L*A#gj~@drnpGhEopN1ys+Uc?_)rX%sXC*
zSAJ#-vnCUC=N4j#C~TJIn@kVQ_y&eL942C`#SXpTBaQZ*EFayqE!zPR<QZ4f3$D*^
zo7i4@Uh;51;hHn#Qqkmc`|jHifyrvwZ3j<Vxh5QlO^*Z}fR5V7{gvd?cUK*3N*?Fj
z8wpKlDZNc-x6^%z+}+RlBn-G{&E~!cn?5!f_96;JBQo{$Kt}NPr{}&vBZ_mcrTvn~
z>N64^*zjS}^(87oobql4<J-Nf;GY&o6pft6Oz#qJUcS{@+dm|^+)zvT;G4$F{d3nm
z@EwBTO!Ul;NEUtK1&-;Vn`z5{QoWH4k3zOy*cz+icXGvRJEqC|mVr6;r%TD0M0H*n
zeS%3lOoP?+n{JP<UlJ&CLZL8p?7z~0Wl3H7py+Dpr`ki)sGj9lB6sWe$G>SRUKWuN
zzomD=J}ZTYIgoJ;6=yfOaL?A~*!VK7w#RMK70KFGV43hx(^q9q^2^^oB-qVtG`h`#
zW5&%3?-zrUG$NWC>?A7(cwM<a25O}WbG1J<7Qzm?f64BnbWdX+d4i@uG>+b8)Jp*v
z6>3+r3K^~5_bFE=yV31jHMmgnWA!&z^okS@p0}wi!@Ik>YfQKzBSe`}0B<KSVzT}Q
zyz*(^DjAD)o0yZtmJj}^ag3>Y^dBKZdptkP&$oEaukxbSB7Zl^xIqZf310`Q^886B
zWPFGk?h8Yk1`8Pyk2=PaA!yNjvARaJt=aX9IOvQ=n~*>%v4lH{6#;9Y4VTO`AI%BW
z62k=p*AJXuOHSUkJQ}a+#(ByW`Fvd0r4k$wEnOJFioe=_X!&=km>AZQ9;0dTR@Jhj
zpt3auqG7>wg3$1`3nGPUtEA!#u<>!x^rxoz-22g!bU*#7Tlr;umEQ2sX{EPYslfY4
z#L!ba@AFJvL3)_)&I4eb8dWICT>6Q7yoUpOS^oq5a~exiT+Uwv3sv}fvh;U%9Vy}?
zle;>!JIL#@d&x}1rG+)sS+Rs31$h{9-3j6f_Ye8r!8d?GP9Ll#&F7~T-4TaYA)>6;
z|3bR?wL5&B{Vbb3{4HryajTi8rin0_3XIu9%Z{K`$6s1c{E5_CaYWJc4)z#xbk6?x
z_sks@kT0|0uXFsZUFx&>qcMHABrFDdnKS8m6Ggf_WU;Em6N$RAN3S?k1_Y*vIxt8F
z27l37({!t@4$(@5CNAES<}IMcjw~5bfAQwq8!prC{>I$JjYk^^yPn{vtDW-q_y2^{
zEbH*)C}%6{mo(kKGv)=fv!{H|ZZ5=FOAaoH%Dm3B#WF4`di3l^;o4iksKq<G^DW3L
z#giP@xWqm~tBi>(gM5Qab~95M;|BSxPvEfDm*M8uXD|)6{T#m!X9R0;27?cSr5)FD
zl55D*ezK%m->UE1sa~Plv0j&8Fcsn2;bDZc&Kk1#T~dAF9qVr>9{)J*AzW;=m9Q7)
zoFadJxz=o0!fXAGMmP7A!g%0Ghm!!wav{VHVsO+*W%}-C(^p#-aGtn=*~i_YvHffr
zZ&|O6sZXVa>*|{SDU<%X2kDDN(G2ByPL?U$M=!7ji6ad*5kjiu$7}}}=n+xo=uRwt
zQ~ueqmWNb1_`y!g(W5xWpJk-krfhAdvQEu4tl%<shY#-hgb^4q4eD0xzCeGUEIOOw
zReh;~xsaKJEPYeWs_(8&n>GW6!B;vfH)+U)MIzZ!Emy4(GzT=Q!XdtkDedlAyOX1I
zs<q*LJ3_vFyIu}}LwCGFO_(XO&3@_H**n?u_Pn0^EcSPgmNO43;gyh9SZ?*7K*e<I
zwfHCQmN8XdBenIyiPp0D0x%pRd<c@mOImT=^fHN-Zc;=gH@?Z{Vm{z`<20%Y!#WtK
zbtmY?)xnwk&h)|pckB&27E#>^=e@Fwjm#$u=o2w@Sh!WhBzoTTQ@dJU>_rBp^oF~r
zrLvhOu92PeV~`)jl6@bUkPsVH{gm}cubZk1*Lgh)uTI>YZ5MId*IWHw-J;yGI2w)P
z>m0h5M>=oTt(+fTKKgZEbd7qV2d?;9z5I+qSenW>s`Q3m8>&zEW~=7oKy2^*eH<Sw
zT%60K*fQC)xE58GJbGKH!1rQRRBh-4B~_#h(i4fCo60kLkKfx4D|<allyvX^xXYYJ
z|AL9&W`ZJA^wnr4PK$m>iuj49@qK`Es=erotKahEsCv@ciH%=L!73HP@Acw0(UIj;
z&mE`rkG~#>4e+l$Tb~zpLSJF9QnNnj{=PZ1S%>#g8^@=oJO_DV>)c>1dnfndQPgL|
z3iW~F&+4w*$nOd{gCYa*%hS#oj<GzVDPk09I%G~N)(3cFoenx~U;2!e`FhiE4tS5T
zL(H$)cFdGUQMZW?*kQ(eKBijw{#>AI1g9^A;lt7~<%?qr--0Y%na0LdeR5NUE6-Oy
zzinseE?0L8=01-jNICbw*r!Y=ZV~JKR+?x_ye#TFW3nQC7PK(5L(xVvMdX!Un7g>L
zB@%lfIA2xKeYqy`d)-Q}xHwYZeD!YmzYr`y`AHS`KYW#)EF)eGY*cWE`1$8}^Nel~
z(!8NxlZ^N(otRUnDU6YZc*Ymue^&OGM)(n5s+;g5Io@K58o4yAXKKHpm4iNaSO&bJ
z?sw<7e-^c~C@P@8Nqu);h<@m@4d(Sa&v-xfy1kkH+?s9Y3%nb2&ggv#<X4z7MYk?|
z;Pt3IwOY4yjkS&~##ZHYL;m?e|Ff%hHiMoHr=8rbe_*_Rj6e9^RbZqn)#!UEx$~ob
z(mvaD>l0ci{+;zTLQK3{&-vat$=NnejJyWF;@yPaI@yDl=G*o26&qCS%ZT!Se%Ajs
z$uN%b4{qdc*Eau$h8|EXKMS;-SO6%)f9Un|1_0T9Xxc_T`EeUn?D8<-qs2BL20Gn)
z&Tjn9|I9hxP2w44rblJ2ps=)QjhwKwz$b#=y=4Cv>_;aXqtk_vWdBY8|G)ltl?{#I
z!X+tPRrqhR|DVjS|I5E?_)xI@fXBAW-tXS!|8}YV{Q7qx;DYrIKC}9*()oYh+`s-s
zypoY-uJDChBZuDal(_%#m&q)J#qEx1dCG77PU7{Sujc9~bon;6$69|+oc&L~fHNty
zMirEE-~GS4eAB7m^1YnCta1NqZU%yLupo}1?|0zE_N;GCNcNBGk{ON0yHh~`cirJd
z<E;=@(#U=PyBF*?!~&~|OZR=~-)|he(|WICe1Io-xr1S+utLH}FC(GgPTc_#a%1KD
zxjw?WSw5zxl&`DiB7+LV>@Nf%hA~+BtI3t10SgVdj_LYcp(kc<xTC@6JQRGbu+NvX
zQ{c%7*%H&VkaztQ2D;~FG$O^P6$PW5sczPHM;se}uEr*ruYI@Pzfl3GuUx+_4#uWd
zL6gI$16Z9omq0f@#?^k?DX~GIB(r!L>){>N-=Qtj3E+ZlkMGo2{W<5se{_u-!r8Ha
zDWp=a3&2`*;ebP<Frx)^Op>oK68MYyfaN3s<6C=VS>BMJ9(chy5XQq^8v|IUz$SAp
z6y@71M^HXt>LEY!fL|aC>|!E7Ou}oaR_l$Vm;-$CbN$-kLitGSs9SvZbU@1X9JngX
zQz({3K-|JSHQN0KSP3|iKX5d02Ufbxz?+s3@^3R!FRZDdK?!aI(5i49Q0IYDO>fs@
zy=>@-buw!`*u$E^br`XleiY_&cE~ct8NF9VQ0)O63bEy=eVnz`j)-Vo!HvKL7}e*U
zW5Dtb14dT;D~BM}wQ9Zi4ZmxJN98I-Qo$CmvU>*E79v5zO$Dz6WbV4MTwT%0EU}54
z_gk8lMqI$5VHfZ*G0#5gH~R8r?18(73!ET|e#si2Le?@<Qa#M?Y;6P<Qq{tKw<(8<
z8-pi?uT5F*U*69{STx*QkasUAXK!`@TShQGbrJ+PW>;?jgCp22&xC@j9y6veK;EW?
zAWp(?!tx{gWaWK7TB`{lbEjyy*jGrFg?(n;)wKRn?tSWi9p8S9^YLrxNeNd8ghd|@
zprVC>hL;5>ipaJ+ZjHR7_ctj9UN&zS5P-7ItSGdWZ3w7kKObsnTjq$P760U*_PcO6
z0m6W0px7X90KN>nfXiEX%%bMU#N^)}O>+kBsW5;kg|1?mJ$pFbngh#8$M-BFO3x+Q
z)*u~G+=pkhNQ%C<h}UiI6BXc2g*63XVoHD2#zD+LoO>hQ^P80XY;VDk$oEpDbe-j$
z?T;^qHx?%1Tm66$lK(~3ccaJ<(2g^)m2}V@8RV|P!E=!MWGcW_{zlc!)8+txtvx*4
znQ2@my}CH)3t$ewss-Srk~{eBG)A~vMX~OE-mO1<4*%2c+y4^WigHSAwST$9+tT}g
z+5;GxduKb44vd;cKojGTihL~b#8ie<JaR4?8Y9>PD{vWiR0*zN)bY<@v_LfRD;T2W
z<0q0WNXsH%TM$xm1G2{5sf1{~A}V%hEQMvEzb-cPe>oZ&D@rGPe*a+FjEq5nPKZ!9
zz64nOcnlW<=IdZ1(OL;R%dN{yjH^<BafBVu_~|3{uivO$cpv`lb76z0#m%{!=^}D*
z-&CssyB`Uhxv8Kb7}tO_QO*d*EBeSWRmeyjGhLw^$L~}qr7`RV#UR+=$Csz&`?$XE
z*2%R<->L<rO*3FFlLDbls_Nl{NTZpbova^kt&(w`Xv-ctnNCp;Qs0W2ASq&VOmVKT
zG<vH=^0%NC4XSZ0*j9?$_00bS@L>V3&>&8IY)kNz+%V4|+>HKy&h?1zWdpL5Apl#w
z|7G1V1BmBkPryRjD3Sj;AHFTy*#X8i9U<)ix)kaL-WB7zWmB!#ail~jmVnPo&8l&4
z&Mx0z=a9+PZjSTjCkjUmIoQ?y*oanxhG!_jNHSIh*n~17yQDeq;>^RDd&r555%0Df
zVq%yt%G+J?p8V^C_+L-6fC42@n~iPj>%RNr@fByf>%S|ft&Ql1rK|yqOzaMs(?Q)3
zF+|-K*zV_A$@FhPqr;CY3YNMDQ72DcnG9YhI$Z}l)gKeV=SA7rj5+!lqCTkmnmNcl
zFw8mrq-aTvrl@K}8UeS=xL}IM*#UG~@!(9rVf3)lW@rVZ)%MmN@Y981IrgtuK6FrP
zpY`%V!^63to1*qw+2(!eRJpjh!Jnkw6a+a3OUFs3XPZp+@lVJkzSs9XZOvMYFDa6X
zrOG*QZ3*!VZUh8DL9c?ab8~P+u-BH!QQfjN82XnNfJaOFW9@1{Hwe3%q)F$)e%_7i
zC%shzX>*`uHxZlI{HR(+g=r19Z5J@1HdOA3op0WcHG2R#5PB-FL&oy*E^68|J&Rz!
z-yFb*xcYXoHu70R6VS7fKoYS%@Z;>~)(dyRb0-_?5PoicVaMEK(O-bef&`vuE5`!I
z^zqLkLXUx8ZkWIi_vSI809&9`!3C%6OA-{8bPzQ1Q2v(6vIC-Q5Ob=K@XAVIZ8FNE
z@(1^T6;q{Qm^@NHFOatnxM)g<8!_E8$U!<qT11OnhoMJ|3W8i<kLNE2r!-Q}BiBD`
zoEXO}Ja7O!S9dQDJ+9b6d{wpkJ7wzYDTS#I?8vbFq_KSg{M?zUXSv-cd$KUzV;7XH
z?>nBs+D$FO$v^Ep^pxGjh2A)=I>LFJcsUQi@iXn{#$nH=x<{a}+V23%@wwuxW&Z70
z6NO-v2db7*3Wh`eEGwMvqGI_0M^`t<sRaqSdLpqoJw5_tx}<9}1~~&@v0L~;tC_pt
zK@Wv&S80Tv6D>A?U4_ejR-B9hI;Z>>+aphuMNb(ZuSMu+zypGzppR*6L20M^fbj2*
z_5qX)J!5naXui>2H+^PehHqIM{=5vbC<S)=okQ9X9fre+)?+|c8k^xO7!`D)w|Jw6
zlR`GafSCgSC?qQ`NkzxbFvblE8c)v^@V07p?~3t!GfT5wMpd;E3%s`93XLcRDcA&W
zo<mxqIiP)V<>r}5irKHHnzMtb7pFulcDXr8juM<+4}t0_Q*O%W&ZXp`xN1D<o4Vbw
zub9TcoH)ouAPgccsN0t(Gun5t_e<zasGxZTlfjeN+^U>$-LnHF06wJVSBt<vS%pk<
zMU6$cIA98EQVddUB9=uI0y0bBpm&S64a8<-u8WO}3YP|!iIM2pX))Tjs{5)jE5OP3
zwH_+mJo79THJ3@2HB}$cAND6h*b+^uMUEo~>J0&!bmVzrb*S+pnM<I)s6LeA9R=#L
z54z9JmM7H<r`~)qw~;=vcK%^&qG(cBKd>|!EVjNv5$xV|ss21}Ge&S8i)=K0a_#1}
zLK|CO4gmD{phQizl~*?PM^V_lPTvi~LI6HJ)@q-w6}1EOE~Ar8Zp&j}nplMJ*4*(P
z>Gr67bW#f@yhu<XZid5F;kBn*E`V)nT72qAc*OxE&lyB1g(-~>FydoZ1WRUtO~s<O
z_HzT)TNINb3lyvuA7)RTfej%BLZS~y#IC%8t3$<X#6`04My|*apKP^%>&i;k{{X6i
zZ(nAs|2esJzK{C&gnpfvv!n965w2e`Ft}(>ieG(+q(%}DT*7WXd9XyR?50EffE^b*
z;1zE~2UdqjU7}o!0wMk&$^0b9;?Rxiq)JO7C<)ICHjtPz6`seBPsx+nd)$5yVGs~k
z3?|~CO8nz3rVmXdf*&IC?|3Z6bD-?*Qw`AQgG(Tid^0WtQd~7s72K(gB)h=UjUqu5
z(MBxM-=t%vb92B0v0jl~+1alc0~uE2OyuM7Ew8I$s{EQAQBin^q`nN#loYZsA(5W2
zk-z}1gLd8{;HckR*aDN6o^4g_UR=K@BE;8+C$m8u^)$IOu{2!{1>SR0K-310q>whH
zyG3&;b-v1qMlDU#=BZ+MaS5~w4yVMH37;|#$ABtMjI5V@2vp)xJ5s@2h}VlGZBC-E
zX(hh_GT}TS;XKyCQ#tnNsTF%~0R5p}UVu1~llsTO9Ebv16}l$6k!`TciXQjdVbz`n
z=6iL}hPekTZhr(72rZ4O;|}e~TTl&&VmjD0z4r1{O!QnV@UCqO-5p!4o9uvh==;&n
zydb)&>Qe{7RAl7@69*wHLzxmmI^t$G9Q!_2qsn$)f=xCnGB2uWG(BQ?W|X_>p}&PI
zB==NP*J{n0+{eV~w!>PK?pV-59^62AK>gs26F-kWpVJQx!Yg;um*peshG2kk1Cj?`
zh{`0s0E18qPp!vJjEziHoN+Cu>=<so8n7k4hlOi}h{2BUj5X8;59Ya@L@)hM<h)L7
zubca3_20Z$o~6j7`(1}Z*N=|<{a2mF-wcse12o0D2*wbP&ztBqoXTK)GN9_0R6dO8
zYylG6QX5^3^Ab?fw7*f;X3;LIwji)UJk}eK&bJ`2wQ6`yG(mU!pq0)|CINxCY>8yl
zSQgN@fRZ()V~fc)NRGeR6?NDRFw|KLv>+bqnx?Vy=<piSPjsDGaPD_i?1lJNS{M$P
zlvt>s)1RB;jv1~=Wj&T1P_1f%Qrv2*TT8xC{qR+gBd4%tJdsK0mOt03c_e0jC6aST
zr~~!V6_CxwMsBr{EPT@DPPwZc{P9(-X5J5r{b;+&@c03fUJKMmWzItbXmHZYw}l<j
zMOfBEhx=)W$Knn^O7@d`#%ST!izdxk-=dD^1KDY$wcOiEy%9+jHLVq8fv%|{#OzA0
z+DjRX3awC%L_1LwR6Un$-Pu#fI(~5Z4K4)wh%6Zxd#&^~`f|9uQwX-tfa=j1oaaGq
z>%-0qClDeoLlh+MqwX5$dZF6q{l+flvCKd=;$l6uQ6LNNXn0^1{^=C;^QNY4!Q2Bg
zRXbh>5$%>L(sHhCHa<6ATESH2^#e&?rpRRq_6b|wIkwcV=`%Vf^<UVO7-LCMMpte{
zT9A1xo8UU7$P^x(cA~9V#ABAG`BYOJsPQGO1%K(8>~@)IPja2SS&WrGfXHnNA=PFE
zDyT0ZjU=km-r8j{Vm}6#&klrsy<OBQ0;0rrDCWUSb7ZOFr(6w2B-$ONTe00^<3{lq
zmlqdis(-JVziIma{=sT0TDa!>zn5Qs6P8JcSn5x3Sjh(CMy5VveC<p&XP1}z;Z;)J
za4TlEKmnx|?zazL&>bSk_OV_*!X>qA*Z*;VtREd5O+Fa5Rsn+4Vl0C_;KpS~8$^pn
zyG|0JM*6-#1lA@|z7LMX<m@b~^Sg65LrK3dYqzvm)w+GRy(4pyrO)<Cqb&U8079oc
zuGI=9oHsDLNo<j#6R>@^j(u_#*iCaI#l!C<y)JorCPnIVvDeP3HrIZGTC~pjiR)s6
z!CURJuNDKJ=~mWka?;Ol&)I=meqWTiMsHnu;vtBn)V2vFLXzw&<w#P$2v*Ioi{ov7
zo@?IFS^1?D$8UV&lewZvY@r*)W4yg}!0lkZl<Jlmt^Abd8~cKx!XjSg#apE6W4TDe
zH6tfIwytd(6+Ppb1|NDtH$CRQ?c}6MAx79}E@g3@>Res;#uR<hXie+O9(Z9Y%Y_zk
zokhILWr`(B{N<1G@M`vnclz<cVrV^eC-9##NwS0a<z?k!uXo!8^M*0dbe9=O{vXQn
z$KotLTS~>R{tsto9T(O5w)-uDgi6W)0s;eqbc%wsGcc4$Bi)TO5&}}vEiK`|C@D3h
zgn-f^NQ)wfbV^8bp0&4ozwhUF_Un5-=RZKjnOSQ+&;4BY_qy(1hiW$=oQY>m*gnHW
zZ(ZFX4?~a`j~Pz?d>yj^)`D*ZP9&gQ;``$HEd->?y;C3I2I*Tfps*r@?A;?gEXr)`
z>#Ntj_?DPcZ_n(;StdR6+^hKf(P15U)-+e-X+NbC$A9B8O?}mO-GX_6#ei|?51S*1
zN=yc04^$`=h(0`Kx}GPmmsRcf+_}BWk>}{%vZmT`0<1cp$Agvo8<?~N<ASRj7guRn
zO%RHHNYPb0gc2fK;H6d8?shi9B7gYF=@x#sW%q?lr56&t0>j<EG($?ls}OtV)<DX<
zAJ{Op7I-rr#fjKW);IC2iO$kjbrRJT*`ID}#+|0@#7|MFEAb;4x%tex(E*3lxdzb4
zrW4UxM-Iw${QfkeCEqkVP{PowFe}?m_TLBXCHcW#&lgz=n4#1%Yr`-8d6T^ry<>Nm
z)eqkXOJa8V+HAi2h;Kj4HA5${j$qU3o(_NR-7esJ6wj0H<Noz+Kvv8cVh>xlBX6sd
zrI`0@-pi;YPAz4pB5DR~B@QO?5l(74B-DIwv*|}Br9>YjT8e|P-x!NG!e&jj?G2e~
zhkNUv@94N8R08g$J4B9af2`!g?D&F(stp)3?|VcqT|6;X4d?9tY}33mk$n0!)$KCy
z0_l~Z$_mmeRMaRkj8~f<bqg>4w#^zQmBI+5P(`;~9+|%+zap~Ei~joA7}iV4=^rr(
zpyqsw3|NOBM*|TGiq*Xig3_}6Tn25xFan-_D9VU3^eyC&Ou!+Zf=6<ZbKQJc78Y<y
zK7~HQQ!H3^hw=w_v^0uuRK6(TUHJF{cwe>3tes|{_<-aed{NiLCqw?2l*;gQj@5s9
zDS;be)5`Lnn1HIc`!$cgEWGjtk|**(dS|O|!Y00);(~6iij~ep`lb~-OB)EG*r;GL
zRjwAgOD9cQ8K%V(@gbYw5$am<{4JltR*QBaIEh*|#)LTef}qZoD(J|McNkJ0K-W!O
z)oB#sUPj*XbZZC5H28{CYiKsJO#YC3Gj@hHkbdyZtl9K>+GPT4Cs<GfoQ#-Wc|0A8
zI(mLOClO?UQIp@l%Ym5=y-cgMbl)geKm56f^#?{me@*$Ssqs;-8X7aS1?Obb)4HcV
z<V?JjEOzr&?KlH0iVIo7Nxth~(2s;ae>0ZVl9}P>$`~({fT4qJuJHb0G})R^7a<JC
z+3s~HVQhvsf{|1~+#aBd1l<(oBKzA)$e}9hNC_3Kgt<y8waCrT;ONIj4(t7+&`nV5
zb~!nD+@H#FMWF}viD0EygyVOLEhCOWoYdGQ{rsX|ori#aWyb8GTLyh$WXxzn>39u?
zXZ%S+V}{f|95Xv<lY*=g@-sDnE~&arS>VY%dmoGP8ibEjjrS8#JQAZUL;eT{<`FrT
z%UMW;AI~yb`4NdeaCVh7HbbGA(y|zIl7n({N=eYO09MC$z*2lG;dT78`n(_4F*))!
zTR<orSYPj974bBxbSybj+Ot(v(KT@iFscl<u+hwrl%Gy*KcD@W;bVv&WLo9S8ig--
zUHuM_&c^o=eW=UwMB|2y-bf1BS-{#|QGOc@GKI1Jj?S%95dgGeN?)abrdDsLw+A5u
zcFN^#EhBNMph!uS=!z~8dm+syRuL%olPBH&o4c!jG4R=hEXi!|>OSLCv9W87ndUrL
zEAeH}GebWBMS{(c;Yg2AJUirM<KAPsxdPPSaZ_KvkZ_}{9~$hw+vd7wUy{9Q9+;tb
zwW^xrxLuVoy3qV=|M@WiZ4!FX;RW~km{A2i9hZj`<llgQ>Wgkw(`S}%D59^nm@fx~
zu;irm!QrfD#3|VCd1ki`M*ZdDv<|?VReqy+ti<li-D>pS7qw-uGNpXoVToi<Wz4vh
z!RTTnaz7F_QRS5J>-_PL68|4xxcT9{*ft7${dW#&{j~sv)@)=)I22@z#b(B*^g95t
zZjK8_C*wVrV!%r42acj*@&ibm>#!AgQ0bF9dC_m~R^$*x{Hq8eQ)1pT?o8A7!a!&r
zSmQZv$*MjE*Av+ng?iHIpDbK{|F#S^73h`*_ihyDL~~{^9jnS+rFaZ)h+Y~>-+Tm)
zx?a<*!ba5ENVv^pEPo7_D$65`FX=ljf$VAZof6~=w{suRvIi$=dp-U@Esu2uRrQV!
z&ADPVX{iF%Xt2kNeF4EPA)(CEsx`=;k+3QbU4@m_lOY8~->cjh3#`xt&xd)ueg!oL
z(X4^;o2=>X*|*@wYrxe0MY}V7L?NMq<ybRUqoIUbO{20M;jB1WT|FsST$P@xowksw
zS4g=Ijk%gy$kxrPsxHfW7#Xu7;A-dI474B~d;zLCMZx(Z%id4&$%;zccrI8}8P5r%
zd6g*m8))w5W}5^=75q~{{j0S4VGHWX?fd4zew`fe;v@x2-ruZ9M>!#UnY~jmXuBj*
z-s<=a$Om5w!Z?SotZ{1E+TZ}eXXxD7-ndz;uwqcbYj267E^?@oLU-FK#}%5Wr>hYg
zeD}&t_y<Z#Tm{9(vGrDl_$P~k@P17W0?X}QK^5$94z~xB41Gsit6F3Ac$R@lK{oRc
zvTEqkOzaIO8znwsZUg5k_39qsej{+>-JFSa76eiNEap3yiZQvtsYo+bxS0}75V{jO
zEC7Dto7!+|U5!QNWeHGKMF<~>#$*C%w~)v`8AlY-PeaDyIGTXI-I%>!tO4dHt)cwk
z&*7vXP+`O34V*|XVp{9S(>^-f^7BG8f&bJuRNbBRS~FzbK2YwRog<t%XG-nEMs_gV
z3A*4~DPOj;cx5|KrbQf`oPLL#KlUh=ASa2)l+k4qq|$2Gq*UyJxpD!(v-~u610cdB
z#aXLVi22i#qmFCh-ky1btHQ|{{6;s;zT9gy5*Gb(1Y1jyq<b<0%1Q&({sqx`^kxTx
z2kDX%$ODOnMr}1L7Z%lO{4X^au@aC*@z?2Y{*{EEb3`VEWg1xp;?k-N1Mj}qOg8sw
z&|I*-%uc;-u;kEtI5ksx9C>Z}aiTa?528WdbnKE~?^jiv?IUc*BlN8EiSN-~bl#lu
z(!K2k1vRSp-f{eJUdngEXly3PeY&>BRr2cLyO)p@aGmXS>~LWjTeosG<Hd@s`j3)o
z-<>a^HFP`J&^oGhAP&jbKUi_-CHWw)UK}Naea32q@jc_k&H!|LS7O>Y=rWw#vMCb*
zy8~p)Oe%K$Z9JZW5-?M$p$0#mvqUlbp!)mD%n?wOt^saxXsA{yosL44q0VxM=~eX*
ze%?K$>FsRLWU8tuztK7!FL!O}r;L`pl#bWpi%>C5Av5uj*h}%;s~vTq;L3!N(S$#7
z2V$!T!#e|(f#cdh{(>j+XGdTmNeR|icYt#_`%{R-9c7@NJm;LgFnRZ6rwxiE-H!i$
z8~^8pBgV#r<(m8tDWHp!Jk1uxKZLU&0&jmi3(gU|g{@=JCHyU2w2cExf`?(Jh>E$x
z?Qb+$#a0p~z2A?5P%nU;N^f<d@>S2{JnMTQ&OMKV_CsplOXP`RniK?FM60s~Kmz3J
zVv@*t&et_Vb<IEmsrY8Y2EQV6rDwt)Nj9C%ZzZfN2dcnp>j!2LStj8Ir*&D9!oF6k
zP|Hr@C)nNxHkMY$^GYELcS?%3jtQPWE5_`jRW4TnzI@s=S+Aa0QEI6Ai)fS+L1m{O
zGy((BK9FMBs~z3@Y3V?^-gUmRcvDVq`qmPtokq7sQY7t-@AXYDAe0+HCe5)Y`sF+`
zcIeiWUN0Vk;bpUzp}Au2Ny|o*+hu6nhzUYsZ>CNRq)=3Qo|=D3NIp=O_L*sQd=Jiy
z`S0mc_eOr-M3?Gd++<5PIWyBn{@pCkIZq0W_zdczfqIJs$xj8LxFZu#+yB`@EF<v6
zz$O*v%Rrj=1w|WLrTvQ_4cXF0Z?N^rx$}YYRv2X}eJ5$Y^5UJw$-u7A8IYH?WuO#s
z5zGz^t0|b|?;E#_CkX12Ye8?}QoIPnT)DId7lk4)0#@X|gX)ur759P<KLZd_tBnUo
z0x^9%!s*XQx%y1_4%%xT;5Hh=Wy+`yS`5SRBh#rakX)mK#AxIA-9QYA81|i2$Gcg>
zmrq1sT+#g$g~rBsEB6w?F29NjY1Ai40sWU$zzixJRT%Rw1R?tfSv~Q54j3#!8EZT^
z9N3leg_r_AsOwSF=e$N(hk9r@eG8((_y;;W@take1<a~-H#P5D1J`HQagY_$3c`Qe
zOfapty+HH(#0RriIf&b1yu_#2LCHM@;FXp+;%UU4YhH6?3x-%n!Lc+Aa;`py0$whn
zEWw<C$Ouq{Lz7<lW_VNI-mkvcpd58+X9Dc23`(!JopI#W>%b9unx-GXxh+FeK%Y?U
zS^4_q#TO{rs$&GSn+>#T2>DxB=oo6~dYliO-`_Uj8UPs^66>`W+TS3ZcKo3kn>F~O
zQM42`-IH$#mpCJ@E;!s`;?!X&<LeXph?`gzD4*vi_7m?Xmsz^Vbt6cnHXfW0B>Zzw
z=MI0`bSS@_Dl?Q3ca<x`$gk2L6i9)Ix4g4IYJ&mJ1gyvl6h8yyy;@jM3&}jAxD`%7
zpt+EE&A0b^2kueJ6#PX9l!&uHv3_}7?r_8$Z7dV*cxz(dDD2eK(8|z3%OtDfhILDD
zg<2Bw=N(t}?Hgu%dRz(?W!SX<ubX;Yn95l0`7V3*bl$(8L%2`$E)SJ>o0>r{Ka87K
z=WqjCIQyz=(r;ti-nhwBf9sNoxLFUn=uoWHW8LIXSnsMu+OjujPB!+jmK^D~ZFivG
zc+*lJvEn@iX)WS9ZpiG?A?Kw^AiXeedwBaf#Y+C1a%~B2HYOuf-)@EG8bIn{b8K2A
zm5%Yb>^BGEEst~@f$-OYxR*_G;PB3S-eui_Xq#Z0&^>K`p_$C{X)5-Cs0#B*i?4%!
z{@3P5wHG4QXq8$oL54HQ$lHW#m6De>53>DC`uipQ6`~LPFAiG_+A0ocP4;d?9P4y5
zyxRDEQU8DS6=Aq>k4T-kaFlFldG%|)c-v8-6gVAbK%bfb7Ts%fL(hA9^(Cxl)k^iz
zF<u#x_#)`iX?zWNXopu2o${qqszFCqKLfzudUuN1KRE(;A;CJeBrfGMmR&C{I<Edk
z!*MlA14A`ZEax$2A^9T_iEtIUoJLMaLihn*h!=VKt%LcHQ3n$^9AceP{c0DPjWP>!
zhr;G3EX^BD+MXU<bXPoCl3_EYf+r!3F2de(LtHbD|J%AD#XT0o2SV4Vh~srh<z9m3
zN;B(gJi-7`5Ry_+A*|CamvW_B3-(w#t5<;(?@cXBxUrP9fHYt6@I3n4ib3dNw-?#u
zq#D;`f?)TxyZ&$vu-8(F#3|6+iMq&A3}sY*hpi*c-$0;iut?<Q6_7SZQ%bBFyOqox
z=Ded?J=a00)(&L&Aj}^gEx~&qVhNtW2mv4a#$`k8hlBBFNI*FmDDUL#WQ^i$$AuHQ
zKj=0}u@Gs>XS}S(@)oVkJ<4<$esG2I-9$M*tC}9@rYzylSKBXre5Fzr_86ntc`x~D
zBg?xhyZCqX5T$R$`>CaVn5`oczBM68WjjrI?#WzhO~HtAXTBQ53ziJN%ThK<w9OEx
zUhTxNc}eC<A%;(j#%y9JN~Ds;T}XhnVoZI*1!G0M%e2lk=`t{;HZ(QBfXCR#?^ioO
z_L#dOTd@b#SKO0gS(bON9o9@pJx!2vY&etF-T0Pzp)P<+F(U%dvPH7<YscH?y?8kn
znCsv_w;S$V2eSZ5l8GiZmav&tO~~^*3%6I|aL5k1hKwfPe^&~Aqj7bKBzOwY4V;^4
zMDSY<sOCHSxlh3vY#6zSRXIIK9DSw2g7U-Pnvzj^db#<I=KqLS{dp!4<pGEoeU0O=
zz+cGCL(~(hWd!v%j7K$TV~?7=rM6d|sG-ufi!kq_q%~`!xO=3x<1Xae2}cHND>At#
zC#3N-Y|2nI+KRlUIog!T*qguTLdQE%8G@2K&Q$`C5jXlGSzm0*$uOZD%bnmGr>Y5$
z+XQ&`3z#Vn1Haj^%fG!bNtO1*4?VDGHW{fT`%F@CfQ~Dq*dIopVbDQ6qZ>!^HON}7
z@%URkH3{;1PrkK}uAF$%vN1z2BMTuY=H5T(o{R{TE38b*F3Y{_z>fS$(($ooimLMw
z*DL3iQ>PHY1{%R)MZ6i((<|+8GXcFtZS+M;ML+%W=#wz?yK!TYT_)AaxZ4%J$&Tkc
zJ`%rUYNsE}8)tA1+K6KBm%q|uD`1H^iaOlJEPgS|E*EE2yR*_b5J^3^asG9Dd5VIn
zf1bl|%DU<tZ7shawb2fqS<vZAM#k;ZwCi^s7LF+Tr278|+3LJqh+(ARAFaU9xLOJ>
z=P2*Sgs<)nztI<Zt!5B){#~u@Ps_<$z5-NR#R{&HKPAlYZQ`1WHPYV;j12708s^X+
zSLXl-mdqtGxXn#paGrB{)_XavMnw4N`0GJ-tatZznvZbOs;}YB5!YF-6+=5+cEMY=
zF3(~+tYpQQl<$FXSxi$`#1?XZVXp_)4)aIEp;4Bfo_|*9`TZJ49|YKmm_ZWTziCc4
z6i%MQb?<Ac&mpTL%v3$p9(#4<H7;sR;h6w#J^(=`)mdv+2sOPz*>d;?XWpPL|G&_C
zi!xC6`9=RQ=GO^Ys*(_zPLOq#e97_clXHEq4R_sTaPkVE{ahSx3xZK^R%oy_h~2Cs
z#<@!2F7=V=_wVG_pWbH!H@th`N0VRQf1oG01Y`pB55!l6@|SGb-kfiK?HQe?*6`mb
z)Hk!&m(wRvQfBDrx7ato#NPin>Ho*KLJ+@ue$y@g*WY$?{|hL6xCVey?Pd)Fqi9^{
zfDN%*A1bW|AX#|C^O?s=H?9zCz01b(<nN!`zy4J8D)jW0$PHintx)~1P|2S^yaCCQ
ze?(~g9u5AF_x4}kO35*Uf1kS>Hu-nH{{Pc8_-_YuJaE}5r3N|vKVSc%HFVoxuc`X{
zzkFQ(@^^BcfcqQghVJ#>cK!a_3;vHcNFNP7AEV5+zx4wA*C+OW1-D)SjH~t^$fAFF
zZGVeKzx$({Trz-A4SFNi{%I~P!f-*p0bz4Lgaqj6m-49nB{`dse%OBHbwX-(3Gg@;
zsg~(Bt$^{-(|!+9Ezep3SHti!%X_&#W=#4&-s69K`x(Ni*2fm=P5;_5?#X-w1AQ(7
zdeNz%3YskO=h${I2tZA@1_*sm)2y=Ub__0NhDrgTyA7CLuVn$Py8okf6}A8uxFR!w
zACL++INlx_tOEh2aR!>s2Vilz2u%f;*!#t!zOm123X>8fI0j-gKt1jeNS^NJYX1$d
z3g<4v+>ZejI0I-!&XR1PgS0|qD6!=6=1@Hl!6og1R)g8l&5wSPC7C4h8IUa|JBx!R
zm3_iR4T%n3fMtY&7NP84@;@OKpnc-EfBV`Jbl$8g`3F!A4+Wv@8R!?@T?8o6BB(6J
z0cF|-o?;_KN*-DX@bUxvpZfz$wPdLlFdA^DFQeToiAv;9##HNPQc4v7GeW1er{FkO
z>%>s<`-Rb(0pQ1zp+&Zoe?7kePYE~4ll%*ysogUm*^mh0Qgtc8H7no)NKsPLs}MkI
zg+K##KueCnMsvv(lcYtY$W@M5@~H{|jUL_x(26#J!Ue1KMjZO`bC(I*KeU4blyET!
z`kMhcO|I(w5Exv)*>Q1u{ZGjV$~3NuoBDByHQgFc<+Z;W_8s5}#NaaeMZO&T0)*X?
z_lIc8FmC|I8cErs9|Nn*xpQ;iRCE~x3px)nagXvd0O7p?do952Z~E27z=xZW1rYWh
zG0Whfc~}O3!DM9J>?wo9W6y%-@F!JZJY)U1nrImXFQXm2A0SM(Fh%2_fz=8oFh4l#
zZv}2-n3_fvHo%v3JoTceqV)bJ0IZxpIRwH_FvR#i_n5R<8+tqs0wbRD=ZUxkT^1%F
zcn!c`bsPGM8K@%VSz!2-1o4>=4@3;65BfK<|C#U%8oX5Vz;CYhk=j26q-O2|EKFe~
z89X&*K!!|dqT+TNTTmg3T&)Kq(&12)+n+(!kX6D%ec}g{#0urQY))8Z<Dqwcym@kv
z;qw?I9=ozv`JbIybj&q30DU7sa?%r6C|aSMAu^o!w?;t6D1#70Fw^_>tm_{laSJ!L
ziCUS+E;NIyRtCl$1%nS&+`q7O16<;!>9&!%cYu)X%vSL#5Yy}&N9>p>eF2@LOj{ns
zu^g4(EvSpd(5g9_rT*qGUqQXtIm6jkok#$B#k}&lJNpUbj~q{uS$TFzPM9ZF8E;M6
zmDagEL%_|SK@qsu_%GO84$nz;0mj&1AUUv3frq$i)v#7YaQQ~D=cU(k$Wzcf`&vPG
zoUS?#8kC#hh2!A{t!t28frzn$vbHMbTCzTCh!uVT4JK^`u5g*yYvKwkqC>4<^4rxo
z2{q}cuUF&jAq+7cP4Dav_aR-1Z{3BGTl`ANH4BoV4z`b50d@ldWyoGum^w|4uByz(
zL%1tX#{Hkr=x|cAKZUab@XWn-0!UDOXQ!Ajh%k5xVV=hTkzrr^m`1heIB(8EKTjB3
z3sPt5xQOV-La#i6*k+8tmn9=~B{>t<o%K7S7jXsi46@k5z8>FgwTo@7qvnlr?gmoo
z2Kg}rg@fb<3ZqA0<oL7jt)_~|<IsHPpbM9ObT;Xef3pB6gF6kE0lby6nS!)$k;0FJ
z(U;z(lgTEfbzuGQm_0cDL05y<W()oiT@{dgtg4UCkbXu?!m)shnQ`<#$~YUu`zaX6
zO3ttvFG%*a!e#!fl&SWQ=;5CL5DHOO5x+^wPVwORv>Ol$o6cG;tZ*V%fJL8JZ1e3~
zR~IO`2;_h~uc1Bn6M)g(S%tF#%@(Ppg=;G2Hdzi~i^vckH>lB=3E9%uU~83&Q@;nW
zK4(IEfCP@)$zPiTeezp?fl`OH%cYq-rO}TtrI#TI>6Al}H0bnGAq3#t*W<&e+7Qg&
zFam>b$7Qdik9LDSV~xxha*~gFl+-rTxN8(YzU07|P8o>HI}t8067{3h|D*?V3(>l+
z0Vr=hm5{I-aou3@Hsz?kYhN5CR~&m7qRUE-E6kff<O*7VX@Ybj_DbWFBf`qoPCDyu
zmef>%=6+9h5K0{dH<Q^nJC$~tnM`vH3w!@OX-yEW!c^aKp(SL|3Cc7Nr4-o~(>v<F
z&DsiMe>l@$Yup5y?r+*ar&Ue#Il%+@b$UMICj;NgbO5J*$3cV3oE80|DrFO2Y)G9M
zsY?0+tTwO8fL9@T=(B5c{3(_bX3nMd8r~a2&EIMv*?x(PSY7(IFIXGfKpgARz&_Z1
z#=g73hc!7r?;fZ}6K;zX#E_Mn=q}wNXUBx2I~ZgbCoLB4;3^@U;QT|52q$iiyy>;E
z((HjugiHVF%!NSu=IywH#aC#8oYlK-n(a7i=3m${d7D)V2_H{XlcqhZrm(=~Z_>?P
zGqOn1Yz5;Q)v=MK-)N+s-JQSlI#W_s^`C;_ze6bhd~NLy?LOJ%{eRa!?qrOGWbsiq
zfSX}!*r8$W`Yc%pRUcS^@_Dt7NXX2FB8?FgxY1CCVeS*nJ24J=4po|-K=HZD+-fAa
zY+aJM)=2*#bhxYU%IV~rzRZGCe)fVYd@A%0dJvZHS4z4T4Ima{2W7dbiC{?`I33wi
zEb4wIx7Q$<*Hdg!FEh#sinaD}lMc;KBna)~>7@#bMVNCZ7{$58F@(E~r}d<go`xC9
zVfDHJ=}fBy{(q_NIgkR@J9Yv%_#KUvQ%vM(XkD<-IHD6#XaLHNp4U53g8iF26?s2o
zWqZ2KEbc5TYT``ABJLv!#lAmNCpt?FB!b$Bu^;3ncZye2IL^Q(`3;9|))BMV*_(1q
ze0P%h`opICw+7E|fHu@7$m6})JTahBXov28DTCWtP2kUd*=-<8T%FP)drI%vki~Wd
zNGfX8+(#jjTr*iLGTb&#Gbr&ye2BS!Z&JdtfJ*y1Mai$L7C<ch1jU~xcmhfiG4BD$
zlf5+NDbgbfUm$ydsd52AzZWi*S^@l1oPuk*%}c^qu^*_Ji)&DY06Y#OMmNBA1`@gB
zsHTt>P-p`H{Cpyn`|_Ff*yAld<CVNk%sC9*wgoaZJWx^kDyns0#De%N?9+H`$~Tlx
z-ZUcrIZ&z^iHE)_gO!iF@X7gwqvfJf{d{UM&oa5W?hTYeKQN6<heRD$?_|1V1IYZ*
zYa4n2QW?egE6?`2{9^!-P<oCqc+m!B(HRCJT6GxJ3{~IB8LkQvhsnyCqjcUry>ZvE
zVe4sWHV9`I;~*o8X?H7-+rL}L=+0BXMdKJ|q5lckPQD3Hy3qK_Jj>hTibA5XObY^r
zr17XFs35^#kF9>zf8*HMH;NY_H-ClY7P<oC0z_gTBC9->b%s2Q0w$}x@1$!g4lYan
zAhM2$5>7&%$$`A{{WQmaGMDj~p8;-8c5W4d>;cd09H~A6KX&$aH2^XC0n-b;M+D+g
z?C>hs=Y<K1kvmqFD^jwibRRFQ^S+L}VYtJ)MnlocFr@B%N-Cd-Dnenx9Il{CVd6m*
zV}qRn119DpV06G}?gKyYiUR6Ft?;=Prwd-D7tXSyw!8OpuPJ!+ThY2wt4?{oWS*zs
z7R+v^D}Xsiq3k=)2z2C!;OMIzoAy=>{P?gW?`%m>m<DCX3Y%XX)CN_{(hx@$FTxQR
zDMHJKXoy_LL7hRlXmsceD(;eTd-V0~slI-}gUqyxq`c~G-N$?2M*7l0Ht@9=o3i)q
z7E%0tK#ey0xD8{QCav^Bnr+?7qr=`Dx6fTG+M(qfk)E?glPAn(EadzRKmGd|N9u?I
zoKH@p+%NnpNb)n@f@*Fdp#}7-XktW{Z<OM{SOWO;-bS%w31a{gf=PBO;V_Ijh5!KI
zg^CvlTn}V4QtlFqykmZ}WlC%E@&?Z3j`){}ojeCaQKD*@`kbndJkQ1G1Xu?M1qXNZ
zcp<95loaf<<=gZfY^p8~!OUK&1ZaqQ3Ts#sp0>_?eBR1<Dljq)%?FL}0L}+t6l*6B
z(Ero3-=Olp-iGsjM@HLQ@B&$j5r@L&Dm%e70tSLdSWI#twanR|Wq<gM(6Dp(%hxOV
zK^8Zu_wKT$7X6r%WH|%;{2){O{kWO9*ti!c`tY1GkRDQ}q^l);$KR%CGxWK1aNeI#
z`!xA(LV3$>d~*DXRWEZFyc281)Xj=MZ~6{m)?Hxe3rkmvFw$JM>Q<DewXk*)*UP>E
z`NCj5@_yC^+Np1%h_YGODGhuY&1l)9%b$GR1O163-_#^6`CAYk(=1s#9amKg-M<HM
z3Y2y6S4L1_VO3raUZ!t$@B?x_U9Lwa$J8%~mOThM&8Td2d73@sOn6rJT6VtoT?jU3
zhej#>K^m<mjL!;QPj)$4C~GZqh$=qnC<6O(>Su4#ly;|M@m*AgaUh{uWyYzz5mYzy
z2h?TPMJ$|)qEQs1mB0Bs+g1H;ShwPl2y5SgSWjvqu+@2{^pLhMJ=#O!GMmnW#P9UN
zO_dl)U;%r4%2vWJ!#*S$f0l3~a>t-x_~_WA*ZC|iCf(<fNxV;n<n3{=WuCa922i15
zJ&Vjuf(+Y7fs$7jFX72wmVQo<fO?ILKeBC%0iA*m?`xo3nz+|B`Fw`8v86!DLHF^5
z{qPB#)<Mt~V}#WA@817jgEPb`W=j@zV1PV>Ds0T+a1~bX=-+SSzmM(1azNJcopO8j
zOV)`Z!$By{Sa7zU3p|e(@YGuJZYc3(#LFVBxKCjc9v`If>)v%V0nq%?IC#?aZ+9}N
z4Buv<A&P%SN=qskLByL%ruG9v<9?rQ@J-m0P<uHnjoK%sEXHFd2Ga0Q0@KG(6XZ`2
zZS)!FCvZYXKOQzxJff`&Td`CVo$%oR(y%|mZ^-G6+B2aU6E3VtO-j|zUH5Hb#3vOg
zzoDiwN#*=B-+*0>d;yvM;&P&rfc#|1EL#G>ES?OlZ^uELOnz*={41#1t?vz86b9zh
zl;v5vJPX;VxpdMn(9ly?IqMLpAT-6_g%iWsF~frxg-PcZC9MpmbSG5+cddWcb(0Z?
z3CZ-5u@zqHl+9>ZWhK3w!%LHk;wgjtK?dHr%|*z7?wTTyLK}4#9_5@Z5VLr`of^%5
z@>WcvKuDlE(Kyg4?EVp$f|SyCK8g^I1(5($2xFzeQI4utUv>clrpBCb!&P{@4`hbw
zX<6Ky0aQf)wii`ejB_ycyT&N}r}iDb_da*Zhglvh=0x|W30=8|n(mY!cpGJsCmJlj
z&N;6?e_w6YB^q1TNaQN!N)NxjErQoYQ9yAZE?zXba@-q&=s*(539GpB2+RXEZUg7q
zAUCJ{13oj!^;iJwytlXy0%KPBwOgpzX<kGe;l4zia3*x59mTJP8Fp7jgr(C3X~>!P
z*bWM=_?~|7TW9dQ#TZR7=mg4Y2_AKG5FF8tCv#(#Aw+nq_^F=Njb1&($8R-5gly06
zKT14kd?yY`nx4~6*>e|VKzklk)8p?yAMfpCZYd8-`>l+q0e*Gs4dKBvt07?N)ah9|
zAtA0cf(|yqazJmuos|E(ZTzpYQO@%xCnVnguoMBr3U3i;>mem<rz8fJzB*`XU}N@h
z1~{W)SgVmfga(tQ6^dAt)8G)ltWTPet`XK)MiWaQi_EY!97EA3;|QhaQ7Pwy@8LAQ
z7vZi~$5QnU!tcKalzYLZH`k;R)<|+NINo|z7K@U%*wra)vnJa>BVV!WWuKT1rZy0d
z-zrMg|LUF|HD;T%lZLz&!TKCQA%m8`#~ou5m6}hgOdJ7ADtzyuWe;3=e)kkZX_PN<
zd<u1aOQZETrb~t)gIaT&UAZZWRTO#n9+6%F5??47e5W$7_p3`Lj27_|U!dg+EHw3k
z*%qv2v?dNmo3Iv28ihqAT%0571Kx=_kE0&EtFR{55GuH?n^$WcsQ7+4{bn_{Opf5K
znbO%hE^&Jyk(${(*;}boS8h$1Q<SzN?iytQ7cN@f4*Xv5PiD0yM4QSOT+_Pr262lB
zGVIGy&j*KAWE0<^Auvi)=*>h5ZKj~|)DNqm@V>#|5tzf*dY~jUSr3^C$#d&^&%qq5
z=ZO+K(c%4r=ub(7+`8J6<8ydf@vLQ3F(&J0z{UG!xsv%4UE-_9QR$tOG@Ps}$n`Xh
z+azX;(;GH-vRQA-Qtah+84S*~S<f03f2T<(3Hn~$e{<&zVaXsmMN5pq;Z_%KI(4?B
zLFm~-oPwXxge~qhBd2d${q@q`UIWADRc~A7Pt6%frpRY&4p>we0hGyj#J7T0nK%Zv
zCiJFo-9j;UZQ#SI<G?6~;KAq8qM?o0(U>uh--!#P0o7!TlOg|<1ycCepL}>)oI+_C
zF-mze+l*@$9ck(n@OG%Pj?ZH7d?Yb<i5+58F=<#2pF9Xt5)CtWi8g*KQd`X;tjjUu
zHaTPkEFOf>eCir;eWc$xlFGS{AM8|ElT{y!Zu_P=R3iU~gpeJA>t19rC#0uf9_#C{
zIUNmZy)d`rr)AczZ%XWn@}Aa)*qf_%qA{5dol82|JMZhO_`o8Ezpr@$R;K%2+qi-C
zi;!8si+(~oSlpR!!fLj5(XShQ>aNI-ksamNB2yvt=c+e-k)Ki0{J`u@E}_4EE>hOj
z#0zpSCf0Hg`m<$uq7@V9v`ch2#uA>Td@iMNod=EIMOP*lH%u#C@Nfv-)R$HUhq5cZ
z(AB>A%GzK8Bn5@6Sgy|*C42|!oa4qf-Fqp)lI9ALk7>BmZMj2}*4i2##&czg|0qO0
z<o;y1lHxs|4jr&U9?UnQTXo8;i+C_FhmFn^QdgRRk#^szJ-wYSYnv(Nib~{MB#fqZ
zc3pnIlAxLLr%4-_JV$>E3<oYeGQTCPxNjD<6?zTEU$<IgwWp8l(I>fd(D(9xLFZ0!
zpAC(jmSX?Dwr)rqXj3oCf0Wgq?1r6?sTWb(ejrpRb=aj~0j690aP!LxDyacic8v}5
zyh|W!^!4O{T=doN?7s5SN|TfJOd}BCzsRn|eO28<-DU!}!Wc*Ky+(GPtgvdoy3ORp
z{UPUQF6?z*PIf_g5!9r?P=FR^UWAYhTPi69YAc0%M`T5QMXw;PejBCOk-nf~b7ooW
zXs2K_l$=1c4F(2X&Px*F7rKu>b_Wz07bw=muJex#Wrb`ZXi;^UDn0h(wtQrKsbTZ%
zSUVdzy><lVaI(8XO?>zDtq1QqNPf;j9cjXR%_7^;aaMP}lPr~`4DkaXJZ(4#I&;3M
zx9I`AzlLp~Ghg_CI<PfrD)bn#V>ZC77_!u#sPp>V;+*bAZiC3I*_7c5o9DO8`mxrp
zvf7_N%NkL8mZ&g>Mqfy5n7oPM9TOG4$es&vN5-m`gTnp#pYFYVdhk@E8ceYWf92#*
zl9`t|4Q$`7K<c5S<}BHDR`q=jrKVrX3;AAWCF2WPWfu$gG*!iSdv6{;WNZ`|b5K?4
znIt(s^z%o9W$~D+19FW{eFc@_Fz6rDh0S_h%e`<t^L9b<L5VYvF1)-g)Fibzb~2HM
zdv%MT`p(B@OdG-a%9kCnt1UYhn^m|9Z9UvGQsm3~z0%s!B5k?zisz46u@}^i3V&ZR
zVR=E1YOP1R>eu=9tq$`OMoZPl$KR`vpKS(-700M`9LY8Vk<V?*Gbwkwy(hYtOWM!`
zc#2kZ^T13=RbSygpAsL%Rc}x>T*&jrqK0i|08R506v9@wl9Kf(0@|xLt;jM1F^`A=
zJ@lXm9)c9P1hdJ%UoEp=8Rj_y{HZ(!K7KC*3H-4WaE`jDL!kNINr5K5xtZDUw`r&=
zQ^sw@!MMqdPsuZW(8oZo*)nV87C1+BK7~jpIw1r^V>9Z!dk6azsWr?zGq*zXsg+@^
zoB(YC&?v&Nw50ju;dRUCHDKm;jt0{vLI&=@wm5<IcD@B3UWc_C6b?eZQ=mRhtc!8B
zC=lxzrSg?Vn5{BBxjuoVaeayLepFxwrl}%dahKdpe-8A}K+LDLidF_0`T<pHkZ_lh
z!xd<mYK;yL#849ViV!qU`=9w{1WU%0um&eFF2HUxBs3wP?D+Jm;e>)T`6-KCD4d|?
z(QfH(|IM(iUMmobp`F$nIUdo_?i;p0HbGQDHw3Ssh1->cSQQ|>E*;BC?a}fXOBMOr
z`G2J0qHGwr2oYK{1#uw<zU_|b$6~6UPfF~Q0ib!A@o?*t4RnZftB0+8Y-om3`58Z%
z;+Om*9sTPU)*RqAP<~&~{i~amV@6u0@sg)|cj+VBCA!;XWvZPh&^!uN3;-eDb~6&c
zZiFBLb&xF?)pH5aREXIm_XSWc`==^S0{>73KTjwr*MQMZ`3hiN?ZgtO$!dSVb+yS)
z7tIyJmxiJ-W0#IApS`^HB0NkRVIZR<EoK)Rc*TT0M7$n0m>xm22AF>m{X%6mtwBE<
zgGTau09lS<JSN+UJGv(Aez^2%K)Ce1OJ%!<95{v<!{0I4GR2Y|l&b|xDhAMny^NUB
zvtyb~$IR?~&fvEqpNU;`M>rK}dHQ{*3Bo5?2iI@x>3-NTcG9-!pfO|>-!WKjo(@q1
zjPKYto1w$~+lOGkx!5ZXEWaf^c|eqLhXO{<&yE8b#jvEOyL#l#7^#H^P|!%Ah}6m#
zYPhnaeE@_p+25SOge6+A_4+b(8&vCf&#sQ~H}&q4CcU0yxG+bH#MN%UU@E9~&Hw)W
z9x)YhU|Qch>Qy6HDB9f+yA|gcPe4Xw$KFSNdJ5vIpMp4zL1tC0$RBBr6!8JEBJ<Q|
zccY=6Q1gT7-^UK6E9ab4URb&GPk+!X{VlKvlX7DWflX*|XqfV}l0elt9tjKIn-xh{
zx<8Ikq@o5hLuXnYs$Lh@5GTTN8^vPjj0$|Kuq|+E*`t#yYe>KMpScTz$$S3J0>C&5
z&E(APfjsz*m0Hjz;#mJUdKu<BxFRH9GICdhcg&c5yKw#DE2!xMa&CH##F{ISl##3i
z-?dYGYp9L~)ee$M>^68|H#yU9tA4cMN>8pCfWM?%cT^ofclkL3IM!iT_abM$38sr<
z5I4m-0hM>g@!3q<z+F>F<VO0Y=G_whYN&cF;|A5^;G*~Ij4j(%qaw|lcJ_M9=8~`E
zZ&MC<?F`CkHx6)K)A@EG(jKO~YFB>CEZD2pycczcd)WkgC5L-Y@PJ>}cF>w^O4q)~
z0x3AJl-Qh$?2JmD#4fXkf`qFEZj9r|$&l>2-u$5K{11(>)P12Dc3@q8_Qeq<UOmgs
z9r$Xyu=cKHki18KFK%7$<RlDqpiCV}PTDH(P$r7y%CAc>wA75MZRnfDFU*2-0=3W1
zA{n<rSvF6f>qu7lga`jU<(w!OVgFeP$eoE7akqylR;~J)1sW*#l@Wvv!W#}NM{z$}
z|FBZ-z+^&h^xL2VgrwO1SXiM#aoi}LtWa<7Tg5B(Q<<owTCwPp1eiH%?XN*f0*w7y
zKl?V`PIWXm4jK8W_f4lU0J6K$x}e)mLiWyY0=C1X9$hD`UkO>M|8=|Na8l@i6)P7L
z>PAyA>8{K1$(k&37PM^CYcr6yBVg|fnB;T6h<a}I^w`Z&L{e*wH(?WiQ$u@ZSNJX{
zcO*{XM*y)_er3O3$=w=)DTs!NcVUPp)e0C@TI_)S!7+saD+tuE#$xl!(+L8^Ob0LL
zB2gT39~KqzxD|crc)^U7LQP-Lb{s_CYlnRA%YHUypg<61J)4*aZnlheU{Nurh;&Mm
zo1KPwRMt!E9TA&y=Tk$eYhWcp_r!OHLt3*j6uZ7$YI@k6l8~`@E#1Pe)`Ne1)IPIs
zf8KQx;ubcnDc@Q#{E8>C?G0>(MAZwlYa?LgHy?uO9Pvl!T&jIt(M@Yr8BXC5bR|;p
zY!Vm>w)>zIAV4>VSd;B_Pu^vCAapYYOS(WxX~!Pjdox*9Tvy{`EfAg#sLN5UWD`}R
zu6mPRRz3SO3<F;6>*HMV?Q=gB?7o9^hA0s1l=D-3TOt1OLz^74n_f<Ho$>u&bApI%
zokej?{F=9TIUiMz^VgthfMq3ON1|XSgYQvyMJdo~QyF@yM$}2Axsjew(i4>JOh+dF
zzp!Q;Knm1Oi?04vFG#&&ehY#R+g14AUvUO2Uu>rdaxj>S-UNyke+-@{adO7NmlDA0
z0vq!IPiY-v&|-3m5Ge^H-qnKPlA6(3;4Jz)D9jCz`Bp4M4CJ!aqz^7{^$gM0U$R=l
zt<1>9WJV>Y4;jgP97X5zueK>Zf|s}Cp~M(%=_-hM(ed_SF3*S@7!zip0e%9Cy4fd6
z4NI{*u`ALZWo}Uzcl%uN1DOmEBkoP*e7L(p70&Z4$vF-uj3D?!@D-WN2^reDYGBRl
zxQ^GKD4yk8Cp${91a^IrsD{Umv8`S<inDLo5@taI;eIub<QFx)uw?vOb1$G|g-@B)
zTk^s1!@$Kjlwu#(&Uw|9unIi3ych-5RnqEOVe{rJai7E5E?HN5Hzg+nzHXP|cL}>u
z0YKw!8BPXb{imZv&|l^*>G>i!;UOeit!@yd4Yy>9$g7L02j#(uCvfh(!-MI(T@XDi
z8S>3qwq0Qxlg&ijMJ@)NhnrQG#bUO1eNu0vBgsywDSE329zkIxh^X1C6(OBI$wN2y
z);^8p=2_H~Pc1%|&l5mr@>|9qeDVGHd5GOQ5+fc>5RRxdXPEbYol>pMfoMf!eQQGU
zH3Mpjb8XUORFDCFpBs#ZJylX3lanM5h-yMOp?Sk`34+ifi9QmasR;J?7|D%vQzS%;
znw;*z-}1k5y$0{1T$A%jm87}V>x)qDg)Mgt)H!&any;8ns1?Ya=0CBT1uFHrFQj3A
ze^j*Z3$;~=%cQ~{Dw?-OBnj7^n}CIr$&1}lStzo&cb7GL1l|cng84PK2%l|$s&>=L
z;dFl|^C*fS^jP{m7|N-7sN5pfy&ZCDd96evA}hBHdu71FtjJnPt1xl*SoUk|#R!&m
z+YV5d>z(#qpETDSNL#jO`c~Hm8iYFF$f$dKPyT56?|Ars2Xz?RCC{~-zAF6t)D$K~
z#b_L;?g`=z0vn$Z@5zX!mXQMFcG8&aNnm3ykHHI!*(E4Oq%5VqMGfbI;!T>>>mFr0
z!7YaS$$OQkdoAx0jnM|FpeYiF6eMD8$4h)3NXXq{=z<@41E!=#sK{NVu~fS%lk5KH
zIBd{>CuEk3xwI!B0i|?-aYvHP;o@I}6XycMg<I4-F8Iu>zIx@GY|1m<V`XJ3@O-;$
zILwcG)!a1LC|jPp1yNoG9I1Gbd=yH_y@r9jdWbwHkN|<ZoCbzxHV`M)<2ZcHZp<fv
znXspR9iP>q&<`XGJ$`|-p}nL$Ou$NZL9tn1m&7Qtf&x054~fE*WC(ULx+K2f@022a
zf9$dz;QGv5Q;OPIgGSPnQO#q%DPOn*VeO##Bg;3)m|9E$<qtP!VH1(Je>ndDz|Ebz
zk;Zcif(2}AL(fzurF|;67R}npPBMt-xJ+M^s;uk_1qh}NQ!2e9WUUoFuak<pLXFRQ
z=K=-l@txuybd~N^?h|l_194OGm~777Ok_S0OuSqpQ(i;@26oKLKc9pbebI`PqEZsQ
z7#%}MA#hB`x|7`BBSWS*n!yO;(iDtT=H0oz9gLCKtv@gq6JXO}K^S#3nxIu?>S$>`
z7065)+uUda;1dnkZ0dj}tNKA>msk~e`+&(0>bj}9v8`dGCjs!Lbz>Y&zH`H+V{q34
z@xI6A@oNg2E$<Np#)GIJLjNzLL+xiZZ^caE@dIAH7o!~Jh1RVSPXgt$BwnV1txb0v
zGz@-d?o;Pm^(#^GQ)M}~(*%@vi_6pAW-yz;?KW16fhhm9bjG<cFMvP1<mi^bzeG){
zG|XM{1{@f11`1;gQ$v4#a+J*0=Du7h>0|u1VL&p`>-+$5<o^72G^Wmh-)d+;ES5cK
z{&6+s=j>j8)}q7h9z@%RH;)`8NzBBOfUk)$b?&>n;Kraz?-`8pmJW;jbje^HYbNKB
zjf)oF*2e`Q3Zj#%B9FM=1KM`!iLZA1rGb%E!Lnr0^hS~~2>m&{Z82rFu)SKh8cX!}
zOBJ@p1k{XMR?Z$ix-X49M_rn=IQBYd0<7@2DF*cXincPIc)i2D=Rdls&&3gSxm|is
z<Ph=}CocFlyflkwH=j&C6M8Y$5sP8qJ*H2bmOMHEl*DClSXEr0<+|zRq2AYX?D1VB
z?REI{gI=$RX@Q&CFX?|*>*mlxNtBmGAEf*h+V;Vh0Ft&_)XI|#_;};F)zl{<ZSrog
zNU;Rk&xG8nP~d3r{EiVyfbk~AM|uuMqm1Ki0+D?`Qob}4S1-yel~oMLu6(^Nln{<&
z-m$xzB@-cys0<V4`SKtS^TbUcLq2@zJD<#zAN1J;j0{_Wz96XMrzWi6hCZMrcog$|
zNoV4^e&DsVp$Y3r-(K{deBa~Fg3p35)Mz6XVKpLs7w38VWtC~Xul@xEhu-zrua;yN
z&IgjY;HxpCan+t61Bk#W@o`!HZSFZjb9qFay>x`kIO;o;8<y-{xl=)}Y7AuS>n5CQ
zV~cF9mvD$GY^nk^g1Le^f#PxkSe@rJ7zTHdz@~s`Wc4MO1E<a$OG;my;78BSQPmBB
zR#@weS)FQds{61JgpKi&<+^*=32Wy>Q}~W9rhffWwW!NeXgQ#2GXaO!?`YZhFt5~g
z#GA(*b0mQ!c@WvB@M(qq)5IZZ;ltXQ3(YF=+%zNv*#!5e7&@6@ltUy1ZrN5VMeb=L
z;Z0Ik6+)at!;+G8CM*Y~+Cb-T4$Yh3qAJf7S4jG#mZ$xatBC7m7vZ5R-#($QR_CL?
zhzRPDRj~Y|+UD8jCFdM#^&&Gcu~q&2n0WT~Tt5_hsfV;lTd^>jyFI_08zb6{SsEgy
zcRR=QypE#lA|d(fopYu)e^Q)rX;%JZOSvy(1-In4s?mx&IQ+R`R82d@U1BrgntmY8
znY*PV%W7q)6k^o@L|?;%W7<qRwZ;h*7K1#stkNhs#gXTq^_|nUdN(e#Q+!KVAaA}d
zDr3^XoG7U6lnoQV0|pON84r5c+42nVMJDSx8?{pm7?|teq9UbMU3i-%_uDSAd{hWI
zq0t*wEx8iirEnw|yB&I&@9Qmt61;-=qx5RV=giBXMq}Z2u{8b?^cnCF-(9W6-Iu6;
zr`|3SLugp2;poj@_E=h+FjN@NMBrR<>`;5;1H+*ofw%PHFr@uI@XDhWZ5pQ7L}_@<
z$j&^|dabxIgQ99!I&<pD0OPRLR>{{SOsqz=?Zv~L5(i--o83hE`s%`0O{+n6_1rac
z{JCqBe_Wh=a;!akAtv<Ju`N5&q4%0xc1wasTOi7F=Nz0UP+c{E>d5k<!^^dTx9qgh
zVSe-J=?2!4TtCcs{ja^|MCNxp-gS9#g5tr7?BcsH`UdJ~%S-i*jf`#_@W0<e{=Dp9
z-fOA2XK3G^dP7uh@9~5MOU|BYtJmWBQ%#Pl=gnE=nUV|%Fd3fJI>U9VcK)8?AR5f1
zZ&qn#)dn`hSvuSD;=*@{ZWxPMyF>b|icb|i!{Ns1>#lRm?P6!O1C>7??$xfkh`U<u
zdCo@<UzqH>y^wxea!vZaLIShveT>z)ZPr#gSJI9FOH&Ukc6|o)jc-rCzss%*KN7jK
z9y#B;x@SA4)Oa9aacOR$8|~cYiS}3P8p!NRwNv{%F(5<*t73m+<dK#ijwsF52XP`M
zn_pWd`pn$Zrq6|WK;ch{R*2(1s&)ZaC2mZWm56)Z(Q}{hqiC=2=^Mya|Gno?$QT>w
z_OR`6a+vBk@G6?*M?_W&y$iVEkJ#rVzg|S2{kh$d7T2ns&tGitYUpaY?yr=@e}%>a
z)@4Bh^wEm5LFvC>@KGi>2sRcxfk86RpSzcW>Tsec%vuM$?$2k9c~y-dXSQ&?D7uAt
zm;Cd7ByZE^q91}$^Y<7MwIH}3qiOm@|1Dp`O%Eq8;`;8bp=n@v6s5z5i(Io{e-)l3
zCnS@!j2k9Y3)*AeM(ZGw^)m8jksqtWFOv3O-`Y8{PH;X+VMG7B>r44M7|C&8#YWuv
zLrw&HPKI}@D8~`lb;LuqK`&CTWAO6|sTw)dNPeMRYWM=gA%0$&b^mQ5HD5BYqSN-!
z=Y0GZ^hZjYO$xUWRPy40QqC<hDtxW-+XkN$?qm{R@;u%bb^mTB``6~+Uq8$d0U!0v
z&|KT!Uy*-R_n;pJcw7gFlEGG?%kLJL|M8!nBeNz35@Fv=zV5%*(Z6}=zutfW_&n{j
z8w&ORmrv?{`u*v5!OiV@yAbyO`v$ZjW{$Fy<Da<iMIKyWJB{Z9{`{vP$`ZY<_)wPK
zJzF2&0|iVg^c8xdzPUoLMSZE3asH<l_0P}i`N7BYxpv!q=^w$Zj|uXRl!2Qr6aapL
zbO5At*PM(JT^4l?!iVPS8n%I?>5s<BqGBq>z~~PuP97zYKy3o%VfEfBQ1?9pEO#qF
z)Mv+r3a)0^*;gK!*+&;mSwwG@0xa7GN;PN&S#WfKa=F<)Fg<c_zxnKh_c4e(g+h5B
zc4Jf-%(fwJeE(zi>F_gAyMnyI6;Pi`wgO)|B_Q1dQcmqetb8`qNqVTw_JOCi4Y+72
z+UXxn?L9;<|MNNh>s?_T2X5#K=a!v*T@Gv30PtJ?ik9X8$GTt9(snP*{))N=vng9Q
zXWH68WTd(_Gi|{)U}gSui03}&*qlrO640I+G&l{`tXHJ}dtCM>@BBT-5GlQOTy#5l
z>FZ~p?QmKKJm1x7J_nZOvA{aZIt7l@Gf;pkg|d(;G{LA1?5C8#VNm`fv*W|9$!#Dw
z9ZW#u!1#_<D{H*xL1?0|zWa!#2@q9yNI<N-rwPz7E(3_HT(NEh0e_dt9CI5EITk1f
z+8*~d=og3qIeG>xmZjJUWIFiYd3r(EGU(jh_1OchDa{dCkBVprN52hPSq*2vZ#~te
zd*+{^H{dH_JDl$X^y)FUMnIGs>HxB@X#EIS7M?=c-3=fj1bpr1<W-@PEnxx$0E2hw
zH)6I;E0T3|3|Y$o%_hAMOcHkigAFA*(t&Cwd%Ce;nzk_p*k$>FLrSOtq_6TwxU2z|
z+xO2b=cY7({}6rpw0eqSD@gec-r39;6i^%t!YV|S02CNtMewr|z+NeZWB-fy9Urn0
z1-XZ%|79ck&`7vB3yJf@h?k~+CRlgs?I!>2Q@F<eDod2qKaO-IX?$74%IePi^x5en
z9Z(vJM~nmb{wr_KC-2MAqXC;WpsV^OKjd~f!0kh!M2eEWd^>;iTc%?&RTRbx*OrEY
z;2M<X1jUKtR9p}sjel7Ugr#&m2HmK|c2DV9X&vdEZ^vNLj8ZBH9&clR^?axhcesJ`
ztobI>8R<<|$cqmeKFC@Sg?TJd<Z%rhy#}_M>q}%Jz#(zs=n1Zh@8gF+KbQfsP_Ygk
zK#addnE*4u_6;ZG9iFOTHO*YlYyM!gH#C)PT<dk{8o|vZ1#ro^zZI&zX&*iWS8n`>
zE7S31nx#tskw*bz#%CZPEfo#CfyS!Hxp9^Bax)?s08BmuqDd}?(tGGUwCgyTT=YE!
zIEVMngLOMJNF-^e``HDe#TRD{2PKMqfUwyHW@XVpVRmmDrG#pMyw27DLT5~JqlJmU
zEu!B?b`v_EIzb_f;I3ou1q;k+Y$tl^==h=Z4~NpBMRyQUt#)V`fCnQ01M3v`Au@Ir
z-aI|3V;1c+pJsI-_}gO}OaHw36^ky%WZ(x9cnoolHyRPz$DjjuKHet{A84?f9#p&^
zZ!Z2`L&<Rfds9_(%HH1*&K}pL+-ME6m>rnb7$u6!T8Z2<6H}_`6ouFfv&PiJJVJI{
zLBhkRt7i?U<BLl7tZku?25KSe`vzlE*7S(;K0t9a7?SCqV>ppu?S@j+0M5#mQqlhI
zpuR^0+!-LsG4Y~3lxC&_I=F4XS+UT{*SY|rG&BV{(1%QIP{igI2*DETbVPM@ild%@
zgy(5$W$ik!&;&yPjC9_MFHAs7k`RKVjf|=yf%hRupHW7QsZm&;G%YOOeMp|I?gQ=^
z2xh1&yZY_{?kJGgZrKC*H{@ss#AVp#Ne2s{JNX|`QNl0v0|8v~J$7$3R<~VK^-afI
zJt(2Z@ifdih*S~bIH8x4i7KX5<>)EHhGb2y%}$b*awI{lzy`@-n8Z34W)M_{q0=~s
z+#;5A_#Q<cMR<vt+x;v-lrM?+PMl)VF+Wbcl)MPUsD}cB$(J#0QHh?6YQ_#1u2z~K
zUqP4jdmvrBEo7*O7;>o2@=m60B*ryoIRg(&--3|@8GD)?>FWdTKO(`6zj2Q6rcwF`
zL%|pC2LjEG;C}BvhHMNMQ}U!19^H^(`AN7v0#e1%sE>rz62B{0a+t}=&SpyfArn9>
zB9v7gvFF2-E~NIT2Pl#TrETT`^(%bm7=ahL)tzB_gajKVF^CAUCP&hVd%GE=e2NmS
zV5`Qy8XRJ>)UH~oz{ERPKd-j?CTjD&DuysIa~7=6)M>n2SFMCpS)pMG>i{%!sCQBP
zMxz`PnT)Pjy*+6%xQIp;M6f13bFdrWWxaw-^at%DiR#pg$_WMD3W{yC%1Tkm^|y3`
zODcSo--W}-cy!+tTHDC=R9iGVH9qfwyIN@1fHN;vHWBBOo08wc-g2yNhh~$w+2Iig
zeW6p{jAZro%6@PKz%7WLbQ4oeY?$vCBUUHbNEaGzZ<7g$172uTHn(@4c6&Lr==(5G
zcM2?uXpCKn=LGD%VEPm#w;G76GFHmgQLxLltGF{Ho5=GGxSOukQ57XYp@Toa`_T#8
z>qOQU(tnC_h&JrtZ9XP11yd)pd6lCc4zfhvrv{e2^-f=X4RG~_KNBc|{~Nrp<fmfu
zrEPNy)~e@o2{Yr%|0|~QZ+pN$H3{^^hb@BOm@$?r)Zc|z4%9**ve((KTofWdExgn*
z(-zu6$eNnNB%M#=Zk}=bWk~aCUDXsF4!9l&;dfE?e-g6-fOnKzcUSCtUecYyt(G6@
zc5=9DS6(8n#dz@nV#wM?fH@E*W5W$PEkYh9OQ4_1X%8E-WDhLrx1yKcw;MRkch{%H
zp=;yh?$*KXlntT}k&10}G%FK<SFBY>XMq{!DmSHrhkA<M!yO5$RevC~g#3IISOvDs
z4kYXWiwlVc2SB!bumq`tYMZciQ!=V=puE{ovzQ^SCNGaV`k>a)4kikV3=Npo3w;yS
zwR@D_7V)~|!O7qa#bDdW*)f?tMF&eBL-0`lANJldEXsCm8&*OFP#8c;Is~K<q`Q$4
zQDLM(N<d0FMoM}R0Vx5c6{JO^8M-B;r8}g17`}7vb+3E9-}|n0Ki~WO`TjirW*au(
znz^p?Jda~P_Q+IH*DN^r+n{si9{~`KoZ>pUM}OAf0SM(;uxmrIyIFWx*BAhvO7XEv
z7l`!0ArgN3-OeNXY?JOpFgw1N#If*cNi{*lb7$66$78Q11SF3N7lI)l#xLg1r1F{i
zddUpe9H_vE*$8kVada!S&&t|v+1<?j#+sI<RJ;F6m1*poqlsffcvQe}0Gl07QIJp+
zaQk_Z1kMKCgpdq2reI(X#nKpZG#+UqLob|7RiP^6&y&4l|D_fL0yME^nj+=&ZF3SN
z$xZCqZ#szJH|yRuc=mM%oNQwKnHQhG6yy=b86W}Kn_lC(3lXSIO!;-1spDR}1OiD2
z+oMMsYw8Vy#DMaujVU|>mAD9iUN0_hpM|^x-{NNqttxr1XLkA{O!Zg-av$)d+aK93
zimSp@BOZ{k)qWp8DvvI`#xln5uU7mEB>GS!TIsYSWq_BZEWwjDp*5a0!Z=n*hr)0H
z=@HR4_2W*>H-hj5c-A|M{;(X)03P?v*J$h?xZXE;cvu9C>C!6)m>*ge>{*IBg0jFr
zI=jGrTmHTdnG4pMo?3yr%mHwh_!(axrCyZwsEs0^up~MAo@L*&P09qolVV1BaV})~
zJvC6dr7>s=2gl8qt*2FWb!S;$76{UTDf%=1*t`7g0vU0@==1x<`SGzd(=Tfs+mVYy
zFywd*oHBxASya3m-hNI&?!P*<Y6&9=IB1yCOIW5~zW=+|p2?4Ut$>UCA-)MfiqbRh
z)e!d^J^Y@#2OpClc7jFRy^%=T!;#5BfE59HvskXew~Dc>kSeM<(wF3&hO&WJfyl>!
zII-NSHx=b{8f6~@kpCd$EW_gcP=tS->YdT(gY$-lk+Gp6sUh3GWY>PZe%I~fwY8!m
zZ!SH*!_hqH!%qP(wYH3^tqtKD@Al2+KF~2vSy2lm{cIUD(7d_2X)6=CE>KQHrIEj(
zGX5J(l0(oB*o*3pCIVP$#FXEOyM8vJ+@+$hLR5vaU+|tibTj&SVAH(^^mHC*!1&Ox
zg#t~h<~ECSRQdaV12z1MbJX~Z<jt&O*`UJ5fA+2HZ*bsYCeVABh)rH6g_prl>;W2N
z5XNOyw&W;Rr0!0Vn+#<TXdsN55Z@21_iZpqjyr4GaBta;;8*wq1Xv^#!LZs&3%~uC
za}SC`umD!-XL`IfF?wI@Z4gKrYD61&1(Wqc2$6%HB_}W*Qwq!tr5_1VJ{s3a#hzD?
zY$FbagspFOw3$+*1=1Z;`-3bFJc-O7xX*V)!AR36c#tw_ic-RzpYhfl4Ed^Yi-qA-
z^>@*8ldYuA65u))3@TZEOa<|izU$ZHQQ{U2_juEjU)%mG$W(Ue-S9pVCJK4M;QL{B
zx^Uo-6sV*BX}5{H0L{EKO7=YtSv|`;d0Aby<?ldEYFHRxBC6<|%_ZOUiPW(o8ZVGQ
z(H^3rdqPt1;02Ii<nL@b!Y6UxhyALe41ge+akM*0UIWTgBQAU@6iP5ag5>DBzJ%pW
zjn)~NwQtw+_R{dbdGRHceb0ZG9E*fcvB@2$DpF}5y-R0CG;b!s>G7eSqq<@3-ny`2
zRoD{}L#Ej;AZ~nGjx`!W7OQDo<qc9feWVm%zd(~j5Tu&&knaH{Q+5tf7D9l_mG}Pi
z^MWtHRB(~WZ3TSaD{fp!j&(MP=i?hZ>vDT^+Z?V7_hZCQBSam6tJu}1G`f{duhd!%
zKTWy_eH7icXK3VZ#QcEL>3#IWV#aml#n48g0XnxMjNW83>1i>%eCe)KBHAT}&L^eb
zo4%5z|Kj^}>lEI+Dvgy&v3IJ%cVm~%+s$BT`-3d-Z9f!?OnQQ&eu(!A(E0Ml_j`?>
zj#e-E>R;)TKDXD_KBI%lS)i%A;^{JILE}`5(nix=z2|rH4A=d-{hX8};U&m|sbfFL
zqtC6Qe!@#2W#?HM=^0;_Qma6xFoR(~I_<iR*IAAh<hEDACUhF6RWh}A=7`}{+*Q@Q
ze``v>uZamJXWQ}q^5^=v&H^5M{rYCoG&<YWCu7ibMmG8DLF9KN_V_o!A?9c7w_z~j
zLPGWg4NV*q@Tpgw@rd)2>j+55s?DKSqz6u+-)0b(VK+#U$;A}$xC%?;XSBmI>i5&O
zt2%IXAoD^k%=!xokVbO}!gGMYiV?)#qDU&?j^`G&l)|Xx4D~Hl5Z79w^zSR^zYe=5
zcTJvE(FQ}KDal$Xiy`uDn-~J($MJHlP$yhDtz@_StHgIHp_gy8c3vxx>9rH%M2Cf!
z&<2O$7D&3kFNxd=yHsOOTf`!0@zqymfj^^L+Y2D}W%LF5USb-DidcgVd3&`p7H^hb
z3<-6`8}xAEd7Jxdf@|SF`EWjpO1v8wf(o#$Nxm3Wl7w<Fi!u>GnxxMLVy+cFB!%zY
z^eyEKg@pLx>K3AJbHRfHTyWI~p-j*~M%vRo!Hm<W$i5CeCnZriujZ$N96Ha^Bz0P<
zi=bhW0URU+R+fP@g-mbr!U&f<4H4eI0TAdBnFi5wehk)(`co!j-+@TuMK@f_kcxV6
zHX6O!kFu6*RuZ*bx`z*p@DXC}8#&oxUg^~Ya-$dtF_zG@-XcfV{!0IGSs3)jMzX$6
zPP$xD<X2rByZAb0k)nhAxZ?cOaBR1G4$$izG!;@U0egY66x}}hw!Jl#OI#ZamNz1!
zDsJd&D5ysMcvluou8OnOOiN)I_Z9YAVKMrHYtoO@ouSuE%L{%0q)3aTnKM(#J>P&(
z9`b5H(Q|$GBC8-iB)1WhKY3P~k=SG{Uxqgh1R`hymDi7hh{QW4V>5lKlrLW4wt}Xj
zWQU@Q`0!5VbH)4vOK8|by}eBUT*y3nslRF)9ck3cat?5S+Pj886=XaLkP)eR9W1UK
z{k%eRH_;8sjDl>u<LJ2C7t-s#iqpddS%v&$^kRRDxdq%|GmRv7`K!nzQ*Kn`eV=%%
z_vwQQ3-T;j-x}dZ*wY9$|FB};C9d-#>xQJScaNJ}RK~l%N~d=4rnogc?SJMdJQoz&
zROqfmMy5zM{0xp@;|?dJ;NI~V3v>~RHJ{n`6&NrV5|Cw;X8W;zPR9%E3SdG45w7d!
zrL&RY?wR;^B{g>L*(qcDQ%DLsg$p$?kQoSj(-Uu;m?7WVJn7Gb>A4_IQKro0HYx$i
zWRyx|pOnf!2tlp(0`O@sL=1XjtVq6W)02HyEYGCU<R<SWg$YH%{TdSePRH7%mnAiK
z^xW0+Mr#nE`1hqyY{KD-dSoJs<(^}N<@lyq;ksvCA7IyA-?3ebu)=j^Bz_>RS-Wg2
zC*58C+bjM20lW4h{g&T7orgk^Np;&<(n+t_Q#dSF9oXDoD!T_S{y3jNK`i$2!g<cu
z@yp+?Yzq%et8RBTTRrAVsk)nT`-jaFzAV^!BwNw`*`T5!wr#>;{L8z$Po{u!DKk7M
zD_lMcf0DUBx#Z`DMOP?Hp+|b;c)Ueppl|`0^eZE^>a5K<`L-AF^;0=Ywv#s#owHTq
zd@Yfqs&#KY_^Sc5Af?$eZ;W0E*w*aQn<ksUC!_Y3a>Qy@H5d@$drPUokpNek^O$!;
zB5m4-+se+Oebo(2v1a9E+T2QCi187I!d#=+R-@coUX1hj|KjHMPD$nPUiD);+DDnV
z$H6k9t>lE$0TIYp7?TucLqVH`FD_-YHkt&f+1wo&t%*<(X2g^fhJa;Vdy!KwqGMAL
zeP!H$6JC@;?<varZrHKdtJf~mK{(vXOgu%k%;m%_>+dU%VUk8<^k3CUV=#$opp3-e
zw@2IuMILorDQ;PVZ?q$3iFa@8-f;~(fo?H9D!16V_M&`2@j^uLuF^cGBDJJ7Mc1|T
zUK3f4&^LH<(Xhb?tB-te^_*?1kPanMtKPi93nh`CyFoiCOF5oHnj++0OQT^lnPfP4
zN`W%LabXjfEV_p>51%rY2o)f5jxFIy&znp}cU1#fTjCc(YdP-Gj2@f<Pn1Y^4M#;$
z*fO-3d9aY>q#%yHYuZ2$65QBRX!Xu(Wwh5jp*`zU$MgJ^BNngY8;+hqgqC}!<%n70
zSWYx!<K2SD*q7pqZxrV3OwONb^OZB-BEcnc4Ksq;FDre8g!Bv=3KjSn!{blo{GPpo
z5^k`R?@{*N|3pZ$wc0Pw*A`3fuutyBr2$1Up9n^VrZaGyXBo!rJzivA!}93@_DzpT
z9HxuK17*_haRu@+-Xr|rRg98w<5^<Qa3pm6y<`5%#rWv?gjN#sBf90c5Bj+Gh-!M3
z4lt=|Yf=!at+1DT8^hlL@692bE{9i1G*mTF^BN#)RTa(0-k@2+3YZzKKl*B;GUI&$
z%$+avm9pO-hW0-m3i4@hdRu`>IITn6<%%9CyximbFSa6lR>r0o9JYu+)<2JDwVmU`
zVtC!n*K~GNTVb7pGwez`p&i_O>797DVy`o&fmVg@GLBHQ1QJ~34BS`30cxx?1^Tmj
zV0E#cDMXPEYbjob5Sn9gf4fH#ZCXkg5K5;vHMXZ|B(v!W`}RQiew3BUgKAF}z7v+d
ze(#ZT<)_YpAL(?~3b)3(@2loL+z~-|vdWQ;+$dPaYoXQu(qr(x2U{WO(c>LG5$(m9
zg!)l(EFHwA&FV+XVZ^TkPt&azw7-EM#7$rn(<U}ne4dS|Ed?Ye6$$exo?s?u^Zj3Z
z{=0<V$~+chRg@*JNgznMZ$vV@TQ2bp`S~rED1gihiMYyhkhyVSlTr+PSt`R$ejsFH
z+^?4!#ND0&?mNo^2+l*%F@gCI=MH20hxKLs`^he^msIxdj0}}Li>Ue>UchzIJb%mU
zIjF<w@)<TcyuSrj0_f~IYJ}C^4J5?(b)(uM_#8Gknk1LH5xMRkeUnWL+s~LYNA8B=
z%1ZD`^d$?ne-X0UFr0P<TrN*>J(9S0brxfXMe{mGL=sq|N>O2Toq}N&8&#jz0y2!s
z6@O^fSgEAc<UBgV`qxqN*O~IazKBzo4Q%MtqzTdZ1Hmw*Aj@YyX<nbyfTHSvZ%A?c
zTUv(DpGfA=R&GT($OCua#!mUs>HRM78Swqa5u9WlfAYCwhHVsDW9krNA)Q)X?a$~k
zH`c5}BsXipa6WdG5U!uv{CuX9JwvjK@RBZoX+E9v1t%_iAxwpo^(TplDa<yByr6{#
zKdUVB-GKqx@t0&QT-vXg3uk^6R%nMlX_wJ5zK%+hf!7n(`tPUCfUm6;cTbpEG`Mct
zcSYhH4d`K3Xc}ZW!0YP{+-Q9?r&WIFv0m{9f~P4L@_8Ob8BIE*U`wZ_Q5nJm)J_!M
z18&|zoqE=3ZtoZz>j35Jdt=OkacO*5$9wTxX1CWKtL*6I`7SlR_xb&OdW5O)yi2cs
ze%=yc!SK#{-#>fX4Hquykpzoib##t*9x)Se0XXJ9OGEi+-n=VVy*Jqqca%4eE5Fr!
zSzW2Ot0;yviMG}DI3I9XS^)JmXH9F&c}{&(ib*VshCL}2+bwSIqV=yu^8%ZG!+aPm
z_)2uK|6xJ%i>>o7THZU1Dxn^&q)ifjoO{^sE+*nf>s9|g>OQP-45GRtB3!Yj|0gPy
zLZ+XD1E=*9U)L86Hq>`CkFD)tnxE^?;`Z>{_L<pV&=W%wMH)azpdnF{0NeF@F@P_C
z<goLxG_##od3=>VlP%cKixAE|Ewd~T6OM@z?GF6uLC17^WbJd+r<<r~7J5xkk&F3s
zsvF;Z_^fLn>@hd$UMHEQt*##Apq0;mpjmWNRa>zV5Z@NZ3X^(MMf$r1+IN>Pol)$-
zGK6K5$Mp^akp#ib*k2sZ7Ri@u;_sx)28Nx5cz4CXbdRuZ^{!OPA{X~LQ$!0EjoB$s
z>@$e@Q>llSEa@KtymgX3^b-_ClD;Cf?S;wgLA2lrkhe#(ROcdY{j6MM5+Ol38}iwv
zlN(tixMSGIsV}N+KQ;<oP8T(d@~kLrt4N0bl%5#iB#LyYP_jhS%AJY>m#gj<;qh+V
zs2gPQOv1&SqI`Y>Zy2h>+M(MQ?1eH9`$9`Q&UHkjs?EKCK&k4hY~hviQmna;e;t(n
z6jT4JhX?O8E|P&Zes4&X=^w6GNg+5CY)uS-fo74@`cr`F>gVe$34g{hQX+lQMEov7
z9D0=*Z2%@+M$kvM3wES~SsE_391EtVJ&2zam8`x5XXT+$@qSJ?!?sty&lHlJIna!#
zTQcnWUZBi;Ezo3qzo&?}O}8vMLhQ^aU&YCb|0N{?XFa=Jbbdf_=QuXTw}r>w@2EV3
zN2$emO51-gPD)A{c4GIIz0258526UeOa&~)N%}A}@)qr2kH>UN?^s&Zia8Pd4-k=P
zjJ-(T4wx&1AE$oE?=ho}sg>zjpA3~jnZ<CmzYul~>oq<bOsqbvVe8-Tu}m~to<o6j
zOp{fm#I*rd6P=k{aiT~ErQe`!-4%^8*a>}=ybvrFA?8yn)!vWE)%3r9z)xbMGnNWL
zb2P*gcL}PnF7(M<`cmq_=Lk(WynBDxxpk|=TDx_7#WG-&HSOURkp-fK=VndBku^$0
zay*t!Nsnvz{MOXd8;XGFhMBLnPu9uUYD?tu0ogmF78V8s3Qz+Ze!&dEM?KBB-!Q5f
z!>f;x`a2)~)H@bWKZ+{*)pk!UMtFOu{D0j&|M%DYsrX>!rRthE|6|~rq>p6;L6@Yo
z&r8TB<sugDiOqlc?8JVVo>;eO`C=E(8LM4!IcB2eKI!}&MJjL!7meFtk|on38nUlA
zLRhY3oKLMRo5<|YeS`oXAuy3SS=VcrP@=&WNib|g{w&Y9*z6$5%2IT_ZCsU5$mZX8
zHS$TAR+plvK9Rz(9_tB<B*Vq3%C8uV(-ya9>_MY99r7ImY8DSuKiRN=n@n%io)tbz
zJhueW80Kq9G#2RsA0Sj%c}_$$c1D%zyXqT2CwMpfwmf2(moy&ZXZHek6IjV~)aM@T
z1Ne>}V+QuEqfO_dW~<v8E^vx27sP<1LpjKXPkoZ}VJrl2l`y102~hStH={;g0<`Bx
zYhwugm{N;GeDmE{>h?k~S8NSIFR(8W650;iS~3?D4QAQa%Ny3VkgCVi`$UycYAtR!
z6QH4b%!d9*vUI4-Zn*Y!{YN=|lOU(C){nXWu^ZTaPNH8CG6|`O>*B>BDjc(I=|&%)
z<{R|wn+5YZEQ3kPS=W|?$}Nkgn1g<ay;jh-ZC*n06U9JlR$h^%#O%x+jr<<*py&?B
z8><2%;823dJI#&&NiZ^ua_itMQuD5`3_nS7k78LbU4}j5iOU)cJ)|ykwo9Rh$sEA`
zIRN;#<c$yFSMkxG%Lo6Nug!lOr{wv|r3Ag|4_J|0zWw4Y*J!^~0TD`Ty9|1lWKJP=
z^_7x!@a?e4l8JZZbWHlq$UZqdx|<|R@nVRymNUxCrdp>^3wq4FR^~}*PZvobgj;9}
zj0OBPSXKDx0X<?DkaZ$4w#!ohWMXdbz@&P<AmL^pDl}!{o)c9GzS44Mqhwi$cJ_hQ
zM{XY(m1So>e`dp3{gJy@jgG{mSu_@_3=7wV+%zv5*NN6H;`1b1mH27x-qV_E;6RFf
z0O<#)Kj)4clYIli7{8mK_Y(c>Ips-`oF!dcP6A9LShz^eohw`hVp~sCrkl49wiOu8
z-t1!nq<uzDzG(c2pYEJtTN?m~MN0Ng9C0VHezyAP;Eb=ox>+ct<;{<4ktQf}ZJH#e
zh~}raDdM2<Z?|_P436G1tT_%zHtm1nRh?Vn8~Nc5A{^_h7_U<=hY>#c1(FFT!<=ra
z(B5%v*$v_IHOo0z6hn;V+LLUzOCOaQk9;>Cy;v4$*cHtdp9K*sRGbw{VL!lpZg3+G
z_(y-&)=->Y<S;IbW0K@`iDzo5(!ZPS-xcvLpHIy|pYeHhL(AYS;6lu`*P5qYw+k7s
zUIH6R1m*xT>JfsYQ6HHLc>j2Kuc)UHH``AJnUX5gTG!XY9^LjMk$%YVMz5^M;S_ns
zDQ~_mF&+A|xi(7+lm0NZcPBFkql{yut$>Y}HLYN9=XA=wQI+tjPsE)jYbTz2M<pz)
z`*WHzsqryIU0wv(aL4aU6xu6gdvEGQ{9d0b(>N?Lt0Nw39p19U{m9{_N=~x+eI<QE
zg;iINSwoX(fJa45cjis8@^?1w-Es#>U`r-d%wA%XM9lsqBF={5s#E0+dWsCTwE(y#
z%!OOzklON8H{&dxVN%m@?~+LB-~lBL5uwl77hCV_l|#M1f8OZ>RGe#g#ybA8Th+L7
zm23dF{3WG^A&EvdWzVh0`~wR7Tm}G_-iP}CcFiEH{6X8}qEXU1ZW>r`Dxcugc{}lX
zOQRy$9EKiI`ReyEX87F{EYw6Xev!d5j7K(!OLlZavKyD|G8mMAY)KuQJB2WiH|kBu
zMnMSJ+8-KbxY=;Ya9)1}r257fFxDSxcECzBB^)wq3dBW6;fn``-$w$&BkG<(ZH(pO
z)OIZ%W;^4bd&J*u*rXV32CD7rm9lgi_9Pk)C^S0AH1d}kCQF;lKn2m|GjN=C+yH81
z>*gSn&{WtxLbeC%HGG%2T9WM-Rs%dqHvI7Jj%|kverF5*4fiH)2Hf(VMcN)t1IGTO
zE1k_P+iXXKY{t&Nt!#OcIPxefG1Y8S;UNy0%jtH%SD))2hKc^OGHdqj1lP#8amdhz
z>FNv`5OR&h@r8D`y+C{yNrs8t-nA$F?2D6Twu3@a6K$9ZlYq^0r{rBo27_kdt<A6@
z<M(9@J-6PjSd;bOl)t9ZFubX;LkjRovHm=^GXksRyJG`cO~C?8-E5~r!rH#ef*HW8
z!%5!6`w!-~qUIbZD~!FP^K$5pH{H~>&6k71m;MO7Mo5Er#?aw=@gnn<=0&E=>89q0
z9<n6{%@!Js?9rO5aLuU6bK|8Te4aYvyB{+{gQ|ZN>b(ZT2$2VQ31Yu7$DQQV@Z(`h
zHfch(H;3?Gi|JY~<L$)@MxI$)<|5e(Me!6-;Upy)J_hhl@W}?nA~O$IWl~LffiK=Y
zW4#9=4^4FRbm}4PQ{P>K@B@&b(Muzi-}ae2u!$??Aht^DX}+K4nEAofT%?#iJQROA
z5axfw5i}8X^mm)w`3pWZwS1}D`F6)7o_vC}OSA`qef`snrtGv&JMe|1(BR*+e6y4M
zQ?%dOiu}j+N8fp3?()9ifPj}DiF&KeETw+6pKIuEopuEq8ixYqy|ZSQir$hWjn;%O
znZFOs{hB8~lXIF&z?j)v;)T3=fBW#I6+UlLzAz?&w$CF8zNo#fIT&BB_SsQ`n{xI(
zHO3f|aHFTNK~q0>WwIrxp2949>$c;=cfF)8`~ig~S=w?9orqx~wnnXD+Ou!A-a?kP
zSkL(wtmK9^o}9TUA^dakS?m0`&ZK<4XnH@p7_k0(sPQ>~)eG9xJpNpyjqy;~S)}~k
zz{Eu;NonJYcJ8y=6v-sU$v-n*MiN4yMvHXw(uyh&Cnz5CwA}Sz)vI!Z%);gROOYd|
z)~5)5M2NHEYe+a$yF@lhH467d`HSVqR&EqaRA&wAi6G%!Ug$Hq_qfmJJ1<(oo#MmU
zs*b_^I&_*St>ooo)C!CBq-gp8qjf1|LI&s{r6`I>HW{a6D?+OV3qLLeN$t=LkTQ!R
z&7b_Z=3ngckx-_G^YN=7n7m6Y!zs=j0rEj`#*T7HFcn!SWT+5v7^vdS+}B7P%m)!y
zBMjC!v1mWNR!*wVIfe-u!M0VCzO~fx4FXR;$YxKNp1`ae?5j8COLY1QHO1n}&LMb#
zkl+VlFG&Jo+3zqc+Lc?a%@^HmGMSWg25g&ZRfk93Z|X7ga*R!4cjmmjX-nZEPq1yg
zcGp+B)OD}OgTi?L!xVebw-77p17=fI<jiNu)Za|fR&qV#xpt_WdVxsCadp28FQNfx
z;1irQh(5lE8@bmc0(R8zM{HUDpaJ9D+k|*_bcQ&Ks95k?SCcp!>wZez9?t;1PkL?r
z`5vz2mAMo;><zPg=90{GP=`3|C&1VbgqsWfy)R_bwEbUS#j;m?HJo*4hG{}MKFu(7
z54RZ%qZ>n}I%T?kf|F~4{Vi+8!x(G2HHBLXs^45SpCN2YA_Te0%s^K`=gl>*Q|?Oz
zB||nAE)dl<thTc)Wp%rxBM*RutjyRz<#Z8wxZaT8X+yS{xyISeU_xhZr1Rhm0*?qH
zTOhjk4c2{QQJ|&5J*s$x4=2IOa!3RxYA+AJd(k@8^9_2K%<Ah9HS^mXo~-vQJ1j&M
zt?|j9i)_7T;biZE>ulmwbv)O&m>$wEO=P!oG#viD9eE=osIaajul`eera^PHg-D-3
zU7i37X)Z|-{&s;LM;WC_<gArkLZCy1<ZU6)@1HZKaN6=zr@VX&+f<QllYp(Z-DT)!
z!ID#LnYo)tj<Vcl@_#(tEg0qm;hL8j%l0%~im57T?3?+Fs49_-QKBi_k9Sr_H&M@#
z?kY3J1}Cl^bHZzOSn5KBWW$0V5O+U;^gx|Sd2wPO{HoTpwz~1s{K2>V-^k3i?}y!c
zokO^$ES*2UcUrlbT)h^1m}e{LiGXL(o+4_ETXXi3$kQeSdU*RnT#v-jZn5mD)`LFr
zEZc8Yl)uM++addCNQ##TgLCf1pn^LHba?w&UfRub9EaJI@%}Uzw^Pda#MO~^A~&X{
zqUdfZUsuZzcHo(Q%tD)mXdL8}@8d*Rmt+!PCox-P`s&2!I)V~<Fe}nm=_E|%@VXh}
zCDfm4hd5EO)i;-p5z+aT$3&^0g`of4$IM!7s&e1Ry$1vqXg*uR9J-IwtbPT{ttg+A
z_?XP9<%=bspDulM_EfL5?`s!5c3TwKyLO0k8g#(!81r0xAHC0gdL0gtaG+l6S>Y_g
z?=Ea`SO~an)FpO^lhNrjeD6T*GeNel=R`K1|F9*##QJy6qUQ$F!cw1um5%>kef0m)
zT%+P%;lMASeboM|E2AUhAVayKQcAr;IbJQmO3^JOv6_aI{z6{G>B-Vq5V;cJHj;eM
zv+T>t#<O30vnr2=6JIR;ii7VjF7khTVT#1GdUN>Mv;W*aHP}k{A?R7apFKWzeycRH
z_c>Ra(*KcP;_p;xEPbD9(1anun3u8$SWr6wr`iMpr3Cg|zBGzu5iS#+o=>RoD;)py
z0w}2Zt8)5}A6gbdin>Gj*N<P`f3-lg(wDZO(l!l10sm}nmHBzmC;PyC?z6AP^SrBZ
z*3A6ZurMvbNt;Ohz<#pDJF%lU3(l3^R7&@0FKe0QQjZ|%zy98T@j{jmixl<sRi4UU
zoVIKDPT7lqfXsmL>e~6%aKR(Ft`oBdpZeZ&w>!9*SVoc?=$V6v*CQZ3vRHb?UY1qB
zN7kFk1W<;Mm|7ger{uTpd%pvEjhy4u|L>P#fVq_0`78gr6bK45$^`I}W`=7;z&T+4
z#ggMJo)I*F`u=PT9A$wqL-PIzKgeR(qPT#b@sxEi*KM5rTZY%ad(#N|X116xVo%|3
zi{O9T3I6eBn+5<~1+d>8|MjH)-@W*O3pTsz4;(W8@#*}}{~Plwraj=~rx7&u;``gV
z`@j3>|G3cq`2zmu39z|+A_4e4&4BMbs3f`=`nMCr|6{ZMYj^3P!iw>kT0mmVhjZo}
zZl9zIT5EX%gYZ0|M(^bR<{#A%3BWYam_ASUR@9Tg2H-L%ipiVL7`(D!`G2?A{#ph6
zzi`q|TG(Ho?yqRu@;}8#`vcH4qyc?ctb5C}?jQ4$zh8I%?x(q_jadMs+u5|1@wp!~
zaDPW6hv{ece|KS1V6bHh_WI1fXVm=lnf|vI{H~bEf$QLXroR`E{aY0acr&RL!_;XF
zFV6orLa+b)4E*b1pana5@B4~}s()?e|5x9H6;KYLt(@n%*8h9=STGcn5Q6MqZa)0m
z&gnnM>YxEI%*GCTe-css-+FKxdziq)wMUIB<k5fYWC7E=|GEADbNm1QUHu{9mw(iL
z2R{_%!nJ+PJpp3>BdHP8f3)<+RqY||lFI_o7e9jjTA@QKv#B$v+y-ATw@a^=UF`)w
z$-(hI8_IaiaCy>h_I>FcAx6@+=2t$4_fxN&DS=$&zgnIFoEU6F-j?b6ip0#RR@DxY
zuo+k-llvdn%!B~aDYxW)a{qN;Y6trxYXv9=r0?bmv|-Bf5RCk43JgK#N@M)&^)Z<O
zfbzwFnPCx0!h!4zlR;PEaq$Y92m)Tc+^+dop{l_UAm+Lpj5!nLCTGw`2e|T%K*+;!
zdk1_I%<p~oI_JQEQri<;6wG_Z{zUHZr@g9WtImJ5+CLBk^x?u~&d2+aw-XIXOPKW*
z_48ps85rrmZgdWg4jRm$Z2EXJ6f-}Vb8z%A&^HgiU6v?ufdWig5kM6)Sv&oXQHN31
ztt!=RpAwsZ&w~UpAZv-GNh&889OU%4tco!naOQsP8AorOWByWO;2dVyyE9MOAS$^c
z){p;3Yz$zfllWh3n1q6GVAl6eNPIK!wej4*w)LKJZVv%t_+KG`kPGyA!=;2TAYu|=
z+)hM6&J41^`0#W63JHqNxrJm-ui@-*6#&PO0DUJ-zeKGZ|H|U~$3t<+y3@fD6O76J
zL!I`+k0N8dhqI8F5l%og+0|n}*od4C0=Yxx7+VnUnL|08&(OyEe1Q!ZIQH?2Q7Q0e
zbTwuHwVnvLWIKuOb+9y_PaoHJIX0ZkM>g9T@2Bh><jK_S2^!6exluhbaZ8ima}>va
zX(XL1$m{*>Z#VD8VX{@jozO$kVEaq(5jRTbAe@R$z=|dwJ_CsM%|MxSPbe2e&#Gcn
z*(iGMfnDN&TSNsfF`2cVU}CvC)4n_NQ_qABVCp|!d6FEz7O7==4^(#{#mwm#Z=#b<
zZGTb<*t1vwcym=fSit4;44BD?CZwE|qlNbS?M9kh$4t{cHW(RWqFtzscO&#X$ELw2
zJTpP;YpR?h$zqo=<Bpx9Zw++^KlKkWIXB2|6ST|9enp;CFYcY}Q((KW#N3D>8u+#2
zTml}hllm(R_&o-=<jrtYJN!Di_PcsR8O+(-$j$Uh)*JYVxiFnxUd&vESW9id{t$d&
z=40m&(`54QFPJ!*X5I9?n~srP6QAqx4wZ23;-nwfu6WUB7BjMAG4p%jVB{Q57_;zI
zM*#IEV1-e=zwrb^d5A=FN)7L7E<l3T3lL;#)0#8^&has}kx|Cnf7_z)^Wcb6qL?I4
zKy9}@VjecYWpdh80yK(8!O!#q4ZM~TLX|1hognm1`Z*XQ7;h|DZaSv73MIn4m_Qla
zV7oA`66IsC|FnG>V_NR60Z$b&rr8LL+p+tZD0*J=(IhgUcS6m6PkrYNYq?K9UqkG_
z6XeAwaf#rFwQv)2q1^EI;_{aPY+EFTFea0743sB)ewUckm$jW}M_-mgAh4)Ahh{qs
zj{%=tD(TZsZKAM;RTu^7b&RqR2QKyeAH8QQPQGr#aosj9cZ2EU1QuG7j~IJ4KVm1G
zjfRWKh7bTaPHq?n@9=qbs-OKHpd_JEnw<eR<X80!$3}M+Y5=3?Utnsy#@YsrQW(6J
zuj5z^w2fHvRs8_gAF?`iqcwZrcaL-F@LYq5AF(H(J747VcZPMZATJd@xHg$c>q8Et
z_hMs$f&aeDrEbnP!$~6^$_s%=yB~U6yu}#<a2l{Ka(n@7@<Kj{KwAkI0f%0q`Y*3z
zo-pkB2G&QLHM<NZ%l#SK?UUxQYk<D4asq@1>lE)rr#+^A!_g~n<N`P#21glBEaXT^
ztzs2!hP`nH=DIFsotO%_ezN3x@CPUpqYCXx8V=lx`RtDxw;BJ3vZ)yb*Ejv(UWQOK
z*e;|uflb|Xy3;U=3SHNqFv=B+IAGar`xF-*2fen0Lz46ei<2QP<t{<upKJX(sp*T?
zWIkXl@;x9rDad1^s|ue(s!dZL4Juh~4V*{Tyc(ZARsC71$_fO8Jl#27k+~d?k!T?x
zrFer;J%q|!tZDlna3h5f3cl;Q8BbN&@X;kv&V6DlRbO*(2`LVK#UzA})-K<h!yF!v
z^*$ire%LgFnVpsZg;JFzk>ahB)88Hq624%c08ZOwYGzcq>&#E6ORH5+AO^9`^Gunt
z+*g=oJ0)6T9KE^?u-eh)Yy8_~fie_~K#hF119tn-p%oA)R6q!WUatoS(Q#Jne*eZF
z=xL8K79<}d11_fN2LapZAx{)XFhm&HN7{84;1sF3Nqi`{MeZ-4rs>mqG2A-*-K#sm
zv*p>9L*f4U{NqHJ74>Bgyn+U!Fzrv_oBN<ilV~@TPR2G52TIz5!1jRyOlmrOMgOed
z@3$)^4|pdl_S={_p$tK}hwCzSD)H>z7e~Kt>Ve8;OF*sgkdp-Lgvs4L44HX)8okIa
zgKvN9gSG-q*E06R2+*622Q}&RJNhorG@D2Aj1W?^L(1fsa(HC80c2waXv@?SXf#;+
z9tCk=20P>JzdL8i=L7R><uV?qIr^mS$B6=a;FjMCwqMi@;js_fw@g5qr`8JoIVL{g
zv^-3{!3Gc&cTs8ZRUjQFstww4ow`Ruvg!j}!SE#f>C|l&$HAGt$gqu)ki<+}V+biy
zO9(E4*(ICKeZUoZog~6f6Oo=3H#)%CJim?1Xrb)>u;?c2_(o(lDB&B(GvhYB`ZCFm
zlo*E#bE|(DtiJ*k(pJC?CNS3GIvdO;GiyBKuq^{;6EUH3mSn8t=I8h9mlLR*^g3V~
zCCQ^oyoc+y1t79jH*rgx<%2Jgv<ujS{2Te@`8BV_Ca~8X&*tqPA?3L*&T9>TJRJL7
z%|4B1*)d>XNzS-Yr3~whd+t!9qt^0W^X>9QY_UIyfT6}rZE8g|8$*5W%y!+ev17Ao
zz-{lsWec&lD5YQOlUfq`!pgZCqg%SrC+oJ;p;^<pM$DS|>>V4Jy`+w(6^V{ZKVWU(
zyAVMf<!Sm%=N7G%`Lar%as6|{<~PQAFWu59jG-mf!lkCa@zB-_UNhff5gU^_{@$-?
zVsMexbwFh;`Bd&l^A~(mr3|vFGFJ;v?_nL4==a}jF81*EUkx>ZbNl#E{xFlWj-z(V
z!I2PCkob%yd6-W>c(F{9)Ftk^JdiBKBn>Qbe>v$0xApuKywq*|)K@64cKlLp7HrBX
z#Wc5N1p4k3#mOWWoC7P^bYt=(K-PBxQZ+oS8jTlVX1doavsQzSXv`+j&&P#hSTh%*
zriCn}tEZq)kopl|`4M}Lfnq~TVyi;FRDl##TGAhg6DgPX-sOJAhpXN9AK~2ez13qD
z*Z|xC1}Nvn_9)C~BQ^D*E%J<nTtiDGQ|&sj8R&z&w4WftU6<<Sd|M_{WQ`^E0kPkz
zLVeW<C~m|h!O8G@pB*2qk6H|~;?$;a6`iNvl_j-ce$tp?9(yIHU%?L}{R)7|z7i%J
zXv0)!PV?dHuR$3%ZUH<Sd_sX=q6FjFM?X$iTH>DU8=15?9K&B;Mv<gG*xpME*Hbiu
zH8q0*Z<)@3qQzFhmZc2vjt(jUPA);c<(V2WW@}ia>7$i|VV+f+ruA3oY<rK^Ab~u7
zBhIHn4TGn?$3>Q7{_%@wN;&zF0gogR$Mi@ts63;VEO$`JwRd4XL}H6j$BpuSw^QnA
z`@6C`?b)o%H(I>IU{JHSG7uTEcucL7cbJoCyl$pBCwqdi6!4yKj`o$i#r%%O$~g&*
zWs4Y#_iggAZT0}{j>cDbp;XS`0RByBPF>$jAh(F2H5~6Sn5|SjM=Bc!)6FXm1{|Ay
zd5gaCBz+%A%tjT)-a}Gax0g_S;6F^N&OImpM3{upmnB123o=th`y`EIaWKySaJ{6Y
zW-m=H4^8Na6(8%r#CSeR+`w{X<$?ZeqBdhf-af0yzU^mB^_Cq%DF0|MgrzYO$5?a0
z84!6xFg&JxXdta(r$$owz-yEWn(+$>BDOM2bZvn=oOp6E9Oql4OFHWLeqvQUIMWZ+
zuI}ANk%&J`d>E5$;YIS=7<guYif`MBojm^=^l`IKlFv};^kVzvp!*w{{JaigFO{<g
z#Vh9<YxTKafh#ox*#7sv6nhGbwZNj>Y_h^c7Mg^PK4@vCe%rH%QTuH7&~O5sv-Lq+
zjt-5H;A|LCJ=o4I6$o^1`x>RQkMK!7xU&;_h60?i!cmw$ntfZa=$SxxWoTu*ki4Ql
zurcs{nse+ij-IuiJgU_+8nQf9^*#6p&BN4+)YQ;zE`;(ZTywQRkgo=)hL4pZZa^zi
zVeQJK!cNcHcgJTAze`}LF!)u(Sct-yxFXhm-&oiUuOG#f46|Z>#ekXPFzcO2Nfk-^
z@`s%CCDK~!i>c&TNa`PZK((oWk-s4rrb;8s?eFQ*xvl%3Z>`vcv;=q&w$a&tpmD!H
zq7l7a;w0$NUV9bpT%iz!KhB+AByij+`lZ9|@n}V_UKBgYTG@9$wwj^^fdseZY-btZ
z&O{JBwAie6-tu}&?dAKf_A_mNY-zJ!piuWC%DG=97DB=z*(u=BeRjq5-s7p4Bgs8(
zS=x<^IEZ?_B}w=#ii_v#dyA`FiEDI&Qi5xIb|MFPZ!M&w?*$ZENK-xO*|q$I;oaEb
zjoZbclTqG10_`W&gon%<u?y=SzF@f!q11^Mi0mKcPO*A`-tlaC_rM6QWI71p9uAb@
z#%JWWn0tlGM=@eU1v6C6Zl5P6EPZs#3(Ot|oE|cjacId|n`&T(o&;PAYSpCR55-4A
zAN@GB{LnEkLYNBnqBySj(vmk{-^UZHkIs<H7y-&+Jx5dy?zKw{5t2`}oFBGmb3y_w
z`C#B~&x%xXzkIM;Ka0z^x(`Z|5hD*H>vk;3XqTrQ)=eoZpxC@?t!Zr^Po7jyx%!6>
z6?ZmHlD#=|MrEXy6rCZr?QV>m&$!~vt?O<g=5iDK?&i#raxGdu%0_-5iM0m0f-&X*
zC6r|hqubp$R!ANdifJ$xtwkh9*gmvRq$s^Pr4>~4QaceoS+NovREfYPr`^ro%H>h1
z&^ACo33joAR^zn9J3YQxDm~4<in3|8<r@V&CCd0x#vrFaxx4+lbkndW=s_ZM&^?Ld
z9R>mBEsj7=+O;Oe;1k9c76(V*qeuDi3$RJ|lLuUd1BpS!tA250TB1!00YEJvP{r0u
zZOB~JeE)6~w3Gi;8;LTGIHVVvGzUFns?XwkM+@i!Y$HHiu}wIKcbe41kx3l<swLDn
zCM;U?HlAZ46?2hdQs9G~UPu4(ygM2A*l`sQagw|@9(#mVoyC=KYQ_ExnHg$Ea|OwS
zVD}d6kDM23*YfqhZ*-BU3<kc;V^B|d>wZwDUlqVU$LKpunaxL0k<V9>p5URm5Wo?I
zYRsjeaD}=*)Gn37A$dIErZN-j&v=M4vYeFPbu0olm75urOOmHU%Y5?&!A<cpT^tlc
zs~gOaoN3LqU2suD?D!oFN{BPchPPS$8d3>qHmVJlmk-%IiQOn@qVffy{mXQ7O;1fn
z&a98pbG@)+s{h<pDDX^LN*5om<IKNr3p&v>Y97O$<9C2nN~L92FK?IH9?QsI<rZR8
z%%y8i;pB!wbzhKR4<ZK87E0u1Pk|Sr(3b8j@lu?#&{Cjq5E_D%S=U<U@*JWcV2~U0
z2l1%~#xs`ZLrQ;^2?tl!g8nGH8>3idWl~DrNvHEPne1shGYXMop}oQt;Rkk9c;<4c
zA-$XWw#tD?R|BL$W~*;)^OTI=Me_W^dbPQw7QC~|9$fx!!A!x<Vtbr4Hc1SoFBoO=
zhL`Jflj#?tRA{aZopPHKEZ0&s3;xddgb?7n@NQEOZF}9b&Cg3MEF!GCAi+Hqb(kcg
zo_C#JOW{PeriJYxtwja6%^N0^Zkm**5nqkl3@mB5%FWUj`;FQyI5&FK=I_^AHOtu1
zbPB|Dq43N{>@UN`;LCdK1EmyAhJ0=mh#O7Ax)m$KO*W*TMY}}&Iu?_-o}t}`B`vDq
z;u{*s=y!pK*O$*oIJUuQjqN&B_?X|MuNub=2lV~{G|u;(lh)v+o+^Io(y#GTQ^c6q
zqM5=N-0rwu_S|R=_-?r;Q62rP#4Ojzbedv|z>Z)zvL*Vk##xm6BO6y`;r8x}5x$3Y
zvVKjzwof;xRHzp7AQfz;H)izHaQUBR4V=(bJ|O=R_iNRQCSIw$n$&^4t8xE{XjXDT
zbZ#L1Ni^Ze8)f8+gz(E3k?{k!25bm_pntg^D3N6=Ni2^xOa^OI5RAyITrL=~twugE
zJGgij8Gt4-==oifeqkOtF+hRKby*(4aqc;{f-#dy(Pw(9w7vaJDbXfv4(T(bWY7^^
zDVkbOf&SWwbsWRLk`@j{i5QqhRbHXjd=%SodxBN3;;K?bYn}~VLcolh90nM(;<_rq
zEci6^Zptc4iYe1XOW|m8v*!=vaiqUc!GnRqamCb@^_n!{LHArw37e6oOqj;Tc^aP;
zb?3&5)c15JJVyK*MX}X>??GvkPMo-fvnM>>a&6$6+5p8<?+Y9<9%ULy<y6SFPtUc+
z*mF%B5(m>?$22AyKW=L3Q#iaLcZic^tu>M;KU!L|HLtM=O0$VO@GCA+t5P*-asTd>
zvNTk)z1))+<F{$y-=InHtmqxQLe5f%#7u~*wRDr3OxD{!hd9v#b&cJi#pQ^<UmSS>
zERqz<ohURp$a@K3M=R4S@QYFSp;PLs+bOTIJkp3y)SeE+ENqLW;j6J|>>)f20W-{k
zI0*mF1>K^=UXiBn0d6bXv)bbOX=ccjt54RJg?Cl`B+ok8cB0taHKvcygWG}yh=E|K
z0pT|XMf)EMBx4UN)WrIwXAT5UW6@<>zA^m05@cV!z4>`LXo40q<|Gb@{T4s>QH)BO
zWIL<>Tq!&L+DUdYq(oIVH|Lsm7UA89&_GnKait_oc6f4c=Wl32Xf#(>7UEFbz8Ue=
zxg7S7i*@oIq8-}3pS@R>?vWUr<W_w*0y8?eqtUUBp5)!MN!?>U)EyaLG!ojiv>msZ
zMDG+8P44Hq^@?_>yvvdP$HYiO`s&2JWgiFUyA#NVsHqj{l`V?_-%k#!doxtq0&N0z
zm(nq*3}jmVh1OOOYR3uYjSjISwMe5bk%OAe7N&j)wp%-=G|Te2$j8aLpJ@<lsxi(j
zSubkYm;40-=e$1k>bpLie|tkEt<>u1d&ts0uqE4GyH680!U>aEl(_7fRkM9mlW$ku
zp@)&P_a`P^d-)pm!%Z?btMC2IVW7c8lWj3x+Oy=ZSTFAk-9s;rUS+$D3T8Q^)b%tu
z{H%9cU7t3x=#8_ilTuZ^A9d=XVVTxROL2xf`EqWpy%vE_wJ*2llf<rKVw&p1vev*a
zHj)jZ8<%b7?N%CZy`fH7kIcls@Rf=|y=}Wu$n~(j?@{7eiRQj{f@j`6+6jr}O=^pA
zE3TEEdt}K(g{_Zw^^U_DACWeacInfFm{sGz$HPij6+#*tDxi*>PA0P5(}~h6xbR20
z+*(PTQ<QgMh06R%6dJGb(Hi)hggY8U&*%u3Dp>?mI%knxfgyLqN33q-?8#6(R~DOW
zosDLg`{-;+lB87!v$^AUvnk`#ygGaZul&|3`7D1^1}vHLraEiP+AC(}k~H(Rw`dHO
z)L6(aYn|raRyg<_GR=J$^~`)EEfa=v*?lzL`_h>D$~Z51#+AMqJ<Er#oZxbuemeA*
z_wOaQdbPy4E0!JO?=`G63_W3471?;2gC&yAWO(2K6-v&@t`+&xzyR3Z^h7PGmI&d4
z@BMy7-&z+)QM)fD>DD804jdzBnRU;uhiUBfxoh~)SJHW$W}MXvI<fOsE|hG0p9Mm!
z0dJR6Q$aCl(e6x4cgd3-seT6Au_n8AVag`gwv^N%>!c3jasBjOkmLc|haPDI5qh;9
zYl<d<)|gJmNx#*Lr)TxB(Z}f>XRvSYtuq|Xs9)IU3UFnnA6cT`&|A^5rX+kVGq_eX
znijV4RO69|HZ-89`BjD++s))L3s;K6yN9mH3Lm;o5Jzt-I&0Jl?e@%Q59vyI<gwVk
zhKiNTQh(BgE-x8T`bDswSnTV;XhKTJuUaMB4WXm5q#U1)MrHZN3k<paG5iSVWNRFY
z^%tk<RS`Q&4m(*(?F^Tt!?KNwgbD2nB*Q+$^aw^zf<<u4gQ|K5ZWgXW$xSihZx0@P
zN?P__Vqr=_c&m(X&s<ODlkKNJz4~%7=UtZ;62^f8CsdbY<B1_lPd4sLcC^yvV&+pc
zjMrOM{pT}{r@lNT5FDTtgw~iY^=d0LJ%7E1x;BKW6Fl`v35EaKFs9rRQlrczZw;-L
zqi34wc_u<0D5Kc$>Bh5iAtLO@YqwJ4mK03~@ikgv>uDV=BdeddU!zmO8<&qx{X~a4
z+}7ha>DnzmU1-waUumE8J1j79bInK+Xf^x&&LWDR%P~4QwdtZp`w}JNlU<B|j>1F8
zO8}JrVyp_|71pa@kh9*9{G;eV)g*nTsvKj$)5t37+<dd(1}$FSgKw>OL^@NX1~)R8
zxX~@8s=0f7=+aqE|K_XvbxH26IQHo*G08tZSdYB7XpEq3mvQ!`+7y+M-WUWI%O%qF
zaueey$N%y92sNIk<;CzHG}|Oi9MKme+`Ff-!L)Lsn_>9n_8aX2P6u2njYO`MSAMIg
zyA1QF3|rH+Yo&1bt?qKs9&6FarQYKm&6Vm84PUq};6+T?QxoG->Do2<_@oe!))gGt
zqjwS)o;$Roe)w#l!>GMAB3u=}@IVrz5##P>9BjEe=jx8Ac4B}+L@@NKOH;keD?>Z^
zUQw|aOq^eRh)kfp;3SJW8R!%)**uT-5uLx8OH1<Bk5?+=!tTi+D$Il?$y<i+oW|;G
zO~%DgB{|Dcb$U~rr)+!Sf>xoEq_A>ldFhouDciJI<sh3ZDoF$JgEn-b!%E{|A6<r~
z-d>`g{79h@MGahyk^(+dgsZGCj9;0G*`VE#bPe#Ay0Gnv(7b=bXS5XFV}}|W6kn1&
zw;$@SoNL{%K?u)}y)o-+`JS%#1Y~@eeLM6Q)QM#<X{!1dQJ$7+T%7t7m*LwZFE+S#
zs$_tPQ7{ujuJ|GOn(X(5B*`|>X4>}2)com<o9kZ-Q{$`IXmYDFFu<maclyG#nDn=v
zffr7wPM~QT^f1x$BS2&}Ij%<u)%S(A3X|ZJ6B$h8+5EAniYN0_cj;74RsA>q;q%+r
z!`vtuRZ4y#wH(^r=1nWr=?74v3<kS|nA;1^6VB!7bT<3Ff$$Q|^(Knf%|Kg2u{xBE
z_CSwsdZf=)=`CZemb(VisN{_=hndna1*Y91s|4PI_1`Vpf_re;+XvoYhK^?w=iQQg
z7VHMxyLM3~J@U~EI`3aRCnIj_azYAFB2h9IX2>=j_Y!^3&t2pTma}CT!%h-UozpwT
zm$6lYl9yMzlLfBI?DbNF2ZgKq8LHA!M@GA7?kVmGO5`chm-QC-*D`xW`nFCrfzift
ztSgC8b~|eei*xet2HHr~j45H;vok_hi>ROu;-0G%KNS|~+Da?Ux>jKN>dYofbQc!u
zVLJO#rz`9JB3blv02SY5p--kUkM1rM_mT#D$=)EIG0L{AO)0mw6ife}#f_-uO+~?&
z!0_$B@Ib#8Ls|kmcKlWdJ7DjO>CA_3pUBmivum9^b&ODvE?nH!ah^?PTMF{B8&Sr`
zbE9#9I*?jCkbt!n3tn2}r3GUE*P!Jshs530%sy7$LRe}EcM?@w48-z9tFQ^4L;wdj
z7>r==&dMh`IL{G2HMD=)cZcl2<oCA)rLC7Z+t?&<iO0-8GEAkDyw05^_NY|KQw}Rl
zQ$N*dn<MK?^RRw?-iBEJ@@n&)N-66(*FBavjTPhFn>}60m`NpMH^RihuE@e}2@{VA
zIi`O2(amYr1|NwiL%jTeR{c$0>=|<GI(S@!&h_%_`W`M*XS~3%kE$lx;IwK)#+33J
zVB0;MSy~tvAx8Aq0j$I?I&_z?{b<l6Z$?uFBg`UnUFoCSh^>#3Nk=WR`$$=5+VvaI
zGTab(sRZJM6ik09XyLFWH&zg$S9hF*g>P65oK8N*fm?O`%#fOWh3#FqwprQ{loYK@
z8H*F00aIB|GeSUOx`IQ3YPm>J0{Lc)78F>1rAQ6$Guc4vZXEbato-*>kM5TAN?c=y
zP~g|VK6YMiS`Utr=#+#B+BbR3;<^X)RndK<fleeR)SEszx%R?hpr;3Yo(c13;{K&^
zGS_&eJN{s;>{2rtdSefS>P#IQpd*uS5Cc7Bfyal>yur&CKRhrelRgzbBSNJm@Xhn2
ztQ$h=6<1>qfiJxD=99}__fIF!M9B_eiD~cebr+0hcxSZpoY+mKA7h=x5Nk-^RplU+
zTH0Q!ftBv#uL7?~v80@aOat%H9`Sf>|G{E8$-$h33#+T6l-#dR)o1Wx$3=1BwG49C
zwxlm!Z)U%S+3n+U>=6Z()w0kZL@S50dm-=z&$JJg_JW>J6z;45Z?5i!%wxYPU|w;*
z8#i6Uvph^W*@7h6pXNfHqJpX<{pFzAH2W5AVKXrVoutap##D!>Z2$m1&KV3gDop<=
zrd<)6Y@^vQ1!{h!Kai#f)YX2Dzg*J&D%Y&Lun1GRhf_j+=T6_0*7IO4`jV9Ikj_Zz
z(EI?0c`1Vv%d(9$im%fyiz2n>X+VtjCM2}}EQ()4-blA4v+o`oH4J6GInm<wGfy?J
z!OTr5L0W0($eMP`YpB|ab}373Vx9WWgP&1$vaOAXVkB9C*TvTVy(Sw(V6TsR^-S5d
zjVbKLstogpAmOfbZrPBLFYSX$*%)|KLlR9rFg@^FkU~ne0SGIpW*K&d%U!~ptA@BC
z#T{76*fu4xz+BEP%`f9Xkz{1kRx9ATaq>qTgM*<X88OwsB@4Yhzr}8xaH8f~As+t=
zg$kr-g3=GOg>wKrPg4HSrG-hD?%&*>7VIrWo4wnYiXO}F7ftw%yS%~&Xulr^Y*X0Z
z=Gkh6M_+Kq$1!wfaDOgillRXozVBUPTh#MaGe#|31%}F97iY#RN7M1v9XzDGQt}&V
zUP(v2W+z?96|2Xg?xj-*J0K1-&evNV6-;O6*+;l2ow7%_FXhKTecvR)@Ur^M`oeDH
z+n;2{pb;LmS%bbd1Qwv=@wPHMOTF&(29JYpIC+>jEVAnLhj6h<vm!O4Az>4Nci+fv
zV(pjqkd%!l92IqxR@%P19_hbl=g@^`iQQn!C(wbEltA$p2J(lAJ)MwU7h4x>VS$ZW
zO747fNDKdgJC_afIjBrK$|xykN)_Fs`$&>`Q7c>3T`Ag~AK9<jB}RjXkUG+*lzM3L
z(DANbG-ObVeMdaZpZajUip~{Lky`O3egI@T3^xlGUg}fig3iGUo>J0tKCmd^08Kp?
zs4~qHIqVYhO8>60>=z-NmhJJib8^uuc=_NfY_xG!ex&sOA?&T=qVBr3Z)p%ILAt>K
zlm<t707)rDQ4ndQyOC~?2I+1@qy?l)x<OLu?igyQch7lV=XKrB{XWlq|LNx=6TjJe
z?X}iE*7qRwKV+lcCGi+}v~8yWD{@ypKB)RxygDr(pH6VJPYF3Ue>I=S?X-wx30>@r
zLRk$xO>s&SW4x&x32}_K#k~n8UdDhbMK_bDEDpnmomYh?L4uZzqn<_awX&N)bae8O
zIIOM(-6>$P%y&h$`N7S3WN50Ub$u(ss$AKl3SW&s)P00f9F~)svsavbaQx@nLY;<L
z%KHsBPLJ~5(E-PIt4W*}JSykL`k{dDPj*^z?2|((c*M0YAlW1IbHpi8%KXmV+i6MH
zs))AX9Mt0*r|YzwW4XtbHZM26@9x3pW99d=Q;-0Pr%C2CQ$86=YWqX8xx+2i7r3C-
z-~TSda~s=)tRh>AAuZ*PjReh@(NY!;s=J)@BN`H@xB~`lJzR~^WJ)8E`kZyIEf2c0
z=qAhAV{P%<!&c0y5lq~RzP$O(XTxHnbc^<JUWr$GGf7d8>`2T;qGP9>r*j*FY6>|9
zK}GfSN$agQ&17hjdqlnBASC2EQET=kndK3a*50MLuvw1M-M34L>a(5xVzBczw}HC0
zmXjTf$_>KJS7Gf*ErXX<)qGD!WtR${kEVI%D>X@_n1mT=wU4!-do5%8UYdrg&D~Nv
zQWJaVrlf{f`{Z!3gCX^kLq)0Cb>|VX6y5M@+gvU$hd7ps{jZvXl|#^L3*4OaImnU4
zmh`5*W8_R(j}bwrl43kU<MbJsD=O&QWU*T~Zaq#(jew$1ni9wgb~=t9hY=i?^wk9-
zW+6o%Z)JY1UzG7*Tg9f0bf78GIR}P;djPgVmPHHS=Fzg(o3HpSy*fwIpMEp%d@1%k
zYYXe|U=+e4PUl}8|6`TWIEBVN^+UXm{>W6YaiUI*^l7&U^jdK3`@oV6zd2L%VK7H_
zI5pPTB=3kKzd8K~JN(=w@H7Sb*<jRyud!wolDmzY&*3kvX>tOrIAyBIDMLGRs>LY>
z9w}Af*?Ds5Jt5N0(6L-Xk*!2s0=*UG5mJL3juu%kp)|1I-PKjx{5JWq56v<CW=2TD
zRd8)n%U~2&H4!8^I>01bnt;dSg{HuX!2)?bzP-{hNR$THY)PJMKh3@H#lU6U7GD_y
zw?9vlbr~x>dH;?^U1b8wKdm8A2N&eC*c+xu=Qg+L%&-@yte3^sJOl})=5k!UXxo~m
z5$ti1J35snq)IP%CN{Wl^>Ft`I_}ZtzwR7(?ok!)679;fMzdl6))6F2YS*dqXRQ$q
zA3Ozq@OSC?2055+%tIWNX5Bx$Y5aJ?zp4BuV<2*RsDBL49z>sG{FQ59%Ez6oK5?88
z;=$-he?nu)>pBK)kXbvW7&FpJIBc1RIzlYbS3}D=&G)F+6}5aaaud?@$&x~uONXC!
z*K^J;HOyNgDHpk4L%g#?K8eo3^%iug#&RrAPrjqh>QW|A(bnXMF<J8%$56Y5-YkPj
z^Us`WH=54rJo~m<@fYi}?&~Hx2p1#E(e*=C%_=RY3f(Hahk4pH1GA8s6GQ+UYJ1h<
zw!v2*)FowZpcyN7e2@$C5N`8r>VW=$2bfNQZMIbpkE!<L#^g3HoRj@N(5$V~mI(a=
zHIyZmkwrwBehYn)U@IgX(lINsRlAa?l4se!?K8R$Zj)oB?!@hY=0>X%TYQsWZ=xG(
zw{*#;g&`+j>DganZmAf?q*(*-wU!QveaNNmn-h1Nl*V{5GMt<9@aF+#Gs+1LUvD`;
zQpC+0>J3H^?K;QT18xahkux9BpFY-6gT`U2Q#uJ|tCP44B^#!vG(K&0WjofsbaR1B
zq1=m@#tiAQlS!R<;q?oj52s=tNHDk0v|J`23892(P={7UR*AN!YG5C<X6^P<)^aEf
z@&<F*vp*UZ3u|{ANndYQbGDsxDo12vdEBj4PRQX3aQVqj`We$l;vp_k)w(n;yEFlm
zCUY6)fjZY^h{btg92g=<7{;A7wD2u_Th)JznHn31h=h0g)r5{-j87GL%|c{PUIiXi
zN4UFr{j~_YBBmiAjDr4O<$WR|(s;q6kX?(2YQOsZ_>v>3iLtuBt1fr|xK9}D$k+b-
zS8@`bSQsiIAvNQlbx$e{akrdl)%x?D*dsmJtV@CXLf~Lhl9l#EVXM(g2|ek9MWB{(
zsu}6c`_CdfO%?NXTdy1*o}177m##!!M#;bXcWaV5XLXa(m4##!XzOV-ep|<#?VTE7
zqWv>!i&Ordk3!Yi-x1jNPxHc*gyZNg?{@-mXa71rA$*kv*B;(TPWLd#+sD?zvt*iG
z<_CGU9LYGyyqbLoeacoYo%^>_R`a6SRjEhk>#h&sza4HL?jm(5WEw{8`lNayjr|C>
z)1#Jks4MEwXzKkGmF}l2ImHGe>EDS*b0fZ{(8f|m3fWq@aZ%kT`bJMJb0<Xw7cmta
zMYzxSi083KKxyh{PxcdqmmcyWCAYF3hHtj-F28UnA5^d0xJdHE(<AN_>EANrdL%vh
z5w<2)^$po6?4JeHz4zTk9=k>E++O0E-6C&2J;Qiq(^t;)Tq~7Cdi6C61`X5OiY=ro
z&8UdruktG*oG5u%J?=tL6Mr`E4BfcCw>)eKHPZ&TiZ71`e6uQ9XH1ktJE#d#;9a{L
zDhzu}N-<0-da2#Y{*;d#6wZuqCwhpd*GJdu*vIU(uD{2k=V+vbIaQTXp=a(y^0o|1
z&q1!>)1jchRvSKsEs-3gVt)nSOguJtrc`~a4LRS=ZrW}m2t$nuw3z6j@?xaLCf9J&
z=evKhR*T_P{DZ*`x}DaT%RIDFdg3{bMU=>mikr6(85kcp6_Nb-=RjtLvOEp0S6=!w
z0id&CerLZ*13<cVTE_=aKWjc85qMKRtl}}Bru^!4>s<m|Eo|sfy~1eYZ@-viX}n^T
zwX~$ZaCU*QV3Hr>*sX3;=)X1#GWNp9^r}Opl>gGgW@d=1wbH3fzZ76{e$FS@nMpE2
z>NnN!Y(Y6l2quxSXK*GaY}xq`d=F6-JtePg&OyGGGK27j=BPi5u~&UJtJ=kM^hf(M
zBc~-5y^lCo3@fGPMoRT!QsGNFo{|tRFT|XY_n;99ges9x25!_Pbip~4Ek6kt!^}}9
z)3vdX70l}k@RIjJ{~}$YK%^Xfn^TfR;ROKF8O$m}(O5gREGnTOY1%lh*hI3mw<=R*
z)#Y#@0g2ty;NBX`bDt3u3$hsr5;n;;a_r}QOZcXPxq8~-zd`e`sT7Rfi_OCMRx<)#
zA2~Z^Rd&OihV}ttcO*ic1L3;x?oWsSB>oH5l%>WbnEP}wPT~F>@I%?-r|(nTMt1r&
zmYK>MKG*j1-2RVN364tD5k_Z+z%4lzuXjppcI3JC+e+_OkQZyK>>Dq18UfUl8e0Lh
z6!|w{7VPRj+vt4ENB^^P?=dHH{w-?jXGK+y(AEBK;z#D8*!d@T@~c26EBmLz$N*Pr
zD({U~`1V!pVH`Uy9WKoplG(fcLY<_+-k;ihru^{CoOZPKiInSG>x7T@r+FAjH3(@y
zJ7OssK?T?9?d(w3(kFgL^R@=2-L-E!XD&PIn!MJf8}?RELKldyDCE{HTxM2quH^rM
z3i69?1kE0nO;1K#y?-hSc|yy?G_#0@1cpi<zf9pua+4HMg&1u}CjQc^`u%u|xg*Ev
zci}o;aR@K1(UtnIUlitVPVDpl1hqysY-De|d)Y8E@kDFoD^&7F2{=!&s%$^BksLj0
zL+EW!ST4Q%41NFgl`&&fXOY3Q5YNpBOqckM?FG=G_2h{?Evc^Ee~8yK&awU(YldY;
zVQnc?$_qQAY)?SBO9VfVM*DJR5<mDmkwP?n`<lk5wl^#)B%(SAAEN0jOQK0lcsJY_
zu?_6|Stq6E)@WiO=LYtDv~D|#6AeU?khgzrLi24bdwud<$%GghrrffE=HQztLg^Ik
zH>LS*vp0F3GA~lno^sSpoQi(bM(>mcQcSvV(IU^g7RlZDOG@*{UK%0=WkwuVFI&bu
zj@Jh486*(e@PvTW;s98ODZ9J=bns8_D)K(@4CP0-89QR)AC&TZGe~}~sPD)L|Jf)z
z5Jn4RlzrH~Ul?tw9_*&y;!DAZt$0Pc>(Bxtt#_p@r|WuLR`%!n3ta$@vF~$JtXok!
zsq9WBJ(VlyC}J&1{1Ze`Od+)x{J86C6vyz04aV96o9kgF*bF1TgUe0Lmat3D!_%=q
zHCA~}K>QbNFmV@sLsvCOmE|ZBpVK!Po>o`c&k(PlK(@<}|EF@7(D5m8siGA9X^t}I
zjF!r5w=xw3p<0k^ykFDf6|tWyX8h8(I>P2iGoi%NExipjwBuehMm@>c8T{$IXP)sz
zA3je+q_mB^W`LJQSIehm6~cpImM^^B{bD++d={obrq+wf&3+CB2c1iJ!fWy@PWB^n
zq#RbcP{E%`%^g9P6vKw41k2HF3hC3p2SaN?N_dD~*vrtO?&<Zzd%;oax{Kr#5rN(7
zRFK4S^_o^ps&hk`kMYfAkkWF$$N;9cRGMI0f{VcI6$J!zT8y^Yd9iFf;C9DzkoMWn
z4qhsbGJRRsfL(yP5{y5R5D}0;Q_dCkV1;n8rYN54;?s?tvsWNNJpo<Q4n&dQ5}-`w
zCigaciU@^~l&L%8Gp@f)a8TtnY-t#(vVc`$DR^7YwR^MfLM#;yu3L>nx}aBv9G{|w
zdS;PAoC2!YK<QXt>aHk`P<NTA&hqCM+Zwl(3<dXEQZPh1Nv9Fs#)OokoVr>M_v^Ze
z&!-{xMCPlTe4mEmk3X?^r*8+<d^_yP3)R1>07~}L-I%p&t;r`<Sv$3P$_sQKn5k&z
z1Ub8=?)Swiu$TMqM+}78UC=xx4{H@bm;q2C754L4F$~M+lsfkCHzQNYZVPROCxv3t
zv8O?DATe@&Ia$R)hlZ1j`?u(yFa3^(IAS-fv?3zhReiRt;)YvxwwD^3;;v4(cK-Ut
zwt-$&zx>7f>P^v$zV}?mWAIm-vW)zOVQyfUw{G6zO(x!BJCHu4aCa(LA)#oV1ooJl
zQ?Rv%Q>?EcP9{{5%)Gig@7Kyq2GmO;%<H+e*R{0fQlKtvezs<(bhr~?iyFdg_LfKJ
zo&d*IK=k>C9t(Gz=&QeC(1$~+F2~W7x)r3it^p<+)`#1~KHaphg$>RO{acKlVT}}H
zs>Zx5ZEHm*yW^}!A_jDw?;W8!$~7)|xR9|;+1a}MQ1@Rltbj?So1S)&sb}?hwG?{D
z>Sp~*DAtu6!ALzM(pw&=?3J}4RcrhGXHFac)>++Z>JNT6Zt*v6vY`gy<3@5!T?3S3
z#r}d8%hU*^pYXIqvi|C2_a^wLLSs{3Mn3N&5w&BfB$%T{JdlV8+2>lc<u(gGe>I{u
z2WbHDhe`2l_8pDB2b@e^{NM4XO=`x}zrCG$8CS~Zcgh4L=uhdqlrEImg(SMBZ8WLU
z=Y5$ErKIH97-f#`TALpSElE>5=22_J5FX?lH~;gQTXF2d<hees!OhU4qPF!PCb$}{
zq;|e)_>FG;OYf%u<i(d`5NkX4{L<1Fme?|QJB=>G?N|mgz>RH1Cuntgg-QGeLo@Ln
zln@Ufim@l6u!u1*w}UaLsIiE};BykSmSo9YznnzSQ>w82$4IlT=YR}ec&bghOg=_8
zma0_<wN=QYaouv5!~z96CFwwK0X@G5VZ}zyhm>KA=*AK>y3M+m{VdVND{F4Yv}h$|
z^mFX=<#=)VgdFG@1OV#QSA&w>BhGQjS7QOU`UQnWv}oe5Fx_z3vYrCn2i9d0ueojI
z=Z}q^Fd>Is2C22VnMDr1OE2n%zbNa3s{5_nBodD8*Ziyz^MXtNrC4A>Ng*MC$ZdT8
zQf<yY<=Z&fd{Bl{t1F!6hbMx2I5qgk229Fe5F1_cRdqpJj_c;xn=v)3@T<{S6Axx*
zTN_v&od@6lhXqiz3K)%-4chr0IeRs&ugmbA%6^{fq7~_82kYYEwn8R-H3;rPM0(xE
zw%k%qQ^`^`=4N$f*ivbP7j`p4a<1GEL+F^T=+8x~H$Ni!>>+!1A)gMaLp{TB%zQPX
z?P$w#2j&4s&6lIoWu8TJ_LmOvbJux)o6*ym?5HypY+Mg41T#eBo>4jdRIdM-XvIF1
zNS!-NnhSPF<v})kZ)E#x?&Tti+iX%Y=YyLkt=eT0>CG9ZsUl{6yGY3g@eI0fQ=983
zEZbUgj^F(L+&YssJoKQe7%T9H{d&G^qsR-+;AV}l*bru1<lPAOW0FQU#WE63Ft)F&
z|LAOzo#_<e(wOqj(>n!Eppi>nJ6pv7m4pCx;oYG3rotD;L9}_84BX+Ag-8umLaK2g
zb?_^R>0B=*Sd19EyJE}o>QlO1&LxTYLCCZp2Z`&oI9<GuZ=J7*gs8OF3?!q${Xldd
znSX2*rGjehtRAP<mxh7L!ah1drI)~UmUL7^g%3Y#NP%t!^1_cJ`!t+`bvX=g!`_(Y
zBvPr<Rl8TaCLmlllz&>GaR`3RMh5?~%epEqwtAh5cdI*CuMB+X=J?z<7*&`FQUZO}
z>kCUQ+J^d?aDvesDIM_~A5MNTuo#)?`gEnNk&XX6qPhk&Wo*qTvF=A8h3|RG%OruE
zxA%3HdIZAG#;l^a3a#bsI|JPFOEUMe``5I_#4vW5S2K9Hmd%ftuSsWkj~)(ojEd2I
za96&~eV&mmf7xC?Q?;%&@dPv!4x@<wnqE~u$DYbB(Z%hXD>Z5&aRx>YShK^+$7`GK
z#Ma=q6q94hNtrC%HP7dx#aOSOBfO0cCucQhF7+aAh%CiC+z{9zeSB5x5OVm<<KwPr
z&sRr7QqWdtv#X}ESL_8EqP`zx*2p1ay9uz(d~v`AOFcH<)VOgXB3MhOxJ}BV@=}?&
zg-=6e0>H#bf7heUnZo8UC8ev4>#LV%%Dwwx!WIU)y4OC3W^>UpVOknpJSavt>vLyN
z*EVJGgCCej-j)@c7fIK;RATPw`mg8u#m1OLAw0Cx0^yg(NVK292$K4>@|FB?jCJ9w
zqh)8rc9^}!v7V@@5VgHpb@X@_H^k_a=jjRYF+tel!cy{t$~V$~)PM3sO3`$x^U$q7
z8X8)2tNTj{0&zJz4gJ1USMv3!)rFE4BFBN}i&N8pm(9Up2U5rty2+x6q^kCQz4CMg
zSMpWzno29HluLva&0EdiXEB&M$)D}F&=FF(UvOS|G{s#IZ&9q>Nt)+laLnYSOvs@%
zxbb!+f#bS51F{gpk@#I5DdR1Gixs(MZ)lV}(bHIao3afLA7M_cwJdz9wQ8x|Hbbx*
zj>D-iFIbeCAf~<;#7|(I;_7C9UA$N>(@jZ-2nVgK$;&q;mg-F1Ybl&t^`P;x-uIEf
zwc6x{C)CH)S{U+Yy+Xq}8mt9Y$ncMa$j*o!H}pG>)UlM<BQ~$qoyOJDy4-L>HW`G@
z*n0~?Fi8=IbSMim=B2y0g^G!~Eez*5#LuMrmr#fqG`jqIki%p&i4JM41h&%fc*kG@
zitaUv#9#hRff?KkG2-lrj=_6{!5ZlMZRIZAneh>%iiu@<(WwH5oTOVCE02gj7`3-Z
zMee@05p(DmFbh=49qwkZ$n{Hbbg%(~D(+;oHfe)6rq5WSB$(2e1$bC1Y3_rbO+@iR
z<;`{_Qb@?&kFK6sfYyJfqMOG@W%f0cLfV*)hs^Rm?r-cq4V!`i1Mv<iVP7^<v|eh<
zK$_WQN#1m6n{L37$0V)}?j#~vA|z$V-XHyLI)sBIDI#TcuLBHLaLuqzkqFLM%<v<y
z;4gLMVcsajmTcJ=q;JSKj(?b-<aW>3P0GDO;3@XE2R5N*bNZ)NEwf`Lsntui(oqo+
z&OAI^E^}`8YmnB_OEg$QV_^2cBPHtBpDLSmYaVe1rtD&5uAkjy-fw&*BDi`71mRlb
zF@G#1V1Z&%6AUtWT(Q)+uNC*&?k4Om=egCL_ZdB?v?80@k(}n<)}nR7Y=oNzqqUWG
zxKXb%;oQi*g`{i*#d6^gc<}c=7D?&`$;@(enn|X%MDepZLrI_7SkHzG`DenYb?oGG
z8CEeY2W4OFv?*&TVJ}D%9kQCkxZ}!3CO?F=X0xQY^&>(iALPEAU>Fyj^iSa~ULqf(
zZ{W0pPZQmVl}(yr<k+x|C0)U3lhBsFigUIE9o14L^17l@RV~0~n>?@evy;?%Rxe9s
z6mRW5rs4Vfm!p<naD*)r{vwV3n(cSu5U(4qUDA^W9;jlmVgg6#yJQA}N34=9jFVL=
z5=r9fz4I%q1Qvk`miK&3AuOi2F;r^CP)jdXT&3A#&h2FADihDB&fh)gfY3VzP6%iA
zah`mGy?6}68y<);+Oz1K3_Vh9_vHkQv*YTtAE<lA)L}3SLVdnx4_-dl<=rJ2I*f!S
z*uqj;Hdz?)`>cAO>*+IzPHuMb81X6ai_J0Z1pC!S+zH}&Z0YH_^G8~w^C53$^SIQ3
z>}`mcr(JH(>l2hs`wEm`hsRVwhg8S7#nGP!9uuAByXwcFd&S<5-0UR|qoS~$Enj34
zj=1Ve<k8u8VCbj3SYmj}P1hy!Xf1+pkN37S70or(Q%q8Ni)-yUpt7J=v}gPsb}aqv
zwBjO@IL~t$jXfzC1Q@GdM!2eM;XrSU{?yJa?Fc8kTQR9B*njIcAAYJo7Z;uy%#^fj
zpdBw*Q$n+^w-@ol8jc|y@dS14Eg=TudC^kxsd>!2r8)!#RuU!9;|S{+6*?mW3U7!n
ztWBEs=fNHO(SEEkGOqhqjxXoGpZJ~7JMi~6T9vh8qBoD1i%Gi;$~?fidCs$mMTDJ(
zd!yUCBH_fX8Ne5=S7FkXO^&BQBU>N_d)bnUNqm3DXoWA$i&Sj0s;pc5;Sh{^Um1@V
z4UP0sCFiBvsSsLl1EJqfFVcjgdog(}JIDUv_>vPu)x}FfmzpU<HhHQm9_~g*SP<+7
zN0N<t`&qBreR*+zh$N5_9HWU}=@Kdg`8C&m9<&Lfxg`(d`64bLNB1|*-lv(kdos(F
zQ@$#yIsH~SO<`7X_BmC8zqmS0Unup_&`}zcs1<t`YiM*%HRCwRyod6^WFI6Z^y=Hk
zZ>DY$w^ck}zEPZt2@J!?pIj)r>&XkNxM8EssNRe?y<HS!r*`yIur!DETV#qeKYwRj
zy3ECH68p8b$Q8F)wK=ky8effOu>dBaz9{cPtbjn1#>MAepDSKY+G&J0W52Td;86>A
zOKCn}#`h0BBeXh_*y>{7d~O9c@Z)oZ{Y5TQEf%io5ZN~eLF%kOvaganA79Qgt1P^o
zTy0=?cPwfDTo;RoE|=Vze>G%yY(6)XT`qBx4yP$6aeR@K(<D13O(ibc+xR6_HHe^(
z$?`}FDx%ar{J}%IDVsgNE|3QoRf{yFx04nm1LN|Fk`;-c;vPWe0&#XLuj+0TKejNg
z@*Uk_?<UAOlDex9D<*Bm5<@3#d2h3w{K=0;jz&jMSGVC7am`9I$0ggz`Kvk4uN)%(
z7IOD9i3<lm6q#Wy?$OP|*Tba}AZ#NOoN<6olqC&CBs|_Vpr<!HmT*e_d=lxD82%o#
zD~>OgcC6iH4t^Thv7eU9y*F%R#eu4y8x!NA!iooVG1Bqo)VU0cpTlC;E0=!(Kssl%
z#8`lBanuKlH|!lM0c!k+40St$9iQP!lfiTc>98L;Y2z({g39c@b(}{X%-A;&4%)Fe
zmK_r-7>^IXsX@r2xZ}-=gR?9Gk8YJ;;@vHQ%h?SB>Jf*nuW@g>&HD+;iFh@eKlJ@-
zJN*$_hY3NpjeMI+db<?Ku6Uf=jKtR>xIgn*L}cLIsp7*HJ7%l|m6~OridHn`F3Yi%
z#nzy(`bdEgm_)l?!9({-%^vx_T!pQys7s48q7$ufEjw}fW%=tkOzHB;=Cz|n{(9Yw
zrrN<IsLrmiZ#Kwuz(9Y%NR~D;N7<;BQBUPT>nF0lwMEy|u?TEBS0dchRLr;<KxOur
zoFH&`RIFh4oTc-AFSP+9l%-NH>(fG{gVEjt4&`?#Y{#0D(z;npx)|brw^RQH&*%mc
zb3euzfsp3l`5U<?l3W#`TacMng=7@fM6QZydK<o#t)lTaIMhl1)S}3u`$jeg$2Ui*
zPNveGZ!9P0;O_=MIBdhBy){0CRR0i~X|v*w)~UjKG9tyCXU)G|IoHMg)h+MrH{g%$
zMTT4|?DOEZj1(K4yY*RBD6x+eUz{gaZqRC8g&DC4SbhKXW`wjm$5Q5aR<WOK;J#-S
zMz=`{?Z?SVPlpEtm9N(gh=~=98uu?Pd>QsrPPC{G18jiT><!Z%<<&H)rl?3*f}%T_
z62G9!7<oye+PVo<#FF58D&M#>@nfl3Z@PJaIm7=wDV<JF-9gjbn;P<RxvW70{yq(!
zSmmPlHeI(tnWAQlDBJQAN83!jjBC};6C^<aIi;7L4|(6OonA2nVgFQZ`gkU=y3KiJ
zE!U(=%8eX&*SM&RNLPlf4js4?8)r|Ow5^%Ed6$#sPQ<2J+3NUmvgJ!CxXsgF2Hr~j
zCMi{BIr~NKiE^;ypY>V$1>rSJj}-Pzi6ePcw#SpIUFnk-xM-b!+kTjno(-n4%#o;4
z`Z%IoO<rsLpF07|*k}9-M$e8vTN)nJ`86-wAV+NYalLF`|8W4rlG4OU(v`mN*vK7^
zq@sfyHApW;em)T%;Ylw17V}nX#Zucao^%;k=e(`WT`%Qbwnpx8(?yw`He1WXr+6tr
zHA;CS5yvM^F%8e?`q8~S#QqNbJOP&chhl{q0{gf+r=c1ckr0tnm&66G%dr;YPKD}_
zpOZfobE$S$dVJg*B+yBXP}Q|#|1K^5h65#Lp_9WagAL+Q-EFK%(Tz+LNp|XxP!0vf
zvrxcLC}TQxw=)AKyq}CtY`4%dpZ<Fe9T1_6(x%Bak>g2Z1JltTpS2IR%>lMG)xfA4
z|5tBOqA(4z5a2<p!~d0Hr1+#Zc(_=rwlG^xJdu=u=&SohpH9SiiX)g&j6_~#k_?!=
zsakHWfSq{A>;#bqVaNz?m7Jw$90VIBi6+O*w_((BF10d#)wn)#Mmi&+;ijSox)Gu!
zsg;=&kdtsuFpf*O$6N=|4@9u;)pzb2sTGiQU15<z24X6LuxJ|+Fa~Xf{5dVqrAU4T
z#scHW_GYq1;t?`i@g3tgFWxW2Js_zXAU4`$Pkuld39}~wv~Tq&OzdYpZ~ZZd<mdhp
zzF-|DM~FnbFh)Jg4T=i-!6d9~O@BKpuQ5{WC@rkbHFLPIInvbpS<T<if?t|MPAisr
zwUwRLRm*3?l6tWCx$7J>G}puRlkLR!N$qVa#snQt*CA(MgKDTJ7=+=<rP)b@`brs0
z`OY<Mk5j0n(y(o!5?qNp@Npih#FZ)j6a~kDe@&>nBN!(AI7gqcj*)-_8Zhp*Y0e$R
zt21L;Yl?MWuCSrAW1sYBULev1?9%Eevg}`K-V22uE?iVQLVR_O_&jiDJZ3(x$T>@;
zV^a}bCOOszPr<h9%t9+YfkHtx^)zlh+fKPI4ktLtK9A<9xluHNXIpOQYiW~^9l>=v
z<D$XoO<jg5!RF-R8E5^@(xEPz^jscd2dK^9Vx;<&4ctYR>2muU6@&d<Ia_Dr&T@KL
zO*IA<Zkvk92|ub`g)hvzdfY@a9(<10HhM7jw6QGOw^~<SP{s4twZc{Zr|L_U6fCSt
zv)OgL6B;T9RWA{)b92r4!wSwTZrp_tb|slg63PJ~7ahS|+N|HvO-9qB%k{dgzCLxL
zRJMpGa2CCx!0m*Fm4|32Hm9B*q&@-RE$R{Q@2C#}-QwRQEx=f;Hs_jJk{>ee@19-w
zqf}*&J_a1~-fx@yNtZ=%Ghy5Cw*X+V=z!A3W|ytq3g#fY*?IXBY8^hSE+PYJS;JVb
zR?-+>j^`iM1%!N;MFjQ~KoqA(*Q7sN-iFS(ECWU&HN!t1kl->}6}38%)?#rqQt!!Q
zaZmpTUC;3CfsI5pUC4(w%2zJMtMQtgQ~EmG5(&3_5(#t(-Q_maKBl4Y)?)lK8pof9
ziPRch`_(2CZ{FMf)D5-pWKQC!!ys<;M}jYIguBjc{oh3|t8+#WKtnynWaO6bN1Uu~
zz%<oun?I#n6GtK^=_4EXP^8N+T6N>!`i4BXB42*+M!nU~%$FVpZoHW`Gi=8xx~69&
zKXN^y#^~iSq2@_-H1g<(R2VD>z{$B{ZQg0z<ZBR&p}<n;JjfjYb1yt?C|J}BU2L&d
z)P6>ceMpM(X<OT+QsKg(Kbh-K|NPcsRo6Ou=1Lq@Ngxo;pf1I+kI>TuC6s7$bms~l
zwd_N1)l%*WPsZUCDGoE{`mu8VHtn`fE>b>N_S3M#N(kKQCiw-u%?@*yb1YYP0yWMc
z`CJFPSeN*W?fCMX+y3q`{`C5oapevPzSy}gmvJt|hcHB^6yzrI>_voP?PuD}9j;>n
zBmkvRdASi{c98P`2j)805!}k2>ni%(b+x<2U2xsRme|YH<1=btGbCm&F3BN{$*-qI
z+VM?YP&35oba`^@JK*JrCJG37Xl>j$X-qN;0Q;yDa@qWp!mi_7oc*hA<I#Ai@5j#K
z<O>nf-IsIU+33bUjGF>-H?_nM0F6Me9Nk`JZVY020XUcXFuPd0n%+lW?^XT$Jqw}L
z8c*_XRc6CBTI8C-yi;0<!zJ7C3O4N7<`OA-dhi2=An9@ZMfw+Npjj_|N@O<{CcT=9
zh&<{W$##{cQu#Pc$G6yANl#$w&q=+ZG8~~wH=4lRSLez_=~Zx(343-MQkN_?065OO
zuB>k3yw^W~`8-D=Atx}^;n4Pa#WhOCjbg^KyCJaEpRcPr?*sqY+U0NA*o1LCsN%jv
zJot5kM(NmTf(y|O`E-U@RV)1%w@z?`YR*^n&4_4bR%NEE!Y=)!bkH~g?<sDwk3r=Z
z<6y4q+rL~&taa^MG00rV&r@Ne2~@M=&YRMf*YrbnBIO#Aa62gxlN!t8kTbl+egx_T
zP5$_7wQG0eNPb3!^Jg+d1)FtCMia`1_sZx-Xh>$<vqz63{=J8B@lY_Iy?t*h%_XKN
zQbyk{PQVkgJGg(HPk>8@f?2{|!BcB25#?h}*)YS|W&?U>m#71^27^+N#WF3An<5K=
z#El9Xmt}n4D>5Gn32(bzRB}hY(1QPB#l#(UNAeT6vgstV1WDgpofp(ju<RnF8+lhS
zyZG3d?#sjgzRIcfTXkywKBhgq&|@TQdcKW6)X6ArimZ1B`!(72h?m4wmt>I!);*(d
zQ)IL1h6>A(?zJuiH}qB93eOueh!Uq?nu$4y6RKCS&OvB{sR_vrp9;?T3=c$*ohn#-
zHXusGxLqMMq_<Oy7*GQ{tb}W&z`XsYZ8GBWaRbK42rwL3ZRW9(6j$r~HtdH<Ht;FE
znV3g%g5jgO<@aA76S|q$@nHuoXpQFXx*{Rn-Wc0Jc7cMv1ZJfL*O1f-V*EvHMpRPw
zSOaY|+c$Q9i_m`>3mr*)bVev7$@rVb+q_?paiwqk!bh0PNFtFbFKDdS*!}Y9RQWP+
zw3U+f1je2(2x00rHB78pxvFsYS@U@SI=cEZ9=GzOvIVGx)36gl(3T9^0{Rk=4{0_a
za`Q}YkzqdhXR)@ZsuS1AR}Eo%6hbQAqy!0zva&kRJm@I2Y|=03XV3m|fHds2Uxd1%
zWhK*`P-$N+k-3Qm-|1r4%AY*FwtOdM;zvYgOk=O_<nOe9Uy?}}@q<N%%`EiC4<(Vh
zLcUkzge{|V;s+5M;T1B8L@QIaO=4K#=N$hWdiP|}cg_H5LXNBO=>)*L`&>4`^SE5E
z<+NJ!jnTtbUTrC^Id2$`U}bs?+@-@3F|+ac;U%12voF83>D+wz$KzTOYDmxgaJMpI
zfN4BLYbC^XJZZXY;^O_dbGMi^>6^wuWh&*0=EEfS5oT-SYX(?ZNIUzLs}5!-Ht(XP
zY<&>mO#K1A0|^On$HiWbVoWnN%MO>c5`>=)gNkz+(rr5_<|3aeq&4}lTncBTOgL+J
z%^)PaPQZC}4;Y%pb%lnb!C5x^5>NWed_t_)oK5+#?TdoLLnoRDHOEfJnlHRg=g@w~
zWV_s2`zjEkOH0MX_<1u+_@YZk(F=weMpZGB_2{AC#9)v}FtbFLVKeHhN3(1AqU)ic
zpb)+0_8GlJjPK8`!<+~E<@t=QKW@A3iLNVdY<kw$=hq{4A8loXvj8NS7PKD0i7vyB
z%6JvNv8A@??@dUEDk+YQ?Gqh%XAYCQ<YDWuH$!|6etqOaCOQ<S6232kxS{W2`wGJr
z+a$60!{VkJbp6}P9ufWZS11hh=zqUNrZ76ayo9HLg-%tb+Gk1w`v!IHw=2TO&w|wm
zY2>i1DXtjU<I5A{$@W&ouvJ~y7C(;aAClOi8<z3LNIl0@r8%?LVw}md&+TDk%gyBd
zqKZE{#OcDPG8jvwerF(NQ9XFzV=sfwLQA$DizG!THv^puy5EbgHm;RIcWoj0<^=O2
zfhbIWrC<dOwnxTI11B>O^%p89Y3WI&1SNXfea11_POm->Ie6Su7K;T+3Sw3&t=OX4
zTRewj?c61Nxp^sX{h8=t#b#y!J<3B-i9fb<rkOg0`<{0#gLuWcB@43lhZE$DB?;wX
z_SZKU_V%TYcB^-254>&g(m#9A9*YT#<|%8%mKEh_y*wMYHlwOi*m+gur@ZsJ5l)aE
zS2Hp~R^fp|bH_-&ckbt=d+BQ7Gvj=b;pQH`I^VjijIqgsRE>$Q<K65+0>>jxW!J(Z
z`I`%Kmd_#O%G%^yy2b_fqEh5T4^f(1eW3wwM?9l0T%iIV9^OW*Q6uj4SMIAu7|&m-
z%}T)>Ag(*otj3-@QA!1cjUR>z_8Y5q9k?=deFu#j2G8L5lDA{DF0klHr)!ZEBbIqn
zuW{oy<Xxkr$EbpFN)N*7`({kI5S+KELZf%M)(AFUA{YgZWK&e^anlt9$yE~Kuo`=0
zFcBP5-B<6KGn}tUQJHX1h;C$5EsNNZ>*zl#;{W=OmP~qYtna|QY62(NUyTX<i=hKk
zCX}acTPKg~|6EbmlH&f&o(x|d0{&T||L>RfdFj^@=?Tw4+F-kouK*6pplEYga@*I;
z0PI2~lD2HW|NhZlOMzGabZ5)y-wXKv`f>mD)`V-jz}^H0>20qLh-MhUKn>w#Mr?-t
z-(Pk-+#B1=k*Ns&?|t<@edwT;%l&4$IWPspecaC%^aTLi#^ZD_tYpM3vnhya{(oCc
z2<%$5t`}_o+e7}J9@vI3c##ttebN8v`~GilAnu4Rb^bLXOZD&Afx<BCjIIqH%mLs(
z1_G#za6d4K@$gEg{VO_<?_whzeeqlGJ}%}i${M~jehrx7T<HJ%Gr9-n*3YSY<FeUt
z{`DNYnSe_g*<VD%9q>kalUmLL858>qVJ#HKsYKczHy;iyqT;d>I?a2WF{EmKP{JM$
zZ2_PjQY&DlC6m3}6-mI?&;4xq?8Qw57$^$51r7{IycT)y^p4vc{AN?Q!^5iunXudJ
z{a6nqU-mYp@wz(H>rdoCi`WeMSFrql|EMSo+;AeoMeB{pXgk(_eM5TWZTBLe<l%ME
zx>*QZ9kdI}tRSIvI^dLS2dcN3Ls3VabMTA;y6C(_iUEGikwsKvZ2N%%wlaVbS-Twd
zu(Tunbs&m~+#1K<g2FA(13%s-3)t!lT&%>lgLH?sd;67&=#xT3Vaf#e0RNc4BIxj3
zTX)sq#UVD-bRImYnmki^`7Y-~J?&%!H*g48kqdwUs5az(YQ6@C8y;7~fn<x@GUzx>
zD#X;u{;xaPn~+=z`Rhm>oS6KxiP9Ktfbo66Gkn1gd9UTp@a+vKDg3R*hd3L6ZWGun
z%}GJnj38~*k(5oF<AJAV3@1IYh7Z^?ylyTI0O>EF+@LEEX)t<ql+c_ik~Hy31_Sk3
zqZcsh^TLgVDA{p&t;99)0~1%g(7Gh>NA18tofzDjeUCU6CfUmQ)b!|)3D<r`wu8)e
zdEIsn88rh!^cp^<th;)HERR!NmO>T*3{^JK6S3(guX^^st#ss8@&m0pFF))iT8@AH
zpqa1GwE+=PT&@YUIaomYow2b2`oj|HKY-m?J0!(obVT&J)&xj*aFJer=SY@Z<O+*I
zvZF5Ye*}TwRK>r!N7W!_ouNI?OAENi>j6<b-5dXAq$FeE55bZ{0WbJE_#SZ_*5K>x
zT<2W7{W*Dwl|Fwl=$z`$*``Bd@(Ayr$Nt}=(0{&EEFFl=5ZKzs?0=5=9^X3*SETB?
zJ`yZfmf#aEB7cqv|7M{wl11wZ{>ztVQH!I^flP~~37!0ZeS}{M(~c@UsL#s9q|P6g
zJf?Yo`hUKk?m3fqt`ET9u~~rmd*n_LeXs#YuLMZ$TRU$b``FlZo#_h03)y-~*TO<6
zHKM4JFP}2^{uS?uZpJbHi3ImvqH`->A4YmnN7W;XZ2(Bips-=TT?U{mTbGN~g!T-;
zH?;w+x&Q#7eZt-dAHC>nLlQdw;v}*aq*iRc|Apq_&q((#25o}&M>cBD^TmK3N&gh*
zIal1FJilDOns31H&v@}Cna;gifH7+Wri+{Yoj{xpLk6zR(RZ-z4qk=xO8`OKTfG1w
zf-XPV>=D;3prn$<zqjkDzVHJ6ry?ScvV4(VLw`kYWY~+kLE10mC+dS{C;Ve8?Dk^F
z23TL!7D$fL?}L?E>1DUa0=o!F2t{ZBYSy2eSw+5q*XOlGtC>5*5JekCHU_>{J>Y_B
ze-HV$Zs)x^*IRFIEW)aQ0dPRj>!!s(D%4%#EIdnalXN3`-99Wi7vm1->6WGo-`+)l
zX9J6v1q#;X3r6@FwS7Mk={pmKr^@r7h@4Y_!LtbR*e@I%Pr2@wyu-hj0c?#MgZ6L6
zGHzD%+)Rfy*zc{xh91by)*lZFw=YM+_Pl^AZ~iuxLw|mX)z;#7Q!b=74s8e?uw2Ck
zSBkNqQt3L%aNHI?vIQ-%k7)}u5xm2JhkJEkPB#EhBEOVnOnxc-=y7XUzNuz?3mlK}
zkg4z_nui{+`E(d6L{@Cb*mf;2XZ3W(B&Q5SWMjJNg#AYNOJ3FQVlI1tZ8kGpllud>
zZw9D%Y`t=~#I}qVOYlFt24_i>C#mP@^YeD5h480y8`r6}n6%uWUE?%aKRzZ4`fhFO
zo0<O(*xU$0-Qt$tX+O-G*S6a+eu1^Xas?20xF777m3v#sUF$aUamN(-lvonBDH*tn
zhy=iq+ztx22Z_&+cu6gTR_ZGS9fYbnu)A&J>7pJxx%P8Jb9$Kj$K9sv=;`ka&F@+#
zsyETum$x8x`2!*L+`K{K0{82^FZE)9UrWB^h~<`Ui67qP`B>uDU$IyAt14vY%FeD_
zPY*ffyBmFtUj~G)vy)*8qr}!-fXl<$J_AN}t0@F!yUvVLV;W~|oxOGbQrHf@ARMJ-
z=9>R|*Px<;=%M57{@K6h5R8b@f<DWFL*apaA4d)RpqVfz9YnnYu&aLKNL_>`p<ivt
z!io$X1`CorpP2{)yfhlxu6!&6<KAx|Ka>&@U9d-bp8+V=n76?+9pIjp#9{yo^fxSs
za`l#$v{qhbpvwl!9i{F4pV%m&a{nkSAHRcMWZkPGLSM-p>;)f@Z3n)ki#FX1K>}=S
znm6#YZDrAuaT|cDbem>q5Vvy!pST^Dt6?GVmJ2XC7m<izzCi(lx^ELIq!!3YZ}C}Q
zb&MW@QO3pmNZ8FesE!E)qCYp-?XHWR{jj;YSSLs{{PhxFL<UJvCWy+SFX2VW-qUg0
zeKItQL?W}i^jxNzw_&I6lMrZMerSoo`+!)l+XUF9tlkO37uv`?0KSNAk&OVb>cig|
z0YqMup5rg^?@3?oRH~eV^M;Ck)cq(n;~w2bw`_yfvBeD3kXH^C)etj2-Mqz$wM!62
zKRp8Z5)0C{nf9}o8)Tm?H_tN&6s9k{Hk7tfom#Ctw-8Cj>8AeAjD0=8amOM+mEli2
zZxAVpDug9(NgXo{#OkxPoVR&{h|D<#2yZ#oGdWLkX$B=fzNtOXsSmKMQ1<w>eaWKt
zO4>R{#f2qvo=n?ivFxb#vG7QuP(W^<KATYN*_{jKFuUg5NwMdiMH5)L*GL;b5pa8^
zEi%K0NVoy6Z0th*)yF;r?)2imFmN7LCh&Xg2OP-KQ9iXw-FAnQ7~n$hMX79kavnuG
zwz&|<|FgKjS%S>C1Qr7?$WjPpQ_4(Sr#~cI1PXM%oRUeL=Xs;wG!rt(v!-fMd%4Qx
zwP~-j#%y8N{`Fi{f#VlXZa)XT*kM;^13+Du?>6>7Qvu7WdeRn!$~N&bl3&l$4NNdX
zKS^t*gHS`+eDcAB@!+ia_K&4#f5sI046ghs2+fCrPgA}o>Nq|IV1g9u&(F`o>0oo~
zrJ%4En<ngXr<hI6)kd!~eY<fxf4%42L=pEx2&%gyl=fV!u6b<T(T?UckXS`zgbQh~
z+YamrCr+U|r&D|^%BdSG3>NRFM5m+hvZ~YYZYP^>zPoL)yAeu}TgXm@fAT*=n(1yj
zEoC25KP$+Y)s%;A=KyD|xcw|otJ>|a+LssL2-$%SIGy+nn#c-tDTjHs%6y8@Pwhdp
z0{8eQH3ZdeGN&SEd}|%hh*+En_o>3Rcg!h6muzZ%twJNf_H}PNvlvB}>Zj2UpbV~S
zGJUdEQYltJE@AS4NdRiqxj3aV-4agwGhK_h81@1`l{eE~W8UERpom$tG$~u1beF$4
z8aUx8Uu=Z_bHHZ>-075d8e?wl?(n__o<+}t?!rJW=*d!4mP2cIConG?!v!+qf@?_l
zzJ@zWi8%o2<ELD$cSnnP=RAn@1+!uiLzxG_yW^P`mj}AVwwuVNGb|t63mYc%P!?SM
z46(M9Hg&QuR%U0o(%bN1MUlWDFi=coW$Jhm{FwZJ=bmb3gD(0V1*a|jOym1q_+E1U
zgq-xK$Q*4CZx0ACwRN2C@Ju7c#_r{))>cu^JptbDBaL`<&b^*Xa{oti$u8G@A9)Mk
z;bRBF(PqwWF8ApUTCjX@pW&}iAGKE|iSR5of~Vy!ONq-!Z7}LTf-`!rSwp1fF0)4W
zckJuH8?=yzTxxw1AET`BMGnMu4t%K5H6ozmZkQ?}4!k4zB0KCVs&8)@8;e$TQSYLJ
zhwK0@C)Gg-cF8yvA{663c0Bos34u!~5K&I6zBIdB(TIxX#vX0nuvQEkvER6hm;$xN
z;)VaYyl^kB49X0atE`FeZ#0U^INRz$gtubfO2$q2NK8{ZDn_m;LeL|k!GJpr<f>RC
zS$}QjpnGb+W*5S3Q$1yqHpnzBzsJSqT__{S{phbH*>efTnh*?)r!SAd0XV!1_qw;)
zuuZ=)2H`F-#t`lD#SgS{U}8q_`g+urYPQiKNUcjYy1<2$a4~Shnj7L#f2kuvXN`5W
z=Fte+*SWHrul}9MlP_Uk0n1h#<c%Lwf<>7wPx{%%`<+x_tFhdJfmIV`VeV}>Ea#9G
z5TB~YXT6?V=Xz9_e4_Ca<g3Of-6U`4jc5GSzV)?q&#?PVx>#at*<xhCdr}St1kav+
zSdURF()jgZPlU)q62<;{zinbAcZM$U8x|?CHJg43SI*;lMux~Lnl-MlXR^sJbS!#1
z#9Ev!sHD_3)TS)Xfw_4(OQii~l4EZ4DCL(77(l1G<6|twyBgL=!rBM@8}Zt7xv0#N
zO2=e3ou?QRhfbDE%Vrx4guM^2h@@4yP}8a&!}+KrM=~}(I&ebyu*au(g`$h<nq!UO
zc6c3vP=~S3veF|@^M_NDWgnQc1v~0MB0e-vQKhouWHDv=(1N{dLdA>yekWUM5aBCk
zk;w4VDS8bz*85Q38b-B~O0lisC(<8^j@6qxgI8>8x4sr#r<!O^A>I+y+GKlsoCC{o
zj&I6ykEa$hKy`uHLt{q)HE5fz4D4_#CT6H3tQN5-pX$F@o?8z;@2bTRHfYW@6YPGP
z5=MR|S!l5AUQxj7Y2a~ZR%{zhTJ^0*Zq>*o^7bkAdSiKm9J*idopwV0)5JBobs2s~
zpxML?IbT^|+eciX9iXM(5I*FMlgMZ%K}oTA{_oiBqv!37hneF_wJlHp?PlTpz<*!8
z>kPgusfisdBeub2dSvCdWKtiJz56NBsFpG+f@nB9dLvsYCYypf8}AcVYU#Etjrz>q
zPm_9r918Yqifc^60lCMAm-DxtRe!=q+DHimx(JwFNu`uY%?i?4)~N93{PZ?&`3cb*
z!FC=AiyjGPKpQ+;4kD6i@X*+pHet}ri5|@++N6xk4qFjUnH1{PW1cX<r5+*W$|ky^
z99j;g(9c%-ltUCituWK96oL6Z*7F6O%!UFC7Z$EgY3cjhp(MgCiuFU5(r{C^+nbi%
zWb={`qdOr+?EUGK2aV5fi*l%;n~~LbNCy1$`Jb(MH@t-KDNAWyg;bMo?B*4VVg8<K
zCrCM$Ili-g-Q69d#5QFJ1L>0ANLyrGlSy6Xg16GJ=nzzbOF+}T@VQ4WNl!Bc>gZ4>
z%eDJbM@TvguIsEo`mJhEQ*Xiaz-0kvqb24euE&n%{*@>i*<Ri#9t{sj(LlB=OXE>x
z<}uagyN$!N9v|Nl+&<EsHFD1_Z_-f3VYb@xsc9~r_jii^3i%6XzG7)6ZG>B|TywN@
zrPP1+D=juR!3QzR(50KMRK8nI;|MdUg3c#0dy&;BIUi8M3rQ_2HLHG}$4`cL#8wlr
z%}8qT`}z+_w0wfCG4-n@RnB%+#7jRdJ&riTq_QBiw4GkRoRYGtT?{l~ZZiq>8G&NN
z<+r{cg<}5%dA87ASv$U%ioS?3?V7@VPdaz$rq~2jg(=<oDE*diNXavtqsqvdhG#wr
zD|s@}B;{>ms*L-_4~wZH3lvdG3X>OO8yAc1D>Ab?WC@@|LNe*=dbGz)iA*MC?gdw~
z&Tn!kP3nEE7^oNtmYQ=ye8B2~P06WQJsY5#+&9B^m?=rb&2gT5XZ|giroG|N?RpmT
zR|MY+wKM0dOioTI0?Ne>%i}PNush?YR!qki^@FLTQnUOE`{JQ@NW%STT}$;++5?B5
zhuQ3&c3#!9T<1rAub9-s>M-gM7W@-?%P(p5mA%7rLYAg3|HGQOz;r1OqXtz8qR0&P
zG>A0X)}e{R$SY;iu%<Lljp5WO6o1*NE|jy5mC|(ZlHB7#WBYmiSb0+Pat_g|dw<VV
zntNHNVYJcj&Tb(E<TiGxnZZ1$1!c-b-Dk_e14o7<YbsTDQB+Mfk8g@wg^{1Dqp!22
z>5m2PKrP`|j%e{4uq~{{&t$U>`{o5<(;7N__Y3CRS+;eGzw(yog-bJ{_Y&$lC;gYj
zfC(%FYBL4Y<S`v3R0HmyO`}Rupq=pQ*?IF&Es+FHdCJkKOL7}53@$Xwao;fr=H7Q#
zDzNNoAkI%DueuMW!HL(Lq^jSTI(+0Lx^ao$g?@@6ot6J(C;mt;8?$`GsV+`z&}j%^
zSg7OQT(l}j0o$uxPE9mfEcIBD)f~K=%hjFoxn8jdURt#3?T>MB1PmgcGe~f(CgHFV
zAG|q<F4<ircJC(+p~AByvo*cHfIG34RSZww39*uY`CBtjtYS!co{cIHrMrsohLuhr
zriS;+Go973y|wFC33HtSZQ1J6OW0GHXU!+ytXC^FXZM8=CchB6w%*c@XTR^AWN=Z3
zcCDQ}oY}iD5neqm!1Zz-tLDiZ0{sOmR`U;vLH;2;bTM{u0sU?Fw*OQ14L?OseD)5z
z)MReWU1DBDkw2U*u|2s<j4p`x)}XcNSFLwH?_2Ku6Uz`EkH=H=5BsO$%3}lmgiBLJ
z3pDpYQqIr6nZ&B}LOT?o{HvCC*hwU_{iDz+N)^eO^jUrY3tT3D%=8c19m0^SXy>&*
z7fM>w>q%IA8ADEWo6K(eHVFK{{?;PGn?&MRnK-;3j=4OkHv11tNiP^HY%6~meR;Dv
z+T2eSIn-FkU>cW~)Sdmoex~B`qo~!~_@nP*+8M&5``cAoe?V7Gj4W_LChb{z*1Ypr
ze9_iT-8QN)Nj1eB-4v9bubSyfuk@#k-kfg=^qzhK8XdO31-{io!kd0R{jUn|(iQw+
z$9uny>fW$is?j-aC%^enXjh_RAa%t5{zUDUC%yzhj8;-v652Z1)SEtkTY+UKT-iT{
zk;YdEQ1_~L!VH#BXB&=dBGE~+_wV8p9(8?<N8^R=YjpN%1QD~T6tlehOyZmw9kB}A
zE+iP0dRx)j$@uhQ=8mCggd-!;)1vfJV>~<Gu58YTmRwY;^Ur;T7lGcAh?!so#((wH
zbOq?W9~(>lsK(t)>DBKPcW$-BlqLvBj8NHA;d8*Ea;eumuu2aP^<u&a03e6qKc8yq
zCP}FpimMfdpUqX(RVUax>W1LjrnvJ5L@8Y`?65LKV~5IhcNE)j=@Is~FMAI^pEe(L
zp7mJQW5U)HHtJ7ws6ADApT9hs#{Cm%i!|#V?NhW$|C@)w0B5ns_%Va^%1D?+0E6M|
zFDz5P`Il?TqI9D*W`e3sII;5ND?p&c)DL<(fL`|dgK!qaqHTB4cLVL)R{7+OG;LA5
zUzxUyAzpP9iG{J=qc>v(N>U;#2^`Gv6S8AUTGaF_cNnCxRTR6U-k`3*pGlg3kF{z~
zkI{}p$XzgqpfCF4d^OL74sm5+VdqH;^HjXaN|!sE%qmrVw^=!MrjqtC<%OL;3^kIM
zZ9C_~q_bN?gxEV-55ja`Zqtn__geloA$MToaLP^KU%DYxXFA*9X5nnRaQ&S1(;LY`
z#v{3nsD?QU$Be;$`s$#|8G6V5Wn06HQIIa}0_vj6z@<CWt7?>oiCuT;`uI+7`;EoM
z1PZv&i0Fac6Y3F9z3Vc$%h9dpbhg<P=r4I(y&)+YCN=kSJ+{#d<}ic~d=C#>Kc*Xf
zx*r-aq>JA<^}Aao9Q0JeEjA3gI$LA7WO@(E^w$EuY+h`V9=}*Ve$a-fPX7NGJL`a^
z|NY;SGNgv+7}5+0sS%FuRsjJK5b2f_snIb?YJf;dBVf=7NHdVG0ZM~(3)1cV@ckao
zIrrRi@BRJlnA>OXe7&B}$Fm}TBW#0xZUnUSz(&4%DsqIMy$}{{v@7-OXLbV>Z`!47
zx~cKnliuL`Frkj0_Y2!Tn8kr5qTEvqKRiFJZTbyz9WE}tXIiJat%1SW25!(TAb*Be
z4Tg>zgBM#F*e&EK!M4!yaTG-<Y}b#;^+k$F&pdr7eJcpAmh|E`khi*#9aFZF7~wGH
zCeC=cWz-ffz<_w;oYAnq0_+{rwfWROVhzgXZGkE}TPw@B->JAs_S9xE-~E;nbx-wP
z{22Z{%Xd1DTuEPHE?=`lik`0K`aX1GSjK$#Sez=o)r57pO0Eb-?(cmTI(08#4j`f0
zoHL<c%JQx1J_61%Yqi%J9<#cQ5Ur(JH<o^TVn7HZK6`p&LoEFj<)w2!q)>Q<8Oa}1
z8g^p&u>U2iTowqp{=IsH<^NL)mInu_6H{|;$bW^7ap7K@4+tp(TooNL^l7hjPQL(k
zgF606_tw`&!y6G|r55ueuXe2KihY8#6pjT4)J=!-R2{k?uVW*wQw0L^c8nO+@aD_3
zZ^VwZ_(0IsT|192nI%tSD&cbDy$d=$)^Z~iJDl$bD%=lb=PfOAJ}h-wO}YDsGJluT
zJPx`ZDnpV&!_Ddfr#_bg3I++Jq2t~sGj_k{XMp{s-HxqqD_jXm#LRZkI(PvTAxEM^
zz;ZpLG_R`-)<SP44i#%L<^{?oH(DA!#v}1um6^b;ZLL2;*412|ZiI2qm8dt6%~3RT
z;F+boKERf5c#9f+j{W}YvbmY8E5*HeJ~1S4FDKwCL}**dV~^ax2734IOJRpIpe~h7
zTQo?$1?huN4Z8z%z!ei_pH~b;=Km~f%i!Wg#a*fVb^r|fNhr}WL2nt9q_}#<%k&C-
znogm2Jl))WyPY^3DRbz-pZWIeDrl-#Uf)DE>gePu<gC}uOPe>@AHJO-@*;-{(;HcB
z+&UHSAs#urlIh<_^;@M*rFTOW?gYnIFEh=$C5tBG<$f*@)7cGdXH?}DH9#!Y-9_pP
zP6~p|Y}Vcl8fgLPyH6`&C6J6BVeMq$ub60i?!j`IuA0n{*7{Non+#(g7@c{E`a4>T
zah(t|BFLX5ysFMmBy-@#B{!a8GTc-SPNCBnV4q!z?~tfygm7M7x~i$0P*EPBqRoIV
z8X4q-*<T;-BxSk~EzHZR`$R4HmxLIhs&D9dE_HIxur()9<S^P90#u*!&n+pks0O)y
zV0=)BfeRD!jOHL@&ad+1u<lKVkXh`5N)J~iM1#9~2u9&@+r+Clr01tCZxFe_fMs#;
z_}KmQN~-16eQm^h{Soq6`&UKR8<gu*5D8D*!Rw*X6RZCFrAVS(vOfFgc~af>vWwBK
zwy~=y)^QIRF@&g;H;=O6HoTJyY&pjCFzuSM{oNln8MeMuVrk>xa0Gg+0<|g-^`Y{k
zBeDO*(1oR?oyXgwijX?jx1LU>6anE1#<q&bxrV!qrsg?FCN3Y+MO#-Z8r$@=YiT>9
zC8w)TM&XUEaTp+!0?aK@xN#f!ESmd>m@U#e)0C`@vRro4hi?!spMoAU&Ze*nZBv7r
zUxjbEMNxWx)BUuU%s%(p_jenlQ{*0HL+_b|cA5pLl5};a5dP1iGv(@w`suwFcr8{G
z2pNOQL`={4>j$(l1-I7tcjkS&LtoIb+$6S`1oVE)dLR>aq8vf?>JiPDCrGQlS1%|Q
zc-yraRM-z>Yk}TZbgciXQ&d<jrTy#=itLhy&Q*ahobOV=Y5>NiCH>#7e=@?%w>Z_m
z!GqjnhwQ>1JWHm!J*Xl_Z%If!T{M;M@mi%z(al6vW%fz7d*9$%Hbd0T=xY*KfudlR
z*XlFD5AOT6?Zx@7>k=j?$^pKEk@t^z!e}%vQmU9n4&zWUr7cdjB7^P+E%+cpbA`|O
z);c`OQnG1z%5q|~{GIX!k2+AhcOG7IJb1xp%+_$oH}vr-`QFY7bTY?8|L}H)19PFJ
zuMqF%TqMncq=R7Pk+a7M&ooL>?fv8T6tF7AJqc)Z^DPJ5I}`JfVKRcbdX*usZLS5_
zT$AFni_1_!7RnpHs~oZ}*v=F?K$x2Mf>X*L9qWHrCgCe~F1^oK^iired^LO0p4usw
z^{VwM6s#d6Z@8exZVc>*t1#F1Ja66hf6jTyg{%DQ%G?izvplL)<9Y#9*dhQ3uBT9t
zawTE6O{6mx;+GSrJP-5)bW)+)hA;vpSdu67{vMbiT4rtmAI6O2;DQYu-$N}f=#A6G
z41QV+{BmzWebHd{^i8ZVNLa_+xce<~@Rtvo`m&QGJHUW=t}3c>1t}b4E)@IS&j~@0
z$>XkUFm|v2l!}cP#b7TKV0=~gy;L&HpQ)-C!wgTMO&kf!pJzLadntNQ$Pbj>IS$8M
zb$M%^!py#Dj{WrT-GY_4vRM^sgZhK=)anm5lpHxcIcLwN>!*}YAr5c+bZ%a!s3Y(V
z^oD9vEejmPZWd5VJlH6|e|sz>Id_-)G=}Gy^71XE!|+I+D!s)oN;^AJHO+V=H8ad~
zj`~nB*tA#8a~PplxJEb7q~18IkN)zBXXflBXm~VYFb3;8^&rEGjKYB+HPp8i^d?=R
zcI`Iol_D%wVXt)9wRjb$mSnTgA4a8M|4?Z#rix?Rlj?4RA@I%YJPmRjIiyb+#mT&8
zRjwNuqei%?CPmiW-b~y$%JL}VQ&)el6(crdx~elHy)OvQ;%$yU;@B42Gof_|e%mBo
z$ORMnLRL(#q%)~Al%N!9f5ag@%QroXZ6)Dl@yUzZ%|X6?{&tHskIyUYko3Fa)+ozU
zg~>MPWJm#PUI}Ggj)O>eRra781+)A64WKIluOiXCUnjGDLu{-w49gOvu#XU<G`KG=
ze!pow*T7Kz>|HFCS7KEUH0ZETFZvhQs{pj}*!K0kHPC)55Zt~!7D#-6B5%L!6>nB4
z$EvigeaFqz=?HIF=h}U6*(JWTB)(wPKA;*Eux=^q;Sx<*eBPkYIpOyJSR!=%K(pQA
zA*YJJUw2)#B`>l!CDwMN#DQjgH}9__s!<7`;~#D3=~20%8y#tPtfeg4UEU(@1+;Qs
z(nUQscGvG&NdHb0@yMx4_JgL=NT$Pen-sqSc72bQL+>AfMDAXZ#k3)y(lhx#CvO=L
zDL#SNoo9JDBZ>klYX_&}T)3fu{8Va_-)7y*TUhB`m{=>XZkJWC<j*YUvR)Q9ti(@G
zHS05mHbZuFKhtr;$9e69KUAj*nWC%H3mnG+xpi1a!jQdp#g;nYkH#IF<uJzITy}7F
zV_($uf+J14-KXnhb8oG^|NEqjAD^>@h=n{)UA+HQ&e7Z0QL&l3;Y%<Tw~Hx(P%JC`
zv+0vN)94(n4?63?hd6=JU*iNYyph#EL>?3SyRC1vu$_PVro7?lKiH_p+U(gCL>L*d
ziXctlcC7%-M7s`GIp{Xgp4d(nd5;Y+{9yr*#!O3O+Nt)qb1Os0?)=&Ll?3s6iF)^R
zElkJdLQmhueicTDr?78y&etNlf06kritvALT!;}rp8+~`vmLLNu{|qTv<*{#nQUm^
z9k4#r>g%{RPwBmfd!h4b-vH&kz^v!B?65L_aefM&42td+#=I)^<?CogCT;%ykg2wY
zD5`igABpI4hyJ650f}cqDiLe7*_W!K&^>uq6W%9?W$RqRy|gEh#efmv%Yx;MpBS%u
zI!ZcR9sN2vGs;yJC9UEJp6DG;tXA%%*#8Km)onPIhysnDw5%TzTPbRhl}_sjV3xD{
z281lJypO-;Kpsyx5QNhbDtAh82;OsFSTf0_AuvYJ=P)Ll4>rC+AqfbWKtkvzHL+*R
zesbM`;SWSB2(oirhxj|$o@HJ>?Rz;ashEit>*acmVp2%)jswErnzz)J8_Kt~z-mnX
zLF^VcCqRd^_U_!h@&Z{6SxuY<afaOzb|+TSr<658ruTzSF^#ygH-ohFrR9^@F9sxm
zPGF{V>=9-${=-t?DS2*Vti*0-cpqIk;r2EAwsTX1Y=1e69H~N*J;-u$;`n@djgEuv
z^o_vNcV`ur>-%Sd0}>176q`8Xhoi)EvDT##iP0POKGJ|()Q~8px~+2=cO4J&q^?oP
zzkM0ROvb|FdlSQ$W`Q*aZg2}YL%NOj5j)D#@5p4Ft>g(a(JO<}5LGsRg9Q&@uG1RI
zTlEfS#m;&j?MP(}@XzG<<_PwgTRt$OUO??P&P}AdPjIHLD6%`qGvpb0YCN*8r|m|h
z432*I>qt<kxT}$7x<y2{IyvHW!&Sgj#xXBbaAqlkn-gwmaAXb4_iuYhic*MdCMVUV
z|5yyxyNE!JBh9j#K!9n(G0!>!ksC65czORfyd>aK{cc@#%z|4YERDMT-g4RCRvv)n
z$Z3*~1-?i|gh%=a(Lje$PfhxZ*T~lW7q|+~5JLPVft8aR-sAP5b;V|%(08n@dWyE?
zZ~)zow)h|I7W^T2_03L4@e?*t2!qR`2R6(Pd;+E;v0MAY8hrQO=F7?Yn1K7t<eIKE
zx|!&Q?1VAX;1k+bPSTtLHulfc?~+SD?v9hpmBhG|;>QtR0u5**8@lpR6-QLT7BH3c
z-c3Q(_yul#?h&5xZ1spT(=(H0<8ynB--_?{YA*kw9u9_MAg}oE$`X@5_vIWCrPwB5
zI?C(z{ex8Z&#grfx6U9c&#FDvW(=p2?1ZZLU)O^VT8z*s?oCZjB*{W^3lZa{NYzsi
zP*Qrgn>XM%Bki{KXJPmhySPLV%rfJyH-dpr76A*wnz83<C73@>@%IzVXK335O*s>#
z*~2!R4!q_o!NFbqbgF-7T;TJRCgi${^rU<SXl1Cffy!i+NS^5JmmrBOncA-S4SiHl
z*6{NR99F^hmr=9&VC(*oeiC*4_F5WwEkDTquffNNH%!}}H71mPHN*?7kaMP;Qk+Pl
zZGphW=K8F}Qn*IlYQEz-%Q?Ss`z0o4-3F}y1QtYx%s8$~^EK?pdzS}s>$17o<9$fI
zCKiZjZCCDOK?X7VGMa5&#j#yvARiiqe+aZz-IYu0zwWgJ+#p{}<Ep2FUYhK-F|L*C
z%|^mKEE?nQT=z&IKT)ifA@520$U)x;e&)pN4ZC3Gnu_&MSBXQzI1Ofa37??v4iYJ-
zbFi4ohhpUKruk#L(p7}_>eYwrJk+L_R+s3nb%jN@pp)6uQb!*9?`zbzn$G6(cGv4d
z5<fq$TYm!@;hQ#oT=k^GrWMWJa?3C-QJvTyO;E-tV|1E_v<lWk>1wtUdR_+8O%|76
zg-fT)i*Y$q>t)<?o3sk**?Bgd5Z?HW+N1Vp3&A!u;pxRHs?973TE5^_*G`n>jO&>u
z<MjIN>jZD2W?vzQv&2iov|`jPzUy1N(efTR0}*#8V)ySSU6QsJg(thkP}Y6Z1+QEs
z3(hiWJ$X{Um7^;@o~QLjsk)<XYo%x?4BT^Er$f)m!e!q51TQ=BQk(4)zidFmV7=v#
zF!R@(J!r1LVL|41dR6wcq}uK{HLBgaZI2{~a<-YTRge#UbOkP`>Fp7osVD`Kahh^Q
zuh$-u3r-{_v}GaguvNBtry-~D$8o36g>6?Su)%<6Be$SfN9*s4I#daC6*(LM9WD2V
zmbe2gb=(e-0gI0+P!ah6Q(1bL^Y_9QNw86Jy}*yn0+~UxpaFt(7t4@AGEaeKo6bdX
zK5id)sinz7qY5enDL!doN|cE`+osBlNs30v=&Vku#VE%?n+zkqvt~NKYzMr{0+oIV
z#jgmZV4MCCmK_^`PZhq2><&5eWAl~>>r#+Lg(QuzjemXU$G@L%zD8C5Dlf)mFce-a
zFs=_+6nvCv6SPSQ={iKz(wwN}4mtL6_yK(gIa6wBXPcvm`*V4-TpwuME>*c$cOY}x
zJ~4j~C08s>g7<{agrqTwy#uHli*>i;(7FXO2cqxgr4Y0Vi<9&Jq_m3KEe}AC;;)_F
z&D?bP3hiQH-eUepy~edq5w*naVshT1mwaotFUR0{t4}!zTB0IN$Xmi`a3Ej%ofO<L
zmup-tZ93#tfoMIuJrHjma|dZQdiSJn19w+T(fEb`w<8Um2DNETL}wn9r$x$XBrA%J
zs3b?~zUYVhhFAR=i=Mdp^Rn>R6|$Mxtr>5(N{5+mp0_QrnSQeL)*)iS^`t3~ToZjz
zvMa@QY31=(42coRp4;2vZv}ta?buM9mlUAu;Dh8=L5Rh9hfhH|YdHbROXYa0o!D}%
zthucnFQey~q}>CKx`I_tVU+aGt28|ypRh0ti|H`+j4f#R-dhyE_{^+cfW@!G6Gbjl
z1{F~QVk7s$%oe8exfYbQU@SdZJjCUfoc?!e(5)_siiaQK4^*n3-|h`bk=D!*$S#tb
zFppA~TnYJtv%)m~@Gg_CsrFoBJ=vX1_0wqjs1v@7HC})nj!UrQs#d3gkdZxaff?yv
zR6Z~VECFtHEE%p7#CwnGv*ccJuskm~!{%lR8GwB6XdJgDFwO+1i>>A1B3`>Xowo4?
zM8}>OW2dAIlcehI4CbAAU3{veYp8E6-J+b!O*DGfKc@1)&~-v--R2)#bX}40@?m}O
z7*2X-fd?Sg@{1Lg3NQ1l&&yy+HZ93-EFG|vB!Vo(U50qnviJt|foN&mWj1eF%1K*1
zO#H0l!sC9sK9&D_Oa5cli^5n(S2fKH4x&?36N(X~)NYUtIy6`(n6xzFb90d%9l58p
z`NX))9g`2<Y*bt@cq=pL*pf@-J8p}dG#qu1=6et;QIRh@w(Qf?Gb2B#m^9Z?mk8#M
zyq71?GV^NGOyxx{ZK*gAua_kEHaPEr->TgO8kqNev@h8~exJ5VRa4JPtn8FMF<Vc>
ziYSIGwIOctt~a*b!@ou)94q`b#b5F)kTcA3!oKh{{Mp9kP++@J8yC@h^JdTOoPX=&
zuz<#Jc<li_atkJjyfY3tcpVW)VwDGVo{ExgVVlE`dB=!AUbhM-sS;#Qej%!-l4*hp
zG)WNqUC|XU-NWX!_GUub!HnKMh%$=l$3Qi)VrhC^3)}A&_d>!n5zjACgODhpk$oRp
zD3M?cksuq<OJ#<rMR2bO*j|n?>gLkgPoT2Z;(m<?c6o2wcr?(cOajQ1m)M3ZqoN@#
z;bbjrv@N9XGVaCP@#)=Jhtjn0Co{8tYB-#T-qhAb1cIzc$*OXBlII4?;O2I40d07E
zjJa9DkxBcINgb5-==*^spuHwj+Sy$!loss21{+2Cl@Hi_W82^$TKWv9YhjZxa!eZ}
zgvoo+5>dn2uT3UG?ZZl{U&0!#DpqFu^cZ>v7ag!{H7!$fM1oH}y?)qb7tqO{MsK<9
zS*Ofp#i@UDLHnmYB1#J&5;P-BGx1KpP0SC?<{l#uM=ST!#tDw8*R@wv7=JhfT2*kE
zLaFoh;^1Bobrg6}th+P21u|bxy012wy_!7K(N}XbDr>`w)8NhB*{-lH2~q^|P>g`j
zmC!SXYtS2kb*DGW;0W3ZIPl4{p6EX}Sv23yUDMO2RrXIQ_>2*Iuf-@a03oZV&-vI_
z$k-dcbdiRT6-Q76wYYy0Yp*hgCnr^!h+orwLz~&K{;_sOT4(<d5UJj<UonH58^ZCe
z;CT9afC}=$@Xd7_65JN|qiZlnFG=bHtgAywrrUvL#;l$1&82C`ZNm<1RbQGc+}s*#
z8Ex+N*`b}O-oLj8T`f(*v+59kBw~Jj=2xZmi~&UI>&Cmdh$(8=<uw=>aZ2k2Y3M(D
z_~)``BE(`6sohfFN3a|f_1-~N7|P$#@v)Z*`XD)MrMbR35nfQG_9*ZOVNS4oZ~FS4
z_MXcvl}Hbv1<lDZs$O%YYE{DJ{3_<=j@9(3-nVb~@fJwi&Bo6K`w<we(np>x;eq)5
zZt;gEhOH7`fuV${AS9880^lwM#+B;sSRT?-x8C<QM2bNU&7H|D^|uJ^xzj;4X5NL*
zfAeGH;BqRZUpr0aBE*sVrH$>K6^=)6iXXraM$%)3iZhA4l%nRT#WBEtIbv7R@O>j|
zHPyZ2Lw=xobANaSr=9L%pq)Lw1w6<B?sS1)ZrR3JEj&K|WfET6haIpr7{e>2FiqeW
zm<%|3dI8s}kRDcXZGp5I7(VsvmsmTQ0yk8z=0rk_hS!-qO{fGkl;;4gQpRNXQ^Qz?
zAqf~EOz9eI=Z(=T-p|g3#nLH!mRnk5@`Ed5cLytq2)w~tBxkfD)i;*Alzd26V~k4e
z6Wy_2YJ0Ff;-qJ=?d!0Qm2Wl7>{>!iP)#v2j5?FIJWYi$QW27`4yV<Do6Mx7us1{6
zARsiQD!p}^*-@r$No=B#(lp3nzGnvULwsA3#GEEdh^ANE5<ZWPwO*wfBQ2c<$d7iL
zy$!EK1wdC~tx49hUv6AVA}*6m2+MX;Bp^Jk{dSL2dL+a_<=}|779}QEYMx8Q0O+Bc
zp%MI##rbz1!8@+U_<wJo{@Li-*w1gqclzpp`TG|e`7z=OGXgb6t_nx&S`CM{f5|Lx
zX0jYGm8c%kdho?J{a%DXw!f))f$s#{S$Pi~{*ZaS9&40pS>5i8#d(n7z34h*C^G)g
zuU9<$i{iAFu4wme$jmz@6Xg|2yBMV)2Xc<O`ZdpjgO=!JO-a3denu)<EFAiaFL~jO
zQtFr*jd%jvhUADb7r#KmhSHwRx?PS^2Dy^Dqg!M@aUc^W^LF5$9hf;3r=4hWeBGi-
zz1)fc<0D|?$`forTK15X`8z+Sb#n@-3v`%f%pARu1gEixrwDAVcmBd0D|<Iob`qD~
z-s6qNWs086L^sKi`530=6>h95cO(<~j7eG`D-?`JQ=U1*^`?sJuFPW3b=G1b{&zm$
zQRohP;66?q(U8UyrqFS6KjBXS#01Q2<`2hG7UjX^(+|}1f*$Kso4j2<*+Qwp4z$>S
z`wk13fOi7#n%M-o10L+Q0v`$G8mgh*^xjfLBaVLP2E2S|o9XEgBX^rx=^(k(d$Y1=
zSN_!H=<hn+fi25`>gte?X9AJ%XaZ_0fcuUPTFlfbJH`XUJl-<?q4|z(%4w$Itw|R_
zQWsZpu63tyqzlv}Wh%9RocHJ}Vbo*;G)W>b-XGfF*@&!j&Lt#Z-BoHZ-ISe;V+?Oa
z8k*OIPJ`s?Y}ldS;P(NJt4<LG+rg@x>oo^jL)F=QLpkufSo!A<PZVOaG<JtRHh$Q+
z00gP>9Oi+-T+7TuWIbz8lpLv#1QF1VTst+A+E~w(s2e@y=oa3-K7re&(dkp!@kC2^
zGQThTrT$baw7(7-wc5cA+HJ%ntNfn#PU&YUS#_G=k0jt$Fg-3e{9W`a`{h5YGTeCF
zH_LY%GT!<a^*#$N6WRkWi$>+r9J$IL2bT_SU<R*Lm?&ukv%riAjAlV|Z(=EP`msVa
zN_wcuxgqeYK>L0#EMDfdxhd7|E%VbkCN%LFs3nvPaZyHt7S$3DRA`;QyZ5-1?*?29
zkOGOb#iOs;>fAZ!j|m?8`s^F#9>mh2jeY~Qqik3j_$kU+Gw~Bqq0X=htI4Lz76C1l
z10misBPK5a#@7BYwO3!rmy~p~>zHOr1A&7xQ>Il(Qx^%V{HDf!#S?}0a%+<`84B$I
zaKZ%fOZYy^BK>Ac5vp$-@-^2Mb~g>ojATzU`Y4nvn3BXe>zVj1Lh)!{E{pSi!LYeo
z5{&Y)78z#BmCaD9TXkcdX4i)uaBfW8LazOF^_sn&bS`>&&-c2w<Q6aA*vByflJUeo
z+}?}5AU~6$pd*9>tJ8y!eQWIXw?Z+#xk-LKg2NLyX-XZ@v`GU>%1wQ=Zxhl>BbVR(
zylRA?8Na*KP^9rARP1>_`Iq9p_6xfFs?_RPwt11jLtRUMk;wkiy2SRN;jd-mdU_WW
z?ZyPMU6Yhgn}=4(f7kPJ&@SpWaxb3rGkQbb2Bfs0>z_v7DhoCXUTEm3h$N5<D~chT
zUnVYHy}|xvkHM^-EKZou79UabChpqak?USvu}h^FTb^8J6ojz5r)%u9_FAQs3Pl0c
zr;ak*+n-7`S3+uod^*PhkDjE!mE`&OmxF&i*DUnP7*3Rw?@(sD^0@YT>Ae5Dwr`RQ
zYtD-fbYsV$_%w{VX2Rl9LXhsB<u=Wqxl83it<WU*ZHw)-4v#4Rhs{h&@Lwzx+;VWx
zPYeE_b3vq~oJI-<Q5#47GqqxY{h&JsD!x{CjICcsH+_N!^4+DTVu1uV;|3EX7_7%j
z@2sxy&gA%5?Y(VGNOC_Ztf0$FWIN%c=mHri9WRK1xIH7<;24h`bGe62Op?xQ4ij7*
zuk!AT+EAMi+J;t<HSTMY3RgO@?G&B5BcyfCbQ4|jd;+K~vkkSb!51hN7&2r<!UA}<
z#Eq>v(?4Xc(yZU~NtxR7S-%zkLccEW@=FO-ws92<s{5W|7x#^-Zoa!J4R}aB=#qPq
zW-PLiWm;@jx^m6&0I}3hYp^FqxS;brdHh2x+b6G+PZhC;PZ>{qD&#ZLd%z*gy44A_
zlZb7h(t?;mF4vm}t&9$ps|`A@l_&VypF&8+TpuZ9@ziQ`{OO-Wu@W$00WJ4(V<3qJ
z{oWWxbo+6;M|~@ebZIh)rnL+BvXg>UI^4_iAegD0J)?1l7dJcfg<ruC$s9#IZDgcp
zmTB-!OmdXDJ@-}9k_wr{A>E5#E+9(nkK{FHQ_(=r@4O*c3p@8<>Cy&ewC&tCX7yCJ
z(7y9BOufU{4SYHFs2c-M`?3>`xNX5odAVvlMr3ec4r23!e5GRbZBj{|%}ujeIp^PQ
zH5bm{2J&l}KQn5w?;-=~0;yd<eD*sVPsRq@u8AQ=%V;X6t?mwOhX!;g$+~*X@P<eQ
z<SG1iu%=KXl8-TiQPVa(Rplc{eX{mhT-QU?i}_p@{?4#M^h;tiaYKmu77D118{V0!
zy@Pk~;Hco-b(c5>y|PN8SjBdF+Z|`d?y4h|&Hj0W9D8T%UL*kviAXg|74>wIWN3#b
z!ZG(F)BZjS*qJFZXJA*$+IK(4u9M=#oSWcD6cJQ6yQwVKL|W;$+$ReO^yj#xo_If9
z=JRioN)NYCN6INhh$V4fvpOlDX>e{B9M)k$75r9}wt3i^kKw)cX@igZ<Umh`Y-3r)
z+-A;nUAW83zI2Zn{p6{=0qJdZfmLgC$14K5U&+yi#ha3^RhCcG#ue8NTt3CPRC@E8
zGqg8>kGo4b?gMT<lW6h|vZ8w4AZBv5vBeDMKIu_8N^fK3YS|Ig$hhrnXTiJKTrjRS
z`o<r9!rzNJJO^BA7*=!n%1V!3b4kk-7(UPjP%^G?s&b-D<VP$BCzl$O`0O82$Yzji
zbG@{pL1|=_33YVV6TQ0zy$8tXIHl_Y))W)-?~UyUzHH9LG){&l*pNKJLRNAexNWA)
zBlWP5)LfoxN(NI-blZ!O4)MRd#lIj7%`KvAvTx=(*heh$w^GnjvQNyrS2{miD}R%@
zI1XvgMegoM?>*6b<<x`nlCY&suP3cKrtYh|f8NS?IjD1gxlj7GEOwZqHf}$95zIqs
zi(MNspE4(zf43qoYMK#R?-*4$(zS$N%zwi!&mJQ6qWM?$WovzjcT@Laq)!HqkY#0`
z<4J4uY$fg{l@hlt-NxaAZ$iCXT0$;1m5SR#%G%0ROh1wLwomYjE7aU5(=9DBaXW@7
zmf4jKBdb!_Z=dkeJD#c5zK*u`fGNSFmP5FQn!M9#X*jF1gT;NT;$S(HJrnB)See2e
zbX9M8ORExpA{go9aH0n<dsI0y%i|^6V<&3Muqk1fG5yh1o6#*PH_0Oz8564%@M`(}
zhsw@<!x>JKe)0Uo5p1#6>9T%zrmaP}YvGc$w5zm3A#Fp?aiHI`Lu-bS97UF2M=$lo
z1?TgVCJ}l`Z$?#GE;Cx&u^;vH9Q<RD^qZW-uwKQzWNK`*;@_h<>iiasvYyROC$#7x
zAkS}p1K1mVm7q~d``rxdtKQP>S8Xy9`3-Zec3n^{rC}W)V^?VJomLS(E#{tfJVR=V
z?cZu22h=H^*tL9fz)zxGNhXI4|Dm=bVuEtnrma%V-H{;Tv8ZnP9VMdgNmkp6_|P`(
zK3TNmb7@jt8+x|Ep)AFCGQ0eFqzu06bNP5}+i@fK=(@Sf<fq4CZ}<liq>}jY%;sM9
z4VzWkXkRw2_<naV=m%m8d>4lYx937|j2^K5Ih*`rZ&6@o!pw$$P`{-F3|aYr5I=|@
z;85V5WB!M_O$HQ&LvRQ<d~Igl!Xo<>F=n+P*AGHI_}#b!4rIH#UzPUesr$S~12&Tj
zB5X`QynQwtlNH}kMvY-Sst~cx2$=`cFjf>KLD#rDD~_ill8JcP7iE6_0M<+r;{<so
z3}A6~Z&Y!bBVWi<O-N;UV8lJ*ROxI8JQEybmmH@OWSKd0@oRmT$cr9{ibajj=Z1+p
z%Hk=&N<9xGH_zV7K*K$N`l(GeOD<A75-k!IIU;+UOM!(jWA^Xr3};cuKF+0=dDNM(
zxN3|pSgoFPtaBJfs%TgAAq?ylnF4C;jaKW^UA5HrBarnGkW0Z->bfzq6CykT1(Uth
z;jY}&;y{$}%G;OiZli{y#4t7L*K}D?`?#Ih<%su-AD}!I`5XM}k{ZW?PAlqp@PT)6
zUOZGFH?XSsooD;^vs4GRtEAR>mHAN8u8U@oJj*CX5EU};Q4=6&^8TI6J3)E*q3Yx2
zUfmtMZ$)Qk*&aw?+2^)*RIQ&v&oDhq`YG0ZHq7t`UB~FclKS25p}keE_G_zoEO@m@
zmi=E;c?|-AI2M#7_k{&VbQl}fQ%8-w``i$VodVgud_caD@164lyQr~`jdDctC?#Qm
z7=i6j^=BKQgM=B=>viQ~t5uoQ?^<ifFD#61Myd<b3-Y~~K_tT#L%sxuL>)*Y!e0}g
z;$pbAn^|^?kqQSMdh9iaAM8H7X_XLbf6``l3U>%m6IwnfC4H5@F#Vl~I9b^#Q#r+Y
zFn@nlepQ`B+PH9-iEob~m(QSZ9TR;E@?AW&qdmVel6?OuqoVEF*p4hfGnmuX?SB7H
zzatX?N`{0WMr@tKINXf??pB@^KdzIWWh5t0{4nR$V%(g?9AMflw|>v{wpSnK{A8Zj
zD_XI}5z=<8h{>z;pYQYN70%){NH#a?63#XccW6cUXTX?STg?r@WoFj%tY^!)73m6A
zIu^e?Uhr)JJL#|Kom$vyvU(brW8IchtPfXi@x&}x1`%>YboJ3ei}CD({Jg3I_cNb0
z0h&u8dZ(55*D-T|JpBjB3ALeXoRep%vxDL|pM=#Mh<1@&Mf3;%7gkDkK)*)24TN8o
z2(I&-b>!rRe6v&asspThzPR2%L_u^HEhxDHha~$sU?<~c70q)gXX+Q!HryeGpeZXl
z^lolG28v!rP9@xq9$aWR4|mEaNnQnTYp8d|R!q8{`AQl^7m}PZG3f3Zy}8j3Se25J
zDC{G{-|tI@bAlOoES7PGC@*j+#KY)Z!V_k?yC2^p1NnRYYU8@G_o}FRW<u2#(7Z=%
z0C1%#ct2nc{mEZaIPgP}gpO$iOTH%-qUg~rBf0Rdf+|o%!JEktfrv{|M|mCRA#W^R
z8a4coaYVtxlX4xG01|<-9B)QHlYD7%Cc*BNSAOjz7_BhQ;Iq?$`%?I{Jmjl;IVCGj
z`18tltZ1Q~K?Ug!yKg`XG(r8I6w<0OmJvW8UP=@X9PY8mJ%(lXCOK~8#<)E84T~c>
z<?x~uAoPtpB1GBL>&){<oCt1Z)Y>Q>f1D^TtmgjN)dQJ%KKx=mN+L;BXF5}@nztd%
z!R}*GRcdutJ#ViRHD6$TE0#UVsoa8Pf^7k>KH;-hl3wBRl|n>h{FYPj_~oxQDvg>^
zal|z}kx4lT8hP2(uNx4rI}d#1R)(Z(1i`s%x4?VaNI(#8<VGe22|_HYAmmsj<!P-x
zDFFtVotx3@G81}Oar$|a=l&JF0+CKTDow-IDKUk1QOOSSYbiSWw-l83g}3QQMz?Cq
z4Ypf=`8UG0zPiF$v`-vC=j3I{;5zKKmT<&i^y!yNWNUEVXeVu2gbh`?{Y|qdwMkiB
ztB5tQ70*=h!mIx0?7&q2uvA)FH`9UreBv|zPQ>9;kj&dtjk<E@#>vwk3=@pLi3c1r
zYpMf+Y;7-LVzEBUA-@P}!}<GkbPaFu^qvHW;xEm|z92+3tLKfLuCh893c6Az2Ynis
zQT|hM`v+(v^d6p707Vs5cTY1oNt#5Tsuu)EgKbOG*IR3Sy6b@nmrH~zNKPntjgSP3
z<#2sclA_70r&44>VjjJB`mZ-5<s;x0vVadzwIJ9DVLu2Ac8HbSNtVkv#^&wdL663O
z^)=W5545b$Gd7Qe^Vzn9lSBsrP1(j4NaVTj+Bl0dTq|xM_%`)8HSF|wjM=*v0tyak
zB?Y?fD)91{l@`d`ppbJ?CGzLbp3{B|;$a|!6?tCT2Z}<o*`Z(hO3DcDYji+PXpeTj
zh6hR!!M61|yB{1%%px;*=0gGDtMC|0dJ<M*W{HpM+`%X~h?4vKE#;lo>@bE2l=58k
zJvsu~9ibd`jvl_6G@?9{$g&o>MTpcb?g7BJ4sbHq2@HVj2r!*J7Ouxp*DMm}?NC;`
zV?>^6UFzIUV}1Nq#L$AE&4jfV{`-8dK)wa@>>7z6Bx+||<jLXE<m@Go>2P3ziZ)Sr
zRo-(wb{LQ$*ukWyG{l9=b>|M`=YY$~wMh%Ar~uliJZc-=9~K$L^-`U=NC+TkjuQ#y
z>GYb-wO3kxem98Hu7PI=Q4=cVlr0s^Q+}4oGvY0cw<7R9z8#PiZ=4{49+oQlZ4UrP
z#`nRLvG{u=d;3&XV`d~yFHa<r*&9sS-kXq2%s|%CR!Np&t6UQyl!O-&Vr`vbC#<P{
zv|Vs$T<MFqh*(Ek_DX*khsdQ8>2nX{FtRjXQj4Mr&KQ+9=BoA)UrFk^5`jzla{}6J
z?Nw12f^Y&)1d45T(1hHOg!043%>3ajsT<dZmY<$7=+V6RT1KE9GklogcWSGVv^Qed
zSOSgU^!a@@>$J}~#O+OlmfET^OyAdH)alV&qE||xoRM9H!q8u0gz&uJm<P$;Kfhei
zn=#x7M&^@_5S~(to5nr1)$RQjy|{eA`OZP%qf%y8Ec1x@rDR6&O2vnt1Nk~hGxxMU
zRvUI7A1}n7!J<ayYh$YnSyvyD&~F^Hh;sv_YVmXIK=j*fuYy&-UB@&Q3{Umc{KZXg
z-qUM_t)3x5RkMC)CJGrQ1gH6|s|)o+YcE$tCW7Ktf&i8DWk}=vg>{~&f6|-J5g!gT
zbw&rBTH=86v`lxP-;e2*Ob2HVKxvY^F>@|HT^~fa=d!tQO~&lmHKTg}@}QWk-|+7$
zTBFdo{q^s<N)x23eiJbBTv&_<W>T)(n?18(wXXIQiSUnFgYrFHT^?M&$OMoGN_Ypg
zOOlRhFY1j_@E@%xeISFZf7ucPZ6$cENVbl#M&FC(UNLuSr&HSH`If?|q2sES=mP)F
z61z{lkMA(4f^GqI`9-~021p<`xbxgg)%ZB;c%;3HC~my-Pw{XX6&)cqcU(sy14H=8
zm%mJ{xQ~+pV3(E-Kgf`1mhM`lB4!K&iVr+85C<9zTSMPZvH@hUzkIdhYr%iNMg{VM
zP2yvJvGYMk;ICe*rS6VyRA24eCLBxwad(@XGj>Cy<V%*Q{f3af*Kz;+Wh*}J-m?W1
zHOK_4thX~q?_^~C_8C4gn$Q-K1qO+N9pg*0`Y>MN{2KJ8K!OsjGYv@+B&eI1c#R-(
z1|UIp1Q5FJ?avX^yOL!tC*IXdR*-JXffKMm=EN!Yz!0DrpZ+FEj)i|fMtu<npX6wK
z_tZ5$<lf%y&0fJ6J|j6TvtLdw!~}#SFUPS9yKX5>5np{eRen%>!Og)V312=LkFW$k
z?Iys$9AKZ*$~TyZl7%sFz^zgK;C6?;UiFThy2V|XvJh{-uW4MoJ82x1%Pj@5Jw$Td
zbDN7i>2mZ#gZp^PXm;+K)r@9wG%WZ8rL15Bcet6OQK5#4GI$DzH*nM=waU(vqQH&z
z8_01>d8`W5)p~=Ysx_Byb02D}Pe3eL7%dfaiW^SXKT2joW=2&B7m374kK~|2vGVGU
z!6JiQ5=&RSJ2~<#+VaigUkj7*sws@J0zp2i-a#s(MKatdv@)_c(yF@p>;SC3YQ20R
zBoh-*qw1^W4jKlhVF@G7iRVayoZdVmabI;(Q9~1a$!H|}(j{?5xTg;3rL|%n4~jM&
zv8<C)x3aAUZ&<y3(f-DA>z$NJDbTyFZ!Hg*^S#E5m&U|u=$xmPsx0U<3NafembC=D
zTU)PI3_5?t?E6c*`9l(F_jwmZNJ=&=x?W3N(6U(4M1;DlHo={;)W^ckRm2y4d-YJi
z<Ue}6{Gg%oGcU0Cosn%T`dhDeegUp+^}|B#@QP(JUKvm270urg)P_AG*H}F2j*JsM
zUg#Lgry#fGuA4;>7Nb*{a0K4W$9f3*Qg<~Cx6qrxY1y9=y9N8rU$VMIQ>vCQo?d?j
zU0?>Q7<?vd;y5ay_SLnP));42YrhR%r&(~jTS*vh5yk0AEo*<##x6DYl3?%A3ISGq
zD91Ogz+g69&bY%C?;==DxIXxX{uiC6e@!`=!~EE^QH^bO^h_6t7v8i24fG6Nwf5Kx
zXwDXJgP~ac-t9qAi8THmC0d8ag?wG!x7jOuU<=kN{ablJqU-~Ml`vkQN=P<ctJ(jj
zg7q(W&VStT2w*s}fr`wKVm-ZJ34BggO8IgE0f@PN)B`;Jm?n2T&iu|u9qa|zuiSzj
z@=9N&_d2rI>$a6?I$^##DntaNv_-%N;qFh^R#%nuY;CC^He)`Uk~gbc89`=*40E4+
z)R{>CA(NKn5Fkym8^QG5QP4m${nk>*p;Br(XelQ{yKZwaWAk36ImEn6JP*+={W?U^
zc!G(fM)d{_*TwSfQ@%`9n@LN{U<n)M)P?~q<!5OV!Uu!Iapj{z<*26Dx>B6R>;p?r
z*E1VIClvOBjIZ$xW%Tt-w)21h3ORkIG8t&@IT$%X%ql67`xM$56VW^_^q#9Tq|z4z
zTP?pU_nUyC8;AULu(XPTr~MLLka|dG(zhZycIkbx@T5H1-vP&D?+bKh3afL-O7c>Y
zeCql#HJZHN$GSR|EOB@dy%Va)wA?P;ytn4P9;^R}XQU+IWs9h{#I*Kj!#Hv`we;$Q
zbMN}qQzJKjsUop`UWe$*3}>~0EXnF&dJ{QuMUTO%_sZe=TPp5rG*P%r!$ZQ&s9zKp
zLDtqCOlzk(8Y(Xr1NpD7WoIH|fh7CKUhgS(KFE(|ng4oxE!2ld9?tapNAi~ZiHDe5
zM-6J=iJWA-QKOT^u=>%F0Kl_u402WTSK3A1M>-sge{B3%ky&t#Rl=T)7z@kMJ}_69
z;HV90)I?n4e30h728+7ag>PXC-(r&n2I4oDLC7H0{)@(3+Ew-_Rj9ZJSHpLrfo$%5
zQCGH;Z1q(G%k7sF%mW(=V!lZUPXHhnD_^bE>Q%{L@FUfd%0CUie-wGZ?d(4gZmW2K
zeZsR6;0f{)Vb2*etQK${lyT-0Ms2imO*upAPv>sM47N3(HT0u81Zj+vh;4akPCzmQ
zExA+n6@3t(7Ez4B({<1diCF(ps3cQs_4p5gE%_5Dd`DP+?tANh{-6Kl(*)L1GSS_h
zYhb!?09?dw>?QlC=2FG1-qu-XE-*v-j@-!I{TFotYN<k9lB;u?ssb*Q&2>Yq6w-76
z{pB%xfbom3x8AW1h)sO!RMfh&aSL}X)hKS*_wR`|NCr$rx8OI${|T7-Z=||VI=m}Q
zY?#gzz$=lEm_YYr39u5d!SGNeQ$VRivJx;i%&hqaffGQc_XZ^PdNGe?6Zj*&UV{tC
z0s*#>)YYq^_dz8O^b|ok0IG}xC?q9>A?=@K@%`JphvHs69lX1mivQ(l{BukF`w!r$
zl%7EH5FTqmidCwuP-|8>39fPgI=&#?{V4$8VaHs-{s1=A_1_DNn-v2L?XirE`59i%
zv+ocn5#RyhNb{?TcP=;MaWfI7ey^YQ4S=(|LLwdA#H)=<Zv3}}$t0Ww7(C!2sO5wI
z!QlDpqwfbb(>m2t&H!IK-=HIkX2Iw(RecpGyXJ#rjxYBze@}H@yw+A<W4N%&KrX|-
zJ=VS#v+)Nia`jD>Zd~<K@+KbZ{|^Sre>|E$1KOnR>T`z|J-zktAD>&M0aOos#lKKo
z03*;N89+>}t37)U_!1w{;2vPR!QWE@)^neTBqsl}B=P6n{@b`fHU9?YRc30?KVBV3
z8TwB$Am6t?WWcBQJ^zy;Sae72&ef#y1r^2~hAydKQ!ef*I~A@Mi&2UIWh+d@fRs2>
z7XZqWyyD_BI6svBA3TOYY~YpbFE>NyKJWyAKLDOK@XLzN0W~)Y4XG0gPoPACxvBk-
zav31tl@7|0gaE2Xp@;XrJ)Q$-cx1E=m72MzD`+7=swkPQSd4!VN=(Hj)UR{m0$i0)
z@)VLnbikQxZl{%^B>weJdMnUi=&nV=6%yJMD9!C5QeN!=D&3GCz9xw_=rFTW?lblh
zG?~ijp=He>7T{#^K1>yQPn>jx+}i}0_i7y*QgAI-tEO&E;h6-%27u|rD*=>tOfL>b
zGvCG*UFfp|MZRPJJP4a6V(?@Fj)M$W>?TLLVd}rX;s0}<nCvqFzyIocyTr$T|Nfyg
zcmSPu8}p)rp8ARj_!mh{bU7I&697Er%Nu~(EMx~D35apm{0N%+%dz(BcQE65R?^PN
zN+!h>l?p2E{>k_^@Z$0}@tJ^|Zw@U_{$gmh=EcHuRqY5;3IZn2OCW}pxTNXQ3o2;!
z@c6hxnqe8RR$P(Uj<+Abn*Duwq%H@b-Y%((d=I|u07%4v67luITnzB!(s_riGr`be
z-+h3NK+<&jt>ktgSnM31@mR+5_r+1eG*4OEjp&yb#{z#nH7rFhyKX8F7?9T_f7J!+
zI_PYaSc&yNH}?c=b+>K(=SYm5HR`~)|Fj(ZKSojWsytBE!|JZwE&F>k13rzNRiMy2
zNf^#|MX6$Podx)Di!(zzr$?QzSk%=RmB9<Z0ku$QnLJ_A19V>A*}Rxpu9Es20MSw!
zb%iSgK>7k+V9(RFg6US)3|7GUH|;zqOC*YQO0j%HtgS9;i4U4!o2j(%@XSJ?uQL;t
zyuK>(E{OTAQQ)<HjvZ>ekwQ#gad9|VXWax$+0n@f;ii+mOPn_k-11g?YkT5?ZS^h|
zB~O7dNd(%5sagTi#HE01qSlXIbxG@gvU&vegTZqpf^S_D|IFg9P#hy{Bk=TVtv{0R
z)--qdF<z`H6G{BOVmW&5v=Fr-ChZsUn!03<Y0<bBP&bgwya9k~#Q-5B?CfvV1^a~F
zL8pK;H(ZLx))~lr@9et`kJwQ%<bePgiXR~5@z1nhiEwuSyJ#5DnCopwT;?shSW#i_
zj(+VjqS*q}yNjytP+qRfXxskVLi}HspsT%Go)GZmhTd1>VEnV2Tv7xAIJbNDRpV9+
zw-10h!+WPzhLrUtTi%B&)?YNbk+_q4ZDB?FH!=XF*!k^!e(K#TfB9w9oup?1hDOIo
zpwKCr4fl2dgsvn&B8qi+*o5B|v+CB*`aSid%TS^k;aMNSD}rGLg+z<29xK5npa&m+
z4N(x3xP7DQ7yy;IU}yk$JJ$of1)i!g7N12Jqirpp_m3+IC=Px-{rs)C!Op2VK;a|G
zPbSa_42`AgADoou#GI*#&|LH5ip+~%Nj@z$%mi;~Gv3vraI|Yt;&hPi?f)y%*fZq{
zuD0`D!}r6tCHZAKf)QY-_H`Aae|K8|%pnSZa0rT(I<;x0(lILB@ZN41jd%uBeco0R
z$!RT!x&W6=60V|3Er6cLxm!(ktVYY!y8{%3sCMj+08rm#s}W!2pqwe-jiacZa(OTY
z%&68dsr~GBUw@#!(VD1hw3clq+9_C*-+9KwEJK{0Zh=*#cH4D0?XxafI^Hn;eY+T7
zB*~<)JU$AgbzF5wK!kf}#4M;%#l8a8N1wwuwOK&7qDD-C!988OLh)|}JX!*9Yo=K|
zno9fn>_K78|5!0O>j8|X(eKsUu02tL{ZJeHk7*qvEHwAUUY7P~q6SRpf~Wj|`R=7>
z)R|cs2XKKTcYfx~E_UrKQu@lAWiuKA@W(m8YH%HOhwsMv+apuycQGD%QvSO;Ny-XM
z|9{|d6z>&z)$pKb=<j=dGhF5e?d-AA@#Rs|wgpG{-69ma2|Wf1Od%`kmITu<DLRyQ
zu>}q$gFi4)%LWC6AIwqsi_p1rGBbF6-lVyr2i5=+nzn$Lkra33M=*sWo!!yWR`u**
z>DxT~lj}k4$^7P+*~>#=3aT^=euA{h6Qn4Bk!uA&&dZ|-{^8{4I^fhm!#4}K5-1yw
z8nzQ}{G0)%n={}z6Pt@I-;+>@MUxZD)&a2wApjatitY-k`$pf2_=?6`1KhLIA0_X~
z-3BW$nah38HL4-cVPV)lzS@7pG)w6<qlEyIiwGapQkDczXjj#<h*KO|2;F3F)9Q>s
zU+^}a?_zvh`T+Xy&#j+zvlot3FJcFa&=P;aIRQJAw{L)S9v#S?@ocOEFJl^@{TU=n
z5i$2tuAf#nT{r}Q_W=X<kZR@#;hHuh+K;zkUyGe0AElBB&IDo;>lt%+VgTeEaoSqy
zKC&5?2o*ua?&Vdf8=is=i+{{XWWhT@e_{x1RgznPBL{Ov#H!U<zpps~^+f}vbN~dz
zz18J=9wYwK8Sf|1bf5$9hIUS{_YHv>;(rli0ZHgQ>REUlnhhJc+O?LS);~!*_ik-g
zQQ)!b@*@sqN~wf1sF>eW%?tI%VIyQ+X-9Nop^fqva$X&2Hcyy57`;}r)i*<qBQjPb
zHCeB^7e`-nUb@kLJSFGq4g>is7OwvPPfvmWhed%6bj1|Ijg|3S+)!W=lp*aVS-b_(
z1m9sIott}}pJL^LVKn;2bbW5^3jWWvAFtrmMQePYR>R#G$aBZ2vZuc$|9p;mOk}!n
zMb=vhl$f(5$)Ro&-34g*)-_28iKV$fI3M`-FY=mRfgX`Le6t_G8F|~HE{ws8gATbE
zrw$ee+>U$IUi!?gvJ$PdH$`Mz6_{A$KDAKYB#@~mir3h!+AQxuNN~yBXd7m+{Z!U|
z+Y~s1yVvjFsL<4G19Zi`eJ0u~TB_PfCu6hS<W-=jNK{fgur=2G@-FnlCFzbmlEQqF
zYwqp2qBbllOKjgl2B8*J#-6Ak(bF~wD589N`NcpwdGP>yBGHd3jeF_aA`uCLqgUBc
z0aJ<w`Rc#c-LH_ee*qKnHa=CT!+=o6z099HF4JzKZt-B>0Qazt(t04imMQ~q;kr`<
zh9i;zy$=)Tyv~jCA;3(yeKPy}W6@hyRsCifmwDi69pH?=mcn^ort;wsn<GTD0nO8|
z-wTxVBB67idZB!@qvXx?q2RSH{9S`)2ioJ;gwiAiMz?4-L6sf!<=S?<qQFgH<Zxb%
zoIU8w`QW`uLRZ+f;>=6%Wiu$YWL3>CH|>hctaQjkm75gz7XpwxOyMaju%TB|5k>ZY
zbNkdS{)wA{<vfu&cj3<-_J2RDC=ugTq1paLEC-a6N;2U}u7z)N`Fz(NV8iXPoqSjj
z7=K9chNci-5VjtxNTMnJ<Jm{XWlEk@BFvK{QW1|fN@az;^_2kumhS;C&*9q!zW`%b
zl+t3v@?7zn1W{D;!in&_`Y>{DAJPQaXOd>@j0kg)QIX4PJgEf*HVQ4cw#m9*Jg<&3
z<cR&qWqpTcq#u-Xa2P%iij`2mXDELerMHGG1-D=&H155NlQwC~@7z1ijfrwiZnEgS
zyzF!GTkr(VEBb0k0%@3F`+)DymsrpXwUuF9pd!6@#Q{6gcq@W^QHHLLUO3@fzUX@)
zsDV^-du}oA3B+|Oz$leq?n_Pwvq0PIsSsZXe!n3A2*!qg!q)%37h~W#ZG+0OVPK{0
zC^rV$zj74b`fKb;&fRSEoJ7z;rKhI+KiJN&4vc-X(t1s6J*$EBKLu|Tn?HI&#Mv}*
z>om&mxx$od%`a6$8dKq?ovZC-=SkUFf?>}^cl8d=j*9Dkt~;buIb~R;F<57(<WjEX
zAnl{CkjWR4Sl2juxR$ffwx41F+hp0@?b7mx<fDE@PZ;R`1HAuVf9zU$oB^7DiFiyv
z%s8@r65cxfl#7X1@UYa;N=zkFzAW2w<-D$$DY7mcxzG^y=;dF$sbNL-0{D&c7{P5@
zY`qAXh|xa&jS&FGz3z|ZKwlYYBbXIO(+a9>Ouf<zbzL*e!9~kyPEs8?;X6cM?6X54
zw`B-=3KVIWQe<ZYHMCBK)tdS$Oi^%vQ<f7F5e>OrvzD6eg);?g3|%%EL#dsykoa#A
z?r2r<TNA`>58ycd6W6zPfg3lk8b*#q{7a;};QN41;uOI2d8)~}#ZrbI!Af+B&b{8N
z<l38|?lU(Pa*B>N;C@}X^0l525tHb<_Kpr9DYa%*ujII@CLjPnmG8$@*SRvsQyWvf
zU{~N|eV#q~9!W=!#@}0p@F=(f`A@gfI@8BVc-mYl(ItLc8vdulazmNOkjN|M@nIG2
zERlhAd`kAy7+jfy6C$^m8H7`L+_t4$Q%D}kmr=|y#0YT{<Ubzec1GSalXfBX><n2&
z3M6;@-)G>Tf7y?@0xP{L|6@35ek2zFf4>}Z7i#-Ks|uuHyNE|Y^z=1&&ZZ#K&x;~*
z%vNOF5d*qWJ|Zy_^0Hr)eG^#B74>cCU_LNpI&fIN-7!SHcfMrmadpNcfSn<shW(UU
zNzPFy$9ot(INzhSH|;A|R7hq<;`cT)O}9Jtj2m|oOHjbILjXG0TcWW-Omv?^2ZKw6
zC{F}i#qX4&_JMQVO9R986_DV4sDT`*HbpDohF>;)NgP;@xy~h2u3lp;$1455xO>mI
zrnYT;SP>+GAfY2Q3DN|qf^>uckt#)zF493ndIxC|s-X0$7`haZE=3UOf`EWDk=}c6
z^3HYk+4r1#@7eqQKfmAnVF<}ubIm!%(*{V-@vqQi3wnRe`TC?)<o7ujbq;e`SMsod
zHea66YpcOeYZdr*mLnNvdfrNsUHs8zVq|aH62aMp=P_pWHnMir`=~#IQ~{58Znx)i
z1Nm0m-5+HMK6}IL24Gmh&w3DRU6WmwA8!iM4mL^#aawG#jo1RF-<2!LE1Tfv?$tN`
zev&R$ClzTn{GAzwP+M?lN`0#p7Q{EPV}f$JxXraR+0gh1RP@`zkC@Ps+}!ScgJEVR
zE^HoG^Jh7l3XEG9Ud&zl?y?91l7_bUf%GJ7*+O=o94XEVH54J6b3Q^3G-ny_$InlW
z^mKOozwER@GHw8q8e%l6(fboanSehqkGq%ZesS?l7zl}Z&^B-`bg@gY=kh7bNm4>#
z34s5^G%N@1>8$2)M2oi3<LW_xX>YsK+%Sg<fW6e>y+Bc&#)|Jzgx!El5YM5g?Z&39
z3U8^NXJ+y2C!5~HbadUfrX%=VSAN7>KXn9YN3KXDx+;d7dQX@eu%0HZjBd_j90kED
zB-CPpk3pEW+9yubgWtahr8l<cJOpGaE*NiD%;Th&dYjY<iHk=o45dH;sW#&Qa$o~I
zz??0Dk#qa6nD7N)GV+JdZix*UIC)61a56ppbiPsI#zkU6dN4|-_TignlJ?Z;Kki_m
zRy;ZXLzJV;f#8+kNs!=bt0c`xOA;i_$5$NFDC<qQot+f+^8i=hwQYC{q(@&c>lg5M
zR`xolPQ1_1rRc~f^k)`;t=a3eUN)BX8xY2jXd1Oy!=q$c;$qR!f0tXaV|3;~$o_>1
z*w2X;?AvM!@dH?09=~}V7;I@@<6Lsd@7{!oD#201h5zT1k`ThtqCD?&{!!;Yq(wZO
z^!^OsjCJ&DSF@1&U#>phBEr$bw4y6FvIoOg0CqbG9xqa83}+J*5oI*Sa;%%E#g5f=
zAASmeMbOMty@}o+TwD&OF@m|faUp%PBy>4?o-P+w3d4W1{Jcv@7Rf?XPSK|1j{P_l
zc74Q9>|Kvs#F*JD)iH{npaY&XBH|G$dxZq4-b}5pbDz1Fe&DNU1g7~RF$uA$fgV}M
z8cRW#G1-2+q3du3kk$Kve(uzB<h<YaA*ci2FTp0<ZZjFRZ}8E&-zJT|D{)?S1Gs(N
zWNK|>^(1DZ<(dE<Gp0NS1?_SNZ9EfN;%dE%I=LOzsg%TCyK>XTyOOV?u}RRYu9rFt
zzc`6Zb(3hg8^y*nu)z>YqcuV>0j5;AvabP-_9<A$qMSl!sf~k4hbNbV9WN`SJbM~G
z|7~-+m2e41kp!OaG>Oy(*FWyf;TCQJd2)?)hx7m3J0Hd(zj+Bg9RJrVcP{utr)MwB
zt{3yf?v=>8La*$&0_wStp)G60K`^iPn?-lAhh~XZpEInOe)qkp$e1%Me(N1W<{_|d
z)+;exu3EVeP&wf%u2(gBFi`A!Gnp_aQ~4w@W;4^n`MQMt8=1Fo{I|g`cqDonCqOo@
z15{LVN5F)CiEZ5Eb>0O1_G6(J*bCjWezbXfS9Ljk=fx-5Gq_Idey(JCAi^a~ZxJP4
z4HW=DaaJ$pb}u%nx9Mp`b9>Vh1>!FZz-Fq}kdr&6@XcoUMq{M2RfGH^T%(m)gVmcv
zNp<<5r(IZ?URbixnZo1nxyhdGGB5=Xjk=(K!g@$?pV(`WyTk^fBTs$^yfC<6BvGuj
znEk?{Ur5tHNYf5?a$*Mzj9D(Rb;V0=Bi*1|W^4cvz6z+vp2u&BXV0uG7rta3u|PdP
zmb&I6pZE^HGd9N6kUefO5ZpMqLWaNrHus_!F~fbGffu#X5zYEPZXEfPfHqi{$!m3<
zvu^VD?>n>K{`A@Y1)vP@7ysTc8QMYuo$5c^=F9i+SuQ2&l(q%@2AF`Z5-gDHg1`CZ
z7aF%x<$8%gRU`jg6Q;c>z>Cb`7kmFMZTM=cSGG~oE1M%ss*ioZ9-EX<0SYS`tZ6uR
zDTZ^hr?yU>SaKOBNz9+xFurYW;-2<{;E-SaVEQ>Ua!_w!+I=Z6M|eDg^sXg%;?gMy
zGrYIjovU{GYnfUAs<_=-```<d$gamR(q|j8bP1Lgy~GkH8x^=>#k8eo;RY~}2<D@2
zAyWVn(Q#)CDh|9NRdsFhN&uT3QKoH@3^K3_06W`phU5@dDt6deyW(+Le=rs9Tj+7(
z^fE1&(j_1%nD~hj5PC%~+yxOWLeD?lk^W_$;j_m%AgTIH`_@zNyn0C*fEJZU>&=Tq
zn#!tSixeuOkRrg=yXegwg5fforJK5}frWzY{dc)qAC_PG$yg?h9ZN8D)z!FnGs>aR
zgUY4UYWm?^xW|uh@XWXBSA=)Xmirb?gVXB`S5QOD#&6Eg@~wZ~g}(;LBB&wog6C$T
z<3GWR64_rMB;TROn3%id8O3XEZadUjbW*zKQeG8+;%;qP$KAUh8v3qT-d)NI%mA4N
zn;{e8!=~hV8fobuXjtZF5`dxuSr1SI)gTy5cr4{K@Y#DHFT8x~eynfDyWrhO>iU~{
z*g@DE_LM8wt7Kbhw<d$EOoY>79RU29rH@J<nM=04v0t<ghV<OTJc{0G6aRS}uV=%_
zZaH}=+WQwB*L8_$Fg?|Ii^gaaKtILqV4e*cV+gucKTvgEODm}WsJ=M0JP2f=cV|ZN
zQZL=KxbzTr#pKcywGrN((WWko4|fmJcF3!Vt^}Fm<C;=JzTGC4W`shnTR_U+*U>33
z#OJ4go4d5i&z({u`c2_rw+f(>>AMp|)*U6e_u~Kz3=%3=mZ)`LD#Znx?76=5;qdYI
zY?Au{g67Lt^3G3KYvC@OcPK0$w279?n}AOEWr{wQQywS^UHbK7@c#7__pLu3&VTzY
zN@_&N=zLdelug}br&n8;*kr~1*`;Q&TJ=4x1@VU@9xnx?z?h;7NB8%Jspp2@KU)z2
z$&HG&*0Q8uiNGWQ_n8>Q8l(2~3(Bwt1391hJJoV`EJ=eMSy$NHQZPNUuiTi7<UrEh
zZ!1JO%JbIs9R#L<?8_!uXh<vaaIdmqbronI(jR;kWZ?u5{@^Ak#JFne`rd<tC$rbx
zzG|MR*;6^){~@yePgtwMbv&)RPxXJi`qy%kgPk|(QNR!>D8NL<p2{TxN+2-iX@Cb8
zk%A^nA)ZL#F_A(vk@!l9TuI4?Mwpoq-u4kq1dNKw@5BLn+9+}EBFc;arF!5I1z~25
zFv`X@8xsI-T~e?h4q<eC@BrW|r&m#*@QD}cKz3l-`P>umjY#HnITGi~&bLMy&~mRV
zJTpDvBiG`ymu+nGXv7dV!XhRqG7ivuFH;geV<n)NMH4_kT~>H_d2XDmi|_O0NN3-T
zK<1zeq_<dT7g1e6LTUo$PdyoHty4Hh1WjwI0xL8ic$`F`h8Pf^9(<a2Q6wE(0IGWq
zOdHa)y{+3OCYq8wPsf20JOHj35PX+Z;eZ%|^QV9A%mN+hC9uUiTq}-+P`Az6yH_4o
z19pHA#p}V;=VP7$N(d9aDZS~P>iKl~h}6LYNrU=l;W3U3f+*FCsHU4ECK_}Rzo?+P
z-aV~%E#fXs1Pfin0Uem48~HH(IKX3uXwI-_l-5P<m~E96;1|V#9#lI=!aZR>ML>W~
zC1)RB>0`dq<P2#L8&Uf;%j-i%8S#kZD%RHmeP3Gj?DcB@IyZw=$=%Z7(G#x$$5h7-
z&UC5OURhc~6n8MtG>R|)kSo&CKg;9bvO>V_o<%wrVcBZUk(KA`U`Y}%g*EX^s5EOn
zE0-6yT;QAO#m(Txw^cfH3I_5lthPyfBtU%=eksxyI}{S*QP%(l2PyxYTlfggCgm%A
zZ5&QI#o#lFZ7<ULc2D85lVNB&3khGgq;Az_i{SH@-3%*^LNf1pUwu?H%lPW{Q-s<^
zTm_&2yXTLX8BLm}9<sU&ecbW*sjtZFQ?~!g`)XO}PQBcGqT-K-Z{A$+Npy~IUtSkN
z;Xj9F8u@w6PJo)xW#5TNNb{$3+;zoE|608NvG)d@9c>H5Lq}X!*Zree_m6pT&P)I%
zO}#aEM>F~r%Lf@({&k<m+SFh$xZUX-834$`DGo~s5ua~!$g&9C4hy7>!!KoN=7Pat
z&6Ob6x-C?|_w`rlEt5$h;^Kgk4Up(Al^DXqK6X1QFzCYM#Vvv${%xCCd`99laTb#8
zlR2|1+dpSK_=_JlN?a5>A8ols_<j>KirdF5Cy|WeuIG|MC}Iym5$pB{ZY^sjQ2b17
zBoOTghWik1XWmyjnD*V~qUBC~_v$W7j11G1M>I1}YVoTI7FrkN%uwxl&$Q%(CO+Sj
zgKW<g*0jd@i={uokwC}t(+=t+B<vD$z@`1(9z|CF>0527=%FKo>?4_3Wnh`#Z_0km
zI%VTukA<q_Jy$uOM9W+3t1*PzKj(I9KuL4U!xE^%pWKm=0GwHx#77X?EdEAVGVtjA
zM&J3ZxL_l-5JK<!oP_vn*l=B4>#Vqd;8;6`0F_I+@f9I6qpoQ_u@!bqd>3ZK_)zxI
z;m4fO@rcPx;?kG)ie-TyAn?br{hk=?TeKMk=cc9fagyJ>=l0;7*^QlR^N4eDR}~a*
zM_ufEcllbH=#N=U6{uD*J!%=>uLW&DBhDxM7YvW2cki5T*US3JHqCh<>vE0{ac45=
zA)joSEYq5!y8TO3>2t>W*;u~MO-K(rd!8EYGTsN`DRWxR^$cyMf0Oq8<?4xqR?k&i
zLvw08=r#T(!RLZ#vDe86H!fb(ZIq3w(aE&VPRh>XHh8t))w-)I011Lm%n3^MkUdXR
zg)jva?K#HISJD$XE-X$RUTI=(Cc)BB`<5?IOYi;kTCXO&wKgrzbPnMPre>OT22}FG
zVhUN|xur8?#Jlv|?}{kCm9}m#Zg5Lb$!Z5O5M<8~s5|SfZ0*nbIq{mQVQgk-N;{rX
zjj-)r@+V)nNrTy)Z=x}~yL9=oHSr0;{ihP1==F(iYvQsCZPMQfcV$OEIkCFn(Q%Ca
z)Y3cb0kN!|>z$Q8M+`1wMTYx&q#df`mcd(Q$LG<^9r3tayP*NaP1$$}2E8iV{nEtr
z(H`{<MbQtM-=o4hl6Dkp62v|`(VvnP+m&6~p=khr1y%c*YIpq&PB27P^pkVR2s*!K
zg?(2}|4q*~w_NojV!wq5HD_M2;NQv4q&|^uoT~vnzDfgb4>XQ>JMG2OzWAUPPqm&0
zMiKvY(fZH-p<RS-OrPz8**^oPSn&0bGFF({YqA%cQb1@(gfKqrjq-!X5*E;yNVT5~
zcC@**7vOhOXt5!_{|ayZt2+<a_=g;j(CBVH$*lZ;X+;tr!+zZids+epPUgQ#EvlLd
zgXX8HSIjTpmVKDZB6F)BU_&E!{J^O&dAB`&iysTdXkJDhc$WQ>kp25_HUe0FWBO;#
ze*zR>|DuCIjL#&<#<i!M8+V)7<%<b&VH9~{@B>H>B+zElcs_8vxA^{mh#UKiS_fpi
z&lP(yB^JiLFWM=zd-rnj{_jna3=c&MSXYIov-wv4bwNIzz4!#igVC~`_^sAms`VK-
zscH}k72?ziRJw+~kW7q-R5EY-#{CAPbTyYU^dfmAN3zwo98pszZOu1AuY(kj+qGxw
zmZ!guz4s;E)}4CXrT1Mkc0GN>j#o@p%QMRNM&<{6*N!D^@7sO>{;o7|x+aw^<ric_
zQV||NSQgsdz&vDw9{`2I%imy|%_g8Rc*<s2F*M@;%rx@?gUB5jJo4yg_^d4FFJb<}
zKmCVQWG3x}-Jazn{PRT*9bqgb5&#=3eDxI+3Zj|JnG-<IFs!#8@R#rM-{1RTKKb2)
zlzd6kkAL-7|NHCz+sC59Bl4r7y`l8qe)xCa;J@B0o9ldVL*9sBh4SBi<Ntd7zkj&9
zh}RoB{*#&ikKgd$KHUEwzHA~l9@Hvq<qaA*eQg)ZnveNoEwE-tf2PNm*T4bE2`dkX
zc(MkK$x6W#GzoAKB$28^Ev#{?SM=Xpwk`>9Lj{X_>K^@L{Ux&D->BVaat5Q?D)o*5
zp)5TP(91;tJXsfxg2_OApjR9|T5H-tlz@gS2Z(1H76D$LE)`jkC3kHDD9I$!fZXkp
z2~Lj`RM|p6e5b8*^9D0`{1<5eMv%B#>^V@-2SzChTcjDY@?FM*Mlkk)v54(EK>Q%8
zQ#q(-n*~f&p0xK~6ih?#(ovhDRG;2i08K4R9hd~F4rxF1myHgZXtIIIuX=Ypcnx-N
zVm-G5=o>-ME150=gx+>r?fopk3{@;s9Tbj$hA7z=47JJvC5kjZlN-=_G(T9)K3URf
z=;(QtaWcaWUU8(Q_bBvQAL#GWe(n~zWr14wV+vR4V|O5QVE^s#8yaF(Fb&nftli=o
z9!pvo$)ut)(IewzRMNY7>RFIPvu@@ZZ{B!o+}}J9d{@B*e>-VY_{VGf{V#2h$N8C8
zgW-;SprJ^RI$cyxDCh&bR$@FL83*?rp$WAZV{o1Nq08-fZ`lcaB6WyZ)1MIMfo7$u
z0F=qg6HLOJu&#_6eXgO{N}<x%@}XPodfi@mv1*SysAC;lpo^?~FdclT!Ugbmb+BH-
zTkbqry*vjlBlp&ZC{?FO{br;yh(3?9y}Amu5-o+r`u@3<pjBXRX6pcYvsf}2Uocr9
zJ4a%c(<LNAzttPkQI!p%qU+}+*Y7|KyWX-w`Lk$mAXh2R_XEQz(*Qu8!mm+Ob!PcB
zd0@#N`LJ#?(!T4F^71F^+Gp=VZ*7BZvEGFWK+#%y0HV;J(5(WHwVuX62gyp{rxpV~
zL!zdMbg#9fg&{~nQ@3houdK-E@*R`hGgkHEQFxYP)EJ;o$pe;w0^p}!$YXa<cfOc8
z`S7u?0j!rrFzD~m5g03%ma>Iomjuufi9M#r>*uzfs_BUj(3z?EolG<9FAoB}3xt7{
z)erG>eV<on3}31NobD}1>uL>v2X#<dtGV<i&aV5cdo8{WBvra1b?^|W#M?le*btg}
z@Xk{5x^f=i6c!UrM;gXJ(-JB@QU%J*e(>SqQTRyzvlUBeWqZ&5ls+J3+iaOX=g88d
zF-H|7hrBEW_H%0>#83x?uJdu8O@lp<`5ZE>v9ka4J3Bo*{C*=Yfl5FaeoFfaYOKqW
zJ3DMQwEl7*{p5V@^@fO{SGTJ&v=x4$NI%EB+1Vx!lun9j`_G&AyWQ#qjm4)f^B!jG
zmE}a50tKF%*&;Xsqcwhm=W!0pkFmwzkH>-}biPx-aH)Vvs@5VOht>om#B8poH^<es
z8H3)v^f6w8Ni2|o2%myD!#l8xf_W?)E5L2WV-jMk_!O1gD+r`Y)L{hF4_MLwGHVUm
z-p7S0O2mW=Pi16}IEWwN4?Q#UYQL7?Otp$VrWg*yyr3-F2imcoeJ+i-8cwK{baKvf
zoB%T&2Sl5+?_Iu8Npk$kV}{uHKsH`A_t>1*OIJ1`4XEw(a%zAWpbt3BlI0cV-e<+@
zaruDdr^@Sh*r|W4GEZ=nB1(5B!~+kp1}5_ECvrF&Fl&y1hdhNhZd2`yJS`4pGT<OR
zpv#hoSlyDijX%#!{*!}f{!uMo3nv87mExa`%}a@#f+1?{%X>i9!~e+J=n((#ipI1n
zFi$fKCn!Ivg=Cbv7q9ceC&MN#VJ&tpH4sME?@jofIVOzsBpsdP1#9TG*H#11)2!$c
zSV+d^oo75-$3*WvimK<7G7g&Ax0nr|N~5fhn~hAO6z7Y$EPor5a1wvR_<-rD_r$i^
zP!ur&F<%Z<-GBsiM*7D?AFVPepwR^B1W%_#p%D)`fWT(2q(8q8$c()7e21h7fi9i2
zJ@!(e9%1xOO_O8|Sd^BFJ0RedS+}gG#_6NlUoRvTcCe886`u|9{^MNMivI*q)g)|-
zqDQw=`l~FCLB!5otas4Pr|{Xm0E~iJy(a6_7rx#$_dFFLz6yA-MP@0Gfz87Bz>0st
zK2q**%(rCq&ikb>*;k_50SPPUq^k;tU{}k?5<xd5X+t{28Eq^U&wLor6MX=5OU(};
z5Fm)*I7|&A(71vKk=15nrAHpeWF{O(E~wN|&xp=kYEF9++_2|FW=(X*Odv>ljv1e(
zC`~j3`=CpAKnk@*nF;1!8DL5GGimCqI>J&{CD@O;Z?QE|b3U2uu6kHs4<yL_O4kd&
zi>1PJv&KGxG4eAW4Afu>!%|sfSa26ws7&LM-zSvhO-)JyxD5aeRg1l6uj?<aRZn32
zj?mm~4{!$ZiFLmMh)Y*m6A?`&x@2n51H2RER*RCJh7x+Yb12>vX(i?LWT>em_S|ua
zz>@T5PhfDdY++1(Fn=qZv~@B*cw=ZYrUBkqUKH)S?^5h#m@XbVV{rPJ1Yg{D9<kl4
z?gEz?4q3>LAQ)+{RTne%&2T0^$hSiXaN4nVSQ$==HcS3a9;#(a14>Vqk25Y37qE_P
z-mT@^tsK<<vAX=D)}oaJg;}rQ-sOLuNcAwX0W3x#7A$4)oF~)Qc;PQNYOU%4^11Dp
zb4JFW?G2Bx3V7|%cZYqJ<6&Bq=_r5#Qov}<3Lyxl5M#5M&#go)M~itRHcjM?N~A4i
zsNAumG{APafr+p)UcpxO=%;dc%QqP}ohg1l41J2mNh5C~D$By7+5~Zi=1jK(ya3rO
z5q2Yy+=YCSY&iH3Z!-q_&Si`ECFvF_hBU`1>GI@@1sC+kEbnFg2_^c+CAX30TR=$l
za>xf9Pu+Qg+X?Ioeu>JMK-C}>+z1swdb;9#!93~qMe{qekrDUfT3c2Y5~A>{m_uJ<
zwLO6J<oj=a0xC!o*$zL#q!`7Wc60z?rk}-m(#=XvF@f!C4WFa;FrlMHxhV_&&>-%<
z>qeVldtVPva5ZfCk9D!<hQl{>vi*&$-UN4WwRk()Z+gY#7TSAs>bE>D@SVJ(T*V%G
zS2=moME-VB`e*aMyCLeJ=9i!t`e#Fsjej2bCAxbD$RNF<emad<=gjWd)~1>>#%yE1
z31}RS4AsY@s-l&shM#LE19Zo#2B3vD##YjR_griu=I!E*RR`mD8RVmB#$3$bMhYu+
zg89Q$s-JV;>T$+cxW}k?nwW7JId2Q3X}e+~4Acz|*W4OXP*2`Rp1O-hl_?&n26X^z
zW4M~M@7Gu9vy<ukXBXzZz?k0;cOjsC25Yb?{_-U>aJWlW;kaT;d{GLV^>Nkl%hCR1
zdP}(1jv0GB0L<q$zCaG^?T_w=hC}irG_JM=`=a)TB?DPxL|b&pGZKyGK+M`*>${kh
z(5(@QeJV7Z#XNZ>FW`GMuGFj+5$|9JF)ANaf(;Zh$gWw(2EWI|3eOm{5F^Ps4S4}6
z(tO&hoD|i>Jk)w1>ujcW+Y!a3uL~vy_smjuWA2PeW0Y01GTwhZZ6CX<5W!kX?Ifp*
zmB2XX3yg5?6P^l&1#Gl>W25EqOQ7m%5fqf$Ra!2n69+#hLuhVheNWOp+>=%dZBzKM
zr+|IQt^3`ee?;@GY?3i!0~(Fm1>+Gxx+dm6t-}+<OSYK%F-q`J4zzKh*Ts9qrqSl<
zpWE^Qm*W&=aR{JWwmGJB#*){5K%|@_k{%#b93j2*Mtk17f<FZQCDKytkyNk1Xn!JV
z02u`jB{^h1Ul3b{&28bA7^SoC@^{IanD2pp)mv)hE)Uw+&@}R<UV(c(SzWna6;n%x
zGe;N1TI*%{EiQwpB5&Rgqm}2X_0udz+$?U$)N;Sos<1Z(a+K{`Q*;J#dmX0-k+Q7c
zq01{OW>a0?Z&}=AoKy2dq1n*a)bYj5(-y8}7DpkCOLsQEI(<X_$0qHM7WOT4VtBTs
z-~QuJ`6l8|Q~QfRSK!vd>%cxB=t^C_ZB+x5om*~$>e|tt$TG>^y5I2uO3`f83kd<`
zi!i6v;M^M!w-Pn>CEPA3sRky*pqQ+N{8*M?@?B%Fy`0=7gyt8P3QLWV!?@R`Dr!UH
z5%lt+I;Rg<mAauCmYdaQL5{hj55Kqa7ZLH&FLU9orgrv>RBBzbn59Y4e3F-!N;^S4
zAJb#VuKkoDCSgQA(#9M#9kxh33hS$!FeVr?%&jadC4g4;tk<GDlp_Rto^a_}X}8qz
z%B%PXk3#te97uL3ju8T6jO3RH=tvJ?bYb6PU__98r&L8OI{@ytc<(^2n??2?m3;pM
z`R>d2jk_`NG0T=+sCHyqL@Aj$Melk3DX3YA|3G#VGxUR#<dI=YcELt#IiGK$?E*2n
z98mhV6h+6t0JskkL>Zcz`5*UuTZJ{Rnx1c51r1)$zR5kkYvVJZ)rwxfvL<52vzDPj
z$;zymhZK34dG!+feT8BCBI}7mmxdx^+*@<RUH*o~$k;|mKH=!8-a_xBpN8vlArY{Q
zH^2}pWi(PI#Z~)To_Au)r0@NPnHF7&t;d>;>ELe8)>{_M)XVysGRlqsKb4lS*Fwh5
zEYR|%i}LLY>TF6p<2MbpMCCmnogXVjpl7pA8ML|&B{~ydT>)(%cLE}!e|;q!rqJAV
zIG0lV4(aE30?Z>R^L~9-QL)HEio;?`^O*V+rG_5n7KZ&Ri@DXUWYpD_hk=2hqeEqo
zF48SVThjS7yUm&U8gD3G=zqhKhrm+YCT}ml;b>er<N!1&7MrVe>pu2KLwiG$+NaEZ
zy9LIDqp1WMzUmwF;Z}j;u_i=*u>6zzG9ZdTB!Iu6VBrjC7V4<dYEoQ3zWAFSc}G!Z
zvRL{F)3raAeIg|uT6xm^deixU3k3ept)!tI6DeSr6ELQ*^B!nSMU=f)I+Z<|pIrg8
zj_xfh>GWI&i~uE)6Cupv`<=>&4~2u)^6fPAIuehml(E+WV;ABcuo2=d!^(ujZk|?6
zM=1j``DbD_<XTt+5|1`VJdM|6_GD-7F_GM|j4*##G}aMkUPz53LRx<##gX8YjruBS
zX3inm2j!PI_vKR&Hac{iq1(u>-=614?dHusiOj$tu{1Ed9maV%qBl}YY-vBC>epNf
zT?w^wg*CSbt&`xWYBu~#h_B8$Go>5Ge;?=xCya^h_}*Knq)_#(U?YT9(XVcralK8G
zZRi;+h%z`a02cnM$tS@5W=FvH+r|P4F$NE-f#JhqG4l5sB|(0jVZZKyDFeAl#Br<x
znwhVKEOL=>?oGBp+m9Or6hU%9BR~!4*(LEQkwupNw8`f-ev80}o|e~<p)rvk#s*Dp
z-fdAY3g*<eY7gJ}N7Br8ZMaJ7E!oRcMhZ9NDyF4!p3JbKKkR}1r8`N8QVlcFCI$Nx
zVS*D2y0dnD&S37Y=89uvnTK|;XnHki%*(^nfSJ(mp(jRC1qv>{bF!J!$Y4}Tg3}Fw
zIY7C-B#_9|7rvv*XDYq+0wUKsU!xn*KTL}Pcr^uNsrE0agdXXzzYyF<4a4hVEveu)
zMK9(>wZ*4pTJ9a6yg40{K6xT@l|6G3gl?p_n$zjVc!fLolZ729zEPDZm=tCt+NSQ^
zIVglQ=cVgx+g_nwQ=mEZS)*TjwHH9vF6h^Ay4N}N<alpOd^#mJ@MLf~`?A)-v>U}1
zXAjP;ZrN}=RqCwq{C+2By9S+{M~l>csr?<?5HE9b#WVcR)vXmPgGI|@1m)4ct;_k$
zUN|i4(T`F?m;()>f@+hzru>>&0utFs6%+Lyn{ZgT>q3msrh+%-KG7OlzL<y7d`68}
zq(Xe7+{E)46Ghc%!FdzIu`>7Gfr&unLlnIhVF$ZfL42A)RT^j<j(-=O{~3w=Nd-@{
z=sH?Tz_!FLKRYc^Y9pxbj^1Q@_fjgnO->%hE)PS<W7MXjOd^7E?*=hwqqXxLMs+u{
zMd~;+1p7>=5f>weL!3U*-S<|rc=JLoV(BOrVN8t>quNo%3YkVti^+O-O-tc|`A46h
z2K&P3@?o4VDg*FIPN&NV+peR~`W*qb<fWxJb1li!cULZPy?Bp(a6btDUeI?<G_DGX
zpU|@O{ALA8_8~}+HJ+|UN<M}=KAGOTEfO@QT|3KT-#!pzm6(vR`_7UIQ5cyghWtW>
zcoM1GO0#ErgV>`0)|mfHwH(dbjqF>dMubwq$BVJKnFC)vJidtvHW!{ehP|(bnYb9x
zR~4;$&x^}r!sRhG@&L6PdTvq#3w;Oct;R067TPQimJ5@5_*<I6R?1BikNtr2YWZL#
zSD1Y!PeMU%hKu$(Vwq}8(0o!N!hg?~w{n{<z!cu(3wP9kZ<)ZSjC;Qd;`9hR>$-35
zc)<$g(Y$sN%yw7Wi(&lrr_Nfxy<oU%ds*1p$HGlQ>ch%=u$3<WjlL^xkt;DX?p+iu
zC7yoWtl<&Zc-E@aT};~=<<a7WMN3|yeOB|9R=bt+=(ymM^x}fW{Z@4AN5j8t8lT~N
zqGZa5|832klcg^;x|f)zuQTKw75Y6~m(rF7hKw|z&PIAZ9Dh^j64>Txe(iM7znRS1
zHXM7SpM&-(X;C31Ju2?0$H<^1=gB7dRg>%@mJ~$C&A4Hn6ZxG|B=Yg>^5gYRn-mn@
zgUx)Mx_lW9YnsU)S0<s>>v}Z9GLb29%#A*>K|7}58l!7Bn|aLMVrB26M;0B1!cqe+
zf7y*FMdOObqClf((`8Njk=J)KJiHv>j|QVHD?j$KbS~wrZk&|s#Jjj$E{%F0<{ofB
z(Mx43KW=mS`Rb^a>cR9?lz<*}JS5NIC)atk;dU??VtMD~v{)QwcI4$&{xOMZ_X+4|
zQ|D!87P6Zdg1!ecNZ3ri5JnTtAu~rKt<}bB_{AfB5f6PGYj4nup8#?eCuTRN8qEW_
z?1;7md9>E*TnYa%U$G)<XBU&lO=M-%@Xyr*-?`*hYGRzkAG*9;g5Q(%-v1D*SLwr5
zEZ1c&;dXUDZG^d(JsdM_#3HVUu16v-h^!eA`N1bE9hik7kHnYz`gQ{j8A~lDm;565
ze}6+E*<Kv3>PqU23k+S5^ZlUv@5uZA;3XdJ;HxaspXrSJvjlWUl0y?2wa|5)i;wRp
z>hs`IP&bXyZ<TNvkGY_jIJJ32u^+z54KCk&H9La2U-gf;h@AmgDO&63Wk6zC`G9Qx
z0BFzna~?!msSKzV&XToh!nh;zr0zsAWE!Yt(eeJ$ZzU=HrhIf?)8eILblFGTg;i|~
z6<wbi5rQpds#tHqY4FuxWnOg6_G?V*NcVf|Gn04gH?w%HmG1<5td<Y4^(j54P0kHm
z2I@s~Ev&2PHU1%ijwbg^65QAm^-czI8zMMgFg9Mx%B?@js0FR%Q&;*pKWAbW5k9l|
zxy~H`oH(+(;`eX9d#}3m@Tg8Q%YJtT(t<qE$LbFZ$mthl?_~9nMQCBXMO{4GGz-gC
zR5^V_My%d8*D7B9rM9S>cRce-y_?hb4v&7+wn<kKd{roipFPVmhjJFz7nz&O$yywf
zy)&z4X^Al30VQ7|GT(%#m-oXP$r6)#MBFN5&`51^$#i|`>I)2@%8*<9(L0sztuDI`
z^-#SVT2>+frpT+ar@KFDt<NvW$*6ulL4wnnt3ib=ChmQHwo@ljOt*Y$pl>M7Zp&!>
z0d-_glq8&UeQmZDN5MWI&~OuE#pR|NC4KKv4XzJw>-4xG2D---*Uo(%wA0D~eW54c
zY18^EY`&i1__hG&peV1?Ehml2;Mjpw>7YThjq!gy?f>B{iim&KG+}k9_wkQY=wSt1
zhAf;z>TuQtJx(_ig&9hGzzpBeM5kV@OrtvO@EqtkDnw<%vM`pZ5X^7!ra;f~Sm31w
zx>7kt?`G$dfz_-Hn%>L(Kscl~{#`E}gP%O!WQ=xJiuoxj2_l>+!YJXheDb(Adc#DM
zyvV66fXe2%0NGJLNHzQl*Wm2Tg*=3FTVTMa`6652qwE~%)5S}4SFHJw=PQh-?$&>X
z>Pa<tA*Ve0i-QZ2P+q&V4Cruh*u+~;yvD<PSyT5sEh;w@y^sM`BfwNo@yMcoOtZr4
zfvFc4k)Odd2t7!hPs1L>L2kG#ENgt$IJP1QiM$_*@Dmu~FG9x6C+@l!ll9Vi!5(Hk
z7K;!S)U312TLwObB5^xg#=aR~R*y!GuTk7TkjtGGC7Q;(NxhbfeVqh>^_14Z6i%c~
z>mbSC=c|p@_h~^T)H4+X-61$;ILo#qHK%_+aC;`3??UybotNA1k#&PW+8%Gn;~E*`
z0oRWis=5L{%htagFQ<ogl`~td*x-)jco8}QYxC8A{QgUj54~T15|YY0ou+;#i+3QM
zNH=l9@+8|jgWcu|HtsX|?}NAJ(bB%+g(XcysyXzP^{M;D+sQ*_w_V3bi~f$zd;_RQ
zdM>r=ul``LR7mh*I)uU<L88Cwi05*lEq!pr4#YYs`$I6h?f#<9V8Q)Cdq#Akp3~Qx
zAG%~_3Z4-0_GTQ9%|{LIhO(Ua?7U@G1^ky8bsHPGd^hM#816*C3|5gN6n75ZzCvBI
zh<}2tvzRi*34MhpW8LMyUu&jp{-8mL!(up7-`c_p^NPMj0YsEnxjZZOUZrt$k@2M*
ziS(X4mJ|49^fA-z-QgZz*K&{P_>h|wIr$r{1QZBT4PuXowil}K-$(*SWHOhdePuA#
zHs-nbx>SVdE0`a;7!{zO)3hB}KMtz2^b!C_JU_smlr>(YpYb9dzVZ|HqDvI}1W>_^
zZNbtw*YqK1PSK;aVSvMbP^kfkX%>j$Vn;+l<U?A`<E{$d`Ky^-kb!lvyx7OiLM{sy
z@lGGu82aDoe$WF6iK!*4#2Rb)*ZL@Q8q|&ihUBzs)ck0_{p3Ja2?&#=#B3b|20WSb
zSkeU^_iZ0)&Y;nPuM-|mSSFBgB&Dm?^J^z{Q&hzm(EevQv396wNR5u7&3?&|in5-)
zX#q0Pf%@U~|L|o0Nm=tf#(Udyb%pNFlOypl9z>VGId{Xb9QV-4H6wm-ME(OHuL*PC
zGze0;JKBt)c~JWgwrp^<!*>~G4a&`Vf>x38;uK*t{%54GG}R`u!C@b>s2}w7Th%G#
zj4ql6=#cGL&JYRQhkuCf<w#J7qfrq;bIxUqM$O)9bTsT*`~&2^$8k6O-aFXsT%X;<
z_050=thTS|VC0c9`YV@g>cWx`T=jY-FEV>zjaj27Ec!GIPz5$kwuObS!{ZPIl!z)y
zL@4EsalfQ1K-#Dfgj5J`7Y$ntsD$KN2{zYa3@n<jp({q5s6H4dX>rA=b<bbEl(U%%
zkzZmz=OYPSs}aZ)D#TvoXbOLm1hs(0ZE@6+37k`=b(Lnz6rLv^j{PP4L|^3kbAHR`
zCx+;tE~I9PtHdl2ZN_*gDosQk$I(~YCn*vpb#~UUO$!F)JOUD??o;YsQ+jnexP)MF
z3nLf3Y-wkgY;?i9)h7TeJ$dxx%IW71AGr}8)QIZ&Ls<QIyzhK6RzjzC*>IhU8Z0zq
zCjNYX+_piE<>PY0h<-ATa4depnH{K=2%MHyxMOTY_f=u8hBr}225R`^O=(-XZ#=TU
zSC*<AL4?Ujp8H6D`u(6p<D#pe>hyb}j(f@Up}wCc{TLDKI%3{rw?E<M2lvyy3n8dU
zPbc;m6OR-3<&JD8zVEutg+ODk<#Al=N3?FSe<^n0)4GI9@|-lP{NtGT6{_+OI8%6$
zl4kKhDQ<}?<v~8u*aSpJiEAbHsWzH2$DcLQclp;BUf$vub>yQ5cE=rYEA;Ssn_@Rw
za2{;5^Oi!``&R4)zdfV@I(Y<{$!?w(SRq6;!|A>XoE7CCCE!5~p#QAFmgUN`!|^;=
zGwG?}h`jjEudl~cAa@-cIyiHj-z`BM8eD@t96D?n_pgiW39xmTj!5UEY>2#jA2}2H
zm>hm%G-hTUvM0kAIW5BhM3ERmTL~{LgWSqC!3`Mp<o!^LojlIIurA8c97l8mdG+Dp
zO+Ql8?wQ}RV(2BCmnFzu>-aB|<B=O7xwW9EqJh6JcL#8`(NklUUAzelp%FUvTT8Zr
za|N=mA<E#uv>vEU%)@pE^^37EFVhh@W_h7<Zfit0g3DFwVv;Q~(i5(s8h*w5*9DIV
zH0vbZaQUc{;>clF{N`ur^zDf8VO=G<R}<GM;vnDW1K?V7nh-4h-aa35v1WnwiV^rh
zQX}weh_5-rnofMDjqj~7RGN;4nZW$uoC5oI`t6%4w4x4cKt%J5X?2A!5z-0=;|29|
z;S_K?xcFmi{;5;=bC>w{TGHPF(BNMmkNvZbOmxL}pX}~D08HWQAAklYo++uZl8uQf
zaR<m?7&sn;74cX{zC)2GYo0GaMhJHsOHH}*RV*QUhPUJ9U}b%|=}ztKrC$taoIW_`
zh%+}3CfdX?17r=zI+OTGWZFX)C%82zi&c}1W%~zfa`{(2E4v}zH{#SJSZo+dNqIA4
z9%CJ~sYhg3zW_HpZzuQc(UN6*%5RP{2gx>4&<~eK*_GwgeanUhZR$3Aj4vDMkQwH3
zO|uY?!C`MMaR!MM1PeRbLI3>$_sJu_^#kuvn<tXd_uw=04fjK-uqF>{<#vB{)-P_r
zMr=A*{-Uy!2i+M+zxA!Mez$OJ&5=F>&5_BAB@HbT>XMZvE05yPe=56V74|5k<V!i?
z7!nkDjA_3WB^6_<x<PwSRQSD_r0K)hVEzXcvrZYu25WCq?CYz`i#&U@NBI|i+*p28
zOU_TMflKYoJqAiWo~T<>*eQ8lPmT2Dh05&X+IIy*>P3-aq0hs<5fY+XY)mpR^C{~l
znO1}H2c{}~L&z~O{p~)%Bg!EFG5-vYE&5Vmw42Xxt*|C?_V&CAG7i2P!5`kW4>q7o
z7orhzYjxrvG}55-%C_{7@@NWC3}4KAl|8%*!W)HG#mol<nUQx3K?`!2q-Vr@ioq8+
zB5w?>XC-I3yfmk>cDPk9QxbAM=Agr6mVMrqH}w_lY1d+E2*%OnxR+^8y@kYyxtFpa
z*mGol%qU%Ypyol8_7Z`g_9wy3?^^0#eLPP#TPxQG4}6O2_Bt3I`jd?-n>Dp45sp23
z9Nvs&HA>4FEihM;V>9MLw)w8w8h7r~KOQ|}1WW8tchuMWFzq-37IJd?iw+QZ4B8XT
zX{7V+h-_PF<#|Ol31~{Zpwo9TEzMIK+@|-JzLEQiR-DldumbazH+tE;BrSD$w;6lv
zS_rnkfXgxE(4f`fbNbht9gBaaJ2}~uU%uX-XQn$I8LW{Lh+CwUHZ{e*5HC_T$$A3H
zDhY{qH!NYe@Op%E#9|PI8fLTamb}s@$~-o$+Ge$wbiOad6NA3+cFth;m$GHb%Di>X
zVO9#9%~g~ywTRwa`=Y!?3IC#r;j#v|(w0%jkhUxeWI^#Wg5;n@ppV9E3P|x(&%ESB
z2YUY|^Ku-OeW{EX!C{7zrh=WMBv->Joj8r8EZGpB&6#4Z@*5M22h+wB^Wx$uGau*y
z|A;0!@eV@yJ}-h-5sfQ5`+dy5r0c!;#fcH&a&6>OgS$osvPqMu(P8X8q6$YwdMFn@
zC`|i2H%1nz660H4&QuRKX9u{OW1$OoP*C(jWwehY^S1{0E@$zTcJ?pC`;?E&e*P{^
zsN?=Zygj&yxR4DJU|7b%76CpmWvSW-;EA_(W5wlz-b$zy0Til>4@o<_9sFHuO(0f%
z<hEBn_kw8){8XR%7P`5qw%(xC5|+w|E22cI5jEZvd?t@YhdPjzniBc5+K1c|Z?!S3
zyP=3H^nMW(^s49Ho*_C{2`>5mo*_omgPU(}fqu=OOvgzdg=a{#$W0=~Cfn9F4|`MT
zVnn_4pi~O~J!M|9E`#SA-XwOgnHQy~x1gM4(Xw63RU)!Gd6N4qrXSx$3!|@g?0}+D
zSibqagif$Z+`8&q9?<*T3f@VSEn0NXNZ8%o@&J;P9I{+Z*y}@2lz>isPSOixsbK+E
zRS-uU@~SoHYbwjjH_wb8GF^({=<(<xj<A~(TAPnP)x?0Yp!}AKLFHv&8>`fbX_b+7
zve1P)O<GEACpABqO~;X>zIe}oXHLY)ZhvK%SCHg3hw(hnP3Ya{pK}6cSPWvRVh~bs
zg<1}1^LBj(#w53N9TbU^@7||p9vOa!ZMJkq_$E(}NX(q;Sn3_=ZHa85uZNEdK1wH@
z$=?6kQ%pwE57gM{J)dcLI{|>5AiLG!BDmhG-gO425|L+ZuN{5ur0P+$WZ9<6+1w+M
z6MQ$ZIL66MOEy9^hV<5YN8)6)p3g;Bt=5wL$*YilNJ5Qnf8emzv*_S$fW*4O$Y@!g
z()C$O2N$num4=OXv%Yp*o2_)AmW10BJ^H={n<>vcSn<10-c43N(4nP|2wZa70nMTk
zqGx_nk4Z~}t<!Eb1F<u=NwL^!>(^SgZhYzsl09%_dRn+`)tKbkn!IIm-;Z0V+{K|2
zx?LUv69R@hb5Va?09%{`6gu6vcB4NOid_tNAkJJPO#89vjl_gDc%F4vxHXQ0RfaVH
zWR#1PD(Sj!b|_o3@))t9Xf`wHV_1;=IBas9vFCQFSSE2?NaP_y8L5#hDhVyr<Zaic
zh2h6q!F$=l-Qd=oh%|0S%}0lhv<RZZsGd?5=FI?~--q)(1xQVeBOnG#H4k-UM&B|k
z-3#~y@5TQ>_KDbY<V&XuD<Lb|Nc%IXXm%Y77P0xX3w;fFY(|&+*wn=jzY@!rNEQ(-
z9(@&lEgBC=q^$?IK`rzbI(BsyWn{-aVok~>1zm{ar_2-4=N*L;xPm!<@g?`xL>Aba
zF5?lETIU342XP0s>G15F0&z^5hvHE$x8gUElF+tz&P=;3Y-+5z;7pgJg!b}+8=VQU
z38lk3J&=X6?9CjzykOZZfO}zMQi~MpwG529v^I_{ijyB$5>1XrAs@y6Os=?6n(oDL
z=VE(tUM*1Ib!A%g118M1Tq-CZa;;C~w8cjd2C3t7tX-Lj!4f51v99SU*me2Xm5MfI
zA&Coi2*KWqdoWDhE7K3MyW`}IS!ZiBOJ(w9&rQ|P3U<wQ-E14iXhp1JZEh8X`ObIQ
z$hW;5Gw%A>)qr_tKs+z+9Bhqt>2xA%MYufy3vz%<uhW3{7}?6N!wj<7b(5d0^2yVX
zH6@A1)JC&RDlxO%caq-Z)x0_XdxjhwZOz<sE1^UJ6JQaFS%_3A>h<&YSkLX}N3;mi
zC8c`2=avxvxnI9u&U$uxKDqSg68p;=nw&j4Ev7h|q9;!bEF`u+`foNL%vLwshvwa-
zU2mJ|q8iI_=>e%SRVewg@uK8TLGH|1;k8cpsp{~{;pgb!*bif!>sNkn;I)_Kl}MY0
zi;2b6@DG<sZ{PC!aqfa|zl(nK)#_qv;wI5dBWZ%$0!T}?ImsV(;0}9`uIAi$@z=E*
z9|er#<ZF=8-qmYRMfb3UOa?9MoJY-CxyNBwn>c%ohu6J;K$mFpT^&n2Trx7@7|lLy
z&OOD<`|5Bd`?8)Q9S<Shb5|Bx<j0K{%W_z2UIWv6Hf9-nSkaAqA_D`hn^zx|tFaQy
zFqsj~FpZoe9(i`6GjuU3MC7D~)5Y!`lVV2WOV`f<826p?NTL8OLw}m%`^0wHtLFaT
zSK+5!H%UHz^L<eEqHb^dsiApL{+=D?Mj3N^uyDeuKLaL%W$@;u+*Oy|qdFfI2k&to
zIeNMe1DbU-A^dDh-5Qd{E_9cyv{i0h@%hw{-S_qIm~M1+Uan8JEm(4>^R(-hMUo5O
zTj!dd(dE<y@l^|hZ#e@^9@Q`M3EC4K^t<T?>;}x2(0n}WPBW+$iL<Jng8e7ERZEUw
z6W(>7i!3E6OL6GF>EJsO_*z>js_V9D58j+=kmwvsl(AUz%A#hK6k4N&r1P5J9NXIb
zQdMz^V8N)kq2O-CKBwtxpQ3m4A`$FIIozvG^L%s3rE5!;qBMi$c{Ro^Nx7lu;<u^J
zePA4cp=;ig^uo>)Qkk-kKulTctk68M*Yv<5YB5`7hdU=~OKRR|%GcCrN<=Ab!&s8t
zG{%&T3lul~{08*<4#%|vi?o^YidU!5vC_zsfzX#}QN>TDD@g(bO%tWl6R~o;4-)=y
z@Bd9`{jYy^2|<$L!qz&Y|D|#H#7e;;hTx12jdBMX4~tOemnq*iooS4{9n^-E^+thV
zEo-{|iHfKuZ*L?s7*xnG{sM;kJMB^;1Nd97J`aER-#vx}_j%1PFOTRK?ZKIr=*WPb
zaes<SW%Fd}h4G|aBC~!9n#&%a>K&?0g>&b`SMQ*2YlDC<Qp)k-?W8fTzfm}>;#2+}
z2|f4^1c?6|auAas=%}>J==$RgKf%sj^&g7nkP6j_XdY5wym)Ilh`_PdgOj-*gk{B0
z>sh@?a@jO0{F9&iUq{UUHURwVLQC{sBD=i~#kc7_Z6wH8NLbqaTRVY*63`Xl+zSo<
zhPwH)8vP$29v^g_KTlfm{I3uAHy7%U*ZRjmm>;Ln^*<d9|M~*|`=<SmKYuv{AjeN{
z>HQtL{1<QX5C?9PLfBBpUoX_Z_)QtS&%kaV!Voq1cRH89xDd3Qpc06gDtG)}YyU5C
zk3ZYuPfx+VN)qz6{LjzD_xTbK{NJ<&QIg7#AnBNmKSJzCs+v&k4qj9!=P*xyy+%}l
z^+d4I|MA{}iQqC`Ui0StbItlslDUKQ&OV?lNE8MxPG=y@+qt*Mx3sX_e1T~PWNfkw
zl>ikP>4R4xl@y7Frc+RgrDum$`pa~nmai1C*7G?#UH8jKKl+|CXPb1N#NKwL<k-|T
zvY?_fd8Zg!^8Fy)CnFvhq|2Jgl~s-oGpyK8fWGJzu-a~xwIRbV&-UfMcz+mSDS?t)
zyrX_qDQ#R^V8P`C$lR2npg>Q6IK^k3J?REQ*&`EON_kJ<JMPO2zmW{(cWQ|A9{tUo
zeJBIg>jk`>r+?grPkfi4$lpbfach0=5FE84lItElwmxnqHsrT}dbS^8>n(t0%?QZs
z_XpZ;FCBmwt~jZ~cJ*v1pYd=E#7W+V5b9$a#bUxj;5J``p1T^LX7Q5ze#*}Btus;P
zJ4A(TH_+5?1X}|^b?WQA_bD-w5GK6?B*!ABkWwEw#qIR~veFM}Chr3k+<S5LK9P~_
zx6G#;89;LHstk$qf-HXnoGV%lh~k|A|77N21<8YzdCIjtw^Jrkv&Phr@TqR#3rGaA
zT0f!mki$fe>O2FF*|<d@f`2lSDxVBMJ5C6mk323hi9PM@tXlffW?iufj7W}Rz$D>x
zwh9JW^uOCZE3Q5ro4+ElAtMGv=pSo-iquY$ScB-Y)vEPk-`xUdkTH_|UlxgYb}$@i
zb*IR!?)q1d;=}mDZlY~-Y#x$M)vVVa_kona$}%8?&2|H61=hd>U^n5DJ3&|%1-L+-
zG?4tZ2uhZ=Db&z!Achx*Jk}od!0YK@Gb+CFn?t>iw-e;N2pYFSkx4rs%kKuHHw?9e
z?Y9vdZ3L8hO8?zG4}$|=(<8=5PXE|N9$HZR`mzb`dIDDE%NB^yd>LJN!~;6k`<|fp
zPeIM?Xc10bA%E0Bt}i{G`lR(_x7be=4ANDILhb<_5pM(c`tH?>g%>96K}17WCiPVF
z=+C_hXOQ)_I}%a9$4<o1#0(GS{*3zeHBn}x3b-lyfapIV`ew90R;V28@NPT@-yWew
zT8PQ7G!bPo&!2(V#YAclYU>Q-rYaogN!B~gj8#Ek%GXrWC{ZXVJN3rLl~V5v7qKh7
zEw11e#P48Jq4Z1fai9S6@O}c8$08*3Hl<qm2ss}tC6{$Yi44Z0Af;a`M>@nxLK1QM
z(T?2`#y8n{5v9^KbULc5$?tY`jD1|zK5Tgd&#F6h?n51Fvk&$@%lflJ66^Jlo(ImT
ze|?lXIuii7XU-s&C;J6}a_#hU)`o7cZv!hJLNQ~uo{h-R(4@;Dg8`h)x?=^SAudp&
z)d*w^0=B-iDjMv?eaHiJ`roV$MJO91z*Q_lvL;9e-VZEA31@KKSg(u;9!&jPu0SKg
zRRxwTd!wC9nyh)C*IWd_E0)wWho=?23iMe$zIz`oO;*)i{vh2^b8OyjA0J36#sM*X
zp&WH+9dGby$z!~d2&IpDUql(yau+wM|IO~Z5c#ah*uHj5|Bq9UPZJMaIXNrZN8Lv{
z^%*?jLqnn-)}8OsPE)fU(!o|b7-`7-ARW?!HC2;7-C<B&1#GSw6d_<dBw@vfzVH#K
z89^#TYP<B2k3a`_s!{;j-C)LPE6^>X)!q9w@fOHpDg9Id1|{KudS0{;7;|)K6%L@m
z>op4U+R|oK2_Vo;w26`Y$`*YezvC71Ul2lgM^)!AeZQ7H4Rkn*&n}Q+m`{EY_5)RR
zN^x<&{3@_n=(5OaBI<+<$da(Rz=Gt%MXdVN2gu$G^+~3<@s`1p8ut@CkV9)9K)AF^
zgb?<w0;TBOT8(T_Bj8gG8NZZi>?w*&=NFB?s$PRw1nRpnc^qi?H6jLq2fERxq*4dJ
z(6Pcq+hjw}-3PSQ#1?Vhe%};#Hk~;Gf6kAIpHQ=0HGE|i_CsArowXu+;kATi^by2T
zaTjj2!q9lvW_K>Jr$*NU7svdrYO9V%KlRLmDEV7*Oi(meI)Ny1pc|+sBKz#h(|zL!
z>qzm<SkIt9xt<1Zq^7_Si@~=3Icg|`hl1S3i|)Q(?CxIY%R`?OyNdQM>CI4ev&ol~
z3n!%V*cmC4dND8)C{oC<Y42~0G1^G7v(KY{p6j0|Q5)bCYrswr%VSVJl`FuaCSQay
z^RHiqoqz4#4m5T9o9a6kP?_rROj$U&{3c+GcU`K1a@unW<E4idYhKEaxuJM<*uP~F
zLocFwece?hXhk~}uy+L8^5|)9*;r0UE;<GNil4d;leI*6^mG_TLv-%e2)uV846AuB
zNL#>Sl%pZ&u6EFgR?td9E47meMu1VB&e}bkj3OK}u8$5X<Uf9n)k34I0E0{w<Lca1
zu|!jp;3RKWw6v5SkCL^8Yd%y5Fos0_?*&(b@=D9n2FO--Lw4i3>`{jgwMJVQmPIU^
z4so9f{+@SV2bL<2uI5I1PA2+{G(Z6sJhb{_BK>)nJu^aH-v@jFGY8VpcG_H#vIjX+
zghcM%T)SX;#Hy(`U!7fpvb(%k`oLx29`V{2wOQZ{Y(I9CP6zIu>7^y-6}}1*Pev$`
zHyrXvc+_9!_cVBglr?z|Ocpd_tuE#dPH&F~mPBwQ$isZ)aR}a#V${q$GV>xe!kIdP
z<nIbS6-hknrJL>4f7ZBsmjSs?K+1h0?|e15f%LP(;5*$f;II{pW)X@I%U4Z(JX!4d
zBq2oZ^(eo8+w(*Xo;o7GHclYw^J}bQ%1bJ^H#>`^OqQ=1s#|lJu84d@P<j{5%7aw3
z@P8eAcO-IyhBNX*i8Ek;z6*!jhEp~&&VK-`a3t7M8Fk9fv3r=K>8?XsF^$tO%x43>
z!2$wKzRY)2&@m0qvnohQf7%HilZaC&+^Kxu?UF6n(s}OdS!V@!z0Ruz1ow3$o&}uK
z-PSD>8~zXfa8}0%<D+Yj==M!;Q9OWavZvhHHcK!uXhpBym@lcZrN8fYxr6+TSmr_G
ztJ4l)L7L*y4la4LGLOz7=ie0KTG(ftKGP(b^8eiO24G+(k;dRUQ9P`R%yQo`@HbpJ
z*#?05?^8BD>Proa+_7Dq1=ax-M90i`M_k7!ps_X4<JJ>AlG{l=Dp2@oGB~qZ9-~gL
zxiqiGIfD$OY%&5(bO7B4^+U*GkgsZgN9-fkJ;4nX{0m|Y#(tC0QRH2a8^jn8TGNUE
ze@_|6`is3c==(e<`UU`?@o7eAcr&**OAuckqBGlvWQlr0JAm3DNfxPuV8HDT8B6C(
zr7R@YtRhTiwA<mjJmVmKB*}8zH*IiNOR<dyDt*)t7tKjQ4DTZ%4U{0;_`){WKq+PA
zO_hv~;(QN)bt?%9ny2gy>jN)N?>~$ox)t<ZO;rsSW+9=et_xL`WuQc3mtrrt?8u1Y
zW|g;sQB+qC?0txQe-ZVA&-YF-<nxL8ScVmguLo5wyJm?%_8t2>#zCwx{VmUFU$tH(
z>=Uu~o~GA=@lP-16)jl)f9-vFJd}Iie@94&8QY<3GumufBf^9kMp;57vWD!0#**z;
zXe=>PcHzVzdm^Do60#f0&IvQNNX8PP=X-UZbMEKdzf-@{{m=7yUa!ZWW3HLGzT5Tr
zeBR4P_gtJ6unFpgXjiz?Rz|Dtz>!wT%@$26xR8+a?lS<dj$4V$ifLP5MfYpyBe%tz
zaI?3-IUmpvUl((xB>h|yJhZ4Xg>>A$)9!tLf^3cl`cqqe%=}u4qkxmNPTn@KptQ-d
zaCs6}fVwVVvh)mSNsaLFkxH=}_v8yn?NIdKok>_MFJYbE`5Hw-;VvZnA1Ngj<SwaL
z*Hwd?;+PGoE={EhQco{-I3t22W((7p^TWE7=Zo%TJ+5KzvM{Pnq__Zg->TqanZ6XY
z?9KP9Y9IR8GIm#-MBdEBrw&xPU_(<>ySi%pC9-zs!f@VV!+^_BTq#Ix1is{Q#7AlJ
zkJ?l2T10pXqz~yv1^6vLjUG3uO?2O~J(~zrrXJKU_oh|y$|9u#2HTcB4w6q?`nQDe
z_tfZX;mH|__sq6!Tz-2~$|=#$i08t`ff8TU^s1(^M&qgLnkpxNiE#5?7f6U>gkEox
z9e{}c$PRn8N`VDrw8R~tnZD<V<fP*yGv8p63iJ0Da0(h`?;7{de-8YrZ-bhl3mGEV
zpIC>CuOw~>pGpUv>h{9vd!(}HDc6a6L8-S#;vA-+!*dzIZM+9bXJkQg)rI$nVrsfp
zfY2isYnaS%-0vRX6PRl7^(I$J5q+c)x35aLow~ZUl7||tH1V|(>989DretJ*x28dj
zaTnQ~hYS0a2`R_~YY5jW_oG$hsY-gI8#UxL&uQd3a_qoT35w5w9jSfFzUnw-!d>eI
zUprCy)Wa#$pnF8eBSxGdVPeJv-TL~DB4aIE81q4e#-2ypEn*{zP@D**F6H%+@0;qF
zka!G}f&Kk`P*(w-odI?@^DL80R6xWQU3JW6RsXO2v*NVvi*F6r2j`O=OS4Vj_<0TH
zSgv19dE83lKH!yFUJ}K8M*<C>ggl>59wuBRUVxXt{GM{}B>NIb00VR#di4gPQg_LX
z6hdHv8|SPsm4S*xn^C2y$#~hMoGD;PpXGT>Lfmdq1@vB69*N?8HNusjNPG55rKM_;
zU(fy=l)_A;f<$9J?CQ_6_h<uGB;>5*ZXu(e154(1S<6uz>*S8uyuGFRoYu1DL(Ll7
zPBp+9HzwtG&^%V_B;v$TgGb02iB>e;*!by<s<*9`3?)Q|izLgAF8)QxJn;mSeo4aP
zJtLTOxBRp|rcaZB?bbcaMy1%ToyYxK&y06NY$(dkl&|)AM^{utfEf8m9+V?3q%Hf>
zI|pYkJryR`zg>J;w)7mB{g`E;XwqG2`s)}s`=Ap3CY*eFYS~NvmmY<Xb6i`$-j+h~
zhf0)Ihpf>ybsYQo@Ta~!wKsj{{a;19f1T3TyCOkpBCS5a?4J?tsUvhbOWIQ>kv*8x
zFZsjweWXue;)^(HL|m2*0+v$z#&6g|KE0HXfm7J450c+gl2&YeqNMQrXeH;kzME5!
z0~%#CUuY2ax~N20Kms@;*(xAVVAHypdZ{z3>ud5L_6s3*Ne%m#b)5-DY9sRDp^+^b
z_Vjtq;Z}f~_vj+Gs5Ymp2$2kHBRIljVMIbVFi<CR<xdyfioJArs28{cj790+s=7hw
z<q~i<XGq{~L9K8;d)I9+o5L2SWrsxWL6pG~j*#~{493nw65AvOW;-=xwv1Xqc<{zR
zvsxll<z9h=joK|5m837ku^C!+oSL$?DoAzY^5#BLWVdztq{7agCfT<K;bU}AaW$-y
z$}ldteibA7VOJnkF7iRVIm(u~4xON>VPpOx{y~U+JZh*uh~dm8!0r+FhAI^(M!(in
zbwhlDe*Hktox^vK0eqaeq5$qtUQTQz`nU^+J3gW%dBUtJPWycFv_A0j<}c&)OduC8
zHP@|hFHb-Ts*1_g((Zn6a=Y<d%>Z<9`@MOy!II&qwP^WfrR$<WoyiU*8Z$sltQOB8
z-)Uk}rOl4|yc8&f%z~@lJzCH%hp$GT#9W*lD-_S3i?eLv_Ll3fH}8;RVvO0Bp=!Q+
zo#loEaUD+{vSh4%Ua;y}3MUdBcEfvNPH?yCBs^VY_gFs6urfxWEf_n-l%9_E80wN^
zO^Kt`ZghBbdWonY?XRl|EL<K7-#Z~h>BHlr2cqLo%#r7V#`0CA)!pqarxmp<)E(`D
z$ARVMvNy?H=&cM2rv4E$I(RS|zIAJZD}VpqgmGD>to8(cfE5h;#Iw1GFW68XmSkUV
z1>(b_5X%;bnE|7YCUhE>6F(C_iWVq&`^rmk`$cfJ-Rlo()WCxt)Hjh?(!GM;4fkY2
z{Y-6shas$^;7qT>1M@Lx7BOV;WjN8hsX`hl$&Em90nhc$<kdG)@M4L6__84mXV8Wi
zgh=>#wgo3|eIg9Q44+03_C~>+3(}%tXQChji4Fdb8r4iXp6xz$VHFp{SkGQnML6Cb
zWJ`jmFid=hy!V%vQO{scVwm`<^_FfsYE8}+DY=e_ZH8UfsC+-~C-y!k7JL-``UUj0
zoc6LHl`y*tXt8eOR2K5!V<htEZ5O#oVfyB0l9tq;;c&NhHAFQR98<<*T7bveVbGrs
zl-JI-K{;5vT-ftmi-?-wfquA#lY2<ZVTQ@$QxQm{VC4`$9nc}PCI?6(BkU6qySd>E
z-Xpxf?y&kxt}{#)rQ-^hE5iX{dH8;DDdG?}9A3T}loby1iy{~)FY-LHy}~7W@Z_Rm
zix#h(nr0=xoB=xiS>PgJ?Ql_s48P&yD1tbkfSIBD+CYHMjp(Or$Q~^53_4#NAI$7u
zH~<DBns76kpb-rlW{zbkJKohiA_U)8c6}h3LP&|i$C`w(nsxh7YYF+w+nve(z+ppy
z-aMEO6S5*#q7f*WY%;_W^EJje0{1KpqE&s^9<f?TSh~mbTDpsd;N?*f-xP?B35Vqj
z85nM*&$E|31Z5LU@8_l}iYbils+&Nz*&tY#eX)4z#jtKubku0B#quWpVe*y=X(o0n
z9HdGiz54K=N<7n}SCIDiLkBo?OYlsw7^d`Us|^cI4xiq4yGw=M;?5oL0&x1zrGBr(
zmuX2GYNT%D4(ykjGyQF&YkRn3qYlxAp|}RQGeE*K^<uDhES`)De6P~9%mjPa^xv|`
zv&<nx@)JXzAy5H|#=G|oEaxLXBgU_)>M=yJJFy!|l7GEDEJ==}x?yE;Uff2E&6u!9
zp3Rk|r{XoI)SpFS%`oU0m+_p!5Ag7#XE@*;d4=2aK65#2LkEaSQ2>@GDJggs>!PWw
ze&PY=4(teV2t;soP0Fo+Fcb42_V(6!Mcv9J#rN|^!w=e>Zqj@jZ&v%(YXi7ch#~bn
zLQf2yBbX571K>N?`8f5a7tYgbNJl?N(R)K#h1ZH%X+OVnzPZ&_#CCn3ho2^o{ZO-m
zRD;kRu~lJfOy8>;!OB2i8MRMgKp2q0;gSTpliYSchjsYm^B5jQc5uTVOLYn9;iQEy
z2OEbrj%I!4hUdsf-zILBS4J7GH2*RgRn2`E{}7PIK1#>!S=^31jaoMWU1(X}S7#>g
zu1Rnlo}a^fnjF9*lNrePT|<`VcRJ16wtLmwU3>*6S}i}&a*b*4!$h;3^Ngar;KV1r
z?xtB#!-W0o>dCjcW;ZMyI=6vf`O*@<{V~ivO)@2pwaP@xFajs=rqa_&_>%?5RWYQx
zj)f44xRvQO5mA$Qo9f{m>v6+}Bq_`FntqCsSe85wd^^ebyV3_{$TsOl`sKU7g_}7`
zzQ86&u>oHZfgvTkj;NgvLme#l@z1U8!OCKlp-5*dS0WWWV5%WwPBGA=sQfah3rj%p
zokY&Z8>va3MtO#A=$@LY3?{*mJ_tW_%|xBZ_==e}DX9bAdiYKclEFSvaVQ*B;$QZP
zCrvtvlT`#`e?`Da>ZyiuO2hckl5DO!c=v$?zf)9UXZwK0*FM&Gol)kH+Stf7?}Tyi
z^s0(O9>3NcAWwFu&I=EX9&LrddHWJh{S!9)z3$<trY2xB19sO(egvy}%uyV>>dM*J
zN4}t@#zSO$nvA_TcP$8=DzeONIKC;#q#qNB-EDU1)x)jqaZenm?{c@?BGT33<r|e!
zB_@Nx3zEq5PIVDj5zBymZxX<ke=})H1ttt4lkqc7Xjx3jT;7V<XSnBWna3cMyPYP4
z<PM6x0U4;q4Zk%QzOIxD0<C=atWV3X>Dc7yov6M((|C`9=vA~8+SyGEy({M0;46q+
zk}ZE>+PqW7<!#YO*Juk#Z+fO8_+o$0wODLu+954Fk0w?xSE-!gonJa_%F#2_Y97%J
zZ_>G7mW$uZDCVFkt_iYj`}!tpM$Q?nPa<=}*I+xApr+^h<}N8t#M#CIahb;mM%M~u
z1&6rEd{UqTjZzj&whO&{-lBH9NYe)~B<Q5JC0YA6bV_6BWhAr4uTbN+V+}$*RQ!-H
zKwk9EVa8&u@gu!@DZ1}GxcUv5b!uDop)bsFy(*Zx5M_8iKWF6a4is;r>OPMZGNI>8
zvs@rA7<=wXH7I6~@Pn+5CM);ZGT0v2%U0$KFke2J$6Y;z*Tb+bC^~8CC9b8G#dVt;
z*#%kYD|vA*^N<amGkj!KsvbAeYCZ{~CznEAALzWU1Mbb#mA7Y(Ron4=#;%_xp$~df
z9UD(?7XJLMG-<0)pdAtvWa98RsK8(TOjmhL-?p+gb~&K5L8w`Pi-_dAM3ODxL27jZ
zR6vR1i`%x7Sihdx|2R_qRZFl($ky<@LY!$cjT;^#Pt^<WD;+ri_-ER^$CO5;X#09T
z&oR4A#L+A3+{88Gp1Om1oDFQ3Y8DC!6g&WVEhsN2U%4c;{UVgaCd6|l%s1FrMiBpA
zv1MDCn)UhjbNH_!W)Tq(F3VP2koaeHWG1s7$O1%GXVFgYVAc4s#xu*gHySlt^^v~Y
zpiSKCSreTQ?M6v7J&9nL1d*1NvN&ZM&^>FZLu3PiQD#4^9zJLRzT@Zr1-!lb6Cmpi
z^%cp?!$l+Y`+u9tp139L_;SeXit}_XN$dUL1}l_pR=&q|?og=AG18<js!DVPou%Fp
zF*3t|+PWO1t30+ALi!Shn2^5gVJzBglXvbHtPkhw+j^@>^O5iOMhCOxdsb(Fic4-~
zl+BtpYBH)N7)wN3XG=oPPwJaC$U_Z@6?34L>RlJ{8n)xu1<KHuimIup4E;BCF*Tkf
z7wUt^49Nw;7JLWWe<sr;$0dt=4AAag0gn}P@#M-(-+}Rf+OGM|I2D~fwT$dhE@bOR
z31_jGeT$_a*sgI(x)G4h6p|h$Ye#brr!{fn2Bc%=XU+zVp;xpQY5EdF^D~c85wCy&
zO=)`%T_jz*&;?<^Kgf;$1}A?_-{&~NX4?$1WBK}|wPxh-L+%Y>Vo0~V&gtm@VJ=e5
z!^ROD`{scSm|u2ZC_Vo36VRN@9`PXUe!~V#8#0TzR=$-Sg!jUr^T0#Wy?xWNZpFAc
zV0>$g{~dGWFUAO2_B-lNmzs~40>YTmexws<3qsyR>_Il*j8FFzSn2$dw8Q%fw<7X6
z2o$6{;m@I(n>-RJuzrK(Sn`%_7E{P}Ok^(Sx)`0O6x3S{W6V4aK^P1;<QVQ!m4}X~
zehALkare6C=yNEmlT)F@8%G6b9(5K&H`%`TjoMs;>hf0NPKsNiEmsPQ9-WR2?(h1@
z?+pxosXkDL!^h_VuCd8y(*A_PuiaQ#vkaP2Nm{*p5RRh1Lf(5fsycNnW2GE&AoAKj
zqtD@JF%0IgDx?}!m8nD2RRWH>kIaLB!Fhjr?qrSShq#*7>tG(|{J<b2l}1*)8s*B`
zg1?BvKgBWJPYifyeC{cvC-V;k`kL{=LSz>o)3jbs@G$$_EsAtWcCEm|I3%EKp8zc+
zT*v!9)%Nf3GPVJmkD`kJUZnAN(P9Du2xNm0Rxh=<gumKzUy*U?yL$miXliVI=%2IO
z%E244q}``tStru4atT|i6ZORXRr(t38OW$R;2{J~EazZ`^WCF5fb7N#z`}*91*{<Y
z!~8**a1McP$-gVWa%;9ZNSg~`HV7Ez=br&eaVNQ{Dt-S2*gI!%`%+jLWukolSxoGm
zHwFS%g|oQHI(Ne~r=&@Bpa<C+NCGoo<M?rXhI$?1GwfOYh{_Q_B84I+R^;c^UZIzD
zvB6=e#x_WdNYhGolw!J1(k4lL>eu;npsA`L<2fjM1l+-fk@`wFPd|t107tE%c9^U3
zx7&axS8y4?ey8!SB<2N^5^;m8z@p6|0!jp{3)af-$+1IDaSQmPq|hLMDi{T@8-9T*
zJ@&s2F<(Bk%(iZG*G?;S6#zqpmBDjUJ4lWF-e!00ju-9B_?jbrmID3uZac-n1Ku@O
z3Fa)_GLNiM0isiG=!9y<oTF?&Fd1c<)3#)o3{Ia>CUbl@>Dm>xTp|Cup1j(4n^B?K
zvFQc3FsZI~sJ)a0YFx)jmmanGyjJ=cB(bB0U>o$pm4uuUrybJ|>XQsl_=>cywt02W
ze*%%!u%QH1FnD-)ERk|KOVh`Ab2d)h6e&dG7s06ZsLW^C-w~_r^@x~zygDrOM%g^^
zn~Op?bP=oyTvw~=Z^89ljek7Qf~b-_V)&|tb!T-2;l~qT@-L_WI7<>HG)LNPsVtHx
z`~?2ESWEy>1h;Vj@_2EAS1Pm9k;t8J`GI4L1nNgfUg^Z0mc@@i!|pqQ7GRxDNSzmc
z0A}T~+MY9se#v&S2;ilX%;XkG@S=pg7;qAQBmAO<vNn2GNxm7L`s#T7eNhclT!<01
zObUlQMo4_D^Gl&%v+QN$tXE_~ugB~YwyeeOaTdvnvx(MN7jqsvUHY(niX#spB!s4R
z^@_PLJf8CImQbgYF&ty9b$#m5**y(hqT<ADY_WOLqvTEAfIf?zEzsVa=syOMt74q1
z3rA|J_wpdvN}yVy*SaRh#>b8(iT5XDE=(4?o_toN4cD^*kSL`tl+TF2M=UZq>}+dQ
z-r`)X$y>wVwUdWW)ZJrgU0c?XuK=02d;4SIh-+fzEQX^Ht&+|&zTvo6m*;KqqK6tL
z+pj%I1JtpJ8ZbCxfu-ej{gC*P;&OEQAExD3a9nOWaHt?Sy)gw|77*c-QP6pH=9+Bd
zB=8Fhi#c;plbTmh(HN*R-awZ>I~IYx2_QIAecI`3vRT|0rV~MdCfl$1<QUOD7r@xh
zvh9w?d|kQqvePdgZoXGf4uj9OCq>~4R-`6GXg}HJn1T8QzAg&e9@xBrzdm4@%qaz`
zM*JIGPP5>F#NW@BVfrTECI!-E`b$)lW@0C<sM!T3%uezRryF%YdGh4*Sl6~p(bhB9
z-I<Wl`{H!&*8LifYF1kRGwN~EM+b|&luA8Ig`;F|T@wk-0Nn(XFtXl<E<2kN+;t9h
z1iq|FeX#QKjxt~-m^_={_RkQ`vI#gDX=Hx7oT>he!_~a`**w&PQmaG~<%$~UmJE3;
z_4;^AE`0eKywm$Q2UHi2?;MR=|2EhN>EM?R`o#UL@CSn}cULH3#t2<=24O-XtQY~E
zG(wexf0TgiuyB6|#Q1~>jCWLvjHqR2E^N~XFE$JMR&G{_GdUp_0Bj*@`O9<=)EAly
zsTY;>e#7e70|Ufx4s5GQpe>3JLIwasg2toM{0E5EMpzF~yRsF~7j}xK-ALlnur>oy
zj;1#sAtwEklk`bkQn|fx!7JH;x=aG1q&pQ$4J>*1mAhjAs+z7D7Ow)&abvu(DAGHJ
zu)kNfP2%nZ1CDtR8GbfVV@e0%Oj#f*oPcQi3+H1tIyaz$h_;8_Uv&goV9KoS)E>NG
z^=qT&dm;F!5%%W7Q{7KN-O3v7y719sn5k?oVv6i7d}|`F9733)P^N14oNM_sRgn1B
z2&+_F2^AkjvsOv4>g6?v3*bb=@Vv&IMQ!stF&Hzkh&Dcd>q*hTpA|jW#e;e)$D`5m
zdQzIKnDmtD*-*96cVv65`ImlRE)zh433^>h8|rEzoHr(&aI-MRA+ivZkNgd~QI2@g
zsVuIBGiCsnp`F+_D~_>i<;vGK#>?cl<`y}0%Xk8F|F66Q+s(<Z%<<zn00OI2(OYc8
zuHT!!^R~5Bqs42sk<&mI;>OT|TFhT=H{rtD6Er4t5Z>ov9W7ZS6~jru=l8m1lUvY-
zD+qL(id6W6<lz_QyglN*fM}4E+?&e0wE&tCR*$xyZ~qyQnFF(<p@z!9V}s(KSIf3Z
zmx1s+L_M-8Gc~{k-(KE+Vg6DTj<mR9*q<3Q4Wp7NNmd!t%v3F^I8|&>YN!m5PV!)(
zWXN=J?QlgM`;gGh81cnhc>ToT)_X!mXPE~jyY8B6GkJ(2pXWj*FI}1l*F1b`s)oDU
z;}F7DbU?TR$NGtO$K9+c<C?-A>64JpC4zN2fUt(?t5WGiidm1{+PZt)+Efz3cz82h
zslrTobe<A?PpWZFr9@+@)tkxts#H$qmhHuL|HtVnG|kVQU1u{~MicSZlE;93>GZT8
zKGC0F?u+)pDlQ|pA&1N$*>U|<8MPF}S>xcec$!{|*WQwmNeDc40W|E|?S~R^LQ~40
ziZKWL*JKr{lSDYYnAH_ZClwi<ry+BEn16bzV&B{qyp?U0h&VgFC4zLMRws<(aN*-A
z2rPB_(&*Y;jrwM`SzOyU7<*7?+ARQBdy6NklYUdE8!-u2Eg|{d<w#)B&#dF*qy4$1
zQsAY?IfLpmJTStsAr>zoGMtXhOL&=07IueDgN5S~x~NsEcsP!7@-n|v6Ok1jNjT!C
z9&@mXWfAY5XvIUD5R%EQjR(3!!s{EtQjPxV9k@rRe)P#y(%v;y0KIsh6e)H~WUe}F
zcQ6*i8-v$0RY3D6V@D=m9dBTpU?dBA*`qaGJl3i{V9#K$RvgOu<@WGcN0dz;+%TVm
zAIV{q_?h<zUL1cPpSh%UQKN|@Y@^jzc7ZlyFvU7|Do}lfBn3-(aD=>(i{}_P>iX58
zw9bxS?;P(FJPHEECriF=ozPP^`4tQ!9gS0TdV?!=Bk17FIzR7?k$JuSqAJ!g(R9(7
zV~0_I6P(Nj<s=2eknZQ86cvyPNL;DM4n^0u6vaq3$}bu$&STS3n|Zbe^{Rm(_A)Pt
z?O+jTZF8c{W0e2W;>%SueOY^@dGisP9y+kFWPf?PCsvcGm?@p9s3NU8s!YHsk+z(C
zQ_L!Pd=W%ST%swoz~(^AxTgxuj-=MPZpPW{r6PQq%t&=^Vp!ibVMYq=aGTM3csXPK
z35rp`1fDL9l0(ipWC$hG_`0M|)oan&2}(8#kQI!4Oc|@6TSGmt9f(hh;<n}>E+DVY
z9nV%3OUV`XAl>`p`ukq<J3UO}FtqChkE0P!lc^qjU@~E^4XnrW<f7G+GO)sv*Zl``
zfq|4N#(DU&dEs4?w(TL*thb$RuyOaXsf&%Cn<(<`BlO+w%#$axD2JPPW!*?0J0@A~
z%VxP!QB27XGJ-EA;@G?dR!}rFdZC?_Xy>2knJMzgbDv6dwL^{M_S&A+^_O=l`?rZ$
zi-3*dFp`QP9|?81V0hs>OIbPta6SDFm;6>ZZ&#peVM8;EW0?0hv;nt{gzz9(%qIyJ
zFa5h0!?_{`!i!F=@UKxA+(}Q`m#q|~w%{!%j!4x+wl{yo3ggj%3JC>ECZo6J=c&*5
zZ-!}JB7U(pw%lX5Q^jNXkT&S{&dg8U0DT~fChsXee^YLPqyM(BVN?=j&0X!5p|-!^
z<}UbN!?>0UOjoe?Dp=mZf4Ui*+pUP(-mQ7jXwT&CVE(wGo_wnr#U^#doS+c}fnJx8
zbMv$7$|qt&IJuhyY-`hu#^TP;%9Eu5&%N`_YM)|^j5?}!9Q^Z`tzDK;tmGUzUYzo4
z?u-F$RVIT_MDS7)>?zJ{2#P}+OH!Vlw=j9qZvNr~sHEznkb}n~uP@+)XFXM>wfSvh
zg~?(_2gPt(d=~TWWTc#Dh)kTE2ksIGZ*5}4hniRAy1E&g*cWd7^yO&IxpvvEpGwLk
zZjM>unnl7*KHRS<GdNHfINx-3uxZU;vnp+xiFSB<c;TWe$^f;lJ^Hw^zir+9gZCPR
zo$*)WwIlO2k})z(j5PYmn>!A!1}hS7rJ1&T4t<)#{)s(Z628WO<67+7JuiIc*t!yp
zzAZ+Tx<B~tHF0dx0+{p%*(Qk(Z8PR`uWd;PRiJ8|qPPA#6#j3N>iRDL8Q((j9Q%VR
zV@IdgFitR#BhL}&uwAEDju@fWjDb-!YI-$!P<t0V%+Z!~surX$&Bq&MWY{9bBTm(O
z9;e4&KzxT*0<FT_6X4_&9aC2@_>EhVz87e!QJ=;^`BN#~&Z6R(`<4SQ?pnaH^|T;*
z{Mt2OFO-`dt1M;#x|Tt03SGG{0-WWgnHen(KIW>#fPR<Xydl1Mf1;wFm6(68;_-_8
zL>6!dNr966dx(B%+<J#2NyDrI>fLIQrz7C}vPhRsMEELj=*@8{sm?tw=ai_krIXrP
zyEY^aa%X*|4qKN!KmY!5e^ICS*VwY20{r=1euBS?1M2<gNb!&re=kHS0Jarha*hBh
z^M+7>+O$%a&K{w_tAHL=-6S-#f3L3QS4mjvahD5)6cGOVngQ)LxuCEl^@3H<DiHRv
zJg%(%oeBQ!1PaWC^;tk|zR9uT_dCb23H}Mti<5SM>wDZK6mVRhIvO&mP^CfF8($9v
zW-FlGe5DzxncsT~AB+zMY(fW@T9)q|A^-Xr5$fQMdAW);{{zLqW$-ga7+qKY)))P0
z{6lSEKrz6Gng8x={Pp^XfvE2IpDxvYQ&I8F1#Qg>6ax|JJ<5OjG2eXqzt#o#B_fAG
zo$+%0g@4*4|IjV-GJ+jsbGCx*_xb$$Utf<0>GW;hhi|^a82Hv&`rkOU^`{!fflYe?
za4Go`Qx*B2jCQgaJ<v6jK-@E3=WIrf9R<KdgP7*-MhD)%{`*7!$FP>C)dQm3y*dD}
z>pRPV599$3IuPwg-}$C$%73s5eK>(#e+E$ybU@c}YXWz?+$NPN2>M^?xIH}b|2rH1
zV&%2F(|t7shFts~^X=Xn(r*<K1lXZF$Mg5C+5N=@f4#B(cr|oT&l(gj^&!}uq+l!j
z&co=^+<!81|Bou=?`<{c7w58rJu};DQhDx&52h>9;8MPk4a=Gzo*P{uaHRF-mAdBt
z@TKk6;8M9NTJ5zT-sQ<^;8@Bvuo^V_@k?KTORrp=4w(JnX_gKGVkEuHx<@%be(6o9
z^!sb)I{7;g`kzkEKfc0x6P)~MYC-mAe*98K2u^sB{vqdw55xKuNbeQ2*_!ynZEfmV
zp@aYC$?-EktZoaP)dUaM&6NbnZ&%U(^(6k&$rQ>5bp>MdtxG!g_xH?ipU?|Q>Q-jO
zFZ~|H{Be@~F$;e^H>c?SZ^Qm^2!3za|8K*7W7+<B&;YvrKeJ(%zWijO#8x5Lej#)h
O_|w7Yqf0cbZ~Pb1%Fw$2

literal 0
HcmV?d00001

diff --git a/_images/IT_sample4.png b/_images/IT_sample4.png
new file mode 100644
index 0000000000000000000000000000000000000000..19caded9fef743eb110b8fc738839eea401389f1
GIT binary patch
literal 334618
zcmeFZWmFu&wlF#bf@OdZf@>g1(81jTfk3d}?(RA;I3YlAcb7m2?(PnQ2X}(IJHwlt
zbM8Ioy|><4_vZcmdUey))m6Kych&Yi1S`l%VxSYD0{{RFsgL4{0KlIjVh$=Y{MJkm
zaRjd@vJewfkP;K4P;j&}wXik;06qrCC!#1QY2f=Fd$6Q`O?Vj+H<n6~4p4k=p8W}h
z_zl`i>2H2nhLU({_DtWt`BLuu{Px8P!MNm`p~2_!8H~zrbo+)KfrC%JZ|d(ypPtS$
zli9aDw$nY18eEUOoX7yLO-j|%5Eu}n-VH{J-*~XEuj3HhAov0}vz~nf^lh4iYOJkw
z0OH@>yFz?WM1Ei8Pq~%)_&n)-2w>^oLI4Ehrge{QndEqqOyLVuyhQ;RbMJ9$r%{n@
zgX1h-Rg1(OhdT|v#R_+tKr(6<mr+I{HX#Bau_ZD{A^1W?9>cVC(NlsB0xHA5vAoV^
zl$fJpXAg}$!;TvsBlsmrHaaSE$1Fd%;z%>#<k;}eF++Rk>r2nCbL+sPi6<w1Se>!y
z$AT@8XnYzy+BcCB`bM+siK;IfjMC^BsfKTSD80WZeMcBqf)1I|%CKhc<GdwL#U>Ky
z8F()oAs~-QYZ>x<t~FYTaUdjWG@j9*DnE%KIYTXljOn@no#*TUQuw(-DUBfvEq(pq
zqCabwpmvcZTB3g+|Jy@L!S-crnIQXp(<M6ZSp{$5XN0pFrvmsZ^-12U-y1cj&xKki
zS;x=Z93tvGnH?nR1$3j<Sgy^sB?kMpvQQk|t?NCpChPsylz_7oFO0M35r8klgl+**
zRO&J8tuMg|ZiV7WF)3a8+1xVitKZzF*$Kg1;447b%FMv;lLR3LtT%U47Au;s07p3$
z6%|U8ty!Z0k{J}=x2adjSAeW<8TUh4Zi-Szd3Ni}2IF?Vud)zivt9uG2+N;=+tKO~
zSXN&=6D9ce>{K6N)lWGpShbaa5yeX64IOf72bTiE&kizs012|BC_B&d5&uuCMD+-^
z{y`$hcO9QVz-9l>s9#QAKF@id5=CYV)OZhmiF4-bnXOOvEF|bv_Un&GQrVkhnm-8W
z{phoZeqi1`=R->mDiv88^ZJ2L^mSPz(*TL4i$fiS%g?MsCh0j!$LAAIp69wq+a1qO
zB<e{@1F$+|Paro$p2P_qENj?E6zHg=<C4WMX*wl3qmbSbMpH0EQ|18+KM+N!<{>r&
z_eZ_iz(EemHY6Zm2qUHe;DoOk;2EeIa_Q59P(ZRtV5qDNiz&f5#-|QTBc_T{bxbbI
z)Gs<=Ox+0v*9P+XEQWSQ_Z7zuJS3!MAx^!|SJxdZ>P_k~>eK5b7J0R~OR>1W^!?^M
zO>lGdB*+LI?Z97mUBy}(UH3S7yop45{)Ma~VpZ}Z#W&R3uWX&7orLeP417V-$}~hc
zI>83wLJsd7a2!}F;@#3`<w>Xuqh#RgLZ?V%fl-+DOQi9mySRH%R+fI&Y7R(J^hdd7
zw)Wb$PLlxL0C{7T%IC?l8IlPx+%ea&qtVB)D~fclaYFE^{bI!CJ{IO`X9;9`D(BH_
zGC01z4JmrffSdZ-<F(#9?D(*Fb?FyzXK{mZ5$|-=h!kn)Q|YSW8wMHT8RF&XS5(M8
zXaAi2`L)par?mp_562Ii4fMKG;!|qzWtv%yl9yjB$#JuZ6z7!5T$o%~E-7m=19Oun
zc&a%xSr0)ap<KoJKew{*a%;s)6c4LY!Xn0GcSziky>K4s9x)%Q8YB2V>14I)wurV!
zKKl0pF<3DCB%&p9du_r;Q-HF1!_LD@DZ?p|H6TqqO{Z!+tHX)LADkTiTwONEb`s-Z
zx$I=-UGsSJ*z-fx*XFL4kvltkH~ZgQz&qMwgcINQ-FNeM6ZSI3Nq^<;L3YqCr7uVJ
zTJ(Rgs)qq3cyrTPi_1#Lvsk9e#*Zh+IYFG?xca%xIozy2vKbT65GN4T5@HhX5(W@3
zasT9Y*O%ciB`P4Y<30kx^bIXvSyhhY&22Ok2EOa#7_DBuh$ifTPFhE6l*}<XrSVP*
zEO{2(V!W8`9H*ShnOJY6w%#><SG{O4Cy2v8X_9YldI)V&HD*XqDOG7vDdSw?o8?<+
zoV%N1%+)}vs)g<)T@Lq-({!1Z4U1{1EQl<4E~qYWgYm!uP0~$TSFb&U?^UjtuF6Om
zNs%y{NRD_FNySKJcx!pRT^zwNJe)ije8jx8d}_QaCKE|(Pz>l0vpT+Aq6%|?won6K
z!&W_+%j3}SAk)%e*24K*e%14;>yaZTZu{-Vnff6C0<vJTCPCwjyA177&CvqgKwZx=
zFI_}NCsm^Tm;3Q`;dR~jU$0SW89bifr(v!CK33Y8+$`E7KGa&f?;dSu=_y;bn_U_A
zNq%YoSc@pMC-^pzkYYwE^$}t^V`^j1l8Ix60Dq#o6Io+3qS<2ji9Hjm4HQ7<#KaBY
z3i1QV%}$s1t>v$UcKW3Ir{V||IF3}IE_~a>CXN2p9}tZ~wIT_Tv<Y!1u)$*rw+;Oq
zl8mo}fBt5gw)U+m9yf#Shs7GV5;p_SPtSzj&=}}-_G_;+ENoqkI;E-c9FKTT-hSHB
z)5xl!FzVP1kbcqdB9ht=#|Q_D#(OQ$!K#5uDlPA4o&i~EUHWy*aSzKT7W(3I>MwrA
z*Bdb#+wYFvX(=Hq2`UXK27R?Q2sm-k6w!2<54tG7c-%E2tRM_e{tE4e#!{+$Zu}wm
ziQ7s_wQg2Xn=XOLQK#Bwy!`FQN*3b5jLfANF)`v?xx(ZN6brjjNn4Ynz8==lNT$yk
zl$sL_zYKTpz9T}sA;K(OjK7(zv`X7NhtHOAi*Y#_vg4%~7E}pK>D8y&=6x>xIP*An
zt-Oco23Q7s>bSr*R(^6FdtX|~AAEkWC^N>rUe?*JExnV3Ms|`nO)r$KRKT>b+jcA~
zWF0h46ET`G)O;!*3m*z9!Uu?(xOi>mYY6K!s&7Z|Mww8gm8H+5?`i|BZRkzn@Zwt+
zYV8-|*@7zf4)YEt4lB=A&k{%___>a66r(jAnkJemE2Nt&y{3ZFAJPSzn!%jP`^xxg
zKFdstUS+*)`GwPYwbod+@<$nF9!6K-2k1kCZRT|Ax|iN{=+zLWIEi_gQ)v<Z4=;Nq
zL2Z^2asiIxvG6XTAT6R|B1@v(@E*x_?A>Vmwzn|HZS%45S|g|tMqhZOT6$Ino#mI5
z1TGBj0<Iy=b*)Bu<P2*o7c0TjqRX;-`Q6N)0wrMgNy3TbFYHEU=OYnmIGMzJnU1Yx
zHU<k_FC6e~Yep@O8ws3G&#2oz;q$4Jnz1zL3AyxK4h|GSi-u>?N;zHlz!^8gr*$|w
zayqgz>1B0`_6_6nf=%}_k!GW}sY2W=OY4h`VEzm6b@~qf_0dQZLsMVd^@Dfs;|!J~
zd6<y)BkKK+or<&Hm9+EnmOk81EuXjE@-OXAF5GgQ?RE+2t_%rRKHW^{orIkgTZ`Cb
z9W^YSEBSl-bL&FA#-9`)9G!oh$j1v8=_a@E!4hZm(z_hG9?Pi9`j)NTif+MA{6VfZ
zRBKf7%Vw?Qj|vx`&wu;LL*+>`jfAnnm|<U@6k$^jR{M2Xc9FeoeZAe<Xs$W9A_@Vq
zu?TKe<N#>_z%OF+Cy;ApX2whNGjjbe?LO-R86NM`0OlgYZd%<!DDBwFvZ}capGob4
zbD9Cxb}Tc6%dR={<oG0aeU%GlX-9V9RDM4F##m1}OHbAM0!4^71&{cTYRJBg)6dl4
z+G~i3x|FG`EPx(fMg<@t5CNXUO9=2w2!Z&&%Mu850K|XPKLY@QEC5LVY9j}~|2d-I
z*PlNByd%a00#M*jui%$k*0cX=jSk2{{I4=%4*VTJR7p%q3Vv5Iax^iqbuzbeUM=`C
z0<S=`|ES>v0N_#oxe%lj-=4tX&sZp{JFCme@EO_JFdG=#8JaM=+1USq0}ycIgBNW~
zoDC@4Y^-ga_}m03|IvaEUj9?fLP_zDCeBuZl<Kky6k>LcCKO!EAZ8Gy5IO|~g@B{6
zDW9Ua#J{@3p9Cq*ot^FZSXf+LU720kne80SSXg;^d09YgENpB{@D@x?5L;&hHzr#r
zs(%9cFF4{RPDYLv_Rbb|wiJKh8W`HSI15rz{u$`MkAKe7#LeR0BiTCr>$Bh=$nxh6
z3oA2-<-cLWy9)fN<x{Y5GqKhXx3GcZ3_gYsFFQ!!AMO8NZ~i^vfAm!UcTYA>uK!!t
z|9JKP?5g5q;wWZk10T~_=-=!0uW|qL<-a-#u>AS-|3Jk*@%)ckIHHBn1z7%j(S*<y
zjHO<}*OA0RTwWP|hwrjK2f{e~58XfS@ba@5fc=&`{8<DbB`&J$hH#LM(oFi{wvCoT
zg(B|d&x>WWx6a?tKL$H1d_$$rf+0Oe5)YwN$)n1lY<ck!H;Rl83AF|JJ(k~?2Sxc4
zB=Cu3>tId^>^jo0y?mcOy1l$qWdhzxPIfh2hM1X(7^X+#AYi}zt8+&M3ArCouJ%3g
z8N)M5ioZD{e1U*?Y;h5!zd5t;UqvD4v&e8f*Z8acB1p!79|C_hZ}``?M8W6oVUY8#
z|L;DXk^)g=^RMs@U#*=0_}ph-;K+*N{Ur!O06o?>U(~-jMOfZaAhzo!nWI<!RexX9
zet-hx@77!d$^8uiA>gW0Tq)|qUxGmP9Yp-gH5+?@2H)}cG$~&g|7N?xM`Z&jll>jJ
zz4X)lD1z+U0vdm9{Ff9Pg^+djcWds8dhs4U_k)1oe2UP&1OX>?l)s^K_%g#u-8>Xw
z)oE1Fj7C0V@r483-%_1IT6>(5oCThSar8RbhFcXAvaj&tuKY0nadRvcfxGfwauRo8
zzhu*tfi#zC4btF`T0p>dhKKF>yj7XY+h=#*{*p|bU--?LWiB<J_7an#$-#l*>v=k`
z8+3k+5EmiG{yW0=vwZ_HAEP*5bQ*Pe%VWa|XkPvD0@_FJvlxxKL-n_0!Gx5b4%TFM
z+~|)$DuAPu6s<vs@-HdmF$vCC?;X||_Wo`*Es1{^v)%l=jlb*vf5!hm`2YX6|3oB#
z09x4;W1+{3N$oh(6bl-5{dQmN<DIpmHz&W8<dbPL+&bS%83qCg4e_+?JC4w_$<c!c
zRmr8HToNB{^TO>nYyNjE<ATC3f&ifFGNrP0HpDWDJFNGxj&Hg%mff&C-nAsE|Le!+
zt()JG+$^m(IqU#!&(oe^a@Yw<)y2doS4qIV0-bX0oq}O4EBzp?%~yu9|D9ybLP2xx
z2AnkRm$KvQI7bumgNqyPkD7)z24dNB1>iVHt(nnXfO4;Rniiyk_bd_HS&e$@dCD7g
ze4NAAg$=t=oi3-;)XK|aejha*+eQdp%?@&=cfZN1#0@giyZ+9y6)TsgLB%^0e7sNm
zzoY4~us>@d{BR-$XPZdpgA8R!IPwDH+Tb~p<h6O5x+Q3TG}Z8GhyO5%6Vw2NdLR+>
z=C_^qIdRoaPG|~_98Wf}|Lin~GvvuE)63K}-9`0$T#{TRz+krAU~AdqxTG<kd=u@5
zb-dXux#y16Z61!r^v4mG_4~WP|J?#szt5sV_B9ZJoo7BJmDJADH!Jh{98~l*dZ*4t
z3d3ggf4JT47Nj0~UpH;_65ADjZ4*x82z}bl^s(JaHX~zqSnsAMA-1XhJoVF8h5LR7
zOSLhL!$Of7JSoeao@^NMj=s$(^M`d0%{=$x{vmvxg%dI(E|kDsJ5O67&P?j5z^Prc
z3oRsZiqv)vL{qjEkdn#DFO~nPp0Tru#AgfWPY7WssaB-nQVrjqR(V3Tk9lj`(}sv8
zFjk(U>cbAO=7gOXH0gzz+ZXS+O-Ruum((rKKOV6>*|t4`okn=Km2*5!I#5!1x6`W5
z_^;p;VV&bO*|wY}&lgQuGT0#ZmHMHVcEDqc>KyDQwGH12Os|q0+V$DPZ%sqx_gVkP
zv(@>>Z0UW5n47F=s8u8Hd%0VqA!WqOl`f}b9tQ@xo~Ix1(_p56S<8pHy=0Z%46iE<
zv#|y!+fdRl_sSv^)CF9M3L|~d)dss{se*%1ZoS@!l~18gP|x}Qnt31)AV$Q8@8=0u
z-IF2{GyaE~`;N{ZRo~EE$7z^z%NG$7ge7#NzZnjL1L1~cdR-+$eur~IIVNR{OyN)H
zRWN5t#P(@Le!FGMVKqUONH{a!k7mtJJ<ov`3Q~_aXiuiQ>xR2$Ryh2TN4H=pZR2$r
za64*6TE0i(kM`r9(%v16GrV8ccBI;9R^!2oy@U4*z6a|U8oo#+IZ}=#qJBIAt65&f
zU+f?!&#9?{)i71e#aFGKEh&t&VTO;0Q;;ES9=b6a-Q-<-6aDP{PaC6~a;yf~m!<8Y
z5^~hA&GwtuepRGuJ%l0Udc|jD5we|bdN5mY<e8sr%61NAk0ccGtWUP7onu?sT>pxO
zv*=N*97O()&1_ZLu()T<J}OYv0PhZkQ-d`gTrcu}vk-u<YYw}rYfB!_l~lXsHST}u
z1ZC}0E-h)?`x&SLBzIf&8>~4&GOC|a)f)D^-AvqSy+2}ykKYC#-W9l_?QvK$>^Q_^
zmf^nl4!`w8&}ZaP!z_KOT!rm;#z`y)lXK<4^s(f~ja;m^Ak%x^>uFM6*u|h-A(96w
z{PciZcHW)o)(mHT@vdDswXHVQcch4Pc~lmiD{#flp!KrEuAw{%(;pcFy~Sz!5K6^K
z&S0ONiUC;YDNQ%Anztw@k$%1|qTz8q%&pUU58qGFF!qt1Gqzikcoo$F4j~)0#jf_v
zcSdU-u(V41>m^ry4H3dTYs%aNJ@4C}XS<oBPS#TZ5EregoEPkNo84dfAwk#Of)+7q
zw;$0M;2@D-W#g%t-xn7T*&tzKOUGxY!o7uW3o67zvt=E~Dd*P8wzg$0Swr&N0SeWf
z#zFWl2C}bR%cbgvqe<pm8)>ZwFuDH?Tnl$?`^5c$ha6#kG*QpHVc+bl9qcTzjYCP6
zQ^l=tjMpn@!LF_4>`0FPa6|R3RGSX+-Y0ZGjOPweOtu_*H1UbW#VZB`td^WsHwr(A
z8M%Mw7s}E)o91;jw<Rfj7k%4-<vk|<j&6SGBVaDL6)j!0sC4O4vS@A0Y3Y-%z~xC7
zR_Y%ax~rR?<|sX*`T^ym9+}{_F9PB`kzXsF>o~bmVwP}ch1N)2+v8!aW7sA{Pb!j<
zjU7+ySYFRy_xEI>QQajeAMP>iAw7q!ig*y>Bgfs-;)P_meF)z7oR`eu_l!s<Tf#0{
z#t8vmS%gjtmrUeFy1+?Im2AcP{&1MF3aWTp5(|^fN{2N0#2&)qbDD@EwFUyPo8L<Z
zo6c-mujjn3mxiTi(~XZ>u64P2{5!w;Dcaqf7wJ6?68`$W;<b|<1a;^m-Hua(R(73|
zX<?1RMWk)2CQt!(8tIcH%f0kV8?qVyhpZiX9WB}vzyS+{dd}k?wz{x{IfS2XN7LWv
z-$8%?4TC-SMgcjWu(sc?`eQ9}vmWqX%^D4>sOhMvY25GYJ<VI!OsU$wSWkCdo&$-b
zZDI;uD3~cEwI;p|X7|jCg@z#kJ@Yf%Eej|D$xk&b3%^4HiEYijAAXakzRORdRC}1w
zc9M1KBX!La-uJdE??5(fd%P-YkR;Ln^5yP(qAr_|><?wBSl+FqF2?Gy_kq$_R09fy
zv!w!xIQ?Jn8|MJ!qzCglwIIr#&%JVjm&$5a@5EwhWF1XV#YDmYbCyI@24`&k)%*wL
z{#wBkv_|~vM~}>ky$?+!15X>oK;n0~-6*%+*ogHX@O6HLk?>AfSC31SdIlcXx-b7k
z(6^TBPj7pAJWggK^|uWxg=5AiMUvd>7a<LMzerm2E`jjVI({37p}71#bqayr4DswB
z1H<_sBom*(Z?|2F3b9>H&wF^$qwg`<x+_aoa9DZEW96wd_I~ZsmXF-!@zAs<dnnbf
z3i?a7V=ubA+t}d&+lM>o&?j#0dR1`P4r{hXa(yzEJlq2CaUr-C^Q<*Gcli48*r)CL
z*%g$1G$}WNZ%`r9`{A@t?uxZsTI<`fzWx{N#57b6N)aKrmDr-)^lf|D``+e{$;kWL
z9Vk82NNii*A?s;)Hc?d8x_4$Zjg+3VlAS4XAB2W@5ekC?$G@Z>^tdQ^?=MM1c2GIQ
zZ2I(g9l72nu)Bm5!LyD!Qzy6)CAP?nP||DqaC2(9Snm+%U6qDf?GHG_qEVF15NI$t
zUeNKWCW?ad`ejhEia%+|t9_``MuCzXZ_h4$VQw-|l87}p!ZWaSJ?0a$178w2@W+ou
z8?&UrI}9=BnVt9Vf`lK|urN*!FiOkvB(amF?z-dDyaG(dJ+<OY;1m|pE4=05)sn8E
zY>LOU%{nRJeX0A5VDlz8q@?$YRCAa~^+M0c!r}C7b*iNNz1~~FyT-*M<CN+ud97-g
z1t)AaH_PAixaC^A?T-p;8vOy!;?hNMW)o>J;T*hm$Shf~8vpi>Tp=hdKYJ$O3*-oh
zz+F@K^SjrOurGG3=)u!Ug7fGWOA~b9C-0~Gl~g#crwLnZ8a5M5mt1CC{HuzOTsxlQ
z*W)7G;`z%hSXYI!Ps4oy=V`2%MX3Q1$NF)BzaJjXM_pkV@T8QZg-C(3k2m!OApEZP
zeAjfa$s-n@+fXI9Gnhp|dP1ws0{~F~l|0ai0GPi77ag0xo6db8g^`;ZT)W|+5<!kn
z)(CNUeGSsgqwF#;7AIKi;xFladV(E4L0x8aw!=I@%V?y?0bd!MJtGn9;sT&FG$aJ*
zwoQk%5mdeJ^Ij~1HrDw9BOiKes3O0vF?cKwBl}j&x#U=98|OEZgJ%s&JwZQ)(H*RZ
zhXEWCIB^DXlFLl>t6z|f%9p?kC5N~9=7i@4JvWhTauB1jj45C+F~e#_4%Q|TqPp=)
zv?Q5D>v{U|jhYjw7AGJqT@qw1YMN=^iH6I&k0{5-T;Bf76q>|(^l`qDugH~vzG{=t
z5iJ)$g0ZGv$rhO=Hgesar3@#h`y=o6hug@$rVgReFeW|L988Q|(=-cYr!$Q{qKKuP
zWyn4$yJKwnu&B@63{cBx(-vsh*ZDfNXO}k93Ule0#dP%1m9y!n<x}7KI1_n@1)zVg
z*ypB;{O2I?S?U?yAhPOr(+^k?;UtIDG`x#&nQe>vZoZyC*eQL0Q0BF$!>pDnqQz^q
zB`X$$GxXlV<(=PqMP=Z?%Z=RKg3Q*7zFm|G_ggrJ-MBM*FwK!;3_2y3j`qjo>z9>Y
z=NH`FNd-|HlLppbB<_?A1h)^+W*%IiS5IBTRsICOtOb2)pZR;ZCWQH{I7_-MDLz2U
z<*-y(6Afj}WF6`5O-c1Udc}yKm>2YC+rdu7w5C;qio1mxNGf_DS*CwODA4!N`NO6y
zf0y#K-Tb}ll1t(wBzNC<li(t&rQ<UGf{v2iXsD>Qv+`prQ&;7}9p;C1^857G<q<}D
z_RzTVSqrKscNT;@_kQADBf=r;d_)7kki*)Ak%Ok=R+~uYgXSgMbom9pixZTPJW0L8
zg>;#if61_3@&<=d!Rz|xb|56rvjJM-qx{*QZSJM~sBuhijbvd-^kcR;zF#RP;BBX|
zwQBPV4EOqo%3!}~Y$}O*l~!B0D$oTC7^|3g-s?aR?iTdsrPeKFxdwHZ3MqlI$@eCF
zUAGRxzBl?7Pza)kX?<lkO&~8p9h@e6lh8t`1Q^PiE7{o$0~w2y9*<nxUXZ)?@~hjj
zD}+_N#9PYlA~=TgN@p$ds!9s<nlU&8pKwv3*Y*~PCV+kliXs}J`-7?@huS!!02Uwt
z8cPv5+s@V!-+u9Lp!HcVxv=^;1qrBYlo3%Rjs$qtff&>oRjuoFIkjZh8H7z)Y)^n{
zAadJs>^1thrFS5pc*@f$#Z3(8HJvG}3=pMbP&T9p(z34TS>)nari){%fZgq@#gh`P
z^rM8)H(N=zzSHbQ#etAO{3`LO`D|llxg7lc`{~2BL8@1Xym1JFlF_(cnsCGcD|jnD
z@GRtgH{n|`RB6$1U7E6a9tQ1cr?Gi>jLm%LE`fc6_EyMx0yUD=Ms6Rro{eYwgu^`s
z&3BCr=hGh6OIMd6#vV_5BJ9?e@rwy%Lf;4!ZR-m;-7`%seF}>rSRy<IiktiGoqDNh
zQ?a$I$5agVo%Ic?2<|tWOfPh>l5kZudl_3g71JZ+B7Io$5hxfk$B*)uc1dw4%$xKn
zHb^(a<MWac=KJ6H{Ky0;ufP?$bVi^W{}#XQ7U!AdwK#>WuMpOyB^;B-!ht0h8ivFG
z_orlFxA^xAF2Kdwsbnor9w{ZFh<u(8`R9Ja>W=O-`+z-D<sF>KQhE8;)>j!vp&~^{
zNyT49@y>q!uu^**rxTEg(s5f081{jklXEB6jaN0!Xgn+Kj<k&*@Z6D~<ohO9qY^r}
zG%`UC;{*$dj!we;o%ho2ucMPIHdXWtG~QnK>CcsT1yKqNwoYcwCqs-)qaAg*m@d3m
z-q^!6<|bC93qW-RmcU7SV6<3SpPMP|-Z^;lk2**6?4XA`G*8keW+7c}9(-wW{!1A0
zTP8D<)nGnQrZr}t7|Ag4Puo^et%@I%`WW(jdk4}6>U74ixsHsK@>$WT6tj}k4~BJm
z(g%3S!KA;Wi^t^+)mDojBScZk4^<37)aN7ocAJD&%CWPei-TTr#cwnX>!zJY)e+l&
zdVm$`&PuV`IGKyxX&wwPv@J8X1)F|;GOS~kv(9C$VDMBCGz!OJ(4C!ot8(o&Lj+Jg
za(bWUsot5(2`F}taGNht{+UtZKM}0*OC_T!@T<x-npP4SWZllpTJ843?16~ErdHZ)
zi58BQhN_@h9fE8ztKlXuzmp)LfSdgkdsh&aeDSH;CZU|+W;DG#)kAr2wu(U=t}Bw>
zzDIPOH9W}=oX)5PzHFn<2b+BF9_jokQc<-07QC9ZM|hU0^SSCQOgI?Z^;m6=%@hRz
z$0|+>_Xq@yQ)yb)+iYC|(^mLX!Oj~ZAm>sta!AgRf>%GfyM~yI8+8al6%ymj%(Y_a
zvxFy{qjOU=ubS=@R6LCa2L&8U>0grGoSr5g)8Mp`%$%(h`<qI6*UXGC31{{lg9SZS
zX0^0awBUgz+#~SG%h2g_cZLUHSs`nycQr+0G+<TxY`fVh8J*TPIO{`iDN7X3B15<%
z{#A7mZ^(B?cw>bDFml&A9m4c(;*#HdpO_Z+Zd>Qo^XIuQP{*dLW7Qfd-lBg&{Y5)g
zdY22skL{LIc5(SME@Vs9=2Aw2;B3D`xiSB#{YW=j-Dv2_Tf9~B4Fv@EdG7gJ<RRNb
z{_fmgjq;C$>F<)~+*`+D9Z;+3t?u40Ti1=y%7t8D=-tIG9yy296v;1)uz7FBt2GKa
zG##SyrQ<z5dQ=QSn$EyJ@&&B~Po=sDW0=gzD-T=TTirPF8`bMunLgfiE|?6gtJa<M
z@$C<LU|2^qm1H9m<?XQ0+$ZJ+W0M{{ReQr!bgRX6(~3LSS+bkNwukQqm$jGETBCa4
zg{q1r;~jOK6S!pzWKkJ86e%^pZK*cmBv8dS?(K(L9A?D|t_%p3j({R8_aR$VCL`S9
zvF~btx62wuoq1{i<L@nck9%s_YPQtk>qn$oRDbm8E8!(ej_LNwN*nrx$r>V`S{f6E
zrm@>8j7Lt@Yf7PPqXfPMuAm5}<AYVV%~uRee;ipHe<CXkGw@v(5)>`h==EGAhtOY#
zc4!)IKyBChVZ?qTRDnugN2nyt=CC?5&SdCTP)jU6at|g{6qlDDoYNN-uM4l$wh3MS
zQetS+kUH*s!BYHSd---0-S;LLx0aG^Hi(7Es(u8oT^?>!OsOVc8q+z_YwK|47elGG
zJw5qctgD~ts}kUmH^<C@`$`VLjcuciIt^FjDCEhXxtG`W#^;cSEmgHVk{~-UNoZ`O
zfQ_#Esmk>ti4!JUQAt&ZhG*dihT|!I!lyy2t&z{yNoc=UK*z0*rOB|=$y8B)Gj;|H
ztz{YRN9V2IkwUCPm>P|Hmmaolqs_dhOzaMw;#H%iJ{;5b<HgN;Au6LMMq3)lDN{B)
zm3x9mKBoIqsbx(qm5-Zo;P3s%HkG9w86MZm4sDRLii%QLx}6_nxRDrxdnIkHsO0A?
zTeAIITkkIpE6Fak8Yj~Y@<O-2!Wxgb&U>EpF2ix>xt1UHr_3Ln)=sAq_1KLvJ*Uf4
zRv4~4=Xdit%eAcsK0u|pyz6GmODnge!JoDW8OnML{ep6jn^&4o?^1Ewc;8^)>lm*6
zv<mkg8*$sKFoQl`3+z1HSx=amjvZ8`G;F0$273KSUdp#g2u&}0;MT2$oCQl}RxbFD
z7$<01Xf!<=sD`O_89{LYRYGyDCLmQIQ(Q|D2GMac!%cHlK1|Xg&MHNx76Edhg+1OP
z?Wbd(r=mEV<tiLdn2NqK#1Clt;@PqKcl+J%MMI@+eF=xv_v#&fxb7O4V{(lSHs^o*
zHYL7?nU{TIJ~!XhssmrZAboXR@t%*DNs{wh#e+SrnLaScjF6cJ{vO?YZgI$vbaPiA
z#?S-fW&mHp-Tmc^o@-@beaDxnio?8wl<UsscG){jYlK76BEPC<yhZz~QG(W3Ga(Fo
zR8jHYG&W=0lLbVCwBa_lj&KNv`L*?lJua52ci_8S(G?rG8w+e9>XyPgy9D<i*?u=)
z?+Dp0H{k>iPQmq<MQG^H`;s^(Rn!)^+Q@K>jX%gT^9_+<Bv$^9$F^<)cda~jmVUI4
z=S}8>@FxF;o~pX(<|f>qGB-fb**$n{)EnlYCDC>@Zxsk$!qc*jPL@PdvQ8-JjRkc=
z`PbLgLat|cDh-`yuiuu9k=NkNFB-+!3`VbzrE(aka!MXmeBrqbbJ{3Rf4KvGf!fx=
zdmCIy$_GX|dAvSxcOy%1;$Q3OK@${7(1M2zbc9A5x63Y--EWSa#PM<%M08`02VB8l
zVuqZoR|e6S)+}pbOCTT3Tun8+IZcwep^$o>2|^9j1#lB3BU;(YAz*GqI%<QQ<nrBZ
z5R+SS%P7=qkjLc>ge_F+b3z8jF;$hAB%wb4QqEye&ZQ(lpz5dv)@KQI6})uZNARHN
z{9Ld;vmQsNQgkEFV@Y$H(I(npTsocBtpGnll--yg<QmS+c#=1Ls@6H&2QGek!o0_g
zX6169BA#|mCQ37)IWk$aXP7#SK0JG36X|zBJc<!;uNdUXe5&HdB8K2Q#Sw_GjM{u*
z9%7@^9`)X{`PbEwvwmN6EK_(l5H2xJ^9AJ=Qm><lf2j?t;uyFXyR|2PF<kI$aQ2+f
z&<^E|&G_pKnN!G2One#z0anHg2alHfJEIodfy_%}tGa~Agi5-XuZN<rwl_>v>nQcA
zK4CxN%?}pa%fJkwopqV;P|y0jl!b7}L&N}^>WUv%c%tlq?ibITr@*u!%bBliloB3U
zsFKUUOCtxt?i>t9CdQ`8j%4{(pXbQCutxLDNqg@r_6*4r*!psA>1l@8*0z@vra8`D
zKsrKN;jRwnDI?a2T$yD|L?zX2F`kf%=S_tr#AkHwx*URrLx+@WW(6<7GA~@htEtAH
zzLP%c+*vMXvuDdjfj>BRy?ODlkDP9vd`CO{_W0QTxyT)wwQ$7n&}@Z~jwUPSpvzu?
z>En9yVM?=s&E_rJA(VhE7L-sZ3CRS5EPNwK;7bU)Gr4#^*Ws&ke?a*`Ky$ORnN$pm
zz&%4!Z*peCwqJC680|tDQRMYoI>38LD#n%%EWO}Au0s#trjq;WOKQB3n6<7Tek0-9
zs;sl7trV*?HloccR|tmK9%0$TWtor5g2Nh1=a)dnl{|k^`W8dNnz8oUqNj$z&b|OD
zT!GF3em64jo3&D>cKz{DADB}W+}yogXaG_e5Mg$%N|^)t;fYrrSLxaOH>>_u3FEzE
z$8et+D!G%Euo9trqDhwW+eo)3P2QXvvD^9n3YX*+%9L$_9fQ@%Xfa6Q_Y>QTh&IXZ
zA8DzEKEDeHF?y2uht5R6q)_@fX<fm#QuZY&!!h0HuIS4!ok9XJe5#`xwc4;#A;dBu
zG%#Y2kh<K}x+?5*p=&>o$Oodc?D;A*7RxLFfda9Bqf{*D44cqeEeXiNukrQN&$Yss
z8nzjKE(A>HsT~5GeBajN@;J7v)ABwitvWG$k+Qa5V#3V(gnoW&mYHgipWf@jOcUm_
z1Z^AhH0{%=b&_zpT=(c>a!qxTYZ()`8^b~V-IQ#WK1cY=>*NS!L?5{MaNq=FlYDv{
zqi4CAh3K&#&EeblKPa8A?gR}#dogOZmEl_Y)9i-T{Av!0RzPlnTcJmI_??_0tqnGJ
z49%0LokNHog>GLRr|`3|{?<#^9}#{ubh@_5)p2XRS^soCAW7a-fAAQzRAR8qu{F*b
znd*P_Smkk@izUoOn*q+#0QZ!Ir|XtWczQgZ0s5Y|(P|~T|MpVL_gA<TG`$ep%A`#*
z(9Jx_CrwLwRYdwV0J2h<bd)-ooLz*;TDRm}z<aavnD!>!hZXO|1OLsnWD3d=q~F6v
z$n6Qy>UCqy9HfqveugMr2Ig7Gd!D(Snp)O2+<bTD#t-83Pqy4Ev|gl5mABDks)4SX
z7DcC~aJE&s5Bq|lo||1aft)V|PwIkF9E|kr15=4@8Z+iLxrI;bh{>9U0a0A|fzV&W
z1kky7jU~JoL=?>85EMjAnK1p6Hwj{*JxBz|5PMIj1)DmTLm}6-rzfL@SDnF(`H$uB
zG>%lbi-FFwJ6)+jp=~1L!|f`H$gaMU<atK=uBwsx#li|~x<=Llankg~!Y8+|&v`4)
z5<vQ|Md}weLDTx3*oSJb^>w4)r{bE*{K`J;xP-^d3TO!OsBx5cLKQOEqlU5sW(-8{
z!-A!%qs0(kX5#<YSLlpcJ~_b*z=$Qe;X-V=LD^T;CJXSr#`;bcctv;fjel3l2Q2^v
z5+J20Al8&Y<rO#^Aw=r%{ZOI_#hU=H!v^j;EXC?bnCr6fD9fP2Fn2X_lXSo9uz&bM
z9p*Y`f?stu$S`xBv*pf!?0fM<N9*a|EPJyX%9iwT`3+|E{YN|G@5$D+za#h?!TQ4}
zFLbl4yCS@4ECNX&@K9emi2(@5NOgV!-1~!lF~`$Lg!@l8_}6Di737J>?S+ns`=tXn
zcpNk(l)15g6J`BN1yZbmK7QjL@Bb`rJ_fqk>=`2>vyj&S_r<#g$>sG(frvFClaXuU
z{_{Ly#}{IW)uI?Sv-UCAD@M<VZ!4nDZ@)Bu^hmt!loDENiesg@{pBQ>OeQ3=LA8gv
zfm4YCkxYUAdS|A3KN?%huBFE3yJPnW3PcX34fhPBo%X^7{q`VmlSCp9-;3e`n_#kL
z2-gI%5iObW1iNmNGmVW{Uj6DBPz=x8@zQRQo1>NM#uani;~do687#kmjRj`OeK=Z0
zMcLkPup{`LX>+Nb2Q!~KzHkChxy}@P5bAnG&4)~9F(aM&!(1l~rn<aRafWpE?dqk)
zH*1=eSsjEl><BCze-U_)sf@v@8^7x#jd;FGI_x$_az*a9*WSyG+{D7{%u8duGvcl{
zBy5q55%8Sy12(4t>O`Jf(ffomKjyDsR*6#!HRU6BpO6HnWvoeDvVNZ%c2Gz$<*4Co
zAi@>`3d!mo+0)WRwA3oTw!Pn^A0_*{)I3%gg(S$cXBE9gbvnV%Gsrz0Opp9EkN?4(
ze?memQAXz`(&r)fCuuUK<lM@K$=u?D>)YUFasCi@GRUXYMJ(mq+z$mnEVpsYAnaX&
zttB=#_Wb6(cL@e$zRz#>)`G~vm$)u?f$_iI$h21Sv;2wv^-f)a%))%HoPhYTCUdQb
zkPsW<!aBlL>e3Bzp0QGcb<E$i-AmUl1}_J{RUZ1W$r#nLd;BDm?Q1KBGAfO9IWy7U
z8{Htpvbhi(d&~_$k{gq*hw9mG>G9wa^!nJ85THn_)E5x@%Um+|Nvh7uFi$eM`SAB=
z)orExrwLgbozYxqg}Hlo+`@;hNKet3k1ND|NE-81)mD=cHf7@$kVg;>EJ7G|^Rq8x
zFR)3LR<jmX_(gFwh#MfDD7BC<&{Az<QFH@Cai@1yy(rtED$p4R%g$$3whh&?8}&mg
zFX@2hR5^*_KByk^LpyjC7;D_UNpb1Un$=Zl+v77AoDEuBdRKMV_hASWMR-{<=sM<r
zZ;mOV{*1Xy=(Okvrq+L4*>Z%WcU=(Ou9xUMc>5MX6F+bf%hV;9fIA=_Nh5w|AX$$7
zSwskuKNV34T`a2656}qClPR)q9zvj?vP5?7ns3GD9RjEnH=VwJO%1t-&6G%?1{qr0
z!a<q})x#>J#f<P*f#FGnXhc+D8yOk*qP<vwIBHpqd&W%p3m>EMedD>DY^m=7>+yax
z$QzHoWZ~#%pdz}@4MC-02>Ml?O0{9p@&wBH*!X-vn)kU_#^};I$L7|cP)e&nKOdU?
zB<{@aekubV1?Rw`a7UIQK-7Le+ByLrA0agDnmM7kyG_7x#HUTE4@?MXy}$C<>b0_E
zeG)j8IX??6tDm|{&@p&d1(R4|_20gu*NgBv<2|o4l4JERK3(QbOgBUxy`9=?GFTLT
zG>(}w?IML%$w#vD!wmB@?ALCG_3s6G9dYzB&DEkj`qNx{KbQu@C$i71DCwv@T?;?4
zAGclmCExU>ZjC)guq_;4AT~YIKa{B~EIFOKuA3htqzPQP+P-l(#N|$6q46=WZQEpa
z8i1%Jg_iLTrJa}gq@o=A%vv|^-kI*~G4N7_r9Qb9K@Oq|pwu&&Hl{>8xsb;iZs?Ew
zW6)?gsaXX7gR#?834V&o%$<8Yiz@2ZNqFqo!nXd-)ag)#I|VJ%gm`t?YoP@mX@i+A
zdu<C=-EvVehTaylZip0#z!sZr)g+qlug16bhRW&~-nN>-TsneHtH{GnK7%UMSJz6w
zCVgsp?j`Hj-c3&f-({;*-a!2BZncLFXQ|{W1G8TwV}onNR3F8TnxB^Mw|4vbQa(L-
z#gp16_mJ4%=&BUE39GhtvK)UK<sv>%kB&cQ;)!gqJ1D^wV|vn_u2=v9KEJA@#P=`k
zHwMOoxC{Us&Om`b(Z>i6w-zXGFB9Esh1K#e*#1J$Me*7|O#gd0D;s@)2Rxk_Mp!@k
zcd|QI`-7gXHNf+ef=ZcWiDTT}G?lg9S-M8z?nm7l5TlCm&VO{?u_$mi5<V7gd$d#l
z);{dY`#Ab5z`UVUgZ7=G1sM_i$V9(}HeBaPf3?v0@fgO%aH$tG>+kf46V1C<6N7Y5
zjwiVcdUTQLB6-%o{hY$|(rwRrD(}K%|8@DbgWnX$#)_7IVhj>o%D*1XkA%E_)*MQi
z=;>$Z&y!VdDnYeEQ?g^$;qQrF1SSF~GMgjn{|L&?Jm$7f7Z?gt>x``lA%_x_e<g-Q
zHV;c15v-1zQnkU8N_uS1aEru(aLOluT`MI7fEY8vv_AD`{)%CH=9!A%2N~0|7Y4>`
zc(){ow8p;NioiCFjWnJZwHu*BlX)qi?l{(#$A${B%YdW|^BX<GIA2j511yw*g`UbO
zw{g2(h3FG(@(Dn45*HxPE?DM`Jjm_SN_6Y|Gq5RNgr`Rjt{P9zyP`#X`#@me^IL<e
zVJ;_st_^Hsa>TVfEX4L~vD-a0@85hj#z4)9qOhQ2`;k2E!P-RJn`m9Ml{PVN;QbDa
zuP;LmiyRLqx04%77WZM07IB>bFo8nV0t(9pvYC_d-3e-&A#-d#7X&`?zMduEA18m`
zt2k0f{89cP@^G*`kNgW88APf@BI5$*7YGpe91`RG{Eix!2Dfur%np)tTUvmhM!()A
zJVXkiY@hR7dYS)N&9vK0s)L#LjJl74k|WvB+3$j@xhpSBhUdjwJ#h;|_XA3%_Thrk
z1?3FQ3pr3<W-N>B#r-{S;!S2$uuQkHei|P?nD_LMgP{F&3zCb({8X$4@YUTw2s~iY
z6#Bx3Ze4QWi(}V@+6I6>JnMcI9*Ub>ZW1k{bi|JwjQb-Z;~TNNWPn-U*j>0Tnb<<O
zp6^C2yF3@R|HVLX^O-Ce_})u~I2w#^A@?ZG?-x8u;A=(*pj0cgGtaHuBgF}o!ddSh
zbp2(vfwIBC*8^_IxzyK~llSm#^y5OX5gkPJf`^O~oi08_4evBY8xTZSS#sZ5tdRMB
zf*Y2fbg?5c3<v5Qc3>9dY$p-pLbpU3VoL~CdkklSv33UE-cYzVjrDIbWQIIO4#b=(
z<sZWxD2a`JTJxmUXsqIW3ix<&EQ7c<mUnp<tmZ81?+HKG3Mz!HtJ(Efnq3(S67>oP
z{)e>uT#M;%QQXKQli9p&a;^eQfOKgeLyR%Jb`)HDHaPB<*OE96HRsU??t+*l+~Woo
zzETH{6v#Y%Pn~Zxn13zEl0sIo0UozMC%b)3`ttD81=fX5;qSuz$p{#=L4HYSYYbY{
zuf*l}aSYd3ISz@(-^s59-4R9Z{A6~-BKx}yoKtu%%69@-Uv3aOLw9vPgw$Gv1oG~;
z5FA1)+1Monr5AVh$Di`8v#_pzlkE(QUmF&+mj+y1zn>X#{4aY`eVgfJIYvbUpB7#D
z-QYu2_>Yuwb!w&IDlq8B7H<&&+;LBDSd>xa9~6A}V@hDV%Lw-~8cf|pZOY?8KhyRp
zO;|JZCT08Kc~gwY@OiV@i&me!wQ)?6>1I0=^q<QW=w~TrH-Ssf>d(@a-ZPyOYWXQ=
zSZuzPe<)G>xs|G#t#BJwl4W^>W835?y-Da}7~jr0Ixo#sf&TXC<16+^WdR$<n0O%}
zy*S?akfwpe{gn}|+LL#9O>zlLMQ*YO6rP5-pkSMmx3$^72McWSXjHLN)Vsz~1dK2M
zR_#3@9&krIt}W*pHP)PHO*j489!$eOyc^76xjuF9{-360qE1B?O=~T}rkxt;VV2==
zl`5gyrHrZuBPGKrM`T(Nx;Tor=fh^xKsQTY%>(y-md2~5c-x^-yUI(6h|B&=+t|g+
z8w{1>@TZI2jVL25=j+ji^|+rNLNeFF{Dbi=Omz!yA+Q!n%ZxP`x>(|;Kp(r36pv+i
zD&6vWs{=eTHRIJQyl~*QA?4%fUD7x`W;{-fCk=C)=om06Th=;FJV|re6^>C8_WJ3F
zA^hYzLwzJ@1yFYpd|37lI*OOM4S&aeEGz_D?zUA*V{ABBA83RBLj%ZeF0Ew-RmQ?m
z`S<$>cyy#;w9c^d#9%a*mbq!S*|_7wmVEJ=p5rhnJ}r(Axfbq-!?8eGn)m5SjNP!X
zJe%iZO^g|hkb=vKPnC7+8O*-<aLwvflV<Mkz%XsY-4C%dw8E*kODBRO|B5t_o_2#T
z(7n3o3AEFSJEXF&LtBke=yB$irprr8@uz=KSNMlYsiuw-5xoU(7JaQ>5bf)xe`ERZ
zMh*<(uJD%%COs)>7=6y)o<1x39qIw2Tobq$6MG!pk>pR>+0(ZAnmpUFNAz?k9@I^5
z@ahgt<1$_)kQ%JA+>QC32FzcE#FbqoJ>6Xjj#d94YKUTkV+DM*4BiIoPi`Vwy;XJ^
z)hkE-60AabpD7#J9a?FKQkEh#gm`0zu}5N6)L!J?M}$CAy;2f0D(K&)VWE4e*g?L5
zc4wv9pgw6=De(c#Mc{hBqEe!WS#N&o;FVRFG7fqXlJtncIkX$|jC$Ihvx1%5Dr2BA
zsvvDrC;)r+gImd2QSYZFFL6T*a~yu2s98V%lzS7FSwGF~Iu#&wb-Kezxr9wH=b&_P
zhv+8GH0nse3=z3<slx<wf(p6CSr=wwWBIu5RfKmICqdzE)q=`<oyHjXJzlpcE>(xa
zLH4#1QQ1w0yFD<VptIPK>i1jujFYm->1u}eH<=9)OHpJ(?cjj4ZlZ_V%y_D^Q6g;q
z-f2+2qo9|e8jADi!-1Kw(_`a_(CrjaPvd>ucW`|r<8Ek7<Uxpt%359T>!Zf>EY;}f
ze)%TSvLeLFb;Sge`O)S|;YXo1qnqeWyId=K-$G7wiwpH757xc)#*8|9D%FhgUYb%_
zKie+Ma#L@@<PVv%M4{9ge7~TH(%RL?$#U|Vzg}DrRcG$hEyUAnKia5GJ1&1lSwM)-
zn*!UDHE7kQ;fU}DcWr*Mk~yA-f=wFefzvx9yH;bv#=J)h@g1+M#ADRE&Qlr#xPo;y
z$1AdJjMNMV&S*q42fg;JF5r^cR54M5N@E5S<GAKtK;HdjLS<#9+eB?nM(MV>@PuiQ
z7enc{W(J0`%FZc^1n?^{Llh7Wzq!VDJzDDxr#(ljy#2E_PPb8X$2coi^@LsiF|s-w
z>=^j3Rr&_<o!flD7>#Ckz<xHXdJWuT^~$!v$ydUI)$AUPWYqH&#->CD%Z|FLSLctF
z-rYaC-ea=vuaB*mZnK`LXe11G&hB{`V;mCR9l;`nzNdHp`iUREZTMYsxta9SKg_6b
zn<eSaIDvswGSAYmMP@?n>&A4(=sX*`gRkfXb``ylc=;@<Y8+M_;TFg73<de#F^ubK
zxoVwD`Tn2rPGz6+bRup2z|!;C#8t?trKE||4RJOzd2K2et_NUK>}0|svJVB38DrOa
zVl;8s{E+1u>aP9W^Vv;-vf|w#NA6+GwjwkTPD^enZMZyL6c9N3omYnz@43<Rie*vB
z7SY}t0(6VJIZ9+-4MdSO*9Qn=E?%6>uKoFf=9hc<8fWM~Y$dvz%Fb?m?Rky;U8q@?
z-@q)g^e)Kwv9=1F{v~{{ewjQx2b2NSpYfR1Gkcl$V@Rv^Mvn1MSe1kE5QxhS3Ns|x
zBuhI)X9?DE>39bfJ4&mY1e9A77F`QT;DGTomfGN%Nn9OtZ_0*J<qbEn$wvf`eg8!J
zR0t`V4u5hkGq{FPGn_L1i*-A90_=|<+cHO9cUv<4KMR)YS1|Y<c7Hsw9SdLW2@Wt;
z4W|{?UV+TS6S<1JQ1_dQ<Yb8muX{M$&yjtH%Brm$=ReR=mw7AN>?Cr%JZ~E!0+9La
zWyT^X0YAbVjvSKH9$X~^-CQdB*I!U`N3;r@HzYAWaGw(Z=2Zw{A={}84Y#l&wyvN|
zJI_;X$8IzqnO4>*?lji#PjYVEfO45u^vBMZm*u}PWomE7O$V;{AIlAN(N}(JBm368
z|EaN!^_#gVC7t-k(B4MQNv>cxf|7gI+XtdUUE~izn53HQ`U6gaHh^yqB+}Zi>%znK
z2(4?1gx(6nGb6ip8l0Lmqlqbu?0;nZ_G;3MWN(BLRUcTGB<es02m|Ic4wEGWg>-=e
z3o3!yON(-dB2DoI@0SN0wF?BJOgJQUyX%KbcQR(<A(gtWGj5o<I$_89D<MPdp#69-
zIN!E&)mzX^O~^5XGK|5H9|%a)w0y_?jrcCSf*yT&%pB=ALJ|z_kuDLKmrIC`thRqI
zl$+U|)#iF(WI?59D~6~B+>d$}P6B(W#y=!j!44rFER%rQv%vY7H?GQq%P036x37{D
z0764GC0P)-O9ZFaASGbbDFPeS=EfbFD2_Nx=Rn$bSMAeNM8^cyjS&wvUlZx~Q2t6b
zXBo@9(RnDYLGl%Q85_^Pd6r4Ih~clE@Xnb7U8ukst%bgC?B>CqSt-nLLX=h;2*-eQ
zDslHuwJoi1$s$K?30_*6<sCqavASL0T1`<_JsSHOr_YG7N)wy9*dv1r=#GBFR1i_b
z7?tf5o`wvT*teh(p-G&bswJ-F^-MA((~`w{8m_D)Z=q^b4=byh@CV}t5)h>16W)UI
zB0p^CXh7mI5!?6&Ui~&~4KlE3k;2*0v0#X0@$g5`A8$pp(QBeK3pdc?3m?m6TT4UJ
z%MN{aFXj`Pfp24W){;>wi6+LF9ITf;q%|@$=-sN*8>#N1-VgX4tnT8lV^?)faQhQD
zWV$xwl<7?qt3`-)<+CFZLImjj7%2d0(&5Xbd`z^p+TR*L3C|dgC0ov9b6mLpy!9;`
z?=p4fUhye*X1wmJmky1^IL*iXQN!Ip4^7(n(mPtac{d%(*q}-p4v(j((CXmB2ufHC
ze}c`7GDP)6k{_OPvR@AVmAjhd3BpbgZuK)dNIT(lgopF7H73<IZK;kL&FU!jIKF0Z
z31ut3&tiXp^Bf&lZJN9aSxoYctndsmJOjU^wy*e3fLZ<_l@Kuh{nkpoGa<RmBC+Nf
z#b{trK{uCzbkLi9?jpt<Ppb>FI+Vf?T81GG_Y9A^MaNIm+nS9D@O%vm88AFF2Bc4S
z9DFwt<(u5rD#Nz#r(psVC^0g#z1xS)$|1H_npj`#jKg)rMeu!P!i3<l7(71%D)3$%
zPQ!PEeL|$X^Gr5Ba~KiZ4^$A&2|Nra0<x#i7?_l1W<$+D7k7_u+J-#?asMyo-a0DE
zukRZjKsptXp}_*_kZzGs5JY7FDQS@IE=g&m8wBYF=?00RySux)&K`gF^So!R^PcCv
z-}Bd5i#7Z)*UVh8_x{HI#3G)6R<cfTm7d(_?e1o$oGTmqgw>r|xLELnux2Db&X)S+
z{y`kk*(<U97zwjk$i8ZgmZsxL^?v-uQ3;y6it&|8`}`k#%C6<*hO<fiKAyBHVqyD#
zEhj2Hg!?5^kNCy-?@a5K#w#eTeUkb2%Ej~q;H$aEc5IAwSKrA9VdmZ81?-T#2l0_q
z^+&$<WhUj(HOX}=RNu^5HpG^R@&jf}a&Ik_x&YYPNko#1YEqITgG@3Ns0wU9YSawn
z=re4{2;y76F2%y%3Dv_(gW`x}DGn86Myk249GVjySVKV@h&~tcflB|E{wTjgA!iq%
zjO+XrXjvuBOg8?aHQDFb{%RX^>rCYtiL7GbfTmYB*lW5pHoOH8>+O-8HK@`<`bPrp
zNYvrlgLr8oHB9Owx`RX+sYqY=m0>Qmo%qcXF+1hoBWaaS{P)=>{7;30>uDr9WzfDZ
z*CJ6HzP?~uI$XxQOZ@5IVEP813SWozf@}$5M&4um{S3$h{G$c%S~{@##m5tPmK)U*
zf<RWXD%<>9QqNyh`}$hi>(`sU&}=U}K@mal=owq~%++6$Oyda;x^i8qwWU0VsZBEm
z2JN<Y6T1FDV;6X7xq2l7YwSlAvlD(lMyJZQ`-c^hj&D}+hMpn;wqhD+Df^{v0Et0#
z%qs7LnwhtrbsA2}C`ZYmEBp6*Rmy!30t!3^y`j}~Asumo`T*0wH)*!<%W<`fH>PXF
zBggxoD7|89ORw+SOe=EaZByN7GQ7LV)4m_8ETOLJ?O~Ltvtbh>LvOl#ZAx3$1SuNu
zKMcCW+eX%{xUylG1R8DTA&2c>H&VBQUR;s=63;#JFXyO9rF<3)l94#Ye<dUAuB|yB
z8G+Ka<^+hjktMdn6?K)kury|!X6gLZUS7|EEo0Sj2&;tSa_1DAjgamY%gpF#PdxFa
zU^1;>EtU96xm&~b0RmJ4L}lo?30V!JKU7>|60PU6y1(1X|GzB^m-cE}3^nsbTaa8{
zC$X3H3gv3y+_lPm6={jF6?g&4AzEgQy-(LkW6%G*esa5BQOxj|0n!VgiOO{JV*p$B
zxFn+xK7*x`sPkZ*8_gZ#RgGEYNB{Q*@}nON<n?|jeogzPeS#4fRGSuRDR{jFSQko9
z^#6pI7#vm#9GK}r{2qiet-Oj#(8pcg0c1MPzJYNJeWy0A_wbfq1Eq9J%{+rVrkPzv
z<KjwMh<2<NwrB)1jbeEx?u6~+ztW5@{fd^A9$L0TDQF1DEh0`>^LgbM{^&xWeGqfd
z^Xz%KcqpL+fQSyOvE1!X-2`v4bfPu(Vgko4J#}V+8WQs^b`-)>bMY-CJnkyi`hE=b
zO@BBeV+)$CMf$=aCP2mQ#%KiLvHnfdNpDo3${l>m-xJKHd~WvpE989H>uQ^NG@{pj
zs9|covKij>%&4$JLN9?0!Wdsj>QXb}#16svwLi92e^dc>e_^<!>>M#oZy%!Pz5TJ0
zMYcX)t|iivQ`fPV7x^`g#OYwimc(QJ3;~&Lj+*WL?v&4iwn%bp^2eRvF6K{Xj=AC3
z5S+Q;D-nuI54jIxy1pC<gx#<Atn?BWT51^*IsYEUb(vPz5#^dsH~Qs={+CgPosO#>
zatS*xUK7^g?K-_PkP(%K>y{Q}?xl_Eth~Bh?VyctGtTT8Amh-&z>q9gFauNfOI<!E
zSzUa6G_BmZT>$-U^$Wpqp*V%7%Fm({pvt)DsB{)f&aHm;ZMw47-0x!?7KIr_Z-+Fo
zkGi|zL>c;sM>Os1KiSbQozunnsx|Tl<kIV|c>)U!O;2&r=9sINAGmc=WOFF7t(H%j
z=2p-4QnBGc2Dbi~=fh{9D+&9KX`T`%yi*$^9FW&`e_Frm;h{Vm<j>#2m=jBOIbZOt
zQJCsq`+z?fMdLD_6ru3>n%d-Coxcx6Zz>;XNphzo^piFQNn7+zRSYX6+VO^%DtKS%
zwO&-1;zC8sk!(q;xi{C)lMsOiXsN3SBlZpj6=V>uP5z=542iU$OPpKE+_oV*4BS%f
zuG_fR9S*4m{H0`@@;^yG>iMi6Mje3~#Vq>Z35?<qd&=CyTxPr*6KhHajY!Q2I%kJH
z-Hn16kv9$eg#nS24u`gkiE??j(%ipojxsdr){fGB&(gA{Pxqn)OIpyvYx;z)jMs?J
z-W|7ejs?zzQ1FkcwRe2y(x3b@_+Q0BoMP#mgZ20I*M^zi45%P#zr@yHIGY;i?PlQ6
z9!V?nj<x@zrPak)8mM9shPk%h5ECOK9D-`>=m}QfG2#_d28R-1;UO~|$X>(Ut<L%W
z0_K#7qwhC&qy`i`o<=5{fW~7tXpalnw!9ta&ZpV9H)`by^FHgN%KoHcVNHL-e=CJb
z&KE&(t8R~@xHlimh(eOrBLBH;(;(1mZoXC>5%Ldvh!j<e+Zytacwi5u+@jI!*0sp!
zyj`Pt_gK)}6X{i;Bswc(dMRFb&kXqb%B)a99M7A*Vq7BcoyS=6dqqg{anJSK__DSF
z3MqLd5_Mo%$%ktZ04f89kc6cR*=kv1B=WBJTLX%_Z^jo-b5vMldaJgRJ1sKF!?ugM
z%1wV&QtcLrv+vIMas>U+BCR`#K$_><%AXjX&nHUz4RZW5fMIq%-9Zn}&PbxaxM_0I
z%LoV87_FpNth3ZJ+UE>IbMN_euZ@0Kxw~W=bB*84cx|47;Q>dbz}I0qx9ru99J#!x
zemLwHM|~_nJL&=FO);3P`xJo>Fjdk#W~}X-=ZrbJ+;?bv$sec}I0ZLDTHs|{hPuQI
zQ;iD+HM?)o^H(d96NZA&+?SOXo8In%El5#mHvosq{;oahEr3$0Zs?0#8SA+QIJ@$k
zG2`2bu-#}*h%s#!teHv(Y`NTYy-1Zez9bqARfY=m<E6{(RTyigWQT3#qq)!gpp>cB
zS`i9lRC@794XZSFVl>?87`tx3E-I`*E*>gDJzH4Y4rSEFId9jB4R#Cj$h18VJ1w2q
zV@KmG-dxnvG`1F&NW!=|3xC<1->bY4AxeQ`egx(5zJ0{I2hN!d&YK4(rJ!bElQLTE
zXeqNJBY`ZU`$$3?X8_LH^Z|Bm`Im=0*6W}lCQfXVi^O#W>%MZBlyK(u*_L!`OCdus
zNe{6ryJBsvy*})^`4tr`uA@idpuqAzQRh3GA;S&Trd`rj7n*w@Augpr<u4)6J(c4>
zL=X@je;Kz%#n#uZFs<q83ePbuN#^l&w<l|>FMZ#i`MFkns6Y}W-jZ~p9<}TL#(8@?
zV(iabkTCQ6^wR3xJ%g!4+ahWE+=k2vt{kTO;%kE(s-&YAP+#^YQ-;kL%f2z{oe^3m
z%GhSyd}nF`G}#4DHQu)Wo_?sG0%+Bv{bqAxwO0U_aY1LXWG7B<&UnQYkvQw><vB)R
z?k~F-e_FRIuBNehaLLj({(TFkVlrZc1JUlBfgO4zZQvnOa6@!LQJW%8qB;GbU6osa
z*JL%9{^1?>JtdC-CSyPkjeA>}HQ#JPT7n~{q}I&O7$1AFq4+9WA2OaZK;fE6vI9t+
zw^q9y5Xnmt*UXA|IQ!2f-oT%OU8jp|{^Im@Ua@u#i^;l>^Chagai^8wb74Tv&)H9F
zeZvoX^QQd(TydczBaAZp@MUv|EE!~#N#!*gT8(TZQ^Y3S@!*9fpjj2BKMES95%U=q
z7y{*`ka~IG{d((KR#)vtjvdxZ?M7U&_G6gWhJLB*+CLFDIFATIa;|68+J7<en1+~C
z6#-a1k5O5nb@n;>iQvxs?Z{f!Ak9anG0T{eyEtUrA1Bighz=yNQ#lOR2$so+cb!~b
z-0FopGiMB_D5EqoNUDXBNCeKho@EODc(H@!FbiQ8xScL=<r(VccplPVYqol`KOwOO
zV(?zsSBR7&Q|&l%=j2v_W`(h9wWfeXNZm68jP;qDfaP5G8UXn}0L6*N2KJs{+=U?3
z*s(kVWR&=W=30m?kBQb~O1yE&o0cYB*s}QSto6{l6Tu2Q11pZ0&c?T+{wj3xhTNl8
zdFSFY$((P}XZalpO5>m?1;*$qt@njtO`ieFh3v4|>=I&}8~Ee-PHjoRt(agt41NNO
z>2#zGIXnR5UGtqoJ)gibQJ98{P4wxE0Yr+*M?ruUB6D_D0$r(!`WvQ|)WS*9PJn!g
z;2LbIyN2CK@UD4F7m#Q>*>Ut4)?5#<T<{`{09iVEtFYt}a`Og_BgP>>{&T7I-apg-
zSTt0OF1-b5pOqq8I*Y0rFlN!Zpl#~KN<Q1XeNV22_AckxuJ>JkJ2Gm&C(LCw<;+v=
z`_n;a=yvloNSdW?c;;0@>pk3t6Nk$S+ne2gro2EGDMisYCL0*XlC>}$7+q3Vu{)Am
zBrfc9UH+razNZGXYDwky7yLCLe!@_Z5?of#UY_wHc3eWAQtRp6RK7d^jvDsE99dS&
z{Fgygx=KHt#404>b7s@E+PDnlJ|>^(qKG_S#3AP?DruIlJI=PNqbO;Zt@vCOEXzTj
z`*npPX~V}c!n`#!_B`Cyq02#eqVKHh{C+MTHt#ZL62=^p<KM`u{Bm1)PnQIuCpZTG
zt+WwFYu4U>(Yo8vYOuw=5OWE?v#Cxb&#+56-rpmNoVvY%N=~9cZ`t(x-reKEJ3jHn
zC0m8vX#b$vL2BCs9bMDPOq-uOIc|fsaW|*cuN#KM>(&f2i%~DY^HKX1{bH&h`>al2
z>|a56IPJK2UBd)WH#;jWmNp-MvG_S#f=nIeM`O3u+#gi%q~8KXMEFU<+b1{e)KZf)
z+k<^ffN{nNcgc$%>NBZa@FCi(4SUf1*eIR_>!k$UEIxf*f2!ZX*>#w%xmnf9qlz4A
z3dX<@P*O4AUamDiA<|<AR<2ucO8B@TWTr@vj?@zvpaGc5oPGPMljGmv@o?%LDvH~i
z)2TjX2JLM>t?&AVtOo*bpY9#D&f(+u%xS+F?hm387E;y@;>>AoHC{EJ(v9HJjGL!I
zF~pJeWrk5=-UM3A8lp7T>IU$%eF0RWm5NiWBciqC3Uo?)>eMG<l3rNcH#lfnpRQ9a
z86>7%gkGq$DP0%>K42auw*)GkQ>=sZHEY?IRixa+A`@6H+8?0=@}t%8n1pjz01s><
z285YphQn~>Z)C<rqPX~jY!Ua*i9f<t!>pWZ!V9KwAr+KWh9y0x@w8BbUqOY}dkx7}
zdD27Sy*%_svA!CYU!O6<aE|vL3O?e5)SjdosY2x%&dp*qD_4U9!ZnTZK8->4zvaqR
z<gU3qnYcN-i>?*An2-_n8|tDGj7>P*3;%(FH^+Yz8^b8j0>gzkxb~sgzSAS;5qlMP
zaW>Q*E#xdP;hISWm@XUP@w*M|pi>BV{*!w<MnNnU)289e?fho(s`fo3v2!xM;dAT|
zW5#^vwUEnepoC)M+0}z{Wf1a#m?oF)XYQH-ZRvv(kn6@@VB{_o(^HyEm1!%mJP$0Q
z5U;_3WPemr^Hum5XW}V76>pwCb%^87W^N5&*coZ>FWw1j0}{V2ntPe@V6ZL3B>EWU
zcFPN+PoJ4)28=MLvf_zJEL11K_HycDHw+_gQpk^x4x{BNJ(Z@zZS4?2*a%l1v01?y
z$@1x2{^!wbJnzVucjg0sgxMNZBI{pP{Y3OgK}9gjqxoKf%GH;`)_kJ~j2wQJ_l%Ey
z*Y$s_?i;T<PD1weg|A|<jiS1Z9-e|){`g*pyQ2@?`ort;(V4FhsHbiq(w!3H&_zZ^
z<468$sjG=?ZU*NM6*}+SjZT@*w1@xXJ7>;vupPdhG!uS<(()YEE-BghXew3q$&D<(
z>zM>fBXos4N;e3P^^?9<Gu7RBJE@?L9K+<m9H7*x-S)9VbOvd6J_XKh@=lflR=xeV
zg4%FqxhX&MNVd<|OX4*|F2W`dI9jfABX6C`>vP$<qsQ!}d9`pX=}cvIrOvOr$G_LD
zyn|h_OYD;zRxK_{5<ya<#Ki$OEJJ0@O&)rbiF<K-wdV`{+G}W+VN<jddc#T8YXOq~
zaJge^!M9MY4w>sWHMJ|XdE0AOy4QXRzTV`Aq;g(+QM@5xh43Oq!qelL9-ViW7ZMHH
zza*0YRCtZB=I%jSzI64ug~+jGUjN8PW;w@m)~K+l*^rWU?O@)vcX^GY5Q;}OWV=~u
z!InF@2wL5S?f^<A#hBLme>L!Wawp6RrC&FAQ&uKve}nEsBRH>{VIUEq?|XagH}kg?
z0ISeIjCr7_v0yt-Ghzo|@aHUtYKU5IKnE#RJ=AjS>X-Hz;d%@b^;<18<?)<$Mfk>(
zHUpNr;}7C&uOyRheh>8x6idd)hgw)SZj$uPQ}vR)SRlqK13Ya$&EKr#SY-(6wvB65
z2pg8q>s0>T%B8Q(O{F-bTJEhfs%EHiYd<Q}TYAg_?bc<ut&AI@T$AlNVp=v~&^p#7
zwi}!V;?D*&6<8@1GCqVTQ43hUw}OEX`NuyEw7xuehILyKU8OInq!J|TnWiM+baxZh
zq(((aWrq`%)qTy;8e%kCJV@nPe|!5lfboNNEeS`+`Cc2JEgVuLHSsvOc-$t(xcE@T
zaYX76?Scs99vtoB#TLoK4B?ATIeL_>;>F#gw-fAbS9IQ`a5~LQ)d#|8Yx+qJdc<<1
zUJj~Z&JW@V7kbzsPboGS&G(Ln<m}j)ndU-`loKmsN=g)XqkWz7^g+g}WpH}jivA;W
zuCn2f`zQ}3oWiYMZH4=+t?o)HsA<?JJjr&?v}mKcb5WXQGf9rDppI*U#4b@`kJY$^
zfWbuP_jKs5aF0?Pnc$nXg-$}k-Q4QFtGk*l;Z8t^vCI84V|_=Ya8{=Qjkv#i!gXug
z47b$e!B~5DrWQtiQsEh1cefwkeWr1<-1*k|ZhvjAs(LAvcsEP+$QdK#PVe%(J6?O5
z-aFp3PX4ev9>02RI>)MTRy$}RSbFrbchl79t5TzY$^Q8?KR#`|X{kK>6?rM|QaxMW
zocc+=qLAw#N!L#GSa0WtE~c<NAUyDC=4|0b*SE6MgOm0=fhNK6%e~7b7j4OB40vkH
z%7dnRPQ#tPPy5CzFP&O1P9<-nx699%O;z*{TfrhLU)5(-d>DE6yxnN8mWPDrNQPMr
zU51TX*(6aR2e8eJF9iUH6U7)BQJNEdmp1f{AAYwwZg7IRT<kV#rvE*xreQUtK9N*9
zc=)K@?smP5y|R0oYqBJWoIEtP%#H1#e);pUTI~;*9HNG_OSEUD(?CXuUth)&w31+^
z+S{HB{5vzK#H#(jcy~5olYK;TmI?i|DT3WOzGY%>sP?jwm4C0~mG)Yim`##YD5bUa
z58!KVaZXm}HNU&v4?a__!&zxJF6~*pL~mu;iJ3;0-gFcm@V1L(_0(Ljt3RTkbz+6I
zy6t)gV@ho}f7hfv*#hg5Qe?%Tr}HepE8#(P*)cohzAg&w^%tm_R;^4^-3}HSMfBA*
z<Wss-eG!*B#9e_rvso7c{avqX_}SfYrZ6we@DaY?aVUpQ>z4&Ux5-o1jM-z2k)894
zJS)-jBV^IQyJ5>s(Dx48cl<?xW%=CdON!3|$y@Qso@0(qGZ||w<kzK0?V+L!Ys)m3
z{xX!>AH>&{o`txopTC1PM<_gRxhC<{J>OwpGfrKn^^w=BOS^-o@G|I!OO}$mxFiRk
zz{Tszj&CD-2*EYPp9;0rZyb8cGdbLI5sfOTARF`)g*3~}5Wyel2(9>!d}4e0T)3FY
z>a)6hMwP^^&LP!wOErqDK#R??T8)kyNmRuB1{Z-1=qO3OtmLlm!Q*#q2~T%tBXz$E
zv+20u2s<ah$+MTjO7nC?>PQct$bE5^YX}jtiVBb?*1wnnshq>(Efv=vZFr-QxLY>I
zE?J&UJQnVW2LVkAcT-eD?2I_XKJ2%=lfuKkw-v%Z>`$G?lA|MQbm6?3)XsL(7QW&7
zWhgC<L%w19<6+<U=$Q6usmgZx!yexa6xvJ91Wcz>)Ghyz=5u+O+t^5J?0Lb3$}sa2
z|4l_u7QcZu>v~Xqomzp6W0~0Lke;N5^YMUWO(zv#<t7_(bNjn7Iw-=_ZBV42XTB%y
zGpo2sb7ns{k25NG6SG|kgc7Uy;-1*J^5a6fiUJ4h8bc86mdJ)VAmwAq&}TjOt)6=7
zZntbm*7V9Sxqb4*)eZL{Wf$0;E}=X*c%2ay8LQjnz&N2%eoLp~+w@XGZF?wVz1X<k
zKa5hmWp^$uS!^aS2{|#=tliwpBDcWUF`{S6t2D2dR^skiPp*s#aUD7z<7Lt{O(y>@
z$)GUGBDyCPLyQ5pZN12@>na^fNeh4olbG~ltE}|;WS{?d;-7#+ns$p>wJdmcfqBDX
z=&?&A$m-*oT91@S=O}xr&*<mvVZ9D7GkWC)Pii;NN1WuA{AAX)UMsW{C{n0xB!{hS
z428kQNBUtBc=ELSvQZ2ldtGKRwvrqj-yK>ofUR<qKW>6Ug7?q}9x{BI+}x&r+*V-x
zW!pWdnBypr=u_-m>D&9IpveqXDGF4ZfsaR$bp!gCsc~k`ssq2+;COFrvBZ<r6i+;M
z7*wlU=_Em|WE7y-AGO))1u3*a`BPSQRi1eq&5`Z`{1OR&rQY^e@-~=l8@*nTnrPp+
zsHB*EZK?OPl5To&?!!yz$TX&9W=5E(i7^egtq9zEhO|~8>44T&BBK}u+J*<VvZoKC
zje=o*(A-+K@|O2jee}_RoF876f7CIi9Ug~Q90qc_T3QrEdSRm{nA@`0TLHx*5+haZ
zB)8iO_|XPo4QDKN^lwUT*b9xSnYY(KOF&aQ@?)E+hNiJ^)gd!2YYB>4n{6JhX3GRt
zr_!&gV)#pKMP2I+eOw_mzObm)p3m}RgIzHDHoSdk&C}wmo-5lBn<IcpfAX2KxJ|*H
ztY?BGyv-C><)DvLl9ZqN!v)1sy3_<IuVuP6RY0QJ&7jh+%m7*bu1Bs@SYw@`jJ7bE
z-NP6P5pI}+!jT^*HK(ma5lu+DKYz)ewfGWUoC^a*bxX}stc<w*E9?mAnz+DX&k9+*
z^R1v@@xwR9B-^|b9-`W0f(lK)Jq6oH6`)ha=8`(>CByti0$O<%DHA}q!f28FBdq-K
zTIkVos%(wikF##c)t&CXzYyQ+A(nz_yN!lBqen{S>Kr=bSdm+{)Fb59$F6Op@(KlN
z&bzv+<z#y=?UaQ96xgM7AY&_b#JBKk`HV#_oI_F7IjHJjGrp%`r?iW)Awt=vC$i_(
zcUQNbcWI1VBCjsUVdEmY$GEglFY2=%TWSAK?CL$4xZ~Krh~%XZ0o^3WYm(9#xpFRB
z8(xkX>mk@Rz8v1voq^S4r+&Q6sM=6>!+b?|f>ocUko}(6(P+F`tVLh_dOe%Y_~Ghu
zX$Y={)tVdsrN(M@dy?&1pMYR*VAUB-=}5=B9&Y1Al4_dVJcDi9)`eeiUYkBj{<~S#
zi)5qj)u}*MY?ks+2dhPu)!&M=l{}2K9mslv`pL0mg%rB3`)NHrZ7QZ;xXy2@BhPpp
zwo_@l4<r+ecWayiNvbai;_;4R5zO4ZDSH`bm!A$Be-yh3quRe3;d21|wwQJhR_^SX
zuz$&UH@mQTpIA>jl|n^g$>F^@(h@jUl%$E%a>iW{`yJl_<Iy%pU+l=}=Re*2a{_u9
z9vpFFw6&UTLMI6hI^EiOX+5Ld4T4{|IqZ8UT#9DcVD(snb$bQq-4aG7H<6#)XQ2XX
z)JN+KXP+kTEtGB?k)Y#n4aF258C|+2%~j0U#v4g{@2LV&gf`L~l1bKC`Ppsvh+>JU
z+Yc*1Le|Z^gLP7*j4l7ZG@$Mq`_A@LZh4hX4U#T%nCMSh+A3^iB>zsdAeQ7uYWtIX
z>aO;pS!qdhY&8=%z^yDGKQ%HnD9+Y)erUXjRvSH8l>(pJs7P=7=|j7++^SelWB%mw
zCnowvWxCs6i44u&-jsX(Xe;=%G8pL!*)ULSwmk`=_{3>DnoN^m`_tdZX{fx^b3$P*
zz0a~4lQsrqgl*p8=i?qqEmviC+i#l+T%`HOWk2UQMpJcAPF8{6FODge&~CHcrI4I#
zwq<uWs!{gNFV0P_x+KV$RraK_I`}hk;-1K(xh-Px&dI~|p5d!HGCrPhBpvObtUv{I
zsL}9xQgPeIHhMGFWUo`6o0mp(dfW8OyS$s?iz=f7rsWcxZKT|lu3kXiebl|@q-&o{
z!j>O9ni4jE5Y}8g^rqn99vv^-JUYfc2&ruqybLJg%oFuo`ilNIP$43g2&U9T4j6-x
z^_WTh!o?^_3OfV^`jJ)z;TqdAhO!L7^#nIbKuE5pRxGyUE!kM+uemm=l^NN6R@>(W
z83|eTQAb{t6C^nWbuudc&W8w*t^TOyf#9d_HbOq9Jsq3#X4)~<!E>>Q>9=c%)Z7s8
zI}Q;DV5DZsw}@oWus6ZA#Jy1iYHy-=zbHz^N%N|^>Q(!xUd?(jGcZ+AJ&`-#3*B39
zc|x}E?#_$dkZ86*NJKqUF)u6&p^WF2Z@Im3oRTj?<;6Ovm~toQn!%h_8@87BV0eT!
zLc?hI!?QxVnqddq8Y@N0e9?e_wP5~TrSDHlaAIvi=b_Shig`jK;+Gmt(*{jmaYin~
z3yyjgnejIo){o4KBvhEyM-6%+rH_TLCrtR>DM#6pi=DRbr~R;hCO0o7c#n$zqHE;O
z==-*f!*$x9?LQMfqeOPIMCJ!@2%0Qt<q3qHynsI{ad(6=kNQpjmMlpMvI#mxcQNCy
zT&OLR5S7(Cxap^9-+qlpi{89d=Doh>lt{hvS_hT=iFJn%%UH`+Li>*z@h;vSRDOq+
z-z0Qf5mw8wh1{5FoEYYgUnc!*-8;u;uyU3u`sOS3(Un#C;j$8T4)M20Ca*V{g3Epx
zECGcF2BDhrfJD{hS;=UznuKyz6V^>$>yhCh+tBvkf`-BaXqkd$=%xVVcNh6C#J0or
ze2Dpk{oCoo1G8l<c+JnloOq={;bNbyf}ZuNR5{&R4_C;JiuJhDcCoL~)s?;$RX|DL
zh@HrtlyGo~Y;v3IrsWOiWx=}Mg(r~-)nv#h^&pKw<~9qw(wM(Ab8~>TURM{v=tbPf
zmLs~)z6Y*>)#JWt+r<X;{+%T<<e~Cq?foD7_cV>+{FnWK1N{;P%m2`kT41m>!nC`^
zmc<a=$?9b$YkHnZ3ERg;8A6S246AGsjs!_1mi!V4+fRuz^%taOe1vc$Rg$=}5?^jj
zR<1LYs^ZL;bFaVd8WmPH{#=KqeOYO%tJ6-#k7tctCR(H^>LxvhlZajY@HqaMWnCrx
zzZptpfT1MQtdT*vawCSv+p2`3I7THY>>)%&2Dvor5WCdvU~RK6TG1q@Hk$q^!yEF<
z$S=fBce&t9tbJA#A!iCb(Ci+`NelOWHQjq?a*pNs4MUj!DrLdS_JtiSH3{pI8EEHQ
zMlxW%CN+Kd8BKG2ePV(f4b|GwJoqg}YS%ozcf16hr`PD^3~oxw^q&|4F|iuP0x~ay
zhjjPOfJ$q;L`bD9kN<;*Z!w7I)fT%N(f!TQqM+nU6*Id{-myq;&?GR)P}H%NJjM3j
z2llgS?T5mVJ)fX)Zp?G^`!>Fs@M>6as-1g#Eib{2G1M{OE@HSvVNs%@T-l_DCIH>n
zeeaJIB#b8hUZB5nRWQP9_CCJR0g!E=JX*f5)MlN?3Ss<E4yE(9V6>%^gr<MBn|N#@
zp^TyxcovF_6A@<T5-T$BhV&<fu6tsWD5(J5D-s@B7FLY6PhP;L!<aPdUwmxyqr%mL
z-wzUk)A_qTjDJb_{`7$)>8$Oa7E&>u*}^sd<?wO#aVt1GBt0SAH?-I`)?#e@PfEB4
z@~ss2B6-vuYr%2=w4{WH*EeAgbF2fhf8d+^GBepZI6O9}Sz_Y6C9Hu1$VV87WihbS
zPfk0}DEdg<&T#q^>XqLP=-p;kowV_HuMppk$!1DV|0dCo8`!oRS67g5CQfkpelZB&
z?tAWHx?8zodh|V#E!mZ?lf-CZ<k0x$Weo@8^)A|Ihp;Xq>>^aqVJ8xEwYijq({4^^
zk`&O6GGcANa^&}552<)IP2}I7c_7WAxBaMAuZpJ2tFY>5=+?H};kc>Ys34OCN5w@K
zWEuH1$`rm9Qxlo0UntAbOl^ah;`cIGt-5)~8OxU+kB;xv=zZffxZZOsurUi6w-hVg
z^=UD3c03%&?7<zI)}pMxlb=$zA9L25A`KT#HeO=SR~{$5&&;i(T<Xu6jzD)C&7KhZ
zm&6!1*bFzh{q~7I!To;2%~Mc4`u?beqMqyg_$(`#s(SCf)yzmmjDFt1Dxinx3ngEB
zb3P#XGR0pS&zYA>w{ek=niJ%Qpg|pKE}Y*wQN;-F+Qg7(x4i6m`7s`sOSx~m>aMU^
zax<y@C8TWX0kj7Bo%BjjFV9YKnYK9+N~#_nZvq$N*HhVns}6M>&(M)!v5t8dxo7K`
zhFhJc`42wgC1ITxC^1lSHdfzMdBKZ!D)>+E>)%Yp*%`JEKG$y~o9Lr{xnncA2j%)r
zS`KU8OBNjvpONN|EvE|)fi5r9l$FY_58AGU@bT`@v6PK%#0(YFGS`N@iV^Y-k*;>3
zxb)O~&}Jv|_8UsK{aC^H>Za%=gG^SlCcz$2;<es<bUE%GrxjE%wY5lvZdl-LyytOx
z&<^jHW7!#Wlt98aR?XjA3O)1WVbyb=mMpM|JJ8DpUP(5z7R+xm-{-~|8YGNm{c+M}
za;tVoWFj8;sAH<E@4c;dp2+PhUL3_sDD<gUTXs4xTJXN<p8?t<387mDGl<ljPi^4i
z`*-J03R(9<4lz(#$HTGSFONr#sj4LJ6L6!MerLfG<BRYwqNcLfhFzkYr^(4&SA{UN
z3_1{OP~C_>hZEozhPD|UGNRE%a8LeFK1A|-T$5uk6#Bm2a4J(RP+;NaDU~(>oa(~!
zn5Mf!MhClXnW3QaC&zK;4^99q?|5_h2&(Xaax!|^aGC7pVMKp=ShQqn?a%zaMw#`{
zkcCz$oO=UR_G5_H`HZk96o_m0J;h%P<v&)i7<Pk7FJ_Ttt(mt*@+`rNmT#aX!ah`h
z8zUzCrx6{?dQ%>by5cFkNi+Fk7#r||s-QL&Q4Ku>f4T{k=bE!3%AZ)aoMo3L1=s}f
zg;#$omwzJ3?=5O!{WDd}S5)vdiP5rw*nVQ(tqo(9N?8JjVd5!m+vnL|Tkj@oy{IUP
z>{4rw#&`y~UYmX(J`gCOs`}ndE5||NbNbLxLsf}t!1l{etjigB*j`h*_2f%0TrnoI
zJa5&e_iV;PJbjO^8R98-e%wH#wL(3#J>2!splmB@@(v>=jD}+$?bsRvr^WaD@Q56w
zH9TzTjSb1l0qdEhbZ`EuEYR@%>N@sdNE+oglpE1W{Rp->Thoedmgbs&;-0l?S4TJd
zih$2{$b#>hd>P)-4b-L>Uvffwz?mpqH<o8Se=t#&rVD>lHSj#UkE{y1A^+UWSCb+4
z=*5&9X3Jt<K)KC;RSFOdpY^IQqS&FP>fMJj2*TkztTMX23zBM5SJ4)Hd;Ow6sa|t;
zJ*{x;9ZM6pX*r+5<r`o>krU*~S8Z3q6GT3I;9G!~$ao|CyqOKJ*Iwx6ej8Rl`jvNb
z$@SSrSf*O|irj>eSDF+*hfnv@ZQV`z*-SG%J8sJ*)K=6PViU1PxFYyHF`SNU$AXoP
z)w$r3qVg<|aBhhrYdnbgc2Mu}X+iwLND?~*@Sag&MM?yJ^BKg@9PeQ35L*lK5JaQd
zXk@p4O-s1)$e%DTLtcO_EA*FfyW2fhlm?u*evA`;DLxPA;{^hGoR^$d7$|67al8@y
zpIo762QGX|)<T%HbFD=`&Id@I;hn3Pg+`ju7-zkd!XNY|NyA`9Yk~S{{kiX|U!`~8
zi+#&!Nrzr~{#fLVWOFOkf=)mAAC$JBR;C^Mc7~3@v+F$pho^Oi!Dngf#e`_?;_yc@
zoteesO}<T=w$V)WV)ZFBKKGe*#~8Hfh?mviPw|$7Ve{!0udB75c2F~5CD^XTv+D*H
zr4bzKM!bB5_qddG@Ygh9bZvo@(03+JmkSWAbYN_rDc!sFO?XfO^<@HwNcO-y<53R)
zQ;F}as@>KWoa=yu3Y;Bc7k=WyJgM)TZxlG{QM0*fITN;KnD!>tr_kB@;h|Aj<h?M-
zz!Fpu;;gImN?&`|tcR#xOXza{i6WIY<5_ojU##7%>samC?iJcngU4v$M1Irws{rQy
zmlOxbKxx93;bNgZT~3&T123VeTcAsXqbrXi9>^GVxvm+B?Xwc>#*Z2l#m4KFE9byO
zyB%+XFbR8x<N|FZ^f`K8dU8OX_9kfiou0;P<hS{;EgjEY=iJI&i3zBapZ!p)+c#;~
zqMuSPo_ZQzWkS9hYgXW6^ZP89E}&x0x!tGqASLmx&6ulBA7^k%<4h8VO~c_p;b@i2
ztg4Jg5N+yMRp-0+R)%;U5S*9~VG#h|=(4ZO-LG4b$hABIQ<Xsxfs<KcDGzkW&kmye
zymgpNEEA3AMMJxRh?VMh&FVMKm_{$4#fY_(z^=Vf*`X6XDbG!;HW?!%hWh~KjlZlg
z`^X@<pRzp#$F;+Ca@!eYLd>ZmF}Um*d8=oW!sKLGX|}Iah+eP<2hO?X0`B>D{1bB3
zINtgl+);y}77rdP*_w8Ib}PY`<ep@6&)m4IjtL(X`S1mio6w;(7;p?o?uRqYizfR!
z@eHEt+LiKE4^&*&vAW)pn)nvYK{PI3r#@jmxfD8<Ue_SNziwxNKXfZ&*{nm?Pcr-S
ztEGcK6~M0c66KPiwYAC0l9xKm8su}2@oB<=PUBCW^mj6o4r^T`CPtGQhKdoolgU3D
z4Dq#fWJoW$ew_u>h6G|PE$vx6?F+Sw<2@9w14@8$=?y(^gS^O4$aKiq!_A(U@tCbT
z@(G4UE|b79O>9{zUMTYD`1S3j);_L}Xz%~kMc|V9zG298B02h5iQupa%@Y?j;#uq^
zA~-knLWqIfzhrslUu$@jC(T=HI8%u~yIiaY)oZokw`!lunS05iS!Wz7jFT`L<YJrV
z?@>@oxNFX0CI`6e;(j5$kiL0!MW+`1Ba^PBXA)jJ26Bs!yQ@rVtz;zc$VO_c>pE>F
zoz{KouvZd6|33YhCE<l=q7P$=tAy)3w+o7yxs)sZ7P1ocM=9cFamsf5LsH#)i7(o1
ztJyaOm?x;B1ScWp#>-k=h+2hAxfxeT(?J67V?*hW@7&gAtrgvMj~QF4N8I@HoT{QS
z_uRXG23`s=u?Ls<B}WWdld*NXIWuHzw>PjCfQ5SK!~mum0PimOXQ7ttNSC5`1OR(2
z@~5>@_5M92QsgB%L1*Z^#qJ}LWgqS$Ovoa8ri8k0l_U@ppT&l~5b>$%4r7?4UIM|)
z6GFg}LAkEbqSVfmRIXCvGF6v*?TR;rTxOL`Jal3-dL|XlUafh(4Mc|qE@=ba<c@uj
ztUBa-c!{wtY7j6WIUnMRd4tl{Xqe1t<uJwBbIstC%ucwPFR{}-Zle_DI1F~3iObPu
zF@`FzNlH`9D9btHHwyfLG-`>DJy?*Fz6aO&BAg$b@bSDq-CMk02}tI-<taQH>EZ0?
z!;3ERLf=Ks2s_vzuQh{GOI2Nv0qk|F;TJ!!6d>?3gp3n^A3V@rrS>GnVZM69<ib|q
zAC6)D4S}tebyk`C$G>Tv|DMu{D`L-?0phFDdy)vHe-1uB^VWT5Erz;nGcxbP3)+8q
z;1pv%G}$V#N2~e4<`AerYB%R{HZB_O1`aoQnqsR*w!n2z4TO~!^G)p#_Exr<Kk1}g
zJRuC}MIRFFj3t|LNP*gHGNukdvDWy-mRp`6Y^a3DX6}3FA>09C`Y+a!&0J4QldDX|
zB?F-Y=bAG_NcRoi0>7wt_c#7=3i5b<Pszd)lQfTByzVC^={xt0atuEaL*zVx7A$XL
zl6JUT-eJJCrj0c3zkC$@-lTx$_|h@L@gT4E5WWf|w#}p&e*VF!xqg6jze7%Br~A5F
zt3h63w+%1g+DI6jd3kbi{XdRs`*)bvmtzfFZZ}&Rx%NfBX;c<EM9Q(GNAO7?u!Vm7
z`yJxo>@=3NY_|at@WXGl31G*uT0Z{=BXp+&sF&kctZa{&ZO8!gCn99KS%*xOU*9Lz
z=~C^+HPI`?(N$3XYB&dRjvD+S6MXXbKKJ8ePCIVpn?BQPI3Xl_D<^zojc_=LH*>q1
z(rdxxehGB0U4@^f(E$gmvVg|(gQXex8)Ft9vD>1^2~T{X_43GB@MwAfZhX3+uS?_a
zeeiI7O|7SLR@_c$wIn?(j{rdl|IEtr7ull$Pp&UhL8fZzs8}B35Mf5AuH~ry*Fh?U
zzynwv4B~#C21qG|Jvp{N4^11L0pmWmx`0Z9@O)Z^J(-ru%|%on;bK7;qt}R|_m)GU
ztM6<Don(5#X|JxQv`<Lo`*Re*YnlLbTv8obfrHl^EjS#XcDG)Gd%}sFCK<vdNp9ga
z<>SLm_mKw&VmD>NUAkMnRXun4U*paY85&0Aay-yRwrHMw%MpC312o)U0+n0ut9-=h
zdraVoFngF;E=0H>EY37~p!7H&_2`ePm<g5w-Qw66nF%G9a<qs;!QcA`k?wNlot0>H
z$4;>4CZM;Ooyko4v=01SzXbAE{qp;!0qX4$HB7D_LTv6M2C}>s9UbD6{fd|Z*ZVD|
zTcJP+sJ9m2CO8Mkxr}2b^x{pr5fZ*pW`zzZ$r?qgyipK0OYji~OdM8}ZHDVk(Jw%`
zd=*%Qz`Z;cje**U6`FJ`{`yZac-4@Bh&+4u>oYX61y5|2+M-y%Xl#+@BmUQj$MOIg
ziLD4pe$x*UD_Z94ch^VpO`hoJiq`aq3tGWo#l)%dygmpvX%FN4TH*MZ&;PHvSu1r)
zxhanSq1Zh~e5BtTPD1%l?VN;g9og>z541Ozmtf`qM>IlYMO6!L-ouV3Gq#UUehRdc
zTg3pW&N!g5Up4!rn)zRoA94Nv$sYj$7(o!R%zqBvq1gvW#CS<chxF_<Y{nvYW!H<9
zc%jVyV+*xyR|8uT*_{_z3IEAS{O^vb3q&|-DSvxW`5#@!|MM##U?|X2{eSTq1W)7t
zcXor-;f+fn!lpl-vH>u;j2+thRInQUhn{}x9}(+y=8unI|Lxzl{suYTtiI|37&jrn
z;RE4K>(yq$V5@uebjqyI%LsTTv4b5$mz4_H|J*&#|2`N?Y^ILV|I5$rD~1e5&&yS?
z_0AejD_2l8){DK=9>TrU7Q_K8RrL6UKaTGqfGYr1&9(}LFdI-^$#FoKRHTS{VqIkt
z|L3ky^?%JR@BOcv;`e%9W$rTS0(emM09h@f&i2t-UN9|tT+U|ndx75msQUW;N5bba
ztcZ1}Y9k{aM|0&2oXNn#`*o}Tbky~q0cXjgy<gc{PycJi^^4u>6~1#mx*M;-H>dTl
z3t4)CjD_LnM=dVi>W}(qn?k-=@smj5$>-vQ6Z&zdKeI&)r{7Sqzo10qj=wJ&GEOm<
zN5W3o<!`9kYwgjq`jKGV+{ruVKq*x+!G1I$IFUGU*UP9>)JP7b_jP-kKqOmq9JvDi
zBwf`ruefbNoR1a57mR>3zFLT(Lb^vF0{Nd_Iv$HazC40Hf8q{7#-Ii-k&pWeHG~n~
zRbr4YzqZ*9zyIrvzwiEUU!Fbz-@zW7hx+%2JW4PP>?huYKaYL!F(byT;bycup1Y5+
z?s(8tYJ{nLyg9FCw_5M}vnZ*qkX5Ix{E81lwzCfU9}Cq5xlJ~Qv-Kv+44-2l4;85K
z4`qL(Ed>(8@o2cD>8oS$@@W#WRaR@&hPinq{u`tD@v#C9B^T>yA%UX@H)nJ4u>8+e
zHpb->dU`<rSXM>oT63}W`|44ju-v=PM)|)3Z(kUy0Z>01hh8@;f|Y^T4ehzyDol>!
zzFw`EbLB9rJ@ENEVzwtrdw{O1;pcK=LjV}(RP+H<UoHWsZKsp*%tl^^uT#K%r?gbn
z;tx@e?Uc!yv+iu2(^=pO=MB(%m5Z+jz=gOOySbT*441H#7Am)E&HC#@=_9}zjN2H>
ztg&SLT-FgOcJ)&-gfc_9RF6cuFOIW|cOzh{pr&`H1;h*e?o7;h)9P%)sdAHTJ6Ej-
zQAcZic9ir1r|saR_4({}w@-ISjq)_tk@}5}{Qr-i5ymw@QmrEJkWF73ZZC5UC(E+s
z8)K0N9#)nEHdHkY3`~o=vg<%f!N~-W3+XL1d#?idOg1qe95$eDYlWnPb?;MS3Npm)
z?#8a7p2BW=0C>0DNueGfZo%rfb_geMk%^TLRqRU=>{*RBGX<zgO74WM57JC4-}cJz
zTpBJnbIDdT3+qn(alH@%oQU|obeNU$%0zn04aE6wu;%T;Gv|d-!(N2$1MN(6B$tI(
zF5vqmkj$`*Ki$T?G+eBQQOCn4ej6!T0R%{V5Z5GKt<t0uiMrd($?AxrhRs%}_YY)D
z-RdSZvPv7nOE655%kzZj&JO^(!|G_g^c&{y9bkep)bTVqUDFd}F7m|P6It?4h`Y0I
zz@!p}y37K^9%4c2#$w0|E$3^HQ#eJ$cvOAUlFsLbN9Xt@J2N#EcQFTXUMt<P96ey9
ztO6!jTM${zHA7>u!_lhg-B(H<NbkHp@;`Ize>;m1lMjRXy$KRBB==qWw||a_&=6^|
z$xN{H-!haF%>8XR)Q0-`h0;4wRRj=89x@Ha_MYS75$~ryODs!(%sW$SoC4MztKY_k
z3Fo<{jF%kaKkg?vMv?)<fdjCgdRM4kr|O=Fy!B9r%lX{=>;{0azL0VmuYarF94p*I
z?t-b=C=k16&SDTf%}U`y?Ri0M45Go|gj8LQZv*i1YjzR3SPt(IMW@gy3kH3{>e_8E
zarAyg!Yl<B1@{|KGR)@cSG@>jrI-qx@yL1lBdvg$3fJa^Irw$EDPTuK)dRBN9sqw`
zbwsWOc3N6a1fS+pk?%(;qE2L22%mh(ntezlldX<Q<*<DJET{HG2Xox;)&hXR>pWb}
zI0lA#mF#pKACSy1p00dc#hU$8q#f{xNfq+b(d3i+0sv|10)Psm6=3LQ=F8Tx&||#6
zdWV##;c^mLwcm_Aw0b}_?70ZMhSr5M>Yi(6Hx7fWcM6q#7yM&Z`Y6aoPw;-_{g+G`
z1tA9pf+Cx(QNxRQlz;o&+(j0U??E1!wmtag!7;*+P+eb0$+YD9gn!FVaz1=x)rf>%
zIeT*oaO(SGyEowHj1V<PT4Nm`7L1{e8uHAUe}K&GRLq%zC&<NJwtZic{Qm7E>JNmP
z^R5DtD^?QrnQi7gH!2x@kzC6<y&P6vSJZH6;<rB!r?VDJ@rAF=xpne=s9_<FSdWGU
zz|yMzm1whPKO^4MQV@j_c#&neT%Bg`9ZLwJ(|?f%FGX^KL_SAZ4rSM%n%i^C*cvNL
z#(p$xo%~QLTSV$aHtV__6>{ltlRKI%^5+#Ygb11m>BXbq)4{396X89X&io>dYNjU*
zzeD(?y>FXX?c@@~>;f7Rt$%H#8r_k0W3ce=^5zG?64I{fxbbTha7j6^9zBJyPR)Md
z`?t6CeGCN2-P?hU`=28@--?Kh(~ry4jb>*Gp@zR;eeYoa$_rHqbr(pA&$)t}@g^q=
zb2c|GD#6z}kY-57iwkzX7?%+{`RX@UnXbNJT7RK8)ZpfpW|G6*Wc;Sdh2GOA`({n}
z&TzO?Q{j!MM<Sn%rRX;VJWUxmHfrmSId4YFUO+j&M#Rtc(Q!wLdAQI^THC7$C&Mj(
z(_3X~7#QB_!n|8XHpq|`hK~f9r>Q?qGrv8O<J)*TR+{E>uQw`(3=-q@%aLH%ybG-F
z1zR_)q%oUEdDM-F2!VQcAF{4sFaAG9X+#Y7C!~fH@d!sr{5>*d8#&qaGJETd2|VHj
zv(h$Lt?k4Pxr|1;PbxE0wF(<<9jz9>)9NkvphIRs#9%IB()Q`UcaoIpUcOg<)oU;#
z^uk<L5ZuQ`gF;7F%DUcIj`owUc3|1nqlkkYy*M{b3*utOS~Z#~&ws8E3xJR)#I>18
zcMm(16fQxTpw?$qZ^tKGh%=_FPhmBppjbr{d2?Jj1_FPTW@Kc$YBnE|J1Lg~u-ID#
zW7@XhGB15z`XnVw9fMXjVZFYq{ALpk0w$oQA7@}0@>6`?<xpc~YcV1t`+3b4p8=FF
zuxIuEbMAt87YlJ$SXoy8xhr3av4D;c$133TS+)$6o6bz0e94@z1!|csS&B8@*u-%F
z`^T&gNm-o$MBZG+t|0E&^=}yZ-7%UV>j;liL~=jb9N_>-s@%J-QwhQC9k3s_3QE%+
zt^jNZLPu$aphQ_8%O#ca>U7S{ZEQ(ZK@8jI{z+D*$eR;J9`pIe;m;C*P)*xOL)tNp
zc|88@&^Yz&_*W<p&QHZY@-Rq>qo}IkuSnGpA6zmnYf<0LRJj%=BYPcGH9hGOT!>8b
zp$9r8>ty{mOt3X~18&F==k_sJq(O!i`PTX0=VHq&;K*B8_diEAY7xCJv;mXe#zPZG
zc9(4o{t7UhDEUQ35A!g^q8{qc0X9!6y$})9GYYrs1BEC_LCjeY(3#x}ESB1`9Z<+?
zsQ*g~fCzF9f@jSJ)~BwhhtnDE<)HfQPXHG>p~t`j*#|+N8-$+xuDd7|(^#9j1s346
zhs!OKpL_~}e;gw5LzP~|@;dHXKKg6`NkLvTtz1AO?ym%ed3-Phf1GZVa@$LT{zkg<
zJ>Yc6KBi`u2(0K7-#$Q0o^NkB%L-+<fms%PthG$`n79w5;Kz5C`~Pl$c+{d|K)U1o
zN&Y!fsTLXQ9vSNwcZ3pfkGj*&%D2{^yq92n27WuS#6`+$tjbJsgfli=9}Xx~_HA;?
zd3fdKm&eH$dM>0x8Zil<SJG%)9d8ic(w_)nAX@=_!x%T<@r5W}EW}PT>H}*B2pYYQ
zAz4ZLlNDxW2vam-fr{1#(yvEju9otYOJ!<ocM{f~MXUaB@(@^~dkT3k$OW%38Y_@M
zip9DOkS6t4Zg~>@6TR^%W2Rj4?qNaF4M=_;D&FIMhZKZm`6=a#5RyBe&+wJbP{zBN
zhCGH<?0=Wyi=^OT6F`)F?S#Z75o<>lqX)2=Z9<)SVZczZRtM6&8C3s7`^E3PD1U`#
z5bNVuz$^%~31|l(UdIEL!6XhDS2qp`d+RocTZVHC5FqJ4XD3L2P?*#M4M|ySiBDgq
z!sUcEzBPd+4$~2E7O6~WEsOLLdPoY%YGg&F3?|`di&PZ5(36f+T#-*ngl`Xl5x(Mh
z{}XlGVFhR+T^6#dj{sR`d0vInoDVce?pUveS$h%wzPwv`rCN$tzz4T#AwU>dGSQ<z
z5UyxPz}7Jq)Qs?l4+vttR(^^HsD4RybdDGTmUa{>k4~ODOk1>9Ed*MjLN=j;Y;;a1
zew84W^52aUaoqo(s-nNcDys!V^wEVxxq-ifhhADlEHlTK6ZON%_QcRjv)O4P5wSa0
zU}?p?E?al`!eA4e4W}f9O@uN4ln){|BOF{y+bNwY8ygUADzj_lvyMh%X-VG}hvVaY
z;$1*|?gq`SMdb2f6cwHeYMzsnOv=fJP=p#m^@5+<`2zYF|2&vw^FF6_fSB9!Jl^<$
z`gGkLcqTzUOt8Tc!1qew1<rBP`gh4bO((j?L)W>E)Wht+9X1abBWkP0el9auXyC#o
zW`0TKdY+2yf22DBl8X;>TRn)((dQE)w+#gUfJMtkk4UV*kM6UeT3=TMRixTlL8`CI
zhg;6qg{vnUU`1u$ES3vi&<tOK|D@(VxvUtFAd3Z+O1x|$Uo05Vg9gHMBD)~v<^+jI
z(SYk<wb0$Q)fuUJ;aouiXTpAes+byY9H-H0gJlb6mx%8BqGhl!Bwfmy{V0kUYC@;l
z{zm^>7!<YWijVRUHp5&YLYG-vEANwkYdn`tFyOOZ|I&OhY$p^%#@&0oIns48nyc7n
zK($dh?DXK@CKB$AbKq&-nl$nMGhM!xxjTA%3>^Wb*}}KBhpZX1WA!(uD#Yd)A|kj`
zPQWiCZg-}J?4-T|L`y^|C#b+o1vv-(k@$n9_L>Ek-$JNqhH96YnjQ#Aa_-gw<n5nt
zklYjb?QC1tW<gIFQO7dR%Zl24R!1_%`ie-c<jpl-z6jkx5R8vDYCxZ@6}7=I*8Br&
z!Z&0hx<}E&T`7)qpNeKc!nV2ku`%r%Hc{;v09eN%s%c!V@kWD~IZzI>m~g@iecr<Q
zlN@6|rf7i*QO(Q;tMMaiS|GxF^>3m{-4w;7k}=8#lKdl3i~SPX3i`%EWSn`Eo+6z5
zDGEIOIn>XWhK_Y0Csx}}M#feIi9c9U$cR5>?-;q|>c85VLqHMTB~f4Jv^N^O&wytO
z<mgN7P@ULkndDS}^d2I=vxe|CMBVT{N;!L@lylYsJXmMw2&{l-sjaR;VFaI0TQJ3z
zBjzWw5K?-3P@UKvWv{gu;77nid@x5@M}#UrFw>(WxewCN{8#(o|Ftx718dlM1?t{E
z{fw*w<W0kOVn;dXoF^ignkGc)6>$38%A~o6=7<y~13B{{B<qV%vNH%Y%@6HQ!0@Vc
z=Qvr;Y!u9w_AtPTGa5|)uEDr7f)2^CtL_Y^jYl*q5Q!0};M!3Ke}5f;qfQwLTD(Eh
z+)V91gFv}MGDZqD=1_YKW2P(9SLPhnD8|B*eJ&K_tgtpe5+oom!&EKAxykdt4LnqZ
za=+Z2uCh|=v$P!9ufp>Bl#+l<LS&ys{q`g#(`{fDW6pjF7o`1X>3K_s>zUb%Sf}Cd
zcwY3&)tzSB@uM<*b)%nu@8JhF59+aJhwEWQsXrXMhl;v_NI9_z+g280^^+`+5+^I@
zb5{qdBg{@-j)MYt2(;$|Ee*uqDD@@_&jmu+nzPO%&9b$nuH=G2bHY~lQ)tQ$nch;Q
zB&^+l=Q~|UC}>dB9JmfRjhDbvAh5Sh|KEp_UPLHa!$*SreGc^RM3AgMX3E-tdOY}p
zd*rvQhc7eI-D}PJ3>QS|%|T8+d(zoQMf`Ma?VxZ0A|m{*5bN>BLaVC(i@ooRiZa`_
zEf7#azyKm3plv`vQHc@-QIL#8$taSOO3q0^C5WI%&XS?Xp~xa286*@L5s5_-Bo^VV
z&pz7S=iYPsocH66@!ohpdJIRC>f3vNYp*reoO1<FDtUwvQe;SJy!E28o{+3vGIb@5
zQ~Df6IMS^YyW4m+R<KdI#v$LR-A?ogouL}RlL0ra%g>sMo8m|IgYF~m!fj4=$Q-OD
z6JH8Hu2@ZL5L~TJjagRJFgZu_A%&5gS}QH-bG^CHZ?n#;PN{+{Id@jld4(r*4GvPi
z%5)r8JQ~WsOQn#$6ED)THa}P<*sM$tx+N!xSMhw_zlQdb?nz8=Qk(;b$@+OSXV>qC
z?te%&JTVs`@6sN@zh*%0mAJNp56Dm#iGI^W>7B;u>HO`wBKPtI9S>CX7jpG~6r0ot
zs#Z70%JKK$&M6NOIe-nUk6!)9&%U{gjN%GIMi#&YO*2_K6~pPv=#)x-RJ9fj6Ynbx
z{Vo-y<^|DZuX~ZPKNOpTny&lOV20Y|Vt=0up^4W@EpVA>Pw;OX{o892S(RYwQ>ZKI
z^1VJka43-gDhK?0aVt#eH9dYw<Lj~-V0-K7sOn?#f}P+Hme}M12?R1f7(s|SoiC-h
zOU(}!^I~<JvrJ&j{Ttk8>3?g*ZmYtIJrGi-|3`JGs8{1fYzUYiyTj#(8suJ@2-Rfg
z=Nn%sFzIT6?299r*R=bi&CI8B`uxZKc2j@sWC2_op}lVle*vWbGD7|N3kpYJKcwF`
zN&Rhx`j_>+kOk%LyH$hvKmV@34UGq0--LAt&dF2!w^r@fF_vP36K3M7F7Wra=b!P#
zb4Q@$I-8kE^_P$S>$m=kW%c|-3XwxYw>R+5zw2+GQ$&*J0P5mFHnzVnkpKLL<NlxC
zHickRHIjL4u1vk<nacbTa*zn}cYE;TTRiE`ca%KRN2E}1&Z(U{h|&V*idMMxniWwz
zUjNzV$R4>1Ey7m7h1EmnUY|uRfx_<U%492>G(q2i|Ljz8--?4Q#}FVRgGM<?LKOG5
z)nOt778Jeczx`y$<HaQ(xS0FlsA0H4D2ucm6pwd9qw<gZn@jVLC7jKHx6syMO3CIw
z`8m%f2!c6}*BSVJR9XMlUcevxrz+#$+d~2(RyBC#n}+Sbp8Us${<nVcpDMwB`!i}m
zG-I<LeP#V0EeJnyCo+tHC;NZ&86QK*o;l<y(e@w3&i~V!W^fR-4VXtO6rJf0venQr
z&_OYIv+}_;ZPj!~<=xgQF7mu=gfh2u!f0!>0G3EJ$Tpfar77}edQ%bLTglMNk~H_R
z{ek*l6!%Pb!;R<l#EI_f0-ur+bQ6!oNVy|YLQLWpt9Y+(FL`(JnE(@k(=0RB#{`?L
z-M^e$!c(AEL%0IpKvS@3*cuUjF7$3foIzNr;XxOGzhOM&bvvx*FvgaXLBK9ko%Z5y
zywJ6>J|f+4Sz{Qtw^8`U(9~ii<=qketUu!L#9;@h=Ud>FHdz~Sw#x5Tl%}p|#3=iw
zRO);mQ+&mi+G;B>i}Osgu((pX`>~*1*I|sf3o-}nWIE5sC80*G#U7HhqJhWn5Y!FI
z*2|7oa_crdMcPevpYFFM<{P$hSXHcs_QUjY|A8^dpMP0I_Q(Zk)Mw};-Gk1FnnAd2
zt|P!JR40kRxk_OL&<N#HOKz`2jDpu*)d#Wie&5qwTO6%w^6Q6!wGZz1^RX%uQBwdu
zUBxH_NDDcyaY51>h<{=7<1bL34QhAf(mKf<d@;X;Lp*@5HIkPXfDKGPuw}GSIu4bJ
z74GZ9_N@T6n<dLdn@+Zb>$`2uFeeJKTxa07xdPL*qWJ&-Y#!)PPR}#cTtonT0P#ZS
zaeXsD!|?BS7!S~|(!bwI;{Yo67Qng9k^#Y?)25zt0s&MQ@I+E3jf5pAdLFvK^rp$M
zj3qRbZf5M|OYh&)T<khp=olONJgD{eJ`i&0zZUrnIPPPM%2=1>=cO<lx)l;$-EP%c
z4ThL4r#rsVx7f8JadzEUGhKR_?CW_@uSG(Z_i-EW?5kiN5lqWz=<m3@IkiHH`i>I@
zCYlT!oSlZ?D!Hg@ox_Q(_F@3lCy}9)8`x(RZtWq!4b61vTwe2JSw!XvwKHQhmq{?+
zTu!YP%CO39M<zusI?vMY^3PD?`Iz+eIn)vYSe_C0zk>AA0cE&nzG)-64sG|yinRf@
zHGIkQs-3mSegtTsWBC#sbNEhB4UAd6pIx~%n@5ZSl%#)QaxwKzVv9d5Cr4g#N~?|1
zwdXd-eC0gZaE^2OKl)##0E%!Va}pqtBrcE0hS3%B+ToC=PVcX4Y9JR_dPerF(@+<m
zT19nw{7q4N(i;O0LS$PIqSvR9d!uf63!3lWVIq43puiYr?mZa)QlhiRj+y@Lt?k64
z&()rj(Cuqief=Ufo1_G=(rzSiUO0U~Xol{sO;#S#g?^)|(MWu^=PRljK%{nRQF#??
z-Rq2PfKVerA?NTyT%HAXs<H5k#$5aLJsZ*TwD&uEkx;*n2z4j;D6P7jcv}RXx+|l5
zyI2=Uki{z=_=dRU+jQk6Cl8d%o2Bz@-uQLdeY~^;mM3LAk5-eDe5+tx$A-uZ!V|;z
zU9$NVLuJ|58VhqP*}J8NnIumhE^X5JaS-<GiF%TG!pz-|nL2ka?f|a2uu#Zo+kIye
z5PM#rhvhj!bv#soHIII39PM;K$s#CAqmTnYs!X!f=Ax{m6!*1$qe)oH$MZ@04T0SJ
zmgB*Uj9BkEQuCfv4pVu>!L;2hAm%NULF^kvT7TJsk8mH2G;*iztfEr!Xwvez&P3lI
zL<?v?hO{S#@BbP6FQkd6L%VqUn}mHXkK^(K#5Ru&NB$nL)BfRWt@aLD#9}aajZxi~
zrdR2r>(=+HY?(P;Ww|;{3kkK(l41<$4T;F)9;A&lh91J7-G?%2i<)8QKnCjg&mTsG
ziM}U7j>m7rHjF)s6@=zCmP)5&xeARt-wy%n!w|xWjnIM+^e?kTKT>YSS;x^ybzifA
z6^VdSg&*_crZ8Kde8t{occrT{Rtas%`h9*7E6;;8ZSLs%F``lfOm??7En{8baAr%8
zaDS+K2PMr_+pJF*TQh7NHSg$`m}C)g24_HlcmTj2{?aDbUmTmUO={P37~44}gRx&+
zAhd6e8GFK;y8P6ax+^!G{({rt6ocl_1zt&^yY=U|Z<<}Z$~_c#lJApHbauOxX}8Lm
zhq5w-uCcv0t(TwtO3#8~J!Zet&5KK}20`~ae}FPlt(YP<Bk;AZ<neZ`!mLX6I;jz+
zI!0nR8iPoT*c0|Lm8NeQJ-3CA_S*X~9Q1^3>?+b#7j>aCX(=vE4LLk~%->l=skVFM
zVJ*2hrAhjJA!341>jQcWDW+$7t_Kj&xOuECD$*g%)iMwMLL;^fd4fSwr|xCpLNjdJ
zv`(J*aL2x$9!be4i3t|~@9Bg}sx~Jh%j~+~El=U^zeoEsfjGrS;Cx!8_kk4!3D0`&
z-rU2do~{~f=<PnTk?OVMC(0PB+?-EGVKGwNAq)}Mc-(nVS2|?}#QmyC_zcaPb>^5c
z45~k>%^U|}n}vM&Ylhn+xbk8c%$;n|#<N9aXZ*z9(c-h5p6~P9k-TY?MV;I^LSFAI
zx`fqdRowv=ekWTTAmcgT$eSjMonGx@4D4!JKr4MW^{r0rbX1!(SJ2dRHnTD4yXVVb
z>f#pCrYAg6O#aH#La?>3y~1f#ieJFbz?EaEA&Z=c0wC3*>NW9*CcpVHfz^uG_NsxA
z%fl{fGYi#gGIV_@UHb&imPC$Hyb9KkkaboXTsM(jIe-1<&4(h~JlMCteINZ&kll`j
zL}H`W*hj6NtlB9hi5KSS^=_Hpw=jBUUAq^)i!`Rp6;@bExb7@E2Nb>|KfkajQl$cy
zyG_*&bJW8dan_<*O?G=KQ{)V6+^5sD2a0KL>B5!2QgCrrCw<M#Mi?Jt3cTloZor`^
z*j(8R`s&y%+)dXfXRNi$)2HWMO>%KEgn05-$}M$1F8g(J1Y!$ofAvAETuX^@*DbBR
z-JO-Dc8HNwfPxF=i22rKOVD^=I!mr~A<kia<>wessT>3G44g%`P@lgH;mzqG23!cJ
z2G}b6n=^XF2rd&29%JX>fM|PoO<KcvpC2rm8uXvvtmhpR5FM}wSkW#{yblvJ0`F}>
zo-citsnob;(N0{P(!79BQoIYc5co|^*Nr>kswpIc2e3}pYAN<b2NWeF!_&5>J)Ys6
zhK{IwvpaN4+&ug9jrOlSH!c?D*1>*Nj)dj6W48KZ`TL2@U*glutaXdKy}2=M>|81-
z6uP$=(mJ*;25z#IR<ff9imsFLk&v=DTVxO3(cRNU=~<Y1Ki_4TwwJ0g*lN`UuQsnV
zr4(@^NmrOf>c`o3M8mR6mRGhlzht!U$i4e3sV=h)fyv-6aoWyJNXyb!Ft9CV#X5NE
zj)e@RJ-*sy@R;7fzI;Al+y4AmBJT+eC{YE&0%d-M;7)Q%At=TwbqQUBpZpB?O|!bp
zAa#M}T7n;Y(_P<QgR<PV(m=&_g*H?*1m^3to~UZY#e2Okkt{;@bNNbJ^F{^Maj`>;
zvh|}(5i@Jgg(AR8{5SogAJ!1lJows5acovqz+#}Fn&LEls)PvfDPbTd)LCZV55<a+
z?ea5kVT?Ny=(H&WUfbyT2FguTWfP;0EAB#m!e0V&wDt)-&-$!6ER$cG(>=WtX*D^Q
z!?tiDr*?;g)rgDwC@R&U!Jqb6-@QideI_>A+%wjYcXBs|m;aIMd}WT?N}TQPbb6Pa
zrLuS7@Ad%gYH?+(${W~V$pKI7;5hqMwK*&<R*APC0}$G^zTD5u%c<3GvfVw94Ap(W
z*zUl1p8=9`c)fV_-LczhXf&mFa|5Vj6sAtj=yqQ8Kh|9VPjp-py0nCFjY-Vi8U!Y~
z*1PXe9S^q}lXIrVlWy1D+fWfY#J+G7j`qktbA)1->Pq;O2qnBeWSu+p5n3U6XKNQ(
zt1r8-_c}A7QiZ<3Jxl^LCygb@->_*uK~fd*S6nuh;-Kwkf-hhvMyaN3lV<h=JT(`~
zRi3>Z+>}@aIrs=)(SfHVDASEaK<9`6AbY8;D;sfF6OJ#(`O81g2@*ZZAv+OG6c6|`
zYg|cD37FUM?Ks_d+l|@oSIcbuUm`OUak!%y0r)ecXX5z{&{;?VUx+3@ziod7Twz6o
zhG&X(660L~pdUkXju^p$RrEy!17oeZZzZhh1Pv<H1?Q>kUP)HP_~^qCVrsUTmDJG<
zxOc_}N?#X4<w0y!xtYB8LhWaCkIN*KHa1&JNHU*+#GR{1jS681&<<Y8n*pkn*Ws?B
zsxcpWx<Srox#V;vT|2&DY<dDu^%OV&VlD9GCUad(yBXaX<S4GSnUwCS65h4@{e7q#
z`j^{=t_wWK$>lv0Fh5h@mt7D*e#Kw&qD4e#?ywBfnTZ5eL}Opf{s4xy7W49N_4Rgj
z7UtLilNFpd17Tm*^>M$w*f>F?h14e^q3<ifh2%eoBr~^SyR^LSl2I9o7ZTaq_@*_K
zaW-U~1tu;(%H&jYPo!zmcdFX|Mjs=pRcL)Egl7l(3S5w>NJb`8W@#4l@SJGx6RGxJ
z+<tPS?bceOET?|`A!{H$MZ!eIPW1Z<OW-EflkAkO7roW(g0ow&xEjwtr9YS-kt?S6
zAXEC7Ic2T*U}0@d$x41(7Zgy50uqBGM_Zts^fp)P^hE33YYjvKR8Q2w+f;ulPyR^+
z)R`%j`8E26%U>JQ+Q)d2vWDkWdtZusE4@|uSoXc|BRm&rk)~f!J<qeJG?LC}9O<PL
z93%X9QCqi<<fhRIFahZY1ro~w9l^l6RhD=7fq9PM06)eSXE@hYm}Y`$yF>e;p9aIO
zIPxWU?M?Q~oI<0eZhdQ_31-3cQ<7v6dU)CH*^xwjl?E?{_q}h_v3FU$vv*uR6{?+{
z!0tlNFH~1>zEC!jr;XR*{<N5F;Ep2WnfT0i6>q|`%OFm>uScxC+tIU%sN2;xsf5Qq
zvsOHbj7C&Smbtnm>z90E-TBjbAl^5b#h$MiPW29S-f>Hli+A#ITSq!xiKB!9Es?om
zX_3jYk;#_~eQ|R5mi<2uBg%8&%tIia(c?e6n0wkpZveDS+v8~c?gnn@Y45AoZ;hLZ
z+oEl>uSdL`o|4Ky<6>@|#_LDh_FhdN{qW9ZP)<tQNUg$m%V~ipy2Lfrgu9|LTkyQP
zeYW(fZQ<~I$#`H|5#R8!3dlWfMsRQ9(VevBs5n}_#}8~C7hYFO5o)%QnKwpg1bbm?
zru$IZ0!9uA3rran3iH2pc0;q>de|0wi`i%=s@svvavumWifKWofS0n+fo)bBGM5qi
zSrtgJ=G@(fLWo}ZNLuAl@{onP!^rs(2b8;q<n0!1YwaQ}iXOdrucqTzH~g7$^x8U^
zGBqmIRuHa5ab65Ux$Gtvc0Z=$c_bqnlRg7YNi(k8VIF)>a)n&1@GvAs4E1Z@r<(N2
z5_m8`ik`uAq9Z$@KrkQR*b2IYQs8|_J(L<bo&K?J<jzp5dBv<xrPq0?kDY)30EM@r
zU={cp^)<-i$Mup@9Yq1JfH?w=;?>$Ov3uK~{_oY~Og6h=+jh&Fgq#CfR(4ax#;0Dl
zHJCnBMJhIbx&#;2Nf-0j<J_#iZNr^O;^xs)(3rE-_ocHF)x|uIvo7$Op+1Q`N?BdH
z(#Z7MV^{XCKJS&}g$(Ad<k3<}7}H~lu0fVREL<p`i4hUze$BQdOL^&|8oBx-f<Pvh
zB~MyGhaBg-)r)5?rI=k13OY?synnW&!}cX(Z-;EW@h5MJF}-@E2AtkX33L4hoCl0q
z-gMD-b#g~2+hRMpv3DdGCJ5KROy-MOd>ENlJQLsa&O=yRGp(~<m~Y-o%phg?!ZXvi
z+IEjTEijc%&`l;~t)ly$;E5J~S=1nC-K%c2`7#ZAA6^88Gifb#<(;G9j9ho(lDy6*
zBD^Uez+m!wZSA?JF{2?ocRC&m<GE($eHe6HS5?;W@5WULdP$SIzoqYwqADLQve3+j
z=4TkdCcZSBb_ZtrMm-9`H~wfFC_Ex`GByP|*JE2f?fv3Ry^5Y&Yx8Cc;5i><z*az^
zYs?@?d1F)AI9ebLQwb#<7T$tI=ou0BsISH)ui0O<V0Jp7uzu%@V!Unf5`QY%eVkFX
zB}JBb?zpRN7Y^5C6>K|&QPzDsh|F;w*!MM+N1(7u<&6{r>aL7)XuEwpfrs^E%a3~Q
zj?u-=Bzzh)zv1<mn>yQ`_=vvbvlm$l{5@Qy;$Zj!Jx-YO;IE=F!z0W14fiH{ss*Ko
zW!aif7G|$`!^molq-ZL}zdQiCoVJb?b;mVJ#>nPdbNmeoh3Ov#<z}8w&9JxVB55IB
zBPrHYtqmWB*UFQe>Ht!KC7U?h#D943u=!Fk(SSHtKeHAgef5bx`biULLX8vRzf0+P
zpWyUqU708#@aXI*jsIdE=hIv~nP1^ZROH1d2rO66i(Imsh2%mt&6lg9a?8n$Z#Vl+
z77L*aTN;oO6y@1;Jf~vM0BRE6(KS>UYgk(>&ooO8FEPhtPmTyr5Hq*cMt~1_T3hKf
zu$~Wbab;Tg!?i_TZ#_EWJa{E)Kr44@97$Vw6xefWENKV3`G$B<#6_>pjSHyHXxvQ0
zOvqCI*3SF%5KEGyMUYVkz>xmsgUmbLAS%0-Q^E4UCFSwVK<!%Q$-e5Mf+!OQ)xmsS
zJ@3Jx$529}%PMm#F6=KF>mZknpIBXdCBb;mTqJVPXf0W`-mZXrUV5i%HN@fx*^6Ns
zo(`6dAO>N|F)YIOdUI>Ale-@}h>QZU+2aY4?)zC2=V?3pjVk~j)7EotvcbBz9iWyi
zrc@_-d40D;^H%&svAn{5o+XY6`nIZh?#mHb3FuqWkNh7z$g4j1l!73$hq)Bac%wZ;
zcI|e{ePP~NMFXrIR^frnJHyd1Zr+pJ*lzU%{dXF?eVikc)AZ>oI-Ng|VbfRD2+EO7
zpwf^-nbe>|Ik3N$kUT4ljatL^azWtNV%=zd``lIOs(BY(TUn;;+NYy~)#(L2ufJ?n
zb!=fclt1p~c=FsTjy|hyzY^Dt<y`bNFzfne*yE}_v}!vc&{!1>2V0F^Z-cinkS=uP
z%F?=SZS#!lQ85(FU?`4LclvO(>D04OiP(8#OV4@jUNp~?S@2Y6NkG5eW`WDBFxMYP
zx!A)gjKyr3CQcZo@J3;<OnczzEA|m{N}8GmCikdNUX??xSE4MZS*htz#=iMvb6qrP
z^Zm@3T0u`nJAhghPI^dLm`d=TGwPdenc9Rtv`Rb~_%9bSwwgI_llf3E%vzqxIDO7V
zUwsts-iZ>)YG3mTJ}y%N!JDY4__t!AL1MRvPC0>js%_9bm^l81x)R_`33;SQ7WK#D
zs8pXQpd1S2RHPBFne^#Fpp%cbrxLMrE+;dmG-aO_pNckSoz!~ZoJxwYIta72OWUdj
zPMZzeZwa4t2>Y;k7mFedT_0RVv!V$xHka=CHnnW%)CAYP<~4nbvp&sBRlqLrK|;@2
zwC!WDm)qRKD)s{k0k`c=1o=DKSa6tjrxsjfN$uLdH&a8bt=5-fI4lUv0`j?I>1k%i
ziBWQqJgxJb+Exxqd#29K()058vu3lWpWk15%8^iCHEsK*Xl5VE)wAB8Hslp^D-8|G
z*g3?G(qN+;H+Fa@-~vmPlX0zkirQdp0Y005qF6|3e-U$C!Kz90BLWY}lXyv{GufYS
zZL3AolMZMF<AqE@T-HrK4XTV6&bM;T-<<{GBw4|maE9f2hKw<NwZ&DvV8&hA>u%+S
zXa3->bpJ*`tZfe0X10!z#D|0Cm&J2eDYa?DaSD$PJ?$OKP?BIxZqqj3;mB_3Pgc=z
z)t<Ou<9>E5{K2rSL0n~wdt_kw3rX(y%Is}sNH54%ZHM=uM9|h4^k58ZeY3rTHI5WL
z3Lm$?h-Gh+1T`zTJmS3gj^E<oTS<#&n1)QpWl70ciSeQ|eE*oB9j^D1B~S9NPj`;`
z?i_&!Mx?8?bLr_TcPn2RX|{6qI2+DUitEQy_Y0p>(c9S`!<I0j<`wK{D+pMw)Ql2g
zw`}e)%-7r&YIt%vyXVT``EUu|UJJDiji=I18sEU^@kOg97o6rR6TDIGOrKNVG+sFQ
zF+RpU2Cn@*i|M^R=6^b5pY4dRq^0y3>s7Qx&$`M_q0L#ITfcG&C5vtT?gj>R&3z9P
zr}VS#3hk39PHX4Lsbwk~+3n7-bYLWIKX+%u-^(CeVVaycYZB5f>E`@p^hh-c#mN)X
zh5=5SN(%2{4pY9BUF=Z`leaStj6>M&j#Q`(h!9)o;S7dgmg%01w~-PA9>V=&iHgme
z7O7KJ=4lxGdvrSmab#}AbSvANhN*MgbmS;y+EqPq2Sx#9MUC1xr@LhC=<%~dwPNuT
zYKz@tt23@8MsjaB8}+ZJ>3_h}tY{3wxiS3x+^nTF1%0)f4-%$rc;!`GD_QYG{3Z8J
zc7MDVC*x7l%MH$dB&JPN9o`DF+PXrY@r<LZOA_BB6k=DGMj>*Bsfn&+)#Be!D@w2y
zA!p{(6d%^?T`yK8`^K8Y(HO~|;Z*e0XmSW{qjd@&p{T2-6Ot)_UEA8)&*T!+d*?Yn
zeIs*UmNJ}{A|n+@0evH=r!9T2Q_MZY)UcP$1y6o%w|3glu4tKzl(J<cx6)nDter9t
z`j}NM_KQuYAm>>JsOS4x5<4v`M>>qWr;Tu$nvk|gw9{;wo1!Rzhf%Dn-2ib|uPxwY
z?|LUjFMPi260ch2=F0mdYEP$P?`aj*nH*1F5$0#NIrsKC35xYXt^1U2Plntz2FI_j
zOjeAaRXv12lv#tUO57SL&}Nj+>`5|5Q+=Q5Y9P0!6GpR6OV`Mm8Y?AQX?fo{I+fsJ
z#dy|MxK~H@rRDArt(!{DTdz7d=yHe{I?d)m$nad=Xk?Bp@%Xs3?W8+h$~&5v$<Ism
zWzCn-;{dOv$@ST-Ga3{O`-Iax+tzneHmPo=-gB?5xN53H&X>N5+gM5qiae;142Qz%
z^2yZcW-2%f;XC$YSRgIONKv?>*CUS0?d`5tbxn32Zg$G~kaD9LJ@zzl&>!gRj&sbm
z{u|53S2SA4pPogkmd!b((sQNSql1nccV7Tv<Uk3(L4|ngc)peT@t;|ah~77{lTcbO
zY%M}@B5NRS%TG*m6HTVF*Y)uEW|9kT$xZud_p2}2ugKD@X*8?8JGQNAq^;V%$q3qc
z^~6ZgmUkAnC&0R5($#3YJ-Om4n}%jXKdK_@F~NPpm^!=e0IV*JcUAK}yP2d&j5^;R
zqoRU}TWC!u*JLO*rh2`^C%5h=FEC#ql30=^(Rf2Ktu1j(@Pf$oDBX%!VOJ-I<0HTL
zFXw1dk3bv|t%6M6Nb97@fhq))KRxYJ>@(gIRFPh=@!E>gBSM~cH<3n<ZjC*Hit0pk
z`?_Az3-yLT2F^-EXPTJt$djmZb=vwGN&$xhW@!CZcUe1V1w*$|Vy?f(eInocFQ=lm
zA|^rAsuZahYEcsH&ZN5opd0rFuk$mP4qQpIz9Y;_8O+aw7mCU{f(L!iQa;h_*aFP>
zIKT)!2NC&>^C~0VB#+A_XMMcrsuEZ%!*qzRZvm@&Y1MngEX*I^dhU|tCN?ETATW@1
z<71r;ul#-#_~9r>qa}xBI(CNv80Tyfdi?d{(xev%_p!uJSeN9nAcUfj#aDJLPno*N
zZS}o0yM}JN8mu~P>XypFOBKx1SST-~7ChJ^?7!QZPH;U6(c<_!awT#9N)F)cA9Mj2
zOEi+lLhIBMuk$X7JMY)OGFG@Lqi&VhTUW_^=Z@ufaR_(8-InYw^~=)rS<&vnOJ$)m
zWHuKlTO*$s9ksQa#SPHAr)XL5ySJ<F*mJ+D%svWvP4@h%ZGh>~T??$^saWL~sOqG_
z{B%+jlE-#2KA@n_ghpLWcbd}5z1xu~uctbfd~8O@r_>gEB2WMNd>2wg0uG>*J&{3G
z{(fh)Pa$3iHSeF2d4rm>*%d7!sQqv&X}!~r)in^3x`nsY3MZo1frq`CAyonLr$*Wp
zmfP~w%H;8Js29^tTxWH;X42}XUA9hApuk{lkKS1{gLM09!^x1X6u-eG4A-F?CvMQ5
zid?Q?10sn+8F3ylO5B-FNdya~4aWs;W)kkBK0B?X2=TxYi3e+r?Y+;0EQhXLm##Nc
z1f=9q=)rg8@j1|{OVMVU);N6<m6{^2cK6;>SoqjvoX;ztR)sh{F6zgeqRpyCx=)J3
zKeQ5MS+zy+<u>eqKKR&3RN3id4+V0zWq6EI#xu2?R2sWl0qv4Z_)VUuDSq+y2H6Yv
zh{s~5O+vKr-OlGqhf8*-=Lwb|J#O2G)B%}?i`}1uJ*l?{ol07<E4zSD7X*X+<6=>T
zCZ%fyRPrV(^{yXND42BENBI?x+AiG3y&ZE=Tdf4#j#a4=FSc@iEc|A4^cH5Vev~D9
zm&USd)46T7DuV?=z^O)zwFMd-3#qcn>aN}Vz=ZAL*AP-?)-uxxQ9X9WV1JPOW_XmQ
z=KwD<(o=_GEL_SFBEH0waje>2I)32n&kG4ys@&&M3Kc6iomoHK3ML#D51^w)^t9Qc
z;~-s?TWRD)Eg`lDB^#f%s81e|b%W6PDj5q3g%bKEI-S#k!&OzY=@|(bjT!rqCh286
zSHRausk-0J$NF$|KP{FgLOC5o(s$2+OF>qsI4Y_Om90F3A7YHCHr`d+y-RAlwK*hs
zd)8%+!*Z_~HWd1Vf7(#L6(6rW8;JD{!segR_duOXo7%Yv%{wVJ(bK%g3iNE(ZaH{M
z4#{+EQFTv&t8(iU$=z$$pN3vn8@K+{;1*0iHTGC-%2l>BKBz!4+jTZ8cxV9t^Vmn*
zjBp*fDJ`rbh<cbr)y%I&fscAF?vAy~;w0VL!<MkCCVMf?3#B3iN5-69x&ukyck08B
zt&^;@O*%U}RM&gCb~Ag{6}hUDUDkakW#4l?=NDMDoe!uwM&MCT4$S$Fdm|I$ii0?u
zQPd;x0W0UyeJI=A&I1NglEpvn>LPQYq(Ok+3YPKKnp17j+HSk{NndxXx~DoHQgpX1
ztjy?U9LqVO!@T&j5c0f1Y$}6_;}C|9h(Fm`?F%M%i-4R!qK&bU#I4FTf2BvI-W%P%
zrM9sOpT5)d(<gPKXlj8);7xcPIV1kTjfT;*1AQ@;5tW~{2<puPK9o!`tFUJX$7Ls3
zD^9OO%gMcoB_*kHiX0(!iOe6q0OFN%2<WZfu1T7swp5|`>t(c~E#5OQ`;cvV`&m*G
zZY2n3`)E|5Bd3z|l;Y*Xon}v$jD*auntV!9eneUt9``go|FR-|zEyBZc-*SNU=^rq
zA2?X^jne#l-6Tn(Cq8_<x{ttPhjX{+i;sTW)cJaVR9|5laeKmu01Zo4YUfS&8Jt>t
z3+B^es2JY!H?WVJ{XEzxPtwC39iC1Xw6RZ-nKp6pu8DhjJuBVg_}45-=@$)zZ8w%k
zm$S|oWQ0(E8bMO@sKxXVHB(n@gC=`-<B_w1A*uOB0z9ecNM(kXjHrF<LFs5pJT_H2
zo96_>x|C#5^y~ae#`#7oXfki>dOqG}(r&}APIsPZ2%gc^1(KX6ztkNmj1Zd|+`F5m
zXP=ZD8CH%#bOj3X`EV)6O4K?&!x1cv`7%+ozyFmY#(gE^W<@E~wqGin%S&qtth%;}
z1}hX3lG~L89><vF{IwSWE)UObgS7pPF9H~Vgk5~jDSdF?XXt}4lvS3*Kj-A2ix8po
zIDKg3HvokS@5XhYssRq+T@y@Pi6HEG1NM&-)+?vdR&qz{dQ3jZrI!m+mg>-~lM_$K
z2bf_35m+vLj(iC?9rDyENOqLUkN>UkD*pU)__uBmbI5*5#*-sjZR>x$XaRkES0A?k
z|LZY<htqvTUMlZgB0=Cb`)wlLmyy(l!Wg(sv-aAH$W$_{l4IcT2kaj$4eIlsac7;b
z?LvXkw1ozNd>T${G%y>B2wR$tPAU0PqN}Hv%gKeJj>`Ui8;Y;Ip0`HcxtM=?=ahsg
zpYI6Sua5M*-oiYNeIOtXJBEJqm8V%<S8eC|wq{1B0F+HcC?NZ!coI_Vcn82+I6;b1
z=3Bk80u7s3!_Pw9;t8E{R1e>>ZH_i7MoZKV>gMQ)in~#J-2O$_Xkt-FBt+TP^p@E#
z`um;L?z;pcvRwH~lk}(!0Es?OJde9qlwblyJFTOFSQFqiLr;x{LYR>Z+J@T^v}epJ
zp<0g{{pT4#14E0;FeN1LP?=d`KyE*0B2scxIiR2%ZCClxvs%d{C;t>G*NtuIazj|r
z?PA&3I!7^4F>T|p3#IMAIuk0)&^LxgWI)AkN6>GT1g*Tmn<P9H^}|EoCeUq_OU<*~
z$QH^M?PM@NA(Q;`_B}26(5k6qdEu-@uvx5=Us7>KRSkG>pFd!w>*+RJW={!J`*(35
zZeOs4vW?6VdM%$pudt>*M*S<RjSD@Hs)h$w24d+;g|BfFTC3~kdq6OW?Q_~gBwC0E
zf%>iGv|3-9w&K9oS#qkkodK+Aeps2|q>H4tETcQ|jv-4j32ia+eQ}j?lpg$Yu1(BW
zIJ9)v=2a0!yY8*oLbxS?t0Ifn#@=Qp6=``Bd;rz`@jizTB$~sM!Ite!%qRtV)ukw>
z{8c>fb~@@-K(ML`Yl9<KxhuKu$+pCij4t}W)I0!&kad<M`C*{DVx)C};qk}QxilsH
zgec7d<5#&(Lb*pPP-W8zejJkmThoc0lY_uz8*(8f^vAu+k&hLo#Wflo(9<qxP>B(+
zn>#U`?5g!C<ZOC@ZxvL-az6-ZBJ;;CaG{ohrP&*=rOarhid??_+64)pP}@8bUUM$u
z59pJ8gPl5!#eFG{Ujo09lG6!XErIeF0xm#G^g7~1MS(ol6$nlrSD*>cYc|wMB;*=-
zNd+<)Y!Bq41?AJ%dX80$CArd6gmxLN+RP}yE}MyQb4*5(UUC<PJcbc3$$Nhaca(P&
z<lFbx{ni4qqcfx==j|=d0Q|9L!cvr>)37?2HU3Vg!l?$>NKC11j{z4dbDyozs{QB>
zQ_Aj^JiMn;?&AB4m!c3)NauD3+eNp`i{%~#YTgL}<%H~Wz7z}*DyK&I%XLNJnl!h~
zyD?A(Iqa6;lv7ss7N+UV;%-N|q{I0;_RV(Y3c8oIQF>6r;I7H!wE=wv*OekKZ)Vi#
zJycmC7uv6wOW<L8S@~gf+jV}*&DRf{Lwyu`x5jr_p0zO-GN&;JW{VxaR0fzZ(c`Aj
zrbm$K;GtzM%@%T!T2S)OFYFqs8X*sA6U`)BYbjoN56)1bcJbZ7!o)hu7Hy8RQ_+L)
z`zoFrs}=GncLSO0ppQ6RT(!V!>;cdk7GVW#-KGt<!X8yJD*qF5l!8kl^)4E3S+&P(
zpp|0Kl0v@(h8YLjCr8+(<!kp@fgRUEa%}9qiq+ocyS_2qzYCwb*$Gy!(c6!JMod>+
z)K*9FrNyHawjAN8;A=E>!-9R_DDgh>(P9WclbJz3U9L5;(?6h)CBT2`7WU24Fg>Bw
z=1LRqMAWMmn!F_49gK2L%e%mad?X~bp?|a1cpE~QWl>4wL0~gkno=Y3$v7Op`;OAE
z>zG=qE#74V3m9X+`@9Zh_;|>X(M?C>9#L8w3(v1V)KAbACs5;OXO@ll%3L=hla{{y
zD4@NwD*x=uout55%AS4>i6UH8?Fr~oI)RX(Majso`SNw4U+tz^!p8%ZR^8a8S2t2H
zV(bmS*)jPcx6tpgzWWM+P{F$qyp50p$oRjuM0?!;)o?(c+4800Ar7+&PB<~`BByOX
z+{Yn_m1~T3g(}?16RZyq*i3miU~6ks!a!JP_|Fl3xU>k1wPjMMsP-AWo3Ktj9(aI!
zhf$}#3T=9mvWjt4wYW!B^IfR>%)7VSYT2z9>7J=YJ7;Lb4~)1}CZK_zDrKQ$^+ODJ
zfC<bZslU3x`~jlbE9WAb5TJy2tJn^#9G+B73jO0LPAMGs%-fHOvH2vq@FhFla=0hw
zNM+n0RHsV3iHWTJF~>XWMuFV;$q;+&gcIx8@&(|6GIMuF8}`=S$Mt4)P2LfEZW)YE
z(y`&%zTqH5s^|hia_6FF45M<8jsxSK@YdIN>_dZ~P`|ne6h=mb%~k`nUnZGIo=XnU
zU-pLNpSh#Uj^3i+PC5<!u*1u>&xsa0E3tbr6TZAZ*1Ayal%|*oc(HJjrk+0gR^2e%
zf5p$oUFCA&nDSidZQK9|;yH^78_!2R;T<1g27sl>Ms95iyOiNnnZXL@Exng%7c5+2
z{8|&21ss>ZE@h^z^ZWWXC$57VkCgb~SSRmiZ8Q{^@9=VRbt{#3CdDNz*}in5TAK;e
z4{W?!9GIe}>)KPA@lBUfb^}#CyFgY|0bO)XeP_C5n{<XPsTx0^Pf9)n4xnL6;xcYh
z@!HIJxuL>WPSn>x06b(HLC7%G9?N7r0>tK#Rwm1#iv`;loqz>D@ZRywr#|}wd`U9M
ziC~I>%iKTD+0VS8WK{VVD);KEoi8O5qp>m5VoKF87*Cz<SbR21h<TLazD2$Vcab;H
zENhgO{}5a$NE2CBtg6!V#p(Hh&X^UgXkf;sK)(hCOH^fn2rM&pWA=7kFH(>Hie)Mv
zcW=MvOvBKxu2#Kl`Y**ez`BJjn?(c_Be3Fpluy8DOUQz@%B9$HIFCKUIbrKk_w+vD
zkMiqkJw6jk!Sm{-gG-w)DuDp%>lTuR1t~Gf?&-$qZMrOE?5uPvZuB)yaJD!F3kTwQ
zmeyFO)ccihj6d(-3-3hW-m>*x1*d}pF7mdR@OO;M-67Hi5)t8?MkH+L^2f9HC2O^A
zg;?@r*vCQ>m781l`xh%9#AcIC19yzGu!Q|Vt=Ph>>E8ajpI!hYk;$7$F}pU8aM*V-
zI|;lkwzYR@YTX$3(ogw<YY|IF-2_bQJ(5aueNTJC7KF}rp02w-(M|4ffDDPKO~c!w
z1@>cJp;K;AJxe9in9vnvkYbJWkyS<Acc6DdM^28qtOQ)9`<;|SeesI@f)<U#oB1?1
z;@}Jk9-&*CP-0_r9`KGhjQd-Aw~3lwdmiKL$4B#Ah4&o4?d<E^3Q)h2fB{HzpANvu
z#MZak-M2D5Hq)J2Sc}RxPfD4JL|%Jz|KVrP{q>V~HkEe`Xew1ZyRtCJ&qAfK7lF-U
z*rbRR?Pol-#QPjHq`mK4(Vu_2el{cKd~a??fr(P_C2i@{xA$U6C>f((?az{s-<J_z
z2$l&8VO`r8p7e#vrNJ_&>p6>3IB;~gg$QO1y`3BaG1DtQXvzdp90o5mqu0<rG6BLl
z@s89)C19OU8hH*;kC(N5a-sygje<zCr>}Kup1VDJ{csm!Gjnu9aP#r`M{~@M{u?eT
zQz{KXZq^k1({X(}zmd5ME$q9MI4Ega#cUD0C@kw#_DJZr=>grUsdTbwf-^ffAT!13
zH1;%qeWRvB`<2uw5VJ!2^N5JVO&(08NE>s(f)c2EbronXOw*XM^rZb*U>YoSbBysn
zA7NV>hr(Z~=Uibn&?YpJgZLG%+Af@(ecNQElJ*H;3e8IV8J6~>Sxcar^74xbTkxwe
z)RtCeNJ1Bl8z2hpG<3$!1(%n1M63FP2SirmJ_AN~!kN9pwGQdD+j1r7tWkBO_>$VT
z-Eoc?IQf3i6j~p%p1M`7_XqR48PmphKc0E%-<sH_X1doUSUIma*y|kdY~1@;;?y<U
z<%Ip`RtFW^R#NT1p^>UOUCZ<g_Nx3JxBVrBP_VgJnjyg<IlbMHQee9i)8#3M^@!7R
zY=^Ldn^lnBwV5y9!MWfNmyBou{oGPrtmO+F3+*cZv~ORR9hE~WquDlbqGeu&*&Nsr
zaZ$&mo)Q_pt90eoMsj((Z;Wym`Od?fxZad|m~OF`mUj_}KpI))!%CY(?c-zjCM$Qh
z%xH*CEIef{_jgD8^#hE+3<*?x^NXg0NRM6oQif#<&L=ab<m`{8jdHd}Ot0GbGeY!s
zDciR5;wT(_;y94|x3QmUl3$-<ApTMg)I0Ul>CWt*93`hdK(^Wt^OCFO3Ee{kaqBMf
z6Jo**h*VUvEH%I7+)lU7brR|5dfTsP+Z<`(IyU3vAX3qj)}YU%N7EMTj*h$9&oV8+
z`#}69<2+%f4aNG(6J3KedXZU^yI9dj4c<QAdiW6*-xEM9S`&Qt6<oGyO*CWK`%vJt
zhR0UzzFD(D(-Dq)@xNgr?oIPlpmaE!d-*|%?{TWnRP1PXwD@gpO_n~Etq$))4>EOP
zYAG$IoxZ&%yL^ZrlH@YEss+%?l%#Y}XVS`ThVDTjY*+Jq%<M*$CErZ^tEIxbxeacg
z8g7@gJS|r%?b*m2jhbwYG~!m3_s1||za@Kg8B5)-(gnuM^WEXEW%AOwm%)7{Co^=n
z<V9t6&U?}t=zPAm+m?vl^#7TDoKjdJrurPmh_A&u18*k0Y=M#E!EJGZAKe6tX*!?W
zbf=fItfpk^lrF#tMF&i2N-Wc6$pKGd>s`CpkeNj)Izwx7j#4N{)qL<duS~+iByO%7
zy8<2LwWpQ6NdVY62HP6#1kMn2YFS*7v{afmlbXmzo1M@s+_z=zHQ<`Ke&-Duc%e0g
zp0jR-MMdgaV$u22rw8GN;hOGBdFh<0Fcno|CtY~;rqb1Ub?~GsrO#Hg2zR<W^p<5S
z9bXY4A%^$aH~k%dj4|dKCTii8>Gvm(qa_k~NQJwFbYm|UmIeRPYt)_2dBS6QSxPD*
z8#Bww>9)O*Ihez18!%rv&_`dE=4k#fsuZXg{;ta(4xPAiid);toYJ#?q+E6B{=j#0
zCUb-8zEd~#t&+>s=jy_8<!et2`o-kRlhNWoB<9kf>q)js-HWi-i-)0Hprt=){JN?n
zNpQe2Dxf0GY;MW0Uf9B9F2F_K8x>IDX<;reP*3uib`Ukb+wXBx|FT|y>p(ph_!sq&
zJ=QPFmdkPiB({DemA6RZI;ZQFjlS;=_gXzBmscgABG<^=`(<tbcRh*S3Fno;OZLFz
zI6>d?N<(6Zeo#7rO=8IYcg|GmCSj40y1uV=t3V>psiwYqdi%;wK8dNr^fp%Jhf8xS
zVpU3GW~2Oo=trmTjZN1=x7G14GC(ILZSw^m6J$$+ef4osSVEra*1r7&ii7@@X>yK<
zxMz4l^S&$k`*_DIUqmOn8v5RNggvunC^IMMmXLkP{H0#~+SUYfUF87dsUh0{*S`1D
zvDrd4^BlIb%KEV}+~7a>5PN#<9$7w2i=}n$`Z%po$E<B61eDu)o|BR7!uXbL09)!w
zNHMsR0$&Sq$$jdDOZ0*VXa=&gP3CIJ7ch6Q&ivxay}%4TG1y^bo)exM5VVj#Pu?|w
z<vEEN*LKV&e_K!DzV+J)XR(&Pl<)wHL$+AeqMD9A#;{@T_}$~?<7v5`-j)?(69rx_
z&9UvHM&C6u!}N6xB$rRTGO`WM9-a?sB)*hB-mz90mMfEsA6{cOM#MV)t3J$B%l3&?
zue7R42U50nZaeAriI^l-X)}~PJ<rfw^j2QY{(_DA<R3D~yW526;{!3__c;?Vfv(Sl
z=LHJs2lu!8yDA7;I2BQ*?;WD!`lt;01uF)Y@|_#FAxQ*8Jy3-v2MmBYb|)TBCs`2f
zP<0<?IR;Sb>W~G0A#}-Ivz;GG_C)uHCly7fD_W$Z2gk|+Tt8w98)tUY=5sz2-h0}t
zJ8h-m&~Luj<6|aZ+qi+j%&YUDI{G>_9v6*ok*7z02Z0)|Y5sS~<CylyOKHQJ7H$*Q
zUW%{8;YTr68P+ij<R1#fX78T+)kdE<4l=bk!hQ%oY6kr_kCH#5VyXPs!IPzozL`|(
z0Q2xyrFVcXLu580TAsO~P~(od(0yGc!gi(4&dS0@$q^bS&7}t~fv+LHJEH7PN@NFy
zHhaE5#loMxlA0>jT_cw8s(HJhIyQrTS=Z&;(_DE0jEKbGt*Tt1t`ZOOEKa9R<Jcsw
zHOaV4To(=LZRNwiw`t-yh?h6BDBsEioF#R_^2zSaEhUS0Ez3TQ@muC=Hyb1^c>(ib
zDu8e59vwH4Aw}v$#(TW3RnypUy!S|SGHzdVGnpi2SD}<@LscbDD81C|kan%|8=Lui
z&LVb{$1mu~V~>%VD~hGj4b119SQz-ATl_(rE%lx}CI9q^?!u_RbFHHw1L^jrmRulv
zBY{;ZV%~=}PM*i{D^My-^WSbLBxF<JAts(3kiuUZtfkY4&0gFte6!}49ckX~!fxEL
zcFIIm*VT4ieMMYJc%pr3k#n%{u5F;}rI4>zkGh>Wr?HzZ%tqNHeOj`?p$Lb&zXt79
z%x>G%mIkjz$Ow%cw8;SLDviv;Y^odQ)sB?-29at9ZvMo`35bqYdCZEY$JU=KWXYhQ
zl1s-;pT1l%8RD{20<tP$+azNh(ll#k?JEDtTd`)+fc-WWUY8nEml>9Iq+p)K-3$1>
zZn~a|97@(k{<4MBvn6QOH1g!4ZfuX8edD>eVF;Q~NIN@rSAYs_Mz%sS8+LP@IYh?P
zPYJJEl;o=Mu|mbRI-#iaMliqW9f2fy{<qVD;c)7d@D(Qk+h_bPFjx$ocvRs)`P{>q
z`oJegVxF%5ghfUAiJCyqT=XgD^e<c+$>HK|5O=S91}J&niB;`N9K*mzp(v_LftG|9
zL>_{VwMk9AU7s~}>11?l`_EtX#|7x>S*N0l_4Wm5FPT0-COQzYG);Ej+cjZGSFTdN
zsGjK|Zrhd0>U2P};v`a5nIsNVm$`-hdbvpvH-N*JCjY9>mpT;}v!DAx3z$6&4*))>
zi9M#Lg{E{$&=ztu4Ws4WW;>-IN#IF$|M?%1rk|)l<AI9{TU&sOPC5~|fKS)(GhMni
zZBe7KJ3YZHEM_b>^}c?5H95MsTO5AvQ35@OI*{~^%#*4!t^5KO>JYDi%u~(3n$|xa
z{1&MUP+V>1`mS<683c;KgKTJgvSQQk*>jb@^68rCP7B%m>`gh;v8clS;q8A}tDs^L
z*;KUI#vqW@!=UY4OWvMws@iR53(?*Fw)pspJ_jM4&la@WIe|!3f%ww40G-w-P-yLL
zJ~Wn@^9g!?>Fbr1K)-GJKKH~=4B^Mq2feP2;<J{4mQR4r8;srjY|N6{OSelJ1-A$N
zefGzI8q*Qzv@M@&yju|oC40$dLV(6%3pW46_WZewbqBj&Z*2kWfY_*7LdTgP(atkh
zzaaujqvLAVLfP+rhU0n#+_YY*o2I>!8Lf0hY(2kvagga8A^RoFI0%L>VOK6>{Nf=0
z$R^UyJVd<D5W89zP|Yr|?1EcTBxqg-GfTd_NQO!%+rAp?2_m5-5oywu91iN32zG5u
z7?kRf;M>y9W!{&=A8GJQ5B8Jcd^q(+Tyi&-o3J|Us(uSl4)wapK7)Uu+O6!(bSgqb
z35$E4Yt$ZN2vsEy_~EeD8Qx|7xrhpga-TRP_t*LUeO1e=!I=L6w6rYG%Y^B5$7Sk}
zKu3m#W&@ifd8`Y{IhwC2DJ|UJt)K>P0jFRC#EiZREMLEJexUm!eEcJis_rh>+5$z0
zY=&x93xeQ|u;hCTm>}6G^fZ(vFXi+7d_Ut%K{%hWHm3RWfBbk%P)8stXxKS>v90MD
zw;iR#N+$y4Q*pb6mctSClX(l&Jij}s{CW-rEO9CtyL!fB`PY%=R)Z51QD*g#_UF_8
z-1+}<a{uxLb;J=*s+Q(|{;q%iYByp5V75?s+UTFZ^6RAhw_kXGIK!}!>;BK*_0L~b
zutq!pO7T8XzwX`t_P77@&jqQ#nq+dhc>Y?rzdZWC_#Gmxh;zf_0{!EETJ^vAVG&hW
zll)Ih#Q)g_{GZ-9&tM`HBKQW+>D0;4i2ra&6$m|{y&nee?rXi)@QjO+C_N-nK$zkv
zuBZCTx&HMI|0`b!(_9hY*0e%xqa1oP^epNjFyya-#}oVimR-m#zvG4<o*nt!E(ELu
zZSz6WmPh3g@Xx2>f8Is~B_c3DHZFoxJnuouSppA(M!x^lbmXrFi~s8Jf7W3?ftS?(
zZ>#awBlG{gQ2x3eB60tBc%l5+Ci9;U`M-J@{;xE!38oH%9PFT1HDV#wPo2+0Wa^1>
zXDZZVR{SCR`1xl4mxuPGV+Jlnz~H%w<_^GDtq^Tx6g}c(&<d`m(GiOvq~!o5T-7Ep
zC58*YMRRx#fWEXvM{ndRQ=lb8hGrnX#0GXy<8VU*Ufy+5c8m~Fi*-6K13Qb|<z8s2
zu_xkVI9BUhhEkw@0LB}f7U17J&2cW;aaTC_xPj}xxB`FPmnW2pP}NAz2KF?ikzXDy
zffBo@U(cKTOQn>oGvaFkq{p{gC*2ZH6-L_)G;*1IIxEQ6xlS7D42@wW+|Yf$OD%|V
z?AA%w709B(!CJAbu_@N;D;=-poUjpFGuZzdtU<9doi4Rp3eECK_;*&`Ax@(<F8AHd
z6iDgsvr>|pBXhG_zyx%m37c;uZw-bEDL<J<GeiiG%4s*(ryO*r=+cZ!9uQC?f$d}%
zS$7igx&Y4S!%<$a5<Ud3cIK6vlj>7benW(hv5qUito|rueJM&GE)GMmspNs&I(7P!
zO3F#Z9~JSL9olykIK<rYg5RF>1Mh`(;Pgxj78*H8Lh4L^bI%{Wpo#hn(k)Xtz6mP0
zj0l-QQ9DXcT*nP&v92TFB+vrWj)w8(d(i;>8UmMW0SDBex)!i&EC8=+*Ix2E=$<rY
zVrJ9(k6(7$LOkyg?R}!2_&fMbqfy|ZzTo=33b8O*q~^K{wt8J)RY7wWo2xIuYgTeG
z;?zsTcnIo|_rq5ffYNF6=DB<5CRkCn@=ZmB*bXCZx|;xigRI#lgy*GGK(P1Pj5xi;
z14igC6)bN4xGELs2zGX7?KMRWg<D}PMYPy^Q8oAnCnWl@on~5yT{2Rl3|o~~svX_|
zCx62T9t%2Nfv#n$>5h2sO`wdmBCg1SZ&QWx(^^RRE1>!bU<^*}RvvzwYV;M_HnqlK
zgUFaJL$Cuh1ZiZtvQO~WGWI>!*F>-1{3_%A`@^b7+yn%T$h0#OfikPlx?oxf&8cF<
zXP2Q9W`-b;no6k>)iy7HhP>^FDi9cNeAXbSP7b|8=?q*Pms#siR^5?)UT4~_jUP&+
z4<nsZI#vIIEjuhX4QWw=0S<ezN`~qr>?H?KCjIb3c?QzFRwWl_TP20+FT^bYznlwl
zmvGuEdLs4wAu$hEd(`MxH$cp}*QdgE?WPbX_~l|i_80R!+6cyg*$|v1Le(!{K}?5l
zYl(>V!|U#(@e%_~?<ia;8b&6h&8JOOK{vGPf`~ACnuI3#COwfSE?QNrcMV8nma56r
zr<R~L#rQHHs5W6}Izztc0WG2GxR^zRQ1jxFa_+=NyZf#_xUO&#qVxkjFx%@#YHR+W
zG5+_l>F5zSw0c~{voi~@09W9NoxW$33j@x)Y3aC1bhqt0!ljS9M(&w$&x-n{J(Qox
z2BI}oi1jdi!lNIpy_c2P>H^xmJ#VjlJs~k(gfICfPtf?sM>eRC51uHFxshUW&0yZQ
z1SOs&T@#>bS^BONq_Qm_I-@Off{&^m_6)+$YZZ~2Nz!Xwj@+Y3j5+{;n)Bk3DH99O
z<jkz`+RP5woGq&U+8ELbPC}~R2t4XSy;0Se9ea_Pzf@?RE=1kV<qKz!2p!uWzTfl3
zk<WeEVcIH1{-)pQ0=$aeld_e#8w_5d|Ir&tH73YrVRc%6wOMyofq5TlZ)2P?G2W5;
zVy+XF^1~5CR_8RO)XqIiz%m&0To|<k;hLLw=%F`m*b1!eO;+R~vnJn6)}f&1wKQJ<
zRZ4eGP3KPL>66rFOOPorh!R%|B4;+Q$m*$*65n<KFsN^ZSHwzr5)6e{B5c?BfST;_
z9(rvZ{pr4}tM#c9%7H}|gZ1p~H<A3>^9ynMF;r^2r0wpf(#1L&U7OlcwKpL^y$VO7
zbUh5h(FE4`Ko}4?PpsZA11;!@)%O{r+lWh1mOUt<ka0M6LN#fRQj{eFB>3GCEtT<s
zUc~tH6jyZY*91m$+Yz-{a64@_<GO4;)yB=gm_Aj^XFXXWay!ci4!6BfDa&19O-0Iv
z*;iMiY~-*CC9^Cg=~O&Dd`j4T(31)jc)P{;a-tl_yd}Pr=^AgdnY%N_kx3u$T%Vib
z%;?^T9>**opvVwwsJbC-KA-?)_YyNSj4DTG2uQfULP^ne5}5i6V-&0Gw6b&w8U~FM
ziY>d(Ec1ifCEuJmi7J7?9wjj-YUK-O=BI#@)1__34r7H>4CRHcn%y3Q*xsitVakMC
zR1pp3R}Iy+>t}r{Kif=9cDzJi(Vc0QT+`iXi@ams+1U!shTL5W(1OO~Hs0SA{$9KU
zH*Trz?0P!DImn0#Yj@mM+1sZ+R~P&9Q_4o#`Wzm3Eu&pCN6)XeWmWCl;hDLsVukIj
zT3K3K6={KCS|&CD=x)S>7j)3PL+jA_!)3qe$tZGi>mF6L(bf<@ZknSYY!iKvmtkLC
zcj03##saNRlDsEWC>T;6cklLCy>nYpZZogK-g|j=P2$tU{XgH-$`RE-^qCj8?|62o
z@LUD(bFxP!niJ)SZQPTN2G5#-vD0JL4Aa5RX%)JI=|#)|QmYW)lc<#}Bzr2Z7nTHC
z3{Q34R)LW?YUvm2UMI3M9ph(2yvqz(x*>e`2C#+_wygw+s{&05?=Ox9lLq@3Y3l2m
zpBip{<7;v!H@!|L>{4RphgF1Io(TZjDSbJjCf1-2R)(LZzMLhX5=$I?ILh*NAlEuz
zHoSX{2Fra*annfXz(opAep*!OSOZmcN~9IQHII2m-C$sleXUlkt&$ng*m8NRp}}ss
zHbb%sUS0PKDiO1j!aZo#adAHQc$fwizX)1|Z3LIg34jiD0B%Q~#+Fp!8nFAcDZSI)
zgl-0SowKo4;>FqKa}TyUsl?f=tOAAk3Yb%smEFmpFVN80vsuyLt)f0u3S9WmQYv!_
z2!^Tjgy62hiW2v{hm2OL=%Zj1>?2N4d-o0{K+c`%69RxvnRODK9revbJv!|rvyWpS
zWABXCwPoJ`w7(TB6%4b%3M=?Kd#XnfsOg(z!miXrH&e?bo{B<lXzTXsJ^e+&&lYUy
z`0JZ&$tR?*h`K&I#d@3C-^QQoj!14CK_VgPGxFEJ2{Sz-JaY5q6>tzre5QBv&_RZG
z&r;(&M2``FuQ))We_vXLEBC=})uPnoHKW*yb-uDF=k^M#)+y^n;j-2k7}lQOYt?(w
zwIHDVI=;XCVyk}^+I8N{H}Bw)Nin9apd*VXaKZN@zTw6w-5tOTTu6pb<<^VS<o<Y<
z5LCF&XwcL2B3-$I9x}xDC}dbDpUEIuLWR7J!rdF$NJF(2TrSL?B!0Jadq7CCaN+5x
z@Z4tvm){?Dt=_j7l;n4d<Tb-Om`H1#Nu2mru;bbxy8kI6395Iw{p>L1>0{LUPAqHB
z%KR_FzB(+*zT0{jx>H(aXplxyx<MKR=|(_7q#5ZDkQ}<ZMN&Ye5rLsoNdaN#9*_pV
zo9B7Xd){-tbH4AdxpZbO;J)wu+k3CQ*IN3U#tiyLT5fGMek@?9;4(ntcwVeA5L$1U
zsBx43p(3xM{oW*YU8-0A)fD<RQ}CPqD}=+;n!WUKM$a=}sWM>ZfDyvIrHX>uD&jcb
zjcZFHc7Vz0FT|&ohIn9iP5chJ{-HhId#K?LB-XIIB6<Q$3E#Z47fYJ2n70e|k~PZz
z9F8;So{TLf|NKt)G!~AV3xgq-mhSZFEY+_YSO-?ii{b?Jc#bB_9sA%++G4-LNHKBz
zk-$Lu6;g)Z{~I%dl|L_aB9>|a5DXSRvxDXasc#pn$Db|FUIAOJ!YGj~5Po$i`#{#S
z)|GoDEWPycyldko&n1~uI!Rw+(4+nJLdgjpns!O#r4(<XhPN28^jb?$F3ahlbYf-i
zr|P>D40etouAS%5a?so>8ysBcma?MH@2ZA*0bGBQCMqWgcbEyzua#)NCd}APdyc@l
zdElXhG|gB?zlC3GIohU?P!5#@trDywHhJd4F5b^`Gu`=@I0vXVx2)UfSCa$T!9@%(
zlGMZor*fy4YAgMTAZ&E}tL{F+eMIh5<9@IC2kwbGcRn9o(y^q@h7mzrTuDvP#}q)V
zEMpPpKB`RmcSTUZKq&Jv&JY9G|0KgdzQ?~XjN!fnrT+t%9hJuA?=LvVtWSoN7O&~z
zIvYp~S#g6th*$|m)e*kMWf5hY*LW=O)H++a_Q8Cv7(3`rq+~t#ts1vEoMb-|U7HEV
zI`8}&yHny*70|aqhHv5seCk0`77J8j8Z~J!uHd5PI&%=Z^4guObR?m^V-P5^i)4*F
zTsty=1T3&`GxiN0FGqdA2B$ZZt&*RpkpRa9F^@3>mtu6@g=?fUH34<U4*)r;gQ&pp
zy9fkHSm+I86~M5Gq+DCBV#aob^KjLWdcs7aC_x-ac*G^!gM#zrD<AqWvE!36nV*r)
zBp;|NjgAIIjWWVicq1=ku3~<2%DxCo4xN@{<B`m>1xwl_SnL;E2aftKK6$*kDvl6R
zvx*m$SM@7qW+yLkSyxH6Pwv5%K14fKC6FVMzgMgfsYp_dz-LkE)s0k+`2>@F1@dK!
zgnu<ZDWsZh@qvaCs9pn<i+Ivhopls<l#?wXLdGA1LMP_Jhe5GASoPUD8j1M3@9*Iy
z1M-kHUCM*v>B|8XCeTU=aXnSo1J!JjQFzKZkthBHo7)osomU*QoD$m%T>A;a0@7nd
z6-kJCAF%Db%l54h7~c&36FOQahLy<YPq?N6Tg@ObN@Kn#s-{#(I}Hx`%|;_V`#!q~
z_lSs$@8WF`He~~DZBTYNSs{1r8?rS)=g3nuxs+yYIy+A`vB@&lRV*=b)IpI${=U4M
z@~@+6@3z4?QO{h#Z|)3s(Qfo^>^IsBP!kBS{2mT*6WmL`MFT?zY%d=p9VftyuC$E5
z?;%d98#d9?eTz}wKB8@v9w-MZWNa|@6{*9Ju9q(V9$fUnx)_EIpFFnoj(@~BW%}Me
zZyy%6H9=apRvyhOh=Xy}*X@Qn61sNQyAa4B@<UYoE6X{wHatfg^IAi&1p2ZRz4jY!
zd8A*#)OdWrNzsqev901a;ES5_2qSL0lnwOxr>M?R&(?-Il6yG{S^O5BS*L>wDKLqG
zm(m7nkGewCYc(#lP7Jv@vK5z%4D9dvj7gfpJkF0Tscn;ejn$qZJ>_?*i?i`h-i*!5
zQI7WEQkoQnR?ivnw#_3W_hWP9#-?j=e!J&gEkstAs`=f10HRETI9U2}sJ(AR+C4f4
zFM?`wweH69qZm+7UR^8NnF68qlnvn`2F6E2qX}JS#mOqiRornIJfswMbot%h`dZPs
zhf<i78fH;xSV+*Mi*=shYYI<DT1(RfNZ+>=Wx2F=V$qpXb6;8;6={B8Z@Bo*&ljrt
zdDWb2_jYnHTPDl7Zl=a$oD$NYM^taHqb$jks<Y1HU4(I&9vMp84psSUF5(p#WEo30
z3|lA5N}Wg=zMP=9<Bu~&Be&CFXg**E0&p2e4E&RnM4aO>Owd)4>_<_Pb*x^m*0t6f
z9mQr2lxFgDD~EiX8D%hQ)B=T)>ASgRHYr-Ar}alp7pS$+m;N_C$5~7CB@XYH_Eb1g
z5ryBwxCSO&D7wbm`M~32{#wPIQK%TIF0JCv93>}+hnXHqu+Oo2i!+8IDA|6@gy!9>
zo}*gico!?|rrw)KY#EgF5KHs~Jtlz%E=6r0Ef<Yhh@&ae`?#M$n1=2RhLIJ}#K$vX
z7}Fac@_TfQlG~C`aE6$yI+W=%#F7_MH6ue0)WjXcBJ=GHi@g#j<TE0J?yb>?o6$aM
zKh8O5c*8lGk<g@1peNLOJR-q25J{$F8jG?v^D9ns-xfq=QRL`ywW_Qn7Yrb+kfsSQ
zRrb@&9Ok5pkwV<U<MB*%%5z=}4J-TuhcpQ)gCPc2SsL1b>`GfW22&CVe+X1i@0~{?
z3f0{H0Gw{YDG?><CXcpt;ToEg`wcv{W<0#KQOr3yHMTSAWueTD`5lFO{MhLRB;w+F
z>`K?sdO*CT2jXYk0Hou_ljv{MN)AIV^9R*y0%Qdi|H@Q<B`j*Wpf44aztVS0-%jh|
zS7HYa1vbiZzaY)-cSaN$9}1I2Se#p6kfjdQ!9OYGQ(I=NW(Gtb+K#4dVV4GTK5`%M
z5C{i=uRyK3U33L;*IFSePYwTXZDT7A@9nwQT@1Fpru}?!&Ncf*zHeNw5C6G&8-OXP
zG{n1@IeA>_DUYwB$)4Sd8^?sxw;_G@MPU{9QKjS7Mq(~$TPijX`G(Xu9(xb_?Iv$u
zRKpV#;OEWLZLg}O?fnm}Q_?mF6#tAphyXEcth4fO<??|obFE@tQLmn>gEJ+mGP)iH
z^4@5LcPfIu@ki^r0UP~M)FUFmYAHuN4QUZaK^(|uSa@2oYK~rDGg|Kcni{M00c?~`
z07=enD-BB$U}g^QRl7x+MLuK4gJUBZIR}z-rUoYFy!k)NKN#E28`$VHWC$*7EmMR+
z-usOJ@%dL2G$`b6*#@~Pa?PptbFAut&;V+&C_UqfEO&zUhXPmh74p>^Y?<>e*Kv_%
z?aTg$%EVP;eEw;5*m^8We9R#E-tp9XK-^-l1{x`%gdC7eRhPW?lcEnAG8W`O$?VlQ
z85P{J=kn&AV^0WBe$SEWupF*M+HB?)*%6FrBUX@Br6d+q=zN?CNEm5v3@OW7l3@bS
zXgvc&3L;cLc%ys??y8G%dLLRVzpYT3Le(*_nLx89&Ua@Ykur)D&1{NS6ic$-FKB0t
zKZYw1y+cXnSkPXV9#|D<CCrHWEdL2W9DAD9+rCR!Mfs}j^F5%^IiJXdif^F7KE3#*
zeM^5n;G54AQ=n;W$s`^5eG<<&$FbM}Ctg1om7Mm>1mk2&JMay%?tgk3%M~qw0`MKm
zQpzn4a}R!-4jGWJR2<<lK}w@m9{5@5p}xtpR1LqV<5lF&+5Me0H}!P%JWJGm=h@#+
z8SL+;Z1(bdqZT|%dc`4RpxDMRN$kc5c$U(6?`j$MpVJyq2vc$pyJniq25@F|$XS+t
z0zH@b`QEuUPGVooU5*z$bJuPSjjI2%_4BM+NtS0H?pls7GaK2Rl!i`UBBI|kRr9!a
zjGjJclExqfbpz}8e!j$M-<>MX!bMxVCc-N1n>d(j?Lyh`nt0O95Vv)#*{v5dr*;=!
zM3P@A-#9Tw4Ei-~4WQ4HB4iy*sF2e;0Ut}C^AIhvnUHvA`q^HB>>aQPXcZtgxXsU1
zZ_nQ2GBA{$@gng4&hSr}d8rWOEWS2e{rd;Krd~Zz^UZsmIiV+D_YyY6NCI#tRvPQ8
znNvlkUl_0nI@5#8(4#D#HDJ|!8(QyY=kZz7{@MnA99SR-*c9h13T<VbX)5V)_1cun
z$$tG6x2cDB_2=iRm(+RvT&VufJVBG*)Oj@QG{=?T`Hg{>!OnzPG9Q6f0Gmg))>(W8
z@BM@2)SNv%C6Ri;iZ|7Rp0H<4&W^Jyse4LUVU8&(NkYZ;Eyz%@3Cd1S6L*K;$o>c=
zqbCeN!+g&~Z&1PVvEG($nC|?0{amo-b1wh#H6_@{LnHWJ*=%zYZzRR@`RRI|-OpnU
z{*giUTZpy&7170;eQ#_S)~CI`=(F+&ra!X=y+gR$;CY893{E?PBC}yEXF&P)Th1JW
zUGs5|f6pMdJEAGHFg8)837r><-%j0T?vXo1f*AT<=AKo0Ab*}^v?nQTbd8sE+0T#P
zB~0=kYFFSpt#yj@3_Y32l_WMBWR4#U{HxBqOLKq4`4n@|ZtwLI{D+Srwj=KdI9-C}
zS`c9XN?COxeu8jfjO32Uj&dqj;u7(m<P}peAxKloSJEs!Q_8&=@ou+qt-UAUilA%j
z@AoueD73NPPyC8S=R|VO?9gR4X6>3(*;vl*1OWX_jjnH+0F4rNq4pkA${cC8TYxDd
zeR3g<>hcTS100%iJVPkYL^I1CmZ^R&AO^Jv|1e~B0(eMPNtz<K#vk$mtq<vP6z10N
z-4MF7taHqO8x98tlD`B=hQxojcd#m5d;3PN4a~P#;*{abe=zp-(;QFGn$ckUAD}hQ
z%t5Ckx|jj>H1@8hT*V>O#es7&6J9k=R5>ntMyEfe>zpQ)yTW1bDP4_F9PY4oNX4fG
z+d*kqBfagN(jHCp%O_GcY0a_DaWIKTiV63@H)Q8`+OI!0;IEsk*aBzPGQ&zQIC_5;
zYn|l8m+tw^c@|bkV{_B^ZknKEX<l?W1~?+UmkiedrayaRWxN<lDE2BR-fDAcT(q#o
zmdkpMz^Cxz*LEphhlr-v0PvsrDym&e%Or%7BWuq9pIEt<*y#k=Q<{ewpB!%Hed(nh
zIhUL7iHp~ez0!Qt*rom!qPgSs>-UecO}@q;>qBXq4FI9AwXtfB1b_xu{-=4^Mowrp
zPO%DBU#TIA$u7+!e2TeuMKYfXE02Q1rFY5^^YYc}Us=h*&hc-;uD<|3>iYnN+cnAE
z4pNuMuc_!e@OK;gtdn&1-BOIi41Z@UcLNRf;EJPfRpkeh)1M;#&joV7?7y5i8;^Jr
zfao^1rBFP7NTaP)6e5|AC0>h%3NdcMe4TZg(7vQpy7@%fW+d#s9ua8BTsBZmCW43b
zim;Vh{9dhX+HZDFnE(%rb@bp2R#2gi7{Fm1CNao<jXuSKAsds*!3<R@?p%Jq=UjCx
z&++3lsC3g3Y7e*KZw3k&^d!Sa8P=-doeL!W!dMDmEJ?Oy&S8BFS+*uP**THCt^hO=
zDk+y9N?=G(oYJ9&PVVriOzkT$eIPggb2)9&r_63Z^?9^Y3)_%tlcP&{u^@)q=<LIi
z``>-BLwKjD7S{7%Q2Cx9`j#wBq+81a{`F8z8Z;dfoQQb;T^qL6qSrO}0IFJ&aI0%-
z6;^@o%;rt<PA79JwgmV!DC{|A1fEV|s=h0ZJ^CESVMx{Y+1D)6eeyxPt0QmA5-a3e
z8rOPMRQ8D!Qh_p_!cNn<ixx>bu`TTRiB~6;!=XX*JB~oZew*4pAPC|qlAwYcUszcK
z94ahdXYV~TB?WeL<0ovPTneq-I7S4hA?L^Au6wCx@Zk(P%g)EhpDbtItf2a36?2SO
zok+6S`X1G&5H<5|%?x>lG9{2=@H;^RbIv}pM_uv^+QZ9AQV4}<6k-*%^iO5{&&zL|
z7-kUa6o>#SuLW0?q*<e<`UG=d*zWu`pO;5)BvbYmJ?YJlVAN)4Q}}7kfD=B{B|u;U
z>f;}>rnJ=<$vCEW9}-Yrco2|xeAe;jrU{h0M2&tYwK*xF;ADH-osCsQb9@FMof5T%
z?=_x{TsbRzrzh+dF76J)Z}vfYp}OfMiT#o2%)bB^VyhW)Q7@bUw2&VxZ+ulLld+?(
zCanD?)abbd<<}!H<TFaeHFL4|pd{(umhsU0s0a6uj0?NeQ^ebty@BgNmWSmkQV#<C
z59tJHPVY?&w<f#jQl&2)5_w_fRsfZZtd$nQEZTjg`Vv_{wHwR;Zyy0%T)Vu>Zi;Yc
zEBUzRn{*c^Pklw=F{!fEb)MTDdwW>VR3?5T*fqSHSMX$O{*E!*7P+hyZ9j_rfo=IT
zBIrF&Se5zYl8$Ibje?l$AGCM26JtV<9-pkKckQV0rX1}SEurrlL9AYyml+h)=(XSD
zOwMPhQny+z>R#qRtEE>ElS2F6wB4J37#9m<N)I9&A{e3tg1kbd+CCg;A1U+}(la8)
z4odB}u4u5wTiQ@5HT1Njj34SYQ(X>uE}eU!mlP{Oei}JSys{l5*3Wxky~oxLCZD2@
z8l>#~`c8*u@X;82g`Kle8o>gjIXpM4Z<G<sF$zj#=5*W_ST=l1Aaz^*i6rc7lI8qJ
zS=voHfLhvlov;}2aPfsbOH%ikru=Rp!MOghb7V=(>So0t_(pK#9)~ag8u1qpjP%1~
z2sRP}rGE1WcwSWy$<BRU^+JtpKlFd_T)oPY{zyG)dM4~=F`qI2m1&oG3KuCbhi@Z5
zByQ-3!B-pynrZSczV=Swyi6Q*_=8%ZuEm#cQjXW|P?u3&8Z=bSQOT7*U6cOJpx1=0
zVo6DoZUlxV$+B@%{~C}y){qC;>XVQwvk`DwLcfudhEnbZqz`h?IMfOmU`e%_EdX^c
zv$D|&aE4@qRQt|SKehG{wBxAG$&2|{PoP6XFd5KwUp%dfl!X^21zSn>Z^w42NZbOw
z=0@B`vFlPvk$sSTimaKetSu&l!P)IV$MfJ^>hJD9xSu3_u)mX=1Y{H(mQWB!f#JrH
zn4ytI3OZ6A^!fI9s$Snu90lJoznn;$`}`TuH(mdv9PlY)!IKeGxS$do4laL@BJ(aR
zh#1$=05=FYG!np>%U}J-K)P{&4F*x^au<oPnk=r8)WO62+Aa?M>R92M=aZ-7DX+r?
zlEtEOcef|1_R_HHVTo5QB%q8zei_+<C`KYZCj4ZbYBpBtUIEEy?Pw6<8f%5opHaEM
z70p_BQ3B}Q>I+m_Qu}ml16KZ{ZuqNw(24E@&Sa*}9nbMf!f62mIjqlsz6zM(SHgk7
zW2@S}orf4;NfVhhPVKt4Oxl5$lQ3tNLjn|kxFx|Qb$5b|z@c@1_e`VHtl&eOxYiH&
za17`Jva!65>oq=6CgRICNbB`E?y)`^Q1l5r=W0PN=ztBSxL<PLU*s55GRlTo6|Gxz
z>p{3QD)^T1EH(57jh;V-ok{tq<w6`Kj*_fzYl;xzW1+T(L0sM>3x5Nz-%pmZ%fa^G
z6zyMy9D`N!X5<fjRA(Q`6qsAa=MM?w%S&=#ty8uVb0#-q8<Dr_<pj;(x&}3e$mM(?
z+8Rc8XKq#f1wyNbE<W`MKM5jTQ)*+1{jyOZ8RvNYO8#^keUR6gA||6I6mpvJtZiGr
z$+B_iBkgYjDeSd(yo9KGzsq=qzCZM;nD=2U+81;wgLH<6tr8%ZNIYuW1VsC0v44j>
z!V~vMygLlU{`+AN{Wxh_cf{ygJ;%<~+H4blu-?U^iu~KG3ca0UWpipT|D{Q!@$L+W
zk#YhT2|M`#e+sP}qe{la*@w4LU)7CIbd(-tpAKFvD3}K==csNqs5$)xa!A_NhcPpO
z)5!>%z_LFPBnpAyA3!j|E5T@DvM<slWzIjf4!orjv3nPQS@m~>I)WJ~dS*>f5$l(q
z*X=7{!)Rl+xY5v)c*2Yhm8mNay5&x|(OVS5b_d&tOXBn%J7Q3x*Y4o|rnwJ=DN@Q2
zg(OFKJNCZ3+^El?-E!vzMnlYh*kR#szrhtN)MFkZ>JdzLW0YC_P7w>!F+2KQ@}%ne
zS4PHjCmB6H>&-O<+0b)c;!XSI(N0pi)1n)l#fF=;!L%&*xOyd8vpJQ4K&p&F;)@vR
z*f&G3<Yr9E2NBFOi{&31aBU4Q#n!PU@kb-t!Cw|anfb|a%9%#hb=He8+6G)+nsgu@
zlCeF`8-E{Ku2p_w-pnE5mQ2c*qpTTTMC+IgB5Q1Z{krX3L3y)W-@a$ii?=(iL>;5E
z;ED{Z5bJ;=tz2PR=`flsB)r;Do}QIn>C0ODwpfwDkCfYY<<vbN;xf~T+hm5xH3y0?
zIykKS3P%xy9Vwv(_kT$9DY!k|pq@Qjyo0KNxRUL_QGEwt%|4@ZRB~dfRH7kL-X9=0
zn|i!Zu%kCIo8Is9!ye9r=zTzgSD1WSqWf|Umh3j!&bM}lCmqg9X?{@IU!y{V8tAKV
zLW|gXFRCV%V%!mY2>K)}p2v}Tty9v~YKl|EHb@#Cr^rmhsR2ADd1cKN9y0`V5)T1J
znnjiM=A|3JgtQ{fHxr55%f1hXhsLO<<f%w=CkdB)js|4jf0pbIKZA?LtL99;*fUQ8
zDi8$R(Qu$vH-m^~SHqmwclnMFsmHBELqGT`zAY6Qz*!!in^HG8p`U!!nCPjvKmAx%
z9dh6xH>4xErHE%5t&P{~Q8CTzsfPc)5T{#LxpD4u#UDM(<dMndRF%Dmo6^6p&fgGS
zTqg;5E{bnpS;(Vm7Xy|quISo6U`lG%1y~l7*&*)3^szqpEW$RFX!45?pcMC)9@9+3
zDFgM0ZYCDR1W)A^z80v$a|$`m7hM6#5&4L0&VIzuUP4PFv~Bmh*7O8DVa^{FX%y)d
z`7FZr5SW*?xGwmw74=`M;ztwMJK-%GOufxVC%qU225j$qo@CarOx861FxE9UA7V?8
ziHubU>x>-f;MQ3}I**)?3!;dmLfYlRq$)p2h6aF46a;kgYRM!0s2lU^eCD^aS`$@Z
zX*6F%0bw`7iI@?`#IbvW(sSia>T$-0;>o5?YD|+vvlSd%Fs~@4H3BIU&k6m}nc`Pv
zLiu#)f9@udf>fjE$A%Ll{yeXSUB2J$XU3O$hJ5qrSAp#L8G9CLW>DgnIl?@;>d*H)
z7KI46+OB6*=wW`V0c8RjVg8d}D1SdfHQ7Caz?QgTqmJYj-)pdu=$15CJm0-{Ast0P
zH9k&>z87~I(PnJEy)ux(6&<g*U85Ra`lN6VSrFfx(>Hzk34tP|SJI$M$kNMZLQ^K4
zEXsx|if^i=?ag4PDt<dJ?ZH-XrP6-ygx0gsoLz)Y@ymq;50&AO$7B}2z-LY}JG>pC
zy3V%YN)*S*zyC70Gg7NQ-j(0KXs}M(NVLQTfR<18L{yM6V!|`!ib#4+u607Do-6`B
zy|UTCsHLq+e~MM-D7$AD1SYipo%rU5Asq1)bq+-Zqjn6T-0B@-n!YBa_%hxYY)T-n
z+g1$P9soTNcs>EmC^x`#W~bJzO{TGfvD1K~F*@I$58wpBX&L#fNj~6n*h7g2dR%_m
zv@P1KXwNPHMq1=O(Sku6j4BlOh&jPn+7t#zr4O4Pv+@6w$5M>^X;BeIP!O0d$~V%*
zSis))lKwHmn?tDK(Qt3MAQ&2jN3m$!NRdGN9=yufi(7^F;f?L)n@L<;K5s@4(-Vj9
z_Q}(>%(;ZD%IqAkp^pU(jzx!YaozBjKv$(!#%tq3_a%W|p*4@f-5DoHgQMc*lz3G$
zbMrvwM-8z`WX%8-`)4^RIme=V`J}U-c^z2WX#&$p;_)@-RWfTo0Sy2MGsE0OFPSzK
zbz+T_J4%TF`T|CfykhF7BhntV%Q%BcWz~s>97j^n1?3IY7MT9%2e%&sdPjnGJiXo4
zz^t_A{CeR($Ati#lTRD0h8u)ReqL`YU1Jy5>6iZY2(@i+L7HnLz=$uK-WP*UeacU9
zoVLz5hs!Q6mc?>N|H#!jm`XrsL|@fiDAIW-=fb8vlwdIxG@~M&ITWA%Q1;{l(<v+W
zI4QSLJ<9+<{}S@h5NvA|i6fz#2t|?DyQh2P#!U%q-+PlR_*M3eR&F3xG|3>&tqvvg
ze`o;&!NIlUvJ53oQMQyTRzx$;U9BD~WZ68ae@4Z9^`c13HNTZSAyovg&Nh@#B=Uri
zGl=!Kf1=ljgxT*H&26v8%*_HGjQalQNOV<f6h^u^jqmPX!G+=Z@4;36zXTV7?<40{
zZih;Pmoj!|W1fy@bj%wibuGK%9v^3X87--B`QCwW`r%E6_mFKPh;6Zseg%l9M10Ml
z3+(qtrBRZFfe=ZIOj8eFktuFrA6=a(=N<qL%Xx&v9FrW6;}@~pEd&jT^0I*QB`1_&
zlXQqk&x7B>L_hqChTDL&dAaU)lCd@K9LJvHp_jZd%AuW;Qo~&1Xs=#L^;$lvpypIJ
z<`ZYC8S-JfCM1~1Z1GvN8cnv%L)E4}$GLed$O$A)Rsmx@1{B#D^bBBD-%^Vqo4sD=
zR+p^;J|fnmNCps1(1bt1AO7V@*et14$tD^Jv9u2;EHX8Wm9}hQKT}SFH>N778=t`y
zHB50h(v`10XN1{xQ?LI*a+D?g%JHklA~}{ut&wYwZTgN~un#N{fH@x4y!FVY#wokc
z|41RSeBjw-z~j6o`G$=<DYb{(Uy<6u*Yc8<6Tvz|$RUZZVR4ZAcSCE32+)m#bgrO*
z>}$Erb??h0x>GMmIV`#^Hl^nS8xJk%A;{3QZCI=`9R}DfsPv9WqgSS>4FWT9=f#Os
zJ`0Am010!zZ>Dr?#-S*77xAqm280v$fkD|ZenUQ)^Md;A8Ak7$7%(;NKyopHw5Q%O
zL^fB~r?`F@ho+rcXAC8M0d12^D+}I$h3|Z*RSZC^k?n^HVBz;##Zdcr%hnBUOH$H0
z3{sf1^+d5w=*Q%VUE&qTHX}q6Oi(Tf#5mS>$^f>W76OM((<qMK+0#~aX$LBjv(C&O
zrXDx;ykv1@K~<|{=Q`)w0fHK8j=@8O?um4#EgcP6JUqFw$;ct=``lT>O$2Oi0dVgg
zK2M~xEBAKmg8+7Kew7^@TQ1z(3|n*3om#K12-Vx4uw|hLj&YGMUbC9Ppu=-agNU!x
zG=R$KX@Wf>qN%YKQ}n~4I%C=uAF##5SoG8wb5N4CX2S<mY^tV?*PwRExY)mg?oY55
znWi+5^(eHygd4(H&8srU^4Qc_j=oW;TZgtQww1r*D~XE5lMiYB$tr0lS1tdJBz$;*
zqwyGY2lQ&B4&fTjl@3M<4vlAUP2VuaGet9fO2tIJUZ30T4b7#?$dK_MnadneHPy%8
ziawG9DNe^S1n?>_Y4cOr9Z3WZH6pxCz?P5;2kWsZk+&i>R$XRhFgM~FR-a|{;#&G0
zTZ$_8SDmKh_Z0;e0PUM_n&Xeno!4M{(6Wg#$hNwihi!O0m_flLWUe(9Gj}G8gKvH9
zg1%wj*Z=3s$EVvNb{5VQt<WCUGW;29hSS;znp2;&$L_Z4Z+uQOpFW#OS)KwCGAeWz
z6IRSC^gHsqFaXJ6I36U%OV3?2Y%697@}$~LqBcY9j^?*o9cd?gFX{js@Vxg%%f|4h
zyv>}?cf%o{v@>$gC4~rWh~91sIU0%o#AjJ!LqqtPsQ=WS>r&2101AvCQc;qUc=Q$6
z(OdGtw6k%57$%MwF@oyLvjpt~BAe*r07$!LOXUa|X^Ie&?zdi31QqA9EcSdxZN8+Z
zKg&_uKjYjfJs4KW%V__3CsZvlw-W3sNkJL&gbzr2{*Ce0ms!@cO46(Ii1VRtx12x$
z&1^vNd%@lA;8Iw{at`O35~5?1^9CM~2nEk9OJs6jITMrEK82VL20D-s?RxXsJcSnw
zysBGjig|Ww(mM`~XOTnKyPX?zq2cZRn0K=aE&93Kyg-gklim&Ni;^G(`c50Cy)fm#
zv3s*^fW{97Cu2xhMgAVKlgyj;!S<ilxAOsuyYAKfj_|sNs?>Obwez`~Tg+U~stL%F
zq0g{9!{tI9eAc)uXd3nm#vVvThx(7rDwu_&t;vzspS*#a{K`iYad>Iz3)xk)C2-TE
zN4D`J5aK>fqmWDztb{)yVe$F*tW$K4vt!ETl=q=_a&UohRBylmlNlB)8ds13>(r47
z{}JMMAPLt3isgrgU28!1-`USn5*(x5BW*m8b6CtS#fi{7gMv9<IxnP5c)zY1B<{87
zg*vm635gRiF<>U7W37J1agyWu^c7gZrBw{U1vI->o#}0}lfa|Bt9NF7kvNmAGztem
z<(eZ`OsoE+nNi*XE<e`G#+eWrDf|m)@;|zYOC`lI8PF)}-O>dhK*nFlg~o9aj7*SO
zF}|YkwP)sGla*eIBV+HI3SQ1lboNTo7<1p>VzUuFf=U!awN2TmRNuX_dCx}3mWN1B
zWT-p87y(w&9So!4)QP5v3==Hhdu<ZZLl6qVuCx~z+;9uz%}}omr)3Bm3!87T_my;1
z%srtk(Jvrzb*}MH3h7i{=WbefLU|d}i#WgLm#`ScnP2i*eo%E3g9;5Iq?O6=){f)h
zMHEu!haiHWKVmLd9C0p?Y+TiZFSv_Asumg<blOe<GtgF~6@nQi86}lCJ~M_3oCHAF
z^*k^+VIMk3D~1PRQN{iVDph!dOtqM>0vr?pZ83-71)qP5`ju>NaeLx3<SQf$#!r31
z<EMb}C`m;P{TRHCffzi|gu%Vv<?pp|GmwTmo1?8a*xTygZmJjO3(lH`S@}w(T{V84
zoshjCb2vj9EeeNH5@S|1H0=m-7|Xm{$g>Tku?F9u9lfM`3$@AazZ??fu(*7(0{<l(
zY3h8zJs9DizbcdPgb*M06z7C>IoL@HCTM|%&1n~^nk1s~IB(~pq+T|FX8$wq`s)QS
zFq@ymbRMN^uruBZ6`Y%+*T5HXVnB3nepSlG8FE!>uJq3nUx~WM6++oN>&F(De!=Q~
zZ@m{$rR}#uj{XR1J!>dSX-trYY}ORx+x7Q}CF69L`j0UKC1+)xx(t#b1hCQr6(Tmc
zm&b48F!CnAT~s-5p9?aS)3{b>HF({|YY7?udL`RNO_h$WYre@IbyQfX639&KD;ItC
zxX3!Le*ue5iDAsDbz8rt{^D7RCcD(=)r?B|OP%=W;ERq(FeTL`)rp#MnC%*qw!X-8
zitFqyV~=;4XLZ$Hh?g~9vE~S4Z$K%2s(+)dlg5yMoyVtoG<Of>p`9`HNJ}ef!LsLZ
zd`#mren+62@FoI6l$gl)aZg0%)ZPiNyA$xaqe`$pu_xE?aAmIgbjx~?W6pAm8ERZb
zBF`Yl;Yn!){Y2ee!6oSb(Z?w>eB~<pj8w&_ReyS*w|3>DdK#~=l=ZA3H#Sf3EuGY&
z*t}q~q{maYu!lm6me@bQzvk+gOVn(0@-flTS|lz-<$;Q;XQ?lto$zL<Pv+KjlC@M!
zt;Op%{^n?6D}3K|7hQct&VGm*AG`2@(B8rOm5SyO>X=wuf9R^vZ(~K*Nee-ShtHv0
z`E;d{J0X<U*n2_N&Bh5bI}G<1c7PLCaR#NbRR8L6??1KJp?11xOqMtF&qJ!tMGSTp
zV&KG;)Cv&osgL&SJwtAOJ2`f%?{c)04QF?rZ2F|n)(Onyg>i;x&`5se^h<>*Ck55f
z+eikFV91*#fwn+kSv-wLYvb$)lPH>dj7*lFzvcITyH~^*CV_DujLr~-&6cZCtmA_W
zRuj4yz$L^H(MaCHIqobrHi?Y23>xRBV&f+f)U`onVIwH06Z&u+FC;Vr6buy|S+o@u
zwm5sy>VmEXGN7`c<c&_yQIJ|m(+EDC)Q8A4!*U<6BeuDxMVayEp8-j^l58amJ?R0)
zU~}hZ78eh+6)gx_JzkV+ay<s(A;}I&ByP;~$sY|ZKtcwC-BJ-+WqjXpz+(T528z@F
zF+0HrT9F^A@=CT~_*;q^d$DJ7zFqKHP0RkHtk*>>_Pfs|6-8uugYt<h>YR3!XhNL$
z`;GB%GM42t^@0>V0RB(u6N!3Rv=tpZZu+3*40+zP;+OO^?g<IhSxNHP^w85JVoIwa
zOf-^NOgD(4{s20CLczI0hU^!Iw<SE(h{?PFZ_aoM6-b}Rg{)N&ibFI}XMaj+n-#5)
z7Ui@Uu+}z?^9<JfMY~PeYP9$s8LObs9=0C4Afq!e4S}9fLUv}~_cc)zz2L*B8bI!B
zi9irhflO$|<RnI?2<%Z}Fq^}Uw$w4VPF63N`7~y*<=rWT8R>3ZlBR{-_p}CCGDZRC
zsE4vYk#S5y+5wy*(G20=W+cM>3V7VF<#1K=osf<_0h21Shs^wes!?Qf0(%*t0Mvk^
zJHbHeEsY(n9q|_(6N_2-jH$GzYu!vchw%Ym8@#PB*~fGxl3m~u?ky!;wPw_qH?Wa*
z_1`wq+%VXH&NHpZTglvp)E|#qQogC1T7vV7CWDIxo0jYAFoz=Ns~C;Qg~6{19t7NY
z@&*we2y41^0=wc+N#1}nrEIQViT&0nN!KJwF~PhS??tz`Aesc3s3B{2g~MDS{+j#`
zsIeXnJW|knKUMo7CiBzvBg#M<8(CsqC6LBwQG``{?*jSC{f-P5If7*qB$bq^8De8v
z@>yv|g|E!NYOv%BdS&1BF1XMbX?}!dd(=KAnD31^Wl{aiQ*+8EUN7T^l5#-Q@{^Cy
z%FR??`ZpXqet*Yyqp!dA!nmf{a`qYRz=~sLe7#tvX(eqpN^Bb^Xft$++z=EU(8Me2
zV^zm>8sS&x$^?>!D>RkkpZwvrwRe?cDmc^qw9`I&!r(=OWI3ta+*M_FBM`jj|9}X4
zz=_)5Qq6cH3xoipouqCW7>yCik^6IaQS?5gB;^lKz{;UStF3XZb)2QjJ23g7c{KWW
z<rOKZai3%EeUvNX6*D&Ebtf7#MrZq+%2AeP-7Zt!kuxO@W7^8X`;m2U8Wt0eGuF>$
zV49IilKv13&FCG(1Un=J$?+04a6IIHA&N_2=$In+CSH!P1PA?*B9<zyhv5fnMbKC3
z-ms+3Pip9lgm1i1fk2<Ow&*t@MbP(UJaGs9OEr)by_yP>Z$eom6FZCqBhe8H%TscQ
zWvH?pQ+ioZRxLyqQbHa=jzPk<V~Xt?)2(Eupx+ws^Y79pp8*Yr3Em2Io5H*_>ur=q
zzwu>4F<%Qxm)_>w<_X-hqAHTb;|4*iNSw(EWPa{kMb_qUy`ib3&VznuU(uQsRfBMC
z(G)R~7jWA(ijoH)L*I9v!JWw*%eKX<ld732!rx)6+A+Y@+?ej!aDA(!aQnhy#>MdD
z4GI@8_ibZiLe;my8mgteRr&*B?ov?THpT^-HWJ^K?XU{pd9a9gMC2@UJ=9}T*-gC0
zaNll^$_8toyFQ3p_2P*(6fJ}`$@LZvIleSM4X~Krbcd5VLQc@s*m^=uQ_{w(_lm?^
zI8J$xu_A-lP)y{=`;^jT;v3x;8x82&wwc%z$8r}%^qOcZcH6`#Us#srA}_3rV=0tz
zX@h2+_(ZcfZ3Blx`N@3MV1(4d5kye(JOn6P%o7+SOAQ)F+XstmD6c*Lo(lc5a4iK+
z0_H-tAu1x0ulNr%(b@#v`C{Bj&gX{EjD)(g)87hGUmwE)E{Rmu=jPbl<;XzCenJ<+
ztwESKW>PdF0uJ)!o$8WEv4IljO=}7}Em<kc1UmEXO0WniP7w*_AWjfBwXN2>x-Zn}
z<VW{wrL^&RD^|HP-{Va^iR2qP!x_RhYMtvPnNQyRjtPK|4qq``@^(LWGKqMR1P#gh
z2Vdh8HCWU>Kf}k-PYSEXC}Z=KD6yyPHwU^L*j8eE``m@lU6-S2mldbEqYpU7i$?dA
zI&!p+!XU26Jtm%2TxcYG-n+^XpTsZM-^>0P)Oxl1I3(`DbPX#QdV_~=G5$6Mb1_t?
zR}52t?QQ$ES5!^B4bbj<^=&N`>}_A;aS^ETCZQaZhZvIaVaT3Ta_4j4$2e{hEwM((
zCcTm5i7wlwMP4$;OHN2a>zQ@y50InXMV23Of#@4`<=6I3TAGA=PM&QPjcP2ta<?71
zW5s11hZO(ejb`V;>QZ+fN4K=tD?0yKYUq_PdM>gThe>|o3lA9tvpCKPxK)#b21!sa
zkK=-{Jc<9kiRFHTVH*}n)=)^%8_nyYDabNmkWAJ2dG>jhFl;!_^S_jS|7<O$vtu;a
zIVbBjp>~*bd+x+!?+AIi;nf8jwqkK}Rjg4M)N8ZPH_ea2>>Go)UL5@x<Kdxv$3&#O
z7xU0VO6CjD;xTVcea*afTXRWZD<l|60ZKoG`P|G`prK0~|22hg-0ErBL@}DF2#@fO
zO{USt8irS`l&d{1{yQcmqX_cP{!g$LCaz42hZ^}EknIROU1WUuTZIIl_aR=Gk!}Q*
zO9@sUNCDXWD5h%VZFBuGuawhv)gb-&8PC@RxrJB1v=?WK2<jRSp8nK$#B!L$0$|JW
zBAF9DfL*z=cv|EfCRY@{;3lzRVT^GMU89m21u1{E=@-E^!BY-5Z3-*<U3YzAp^AA<
zqJ$rXkx^R>Cme~7DQ3T%n9uf9UpN@Et16`VoqQiD(`BDacC6FLOjV)cj$YE*(42%X
z4#xX`OXw5bKVz{;N1UhF2sm3rpBEt>B-FJ%aF`}YqJu-C<ZPvPY_@DR=^7A?=aI4a
zF}es06pNTgM@_R7g<~JN_|ZG22g-W`q-!&s9ve)2aR`MA`^(P6b49TjRP(b(L@xw<
zC@v6}du%smxg)L}Ydzz=j-;pRkb4(=FVNnle8CAS8AVbljn&&XBE&jOuUbZ3Yflw~
zMavj>N=N@`uJ6Qevxf4OFjCwoeD(mww80N1P@G@s9q#t|Z`_Rs+ghZ9sph1k2a2*t
zE_>RIHv6$^!6;^W^nS23Y3F3v-TvsmmR%_^!5R!1F%EX_3jq6919ska;P0Y~ta4=I
zU$%sg8w{5c{O}^yEIQoy3ljU+_x@==)=7Z=hkGRi=Xha5T6BQ!d^!G0l_Z)H%sYOr
znRg)CJ-IJauY2-Y$E7~=i+*OqFhzY9VXKoHR9c6VEv<7S?b_St9BJ?gBpS;d@$m?t
zuD(NgygCzbv*L)ze_KE=7$n(lLOU4g<R`i&mpjC)ya52T)GcMCWT|kdz*zsw@v%?Q
z#eD#dJB)p<Tg^F2f9Iy4o+i0SXvWtaib|Yr;}}IAMKZ@l0?Ue5IIgk`L#qI%kp6DJ
zpnT6<rCy;G@af#x%L2&P0h|eNd!#eRz<g^76~Qs|fZt#dPSv2fkZ*oSaF7d#4(rY<
z&q2}^cvGlOB%{t_>?ONbmW?K|u~oyVO)9)m()SJk_~ZH%;X?4;FG6OU%Cp(o;jb}n
zwX@I~xi#TMnM@G8GKXb1wH2%kbE)5%W3+uO)s_~QtZmMBK^Tm^8gTSz?FoU|j)?1o
zNJEPG!X0v$rDXGH6iyjgF__8&xUg#qm}Psjz7Vb;_eU1Rdug?|{CHPw_n5YMs#3a^
zh0w0RF(;pw@@l{SRCjk{0DxpNy<YP($vB%XmVd|_URB$Y0EjWMV@eVU&puw0CXwE8
z&vu9#SU>-+gVBXHc#bPk_1H`87%7KvF}DKTC7W}=kCzJAr98md!k-|q*#X%13;-!Z
zd&i~W_c`Kx;@MAz0YhT;nC>}e_I=<omr8WpUhcfK!TeNhh(g23B3Yw(@+5y@zj>#c
zgD*^X(W}bBWn!tshfQ3bL?vL-pijXws<~6_^7K>3Pg~XPoYw8s0d$0pV@!FX{bz08
zYCQkgQ2nF$mc_IW5=;K;pZ&DlAYYWtpA~r(#57ph-cZ0XE2vH&b4TO0x|HrZ?MSJ&
z(sL@$xUb-&2tIe7y>+`Vl{x(4=h-o1(|dqkl`Fu&jvC5yV1@-|)*coy^Dp5HZBcf<
z!1<3Y(!ZYWP2hGw{Kib()W@2{SPf?>e`oT~H>EmZG8eRdIv93S)LYKVI!#vQeCDKX
zoB%?f4^(xuG|4T!`fuI&e;m;~7`B%9PM)5CGxY4K3rbo*6?$yml-Uf>8=djxX+DHX
z{>Gq5z#c`)%=E_d?9J$<RLQNOA=bl~C@^UDLR!CJ!uoMgln^b+qi@g^gWER%Q;R%=
z;HR8zH&i?HmutMQDGBcQ1XRy*tgr=&Ze^)H4agVvtuH$%aVKW>xF58gVbPFc&ws4P
z+)B#3MkZ8QD2aUBYf`nyW3*B;?@Fq&P2)BaJP(UWZ+d&QEZ-9^jdBH;wBgzTs3Zl8
z^L5b<nmfu8v<65F2f=-JT3_>G8+`eHEa#<3dB5H{93Z6-17SS5O2r>L0862V2%i`8
zQthkv52#zUG9Z9ffx&+3XzFw1A7H7BQ1EolUGN$^0U!NyKL_MaBcXQM+!vEZ4EqQH
zclrGU*#=#)2l!n)+&N5O3|JT*6ZITrk`tuQplg%KhsYnpkuB1e%%IxT(KG$g$yAxN
zu;k@<Jkch9xrI}ua$7Px=ll!9#XYPgzFU%CJhf76av0Q;#K9I>{99f%{#I(VEls*B
za23^;^Ym*MJoJDdfN7JLN{)TB+Q88rL15=XhJOQ=07VT3^9aoD8)5Ai>|_s4E{_3p
z78Bc*!;Dcswmaarbks)T@ntq3SP*!4e~xJB%i&l2P_NImkCM-XL$UE?nqPf?1mF!H
zfoZS0g!b%#9(!cb%M7ooXZ}_hhlqHf+*zFmbjGt(bmycMwFnP<m##kW(Y^WHjk_u}
z-{EFUVAG=3%imO{vC6s^zSp6YO0;y|D;lCIF2c)yx-Q4!r4Cu;cg||}2DbJIoQ@x8
z`~2{Z#<8~{_Lp9L%m6GGGv&=wcb&G!2s;%#n5esXXG-G*xR-Vs6rODLjH&$O?VbHi
zPZNF{e)JOH_}A^lu8B8_%vBqbiLZW89Tnk?I%Fu9LA*eC(<kh4*dqmdhtw`x<`k)3
zF?BQg>V!Ly#ycsChhQ(+YglsGlcI&TYG$?SBx9E!FkP4ospVQ#I-(LI>Rp*Tmiolf
z`=(F4pBbec414Fz3XSTIGGDzLa=Zzaw8`AnHYp|k+Ww6c7PnXA$PWC5Im3K2L}G7&
z${7HD2JB;sPJE3G$rZt(8Bx-S&UG0Z(4zY9{Pn`e4`IeyB<JI#>9`o#ATjcu<#+#?
zr-DUr@)2nf5>I?e&mh0cRWPewiL6N{A)5x4D3;{K3lGQN56Zdqi$*afKjOHTs26w`
z^yfdn{dXz-*Kv@}iNRku&=N*tH%NAhP?4S>_ya%XZ?%efk4IhH4lKk`yMmZgBq=a)
zOj|>bqQqAKy=Mm&UV=`gP?>OP7^WbY)ZGOZEf5Mh)=(9LWANWve_)1-I78^jtnfpd
zCHvB?K0Jsu?K33!rO7a~&G6FCg+7(~2V-n3zCUsm&=w&?eM7v`Se3T-v4n@0>k%Y(
z1A^*Q4f;h1YfuYIk8dP{$8dR(B|mrO^M2;V<==U?CuPByi6riN`3#3SC-jOW1855N
zc+}U79!Ru8s3UxXiJ1$I1>5<E3Vq5ks-r9v#-f_<x5UbXj-dw@(-Bp}zgTV|!RZRO
zuSg>igtioW@AiC^L9n0{?G!ORWKj#Y8*lsAq;r4*pFP!V-eG{Ag>W0o?gJ0c<Qg<y
z*g(s&Vs9I3dXg9iLL$s)u0=LhCnq_gFx^b}MSGFun4{{NQLqFgN1|tq*VE2s#>bGA
z*cNr5ZOradGwRpiyaBtM|AD)*X<TZyqn~HEo_^Tkm=gr|>!q=ex!*?(VH|l%ii+M!
zM`pKEGl}|?%0kq#-r8yg)w%;h6htN3=9Mv$Ig4rl5F!n*gt0EQFU^6jlDPK-5OU$%
z1#(fzn-fJ)m@Q65*1?3!A-ju_^0`d`s2+D!PG8m0;ZeAL&vc(BOzNSvRIxJ5JeOl2
zIbH<FXO{su!<$EpJi0^(JIkzInM?P9fQj2zg<<x@{St0)m4`{Is>Ghu^;EyHQDPYU
zI4f$eI0w+V4Z6^}Ki2|E5VS_`x4*RacD+CEGdp;&L-xn-EMpZlgOtYSFQ-c8@h5Rd
z-^ToAwSIpl9lq5Wt7Fg(Y#{-&<>XNWoj)D&#X8!Y>(Hfg<Rk)EVS-!cfN=B}o|B@n
zgGQ_?d@}zj%zzh5CXDnjK><0(1Tgt?3dXraMl)m9IGO8T#N7ZAj0IET3}b?<T>wO2
zk<a@w|M|Uts}7%2p%G|OD-MFgJ=dE(zjdtm&Jhhp8dD>oX!Q<pUzh^n7u%G>0@Js1
z)9sY$F7gv!yNQpTaR8MRe$&;xP}$7HXG|mg=!;JL`QI_y<HuW6Wk2}ZPxyKou4KF-
zYm<XqK#5wnSrp!wU3Q_P`{{$UB{!{NG<uCt2?M1=$7HTY{;fB~AXk#mR}7Kl=r8a{
z4&Ni_HzCkKotm+3$zQnIwWvM`yy8y92SuhAS`RF6!Zx`O?6zdv#Gc77r!*t;eK0jf
z%Ni)+ed@m+v!ETQ`V4H7Wy0God+XKdSVw)F3;p$6P^TJ3dZkl@wLJJ!jF)fK<c%bO
z`pev_Tmi-rfK5i=ulDcHrzlW%Z4^06Xu~%3*qc*19XIuAE_(&7zSaR{*XVP48qv4{
zZ#&c7hlj&}XozL`5^o%_-Bu{Svh*O>HQ;J~i6I7Z(h~EQB!ax6cDF4Xbly-V<$S10
zJK6htI+W_9=;hCm)$Ply=5Q6$!><yrj63<&TAk}OTpuzh@V`}6%CgzD87g52W?z+D
zN^j^;KN1nw`*cBM1H$M`#<EoaulhY$EtKq>i6s!@g}yZCXT4}K>+6!NF70SRH2b}5
zzZxpNxPK<O7xE_9+p)Gl8oD&80+m=17UJQMyDcW>3~$;i;!5=&wCA*_xR3O|k749|
zj~@Q=cxUUr%p>seyC8DV2D$&ArvI`$_0JB~|M=4R4X5?;dyR-o$emqBXA&vKH(ths
zwne~Glx~Z(kkI7e<tschot$}Z#-iolBr9maJXN#xe>EbNVLmwynDZKKQ{Fl06+^>T
z*6(qG+|$D<{G#?s0U2KCj_Brr)L)yU{J3E5m^om|6?-lUNQSDdoY>O?{myH!WuC(u
z(|>QT|N8j9=qmsHUr!W*s@P^jGWYlxUwaTb{fQK%YF${qhtDVKl+e1CYT^|;x0J_>
zad_>HV2mK}!u@p}gieHW#~d{->D*D|nt8rm(!TRe!|oLk`A2HpARpDYouSw!sE9M;
zznlCi#-NSvJ3BINi+0Iuz!~LN!=x5IP-(vY$rUh4nof0>!}w~iRQ|_?>wgT>Os3bU
ze)jnvQ^2Gm$AN3Z_2{GBc3S&MNkL!J;+vh?^~{$<X2(ghN+9`BP0^ne0Y}jm1Y}=;
zx87=s%-X(QL`F=%V{;f@|L8ycZgyzxMDNbSBGbDSKn3{jR0gbo8UDvT>Dm1|wjdsm
zntz^1FS-u_qDmi~kF%Okww0H_et_t*k{2LEKmgQ;CNn^8^1Xepa>pdYrwInAkTQP1
zD=qHZ0xV~fk4wO!|1Uc}Un9dNCnG+=&1Ll4ZqiL$;0!y(C{|hhFDt?S>zxAbDZd#8
zz*dI5wYWv4{i*}#5*{+hy1(d1wNXs30l9*RB&|pNm&VY}hYx@syeQ&5Yeywvf%6CY
zu>HXQ0Ken*cMW#G=KR~F8lZfWX+^T~(JUN8{(4%Lw+<$O6bg!rkVG<C^(V46)h@r~
z^Rx1~UWLef-jJaWHj?ke4f^<$Hkb0>JL>=O4*g&MX_-sXIe*WjioMV@H-=s2q91?Y
z_&cCybbNRg_~(of1{ym4dT>Wq_)ynN+cOxj4_uQ<e(=9eZyd>!0%F2=*pN-k`KP%T
zm4ww-#446QNM3`NKcqyK_5Eju|9}1He>>m*_xI94z|GJCCCdEQ{p|nO&G=v58=?hk
zfwCLPj!ym+YWM`ffj85Egn*{;*|27N%YcRD+fRt3tR}A+U>A-GrX}TjcXRV_d@LU3
zZC-t{zmMhu`DiYmEB*hu(1rv!VrZ`gk_$Jw{^J6qt6)f9^lOLX^L~c50#zCT5PIGl
z$=3h^(QtlyEc0h_<<9T#D?o9a5np^uh8B(*4LJQ6ij+n&65|58W%h<Is~WMUDR<%l
zH$HB^vzDd%1B;M9N0100m%rMaD`w{d;;H-3_2{3lJF%^aXK!Hp0FhPhP7Y~zp7Sg3
zg{mu1oNl5A-+l1cAn>q4-vj0%kDLHy^icOd<MhE7+b=WArj|D8f&oTj7+_K(_>G$9
zuL!<-@caXx%k^#S*~n4lu`m9|-}OJ%sIM_7ae%9B3dLOO|7S#O*hqT5c-Bt!3b1Jp
z4u+8lF<g#CV8v7eE$|#m{HK8R;v_9R09U|g&wed0ww`Wd+?Qm2-NP3^x(IX`;ddrG
z3&16_%JT;ddu{Y_0UJtJpB<AW+B?l)u9iP%e{L<rZqHY{DNUT}HW|->D|-AI@FyTZ
zT@1SfJZwE}tRVypPfnJNlOP>8^BwmC02#Chi<|9dn&vy#+@5D+3dDpme=bIOs!K2K
zmi@opwVQQYLG*)5)<l3YjAfAJopuu6x>MZnJUX!?i8S!x`1|PP$9z6r_+Pt$)tLb?
zAQ$A{!LRfFy%Ev8%^Xg^8I&@56?=p;^vvR+w4M9$H{d|Bx0!-mcR_}RvCB?UL@qu9
z$D1bB<)^@_Pql7q1BVsb{tui0YqKYhe|?Sd(szO2f1Hy1{{KlGwPSZNC(k9IUGsNF
z?0oR96QEMvOXW4A1s<flw~(7nm0xE82WMZI8IB?!b3++^$sMd`xe!f?Ao5&$X@2dw
z1hypnfmVAG`Z!&ErD`V$0FD^!$OmxhQGgdzJ<gkcCIfV~;lLk1rYc?$!}v#P-!<NZ
zALfuX)^XL+@o>fd4WuUVE`|62-D4K7@S_#s1d-!z$ZhSouNkCtPwq#^&|`;0;#So=
zZ@*uV+g}i@ekifb&6ajI<YpX#iBrJ23-Fxpxc}N^(I_9l8#3$DBh?p`Zk;ce0L0{F
z*pD#u`5J|jk*jlHi#cIkF2%;k(#dE0NgnP0A?&TAqHNo?@1a4uL277_?vA0myyz|g
zDW!+*7KV_JZUt#XI+PefS|lW-Luvq#{?57X`+2_iUF%!xTWkDN98b<;Kla%6-}bOy
z^f2Nfchx?LAMjc`M0-&x)Lk0&{QHk&-NlQd;;<N>K)^_~<<Gq2{XZ#>3d_Orcb$Nj
zHri@Xs^r1@)Aq>_Ug<;A$)}I~hI<Ky?I9l(27jCWe|EY5-*-L&sshB`a|_S|^nXHv
z4GDVB<NZk-(&q}PR{I5X;)IyJ3wHV7sz}qxzq?@%yQd4ovVQ5hEulL!Yky;b<+cqU
zz@pBTPIoa0=@S=T-1LkD#2C0UP$(EoaVvuDSs!{?V@?W^dpJvtdwe)~Y!%-jt*N~3
z!~NcSH`a@nj`Xmm`EUq4>Q{dY)~*ZsFM$hI`1ErQ8p|;9@1pRT*La0DV3p8B`aJOr
z_)0VG&44~jc6xUO{G$2SU)ON|481*y;0a@Vr@-PA^c#S9Rs7cPod-pj?z=u^Y5%Nx
zHg&zlF1IMX)(zZF`#h`dDeqlnfxDeA#=UiM2q&Bs<MXr6sU+fe>HohO*>ec+(4I$p
zXa3)(TN%|?j=6)3jlY4uowougUq|Or%!nPt`}>HjbN)@`KzrR<-u$iQ`B@2J2uIhx
zwx?iyA*|J?;<Vbh``J3RmmxbS>n$`da8PvlBw$#IyCWM|wy_1!6>lB-n>+z|S8;sb
z=id9H)66Pq$9_@}0OK*Z{up0W^nY&+`zV4Y_qg9#+P@L`U2pDoe)WfY%bsF}Ka6$V
z+7F#JLY=q|O%_)ZKq+pjTc7RHMqosQr$*tH=18X@uL?}ywn{;Pd;aeaZH`9%$8<vu
zp8uMB_&xdMs^_Y^;pU)*uqj-u1%P~t-W?eCHo3a7ckC7hEwz^Z{OY|9c>4IBziS1k
zYtcY6O6Csubw8}%Uql@p>ox4y0<;1=-Jqpegr<GirVPw8U8hwu1WC?We(nJ<H(Oa9
z=}=(O!9ZwyvUay4_SGN2VVp)W+<}o-4Ic8Q9;YF;svT|Z-;zQ)Pl3=#My~$)(F4ss
z@#iMMR`yqbOPQ%t<)>v1(%nEg<+8pWwHdQ$rFD>+LVAqdZOT2Ay(KhhSlI<;J1<vz
z4oJSu7iV&rPo|$(1FNR7tOF@>*BNp+NPvr~l~av**DPzjL+_<|FL~kpZzsa<?H44=
zlACFQ7zaOu3eS>#%mUpsUfJQQc9n1cVkCsdyX)q97UZqz_Ac682^C!@b2dDj5<X5L
z@Op?Q-&><2_mbx>*-Bhw6JLulg8XVpZBBtFbdnYQj|!1}4-n{M-f|h6bu-Di*0onc
zJIw$JVA6;Pp~;7%Q@7VrTGBVa+_^B8$d9-%GJy$|S5iUfIvPX6P4JRSRmHNRYS$u<
zX_LdA*$R%My4Cy10G|7mwfmKhVW%6byg+5ssofc|DMEmbHTm(I{VMM%n<h=4;x_U^
zc+UXfhR(l`0`R0)0QQpL#_IsRaGk$Tn~SBs?ZVfs|0bGUx?jyXle6va+uGf?;?$}l
z#@a!7nclM@np)nOH6U6BAg{BU<qIc6Tcew&H!fY0q1R*9qV*+PD4xMvDBM)cA5PGR
z5J(2>P(pRENV^VAMeR&|@7WtqJmd)MTc?8nf%M{xXUbtzXC`vyK5;v$%3CyD#i%&t
zA4zQg1VQYjemwYzyTE;DqbZ=nN|DXsuB^WDq^e*VOBx6-gn-3DnY*U06mqRxRrY&D
zh!5S@#019sOaZSQ;7%lU#k$Z=rVE(iZ9Tuc1<a1Sp=~O>Y+x=X%E#YMa=5@P7DfZJ
zH8yt_IplLTA8ZT2;^t3DBam*NsWQOo^8#)}E^V|E=)XbK#)#<ev41bk!%h8TApe@?
zejsowdxqFZ)=OFWEdsJ>M0mb}++II2YeC+}I|2-od&ta*L4QJ%wi%Xcs`j-j{Dk3%
zE}||ZjA2vvB-bpOxKwpG0d!-D3hdJdZGK6zzO&1~UoiX<#oUlXv=>C%1%y~kNsm{O
zr+sL{OU@Leg@u$r^v+>GRy=ZTx5ocd@kAJ5Z(;7_|AY2{P;na>T9YUZ27neRO^U=$
zPG1&`QyPHb{Sv8%YX7Z&yy0WMwq1|OiP2BrC51jQK*@LnjD?tKgMK$Mv*aQ}&oAUR
zNu0-4bo(BNEZXYc+@2-q1~XX%pAEhMhamxHTe#S6u#wiKmt4%GHdODIP;+>;9}emt
zMrQ~Tu%ez_+a3Nh(I0n5sm?)VpX(Mu{7XMI@Si@?c_b-rcYh7ZN-L*GUM<JvUcIdI
zuULLACCUgwnv(qjD5KawviL}xBEEsmm4Wopzz$f)P6OEk+vO*2Ca@n(pj@oT??^n{
z->SfYTOu}rz+T`m(J|tltsYukb;x%!$ycL#_ZZ^^(0#=&IvrT|2CipH1{z6EjaQL-
zs08J?^7lJpzFjXh=Pn@F1TYhD9i%Wq7z)gB7H{WxPFadz1|PU!AU1sMb3mStgVLZ`
z08XVkUwD}>iGc|?@wtlwM9btYwAc&$kBDtq@$TP_e7Jmp(@C*#nM{;Rm~3<hF#FoY
zbKFN-r2}$Xr=zErC_h~?IOX$8YSZYkDTt!>NJmlk3+CWx`U!p^Kz&Dz6YbLrq=39h
zeh>>+vwCo!-iu7S5D|C!NH@#kB8M8=C5(Hld(KfVxC6{Z53hOq)cRrD+RR$Z`g)V$
zQCu3x@(8Q)y+#-OjIWNWgIdn7fYIa*b$aUv0}Jhv<UmbXs;uN4kb8Vc^HCFW2)L=2
zmftI}g#3xf$&z-R*`0ozUbC0yPGR6oG2L_3n~S;yyZ~KTgs@MNuUd3ZZL}&0%Zo6u
zj)f)^RSubHeWoDuS$%jUOaqG8nS-F%;LmOXkib?~XnDXAEpU|6Ct=!Dnv9MH^qCkT
zjF%a*Z*OuB%mupLhd1n`Ehib<?gy>!Vu&5~<4B<y4-C{VfxM#m?$teD_(pqek^N`d
z5pRrCL8_P=Tq_@4fv-oOiMC0=;)k7kfsf?zIsGj0Aa&Nfz(-mlroBwa?uqLmHelH_
zo7Pid7|B7qCe%vv-S-A$O7c$z&BZp~|7VK(f;<A*N}z>qrOgwjt59fzpEK#|cEB(R
z=f=^V+n_Mq^Vx4ylOgTTGKz+`sSy}q^lYl@|IPWE&!|EM6!6{8`%!CoS{IlaPAgrV
z^J44PF0UvVP%UpCWQU4m^^^(YrBf$D1Tc40_qL9EA0I$FfQgE3F#{E4dMLjH71dSO
znm@W1x=C=#eWiNe@>XS@$I{kYqsc!q1RLVN=AP?{SaSXWte37TvR&Uo!bX4+#t7M|
zZ_^-acAEK5L1WVhhT)A9Igp?b&CsX}`}f|i?5wbFlcO<P6(Yj>)4oHsraaqzVKR%C
zaNnKA$+-=^#1<e5<v-Y<ORN>keQzWF4UJSJ6OB{8G$Q^SD0svyt7S;URwMa4dN#=G
z`A>qV3<R<icp2XWWZ4C2z-0gfpdGc~k8mb;D!nuDm?qLq;SA2?Ng@J0)bj#qoA_aG
z+-PA0C=DG@lM$kq6QeyytzA#_jjE`jmb)yK;GhucS|OKyq5T<@m@nbH$Ytbjdicax
z%pxYXb1EO6;{~vP`5Os|430!2)jt58M4_B);=mBB$7JtyiHJCZ;P2diXxww2qpOE}
zr6F+vFmdvwJ>O$0(b>Bqx9NGG(^ZXoqlGPs3A~6vq2E0v!E(V!M^geP-zFrhg!Fq8
zwr$|M|97vK2-G=;gUEp^4KsvwrLj8S!aAJeU<{1%^3P&eu-hX85rtP>!|@%QE14)q
zr}rmey|!=Rq5HDCBQ;Dzy?|cd0}fJ?DM^j$gRCh@#aQT5yic9$R!6<3h_*wh_Z_ki
z5Fr|f6lVjK?<-|D8&gB&t~yNKhTVV;&>g^P1wehAmFCu;;#D3DUV*6mJ6#UOYs?Ep
zISJg9#xn}}d?2Z0iCu{V2`XMIx|O|Ar07N#$9W%ZhM6gq<Zqa7D#*Yh3Zo^SitXA@
zXa7PN6hhlo!_-!InF%NTtuvM|Rmjy^FDfpc2Y}0Tt3JAvwHV_aXOb!jOA9F|95AoD
z<OqN&c#=@5n9r(>@L+Ywg?FIO9*}x2J6guQs$PHjHkGC5mjGoGWfOh~TL=e>K#$@|
zWXp(&DW2ufhMk4w1f#{Q|7@?25wrEo4{(|OY5>j)6f9MQYA07L?@`$pRaO;RDl-v?
zs@{cx=${X_7vuq?C2yvxSWz{eVj)gu|9t3#<qIT3v{bW-C!WVrp>UeV3&5oI!w{gw
z8$e|*bay(GJl=N;--XHSb1_b(Q(!0jXWX2lj1Ifqr^$#YYW)4Xd9wAv=jYwCZL9)?
zIw_kp4BTqA?{wt>6bRzQH`jcdPEX*t>to`1LyX5q|5LD~UcW{~M<Vw!RH<*1Zi;ZM
zigHz#w$R!)F!%x@czT{4v@AJ%)PA}YJiKOl2Y|zFZ>hp=<a056_pnUBT~uf{0u<V1
zx3)T|^yOMu*5gtgj6(u9V8Rsn=jc>5xv9@ja$JQx>v6|B8~wR0J6et#t&T5c2%4x0
z28Z|b_I1S>ve&ExH`0LqPG9w#aJJ(AV~lPS0S4Cx>8EREVx|90OAQ`)P%>$9H5d>Z
z1zOpE3h}!1(G%e9Mxq28L^$lBH?u`4OmFe)4DX3-xnClTe1rBnjL1}=h6zlTN@;p3
z8`;FtiRhHu>Kq!x^xL)`d-py}O*SlV-f>iFMwNAuC{h0^!?uWW9b{k{P$p(7#akRS
zkMgEwE$yOE=2XP|QF$NbRgtW$mbdqO$!wx98!?DrVqKgHTIWgEA{%+V7br>GFJKUg
zm?w9H>11Q6j&1@Ho1}O7Drt_T*i}(M`*E$HWI<TfbE(7}D!D(6smh~F7{oHAU295E
zdd=*YsoDpIrb<j^RHA}^vsZY#*>t7<OrT)V5;+!A2gbI_FL5%s3rPCZA3M~YPXj0}
z$^+bMvUc7?XDNeYCQ&6~mohbSeU^9*g#m1Ire)={29>&9O6Mphf$i+r&@p+cZ4~=G
zvt?3uC+h7F0|@0+C5t2`{LgPZR@Z<Vc5k?a+%@ydhu2HW<w;B#i{fcR=r_1&d8P4O
z`sGKEe1jC4A*NS^b(fcl4(A+AR%HD_<*DT-8O947k!2y)1AA$cqXBf=)RQ;TX&a?v
zp3IYQJ>s>CN-<?(+_KC&=BqK@abN2)@~`uM9DXMmu&8hz#ig^R2I|Ldawrtr3VaLX
zC8-TjME_hOtmP$Gi^$np9&`e5fo<u*Y)!%FNpm_g{v4#x_2fzDg!fEEi9Lh;nO-!H
z?)xz&CM9TC3gTFpBq>>X{>n8XVI+l_Fw+9y0(PX>BDft94EF`_E#^(iOPwLi3>SC}
z{FmZ~7r&j7VR3zo67pO2oxHkBQ9V(u%2IO*nI+;gIF}kKjXsj%>6kZMJ!KdeQHMZ-
z8OVGiOjV0rimbAVDh1t4X>rfqQK$n@ExNmeJ*Q2Opi3y7>71evI#uU%nX`aPGo{n-
zeL%R)9xFYqLcuo{nkSq&xcP%Ng6Y3^quo0;i638zR(%XIf<0x6rWGbN=z;|-_cW%)
z0$!mnfx&|LBeHuz(-~V{dW)VrM}~h&MAzzroXj@Y0ox1DP2XJlCSDn!>5u$oqS))E
zU;n$XkDQ}$1M1lbYK81p(@%2P!5nu>Kx5)B{OfXjL<<g!HkpK11^u=m>!_6GHama5
z>$3XT4;E5sybn|+*P7@e&@w@0zwfGBC~S7`vz^`pY&@zQm;g=;k&%iZx9r*vs5xqJ
z7o7_9*E8+dE_|xA#sJcf;FI4#!&$%PIPo*qt~6FZDJqd~0~L51Se11MC*4N5b_Et%
zsSL17GArw{2KAFO&e<-jf8UEfQ5dB`a1M+#ffz~bs-;n~fEQo%$F<s4f~YT(BM24j
zpsm(th%!S=6&2SNkt{9hiY|Xc4)^4E^{J^{jCn+DXm?KmxLf3^+*grkz;wKS(SXUy
zkpdwcTWeh>C{9b){|6bcS^7~K`t4Z=GDS)}0az!DoARh{vHJQ=^eNJ$@t@)tnY+A!
z9gY?j9rj?NO8pB0)J(BdbWwjKplH?r<r_i&i8Syo<3k1GV!e-ZjOZBqYk(K>W)Q&T
z6e?ypL9o3ly8asW=&bQgLWwuNu#~vm?n%Z@U%~BvVK>$T7=t9yMdd<)IFD&RXEEg{
zidgJAMX!k{X&V*nv`y5%3R&6N%B~sQ7hqY6b>KY!n(Q_u;Hef}zA@wH)UDhpgRg1{
zxhc^I3gbMmecsx3hODglp|hzLyROMuhrB3CGGJGA4J~v7nlVLWH^!7Ss2O{2M$YZt
z=K-Bvw@*LLH?hY>QF|~NgzNq{&_={W$}icaxVtD1H8?bfP?--jkQ18cYCnDDGU4|T
z;XvUBq~zeH$@2OX>ETupM78j=sI1#qe5ZHoD+kT?>b{WHKV&EkKDQAqbOV6jg@-x8
zD;=;-tHQvi4(a1L57oX_MK{u~M*eQ+Sy-)+j;<E1LPz#oZ>5hr*0fteZ1mLC?zlRw
zBB2lGMb=M6mKB?Q>6U$VAXepg_O1Py#-(@ja8KfmNO)fBC%omA?T91L>%mE*a3*a`
zs2MKPL=H{ZXkM-H;yZ-9XOOu#u)pbcDAjsIOZjfDrm5|nTS4Bc)7TWy9M1`Q@Tm=O
z(9~Q4U)rbh|F{N*um+*dL2ldqqNR@J#g)4YlDV~V3kILswE4_mJl}OrGPI@5a%+aq
z3(S3xMK#yd>j^fd&y&0i3MsGBqy?k}8J^7;!poA)7#8#k583VtE<ZS(Z(Nsi5~4}B
z)cGurgUjo)uNIKq<sTlyIw7{V9<3@)T!+$XzbLl4U8jt<uM&GdOu99WW%+Kl?&ci!
z#0r>Oy{eT)@N6bRZEopXxhPRyIlH!Rcw0^^Hs$ebD4(|S#O4i%)FPbIV=vsYgNcCZ
zVxcErRG)$hO&P(Erg`%;X{={)>z(v&u5`cAm7AqGrP56Ekt_S+PUpWbW8-%bFD!HF
z=9m!6;x|I`jkSEs4x}(gSmXA+D3Bnm`|@Npx({wIF0`I&i@~tq%*!O?XgQ<IO>d;!
zO1PV<-pIk_;ni*-0~CjDxcmPZmJrzxaM$mn@h`|);Lc<2?qiu(t^nhh{W%A-9ZNTV
zD?E4)(6!MlepCpB#oaaxJ#~#nPBGx;iQi@uPyQ(oJDMHa$kn87{e9c)1-#+QTQRRw
zU*`u#1gFwt&^<x$2BtPi_Rsse&^Q50*+)#%pNNoZm09*iau!LPA^o8_prOg8$3Bgw
z#rzXc971p@VoP7#&D+J6xDJvclJfiVm7~jx3z*%GCxg#{Y&=Cqr!#~7SHuKQs$z6#
z2JwQZ5aD5cZ_kMS{?j$0eYt|A*8pn|$IMyGr`BE8dDbl7xkKb45Kh*zZwI^b8Pi<J
zFL8hR*T_l?7LnEs3)vgyVBkr|Mm7z<BG)O0+-WAxk=_^GNk7YDghVIh#UQ;ATr?KG
z|HA?}Ho$e~PFlw<3_<rnM>h4q9vkIWNv{OSWV}LoPsqolWwcgGL-^&ux@KUfeRjDP
zhMP6JVs6e^!1vZB8#U0{C*#BFbY0G&HdlBvyaL5E%Ma2seFS4WLbVk{ut;`GhX7?*
z`){FhguIwm$h-#_>WA6u;%`PIBhGg_U8$2fID7f@UPzK|3*|t@H~)`%h-#<_0O5(k
zHm#5h7$V@|pwmG0LkC2Aht>fluDzPHA%8?#*l%QZ3g+A|Sb;xJlmG1)UOTuHyO%NO
zLU7m6!q9nG=-}pXvlf4eS8Az$Lfy>kZw8uNE7_)L&hZ^c57;^qr7sNWEr<Fu>3Pgr
zMHPD0yUju_zt@K7%Te(GHl2~h$|Y~jL7$PbcR{+2Te9~-aO-wn^bpm$No984VI~M}
z`K&8aC>fy~3n&UGO(O#R{yUAN5JPpue37Y6#lh$#(4H?0C{RhspniE5pJsO7ad_qE
z3BAs6YMg1CQD)oH>R@!uhwkL=H__#)b??fYG>+gtF>$|ko)nDuCBxqzps<=FZ-_y*
z3Yf@5QG4lEvc|BC(vx{`r1AkiLqhXLgVR@+t>}GRb!|+)fjLLTh{V|d>6->47LXNS
zRt2Fy`6$itwIf^sow*<A^++oOD^kaj=$T_+VVlD|_^ZocUBJjHvUNj1`AJTxLpoMp
zhc5oH$Oq%7>?Q_xvS=cZ{u2DFF2(HW<m>WJzpA1*gV<T&6l>dEF0v%hQtbk@HPw+6
zl6r!;3+`<ezH~(v{$uS@ICY%k8X)J`8GQA7N%ox$KQ6kwdp5Ni){w*yth(<nNYoKs
z9?FHk^zJ8Ic2n&H@zSOIC0T@H?A6_^_A<wb-@YrqN2yY@k1-|KCcX7P@Y@6a4;^BZ
z=zSHM*_iI<Vocc=A{55k?lXr!Ed2{OzYCa@;T>}cyE9l0H@p#B71t}=T;jP70{m#v
zyjV1DSkV3`g9g-i_##!J!~cQwMwgI9J!8SyVjOY-hW3WD()6k}ufZTFb#C1C$FBQ#
zujvs1`W_|q(I4^qvAqN)?a00*EK<dq?hV<2;%TL$errFXwW}0HYNz%weVM~bQEUX#
zGKw{e(FVu0qPMOgHjBc1MDnI>gI$G<*k7=GaSL#KQSY&aTzt@>uqzju=BfGya9oT8
z`T=K0d0C3&Gk9~|cM*6re3NL!RIeGRX?OH5|CrMU#9cgT!Pxn^sdhPj$vB<{<Z%jG
zvqo9-ylPVDTw7@I{Clz{0U2?CR0rGzr~$;rorA`6`_H!d!rEmfo8hb64&8uBh!*>x
zgdm8823kH@hU<RLnEHo8?uKs^fEA!O+l9NuN~1CsKWouOJKOf%1hnQ2f=K6$<iDV-
zazX@Q6Kj&^e%vVPWt?pCphYHH7?JvZy@}7zhUSj$ocV?g%F7IlDaSRfAMYQ5uTh-V
z)uwU>g;wr5OYKp+a1n7$^>%9l|7iHco596B$`4#~_&1hyQ5fe7NWTT<Et3t$Mk3!t
zKb2g4YlGr%2337geyYK+fJ$;IDXOHYTm_;MX*C)ak_pBjgF;7#p=x2@qubr4$>`te
ziw1GIG*i7(S8p+H&@RXfOWX_IC)OYNQg{GH+=HUJrB+Q-*@0U0BQz{_7vTVIa*4w|
zUsA1@9&w9n{a4|r)VA<w>4ZepIt2fSL~s|`!-VSoAzEU;E?JrWyoDfJ(?yYZ1Er4$
zZ0w~K)Wd3c&0k*yz@r3*8uoJh@IN%1OjTH2#yB!SpUu)J>Qzi^okN%8?wyFodPd8}
z#IXs)JNk5#gKWlC>niIVjwf>z=zirsL$xu#PWBYC$_JW~^K34&aO;JvJ@4l>7Ag#b
zf0b8kUxQjj<NqzLZC*32qxTJ())jL7XG|EOoG4F?0{ggHPTs1ebC?;ZthSQ`Qj;__
zjZ^r@%YO~Y_31;wN+trGwMP1LlFfd6Y4@E8stSoGiS#$E7&-VQi3(mD#sw=8lQ9Oo
zVdTg8GDf#0qUP#Ny^{`tA82F4IQrOGQw1f)$<u7UMpM}BB5hM*-MStH_6NAAtk84N
z%cNBq-*`pi+_2cnl0{ppNf<)k9R~Zb>cR-g!af?QLSw#kJS(1y3{QHE<!D1cN+xj~
z5-Nq#5D(wlwLJX3W&Nabs?d`#{G-v}l%A>^t8T#aveakHAiMl>>r#(Y^Zh671d|O^
zC88@nvu15eRLmV0BCrR2(Xy7h^Xk&0Tx|)N8uUH@KkRO6_%mM@)yCTXwemb$Pjb^c
zfglT-Jn=7&K@I;V84zJG!B(F-8p|J98IS_UoS*HBy}@vFPU1yMG}0sD8AeQyw3rmi
zs~H0XFpx7uL!r=c$y3bBn~?*bFs72>gv0oAbnRcou48?02K-WJ5%Io@j@i@@G!jIT
z)ddqXw(Ug~;SDf`aP_vti@bOaa~*=7mY~1;v+r0Tnl0$Ic;&Qh#$Y1)4R?SkzAO{P
zs_4`wF70UnPC2#@6AU~RB~PS{B0*a~iV!I$K49@;b4c({U{R!<Q>XAOF}6vGm&Nd+
z!+lmoY|;ShA|a8()#Ps|Tl;_;(7+xB#|aQs+>O3^1-GF3M*0g&@Tq&;9}FE%8)>H!
zBKl$^@S1sAfg&m=oIYBf7XEzw6>vSg`wXE_XTqJLIG++_7&lbIZBw)eCo<VVfnAGk
zV?@;aHv*tpGTey6p#j|r$NlD<R2%BlGdLYpcHw&wyeK5JTiB~H{OZ;i*NKu!KLd0X
z9*3o$G`-Xz;2c`q%0F4}PQTGEDE)oLZ}JldoiBnLro=E}XUwm<n}%BMN^#o&qV(dt
zQTDO1(KY>5k<hhHo`~_Y{l_b8KJK9i;ZXv5<@yb7gUVGVRf9(`7_yOLRNQqMi$4L&
z0Vkch`=Ao_gY&t~TgTPxGb&4ff+IVG&0BF4{Fd{wv~5#)RP2)c-qcj><nsqo$Fon}
zDl0MTs@C`@<aCmkBr6iQtw`sknz6BLKCu<u*Tk}G%5jlDBObEDd*~3tM{xm%3O_=#
z+iv$<omJsdEv!rY?zFlMaJdV^7S97&fD)Q6coGtDP+mZ?ipcs*WlHr$++=EacCD1z
zG1&Hu)BB68{}8a3e`p90NK#}Ka#wR=_%6KLH?eQ@hREFTp?fvb9m!EG%tM4~vlq88
zvJbc)CRrHDi+aesL;^uuNzyHEC*z|;weF=Z=bpdmOVU!fpXhFBYKV|6XvaX{ZVI@3
z+s>1gc~}MtZVD9Cr`*#2D@jV}Lxsp=MsS=lYY<OM3>p-gSiE!aNzo+T;c)z-bOtZ8
z&a@NJfa=|Lk-f^2t@WY{)EUJHvgT?mgR3iEAxMdFOUYcXD}c}zs-%5f1`uCs63Q~y
z^hcLD&@?g;O&0~#NjB=i27irY1pjofb!2&Rstw7Gp&2kOog45Q%kp8^&9OkYddOz<
zHHtV`-KvC18#8J3N7Fbe+%wZ95Zct}8(HQa)D`h5W(98hbJeqHKL1!c8`O_(z|O{d
z)C+Qj<NtYDVNn-i!4kni821Sv*6&I(SRN>8d?s7fj~(KOG3*fbDlbg>=|c6qXR!F~
zSAC`AjqU#SOhzNV!-mYJ@uki6iWE7u=FEBhtq&;z-yRQ<#StwWPEtWj2URZID}{F}
z{^-wnt-*3sa%ic$N9q{R(4(@7rBRk4xO5}b!-w$`E@Y5sy41pR3Gco`h1&d1gYfGJ
zs!aG=GnYNjM3EaM%3}}O`y3C`AC(^Cb9Xm`(8v6y5mXR;7-$(Sd#!MxkR5#5t-d8x
z`@r{B^M*EQc}u378O_TfID=Ti@?d+R1vwN|5|1}?WN^%Jc^jIBRjZ>3k-g|4yYaKc
zTq-#f{H@CQH?A%r-hPwj1dRI@!$Iff8RxGtk6RJo1~7>shS@INVC@)kg0UF0$Nm=b
zF6vv=lqSdPm)DjYGBK&9y(^_?&OI*XkYW@{xTCxRyaqW8R5E{g@7-5tRE7kaN_%sD
z%?b%Ks=JJE%<BT57oT%r9w<FMR9bkpn&J{Xr$BaFKu`2~?0_qfp8C^~U{h6VOtPR~
z)o^8~n^}F%l@Nhz|Gr+R8b=@)8sBuwWcId_?pk~6JqL*ycWeCx>Jd>vW*14NrhTgR
z`LtmykEY(NFw3HQ?Wg&~y`Zf#PY$y*d@bYXloIavE|+uZv{5^KLKTZSP2!gyrAK!H
z9ACG#vnfQ6eDRg!N{cL|U{K}Br@oHH)~_@t`@l4s6kM*o4Pi;+Ahez5RZ4j?S7Mjr
zprc>0#x?YPJHI6n-T2us6K>MFoxN6ur*N(DVB2oZ<nPjqgEAem*Uo`Dw;6GAv)+jg
zlBK&8oV+g?1IbK960w~EVd_28mC_x5IxTOXUU1BPeXb+Hns_0buH7&-A<AaV$Y6MR
zw6DCyK7`URYv-TwdFE%Tf3EIJh)ZDCXM2XrYCv%8h}kVbWrwRCw&2u2BZ422rJC`O
zC~GQZl3O4z#e&aD169NIFM~h+!Shm0e;Z3brw(KR0qO3zx#b#hoFq1xpVv83EA-J3
z8@u;NkQ@ayYQx}vw88%>6M+M6L*h79Zko%7T5Lq|bU!)4pPUngxuq3b0K+;B1ebwI
z3r3n6T76P>`1vBwT?7Bf@uYY!s-@D<6yTK~42q|K4U(~#FqNokTN1alYdRBCwR~i=
z(dE^6%PAdWjH2f&Q=O2CQsZaxEi>GsWkz$=w;c5#G*-z(a<y2qGw0EAKp-(6%HF%^
z@TH(PJM^=mngj~JO6GZs5WIm(CY^-K7WV@Jm>3Fge=>3#5=^f!qE&&)JY&r+)yra6
zt0DD8m&X&F(><{MA^&t6$~ejl$*jr<06G)?Z@jG5PJmj0C6od?yqwZbY(VUQz?XwC
zvtF843Ci_Bm|*(RCZ}{lFzcTfq^;Wl-TfwuZQbots3mZiJT_UpAOgh>v6*kY(t4{K
zTmM?+d6Vr1qy5f{1u}(MpVMIAU_P)QOv!=W=f<3t7W_HV%p(LgSz6gaWLINOWM}f7
zWpKUAt=dq$C@#%kQB2e1BO>CVv~B^xGUrJ}CUGACDAj<v`Rcm@NxUSi+(k+UG(LTW
z#DIfPsW5Z%1g&);t9Kg!>)+<yhu5FE#$K8S^M-22W-~qnPNtc~oj_#7hNFFZ(ImcW
z5X*D>@X_KBZLAJj!plf1WW|KV&nqap?aLR!{3aFFg6PYK`EN;Z_PXIdNkqRc!3vah
z0Y=cJ{FBkN&ndu>^08@RZB5I5Zw6nrd}k66>jo3Li2A!IfB=$%DHAAt?U?}zi<(23
zUHE(X!tHC%Xn1~MjM{Ld{lTIzPWUD^*~K@+Z^{o+%J<m7<B?~6rchV+1bz<><?ety
zn&4%ObM8cfa;k9e81c`uvGwEV1K+0&pYIOcdzb-8>~F?Ji!lI_uz+J@>cV1`O<OgI
zR-Zhlx$76cXM?EGT)%Pm?7xS}rEsCLzT~)+D}DEh!!yiqu}M-VjM>v$bhtiPPYj%I
zD;JJ!Es{0Yz4pfnI7GNv?z5JE=mkGYf6w*6l_Cn`c5v4O{jQN_JF173yf@yX8lx~S
zmu6<J!00eU@4Mh0W<!_;@}(^>e#^%AEjs6leSmBz+MF>XlV<CoLdE|{n!`aFOh$1E
zGYb1LrPwhJH<}tbuT8hxF9=A{Hqy;+`ak*9YvsP(h+W7k)KtZ*<-nPNN@fkBxu6N+
zWJr)z25f3X1PgLTs>Ma1m12!jY-rXv<)Z(3i$fR=tdUNxOFnVk_#zb9`i~F>)@o^h
zm{@wb=k{><UX*Y=rPKy*pgz2a@AeH7+845%bdUHHcLh>Zu*s-zZ^-%G`6Qo+&v|mI
zIKQXI93%e(npLTZdtA1xktBt&MHzOf7TBdRvFL5ABTAGMbQldooNbVgJ|HLDyTEXH
zKTsg0?lSQP@?fyCX<|5>G{88cV7VRQ?noxK<jkB@rS|R@^~PWx)03*#m(R9e8w^4M
z3k{Dq5fHQId1@!Gf0yYH5u(dKaCl><4byrjf8tT|p8NEbQ10gvhm~@Rh(vudI-Hl|
z+PjpFKD9*#GG|1szno%H@ZoE^EUE%0*LG65SteVHP#2$eX<u<l)3c1gM1i;#%CP_z
zQX#JhABsbN(_Zs|_o-WD1ErSh-<scZ45ERfNAn#N8PJAcJe(&0aL@$+4kFl4`(FT(
zJ7PYGC4$id7WeE9^#V>If(YhS>AegpC;o;lUl1CZ)jk*^K#`et_@bm9WfMTqO4T(_
zNLGSloZn!{c70lhYMJLOJ_}pob4=T9oJ4?mZy*OuYXhL%uq0X1$OUSreA|J)8=r4E
zQtk5HCjsXYzqBo(E=tN_Dunv&pq@&R+M|^l4bVt;KQCIHHImwiz~F-uoqD0tToRmo
zOGar?Ox_Vo<j#u&Hu_Qpl7)p{Fece1w*%r{RR;z{i8}#vVe{!c40ZcUfI_G3&yn+@
zE+$CU$f1Ni(Y6M4m3&?~!$v`u?9?#$`7cM}#iSFP%*~b3;H(Wq889#T7>GT^0Je9@
z>w)Z?%|nVJ;*NzG%1&%hvRZBEq_aaCh~h$tB4rT<D-5ghu(YwAW6mBQH~UlFlS=#9
zcKtxlM!MU+CVV#7-%s6T#MLHbpMbi5DIgy2+h1iFF6JtBR2$03;Jgc+8=cv}m<N?*
zzj+n8iF3IQ;R`dka=V*+Is7U70OmRd%y}3*{*5C{SjHF^>W-ZT>(mL7*@t+e8}ddw
z?M)I^i|6hDq#~vjg)SiyO0k9G^J-9@dN%td&F_ucG;bddn0+n{+1HN%QGHEl@<nh&
zS3=h~L!h;Lh&w$f7flog@a7OWzCI9ICIQ`6lPc4`D{SV%y19HU=@DPk(&z%HEH63}
zTIMV3A%qF^ME8aTS_3jWFKzmBOen#iV?eg9==Y{$_qOc^z!Z;kWB>%}@}Tn24XDU{
ziTTU_2^oN@@kI^Sp4tIwlZ+MUgE0^?S`nr<9!>6zL7LN65n4^ak~a;C9|=@$a9<Uf
zU9|xqKXv1(&HokOze0aOg+h%RWn_y*{U$1t|ChuhRA)-#9+D_x`h;m2Z=xpD`O*tU
zDb`tgq!yX!4vCEcEB~D0TA~NTGnS`BbR4kp4e+qESz?bZy+&9e4uyR|(D#X|qn+SE
zGuQ^V^I<$#R|TRHSFElF`#~+0NJiH3#vHaASv;ai1mRI5ww*2le0v})DZm6?YN69i
znY;x*rkmZs2F4x@7POf%8i9D+Mzu!b8i-?3Ll9f|_wz>dkc_)x@}Cg;NxaPxtf72o
z+*GjR-*wJWO)d_s-HeB+d0YtQKDfe$D54P|LQD|ZWoueMg!3$Zoa{Z8pc0mL=8+8b
zyQt)~h#RyWgJY~(KRl)wyuq5c#gL&AFh=VyZbDPYXjh-udvAkhSe#gqt!Yo#3j^#H
zd}g)dR(F-2eCZ3K{Zu`Efb&CSM1<S%eUs-{@x*q?lxjDL3!|F46q#rBrokyX=Ly!8
zgE+2%8D<(K<_{%Wd<2YG82UNb)z0HmQ)8lEXkgNRYa1L-OQuBf)Vw~xRHNh+5WPz3
z4G&=`c^`DE8AqwsAWDw1BQw*Yc5gnC8jNB<Bbr)}J4uj%8Zc|beZqjt_A@&hSmg53
zDVTVX@Sq@>5#mq_HmJQ6nrlQ7sSe(C)nIUD*2)^%s&|vxE-bOS%=oQDAkY|=+Q?WS
zPa43EU2l#5T&EQ=fntHdQXz((UutNScsV*P)Nh`5DtlykNUv%1#`4DN26{6qi|TGo
zj~OYQZ`P~3#;6$cJ#2&W7jL`?{obR&8d{Hy26UQl+6iD9s(Nt~xW`c{RW}M(s44F?
zi}nmsm@h1_+G&de#s!gT@%w50X+7l0<rXcklg<>l{G!M<AWwv7vZ=84nBn#DD^dQ}
z0$d^ZA+=;{5SB_XYWOGh8GMRlUqmQ{!~0LK6);CW%zLMeYpay#+DWU*5>tPlVI>JL
zY4@ra!D4zzU^}Az13{J0@HGY)5PSTRNr5|=UAtdbdD?5S@Ylg%0KKwNHYqgkK^1+s
zWbTgu1)=i7Ms%o^O4y;N(#vMAGHGhXkusyUJgo>fl{M}Sc^HcKX$2rOnH11kUXCUV
zUPnW@ju%i{b;kEzYR+8yqiX%-Zv-v3eG@Q<1=i3!8n0lV_P=siqY8qcS^nROW7r^t
zFMoG*$(qZlw<A=g<SCT@fJ28gsH4&B+3g87!Cu0z!TyOB?g;bT;1>+1VL;~9puq&@
zTM*5zkb@j6?<@hk>`{0Nkm%CzU|tfx1(&LVNVPo(<{V@G-c?J8{gjc7;*T+%J)!;V
z9B;7d9L7!?;<v|SYEbL5LN@QRYH^CMm$D6RiD-cYJ7nV}E%K8)9}r#kn{U-lMSskq
zf-*3aZm1Ir{gt#t3b5=gpST1oF+>Gtpl&ilm2?V{rdq57Dw)d!oob)J=eL+gP#q@)
z#AXyU^75QfR@T#iULU)2fBGj~sPIz@hO06cS?AsHv`<v<o7mT9x?a65PjkGdRQ3Ip
zesbKSdn+_3(T}2?OA_a66T-VbKGRE8qZBk>q^&>buKbbq-49ln*1<iQ5+zIAP36US
zuC?=P?u|5<S*S{B_p7vzh0SVWkFI$WflT$_TL|vU+DBDY1B&Vf?x5ChA`LGJBYUzM
zA-GWOUirdYp|W1Z6qU7{!`h*c8kg>8Lt+deY%cP<tbBz=s!1-@QYSS!b0PNWK^KHA
zJQ5`DL-L!HQ={o2EN4qj?Ti1ZnQor?Of!YjP|OW>gg-b8CXz#vR9tVvB-k4yO5!7Z
zr!NER>jRMxC86kt=+}@=e*98p&uA-@0?UjkC#cqvwXJ4xYTr&Ifc@ad6bvmfhG32<
zQeJbX#tFo~MgVHcwndv68RGZfJmC2yFWuB9Z>Et25*Bh*h<1Cg{offi5Wlyui<9Hl
zBN-dH<>3lv;=KHjVP}3`8xETpYdo#doR7sn3t0F+sA0}*oD&6Y)xu|@B5}&NM=|50
zi5|cNw(?GPO7PhM5va=(*r?tBgtwpSP1-vg%5wbp6fG(oj!%q@aK*@i?0d}&>tuEg
zm4iZ_AcjFP4J&MxB?=aJSrR9FtgyCIloLhIYU9gHg{Jm()`lV@mV_YnrvAFT-N@Tx
zs+><uyxn%HnKXzQhb@d(;cX~5;eWyD@*GNwsp%w&erZG*OkZIDqLVjMi?cX+KGB0&
z#YUVjR`%en^q-s!nY7NtHTWc>dc^a(i;_FN(;bvSPo0J+aD0<kjQRojq)xRs0Cdq^
zfh-F1>ZN2lJqejlPy(^^Qjx0>4ng3OFa<d2J%K|}wS0FZw|Y?I0%Y~QE=-`0LYdGv
z22uYd-Cr;IG;TVfLv^{lTk&^AZy!}6apE~3;*GW7^wgVoni`&?_CDBncNoNxzpcy@
z?qXTY7j2<UeM!t8uE)T5P29&2g?Wz>@t|y8WoDgrCBvbiUSsraDMw!zwlGHbFg-ct
z1lcO_DB1hE9{DSwgL4v3CA3bE;7oRiA~Si;^+x01>1Cn-y1it%3cPZVUo!+B-Br!;
zkt32f(PjphfM^nqrKB3akvS-wP?sFU;lfyLK)<dgwM{t4w*958kVe_t0uZp%>;71C
zNA&+qjC>o)jWI-ZK!2GYE=a|!u^I|MvkSm{zWd}-UcIS}0UR?F>@dx=dOZTgg;=8S
zJ9YXG=G>z4Y@789Cw?vrAt|;FQs23-1>v<o!YF@VUIgY#W}x!U_=&EcAzcn4et2y5
zr#ST@cVeLb*mkaOk<=*`0Uor}e3f0w3V7%cft|54O9u;zm7hS;K{2KMjA~N~5ln;T
zTfryit<5q)$H%KHGEOHed3U>4o&J;bxkBg3`OQL)(6zR{*e1v;#o|=?b=7rCA)WVU
z6Mg4zKys_zP=*S({UYdO#cyKDCQz)h$c0R_#)(s?{G*aQ`j95-3-VRUj57iciV}<P
zanEO=%~QI$A_iftr$M<rlCs`R+|7SCaZ!jAf6(m3%_LPmv-~k09VS|+5=!G(PnX?8
z+22G=pUiodztK&JBFCg(7}h0rp0BX?RU8-9mjX=<M0F>o<Af?t(a;1>zKZ-?*e8WG
z30;*iVOfGN7?d5ARpEk&a|Jw1G^N#Kc0=A>wHo1q+KtKd8m6A~i$wTChe%9LzjoS-
z<_tPLt|;C#o%~(&zD@xrU1?E(Dio!#K_AX{R8|KKy3Pf6zWha<#2N_Odh3MSOUVM#
z%C@@p)Aucvo*(KJh}E<%qB|KMQVN7-+n{`D;{ZA6)A!C2y_j|SK{{^t?n>%Afb#XG
zx8wwIS6dBAz|2BCYX*m2)v|!LETdnm?0}!*C}*j5Pul$zz#5G79$5Yq*`D=BKoXHn
z0on<X(_9luT*w`^m5v;gyqJ~#oKpqyvA+20cJr`AfkN!efJy>t)ivc{WdfrjKYAG;
z(!2eU&5Mzbw1al$PAF#t8RrEUm{`@OFuNuiMr|3CGgUt?@esO<@sAxU1O4*CsIqJh
z*V98RJQ-(r{KRnUB_{e^p|gHM>H*H5Nrxr{qadzpmR|JGpG2Ou;Q=UPK_*EfPqtoK
zQl<*nq^QN=IB}0L7QbYh;{vgGe1w-Pm+;%a;-w>s;r~F7#+6IGRZm8xW`mj_tv8fO
z=Nln6TA5D8Mn@<o%!$kPd3M(;Y(~FEcB|I71wUXQ;lih*X}SE3a3z{=T5y%3Xm>or
zI;o<}Wuo^Xop}6N<)Oz<<NXYjsE<=M?<LbHK$R!>yFQDRsfJ6Hp)<Tx>+5DST6E<F
zOl)QM%ytLKtXX?-n(cx13yceWyqD#-F&ZvvC-#xi{p6VS#LW6bGK_llP|Ut+Bd~P?
zk0AePEI3C#cRG7$<kteQ$gRt5%+^g*{z=f`kmX+X60fFSzwP<{Lqoajl-Tqmc&U_E
zfMck++|2n$58I!ROntogq;bOrb0@Wv)ZwJN0aSDa_P^(}>i__;a7CK*mpPZPIsK;%
zw;hwRYw$`-+C{VYZSse?P-2;LxA$ZS*(9#P#c|mi+8d80tPoJt@I`xkTTM6fjyr?0
z%vZZmPKKMZg$Wm|xsksUIp46hC#78Q;^*zdV6s`Ea_b>DD(I_kXEN^iw;uANK>V_3
z5~<-0?5!(iz?p1%NC?MHHbxJqR7sNI4o)hSlWqKGVuLDF3~=qNdAR!~yPNAU@I`84
z`{K*TE}|r&T9KP&j=UG4{^uXd`M&K)1$jt?@kKf^Un@35h2zxl&z1a@ElP<JpeVc|
z0X@V$>6MZq1tE^39D^vnfI&79v`J_-$)A+D(L_j|x--aFg5ZGPNP1&>JS*#Jb`#1F
zuu<Z<vSDJ34Vp!LqV&84H(D|I6XTakD+ja_ZxgJo&*IzG`}qO*jze}l2NWT`l1eCG
zJ65Igx2U)$CT^Kl;s5M+z4*guwVfNWK$4`tcs5AI-rG!EtqGz^@VVoz*6~ShZ<N8l
zF@e2_7k#MvK&D70_zUV}K?X7~rp^w>LjA$C%CNrp9ks9Yz)%Vo#PxkL@DI5u{iSRY
zb#HM|xo#cn!wU~wh{y?!DT>YNRWo-Uhn*AT5m-4r?!_Vis0oxy&GKpxbR`>UMjJvr
z3}Pab|9+_IaTW&L>}r@uo0+l}q7^<jH9&;5`dJiX@zk~2(oR>n|M7Kw=W{-UodaTP
z#1Z3@8=X=LOu+vwN}(=R<i621U~%T-{mdtb*g0SRx8gU1(I#c8qHzwGvoWD!z~XbN
zV!MG65Mq%Zn>(-#Q}O^C18JU#j+Re;PCQ&_;snn;`Pa$5Xj(ze_lHDzw@3kj{$_^H
z7{cU}2~r8O6w3qkB68Y98@-0-cQ4rFzCDOOmGE+~%c>+LwP0l{RR2r2_FR#en`Wfb
zR0gqDA&O&pCP*s5%9ePF5J9nt^VZ^CA098YToh<#+r0g@<gxgGf`-Y`3Ttq)=IRwT
zs~;yXPeu8zk&(bcYW~kZhh%auvZhETL{Q}I>^=?rTxWYB@$6Sf*YuRLtQ%wExzk<A
z`zhzA<DGF;{pd=D88-A1B<4i_oVA&9>$*+Hq(<bnef~{*^AzzSY4nbs4a6<ATFv*u
zP)>l@+RmsjL~A0tF~Kyo9x65SBB_{!RM1E!_3k&Y=aj%Rsy&__6nE}+R=zv?^R1uw
z5CLff`wP?VXJd(!A`P<?-&oj^H8|>D_J6cud$B)SB=CLWLrd>;;M+H4HMipdVWi2d
z(w^?N`acb?oQQo(kMj&ArYYLlTlr^;axZk>qHcRIS=0xyfj;_Z#D#pzR9y`f)3sK-
zK0B5-edpLhaIJo_Zebpuw(w=>;+SB$t17pl=j(&f2}h~bwymkd!#M79yDx1DrQA)@
z2S@d8B@T8JtfZvj?EJM9<9pG)Q37lBsC`Ydh@PUi$}g<Pb5!gUjt7*s5^$D6ET%q6
z`I{9rpY3&G$IDN$^}V$JzhA%{LktEYWK*cK04++W6;-m4e27;~SS)kIK2>J^Xx{+!
zfZc{7jKmvdeLL>`oJA)Kdr`30J#kBaE^wPb7&&l!No~TjAcueyOjf{ELN+(MAo`<J
z6MZOC)Etn-*6a*$r?KuNevI2y8=?Hqn;&tO0qRBj=Jj3|o~QhQ1bu*6OY7a>;ug?~
z=RVT)1S14BS$L}@wqT0s$Q6sIfPU^mww(<@0j|FzfSFx-&(F`_I%;uq#x(*-9*G9|
zp)M4<8n@zE<VOQ=qP;=aP4+KJ=$L0gcMc?<n!^J?ub;Tc^AC9L7LtrV3pFj|YE<IK
zyfQ-f*3ZfD{e<3c)zLNn#T_v~s>`BE4=vZV`RaGN-f#6s|1=Uz{)>2+h^<07S|8-R
zN^NmRF`EK{#k~3!zivnSUG{vA>wwy23Fstb0QK<r7z<b8FWLjv$WMSorg~Pn^i<#~
zQ1({}c1rLMX4>(LR5&Jt+WpBiqr$3>6+F%qMSDW#%7XH7z;&hwIxBA}DAr>o23V>g
z%G1|#yi;UvoKKo<7!D1!ur}E<<4kjPh!iXUnN&jik~Wc6j<Po&yX=QvOIA=a>XT>>
z?&>9*D0~v7rcbDM-y9w7jz5t$7aJK}jR?7;H!^=QIUv7eqZ!vQja+yI&_;9?*k17@
zMG{jmWtN;&c^KK{zNw!+nPYD*-M*$1dZG0}uGET=g%KZHoddV|&+$>TYZIeq_p^=b
z4DIL5Yt&=Ry%!-{Ty#W;Q`*wrvefZdrUNC>36<X2hxfWKMplMdIftp<n_oO)A5xk+
z5Qo*hs|n!yana1-2_>68p_u-WL#`Ayf#gzZA>9e1F^Km2n(^r`){^&`jwdkd5>9wp
z9B)4Ul?hFsH;V1Ltj5lQo)33ob&WF>;3hIl$u@5r99k&@C+T67+buB6dHxSc^HcK!
zF?F$zP-APLd3B$9#W7H^-@`I_YbqkDKhV?6!(wN_-x*N0<h#U&%EW(61jDsCd`<q}
z_y{DT45)1*eU2(iJiNzdubwAzfh=%xUoov0)=QxttGv2RBM=pz_7Z!yf9oQ}fO3#!
zTE)?>&Dx4VX)(1Du$|QY$#{9asFSB5B0juE;HUqR{O2_?<0mb+79G#x<^4$;a9}n@
z1}n9~E{#RI$A3aD_NI>mTf-h7`q}zBq;f62!t$KG{bkQyh&mtfx@e<Qxur`lY^8Tf
zzZt&d{5M|%Qno1}Y+C=HuWi@+BKo->Fi2DWN)d1}2-rn;j&0q??aI_ea$&4yzv6na
zf$qQNQJNVJ*q*KaI?nF}oJ^Ni7LevJ{m~+h)9XVd{@MGdqg?gq=J^yrhq6<UvWZO%
zr=0RCP~cR@uC%~r1J*Vs_gX!+v*oRk*#|edHRN!%DP)n4EfG&N!F1^j%?f~(`FzoB
zDaOI|^Q3ac(a7Ef7aobj|70>T0owIb3r;+K8_{X$=oWp3WwD^I!}c484IK~A{@L*h
zs3F&GqY}VhfHpTVec&Ery(X|G?#JI~&JW41N|UtPNN@OVWPW`qy7>F{n}R(*mt7$x
zLa`?i8+u6Y@!|wn#tBa{{5iQ%O?g$v7Rke!Dy{!m--Xcd|GreETVxXnvnidOkyykx
zkKd{bw{?mXWthzcme882-uCQ%Q{`k@F6MT|<sZTjus6G_6pJtk0O<2y@;rO?kG`S%
z?vE*~3inN-yZ58hSc$!-oTDLC#2}@)?$+8O){3Up`e(D(6tK#zGU(P|kL`gEl*^`&
zTy3U2HU7?4ecppY*_o`q{<4t^)q~81u<d;CNYb~r*(PhSYi&?Q>T*cw&8q(<;%qzH
z`j0GSc5+){x!eTi%@a~67HJ;FIedLn?NkI(dUe&>HaCW!UDuYK;x0_1TX>S34J^^+
zuJXRuEuW?M(Y~y4LtH3vL6cnO0W3cj_WIny&Fj^IR(Ev^@0{vIYOGd=^$CI2m4tuV
zxKQc?EkP#N{lQd;Vu9(On=W4_8f&4o*N?7|#Wr@9{yf_4I$|q6R+U|<dUDia>k@-G
zMW(rU$%>l<Ik3;th(Ddme?*Y{p-!dkb|6R&3le{lBXo9UtHaG~hZ<lb{i*nW8R1AA
zEuoi0G(JbYm*pLYR#u{2GWYaWTv4{KxOG2EcPzX+{aJc{sx#cFMjOO7fy16ZwNiF2
z{ee7@BbDifi?r=VXdyaOHf2I6>B`nT|L_E@Me66HVBN%d(rd|~WfyO0W<*z!;U?Ph
z_oJ2AE8<oB@prXD=QaH^p~O_GIBFpA6pD{ctp9Ub^m4;%EiM_-@7||IKtvz@rn$_M
zMYeb0w7!(0pRdABrbMS2klcu84nB==XT!Q?<`3u(XNxL)!ZfvlPL<<`80OZ>Ot(AE
z`T|Iew@HPxTKV@RWzm~qzoMW5^;>!Ba%?Zg(^Xs$JoG=Ru;au3{m4&(`$`*-01I1S
zA=sDEQcEQADZpA@@8b>{P^6QEIz5#QZcD->ew8!G`S@)Ny|-QCJmJJK%XA3C^2JEy
z@4)64WaA1ck2<_kdcTrP%Mi+b1Jr1w9a=tKYk{aIp8C>8N5RpiS`AFGwh?bltG#8v
zPk3hECyT)58|{bv-jySbg_LCq3Ks@&urWOV3wE1%r(ni6rk}isu@yuHw0C$X4xjF`
zf_a%_x;zaS?Bt`B=vZ=M+HEMRLh^7t21?M;nYF}8i9mwHmP)`Y^<MEsNfkF|IS)^V
z5;$H@1d}3;>BYzI^F)T=0|hPl6(uTTuB~sXk{(NF8?xx8<xzjM&8~&GO8dF#I*Nxs
z-c|8=Pb(LSj($<4_<uNi3#h8Qt$X;;AV`RmbO<O73P^*1(x8HLm(tzcEz*rBg3=+~
z974K5y1N?=@!#jE_ul*5_ulvW#&9@}sPK!u*IsL`Ip@-jULz)E6A>YV;g=<^E$Tvp
zIiRitjU5V?Oq=+NhKD|?Et<Ekmm3;A7KLFLwbBq%%cASq)vTc{+k(R=7hN)x6510=
zJiLp{BIA5KYH|k)aC?~L)$b8(*soKo(NAv@=A1OU%OIkx9-F`!WI5rjff~!M>nSa&
zfoQIpu$>LVHL^qKT%z&a!C?_fF$#=dr%K{8eXy%<3>hcs&%J^nvBFT#@Mt9>-4<eG
zU&r*%!RF`bcO^zb%u{>abA*O&%S=C>z7)bm_~?o36;4j2{`m@{-t7z<$o*5S)>xwI
zt}UZI?%E+9a~T^w8ED6op`Fj#6I+gOPO1nD#?;dx!$(2Zf1JJoRbtwghi0K)&Zz_#
zZz)GJA1z0q)bAJF*Nx<jtWjgSuS6kn{Ps(<lih43o=GDRuirHEa)t4Ef4~#v>F2$4
zM07CB{87n>Y;}6wLz{OP)keaMn6<a~3}EXt?4EcZAl6rptyWZi%l$$6--OP-pRVnZ
z5!e#tsdAp}wVV@Wrh(jUgou^N&KP|c%Ub`bxG3_Z55#XDOPA}@eI&mhviC|{-Q4d8
zbPvV+#G>O|Xm5H-$aVj(<I<HLVb-pAF(}hp;{AsA$^&{Y{uHv62Ys!G6TpDzP4e-Y
zM+mQrlNdzt5qop5M;3AZ&@&rK#5kA7=<2v6?A6)|RDy9({0*~`S3xVDo<+qX7-q79
z52=tj-#VnxC;VJgVU=7yqd4)g6`g3kPV|<Xem-B%8UGBH)MTtrs-%K-P@l55VX(m&
zPPon+RJDbtU}Gsrtx&Izpmcg>$u~1iDGI?q=U|ugkIU^t9Jgk%czkF6y;ckZ<uB()
zXyR%k9QoLT@x|=o8aV~~q@P}Orl4M;BU^G%*>|7H918}^>j$4a_AEM<x{c8ukZ8!w
zQ%w$A9F`s@sK!ysu72r+fFd>_iMT_~y_{AG7{9_QSb1?0e4gT}N~!)$*v2EG6~h*J
z6!9C*TQjQ{ek*F8X0I|9k9<*_%2({YRK9rt7(+t9YqQs^eyt=pl`|WMy1$%llCQD$
z*DmvN0e)JWPQT5N{H=nQJjpQ6>mp~O%pDr~NMG&C$i$+$6(#XT4SOAfLbdvgH5ZYT
ztcW42BeVqjTl`-wKA!1k-_?E0O18%cj_&l1<YlRi=_~E1)=+k*Bk{~W`WEv;Olc*$
zvQauG+KEwoa)fC%&iQ7jz1XyyCbfz&P~b3Mc0Ag6wtwhM-fDQYxGT(=0#7%3Pj>@m
zs(~z?=>g*GM1#@iSO66K5%m7UhbCvK`Nc3cM-B?Yp5ixSo8Hd)z&O|5hiDN4N_yuH
zBx(R3ecqAO=}<LHuJ-LN#=Ex<Bt(!ysi_cjImigY{trI|T!08RqG}eDb2u1}cKVQ#
zkO`jaeC&pjDLJ`rJRR4>hfA`&w}v9PCv%679DM)yrsYTo%)2zY9f;KZvKSLnpWeix
z;;p*^pv+OwZbs?P-Sc}1{lEO)?uJ+;K*?$NvZ;U;0fB0l<wX?5&h9hV*&Zb~m`2|6
zY@xzsAXC)HGL+JNU(|^w5*^nm^ZN3{?yNRoP7^Qmk8=YX@v~%vEl;BW7e>{C2B6n&
z0bb06W2y!b-k31ZBV_0xxBmuiQ-KaV76s;~*=ql%t8?Y`i4wk<2ym>Li_ilp%U!$P
zr*Bn(SIqMaA&(Yue7XBL1g9bM`=DcsS=4yr`bggY{*@f?y4L%eWPAti<dpdS`#0&1
zAJl)Gsyia{Y+e^Kyw}IWGK(%l$HNN3E<?_NJEh&MdSNUTra@mG-vzH#>!k={6RUrL
z#{X%Rc%xqhL40e0+Zl&D@DY4SSH%w3nY3bY^(an`hem*tEf6@(f6*Ti2IjIYA7xlc
z=o58)Lpy%F{^KdsghBdc)6)hKb-Brb9aW|9i!#`W7Qj8G>8TLrtF#)r(*AGj=wH@y
z6R$M55u0+|IF>(w8=rRcdCO%$JULcY%4b*k3(ooPAAB4cz{{%z4*U#%<|qRXkpu6N
z5*Ai1hQgcX{tX_m)rPh8!OyMFaRlFlnnVenqQ92qkl3i@{wwtU_YYKT2pEu~dm|@*
zY%nTB2Yg`Sg45C&P5b;Kor4Y+(A@;}^2`aya{dH=9PSD6L&L!aZWwQfpO`Abjp>p@
zvbK}qZWpO=(=<R|0hTvYb1hu`0{F(s@4lAW9ZUS57U1s}dx-l1!j2=T#q`&G<09lQ
zp6tyDgElD70$4;2Va{Mexx*-L0{|oTK@t<B^jR_hc;7c{x=4`m0_IX{^b3lO#YXUy
zn4Mz1Zeeq+F$zK?aElCf`}KjQGx%VE>Lq35zwY_Jf4~K?j96;@&qXAjxq>WowciRH
zb8!4Q4BGvHtb`SARjwEEFh19HE{M{TR}+jF)uKsv1EV2Y!LX8hAj7UMJ|2>JEBE>{
znl%$pu=6`fQ9O`K>H7oi_!I5>56ffy7cw1W``1d&NZ23IAc}4u_+vdi-W+9uyYGBb
z$PiLKVrLCrYYz~3!$L*=AHLbY{MEV&Jeo0i&S?Jex{_)4!9)gsJX%SAu&rvCW~3O>
zp7d~Zkm)s}5ZUX^AmhspP%i@NIp?&~oCmekNA_V^t^nn?TBBMcLf|zl0A9iAkx7zK
zd@*ps43HqwDsMfJ*v)+laUPcA2>>>YtnNU^q!&nH%=mBtcxAn<0;XRrIKo%+F#1&M
zmSXsTNTFweD@{b&&drr;pe!qHMkV$C_BHvty~p%9aBRF+ES-qG?6D9)B)|8)`g-)O
zPIcJgTWyzzz^U(c-=4L&2~jUvf)E3frCf1c7V?rp8yI>uox~#9xC2W)Jz!al1uRg#
z;8v0?PljcOsqX^o%f?n<Gp42Agwf4hCY$Iq>(bpPVJMZ#@!>1LZkRns|4QNiiU3W{
zBJ<!;9Upju-`xc|3rcx~@XpVdqCCt^;HKACkA6S@!ePiGXHhVte6rtz`4f2%GTM_i
z?)(ukz$-wXx&bySRn9ogQ}<uXE^J(!PFc0azLqC}9})14Z%=#--wCb2o^DN_-QO2z
za^<T4;*KO%3Z5796^OG7z%&&H{2QGKur0NojDHVT!H3XnY8b^S3s6qOfI9R}f~Kt$
zeDk)zO*LQHzJ%*x2*IQeexeKN%T|xQxtmiSVXu)~CC^=ODiAl^T$-71>Na_^Uo6KY
zEb1<SneSLp0C##v+0q4E6`zuUu~yk0Ke%te;|l`2?blv!JtO9>Pp%dGC;`4SQMES(
zD~RXGW`!7V>1Ixv*(1!ulff61E}E(@>280CZiBE*2iN-b1wXNE1d^q(Zq~F_KO7-m
z0dRsWfIVxH=)4$`*4E^MG@JuZF#hG3)g@ynzF19x?cD&BqG08jQORPO)9i-tbIS?<
z5CniRJq0Ugqj%dC5L*N99<?Qe2b}CoRkEUk`3u@r!1%CU?hBhPnaM{fda1qdH6Gyk
zp)E=U0Zs>Ne%)c$hE{-8BiOwSO+1oyV7CW?fFtzBS^)JS0DC$x*}Dv*s58%2C|<vj
z&P6|F1pDbUckz`z$Wg$>wmTuBFXGw@f`F=*Vo`wsOiAl-4ylD&KawMDlNsnz>OF`@
zFgOSAuNCf73GD8eE1v{qnr~z$IEqf>m-TJ>rNea|@B_6KSkl^2qPNlLGKoDzpvtOz
zI;KjS3Leh_e5A`p1^P^JU;=jVuf%c@(=!mGsy^79?=LK5g|Qx9SylWp{^YbbYl(-;
z<kBw$Gp_<8M6~D<BGJNvf!~z;yKyTom<G{Kn(OAbY>!cHq1&q?xXPx+*$N!n)?uMv
zVoF7Voq?BV)w<<w^%{2Y6)tAWQ&Qr5DhpNeWldBI12Qm(xDN9H+(*)SLN6#zE426k
ztmCsMAJDoifYxY-eL(i+8$K`-+N;xsBkOvgJYx!z(fQYR3Z9w;VmQve2ABGfm67Ga
zlcqgY&)PC#ysaCig7R0xmCI*0cbM|owZl?pJ<hbw9KGKLsr%RhH^PW+LZz?$CZZ4V
z`<NTZ-nPs)`_(txeek4{NO3Ht=&g>&Q!>}}PYt=TM<iUWFet#3#`F$Mf1JyMjAkEr
zpT=P?{nCTGDuWuu{+$jJ_r7s+{X3m-v<R|ioL7rPjizk{_t%_dMXq<Nl+QO|w{O#z
zq-QAJ9v(D)+-NA+)mJ-)*`DEzq=A<|8RCuXz4PXQ^q(2(2vw7a;3A8-4wfZnW(q;<
z!Dj?KGE0+w{yhPwjFbplfV*M`_rViUYha94l{e{(Jq7_%t?7XxoyPqf;p?A!cu@ow
zBq-kCpsA{5p|?b~K;H-YUPMavf-}!AMBRa^U9HE$V=!D1n)h_<#dnFOAvp!;DxtS<
zv&?BkT|vYqV8Q$-6@XJhx;>BWpl%$xatA*dGLCKasR&DSpCV8pRRoi*0WBYPUQc!4
z=$0xynV_qEc*Ax~xjyy&_q@V8u@>1Pd>}fUH%0x@u6I@DlYsr3_os(=wR#15Tvul%
zd1XKMG{vQ>$XA_i!Ny*1{FPG8)|fzYJMUzX$CCFnrlGI~bq~dvTi2WswOyq6IHTf*
zyZ6cPc_MB`Tnj*?Jff*6sN3V3s&FYNh)I94rsv0ZfGshE5~qaQ?W^AYnK&gho%_2M
z=~>ytllt7)5^90>BV1qaFPW|$QQn@v(A9*GnX#j$3Ug6=^PLJ&0#C6~<NnXnkQW}e
zUS0Z{-=>zNg(g1**fVs~%q2|uWDmiiR0?qBnn&3xKK|qGGG8NF)g4Hx>W4g>79(FQ
zD|cu-Zd1jW20&#i$di}-eDG!2MK6oPpv1iGwzG(s%!-cTY*=(O%!W<(e87&q4`KZ=
z@(P|eA8Jqu8LrruA&N}rAMef0F`#}g*<fZnxdS0F)G<ynsuLi7k^v6L$C44~A42Z<
zyN@D!Z(x#it_4vdIQ1<fDJ=tQkS$i9;k9;){JVvq^SVnz>_a)8jTBG)R?A@&Vm72+
z#FS!B2Jh?t>v{hRxmfnL$m_bBsNO+Ye$nby08k%JdWCPVEG2NaU`uB6mAAP{pSv{2
zb^(gIdktAo!(w)y22G5-NRl1kj=lg+<b!~Z2g&I-A&^~=bYX;JE^K5afdlM{dK;u$
zcQGJkXWyJ;AGv9j&8bn<!4Gr{NXo<9t?%$BA0gX6Q`B%PY{XLtrd~g!+!@H4^fAKQ
z>)vU4^$uLEDGr>7@ZuRG!5<0Qz#|_JH3hZ-4lPt%Iq!=(CO}!qP=%!T&~{9kN@ej}
z&yTl2UAIX$!>kzs5rIgFi@qUFC@HYjxaT!u9R)Iwk(dLJl?1}=T;Cy35n`4c2JM&p
zQry#Ji6H{k%j%i3#3J7at$SOM+rTlYt_&f5n+8_+l4Xd`o<F$!o&w@ouiI;oaA@bg
z%cfRGFcdQZ|L?I3HfzSC4!A9KT}8Yis>HvEWa{!%H>dry5JL{ST{#<C0U^*LCQzw2
zAl@(oNPUO*3ZKA3#;Yh0DsBxvv4H2pzdMR>a((+~*e2UDis`~MD1Ho^Im}td5-ncZ
z4-|NAAXP3F>=OO<$BXrm$I!*V<|zWM&Pq8nJempS<%5#DwmBfy8`;o8;90}Wip(#c
zk%Yo~kgp5~#p1<d>Sonb)5<Dp595+2QwCA?L0}whzSP=b;-@HZ+E2CBo%dHhQXR`9
zz<svpS@{t~B%cLtpSbIY%EJXO5EFra^|UgF@5@B^DeY6I7uld5tO5Z^9~d1h;q+AX
zv3PDUQzCGfEPPGWIl4IR`LD|m*Qb!WNxyzwr5vQLyNeFXozVzpx2n!?3+9>RTfVLQ
z*Tc$>8mUW&6R?>RTcxjo*J0wUmR*Y5qNq2_pu`iHPg;vojE$Y&?bWYamxOCs?;iDV
zDLiU8_Fo01@oJ%WlIcq2<5|dgK^HFsbG{S-2O+h?Y6FxH{aUraqQ0>weRbFBVrJ@P
zI-mH^bEgshK!g|0_f+X#s=<MD0=v6Q@a$o+M1Y@UYQBMk-7}iJXmJA6yo8eB#CQrv
z2Ssq>Y&|d%k9EfYiD-DF*Iw7>U0I>ETMYykhf+cC<c%{hByI(#%xXN14XoGmGP-XP
z4Kv^yExgn}2XcwlH?QSZ(&=+?oN0f$5=q&y>8A+!mp2}EF~QSl$340^uiN@FCsjyx
zM(*lZO)#v3_wd*$e4*r=`!tBXLyR_q%<X(3$is(i>Ntj3KhQ4wLyo{<v3aFuLcl0<
z=&G97uck=Ps5M-J4p5odp}C*UIO&`Nz{OoS)rIw7+U4eA%|js7IoROzmkCb2>&YEZ
z96QM@zI-t$E3|mBd<$~x2%T1%Fy=+CZC!Wa3XRQsvO`NF635r!GhCi>-lW%3eC*uv
z9uYn!*J79~_Zu%wl=hyAkD1S9R9Y@ouY(xE`on6-*C`VWgGlIo_3}ucfm!#}o6eLd
zb3RihxPDJT_<E!OZMcD5NEBx3Fh@Cj1pZUaAm&i>>2iz$g|k9t!x34QRkr6_SM2=;
z7`S*>ibUGxD#X1w;x<*6mAE{i)=*<9?d0AU>1QjQiW%^*@h_FzWNH&|1*PZ{zICVE
zpMRlP(A4G3++6bA?dQV;X8s>qDBg4*r6NGfrC}4IswQ~@%jQS+2KzHeA$eXukh1kh
zi(U{pp}g?*aVVk))Ta?xTepH3y4B>(?T3mrG5<9|lm%6bB-NxeJ09`@W`ME}W@<S&
zH4i>+0ryRWEqFX&4V)dSaf?;>9Ay8N#Ad=UfVbu&pSe_KWPZ@sZUJtZn=R=i#)-`m
zg3N>lwqVnZV*yo*Y$7<indl#sAdPdqlbh$EpSp@nvmcUT(vF#tBD+dAGWYrU=H|R!
zT0qv+z#^R^(5m60eYj-CaVntxbTaS0T<QQ+{Tj6HS26L_tY4Bb`_fYQhePp+O54If
z?WRD)8Lh*-N{gufijNf6TN5-&??%uXF0Fr?+aiF_yd>te&K#$lZsS~N9Fk^I?l~2L
zZR1~_?r0pSrl$c~%5O>j`14KXB}y+)i}5pCdw@JvtkLsYAOfT{Y2!?rR0a|o*2~c^
zT&IFzox^Zr7i`{n@2b2{{prk*FBTl>f%_N-MEajIJTD&&I1&qk`seP<lH|jZ`)fl&
z2bw;RsM1dnL)>^*G#|y>U)1(h3<g4+Y&qX$dQ#0&^&UTs6%Nb1h+Gf8DSSo27gDp^
z>H@0;v%uyAD?eCXYTT?_{lxcpbKme5lxefKN~}nzSo>FP;)F^RE~*u;8^a*SzL>1m
z3piO1#G6LsPG48zeYFHgp-}n*3F#5H%VJRVpL4~oVTvt*`2#&2MlPq#lzjpQc+WWK
zYeBS#6EBs3;&sRL8y`kDq&MJy67=!3BsxlePGJSc3s^5j@YJk_lo*P;29rebE(R9^
zyOPTbu$k&!P_eEiMa=I2j>B=)l5idFLVch}hL5fn0C<9fuFG=I9?+Oo?DcC@k|Z$b
zD0Rx~RD()75De(9WiA3<^TlkdL2ZDduzk(eKbIvo>;7ffrz+Nj5>d{-t0XGNL+;s1
zZ2Bp=<HV=zYu1ega;x=$PqUIP!e0c3C6kD#$$RSe-T9Zoxipaz`;gX}$QG#fvX1eY
z9Y57EA`zl&1K47=Z+N^!i%w>-J0?nCuX|bpP?kiHhd`DjJ3LVFCtk{dfND)VG5_MJ
ze#v_u%9p^?->M3fXV)D#!m&f2(onoN?8j<n$Pq<UXt`7)jD%6C@=#D=j7+FkrG$_R
zrBG5@R0Uven~KeZYLJRG&CWQR1YuW4fU6)ptivg{@`@U2I{3{eiUad+IdWTZOjWKs
zM9UK?JpRB*|0>SC(aR$tDp|Nk#Z8!l_>D|>V|t@Ob7!hApXayk3DA!wnHSwJwvDZj
zY-=e6-*l4UV2REuduvgM8I$86YK5C#p;|cp?RkB0Fc#C%mpxz{ygyeT{*-+%xnh%$
z;Rzn?K4~IH1h}<+3CUmr)w*CbBu{(;BVlsh_GK1CD*CIW+yX-cTv2pPDm<(Rt54bp
zRAWWBzU<oepzl{5ASF%;ct7C|9a%_XDqR;~THjr8uvBP#%$iQjRY2=XaSf{qaE)Yy
zFuhkjltzO>24B&`4w8Ks1KXl|{lQid_iw4fk7oxn`Yb97F_79n;bK$cE~hs8q1B#q
z_Xd?bZ~svo>yLit6AR*<$qRr8tYJ#y-BLlsbHiSKA@I4`1_LBdgpl!uLFKt8#?oGc
zyGotOznS}a`kpD%u8z3Z3oi#STBaJE>$I{p2<MFZ46H6*oy!-*`chbWPhI50s=^qf
z1em9?=OR2t0Y<b;BS^?c3(3g{;u>pkMx_qV(SDI<AAIKgRrey2Ln9q$Um`f5IB5;1
zUrGEp7@5Q@M_|9N6ilIX9eqp_srn+N_eEX0LVm2?XpHrMD4L4~J*7Z@3A(Is=nIne
z?+2=Wx@G{pIXenlhiyc?3qr@+V91*2x6-{m@2kf+#s{NrQqKq)mKtwEU5lt|jA1u*
z4k<#v)t^pW>4NCPC9EL<)%KYd!?S9~jc~-Ewj0=5B?vtncbs*HrqCLS^S)58{AEn)
zl@WH$x))VW8ySr@`Jx{4h5prv=b@!-3kJh^EsI603GU7VJie^=o5FiYstjLxO^6t&
zmW>7>d(Tzt+TK&`RrP$PB3n)Kdv-6JmdaEiC`<L8$*lXbi;n!7)D-+|nn&J70o}Ae
z>k2g7;^tiFv=iOwXO0qt!S9d27|4*jp>*6;Zx6^cekD#&d9)eW<P=m$Z+E=CQa{hR
zdqsar^Kl~T^R&$rRY8Z5@VRY=5o_?hvWf9`0;jZ3ILeMf<Sgm9*WaADr=Uxj6MW6-
zN7e4``-~;0nE;0ij9|k$V76;DRHgdiX)8~C?Pi7C4fuI+6F5syaler3c{y{cubUTf
z*c0Nf_4A18Zzgml^pVWs<cEd+p2!j6%{$RICXpaj0R<X9K2z+>&uIjD`VndK!)kM{
z0yg@FLtcw}wPWwH4d`>F%{h5F7UcBti?XlHKTwp7#U-vcEj=mG|8;Qw@aD9#%_>B$
zoSxn|Jm_jTzn{+7=Ml%?t6w+AL}k*eXY;NBwumQ@dS8k-ZC>ifmkrg{m`~Bq*E7q#
zQLkZe=oV)R36YhCO|Z+zgoMc5OQ~X;(%F&WVji-2Iw>?X-o8`!F8E*s_HGe)?ecyI
zalp6Bm|A$Hno>13Rf_SrJJRk3Oqxtfv>xbK>#^1PWm86<n0U^{0e;qEDAb?F2vBkL
z!xX_SJ`Ark@~eI*FDt(Nu9t#o(+%%hi;u-7f5`uxkYlG4ep#$Mby4_CHVm58^%*^W
zW@azhn{%OBucZ;H*+4(iW=|Dme9g*HVydrla}Pf@2M5O8|04Y}y{WvcX_Da<&r&D-
zy0onIvO!&3`{Yr(lDN<i$!Bs-!rrPzHN;~_(kZM9+6;PTIh|LNeN$&CBAg`>yV>WD
zEk=?ROxxV<zB4VK>jfu;8$(82iZUj>_}K*n-}o*6!?PQKi%h|!Hcb<gFvO4BsVE_z
zDUv=sDG8ti&&w{S@)h#&j)0dl1IL1)KBfd;s-)1jt#TuLh$C+wb{|KxPLRQJjFWO<
zuh8uf!<wFyfW-3W;IkF`^l}z>R~cV-G~v(2S+EEWl`iX@k09$Dme4{kKi6w&ASSMw
zM<el*kwvx^AScGv6)^;Wuw0L_f_@%#cj205pI&1@xU9L6#j8E2$feMa3JZel93M)D
zt8@M0Fj!mTwyx}tDqc42RvfFv($T<dj{{XQ<Zt(JS}L5iuf$EyQs%C{w%$k|TwU~g
zog%!^z*s{X-DqY+&FXn5roWqT9d*o65=}VVJf@jIM`~Uys-Gt><EIw;tswsxVIMU9
z7)uBoJt>DK)LbSu%_*s#KY2+`wd;!3O-qZMg^n*(R&XhAeqt-|Md9yzfRzfDq3c4+
zau<ndc2G0z64|cJkHC?e##-OqN9qIslGU45(w<0V-jQ>WX&8_sEQMQRd4?!AD&t(j
zL-`(+FelB4x!ippfi(i3<%PgbNXY~JT;iTZwbD+Suya)v*B&HMh;NMRbNFKB83Kzr
zyYlEOHThe?DJ4iJjZT^Uo0EH&_(%E+2I3nro}l@`V60T|S}#LwUZ&@{e^|-Tc;)+)
z21E`tFPl_&`@$YipJ`X^jaP!a#(fBl5LcLcw)>a<pO%$b0?9_dX08j>+#P(LA&$KF
z;V>?v-_&PrPbfn;NLM31DOuYtIZxv?g&Cu;bA*}LIBjrQk%5H}OFSh4>D0GEZe&I_
zrCZ*Ch24V?UT8nRA>DC08r$JLwHy<`zl=1Wf7a20+CI+I%V5}Vb?TPxZe38VqR@k=
z+uML)aKGG#Kt+}MQ^9?5)`oL^E4MTZtPD_$8GZ^Nl<uL?%aNKFMC2%{?ml&4BP7{R
zPxjpFl>8X<(1sv7iV;ldYDn!(XaG%;@J8fi#0gi=Kgq1?U|gtEWQdD~(Q2WVg~&*6
zf<YJAI9cLIM?J#oDSVCz=UG7&LX+$bfIg=4`3eP9c!)|WLchm)OpEI-UIQka{h3ge
zViOksu5c)NVq_(y=58p$@KzOlbFVoT_d2Py%aoe8560%z6AkLEU0Lgg{xU)Ks_(Mn
zrcq>FDW|FwTYBoR8JYS0AGv74Y_<5_gm!pB-Lz%b*9xV!JBn*R7dQX(Ch_P(JZq#s
zw5!BFZJq0HH3}(EJXyD?Hpr4^L!0K&{W_}>QJq4lsifFYrPdTKuzGHzF{tJ;zl(bI
zas)}UzNso2Onsd}k*W(kY{@SFYcu|p6^V@FI)jYdmu1?{$*zJue^IVC*&Sr{y6ir>
zZA-t>t}0qts2ctF?GUa&f(+Z!*ElD9UtQQ5kgbMmm*8Bb(G$2n6h2ozC}D6))(+|M
zt*<rfrdZ-<#p=BmTZ`F#_e;wy`ut&E{`r&TQ^L31u_-QO=Q`qA<wnESdIxPpIH0)_
zUi*EyBHV$!J%CAf`com8sHRQSbD155vmjZLK7bWwUl&mxb?mSKqV>xT<$ms6Mw_S}
zUc6lwvjEVcnVDhaT)nULY$%<W0s6zlQGkV%=&UI_3xkQT3a>AHs>eyQloLn*TRqO@
zn^r)Zq;Jzwk_2s8b?{T}4QpB<sBgz-%D7lf!tD5~YUDFWE5jZ~(Wp9_G@4<4_6zZt
zP7`2VH46X;dC!a|C%yZann_$*D3jeL=2BisUqagc99#I0;W>8sRELOkhRCeLXn8J!
z9{TeW6Ju8aK@a&KJ9Qqydx&*bhFXxEy$;nU9*)^JxdeMpr)vj2D}!r#C-@78ZKxwK
zSM4(?K6+^~n9LyM^mo^$F;FJxc)Co07Dx*~yS7;?%(4TxMQ@{~3^YZENZ=eycH8Bb
zPBsx{yYHL*dlrBx*8a0d=`{`>aRGylNC%>Yo?b<wUe`~;!saE?>?PL40bB%D(LoRK
z2&1EDW)ql<<@o_B4fIs^Adv}eg|ff*5|0fLjWJlUD21lItW#jY-se%GZ&2HMHdDg-
zaBVo8ld*$g=LbpH=nxAO-FHMtUVzy=A9OF~PQ_#9-L8|LIHjMS2cddaDc>>XV4L_p
zD;?Bz%E8X>dc0oh5!W#BUf@<VMX=o)B?ceMeGES_mpu=eT>eL#r#7K%=>GEw*-<pl
z;6p!R4uiP{53TLsmmiwEALBnkzecZk%s%nBQkRwbN&xi>;|o+9Mj7!`ENhxBx{=ou
zHX>~*#+UP!pN(D!QrDwPTeRsD3piV=g53K0=JK!RImoqs|25!>c;G~ipAM!`>9vuj
zgMwLTaP|&DX0`0mTv;$wHBp5nF##dNod?tVhLqn;TM$3_8g_=W>Rb4|m|+iU&Mmhz
zbS4>qvn*JIPRAzfZ;CF4w?fq?x*kgPfrA2B5k9S+7|?8^ERxQGn<gX$Kamh_IJ=cX
zidBU1j!ef^-0OOoQ7#viPifn~cyKt5sp($f9k4ICreOs0hcUUOdM*Q~!x4e%jD37j
z56p(MLqIb~XI^~zw%v!LN(CJfwz)TS0icH=O{wCKLOhm4fd5sp-RDUZy#Y+>cUx9H
zuP-G8l!etkSnO_kxO02lTNEtYCQ=QbRm;4Eb7|{ur7}`}U#>WaBJDW}{5$2xyEPR@
zr`Hk~5VMoF5e?%X4kTeWG@IOECrPjM+%u}<Cm1+Q3|KAvs^6Vn5Qv)=+;KV7W9`@S
zS0L~0qcj@sPDRF4>LGB_>JQyM`O$E+y>Npgc&u13{GQ^up$9-#suH6wh&Pdr7<vRb
z6#Nj-HZ_@0ynoLbk~8fU0sEqk2Cy%jm8ZM%e=8uHs>GUl3Sm^_!Xw1fqN?q>XnBhy
zSh@Fmj;q6ZhB|s=8#Lm=V2QjUorFljZ`4rK44e2U@~D_l4s|V^naCecb!^J2xHu9T
zr`kwTKFPPInC^LKmHnLj)aaR!f^GnbzL@8_0n6Yt(K8-Y$ax7#K;WwSI1APx36fvZ
zs(c$7huWw1ddBCd0(g7T3WqYeIRdKWwH*|0$aXWNV>({f&M7s*B<5`oATHFDSdM?R
zcmI*(WKZt1O9;^q2@A0PLBIJ)&+?aM-9rT3@zIFmv<C!2))v}$FE4L;%?)<&*pR|M
zCwA8sgC1n9!wW)M8g^axWl(1ywZ()hD3dOA&DMfN#OqoL!9^g|W_XW+{2bo)iBf2j
z_@=7yrRb-+d>#P~%{M9fq%UKS?ozhu6C+9-dNbS=u#<f{`o@zI>XqSjQ=JptR#t!u
zAsn5{=9&$M6bBzNh8j`WC?!AVnu_HSqey2`Roh~wdZnZ5gW_9_&&D5j&;xq$1F&Ru
zG3NFl62G>gr0Ovi=HDXS2W0Pe4{m>1mhZ8Bw=_Wcz~>5R(KXI#Bm3KY#{<uiw4-nl
zD>#>bqV*#}Ps^kFu+ZRr1Iv{U)reR;_et1wo9kW((#%MoNqmUZ!(3H4oit-3`oTry
zzn)ypu*%Ec;O$)!huBniL1lFhqVia82OqQplAF%Au2rA?4_(5?Cd?b-0lpndbpDSq
z^^ogM@=DH}>3@6;@Xt-oV({*~ta+0#2ZHt}S;~(aC?S%HV7j0+02rK8Pp(wTB}_wD
z24s4EXz62)mMUF1l@)%3PhO+Qk@I~k(1EP&InCUlEYK%!hmblU*7UstECGk8R{BqQ
z>BF4}8j4AYC+Sc^DQ;UejdU|ibbLFDam7jVQUQ);+2`Z;T}m;o>I4ApLs+D<I{TCZ
zL0&(M7+-@qDN$p}5KyvZdhkWX)J92@^`awdqfet7SnJbx?rv0V{^IU_(Ps-9)P*j5
z7Ng7|^s5#$9A+=G?c{Pl%7T6S69CF8ljEbL$1g?@>|#<98Ot|!<ceRFW#qc*k)wrz
zSW(F-`Yj<s4I;Z&Se;fW5NfDq->AX0l;l_LcJy-5AcY>%OU805%_p}VlSrI*<?W#X
z=1k$2vBqmUj4G2ywvkVF=PS6X34*>qJi${SlURW#cz}N|CispXBzL_J%=BqSzgKcM
z$2G<z)=q@3l`qx@l0Kc>BUG;JwUgHmB)+G?96hmk3x)o?F@IXQEoE|2L6eE<wE8p1
zh4zroYB$wvHjeKKB<=JZ3-5+r>ZouHjfj?sCno+Fhl;y?t3R<yPIy#gNAJt#R*`Lx
zY3`iCjiBr5^vjVVUObkE5~>7+aUL+jUo`~Tnk4tHU=Kd9W8==F)m7u3A{^OJ@pt_#
z4+>8}R_9Qf&27SHj!>&o2*e`!nE3E>sjY~qzNH(!-S9eoUcr4xa`&r7w6ku~muCkZ
zlo{rgGmar@VNy3&Yr<<0r5Z(6Z_Y*Zdr_?P$8JdPjT0nE(J$eZB$CPyStd$|MN#0z
zyJWru*yz><d_wPo8ViuR^O%-<d?|{Uav8>5`gX>32l7avEM4`4vn+gdu^I`RH*-4`
zu;-y?tu<Vt%qxB`v(|Of<uuPjGON|u-!4lPY1gaMLzWvPj-57+d1xU$8Wa+x%%yX3
zcHf**Oztd>s4og$9)X7OFp4+4`w*0iw@3Y0a$i<MDop2D(iRXvbQyx)y?jr-MO(m#
z%Mv3E<wp1l9%OX~2H5*Y(CIgC-5vKMn`(WlHRUI@AgYElLMySa@0GD8QJL~=RfM=8
zO7r25W_NS-JEptW0Tkt?E_-&^>J28XWcdEt0mxD`c49SQy|1#4;8@Z<<S(O$dv}Ac
zP;n?U9q)uG_NbdHH)pG%%J$Tl=O8u-FiR=Uy5j~Rdn+n}b{1nOXoE>azY+N3*amNM
zAOM-TN}p;Y(+~Ucbz9swSwk!`UiN01%)h-;AYY?y60fQLa*WXpCs|~%2mPjzs<iY!
zd6HlL1&Oe6;ugHfgqy_N_B&l*WSiDx$3cD-Tl*21Hun$=_4^h^n`l@M*d87qj{x5H
z5?kne-ms*N@TecaTO(d6?OTgE)ebVd4KT|_29=r8GriJ;Clt=}K0-b+>WRy}sNxS)
z(V@O1h+mHBI2yxBCJlGF@$Ed4>dRZ5khxV;e2~OY>;~Tw>I7j<n!Z?$V|`tvo5oOI
zRAvLo*HSVDRfs_l2U7mNZirQrH^NgCBbSoI+B&GFyX&)Q&_&2z$33FTYJEQrx{#`q
z88cK-8v<ooAt<o2EXu{lSsu%B&$$&M+@m&LkV$B9SB8gF_sQX2X0s@G04~~3qXez^
zM5Yl#U{9xjOPGVSmK>%`beX(0ybPYn-atZPSSM6dlOD`zdjY7LwO`oS3{y@c93C-4
zB1yQ~y90y2?Wu0kMXfdrZZWG!J$haEh;Mmk?<LOID}(y{u{o8z7|o2|0*zs#%<O}^
z$0tnr#lDz*Jq6!|ne@p7_KN93O^2sF=!1hE3-S}wu2=yq+|laMh{Gm+i&c7oKnEdQ
z2!@>q>m~O1Wirbf+KIL}rCQX<*1rT~Ud9<kMoH&;2_Ux~bH&rh4mh%=6m%_*l1bUJ
z+>2t2GADiL;WB5|DTxON_742&2;!M_Bd=s!gag;}nZ(lF2O8lmJyunyFY{vAaV^2v
z4Lhf|!}G(o0~O@WHt_y;_K<>M>p{Z^@xg8-_C0oy6{*;(ld~CjgOP?UypK~f7M20|
zNtRICS+6Jos<YvdbbeU8d|7wFxapBp{znv|pU=2{317&6{aP1A-VSmprfh0PjgkJg
z`Xf>}r4yaol%l83qqsR~$j0JdPb~mjvmvA)W6{4y=l(`18%!XZNhq6%EBk?<NA4@b
zG(ERph=Z@3WgyTh(QkuhydHeGFE@vJS7ONbmSVkD>A7FdD{*gIK#Pa4=Ta-t@S~J;
zhpagPs1~YpRiM(a#dbiZBsbT88V(pOyLXHi+KYH>B;j36U%A3SkVc}E<-g_Y@HcEm
z4)swe3%}RXCCh&HxtZ`gO4;G+TEv5=k3kkcv25uKR+V2h16ElNrwn_v*rW6+QB7kd
zCis!82PSYHRmq~`nsf3b!iIFm%v+?H3hv9M60?0fe}~vq-1Ru`e9u^Z-jkm5aBhHr
z&Gd8Hs^l|xyUFNL4S<AwLs$j&4!o-auz@hmmHm^!wDRdYnqpo2-R68k6#PIGJ3ZoD
zq}7H2HAO;tcFa|EwnWqX?a8bL#W{T~c4`H786Aof!1Qhb#IubR1C=>nwcSl?8GI=e
z{9WXfVm5x25EKSQw%nYJ50xJGqXX?-AIx~p`=Heori^Aaor1JsduImf3g=7B#*=Vy
zmRMz`CGm15QnBRMY<?%VJh_4G1p&9VZWw8qKVeTxLM3KdDq(Px@Fl_~6i}Wh&8go9
z@gDBN$+GMmq2oN0JUfSg`?7!2D(<6=W13U$10NT=6{bdlY5eKQUO{)ERYhZ1^uVsR
z^8(841DHnQApP*CS83rLNAWMBnZq`n3}zo(KS8J-?}d|?5^*0X3U4qu7Y-;WX&c9}
zCKmT^$#Q#MR-zE9QYCgzTZ6VP8a@#UJ~C_$G`}ciUlPabcwXqLykx4lyU0WnDR4ox
zCx9=-v~amqgaf#cjX6_JS7($S&fBktVZ&1{Fwfa%=1SfPei)AR+QS`i*6QqdVjQo6
zAI6WmI(<?W;M~*cS=m`*v76(E(R-EjKAyf6G%?cJXNPtKF4*<HH8LsYI7iL9n5m@s
zh3xubkosY83n~Z{xil^(4TLti<%b+0?wJTQI-@2s3hYOA90fUc!yx@^8g#g_+uG8o
z{b;lAT!aSW-;tJEq!W$E&Em_7?5>n8hB95k>`y(ONyY=z=SfOb)?9*t=P^(~0{juK
zYGL(S(f%K39iN!rF&}5>Z5N{71y)V@CYo!u+xv{)YK^}z5%59e2)Rf9?fP7!mu?n6
zs^=8-CVRJVL)cF>dPBU%wkH2o@R%qYQSHzk^c=pGT36vy)I+klbC6=zeCLH<7+!Bz
zx&o9lx*y6M12sk%C`tmoT&)j@gpiNi{$jGw_9hSBpe+p9_++CTZ^cJ}AQE-7k-_!H
zoDYEHg=&cXmtjl*U|~()_IymuF#`J++qmG7_N#*!3?nF&@dk4ufdCN~`^FT03m{sP
z_6+Hf6iIK;>jc+-XF)X_q@qE&!5w5bhR7t+6pEr3_i%PsO>o{hdOX9>FH}}R(4(af
zuZ*>Nh?c~r->ZC;Mid{4yF2n|6A!V(6|sfdjz`n+{VY&AJ?OeOGEd~MfS#Pi7<?Lr
z9>ghu8<senKNu&TetZ|5DGvp?1N4G$(AgK>tJG`Wmr2Xi5h<5R6Zt>}`}E%CX&EJg
zzgmhd2kpC0*mhRpt;QZ*Z4BsJxR&<-=_y&4)vtb*+pbZ4-1=4$R!dAtV^kJM*r;+X
zzN~MVkVnz~8lKI3pDRcnl*d+V#7k4!4IW>I;awK5lSDHj8bMbCcpe8Z&ORr*YA3Zq
zZdb%prE$;l`eMmt89JY<&In)CB3J&rkFQhnV^z|N>WKCQ@(B-5ufH^F(xcfqZ5fHW
z2V3JAx0lXe2i$PXNQSu<Ju01JDYb+%Ja4e!y(qpE>HdQEMjxHLKu@TleVWj1YX178
zZT4;Uo`ksxp_@mv$ZYvoO?m)?nefwIw#TcntLA%nY<rH(v-9;_q_`8apr^#z@HO)A
zN*4k78c|=?-;enLm5%>5%r=~jO=j^OJ*7k=TgMfgQ?7dj*B+c}-E`IeZAJdm>P*3H
zdXGU9M2yQ%)g>JS2)K-JLlk{<e`gMM3~2$I7Ovf*<B*li_lTj}7trnt4tAxmd5DZK
zdZoU|zPCfEy-?{HwM8&#R&a<J-rYC&G&I=Hk)eWkJk;`B37M8NSX%*VOd1?4zZCqb
zRM%SW(8P49_U~bQy7${EgA;zTpBusU=|#vd0Q*9d5mko1x8%uBM<yg1B!Avmy?Ehw
zS@MQk?tFr>iL|Tv47gq@ovPxn&XhpcA?+Qq^8rVFm}F*<Vq%3)u#fO0#R?cnf;n~{
zrOgeOA^ypRMki^DA~RYG&Y<Pfu#X3qbr~ZVP<*92;yoEHNo3#iuq(VyvvRuRvxXmq
zMU{<VHXnqBeyNFw|03xZSW%+u&bplGOB~_Detn6uNx$()@MyaE(!9V^m}OR`BZFo9
zxMbTA&j!ky@)PZy#I%e@E`f6c4K&XL&pEvz!xM#rPwe%O<g@)7QT~<LW9lzYrz9UW
z9>YYq_MR>`f*6-u)4Pal7rm?UW*C#>ecapVj>A>I!^L{W)9dm_>Ex<tk(k_}vQ|H;
zI?-sfx+XL?U8m^D>*kaihxXEc@j6qyI&)?dfe>#!q_IUjXkRk|v5Wkzc06>{RA^oY
zutoV4;4D32VYociGrraqv{L@Dl=<6dIsB#*M*KNdo#v;%C&A%)$Rl$n`5z1DXU*t{
zB%p(x*Y=kEG~V76Ti?-Lt$_59z3u+%<9sZ}AHRR+B|4z~==iCx{Ppj|&42nzrYfSU
zQnMK5u2{)c(2qEfGS3u12;3Oj=;7BmBO`LHb#~ZqcSZwslYmAu;M7#qV*oAeQ$4^k
z!U1$^Oj3(-B{v3@Z=j3^DG{+1I=IK#`#(4z|Cf6_!EaLD`au1BdWP~Xb#Qas0ua2+
z5E$IvT(<_2TB>^a^Zh2tn2hzltmS4&<ckU5?_G)#p4IZv26JzAyX?ViwEa9AlSx3F
zg$LKs@Pc68_p>5sZgf6-jkiKr51uf(gZw4?SQw}$cJGnkQSBOk(*Z&Txs9qng*2J`
z;BqI(M;!j%@A-!e58Rij9U`E+lqkGk+GGs`K7HIhKv1R!kP&Q;%VP|DxhG10^o-U^
z;BRUo>&TyWmLCX<TM{`85T^ib)aTa58W7shp9f*zg)5le8<+ldz2IAOkU381Eu4(&
zuw`eU;QdGW)r=_67VKsgy8WFG@lUCyND98LEes?&SaY5*i$~dTZB0}1&~JjOoV)P1
zwY~wr*yDRx3vlHC`m-{<!=DT|VQdY~j!tcR5>O=${hjSW<n~B%QXv2ngt~~R)&N;r
zJaP_!!jSwQDeyWXTwQ`-I3xDwnfx#G{r~pIP!J+@2gM2gerEq;g#6z>+arLrUJEI;
z{8OLypEuXv|5gHI6~(4mp$h*$Z=l8-)U;X<waUM0g8y$He6m1bP<^7*_IFhKZ=vDu
zR}a3TB?49it9QP}KfV9|{fU*wHvv}_GPur(9S|yEI%a!cCqUz%#?O1w#Yu5o@ONFq
zWEz~^t}Q10U*C+#C*+3P8<=$)!CQYIA)H*&0u<_&2WI(7>w<v)RkIJ5gTSZ9m`nh&
zDrSUEbcjeEy1fL<RCzM!F6pX;TM)S;x(DpkV$|Y(5xPBw5Ab1lRtMw@MrZ1l)hjKa
zIj31sq}BLk`1UB{wyv1_o8wL6y;~sF_$^(6kB-_06CQ&?=#v>`!CHj_?3lM_UbjT9
zCz}Ltq2R4nW&tZLJLp>cCIZ;41emY?qpyis$6BQDhpPAltfDab@8|y%9{=6iolI_m
zlLL#vt_lDeB8EDk(fLjH0rYdMPkP}BI%hx_Z2_^AJYN0%H)}u;C3?_EgYTn)w;p;L
zEf3pYXl(`J2HdZJ0e8Inwc=ezdA+C~YJfrHv#;NH?Ujy@f#zXmf4>zch)(LgB<IGL
zm_Yc5s4FmbKF?+L+;bmLWf}2t>j8=MFDXNm2>!4_eIlqDTY$<$`DM*$->n}g{JpS%
z$V#sYaAL2u41kj8$TL>x3c$?hU}S(%)j>OHE2uFw(48cPh<_Vx{qMzo?KIzxwrMgx
zJIuby0ea4IYVfqCiWOt&{-+TB_lw?V46jMQ56I*eV@d%ZB}Ze8A81kT)Kf1ug0e*z
zKs^I22h)Ig9nfCXDW0i}T$AMiE%yR}h!q(Cp~H7d-a7(3D-e8#p*AnDOjPS4gldDE
z0DbZSDi$#=E`6JK?N|sydX?<_GaxCb%9a!&_>SLHN-B0jTx?zq)=&_Ld{1}ZZ}lhM
zz{ez9ZRy<JQ4lx~r`}NJzr~T9Wh#DLx!-~rB;;|4oPXCqZx_6|yE+~xaa?P>Vj=;B
z4Etw|w}Jv70{YlK;IlWK=Z|L(!<G<!5hkCp2+qBAK6nHqH=B|AZviB;7rh6daC$Qi
z<98)6AqL#z8dmk*GRhbbT0D<aI>Z7?O3>%K3s<gHZG25>A=EIBChdjy3KzAfzpw(}
zs(R@)as*ye*C=RWBhm~JpMrkrK&{iBw=)YW^Q{F~_6RcGcfjGA4Vc-xlLpA%946pr
zI5-22I3_;id)jT#ww31*Y2{5&-pa!S%^|H@$yN6uYz839J7%?TLh|`-G8?D{S@0o*
z*JL}W;x*ZzdJ+XBKxXtWS$O|^F8=+Uue%_tvsizG6t9h9WS@H>tIr6*MLYXpI#*fj
z2=lkE3JV;m2VP5<?X<9^MW%O^nj+{X|2(g&2jdN^niMdxCTj`H)gifSrfvz1uNV-u
zq(E#d{8HfdUsS{*D2P=r-Pb<_Nxx9hu=v3ah>r0MOMN?l1ocQwIa${iW7@((03lBX
zuq{=u9^j_wodkHO&bR<Yj@U;!J;052GSkQ_ta<+%p0ZY{kCb4@2KWh0oERIh!)~@#
z!x*{M-F_8qmz84ZJsIS~=)0*v9C_yAY67&dBI-7uPJy$20B9{tDP#aI55#anE&IXk
z74imyN9%x?jQtxQ94sXARlpSi@<mL4@UQ4)1y4pkjUwihL-xU5#bK-4_f7Z-*JOi%
z7!^&(LL1)*l8r6^fzRz%&eGRuxE1Kfl>)G+**+B7Q&}`#juxdVxC)=_!xh=^zJCia
z9>H-sk`k0?l6VFTFzh#mHA-ig!)WmR++HUx^c{kYV;iqpq5(v#NbzfM+2lT8sl!-u
zlHWfz3Beq(&lG$`baOf07zpnXX+5RcO#*SL<#AsAJ!62d9d8g+h6R7FokY7PzouZt
zI|G}=yLMIwu<gkYp8=_Wn?{jQTPPn!t9_<b6N~W=k&Js<c`c*~^RZ|?4SFDJLkBQ)
zgvPip-p<#zMcUf&ahEKGW<jT-e3LBpL8b*so;26n&2kL=Neit(LEsVd^Z<s@jb@rz
zVKw6Qd-tzB0R6VCE=@mG7q5QOeF)MO^xvUr03>aC5CV&7!jl^Hw8(dTQ{#px-cTpd
zW~oV;bjF#*83U>7P}!s1sEponyF$4HvstH5n#-HS3}a#V>+!m=72NHEyPG-HrbR^E
z#5#q)|MZK`hsk92h1dB4T2yj_ON<=%fs=gXAose=udg!$35R5K?jXQx?+%JFo<urT
z!wKC{Ik$rhNAAHsRq;TNay}}Y@@|d8SKuPp2C6X4VsLRB_g4F$lv=oMoCT17^Z@b2
z)@iMv7^1pt7hCs4gMkJO&ULjHhUYbnSh0xzF~<5Fq(XdsX(M@Q;LLv^YEO7q7X6{V
zsCZBw53;$e`r<1>1Q{$@bQ}<Bqqv$~0Xa1QgoIIv0NFmypI!d^2eiBDzz(1l_;B6B
zWh<v~2hh?2kcvB&AkaMUcbrc6Iq3z)%(C|AaUKInvO~}mt*SV#e-oAsqW%v!vGyoX
z#5-gOPQeav12oi11y-lv<wpL(#+yrPkRzk!fBt|vEk^^Ti3CPcKP6tESE%nGCUUmk
zrC;%V8SJh|zwGyO0rW<w@`28Ype#CQ1r!D?88^oa86(+OIcbiQ?4U&v$W&BU`eZ@K
zcDd8~>YxK=C?p0-5l%I)u17ika9)2by*N8K{=SAuFT4;yK73v;{sf4q+~Txe*HmRj
z2r%D~_N}ak+gih+c2?aKAYxA?amNc;NRK(B;Iq#?TIJm#S&+gipTlIWYLd8boo(5Y
zcD*l7RtpZ#@%t$RcrTnPqQL6ossmlkbRB3cIwKyP5mL+Ja|r~gQ{FkeW`WD~&)Ls|
zi5bx*m4Ap_Hvm0pU1K&|FpX+gWQ_-qanuIKc*dSedp!ZPl|XRjSDW&XTJe`hnI})K
zxRqu7Fdm*W7xqz1_e?JRTJ&W%f<ka|i^7IxtlffRhf;~Z|07_nFvI0P3*gkwd#q{3
zO&M2c7Why<U-)VP=cca-)&y4<D-Dk;jsrRFbEkeyIjE|*P7L7oQ%r%ZLTf@n-*oOO
z&Fkhuc}^zz-5OMDgdc?rj-fDE@xp}(;Iv#BR20U0_%&mc<HqCk7sK(+6ptqhtmid@
zv(+!q>ALdxc=J&h>ROI*`QG0ed8iG>cwz?n8LV%<WGrjN^P)39fJ=}y&o}sA-A00V
z>lL(;7G2Taf|s7h@nouR>BQv_Z)qMsdJ!OE+gP1kJD5uJ4ym60w)3fQCWL$B>Nos%
z6!4;XT^oM3vU$w_BZq&NjQ#b+X+HA!KwK!C`-&i{1Dq|S4U1zPfFR?-|8WbgxR$JB
zhqsYHK8&5j9mM)tDg;bJ)r_z)WIRSZ92F;Pq9A2LWG$smf~Cic_-2g2oC7nMpxD><
z;X`@}JP0R3I3vyhf=kO|JwDyE;A@jKhj@*v=dp0rKa7qOkqm00y37ZH-%K(UBKcx}
zAduBfp?FRcaS+7c;vTZU!^#2Vp56<%cb|}y=d8<`(PzeGexoMn<9Fh8K?6}1U4GZg
zNRsezzY>_PE~+%4z64n)yAD9dj?Dbs0(7Vz;-O~|iF>0<<AqRt*R*LP&~guKBS$=0
zUH=?)YWoYM)gCkyqNMuwkl9sRNV&}uH4p{BP`1`rGAt|O1lMdhVKhTxNz{u13;y_Z
z%sf-(TtorXtKk;*J&7E%UyFNMtAi02B08gRmv}+CsX5=p;cqn}=Z~%0TL(Zh&CF-c
z`}3uqRI8tYd24Dt;649z9cD(FZP8A}pNLc~odp$>T^G<R6X6o#2L%^4+m(&i1!1hY
zVWWsD-*eA`WpPpMaG+YjYG}!c$)0}YJ5@`OmeF`h6Q6=8Hhm4d7|18HJ-!T&$WQeK
zj~#w#{N?G}5WdFLuZXMyFB?S6+sgb*=yzPelQ=Tj^FgCn2P|z8u~+GhapI~jhe{ih
zl4?VZwCy=jv*6?uh!8-+eXRF1=E^qQlN=A<f|@*Z-7J`m2_kMDIm$apHD-+WeQ@kI
zQs7p{M){NBzUqcxr_fGng}IDAnRJW%J?aYDeKJvADUTl1^j2o;&cHec-?jUD3h2|I
z9f+Wl*$iXa@vDK_m4F7qjc1eHD)7cTIg^K4Ys$EZ-hVN<uJf$&;}4ZZ;{#MoF=tB2
z0lDdaSo$6xO*-%LOwaebFjE^4B%Ml-9B6lVsG7q?KvlT7TUinTS?Ql$w#1b#l;EvW
z6x?F8_`ml<|E0teA;o2HR?&Zc8-fAJ``V7|8>067oAb24kEdWq?<-~fcq_hC6V|SK
z_{enhS28!}9m0;h+9TL@e#wdRMbGWI-=$6%e#;q#x{*hAkQ<W^^m8zYhE%6U)LsuW
z_tDX7A1^IGBZb2+(0td2^qnuw_4yXA?Ae>;Z2|9(fxb>!78;)@k$k)$2HZI_=UVAY
z@KbtN!=}VPXv=^3q>{8|^sYsyh>M#<%klP@XvhSA9Amq5!Ko3ZrSIDa%?F0E^F0*u
z^DaQ>_mI0JEwPi1Lnem2@6r1g+Jr7c2OtliGycZP5iPCn`x6SyH5yFn<mBj14#7aI
zw@gOQ+~(Dh9r=*hy$tpD3aKuaHO1M`_4Qo_NV2Rf`-jAnZm7S>Mr%VFPo(2}%2)=E
zbPN){<+5MoBg%G<rZ-lOnWo*q)VIf52C~O1u--24yix=xx<1<GXqJ9Tn|#4-v9n)4
zd&tOH;Btw9pLcN-?a3v?zkVj$6q~3k$Z_!c(KmYM!y0|D!=VNh%O@=#8`_xlT<0tn
z^IvJn=k!C>eu9=#Mq>65lzHbX&V}P_1HEMDQR*5;v;U6}H4mAdG}}_3;(ACQzdMl7
zzV+UNyT#Y}u<rH!!L#u$nX(SqCuxb!WxL&Bx%e9?gOcmd<uVfzgV`2p)yA08TKIAH
z$V<rEkSaQ<IGP=<HZ$l^<qAP3YZ>&U7<YK|)b+<cUwaglbNz_UNz2JPu;+=vneFHe
z)zfbo@Gt+0)E^ECVrGuPtc>SZ2QD9lih_3p`UkVN#PKea$iNEh#n;V_{=ck1M&Q}T
z>Zuj0%pzmV!OoHAiV@SHh1YX8RpBzltj}LcsjPD;rd^isEvD=k?LZ#g%cTe6Cg^KD
zzS=>IJ+|ul@!=mq@n79qQssy7&hBh|!}JGaVERhszQIRA@nl)7ai(_jz=!yo#%)~z
z^2X0NX&60TQTEhig^ibHaYNmMwJU&L8_H}LSNPM?nj(wIKYL`U5WTm!GltcUbc-ev
zJ}^%CGhjfnb(MdF3^31B@3%3BD@XWovNIWmsNvWjh@V31Y|3B5I%zdKZ9{-84pjp1
z2n)g!E_13oPY<6@$6*_P8;~g7@YtCa8rs<I+i~swto+WrJVOLhwVoA5Z=7AoEQRib
z$C3?GOs6eBnFHP<Mr#$DGPKO^G*(0ZA7^I)Rb{^Je?*Xth=8bwNNp6PK`ChoQ4tXd
zK|(q=rF4p*f`A|$5)vXM-67o#(%ndRiTt0}IrpA(?>%Q+>%Z2lHO>s9d%y25p7?%0
zPY>>Kp2UElI}#k~zP_L|PgB!8)F*Kk0JBQ()h*g{jz9`=zh%-$Fma;&jxw=*^zSJ3
z%}wQcT;J1J$otoYPo%Ob-E!JuWYRUKE24#G2;-%qwmnb$DN9)!z$o{+jyS6<2A>G6
zp7+h)EZrAvXv!H)6MUk!bZ%<ymbOxi5cTqPESBPjHBO6AXpMD5HQ;6nasLzU_TLxl
z|M*2LB`PfQIxuNc-3i961}*mWFD5mqVbxDT-cwTYy*eu0YjASqU5wNfOOFyKZWGqZ
zv||N9hW3X!T=OeOP)J;4>v-^Tob6X3*;gkSmf5ctEf<v-$8Y$XSjwofj^7}@SePV(
zr*?W!!o==iA<3lp><pjTyyo}&WEF)bb+2_3Uoz_s`%c<qU-m9#DPs4r*!LlrhiJNp
zg;9Ti{mdnfiV<b~k<5r0Is=cxyQCo@>Y+#4+O02#o=#S)8#9ZMutdm^ndMmSzdckQ
zrhY<D2-7q2@UfcPrb@hSkIYP5c%mk^{Cvz$;IW$cRQYAwe$MkGQUZBIYJw4&Oyf<&
zwhfXV4kHPZR6`ybqx3DOH5Ymdb3Pw)PGG_vEe|e}mf{SbsP8aaru&>4CA`S$hwlEE
z6jMRTx*}jSxA5}tnc9bF2)q(Uc*<5yhjCt3vMdDe6d1>ogdoCT4(5=RQ!aquNejQ4
zb7|>ot7RgkdJv#;Ya+X%Exg}*E949=u6SA|nP8{b-LH=E&7|27Ts?1p`0LHhCv&)^
z>>I**-jqLz(dJPNqtlR$T9cIpG5WMTeYO+yZVc*j2b#Uq($<@R<shIJwI*8<jIr3o
zgjk(iP0r=bsZd!L;+7SVWk?Hm3j02bkbQ8&Rv^k;m~YBoptH+ge^}vYgn6dNF!M8D
z?$%SQsD|(h>TK9(|4`B0`Lotf{x(a2vVj|mTO144sAWCnz!w3hdhHLboIypZix!Uq
z1uQ;zCSE0_EHvR8j+(q|4OJeCz9i@@d>Er(Hux@<dr4uL?=Z{%&0-#=LWbF_kDmZ?
zNC!$7&ifA*fQ&Y}Dx?G}AAdRJUL_&N%p+e(5|*NUD83IngyT0gr3p?vzi9HLFtW9H
zVTwQPz`y)jh%%a%I`0?ehF~~E7|bE~mNSHsS%y56$_#OAwihju&fFyJ2#Pv@+`X82
zmR9^t&~nBOZfLmj=7YEBNyDU_&a)S>WX?z4QF|^IZ6n!;Pam+Xc_I@o3KjandFl8N
zm=Y&sE}|utRU^?w8E{(!DMf`Y*FR-2Eu&<`LPSN4P>b)VY_(H4^$SXgr#9SFH!eY(
zqOB&_Zt44;Tdk{Rvks?l-9kRbw>R~#H~smBa=oPQQ_zJmWZ!#t5$hWiAD2EwjAikA
z;Tn)u9oi!%xo_>AJD($rQ*fj|;E?(#<taoJL>&_TI?M=nl90|Se7nN3wiN4}7mr&3
z=eq%m6>y<q^T16}nCEl>AmRTjs`#J(*dr5r{lqxfH@g>vTE$egL++j%{(ONPY!;MP
zw7oznPOsm1jU0qKB^xtgDqGTV{jm}AzFX7bX>-3&XZQ&Ub0sDEjmaun;ULq-$MazF
z<RE#LFv6w*87SNRn~Pi*ECIyQ#qgCR4``0A`YKu4Y9RM*j2o~@E1HeCFruqJH3YSo
zk19%7cb&y8jC=F5Llu6OgX5h+UnX-&nFKky=NGUmU>1sAa6)T27gJAD;7{~{m!%|@
zr}R#8QZ)bV`B4?5^jQWV{{aK(BxH{@*hRMsZnm&pm%?f-!84obZjZAwEBoh@ci8Ni
zw}<0*v)2oA5JA#Lf@hWsR4X&r-73O0B+6d(Eo`14zehPSVg*N+?7{Y8Xj!_ql(${G
z4+&hhIfcunvw+g@>Qp$Ib}j)wx@4bM3b7SrHL=o28|qq^HR&Jr(ELZy^H}dMbEo`h
zM6l=t`AkuoX8le9mcfcj_k@9FaX)EK_yt*rW~V-Auf94w%)ui*k`Y|5F_ZN68o7Vs
zHg@2r^mAAdL)A6Ilh>3vQZ7v5n3W)Jf}A|edqEefuCifk*EH9O2m}yD)&|gsZbmEZ
z7_B@EL>N@Ju{=F9OT;=^A{yToy|lhf7?)vu!bBAGLEe3<X;{QA%C%g!ITzAdKTk<q
z;)R@@J}KJ4*r6cw_9)+Avh$YJ3C+|&JnP!w4tkh(FR8M>Tf^>qI&Wh-%hN?$jRXsc
zEk}vM5BDSPCYuPJB$-cSExOS8td@eVTosf<Wyf(3<sWVZqCSG`dEf4-`0uv^Vq#}k
zadMOK={vYDA!Oz;YDLXyUx6ZEFu}1h$IJOuu(dw0GeDB2kss9Fl6JCOww3DvT?Qn}
zrRT#<xdQzPMje{dpW4vR9|M;lV6@E9QDYD|f}PgrRz<|msV<8YAq~C?{Hf(8iLE@Q
zRYhtd7>vELZz1DZ@%6DMb_Jp>l+<|eNdM$TTVgCN{FTp<W=&LWN0gr~u?;g71zvlu
zHhO2b8s5z0<+Gia19wzNlv>7tPZXs29dC4*35(^OUdzou1h8LTvLLDE;ZOK5Z}<rs
zd%$MMH1O7J7+n##uyqtJh}_x*<>1}0H8%e9t#RR7v<l%s2Phq->1{nV$R2m<$9dP_
zAEQ9ZiTo`x2BKSVS?LyKvlLR9C%}pUFjUvC!VkX#WEjTyHK!QD8SS0tuD#4>AaxD>
zIcl$lRS@U==ohN205V!#;3(byt(MjAiX!O&t!lzQ-ekFvnUZB=IcO!W39hER2GpG0
z@k5aqLi@vrOV5;?cA;KFk7Uu_SWu$uzG`aEq0v2H5Wme7`R4w*4#^%gIOZ+Bc$R+?
zPrzL=RKQY}h`P&ALO!8Z`pf5L+ZrxDkjG@J1elLFkwVDkZUB(c#zt@gXEE2~qLQXC
z^wg!YS^fHebNOo2c#gNelyvi92h$17=sCmxpSK$0v~lkSwAinri2LbjIPAfutj^)V
zJOKTs2Hcm*(UZ!^(2>A+W&FjVf(%uj+iNcnjMjI`;lN3kRWN>-d08{kn}@fQO}wE9
zKp<K{f-D3RqsyQ;+`^Zq7sn%_`l!2EnZR^NjLVYWuWi=I!JII8h2;v`Ll9K9{Dqve
z<s3DqY4#<SE}m?;TMyRmWhG+E4}}1ZvvU&pxh5{iy32rlx2q89=vSdBy>8kSb=H#x
zeoiRnlJH95RB21MUPP$gfrcf;ZZqjaKCS0)N|j`qe=VBbIJn^V<e(;nGH}?F5TWr;
zgMg4#z_+yO`~;5DHEOu3vZo}SjBvg0bW08(_`^mxDW}k)T^^nncN<}f5)d2$k=h=)
z0x6dEx)}tucP&;Oq0V8i?zceW;z17$C%YR^lXn>TI$aKK!sQNO?P!t`+B<J3_wN^2
z4-zQF{+@)eSqGC4l)olKB_4Yc$(e(#Zea)S3C{N*(pQVQn&IW%M?sSOLV$Fzpv60q
z=RZFpY{7U7R*?N=!&C(Gqr%^{YgS-tSIN*?{@7k2hcALx#QqRJx<t@1Q|t}FuvJC6
zP+Uu{;+;VjYurlx#Qccz^3G-bMVnG0Jxd_^9SGpI3rI9C9JJgdUwj4uQsvWoJs%4)
zJ~_*iX%_$7Xs%D#XJ{9R7BAgg8y+VQaHjSoo7TEr(th|*PGXFDGhjE+fqaO}3_*d6
zbnO;?dwR$e_mF1IZx`qc@}<83G%<v@T8qD>2N@i|^w`ySK(ah9sQ9=G+B*9vO_5cG
zLV~t8rOO}p5Pd9W_`FWz1~f75b*=gf8bE(QtiZ`@TY94j+sjxluvLd{Y*SDQQUc!u
z#W^^gptXx5VNUt}&zTx8sb<3{iz_Z^0uJdaF6laM<O4$ba|8e*Q&UEaBKV50Wcc4u
z+75GKfNN1n-zJ^R6PjO%<1&FXDtN-F9uQx`h}{U|pzXh!avc8sqo&e$6&#;b4|E?F
z=YR19j0O(ry*+2sDv=f-Pq{oJoxqK}dX`6yoM+H5UQ?_>6YZ_})w6j+E2Ogu3X^0l
zgwHd*Rx$yx@fwgcLQ9}gPP}@Sf`<hu*G@nO_zb-oie3{%FHAr`%N(T>_fa#jdje^#
z%+B(J&G58msAyV-I7QfTAON_^v~iHRUV1Qo@d2O0JG%BaDqIBg9H;fwS;FjS`mdJF
zA<dg@Lq?Mo(~xj%dkrv0xXYZwBQ>_;=VN4VzQ4fRDaAo+lzDGnGY*x3H>;oDm8^FN
zBU)!gqpyqeh3h42R#CCj{rG|}`AQ}9j5K?h6;e18WY;5^{@HaJ2F5hjh`8C<K>eFv
zq^Wn9>=bF1-O<g(h)Ndn-%neah<^Ns1w~H6rIrIlWTbf63haR4$~5d#Kc=DOcctZ;
zHplJTNFnuAQR9+@kibS==g2Z$R5EmM_``5NYkvJu7S#rcU*dt0w*MNC*iMluoytiF
zDsMTgf~>#)FlfsSx`1pg&~bH<aS*W#%VQkV>rJWxuI%{-S15ZS=F&u$i1MSh*y(A`
z>bG_tra*6MU^h~-Q`3ye9sdpZsp}r;9&%_s;Ann)$#Q)@`Q%ni@Q=r(t|d;OoOho9
z3@@Yf!^91absLWLCll?lf=4*cmu5^t<jNkdV}!tld_*q(=7B%e1FrL6vy9Tz?$E@3
z!A{py>!GaicK+}I*0Ydf{qnVs=J{qxs|MPpa~Ly)G8vYvnsZ1?j!56DZ;aVaxXU#h
zYh>&i(#PMLDUV9Zm?%2)BzmsYaZMxOEI?`2ZBLBK8_IIDmiJBT$@K_Gk~q8aueM?~
z)0W4+eG^eF`~}U>jScBt$)%(RKxIL>9#1;*17kZ^l{DKv2$>pT2Oz{MMAvKBz2LuM
z-hDX5BWUk89{4Nl;Nf`q{TK`Zekt3U@JB-XL3^K6Pp7bk&KPz2EdrQlLBB$z&pL33
zpHs2C8dGW>jy%-~O{MEcf}=wAD`eym&hG=Dd>Gf@be#>-ASY<0w$#f$`3LCre}>op
z48B2A5%x&$MyHx<abX3^uOd`K8qq$^5rANbTYiL%1$sTNapHta2aTpDBiXkQc72e7
z-V;Z?y=wyf5TDQszq~Xn(&F~7a-y%}kPHv8B<<I*y%MB%jIX#BMO4G9e=m_m28Bg3
zQu9)jSnl2#5JveD@Y2IY2SYzd5<a}|jF&1=Ca^rYAbQ1PyglvQYqyhl#cS^f9ClYI
zi7hUW5Hp}1?&7T!!3sNI+jy^xV=2#%z(TK8;va5iKB$izy>sRj$A{mXL)1;@(k`AE
z!`^;1i<cFICujr5P=v%9=;7G+H%`1?%L(vLcwFkcWC9n=3UzxTYo~DdxKzbG3P}a`
zuoJhLs$M<C7X2!Pa`z18f7fy0#2GPkjYf~{i=l<(Yoj+0RV1#A3<!7a-yZiVBwD}3
zLOk6S0^<iBv6aRx(*(%iNwj)0&5!-SN4<^@TLt`swVPS#SxatkacuvD*5|Hi4FxI!
zKPXh?@wVv!T@&LQ^bnpYfhC8S*}=nu<m(T8c1`+ZU(8%?(bl;atH2sHP2iSaRF`IR
zzxyz9nSod(Lh&its>O?gOpw;wf2*^Vcs^^~eOL<J9o(F*Ml&)~`Cq8vFMeBI_TtgS
zQDwYwSx9FShWUzwachwsumAiM)%i@Wo`k#0Pw}+<!R4IiKb)_vZw$Bg>eQ$QJ>gFn
z!|RU;mHhBK)wdFk9QbS=ONEX|y>MMxITE+*UD8Z4sb`*j(hruNP2pNG|Lev-YuR1o
z-UZmIT`XR^BH^jd2u&9qHOU!w7{|=sb=W4by7!7=G{fE|tKxO7;Pi)G2W+u>qXMa3
zhdKwBk1#v8F0TeAJ*?qFMAschuwIvb-FS5Re^i40Z!;#=2P>o^uHm#L1?g)~@Ba58
zLxnGojA)-dg?gZ>wB?;o>dX=<iIKWI#8R|obY-4GVP~$DDet0vC<XF?Ff`}xWlE3k
zKB0wsu}fQ&xI)6@42o@Bo{^Y(VGoor>U$MOx5&I%P+!BiYO9D(c-tsjGNKcj8S6q@
z@w6FOG7?Ur8`Fs6Re-nOJ@i1i+WqDAc#>xA+1|DR`8(4ep5I1QQ)6<BWXB3Tm*Z5X
zW#kj{Ul)E?!^Pw4Gky^ya`6HQ$#XUqBZl0OH10&>g%%1a+dMyvOM{3~q9Kv8iaW09
z2OclUL}S5B?R!FL3<*QeiU}=P(Mkt6Vh73x5mu3(zZ&4ytEnm+g4}@@TYC*$n>DZf
z65<(f22Yxv|D=#ETAtz+XsrHZHB0{xAIAk1iJz+;eIX(ErKo@LJ=c5R$WOMDZFt9?
zrL~azG*GO*@GHTOU-FrFX>QW7!Kz@zEK`bzlh~Ou_uKAoAgn7REvD&`yD#ysi}v;j
zcPlZu<j1eLe67ymEXce=;h_CR=P46T-+g%~vbygWT5`DTGPW{W^6aQ}(`lfjpe4%Q
zV76sCd@Q<pPs}QK=g_!aubo@RM6kuYG|(w!-Hkylhp~j}Zu^{;+*SmdRi0`nX1241
z;gvxte?jPc$xl>5*?!@Gu$%8^cB>#|Qmbh_eoFb^#pv8I-DUH<U2cK4#?%G}e`=Gb
zgKzRkHt{zv?wnk$$bG`3kYt)k?Q}6JB1)?JIoA+F*Bmn2sAC&hWCuKt@c*>Ph_Ul<
zUcT1-`T%~A|LPh%hyl=JlSLkG+ui%YAwKe#Vy<HuzhTGrkpA%rM9LQ1hba8|c(a(U
z|E(&u;|Y?7KY99ed&vxE4Sh?i3WqM2WkLt=bz3OH!I7rqjla(4^HQSu-}pkB{u2Dq
zP(!1BbG^G8(UJYlUmY_zfGT$>fs&2sWQ~^%piu@8I`hO4igU-E`ey|qTw!C6KTU&O
z=2Za#pCyR2*cI_fP!r<yz7z&vGK?aGc&Mr`#H_>Zcb!JmkAY@rY%2261w@JCrXdY^
zs>V=mO4#34aYUcdBI2?kn7*SK$RPWkNtUzr{JIyG1*A0vph5kX5&K9;GFBcnzOV{W
zDQj7(>+vrh!wwo%ko938vB_O^*)2&l;r~niGf#=5d*~=7ka7|E_8k<+bO;JnMeB2a
zs}%xy{F?AikG$+a-!s>d&!ZqviV^&f_bHAP?eA9s>)RlZUWQ3;Bi-D&84<$}|02A^
z5o@J^_tzsXd;1dOx@si-T>@zgJ5azYa8&h95&pXu@Sop}^DB@#I>&`8-upuZ=Kl(w
zen0rpJMiE~d*f34{`SHD$Gbj;1~jlShWd(+|Ip_C_uC`30LHy!7UX^TLHD;S@IO6(
z4hCf5GeoEapa0u;_V2$vbm>iz?Q6=Ka`4ai<v)GbfA{aV@S)4U<E5{3?Jwo&zyFrV
zPF_SIMx`CF^9=wld@5t{ejG0ZQd|1i+1zDC){&`^2NExaw}MbxSFtb82_@$#D8ar^
zLLQHzZg9q*-i|2oB`960L#-np{+J<%2icj>9b9gpJ*q>BQ`vK&Fis1$>G!6#?*~;m
zkmW@=gZxLY02mFzJit-Uei5~;_j7OE28LLKYV>_`bsYj!>C)lYvCD#*xzP1uq2jJy
zxbk`tI#>4RrtS<F{BYV4q&w34gj9n4`A~Q7+8uR>74B43&0p<J&6d>}gF53%IY9z0
zH+$7i4jV`u7?yylAL<N{{W>*4P(AiTZxjP)_X1O>1ZF`Tk{7$x^m{G+V}?rbe-Wzx
z;|mfc#0C-ID>z_UfPJck$XI<#te)cV(^~+<b^tJhs>SQT1<rzB4ZR_v*5L0Ot@OYr
zUGf({n}xvjEr&oBYwie_9|lghtj@8~ob+NFa5@b_kKY(1HJYBpNnzTj54SU1O~E=U
zmZ~vey21L_1A&q+=zK#oRVBWO^&CK>G}I&3o2~dul*_@w$miv!ip1OaIiMtNMg(gq
zo=917$S~d)1A@m?hq4O{=D-=xltPS(%-O5k_Me>(U^+)GKs!e7_+Hink)K9JhOL1T
zN5=B;kJOAM2>S{s%u72#IDAd#W9Qhny-5N@<)>46G{f$i1y-iyo2iMmh~`jK9M(1l
z6>9c>LCgPZI~WF4fC#`8?A^Ma$He@MUZOcTYi*nG+lErrr%nUrdKoG~Y*8kw<KSe-
zhmsvT&sLE+2rz7*QZfJ<E9yvKe-1h3>h-ZfqCW&6n`QXF3qVswJL<l-DloSHuz7Sq
z{RZ3bhlc1oc7P^{Gl)L8Kt&U(tc_poO@lwt2$o3l`m%j$cw~?(7GJfh<8g-4E)UQ}
zC30ke9I*q-aTL)ZK3q5j>bPiANNa5%ZRlAx%9B6c>#h;7vv4nvegH-h7(jh4kdXXF
z0^ax7Lo)7+<$414O;F3SV@%Sr?w~SMND(~^=dZm^tUdvGyP>1xR1Y|VM||Nx9Z8o|
zp!&msI<JQl?Q&q`2-ZxWN9jl!&7K;S(jAP0YTkuW4~1qYH1$+Sx@a2a&ti*~)@@r8
zsoD?4m-I)JZpX%;v4-JlSc1i$wv_Mu!%-tdC`)#H<~+M@0ss_vaCA_wyK|w=KLooo
z+Qk%bf_Vte5y5HC`>{H!u(yBg$OiU(7|u&p!@1I_pFm(YvO6xBjWEH0j1DtJ61G-h
zH>Zf)WNs@D;EZ1x=4Xoh(<y?i8W}jzdL_E)*rzWzT8G{zlOw{%I{wD(tvNt>Aepnx
zx6`j9KJIh3i1{3?3Bs1a$0B&ienRsEv1(8)ppi+jMFdYU{a$~v*WhlDHN6%5e<vgS
zYY-mWcMho{EJLQ7johefr|&dmLC!V^KKDyn_ero>ps>fiDBF+?na5Z^#s!X?Wl-;B
zL&1fWXC0sR7FYrLoOsIod5K1k;ebsXsGWnk0+%PgH$7BxqE#3L<(K}w0FSqHO^1H2
zKdb9}DV}V2N*MNl9k{t6qfDZV+GS765x}oY05;|kW0-6-gl6U$a}TH~eMC4DA8J9!
zYK|bIoflQdjM<xuGkmE<Mx#Mv)^+rKBk6%hyi7NY#4tkkD@FTFwPO%BEkU-zRm$1P
z@TGaTe7EdNGbmBDL7yd^!{}TNhjb!Gz()`ABO009b=HDmZM%Hh5B2crqcqrl20jX~
z_;eO()DfM%i2KDOUky%f^srH$L+%s*a+nYN+^HfszhbB#&H^n^fVy?cW^see@cH+h
zik+X!=B~_(H@8xuZjmQ*b7C7Jlrfg57&y0w@>GZWeeR|R^hZ3tc5K`6K!|A*rk^NV
z#TeG@AP$9Gm==<yfQ#7zoc_~;XsS|hM`+b|ts=;wlE^pZ58DkyBDDrEG2>+|yhdS`
zVs1AWIH=`CX1Or)fR)!oC`RI1)se|~GmK(b{quZtW&OCb#B~5iUNLz~<zU#4`&Exr
zLptk(q@z^xj!6}DcM=TxQ*wmk_KfN)vKoMdQaJnBpQqsIQ+x3`R1Bk>*YjZPYlv*l
zTnawcg2`49rb>DoaLz$O7dG|eOz3WyS;=%74?R;F;QuyFK36&kSB@9=je6uXODl8P
zPJOnX$HGZ{_A=L(CACp@Xy3d(f`dE{DE-+%-)}Ut=pbnB1p&N)8yitBw~xO(NDIdo
zKL#1;n<%s{=OSIN7c4f1mc`(9s{(Fq&@_XhFV+%v2~XPt+5dPe{S_^UHlt`ZiosEc
zM8tlC+X~9=d~baPJ)U7aVxW1e*jzEY2VM{}giYjWMNAaXzo%dQfv^s_st{LnO8bj!
zPr?2NM{w?GIN(LLdKiy@<v@Bag|ZaRGFl$dLeED;A8Mg1L8T|f@<y3Z3*q#bfLSXK
zG}O6Id3fwMbYZB4i-As^4?amX^dJmTOZ`_VIW$FsUK4iRY~!m2IVt_r0ifs=l$9kn
z_+hw)uPqJD{)+846AueraUL0Pb$+3=tRdNfn&lmPlE)oigDH+NKR~v3^<{%Lj8pMt
zF_a&)ZSbc>f{%iMamh=TAQ>Hk<G~DRK=k<S$dbkpSB-YEUuS4xdW~8rvlm(ThDT}j
z2-PNhv?VXT=Dvp~N3;FRZsIj;R+}8)2Y>zaWE&=N%T2jkBKDkqkVG1ww8&5T5FKDI
zbHW&iLB#l+k6t$CI+O1vTz%;(Exj;w9K;GQGu$}-2@Lf!Fl~|Z6tVq{(Lw1}&xGQ^
zUWHWIo7-Pnk~HA=b$+$WYjP;{{)!Qz5w3pU5lWl|N>o=*&ggV9>@O#Y1UYJ?`me2k
z&Ru>e$}a+0^s#Bkv``PjT74>*6U6DKl^X6H?wVQR*-dQHy_+m9lpOBhZ%%bWGiU1K
z?h0slc`PSuzOaiq)NH2RW+K%Vm=&n0mZZlCMT~S}vA8;;o513(bqteT>(7U|E_H%_
zu~#VEE35#)lY>}3dhZWxa19mY*iQ?6rN@%!d0QADK^fMq(ucV2^fH*Af8@>iY?!yq
zo{EWjw&C+aqjFCl6F9H>(=cIh&@RS#CkswA;T7f|)V;#6{S8pwje)}D(0ZvU<~cmu
zE0Z?UWgoOX3KNt9%dYjB%|+R2{ET(C(RQwNYKu7Qm<NV|{)Ym~=d~A?{8);+FK4BR
z9iSmqLOe&2care_6!j`&V|pgOpK7<n@MTZDpQETzv7@TW!J}BlJVLLfKCs)o1QSJl
zDFt{0i)ahO;W!oNs;lFRUc2K*MH!6+^LKPrjgu3Za!^@nqk;`VGRr;Q=f;rtB^707
z?ufr?6LbuB=ovp(%mF)?OlY5rOT5LJf*p0X6NMq>p!IB_mmu=I?gxYkVnMUK?@F{4
zI#Z-~ripZT2G^h0lzCzURX4JquXXnO5@%#1iVWCeTu6MTtwXlzfFkVn@H}m9jP(0t
zpkLs@5Jam=EkTUI>b4O%==6(QCOMrgClez&-oU*f3{!NPC5Y@Ks&vkw-;e#A<(;^9
z%1_DYzKTEDK0*r5()y^uzCp@KSVfKFXxAS19@X5BV^Dnn4FVru<H1wwDrLzTVCA|*
zommF=xL9e;C2c276Jo;g8GNFI+>=$PnyP;Ho)R*&2LNFh>ox$LW&l^Cw4H-^g-SSd
z%Ffa`T_3HUxAk<jf8v--{z2i%V8l~vqt8^q_TQPYSW;xC!$pW4=?e9)=(Jx7g*K=u
z#R(b>+4bM3tT)Sdh>KyIl!sJ7(?QbMUx$2@U0_MG&@Wm|y{~$|MnD0hQ?A>l$)-1|
z#U{ye_xLLVw>;5CD;*Zb=Huol9FzPKQn}<}%s6g#VyO>MzJ?|^2a@!BH`zO5G9`WR
za8x~|@w=N!eHpkU9^zfG5Jhn<xb`!+-2HIj`OLzQWN&f=mcNU{an-NSi^78BXpYv;
zadq}NChwGneqq!Xh1zbm+2iG-)~_!B0Fua((K-1>mXdxs^9{~E#HyoZpv+9?QJ2HJ
zxC3_tjY%oeZ|Jr65&sqA|JkZcdVjz3jLBV(Vjro&Ga1b=j6~7L5BiC;vy)o`E4g<b
zG$Zs>w~si(*RTJ<vjZN7lh|w__tX+X_Vg_#sWu0LJQtD7VDeQqPaod|g@+fxuXDRD
zfrVdET7}B-qY-?Fm$%C~vcPLAZpp1elGB~^@*=m3hMQ&bq!rtX7z%3&L<59&`svnX
zHnSbo{67?iRu`_s<z<^VS+06{5kv$a6YnhjY`nZrHLqB%@-`wJt2Fz!1mjI#$+l}>
zD>OoC5My_}r3TkRF~^!jGw6F+^Hnz8>m~K#)z?<IIs1$HY1be5==JUY9v2T;-m}T7
zsv6w4re9xF)!wRX5^q#2Yr98=qF-_TGP{jtD@-b?!F4swj*DIqVriPMis0!R66iGi
zmeSHJDZQ>S=95v{V>C7*|1_u)m&*Fv&GtPz=tG_(`%X`lfAf>9!r+O)YFu6|&>{vG
zHpIyk30|?E94%;SD=}m$f*#p9vI{|NCH+?|W&Jocq#1^8U5^vGyi97Z+?#TrWZQy7
zYEnDsX@=noVRNC4K^HRLJ0dn2j}x;ylNLzXav7thDukXv1vqXud-Hqw`O550;fc8L
zFW%m50U}2WvR;lqnVBO9@_{pQjjf+iU~Dv0e{o{8{2ZBdL8$e26G@*JH|or8$`;dP
z{&0NiyaD53v?i;XYjA_`Ux;{Z`#L_5R%3})gOn{1UHFKp$pLi2p>3A_1?Hi8Ayd+W
z8EOfaiiSbEo)xVVH7-`~!TnT<v7L`yTc(p~XiTnbv1oLjuYGo5vbV<d?gydojP-2j
zhd=SIYZ(}qb-3~c7QStr1}(VZ3vYsDou~z1FN>oYaxHm^3R<QSL)2tkSB;P{!-~3q
z<E{54&6l%eDy_h#9IJ(0Sp4Mg%Ttis&7Z;{c5j6S+wX)bF5N0eK$VN<eX3~<-K5sd
zpLimBw<n`tWcvkCzMvKp<mnmMYW|GgXqXKZGPBYNkuk&JK(VG=<7paUR+-e0A5_y9
zzhL>&@d4%qtB~6lVhzBLZG5$QLFT5KH%)~d&)Q#@{s6^}U^Hrku<}Wa^j5b!UeZe3
zzVejOX1aW(u*iG7O-TAR^grmuXa})BNYN#Cx+sR*=EQw$a@ue*SHkLF+S`MmHL}n)
zHl?Z@BZLfe(S^aI_*Ji(rm2LcBHo^PFYS4jn4TSMM_!JMtRwH!<J6f=WftDNb4L%x
zcM3!3tZ@F@R-UH=7o0}<vgfr9Nv@ah7G`rUQO(E67UDHrWzOFA58{5kbj|Zv;QA}Q
zcxUtlV<;M=`Em%b#yIAlk$ZSh7T>PF*qH_-Ld%Cy#qL-ZU^z0StCHY)i@L@9(V*3%
z?9+-7Q=EB{{mjvW!(hjX2au!(r{#9tUynaaR3GgG$^>h@n0k#1S)UDqmori~D2IDs
zA?WuhV#BPH>#x+U+g<RNquUAwo|i~f=&p@d(hjyqw^QY?Noc<7@GpN7UOaU<E8FJ)
z<cW`{PrSz3QG6Njhpf#C58Ffg+L=oe=FZguRrW<R9AYrcqkh=^Eo;$qz-;uOPp0F*
zw8ko0f?P|Rw_nqJDUqk%Bl6n$Fvrum;%9FqSe1DS-Vay2!gV?2JK)q4?gh38?I})g
zwLZC2Xb^*vJ7`tOfnc&{c=4Xc^9SOg`WZ>}9<A4c=!S<%*vr0!%T_g~!vqhKrf`c=
zk1`q3PZ!jej1V)vN88miu%<;TTs8t@BpP(YO7cRH@ier4_}8SA`!Hy$Ey#(w=y5D|
zsH*BK11#R9-(UIAG%T%_%lv6UEsW$SEVIGG)<(<N!||#DrtJ5!G0)Mu#k|?Lc?79E
z*RH3&huziE7eb8fdo&U>pGYB|BhODL+a<XZ$<ey?9dm4zVMvl6U@^SaYB^OGPVlG|
z_jEq^Lhk(X7)EU5%b+&(<!B#R>&qKek#OW1WPk}ZdO|-4Pkaoal@)rIc~s6#nU8^q
zk(|2+WVPDDaT$eV4Jg}k7=|a|8fUgeh<PZL%zPkmIVG4l3aY&--_hrkMr!S?hg$R(
z<z!KsON@)yV&{{RVM%FttlMO%d*INww>?~;<gLm*7jY{Rbu4XYxg;Gi@7^-QhQ+n(
zLy_b?KV<aFUZmSa^12mf4lHkaD@A>3f+O^5HtDuA<a3f~j8F9rO}-a=*z_Izg&9rz
zp!}AYrz9uPaW5zSJ%5uL>heTx&g$IpNg(3*S=0tkxT%z|<#ve8YJ6UXK4PnI?`)X=
zAOTC}OJr!)x~tC;;fS7Z7;E5r2<Rf{AZ`C$iv0WJCf5TCO+#d2f_YkE;7@$7$l%h8
zcAnYM@fLC0L0Q*>?tcIFixuKd;f^TI-85aytNhS@E(tYWn0I<(U5&)P$H6?FA_w!l
z#6V!v-z6{m%(^m+c$>VpD{=AF>j<JlTpVoA8&4UrrjIZy$V4QcR`uNuWu^c}kJBVA
zP(<t_z7fRsbG1Tf*~9<T`UUKWoi0$#ihY@k#?iUS=V)%*Kb$CAVo*pdMDr*?ga-AZ
z-SW=y9lqI7)@xW8QRK)GwRrahuTxa2u`D3M(d=eTq{@2A+Z}GA98;>t!yv05LkfPB
z*a{MPRHu8XfIgQgN2Sj<*ei6#dMl-HH0>@^q`kpl#FAr9y<|^3IYj-=ao2}dX@|rO
ze?r~nO$X73i?2ivPEK;xJO#hKfu}^4BPxdfzGNmQS2jV671IT7!IX<xa-CK=+h3%x
z8i?@rvR=ORVGquO#iB0+j6;PnHn&zw{~cxwV^z8rDZ?Q6#SNjwU>jahT=ryIh2pP+
z#yq2KTf@Zaqrv9#{S%r8;vZI><^*WBr01UQTA8+dB497ggwbn?rY?KcgB?OD^_MpR
zv)-bV^n(-a?4n!YCzqWpkVcX-)0qVYpiQJ~7#$V;hl;ve2#Kg&_3Rj^{mw^M^3_pL
zMGk=f!QO7I!J(%ZYJ}C@!${JPB+u$I+9xkt-idhEK-ZeOk<;lrAQ(9eqCgi5M=(C0
zVb{$IcuRC9OD`2>u*|ZJ>JgV^7I{D6=CN5%o1SnR!Yn<WMMmf1+_b|N`#GSo@hElT
z)+{^<6#It(#}bDZWb6G<Y{5Xjqk^I&{!F7(sd{0FB+NdngqXtt5G;5Vm_<dAwwuCh
zRd0xK+r_r&JyKPOUZQvxN+L1Y_dIz>E{?mMaU;`sN1R8)vE!eM)HvT3R62})N8YmS
zKi~3AP`?hi5E+>M&fqlP&j-4^GrbI@ciM+LnO!zOZ}pfHEDb&jAV}}@AC@F66w9Q^
zk5ZFRSmDx_OzZ-s<w>SKmO2z;Wqv7-dfvJjzQgPh@l(^i{0M0%RpL=!5m{;&?0>r2
zsB`G8ET7x7)0bjSH%CD^H&9yg80+q>F6wt`FjtB;WcEt_Irb6HALLWP__BRJV_#8+
zjo;3UuOaeS!&Y9H$xQ9@qzqQ=gAIL|=FzhR&5>(=>acVeW0%WC71o!eWb8w3O}e`h
zR9J|%JfSj4CD^orRQdxE4YK74&8o821oXz2?NKNj3#}7p<gpdpT8*4)gqrWztkaH%
zc?bhl9G4^0`w?c%8!!p`DwSD&Yu=WJqw<5{q5G4nOkUm{UEQpQf*?FPN2Ky`ug(N&
zn49f>Hpn1T<MQs>D{|%F&n}x4Qj0AcVb-10j(`fZu+fVQ0!uXY!ISi9s^N2F7T=3M
zF8(Ap_()=3(!psd>#Df<oc*QT9WpGXGL1#u_@k5?)Vo!eR;@|2XA+Cct4DF8y$Gxq
zi@y!ZutmJPQMlV~9GV*4d{bEMk$(#e+x2pcNr|v6V}J_8kZy8f$o*e1t>Mntj{ld=
za-a)nX9Ko^flc|ngMCAT1Bi_C@z{-XWv8`NQne*}s&UmDPvBj7w0Oa^Z7x|#E`JQy
zrE?Z+yryEu7M(A|MxnOpJgL1@BmRS)>KB(Kd%AVI{O_Z{hgwAV#{_l-&&O`kt=&5q
z{haMJj%v;)39IQlMKH_i8e11@i9!#bE8rwvy(YNTys?_b&SodArZ7Bfy+un#9vW3o
zbu-BkHQu!!cVNR<eFe|x+kF+}dh<Gtr6ULWc$MHnbhJQ=5QM14GM%S~#<Xe|3qN&Q
zMz;psY8L9(m$m+p<@ocVnnH}7@}kGK|MA_cmX#XWkPh}Nk!|hblBsd5Q<wFWGNj{7
z^}AYqDD2=6Qao7t!L6N4gEUhcWGbG|#h&?(?XUl;_nxL5*XTM&lr->_V-jR?-h^gE
zWhy%ld9ZC^C@W5<r(-yCiPHXS$C1Sc8-PU#RcAu`WuLkE?D9xhYfhstN)P%CGEgKf
z^YCqLn%3A5`}pcnYz|DIoLxbB2d_9i<dHi#lBXeL8<)(fm@HEl+By#fdsKHG4`jXV
z@v%udO1<_&T4=;kH?P^tarU&EB%O9h3i)?Fuj|$jy=4^LUXDH?o_U+xF#wCWlcuT>
z5fZKwzi)*6OU+&MFiAiaYY$G(tAJQEfw1Ohg7^eFPQv5l_b$LIg%lbz>B2E|4{<Tn
zl&RL5CVhN<-}8e|jm>Wr?$(rpd=5Gp^DD~8Bzv`$icHS0?L5{+y+e|(U@X#wg-1Y3
zze(=p;=wuuJ&bCl&fOm}|F|#GVX_j7@ke9;AEzs#W!>gc`LsK80r=Sk-d_~*Tx$J<
ziS0Rs@4UaF@*?`dSd4+GLQh28g?y3pXVL9<>g{*b)mgC;SQdyRi67;#jn@lP9%0p#
z9lwr^Ii~2jpQZ@j@C;%sVQ_+*nk~Ysjl^E?fcu~uxeqQ!QT^#Y(8fm!^^Gh!Edk?{
zZPn!=dhh4BYA{f%Flv<%EFENpisgheftzNc6lZC3=73M1U}>AertlL9Dy05CN$cJG
zi}=b!;@1tQL@B=VsjT|yO?v2YC=Wkd_~hr<fLA#P5Yd49tEQN}7`RUxA#|fei{jj~
z8!m!~uBy2WuG4dZ^p(hh;L<NH^5#1=(S!6Bvw6P$6e~l-5dAoH-6YE~dTs?Gf(PTA
zr*3O;XVKYD-h1P#>v8@{pyssJXHp(tL;{-t%7IGK9W~|~mDFyRsIY`|)fr8iWL*U}
z8tCH3x-LS?Q7qRr@5c1PlhPh>vOA*#!gnLrMng3mlGDufn^l-BYENKCLfi3HwAB18
z?LpIE(&SFWGYS%{i=4ulC#l#tN_8FP8*$hbf=jbRsNQi5K+}wrXhJn_sfXHz>J`WO
znpfj=W8jpqAVJp{pPxbMmXjX6WUf@SupCZv6My2EH*xy!st%Q8zsrwrAkV%t)XiJx
zLmr66KXDqH5<~bxEwAUh^Q#-2JF8&po*#<8<D0@1FcS?TE-CdNdEvfW__Mh^5AFMV
zam))Ac8z1wD4B$?E^@aq5^VBr(rb?|U6B~8?w*4~K5E>)V7|tw_9<3I96-ukRCesJ
zB%GqQ%imGPkI_QkHR3azJQ2_0V}iV00(MgGYcZY)7nuce(lp#_zT3}cia&c=YV<^)
zGnUPRp5+~G_W$A%IZ)t_MrP3K#{2RxIm@<hMz`IE)fpzM(w#Lrhj(uITHQ3@hWuH;
zB;y4spi9W8?6)aA&VGNS_0!utVEJLR@4HD&?v(>roDH(E)GB)tU#|Y~o~y-CYtTc8
zB}I7GHani>1Jfjoz^8?GV;NvY9J^gwE4S;09RrG9)~iYMey*X??j4r*D~fF#j`u2h
zLcZDRxas?FlTc=}wW>h}%mw)f-Yo=5JYt)%X;9Sm-=DQsDTn$}a4P^nW~)lFH=S&W
zv=d*j-OYibTdSa&Y0yBm<REM}Q#LWCk(&-%hA*_dj@JELxyG(?7Z!c#6kIZ2pd!N!
zdmG6`#u;v28foqk@phb>jK0>H!^%%@mLh&Oa&i{?YgbM>|JtK0xlm}^bPDQ57KM^+
zX0|Suw@H4SeTB8jky7dBGJ!s^DWkR7s{P{j?Vd%bH*=G5MZ6ur1X8QsUOB2AJ{ent
z$F}Kj*-2*SGF)ic*>1?Yt7)*3)vR{+oPA#7mqt=RH>4Fks#Y7;q3Rs~xvuaFxv)xi
zm;28w%{^zuL|=S*i{L;^oBRu&$yptoab}1F-U>Mq+Qm7sWizTj{}52wBdMJDA{dYD
zRVHYpDGewr8^!NSMO|K&2FTN=qK-C06EdtP<)4=v<)*4@V{D+36!BI+4~$dIYJ!?U
zF~6oJzwK>yly~X_TPd(8^i%fpwbz_)oF}@4#A*xY1a(XS(857>NlKt=H6^ih`9p#1
z`P%HKBgnX+NJmkha^F@X`brzP9K39dSNGA2ay36I3)eIAI(gNcw5iPa->pX9Z!OEN
zv#VUuY-W;wvWq`^Zx(0ve%%}I>?$glrTXm#k-?Q(>xt|li(a)-3>!-hoDzAyWhrT7
zV-yRdEe0*$ua2HzAC+gW^vza!pNFoywQ7~66BJ|LT(&qEd)vL`wH#+b@i=cUYs(BA
z6GrK19;fI7$1uGTXe?@9;rwQQcPVVMvBi}}bEOvarp88xn%-I8*k(^trCF*uQLVDY
zne(c79Se<9h3Zh*P8e=6lVL>=Fi8(&l_SP7ZC^aqI{X*6ZPev^kV@2tzMW8}(dpY7
zL+vKyQ($bBNITvWwpr>|mhP<2@v@$RZVuIA$kM(NrK9JlGQs*NH1B9<yWWq^XnfO&
z4&%@gOM4sMomB!(oX<kQh^SLrned%DCT!UZPKM|*c&5`9>ZP-YYD!Q*2*hyzNu~4q
z-@iU*k>Q+0-Z)FIjW3~rvB<eKH@YJB^5vyA@v1k<NrTNKIYqqT{VfvXb@XM9@4b9w
zaGrZ!8h-J<sNSDgN0HQLrtzd!sN)!ZrFPziBKR2QsMhZ%kr<9Z89Z_I$|D0JKQ;ch
zqQtkgB4ETu9r0;}(`XX3ec?@Ui#b{=LOHU@TOQmfJd$4P&shE&#bo8gSeaRiqxxM;
zxrdE+wgDt)Om(`A*s6nlp41aB`K{v`0SV4_YA|teQD6FaWzKN(9S&y@d1B-223+!^
z!;<iMq^s>dD8zAD40S*jccvzNC6SY-P>Aa}s^L0Ca`qYf!wdEfUvzYQ5?9C7(rfG}
zIJG4_ElozM`}u)Z(jqi3={zj2h|)72dA8_8hh2fLmm18-<j(!(c<S`}nWNJrH>Px<
z?toA0>qwOH)JY;T*<Tv~H|!(g&%_?ROCJOamGoQ=p8XJ=dCc_yy?l);yy%@2C!QmX
z46Mushv}>%qap+j^dcifOO@C?34h{>*kz9j1DHRec*60?t``hS=aabIQfVaS>CSq$
z9&3l}16|bym*OB?H9I*UgeRg)tCS34R_8BuswzE*8GTm6C&lu<a1|P?_*%j|ul7Sk
zcDKm2OTR0rwhohO9Sjb~plU|wFllZLK6y=0-n?}5ZZ+5KEyMdfnSAGARkIe(71JBN
z=v?Y#&vHs~?sX!OB@5ZYtcz3Y&tQH%ZSljL=R}y!`b%_1DQbP^Z?$y`e=v~wP`{I8
zjoG=Gu{>ASGPXrq<FgEHF_h*=q^V=U1sh*$fk~y_w?FSDJXcp;mQ1rIxk`sKD<y^s
z&&u)CQklzoqtB*arpRo)eK||1$O`z&v7AWnwocmiPdfHyROY_-nkS)Dsm*vYt{>;e
z>PYBZ+2N+q52mu5Z!ZH=4Q-oNeBc}j$J*Ps;X5_jMjEZNZXB;{djtCuWQ8OwMIy5>
z4-MpJgfPDz7jR`&`YC@d3wl?cy+oX3_eIm)amUGVf4b<-$Xb_2;px@K+nA-#t2*wx
z5cv|((m5@V+M&d(xGlyjw(@O6)=ve#6C)wZ<UGUs-Y7$%qiP>OYhOo7BZ@qUNojFb
z#jx(W&Z=)obwT~^{$#1KUPE_A1N8|M6^t*gD0=F(-Z`poe{jXLs_xe4Ns@0JPbg%3
z0a-&uy~R9;<Hq7dj%Rl!k!p)sB4YWP1ixpHR&X98D!C!&`I7!7Q=HJomRFpL{$j6Z
zKl7p5P9bzqnSXSqEhqg%O%pTe^fA|;obnTX%iC|9oO+0(7w6+?03NH)*`-^i2X+xD
z1F2h24+ieW+^v^AzzD_kW18exXDH{U8wI^yMq1<RMDOUGt<o5048D~-?@{4L&EcOZ
z<KxvMZ({7oG=%4P$C-o?y_fO8{=i00!`G>Ohk3TNX<y{RZj!yySCe;Ca~w@MpYQJT
z>Ku0S&+fOt-*!J&?QS{d$sSNrZD(()<{iG{f7@}<uOf|v;)wQCmB<14j@NqACB3LT
zXruRW3zs<fDxVjhp~H3#M+B8c+lp|u{jH>MJ|1VSP&`bC$CesOLk~ZRqC~{jCn-%q
z9J?H3`{UY3j3^V<m0QnsQS2m_J4Cwt<43-oVxq-a`}{igjaBGd)@wicCek-Ou;_=G
zXv6N=#a4L-K5nOdWEC@ii~GqvByYd=vLVo#vC0wWWosLyZ~2p5L@-E&C=R3Z6=eRH
zZ1K~G8rI{J()R;O6$47VuZSk?PD6zfgCKuamh`1r5Rb@Fl%$^;b>CpP!Zmsw{BLBe
zf=@}ho<qZZ>2}^DIiwv**TaoZi-(>9yjZqxQij95bWkBp=&^0D-78#y04Ha(-uf9Y
z4f4THDLy8FkM&ey9JD)`(9Wtx?Uq8UmN(DZqy#&U7m+jb7JU8j8`AqX_l5Jv|CSNU
ziDKPVu7u((;u$pna<4b>n?#D!{Dnw~$zA%g*$#OA@<4Le0MJ2BRp`~CbB7QTZ^uxO
zj4ZwA^;%>EaoXwM>X@})J>CWB0`yA;xhiFYZqO}%x0-Lr(y710H^L<%W^JBXd1~Q_
z^9~la<gORhBfSf9@Tt#YgVy-$Cs&@JJ@7f{AOuEFEPPUDP<~yudG=dM{R5FYl_gmH
z893@k_cG}lXuR?j%&JnevtB=%E5rH+3!oV3wnIETkj;?_OtNoL;A-80XeoFUgqqLl
zIjKf%#++@3Ksa{4&+1?A+Kxl~OddfPa<mMD1ZKq$YM-_G%a)*SD?sJ|Ah6CZYf7Y;
zD=LM^T|NE3{^96@Wu}bnG11~MgRc33$!Wb<_3gyTQ@LpxklKcW@=2KIMQWKY7%;QB
zat-np`Tv#w+93cRM$h*tD>BwOAE>)eVZ?Tddhab{KYh6mQnVChtN%Z>Fo0fp5wx7`
z;TNmFSv-IDpB@Fn_EyZAPxwPq<v*3at;}Kj*{tr!E&Or&{rew?WrA;zs@?hjr`Lww
z-YD`4N0;Nf{@A1b_uoNO5x#*^ZCv2rJ*IzMUH|LX9c4(sS2iS-_!l|*?@mcKG5r6j
zdhTcm4o6MXaJc-mlMbl42x6=&Me}mMPg=KV3GgN`<b^Z=fxAN;`9suI@67zU*?9vS
zlw!VRU`|;71(x#aeBj%%F-v!{Q~iq~dN;-4LThs8PkPztD{6?)ttMWYCCeinHi)l&
zchc$@crlitBAt&|)y>WhWJL()0*+_EI5j=Q#dJ5&(go0`wT?NwEB}y7FC~Xi&8168
z6FiZc9oVqtjoe2Z9RSR+bp*NHGLTRUj6R3gfnlCM5(b2eWB3g_V)>1~r~)-QQiZEC
zI`Y}g{BLu;O4pF#VQ#&|a58eM1*iULU5XO@x2)L&?0@fC;Yi|vpq@%Caw9Irq6(h}
z)6W5+rU#{UkK7E|(7p|55(goR)<iVM0=C;D(uf3Z87cxhP_7+9gwRqV2OlbnFL?dh
zg95Zw7&lttm&1K1U?ndhGD}1#w}MX6;dUPX7Zr7MwDE3WJ`%|;A?oSj-Bs7)Ay8ou
zM)Fx|qiP483oE?s7v(hNBAS?3U2Sv1b5|LYK59eES?COPHfT=gr~osaaGiXVo<xjp
zN8yJPc9hZ0<fz%qHuEWmv3{TB3|{k*-zSUnYvfK&OjP_6?VQ-H7$2Y!q_orgP+1nc
zv;{8po+FIyw{_^p<O7NKMvW^XLR7D<b?=2lmP#zbK`8ZmuqAsje!Df1;M&VF&^zWs
z(UzRb7a8?lW#VaQgqTZQj~1ULmXh!|1Hf?yTGd0J!?fRZ?L!Lq9S-H)uH6-vwWC#6
z5s9Z;1rGB3O5KMfR)m!NJr9BDyrobQV*$_m@P1r1n2Y<t6>MPl{3D{6cFu1=o|3Lg
zL?bgirR;E@Mqn$|uE4HYvnlQnBY2JFznRN!`VG#}CV7LKgc=n$qx5bvrp#tyUru|i
z!liSls(eN7H<w>10G^Ui_JD%nA0iiGw>Fx!YU*Dd$Uv7x8SDz+kMesMd)!0p1B+FB
zMqO8-k+Z)9dh>i31q1f5gAExF_GOp?zLAv#tLlQRlKHnI8~k;oxP}sO0<mmAlzG(o
z{R+LlMF)T{;2rTWLTNKAZt$eBFNa`p#v94Rk(y>75Kfw&Pp3=ixkPQ32@JGAF5F0b
zL>neFAM3sZO7ZvGi+vd$*e9MUai|u@x<6#-Fpf;VmDU}Z`-$wL)**$%r{9WpH;ub+
zmuU2VN%>QQ+*6bjr#LVyG3+vTXftd~b7ZIt&P(e_lf1CR>`Nn(xk@TuzwSjp;p<aZ
zo=ul}k~tSO(x{K1tn;BJI)q&Mh-Ul=ZKvZR&0nA9{hbchzYb8L2;{LB?<?5-?PGTv
zy*myYCBv73w{1<?eYXgK)PhC&>>=nZDL$RXSG~(u0u-k#neA^sH>5$VH@5`ajkA=x
z3<4ccaZaNgJ@F8e(-@N8^0ZsNa)S8@Jajelu(r*>=kfUnWapmQ{$%(>_f69CLRf);
zv0m2TYS)LDvQ-f6_(b9gba39SB9m?r<T44>7C`2>O8dc~RH3AzcR|&5#JahZKT+>h
z=id61`xFE!7q_bgF0<h4lt6T%;LPS6-F}~w0JE0A0|?9@GrD$|5!sGZySs{(Dq&KS
zK143A9+*TS_~TLyiEykrkm*)A1qIV#yWikv_3XRfKtIxAps0x*xI!6au&{PuD-I&P
zLXR^*DGp{f&r#72zjh#*T4uyO^jUJ(^nhARXAI7kLm2Uaf!{9+W(ifyC*5q}aL;?%
z@^}C%>e^E}4AT*HyV2FsWGy27DC^!PbnRZ^Sb`Be$b1+u#&dKzCYJ%7j`;zTbU;9t
z&pvaZ&H7?9CMuPZSDt8zlxGtTKcDqBH8J6_!_nO{k^GQ&8D}sFg!1~KG5RWQmw&!0
zt1|mQ<@(bSIbQ+pYINtgnf`!fr=ZMb2SD+Xdz!P8uCUCtrnHJc^IJC@xJ@6{O0j~_
zF&WdiR{OzFGpVjtZPW4aiagtFKIjyMv-jpij*SptGo!%lFn~?+wuFRediXy1cGnbY
zvDN(2mCxY`6x`^cQ$6**q)Gpx&|B^y1}6LI{kDkrx1xk=B3{{Oe<Y$3<U7EcqAn2x
zq{Me8Lg&joEr>4s_Lj)f$xTNgllaob2*G@j5+#ZLL&TT{1@o_$c+*}W2DLCh0IpV_
ztuB0V49yOb_UvALryQPY@f#1!3n$)m@#+6c$))2yVQb-LU5zx}ACC^V9k}I5@X%^}
z_U-nN>V`HE%hA&VOr4p5={53B1I<wS!Ak-mQLBKbS=`DNhFqo?;jWix#ecc`_hE-~
zC~_Jhic2NTl0h{o`m#=SkbgSn3i|pUEbsnP=e2qeZVSaL<a(vAa&v)_2nk!zDkg*q
zE<DB8G@gVc_E`g-6>hEFu<iv%-!}Mel>Rzcq%}hj=$;n`&(qX<1wQ0&G=mL5+?NjG
zogMHLN||BIay~di8)^V)G(rnNy~q1G`klmh7DzpU%EAdv#e}mP3FRe7P**h@=-I1~
zE@3M&`#5`!>6^kk6=LNT0MyXFo7Rw^_!$Ko7*Vh*3Y#IfG?IDkbchq-ezOvf_EjAG
zKsfea5^tXNi}sUnPvyetu$st1!bwaY+!3~%2#T6!`iCD4gF<&iC7Mpz>xuCpa-?7h
zmf$|`z33zv@1-G7S;Q=Xew0h_erE=;3tCjusl%M)9-7Kf4>f*28g@mz$njQ^2%&%|
zIdO)zEOCX>X9Rdn%b`_TiPyOdty4K0im81NyT8w>z^&t-nuM~=7sfBF&9P0P`W`O^
zF#O(nJ5k@?Q>5d?#llu)$Dzn2b$$Vsn=>c)rpV;lK^?3h&mlt&t^JJFw7R$tFHrMm
z0qgRGAWrDu(&p@go2?2VGEzJQmKU5)WRYwlf)k4N>T?3Zx(&oba22MuCAI{L>^I8#
z6om@6QaeZ3lM95u(+KY)`5w@h0s8&097NRfg0KCNKF)2|y=rRhUrkEmun<Ejf$#E$
zbxvs55ylpTv|h8v(62Vh`(ac{y7VvJdONc~szPP|o+PKc%~J%XO_m!FR{5P%j$z5`
zdAzQ3viE8lI}@oGXRlXU{9YF7=#+rlNUc&k-1_U{B)!*+VlrCdJ*{QHK&;Mu!!<MC
zeydYT+_O!v<<e`?&2yXbGySrQ;^dn90BNd;2(V@oVR6Gnmq2r+-cM9nJfI%N+Fe@A
z+u9e5Nt4`#UKKG_0MY(xX7$;+LH<bPdhPBm`}3-nEu8hb=0<~NN%`jvZy#6k=pCL}
zG?;a?)AS3Y&U||xcFRa8TreXArV2Oy{(=#{^U~Le=Xl_w<d?T!2uV<p0hnJ|z+vQ%
zvAa|q|4<K93RZUWV&iXUd|S07PN9hUmvr6D`!ls@AE!87%&O_n(pg5t)T?f&tHcj8
zCp<ge`znq$zG2kZ>KCmOkdwpDxNB#5o6qBmF&KLDz_W6Lhk9U5_vQg;%<~~^xmKz|
zizNsVq1YF@?}|@?FSxfps3BRazDY4NZO@}VP<p4LmY#E4uewAj<$O+NvJV48lUG3^
zqEP2)M`nE9XlI;$YkPsm-yTmxo>+83g@hA-G?L1Gnk2q#jpqe8!RqUr)1OZZ`5Dyr
zl%Va8;2<bKIM1U)8ly0K&U#CzkPa>JK(6&94L0_mcllsS5bTp5Lv~BI$EBW$J$TLc
zmY*)y9*7xBAaE#vZcjfO^p@r+z)tWm)gowSNF|PYi5RVZa2739haPQAu@9(8J%_$0
zQQLpB<}=tQ6;+DX%Y(Lq>TY_zCz1SqE+$Ct(TsF_RZ$ZQ?35Wt=5M@@i6n%`)GXX6
zU4O3Ajv`cFu`iKrly+jZcP6)2fDiTM^4S7U!7t)ekt&U`;j5l;V1Vfj+a<=(R@_m0
z!q{Qfu3t7!wMRUv8SlCkPyPA}Ta@%d4)`~`O!~J;Wd*Z*m4hu`v5NW!?-(^nP<&vR
zdTCj$_EP5dOAiC63VDWu$9b&wO0D0CdAxoloTUK|8T}<Sz|?V)(S6l3YjU{WtaH=;
zNwh<P-m-C|O3#&$0@(z?4VP@2|Hs^0M`g9HYvX{3FCy{MNP~h3(xJ43sDOZhbeGcI
zCEY3w0xGt2w=~kyos!bs^4)Xo_|CV_*~_!n@1NfoYpj9s;$+VG+|T`7_Z8K@a>b@)
z&->!wpT%TY3`gSp4hG<apg)=eE^7HMqYxfM^5$Bf^_G<MagcvbiAaa*X>M99i9XJ)
zPV*~Iw4xe!O!`BPoUv@qPl?n;2HC0U<cJx0_^Up<{*h-r4WrICXkw2W)iFg&O^|Mv
z6F2Pv^RotZ!m<XDM{U$5Mq}IVbE3|9#uY6?VGjIsIBKlMBtOn1*}&3k6neo=X%@ye
zlQaA&V_KH(>wcTnkl&pT84H9*^jJxOg3{97e=L`HZ|c(*|8%J!2QEvAD#2rt@g+sg
z^k;hknn}bZ6ff!o&yf&z#_ZiROj>+t&3JF>O;VkxV}8j>l1Ff$h`eS`|E&6spIg10
zyBV{MqEm1veqJ2$xT%7s61v}mH+xAPt>a9;x}#KYw7L^MbZu2ThM17;eAb3SQ=Pw4
zFrM(|%Atzk3mki9c5O0L!EU$1j@+Y)+qRc*Rrgq|U)om7OpO^56XwP38QV=wZW2?-
z7+Y})@8Ayu4Z8^rPcMq+vzN~6xZCjquTzeAJV;T~@I^62tiW#mPwW|6Mvc+I+82&C
z+%lsb`BM$$7KtCM%04pL5PEgeYqpXYFlxvSZ31ahend)t>P>K+=$s*E(H7<Kjl<%P
zuimZR=G|TU5gcmQowdI3W?&lF8FJaYk8L~Ur)u71Q!SM2h=qQXP4g(r6_JU4&C_9A
z2`@TD!bOe?cQyCooK3%DQ^zT^;(6`2=+vl`nqIP5x8Jd)p?H3VY9=Sa;%v;OU#N+`
zW^q;K%1hn3G38pvGTkh`aYJ4P#8`^crdnz0X)i)d<+v2iQ;ZUrwWusJsp#@(?dO<c
z%VXsz4K~qy`Jl8zrj@%=a|=R1ecaHY*XvF%8_jidx>y(|g0N@)t^jWbjf#EhtptJ1
z1^t3}9Zk|c<KyasFh^S(vjLyxqf}@`RZm<#_SZ?Z6sp-NDN>x&yZRM;Jh6Vo6Z-by
zV&cu^%+N%GFn4I;6psSAgHNudrKfL+Qfy!3_{=7p>bXvT;{wM|g(F#RTm!bGZ=w85
zW}>qXDHe=4DO{JZ>gy^9Zqhz6S`sI^J<?(@r7b-?a?fwnvD2m}N=%upM1spkv9f?;
zu^f;%K}Oxtw=`D*eskX6dW?l~BHCsow(hvZuf}lAu7kY^gp%r;DO*@cC8&u#xq2f+
zNfC6e+wHRl(}YA_9nO;izwfv{QEu#?9IyA6Jiz$&-dw<?_{YF2!*1cE5jHIp6KLsZ
zYbh{O-PGiX2y!P|8aR?>mft;Seab*`gF@4DJyhp{xmZ!$#B$2%DEgy`Gr?|YpIq2>
z8SCt__;Hi6lVi@KDU{!fOa-g-eo7E%>n<9#Q=dw?OtSmxLEaq{m*-Ais?7M$HY|xc
zk`<TF7x)_boWK1r1b1KD8!~56)t7#xk5!I-L|yPu+{x>#O!kQV!PKmjV8T024RbgD
zvpHv{lGTGtYDJs<2e7NvjIJ;dZkYEOtHgc?T5UsN<TxN<Xj60;a+}5lxy5I4h$u3$
zYtCBusSY0ApM2@oDnG)kR_5})gV9#DE+#oOTR4of52f}xE6_0_Cr)g5%!F5fw=FBd
z)t(gy8U$>mif`qU_6^%=4;JAlMZoxYCX$F~xAal-p3e;<Fg{zhBH2Bw{%|0!qQ62W
z&_h<~r9y^gKi?Cl`<SoX^6$bOEsTF>uq#AeQ_>ZDLR)OoO0>uHKK<fLqq_zceNW&o
zX`NdybR(C)^vI}&oE-uN^&WjW&i;TZyNaoM(ZPelCq+(z(R=g>+ixH}iAN;X9f-+Z
zs0<~SWPf2CYVOvadQH?r7RerBx@DN>MpzfQv!c(w;HpG?jnU-Km5K|9U$?U;BpRN;
zdHl&Jv7Xf^iM(IX7nl1a6sU?qut~gmO?nqgwb@6>GCK5`(O`CMbhSV3&1)0Q*!Ly3
z?@>rn9n3mJCvjP2Geom7lH)3<v#sCV-7;CZ&Rvs=CfEQ!d^16FE*DX7@kBGl93l*J
z^}$9L3iX?+cXvrC8N-JwN?wJA`%%oq_}N~3z&%Nd%Q^J$vQ^UT!mywFPTL-e>YMZD
z7}Vuk&YBb$>Q;D|FZ&Xdru{Sx%S*3SgIL#u!!N{o)S7A99+2cE=+%pWDkE59$-5$S
zC(LL!{XJ0?!RImVJI)RM<m#6s?-OU{R!E!5OTHngOjKHZ3=}fHEMp7f1NY~te%~%?
zvM@Tv0a>FomyZ8QbjgRod*ha?ufgZB&RqhfDEV1Mj-w}p=&fG|5*0A)w~?_ECP!!v
z`i~-ry=fTvVmwl#SyRkR$~E~E7qdlVh3>olOKu7kIiv`g{7VsPI%qpRumg<iL1@xx
z78z^qLzW3w-$f1-|1Pl>zn_quV4=(Ox978CBSRjdi!=umuiuYKX^jjwJ1C?Tj(f2a
zn;F;4TNVx)KU|bl9Y$id6b97$8duu-5kt3J>(ZD6UBM9pN52I`)2iiClD)bz$yvjr
z1D2ua72zaq$3&&pf!yY`Dw)A$n1g(WLw<|^qj%m6L%E2+P59!0Jv-KDB+_AKDLi2M
z({0m7>{gXjNTS>KkhVcM$!A^03Yr+%ltW>Rc1ODG@gzZUmD&o>{az8}@7z=oW`7C8
ze)Z<;*v^yK+6t0-V+QIdnfGt7y@Rffgz5%eBQj=qf5UrPTdP&-a-b5-GA%#LVa&hg
zm71PTA(XOj3P^V|GcOtv`H_+x*5*7ms;y5#@q7xr2=FK#^toD9Oud2DdMcPNvOHBL
z|MJvZ<3j<gTSMIy9||<c8qWV9Kr363VYr&=AW&ATl<F|pL+S@H6NQ;Wvo$*@w+v@B
zyy}mEiD5ajbOA#~vD)SjuwRJ99h>J{nWWAX2k|_Z9c<loUG$M?tO!5P^dh3}NA#)J
zEKeC{U-FSXmrbNFNT*6l5Yw49;e|T)Fla^h=|E-$pK)tE5xvX643Vx%nIKs^_iE{f
z{wGBbME{6wV?el6;rog7C<c_FIpegXsvAfy&oc?{+EXqDHiv~|Cp=DR>q!GtOvFyL
z(}T%qXo*#PV<HfD3xC&`#!R@+=?DRQXZ0#AfR0OucHE&2x-&U3Hk<q53!Pj1Pa=J0
z1_YS+kpgqk^95y}FrGP#AT0B~*RiH;wTDB9QCy9~?3Q{kKfmLE6h<S>qs7YdIUToU
zehuglrCNjA?zrU58WI>)xIhnIuP%0&`DUpQ+gt6k5V;TC-B)u$oDZ7dK>t5aH~#kf
zuXc&7F>m;t-)w7r&jteQ<103LvVN$Ywwh^h)Cf^3@Y}ffHpz;T6_FtFK%h+FVFaI_
zftz57Y<%5m@#WF3o&rAXY`1Gy+{OgWeIwm4*o767M~4*#uT!8|{50$ppEe<^-^mqC
zJ~U%^$!rQf$+bgrMZY?Bx8f;0_L021ewOxG*MmcGnbWOgp|o!5PJaaemgSMQi;<Ss
z%Js~t<&49jTh7XNp0m2^qYEy5R@j6|bu=EEP&xr)DiwXjb!;w1ia8*aeq`3eje|5i
zN;|)K7W>Nv{?=&&%oOh&zOh%l?&Ny5U^m!t^U!feEh6-ilu=5XdsY$MmSB9DU}>rL
z52jaj{)tAr<nX%a<q7b5RDSC)mc$_0u?wIwzNL9-Oix*ph|F<J)W5g9-(SH-v0yEQ
zpMC=mP7gN*p2M|Xyger_`J{}ss;=VKMFh@y5isC^;Q59Rqub<pE#K1tPC1)U8?r)j
zxmzT4;1LA|3&+?UO}^OdS-WZt$y-l$Hxi1k6)5a*@Puy5SC|p6v;-tu02xmi#QYeR
zR&7gk7$m#)0|BSZ^!mJ(MO&oOsY^Oeds}Q=I;aEpUsA`H+$k&+qbwXa@)O!u$&8A~
zu!_%XC6oB1+^%vp&UyIcPb&tDtF*93qGZi?=v}6tD?AKq(<L-Zat^Alt7z(N{Q+m;
z#ceDJX;AMnC4RHiHhImk0&WEJubHE=zWx&02NMNJrFq66DGA??xhl3D+`)mt3Pczi
z=1y|kA)#Hib7;l40EDr_!-FD}rW2O)L_f)VROEu*R>l}R#@ouEVDv9WO&RZifv4+*
z5YNp*m&oH+7{*PgkyzioVw1-7>`4OuP!Zear8nndTm&{N#$0uEM9$ewF?N&eV)U(i
zp=@07=diKzG1^6zJmY^^@>K}9)UXPN*D*_F-3CfohOw05EUHNwPGe8KqvFO+O0(A0
zR<=J#-;)I-ORX?#CX?KKVkfUG!0Qjj3#CzQt-OI`qpbJdhkJH)Am`A;A1@sk^d%0?
z?!4lmwf)A}1mH=u{L5=O8~3!vHA*rc_pHpKF-rUEh(7QJdI0eCaB*eqjlprB8deo@
zR<nsrXoBoY^ppYN*O$dkHHN^R{Wv%VrT+njGt`SmM%A}>H#8o~t8HRx!Q5`17#U_l
zjQWE=UCt8N&A85#0|R@tb}hv=KbkpJ+-vg)$9C5Y<5;l$g0P4td+tn^HDCF%<%-Gh
zrVLQc%Gvjy0!<Y>4~NJ5u3~6<?$5~G5{LevP6~rH2m|2ImbbBUtOCXe>wDU5{80NT
z>Rm-$g`(*8gSPPbofL~5b*S%rUgb(T=!i$3GjGWhJf_2BsG*il2DyoSjdMmUWguOR
zu<EuW<^)2!h8xA#BzCwIB@ZSsxMRY%5!$u&5KJjvKvX7^3)m@7xr?6gUBUVkYXXND
zp2&YToM-4h6YV(94<%+v(&WSB#~%X;{j{*tLAyMAJVW`V-?;Fy5+-XR8ly{RJ-Y}@
zSy;p7)SGHo13Vhri{ijg{W(381EF~05@3P29@SD7jb=og%3_3i=T~J7AHka8?JiIH
zBn!hGm)Yz$htkdYor)sqxi{+;bu#KkK$V4ek?w49+f^7pNu;05^B%o=T_G(l#h38+
zE)(AEPbA_2)!nRPPuAZjC=A@`6QWZWo!w|(!}xZ1pVK?Y^JFbh+;8`5gx@!r0gJwz
z^wl*}^{bPw-9hjkg^_)p2G*-3V)SL%$8aU$Pfr{Y;}<2q2}l}Gt^0mARI|EkUOwew
z-t$qJQAA2L0E1D%Ty-iKMd<D4U+ab&+%^IC>#83kV907H2j@nxYb9Y6>Ym%LbEBZs
zCMvuNt*VXUyYInn&g6t{K;1Blk`?%RGZHQG_Vu&Dpw>Qde4n64Z_N`>m|+8vwjXdY
ziJx9bX|i0J5)H(+FJ}iYp_-h@+bEe7e=C9NDAyV$HjHUH?5l{0l6gDQ0X;V<5CK4b
z9WwI1bHJ(C4}aVIjc4>3Zq|M9exe<0@u!9SyFg`%pHaWiIOl!T4@8gCXSmUHBD{v*
zo)Mik9QHOLN+ZrBu+e0@HzsiPja7FXh@|KPoZbP@!rkyt{m}uT#dXlE{c(EHs|J}7
zjeB=a{Fp`^s|S|_uNayj+vfy2yqNZGR*m(Q{G*`T&a8SlXmS$rct5AhHRQIQpz?gT
zxnQ@zYLpg6+v~eHX!BC6!(^;dOPVTo=JC!oD^a4KL<do8l?V!zAo|uyUl3hR#6er8
zb+z1LOaz*Owv86CKrMyC2SQf58r^maCPO-T9kyW>!obQ*Bjin4kBBL7YGroztG!P_
z$5o=8uy!T3^25-8d{%F6(PK|P8WS4s_FP6Sv@O%wC76HMLhbla6z4HpkYE^F^p8zs
zp2R4;dpZtbfgWI`(9?q%u>f5J;Bsx)uUYhUC1oD5G&Cs7(vod<!>Ht?2<S^J`oz+C
zagWP@UBm(tPB$a++vvr_a>qDN2bhYU#_Ox=tqU9Bb>CZ~WbW-VLB<^Et0<#L)gAra
z^lLD-o`zeeKp-^#O7+xclX@;N_diic%-%Q0TXgZv?$T~HD{7*)@@*YVx4Oo{ZH3_Y
z>t2Ci1tNGYcO|QVhwi&Lf{;Fy5Wh1Sit~Hahl6E8A43>AjwLV%;-M%Qyu+4Z2yj;}
z&|BiImJ;2Vc_^q7+!%SsKjt9-z2<l2E89V6=2m~aKSqIwx5~=9cEoMKxlhJtcCxOu
z-rEsug5s?RlB|f6LOWV3UP;rfh&F>iqQEAX?hZJDC0X&vYFFZFx;8>oc$+_Jr&0V@
zr@L>LCbn{{>_P(z!AM&CD2V!IwSCQj#hi`U61krew4JKx8I&qaz@uubV5)bac$wfP
z%15D5pBh-f`?C~uQi)bPNuj&uunc%9+@wdf7>qcugG4k@rgWJ-n%PF;d)0)kGS2Vr
z)z%BD`1<2red@OGhQ7F;@`uY<wXo>4c9KZ(2{bs|yB7?g{qUYmNow`CukM#&SSQ#u
zGyq4dd<4Rrea)Mf)#R~6{cAR9KzYvG_#edf&c@ui1-q5bt8Sm|<u!_n8Gg#0ys;{^
zZgQ6z(A+}!^^((b)bsC5LWI7-6mQ#tLOdSNwYC0CA-G#KTtkiv24N+EZF}}UF)knL
zDL<_6vkr~c$I85}Dv_ODrQUug0(cFT-HlsYZc8$W^GRXhDi<iUq`hXNa^%FBORO0h
zvO+WU<%{EselF%Phb@|aen?0mGzg8o8HYQ^sJ8TYyGFo0JzE-5WkT)Mv6a?}s38C|
zYdFfXFPe+k1-gav6ZV>en5dux<I-%v6PflXS!IDd^^)_~8;_c3Wc4cz?%GY}>NVy)
z28XviFzOeFA9YFJ!tQyI%ihoIsMh2a{8r5{1CFlX^;NyBG({e(;@#1%Jesw1H-~V$
z$EjnmH?^MTGukaMu_}2k<^^DX-8DP3Nj~eG){)i9V?J|C$H0w;@5=?EjzT`lI%OwZ
zPJ+gNOze3qW5)+KU%cUFp=z%+^m6-9t<00zr$v;9znDYjztGJNYu@Ucb>R4UR=8RS
zk9zXWWcHaYu;i_$FML{n@6lIpWwG(CEVIeLNiC@~eo$3h`3#IXCXX5!atPA0IlA~M
zt~Hc~|0)9Jxr$b*vLXfUFi;sDeon)i)NfmU`coBeKT>81WKMj9J{qMTV$S4Ko5?MO
z;)$hI2)pXgl_`v+U)FqJ-rBpO)+6%PeXYJvPiJKDEw|z*REyb^TVHaQK8BToi2eI@
zgsw!>?)p0$zaAM3a7LjL*wXggH?mVy$U%pC<F8XvCnM3eM~o9KEPiw=371k#?4lL)
z>Ykzsh!z<iNJLl7vOR8~GC8;YGH^X?HOL0>;=6YD<c54a84=@q36a?M%_3c!ZSJLK
z$PG4y6wHO_JohHodqrCZfhtGcxDtNB@2>l64mysj?f36C%RL>3+Cs7Iuqs-O^*T1N
zJ{139odzoQo;WTFLvYYA$~o3+BsA6~zI87`iGz&ky<oO-vvBSw9>UCZ%jT67qgJW6
z0lEaQS(D08oP*^$_MI}}ZN_yWF_JKgX`|`38YsSrQdgAxo)E&uwravlX+uACUSM)T
zV~h6Jh+Irb|1GJtLh!~=5mW!RD`KH;IDRwx_P8s|D<qF@C3P0fv-XTv+zc=cT)(AX
z?)t4>lJXg@N1y1UA>jN9wML04d^LY4`O!P$`cx3JIn;R0idi6W5P+yy8NbtVL2}kR
z2T56tZqS{*mn7wPAN5l7!jlDcw1==2)gAvP8zpH`XVDrt?5P3igG&aa=*U1l0w2P{
z=Q~w6eRo_oE;SqXUD2qRL!Tan0!TX+fsq|MkPx`g4|CoxjHQx%{(F%*fujauMs&7>
z9s$=1co?G_#O$Mma`EYY_KJ|2wxT~(#fqS>aSL`;B0Thgi6^Vg2E0vKA6@MW+T;hP
zlWT!zoPW+mETM7#IZB*GB7X;Dq121Gb!w=u`vzC5Rb?q3nTbB{JH>d}CAYoq?)Rdc
zoOJ#;;0t_XK4gx{UvlZ^cUj7G`x?F<AoL-4n{-o3U8Uj5l$z(uXYZdZ@@7^)<v%|q
zYtB-0r>-*HM3)=p<^J*6u<xXkxa{7fwjebdw)aTsOB*twD@&dYR2r-ZZYZS?4%GS~
zE1j;m|Mk?dpG)$CP)MZsW2_wk$ca{?w$*Pu7JU-2MMTt`NeoU!;cJ(_U?8tGz|&7t
z4-QupIXy|+TD$At&n&<vN@@~>Jup!hg&(YNU%<87Z|_;#ck>#vdn%?us;Bkz6R&LE
z>JZ!v{^+AHz{mJ3ZRzb&fqB9`Li-Wl@7!f-XQj%!YzXylNk1_+wWY_!e0T+8T7<gd
zv3?X)WatHf`az<0mVqg)D@j*2ziHPB6OK&Cx#%_<^Ls^qNDR7g=$>M5K(w09YBZ=)
zmg*=WeTV)j=JwjNuE}1aRr46S%+E8==QQ&y)a}Pv#1F`iW4L9hd1B|YP?w1k-(sB>
z5v%$5Et<A*c}6a^G0F4`v4OXz_eb{O!0z>KZ7)gbrzWlT=L#R)+{k3nY#A<s!?N2<
z=C1mftKqoRxHsEQD6p~L(N8pe9$O!jr+b`qKNd(Ag?GbgcvmRBXfMuoKHz$=sO`re
zl&*W5x;io^Y=A#G>9c64ho7Eo@U9bIDzn&Id@8v?Y3lo8evQlTj4fDP$R84J{Ctvh
zo9dR!T(<FH-N{$cldoG+2Ut(p4=|3yt_N)s9~(l5aY#qRn6zkoZC7Qu>a&uN@`d5$
z?H0aOV|bO8Iks;eC9=$pG}}m2WZ!=L#L6RfOsXVzlk%8;IM}IstMSr8MaIFjEIFO$
z^>S)sujqRnhlx4EgYNsCk_S>`o+GDt`}k>0`iHi&TW7x*z~reN@mHA3=}Y=E2<3kQ
zSW3BlfH}ie=j?n`pJ(-jgh@@(jk&Te>azHe@*hTzHxV;j)YVb8-$-%(m4fm83u@vy
zIcKP!Y{Mq~SQ_r-i2gE_51@Fi0kcO!YM@a94a_*hGkF8Jiu3+RMfV3O;cw9Dzd7Lj
zm;XOtl?YmX&MbR=((5(~1NYq_XwhUY%|HTlO~5&-d&iSBCaj7n=8%bf3Gco;G*tu8
zgQJ<Mz>j_qQIkwM*_GdngshM)M8UC>gY(GYH$c?C+{IU%B+h`hNt(*;jq-u{th|8p
zn(nZ!FTQy=e_!^m<HHYeXAh;Vok6%Z1Q7G1W|DPAulw@c{%vrybc7g)L9l_;Da|~E
zfFAQfu=HusT@JrNK>n2<>!U4&Q9TOms%2o3vh_X&C@33oyG+k(R~PCJ@kQ)Qp&5L{
zzX&70E|3-820ENh@1oEj0;^tl#H!&T?f>b$|2r30LXZ_+K0zx6GR5B<`TyO^`5AE}
zK2MhXd-Cu9sGjh#yo;Pe$0V+#{OLLLzq~g8LSue_EFb<W{}1Q<Unt6d`peOm2-scj
z^FZW3d~pBxc|Ia1=_e2D@BT0D;42oN^&gEIFD?{9%v*oCEX4si=5}yp@7U)I)RrMu
z?jzBX?FWv3T!_yTSXb#cW#<0fo&WpqzY0Vb|NQ&~u_}$Yhp^3xK)O~M*!ya`z5*bw
zrBSExs}-hj<}3tVR!rsYtnWWPps#QM^n&;_$0UE*#XCk~x&GigkI9}biQGxc+FaqA
z`E-Wq88|P9wlyE@u3BlB=YY!6)$LyDxAB^a`W1a@7(SME73-fv!tD1kF9%BEni4GT
z+YKP>5k!5kNC5#?k%;5@l4ga@!3K>AC-Z3kb?gxBREN8tw#gAh5bB54@~5GHd9eTT
z%Pngx00B!C68V}`bUd~Lp6~#%*$s#Q%kSc}bKk|PH~#`=6aSXGPBB%^Fj&Lt5IZ51
zM@L{6@A`iphU32p5Dcw8WkNb$*fN~v1mGCjRdSO6*os#U5i@0g@XV)kMs_Pf$>msu
zd1e6ki{vgd8I2Ik`51V@AFsl!j<E`uzL%FEKfVB&GM2Oqjq=VfdQ5@2TrOzaKfN|i
zgu#$`l!Cxz(Qb{Um47%(Bm7Rt%yglKDHA*)*LSWzU%3%|r|H|A%Z4I21pFCd+Mhre
zFRB1c35+78kzDu0pN8A*dXN%VG76S~@2;2WeD%y}PbV$AUx=_JAH*WX5QM{82brfr
z9d_VMq9f^}>IzW(Rgbv|FkM28xf^!_v73G|e5qv2R&(nhy~tWwRQeS!!l+Nk--aZY
zf4s^6;}`SZRe-JQMT%?$X#sFYJEG=EwN5`(4a27fW8hx$WEsym?TlKP+jBzLO+XQC
z7l3bu@QxX$Vj#|Plu3Dyhk(~*G5`IwBT#led^msLHv^<Tbrw?7OTH{tzaX2>0MJMs
z^UNrjF;`9S{450gpusswFM#UJgP@bj&E!=+wVISoRh?B5R0hHfA{uMG_9N_FfRtv)
zwYA~#Uxkb)1(K;^L>|2MGXDLTz7qm3Qa<7u1jtX4@;ZP!`H2ThEZBU3+eg=zfb^OR
zI%7uH9<T^nf<rv-Yylk23O{M9tkML0B<vK&8gGhXMckGxzwXs@4mpP+?B8%6{WA$9
z=;Rcbs3^o_6p7VEUWQco0o^UYKL$YD{@Cm22fana_1Q}g_ra8!TeXH;_`?z~+z<+R
z%*vv|8JVmOR}$)88Em0YC_BkA>Z|BKz57HyLZ~I+X6}H);V^)$Ka0zjuYyj?#(c%N
zHnFhci&e@EksR=L8vt_fpdhB<lknmNz!Pk`u_Ppw!OZ+5T!d%P@j!3P_3ad{2>0Y@
z2Iq%zb~p7@B$gJR^A#+VuNgqle)s+>9R9*QZGay;1e4B3LS7&k3Y;f9)J$*<9X*Ek
z-RN%y(=qrX3YBIAhwm9(eJc)t<WB!|m<hJNHoc(lfWaA;0Y`cn=puRazt#dBdz5AC
zBbh2+Ke0*P3?f|*C<Cg>5XiCOE8ca`lkgmaZ6QI{u=Deqm)aw(Jb^6|I{;>Htd3yQ
z5%6Ie5X5c{Sj6%7V?iOU@%UZLo!#wO-*N-Uc~u^je)SM)hdG(r%<noJ1)>4q{GRL3
z1f?ftmapuElsz0JtHJm-4REtQv`l4i-(m`7@4C69@<nmN!WL}rq0@=<IJN90ON>!?
z;|<AGhF8~@s6V<`oCscb7*#eoi_xvFUt0*2&o$CxVC)Xg-yigIDx-BAsO|$_mp>eK
zL5JqClN-T{O|*vu-!sm`K&>Bnn0a9B%3;@dLaz<39g`mGuxbfcEG*~WvAWh$F27Sq
zd$O&MbTm^94}1T-`n%#_O8Vz4+^G>Bg|%+#O!g^O1x8xbe5OhTayA@n(k_5@GDlOV
zXhQX-??B`&h`T|LyI6&HQ9L~R@io*oJc0m<D3Il~%|f9;CCSB4K3SkY*^|B8vjpP<
zz43#-qH~ezJqu8_cs3@EoueNMOluYF_St2k*QK9tn+D8?pX_(&#`c9i%SVpq2L)SA
z0X$<yUDCo?usxtlIYl2WgUgdz&6%lnmLz8i#s_(N_wxF?e=w=NcX<!tm&_g<$LSY$
z?qr#Who;N&dSG!Xd%>_^33j9iec=Ch^r@R&MRYTH^<h8G&^cF)0j|)s4;#!d5<cCv
z`{Yb`ZfiFy>be_n5!GH?K!?0g8d;OsI|4jWW_^CMhmg}F{H^O^^}yh$6ki0m^aIDd
zvWxA>Hm><*T`-9h>HX4uuQM_OTFy-EjORG@i76Gp=@+f?g6pLp=Z7}7`_I|GxL{nc
zzh)7o;Ebzj`r1GDgDw62UDzTDqX}!6&8h5eC60lf0rl{uXR2-UWeb=qhjO(qlx#(D
zwGZ5M1}W(0Pw#VAJyzWcfebzbPIvcWj2t;w=nv@aR1ztR5LWmg<p>Cmx?8TE^w8|!
zPq{X_97xl-J$$>Y*iviq2!-LYdNEE@9|Qnd#|Vv=v#NFGLPHb6EH#Tn>{iQt8vL)E
z=$7G<9RY{8VTd5IDj(Ot9PHGKpr=%EX<qun4f`TRNveasz@%ObchY<H!T={laI>wr
zS~r3!y`Ao&cK<8vK#aO?MdYp)JXjceLvRms0!eX!`5c~`4KVZqGMXyWm@##E3&n?7
zKhfKX$=~{igkUl!mS<|&vC11$Z#Vl>?<_jS*R47Hd&7Y4(DM_go<uL_OqD10VSUZA
zbI`2+MB2~ebo&p#Y5u@lj{1!sq0~U_P1+iNb;cT~P4tL*#4Pp{<5}t)#C<DFK^7Y?
z=+?lR8^|N2@^5@=0mJ1oV>;kl-$4-}==G)h=W(LSk%Zj=%@OFSE%s;PtfCr`xh7o!
z%F7W_D28DJ-bF<P+zLK;e&#8eG__ws5)jLWraM2Aga@&Pe3s2)M_-&w4lMNRgy={E
zgnumgEn&u2cMj08FtJV0YByFO1Bu?kCb9nch9zK+Q!U><jJib8XfDK@{+N@jK6gm?
zJm_8M3hK0DcGoLDx<Y)AGeI@Zk9a>+9Cw}SXL&y+L*^3*emI9ohLsDR+-VzyxvPJ?
z;Y6kk5;ru%81QbaAtv7*XWjfWUZnsNI}7ARe0ArH{4M6>n75swF$#Yc_6Z}v+-w`y
zFL)pmM8D;2!<$1|b>x2-ZF7Oqb~n-5Jr+O4{7EDRp=`M@2{GHY7}L7Y@&ro)97aGI
zRZhW8w~0ALjLj*-B58<(e(PyCcQ}1L95uxo;#J7+N6XTn<q;X3UKzrb7SNaW;|lsh
z)Wo24w|=%&JlYFvzyo9(?>r~2S%jZxB>P>fUNkPZLL$*OE}R^X0f}a%%MtRc7WE%y
zv#mY~@U**JiLAKNyyJEM;6(iA6-)*XH=qQD7M%B>;dH$4(*@=@^_qa`x*d5jemRo+
zD!y92DS57?oDY1(DD^elc^imA5mGthL<VV64u>HPv#E(W_iWzdx*#-?Wzu-roPM(o
zmO(<r#RxTKq^+`PO{C%$Tr`lRIf2j&T#XR#@ax8)W(~0k|6cwr`YZ;T^JGEME4TZu
z@Ktk_ne!E*;O&RJ`uitX;(LsGaoIKlCI=$gxC!?sy%qd9#j%@N>Rvw5w2>J(1fqAn
z(2JeA_D&oycsFtOGaOAF>RQ??h@hrTE7<6C0R-RxS}|2|w$a~NW?W@l1N1BK5vjoB
zz@@q$#7XwO#M<(o{VC;EWrCjWVXmr~XEJU5xVGK*^x8Q>ck>r2n^S>j-U0=E7}|0C
zj+>}JV^Hi8DNk@OLV$A)B;y6LR>2Ee68<O#ib}_dZKN0K84)U%Dp;1Px$XT0mu$w{
zbmqFbCBtf;pw^?ezts{?UYOj?gf)eXO=|8F=pE)tUM!wO<3&EmX9XX}oP@~6qMdu-
z<>+iREef9XMuu*X<?W{%s&Et`;1{Z^_Jt$7FO4TvT|cY&a8FwrtORfcrmPV4#dY0N
zD2j+jx9ap_f_JJjxhmJP{pOt@4)t?+#*$ka`{ov|h=QKM+ULM*0!DI45kIvc(tF%|
zJ5QkH3aSjJ{ldzx9@#7c_A3i{0M$ID@hgv}J(Sm?g3^f|&^<%cS9Om2rT|$}ZXyPe
zMmWCgv~LxTRWRzwXI#a+nI~KE-<yLZ(4FR>YeNS_UpECLncc|$5-Df=fONSRcOXcg
z1*+L97G!n>LgQbQ`~;z-a;p>~q;$@C*AYR*%aZwg7kmg+S#%o=24K59pF-!gnKWKk
z{&HKXVecJ2B7*vM29x;BhxR*(wUvycU<PZD1+Jpw(ML&XnZy|Pu%yjUAyhg?1Kv6Q
z*7#PoK~M62tRxX@q+e9~fJ`<S$Bd6Aw+Ww9*^%yRgNNf}!VXZ#O@B;AI$+b?MX%`%
z-_OxGAgKJHt&4t13r++%EDBj4Qc)x8-|iehnjU%wtNNM%c2;tF91bBJjSK9aMvQwh
zTdIb4PJ16&pUGjgV>h#CHxYFxV#P4;+c7!3AYSvT5?q;#f_C<9pf}GfO7-F9U&g8e
ztJn*5JNb$NT6S&x>jqduzCyY`a(2^Kk$j<Smmq=1gulGLKkuL8TfNz6dhim#M{S9$
zncx22kVyem{OC<x#RopE^AvY4R9=rOUSqYvR8kY;V7RTRO4&cUs5@A&)dF+*<l+fV
zB#@SvBIWG|xl|7CTO#h7Q`G*!p=`vv(GN;v>Gp7}r!**nL&=DDLtYg3r*GYJFSTa2
zxuHA7c)O*&=pfYU9xK9_9nEUeh%9H<G9{)NWqbIkm1F0&JFoD%!dBOHf)S~^`M4IE
zif_V4zdRBxQmJUH4^|c!XUKG|*f7+NjoE)UzS&n>apk7ygt}rgd>e4tLR62zR$Fmo
z{W3+JoSIb+gjOX#)}2lZ4kY1(<QxXhmYZOVVS{;R&B^M1aA-$!$lTQACE>z4e-)$t
zp~DwNbfuF#H!=S11rWZ2+x((`<wc;I_8rZ&Mm;s#PKzn5<cVpf@m#Y(PMK`^!y-Hd
z=0#{~uK%2UJx)R~dU4=Hg@q#NLV1gS#!2ItOV#MTXeVbB<_%KRZH($$+*_IX)K;W#
zQQ9Ham#_GZ`7_*+aOIUCB9b&cFM*A9%k&BZc8yGxCedlmSETPAb9EA-ag~Gaj?L9y
zcj`zHHEP)(+V5E1^)OjnYR}8#&Eq|B88@wouaQeox|}F+F;Vq$;$G%5r7D5~Xa^?D
zvm{b46&|O%ueyzgvS5d>SckBX{~YMRA<-3HwcNhY7%#z<_)e0%k^Lh<QfTRln|5LO
zoz<IB%Za(WT!Pznai60fEtn74vXF+dEQPaF?(8;euDu`i|H{I#>(t}twGM;T$a1?u
zU2+IToCswhjxZkNPh4X!D=Uysc)Vh&u_%K#TPTnyQ}ag4Vp!gkTmePzBHl`@J6*xQ
z=-_$I;sFW%q1(9Qy;E#Gay^PH7+V9EJ^=Bm+LWoplxe4rthM<yy$Ok~`fx7Z!xqn<
z@|j<OqgdFRYHZY5v<jU@2G4it78@XA?aaNX*g=p=1$X+k(Gt6sq6L>qmS3AYVlUg6
z%y+g@S{%kZQ=Xea&m+Nz`QY2xe}!NF8(c-=CqaT$1+mnZ6E(iokme*~O5ucW`uK?F
zagdp`zoaU|e_%_{$VkT<>d6uPtuYFx84^0?34Cq}^|8N@$%}uX(l}l8+4;>Kp+wza
z=lYRsnLOg2DT7$`DQH58+paup4Ot}$=XoV&`carmL#3E2T|`-s)~KGup{r9mWn!NW
zdZ*5GyaH^3bKf_YnkZ)5b7D?se*NX@)sN=IC;pksp2S+rDs5IC7xXj~>nSgGn=x*-
zsncX(>B1%W6WqN=SDiwMtII6<lHQaQhM@$BKHjr5X?c3&qL3}WKg=G|fRLvW&6<iq
zA1Gr-NY%z~^Xr9_n&1~t^dzBv;u~FZtcsNz?$Ly($3zqHr>i7wT!m6xi7<s8x14q8
z`t^wg;{k1K%F@Vq?#x41!Oz(*-FkQ4<?ep>+(|Nf*!+c*Fp8I`HL*a@>qF~?&QR|*
z5}6!stkxl;H%c&-EXmdCCa-pnG?5&G$xwQ)FN5Q9X<+H_1~AcD<iohx3T<|l1~cXl
zaWOQ`3`r*BisNGBZHw7x0jE}5C!Vawyl4GmT11F3ZH@&ZV54YrHPF#ZEg{))JX6zg
zLu$0t`TOGhHQ47%;I#U(MN;lJ51eZSJCO)uC-?Kypi_*vzIpq)9F$t-1}k(w;0)Wh
z8R;WEReeua?5&Tjop0=TuX^L>7Q>vB2N!7-vJHtm0Qw0>QXoR9Rk9``j0D-&iL$eO
z=j%r*3nCVal$>-6)2;PXUEWv<HHWbNe47?5Qf<e->+vfEl<5!>oMRfG6zyJJmWr(a
z7GZuG*LjP=j&privs!Y^QzqU7^|dD3bR1ImoP7}T7QV$SZsxFJYuA?%h23zn%h=a~
zzO?n~k3btalFO<u&^%~4<0$kMH}<ku67?;xnM>L;<**Y@cYkL_@Iz1ylW%v^g2@Cl
zNSeo$M4wpkbOzn3-cUX5+_E&>ory!@KD?|#(y{OIW4`*CebgP9W9||&7Jh|YIP2GJ
zZ0@cSsbb8=6>7o-9*V)|yBYm;gzY!XBeoUn*pkBESW|WK=;v9g#2v;YNDG+`MjDkO
z617CL$r?ARzJ_v)XN$Ty3Bi2PuOGj-DUPYzi#*pH>v;Soee*shu0wDp*dxGvRU+wx
zd5Qe~VZouxP7kh(k=4$@;}T1)ms6dP!B;f_Pn70mOvLZJFkh2U7Ge-l<c#`VK$7-W
zYZGZC-)MUk<JFRBKX)tSQ%V`RMSdemeud=A+qBEE#n!bKSyZaBQ4JD;mfL$s#1HLN
zNr{vN!OHb;rXK&{p9HP3WE|(%6nhk8)z+nx?fGI@v$X@hEwt$T$`v0*<gEV0h%=r>
zFXBP(aR_-qzd*c8O)hrwCfq?SwvA-u)SChrZWqlvst@Ck`Mf3h`92MvaOrU!dquNW
z`&Y;p8_n1R-=6tC(VSdAHo2dtVv7JP&YTgLBT3Xv*bwZ?hY<*EhoVUTjnnj2W4Dh_
zR57J{LfAUZtJKMM``Y+9_0vEKY|c>4G$S-){QP3|-xg|{j1@=R)vT^EWq+f&-Y=t4
z$|A5x-9=hB%F;u_H-|SDQPv5?h+oZ|6Itf{LMs?H`5zv{B)sG^9aMCi2gW~Vc{`UN
zIW7izqMPIW%R3JCkT{l{`xO^+0t`B4$uZmtcDuz!*t=I0YZ^Sw7yIgur}{l%1>VKl
z%x*d__`fK*-ri~|xW=7qtqHszx<<_bznKJiAPV&Gk0$qL1Nh<#ign2`d`l)Rgl7>0
zG-``{<-IyqG4y-I)Um>>;gFJwb1bBfhb??xRO68O;ztKk@fznZ%dsJ6!)ud%BuCoa
zg?cALRYqLg3fDjndZzRNJjVJioZZ?RXJ|<gmAvkoXkDf(-Apat;HIazkDito;TiL@
z$8Uc{V(a|iK<EA{f6&{7tKVx9a-1mV_l|DNhVaoq6j3$KQ(t`hIVY#|e!W4onz7WV
zJrX&h2i-reamJ;1vbXZB^`TJSAEa=x$yZCiu4bp(cQdz?_YLN$f#fIB){y(x;C$#C
zEah9$htI-<$I^G{uDqq0R;<w`b{wmC(kt%xFEt~hty=ty@o{7j$k#d&E*kr{cSz-D
zbQ$7`#PwQ->$xshi<ce%5QN6p+2TXDb!&9I8e8vZaiMkfN|KfMs(EItnsG=-aZ(lx
zB1Oqt3yiN#c%OQTYuhywa9|c&Ue3YuD6!hFO}n9dt4prhZ!TrGt>z(v&a|&u#nx_>
z-yQ9anO&3q%d)^7J(Fbx66?05Ox+dX+Wg1ePjT^mOEj#8<5X}mYAHhmm6e39#ybmw
zs8u9mn+NZ6?QGZQ!<0W>#ZJSr_e-L!3dbYwuFUj~<c({-5gP;;^mw#swIvfA7+Um#
zFUF$aAft&S-sP2@jkDmnt*YlwR)a4%@;&FPPC^r|Z|rft8)E;V8+%VUleZP$7@y~l
zBLn;1X_CeQ4x`xJ@{Gr^Iu*3?R5u4L$;o}yzMIiLDq5N=HNGeme_F!J^RoU>V9%M6
zM#!U25O5Q{lfwI@Gs6M_d>O0=hd5@X^QwfDw3K#fEk@sx12%}m`u!7b5BYB}d?KH9
zCax5644aE(fc{C7U=)r7sZ4U6le?Z(dGMr;-_X`deWp^LsJE9^OW7S>3sO0fy^@&E
z@b7Xbs$NWFXwdql&)*<@0qx{hoLKwNno7XoYMwmy;%$*U*yGJaA2Utkb}4A9g3rXr
zC_`*l_Q&9uLd2UQ<CiG-P^DO9R_v$~&|m5?QWH?LzOSH~(ZZVfx9wxRBk>_8o;PqE
zBQ$zzEDQwY+<c3PET0-RkF!MBr4x$n=^~iekFv4t%x@mQW(B`C{h%9i&BZdEV*HAu
zlArxKMl+Q{YYzhIAUgzW@6Bk2MRW{{X&4JlNYB;%XY3e>rh;F;6hz=n-WVA$sDCEJ
zf4p6KqB5;T{vfdJ^!;5?P7UpMP6k)hG3pA;axaO*S<ex#;;Y<IU>_)A--84&N`*Qa
zix)RcqznCdy6a5*4+nyJ4aTyV*t++l<V^+i*fHL_rM2BWP>7~M2#uU^8yvBVvBxYU
zo6d{H7!9^l04)<o%}m&zC>-^9$R-t^sb|DWyD7opExR1dR`KWE07+sM7zxo*O>;8O
zHnV(;Ao4f!Tr!~FT->!|ZOJjFx{@eUy)@%dvf#6!mS?;_!MAIya_R?U#<WD35#;Kz
z*OMsZNbl876si^3<RqyDZM{vb7jmjcg4qr?YaUuBKu6%n6vTSVYSccO%&oU+SYZaY
z%L6LAOx`pVaFDC^VvEWzGGQ~spT0n`p6GDXiR37~1dopIZ_iRkA*gR-;Y(6PuFKQK
z{>x72fQn@ctHmoHZ-@jfYVe*Ax7nJBm1JzQsFKfE<+W3`HOIsW7XPAQLo+<J8cF@p
zJj9MwIzjg6n~cNVrxKoxGVjzV?BW(4yZzb)TzbbgO6VPwD{UCl-*_2OZ;`DwOmhq4
z*4V`0df@k2U;*6fjQ;hIR)9inbhq5x<>8iPYekY^9FaaD(^-YNn2JDq?VX}KE-|91
zGdE6q)0p+0h|ueRr${<Ej%FSF1C<=!hwV<4_nlk8O^%{%=&_zgKRbSDlAh+Mq6#L1
zh2W5xF*o-LP`Q8onmg_7(DI7~+ZzGIM=XRazJVm;-J06F@O)vkBf<Hfm1(}(vW+l`
zZw33e2~BeK-<VbT*swyS()#YZ#=vRMEBFlEV%13R<u2|03_pm+6H?8}WXSB)=389M
z?tMMz%^C@zlk|hF{c@sQeMEIX1R7pQWaBMj+Ih$6ahYc#4qx-k4U})LYCy9Ful~di
z7^rMkw?k)LZY}ZLpf%a#C75K7Ceqe?)W+1q>su*aEdniL`>X!-(VZByBH#0fQug~o
z0^{y>^wKQH2pT=KbZY5bb`5$_brfqqsQmdcjDPD#uv8qWTlPhyUe4O~rQY9BO}DcV
z?+`z6*C@Ac31m^7{UN{Ta3;Pbh&Az+?0uJ~H{AA^Ik+CZ>m2bECQjeey%@f%Y9TjF
zlC>o~=JM1%Af^?%MS;XJ>kcVZ_hy_3#tz1h*THj~`OS+Pk})DaYm76iFd^J*-+`&!
z6N^QaoBYZvOheLj!KbU14t<S;num~#lNQ62T^Ql@DH{t^zt~zAxi^?xWE&vUMCd+o
z3W2O6Oj2Bo`6$XxiAUZ@Q7bZdK_-Fy9i_b)o4GOfJkZ&*{FEwR?F$_NooZ4xpPuKc
z-wnN#Us+xtuuEHj!)et#?6zU3R59Stx|Ya|`Dqj#-4K~oAK4`7=okri@?Z^JVg9P%
zFfih3ex*n6$3=|kWQOtgJXV87aes8<X32&6Mcx>#Bkg6*tFU`-Id|ohpHaQq@H4rD
zd)JnlQB(OEIEicRxg%Z*j&@4`mC_(C7!5EhY9CLsG!nPywm;Ft<JMAF{5rUi@9;c;
z{F2-yJm)OPz19Bes7(|8EZ&t;E{!IR0;BKpp<cZc#`;FZvkCzR++i2;!Hr0(Gj+2f
zhosj!Mqo%Xq*)=GUOC&s+k6OfgQ1ozBx@{rXt(H&)`;LeS~!A!kE#+`D!`Js@jF_B
zE*23N8*jDrnd%YO03_M8nw{Y^iSua^V(h%9q@HWO^w=bZr|oFT1%^!1`e?mCO?*y(
zl2V=lZNA14-}_p3MFQz~rW?6vbTcb!IzTO~>l-^My6U2+3doXg@P!(f7Y>93p0+EY
z4C<)Us_bCDa+=zlP{5hWT$g_U!+Jst#;eQLVEW-I^0LvBdT-l6{v)M47cru4)b)+1
z*6g>vm3^aI*Eg)%?`n0OQZ;*;UjnZQ4hhM@uQy&S-kbLCJdo^KHm%3Q3}4?G08C5>
zBWTNx+9Y||hp~gyd^yI~z(?%VuGOj1>hxRtmU;SQ0)vJmVWNK5?PmS>`&60T5;=z_
zEcbdoh>^*Q^cJoHlNwi`icSFJuZ7F$Ne@d3Em(xkAwDw0g{&#gOz{`LtbxO~NGr^<
zGd!jruJrp|sm!<?n-OYA-YTR&0E8po5~@~nTK9Em25?!LaamYUEIkXZ`cVs}Op5Wk
zZJVdQ8smRY<reHH`wG*$(<HM8ZbmiU5q#lJDH8#gd=b|$T@ICs5iY?#?UaP#R}NpN
zuWyWjsT5j1olHKTOx~GB9(zRSz3FT76fPgXxZ5n`>+|x`RPwr1t)<3Xar2eSU<;NV
z*RsC|YwnNpA)Rt2GSaI~w>eE>>FT0L)mmm2y9&zJ3i%K6ZLtXo*?eZ1)qWnBIMJgl
z80UD6TTL(SB4NY>wk=nEan)<?$z7e9Ay{o;=tv{G?SJ~G$WOqB+&JQgJ9RDjH9FpE
z*q67wtsl{GqmnH#t*O_ICG}M!JI7+Xbr544;@Vg@L#wR(KJj#3)uEIaSJ4OUEE_SS
zWw5g&a*mWQCv9zhG`#ukrg5x&GWGDsQ!xDBsj(GuyPIXyt7o?$?=KWq7^Y^n?KFb@
z!H(pS{^RdL?9?-YU>#wN__hluvnJfSDxWGgT4ckSe|<oO^jm*^n*4?3nGMK={jp4(
z^mfXkXBF%;s@acDlf?SNFfRAfS*uft7$VjjGhc!qSG4WJ;@-<ro7-}Gm7K_VSdys_
zyM7Sy&87dc;AQbVpB>b~M8#vFcyI1_uAxQsX7tcEmbY7r@641Vl*nz^Tk@I7wd_0d
zg?la^e%e)Vo}b}&5L>oLQvG3b-)Vz3r<Oq2c~lSU$+#tXzOUM+bMPp>oG8C1mLFK^
zh{{c!(~G4#E>RXp-0L`Te&3_aLW*yEx~pSxH_KdWgi~x8q3bCETj^JrJtR5&drlSn
zQ5Y2_4lc6m9MJ2F<4m+}JP)NK@?;!R2P-DcbI<*BM`)O+>i7nAInHX))C)|>r0)V-
z!k7>@kj6GB)GU`8UB{K}0&p0P9XENax+)jjeU3y4ci(4v08qeHHJs6m@~%m5@au4&
z<>d&wwa#cL0yyqdHPdy4ATA>9_|q@!gF;z};_UIx87KM+Q&=(7aj~g-9$Pj#4U2tr
ze$!<Cz$ihXc4;urSMBpAAM>mnCilftEw_p8h&7$@lLmKMTZt;(Te`1A@yS2@K*uKX
z-|CQhtWl(cv+go>?ua<iP|d!r^i->F2y3FeEQt#iIjV+edZV3@^zI~>!8qx?D7oQ)
zDKV7d*(j`NO<&#}^g0*wHARgdDG`ck<F+11$Q;)tmBkbr*E9qa84{&QeFic$Bwjsr
zzDA#T#R=Vm=*2M^>Go@(!BN7#fjdQD=ex+H*ldG^RrLVbrCh7im0hkY9vS<z$S|;z
z5WU*ezCZhxYku=q<D+AlZ%?`WUD0}n3_iP4FyKW_o^hcXADwgqUy*QsQd`4AwV+WB
zctI~eIkozKN^QvrO89t4bfoltBNLt5!Aw1_7f3s+?@_p3`J&^VM}ny3cUS|*w-w$C
znU`aGqg*(PC$KopLV`9@7*^UsdUG3<jd_Ak`3OYaAS&!6;Et>K6`>S+|3t~t0WaDv
zIyK1kdg^Y}&PA+zdoc&<?T3}=H;CkgpIvi3J;a<Z*v4d;ReNRGpbJy7&YYnlh%OHk
zlXkU2Z3%C-Ms(Ug<zdS&INi6h9=|nXyz^f4)~jPr@UGhDF`^~mM~QGclYBIrd?|T_
zH$hc$u;F+^qzSL*Q?#hN8v~#G^1i)g(Ri+dO_6gxBrWS&12cl4{f;c~KL6|uu8K!j
zHPbyMWp`(MQ#(ZLn-jDe!XCre)e`cvSlRAtZ!USGkZ-!GY%>lHds#3;HjY0(ISx)R
zeKq4LuOq$FpXWtZ5_~2|DJ7)y<MCA)rIli^-h?220vV1+`k0}ZW-F&!Y*lZj6XTF6
z5;=7*e5~Plhh9HRu0GMh+lI3eRlO(jV~KkjFJ+v8G1cT+;0?#js5;q-DMrQVGsj8B
z2^F5Bn6uE5YUVB2_cW1=5cYL?f~TbL>QdRavU*{aE%xh?T+krdhVNdd-q;OY15f7D
z%{Rs-Jzw8Rd8PT*eTh8V@1dbfc9SiIGM=Q!lr%<=sekVC+}JfZ-{y~1n<osCwwV8#
zO>}vE(TK=M`K+&4q`iBLPFKK$pGgRc%S*9h>aFHcXQb!!D%L(NH<1IY{-9-S+s3+H
zCwuS&nz1@r<eM#kFsZU3zKaGG=X^-}rJoIad3kqk``{sMUub*M;l>O~gY^lmfHs!I
zPuphd4O8VypBqCX(puk+ysErGBTwybYX(8yJG&Y4FP8*{Lh_fLJDhKRYUM2|;fQ`G
zGEKK>gx`2i(7x=%9>2@yRd&R@Kz+=~yI(gclksP@i<HcRW|Y_iHhy)pv<F@>5-&$n
zyq7qRZZ24=TsnMqH7}&u*}3NB-0cKhL3V25Gd4~ne(&ABov3^Y-Lx@9Qn_isWUghs
ztMku~Sn#J&(HK*~Fi#)TihquclaO;SjmrhaaINa6%K5#<WIm42#ZnC5{N`!-h3I0K
zSv47LPc!nCIt|&UP-?v(;hZTJY#`%Y6<M7zJX`#tA%;a#rI{lAN+15`+MmQUC!0BU
zg1KV2%=*%;0SFPC!cuh@*b^UC{(7pnJ}u=`^sitED9XWWr1pfry;3PwE{_BIB9Zv?
zw-e3lNaW3dgM-g3O4kBa)Z0W%wTPs_wQOq*vQ4Ux7B!>HvWWs5=rof7u0JqJ>stRI
z{c1sk`)<>BBRlMD!xFG@DxRs&@>n*b&E&NUR4jg|DP#pC06nm`#YfEj@UHsCR6Oq0
zGYVEH|Is0|nC?CN2#cH}hiMno*s-~m2?{Ykv)S>lMzUh|*8jMz(@*Xr2;=OJFaO&V
z<6nO?E@EuN=qlDa4Esle4NO+4gYy+*q?yjvJm~`7&)A(V03Rv)R1L!WToX=@%(thA
zmMupE+QP*S(asjb`8+nbe95dVmUlkvB#%P7v@$DLa^4F#RyDu~JrDZof^6Wi1R|&}
zB8rQ`sX-7XP;8cSfq7OokmmSo$3NGeufCS~91$<Bv<thD(C>efmFx?5bJY2Y)JxRk
z6IHa5mWaw*2U_$z!c|@|&zJpOOZ-osx$ZaYm9HnitRcQ605M0+V7vHeTy=H?E~Eym
z7siJKJ1K580*ik_e+rl*UT?-TSN_Sw)_no?=+-a9B}Oa%wAB5VLMU&HRlMy9p0rZA
z81afN(6x}fINUiB%lN+r-&G%x@7iPM-NpGI$^!oOM7b`*<Gps7KFL!3e|;PNc|YI_
z-5%ugKd(zHAV$ak`G4`z$P^_yQbF(!m)n1RtA795{X{VPZJJkG|M#B!mpk}pd&(bw
zP5usSKzExP-zWV><MVHRG)f2fe1t23(!qZwR`|y~WVr}MjdjMLTj8%=f0z(_URLzX
zrQ`qj*8OpBqYzu+sEwS3_kZpBZzjX%k*WuH{J*(HfA;%6w{OEDyuCkQ-~HFFKkcsW
zC9JB!Jd;4gi=k)m%3r(>%r|wRsNi+KSD=o1Kq!N8&2F}J4cLa^x3TQef9XfVM8{vk
zM%sUC$czYYT0SR(0cJ|VcQCH8L@HAJh9~@QZe{!}Y$A*?;5ZrB%1NOy#MY`#*<y<P
zfBJq1NfcrcsKc7B;=_;4;*=Ts+ePy~|Ai_PQi}ZOD_?*z^tTK5zyAd;7o?Q9q$aOu
z{<Tu%hkU+<Q0MkvEJfu=DJt){>+~0|zXmBqlYv=p|JqWP*G5Xwn8^p>zjpnUdPpf+
zt9{h<*OoF1CsK-5^SYD&+VzKVA*E<<l5OFyJbmce5GT~A-NA+SzjpmM<KgqnM{6<m
ze`&q?G|53JYC2r9pZjaq&nyq0M|<wwDgEE>@Bhy_s+vY%=4lVdQoFx){b&~WJovR=
zgkJd{+sl9Oh3*D1EZ+?=;n9-6cK!c9S^i#>w@*e_Jt;tGK9dd9?1OQx%-7fc`AP-I
zO{_Q46LL3X19^{dS2Dyy9(4pxvxF!6sV7vBSz&WR9TiG2$#nlBt&~LCB^ieKV<nb9
z+ryw%`*b4DZeHxLF#2SQ;wuUWJErlxCU;eVj#rDs&=h_hZw3JYttCKMYk;f-ocvln
zA+U03(WoJoV@JPNNFoOSGKvetFa<K}NYuaxa8EKUytW^Ck^P#F{PnRQ^Yq_e3rajJ
zaP*eiYf|*=a{Poqw3EO>*BLIn!Un1&>taw6X?#F}EGEGU+{~<^$^U(K4oDkLOCJL`
zX*WLXHx7C4(~Wj(5YzC(BDj+Er#CmVH5?KAkwKFlfSUNwRd2>?6)oT?#Mtw-$*e1g
zVm<sxSV4cElm3#-Z<YtwB_D1Qj9>j;$8mU-l=2*B5fOnZs3k%j3kus3`Ps^WvYoNL
zRD*@bP9(PVn$RCPh?kXg&L9$#Nf6(d*9TwIg>baT8FLhL-~zptMt}TG(9R167BXfN
zUp3T;+S^`U=cVse>N}b9`o%f|3?Pn!!)iiLA{ew&ebvDU099D4PnHXC#{<0%K6sr)
z*e844;>Y(-+(3&`1Luo5$tuK~*d0u^31EC?#nui23UP==OU3_sG{ad7Z9LJ$I(+-N
zjAQk`J^xGphqAYhszUABcNY!PE+s@7K|<OE(v2t~($Wn|N=dhLiP9w<(nxoAcXxL;
z=UIEd?{9qPjB~~}&VM+@)~#zjF`qf_`?_fl;gs?oi$A4jk4EVLTf!~|NT{7jVaGSw
zK_ygm@J{e9U^6bo=!fR?t!v+}X*a7GCqBiSdbnwLaQ?O$A=#XK+m$>TTJh2=b42+5
z3WQCNfTeI>xcSw5CJ96WSpkcObE0g`25@;f9J$X3N1Cm{8%Ch_Hvcm=!D`wx%NLl_
zcoKjqVF|ce5)I_^_xF@#&=zpZY_jB%1-RH>$$Ums*dy%|3}!f!9J`=)H~$|GhX0S^
zSXuy4pB7>ubO(0@z~Q?Ik1aa!+Th8L^`XBi-MshUk*Ht7$&RMMgkf8G-QUA=8ze0V
zGJ$D2*#hgm?$SO8iNN+J#^Bwr)Q7j%UEe$$fT8iloj>ZsBJMp!Eiisz!Qp4B0ON2w
zX!dCm5+-=tB8YX;%ONN9;Q}Nj@RA#@>|im$!6f(6Q2@(U1Yjc9{QF?|wFG}0Z#t>o
ztV~%ifW@SuYW{bunf&31yvD@Ca~$M9vG*qN^J;*Q3*DYq6=rQ1qUF>-#L7G%)}Uu$
zu{YuM&#5NF=IMS@;eJv=Vy*nB@-7Xh>(fU~;8B|c+xF2M23*f*L+QkWh-nH|gFA4g
zaSUg6H-P`k13p|gX9YHgFHnBd6@}W(>iT4k92vWx8!rLDASO%Q>BohcyRkyIwN%CJ
zJ8xV8ZWqu3!Uaw9O=n<0_u!;Xx|`h)nDBqm84xZzogztU_cQ3>e%<okh-&s@hI4it
z&U}L<Fv=`};didBW+G=Xx^$ZZr(ET`JNdJ_PHpv_|7GWaJ2b2@1_MHzEo>FV)U|m0
z6oJ%Y86B`*Y$pNI!TE8k$N>1x4mxJRuLK%h04$=czo~h?h-t?Pe|Bd1(-t7rV_^kT
z?~3{#g*w!x+2NWlfw8K66A8f){5-V+ZPII=Wo{AJD^;6j224JyB?ll#s(F$4WTMCo
z9}_;>-_OL|arVcV`1B>;qpcqF#IUaQUxAX93k%Mg)%1_8)1Fl^2lj*|aLhFXA(Y;z
z-cL`|uRDU<laKhHl`6!Bz|;lr0lad^du}%<jON>L8Ps5Pk+j9c^dNsv6|O>*069Ac
zLY*mp2&?APi9G)uKYEDt#JdK&5l=&HTJuJ|L3m)x0v)2XJpzS@#I%@nXNnBxQ^dVL
zmaaSM2p8jbOcrvhr^+%N0Ohkzz`X&KK@pKz&LrWDiY}5hFY2V>pZTV}x%vNEjK%OF
zw@X7}B)!_W|LiFK75O?WMR#?jiuk;zvGLJ<ATIyuhl(h_FAYFpoxVq4)pH4&bffu8
z6fUBdF@I3YNiT1nwtA{ciaq%TOsdHJ9X#_zbFi<8@OF(Z$X+p<BWP}CM--B64Y4d)
zoeX?g>QMxJ?}?6cA`Ct-{C__%M?||et+hgKy8$B99(Kgg%uXC?(z|{0Ty%)Fes8OP
z4lw0j>}FoO(UJ|}{DN40=G&5YlI%#kk9k%lmh=C(-H2Ea;XyuO*6=8}6+6u8ePn8n
zi)XbduYlR+o~y}1u0p}HQPFXLPK1ZHnP0IJkw^TWl0(PqF_9Dkg)+KzUy!;U(<7>W
z_+dM1_liIp21p!(eTI6EA#Q7o+896pkfeSAPtBPQbq4^HX82Kwib?{X#ea9d?Qq{^
zfVBEjQ#FBGkn47*;Xbbhz4aLd0rCR4??N2QbEqJ@c9HNybtlRQNJ$*3@YvnWlp&SP
zj~IOpTsC;`&63=%O)D(g?utm0R5cX;{U>hHuqx{n!bd(!U{3XAt`W@QwxZC~N!%I^
zgS2m(cZ$mrd2kqu;k$4obNm3DD$5;uLm0)}1egm~{SM#3w2e(V9@eMXTY+<b*9@OE
zDc8N5&Y><@0u~I6)@a22EtOBE8D-Na6_0uVwQnq+eYhY+rWjCt9q0Z6ir%GNhiIcb
zj8ye=dx-(;aMsF~*W^x}2I2y?)-3i{*u2|-w{b_he=_ht&-Xjc%*?MRdaIIX>4&_2
zbRq`3&+XBO#F1ip?Y0a}K9pk6;2^j9d*_MklEM<r;K=Kq&$9y$mvbmETG4sj1Guya
zbBdM0JD-zCJuff`S4xfgjxUmN)R{hw2+RP}U0r^Bh)>rWX}2p-GDXkr9v}_~_=3xy
zqwysP+=_RYbWuxJL$4@<Ht@svU1P~>!wF!Iyeh98c{JddlM(eU&$pq>w2JWf&A}V<
zZ`>3F8Lt`ubeJGX#bE_>{m6j8(X#AMHgo~B8lmto=<@8+H7eehCndEbP12BMb>A)t
zzV}S7AGJR0%O(rd6;TWH{P@=S(3SgUr>F|B!_@~80<^#oowVi|Hn`JJWaqBawr@P2
zI?M|nBN3c-wbWkt8$`CbZhuR9y(S@VLKSLO3k2QAtXC|K&r<rEdms(FE9!}S1MG&q
zeuB44LUrtaK4k1##;p|%XMX!XRiT&;160Vv8l3Kbf4ogb-+R|im+92VI40+I99eF#
zES$1^`DDTS6k`^AX6Dv7&c2ps@JyaF^X{|lbp4+v$X81Mar;>xV*m-4j}<-yW{P_B
zO*zfo_CM8>O^(gZ^I6-ZY#WwukvHcR4Re8B_1WNhZ1^iMMbN&(Zy7ft_O?T;(l4`$
zp}v@IP^@as)>&tlelrjp`grhdWwD2MxFmtU<ABLj<e}aSJaTU9J6N`4z!l<)q$36K
zw+iWZ)d<920HGbD-SY4S)I!*BDNb&tqs@XqL4`64O9F8U7I$HuxTpex_eWthzD>&M
z`elRZkZ=UsKO}DZHp!Uo4;Sv_@Hs)KA`QYK_}45V^<asyeQVKqLLPwi@;DeEaPi@G
z+GgWn{2Unk3DfA0|B~@MH_)#-@H$f1mZ^bC=yzzE0w)Pxr_NNA)8qH0&x2|%fjX7T
zIR>{A^c$|2d2~|9hWI>7+yS5ExFk80(#X#U>5R$C+D8IixBk$V#lXQ&LH13?m}$_j
zPpul_b{}S>8no$qepK%RNQc8kF1puSq%#6L;;l^mV13$Z%tb(s<ZOIejPGgO%Wk2M
zW=E|_14-AC8&ATIIMTq^es0~0-#{mD;a8n(4<!wj9`zs}53j;aS^+(ty+}^by=_p|
zigmWX%K*-BE>;B3+Y4a1tR~SBlfM#jgNqfNAHBtkSL~X(`ZA^_xSgGbm>5KVD6l;I
z#<GfMZ{YrZZsAU|@fXK3%`wW+VLzpX%pM0aW4u{OKt-&hGW{*!bDz6}W&gZLb|5)A
z=nyzBE&^d>H@dZqs&(JFCUg?JxM~OBh`_*Z_q<Fb;aOlj+PrvwlXwbGtu(jJdOAO%
zMLL*{aYV7FG%P1T8?MrsJj->(uHBw-xpUhbag2#TjWDhjRX=p^x~-7>$iK`{<=cb_
zZ{)q1PV)#5yzJ`H54YMZk?QlB3hUH>0bZpJ3BLBDR%cL^UZ1QwMm$K_JaTC`<kJ<N
z5_a+Lio*dL&OG%#``V2U3`?}Q*onLr$=D7+iqYCYxTXH_6)-6iMQOV^x?9p$ocs>p
z_Aj-V8gShQ)liaPm;0@}!%@DRmdvt!H}N!fEeQ}q)lQ6EagJ#gemduBE7`x>arx&-
z;~608VgFT`CY*NcsP0`#{uwAF|60UcI|GM0QMbd#9jZ@wp7gSd0P1jlk2&3eo^Tn(
z59IAFpxWSSQy^<L28PjbWS+-8$SU>LVIU>&ooN<%z_~+Qx&r-=M|5lvM}-g9g?Bem
zBRiS!&1%;RVCwq|Dn$mvFsNNlatT@6B-dMR;4?)0F1J}~xNG-EWR6dG0VA^e(qivv
z=ZSRc{@PfGKnOhFJ?(e!5=lbwQrEM8cCE@U+d*7mZ6c^?Ji5~U?|>bK&>ZoUfUh-M
z=R!)&4zt#N6%DMdDWKi;A*u20BS@^dm*Hck6eeIZ^F;K)eT<4h1TjwWXg7IQ)UVg~
z2x6L``wAfs<B0_4zQ1=Y2^K5R!u-9D7<~`vOu=qOnf<}mDJ}Y#u-0~!0;l)WMCW`*
zQPd9jZCD&{dp%fwVMk%XlZ!x7Ae}8b_gm4&5ccpcA#7z@@Ydrn?sh|MuZODuJQzC;
ze}6Fij&Du>&kk06@j<yv`4`gK4ZI>p=4r&Ux0wU>_urxr{}i$>d|dP2(t@h+PFcS^
z>*1xRFcL0mPHBhx<p}%JuZi*$fxay0Kp*uvBCmjt0)@r4I&uIAU=(}58-329&#T6o
z+H#y8Df_H)qy(!cTHEc-bMV112d3|5$_UjZPCrQ}XTvea_Awtdojp+XVc%MG(@FZ9
z;-jR;lZyAAM%Sp}hWRmMcwO%!k_fe!vg)VrXwAhgI^T&CJy@?hB?Xo{zLH%uVwHmj
z*_gOV4m7U+k`{^C-BD{*d)nBx_o*5ry-u>;I6VL5XxEJRk<#4vOLny*>QGT!{?X(n
zz{42gdN*xT+Ph4CMIuk)^(<)x6t8Q`Tz_1sMpv(>2*eEH5wRTIP0^0`b=Bfw&zDO$
zt-LDHJt>!|eYnfqiB*^GmZpqF$7~mA3GgYbIoQw$Y&i*0hJBShQa#RiQTaRFe@dzf
zrR>Swdm|YVJugB=S6}|j9uAVhPMF>z>3ycfTzkc)^CH~&cVD0D4|if@SxReeunja1
z#betl(5_;PzZ)Vky`VTwuS@*<=N+Bk0*~$K!`{MuzD?i0AHK`Wn{Vf3bNC1_(#&Dr
zX+?PHs&n|=_`99-_IsZ`5c5_Zv@#DW`ZAx=RjzJ>c?Tv$BT=X|p7vZx@@<eIJ}RNY
zoVeRs-9}=H^ioZtL)3-OfyFgqN%NXnbuRM+{Y|`{+S$C_zfhmXR_ewI;yu`Ok7}HR
znQ%?4r7U73&!&UXM;z#z%>hHi$IGsuR{@AOfZ6&u*+pC%-4!3XUiV~g@pG}<Zm9{i
z@eOWFa1V05@HKAWM~T=|Ks-UAEekPRyotTXqRqnJ{%4n-VKPPfM_RgWC$Zo8Oc}d_
zTKJnb$(@_{O2L$VyhcGh-6m$%eMSQ{%C%6;x7eqdtT|6H{pR?3<W*LryzkAL{RB=a
z58l5slnHcZzYCpkp65(s*wXpqeUG{wL@V;k*<`soQ2#T&N$RXcHU9@+R&zAuF{0@1
z3wM<Oxefouw~vdOQmiphd8g6D*1p7x2bmxZ|LU(mh&J)|8WKR}0Y4U)B+?I|3h|s5
zH=$O<ZsVI%^j{<cuE7v$e`)}>%iJGtF=+PR@xD(IWlhox(GSK1oT;y`Q?aQk_DT>4
z8opyNO_v+1^s>uxVj>|uSmz6SVIxb#zj)pG+qc8&0JcN<$-8Ytr@%c8<K|JAVH!5x
zwKXh;tgl(ROihinBW^6hWdCQ{>ul;H#(!IhWl&bOcV+#N=OLn>IGQ}7XMXn4Pd3`s
z9(OaQVSh}Qx&NY^mx_UwYWKCwyrMtNB&w6@!W`*z$j93JMy)t3<QwqJOYU$7oU_dp
zCu5JRJ?{H!wNzgg!82vVoH174V5l00M?0+(++m6IU~aA6@qzP;?vM`}@Z0&N+LkAd
zF20MkeWE$;lXRbupY{iM#$Zsqt2eZ}$9J&2vE&Ghv{Vd<G`sMpNgWc~?$cT@e2O>|
z#%VFHa{Vq2BTwcyRnE9dwz$m<bN5Kbvf^)Av@ixza|OqmQ(rW0plW+N5D81fW|r6y
zGpPy0xJ?b1OAQ#BF3);O@+TNJMH_dRx#F0q|81X^JILr4tNHegI*e4zB=LR-+23`A
znWquiPD@!WOSCL@fz&U&u;Da=#ZgW2zIS6T6v|-1`a8p2|B$H~r9Qv+uc_|LjE{VT
z!;p{r&}_TJ@n_hf>*>J5?Z6On_0XRt@g!{l)sBh=lG?O6#I;mAABUN_&hwnZP@KH`
zmp5*^V9;|f4aXBb^!Eu+K<_2$&FOKkjCKgNTZe3?ESu39w4vEfS1r|jy_R!4t`OS(
zu=M<{LB%O;Kf}l^d5wSH`73gbf4aq{-A7waqqTrNEduu85$Nz0S<V1kOsZPe`I^>(
z`WD|E(9YP>=}0`kO-u{>GAcTC5Xi@Jebn12;dT;JI3eSA+&U|%oRMs`KV=!3*Bp8M
zg->*fPZ#>5PCL$t`!3dG;D&dEr{Q$$bQ?D{Y?ANQH{zDca+!A|a#$f6&Z)H<c2Lj@
z4OUUvGiIt@U;sX?u<`(*D51Hp@s-_Cu~R!LnwbX$+mq5`U$*)1V?HAn1+p=FMmqpq
z(iHNDn&oG`hxgc=$W~a)gucM1<tc<J0k=64bE^YqF*Onntffot;01&e6M2XU-xe!7
z=5>S#2P{o(&)<kLV+DqH66;)G@@UiZp|@I8l24pu`w@g#hm8>@qP<ql+nQ{>dcya)
z_{4iep;BfA?(DdJUA$=x83`M~U$8Bef1ag_7s%};l%f9ELon}c`VaesCEW~K4~8iE
zF6h+5J!q)Ry<Wncm>&GRlD#jIf`ZK2?Q~talfa+%n6U3f{=`@rbu~&nHj#OGrF&U$
zAOz344bE!ld-ip8D)y{Jxh9(=3p5Z1=oC+(=dl^<m7^cuSNZ3tYuHsx;1j<7Rk4%#
zd-JwrP3$owPHER<d^RxEc8ScZ)S~H+gOg(fAzIc-KzyhIa)bBxk}Zn)TtuOI>~906
z8zTY-1l&0IFz-VD)$jP5wi1ngDJ3R)xNAXGxq;~ZfwXU8#Tz?YpZk=}Auf>m+u_Te
zkZrUm;z4A69BgCGB&ROCZ<{a3z2;Lr!aWX@N|9B|W-7)4(zf8%xV}_l&YlGR86r)n
zqOPRERW44?ql$cmf9D0X+@7Oka?n447SV3Tw4*P<;WA1`Q9PmX-LQdiJ?Tv`a0ZDw
z>g}JHiM(a!DnC-s1-{S|2J0BfM+YQE8nd+q!BZ96mk~+RDh1O6+<$Xk+Cf8vplkNO
zjtiLmc=S2!3iD6Ib#n1y2AMs858vBjoOT<l@<|Kw8`^{#lvxDbz$w&~X&fn|5Aaw7
zF42A-1VYtHzGk}=wFyQ}`6N38vr?uzK569@JrFC4bL`wa%3(2#o_y77?B)8hvIlh`
z{K}Y0z#+o-{fpT*21o@PMgvKE1s>+CEHyovypxL(jY@+l2+`oBRHm@d!uC4&J?pIS
z7&dJvR%J%B7|bHtDnG%UpbLf7ZOyuHoH54mkpo=-Ny#6^POS}-e`4T+5j{)xwRTvG
zQ9rO2JF5a6rI*^>Nw(Ee(9Lhqw3wLNHm^Nu*Yf+AHR<6~lisT}(yFUlSa<$tH0O$k
z+4*@-<!*Xt<5Q`5SLXd!)*MoGbFUT)o<_&qZO7Q&r!ZN+`aHgkXgUSd^ut@E4Tq^(
z83kOPvXakuWKB$3Qnohj-%Jx5c^3ct61Iv+<PgNCKKp0}OTyuc@ciae9kF<XivF@W
zCzNBP%th<5FCrOQe#+B+JF2r)%N9Rx7@%?8Glm6kHj52U#GI1K#y-Y#xxboiic{Jz
z*w=Ejv`&MgxNG9&G^{ckPdFm7YfA6BRqBVA-tmz}KfH>^>92^b+CR7HPk3{+=~N&H
zr0)CAV%RN!>SxE1i8;(VG052qNZxBVij^FW_q=PKYXq)*=OHnvQRrV{W<DT8P<hRm
zU-qM-<4ShxsnwTY^=I`8#|ua<@u0Vo--2QV6VMowU&jGIRredXOZus~|8y0+6yU@Z
zz3+>?Tz9`2OS4-xsow3#f5O~lEou(!KxI}rHf64v|K|vuavneW_Xve13w;iYhp_g>
zu%#V--*U*5i#tfOK&bFGLg2w8XZAHEiK}?_u-=qmRWHZWrjOWqxEd*!Ii>@3_flK5
zywOUN2@C3cV!~xs(nY&p!fpO#yW$;BuZ=M4mpk){My_<VT8pz+Gh*?LM8dGcyLX?0
zG&qLstHH^SbT?vz)mX&@@0iW$M^bpJee0GO`vk*momb}4|1sE%2Z7)#N{I<8BG-pl
zrU3#c?O~4qH#=F=F@SCDm=ni9t&Bf^NB&g6JtdJGEE8p1{M5wwI9OqkL<eyH-{+Y(
zz)3ijfb&uJs*?!Gbq~hYdSha8LYU#r_N0DhYoiE3*FowIo=RF@3MbWtkvmkK{`ynB
zoohb9_$jBm<YfVE`3r}_!_Mk+roebxy67q+tmcIv;5%}3OJnhI2tHsqj#Il~2u%H(
z|8Cw$x|hixCBAj-=&*Lt!fR2v*0iaX!_i&w6P<hPi-)pLm+I#=UhJEoDQg-Obb(Ij
zHs$6~j^x%bY+K*0m1f3MvQeTzE614_T5`})bRzc4O=O#F-p?$>jMWl<N`|yYrnFZp
z#UhEKfWi8M?1z2@*oUNe;h($ac1*^^gXFSPPQa~)+^*s1w*{?`O7e6H{gxBio92C?
z|B()xdem`X2gudOH<U(np)_^(ey{Ef%UF89h){mM7BNS3UJ`&Q*P-?It%p?Gv6+iZ
z3srJt%1Ej&g05FZ1&W!TAuFYFQP-tLc@gO*FT?$p>O{2+%}8^uBdwDpBc!|LpKbJm
znW^Tl{|ME4U`~jUx0Fg$h!6{DTNc_I4e=|1=kJjs{lT#{R<-O++(j@O=k@vE{2}ui
zk~!xz-`zB!(y`e6r%!~QUGH?7A3?564DfpXe0btkiA4qBe2+4p!p`NLztJLDq(*%t
zfghsxOv=`FP`IgsfP9LRwzot=cS9-y5^ipqf{j4wt9U)RO$&aTBZ55H#P8A<<eNIb
z5t=BT={346YF&|VQq+=&aZZ4qj&qu~7iY{&piV)D%OI=sfD~jhkAsoal7Pxf&G=M;
zne_30oU@jXiF&?stB8g7p*^;HN>1x?J9uSg@w_%|32^j}Lk)O(w?<jtyPAh>gXLs*
zL_r>P=c5aM@&$!$a*9YNVXx3b#BTkw{c(kDyyNQfLk(M5fnxi(6CJWOyRLr;GDFT`
z6Yz9!6%bzfL|*Qm*rBluyX?0~OYBZ(J?WPC#>I@Xq{gV?{Juk&(s}eNI#gn-=VH6!
zGkqGXsHmbP-dNg~&4jo5N$g(2&BCHnws@9Gz|Gx?Z4}YNt6FJMD3nD^$_Yj`4~<?c
zT1QX!(oT%iLSV<C+CFZk6aoGCJOZ0y4PcmbTSMzCVE8IB7Y~ojeo}Wz)3?W{Y6Puz
zQ=|WT4h19u{a}-IC#>WE&5NV7d?&GP$~aMzObnv>IyAHAo(8f?!Yg>h6rY(H9p9}Y
z!xxr0l6m>>o!K7NNU?Y120%-QShDdu!i0}9t|J5tug$JPH=Og|`ZtPMOz)ol;*bgw
zuxw?WvM*MPKJPk^ZZ<|lNBRorg}N)=X-Xqa$Q-n&tU&u~4&wBVU|v;U%+BUc5}Pl0
zn)y&anch&;rm^t$_@2c2x2y`Fwgv?0FTK4swS?n#`Z@lNlvep2sM3nWomm+_Qt?qM
zJJL;TRd7)`tP9$Vlsa4IZ@SThqqfKyDP#-G3DR$8sGYG-N)}h1b-hHiX4PdR$9u;9
zGMK`<X~ec+74sN<Bi1f0oG@Z=U0*2b=qarYm4*H=ITlfhM6d);Uunbw36wohL^Dd<
z8djViJQ;M`^bK`S)@FpX_mk72tBy1}P7#gZQGO2%bCQ~Tc3j1q0}mt|tNGTROM(gd
z>vDnp$`zMdhBeIg;(oxIN}Vw^XS~t$6hg*XMM1oZH$48<z?PGFthBSvfoV7r@n(#$
zXEh*m7(8x??31#oXmm9q9jH;d_1Zt<zF7?~%Q>u|j}niV=!3SZ!Ejm~f5!eKOiy&G
z(st+8Zy`$fU7K;6!k4*_c#U!NCh`E@CD9?MuO<7PWF$fd{T3TDeNe}!H)5%tcu{K+
zJ7wUi^_n!;_e0jl_B0F%L5||8m(Rf{S{F<%n5Wlfa_5ogxX#=LEvO$@&qW%Rflp+2
z?^!nm?fb)mk5LNR;b_3yjxwYR{oQ46G-i(tdX-#z%+sx9&ZP(~Pr43~?OvKwOzXFx
z*G`nJY^uX>BLy&gzj_sO704tXfRWq&-Ke)#!@H`)uo^SRXu*cwYRAH(Bbw<C=nD90
z1g63-ak~UfQ>BMq=dK5&R+h;i10f+}Z^sLwKS|6UGE4;K>mNC-g_rKs#7Xv6%B-}f
zO`5AsqArJXAEy0m%KH6H5=n+oJ5%jE#55Qd&pR3BOqT0R2wqoL%N6w#a*dfa|5dI%
z*WG6xw6Zboei*kWsIG#vyHwge7Q2B2jdy*xH1<x%zX(vC{M+`X_o;@^5@J=C4kE<(
zd)1W7TV{{G<M(bV9X!lwU0mZj`;%GN@aveh8@YY@Nrf7b6=B49)f4&#`uM(2pEw!S
zhJ03s1XNSinghqD@XRxfJJOnzXe|Wy^fUFf-MlfZ`0l9)f3IvI#@Iu)%KLk7Upf4V
zf6Xdruc!eS9>TRCtu77wnOMB*HnVvCN`?1(d+&g_RM*-zFsXK0Y;dzST45>G=;(C;
zFUj6J94E7>uU_YDj<wM}Y^4yQ{K(>sSBLw@_3^9)b_sTW#58QD8(W0dgm%Q#UxHF6
ze><)P6)Dlv?hq@1#hgn1+@3_cTvp-XdrGfNlbGDSX6qdR$}Hi@oM%t0uvb9IGO(Ub
zSZ^v?A@JeP59yAR8!q5#^iMb1ca-*9G`=WP)9SPMtX=?>95HOLv$*E{g2ZTk!M*zn
zEOgSS7TL7re+Q04&WPS+9nMvdaw&=-)ZtW&kyJH@i}OEfSV6FG;;a634d`casI_?f
z<amV8^j{e3wjJZbJS>N98e%YGO{;aXcd9!>p^?sPQN2(u%X6v4g&<cZta;GFDvC7J
z>K<_r@x+XRUkeCSB4C0Gf-V(SI5WV|bOC3`m^Mu6d`p*+yqQ7mUjd!t=Q95ITPj(}
z$;DnSDaoyWP@QLu%x4M>p%zqU4Qe)l$+xR?b?&~#`rbVe)~NlIgXK%=Q%qydHA%KH
zZCST3mO&M}K4C1~#`NmO>7HU(Ts_%2qtNRJKVD;X;m}G)!{^}Jen)GddDaX?*-6@+
z2~zVAt+&LmUUR3(I5NNG(1V<l^pm02ETe`~_6pbqKdmpii9UYa!oY2}+C)oZX>Cmy
z@=3K0<3IPO%?=jVr<%1-td*HB3J|}~mr0@-h@;2|M?I(wcy>!#6i^LR<d#f)%r(Wk
z4k&k;^zJq+^kK#ef6Z>E*1&fF>o3!Qk#5zVs)J(XS7<K0SV6vuVgDL#9Cs<d5l%i#
zeulgtdncm66jY$VC5LA`#hp#3W87|%Dr3Z4ogZp$6foMOzKsa|>|7;1B{wH}B|W7#
z536-c+XV{K_&RN}cB8(0>Uvt(|0r67-HQW4`2^#YN=yjuKRNQQ9i4tbyh-ZJn@5U(
z9l6E$56v(Kuf)98lOr@ok5Avb!r>Y#+v}xD+R$wGGC_;mEH<w0-ZjfjRLjy7b#gQ+
zGtMSIY5GG}kyIRT<a8dgHt?#(A`-Rs&bZmsove(`#P}oK;C{O1UvMQ#lxpx>jVg8T
zxvU!Y)ZAVfj+-K#@R`WT8y%BI&gbm1m|tmC$MQI)_TFu);>UC%$K1id`<_r(r9#NC
zh0Xssu$cl;=us=MV3x~fS${5ln${-}l$mN_KEMJWyJR4HB!qc{hpoBLNrEay7E(Hl
zZ%Q-Bnx(^X%@O1vkUb5aW$p_ye{e-1F=CONVkeXU$>ZI8rLS6{7Q;-FBLDgi!b9rL
z=Xkh+LmCxPe+thJ&5py4Gx57EjzYzBu-JiFN?L3`x(yOZU%u$7H#@o6_sV2BI)lQ-
zDp=Z-udcZF6_&_ka7TgshXpFx?H3xI$$T&0#|LadpI&2p-el#<vrp0z*zaJ#iVL;4
zP*_7^!~Srj`>rVV)6i_)H#9=>T-cYQnKhTiCL^c>Eq1U&HZ{bI!4lV7c#1C~$ygIu
z%ehhi_G$`q?gS=rF0amUI+SHZeu9x|q~t=vzxFYyUFzt_t0`ZvlW=>6lA+#j`iP3j
zbdY<?*iPCxcT_sWo0IiXNloYv=SMtjy=A_Z1?>#fvT77mzMBA%nnd2s^RD|)oKkpm
ziIb8$4~pj9FB5eKjLBxXyM--Z2QU6Hs>44XGQ=GIelc^gm*Vbl%G-p1S*h3o6-p`c
zli1p5T%)}<G#k?_j#w6!^X1FY^VRtB&G-cU-eq-8Owu8yIk8gOmP`pz=vGRI5Xs-@
zLkv)48J!B;yCCA3#n6!-3~*rzk^Z?{XTxOg$Jc$1)Ltj@xpXj-D~Opn!AJj*3zv1_
z)j5BpXH0@MaBgIZ^17LJP($eB{&c#s>xRLoy|2VwpGYr?uzrqmqcTrO6^i#2pgh6B
z-?|t;-JklSe&u$3ck<gb>gsamPc*)>b>e9#Rmx&4YkeekFkx&5I-$HlMb;YaF$|HS
zHEXI-_3?7!F89*~3V*_HFmIodg{N2Cd8hPaT`E^QS!e(9sw@D9<;V@dbrc;lx89^L
zlnkWd^w{>@T2484CJ#7GalZ@`)LCgd4T-`wdNF@vlFjuoQ7EzBffLc)$DN+yHqEU5
zkIr)#>By?(6rX%W@9O-{_=flvOFYhU_Vj?Nt!uebl^@Rm8dT7vX2>#3m2F|ebbLgk
zwVxZ;G1o*-w4&_ilvYLJlwyq$kpykG)!UBY1t~#Gf$)9<A8!m)@9zv(4CD?|$IR|-
zzs`8>>Sfjf7%uW{#B`|eiRck1N^`S|ij;?60euRcEO;G%*^qi%N@+nv^91m6WOeF_
zw7tW=KPScQy(Y=~pbUAtl9F9YF~_3EZ<K!M8G8T#TcB#KUY}=696xdfDx6CWHs}=T
zP1!$2aEZi5RjyY2#K9bMqmH_fmf0`Y)PNsBnrwTkP!R7N-z?d}^gNV<uwG!L>X2fl
z6YJ};?zoz~&Nxq0?lF1J^PtB0S)@y*E?p2}w*XK%`-x|^I<}bT37ak(`c*2fQ+v0$
zGdt@64CMYf)O3tQX(t%Ej|bv6B6#wSw^BKJHP*fX3Q~NyOS|7<5#k<85+M?mp(wbf
zckR&L)SD%}bV|C~!O}wrw<+4~6H+Sq?rHbd2D;b0<dSeV-jnHlNfeXHM9;wJq#3qC
z@6H{q_PY+TV^~<3(`tIBmf;3a0Z>th!xw8(lT)7Mt=Qsm;EldwRSTT@g$^NZ>_tOG
za^#AMP<vw1qd#Un3B<qu-D88ge1NYK&wMkY<HY4DB42N=r><b=4jo7fIDLuppP{8N
z#orEmyjb;tW<k`4tI%$=>GWNFJ-+dj8kEqMXV5)CTIVOeG(0=&b!#B~;rPV)Y=@wY
zx3fHR*c0zF2r@dsU=ycm12&;~I+$K@{${+V)U`C$Hr;O8KyufP>cY;`kMSsZn?OJ?
z*SLUphuXbmqqB@_9+--8>LMBce#wM+-AOC596QRrh)=22`|?UcWKsDEUbJ5EHt-DY
zPR?=mgkp|Peo)^1H9rI-gVlMLzlyQ0eU5+U)VhKY45y^`1M!}gK3J-Gtsj*hNsq3o
z7tA;(g}!e9UA|$*t%`Ta9}``E6~sT}|Az-6B?$%B;j!2+<4+fkqq?yT_MZyl>HM?I
zp1UNoN0hvsjNH$gVL=q4o!mfam&Ovmjn5z1v{~b)BM%V==(z2lvGn|QJ10-%&MVA#
zxE)f_A$j9%WqX^{3H;9X1mZL7D_3PNx)fR-L%ufg?Db|&zlMrKVq77oL#d{^3CQ=s
zeuqpZqBzfTSNB|I|GBG_#&^&m@!e3ReRVLg5u-YMfIuJ*4~~9I82@E0?j~Z?Iw+hy
zM6>;<lE%mH9;J@ucrUfy9pfi0|KusIROXM)L4$06_MhxODSuMx8}vRQcFLw>=)0v!
zd~hmszu5s<B@0Ih&IcDavja|b`*%s!$#sIxd!~B>S~DeVKLBC4VG1s`*bQ^P)xV8Z
zI|p#tYU`wetOZQwN{W?O?YnK25Tg4X?T62N?m*0D2q%oL@YO{>xB_)g6kG}?bHGy6
zh&bguf06``BCtD8ez>LEuLGQ-DY#f^w`Sqa;(Y$1p38YGPwQ775a{5;7Nm~>KGqm6
z>22Ks;>eI2cwEHDUgiQE-)-#<l*T<;Cvg*E(@<16`WFUb)eXVeRZ_#!>BybV=lv(R
zCd&-Sj{4mI)YJf;=r)z+s!ct2#kb@f#8mPYU{Y$@Dxgu0Sb^lu;LlEmbKr@;;VEk+
z0jN8C$kEf7V65eG)YfiZwBZPz{BY@$88~Em;gt8x4InXbH-H=EN|lt=UZ}g@os|OG
zcMAS@93KxnJVYqIXGyg)D2jRnZkKSm#^&IVaEYf$GO#Hcb-+?j!-12UNEYB}Eu|Cn
z=YNO#6Rq2cLw2r^hOL0e&alw!EDs*!$6EuQ02~D`u7Ea)c&{61jUwyL@-0JdwzTfy
zGLBH#IZ#5GX=LUDA!$u1_?}j#NER;J{mJ(%)YMTWa`M5NSN(bbsA>)?BwHQ_!`hQT
zTTlvw%eg~^i}!FTUhwHi{gI@)!^7Q$aash6(n!Sgst-a#*+T0T0I!%sCDQX==D6Dc
zuTL9qw5Gk<2sK<qTnheTEY3g%nw$tm@>2j+#X%*wN&Ot&dH@Lnv93VH)d?4c2V0#q
zJHw$B<$#5306+j=?v?X9HXk_^0TzV#17l(QCc@U$^+W@Z@RTN#9?ezTI*T>(Yh~v`
zS8cinfZCgvap&L|%hmC^>~t&e>Q@;1+GEmhBk}V;f0Lhvz(o<a)EOP%uZ?)uOjwV1
zcLP^rQg<)HHP^d91k;Bjczd|Hk>xOa0J1+K>wpBfAMV!Nj1&Mkvn}P@)SA;RsR6)Q
zPtpX61epuTl~vH!!C`_ky0WQ>XKbI*u|}x^hHdFnWmQUEFujhy-diQrXKq&knp1a)
zM5&~wzDL++=V$_lp27M)4)EnFnVeM5cg+eEzz(AUdXseta=3<Nq33Duu2PuOX5n#f
zQ}DElGne`TYQb;c&pmFLzDVw%r-fEOz!mh*_7wBV^OSh!gXaG_mw%qqMvK~8&T<wJ
zLPtlSWKBTVK{Awv8Fv5Ng{RPNk=vWKGmGm=^8r<_5nQDV5B>Rj$GaLXy37kl^{~Zo
z%ef8!`HFKZcPQI5SnVX|4%4O88U40{$+18?rJtY>ya6;-ZDs&s_r=s%r*}a046Gyx
zb*3gKAcyM)&d{WfBqG=Q@a5Wg3s*=)l2XoDUc7hrdlLa9<yNiwaFziKAl80nM)(yM
zrxErB+5P~`8wNNKpQHV~xUK+1-IbRz)KI^^y4Z}+PwT)H?9X5q`p~OSzLDLV@a3x<
zhKzce74$LMg2OLrU<9~?xsF4sy^qwMWpRwh18JH>uwT&8yN$^+fwLUBjFBXFIQ#)M
z%CNjr*;8pKPgz%4MzWAcm@hXyE%cb}6VH%b62VEv#BH{h;nB=i?F4uF#c_k3PXun^
z9NRhJ^g=2V==bR=3}|Q7Dx5bQ|0-N%37fVq<dM(l@1;w9=bVH~M|uGS%y498DO@Fq
zB7aqp%&kaIeuz^<Pp~J{=3@r>*1r`JlNjqKdK`bb8EM|dIn?cF4jGjPIpn#?)HpcR
zU4Z_gZ(r281AI`Tg>pgxjtd<L!NlV`K=Q8I(YYDQ(_7S6Tgb4~Z=uZxaz5OVi~{$a
z;eZ0cY&73S4=Prqqhl}5?d8b%^~m~=gG+-{c*2Z}UW4^3_r+iJce}=d9Ld4#$=QMG
z0w$a+m)5(H)^NmIEVSpi@HHtfWs^(EkV{FS^&Rj-!{|t3BD>7sDrw+Q=maJWBY~LL
z<hOBL)9VI8H&+Ah7C<&u2ajxTcW@22)V>~$-UW|K&SaXh5ZH7lF#pM)cotnPPP?%G
zISK*hfM%C)Ga+~=e4NVkIw|4|kYVq^aVkDbtK2hychrk!5`jt{{(9aW08DxrZEcR#
zYjN#m4LFlI{{mr^%{uf;h;iQYOXQ41AVQZBt~{a`UJG$waU%0>Z+jrE9BOi5pRYyI
z5jf<cV~mq~KCOL!9Nn3xeY;G!9vOnVFW@-+jo<YKpqJP~8^8%kykOrFkq;6{@zI)X
zmp$S1^!0Y86?KpuVnK~X3N{f1GA$KinH(Iqq%b^JsV{3^I;NssXFZ1<JhWfERArst
z?@b-XHyp~nj1K2XrRHbm3&B%_OsGwfOtVZ?dRnn_@dcAR{VPurpsl@9OcLacU6MoX
z$E0*)lUN+RtW-pFb1{*GW=Cy=Ty%{@rkQgwJa(UN_kj@sUy^PyHWGtm?+wD6EsE(+
zToN!}_3tlmJ@$Zmf8_L!oK+0TeqVswx>%#h!6peeag)i06w+P|bSX`H5$WVxKlW4n
zjm*Ytz8q@OQy|+oIR>yv67xzqqF7ujl8R05aJqC3sqK|FluXxeWe?fu1~O)b2EMj<
zRZ&j6{1)Cs`<8L9<TNNHX5USpB}TMLe<yfty2^itbiZG{!)(*6clul4Bz7QO<;#Pe
zp8D!5SFY=z)|oEYx(_it$|$6sj7NS~5<h~+w&tBz?l!V3ck**qx?VA@!DzQ0^?5zn
z@E|ubYnvYhZ{$Z_dY^V}g^j3W{3(Cj&13i#Y?h#7YoL6c0tTU>?W0fKCXZ%Ag4ZQR
zvK2N9hqEN26I%)H+^)AbpSn7=ql+J*Fsw(q$`^9uvu^WWXX>EH2FqQd>c_)jTM~&J
z)60=R*CYQ%&?a-2Ie-WrQo*ap<Xf8VN$}-8wxe1Lu8z0HBl=M9(%h5~Ok!9_L~&Lf
zS*vgM7qm&I;KuW<tp3b;ZQ!n;UPzcVf?poI+hT0DTK#@i!9V(Q_aw)@?gtg=X>@aA
z^4q6bY}QBSkxU8I2;Cee9=Oo8@3z^Og|c3%j#ZPBAnoD(m}YLDI>k8tI3Bujpu%79
z@nXW)aEx=)6)ZSV8xS!elYzmnO=|tq4wWlao;R%5Ij3Jvyj~?7(XH%G)a~0{sZ)m*
zHwb`Op!J}WU;*XW!3%?1yLrbAadH;^6=={p=1H&{t*h@pXE$2=`YqgDa?RD%-pI`|
zm73W3^lnFwhbUeEC|Qp*j<k*v>ap!nuU}kUdEMI7r}PROM})Ojc4IhSsgs7qR_OFL
zt4<M8!RnhTeTu=Iu=}i6?a7)x`Ay8w!8g!R=j~@1#I^Oxsp9a#-!d|5i-|eXVO+YY
zH?i=M9SbqcJGih(yuwH}31j1j1<4M}5F5In<MdvvW-!mSpRcAXoM0Rfd9%^^=KEUD
zfx}Z*mNWQ>rMiizm2h!AZRj}0sU=a2$gqLDN%<U@2ROS$^1gn+stDA2+VT!eyoT~*
z59b)RveR6GqhwDbv*^i(Eeo&YQF>1MmMsgdU)B#elIs5SRkz2U!8{Xyt=A3;rp)mJ
zwd1qs;wjdJUF61>?ja;;wu5=Hia8u7QrW{v8!b{Db(!85MUJ`~?F7nxr%*qJwHAzY
z-5onGrI}5IFeuk`;O4lARKs0w_4VshY}2fz^s*nycX7%C7fz%z%r<V5#&cT|qa3oc
zxXyoBhC^WqY@Sd^AE}((Qx`!eC9H2uJ2j`~*RzxRLj&;Xr{?~41#{x&w!YI_bSzdx
zeF=!chfS=t>~yClZz&^m1b^o|bP)}f1CFlf<RW9Qp1LuVtzge$=Nk^~B|>OTIDTxU
z0ivfcvj__2R=)~B?dDDn=|M!R(djUJL{J^ViH@5ME?-Y*a3FevEGRBwsTdD~ly-wp
zS`-tcO9nyJ=pI;$#LtJ>%;-S5JD^L00HwvE<o&wOD5KYmGK_)xgq9`<`kPN~<EJq`
zx)&bIWHm;$!wSXIOX3%w4>x4%BNcq%#%!XqR<^44Hs&P`G7Bfz%c<{??c9*9l4@go
z!6Kr}r}HFgMHZqO8FVXuJiP?^(>$muL|vv_rjaJmrje8bRas3@dt^jH(v`p?=iS&?
zgLh8(R|n7du-yk8sZvk9{WVrv+0-1m#M9}W)Nrj~$da#pFaj~E3+m_aOxDqG*|%Qk
zKFHHwgU{^Vl5TC53x+St#rG|;NEL@|%oCCXaeey^OLS;tkYW=YK1+w<Np+DNwl;dm
z9?GC+0kgwbX|tk5oiZN*qE!L(ig5i)f$Cmz$DrSRDvNQ>@ysD2@6f(`yV=%o4l=VS
zk6{p`e6XM(dhuIOf{eD8KJRphQjFFU^42~7mj`t#d$TV$YFAOBqKu?~**5BR%&qQQ
z-IB0c8g0r(;S`BN3Kv>-8z5DFc5zDteM!W$b-?b3$UOSybSMW??erK(<m-MGCEyl@
zt{8BiDdozemFImKx!82zkm}$@Yo>A((Uc)hlYMt)#H&XOjU6>Uy=nSojdAe9tzG`-
zjDgT@I+R}P?iR;}SY|7K*iG>jdH78XYd_}IK<Zx_CJp^#ZmHt=0_xx=_PHE4IX7%3
z?8Q@3MA2@WqU77Bsu?%*q;=zWc6w^`2u^D7YLWsO^^xc3!j2yVy(_t#^v4u05L6Zn
zuO;24%4wWiQ6&x?`0o1gCs>oEQ_*hLKinNeKkn~t-d_8ZxRrA7PTPYG0rPgNf$G#D
zrqeIH4Y7rN3Pe)|-|94dsu0oUZir!i(0YnUv6LS0m0&F^Kz?0D0L}}H@pYkxw(Cr;
zsA)`?-U=n(!>yip&jao67}C9~i)j+YZ&Vv>=<0sdRnw>CqkGtG!B*aMQ|Md2MF^HE
z=8YJbWSS*$3wKs)?b1sI82e)B*Ag0eJWC6>TPcBBA#WQWn86oR$kSvDP}ToiC0LB^
zK|Pc+oD_!D=e0P_sOvb1;V~)bS`n>!n@k|~Q&+UX-wnFAYw{R!vu8X$R!D4#NKvl5
zYdtwLccR^!r{irLim|Jvidf<6OH2icGTmmJ*d7j05gCoZBa%gK4~C%geOvpDG#$h`
z)U*1m*@uR4)Cr%VbD4(s*UKanXjiXh%y)+*uG%GRwo60y;HS6f3$!?Ia}8Ej-5)4a
zSL+tjou%Kc(aBbYUJxy7w|eZf33FMuM9ONh7U-Ddyr-CCqnGi^NnB+!IeQLO{Zk9!
z4W&#N6w>8m$P8hP@o*Vd$<DCMazS+{TER_FHSeHdi`;ws?~P=&Frj6v;7*HZ$0*Ip
zIZ_Z3H5fJgT`1}|X&>eu|B<)ih}5@GCH}Io@BmiDj~Dultu|V8zw<qwbW!*t;P0XA
z378r~FyDk&{`_m`f=eaGCGxDtUwBKCk5<S<_n2G+&lPDJHf{0NEWqS#LSx3<B9HIX
zp5>UW2Z>|U^s4<rTey}4PWQ!8_x0-l<mhJRo-o_6TdIX0vK~+8$3ME8m(_mCJjl|%
z=3sjC+fpPs;8v0=0P{-yl=;{E`vxUDUf+=cOQpol4Z4MnP6eicRb1B-nx8eY^NbW-
zA{tb<&{)<w?Q`|lq;0H6qmt9CUw?bA(pmp@@3W@+Z7KWg<;E!KCi2!IZry=HYR-s5
z&%Wq~HeQsZt{Hrmp81Rh9Gq23Ct03s-XnrxsX~mf>!>bfR*xqV^9l>ry|0h4w<PG6
zn*<)PZHW8msuuov-eIM5IMSEf{8T#S|IbE*HQbHgjSg+v()~^d6~7sho73isN-P@u
zZ%u%MTun*%_ZZ@@8OzoKMx<m(9@|9jw8O+cZ)$^<DC66-@t#oT2Y(+;#HpQys?c6b
z;RZRWR`yq}0{~puy_p)13uF0PmLRVpCB<L4$c?vRZ-RJ=281pRdfS^3ugwj;&?Na@
zl&VrTkjNjYnIXY*xt1^z>pxp3sQCU=;EI}5e~EV5gtrV&>^>J)-2AmWCPQp44*8={
zadaLain8D}eW)7NA#C;iwnvprE|E#Gg;CIPB1G{8_n9a`UaCs|u;Oddx^3j6>>r`k
z-5n;IVuFsJdZV>Y(jG;%S!SJRt%T3_JdEQ@2-MnlfDMv1V(F|!@JX<me-~LO0V;&v
zable(j?`Nr6#U+XIZX;eqkX*v-)NMHrx;WDMbex4jVK5e%JU^X^E>CyFr`&ng`FmU
zr<oyOxbm;Zib2f6K5Oy}-7{7*B7cV<kPk}z(}(%@x%^Mzg@-7jgN|qQLvtO<9a}ir
z=i)hx*QzT21o40TH=9opE+%^e_j)nxUuoMN+{ChccxJso$50CV0+cq7b=JvlBxVP+
z7uB>ce_o|#-LBzx)T{D>r%XAhnvG<IdS<`4gAjS1dM>IIGG_2?CC~P`97(SSsZe3d
zUiZv0ep{b;Q64e|kI>|<L2K#_xG_&YYr}|T;ro0O9EnY3=_>qwW@CEpl$4yE?%J@P
zT4cS3!$5#O;Z1Sj9Ahj?Sy0)Hn8SLI=O=+3tyocq!QbrGU*&@(yZ1A`2skEQ;Ye<k
z9jS9g`D`FhG)5km04j@7eP{7eofb}G1Z!uDMS48@Ht&1G`p(Z{R(9Lx(e<Ypy`-!L
z^_a@*$iBS+0Z04`e;H0^FbZR#TvQK|V(D!t+RuN!A<l1`>yJ*fB~m?S*UAT9x8DDc
zuUpoucQ$^*Uvf6c9G(TrrJ1Gm8)s)U^L3r#Qo2fF_D5rav?;J0u>4N6Mw)o*8>yzH
zSPMn314MrvFX8+s0q{!p6U__ug=cPJj}vaAs|Xh*W~MD`c&HX(g#WIRBi~oCmQ6{&
z*0NYo2NIZw_e|6pGG%^L*RPS)a5&`ttWwAsc1XA-T#Gy2Lv9>4?q)tuQ)#nmFR6dO
z;0AE4A{Q>j^3|hfmS7E?p`E3G_X6_aITil@TpVBlf#!=RrO+b0B^mHsJBJs!Sm}u`
zyx;fVHRL4j(bW#ZRq1IP`oTl4x6b!93WC}|-IWm_wQ5Tfebq?dtOXcR_KU-WD#@bV
z8PGyR{|Fse0HLqK!_&Zm$8q5_d>T>`DaI4e=lB&b6r_8_m2`ZrQ+D8?P!##j1AzQq
zG+Q-j2K&YJ<_S&P;4RrRJL7&HAm>6=16+lY(vNS-t79#@;;W+fdn-PX;%jwt96x#<
z+iF)h8xf}Q-aY(Tbp)q;Js6Lo5}&}~we85(yt;@Fz{T>YgrXIn_U<I|8t*aj7tq@0
z!8eL{A~ZXTX}7M9Xao?=F0F4K1QynbhY`2-Ku3eLSSJUE>BQ95(gEAc%h?4dv47!U
zt&*@rr8qbWU(~UkAH{H=Ucid+G3NHL`CI+XNN2l&L9~li(k}Z#kBA8k$$^nGDq1Gi
z_t(*GlmzgQIx_1zX+c-x(6`!7(F@8vIJDw9qJ@FFnJ2lezFM}ap|+z+0=e5_V|*wN
z3{FaDQBr7u4dCFhdw!;+)zH(8ww14La+71J0;p=eb@N;(xCnU7Hud{b$c^w$l+8i5
z%zz4iPA=4LG`5ZNbR<bd24Dj%3(0(T;4i=5J6k*r`)n*Hb37$L=E+wB=GrF}%axy>
zZ@(np>;`SYq$_4YVMHI+dqA9Jg@=TefLyXSGXS4o0@?@sY3^@tKAP=F(&T1h)}3<%
zsQhx134~#OrdZSjH1DmCKYs&jFt3f}M69za&U0|*g%C14_Yg$BKLAZbwfe+#Z3*qW
za8aB90e;6Nt<d*3Pn6CAH&t>f7-`&F^O(p|lEiV<6eA}DPZ#*V8t=>Srs1mothP-h
zD(LrZETUx1REIdP(fq$HTxL8Sl5zO3#I$A&G~qXj&)%q47oQy>pgkY_T2sO-@LTf#
zc4nfYBb1j50ugd=7wS5+v-mZMdJV{2O)3Wwm*Uc{!{Z4Fh5*!^1?T3&DeW`hEv+xW
zHudvyxs_p_7F}H1c*_1&zl%uUCF;^mdR&zxSv;rdk7A{*cZ~U?r*20y=KME6lN#ro
z2D=b2-WM*sRNdRK%WR=j(7#bnUge~-YlU^0@lHr3(3qGFNhHA)r|U_2X24j<Uan$)
ztbGf>9^g1HD(>4i2HlEGibN}ueN(mqe!$3UKle53*b%GB+;C~0n~hW-`KwIVIz#8H
z-?Kn|w<%eiXq{&~Vs~L--0HihGn5COrbAkw7v2GDmSy0iOOp5osO9Nbt?ub8b?}VT
z2{!=Km<CKgJf2pVof-eGPsiz7Ghr7H%s2y@4x~0L5$gsbHlP{3+SSgXRBVR4iYEK5
z6pR^OG4BDeUU9~6=>Bn-t?TTbeW9V@=_^aepf^)31X|ygF3FtM4dhP(oNZG78AFus
zmCvANUZIcuo{@#Hcf%U?uljJ{czYixp9tOLaJVI}{ELSzZu@36zJtg;nTyG7k~<;t
zlaHCnt6b`G8!xshp+Dow*=Z~XUzNX<`xvd2ad_OIRPlLFneNg^6aJW19Ksm>_*yP+
zTS-&(M^x%*E_716A7Di?b-KIJ?B3YzfS*TB%^K*+2KI$H%d@gvK9VbR$QM${HpA$m
zW@zKhhO?a$67Xi+;D&=ChZIfYnPHyvK-p0><MokYIcKdGq*s<B4thbKFIvAW?vecn
zxenQz+hQEzN)uc(s5YTHIZzG})m-y-@_OardL(k{XTtFoccypPsAfW6KhOC(T{id8
zTz8f}XXwqAvGUeLI_ID=?-cCXp+1G`+UPVfOE&SDtI37`^O=WXIp&>-bgS#TyR$-2
zEk<t!A6)8zK+vDMxTv_x!la8@6=?$-K00Xw^1BL(V|lq}_6??%XJqQHO8L1~urHK}
z&ksf&E>LjRlkQW)z5=~oFnTIZA`yNzGdS%<g^(L5dsxL){(UL^pB>^E;0cj{EjZEB
z8Y={{ZPT_C6l!rT+&8&oe6G&d{<pIJ?~Ng`2tsQ8WMbLdV*Go8oL2&+gbRdMkMdnl
zD!d_`H0D~H@2jJ+YCs6O-s9*1j*A3*Xsz_Vi1Jq!k#dljE-@KC#e)Hz4_~B60A1yS
z?bC0sE_P(mR=F;;h^)QS+r&O|A(|co4{15*ZuY~Qx%;~Gw5^oFA`G5Be}O{^?WnnK
z=ru|gO{Vh?(`sdV#I8uG3k#nB$F~%#NCwgisO9k!Ubj*_E<HNlW~)g^`Yyse6wziN
zhew!!{_z%tqb3-B$Sin3&=Fo^APaZ0NrBs=fTH-xiZV*i6PUH%Wp`_(b7c_UTCMh0
z)UJ}U1f@)o72_IjjEJqbL7Y#Q(}EXIV!NegH8%be=^S2)R2u6r*>$pz-U?UPLO+|P
z#jOadm8I_Qfsi0`90~#HB)+OGj0-EneM3&$>@A5d3>tc?)LNPxg~E?9TNNrLDo!kS
z_IXe~ZDSsAF5hh<o94u2DV^oYGD*p4IgMc~JQ|xFQN&9*UV0KnIbC&8GHu3z=RP~4
z5`KG0u*VT3Xj+QDO}IeIF@54Pw=%CuT>$74GovzgZ=WV+m-aJdW=x#dzT-JwXs4AF
z^cU=OTgBabzPAYnwIF}KMhcJS9(X^-J*p-^aaSDU<4ZaqaAOuur1nI}{rf#6eY}_k
z;-w&X*Q;oJK3ynvC8FNjjB@=EPq|C`x{K*DC0Uf@-OanX@m%f5=Ix7Uw>oE?;5MpO
zKOPj3BRDh3;2!QMN55&sra!08-BucNs#&nDa0(OuzZiS#s3_OAeV7hGDJ2Dl92Jo6
zZV(tmC6q1!C8WENZcw@rq|-rCB!-mkAta<5q`~i+y`TNOy?)=Hv)9_ooBO`vJdg8;
zhq0Fs;{<3TM=z!C?zGUww-)gRREh=L?c8EmB2ICHj8P~tG5Whu9u<vy|1L`YUb99f
z4ErFPyX(<X^$XK}bL=;Q317_jwo?C7^%B(jlfkc!sL5~Tea{qMT+AeWH@drVu7f`|
zdAH%4`<E^zi=9C|TuOeb3bHSjh+;<MbG4T{+tBAtCV)N_vL}VX9^Y^)X42=Cw#Ytp
z(4+(wr$r0PyGeWV>iZQ;s)@3@x%AQ6^4;V=7kV$r=9Q9Lx0zRzOSr=Kc(^xp@8B}j
zD6z9i=!!vU0pCrBiI%u#6@-&RV!n(p9n#bG0`x0^CRqw$3L|wQpW{gIsZuqTzM2B$
z>*c$gfo>5+j37lI-25)BOwwUMw3L@}Kvy>y4l-p3Tx*TnY+`O=hhm%`_Me6+a%UaV
zE5;O2G{jZEgFNz6WMru&2>D9V)ihS7eO~Fw;`wU<6fqf}MQJtr6v2u}Ei}@Wuh>My
zpD}M}LWkaVYpf`W0vD&30im<IDF%De`+>DFLlaMv4r!Tx*;Vc4cr%vAnnN7MSu~w<
zcSU$Vm4M~+Mr6cwlrxiBnMOp#Q&IE@%>V{LbYK3P`wAsq`66O=h8=nA28G{+eVf_!
zo%NV(u1|P}FAXmF&xJ<!g7>#{g>b{j<AFGpfKbWvk?zg~|M~ufa6$Q^Xl)NM+VlJC
zB}`Kx&n52tRWR(Aj3Jv0>#|+l1`CMP=tFK;e}(%9W(GwV)J7~+6q9xCJ3Yn<#z!YX
z3wc;~SNlI7SLpxi57S6A<UrLuZiCJ#Ml~KEJTB#yVFeoOVKg3A-4iuFl$fXO=)M8*
zL3+Z17Mz})bb70UOr0>7&W!(9%w`Lzp6p>^P2y*!a-G=y%I(ZdWycnN{y3t@DVOlW
zVuMy4j8J>?o-V_XB!;zxswC#2t{`3H<)0BYyW*_#TjV07sKxw38l1(C);IM3*C+g^
z1pjNbk8)$Uk(x46st2OXAfU{P*0I&6AZ4It&UHZ?P79Ex1yLmpnwqbwseBK`+S7qM
zm9%Qvg)!Wr14&{4-+vTn!2zGp13pZ}IAy0@c<5sGz){Pms*FEtCEdXRXh~1+VQ$D7
zR^A>{ao*~XVu8hxVC4C=qO#3n#12C<z`)Z5_*qSZ$YwKZ4)K{>pH)ky(tXe@BrD;C
zq6x)0=vw{v9s1X37lQTgD+Zlm2*;ns_A3MYqnd7%_@ApI7chjSo}U0w4Vb2V(|}Bn
zuZGv4Nr7QqLO{atEn^!s5t%B!<3Og{PW!GET>uQQ$A9<a;H?2d1FzU!a09Wx152o_
z;A2go$Vd%%39MJ%%qi47e3vm6jb^?A&;&c{v|ikgFJDvX0$Jwe#Yd-H&MBPDPYuo{
z*`$Ttc>dJHLclR$2c?Ax!a|STDcoz^lR!}@>`+V$4jb!b%s-+}C2K-M2mb1eeltKY
zjJ$OoG5^0mnru|-55Wa%;Kmhz;)g^z<=gNdM56Ar86fc={aP3}EbAwR3a&6m*?3XY
z`zQC%yB`3DrHK|mNI<>?oDjry*t>mjOjE6NV*^7jMrLiRe>(i9q8M%$jR1V6SsdUS
zRjsN-TRH~5Emo+}^|C<*?xw{=SV7xgD~*o{JTxzz2-$x-L;vR&;xZ-y7F!hCAm9ft
z{v{P2SO8x+0O7^F&dKI3u^TWrZb$0B90HwGNm=d469JI@oH#eZ|DRq)BBLKv|AJEa
zHJC%D0Rxf1GeHnhp5OQ7PjS{u(fnIaQH95l6j=H=cSZiA)BI1X<i9ML|N94*7;52(
zxqnK&^}p}&|FVPruOF5W@M<t`{XIba>mjZo`(2VZ!>Am%sT)OIwBLc*^HdyIBOf){
zN-Ntey=4~fP+=5T1hco_{r|CZQU9)gG<eb}5=Qa=_+x5ie*zssEf@`Vxy-v;fVz5T
zMto+xImy7|VcR7j7Tp`H06DgBVE5|ud-r&sIR8r)7>P&a_k&Lj2jPCerHEtVt3WfI
zp3|`W32<5wWU>38sZ|9i5se$K>QIh%U#?HT{rLhc#cGk-Z`^m)eJ>L90axc>=dRHQ
zCnf0SXjsEmN#eH*pRI{7*uv<)z`Os~C-Dyp^uzNbgI0Wa_V-t(G{HiZXu#CB+5F{%
z^CQ%U4@Bb!AKmF1CP{=cy>S5?wiyr_Y!gE3q9ZBI*L(!JLnLZ)H#JuDd<Xb3Re|GZ
z{tO`(2M8z&C?dpPlT3F(!2g)hP^2{n{GLrGqVL>Y_f3Jpg(*`8N?i?js|J5lxXO|O
zwVO9JPXGf23C@fx1A~$)_J&4a+k!-y0~%|pg*fF(WPEM}et0j7H#?rIi<=A1IX&+Z
zK+6C|jz4Swy<|t=jqLd>=y_?Km%z~~ns`vbG%^zQ`tB#PA8zZa|E*sBA4`sv9D@-J
zR(hrN=5GYWNQnd``LpRwe0T!bUr4}=vOuw*Y(PP7okAU|=r+^{av`m2@kvpnr5(^C
z;~&y@d?^m-f&)5mC!hm-n*zrY6*B@n?Uy%|vPU;{EkMEuEX=pTiD0wBz48r^9V{(J
zhV0}(7STM!r{MJ9Q9Xd9kSOAdQ&FK|6YI0rH^4IU^w?lL<e_Km7f21G<|so(@=SMp
zvCf3PAwT29Vs<eM<p$>h><n>p_`Sn-A{ji^w$1=h!zT9mP_+5hFRJ)u;KaF5=Q^~3
zPP5U$R{tQw4V!=#|9`xM|GJA?JAmM-H7sTC$(O&YGD0d0;Px1aVkMmblHE+S*rAdr
z=*T0>>~-;1@__B;9N;)fP6ZG=p4zN>RW1oyap@`3h*9MP)^XC;x6Duk4-^wcfCS%<
zSRACu2ldEZc+YJaSO99C?gCaTO5Kg3x#5|L*_Wh7%b_L0k~KRW__j(6oJuF1P_tzM
z89}TKW)x(MqEkq7sZH)*)q$`l>P)mkZFg|O*#LS_HHvHn2Tt{>6+PR@kHOzetPC~y
zE~EH=og@Ej8!m3ZIellc+W3qE>#w&V8{P_viPtC(s%!8i>cCA}X#NN+g8KjvJ4_}<
z(^bd~rU6T5c85bF%Ru=V|L57ySQ`AajA$+K9GX>LEcplG#Z0u$LEJ45j!%CB(pmrf
zae<(k8-@{;$4FS1+AWSzeEolp_E^QA;Owqs@3Ve?PpjK-qKcT(18YBepluqPTnBO!
zEFmqMi8jtJo9Q3k@?T2jA)x#DIrKV%@$ui!4Jd!M_<n8*GXTo7nW?n``D91@*=;m4
zz*NW_&=!4{W<d6(C18@gKj;$|!<`3NsrF4Bq$*Q@8%gK+<E_D{n<OpDh09n^ZSCqY
zASIoE3_)SvqGj&+uV<_WW{qR0R8=@IY(fIg-+kl)oCz&Hc;RYQ@C1-QkigkZ_i{Ar
z#thgs#{ZD~bG1nO26)uLt8KW>2T*QbI5=|gdj~sBQq|1;%%}`g4nRT#p&mY-44&6u
z_gGt}5HKX}NL7HqWY5DMD3fn1hL|AmZk-Zn(9fS1NcHjtg%8sl=eH{dBmCjl0Dbua
z4t<M`E$g=)Klmqg4*$B}Q2fHle8&wIxHE#uob-6h#?Dh$gyVYWLw}n<thHxGQp)Sp
zhj0S<E@Tg=&WJ7nsg#<`1QqST^y8-|pe()jIe3iZN|11Brq}>;o60T~Xee&x7h9k@
ze~{-3D2I6xM$gQn_$Tl<^H<yraMYloy5D|}L$<IsHnl2>3j!Suz)7B={_0U(>^?X`
z45v{k*Uq=?-NFA$kM$eqe#A0={OOx46w_s{TmUCc&f5*z%~zYVP1zx;+Zr)mD@^p<
z@N`9@h=8>hy;g-hS@U_<5RNf>x9SYQl^ZR~xuj*wXubT&>2nZ)BY`5aV%nAGkkv}Z
z5)9!t@s9-e128OVO&G%2Gs?U|Wurwp!+gPZeqs*1PGNP+uC-19H$$|pvn0r_3Q!g>
zUEr$44Tb${efir1zc$W@n{`j7?K>y~bEimrn};@AAS`0$;3c>egYo^rDb3Z}0U{!-
z0gp!i*Jtq7=G45$K{#4op4z8F#lk!^M0NT1*Y*IQ>N>;3LzxRsgLp?C+MJ^Nzh=H*
zv#m$Af+NHxNzXyKJ^JJPltq@e!qm!$`1Ktq#&tAy2=AA0XGzpmO!US>@n7e8s}RFb
zH~=lq!I1b)nrXB!<7?HHEU*j!?=yr-SBCW*m{Z!(Ukrt8qiBo)D_zWj9qg<}?h<i@
z5MS;T6zOvscvMLQ5ZiY%giHbN*VKU94tTm>c0qw1^)}-UaQ)qz1{o7}#LBCHNIF$b
zia|q0iW0oVib70lXE2g?^uK{2wBjs-bw4N!<I2Hv3oH?Q4?33B>Lp<2j;u2Z>Icj<
zI0&JdLS<5}Z7hI2Zixyex(2Sa1SYgPDD}?9$>-}6nG+G9)!YVz$Qsl|_!o|TxB^}*
zAqJh0Ot#-u4XV!}&;OzdENmZ&ZsaEf41M=hj&Y-JW5()fKAGv82P|O0<^fOC33$6%
z1~TL)yuj9r0wlSzp<;k>X(I|8y&_`xGh9_(D|1ROq2n$^W-r{&Th4&Xy|2Y6!W&sb
z(U9_cz(N$9SFCG&?<Dg?UK`To!q!}nN4e{^j<rJDOBswA3SYnDjPDj7LhK2d&ZR)z
zjx-u{d>Igv>j~|^QnLIK-R9!>JI8dgB*?+%gHo)ok&7t5N9m3{fNo}U&K7PrffrX-
z2A?So^HZ$Px6Rehkq`zogOZGFG%$8X8Sjy=ty7rCd5Ca&?$#HIgM34ciaF3?OS9{O
z!s#fcgB#?0y`*w)KI1Zz+4N}p<-T-X72{ocHZ_zU^B#tV69k))-NZaRMsY{S?hzkC
z3}grupzCj?+I-eT7pwq(uH;uwPHhD)z1KUjz};Ajbp=w^LrcRm9A{w!htjk=)$I(3
z*XAs0Qw`Vox1wrj!A@~{Tv~#EU7wUhSeEn~oN5^&<8<yTBuMadJHWMXhw{Uis|o>z
zXiGM3kklP?avWp34f%b6o`CRS^5Z{XTEUOFH^p3hi+~p8k28%T5LOCT8HK$Mh~Fiq
z>}v>?Pg@aRRzkF}haw85D)fMd3krUtCg@{LHM%q|&7eRyf_BV!ihkrbpruakA0|_f
z8<}!;BeRcAo!PNAWzdl9Cm5^J%*!#Ya^<{T^w+DDT&!m+P)RBQzFjduHm=I1Pg~>S
zbg_kf8j#9fsLCEn6GBzx_J^+?{@o7xZ=m_}U_i*WOHcj)OI-?ultk4~TX#DVy)Fpi
zjW6L1SWsos?@Y20E8_FE5J{<BBMf>@cS>h2FY+)`N}m*KB@APX`0l?enjKE#<!1b)
zA>ambcitKDa?v>Q!$b#3j0`aK&?=pm090nhpsrhXOTm-UO4yWZkX)(PL3lVRXiSiW
z9s;xX0JrSIVTj^O2NdtcusbrZ3__WbQAX1ITAZ?c5Lg}ufL!)7L%SqLFocQXu`=mV
zHm5jYM5g>$k1GS%yO}jK_S=rZ2|rT~%5x>CkkOO^t{Xj}8@;;U3a2qtItv1cKaJ&=
z1bLOgH;L{6atjboQrzy8Soz^Pml2K>miOO3O1cvzDR=c8i79i4XZ<mMN{CaK?(oFz
zDo~EkY9t@>{c+QNn>1R7Da%DQa7D){cf;CN@K96a?ys7_Z}A2BU7a2HRxDYN?^0vf
zDq42~?gc%@^IKOxgbP@TNWZ!$tH5~)Rno5W&}5+{Mms2w61N}IP<+pwb%ItckjW3}
zkf98=Bvd9IdC&mt8@(91ksV|?dd~#2+?K*Ld`PGF{w#cf{8<bmk}d=VTK7<iQ<fpC
z!w@GcEratjXp%dcSsQu<QSHMFtT{h3m<Wm3cb<*7<NDFdKo?(jrvSbLZ6Jivubo6K
z$erdrZ@b#Y+PIzsdMv5qV3*XZV7}YKVkG?sS1NxnWtp!jl-|1!#x|*l%Qi;exmj0~
zyhUn6cQibN-_~VnJTJ;{jZZGD^iN$$cnjlLUuUJmElPr9dAyq2k-G&r4<@6qgvDQQ
zH}y4{Y%+`~ElSK>S5+|{Eo&%ydAkHRpq0DCj63XslG7uS1oG#ejnuW5-+LK$i5&XA
zHu(Y3Pjy*)sug>Y=KAfFBo%*%NA8HMXOk?48*m|_8^FY@3nM!>sxs+a@b8z19rj}r
zT(s?;$jkFQ!WGvxqpG{6%s~R!o@fHH*4m7{ci!$2sW%NPT95ZzDIqBQ!wf8<IWL&L
z!Sws}G+wO-a_oR+jY!w0&^<0LXQLOwG_<GlE_O8k>v*Ie%__^k>&w<y$WMb)p1oUU
zwqw0503SePk$qfBbzS#-(utz1yA48Td7;+k>6ech6ewX6y^p~-sbASV>TTJb+17;m
zxG&>71zRoD50A#brjSkWWL?pdw5VTL*z)zx);wQjj_eml<8RJ_rL!lKC<=-dszez=
z$;~AEvxNOi?91GK27T?XWA2|NGd|E-p2fFI@qFLw<?zYgrk1E%`*2EGRNuQdD*d2|
zMu&+*V(P62QBy{3(z#2c+T>^$*?r!_LGk<MyUiQ`36H1W-C=VybzRM5;en<L?0yOy
z*V5%cf>_E?t1A(1ZpS<$1p{XLiing7x`A}8_{U$acjPO2F)lW>zfn&KS(R<OZ1A5N
zkne2+Nl1!a$m}tpx&PM$Ntl9F!{l!M8xh1SsF&`}15wyYPkJIR<KSFu?0JX!dLUUe
z4^$&c=_Y#<i)4pi5=R4yV4@vk0DM<(UZ``_5W(rvN8x%A$7;(}wklreCipK<r1`G6
zQOEBq?9E@G7GU))+L*%kAR&VK^6T%dbu`PHwcHYDW3p-a{cbg=#wUO})D`int_b)0
z59oWIQ+RxRR}_iR)2y#ao&90H!xj%EVlVi|_{%!*0}i6JPPfMJTNVk(s70hDJHpD!
zB+9n_`Pv9tbi>vmhkm`tboGxGu$c!26ZORFx6q;O<|s2`o&^UHM=(_HcQ27dgqLj#
z>w8y*gito}+MZy45^owmA4aK4F2HrQeN~xh3E`r1RufzfSHRM}Z)9rb^}I``rW1;D
zQSGVO=hYR_+A40OFpVe=(dXg>4#d8BvzIbI`^8%=S~$G5Y86)(hO3A_6QDaqyD0J8
z;He&qI9@qsdlO!yP*~CD?%rX<XpY3j<-Q3YU5B(zi2aLo1^OQ&C1NEU4m|yqA8)Rt
zf3{1FbD|UTgm}8qNUyGA#~Uc&w(zU0BK^&FjZ1$)A+L78;5e?_mJmT4?=}ANiX~v?
z=5jB~_?<1wZYTTEz2X#4v3^MTU2IHF1)FY3bLRG?Bj=1$l@kZMkQBQ(9qQYsa0)rZ
zl3L|qnO+TOu>_7ZuR4)Yj+uc!@%=Gr*~2~k`pI9K#o@qKaOJY<ntpMpk*QK@43AfH
zg&RA}M;Lr)UKo^0x(fEXWnZnz<t2k>ukL4Fr&6J*)}^zFg?|Qol;Z18YoD1IO8D<u
z@g91o*}!`YqnD+}@fj!I`Vi*AA{9j-Xfi+FQ_er}!ZuN;&x<^JGD;G2n;Jb9j?j*=
zpp2dK92inRj{IT^#mKfh5H)%BAnSaWn@g#4w~E_WIQsdw*RgnhEjya4NG;r+Y4L3V
zj%dm_H2r^flLL5`BNpkrMi@RVX%ga<szePBF@Ekpa06c#R#fC^tWmgkd&!&pw{e{#
zQL@^wqk2TjUPD}c*SCT#_D7jO3lQwVp#MC;iDw^g8-N;&Xm^wz7Y*ExjkF|*@>!LC
zh4rZqGqUw~Xo1eiEkxkW491ja&A-2Bdw>+EPHn74Pe6ITAF$J~RlYcqRz}u%53Xb^
zss$f6eJcTVak|_LU2B+giKUpkd~TiP01p2zbGdPs)@)mVnz9oauj`ogIff%rNO9^U
zjzf34xKsNRd~QCJ!)BT>*oJMIhlkS|yKkbv+MoNQuJ3X69NWR~54)=dB__J2aAS5d
zxV9M$5fo2gt_2WHFicOkEnq(_8??pjWzvP+dEqkWpdOiaLN-4@Oh{PAZRIOFR=h&h
zIS9DaiXhVZBHT#y>}Me0P{Vjn`7A-OvG#LHb7=Fr=Je(6_=v>CaLmcd<_#wOncsQw
z+SG0#yJ~q==L2)%>ve_V0#kP9fMcN?&cn!^SJS(nGQJkVADCMAZDt~=W9}5gh`$>g
znG!pjh>b{l_n**N|K4KWRWP6+nsP|y$eP_stJnTofV3u25tqB3eIoLK)6n=nlA!9E
zt>KmCR7)h!pzAcO<%zBi|6|Ps`nPi@Jq$<y>*Yojl-WYZ{*3kPT750LQNK=IB5ysf
zlmJb)Vur4D`v|Y|7S>|*Co$xNu=2c@44e0pZ~s0>?xQ1gtk-^Jj+=V@N)Y>}xwZyC
zV<AWVO8jb2=?FKkcO9tvavlyGJ5Mc*P+0Tnb7pe2yuQCyLs#`F&vPn<o77YsdfK&2
zz)>4pKe-=llS_%C`x%xY!vFZPt&>vsYoVLt;QsJLbtkhG=%>|6M<>jt#HUsh4mfRU
zJ#C^n9!<N%fBo{dqKbekQ*DW0JW{eloDvx&I8S-jaO&%*KCw#hcr@?Jp8)7V4w0O1
zJ4JrSSqDW<XB~x(k^hL@M40&ouE*CYwXDpav$D_E3pg|1-fAwtlA>?r;+*R73W>P-
z1n+90<m(OfEW6X3YS(fu109YrPtu>5ElE)W)7D?)t%=<5OmndVAjo_Sm+55fboo9N
z*NUdAq~R7~uSx1egjDvq`azjiofUKt&WR;(YLxj{^QKItqIit|TrM^X)#!|oSnsIR
zXGijqo0kS&k5Hr|*jl81eAF=sLu{SfV+-&gC?d(cNG!&T*E@llNbgNZ{+*w~a1R8m
zS74;Yt%$@$BpFaj@&>kE_!g{%93+WkiVRWMbe$YNyxqd}Xwzr&E!5~!-$@RU_XD`;
zZ|vNtx96C>3dN@0Pzu?}qP5mK&GYVWhF#ph**6#;kDafRo8Ltq;LunT|01V7$zXpX
zIcX<ckw?oB^)!j%XaU+>JKaL2K4tb$bnKU%TRRQmrRS!F5!v^Fkz$VG5%NOI(>$_7
z^<%v62I`V+u9nKu0@t~^ElSNBuV|s?Ld0~}^cUR9t--=r@Kw3bmv!?VTTY@gp(ssK
zl;eKk<3(x@<<xa1qrm+P{#?6+x*7q4(t~WnIl4WOn#Xf=LGa`j0c1H-q6*;0KHhwB
z9N-dnLe#7Swp_dIblI(AUe1vU*xlt9U*^m5${!+%CXaloy0;2B$D*rMeM^$KUr3Z|
zA_C6eIf+%FroJY>>YlOv3q*^Sn$LcZPMaeeMMB^}j@YziAwlDIYp|%ZHDvQHrx-y;
zh(?O-$`|vMI>kjxD0oFG7iN0AX}JH5^6DEXa)bQa>RfLJVh@@S{}Si=?k7F7URp;=
zSPP8reJG_V^9z-6aefo7a1$E<fSkm(gln^9o^yH&)kOmgu#``vteXr~()Tx~$@|2H
z>;&PHNAk7McH=G4DBOx9YtgK%hjNGlt8qdT%HU|W@~>@PG?`p80T0b+3{&QIpGSFu
z^Je+eth6MiezyT(<sKMd=_&mP$-sN^L;Vyc#MUDDQRI7_wvO?8E`hJE{U$FI(q&h3
z2ziX-dx?l{gM+5d;P;!q4w~=z1c)9N^FTKSqwP%j!2QcV6<2@$%&C&QJTlCaKI!eU
zPmbw&RY!(isehTUfBd0qb>7m{AtNk=+59@yiSru%rhUYFLrzBKUuOnaGec``DuCT3
zFIHo(b%M~PLwkOqm|sedj)z~7prwd8RlvD%Fli7ykzFX6C!9dU?ISmd93^6&{V?Eo
z5$S=rBu)>-G)2b?jU;*bLMIWhGzZq9T#uo2k(eNz2o`A8c^NsE!W=l((bfbQSB;c%
z={0_lVLq%q?wr${Zdu1t!DiemDDvRQ>B}iBdpzU#57>!dE~h@lOi9K&@IF!I6@z5u
zJM?G9b8RL#3-t_m82MphpZ5BkM*=d;3Nu`8G3vzeX;FN1&MzueCz8IG(G$Z!6_ZfF
z_|f@S7emPX5i?zgoh!hW)*W3EVBH2&A07}*iBb;Alm{9GKh9+m8x2RpO9_)#Uyt#W
zkGKjlinx6%faHXEzH{~2-Gy-E0LYhlkdSB<suX#-F$m2mRRy!hHW@`omN89WvQE<2
zCvDBm)I{<&KXwm*#p5W)c;FlonBSM7C*Yz=3*<kq0WU?NUVX379-F|ch>f!DSqmO!
z8usGT7!ugL%lFIO+w*kZ$HNao*m)2w*yoiGVspDiHzPJQb%t<_{Y55ST|&(+q8X$k
z<?9%VS1FEJo;`B*Hh}QNQ2@9rtUjr-oA)r#E89!z_4kizbNc-;UP78N!%3)^C>H>1
z%3sEHrzN_N@PY<2lx)&l^oS&@<k`TMrJ|zT=MaLaUtX>mmdFRZ>zpSqkCU9FeP%&@
z9#tiBV!z3FnA-UZXl}1UePqVrU&CQi>*8$`@KIbpLRoKh$<L~v8|%RQs#^C$IrpH{
zpH|nNFrVAb=+`mKwb<;Da$OsC8g{Z5%JCzu-kbS$UduEZ5|Ggx-`YzfxwBdp_OZfm
zGT+Sj^22Ro>OUUgp`l5kp}ZDdGhQay#<=Q6O5|0r<7HQ<(#T7OpWnNnKfWv~86wyn
zr7u*k0*!>TUU%TYu?XigdFY|YD{!*Bnp{~P*|WdLzFB%%5{-SS_HCZV2TKm^{*Bm+
zTYvYFMi>mSgs<e)g(oi`c|^`NQ|2Hd(3HepGT<(jXABJ^gsQ-FAO?cA<o%=WVU#ox
z0@?(Lbs_X5taeYj=U;l?RuOIG(hc2=g^i4@C+)6+X49;+L13-GN1l&Bn=f6uho!{o
z(!dg9Vm&}3)QW>C+%W)Ry!Z&;y8UT?7gQ&y)ol90;u#+1BLo5A&74w&*+-M)<~vUA
zzQjESrbT2%ek8<$)QLH^*wxxT=9{2>+1Svm_`7{keOz;RH;*;=Qmd`wVM+JFO_!XV
zK`$%8(JAn5cA*oHKY7G>(tED6<-&-@V=N^)0^8619(3y>A?v5rctuYL3x*Nh?SCOp
zt#Z;op_1py#8;Mr4DY#D-xIg%B;j9yfgxYhX*!rwa1ZP2@!`Ub=O|sFF)t%y7!J4(
z((PjUxZY#6xg~R>)BT}LlT9=lrZW4)hfidTJ17$?cF648){#dHzlA_fp;MN~&U|mY
z%PbL!)8q=q%G7fC)i+oXN04n|3i4#`mZ0=DGvPhi!$ogIUgS<+(<yiToqF7J5Wrw7
zx?L$tOz|B9iPeZc<jaz=aJ}o=-gqx_W$P}hUCMT$VLR4iI4+`V<)Im6(@^CZu>*62
z=xD&;-4KyVRb$O-j;&E&pvA5Q9N#9Z9OwpPR@Lp5PvuvwZ()aA%4e}3l9d#RMDTy%
zHD3W&2zqyPWfFa(Tv|p#0<AE~4gw`r)sDP1cVnI#;DZ->z>>*6?Av=!-~nqz6}R-Z
zoGXLqulGEWH{Kg6hPN;{MZ$d6T%Y(Ed4j1yP!u-N$3&9j0-b@w#S;&`Bm>Q&x1qcV
zzanWFrleBy?+0Q#-!r4{#c{@?qpCvZQ3x?gfu<MOW4q&@`+1NB@N_^CpWR{=6{#94
z-|=(KGJQK}z^l6Jppt?sg(x1QNJMUEKENE41!Gmkb?8RN+xv2k`vv6zLyF>&*_Xs0
zD!JFNi2{fl9}325LNsMOSk<$9E;b%e#Og7iJ?na2ILQ&)Q7QQ>xafJ6GxuxQ@5e5j
z-S@6?+16ia6ek{u%nX}&c?cIq$*nO+1@MdVKOI<Cms-kK3(zCbumO$=F<yo90sZ*K
z@jOZS(WukF2-oB>^OsL!*PXP`XL?1Fm4@RS#B&b`j-gXlg+hXu3S*~<4DT!!(Wh8*
zV$jRS{;c53se}&J7WJe-Hkj^j`W%LK-ix&#Z40@pMf*+D$+KyzA2=p7Ax$nu)^E8-
z^J33X&ahvo#p)C#b$gQcNs;D1&MP;|d@oX;@b|udrdK@C=sd>%@Uf;(iIIl(oMD$v
z<t-c!*KU%19QnV*et-2-g!mXOCK5;GimQSBz1%0#QbE6HXL^n67okTz*X3>YF+M5k
zlxDUN>&qrG0!E(}kMa#+LGsQ^3A}-Hr6DIyUk8*Wh;41J;UjUE_UEe!*eNuDJc6XJ
zZA*W6eehUCQoojv*0_L`L>MW%N%+0}t<<$;D-`2&X_T{EBlZ(f=Qe+vk0Z;qU2Zk7
zi{x|;FSUM=?I`Ty6MI%8!GPDcW<9e;Y3F|Y%V{me+(#UIJc8-3TQ<3uHV|J(4lJE>
zD4HJ;wBK$#P;)BKJ^0|1LQ_8@u`5^CFMc2H&FWhNfRJa<(7FUeUO^!gW(3V!Tb3Tf
zCmsH$RrR6?@7HFqka#%rjN<d<Eb+Zx(L9EC1)rMRI&m7<rZ?vBZ5F%y>VwiSdI<^%
z%$GPwrF+M7Ix=Z~Fv3}ytD3M8v{W<@@)vVt(pMpz)yX;+c#~o$D72I^R$(u55$d@S
zKqI4qtaSQjn*o^>nK`;EdHoMF0|-%|mu9=utOyo~X)zPMCyiSXK|Ih{a|x2<dtLoy
zEt7j^cr-zr{e9`o!*}_X&9joDZ_us-c$Aklp>$@E*xsZy8{8wkLh&uHQ_k9A*S|Fi
z=JqRVGSQ(EoF>K#_b{VSbHFs^T}rvahWeVp?%&$>vVR4jz&ggJt%ThaPe|XlXMI%(
zob9_mvC8}F%73IMPTwsj8{s)sNETjXcO>GXGX0(VqO+RP;M>~q#`jqJHy7=h0U3qb
zMZA(88w92m`})i#dV85cskuLrJH)%qVXwk1hJ2xD0^$j<^$-);N4L(mR@xrWz#KT{
zqmRrY9G(HiR3DuK(Ajhp^@pLQkSX`l6?xh49T!hFAbQeLJcasaX`XA<zSpqu`;_2>
zsSdUMSRfq5R=LNm#SWO9TO0j7btwtsPZrjeEF;-o`B?keak(7HXGT$q*o!zWS-Ai~
zvmzbyC6jYuf&nj-LooAKt7~X?I#;G}(ZkItlV;WPcQUWWrpM4>t2Q3r%E^vn_W^NY
z|JfzEfRXhe>?TU2Y}~aZ(Xp+g;Ke6en|l?u6{5xhj-(n0F{wuswS`mH>FMd`0Rak6
zVZ||DFO~uj{CA4soz9y<cG9#imj!syi^|1sXMX&st!^j#VW#*Uw#SUv26;G?B%b>Y
zL^W}CwKaKwG>F>NB2J!9XG|`>kCDL)qFOAg<~tPRz2&peXwu0hM2qDsY`D?2#>0w#
z+m1IBFg*|#dwMV8k@FwRC~Qxq7CKr7ETevve_&w1W0uZNlYU*NOvQ--3C32ka9Zas
z{fHk(xuC#^BhlGIxLdD#1xfzU-1h}s<(#^p1!-hNJ8Y`RzcZ3rad-z)0EgYss;jKL
zQ=G<Z_t)d256CzHL-E<)jpw*s@!Qx3@6UMbS4zSUhbis*YbaGCFo%w_EA$4v8q@o=
zXDxPfTeyWg<Zhqu(A_n0DDJO}YkP2WV$#al(&sLTIVa8R8TAR~muBaG*BrOn7W(Wi
z!X2l>oF^2xcmP`<>9Mc_C0xXVO#{3UO7>AXt6tAg`0gSHD~C6>qn1LGa|L;d9k>1y
zq1Pj$+G~p9!_jD=In;T{6<WZy>oQ45=CPwXUy3JPcR5IlXlbVoq5$=N!eXjOmONT#
z&vPzgnYacl86PUE1bmmn1pGZ7<A7Q5>sl7Hi6XL&obr<<cJT(*zxyhgY#J=c^P!f8
zy7Z4RykUE~mmi{2beqTa&O#3u$u!I{(1}QMJu%oD=X5GjBpx_37m7ES;(1JeAYrQB
zjZdsGi|h-%+`t$drttsi_ai?;3dgu<&mi_yLDAgh^}J(@iL_C`@yHWO$42e!NNkHK
z*6{eySJy@_2}^o~v7}B1$OUM&1+Jv;Y$l=UX!|%JDG&ug2HdY7GVQTgL)u~=Ulaqa
z?8ZwUDqUg$Q%h_8$PexmNL{w?<nlhU@7Tx?W9$6hOT+QQLnQ{A)JnHk=2m(Vcu$jc
zQdRLx6Ycl6<<OjnVwv#Z51+P3drv$;r1g(HGBEX~C9U4uCdtCS1k8`e<k!`)$mX;6
z-Z5#!FM7yMpN!v?U#Mc)_R9$M-o0kmHH(wy2_1GO_(hV-v$EnsFjVAPg^ilA>dWV3
zIYg@!JrSMJ?3fY5QVthmC*5c0yC82>v*Gn7I7}2%eWs|ZR|r$(#FFan-XVY1(A5eV
zss8N4o=^W*<6LYD)x*u<fJKz|UHCrC9Mi6?IBuMCG-{YVtV@D_Wsfy*&VAFYtEm6{
zW3)*5?{Mse(r2eJ6FQZ5*zY3V0ruL@cPj}d5C28lxQ{M!R*Dvq-zk;;9WAiqSP{3L
zX0iVGP9RO2iBwAYrv=6;CVGJm__5xFxe0x)fr$Dy^F^scBfnsKN_7o4gC{Z~+|@Gq
zrkxLiS#VB*LWc2k+INsd=5A<Cr)yDLlsfnXdoOksBw1B;r28A2MqwjwIF$n3=pox$
z!ChL5Lq#F?F$+3GzXeo}i<(aYB8LAbyJ^w|GZtH_mk}0TqEY*^@9+5A8=!<Y0ET_q
z>L_FXg&#qGh$VApK5bv`DQhf3Ni>t9mAeeq6>%TOyS&CkRG%|gL$hLtXe7xg=5Gm=
zw6;6_a*~}t$>Lr7VlmmTr6Xn|pArl{**S@l#AR~cM1O(BuC-{YtaFo@g7tcd*ZkdY
zQ#a78g+n!XEs}@IE&~G7Jbn}FgrU#A>H4x+S3#_tjg~U+ERU{Lg;hkxo4_jiB4oU0
z47qjw$ale~NR_1;6%)$pdL$>RHFpW8p)$bsVoUw3IP-zCm)7cZXDz@hm9sGa-e6$M
z^;6<16{fhx(CDS}zy<<LYWu;EBu}TGcW|Ylr0B$))o!D$y!d44wZ6yCVG`#n`vof3
z>Yj!Z3fvn{Xr#YE)LKVi+$ljm=j4XNJ{o%;zZjdTW{60;Ie%FTV);i2HAIBrrdnV)
zZl!uFCDKRLeps)>?u3^_sWDVA-eqJ!)`O8^@k{;(t_k)d9d-IR3F-*`0pPX;_ieGU
zF#3*B>w<xfO1rnImaX+{iS~@tJeXQSM!OJ39|+BV_7X*q3~GIMkTbgXSPD%}@?bsr
zW<Bjx1+A1#vVqoh**(>vow6Wf2--;i8H6*UU3|<*vHM25QY*)364^h6r@x}Flu^b!
zovX9|td^5&73EzUh}I`$QELV+zoZ5nyNI8O^h_T=-9z3RbY?_@AN7t~6~-R0)+B!N
z6f4hZk6a`ci>vCxou?vMFcG#69eW#j*+aG+?{2(E@1jAzlQEvNTVQLZYET#bT);|F
z$81VjXFGDP=EDn)HBIEO*1_h9h^ftHaxO4Cy~(856hGusM=Q<aD1Qzupum9GC_-}v
z0<a8TZ*3jh3Z9m1p14h>0O_?cAD#{O$wcQIfm5Myi<xXYD^l6RD&o10rqkbhFkwWa
z%kCcgna_s?-&#a~^UqwaU`&6i&;r5rf1~CypnrQO8+4xEAbj2pro^|7G)-1Zm}$6O
zmP>@Ia_#3gTeW|s(FiQTY3%!N+rUR*qA4`UG=qbk;C}IXe*NW+FzMA`RmXO-(Pm&v
z*P8y-xG5<~Yw_#i08pseGV=)1O@lB@E8Bf;!QxcUy$D07U$UM2n_cmN*{8;^r@UEH
zp+D#3Th6s);UF8LtFh$~5)d@W_k<q~&B}fdrW%^vw`jQ?dk)80O}Mp+^V#b-HQM__
z<$1n;y>Vaq7%`=w_|Rg9c6X{B#h6?8l|u|+)R8Tb7aP>kG_8JdA-#3#kpI-r$QBPT
z<6ZT8Vf)S@%!Qe@*?qdb^Cxc{F_Lm5jds~y{b+_Jlewj~IS-Pq1`9j#SpM8b@}`hk
zN7`jON@NQP-b%AuG)C$6Y(9O^dgGBr26ds_JJ>22lU4t^i^yvo2@5>1IX&$7#R~M2
zUcb^%(Z6bothA^2Oh?K^2=N8P7>Yi4bnk1_TNod=VTQE?74e3PtnYeK0+M==SYpAR
z(_&Rl<LE;N?eFj;@Aw|z_ADsdbrE+kgb!VqdRUllvi^|<(%%fH`qi&ozDh=Nu!a;*
zqw&tWW6yLInjcjM1L`X25&TUa!LNeGdRlwWlAaGf&tffNqm?*#i;C(U>jThJYJHuc
zBa`=)aFt+h0~zKIs6d{9`oiF!XB5Ic0#3bQ6T>G@_rM0`-=sfPdZT#XUT%zgr#C4f
zHBSu{?wfZzTihw|v#@l3x|6lK*nwQFA}U+(=7=de1I@C<87E?CiB1HDTqMJ0%A~C&
z2+Ssvh_ge{?u3<vW~gRhn^;h1YLV|*>K&=-o2hm*Q~62w>XoiBLJG*ASTc@`#<ilh
zeo;4QB8o3h@V^h0RSQH`ZjRu$6_<!Z+5gSa@*#d#A;w+cEaQ!w%z#ktlz1Zfvy^Nv
z*#)Wc8Z1sI2Sw4pEgy;fOeh(y%cY-uLMhuA^2ahJV(<+9eU+fJzb=vw8g|dFOW0I(
zzHl#=;sRH~om_{j;0?4zsEy9|%HEMj)=F%ou_bwkRXx)hHt-}#;d-*UEyXiE;@~Gr
zvw7gT>&|A4-g50YTBBKMee3g5nCC!5)1f{YHN)`U>(cF@R>T+EJDdG6wq0|PF_bk%
zOvZW(upyOGN;)_zRE}T#$wCr$syZ>v=~;`Xg7t9-6<f|J42|#9Vfxti21^uu+3GZx
zTp+<|5W3qwutwnDeiB<gc`ZuqF|MM&JWphL-;aY>*MQBq6c=G6pLw^l$8t47qy0{|
zPPJ-*Y6>Zw6Z$l94(0mS|JirWvA$lrVZ?&9O@6;~cQeA`2gR)reG*+ZIK$>!#P)9U
z;B1=b3hlD{w!d;EJ-7w!G9G`q>O2XGqPo<t+j-X4>LB=8GNbgqi^S=()W%VyG3~FW
zH=;ATk%|&OglI%@1<sm=1fFYE?lBvCMT>Xf`wbiZ^q{q|2RFfW)UJj_WQf!A$=c0H
z;srBBt4Yo$8@aUE*G2=Fo|+lYLI84>@lvzy$SoQmXg@wwPnLhnSW?jqwP81O8Ew9@
z<Fo<ckbf21w9aZ0*P78KCg1By#;)`D$x8#m=4l_J0JHL|XV(AXG6yjFF?5J;1Tj<*
zV&J`Ok!jl28s!(cX-O^Ecr}f$K+Q@L9w1F8ON=>aizSPJ_|4=0)CF*L<-E97q}c;%
zG2)2Z=$T0(pTuIsd8L1K0k=POJ`diT7V&4q*bM$gt?z&043l8;0mRiY>oWbE{ENN8
z2`7~}H!=6{$nJFYDMJX#P7=oo9V4)DQ@~K|<%uQo?Oc4b*A`B@8$LH9J{c$F#;mE7
zo{0=zF3lOMu^%246Q}COual*aqyS7g>d5gRTq{x;1}kGmi~u5cO;+LuJw!{LQY`q0
z=iM}_cboRS_f(Q-zB#XdunQv#?vY!u!ZL6`Q@)i>$pvNj1K8)qPSoGPxS-{{cLc$Y
z7Z6(>@06!4*5`@{K+`FN5!)Tcf4>AWHwtsQXG91*rTTn)%smvXb!9Z)5uD!#V`0S}
z{7NGJ_M^=+ROVkYo>It^pKy#($BcgfH069kqZQMDd#2)M<}D$xA#&+;Tb2%b3=TlU
zGVNt^CM3g86x_tyBswhDKhMoHk0&9$znGmcHhsl?9?X#D(xE*B4}YRpSRQD3$N%=W
zr7I-<b|7H@jJcgZA#;Tx;qx)kS>TmgTC{`|-Kpjlhw-NvSzx&tDFcOAna&o><%mPY
zW=Qf&Z!df3i%9GI_evwOe%+9w3))!;v5?9Ru#^-hMSGbpPe0szyDK-!<f3n2&h%BJ
zM(?hD8LaJPiQV0b1odMLhR5|?AD4DW)?%}Du)fPQH!-Kli;e(DQd*+)QI|QElr^8a
z`0p1}gdX^f16S)Vv$uVYxnB}u0b<%2yr!x({P-8X-?3rUJtIf^LEkE)XdacUByOT0
z*7<|Eha#kC46E5|Q~Hog>41Q21f%-)BjT0HACVs;gAIPmi1;N2?NnwMB)A}`r|7|2
z=1_BE(NprBs?^b2%e*i}e?ZV(h0FcAT_DsOgXV5TY|%2MI@EqpKG>k@bFHp%dXnK+
z<c?=qW}cTR*?PWq+5DP^{;#H6`1$%*4b9pgm^J^SsP{g$o58ms7U0{qb0#(08rj(h
zLlc^};TF&)pG}zDneke6^9uQtJlVP-E4uNCqg)pHH(a%|zkXsstFvGBcbD7PDK)5u
zi+7eEA@^^o0VM|3Y@UX-YpCb%CBqqkisw9T<!83j^<<A;;rz<aE^mYA31WO9459g2
z>B?lVN=B2CfNol%k+p7T{e2y01yL8SJ6dhSto#0}Vn_e&<rFe$lef3ZnZ&Y2R!cON
zUp4w#|7n|$51NemTG!mKY#Db4p1rN@F}?e{2byJ_A6I(;vG1^$7GPa$<vg%vS!qmo
z!UJtEhrvS)XDb5fk)BSZP@|5z@WJXGrUlzVBB1QIYgo*ASa$BT7V}U=UYg=)FtIa<
zL-}wJ>=my<WLapq{gbujA~Feg5~~wwuho(T9{8@z;IvPpzIfb{M5c)J{o?g;$Oaa&
z@@dP_g|<uo$6a}F-&!j!1}~fJ>yO_)PfDMqKpr<RNWjC*k&sQEk<ff6F#3|1j-3NU
z?G&<Rn&0=3KQ&s;73wgeJKnqCc~g;7VP#~)-X$m@;FRtRwhtwbCXuwGCyf*!P|qrG
zDj0uG-cmHarS5Gz@HlFiIda(;7ynz^LW}{DoIBmBG9H=MyT~l0!`*7@&`y*;y9|fg
z^C8bZD$f`s-e|EMbb5T!MZ09SSS1kW$%5?F+YN8(!btnP(3MJ%c@7yqah#<6v!2eY
z1ZOioff7W>>w0J^$)M)n6?P@|elU%#4nvT(K^ZTqzUDpImyb%}yyeBh$f9W7zDA&t
z7b%=lv4L-xO(U|k()3X`$s{e;3y#DP!rqMdYLcdDtsBZ2-IKl@cb+?W-2M4F8f=2<
z9SB@>;hOQNGnb=IqGMiG94!-u#@-oF#c*168&0|)%yKmM_|;rAH?FY}QHctrt`WtC
zv)7Bs(@Xr214^N<f(29YvlKo20n!y2UZhCVM#brViU9uIPnerJ3t=pd7{(2JtKW9r
z`Da++A$!`JkJhX)Uw)$95xm(DhvvMP!T1UQVps-cB8Oi_X(EK~U~7Qv$tAy7^7exk
z8hKrsy8^Zg8&Jo#i_rZNgU>2;8#QD47ANuW;s*Vz2{(A>l%6-Y^{PR5#PFN+(P*zl
z0z<t1#uU%Z|Fnm0S3bHw=8v6K@pns%!S3PQO>FeO`9HQ0@GhoCSHht*<?6nCvSZ)L
z7}pFFJLuR3B)mZ(S5Dn!YN&UUAQqR=<;UJQ-<2Fs#Xhg`*LMCZnIB;*t&}6}Kq)9@
zWGRn!Kj!OtTA;raj~)fSUm%HSr_4VL@_oj4zylQL;71+34SO_OL9nvG!UPv9dZ9Mp
zy=I^W8E_D^{(6_H3{QGgc(B?JZA447DB<12u4t!c>_<pPZQ4{Pc@<!{;GoC8=a8qm
zovU>~GnY;z3VwPMbIt=rgoV+}li6jr$X^VCck%sVhbh<U7@Zm{TKnsUA`ho;qZ0=t
zdtUZU00Z*=W@7V}8u`_EOwmoplCP%@-k{|I19E$DAhh8*CCoC$WqW-8lb(^`flV#u
z;4m{<HD1cqs;6@|h&azWhp%YFXk-;)!<>V}Q;jQslHG@c{`t(AGfFKh(C^p0DDSBr
zl?c*lR9#2v)-WjsCP4sF`Bl^0Zi>%tRr0vF%IjS<{-ZJzB+hp?a#GK@r_Z&8D`YUN
zXbv9RX|X7e50g(8k?sB#y1?#uD;D(rD3J)S+Pp2->QrJn4sG3lw_S_yVe^BrzC`+7
zjf|j`<BSHUMxZ$EtUxZH*2TbI>*8hlPlN10<|l?MoVD+FZ}TWk?uONuoR=8PEQ$Qn
zsdbeuEQ?hBL~3r{3{w-FM1;-LNT8~k7P04%e^)gqWnL=D{!Wl-05;LBWe%tPq*rf>
zxbmfyW&x+AZFJGCOJxBh;6)0S4l(~ZYya7OtJz0hWSG+-@W^0Jgq!Xm8Tvc1r}W!)
z;b=lr=Xq0q>!Dir8PU*#mDLtV`4A+;0r6bIE~q@TBCJ8K*w+x#pFKi*VHu`kR`0|?
z6mgL9+a`s=w^!u&f?g3}it`I#>{<gdD<v{5OgK5k+Ao0*wId37i%yPY)(d6wB8#~+
zmm|&uQ#AwiGXr~>ELu#+o3bhdb&8z|Rp!l+y`LX8U92Ew;0JfrK%@APEo5ylw8|;d
zr;n@Kw{Vl&z`+aI|Cct%t8rE7?X3j9KY(X>i}daRW?yZMv`4`m_UAg+H%mQ)O<rtW
z{dEI2M!o!8yIPywM4=!zZG=+}zq=A7nUX_?!`41%8&Dan!2wTXA1jkn)F=wK<%8C%
zhc`nDO|Ea|C4uE9Z#Rt_TL0D^%A6QCT3I{yQ$TFz7`JKbQ8^Ujb3a_M1_<^b$`j~n
zM?<(=wxW1~pE>e?$NBdBO0t{eR%wli>m8+Jl(+Dh^i_a66?*I(J_J8d(52ZclUTx;
z;&&e)4bpe0Ipygf-0lQmd(;9w#(Tk5uZ$2nY2u?YJKk=uIlP2F#3G>-&iwEUV`8zA
z=g{Z$9M#>7RBP`3K{BbW;U|S)!L`h}=TFytj+`?HcNV3y9Rz|SYarY2MorI6h`*%Q
z=%UTD)B)M${zVpao0NV@3K;~<M7{y1tkf3WFX@xR(E{?HW5`@E7`;n?A|X^?45)GI
zt--gqev|h7eN$(keoN#Fn+U)H-7Xo!Cf*Ovf7Lyc$h(j#;=;LsSsh*B_mJ+@M&T#g
zl-8)u%h;=fSe5e}yB8A`$}V>|MVc^cMuY9fsSg0nH-fh*Ca0!l$Lkoq?*c~4G`;9w
z+>jAa(Kw*G=yIf+WJ~dlpU#HczIC;i-uq?xi>D#E>c`rj3oS7>PAdSdhf1{Q?x#1p
zdP%SsWv(R?{PKDi!WH52w9b;($;GDMeMZV%3+S<4Lmtj&{|Dxd&cMY$ykExfqt1g{
z(1?FzOE)OF;wP`6yru1IHRc_mR+NVZ;<`#~zlZk7UvCuZUES{W#c`BBgYYrN`p1f6
z0s4J@#tjA9BTcWUBZE8nWX;+9$%7MRWTqdU2%4nQ`22_}UX`<QT&{0k6W<FPybS;I
znax`lUKJkQ!VAGW3wJrvTwm1{NwHhAkWi*-SWM7GdCR0$xX{NYKcfDJ0vAkyo`uGn
z6~qS}O7X<50d89(K-5Kd)MMjlr*N<($9-h;JMxBdhjUp0DkY!mjWK}Y+q0t4`J5SF
zc5?a{ODx*`<@=wh{sIZ4sa7yQOT8my#UmTYkGikC(8(z+T^sY=ou2g1!*0R*=g|q$
z{o)@}Jau`XODSafOSGq(=CE4stVqGEO1`YGd|8QtPyxWu{>TFjNhT|ZJl+Ms@o>_9
z&mcljXx~QThqyxu83j`mipVZ}z}zBz1}o=1{b-uNs#c8>z^`_cAudrOeBqa^>ywNN
zm-DCn6tgMn<pP<DXoM&?XQu68Qs{$e6r$0AUz)ycwtJ83a76s+G`J4TVM@GN$6tU@
z^fSrr{R)CUg>3KpLK>r#X(HS|DogV1(WQR?Zp*5{9#aewTVK&fD6YecaW>s;=(q@l
zJ^@j#;0&;0nRm)K0#dQU4Qf+os9rMJbuRtZO8=1WDZ~n<5aF=Ke8;rE-K{%J@8G_9
zw3j21l}5Bzo46Bx1_s%_w0c;Y=RPkKBbCe!gW{q`er`Z9-1VWU*8&A;Dw%Yy^;%ZZ
z(bvb##jk)bPR<|LXF}a44Xey)pG&&w4~LXLaO1G3$3KPUrd^@3l~*(4+RBqS2ggi-
zv(ag(RU$vMOr9#w_Q5UV9GVsM-;V`k5bdy)u!BLAN={^TrDLgtX~0|BE-}d|@ZBkJ
zOmGsZ{H97Jdj4S3?i9lkHug5W#IuLTT5oq_<N;Yrl;@~7ES>P)Z`$9I=7^p_e_>Mv
z=zJ%=JB$iAAY}B1WiYjU_V9Pur4!l#=^cEhQ&i^)J|j0doV&_`GxRf0p|7S4Ja^1t
z9@bv)H^LP&(E^MkYv?vi?t$L%B4HO@%9X<gWcdk~Q6fa*U`pVw?%mS+s6?(&Ep5}R
zF;w~kJQ)q(Gc%rtZ-FW~%B|sa%(6X$nB1Am{C=|oqyAVC$vO2O|2WSrekde0;WU?h
zh$2UCNb*Sjr-JCO4vx#8xROG-hRe+6jsaOHsZ)+G3@`FbWRYivWQKHMWF8eJkzAO)
zr(O4Ae8<acvy}H}5iVM;TX(cTy#*KYPU8vL9jcH7jM-dHDOv<QpsRmuVH4^6n*Bf%
zxdgnq-ZZN;fP-4z+Cv<i8v4$@<5fl#8xu#P6etO_;d#;6KZ-nJ3|iv-drij4uPc*-
zzI*aGaCIZ3g6x(4cHr&CGcpnZ++5Ru2-C>lR}8AercXQ9AEoHv8GUw9XEhFD{-sI`
zSx11sfg-`&Ui!~}%>SO?arJlugkcJNPTCc+nOCd<48qc9jujHpuL8UMN`bO>RTt68
z9dwNnjByV{5!l?+C#C8zuOXKOG6QB^$2GP22ZVJ%&_$q7!ebfYAg2#O@=me0XFXrC
ze@){tf1(;tczI0oGBOE|JXR4xCsV47p&YKf*Tfa<I;bnX>kg!&fsyT4OpDS2K~L8+
zBMIY)6Y`3_J)w*(TCXEk5Jksa!q_mc4Z7hoQ)IR3(Z$n=y}Pa=eTHe+t)jwSCHMTU
zj_M1Fk80?rZcE4->a!&LWLYd3L8LP6`FA-XtV*lVf-jdZmg&q?$V=~HV{D>?>C90@
z&kLznB2X?cPH5GSI_@aZ#$u5;t>8h!SRGh{V7-N&xP;dM%(Caqf98R30D&bgCvUB$
zlw#WzH7lvaL0kX{I@9W_7yV5?0T^prsb0OUY1Mb6BM;Hh6FMu2KYY&WrdMir$AJ9C
zUEw}v#CPumR}EL#Zl++`Ci)?MwT%YkKx)@Chs)mYzCs3q{x_z5&W*FWTBo*R`tz7v
z$uKTOeK~!8LJWag4&WiDMmC-F6)K8;mR&>K-5Ye3A6akF{GcL{dsCE)MUp`e1mpHf
z^Qr8}ycE(buRrB;9IRAuSNoOjLE*i4ndm$NyMdTde0qY~MC8jOTf%azPIuzzWyF63
zoaP7Cy0^T4Tl?u8K3!K9T^-<^?vQ9Lp*gd3_f*6wK;(l+Jn%Wnk2nl7hzqHeUBAaX
zUoEC)Uv$_PaI04!-(;zs5n~3Y^1Y?<Wk;@ZzRoEW17aU?U*)>v=t=`_sR*I3`Rp%Z
zR||+~Ys)VzGE#`@=IrNCt%(l2a|0QMru6Nis*A1$Fyq9OYIgaBS3iH|efcxMHc=U%
zbowG1f3sZMKPSD_to&TFTfsxV&C=}Qa)EN-+Tc|G!1zy#DU*$fL71yUrNc@oK>73+
z^mcFN=wFT+{~u{@9Tnx;w+#=aqo9CDi8LZeNJvVjD2SlI2m{g~Fn~jdlF|)IBcPN@
zH$!&_gLDrdAYGCY0^fP<`@Z+{J?nYiz4yD`wfJKc*34Yjb;j>Jen&JY$atxUhn6Zf
zHgw+d_}U`xHMFMcX#Ae{;*|Q&4%Yv)f|v+bWW7}GNIaAc^yUlr#d43BOpi&EZ|nX<
zXz#hXB3h$u5_Qucx2x28zfESZ38)$zpo&Q0v`o?om%Z`j6tA@PyOO?gV@y-LGONcH
zIOBbRMe_=Kub$e=V`s#c<gA}E?<~HRue3cgy&{l$B8Kh0N_SPOwnX^R!}>`^Bc1PA
zW)K*YPjxf##;8mpV%^gR%6<?5k&3;1K*yZ3rjCcq5sTF1Jq^2%O+&^vFTD+9Im|v%
zybAp#27iE)zsGu%G(z5XKU#ejA1~5PA)rs$x9~Bo&8G>82MNuO+28$+gwfiCl@ES|
z#;dj!8_q4kpAfI^C9PSM?v+kqwk$XO$GHO?1lzZsk;)*>tqOcZ3QWx;MSkK)AIuvp
zT2jt=p-VzsV7(^%2DQaN7-1AOSnlOIhHnKIH6-v~cYcJ2?u75r&&O^PgahhwQQuFL
zwm<(N>BlHh8KzHe@(iD*Hm{zaTK@!<c(Iw8an{k7`a`@Vo%FV0@3#v_h2Lp!Uy{=w
zL)|$kZ9Nnn$)w<3`*fMW;)pR?WBaX}n)}D!E~`>Xsr_k$oiHbfgBS1ZZihH|roir1
zr?Qn%9tyKh(&bHiI91c6R;^%-G3(!y<tq-SUY54HiKxSVNT?L?sE&lxJC(IMEfa3B
z=6bB^PZj8nkSS3vQ-@TodNidQY&dZvZmhp@Ie@$e4t5@K{zpH;^zSzR|1^5}<2N*N
zgrfrCwl}t}#Y`Kx!ruwp(l{Nis567OWiRzxj=%hIj^+!6mjdC4iEdWAXoQX)4xDx*
zjdSmS(+=MeB=$U_Stb5Y@%85ljrZs1{%@cC;R0TbA*!@bX~vuE1$^o&8`TuW)@U15
z@ypZZE|!Du8V~;18fcWDqZ$rL*HexlY-k)JZdHYou!7y?vPnMYw}1Fy@_r4F!QsU3
zEFQ+O=LcY~E(Is6DadUIoL|*!->C`HX5SeiADafjP|mYKH`}Nde9ueSBmnKXOS^E*
z<1E!8i2WU|&Pf7YO+!43{Z6+?K){GHqx&Fl^_l==X=<Z|jjad#*StI+o;JD0=#I4k
z+_my%DZmCd?plI0(7Vj8F!PXIP$M&y&uSa|aZG%r2$lf@@`0UycrkJxlnI&MK)TgS
zX><khgfiyu9h4>j;A{(2NkKb7yPHs^6&b@RU^9xm1!{5h1l`onbFU$dGB`$scLU#;
zu-E@G+#w;x2Tk#6dhnR{L7tFe^p{m~xEFAAO|=D<1P=M0CIJmED_TpBfu_0e1SA=S
zOTnrd7>falF%BVv!2M@+6ff;IGy*0e27=?enESw}4gVipOT!12pTPNMlz>!%jA5g5
z7IM$0oAP8&{JuqsY`FOqDCB7oSP+%9{MO@-W4e66A1+joP|as8S-nB?0gqYs+;(7S
z_ZCxWtGd)-hOEc!&b5*r6tL%!+baH-O%rhD!8g?5TW0>>hv9!4lmEwm`1k>*8=n})
z;IMyoZU2Ag+&?FRcRtWE6mrGyM*hFNNuwMXE{&a}txA7_!~f`Z|F^FsqXU^AP^yC5
zhW{$i{=qZ*Z*S6I2yX3n!}$PX*1uj{{_^{M-~tnT?IOD*+dqAia}>by;AKNGHxycB
z#K-y>vFF<W^xVp?!RgKhSo#Ilrx@gK87*st8Q$UmkNw;=ydf8YWXa(7;Ku8NVQNUt
zv`qf@<DuEV0hl;Ehw|50$E|aV{J~-VdDxEgd3EWNQvDZSZ|sBhqGGuSoQB%XfoWcr
zb>2e|qQ=7dz9tvAhnDY;TQA;T#$3X8B>|3IvVT}RN#yDIgf{>OFGI*#Q2|}lN>u_r
zq^g{Qf@*#b%)vY!tP$xFaKtEvETK+;^QWSwHqt?H0wfP8-FOPjeGdbn4mkNQ;2`LV
z?5f?pmC7Yfg+CV{QxrZ1L}NBVH)3h(;c5ZT^3&_m>Fa<C(0b!}#?pH5UQnLg7&Jpb
zcvl2SWGN^C=^6v(@kaj3WRf>j01PH81BRDu2T`l+-;WH|pM#Ip#{i-IYD&#G-gdT5
zZ71-CK|`5XdQ#;8#T<uVp<m4F$-r>|GS@A0lk%K{Jf0xami^g~+immUum14$i~uGL
zs-_F6woYzX{`OHI5F5puK;|#d0SZd$qZ@JTW$^Z8e$}XO26@_^5>3)m5PATAwu!k2
zY1HP+_Pei+ralHRJcq1%Pe2q!88Au8+Xu4qf=|CDBvq4t{>Kx_IYRC8YbXM6ot5!0
z5OQ1iJOHrG*LOj%j53jr_zKpoCGh6a3lIl1VH7|gTQ8K=-k-zoS3Mk|4~_{+`daKj
z43q>pmKK8oJblf1edH)l&^n@n?0o!6d_Az@0PhqrHfE7(OJ`f!NA}6`XPQA-UHT;{
z;7G9sJjEPJccM!`>eh&5h-DI|9V95S0j*Gmniei4Fb!3yq1{y9H2n(m-%W}E0{oZj
zcY!UiPsh9dX6;A_B~)d}vIaayotFW#6{^68mq-sp0Ij>!AM4}H(-<Q(!N9T@n`6yC
znPWVIvlz2b=zAQ_J0C573b>i412V^0Q*Kv#oxd6=qVm(U>`xOZm3n%zdo&DSHpsSN
zUpub-%tl49;6uS^FGqZzOoYgs`oPgG<jTGxA0UIW+M22TrxF*XfFOS#MEq5*qbZiz
zQ{sVci0ZiO*{LyE+SZ06FenedNchK?b_WhN74nzG*g8B@j|M+&SG4USUI}H{j~o^C
zc<h&4x*}!32bm3BOED!6eE95md;|*l%H2!K!(XlVW<xopP?NGk)3FG7AVZmlkX*}#
zMUMZuRle?YU)Cn5&>`2MF>sQmXO%d@?tbO;<j$TsMUdMmU*aI*2D^F#WPH?M9>($?
zPUQb_KcKjT2Y)SsvzH$jt_HMHyO$e2GI4}Gf28-S&VI5!oCYCdlcU=rUB&Tkrh2_#
zIE$X$Q*S3DinsSy=;gsHJisfL0wSbI1sFH}K<gpM*TZ`a3XjBc$BkJbNrC%th*ap@
zBl&9-)F2}x)Kft9KsPZktOAn&LaH}N8QOQD3a~T^g{3|g7*#!@?3)|FhTqh5lrICg
z=+-%D^F=UZeX90JXS}KH(mS9;aNi<3_f@^H%}1oQ`(*{T?;{puhs4o$=7epGO3K#2
z?6{5pox>D4bhH5+!0lh2RfBEb25@q&&G6G{9s!Rh(Tn^#k<d2beHH;W#D2oRUqcB(
zWu#lg%EM2v7E%;n#=&%21Mch3RjAh3m$klfyQQZ(I#N^w{S1Zn*DlDY{L*Hq4=DM#
zeb3z%ZdQU_mms7}H&XiE!5TrbEy;Ge6`Ad{@Uq4AQ9aN<e^>||gG7*o_2OSoe3EoQ
zaR`<$$ZJd%OW6w6GrYh4h7D(AO97(o!zRE2?j7VTwGRTeS|HyWohEer1uvV*a~dJG
zJ3rHYfxJkba#<u;cuk69ZTCinoj}-^bx<GJ0bGFd%PLDKSjl)hqC<GAKJx(3upD5Z
z;~*-ncUvFzJ!lJH(~vZ(I|cX&{Tx)PyV3@^4Cxtq6OGqBV}o6p0_!k^US}q?MUt)X
zgk^fzEpo}nh&+!iT^CSjD(iv;P+F*e%U`6XQ5a|6d~X`EqH=y+ZBWGK2^n~w^S60c
z*(*mp2EYMZ&`gi!DQdamPt8^52}Ia_-_(<YyQ!YJIZT)*+3L=AI=f{Mun;Z31Qtr^
zxi!mbcEyk_Z}}pRaf3GQ<)GX{vH7czPu4nc&c7u^wfJ%x35c)vHy!vl@yP6wm(Ep*
zfsF*Ttl*<f)yW@9UJ`c`ZvyG<0=*>3zx=j()btyv=71BRd&Y&6jH&O(NjAkQ$`Pp&
zp1`WFtvZ^=@!It*WI*^Cu=yx1jX9W@+Ar@YaeKci-e$t!#Bg{9*#bg{qtZ2P1K0V{
zqZ%M%WK6cC{^hZv>Uc$Brx?i6PNet0;bY&<u1ue&isQa)bYX*6miieRh8=<b0P#KI
z+(_$%4HYjLrUTwaGwNStIA%<>wU^<E^yXn*)wAi8*A@#?6M}cQa~2cUH&!-QvMU@@
zk0#Mg$)-~I>FA?#5;Qxasi4eGp9Tj82_xG5tRz&h!>3r{L+x=bsb^vP=&4#T`O~Fe
zJe>GwyFj2V!uj|jLKaGL>^DoA1HOxt>(?qKLMJejrnzRh?QeOEMaNA&;dvKfNz068
z;R)Tt8XmbO`K>oZ=Xkt}jDTc_WMmm~OxaBUR%;>)jteEPcEoKXE4_*qfUQG`3VQ<E
zIOW_aJ~8*BU$y34kK77kVL!pz!q=HdT6UzIqMT#mDY$y==(;sE_YAoqHqj`UGpoZw
z^4pgju#6X6>2Z7J?8#$N(-onIv9+j?1c)2g;lrEsdBpwbu<O9<iuU#C;UZT-FO^j_
zcA3)+AY1$*S=QWvCE9GMt-V-G^5$ZOvU{=rJ?|nPv765p>1+yRt0r%>Ymc0i5|8$e
z6k)%KM+*sC_cr>RA6Z{K+`Bcb^;Ero@j&`saa-p#DkbdtMV4MlLN-wGW#4~H2xjVE
z{M={h{S1p#v+}ii=DrsLE$Ymgax{bYEtRmEzYWsoIqIgdpjGK<3%GZ>fxou%N66fY
z8Uw7=TI4XM)!=Yp?SWqeo425dSbR%tDl9}E4WWVluNPjiNGYHpyEPMntNh=<Ab)Pt
za7FzZ{9HNc>Wr4=-4x}x2qS-D3-%C7Vn>^!Z5L2k`XA((%j$-jn(X*eJcP%V7J&nh
zM4ifAR%sxPe*|CYAZQBK{YV=NU$^%Odt~>$!MpJgY&s!!taUQn0e9D+gx?~KuC`yF
zQazqq7u1&1d;Z?CC5#%xYpMnHER%4?7kPXqSv{sr$oPz|k2~o%WIS8rg9#4X2EB`7
zZGOl#vk)@JU65oUhZW1JRYF-D2f|5RgVM$~6x4R1Ai5&dHE01?2fXSi3kDX+?FdU_
z;51cm%pejDT-59tRVJg_z@}mWItljz`Ic@LeyAhAK*A^`ff>0UDbdSb=zs6@P+mq}
ziY)c|%kSPw6Pb*fE=hKEu7!Cq+FTC6nDk2+4S$mXC>oZYMW;H;JpG&w1^N;O?IB?z
z6YS$%Xh9S9S@IPYAlTwfwjxfx1i!B?HXlv)?Gbe;r?2=f<+%78Gr}?uu4x<y3t2Aj
z&)3xL-bGJk^7LOOztBh81Nh0&rq?4xX9I&LY+ADJu9v>ML?@iPlHllVm{jLy8R>BM
zLg;s|dojXq$kMBka-ZD7zL;>~H<%wQCzzO*;I>WZP4G^98eXb+Lf&to!k&9WI_(Fo
z63r0=meyFFLmBBNMed$D;5f*i#UCd))yHuD(lyhZ+~WJ84pzyLp7psqy7)3jcjm5p
z7ey2^qcYqMBNzKRN2k}S{L6ilkPBi>pfgN`CPAz58@68ic*BO1qKT`mTtJ?7PXAg(
z2emcDa9o*(<3L~4hlCeU{mcKbKgCF4qP1IOb^R*M<6oda_<pap$k_zP75piQzT$AJ
z6}Z}1gBH8(?|L7J`4O+crO)W23x{l^ibc;mt!bf2h9!ppnZUDqAwqZ00%p%hN?XBW
zo&d)Zu48agm}Ia19qHp5<luW2u*Dh-EPF*eza0e%U9e4MWU~Q@#h{>NOM{bJMtKo>
z@!$`xf`aF-etKgFuTg#rV~`YAutF#{Yw9|@D@dpjW;N&Z_|*5}>a^YNwJj_#g1$<(
zkEk*dOqD3S(9X_~v;+?jEv60loI^vFT10*T{K>|^(s?bZ$|j|_F(flzBVDOgGB@Em
z8q06%s{=nr_VK#YphgjGpq6;cO-@n-4XxM<)4<2Tw{1z7szB2t7+<F4DKJY$Au5{2
ziH!w);Rw$zJp+}^I0ps8)^ozw$AFxiK21Ki-q0%KMID&<O2>M@0D9c5_-Sn5;rGzU
zcVTOgcLDR&rY1Vn{sFJCpl}Q@%^NXPF#w*BM`T<JBC}t)((Y{EOJ?S1xKE7FTqiI2
z3~<$r8i9viII)JwE%>YeA<vhGMHt_E8QL-=K}0pL@bH+D*5Q$(9Lh`WVpbFppDz=;
z>F7IYsagAGJUDL7eps9zKq)1hNUrY0Ce^4pm5kaV^(Jx~jaGiITt^<>fFDK7O%_`C
zQJ=)dk=4y`C$-ytsx|HU^rzc>`6|c2`V{WKWg8#11MJwd;-5SiL5M0I$R&{X_wz12
zvzcXW!+;t?&OV6F@kmAE=g|Em%iD9^p$UO6Ep<3S%C2Xil(OEM>Ie7P)8wP{T(Fku
z?Vk6g4!le@(3@50uY59i<qMd39VR6w?_I2V5PHVAwtmgNF!qtJ!B<%H6>N7Gjb$3{
zGd2rQnNVH77vcQyu&^cgp5J~D8AjUIoX9J0H$tEtYre`MJ<NOr5t!xq9;S*85>1BE
zA?`IR@>AiHZ>#}hUjFtz<?R{3fFvopW@~k_Hxl9OCmP5fAr|TCCl*|;Qf+Fm9Q^13
ztOL<a^Uy;xsr?E{0};2_w>Fx(&|p3+lv^90^pzFOhD)G`rwjHL3lnbn-F)pb^Br6l
zNnW_ij~iin@0)#{$gQJ{mn`)kxz!1dU>&f>@G5eQ9f*7`D*vkIVz$=4!eY%aY>8dY
zReXxh=<URGlgb^`w?V$<Ae))dBsscbs*^@;+Hex9dpV~ecch9Xx@coDfE<_;z5ea#
z?L_mQ<|Y}8#n8<C>vj1AvXcb@G}jg)$UJ{GK=yZUAmc^mZwjyuFxwdB;H$ZAl0OFe
zE8%z48qQl00WkJ2f4mv)HxS)VAV!KO->rbfR=)m%<;iqBN)BgmTm{NX1=G5|$-Z=_
zO%dJjstZor=Ukl|pDpB%3QWh_?8NHrci`5aExf73ilk29T#U39hARZN*I7!i`h8`t
zZ9J)FAbK9Xkd&8T*Hm7)NjI8ME#K`{B82{xLTk0)6B1nvl+hA9VMka+r-TsF?Fz0%
zloi`f_YuaqNNz!UOn~~OdEHEE<>#9_YOS3m`cAVf1z4g>2_Zkm-Fp0l5{Od-sF{_-
zLR>j~GKo6`s6Vy3956JBm<U#I#~rfUEXWmfT6|?)A5wf4Bl_^N*pJB1>J=n3o7$7R
z#M7U@0YAzbtcbZFwVOGrFYDkQvbwb$%Of&eGm>y``ZH-PuZ`v9=bNy9djWJHzdr)h
zLT(FQ?$Bf2rR=JaDaCa6!PZZW9Yiq+#72VDG1@ma<-D;(#0k~N++cNT-iKXClBCMW
zTRUo-SLv1ekHG1h+&21^{YRsrlVaYEbiKXcx*2?5?y`_*slDalNCE1}bLtgx(I~BO
zv$CN3Z_Zx#Te@GJUaX4ZA5f&>-x%Q4znZ{bN9m)-)8&7e{@UKgY}V|0;z(F@dZ@=k
z_9aGMoBV{-tINZe1gY<4b!D@<ck(c{9W1?x%zllu_XLT_5u2rzW6G?^wb;tQ%$fd}
zA!%KDLAq}*H;AaQib*c3ZnTS>7%#C|C~IA3!G)UBj5i}fOiZ$U3#|2B$PNnkP=`uU
zk=}j%-_$$jRxeZUEIhltClNNun-St@2ck_ZyK;q)KPYz)uNhT0Megp$73DjK;1cvU
z@;b9Z`b)qOg~NpVL8G5lg5I+(+;h9&m3|Y}IsLp)Owu&g6ZvRWJFd0z;LdOYZw3}w
zqrG#N7axm+Cs!`rMQ`a337UmXplXGW9=~3tqK2EJz7(z;n;AUv%*#D~Y!J_-|1*x4
zOBqpAZlpgqTYKMa%U9@b4<>UXn?GO?@u@eZa?`kDHm8cpJNCj)YnFd9H9nA&!xaq)
zeu9MR>`K=m8*;tLtMk=ivFT$q+YEAw3hpRBHZ<{Vr+0Fqi^H#qK?GfKx1bH|q3Dcn
zf*#DKWGvNll;@-0y^FFc8i5snC*lvkXe3kHaMxy>)zV=VDa7~yU&c2yFrMrV_$7*#
z?r@S4CutS<`Jwd5ZfN%p<EP}et0-F%XMF?^EU@<gUYdXpkx4SKb=W;GY3y8Bf#9i&
z41{!WB9U>+bSzJX)@;a4iGc4^Ma*>LWpX*jRD=&v34-bCgBM%gok8$W8EPcdx-^2F
zyaVVR9<}nOV)T%mxI_)9Ti<9+mU;r^$hn6v+C(=;y>N_y_2!wIqlUQ9UC6RxGKW$k
zm$K`Jpg!%gp=c$!qs!R(s}g03Ragp?sfZ;LOD(T%2`dk{S1$N5>P2*4&)3H}_%dxb
zeJ7!av)!3AzGyq9lqwL&Bf3ugkrvM|;Us;Cza-}VeXz%N9@+zIhBnNjHDat&X{wtc
zuR~WF{l#Nkrp>~zgZyr)XFYc3j&q5DC;)Zj3A>M)Cn-j*o>J8(PuZET-OLjh7J_%o
z{D`GteT}8wJqAbJn+o%sJ-ZqUk+0AWukv&M7-e?h7x0)4?Q=Ye{1x77b7A}%5CcEu
zqm4PNaSQ|a4Fba_io(cuywWwYs^K#6jG!Pr?0Zn#keKWwF0J{Q^1uOqr%Vq&*GbQT
zrS}S6C2XG|`>(4QCVa^l7|MF`3dpXwWU3RLPU{le>gH%6hSOEY__cOZ<=2c8*U>U;
z)Yu2ak=7NVP_!I4r*5&u_lJ76-yDcSR@HDWv%?`?)q=R&b5oZ^KCcrYooAnuP-kH-
zbDhI|nynik;+c`U<kI3+u;h_@V306^$!fj9iRyNHcNZ56QUa(gel+t$_#GP|Gb!bB
zxK&79GBM{{{fwl(pHjP^p`i8~<1XKtPSQezu{?K_YKT*t=7EqBgcEg^lmWhSzytG&
zt27%-AwA#Ad-J5nK``6XnxQY&xOcPo1xvGcYC~U(iOv?2Rs<wp2;1-bq12yeI6iJ&
zpua<b*=bfY@6(?KqD4!TYY%Sw2}U>XuvobX{>ue(O$s1~tCGWx5*r1R0u~tNwspH&
zuOwrfS{1CYlgK{7Htk?tX*K;oY9uT)0dv-pCy1hPm_y#i^s*!E`_GTR=B=H+C)T;N
z=vI+_+9-&Ta9MQ}(uWD|3+xNVR#}ST3g)i4gF`5MWL~h+dJy2=y|^n{8S<~(B^}#{
z_%@yzy&YHl_ZU2CoM&H$iF`&86~4WSLMP}s7$v$it5l~=&RrHILT1gT(fu0adw!cw
zWRCahNE7dx-DY3bMrIM63nu!KNIAya2`{;_tmg6nSo8$<<I3_Ka!O{3<A0ab8S^@k
zr_UrNIM5HHS_k$4QzU-Ep*GRQdcue~$?ZebR{rqKSc6x~_IttAOB*D)VE}M%MR?AJ
zi_~9Er!RJX57H}|-EE81Vuet1y2;yf!BU&}qCI%jUv~>;6IsqwOBT&?j|VDLLJoHU
z<jvovNG;!vl%;ElDLBJ?oZO?GySNG6Y0j_0?fVOb!lYbsgx8pR-?}}z9pVKK{L3Fl
zPH&}L87D0z-Lhc0x~h`vP+9?E$+Gp*G*}Q5D-AQkCZ=>li`eI13u<lH6-fl<2vU;_
zzRFa)&g-EcJ;0s<&kg$Ce<jD^#|4+hItPQDvHZ}M$wIsD{_`R@wzm<KxXC-(cJ>J&
zewk@zGTQ^U9`rZhfB)X-BbIj&L4*U8o1EDWg1x8;*phKXV6)<_4aJG$bXAOU?xJz2
zmbEthQ_<#ZhQoFNQw#ZY_d8F$@1xkP-9M#NOCis_lcu&Ee3f@4Acihw^%k)g=8hw|
z>ip}yhxP#>uUFED8U!X3wkE=cmABmU>1YM@Uq<m%uN9ZQwc^bG*Xm`XeBxs<j>f~f
z<3vs;^}S52NODrG^UAc#`|op^VZL4$Fumu0JffRF<63e19QC&I5<bnyk}H?1*?xp}
zJ63yz?J3L=&rnO?YmCLzJU8iefN9ur!>W2ExMAfYb{Ve`jOM2eFAzGucbw<?U85AK
zE^y`T4w$MHq}yfvN}@gP(j76Jry{mIoeS*9Cw`KyF4$w@B)q4V`?9!hDZjVlr4E<6
zNhdy&YvV<x9USN?8^j0=%a+Ex%6IL$R?%JxE^+w8s~S}5P1&WykDG%rA)X~4<paz^
zjou%VvWp{AZ{$pBQEFh8EH86O<Ub%^x`X!=wtMEv@z=v5P)7YmZy@F8{9E!l+8P}?
zF0peUJR+_V!Og8-`7@o98}iITq%>BASL35ZF1b9mptGxSKu9zfU!4J160tQmLYhv>
z?Pc{Qnu7Qh0x#E0?3^{r#;{oK6{8F7BConzrHr&Lbtz5KB2~@oYWHG|%E-Dnt`|AZ
zZot#Az@<1i4{&w0o<5Jc5mfdZkIVznF$OqsZ}hd!ts)H@`-rG)H_D%PXR{JpAd)_I
zI|UHQP<tb4r8IkVNwsN;ehy4k%QU~C`CSF>o%Y?LwFSB}TM&3A`>aKC9su``&F~7b
z!ck`oi<6S(RWu?;Y1M)}=9!tRN}U=?(l<-!_efKgbNAR>&fD-R2)RsKZdCDqE9(5g
zLW?FxBQC66?lS)2s_Op@0dt%wshsF6WLGxmT4T%$Psl6eYR3N%qI{{#wTydrxGRZR
zy?PLzDfXkj<yDhQRIl_0uW3IFp_&Rld_V=8psQaEJ2A_%E>sW)H+15fbuRhN_2IL^
z+`t>phEeU;(V>-v(&OnN1x?$=Mc4HmCc%JMJ^o`rsA$8A9=U*=0)I*wZjJr|qxCl<
ziB#(UH7XFN=#S43MAIPgrRh0~HA!xo@kzg8t&tN=&e!$49z;oX$Iy6E#NE3AL(eUO
zDZHI8IO^W|d>+sR{sJ=tzp6@>Vb%Unbqk%RtH}Ip^66z4SE^yt{b|p$tSHQb+QRJ_
zl<tHYfi)UIh4$9i2Z*s&$_)l$mW>yD#{hQK3o3Bh28~Fwy2kj*q($8daJz_!o2ycY
zp9)ZqnQavFL9NJQ;HsCsaMw&plQ-9Di0gc8cwyZvU601pY=xeg%BwtGn4?R$QR7!4
zFQL4vR6DpSr<~zj@nZYo0UlUY<Sq2Co`etsL29mHg?ao4lHod)F*7RIq@qiySgvLn
zHEzi4N4h061T=noxI`o%ssMicRlp?(iM{{e-=((sG9zL&hZ-43KCHF+bjfi>ccCdH
zu|SZwer34b8RyVm7(BlrU@fH+`a_a>)HJldmfx`r|3mfOW$U0XZvR>zqCc63n;~F0
z0LK-36P(xP9uRsLQ9PwW_4{|g`KXL))1JbPyP#*+HvVBsBDpSg%B~IHFk$uDLIj}E
za-gZ#jVdpPlpy$##z8llR#J(bOmJ<pHZFcjZPmK2*WOXVzN#1FFG;fntwFh2{NErF
zW?3<v11O=t;xo<J)!KuomG)8nO%JXuu!B6+GpVFo{jkUV+*l*tN$oXNBA5m-iG$U7
zR|>T$dXN7KtNEh_(pUa)*AKmSzi%arW7*k4j4Tl2SB&LZaazi{s#6758MXO#UV#O(
zg|>b|OBb@&1cDl(Du*&a#H+CN4m@9-{1Ox*WV?|GZ$X20g)JP$==vk-s=gC9iXFgx
zaS3`X7s<ba-b0+^RttqzPn^4V{JB`vOeU(IuU&Bq;N-Z41OY_(EtmpEhF(JlMgK{S
zVTW8|l`3W>cmnqpnbC~y_MPH|WnfW497&G-D$T8BdVbkWGD!f)EGd!oIp_2RMyBIV
zp~@%g3{n(KOY3#L>q5G8gx5kqZ3yiaU_l6n5UD%ouiMM$Q~e;{!k_hr(qH4}|2&0}
z2}7)XA-}3pAzb#HRA~x~tSCY6ea@lBEjAMYTPW4%fre8=sf8~W1Z*y7e=LT4P&Y~>
z_ml*|=Ntnh-JA>AMFy*KDzbq<bTWDuHhe{<u+Q~5ay)p9=v`oeOJ_}Wr2LwtL3fjS
zN4MwXZcj{Ez8JVKf)$QQ>ma^(45SjKYN!(;H3aiP)9zbe53NoW*BybWq0WJLD)80|
z*t`=n%tNKZ(1t5<WqZCy(I0ailAJh9p7#+s%sK-5Ky%b&Q5T^X7#oLu_`IK2_ZU6i
zdp|I{0DGLbGD>mb9&D7tTG92&U&0GZafG7{O)KHm+_}GtmP74tFfQinRELtrRtLK_
zWW(to2<(-XRB%&jH0O3)&+WysPFXieE;`OU^N4h%u!~$~Ho+##?DZc53h{D`7E?Vv
z{Jkz~jMDS|=sLeL3vzBRuX1h#^EOwh7?}EMNpOE+x?}n-=D5`h-k?9>7e5aghw%Jm
z+HJ2rDc>g>#@zB4AJLnPWGM}S5>?d$zw=1QzhAidp6sysh!h{T66V--`47j{GmzPq
zD*3a4z;=Zs$7DnNc1t+OpSDWssp#yV93>*)CaCP1$=;nLdI3jor;*U~8j!F5NT_-+
z4t3Hmugbwa$)5ipuq6zKhZySiJP)tFNz#Z<H`!VTC-$|1G8)?aU8EwV_JW)Q$JO%o
z)oY7`)K4dEEtJMWoW`FsW=>aupoeqrOg+5waTZWatxba$p3uY)SSzTU1VI1GFH@RZ
zfkmzwIP(_+V4W3=q{G?3LW0J7I)vMSd{?{P4PJN?aMe&rzu#ps!6Jo>i-p8T?6)RI
zx>~6&<B!ZCe!-%6kfkS&Q$G1Y&_qPo2C^4i{Dm~eUHD6N!?llHJ9_{mOve&h(;M#b
z8!!sYpll;ykf)_EVTLsg<@Q&k*1yx{k^77?PhMb}PrEv4+gWks@D91Bk$#j^&`*1i
z{PQ`Hc)MpmW9N=%FKA-mgltMZ^Q})ByC;XWDL;mu%D(9iPtZ!RntkKkEUG%mZ?HIm
z2R90`VEERv9y%H56|esId+khfK)dK{zO!KN24%b1_;?^WYq3$Yzu88)aINlm)MnJD
zJI>TR2V1P?#@U;!htK`^G%DRmbF8O7!p5}Abs%V7xr$pm-mH(%yIpr+oNFD}o^2GY
zz8T|W<5ctD&`|B|(EuQ>P*JH4IEhn@Tv(pW7iH)20y)=OaBe$bH0mt<co@^IYnB$o
z9xRrC#<v}wENqW3ggS!=bwi_qoZ;-e{A2ReVsj&#4+R+wuTu92wOgOc_oK-BsJj|}
z%U!p29+nyO^sr%;OkrlYyth!et52`^ffnCn>#d09--8hRs^2%NH_HKqui_Rpt8(}q
zYKi#j#hjMowy48%1(h5hujmX1*?kxTzQfuvAd*Ycg7>#Tg_kKW%V!+gA+qw`SEP(i
z&`#J3G+yLk8~R50JKcGvloypmG!GwKs{3lbk;e7%$lSxt0+28weplb?!U@D5`lnQ8
zf^sIR((ERh*+7c{$wc*jl1P8B5%m8I>nc&aS^U&i33e?L0D7e+K|V48khk{GF%nZI
zhPHO!2i&^&kUA{U;{|QJFbT|_Ec%zr0#yOFm1qIwo-^gboOoznUHL!=O&lCG{hI)_
z-0mjd@5VZT8!D_M>qIsf{m8_AM@8B~{CMii3Dx4lqCa`2IfSO!!ip<9SVl2)COh4~
zg;VWfZ_K+`D>yGEyC@~4*+4NUl}iG3GfBA%!?re{OD;%_Qm7j}dhs#*(7dLSsVQ81
z<#PwuZH%=9dew2kSQ~9ExuDI^-TGxA>0lwLxN(AqZ47+*>zf)2#^0rE{i_z~P!8=&
z#C6NIymBG0G9TiS-{*nMjhK-`-uG_uQ*I_W+?5ZNpZjq1Rub}clS8JZY88*nc{?sF
z@EV8IWclhYr$-O8%4ld)-#W)g-JvzQb!mD{n7UHoHFzAhV0;NpP}+`&bf%*;r;so*
zcdTDQUx9lG&UxD11<h~%Dr1oK5Zx%ht61^M6fhpGLqm280PjWMD8a7yyv|XT9@E8G
z-t7Dz=|jSO_uV6dirZ8X7>-6H!URlJ)xr<;RR=9d5>wubB<%Jta?%&xgkB_gsLONS
z>dr)us|Zn>8TF$U#Il^ENnLt<-Buf&gX4pg1*-KlJH9M|hgCwIyRSk^3hIus)g<=@
z`L!$e`YkE!JpePcn@5Y*eKfDK6|<_{acgr}+<Ou9NJBJ|G2NnS(iWl1UzxiZe?*nb
zdVaV)LOduvmq|RI&0tmL8Hp${G1pQ`#MpD8A&v=+2|b!-9&LoqXb}%^!0GoT>Zu4)
zt18y$T@4d!Cu(kEL~}Lzsq)9YquRmfeF+sya!oyLWi(+iIiu3EG2(R#|J8Zap4Z>@
z%2zB-0t4#>T(7+Q+okJ5!wA-K5o`Jix91z#wf)ka*N^p{j%1Kj|CCsG7CuO^s)Ubb
zb(+Wj9q$sn&R;vdQWUxsF4PRFBn#f{;57_fYMAItcZNi<w!P6GgdWc9uJBN`+`BZJ
zU)w?a809K%UkcHN=|1SFk#n~#{>m;4EK}OVeaZw(Nay%W3W=+`c}j<ok4xj)zu~N%
zCIN8O(|(3tYCcU*`h1)EL)*)q>n&&2G+YRm5@(b3=x68>RQ2Alo=NvZ&mG2Y1$_(?
z=B;}h{rHq0wE@lQs4f~YZ<HJ;LjhXYHu#AK<bS#Il6mJddGXBvg>ium`|$C(CN72B
z@QKdL@MxU4uHss+!4X^#$hHE>bTbx5x#icI%^fZ9lb%`4)e82AY?IJ+*|3y64SU&=
zhu69w;i_=Zx<D^SdrCGxbkXML6v$KahhN5n%}NW4lT$Fu$t(CAkl-Qm8q9^?PznDg
zlP%=zSYVkY_hqvgtV%LMk5t+kxa4>*Xjii^kXv5i%{V`1nQ3#_V9v(kB|zX56n~`=
zm(1Jp8kBiqVcNoIa^LNM?;*uZg`aQuvPpLrnhMfsP_RhjTd{JWyF4AjV0$3bG*r!H
z-1v$EOQOkRh8ec?s7o*L9Z;NKM?bs7`7np4apRN~d<&yWIZ5dn$~xsvs<m1U<XgT9
z1_oMEl(Mco2j}ge?q6=z#?93H7E(<2CO~P)8th2U#bVGIpzl1E$?}q|E=FpF>YRL`
z%w(p_*v^{eoW0Wn;pM2&ANfQ1pC7+u@%U)VDk{$w!bgEO43c@T8#m=&fxVSI8iJ@W
zC}!SmRd=TFEQaqP@g$8jX)^dH^jkU?^U8y?S9~O*#muXON-QS^FQi@WUx?Tbf)QwN
z8Vj#{qS571zF&~wkuo)Z&%IZNEWbBviF#K}LC2~_!BM^v5Ykfp5L~xsO-@-Zjbrqj
z!}yc7d>rFTmgYE<)0V_jBOJrMt{CO(xf3(<uYVDx`egCCpI_S3QTQ1=G33o}?c?*)
zC}-@NP2H;GVyucujq0yu0(cR8HwrLK1ivM;A9@pbS0<J#%fRmGsW6w>j#KWiSYKRt
z(dN;1cbGEt1nXnPdw;Y1m8pR}u^Iobx3PN7B;>~Tsqw_;MDaJSZl9!Mi|S*X)H>;u
zeV_1UO0#CD&jfSj!#~nkvl1rilXuc=($K|oJ+&^UMlcGPV-%bB8MxLNnzR+tavvXj
zl(mq1KGB`;ba+mh>P1&h5h}CiF{%wQ{N(pbA1V3CD^<pM_p88``}A>1ii~~NX!}_O
zLLIwi?dbajzC@m*DJ55TJ@6P+wVV~Ge#Ym)7oOoI6UnHM8M&Fx`B0hM(hu~X^LhuW
z|K>pRhy_nLx^TR4;rVO4MlBs%w&EGT%#hN`Ap@yVoXO){8nVkp{No8;GF-Rcu3+zm
zNfA75k^VtkM6KOP=FP;h97?IULeNL2O`t+Rej(Orv{32iV1id9QGQ0_IY~2TY8QXM
zJN-G7qlJsE`1l0AT8CbriM)|gg=4`>*=sVDU$}(rInU3_a`^KF6k`EHD>=K}9KrH4
zeF`2}P4<-SL3A|Z<v>0%w@fd@8@r*|HbI1j+||x35!)C6rqS=Dt15eh$(-)th?~Ad
zbl(D(z?i=|PPN$D;PO45v~{0U$5-!AhOuk$(~#8t6!MQ*k4%*%*2DMX<SfQ14?7p%
zz36Q#QixR$;jAr=7{qf5CyR1cq3?gVlp0OSrJ%AK*xIhUH(--wSteyDTg_pOHFa~H
zkI`XK>x^#W;w}@g`?1RLjMOZeHFHE&T7C0HhC5531&;k$DeE}Xp#{mklOzvi$wh+s
zPuOS*Z{Br2hkJ9pa(w;EHio~IaZMPXz_%U~1M(iHOEv}xcfTXK?o3E;TM3DqFHGAm
zyf1HvkHp{6N-Wv=wLNF=@!tIU*N_J)0p)7P4_5KU$xSX=MLJsyU*(hzW7siq$*aj^
z<jimkIbl7M>M0f97b(eCah;1_*qpr{H;$?^*xrnBFc^Lr=lfxOK1jIfRms%o#8h?I
zlnJkbS1=PLxna>sd7MGcoUoOk`6oslCX|VUQRsribcJ|HmgVgN`blfE$)4(6hwTs;
zU$^wG82*m+k+f#r^+F4^1kzUFQ2XREi;JyNiL*{uz0ILh6Sr9H&A+5R{Mk}?^sWJP
z0!*6yFA-adBh()M*~(oK$H#nZrMw-xZ~rkL{CRIZ9K3P;ie4t=UvCQk`mHxFvzONO
z+_4!JiY`w?PbJ_GuIN&(zE$PW_;;^($%3%(y{KtNWFZz?6H;ak%2e-DMH&2tQ?2P+
zd4B^N{|h~gmV?E;yP52$iLtsNOX!91b6YEFZ_3+-HdqKn<($@J`Ew4Q$$tlk@<oyQ
zuk`B>s|pYr^T87D20^W30QoWkV)jJ)ATRhXEB+POetM_2#@}L55K{Gl8=ukR;Nv<t
z%j^Rh>XWzEMeRR8Qe!BaWgOTXSP+~e`HUL~`p+fFtP*b2{xY~b<4bW1-%a!P@~2dm
z89=3)0R7hIP)^+{qvl2MwQK#sGE$qB9sQp>_UE(tUtcVV5f)190*qo5T-4*QgAnfv
z9OJ^&_*<{Wh$jDuw?->_xyBueJcV&*NkBA5Br<fQ3?3*meS-<la5_&fM)hw1xc?oV
z`#?$&2`ePk$q#1#XCv>=A^5Ksy)Mweby?NELGe$<*Z=sNrDlkB*Bq0R1|H#`7yHj&
z?CL{LeX>gH>;Haze+Nq*@!?&79ox@mTmCuX{;yuR$APEqUV+jQ_)9Lu|G4gdj|Fcj
zRWNm=kd?O*|L*DmEBF9N57tk9O;=cDC9t2iD1uRp{3Eg0?5R0O$ndPZz$Ns@*YK87
zli5*D&7aD;_({0$s&MUK5A>LN=)8h&|IK3#Y5){bGo5GdWhbCQBrgLiz&IG4?@_;2
z=D<PH9B4+o|Ijf&j6a>Y%ui;gkjxZHuY8*aj^ASn6lzZ@H!j=+2hvhfsRO3s?T}}`
zn}Dj#CnYjZD&~B=fHM;40Nk^Yw?d(79%QkHgwjB>_M}n>pqk?VVR=@$k#n=P%=p&g
z<-2Q!AesVIWwRujwGh#lt~>#p4WGn}Xm@&BLXGj93=zWym<dmJReC`HS3ym`CEZ~m
zAWV*dUeS(5vJ~TE+c#3(iTo$e0!b<~!yi&3NP}fAfI-KNm-hC5oss|bpWbyc-$8|4
z2*r*9$p9LvLf1j9UcoHyzw4Ja21SMgyw8gsq?-!69sGP4w+!SjMLGtqt+Aj^d{{p)
z%}@gjeAhq)SPJAhgHwP<_Zo_D&i<0MqCxZdSee1K#lZ8#Nzd}5-x`3o>C+FZL7<;4
zAY@bkCZ}8IeEM3sBrqzOh>$+D1lHKiG+ghbxS?k_sR%Qo*fF@`(hbNe&VU!~GP2k{
z>e8>f1=3pWe$0SwY>~^f`v{1eR!jnf;iS%y_(bW4;>&E+%UMbHx!M6<Vgtaua<GvI
zwnT#1?s9sMgBkv55uz;WTE`zz%GxJ1KLEL-1l;Ct5Y&`)pFeB71$1nTpyRYVpnykh
zns<5&{QzIIAFPd2g;Ec7fn8`G^A!US<|)r12I5<_+rqP8%QwhB%YuP^?a|d6e>470
zq8j3!c~}ACO=<a`<sdJ60+4C#plu;l#sL5k6Of?_I9*uWrTc>r4)g_UK(n*4#&xn$
z3HZc4z$as-@D;)B{IR7q6}&CRHx=@vl7s93s=q}nIhou8ZsaH=x3~-LYEpUVW2KZX
z)I^|R-RY5ax*N2Bv-FN!;9KA+Y{2(#6>VzpCa?vMeRH+7yi{A&coWbB$3Ra@=NNcj
z43zFo^9y-OfSIxnJSdcsV>H4ceN^EaY$CZgHD1vf&~Z{Px*w(jQfU)ePo`JvIus-y
zes=?Nb*R@twd}$;{~7fWKyg^=!(s`x822J*b^$r91WfQYGC^aF$USmtjyDwTKrmUt
z%>D{ND9!?wdu3n8XMeNWAJKY4OdTi{dyf<EJ7}(1gB17rrmztJV&ImqO@bY+IgLJ6
z;_CQKuaJ45A<I_22z&g^6N+yJVn%5tA}>Vrd)xA{HBg}!0--9OS>8m1S%odm$e8>&
z$W`9hFJAKb<91CFl5z0+r@mVGGI_Ghuvc6Ed;WG?gc1nn+4nt64}HOg$D~c(p5Zs0
z(kqh%Vto;$cer3{7i{sv!(!(E)v6Q#{`@V4@&T7GSw~W+-O)hkD6_dvxi0#}>;WCu
z6KUR+3zy0w=4HW3sr~U*_}{FQrT0JF24r@#Z4j;XZM5(mtkC4m6__(N@TPr%B*60{
z9PmDjJ1?$0P=~f086L2KoRY=}_@c`WoL!K#^O3NhXV3FAW^iY<V<q3N@tl`_g8`zs
zb#UWKwX~l?Nm@1cz3YxvxF9Kt98(~SEGqpCJX!?|Z~?1ZJb$whSVqF~2T`?q+RbKH
zQI8!coWVSr04S{$IAoO=71^}WT9kst@D5TF;_*Y;IKLrU+<i&{-Z4N1xB(X0#WgS`
zDu6bPgF8hHFqn;;Lc2dh>ZaC_?0oABFb5sKq!G7<``^Jj8xxT&-rhe3#*fL5!~f9r
zkw!dagPJqKcDY4u*{|x>LZEnEC?GxC6yRkBN#HYbW`n6gePr|_=_U8^_oh_<sPjru
zjYOM)DOliIkb4j5FA3y&Gl_>_Kwtj#8fuM{#sPwTMb}&L2fSmzTS$z~;=KdODqHjg
z;Zn#Fq8a$olwH>URpj6b<OGTnY5Gu&v$Y|ygQZyikVEKI>b{+X?`{e-iz;z3a}XgN
z5;@Ml0tk2cb|;q(?xzE*9Uh?nGnW21jj;IA4w*T1K2gbNQz_Y#efGh2hQ$?-?gfw<
z10KP6IkG@-`z8!v;H`Q*&*H|)XJlS?oEM#)!qnC*PO$O*EX1lvl7R$9a>#zaIp!x&
zn^+(=u7l6C1%7GaRk<SOp<jypJr2Q&uS3Sy*0-=^@h|m(^C#;Tn7ltp5$F)`DQbm3
z150eFQKt_PC+^%gzGf05S@^lKPESatEvn7}r(9Y*{0X`)?W0k%;@2t$(0SL32!kvW
zUwk*%0o!)O)(ViKx$g2WNRIbhdBEMbAee_WIxHVrc{##vqNc<bHvoE;Vte0c+Au0-
z^<eRCqBN~_t?U!l*iU?8><PwCn_K6&SJ4^pdI+r~U>4-HbI6%ul(!)AIuuWMp8e$w
zt!<1{lX#d+k|E9_tY5=2ORDp|a>${zlIY0V@5YMi>{OA*;VdCapr;d?{;BK&=FYA6
zt3kGD2`>z1Tzfyr^2e31-#RzkpX+{bHVQ8KQH_2juZjL+rE4)H%NaUkpoG+U@=k#B
zFW2^7oeNAhO<GrzRtU~tCh^9B9f!(TOaTplJFI6XdAyT)W)oOX7R%(Q`+B`>9o#Qc
z?|UhihJm;$!n36Weu*sD9caM);eOT%<jn!35t~1zyHZE4z*saD3Kf`0ft!v4+zI$~
zrT)%w2#=3)g94?Ql`@cl2B5~TCVNxJ-_|w)OqNGJF6FAtUQ$@AKRX3r>MyM+nS0)5
z;SjKMMQ4>zwi}Xb0+QWhxX4SpKh%7mp}tx9SN5dA8XYAzTSMTx1>fobrxAK%Vd1M|
z9EwkT;>jzjAuX%5#|CJLT+;bwBMK&JN+TpJy|BeFZ!$pRfd7VVoXAOwi%IwHLcEDw
zuHONHi+SnI7T}$AK4tkLbpv!JDz#EjbhD0qYF-3U^0aY?%*hS#{gvGN7F=VyhoI16
zV9fzzqd&rdOQoOx6v34H3d~x$j(f)lU_ZpQ`9odzO9T+{4TGJ(^?-d=UO0%{g_59Y
z3UmNO+Pr3KwF|mzCLnLqQqChGy!`j9OY#`)+dn4QPA*arPFWz0RGDqtNLT!e$Y(~$
zm9LY}8A(m@OAbF$hf=J6Bs)a8vA|>xBJodb0eL^<H2BzxFa=C?CAzIUEcXaG0_Dga
z^|!-!H87B{Wgo<*m3NP-)i2OtrWIsVJht&r6_d_-3jF3wQuM+$h|?&|9@<YJF}svZ
zny-zGpJVRa92M`V)y)<Y<}Wn53C-KKIWb%L@t@2KtR|o!>d-LwQ(6nll2Fz`&a{UM
z!{EX56Ee4M-Wg!;G<eLUs{lJ|ARujdf3NOB>4TLh;dW#SqeaO4QqD*+C5+`w<}`mS
zM5ANI)B|n552MNT6W>D%P!W`Yl}u__doNx{kJ0Q~dTWgk>w9bG)9(}dLp@2H%2A4}
z_zlIKeH0Uf$ZM3a{TAUqRk4FDs#Q^ifzabJd}(``-D>cT6_xsJRav2T#!XepYu`X;
z!L-C^mA~8+yzB16_@_T>q|5Pn2l`wx^Cb>^PiD)c*<?)76a`GOgnac_l|tbf)WhF`
zlQ!b?ofDw#o0+6plgXF@kmt{jA>|8SG~k_o(&AQz83sV(#zK-E1OA;BgTT0Zj5B}=
z05BM4_U5!1P+T+ifFdx1`H?Hw#BFw$`j`>2ckI7=pVua2X?4`CRYFANPDc$hp8gCV
z%LY5M!#aCdh_<h$&4<R+J7j%>7opTm9-v7oL&aPPuqV}^qx*tbF=-|<B9pr&>~2L2
z=Qj*4ET%aHvRMDtdY{YU9OD;RxK^UvSbuytVJ=Pc7z<U(ctjGY^^(zO890m542Kk6
zj~@pe9ev7K_Hlw)dS9<W_osN|H}Mgj!G&-th$BRY$HbzOYZb-V>HE_hXe5M$O&-g@
z0zaP%hwG3XIXu*-h%hAkmZ8y=t;MwmrhyLO&FvsjN?Ai5Vy{3mPR6FlV)+DHI^wrK
zl?d-+PWC4*X%NemvKi#WoVg__aNK)JF2>3(_p+;+uSt<#`vqAyq!D0Fgft&oZ%tn1
z%I2yiklW{hT_2G5lOTELBQrBfBT~?}{{@Z+fgVcIX_J{=mc$=7LAhy)Au>g(n~1zm
zV8tkp!yR_;AFOz7(R-7g@CKb|jw5e!nS@pX*%TtQdf6jExuxQVj{~eo4Bz>P%a(SF
z8Qlex_SckK9wOSf{IwSf-6q}`3C`*BOLrRx23!Z+2mUseD;D2cp5I!3FYK-OsEw9e
zvOhmWZrgl9A;J10x4(e0V14g?p=b!v-IgS`0AUq(qYII)F|L`?e%~&_zL?MeCP!VM
zD4dx)%|M}fx&0hgde>;7;P*jfe7Pnx_&KGQM?>xN#}0k$oVBFwvaScoiVL&AxYMso
zmZ>$QYv38G+fZ~z`-sK5MsB4#)W(+W*N94$B|CEeu`6&&i|-5IpE6$iM@>wT`a`va
zH@R35CP8W*1r+joOeJLr)=z&7fSZrMezdEysH2ZMIob3qXvw6Ywl?G4{)#k>OuMD*
z(wbCRU56D(;c_T@h|4p}06O@Q_H_WpFrytFEF}F-2tf!;+vo1-wp6ap+?&=PGvoE@
zz+KTBD&eK-L@vcwcHG$MZp13S()v}s-Q-7{pvP~sM1gYPf;6z#UVT~%n9M{kyr~!d
z+SIS;s-eS6V~Lhl%s6rc8id&-EYE4Co`a)?=0a1L4T!zZ1I&I%wIN8|Y*v?$P|0ya
zvAs{xD?!g6d=_m_FRwlfv8PH)&6^VVqKdBh{tMph4&cm?h9!CgL<qR5rwc5>SidN(
zEo5)Fe$Cs;Uj8GMQ@j+Hj~2L3>%+u(!{;}fzY=-xN%iq;fkK?xN~sbF8#OxBhg0=b
z-RKS7xr+RK9XkDOa;*mnFQEJcspGTSXGb5mU85FiA6-Iy5u^^(p6A!ceCI;rAr_3i
zu&p>w@$g-f3e`k@smkvd&5hskkjFoC?y}fm7e+T4oodg&FJIpX;2Ht9_?pJ<obIpA
z!1POe`HX<Hn4ni}^Daa0JQlaDv4B5|&5J9qR9DDBBXZ{SyH$l~z9PG>ci<cV-ON0?
znug{)Bu!YKB=XX#=(40R%jJZG7nX4-TTv59O_E$9NOnPEmoN~r<GyuR=e$mZ6g)Fh
z`V=xB`zpPz0?fN?hX_K%ugc6;45Zi{RxKGi;;?4mG0%@Gsy{$gq3(p`FD-^GFwaYR
zI<B^81mvoVQ-Gd{x-CU?ZGw<X%G30{zFS?tD|QycVttDGi3W;ze=-WRyk8wg37++7
z=ia5AUiAZnakC7K+n&LjhlB@;@80~)aU-yShP}c31TbkPl-~O-eakb^3Q`*7P?7=S
z=#wv#$da$V_M!Eq{5|GE_+k5U;|l0ig|WlM)YQtPZ`jEvW~iYrS&_`CRIV?q)gfB@
zH6vkgp0q$+;5vI1mm*RXq-@?vrBbFV)1rl<=RrTvqiy04P5{fS@~YpHRMcTnHvN36
zOGU*P+dI{_5y9v>KiCk@;?Py*ENZgMs@AlQ2Rd+=$^g%tR?Gm;ds9%ft1ra171u10
z^j$Lktm}OM7?WJ+IFJFZEFm%0vZ)HHLpB&z?l-<g#XViW1hp5lR_ce}sVL;{#73;D
zn1t!|V97?9JMt)Ta)$7l$s2}A6Q$=}J|$NYHFSzjE-*)Y=w|6R<91$^wPm0k0D49I
zG^wL>y%*2l)nApJn!I9E4j_@N&sFW_3x(Qk)m5}sKmGOah?yNEi_GtDY_&5+Y<(V`
z0s|y3pA#X-w{HjRm=g4kA3Q~&e~9_+{q}0i!<JP}$FU}o!?rJ{9qZ!B-oSUNARJwA
zrrpyRme9o_)8+DJkC7sYN;<ncN;{8ou6Kh2;nUQ;AHsM^=EFGGEI{tnDGf63YK+iA
zP&0)Z*p>~=w(A5oP0l(f>5u3V|Ab@%W*Jch=aYZ&;=jOKI}3UAU%1f!ispIqtHDGx
zFb*v@5D<Z0C-`3k`U;jNqEx%>c)%xXLC+3PvV;Ma4<D}Og<1m<YyN`$`JkM~$c|y+
zuYh;{+MzE{j~9omY5?@<A>(V9@|Hc`jRJ5Yd_TLMZH+UDBE`8^z06uM&_Ui=2@jR{
z?mw7kHwKWNG0LYaPZCkxthj?Za8#O68W3^6BvORB#Jw`uN<CvRsOTm%@L+E_w!#KX
zoczO3y+h((Ihe$6=W&RmtjtQXL1f{(dM&4iNf{ds`b#W6n)<LE=KfTOVs2ZYcM@VZ
zcZ<>`4M5^>>KqJRrsPbl11Oz^DncK}DD&FmsnUj^GzWYzkK2fNHgs3FY|5nUn?t%(
zk@(Fvs(!^2&a4|0v^vP2*Y_5*VO((gn=4H>(8SNRc=^6O=x?XDzbI*yYqZvud}re>
zq}abj-4SYs0?ysHQMn-|^zL0e3~h+JpoyL}J0vk1{N910;)fd-qxka)i7pJ@16)-s
z{Sq}+2gBcL>-xIU@7hdN%~Df}i*9nWF6|~z{<8kEFYy3Ha-yCriO%ZYl#_RQWv0HK
z(~wq93Kt+gRb|;b7-K3d-rw*9DhfSDkaW8ci{Gs8^15d~^L4Tl-uq7MXq&_-z+XRo
z7H)s#UK;wZg{SI&E<83ZM0}fNn&Bxpb`Cv3Yt+<(k9|j$er8AJc_t<EhLT@jMeLER
z8lT>7mH2wLgGAb%Hhb9<rJR}5t|kifcPy$b)`mb;V@_SP-d|UZQujX$b-yZP3I~U4
z6Jzw(JpM7jY6GNWAq7Soh01vubk5ZU)UCTWU<&a0;eNRmnJQG-m#~+a2>QU%(Vs6|
z;#>7>^Cx+-qSS?34zzUw%!RkWQX{0R<U(YLy1~;T3-{n4Xp;72-O%pG4WZ%p#wl_$
z)iiwBAvms^8`8%zliUKL^od5VFmo3Q<j&;Ppu#0^mo~59F-_Z71o<;<fu$~DKV;Mr
zbkzY+{F>X>G0=3fHL;TJ3A!SQ8KLI-pgmkcGmvZYnw~zMA?+SB-Ur3kC@t9NKwBa}
ze!pE7Ut&T(__Clqp!rU)!uaZ&C*y3!8^#B~4a9I)(DzN?(+OgT5J+C2NBUf_&3Uj~
zr6L9Mxz^5kpP@-vVHVyHKtkI;Hj<sfWpdfttR|;Zm6#gqIXDk6k%9k@y|<34vfJ8+
zH|a(|N=h0OX$hsJq(cM*X$dLm?yilLNJvPdsC0KoDk0t7B@NQ>t=s3k=RN1=d7kI@
zz29Ho81ESV!PtQ8z3#Q<oa>s`b<L+tMx`NYBR2u$SXI@jX&?HTW4rVjZiOUSQ#rT}
z)XTNJ-}H2spEj3-t;Pv<vkDD&JJktks(Hq6L=th$e0-7viBLZD5nj~x2@Wzzl>oK(
zu*t(Jg~kt1zA*#&hR`x2%r1+ws9v%*5@vI}Ge=XnEQNDIg%(qk&qx;@V?yCh1$q$f
z&yyM3Avo>N|9QR^$w@k-7WD~K72)2*6Yb{gWLdV-F|^0nodJ4sV~G3=ae0&RVE|tm
zjv`93QOAaOUz=AK4Oh>~G*ol1=N1-Wv^(>Y;dU<aH1eP6X7)TNA?2e-^MTV%q1mc;
zh<SYbFdd`vm^3Z}SDQMe<uhFEqFNUA^}_*BzZ*gjq6Gj)tcdyI;6;~1gha5D!kXa{
z2*3-V-+IZ((yA{JiTxB+X-jP!nn_*w;w$|Iuyz}uZ=eXt8Yh6osZ!y?;^<&M%nTKZ
zkrD4zTf7j7MbaT?3M@j^?z6bFgk(#AbW1>_TXZAfxU^2h-I>@=Ki1(38gZ^$X=ZtM
z^I?mGD#ioXFC-i;JF~dAM+j~?$2gGMt3L}7;qP$2tgrF7@0>J|-^+aEv0EF)-M5}N
z4>b^T;vl^~B;;si8LnNvhJq+QlUsd^q^EHRrb%k%&PWDoLZY>Qlo07^eAOzj(dY?O
zi$o3kl>}~0DxE%h00!>!5?v7i`yvme&)@No4fGSs{SKdA7Q~7DI=<2;eMQ9hHYx?b
zfaE`al;9~F#{1L+b!rRt(Gy$UVYbiVSKWm`hh;7NgwPA7T{iSiZI?HYoFSnmKBt9P
zQCs~rgqO9hel~aRAz7;#y*)pbY_m*T*u{F?7`hiP8#H}h*(V0z$un-7*~Ts%s^v`g
zmXoI{3b9qvdFi#4uzU8H(%5w>!_9?GJRJ#`!JHKb&$u)ykVyd9sWI^=ME-EipuW1N
z5_9%gZPyp@U0KP1!5F9WL(N)M$IYgyt~^}{U2T}UV?iz^@v+1yBr9L0{b;&v#lZrJ
zNoXUt=6UTahDbewgp`4XuDGkUsLV>@omNL~XG!j<3)ngHWS_i;y5)&y7-}qQ-q$b?
z8!AbC;aOi81776s*OkWk4U??Xp0)leE<>N^%)8C2+k@_~F<WsbG*0sChK5OUY!xoM
zeV?Q1^gi*H?!%5BR-g9!EWeu#ur+?y9!;zw9N|7Po2cKNp%;><H03?5tigS(^qz9c
zPk&i#%L2>@wq}|t{#AN>;M85XQ!)Ft>ToJ|tf=tUX7MX>^uwV43O=Q?V4l*!jZD*a
zwzzJ{)ci*3<{C~HfcM70ei~o1IF01ubatQG+CNCw^eoLiWiX^<B})4}AvpLNg{`D|
z?8yjKg9%bgtKTgBka<d$zx%%|71GpCxU$-<jj}tAd}2VMY2BUD@8rDJ_c+C4@7qnn
z_CHF|SVo-#CgkB9yPU39f88V%p}Nz%Zv1c_Ge8qwwT{M}IQdE|8_mb{OMFf4AJZ3D
zr(ZxQlMY^*mhY<rs72sh`K1S)(ct6O^>B;%_kvY5wP6aEZhplF+Rw=GbSu%?t8?*k
zXpkk?hg6Y8xH+ayRk(qBWAN)jUA}vjvwmNhUCq=El<#fcj2!J_?RWw4V=Y)}9NQ`-
z1Tu80TXQ}#kL%X;zJe8k4FI6Z%<qsph}+nYxMQd+NIEov!`$c9_W=fIWVLxu3lRDr
zEiPJ>a8>js(7etghB`^64OU<78Il(ePu4k>ZbP1z%XE;*%d(o7d4#If#e=p@kXmX!
zRDLw}!<M9FcOY&b13nH>gc8`TSt-B>?`wNAj<9x>>M;4RD(!1oNda3<w}Vw&KT4ZM
zBvS2lw6^==1psq*7^3?l9dx;tK3Ep6Fx7}{m`CfWj-7Ry;M=C&bKuz>1Hej3mFmO~
z09-K859Q?7sKC_!NC(WHet-5qK!8!SB49T&?oKutGxc<kNHQ%Y|0QCWvj#*YD2lB4
zpQ|sK2%l*ztNG<@#s10)pXJvwp5Z~Ts_I)#=D$h`w7tX_g-E4f#(@!J+q`UrlJg4j
zfQ!3xf7ovH;E`#oBtMx9D)(>^Rv=m<pveoJ%>`4@Qtl!RJ3)6(+>9_V2Vo4r`zfA4
z^|AX{@Sx%9H5-ux1m8%Wx*}0WQV-iW$Bm_l`g1xTCRXL-aGU8-KzVEoP!$rZI%lyq
ze6x~pYIzJ=5Ei3_`eVK6kM=8M1t3ymW>^f@h5iXHrsf(y#YD-VEpX9$!CrH|*U8o2
zbuGY3J++xXp(d?+hG~f$lg5M07ml2Av+y#Inm|qqzHf>Nm;P8oYd$S^=n5Qwo|R?B
zQ1(FPFJnGPC@(+wO8ayBHzE|e2g_Yjor~<2x1<*$YpGEgi)CkC9<Js;C;`$h?he71
zgCAw>EUJsg<Vn9Oa`du&vS!JmQJ$|=-+u0OVie>P%1Q1ts$ayBw4x#D`TE`|lV#sj
zRuGK)9xkooLCJuk@I!m&kF5qVZd0-oAFz9;m8s?thXSK@V0fYG?n${El694u>wkFx
zkd&<66O?P{(7LHW1S?XGT@j;tgw@HTvdU<v(4oi=2;^|psrS}urFQWta#9_xpl=)e
zUL?1!+I{15@xV+GT&hOm$=3@_eNbIvOS&o%=28K36$^lvc20Kdu}j5BlKprklKLW7
zgys8rRDsqhzM=2}?mor*CjsrjcrlO%fz?ZgBD2(<FD$a?RS~J&=Mx{7GYEwaEe%Un
zAMFlSn-C{8zrME*btK4329@QEpr(&n=}XBAPGbF{2+mTIwKoU2AM{A0>LW=U(Vi<V
zp7He@eCX9cqF&%PB;ND_6?*7yk-AliK6&f9^{;y9zCffhPhQx?Co3(5%$!r#Ouv$+
zDd$CaprTt^#jy0-tEgVy`ef->{DAhw;?;vN>vNlB&@#0uS#A?MQ<9<}T$lwdBbhzb
z^e{&SQnRN<vcz?-8r2tv$<qHXF5uL&L_1x1KgU4#-HgqLeMX$enbCyp05N%NTvT<)
zu#c@}j~7;z^xSI_=xVH>Nd_nA)hEVps-ARBD&$nXZh^ckA%{pya3;B&-<z>L<swE)
ztSNp#j4}>{xt2iV@u@0DcR$?SzDU=jGf-=}NoZd|#$G?E%BNX7>uI`}CS=74>5rh>
z`^0OSM_sLys%ASo0Mal7h1tt24$#R<(0c!>wlYx69!dKA;puATGMQI-E0NZy=rxL`
zCz7Oh>rZ}p8}D>w^Vls)gZIo@wfN>_wQ?Fb)P0`s57fHJL2MUabBHx<HPxy8!kDcP
z9%uE9sbg+S-{0M>AdJp&dA>(_W(s{Ah^LH!w$9UIL_$qTsDjOy4fQ^8<T&WH83OZ1
zrBa&uZ3AWdw>`@lejhQ%ulHm*wQ#YqaT32w4CU?OdDG2vJEwn^l$d@LG#rTS^>qf1
zK$G2y%>&`I)&vB4=~Y@f_DG+r7=|Xh^#xaG(K9bQvDmswpjP|jd}@_6%Q*X5&1;5n
zI?v|3NC~YT?~%OCgwXF@t?J8Sx9#P8zCeRV!rEPWramNS(QpxWj%S3jy=3d+Abyb$
z`g}e?ZG$eAidY+c0CZgMoHAW}<LT{|pk3Z7*#rdIaHq96!u^6+w`ti4r*wdaj~S_+
zxI&Fl;K7RE-mqd}K3csuNM;#i1KbP8L7`7zX$SD-a1Ja=)c&l7s!)?m%NhjRybc!W
zLRJkc_vNf4$$+84i;Nu{Ll#4JPk)c<&X@MjJ_lxhzNe#3Xt)jGv4%Gd%P{l00}Yll
z23HNkL~UL~=&`V(pB7?<W%N|xPHIcP2qUnr-IXMtO4K@_7%vOO05)r_2Z>e2xi;om
z&{V-hOzuqVfpRmPu15#=_AqeP$MPBzYMk;tKF9Lq=I-XdNgz@rQPep%VRPehj-<SJ
z$MMGNyP))$Uap;a)H3oi$)o*Li*kjSst?q{Ti?e?gw5o_0>kRq^de3Is{*$<;#k0_
zkRqaV&mqi08-4j)eV3pI*DnGO<*l&_M{Sz14eJJUFB4Rt?sE;WF7iv#x_ZUY<PB4D
z4KIqcJp>)@u*2uzJWG0l^ZaH^t?lyZs9OJOhu3$06A&aK%3^D*6vr%AUB2_J#D_M9
zU3Q*U>N1p&8&)gC6#Y8ir}mu#<$$U_0lea#74S3fZe31k9}|1V^F>}|L-aaH_Y8#{
zZ}j1e+%=7R*;gB?_&d>mU@bk--`#Pbn~<UoWA<NMgTS{uU&E4S5U98(c(_hKKzD38
z?rJ_UT+^BK7U+54$*}mce`Ea`uk`_;aJz-Zg1hQHgFUk#&8gT;L@zQDdXhX`3iqU_
zPtxW{h32y@v#M?Q>esw?fwQ6Tid%&9b(L#Z)NXV*h&E-04^i`Y=#zldPSYjEyBhb=
z05~xgCbYGk(1x<$&!;~9aA1`qC2nVtyF2wAm3ut8_v5i(Qy~yzabJ#BW_ry^@Zmn8
zWym6jdseauRBp76%4bb~v|b-%Wf%jbDhot5ZX~tS?Nn4$AH;nOtG@3)C#;1H?YYs$
zW$Gzt*fYXu1m$Q_C0}KJ|D;&KM&w7kZVxszcZCgMhM02$ndOZrYNXx1bhg1qvr5L5
zSC?nqtUtuB(k2;S3s?B4;TeA5PTr#XEaDM`<9`h&LP2MKHu!a%&}$rU3z8c`w`4wy
zQY2O)o)>DVgd>?ME+z=a)1Ke)hU3*RmG+m9H3|WJAk$`konSwasr`X_6R?#$8igrt
zS9Xqm6v&@CO-}}H8n&29jL7Q9hD2ig$jBn7PK;sSfb2irMSL;{-0V<_%uK|X+Ce*#
z2pM)t?<hW@cOG<2%my{K40!=%ll*Zh<Fv$NIGk8_%Ye5HrLEm<!70yyEZp9_jN@A0
zGh8Z8HR_+sHDPCn*LW?v&&av75}G^6w{Nl9*R9|&)(Fg+jv$2f8mTVcIVoOA;!1fJ
z5GI-t809m5CD?rBCBrxA=nN;pbK`v5TYs#Cyg?W+htPZ(qM|Z}D@<GJ)Y-`K7C;d5
zGY9h4{7z~xL}gFjiW*_K&^PrsWMomVtk+`yJPj4lX(<178m5y0)1#B)!vw0KW{&w*
z$~e7gNjMyq(C{2MagyNRA=^FH;Y#J=F17-UjrWE^lH}_5vWilvD(;_of8^}{_*S^}
zNE_lUPF5>@|7+gjDFza%m&HedQECGfK8;=jE-G7UB0Joq_dZWB<ctxI3U9_DC&>f*
ztc_htWYOFk`o740NKLo_a8&M{6Ez{drE5j#NORQNizsS$jwP>`^-CXpr8;R@CqU-v
z2wF;uxf@`iLwHPS;Qt+UPvvID&6(y7JPTYPRAQ-l?cIX2so~1_VW8_lT*j=lEAW}Z
zeWWgRfjjwLT>>E?NEXJpxDj1xAV3>&s_)MkXmE=$g>J@x>rJ>s3zfQKCbs?*YPur6
zj`B$9vj`*faZ1bAry0|GzQz5^dl$E-f^~Ekr9FzK_dV(AcuFu?L;Z5p*6H4N;`Pxs
zJwB!LjOFZ$Tcxv+;2ohzPbd$cg-s*6_^<Ey8(xng!!%{=jUdT{pSYi#ej9n7XKN-O
z@!%8r+E)_aG6iuI;NNFx>XttoWmgMo%~+0TZjq$MJyV0H8+A8<jdP${fWx5DEoPFd
zwNSHZQ|rmcW@T<!sjqQPDfY<>Q5Wv`wft<m1dX@i_8K+P9U;p^TY?+vBMOePRN1<(
z7UXu5x=TYiM*xp}kwWyYq3}i2?)6<E?}Kj<9Z~)1o`$S34{jiTo##zS*}!fG4L4GP
z**Kh&-EbH2r#lMinL^{TzvAmZcGSS?XD_g$MATz+A~=TV8RY)hQVd#j$Rf-9!}f3&
zjR_8-><8Rxwh29WU+kxq7=AU`D3(-?+=-xospiSR^Vb3MpWm&Efo%!duDO8@DM`Ky
zA=rGU^hl^aKtPsb30%LGL{O8}$Dgh~L@gW*Vu-Z?K_Ajp%Gx4%^vm3BjpV>><HVu%
zQd^}!J?4P#^UU)#rMPdIK|=(ywP8mQ_)Kp-6Cg%xm6Iy>U!s@)#L#`MXngK;14d<l
zV;9ItY`N2poxz8AfL3mG*XPXm)T8)&2=k(f;={6u{d62|#aTe6ygwOBP&kM<w8%B(
zF{pZQ<v3XnSt;w2lA5Qg>j=#w=peuGufiYpDdl0EVNkeh9p@VX6JNqNfpgkJ-jR+(
z_&JnphVcLT{QdWL(?pShIp4F&6NUWOw*H<Vn~#IMV8WbN`UaYB2iy*G9>jm9Bu2tp
z#YUBkq}0K~h<ZNh1&!vdquJ<KxWadVXzmiulwUy;e#Y$W@W~OcH(<1E0!CxOE8Ul$
z-dg~wjjC<+Y85K(p#e6&b3J1hAS#qMB4cLB0byu6KDZFJtA%s^UsAPuRWK$2v&*Ro
z<L)S6J*s~*up>h(Wdom&acH1tRW`xDJ)za>AS5}VslNYf8qYsLt=?=5umC(iK_XKx
zhN-jskmu@tYKx;;n{EDY|1SdsXB$O51=PjAyvY99cYDJjin!$HY0Tes;{F62iGkz^
z#nMa^|Ms^2rk(T8{Ry6@B(yMe7M`+yL%RI$&GPTG^A3o#sc}2@@0g2!MtgUl13R?_
zuUzQw$g_X8!bT93!16{`>VNnFsv&F6DY#bZUu*iICG}Oq@Bp@^+(0HV4zB5%3bNb3
zy9piGLD_Hb`qQ6(tJS!SYy_-s1Ay=svvkthfmU)kP+nYKQLZkTfMPp{0?;A2o3Xf{
zJJbtEYkcosRPpa5_ulJZ3ANI*o&WX^R);kLeTfCIV60RSI$2D<Yy<r@Gl=Hx2|x&l
z=Q|GkZ1v^Xs$OKzY_H+yg4@T>q?z+JKr!ZY2b@1kfFG`|X{cTPz8$JxgLTw_Ka=@=
zN&b&#e_v`A3=k*>SJL(f;Y$}XK^L2B0?}h<K-eA!L{USC$2tZ4a?63^wBzYT%H=Fz
zgIPizz3P`|po)YhrEL6ve-p50^^irU`pyKnJ&-+KgdD@cU8uST19q?OK>6348W3Je
zfmuVfKAW59Z~H6F8rpi+L)o%mJkFn;@L#Wc`ZWSKI19j0twHq(9Iu@VBC`PI)1?5!
z8J-4VtVu`bcl7gs2iiy#x1PEF^>cM2_uZ~P`)&npXAu6y<f;X%!z}>I?vu8?pTGp*
z(tgC}8PT>M0}<_6)#JMwabp1Zu~h%!4)EelK&dEzWX`|#84xp@gR8uPTj_edV(Ln0
zzCD%qxotD&^Xnc-f-$5ZFHoL=4%nl~azKS21Ml2Yj<xJ*lA=7=Oyy12nMVOXaB7(n
z&lS^WF}j$B>GtNQ4~ZSiDehKUMzN|EVDo%D@e+E&8C?Le_W@SQmGt6`j33R3Cg7wr
zQCMHPes4J6lY{iY;nlP5V9jbD0$spQ0O5^;0q~TOiolCYDG1DQs5o|;GM^+<2?mz7
z!F9lZ)-iPe`?nbAdnpRS<E2s$)BC@SuOc|~JD@o~QK;Vxaic=DZ;<o$0#hWe3D9!N
zv8w~Uy-{Fz@)GontHOaRKQvln3{WXG54!-L+!`{r-UNob#X#mXKo1~j!0Or!I|kie
zdQ@{Qi=x0h0PkA{m_tI6hkIY?ex_cXx8qW$fD@o<Mg;^GfVo*ALGCJ&CRN{w1rXG{
zI)KSXjr{7S!|1eG(8VFMdn?55tE0-t$c^B9wotmvBDq5Y2%3RiF4@u}hboYy#{foP
z3bD-dZj05npFv8OI^ei0S=ac+neht1oMPgYQ%)0QG7A7{z!ri-+U@Vh!_cULZCZXv
z_Voi2d4N8|gLTpzuVjp0#`XXShitk7(+at^S4n?g<a#-<$n{zkObLIz$oFV`W`X1;
zzXb5+zD~O=U_*K{h+I?#xYf^MVa&epF~FL^h1`y?iY5!Z4s$xV)RP22JlOo08F9UD
z3}V!&UE#i?QU5i0X7Otjck4}+QJ__sYHOm(_yR^NO+W-d8;+zFuQe3yyBm&(X&|#H
z#O-~5W<*{9JPf{WfK}B4))?L~MR8~%0WJ^&PY2UW28LLw14n%v_aSZhqn2QX$nCIO
z4>~S^R%2q@>!>DCqXCc)ECDwB#Q-p|a5!lFWY+=9aJm@ic7pZriQpLf!*Jn~#pp_J
zFk}Or3pEt;Uabv+8;lW64TkN25EfqDmVA%)5eu@@%~@f<Rhf`w&U*pgvb$8|M*VL~
zCqi`>WLByIJwnCbBG133;It!LUSOFlX5$1_a?ILmKPizE8vnZq4r&G9-j6{(d!J3f
zkKA|^78J;vdjbJ(+<J{Z9b5>Qtz73r?IVLJs}9EO+)2f|i0efp>NjB@&dBj*X~e~k
zhY69K>p<Lchaci{a^HFbdwm~xnyYJpP8?kiBSp*WECCQO&^m9XMgRk#S_osLA!*s0
z16*%?5ahk8Ex~)qgs?+N&qlKr1Rgh#S9V-Dh&_aIOF8IwpdLZCl~dQveE=)I8qzE}
z0UF0RXvwNnJ=;blZmAU31$Jm+_tm%gT7S<i2RKlajC50D{F^9=gp|7tnOFps@uGS`
zglg4%2t;hom=eQeeNh?RnrtaVecAkK2I6WGq3$2`^Qj>hxdp>_z@(#K{jq$K(>E|&
zr91i2`(j_8PJ>D_$OY%ww^kTr#XgY_Osx}}>JQGL(3(?FU#<ThH^;Bl2etz~5T>~&
z2vbtKPU>_m$kZoUkNK|yxzi&!UXiH_X#P4JRBru;4`vH}u-)rwyT9k!e|^P`*!Kk*
z?uJKYW%sqoZw9tFQP2AxvJ<Tpu2*z%b~-R%odK;IHpeYm&VH7l`rEKtKp3B><bS*b
zuwTzjFrqWb6R5TrKijqNQxD_Y1`C4T_S1RuWQdCyg>DrZnjw$s1BOMuf+BI5Zg{0+
z9^-rv$?iO4!dh*%9`yUpy`ly?_v)ibtny#)+_$*WYOrj8KjWS%-0u^Q+Q$(-?*$X4
z&>)qDB{<!xTNbXcU8q0P^oE?xNZ>lCTwE&OwD#+991tVTLZVtg4scRHngtCR>q;8Q
zCu2%3;9dt=|3?C7)IM&Vxy~Q91EO6Su=Zz=DC6U1g0&>Gt6h`P`VR@+8xo9qk2J{s
zUS$1ie(y!W&_mlv^S@DyM<+=Vw+aUg@gT5ib-AI=8!}PJ1vu@b`7d0pLAYq0u*@e`
z*Yhnzb^%)P@~XQ?X=BjAhsGh9Yf4+7{P3pZ1}?UVM6gaG7-Y)DeK@Ex0b1Lt_n2RS
zE7$)~%&*>st_*hx^9+738UD4<I-o@Uc6}C_VqA!P4643Spmj0@b5cSfNZB&?N(t1Q
zP5iSEs|FOHpHJjzeT-^=$=cUnAs7@TDh^8s*bcZ-mH_AcI{df7Cm=sqfV~2V&hBgD
zuK+ut$CS|e*8q20Cfcu({q+E*g9Ik!Keb^MCP=t!>epHv@12-(JAekA@qWh9zi!+=
zzCNV{#p{`F@@MhCDWQBsijko%dr;^&pp6fKJ~h7<V194JZO!`QK=nvBkPkGB0(%st
z{VFJsgQUt+v-n0%jJyJ9naW*n<oE+Kq>P6kL~{?T8>;@YCfNiW5^VE;U@9+{w-_)m
z<b@TQLybpU3*?@vOX*z1V0OrM&9)E}@GL>xuTXi+*9ua6e|0LPA>D+Oh=9DA*Ue>w
z3B&V;tndLrK_m&pv<VMIyx%|51$1M)$f;M={68P;zddJPT`_76q?RG(j5=V^7WkkR
zgFZgZ3ZpO)qKcBz79~54fAa!m#{|S@=`p0)0d`kKAmQ`^^qL&1TAEdNz|@RU`iCqF
zG_)<_)Wa~0EJ*D$0XR$6pc)LH8~*8oas!Su3q-Ah^x<lB3|z<}xyQ*s>ZX5T1FrKC
zFMx>T4EL@O7+UfZ6vx4Ln$C}Gg3o~fL3|odfvsZFjwW8NNAC-6A0@J6OIVB%z;DY1
zF<Gc7K}oZ_2}vut$6gSeE#33hxPv$ckEuy^9-yVhAo%%1oM**U_Pt|Je2jup%MjGu
zZ>B(H2W511CKQ}KEI(P6hg%@`BpYCaLWzt%;RXH#(CJzL*QY!WmAn68aU-F{E%>f#
z@OO&~WA=e+fXaQKe>I03mp0i~c&1$p5yOS}!8?{85{$3!PJlyT0*1!rLhY>bFX&_e
z$53B8Ct(_*!GqkdTMIEm#sTtg2wf8DASWU57l^YN;JV9RZwq9m<>oX6%ToM!u0Vqj
zglikCss<HTLmyJTPI0>c6HiJ%9d~Z}jU1a9G_UT?gN|?R<WiB86tp7KbPe$97!WFh
zIHUxjSZ@!AO@{TLF7wm=xkXT>VZ^~H=OAO~c6)URahI=Df_PW@w!0uKeXjL}Y5`E3
z+?o-yK->|X<i(-jFWtjJmYaM)*U@r^>u?CK6OBl{F~#e2O&Y~_erK+kd2LPnw{V?y
z1zmBjhN@(L>2iNx@qfJ1D+fP6o4G0n`raP3=bnx=z-Ztz%is_*^6)0^4h>85Jk>lG
z>Pz*MM~gvl>>RJns>lmaa6gJ^<8h_h@U3i=Kz%#iE=x`H12TFU0~xpd@lu{{%0}w~
zI*IKJNL3Ukji)qfr^iXazs<iJ2%uTF+emDw#RT^Nnk|`op}^IBAMX(KeJ22;R6m9b
z&1DdO0BZZ|+in9*Y9UB=Jh>cAZ))|D@s#F3i^Bk#S=l;HYeu^TiH4Q;Yy&6^ns4I5
zBgpwUb(~%;0q~FX<jeXz=ufH@^{UH=|ECT23fgd=t5wzidc#eN!*-$DMbpw2@O9{W
zGK{ySUxvd|2FC2Dsw0Wt=YeP^5)BdfkM17>3^3r%!vMS5P#+sWkHzAtRG)cd!3zpz
zRhA9M7-O!ZZvsTy;KoG&U4}EsK?riK97xs=A{~zj6s!Y}fW}RJhKyZLBz6{0bOCbD
zJ5lVRej(objQKVFQs5orM=Y@U@6JG{eO+k|3{@FkelU4OOWAcTg^Jfk@6PooS~WtI
zRV=yngHPEh#o#dsPhKrg5oJ3#-26Qo2}9XP*fH<T-((|NWcb%0EK&nRr;TW6E6<;Z
zC%%FN%&z!o<xe--A8EMv5nLNggM`Y*0Om1}cx!?eWyk(4fd1DzU6p~}piEhM?Y}0+
z-@iVPL$36D2+-9Z;MSQNa|LwEH;8KyFi@1bO#m<go2TqgcyF(P3rW26D#PDk0EUQ0
zNaqe>N(FQ$izR=dVSElrIJt+Osz6S@#?kVeGE4Q3f6tM>UO^Tq`>)URA3rE~aTN&_
z81xaIfm2|<`~sjPC6N7h8`yD)M}+?o`Ro;jyc2{sblwyF-GBZc|LE!xw7N@#AAkH_
z!2Z6*zrS*z2o<?KZqGjbb@l!yYl7;=2dW2mS2U)-cWV8QTk0YjI=>~;sxSZY{QlY6
z`073eS6^LS-;Uq!Z+`;V{@WUeT-}B$<y_}j&EG=rzuxiw?CZ3*P{I24S=xU<1^=~R
z{j;zC_wkFog~KBGmPyV41?2Ru(*N{1%;3kn%!h1((&$Iro^uGy2GU$Y6X3M2r$zqv
zBX7Ji*_NNpOIp8bxK5}X^OcT<GZKTe;~iEMqxjc&zNi<k&<%qa(%*2<s?Zpwh)KvW
zCUnS~;R-7!3}2mHUTU=qoo#XbC>g2isp`4N9od*JS*aU#FL3o7cK&~5F@uHa3t>Eh
z0E;r5PgcT%rPy^a&H+s?Su||*&VM{9fBzyo3o10%wh(}2VIj_v60l>4a}ly~i2px2
zTS#~$+{gv)QyXn9a+$O+a)Q*2|M5HX1yPRISGeAf?QeqgfBuy0UeND*H6(ld4}brE
z|Nj53|Nrj4|KFcK9j8<X1fByp$PIw=)m>qq`T_idyLJxS8a@E*T+$+YJ^v9x*x4{9
zf$IiM8wo%pifq@sUH_TlsZnEBUZMTxQ1f4zdLaZ9CJSrk;cuRAgKM}A=oSS)?Epc5
z4b6eUgP!q#`rUk;ITsf|8R7K-9tetjaFyN3d49npc+mTZjr!LJm0+QRbYS*_8Mbo*
ziQa7>1S0~{@!|j^NY4B@*&pvBQ*jou={N>A^)a`;v+1MyLeP6F`&p)x+9*k8c@uXK
z(2180t%oTV5<!j=#07c<Sz;gj*WfeXmruNJrX4JkLhMV9`gFTTRH=u=9N9l+<*l}z
z3`7eUudN0^WB@QU`LwC-8tl$i^MkWM*v85qn;Q-(HcvP=|E72pkwP}4hd~$2Ex&u{
ztvQnrLoH78<)_aod4|z<6>mVgUh{@Z-_|N{7ZB(I7xLC-jS6sy6b^#ADTng>d!T=^
zHSFcO0J!+r-e7+3Is4@!EFREYJ@(2jq^_R;_n?l*TV)r}2uF{jni*`6y{%m)CZ?CY
zA=D`Jc&8JX<B&t6Y}mW+4>%m4%*2>@00ibpbQ@^Y=s=&~YWGVd(983ZJM@a`19J_F
znk&Hj#dv<&|F#MAH3^_=JPZfb-|mk^pTZCXxWXJD-CKo(@#`p1UV;&VytQCP{Dy<f
zD7X+bMS3032}ceM1I6;aTmW5D6pVySqt0G}!O_9cP|%}Rl{rFrbaepewby<0er+=e
z@RK?4fDWp$qnSP$vVvxs%G|Sj;w(}1VfW^iszfJTu>mkI$5k6m^YGI4uxfGa3Xdd9
zFx%4|Y~L~x$VNjS;HN)-+HB~xLNiN=A=chK;Z<SYjocCsTHV(7{I~xmhsL6|$-|^q
zjsTQNa0#ASwG>c%$ULY2*=_ArvTbN~BVC>62r%&Lhte+#yf2sq!QI#DQ-u3!nv~Da
zeCN8H3A$U!&9c*wJqIq=JivxZY)18iPn=F>0Khf7f`?rXhG}C%^XHowv)|z_1BQ1)
zDGAYyM#FKlZ0rKpZgY=D7r^6Y7Unz?8or$UE&!S=XN3R+41xyKvuW$u>_V_#!*T1q
zCV&>;fnqH=V!PdK5O}VU83QAbt2T9J5!lyV(DUQhQ+j1mRBB_#zBq=sP0COO;PYnC
zF^GA*@$6U4o`!;y`;)?LH9h57wj7va-QY^9T$*+Ifv1k~9XO-DGfYQd#)m-d+0^z8
zZ~?4Ll@y22nZN|&#82yX6dTcvz{-d)1*lPXBE8QudjvkDbi%nZ1|ZY2?6OS#8;5Cg
zddetbcjLk*fsIkKBG|n<06X@M_vc^tB8yr2=x*_@9%2q1oIWtv++BRSO{Mr;&PxFm
zow3LJY+~kgvHTzmU_t!Qfb+R{!w8vRCht_wpBkx?qGTSB-^(+=5LX917N&wB(Z8wH
zKGL!qBa5KJ3V=;ooCJBChlEJ`9(N_=vz7~JLaJUC6o%XvK#}s}4BDa2VbD9UN^WnW
zq%G~UQ@0l--yQ4m;%2AzDa4^cC))1V0;kHMk6i5#f}(c9tv#BL@J8oe>3>NiATpf&
zQY<P(;TWR@V-V&#5<<a#e$w$wJFVh$J*Ngyd<=Pw(OH1`dvhpM$3B?HuU1}7fR_wG
z!x9sD=0#rF@XKT|m@NQ-uAwt)^na8Uhm@7Q3m~+zT8u*%THkzB+#eT<cZJPZ3EEqM
z*?*GOmSGwnAInu#Fx~3{JM~hDv~K(@^=zYW<G`_@WL=peM-&!XDerXLa#u)cA^+2Y
zasZ~hUgI7GQ9yf+`K=rDyt#!<*YRPRsHPouwQcErdq-dg6M(kd3~H`IgjBZMmCN}{
zf{`A+4*OH&k2nYTF!~Obx4?2614e*%@i!=noL(2;8{z0XMnvrNcr0@*sd#SEu`GD8
zAVs`?^s&vZ6@nME4!|NTK;*TtWwe%?Py@PWUP>@{heHF~qGrci0D)2zfneAHeBZ24
z4idqwatp&Ddvv~_vcwmGMeAPT3F_Bd5=CPn+oB(xXT>cP02n4AqL3r8z2OVy)#<7W
zE=}BmllzKagYM;N+`iQ)UA3@{oM3W0M1y%JYX#QK$#-Y|S7_$$(me9g$T26t3hJto
z7{-kq40=hDS07lOd%h{BZbW5TU&Mzo2_YJfsh?DM(Id?sALsA=Wv?EK&KH9b<^dW3
z$a4F|EF45=cq>r1y#Aau&b-iP(Ck~wZJ2l$WExoG*Z^STs?Ae~+gra#=I@}cdOBQD
zKcoaZ1|hU2unzK?JO^3A$=p>PrNwU+^kwc#EMRp$VuL)ylD9pwLrHQ3oA;ccImNu7
zSuWPM_q3Xb<8~*KfdV@#SHkyQt(&mon@d1XyEU^6Kyu&~s;;@nE4Qj|tMvF+e9*fM
zEsvnDN9-PEXDMjOJGyMU&j{>&38YgGhlDmdSy0m_7|grhoC6c&u+@u<)|_*9ly%R6
zBK7AAEp964hE3s50@;I_i}G|g<Ve`r1GTfLUwXF-OwZAlf_b&{jZ4>rk3~CAcigA<
zDUZy@u3a5@U#%>$4R_$y89s<wVF7i3)wggN%w%$+h@!9R4l!#no^>60v??qlTxPS+
z8ct(F-QTnt@~V92JV{H+6!KMVha8#aK#a}__T|Uc_dz>MnYq|<zC`fU?}oE=-3$F$
z^RQupkDk2{>k#_ULJ%9z7H{Or#ZkyTC5Y42PV|V}6WI8%a$QOZhPZQgT^2s4*W#=T
z3?XatPl9-A({qkK1{Y<q`{AyC_Au;crj?uT0F=AQ!%OYEkT!_jb#HM8L^hYxc-uBo
z6f-w4I$+<8E!tLEOX6*)k(+TeuV~#l=&;H~UI)3M@X8jpF?}3+{>q^U)M;leal@^F
z;f#k55j@9uM?(c(4_=QX1G`AH)VgY!>ZJQ-{W;f|%0DJG2e`;yhO{d0$?gd@2`8R7
z4!8-MF)Dm+>RGx|UKAlMltR8{4H~ubHPG(g*^4OGvja|@4)I_z<v{)w(M&T^%9j$!
z=05a0$@>a$ty9XFZ27c#fhg-YTl`hkYyNi2K%}(WzT52$(Cw2mK3F6u4M*M<uekU9
zOa`XiwJa^d$sNs%Q9^dXnKG`^YZHTcsj3*T6SaV{)ehSlm&X(TEEgAo8$thT$&&?0
zeMXOWOFfv7J#n8*ZlUy})&$gJpHi>dlXQCpV+5&wNtT5(^t_W5%JSo2LX8oUM9ZP%
zk0k&$(jsyDqK(1SmcZ~12dnWCDqWE5f0D}cL-XmGGh|?2ec=AAEUm5kLZ^L^<ry7d
ziOK3ux~;2We)dzSK3GqoPWl|c8J#!0u{1dlr$v#TU94E;l>BXESo4MnJCIQrB*xTB
zm|=1T`85#g23-p#>CLwT<d>CZupXq43@?JDRu;v#V%jYShmDR+|4GD`Z=gK6!yezl
zqX9C~`NNmwPw|F1K%@?NoVEqyG9V7n2lEr&X%$Jx`<dn?J#uJLZie41B$x9Ft0e2g
zlAAK*h`STq04fC6lUJ2t9*}L$QE}FQa?8~8*ao>DcU*BhEs+tQH=GdHm7UkCSM-|e
zN!Oc$<Ah>IQ2AK7vt)#V9ls$|8NrS{Cls-F?F-rx4)W3CB^<A`$(3Z@7BUOEkE?{n
zEmH7hy^f>f*=xciz}!W!A+_+y4ccnAk(y~jP>>N58y(BZUg5rOya0ok9im91NeXG@
zHLl~tQMEh=mYEs$v*)1pm@_Et&qN=e0_7fZlV?1Qn79if{^Jk5ukImS$=U<keQQ-0
z<UNEDw_#Z~q%Y8w=x6+`ebpn5zf{nBeK=P8N!6s;f9Hq*=gsl8w#i88;L1t4N-Zy|
zDNp*WC)w}oWlvo?wWjAIsk`^XLlzKEKXDQ1u*C_Mn7vDj2*)Lnno4#9yyUI*xE#n@
zr$#S~SB77_^c<w<VYE1}=BrDEZ`-~kkVV|Si-<q~yPi%>2Oi{6UArDbMOVPPSvt@J
z^5Vk$CMajVEzq}zXhALd(U&=S+AhN~t3$hUp6^UGx9=7h72~cU?%RMJS5bEeTyqUT
z+9o?E@w3?*F(5<tB|eKW;q5?UnSIkhGzH9M%!o_&u{t<Akm^OPg5mEt!yF{5!lV-`
zrN`vmO+jt#fV2ae6YFz%ZdcT63#*xtKkpd3;@F0GkV)~0>Xv}UWsTlBsCV}5VoN!|
z)Hu{*<>9weF|t%u&3##d_qb%v95+mW=y$x`YzTFpZU!DjKO^%;zy>Kz&x2spG&8(5
z#T(e8MsEAu^iUL5N^ES=U!O@HuM3<AGKfZ69!zO`bP#P>_yrva1x=*}YBeKAU%iU>
zmgRrzW4y9$Qh6tU-)GmD0^iH<(%ZeG7V4$XQvAHS>YU~cw6&ldZ2~lbP0!q^Q`U8%
zijl*r1@VHTCBYtb4$r!S?RoAghk=x6A^ZC0JumOFC(ankE+uTz45q-tZXlf+LbRfq
zU3e+4snF_#!8>0W_B#3!9Iewds^mSEyeG2WZmQmEtG$CN5^2K71j;aJqi7>nyj+gn
zc0EirRb15m-JaR|YRAe_?s!HZINRW0vZ&jCxsLB&BZ#s=hV{V%B&q!`Q&z~0J)T?k
zLN0+WzP?YH>;YS<2osOQqP4$MosUH{3L_fHkTLKJLkl_lm>+|x9@wK^i!;o}Kf2?6
zH;Q`L8kggy&Mo+h9qb}hA~tA9NgM+b4XuAUmXAHQSd-{2l2+%JT2P++z>Vy9Exw?7
zor(Cqo<3qP$CTz10ok4_KLu_eU1J)`O%C+9=v8Wo+X1T)*LQT?H<t*HK~*s$fgW?W
zDo_@&;v!6R4Km*&mZi4HoLA6L2&UT3;*1!g4LFTCi<$aHgDkr0D`F0!=FWynvX926
zfVcMdIib-wvZ7QYG#_M$-WM*1B5=G9))(8xsb`RLbk)x@b6fY=3<z3?9E$OXP-)e5
zvKFw5RQr^yeICm3(;-g2N(s#$(N60-vGz283uB#<#9hG=<p|{wk@M(g9pF)D<%!qn
zKWL8Yt*QabN)b09oNr@#yWsxxVa1GLRZ|qI&mbOND(%ONOaD<sgw*t-NwLz9ven{+
z8k!r66L?3fHre?hut*2#cUf$qcNz^N_&#K|R`1zwY|+7+i|_6R-_Q+6Zh0gw*5ckI
zlbqln`7m$QuBZ#|F5(FO)Hd~WFuPXh+YxoLau5CEH^gtyLFm-p4x0vd3(YSH{WJcj
zT(pfb!$s&pxH%m(B!)Jb9q-Q~dFSE8<k8o9e^NiBbuiTP_lW*Zygp&H;ilS?>z->?
zms%TFPL%O@-oV!8Mnhq0ZZ_@f_u*oivJoDn!sK>eX{?YUYhy#6Es3(`&cFF?iV}tX
zZ9p2cio)o&S?FV0@h=W8z~k{d$EcW>LQ)v$$2ty^?w*PcT(bmonT7~%AY5cnN>6Z!
z=^V0t&<fx6ZW5zD83v4o6|dV7H>YCcsrZ^j;}c_6B<Nc=s}D<<z-;4(1SAY4D!TGi
zP_IkLBewjHZqf%@WO=N#+z*#(PNf*81j<BDf;S&T6?j~UPi(gp6JVkq4I;H?TqRVK
z7~+2tPLED)8?bHD2Db`c>l$Te6A%BNKU{1Edq{B%{OD&=4}J#T^Pph?f{TSOXW{44
zsme-~-#Ro(KfT)g#D5Pu@IN^{-yE3CCI%VOwquNEl%lMmH1z>wME2_Xe0X()7kZoa
zT2JF~hkDfqTR#lz(uX-6iv}c!{ao}%GbU`#ApP)jom#A3%Pz<p2<X+`e#g+GFBnjp
zZ0qASW$G2l7;EyA%t_}ivOj(r%a`7phoWsVtD)H!O(e7@OhG)W8|%wKte(Bz9a;7r
z#7*s$v_|jV!LbOD<n&ZpNCQm$<4_tV(C=noKgM!s&|nSWjwfP`im04nD(XYG$i!}p
z<ZVaVh{-;>Gd>%YT7hm^>RV!p^0?l5STo}RuGX2c#BI%><&}=Bu641}CCg0g>eA$5
zBrosa2khe!#LgCs4W>j?r<V&lb>!P+qA8{sH@q7hZ%Ys?Io5TfpJrl9I@F=7v6o!$
zM{LMXuB_x5kSlz}%TwZ>zeV6I4$RxtNJEeRQ8K3``#64(+{m&J9yN-rokAVyV`*}%
z`g|ILv6j)@8)E$6p61=x&tD|d^tCA10>Voew|$*sCI*9G<kffQ`T_npb3?!-|7rIH
zRl1KqeVo!<EjB~rG(5&IMws5tRG=rqu~4ccj9A}<Hq_K_A8#a{S2z)uA+eA>QQfrV
z<TDEW`kr%QZ`QiQQW5AdcWsXW3P5d&QZ2k~kQIK-$aqX+fiaPtS@W(+=^f<E5cCPQ
z^n5*{o(v4+3sUaNUMHcI8;%6$W)#Y0R05handMSHg6QXFJ-HK__~J!+G_oAtoHy|l
zAqAQeQ1+*nKPRAYtu1Y{y=@trjA1noh&-eUFv2`vQp<VClQ%49gRvnz^nxMAj#e#P
zeAdZEwt#6yY0E}o+GKZsJ|M(41H(5HV@6OXzA%4?(kByJO`_kk54-cIQE`|N-%?G#
z&)w5HiI)%muEH^~<u{EUlu(`lA<j@lA9{3VNMA-s{j2N}i<Qqu1S|5RQ$vcS?H0H>
zd$&r+S2iCc<*K@-_?N0EgpHK24r2F<^1@&uFd1>tr@Kg=@7Q<UN<~|xH8j50eB)s-
zF{9h9>&Q1dmA3qKRnD7Ce<&$b59}x_IwU%<J4ogw#}w8MqL|5-Y>I@XU3I-4Pn{-U
z*d|{A$-!JCs58A5U0KXV==H?+y!n2?5Wb&MVHatG?G|=Yr82sG^q8fWs}=8O61i_X
zCVzJNF{t?Q3YR|42I9MCDCoW{VzE=0;Ew9vm{<}Dx|~tpmL0#tjg&pD@W|_dA#)ho
z=pu8$NLV)SOG)y{o`b7ir@6=-c~q$x<?CDx2mV8ACM;Zu;1qMPC-n|pmm9E=U*r`P
zeyt#BUUa|nBt>Zbz?o<PZBwgWDw<<%e*bjJ>ek)WN7+;FIxxJt4Pqmyh$F|cs!he}
zyy}KV91jLYQ4LHa1(JnWs+Lh#w>9+x2{SMR%pY3ZZ)lI{;4(+*ZKZt@4Z8>lV~$V9
zEFd1gE+%Ab5*@k#E`pX+Nm&imZ^_CJL@O=&1Kvi4^?2+>r~V*BO(yo`xa-s8^oo~>
zMS;m(0oA7!#^qwLEOl`YJ8hgL{VPmAR?Z&_RPc=~epO(O!EI%vv^^6dt_;S6i=S4=
zBDd(UPJ2Hx@t17jcXP_zmm{D<_pgkQ&C!V7^(8?!JU!vs5Nmz*)M|p$zptxK>K6A0
z-JgZjw-|5&h^!@Hya2Ez#t9!ap7q<FQh-TclOLjXb|dkpF*afvx070X?e$0Gy_ABA
z^`)<hUV8-h&ep}mVbbk<v^5!qHi_SFB-IV@hxz!PUjbUb6-8%}?-d5uJVw%x)`@h7
zGe5JdcPtE7|4p}gxG_>;j!%cNx?~CNl2~@JNalL0KOCK;Hl4<hE9f)j8p`D$ahK*0
zVjr%7`#kYMIo?V?(mET0OP?bbxv}q*GDI|ahCbuLbTmk__94Nl5tW@h_XO56Jeb#G
z2d@e6<<oi5wJmvO&V3~23<*S%oK_;@+<Pp}G=-X~pdKOlF3qb4l~(+5a=%v861ykP
z`H(owRPBfMO-mx}P*IBs`p8Ew0$jGhXah@uw}!#oI44gA+`Ltg(lkcwIkGeY4xLo_
zEU=Vm`|nuiC$g*6d-ZybD=lt9enz+JVmnFr-*=x{ykE_Mj*O=GfrK*%Wyd^XG3pM-
zWrCk9N&2NCY;HDdOj3e^b-KrC>Qn=`-(GgT^18oqL-(ri&XU#z=907NA9neA>vSSI
z0kx^z<r>ZrF+FRM#Ys*yn;0!8h@agCM}yaBR1o-b^sG;CrT~bj#X>MJZ;f9W7Hf)~
zu2Cfv$GEC;&M|1DC#1$=kIEZT*Cq=~Z0^ZC468mIEfx;C0|r_L&$r|}FE=|wl#>Zs
zChR<;;zuiAZ^JmnLG-j1Bl|NP$lZ+O`0|>2&$M!|hXF~CvKqK8daMtpj%_Mr?mi5<
z)ANJKDw)8y8r5Ryjp^`NhDLPWkE0*X?_S0aT&pI;fNiv4=?D{{25G{imp(^tNWl=)
zI8h>F@5h)04ME-FfJqIXlFllB8iV8Dl<4^uofMGKO`=^nj=wh|64)$D@COxOh&|Fy
zd|+N(r?vUTPVuyv35)RC=f~tNf$D*Co)?#nB{oXS&GiXvhd1Ep*?8_xZ5R2)3Dre@
z+;8lDpst@GhE`K7hvJ*e;&8JdTLqKZCsQ?wtzPce<(oYPKDb$HiqGAFs>Np8LWhgi
zuc7OJApLq}j|sD^Qu1a)-nWV%?W%2%g(1)$1Q4q(S)3%4FNSLAuV*i~y}rRAPLK7&
z!9F=xW32eKka^KF|8H^VPDFM28_U7ED~X6Y$qn~|J^P^(kSkm-98fuSo~n*y(Z#sP
zQ0xjA^Oa0cIf~5lI5zd!3~u65ame+=C59PC2#<5J_QgF9?;50g*L9wck4}GeU!U)m
zEiw|dSzBC&tnQCf0r?q}$Qw+sMJ)ryo1x=%<o-JFx_I2!iKWj+JcKa({1<4wanC65
z<Z~Bj^YaAAJX-wS$d9MCy;kC|PPpZ_#NuU>Rsq@dNaM1!onPabX`+yF5>d`uiMQjH
zMsjxn6156wr0iLd<Z}!`4V4c;Y_Cm)N;7b2?b5aRg65e)K}1zuR@S*a_y<p>aqtmr
z<9-Z=Q9CNyco7^aA*NeAvhQ8#6R#_m$?fcL$;!4&Q%m)Rc9$27MI$*ypR6AK91;jN
z&oSe@c~|!$g>PJ{K*+Mg@<BMkv0yvbI__!(R?dD`h=*l#y=>*0yZqv4X5k18fRA2K
z)x;;ZcH+Z^L-4AdSTJgX-j2=A#S3UDv&0+4=ncy;&0GMXL@;FK98dU44<(B(pCdhG
z#@e5j?)!U`VZ8Z`kO_S6wI}Ry*T0~1`D3r5Zj~(|`8&p;YUwh@6Q=9=$A5{t2B&wy
zG$t){d_G<YFfjiRaz6}MGmye=2lE*lz-P)u`}IBuYk3LH`k?qdxOrZ;MtYtY8?4uf
zb>`kxEV~<HttIuC-O&L|CZzMj-ihFie15VqGDfYYmwdqgoj>yYf?I^B3(LGPA{(z_
z-?J;LA$SF%?j!ce3*Te<Yp7p`BuM_)wdFb}?bzI-)bVHug}^qcESY&b-skqH-V1g)
z-!B0nO9SH&q1+ck&ZjoEIKs}cbP6)lmKR)DGTqe)Wve!2$E=(iR8h2~`0z^K;DgAe
zrQBM11kV(^uUI3Mu?F|3T|5PqR00ke0SujBw@k-G4jv48MkAf{1ltwhkB!dOygez$
zGGq@`S{krCE6)lHj^z*1xMeOdV%9Dm`-KG?y`*w*4Zk2eeh9gMjPPE@<6=OWzLr#`
zE+}X~c6r>5>T{_esr`D+r#lVH_<hrbL1);o+q|h~q-M(TOshiE7rL%nnIj3gwi{*{
z9XNwHmTPNxjtW|S58|l@7aXg;KlQpXvj=2QaQa#byn7uZH>#}-3iUGb^bE%J2<-h7
zn&8}cts9-jq5~Vn+`erM?n45Jh2-J99H7mFRuYFVe_vy13a>qA(>>njUc8Z2&zYo4
zc_-5Ptg%LaCjvBdvO8n1u3}g>$u0U7pxGJjx>_rK+J{~xZz#LQp=Jabe|<{1j%rGR
zRT?>D!-=x_);z!JL+tYj{2YD20LCbP>cF0!vr@N~(s%q`L3SdloN6t+-1IaFxT^ZG
zUJ+YlXr^syJ17iDy&On{v5(qG=1C<{b3UO@fc1`L5fo-&rE4o#{Ynp6J2BqZ7<)Q<
zK57Ybi5jK}NnCtE^_c@e^ih2%;0W`Liu>#>l>H%RV2>q8+cfKxMJvJ5B2^xLcaW1*
z)@TESpVyLBm0c+=+DnZXO&B`Oq(%d93YrJ*Dm1w8n1~(d`7h0ZUINC!$UfG(fnYN|
zr;AQgoa2vU1ca-9$+fQdKH#;4BP9@Djq8Y$beqcJ)3lx(RYiVxS)U?VxN~2A`8V~g
z)IAsR2pjfcy&R5Gw9req)>O8dK(dqGyfHfNI|@q*t)|h5H%rxqZt0|yaPDs{&T9<`
z5Y-p^`zc8apt!Eh9Ds;aF(T@IXu|F!rj8(P!Emv?HxuX91)(UG_mZU7c6m?IkLAds
zkGLL->BR41HWh&`wyP(C*NU0*rZwr1rB(#n4<MJ@A|8J<rTsY0KTB!!$=iyMiY<kD
zzRh^M+NCU3{aeR3zV&@yZ7s8USS~YJPAKU|`nB<i47<K6^>e>_o?eT5yyKPukwS(1
z`BuCgJ^2K80cLB(PqoR*A$^7@4z7K_dK^?*QByU^61x5J)?F912$P!^gXT5rL4CSe
z4N;fer}w0}cCvJ=gEZY9GR51zvjM_iH*eY#Hp8~CvT-K(^ji%!;Z=d6kmJ0SeyT^$
zh1(gLsfsx)`q?}m4`N6+)Q#smE;wy0Ve>HX=u3K;GS9kFys|6#@%8GPHQXe&bN6B5
zVuFOQ5*1ytl2kg&2jP(8kS%>6Ts{(d*(7$7*h$zaylxr8QA<?;h^W`JT6@kS)=;u~
zf{uO4VOKE=&-MnWm-MQ6F1Yhl?i-Dj*30vL%$xyo3L#1H=WJ`y4icyL7w~kcD!#Q)
zzF6;Of6w;Jflr+%^_g?heM!w`j+7lBq2Pdqjrdke^(j54Y4|8>qR=nJ;l-V|G5>s?
z-Ls-pwYMvNwr8WJ#gS{C(BeUl9>I`5tF<<ku6d1F?aPPKWS}-~>;-=A0r6#c*5Yrf
zVs(U(MXqx*-7}^vbrkR?hU1ZF-?6`4^hQY;uEg7KTE1kbvm;5b;$-$ZodK4W9te-{
zn_;o9lxCe8k%37s=mTsB)0n2vFmM%u1#b{?o_jG_uO;(&ug#|KT5);QDnIr5x%Boq
zr3LqAp5Zy2B@&VkT(7PhRj~tr8o_r=KSl0eX3SyI;`6k^-QgEhNRRAW^uvV&9f>;D
zifN~=W#K;O$FOG@Fh)gO=6<}u*PoIMarHoB-EpoPntigVOe{@Zp1F|sEZEiU2>}u{
zg`mT9?{5Dmyl!_3S7Ty3>>QX)X}y$z_DV&FY537i0Y5~xYJN2AHPTIS83*yckp4(R
zn?M(##0VtPTRSZe*6l3@(j;LmdW4QEpG6z!S)}M54-&#|OdOFsg-M?UeKFh<E14hV
z`b-Tl5O08*9@Lh9N2^-D;6wE}ChHTFCt+92Fygo9s&=mgg#Fs)ipR(3bwo<66tU$$
zoL#{5^7FShY|+n7oQ$J-1)hZVSVNh=9MBv8s7<SgDz1^vGW}O=%K6E^oWm=PZYuW?
z`uJcSx^l<%$UpM^F@3KbOnMbJv;Z`Tr-QACMkG<=Wd`NDKGaEHIo^U9+q@QK8OMWn
z&j9jCL?sVJn&Mk8Epb?B@>-^`0WM1PkAda%2pNDrC!pC{huTH_M2#lvs;(SyG*x`E
zoo=1tYpyjsde%XT+qK4D))x;Kt6{bk{i-Lo=jIasMA?C_=}Q{x68{g|Fp0wMSyy3#
zBRq|rha_BSJ=1Ep_b`|R{oQUOW1Yf}Kd4^d$?s)Tq3u#$vNKE00s4b;AIYx2i|<Yl
z2WdTHPk|lcE_<O*!<yl&P74}(Lsjs0=L7lr`M2fLV;n!5vv^M}NzN6T9;`Qntu&(a
zJ<CQK+S|ALF@op5WWZ;7Ye{H5xOB)C?s{VEQB$1R-K-~Q8&E`wrP15Ay1YJYnABn^
zE`H$q`WpL<ueTZ6_7NkCJS_@M->3x$sPWz&Y^(=mo=@`I(mvxwS6IoUxQlK^r9T?r
zLC@C#8|N=)(&B%5k`!t*)ST3Yk&>`Ls_M1jz!Iy!conlEPn{i>;VFH^Wfth*FPi+h
zR`}uD@;ft8mx6m!4y>-iOVsF^Br|MrD+p{HfgkMN*S2YAK48%*&XFES5SV--rZ%mJ
z-K>3=EqJd+biwI_&wnea^+f>>o^IPDo|-CYBF*_A+Q+1)Rpc|7GARk?;;!L%w>wl?
zWZ=ZRTt<RR90W|OhI;38CJnYl=a3U-@*S^?$BzUv8eZ+-DSmDKHo@^O05^=^d<<Lj
z22tuj>EkGWqk&USF%Ab3l8m&OWG|<FzO|sz$I}>w9oFlb@9w71B0pxk)_IUC>djFY
zxD~j=9wegONTOEtc#Jjqfecma@mN;A+_9yR!Zw$&2TBo@qsqF{pd8{%zC|p4Oa>ND
zr;B|`;gAY{?JAFFvI+n#2M?akQ`_)62OZ|Q{djqP%C2k_rKlu4V=tmhH{=+K!V(_f
zXE?HzM!e<8_4%wF7rv3(=-@pDQ=72lj^@zJ7A^HF-*g$mZyaWippOgXuA$b<<kqBr
za?<`LBrc)ln=!qkY}HNOxU?KQ?$!BD%I4e-Jnp0u;yk)f>kjgAzOiq^S)N-E#+efK
z@|yMEW;Is|%JhWFHAOv^x0ED7cYN}8XXTob7z!4OcVftfeOHaj9i1W&bi)a&D_wqr
z2>VtPwe-m0Pq*@v@bi##1XZgZetkp2{dK13o&3F`pp<*q2|ou46f6+Kur<wRg*OgO
zDOM<T*Qqm?5?RvOHD?T87jk1<+ukr?gs%s4H*z$&0|+V4zGZ>U)>@XI;7EMqY3>2C
zmphnKm8oF7ds?aSDyKFdqq7e|Ff2o+FsZsW^4Y)n;Ni|pnjCDu&Vu^RxSrm~og~+5
zGhO^Ys>=r|t#JPz_TD-w>Tuor9zwcD38j$~Y3Y`f77>sdKuQ{=n?Y$1k?ux7TDntO
zLb^jrY5-|??s4z4pR=BKpL5=4z5l*zxt6*_)-ZGbuKSAb_4&pZEewNq$ZBc?ckOLe
zrjBY&GVh21Us3J|#@FnSgDjPsP$r*n^VCM>e7oV@4a#-w-A6l(MU8X96HT^?v$6Xx
zU67g4>Jwd5x4&WN^K{;|RNv7%`-`u75=u&t*-kXyK>TMiP+j@tCK}zOdQC6I{YST-
z{-_ia+h%<AdiKv<fmJ`rd`7>1!$stzr$k%a4pq#EdP>H4gz1ZH@-Y$Xr4?H80mmp4
zw-i`TmCPBO>P<WwPB)zpxf0*Yzfm*je(7~+7Tj=hSou3X#PTYtx+r8UE2O7!!7L`7
z-i+f!=Avz@e?y?A=uLJar1kPwzc=c*W1)VeLjw6wJcX(#h=h;q^EFR*dNkxS(y8pm
z@yXL+bFCO*t6&&NYP({@$NjYG9+8RG>>~Pu5sanl)Z?F;r$~N4wu(<k)vZmLc#`DU
zESgt5n?=k^QSlLZX`>K)H_xV6iv+ID9s^|<>ENfVmLRTpDrbAByBrilt{Vxrqm0rF
z3}VzC#4+MC1`Qy;bv=3KmLaE7Asv3j>~PlULed2q882E@ZdJ|oV5C)<M~+c1o{3Dd
z<qK!s*>cbkU~)`$A$Rl>R7{>{PmvwPaNGeRo@3FlGs_9MXcS9S%Qw)_)D)o1ofQ-3
zkyDg-%^ClQ5}tQ75($M@MX|hn?$c8m5|f3^<dJz?eO}g$ExO2Oc!I88PYXg-at;*w
zN1k_J<y>O@9O&mLRPgQ&BHnBbX}fN9WH(DvNVOz<b^FoRz&gBTDllTm_f){vdzUHv
z6tVEq{77{EW!KRT`LN3QEm->9wQ@_gtqvI%Qnu8d4idMV^LcW2uo@;Snlr~>usUpV
zr99gywvCmlAGtsBa9H^!pe3X*i()ebv}i)(m(5jv?s%Fj_w0NKj6n$HKZ^5Z3PxNd
z)zh1ic-xqZw)M~{>~w0@CJAyUX@n3jSUxAn3tBJGT!T&LG%GLSpVR`tb^ql<l_|23
z?>SYn_XYZ!7JPdJa+{sO&02kOvIa{`;6H!T)_fVUZ&&og?Ue`+Zp!C;v)d|GykO}b
zmfEQgq`5&7Jtt)c5BzJiEN>4|RpzJ%lfKr#gJ;Ksiu5Dhv`+M?p#DBs$}A|f6d8jG
z!AK#7NCo%UN;E-Sp_H!OGdP|Z3M?P<B{qck3B6%X=E7j&E0Gx|oQ2I@m-&nhn>_jA
zH-Y&Z(*wk%2h#)igs#GdSGf+d-@glQn9F`~>#G@b3dTj>n)>2oMf)){Kas7ZKIeq8
zIb@Jv+bu#Bx!lky!^jm!Mc6?LSrcIpwK!-sUv5ZfKHJ?A&8L&eM~lb;Ga4!yzQTb|
zFLHc-^o~p`P{=+lq7hIC(g9hJGY*EWO<WD`aGy`rx^xp1li*4VANBJ-#k+J5r8j{S
z-rb=3tG`*g0DY00apF#h1~W3FgQwTZrDR7h4my4vT(I3q*m3cme;YhFu%f>vmP(l_
z;`LUTs!$&qmrcyPjROhV@@#blu0#j5BfH5jyfGuA2}ilV3+$|Y7ukZyLM-DXg$?)*
z4nZ9!{B<kv%Svko<~l0chbDAC^iJ~+a>*Jya2Vd%h}<JWR6@!n4gva-Cy&BL%p+YJ
zL$7Aoy)GoR&$IHj7`|C0V?l!Gyw|Yg6K!W`azxcm^A`VhLiqYcw}?49<vu@4eP26Y
ziuLh$U`YbLkM(%7@$*0Ig~^T3FYZA`hX)^KAp)N;g=vs-V<|T^{vN0go?In1Mk36p
zX77!c4j^RL18o8Yu}B6g<j)UdFUNaqzetyk(eq^w%G~k}6=T*(M)TX)#m`tCGpKK@
zMzx8~@Zi->IM=mn<To2D?Fw6j<Fw)%%Jw`|*go)5x(dR|=*zvpaK^2)S&_pIoitu+
zEyg&Pe0%6N!_=JkO{mnG0gqL9yVN|juM0~bZLzQ%a|LS3H?CR8#OB}wmT@DLuPBtX
zLBp;ksbU~nj#!cYid|SHv+6QbTB!M+sVG*^g`@LM`MSkK$n2KkMRzc7mFJfYGAd`-
z)sXv4>~($Ka=`=U1ML`xHHD4rd(0T;dF%TxPCfaIb$wiCN_T?Z8667r@P-b9hDIM(
zr&jQ0pRM!-M&hy6#3>t%>^W?rqfu4T*IlbwdAk2KN?@R}AbZ1jddF*}z_nw=&s2x;
zAt=YJ(JrAJ@^c*GrM)gOy?u=}Tc-txOWLje;@=*NWY<dTH{2`1x}L{3C3dTOT0UDF
zI@@8!?QghS_`|Hz)a-#B+*dcbdB2c|x5=$lzny?T)b?EvGQV*068Eq_7iNC<V0~Zl
ze*S{;e27!o8!u+ee#EI&{qPjVjqR(N(q!?ZiQxiVURG=rx~vwbG^GbP{tT-E#}ck2
zkYB;tDlPtd>9v1_vmd(fgh?=bAA0aA&dsRxjB)Z3)#h2Ed{q7YE)7GZGH%r8Cuitp
z#T`lWd}EXYz~G5^;pOl;ZXi70xDyZK#ZG~XJJY^0LHUFCj5k>%V4Nk#Hp#mhbc8jL
zFfBa=g#-Eaj|cj92U5-NA@ui*fbSiB98JOJcgfBluXHxzari2Eu#~20_;4tQ;AE4Z
zd23fZZ9=eXD%yU3e3(UB%y(v6hf|}YzDMLIwy%uVGk^obxF68A(ojbGJ)<H~<77JG
z_1s^DGgmaWx_PwkbGaM|^wAf5W8y8<Rq`&%pKaf<k0y2&?`N4#DAgUsRrvA5pjh_(
zNQrd1)AGZub9B+wrkb92CSg~>0*&{>d9HFP38X!G=xHbQz^j}?LmN?UH|g_WGZx_%
zAtRSB?2nxCe?J@m<Oqk^(+pxA7rb`|kFhat)`esx9K<FaQV2-53A;S7H;f1T?(k+3
zqwv%|wIJd>O|<ko(ZA2mS9Pd;elE%o-(lY=yP(*Vd&bc&g%(E<!l%$5X`Sp7h2X`U
zw<jSOm3=eEqnmRzkIlT8l2|!TcwW6;C6%2n#m%Inj>$I6m!Bp>MdAouJn50RIn0pS
z^`JSi-0sl9VNpXYd46Il&~$}{Juyyvq}s}vtHCt!(J1HoSnozvma79Hb^wLG`97BL
zO@C;0tJTc>aV1-k@To!!UKU+SrCF6Ib)Lhgs2{(wyzej|lAIcqCkYVAh&~Xi(5rIN
zvof%67XXk}+&v^jam`iB-F*7C7l#(@_u5J4)|_n#LS$Mkbz@*WZ%`(QO}Lpk^;<AM
zMd#xNrY*FWUsROz*{YiodT~Tn(DoG~-C}2|w#ZvFdA!%~<U_Mr)s|{h-3f3<UmyBZ
z7j0Wma{Ahj(wcPu`OojGFjUDjmGin#{6D<qZE<IZqILQ-OZ8cmjOShsY5R&n4^PQj
ztPWq<{zKZ2$L#L|On0&ATGYHN0x=xP*D<bF)~h|atMr53lWret*)FVnv(Ypk<!m(~
z$r!1Ya+ru(6`onbBjT;)e;HZU7JNayYkp(7^VUWk1s3TrPVu9YhhBPEwY|CXlLzIa
zO&hT7x}`pAfB$Z7iA{gyw^St&D^Z5se4`ce!_bh=o&Ktkf8M-$pYhBxz?po8Dp>xU
zzw5%%gXx6sg1>w8=RHs4XA^-mdbWMU1VWgq3G1guIw2jB#Y@Ev;H9$kg4vE)g6Cxy
z%p<?i--GYu)pp1KW!x&>{nTAFSIvaMaQ#B)0&bJ&vd^q$=K9l_FsUG_S@I&%GMm(3
zyF>D|!mHnFn}in6A>K){@eFok>65P?P_G5S<G58?<YN~x6^(T07}Rn~Q=PtD4Q`rd
z(3H=f79x}+dOktQbcP&gUe3dm#4<Jy9<BR~tg~Uh9cOi_EY>s1yA8Q%v5N^bP762q
zeCgs;&hC>=IcpXE`WZF^^UDs2U2)*(ZKPptO!i)1pF-y;08Pyu7RDI8(?>7FYWx5v
zm6D%Nl%Af<R^#LuNhrOv#ovEdusexi8lY=*aMEU*(KkFTX;IcbU1M8Dm+-9_wo+An
z87fj7YezNThQa?TNY`QkBLSW$M`+H?tYgYIWm5o)7=dNA!t(avAt>;{99;gXm7PyV
zh7+gNxdZ`;g!ufmq{Z2^I`8`?VKpd!U=cSHBE?yEm4Exy@=Oghz<y1_bv3Z_J`heQ
zoLVF}&?T@Q!R}NAwRPHmz0`d$FN5oU(pE>~GGPq0`5<4W>(XX!{jJ2d!uWk?sgd3L
z_obS4(jBn(4rP&a8#Zu_Ha1TbbAF=vL#>IWQ)?U65EvDttymZR!AER<_2pq9ppAsH
z-puaRv~9!aBVfGhXtQ~71Y`#9G|EtZ!Lp<*!di;k$!)L$U~p!X+gCl3vu~^Zg5jVM
zkoA}+B{KoGS?92TAeeONgd;TEz0SI_M6V~wM*ZeF#<cPZ3_jErDBmW0@szDcrIoa}
zoY>|(119Y+nJ6u!AY?&r#~B5bIl-{l4W<RlhSSNG8)vTafJ@fY4AE+AR70O#K-Mz!
zNvw0^j30ZlH5n*1^xeEkL}D8<@ezmV25bDO4edhzIZ)=$by!yxkAA&T5`$M1ZQ970
zW?b=c=Ug(RX8-;rG*0!2mpgnCmt}*+Qw3|~1e!b&%W=4IL~Si8V*MZo>x<zhjtWJz
zwoY@yJiS$M9r)rkKJGT>=e7w2Jf=PD&tJV_Mx6NT7!oYIc0|W3ZI<W~Sm>8Z0c%4-
zU7kMeIVMX(-eUGuBhw{35MKZA^w8SmMJbfnM{krR!THW9!QxD3q3y}j?W_s9v0nz(
zFpT8bofFy}zN5=8xWsD3tKB=={8LNLR->^hIhG9)Jp_&s`)2TWh7x%`@;)Z``lTXz
zt71!-D`d`MJIcjgyR__M@T6|+2Z9dOIoOm|^X4xwlH1}9ms@g%S6ttSSc!BTfbiQm
zh*OkrkI<UII7_V;+$Ec#oanu4nOqF`rgNjeidPvEa$_K5&864{O<$j;tQ#U;RF-Tk
z#<(T+ua}(a$(wL3S6wcmd-@|H`8yb*L|0nOWye~Wx3buV*@OCpN8rwVg4#Rd1QVGC
z%?~fdcl8(GG`8)r`P_XfeG1T=3jv<!w0Om{VRQ(?6_@1&OSRP@+OF9sYa~yEpn3l-
zdNMhg4N+TEvcv2Sq{>ROVM1-Hg*8^(=|i@ayK1w-W$J~-h}O?xK{X)-D#p5T?e?1d
z(l(4nc*xe6)$m(e6BzAcu6`zGJNgX6Hrm$*HXZ1G#(+iir=Gk=acm8mt&7RhZ1his
z@gVjOi8mAH0gorAPmGyhxcmoe&eAh@f)|5dI7e&di|n4x|G=mZP+`KYW?u?+9g$F(
zor=plbuEkH<xZ-({4%K?r@_3tUT~^TCZG`!Z`D*X?7V>SqDfpfwVR3a!qn<??9YdE
z-Ttp9h5E}GyEdfpjRYE$dIgAZ?2E?RR(4d+>zeJPcj4k_%|Y0IW-b^7)fOSw0>^af
zi$zYy*FVHf!m~Cvqq6qC*Q(ZVQ(Ocbl&UBn|KUEY-Q0}lB95<7IiohO>+5+K+%oQL
znXj&vmu5dvA665&&A$q*zH-P&Q!yxtw{GXrzMmVRbPL@T667)Lc8gRI#10qWCY3dc
zQ^~?Xx=lrOCA=>+lsfW(RFacghT*MjckUj0&Cn5w9{5GE=9;b>t)OiQelzOZ=)XL+
z;KP=m7i_ys4rGuQruixeN(<Wyv7|Eg7Li?n2&gD?5AP@6i*pq-a(6CoO07sf)#+DS
zq%js_uJMSF!pkx4?f}-On?|g^=08)ICyGx6J}-Tnvs2`Ah<n%-s7tplAhhE0vjIcM
zoOHEO0?fz^U;gx#8{5}7X_gu#bLNCzRbNC}9cj8~$QygX9SOQGJ%4f*UXAULJBA=;
zU&bDA`@Qqt)|7s$@mu@hB2#!c;lubM$^ni1?`;)|-iggB7w`gm*PY7D;kbG&cQ0nL
z{$pFGYeV22E-o%@^fIPAtbQw7@jARr(>JR$#m8wzB$aZ*%6^1^!JfT=^04qqc{fb5
z1)08nhtN}D^NDW@V)U~XJ3ORCcXXx<DV?fQ&3Y0h&LM0HiIpNH;fhd_!Na<*@EH$v
zsUYLWrkVO!ahV&2zL&&%=4CdV`z&+h;kgW%o12~alI*-NSgwwv=~Cn~(iM1?sd{j?
zD+dZK8saX?>*o$;=_t?dPvRYnj(o<*3%x0r=`KB3)_uX_jzN#PX94+=qUb~wF~mo!
z)b;5ZBis%P-3>#U<i4}lQw&k^&5y{s3+T%Ia4H9XBSL)yNIxrK<qw*u)%0J8_hfLx
zQP&0qg^7qKa);jbS<B3?&8}_naXFeB3}3&gWH1PfZvfdVWt$MHQ>Vt4Xr&;2DjxG*
z+5~6+XDGBct3JHzyD<ct2^qihDwCYTJW%fW$V6zFajxL?iI?oJvyd$Zq<_lRP<~pl
zNwiB_q7XUs3LU-o-VP(kz&_`HUMHMJM-x6j<UC;xv<QR`os<(np3_Lq$>-l)M2j5H
zJj-4U-me&a_UKNs#;ao=>xSM_OR<L?-L%UY(2io2YWz<J^NDw*8#_m6^Tm~(iV0ON
zYz7on4RdaUOk!bMRXvx_zEQZ48dpsHSS8;hj%I1*_~uX-ZU?P7#B7fPGT)IgjIH!D
z)!GBJ!`NP$F4a`TC5ExCmmFBR@St5}LvV<T*g~lhWhNWu!ae(3UW8XDp4)v_Bh%fk
z%(l@(HhBfOiC_iHYTwJW<@a}HqT!JtDP@KGx${yRB#T_$YAdf*;~N%+PbKdT!UT=|
zaJK^d+wXorBM@vIGszdNByS$)ejUeimB>Y9EAM#-{CIG%yqJ^A(g-X^1-I~tJxMB{
z)EK%Vxrb7tnDnpvO4o=^N2Q_|7BeG2i`<lpSUyE@1+9<eDs~R2zJ)qQszfauQX13t
zXQHxq(>r7JjrJaK_g~Pd1gweLnr;)BrzRM<Hm&flW6%?H-XF&%9{eMsp0to@^baK(
zt5Sv$5cV7#NoipaXnH319+(UDV)};J{wO5Uh-0YF+%@EQWVy<v8**bBxVP`MA<{IQ
zaQed~Z%^7Fil^||J9QtLUHU%Bi_xA@E&Pd9SEijYLrFfPo^>4gR1}qMjB$z(vL;IJ
z2VR^^EOD4On)5ZH%Nb|N-A~AiIdLJ0ooB!)es1=g6_4d%^9ui|ri2N5USaC{?~yL}
zwy$TxtDbPPw7O=B+mYsYNyaLwnXYe#HYJq}#L8J6@$^Rg5n06e75<2hBtxRDIH%#q
zd;UAS_?7xU$%gV!^MxHn+_B2ZPE2Ns0rhX#jR-<;Kaz6Gt7@f_cCqS;YNxx%I2;=I
z1&li>xA(MTjn^%HV7AMw6jqQ&lb^e8b8;-|0T<sW!4!U{@<v6}Ry#9!k4W80Tyw>V
z*`6;gT=H|O!}b><q~Ko`v-@u0d7Rr8<d55HiPzc8EHAR{qqy;zMBZ&{{!L<?upn<^
z;b#K!sae?lc_(HUXM;}mlAmV=xvc_grRSSi?hf=;%wF!KC69vdcdnIR&taUi7vjf%
z0(Zd?u%)Zw!zwO9_nVu1Xz1)w8W|=jfs9MR7G(piw3<{st<Kwxl2R&+xsj$z5?VZ?
z-VdMg1}%~#(ml&qR%OP(SH}3n!sXB?qHCMEtepLAb5oh~)sf<cW$p7Hq*U}zuZ+xF
z4-t~kdrySzLL)Im;9u*gQuzk{Je;#(AeE{u2b+hH2ii5Ryx%2*h*R<KWZLiSNZ`R@
zsn9<1$$he<_qM3*h?EfIHuayj)}p$zZt>=?e2WPc1)+i%eweX$*BoKk_br9M--x7q
z%;Dj_s=Fwvl(3DJ7c7EHNVo|k_0lBA`ETodt*-9A_xQ~EndY4jljGB$B8M~)zI~SW
zZG!z-d<%XvFM}JRF{)mLt-r8DvFAWr_XT`iyz_yqd}<EMTxJKAzn4>&hobPy8<iVY
z{=IVLKr^(y^OZwHlh7Y`%ox@swG>*>gV+2L&S+;YZ{a>scEglw`tuG;_M-~rY1%h?
zT3ckq^7&Fi+W}s_E7&LOI?v@-qb^Ljn+Zv}?M{h&n(ysrm<H{K_zHfIP|yP%pb#}@
z-MH%~mMOFm?NOHvLhZqB#o9V~Yo{z1G-1VSVrz73*g2fSl$Ok8^Zgnd8$Jk8a4)6l
zT<yh*dx|w|i(b8*yi~NnTCqw#Obv;uI%hdOi~RC?i+Nb4Dm<5vy_9>)vY2um-(5V5
zyBTxwy$hR(Tr;*Uw~r@s1Is?rQe0>uzOR0z&ct2j2JY*qWvd+-<$S?%5qm0tEkA}x
zIqkqyy{@I7eJAEf8%v`VKKAhaJ8ucc<a&Ix+wq@73rB|kRHYt1&7eV{-IWa#J5C&%
zOr?zLA^r^r%zNHpvjv`9i8-4@GgB4b989mBAhzBY5v+Qfr+HUB?R=!^MKYLO7Uii*
zg4#ZIrPrz}A-(*>aBMASX4PNFFmU_xrQRTOP+NmhPpI^lVW#;pnW}C{wr*${pT1e9
za1-1p@Qbp{?}rfO$>%+pO-i#L_?c&wpBZx{yAm7D49rtH*JVAZN)?SmZPmB>3~VN}
zt4tGMoe!mE-YIVzBN$N)JDZNXY%J)B*cJY)CgG%jlx8j6_^#*A2xuT!QgO;T8VQBM
zy~d7h$L#0beg>L(Se)0C(Da>Frh{4H@b0@}O^TO|&B;sQ*vo6!$7r8>GQ);p4^T)j
z4EjhL03Owm#yQL`@+w_1Is6_ou}5U>{BElAkctYbWP5~L+Tpv4&zmtzWPgI=3-ZgK
zVUhlFs;306bsfg7ud;I&USadf$Xpudg!X0Z_JqJ-rz#T|?y}-Q<l$z~I~lX9S`dX<
zzw-Es<wzdPQ8L>ADwnCuY{*AXj0o-ZQ!69%H^gQB3jg?!z|D?AoUn~GHHw+=E6n`9
zcOkA`DkPDGZszxuCGY0WT%I*jsR}D(hNWhY2)krCSIZ{<e)2O&8|7b+Gja)ZW!OCl
zM0C`<lQR50IcTYD<=mtTm91oQAerq)@tN2D)o-u&)CDTAnMN=ovhg3fHVy{f8Da8t
z7Bvzz&I++4eG?Yc{d%@I&3(~Nw{RHItiQ}f_x`dd-GgP$#z~<n_U--mZS4M+A}Rxu
zheFl&pYst<2Osf&VV-Gy$JgJ<Fm#V_N~YBk^`&nVHY6<HYZjw0Ti5-8V+va5B$5WP
z2Nu>x0vSeL+7vW$Pz9#yCLwB7y_m&kTHS;Vb_7o75{b%7FTphsEf>SAQoQOdZXUrq
z^Dz4RX=QJNBUq*kGj|kC${1~V{}?~nZtL6{DxVeN=3zLw6#y;;v5w+6k6HFz?dznB
zCo>Nh3pHbVl0IbHrL;m-+M^$(M#-g+uO-JxJe%Ji2{b#Ow`wV&8TIa3q#R}uFP(Ie
zFiTMJ+SehS4RuSyoCAz?10&fRYb9MdXq8N|!=%gCkjN_vb8dko^@d%2$DU5-5%xq`
zr~?_H=X82Dp0($up9ue#i4GH!9{;o-$DS{ms_ZUSE~g5yMwm=rhz(uVC^xR12wpTe
zi6kp8;n{X|RLdhhYJBFH>Z_?Fz$hgfyKp8FG(eIkSV3P}!-yZnxJvQs2vc4aH``u9
z*P10XJ=MN0EhfzD&|9dwwE5!AX(@K-1F;gj3(Myc9ogn1!()*lc?}n{fg{x7h`|c$
z`=x!uW6fCOjDA1w|L1eYhV?_&_~lo6s$}!RSrNkih@mg^8Q$mts;9srjd$kx;E1`K
zT=MR8k3s!k#v&5|if}sxW&W7lzgEs9(LaPpa-z_h@S}@YUp#wFTfJ#XW)D81kzyZN
zjKzAPk|(%((SJ-wMoQ#<zfUNL$!-dfV1)dZDY!foMoThx9v0$dv3hAzFHVl~05ynh
z^0oS14T=@QN=j;XRb~s)YJ-HnzyUUtB>=#nvaOpbS$UyXGK#4+3~bX>OAl8a=+Y1z
zH{yTFqatuYhm9|(6)E6Z*r){ifpU6Z3u9@fp3g2!lry4b>*9nGCY4nfP5!ehhMzrU
zU3|x{ZuH!;GeoM{>Y{q9#;);xx1f~H_6LUdzCRZ80(eK8zpg+fjx!0})7_bU=iF8!
z4DB;+)Lp<NhZ-(OF6B0@!J#gv!ORl2<J7f_5W$?T`Ki~~SjHHGStrJSO|8ZGHMyB+
zWVVofr1<Gev7snB65gGmk(Up*-3Bw5|B*C+7^0&4tAI24o_*<XDKIXfZs}XRgF@B5
z_hOP27xK28y`^UB#_b+t28Y|wS2{r#Bz*_-=Z0P}Y$&iXx7dYjBVzK;fDQovH1;1H
z7g|)!6u-w}l)(I8@$8G5ci-(F&~vW^dm*)FQIT7Al{5LZ8@!%=k%yiBM#8>8dwgxF
zWFpL<_lfoB!HFO;41qX8UM9Ns@#zdgQIGV=42iKLSoMhn(<q%z`izwS%}-WVM8#V%
z?nQgl1)z=9dNXY^1is34re&v9W2#}*o%Tqc1OwWX7F<&YN2=ic9j=*Bvm$ey2xA8v
z>sLD4<z`;90THs$V#q%S1J%Vj=^F-B9mJL+3o>|I{p^*sw}tG}hYM@I#VxSh$Dlo`
zCtb8VU*|&7&j$(V>O5BU>Cfu1+qb<>a0c|C`w{?9;=ObLTm|ez47=d_&S3E0@Sgla
z(hi^!h*5>vW08~@(0z6dub$fpUOJdVa#|PN!NRW=4t5<z*r)O6Lr%P~J$Z?XXy5$^
z{uve%1C}n_lH~=Y_asMoC&^7u2FuE*6IHUXpyLPY{t2yuN`?~J1*DzYArRkiHv4q)
z!z-;MQ_f^lsK<W-84YZjFCx!@zStk-93?<s@BW;c@ne!lp_KVXEL3Wzxkk-u+=dkq
zenBX^2*+l%=_e7hqA|7RN)Nha$qdF9b%(}kF<M}W2qdd**27aX^JS`FGy|p3sL*1*
zRf^?>%Hh)(w^%r%Pl1#3D!OBNimP^Rl#{sj$(~f?v{MPk8gm4%V{~Kw=BHbr{^TSx
zj8V0z>s?_qlotdzdr^rp#x}92$9dfBTHHbi0J&|^gEA6wGyp9$o)K_4SV0PIZFM(;
z)jw4Hj7y9FUa$Te0@cT7!3aj}IgU%~kKWsCULC*zVYmw+OWG+`oQH(7MI&;5J9$M>
zZawlQWS5<-D7QQLwyE8D>P1t`VFxCIU78U!Aw~en8x}8ulx5Bq6EWfXVa<f<mqUWP
z-i$sUS!elIAj~U~ou@8m%Hl!43bxe!8CoT)ZvYfIP0gx?>aUaCe!&vcey+t=rEL?i
zivQ?V=oz)o7giiQj?VV!=C#50s6kcmk!xTEqdjL8#Pi}*lR9Ezw&nKy-X{^-TS6E2
z{H1`*`oZIlzxDOsi<rpYQNERJvW<Q2PBO|OqTz`h+}ZBKw~1GebT(wmB6lY0zPNWN
z-(v2(v$_-Qpq9kPn!wlzqx#YoXa6R3>@Q(lnGbbBiqt00gUg!bYsT2!U&HJs=nt^4
zkW1!es=X8Lw*LvUrvojM%7wwJmkWzZaF|MP`H7|ws3#N;#_^*>5ej8ymK{ja%fNHB
zo{N6QczqIOT5_P4P{tx3Kvgf3+J~V{`B(Fu&{s}Dk(42aPGWg<sNJ`Pt`Cn-?kn>V
zDcg__jKa!tdjdj;<_Vc_j;r(VL+@0|k=ytQ{U8;PGA?W5@*P6bb8ME_r(lWu$+<UN
zi(W@}paK^@7xJ?|nM`YMN6RT&LKuzTGkc>~MaNY7H^BOt^5!F3gxVcJZ|(ySV;AgF
z0W^+Ung1p1<xaq3+ZkcZDA2zZO}NuD!OforfP)7}ra{6I{QzK{8-r7DWN$)x5P-dP
zH5})U&kvyh5|biWjM1ypqd{AD{oB2{mOi#3^mcnV;%hT1)E^y6hEY2$=~UJmYaFA(
zkIh76E7sFVOf8|(NT%FL7H8n1dzv4@3fB$9VdBCARC^DuL(8*vvSy-zfOsQ|tjDLk
zgjwmo+AwL31(ki@>vH1&%+cF2Dbpw;^R4dHT-{GNOdfdESyDUK>-f;dJJ7BRV`yt2
z(idP!JOk+0XWKes+B(zrHV_S?K434s)aw2^#c@z{_4$*2`{l9p8@?d1dEyP*rTLa@
z>?c=lhcp9q3+Q-V9m@f@99RI0WY5(mHS74tw!&(*JkN2qvvV1U?g^{Ivo}<jEDuF3
zK$x;_3aaYx1iF`$GO(fTLCN;($yZ#Z2Wv6u0O8UOcEdG=#}@5c90T$}DIW7KP5S3U
z)r0W7#VV`DzAm0amYazP=x2M26i2y3;@GW@&M(W4jLt(h%xuX7rZO7^q^jfVJ{z02
z4fmIsxGUCYN=$W*E+gTwOfrW~{uMQw-zN`f*14>b8<Fchcu4Ad2Ux%T<zDlcZqJO+
zU3~r7Na7h+LMvy&*<3#sw8l->6bRO(sc*B@%)n|aHqw8LF^JN!VKbj;)*-jVtbf(x
z`a`{ch#(w$JRl^$DV)BAH>~>W<k|(b$`~1S#d=DU>eHqOMBKh&i^Pc>l=IH^W2Vs~
zd*vH>-Nb&4p}BWsDu!g4;hEwnYLxEZu=4|Si*WfBJ9oZN3UHfog^f~6gk(FYl~@T)
z{f7vT{d<j~1%GtMDxI^%Xb$#ylc7rLcXmz>gUCOS>jbmO&t~WD@ZT{;pX^sG8PuDT
z{YEbJ!-n<D6U~qm8eP$~Ss?!%x}Z3LB^Ru^uVRWHpbRq?6n!F0xbyBd*RBEJneOQ6
zG8I>iK3kZm=!H^8QXalRr=%kCzYqAb#VAVg|Kj4{k)Z@Wz|;9*98C$w=kuYMN>CVL
zzj}{J?Ur%xwQ#x(Mx!g_E++I9ndMXDS~(Lr3PJ(V$0px%KQ(_na<9<ZS@|!P1tkBH
z=yG9@EUwBw%V;}V`)*^&3u`7cGM;$+8bMoY<g|QB{2V@}KR&kEqU+B+ZAffektNmb
zl+_PL^qtUR)D5tG#0@4I;Vv7XB|}>eb|At|crQQ}wuE`r4Y4P&p~DgZN-eaT&0m%c
z^lrRLX7e0H-X&y8+djeF9(9z<Fxmxn?u)e;!+kg=ycbKWFLK`jEOuh(6^?W{1#^<4
z|7vg$lP&{>L^GIcTZ9UGK;@2)bZQFwJ<7d;vadKfQrhk+35liz#*|dDLFz;j2~Q-+
zvR5!GCpH!dH8OQSIAzk&OHnQh4WOrCDL6XNs?w6sQ#O*J&|Qn@D>dr>l*tb^7!~s|
z-J&resgzk?dBWsx2!P|wM)k$}^FCdMLc07a*{lxfmvb8_LW^!-Q9vySu1kd2M^{3p
zFbi_P%w%d`@S&8(4eXJf<g;TS=eGXTC4X@$|L_wfDymFKZ|qUhyT&U$?cZ9>EsKI}
z-a;MRmcpDDzCQ%pc$0j_@($UU`{|+8JKfmmz2^8O7D968!8#8&0;LfqfpZ7Sk)ogY
zlv)G`qWh4%8UEYEd)#Dd`JSWpq1srQ-z9oIdtF&2SjLF$eGlL?#dkHBQE2I(qHi|F
zd%-0FpJACMut?4aA~Ir>d<yhFID0H)^bm{D?>#LDw8X*~H#p1YFqu$jiZq*8e^vVZ
zL@X(fQxx5msf?%26tUL33@}?Ullv^%6lhC?{^Dk=s7+4SCHD$Y_=!?TKCBTP=J(zm
z7XtP;FC=8WOajmdG_-kMNiIaTe6oH;0Ce*}1Xk6YqVqA1=;--=6{ES>WBT!2JNf&v
zo&B!~611=($rZ{smWcj`P(+v^0Se4r8s(Hvkw$38;_N-Qd}TX1(yyiTK}_WL9U_e}
zjE*IY6TM~$J78^yhxNo!XeCxc;Z_DF3FyfLSgJMu1N>D|6&3G_J6L<2T*sn9Phi>T
zT)emq_T;0#tBjw{_S54o8k1(bmq)}{Q_t<hLr|khNqvZ@X=_a~QvBbt<1QaS;h%en
zF&2{9A$hCoMhj6Hv>Om-4|#z&ow6N^6)=uJz+>m8uiji=^7qS*fisgHlT*rlgb1`p
zH{-)&v?5PUsknqNq9{7qh0}T=Z=A3WghPZ9^gAHThd5HKq`2odIz>`<zY9^`tcq;`
zb*4<Gv><T|k#BG|C4(J(v!wdR5;Y&zTa#X#{usgHHbz{T^_`3PQieZ$gB6PR8NGC-
z>AXqoL$?y`dj%NnTC4krgT9kpIBE@B-LEP4kfCI&M|cuRM#n$Y9<m-iMA#dYYoFf$
z5w$B)M9qq)vLZ8PA4-$dSv1EEh5CTwh_LC4He4qUc+}UG66q=qvS8bDmpZQk1P!Wg
zyY9LhlS%mtpO5>`N$Dw7v)T(ULs46Sb4G!)BHNh9`XrdJWm^HEQBg0y8F36b3mlwP
z2F)fv1RlR^{Ep|Eg3(zRpPZqhu2Q0mcewC_b#7x_uq7w*>$1z?=lq8UFN0!~K9Emw
zrak(#+FMUQslKlam7P+v`{BeTpGuF}&vdfhnQ^8UTU|C~t#n6Z!_LZ1J+AANus#Co
zp1&=pyulL{Nmj#IKJT@XbPZ~a4W=Oge*x35Ud}J_(3kDu*&M3WpMZ2QHZ9kPxBzM1
zRNwio>3^3Y|2vWTKN!}ncj;P3OLT79o{s@>rI<<YT7B9x6V5ndx#Qp;=JU>z&S8Ph
z>A5@kcqr&xI$|;fa8pHVcDyGBs5*`&eO|+q_JdT+PG?r4;@L3}2hk6*KBWGK3VuJ8
z0mBkMozh=c_-|3YDF<GX-MfG6Sb!+~-~EDFVwH_QoJiKWamr71dGouJ=4uHOWk*=Y
zr40d%Gti-T2PL%im5mt81`Bp>f!rf~B-R*xjcgr#6E!n1Vo@Du<oBA~h=g{KwU-|P
zB8O>*u063o4@z-ADEYkobK`Xl&w-T&7Sd_+pR7vB2LL-h>tOl+#&Zy+gG~vX9-8@1
zP%tTfOg<lKYH$1W1<4+q;qV0O4rIT82UoS}422;oySpe7cm=S@Ribr(+8GGEF&e)a
zSv+;jBKW%8$nTm$KBOL*R%%zbPFhpw4(!{49+lM6+kh;mxs$O!%C4f9$v8Lm<^87k
zonCq_6zmk>Ft#kva<1}d4GW@+a|Jj>ct8-VD|pa~Rypqeit`7^ty^Cxg7^CiqLNMh
zUy!N)@()}m#BRXes{NstaPEJw008la+#CP}zi>L>L?e$89*RG}Zazjj2&GM~P_bvP
z5J60A!8CF5*!vm`-M&ACDET61Lp#R6JW3^j&*~HIv;b%kaRE|K)sWTSn_C#^JOQPk
z`hPm$Ai;hA_Lo1l3}nZY{o4&7E`?*A0KYqCrPlz~<cGz_fP^bDh=QN{Yml<YZq@Qv
zd+#FLTVUE&KjHYdbdW^a?YU%Uzx^+N_WxkB{`)WB$^oeg<tQP8{r?b+|Br|2Uw{4o
zU%>yTcR(GE<f7d(2}c$}MUL?t_x#&{GlL+x!2JJcEy}fq&H+r(kkqno=T3=caoE86
z_W5kNorqA{$nN>-Iwd71c>E8*xvFw^pEPjA2U}eRv*oHHCetC|k!pkJ;JS`<Kr&PV
zz#l|DP?#rVek+Y|K@~zI{b7D%FA8Oedw{f2n*xWj=dDBcw?$4ssXPEGo}V3Dz>eoV
z=#5zF2U#SD-U6P$d!!|>R!@?7k`?Mn(InuVcB~B&`ygFp+5x_G3spcBP~-B9M{^Vz
zN|L{0Dlm_yast?ENVt?S4FfpGk;Ilwa{q1ML^i)|u)VgFjNU*O01fpo_mu#X>Ye%`
zb?;$#vk`wM*#d}v#z6Rr3)o=3^YO@R=L|3*Dfe1#Z@7T{(dqiz@~hhfqz&)^azFfa
zH2^VU-mKL_L2b&6N*@SbYu)90&CTBccSnY#bshjeDqd0GPWLw|C4lfG-z6|<=xK{&
z*mkf3C^6eGO<2&c@77LdZ>LtZBDo@iEOMMyfx{m@k`+vQ``KC!_M_Gt>8As1sl-Hq
zDH)Qri-%ksOLEUyo`-+PUbddK8`es@1w36W&`EPqQaJS|^rB4x9L1-ckVa59q8%bR
zE7*kMXRuoEnmYgva<SX8ad$0J8wr2w03INeMDBedt(L%G!J!I!=ltJ4^#3U&Xa=?-
zE82ta3{h1JO`v}h1dO&ADS`3ZSJNhbC(-Olk?xLjBrbXmoPn<>%IwW4V8M0S==MT;
zablp8=WOW)6iXE3zt#4DM-$<JH-HaM@K3cMO-=wkkHa8ukpNv)^sw`w-{^)X1+WLH
zN)jr9Zh`AnpwacVHrPD2!#a!gW=~vqL86b_gCwq&kw#7v6;4UBG%D%w45e1vn0qvT
z++rERSC=Wi%n5?b#=uYYRs7#KHR6oWaU7s-NX^sS&4<U5&U2=b@8-0gQB$&+!Q0pk
zK%R7(0&nQmGFCT7{U-hJu#xwygB{&f!+M+^YX2PaTr4BuZEm-hixMemyJOmQtM?{&
zzHDg7_5a_zmrtc2{y&+;4?|u7bK{=3qS;pihqbo=do2cj9O`fAe(MgZFXD8pl=@lP
z(QzG-9%N^~1X?H-4n7D}aXkiYMSqF9J#?aHCR&6_G9ACqy1{+4TYwEI4-f=>n7TW8
z5y0#zL1P@D7ulFMfqUzE=oX?45406sI077+@yFqckg2O3&^n~K_z~2OWI}h{@l)v-
zNqLP`*LQR>X|xzW+NJiuuGSc7z;sz%lEOf#l70xdcJxU6tLA$Mmuut2o@>j_irEZ}
z6Ft`q@^Y|k_xK-uLR!^{zS9ZzABJc8jr@QL$q1R8-%3am#NkV9)HYwQ(%ibWKAVyu
zgWPPX`QiWoS_`^gqRA0>^@#x@$5GJ`u0`fI_iKV=Aya&m=fTYvIbbyk?C{JS*|B#v
zZK>Eh(Jhs6Slzx+n61fU0@Q}?BA1yNsQ(aneZL3fEdIM(3mGz$u^K^svJ9Mp?H-r*
zDS=CTi)+8}sJbaV_uCEVZPY|Q264?80FG^U^YP;tFG1v>A^+$JaIEtPbdw_2##JhR
zsjg}B<i`dxfAqY>C*$J}+xtjg=6zb7KQrc=16<9!H%M?dcvzW{PdBw4`%vS_S9|A2
z+Q>C}!p{@#t^{qmkrv0R8{pZQTn0X@ZwxK~cv%d|$o_g=b%+cy$O(g$0I==mLvB73
zg5NcFbY_-j0aQ<A8GON8SnD~U4QKbfKW*CgHa_ekBSey;M{~t|F)7wKB;FP4#4U>0
zTng#@d61WKSg3g4EO8Qe&@RNwAbndqkUb0|CNg50Eh_mGBj9e^4xr>b^lnJru_(ve
z6#!x{%#Vet3L1AhQ_}i?p5_;XzVI19LC?l>zqT?*OvM}dOds7xcRq_)=kNsH8=f3j
z07QH!8fMYB{@5NHX*${gQo-NnYnHl@hbkTL-1i_JpB^L*dv$l{|Jeu8az)hWa%M74
z?b|NlE1xQ{hpR&wTc#hcI9wa=qd(odfcTuX++Ip>GQ$vn>2N2GJXahEgRlKr0vy5;
zw$TDJ*&z?pOvGZ}#Z!{Lm;^~Up^@+eU#a4-VvDox&5qHnlW#Kssor4e#FR!$e4Zcn
z`NON$vvme$M$m<$w9}k<xfY>S7Y&FTFtBy^K0<Z^FEHJA0sirqyO97PcvV12E^v$Q
zVfXPWw8G)*w+V#y9-v7sKUh4IO8|FRy@$WqI*t1pamxn}F~=YeiO4@4K*+k&KLl{~
z5_sSTu8|<b^&TfI8f{pdS~Ai=j)1SwG=vw!cziP7lj4lehP{j|GK|l@atE=YeXTdf
zyY3C50iZQUcIp9&SRYuDP|P_w3mL43Yhww0!M#i6d_?Gb){&`wu6I>Z)s~m)@`@05
z7iH#cm4AV@7ic9Mm)%{|SjeM(deQlY?PY%NX*r2<a|>YDyFJ)(ZMjTcl=JkfZM$6S
z764U1&*^E*{y(5c|JAdThl+|vx(o6G4ih>-q(N&3a;*(qc=6Rgya9@y0Lv}S{TBsI
z4w87<4g&Cm3qE}va5Q`j$ly<w!QJOfH??DUDr{7OAB`(r`n7td(22+aoCSOUHg6z|
z<o0tzay7*+kz7sdUV$sWDRkec52n4v0E)|nyl$KE?!!Oa0^X_@ku~^ISH~6bQO#)#
z=&a{c22@fG7a;V`r-0Mf{_-t2p*x2t+?xQE{+I^FX&45x0=d1TKc-i(2ok^Djx1Ux
z%<u}@kC2P>C=w-fs<eE|BFGM_I9CI>Tu&+0l?ka{e-xQd_UQ>V^H?mq`30M{pK|Ih
zVzRnhj&+VHErgLxF(3{Chp6hO3F3ZmsX><w_C4Y^;K3B31db@~fXHYk`ML56(ND%{
zWxEfKgz?d!fc<W~z;|Y1LVz~;dL!D2Du_@T#d4>!%1+4Z8<zBP9oLYr!-7k>%3!Q%
z2J0)>pZQzMnPM*ivqV-TJAKax<C0=Yeoo5;t7TYl6_4yDZaDXvX<gn!Zf_B(aP*1i
zHN@|uv+#yU(BcHpwUcx*cdAexCfBNpnE<~(YK=GoO>f9;+#SE1$DPl65zMHhH>$uq
zn+x>m--b2hxvO{(w7|<A`ksR)dFz~FSW;Pe{!XUJG!?)ia@eRkHv$&~))bH(t@mTh
z3{%t)L!Jvv14~4&EsqtFbN+T7ge}vD-yj$mBP|IxC#CwMQkFM2>oAzNe`z<qLFQs$
z*p?Eup!+<*!-LZmI0zz3!1g<3pS{<v1)FiVJU~@Q7~D{BdT7gHy#Z+Nl48{U=63Pu
zeqP#r&t)ved5L-G19I7OQzSMs;IG`(rwN7nXqY;^0H7RCS%<zwNm`GoNu7f}z|;wa
zzCWv<qI601Mlr$B3tCNVg}AXRIWutYMg~y(cw&NWk(?(ogKIQv4C9=Il<Y|CZru~b
z&Ip~jHBWVM1fXb<`$o4M;0<X?NEWjSG$Va<np@Jn{(FB_u6sl<2fRmAcsC{^XeeY8
ztmIdZg^SU73jTl-EN)XZB8Re!iB{grUDnY$32XwEvy#AlK84vxeq}0o4tTaWz)jJg
zFqjGypbYF+boXT(f)p(OOk3E8T!Th9V6?<JbD(nr)6ZJMU%5v@zMLSLsY!zWB*Hi<
zH+)})am!uXE%_5rWHu*CF1+pkdEFRcT$4A7fcw5fkva9Q+de2Bg{2|ET7=EBpiR4t
z)`a)#FAcU|s@DLx(}}Bp7q~_OLQ(11(Mk5pK0ATW^c=#l__=M`ko1|`9xZ&XR|25E
zb@EQ}ckBt`u0K4sXngZYuDtJQV}}O~=`B(IW56Njkdw~&80fwBzEe*I4{wG!$czr-
zw`Q7rb4XccAJWU&ZZ~h65^c59{YpCRSz13t&<e7Xv{n-cRq_M%wDOX!Sx}i@<AXLH
zcMejt^3fQ?qfw@F&U5k=tU$j&*7ydH0AHW4-9h$w8}(baji<TK)TpK1)z?Y-!TBq+
z_$ByKz?-jk->IvJX?aMCrj?K|@h1xt0Z~-VlnE|bd0xd`>*s*loSV+D#{hdAs`=rI
zxTH=Ws-z6pXXlu~QL?xqqH~g-@(S4tr4lQ{m^5A%rBQL|$1<%t^Cs>Z#plm!SR0U*
zk9JS+9r-3&h(<KbHyngk`A<|kKd$v+<q%(#t5CN)B5aF4N1r-ZqzPM|NApq&)yAc3
z2(+yDu#ZK{5ehCrN?T4%Gxx!K^-rzZ^wWAq`$eSa#PWk6btb17>LEU^KNf`eK7D&O
z)yG*z6x#JgkUs50BZv+dI9BIBYpw~5TOVppue10P^$-gHfTW?Ceeg1?HF16P;<Ee<
zLGU>-E2+VdTUWz(31Ze&g-K-xT6;Cr-zyDavt9e45{H*P;(xeHJ!yaE7)pBwAH#J%
zHfH@M>F4y`Qc4r=eOjPRTv~9XzUWUp+!4UFD<|VWZd6zt6rWytO~(u)mi$jrkbi%Z
zz_75UV5(J831=W<Db2Q54kQ`uQ{hu0VJIn=%CnS?wS)^Ef~zHj|7_eP0$Xe46hNIV
zr9WH3{YEg>XNK}qk?trnejYPlue#FtYhSo?^hjO3L30`@HV=18d2IZ^_9J%^%(Ao2
z4pVswYU@5=7k?}oze*uyglNIF`N%QDrO8(m^;J)KfCx0w<}Bf1UrlWi7`rXeQb#Z(
zXCr4x`i6SrD8rc&Ld_Y=c10aKi^b)jK{C)aS-sGbU9Wtq2c6C>(|+dpN*>fSN^xRh
zymvT(1U>TfZRewfc!mhO>S=T8sDV$~eAaQcZDs!ZmT~7-SA;$X@anf+RF#ILgC@@p
z)fZ=(@*+EtQ#Agh+!Er`X_z}mNO;9;qviUxf*v3l#PZNOHy=y-%xa4SEUu&<v@?)K
zqo*GHD6rWlbuKo&?<~xu!~Cfei>}Bw*A46nW5k_)xvH#a#V;Lu7g~AyHO?1{x?l18
z%bPV`0?yWhQ_vW&oJh~Ye!+Ilk9UG1DraaMKz^Eyot&lH5{gv&W`~$&h3I63e0X%7
zPQc%5?Nr7Wt<C5E5)pZu<Tld(cAKv^I%vu%Gwn3jw9YrV+ZK>JbFxFC8q;1L(>N1B
zO)uTIDZw5(BHKz(`21_{q42quN0jd6pIT$3cll4fub-Pg&zl;pw_VTIjXYWfrLXs+
zc}>zp!HJz)Z?VQ-vix<ei;52dd4drQo66+(2WRZ7awEfn^kzYHpTT2tIudOwMy*Lp
z<ri!LJ|J={_5$%DhCuhnuiLLX1Q6Z})O_V{lU;$=akFI%ahA$mYmqdUfeT~4!)#ys
z$}3@d`!ziqz|a{AJ$pL$G1&tvqciRwd&~d67x%Y8M~;eW9x-^w%bbcy99)qMUV#pQ
zD0~<dMS+gy#6K`QiQhRp;eAS$6(YNz<Q&)6|G}(P!acIhRgZM!_7lXSv<Wa=U)ATO
zFaAwZHhW6-J5Kq-<VM*0s_eLTUOECCf227T3u%`tRz@Ml6E$FpYDwDSfpa-fXp#cD
zied74oI4d7u9B788m7G!+t(dV?3n$+#=XR)eftmiB0j}WS9nFE?6L(x2s1-;KNkw`
z_FzyIGjo=&m##7;+mH+vs`hNWrD`{>jDr#j-KCO#m}dMkNYs}t?U7xwbG{{Mg_pXo
z4$waJMVZvTwV4DHoFjlo0ey?3@VpZg^vr5JXFgoN*!G`mYI^jO?;m2B7RtQqcuu2$
zDyg?8JoBkO&E9KUGd(oe?7~q6kwnRkv>oC!)trwIg;sywFpn+1hu*0f?9}J3vJ-&!
zKxLKxrrOMh!7;zjF@?)(M{luESZrk~V_kmCA^bq$1DZ3U=vX+=Ykqros~%xb0wG-`
zoCN)4p`moA{LKMr8@`={uBX%0fGMQ-%xjpf1~e8t>c0wu^tpUJU(8FHs9nO0n126>
zbqa@iG2*Fk+}}f@2)LI+*^k$g8-M<>o{;k9(V1<#-KC`#=norVr+%6G$F!4bn*+2@
zW%%sWEU{r>KmX}*^j~&OfLWl9`6&DB+4E@jEv<L#@x4y5?2ok!sT0<jY$Ye+<3xZb
zMpz9XjbgLLS<Hfh!V7x{afdzA46e!$2jMX6dG5(B8}2LDQQk<SWIPY(RBfHU{@#@R
z+7+Mg^%GZ)Q*K;13&o`j9)|l|TW72Nl}9i(4s53{cJoA<(~ZWXHt*H%v?EIuyn*Km
zXTL2Zb{2EayyA{|ihP=jkAUV%PxOgN@_PGHZ-pZ}=ENX=#el)NJ2*#Id&V=vZ566+
zPX7p76~+0wVmRixd8R}j>rdF3E#n5-0eo%xkV1=j;8`SRuVNBUk&IS+Rb-+hcbQGS
zF3z#p-7)MqI{uxc<>JaA9KfDa5f)t``;{Xbx5tYDn_%=C>(^=jm>a75CffC<qfmd8
z@c!<`jPMqNg^z$`<2JMMwIQ#ix8DYoc@S0SFLO|Za`WO@+%VIv1?w&XG2Wq0(|WXa
zk6pW$B-@;LJf5$6n8{@8e)YQwMZ*|4<2t{Rd)~g3k6D(IQAI3cIBULcR*cgXC=>C`
zqxlldy4S9YTU1tEN(I}V7qdIRiBa*QSY;dDV37a{&^{xTsGq&1NzGxkRp05u(6<B3
z@06Vg#sx7pCz~l6B?+Zw3`O4Rd=jVuI2KK&3NPna2D@aIcQQ?4+UFf_b$pVlN%X&;
z>JVH#ws%)5!{#vFtJ@$@VlS7;;81SLis`2+1!Pg^#(SXH;=4lDfuQjA$rTtxIKaMG
z%)BwH0iPU3W!aW3OR}Vf`238;mh1LIhaDG#9SqxWTx9<<Z1|ztzcd5?cA%wv7Zuns
zx#<8xpZ5=fOd5EBU}fOCbPO2evFL+w90Nuvku5h-Bn$@1pRJHE{|#O9L`Hq!1J6*h
z`=quXzez@jf%?@d1RjI5C?YgWj0F;&t2=roV~f456}l%_)$Dd$k3twQNE(`$f|l5O
zc<ay0y@ulC@bL<Ogs_SFCB_|O|HJ`g^tInHGh0CpOHM?j?|t&Kjs2)f3>Ag0owkx~
z+NJSLPjNoF+bmDgS2t~Es-A5E%a>vn#kw!gjmaa|VY&j8Dyv#0e4z?$#j<xQUyO5O
zH6fCuFn4T)j$tp`0(6J*I=k#5QJexpt6!KVZ&Da@?8_@#Dj}Rb1H<079K7em%Y|@R
zoR8dVhcNHtM5Ub1XB@uqqjhOC`*8sw<O_T2f7V;`jnPb`84g9JdAKTI02N1U*7doB
z@L^Pp_D6H#6Lp%3bw7j6r>O-rlMhRK+eYL5;8(L-lreUR4byT;wgzMM#F?G)P@!-J
zm{gLG4+Q8+wB6rK%5!YcPT?y0(pIeIau@Q84V*4hS~oFpBO9H#y%-)_Re`v*qF|!y
zMS6k5V{v{24rrZZ0?%}D0mjM-GJo5}SXq0n^!VODKtb5L2+9!h=!86k@L9h%22I$a
z<6p@7z&R8UNLJ)4*|%d**z9i>rSxv*cM!;N4<EsBgx>6XMD71D{#4FI*J<vR;k504
zkqYE^Bb|PZKqU~HX^yS5$E_qDyU~xS!a{7jROdBynt%Ms7eG-h`5aA<hHTj(RI9*#
zYrFrdT$x4bXVSe4%L(Fa-IIc2-+cH#B9!V)k+2;+w^_eZ7cG>`&@Gd5vWvvu{2P1H
z&lby<;#s;(e?3W70S8fNg_wycqH>U5PZZgCyiroR8|kLAEIH3Oh*ML|-;-pv)Ah6(
zu(RfAfgG=H)biCtW%BLdeKWpH<sKyrgOaGm@eeMmK%unYI>WW>m@F|M9VM}9A#tfY
zdsvtXaNWbDRNahzZ{ujRjP0F@RWH6LF5=%WO$z+By&$!XCK_Z`WRYZ*XSotz*i%i#
zYn<AnQkgUDl9I}5K%+oSUz)+ryJLO`tYM<ezx;^bFR4T?3&9v$W%CtTe<Gw)@i||Y
z_Gmqf)l`LxG9Elp-+8`{&tt`4A2jWm#t5=q$M3+#YUdxRPRZa$sR01=Xg|1H0UzX8
zzXJ{~w-2SER@ksg$kd;L7uw)vwe49Jp8?FLz}RY6Qr7`gm|2rIUmp!rO&*FOcV+r=
zkKUOYdf<6h0~)|*z*^{T(?oUZ-#ZO_o47}NYv8}|DSe2|2i>oTDJ#X(6?~*3GhY)5
zAtki-Oo3vp&fOyZr>?ECGy1?ASw1DMk+=~=o2Opf$!)nS_0sdZ(~IK&i@i4wr?PGP
zzZWS>l(i^BvP@CRT;_QeAw<YLgpe`wn1v;?gv^l;naND%c_^_AnKRGx?01~kec#vf
zyubT?uIu^h{pWqR?b^0%t4mJjc^v1l@8A8ie|HS!b%wilpAVtm<Pq3*BOMwNBf0am
zzM+HfJy&sl)os+nZ}tsCpv6Fc`IEZTQ{7#gmG$af>)szfoS(0CIrb`^EV&wdQhW9D
zS!RRh67Rg!TE<<Y;+C3;ce(3eG9;?PEU~T8=UvThFWZxYO_f+SUHW}8ij=!YCJsZA
zon%Q0r~<8~7Bxl{rL8mV+G3nFIxhxd!|N5TUy8y}pg6(EOVTMaNxp?h{2eKq!SyMh
z&PD2@BjA%S^)pzFxhpvW@L3kc#v$QT0%{tvYnu!Wm%bD9s|@Aorh{08-IMu;-509(
z4853ljJu|9_d#UJ`p&b8KNi1J`(8qG-quR`7iQQ;8dteZS{Tbu=g)>qdNmTFR!T0b
zdKmis2FfFQFhXmN-}_19ml*dKi!`ob&aS~?FR~rB#@z-l2to6E@ZH@{He25ga~~>=
zR>|hl<~ih7D&~EibEDOwXt_*0K{?0Yax&EzC<`Pa{R6rSGg=PJotkgS6}-xQHi+ps
z(?q3&FgQ=2S(loJat~sJuDO}tCng`$dBvsM`(jiJ&&RcgMr%2e!03ITC1rkxiOTL`
z7(J}Z^2`-zWqzq;*Tlv*fnFvBATxsl&Ufyj?602j#eUWO`GQpIq7Q0`A5+(PY7in%
zA7*kgyG}>|>x`GEss@w2tE_mg|2g*aKY|RY>7<yNGuo3%pma3`s5?*{g4sYu2zOU(
zU6BjcgPOxY3;}~QSjP3n`?M(BVGf1cLOoO2pfdQ$0t4%l^t~v4g=*&acWj|?Itb$(
zJ|ke$HSLXWWq^KehTw2BFY;+_OuR_!b9%fVP6iG(b8wt72NCv~xX91CIY{ttN!`>|
z{QOF?_(?Lj6WGGQ$03k3Imx^Qer1;3H%@kJ#4)8B4}x`n56c!~?573X`DXg?<wKut
z3TKE^c!Q;Kwb$)nMfK~o9=mnWIEtj3RbXlo^gP}R6(X|OP>rvSIq?Fue<^6>sN_W-
z7Xs7!G|IBOAFJoW-Ntnn-)kSqCWEmNPU_igyLvfh!amrG?1iX2?zG+*1(bxy4$hH8
z_CWlyuZin-<7S`;lP3Y_T)e3WbW2HRRFiACiMX<!>GfPN&|~QZB4k@*gbV=X#5P8P
zeS#eUihhx^C~%g@^4|DRTlCMeVmNF&QM70<9T@^1c2R9zIvh3?(R$)%8_sQ?Q3~^>
z;Z%0JvsQGlU6!M5O`iipi0>E$vyx2_^hpNsX5ThuG&13ngJJQMt3>I*Bes!H|Bp}h
z&ks`>NT5gT*l0|I+Jxt5Fw>3GuaHjHKvgA|H^J*>k|Bz8Li0#mtBuKTrji#(J`onm
zt3L5JSODpI@KwNnnuBu6l?0htU&Os$DKGS4dM#)-d$|LGqJWjn$G%^AdDeMAS*yTR
zDQ>d;?d@2W3sW`aF(-kDtID$N4j$~}TReb!HYo0wq4Kv4v>Jdc;Qd*U_Vj7XV-*gL
zV3r<P&+WQN&MgRY85au!bHUAb7Wx%=WC^GrVJ5+uz)t@_TcH1vi-Ll^WQ2q(lTxCl
zBg)r2JDOJkEdtfG?_4k%0GC_!%Y{?C{QFG>m)0`<H)g2$rP3UsqdT5q0`y~AAZox3
zFCo9N=1V#*b1#$|vX-MDh8xFUT@L&zTy08nv;iPT>(^J<`7rw%wf?j)q!m}cH@rXS
z>klyNjNT0tm7`)Com(EZ(4$o$#~+odewu{@7q%H1&r5WMm5GX6UuN)D8z_eI_3LP7
z6}Qlg<IQl!aOYH2=KRC-C*wbpI|T-3Imeq#yN#aOp}^eqVpQsBFKE))n8A_Ir^kCZ
zf;WD)VHKEZI7pF!rjM@U9=cFXQ}z6&L9SfjhAE3M$Vw!s7gKiBTi$(y-3c#bVvp&1
z7J`Fje{L<SX+VN<10K$Ewzab1DF5v#qYGy_LuHHlB^Oa?g;ceTyzk9PSKhfnJU<eZ
z`*t<5^#oa$KknAlJHI)sz0GfFKlgy|{(e7?-5W_*@tpFW>zw%D-cF<tDV+zhUyZC<
zHFL-}+SeHvu78+1#a{WX{7Gmo1r*u&yPCTyaxLEy2%3h>*boY(#vbEb>oqCN$#tMu
z0T=YX-o6QgWb`g?6mF&Q_Ru`eY!&QK_J&4Qr2(k*76yi-01WOb-8HUcH7$nR+M^{y
zMacw4YxMlP5nf-VYZFu&5C0A)NDW19AZmEb`)Icx+c^P&2n!0*SStcg(w?z<VWCvK
zJsy68WO^=!+lYcW2@3b))vw)>z3&OtyQB^0rc&hR<`Wj?5>}_QCRpy1PsS0}SOL6Y
z>WA>f;Rrg(u8XWw)NEmeU{Y(TiQaDFj&1y0!^LO7y{fB56K(GW#N7{9*)BgI4zyys
zR1(P`JEA1GdTl9|#m@#k0zyE;+J2p#BS8DK>r|W9ZmK&b8}GNNIkJTdz2WMzolIEp
z{p#+(HPLrvmr-zk%w%oqf*J9g9`Zu0YW^WW8~hZ@5qC?n38gNm5z|@KKD=>eNM37a
z0fwOFRNRh9v(j-i2Q%y}`zt_IzE#g-Q`&*pV@~hw<sqFA^XQ=(60#od*BOM&UBdbP
zmD}`IlnJ(+22(ZGn}pWC4duI~k^a-~j2EY>yM;%tt#uwNkX3bz$m+IPg5j)ju5BU?
zn?9}-%Y75NKH6DKJb|maAPJen%_uSsb?+>sb{^vlc?$+qzHvQpY`kh1qR6pkI9j_&
zJ+9*N?Ywpjbmz?JZ8u=&B*fBFDk9jTJN+-uHQ>!q1H>f?R7NZe#}wwK8gP$+ag`yP
z00%j4d~ku^Y1LP;ZN#^^gu^mWY<W+Z&|RYu@6JWIUy;SR(`urpRabR~Hu~bI;mXQ*
zO;rW9Pv@2a8yD9;3t5ZkzMSTkx@#8=<NdrXlflA(ArtEL>9psiIeaPABFkFD<*7e)
zXw6WnJYAk$NrWr~K8N81UDDd$c(n96$jW_s;z++|mkwR%THO(ExlCS7_4^^yzTC~h
zhdXvFUI0RdGz+$Yz~ikOmRkmkdM=Fz4Vms2h?cK)EWA!CsCDqQyHiD2$@i<^Nxt=c
zrOJga6u|EuKi|hQN7Z75a$nrB6soI;t*ifaQOmH}5$7{!Sqt;>336V|MtFL@*sEpN
zI`r`7F(vnl=)$}in+xYiH0RmdIKlU%zd92RiJx16l}$9Sdt$w&E0}vv^I4yM5z|@O
zWQ~UifXn7kvW?Y^6Y#E#TSsf^PjyY%Dhana*PnOKNmQi`nI(B2Bn0yJ)I1N|-H1|P
zvpADP8ei#{K&&0Ib^TsL3Z7h<9p?)kr=bB`m&Jj?lI}b*rf@A)*Ww45@Y$kMPT9iU
zG2dXlO|y-*>+`NR03$MG`(4O~d^ua1<K{%y2)0FyiHvIqLmMY7Ic$leraakTR@;{I
z<$6a}$)K;T`~G+8|51|p2g}lT7Z00gx31a3loN<sW}l$m-cZ=yc!@vZv6WHbho%<3
zhC4iCgLW#^RE}$u^oN@L&OCmIUmHrvW^YM-X^a{U4gvqvGv*!&uoWvq$-?-e>op%6
zaHxXT-nU_i%owZy8&v?Rvz%6U>G348CiO}rsN0!zFLn6$1zf3nL6Q)#dC9`V8|B*M
zYI^Dc1ir3Du~ATHe1;&3xJ~7|mXuBV^7FtYTaVdD+1KU>wlz;iES}CVi)nh<(|kR{
z#(rTbvg_}!c<+a4AV#n2KBA^2Gj&K2ni+t{VB!(qD`XoDXy6(%-RlQn`@OAwV1?yK
z;4qXZZXN|-Kx50xV%|Yl$v^fV@=p2J>a?i)Ww$Pd!|{--EzG}qXSO+{X}d+6yfASK
zgjk+wGpRZ(j}d;@kD$$-nwg86xW;T{{#oJO_ASDA(mTRv&o$rNk8*En<<9MGe8lf4
zp4+oa&L!72T&fLRx@t^GFHUl5f+4)s5mfNQG<T0eg?w71sXycwhtI|HJA~U0Q$#$8
z43)B@TTfK(PQ2(frM8^jH=$gg7AhK1PgK@R+0DgwaqgN1g<IljxhC&{E29P?(uqss
zVt(djPqd!73r~)R_}OBL`eT302pRFCCEL-kh|Rar^O|qpG5(&CrJilPak<bwaZK_l
zg{Wo<-PjA0&qHNe<n(j$Of7GL8Jb`sy}s0UOhMFQgydC%GP0CT0q>^6foI_dhq)=$
zb)|C-c&|cY?Ic!O+#;Gh_8{=WBh!0Z@(PJJc|qQ7&LOV(=Zcul!+!X3H6LeP;U(D}
zC5bE<(dR5dq~3lTk{?Yv1kMYTepAEG4t$A>sU<u{g19(i)y$4GGb(t4bxq$ljwP5-
zkl*o6)C?^66gKpGfZ!X?*i7`z^=Su>2GV{;Q7N(LPGSaxPw}|fitNp>NyKyO)7Ja#
zZj0`wKnT*9OV9bc75@ScVeg_jI7vW1-j!Fzff3ypY5hbzeGSbH7ZVyY#lkJwXr%mH
zIt|iey6KWl8@n2RT*M)ws<vV)<iGBIXwKHFombrM|5Z>Zr@%LGo`#3iRn&-VDNq)r
zn$1<f9}tllFEyTl7I&Rh={`=V!Y6YnO$&+Ues69(f@3J<5Oi$LYxTO4F1q1|Ru`Ev
za|X$3uRHddHCy4gww9Ly+^>HF_beNi-o^`buRlFVaHJgRu>0<G?_H3{{M6CiUr~$N
z8dV+NWjB@l;7sqopCZI@^kiqAd`@W{I%ra$aFQP%3+5HEidXcm<&71})pMcwZM2eF
zSGd?CDzs=g+Rcbgf?Ks;*rZM7prqNX^t;1B#`t4@q1a<CiDDfocVC3dNtJj>SwB+<
zP8YABH6K2=#qKN&BA=eB@l7@_k+ZHBPwhmsM2mCr4ZgWhhM!_#Jiahivc)HN^)2)8
z5zUWZlB7xFv9zN@XBO|j*UX`<XZ$TZXkz+Mi~0VxiT$IX<@YH3x4Rw3?qABf%iDC4
zA1!Xh#7}&G_4|-zTXyTzBfk{2IA44-_Ong)$yRbN<^1O4&D(Pc<?Off8Aopqdv#gl
zW)w3WHOkU>Uh~3`%)isaFLIJXx7654^rf79kn>3Hb(^PI>c+MlA9BJRXtxFC<M1sm
zprd~FRPAEb2l+#|iDs-55px+s`yqiL<aY95&YKlGX4PKGgVf54sJnb8p>aYfUK*{d
z!Qdsjnp5*c`+p#i|A8g@W}{&VgJ@^zdWk!C@NUWfxd7G3alYVx%G=YhhKP~WNW25P
z>)yunC*V9Ex^Z~-_pb(kVObeYx+VRM9Q8zU5(HNTpv90Eoj(eA!W>tr{Xb<+M-E(V
zV#k6@>aU<eV(N7`LeFKa+)f5=(IAxqA`~9m$$E`DL3+b?zz*Sgle=~OKjPGXe&~Cb
z0I_7h(Dy-_E=eZrCZ)s%J?^b1Lr_~Vx=WShU;p1i99{`(A#U}vHbg*+DUecd8lgie
z?|<6R#`8g2p#+_Bp-HklE*l03pd>*!w3~4X@XCVIuOD9gH<0qryY*jR^z{&&fM#GP
zfIEyHY|apJ2c@+g@OD4=fd%BMHB{{H=IUheUp}J8{pUCJ6~fJh`wB<YN|E65d<W@$
zGV4H?U%YYDlOz~O^KN%#_y=Ubod8lv!l9iEV+6Pm<C63jbpQGn{^J#a3wadB@$J*2
z_o_u?poO{S3!wc5v(+-AhF+=uD?m;)M)bL*{`rr6dFWGj@4fQ$128Oi1Q0#f(cTa}
z?4g}PjRDJvI*U{N`d==mf8F2zyU^yF9SFW(P5D{Qr~kTD|M9Z_hY$b%pa1{T{Rcr6
zlX9|EiuYN?tK~t~6U%oJ3mNo%Fx;ZmOJeFVOjS|HKQ}M<Vn6~kWN4c&>ix~Ki0dGz
z20iFJF9x!3{UPNRfV*=IcG>LEJZ)XTj)5-cDy{RgQ-xn?4)lF&1hdwoyF2+#1WNvS
za<n#8pAWcnbA##uIZiTH48%kcp`l_k<!7J=(6U34_p6XED3D-5sfF$MbblgP&kHE=
zSP&&Wb<7*|izx|`8B8B^b1pe93<a(Szd)M6dJF;p<0L`nGH4n!XbEwGMjBZG?WaU3
zN#Iq<iyRSt(zf()$gfsLq`d?*BWr>XCn|GC7cU;#fCMB-|Mw#Qx5a-)b`lVArQik9
zT6=b6c?^aKT6M*7@__nsY?W{#4Ubja6_gM*;41Zt+ljUk&F-&M@V?`T9IxDW2aPK%
zZ_;)$idq3&_r)5B=<ofhP4zeG2wg{=gimuSxR%_G86v*wmWtw9e{B*yKPjem>=Wk!
zvlys&UnYY6t0d$8I|=lIF}MQ@_jv04=W&LYMu9qH1%$%hDme9B(UGOB^su57uOsGx
zWzaFrH5UV^q?qcWCuJE{ueo@NP+Alv`)1z9tGrcU`pzn-1eFGn-X00L-(wEs*ueg>
zy0fF4SBu}90kO=pJW^`up8vUH!T{HGryuGHSZLJ%*j(5uK&_pCD9T&D(@|Rky+v!j
zrgr=_Rz-`;DmOWCe)hEnh>d$dp&#9tY7CtLC5o>`1Q5?n^}zfw;+z_!OV`(bN+ab+
z>+eC2%1rcR!;fyHer6WbD&uRv*q(q6?i=G3$zDf`1-|;_eLn2}xPLf5Rb#$90t@TK
z?<L2qp&3SqWFw?Mz`1Y<?*kxg^%LEx!5(lGT>SW!$WS#AekHV=68!lb^dwD2Mkr@&
z;Q<?^D%vcmF6l}z!h03uNCXbj4tgP-^GE0TCDDo-Wnj@&xH-)(=UQx<*MJ*13T8^1
z60(*&*0d~0hVWV{bh?f#JnI(jJNRtB?50#un~;*rV;Vl&)FdE;Y%+Vk5qk)Zy_lx7
zw}1QAd<RE-cLw=ht~~4joIm*A@TPJ!v@J5^hnM2v59l@kr$0)nEMy4exICyV<kZF4
zN@2*`$yoiIZY4`5QJXFNMK5nhndPGX&lvUnQX1UQqMSX+RL^Kf(4<lfL2EbYU`>r|
z^m8U{Bakz<+y0t0MtmAQUk519jt#u5C&IqoCo?ooz<Fby9`>lVG?FLML;0jeWf!xZ
zqAhZMGB1k1?v2*d1@lGYUHm`NC_$(*;nkn_Wx<kBk$4P1rg7jgj%B0fzDEEPe9`$?
z?{9wZKG0TrhZ~Vs#P)E~pIom1$jU#UV_#?|y1HH232$@-{_4(<hyXh-wZE7`aLp~;
z_Fnrf#P8_4iVCAr=}Rh~Us4~A0c4X3&&vh(35l-|0eBMJd-KEe)Pn5b8k~+lJN@-F
ztcKJ_Gpnrtr|A*`QYH$N51f&ECggqqN0sN%4ncCSz^kQ!4Q=EW(YhqG8$7|7cAoh}
zhY={L1~w-lr=a!2T*G8{rMUM-@@SL=5VWsM{fwIbK9Ri+xS{7N*M_0U!5h%G5^Yn3
z#;qewL8OhVpnj0@rDktL#w&w_?#)b}7TTV{WEyNcQyQ-W!#h8q(|f51l2H*h8C!D6
z>8vuLH%x(`{{D1c^t=$1vzjYkEkC!+EtF~AJ6Oz2SphLp8Q(1JxN#X6MOM(>#Vz!+
zX0wHgu4D}0;zm#}UYp?Inou2hAAH>dud+*{^wDgnbRg7=KYNj#@wy-{TK9K9%Jt_5
zZJf!~uM;n!E<|(o6&@wmQrVg^>R7+t+*87Y8=_tw3n@E<M+@L(ZiL;N4rYG|)v?}P
zCO86B(osob;U@(TJHF>UWq;<_FMX&jhhb^Iue5Ur8aLQPx^~|pf;l_^;o2Zb;3_X1
zqiGb>cMiHL_L$b)0X;A2t)vIR+94v;?h?n~#bFK9;r3^_4)Z;bI>b#tr>jia#@BnM
zlCS$)>(bIS_c4>|69D2hP>4U9{B|AL0BA9_0!XPGrl{I_v^&CM+UCB}O3E<vz*y8X
z(5pd6E`^TB6|5y~c{~aDE{NnyCOso6eEE6#lUuBn_)Mz5v=F%0vwhD_J`9X#?W(a$
zBMP}<mo~xG+M-9oJP))6QvJLGt^7NgUxuY>cC$)_zw;m)6??DbPP7TW?*O08yyRME
znRvTBmMZS|dPub4WhSX*&DyW3Ol9&gt^A@G<IskKc5T1fj-$^u6Samln&tev{_`!N
zR8RRjZzOns1ieD31$!pH9_j($)@W)xSN}#KD(h!nmShq2WJ{&+`<4(&(!npA{DbsZ
z-E`G)@4%m<vTt`SvkOHJl{{@ypvXRD^^@v94AKAlasS`Hb`-<KbcO++=YsxS;Prv$
zi}U(APddVg1-$F%R@Oi@ps#PF_`NqcZvqQyFm}+1U%>f>Bi(W+&k3B*=qu;)bUwq6
zJb_R~_}fG+VnaRfroG&hPBy!h3kn=&;&3#_74crs!a~zT4yPMlIEUoT4Wr3=U-1XA
zJ@eokDgf_bc1$i*26!9>3jT~3a+hR9GpYNa;82@n+d%L^X;Fypx!OLIf&dBh9_CAy
z;1t>LgA8FZO#@%rk`Yi}P`FIMJP%G?<MVy*b8;T^@HH70--2z>K}UqEFn+<9XXto$
z+006$_rZ@xi9Otf?OjZ_a4FH45@rH2-AYG-b*9`~4J{@e>B3QmriXe;FboVGYPTdV
zQI4lS60`S4O>b2t-@5g8Te`PlEbzTiC}3Riuk}sx1n0~*FRWPR+2WoF%P^SRaCw^F
z$_-Shc#c%LIIq$bHTEF;j1nXRh#rndb?%3Y9JPVASRr@teUcMlqll`HDRd=i-lxEC
zlf`WU9AH#v^N@QxYD?dWp++_FkLZPc&?9Dv$myRbyeM@ogt+}=iS@$8j$-ct%crg}
zP(xR25^!qT16CyX2@QD&jqhX_wlKh&jDo{7^pIc<Zf%hwq*ha=NV#?6xP2LHlrqS0
zbsXF_4sr&T40et{NP>sb9efp&bP1SRfa42hB;50LL{`ZATT<mAx4YgiI)ZY@rASpF
zjSWeEQ2ZqSrLlg}NZ|;&AOZ@n^goypFqTL}%&RG_ygj^*eBhwNosCjadR7wTwxpu$
zZa|uQ8YdZNsVPhQTCs;F*VH^5ic@my_+&6j2^rS(o#oFIk2M^Kmn8=SK7wDIHMcp+
zK{644cWt5$R1kK8x*^6{93!7((4f=kI#-x1>^=g}>l^6#3@;O8uc=9gM;Z9Oi#;IW
z%!K%?sdbz=R=h*dl3=?)e8RyL9=-3DimFOO*$Eb0ZP8!@D?DxK3bnw1QEy1d{gcpo
zfltV6B-lnjWPiMZw$Z5`^;c1FbeGlfs0ICX%Z0!l9G-GeA=;Sa$77dtsg%rWW}>DF
z#vrkeJwWT=>w{b*1@^>9lE$tdasm>vow5rI)j?ZY$rs_Y%G{E;()0*&mwB5ud68+a
zp|!aVFr|iS05a{hvbWj7OJ2;H0B1jq&>5Wp2nXK1hFO69Mm$X@v_Y#nRfE}{jDF-|
zC2=AtjO~)9@OgZ(ml=Zw(dzQX7$8Az;4xS+#mEDDHCcr4Ml$$Bz(vE@Ay!eoI#@-C
zCmdN%8)7Ia^nMbmz2oD22N4Dm9RezN7|>B@x(D@#3c2sEQ71Rvqe`QyW~0k(zfXse
z>}wWu9<j)*HwohhE9}kJc*t~Jr_q6=p%?3S!?Z13c<ZwQGY+i;p2Yqspjrv3RyL~^
z?mtvkN$kxLUMjWjh<JM5;8b{B>f26px$t5-rX0~hgc9PzM7!7|7W@Lp67SBHquZx&
z8(oZ36oGq_o)Q2qCsi!#eQ?U!nG$Z<w4YF~>+Mh-A<JYbH;x{PriE{LEd}lKw}gJ2
zXJ~2;x!?^N1O4VRK=6K>!=UR;^B1h}4D2;6&m@vR52WMYls4p~J5(<v_x@uuy~ClS
zlOU3Mim93kb-_>Y-BslMb*<%UgW8NVl8$z-l_JIr=#|Wpdd8xA;q}DNB&z-9MVlga
z^wAO_%R7NNd6jqJer&Y|(?ZDT?wkxM+K_jc1*FK-p&huh#H)|oGf5q&=pW9hpj$B~
z@7ZUfP0zk<(>pM|=4v@`Ja9f@pTS=yo#U)lqp$wV^z&|qMGlXOrSOt9u@!EMOOD|0
zXtjoL8pJuB?H_Yq?VAd*SB$v5x_tPkkUS1{-<gWSs(zN(aPNca;@qR}&{!|LX^0GZ
zD@#eZh`%Uxs@c2ABC2V4_HCKQ_+s_OTz7fUe3xNk0QaVqtd~o9Q|&{@B@XdqS}U5S
zw%!^9313mdZs^?&&`@i3@_1&_kNE1o*(rPPu2_weH1)KT7*V^MAlY&A?hM}^tjCS&
z*m$)MGk~dUe6gCIvGh~b*gM;KSx1mz^1|DwztNa5VpMdKhQbWw?A*+u?i0Z}(d36b
zQnP7anTECnrV>^n-H(Hy=9e~E%W&+!ya0H2-W2UB?zIWOlCF)UTKmAl24ktr9P=e~
zrzd|BT{Sja?aoj13Nlt;AikpQ`0Ju)r27@ZA<xFT1#OQ>s{E-V2}0Iu6+_8w!amKh
z4VgAZvbEV%6$jZ19v+h)!Wqa>GAMfv`+o6RiQ)s*gO9>GC*&2WY+IofZOwwxLqG6q
zaVR=waD&>UuOVC4bhB*}2=CC7Kb09^?UmyHwQYpwJ(0yWtj8u4;vvazU#L8r(8H*o
zQYtL}Tzlprx#z_IsAIwrgp+~ZY}c;Lk6ln!toS{QD#beKop$=mA$ew!^Wz6brSC_&
z!!PfR&Q6~zm1mZ;30L{os;r^Ne@U;o-m%Yr_xokL<9M_GH@04Z<=nQ5J&IfQk#de8
zl#*qG!%ta$;k)`cIp$k^Sbxwy>3v%#?pj3zeOhm(8p@QmS7s+%d*yC8SGk;^J)Onu
zbbp()zjPLb=zDJ+49QVKa87XEylVsi8NyDnLGA~^Izp(5gWUGG5M4dH=IfD!p7i87
zGHaVGy-QEKax@aM;<T=l>vB+8m;Tg{cCO6gv8PjQ;v>C1czNv$UsvZQzsPpH_jF-O
zMY)cVl_^UWPlXE~wkS9?-t!~ts1)licyH*8kq>gPomdyKy%8_+vhRO3Yt(n}q~A<$
z=B3IfN@Afdn^}pScze4|%URsrHh=HPKwILy<Sw7Wu9^dBw(sF%39s;~<bz6YElakL
z?c-4lCU4&CJ(*8C$Lwm4GcUW*5?7@uttoQb+qpI`6~3HWe=GB;`uu86MtDrlQ-$!;
z-yI*ety5qfmW+QRh#-7<<a|Och!-hcH4EUI)M-Rg&KSKLkuI@;rJ-ZRDi4uHCrbNW
z*DHo<9>>UBh7IGfGLC$>iQz!#m*W?gYf7gS!kcPRrPq*IU^&*$J;2RWT?9RtA!ES`
z#SJ3+Wn{0`$Azt6u^qQX$O(JKf*;MI@;Y87w(igGw#$49#DtDrFlctJ+|Q>Ca8wqF
zbl<eta>G;78cBZ`@G21_O$+B0zGiw5m>MObk`Lnu-^+3`E>k%T?sk@HJPDSzUu(Vt
zSL-x%B90L+_rWv@k3Xmg^;$)RzAj)TqkT}p+(^*Tu2~EYQ|YzUd@H4~KwGS^vrLJ%
z0G)*Wx>f%rDyzpuDOE$MZNh<<IHgDOw(njreIa$8HJE3_q}AbbWkX~iY?s(!zRon|
zm5{R3I&M{1oQH*QS<<T&XJy?;cX99wSQXuoo<P2!ahde>m6j!g(O)+83T}msT!`!e
zA}c96jWZhRyf2X3@2{Rqaaplp#W8qR7Np%I?OGb>emP92%X&6>)QI%Xk;$*$<53E4
zes0}#4%Um`Ohw_Oy0=_OGxs7=W#e3jAX{<U-f*G2jCek4uO8Sd=b8@HzDv&E*B;yz
z({uW%f#>ru%=Ls?LxD!IR`9R}e7dFOnNkV4I=m|V`r_VSOSLToE?LcOjRcoI_y<<L
z*@Id*p#C{rHX&2GT7&zA$VHH#T|Eq^;7l{xvofE)7*X^<6^7x<QT;qiaEr{>rTj;c
z)}-{u#8z4%(2Ei}R1MCP&(n24#}IOn3Xvws(4!!PZE9jLX&EG_0faoHU3Axem)!MN
zJXN)RQ+`Lm{lW3p-Iw6GyAU_zFq~)Uli?VkR#jG8tddU&RSE)YLjs5>Nvc_<STWw1
zNFgD;4HyRhlRWd@ksmF0QUEaxO3E3f%I(P<YfPMeW97S~!_$>#J2m2dm;8C2pWJ(g
zNT9W6RoAGdpr2JFeA{-rm+L{A_Y=#foNrl7>x=K5E&zlpr1;(;|MRXmTYZL$K3dj-
z%h+|gX21F`i3qMsehc`|skLAWQwqUz#MR++Cu!p@OK+r0T}+$(as!1!H_^TS_K0O!
za$5v%i7Gw()GK-}w1rrnl0{|=3b=-W2?Wni^>Sblm4$oT8`$pJc$)P|e=<X_3^WX~
z#t0Q+vC9~}Vb9v>M$29Wo?iVw_qgSl_RxjjTm6S&%0TxFn`rs`;V+LSF+n<8aDE5c
zy1R@w6jK^DBX(Cn6gPDY0%;V_I29udQbi@tFhcPPkUeDi*J#b$f#TAadUVg}@QgE_
zgu*t!v$~fcoe|*+wFo1w(%GA@(a?>;xpkPM4On0-zpcMm;W(7n2i%t^P^H>jTru=7
z@O$jE7!C5u!=Da1@_4U#Kh%02%cRSMerE0z8co+8?aA?;Iwc_%M>9ian(3ha%_1S!
zs@Y%XeNNlM_{7oyw;sDpd=ga_oGBsXLHD{_@Ybs_In33<QV=@xjMl*$70>UI?-ASG
zn|=eFOiF3xriIb^Iw0heTxeF^Z7U`UU7-0nqTcmDffDiKJIc5<EEGK3pjA8lbbv5o
zomgPG&MUG5e)TmKJ5cV(sRMTJG$k3%Ay9BgZE4w_kN*N0h~NGg6zV75p5?>+?AT#>
z#>~B!IiYSLqCC-Q$C{Q{W<^>$z>vS^LxYMwjFcKVrXF`WG;9X4NsG0jYVyp>w_W!W
zv(#ajzDPG<SW%%~1F9(?71Z3+!}9?s{F7$)8?QM3@sJ5|>t+7fz^p^HrFr-rztgv=
zQy|g3E>vCeFJI=qPGYcdQWs6hg4g@3Jj$EV5^$Y7NVrsC>3b0kL)f@t1fie)JgD$t
zdkwWZth`5HS?X6M@E-@Gz&@X=&)^hfT=YS&n7l<5$~!EA1wv^3CN|TtgPu)(UVYt;
zAK3<QYD!Q-bLKW3y#ZF^5&85e|6CkHgXA4l<;HlxLQvuBhye^8XtF+je$fZkxJ-<y
zflpCp0&mVcXgUudCr_(Yct`Qhw2I?XKY4ZT1`c?Im<#MC7#gHd1Dz6(wQm41{i(<2
zh5LVO&4J3Plnf#)(W=VB2lzsBx7IA9VI0JxJ$GRysBtic_Kj7>0M3UTWaBFnoJmNQ
zX9aExAHuq9(E5SBkTXKdl_q6OAZ7SfklHZCZN+<cgr2sO^QgkGG1h5hb}D+SIL?Cv
zxvZD2-qbX|WQC&v<TB=}xi*I})@r^$_R`C+#q+Kt$-t{$mR2#l$e&N;AvjlOHxGZS
zC})ri#<R3upnYV{f1LG7_%Rzz()t193vpnbd*iR0DM~KjV$KjbK;vRn;isddXKNx4
zB?0yWQ-La(N=fm*`kI!t*T-ulwFExH5WdC`JdABr(ID+mv(-@0$^(*jpAyuK(cFqq
zp)(QR(UkufC#~~jvzc2bR|Zej+R3#{wv?*ab%z~~4Cq<j8%Ml`4d|kujJM1jP%u|M
z@WO0WX^CZ4`-GS8@$6-VX=C;I<~X>6TrOBC7)2oRw@1w)&@$nMK1lkAjf23aMS3-E
z+yn|Y>KOj@(zNp|iuOgk16fZPS8c4k3o@hT1|zJiv#j_sY#)G$eXScC>deir$HQ1)
zAJ4FO4WfAE@MOeDp^O%Yt$?KDmFS~5Db1Ax9t()4IF^2s(5+}mswcca{w^_1$jU&u
zmg0v{5>q(Lf##qEP*NerA`w3>6}GmiT1U>8Kf8PG*83AIh4F@G``{?=ThZb7IyuNv
zh-E|4E#Aujk7uw7)z+DZfRVhc(xD%Fu&3TbWm_@-S$HA~1o(^TzdN*--BNW#1SU2?
zqcwJORFnHpGyDHoS24%OLudeF0~S+W5^qpVSOBpO5!EVqbVb?9e2*nYn1n4dHVZ2h
zw1#)qvv9?#Y3WmqX3oL!ugGvmKt`zXl1)Z*=49@7_wPgSDY=6IMV{eX^d*)RX!4qi
z<T@N9pthhe9tEBxSEGPZ3w;Ct^9DB^HyQqXoFj-xHu^;SAdWm>?(tk5S8NNgvQXDV
z|L<6RRP{OcZ3@=?wHLh%$@PN;6(D!Ynl>Ig2IGdKuYcO)u;K-F9tO)P9$OF5vDpw+
z6!QKzSMR^xlsa{}reY7?$g)>gB1QP$Bj%y}xgVF1rH=#0K(8zb%oBjiS<M64H4Kp$
zEg$^~u;#=p0+p~0&@x9bJpzW&(8dG%h%``7yGP5voLh{W=|?QWhbDNQ_b{`1j;6Z1
zaJ*i@I$H|sR^7ub|6jWh`Slx=wO1R^*M7|3Ssr3PYNR<s;EitzFZK$x{&F1G3bVWp
z+hP;?uq7D@9M<zzawrAyi6{_wK8v>Pi-FqDU~;ZYTEw%aPhI4i)aZGll)4^vDWY%L
zM8gVHfF|-?_Q!5mxTP*}Q}3d~5(6#H<iOC}=fb}}H>qybKNm<?c7Z*5!@kiXQw{|Z
z7Y$JflZ^pa2&r;2c-o6JRzb)>l$?sixP#5>7aPl`-0=zk!ZP=A3nUI?wGeXvHsIZ4
zoq9V1NMrI-$wz^Y={-?XN?dj!fpMT_aaS8)z73}q$V>YX#0jT<_{3M~;BQ?Y_kc<u
z6a11)2!)Djed`|>>Oh6Vd=$$jty2TRZ9MU=NLdtRC`!1p{nIT}8vK4Gc<cHjKDKjh
zP%CTOc@97E9|9UZ;rz6v#id1$qMMCtz{rU(jZI>gE=^&psdi)hH9lXpke%@RE%R^_
zuU(JUo<ah&5r<S3S@BG9Vr$V(75EF5GDX*-J@zI%7mG2g;_fuaV5CpfLJe&LM?wrW
zN6cjZ6%ipd=4L%_#1*3pv!Rpa{T86UX!uCd#STi~;1+L#MB1>e&(F!8St03ovZFUd
z{P`w9rQzyy14z!LK<&>WLpCkfv@ORQaD3GrK7sNrzx;(~?@3@FMZygdpm_5c!uaSz
zT0OXwx#{t2QDsT3SY;G&$THr*h0Htxy{=OsjgdU<nw4OGq7My0>n-6aQMkD))6s(&
zJ^S2y+~J?*Cxkl1^e|JD&YL?e6yktC9=H30KTfVCOCcsd-`5{Jyxl|C`>{IMX-Y7t
zEY<t|98jCw4B5;SVZ}Qmn6ih(EWEdWoGlJ$SkYSwZ@mbQBi1tps@<x25MvB{s`1cJ
zj(a*ccjo2ES#An?GxV7JL-3{;IZbQIyCZVZ6R3I4&wLeB8?N$@#S&1R06(+qbh*i3
zFE?mk*IR6=Ca()Vrf{E`IVg@+{IuP^T%HVt_OBq3l7_rc=?!TYT-Y0H7$^<SRC9`8
z`7g&a#{fOrbk-ackv{}XQPF^YuM-h&QdWTS$bW$}|5Nqoi(GxPP+2EiRh$iK7+4C(
zKJ(aY?&clnmKL5%<8-h?R}{4Bz|B!^y*eHSz_TW)f`e&DwJNS@OfbHpuUALRq_)Ch
zEu&LM$dnDiNj~`ArYs>iqN}Dh&OPb!{$W63g>92gi7^`q!EZ&f6<`3tHBns*d11N=
z+^jPD(7->@^|ck{-X%1UayXZ34LZ=uKv;qWH&nfR9!y>+1)AJMUd^7i!?&mkmu`<A
z!(;C1E-kz|-MmWOHG6S3O86QeWQ}rN7>EcA9`BO&oRhe^9cn%bqA&R(M>}7r^?{CN
zX)gJpRqf&2{r&pqY-fb$LM%q+#UN(<sdG*34uif9#)K_`Xnx#vo*I(1LyvVnK6a#e
z6h-ugZO(R*s_O#_Dvh&5yf<06TiEq|aCH~2yJUi!xhy1Ty%tuf<1@7m=SyKpDIw)~
zm{3XE&NF5!xjbIuPN#76+U6#CSf^+E2ujO3!QOOLf0om6;jx2?*#dbmYnf(2d#}iG
z*BDhMXINit2-O;MKLCtxI}l6z17R&ADv*iDn9#`<R8;X=G63j22H<7GH7&LfJ2t!r
z`x+K(_LooN_MCkED`a_}0bO+kRNU{A)5Nmkp2iop^Y(T(i<E&{>y6zPY^KbTs8fa<
z-QOw!ambnX!F}DrcUm&L{GalGBf-DI$F4)TX#`Z?P%Y;Tiu|7KMx$NiuA!%0glq^x
z<MhBNo}*!~Wl=|Q?pUIfE?3mu@v&#A75n^+Ub?f%qy&Vg`=C_8(vd$P)Zx*9l{vf-
zId<XYLaaw6Z#_rdHj13j->$*G!tl&--2(_nre{I-rFtgy211;7)+Z)}c3#90$JLpK
z19N$<TwxtP6VO)g{`N3&7k4?p<juUvpf1)q@kAEKkj>=R+eg;ln6|QHQLB-&4V4Ww
z6#=i9Ev1NCUk0-j$!*0%O%#VzG2YMJytUm47310T2$hS<49>PAYNh$@!geoOZmG6=
z#~}uTW?DiCEma2JbX9b<g1Sc;=cgm$7hT?uN#>SJP(m~GGkvjq#0uk$6R>u);#Eqn
zTRV-jhcce|X2KDHx`l%dNVkJ$4c-{UV8_YneT(lO3uNkhChG-gE7p30B~jjirmN{%
zLl;~NAB(KUf*5}F>*^ExzmDOr%J_sf{ZS!m!A3hOD;TP!bU%?&h9eO}&2_7?umwYQ
zhfJS{Iq>S|<oPk|;y`?JFhhadfmlrZQR^k4giH+HLpTB(EqOE09BD}V`Xk_jU7*IM
zDGi>*U)u&#aqMu|uTIseo4Dd=gk8LK*>IO++6Zv{G6C*Ly4V3zQdT9aNqQE$^}AEo
z8hD#tV$jo1!Weha%~0&Wgh#l~MJ7dq%44-^+gtj14}N9lWJU=2rD8tnpkT!eM6BJC
zS7K|DTm8#XsF9xto(7npMD=S7$<%-hm9>f0gWGC@dvbH?>(1yKje*3(5}m591(Y5O
z9|B`4pE$Bvx|@`)mUCzwk-O5e*ye`@Yy>!&CZ<v&rN75uUL%u>kc#qq;kIc_-l)MJ
zV>6#B@NYQ~2Y+CHSI)Yx7UBTHS9^q$>ZK-k%&o;@xEpen4OD-{Jp%5~Gb^#GQ{Y_#
z#Vhl2bQ`&yxdXET<<<g?)+!JVZ&4bml@QRINxj0<Ac>4=LR`Bp7MLdX3Yr$KR=0Yg
zGrL+trn;_UQS^f775EwAGsn8hbJg*g^1zx>XzT#(<GZc#XW^RtHTmzV7`I)$TzqtC
z<Lqs0{7+s21*Je~1F>BOa3k`b%(?-=^IO}7D1+`P!zYbn%{tBTew)6W;oSJY=`!XW
zh>=!L^*vu1u10t9o+j-QEnl<}PSPG_+WHJuJF?g?L~N`jp3snqe(Z&6jNRAxlZ2{3
zu>5Ho2IUC^9c=pN^w<=}0}y0WH?~>#aEQIy8U>KYCa3b}cNpJDaV(;05>{uhw@gj1
zjFV)K?}}+uH833*5?dnpieyZweX_Xb+9;Y5wYn=mp)|D4uXc{e7U$tFWfvzeYC>s^
z=4$rHf3Lm%``0B7xK?m4fL@5h8(T!EysYS3P`yl#;vQq0=7fwwWI%h-CxHYvtA;5g
zKC!L~S^`cUe1edneS>M{$1q}bzA$xz%J;Wt@*eC7h1KluzwZ}_2XJsIVf^nrP28Yp
zxMY=x#bkK2?*L5+uh76Fk3@mT#ey_B$e-4QIb8>1+~KO%z;HPs4p6#&zqicShPCt5
z!uKb<>mC#HyxZKur24r<%e$7yW_wr)%hnk<ls`$kS|aPf9KP3PG`hI;bF{7vA15=A
z9Q4IdRy(wmVdQY=Se#@%A1<CDl|G$S&Ba@GNowKS#D*fSXX*{j_DS9h5K){kTdDUW
zIeqON;%Q;{;w{mcGp-$D12>6ha;?7g=S(8k-EbDCcTXwh?b3P~hmF;`9=hm!X<2@r
zEtDV_+g!GdIyUUYhAXPDSw&U^Y?6#NT+n6VF(Wc!>9|MdWjJO$mKVLD_f#qI$zp|e
z-jvEU_IgZQ)Sk~mlq;vOY-~qW^5N|yVxvyYJ0UQPYr@{7;LN2{OmuQ@p|KUi*`}1h
z#4P7-dh%~o%Mqz$NS`*6$!*n)Eev~jj-ap295WN1d0A2YLb&%vOsQ2yQzaRW_V+b~
zU%iRJq5U**fpLu#_cr4?v~!-M2^$XS(9HjerqeTy?YP(Lpfz|wYxPa3;I!G1fX8&5
zFU3rj=nUt2)^lAocG8w#ZeR_ihd<XM9(ZOZJY)xSNUYt6DmO%!{_TIh@o-vmEzr&C
z=i*l|4EJ~^m3qw=kOcH_>CKQpe4&3x+rD9vV#~f^KVm+&5B%ZET>MMj@_+HK;~)!w
z8RfUVU(Wgr(JMd)W22{kECtKAaJfZKB#?Z*@d#A!_AjwpO7kxLqcJg6mlQy%y1Xfu
z*>Rb0QbYc1SNzK_fR!EE;--M3xyr#Z$mIee+%Yf$J2#Lj1V+TG;apWO#Pd}aJss!)
z%U+_*8eqR)JpkJ*q1g#_g&gfJKb?M<^bcUzoE^}@=IpE)_HV>?!FJ3|ez*45+3G)U
zvagfEB#>l^)5AqZTEc1cvo7GWpD6<B2|@rzLE32mHJVOU);`jOsLPvDsp!`sYCv-M
z0=n|bfJ^EKgp2`7F^qjt?GI)*m6`<H3u>PDH!Qf_5Qj)9R>|_0;_|<MT5A7#Gy$nq
zM>I2I&VAsBJ9_zEt^q(D@4D*V78Jp0O0}sy{B#P48UGFt2j&ibdB*z(`aHiQ{g0pQ
z2|WeRuPO5ExE7$a{V8Zc^IsV8Up|O`cZeFnfjr5@8L~0|Q*it@pGPo(j^%hv*k4!H
z|NRI5^OIC+=+03;h<bD9e|78r{c8R{a|6Jt0fdtYNH;?vgEt<qaK37m{yz!fLh@zl
zf4@ca5Ss_%0c;-G-@S3}Fxdh-#9<J4TE^<w2Z6#>&4JN{@Rv2a8qjMW@^UIZJ3Z#z
zQ0(}lMO{0r9#q#&^g;8Y%M>WWnn0#Boi#c(FC@fP`|*O5!14Hm7BVZrf*w*G$JV<!
zxkiw@D3IxQL;!?Y_9+`52qHH5{!fT_x%aX18!;+Iu&k)8DXuI0<-#^ty0ro7F(Y6*
z<pTbSnnWyLuLc5p#qluY^)d25qA336hd<i)L2?zeDVt{xEeTuu)PrsS2u71;TpX(?
zhH_gTFwpK=4-gcW_rPnAEbNdA{2s#~S{q0kzZ-lx)z3+C5`>$DpgHg79I!_uW4%8!
zaxcHyi1fBn$@Ql?{VzdYMdn$jr}0Q^Ri#T(min!jWvDZC<Xm=U@(UpM-+i(Q<hw#3
z-%S9f=b!=P7BZ9bDc)s%R|VRb6yr@m7+wteD%%gL#2(@a+yJ{$P&@du$bWd4|9Xpn
zzgms22Qcm`@DVJ6C@bBebe*XoZA(@vK_@FjE(>rp{{6D^^aL2<nD||~a(TKAy!{0B
zPu>}hGHz@%5|4rw*%ctml~&BfcLGP#k&W=$GT`$?cG(MzWf%tF9c7sJ!4y8IP^}YD
z)l9#+X9X;G*v2y1=_8mpOgpYY5c3Madtcb)D`L(OVa79{tGh}VScfi7FP2{<$P(5@
zl<EDvi>K{SZTapp??J}N0nAW4apQ@w?WC9>AY*3Kj=JO~L2U9BSlRYKXf_ovFW88N
zGElQF_3&V^Q6N(pj9QYull$wHaF?k7<j{D;u9#T?BY=@sg>=sV^l`k>X?4U%-2*@n
zjdL`HKOqp8%spRk;0RCTVDf?>;6o*_ix3wMF__>bTm<;qaye|^ejUUwU87i+f#b@C
zP#}5Um;sCT!F-H--*bY=(pTUDzk+-Bs|*xYHJImt@0SspALwlA8@#UBBk2H2U!cVs
zX>b`5e;kv0DLDvKPizQa3g6qk>;c6~xGYKlD=vO=)jAv`Mb`O>Pw4;h!1m3Ks|Qcv
zrOSoIEHHMV+(La<;}&G*dXd-~MEpf`xS^zaIOpr<XT8spUTGv)e0o3q<6FWA#2Ynz
zhUO(qRo7y~m~p{g^V%R|#xY<PP@Cr`fMxbLrQCS&i>#$8tV;{GzNIV=&nNVE&e$_3
zS^LtdZ}WGP6cU77i$ZU3-;EG++yR+0B;G!a>bhZ>Yxt|mncbCDgi<;Zf>-x|N{j_7
z{-x;Y&fCHd!0(w*d!p9U55MGKT*i|mDr*0!Jj5(R=dGztyLLEadOjd}h$~SGL0;`@
zjCBmxf+dTPWpM?Ckg2>mRUC=D6L4`a0WSD$5YiasI^Z7FR}A$(tDJ-D7y^4)q><ql
z^x6bzF@}!CcNEa|FTiFG0jQl>88D^B?=^P$dM|ck#4pM2-1}^srZI<l-&jr9eUC_Q
zvqN}vyq6(M7Z4Jwz=$4>I-Nd(d{}wxrknc0(7%Fz09KJy!tNZH9=}n(?pby{28%b&
z%1k_g<TPRjNbw`{rwuee=FI@Io3`cPb6cqzwY35a(Mcpwz)Cjc+uYI4qS_!+)ZRn$
zS<%x)Wf^Z@PL_w!h9LT3yIgpa?ZJX*3*Wf1?#96ZpqkNcU7)+54D?n6RH{q>Ba1H&
z3A|2spHhi>Ro4LW?8Aw^*P-dgM_HQIoC&C`A4NmipF#or9M`=^L9xjMO`a*5@dLQa
zBV8WcX6jZA|2Sy{-h+l$eWd4mk{T0n@LO_aKF;2hIV%x9ncjhb{W3WU84?uVSKz&U
zU|&#2uJ2K3`$*TInSZH1vCX9%nm5ApZ4R1<(Qs<LsQ~KKHZ-O$eYCi~1s}cuUAyp1
z@VYZ@REi>a1{oE<rWJ5a-ZK_~u;GxiO|XId_ZX!Fti=}=Wis8gX0_Ost`q}*t9F22
z1=mM$8f9n1TyC&$iFD~P@bYK0#3%7g5UF>OY^VkFX6Hsb_niHpwTD^q1dwGLZc@Kg
zI*lzAUNU{Z?-Ki2qd2g=;&y(iibjmqT)6%9m-)E$ZmV~P^IUV8HA$N-XFylMFqy7n
zKga$qNaI7S0pDLNl(Y_q(meQ=EP;Pd?f&~GC6>4hvSCE1R)lPUo&VD!An4VYBwYoT
zqRRkwY`09jg*OY03FL`K;!O=hy&lB#!wXUirQqEZ5NaVs5h{<ls3frXk@lKgo!^xo
zU}uh>9Szl8%C}lvzB#WF4s`ZmP`1)<L*=2>1;HfH%kv$cbS5l%GiS#W=Od8vz2yp~
z{#ijI;^@gVi!3U@0X9l-J2AqN&lpVw8P6cJ5lobks1Fv%YqVpMZN*^G41;ElecD8E
zgAIW5j)E=n68Ir#h)*Jiu1s}Z;T<zT>&v8yC1dfZ(%~71pS1Q6nCn1(zs1yiwHM60
zy^B7U);x}_J7K(r{~Vr%Es@EJBubL6qW{!*DJq+_PzuB4aj-!vA~`1M{mBC>2h4?_
zW*#C5)dmy;?jxo({%H)uYlAglqzj+Ho2Jn;AjOz~AY6GAm{>=J`>(@`5LI+|5(lrd
z7NEI&duX2sJ0cv8gp~{USBWHInBLF2yx!6W90-KI2ROIV3pNQ11V&6?1?GYG;|P<@
zCmX1w_<_BDZU84g0`9>l1A9cn&X@;$aeP2ugL{jMm^<wy3G*Z+A$M^9Vy^GldJnqM
z@<|FcVeJ6)@6V+|zVdqS0II&U)B@l3+q%_Vt?-^O;XQ19DCgTs`Z5e@1K5j)y&tIW
zFw6+1_4745(5^O(i5*hrBX0J;EVTZ?hFt{ZZFA71`4l12>RM9To4XpB5e`+KA)5<y
zDV{N|v4}D5UhhpAiN2njT^?OJEs6Cv8^;3qM3r(GD0q>^-@mp8qPu_SmE3ob-Fkm8
zTrE;u(3V4lWpP*okpzuY9p<y)CH)rtK{70lF;{hklBigpR2nfoOKU(yxHtGIIuy?3
z>kzQA&-t<+A7oPGgO5YR+X0YJRM6#6tuUB9vhJE6T2%CL4B8^QFT-)`L6SZn4-Fjs
z125iCZJ!W@Cu=d>ciMEmPVZSQW&<kapAhxGSsu@i-iNXWhvn>*zm8L=U&E->;j~ye
znFF8siVu>3F}YNFT~Tz8-tsp0rU0>5Jgyk{GhE`lOplNfb~y%9N)lef46p@;;@Bt;
zk>2%zsi51*a4>AzWB88L6K7CS!H-lqa?{(b57t|04tJ(Kkq`rYI%c2rhi56)>EO3E
z5P=20-AB+c4VOqsn6C?wZwCnUZs3u^)3)SiL)%~~q-l+iYu`pgFabTcl?gDXBczW9
zEe{V$Zc`;=ej>y({3%>&B<s<wiQqd0Fws~FX%ipvEU>L;tMQSopnskQkJt5Q#+O;Y
zEZnJ1;DiF^%*(l`n~;_}uaAGQp$!XZz5&00qy(=RX=!!ydn}af5c+5k2nRBriYOjG
zToIx43mjAd8<8R@WLFrz3xLrDerAuC-n5DcY0<|?Uh@^E+xW(Vz%Y@z+M{eTTuvm%
z0@|*AyNud5qT2@LWooN1FNw2IIgA@pkgh>2E0B+DQ%Y!Vs*p$=mI5b1YG#sp9zb(Y
z(qx4!roRxof4fINUX7TC2wgb-_g<2%*uv|Dw6BfvxH5!b^w&nfZy{On-PJZxRv3{u
zB~CPmDMB1=CT&J<m*886>dvkfUdd0}x!n{2jVx&Jka2G!Aoj%)=}?X$GU>A3kUBMC
z{rleA5p&*O$yQrYjBtw}wr=Lb1U^7p6w<Qx07uT}*5NFk*QNx^!VE&o%(@!$v`Cv!
zmPj?uR=TDWU?+lXs&r|SBC5f-r!ss?;P)+SG-gN^r({5yaw$&{grE7{7MoKayd5g1
zr8%pT?4sBZr!ZlJGR>8YyD_>45*I3~$ii6u8!EKb?=2sOkb+vBGeW0`N8*8E%9I{o
z)NKc(6^+{hojWMo`h8(w3Q+?Ry1%L3jIK_Kw&=DF)$eanTW0S?JAT#)UxpQ9)%7ed
zA6>x(=djm8Q~Y~i_2j_uweOrFHXjVIDPW3%6(n9<@~C2(B#rssv9WRt``@2FweV2m
z-wHiw<4t1NuP3iCOG)$aAZ4ipyMUg&LTat_o;|Ui^w6`VddUTu+6s~iSi$T)Ds^IV
zHmO>E&8?&irAh;?=pPqdB7=b+TGSKA(Q>;PdU7Etkd)Z>A68#=0v0Z9&8(5Z_vy)L
z1FUotuUF90+N)^HCcO85TH!+*{gao6aQ<N0P?3S$Oqwk%Naq1r#r}u1Wh^F`C>`>?
z-%^}!lYT2bB>3GousIZvFjW%oAYZAri>mz?Bm=G>`QrdfwzOc63;gzxR#SSlCD&uv
z$}X?A^G#8WN4kgIyL%g-?Rm(Uq;^&;AuK^BeBJJw{##2d+P#Uex6&oQefI$Ej!(3|
z0-jkU;mY)nDN^@Wb{getM)_%$uBh^QX6zSts!$z|!CRKtF`z_62BwFO<ruzQPBhSQ
zL^?@TWs>F!H(4x5E?Tx5yWIa&A;#odxwP&&(Ab<C85~GY&JcKXOMlI%r%n3zl)#Kw
zdK6(VZlL_St*^l>>b*P}&NE+zA?AGdfEV^Or+qa~0}3}mK5AoM^Cc0;g9SM(p3xN-
zV%UyYf-ky%@a9O#_|bkl<tC?1rP(m1;&zgb)b%+#^4LKBTf%i8*L&>~-$}<fhg05X
zvX6XK(oFVDiY+cgXI3XVDVU|vW%+fzK^>bTYh|)Xs&0V;d3%6ekTLm7@|VU!Y2DMV
zqO}oE-Azp2zp4&iR$DhiS*BNO&T)Ew<R7TiP!yfeZUqv3XZ-8UPT>>jvkP%;{-ugX
zSB-yy%oAA=IL<*_DLXRQ@Nxf*m?XhQ4)Om2c$J_HhV|7bOAEX0z+dOt|1_K#5oVEI
zOZt>{Cv|MwHPJ>1hRt#tsLZ}W2>W&Xbawo+yvDCa0$CJK;cRK{?crHH07WFgiIJ{u
z@X0hTx@%&OZGr6ySxt=bzR&mEg5DZyX%PPtPxq%03H0Hl4S~#=djK9uW%cXcGqiAJ
z?572YLLWjH)qk()xc;;DOMwX~)rVO3h`V?7k6`_80zE?!fjqHeR3@`}EA;zP<aBy=
z;X=MR4bk>N-EKTT6bJ%c0<JiBEGEx)m_{xNnaT_VVgh(rVkx%As!U5jDi~drVT-B~
zvu&=^?3&f%@oOuvRSRO*!sB%PdEp!7>Zv$E|4r#RA*Ww*{W^NMV*9m}|A(`;4y$Ty
z*S@ESG7*(IB}FABQWBz4A}x~?5D-BD2|>D~J0zt<5kX23q`Nz%L>eRnM3hcRDZgvD
zzP<OmpS_>8-s4?|e<-d)7;}tq-}iN1zw^8nQwis-+l7xGSAOrR$>AuUnb=h+JU{mi
ztuaP&_dCOQ@N_Ws>GNYRv@y9Uc#9!X30uj&WoJ&0<TjRVhAinU#l2F=i~xMpV`q);
z^i`hld9*)zf~I+}k#Bm;Yx&dl$E1(_^52_Jd%RU%QYQV#F{Pa%c-NJ4VI)nNV<BF&
zcQeO6`?W_6=lRbM)$a|G@9Q=;9!M>xl$&qfV?n^$x8ZUq4ayf>&|o-mWU(;spVpfq
z3Ie2qh(6$RSR8u&RMHqv<>`xg40(SFAB>3KNwLaE7+{+VQ^+Q-5=(1dBD7{)P&U)x
zZSX&1`dK*8?^$Gu0CxFt>oW%YB!~Q%de`-ak~991)=h1aG3Z<vs&H6nJ+hrCF)K!u
z9z!yGVun~;hSi`Uxq>CF;g^aI+G@5T71J5H>UV(?oA9z^AA;pAe9g4La!SyMq{R51
z{lG>1N>=7pLs>)ovsX5z>>R{+g;~Xh?7JsC=k<Ltl&^0?!`D{HlpzuPoW84(Tu-?r
z20FpE$56%eb_3lqkNZq5-jf>49bz=8#ubt1djEu;<i|1973X#(9r_~9$q;s?hD5Sn
zfeW;;Mqdhzj2C*|NfYhOL`%+!HMpIzq^x{n6)Ep)!*cQx!`1v3_f;Nj(ibu4*|S-t
zOCAu1YNRX33#)&;P9-z^+AL@IBBAwFQi=YDWNnzbHT&He2+cR~ePrh?<yP^11}03M
zllXk*Crzw{8PZ4@k=Bzy>qsz>30}XRVDm|MFUGa&<m9Wqx0%{v=3KY0ZJ$2Z@YdF|
z{5ww9_K0HZ3i!V-IcmONcg{;IMft`uhp^fbFt~hVYJgd<arU=T_RyRw2~;;(&3`UK
z`d&E)$;`TWcovve9sN!i?mV}*h-CRFG#$?K^#13Jxf#s`flO_~d#ad<xZ*J=d1<DX
zAD=AiENk9Rf|DEToSQ2m+qzl`ztsL)Rpm`1HccHB=+su-H3Xy2hO?-T8!r=o0txK&
z`t5t1oq4eRMQo<otMXE!T#PjqAVN=8@nsVS6J6S*!JvZWgJO{Q&c$w4yKY-<eL|aj
zF$>NQ3cj88E~WNxGk01K|K!=xwK=4c@%5rTGiogD8@Eh+1#PV-3hw#!)vjYFE&P})
zf^-e~)0F)R26jK(%pL(*$^;tK4w$K4e&GISHuB1+C#|12<vrI3`Q);F$!J+GZ*k!q
z8_Hnu@ntjduN6=fz2dx?LPC3dOcBHf`bFq?JjwV3rMoyfNVik?_%ZQ%#A4)6j9O14
zvNu<>_OG*jAM?nRil9#Jq2-qZTE6Rns$(DHomp3Q>|%d)jL?EdT9>}_VaFRG5Q(y#
zqetpyW36h4XU2k!Fj9T$FA#9_+IFiUt?3tPIv~}p0=<b&pC7{HTRq^oH1M{Ucbn1-
zLMqX2>|+5t<JIr)a-s~h*qA!GHp8k$7)hg+v{BJiA0g_}&h_)39e}3Ac}Oja_ZWd&
z^Ci-F?fl)2g^o3d6$~fzqL7`_cSYn<SBvW}HUIv0-aLccPDh%m`FM~q5QK{7qkrht
zDwIoE)}xhAd;W5O-a<fw!ax{3N((5|P?!F1b3Hdj?n`f=0%u9BJe@k>2moP5%_GPF
zPa)oFNEsC;4>n>J+WksFqPg+x@v(I)b`Igl{K*f6&sTG9&z{l2IJu;)eZ_X&oH`yc
z`zOnU743{^n3OdFDlC_k@?*h0og@w3L>EvOeWVLboAiHNZ6Ukx;2KtNQoOY><#oWu
z>8<b`(sOTCkeNTH$WYVDr1P)TtnY>?bIz>)L<N?SlA63cLwgBZ#301ZqS2i2BArIb
ztPP-`u3lQ_P387*M~x;fv>oOY<0D2uJrtT<j$Vljh$NkP{#zJ?bcDC@b(kG3wfd<t
zyyC>LJdmkAys)X&`T{S!GIN&ntvM~Na0gYsSHYbJ*1opIv?wzyY8o8Y`Q|GGbwV1w
zht8kqOf)m>vuQ+?bhdAgujy%w-YNLp5PG&2XgpZVowf}xtb^vNo^{IBrPOlF9JITa
zs5~5-ws|Sxh%@>jh0fpJIx#{)@*`B!l3bmc<%%_PR7eXT!s?DRmsa`?MEFiIZVQr?
zm)1Sq=ABO>$;C}*1Zh1R_)Za;)dH6F3+2YiZ8tu~ujajd8^^_c$x;40gBK%dw&s>g
z6iLll8t2w#=6hvcjE|M%@iUXhRjIB~#BW=NAaT8nRMiymT`pf6{ANK83Ujh@m8Cs#
zo%?aU{P%MoA_AH@$j7S3@yiFCBQRZwar+y2xq4kg@s9jEjpkqKd%qBw;4vv*MDRw2
zd-7}zI8KRTtHC~w+ul&O;}r2C+WJn5no?B|(dME0_1^WxdK($n#ZaLlyPtO~C$weG
zNBjok^?ZAuUsy!IPN)!}(RW-yfip56u?J|&ds~83)T)y|7`)iW)#vpGU7DOL_?ues
zA!~Xsl(yaa&13nT)0)Y^a^U6^Kw{o5GM6CmF)W5F;Y3&ESB}?0&X0$Mw7kg6rAv%~
zLt=F$U0RsTWTZOd?^*PYq_BNFaodk$DMFntbm4nLM{Bhx<+SJ^8nOT5pwF^Q%w3Z`
z+gBZnr#*fn&+L<ph!rLNW!BF+zXv^_iPJ!(r)rqEnpY6vnsjn&G%mjURh>26x^XeG
zJp>y@t>ujD-mSR2Fv3b21yRCt6@wLbLlccG$if}nDy)TT`n$WT7U9&_fbY~3nOO;z
zYisNXsF{Il9eZzv>$W6*v#b7}DN6sYm2AvN_=O$+o#EH~nY-oNwe(vRpdPjl`YU`y
z;x)ioBh2oq!+8158Pv~lse%9h#kX707q4oXPIKkX7|POd-&Dg~RjX>djPnumDg&8O
z8qL^FEc5zfYD#V`v<CF19ASv^i9}FUZz*Fj6BS0$t<?{X+1%>Eiy9Mt+_;f(-2>o7
z9O>t?5V2C0^oAHqhpPp<B86cvcU|?DmlP)$$#2Sx3?B0eG#`X6YpZaK6xb9vGal~N
z-Mm1=@D1i_YL)q}c2+YU!&*%m%#-}efsVze6DU!bL|L~0Ie+p$y7%?q_FJ{$1cKLF
zcYh%+>eFatxfDc24t)JF=>0>u*zuG4#o+JFljRHf7-AH7m7SPtF1P5iLq{zL1u6QT
zCO?xxSPF`yAGq`65J*fwXN%|Ov*{j0ux++(ayO}WauZDqJfxnw7cR8XeX~top~m)4
zc(HTz9P{Ubgw|NR@Q=0b)gC1J*Ewws9bqP@hlE?Z*AGHuM+cZ`^<xaH<O?%r1Tul)
zNAA1H@!}k{shP93-`2ulv{b;<j>=8Z^@2<ef;u?7-9Sg(dl9wZS#;JNbXXK$|HVjW
z_A}5jrI=h8_<PX>dNkG`I!Xu9VQwe;sX6WIzxW6V&uF$@;#>|uN==JkZ<Yo@*T(nN
z3xWK}HfFvD&sIJBozUL5&620V9MXT(!ReXY;Z5#__c}k%PslrquXF2R39j&+fQE>&
zicpKNQ^b1Sb8p^|bvf1dw}OP?bc;qzGx4;`#jKj3COncovFpTd_@VM{f%e+QcZRo5
zwv!mcg|5;FyV#N0ur1k4h4M>5V@9V4c2SDz%>E;>MfrKhb7h7MTxDoS_=_mECg9`{
zIA!qDI?0>C!H1eVo~HilL6~V+{tZG5?av#5d%ZX|LbpKIHHY$s6YM-|&mADT+3(76
z8+zvt#8!7`?+Q5eYr4#(93^Rb62d4u`o7ls^qv;L#uA=#R%N!N?j&!i9-eBLl1fH=
z&#xbbT~FdH{c^QI)u?xm&9J#-t~Qc_p%1XHd#uL|C02uo#G2kho6xLGN*g1~ki5m+
z>P@loM&!qQz4PLGomvlZ$=L5lNPGim9|)o!ksOn8Yl9fPTf;(<_87@1n!Y@HA&09{
zaY&T!lzf2#8NO}frJL}1t30Ke7fxLeZA|0z?d`L*GnF!<6x+=NVuVaiMN6P2`<C)F
zM?(_XwT3=&i_oQO>dY--2};bq^!=|qH=NBYnN&c;C=QRxU^#dQ8Q(#jaJZpgY@A|S
zQem4M2+P-n1ama4D>jIQ%-)avfyi3(5IX!}sBT2oP2P$8NqMgFi}wv8@IGFFR~baA
zL7?_g=v@H_YqJjJD!=txz_Tf0L*X#qrul3=7YoEV3sC8>7t;o<x57~<ze8~R!)E%-
zwLeNb?1T3{XI!~i#;x?>A?-prbR?K+zI1nGIlOFnje9^fIq(VsI)oHAbeh>b%6w)t
z2D7u)W^LE1=q56iu*NHdg33qe(2@TZ82AsqVFYz#+tPq*`O5aHArfPYjBlOxAzuF^
zync$)R_X5axx3R6Xj>&i1-Zopm&N_{u2P$+<`~zOtE7ge%fwAuLevQ`Geg}1S~FLu
z{9*z!&3?Mc$5~%KKTU9Y9aC2V{m3Xc!zh<B)_O?RK(C(~Nvg7{m}5qgP(Ja`2Y1@K
zL2?FpXZhDoEmo77>&9F{`H*Ej>ekn-<z!M*PQ7fKA6fpU(@nL<jd2)^gruOLEOp>9
zes1#z-w0&ZyBnqxwj?gJu6N%gErv?|ioskGyX_Tk3z$^rcQt=(>RXuZwaD#1Fza*w
zJ?-yU$a~Z!<-(k@Siy<ddf(Xz(AHJ|6DCa5{1iV^jc|RFv{bW0FoI4)z%o-u*LSQZ
zXZcvS)`rHNJ5(|;!C@W8?52~7yN~1fcf`jz`hKaqhI}d)==$<8W^;LL^vjKx<<XzC
zZed3H@ZQto3muJauU+?fs2)%hIONBcd+`*N2CNnXb@oMz71K;bl}s3GI~nV2tZjS5
zH18nUqLRB)D6YUcTK!$F)6jQr%5tZo)i6z~>H)3(M_d^v9X@J&$1$vjZZ`-!yV*>7
zIY;YP!s+o-HI0AG(l^ER?mE4WhgomKD@qHwbC9XP)B1T0nE?>V1vwlKSVd2sB@=}s
z?<z-}(0SLdS08_G)x8#y@?(BdU-#W*C+e+k+Q=tc)^B5-(UyB4E=c^@B^vgoF5jD<
zd_eT+*yLNKithR^BzKHU%aruDX=_Gw#YeNNezs3H+_t>}kT*M*1Xp4eiL33A%5ta6
z8!Wr$+(}&RtU{Z2_Tk?ff&S~bLC3~w$TWed?c2x3{8WU;Z`do_<|!}sS3it#ExfiJ
zWk!3V(!6}k_MP+WsCE3>Ivl=E*_sR!J+62r?P>sS`izFS4+5Cci~F50eN3g4GQJWR
zrx3v_e7cvH@!eXBG`k`eAKgAw7^b|udt&``$x9<s!eE#78GY48kr@8jz75_%6E3Gy
zQ?Ff9iV%ZI)&(sT_4L$tS9s`_nXbeknI#?7g?_&VKjgKIHc&@sg_CP<Jb_~j3nDmF
zC=SrhRP(hrJilrbFIteJ_Tr1iwQF3@HBv=Dde%FX8}Cb70kJ8U_t*lC>htU0PApmO
z(S0$DJ9`1Y0+a?FIs&qYp=n9h%=26C_PcIY9jvhplCB}}xcc@#9AMkY#KgR3y*&dD
zFeZ?~cuTMF#2PU{30SIXLT(*Fp~O(KlHzxcMG&{lHNEHYp=#ncZ)fZlAMz78Y4cH#
zi{^{L3xy8h;bc0C{D+c8ktxqn-15ShpfZ*!`tRxQRRUi;myq0!G6dE{3zN~*RoK(@
z$clqj;^K5dkq?HxhA84r)sb_FnKNW>S-x>edR#RB3g>Zt@HU4z=LUQw`S!#h2D3pg
zBu0PpfXy}zV+u(&jMKE;1XP5)Lxiz!fL2h#*z2kX5={v<cfA+II%>nlBHsT*H>{x9
zP<~=`FbrE$KwJe9?rT>M_Di&Uj)-AF<O!PU?Zb2^&sE@c>lAi*9dMLyo6Bf9CX@b|
zewCQ9Uw0;;EB?AX@feb_?F@RE`bU`H0_D!KJfP0}KiNtag-9nSTl5_Uk~|Nx`5-(`
zuE>Cqv_6-Lj6+6WV$&Qa)(dl`y*AGSHFIOh*>!3n3_HU({$gA0MsyIT7SMx}&Lb43
zH;9{D(U#Z=+=!kKlUfE5j=#x0@Yq?$wt(HAq?2q*USGW&*j~<O6XE2WqT8BWw;*KB
z`YE3zCIodQ=JfUN&wSDZoJAHKn-oGwm;J5|krFsGyEdbpaCvqsCX>9fmER;ePzK94
zvqIiH+508^yGaUbPtUi|n<77)gzGmnmmpkDSoRTkJ{2VcDO6D+_j3~XC){kst3tiv
zA!l9RO*cd@OZ=<97pDHwI*HGNCaI!qS2{kz4;|{xsZx1qsUbu34xx24;+!SbNZZ-Q
ze~lo5PL!6F2_<=cf5a@8ckg~PTa)h(l|kfuJ-C*mb#N)Z%he+CrQzT&#)bR5ZBKb|
zq9<_-<1^A1LuZ5A9GPp4@-u}#lkbl;j`-L3-=;d=ac+O9scLrI%K~ICy$dqGVi&C1
z1m`Q!MMTI`9&lIX?BA1a?*$2*1nZ|xCt>(Tgp1UNn)D0fS3WTjvI=!E^Coy=YUm}_
ztUaQ(1p2fdIO|ADdQtl;2=Yx`N^x1JZ>E#*Gd<%|Uc`-d2n_t@s}Xjp$-swGzg~dh
zar#BMpq0#NNXy2NoH-Ybj#)L8(kDA_qg!zc?S^5c7m|p4rl>VV+kU|+<Xc<N?v!wt
zLa*tWq}PzLDD2d`o|jR-pv&7;?B|N%{AwRi<JD4T_jDNdRB};`>&J%zt;2zYR4b)u
zZib@NfI1TQ*fnnl@f}~W6fKD+AK`8#OiexEIjMaqC@fLYMHbB$n;y4CF5g6-!%Te$
zx^nWIr9qMi`f8rV?&*Yb>dV;YY6TWeRe@}41#-1d_m1vkR<0ap$h*j^PC9b5o__lU
z)zpb4anW}Q8E0M1vZWV$INSKfXxjvvwi&jv4CvO}H{E(q=WB#`_#Sj0x+z8Ro2XZ@
zfS{A}RMk9S;KY4Q#pI6{zAaK!FN^QU4PN#zsz$Jb(23Z?CK2q{7tvkOH!~DME^nCZ
zir6=mV1w7gue?((A%At*iqfWL(<7=tR<aKQdb7%VOWLEwM9sPC3wxa?L=#hx^LYzI
zqo<B^No9pXk5tgpKs&P`bQ7JxkFA~LjiYl-{+v1R^V4IUYKNEGuYLA4M!5%jL_(B0
zM6+zqu79(qw>PxRnYPN{(AvmC7hQz@sOIyMai)KQ8!=UpUzq~(^s@71)0zJ&=8Rnl
z+lf(ToNVKyYuv%z=ezC%UIzX*f)Inrg(IkVo=#|LtyC1=SxK7^eWex-^F+C&zJj74
z&%EbCWn@E`t~W3|$H@OH-}e7N$NpRG&@DsYRObxN<i+F{iYLZDCjcGe?|RDAz&*0`
zk2z4xgAfK^oxxEng&+jTDQa%|T7pX~SQ%{HyG7_cEoNlq6aYPw=m7Si;X3c5s*ZEF
z32CHJRwD=yh^}`>P$jFHP}kVUz#tlb1jNGpWA?iOB=h%geu>>+^6FlJo;jN@CWXfL
z-nna!ZlU>IU@rV9ASWY=W~usrC8K@Eqps{ANq0^K5_JV|cJt2*@l_s!6IYf}q6lyG
zJS3E%8REmiD~I`(|31LJ5U@mU{ieXYupldOl#Ip*66Xi)EnV;)Mt%BVy)ld<#N5zv
zn)DmY)e#*9o<Q?^yh{*#sG}DVAkCcWrQAR4cO$&s;Hzr}I^#SPITI%~XM0nEH)neq
z5oZ(}i#E;SsLrO<NA-vN@=kwIH{0{h0-D|ozD$JQ%8wlR$`Ei~9|HNr6?9km_Mi%G
zCA+K8Z)eL(xVqs8`#h!P67Y+_XpXsTEt<ffqL-h55ArjE8_p^2Wxtgy`@-D&7-b#+
z05wbx(jDQGE>cF8u^xZr{C_xMzY!+efe!Z)U`HH3Q39~}X(Fl>P*x^gYO2G3ErO3P
zdQtIKX&fBxttdR!2GiUR<!oY#cw<fiC<pUFvnPKb`rq@55en_E`wdCYP{yVNEx7-;
zNAcf1qW|;j&yK>&W$Ts3@efz>9~HyD{%>zm5W914)27JZl6U`}m~BN~uj^uWDE}WX
zpt~1&3rxPbB>wF?@b6!b?_!XAai0DX$^YX87$fX-36D*+(my@PY~Bbf1+j)k+LaE)
zfK8;u4@!LaPny{H6u}D?r4rEDz5qo@c~~rUcB8XI`YSv<MXXQFAsETZ)lW4DxdU(F
z5#T$XC-N!VKdgz_3SR-KDu&v7_;4dtbQmaxv5^lwd$A1MdJTk?ko4O4Ua%DL77c<(
z@F-zE`V~^2?SZdU;s++z5`ua!BqhPmRD~qI`c-7*H`72`SSCf-CSdgRbOiigte+qm
zpne@L8?$=wLfTK3z4Q15-2D<n(iKd2{;p1Y3NxX+KnJqJ&tzOLgIQ)v&<V;1<nz<+
z_KMCY#`3ZCx}&5+AixIy%q*ofz{R$<FzdF(g@lelM#T9NjrLLIdA1$IvA9uaaNqEy
z8nPa>sqS(iSN{o7a_&i1i2uXqFNq>B@4z_?0V>kiT$)y7K>+$#F#<Nj3F;x8+p6~%
za2qAN^O>bf68x$ah?xSzx3x4{8gnD~a^oblO!ASq;HT_4#>km#?bOnSG32jpL0RM9
z%h-%?W7)v0EEVxxDnS=%69n1-io4*mJ!<$$dk}&St4cJ##>x4Ao_H4~*bV`{sn4cB
z9r%-BiU3u70GJN!M}em2=WGIr??_{zf8Q;N2pk<+h$Y-MohYW2`DZ?{MkjY8h@n=s
z!?1_7>jkH{SbT%l8T)=3z4j2qJr9GbYXZ<)EvPV$2f^yS!HI84O>oHo#5BWt`Cl07
zKPl~|;Hc3!#WcqF56=@$*GW!%zqIl6jUh-0;KEtNp&DvI9vnnxJ3<2QP-_IEMRxI^
z=~PONRuv*thdZLjOl5&<CI#DL(<2C`Mh426Iaeu+?|~NnCpH!d5n!EwWC3%h&#!?L
z8#QoNW5pVG5w4ye0ieC{;rVpP*sMQi>6)X;Yxo;V$VKR*v7rlBFS!D`It1ji)wM1m
zTi$!*W417`YZ%EH-N{W6Cawi$R{W%rcfsHjAp^4CI$5u%K0=->k|IwqLMgvFLRj*#
zf<So{9&mFcb?|LR{xcgS#E;h5aoCVbQ$L95#OFSp2c8G^u{<U!lRV>Bpao?NJJO@c
zxIc@q7%yU=@qfPjmlbKpb1|N%>>_@^pX6EG6B3U0GbihC%HMD%dTl1+;*X`0p4&c;
zt4FcR92dX(oG|G56qoGJGkA<-FR4t%?<#4t1o*F<Y<D@ndOLIGUPWp}WyO59YHDn1
zt!vodEO63FjPq?sk0MIP!x?*42`vN@ar6NDoQL)sEFeN5fK*_{^5P5TGcx@z*X$(e
zpiuzuycKJopngYU6WCqeE5}~^1pgT}mJKmf{%5$6jgXo**cfxLw`v8PQXW{li#Snk
z0EjifTS+BjQ_(F!vUxU3u|>SaknDGS%~g%(iZQG&U<(~^(jx<(P~~ssBRJt*cX08k
z??UnqlDqLvI{Y*eODX;I;{92;PnTwRMt)86*Up`XWn5CW1(vs`$g?PkxBurIp<4;P
z!E)p)gjiDsMll}IMJJ!Sm2?zmi|~O7Qu&F|+5QX!+ITVj+~11QXZ3R}qld0GGsHeq
z=-+!sAnA3vdPAiQfP4azTqLj(LEm`{cR?WblVyQ36*;oVqMl~_>z?;7d*PeQ#32`M
z@O^;~WfFYv0n*qQ(~NYX$LO9K8F7JnLb86la2|Kb8Ah5RaL0a6C#G}$j-`^ThK0uC
zkfX5U479+@xxtY#pp}?QUqOt$HBX`Pdu_5%`4^Pdjj-xrsk?yOnfaD%<-VXf={m8O
zAf3!kLo{ca2-hiB7~=kcLjB~#PBZ@{dnA-vm`N`q+Tgit1K1iGf$mJmkQ*zxMwfmR
z0h^rwxo-r6!2*^XPLtZ<V*CK_yIjQMN)R8{+(v~au7lp?-DSi9OJ9ANZ3amAL723_
zPB@*|?+ooqCxvtOj(i0#i0~PDy7v=cbT54g)L;u^3nY*j@P3Zc?g>-qRvJ9&H>;Uj
z^;3@iwW6|KD4=p>v(G*oiVl*AUrzZQYn&9cFR1bSZfR<7jskZ$vYs-<PlLOJdJr7@
zbx{wKN0OBmyRNjH*8_c#G|M|lLeQwQCF&5aFM#fFH2|N=o8=19w~z#OF5=!hRmJNm
zdo+uIyo#^8V60lQ-Z{hbZUCnpji~?HE@F1plNZ9OT_SHrbAm(7{<B=BgP+wcHY$@>
zfd<y891!)>8E+T)S=%vw+YGW@9TNR6%XtL%e}v>=UBjIr%5|9wX5=?M&~*VeQ8V}o
z8XaOcHcu>2{U$m(@<^eO`Q8keWERvWCPF<*X;W)5tnMzVZ55VX&{>D<v+@I`yICpV
zFA+S*GMCk9za~hrPtP~b8tQbT;krr>?7gFNd^JCWeT|fqm@4nCIaiTj)IvA$u{)w}
zTL6jOmy`X{De%*Mi9MG7;Kr@rTcx$f#@R5j_1-*%J7;^p%*dAx6CN(ZcAGKJyj2R_
zpK`iv!Do9gMKHoWwbj1Q^QRUvlAsaRY0Bs)kK}F;)H#^rMqfE3iVpJ)t!%_6_1S#i
zdpCjO?nVOMx*usx{ry;y+4}^6c$p^-KqQwLn?v&HslZXlL${@JW4I{JjTn?ayUj{j
z7r>Ib*P=`4;C8YWNc4mIk{5M64~kRSW08~!s1v>tymFzRC9+tvDNm!f2KuS-6hHZe
zEx{%pH{F?=oJrJ8lH?-G3|H?z4CQTrTIK$WGfPN546MKJN#5*qhu1A3;U8`og&zo>
z2?avpHHP>Fh*hTR1ZTi1Ap4^6D(xjNe#u<W6}eowc$I2{-c+A!2NMz)F@dHzEjasz
z=JcFP2|B`KGgGJh_7i7c7QgUAw<4xc9G}Di#1_rA6#SMY5Zm<;m;+;mL#0HBZ(Xoi
z{wqj~&%3HLyzp_v>N7J!ocqI`*kM5(L9L(56s3`rr->wr48awVge$kTxHQ?{SSnH#
zfFPSMv@v9o+JRs5-a>2yFNY4x2m4gMdSQt3HY+`{TQAyVgs#auxEQEw8FOy14XPIv
zMoawp4#D4WrmC(*<FHCaDwxAMugLbH&Xd<VzP@7I$EP8={=0%k7MqX61CbikA7sss
z#~ybai<={=@!1#%j8zhxyGV6pdapN38<rHCNoejldJLU$f5T1p#TN3$iIqftCGY7k
z9*sCvGaJYlW?j!Y!(3@_z^$Iw<j8o}&xnR-A-?<`u=0(tV4a2ZyoMiYe)(rEoB+Sv
zg(H-o^J=hrxX2Oc+BJlO2PI=+N7;X6g}GMqS{+xqF}x7XZrtA$-97>Tx7c-IF=jgu
zCWc<yKisa(UBe1ZA&BXA`ViYp0zeg~J~S<%XSU$<X#M+OTm8`)I>iiiHw)zTKp=Ok
z07uk5Mc2)7HYb+JaFkQjReyM@r&>#I<XHH?7>VYIbq=Hb9rhFFSMlg*XXy@-hRRRO
ztGOL4?(cOo7uSWp5BH_ny-?V=h%z`2wRNcIxz(9Zk#~=IEsaRd%d>PP_8%WroM02Y
zAYF2Kgv#4n4r&B%!F8Iy3i*FmOo)BPmcjnIf0PfKIH>KY9ZutiA+#Y@=0$SR>H@4p
zmT`k}65PinBC%m)455LZ;3s>mM@@k9Wp`th68g~MJ|dmw6PTL@`stSdne4PUh#a96
zCV6{Pnc*s_KBe!Ya~4Gro`F)y0ck;j!bP9*;VgamJdC1KqjLqv&r$=vNd)>ha`eS)
zGluWJGbHVP)H(c?R0fVAk<}|s-&-T>TU7g79?c8Jg!do90dCnHU-x8{#YSxP1Cg~a
zKQg;@2)3F&CN)jpGj3(lRRJ2$(DrWUr6PT~(MMX4SeFUhvO(B76EQ02q;Hw=3k5#n
zxZv9adVO#_p0oWKJF(CFOuv+0j)hOWNTTWP5EG6aSIW?I=@JQ7nXU%3eS%LrU!YsS
zhUo2~nJ~NgN~8USB7w~Iu*y?ODbM#aM;DsqRu3|w&KQzKEW-iz+myaS-f~d?hY{us
zy1B6Y&8jrN;?5C7T{+mj`yn?FuPa`pAo;?!OSp9Jm?U+Uvib2*Ny+D-kJYKnV~L{G
zRnC$Hot$Jo5F2`e@wE?&4lA##MMP)bZ=DWHPg{c|v5bCK^<SLg8=?o|K+r>{={-7X
z%`=M#M_8Z;X@-}n+IvY)cu>|K6(RBFQclq5vu5hOgoREAPUC6-csXN9a<MBHdlVb|
zpOG<8nw{HgWS%?fgO$W0=1&s0Xh+|#!zzDXh1d(4{7d}oTbCY+P&=~J9M8XSc)qfp
zFQ469=<=E7=AfK5nyGjc0~b~{NaTKQ5o6rVx>%fL?=nz=w%a$5`JvULcjgPtC_H_`
zN1UcYkJcpE`>p`cW9CftHJs`-+foSQ#Bx~)u^aJ<$Ig-Q;J}k|O-2b+)?d#O5aEtA
zP)eoS#VV}EwMjuf|IKm7KDv<Pnwiou12SP^6gKto>8>y1fvBM%g^M}yLY^w;;%ihR
zuY~neS>ScW$nvO^{&856w+OQ?oBlie<6!3VLCjBsC__^-q@(2cUhiCf{f29#W#5>t
zKQk<qj*{WhNI<wTRyrxDQKi^wsnMBdEr{-SU*h2%`>dCA%+ybrF|1X-9`_9rjp<U}
z1r;aZH#4qxNh;XD-W$X`AI$8PaotkC#h5Pd9r4>F_vozaihe;)l7jq_g8Y;4M>vF1
zU$hrYH{voS$@1`IJOg$uNo9DlAHS+TpWoN<3leav;XI**%s4bvye6(w#FNs~B-JD7
z&)broNBUc2hMPBEr|ERT2Vbk6q&9r;?Wk-ni`tX1GV%Fo5k0-SsfnCc3U_D}IIs#A
zzpX#Fs1wnZCZpW78Pu(bG$c#cCVR6}6`j{N6E-nqe`zR`*w%D9c#4v#&E1dKp?8kc
z!ekWJk0<MuB#Rl;`nU#|Ge<*a>4~k{xL-G>`<lYAnw?bgi5ECc6Kt<eY7a~*E}i%t
zm|K*Vp;ptIq~ctSH_h6)Om<mXUg4}lljZ=C$5(li7F<-tK{R*8^m+3Z$(|lU`5s*U
z!1+|4{MTP|r>UokSd%<R$pi<^D^bMn9$KzensAJj6{khp!1HTXVHV#ZeYU1qQs43#
zvtY)?a8uA)QZUz3=BkYTe8f{Cwb$qsNV=H5daGevA#I>9-9CKXLpN!=qGg?k17AGR
zN9fo?xY8r3rj|NHM<7X7iI-P@pf)ipeRhia6<@u^Y2l6Ao~445zGO+nPt5#{a2ZY@
zasJZ!nG@|9zW6BC*j}XK1GAIV{gK7j<!>4eyBTRAz_91=8OsPA6?uZ{o#FP~*?#8U
zgB$LLcDstog95!yMHH|sd({dl|6>Nk>_Y+slzfM`&tzYeA1g$w+Q7Ky9^80JMh0in
zQ4zUy@v5{{Z|Wm=*y$M8mv{i<!S<+(Rtz!yUb{Ye8h2J?knw84tncv68mkDmq?9!K
zN%>FWnR#e$uJ`3oxCfth)KVXGGq~aNjL4hx_nC8qOlZO269goZ?&v0t`-(km4~RsA
zPn?BA?s|jmb5u8u0Ka?t4o7v`s6tP^;zD}?@vgEzsypciela)gc6eEeOtM;u!o?$R
z@EjqO#p(KF_=EP(3?apK^?pA2`{nprEDeqZ@AcIEJN~WKL$=4--1qI=&)P?gjJhjF
zpan*2Qdq+}V(_KycPn*s$aG#1dQr~hWoXO4TjG0Go0xLN{*fBXX!u2xq*$<LwmOIY
zn_zcXs@hyq@66xZqm2LbAtsO%YpM5Y=vFndZkg3;TA7dCo(LPt95b>xTXTHv;Npv0
zDu&l6)wmH^<M3A7i5<)&AN!VtTACV<V>UXjOeCn9<G!VKeQgR^n5qJ=TF=sng?8$e
zvLfQ~vmCodNI&fzevkb;*~_?qvNQQtYJ@c_hMDijhg;?o51v{evT@ioir_y8x8N8l
z=!(HV=Oj0O1jmty#a-4$=7Md7y(J+}LFd%jY@6gYV|Ie0L0&4T(b0mZe^VnPf3?AT
zL}lnXbGThDrEG?tqfCsfsZ8&7j7-k_m&E<=EP9JW`I7+Bre|~=ax07F8J|m#<I<d1
zJVYiM!xWbHy87v_<z-0CBMOBmvhgo7uP0>Fhq6Cp;H5VSD)b{SPc}+PAWt8Tv<`{2
zD?t>PRDq6bttEZS9{1M8clmn<gn@VxG1`Cstl?FENAfYIO^)@L%aq41orK{FeZ9|V
zi<I-jClmC;$b)`8S3$k7e96;*H-tj%6tfHaqIA!%?EaUAWM5uHW+i`cheGB4oWzeW
z3B8*U&8IaPE#4E$UHY7_8l{jd1>%*RlDY}ORi#t0N?kFX3!$8IWwMU9+E;%~$6n66
zD|=6VSDTF5wPOAlrAD1Q_ArV6go%)Bqge!}naa0ED@I8Jy82DJ1eF=<g2n8;hSwZ=
z*YA_pwR6h84WcVX`>C0HSc_q<Tequw`NE<FXYDf5f2~u1aevWB^doJ?uu*)Ee-Kkf
z?`vD!!UftLD;0b!E3PhKV`k?>uP@cGVTJ73n%dk0OB#88%UFf&*Sq6wZjqB_w^qM)
z$*9C@_KU7W;r-Sl*LG+Y6C!^Kxl*4i2@2i5lcy8>_1vwCcQ&<_=PDL$RL*E7>nf?n
zX6KeFWw<8zW$~(5P`w0o?b5TFQ5*r)g`9Nuhv%nynQx(&=9Xe^y6L^s_e7lmeJ5Hs
zf<V~bS+F$LB5B~)EM@H1_-{!1qR?g4e9`t*yV0uH({v(V`aZ_4@?BWA8U6FvsTc{u
z^7B(sdHO$=pZ~j4p*mKq;}KhtPfGPtzD}U@&Q2#$WfpfB3HAbgN8zlR{pg&+k*~-W
zPj!yIcR{w>4yvG?7PsJU>H?O*8dRsjV!s+7y3e4|AKUK4p%M~I!C_&nEw1_FJD+2F
zbT}NG`H?dq7s{Xp^tCN+Ky%B}h~CEFwMn89u4wR-N=#m8kPhaWYPc<eE7g3BMJ<Ck
zQ?J5Dwe@an@BKo1QP4PKjaL+;J@%8xylRfSq8Cf()lrK#Yp11Q*`rkBq8Ir-Rhi;x
zUmue)X<zB<opRApJT@h4`Q_o)g;gwaa@SmPG*vewL1Ec)Kp{TW;<F;5@tTV2Dk)`9
zw_jgi0@jOu|K@&coW=<oD)Ft0D2~SM?UV@%bZ$6l3FDTjqe%NtUZlw-wsQb7R=Jh5
zWB|Q2xT?8~X?acWlmXu1lfr#UNXdD8<Ia5wQ;OoxS#@mMu}u#`2Zx~aTOw$-q(YUL
zk*RCh;PY4$>yti}d&-zUEeXuWoLPa%m%N^{J=uIc8QS*23iPO=*_?WnD)sk1qTeR9
zt-9ZcAD*|^{51}D`6F|?(;@ec!vtG9jkF;_5hKCTXlPIzb^6I`_qPoK?_~zDi?kxf
zXCg9(eOS5Z4svpjc>!pnBz8xa%}iTnz?cK(`4|rAJ~<_nq{-(ZSHX;bhufw2M%WZ4
z$&Ms<5+o&q#Mbe_Zf^f-sbL3EPbA07cb-Mkx>sPl)^T~|SzG^|6_C)jf7L*NP2%SW
z@yR;WFmsQ@)8z~$FO91X3F>1}Pz^aMN{?*VdEV@Z?mUT)zP$E1pi*xB_UaPBO8#Q4
zMdX1{oD_Ry2=x`5#-)$``ah($7yEPC;SpFK1#$eG{qVb;dr$SKW-hMrNm$BTEMbK>
zUzSQ+Fw!t4VSR-H132wdgHu}kP<fXREN?owyIE_UKz;nlD{0T9e_5Yfo@l|7zRva)
z>nrpMD7$T9-{o*K*dVHU6{Qrlw|A9YKb?r{Egr*d$sNd!+L(|<{TzFTn|c@pBfR?k
z{59p{mwmbxWyeg(vPaZPha{Dh{R$Uj$E+OOBlXA(g*^)%nr+4>h1eLZ<ho<2y9tl0
z)I1DT`v>j6;$<DkN)&+Dju_8z-<VIr2L?qQFVQ>4OcN62QaX7(^z<@3WQg?3D)a-I
zA7NXQO!rMk8>Z5GRg&@Yug)r1V>k)!XXFgiLh;;#qihEFvppP{YuDp@tw++9cP8py
z&zxJdG0SV~D>*PisHcg_zVZ?7y|IZpNw_BYhgL0QaG+mHY&OqiKvxjnC&M);#WQO(
zyn>3?`OU3Z?zeY}&@u|+<y$<#))WdM_guev@>0h1+>7bXn2uAlz)id%rrm!Kx+v<Y
zaIbdhiE9z-Tfa~Fd?4!mFm7PbF{2DBSglfI=k(DjVwLS_vasMCNY{?>YWk&lsXt>c
zHg9NwQ%*<i00f0I3eoCnEu_H--)CZ3SzbF8Ogm-dp>unB#iYp8-d~zXaG6i<!Aaql
z%_0sn!cyock9Dso8@!r}yXx+F&x1<ys<<AHjmiA9Q{};6X$^%Vq<OyIn}8MkhuN*K
zGY+;;%Sn@s?p<rc^)*8=$EV0$PuQa}-wcye`0rm@Y}Mjby)`>;@iWA%v;z>P9+mt2
z?&T?E+phNs>nc8LoY2*gwpVJs!1`zHjxL4yHXN@}d2{9Mr-nTpG8Kvh>BjWQgsWA_
zKZz^Zf658JYMzueqr|d8ZW!{8<-H&m!BIakYSyCz0)J1LNWO_B)+pkq(nZ2Tg`YBU
ze7XzWS1OW#zM$DRdG$W+QUqeYHz_Gt?%#<F^vZsaU-Tp*d<IFFScw0&8TFdyeo<{1
zB)wETGpIjC6@sKF7v0G|mAkN^o4Q^<gw>#zOH)VGjIw?e`2mXYdmFhs@8}L#J+><+
zLgPl$G<dF2+28+O^15&LK1$=gg2t{yY3!p9mz=4E-LtG9KB1rar~8~dj17_tem`_u
z<}tK3%4eq`d-tPbb>sy9xYN5}S5kqmUV+)4*#jpqEf;|3S4*|~vk)Wc`&567C29xX
ziS#MlZs_X-*WI;0FCLCR9!g13lp%6Z$ISUWt?NS>L^e!1VgnC;zN!s@X;?Nzt^RAE
zw#nV3xm-G>vZxl7tGA@VlV67TF(vU?&L>tdLaEUBySbe@^r-O9OI`UdFSXFRlg_sU
zO6+y}x7Huak))AHSC!Hc$xMd<PQISL8~91W&D+C&w$*l*KbvZ^fQwj0BaT}aksx*7
z-#_{f667O-Kv6`$699)m3uK;M6M>@@@3-^vnddH-e}7&EWlJ;22L+}Al3}=~%@6@b
z1{CCcMB2|Jnob4@ik|nFv7&SJsS&7Wt_i>r(_S}+fj@zPT?Gg9BWEKjF(nMsQ%*`u
z_;(Ub+`qoRXmM-GNyxF9aY|i3qM3B=%#l?XP!<etUy9>VO35Y(Ex%{jo^FsHSJqW{
z=#aK3oety{F!qEO;21<+>yv>9Q`2>$!Ry8?jTp`-{ecezl7<&YUZh;;q}?j7w+uA*
zLlf_b(cgZ=*2fGX2^vO^=1iuD=-h4(DB#+#!%t-_qQ(p6slQOirVQGRJ`|<1Sb`*?
zr|3joduR?y%kyg&AHPvdEa+?f(siA*a{2+pOqix{UHG`2ZOYBGuK{TCmz=<~pi_n<
z+mE#ME2I4RV4Ui^;l`Auhrv+=LI_LAwfvD}22*2Wh4Z`jboF(sbNmnmUsz}zoo0JZ
zCi#(PGWYIH_-W@$X<QzDjzi&+cgDACW|A&7#Sh^kJ_qJC?Ha}nS&7h$Lwv?pzEAFj
z3uHR7ky1FXPH4}Kt4lunl31L0VtwFTO|5Eea9}7poI3%A+D!o%qW<C9rl;F56cHef
zw6F38Wu}Zcj6S4wN=Pn%PH{P_brJ{)JKjkRwj`r+9_bMi?IV^YocnfLE?d}!Ge-n;
zbiQvJZ$szD=Ek>suXIGe1lyj5E?BF^&jrTOSBoB*K81_LCZ1X&gxlw@Jqp?LK>}iN
z9a60^oI|I8FYJ<}$);i}&$UQWt(-p|TPOOkwn}w1;QT$!&MtX}u%CPJIpK;HNqsX)
zG)k!}<ndvQn>RJF)U$-lC@<kvwSQ<5i4pEb*MYn3EL0Ljx(yNB2CdJyfA6mcz^=!n
zaj{$%cj5Oo^a_scInh8H<%KoUGjlQbDX@l&m*et{6t^}gII?v#V6lHfe2jfY3w~2i
zh?+!F<C$wnm>Wp8-U>v|<~>%tb|Z?0M}sGI6>Qj436^xCjgb04eYn#i>W}EH8tsGJ
zaL_tNwn0dyacnP|jw9QOJ3=ilykDfI!T(U*#OFF)FtKlg9D6qTtvh5K6c66X%UvIA
zbL;j-VzZ;Ed-p`Z`}#}k{{0?ZiDKO(v6+S;4qmGvn4#d4;*}FaN+wAdyh{PVPMq$y
zcX>@`<b9#XP9%S7aY-Y&UYBk^J|HOF2S0Km>V1Dpm-{X9$(ega#k$I84@N*i=g`<-
z5hZfH4l3^x;%py0_9W$p4|^sOOrj44J)uwz-X)`E*(#h(3O*iu`s}R@?5!*+-8O0s
zI_8SU%jv(75Ekxwb?!A|?zKC4X~KH2o|6t;_3JW>a=2FaqS%10@JUdoPmsowAWoHQ
zZ&hP;wKWWwtyUle;uJDvGS`so$KB=n)9dCEqZt)fJ@dBi)ly#QSn5u?b)3xX7}+yI
z^9=xcNb^csb?bfG>U~QM$7_0{+&4&>O-Yz3NnlWfoo;-zHD5Gq!@&i!@g$w&!6=$r
zdE^}Kwm<GWj1*_SP!wcCT7}DIa2))h`;AA3y(lQLshPvjlsm$bd-BS_B%ti4D0})o
zqZbHuB(LaMaC_@;_mt$~$p|Y>y>8`3_Nc=yoS+}X&~R!H%FirP9j=J(5ClJ;`K+<K
z#SI_b9`8_L(9I{rPpavd3GHHwc^dSf7v?x2OIH13@5W|jM|ui=s~9ip9M$ta=*X|@
z;}mYw(C|GtJC?R?_(?johIun$#{B0mX4B1?lP$M6Xki}C@_Hl9)b;qwRhRQ<3z&X!
znn%stzO<vaY?xGb=f|M|<V9JZ%egbs34N~lmE$|*=52ez<Ax9rm8kpel1^rp>f%IG
znI0HRHsuupDHg+--IgYmklgo-%?Jz8N-NwL(qVR*cpt6*bn)%`?UrJN?tCP)eO0KD
zf7LvRlsRo<TSVZVo$>Y<GIUq~)~<u=QFOO;MhM;KnsN{Usk9++P0>4%zuxWGHIYu=
z1<TBr?LHS0O5?M_`wdw5dgj(yHs6LgK2>av*nOLgK4}N&&S^W6?w!9sC{#Q}(OFxS
z|DgOW5RM_j64Q}ks~83+AFjaLx9Ico;)XuTEx8s7>&)hYDiBgLl6}OUk*=geyO)Q{
z*hKBi6Ye&Z?g0u^uLn(wMd{dd5?hmJU=}sex0}8~lzz=0G}A>xOVE!oQn(!PD&8D9
z8Y2CmyVFk&8R`m+1L~7j1xSOpW~QX;h>kXe2iEUUE2i_)QLeR)di{dq)ij@XYiM}q
z&`}?Blj{CW$>^teiN_JFGbujZ9leh$TAvCMk{QL>Nae{G%|&v<khQYeOcmu`Mg8F*
z1>KuNsNa}9hOiGF`RvVMUn4iY1=v3l^swu<n4je{*BCO*%&vOsfR+D_&qhoVU20nu
zJYf#z1U+ql*%f}GSibfuaK$B4V}mu(mJl8CWYeUF*gaIAG0<$4E>H?<3H_)@U$KE+
z^~yJwgQla^eS-q8V4vSJ=xeY{{$>0j2UCqhEBNgp@fC+JJVI!A+qc(L(ebQVs!c8A
z`<L8UWl@2`xG#?sCfRoahRF4p8l(uXQ!4l5w*3w$c1QY;qPm>l*vrC-1I4gME=ph*
zBYjX9afTFrcTW}%DlALVecy<XmZZMBl2PMpYm1_vo>sTVg&T}6?Y#Eg29F0C3HEA)
z$qF-kk@MJ2y>(e`?G4&GntJt%mGYk{E!NHjjyo&K`W1Jq`xkfL5|@l5ERC~LzSjt>
z(2PWbo$xoS<rm|gtJ>%hr}32M+H~n$wHICy)Lpe8`0W+mV053xmP&p-fWUe({QBxx
z$0lNG!;r6sbiAr5BHrx{LJHD2sFFHGyP_tzOIb$`MO5sB2J9<DRBptNNFuGMfUDz3
zCu%;p1jqdha65r#Mmi7wK2dH~jv!D9gTsb~gpau0y4XOXoE#YD$9+);6U0E-O4-@<
z^v%vtpce;)2sNXraWVY~XOO~O$gf@rgV?b2tEUMGqLv5Wm0SsXMDmdFCusw~5_k!}
zN28`QRSwgPV?;It1T46IhCDf20jP2J1Ogx3B)=x7v@XuWBk4CpNVg9&$yz)lrSW{?
zwb1glr*cX@#H5&<CZmA4%1<1w^HcZ+-or{~Tz;q-rGWVnh?W^T{w0r{y6(w=n|Meh
zMWD4%tw@1wL{0!j$Ile)Slyg*C8c9F@;z}|cG`t4*xz}yXU{e8?i!j<RG&`Nn={pR
zQ@==SvPGL6C+*oBc45Iba<^XSj6z7U3J=lnmS!>gvh<ySJDARkiIKul#SvMj7``cZ
zu(G3y?o_9@JvGj_=bv?RUu0Vx-Vgek?+?g#E;3MV%HA5%tf@U&8~mF^|9r$Z%;N@I
zH%@R&eoP2=a98-&aV^l;Cu>2-(~WhBF^=mJ`mq4Zuy)V&bgtj#vYR{F<`pr|wXNpd
zmcwPPP#n_ZCUnMMV*LM)XQaTwC;uypbK+DrX-&uFu2YAT*Jyrq)RY7%E@hlE$L$HV
zj#7#wItDySig&0GDcD|aVjginOx}C-OR=X-K*jTq1pa@eK>@QW?iKz?lI7+z={s+2
zlx5OocDo?urr^@2k+QPVWd;QM3Ptymq4+8uxd}kiR#+a53i@6Kwb#xO-*m~=J7+!7
z7$$P%OFuyJJqdELYzv#0A=y;dKGkNCc`R#1z&vsU($&H~Ls@G@47&frfk;<~S*+>0
zNE`@O!;^4o-tI4flg@+O@W3<?$aCNBcR*VNx}@TeA8btGjpzx~4@a}?V4i+!*$A9$
zp$k9<0|{xbbC|4qWm%5(#;XLjOB+1EZM-*5$ac`;>$#M_e52MpcD4qaAr1co2*jY`
zn3V9|{aY0WsA?Znx=%PvfzF6@L8Bn#tu{cK*13t1ehC;o-Z{cp<BAqlu>W!?t;edI
z_5g@_5+M>`uG1-GPKyS>PYu9qxR^_`AdIe<)tNGA5eQ#mo9jk^^E{}boOVrmMUlZv
zdVnccJMXECMolc8oq@<qYMJ=gC(?V7lfj1S=$aS8G7G<SD9Hpfp=G08`LON!UO|vh
z#M+AEFyEC!?Co&axOT|^;oa*qH|6*TVCpu3H<CQJ2nM^Ccbs{yLp#PFIevD4hOY;j
z<Ho@mOSL}2q}MP%=Sp|L9iSJ@CQ?X|nNHN#9~%fJ0}T%nI<Ys*0%Aqn<45Xwj{TMt
zN0lLAd@t|bZ}SW%{T(rJf$ov4`+wz0d+#EIT$hyeKRpq3XQ*#DWTm#qqcEi%fNyZ^
z$C?+kso2kH;y93e0R}v(V?q}!B`?WCem&gfL>`T)r*#N1zuK>=<ahCQ=eQ%JADq6{
z!F{tNBXj$QW2K8r|Mma6<FgU)6bR#biT~k~`k$QiUtj5Vhf7(kxvm`k57+W9h}eJr
z@r^W`IJN?1{yZE14kP%RuZwNN>y@`REh+quruF~b%<fH#@Y;;H_<wha|Mc}SX;^TU
zW63uEK1udJulRrb#swk&!vpPn{-ITaygmQhFFQJT+$*gV`x*c1zW(pm3edSXEgIyK
z8<XH1I1RAHPfy{voJM3hc;>a=go?8}<s5egnfY@w<O-4sI<-+#5%cHUit)mC$#kB5
z^}qiR|M`<`uz4eARNy<#sktL~Lt~y^W5)c(yGmm203;)52RNJRcW_K|fW+hL6_tHc
z_g1*TX9CHKRKAEe-n?wm-(oZ2mos3ojT47krK?~B2nb?+&3X$)JnKQziDY7bn!iy+
zo|A$(TGyv!WDt+zuEET99cG*qy;9Thkbip}Yz-%od5A(}YsCn66b)5%W%;O@`?-k;
z5Hu1&rX2OKyepnh2a9Y%*8(g7v7*Tnf4v4%&)}_?!jxG4lfgz9D*$f3Mi?QieSHZj
znGjRH0Fb8a_EY*uk`FO5h$b4)=9UjjyJNx$fUo3%E8P!DEBY$n$0xx-ObV;}isRJc
z$0LYbHR*NqG+O$+jikWrn<w{|+X0BCK!gZu$WHcyKk(jyYS?lGz<Fn}tgA32#~%sq
zzXe4!XK6Nxq^=vTJMloR(&HOIt|x)#J_UKP9XmWPQ?@sOC$_V(Q#$qgE0HLAKT{oZ
zFSnrm5-nSpZ4~f()*z;ZsuR>%I*Vqul%?+W{|1lO*l&98a?W?RAWzhLhOJBCFW$ks
z21!qz$p0Mw_ho|Q8-G+!E9lI{a>6~I9ALDPHQZD6c^B6i7)-COVU}GCNO#<u)+ZJX
zhqLd(8$h+1B6yC(D<b_+J+PM>7xHS~mHZ0(At_kt;oyPz2a{B320;R`&;y<I6k?#<
z=?V8BuYp100c`K1ac=PZP6A4O8j>_U_XQ-xD6q3#gu8{crk|DKK*^m1ZtXOZP~e%c
z>;I538y4&3d8|rb5Q1ITJqUdEI&{1gikASyvmbxA@L)67^8$Q3{#<EpDFdnTYt5;>
zr&f&OSLF^wK_uV*<?sGV*~vkls_I6@U+=-j1!M;bt6u)6<bL5ZUm5rAYFh9r9`lkE
zehu_$BVa>krNYh+Kcusla3_!yrV+GDBlOzD7Q253#+w-FanR$pAQZj-@eR?v;Vz+F
zIuk6FhVBuigIEjSjk<f@@>|hc_&1_wX2TCNv_-w&SU_?)l5{y6t~$069e8{n$=G^)
zsB`#}N~zid3Q*_%emfLhd##HhlBoJese1szK?4XvDpR%#+N0T<?rR=*j|U?y;YpyR
z9K6jS>THtK(^{NiCF9axZ;d;$5Y{}*z44cK>OWV6fBlVEIN|QMqu1EXYS-geVC|h+
z+;EfdB=6r4nqQj=p}4D2f6POY;0{m>XxcdipC7;PH~it5qxrgzdHu3q8=W}h!whWw
zK3w@Nq;GoY)UyJ(`@{rWAci0Ozh#&_(d4_!!P|wnn(V=soGA0nL&3j7aNanAk~(xl
zYartv*04y4H}Rw$$=rwuLAzrO@OppboPcb>hetdh%JmeCkk!_kScb{1t8n`BMJ&^c
zu&#O+WkBeG!Kd;*B~f<`Xkp2Phi$bohTrL>V?0e?F{-L0{9_w&e>Ki$<}RB|!-H2L
zb|)3?jQ|HE@8}`eGTAL|aRt$I`qvCoyGU*&hT#_Pz#Tp4$%*99?SCb*&t*5y3T?+7
zp=V1tYgR&E@e7V*iP=AGRRoPy`aP!3O`riMH*$gq+6GvTmTY&&Ry_Ejx0S@k&<yFg
z)bPljHu?##(Xn1`j|I(%6BB#m+i?#Q7zjv{gt5fo4K~y65V4##&mtPlzR3I%8AHlZ
z(2AQgGSOUoop1CJsSQos3hg(yej~xcgM@w?_28#*7Jx8eQ^l5LB2ro{)d|@T@F844
zz;ijWAzaDFs`nDfoGQ|(SKBVW@C73Tl~g-Um}OxKN=|F$WR7MX?$j*0MIjxHiL8J-
z(MoPNFYSbZK3Hzr`58<!c4ePS;J8?kYcj~b*MY2QXJB&W1D(u`peqct?NpNZydNV-
z<(iOSSJ1{x0`h8MDtB&ar|d&W5Gp+`&<TB&BNMF^lB=8W+3#A+9u3LMFOc6l+<57V
zwFq4+^th{SKfn#eTL3aN&4PfA&!80(-hQ8_^uJN!2Q8g|?vf<x;3{&8dgc7~3#R6q
zbPPx5%xXt-5?gb^`xI|{a$jOgjeCCKzF{rDn`Fuw*SLu^s#Q=EP*L|gNK+Rcq-*=X
z#*Cv4bHk;!O}8~}tQalrEV=y_v|zl9N)}Gor1RXj`EhsWc7kvf%XboyZG>gcNB~G&
zD8*Ju*|u<fBz@w?*w)(s+MhE1LO15vYr-69a7U*|ku_|^sAs*z6t(-Oh&k|r&Fnz;
z3@_o|9uK?^<=+gDb_7aeID~e-KH>Tn11*3HryCN$2_i(FJPOr)BdO<3BaJaEy4vsv
zQxVMkRDMaC9vJlV*jgKin^XD~`}v9<bkGh=xb!N#Tb?5y1cfaO-<2>ZX<9_59{_Tl
z;DYh~h~yYS`bD4^PpSFj6F!qUfmh*m@VQxt-8u#{EqCZUfGXz-^!`AudMl#3sQ&;b
zeo7zR^pd<uC*+h3<J89}j&P*G@dYBJC!yJGe}pzkXgVl-&MjFKc1^dC;f2CI^aqS3
zz4f=F?ljjdtba!^QGfl9ECAj?4MuY~MyMrSF%RM?N5-P?`&mcsS=YK<hfCMiIIh{u
zV?LB;4cPFs(!vgZlCVIsPg+FL6L?+CGEB?-7*Dc+ZNtIFyyyN9%t32}Namru?7O4|
zk6^35Rcb`7R|#l&N${46=7t5}?6&9dRenR1`L`aNZ_l4NSa#$Ze3<Y)!Ns)HYCc9)
z(h83CvY`T$SB7ChIR_yt*Qi{(S==Npp-ZD}2Z<NLwtngl*je7wbURhz<3+A~SweO2
zvX8boS;S?tY4FxQ#_Dd>#1+DGTlW8P_SSJ(rtA9m1HuqWJdz^$pfm^ysGyW|N=f$s
z4-L{Cf`IhUD4=wANjD18-Q6JF^<KBL_WG^8_u4aizyI)I)FJNsiZhP$dz|ioooE9X
zMfh2|LD1DW_8;=&EyH>_Y-!(~Q#cwerU{sM(Wl~FAQ>U8;0IBeABDqM(<oAgmO`mk
zUuXuhjMxeDY%bJ~#8f6F;R{j9ZaC|<(i{uuz66kP+%$!kAXhT=eSTU4b?AFV1iq8T
z8T)cviVFWYXT6J;p=Y;Hh)mVEw5;(!_1*;p?iJgxZ{#=4rCC{r@530X<tm5EX<<_!
z3Us@oi(tOa&Qv%)Ut9OsCbv4e%F}8{+}5~fu>{af{&1rVWb7&RHJH@u6zLehC=JUB
zHD>I^!tTXBK6~YbX!@Ld*l-iI>^Sdb^&p?KS*vxvc5(ilQ|PX?K!a1ooX+!}1Kj>E
zzjRQCDp0a98b4h1*V)QBy7wDC_5=7K7{a>)x{3zWrX&a!J_p39(6fCog5nRN%PCqE
zN?=o~2cp(IeLHa2RwI9F!bHVJ^A>@_&c0PQ{dT{qF)ha%*AxG2OE$>bKQ1`z+te+T
zJ5(`nLu&O@t2CnrWTed$zUY{akP|^pjSeudWOJY?838sEoBE={3c}^kJjW-vJ~NsZ
z`zE8QjPfGuCBxkzb_~#yEDnmskDxAlP@<x(8oTT*9uhXL*-=XqtS<|HHbAK_*eWf8
zPJ@nMT=7f|;eNYE-~2db6<;O{Z;iag2qKPmVz{9xfe6q2bC~d;_r-wJBP~*k9WX<O
zmlm3VhZ%yu)Ks_SgU=!&lD3@;hG=H^x4y;;J$RndDjy1nFeJ|*n{%$%<~WX6HGD@J
z_E+C3I+=2Plfh7B&x`xRxJ~yW;-C=%KhWiyS+xgBYf1{XBXsW;HRHtPE3v9}8{cZo
z`nV@4^Tm__Lr!2+?$j;KED#|U_1w1}a_$z`ZFNic!E5r=y+76lzAQZnO2(q0)z=#!
zM_aE5p8KO*ph4}Rh+2If%yUu%sD6*zPmykI26fY#AcnyAfsL5_-Y|SU2$sL#Bi`=2
z_EZBI>4K<(@8Ri9f$(WbU$r$J69)H&n1z%w;pL(ihD0Neg5%BH`YRi*EJ%ibfJ4o@
zZ0H9^gdn_HH`QA2MV<u0dBrdH6vZ5)pZ5AezReaz6YfGR@aFxD;A0~bqrd)eT-~-)
zu}QzHH}<=_!6_Nqiby)R1Nz;o^9v|Y=75mmXD?}M_vy4-*?0;^Oda2cpZz9U%sa;4
zGe|viJ$R2wgSnB+k6=r9SYYnQ!t(~f#Oe`qQd>Tvp35X;zJalW^Fj)atKf8fU~Wdo
zYIo1$sm<KDw^yPgjvKc_bdsL%<nqfvFfc(VPu(jybL&?d?(^(jc8u&_m7~AP^ZU}#
zq#H?C)io9CctPu!c2&bJ3%L-RV7AqQd;z9~QUVuJu$odt{BxfBV+X~NY`2JW3hK`9
z&E#>Se3+daP!>H_EJi-jZU7>Du^_^gVLF8m$KE^%V_=X;L(2P}%r^p7dt*)xvF>Nv
zsrH;rAy10%PWxLC%A>JoEv>3ERs`ElrC=ylk#kSwvd6KWK<1PnK2r3dmes;%(5<`X
znD9T(8I*ZIEE8~-b=p$d{k%z)^={~-pF3J%kyOVxA8?5IQw}X0bPVTr{dOW#3zrE{
zQ??T7b^_Ur2tkzch76LF^<2FSGb*}zX2!(Oed$7fVFa1d;bvcW&Apf1TZhG%8aRr*
zW7u4Y6skfZf{_b_3AWPa^A>S&H0n8n-^h)FM8blAtje$OHbI9dL8lSkMZJi;UCr<A
zL(Fve!Nm3jGxMRF#S|l1BDaE|T1=|QZmtP|P8k3cy6DKjRBb<D_i!}nbaPax3cqL=
zX?5l5Q1lbQIJ9L!eG}a~UlnLMowVmq_g*?Ss+x;Kmh+1wr(X1ISo_{AaXC{0k^@Sp
zqA%iKk6u7%%QqVXxCfS~7089}O3LHS6AI)tI(bUOTMs-3LmM{e8a`5~^LjKRQmQRh
zCdt+sST_<TKL^$&i0Z*PukPu?ML*WlC8~~`i*HANwP-8o+?_NmYdYO?rlWQsJQgGB
zN4yi2dP=#CcP9mFB_%Lfqj@M)#%{r}BlzB8ulURji;#tkC+k8n|5IuFwTjM~lwSw!
z9W>NS^ml)!Jq-|uy^B1)^+cbq6#<NfJj%X3w-;-1#>9qyqDEz6ng+5oq_VPB4e(RB
zdc}qsXI?Ds2UuaoKi;%~^l`L5<;xMnR%w<hBwmKoC|gB8$SP(q+5;(dSd8prwu&+R
z{sDVp?4<oG@llPX0wE64RgJPb<ofx3?77v-z<{S{5KNWU_lLHw$(g3G`0PH_4QpqT
ztz<L8S2=)m6-))kCDNk-yNZo*<zo-S?WZ+hw~ms4*(bezvQ!b;w>IkX4J<;R3*}nf
zMcNRy)vCAPW_8Wv2=-;ZG3|Q0cd)(zDzTtiH3pvUW0GS3bb-*b^5mEF+q9BGefrrN
zv?+K&+{@g7RR!9D0s3%4TK$O>AMscZ^#_e9Efk|P)(IdqF>S32@P*G9iG0Sqt7dRo
zCF;lZf#DBNi8Tv(G+Lik(T8Ur9WIrho=snpPVN!hg{?4NpE&aBKWe=Uxm4auyw;e!
zznR*9rT%>6gI4r>Wf$_@-b0|H_6Y$hv#Z%h9(zIG&aE=b3<wfWwUWMIra?sL>gl!>
zC;E0Bz=bA>#0B}=;S*Fz6mxBzKaJx4TLtT%w`p~>Kq=#j^n*o4DT%#mK;e2yj+-qa
zfny59rGA#zzYkC+oIVJ@8y97=SQ7qru$4Wjb07uw#!|*biU1W&A<k&fT79TRKvHM`
zB~vI_0>v3rlQDvHV{&d+riIgLF=YEmz*QWI!}^liBMI$;Lx!BwI%(1KM7lQj9M;c4
z9?zBs5RQJUk)6?m-_x+WNeC!@3rbi!q1)C?70Quv3P1j1=zF&6lAzp6=^K+p$o+wi
z@Q^b<?Y1vh_6cw-oMks1Vvws#kTLr>u^wVkx|zB*BgI-Cm5KAfV!etb?RPC8Gf3oE
zcal*1i0y!km}my8ZD(EkfZIgqy{`T<xi&Wt5TtcsSEltnVEqZ~n-)wgOX=gS5$qlV
ziAIVr<6au&hnu<I3QAVXYvg#$f*!1|2S^O77RJSO56&!%e&{LvW;+uqLUKU(jn3)d
zFnWyIghpm}%fvxKN=s>%S<@^c?E|hW#X7s5VD)av?RD;AECmna+v}cSHcpA@IB{Yy
z`r{az>QIl;w#I<dKs&IWSmFzrB~*CT;$7BEe6K6)%-iSYKOGJ}QLDrh!&h4Mb>5dX
ziG+=byG&;8b_4z{qugdHhNs()C&U*US0l(`jd4t7oUN9|ljS0ldUi}Ef1Lyn&rcUF
zH$5Bwtj6+$rP+$wo_C8ZQKb@v0;)oK?*&7}v~l0;n!mbxKm{^Lpp+_(b=kNmfT?M~
z<hKG7rBl)mb8Q3rMA+)>jwsX%jP1H6Z|>{u8pK#~sSXu+1u!tTZtOMF$%*@rO^8+*
z!X29-UUmgAYtKOQUw4uEes4t8Af&m5`>v_lVZCm)i`Lv&<lHWZ)~zGHh$Q9e8)v(}
zOHadz;p3tHaJKx&b6p>WuV5C;5$QonThem70ygZjwYPp(TOplBqhkGa7m_e>9mr}8
zlA~Y*q1|IKHBNUWmiX=$6?3=pb)3Hde#CF1z!v&v^ybDk_0T6!`nx66V!r+x3iT$z
zN8x}0OTk)wc?PYy4@e?cfm45Ep8|Hx%Mu3`GJWI&SQ%zq6XMRuwrC{~4ht-*;KLSr
zSrU!ofYM$Q^9%r2ehVTer3BJuU6ag@0J5=ISp9uxg}#J>p9V!YX~EO8lJ=O%GW+Yq
zH8l_vTgCQ^B0^6?9YxWui03&~d2~qvdqHBn9u-ruJ%;9v9NpO$LlAq9-H4~li@_xZ
zpJKu8%MWmXmv$`-YB|@OAIXCLvQ#3_K--sRJr9!8{VaXZG_h4RiZ@JH@S?T=`C-9x
zM>i5!0T`7=nV$la^KYJ1xDW*t?e~vtBu+}Y-`xeBp*L)OO9q_sRmVW%tebt}QfpjY
zuZ1!PQlLe6YrqR+w%2`!#DFkBtML}AQ`hxIXIe#MAP`HS3%?@7By1@B=E37}P+-<{
z@{JoB$a)Vdb_2AJ-mv{#P6Y2u{^_{_EG`9hHy)>$%7+;tk4W)70_=&Eg?zV$Ko{RW
zoV__<U=_u$d2nBlT^@EIk6R-Mdj|MD0#VJP%{2)qhD`X|>IoxWjW~j=sGL$5Kjl&3
zZq&Z6proKVG%*W`cJvRJs1epFZy|Y4(|}Knr{Jak^c)^Qi63VGU+U9b%l(>rPxhuA
zE`0&}^ux{7G?*HxN?$R(cn2%G)1W1KKk(TC@65gUE)a~rRF%~7v=W{Qt11{C0J{qT
zd%#nc?u-X1am1(!Ms=jV{Dk8^gyo5eAA#L;6)M0lWs*1lUn^|?rAr{tfM5qG4G}55
z@x?e6RHZVduzobJJ#gjz+`AN|uLYCH{d@t!1ioB6`6{9DbKN^v0I-~%<4z##T$c_j
z$l(Wtos>Y2oDDGiSNIH|cSM6KRFr2)_>;i$W{QbPdQt%5(|&|q8z^vbB|OX2s~k9@
zgXOs<rNct$zA?YBTz`Lc9dli}>y*=KBpptQfkQ^fDyf=NU@k*SN}_4qgp_Yh5qT9f
zbpS?0!l0RYFn#kdHC^uLLbZCXlWS2wbNonth~(Tt{!S2XNF)z|)-x%B)w3(8GL95#
zL?ZI5PmGgTOcgm%gu^OYcq*d4_NSnalo`vDxz|ghDcG_o$C0KLXr<fojH}yqBIP$y
z0uLul-ygl3Uc6t_mZC2YbN@sKW8vAdr%MPTM#=qYgB$nknfiwx?^nxQ`+V8^eapCF
zKjPCT6jkITopn6@I98}&CDefw8bF$Gq>4D?53M>%-yt`S(|hk`KfJSK^~a7?n~r_t
zY(81k6|zGvu8K3ki+fgSvSBY5SRILW^sZL$z8>%+P9L(suenc>K6ups(cOIH&j2h!
zb?!iVX{Q1y#N$Z<`~lT2olgO`AY(9LrsCI4P4J=vSU~cqW1m&K1M!~p@6xAI1hA2p
zav0EE{Fxt)d1b=>H=7q+>rYYh-E^<l<vht)u{Fv9^l><LX~t9f+Isme@^)VYVJ=e1
z^a7n|^J1Nv+I!e8IM!^xbd6|OQJhsxXVrr0DEx~O6n24eCzb^@O%RtZg4Ai-30&Jx
z3EvWFCGVKgqRfG8dFw<CWVfm5_&;>i&)fk^v1(iQmB=!vU3?%GWdEf!2VLus9|X&Q
zcwjxBojwsnAp1;j$z!pE6GUC-L!+nytpdc`k096aegtZD<6BUw&Z}V?BxN~IzOBhx
zZkv-cj_4T&Xf93IDd;7Pkb~^J!#zZU>|26?k}4UJBz>45hFE(@KRrFn0c8mVgS_wc
zXYT;Kh*`i?P_)Nr?IB(+_<3IZ&5JGS6>K0G-79Ysi!QEF<g2pRV}3}rjD$b>Nc@jO
z*XaGvn(fm-VDjGXj_B=@y)E>x_0y?Js}#WI2xiJ0j~{IZ)bXsFd*Z_$Uq8NdxM$H9
zTYLHw(x(#B458Z+b3NL3L#Bhc>x_?Zx-e;%EMfs-*!$yAQNl}CZ(ps35$|4saBh7K
zj93eJLdFW7Nm?N?f2r%&twne-%z?l)`&_J*4pF+i8HCV=@BI$=T6+QJAfnY4o{C`=
z{`ksG^Z|29-*6D;d#8X4Xh4N6v`bWOzFyds7xb02zJ*&f47&UBVTPUR2C*~Vr+6d+
zTWYbY;1`AoqY0z6DpmCiq`LT$N|oOa)Q={yQI(u>0^FHaAm!p#pO~;SG12%bmJE^;
zN-DWGuUlkvtAKS{8$_0bo@2!I5-pQjXG7-^s1100iHrE9OMTWaOmKF!zvw#*YNrbk
z%lMyni(hnwWhn_`U4r(`5^7mW*!m%f4c>wNy0MlxWK;fnTb?*jUF@Yij$$1Z%Nc)S
z0baEMqcO9Y3DnDi=DwF5)eROEK1HU$EP*{HThR}(dKRGo-QfXtF)8Q!kXa(rr17;H
zt*|TYW4Z;N5#@T-YQ0LVR<13@s$<YW3&y<78j?3jh$^u4%u40#c;YN%`>m%rhRrO{
ziYN@<LFH==5*B!<PQ=HofuD>JZF}1HhG~q`!x9YXJ9-984|sTxpFZBeuZ70hwyA5l
zkzuw)x1TjWZV(-OIXT^&bU`3NF!_TR0fQ#ul{b@|bzcY@OCLxk2TVzzH3H%sn|i0E
z)%#*<7DGOGOtUFyZJq2bbqLlZp67z0)jPF!_yu`E;V<vWqFU4mmT-P8IHPHMsgl%Q
z61B<(aK`f@Vr{lh74ZEZ&VD0KZhNK;@WhfYD<ZcFe%M48^L1{>WjZLopZ!^eGxE|l
zp!d7jOT#2*l6mLQ!Z!aEcRH4M;|<pmtWe{tw+a4_FWIg)92+jQK7r1yCfc`BudH-X
zJPZ+gvNUZU_%LSIMmN{FA?ZBjPhG@QR04KtTep9-hyEh^a~E%HCC~$cNz7yKiX0N4
zxH5?Y-l^~0f8^z?R*D$p<}CWB6W|bkeKgi_rVH=!5sAM|w{n|~4eMBtdkkYqSAMw}
zRCEYIVQNjcIU#{Ni(Ovrblb}FKFct&NSs*@B#2GG72sm`K1;c!6M%Eo$$?u(aQ<jp
zNj>VTHR!2f5QzDL&fD`ifu0LNjaUE;VlRGOl*QK6%F=|U=y5wJgOCMPyT~r&I_z3O
z`WqojN=PLakSv4wN8R-~pFnn>!0pTA^irH-4ql+qpld>Gp4uB;bs$)NyNILQx>a@3
zgyz-6C#jN9sno6q0-xK&+3n-%4?gNH09Ei0T~v7TBLbVVPffRoA?y-+lSGkS@=PdJ
zq7N4NJNGO+y6Zcj8RD!8+M@5=YYV6RDF}RAbXZ;%PX3WzeuwVG9l9cHVn4PxaSm1W
zS!*DFzbI$#Bne7et7wgJZhpgf4Dh6#t!W-s1YJvySeG5|ZQi<=1wo2WAWE$qv3E~L
zqB)&CAu#~&Tg8EvjiPd)h0JFn08cv8U3jWv27-M1BBhy^GrcKp)$TwTaM0BH5#@em
z*=Uom#5~lO9IHrwic9BfU(@^a+p1s?K51MbaD8M~RoAQJ+G<K^^V<Pa9w;kNd1nl`
zYGQ9Q%^LAG6l)Nyq7%fu2O8N4_mIHxsnH_Yc_B|0$&`~(-@xd}xG`4nMn;PQ)!gc(
zy~uPw+_c0lFiT^SO6#s}(%0}wlac@v=@DI%aQ8Y$kI6?uhgmRntHBnicYJR@s%2Nj
zHaAWjK`y6pJMOlfvIXI;x*Jaa1f@$4ssv7X{UU~Zz19GDw0&=_T3Wa6g7OL0sWri|
z-!El%-{?hHL;L8wK@rBY;+6UR(S~a-bAZvdWFEJ%+bF#2Q-54zcpAt_oD})Ob{Au}
z0-4z4JJKX<(({LX`$3C>JnxB}R)5TMP8+Lb{N52zRL|b8(Qqt!Dz&|3AnWc1q`)np
z-LYoYizrJsJ<)mL-^FAIH1!usF<SN;mG{KxEs9p=KlGMn#a;glH#C3;PV4B8z94ZP
z{dO+h(?+#GQktG5Q>5)wSoCcuda!H1rZxz2S82KqZvj|+621Hcum0IDEU0!x;%(k1
z3`u9Fe>U*fb|BTx&Xc^ozXoLnlc3=js4CHuSa1&N3G+*!#lB3&OC;7}qqbaFq)&Mn
zRJA}Q3Xj<n97RZ>2T>E_cM}HIA@OMN7x>U<12EtZ)v#4>jNvt!pB{sDO3^n&(C=2W
z{nubj<UUJ+LOPDHIA$?JunBPCsSm@%oX7JE>kr6a2*8VZ%Aw4JSr7WBF}mv(iMb*_
z_QM*3mjOy6<Msrxwv!&ofNEKpof@MC+8@Nx?c;-UppJ2ygeIdM9zi`J*c#gjnth8t
z)Xy|HDM@l^UWogJO=?D-fA9(m+YETDIB)Q>x!FyE1;h&6Bw}w_7jRXn&3zaVM0{hT
z%$V7K1mJQ=GZ8y?`02+OjiD!G*Z;7%xB0Fd_Q>~E2=*QPMs?v9(adOvI}vb1FEsqr
zgt*H+%%+dwn>@xZ^l*eB1xFY+pg|Smhrft;&Lm|@2*|mfs7Af4L~rjdKgDQ8Vt$L$
zkbGH_GwF02gGK+#1@p)=edUz|&%9f1Fg}n$#n>cAS`l*ui@0NBj_tY!%!An@9;&AZ
zecyKh%Ek7Q4&(u<kTrM?-FyrY2f#M_CQ3SlWK=s;V(#PfBlc70f}j!9)m|&};eq0s
zot^^Mprv&Kk;axMw;H_qdJ)dQ`Bvr0KP#$$Tb9<x%aQ6min@HpbEGn}c5@x+CZlsY
z{rFOt>^65x!V;A}{|d%OUuJy`-?JC|_>y}y@=YVEW-nCWlaF`uN-ZC;e=Y<vlWjme
zc2w0djD9>Smg&ya1;w$6WqNRf>DiWB7t7CMSuvaUvd8~8iXS=%HG*d_^KrRaddwm+
z&v|d}Xk<sNJL)wZY!VTC&|tFj^qdd7f7(KFN$Kee@>)b8xK0`csHsur1mX-s<o_lD
zqV@{E9ii|9eNT+GA0=WH6ic~C-u2O~BFGo1Qo==r3hg{wO@J17nOflsPRbBmK0JWU
zZHl$jnm@%O-A6Y9sT;Lbyft=D<8C7d8yeYoGlqUEs~h>s3t{rZWNzgB;@4J0Dd9Ao
zDTGHMNHxqo1>hy*zX@D<6#c7$!j8wAtMG4U_;{IhGG<Ei=T;Ost*6H_vBdGO5OC-$
zXxa<@#PGSZrHe}tZMf}!F&MU0yTk^H9^gEdzc17$ihiFaMcW}bq-SS|R9`rS7ZEOi
z7FRi&G$DHGpV*kp)l4Sx*(|&9qk1xC#~-gr<w=j<;glq6X@qR$Wb$#_eyXQ^tRK3k
z^__d~2`0z=ZuAAbk!M#Tr|qZfkIX`aY9wHB6c}6ZTtlgu_wegEF~#qN{YtYxiM^>%
zEwCLhE(n!b^D8f_jq~yheUSlL3Qe)p8N!-Bcx3o+J;sPCd>&;IhR|+WP3);;Q<+~3
z?8MB|ml)9KlbIKhRLwtewxrH5ZipA+luW79w;JPwqj$vqQT|cX-Dp|{Xc$6dP0m!~
zH&d5CuCg%<7VsQR+T|y550oZs9$dfUx%|z8UZaFB(#3Zq&woj=bamxqZ?g8$?^0H0
z(?e93NTz@WR^=l8`=v+gRTWS$(KidpIcANN7XKD_0SZy1k1)e=0jphjLIXrV9YK7#
zk^hKD`nu8Zh48>|yWcQq>i%3Q$gzU{<~146@JPMNe*e$v=3jmC`8levfzW4xv@WO$
zDb3MF*sYaS`p^Hv6cb?DZm>Wv-R&6|M`_8DPi+Qa?p^VZ1;5%dw;<^2Exi~n(Z6ic
zgL)cNQsaAQ4ajWH3A0f-;YzJw7+RK;#}26qEby8^VT4#Tn6y9BeW?2%;w)V^8tdz`
zReuhHcBSedSol!C+>1To)4ecHa$qa%1<09CltRtKplx*yIqzN^Y?+$><JtVB_)$x6
z4yZ_Q=#!lEPl2X_=?e14arXq0t~>uBkzI~+_~$d&AfO(Y1-&}0=Xzjr4I2=VarsZ1
z@S34PM2M-^$*~O;G6X#7p<E|fFXZp|vCvok+>7-~?geUNxQwW}^{-yNzk>yYjh!l+
z|3p&>pm=*;ZR$WQi)dPi3fK&=&1=mKp!{2<P)GqiQ&PyCn;f$=kK@CCDW3fA{_~I*
z>MYkV4txFc*Zn_z@*666@EFZ3X2JiP2mjyy%~dEQ^01WTqW)j3pFda8zkc-ghM2$J
z>|JE3e_8MQpB@IN1A=$f@KFKh|DOm}ml}{It*L*Mv@`r2ZuQ@;$i5xuDA>tRy7j+N
z0h5xT0oYFBE@*c912^=u1{)i>=~rm&MBYGQ#B(`&bEOjpWP%;TtoF$~XiH`rUiapP
zA$~{z$NMvovU)}K?f$v99>Ah-{;v2&LU0NKM<U31WEYf9{6KAm<(H5ZTo8a($NtW9
zm*Rz!HOWc+jl(u5ePM#k#5>4<bQ#ZM_Z`HUJe)-_``r%>Xk_13zWWb>tlw&u0$Ku%
z^{}#vA2j#8F(g8w#d+sGcV14x3|>LqN`e_`N}c&Qq)uriko%Nx>f#CFEJ!W$OQfo5
zUfh;Y&*r)Sp(A&J!tG5CgvuxaLjE@%C!>iatErFNyCDU>HW;tV$=+x!_?=)mzaO-A
zdq*#UT2_OLzwRm(%oA*1=eYl0woq^^8YF`SdR%&6a3uKx%H9J=S@-Ku2<Qzs^;ruD
z+htlzY6Op(5+c9vf~q6>iV4s$V1r>T=|a1uf6=x|7nT)@PSQ7iqOxv8nFp9YObiOU
z*B3bJohWj1KzerviiG@r0X~~+AV#6WC=!DovWD`5(4Za@fbK?e4Gu$J5^bj@zr~P(
ze8=D{ePPm;9>`+L1yD7`8c%mTZ+P8+HTCfUa4@g82Ct5@HiMs>#g*@TU=*5Gng@S=
zY{J;`=TnlPPvmq1L;v%j^hKW8l|k`QK=>R?F)ANLI!-{QJZ=XcXY+<YA7SyQBNI^t
z7>gV9h0}`qO3=kW(8&MwJn_BnfHUC<dl~cJY>8<hF9uoQH>3tZL0*)M0Yg-G5?Dr5
zFqQkEyeg=f6WSMq(FHqKfwqL^7&JSKp0fK1G*asZK0@`{7r>%r9ug)*9}goh&>%C*
zB7niJIknoUJF#q_(Rx0IMeQQ^`V%(l4v-1aK!vsJocS1INUp00N1X;p$9Yf{NEBmC
z)c?Rprme9v2QoZjd}`XaFE_GMHxfsRDCeo*3X!Vr5E&CFhRj>c&Rd%PTw~xzAkqdL
z*5sJSTaw&+I>p**EN8diDI#HApaf2hcnO>bnt^U{ZKo7EyHRhPIs{2Sb%Nzi2r=+)
zzK^L&&7So*HUs@Ng%M!i+a>%852#U}gEM)>Yc2C{jx@Vl-jJ@E`g{`L7+OGg1rCMa
zfY3moRQ(=i0oG~W<%sMoJPu-XGUetCnEfIMPo)a*0ozFTr{dFJ!y7?}zbN!tp`5L(
zPw3a)oS+d9&daZu$T30Dy`}_3QL7eEyBvssS%VdgBaC<fyqj7;LG@#Uh++fWHH7^6
z<4?e{UM6u?M{dIo(7#6QcyL!#zyo1^^zPw$RN#4F;acv%)-(JF2)Mx}%?ALN&43qt
zSr>DXnO78+l?tMl<9SeMV?`011LxF^Cz4}WN%3fy-ChUUpldGRc{p4VX21LI_s75f
z$*qV+z~TIr`|m<2@8y~pKUn}~`$vwG3Rt*!xG+W}%(|8X%U%hMhprC?wGq%gn-v}I
zP9GZcU#Mvq{4uuKFqS|hQI76RuC3Jm?014&qjb=8PZ1!k_<`A>2h^YB{Ia~jaia#E
zv$+Z}1RVH#>~P#8u}6V1zh2&JX>cFuv*r9X@&Mo4Mi#1eVu3wgIqL|?oS_3iSRx+O
zr*h68gUH4SyUm5_l|iugp*h}*T>&RRM19F~NB<899IOXk^IjISGLDf4zQlV@3sCHs
zCA6P1*e)ME#w#SJZF^u*D%bD^^f$v1CD8kVi09d5qC?G~Tv5HxyA5gKj)&UUPlO3%
z2;TtE&du(}jj9`Ph_Cf$2E0Fogu_5+*F4zX4YrCN<h=y?Ffu%KMp9Hjz<(zBG9vF(
zj0PL$(+xO)oM!b(=6r6y>FM6VcMi0%y@t-e6);P+3ki53B=5}X3_i@r!aHo%`Xx~Q
z?`s{*LULnLaDB`dn?Ix)lVE)5ur;;|qB1{l1Pxg%05r0;s%dt54no?fklFSI9O#fb
zI5U$d9S<S|PcDLuPbsc=?s;?KsTUzaY~mz!p1ksls`LE8ANb_rTnQmjYD0aHiQG@1
zhu#JC{qAiL?bW}nL9|h4OJshutWN1-ve)dY?j#-{fw%_aISK9qq%%Nw>wP3qM*Gtb
z6IG9%TUM`|1<==(*YVBB&UMh;s{pfSnj4^;ZY4{#du90?U1#!NWtW8NC7jS6O>Ac;
ztx5L+&igXw2>)iRQfT(&LGsNZSEPp=5D@E~?#?SWhuY;)ct&ecmE9`6XZ(qS>t39}
zC252L;~1JrwQDY@(0n)_2Fnw?v~i1=!$k<>A^+4h&kRBlpM#!E)Y+Y`cP3)bMhgn(
zXHH|H6cYFmy6y|E&M|Yms*hfDfH-=-{^{(jOT*+nvHE-?3!lwSaxvcehDmc&a0*iY
z!CL!@2254S#`Um98k$C0FmpE@gr4Ms`Inqdde8`Hx+ZU>l<-XIgjyyLz;737;^cG^
z6Tb*k-Gt16Un=hc`zc~0P@rcAC6f0$w-ReNjh^pcI|0PmnJWjyVqN%&z5<C3DC>Bi
zbbyF;Y_`}*_W004;ohfTq4j4Ru&XOjWMcm=4za?Yd-cSG7+y^iy6}$PL<B<r<V?~y
z@PPb*skvrLu;2=;v(<%I(1)BxHx01LT3vjKnhhA%t?0ZWR0uGb-rb0yf+*hgj`DPy
zT3Tc}k)}$5%g~Jcf)N9X&O&n%5T8ER1Rm}lIPRhhCt!Q<?b`5Y;A?M#la5BjZGH-x
zq`L(0uDnm8>bw&R^^k6SE|tw)NVU}p4F|fqwK!<q1z{rdd{7SYOI-`;`wFTSuy1sO
zt+Djx!)ac9;`PaMJ6oi>A!P$(lU*QQd@dzyi^+&j45u&yx0XYb@;IYc;H6WXNc}vY
zn}E@sslqJw0#J}FT6>+2EBJtF{meQ<99B$${!zfU<HdeI#pFi5&*?5uChyy|AEMAO
zqr<5dl*SNVv=gfGf(-AqKy!wR4Y$p*$O71uLx6Q|lJ}5HMD*VBq@E0Q8hG0niL3^h
zjouEYPQ6EJ8b2zF!7b_!O|2+&PlbC#*iDCEHxGj2YTN?1|Dh0uuksYhg;9Z;C#Iz1
zm>vcRe8>!MwlRU`@q^KQya*Hv910#za-hbUr{_Nl48ouDbqcbHmfyVvv7QTUN17oO
zHE?7!C!xqV23LK55zsxH)y}zQBzw3ds|+L<OVwO6ok_Yi15D~xec)FqbU{q`=A1X)
zBHx&YRkZAtvV21|m{kOp3^lJYXu5y}h1yUz<z@nrYjimu&mMsxU(3F(I$R>%Z1RXp
zQ!Yn2CBK*tbn4|!)bgY*+3J5w_FucJ$JVQGDf}co(PnAVVuY@$AkU_3^8$hkU;Uu3
z5ApFdNIv0A`Vu0k#&U3igpdRxuwL6!kPUx*+D7F^Eh^IIFAf!CbEgnk9mUJ$^L0=s
zVAt_^tAjA^l2#<Q6O}-C1DF!^1+d+Y@h+x|Ut1)DA)E7u=B{iph^J34cz2jye01(z
zdj6g2C=Ue9(R|2X32i6THyaloPWqb|piIIG(-xdgQUaR(x<Pnp0h{<)hPP8P0Ry*@
zpT%es&PQFu1NaKJP*Wpl#j1-w;Dx*4KUC<1*bb;(;I_$NFK1uX0u4~g=fP|Gm{Ds4
z+jsp8!OYfCmIr1P1Zl5<tv~WS3kt%(7r6p>f19SMLXRV6614u_S{Oc@s{bqg;-2>*
z7~4EzcsGt|2(=(SYw;~S_COQ17R3LCVJ$2UxrW@{Ex9wm2Vc;3nn%Am3(~nnTml;t
zdp)8>ntk|H0#Mq&T=)TkqosvPXpTzjr$)PfBVI9iiUe9n(NfC+7@$oY1ingSkckG#
z+%Q3bGS|cs9-4muZ1B_F?}1?*B7R7oaV3{C+kn$n|F<hvkbc-kV?p9U)_u#WrZ70o
zHo&rM%L|8FHN?*b$mohJJwpo;ws|0lNA+X<RzFj`qOW7X!3S*kJNz7L3J3A~Ziiw4
zlD{{B^Ou9#T5#<>jExE>Fw17hmeSy%5WWyv&nqt&)mA8j3&iuC&I%XalZq%`qeuMK
z-pU2G48~6xVKKxtO@;7W(CJ+pfxE7VErZ}Z9JGitpV!zZn22X+xIg&&+2_4fqquN8
z09LuNjSv$>O-!W!*@4B=gZq=l6!1jCd1kZQN#g;Z0+>}BPEimxh`2^7!;VPv8gPfW
zJv6(_N)64Ci0;vBAxrjlr~OS7ZA>Fw(NlS-&=MbpFn<a3ubQV?55n0t9P10zidED$
zQwjLxSDvu+ej$obs;<z|X*m+kuNy>M!_{y+y$VGzgBt|We0y6ubZheDTEBNkIkZPu
zRjPfQtA5<xAGou_{c+AYK4jq2%;4^g#8~obqDO4g<MQa~3Vnjp--WY)t!F;}E2O`p
zbkf5YsQ|ll{-oCkyIBAc0f{cId9N_^m_5aPl9$MnSc9ohsP*Npt>2yCCf&NU5vaG@
zFZj3|u=dBvcVHgU;TLx~|J&8vOGn^cbC5NO2|Ixp&Q}L+Z=k}Nw>GWrqtU-_(rJk{
zwi@^I;R*-s4%_u^y|U)qQ{em}Q#Rz?KcMhN%dEK^*}s#51xRvKLC5kX8&22$?n6g>
zsz7mxn)1zx;8;oaNrXBlT!&LnFujMb;+%_4HfJclh&JIH+l1JrBn9s=%Cm$%#tE^B
z92bv1m5Iu4+*}>fmjp7wcV%8E+va68^WnP!g{EeixL3cgeY69sk1FRE1sr1nR*|5s
zyE4z<+lzL?q+k89uv0`{29=qhmLU7vKUiu>CW@>C5q~YtuP8zx*b@f*IGNUavmiW4
za46y+;1?yZ79nsLCV!L)Jn{k}`_UNs5i7%--+y6<6EHuokP<<9h~CO{m!mU-nG!!6
z2;SVt@pzW*(|5|_E1<9&dj1LIv;AnOTpjU>1iW8m24t@(xkxSxQ7h;O^Bt=)W)OGJ
z$Uy(wB1NPnkEcid@q8iG(|+~g;=>j+|9-Sj{b-R9<<YFf{mkl1KadM=tXf+6Ar95Y
zxdHbOEbbI64(!P`l{oPyka%UGCc(CISB0);&!!{;)XXtOgOpTvtQn6C&vS>qyxC^b
zimjk)ci^VGBdN@4Rm_$2itgO}Y{CwO?=65EOYyr-Ig|jR6krFo?B8NF;8QQhB2YZL
zjGtpu=bQ}i{6?H)ONSX8gLt7aZQT%TQ)Wt#7@Dd$JKN3Ozu{)rS@P|f`gRMLcDv77
zjZ9~}pbbHy);oa;V@;iU89#sfMRMSJ%2ku(^3+x+7J4eyb%}C)?**(^&@DDl(=2qA
zdBS;gd0gg#IloL{*mkP*Oxr4=jCn_c&Lub|Wqh{o;AF%K%2ya7VV=)3{$90ih&BO-
zS%Tk;?NSg^xllE#7%BfXCH=Pjo_5_qmK>D`Yqe$>gP;sn3VxiLuTTSbxdl8PLFER&
zYt+UNFtLGEKTvLd7S|=Fef?0L*ypaVb~j<qBJE=;BlrtW#nU<S2rYzrp$fhk5I)IX
z7guVad%00pMt$!`z}}1zg><BJe#JzCQ*tXf2;MP@AknG{L^3mz$($^t-hj|E-s6o0
zRDBc+s#j&$>PkZP(=5(4z0*SbnBQT^f*l@&%d)S0{Cb#gm1d<BQ!T=STTz{8nc7+i
zrr1jiM<AHW#l5&H#mqadSoU?kSjXH}eH4shBV>gI`x+*KW6ST;my)IGE3g_s`ZX(R
zzOY&ZRQPo=ubIk+bF|Vo2P%~1Wvkh0+B}MA(W<v&-n3~0^pETnn!#3xG3j;;2UY%r
z@LgB6vMC`r+f~WXvQ18QhJIzXLi;oI@<7>k$gtoSjhaVKtye&O`<ACybIot)p3gxv
zC=!%H!dk6YmQ{a^fknPphjL@UIC3^#v}<W1e{zX;2Y-qtbX9%Ez?Escc9RZ3`!3DD
zLQ?K}V)<tp6s(Dr4Y!n)H~gWSIZ9f9(X0C)_wQoZN;q}feMUuA9u26j;j%?(GF%U4
z{#>!S6g;_^(ynO@$xJLH#MBGJ--eWVrS)opQootP8L^h}EC;02r2@^vc>z-`Y!hSF
z^KczG^*(6abg(SqH){wfOq`gR!5E~plzl_#vPQ!n3skeaf3O{a;N*bwdz$NJcDQLY
zVE{B`_%5);g=adf5SZmO@twOSDF0L|41XVeSVIT9W0(Q!nF-^3JJtk(-H3ck(^<p@
zxF8p^^(1bC+fDxzRHY6NE7Cr*mg**_RkY1)88l>O(Gfk#hQ&wH#H`%otrWNLCLBEz
zI#(UgO9I6u57}#{<EI9VY~oGELMr^6Mn6cC>CGeTd_q289iTFnuqu4c+%~gg>yXwZ
z6OW!y#GV_%WpwTcrMOS&ycD6NT<V9i=@Zp9-t1EJEedKTzP$;588h9XE&fRQycf&?
z$Bw^B-;NA*1<HsMyS6mz2itl~Mf8R&^<GNBCX@qvC6`3msj=#&*$IS=>wM-C9Cx~A
zq){|88YJDVh6@JlP0Uha1k?|V6y9Z+?Q0CYh^+pDFeY!*L6Ui0+oU1#s6RV-?iw67
zNM;Sd=Y3<*EG(%9487*Tq|<GP5?R3;?XG}~tSAF2zkjpxI1;90bKTkZDkmT8sDh;M
z6aR%+2-$M4e9-=8?@!?4)tMXt65+Il+qn0+SsX|4Sr#%gzpgk-O6m}}o?qNrUvrk@
zq^ZsP@-)J%yL|kE!;C!u0NnG8olhJA^791RbZ)+2&ZF_DlNayid><+Q<63-O^7(c5
zUw6o%wJY#>7}HBwmZm!$D=nd#H`_u6b9h(tLUlbtdL#Wb&ygd$S6oDzkQh;d?;RHK
z$Q%mx8-fp$kN0RF_qfrZlJ%d9H-Ty`a^aO6h19wBKp-?0Ou63IYH%iX+tXHupqP*{
zS4Cn&;pd<t;4aF&SkS8!g*{L}d<rNTRHQ+rvQ>DbuOT;jV)qF%6i9|FALKKv-$2r5
zq<)_`**F-sz4+7sww_;iTtFKwEP?&W@nM+Va9ZsT(T@*&9vaYY{4O^dX%x39Bd$6Y
z>t>iM|24q6{oG6`pij-tbHM2_F2OtQ>M2DfTWu>*SlTUNlm`p+*=?nM;^xBBd9ewV
zES|$Sa9G!|N}xfmLv(@E<eSvD>6;~!fV|L>YC@c@-<vdwY#2sb{W^Sv$;2{yi-u$!
zG}_(3<-$0jA(@@9EjMAsCE{>z5jjl9b5BgIzCq&`iQ3n&B>u8un2#K!&)4;xH4jJc
zTAYD%b<e0+EXgzNK(@CA)N)xpyPQDTrX7SkfRNAaWM@KfBRT_z)FvltbWTUc58;Ks
z&Y=uUFCa`c<x8jJM$!GneoevjQWidZ1zJf@F&R)lU(JU7-Q0@ZeeXB4uzHZ)Q3eJD
z=s-zpSPKVDlo(1GVWoYqjJ*67twm8(mVIv*74<+;dixQBF<svtk|Ab-3;$kQt6|eX
znM>J=6x?zDX#wAQa~g#YSOGp*f~XOvP`O8!5&Pwbxb#q*;Aru;bcjQ7t8ufSW0P5h
z3>V?bltmGH;X4l&hz|O)`bZ{bK}b>k7!?vs69nWL_~hyW#Wc=CcCncEh3i?7IQPx0
z`v%U@31s|$pBKAo0^ck>SyJG)W_7dtWm_q<WOEwR2N6MJB6&ZOdg}x4Fh{IHB%iQ1
zV!~ZObxP-HI&XzAz5Bf+bHG%IE#;P4=dvy%>L$0=Bh)!i(IQ}pc3>BjV@ENZ-VAgR
znRR`FO8rz$9u`>fyS{c9s`#Nu%CmUPXK|z{>?sCn*E#nQu$b+_fwT%!mG$Z}3oa%V
za5X(F`I3Hfg#2Ut{LQgO$nnT`U~C|lW@CSTM{4_bfPEIr<&H?H<Ju?57yZU!Ko0IJ
z`bCDW6USfJ+r*+D6SfWw!+f7WRf}@F7wu-Aq~X-BAsL!s)`L2ckbQm8VWann++`Tn
z>SbcDuurtc-Lqa_k9jbk4)sFn#1fX>0ZTol>+Vt1nvZZDCyFkT8&#Ki2Vm}0<8h&`
zq#c3x%xj_?nZN6<dG}zJk)iAYY4x!<L(B)31o49TZXM7;K~hPkrn6W$C4rb2lPO;m
z#_64`<M@jU93?`4AtL3v0@r|OCdhTnm(@w>J|Ty^8UYjSXTn#dWtQWwlrQMa5574l
zN85Xp5uk)4`48WT8-`#T6z}jAYtX`=Cd~}LRQwc@ln*bssWrHG_{?%__XalslQv;4
ze^<$T4qzKX>Bx?<Y+__9R$d^ULEOz^ZQsU(+i!`+lGERYf_+i^bt;?n#vHsDdU%gC
z`#n(gbJ0<W`B_v{DD3+Cr!hk-F|hp^u|Ed}*6~Aih+bNGzRl>8OXUx+0QX_$7}W7p
znPJ`;33sd0+@a(@L(zQEWy#U#It#7|4f@SgMgEBPneR_n$|Gj~j6R+UyLQ>mIp%T$
zf>-DM)t{JadX}J-KdRqcK(NpAgTQaabW=ew8}$R6^3h<h+QCz-jBv#U0v{Qetb&x;
zK3TS`&ZS?%#NYTc0!C*#0lK(g25tUkvIfuTc}_{#qFfd#owt{l;5wR+5S|F8%$)&W
zUplRD7{UilxcWVU;Xwq$K#%|-{`%cJZfha}4-mpa{+Pl6l8Q=*9~YO&v8Bo5b)_di
zXQpK=&$?F$9335wf5h*6il08I^Gq7NkNby#Fbo9^U*O+;_*{cEbRRaf$GDXmJ_Ouz
zb0`Z?nW0ky4OTZGIlO%+YJAJ<yd_u2QQQ8JqOKkYMMAC^YSXfnxk3t2Vc6r!YzdE5
zX#{0%saW(MX6l?alc<xyh7R9i2Rqvr#&uR@J=)04D6p`H_2u4vuVScLk|h_tNthRT
z2FB<K202vfleNuq<PsyHnfZLvjns_g!Dq-FQcg?5^<nRctqZ_I2|dW?Sd6|kS(~i$
z#U>-%%%k7K-khgB#Uzut(jACSOHEN~E-PHLtr#*RydsY?cg<$&-3_X(3ez%`(?v#4
zfExD`<Rqu}oF3Jyu<lr0`m7G*JBHeDb`5}ec`VZ~7NPUxrwZuy{Z4MYOxTJgJg8;`
zgp#bcVHFD=(dXJbRg%e4VMp)Z4y^8nCB)cn3}qV*nyk(iE<4xh6^a&wiS7`rJCU8?
zN?9oJaWA?J@zPZbjpYsq7H(1ygpiAs{=Cw%mmcCRAS-6!iB(szqJkWes;qXvOwit1
zv_)6K$PW5d2d0L3EDbs<493eYAuOm=ERwMk-6?|NSct84#P6`z=@p;xg&wBL*kOi-
zzqXEw26H{Xw{rULUo3dp)gfL?Cuc)hdveh-haFPN{gRsli(BKRx5Fg9&?ob--Uq@A
z2ap?y4cC=@xyCHO=Ela$Ug!kWoA7nKiRd69UK9SDt#rOxV1Afxzg}mmtfJrwnZ{NC
zM0VS%Rn2xq8Beo6HV4Eci)ZpvX2WJ*$7{%eW%1w5O^7E>9${@3!4729TpDZ=dB}Mr
zmYl6gh$`hiytjeZzHaZ5cSmk{vd@)|$WRI85&&CBFJyPR6o$8OpD%@*Lvxeu15|m#
zmK#<R75Bm>8bVcQGNEQlo<#Br8x!+6r*nL>Y;@3+$K@gwU>6>#5Dk{GvF03SGI{gm
zo1;}8dN9X85c8hc9%n;kk&uVmDyy2&j6pux)!c5Tn?(>$8R8P;$;7?hhNQ>i5G4~Z
z`y0CTPRpvsfm|iV0!S+1kB3OVKbr9G*#tt4q%);fo9*S<QQNc3y+}%oJEZtT)PtLc
zs?H&6<vH?D8a57!z$uBBUteELh$bh4G06ooR<cQK!)zR7(SfRZ;96e#J(AuSG8k|e
zk{1)1bQy|^CqNN6;#EdbS{uzw==oWDAe(RLXwrVwnR!37INwaUw|wpV&?JKvb6YW>
zqfRubWYo1=Gr_*W^mom5C%f7Hz4YSD4%3sJ<Dah1_d}k11@gg)y}JH!SB-qVLL*J8
z?kc$l?|MA^xi;N6dFbqti1G#;PeHR+p7hfQcTRoA&Z2r@m`jWH_a`^1n{j={2{z>2
zRVTTRY+a=}T@nG$3O8D+&Y(H-ID{i}*60=n?vF#B22C(F!C`qcoiqC;x(n)X&RU%(
z<{3x^dGxOFUIGKWupO6>E6{W)wKjjcuc6YAc;qUzz3CYLlh;I|Vfx6G_vWf#7KzQe
z!DAo@)~ou%J~5n&?m+ba>97L-W~HPN3`rRQs*dMjGqXv&T%-|nO*-oq<QYS>ED8{c
z8Ey{TF0?cvpqDKJ5ogbfAw$Z$j|;Jpl!T?=UT6n0f2{jlrcusuoiW-j>xabyi@?wQ
zM{aW4=K5gfBp8D*1%3BQXRnO1kR%BB<*@d}KPJDc;Th;_nFFED%JAxs8+<bNz6))W
zo?6o+Y{EE|{ZfeBaZ~K%0ous%{v$tf8^(b5FzMB&=e9=V@BC30pX^6OtAqzuyeu=+
zCSv6)G&t$|<UB^N4ROaSfS)TgD%N?t{Ka2^s%z$S$<P0{AFnu9Xs{2MIQwHE0mb>3
z`mUjOQ7eP82`aZa3ptG_*`6dzR%U&bEj;+}az&6_1LU33s~&*&z<o4*OO6gBQ3da}
zIJqDaqiEj!&mY%oJAn%84JLmfzOnYhj#?KD^5lt;=Ld5iozdGn>Bv0ti%$WD`lHJG
z;428eMJILB@^yX6<>Avo8yCL@(vOq~vr;)c`?j&;jbM&~muqAOuda@EbuI@_WBl6l
z{AVZ1<_CuoRK7jID{ucgaC#W-T?hJTD_<B7UaKEMoy!W~I@2-&x)lodH*zsHf4tI3
zw|9YtWqflB0xuPuL!6wr3QUp<xR0N{UHbM+hw=@J`xeN<p5XmRz@YxYyxzw3iE7g1
z6-6cn&K77S%H*1ZS-uL;Po8~&Mn0P1PXXbiRut!ADoln6LSAVVQj$GMem~`QzEefV
zGoh1u!aOZjo09E}*-~4Qg+s>4(39F;ULYUxB(V~#qUl=H^zD)^HjZ}Q4V9NWPhA>J
z?KHCs$)1q$o;{py0F}u?Qj+zT)4X>1u9(dEOm0!xDvLoUcKJPV9I19;td|1O(besc
zDx}IG*ZgkoM|*e9MZPb%T>BIwYiCG1={1iPy_+gxyM`&yuj+2uSfP1vkiFe9*`lzb
zXL`^WwO#{tZF1MocRL$b6ieIPU!Q#HVJQy{G@=xvJ-N0EwzZYc!V3w`HBzSwJ5=3?
z>vNyHxfMa0e2%l)bTy>(V&gjB^{Yqf+S3>nUp7~rhFkVJgN}6`qw6P$H~t)ZuHv)X
z=b^_puUlVWBEj6EVDMF^t0!%PeMcCT6=*?wtxu3~=nHG=P^N|OZSdvInn6&R!=_hz
z#WT4Q9X3v%`w!)2BUv!1y#z|pF=5;qQOEz|?SsFT=5-zC1U99^sfk}>j^7TU%OKU^
zk-t8A&Y_sjRfHXAECv=u3Gwhz<go(Cp5H}3)m!(@j^>knW*{3{#57wTT~6i)5IYCR
zHyZSHDZgYNCJAAHGuxc8=ndhaLQzldK)ecnFh9QQ!%|EF=kz#C5yrHP@RU$_sRoa$
z1-#sBR=@DtjUeZv=(5RXF+Z{+4EHOV>Ip;(Xbm$;*S?kVT_k)jA1>?7U)V}aBXp!D
zCB`;<JUHWWReT8;d4ZnT7mT|Z2Fh!%=Vgx!ff6u~pJCI!KgD_y_;Tuj&Sxwmbt->e
zu2%V%&+#gfPYNvwo0?O4pH}x|9)%oy(G3#%Jkg*s7s$8vf+6aUb|ZOqu5Eq_|B2Vu
zeaw5UV<R>UnpwJmEQC&b8$#;_>0%*)pwTc@tDkfQ_;B2XC5GD0EX~o_;W<=frP%$9
zs^NIAm~=n0Cmw5brSut`FE&)Du<Le4F&a`0b1>ftC-(CoVvv?v9+TZXVz^Ba)Erjb
zQ};H5acq4Uy`?#qQMA!m77aE`<wt3UH>e7ENaQI~u!YPNmfb?1t}RJ(Uj9ZZhh1fI
zXZ^*7_Ll>D03sLDAw=E#;6QPln}kDFr^CflFr-o_YFOQFC*xSQ=Awv{AC+<qsuv?f
z4DK<GZ7Ca?VcWl{P%|6LjUN?DJ;9=QMLVk(k`yM|;j5xl3S>ZR&4uf33`_Y3Nk4<C
zF*YycX3b2pm(R1FZbeV@@Q?7&B{-9ZOkF6>uCBzrrXsti(p@G^xnBIlK<Ebk;U_CL
zhU?KRwANh(8wif-GIofrxItV&IcoLfig@Vj@Q&~m_n?)A-~J($1w&Et&%kJdnuytI
zZTqJ^A{&uDD;tHx?({l<=(wlZkV~jElro!(-EP+JSYLmk1{Slo(QSN~c?H8ZjTa0R
zEcAc#p-%$EFIDL*Emg#GN!qiuup<B3I!BVhs_rO;(JoUr&c`wb`;A)nquKX4ZR@2J
ztP>6A@)q({0)kJrD@%O7oTu~eGF82?=wDTS{#yAbA)(@Z>}RZCb`b8NNHwg&ip|aN
zI-K~5Lv=W(EbN#V-A{>zgKD2?;u=UCf;Q~2)w6Z55vHCSVLs73#ypPdN;1&=7_(`{
zynT)=J@1iNN8Z&MIYZf-F23bGAfafFd%gB*Q6O1SM36#?nIuGTTW$rLV`J)U$i0{?
zMCJ-!mumZkZ{wMg%xeQ)zsJi-?c%PBk10}*^EpyJhYNXaylXGJUcbyq7702Ar98DE
zs`8J-iBGp5H7Hr8&V`*v$8;7aj;QdN{#d<W(b3w;p^WpvicU6JF{Ah>6pJ0rh!)pe
zRsJ)z^dxtv-sb^hUv^&_s#mq>k>)D1-={L|rf)`%LZ%9bVny4d`3esTU0-3>AhpUx
z7KLfzLy6H1lu_Oom;fzd$WCY|pW>kh6w&BVh0M@{c-yQUI#-b|7%r2E-Bpx5RyQbf
zbXC<&zw+@A97}FzcBpswR#+X|*d5Ol*8kY%=wGWctw@(Hvon%To0U0F+RwJRY4|dA
zlarI@ak0)Ta<Z2qa#QT@aeIxj5cYXufON#MDh8r#+sWmTPgWP&A6FKD=;lOM{g4VT
zZT9uq$^Pul*$&N7;mpprvPYC~uz-%yFkuy$Ew2B5@%(c^**)<F1_VQ*VU!d^e<-Dw
z3<qUc4J*~=l&Pq7XAfUwTotf>E{e@~J6blSPPp)fNIm8!?dQ&7>@c&hqnQ+=QTYi*
z<q?ct_Xws}A*%DZK?;&8?-i8eWyyKLbltVhL;lU${pYs_M|dp3>!YX2ID?1ZaxODJ
zh$kp$Mu(|?&GH!o&_}}@pR_A`Zo;8XzJcQ@0#ZIVC#85Pq7qK7gVsQTRc-$<+iBFr
zj<BaPUvosF$&}+&h+KFz+JZUcPZ-zUQX!Ay*4|=>`$HSbl0g7gbU%r+I`}Vtp1vA%
z&bDJ;ECT&=6$u7d79&okY$BN=9>i>eOsbz!tOKD&kcr2jT`AD7c!LhIA%wx31fntv
zz<n#vubin~zesQT$;j@!zZA#Ab)-G$91481>XwS{W3DBotScRK0pkQ9dF*W%?YC_g
z78BCCxiSh1EbweS#0L*<?HuG(HP_Em{Qu_NMPdC&;Oc4*zDGdxb>RNh#Yq_B8=|5K
zyFnRViav>LqL+}p4Y%b~mBQ}pNjEU-Tf7d2!3`k2#WrY8iBK6*{Oj3LCIDX8?bjxE
z|03{(p5z<sEY#UDlfllMToW>FnQ|W8Mz5Id^Vyqg$E~Nzu)L{$@GMB&vow1_g82W^
zePjp>rE#Qcfazbq=f6GK{$+K*P+5tnVPgu~WMa^o`hQ>TZ$A6X1)ibCqGayhJi`C-
z{mYcVBam`7A^*R8FPQfTCOE7^nDY_;=_miYZ{4SWg6hS(oNn`f_`RU~gcdAhu7a`;
zf4z>|&@*H{8mYyM_kDR)<0Q9E`JW#9YZmZD#{A0v;%Dr8g>KZFWC0SybkLl;`9P})
z9M#RcSSz=<#V&wH^Z?)vH$<O<8>4FitY@i0*ZoLH{s6I;N?vJ#d&nF(Ju_;kYzE1v
zijIST!W_7n+U8T4bUH-R%T)j|@t3@Os5QI-lzt`l<@K+__ds2r$$}3k2_`~$e<C5O
zn$)I~Pva;2zPfdwz6so9+AfSR<1BBY{pYVEU=JSSu!R%GU%WGy&y7GIiv(s(aouqo
zEdaUse9|E7<<^fUz2LXWDVhXvw-&ZzKJZ{W2AZj=$-{<-P)J2^4ET0Y32}Vx)f7O!
z<N)k5V@8G#An&d+P&WMrxS}`12gUCz6WR5AWa1w`oHzp%F1;tlJ!L?eI76(C{32Tp
za(b0XFhOqshAuwf&qG@fG-Pt}f!@<_`wGZ>ih+(R-LBUC{ID1_B4XPWCRE)ofHEve
zd!o_%mIVMLGt{jFjsr!ZDIRde4b%rss7jzytU6Sevn!0Kq&(;XS=)l<^5{|2YIPx{
zO)LBul4Ij$_nVbvR#}t($6FR)2H|prLTm-NApWPP^|=cxYhkhlkDr{T`W-IMHzo9+
z5Wwf{Ro)1Vvc2N4kaF<gK!c|zfOK}}Sj?o3@Pmd6d`cRl97q7hfW}Q9GSpF59!+K8
z2NmX$A`LAZySIy=PC5ZJ%*&^TU;^F<yn8*RG)=(B<Q-lB3L#S<%&pM51hB<Qt6Ai$
z`36Wnn)lN0oG=cuu91mmk5$#j;srtY;{Vs)cZW53wte5As7-6pM~0%JwHAbkY*a=-
z6letm3}Np*RRobOL!`>ksvtWckU+weFl7}L6bLKKAWO?0f&r1CyyxBHQJ>dpzo&nG
z-|_jE2!?xH*L9BHIQgEQF9lp7VI_0hJtltKW9{wyvZl{FcMct`<0E&)d_cnQ0<2YZ
z3`|1|;BmL<*@lH0F&8_k!Hl5=$BJv|ftt;=d}-maN;CnbAf6pBfafI0=ce)TIsKo@
z3Z<03E!lu=z5!31vOW$3HIz*_=4Msmc8;LNO_Ci7+#tIb>F{U(C=~n!^%e`#Jg>Zg
z9%)I{Vh(8|LeRVb=@PYEJ+u|7EzgHw2JYIP{!1~)i~E3>$`#bU7Pe5;g-y>q9Sry%
z37t3h=Y49b_(S)z7ZoLB)?`t_L0MP?6u~kZ7I+pjP^oXGgf}37J!pdh6-wk(zF`&*
zU)vp1JoQYAuTs_b!Pfd~AI=UbbvW<}4?6zc<|$8tgZtd&h6dT^J8D(jC6L%T^k{*i
zj4Dz-e*X>C{2hhPRYN&f0(}nVy{d(u<UqhR1B1CeRu0|dX2-mp0p9)gabHN|X8>A#
z01AmbD{LW<fR9l`omS43EyZLAnxs{Zn7AeW`xUhM!%FoWbXRNn>QeQI4J1AH0|$lE
zYSLbF77RrZ%Hs^}=B7BwA0s6F-d;ePs;B!#iJ1-EaLCx8HglkNUp3Yx8o`=!h5kfT
zmX-52#0v1S_5^)<ANUnmfdC-af7YEC-h%D9uT&3aIaa_yUAHO$%~~#69xfkwe3ydR
z1$LC`B<&^$rRRw&9b_vX4NXCckjv`A1pZCjf--k6Nh9v8lX-LzSAJXC0Au0ygLe0s
z%g_#^@@e~w#@So>midDOy%@wRJwTYZn#wLbqv2|#j%Y_$0g2cQwYAjX?&2}D*=`Iz
zFRme5WyCalDMF&QVq_joV`FEI8aPlZq-<y5fjY71N6q(yH&dGn>O$pp`*YGJ?ByhL
zQ6q_+C!eX}+fnC)3EK&s)I)@y)f*x7JEMC|ozY8LjD~)4xD-b&rFevQW(sS*d^#d;
zZKYw1PHQSrILaQ#T!&@lE(Z#O>D_~GRdu!RhVVqr!;-=__bqmtp9eA7^+lESJGO=<
z;Wy=9{~RG`dPi2U+<crrV=CRD40h_vf%GL>Wu0BXpKR2nDB2l+18CVX8r!W4XX7$9
z$;}_+K!7HIS)ILH?Tq>vn}Q2)G4Q-&in??VZR^4TU%LQ}24j3e<EijjxALP-H=GxB
zmTJ+4OVwPc=;I>7%=^B$sfuYOtXYR|q@G{my$?GXVV8#3oXvJ!^Xs`o><;!F0vf+s
zB)wULlc5)^^v}bz3I~nuWe9t+?BeLM8ibU_9<BH{uDtd~qu;+6I%wzi4NUe@Gthj6
z?MOuum4ATId-+=dU|%nIFa`q*??&!E+Kkc@w#*2Xl81i9<?U|<g7Qv$n6Lqyq3c_|
z&7#xbb;$kMfxS|s35bE_@Bt1JE&&s;ewubRUyX;Egy$xHw7rR8`a>8lK@~aMU*j`>
z{+);88|hFx%mv`Js%@|81NY$>;Ib!+fV{$X1~r=TMB@%@vYr0g`_<Tvl{u`?frcp{
ze5vRtN4~NnX7<pnbmT=K+XkzBvZ|qp9VBH}2wDNXRmRKrlaZmD0SiB_($5E2$KO~0
zx0qtolvms~E9&I;e7qR@b70UQK0<A_Z)@*$%Mj1OxXXq)8M8G3$>3n!$pA$RHENfn
zKwV|?!{W}JF6R>`S?D?|9!V+4R(aR%9+}rrqvvvKm1t);8XrM+$0C{HsaavtI1f`H
z8>&jZHvlSq;1^}RncvGI8@od6Ewt|o<u%)BX&T9CI`-}1{6NRbx@hGp=#gHw?9mug
z!jB2dQ+?Z^MX6gHjyg;OOq**!_PrK-1A@zuK2_0pw-y(MLlX8ZqCKq8?_LMpCbXvV
z2Mbma=d4P^!v0m<vUrv+6EXvy)s9j<ZNs2bk=}%#I{Zxr_Y&Vt+iO;#jBCU7h5Low
z-U-((VkawA-S>?Meab~5A?<X&Y9;ygl+H*b1#<;exlWG)%OyhS7$J2Q09h4XC&t|*
z5dLIW0?r}h*nY4NV7vo0cUGjGyRx^8X(r3}sz&nJQfMVG9`DozP42D$IGIpE4sY~Y
zGipJE38qM8UWgJd5lSZuhrC-^3{Qy?{@D~*+1ne=hGGd0eK14r$JTp!lO&7ZYE3W!
zndOP>7b!=`ND8|EW<+RzdcYCDr?OTv)HniA@q8wdDSTD-Jx~;W4EB<}2DDyP?&BP<
z13=}asyz>`?h&m9PN$~0=NJLp%4g>^*miQblAv0eIS({SPe@YKw8@8R$T+72xD}b;
zSd`ADE2dkj+R4&g)c9Q}w1cZO5wEVmrbJNJU93f!@he$X)bfO8D!Vc)vdxYrhcwbt
z2qW!JOV18z<DX6@uKyvZcwIczEs;Xc@P}Auy!{{=I@1&}M4?+G%*`1QSY{pruRDKY
zfTK%J8s=Ae{$bgeGHyiIoXZHmO~1P|I36=7TuO4fSS~=Ta?lx3<X|=*d->gNDF<!c
zblv@W-LbjmMO5F=U=p<H?Gi9oWH)0LglkFwZHlf9a5TrPPvFQZ!!JM_^3$yA#Jh!_
zaRtdD##3tyGWk9;uy|JMb=ykNhRreY=R}d7Ui9n|uAw+PgPP!m_Eb-SwYA0+lL-*g
z%S2)1ZPF`0LE(VOwLk8u=Cb_cqFbgaRk4cyQ8@+iZCu-Y=-ZD=>_g=KRJnMFvvF$L
zm<frpv^7Vg5lNziB<r+on|j`@81Eqw7RKMU@;xM~MDHO{IWTXR1W>J#%*#Lb)Q{QH
zwBcwk=(&E&N@^H^sR5qg2pVq7@6zZe-1?xT@~p||52;Nje~nxJ<^W6Euy=KtSF~bI
zc{Hecjp8-S*rc|5rDAgnv%<Wr_K{A3DDi-l`~vgj=?r>Bx%p^nK~sop#wA?j#nH}u
z3C>Yleh>p{?SA$<)+010Y>On3y=b9#ZAZ}z!^Vz3wz1-X_4^#J+$GRfylcX4DfZ5b
z#!3mG9_aCxhb^<k7KSWFxf>*s>emj4a&EN4MY;uCn($70*$|RkaD_8EhS|~cdYkj;
ziTrk1g?Zug;y49A!{RFDw~?1Qee`@EK3e2I6H(=j<UevPI*}HmB9~Rdsp9HIcycVN
z!MWI>ui;X;px0Y*mQz}sKBGx~Jl^|BV*|ax%=dx>OX!It@rEpGqE6DZd-+L7ah$#g
zz2dCnC_~+zU%s*0)F;=uKICymw~AK9io+=PH{k=q#@%tgc7_|0{Dv8+VU?SaG>-ls
zEBPEpqt8m_Jur0Yr0%Kn<S!S=X;hq;R$x%=RUewKwBM6Z&<fvNW?iX|r@x1}0d-D!
z@$t(Z-EL;>o)ZOaQ}RXl<(_eAqh9{JPgxNYGhVG^i2_$&iEu$j1X9Gpm5*cG4yAOq
z-4`S3Ga8|3aWC7b=9H_090#dl#dx%{w;_H}Sa<vWg@!AJMMz9AZpBZHt;X*WM&Fy8
zvZYNZ@cymaxI8g)HFXS<3&R}Z^s}(viRcv%bJt7cU5iE)j?w3%&9Gu`A!JCd=!hVh
zDX3&J8T`NOETcJhk*@izbWDT%T}yQAsWrg)f{7QDM;V1YZ|1=-yrsg&FkvZPJeLdi
zI?J-Lxuls66i`VHIo7ZoQKmRo$u@2HzR#n}LP*c9MwwD#b|$>O8L59Ey89Pk9L!Tc
z%wVag*GTn!)4MrxU~qq6%QWCYIW5)NZ|PK=9cE4_-IyjNhX6OOFjFzfP=6QF;jhc)
zOmsTR5>Cw!ki#G`G7Ex>&n5AQcIwR9Md<WBc}J&`M)mDc&i(W5*VqKNq_Mt6#q$e%
z^pYp;A0Ky+6&ne8v(CBUZ}zCErUifp5kkfaJ$RW&o1y@pCr4Pf620dhgMv!Owb0;I
zP;G1s1Itn8am(=~<QYvB7nA!Tq>C@Q8ksMTM{jlo;3~?|7LtmNhLV?y3lj|M$zBLw
zLZpq0f!Dpa942usT4-prP=zcNy!75K(_TYf5$@=yU}KF=BS^@yW_T{EX=+ZV7RJ^Y
z;BV_0OVJCO<dL0>B)xNEmMXiCo=enBJl2M4`C=zKHUx?1WRc!1OxhKLPMQfo5pTWT
z^0HoKu06(L@%IhN(DS+#d)osl>O<3Apu^t4xSPtq?XYk4u9pq#r)*)iSrN)KBjt8|
zFLkDp&oWU<bFc#Jo$I*TWQ!rdc2)2h8Y)sM9Xxz=IAiOinEa)a7vQ5)w>LV~YoFhD
zgE@aAX0&q${ZY{<w|uya<K?6%b;f(AMBmI#9#fAtO5apDnnqglozvtP7`QJ)lHOWM
z8t%HJTjp@XHS)OTvaxHM|H+rXV<!oH2fMm`TU<&CWCxnG*ro}_FYm<LJ;%<(>2HY?
zYLc$>@WmrjCzC4eohB`lj+&~OapZTnYfj||G3}cZcQNwGM{Tmi$rFK$_6XR6gfHzu
z=>Is^@_k$rf7y(q<$<+_&91j_wNq{9CuE=o_LR&^vBQ|qN)FYeSqwnke3pkYpM{-4
zt0{2IxpKpkQj|LM>ziRbVhmZZr2?QG7}8shM+L5cDK$|yGa=cuz(u=VjhOrf@gaz?
zD50I@y2S%zfynp1odpjTqeTH~-I(Neqp222Db`OpU7r@%1oH&!FcOnHEBI2DrHI?F
z>IQ1BLJWrei1A$J$_wq6hrt^MYzVs$m%7sS?@>i4%P||B+!UEYWWSEtD12yy@s!EI
zYhnRK-c6y3We6Kx6%r<vidcOaz9W>Qe)re|lM1q@eSJ*N??ag?k(-dtE(|t%8=C}2
zC)2nhp6p$uBL@|a3&5l^u6(bAHj=gPYt{GL_1SGkrpoM#Co}~0xi}DRpHAtFD|~&Z
z<<>*C8r9i2I-q@1vZ4oE7WhP*?p*qm<&Qw#)u`YV`q$|A?>MP+Z(HV31l0`o@))~p
z?|iNMj2fdtsNkwS-5A1&bHM%Ne3eDvctXpmy|7{e51fz<=2QFdrL#jJoKX$uDe>gF
z<lrz0jWp>;In_-;hM?u5v6?7hds>!OTMy-b{U6yE;<bE@N;Qop7;Jseb(lt>677^^
zRK!jiiV*hCDvwi+>g3@F_B-9t-UqWMZY~1Wzfhhm2vv1Ygj<F+{wH<f#pA(b+@goB
z04wW`CZPc(TKmGBjTd_|F~bqNBwti?ShYKpW)94mYJHCgKc-+VVzP@$#>;i`^}E+m
zQR@I{K3Vx0g?*fXE9osJZ$d7mP%6k5sO*n05yCwb-}{sUqu9eTJ}oY)NWLLtvOA`+
z&=?43+I;E<D}^e}H`QdT{7qab6B<grJ}o!6e1>j1NAZi(x|{&wcgwiVbf@<w0Vnf3
z?|EnBx_}J}y`tUpeNMZ3-V5H<ImM$k{D^RlmKJIU+eCg$uZXu8jeg%g`9O{Hu?^%C
zNMjdo<y@ei%M>>fK5)cjjQgDg(e9Wp6uI9HYGErc>eFVp*V-xY^G|G0rg>vm2CoFP
zmhJ)KZ;Pw6AMK>K8;9QH=lbqw(!O^OVfB1lLbV+k!!4CfgZMeqmeFI!q&d1J@+(tT
zb`jZ+H5-p?qVLpqVr&R#+nm?1^QHQjnky>ijNAN!WKuZfy>_^0N7I*=!NBin6b$=m
zIM=<l0)D)a;DJ{}YLT)sbltr8CN*;}r$L?DIN$8_17K1Bzk>GYrt>r>0Q{_cxJR+R
z{`XMAJe6Nd@Gh#UovCQI9%U%<R*|Fm`J&v=ibua+pi<|Rf~0?Ln_Aug?;HNd&WyWt
z52Mx}4!%3*2_yqZ8j!<%xoy6ru(N5VME25b)Bcc4#Krwqy(_Q$U-`yuPH*gzh8^O(
zaECY2EKSAr!S^4}q7;*w2J_dLk6-`w)s_b<Ra)|dUPQfyHHO9Vq&jDRaOi{9UAfpF
z<doA;ofdwxL%_ytf||WIG~LgqwTZkQVet#>mf>QpD=Q2_$O*GuVymcID)Br40;xpk
z59k|G3(TC}kMka8#M&9V77-TGjF49BE<4;M`-efaU+xH~>{f(9WZ|B8BXY`}+GCLV
zTv?w69}ba9{a1I6!%@^#dUhOd%_KX}xwhP<@^Exj+_xOXYt{D<yb6;u+Aio+4$H9z
zu3z!=m~r0YyrysLQH>6VJAf>ODXyhnD1-(y8j?gR>O9ul{GU>7?fRvWOt-v(S?(Q$
zfa-|mt~VeSX-iFZ8<B)dNxVsasOGiZmoW_cc%En<frk0>5dBX_&gXw?l4c7{^>gZd
zOU7touBCTR6iD2?_bOY9o&X*PPQ9wTUGECr(lgwu0jG4^8B0E5adAGTADT|L|D<c7
zn8_+MnEp2{%NFgg)Wd0;pU*x7hj7w_0)g1DMBrYdy@L5|xRuQP&%s-87h#AC_e_+6
zi;Y@RQ(V!xYKJn(+RGQriZCI}SD>=zCm=AIXrCSUxA<gv&#1rC&M1T#>uNnZClrga
zCa*1{O3cU`zw(-T=p;A=GfT8=6VNC%vnjD-0HB>E2SE-*?IwQ`Jy}7nnp&RRy#u)$
zsx`Cc8q9f68O84`>AhEc9NZoX!8fL4&QHqP#F31$^fspcaRD-r=#|u4vC7KK>TIeA
zG}+E2h>HPiq`}rKCD3l&*3-I7$S9;m@>3)8)_OGGIbO4M!mb;f-y*)=yHFv?jSx&l
zUlqW^u}ACuXJbH*)1sGn>)KXB4@$At7w7LuyCc&sO}cRtvttV93AAP#Of9#_pB<ZP
zsMB<NAJY@~lD#k*B`yc#ui~Dasyfkis;}ZQl008upLG`6xCXuctERSu$AUyZ31di+
zC5h)M#E_n+3V1zSs3(mooVM&w#stdp%62>Ia43b+F3qyk3!!XF@HvCzpS{}sVM!>l
zI94)$Sy!YPd?hDG_n^fQj68(UQA=sooW{Znp!G--Z!@lK-L2P!<@}`=3PngiqjR)G
z&y@~uVhQ_?1|4VZw{RKzL5M}cj|K^{hN8T=C5@rcGXrK#`O@7clK!-Xw_s%x8>WWx
zl0<S?j^|u~S$UN6T@Q@Ga^jbwft^eb*diM-RkeWoY1jwpK+_|=w#$1%!up#|dbJ61
z&i8yTH*LYck%kI;<*(WUz7y`fzo4TTWv5}e>QHuCf_|eUr$C&RhzWd-%<~J6#?^Zp
zw9LF>pU6MA<qv@Bzpn{3doYIKmNR)*-ckQdXw*2l<FpE0%tTsYKI5Ia*J6fQWnk9R
zfm)le8ZH&r@j0Wt(0{plJ+1RI(qvg^8TfR{%7KZ8OKK>Ou)oauW2^)d+F`q?ev`%t
z5P50$21&_dvP(exEQI8>p#SDXB5<ObAL^^nzyL?SLIJUv4CCPx>xIL_mH&oRoOkpC
znX!|7($*$PFuW_HU@G+>3v<zli5l^;oFascwvVZHMTa*j1Um64E{_1@B+hyLiNVu6
zQNG>$vNY7WtU4}ZK0};34=#`PlX+nx)&xlr2Ch9#%iVQwrK{vck4MrayE5nx(?c>w
zgO0HrE1aEPKvS^<ZSK%=nhU}@T)>>6;EaD;rRvlTo=E4&6a5c8a=aXlHD~3D6Tk2F
z*f)>4;^ul7n#?Mik4srKZhl>lXl{{H1Jc%Y5vAr{1jcyCES?z_c8`v|t7=Ryx99Q9
zu8Zckph2zj-0$?O(y$4$sS^$3?oP@tjh3WbsbhTV<-8dgzxj^7oqtIW2_sGUf~lul
zq;Kre_%4H`{j}w~vu-VmlL-X_*tB@9WmkT`s4KW=_sP@vT-I4@qI}ys7Svf{#7*U|
zMa+Kuh;vH)0{Fq5{f?Wy8dZW+H&(n(_EwQ+oz|R@N!ykf{0lC_X4rG3f?de0sKw9e
zXBNeAw6oPEh=b?Xp@?8}fVq4&E^{53W{|Zk<uKn;k^nZ8WXEFImOH9Pxse~Ym4Hp=
zsSDW>NM;?^cQB<a;lN|^doPumEft4$rro;1?u~KJ;N!p`q`*X6p3+@?gu(E<2Y6Wz
zR~5DR(0FD*rbwPAcRgh&lHD4!Cgdg6wc!lruw#}>3-wUlvi4hzAEb#@Jr%N+3&vw7
zVvJo1LS11=>xm4ddvf2E2Q!ETso<NKL0$Kok+N@zMjHaney0Omc?`eKxbJoWWnSvr
zg%e~2A_bSzGr+^zU$U<A$D8t2W?4t-h|O1wxRiB|oley`&dnjjeW23-KldO;#j=3Q
z>vW%Qh7{wt=F1~FqCOqV(RXUqjp7{*%c{}#Rp25NbX1=f9WlM2hkgbkgO@|2x%Uvj
ztSLU&RW=M~d)|a|g4ue>+_kN=e3C+)hFPW&EmO}A4UxqxS1RiAo3B9KaG*4%4tU^k
zohy2cKA}vmQ%$0>W*ot+@Y%HT*-9(oa>tn7u7zpnNnmO5b(f5|&UO{0t1P_&$@u={
zEA!#z5f;T}<9^SK!<7Rso>7;~Wp(KJ97JLddVM7K8wo<Byi`p4I?qCh&W<t5oFq^X
zgP9PD2I+!^8k+<$SBi<hrS*dn!5BZTzP{=EJq}LfE*1aA^LfJ7nEk?g?$!!Ky9y<y
zO?op+0GBZP;rNVkBXC!AF;<S1?4gc2tC$^^XyXQy&^zt+{tbQOVeH2+4;|MrtwODt
zDeZ95>0zE``RG^q#0NWKf+u<iZ5r=xQ@ftw>(gsAeaDo?@STlZw~PRMn1Q$XD`GWi
zZqM($6`{Bi@c2Vm5m`Hq{j5faY20ijwA|Nz>5D!tHYU&hbc~TB@<<K3Ea3H`;eyj|
z-LJ!ogfq3cH!<Gu)xXaC3CVOf+hyWhY&+Tg9D_8R=yojfJ@ea%N%!|U-4qw8C=YtN
z^GSm8r5q2MgP(9j=KkDfopSxRv@$o>3EX|0a0R7+7-KgmiGR2#Kj||kRMm)k!I-KO
zzninbJ7Jb4o;#I)aiU29KTuwhp)&kTw?KT|?7dH`^j-Gq@_i&FTc?vZBTdGZJr@!Y
zi;OKFyGKEzD9fx$NaX9ph1N8Vc8@elE?L;;a3=@Yl6j+ieO<8x0KE5tbh7hk?)=`+
z6se2?l1kV8l)IE|4?Ok-0XEo(3U0>pNld`hsR<Gs`E{|iG|Vf<)BRGN?n}R=lKV}S
zf?7caRK<0;lj~)?OtMhm!_~Nq^?wJ^)@T#~94S0G`1Ke%et@0AL!#)!WAvlfbvs{y
ziHz&z$-X1)i)GBMp$9<$X&ZlT`DpdBrloDZkK8f8w{w|2_k{f&Pmf*SA=uUTU0;hz
z(EQ8#dfnd1V2Uo~#&fXRA{a#G#GD&|d`vv2xHgqiPR6*$*Lnd0mpSs%B}9p_QFkjk
z;_l`Rs*`RW6v#D{POF4sYd+-<3AIlF0H$VsuaA3ty&*vP#(|<o*ujuWFycaVZOK`9
zavc{DZ2uF{X^5S;<`u1tKtOQ(Gnv*y=QHC?1hkS(Fx{EI2fSP9g?aJu?J|)~yVy+C
z;y1}vN9;NNz=bli=DSw`46wUrAkPu%a{*(D2cS|Tx$G6)=1^3F!j%-KWe~H(cuG68
z?l%TT$LT!p;E!2us|j!^=!@ynU%zmlt2I#CqNTe|(&4?@>+j!Lmm7xCW%@lBXTqY-
z`c@9>ur=(Gv^%xU8BX-^tzk{#pRKN@6Yxviw<VbV4Y$#Bqwf59DAH`BM_1m=ILCW>
z)&#QkkBE&6e%=UIFoFnl6IZ!s&*$Iykd1x3yN^)|SZ7)&tq9+8xc=)cgD?C)*bQ5d
zH{*=P-@ktC|M6Qse8Pe*+_A!9(?yfNyQU96{jszE|8MQHNAiEkTTA(gtp?R_fXK;L
zYR~3RMdTNj#>bcaVPz;~)IjIK9=u<3P5N(li+qF3mJ9~G0VmFq^hZ*EPx|`&^Of|u
z#|NvcHRGimqdfB9ssNT^0zqQzDO>)IuW#`$9k}H;YV2{WUFcY!(Vs6H{R`OxL$@n|
zI^q5Y((}zf*?I`Osfq=3Qu0xFEy{xomj!8cGuWJk9KVQ6Z~ST|`K50J<sQgyRsOaY
zDqNgeIpa1mv}fYD+~1%fU-*y|UWkIzPXvi4{}IP<yb7ghzA8%_`|#>MTdiOIveE#~
zD^c9?9@&5L(3F<Y^UA+vufjjO#wT!%zVpw<T0a3sUwXTr{abk+1h84(+41B5=o&?N
z;2OiVaA8qjzV$CZr~iWL^V#M7`Q?8TWdHjAf@ZOhovtR!`TVnM3`fuX7k64xKb1az
z*|~qcX%BNa_xqQgw|zp3zOep2qzxagHpmPmXYvyLzTf*_{Pf2kNm<3fGOiknO8!UJ
zxL^g>So!u3_wg^kp3l#ef8S-lJ}l!Uh06>7h${R4_EuZpd~r228>6$@hrpj-RL-2r
IRkjHJ9|;1VB>(^b

literal 0
HcmV?d00001

diff --git a/_images/IT_sample5.png b/_images/IT_sample5.png
new file mode 100644
index 0000000000000000000000000000000000000000..b27c3eb11eb560f49d90f17baaeae9baf320949e
GIT binary patch
literal 632293
zcmeFYWmJ^W+Aut{N-8yUsUSHp<dBM#lz?=%ICKsT(hVZrqM*{<-7s`_58d7Gc+Ppw
zbJq9$c^<!C-&*gz?tAS$d+%%K)z_Yo_X<+DPspDD003MW>9@)Nz@IE8(IZS`LSLpt
zhb+jm5Ep+hBQ8$+-rmN{!qOA~kPeAUz<RHuN$PXp&XMLHPZ0WbD1|l+p#0`zmMj(p
z-D84x!M>SBQY7lOY{9`k^qaqfKSNPW3WALcm5L{EOM{tqj9LPFAAIPlZw4P8PBN0f
z8}1uv?)x<^`<@Qe03y>ujZ_pCwCGp8F>f#3!7D3-z)KV#08b{WG@xtUG)!}OxdjmS
z>c+*@8%u2UJZBtM=<WTWFA>1ejX(heWT&<dB22$|P>z!dl|07+nDA}$>7+7HZ#aLo
zcvdd<^&rBb_c?xq!w9-@^IN$O=oF^p0CcVd78w*DxY%8|j^2~xpxuDdh+qz)ELO>B
z1~51*>iFr`{vqI)6!qYs+%>ym@1p(79tZoHSN7>Tef|U<{?jWs`y&qyg7;M>X41I`
z(CfHV=EuQe1<bYP7b9h#*I3^@!A&uu^QQOutnve8SOwl^#wf>`u|xQrCgmx)P)E-j
z(MTahJjPF<7}Je0Dy%)B(SvcUhGjX4EJ^9=$<%BYxlj0yZ(V;Wm8hh$gg;JO**)#f
z+yYk4al}Y=?~n%HHso%c#TE&JZ|csT2ppFP{8B<Wu5idjR%wh1l&!8+9AStwj&cqk
z!|WofJlO3dtA+HUmpLxXHza$z5Sdu^ZkE*^_@mXn%PKfiwAdz@%qTbn;UZUnXa<ch
z;6?)Hc-XJEiC>c246^v-nwNrM6JRoDUgvW__~K;G>S4Ty9X{RlxW%H@Gr<10l9CdY
z(Z<X{0Ocf>&-0Y?q;o)KaQaQ3HcVM&|GUi!yWy~n&$CPv`AlpaU$SCU=jO-NC>%@J
zsIP&+s7D4UOTHhXL)039tXNPnIws7N7T)(Lzgwtn0hE|hufhBn1Aekg<kcwFenDcG
z*DbOjoCQCnN1qP~FuuJ>j;1!j(R|}fKzQuqk!8Sy8XEL0i%1$>CTo32>nD)empPOC
zC*Cy%<Z)V1q1gP8=TA~{{{^uOL-dzz+!|QCzUD1*i5OTdN{1f&7<%X%Ef|NA)s%$+
z_$~5>u9xH<6!9$_%TLj1pFE-(mdYo1*(%u@jsBc0hL$CU{yWYu3G!&Q?`Sn4-O+Sw
zgqT5DMnE7-IK@i<VZ^c_iJ_ViuK_a%3nZWD440SVFaw_8%C>wmW-BSwz~jYB`K%kx
z)*f$oVW?=pVPs=`Q*vO(Pf2AS>d=X?v|?vbZCZ_6omMS5C!oVuh|l-AYnA6H9_Hc!
zOb;7uAzg7<!e1U-aX-Afj6%owOx+T>BqdE7{OHP`tMzp&*&BRAAJDrGFUbjYLk!=F
z*uAnNwBsm=gQZO=Qoi^VEr&c8CS`I9++V3<Vztt4Z{6}TGYvAAzJa7(|18$Z(pe5}
zH4V@UP&833#YmD*mx}+w_vIpXFy<h3QJIN|FqHI#?-%iD>0jA8nL=3}AHFkdvDgz`
zh2{~l5T_8i6Y0Nt8W$d?@ecdz@z>t3k*{>s$(3I+r!bYp)%3E&vBW7dFRD^2W&NJ|
z?f=W<x8-|*pY{^#HOzYBZ^zZ+inKCorOrNoq9M*ASDyYr?ZoE9aYkQ}5tyAc!e7p<
z#kmJ62;<Gq`HjdV$*z1`puAU}93D9&ze(wa=}CCUbcc6WRvQVC)y-_wYk1utb?4WK
zgUf;ID;Xo1-Dwpun2aN@-|yJZmfW8lRRPi>(Q+s!f$oje{^a5I<88CTw2>SR&jwR}
zY?~pOc{<Zqe(}-eQ`F|>_T^5nlk=v|5ZTC&9k;EVt@!QqVXCq0ZP(4mXYbAiwi^t7
za%zO*ND5@9apo5l&}4Fq7Y!eb(C~nGf_c07K61k>rMXPVUsA-ASCZjTY>@>3+4z3*
zxf#fDn~~>|+wkp!?hTASJ%g4Ge4k#c`4#x8i+iwq;WUP<13qdQqggP`=8!5dDm3qr
zdxeWV)jCW+{%vHX_J!q^$*b}?i)mp(!BNwkk7j%DIyDoPc-2DH2Gt^-dB_xGv3B};
zoHbkXaakpNJMpZ)bNFSOSy8{Zw(6|dtjDa{ET1!pvryf;I_+~JchMWwbGGv$Dpo3V
zygJH#0c9$2%1MDr0WT+e=P&#`{J0Pb0Y-?rz@q6$;xZf;-p8&%YLlSKo~t8L1F1n&
zQ#;-D_4l&P?`6)OOy`tglwAz$JMh_V)J|6S2?41?sOyAH(y!BX2DJur^#b)giahnu
zSRK^JcL;Xksv@e|Z~QN?Dp}kyZc_1ARu5FxM%VMUDfYCNZ`uc&IXa3KY^D~6y^|hl
z0G49!o8x`zD5>zGRJzFU9PxCXPEo(b3&r{U$c^0cDeGfvTwifi@ybA<Cp>t>0lY!J
zAcd)k;;!YK<*-)oG`|!=kzD(MvPZMQ>rbg-#<~MyuoxDlT&1i+-GEjkY!TLBN})-l
zDx@cL6O5J5)kyeQtR?0uU<EKk4_Q<Zx|fFft=&3{HM5BGL5Eaz{(}LJ(JNVmzGh|x
zt#Qj%z&q?3?5Gz;gvNyUFTIun?VvRbGO6Exe>bE~sY<)}a?ruCj{js1<HcuRlZ&-4
zYa6fjUummgstBv}DF^vm8U`FXX^Ck$%><nmpWba5la-J~B>BVJ;j#3pO0_?QW%-~o
zYE@IpI!y6w_PXU(!^O{~OF1Zd(=+D3h>KHXEBs12#j>y|l(IJ6@9N+Ti(*sKq}Lj$
z88h0t{(<J|<tob2#=6P|)h=xE=s#W{&L`$!$%>O<nN=ezWY!pOn(;pK<@wID1@-Ey
z>fz||uHtpJg8C}7Y=3T~x%K|Z@qzU#(Sq(qW#P3HJgSwZZeq4*vE*L+Y18JX5_vn#
zqXgV~Ty^i#yI;4tB@sOob-V&rGZkc2n&npmB!g^N?>@Xces^6NXlcc4`jsTEakkQS
zHjXQ(bbIgn-pF3*@zQZTm82l=!KHGHmR;RQU1`a?x=)_tL20*X!gck|JRf#Gkg9tx
zu+4cEbvEVvn)qI6iEpjApKk7MeC~V;zpb&(m}p$_)V~Nj@56gb`LW2MFi-HOr>%;x
z4o3lv5ck1QM4L#EHhDhzC-SX`4yop+TQQ_f&+qLwJ`N368pDloyCS00(=yYUK7CG(
z=f&mA<?Yj2(XLg*Ot*wOL4}X<&Wdgnw=#YU6*#+%l8q#N1{<573`C|9W>7#f>>G=$
z3}@T0?MSUF1}zS1fsRMVFPdaYAsSTX9Ci94P90~xJ$dlF{>jur9w&%%`epx76``(z
zuKZ+LQPrGn&G3wH-Hlw7`QTNG2p`A%%3Q6p;HmRP+NR*e{y-f|U02h^tykyWB)&aO
zxQNc(qnn?bCC96!j5CU#y!jp)ln~Da=XXYDuegu5+C=mg`$S70E=Tka!;kYV#cVS7
zYvxZ>{Ji}5^x&Su56ZXpjz13-<3#iHk{TfQ36uJ1ZFX&UMK6lF7A#?TSI!TDK`vGd
z%M6MO=8ZIW?@yIZR(%!Wic}fKqWIzL_n#k>@5gVUJ5`xBQJq{}o$Wf0UA_^Ey$^_u
zMS+#k0Nx1!#wb2MfLuy5(g{8u(-?ei_Fm~pcYl)#_$by7({2~RYJU1bUM;&uiOMGA
zTRp(ihGX*Ag3C8W8dA#ZuF_fa)P0)>24C-P6Z{9=`G;}?p**z9+&j`cbxfby2~>5Y
z_8MxcA!8;l4`4>-9|6!&$N?D0918LiL816}UJ``~fcB4aQ~)5z0)YOnHVVk}&l8RO
z{=ob*Mf(y6z(W3dhWucesQ=aa2_O^gzw($w$U4Al6>%9EWU69pZ)$4o@X^L`PreL>
zEO=}yt?2*&ki7Wwp~xsdKScIFW${77QA1u1Vr*l@ZfIg-WXcY+vi;K!KnMmwX01#e
z4QXLkmevjsm@xf6T0oHbKgAsMwEt-02o<K+kbh4rZewps%gYX82hoc>p{1o2vNth<
zD8H5b7dY~lF#ShIM_UL7hl`5~y9=1z#@?KRQ$Rp~1H{F_#l?nf!RFv<?Pv&Nvvy$k
zXD9#F&s$RmV|xo*M++Nk+CTjo8re8G3e(g78R*~7Kj&!*v-t0ktR4QfEaU<?{?u@A
zvV%DO-8T|c=uau+y#>tFQuD2a6>`szV~B9_@bC)#1K|Hv_1{DO2dKt>gL3ikg8mcq
zKdSzppsEh0_Tn~H$Uz-N{`<iG1^l0t{{j@^__Ox^*o%K|^gl|GJ1z1=h~wX<Ch{aw
z@bv=nASo^0Dt<twNHqKNp#8D4nEsg}^XNBqcOQw6Xe9=adHebU3}rVBtDdrAyvnp0
z2<kLwK3(ASU!at;$r9ge@c$wPdzBTKP5_}3dqkkcMCVWK_Ay7@7&MhL>%%$k2+D!y
zN_otD+!)xOPa590^1gCuC@TY#Zz%~vAdpd(k`|_5pGSZ9I=(_}Fab>c=#>p2MtMr`
zH?Nc!rqAz3^t7md^A2G=?gkWNlO%s({VRA{R4%{=s=u1|QvzRhY2@5>i-*3N{N2)@
zMx$hI|Mk|tM$Ucrb3o4Xu)pjA2cU}eRt)`b-c0Og(I^I_tlTd&{|a6V-30Ix@^@!W
zi>e##g9E5D$a>ZDH#qZWR5Y>izroc%$GrI&Irk|HY<U@yzwCkx@Eh;%$Sqim^f@hB
zGrPveccp(xvE6|8PJegiV(7KBC}aTVmNzP2B>u7sB&nnQ<(d8eoYX)5Mp=qg6tQ?I
zpZvV{Z?_=((dqvDY<~T5uYA9Q3ND~(<`WCpu+=}uOf2Rvxqz5C!%&mAx5KDB9}kbF
zJP#n%^L)C$?x1mY42Q}7FZtT0CAwL<tJz5l77h15I@NLL=SfNT%^6Wz`^&Yz_i6S)
zL4&v9i|Q;5blO-Tx3iyb(X#Ohk<zJ>&GMHBnMpygIhq?aZQsY1k{p{u4&dck57qfg
zmgWXRlE?W{tNO&>&8C+94|&a(s`UOH{Qn*Q|L*_)H~jOBCkH4bbN3u3Fv<z;mAkS}
zR-}bDr`R-FT*`iFXRSYMHLIL9+c9~yOWmV1yG^}5wX5?pojav?h4g<V3w<Y^ef)*8
zB#tj^5{O5Y%n~$fSr(z`I4t2XpB*NM<-6P#loG~X+Z|3VJVORxuiGs}_@Lo|PWt)u
zH>3^t-UZC4@iD_+nn|2}53wb#_}_^_N5G$D>UrOtAxit$!K<ASU<Rp3ZqeIAakKc`
z<WJ0HW6H8gnr2BIl=j_pW)<UVef5RlRFCyIbserl>q$cKhTDVkQ3b?yUQwaJ)_7im
z!2RWFe@Pd$wm_~QKl)hZitX-xiDQ@BNGMm{<Y2|hm0RHdZi-{{L&|8)ilTR8;Z%YX
z_2)BCME%JC7+>Vlz^t}%-dVeBr?4)$W+Q#}?rdk0<D@oHB!q62vy0Lgk?Npu5+r(?
zl^r{JyfNAL;IimxZkBBE^5kkGL;VL7xd`4U(R<jqfafbkF{?bGwsZn<mpsM-rZTrX
z9hUSAFG(XW$A^)PGhOtznVzVzNj>r$H;S+Sr!29S@XzK3E)~{pGnbtV2nOXccMS@i
zglk&YtTQhPuC}y?5cWT2Of^stKKoEMp#@EHKWL}|zbvz`_{8xL`%>PXnv#?MNgHG*
z-*_Lnz{3w!qMf5JFk?hcw??~`a7#2_Tjgn*)T!6&Hl&w-<DS{s(25B>1vi2T`@gT{
zomzMzG4Z31SccvmoXD&fvvBOi|5V@%IX!YrL|G~}Y=a;7ab(DB_i;3Fd)}P)-xh#V
z<-WB4kzr;{cjsVWbwR`VjBqcYc=+}JDJc=i=6s7@Hw7$}WM+visUF8YW*MI6TR0t4
zrQAXna}J0B!GnAun}dc6Uhn|F@w)piT1COs+d6X(-5TT~-jeSj7ZHC*w)i|FsR-_W
z_%#wK0LY&Y6-|nbb~)-@m9|)Ja{W(5@lX08Lrfoz#>;P-_B}81o~<@5u~~(_$~n{*
zoZt9EaUg37VYhNwZACD^lKY{k@%neo(r4^BXvRNq|8gZP<?do}@$_J^X*A9POn7?H
z^e}I5x`B3PHY`DUrl0No+e$)ehK*sj_v||{#>hv}$<T8DL8Lh#E#5^R9fefQPmV^$
z|A)J->xXRb5I;(&C}bE3*=9OvM#KNGzbk|3<9ag4!j0TjD{w#eXyUYaM#Hb7#sVT~
zMnU9I#u#6!ta(WC3V%x=O|Xh?;Aqm$Hyi*oA;rrgquD&e%UUGu@K=&o^J&h*>1c(w
zIA7g2Mj#-P4BQzv<8jhI+5b*)dP^Ghw>p)K=zZCdfv90CI@q8Y0KPjN?eg|9BKi-0
z)-54+tjVrZtAV75Fv94_tT0l8v^DR;7OX(N+kT!&og9~g1*gI>MNRbQ_$Erj0#7eP
zd}e|^eSXBPkU2^Lpa&s|S1?H+*?GpI{|z>qN#p%B6l~P~$R*9&^AHXmw2ap@BLmMn
zR*+U}Uc#0GNZ~;=ULPB;w^>GW6=^O$P5hyBf(wQyAtN<SQ{uLZgQkatIV5Ex2EFgg
zYHs%H@<z&5LRlb+F86Dhul?rEIGS!tMODuP*1xKd7cLXW!u;H~ek8z2^*s!)&vwk?
zwLZNnYevOR<XiL@WrKg1zkFyq6ui1dq}a?gXinN%mvoSTVGDW0)ifNQz|A8My>VG`
zjYssJWpFOp^$DS?Agc9fV%Qm*5Hvth<f{Eg5ZCfS7mYVJ8Wx_T#d0I-fp334L&Zqa
zhKD&Xz>=iStdea;5!j7pwO2MM;&2w}7T)seEc*KMMo)f6zuCI5oOU`d$-+idR|JK*
zWF)19$o-}9WNtM641s5Je%Zh_2g~X7PZR@71tblnyI#z@fJf!-L0WW+r~Ri`NY3ro
z2ll*f#pamH!4IBRM{I3uGK@NLf>mM^djTM>A!?HO1{XXYmpNF~w84o}zZ*rtc$>YA
zoFWKtdsgJVXOfG}QD58@SHv|Zgi57Ul!zsCd)Q{(NQAIN${aqV-tG$fEky%PFs9yZ
z>!h&RMU>yWPi&;QELd3AA4#J1Mexor?L12(^p!*1<QNLb6r_U-Y^0j~h*RsWS7X!`
z2T&2evqe4mF|MXvz=OY!P;8+PT73^p44ZHZlYvonr8(i;+@1{Lx&cmk{xHE+42bw?
z7`)9q*&@eL?NpfFw@|BS>_RFZ-RavG4>hbvuV=p9=%Mf|dd93SsmZ61Kx+I334ns&
z9(NVuwoCNB<dbKOMAX4?RE*Hjdf0oikMt-NQ7>Cn$L1r&a`^!DtC5264pRNp$LW(l
z?-6`WoDbKNO{sPZ(4q!JxX*jNexxuln|GSznYC$Z0LRNn`DA)uttSXFh~nTh98qA}
zhOw3ggMrsb1T;o+H=(vSPePN$@dVn1PT=6JpQ+!R##Pk|Qa389m^$zUcR}X$%2&$|
zG?@l<NIb=HW!P&A#B~T{2R3CE3-4F2!mX<2&HnKF(}I1$qS$!@Kit$9_<`4jU~Y9^
zErDVHk{5eSC|YMdTE81r6x|EY1(l1V1RG&>9QCpkSx#u0Q<dFc@1*p+Okq{OcQqqn
zd{&A`h9)oGpI2Fpw3i*#o-IDyPa^TOB{+KWbsNU2C{7LUtR=YYbVOEtelZ`Sd=~vo
zA5w;50G{Hhvoa)j*IXUxd_1P2Si|ejm4dMxC34+8ZoG~A#FFF(Xyiw~e_dp9%tdi>
zF)$@JBH*Gl*kPxjs<wWZ9puljSKcvgmeOD0%T_fvQlT<H{5XOf(1~zF;!@o`2TQ-`
z&Gg{h8}7T+>?n~>Adz;aD525ZQ4XuRy~=^Z*~N#draLw*{iLdl3#Z}SAH*4FH~`V>
zC1bFCyDTr6%>4J$ec=F2yTfC%CpXiYZ4Qc}HY7TZufbMLjcGg~Y_jTdB<2~nd=4WA
zihNS3uwsF=%{%S*pn>DeQcRz2=~H<aA0C>qJW~p7U2zCH8UjSCUy4?l<lE0Gl+ai<
z+*~c-bqf=x$)`otL=PRy%ZDAO8pnR%AACK~W`J__Y5Q010HJDCBgr}`#n=>FLfh;1
zsB4yIiT`RXHcjeA^mZ8^v6&sUKj8s^B5FT1g$)4i`gV$%_B(M;*~~LMoy|MuY-@)?
zGPU23jwxk*OyfTt67P2y6gFS*e(+=r14St_DFV{wjH86rLN*Jk7P#DZOM(QD1j(|o
zP373h86MG;pzE9zt3;-FbH2C3TYk1Z-baRNT{Z7))<xx#pTKsf_C-v+Kz+XAfI~BN
z|2%OuXsrzR;v4_lOiok@#QTVj{5<j(LrAKT9L?G3ecJpA|Adgzx!5<SEa=-wvDv!p
zu~=C*^TNKE*j~2a*Kfx|%&2J#0D>R0FA{W&3hVYERmD#~x1bSXeu4=c)eW^R9u4@H
z%#oR=Dy=_Wj^JAWF{U|w0R!H~lw5D;3AQ~$SM(=zR=T+?Y&h2{>*t*XhRaD~?f+7W
z4=_YZX4~`|{w~fyLDu@xSzlwKHN$4<$3@sp4|fOi`?H69k%OWS9_Ic*zAJlGi|O~%
zmmTEc#sfkpa#g_o&yJAWRhkD7QbU%#E)R>X=Co};#LKb_W2>~Ce9DJCi9V3wG@)r0
z!86ev_kh$|`pEbe97pA?>YbyhM~E@gU65p2a{{~0lmpGon8eCcO7#LTm@WC{9aRz-
z!?_wRqRhr(K%S;if=JS^bT0U)4g9HW!_jc6JlXj%7S<t77nKFXZZD@GvO_1DIVCRV
zn<?c>3`@HxPCvY@w}LN59npA_ht;sIE%is&8v=q}FArsj>r%tc_)!OuR`-TY|Jgq|
zJ5ssL)P_|fshH(1Uy)xA3v}8+0K^Z1iL|%n;R6-RK(95~tCv3Sb00NBB>zxHE_)?p
z3=E(=Z?|<oWZY|+I-b^euCsWWE;dIOE|z=tPm^XCMeO8QIUQ5DEs$n-8=g-e(K<}o
zVI}@ZIGDziDbc*WOZ83^y}#!^sWZj{H3?g01ombN>VB925V)Embv*~_cgtLDIADGu
zU=oe@?a_YX!=ZGawb?;TkYadsOcYY0=k_}F^JgBki=}gOlW&S=cpS?x?(n!p74H0l
zdP@DLZf*P6sy^^S=pO#4kP6OXyvmm$%~K`D^OBF&cxJfL<Hb4lHB$!!HYm*ju-%fb
zT3o2+<fe|pW5;nXQgIkBR*5FQ=Mx)Q194+@JUErhlpz*h<m8Odbm%Q{0pVuh5Z`d2
z#f^tLlcG<AVg0W*EM7E**{oIQbH_@~@X9cyyEt{dty6b+-Edm5g+H%Pes2x}NmsQo
zyIaYh(Q7|7Y)5)}x6tR;vYlVjUFAP1Y77=rc}$cf&cpLs7<3#t#Ca+Pr~0zq;)hii
zk&<6_bC!Z9BkhOJ2Em7duICbDsJkQN_)ck2KhX&PERGu=L5DP;&<ebD4O}K&_|Uqf
ze-KtOh17wdf%r5FohOt<C=C}2?s<ChZAMs6pg8@9v@L}hbmD>h*~v=Tk^)8EKr6%c
z4O{@HSX&vHqyVc}fZ_eshRcRP7wUoAPQeBay9SfUvOnQmS?dm`*otaXdk}pG1;sVg
zdC}9we9k`zn<VZ?0f=b>=oZ*#ywfbQ>E!~nv!K_VKW%=rYrg@(0rYT4N&-o)#p+?2
zeJS2whk$COkB9?V(V8i&t6L~xOTGiybd$<2fgaaF%P3zY9B4oUt(Pnw^Y3>dD;dPA
zWnc=xm!`yBt?^2jFzX(ClB1s4a40Z_$?7vC>=<GxxXa&KY?<zg_(cST*WXBoRdRZZ
zN>(wZ!;Cs$XhlgtoYi510M5V?Ci2I?Hcpuqn1Id4F+eoT*kL73`cyzOun!Fr$JR86
zky-B`Oo>|0s<jyvzLlG?A1)|BlUD?v9du<_16lQp6q*D0&(HY!y$hR0Z!g=T;!ILa
zaZ$0@Dt>BuXIHRnT+0c^O?xg<!9Sdrfdly#VSTJkOXP^sok2bdo0`|=az=e4L@6DI
zx#@f^i_T4&R(maGhugvjah{wwfTR32ROYapg+%-S5=?{$?IfMHE1H;|yRVSTT8va}
z1j~J`nMb<Y_OAf|h+Eq7K%m>KJK*%k6Z&Y0OzDQ+&|G*;)BRbI*~8sM0*V1{@Tq*=
zel1iS(+4&vsb&U#e8LkjLAZ%`$DR^RqV+M09eH4wl^7%%&6%;yZEjKlrB0|e7n>PK
ze|zdPlT}i;T1i~M=&#L9PMn6vXOuSAU6c;tqZy@2BUp~v9{7V-i>K2<sJbP3P6>^o
zlD|5sMW1JMFcEjqp<X>&OA=oX(^RjL$^>r$XJoUzI3#^4Qjt3J<n7|aU53Mq5^Ma#
zys9%biQtbfSVn}Z9`i^gUdn}8e*MQ5dVYyq5IVHEP4Ef$A%CS*KKXp!Ws!GMYc;H$
zdHLB*D7>A#{rN<;t`V-!jE*^}9?OyDpkava&jKCeh)0CbaOJBsw7P8JtE)lp{lXGV
zAJwMmj)mR^kvr$3@7u|TVP{k02)F&!n)5Td5MuvlJ00J)$#1q7-Vv$MeK@1I&_ZkO
zJ)qQ3<ghAShSXi^I8j}dORA_&esQkbsj^Sw`JHz?Pfqn>0#tnLo_?zx1yHq%dCBr#
zb!DZw@!bneqmUQB3(GLQR0qq8P!;kM(tbnpkD^&lK&2W^&%?o;M-rsEAL5*$CPWM~
z$#!d2+HZ;cbXjKOdA{TseG&$|bPn<gc<eOM$XmNjzP&iEuCH6uNz>k47nNs8QC7sM
zn7<MD+i`X>`eu_bFHv!q{de!dDV1|fE^;z{^=HptGLFbBN+tPwpKeJ}+PyW>Evm~O
zs<J{lR#Yc#g~Lbon9TH(ax?ncggPtqA1*D9m!oXBZ}|*z$iCcpRjId4ZiLinPKw-Z
zqZPsnPz(~@xqxB8601%PPwxq1IgGj6+O1clc^5rSepdrs@=F|zBJO_=Xp1{CZfLHX
zjb9C%28h`7yhzxUlTp@sn~96EMrQ(m8q6a2(xm6=@kJ)O#rbOizWYN5e;iFAJTJFV
z%Xb)Ku1KxA$N>{g#XU{&1%&7Qz#c7V!T8#T^SRAGT6ue;bq&ZTlDcg-g;z(I<F_DF
zs@=Rsf|z=~`uvZBB5e@v{gh^B6lQE*A2Ptk^{Ci6N8ke{FzuNy_1nx{p?s_?Wwn30
zd25k@#ATGBCo9%nfPLXMp9^AEuMMOd$0b=eufDgHJT4$Xz~&LGhkj{=LXoY0`?C)g
zV}1Ny^=hQGRKB3aL)!^W9T9TsTqj}rx2+kxf(P|w?z&d%{=&`fsCt-Xs6};a#I;++
z1CX8SNv)$~UOn?Gq?HYXZX%qASEoM(5Ubf%!I5Sx@gPhjatMuf!_qUaeqv0&jd8(z
zs%6QK$LlvN?`Q)SfV6mNcAF}f!>hRS#%$BatC<w<3s7g)^PbfRH-iH>um?^&gtq*B
z)eoa5KuZwv!A~NGZsvMuB>N+|^+sb)AHp><(&o`<ufkgj<E;9~2P>yNqjI~2FXqaW
zEm?}{OWl7MBBx#4X}Vugvs?poy*P@qGDfO&sVDavtE+?@u4CRb-3-(q&ncw6pd-58
z#CM(4g+^=Fx%SrMD@NU2s{!#6Vm(2h7lv}`J4wAdOiy*#8M^VF#^^lG^;+QJR?Tzb
z1dtdPD1pgtUb$$+;&5vrWp<|{aM3?Wh*Uwtix&yOx)1jQ&?P8GgLS8tjR#HrMOK}4
zAj)XUi+H^`IM0EivGqD4+%HPxcB&ra@jdBqI2X~!(a5bLKi+kkVF)QZG{R02QIWT3
zyhCV%A6idT=LJ5s%Oc7YMZpJ+CykyMhrWnlLVR97^&1OT9FDc|j_=v=L7@jPJ<S6}
z?>tOX55}QfWh;XT4y(_bBD|JgoFEeI&MUqwB7Kr6)2vm;ToWg|iVFmyWkqr?scdOk
zOU71m^dG#Cy2(pV!5$%K!{&>Si_)pS{L-1wxWzI&kw3P>e!m@J{)h$`nVVW#m1Nkt
z)vt2`w;&-ak8WXEG?HaE%GtVUQrGG<dhI6JbXa<I8mpN1g?CcOAxU5(sbO2kUH?$>
za->_mKbKIXg4(Q+?=ClE0M2deBY*SF?3(P_e1W=7?<(j$ql}{d4P*y{G-*+8+Wg&v
z%KTs{HNbNX_5*Pigf#k2a&;CvpyP+fat{c7B?!{e*w9T<f4<`M0tXfW<g?aUC*n%D
zQ0}vy-pKG?Xn2xVo$srM$O9if+G&8?1WTi>270eIh|+kiqz$>>wu#fFr^L9Hp{6Hn
zWz1r)#qqm(3Z`oxZOVBM+`;+^$IWKLs)iR0InCfeaY#O0%?VLZkdEFsj;PRLIsK9n
zg<{8di~)}Wn;TWV{@}^4F$_}4KN9q2&ciAiioNd6CJ{)-las6O3+hNLJj##29ZX#E
zF#=g3w^bQZYDxydSzN-x$fsBiOx34*vsJk9-_DrBnVenAd)=SyX1F5A5{S3);074t
zdMf@{_PxhQJ<M)h&U54DQ(l3ZD;uoa*H}ndg^>w=8{I)oXc`>U!6HPxGm8x;v)E~G
zVQIX(J0)`-4YL?@#~AF&b^x}DFbZWj!RM0hLtS~(zccA?^KLc!2{{Sov72_pT~MyS
zt|4GTt9Kj48Vwpz|H3f?O#iyu$)!%cC+z2L5k~k@&Q@>quw0%ZTmEvgZZZtbW}@EH
zr!kQCojanjgG3unrK0y8jMKmA(sp@5onV!nMaskvihR#r{MeP3%WI7-D*DA$4wVv?
zM`r)9^>FN^7zHlcD)CXY*(v^}#t4#|+sI`OT6Y3T5ZSvLlx~3bJOMe2yl&J}<B5;C
z-}tZvzDF6Ks&Da!d2oOOd9s_fL(^wkuO-}zF~bpGWZ5kk6nT}qPAZl35p(Kl6K8fE
zm#iv9FfW`dR=AYZSYF}Wg}XE7wrY|WJ>pVHO$0k>Ee`r7DS`8A$yfD^n<KyKq&`l}
z0qA1wSm^ujMO$)LOe#f2Cmk%HQZk3vYo-gzOB=l1=pMPEV$QgJ^K&-oUe{Y<854aP
zjARPwOMIe(>xoVRjtB6YvNtgzu>vH=VfbUO*KQoe?j=bsP+WsNT!q6$^_}q~ge2Cy
z9y}g2{V}d-5uWHK(1ovrm=ilp`ZOPN;n<^ek{Mg8tl&uH#n8<{7_TSc;LB#;Go9Xy
z+E$&R!3a&of?V@uuR^JWu7+~7P9sEFz3!8N{I;R^$MB6r@6Jahh;q(ZrYq)=12<|o
ziff8q13>-sMa7%^8mWa^vfduk4a17#_ze#-S6Z{1?pQ=Vw68Zw(~j3Z(koe@7=Lpw
zu}MXg4DE_xXyK2=%{V(0HHLj!B;hnyay?Ify<Th`+eBacZ16?Aj>>Zy5nK-v3mQU3
z1;=B#pM>=6r`uBmR|Mz%_o<VD*9Z4tHShIR=o*-XrW+CNzSZm|3Q?dbJyt50d)3C6
z;Ze#Za*VU^=rb!Y@V)c0N3sj7ATmWUQ<eh2j&v7V+p)I$8|q$HFs4{lCe<LrwezNq
z*Q{;<z7nJ)rZp}|r;^JPPISw!#=yh?!uk3@y82S^&iFgvm1`T{*8w2Gu^hS0Pgtbs
z*c+*!*Lv4y9WL`u>f<|`AdU7%Ari7L1AzpAn4evb7t>K_!}Qk@@Pu%LMDMmCI9Qrf
zPue)VuVfvNbgZsmO2$O|iy`h21;0UQOQsgboY&3Xz|gS1ABxF@Km1{U@iQkjLgI~H
z8+-cGKiUt;dgI;sY+ZoV$lw~@YTXd6BQ6vs>xqX})MSPtfWvvL;9)uoya1P&pweJ|
zg#p=aY@1QA5|NI?A#bNMf9P=_QuQHW%uvt^aP57Sds(y}D?=9>4=OH>G#EhV!?{=8
zy4T5X{Bn7G>TrKmaBUfnvQ%BwnPQB0oqYvoHX+fl)jJ96pWvC)_Y6}l_E-yF*eq(g
zGW(|KR`a4CZz7W*yG9G8t%;)WD#~@Yg|WXqWD(+5)svez*N_Y=d5xX&D>rH8bZ>H!
z(!F<CRf*DbDMan{RnFCQUQxfXc!r~-odbM+LfUGnISJmgL2X0qKK-r}fAOK4@;t>D
z{#M%@ba}IULj$jH(fK4&*%|Dz=vGv7P>9vf2rQXibi3HMs+@8*E<4ei6<{ch8Ei69
z5V&Ve**Ji|O<9b6uus*zH4jr8d7PV#C)7gLTvJ`3yqV*khdaSSI_sw{*<^O_I_+e(
z3$;lq%0EJNO+iLY@{gp9%+jnfHLVI-5a%fm(8%l67PDXL_qp(}j?d+(Znyeyhc?ju
z?W^d(0+*?|-P1g)yI-{#aya|-SJYir-+9t^O^fzT<MM!93Wj}79T^-pymO9JCtz*9
z5jwMmV>=aR^M@@D@NvL!PE*>h<!v7>oMhH&#k=r^tPf!OC%7}I>yqY*v{zU#QM|;_
zo?wz_#xOpfkgX`8j_F5wq!1!)=*L<(Cg`Z4QaA*gv-t@m#tmA649`0;zjk!Xp0>%&
zwKY#cg{9Ox#joG>?nVmk|GaU7jW~_SFxHe{zhEuJSv6CeLK5ZN&PJKn$p+HtNqzB7
z>E!^37T06MC9}9HJw%6c@0GEqhpXuHTZc|Eb6_Vgk+-d}{S(0vpW~o@5olqJ*}_i;
z-NBQ4q&)$-XVo*P4th$zWyp!zi3AUa_PR2KP6az6S0UI&|IEiI^ggtw?(Fj(>|1If
z@p})$p%A&(^D9yO1}%x9;a<p_A2kr4r-~4w0(Fsev#(NlH~>yt!Ocv+xq3JUCaCJW
zLRoQp{g-~jZX+f%C4f?+^JKNaTC{j-){l*5#<Lj6Yf2`g>UhzWZwesAIi!I;|5+TS
z4}*Jt-SB?6(Kk2fyJK<(%`^u`d--t{T|6;BsFf?y{3I+7(ZAbPzyC0=m@c2@dN!`k
z;wt&H;~RM@@61PrvPnJn8j0B2r|Y*d5Lwq^ADJ7YRjR#2-Zxq>k+X<Kj2zZ$gYUv7
z-O=wJC`G>5zvXDAOgv_=rNJyMpHx-X9iPt_XZH|-KTHL8){|r2>#gK8W$V03DNm)S
zTRhu(bxf8W;f5yGNgF=5#TQJPhIvQqNwZ3G0v3IlGpbcC^q%<}IP*>kYYH969y)>y
z9BJO3uNrAdwM)`3N-4(!z0LZElM*tJ|4}Q{8QtQz5jIO}21M5%yQ!aQu-d;D{j~`C
z#(F+wjPK|5T`hQ6e>_xh@T$~U1ANVLEpR<kc>2U(Q6V}=gGgM&)K7Y*n{4_s?^oXB
zOn4|#`Ymvp=+dZ950;)W2YSR0i0AF7svq14Hv!Ln%ZWa9xMgC=h_lGfg7=|PbFsUr
zBC5lSx(&i<!k6o&Bf5ZYJS>h09$$Z{b53*JKin)n@WO{<`atpGl488ZavE&|GBJ6t
z%uP~3@9#GUt|Hv4?~xvS#}qy5hBGx6_?-V@^hM>YwIClB?LLTT4_wftvjIXQi#&ty
z9&x%TP<$$OO^rx^TS#M<US!;#GBOTcIkqQ?6UW_nl3%j+q6X20ua}Fj>F~A&p->Uw
zNjy3|fbS&6=7@)j#Pat=W}7<Vtfie}U{AaX!S;OY3-Q?x5lHe>339+wf*O;X`l?`<
zcyIHgpnWXVx9cG8UvNLzn|R}jcNSH_h=7}l!v(DJd9`vD&xAmjuZyJbM|>Oo8iUVh
zPJK1$Cg!rG=fef?`+Bt(98A}IHTD>6l1C@IqQRqfyw_B1x~ClBRAYjjU!<sA+avtK
zeQ&d*vNH_W{Rcj-M(`C{q-v>kQDrZ({nkHs5WD8TF8w1Tg#&d|xc?<1fwtcW-{8Gb
zID}~2R0=RoKI8BtwHG!<+gcM`MWJ{yMN1y2ai<D2XXsi(a|hrQKJFIm7PTCZ#tKi`
zn*A8oEf2pNy<*bJ0S=Z?BlxI6lJsaf>v*Srn_8aaLmBRZKONo98m`c5(@<!iAcGjK
z1mu8`>FlX~u9;RJ6J$8XDeMVr=b<pwF?rw^(xnA@#eFI(+T|T(I`;WYSP!?}c4t-U
z@n!?AedOzBita!nySqRNYmJIn7RCEnFu;U9u$6`Fd5*Aq3_E0{>2{riQVlh>c}U~8
z655aRZRvFsTuKAl9)J=DZDP=t3E)?S;6RtSu!&N(bT@RW18h_bHJxo`+HEIPbc|KR
zEJL~t##~2CXmf=?2BzQOtp>neAfQs+rni!F-`t#muowhVdO~_-Q-9RSJQ&Ed{Z#Ce
zymec3>nc(*N>V+}$S?&+3vgM;VqMddu(kTM_Od|uk<P8>g+h0flT}>YeSkwWD7Q=h
zkIAH+!p0tt`)N|M2?4W;bihp#&jgtfBHaaVQZWj6|E=acG$sF{p#zSgo@RCUC=n0V
zJ?nmh2G7P9w%{$7^gm$(6`eWq1*O@_aoaRqMY<kcDIwBlX+VAX)n99Fjwtf1n=a>A
z@?#dzcMSIl9gg&ixOZ%7D>dzZee(0PZcbU!!`jta=i~?v=4WqO%0Rg9{Mes1Pv^T>
zyvp>Pz@1<kW@fRuD)KxZG3!u`FZ+CGg79^x=3Oo5YSG?*$n85wF<#)E)ORX#b>HQ8
zsNk8aJ8SVyiVM0OY+I~5s5dXHOa(D}tPJP&ui^`vE~I>zAG|CN_wXtgI0z4epS0<L
z^0Xtd*&t0xuJK9mstKlN`2KC_l-3XFWqQx7ydInir!7+=ZVO(LVAIzxXH+Fl>QSqN
z+|+|PxZLl}Hk=#Y*rUa+*4;Z42^<Bhz{{OKm&`!?`s+{nt#Xo_$-&EQj|Wm#{4PX4
zk?7lVI<!B=5*&<)!<zQfKw2~>cG^|E`tDNZ?|ZK#Nc4w<xL4TuVz7()Ad#kkVC7|G
zw8y#hIUzM_iu9pvE3o~tANaOeI^gls6k$;P6EsOCrmv5IMSEtXY8shMf9@Kz6KVp`
zY&*GdFud<}^#9y95Djm2MQb~uJ*M!Tzh>H0`fSFwcAb(GPRGIEcNCHZFg8&W)9cvG
zQ4`A%-Yss^A>T|nqHl(ERtMhtvBLL1)n6ESG;2@^(oQnOX4OlGq`EA)&VNtzFywDX
zC&iaOCyr7$55M!Eq1lze!eNpf@wE@B<#gWFFrT_QpS7vu>OHo`&LvdfZ;5+s&yC3o
z%*Yh#IH3srwEe8tGyI!|qYTzpVM}f!bwQ`#k~seNpXq3_aKx0tj(x6cknx=F*-Sb`
z9&h`ZqSVL$^i^d{w#cxGz+if@_z}JoucaY>Cn}Q_l>}B(87Ian<ZI+ICN)2(sAlt9
zSn5gu#McF71~6zyEt#Y|0sD^6=K+gnl7ByyhP;T$j`vjv2z}4Uq(~GC&rMux4Tii;
z9zkJvOVaC+Ir!Rj^d~1ab-(6mwnD5z`fKu!`mD`Eox!oZ)Zig0L-<<?dfYXtD0`-~
z32hr;bN&#{=e<lLz8fEbtES{_iy%;rp^m)1|48_G%Ygp1-I^Tl55wnybikAF@)jOD
z&V(G1p@Wj<x88n?5=43BN;to_iZsQ(6$H?kHBoeGuklM#{P6z9-WFXFE)sD)SRZg7
z&=`E&?2~r<Xcp6gU4?qgbB5`Tc`wcmGh2~nlTkpCO5$knz$^81vUo`zGAvUp)=Uqu
zkjDy_RjNj$L4U0FTvL}RG-sR#D}kJn8$X{vF*iXo!En?`nC%bNbACKZl-=}XmBo$H
z7zZlpJ^o3&;M;BMjfdcoCr_YoH@=1ISIB^lC>vkAMYaglKSXHKzeH%<z1n16;lhP`
z;J>X$NfDM=N%5qF4y83lgmSl3I1N&S9@OgwAH|6H3+U|*=^mpbaGf!esGkD95<6hk
zA2Q(6QVex5T+rqj@dwQH9&`^HE|H%;?UvW8@{#L9kj@~5F?R3fO3Xt{%}~FxiwD!F
zIla_qXEfa~CuqYU{2c{=E4Nb#y;N|}+>T7#lf$MnQ9Lf@#<*C_Sv=#7E7zj5jhP`@
z<YZRa=HP9|PS)+oRrs)ZJe8CTmQ!KyeyF=m%{qxXBj|Tl6g|7+hcPL{=ln3xps~n+
zq_CsymXGx#$-`z^-!lQnx8V_Rv|mz|aAPmH5WzG#+NM0Og=L%Fwd?XlqQeZ(s@$?^
z=OoYzB~96lEo^ZZYXqP3!XVoS(w^*2RFJ~CVYpRKe5EHz9Xu=&`dz?N5k~+_WuCpb
zh0W>FRr1J_!GYS*wA{4w^}&zu60~`MF%1M39ydZUZoFA-d|D@Q*5`$%A*sGn(-jR3
zMYQGm4$|-WH9pg&UuCkvYsU<1%((Vpj}cq!u-POf)8C}fbboySER1R2E}9c}xR(-<
zW(f4)A2gI}h4GYc9Vv~_lp(y2dl>5?GP)%UMXC!wCXM7~&vr@%380@tywEDVMKa(N
z8V)`i<jxVn&hb2${ui6oA?TJr_M;6j^4iT<P6idYX1@0@>(1&xT{x<02HZM6I;8Oc
z@@og!ou8{|&Lbch)Pl6nb&N?(fA6n`>26v$WDxVu>TD=Zp=9@;33V}>-@R}W-Fsfp
ze^*(eK5rb*%{zIBGdsmRIUShc8etV42PGHij`iM2%jySP85aq4HqPYtAAFxx;AWY;
zM1zjcA+(?+8B>v6`d`}f8RDd81-hAO^rLz`W&(aVr|aBo)LVrm8kHbLY$>+Y<FiTj
z?dQ%%+$CXRAw0~oyR?ZKFC@(<s?hW(Y)g*yDkH|3`Ran>rvfIDy(gdsMsX}w$Kwkj
zT7%gOymk05tu(D>*bB{H&axL)T0E=?;%HixNbAd_x?b0Sa0~eFue%gu@x48=mzz+)
z{n5J@bq(Z=UZVwkj`@?bCVOj5igO>F76li>74U?P`9hpf5_DWHM1OmoSh(%?OMTIc
zJ$uJH<(RfM9u-(sgUH%AkR#Eu^fl?8@teIxtirCxv2@7U>+WxpCoiG1H$rFiggQD+
zQyvW_g|23b_XolHjBR?}Kc8`I3GCGq#@gve6{a!_)KS+h%$wr^@&<wi1-GM`Wb@72
z;EVcKo73%XIkE^{gTl4?zCr0uulqOSPBgcNZpTXwF3)$Y>f_x97K;ROw_P4m6Ine{
zUANos2IrD-v_6%X9B#9`P=DAs=&6g<`}J<4j|NXRKcU@ZDIfeo+qy~SxHET9{cb-)
zvr_L0c3Ho3MDgsVSHCOb{!{wZhTOu#eFc0Y@WM`r^NQ;oxZ;V%g-4!-?U80%OgnQu
z;f@ny7EKwZL)?bOx*;;?K$s^~nA!&^#`XzOWv@NzQ;;Tv?&{Wb?Z1)THQ_6H7n-k~
zH=xOYnZTQOSp?BP!1he%f&Frk>0iuW;VHqy2+<|7w|H4#XdqM)y4_AL{ql6w`6=D4
zz;=o1?F+SEvo96uMhAQ1$4H^|<A{qy#{4fmJv0;VSzn9bquGb?>?2oJ;(zrTqR}C4
z6sH)Xt0B9)-`UO#{mo7%vhAb#j!{i1Es3=5HGK(nDPgBP0>#-uGKGK1Bo3o6U86%|
z{Gg(d=$Cp)U$!*)r}f;JlL^pBg)VeAMPpUQC5)$2K7U!x2zTgN9`~=WCAzG(LVNTd
zRL3T0IkK#@IWR)b9%-IY=;teG8xez%aT+_DH@fmzd7F0Y5jfDeud1Rd=08VI1%!ei
z3qR$gsdsm1^H%aM0=)31hsz!n{V3EjE^a5649v}@nUx?<kjgu?_cW@_C<~h`)Y(O?
zOu^U4ZvM@!kv-MU?4z4u8bLPOU8hzOuVLKR&V2I>Vbq?0H=!K~U7gTi&1(4Jrz<U;
zk7BE=DB3J+dOd#%4)fxem3)O?2i&=Kt{YP0;b4>>BZG#gf)YyJH9gnW<Lqs-)^T+D
zws85eLAlziii9zbz}Zd%?MN%3e9dDTVkiu)uI&Me3FDyc?yc0Y6tPl~PIhh<wVIO%
zMDqOcWcJRQ=bS8VbF$ufbUx{PG~<~~CrN>uK`c9gyK1>sZP^e?mofg%9O=Z|AG!z<
z3Y*9S-n|8PMhQ95t5KL*zkcvx=V_4ind`c3cbj~dU!oo?GGph`gC%A(mRD+!w5Tb5
z(C)@1n_r@zvYAN374_O%Nfv9z<_&MA^^`nE2#=Tp#HH?W@5AmIZlPiN4${~n)P3Z{
zo_bS!FHR-b$C)oah3jIq1_wI)EVcK>A+eR{ETiZYT7fIpae_)DeUh327{gAJOuL2;
zua@ofrn@05c4xzo$I|(Rnnt)>t((s#$TUSuI;m7Q?np1+A6&@|CMEH4;hRv)q)Amx
z=D&AVpvcLSrdYq%xU)`-KP}-Quwz9=>#i9uggfUL)aF?Q@%!KgJ+ZE9DAcNzo?qpx
z*UwAs&0c)_j5R?pH7<Bcv^li&1#k05fw|%&%@ppDswkg$m+tq)7fPn5eRDM8>d(Xr
zU!~w5f5CH|2|{A~Zs1lsxs&>Y=E#xK{wB&^;QguMzI*sRGkArPv4C-@y|JTw8?R2O
zt49(U0gTo+L3Qlc%(6_H!5x3>Ea03isdl<gY5x*x*stgJD+zG-p3X-V8ss0zEmER3
zWNieUkF7&NX<@(XdfV36%d|WLq9yoVRFJz=WC}=(<nxe(QTF`|2<UdKy&A|&WL)d<
zcE|LYcp?gxKPOhV`r_Y8$8zU5AkxmYc6`KhY~dnNi-AGn0KKEL^|#n?FM!^;Leu8E
zSHVR+vz5WmN$S)p*<OyE$Av!bBxd>znhO9+;!NB~n~so_+T@@ZO!KPc3DZW;%s=6I
zNR*yVi%4?HTIfOAQEF627QM!R?l+$_A}Mo;RZFhljU?n|oLpIqo3r2m%u`*(>>rA?
z!MMNmfEU%^!<#luDf4>~4|tnQ(^iiBL=~n_RK2as@qX1frZLbk&7sGlhiv14^Tf+5
z%HkA(3n0}o5i$RHK9P|9NraD^garzK9aU^V0@h``MRN8}ti3N6dr}<+)922IDv~BU
zOqNd)x&XEZXwI@`>871)IX8s^ax`{E9xjIm7m99*({(w9Oqf2V+i{A58NhG*?nBf)
z$vgNmpbs5B;M;};J61F?9Nf*ITa&K+U5BuJXLFCwPw_F|Fx5R``d;Aj-MQ`&5o-4j
zU|ME1EBP_AlB)ulDd{R9p=e!SzV(+q1xs`8fTz+ds9^+5cARd_wvmt7O`VaqNBGW;
zfEvX;!4lZvAtSya{H2cq^|uUMzSY+HwbEHhlj((_Vl!OkO91Lt$IpbB+IKK`6*B=P
zMJa_Yn(I~cXTR;sgoBl+O-BiGa$}5f0DC(_I>Jwi%wr#ifwD@dud=-~ZHVJ!Ij=hi
z2v^m2VlwSTJjFB*kXIT*&UA)4KL$bI^q7H8nw<qA>?YDUQZnLhoNfBC_wa0x>|z*B
zq#Y%;T{@a4RTyVK4r}~S<8I+-a#U7>Hb?%zP>&nmuiD6Si-zFmTPMpBYq;5p*A@8z
z*5OpIkViayyLv*?X(&AS!nq0x``vGs<W(ha@>GGyBHQaIRHD4RrA+ae$8KI$+VP-U
z&uQ+wdH8W~))+FTDWxh&ajRUZFvKxjWX@(Ejn2z9BxSOtWpt{JYgV>z+7E0qI#V~j
zmgAa7GB(`WU=)LJUh*7<_<V}a>2AAeH)3&-ytskLLUrAW1AN!cpXw4fA!gw;uG??(
zc|%cXDWUzZ3xfC<y)}y_=&nC@&A*_rpq1EBg`^8DT8vuu8n$6~LMu@1DIBO$KaOE9
zR0sF-0i}NMIhf2-WDh>yW7W%s9-`FDxw$)V!K&#(<iA=<<A~=Ae9){B#l$`)o&|u4
z7JlvWlgG=$Qq_5nF%>2RRS=Zn)tnw3-5gXODS(|^L3yfCNma$^{`GT~Xw6&>Gaws@
z8K#S_ZI`z4?@>EsfX-z1_TMSUfXY^V*JZjC^^&XUj8yokS%=cw0eC3mpaAkx3e5Lb
znFYirxfpc!2AG_+%FdiOq7b&7(4&H*E7P6xQ%8lD#*qBhZsEKIQhs*2nRv<Hs(8~G
z+V1vFL8DGO2<a1oEYvSvOq<hGHJUWs)M;Ot;D*$CAw4>(-;lRHXXp6Oedd2I=rJkh
zScaQZQEw2JA2~JypLSUxEjM|;zEk1r1tC=75dcA|Mt0eILtVeRNTOx-yP&5AT1(=z
z()yFcH~`H}+1!+aN3*VP(r>HG0UCX1v|%r|qD@)fUf-b#PImFWf0s+QBZy5s(eE)Y
zx}#hfDjXYfzW-QoFMne~Fk!MdtlFp)DluzhTe@FBIQ5O_&?!ZRQ}PiGHxuC2^<i2L
zt+~_+2c<0+d<&nI+!>EJtU@v%Gb%bNO)H4i*v4pBmQ}aEG09jP@;|DUMLw?K4WQK4
z_dDi&>`J4VPX?V}4T8t`K_+^BEi{tGVBA#BX`{M`*i}VA`l}NOM#gjEs-B1oz4dU^
z#p_QWDE<$7?->+T*KG?oK|o0&3QbN52uKu=j3A(b1eK)Zj3mi4&;*g(B*~J4q7qti
z&Ka7F<RCfcobPJi=M8nv{i@D+{@g!bl~q_24SVmk*PLU`ImZ}bZfSjL3K@bnkmJ~3
zrDs!To4-&G{d*2ujREm-+_)3_5b618gQQ1dDJhVPCF|ng$(oSO6O%E6g6$s5B1OHC
z(&bd&L>B3e$?22>q1DN(#2M2EHj&0_s;BWTmQ_pN6C(ZPCZf%aKG#fv3K}GUG0r1^
z<^?vh{vrRo^{kKjeUC;6!Yhk^ZFhoZ_>9=s%Q0!{r~mRB+XdD1TS^qh!$czM;Unti
z6+9=P%>Q)tB2BAavQ0_NVj*oA)^W{H0dao0pCxe*jAvOp3(9IvSuo*~cM%{Kpv^xi
zylq2w>qzOIvqo{X<!b6QKPBD8??~cMB3yY$Dj4K;w%YL{#R+t}OwHn_2n!k{U;R@H
zK-V$i;sYj@VsZbcnwqE8LE5x3{o&nm$y*#j#xP>N&8G>(SixpRXT6Y{zKjd)*2AtF
zAEv({CsOiKbo7Ulc_{37c)4vsZ7tmQR=>|qucm2I^oTl{h_r~n?EBCfBBmx2oG~&g
zk5!7cU3bJibGkGZGbkbD!!}r3j&aW70FgA}yg>AB0ryAD$Qo45n+{l+AxG-P?8fa3
z$^O_cczZj0nqQ|0=kD8Um4v*cbvbIRtI*No`9SzSP2VnOciZX-EyVK;wIY|^#|0^>
z?p4z=W?9cW;s7b;@Z?Zjs7}fFJ@iov(_dLy;)fu9$u>GPeR6DkL?y)e7HQCCTHHTM
za_~cB|AL_hG}~n*`XX?=K&uqT3$X$s32DNwKF?Mn5NML+5v~h!Y`f2ro-8(;*}8cX
z9#}H|Ag{6|!66Cho-y2yAH)~X&VSZ{8yaEQmU4vIrWMl_iwbG{z0g?j7e<<AFTv2$
zseUzT`;a4H`w`M|o-lD$h)@4g&NG@QM>W4Y;qWlJCiF%1VBx;&`_l^E=}wgSK+l7I
zS;NE1mHm&?{|YZj`qs1sbB~Y+jj!;`6zxSfO`$?mVfio>M4-UjyI65wJk2oLWVtN3
z;|BmcKer9_DYTyvlz+7pZ<80Ex1u||)3@2hW?81RhSisyv1FVpw3h3Vf12$;yX(<<
z{U+~CD@5H}6Yh6LsD`ph+!ihbg*4LZQc&D5>b?%7uiT_PBFFpp#TF=a5z}SQ5i$18
zfpD=cwJd#KY0l*FO`n`MeqJtC*F}6qao$Ino9+NSfkT5$BXt~mq+0lo-YO}-un)Hw
zH6O-PTr$1>;R7)5@UB(K66=ctWspa!yiA-EPNIolm7jLzJsdQcak`v26qpb4kJ|qA
zAy)vxG2RjfH7-_hD&c3E&t>6ReHs@gisQw>>1W)<gTJZ|NIjxhbzfq23-a=ZZ&zo1
zGlvAE{c77$V)m!=)GJs~t9_3Q)o!j4k5O;StJ&;HQlUp8Xm$<_`BO*+mb<7RH7Cu%
z{3%NzjSW3u%$+^ALaJcmx%`r=$+5d9b);`tp$b-=`i2E!XVrP)yWGR5u*dRT$vGFn
zzxa^Nke`;-yDMCKpb2=B;?|b3!jBPnsD6SoX&3`GJVx-QLC$WiFI?>u>I+H0>hP<2
z{yJG*cGNeGs(||W%Hbt6i|4a>k4qLF7GfK4Q*>FVQk55?zpcv}W<Yko35GtW(zpXt
zb34-e`1EAY)f-H@mKFWf4A@t0QTWhU?i&_1tt_73+pF`N0^z$d?2v}iHUAPlRw;J0
z1zS_)>tM2>v@_*(l-2~Jmn_dIc3jsF$8ds}$8eAVpc{S399pzgv`y*hjvmRhNYIA_
zJidS9_3A^Ncy&M)DouzK4q<uj<g{CnsRo72{>qG~vnYT|pbl7TmrHpxEbl5?&vUWs
z={?oV*Xwp~ysF2@kUWj&;7ZZg0!JtA<n&gaF;@74#4&sh=-xYlQ!_e)XgPu}PJ9H9
z+{O3Y!ED?<pJuW?oHAT{S*f6<9(ob&sTkh_HPk3P&8`Y2Fel2TZS3#qv4B9cL;O|(
z?ydQOIR&i}o#ybKcV7yB9sH<7Z)U~=HZ_Sp#7;(^FiBwuW6y7yTphFhgY<*hw$t3o
zd7rB0AL^QklwwQM3ig0_Bj74OxmIPXVn(fUgmY$?BTmXhAh5-4(~?0aJbgz^qkiQL
zk<k3jSONqz!YYo+pJ9>u2V#OQSjLT)7G_fvLbk#OT2cC=A6sF@Og6tiI%mAWm>UMz
zliEsIU4t}_{5*l_vlT?cK`}wmEkb~@aQa0S%ehVugnQ_(!wmi&&03iXtWE-{aJFq5
zb%9_ojd|fC2r;tvaPx-oTu2?4grIlyex>*ZsE1AHs;S?4m3KdqxQ7v>Pw_Hma2HO(
z@&u<1+epug&Je1p2<c<0R`jN>g4Mc7KM_s2g`V;^m9P7ZyQZvz+b;a{ftYp<?|}{k
z=JT%XgaSsWNm%`QIakO=#}N+DK{q74{(W|Ci5-&Y01A^KkHo6A@Cu#2VvsbvvxbG~
zebwbzGKQt+Y8%iVt|WgPSe#3|zNx2w|2)aC`=q;G_w5$npEl9F5LD;uz(;a~gI+y7
zoJL9qYCy&6Ya1)Hc^}6s_uX3##{AUeJmM%%s0Kn_cR+#I%XnNP_spCLs%~-(fxUhE
z_1aO6+j)7O0i<m{q1~;5qvz2aLCT$@cw>l0Ch~59=!KRl<D2a-caV@!{cpalWu`h*
zpU+K~`P&x&p{{y2cDTt}PnS^i!t1i#hIm}$yKxlxz;bQ|0R$JB|FWQgt<!|Eae+rK
zxV=H56m09*DbJRY>h?%OC84S`caCyYs!>wN5UQ%Ak^<5{$m8_bI+Lrpa~O;`zFNTu
z$A`(VN75r`U6PF4(hySg2odD1Xtz$*N^b0i<)!@rC7WYpTil&R9yn@KFmCR_VX6<2
z9q$F8ElgVEY!lT<;Z1#^kSidd>)&ZU?q-hFZ}hvzQKJQ0h`w$}*7DS)!kDq!Kt}kt
zD&m<NQ>0N9ePqh}Ijo%-t53XvG`oY@Ced=q51ZON(c76sks^QeN8jO&<mw$l*3^Oy
zA$CtW8po*05(*F-IrNlU^Iop<`rUPrt4k26gsCj*a}3|Wk`!1fHoY7C65obCEsIj8
z5h2>1Ek=30ssPhiEw}Fc>|j{eIpKnpMHJCH+=xfC9*|qA0Xn9+(O+eITHj?rA-h54
zhnSV$zDvw3CTT3FsC|DimY`qPM;n<l-8mF^Nal}DT+7IU-KZli1W%6Ug=&P1ANLj5
z$PK3UiNg?XH7ddZlK)cLU|&;pUOkgemRVKJ@kpnblgn^qG@;sfidDDyiFnm0$Mw$8
zR#hv?J6?o4uN7}T@OZVoXl1Esm2W2})R_ViupOr;tg+Ov_Naoa|D(Fy(1yz|Yp6j_
zT2RMDzD5ztT*Q$HB?J`18w;+7?RCf+6C_sITE>*7bHB)t4Y6~~9mVmISt`<hBgFaN
z^Dlt+`Zb0$f^5XcOVQ?Z;-S1L`SU1V^WQo);5ge%UP^W6l*-{S#q$!jFoxc>Z9k&1
zM|~JijW@Zn*#jR8-|muYYoJ@QqMPvFyw&5fyy`A_r6}=X&kqo78&dL45x;a*W#0V8
zx42)}Eo@=;+Z$IHiSofEcA_fkW6D@-!tp)V+j`ZRri?=GUkY8go)rP|V~!@P0}&>T
z3FMLc@p~3!hPtdPsce<7`!TlUdC4#xHKw|;gkcJXhi4D@4Ilx#<Njw`wG!okLC2QY
zX90seQR#hJ*ZDyy_{s=Z1-fmt?1?YI1xWhayh{=E=aY1F&J(t(vZ0sE!p0NPVSxLZ
z((w@SJ^D(P07CK&E(~^DZoNd@_2jEbPo4Mm*Dg(uD6&Ie%TUh9XA?s9N*)wIeYCU3
z$PP$|X=`YYdwNIknI7kKUD#X}#^ZRsZ(8&BR68CuqZ+qur5T?HQ?J{%mQ8M+Wf8gU
zanh;FjCTT)o^rjF5&F>_wF{Ec3swjFYR@E-w#eZZ1{x%OaqxSjwc9Z2;nk&vhefkx
zg;WqC8V|YM<7GF#kUH)g?(@x+Nv3r*7pKh5qPt**z<{!ve*m%_KuTnXls@!xoLSmD
z+l)g!+VjHZa-BJ;YUR;9HWhpaX=C?V`fPe=<kL0uVk&9>`J_9-<!mQY_sTr7@d4>o
z;x=Ci2EF6T0<qsc8@LeL9&DztaZZC4V!no+LEhu==OOT8-m2PD{DNQCcpEMlDX+sf
zt9wVQoPI7fN2*RPUOlh_ZXBgud$D1Yq%W7;byqw$Up&Hjri=0MgE^rpqW)OH0Z8}>
zelTe<A`&{yjnj0~1W<r1BjJeK-?_|@9#))1EOGdH9057BIE#XV_naJC31F+Lg&j;v
zlm_$FIT*i>cjl6+K6~6#O)jcF-|fs(EIK;9b^EakNoHuP03`E8zGVGcIIP+#_x*^z
ze{mD!X0w``&I{$kh7UB$;y?R<`pV)nz;7%>4Z9w%9cN0cP1kzb%?7ggWli!>+4PI}
zUC3|{F-8`KhvsxZmD!_KoGqG1mrpdTZjVBdUp5Y|T*m5@#V4vjO`B5X-K)^Sgg0pv
zIMl_-s#~BO+BY@(UJR=adZAif#ee#ClqV<tA0cA;e-<JDtU>+P5D^x4z?L%&3+sEv
z&9%t?jYo=PE?oo8Of?z46mjM}ZY3z_FRL^2LI)MExjp4HZWRf5BJ^prz6?XuKXU%c
z>?eOpW&cF$uUqN@g?RT00qrgOE9tdIr;Z&aR1mVH0l)J{|H~NBW}+zIH*^CLn?sW!
zhcC_NFaa}G?SG|Kt3Yk+a4&!UGxUU>YbCrC{$7IW)lR6tPA_t1+tp~DKtt)<9Z*9l
zp6#)xrJ|TR^-=uv;#Splzk)Y_l#ytN?NHM8G|iBk&J*PI9T=BaW?-4-ULC6~n1({`
z-dIK=kDiIuKhkCp_a@?l%->1I+tu{60dqPDiM**<$OTn{mEYtN4Hk`)XH!qF+o)(F
zUl}lj5H(Hr<fU`dvz|_$v*5pXk<_zmn8icVG(rM@JI5VcT(GnVhG19SKXyB3@pFW-
z7qo2y=C|+`&SmNLK;@I89(}vpi(R@Rog9T_!<)0sUf1qYQg7XCC8fg`XT^tuCYR{`
ziRrNtL1x1?U_;+68#<`n3#3|qzuO|U3OH`6j)ZEXX>K~R2vxa;{pLs3#Ta^1C3k{B
zFd2d$O{%MsyHDyvOh5Hx5;dx>76V+eS#!3jc#L07+?*q5JU#+ce%8s9;I0=e_?FN+
zy`0~VptUMUGEF0rqv2%yvmDpN(&VQ3C$0q`kd<+k&wqM;)!rS<k@1|&rcwOE@js^h
zlP5cRM*)e<eSEF7a)`g9Vtchvdt7v>e}EV1X1&g|d1TMkc`5dA`Q298H;ms{jn<Cl
zWxI>`d;`bZ0hON|b4x4RbN#Dw9%{L7z&1-GFWnVBFlZ0)C%P>q#C}THtQ0;Ok2W>d
zA&vA^lFq>K(z?pNaDHYIXL~xY>#P+MUz4)SEiAPBZTjx37n3aI7c-U_-ElZxb2YR{
ziEg7S?HK+Wk>{70(vg*PwfQDv?#UOX3Rse4&y@w{U__%_1Z^&?Q#(#iB&OPk&KYlb
zZ^vIg0zl;qO0|=$ewUuf03ww_>x$|Q3T{tLF*3c-*{8noxdwhV^zhOUb%b;`|B69B
zMZZJYVwzhw;!7@hn`njGb>FP?Rah6KtP6Q3&VGAr<%Pg#X4k!{8xZ4NFz7yMl1RX1
zV%PokOLkrbo2d5QBc<whYOwid(p`x8<Fp(M`n`u_7OdHa2;uS=Ar;n?Pp@nEmLBs6
zF^_~tcjF+VTl7LMnL%((j#Huk_4cU|(KDU5GZ^8b{YwTDOH1R)xCfsvV_q<@Huony
zui`CD#$d2LzdWjSMhXxU;$85C`;Xp!#dSO!s_ist%+JL>ur&lG5$FtxRTf#IaPf_b
z2~r{!DDc0h>}%_TT5QuZ%6A*wUnOM8U-v+6!f)7Wd7Lch@4oTSDp4#deS`@0cQ*;~
z;W^EQ)^&vHcda9XICwuQyb<&B?OA+;#GpTFGRtV1ljhJIZXI4W{i~Weq&AejIS4v5
zdqdfA|L)KL7$ezHATrNnQhj61dCG9{vrB@hVvxHIm$IiOeO__CuUjybycvL*?N-<E
z%d@SmshXc)qP}ch@AXG#s(JZ+o+`FSGzML19cVc7TR)Mv>&E*AAxMIi)Yg*bupOsO
zO1qeGtIBNK4VZBTFOFhcMcbImnL%F6;r_=&#LP`s9*Nra=k^GD4<c0u&}&Kicdx}3
z#9|HmFb&7Tby1F_Bk1&A8<r?ImIOphY1o;J7P(;6ea}iBg9W}AC%!3B!9SJi&6(iX
z#{KY5f8{~8F0;}<qZnRiD;nioiYKn8Qc1CVKz$_ep1dyW0Y=X+f2f_qy7xZ2B`Pa>
z@|+(XzIa4upp89Ay(<}GNuI6HxIcUCW|(IVmllxL*(@8%j%qw>FdXVt6hE1t@0BKY
z+{mvk{<5t@Q>v7@faAsLulc)<lpIrah>o+0<d&HhHk)aTpb#6sTJEJV*`1|iw@m@t
zg7;GH*fyC>_CTn@-t_tg80Ddob?4C87D>VgVIjy$w0oX#4A;`AG=RB*B7f!Exc7q<
z$)b!}CFCwIR6djBJ%=?@(A<e6d6bEq5s9gFcvpj!@v8B{WqGD|+CTqxBuldY3|(b*
z02nGC&Rykxw}2a?e;NPOHEmSFM~UpGyQ}+D{CnP;r)w_O%RJt`)WV!sB03Fm6CW8&
zXs*wN*iNuv)iGKHpEnx}|0v-%XbO0)z3uox38z+jV-1Sd>nR=SFfcwAX=Rl%8-S9u
zpeln=m+B=UwpU>MJa$wzetm$`{0v%`ww*eec}+DX=1q#s>a!f-TVL;k3LS65w}|0y
z`EgzBd#uAhsln9n;oVMSPgG4op9d#ODLWM4bOW=)7M30sB(5jaW@9HI=|^|<$wHd|
z&Z{j&*I$#BGmC5Yf*+IzREk@b9ZoqTEB0RPCo|4C)<6wljcivZg~qt|_<DE^?S4J*
zRMNR{arw115r$i5<vhx`!l#-tU*eyadpg<5vsg&}$vi7TVDw#U*DcY`nT@V1S>Y5Y
zT-UBSg#7N3mfgFBs;i`)9a7!A`J;~mVGajN&Cbc+68GGnDuH2O(cMXw4Ht3ag3hV&
zdxqjvD&?Ee#3?WGD$jz9_N)Y)a6F)p2GOe6xma-*W=*T4dUQ`ICegDwz8%zn)U@}y
zn_@q)=u~hCog~ghMyqKvXu>B`K<j>vz4JrC;mJXYrsIccy|t7z2|D-9P@^=NB_%cL
zDlNO_7vy(s+}|V(>mw!N5)Jma2kJ1|i7IWP5f`(~KOPzUVo#wJr*k<JI1)X*X5>RC
z@HQ(i!q~+Igkk|r`Y`>$aJ3@FD+a8Y&q)p@FRGULxE1x~7(EqqRN{`eE(@pDoWgZ4
z94{k1B25`BPVC)>f|rix!kk+;t9V((`WhmsWw{ajntoQ$&UHX~R*VRW<&B!q;XJxv
zO#U%?(}(;k?^Lh5`-;-6OnKu+zSr}Nrj4}ai92+hh&$V6vfDw-h&GrbB1QMuwB9?A
z+c*T3NYn;dpUTkDw``{Nw-^82d~*&?CNYBD+~Vb+S7pL?dyk=t@N2a8;0poypOq0I
z!fhuj%Anf98dhNbbm{(<v%trC3Z{aeiX=U=!CcAffS0Xm<twaoVL&94u#KU#y{BQ=
zLnt`W3qwAo%FM@#7HHH8aEkKy!ndQB8CZ(B=w8(?oTFZ3u}Tw+%i$G)(;+w+dO~}M
zOFb8OE1nOY{LmMJjH4zEY8aV}jGT`@l}$1t3hej4c4yYYD3nS(qtB2`ek4)O<NMM&
z@6t(CC^SQyKv!MMR%Upra158qguBjIoy~r}2=|4<?IJ>>b+N+{ro!pFgs6lW#RMWO
zW1m?#kbkfh9T|D`FWg#SH8{+obA_J`RIa9@Wyb2LdY&Sl?ZZ$#4Hfcl45M&ykjuI*
zK9VCf;-||@>Vn$@K8pSlt0ro{#iOpEx^rFJ)Qx7u2qu*M&TKzQxk!<_l*z{OsU8zs
z>xvWV7=!&3FRF!sJtYfNx5BbPvf$wWq?olw92dG7aAw%nG7oC+$k9i1;(a4w-uaH*
z@4b160jq&&x{Q#Q!{_!vZ584VjKw=a(|?y@Woy~>SJLSvm@QM2cpM)?BhfJwSYQ>E
z|KTuDJSF6W+D$0zoW$L;MHmy=Vc~}urr9Ma)VX+4b9_}59aN5GAKXgomh+rhvgNTy
zbDekjHT#x*6Pd-}n&}odz5?ktrFC1yEujE9V~1Ou3)4|uz(+u=A$4f<Ael}fmM}#9
zBLy9+U*}ue@fKoH#AIdKUygc}3#O`j=3(d@v<(kLzI9*h)8^T_cQsGS6~3RcQpXRK
z82oe=@0@|9FCH+Ja>7VkU`D)Lpi!7<-8aoYj-?kbWHl;qqqN|{`ZIj!$Q2+WI>Swp
z4C_xZ>3iIpAtW6gSUh8LqwZ1Be4XOc(_04xRgS(jBx2b1u>CQV4>zTZvsX=W4S()H
z;Fmeh@vT$$+rPLPGZ1-qW3)(_=#0nxkyP&eYS8gMV|~vLd8Cbnhf;{kAJwK0cSY7{
z5k76QBceRhEWKG7kup~b0L*~Kn=xg}S~VnnkCxXcVBnWYNY9Me${Zk^4p<4Z=#Cgi
z$oaZpyOKovj(^gtR*T1_b*a<9_@!|zDxb!3(LQ%r$WYh9JSSG%(3jUaOuLHD3FB+m
z!H`H)eo}FMkc(mTWy>ZlkCW6Z$E4izmQ5(HQ+|s^b4@<J@TPO+d&~8P*!<)SgsLnB
zdmOa#*nFLxB}DtFgZWRNSOziiaNNqhmXY=_3DuLLsceKQH5L0q0}b#)(}*M&ib^ka
zK(ZcbW{Ra9pMkEGqf)H*%<n95(hMzs@3I-HM9kRa77y_b(YK5Q%KYMY!sz3uJ>PA)
z!P7#iTru5(pfJn{U)wb!zm}axTpR8=GUEHDu!Al8wa2@=S3^vJ@?qLh=C8<_EO0f|
zX`RwKnl0|&wNUa=THLb?pmU<K{OLW{$g?wY?Y1Fd1>%5c89FXw7wgRE6*5_EXz+_<
z9HzEQHRt83$1(-#+#ef%t#t(A;mUwoK{uOwdVZl_d<$h(`NYFQtI@#3e5|1Sp_%yk
zb;j^~d>=n28y0)3yfoc>QX|7^pC_Fn@L(hW{ZOkZN>-@FUFE{CyRY`;o$7ySFx{3;
zrn(OD;kx*UeSs`d9@mWltLWD)oOdf~Mkoc#F2`=5qU9f%M!(LxO06Z726-{UBh+Jx
z^77*|v-TaqcGE@nNHM3H;|W&#@et>6hjSZ^t2IsXgl3!vmP2nmi=UsmzI8gIEvsA+
zv6}7xLRysXl07WTs;?CHW^n1K*4lf}bX${2LQhJOvhB1pTD~x`<BunMDcyqEIW_5;
zHjTxo&DyDU1+lKz(Jlg7j_i9a*@j(J_$VvC9H9}8vTgT6p@ET_MYp{je^S(dlBfl3
z)saf!rL0A~M3TLhDkWRQh>y>qc>fpUH5vgy{09}O6x@3Ri)+Xlo)pDn6oQ(kqiTD?
zjBmKgJ}%q((#<77q~>xpq054+b|%zcV(&xw6<}mx_-|Rb9wn!*j@^r)H4>PzyZB*$
ziEr1GS9@Lz+l|QT62B_-*VIdcb(yBPTTi*#C&Jd#vb40O$3+R<c7^PFRn9bPq)|Z<
zx>M7e;`)xKflG0vEunl=Rwc=r@B6MPT&-jko*a+48);VYuB!6wPsyRAf*+VUo&0h#
z*|QSNtC;q@-!bbpUYp`MbZ-+T&M98FioF!8;cDAyh((m$P>U?<(>g89Mycz17<*p%
z<|D5O4D2VhxbN;{l~BumcH4BR<EZT@SZo>yD=_9Y{Z*a6stR3UA{MnS(Qs?+xPSJ;
zm=CmY3FZ{t(sYsDU&Bsxt8KPUU2CrvxIB5lzD;$@JJ?*9&)s&l5H-j-VqwYFSooUX
zf|}`B1hP2-L*MckS(0Fj@hysyA^h*YS9G({B~3h;jw4b_v&ktaSS{kGw_950<RCqa
zY%1KUIis|Aq~j>2`&44LD2cUViLa(^K#?ef|CncJxs9HW*NKrY%jbYo4>W!sk%z(g
z3uby0#}X$@#6~GBlxK`!My4Y79d_Rchw@IZ6z1A~O2DrR+h-z;G%@15%uxvk7aKkw
zdh$dqfFis(aG1?r8~(G!FjR0o_fvhBXHG1eI?MW|Kr7>gdv+>M6wThfepA9|Eo!IT
z%|+S?n-Ja*_~egBjPQ_wdNyVhT28Sgl*VHHLvo~0?#srJXJz>}UK)mwCMAYa8$~~B
zp=^6Cqw*I7_|~D^%{3IzKGG#(t@D8ZlWCQl7f-)8^>H47P^--lYg3oD8Q&13m77>f
zE$xCQOy9#GghxB6D7g)xItZPoS;aOszr)|c#z<&a1c#U92j5b?|8+fAPB@mfvb?ja
zXPG!By|q|rac~f}sJhRzFJEMz{lwM^Q*+u{)P8|Z7F9g&E2Iw1$<N8}=--agpK=5W
zdSxD%<>9bG{-1VM<D8$Nm9S2@X}5*0(#AX!l$$(+A`bUWSWw<3-&U4qnvJiRE2abf
z`=27m<MZ`dKIKBjn4bHD`)-*j1wWA|yiQm_31|IaUe4O{?LwrJd9>E15USpMD-)x;
zh|5L8$J8*yeezsj^5`s}<FHaZnK8w9jCGe}onkm{E`306J>7<}XYl^KVaJCh4PWG%
ze~xWVw&2Qs%md<nX?XDsxcbp|x<UL(ioRa*{;QV)+h1vZt)9ha55)8?D7TtR@5j&k
z2t1!m6%X$*@lAJZpe&@Uteoz$pg4<*eaSp%gL!e2HyHjUNPj(TiZ`}|Hbo0|a^AFz
zbk_p-AXMM@EE@_pkqgylaNGaf=8J8_?_h!;95Q39LTNP8snh*Xyu9Kv9y24*6l8xP
z(;c0wZoE*~t&FUY2UtTVdV{4wSi|8wK!xS0T)aSs=%>q|oiHBE(GVE*9zWgsqPbx5
zn>pJ;ph`t$G-}f3GRZ*4vlhLiJ3m0l1zn?Pp{JuKXE@__mW&Op>P(!lPPMg%4_yln
zGXdn?j*J}J=zOZ631{S6JuAF;HUdq1oGZ6c@qyO4gO<!`>w@oXpAa@qhE<Way{a0G
z4?N`;3fHNF0u0(Ncr<#6;tMl$f9kAGYDEcUPiD@Jaa(w_EofRjlufxFTrg)wF&Y4)
zE#QFF0D?#&qSG}iB{))ih%A2ZR6ILo!iEq(F`4LuHtEh<R9l^>4=jm4{o$b1P^ku~
zwJfVhR!E<0yE+(rBH=`^=M3wDLO_e~x|zK^AWV-|JZlK*ZX7fEQ5RUSD~UQ(p+AX!
z+tC&u{gjcKTU@X5af**Mu2}Bm`cjLeMh3ei7|D*~VU{1(`9lT^Nz>&7EdYk$s4ltD
z!wA(cy+@3lj97h#efQFae~^RX-a&F-o6T}oPOFrmpo?(z6;Og*vz8`qI>q26DL4g6
z6)_ar2;LWR2t>f;ZIOGGC+S~x7~m;aq3a(MCvew4KFyM%#(cmJF`D@e5V_L_Ri<{i
z_#Vcm8W;te(t*tlxKeMbz0&CKL|mg{mSjES8LHn73najgBEZ9A!TZIb_mlM|4yA}x
zRx!16FegC)UhsrZT4Jrgj9e7bp{N+shwz5R$9`9g#z!CTVY@L~WMX`I*wf^7lOf1E
zq!}s<Axm$=?QkYLkE`-eg5nEsJmHyP+QKtp-Qi(0c-*l$>%3FoLNg^tN!4Bp?|Sof
zMm0{3jG4j=o*G$@OocdDUn1jpMrzyugBG&3NG+(nwOSxci7z{q)YPycytG&}O%`sr
z%WGohi=h`$BBO2m;dU^M4*`us;W(MG&xfF9kumJxLg-l^^UX1hZ9M~p%BDv|8M<SH
z8(4D^WP#j?hPIq)Q!8&9Twi?+iI5n(=ZD0jXT?9x$XPDTalTWk;S(-3ZZmis<`O&@
zIRgK=G=E1ZvlExhN=PZie+Kgca*}aTsQg&MwE9Ppis;Fi;X!oBzWqvmeOB5iT39c|
z<@mn+6on^JWHY?xzSZ<;th|zai<x+}x{gCg`QDJ*h5AwXVnud*HPP|(-HRB{Y3sb|
zUFNfKB^<wtuWP2=72BC?Of~jOmA&p~8+Z3Y?mY1qJE>aXA606(FLAk@5pUF^WJ>ru
zrqs`Oe51ga$}oR>#!!dG;?i|-I@f!Ym1<b<>lMM!NPlR2lOM^XJAU^Zg492U|4~&J
zO^WOJ=;Db!SCLSPIFCzlH0sj-vgeV=#o2aNN~w9-2m)r66AuLWa`Ky<oh&y@7xQJU
z-)<De^G7A><yFt(IFApBSX-BjSV#U~{U)a*afm3N`TVx4Ged`uv^L%>=jcf+9bUKi
zQqIsLdy>M-77HP)iXRC+Tr!$<XT7>~R)n|2i43ygWvi6z3DeXC>IwKzfB(xP68pVc
z8$O#**!Y7?rj!?n$GP!&o)U%YH{|23oTeH$3XY|$HbYdA4FM9DW*<&AX#L~W<!V<?
zKl~cHEEigsAZ?I;N89=kLM`m@E#Wu!WvZ9_QL{pcvf{1?{FN}bhyL=lX;+N$=&lEC
z&MRKLKBmgUWB5&my2P_B_4~zbKNP&4@xi%p*ysISp4Co@EfS1{r%=cjV)8zu9{K65
z@Qmhh_dU<Hn6?~NbDpPo<j{iI4!2N;^s1)!wB^YUBXx~rl6|!rMtY8CG&g{db9dbR
zt-@99p2B$dLS-+5(H<_nI=UcyBTWQrglQq-D1fe6Akt(}LSeHV_vob}A`)?xDN&6#
zHy}%^;sgpK${)__idpgS+N7YL^m9n9cXy6rbJck|GAt?Gb%NXr(jZv!1nYP}AR=#2
z7c-&5jHQTVjM)((YH0aPXh1f6#yzD@Xi*Cl_Khj(-a+K<d?bu&A!vuqvna_S^+jKe
z#<%7+#2M==qe)n<$}MxL!y=vA6S;#Fi`i{#93E*CQeWTQ{pt1boEiu5of`_gNE(F}
z<A;KMBA<O9D0;BQFgP+fxGCU~rS#qSoUA2jX@o0=Th|B++ViBF#b%#x&}uPuaO=%R
zXDqJyf=<U=7LQeWf`}mpidtL@_i&}@De2cDmNnn>mpV<Iu*In&_BdsMVkv}2soKOw
z64!w2v(f|V6~qXPs48APpT}=kGOe2*+Ef4Gn&w%@JSyM+tTP7JR+&rXQNNQpX_59S
z{VBug^Hu+Yg7z<)A~;(sUMeh(iG(Im5B47D`nTcYFE79DoqWiCJD0S`@%Ea&WZ*mJ
zg34D>f^h7OM+B&HD6-*qXNH33ofh@^xcwN4#*RaEo>NpPb)(dx<-^D7LppW8vtl@Z
z4F@v9O$angtBV9r-9}xT?K|2`lhXbn4OY!IuJrLBZ~2%cLoqy**Xp(=q4WhecaPr0
zp?67-Xs5qP`#)2o_C6`iBjeTy4579juHF>BBS~3p!w~+7mU}h+GoaEVco=Or^Bpy(
zJk(Z^ssL0!8sqRONJnW+$;5xEL=Z}#<hZSo{7bBS{<E6-^Opr2amaA8p{};VSe3qf
zWg?VFcxyT5q%4?6`^hR28WRx`r~5N+X=y3bzsqU4w8}bAh|OLn?;tIv8+Y=Kg;bH|
z*o`$p%R^~e1Mz3@0-eUs*61M>s_c-M6=$(I*OYy+!M;ulD8vq_Bpv@J`3ZyMS5|7t
z%1+#5O((PWO17Kh6qR#Ug+WhMZ{7n9NcWfamxD{|wl1=qcDsg(b^M`l<?nZbnlR7&
z=YkW~aq)sxJUA)QE$FM9*r!?YyL0b`4lcD2mmRIDi=+uOyLHTOx6J&He^wgo;db{x
z`3j{#zH1GBEatbj{J2kEzc!rZXP)mArnaCJSv7Tu<_%lCqCv>bHB-^R`RjPvc(w<a
zI2L#w4tWQd&-Iu^hOEk7UG}#Za<_a;k|8B))0l7K!;;c<y){4gHNUW~J(Lwkv%gJD
zA^B#}b-a0vo2BF<Rt*i$+`MIoge?iYZ$398;J_Jxf^<^vf1dx=SM;Wh+1=RClt&}&
ziW#Po8Qy8!a-&`J0s0Kng;H)Ot&L1OzwjOMG=tj|f>`BNZ+PMnvrcM44QB}Jd1%V~
zA6MzLy^`w}NndLoZ?nnehH&(k=snkLF~G%VR#y1=d9=iZ=pa@`FzS}V&^Gt%ICij)
zM{0ha`=r5bobo94U_m?=jzodVsVH;9c@m7)QkAj$7xY>@cA1{L$!29Geae~1O`AUO
zGqgzI91f<i4#&3g6W100Z`tR4>Lo1U&8zI=YfF^Iv}YBds4`fto|8tAa?lV3|N0Q0
zlh^atJ#bPSM9>}S{3_Qpp))f{G?!eU{;*@gbuq6Q7J0~DV}h|Z!$pt+KP_JoGq&OL
zwKD7xKd3tST}@1g^A;h8W%9Wm7S_0ztS;$y^fgT4YTbTsFr())&;sd&UCmv_<8+(h
zbPE^mwF&z2d$ipU-OC^z6Ld{57D%X#zlyERDVr*pI@{hAoXJo^z#a2mTu}@3071>&
zgUjO-0=C7XB_6-d8}o*P0kK_fbJNbrSKM`p=42Mz`QZm6Mko5WM|4;_KA3(wlx@hd
zM;YLt9!Yn}OSoS#?XAiq<9BRoO<7Gl47<xs1v_)L`4ars3j$P;{ehJHv-f<wnQ;WN
z^U8x5HbOpv=2}}MA|(?eo|$I3kcYGWDT{o!wBnP{OyqZRGm+v`@!8a#Cu;AwZYn+S
zA=|u1bgDGlxQuY;nc8l}d>nHg?ZFHurmb79q%d4Ts<icQF%=qon?}`@)$c{DMBs;a
zJrm)jKbaHkxzYH|wIexwr6~jRM5Y}>o_S&L+PuZQj!_iHiQt8(#!FgMLO`VJ6N;bh
z44u%2aEWP6C<D=^9OscOJwA`_Z#Ok?107LvIS-8%kP-GpFXp*151=|zsbQ0KP(cRy
z*$ou#VpNf-k+M;YLn`M@E-N454$;U8DJp#;^=wgh(Hf#xr2#&|GMc-s=aGc3&Smuw
z_pEeG(uuZFvX}{+1&ej_@get{niJz4UXe#khTq|FSqSqY;Y1nu_d>W4RGU$9$&zm_
z5+j=*kKKdg;8+}Ve6Z*f*%oyg<`wjR0OzpnEloXt_oO?pZ)>sL{&Xze>LG{4sfvx+
z^KnUHFKh1;k)=M>y6kG@B2o7$MC-DS`}pkW;Nw-^rEjI!-NUPAD}3=HpU7547TTc!
z*1k?^NS$Mcs0)slEDh%;O9Q1J4PR)OTv8eZzk#uh>z<fKop1>I?Vt)SR7b@NGg`TO
zl4>ZmP2Dg~P+Y0Q6(vupe5ExtOVwP1lDU8$lnKR2P9ac(nmTIzp>BK@_>+k2PZ5=y
z5RCTQi?^(EXSNbMNY1>GFEkQO@ykmFvjrV|_xGN;$;qbY%}|6NFfG!wZM8kp#26(L
zO0BLd>zg{SQAlaut5&=Q3l(|}{T92A3&kD6H(pfcSpY$sd(+T|e&Jb7bsm=X?L8s~
z8c`+?vk&5vZToDa9B%*8R`F{iO~atK=h=VjQ6qoaEA5NpbujOdlCm#)ON@?pe`n|>
znbPMjJmaN5!cOe{TLKN&D56fFRO!{GyKjz~<IucY@fR|S`j=A$DL}izq$G^-gpJLV
z45l_D27a|+67hc6N7?;{Gh#9hK*mXXRfgb#cTQI2uR-CCB4XW`6(u9QmMGc9`W3f7
zI$L}0N5bX~HEFa_J7Kom4?45vLRG%%Ct5P6hnY6^f^lHF%fVhAxZp0KtM$IMWd&jD
zp&w(P4F7cyo`0IK?fc+47-Ft#G|r_FyEy8nf2GG1i<bo5#$?o&>Z)O2Ah4x#!HFrB
zBnh|y$=xqc&^936YNX$aHTqW_mSO+h1Yj~4Vv{&)Md-3LxAC$IPs~^mElmFzv_{5-
z7Td;dPZE>_y~2J+0SQ9j1xUJMR*4x#z7Vjb5_nJ0EKBO!jh6dQ$`+0KZ2c4xMxA!%
zF3RvbUiLAzE3Y>g-gHWz<&jSl05&i~CK<X<QpJEpnq^qH$DhmUmQT&%h;f*@rQGpf
zEzeTjNb1Jtwr3jAvgdPE{?lMeq1)bt&u(;bsmMiTK12-&SAKMp$FPC@Sa7VOaKB^P
z!wg6c$=a74aEt2`U;wK7jLxQRJ~#YE650VlL^qyQ27ULv8`r5FdLJ*Y7xv_JupkQ?
zqv1@*8`%~4xbXNmck3G7km~+oTzwTanN$YMfy2+E`#ijLHh3|dx6n^gQPNAj7%lMp
z9=R`iF!*Y6tphD_8_TQ1R|#YXo?5SCFJ#{RuU=x3rYzfia&+-P%YA5j^)=?+3!Rlm
z4$I1>KU-Xztt&ScJ%SH(os$L(9-aW@1~u;b+cjwM<&*73;x2uE8WTV}81C2-zUgg3
zf;J=2s;1&G_*f5J0Y5AN6upAoCV=yab&Qr>yNTGFogm3Ddhkh!I3UqJ0jw1SqtDR_
z(~C3ybOLDU#(2X}ArxS-Se+hjygkvzrBMP)NTZAM+#BtQvd@aCfM-h#j!vcO3vPho
zEm(%H1Mt<4?zPS<Ja-%B_-F0vS0ZUl;A!|>-(TU+ji1=JK6ce+{p0u7n~$x*F$%Q9
zyTHwwC_h3gfh!1%8gZ(C@0EhV5P8&VAk3H_I5_a%J(YyTyn#wYYZq7T=_<>lu*GdC
zhaN}K%GZR?;MsCiv{q~iXH^5E6;wM*$~uDezKaG3jtgk{f1Ql=rOl)o16FVC&vih(
z!bb7Y4E7$7ge`Yh-s+{<ab8aIuW)SROnQ7z2U|D}Vr3B~cF1km8k(E=F+ap3h5y1E
z3zt&oVyf2j3Owm$7qpe%F|YCs^oyC=dQ%T0c2XaBK$M}CsK)2rfPk5xi<747%9k$2
zV2=Jp=cNC%EL;}CmJOjaZV90pW^No%6v?F!b$Oq#C(Q7Fu|zb-gmQ@PG?4<y%G#Xs
zCaS;@twk|ERgVCdUYeyMs0uJ%n7e}|`5xHH^yD{DMSZ|43aT5{_1suxOpqPW|NbTB
z-Meu`38iV7)XH1v$;mD-sm536zjGAMAf4D}wRlZ<lmTs!$1rlB{KqaY9vfJ{kl6$M
zD`Uzl7$dL;2(u>%YbC$-KQ%P@>sGzTz^~Ov2=O5KW6cqupf}$Zv6$vh`%ZA^u@6o>
z+FD_9&Ed~m^<Q_y|F=Qnz7hJ-w|_xa{XPFQ2Tk<n3uLlB{=Lut|9}7g&GWCjGTh>*
zyLUgYRvY`RQAG134Vm-xTcS_#HaAPkEW)v!oe4QMZ%xo(%6o;&2nMfn5}imdZ&gJj
zO)OkptCptX%WGYe-72#xD=JoVH}hU)DJG#KHack6D1ZP5FF#au`d_}`e|<jm9>^@A
zxx&VO&xbCRixlmx^}F6@l33!rH}>Sm`hQQu|C8^)!zF7yE+#;{o9O>kItY3HyUu$P
z*tb}iV?>qx@BFO(wI7S9yn#grijaag^ULTvaG7Yg8g*UxuYOEc+rRa=fZLNG=Ed2O
zGg{|TzVH|D%>o+o<+`31#b2%|&;Czt^#AfRz1G+*P(uPiy*jTLlk6`zUJts;Z~gB)
z&i=8Q=odo#&kIqutLpH#?$Em%{X%r-iY)xi7lN1+kgMqO-0uKOufO5jnQD*r*tK6%
z6#9D)V7U>U5B%A6W&Y;pwa_7AqT&?*e}Cr&PXE|4&UC2v{k_i%$D#pU>1LGu-&?-_
z_shQwI5Jn~OA?zvQaLZ+OtAJOgbJ8hu(z}ep8y<Cmf~@Gf9>_mh=yUv0<ectKb~Ao
zLR(DTd-n2BAS!Zqk%U!*prB#IFl?iR#(5&=l`!)^(AQ(^cjj_)vqg<ZR{*WNZ}ca2
z23C3wy6}VOj0eeP>46gnzV2ma;ED05QoH=h^XkYmwAQuZBdMZ>l0&nVG$x(LaoQZQ
z(0ky0;V_`U_kwD%AfE$(Z|4wgjhP>^3s*OuEqeS&SbONK@m%>o7C*HW?++{nSN{F2
zzV97sVJ{PD-POkI+Nt${{`g^M`)Nfmgg8+D+O6~39$HEILG^yW&Qgc-0}BOSak4Ew
z*?75el?N#^=V*z_qq#5%zZBI{F`lpE3L-&yZf)TV)<81e0bo~?U%RXy{vGa$K1+sC
zLZ-k(%_cB<`KIyJwuPc~1DszRsW8O1>#%fsbE6)$c>3TFuJU*vM!iz}&WhRu9U<$2
zcIND;h0i^Nw4x*UpEb!jiU(g>qs;|{zZm`_0?|B4jd6O&No4iz>hc1f5Tr6nz=!$`
zCbPvoIYNhjT%f_~By1^Z69CNvSufD!`eNgQO91^1oQ^n4C+s@Q0mEs4{o9*p|0%_z
zK$hh90Cj3;kB+#^g-lM^_h=9E2Th1$>%b3GodI&IES>fkXV)eNQ`KDK-MpO~8<FkI
zc6Q(}!*A<&_~n%XkMB5WUKJ>n)c08<2LvQ@IQ|vK!G9h~{sy?v5%}L#4ws@$W~{;Q
zw4smgi`_34uR2}JSgS#<JG!CQp~*3U=xcxF$>|!h64?C>gX@QHg8Om;lk72Wz)$D~
zj#qw~Zutdpbghze3<vLC4KRvWW4n{ROx+Sls=)0{1w;g^Jg9t{<4r!dE0uH^VmcOX
zu(o(2gfcp1TE##apKL)_;@{ydHS79580rQlTyuXER)J5|0K>BJHIrwe$G?57fem7}
za1x9l>}fW#ros&v81ePYT@cmN&8&@CAUAlpyhPT2`lNWD^viRVlQ79f&4<L_gG$bm
zD2nbp0qd|-4p?8&cl{x*rz=^p)Wd(J?rug&5m%rf3u{aVwrH7_iz**~+gc8OQ?Xv|
zO(XJc1Hy^x(M%{jZE&uFSB^g4`U#at0>ZbOUlXj^T$Z&ha*L~mYr!RrEy2M!t8<a=
zT27(jXo17&b%K3l+3XEDcv^trR|_@0PYoaNIhVfzx60D1%QLqvLd~7A87NK?w&?gl
zvvThFTkCS;2wr;xPs@9>NruSkw+vxg@}RyPiY|BExWR@KaN?scOho&&p;Kk!O(@r>
zZJcQ`tV7W#aPDBpLvYyKhUXXjKSSnc7RhPsj9QzTla>uIHMp4`%=zw7pULU=uhXZ5
zrazj=oeVu`jFR5a!CpHT7yM%jUWq-P%=Ec`q^LX`Cb}C?x!r(22}}s*4}PrkU8kC>
zK<od7=C|5;-_h)StjODbcN3f?k6dT%Z1==N`({bxsAmo`9|fmzXkuj40{@e#6sg(}
z=Qqsdg+B-)zfx=g3pUtf5PROp;0?a6k_?yMli+|$y)b53*fjw}OxMu1A~}pJasS@H
z`*_~8{E&0twl*n82cP#%lekz>GRDP=e_-24C^%Rpy2;lgI;Zx={7ZrS`n7ue9y#d5
zLh-tFG36;J_fG<tvB(KxiNZC7WE?Mo610z~j-JwzJdT$+<LLu*TugfACE)wzxyBf}
zKj5tEz9aMOv9cdf+w`<`p~{%7UWj=+0m7(CfIlOFVFDJhXfIDEDVKLH9x(DlHex--
zd!;D!rSUD7T^-QUB_|T88WabvgknPi<kl0?BKi~9bP$Ln<k^$QuQ{$grs|saN1t`q
zjeNl{f0Grkm+~+-ijn(DJI<rm<^njuuJ@>W^mOrmbzJTBNVqjKVZFEgE#!GPYC0!P
zQ{MBEH-Bb3Gg5IC?Lk;OcwM&D-Cmj2)EDjB;&J~ybho<}teK9tycXiwjzE8C+W1-#
zO+35Oe}blB620_3bl01SFMb`k%Izm#SyWa$Qp<oJzk_GaRu&B-t=Ev!rIBkHqgfem
zMk}jrvMx`TS-!n>;LebL=t}G(#O%+vath?}cPg41i9}Fcz(!7I>BH0khzR=?BKr|5
zb(xC$0v0*>eun%wEeeKWO2QuC^nAL)0-S`lPHMriw~MDF;ZAQvUO@-jRKg4>=2X`+
z$%a+6V(*UwPG0u7LGXhr7Ok;5St;5LR#(?C1M2bAJ+wUfn5%&^Q=&u2Ki?5aToUf#
zp6XkcYEKN}c&{uMpBeag@h@~@Pc?n4M>$BHf|N0B@jP2Kg4W}B26Y~fp_f<s5Xp9Z
zU8=ls2-sOxhutFm6A_lRF^k}!GU3#Il)P1QzdVT7^m$5Cl;l0zE_poz9GHyl-l_ic
z;ijrhvPt_<<Jg0C;BPo3)7rJCy$e#(qT^bL;?ie}y!&i!W?7@3OMjODAIXelz-$h0
zin0v6KQks?3p4#0vzws4s-U^<TzjTHHM$bLkFwXDp%7sD)l@~vEYMjVPM%P${fJIt
zdi@Gmm5S58efy4m)Nxpw^p)t@F+lWg3?WYAD~G~Xf>Rw~|9lGI%q(ezjbeIfpc5W6
zzx7H<thgljRX*iL<8Z&+^T{7po=&cwG(RU_(;bMj7DrW&pE<1}!)Ptm$!HXY+_pin
zh`A&^=<3j>s89C#^j40t_4zMm8)gA>{<TlE5`f?ld#6}GpZ5C2OKMb|*yB>_R{iN1
z?8A!Zd>n|3;i42%kyCK7EP&7axo6tO^K86{xq2uHSLUx{78+PrUFY^mSN{Nm>^Fp$
zTXK?&-JC#GD!e!;`0|nTtsO^wIKzYUopDjBJ2=!9&wpNvK+VaUOzjFl?yfB!R?O>w
z+Ti0aY|qfBZrK<Um&QREER&NB_={}qqqJ`85fsiKB5(vIv|{Y-tJ+Dty&Pt-(wPhD
zj?ZeqzGtb62R$)vXQKwYc593hLpG}f@wC1y7YT}wuPy<D<zD{ifNzBW7cB5WbN|r6
zTZ_dw%Uj7aG)^N7`S&z}ICGmJ+*;h0BR`vB-tDaH(3matb~%o9Pecp;2hQ}~#RdUQ
zKthPUZOM{cL(dyKzIv+nDP2vn_{-{(m5}4430;5n(qD37J(6O^TZ=VGr)yVykacDn
zg;TLQ^BXP3?ps!1)(qHvJ(s(GPBe$KrK-QYCC{M}Ffc9>T~n!PH41QyM(?xc1u5KC
z>|<By<PFv%6Y3FBl2g%3K+fT^NYu0uzoYwa^tH;+Yk}icBAQ12aY)Plg-s?ozUBJd
z^?7tj=}ioM+t^FL_=aha>$Lbm;nW4>U*5f;8(4MM$hCg{dFqaY?C?KN9fn!*<S}^Y
zhos1|mG>LP!zL|Bz>$IUE#qC9>Xqp<tI%<$+jJj+V8UG}ZNCm0!?lByKCpiwd=EqP
zR;$7+{B1*eZtxBWv@9hK4h&cDF38X1-CD3F?P!99<<EMYzz2m%+u|erFRip|i%qOS
z)##E!vgT{zigFkR{qPhBo*uoQ0!25dE6Nqe<12Unfo(zU7W@l;0YQ#wW@G-&1zrF)
zJvHAN`eaYH5B4M2CMv8H9ql}60w&ey4n&~`QPm0P4d9$Y#Oy)5y<G+Z(w%*~WnXq*
z?n}xw+EKKm1Dp0x#uB%g!V=ecH!tK#DVf?Qyq$(akcyD(Td>lM32H;57K3)ujUABp
z`qsM5S3jh;{VULFKTaJcgs5<HfU;)%2ijy4wrj%t9FBGLcRRzb>2pmHLAg_3-V(D@
z#PJgQpecwhlX;8Fq|##50W8H{7;@fa<MC>?o}D-VD$#7kb5;)t-@gI^^2~syd?<hO
z=+2gBROKt&+i~?lAGq@dYQUlG>4~OCfcLc+gU1*n4`x96IO@UR>n@b42K1eu@iZYk
zc9s6txdTp1dFO%uQDcj`RX5OwO{$<gD2IQ+U$r$XQlB#=?#=|y?0;D_9HF>@l+LGG
z%NWYah$wKj_zS5wZDEkIHohZONIwU$V|L))*#ZG%+#F`K-<Y=X+@_EKa{n$Z?SSn`
zH89hS<^JYCzvc9BVhzMg`LBQmlp?2=C(mU}J!{VXXXXB^UoImz&h-SZ!MPa##BEF5
z(?mcKJ&9slD53i0C)~GWs7cSOi*D@kPxjPhpzb>XVAv{JD%Tn9)Cm$=<=tcoNCg`I
zQHK#DDET|!04Z(;bJc8n3Mj8BJhc~~UqMBe^ypGcSfp5>GE%P2z-s~wgf1o@4k*l_
z-i8*&b6A6-DO)+>7<m7wx9_~K!wE?$-#I)frMPeX=^3*v^TP_XS?CL~gF%KjeNJh<
zWOiU|!2xBGR0f<*tCxW&(Vs`_Am9!B2ZrKF^qw)Guu>~`JJ^Fm$L+*xsyT$sB+Qsz
zJYKduG&+4f<1A)HV)Y%i<%q7n^C7W(AC72(dALa+^s^B5f~Wa*H-Se`_g+TC;YbYq
z!G`5C11G&i+ZJ1J&vcPA6Tt9Qdr&q3S86)>D>6H)YQ-Y^LSW%j;k|g?j8RAk70AIi
zmcGRTbMm5X5kQV=xIRg3kD6hvy#L8v^sUs$C?wox16c8{Gk-QqKmL>sPmr9keBan9
zYT~u9z%h?+MBDEHAb%pv)B#<qJ{zCDT<iw=5J$k1>h>3XdA~<*hf3ZPbhoj!iE5_B
zR2&09NCWG<(Iy_%e*2(?ws7`6L3y@rt;d4m+1d+=r3LYMem=<qUMgNPW648&Z{XSL
zvD#_E<lh-w;%(Dczbwbp<H4R1!D=B>IoJby#plqzucd%rMwT>(+w-f=8%sXa9+Ip%
z{TT$p5jZfpLL*zY^1Lb6o`iNF`(W{7plT}tX9q0D$<ih8ku9CmUYcA=noUZvu2`7>
z2F~j(z#KBY6O_uT(x*@Vv+mw2hD{{4uj~vu8>VD<<>BHq7Y*;yz-Fk1-!&R?_*UGL
z<g6=^6t=&mxP~0?L}rPb#2b^zT`{u{_UkTHs2z>2Y-QK#XlF&L`cqpRT?)fcg<BJ`
zr)$fbL%N>uBk*v1`ixgRE&`Wzp`IZ`C)#`v2V(FSPu~M;kUmr2glcd;=vMoHFA-c~
z0?|kxe^MJf#b;6F4BS%f7q9(f4S6oH`%bQKQ{V3x&BEQw0@ilANcFCutiO(>=m6Av
z1Jg<$QU+@5LO7pRp0s#(FQLsaaMR}I*2h2j5<TwFs;U&(ENjdKuGKPFOAH?Ouq6Ar
zx7;^@W!OH}5WxiRw-$O)ta^t`AbX-?qZgpBUA_${r{%z{Y(VUEIb##pF70CZYALmT
z#&Gzk(Gf=kwm*fafX$0dUvi!86X}4KvOZCMSccTDL1zc}659M2C_fQypzH3EN4LJU
zLbk)DiEKSD_fq8bEWv4e6D$XZP`9`27-N*Yz}Z+eur??9;z7-=mG$B-N6qmpk&HZE
zS5YhLd7Sb#SXIHh>LUhtPJns+=a3q^2aLoHW&Mi!*mHh<5CNAsi9U|c3BH_a2zfX0
z-Bi+BR2Ot);_oLM8J2>CbH=NmYQkrtt&GqoC|doI?TYmPk=2i3{D0Vc>!_-`ZVhx3
zQqrJ;v`9A_$xRDL2q;QQx0Hmyrje5Fl9Uhw1nJmxOC#MKD&2k8*7y5f&v(xG?z#8=
zamO9^jNu;$4EEl?wbop7KF>3sDU6YYi$!;YjwZ;+$<2bo-@ym=-e--{jryxgy>j_H
zzb2ci<peg$HZsBh6X;8OfPN%Cm6}k}t7yAzb-eoB2MvGQ!C-!6pcjyHD>rqlzpQnv
z>4q`zm@J*Ns77~Ez}zK$J52CM-*&`xXsEz;{6FQL3rxN6OH`Dbwck&Nt2()Zt91?S
z%DXVxy#8t519`@5G#J;5#(v(e*A+x)5_Q))7Wn-{tG``2|0p;@BpZ|}>C>~dynJ)y
zxp-C!#KP$W7pQeExc<CCAxGhMlv_d<H)iGqw>|q#F<R_Px&98gclzPz+`E#I5h&=f
zeaSx)@R6f?x4z{@=6R2T_MM8z7W6a1?;DP*!=nU8h2<Nim(>Nusu~&bb-ZqG&77ES
zJ$oV9b^7yr;Z@(*hd0W3gN{?17cY0y&1b8)4n_wq3aVD#cuf&UAi>^u5y6ghr4x`M
za!>URV{j#b34q})QLlC!g^}*>dKz!U(ho%XocY*1bxudqK%lxLtu8@gK`O<)A)o%(
z-BBVa6#GBp37@H^U`lvjAuF;PnR3*dHUJ?x>pq{Kp-LQGX=M6>Qq<yKm-RzRcsT*W
zk8-aH7&DM0a2sRaUnn8vQ6w(O&%(1lHC_p4;xVpifsvoM=Rl3y;GKX8O&{}t4_ts@
zNxLn=*1_aF3G6AicFiz+)|ZF}OUP5bg!mT5^Ltug`FFwO;KxSPL>HnRg?O7xX4aYx
zqA;mbLn~^C;==mxJ66J+B(nf#_QchyL?RTYjhEky@-gu!!ifQ34knMLwzCJyEkGTi
zd{ftHAcjBse1O^^qQ26`u3VIw-&)hCgoOm^#KMvdr%<$xk>I^L-%CsdbGM*Fx2gtq
zG4#r8KZI6(UbxsNeu#|mt0uCd45LYT)&2}`zH*$PTbW?)-Bypgr<Q-h=Qrht>jdH@
z6R@m#QVal_o4MwC^rAH1uWPaPvYmCvy%meju)gUsXsw$huxbJ}A!pUCBN0F`dTLn@
z${5K8b{2PWrjT!a{PdZX)$!P{Eua(3t60mB9)#MF1;an?d-Hc@Mcr(-kqW90Tg^`2
z1z5#_o*AWa7!S_+HCr9K+Ls*)hLb-_W<$(#qN*@lG$-!H3DwG|_u;@t+-H=Y7l6o3
zAviGmMe7oKme;iyu&PuspuUG9%b)9AS2<dHDK46lf9JOMGtaQb?uv0y++j)w!+n!?
zE8vbQ){pgun||+KQr`Xt88zdVrj;%QlOSZPeRliaj7ym^lD0JN68b~t;$UH?$U!(Q
zNsIohc3PjQ4}O?-_SoAK+{NstBPrkFK1NEAc|ZJdeOD8;QVVQck6J}Y39dsm?A)wV
zSLX)DU7eRH<|1!iLhIt~#ri<9n{V+K+WIS_;h^&wjn%`Jgy3QLI2GHl%-v~Z#<DAC
z*OaFwKVLv_*!<`lwt@({ZqK!t>631ncbPCa@k>_%A{=M6lAw)bg@ZfwP-l-aHtK30
zf&eBV*0Q=rP<!IN^Ht1FV_&YH=H!s9%3z@+7a}{-4!j;Td84k`8@r}mG>N#IU67*p
zA^XzEr&XOGAq0$C-|M+=+$(%i0}x2!J=CWb0v6ugi@HXNu8hp|)+;WI8M2di!k%(h
z>VR7<WV6%Ll0v9m-`X`v&OK}a+xWb{E|4S8mdt6qxSYY}o;`BvjQ*u;R|<AvQ==OO
z>4P@29L5#X!klLvsOLQ726<`D3gFMnQ+`cN(20QUb~NQ+E>V*P&Xc=>;p@aYoY}M`
z1~5L84nlo>LsY4`ug+jFJjYO8&m5;xj72%8>KIQWecdgP`4Vz==r6~4X<_*E0gEjo
z>X_@6Mo-}2kHkEAHO~;zbnJzY_Lu!?mwt_8&si&u+7djLm7Slf7h%V*1a=CeQLfo0
zn%A!O<ctXSoYmouy&3JwGVer$xJ&CKc>)|Xd8-coK*d_ibk>eG{SWjQ+*E{_S5o3p
zb)247IuG->gJGNBiX!!@wcl3|%s!wAhVf0T+Ln836`b#!fbA7~<o1J}ART@<&Fpvj
zqwI&%+^~{G(53K6AIJi3o&HeTe5m4PMr>O~*rxx@Mw%MK2fkL);az~?vxb5cv9zfF
zWFCxG<#&3mWQ%{=^ZePaEx76vZB_bX+5QG|B~*wkD9|MQ1<g3$U3=_htaK~rt#r_7
z<HUz|Z&;BUk@F93xR5Hw4LmHg`Y~uQ$F)=?;q@Gulapqe%c!K3bt(1j9BJwWShi2q
z&*Cl|`=o-(iRYhgv|{gk!|=gEtnXW6*GXw}hlI0k<6CUjUtb-MMlm45zxc(Mer8bZ
zg%EC)cmqWLBGZQh^U2Vq!Ep<45=Sj#PlQd}*`%4ME-G8LRM<?`a93(7FSY6vtr7@y
zV15NaESjyS0!u*!9KMV34|~dM8LqK8(iT8M_F@hpWHAk08kCWq2p>blYhN0NV2*Je
z?y1nlg<(N-(hD+zfV~%uKFs*#cV9G?==~J61{OW+&u)FzH;sR~pcr~cQ4kvSnuF_^
zL;V;S%q)I+Y%EW?n*hvr!YqZt5bYt7FtR|0FRh;(*-1upymY=-Gm8!?Q9xEsey$#b
zp<nh`Q&6eSrT>mYAJo1^t$Ny=?0)Qv=E!(P9Mt(}KYp!?`R_IG!E(Cg3z*P*hP<~t
zYYOWxCvJgF8+pd}6H2}ba`FOfi2~2#C{0n%9J80WJ%?P<>7Q&w%IcM?zJG{>xvmVm
zsCm~qrzw}MRwV3g4C*r#g2+=-Ti;7&q}l0bC-O;?TUnE%^ecp0#Zw(5XFB`>z_Cn&
zSjTN9vI0O1<??Q=c#Kt`^KJ7U9D#?>9*zjyAIGN=Lp;#%6Qi!p|6>>Lqx+WgKyr#)
zqYUM%N}EuPYcq#vsTG%qG3f&dzw<y^tn^Q*Q|!PI_*nGNh5ALk7kIR!y->0=Wz51?
zxC-rb8@oc!V)-E^*mW;Bt?fgpF%ur*u8!R~Zu<{89kV$wy)w=Y2hP4e%;$^C4|%Mc
z#;BW59b|prr7~ss)JwzEYF(}+p_A=JecdB~HV``jm{BYed`Wif6(UKHHaeNSrRUnn
z<xQ~=Xj5xrSD>OrkA}nENxFNZoXpt^lyd=-M&102VbFw@&_usBT4r{71>v7GAa9w0
z6h}pE6yaiEM0c3b4MCF{?uC!YRO<hHD-<VW19@ExO%1KEL~8)}Uin?>8K#9OGZ^5o
z;C()H%C*bw?vANUhl|y;)3A&Sr^_mR`o0IucF;*P4AR|vkYf!!&U?|dSLaHkA})t0
zv9AQCUJ45)<T~x67Joo3_W}8e3%pmmyp_<geJ`y5A+N9`g)6(As;BtAjDTkFV_xJ?
ze#BdFA$2&G+2kFLvS^4NL}T_LPHjOFwe=xiCQohSq+r>jW`DHPD&2F{cB^m}xF^ZX
zx;WL8`GavaD<Aqknm{V&11a37_zy$?<9!EX?WGnK-YN^dY}os2a&*d{CI*`DtT7kQ
zO2vZYM{)j%NZjcsL7(^vi9&G;yD`{A1QoHZzi{p|27HH?ke->y!W}VxZ9K@jDo0TR
zD#MW=NTH=42XM@255gtsqzvZB3w@V+ZxMf?g6-!l?uY<fu67Kn^13g^0|`suc$0DY
zJ6d;2Q<^}?zV+83aY-jGUk3~y5T^BB1!<cpWZgB=(jcX`pY;7ADtiZ5cxVu1A||r5
zF$i{_M@pkm1CBF`H*h!I;VTPWNCuG`*zb(yX)hD>siB3tH*Hl-nF>*^ndCY(e9~oh
z9NT60_kj^b1aWksFLEq^JmdjDmPKg}03Ea21vTU*lM?mAe)jEwuD|=29xHC&(;n@*
zW8m}>X}5}ZlLxEyHL(z@4Xvi#$tRYKJ+H$O($#Pny^QMZ^8rpY960oYdLD{K`Vq*<
zWE5w#NRRC+LBgc4uub_#AO@e11|`A6q<9dZCsDL6K|UI90bB6-K4;Es1v0;a3lNOv
zRFv({FIqj{pkE$9-WBAlHG^Krl~*e`kemXcV8C3%J8goD>f^eTY`e8bq3-FAINH8C
z$BLs-uFAPrSzBjAYAU+lr27FThd9uR$A!R|2C1)Ji8%s=hnM`^8;SeT+T#(j6+1JR
zU>@cnkiUgx|G6u*Witx^9K%q;|4NnAZdBWfV=Md`fd`fssFb9lCzrsYCle9ZbmJch
zf`g?%Tk_06kft0k6Dg7HG45ZO(q4$A`;rsqw?YUBf|duH)zpp^xzyTu8+M5Gc_Fw%
zWjB!6)g~jj7q~Ypahu_YJX83XjNvZe=59zV58oW-b&b++r{vXF4({1;pex)>P!#t>
z#G~POA2-t|m?Vrr2&)3Lnk4mtD_p}{nzsQj8e0@6CyYGAW#L0I9(^zjk_VBefQ7MF
z{2I2}F34>lo8BjWb#Agf*VqYkGlUtRt+)fO|6WR~r>Pq5vLQIrbCJ&EsvRlvWx{xs
zyNIpZR{4*1f2M=VVqV=VRMPRBv!suA`;YTMi#ja_v0LqM2gY&^uZUzRKZEj^xGJhr
z=`<w#+M?CMh4$&T{kitpYxsUu%7%X1$>-A94;L@LKC$v76ae@B{`jzd9=~i*-H?9X
z&|d*v0b*ZYKO*K0tl#%W%%6g$0iAWwH>mgIumR<Q@wCQGnr(7GkgIfnuFz0Xq~KTD
z`1{C;N@i~0LhJ*=2r1CBi(31`Q6@n}54>}EA5aB>Mbig2LQH8@J1Drxm{yMn0L>6n
z$&S`ddkVyT3cb<zfo|)np(qAI3}{X4=_p+nU*eEF%k_#l6a$n1=U?K*PuN4xAz0|7
ze4*^c56o%;vIPDAK9VzCEMIm4H941_&N!^+MJCqJ>^6g^o*J>~z50Q48bAuXOp@qY
zp{B;cvHVe=%(;D)1RmeApWh=>qOAoMK+wJ9b`dktw==sMYGSJwKUYt}hMF1JKI1|Z
z6My#ZwrO_F;d_Q`+Xij`OkU4lRPk+}PthM2*J}wEB7>JnbJORbsD6N??y~B^D8Rdh
zWvR3RZ8`}e`8S`hH@903O~|p1Vt$sJowmN>?RCk$aDWGxiX8ptBldiqJ|bdzOb;1d
zJ=5L7o(m^Qu=Tkgns{H#`4MoQmhW8-6BqAa^XT(Bs+k>jfYaf}npyf`!GDyn0_&c;
z9w3%6#wrV2K01*;3;Z0wGoda039KHCt03M9?j%4uxgxTDVdcnF322?jeaoScoE%Wl
zYXtlFXO=c%P~5#ZJy0}DdpUvx;Mre2{)93I7}mICFoxpUqd01D;8I@HT?v#HhPwbQ
zGEWx)!{rqQKLFF?XLaRcMxjff24g!5S|xx(*d;%s{epWA(7`p1jzRt0Z;)=r$RILN
zLQ2o(5r-e<^5#=J_n%PPtFJ6y7IUqor;0cdgWM)(qV&pa><E|}jKKTo+z+{xj*Vb1
z1jc*v-3955=wy74Z}HS1&s<?@zB538+A2O(>;H<BU7{}GVrOvHFQ1y;6Tj63x1-tP
z0%f!W*j88OCell8e_GG_OKXf>{l&;E{_;J=>HwF-4Io8Hd+>qsxwOvTT*(?WDlp?T
zdP{QzAVvOh;KZVMsd-y40k2t%dNv4;LY+vK!}D0sSNrV2@~POtzddl60QSup=yaT(
z(H945+0eQTrI`YTC14Uye^b4iHty`5z{kFiW{}rAbjGdA38*t5gF63CQ?_mKQsnk3
zT^HIMs4s8H4g9*bekXaQVG=sGA-u8v3^=c{?$q{Zn^mZzH8_+!;RL=N#2s4Hkj^du
zYb4CQ5>Dgt0>YMchV>yM$T2XaHGkJ0({X?fhvWw(EQBEOu2tgqqYy&l-F)XzW3Kx1
zV$5g_ohMy$l6o8V?;B;~#^CV+OK&I2D=<krXk=wvPf`FN(&dJB|BEvqgl~qZ^%<bv
zy7%~2K+3o>wI6oQ5x5Xc#r6)qsvjiOzu6VqG8b3LE~%uxJ)VeS;6)8d=IOmpgYXHm
z6HwItHfm7x6U^$C$G|016%67z0=?lYXDs1EX`g>9bN||@`lzn2WEt;og&w_g@b2)J
zbPlUPqC2fk!-GMQphF-sPVQGBow}uPmNucEr&KnAX}>p@ESI5s@$$K`6VR-Iq6?fP
zDm28gJMa*Tjr4lK`g0lhThqsBnI$u58Tej1Pum3{5I7^gQ&QCfH)|*vI<;V^p{c<4
zY;ELH%+fTsMzt5Ux<3A+#&>)5YbB(Ofb#%5Q4C=MW4fzReU8p;Wl%R&M)YbE(^+vJ
z12~$!6Rm`WXbxBgWyd%!ZGl2a*_;C$9K!4-&?iuGw`baUvfys50wxe92S@ZGDG*TK
zft86$4wPJvn#pRQTl(i^^}Ljb!C7w^5B|P<qX*?0BEaw*DLY~Wf+;0e>T-ZI6$kn-
zAHPEVj|ax?czZs%_=fD`1An(J{T>MSmbpHL4`&qs7(+ht*)a}(=bM6UoeeTI^%JOI
z&DVqv(45RunKlgp-FrMckn49!Vp7E+0838CHi6tb|M=&G?Th5GJAqa|5va5TI|2Hg
z1Uvg#VtU1mi|b&F(Dm49y`G*^QZxh>tnkzSaKRosiVo|8i4KkZB5OZn#U*py;bahM
zaL?8~d>c-?RM7Xa6%uuZPYn|^TtBYVy?(YPOXgib>Xt<nzSZ{?kIvih3@|XQkN!jz
zIq3U;fI2cuYaU6$QI_h0zzw}N1yVs|=#<}et;K$&F}9!}LWOKyh;aqgMrvdrkpbXg
z6nht)g{hu{YNeTbIZiDhM^2Btv=3)#zHC&FO_&R<f&KAscx75HQxB~Hsl3Pue~XuD
z6vBdoh5|49Rb>oN2feYmADpHEm!|$`<L)SdB16)3Uld5$1FQkFlGq6hmotFkgdZFF
z+E1w^A@@FjcHL<zK;+}J$l=6EE>6pD^ar;af5d`mqhP(EI6pd@Mtz44zyswUxy5@8
z>0D}uweETx?(qwMYRQ_YLWn~6A8#%3*fEw3S+Z@0I$VI3i$e_W`jZ~Jb7uf1?JHK(
zIFA{2K9Zw~PX$0#ryqe>(m99~!yn~V{}k;AM?q%7*S6R}qz9hV`B_t7y^|Fgc-POt
z71>+eLh}if5yap)KKZPo9$ANkw*Fa3{4<~AI<kIWzll?wd@!OD$=bzHs_31Dx;$>a
zb$EFSEQ$4987;1BmeH^MU=^k8xeck?A5`>OpgV#Ar0;sgSGA0%pHP5zuIn~i1?l#d
z_m#Uic4buOkg`f?_6`S+Ag-H2IcW0H>QTJWzW!&PdHWa93=q+nTd#^?D~|}x<ma@7
zZ^o&r12vmy^0G-kTLi`7&R8NoyVb4UpbgHNpD_^$w1R<qhM`4dQX~0aPk<d*-zB#F
z*o;FL2g7<4b=qe^aGTs@*kOjYMXmevcjc>)^6mU2B8R+_U$6L8V=$rMic3wB+EoAu
zBL>y@ZsY=Y#I3_-_TCqdsBIpN%%P-vexoZqOmYIr%p=C9IPkcN$EQ10$lARG|NV8?
zo@oFgE(_2rDfKQ1=7ywdFcq?kfYq!wD085B<8j9(eI#rlJkGTj3zU-&UI3?FmL>Y7
zRgjTj<z3w53Y>q2s0EWjLl$hR*WV@YM2VM>z9V~faZu=epcv-wUyPx~8ee~N*)hG(
zOYc`rsYBXg@7Z!De!eZ;nzQ#;aY@SAcjg5sJ#3P=?L*Xd;QuHL2P9N3-{)h}^_)M#
zz%7*hMb|B9Y8Xh>Ki;$j#akx`2T3fkg=Hh!rcB{O6F~&z??IVq{VQDmb94M%W=49k
zx;LR2fE<YB<?jkJyGFsi-W~Mx$ce*D50M6t-Q_+!J<olI`;0_)fo4mq@~(>K2*@gV
z_a3CQ=PU~vV38L_5dKI>;tsBmYE%05Hg96!xK7X}c&}Zs<c|z)pVU_m3^E2TCCG`*
zoI2_-?Uq-AfWw1gTFHZyON=>$;xq8k=P|lJAPOW*-tlC&IIGvPWCGo>LZRK<(vRLu
zWk~K4COz%ky*{sJeR2V=z5r(Iqp#l^h>k$y;b<`qk`zri2>E85A-Mw7iUM#z$Xr|?
zEk%|<nyri!TVA!tc%fs=n7^@n7xN>TD*Jbi-LH!d27X7zY>C>mfl5X9vFb!3P+?Zy
zy##GhiBmxID6^Ad6Z*g!<VY-4qaf|laiZ}87)I;}=%0G#pyHa@Qb@YXY?r&g=;*qx
zP_Z*F@U<-8r%~@UNjXxcBR`ZckG)euhzm*n5IHD7#qAMV3jxyE)Bt9YjZ!svYE`jf
zr+q+AC#hB-NvlNZ%J&1+SOP7Yd~x@zAFbW-`_kPBgt|%R>nh%?!O~O6!Mk@CD<Z{C
zEgn@#xv{6X0kH-nGvQ8Y7z(4qO$-fo8VAsUbZhWK#kp%J{t>ng2LD3TYDvZ@lz6pp
zz;kI>b!9)n?bAOYjW~|Jv*K<7#WVB>M2!+_Qz(-s3hCE#9-`M<n0~eYu`-f@OhaRv
zqp@efLw<wPp3^`p-8Q+?Mp5`agM>-libm|2nhMQ_aWR*Yi>|0u&~>!&3vEZjp$F4L
z`ZO0$)W5BVMT{RKvu_qju=DEr_2uesdG~4cWNQoV_e+O<#vbMR(n4&?bpZlFZ7&Mi
z!?veYfgHw#bW0!l+<+Z6?_P&zNf>2vQA+MCqZr%Nc37UYN?G=fXQ%mG&moXlzE-jN
z1CG&F$ZW@hkQ=n-UIM-BVgM~h$(q4pXMp6iAh+u4icng1q<m$^vOaQPGB3mH4SM7&
zI$MM&5OE@|TRiIuY9O}93ZSD-M@6Q0FNER1CLTwM(|)_ox4l--1MD+-^;PTedm|6X
zEb12{L}I{6sp`m9uB6dPfXXuCp*b%R_(hDN0gU8l0+8nNlk-_bWqA6Dc<g!uJ=F=j
zJqOyUrQiL}$ieBcAQ`^XTYdQZG7ye~K*f0wO@gpeqMe)4FJQ%ZRKk2fOzaYaREKEU
zlnpB^Eb!b6cvQV%^Ns^`9hr(6Xp*f4Us%3NW`u)CqPUp|V0#pes^o$W{AI+C=vlKa
z#oNQEu{f$xhEW+S#9px~W3300lWF&2$&(q(f+^#cmwkh;=g4DW_)HO;ECRZhgY0Lt
zIPOW)UH>m_x7mV4`4Wv4g}aTAb$Be0m(0%QC0B1_kJXC<1c)J(+g}Pp7-S^1S!5W#
zx`<X1XnHJ2<X2*|y}&4~45e`#dWi(JWiPTzI`imFKq9?Cio&tRS<b9Y<7viUt|Xgt
zc2ekxNTo|P!C_}`F?u@vaBFzr@yZjowjE7gkK^quY=NEH-H%8>PCE}idxWgo;1MfH
z#`ssF=8R%`+Q+~|y$5zgA0c?B3SeX*>C;Jte?8ORYGdIbh%6Z>Vm%tqXu7C*iTp3s
zbA4{Dvp2u#H_!48ZqF4QnhaW<a65h7vtPBhrr&1TGqJUQ=?3EGt-oM8JNEpWY04ud
zIzPSu2~q4S?5x)_w_aIaoXl@eRIcwSXC|6y7OnBC`AHf(a*g|TVO}dpP%8a!?m<fP
zck{KVf$M=*WPUjhL!?Vi+1s4NL9)CArdF^{F->t<Cm$=O7ENFstx~ip*wQJx)qvai
zEI4LO-MyG9e7uOpT^^Z0Tr4d*QC{5qRb~Ey<<*Hm+kiyEmZ{H6a5=LO-${Py0KDc;
zBz1-F+Bf*%OGZdel)6w(NkEf2!s3cx>e_3I0%%`c{Tm6Ra$jMyO9vJkftyxQHH(m-
z8QDh3S#>yJ6$?o98Ja~(`$`pd>p<8Y_Cg@nlxchQ40<%>$OP)$LV=EEHf8lX#?%*0
zCV-3QdW_&siEUit#>#j})xNFl*o4hQ`qi(|Xv6`pWn;F!{?@xs9DZb9a|(XlX%g<l
z2JiTB2yr8ul)CesQ6sIH@0SC8&7H^!vPXu3l^}s(0(vJ-Uj9(UDGMEc0ZBl-kr&6R
zATkXbT4SJee3R71_kpRwIYZ+%Tj(_#o>+7Xrmg_lE*;L=W5Ko)U^yI;5v~8hCQmnL
zI@vsZeHj&TW(2uUk^YoZ)2JoL&=3X{h?p$Y7tdg{Y{l?_iFD9RT~Ef0>}p&@)1%WA
zd+j8hiOZZ1nkK|Z2#}0dAzybkXC>2t?~YNW971NkQ#5+qXxaRAGc;l<z~d&WsbqQ!
z*1eXliN}Yas2nB853m{^(Y{1ZAMh&T_%VHV0}wp|!KPo#GA<Aq(Usi?Kh<n!B<qBb
zIh9r$5_U^$=W%YG<D=Sr@50!&bTR|>dcpfphjg9d^ywG4Tgx7^aI+hKi{~ZKg%crc
z^+hIjQF~(xC^*e*j3>Ehu|vsmt@l6YW5emFPFdMIL*?#D(R;dn29?46XW*+%#A)JH
zm*k=M*MaqmmkR2No^y?#w5H`VljLSOkU&=9!#t9)(Hfh)OW~J*(NEqY!{{4!;)c(}
zmSTxc>zh-?@qFFpFpsfxqm1_e9-WR}=kdX>v?8RM4qP8j0RRB;c7iU5azSWt1xY^z
zxcgU<5x<j$p5{l2UkXZcm5%DUKOOouFBmA;(rUqnUMH^foZv*!j-1}~z(4)DoOIoS
z&6F37>zwZn%P-l;?@~yj3p=clO6kG(XafZ-bK9$~nJ%*g!M!)D)={N|?PlqMjv5=?
z5BifE_2vad7}<p>oSgZyv)5d!X!uosT0^Fi=hGsMO1e{7gj3;%jw9I4V2Vf;H2YdR
zaIUx*Nq+P;1<2~Bki2$xSuMS{RwKq-8VCKV(y<f2ao4!DqB&Y-Z`jRgsp}j|-HJg_
zdR%j~zRNybNGeC#s#2!&&lHKxts8^o$%#l=l0$i1QAx7kWuMum;GV)_A&(4@m<ftK
z+^b&Xf|>?!DL$aL0^CKrFx0yH@5T~|IgsVNToEc?Nqr+(4fSK5-5fX^XIGg`p9p(U
zug~%pXk;eHp_`xL^0PVk>DkN`Ql$`P6UI6>N0Ju`ve+R(K$<;H1=P*~?(bY~v)BoT
z=g;yhj_Lyai&ABWu&zTlX<{!f3f0CX$VSv0#Kek#idzs$>9lgfCibk@4(ca*1f<;H
zS!E<Lb?c;LcXx>LXOBGO8@R7$tL4ziaB<NSYeGl0Ya7dqz%`BKL4^6|2c|Zc-1iF3
z(9nHh3{`!pp$E@}s(9G8ye3M|=n+0V;-gpRLffGnFK+D(OGiw}kXy45av3KV_ayI_
zzr}sW{&Mu0V>q<%TisNL2miNmI=STN+!WBLLbVQfibI*vXo-U5!sQF4E~6gUSw|_j
z9y&gSqo?C79ciO&Ws{o-rw~4IkS{stJ&NFC4RD*GFgbfZ`|Nw%D%)0dCD49QZGFaD
zQD;?N5t_l<+x|X#(2XVhyDj~%d~gJPt>!IBFAo2FCzWG2k4#?O<*8MnetGbgOncWw
zQM_5|Lv+XR!+EO>`qrbuioP^+fHuk@X7K<5tfuIh#G}et9otYiUZ*n!ApP2E<vfbh
zA463IPhxat;eh*e;qT;y#UxJxs%NsexHsYKP*})`Cd_J(x}`BSMKFbooaVV(hMV+O
zB7XSy66=?O=4>8Mh8Ek7q*R`_dFa{QirCwz1JZz|z8DT0N}C6u9$QbzZs*IEFLb|e
z9!ec!{`61S1a$h}hD;En^_p!xWuJVk$gzz1LHaJbd`i`VwWWx714Ts1J%L`NT<bN#
z(`j=95u1*n0qR%eVD`;`Wt&7FEM720YY{W$6GQh)(9RR+4pPetiF<8cOUY+EmA0ED
z(?}^D=64${<96C|P!^@+SCu#e6!8u8X!0_$Q5>U;XU6A|wgoa%O6&1wNug#!X5-?j
zdOR`8zp;%{4qxrM1N#-8(#FbQ{&0SO&xqmM-)%ylTdr0K*d%k9YQ==hP8jd?{zzh1
z<uG^=`N%t0)eg><d!v_*#(+;lMdDK8($GOF*Jt3if`>_x0!jc$EQ#vPH!k8=>|+kJ
zUJIxp&S7}b52JLVc#>M8@Rx$*DVeRJm${i4#1uGix3J(N*VgGYlJf)$i;RnddM*!&
zX^U_4@y60HPo_HduJ5T(<(a|;v3>boSqBWzJCFp1DRlr{_ca6<fQQZSti3#8cXZb)
zA-454{IH6#&%6ZzYm@m5&wxcGx}o+7_B~U#csH|W_Uc?W)T5_@eyf^H$RWpgmo)Gb
z?%D{&8hx^cj~kbo$KSXpp)WtJgm?!+#B>Ecjn^|%uzyt2J3_f8f$+4wXMf&moj!N(
zu}X%>)}^U1*ZF3`rLgG-b((^?ai!UCSi9uKL6Yo_m|LXnTA|uUfDe%c?o2(wSu?24
zt&iBmZx+IxWv(b#nx{%Jz$HVKwOe@#rh}S>a&o5MO;wH)E-<$qMqpMWi{mt<)>%QV
zMj-{Pp$~8|JaNVHH5Q)NTfxqJbwaa-I=QXMa(aaor8lNR7#rH{%l=T672N3u^EC=s
z{Rff6(4YT&ZRvcZC<||QJGEbb$1xBq6di>9hzzUlWF<w0q`(?(XWmuw@;ZL}FU`uI
z7aaI{;+2Hm!DzB;C9u<W-SQTsGF*nJ-)bg%{PqaDh_x4k!g}r{U(L_r(2Pn-3X4N8
z{CKo~ytr?j1hU8bZNgL8e|+oryYh{oZLoHZI|jnBnxh5c1dFcS(}pl-7+efghMS<x
zW|w*pRQmt`KsFdfFc<Q&{2`wE2UU|Mky&x}1BvU3oNqV(>X%vps>wo3O@RCV(B8=<
z5p>a}-g)2(v{z@tY9qQXx-M<x=5hHJqn4(Rz5WUR2*)BjS4(E%tp4R)a^7tq7H0oK
zl>jxuPDVDz7|HXXF5OPJ_67)ADZ4yM_G|vAx&H|nzLpR}pLko}%ew=Lo9Y^n)1Aj`
zGo`y`>3ALto1!PF+!&3=aid@~G1_jw-z|Q*6S;|;`v7n<72S2_;=e(MC8UiDJo<_7
z-&FEnA@i>9d~=XH!zFltKh(5A8FP_;2^O>OU~9IX`2g9J`URJ;<cdIW4k9r8M6C+Q
z^81iNN|~MCx}<+u<NKLHs6Y%s1B?c1fUoEj=setDOfo=4>SnEo4IH0N_74N;?w3h`
zicP;JqNhwJOKuPXEdIT?-xe%?Eh`MG1Hw%ZqVS*gk;5Hsi+)-cAZ<Hc_G||&<Kija
zpmB76<^pueKC9>31{%60v8!`O0ia4c0!n|81S}-)G!4dEtb<<LL^b6&;B@m3Q1<UH
zBxw6NFAe}}(+EIjo6TO4EhnUGGz9>YQa8&s=unXnv>9zt+p1uhf6M?)VBVkPus`i3
z#otTafwZWx<`icA*DY2n1!=Bh`&`*ndwKd}6JQTcNa<wroL&Bxwq2a%opzep)$5BA
zh)WJY7vrNfB6Id|01?;5m`^|XLo4@>cI@gjiJF5^ts4fQ8S?1#m96=#?Xt86^!X4o
z{KBdt<dN53tCIQOb)sjAV&9ct09|fsM*!iiiKdqmHajX=+DbBxSO+rlJM{>HNpkbl
z0c2a)h8EVc!hiK9!hBAD=&HYEjrjX<I$;AVAZK}*k7WP4(f)DN{{8noL%{($Q}_H-
z`tN`5zx%1K6xipwkE2-w|9($@FY$j~^ndw|#Oxigf^|=4S$`Kd|MKJi)Bj2&5&{+_
z@kvg|pOW<d*+)lyr2Y!9?ThS&D-{0dsrc_c>woe6eg<UcKf{d=M*kN+^dFyUp99(X
zuX9i^_$P7kKmS|*?=OEQu#TYd*!DdGN2nS-{{QAW(4v+Rk%0tq8t^%fb(S!U%YYjP
zNS<6kqy1Aqt#79P#k2pPta5)c`svZu7ML>HonTf6-YSwfIRhZ4wl;);^IxOJ|L{dN
zWPyYTWfR~tHkReu&mpeg9tu<;W9<JkI}Q*lSAnrp?El(v{I4Cy|Hth(KurE$H<thZ
zy0QGr^6%c=9@fyW0y$4L3K~YGJY(|RzMOb<8ph1kk7umjz&N5#70~#KM>mui{_cJL
zotL&*=G^{=7r>uo(LcZTrAFp})EF}n74CrAnF4vc_B|`W^hM|ZAtcUD%w2l|*tX9Z
zkTG6-LE0;DfacPGy@}sSjx`;0KoX0&0FhEPptxrXfS||o(>AM37uNoW;{oXOuLLbE
zgLGRhzN_xtJndlw2jdr4pl5g!{5==(i~U%0Zoqw114N*bH4@~pJr}?N1PNn(ysbfW
zpOz8SFEo{j`d(EpEPxY=Bd~GeIv-K)BRNJk1#%BDDuHfA+khMDYbQ!h<5YVV-s!Fj
z*mA=5NkH5?je4D(kmmb<XB}Xmuj&>>FG1?)sUy5uxmXP*V)Uy}CVO9=xbuluC;z+T
z%ZbF7S^#nm-q*AuSM3wtq}!Ru`_LA^D+J=uio&|nk4Q3cMJQ}2LDL9(v{1jkdk;?8
zqPYpiig4Zy$Z`YCjGX7nTe#%BEqE0C-89Oph0nCwsNJo9>5$T%D0*#GE=C|>kr!$Y
z?>!yg-&x}W=2FMi*P~VEka}yqV3iYf-Iq^Y0ACUK?JtqWk`hEZ2U%trSd+v6oj@lN
zASCsw2H@g2Xcbk%UrZJ32D(PuKytI!1k2Ulz+kTmX`Mg?RIi(_=_B3?ZGZ}Lup$<C
zeB__pmiZ}ye%tf&8wUS;mj)Ef%B!>bt37#7!0XI3FN40td1UjVS?pHGkE^|mt4dv$
z>0wX67hl=|`LzQO96TE?Hr#GJS_iB30W6-j^c+}zCqToyw3GK44mkl>z7l+xUZih{
z7cX&OUK;hpTd~_c&Ayn_YsfAUBx3{DS^_$T&He9d2j{B;mz19S1DgOtXNg<RjudkM
zD)(9u;LdWr-OT}Wx)G2<^P6A_Z=s^(yENb><Z+YxW4rr4c-f!fDDJ%a)rtRhU9Th8
z)dLV*h(RITcM+g@*N{}85uL>K9a%KqZNB$+^l`7Jdmd`hthUp5?*|eBDanVc@cCI#
zXUv-cVmHK+&v6&n3p#+VlQ+#t(M6y^xZ3k@oCzu)#trNXZWS)I`PhPA+&M`B9zdSw
zmF$#0?RH*!6apif##I!ym5mPpE1SH&0)G+zRDEX#<q)VAiaYgy_lqqM^<0ydy^~d1
z4}>|jM-mtd33_gpE`ZWbcqq;EyIguhFOgCVu!+k0tBWnMxBZ8P$6nx&krL;5HZ`kn
zcQoH@-c56!=tOo82DrkvQo);>fo#9ej&MMF8T51_<(!*<_;f+)u6jn9jb`ck*Bpo0
z=}dUTZoKL~1za_J9Jj9>sF1u$I$c^1$abayl`yl7xu}l|v4Te?3;N;Gn5=*H8ITNU
zNO}qO%F#v}u+bQi6mqGX_0>?t?FDryY_`2A>PxJ&$dDWq-hf7{H+`%QeH%85(?ise
zebK0SNON^8M@J7%X?-Hz)ox-vX~e-2eyl{rxEY0=69Saw!uxb&5TowU)DR{u+ztRJ
zETnfQY%=`21XNcV$;5&O<BfzNK659z9nS~V%wBA|*yhHo@2c)xRsbql1088BDuVb<
zL=9geIsz^MH9&<ka@E#>-|E1Iw>smo_s|6hc-*DySn^T_fq{yNSFm9Dx=$8`iWiw6
zuR)kx4&YfwC{T$7IIk=2&q>+=4f&y|k?QruMs>_xe?u%i)MC^bUCmy+cR)4#LY=e%
zNH?m5R+6MKe3G4<etA~y()oaqs4uq_0wv+xD~c6qr2-n@xHm$!RlHAQk0RWWZlT}~
z5Y6Z|$$2H973Ry$Uf3SG0n`W^;X>Q_z;)LWHC#+%-RaFm__H<>Y;uTMY|C|u*+gwz
z2-4Tf9j#&Pg^<WoDw}J*NJAZ4z<4<VL#WV613x@%`_P0uI4~uu0_;%*5Dibh72EA-
z2HNF$QNk_=_f@-i<YCZqas{BGMs+ZJly(t-r>mu`0SI@Z>uGZ)rHapgAnyNj9S}~m
zW{=k3#LqJ(jomXS8IF#7Rhiy6mX6Eb1qE}o-Hhw3K3BEr0yM|=m$^n0*fl?qwiXqj
zTj+F|Olzg+nFL)Oh+eLRZmzCU@{zvuZSVqISL=o!bD-vlfu{hgscDMi`Gca=JZbRp
zv%X0paSjV?l!UY8l!s@;sAzdWL06#vS@bad)e$f3n$ePOq()tot%|j!xTO^+L%}qG
zNLlP9L!W`QT9Nr`(9k6E7B-zKXX1_9ZVDNrE}OzXh?b)UzgfM?5_y<c0YvQk!Ac?K
zQoo%_eK%>jfr+kul*8%LCbOw|9I9P4((DWk1NTL&60f|uiFTE>kgqLVt|pg~qwvKv
zVA0}yCW`P2D)Bq`KlO!y_UM_*w`j$m*ndelZP8ClQE1zUL2M5`L-tthma}{A0}1bO
zt_AQR5VqUDnZT$_LV)4J{w?$tNx~~*%Q;thK^RLBJEtEV`%9#QN)?d6)nA@|(gEHA
z7mpErqDQICIwXPN?7<zxLbc=_*A>r;hk!>%PkeGbXky~!eZ3DZ`P1D@*9Nr(aDUDr
zt=YPNy!>sgmg$V@D!M>%23lxtak_!XNr|ye=0xMrK^JgnF;-+4!5_{rApmMHDGv-o
zq^*&h&VvS88snFrS1G%2vrUX??EDlt%Hoh-T{xpbR1068=u(1E%Gj=-E3qm@c0SPC
zZ)>2Gaf7?EaKvk_AOYb44(-ey-3F^-*F%sl)KMY3bW37-H#DKVGPcRbSJ?B9j={+|
z;Kh`E3uMlFvYAN#*x9^2q=;D^!NJ=NdO(*kiMC<n70ZU}>mk@<oq&s<8pZ6RJwv8B
zM{l}IcmTR31UsXnFzM6=$B0g);C&<vms{$R)kdqBNY5X5@t&@}Ih~_wo#ZWMczFz1
z2-iASr8gjbm<Ecig+-Ixhg92j;bc5b6gYdFP{Ed(fO%?5jq3;<I=pmjnDm}kG}F?o
z0<fPS>xj0W(*fyP*HFGc9bTP$+%>J(PH|>?S88~~w#8^j!<qjSO%ssMuEXcUDf?E=
ziMT)eVq;7UBQ!4nCb@~UKr)@Dq5>`ph<sQxF7((XJK@w~Ni^^B_;Apx2z?u%b6e_~
zh5@RjP37KKmvxk7OW7|TrHS;^Wq-kxNN)r(qtT-u%oMZUr_`_vgfz@5UW)#}cfd`v
z>CG1XQE)TRR;cLL&yOL=mP8m{{dl>w=TdSI*dH!|TO&ZvFX|0j-V4uC{ultO(-_`-
z)OBJuQG}@1fR~Eb8bZkGnLxK#ivFcQ7n<Qcp~7evz9F7II^hgLBl7qy0eFs04plZs
zqoDS0>L`;R)nhWc2C2cC;8e{|AVOvdQ^1bqJJP?P0)(21pJED?C0I^fA`=|TYx#M3
z?hak(clVjl_vP~I-t`Nkf?j4l$=N<#FCbTSpVrM4Ss4x-ATU1=!(AZnlQ4g+oVtU;
z$o7OnN8+8K#{$rRZC9%~x#u({p}NsH72>2x&jIn&9zS`Qm5tRW-tx!o<3dm3)If(x
zom8d@1w;&_95e33QrLlL@-av+#7=^bS+sf02w+cP+M9YExO|BreLrl$`#9{PZc}7K
z%2{Uw37Z>`<^@O&)D(z<N{v*}2_CvioAjrl$ih&J490YP8-*~;$9YxtZ6O>$1Pk<<
zGe8JEfJ~DJ!sZ(ZC4*6{639VAf2-Rk!RkU}I^%Q=Wg-x?&W#U5W2fChlos=G*PrMG
z*>T7&dwBEQG@?KqMY=T9U1Y0;p}?qYajh&S#|Ws*>K0RMt3TM0oTRt=GQhMp#n8hX
zfaLRf1&|R1N+;{=2G|SPa0N<|<>rG{vt)Z0kib;*3jZ)kU!?Wh2hp|GcwBN<>CO3o
z8Bb=HLkopX=_4)r!(!}F`z-S0tcb}c;UV)hu_VA2iHGZ(fm4t@(S~3Mj;q0xBhuR9
zoeTnGZ`0GjPUTtv9;LJ}5=qi`j&-hoETP<#NkUJSj(w?`nyJ+I04IQgnjMZ&JotRS
z3J3mN#uJ2OqFoJ3caZ&7BL{GHWwYm9fPm<>h#t1dO=!%rWw3o+1Gj_w{r$lE2}+nt
zstGGAqRVFzR3*8bu8dK-@Ossd`0K7YD=8#1FVnGcvET`=^_p!d*leIrUzYQ}?2?G|
zT#B^VCh&^+7$UJiua*vycD?BKv);3Z+GGK$_$Fp`B5Vbv!~>3a$ewp^{W%r!d62Cg
z=+(~SxL<Y<Y+-#h!7Y9CLiyhCljM^IvgJbWi%&<pJ!R-Kcy%R8WT&7Jc1|ek8T+l<
zo_&CwoNW^66hJjj1j*!=CH|!g!2IpZhQ`DSQCS>Pf!1xZroCEho0~S+eK!w$iD*3u
zPMwxr`L$_)IImKGvWe{0Vm-jyvtQ%d@NRMXt|s!^g9pkU@(o;3=G5GfrD<@5)7Qq_
zSY(|cue<d^3S%6&9Zdt*m1iw<m(lG<tjdK07vR*Ydlc?kc)K-2xDiZ4XtR%Jdf2lU
z6&&=kw=CYfC*;-xO5t2(LMbdm3+Dv1dF>0|p14;xU&=)GqGUnRPNe}CKB|ty+KaJh
z(e@t7@E4FKr8eucp;e>QZAiA+E!W05o1!8Gx?6TUjjPD2gHp@<4EL{FFdXK(W|2%A
zpZ;Rxfwt$tZ70*U79P|VEOxzzd4;@|rd0zynKY8Kt$RS@)u(<B(Upy797Zj=54^^f
z+cA=?BlNH?m}Uvv?)6h7Twl8h$|(F<x%)H{G(^pksKO#nnYZ*kz6fBjCk94T02yl_
zs8iE3io!{l)BT+BV0IlxYyAs81FUW#gi%-Uw&?mrpGC$tTvB9uO-r~2T9Xr>IRpzh
zB#py(xxb|}a>m`%Jb|iqHjw&-<37njlY&5pE$`pd_plewY@}sXwB+##k_cvE7&Lnk
z+v~!B)<B{H8=yGlc`c(yC<x}B-P;<!mg|H_H5TS7IPYl?T{h}J1lc&NI~&Fp`IO{q
zGu0Rr+GFmqz|8h;H;fX?d~S=S)ZaXtgfTkzOqOnAKb>p2vE@BjDz|EggP)`ZY9OPN
z!%m3*{8KEC6fVTJl=urDCJEs+vSmY9Aj%l<@tVhLSYH~Xb`WjDT|dNX772@%%A9VI
z;OS~seBFGrP?}lSVr$HZ*+<ey-Jpm}`h7kGjr6)x6OU$l10*u2RGv)82silcbqw=T
z<AU9vgJDY1Pe02>@FW`o-H;>D)l_DjM9T5OrVU(ukN9yp$kFp`Km2gYEa~(Q0Fic#
z!nvUO`^P%nRdVUijchcs31oOYoo}ZgLN=Emqh1hw64M6G(S|9z@E6w?i!-i*E96Tj
zQ^1h)o7N6SS~yID?ydTe95n_Rop*dWb81vFw|0X83ETn9w%u*3Cd;6+@-!%U1d2|$
z-Gjy&M?f&S6J*4R;KQ!7UxdFhkbxGpUcalrX4ZOVk%#0qf$osdF6Y)mVO0%~_h%gK
z%ZqJO2j2-dN9A(v_&_GGqk9Sjf@^ze%6_W^1i6NpydI9+ZJA-is9Qp^fgawjpQ8(T
z#BzRiY<OgEa|cpJL@TDFBwvAa%)!;HC6?8+^+E7zMztX$3yLW{Y+c&g3aS{L_Dwdv
z2+8kW2t?Wu?>eq|H@vpHF{{1F1DL_tya$Q7$w!pgVlyJRx$p#T^{^ZA^Lk4uoe#BU
zgrrl=D!{!BoVd(mw{G@bYkU7T&5vo8EzSd=HF?eb;fL(_Jwup%(5-%uysy@mOtIm^
z*kB^)h^gHk+zp4<qrU>e+SQKO!g8cD{eFO@j_azSER`-O{jQBO6d9!aBNj~l$5_w`
z2#L-5V6wSfX(6G)WEZ|EwB(x?=qTU{hZf)A<(BpCVCe`{!jX|}YQ~+#{gnb+XrPk4
z*(+#rx5O$fD;39e!aNDbCfx7RAx}a`=|f#*Y=6fCq&bVEi^`RXBsy(B$CgD!Ty}I`
zOdW|Z`MxwxIFW>kpBQE>Pn*I1!yDRi5KjhSXc#BlYyDe%3N3}hfHg%RrN0{bX~vq^
zKuXr9<_9<zOuWP-6ZZJ3oSBnTpxX%XnaGnS)|3Zwe{;djX=3f=;NqR-!EZY=Og=&1
zk-$a$(f-9Z;fa~P`8CQ+E;(ECp8<sG`bM~zOxSd^j^7C8lJ6F^QeTr(Uzcy6zik8f
z@>6CMD@yl!WQQ$b=(l_QJkN<K1g+QHC@PhH`n}Oy)OD!EW(~&s%22*!*eQfRHm4^w
zW3%0vA)taOMq_3%=biFO<cs9HmayCBb`+WqtqlgZ@~*TNW0O>)$1>%O0to}YF&5or
zjAL9W%YDIKYurX1Ozu8(ZXS~CM0~@0GE>zk-H#Zmluy<N>pk9`@KG_svod=1^pS^V
z3Xg;lhHo&-GzrBvhj1ay@8YYBgxe2o79$z8S-(YKL^u=MCYoG_fj>(7K^Xr?!kJRH
zMGo?1(7MD#bvPM${Jyln!cLqSTA;?Y{(@ztbo@{j9+mR>!8he2lJec}^WRmjKUnI*
zC?kQQ%6F&Oo;pd1SVFN$pl%A*5&}G1_ZjSW1^Yp0Xupte#?>W;&M9=i2X;7bV@y8C
zK_QCWNyST)m=-^x<+>%By97o~*h9<%rGr^aiPh>D%`B2M^GeoW;+2Kuz6@qzvCm!(
zKMn3VW)Ja8HEz2R41ciKmrCAAn#OgMJ+u1p#(^P*npmGGT{F7SHyE7p-OmF}19QC?
z59)4XeIpFz%Sx6~Vko0xce5OkSg%^5*<AdXgDUo}8RdI@sJ244tDvM8l)=JdVk_lQ
zN}`GlqoD&J=My~8x-~g?xJZI~c=sJLS=1E<JB0hO6w@i7$87QuE)%#yuv>f2v>e0=
z(acO%wtPM8+P((Ee{jm*4&=`s^Ny6Jp}n7#-p$<a9lBQEKl&Q&z$#?Dg*2wFPxUi<
zPu=R&H{4MVt!@L}Ua81UfwPFf(;{+2I{fq`AdkF&36#HPM(Gd|C?^}MzTSbM9@oX0
zMx5*SAeKd9D5??_)kO)_cc2Psc*Xyk-{+7K-$S6M<js|`|D}V6Smf5zRK4soc$p#G
zyYTjj**o%~$%uD_c83CPH}+m=JMzM7vv=^KZ&_+sa-Hz^^JAPO@eeYV9dEtx2>+#2
z&YZ^Dt`!D8KO}oOddDT$jC$qLg_IFT{;B39KEx>=W45nPO}y5iSp>z{GM>fftiWpS
zcNsfg_T*#@+-$=|b)^npJ-67aZ=Ry!h)P4!xEEJ}o7h2MPr2b5&KDwXl`StAnx`t|
zn$Ek(*46O}`D|=YGshLRtgaP=5WZ1%za;wF>>h5GK)(7_tuDLM<ch@CNfmBqk|nrf
zs%qRl=z{}er+JvM*V<**dQ>Rq2*EG+LWo9r?QOSj5N=LEDG1s)+2?DEkdn$nG+BE@
z5e<f!9}2B3CN5|45h)jK=A2o>JzNhBUI<@effqNrUxeTov3W53%Ql+t&Q4L}pgCYn
zRh)r$^P8*g617UiI}X1lvRL$1*Q!u%NDe3Y6zgbG(=hu$EnP^@d*;Nrh~E=Ep3wP3
zW6eI>BVI9n<3b|7W}7~Yh)#je$akNmiOcDjL1$Utk>Nskb{iW$k)!3wVkA-AgXGvW
z^@o**@RvNxtepa$L&Pan8b)!dn{Uh_M%mQ(91D-NaYLBtF?@iTvYBLh@FKTlqZxjC
z5Y8Ey`B;8;OM*KGrQ((<|DITKl1p$4;&ZMF(+W$Eie8bawPJ58W|XHjzp0Vpc56`A
zroisAfR^2lze?Klk2Zd(oi*e^nG%Z>ghA{=d!Wl??;uY7X3OXoJ4IYqhFc_jgTB1R
z2oq@Tw>+GB9^4s|*oWr=AL2250w^Z%k0|dwkGoQ*b1_nC6Nql?R}V0FVOGLLf=knN
z4Yy<!2r^m{=s81P@i5Gq?U`sAg^|m(6!jz66YjBE9g42%!O7fP;kM58$f^+YB947=
zD~d_SmhnsSY%u&bUeGyz;b|1goJD@oTur(ur6T(b-i(?~g89D8l{XjI=<->3lom1+
z3E<WO5KOXjDM}ne;*b?2X;kF4IXap|+?A3MH$nMd`dtY*V1)HHJeRBRH`@)&_HGX+
zIurPwE8IbAIM*DBdEBo*Vg%7tbc<ie)RDzEigC>L?c6ONRLD;JQO$qX$Cx4g`aUF$
z_(iu5*e?biA^1~>i`Jmp9*n@k@&%&-`(G9{C|F^Q;%8Wg+6^1LUFdHrGK=8`A6V5i
z&C|`vIX(**2XyaIwmd9^e_^xVd5(eNg)Q|=ViI$xMv+0BJ<%ZimA)*lM(a*7&OGA|
zal=Y$8%<jeYZuwXPr3Rhw?Bv4w<`!PMM;0<L<ysqKze_3>9=_<`b=sh<QtCH?cd&v
z8%OY+AlgTtOy~optd-tipy{w?d8EUaXUe$^#AfUwor6Nl-$Q=bNKE2U3ibm`*s>QF
zkCowr<L}$T6sxFozPlWqQIWa$$=waCuAUSke|G@Aq`@&ASB6`X^KR6uh*+dYea%@(
zCGT++SmcLnO;6_~%Coy0wqZ~3q}Xe`#GX5ToHFn|F~nSJ&J9(ur8ozsm}m>^Rc~<R
zQklvtPpklP=#$ZutTI2n8lOXbzSWCck+4?YUPJh!J;aYh_DmbI&TD&ai@LLU(=#bV
z@pc)ON66hz!}Noi(%8~8!S|k>6T?CGpd0&Ja}3!Vi84j<C=shYh}@I&_!<pCy)DP#
zkW+4B7?(7i_i(YbO@}-EB&~U|HAa_vki>M62`}&Mt~r07pYoSc$<jU?T{&-7>$-DI
zq+#V9ChYV`qungi$34<9SD%4Fz;esNkES|uqSIx*<D0|9d=Kf_Pd0a`W91SAGpViL
zWqbE(jwJ1tf4_K_YS^ijKxr;Cu7Hxx{!&U0O(22DEZ8>Tp_x#$0?JAHnQEWmco1QI
z@)0bjk+*;y-Wk(5T$eWg27yfN-im3eCo(v*U8_tnEYW|LoV$YsAI9vOdCauV;InxC
zLS<%PpSXce8Gz^1aGz_l&YI@ezNa41{SfV0bdP`8=N7p<9JqW@`+O-SnbW;?!{YSR
z%)8_#XBRhJ#t8AQo7|%Nxc@`*Slv`SSvZ_oEbq6mt_do;V$ieH;kI{O0h=RX2h$zV
z^tHIs-qBmx@FlKjk#D2LVbr^>l-*~ANNX<2Nd=VYnuim6kNU4?J4(#<4En4sX_@sQ
zVrl^0g1GOk&oZsuH{kvH0Iy?lhu4e0sy{i=h?0Rc#w7Sup^=g9gts3PiU|>$?buwY
zYTZgPe77Z_p^%*nypb?#1jj<~BkG^Ot@uHoCcbq4RDM6hpi>aPkHMG@Pzw6*yzUeV
z%Xpc}tER*)m9&oagOK$j!?L##4!BRfUYUe1@_3h?iivNdXV=~4^bpxgjg0X^OHAqX
zZFx~#aJ!WZZ!hx*GTvmqy@<9GyTfsQ;{a5F@4>SOyR{rL=>-at2<X?kAMp6>=+zJ_
zO`JH1^b>Vem7(jiY^bN0yODb3H|V^|*MAIZl*m6xu8?BVwq)mgHq>1qp7W@Wf2G4^
zKwWlkR&xTyNh|EpE-bTsrONy~e%8a~n=s*K*O?_7LQsdOaTKi&tM7(QI}j4CW_-$Y
zXGv6XvP5(tw&;rbL~f#bx}-l!7-GM5tUQn&WnQOIS`Q-`lIgEXsNs)Xj7xKfBAB3Y
zGUo5>bj;?c2;PNHy?cK9g*4;bq`w$X>0+rQdq$V*5r26O=N29qrqEMrm~*3^sdRR&
zc>tRIBSac>I$xZ9b-72??O5L@(oWva#95B9uDT&=wV&Kkav>Sv%W?ob4$t2zA4iP)
zFFAcV-*OTe%J*m!QKmL%65Ar(67RdVA@s7<QG3(Vrj$}E0NSj2^>9ATAT^8RmC(x<
z{OeQWtPcZz@?N&?j&w|nCHwPWlkHuU49BNF?B;4A5P=utT8`A%thmRoo(-&tXxqVo
zW%NX_#3Tje4sFsca&ZGGv{d{uGDy~(%!{Q(a#ShoS0!+jT|8nOg4>e9NgiaDYuxJ6
zQ6GjOguVl2IL8FsuQK39tXKlT2b5^=rk<<eEVJ?ZXnc>g&LED~6wAt>N0{Da7{Z)O
z1qlY!*i^FQVA@5J9W$G*pFx<a@9w+D5}M}Qz8D`8P!<#{*He+`+Wu2i=Hrv}lu@o(
zu|az&o(Jx7HLtRww(>bb*W;fuzZ2Bp1eeT2tOHe%!|1z60p5nG_b9Y}9pyurTOH;s
zBnyav%goHiqDxTLhp61Eoy}JoaKgw!k4gbO4NwA~0F;}`LHGiGP_pxW$uKlD3FeYV
zoa%4lh(1d?(}@M<*%yt+c+p~$Gxw?XISzk3?P!K%t26!|?7d}JlyBGeJ#;DUfOJVW
zN=eMnT}pR@lpsiV_YB<)f*>s|AUTAzfPjE>cbD(U|GKX`p0}Pa@3;4p8_sMnbDrmM
z9Bb`st=|p|$n#c*1r^S9KJ%2wSNgQjHuP*Dl2pNDFS16nIG~F_cV5I{nqleZ5Fv`>
zljqN`AX>6koBF-pvB%Fw1$ccqf^(7w)wd0}1<lDB|0YGpFSG+yMUv<5PK=HnPhu;J
zlo)?JvRQw{s9F!@37!h-xMa#Kox;?p`iyA5$TWDi0o&9tN+lqrpR-z_m@AxS96W=4
zfBzq6_1RCIx4bHdRqt2AOtFXS<6&P{yfqI6_@W`A`!>_(Kd8!QK-^XCFGmEzP@hO-
z1F)rI(|=J}`-w`Y`og+s1?UBS5t^b&3ZfIj3*l9%`i??X0<s<YoO!uc8k905sOTGV
zsbaK!7|F;dX`Zl4uZzdF-rbh&BBUF_C&P&(7A7~uKXzVBcmi{VSkJfa#?L36AP<;G
zg@la9K$GJ7=a=76?gfYrsra&m?oebsO!9h{*;h#$Ty2}A59R?AmlBgwsg?H@8fb}9
z<)Mp}@q+%OjOjC8;*qH4WXP+$XX&ut&$)P1C*x`|1=ln`DZO6+3ZeIP(BG#%OMsx>
zz9f?y?0fbC;AB3^tnMjl`|0tgdXBj~TsJi$i7_lZ&Z*AKIg&Bd>7-!vhce?U6Ug_v
z7y=VvU{4CG&+u*FxvRA#!AVTIZtuii)mw}P*ev7m>5tQ*izciK0*f2!aWc7V?P!2+
z%!v$iOKaPXSElbn*JbcA*UnH%g4_HYs?HOMQ{1gddr#TXCn#(&p35oE($a<)j$CNt
zo&1~Y*(l?MCh(vb1AqbJ*hObD)+Z7}SGZH6$q=6z6`t=>w$nQb8}avTL8oy1R3|Ps
zC@_lht-+sbaw$?U7N|rW-aNJ>KNX4WPh(kfSX9(c=4h|49%V2*W!>obR<EBF&M}I%
z<kzk_CF?ceMjOcJJ~qEv`eMt+R7S|-5P=ewT#wQ~C!Og2*+*zes&BfCq+NFn#hmZT
zbU5A|bd~{NrZR225r#G>%;GXbq$&BQD&(+<xUmh4!m=g{r3nT~<a>`DKE-goZBe_8
zbVT|6_p!J`&O=f+N#CRXR+(mpNF6}?ytim-y4%}i!Bg85A=xUht-C4|-Tbwd-KC3k
z$g|YQ?pc@aQZ&@-w3EWa)5o1=F@_cB%*Ocl)E>1^GmC}teJ9d`z^KqZUSI>VC{Bw@
zgEJ!2GV9})+aG>aCtl(ip0aiv<tvep*Mv6S03>f^7>8Y_fZp|!BnNuHOLk1EVPQ-p
zbBrgTv~lUexPY*Li1(4p5n}<M!A^Yu9`e$HlxHk$?0FF%SOUOfS~buo9m|F@vze5Q
zuqOa*1`C^9Y$|MVw8Ue1WACl8<c>vcs|jEee9%?FJ$hK^$m^W)f7D3`Dj7YbHO!h@
z7&N>lK`T8kx!lB&z42PuwR%g00%-8Cbebvyn0RDrmCL_b6tB}YO@!swiLG^y?S!T)
zl+=ercOoU3a)0Fw$4Q6fclYe^JfU&eHmo9k4ammP=$bHZ53Q{bYQUIO(+R_ArSK{W
zMS(HN*;VZH$vG)!{b#Wg?#|L=c>pKL+zyyxj^#}UjO$g{tj~9KO19)oZjYB)%P=hp
z+NCe6<0)Fj@nS}dlLd==p#AwJ*&JaI`G;=M^58cSpsyNvm5r)qK<_dyqj?i9xdXAP
z0!hux+q5035h;!5*v^b3XuT>~Vz;qvo3ycu{Gj30M|GST?>TcNvDz|UM7IDBsraTQ
z_0O0^1j*rB<71G{0J+(V)exXEm7Ei(H~MNm?CXxV=BPlOf$pxY>`?Le@UhoZ&?yiY
zi2%jzl6)W@3Us7XD~m18Lc}6cm1v~+4FLh-v6Xg(>E{?;kNK0}J|%V$plxW-tGpz;
z8wG1$6iN5lDl@uA`F3pja!@{zFWP~s9;pb=FSO>IK!p<oNaW@!Z~jPN@DKI9IU_O8
zpKYLI>K-Ftaz%}NZy7h)kvBSgtu}(sWWRk_5DZa`r--tCE;7%Z18E+MU452YyDrMZ
zxfaL~e1CQ+(aDVZTQ%jf0C=}|tC)n?7CBEtuoATvkBJM@IcTVVxHF;UdntjZSdLt>
zb~Z0GdZ2E~-J_&T;E4!NMlls!lgA8#A??RKWv<~^M_Y-pNZVI74ItU7-UtCBl=t3)
zd_#4ESI?+cc(Tm9oTi%;_zsyZ2UdS=WL5*iu->X^2!FoJRFNkxThZ#^#Sdr6J%wD!
z&oVe<`9`1J@U^4)aHTcL{Drq{v`Xue{_U1aXcrB8v1RtWafmOkbv1k0$iB=y=4~A^
zusKXrz-Ai7+4kS{o5lHj@~FCY1bpM!W=vw@jv2x-NbjBf0J_>@b=*f#yq)v1rz{49
za`_L357U@~3HM;@fB^nCwPqMUNGd+Gn{?6IHA^J^>^A@|vKId%l#RROb5`ubHpaHP
zpWGtL6#AWLAAOX{f#bUKiuEvnhxw3Xl)(W{#Ae1*#smq&^+Ux4<Zsn8B*(4AvAQFN
zwbot_t|f=|A=3HxPHmux_)x6#<bQ=xbD#soU!EkkY-1#SNajjU?&{WvEi#s$$6e+$
zJu)nPnwPjPb6(vlt&Kk%DVg;2hnVjKA95~n<cKdy_FdX4+M3sd%X!Pi%;#IJ@x|<?
z*@R2dPy8|bv<O!(W1G{o=V0cO`P&xnY|}vDtY?~gyOK<VB1h^fVX}gFv^RR%5j-Q)
zcWoniZPeEyEl9^jpV~9(%cjHMRr%{slj=~m?$CyQefkyM7(c2K{@za$8p7SQlpn!v
z=ttjj(|GCl;(idzRUD6qkl@Y@hxdq}hUP}T5{_gMCL~fJ%;ZoP%vt0K<m92{^qu|7
z|9xU}2z9jJ)aouZZ5gJss^PkT+R+nIRbXz#otlOL?;&Q7WHGoSJGfA1|21dgW7cJD
zcF-SP>9mfQ$pT9Y65+`&E16G&Wu<j7)`XRs`4FKJ5{VHKA=vsR|1na*4;(~QQcswL
zeEplF7f!=t+VLJ~|1%2{(VGkgetl*hY56|}({XMH-%P9QtBOfCpoYgpcsi#s^QvU;
z*unJCjqJnH{=6(-ZPn$(!pbJ?o3-TU<Ti5lT~byqC4x%!>B&7Yw?uDnudV6hB9zR6
z!I3|@l$Fw-<+A?z^hLn(jajhdi%t81kRD>u$mdikS#}$cFQb}lyWPY`527_ll~z$2
zlV71DFMD;>eKG}CPE7{C6|9!@u)z5o%Iy3rGVHaI)ONy^)#R1((`F|$%J_Q4e|U$F
zrg9(|_!JCRdFjl7^8V(eYm0JqTfeZa%f7!?#!7b+PBrIO`*Uc@H}tK-h*Q12*QzKj
zEvSS!iiPO@a+cUgy1&jZ&D7nDD3Or-J+sj7VQ(rzKD19<(JZ0e`Bd{8=YU|yJ?i)D
z)NydRhd3H*3IBjn0uTyiwjx<HBx_&jw}0VR1Eipz2_ODIoTg=SJWKn!(0hm4KY998
zN$6($;{UIG@Y>EncF1sP&v{LYZ>3Twy=`&)Dh$EoWBkvAciuea+jr(HyMX_x!E!_9
z;9W<l-1s%b=+nEmZCuGw0@h~uJZ5V0>2yCC!nq$@-g(f6PrX!um!`?%h89VY5l7j`
zY-%(2$7w`W{5|wm-pz44^gg_;zHHCk^vk)_%N;%4^Uj?%+HF*sX_<C1Oa+@4@?S`C
z1jwAt_Fm_JXX#w(!AFIdWW~HT&#emMREK59j|-i^&UvyWGM_3#Z6{|^eXC;a9xKwt
zP^s9#t+8&qM89=k2Ql8?4H3Yoq7mG^Ac9uCC~Hu#&)Iyc#;;`vMMxL?!eKx3oD$yM
zVge0^3gvg3C`*UrSKe+H#qugUzxbUNkzGPPZq-tXcIo<Bqd`P6(~hmExr5)Y_oX(&
zVXdI0o91*+0jBeX5V!dOtwgt4F7?;g2K7c=MYDZbhAKI+6ObdFkKXM>fsbGL;i3A;
zxYoCV#c|_6f7a9_xsBv0!?Q1R2E{|>LeoD}6VjZ|?+Y!Gnw!{nr3v(ePn~s-qql0d
z>&CP{dAa=5)PBCfyE1W#HLapcJ+gVWO?!j;oGEfx;0A3l>7xv!|AmBCQa{e*?})v5
zd0AbGdT5^-T&%EdT*_6=)5PEzd~szW?4Fs|*{}8IQbiTmEyxF=rBH|%4l~vC|0N?)
z&oX^u0oS0v*QT=Z_t6g#ntVDLCIuR%F1r)mAx)9(5x54=PCYFg(Vnj)a_gsFG|yq%
z9aQK1>RlVJ;_*EQjemM@rTAy!ilp*7(llt<I3kHmIxenOYyKQ-PS6sCB*-OUXs+@$
zs~#S2R#e1xzS`@Z=*3A_UM%$GaO|=My)bq|y-2}MpJl(|?|4U4TQ27>X9}H<z9u1w
z4L9Y&`-S`b!V(Sl>ar$u_$DOF_2|OOWsIN+Te`d<G7ZbcuXZ%G5j?C&)vR5-$TZQS
zu#KDI;49Lm?X|CyUs!LZtA-|`pT49B$HE7(D5BGa{HX2h9A$vD<xL$gV+tb21<<Qe
zBN3A*Rj9E_+KdK&EF?1x6H9H^eZ7A=FWQlaFKUPiX-4NxP*?3jH!wjS(BZ)OE{ff}
zQZ5p`(jdsfDkiYwNzU6_x>VsswGzbMr1`sHch1z-O@i^-J4X@q3BU1Mv?K=?xCxPx
zuFRI_lOVF_7wlVkS7q2eD7H$ipUbJXTf{2pY_isQq9&Tzj$f%v&_RfhX|TE1lrU?9
z<)Y_m7NqG^<gMOBSNxPdG1AOoR!Ma6p+JS~6dMF2vwF!hIp`e_R2ZqEA>pq_Rd6P;
z8j!q0!N);b_5;Q|8A$h;%i|hmvC3+x^9i4)i^Yb9)sEF?6EP{^_k}%9bM6tDf1B1K
zS4M?Yf1YEOk*OaNcsJub#d%%n^Hr0I{cbQ@PJ3ZHgy&f=BaX>EX27H_70VZgqI@i*
z>evU?_2Pk=ZBqWqhtILc*0av_Sbq15ly1=CFZ>yw2CxEz$R6ei-lJL(AqQgy)cTm8
z;Gb4~4yeF`20z{ECe)Lm%$-2yJF!nb&`C*EaxzFBDVdq4iH7n+7iDpKsv){YG_{hP
z>4V%mI^w_AY1!V(AWNulCD^(w$T2Zwe^wT7NV2^#nuz_j_JWH?V=vFs)vIcN=g0l^
zM=Bb1Zfmv#%^^UTyF6JcY#8U?;lF=+N-d`Wi;fPAZNm?j`Bf^YAU5||drJ^TSw3Ao
zdnGGo=~<rEt2LU559<GU8J_^f3>f|uEsQT$F-(2>z!z(2|FB0wS~bfaNyYY!Yxh%$
z)Apqi9Z2I+Cq6l}h1o6xE@q(y>FB;<j@I|3wxk#Htyz$~^sW4oX@cHxrn&WPym8D4
zu`8jUyA>7g1f8;W)RvJ@!YdrIFwCwW)<QO!cPkjV3KmQHO3P0(l;9H7v56pM8)ao|
zg;;Qfm7IXQh_ApEXlFPT^kWBW$JM(v4JYHz%2Xm!UGu}USps=-KPFC}&@k~$BZACA
zI<Y+UQYblIpLvtE40bnF4|k|vx+gw-R=Krljh+6{*s+PJM1nji0g{3!rW|YyTQt;v
z$f^8Ps3fi*p31dXZIg9SS31yTA<Xr<wvN6Lgt)BvO#7i7o~>*P<-?L^7N$`~tT`n`
zx<qG=u%NY%j!**`TE9R!V+dx!LMje9xk=&+yg1&?ciD{#Ynlg@3`VGI>B$Wv6iCS`
zJ_~_&b@W7v0r$w;^!|Gc>z*O*?9^*1r5@C^g-W1I(6A$Fg|o+dR0<^7PJ+^SkXC$k
z08Uwp9J@(Y=R1<uSI^5QD)aB`L=yXyDr!$CqMBSeC~wbaY@f)Dx>#8A6T$AtI-<4E
zPwK1Z$u*uz#jbb^I&v^ZVD2}w!xpV)judJQm;_h|R>)QzsjfJq5zMN(N;cjkR&Ubj
zRP830V8^sAkQkP@r6hVHucTdsK<@zml9e{NbzqH`hWEP?uc878ssgcd%%zJp!4PI4
zbEcXIcYe5t)>};V!?lO*6<t6dDOf-*+Irp3c>ejz38>19QYMbYj>%HcmG{l6ET^>w
zuIIcw7-Tys@3&5*1F|4s<>Fh(AE6&ANGhSE+1&{A>KgH>5K7UVkTO;EEKm!mkiRa0
z7!;xOvRrP*19QFt-5iGk8D{cxThuz$!|rgp%u`?D%)PD>`#7Nu(px_T^UQuJ!FizB
zhaN`)QRmWceour=Hi*vl^sJS6F4_Th(iMa%82CIo)bn`(%p5l;l%4o<s^Ldn+mp<O
z4|K)3MeTX0=8o_goE`F}1OYJTJg)LQbJ{pR?X>y^^K11vJ}|?&xGlPBAFA95{#lg5
zeZydIt(8$EFDKd@6?qe>ei>U)BHFo+SbEO7VD>tEVw~aWvPycutJn8KeLjOIz2Z71
znx6!<8pn-`+g=LsQ+Cg7Rb^bHuXPLv&Y^a!w4VL4W$)$PNzr%wC8Ro2-Z$1T$Lz-8
z8{I1?m$5BESw5jQ9H0C8aTDjJ#H4vmYrXLy$V?c<Lux+6-#wQ52v9B|8MSOTqsZdZ
zqNm5T8}}#Hz&-wiX0^8SIUs9c^v39?cejyKFnicu@DoIvPUQ^1F?hx{jadx}&5GZ9
z^Wyet=Cpclxgsd5@&v>M{#x^1jJfDMcZDMQa*i3WmaFfKd0@$%EV*43WJESe{Nb;;
zP4-VLKnvlq?p=MhE<-lfuRhAm7J<{eSirh}eg0205X3Gc_8Iw}#aC0ETL0YV_pTdO
z!0O(u>;MjG=Sqp&VrsPUWbT@CAd9)~b0paH)AP+YbHl@7O0FgqEap04b~@m4-nE2w
zXnKKo$zK^Rl*r6iL#D{P2#+`L*0+;C^Qy{jf{zea+?8Ue1)MDsuBL!{X}!Zl3007O
zEtM7{-jl>p=xEWyNBr;EHyGa&PqMzCb~fxvWglbwrd(yVyVgACG5l2db|6Z5ozL|<
zf@Zkt7Ht}}w6T{b?~6~8Z)OATwoeJ_oWOcgn`geY5n!sURMHbAa7i;uv~{^0yJniD
zv$6F2!d?GfGy7uvBRM7i=tCAgHwz}ajKq|!l5VGJbBJF=_=U3a-VcmcFpk0`4J~^X
z(8mIeelR^)XugPHG=*wm?+J*teT{JN8&{+sljbl&+gHG(Bl<`MUzu6yl@)&2dId@U
z>&X#>hQ$j_k=Zjyyr2=_fucs|QAR;gWMDMUq+~F8%k0IbUreFF@%M$|SZu~$M?4{6
zXhUZKbv+eWz_)nDAh-22;|Ur^1)B0n5Thitk}P=P08N>FVizi5HWuUnTUl5nQRjv^
zaxkFj>eRRk<27Cd*N&U%!4kx5Cf+{hyCy7Ox*NBHU=uC`M+88F-j6L46`AlZycELY
z6{31Mj>DTz&0tTum`rf{6xhmu9W5{rGs|^C9=djt7iomgA{w!`cpitT2UWl)XM2f+
zGe<k+mRqAF#oo%OVE{X)B#&SFT)itqU>AhRWjG@^jw}V6cp-;1>2?T=C^SC!M{_i6
zwHQ6a!-M<vnX{M$A!MKybVI3(Ay6q_a`ESHq)RaML=K`-c@z`-X=-(0L9_@_Yy4|=
zIUb_=JaQ?TCv**H1g>3xDnVUDflZg=);oSy6sa_RHh+DKAaN-+HE;UZkFlT7=<=yT
z>K*sBj0G2YVeyllGNPUr+T>wOv$m%p$ymGuf(ks;k7l>r{trHEIB=!YQFlN3%z-b8
zV+^BxzH96R78R8+TnlQJ#^43IQ1|xJZ0K6{NUOtCIsYI^dNe;gd`hK$Rm^77X%p;#
z-mKKx{dy+j4zaF8P+y^Jb6G9TqUa9Ao}vl0gz<ivq$;|4d5tafCS_IlK9F4x<6yiG
z7J8ki@zMsvVhlYoV8I&*ozdX_!E*Il7rpgwauB<SBaJllHKREaH;E!9yWF+hi%1cU
z7oP{}5I!L#I&_lnCa9zw_s{U^-EL6fWc!D&syt(hDV45%z%RzEU<JJ=GJZA~Q>Gj+
zbe8)s;;5?cAJMYG5q?nYrJPa4<L9N~CMq{l4PU+JU!)}?@xE-L0t7@Twn%gpsXY09
zmY_85Qx_zPQuBKC9j<#Jm)E|G``P>|8FRSGbIj*~FC$H?Pl*XWIK(j}l%D^6?1_y<
zn{qf|GSh|eibUxAN{r<n3>KpDods}_4i1Sz+58e{EWaxbfCl=$K+g$)8D+DrcsgKa
zR#GRTp|?{4jmCqf@C30BlU)GRzVaVf&m}QqCK2p&aH=&gMsBfjl`YjYu97hs|Kyc-
zF7~`(U&oDK?eoo(%4RO7E=pfWO+!9GAxx65lkEo1A}Ggmx%+4fpGi*gv*K_g{U-3j
zLFQ$>!M6Az!X2krV>IWqIeP8%&jVh9g0$vwqq0FaXRn=sx<+}Ub7QzxAwn}xOHGRx
zIh(q~djBPs=%NiAqj<PU@|BdtBDTju?rc36`4rOU=b6dVT)OA!+Ba-WI@(8&*1?@C
zEgm4E0S=Q#F`u&ilnvGqfb_`+deW_jWW`ez+xsYvHuy^Aq`IR`dOjh#bWO!LCwTI8
zo1OSvt@n)nFJDKIDEHEkq`sv>)>cY={!dr=EoccH_a{r`&}&Q1!lGvJQzHl7U53yL
z=sZdz)O3np!N+0BVNfx%#{ND|i1E~7Fzo_L-ARG<n`$N;t8lIGi&yBP<)5z5=}WC*
z_?;TnNrL*h-0JS|m$clWcQbO%v2};kFKCC8)2ZuHGB9?Ke7SeaS-72~`Z?$_V>z~k
zTTIX9Ya=Y(u<z-F9bI|0G}3}6b0b|HZj#Q7j^)w|=<(D0&P*$eCzcAMsfd9AvDi*w
zx1VGmAH+`e93*(Nnaw|B2EMX%7<pxga*cs;!|G}glQ|{-o;49;P7Y)DY-<xBvy{*@
z4*3>Ji4WZ6ilO-~=5YhrT)%9NZg}1oH7z&fTBMT%gr?-T#gEygd=D_KGtP?UCtYd6
z65?H?Ovnz}RGEV)^_Dx-tIpYDT4ICkv3mhI24MUlN7aQ_7G6!vTXE0zrAh>vpQ$`u
z6%+Qn(Y#S{B~KE))F5R_*(UF%n%}G1sgG}%->Sa8*rW)Qhnfsx1t`U=Gk#o)6bbSp
z?8C26uZ$<uVK?XDM8gbW4p7glM*Kjg8no4wuR*ibMet1If7G30Td>xMQbY3uNkgeD
zl>5OyK7?W<q9TGZ3p`E6O#f7u4&xL31{+V-Eo_nzeG_S0M&rVqv*h6i5-68_PF_=7
zXcOxZ1u{yM8#0P$bTSM~YhQ~$$ftCMKTJzJ_{ubm&ss$@$?x~$zKkSmB8LjZ4DaX{
zxE~;-gIHTqfg0HZpvw=)K2b(mm-bybBjs^0@E{nQbe@KHZSdf;U{hR>suqqSwP}S(
z%!?neK2o~@yg}-T_*$YMdIfY$nchwYt&}=sY>hsnK2>5!&kzBcNYBTlvb%h>cYH75
zEOSa2Cs^xrb}0PYD%Ysgm3-jBF-TwF>>Ds)3PK#ey)(gM0+rbEHMSh_lxCnyaA$y-
zTq~wBR(NfyD*+5;b}BFi>UA<E^X*-s2yD_ue9clVFOK$8!;2+^glnoxQY<y10yU{*
zfhljn?0YPYolObqng9cO5;+Sbq#)j;`$@=?K_cnbmQUe$F9`N=iU|fQYdbeciLqfB
zYM&>>1_T%c*2LT8_0;wVPy7m5Qj(B*-X@XOJKdnR$JQpENnDsa$9iQC;s}8?oHJ<&
zK)QMP6|`7zsvGv|A}0wR$wl?)w`klFrb7nE#XfZS6UO7J4ghe8F<hG?=<E21Jh_GG
zPO?Ft{=5Zl64+#mO^p?b9a{18qs@d2tt#&fhzof?$rI!Lsay6n#XY7ki35fNs~he5
zm$cN9_3!S?MBt{Yk2kpcQ7w>f7JfY+u6=6L;<nl2lTuo20~Y#&EU8Y!0zS(Bg*0yl
zHhaQL=Yh@6zi`B9Lf`#~U17(&IqL?Vxasd-u5br!UD0}D>3TP}MrluKZS&b*+7`y>
zGLLK*-l9#=4DmLQ&K?Q@851BGbh^@X8c<D$eBk3BOh!AKL!3)bppMoe;fU6<AI&i)
z)?S|4*(+^&Zfpd1F&9y+yld$cVHWLPa7twFYZ~<dU56@*)Y@B}(t>Bx8*6^WzJis}
zCB!P^q#c}?NV0uULeNV4YSpmwum=1=opz4OG@fUl6vi3ut&j5(vqgnWEXn*11EHB>
zXVbdS?gw9^pW1AknHWxCXqpNr7*A?_%h*QkbiTOPTz~uV+f!s3H?%Io53-mZTSVi!
zkdKn$(ndU8$k4R*CRVi)`zWh!-pn7g&kEQ~?i&<jGggA8elXtn_zL|l7p}!sAe+Y?
z{zS_I<+cc>xBC(^V{4enFWb(I$<2vOgYGcak$m3U>my+lWFn-G72q%vDZGkjN4q8U
zi#dTD%$g$YH$vJnE<?VaN;YI&ah(?qiPvDNPmZs3GEc&y<aDqsL7f=7dFD&RAntl2
z-~YqyiWFuzKqO4UzyFC_g?&;WOsL_v@lRW7hsJT!ll@Ua<iwE9BqGp&NR&=GF*Ozj
zahDP<CjJoNw*9P}PJ|+A*BS1h4da@QC{)LS08UAx^F6?)vnNCxc}YcKyFQq{w;TSO
zf<5UO3K99rNbtg1hD%-qQna6RF=$OZWL6|^s!Zc%ZS>Xk4>QJ)=nzj$10fkKb98T1
zb@mT=s8AXVZ+7!~mNTv5fvSI!j%sEaSRWm6F^nQAP#OfrqJf*(ej!Z@z>{1wx0iw&
zfJw!}f4L2}?)g;J>U8=N$1o47`$<I68pEQmBnkLzn8=-B)Wp2(;6%>$ev1eZZbvPG
zcOH$lr9ju;z2k$}8fTv_xHf%Rqi-4X9n*!-oac2nArH5I<5lJ{rRP)6Xb(^7p(y6G
zuwJ-fOA^aDs&3dNv+UXni_f1u=1R4dhfwJ7+~8h9%mGf~5^H&v(U;#jm<A!k5@HgY
z0muJl0lZkdES-<UR}=~+M-n0u;>Uv(&xS~PPKeR4jB`c=H<izMPyykepd9(t3FRQ%
z3z{v_oGX`ufDIrU>Fd-&dM^jq$Q`dZeNndavYJu{qgxi^LsB>GW|xGJV9gTPkrDh#
zt%HC+cV1TUg2p_gCUH<$R9~dw*GQS_9-`>mo02J}!Pt|+Y^|^3W*IwKGY*Bz%7$`H
zd=ETEFpLeV;cqnaaJ-L2Z&cCIRzB-f_58S`ab3*6COQ}WsXQU`Ee~j0@DbBZDS2G?
zIl{Jx+gxJRMR8}2T)q-(nDXEbsZOA7;sdL1;Q#^qPK>EfoxJ}~ncQD}K#tjZ##w~f
z_WTSdMm8oN!Ks<=NRFns)&)g{^f~XC(;aCQcmL~F1Nr3>Nstt}#DU2he2Zw2v2Z#r
zK0F4_o3_uLB1q**_6xyH{LC>DgS3^~j9Kt7Q-m3N5imy#XIwp2;i=w=GSDswHx=s=
zIeG8U)Kehh&kj+MP6JjmLI9@AoLHdQL1$#IaKFYRZu>zU!RcvJsPx9i4V*UF>K30U
z1+v#`{3SRT#({}#=_2MrJB>uUU1N%AXWt#7ev{F#^HYfl6bA$yCBqYZ%2)vvd~~1q
zIzra^Nj1T<XOh%B5uk`1A}5UD^I3!FWEZ)MGoJftecDsg;~ak8<P^ixwpwQ@a~=<i
zU=t&mEGtSf-(O9b0SNM~?jua8*Ok~6-mCc-UF-HQKn;dWJ2B$#F~*_+)<e6ML(OLq
zbVDcBTy-uukFBvs_VZW3@>cJ7T{bEi7w#lME|DYk2}&(za>^iO;R^F;Dlj&gbjD5$
zvYh{q#88>^cAOxM2h+xE=lj3T;D1u2N73h>6FCnVNjE#-DcM)Zo8u7UkN=W`{(0r`
z4Rg{EtpjEB<X4A-5z=N6bQ-c8!>MBQjh*V5s|D=f)Wyw$h}Y`VUIk!^LjECbvD78l
z`CG+I{*P;^Z2m*SZK7PeLCK>C4O3$IniIcC^2d>Ij;D0A$ucj7Ce(2%eJ5#DG1a4^
ziVr_U(?!j?OUWpt_W4a&<uDTtjV+}E`?XQA^6O*3EB!-KP57Ye8>_i2V$1dRvvmH!
zm(uJ7>B*V=QzeS#;!!|DQj+w3rPBagu;GlE-;A5CMBT-LTg{5a6#rbk<D}o3jEGAG
z_FNi)G)>f8z4p1>=)aT#LKgNx4J}o@6C64rnP%-#uCEXJj^7F~#lEgkYA=Z^m`N4I
z)Go}NM)8BBJrlNCI425zdg|8g8_Za(lqw6W_w0>)UBP=!7g-wwmKI@M+!5wVVY!cs
z2%gT7i{}6OQ8YI<Vwh4|Ebb8lK=E?Jw%d<5U(Eko@>0^LnEqK0LvTJKfoI?e9VkU~
z;)-kurvIb##~<~~Z>+0{nzyCO&00BhAT8j^lDa(^S}OJ#n~BUzH>TRY1kAIEEOni?
z+~<J0K_B(p6SrI*4KVj@Nw&2mGo40Q@7P-^lQxNuL_)77`nAz4egIQ)yXr>jEmXE_
z)Q0e#kVD(V_v~?ny_<~tS?vma1uMX0&D^B_V084r9o@ZdcIxeJ<yt(>P-XlR=ZTO0
znxA~@f6Ig;8pKyHm57mb29N$j-~XTgs09WPf!G3$hXB#&zrA8E2|3`6bp^>sNw;UE
zG3?GC1C52zwS$zF&TNAxwzk4L_~GyA6<v=39!XTK&(t{uP8U8Wjro9^$&A26a1V##
zl%apEgjsZat;oEasH(YFu3hbv)#y;NE=ypLwT0&x7=Ar1h%AMwtIWHC#SlwuF)ldh
zpqKQO^;Qh$m!>T&^BeQf1^ha&qUaOz9;OO=QOEfgr`TFT(pv(7RwIgIRAhx35gO0I
zX9lBgy5$FvzJ!A)2>}(eQo7~5F7fPuSrta<QW9233`xgTuhxt=Zvl?-5HYV**9ia=
zVf!+<Q1#B((w6LkYz<Y89ADce(^K~aavnRY+aF{kd^(L9N&yET7qb1W^1(`%DH@+V
zI&IvD6F~Z5j#~qxa-5-MwqhC6Y=Z$OkM>0?ZJwul?(ngM2~4C%C+*1E7Ojj_e$K(0
zM1nd1gR+z3_h$MarrP(BG?~CTzjd3r7AUN6o2*^Uyc=(*4h1a$GfJ%|sLr7cG>tWs
z3FOX*Rvn+o%thjLX#OPF1MEWMiz-EWqwhIdeZrc8vCp1t3!fTr)BX1KvaAtbu376z
zPYuAkm=#bFIhOK{d$|^P!x8PS4z`G8PFu2D6*!id2!|xE%KL3IS<tc#&Q+Ou06dJO
zjlx)0$YMfx3#O^gqIJuhkU7phs7Mu4Dd|&1s#`Ljt&MOPL3qUzJt`J1U81i}1FU+R
zWHZ-n7m{t6$bu!CG_M>1f=2eKYqm&oDF2QW`6`~viUQqRCjd#)FixN9ASd8Uc78LN
z`%wSWwAtxkP=)*i<Uk<wc9DAppdC+2Hqqx!p;##Eeu>U>XwcOr6SU2U&%tzi93YK_
zl|FJ$WY)%NoBBOt#jt*15<0*wa8qDi>&?A8A`Ts-eSwz8<iR8JfK1vD&p)@7=X2vL
zYmz{a1}5nwyAu7Up7r0J>PKO8924OMg`M;m1m*w@WKisfIK`8v>kV?z$VAzGs#*Vi
z5yW*4OM~3QfZih?m}V}GizMt>eIA<DhpI;+1R6Y~3!uF@j@ko#+{w>HfiX;!gSz@L
z^2Mt@0dx*9hAttmdt{Sj($}1TAs&Lf;cNq*0syx$vhZtT=4*ug%=>RmJ%oy`j(=`W
z4G+DvJ6~`hHRe7#Sa<dw0povWOC(s`2BcB-R-?-e_^CmU1m}UpS{YJ{pNtVwCYSR!
zOkFawfS4UzL6LDQq5A`?7TL1W5`Hb;3LXlU)6HjnS|ks)3_?6L1z6@-bS7bdN&MX6
zB%MbS0IvM8pO2}s5eD~mK%c<SqEmB~Ruug%`e=|o?nZezx+WAV&x&Utsp0>g9H+8L
zv1HL^Tp&W{-GP}c3I@gdfaa#e7u89zBQ>_^QLuLh=MmgLZFR4AoK5w6Y7zs`#jp6?
z;2&uds+vc9=jRu#cqTTSC0l{MN_~BNp5Dqw^BpK^*?WLsr0);Wrvp+8aNxtMFY~o#
zHnzpeFiwJw3K3iFGtnlzUbJfVl#Esd0<DQpulxB=#*2&w%wVRG-YWnSMLYuNHhIpz
zW~&SgOQdA5nJ>INp{UfJ$32q<hRf^De>G|KDRcO?M>s=6nrM)<I@jra-9OWNl4V@n
zdm0qpm2~u@F*rh{<!T>{7}{xgtE453rA3#?4(y-f)JHgaCjMkApEnttrBl>^N>Ozq
zUDnaT(BTc;z=*QLJ%{=8O*(?`M;DNYmy?=QbaHyss_7as^`|0P04JN@YP>Olikp+%
z-@=xA?17nR2i3pnb+NG-1+D0r<(c6WN#6p0H=z1aD&}HtO>4tc;BUU%NgwfEo;JPN
zY%R)Nvb?s*V7#&JlO6mBD8NvQ%DktjNR1tUuEFRK1CLQ`!Y(oHUw>`uBpg<J=D)3I
zyp#08GC10*9d1xuvN`ta&qgz;)KMLe9xAjo^tH`G<pHpT8^*7^OZRIYKL0Q&Jl=OD
z*_uBr-Bx!4@ej#jPAEBVjmdAS?6YL|6a+>xf4H%y+rOcLgbA-i(02huAp)s9e(iC(
z!;Fd``;bi2K1>6z$qi?`oN&f##Oi=QAMUs#CJvY#GiINUmWFTAacg(mX+kh}qVk6@
z6kPHJub7Htec0I~y<K##CKjlWUX0hr;$|BAj}W9up=N;uBa2sTUH+Glh9b~b7SdX-
z92uIB5P*!1#^vu)V;Qw=t6IeLKO-ZPX*$q2PC+g^<@3lm>5salb-f=W_}L32p6rp(
zU>%7aBUW-bI_YNyg~Q-4Xs{7vZ(y1w6E-xM5*lXe&T0WuwAzNIBUwTh%srUkb$NGe
zn6QTxu(#DI!h&hMkL<CXJcsmSj^HEn!V(Vw?DX#O4YR5(7V1ARaeRm-;@Q_{HwPy=
zSA<Y8bxUz--`Sj^P@$Vw)-69x*U$oW53*I+AJ<>UI|V%3reL%OBuePCg>bxzo+pGj
zL7Ii02@#a?(_zC1|CR=34+sK*E&)<n&oYwA*egOGyJlH;M%?)QdH&k*A$%JNKyR|R
zJO>nGxVwdVAO}P=Zta)f4!<q|)RCM&y^#wkALslM`QH)j8c()-d;h141}9X8!Gg*c
zFtyfg9_PO*G@$i|yqyO|tL#;>+q2^w)`FXMrrG4|YE4oS%w>1-7W~{R$Kxd`6aycE
zFQXV@TssEfA0Z1a*^WJA?MSqd?K)bAK^2haLZ~ZbJxyjdbC)Em<QG{F&sH3FF2{2r
zIf_!?9X=`A6FQPnelUzFLb52mEeqfVeEhp9{<rxn4sgz(c}UI$5v%$ZR9UVvu1hed
zU@R#!(DW%`PI(Ff796ptxyEd}x6OD1ctE#ID|&>0ytq^btpW5seLVjD<W)-l{F=a(
z+M$Is-`F&>f1myuQajo&yjlT|qM8;~YUPRVt!tDNG{X`Hbd+b~_*<lx%*Jx*wZ!(h
zbg{6ES)Vc2CyKV=vz~uKNlGRM>7|tUPrxYKi9Z*pczutBPGrPNUt8wOXWQ7_$?sFF
z2)YHgr?^h~zYj4P^1zNv6W`&;SN04L#t0;@tz@Ug;Qy$`1WU$5gy75kq6mz63*Ol%
zWaq4;1xyw8zfNnBX41SF4WWsyV*E7GXKQ_y=xHJXVcZmxTBsf4PqAFua{mIXhc6f7
zMX?t{g-ah$87YyNKq_8Xo;Uy0d;c@Qf7Imk@FBXP&+w9%daD6|`^O3-@Zojy|2Vb+
z=*Jk5b?$=S!olveWC%;cGR=(7hY|>QYc<4xeOa=$Q{&0;8NCj)p|eYk?%CVkQ8)0-
z2RhnA#JPSrW)~m3-071d7_r<N6%ZgqzWC&f8o{Ka83^j^5?nHK&RsI-4kF5Df4Ynw
z2p2K&#j!JQlWhN<=w1y@1<&@w{U-Va!m)JjXpLEv%se3)kG3H{yofyuIW;=coxW6e
zVY6)e#=8wAosfqqVSQ)R{koS4BXBh9ed_WC8$UzC@6=nTUSX}W&9D6^lxq(*l2Jc?
zb`P{Gtjl_Bf7;^f1W$lH>GABoeAQy&+HSqBiS;SE+L>aLD}|w*z*a}y<Q&clSs-1d
zbQs0f85Fj4^>{}NzA$KYQy;Yj+uUHDTH_%#u~WzG*)8lF=GnPPQUtq>Wzd3(c+{&-
zVjA3h@9&?fLyCz#z#8o5IA#R<@P;61h#Y(^UR<I^gjAc<7!w$KlvyZZ!n5@CZ&Cf$
zHAmwt@$!XJQMi{x#@qqZg!?3c;ya(*c2IO!%%3Tg8<ZXFOF_b$b^2cI%d~cBeGH5_
zK`poI;E+f+p#uU^JeJLu8*k#=`m2+Sc95n$Fwh2rPgrn@;$1Irik^Fhw&*?EnT~Ae
zr@qmjqg+R7kGY>xJc-xrE69?4-oTN|^Ksl*y$>ci>7nv150AMd|H-8KsX7>AW{YoI
zz=mnwvDLx4t-{Msqtg7<v-Gfsgjl5B&a^=*s_GEb-(2;fmK{!KtusX(@9KP~918DG
zHZB2qTDB7_Akj!7N9^B;=l|Sn|M!2&Lj|7Xlz^4g|I>r~zyCY_dFTKCng07O@PB{t
z{=dIYb5IkY*aIGCEw7kR389MqXYV133b=G9(&Oe}*d3Rpe){Mqt^>e*x=(X1+rVaL
z%>O&LKkx$oJ+g_nj+fHElhIDP@J@i!NJItr&Q$tT0kjO$#WqjFnVal-0@;qMp@Ux1
zZteDy<Rbw8G`}zOVIu^vrfl6)@0i991V$>^9!(Vg0QM~oq=_8|fL48vf3<7;m+a9s
z_6R{AKWonA^9QG!&FDDFuj=Ys3c=Asr6q)kxrO>}8RHL7PR6V9{9)Jidg*!uOCg=9
z_y0<f`Csn@=69?BBb@*Gx&Xr|obsd2qhY!5;SyjPW(pu~@Bmhi$>v-hwL=fj7J2V2
z{=_06wS?zBoPGKO*e6-e0=`*Yvf?-6E9TK_*MPYwsnN|uYM5n;Zb|YT7l0W>GMjjW
z@hSLFMFN*UiT(&USB2Vk4G*z_iJQjzeIUpE1AIYt|JkZIG^E4BKIB~~DX-&RlJNb8
zc}{>F$NA?&?(Kq9QJFCvmCq40(dn_Q<`;i0`v1P*19gJ{&A?h7>c#)>uNk-v0qy$g
z^vat8<}f%q5MdyEoGcFP=U*Ycf_I_lfZBjR*~R5|9Q}SuT{iUkbTj<XuIgy2!2RZS
z>91JZwz3m8!0++!_*riB6B{(d*=m0UB#p$6&d*;7b7fb6o41Y$ssHxqxB`q9k>CR4
zDG<`pTB0iF|M-py^K-y8{$l96^CmYV;^nzwkFIt0vv$Dm$(wo%kKtbxi~k{Y{QFt^
zkN@msmjL#@x>9aYwtv6U=}LdJuqh_c2K*N4NRW=w2XbfP)jjuiU{oNq<F>Wq19BeZ
z=GUkn*yyZ)B=qr_G6fJShZyr?myfMS^dJe=v&S^_O)%&Hz>XdMT^ta9gW|r)wz%Y$
zm*yruyBDx7&c3q`82=FEb~Lu1fB6IS_+T|IdhPan`bo6)SY`Qt`pm$APXTPR{QYC*
z|Fua3cF>NYdsf8?%CO)wfgY&!nxug1Z@j>YboAHk(ir%K+?tqzNbf(xE)Ic8LS$!U
z>jBwgASs0Oo;SMx&HBIo>KGw14hd&{nwG9*a*ElR%|ExC6c2gPI2J=Y{wJ^`81aAy
zjIl~-PGKC#hNvHo!o@?LY3b04Y|E>@W7CW82i~J616))Nl$~**e}9uVk(~gyl#IsV
z9-vx20j4cfJ%Fj|Zob^>;iElk@z-r9$yN0q7Fyc(EV;^+z*<WW0Ce^B-OWq~ZqH?m
zwFte)q30;h>Bf(`CRSti=5cgSrzyer-j9Zp;XwAsw*WXJ2HriQW&YTzdh6)`7L+Kc
z9iPO`%U+b-FDVKkJ@Rv3ia0;L1I&tDO^>rSMLzMyxfXJ!_+JkhcRj8-1db_S!iEDR
z?T$KtpD)TKp!76!uB9&l%?JiM00X!S@bO6&n-@xO&nXmxk?eE>Y&zR@IZw0SRqQ6T
zTa6v}Uz(-*(2l~z&hi13fc#9P1)$r704sbwr9CIpzZF7OEu4gLay&QFK7EpCe0X^q
z?|)~0b>QA%=jjQMExU+vucXRTmAC;9nDdF3M<dDtzb(ea?zdI(04zpHb7_e|DS3#s
z=&o+cKQ2=v{r67%YlV*A-jUBU-cz~-_|b(XYw58Z$>tq~FrSti-?NIWv#8dz<I&k`
zl77wpyCo~WGbO%KggcV`x{=9i=laoTs#&EqpPEUz_XvR5E<}hHJvT5q3L*&sw%4>*
zZ*r)mKt=IN_t*}(ALfREclhk`Jtwl+RrNbbjLtpCnkk!K&RPiMn>JDAz!Se7P_6j!
z$*z79nAzjW{ut+Qo-Q?8Wb|tLaBi=P3K+(E4=faY|I5s!yS}9NQ&=}csqJUQT7ag7
z+K0RIlPKp1^|jMUymP`~Ni~4G(_gw1|9dJvU1_6_+UGVQnf>|Zz_npK>A}@v51Oz0
zj8uETuA@_McvgYD*dh3}gFhfc)m;n-iZkcw6n`Ni=|}}I#EFMnt)v)EO@GjLs}f=)
z^;N)sAI^>c{#;9Eb?+@QCvCP;BfD?^S47=e5DZ#<5CE)$@y+)J&wwsNZJ~B|(bB|6
zDCfqn>vK(Q!=HV1cXQPpmp~s%0EH_+G{(IhtnkHcJ_B~P_H8#<GF%(BJ{}E$ojLor
zNiWqqHjcONtidkslB;}rU1i%wvs_?V`kSvGR4^BfJvrW^mY^fK?ml9hJOIP-z2L2W
z;C#|n`q6*S76u%$b&YfsfJ@n*e(?uq$Yu!cuVCP#DlsCffU<^Nj^F<00z)A%p}Wof
zd!73Sxb2?_SanQPqKps}k8rvGlYh<&AHebPFftzNSenGpB5|eO<jJ29G%JJ_pl=#^
zbdxWdSio{p**Ieq7TKb$rUWYi_}zOTUz%GBZY$_S%@hY7Lm6QYfSF3Ycs=4ymR_~z
z9SM5$)yX6`iNImMlhiO}B8oZ%cKb}l1?_J;nwILM&Q(|`z!Ej~05*09OX=qte#Q?c
zsj)qm#azAvev)HrRJM;BIEQKx2ik<sf^&cR1a$#O5FgJiK$;+_2|xsAb_2BKETe6T
zC4iQV5g~w&4hXF8^aEJve{<IqawZcbXjm*`r1)e)GRo|L_RVzD^qgJy5X=+mzsJ;n
zfaAY~fXM?qf$3=sP(ZuI@4Cg)yx(9-c?O`ijEddr{Q5KPIq%T;H1~eb8s$nupfk`5
zD%8I3;KWB~`QSI-VJ#_^0kmix+s`d!17ztx6N+448PrE<TUIKwq=z4mI2nrKA7e%|
zEc*a5Plo2lsaoUcU8DbP<9p!vB_FrnE>7jAJ9wmQ*Qq_C@3<*|QNn3lTLFAWz)N;3
z2Tgsh7dXwYaepQIk(&{W0uO2!(o8PQxZD7gC~J^D!AQ$Au{4<q^Pl54?|(cjAY`;y
zK)<!y?ce*3zt$FtbF^z60A8(z?FOI|i#;NPKE}3O#<{oAx2<zHZF_|!^4}<}*y>ef
zoVh!kYflY5x`+0C803px#TWkd1Y+DDwDPDxT6pkdjm3YkC#wWvsiQx5{<Ay$IvO*7
zcRL=2)(9;KA4;_m&wTLG%^n4DO+poG@*mDL-+%?3`!)8?>qhz3G_eP`m#LiR6!0O=
z<~2gQLi~GX`nVV<%-=Q>@CH>j4&vxhKzQ5Qmn@l4FyJDRGVN<42gGD4-QLQo6#jNm
zp`uxl#4Hj`4FxO>CnwZCj#mBW^KFUA&Hw0L#NV+h7Q(ktx<xq1|0%iG8SvQ*eZ(4n
z!11*CuTYCw9A^65dUD}IlDqX}9T1+T4m?GeCShfwz&c>!otXDXEg-yA$voTNj)lHI
z2PkV_w3OD@cmN<b9P0Zm*<GnuyNjlYoS9{5){O*h9q&-2<GTq+V?Z}c?%#F|8we6&
zJApBAb3tH66r*EHwf=FXet`-(6F^I(mqKtel|>S%xMAN*=2p<p<W(C#jS|>I0Z~*Y
z-PYkhwe!dDU@`!N^Bq4^ZMY`!sh#uQRA^hENi+?vJn9h7LU<7j0~}8KEzd?aJv1TS
z@DJ<_4YtYi9Y9!mxq_8wfnWe?p99{O&T(Y@0H&6TX5b6EqPot0P*EQ;2wUmoDHPv~
z0td$u37IAY>m>gC3rE9jmmC*tsFN^Sb1x!{*cB)ToZ*|zbtW7_Mmh<fDaauF+1GM`
z9#okfj~{kM=HN4(J0@`*<`4#y@APQk8`_8WPZj-_p}))mvDiKEco8J+M`-jiC6|H{
z=(JDS!I0&EBB)<95)SgA_tNc+-rJ&)Yydr<Lb9lOoj-Cxk6YqoT6PFv54p`yP%#FX
zl3AQ3)foQb2h-fHZ*0(_WqWR<{{g~*7&xc~ghe&wbOdQue(n#{5*qHt!)R1u>4#H%
zUY2P){F$ropr^!lO>`5zL6K9(_qKJ%_bR17quw(MVMaLqyaa-pY@mN-vqzS^kXmRt
z0Hs1qlAB#m!u-zxz)oyvPrdko%*N206>_=n<o|jPsN^ECIq5rmO=i*8SuR}ad-oj7
zZA%WT{YTlC)9i)(l`syp)Zxx^fQ(4+cmaU-*;$C5N8Cd-X~vYp4LYWBy=X-m&&V2*
zhGs2uzoSMUwOnxT7*Jye`jRG=fWAU?F4|PTWfbaaujCtHu=h3-9EB!Pqg})uK^s16
z#_iS$^r=|(%7i@TAu<t#2T(srD+uRzGmgNU5KZDfi#aNW;{vyVZTmvcd*n%gcG*3V
zWvto(&)y<QRLpc0rR+G4b<%llEfVe{``SD!kOBG|5FPMONL;W^WO-=Zeb(ryv4E_`
zy~4@t58N3);l+})SRPj%B_!o<zyF<VG%_+eIUVV)An!w6o<6wxq!j?Tr~E##`*#=$
zVL?XZB8U)R)D6pF?eOf)W877sK!OcBfVb8#YA{s`#l*fC=yJ^#w7PkX;2b80)xkx_
z>}Z0aF;Wi@N7D~XsF?qKxt)i{nhsOW%5~jUFUh0|FQr8uj(wbc`4|$;qfhXkEZ=LA
zGka0Qb()}({+)$|kD&cVSv5{~w12PkAZ3W!)=?h4u+|YC&g80EA$oC|!^a{d1h^+C
z`EU4k79~Y|G9Sc;!4CghN!^MRSpy-h8xRQ`rLsw#zZ1Q5e@D{wNyBpsz;WIOcIra!
z0>uXP-V%FqzRPdfea^w!AMd+ou1?g@1SBU?EK{uq+S!ONO2h>DI~Qs*vaymqsmxtw
zQ(YdacXV}91c(BIjf(|Ek~M6es0bqIycics;^HmE#yz<^y@+@^R2c%20<d7qL1XMX
zlh`qq536*uMid>Q$NxIjKeD#r=XD9AF5J3)Y`i(eFXAQs*X3N5*RmWFQUuxTM$OPI
z1Fi>rq3<~a^3@*Y7tNTVU~7WLTFm02NR1IG_=^JKFYptf2=r7v>YRC9nK+$CYd!SV
zgc(Osogq>LyPU~K5;dJ!!(!%JgNl`+7$KGNu4dGygBn|Z?Ym03K6NdcLd5tZH8tju
zem{k4>aAqa$-L{KHF|QH(eR-inoWfCnTW&pshgUk$m^M_V<5?Aipu&qITIoEOd1jW
zC{7p1Rf_rsfE@}J++~}6zltfp&S_myW}>r|hR3bUv|oIc13D`2qcEG)wU7!Ul&(2$
ze{20tL91sDa#Fp!MuU47$^~2(Pb|6J4ma#RizF}ZP)xU-FpjJy`x^b~o%+mR;>kca
zV5~uH={!1^DOTel?%y!r1YfBbwPb(xhkY>ARn?U)SXMWOr&(&PKIis2blaxz`ttT3
zEKeKtQvW6bGoMjA65@1^eJ@W1`zqxXTjl+GS7`&3Xa*ZvW<GHwhKom4P{B7(AeqIg
zB?c^ScARx#>%)Rhi*2>~en0-L1=yek3cIKtcNdP_M<3cvDkiOeU0Ijj$-3^~pObc=
z;Qa<wZ<l>wGtw^tWt0r_f0=ivuTrmzbF>Gn5y;NJ>Rz|;k7rc;E^!>bE!Wl66nR4w
z7a(_BqaeWlM?QCa;Pk;!hQxTO=h1S)8`urRNp)XmVVMF2NVng!O3civ=DV27RDF*z
z;D02E!!)QgHjsD0olL1h$t<!1Y~|eQ2O8cv@bfZeJ6`qaK(B*h9OfuX4(36(;W4SQ
z5(Sm^-we1DPjm^)o)!rpTaY#H)-757RdcD0z#T7A;88Jfp{H1%V;=0%(thb+o{nIL
z`Oui+N@ndUj4i2@GR;Xl9wu@{#H=Mo>S?AqBGaHU>E_4Os(_<=B1L|{6aheR0t7W5
z9XXP{V7<!CPckV8^y0EG+s@%3jpLfRyGZ35Yqa&Oox_=XuammKT3i?6ho)Q6mCEEt
z72~J*$|=3V>XHwMZllc6nTDuVh)7snbf7fNir{|zWK^T@=5lYrQ-^N~1M)_te_@0}
z#QZL?CJdaqsCzw`e)XD-{ZJ4^@hFAtH{?QL95>|~wmwYVfC>i5L`J!>nDVJeXChG9
z=vkE}6zpHHP$%15Q03KAsVYmppP9F3C5}w9Po6HXbzLjq<v-|Fn~ar-9R&qHi)ClW
z8ST|K+;ll6(Kc}IWfs^loSBRy0ud4Vx&Et!MM{F~ww7b2moa=F?ac9c6M>BIwC6qI
z7Q%e_uZLj#J@WEjT7KN$vwwRJI)J>|<TIduEx)DRUvfreBtn{rRkmvO_}mvC&<dJ<
z@=uum&n@Oo#P7^5kHFm^BY1OhA5`%-O_54Bb6bRwR|nRc7ogZY>i=+xq#dY<^x>LL
z?=f=vz}zUJS1*v7>}Gfci&J0i^kpjPcYYBvPi(LNV_GmHB;j<b0-EA!%7u%6_4HP-
z@1^gmve+u-{-LYH`M&=sgwc)7epv0)K>bEDQP51Y02n)XEre(>stLT8jl}v$;PC`w
zfhV%w#__!~lzCSX8UjBW{d+fBJg(Nry@YE8@U4Bf`3jAQ^T!ST56<2?D(bcU|D{2Y
z5*R>gXpjL3rMp2wa7!rNNSA`r4MPkKA|a`Q#4sQT0)pT$bO=gIBOua^=N_NEpYQXW
z^Ihk6*7<8M*V@}HAMd!X_x*mquGguVN<yJt-2Cvg6^t)p%Gp%?D#x!JcxREoooV(|
z`f1)w!nm(1L^Mx??0>5D&N41fy7p*bd`1C>?zImnyEB6BMOtmBco8PT_my5IKGtLA
zkSC-r=-^d2(K_S$Y3Ox3Ucl<~hjH8Kmz6TIi7$;#<~3GI#v^ubh0j61i@Z#_e0w9-
z%+9@46$j?$qa32i778DkXuCXEk!OAK9mRR7Bm7ZrV>JH|m9GyLke6WJ5nL#Ak1}49
z&TiP=is|jBV^Xn2%Avt&bNU3&1)Y07J%$@rRnTkJQ|58^iJiz7vyTVQgya-{7-UBF
z`^g6Ff==+VwOTBAAw0tC%^TJ~t7~nON661Bnfzy0I|CV{@#WymD1E^qGqa~L2}P5P
zry19xIP;T_jWyMzjg}?9=X34d?l-@{uZm1bmi=d^QxG8X5Y`Mbo_ho5q$j$W!x6DE
zkCUdfrw8dH#BmASev(Y4+f=Ou`~=pJIuU|zX;NNa8qE>}@+MNAjWn-I{+KTC6`SkF
zN0R?U6ZJ^0k^OA@9sDdKKyGmx=m4(%g&J0}8OAtiJt<LVgm!1x88gKZR?}HA(a<~7
zWsbNmzE=rW7Sr|RXOG~|vY)NVjxgMGSx~ZI&=KN$Z==$kWxvoW&9=`!No#+gof#7W
z3nR7@*+riIs4#AY!#G^Jc`Q>Me1);qdItW{Qt8Jf?%>5#6AQ+xaO+*;)f}EJ&Y)z`
zQHRc7plzF6xv`69ZGwJ-tgqjn7XECo@mOR)daJ@&rku@c;0D>K5>r=Y+sE5A;%U*$
z@ZhI)>B;X+8tzxG2_PocZ}yN~l#1g%kTT3iHsqyF`jwaux1uG4LhY>;um%qEK~GVg
zsYfH0%WRdS>H#rQ9AQ#~r?*cPG0U^hw(B?6NY>emECTFh*pAx1A(G!8ym52Q-G4HM
z+x?1mDS+pfhGK&iMcgIxg`iuhWL3$LWYXlouw&`z=d-boO+Kr;bppZ=5&81=;{nMB
zW%txAwVJv6=Q~XLO!O?f_txH_aO?k)P(Md2JGT^V`RL`Q3=^a!GSE2jUMz)6`!T&3
zCZ<}mOu4fgDj#|&KsfnY9H(q_p!ThN8cFV01<!?f?!_119>;_<bQFIj=+$`WYco}w
z7m`C=6<JF4?)OBEepA19Dkb_j^)fi@Kx<OFO;JwIExjSpOlQi{t^P}8(BZ=@ogI$0
z(eXI=+!{RR_s|pe3J$VS%hDztA>Q8km>#i$)3I_xUvrHSkp#2RgVX{^JTV-gISzy>
z?{Z21BkD=vQSm_cYjMcqY{o)GX$WC>!to1A-|4FTqvS%s%nkCVzIMA7rjf;b+Bq=J
z_K;-(%i1`cGzU|JSZTf4(@DGdp7=_+18y8qrM?U@`*r0U?MXZHiegCy;>@_d-57=Q
zmttQ0cl39zm#}FPBc9!zjKHU18|mlD@1d5|6~#rI@@+VX_b{vGHy$s*Bc{2CP2Z6s
zrN0moM_y0=0Mq;{%DPAG{`OI`{KH)()~Q#<7d6*?t3lZVQ6PJOa9N&g-h4b*-!Lhq
z%90+)N59^DS8s^EH__#|Jmj?86Q~i4)Cj^pxX^n(S0DwNJ1f=Vu``7mv>_l7aeG)T
zIw<|HlOY^0YSo}|XvK-1Pak?i`MSUjo5?p*i%^`T5$wT%>zfz#9vmAO?;nvM&zkA;
zp5*n5+Nd3?p#}e^j_I}>INs?lldaXkK%Sw*_D(vu?}2=tDKBI0hPovZ1qZ2fzAb~X
zEWvA#A`wC^ZSSV{+Yhc<g7lSJ-{ttb*873>B|g9bOOU0oiLDCSKXGDTI|#&W0uL%I
zZ%d^T-hu?NqiRe(05duv<ZceW*f=BtrXNFgvt?*WFW*{))=O01A*r{F%CzO3+S)?%
z{Z;H$W>(yH5s=j+mkaF?!@Y{|7`h*-*4D00=m{M>``KLG#hpxYsGS;<5|v}ip;4~(
z$>v4iHkjHg5x^pfkEGjHVz3gqCaOU8vHXR1d>5`(nkc{f1A%TCjFb<lI3^T-&gOfX
zl<E!>`a$3CUQL?7WHkp<A&Pm$JdPS@e@p0-W!;Wgnl;n-u>K;n8t_Pmp@iG4J~HGW
z=K6{oH3G4gpD(pa%YU5@2(Ri+5!++V;pAd*COH_o8O+m-w~o2yP&7G{`rD~4{|!`i
z*_JFdj<YQS$JwK=HoMs{I(_E+WY4YcRpmH+n(L}=tt&X<9ahG%DpMk#Vu;2?NwVq@
zu}DOdl+Ub<l^D&gY)}Unuj1x&C^|wONj&Lye~M$4=i<J$@r*d>w`T=GhqcHEk|_7@
zbp;o$IF^$z$su~fq<vE=-_MK={}PdE-6rQ=a@uq<hCQVizAv;CQP~syp9)f~*V{&=
zW_Ot^`_WJ#QkL7&F0qZgFFchgsegd2U@f^;7y%1PVE-c=Ik_ZDWhR(2h-=yzpSs1W
z$=<BKt~6^8+qzGnxO?0*7&Kr<((QBl>{M!3e@eZ<#zOCJ)hM@rEnyj@vQL8M3ur%X
z2ZE-3)~jc(&`j>SbLz7h^#Uzb&>o~+R=vG^Bn{%PtKQFC{g_~{H+Cnjyk&OjLdFD^
z)a-5tgp16WG6uLyPdQn9H1Ux7@v7Lh1zo+DE0Pgt^CIl<K3(=ZGk{<nS0BE4Po0@D
zV#sm#r}&iXXqFXOZEAhnE%8wg_KIu7b{H^8>eGs3$?YCz)Wd;~FA`O#@SF$jaf^9N
zK1Of~Pho2$R{?{;<i8c#N{xA@x#Q?7ZxK!6U|3Dc9z2`@mr?0B6rY0OSiX7FpEWG#
zuG5<(u1kZ3&>>Hts!!_qK&EqK3tu=WcX2&$!49-gzqa9si@2gWoT?~Nt>d10?R<Y!
z|J~GLJCXvI>KY}S)+AC<E16pab5($-geeu5<_9o=1r=Qh<$G9*=}sC_BE#n^yKAGo
zD{!MAm+Ddcvc+m(12!3VpSW{hZ}d{AD~(Nqo4)?bfB_Cxv>J!h>NR~adT1J?VI7C0
z?q#B~0<#1aA12sNc(dMpZxd&G*D}Ey(WNo@j1Cv!zfOO*u;y86lr(J&g8Ji&nf`q>
z*srCRZ{N$!fNaZ{fajaKyWj~Q#HbZb=WZ&R#oTH)k<i0~jtervPopS<e?>&oT#&_|
zAcesZ0u$_V0Y?a|rOqVmn(tQ}#%np;gb)tX^oYJWE=k;)D^D-0{(RfCC>%0#Dv2f5
zS(v?*J-Jj$m$RMZN-Zp0tifhITJ5Wi)g>vvi@ho0<8w-~q&}J)g<aWC>?C67<j!c&
z6N&sZSh;Mu!#XIn%KXr3`9%{{RZRGY#;5P5n#5~&5y=u$u3~T(FyXeo->xq=B`(Jn
z84IVt?RX$kgV2<9jmK44z`X|%t7TA^QZ%yi%37t6TB2fFt^H}vwMw_hy8NQw-SzU0
zz2-$Ib0`Ir61|U*z|>d|1bw2`_*%pdFNZxZdx|??3_n-RkEQ3JCa96UdjyqIUea7{
z58r57&3Cbr7GGy)s@1%6!6AgcdI4iz-Jv{dc0McA;eWDXg)U2fZKi%*Dk=b)`|hjk
zr<Jz=GLDQPxtB43Ej{x8ReDee8$*A)BJ2l<)(KgdmdQ<{#0mJ<u(P+>KcT|xnGWKk
zf@2lNZj-2^3@{Kfo&A6@jg?;i*1{LNIIoj$DO%WHQfNh-L-5_mq*~+o`u0GBGV3v-
zo-<Vd^sjM!#raPo0{$iPWD5lL373iI!!e34u4{n4$;q&KvZLNPL_-=XQ%?RX;fzXg
z_-v>^CAK&T%3H`3&esUcM>#R^=`p}Be(Yqs&snJ<ydtblC-`y(c#XgROZoD`m%ENt
zzKEr+h$=t=hu_wUq#kOvCAyOvLKo>vV6wdF>)i;pHG!HJSBduN626#dwx}>2FfeJ7
z4Q4P}IT^l7u^;+8d{|@CEty8F`L(BrnTVMo1S9(uHSE}6@JC_ZHy#x%Ffdx1u0d&k
zGdO~_z5J}^?zkrrcIw-Idm?*NcPhfu5+x*+=XR?)Vy)pTS2$}e!w|37O{qJxa$l`1
zAw$h011U7@bHf7+j;TV|nVnS7mVF8s-wZU1>#RnM-dv=F*eN+W2Hk<q+b2ZfkHiOO
zv^!9U%ZrSA*_~@-8AD9;?d0b_QfK>-yt4iC68pD~B}RhFuUE|T*`SK=QR##5XJbpE
z@v?N>kw=LtX|sIX24kvq+dk`MIu08Aoukfg7i+vUza|^%sOy##Jw7xFzLS>t>Wcbu
z5H)i#?PW^KA=z_ZRbBNWTCTj$-C?D)9}MsuY2(U4uv-D8XziyQpH3oVuN2P=H-_L~
z+^25yvu!f+D+?4b9z>^?LCR@K@hvmLMIY^OJ=zujWy&8~L_2JN>C1Nn6O%sk8Cne<
zoQFe%i_-Z{O2>l-GXqigtDoacq1pt_8q&{GPQ_fED)L;c`zFKFV+`W+U#Z3L9T6Ik
zDpVX9x~fyx7eV+r=3qpo{IQ0b>im-Mckc)4e>u&ikj+<ZyohCBv1jLgWQ|;9r7ww%
zGgN&rxkS{mn$z&3;c<Yv;H|??1wx45)>BS-@;=N5@u=jHxzZN#^=bx>(!wOe`SP`=
zNrux?*@KOSM%7O(UU@0SdsX-#S!wHq4H@wp;f1XBU5&ep?mEVxtA5osGlmyi--&cw
zelRsT01YVGobgRO?r9SjEV)ykgruOVziYl>9@6tKg~bX<HKc4qTPuKVVhrrs@;~*w
z(WKq^yt4IxVWVk~P%u`5I!Ajo{=rfb$EB)i=#8LvKWtc<${8kVe10!HTMrz&h0ZAG
zflazn)q8&9d<VodwLoDOl~eU#)vDCed_K76j*I2lgJf5~i-QRbQx)+KA0_RpWp$Ac
z90QasMO+nm#)|6JwI~$Me`&IyBOp4(5C=~08$OWM;hS6mX5%!vKgS$m`6@K?j8Rlf
z0ntsGQ!ky*hKjt$@Pg9oU-6~7CMW5@G1M&a;t}2EkH3}v_4{sS`Rl5wl9$dh8WPb}
z1_X<-LZMC8tg)74JJ?jz!?7JyqJ7#^wP!=&A^Cx4{jw&iye?uFKMs61x+fnMqpVLk
zOeKB{*S(Y}VN)dLHMFBNax(*Gf~Lu>x5oVW??0DanC~?A$qeT`+EYv?j^!J~Gsqp@
z6!AqPlN@c8E-gkM4@kW7nG76?%WRF7=5uMtb2&%w;*UO|J-eRrs)!`Oz(TYL{Y%c4
z>?!fM4k~Q?6eT*tumdl#U%b62wP?Z0P_3Gpj3i}Hv%`3N+iv)A2%TYIW4bRP{=;PY
z@du9nVvJ+UXi3|1+7sA%H^Ut@NjN&ME7YCgLTJN{qnys;NglQb)>iC_)65lbh{Z8E
zjr^m>S-Jnw<4lri*T48ipqZ@<udaPkoxRP{g#M9oqQ~k~x8H-FN$?q4M9#mj75(dc
zvM3aXNBBRS{_R(lL$!GDP-c7TTNYcn-URNEc3X-+o(c&Xs;Isu<%>Wy!-VR`3=JH5
zD_A$)&;He;W!Kpqwo6^Aq`<*-ES$rtiepy<tx(M<Vc=opBg%eJ%ZG?*sMXOPFdzXh
z8Zs<*=zTQ>NSRF;>#i68edvG?_J+qZpf%<~4pU+&v%&!v(TO}udBKO@BqGA?U@HBb
z>Jesv=>V%@JUn>x*ZJ=1)Y|s$@v+>Err0NwPOB(q%?c#WwR(I7%mZEn4O-r@XQvG4
z5buvR6Y$^Al^)SD{*m@l*pPebNTp6sPsdGL;?Hjsr(vwss1a?sWS~&kYkeFU-orJ=
z;iZ+Js)}J>A72;yY*3Bv<-0J)JWIROHd;uaDm1SujaP$ILDPq;_l)m*044%!S}IL|
z(oZjK`YDvm1SA_eU4Wg&s&E@r2_8k4kL%cQ5o)Ttyie5R$cnleO_!{!ix^NY(-1}|
z_rAbQyIbg1m_L}(r}>KOMyA~M#2;@6Foe+HLJHYemF}jh9br_FB+;^nrCtoNU&Kz7
z?q3ON&i6>*o9N==NNkO8X}x;cbu@DF%2^YH4ENHHXpN<nQFgPIknB5R+dnm!ZN!*P
zu9t6SlqaHoy#;>*2;QY&5zMXuObD~$6_HhW<6x$-b5DykCIMGn4XMuRnA<kx#D<io
zv)Aojdh-S~>|oa~f_bzDluoecx=<6&t5i6z@AQ3)x67tg*i!%5p{vhF_HZ4^GS!WI
z`&3QNI$!4lm*v{Eo^sgilTrnR9Ctjchi>UjJU$ggtYXU<CnS6BO>DYZ599_i1JlC{
z%Plgx#5U{2Pn#Fsy<%F+hDsQJNm-E5osvRcjchRE#W~S8vgLeQ6nIZ&=@YryfDlo4
z@B)H3apCQNlz;8V%m#}xrbVS)<%^hfl5fonUu($XDqJHhANoCGCTKV4Uci=j%=hIN
z^LQ@4l%O58B-De1MvoBscpJWzE@=Euzb#%JE3mksIQY%8h(Sv;ht@%2CJroiRE^&0
zIyl@@5-DV|ouQmDX={JJ94}DCu*_JxF9J-N9SijIxWIScNJ5Kr9RB;XJ!FU8J2nu4
z^^xKbgifcjkqS0d2$$Z9cZ;;jnb^j|T^1&wY_f|J6`ZYu=gdr+q}fs{IyBySJN#0%
zb4360^G{C*<*v*5z-Sh4i54^Lfnr%0ECP>RFr7Dwm6U+wwrHul43QQw%@pm(J5wF@
zf=z$BEFZ7%=!InS0Y1-V3n->vFTNwm&FYHMP3!0=C0bg(>m&8PuwWT(0<uJ^n^62E
znIiPfG;P{kgnFED)dSe<zI(78HOVd>fh;X^(@2&mm(b&KYr;wxqBVWom!KJWo=(lo
zuTn{fR!gClsol)RWNg<;Dm5W%qU?!aIm=<UcFiBl*qp@9ot(+$j>IBKxPvWtcN?Fd
z=YuM%!s7PF)F-)zPd+Kl==r5KTzz>nuiVx&Z^>L4FiiZH9<-m@O{!LrD1DE&r+-Xu
zoZ@SsX(L76TaE}eY19mB&_0W#>JOS60@fjifWEWDk_4OKXqJf+xdod|#gi-Qn?<nb
zGC!HFn4*l!hd-~6aNpa<&>>At)vT|m-*-bf7xZZ4j#rFjm7I2ZrC(MhUdDr&%7#x~
zsdzUzG>;ktDxGG^SkxItl$!m59mcTyV!G5kAgxGAXR+nBC5)G>{9M>L1qMy`dz0Gh
z%LznlITv^N(hZ&gGC^hNt<d}GkYQ-w$|9dv$`jbtyY5tk@UX|aG8&H~_8+3mlxMO8
zAZf!j@>~*pU3{C&F7jwX@41OOl0m&k7@|^F1yN7uPr?7J9BpbVp?JEK;w0%u^G}A!
z&@(3MK7B#8dPlGIgD~jR7}lpwvYR!X3*Z`enLwPKx;+<wqXrh-Me<+sXmR5~k$3JZ
z=BxAEBtuKPmN9bOv(9mPfi%u<kS0)EVMnlHVA?BQdrSd@;hH(;nfbb^f_CtQa}GJN
zB-_zLB`GZ1%E7vfua0h1K*Fy~fN^}W>#qs7kttN33XWY;=Aal{(o;{S$Pz=p{5cF<
zu2J{z_~xy3B*~vTO-)~|rIzD`96J}>-*exU1U(f1Yeini)Ze(fy}lj&&EJ?#qiv&(
zn~gvD&r2D|J%tz_>Y~;9_=y?mHMbjv2jcKAWZuWq9_!|oSZW#J44<g{bu(8?toxdV
z0t>5`*edtz4Ls-)=Opo|ZEJ*C1T~=@fgO*gp1p-*Y%^%%DDdX_Y<DDo&>fGZQ0ML&
zr*RZ;&wLdwdDN65(Ntt9m!BlAeL_wOsDYJVYTW~XlVoa=d_&@X2o<|&mH;Ms4V6+l
zRDVdehGy8Pn0!NnbM|ue$UzJbo+3pa&iePc4E|w7#9pYj0yQRAB7F;0l5*-`;DQnJ
zyN5#_N1$z$R9{2COJL(CN}1;!-o1ho87e7k;RH*Caq0-IEvSpWQuDX1ApY=95GA|5
zGFZT-s!8T8^X;H2mQ2YgIGi@Yd?H!l@)%0^XvnT*<2~9@DXq53#qAUOHGQh@?3*on
z14@zIg>gz$T0F5WkXxqj&OL=VOcA|<g?Z;n@FS&?cBsAm5Elp9+|UHblHq_KMUz(8
zMcIo#sNKV|Gz)dYqgTB_X2?7#@m{njjmP9#7A-=F>=g30FQ1ryCnB9viR*TA=4Hlr
zwA{$6p8G=s2hasYDqKFz*Vj~;NjRWi3mzAk2HZ0~?l%Ad2ftOmU1bE>JoH+j^_!lu
z3^9{f0=ietw&#KWumG6Gh-+>T2}+wz?rrcZU`@M2KdT1>cX_yVpqH|mwXFV!Zz`&g
z`*9_Y2Az~051wxueqNrGik)XfF;(tKH>MDcfT^BG4^4YB?tgt$@m6{e{V5P}7G0=@
zD$FZU5)#;wqArsjMD5rD*IxhAF9f4y7kQ6Yi1^#X)2A}8p7Kx1xN%dCX{DrB-t4a~
zp97c5PVD@3`Sb6TQxYi8k#+qg_Zw(IncTW?+I%H<!%@fPq%m_rZK4!ZM7r(x?>nz;
z+?4I4pJhE(Yts#R#WJQus=Lt^s|oKIKeX@RbdvBh0Rfe_K`BqV@u+`Qy4~5i4uP2r
zeR*uC{CsSSmU!!KbM4j{(un3bA3dSOrS8%>N$3*fbDzm=(lIru?BI=a$b?ad+FYNc
z`kW?9R_gTW1JhNKggV3L-x@XBFpY&>vwvNE%ksKy?6$Dd<HGzn9e2?y&WJAZjS2v%
z4RXOp3j}1B05)dD<~tY;fvG+P@S4J!Sy2AhR9}%GFKHM;j%i+Y>{3#oc@sR)&I!6`
z>n=r^tmc=6ggh?IXo&|2^|dBsNTnt7&SU$V&rGRuCMFF<C|Z+UBVDtW@Gz`RcXN3v
zs}<D?s}arR3md^$7dY1X%1#0m8%`EGq+R{m$&=D*`Khf?H%tjK=cc1%q4=bonM^97
zR*k7xr;J(LBE=^h_HgR>uO~uXW_OKTB4T4ZgBv!Q3`P~@hmEnCmN?~R{H6zQ-w9)H
zigi2|*oG_hfXN{q+hg&MbRT>5Se$(3GyXy3<JZJH;Bo&F5+{iEPVI?R%Tw9mw$t%t
zM?==AF?hRI103YT^qk))j~<3GC+6!cmE-V4grf@p9IBRB?vYRB>q_u`>|@GQfAR_I
z)RYJrbD0#{Yf96X|CY}}oawibp(&@twBqWnte_F1@qudw<+5*xVxq>um+nG_>Qw8N
zmp#1kL$o)aZDo@0;y$*n5#NLCd*8>EUE3RCddC9n0PUrTU?t9h9^%`(oUG!TL+ahg
zDY=QsN2_bOjCCl+{*`i$dDFQY5>-lein+x_qZ({|?}^_dWt~P}CU&i{pAwG4P*Q^n
zIzjh$0=gybU`w3&8lU?bHJcSJ6eYR>r6%NPG#>ip?&Jo??Y<@LY#5dfS^-Dj2?KcJ
zKXD(i)Qu0!^tEL+wfsS^s!7R#J2}}84C@b1*L>@TM?TrnoBCLj|ER1RK+1D9+bg7K
zs0ooz5?Ho;@fq;1ApcnQdxU3vAZR}ZMeaMJ+W}n7k1qPb%;ZU~>md`=*<M)Bb2nWM
z@7I4`@Q*DCph`V&lBlbwT1JcHmFr+Tj9M<miQ=TdqL!FP@1=4=sD+S(ra}LrAub_7
zd+<G>PG$d0hfhXGwSN2)fcf`840qzdiOEY?DW)$%l+0c%5`T*Uc}1RV(c9UUlyFyb
zUOwI(!gu$siZoRCb*N4m%Y#E%_&ybX_IceJ1!IiXw--`An<oox;7FVJ9>d1EBwxzl
zHYjD6E!uUfgdME~Q!Sj1m`5v9VDxAw5f^tKglkNmy=1woPQe#U5Ji?<w4c!>*D1;`
z-HzA6UlIi*)RN;j)Czi1%jp9WPg&7I2zgpO`)p+bWov<61N9CrwlRH-s*qvgnD!|g
zPM0nSxMhl=rJII@LZSCbU+Wdpy36$K2ck@?pW958v_<@tiPpauQOX+=`T_!60_pPO
zfSJpWxqZr*evb<Gp4G=_LXOkw!JwZ$4Vo!G>5}=B>jl1{y<uY*j+pu<v?}}d2nN%H
z(o5>y;D}2lh-HGA0Q{v=v-n$<6p2eG$NfA$$tS?L$I9)feP`x<-t4LjY`@5fd+>F#
zQkd$~^@_E$fU3$w@h!1?($B{PV%>F5o5jxrFiG#bio_SM2k``P$9w%C2$C*w&cB0H
z?kG&OHS|`NEZ2`|S2vGn@0MUME*dL$O}QCwojU0x;?}hXkS6&tOT=3DGa1~9akKAz
z{)pEBe2RENxi{l;#rz9e=|#Djn4zSj1B}}YX;-Xt*us@_60h5b?W9T*eGK!qN+_;|
zy4w-Z_LYS7<EJMR|0_2U)}tCKk|8B$N4Ij@509)v>(Z7zNhC6Wg$$D*g6OrsGDl*8
zRtNGZyt%8<@j5P5Q+zuEP;3)yW~GF_3H*M{f;SHyi-U9ZYify43N5XI$xra|$dD?g
zuT-S@;ErQ1;AO+lG85(I14Zr@ws+si*J!;?TlOB9iU-(n3?0Aif4eChqLP{JNHi7u
zsSXI(dLNWENWw^zao1I{G;+Aj*-s6{JJF*(sx&xLkBW6kZeRRpuQ>wy9-x0a7mN<g
zMlY*i0|q^6ZuLI<?ye){g2){P7BOj|0FfHq*WMbgKBoKrmBzFO-L7kD9ulu^I^=~O
zP-As!Dzi0us6cIZ7&Y#`-CFm@7foV8241tk*OTv#HxG7r_V|AkH`wQTCU|?%jlf?K
zY`0#lSq{-;M6Ew6ws-ESCg;FWp6wbU%$<)RubdH<&<d+p9j6#hynRIv$2B~^O(u;R
zyaMP5$>}%$V|;-UnEZ0|qIVHR9A)3K!leLRI4uz&PXnw~<$5~@E7?bHImdV;$HyDV
z@NE$hRvsA00F`om5?=@{;jO#O41oxn=x9(YUYh>m2|1_EZmhEfmfILaXa+M6gpaf~
zybDjBjMMD1wvqgmS`zKcFJ0HV&V=o>ccVF)Zrr7>d92x_=v~5WoLMj5q+X*I@)K0F
zCpS24s3h#T9=}Uv`&FuUH|y@MGEU_idcJV>&=QZoO4Bx|{=%S3*ICeHn*+-1s&e0R
z5j?3yymDD=-x9T(mT~Pp^xWHTKn~=hTJ%a>+*pD482lx<%|9q_E4C=|w-PC~EYwpD
zzGH`a<G9h_v!H?-heyNBbz#H$Dyt+z?VD2Ad%b7s2C7V@iKFj^SJ;_ew&>WgPrAuN
zhVq^0sZdVU`*J1lykQg_BzRZ+UPh+0pG-2DHyf8__s!kBQ;-N=3OnCPM2b%u737TE
zdx!PycYm=AWAu}({>(R^FTKw7aA~l*{9Hnghnv@5_^?dz0ZD{rOxl?Gy~&Cs69FsN
zjt5F@inFgf(MD7bAM-?BG{hu~c);{`L?L^ElQ6iTGqW9Y*$3mquJxd9o7r-+S01om
zqx8RGT-&q%A>ddcsjaRms!N+Ob%PfGL<&vsIpidBhd7mO+!cd|#Diz(;aqh%Xv5$^
zxkNl|ILXIPdhxs^jhIqKc{6Zg)igB?P42%&{>a+xCQ{wi35FVEL`kc@Mf$6M7)J#`
zt3{HFV1o+314j-1D#Z;*I_&H=f@~2-z2?Gc_%e_~Y#5TmM7R1TAUNe_Z8SQ%3N`$M
z*<}2+v1_PxCk3lQO}e?dr<zA>t81zt_z%><>(8!e_1Uw3z5dW^%(`G5E4cyR`5<7w
zt-6NByy%U-+cZMHMw4x6%U{qxn^5c8uq3s=rxt`aSXk$<DE;*8jbfmy#pgTfDLClU
zD@Tmgy(?cR<@ON)rlD(})zxMD|2+M~;*hMYeTuFfF)jaNZz=~Ic)dSYr$3X;J*O}4
z$r^3VbI$ySUXhM|QLh+DtD*$kYQ@$^CR6^bhlalu$!RNvWPI=4@@R})52q6TqD^9v
z8e#bLKg?u*J5ByEoQM*ra=^WhSnH1!8ea5>>ZIFn#XE^Nc<3m9LQ*iR4J9^%hiWda
z&!d04(P~JbP&`vwDgJm@ntFq`9!Y9L@yJ4&`n%%+dIY<C*Pfma#mJ8}8J#P&AQ{-|
z8;x4N9GO7^UC$>R(pv`)nqTYk8W7Y4C+~ldcy!&)bYcWnp_H-1>NjR!z!agq-Ak%P
zn<E<k7foCRgKB>KZo1W+qIl7HHwy^(d7tG&Ok*x-MROq;#p2?sOSx~L0y+jvltnGt
zf63WQ)DL+$50w?Q4%P?NIiEr0ym6W8r3B(*4Rd6v)rzKz)tk!TlYtLT%z|xiY&?Fr
zWd@;m%dF;^Fe#biSibkLB%ANfmO)9B<^@`13cDOilzs+t+8aSsGZT=a4kxs@HJjAU
z>LJ=pna?^gj_Myov_rlaG0k{=W+{d}In;fF);DmkIg`J4wDLz6d?S?9GNBIqm#wiF
zgUsXjth0P@qB9DpPQ!Q2sMRRqk>k#DV-63Qa@TB^hM|}TLCy~b06~S%+oMGHeqvpM
z+gQkCElH@{POPUBk`5~nPR=vbhqbPz0CkpW_7~9|jX`2x+!q7jotzJE>8=|}uRtp4
zshWlQ`nKWhKP@N2ZU;yvP;G<fvGflE2D!D1ASK0I|C%Eo2!nYW_6lVciJ7I`ZR!uV
zB-A-_Q`;gar1>bG^P=$_$9UkfPtOGJO!6edhb+Gh1R!=SO1|^y8_IckD$PDF5^v9(
z-Ge8!yQ<@m3I0b*`riRct4wJ70)*_8?F6miTASLn=@nJ7q<Tf3DWdYey6Kw)V73NB
zCBq~BzHk4S<|F}orMU9^v)tiVPGoyeV|qBOQ`Iye5r$HdO{;I`%c@y_Y{!WM--H$Q
zBA+_qCLTe_&Aqzl)K^mca`;?^3|MiCDsBB{^%d;iPW<#zxL~)fIp!7M$#a>m;0UY5
z$)45Kh&2Xrvu97MRs&oG$V=`(wqgB9jNyCWgg_KPaZ<Z&aNdFE!4b_+R}x8JxGSDs
zRcj&;H%zJg(RF*=q*L57RF_WVCBycsRYcvjXr#7XK&t)LK)r9Ww*Xnm9rtfVm4%E_
ze;L7>M{e9%Fp_}Uw;XrfOv2K~^>3k{ugjn{?ft^e?mMr=)vQGEY(?lRzlfY?f8vJ)
zFd@a$c-6+I7O{yS&Qf%uo%2Qh&-XK^yz>3tsqp1{M%2Zy=tCG<#j|uq#eEt)TtCA!
zki&Av1txhvk)d%AboUA^BbWurOx53%wX*v3iIi@=cmm~8v<T%AxJi@JP=I$q<7Mvu
zJ!+><Y)~hwIZ5}dugop|*2txu^^JjCiNRp{v*L7hFw#Ky;pEX1`#0S^|60^Uk>=I6
zwzgvv-shdVPl`si`sOFyUzFSiOFPanf?QlUC#r!@Ak+-m{*`n%->WgcM4(#2AoHxK
zg^h8YIDm(6^2SzLBIUH>i?j@tUR!xqZkA|Rl)9<h*7_jgEx<A1ZEQDylq7A?>Q|I|
zdF<irGmo@UguY?u5La(ZSqLKLDwAc!q|>N;nuqvX<JUK}4q_JhL~WS<NU5(a+(2A4
z1!-GpS826}<9)rP0hlGlF!%11Y`%8v66SY9WmlaCQsE+``&y!!y5e~AopPi0ywV%d
zXTZUHJFSA96oS9?{PDmv5sO)hj@C@pJGQjpaWD(8@tZ>~U2%<%sewe({Ev$2pD7pi
zC({@r4u^k4vusa-X~G(=tCas(>i;|7*{-9eNcMXLDNxErKcsNElN<p#jDX<~ZOMyE
z{9F%a(wN?N8lIZ&!$OQ!o(nf+%-nVNMtvRYb}YL^!M_W9A=%AK`_}w7BhcyCjMD8R
zg$J-(#_{W|Y%G7%<ifl-+9kQ}uq!Q<gfjV?Ziqek`G8p>>q8^8ZOdoB^Oal=R=d}I
z(*xzVoV^rcUqP_rpP@zfz?;<2LIBlkZZlvwQW{5YGEW>J3G9BNRe5gF=kQZdQ{n>|
z-o0f<y*c`9o;%Z|$cN1MY2#Gy%#4fIgdH_?qV(X-BJh+-ynEBw-it3}TJQhdl2msj
zRG=+QPnv(eo9<n>N8lpZ@isZ|y8u)!(@3ah@qQO0pxcV8QKO!OiGaZHy$8W`Rc2H+
zW2VJZ=iWBWfUv-LCoBJOGH<nEKWkRwnYqSn{17}*QulGsb$G?9EEL{Tq!;{p?+h6F
z$V+x^G%0c=bF`c0I4gZI(@j+{Q(iREAI}blX+54@Ca&5-{dzQEXg-AJo-ELXL~h2v
z`mQy4_;FP&n50pUcD?IIbfuhWGxYbnX>WK>=-%oeC5JSTY0A3xj_<@)diMJ*=84qC
zUOMZ}9|u79fv6IoP6bEc_4;PChTCm>dD{djOnq+OK~&x?Nb59MTKr?GB*2M{@OAiD
z>4m%KQ-@Zv#7k14OnWXppXmmSww{=e4`{*BWGjdFxD+>J)<@Sm+m*TaDu6;YsV#DA
zz1q+WV6!i7JGZ;>S6yEpodB`t8elLfU6TpQ6OaGKOq`X`)!Me_8NpV^&gM2=L3B5|
z>|)0J$qb#UZ*48ySD0vY)>$CHikV)k#||6c0_O)!vgX=%5K*fMQi1r%TwGb4YM63|
ze=qXvB$vppnBSi{k7ZTnYJ;XWB>JE=CSlf#ke^!t^rn%)Oidb@hvAhk9{g65Z8R_4
z!bbObD6!r|aiUZpHknb5HWd7d<c%aGgZS3xr6=(K4Lq!`c2rM>ur;~Yy)~wi<63e3
zcM!w<dg`z*V2_o;wWObqR{&Y!;h)`$m6SqM<!Ewbp>>=$(W0e2_cp5ATI+Gm&R4Kq
zUVhcWHZU3ou%7ILTs7I;<>5QGbL*=e88>&iB!wxJi0oNAzkX^oyNcT5Op@eqQmLvp
zUlD1%aDYLSgkxJ=7<w1udiGJJHuGnFVTi3%vryN87OmXkf2m1dfjkPYK5OC>w#Bw0
z34+r4jsLVF|F+`)DVwNVKR%Fw%iJyeOJz8bBSGyUq-~M{S!s=tq|G5ncM8l>Xxbz5
zrk^<E{mjZ1km5&wPk4n*#h|AM4_?7X@jow{_u_C=xlGALI#}TDaewg$u;b!QVK`p8
zqK4JQ4|$XyY8RN9PI4%42}0^kW7F$1p%u(Lye$BlM~*m+D`>v}pSC~Njh_M{!ah(7
zRp6`f7nJbD5|8pU*fW%1>|31#kd`Ch@vYLxEeVTKC-Fw?8-}7ZrLiMNn+e2AQXKF3
z_dcApfk%T`6YPhzPJ<ns@Os!L#@rfFpVeL@R6-i-vj|GedGvB?deuTomys#?4U3;I
zR74TIC}_nk^+(!dV^mB%l#zzQp;QJ}k*}^{!Q2bW`97Lt=%~f3x|77~MK7SkBSb)K
zp&aRkswoCXa5VsbFcI3TOL)dlrnDTm;ruG00xyh`<KU;g(@KzMm6nBaciBu_9fvQz
zR0sR{&z;Rqd}+ORT@q}nBXG%`@}aem({vCaGX|*S2(Q6A=MAc0DE;Vm>1E}cyt<tD
zhCnb%nHZq097PA&LkWHGHOa6!MDyO?x7Vw8{eLtVoO+_42NcTPd(3}#?&FOpWUIoA
z%hMO@Cr))1x4V#9A;hm?DWhkl5#pTmRbcpo5noWHvkb3c_ubPR;QqZ%EgVE*m}K_&
z`3t-zmeip8Ry3XE3K!0ty-EvN53O9b5pULdp(2?68tNahoNUQ!Y!q_)cRAaOClG34
zOdDvXkOcfSLnr~350KYjse@HW7QV{>+@C9g>1nUe4_l~6n#JjjQ@eZtmtt@?gYEQ3
znso?K`=(nSNEAi~Qhz>6RwS;#h~!#A&cc!_#`kauZUeN}2RaXf<*(u*R@qOK3^Phy
zr<@SJ(N?(DSx*O|nZ;MQ0%U(v1l_dL78g~bEz9Ff6Xl_4&z6rQCvzR+BSQ<1)!q?m
z`h*FSk7^BXOrf?*93E+3!Nq(APQ8bJoO=9PgE9Zx#%Sk=!;vvEZPCXqWiYaW5)@8f
z*HB!K9${HF7qQZUn&QJiQR?GKFi2~Hqr`Y66PEh)-)w_2GF%eJCr;GVE{upY6WNHR
zWVM7Mx)j+&#FM4qzFFIy1h2%Xpjlh?geZ?o5?Hh~H7_L<k!hb!WX`$8j)ZEiXc4V*
z_wKx~^~rLUgpKW`YuYV@d*zp9@C5y2BnL8e0<mj-fkU$1<{BohPfrH_l=2z>BOW}$
zD$jCo&Sf{(pPiSS)C4B*@NWMc=LO(2RG2EgOhB-wd`B$5L9?|aFBe;r<OM3cBgzSB
z-kS59($FS$3+;OH0~fMmLXx>5#rYv41v>g{)bt%#r{iv`0cYi{_qN%^Z9WhSm;gkR
z|Dv?f8*g~Q$<f}NU#)9}g#mrdBkh8Sb3G}J-rf(j2e#99E==1VQ!uWB*m?aWb>M57
z{dy`QD2~w=I543makg!s3}Ds<?g3Wr*=bB`l&`n>(+@7zEtoz7|9ZoZ`%M1h>)Wrg
z5~A$)j%0Nc77=qqiI#p-Qgh+Hl;5B0k|mr36k4Pj#mYtC`Q;s6z4BaKn@J=3WxtAe
z>c%dGY3$?NUhVxE6dB0_v}A&F2^sRHoBiAR@gI-G?mNG}_inEddZhG8qm^^+b(_sS
z+yH1uU~y$)!xzW0{js;kx^|zu4vekDzwDao;iz`&g`ueU27|Gub{hz7jSLKccARtj
z;8|wNy_l4@&^|*RbTmqCsn6O@wcEz-{v<5(cWNa@Qb+nt_d>c*_p{mZe8ah+lo70O
zAbig+i$4UUa4=Y!4C9UeiE90O{rt~+*nc)3VeHiHH#m(xnpoMf&-`}}+`i8tOt))F
zhwJ&&XZCVpP<F4c220!Iu|s!YTN$fe>{?&Sh$~0`qbmQu|IZw7@JvQK79dp`UIMWj
zGZB>X{zN6yY5z(3Xx~(|y0KjM4M4b513(rp65DFY?hmPaZw6dm#f@imbw&YI%wau$
zSWTAd`O5!ibPKlf%f=zVw#Pn#*{QprV;G*^wE(@4H82H|18;G7(-qyLi~sy{fX5OG
zf-)6(V0fSEkZ9c@_Yr$`x?VoTgIF?p3<`m(fWg7!QTCbY47^A%W254FeT~CG00SUc
z(D+>r#BDpk<M(*ewgUD;A+aCIE+eMA`~TaA;-SSQ38n`F%z6F5U>6hr0iS#t&!|Ie
zn7tB-xsmtmvh%-wBdc=~otVJGgX$1DX0K1w*b1xus!;3k0&WwcjWTc^-x_d6838Zu
z5>FKaijPR_WMP|VoMV8RmQr+`21fBtJ((?9bUC)e%+|qj33HYgUjIM1g#Sh||JM?_
zqB0LAL0>Tm)cUTa|J2a`Bbfyi$c-%50B%iL??2J7zt=`K2ll&mlI~+;jF*6Ha=7FT
z9Fez1%mx#$v|*V>5A(0#B9?^e`b4d}=!~!S-fTOMET2uoV)ylM_ND*3(D?uHt3E~S
z&xcRepY|sxfNeo&sgFSlnkg7y9b&Lxk?)Ly{^~moV5}7U)*lkgc{5X)+WYJNE1x4n
zlND_7#*6(3E~7(}tu5&O_kHzWUy6u*o}gWc<Rs{y#cD+a{k42EM9R$bBcU}|V`NzF
zv7=(VTwGH77<&ycu@dTx7XXX)y(5F)BOu8D?)cm1mq%^eC3~PFw=e-+h|Fr;!DoC*
zS*q5PevfDdTVn{LLO!T2ea)Fx?4&;zcm!=c{0=z4YQ~})E5!n-xV(U-Y9D&6>HpKG
zfJywTRRpd_U**((78YsIi6vdKK3!w4pl?Jz1Pm*;w2#wGKA%%x0_F$LU4^}xA8#t6
z1NK&yvu=dEGk+(baH~rq@M^A2`-RW{-pl<*XZOE;a1#(ulRTb}=)`hjJV6AU)&mrl
zHJY3_Ud!NXxcG0@eug+R%kIsPwEXm|CZm);E~IX3#DBg1iAH}j*a?z910JBP!9z18
zmR$pTf`HaSp8+UCbZO>;M@!85tH*mO$$?xTr%1=Sx(I4`G5|)H(+n{8>Ax-Gw~Wtr
ztk}4m`4E<bCDUM?!L0mMTEPsvbJ!NE6$lYTKS))F%jLG5Gwn!<Os>m=<YmIuuaD&^
zWHxI&!O7uQ^a1dkIspWol|uGc`#^oL!erIdMeFU1Bk*OkQ5O&Y&tK-xRuk*(zqj>I
zlGT^}S8Z}oA4w?IJ0b}VlJ`GP;y7zwan=BpF&g`K_5DfP*-Q~d)T)At6)3w%{@}a-
zOU9x36!3g_T|s;QjizpI6<|Y{IQ_Zt=0RYCWdJn!*v%DF4p=yxj!&gOv!QLUF%QB1
z;wAhdS)=p(lRn4C%k&k&cVM-u^c08~b}g(b!g4311s)~4dwIU9Ft_j{-ifGS+ILjf
z`rXW9vR+u7PUt=$qzGKc{Y4pa9APRK2<R;xDH^Pd`2g$nF8>Y;=Mv<1*FH7R1)mXC
z&ar^xd7jHfFv|ShS&++(?Gpy^45zbq|HnO)or8VBbjZvtpZw?KTT`9~<BmwO%)GU}
zf<I6v_{CShL4D>hH5S}h87dq~;9XT?Asq~f#vxE73Bf{!ANviToy6$}a<l;A(_3Af
z7wJ2W>4BT5Q9ay_dC(!x+@Ig4tEg^$`W2jrG7}*nI21*rC&_LGR3O}=j)1Q=aKR1$
zaFVVd2?g?-{&An9ysZ(rS6t%MoUCGO<>C$LWZSQEjmE)eQ*-jSaBCR0`TUBGe{$uW
zF^FrGes}izzk^=?>zKe^6{6V2$}SULj``>3{P&&O9ShhQzv-vHDW_iNW~0IhIjk#)
zV)AIIt%%O}636!Y__H+_cV%uz6580^EzLE4TqXm)+50=~ZQK7WuK!lH|MOSE_po=A
z4nMzs`9Fp0UB!9qu4!!U?bH@{OMARQe|h{_fynHf+Vgu1T*2(8QgkE08*iP~<sVf3
z0zfO-%9dTsWBGxQMkBLEPwQiF!ZBhwYjw)?;3jzP(hbg!DEbd;YPKh`455U$M^C9_
zv2-p53LqXPTa(FCQLPN*xwx~QaW8>5uTP;)@a0}XZtf@Sbv4a1kR$e4Sm=QPW-;6q
zmzuCwJF*M%5LO8yyw<=PB*8(n6HYxS2($_7JiqZPcLeMIVpUr`!k&=sttSqhyDwu{
zNwL5t^4_cuoA!rHcl?)bQ-Z$fNcyK&!A^o>B+x85+gQNAQDAaX0v{urR?Z~c_57Qy
zHTr{Q`RuE0G9v_BNP#AR#MyBV!~>DW%~fA|G~z=&W<hhjK9*#4g5_EXZ@o{+o1aYe
zjumk*Rtcp=VQy6TRxSJSn(O%S<&|?DoC3j&+=RMVmQP}G0*ECJKOOlu8ak7M_lxx&
z^0>EZum};2sF-WvCRm(cj2|~CEL+ToYv+9cZK2_Q=97iNw=v(n=>iS?7ibEc{37M8
zre7v(-aqVbIO~)4E7}t6-;*Ns68az!w>Zdm0*tg5LBprO!Pofv@(e%oGB)J${NxIr
z%nAg@0~{9vlG+cSg4N`<6(?eK*GV+^YW&&nx+ko^8tA@UuqA49+0>{KF0#R(q_MAP
zGv0)gQ{fU)mvTZm3`_SEoec@q;$wZ&t4K5hpcs{On~&=I)FL8$?AM1FlvaPARoUqs
zetWX*GcfyXpm}wsd3@Dp`<vfkpU3I*xysv^3mdl3Ui0bSMolxNGadI2q1G@j!0e6E
zU=8?G$vIN~hj_NXYM<2`(eZuO_Zm)a8K)zQ+f+;#<DwJ>rfjMA?2gRfagxX)mZGMZ
zV`iUm-&Mzoa{SzuS*3|bXBG>^yPgS7i47WTQ@^qC@ybgNfp^=tKc3qE`#x4@Zx4+2
zqVy5!mJ|B>)J{umFF{SM^&yH{y^twP`kOr53f*b%zcIea+WMBWo&N}1DlubA38{eZ
zBLCjm*V#j5Mpc&<yA*(5>HEcYNC<f1XCaEl7a^k$Y5RS#-6`4&qnSOZdJ;t@)Ck4a
zwzz;c*NXULarAy7%B|sBH`)Oof75dhcm~OU@^5wwAVf}N<5bmPC~hM=YsK4!=cUfz
z{@MDK@w)n&IKw$OncsPQm(KCP+H!DR^h*jFmGUFdAC*tkncQeSV@yD5srI6S?B*>A
z1mABh;*gf+wIcZQhbu=dt8(ALE<G7TuaSP;e!o!_t-z%q+z6m!08WJ^btPTNdIoFv
z07N)z@JG37pGpfqr#JUaX=;e<J4BVs2`at{PbbDrXjI&kJY6e3Tkx6>BPJq&mN0d+
z1LgSZli%1Jz0c}Ez015Z;}lrvr}<jEmcLIY`VAqHhWYJbja1@xN!USvkRv5Hr`u@C
z!mnbs3qEErHZPiUqMiIIpk`F-euRDKy!i=c7H%lv&$~S5kn85)7v{((1Fn-8>L5d@
zs3DS_?*OYapOQZfgu0dIxlgs5*eTAh5uNOSaOiDUz>`wZV)%fgAuD+>cCQl~_*-I%
zEk`(cSwG;Kr5@7<jGbRQ$syB|6eD-OVTneGuhQhdlqT^=6h-#GW3LXzGI+XbY2{V>
zRVd}g78IMx&F)@8BJoi1n;8rD&-=eQGzij2?V+;8lX9t6N1D^V{=TI}9H4y>1*T=a
z5mgHM03zU>U?PT?y#YNQbop#Pf|8=(_xURRlGminB*mmGNd#bUnVSSGXTS^AWqwYE
z>=Tafavk8dT)y=P>YaR3573L+PHd+`F_rz#y;bO)=BmJfi!DVCG(XHEPG4dlT}}-&
z=Zi7ER5(-r>=S`4-V5Pnk3qSktC4;{^kG^`pjnSV(@dLU_kMk3pl9R$$Q?m_c8RhU
zJ6nP!qwW~01BIieK~n~Vj6N{abaM?F^oA?^nR^AOixgo1&fOzkz9W9|=>G3nA-fO(
z_V$(wFgx;HYev&gK4rG2;OEf&=CA(SEWhqcR7|}?2R|v(#aCYu_Z*b>?L3nNCKt*!
zM!YU$`|-Z6$J2XtIOpr8VT!qrpE0-BoKOUH0+^G0(mefhVP0+p$n|zec<j?}79^rQ
z#n>KA>^e4ry|#Jq>2EAnpB(}8h0?qhUdm6P+j^FM$=|=)hUg!VKikBSA}*pa`erg@
z4o2~Ij376ocX9E?u~l&$9*GoUW^<A1@ktQyVxYEF3>VH22&jSLfGu@y)CvIJ^g&?W
zU17JB$J-5OavbxWAvm><2yk6NVn(8f)yTWXzjNUPbTJ;1ETB$O*M@aR+^YFAa63Cn
ziiFJyV<TmPWuK~}Rd)PAfg6dvcDg3E7<yx1MZht-=nzJ+40hGDwVgZ(T}9RVH&OjI
zIN4TBq{^%(nFPoga5c4hf{J!>0pH(-{Mt1fqbuwWu;nWRVokpcjHO}%AxpQn3x{rE
zK+WoE6xw)*EpC{DkH6z~Xl8`+mZ<HfqU3*{3`%gYo?OZNRDdMqFVlWs!WpIUREx%+
z!{h+}9-0yX-G~e<G_%_I1?**_3#(IeAOz73tQ}A_&t#<cq;vVVBJy~07>@FSj<P(K
z6a^uURWY$DVYU7i_sc-XHpLmNBQY4?&8En}=vbvDZi*$`Pu;sV*Tzpk*=g46ff)tG
z6-&d+<A-&6*$X43N!NJ%4XbJ%oPdtDbMlqVv+x7HpQHIH9^m(W_7!}xx92622^;U?
zH}4<3HFI;w5={})?{#)nesO18&*t|i4-3EC`jauyl>8VR&8cX$+vo-p_tLG1W1VId
z85B6ZVgPgN^Uw055s#!x&ZsZT5(l`EACGZ%&j4Xmid$0?pK7I7`6;{BNd9n)Sbpdy
zG9H+Ytw?Hc6ruAeC5$qegX}rZS8f)m4W}vjt2*o<qlcpkDU^|F`65YHX87OJTf*0R
zlxdhqF?bY_by$Gj@09Ks1TkZXB46Y&rs&2X3&7i{-@p<>Uf`GzjSl{KenLm$?9r~U
z+27gf3@;aSWJUW!$bKhWl;4{yvz1<y-JAjlxO6vpTQP>91-|eo^A(HzauOd^@pEg&
z4+A5hTNF3I_B<bYT|KRn`u$y}MV*%)N!Kz+jI#!d5g3|UQFH_@fv|%UBve68fXZpq
z9bf0Ce1h)G8pF4%GP$L0!F>*W7-x;&pNNe2Mbe-a4ZQ61vJ~LH6Rw~}cu#cwPss_S
z8V)dQ)kZ@v4mR?A^%e%t^!)H$+ssiuj6u!xF8sk}B8Th2Om7fJua7}w8~;abMpm`>
zI!R2?Q2JSu5ZXI8NM-z_NK?>&tG2aJEj59oUM=PHsz3_f*-yCdH8sOnB#OqKyL_9K
zmxEEAU#v(|-vGXt;!A$5gVPdqrVyV%e>4~Jn=kdRZ8||x)qLvO>L-c#{O84*b_SWG
z#nb(VGmDW&s>*N<Je{U2AvA|Mt$|3M(Ip1Yn)XV~E-QVhQzFMRRm@#eEJc}bL*Kxz
zX7}xEK$7o1-q{?!se`CaZ@|M~p*$BaK}?qyZ4+-iB3tO_BK|LGwNp4p&7t{f<TH*7
zRY9bhX+$jY*7#9EY@C5z_HpLB7VdaS+R|6+%>|2*+lC=(?>7m{qqr*uIZ6+!yr2u7
zv?;dQ-}EW{UKb+?oSkV?ytQ+5zu$?wUZ2g#AL8d<?&xzF{(&ThllWU{f1d%5_lmQ~
zu<nM&&Wx|@l?(kO-&=w->86u20D3w$`8ATb4JE-bt4{4<+cZ$R>V3q<KP}XT8r%kq
zU?xH@KtuwcwQ`0AS9YYEE=pQxqhQ=2{uS%Gu|`VS@*8G{QPJVsENO-s%i-TnWt%LL
zuw4-CFpBay*l9BHg=}gjTg~};PVb!4?CV8l{TCAr;D+m&7vOpOraK^n7#%WJr*9w0
zZSwwI6q!;TchI*|)oYb`nV=Rq0zWSUWapY}wb#6$g;Q=4q~L*7^$S3VG=O|B7AH$O
z*G;+LFhNM1i^v!0v-q`QVo*)5j^wSK7S%X|hR`W*{XIu8Uwmy0aFZ>ctAi!1xXo3~
zGs-~od!LJQsBDHuuIWtJZ9H)j1mo`tA{y)alL@aiMa^@r8QX>V!`?49^c*2)AW|WN
z*rgjLOJbGR{@<%ummb4jr)Xx&-l%Q2+y!}JKU4MN(4<P~QG`{@!b79e1f6J(a{w#Y
zG-=f+q{3nMM2A~$NyNe*Ut>y9$@F+SsAQB{D#z*F^G0v`=?M;Aw%JyW4Kc?oR0}kf
zaa26&<^-W;F$*Mkg=RO5uaVn-<Hy{%yS3^=g#!;$AyuTt-GNcFvF9pjze;0X=Zl$1
zC+IlyPGqtt6j1hkbq(Tgrz?z!A4#7=KBZr_5y6!1r$6f=Yq|3wL!fJ68TOfqpWWqJ
z78fGGo)Cl4Oeu`BVeVMq8!1b|kgARS9DE=uvTT?sRAhE5`C63g;4SWFe$HY(>2Y4m
z&bg-s1Cd&5A>Jj$&N+qsk+7@Z-)Lido3A!PWQ6vN8ov!L)pPg62^3?K`267)wkK2j
z#K?GCx5ZaXWY04zb|rE%<EqxP)1{vSi-X?O#<Sk9*iPBND6%(e>qRZGYP3@B&99Qm
zmmocVO}XuldaC%CK>k-4rK;(O8_Q3=y_WGX3IAFb>Fa@y!sPxh#@;%v>TYZM1tb=r
zbcZa`MVEAUqo8z`NK1p#-7P6dBZ>%%?nY86X#r_aQaaCE-uJ!te$IK$`|iK+gG<20
zZ_YX97}xc^l;0vKsh<rT8;Xtvvf}jKi%+=EKP6tTuuIk~A2@h}!#3c&|9eX=prMHn
zsYC#m>HMd|+|Io3G-6)dxUSS|E7kqc!H+GXp8f7(f(}+z+Tg3PZrQ29lWxmUg>LWw
zJYz=LMbV|OBmGxd&jcNr&{_aA5o1>)ShT?F{;BCc!$xWl#Z$9u%xYrqg@5W+&)8Xn
zzl$g5#aky(U)Q6xiQpFH_H6Eh?opvxUx8S8c86G6SMdDKQaSNdT>Y_9x$ldPs{x(X
zF~c2uct6t}2}KGkE$8R@vK|iwO6%3XgXTd=y~D>#$gZPvkL#!$DcbzK<Z^IC&l}z-
zAv-!=86PI4p@O=8p;GyUwUVF}X?u^goIN(3%w#HNBK5F)Vg%>9bcX0!79C+arS-hf
z%6mm^tmuxDIM>1g(&^j#5l_iqs431+tfJJP-9Jb>F32@XS#M2#$k4&}@Dy>?ab>?s
zDG_D-4mTNKb4+lqVka<OY&pa%_m)elJ~Z(dHSnwL7ji65(|eXt0T}I>gXRn0!7XKo
zk#iBha@3G&#Sts*+fwjcEX$U@g0r`(=nFmh8`43VHJMbrVTV-&Whx*m0*z7adyB`P
z2kwoxes{gvHLUgP0mG4m(Aa3T>d3R@#(Qb+w4`4t&gpy$k1e=;7bP=pS6^vj9?2L~
zHWPcqvX;~s+mr$BAA^$%Q!Npf9ad>Z{63p9G+2l+-*=0apK60zdp);e&o!A`3N19Z
z-?6o!>%HXBX@egchV^{0>K7OSO@WK6!#nWj<Opw0>*e>oFsa=ws~NrshUbnS(8Y9c
z7XLbZU!Z2vL&u~Nv>dan`TRvfG}gTxIvp<5r`#T1FLWOIBy+16Wj4lTZFf?;`DzYm
z3B`>E$_e54iMK$wJgr5e4OB@Lprj}Z;cL=kUHy>(kN_!?8?{AYdk4A=PcgQ?p1Ht&
zIYfv=mcV`9e;&eId=$0UZwg@DVTmmiauv&FdXl_JL?K*?PhAWwEHU-4;Bq#tpr1RX
zlk|Wk$j%)zpK6tZq%j;*wsDw$j;)WcQVdHfDAc9lC123@+*76GvRm?oJ&?Cn>iZBM
ziRRTW?ymfkjws3mBCCF`J`~dgxy}W`r87h;u=&S5CkX5g)JLlyhIJ)ng+iCbW~7%R
zMG+jJn3Gfo_pq{*kHCb}>8Kzv7#odis2j|o5jULPn=*8b#V4^!tRyXCyA1tUJImq&
zqL#2+Ue7e;;6S@1UmAe+$5_#<2@zkyzUBEy#HSkE+Ir-zm{92kOa7$1vM{nrg<a}k
zHqEe=IFT<A0V&a}pPG3PsQ%2u*M{y-Nr;0DH7IgiWWHa1K_fLZ?Us#@abccmYK7ut
zCbcwHXEY4R#)-09%+>l4b!CLyJWvqMOpP>iC}wleZz-okiYvH<tt-u~6r)NuR8)DO
zNs}`inZbcrX6`rqOz78WZ}*r>fZBj;xtppqwHm6V;<tq9532wN^pHoQ@Dg4I<oxmz
zRPrdV<rdr1Ll0sJkp^KGf91)1$2Tpw`&%KzFV*{w>qi!|Cls#Vgqh>?5MZKkldEM?
z{*C-=5dQe#Oe!_ur*YlS>Qk^9J_WHY%r*^~<#;V)&b{cpO7iBTq!~!!rCzI05W5AG
z{AbSUU#mtFfXW>3j-S_W<h**B)`4L^kkyxNQpsEfL$s@wKE*yHubK#5bQ3iuRZspY
zG(fy)!3(p>Rts81rsVCXA8?=To@OvEFg*Y@I{#zRq_Wys+hS64s$&Ey?518p*ZxXo
zBiXgf_=4lN`w<LsLAV|(C-n(QTq$+3_^BTBQOOg((&Us!eNGc%UuFzh?Y?U(STWCz
zY0jqzFeZeNPK#z7J?ehDH{P3TZSjQeH@2JP=Jg0+ApLQCqt^WM4Dq?zk4k&L4fs66
z?<$HW<<8}Qe?M0pi4$FoXI)`g72oWrYtS&J%$iZo8p>C0GOKr(#PcVnqF9?>P5h_c
zSFBCX)kI6SxAzsUO<muJj_B&Y+t|chqh{d3u>P&7r3Ygwx0`XCbKLmV)%S=$oHW!A
zJ$pX){AmO3itl!!sL#=Wlh9YM;&R*-<6Vz5t}Ml@tK7Sc-^JWPm}9o*g3ZoXY=CW-
z6j3+8Fb;Y2wo|B4FGpcO(B0wE^_#}*<5`Ns(3Qf^)`n8o(CmrxWx;5_<U?j$bu%z(
zysJRvEL(K{CFpfjdHCy6+a>w0S<j_9bTY5AUK8)IUMo|4IJ|3ta$$DSjz;T~bwePA
zFHCr0J<ECseO=@t>T<l{lXc*cw0xFyfVr4w&&$s!o*-fLr?m~E2Lm7;x_^vVOAxj~
zNKF1J8`DQ+TJ7eq2T-`q`AqSKsy!`jWdd@f?Q1P<?j9B_J|Yp-`+A1cV3gPkM#kRp
ze8s{&G&VBKgZU8IV$j(fNn#ZUlX7(G1Xo4zC&qC>f%6F=Ac(ExWRwF{5~at59!r`j
zS`|`RAHsvw(~dy8SAZ3X-T>QvLQ&X9M0~cTfA0A>1G<|4a6-KTnPbRB9LbgoQlnA%
zYXAm>>#}F+%o=Vax|bAZE=W~p^jJ!&1xODwj+GiXD@-*|f(ez8`ke8o27?ac4y!?P
zG0o{)&R*-~ZrGEZJM4~qub)nXNP>^|&9;71?|CVlEt9rA*KT|qk>?Z2#X{Ortjo-K
zu~t^lj|IibQsHcwV7u1%Tq?mPZU`KDzNyFt40Q*rYW>DWvqOef`F1(lffI-wA@kom
z!a5X~EH2{BbE?H0m5~aKtTm33-(Fcx1HEUsUuvj$PWT&sRZz~jc>S{->?k7{f;;rR
z0T2@O@vhC{Inew$%6dIOI_X&ctIV~OS&9?ct+;vZjUR$c;{9_yQFrQN)rE2hS(Rx`
zx;Hd+T~E^T4T?~JH^sHYZ^wuW_s(gA$sA(OWUemT3<*Bns-n&2owB^qNqMsc3!7U_
zzd1)(Cglv*of4j@3;qaU4&hjTF?3o<g=@@~DU)d(SOW|1SCOWpeUZH}gBs9t%s*y1
zq)F;G<^05`KDc}XrXw)_A~+VigU|(K1<Q}tAn%cBtKy8Kv@RtY_WI$_f!=fp)f4%V
zG(4I<r7MlzQLBpSHUK)C3FGTy5JLM-5Pd#D-{<-u$IIMgAGFhb(nUPWjVzL3Rx!g7
zS}2jrpgVZizS^u6w*w1KwiN)Ayp-}S{3TO9#AttHvBF_vT15kh`DnF6pGKk|*8x4X
z=m%C)>^!uu1Vi@*#%ez@hds$@DzNDOiCjp4n#6pZRnIHHjEzcF(Jg}BNW--n&riq(
z#+o}&fv&uIjcO|SDc8d+wJe)p&cz4RtGvf|mI-!n7D!xFb}x|gLSw6SzdX3bhuCkL
z(mLi57MG`8#;251Sv5H{Wmn-^<b+w19pdM`%^NQGf^!%qbs4heenBySd`-57cWp~Y
zKkA3*$dKIk3S|FU5tus<ijVJV-hA6mSy!g9OBOVj>028%8#6wV)M~#q`1ve~?i-E-
zmQ-SaEV@@aRy*Eb;MpJIq~)!HF<w471i5Dz$8SzBIVED+Mj9mb)#0ExIa&ZIc?N86
zKR}j`3=n9y*WpKwyNk6}SzbiHiN^%b`<^*Z@&fU3uD_X>6_~zbBF0L)o2Q)xY?V?u
z7xLuN&riv*M@p9OMI1<S(~J!^T>M6e)%d>>a#$g|Y=FrN<;Yh<-{WumH-*%>IRzG<
ztMCey{wi&wz%Z;LB8=u=5Or*BlmeJgbU!O=chu<^Vmo<Ryu;aD0F;j3h#c97o-&K1
zFXdIdgXc_~G_9ScS96zkS+QdM5;n_2wPIB!hJXyi2fxG-rg^*XC6l2#o-ORictzaG
z2P|j|gfc`}Fx~X-Lk!wdwyr)l)y8xA*Vv{^YoEM&SXG4r;b@_buO<0K0D9UMGeBre
z<Hiw`5h*TgmY@*08?hXUDZpW;(On=}PSEEAi&j#}uIYJy>SAJnk7mxkDxC5$#o41H
zKS2RG?2y<lRrE6ovT|(xYIy-$F(PZAxKA=L^qWK(+j8y8l~@4|zn4qgYeG97g}%@K
z$If8(?f<+pn6&e#(47KEG-2Q7(LLBAE%a_O;)-mX%#+AUo%~p#i2#Y#W$e+oLc%Al
z&Sg|zTG@l@CC%eECa!-C?ViUYxjrTF*(x}*`qh5bpj_ejq#ADI7sicExi`Ik$2s7f
z?YD04H{G6~^gDTLZ(P5W&M(P&S?72c4?pn2wWDW)@tV*XUffp{Nzj54kT6;cNB!Uz
z>U}Q~y_k@)xe+0PUCdTNx(2KVcR_Vfj*pejTbU6hH^rkD)o|ano1u_$upUtsH%@HU
z^`{qcES0YmH@?6y+S4z0Y&q5*#Emfj8Sy4e9GO+Zp2(sMu+V+F*aY%rat&W^d~kOL
z`vge8I||410K>AZatzJ8(VG+TK&jjrRHBN9J0RG7r%yz!FT&kfb*l7LAu;@pwjZ>L
zD72nYB^1va0RmbGn^<<G;@z|MyB8O;Rr(<zc$zBeGjhb8!tjm0CPS2l)J8O$z;mIq
z%5yww{O%$|2Jwk6f5wRRXsVR7B`@ZVvfgYhU(bsHSI-FH1paqZ2{=j;7hkeHW=qqi
zwtV|0@wi9Ba|Q0G90{jLK$vxxgY1LVgYQ+Wz*}WS`yR2GOX)L7(6R^7>~J`6kc@CN
z{o<iFD>>UZ8`hK9-{-deLiO`=`(0}LZHaW10LI_4M;{ZK-t#}(fX#u6<QreIt)egE
z>$FF0h4+%aLcf#vlC8y^An`B?0^@(5g1EU=P{s3tk;+8R&0DCJ^i@F)@nwZ`O;dMX
zWOYC8_995-b;t?X3}#la?=t}bz=*V*=Z%hiW~L!uk}vHgpT*%E;8MAy?ohcCwHzQa
zciM{`_;Zs<mSZ_0hp+J-&&bB3JwAM0%Z|B1Q7)rd{IMwZLuT8!bzU-?KCY_S>V<<r
z?YSSf33sdHE0rdU`_->vf-Aoag?WrA`1kA5)^RZvzqK5aG8jDFY9kY<G9&U8_I-F9
z)mpFGIQZ?YjHK%1J8~1AlUy2Uf84EjuhG2<+VEO4;HD@wdfOM97>?-QW8<UPP8j}s
z+zx~&&@>5+-BWq>NRJQlJ2@0*%>x-BZ~tKd5OHcK_Xd8gZef;_n^@(w)ifXvNeR^$
zh<<0L1>8P|8C0Z2S5;1mO5}38(fZw+*r{hq8RSK>v9o5BMKQbIrx+gErrA#5)V#y=
z-h25C`KvM}zS0yl@o;gO9&Ii6bMgFm2&b*I#M;Y)AR;!BS`QKu&eOOewg~vehzRF>
zJR0V#{FI#6R?^8o-eq5Ca9~7zI;K`pvhlLXU+`@4N}{Lk*%wTu>mFk-(m$UZaaC(B
zfUs+$%eh3%dXvf#M7az;zAT5NYNWCiK%AVX1(+XcdQ66WTSNC|Dl!EvL*+n%BBbi2
zLc>TGBfKKhSWO0l%=fLyO>xRX_%)kkRvomO#?`Jow)mzn`JVbnBB~lL@}h)DcrSj@
zUDgFhH&w&1ZyDwR|I{ZW<H>PN6Y)g7|GEU<*x}Q6aIK{s>Q5oA!07~mB&3C0do{!b
zt?CAOJtgmYqGxqBRch#k2WG>LYie*w93QglgzD^6*ktr2#wookaM7|@Q`aULOgQ>t
zCDwjXyAawL2-(vJIepg?s;Fx<O6Okqso4Iv!r~?(FAUMlHAbyYpsrb(zMzlMOaJTu
z)W#Z?VXZIHKSkUnl?#ieBRl%en#1KcwQRk1Wwe?P(^Rme;#O=dws<_<wcB6QGOQz)
zv?(@VsLGJZyhW11870VfC~74YulEuYM&%YOL;Effw<(1+5iMV_L?ik<v3WG#W6@JC
zAyWT&{Ok);S4DiQi(On{o1jpVO9xSzbJC7lmtS3VRyOQW33zh7)JI#Br(wga5U+H<
zxO1^%L)r8`!vfflifkwR;4MV5$3QQxR3`>|l$+^0nR9}>W>XwafT#kQ2v7JJQ~rR{
z#&f$US5NP&2)YNopkXjMKq#EUBi<8{n@kw_(6+#7LcV{QU4$f$2Vt4RZy{+U(R0o^
zR4vD*!;r(jDMtM)AcxLvKgt)Q(m08Gs9&hprA(Pb?<=R#hk<WTf8OH*=U`uf&8f{s
z5^c=;>W-hYxC+i9&-Mx<I~^s`;i-9BzS-GLiRWt`VUS2K$=5*;NnctJ9L3Vv%5~yL
z*6V!U04j!K+0~m?GI#bl<_Zq?r<7R^hGE9qtF^G7>CC#9MSgfU=#=)RHj)oa7>B;v
z_{*V}H;S=OWu`jNDTzo5Z*po%B+<o-N5Y32@2u21M<}Li6pB-3J-_2?Nu^1Y$nM}E
zsjX5FQKDhlv%It59Css%CT}9!rtHuBGnL<1`KNMBmhML0J@-Fj>D(A3k?a3l^t3Wa
zj4d9S$JcpmQW{3B#yJ%)zVO29t)d}RX{1iH9nRD}D1^2tp$&3PMfRN>*u1{UhOI?L
z(;BJ!s^mkfj~wr^i8riBdf_2yph_djHn9D4=g$byMa_`nkXi<XuB~&TK^U`!q<7<l
zri$Foi`tE7RH^&uPCH!H{mmYvpILhNlIv2OX1`kz8bh9K>0huX6GkV}QqMAyX_!Ah
zH91I4q|oc0+68OgUqk!X#cl~9i61nESV>D2uJMw9Q;)Q|Uh6d`Q)@G}l=}MRD=Vx}
zG%Q1=P1w$vcr<fi>3c(WH3gO?OkTg5EI7fF2#T%{>j-dD*&(ZWh(kXcMDDF7T;n=_
z*j8rB6#sbW2`&^XPNlj_d;l$^$fQZ`)mte&<PQ2o%2xtd7%T4z@J?ZOA|Zv}>q9wl
zBxEBoy%RkGd+s%aZN9Gg;dCm&%`(HPkd;xT>efF_b!?$|`Ia9`{-V{~Kq@;yZg!#O
zP}-73imiS>hiympFnGnqWUvYDVQNJz=iA@}Eu*?th<cJ{LNtttp?8n3$6WYPTkxz(
z%Ghk_eJfgec&@f~h_v6-uG@zu_yWO)&TY+d#aD6gY<lW$E8(B8>9O!}d0T5AM(03l
z%fQiD|HC@ABc1Bo35*LYJ)0W%z;s|NIj}eQrbG)<3LIn~lEJBRfUNOPJKs$qmvt+k
zM{ylhtJ~51^Hmmqi(`SV5L)&1(P@km!8w`{>dKMS7$0?T=Y5ZOA(o4KdMA5g15~dY
z#3_G}eR^P4N$fKFZY-q!G4Zo3XM3SH<<78hfdTI{7di1h6a(O=|Ipe`^`^QXMg7~(
zfSp=}A_Lw)M{cb51U|ptz<1j<fhOZ2DKg2oT~y@KjIyEdT#KMpNFaM11E2;q<V)J-
zIWQ9vjbP_?Q878pWy&Hke6-grx=v=SM&wZhzLgcgAjkNgJC8IaK4=)v%`N+Zr?l~;
zC{AGHt<*Eig_I+E)976{!?7B_zU9J1_=w!(@W)6IU&5C&jl=GLt?&@1UGXxTISHVL
zf&DC$r}!|2Ti~D2!!tLc!>2%L-IT>;dpXA-0Yw_S|AZzAMJcNGVXR8MLQ7swZahqR
z3T||lm0I9q0bJTHgnyD69kUnfdDphvId3Eq<i`$B29}2ei)V`Ch+PVXc!aqVPnK?n
zyzI$@tfUFknS({^OC!3<J$S_9USDW^J@sdXL5JvJx3v8gFbaN66<oB?q`-#3gs~FN
z*?NYa8snMQ80(_~GnUB<RkIp4yQ<EFw4tdPSf6FZIfRo;$EfO%=-AU)q4T-snm!3C
zK}Wmq+S2Ptfe~PCMINVWJ5w=31Y7m+_X9@zcE)n1M;=3*yRt4=yAKKlxjYFo-7uK(
zhMJq0UP~21c#Gnj?it<^IjcI4v=1ytlzJ}XYG<ZrHF&YZvp#4hVhuUmefG;4n_<v;
z%l)#)$Jd261RHV)@hsKP-F#L4CaouHDd^mDIm+N3^H(<N6jJ>CAfoJb0x1!U;UXTU
zUxU)EEBB>SafUx4aU{Oz2{fQGTvDEK;LxuoGv8uM!_<4k<5_NCUm4oKo`CXs9)HOt
zY;VO8=@b>Atr^ZHV7eYOb5p}OKyAxsP~~2U&q`xhY66-}T=%q~pQOtTHy1y%exl7w
zk{UfP+>dIjw-f?un;q|MQvZ6`h@%{<g=v2FJ1J$a53f>wJNMmO?$H1;)@E)`J+#Oc
z`Ur0xIa<`(vF!z8(1#^>OhMf8LWE05G)60Ps1m&Zk*WN-I-ONSLuh_0|7(r!E$g<t
zuCF6Gyk~3tR9$Yzd)+C@O$*O<h^Y;qj?)z0BzPq8OJj?)SPQl@hNn2KZ)bVj0u5?F
z!)<~V^{Y6)z=}-^E9%YxHjYo#c;7beiz(cWl2(#DpbOA@n~R3}iZV@jR?Vs0EviFl
z>v_4t=I8gQp5E0KN3WKk=FQ(6=G_-Tqh&o4wT`i#t0PuLqcN4Wb6dami4S7j&u!$M
z!i~Ka=SO+UwHge$D^;lNiYtFr4KuUod-#Qre6Ck_t&-R=<V9PUiU2#;tXI&@lW693
zxGhj0)ZV=Ddo4Utd{5o${L^3B!1foYWOiI7-Na^;QVNokaj#BU@VUD#9P}0h(t_N-
z?)$EX4P}taaMhW`eNLykM!Dwp_0M)wQ!uJ8Ff5U<c=Op1Rp&H<{~a1W-4HR?Z>{wQ
zbbOhYLV71~=Xsb}vfmL})hODFITyz0m;4XJZm)iqfN~_=PD4E6g{|ubd~W@)>Wx4@
z?s`wXTKC$|`>%d!5F74M93=3)j8<^eiJH*p?T#7U4yhy)M(=gflD?LCAS1Y0RQ+T4
zNV6UpoiNd2$k*B1@d>f^xZUR7NaB(_`L!{DVbU_3psog<nZjWpoZgm4!-Bn|F;yGp
zl)o|3YfU*$@L(RNfRmD8b4?M0qvXSpe~)Lz$`ZlL$4z=Zt;5`oeCIeX>1Pd(+K0Z4
z6y=%Z-;e!$Y4j7rvDSHXV7}oOM^y}lD{PzADwM|{#W`h%X-76n1#CXCICngM59y7n
zq;ku7Dp5jJ_Scyn_J~}^Ree`me2&<BMhtj2(1P<Tn1U8U8(t8&aSUlWr>V5%6xz+G
z|8OfUsQ2gzWqql+?Up_Al}DTQ_Wi7rTWo!Z6%~Kk_xG3byFVQU_;>hiq)3ezR~fH!
z%OWRFg0=QyU1GWO`Gscn(XUBX=7go*4rfNQ&1=V9x*Tn6&21i9#dOTmjE6V<RZ_A5
z={*%Ov;P;w9`(mifrna6|E*6E4_h9eXdZOyboMleP~<xA2tfJ!iJ$F}r#yraC2@7E
zwy=!pzAWD2Bot`P``$0rM5cxx$)iKkALQ|HHs~Rt&JGb@62T9<I;#M~)F%YUkBg7g
z`eAkQQj<|J^_4(yn=wX&2H-nok42Cz4=CvflV!((I$u|F1+MCdWT}ltj3u@mN(&|H
zM8LXGm$S>hYKixn-?6%5Ra$m?)G9S!&+*d-`4G)3wV#w(a#>9f?n4F*F2wxg(N2z^
z;;F~*qv1VI;PM?hmJks|sHlt}Y;RS1ruANH3TGipO!=0icMv@MkWaObGKB!vi-GBW
zj{r}ZlxYD^mw-QbA$S#|;Yrg&H9eKq-}k+xXHazd$P*sIey`w>;62XQ^rJ;#GD8lP
zV>wjdxWn@Eq<;IIA%2#0QqO%NPqh#<l;vTRJE-d0gi+1>JS@mu)>#<^h#I2&*dpmq
z4MEaD>o=~ggIVY~OGu&<h0-6@^D&=j&UzbXz-qQTzv7(+lfL_oMAVNup<?yQh{p?0
zALB%yXih#=8`W$5?5&KkBjO9YMs>(3+e+u5l$O9|(pudrUyIuFWS<KObW{3#8-i#5
zqt>Bmj*&Wn+cL8I;yn>5(dtC(#ODzKndcib4DJ@J6}c^ma%s??)coQWqbYs~&iL2Y
zpuhVQwH>wa_D==|mpp1?5>6Pq{FsRhODTO4muG@|`wKA8nmY??U~EKQ-#uE1A>8DI
zW<Fv;PWWYZKt;#ssyW6`H!!IHPmB{!@xCuBsmE7)kH$WoLM{`J^nIrtZybC0I3Ld1
zzP!+$X1VV3@FVD8zl;PLEs}Zhe7L?4^@%hg8v$mB3BL*s!Fu-Q@PK8l)LzU>C$iKK
z-Wa4lUi$@5R)~fih+SZ;4>w~Ap<n0DL*fd*LYp-hNs+m94DKYN^4xXx>up)x0djSl
ztQ7Pj%xi`Va{k5qPvi_SuW4C$(u7wNWYEQCw{yx&(A}647N6VMxYFhv8m;TEGCQ~!
zHGSVMC`tpDh@-U4C+V)K0)lDcR#v2)#fJW~==<+4g$6y6J?MIh>`dNtDz!Px`g!+2
zK$9$P_@k8+7n+r@F8`pW_O#$2dMiAp>N~sthb`~qt3D;9Ut=ep9}9*`O9waioBCAg
zkl@9gnbO%YW4MzoBAddL(9sMZAp3o1<jLIH_GjpINTk(l|GDA96(ty{Z(YUB`7aUc
zLn$zfR2w<8{+Ih5Q!3#5TP+c*5BcFW_aV=f(kCRPnWN^LkJ*Bqsq+a%E|dGLtatML
z5$!{4G~Ez-u>|&t91N-0Y_$YfeOjNPE)hjFYq=U#6l(kOCG4AS-*Y;=o@wqOb=k2y
zyQYH-1lZy3Sey+uT%A8EobYj!6IsjzxCqUbht#b*8E|0!6&L6pFz&(CE*D)G&1z0h
z%Fl!@O}~nzuT85MmKlvafk_!g=_wi{0TH~)It}3uLm5TZ&Sc(qZf4>v#~cd1&TzSi
z+Aw`0C(kUN99N1vfxZ7gtpH0mkbD($01e9vz;pssjQrv#ZYnmQuOxf(*izX&6Pfmd
zs00;VCN%~WG{XSSlkwoiW5fFJB<fH+EHx6Yfh~#?zeWkL1U<86YW|~)A8<VkIvjR-
zJ@QP(m;>8^73hBE<4NWkkEqPP9p7=dj3@F)Gf64aji+R&HmXkRx{w=Fo0HB`c%^<T
zBM~H*-B|tmJI7koscVWur_vZdb+ZIGhyCKYA--A!htWRtw@C>CatH_}>o&xUsH@|G
z#!~m<V)^)p`qurgZMH95QfEvGFx>P=Ddmy62RKnl`FSL#&2Nw}6V4x8Bav)8lDKxG
zcU0v;#)EewRCcR4(`-Yg$9f#f4)lv({Z4qRG|}`z#p8q~Dq_u<!?W;6$3%J-`B$cw
zFR46ZWrUhFzJR;DVrq=RG?e$fWNX3DJ;0b`2qM&`r47VM0w%p_er?He3mFr2@-W6l
ziq~gIdIT<vzp>ZZW9(=fme=l3-#yM4&J+53Ss~;3b^)$iS*wicGhXgvf154JpCPd=
zUiUj}zuOuA&TRaAQC?L_D6OrY7o%~X_M{>(se$PCaCLvGXfbt{=8qsKQtd|?-EW1l
zauRxSimc|n^6rXLn=jj|l|gUq&#5`pDj1sVH&=Z<GzhXZc<DdL=ssE_cVU=Rtt0}7
zbQW^!CWGS^rm4(T1$7kDIe2zKT>e)}LNta&j%Cz#j?L?@O)(ZW76&0cOrOo9Z`*(C
z&atU{xtYfduyVVe2Pd4oyri=Ezf>d$uiG<rl+0>5X542E7h*r++Jj^zvDq}DaWqo#
zNms2|+ltUy-pRGW<f^W1rX1E>{hVF!6V@h-Z~Cv%pGg!b7AI4h#b?lS$>1fpr@{yn
ziAvqW3j30OAo?}!mh@1&j~bKP`U;vMsjxxGn!V_5>lGNjQn4t#i|NEOe!J6}N}GXe
zH!1954ejJ{n>D8?x2Nc{OZgFusp|NQ%iujAtWjdIB@7x?lYJHs&$Mhr)@zQP#S?Vt
zo88B=+Ol2U7{rZApZ2Dj16!(Ktjj`?VZU0@k59v;1wQXbdDSXRIr^xiIa~|)KTY^4
zWC7itt_I1b-P<h}3$oM{1&88Dxcbv`^6d&Xd^rcy(q_)(r=ok!zwf$w&%X019+Y!<
z&qHO!*uqJshf((vo;mGl6k;B&a$x?Yc3o`GJH94ksra66;z^_nt;)2<>cX=1Wt<JJ
zY;_F6vvOI^xk-rszJUKwRxdkdNt5yL#%x-<W=`vCUBhIJ-y7ePnZ#Nx?dvc<<JBRU
zxwmDGMvC+6KM%qqP(C@#JrX0#<B2M>)EJlFg|Vh-I4Tt=o-sE*@%ucptA4KAg{^id
zgK9#VoFb7I>xAF(SVK0w`lYlH3%I?32V?Ah<e4DX*C^+55t1KsW3Ue;?(uB<e@dp>
zzP39n!tZFP&XyChj)-T;$mOKtCQ76_>*acxmjSTQ_*{4DrkM=37F;{}M|Boh3^Aqd
z`40gaRiIupFzr}7Oup(Q<wC)$TDV#l$dWE8n=oqEsQlW0QIe>$iXwJc2Ri=h3J<Tc
zsPPgg?+%I_#$Qd~%=bx0Ag+Ni7|43C2sjsB=_?oF83Q+puT!E=|CXDCv?B&ZR%;Kk
z`ERIe$%G=1U>{)eOM=T0^tNp(i|qno_()B~1^g)V3kk1{>C=CKmuqNu7qHpQWWf_i
ztgIM!fxjOAa1kMlHIKOctxLEpt<~;8GBEf9bo&0<9X<fl3CBQY2L8t|ho!(~bY!Qe
zHbHW9fn4Q6KSBu2Gv4(ISh0)T2a0_X6TK-4K8o0!<h62GTF~pg$Bvx9h6;<nTLY9D
zHMOVb9|<8iheM_e;f>a_#$Tn7hGr&HA{pR*)K)X&is5)6(4nW8mb3%Mz$C$&Qr%mX
zfJsiXK@y?wa%!M*vyz3#%_b#7u5h?cuu`-$So0H0!f<y179GlUO2_&iDGI$>wOQ4K
z%$g0C6}1i$g^&3y`T^NyVIHJ^Xxw+VoeL+rb*^pNqIB@j`uFZ2oseps={2VG;NHBO
zq#wxB8P?$FVEyr$PudO}!J8Vm9Cxl%o8hRN(>l0Mik#rY7{rz(yB*RjxZi$9jg)Rr
z;dQ+G&e!Nb;eW<*l7zq*GL6daKlx9=4E7LI6qxu-<N%IjS%4JEYPy7nT)a%60DZu3
zSwWP0cz-k@$5D+vyOt?@E|kxv*4{TtJwK<hI*{LGedX^yP<>@`g6B#bKv10?PBo?1
z(s!(e`x!r74U&7MzC!O8FSJfZuh93=HN~!aK<;oRB);Su;+qGeg|-82*R$wYV2v6i
zviq!iX@;Lv>5y^NRS)^_%)Zm??ic47IA+#^)tph1c=As&wJcbao)rVA%<3EuUPVtN
zm_YCMSrs4O(>?c_tje)k40x=|=rFE9tUYJ9Bq<U0tq#x-3Q6x?Q)DUW#?V-5V8bZ{
zNp33`5XP1IJ$>RypI^WWl72|l{c<_@xYE2HukC%I9o73zdlz<v26KF0kO~Utab1pl
zARi%}Qkrktpyp^eJfNN}S6i+NdO_L;(D4f9Utq#sA$$zRTJ+K*0bD0Rq|H#^mZIUa
zbLkOW7iO~G)*EFJBkSjL7u~I2VRKALM0m%wzkN!H0~&j7>E41j#As~5Ozct>6kb&E
zMq0FTU;YZ=plIc1(nGrHf8;l2CvwJ5f<zlYB7naB+%5v!{W%yp=9nu*{RB6hG;ao^
z2Ln`719Uq5&dhp2od!r2KU}gDDF>Qd9PF&f2J;)x1~u`L)Av9Vp`c}{|5ZiGht)HX
z$vDep?PL5pZb5|_N3+4!Kw_k2V7M-E9GYk$mud$znwv6Ei?&}@oV={5TgDe^&YPOG
zO|^u~97Y!vR1W@87;RqP0*F)>+dymMGa@DZ&PRho^>UB2ibln>tK96|W`qvN*FE^m
z6#21Z=4G$a5o@&?Fi=_2<nh1$P-`WkKy7t5R7<ayp-e6XbF)nC>ZnAb=KPGpt_*()
zfSZ)3Z3>=0qMbF*E}u>tGah|5@~IMs2WLyaHqyejuF_-JxiTC2Ws7n%<>y65xedaX
zV9ns&Fi_iCHZ6hAAL8UjXHS-Rd8{iS|8w?j@>|n(nr_{Q$Su+9N#7<`^qW?&iw#ks
z=E!PJAUFOG*P$KWh@~LHu`1Q%7Yuq%9LzyP4^sWqnuCd8QrKiw=(vyG(R1OTRyPS!
zfo^`_MHttP;%66Y!s-NqRcR}#w<lbG*S0)RsEVLeG+Ijn8K)8^_3DK{pR#6xO5cK*
z#NSs|G8z=RsSpqWP8X3vP_1jf!}(i>T^&8~Fo=H=m7NL$KeQ9bs!oh$WqUE<Wkej!
z`_fsbwq~MbAaTMBk=SHS_SWu>+}W^`pT&wt111izMn6j+POZ2)96aqDUACwPt|~Sh
zIP{rsM;ERHRrLtf`fwJhWhu)G4i|-l4{+$zbE)#3AU1;I&ZS%j;R}>4FFGD&gl{UC
z<VBb7>33@z+P+UdYphb`i}qhoa$HV@*HrWialmAO^XR(>aE#phRM)cO;5#KZf=Neu
zWYlzhGI!^9VSa?9<&<R}Ti^Bby;PFM;0S6d64RGcF5P*gx`V-nYs#tgUBLWkyJmj;
zO$Ya?KdNxc+LC=}U<x?AgCi29EX$x{PfTK5O{H1~^&de^${W@*p;MVsvQbdF=TM*J
zj<fKWQZsq;F)|2pO$89lU?l1Xl^LL|1F{^M?0iN2M_|JI0WfKI?LM`5Bs48EA4hu)
zx)>#3k3dPINO0D|co^~}@e;WV-M?v(aJahd4>8iwp(Q4CZHS*XKvAM<G+ZhjItG=q
zixfg2(I&QU`#kYog4kHm^?a)Ye@jo*q8sIunV~gwR-c47Y7y!6!G+}(q9*JenCts9
z_4l}E<pwe#LX(a`0xvs(DRiiUXiGQ(Mj(rK56(al2M~Es7xlesDh&AMvjf7?4^!&}
zO@-N{=OZ`TM($eC4-Pi!-^5WJ^l3=Qa^Y4xOjx}jY3QAOn3z@O_=Kb5sbi*tpOi1_
zpMoJ<@nCB#9O%iOP>p?xkPIt|$e+|6@$YAMXf(Q<tNH%%o>w=e(P1$EK@9H~NE*O&
z;VH315ZQupkjAC(3dyYcHvg&Ympbw61JDoDHRzfS@vF6)`J$yDCy3*^K7@%#M8k6y
z;_4+JpYJ87gP`L#el#(^QOn$X@Lo3LrQPZCV9el?x0O+Si<8FRLC?=@t0=A3FL(Qz
zW2!OLp;tCUs%h9OMbupM@vo9i`VM7EE!)cu4jaWiz-VZgEFW{?(S9~ROqRxvb(j1>
z^Ud+6x&wYMZKbrED<D0n<|{1}vG(b&&#v^>*qj?Jjki<Pyh=qVP1-vnK`BG}#pk!~
zzjP;Yk_xElvQ?%Usn((^<i)bgCa12TGI<_`G{8iHn!KEU#gA2naQv<f{-b>Z5b}yX
zn4M<Jbnf$1cLoDdUCTtdms8h&6D3TWAzV4nMlU~@T)$^0;I%HI4tv+x%cu8%7v)~`
z0$xqB<ZiI21`$LS_^;!QmVFz#_&QY&$RpI{r0<4MkkHbOSHQ5OlqYqA*lx&~uYzf=
za*s*jCsrLV-kuSB%?H~ecTD+bwxNM#LtOwZih-ne2CFik#-(yVlGo~gQ-u(d6zk>Q
zmxDU=FY7KAu-%tQLAf+TN&ErarK$;2K??;!%OlujYf|MdJsYF_l28O81SNzdf&wo9
zDk4=MgoKtC8+wKyz;X5pAtoM*T1$63Oo884`)kCVV<nQvm={Qu?uIDL=yVxuU-ZaT
z8DL|NvWwy>&x1DJZ<9KwYKeWQgHw>MPR{oE7Y0f!Slrx0cB&fY=Z$eKLX&1kZD)WN
zKHovxFx(JFkxD&5PgGffVqbaOjYKAZj@5h&)P-<k1m49otAl}aH4|SjYHB(K($NSr
zWg2utMxmb$qB&0Fk1>gmb0Kvsyc<DoBjd~+!|}Nq7f-{oX0-K@$X8F=VohA=r4&c8
zVPS0{3$@X5dsD+UMmPH*<rqUqp=<@(qZbGeo`WmQH5Ts1J|_fbr6nA`x>L0hJ$Yx?
z8pj}9D$^**z4Lxz?V0K~=QNKxmS(1@##M~0Guo&DtdG7)9!bKFXaYxCMWJ-mEA$Py
zHI35O0lI0ku8K+Xd-}CNuGtxQ)yss<&)7FmQu7@iXJeCZV%|(Q_)%W8Dw_Bfk=kqL
zxfaQ|&#_W95ibc#F>s}ys2ruQ13ym;DG)Dn>Q`eT3}t&1D=M?y3|iNwjZ<p^HP=3m
zNN&q^N_^9fRp;hWW-^B#!uj<sGdQk=3jHDD-4+QWG>?0YO2i?IZ5xmwM})(->G$H3
z|Hee$0jW^D0-oPRHVLU*Q)J3Pnk#m8RtCJ#fgmmMvJzr6I?qG~h(<(Pz9bO~A@}tB
zos95(-Q2K{&QAK&XV5|7uD~*s^J3-M!UC2zwNp?o_&e-CjGsnwW<Q9{>7#etr7t?%
ziD-RVpAdD1+oR)jp(d$*qjQ}SynDG@0s2kEi$PC=j}_S#C5eKnCC~HAG<%~+@CKcK
zM#&|jO6#dhze$+yE+_K)##x9A(Zr2tmlo>SrgEp0ranNTU)IuTYeyM^s%rE?-)k<=
z20h*|xtyIk`F8zYu(|==n`GITpfybqv*%S$ZHaCe!o+Zd@7Pq$r)ytGMXT!&WT;Kl
z*^8pDQL+D2VHZ`VVX4|h=}qa^!r;S5#$PTXdUpjrb5o}Ed$z{PeS6B(9>X8<s+65P
zC9c-FUrz$obY!Ocv8XJF`kG|&qo#LhwDlDHdrda1_k#mXYQYX?Z?J{p_)_GE6Q9ke
z#_Fb%4xC%?JRRc`&`O(h=msC)Nq;YnlZj)|=MK}w$a_D{w8}EqunluyJicRJag_!h
zU$HUZbC)4oyqK)B7e&XZzKp@2VUQLq2r-l7znZGC=PhQZd=5JymBko)U{THB4j!1r
z<1}ve#R7@EoIu7M&dEH5j!FlAoYa&9GeFG>kNHN}UuF058o_l*_eKyZ*4Ik^TN)B*
zqvxUr3KJ*}ulUX2r)xpHuG6(<#6A%Gw0fm&Rrlb}%^^@P9C3d%N_`ddPY>_^jg0tD
z6wjZ|d<#Sn)AD4!rNl=N#aJTDV)YT*8;IA8vEc7vi@z}=Toj0Y{=EnO|6j_)zqLXC
z^9!5O$^($cnN<Dpq`~moGVsHa;|Oo&=1}g`Oc{%m0SE}T^S0=AoTLD1v!7en*Toz|
z)MoofjtgExe}BnC1O>-O_od?BUb6q<17TD<g0HkbltFmk|H?V~2hjGe?iW4~>%Tbx
zdaQ~cVU+66hs5;b+RyO!FSz|;uCd311JZW3I=;7U|D}ehFhvkGf_moK*8lp%e?bi5
zx(B%eUT_5D2^#^IDfaO#dab>!cjt}EIHw>=6%z6%nks0rwGBaw{nxRWQ_DXGBG(`S
zp}{uz;?ZA8qW{JJ?|*+u`;a8)+w#)Wh!Ow&f-0&&fWPwm0Xo%~pr=>WwH(bUJ~+C+
z`qwumLw4fDSgGP#$d>SLmAmW#ak`ykzVTImP3!)H#PgrE_kX+(AHiE>YPWeM_m5}A
z)CCU#0q6qmzWe;Z7P$Y#v(?pwZk=229o`>O5>TYq;#vHP7FgSMx><L+C?Bp(23hi-
zet_$VH*j2HATTMmUTp9_Is{88EYVR19Je#*{nFVBTeiGyK%I!Xt*4O+q5&%aK8A#I
zczfk?uT8?|_HuF03^)%dQMYHgKzOzh%JC5pqSKtbTl7RNhQ0uGI9fzUpTt#{L<qtU
zKfvqP@$W14|KGz4mgiH%K3*!ca_}#x6$SLggoZ-s9UMXEA8J-^FplkhvBWmyiUtJ%
zpY1}lQxVvkxt`mrueT2elM<W}RGZty;HqB`%~8M*AVwrGA<5%00Q7`41r{ECWwUoL
zxiuO2i&Uwq5hl;Z*&zT6csLJSyJGWx;7JMt)dI(_x<$VhyQ*s>mvsbPS9UI(#!qkv
zST03@>7E)3o1p^`i8z0NzazUSli<3K3g`%XAwX};!B#tMzH;6U$*Ya+0`7Lg-ImjZ
z{Sg`5P=q7v$Gq!T>K}G1uL>Q6dr9eWXa4W+i)j*KaWHJeeIEYj#bo$_2o{1ok|-E{
zbb?8b$g;srI{ip|yZ_Z^f3i__d?Y8JeSQIKZJ~%Uyczhzj=vX4w(n`v;yrMgtVz_@
z*0(9zzr9(!<waSth!=ZCQUKmIiBNDU61}=!yuD_Vw)&yI?N*ba``5Mmj?<@>6D-|w
zk&|hQS5wa}k+u1q>VKfqE9U?o_m9P!oyE@IfRvC|(Y%Dl3AsHsxh3`k4+^gDBu!sD
zQpf>-zVJ{1Q{NBwPp^z_aAWYy(*#YC@mN1q-%AnrQlipWXwkU*Yw_=%`rp^V|Mf$<
zG2$_4IlG+tXS)q#gUka*`VUZ$+Is!YTBMo;8gVwWm(aL&Yw_P16~IL>{1Rp9v-nBn
zi<|R}EV*}lz&;gKBvBSO>_OU(ZQziR!cVAEfN&cSBC%rOkxWAu;*^m@*V8RxkRZsz
zRvq;F(I!`P%|}I?_YmysY0G3a5+Br~9x7+Axw!=ru=5?*Fe0rhey+9*@SphS7xlmX
z{C_@aI*Ztds-^a5JpO|>1QG?nXKz=(8Y~==xJJhu^8-KI18@$MDO52dgS@l0Dkxt2
z%Y{MX6FB1AMRYOXA-NV>NUb*b{|9a^NH4@6!FWP=W-CD6ya>a;H!&`3{Il<SexyF4
zT-dw|f8o{IQR%?8k2ypXpY+dm#D#?-an^y0!eI~%2Lds)8~NA3m_Ud31!R${IS&gt
zsyh3h0*Velg4+E9fj2#yYrCG)2)4hXD&IZ3X}kH^76#VLblEUFVuR89`chjLV7`X)
zAFoInH18wKQD7~Ik%TMI_)d(eQW+ma=}dE%GlF+_W+0&w=@k4jS*`%F=%JlfAtoDQ
z@rIGM?uUl0Ena_LRP!d)$puasr(YKzjE`9cvDt1#x0D`XA!t`#fDq?I^3f4=EcU;G
zUE!IPBLczo$$QxD7rAk_*bZoED`4LtP47N1xmhzAA{axk%vkAHnPur0sC!-)qZRu*
zU<TZRRp4Ml`voS6AmCAu+@8hXlD){gjIkD__cbqTGgTs&M-G;IKw_B#7%h4YbWRxk
zr>&<gIjkCn)f`|Gc&$B)-~xpq?$5zNw!WVn8f>pSkU2d{X_%)u&ZWp!X}ig0u?G0d
zOZz`>h;9gJKM*Y0)v*3hXqP1vccuFnwVldAjr_^s4@B_F*{T2SsbtnUSft-c#7hY|
zz57*OIy^mZaZ}HFxz}>4KNmJ<JgYJbx!{_&$E8O~nrb#l0Z}IXz+dPkv*byXQGOGC
zakN5nTTj{p_{+ol7o+ihmR4DZv5&?7pm)N<^~j58>MwVjKKdolJn|lRQr7ZnG}@Rb
z)x_^S?|jrlRFXV5yJWEgZ69BO3Lk9jAB|_nkR4zvm=|^PNxV36C{S<Nrch$#fv2{1
zu#(_HhlYWSIH`l*um1h`CEx~v{sX_E-}#yu=-c7!`D}>%M*Br8gn|u0UUvtl`A^?_
zpXgS`oR7qpF~GkYfgtlU6eviJ1h;IdIoTeyO3dF41jo?gk<r%2d;HVT3T<vhml%wZ
zw*{v@;80Wj8g@ToFJD2SB-YqJeHCIcEpXLJKZ1sQ^b<Jojn2m3-dJ|z$^i7?>n2p8
z#J#a+iym_hw7t&{Uae3rlIb?NI-`=Hf+c5&2fGyb=dcRoh6ZGTbY~@KTQlGT;R3vz
zOyk3SQh0kPsWBM>e-jGUnzcNMsb+t!8`y5|fg!SbJ@9e;i7kM-#mE2&Oyvs@m>f!T
zvmY)WvND%|7ePs=t%zTU00mKh-gspueT9Jjbb_tYJUJdgwF|=_Bw(6aLiYZ#SrEV5
z@<j9uJc{?rK_M3qgmg%`WlC_T0}TZO8(9RjQ;DLH5a@t<ps3mrniu$*Ei%>j5HB0*
z@C)R+|3Gkk1;Icl<`-(|9lFnmUQ@^eu1+!b0c0-vyA_Rq{>8@OLlZ;$LfTjt@R>UQ
zaEUBLurLemA>&EYuWc7+Ir1!1>+XL(?ZCG>e6{d92LTjohL}xS9PSK&jl7>B1S@2(
z#K*`YmhUcxV!68u0yLx;&{x-b{82Jpt1uPUcT>g;pgGI1i^q>^*WQmSuo~~`OSphL
z0x+*$XaWKdWlXw~(HS^O?4!75W85c<H89@!1NV-y9E<ek?6Wv}2@)V`w=;Nxp5oKK
zBkGIifayKN=|JGHZm@Wv{ejed)mgE@1P~BV(xst=u>nYmBG?UBb$tn+&>N7#*gV^j
zF?8QbUBPg>gT)=v%~Ll@s&4bt@+9&;me?a`jz0jUulTjo0KdPfbXKU#TMx=mSL#5@
zB#NCIQ0Ch2C5aRDJOaEbwT7cz&$hY4!Y?dpX$S4#w~K$X_4cO!0(S8@;u24AP<p;J
z>*!7cqcs&2Ni!8k{A_9e{Hy;0Bx4od7vjs(eFe&la&4iRBEvOR1;-dM(SHiNcKmnp
zOYaKe;M~kvP(=N{9nMnjw;{5K2^^4bV*ObiFN^T;OoOKoWlLxB?J0zkAR~DAsm+t}
z^gZ5B(j|TJ3`atnaR@arC=w-cht^*}7(u-mt%e~cRGH<?9XR9(8u^)iXclM*h2i)G
zb8#31FE_`_0So;FXg_9&vu*X)$NOQZbXmV~?rm2)7V1<yI51H#HH73gA#gdv>FDTu
zx(|e}_PahGH`rj4wt#?xJ%<%mel<ioe61h11O6fIR~E!U{Q*8l`15bodorFRGlmcw
zuuTIJVb^Qzfid156Q(Mft0jzKeG=?hU@xr3zena(H6DYx0mj;7UU6s#&JgY3h&v?P
z#usC0`c%+>T(^)3bRpDe5yVk3ol>nT40_$Cp`fh!=CMPF%l5c(BpywKO%*}mLD2U2
zq%DbY?k=?*zW$_p<s1kZMb;#+2`H2n?xRt>BWGrgi@Xn*3A~!9B7qABah_z6n9>ks
z>O+*y`>VV!0uvfQFLkakFj5Q^?J!2LA{r8kDpHRK38bHqdf%bRaIj>c%kZaCFw~1t
z`R4ot<SRu6p@yPr*TZytaF<b|&Ox#XuOyyX$u?~`mH95+ht|0@3M_(wT~KmbM>7Nc
zUe1p`M<pUFHUdfYkiyQ@B(wmQKsU@%G$GJIi<OOZ{{T<m#N+%->oBPGrQYo$jnH9m
zXaoXHCTVV=13gAfB9o;fa-PtLcr>B2`(*MzPLOg3HT*}j1E6s|)n|vtSy>-JNASUV
zg50Cnf;(1NF-zf^6AQHzQB9`4ttPa=U~piXG^HgC!e-!^g?7HDYfW67c%tYAi8eaY
zNlDXCd^SFFn~NGW5@ppJ&DlB~aaovk0ufEsHOEvT^yTu>EUue*Mrn}Fn#|H+4G`|N
zNRo3{l|si-rm|Vl`H{oOc)he}sYQccNSZXqc$yEcNi5~qu3v|GA^fO$0~C%3i>=T~
zyH|K?7Jryu@Rg~Ltbsgv!H|@t#Ks#}7UDgl0j)1?4|_)5dj6|!3iN<*m4}M<T%~d!
z!aD0S8W<^45TLf$!zhU-fk+`L>nBz}Z(=@AD?WKcSfK%#9}Ru#@aqG_FRy8$){r|@
z;cciNzE8S;Zm<i`;s^Fi1-y*BLGoqI;>^fujKM{AJJ@^<HR3t_G_FEyG4f3GvCrFk
zwe~?)MyTKMgt^@=Hb*w6rqd(tXYr*7_v6bOP5XVv?+Zn;;NgE=rQJ<NfdaL&;8oL&
zm!Wy|Drh0~VN=b1M(QBpoBPc0zpTf9RfNFiv#c|3{ni)weEN_yx?=$9ML-_E|6+mg
zoZZ%xVojPW>er|z=V@eFnH>IHF)Ayz=z;6oODcYN!BEhnhPOY1=1(Ya=-6fT@-qp_
z*!tw~4f7R^pX?T4)1}Jc59TwiY-#T41Yg0V<q@XES5r|KbRN1CMQ8Dd41yh!bJ{@{
z52MMod_^h`Bf>&hkOEPEDOm{hAbW)v=io~X*%BCm>LRa8CApkw6y2&^=LAFHp5M2~
zJ}Q!DLS^r*8<^ngM9vR(K}L^;He-n>Q@;5`TptYn7{OSg^L}FhFc#%7IVB<~dgVxJ
zIh^gwkd!$M<qR0ylIVapXpoC`&r-^6+G1wUlUb*MZoI~CwGS7+Fn>O`7x>SZqkai6
z+Q3tvSs5r-z*s4s>`oay|8pQ@_U3W07Z`pWog5+2sFU4~a&afI5-Q&tpAca0?5S&B
zW7j}^4=Zw8^i}@?*lsX$PsFI1-kAl<foi9PWUY)n?>a5N=1g(NdHf_Qpt$K4UvFks
zcW0enbQSKSe9`_m^#%Q!LF%*1-QctotI1+pWTCM|qR>`ZY&4C5hh5F(oWtDr@y^G`
z=sNwlv9ON>6{sv*#(Gf!gUzP!mJ`eOhl0e#sG?uwh;j2b&-og|HAMxJkShca<?jNc
zHw?+uOAjY*){Xm6pnAr1_Olwjf=3XI&)%y=%D14t>_Jd5ef{ZFWgX;yHHF0mHOP63
zigoncQPq^MDU+1nrC-NI2+FuJAY63{{=eX+cz=xCJIk<_rTi;_<}9oD(0uw7o#=?1
zDz@2&2CWa*fM4ec$}x6aI~jh$K{U)Rz~ZX&2qU&Ujr$!WV4Q|LzN1VkAp6%putmgI
zzk~N&nZXR75+s`8C~8<G5)08YNY3g%MmfMUr-=TAM+%+ilDzq4Eg@11lT4SCx{Lg{
zJc3op2988pLugYdDA^bAwD|i}ONC{|DABkUGJ_@#a1k&t%IKZ5e9)I)g^3Sxqe>n=
zX_6YOd%0oCQim^ZoX}IfaNHQo6hfgF6%EgM-c8ehT#h{kkpZ5G7Bgceor%_oCG<7@
z$xq9<M~Mn>5<Dwn=z#Fb$Vf>vl5ULdxQmug-JtW-#VDV|b3*8d^ma&$?t>Hul7LJ+
zzO34e9;c0gcn)V+uXV}_l|w9BoPJAU(@fyolxTw|h6A=ni>}ps@Fd%iO_pWPm%1V!
zV<-4Q^7_U7w(-$h29#X3A$h$mR{4@WU6Jb*1UGVRxn-}B4OL#RG#Xwg5*nlDgJ(4k
zazgD;50{VK-dKooVHVa<IR_xU%RZ@$iorBqMO)M4C?oW_<3E@3fohA4jJa*=-+i$0
zsDRUB?<MOS9k~!Q`XshTr@kTgqGEnL;{jz*gA0nON#X&af+^jf7TR!XVHQm&(lp;Q
zvik|g-_$!+Aq7uRudEbuX+cSA4@sBUg~UVzG$Bn6ef^gPCLI&<(48@o{4(BWm400u
z|Mz%(Q`eB_?h^>MPG<eDuK6oLG}6%e{cZ<#rX9H(Y{Y3XTqx$qXm#E<FGn!t4B4f(
z65=)lb?<s6$!R1w80w1*R;gTzASCs=uqI>GgGAyoC{I>{MSHWr){?g&br2&*Oossp
z`Z#NgiJ??0IGP_$p};f0HQgYP+lqM(j{eR&S+c0&uF;WMsrh(&tHR`0+o*MDDuday
zyDn#x3~Gi*D;!~NONXw>YW(CxWorUV3(oNgye4ue*q>g0`w&#tj<!#ayN$~Go30yG
zPoHX=pHio{ew9eCmZQGX5GYn%<ld=V_a)NQFQ;g!xVlPZkTd94RGbv(#-ocZRu5At
z{6CbvbzD?!*Y^!b3=RzrAT@-Pba$7efRxgWNOyO)w30)rh;(-jp_GaOQql-WDAMmS
z=XG9jzt8nP_vigfVr2H7z4toSvDW&1f58Gt{Y58u?b#)7izSC7Kb~zCnr*8#QQ)MF
zh9f6GH;q}o_|^te{Bc2%YMO@r`aix)yf!#KTP>Uw?_ay4FPw&JzGyJm{<ggz(nFHn
z+M+qDxkvVP)a>WLFT5Zd_mTuE3*#?<qx!9Je_Yob>}k3I4*~ImnRD1Tw2j@p;+IeV
zc_>-MLbvE@DAe4jLa=#N@34dh?MC*>VAeg8kw(w$)ba1F6;v1_acrBB0y<g9?^3Rn
zZL?nV-rPGG`q1dzJB=+=@&|t&;0uZZ>V!QV2C$EPU{XWGCQt_6LiKTj3!37WRkeuI
zXI@bD9#^QE^y{~cwv&*=Gkrq~tOUY|Nte}Rf(LGycTAw^qWdTk?sti_dX8rsV~H28
zQ;(_NA=|2x?&HRYkY*!d|5&$_I*$9e<r`$UKaSl6OlvDEjg6mMl9O>LFn;6lH5skQ
zPfcw=0Dsl91edi7t(KvZ9{+1oslA;H>rCWu*2<k2J=aQ&DlU4j(m9bzIIfqVGmx;$
z?LU<e4l_(h8IS!AYF3|XB+k~-FQ(=FDN|tYbdxDJdF{-D&F%qG8!z4l_jog_T}Bd0
z<SVQH13%z|bk5MxE5>JX_MzPjHCt4PMy5VtrCJ7QpZYq9PwL#CblSa(y~ou$<_^`-
zZ?s40{KZDuJ;aPPbt+tqk^<Ic!Ek8ghzRv;`K!d4<PK0B#hd%_^t3Vncq}zrjdlt-
z-Ytk&t%o11AkzOhIyIdd)l^Ntd3vj^ad2C9PQmk$P4Rn{6bCw+JLA&{)nr#sruDLx
zD2GM8V(T6Rij-4N-c0Uu0`JeTru;6Z_j(C#MJ^0WzT_s%SF;QW9F}_$2lo7Khlh)9
zwS>iZI)s&0AN5W!A%>y6d%@0Z!QE<fS*wbz<(!in`U%o4lhwLXYj5ur&A#p9?wsKE
zx2!46ySI|lC4%cGZ`dI3#rJLfwNq}Tn%4a8veFafujOymoVeCs2pxc$j9_zzm;bA#
z?_p$1mLZ*hvH7J9FX)L;<w2G9qX)@6kY=Y`<X7kBJ!$vk!ybfM!0?6BOmDkrsoIlt
zDzzQ`o&KCL9wEz)_kX;=J{SmXsNCvX(LQ{0PNy@Ozwd->4zAn6$nj$PJRRR>n9utx
zzu|3(*UH(i+$2;N#rxWS=kGPqmKeh+mdOBg|50^rCqx1UHb|Q>GO$=(t({mquJ&9C
z5xyU=-c%`2rePk1E#SYDBelUHy~B((a0>%j6!n|h&4Ph@%nu!lI=zGGLGiC>;f`K1
zB&+pkc(_bfC`W<%Q^QmlUN!%mpT!?-ic)ke7QoWO%~_N2-AF>&p$0gmVYM1^nHIP&
zL_WMFs>wyFaVV5nC4bF#mTohE!r`l*InP4@LILx7fJMrp^n=9;imv^;W8pqtpr;rl
z)MV^a52j$p*CtdV)0&P?bZ%c^STn+=F*okEuADC7sS!Yc0Z(|#QB%h%u5MVQ^nl_1
z^G-#|*v}{eTca*(DGqL=Rx2JI6}l_Qd7%Q@nh)@%qrg-aS<lc7=g2)t>!W`+<5L^U
zr_!p`b8?CTjdR`|TV2j@@uDGA$8g|Orna}E2$}r-2sgo>WA1a3nmcNRsf=&`Co#d$
zm$>LA><{<-ktP{@+(;|~2w5u5XCpfr#(CZw>+y%M4KtJ_K?_LTp3x2kK?KKra+|Ab
zhwe796-!8oZ<nS&D;P7@*SF~dgDOs%YGxxCcc}0TJH$|>hDq6uNyZun{BTmHn3SL&
zRxrDv&hd(Et2n}?klFS{oAc-yE$MHoikI{so;7~bw*VF9x<ACT0Gh*7)1x6lky7V}
zU<D+Uj=?0y6v^L)C`vO&;wpa!<L&`0dZBlY+C=~RURo&>op4C4&QLlv(ha=i*ZX$A
z!X&;xpVXR_h+2V(B>??a%3>qE<5zelaE9!&*M?{D#@M10$sW{zK@{}|_rKe3%$tWv
zVdMk;^??cAQzdHMF2k)TbAt?a47jzJ(Na@e_=j><r(NaS5vSapCpvGeQjK)Kh1X!p
zK1Fduj+`yd!Ke>W9CC?a<c{P=P!qEM72zlFD^$7<L5(sOlg;XW=3#v9e2xk*B5`O)
zZ2h5<k9s@QPWTwSdazTc85Sq4OwR{6E4A=0Fv|IYE9qdms~BqD)t?t<^>c;xm~Qk2
z0B`J_Ec60lCxSc79!(7v<l|DH2V$rZSp82H099NqHTj9YdBF~wzp_fNV)6SJp3nlq
zAUc~Qvr&F)M-@`k2<F3SciwTK;VqKMfb!o7LWxaLzA_uWGcOm(tWEflG|j{l@k{U^
z6xUj7Nwt;P>{WXQA#vVs(kl<6_#lWi@p&THd-jatUSS>RDuFqK33%j2#_B2P4zp>V
z$S!*V!-pr7IM8#SX_-IIK2%6#&sUyEb=zRT4fnxpt5V{cl!x9Z&6Tj9?!6h4>b8uy
z*C!2~N>Aj#TTRZU18jaTVfZw7rQ^Up@JDac=l<j^k-C~UB%D{GR+Mb8J^!Uc)8yTA
zY70*`($={>t}Epayr$W|lP|ZQC%?XQ@^}kE5vkhHR|<O-N4-P`xRdTM#Vqn=9C5n`
z37SVU7=?HPF099EQI8G7-_wQD-u1)<ybm*hPD^j)tT8tQ6^Mdq18)D_U$Di+aEfCh
z5_q`EBP;u0Z~)(C036m+Ri@UZx8LZ$VLe8=vwqg*Kqy#ONRHw%&$IHxl1F%RjDF&5
z@Hq=UGhsgefv^^4CJk{lZ|09b2A&Tu%OgdBuSNw2W+)+(koP_Q2g<lHb<u;jGDKMY
zjV8@2u3K<JIUqG?{IL*>G+ae&Rh0}(Ac;ax9l1v;RHa?}g+IQi^oeqUUMpr5n>;HG
z$~L38mqe>TxmZl4JOb@YqcOg!*2@5=9p&W9A{s=Z45ld<2ANmr{GHG2w9#@@@__m&
zDc@=mHKq*gp1qaf4;no~v|>qxhu&665i~m=Z)D{zJVCnfnf>wDfsn&H=CYdym_79$
z#KzA1-Cx9So=BIbwUltk8~#XFyW3ysOw1M48_oh{j;J}{qs_sbew-}gNsqR{ZXi0H
zcJwXe?@tmIj0t}%1N@=s9JR*9b0=hPTV5xx--UhIifNf((AB0G<|p_9h}X~XYEc9$
zK@p@?;%u$yttp)Q>Y$n6Tc~Ft1mPhV_UB*vKF^^k+kpawb5aWhAZ_yBN7btPf@*6d
zDAP&gfhv?x1!%w3JI&Bpsl+;slhZ%@1~$}6Dv>o|f-EW_3o$B#a`ey@!wy+E1)?4e
zr|Kvp?ws5#u$FWBn4|RWw06NRYz6~+8S7PA67d#%5;XajY_17vD>dOy(+ZeUpTIZt
zjP@Q({Df{NNJMy|1sjGBx$7DIl<1*odxjfdd-tVu&9j)U@LF>Q7g)b}VC;2RpcNMu
zFE&R1vov-LbCbFDkA!_|L8CA#57zFCw`DeQc+X+yOgqfhg2-Tr7;@|lB_eMUbRx$P
zj+XT1@#aZsnJk;)yuhKlIZlQ|xjMPi$WMC+Pfi22Dg`v_E~TK7VDqFx{8FO%N;*T@
zVFs8&SshT`u4qetsk}Xs;|GnDUL928QyKqgz++iUO+!XeA4~Anoh0%Pe{%#9$D7Fe
z%T`oZJqK{>h)Y}jax%2>2&ps}72-+1)z}m6pXg-}OX~0;x=cyPak{BQxU3^l9&Q!_
z$o;o{zIkP5y3DzZiVX;QR<055;XANs5Pgi7YYknK1zfR|^WH!qKOTBI`cXLvMQI(f
z(U2YaNgEw*2+JpaNT{KE+4~aCvJ|uuW1Y+3Y$=Yxm|$WWG>V5Qy~J0-$I|D{HAT|&
z(z~VVXpwnxW^$4;!y|!NLbj;UDNC9x9+qNbFE&!I7J*Iz=PE&-CWMAXV8k8GqX>8W
zuhB=42XvC^`B*)FB{TwED|vOAHxuPw>uMHdUK-ys>EC+H?A|bMa*<+e)j_>yl6~F5
z9x`vdJN<dB?|E<-s8#Fn`t<a1et*0{n|P06x87|7)Z&FAND_@aRQA2al>_t?!KdGE
z9C3E%JJ*Kr$+6yXO0|~&PBrb2Tx=*a+yC4?r|NL2Zj0!sDyS#a`NPU$s{&1LmsR3g
zXOJ*IC3<|OKAWKPTA6JRk&&q#UoyJfz)03jshxf_dLhd7w3lhL!_VfS4zpsbO8lGy
zO<7U=W4x_^WZy9fp9}<i)Zcumb2To*TU)hqwQRAL89hshjog1$BsB`BovX}n$%`$J
z)9jpUM<Kp|br)az_Vm%)E)O<FPUEz@7u5#&_S$^~pAfpA_PQK(Aj)jHh}fv0S@N|w
zZ&f9Y9E1v&w(4|PA#93Y)s3Q2MBt~W^HXj1_)@WxR%607R)bc+3`eCoY$tE?Q!ykB
zp8KyPg6a&vv{I5h_cRTGdxN0+SEoI~UlmCz4z!IFurea5KW)GQn?w5X1*76_kKN3B
z=Dkm)C(@4<>II&nbE>||c+R(;?O&wG*k9_ad)u;KH@<OoebmY*D2|~*nKSR?+k_zP
zBtvnTB2+kNg&Ubp?M9R)`P8jZXl_|gm}^9@8gO-m(1ZVOjG7nA%d9n3YN(-bD7Ah@
z)?VCCizhaO7;H0Kl+%O|3(QJJbFmDx>bWcqE1x&!AOA^a*wmoBU&>o5<T4<zVP3%D
zDDasjewjqEy#9dix`OTwaV<A31BZf9Q!In@uZ`<DPYOl3#Kf35X$uAhUc(j!%3o2v
zBQ?P<>>4P!;xUmcWERd^(p=5M+6mXXLuQXt6B%s2Tn3K<t62e>8K1pf#Xq&J&B)xw
zS`1yD>`%jUcD6hC#MUK{490u!9IuL~$JlMii236`-hFjTF<02dlnk9CKF`MHU0|V|
zB3B&43kA*$Wg{YvF6ZI#P6ZPF{O7R|js6`#P1>R(-IF51yU%(1RIxgdf$9zb_;{2t
zVyXBJDC>)AOq6KYYg?~)ia)J)wvh-RRR4}{Bd4&{MFGrp6@SLhQ7p74(<TWeTyoxa
zVs~m%anfECVrZK}6djNd&uzr4Wh0*t&VNmgdm|B~Kzxc&@lLxgXKk2g7&oM3dAIfM
zb)?;Bn?`{=3uj-&`f*Xy?R<`vnWfTT)*zgqq^g3s4vN5G`qpzEVl#xwX=iW^g+bbO
zN!8C6*R{T#fq1Z!;n-L~?b2LU%cM-KH0yvMhcn3!EM>;T+0{1TJZ+FRx6cv^ZApn9
z@p9rCeq`Gp=4?k7EB#-Ue-B&r@z^SO8J*Uy-r&^iaYAP@>?=Cg4j|82aZSS}^Dn9m
zL0cw5+j@<aAy5LPVNYb4bVF>~srSw;1Ag9u(5uFS^{35eGPA$gSJ-Zm!OVy^OJWbR
zf<NQ+mT0=qPPHBU`IQ@k)}Ha7=oW1Fr1kT-K3Q>fjl+l8p3EH_aGA=xF)~b39|~0m
zah4*`82DGp*C<)gT|o<j@-_A6WT}K0<69@Py^**1#^CPHz1`>J27$^BkrA2*Y1L|n
zU986IBw}IgF~_^t64Uz;q=RZ0SqKLb$vI3GWqc+sk+7O2@qsh0NFWx13#)@zKp9Gi
z=wGrwY%6C|Zr!;Xa}u0=E=yr@9N`Tq6W+Vy<&et5ro?8Vyv5lq{`95jhf4||J8Nkc
z)k<hKK@PFX+RzWicJ;^hHt^cq!}cl|Btkv|``mXLLtt{kD&dO;*;|ddHxIT$fDPQB
zk<r*IV={CWm_W}JXxs(T+1BmYL%+D^T94jzy^JD-)jPa2;$c&(K1BPKaw!Jt_@jk}
z=9{9cJc))pv1mDOKs|IA<JLFnwZGd$`j}fja7>xybg#=f7v65m;x|ucrcotS*8B3>
zS;(hLcH{V>DflKf7>al0>ne@NhQ6I#)ye5~P}5h`R+^Hn4jE$B$!WuHPoKhsnajJf
zlEQ5$KWiOdvSP|?hzz>Gx^GQAPqNe~-OlN{uMM51ETwDxonKd}+F>+cd(Cv;$ST)E
zxYt&!z)246$7AV9z2e!s3Z~&T3Y5iN4`<{1E+OuBUKUzM)%m_TpuotHu>X>Ww)!L!
zEX5$RFaOYnpIN6)YSZ~_bkn1p5vw4Ui3rl;_vMnmQy90KsPrpG-iiCATum01{ESDr
zh#Yp2@p!(tut}ZSMo`0WUFED#XJ8+*&1hCtV3v!GsGSv`m=<Iv0<-f4N7*14QE{5I
z@=g&d=nB0oJ&;?%z&>hwc)?XFtWr1b_y)7{7Sp^175<w=n)S!kQdNIA=P-mei-jLt
zrE!eRU3UF3J`*8EqsJ~u!1h6we_XQQV*`A>2_YK<wNkprD4lhPXQ$r4M>1F&zvo%T
zM6N2@xC!%I!+h&`ibCB}@uA(h#dTF>_K;Q}?G!YrRgfO@HO@+MOqcDf*h2&-C7|V4
z?Q3hwyhJ@NyYaRyx7?W=tLy%VMH!l1o9>G}2B2p<E>MZgYxUb@^Y};gJEs>?#M=A#
z;UVEY@zL;A!Q?k&K5?EX2I`z<b7_Vu$a$0Y_!QFyoY&nTS+j7-8Y1ll#Zj5N7=)vA
zvC`_8$)X!MZ0*M$gScaU9I>Ie+g!I>cvS2RlE;5)Y1<Z8^&*~+AJ2!Y>ufOJgC(%=
z**PDKb|#dCo!kqHjpoY{sMD>_F&F{6lTuedLTD(F?5KX83QoHeJsLt{h=h2_&Vblj
zgv^_js+Vjx#fJXvdL5%M9ZcB!9{ItDp4^$2c=m&%mJCP<-=6alM>^gEIxwJ~U}#7=
z%w^+k!yES;iqqeaNID=7ORqVHH{Y%0;>;MyutlnJRF@u~JUMl~CxXxA&1rxyC3rH$
z^(&&sV-4rCbWP8L+*Y6+x!A$=O7b5IhpN(M6zXO1puo#VC2G{swjWb*+#3ZW<)tPI
zij6Cz(s2~&rE7v^smyGMGrkqBv7FuW(eG3w?km!qg3Tn?5-H8cev+vH5{chWb<PVv
zl3O`sM{!mKBQ@&MT8z{y=*$lx2@)KdSr@G5*c=u}n@F~sT7@S+xD8oL=|b^B!`ol)
zX}N8Q(H3=9xP&JnEfs2wCx6dmrwLWTlGWM<!W7Bpyjwrj=3i5jUs$i=6|i2jWxESX
zN2wIPG4^^e+*0$)_3v#!ISx8R(r~C~iIO=neT+(yf}8^szHED6eGxW@QlwKhGyTn;
z3JG1WnF(y{@HMrmr9Hy@-d=|se3#@EKO4sqv?qcani#T!DTkHy>m@Rr8I43ZHY)*4
zGG<6@(jGf`az1;Vn}p&u0E96$;nzA5?00>6gQlZiXQD$$qU=d3UVXX~Ksi=^OyNak
z?Lmw-wfhlwPyW-!=e%u|B~1kqSofIN*vWY5bsbxL6V)4MNEXc$n+|q7j0|`jn!g;o
z^A_VsLQH96JQWK&6L(+MRi;~zAaPA>0TMl)gpE5u?%85<^~4{CdeZL*Z#@UU)$nOq
ze)jxg*!l||cxUm;*QDq8b9-XD@4OC@{3Q+OqD^lh7Rei0^YBAYvQe&r85h3o(bt`?
zgYSIDg>9Kk5f<STNob!cr#)aZ7-)3NrMTDWHNjNJ|CtmAcTbRtu(Fvy6D<d<a}LcO
zzv$wuKY1t4Hg=9>Cd=Bme+>9*spH$QNS~<jB1L}x_|A<O&0}1$r^7KI^QS`LhaoPG
zB7RNQU20${;6wBFJlgDKE<_^E`c-)AzsFH~<ibO-V`SsQi+Tgi5>tRX4tza$65f8$
z1~HRW9RXS6B94Y-1jI_V`c^ncAJ$$u-dsMZfC!<ECKlU^a58<_YSwUwNzlQjNZOms
zBH3p|XI!y0um*7<dv}_KP$C8PT500X;baOhgXFk+>;qA$C1}LYh6t(HXC!OYNS-Mf
ziO_m)1U4IXIFds)J_oMoxNNFzeaM1pJjjX{N5tC6j^4n9Rb??D0S4{8d&UEaaPmai
z&zRw{+_Ld6;ED#5)-1?%MIwhj*U1kiP}9$Z5gF>Tsx)fN5PNTv(PDU%QL17k+)*<t
zWMLAMLFJIVRD^)x0&CB{p+z;9TEP~|kZgDtxGK)Gt%sd9nv*FYR%=)ow!R8-a$)XC
z?VD8O*}L1?l>2Do;N2hLZcS<G$apdam(x#OogU51cp)2E1GiN3U4fcWl&%3(yDlEX
zSS`c~MmSoxkcr7QjB(a;x-N9zo}LM%`5~NhxYH02G%8%S(9n^E@o7J<hNJ-~ORd@u
zj5u2`j}jWzvbfgS?Ti84Vs%<jT0yM-h%2f#(`^o&ttEo2Bdda;5<4v~Jmb9u0DTbZ
zFvp-2FF5kaLiLWpYq9d&vYZ6&O@eg;6!tL+rUY}mP9ZuvI7Zkt-Z4!s`wOjMfyD<!
z8m#d!9F7#O&XZ{XWWSGT#9VtxhJGJer;!RZ+9kA4-<|~np@fc()`xxP(7v5HQw`M{
zM&Ur&+W~5Xdn7^iu*J8}bZza2d%b{zEFQx-rKTc>XcmLmq*8?FFTFLperVpvt#kJ|
zsn*Pz>#(2k&pjmIHd*U*SCm)u-)|4GpGY@cHFT@Jd4Esy`|~EZqM$UzICw($=a8V~
zTXc}RChH)GTUm%WXaq!X)PHG-J;XReU{O#79IAhC|0F~+(EoRv%X|Y4Cr=Sl{uo`&
zc6_O=^MG!=#jrFMzWQO(K*73hQ;~gJa%{m^L6WM1$3^S;#+&h_Mk9KYCN~5u)QKRE
zYqU=}HH!J=MY-c8qPs}iMk+5-DqFNFmR@#WbH?UEakiHN*LWw=@4l7Z2Br;h?o=1s
zn<rtdPcwODH-ftOU-N8O6uiImCbN)u`<_zBnSJ-El3$^n!XF5z_EjetP#7c_TfgJW
zw$U7*&I`2uAU3bSwXVL|`^71Bt&(osZt~5D#%_@omec1(KVmWNQ6G{;dAqPhf^T2o
zD<(!a^)(9AFOKYIQ9J2!7Fjd-&MFe8pSRt)osJV$uRu^+;Y-03J`vmUfH6-o3pom}
zlgw^Fw_I~mM#}6{?=Oc26G@V^;ct)|-;b}_ctW#zW+B(*cuSi|w)aH8zre-4&|T(<
z%2)gQbvhIcxRV<lTdtF@o)>1=^_9Q>%&a-nTkP|S#IppXh=g0LyVeiq%}zg5&U&0!
zY`syqtZx`y4Vbib@5fNjX)g(>NSQ7_Wc&4oLaRo`PEI_g3(Z9E3g$n7%0<|fb1AYu
z%brfH)YBe<G1Xbm?T7-!(+_PB|7;5<ffWH7LbC3pf)~8=HP@woG|yTbC|bu9UV4ig
z+)R1>9r4@?ge1W%fajnNRi4wc$PXT=Mcf~;k3+i*A{S&+gu1dzzHO9%5?~~YW7x@6
zmGCIO-xp9Bdb>xL(X!y>IoW)<I4*}8?!PZBDfoAL4af<|ZV@3Lk|Z&(Mx6p@^DbrW
zyds`R4mdP$1<ca%9nsy+DUD2f^961D8A=C+A!mk0srm4SJf3B^Lf}KglAkhXy^-XK
z6r-+2OZ8DcQmtSHlX8lye=|U~s1!%n=r?bfzRpnity%3+o-k%wB65%lVL!he&q|h_
zDhyBN4MUe-OivhuLveqzpW0x2Ba;M!vz|)~5K^W-IXJdEokcMtXK;~>f?qU}FCz_2
zR1;HXzl3b#Wu3R(I_7$}ckfFQaxAZ0w7!c}Ekj>=vuND20R(|X7S;mXsur%$g_!*A
zw})&rGkPt+z>uJPd9MTD1w^&n&^&nH1c5jVl$h$s(o}In<ZBX09_x!Srun1#QZ;=N
zC!Hr<v%fmjT6fO(R#7x93$OQteW5)H!E!ofdur?0io!PWZLMJP!Z2vLgH?2Hy!tGS
zkzi0j5ZD`h{^_I=oO*kGm8Ax$5<FR*I@=)3szje%pQFcQRqlICaFAZeDpjdQlt4fC
zJiEc?qx=rJ98f@+>OA#v`bSYx2KL!O%f_jHP6OuBLn3zu8k-bg{@cm1og{eO4EAxG
z+_1`up0vaM)ZT|)rw*cigF9q1B1#*ou0=P50SjWkGT(98Qloz*e>+M-n9G3*XO!F;
zi3kl~P<v?0n%#h<7?v|)czeXCdak^LLRk9md2Pc##OBX>D9GJ0ul#f^@=)%n*hWMG
zM<X?-HWniVG4SjG+)&)O^QCg=wMpD#Hs;jYFPFn#4D}QD_|DOgvF<casWhhR>$iO#
zVVEbMr^ULDCt%}3r^f9lzgM(!Oq7Vro(|f|?<}$QNhEj|bC`_jkz<&D@GSLgq<>!O
zGsAscj3BG!Hk}9>-w@iUenw{dolQXs`M5L=-}Mr4-@2c}!}dc$Upbdb<hY<xCa4d0
zx_sZSJizxS2;ytQ6IyO0-q0ixXMI_xO1Eaxf1Xn+tXSC3EaSJCa|qgCBHJr$gj=?`
zjPk~jC&~d#T3k(2Bv%IcGHBN6=YSo9C$18e!7l^~@{~;J53AiX<7Bs~y|oH(pLL>3
zO6_gKE%?zf$NsT@VBOS+`+En`K2qWskbR#fV)Q%fDJTh<)LQd%TGg%wBG~t<N{ag{
z2y+xL%&Won0oUTCRH2vb$!f#*lcX)zT_%v1TqK(n{=X$$<}i4u9vzan1R}0B8cDIp
zhzGl6c&Yq~@Ktn}F!XmS0X*YS+}VO?eBWTYUz38&oR!m>eE~Mpnyuo%Ith1OC&OM~
zHHH@<Y05Nq;-C>O(-Xp9lO$ENA=-F-v^j{Du%CSjI8c=z&QazgbQn`<G^>#Ix_E1h
zn$6B5$s@LPEVu9C5Q=*Vk%|~a-!7LzSaTlC^Mp01h^iP8`qMt!L3(UUj*{M1i)9<g
z<x>&s!FEj1Q^+`Rt~TSz#=ZUt946-mZsW0;vC_l@75ay*56EKFVo@vg`aBfzI-SdU
z)An=DctTrx+NXUDw4Z)sxRZII$+1}TkCJ-fzX|o~JI?6WF0Fq5;{_eqP^(_PDH!?s
zx%8)zdT~&Bu=WJNFS+c9?omLCBO;rkPhUB9XkFmQCQ8FFD?=9K?L%sb(vjA!b8o<m
zLOiC-sO$5Fhk}($4<^dZ7Q-&zW8Ah$5+Y!*)#)hPd>!m3HR~H%H$-5xBkmRLo!u5^
z13&UMoPTnsyL@(Tp!iF3`N5qF)9N(l>v)W*?!0p2_179=@ovXf|CnqIi=sn6x8LB7
zq473E@p$jD#2%u1C~edm#!)r&4EQD<!d~L6XVsm)jfz84@y~Ucz=noeKdi&ZIU=7X
zeu+c*vSy0e2<wN<M(C}fja1Fak)O>xP!D3Ebb@-c98moJgnyswS=Bwd*O7si$Bb=-
zfVTvFvDN*1!9%N_Pw&@t<!pI6S`I%ATPNr%Z<_>9k(1$KeZ&!(63Vr@)9>h?I82~o
z;y&W!!FQC?gxg1aEHKLQvJ=2Xs2Oj;`eVjpY}~6VOA2$izJl?ww+AxsqCee~V~m9H
zz}XMLsCOpp!Bp||#<H7D`D3(5g^`Z5vwKX^=X316b<%qyJYXy|%qwUk;o0v2N8GIK
zuN~3BRhrDj#_Q~)UJlgjVx}pR-R8NZDWqO<h4GE%knI0V=(j8?p8!pQ$$Y!<cIEvP
zFzogTEO5{9ky}__sqv;-HGf+7VOw;l-?`Xr+k9y}VFB=IAhow<gg6OWPXSTS!kqtx
zZz{&_hJEX<Jt^MR*C=Xqjs5?qTA`GLFn7#`ClXKH>r*TGO8(N)_WY1t-g)zGXzl68
zzVjas8Y=!n+5!kn*2Szk+8K_EtnL#%rwUy0pae;L>%4%pnBghEeg@S_*X|%Gu??bX
zLBYr>dG3pT$DKxdo9}mwM)%sQ6*=0gu_n59>s+mu?we)|B3Khz2i3f*-apcg%<I}^
zurX=@Gc-Q==}AI!$eJABrP`3oJ9=&cd;mw+-99lMf{D@knnn?0zruvNEQ&^mK|uFT
zw2*`XG52d{JyXVa6u094==yb!C=x3nVl}ie8(5!L(2R}ArNa>r5g2*j^Rg0FHP@uY
z!iE?dBk`xbo~8C)Z|Uk1VTwC(8E&HnMhe#nN6Qvx8U?x)337hfSUr_Vd?;5R@59d-
zRt1?c5tUfiY^N=gjmGa^3Qw%Ey;;p%@krxiVEB@8Av%+Wkv)r>K<L$;6zfaW7*-Y2
zDUqk%+pW=3(k1@{f#0sprBI^Ms=cqYr<VpgAInO_Z{K#ZH-D92!f^Rm-okpnQ2XU;
z>|h=J69i=8R-r>cx_80vhnQVqU#4Ovu}9d0mMSXEN;5C?<czXTiN+X-(caM3I)VW~
zViv~(k6VembJt<q$*%Q|bj%EunS+eUB8cCPCx;DKdcWTM(V>JIWi|Lekk$%OeW}ax
zj|ko-knEoC&eHrs`(^)kp!0DwOY{21%802irHED)RP9@`?OXI`AKVVKHuzP(uMXX2
zc-*t&g!a6vuS<3*-u)HWllVJ-%2hX&-_P32_c1jF+<?_r6$OvyeI~u93V4M`&}&lT
zC1OnGoh;MdbiVFO_q*f4NQoQgu12dUQmyU_N<?SG-6D*{J);T#!t4H}g6xq?MxhOk
z@J_Z+po1!TkfGIcjY;<YouMVmSYZMu(*@Qyvu>a<6rY-e(qx%Za0p2OqhKXim^G7Q
znD%~nz^P~|)+<&uVQsF`)l36ZN*mtFgX<f+1NCK}m#=TS@5x8dDMl3UbJiQ{<k}&w
z+ZePNtdNg}Qmxe~+e&dM83l6hj`P{5Xct#LF`@T(QCj-r#-Dqy;tH5~eHWeZx%@G^
zdm*_g+AkFl@_TM(uCNmp$$<2>trdO&<zksG!+|>iaG6`O&#?;&KqzPCeZ)j3T2(zK
zP=+;tNsD!RSb^@x;7GI<U@Rs;Y>Xsbiip)`O8Nm5yi*);0>mGEhpNsLJnAVyo9;o#
zY}GxnSDZH#OboG313`;^n}R|7lEeI|?(mMBdncZ)+Yri_fHS%QsE6OIXVQhJsUH)O
zcaJ?7IZq~(JbB18!S<6N=S4!Fy?*}mded+&znv1qh03WxTfnEGE9XlR&!a><1lc;F
z5D_biW?at4DJA_mm@&l9PX@{FXqDtRA#toe&t&k5PC@l+?j3R#^cPfSW&?t=-=>T)
z#A`<u3dN=(6-|G>-&Iv5o6yST8yq7-a2=)LScl`w69qcpj62x`x4W4tR#(2{P)Z*H
z(?(Tg)y#?SA7?cyFPl^qdad2E)MQuN5!mKct4wm-E-;F8S!@3`0*tqJ4@hyC3V@oZ
z>060>D*=77%M6#ikZQsOlDgDnbfKy=n?$TPt4`?t+2H`G*2&StJT8g)yi3-I@iqK5
z0iz5v?`~!R4unRrPk7y?K)Qp_2<TQ2t-{W#*j~4QFYwx)bg9*xXgzF$$;TDZ!MY>?
zzc8NnoY5#TQad@eq>+R9Vq=&a%@+;}BmHQCDBLD}oWK;z6vuDjIrLGo#P@d}EjU!<
z8*{OG9Nkj#lVK)-De^JRzW0CEtu?P4(A#K90d?2YV~+Q`dm?PDH>97KKTh><e4cCw
zW9zAXGr5VU_Pb%EU#-hqD+!s-qN+a;sSs_^E6rhNrY&~2kMyKo4fA>zDPlvudh_De
zwNIt6GdE96X`D1@3}3|HABc}Lydv%I6~da6oC|?WEeJzv+<+q3qjh4}!aq3Gie^t}
z>1|0T1E<5a_Clxv-3O(2C@jk+^U1D_(~ThoA)XF_znv#>D%70HgdKuT566rQB_hI*
zGnZg6PiOXFu~dyQALa9@Z|Nrx_todAb#BjJ=sffte>u&8dSy-}WC6@J_L4>H*PQ}^
zg-3R9OBe%c7-s+WQ^bbJ83~4*4CJrBdoJGtvx_iFy{*&(wY|rY69q3kU17g19M*z6
zl$pWF6NN9lz>{;YyJk5YDik!J5{cgYtf_L-xjfOAWSl*{JB3QN-klV1Vm_HiQu!#q
zFcm@=^CX0>Waw(*D(QO&m!l?ymxSJENG%EI6jY4B-d>$sY^+Wr0}3{PK>2=-Ag>R-
z!y1zhx&vmcIn?Buqk7J^Fz9!W;DY?!Gy*HEeS`<9gwitEFnBhU_ZcFAo^7hceByG<
zRh{5O&r2c7R|V)gVYPR@vn~8S<f^{s<hJn5>4?Lx_Y=%nWJdTwB4<mhxJ*~C2tIZ1
z90aXE|C*9=u+ZW<KZcUuwq;&);a~$jyQT%f32n+DM^TN2?-)d)##k{9=T-~buS$(w
z@dE%vCaZJN^F319)O*>Y6x)aU^NCD0-3Znb|HOvneb-Xsvho80e}S>5y~w*svu<xy
zO?Y)m&G_#>+c*87lAUCClrFog93R0a4<LM=0)Z|dwZUsiQLuexA^yeEx2LlrE&H*n
z;>N2&rJAD)Bv*BJQr@apmVfen!yS@+^AsR5Gq-+oa#sEw?yh+LURA3Fn2Gfp+2yZq
z4|<7koh4H29DW~~P<a)0ZegAKms9(<FLw_c53j_IWM5h-P^HYoChw2lJsW!KMJx8{
z?FqFrR+<yWKE$n2jfV0ST6GJ6;5|ig@xs$eu4a&-#C0^vqj!BWZP2qnG*l?{5!ikj
zyK(Jz)W?r7(!8xi`}=YV2fmA(<iXgZ-&U4NjNPo(of4ilrF_M>JU59f*<Z~P$`lfd
z0tend@3Ya?(mN!{c{Z%ayuW_-#Gx!xVy~e&Z*#BONb0C5zv<|3nC?`FhC}DuhWM-Z
zLJ0|+P9dtpMat)R#xgIwK0RzC<X6+_t&)fW`n+jg0X4Tpu7}CCK{FTDb8!IM!-A68
z^{Tu{l1e-&bQ_}FSEA<FnysYslj(X@-ynMO$~@mHw}8f}^))*w8=YQ?(&OXcf;u2!
zvWZAD=<F={CiJ}gbi&*ovqy$5<&F8d5~EkZyuq+<4<va1d`?BHfW}yq%0h@Dw}UOf
z_5tTJ7pr%CKFeMM+Iq}8_RVij=)95+(K=PZIHulF?}b++yOYK@!@DEgUz8Z>viZ2I
zjQewjAa08|)3&^)MUn#dmwFSQ_)6<y#fZ)Cz23Xy*ZiJpF@*PqwXFghx`CT`VDGpk
zy}dbed2i)mV{*B#XPz-67EPpe^hxWTe?7R&L>R2u+EO_QI0lGfh`Opj*$`{&GrL2~
z(b_r&R1l!Jf#hIenkmRSs26=JzuK5s7(TN+{WDSK{?BM41Nj(*234?in;79qg-_=Z
z%3+kYzEUE4Vwdj~h}U~y8S0AuAXUUL(GpLNIu%M!DbvQI7TC*iB-rUc&O0oE&SN3~
z4d^2J;1V0Qz#<cf^2fSbSokT1n?(20rK^>+-tZv{&*!j7jI-n(@l;jAh3HUBx3ocB
z73R`htX^F$$xjEz65G+2*80;J(2W7*P-Bd3mwBgVd|Snl!iXPOZnyYg)|4CX*=Ra$
z$&ML+5=1^vWi)dA{*?@lFZ)>qy$@~TA$BJwB+fST)roff!lt2IRXE&Qzwqu^m~lGf
zS!zxj7Se(U0e6!d2wV8773$fiB!?6sdJEsl(L)9!C_dRyK6#m-!?Kil53>#r`i|pu
zRup(nY02GW6vFX-fej)yq>W=k$8MajB>NXS4)wBo{*KFG&<^XsoKd}PE1tZ*{APce
zzeZpNeu#}5(GuJpE9@DV2s;(aoE{K<l}_{p!l7BOqUEm>s-jLaD&OZm8Tgk0y_{XM
zOpx8;M=toxvLe-k3v+$^e8b88GYb>pH?J2>W#WbIhbB$r1)pyNBGzu@qL837nI=3&
zwwgELIXZ%HU9%nM0}&gNYLBgSgW<59=u?lhLpr^oyeYqMv}RU$e$h9E94sA{hZ==a
z-{y-h<vXCgpBJ9D8<!L+G*!}-l84`2!<lm~xVHw(qI@;?*QFGEJPzN#!Z{c0N|Rgs
zUEvr(JJqP;qPwJBM#xY8ZIM7`p`@EW|HP`{3Z)dZ>QRzCl4sqB=pyC2-D|8QM)@t|
z@^op0rSOOK_2uJao8TaU_1CY}*Ul>aHS-a{vI0@#@y{6MQ-7AvwZ!w@b{YXs{6V3d
zGkwm??@zq6aA%c|gBn<afoWXRGOgtDG;Tz8cXBJ=e4}l+sdP-lA<&x>Uu401kwU&$
zeci44M3*IN4bz<;Q$%Q|W-9tn>S%leV@PSEX<?VLR8p?F>l@w9(Yp`c?N0-+isv~J
zE^?Bsbq%JeW0z7_zN%!N>sH^H53VMD{c(!ia`I**8o1S_za7Xl`}aaCM;{uV`#^@b
zX0o^166+s7r+5M>T974<wydE5x@<o)B%A51%fJ4M420)2f(OR}=ZF@Qh~uNVESJ&F
zPUJKgmfuKlCCV`T6Ky^`#OonGYiW_-Tf>f&v%F1ohlHNB+m6dXlB>Xq_r(cu4Z{$E
z7P^j!B;_!#M^ouRU_qBSur2X<uv9?14qhj_I&zCi2C1@)<`;fa|B!I%wmV*eQ)&kM
zMd_eB>}L&omJ0R2urR#@GYS1{&pj5jIwYYiAK_Y(#BY+C&$>!T^@H#g6f%aD+_gx8
zRoMf&gV}@lRDWW62$99cAY#fvtQW1;?*%V`ZML&^d|g<Jp{+yZxAu~t{G)Q`wY3Hb
zjb%u0`ID{Y;D&@0Pq;zrGm9(2RS0zX?3X2Vs#==f4!(Qz<-=~cn)Yl!T3qmY{K?V#
z2GKYHS<eyNUK@(^CU18>d#M_Uuh3qYM9t7Au0R0B@N$$}SOW69Jl{{ar{0H;It%7n
ztoH;4mBr-)(wp<Q!QuUGaFj?fP9;2=Xm1$KcV!*YTNid$6qluf7~M9C422lwHZ`R`
z(~%yz#fMKX+&F(i+%U25T?ILBQpF1Um9IE%KTDron%aD|{Wz1+tE5^g(VO;2f?^u>
zFgb%hNJPwd4t@<83}nh(F*RCJSwKFab>}$V%}Z%A#%CuT419Ezzecz!_$$GUfbTt>
z0yPK1Ev1JdqAhRkx!BOHf%n2<SY65i<1J+zABk}e2|!_?ah}cP>NlNN@}bV%*9|^w
z$-&2Ii*LS**gxUN<8T;u_@1kk$~6~sj(3i}Ll)Vo03=}ZV}a#VN?uL#fp1g;6mK#%
z=+^R^qzIplhm#e^)$VD{iV{3lJ{Iwm@*^TbcRk8zbZ19JdXO0Yoz4B%Z`9E+4Fv}E
z_!44J#>MG!EHX{Zk6<-&F=INgULys43MmZ3Pkr~eN*{3CXKw!9e^b}>?-k;&BvFYR
zofu7-JS52EA4%hXeVc!$sX_Fij=EJhKse~IFIN99``6Eh#67Ff&ttA3{#(T{(eebM
zOw@rtd*!dn8A8y}t6qT3Q$9*&Pk=+?nQBh_KTjte8@xG-hW7oE|La-)<L!HV<<Xjv
z;7)34PgZy61^|wJ4dHKO1mZt<e1_XVlHC%3^fVhzlV1k4YF7jA?F2(RP*N>_QuGAK
zGW&5^@|O4WXM$*AVtT+l`kwPsyYU?Q@fe(XOW1!N6-5MULB@tEKKOSg{eSsT&>#}0
z3mz6=svGZgDJ0}p>Skex)Br{QkNOE()2khW82$+f{&65xPP!1I6vc%9kwvOoG<~&&
zLOE(=4E?vu`ZdM(*7ogZBxzz2dH_1;ZUNvv^Vp1LtOTpnLec*t20_AjQPIq*N+yH)
zKY#o`Uh|*dtU*vLkQKmd{=hBx0f3hGYc^guV#nIuW=!c}Opzej2gRPDn9Urk<eEu2
zXw)m+if919eB)-q@_$^9FbjI%H(Qgc*YSV+>ObG|-*?J?eOYuYI01P~1MUBLV*fw>
z0=N_uFboDX0;tj8ea!!QVbmXEqS!?-6$Sq69Y6zzcMQT}>7XnnImUlIsDJ&DuoVVy
zj;@D#KmPSP8FEbLj>Dm!V5=)2**HL=7U?ikE@(ovHVo|7`%xyNzFy}i+ZF&E`t-8e
z-5t#6$AG-Jgm^HT%9&yVu)>C1@)I>MquEHU_=i<9!2_LsC+r3s-U_%&A*e5C!iBpI
zlK_!XKc~tyQMC(jdXN5M41v2_0A(D0H&bBie|<oB4v9hBSk10=XZ`zAF;{1F1-gwW
zl;|aG2#3}e{sWX#S9)#{J~O5|iHtfW)so=7+;a(mYDt}0qu^%!7*Gx#dryr6Sz`%+
zX6*+u5*k#w$8|l^02_JdwiQ*F#i#=BCUOA<c?I0;)>Y1{f>MYIGqXc)icS;Q+kLG8
zjn`-}YJt1~td;~<FbtBc^OSkhPWm`^A`-9%{({-A01x-vJUHv@zdq;xk4VSkiK&1_
zEjDoe@83`2qQ`oIG|cRK8kpnZMq#~Iz<D`>bHcnIWMCH5NQyTBM$~_Tpc6gYu^||t
zG<&j$qJyC20tM}b(Vyw;E;1B8ss<2v1SL!8tO)-)t!Csx3ZN>x9*wY&yKJ8_Da!Ei
zK7a;i)Fol{90VN%tbpQ(2gzMbnMP&CClmn6v`KS;Vnt8``Zwc+Tad>87C@Qt!F{Xw
z0->)I_4nx{5@MSEdTkIz<Nc)=hG9N>2#DLH;Dp*A<A$>B9|WJauYhiCGb9u@)y<u{
zT%8Hh^v(MTOYDYoC5RU+EEvz@MvV5Y=w}TpgNDTy=`ofe<U@PSOZ!aYzh=#(QB?uP
zORpf!|MkeAs3XgIt}pekBO^t9@GMw@AJRohFngSuzA|5edzBCv(0f?C;yrsFcenS^
zqaY9y%*|Z4E&lg#^!+vn^}~8@Y_<Qq<}ZkSfu_CBCFEirWt1KRsLf0$f+QizT(|G?
zc!F;6D69F;0Sc~$BE32X^IrUFU%KJ~PW0UnEy-Vx;9USB!QBBo=m}=ZHI7^Al&?-E
zLaseXiyh9^vk3t++U0vKSTKal51EI)v@`*>j^Ky(d^|a{%VFv3m8{Pwlvd1N#Ic*}
zgPT4e2qY<2&m}RvIiHQOc;4v_4F#fG-#@?Jh0xG~<t;Q}8ASKRS|(0_3^Dcv5TB2{
z7v?_TX2dx{|Ikr*gz<qJK=Z`Aq_;Fwo-QjDT|RK61+tW-mc%kg-3m=F<C%li!=;<+
z=1r^5RYnDb!HkS8X2H8pmUfPQZwPmbj)O|;paE5PnnSEd$mMsseiT9A=kx3iroZAC
z3d0nN<ePt1FId1SPRW^z+#Lo&OtyE0U{|MaL?X$JT2!_>FD<*h1VVf!j{!1}Wd+2b
z!&wy0G8$!hYxAOE^7G}kR>%=wSr_n;t#dMQyxBFoxsBq!-Kti>&6o$J=zXuS0HnBk
z-&r`6i$dSX<_@=GNXyTZ6BG}OD*-tGs9l!t%-#ij+wFU~cv23_UG_VYRoP>3n)#k0
zO^ciX+?pTT<sY6CSmRTcb*`Ry7cU)AM5{k~OSgCQdDrXw&S;V+N@cn@I}Rpalc|D+
z5eZy#j*l`6OC%ilA3!$&{jB0Ntl<WUWuy55V7amdgA&|J-)e?!o-x9|H-JMxfo(!U
zfMvZ}N)z=(W?abiaiJ;qQSfu2wBKD6;Lf5wD4hBAh;HV<6EotY?Q}MG!=&6I;WT$z
zj5B7B*qqb!G&jueYWYfbn9Qq5u+-{d>#E5d=^RpG?I_c1V&YgRL5h?E7}-e;pG+y!
z{Zih>NxO5tq8WDB4`RpeL3$RLkL<Sz|Ly*&EX)0CMHS@aT_Pp8+kR_m?-cwiaeGLL
zp(Sys(6U|r@`Wd}LMQlFe?KMH#{^Kj!Va0kc$b00)AA<iI-S+l{dXPHoA5Q<AgG5p
z7JVvF%i$2!tzR63TLL-c?2xc2B{Uj|;|{V(5AW&uicQZdoG?qe#AFi5guXutXeP@*
zJvFByQ9$lL)hKK_8AoIvAtRM#o^Rgy>$Br=DHk-XHU^bGyi?C$_~-K~NgpbA3eY^Z
zE$2IL@;k8Tqp@67C^**M0ZEePa@eEaQQU#W_cY1+QR<cZG3il!kf*6E@Xu91s<=M~
z>(i%&E28joVZQW6LEr+oFeFL>V*r%o(4CKddqo@=pF7bi4j%{Oh7ys}8-iM*p5|v6
zO1wDB0>2G6MV|dgBTTv`=rXF7)a?PDT5+RWI=B3Av>=&40%+t5^o;A6?R;o(s=Scv
zv*bH|40M;;%&EdTJ<j0LTa4Vx<_PGVZqx#<K$_mg67)CuzN>K>hwpQE8I6H{CU|eV
zqL4s*uR-?@KzM#RQ9nly3kx&`)gjGjg{;&EM{-iBV<6sg=DGIaYKE#EtSeie2P|C_
zq&LG8!x)tz_^Iz*ty;l#Xue03T&|wl3^KI2;|qX*8^=vuYvl<1(@*z1Fyq_zxgxIy
zlfX_>oK3I@4v!>cuvFg%WUOyLQPlqNJkgn5Nq4j`(lcXl-8<hJ0*KBN72m(qm!O*4
z*;yo6<b+6;xf>zzatC?}uY3kTgBGCI<DEM)1J#Z?xB-XnD#v;XfAf<dz73EoB)Q0m
z40Ve*8R$}<I2gc8eJ$RZEvvSn+6jO02caPQc}-ejKa;9!9RQeZ$89fHS4HgoKC+RP
zmOQ<JC$opu-GKFw3SK_t97eewUj-NE=yC%03Kzd#*UVUZ!(PKpY6gAJag1>)0mPNn
zj;rgB<-7F3t{-dZS_B}7onG;GG=p)Q)(bzFRX5-?`d-K{BFQ7~p9?~a{3sqVKYWNX
zDfs0t6L?~P%Kp=>j{k_ps3hWriTgbPHT^zqWtvYX6u(wch^dI5@qzMWjHc%V##-sW
zy}3en?VE815!JkkCQkT0`f2wUfe&{-l})e9T7sDD(pR=MC)j|-TqOwd+^NEIENd@<
z&z)QXo_qoHpL`w@7%-0jsm=VbFHv3^&JaWD?XU2@b`JkM_fa;@M8&ibky_Hj)*s?+
z%>T+E@uJbJlCDmdmVOBj^laSSsM)fCZy4YvQhI;$GGMu=|B6M^2PmXQ_eOb7xvUsA
zavb=7U{gIK0vx)Dq^DxJOMFAz&gDN?w4dc!m0(->Mm>cT)pr1P<=r%%4~uTYH2q;X
zv|n1v(`!@E+Ng!R__(>D&*FC)wBP;6XP@1v+zbo>)B>*{ip}xEZa_qAT})cK-LDM)
z)jK}(wIgKCp(4s_Zp$gA@Wt{=`A@&QLr>sW%P44s>(5@C_Qk089aRsCDhhsB5OJ`G
z+YVk<)MB333wNyGSDSH3l`P4bSD~_7Lr%&vwZHQL863=<(mK|}=jk-xe&}1x@?+j7
zBOywGW117kUZ9r6mzMH=7`^E0<fDpY<#^V$Ji|q>UBT4fUuY8QL|7!ve{BQW_2a+l
z)hH_N7yMg?ELhAF#xX=k{dCup2Ubayvy+5?bB2kj;2KJn=Dm05G>xkr{ig;aRt^g)
zX87kdKNB>R)wPFg%z%gsUP#1xt43i3<p&ZCg6#Yq3BS8=s9f8{s6p52nIQ-s`Bei5
zN$8DXO&j2`<&V3AJ?H1@WihY?%a@5zw<xPr`Em(hZu^mtX*^1-+)x2S7~E4<$l{d!
zOokrkRp=m&HFhFx;7?hW;7)MMK8srgc{&7#+BK9^MZ1THPyq&0grC7(G8pR;Fvt^+
zc;@(e5%`geuzMniKyx<)hq(iRyNOYyj`G;*?&e(w>F<gB9peupWoxWXnV@%p47fP`
zA^lfNz4#`qU1;hqiog#WgElCqs@vasC<UiW&E;xY#$c&HSu1cIvr5t*P6x{fy?-08
zv#3n%7Wo~w7|KBYE5~2+3gP^dcI_@CBozbqBh4Jc93VID7+inAk@|^l_6Tg#9P=&{
z2W62_sn@Zj(|ETxP-WS;qQ_+tJI`^C9yJCNDSwg)P^76yv_Gi)e!_?pOK{C#WGaDj
zcTCPNy<T!Wl^Uo8tF@F&S2>})g({_7as^QQq8?elLWti9$au|k9}z8Ulaas$<l+(U
z<xJ@Zh*=VFQ%&h_qXfr`(G@DMk(j?$0i@Q4fv(gU@_wd5i(60Uxi>Nk1(sa!hnD{C
z(4bE_k^VT08aH_BWx1EyMV*j~<oEgxHl4VrwEEV-uw~_(vq3NFS*LNrZ+Vn?e^q8h
z{bf4FQR6UeKqOEsaG9`6ZaBdbOd8&N<hM@qSDH^85B~MW%u9zoNa)InmNfwsE(HRA
zSkWDyr!Gb*Xo-(}A?g#n;+vpxN78E&SO+qIX|U1{|IBXj2+@a}PWn$qFgN(eQznm#
zW?qqef%Mpl4c+00l}F>Cy3souvhN=AjlqvebovrFOhKDJXdhiaAvM2A8)2E$6@mZY
z8IwhQjGQ3mB-J|X^Oxn}!0LJ5jyZ#0-rpcR6>_K9bE<QiLPGa;ugAS6(UTrkKAw2M
z!sH}zr%0T>C><d=V?4e>krwibP)RwdI!!7a|DuHW<L?{7U$V6_<O6efdszP4t74yS
zx_JCPU=O+<!9PCHoG}dlCFc3G%8?Zk<`xmEkwf<Ht*cvfD4TEz@%Spp=}3}(%s%(8
z$EHYXA$3|2xyy!F?;>VKq@5#(KOr=pBhoxx6x5FB3Rs#5tZE7LsW(Jzm0w))9Pd2q
zO%$SVZq4S(;S+j+)pAl_Q9_)A;50rxsUl`Y5NjaP=n=$eFBW+ReFSEOr&RQ(R6+~%
z8A${sb%zNV8~he**STIq722nVmD@Y@O5H7Hdr?+vpM_RJ?E7N>xE~o;pv35=Q2%+x
zeF^C?g`}H;otJP<E9(X~O6U$_)f3!FGXKQO&S;>>h~5-56w5rN;x`4+PwUw1<seto
zt!`1zSTVQe%6?Gabcefb3t`Wccj=X_%g8d++<mn-1$oD5c>L2aH7>)RF;kIos=P3W
zX1B5=D5&Or{|0P&eO^U%N`-R@R`#Nx)T!t#FG5v-L{~uZ@yyjFj=Sbdi|V)F6u*O=
z!J++H`w*j2#&KI{Ua5>^Q`Doe*J=M`0o>0HdV&{MUZmE18JjqT{k1Qv8SCSjCC{pX
zfZE-uSgt9gMew|e$rNNF?RWGEi&0tib0rU|MKTSOb!|oFok}{5+}A(*Rdl8xA)LIE
zLX=@2cscGTWRYWcm{<Fiq<uvY6P2VDGX$B^vT9A+mx^gYgKR5g=q#X4Q;_3D{h8|=
z%SE$d8oA#IK_)bO7BIP0z%jipFQx3ge>aZr`7l{xAeBSFHqMc`^^U#5j>?q)!f||;
zVB#ynUcfPIYN1Ps@vBljc6#@s$h&E0(g!q#A4SV4O5f7cd5zt1BRQv>{VB0Wxku#m
znpmRsBc#EiA+9{oQ`BMM&jGFzmAj38UMF2S{`*@Zy4+h8XG({`*H!;|t{`D)vdds^
zfH)l+Yz6*9^UM=X-@9MCrDUGwzJm?Vl%zF*ouut@tMs-YU+P$>6XI^tn${ne-{Q1|
zKSLAK@%RXvG(w2K@r8_=P2!X<H5oZ8c<3}1q#K@?I&j&V3L^Y|Q%*^2S>4F@81uHP
zQ=<P2Ay@vR?sRj5_AXwLTdeFR$jP!$Z_wScWF};P^8aD%t;3>xyLa!Qy95W21}Pc3
zyF)reL277_?nXL?k_PDz1!?Kdk(7`QNeSta^u6aB&-2@RANzRspB~2y&fIfd>sr^j
z*7;cz1-vf_kvFiAMwr?Dt4wWRB7uL8BBv4m9{FUrm?9e#^xLeG+aQ<wRm22@QN9j!
z%9YayszufVJE(*hSsYrL@v9=!$}!Z9kvKG3uo5wD^;UpHd@$uFxPYvdeHiY?O7-Il
zP(=651ZF$YWlEVzM_V59of)Y8Ow^P(FCdVwYum;yD$VA<FAD`l*%yq2NTDRvlWG2t
zo(r4S!?p4l+xndSXuP6MxS$8($uH0S(t@7x6ZU+0yhzrh))qVjd&{<Uc|k$ez})I2
z2Ri*x?RhaJT$vE&r~f152ddUMWb6x}%o%DylYv<@qcT=SedxA&UHE&<d7R8~tS|Ld
z^^dplc@rdNbg6YM3g6t@3~9zmcEMT{eduwle8#vDJYd6(S6M|e&+(ClDmn#LS?+k)
z>L(yw3U2!bO;dNo0GP}X^T>+7>e?#X!~-7IjFbQw$>LY}gRM9TP2(<VK6mgZ59f2m
zlIACXfMs!iy^v(30n7VzQXy3<x2pN>brp_r_tQGyU6&bDMkRkc3Sk(Mx2Vm5$owic
z5*Y60^ryYmSbPKK_D7NPY~Cf&rn8k{k^|X?C`kSr;$YYa)erwxj}G)Q^JkLZndc94
zULp`idbm1b{Tz{`AI4E5kC=?aY?LwPzY%9%<Fy(&Zc#U$8+o<#TkK6{b{Gvc4xrxL
z{~>Psq<3DZbF(_jbg8ckVHOa%|EZ4J^}szvQIrs52wGH9k^Uz|>hhN9@;qt6QD3SC
zoY*?mUlUrBxLx%(xEX9v>+m-YAH_j`FjY_p;xISGaciq_y_1P8$6aDue}m>W*NpQ+
zEC!LFbw^1bKJa#~>@I=ZQ8@5ho41u#2m51^o-8Ol4|Iu*%GNfFUZ<}eO9bFsk4UeY
za*z)dT{<Ii43}OKa%pvr@C-MJE&`296;u0%3z<BF2PcXWr%CbS_^!n1ZB1|3TR>wQ
z^vfx>qNH8GptINyao)}aL3VGJLf6MaYWo?#9Jt~&C9<Q}<?mSCy3A)SHG%V<cg8tC
za!!gec)c|EXdp(lm~Y__>ct$5GZ!dh<WZmJ)Zhp9YW@19Tz|suDr1)2@Ws0NyUto@
z7?)RQwBH&6va021IbLd~0LPl9FwESpgacvJ28_0xDzH1c{DJKH>$2?HJ28SLqk@B;
zEjw%haks`v1(1D{-{(3pz}z+Y>Puj;*o;fx1yAMn$G1(W0;eWY63(w;RJ12WDm-J(
zH!4XR+e)Y3xRM%;P*jE_3Qp?GY#09=6)C?eZ+cl0w+9CDoDS<gN3TETFS(bOx$b!#
zVTe%yt#(3_A^TniEXN%TUe|BZEJKOLQ5;lVLtmtM8oth!7dm1aUV5i%+qmCD|7Fz!
z#)@uh*Y(-4S=n-8zlljaq4mvxua!Qx(HoIT^$F4q`pP-djcuQ_=B|N@x+K)>mEpIv
zqE;omz$m^Aw`)<6zeVNX3HdH1WC<p#JIk4bQ2hl0nCZx5RV#5Hzb^NWg<N4`Q$VY7
z6PJ3xHD_H{2|j6a$W@Z0P#k#C)Lytsd;IzPE0g70!oy|YGPv@`*)e-%sl@a{vn<0p
zwK^H;iNj4N$%vW%?%W}u;bS4BFJDN8BFd~aAIII51tCO-(yCi*&Z^;@mn&i;p-T*P
zWb{*Y>nXq8akl`wK(2$ASvINvVRuFg4QUtbD>=rg`1a~HW6#gkhU>2zriOMDxe?pQ
zWIiCA#f-f(X~VJk2rwr_9U)DIiszG}hTfvQegDwAb!0NZAXtGjo_+Q^!|jg@DWODN
zA_}nv>YHS$nX6m_mJdM^p+02}%62|^KPL@oj0jgH09^Cy+zn)ryjs3viIYStLs0g`
zA=DW1dR^*6$Wr?Ev%59v)1(!e2>TCBtyHX4z)+{glR1QX<ioE)V_a2cfI9FUF0A)d
zTyvuYpZPw1a6iRiacAFE-e6dG!~4u){QzAg8t6k>wyZZYT5EqXZhLEMkV#ChGkxO7
zGl7)m;uYIg$R<$5jZzXo2-z?D>0>6i<lGCT_}(5+6E}shaE&`%xvK|ok0Y&Y(-H6J
zO$4(8uB<@EY>>)rM=rh7**Cdgv9=~<q0KnZ8R>6RKf47SZuHnXUtY8wHsqaui=g;8
zF(?hgBD*S);F?WJE6I^a5Fne~ZRno$IIaHz;~xDHe09;Mk&nl78lXZUX7fjf6x^&H
z4RDbd1K2j8?x`@R(HH}B@lsrjO1vzB6{^t_p4BQGm38_T<eZ&^vVZ<0N#7_V-)Ie3
z!Kn`<H^gV?Z8yG|^H`Zfs(;lEx2UFFM8ec2k|rK_)h#L<D7xB5E&Od1gj<*%QBk>U
zas&x8Q?Y()<YjVb%+-~zoD}>-z28;%=E?a398uDNn||iRLi$?}Dnmnp>l5B^02K(3
zY58i93}uNJ_K<N36I(@fh~!&fC<TAOvpl^xbfjZI7<5Ksja|a67DZEX+ZJ&tL+t{F
zAwIRQ{mPW{Zf&y|5a-94sUQl3n!l@|<D2yq!bFTulbeD2P%jo8m83w|D1h?vznoOz
z^tiU5ALxt0A|s|{l4W@{1E?mT>wF<@bF)GE7nwb+8B{6zs)i$s6>Kz3lEBee;-Og!
z@XPp0iiVO}<b}1yThPFK0>JD_CK;m#b^0x_`|~MrPSgp7Ob_0?;>K*b4^FgR+r~l}
z9_3K*^yHRiDoD@T2lpAVMJaF=g0xRidB~p{rLaT9>c|Gn&<==|750R~_DvswcHMr1
z?=`8AN?3|txN%PFxO;q+qYtWs+KMV8>WPH-LO(!GC|WypQKRduIE^e_O=%c)rMoc!
zQ9C)v2VB@D@CZv(sTHdF<w~4bkwKEsdVZ*8qahJ<!7ADy&(=B)_x+LNMFk9&u$~-8
z0b>_I8zA~2*M1;J@{1($@b6!x3r^S7ldk{#YSaR9NvwBl=0DOlS@Ce|9o)YnCPfmN
zIDq`v{COR9_Ua`bx|Qi%6}Q4F&>tD1FZ9MYf(8<m{n*Qd#L|WM9qA1wc#<sAc`PHy
zuh<q)uyKK^WF2Sq7rB|AXy9KG0J#39(fcv|v@{0QCTgGAiya{{okY@OeL$*tubC<-
z$dbm2KcH<dN5?e~Qs0lRP$xu%7|-#$4Xc(uwshoVzG=pRVG1A&2!Lw~ct@vp#_7<4
z#{pLt)ksLoTdxe%uJX#3XNLYK{T<Qa;cBON&ryBM?|LJoHuG81v4%zBqd`$)k=*=Y
z_(OSUZ-E|XV{q)e{nwti`q1|Qz<`Nc1*Odl|6`0&LZ0sA8WHm&F<QIri=Mo28<3Vw
zj>pVb8r%1EA3XuQuj8FMGa{7iOC^@<S9PvLLoJ^gy35;Z@5>QV@zJod_a+{Ycr`M(
zjhM#^5B&6oC+hd(+W!b;k<T8avFsts@WIP2Db>hZ(Tz;&SUmD><Cu@oL}4WnC$$`*
zaUiy!fb%2p(zA4%WvW%rALy91#wUikJb%J(VYeSw;g3<PzY@Z4^HIB`YbuIzKhy?S
z;IY1M2vLi%$CRVK!P-In7BFe7`KpX~$1)MmB19l0FzDumbw*Tx3pr9E_XIEpVC|ch
zK;kmjvo5l-F-_}5jrlsJZgj~RT+DNm=K^!M%Gvre1ILgfpmwr)DKrPoS_$XF(D(6M
z48qw>0iOJWP3UfdtF~$C!v_w?JN2)ST@kvx3*G?+m>SkJ&cHArx0)}W*rxg)$4yXj
zp53=pKdt<<VYyEBG~)y5EUS(uVe!Xg+ajE>7Xp8ZRDi04Tf2s(f7ff`8}6u&K2WjW
z^W%Xt9_1<Q8<^W%u!&v8d3z~?5fR-Fn$r_RgTNuR8>YWflq1!lBpTqy{UWv83^+i5
zJs{oj*-HTf(Q$y`6`=cXj@5+;gk-$R7h%xuvF<>WWC8RdmJHe_?l`!)KGY>T)RTRm
z@a@EX+S;FOYw{KKSq`d+$~(rShZ9r~;ZGsc?gsl$vvLv50ouF&Ps(C^<QVxBPd#Ba
z^%o2M@_~!2X^ZVs!xt?;w83$uR#s;chTw8z<qEr5Nd;=D%ZN!4<zawrGWxhP_xY>I
z!n8xV5HTj;KQV>~ji^<UHC-K546Iij-(_YvL9cuBKwC_iLdqilu@$o7jtyy&?qP4h
z(!~n$NE^)Ok?ts;Nl>4=@Fmc8s@02Jx9@-a>ihv|pNvJS>`D(LT_6D5Cx*zu<aBg1
zL^q6}8Af6j`WmucdSPW&`1rKo-VmTG{!j!v>7$0Rj9rLTXTXqwaic%So#P8J`q{T8
z<3Vt@(E3FWQ>~vFW{C|EceCuT`kwE27@qY^(UGLa?)nB6ud*?KO7X}O2}RHr<RKM>
zzZAQxX#W;F%}J34x*-a@v%!#_-;_Lr=9~jovlLh$9OlHU47+V}Pw-$ya*o%EkHj|1
zYu`w&)Cx!G+}>fSdA;0gN}NvnE@%bbT0#fKBjngI`wr8JAv57KgUQ|ys@OqYlvN2J
z{N}(hkZ@{o`e+6U`a#d%;gQ%sm@_c+wfRqUV5jXGYzjIILGK27gw9?cn@T5fQ=$@V
zcMzg^0sVmymM0{NCv%)giFg{o%#5l83OXjK2Wzcd7xJ_zIvhFzLMbv?&nJH#Pfd;=
zO{t5HV&&_Mx<2($cl~N59`zWGOva5)^)g?AE3I3Gcq>F5W!NSWKZy#ndS{H*95E~+
zkW&LkjxwsoUPLy6RmGjZFB%_Z-+r2VkNRcvS8HrS_Of&cy<4BM5r@;Ra6Ig67!n{|
zNSi}QK-VrkN&C4!2p?@(QIyR^_cW=4A9r5e#Y=P&pIe~+6!2xA;+V!><IgC}_o;Se
zql+kQ#<pOb)RCSVbZ(r8SWwKHhmyZ+w<WDEl&1==4`-xX(A?wM(pc=Lq$SuH`!+v$
zSZUs;UC}Cis^%u(oUbMdlv3K`)grj>fl%^HZ$@ier@0jG9*%S`D=PESx5HZwBjhmS
zF8}R->*E{LC>|IFEj=-30MDuJKe2?6p7Mq~Dy2d=h|?p%&Lg5_ArMn-N5&R|mKLCT
z)6{CEb0f`6C5l3FxChuAm3{;Wgw|d^Y2FODao>AlE-XNi-x%_1(9PRlniT1p6|llw
zN_v<kxhtnon48=f4#e5NjO@Zzhb1A*T&ww-knvB~v=Vz?ZQ|0e1C8`%xQ&^_Ls2(j
zE)?dk_UOz(11<G$VabV3FzNhTGyDFz*wN{ork&O^smwwhO9>gjXnP$E@5q{G8!y~T
z5*5gq>v80%s6<{4*o}DT4V(XZ-aZFR#d@-lO7e=zHFf=yYXwg1Guv;i46q>nW9)y+
z%3)nXGF?p236lK@!%*CmHwM4r1cKufQld+h_Q)ce-J%7Q;su`D`;0B&Y=kmrO@7=m
zZt3!CAm+aRwQECMXG1(*IsSx>bs>x|tF~dIPCz=!ihE+}4RHqIzBR>^NZqJ^x&OO`
z{hM4im)xl}cIQBy09Ks9^C*S)Zwx6u6En~&=-4;ji`AOrC{KVRmGihubof+~CQ?vJ
zuGF;19f+PSY(85EYuh4sv-@?G*eczlmjuwo>T-$HzP<~*^sgw7zr5?lCs>f1*{^0l
z_3(bCV32a=3vE4Ll(5Dzcz~&;EK2CKxd6Q-zT$%Y-X|}IkASL$<bsR_ACK9T_+Abx
zZdT<<FF2x5-^`}PF;*~FK_`o^a#o)@bO<&2<7>$$<e!{f_db5`{evRmBbugYi<1{$
zKm<A){*t>W^M-FrwcaTIOP`7%T%)Q$J+m6^mrbjB5xb{6L$d;dlxRbPpJY}K&jNkJ
zED}Rc!_#rkPYOCvc?P_OrD4vD$(dmbpzJludF%7tvy}s~N2!B5IRjEyc1`U%^e{}$
z6~y74h<ezy-mNh*8TtK{<YWTerLwwFNc6~`5>6I>s`GcpTX0P*=%5}T?I%eTHU}m&
z{P>=tH&8h;xGS1ro_Ud8db_LG=tDjJW#2%XayR2fpMl2vZO>ePMOg9T1pHBwHJSM|
z`ACRqeSdo?%kGi=Nu{kS{4T4F46Ax~zNLDE!4mg*o`a)Mk$zKxRj^>;TUstAfpS->
z3r1EBgKy_GS^5|%=?m0xpvMjc+zW-CeC12*LPeZ1uPn0V{FjUc9_%Q5@&sXqgmH$N
zBAik$<CY%1(zi^NFh4kQ*)HE1wC8<oQ5iBuM^HSWq$dGrta=bLScP`HLK)2Jn&s{H
z@d-6rm1$`PvHI^GFEW3g&@r$z;m*;!t32^-)_f%~*|_)afW)L~u-K$XHBOa$G{<td
zTtOy&B=no^lat!5!1?gN!3|x@Y@bg_PDMoc^k+hKn$sze1xE%<h3yd7*SPaJ_f@#z
z1VlrDA=I~3#_t`5d$YnG(AZT=z?4iEl&TN<#a8ZNrvD|^1Smxsm!9v#r8>I>3=|lA
z6&P~)F8o-jZ%i1<o2luH%5Q~|Zp;=)ErtYMh^Q*i#wvhMAEw}qNe}LG1#cFeP&aIf
zC7d`%=-q}z5{M@0TL5nlRh-yDA*Z~cRAyFcSCl{a0%gb6=0jqu=S^tKP2MNHr#X)t
z%;tqsZeej{l9mOP{>4qv1S1N*Z6jCZP+{1KYftfPQIZ>#=)jz<c#r@wl|;sZRld@q
zI&FJ}B{!H8=lqMpZ@cFHh14W|@ea?k59ivIMY4b1Z1&k?JGBusj1p5Vp0#W}E3pL}
zIeNIKM(w;Jn9x)0e`Ot@iA)ibt6Cn*LUf_%;N@<HD|#<WHZ6FQj_A2ak}s{-+~FS^
zs|s`SkAt3S*R=DV+nZW~V_K-$!|A|I(W>Bm{w3#+3(&(6cLdh3zbaxw&!ojo1~maq
ziLIxtzwgf%J>wJ?)qW0zxyLaOlYuR~CDq=X-d+oj*EYQaKR)&Ic+ZQaP&tHL%%1X2
z^=6R>#Ok*4bB=?yv*B!>G&W|OYmG<6Og#efY&&+maE|ktIYOhU&+5lw{#m?wPYKA{
z_x@GOb&xVmFWPSt_)-?5C<kE&BW{cgPfOQBR37ZL1J7kad@7)cin`{IPU5f8mr){N
zkMhYrJpx5nRi2j76YK6P5~WKAhKB0ruo!JJNbKb|)4OGoGa-Geko8_Qd#P(clH${m
zr~!9^#Cu}j1Me&6$mFtXoA9*0U6+XcZdKEt{AOmEavw26VdW5f>p~r(iD#QAZ@w^7
zE7dNoFE7qv0IFPYJR23ME^|}mRx*&4*h^zue|p~fBZH09B&y)zt?0W5{eGC$mw7?T
z6!wfl%!r*RvEx~F$Jd=@Y=1{|t|ypMmuBy~Lfn%;?><Bd<b{`cW;aJjgTjIaysH-5
z*fp$E-N5%Fdy3`~J#AtA_I2)?8{;H`IM?Sch|yFnn}dIE_^)~}cdj&YKH#}*>JK1U
z)l?q*q&1OAvlo1@yEO}^fU(X+_4{)<U@ifD-ToY}=oTugO0S=ZV4pKu2A^lJnaw(s
z)Ci1odj;qBM`)cHYF>UaDLTx2>^`e2Rn3Z{8!fF1-rpZl95OJf%~{{S$iNJm-Y@K6
zP6%x#I(u99Bge$iZ{^5aEWw6w-paT;!g9Mj2U!GMcS$|!>osK#%mj*XA`uu4Qk#vH
z9Py|f=cjE{kC4iq;Y6xdhm_B_)pe23UE_bsj?jx6BUxsrK+`eEP8ZL%q5FW#rdr$q
z0Lb)bp#i9_;x}*V3>Re<sKB|p3xSjMKnuq8YZTuq&tQQO$dylaPieFmSG?@tY&59<
zHJXD>=0pD2!-r>J;Pw?9?upJh^3MEYZpWDaQ~zY+H^qa){+-xqy=0uss=ne($h2hJ
z!Tjo*)i$s`2w<iuyPL0d8yMVC%53Gb3j3wuzb#+<>tx_dagn(WX14uKBe^w$TTTzx
z=BKHvn`S{4@V&q6csvK7<i>M6+_e1DB@-M>$mO;*q54vhmz_%g*C?|9E_5^?N@GQD
z(^3;#nwj9&Z>Y8%hdRBeZVb4NHq&$sYdx;nF4{7TiUiEkntAuLn8q`myTF1Lb|Lwo
z>6(IuzO*#_r`j;6P^fdjCwGv9v>4x+Hg6lKMv-*|Ou|kH*B>bxp`YtTmD>e}VDasS
zkSH$s1HBh)5loD@LuMpcdafaquLOjD<gn_L%JJGBj&FOvBR$0@iiyUzl&Dy4XTh(|
zAE+E8q5W1(OpS(_{PHwrJ%6N7R(qz<(7N{J(^$H6J$WgYF8m&|9|oHQ`|^68PsimM
zC0wf6i1y!Ez9d%EVT!r#R>nYm4bPKrHdhn^WR@h$u&*L}Hr(Q&cUIEj{c?LmI1GBa
z`9PN#oFCgE<t*`C&4V7Us;7>TLPR1>g=?|hu3kzi3Bd(!Ila~8D)T&xaB~{+MeKQo
z=z`5*#_LI`#Q|}6biWaB>EKImQ?m?hOHyznUy*YyFBkg+m!rC1SQFSu@{Lq2G8{Q6
z7Kaj{mtfU`BDT$jhkd>|dH51A;+-n}LC30W$|i0uYX3lE&Ro~bZ5ti$De#rX;nCBE
zb5W|{n+_5!Bc7y+vm{cbgXhs2ciUq_x9FaB$sf$}JJB$}y|%YqjA#~_JkrAb0ejzo
zv2Z_qM)nQ9duAVf50?MVE&shC!hf3XmF@ul`30O>w6BJG*k@%WNnt2{A99y|(VrrE
z0o0%^P~Ap%HXeZ=)BUXVj(a~*s-qEma)$S>e?t^g>H&(9)_k#_#j>q1GM>r#xR(nv
zXvF-KTmB^5;|~H_ygX@88v>_Ht$B2>?>HJhN(}RhNh{p%;Tb1=$~us>EMT>Ll#+8L
z3ku5*N=;VZc+t^qNNhhO5R_74nrH{}`{wlCV@7{ssY>20iDJGFSvt~g<htg);7FJM
zV=4s+e2!acIf9CztuV;I&@`u;;QpD?cHO~422F@{tD5?M0z^+G$GrN<g|DhxvSY49
z)VLwNDETJ?qaX7?lN91iCtbBzT4CjiG{rZX{^|&LUn}>+GN#p;raYn&cN`UK#Sc?H
zn64DzdoqB)iy<O46|cw3%|a_d?d)54Nm%=(+owE}ou)*O0?0*bc>)~1JAJrf);4*M
z8MISSqc<FZbNZ^`A@S7Ysz!Nn-0(ro$v4@6f_-l+c2{l{*{6QT1?xzxLf3r?hwC!`
z;c;Gh-Jg#kXxD0}QZsrBXs0Fu93OdUb{C^TLX#Vo9VTeQrm%Ekrgmy#q+_6Q(9svo
zRoqo?8j1J8Lc9jT<pe!Q?ntYl98jO*3+edXb~d&z)Sz-A4SH|3?^)kD!kyt2<kHB6
z+q`!<s(uCNk~()Sf(?WjahrPKTK_da3=knJpki(F(9VGbGHRJKWF#|yP&ChxQ7B5*
z%|yn^l<|nw9vE}ijoxV5jYeueRFH%jiWeBlIvmnmiYbtY*RVgl+K3hbO*`xPR9^W6
zR!=<A51negtNv<P9c{+d?>YzQMdDNqu}m4?yP?USp|MllOrzbNG1phmi`az(*Vl4c
zqeEF|p3Dkma+@nuW7z}mobq5u{e5*BTM9XA0t(Ph7;fogu+2504DF{Gc41IawooFP
zkfbv)<4N0?5h%`&?uwHwt&0FhtIJT<jSzM*)ND#Loi@}-)6EQM8UdA4HW|hu)#h&g
zh<ykV3^skr9$}smPNa`~blj079g;-R?$4sbpohA+MgsJEtwFJ?pg4H!fTISAT?BnP
z0JZoCeK{4v3J*PK!51*aQsBdDXcftDWks|B1BG{E^z7e!5_ZJA>GBLPG~TJhlOVy`
zp!dO#Gl;61C$JQ}?c1`|?N6T06$TH-HRE94&2Ng$9qtDC2mb^dLqB=OZhxdjPNYPO
zQN3rlEd`3>mQShD<qe?6zR~_qZ29Mkzwc)>!=80ef0~&{!^CKUDsxpWHn2Rhujw1^
zN-+!eCqshFAs5YZ*4fW$OJ2WMD-9*8xf@n6#A_($n_Dj)mdkx#dy{jRUHELyQG(NJ
zfsld2PrS6kc0ZAIFRFP5vzS<)j<(f`ym!AOsXc^9CzSPVC=uGd!~t$Coc}PhHfU5U
zSCHWsdL0v4qKn(a!T78ewXP*l?>vUR_^sHAVO*d0ywtWX(VXGeO&*a<yYb|<o3Z^Z
zn}5fy?oB2@a-h#ZW*{?tM)pyBZawdcwe*~zlncB5`8M+4aC!qf+Tz3$Rji-)WV1V7
zBMvb?S;c#k8ith)SE@E=r+lb*q^WoWpLUa6<kILE`keoWshCp##+<>QeLi2xAVl_+
zlgvdjLwjhSEdax#C#vP4gZ<7|)I3(Rn?$!8tW3pA|EXx1Dfg4!3o4!IiJ8QNXRf*)
zhO)OxCd0tB<{QvkN0V`tdOlS=0a0u?h)rU>d9$G4TryfpWo|F`#6g;e%x8L0cN2DL
z_)m^Iz=YhOxx1mi;4r`lh|v<}^aI(j?$fTNEQz4NVx#;jD0RePz7F)klH$@oF!BLf
zjO>QJ*hOrQlAGZ3jgjlY#P0K#kY;3j1^?q_9&?Qwt;($e4#=6%MEAF<LY%Tk_DMn+
ze)=?n+h_Z=LJm5{&<=;nhapG7$^z?}IFHI(E<4S*eUF@cJDztY)uW}AC}xyn87Mg<
zXMEfGu)(;f^^M7v4g*-^^|Or}Gau4C^N~=;5Mk+_UIFG74p_5|;633{_Jw5nJ)iM-
zrzyp-p>;2BDC)M)hz>bMq4hrVbY{Z&a*Ks%|Li_dvsFjW1ZFXnNnQUf^r=){>_WuH
z_;Kya9muy54;&BpWtLs#t^>zejdX~^9F_vd@#W{mTR8p={&xQk1GB5ttStFtZaSK~
z;?C!t!3<*Jy}sHv;+!gOowINTUyd8Gn--Ft-U-_R*?c@{FEJPwa3Q*3IS&C6*>LJz
zp#SNrh9qE42`~pso(0bM>e<=6Vg?Lo6XSV;r_LKZcC0Yg9YsLg4}krNxd|1y&3mSo
z=(%7pQ==zZfX9KeQ#6s9&p9lcbFFj|KjWQw^s3vHY-20zd6nY%FzkBbRiU4Hx4@Sh
z-^U_5ZL8J~J$hN!+SDV;j@6@+mDc9|Lqe&|IIbgYQJ%>z6%L1(4Xk2szwEx}=NYix
zan#Q?^Eb1QF*&mJtv)s<bJhJVH!iu~Q}~l$uUkt|3bjj%6PfW}k69NKlS{uQhf}@G
zbTSX@taZn=oA;);9XRy{pF{3A(9v5HQKM1TK`18cL1hBKFp`@>q#-=?r32(JYRnLp
zQcb>^CVSDL_d6D%3`XRpFZb^}0%aqVkxIBWvkXm>4pXV<@&JEAAbX>|;^$<To{)9z
z*wdVTqDd5VmdQ-e(-JY1hh%k;kNdRkTe9IeiQnvBxrRysJ)0j;>o&n73-QU*_E(?T
zI=UY-iYu$0=E0wb@A#ie`*3Au^4q1QWB5pCjwbPh!Qp+8oiA4Uy+7SCeP)>&sWqUz
zXZ_65OLw2S&&sTKLb(N-jQ*;FT})vq^pw5s&0|0q<mSY&U{|Z8-H`HJ^gkP=IQeij
zMf8rw0>4g=gvI1k`E?WuBlT`D6tLAR*)zsZyo*kGa4-JVg}IJDq*u*TPN~krYxh_}
zpOJ>}6YVyMER{z+nRRC;F;Zf<0GKz?WdUc8-LeKwAKPSxR-*MT{f+`LdVxz&<+^0j
zZL>VN2b+f`*^@+=rMYYnPDiH0Nh>lnu!&PtuL8lI<RDU_QAXlo{wAQwkrJ)H?5rOG
zg7>h_=`dNHd^<Yxr`iJi>Pm{=Z=@+h(T{_7!XJg+#8Xa5Xi-nggAKI=rd}RH3g#{e
z0G6CzAAx?1z^td^?K=GyK~j8>J$*Whl_v5wk~VL#_E5bJtrHwwIk#3bQkAeuf8jzT
z961^-hD1|B_B>Ju-=*ECI%h1nBh9}C9;2G<2|m>HmobxD`KTDy*75zj<7}{eDxzgq
z?Q!!bk;;Nju5xCze$~Z+9xH6vd|4=K%0>TDkr=H{bs5rHs)(||S8vqJpmXJ0xLYro
zM)CRXKjx+HgisWU05P10^P?Q(060^&{i7X?DQ`h>Ng~W3*zSBi{QVOu)ILe@njX}W
zfmU%K-@e#97k(r*9d1c&ZS?eGAZC#LYg79LM>i#98j9m@bX;SqVU6xp@eIF9n<Xu~
zn<yU9#(sn;PK<J;ha|o>w5i&-WyRkvhgr|{EwJXVLTfTy_+$T|o^t{nC2wk-5_iu7
zTNIaN#$V$8CyyI?^`DF0NT|r@+o<u1fSVA>>_p=E@@20ZI?xQ=;tppf1(ylcQ>pb3
z8_N|!-@qJUlnW<%40n+nARmkE%Jt#S-opAzh?&(D7}3uhETSk3kiQx-q{*QIj*cJ7
z_hjhL-_l?7BZ^~DE+}U%=U?slN~)mv;=eRY4nE}Z5C7>9PHYZn>{8Io9Uml!W{G@g
z*9j6!;3zDUEc+{Yv#pW=Rq)baDc-!BnB(gGuf$?bg(@|G4rAyFxoKOEVQaZ7k~LR_
z->?pllQCb(fcTtVtOy+AQ$RBk5EKT}PoKpbaC#HVJy-YK+tnc=0g6`x2rpjz=Tpr$
z|Muye$AO4vZsYnw7$I;!mX2!QNdBiOp3BHF$SDjELf(n+^N`8+!7l2hwxk5B3LkO4
z;-m2@`*<d}jPXy({}Acx5p%9;5@qhp%d4YE5E@1vBJh+f1WE69t9zx0-*HzDYLj1K
zLfq?mt6&u`fF2b)LZ7Z&hG2PMIAs^HiWjf-n0v@0%OLG1s(x66<(lV731O_n?r{DT
zJ5I~VVXp;ol#kdY<Xb~wQSQ8%bZ?;;W1Hmb1Y3r<DXhC_PE7Wjk^LA$sPo9NJhcYg
z)E(W8i7M6kZCyc4`o~hii>$B+)s&J*_xvyvoG3duCQmm_lV=^D)O+218~qmF_B3M9
znV05`6xSYDui{turpml0n~nuCB-5V3dK&FmB^GLFTfkA^%}gr`w|^}rj}>Hz3s5`~
zku7*mBWp=|RJmWUCx{@Lv{<%3uCCSxOlvH#c&dEx>{&_@K1uM~OxB#ayt-{yA2%;R
zYL;Ts3}R}|zyHuSyr8-uU=iwiNTT^$=Jnx>$&1wJr?Fq3St=99F4PCt`z<f1Es#1Z
z)`S}BfH(7M3l1U(v*AgQ!0nCD7QhQ*^cd$RS9$>9gxNGFY+e_yrgS%yC9jQ2jx?LE
z`_YrZ9_|_7(gO4}B9o78ks(8}>Gbl)t9~Bs8vb3yQ=mZtXXv$=Y1P?$ugVY+n3Bpe
zg>+F05iDZ{oj*QjHJ=AaZQU}08{$eyBS@D3IB@`-W*hC?0dQ{NAWXf<NN%`$Jv<4?
zC(}v>APYCVmtxi47;BYA(B5_5{m-P;s78WFiPw+KEe2Q<?@pUv<Wdh%z&EP_WVKsn
zo_no#zdzExVq)avfHnlq#jF^Atf|TPf)QO;zoH6Q`=Nhl3fx4XnxL`LZ-w~W{k<94
zaehmKGMbIw*pR9@=2w8`n`L684NtMx=M=POIe_8BQMWeP@QQ*{#QeNy*`4!#Kva92
z*J&3R3}Y@t5kKFA&6=snoTj_)qx=3CSJoshuz~DbXTF!%lE7LB%G*SmNFpX>6V*=X
zKWH_0PS}Ya@nWY+M|E99+CQqF2LbBLfa6X;wOOoUp}MVnmAdinAKJ)LkV|4<mfBsP
z4tU3$TdwPOtrDo>uWh_r`S&*NJy8Zx4tNWjPVUzLgU-T?Z_FxsWMb6qkbItei!7Kr
z3<oK_9kU{K4<Dp-Qx!TJ;cZ`rKJ@z<>61kmrUrai{324bbIHC;Ba1oLCG}rP^xx~&
zw~(L%!nlU1x~&$Df2`NP<+@1HCLk9I22C`1<eH!UYNSraL^1%yD)dDIGon4}<B%NO
z0s+U0!r~pFgI32md7T1H+}#V`aREE|x=SjDUMW67*OKu~($Uvctzz^?aqdEZ$FxXa
zHvxmj-??6mdIA#X_XyTR=5VH7FYhoO$v*kz>1QC}jEa*Z9=%Zvw9R*`6FSIPV2ZCi
z$^*zS6JTzJqd6J>-7|Ph61kW<11L>GXR9!YtBA;D9bqjIcwzugurlI&a{cju*JBCV
z5hJF3*`QU(r_3PMj~IRXL?5F4N}KC6>_bwEIfdrr)+EDt7UP`_YR%;Bwz0sI%<6Mv
zPSQBNwQH`^H=GHA;yTt6+I=2hJ!7h7i}ZOsC`XDE`2<%dz~ctSUq(w;(8(r*a{)#T
z`_(YVoYg0r*mKLqtjh0<iwZ0;3Iov!Vm1aE*VRO)g+xw;KB6|tkx7m8*Sub-H3qHF
zm$1SlzpNF`Thrn-8@X45NJ}mK2H-&(lcGurq-qDAk?iQ<1R@Z{_;j@8ci$DI=^*B9
zdFCetH5-cxlpeUSzNwSuXM)ViCidBHso3hGW`lY$`z2mAjHXPqsns96dcyfa?BnF+
znXQ|_lTkexxywbg1#-7oD%)41G8R+jRwaB-G9|NOr!cAep_`piOIlaI+2Td_e!cx<
zW}3?_7H2h`pAk3Uw3wkO@|OT$UL<T>8M4JSFpQOKoS-mvQClpr9o+F)7!r>+LjPqp
z^iF>d-3?{46T+3t%V|iRM+6$Cwly?YbpYp6bON%V%Cz`d)1!~fmZm^XdYR)myVqB^
z*y#TD7j2Lw`ai19zbg``TT3A<str<wI0#iZ*d-c-tqum<3lH9}9_pNn{w-`%Ac0?x
zqFM0mlc+cF*mbJKR#`0zX7y*;RDNDJi)9CM0*1#)+yeylt)gGPE`G9UJS0QEGGSX?
zFpHng1SGndC|juCYa^Ddz9Tzg5cKCnEY`^?xZH((nU`&NH?@G0Zs^#d&Y3bDFh39U
zwCT&oSUqM{&*WjuD%sF}_tHpDrs4T@_=~>6*Ec!wIq)QCO0J6H%{Gtf@H)3j?*!yT
zu1;XRm=05gwfY<8HIW(hRT4#==FVz)o&B6@kl0_Ovx+)Fi+4xK)E19=s4UHLL3g*7
zuBnsaaH7*kzcU$?n7?K{F1DG$G>FoK425u7TBbdi5k&&dlkP36qBsFpWVEepjI%^S
zeO$rvWEWr*O7ymwXy>w!pq-XP2f{XE5T$E{H1V7?vC`E{d&5$6t(CBY_BKm?2Q!T|
z5q;=lJ(e@x%7w|uQDi15;ZyB$N<Pxtr^r97vK_V$9$6}M(0!KLgP!E|djbU1Ed6he
z*V^ymUUUl8Dmhh>8SGKdZ5MFNmr`qL*p|~CfWKnbf3!-H2n6WtQFawI3VedwK0NQM
zgW9y-ssp~r^ln2Y?~Jx=l^z@L@Ozdy66vo5bxiwxMhj1(n(IGo&}lM#ggn1BRrsym
z8LL<qvh84NJifEP*8|+os$v9hgzG#hJ~GSCe{cxx`T}&N`YzgW7?lTB9^4Waa4J-i
z5!2}iBPFWpRE<2zO(-9E^wHeNizcn>^TH+X12=)@M1}YOqHFD4m6(*_ha6Mh=w=az
z`8MfR846^qV-%X-LCMs+8i_P;t$GSr<D3-cOL=74)lWr#=yIRwNu5VR6rPadoQV<u
z<Jx<c78TCkZ~c@v)1$g3W82sQ3ikZQV<$&H-@BUXpj$?({=9dm2k|fsSrQ@BnwJ?q
zB-trd)xT;M(@ghSh4A@`{}cu6E)~Oa+&RT)&D;Rp0#zz^qeXxE{Bj0@>nl0>#jH%W
zGREcdwiO?8b6=WE6M(vjJolzC<UeN9f3^W9Efky9tK2f?44Mq}di+K-=bp*2)ZW@A
z1F+SirT_hMm?^6BZCGnAi8GS8+qSBgg9Y{-^rbHHVgBR4Q#q(TLr}J#;m7sK)JqD0
zGUrq;uiz#A7I0#>w2CY>k?($*{&EZKB#MzwUV1)TsR0m4Kx4&N>Yhi|<J#vmUG|v8
z1CWug_52RLpi~NuF2uI>DZ_unOu?T$OkrvszWxT$L96RYr#@jg+<-Mrxqlv`QSin@
zn2M1D7JwcoBKz*~qDFbC!L7?hr>pZtv%p>Ja<gj$xrxCAZCfP_0eNFHGu+rhGK3Xx
zo|OSZaDJ%K<V1$CZoMC`i^3r7koU@I$N)4sQhU?Y<%CSMRw&K_2S!sD3dRa=trge0
z0T)hUcND`R5C0#>R5u0*^tyPGh9%cw=qAda_$1=cq}dStH7qL+#^Xh$^6rYKE#-i|
zk9QMrJXn&=Uf;6BQ8&V4j6WT|JN2P8Th_3bp3`f|MUkr(_Va!!@1gz!@1IKcakQm#
z^5?@w6!d=-7F@;Rf_5|3J*VK-hh}6~MGA96MaOOi6`l~lc)niw$OXgtCI7PS`=qDs
zs?KPNnLi5aA!N^44ZYXGyOo0CS+~aCzxPi>HymH9&F}}yeg}l}#eP1#Q_*HnX|6wD
zXZO15Xec|XWRF879NCL+bg`>!w&HSiSE1S<A87iKhOzhTkkchHH?mfs9N49~lj_TF
z|J-xDohCkKEgr=TIc#`x)!#HVAWrAtPVF3NYta~F{~4Gs{17Q>y7VIi#Um~d$qTLk
z^$;Fg<;<PxU>Ia|ck{`}r5I&!pyXIhazZ|!zSaYjBjx1&rASX!FQk!B4^w?baZWYw
z1X6Mot~>)^ti*DEtNuqEtH*-0S8x)Cs-)RAs$&0UK^WJ{?fTPKh^USj#fof>zVJZe
zY%s1Daf3WFt+r=u6rgvbF7nd$HhSZ|+t|sCMqc=$*$v$d^i;Z2+Ms3cQ@QUSSsHEJ
z6lQa&8}QAm&4`*P{qG+0wQ;B`dAtgyZSt#Wk)Q50sAp;Q5(DdEemQ*B@4zi4orS|f
zovk&vaLT&jo53es=1*@37^ct}A_OH&6eR|RyfantLM)Ms_PmArdErF3apy-K$tQ&7
zzW^;`rHDuS)3~`kiWC>yRNM(k0He8B9U--|mgCHg|D833Zrl6lP#5ykVZ9Ln&)=)c
zY@athBAbMMDv4$zn+MjF?wIo$qVXD!hw3|c3K!vR8~7KIDFXd1Z`~5z26}XB$TnLM
zyhs4<;eIkZ`>%uTpS&zX1{9ufaO%E}nQTrn^1Qp(43ijIx6M^LS<(0ROQd*_A%TUP
z>cI_^k3g|7LOD(Kf2fmPzokI6xy@1M{Wf5c(J#fELp+YwF222Zg19tyKQIh1eOSF9
zs?vr$q%PeRsQgMHkU*JIu2|_6!DC8`>?;2)Hnl#aI$x5JWyhGxgQTXBQLDvUk0mY~
zfna$*@Xg3GHHIO8AR?Yok_eq44B(h*(h~X@!Wj0o9%LNVu*;090xqIs1Eb)a5aK^7
zDDjs3qb9a3!vTz&p})42jNC!uj^NAWPlh@Mqm0+h;R$0SuOk`@<ur`TTJ7}+!(a59
zu@s|hqR;chT1w1FmC~+dd;x1XWg1?`AwAGtK^A0RGbEY}2w1U>c47&~Sku0iZh8&4
zuS1(hA^k~LnP*wwd%Ai~EA;Ey9D^BMocZ=$jK}q)wmC=Rd<4;QOID<w1|OpZhwv1<
z&lIiO4MDR7L&+0T(W1nm*jh;M*z|Y;84?MiatYgKfCRX9mTjvDf1X3_MII@j`Skm2
zQa_@yL@rGso>R*f7~LX&XU#1Iizt?*m{1A2vj_WUQGZrxVD*cF5l)WqWY6j-wWqKL
z-?0M;iuz2*dmS+CwR8lRouRt7{AD6mb<nvAMZBT(jlOd?+Hj$bVbQX5I*Jy|iNP+h
zXV5iB%KosuUHvjyeL+o0)29q&SY*#9w={5kcQDDt0<+fZO>81Y<&-+DN`0q}N|f<^
zVAa&wMl&bJ&|`DTJuPg)Fj}&bmZ1xg+v@Wi_loj^51rvla6SBau6;}Oj>x%sz<h!V
zKRrO^K@}b|tnytQ5=uTP(T}>6NV^P)V-ic><Vc9!*S*zRr$EI~`qp~X(V8HwodvXZ
zxsX|?WB#6ABe7wyVLR32r!lu}6T?gs2I~$C`-t7=(zuOqtAkXJ=tAK%NM^KGScEs^
zKOc1f$d6mGXt#mRM9^+1-JmfLGbG}nkO)KC#q)H+3)-7W<5F&6H&%g$Vo(j?DJ4=5
zzet1Fd;im~>eZuFk2ii8>oyN<&}_EhL(s1)SICVJG!vt{3lF+~j>vy{3$_WRc0)tz
z$@Jh|dr|+Kqc8_ly(Z|II2xF8^h*LKdJXO8f8Ml#g9K(jYT2(0;^zj7+RoCw&A^$~
zF4X7tf?(Exu#?aW!4!V9nAOz-W*j){{-J0*;O5gH#zbRGen{El&)>IBrd{G0nWK#K
z&dlgI(UeRGdv^90h$<%QLd~!o^<*sQs&t?(L_GoYPFv%TFg8o;=Q38=a)~uQp!vb`
zQz*h1xw_Z!EEVUqOZ9q{QPDANzCKbv5e<FV&V~%oCnOTGzEw)l6ER=;J!hV9O>>g|
z^?Uy6R=b#E6cn;48ze0;^5x68C0(yG)jJ(W{hDR3B$uIdd8Z0enLF8BI4K2lnOhFB
zr7~5CKP$z60dx#tU*0URXWX_p0|X_%hb~h6?SLTyWN0HF;UC)TGTDaKt<UMV!bpU|
z%5+HrLs87TneKAbtewU^Jvf}dwha9n)aOf8lW-={t~tIer!Zf9G01%91!RpBwa;5a
zROz#)8;utYC#7kmDE%quZS^U1`^@U!%BuH2(w9Usv~`Oqz(-g%`rR*XD9S(7Tb>8#
zaThr$%<Qt+u-P0{znNDN-@V8_$fY<6aN&4$o^!766QqREEFRu+TTS95GBa^G#rckN
z0owvC6DkJ6>M#&Zp6pKYm03(|B(M(@0;X)YRUtGC7YU_5I<O)Wh!SUj;KmY3%m#C=
z^!O+@+Ku;3P%W2@#K5F?rvF8(B@xc?C!Q;of!SfF$&n@jGy&OSJpSu;|27R#B^ni$
z8S9}Oi4^PV_4I8?F-5<zc0Q&Lc%y=(#m#6ZX1c`M_S9r--@w}iWQu`Kn!x%jb2Ubb
zeYX0&wI~z-a}<TPcKp9E-~Nj0OybA~0N9Y44C$}6Nydo!0NCDJF7%@g2&4Y71w(%^
z0RE0SPKFZAe>$!IOHuk?w!we@p{fLIl=q4KpHu<!0T(_|*C|u3sdK^qf_B^BVpgt2
z<dD!)gnwj{&23BX|Ml;~o*{d!26zFLT8gFbKLDG;wp{Wij}LGmLlAr@zMt=I{B^&J
zj_-Q>%8&@b&f%-Q9^oAi0~p#72>U=q$Iaea7_qvu*LfU<`(HFvLS%C=&{95VKfz-9
z_rCt`7yA1U{6D`p)QtW(l;)0>(xs1Rt@&r}8E(gpJSR3VV<!6#FjAcc2N?Q((l^rN
z%7|#0CS$;qI&KA;yC^;u;2A@M(Ru*e3KhV>`*0i~*jC`8@eePJs{wIN+ovaJGyexG
z@qhaS|CC&qD3SHDS`hFp1i&V(jUDjb(CMStH-G%=ri(<UiG!Q~s775+?684Ysi<dD
z5Z*KWQK#3?Z8Vi*{~x`41B7@0(sZoR|E~|=--^%wd^9lt`A$zxuJb=hMlg497Xs*$
z3$RL$TV~ris`pi;F=0%|bylON<HK9sHfbRp;#x$S_gUVI<vcmwmY*p9^%DJ$yZitA
zKf=1^<bk47je&r2=D+Pf&*QoPJO-)N*n4t`i;o&#mj{F1IbVE#6^TGU#Wf<h2ay0K
zsP4qCFjowqj<1zwJ7RMWltLU^c1uk!=dGPuX7&?+`C;scX`^e?IvNKTsjqVZEcap$
z0{m6l^l7t*web1n#sGyAE?j{+P0uWv%VDAZ?%kLoQt-qllXuaF;cAlpK))D2^?Qi`
zUyVZ(g>qF2tnbnP``zmQc)r2U!+_13TvnZ`OZU&Z+rSEb{8EqZ4#2D6bOLlL<fl0`
z)PVrcDY7K(^)p~Ki<vWhL~yTqlQksB5d<)wtTX^w?7SY$9G`iP$3}Ox;^#C6B<8)q
z6X5)V`V<b(T!x-QIDNjVpx^uge)`A2V)+yH7D)I}+9h{X25H{<Hv#GIvn>>rETD1U
z+8Y34^q4%Hk}@Abs=JU8pd8#-_J)kvy2g%+&!J2%OKo2Nfy1r*Z*VwHX5!DPisgZE
zX{u6tB{|B$4cA|)0vvAUT>&P@TXO2a-=>`sz`W6j)R(VW5zH#~alDP5t#p&a`vHJh
zC3vLa2ykSiZYS7bFEc$=(hvXtX@40}2<jq{imuPU)I}~NLOq1@s2UB6@ZF>yKH*5i
z!kcpE-)oVPjqL{xz!PlJ2;87pC2LsmwbLLjjstQO`;lTNJ*;cs>P3PWtp?cCZuYAP
zNDZB(e*-3YYU7*9bo_IDxbUCfgOcy11T|4nI#5jyfUNSQ5#uWc_-IUM1BQUz`97GO
z7m@-8+kR9Nud!vSZVw_FbpStQ#PvzEcLfL}1$r(g{Q_WabiIgAn619_1kf$~k6Kp)
zZ_z4JOdZ_C!lU&{i*!2`RG6yAM9c?Z3VyW!3MK+*#VCJ)i;ypm*9hLuVTA$_$3PEA
z>T(jVfB@cce%->k0{rm7=ab+ZfF!gApr;PLK0C5WwOzh`<^O-IzJH8+|J`f@D2Reg
zq_*~xe|NE}463@s_2x00Awh`M4DeCWOkv&~dUfK&a<teq9rxf2l5`_v<W8XQYOA^-
z_^U5qSH%Py_=whw$UhJh0Z^rsa*H0{jADRsTnYjZ!{G4*=4dMb#=GfSIH{*fMCbLW
zQxC91iGfbgJXDC^J@8fV0d(4L8%u{59kG@p0N~El5I}vt$&3%b2`7wC0Tz}xAWY}L
zBkeqIgqG6MFFd9v05L(d?UyCznZPdq5Qv4q%M&9h)~*A0Jz-N|H}7YHgkBE0Kjofm
z{I}Zle~+%PzbaGU)0@iwY+)`_!ruTCX#XX)Jcu$|eu2})(SDmgEbxrmZl>ZG0boOL
zt<;f6*=@~`2xE-__<Xqm86{_B!#*TCiQeyfrd6tKj!nY=|2v*woFA5H`U7A60alHp
z5a4s|0hZs?yE0%KGa@W*ra=56ki1(XSjKZ2mSu1*Fah9vPoL-g_m}#=e}#uwAfi0_
zj|K|x3Lr>`r<H5Aw9j7C_Xt=yzh%3Ssb)@{14Mu#e9Pk^z!=f+IlJTS{l3W>K(MGW
zKxW`s!j|Fa5@)$WP(DEuVQx!!>rCg2oIcN4wTpiObC_{~&sS~m$nPZ9gtFQu?4nt)
zeOHazYd7{!`)`2p6aj>f^c-p|Kp(SMY&%{D=qLNO4*;N#<}bP?f^&Me`sdz64{+-B
zoDOr+7z3~Og5_m5>4(7;CAuDY)gVZIOY~C)E2rQ;apkKe>BcQY2q@OpY7wq8lK-@p
zVsq-30hB-D06oa(eH7Xl1G(=PkP<n)0qQ2kh;wR;;o4!+F#!LWJmd1%d+}#^ca?T4
z7_H@;eSO;x?#vDSeE31~-gbkBN9Eex^<a_HB_7Udx1{ki*l8^*%lre#6VMWxBx?ah
zCw$&IZ+S}fp{(e+z$*a4#RYiUykb7>8h2j*9uVvxa%R8DIr3+=2VkFSE+sSwydw&@
zzbte*y%q$n$C$?3WAiWvD4(Y{+ztzTje+|9RZ&O|*r#$-U6Vh6ma;vdF_BK&sXOYl
z6rv;kvAmtFl>MGM>=%EO{fDFPE*0gf&vCYZCw)gRb}_B9^00=S)sW3oa<C6Zx8>D&
zue<ju2E{H4F+IflM=(a)0qqzfxEps?yI;>Vc6|T5UGmdU7@wZFz0crMl^(R7OKIm*
zZ`!mly2vvW)p9!K)xrDi=f+sJh#t5s@BBG}9@x%VI$lZ@ZV|=y`$g5z%GMK(G264U
z1b|Rz^5kGTg;fXnFrt+Rw9hgny^!U3J@MbmQNB1b>;TKjyO&}?vRG6mUjy}9?;UX9
z*gqM+BoC+l%kr-J1a+EJ;~$zO846~#RTMs@5vzzZ8XZ>>A`PNXuRQ&Su0@QEl=2Y}
z2NRN~Y3)CEKHU5z4id@$C;o2im4@VfJ#h>zi!_HXz>B7u0gkt{@IQbS3i$=BT=n;%
zn;5q1zyk9!>-OrrHIlTJ8fizc_nylk46Mn4u9|er$M*=$LS5(wwH)<IaK&-%RU32L
z2G!AyP}##oLLf=J#ElK}!Iq1gdW$A4;^AC*$6Q_GP3!@jNdlll`ZK|Ljr{-(nN)F?
z$2!=+e(h;b?Nu|dFo4J=a*e!>@|5=!|7$c~1p<uM7It@r0OJvj0w7^3=MrqXJ>Z=2
zTE~@(HuD~}Gu}Ab>Wg3J?ft#XhD3;rS1Z3lXYBy^qSt`-M;thokHRHwWZQv*lbf9`
zq0C~q^za6Pr`U8psge_gYD%0w)Kn>-dm2>I5{fSZ=l%ereHC~!mFM~E?C0rOV6<Bs
z4&5Ojf?l5f6kH8)b_ZP9U#3wGf35+HT3o;vooe22)N$);I8B6Hi%evVm`0K9*kb$p
zTcpG^@AHqV{s{7Xu&*CrO054Zi5R>qbWlGfR%K}+w~#U(cv<Jjm{?olsKsm(niva=
z@)5QN4)Yc87McH{TOJNDOsB+L%D@9qcZIpJCdpi%=afQf{Lz`Fg)w`98DtsBpndzJ
zQ}?Z(_wU-#QxER4b;tKvbT)s#Q{ixF4&qYCZ<k}Db2jIG*MWj;)XHU(ow;7F-KkAE
z%?;U@_%oqSmt7p&5K|j~Apf1NCTM>!$yBl>-vqkp&zq-v5roz?0MKWwV<Zj?Aj+Y{
z<R+x&+%i!BZAu);WR(OIz=X9EXlEt~WRW?_H;Qu}sByxe*ujJlx@S`zGPnb}0=oyB
z$W3oD@fT)rL)usVB~o)~5$2CuSB#TIfp*^lEZ&{0b|9e~9mE1MuQ#*Si{f2Ep?K~9
z;<+<nE3=XNtrf@M?{a(ISKJ)%KR_r+NRXer%Ra&0AcRLFAe<+U@2Pt+xdx%36Tqka
z3pE?Jmg1@Yq?l5mxwOkL8n4bi?UQK!pqX@5m!ZYyFEgKRi8pvU02bk>py<zN#kbni
z1Y1`gD>TR>MadFDf*%lW<k|15r04E}sCzrHtpVn3ONt-5GicF(Ml#+AqUpni!XaN!
zQ$CI|QW3E|A+^RDBHKm4Cw1Q4e#JOi0cPx|2@%d$-_kfMM#cxe52D5oXq9QltiJS_
zkSF!=DMY|>h{=JQ1hebzGpE0E4Xy%0sqXj-8VI%YTrf0JhO%4|0XZXH(o8Wog@t=F
z;p<8F)M~m+f`WelgJE21AE$xHK}FEsM?}htKz{g@J*U7hlrSh<6jMwuQU%G*3V}Hz
z<^nV`apqVl_fX0(G&)R|wno6qDcp>!4b=}qMU77Ek2Qmy5+v;a7;VpW=@W?<ZTL%*
z>;I3jw+xH&@7DeYq`M@fhVDkXyHiTKK|(2!?yea?Qo04{2I&||8l;g90U3}+pNspx
ze|!J;K92u$Jg<fqD8q1F-)pUNt@Cq^8i;*0RH2<@cA%+Q13-VdI5XZrq1&FlZbhM1
z4emInBv#Y<$vYidg{omF7uqSG+lt%FB89R;1|!7#4xZR_Zz44u2wpv}Pybr=!Ku94
zj{y`q|AemVx@xKsM#sZqIesc_)LzB~!Kn1FpDI4zKRt14I)hpzVqJe9&qS38+Qc`5
zpRI0foB5s07j8x{hcU1c5yE!<+@GE85LKBpVLlS<M@PXUKcQGAK#tJiGYK$sDAfN+
zoJyo|lMXF1(^jNOmfmXv;I+@<Tn!J887{Mvo*p<FcOg^1V>JEyz8L5T#)h;nYCttI
zU4g=f6Boh@MOBhE9I@$5y0QdnINa85yI$G|if6wH&0HF%6*T4wt80O)i3UTQSU4KD
zZy%_41K(~OIiKld#SRC;%=BzJZq``4E7Mq17;Me}2=t|UaiT6U`&{8cpAMZDCIj^j
zUsVXP>_-8<LA{c|5}^ZBnn{|J))tLNTz6&B1ZNquHK0kGWiw5Bgvl+!+QlG*Aex`E
zL3LUhriI<bV&#GWM4u;3WDG4BAZ&!SWH#y@lNwR7H8oCN=Hbn02hW)6?pbI%KLf8@
zqNhsWVV3!)oFfMa$K^3^^&-2Ay5PBmcmdllv(hVBRw9?i|CqyVWP;W4o+`0pv?;>c
z^8_A~7Laz5FzE3S98xE;9WeAU-f?CRVav*_*Dzezmr3<u{P=tDuP6TC{_Ob2B}gA5
z%FHlDOo9T&HY$Pumdqz>tZxdf=P!Uh=<O|v@Ffe3FX)4x#kiUG&G%u6JrO2P-h<>R
zFXr%f59h=8{y6Lzi}@a5RSVzvVz`f~5iWZUbeG~|%r1Z{`1xUb5GmVN?Iy13%X6M9
z_*05qdVjHjoM5Ow4qf@0i~RF|Z$Nvo^j#~o0CcH%03Fh`&=t<L(>V~Nj(0S&x6s9~
zEWV>2nkPL{%B`51FD%miU@0D{&QK-)3j2wn=kGt9fkRmtikC5k_8rOezn3*g?a?x_
z)tYVOJrc+#?o9b*6m)T%_x1VR<@jk0_$wu+l|t07FQ}XI&)k+=OC75J6dLWt+AY<b
zN3$%^PG;zp481gFkG7-a$g+|I8oD3T(P9aHpS}+;yf_3eBg*l+Yw=eisbBj&TB>Mi
zX5)iY&1p<aDWXeF68B`9cgc*AVgy>RJ(uMKmXWG=Gi!jt)2$$aI2MqvRr%}9#{zUC
zh0_$G;{<*<v`r$@8GR5*Xw~)&YUF$Y<#};lyFaOG27AtHoko_u&R0obw$}_MOfZ}U
zHf+jYQ#E6<5p+YnbyBApSN@38c?eXy)dSUloQLi*?Y%lKKv>8BCT2e9t8?Yt;q*9U
znTGq2-R$&oxmf%~TM5e>W6~If*RcxCj_vD*irtR+R8)X&f-hW)+z7jem2=d~)2%Dm
zbfjApH#i`DO;fDURG`1|Pn&le#rLU?$d43*mICE<e;}nWA4AHpYPOZ!ngjhx+S%OF
z`Yx0_L?mmAt9$OM^RshHLhB^C-q>ZL=xMaVQJfvw%NIVJFC42_kV|ip6j0(6KylPJ
zJ*cz+$~<$^cEsC14Vry6(A=ld;-JkHqz5U0J(b_7WYCy0=$ej@lJ4k--Dm35XFpQ%
z9&GsQoz+}&p!G!@0tyZ82iNN7M~@YMZ_Xr-aBUsqBA{2Z>>hh9U;PuN$ka1WPlE9?
zvjM!GQUp~E3widk*{L>H#frkr@2<2vXCx_CFYK_?!+aO@#eswWY8MHK5a@o%uk?_(
z=t8ge6;OV%lqu{bHS<)I*^g0KsacZO7P|0X`dmKYrKx4JY_A-$+x&4}fBL&~&RuGY
zGij15%nU5w1o3YfHYWJ}=`uQ)G@rf|k_r|^+LmwlE{UVFA`ugxG%_4d#@Fep-EB9K
zHUB39!yBleU;K?U>c75y7g5eQSJ`Y-A0mFCGg=fYVJ{P+t7`{wq<b^`G*&x5@ya!r
z-G81yqt(Qfs{}4=kvo&}8{6Rief-ATm?4TM+vV=<=38<h--Yz%Apu_l=P+kFT$pzE
zEqA?7;{;2BkizG`mL|?-;tw6ow*t$1H`CADumZ?(8jX(PY}?1BlYM5CndDdA5ltA5
zf@_W`X}GKLETVG%dz#Mu4KEnaTHoKqXqNL(jmmjHg%E8#)U;#x(`pniHZRRUl3D7T
zX5$SeZ9B56h<A`6G=kL3*VKm4r5KRI0#2b3y@yqp(W^1Z6;B+RzSXKbL{yt0F{wA<
z^oTS=?<00wvcC1gpD>&J=|k#^vq>?)S{Xiy`)ecaxI|8G_HV!r@vbyk#%`rU8YlU*
ztN{O*&vX@=0Db`YqqP4a1D1>&%h1YO3gCxvimSww4d5OUSvl4eZ1Cf8?U)mD0{v)%
z!@^Zz8+MXi3|AK0qyw>7k<sZvCYqA}9NAWJPQIB`U*Y2o0Dl`lDTSbEK=$FmIx*g0
z^!<YP&4EXem5v9$ugPDJASVwUdXtr@6-M)(3?5|?2s7y?;qi7MB@sW34YG{}Ebx}T
zZG$J6Bf(fqg@W{!XswUoX10ZSLM!6gM~bu{87ro*&11{Q^gHs=0Wsyq`jQT8EKCiB
zE2n&cImrC*p)Cz62tzc?zcPL&JNG49utss&`c*skt0uL3M<*Q-FCxVM)FPSq+?C(z
z3=qND;Zx7CUR(hx8MSf7$E_hlh9pcYtFoTfumoJupS59cC2Y(9?C<+Qr`v#>Bz2^z
zM{culG?83c;N2jRR{Toz?zFb}x};R%*4U-%WybURkcw+V{VvmN-)Eb_e48dm?|$nX
zGHf>c{S4*GZ2RQ$)AHKYL42`mRRHVfjf$+K>4TuDkT}qd%u=?z9JRrMJlfEA<m{RW
zCnK*PI#@Cm@pNF49sxGJf7*TSbv*8w*Zs)$OeCbGJ#N<_8BMPC%gfQCz0m}xX5W>h
zT}(svNS9n6EIRU6zp7@&!5)knn4frU=Z?fHaAVU+w~w)vy}TqntvEv`Y>bBqt8+Y@
zt2bGLGTt6H&?NujbN15tgqs7`<kk(AO`LuuO@tPtK`F7;{6=);#YUnH%Dpo)Q&1T{
z>B#iNME1v8&x%8b?bqI8Hu}_SCC-5jiSSza+*&~6My0E0t5n?@j^MAZaqEyJ+kB5o
z_|VESB#yAR(t%8uf7_61E_nU*Wa~%}hwd6_W$?_kZhXenvC)pDf6akOu)hS?mgHVt
z_UcUrz*yl9g?3yHh?D+?@^1;m-@T<@`UT+og@Q`l64}s>@J@%^X3-NHMO$NrzLHe8
zy`#xm<S^6V)Nlj7AU9#9aVkbIC}R0eTiii9_^3t&)HCo8ZX8x|iBFjiV_Yqt7HL*E
zpbPH`Hh4<_SHY?~!XkPq!s+}nvWeCbq8YbSg8I$eyd813XtXnV+sc|0uy`8K8|}=Q
zy@P)3mdbI$Q!V)iPy>l27q7f)c&7zIC0vW|Ldds)jgI?lpuhCsjtdHD4_@Iv`Us7F
z%r(++bVE>E+YVT5*%oNk?I~#oHZ`bba|wVlO~6|`zQcVLwHXtJ!`bQP@~yRS9Qob(
zn)ImWRCDQ#%i}*eGu&kO+^WmH#0}3L?`oe}`>hr&MdpS`M68&vd?FgxapPv&d^T8+
zJ_O(l=Ys%6?dB7X!Tar(`9V^+AuCIPUWQ?Nw}4n-FU2L5qvfDr0r8Rdz`C19M_AP;
z!aViLU1Z^0BSZ|NovQicJ9V~mR<SlQ?I1y|yk>Zje*CESlDcU6Z~ilTmx0b200tx!
z?^0FWk|4|caYy+no+YJr>Z9$yrEDN)j+-p_AlSXhxOPp$5|{$w?tN~!oR$Om-yN3a
z2x;GG{U~gkz}jP_m6|(L5aGtO?p?)1U9lw`Pd270{U#}j51LrB{BY*CK|U^{W563~
zg!irxO^(TfW$uP~RF9bJ2{qO;5H!WD&qV*oss*;4$a=|^;rP2*E2D{psSLon=Q;S6
zCc?ooXdHVeyctfKOqFWc0vznEo1q#OVa0Z?1)tDb0bv-~!~40wQ<kb{>SMWe-}=AK
zMaGA)%t|g*eeN-|#9bvkP#k(b(APSW^emgiM7>-vEV?n|9<$Wo3{^swo?2!4AS#_h
zy9Rq!oNq}8I(~Bo$GZBA_F1+0;w|oWr!inKlQ#|hphCXRQB~WA@u$=1y`$J`gY2c~
zU)@ZP3tZ98dujc)xHI1Y_sD*IOuWpSh)o4bem3{*Sdh|62e5c2s!f*6f%H|b89?*M
z9Pnc{_L}Sx(YV~IOCtL$uhNNzq@x0G=+IgKs(x!2>Ly<XQl{$js=<_fOmqC$T&mmd
zrTI@)EGHezlVsF*U5CB>D}wknbH*TrIlWDaE)afe4HCHA=@r93+IvOehdw?{8ixNp
zpBqS&<#!=ccB@s!rIh2Tg+^Zzx^_Rlr%sY-w4yVmicS{}?3iGU|DDktjOtde-MdOA
zO6;s*(>25SFRj=)QpIvKl_f+HF*^lDm#x^7;^V;MawwkYoT##ENt_+vKe)#X*EC8k
z$-`z>j_S8#of)$uo*5*KlcLxOI-0Ba(=jW#6U;;9Nb~4C7g{R-3~Eptz}wv{_fM*z
z3xC*7k*Q6*^~B5vA)uY^N0q$keV>$YdmKgpnh*)R0Rk(Ie0goGfLw<wI4I~_W_In!
zXZ9UAA^{AB%>2MVL08|NIj0q>{6%JtE-XC8AXTCTYa1Xiti_?R5$NHiQXr!zOUo$}
z>=MkGCKG%F8_8?+#O0f4LrvwdTK(|j^Q(;v2JgDNFUstkeB13y%kIa@a`*2`t-a=r
z^~yeLeD%^`)m--L?xD!J<Cw(I&tvv8+HSn+MNt{L2O4>dqGnvUdX_H4>0RaD@!|6H
z2mFj6x|Z?{M_C=;vW>TEKJu|nb5{s$nLcD62pq<q=wTx&r!+xR*>2+xuNUS-FDfrj
z9!N}fY5LGgieuU;*}JpjS?DtA-S|eHZAmj$0mBcU)c0zXMO`IQrUh#xf?svcAb(+M
zHXoAHVG?awSp9gQ3PqL;nCUmD<+xsj!A_EUNv;9%PF`n1Mknpb_*yT*SB7l+F&n$}
zvtaUz4KZ#Kg~*cNReH@q?l)_kL;bhqzhk|#F9$E5ANZqfN|6^nRIoWmtL<XocbE(R
zl6Yiv5pXG$tI=F3X8+C|OM$?L=!<@&mm}cBweVnI5R_Ax=g_8_NsC(^RRx{G<pS1t
z81KmxVIJBlZzkUo8u2Z-dm9Eg^61WK+hBiq7ym6o0?S8M)cygLGwiNVgts9h^1*-3
zzgS4&FevM45H=?VD%RlnmZDfHmAU_@xMV0=BKo*+5(K|qCgn<6mLN5sU4*>#Ok>LN
z)TvnanEDj*Fy&1{cL7QIdvW}jsCd}FOm%&&nz!MB6pJQ=P#YbcCyMx=RBK%j0^>6*
zvpTi;q~IubR0f)X5L!z%TMr_{e3L&&Z86;P&j?onfAHnq#zPym@<&{px)uFSd5o&+
zLClM_c0WL>;mmZA=>9=hKbLY8pGf&D!bqh+UQ?sT*V7O}kgP+ISdZ!6?|vY9%T>jR
z9pB0;wmou?m#BalizaWBvXrG5g{3#fBNbirY`4IbA|(VAot%!$OvI^&5Kr^RQaMFz
z51|CRSoNr7)ja#Rd}=5&vGxrw)ZQ(`dpHa9J*{)>La>Uq{7ZXIkl)-kqit+`G+Y$O
zQ|Si+3WT}QVS2ey16eZ0ynKV{fyd2UcO0-)_5rCtYiT5TQm-6j9Jnvwk|rJTVl@KE
z;dcWW3fZ4lETbsn!JoKU=|J@klR_ZXj{ZnMe=MgXtQCkxUbDQjAHsvA4Tplhzgob`
z_73_6B->LUDE8oId`6S{Y$cHjv&It#N7Vj$2@F$efy8Oj%&JjDum>URA^rzCyGVo7
zh(eO|#cep5H3EzjDf+(Cycrz<bH<i<%7=t)xHzQlMO;t3Z7dQgdpKVUbF$$L=N*Co
zXB|t#1czbyKCajYDt3T6FWLZ~yci7j8UPls2RR#*CJSkad(VKhmxfiGu7TlDN*iK!
z=5!yio<3n~o7|{IMB)eDG;0$T;b3WKc<<^VFj(Rtz=`W{w!5DzlXza+g<S`|CL1ek
z`R>xP7H{}!Tvp`onE4-qU*l4*r|zZ73rtuvxpO(bqTYhnpTx(*rs@sti|Oae@B2PM
z)qR~2*WOfqsTb8fen$i_qA3IxR{^thNWG;QVsA{lbWVQYjM*-{fd~sNI5p2?HWL*|
z$zwovtyBZZWq4cbBbeRz8Sz4Y=-M#0Y_(!+p!a?}U@0Sd&}~oM)^hGDL=0?>S9C)q
zLRsan(+!*|P-5<aueyS@!MOq=A4?dz^}VZM9mUpjSJHlf;fhp;v5`ptTXvx5PGlHH
zeOi!YyN*&VZ=|v`OZ6lp;oN4z)x4gGV;SaeOgy`pj_e_n;X^6}B<;$m;6X8nFl~K&
zqgHoDjdZngh-bZ9Vta3AhPzZ=f5X8HBnYqf4!Bwp!QO{lQzP0nBW!NkHPDyNCUgl0
zd83H*slcY`5L(|HDq=YAFWEghQAqY+d*&VG5qEJ5FgogByA$@<3vct-Uw$djWZl*f
z?=SXd?es1~BWn)8CC`wgc#q<<|00$$iN{8pL%0JU5tIL#-qypfDo+PSMmh89qkQwq
zXgPRI2Yv@dOYCYCjYfA6D{~iW*Ygc#q@I|xjyYk03dzeUSpkUxbugC$-{n4}U}+~Y
zJ-dENZXv#Dl~Uh4Yai1nv@;|_HD@R3Jr3~!aovp?b%aJe^L4cStIW)@f)8RU67c^V
zK`|g$D<xZSqZhzJ6SrZ61F*FA&d*-WrH0KkeA$h-3zWLzRwnY@`U<CZ6e0es*?h(B
zVrFJlGY>INbE?Y>kAt<Aa}re(ZhvbXfA3R-ga;FX2nqfiqY3@{98#k}%D}26N7>bt
z9P?6ki-BVo${CSHfMDS`5r`_e()y3J{J+nWf>z%L7dU(-yaT{P(VUfQIm{A~ZqJOb
zOy3vtgjn!u?b5dW{JPat2L=@y%J2-?c+F`Ugv<;eT9EHbOJt_jSwAmf#gI7?>S)qB
zDuBuMM$pFYC)dBhlPFUA9_O}@5QZ+&LCyr+s)VnJ5seyIjWbklW%2vBXfRlaczl8#
zVz%yq0W6E2!v*`OBIB~PI>=ddefO4JAAe4N?OV6_sM{=W39#AUXT^wNPIRpoVwyI$
z0fd*26Bo>QxQvVwncsbx?J|t+jYYwbQwl6@V3#5&J<C)<l7fd2WUqZ=N1Gll(NX|~
zj}Kd6VICR&E+>Z+{QPBKhk|UZXCWPf-2X(YR6`3^w^i3`e{84Q{PzdQO^Mg@M$p4d
zV4zkQ1JjCqv_Wcek9kj?-AcNrS4JZ4{UB(Gs?r%x1r72y^DR%;&lZQY;r@H}A^fK#
z8@d+hd^VIlD#K7V6!sl|pvrCp?z&054JASAb}^|eYaCrQk~Ar>6t`=z5NwNSn;M46
zUMS0K1BD>l3B;xvorT5s8^-k%ip!yHSq6@TJ_06Ht262RQ0C~8{%1HsU`;a%S*&cw
z9)c~*ZJj+`=R`#6RoVv2&^l)F5+!qtV#1$2ArU4sxQC%<yr<Tk8h778!x@zE(H*4B
zf~P#1*8WM=sJyrt(mGbNco%k~Irro0aRtf}v2G<EN^*vT)~7fEx-aR)U+h#^Dshc#
ze$T)tg}enjTbf3;@!CrHXyr}i6v(n+9{hG7Nu>j}ds9Ua3c{~c?!z{L)X8-H*KM5#
zg>h;@eSUL6#6vVicwAl}9mL<czDC3fF4PbS@QS56Rz2d$3pn+T5j^ChSRb$n<a!d;
zICTXUBvP)oJ^ajPu=7}Y5+)F?mY=|6q}(R`3yE$NpA4xESuImkci(Z0Ma$JWQUOP*
zxhR1P=*{%s!(2U0@G@y$egaaOOy#VPFv3v;Sm~UY6>Y4h5y?NytVc(kH9(1F%xQl}
zcS&i6eFA??MYa(=LV{=T$}oj$5XDyt_)Z(-8_O)va8;N$VphqDvmV2DQW}4mWCmt^
z&=mTCG%HAxH*H_mu6!@$8b<RQCDgNPMEtiXOmvyZXX4rKggeEaSp>s0mq6+!*Zqw*
zuWO<!mCdnOQbnYxH3Mil?_M9#wZFgIPemT!zNE*~jSKNr7CSg(37fl?qAJEZ5X`vO
z?GlU1rO=^y+~lsWz4<D#@AZ2zMf#MNv%qEQ3VPp&y(}ssx5EiX(AfoI_{=}Ayrp#-
zdVPvh$)#=^uW<xO)9)-&gnwg*6MDrrVeeq`rk=C=C@fCi$x>y(#>H~C+}8YwHfmP>
z{tBG!A|)I74wVSwp>coSh9mvrq34BBkI*^595{nLq+SWsI?LJ#I8tAR?0j@1E&c$5
z@3fGT%I1@Np%uFj6VXMxVByB;ll@L+{|U<Z`_W}4s_EtFS+-R7A=)b%u;pg&KTq*@
zNTsBR_2WeIEs8Af3Sj}-7@w*s973PJB#YLWk~<NK(^2qVo7CjbtZOZ%80ESaUy$P)
z)svk|%q)+H9<^@*<2N;kzq$6?<$H8>$ai*3Np9HZVHqxZAN~3vOYUlQlIvI1`(rs4
zH29ddXkj5Eo_0rh5#zq;bu5!kwH(V<QEJwpG1M<r(MZIPifSvA@q>2p5OaxIDqL7)
znt#CfYm$-=#Y_C7ui9?%3j~Gllk`>}|E8D3$LUo47MoS@!DIQYm9or?tUDK~G^b&4
zPwW*U!c=0BYQ^xTu7U|$kA&Z-5l;48aqA0t6q?wTIsmgwe%^G<6C1AQ)BT|c7|eAx
zoT(m?G-c?;fA**zj4odb)0V6G*5vsPn8z!q7t+Ct*{5=CVg|bC2Hhv=bbhXgv!r#L
zI=b~#tFkP3nPPfrkJeQ?eQIPmgTr<?dzTFZzx-_d(r!xZo#W*4k%d%M*!<RU)L7+$
z*??=Idk43ff$u^{xLUWR$%3FHKhYgQJHFH-QLHRzD%5d*nH#l#A;xm+#iiOCYFc4j
z-?>w#^YXF)Se9IBy<{wfP;H(Cg%suJGZ(`&qC%xbZIz`5{lz-sg2L401=o^?qjm!?
zoierT(X7w>H(o%M_OD{i;xipsM}Dr)+{CfuW1*6@c6ebm`!s+CpXu`<5~c^j?z&8`
zD>aTTXKmeJySDbZ`tO6LUJh}O=QTwOA54D>xFmiJOQ4liaIrct(L|*!`c9E`PVJ!#
zC7t;)W4>i|$`)#}>fA7~g)O)`HBG`DG!tq`lv#6Zy7vfd4DwPUPLk{!(PGKtzuJ0=
z4TPDP>=aaA&!S(r!jp1~%f~F(IjvF?yydBsYCIofFA3y7#RFRdnl=&!Sj$SJO(2(T
zL5@C>x7mLR6a0zj?9>#y<Uq{pSvvd6ayGY9eHU7XX%Xq=gLmb~N97ZW;%5$g*@t^_
z3x`0W4YxN94)g=v<UWFUB$~y-ki6%_TQ5pJO@g{A&p(`n`a~7rnfFoXMUA-7BrHK2
zr~rY2C+(nAwp0>qRfZfdu0?h7c(k(ggicd#CO!5$S-O#3CiB>Fsj=FmmcE#7>O>;~
zK3Lt1dqY-WM|vPK8b?07lNs(nL?hJg0*@Bv@)#n@n4}?$^^%f{WOmn|^oRJ4ujbXm
zwrM2xe#0#Bo7JmlTr2l{4mQiPm$8Va6Lv!U<5ul`Iq)qmjNgxYL+oL;E}e;U{bG33
z?!=i&ua$L;@&ls94BuCJd{Hr2^4UBL(>@ZYX*R0S$vRp%DYlmrWVNy<>SdI-?LOF)
z|I^eIT1~E*7rp9w5#%D!2ru@BbH82hWIq0+zdjeU_4u&VoOFCt!xXn_OQAdG5{hk&
zhaaQPs6T&9(XVYgVK}lT(#~@ye#hzgyDFO-W#rij#iw%PIksgT9{!J+gEuoiIsV1o
z@QrqtRPXGo!GoSdMMpx~i>s<b{BBrU^Ltv}fTW*q$;Y&984dYesFw|`y%gwQ7dkP_
zRS>%Twp1T}{9?08m+6L98wU&f=NZ3=rp5ylNe=tn>POm~d^A${#guvqKc(>VO@m7_
z(y}9SSL+Uv3WFt2{Ofuf*+spoOZ?=mEMub)xMR6axLc4>Gds?7u=3HE#33zO5~(FL
zgY#!C)CiDkE$)Kw6b7U*Z|FgEGIl*Jp;qd)EyjnFjD+O^`|C3xeT;LEBIQ+Tm)b(p
zz2qDFDtxv$$Gr{+x0X%)F9_S9(z9ae)a>OGbN{AGlZEH%KPCrA<ul(~rb!4MY}H~A
z%_wCS&CTK1h;p*}mSc*A>8N)|`4N38E=hocsZt>p>VKN}AgaV#@%2*(j@9w+hGbcC
zDsV2w1~OV>y*gtrm-%tFN;u%^C$CO3<5M(pp?=h3n2r>TbJ}8BxVI82uTJn1FEDqg
zQni^A0icSX?u-0!4wREz0RqzBTn3x@J0%)pzU3pNT}7r*`jR>n>?U;?gW@L?Ki?R{
zfRreF#KS;JqO7dl*`uN}1BJ?FOhPE+sk?*6^1EVZEqX0L1v{mXS?F(xuSE>1GKn>m
zO<mAXrCS;04WFO2^35I{FFlFlmnDcqik}gjr`h3*<-VgLW@#jxBbH&_*b2q+dGnen
zwslT;as`z)kzXG__0YVP3o}6cB<#b%is3J)JjL-D{7)4Bbkua0<0Kw(90D&w062hx
zHw3PXARHl+If26L;W)W0yppm|Y74+{iyQB1{<$o;De)~HCLdJw&sZCVg`#-GA1^%_
zK!x2F?4hT!`iZP{Qn<Y)S;F#AIX(XvWOV@<p5btQ9yJS(+$tKc3GK1<6y#DEA{}Of
zZAgSvm-%ng=o~y((v9cO0Du~%JHGXg&pjDh#P3yNvRrnHZd?TLwQ*@O)poVgL1+~*
zBB%&>YldXEM5cHb@Xw%LbV?;M$3*W8(XbPSbQ7KD@{3-3O5@oXs>PSw*AYVeYNRCr
z^t2f5u&tq@hwPn*gg&?o#JpOpx(L;=C}0O2$nsJ8evtop&-qrALiT&=@5-)ru~D?O
z7rYU|&QuO7u)yUv7-G+w=lJ$QSfp%n8LZr?zLzI(68G7Y3A2cg*{ORb&V6A;1p7!+
zN&7~&8e$zRfyoMMm{F8|);o~dGJJ{R&XH98syJ*a4|mE285&%*nAYg5AJ4h0D<etm
zomQBZwa#$eA)qzehKP(aZUlz&Uw*hWpNlDWVu`Cfaq2&+ne}EWS}n<)x`GfYOCl#8
zXrM;?=S-ky@$AG)Gm1MR;kJ;L^qv&Iz7$({Wp^caa~UCW%N$qaO3pFFqhkI^px4^J
zpnOC1eD=?-@4DwzJyI^`%G(vgL|&=`m8n^#qQ;~cMo{y7)JwAKbj!m<iV#b@cO*SW
zA>#jOmxK^?u@S~YR4wf|3Eu4oi-K3fl~}A@t$Cmv$Zc&wA68b@7gEGmndmPTTN|f#
zBsa~pFiNZKd1_3rBycSe$-Rj^cHbZx5VWRF6@!(8soxRCbbw5_w(1cLa9WAV$?w`g
zjGWl-5jo@CFbkAFwlUAy%&48lwOVYD{1{w~-!wzLSZ~!E4<oYQ^F_MYvtS+6f#}Sw
zS|<m>Y2~xq4wiuZnu#1acgy5qI)=E+-z4Z$YHMK;jE{EJpZGVgki9{hyv#~I_Kq5p
z%F2X|*%CQ8kD^tW$fM>vimH+s+eb>VG9El1B_b=aq4}Xz3z^4ctI}|j)2xaF_@*Ci
zdKM<zs5C}(K6(ekGw&Y2a@O>*Bbs%ORj7pPF&P)067_1hqNcS*DdetNzMjeeOMR=y
z<pQuaIGysnBM&3zWU-lus7ABDa#Z)7*tP2;B4L;lUDoafM?l)v^UZc!)4zsg`W%V5
z(Na6)!LUtGqBrea>u<VJ{T3wLPPF+mYfQrsTFhEqrmV8tsPl7vd_(jmeE8uSE}(QL
ztZ^$fwCPKM<v`e3gUqe^xe6t#1C@p0r;pKU?eQ?Bx5<;Avr2sQMdXj6gi?6o!`2qj
zDxBN?8-)k8v8S#s<Ms0AmD}rMsut~1e?md&R5IS=I~IbbkK%C7wz9q^TvvP`qG9S4
zQ0khqaKL0uxPL65EYiZzR%IQgX&e^YccVNwdNSGQP3m{xpA$r&GrXg3*Ck-j0+NVn
z*uOlKN*VVDB&7!KouaixRaSKdY`%_cmNiyzy(j|?A`i!V3mHju^T-vcQ3>43B^Y>B
zrUnpP_o|VX*dnZ^`ci&lwo+NDNoV)!aX)4gFRBaR8O-v!&^p08Z|eT*P+G&Y_{8qy
zEp5}BTj!f=VDg7&($-aX4^b#6ZCm6bcB|tPpT3O-yNuG(Nyy?m(;?DNKJ``jxc*dY
z$fuL<OkbDeDM7h&Z9DlLa3($|7E^h&#AUBqB3``MD{lS8aikG@#<z%%Fg9c%znIh7
zWsNmVTF%aWZL!W8XC!{g?vv!XeU$8fg&7~0`HY8TOxE}ow!K3NWC^=^V#Je_y}>&i
z-Gd#>N4oiqtRkjojcX>y>F%x3A}@<Y-12*2<rsb(GhTkS{nq4N8Zy*a9p$oeQ7o(q
zulJiL2q>>MY;rYyHY1bWL@&CeEmN<gS9;W`tJ^dAE-@Q%eWq%!wC7}rHY(|AmK~^X
zuQNq|utG%YDospCR^0zMi`bn;npo$?o)+bCrV2~Au07>6dAHHg<LpJ#Dx=`*aJ;$v
zTpY$f)oeSsW|CBN-78<&8C^`D(V@P>|A_D?qqiwKjt?C<_)t+fD3xRRM_pVIm$@+Q
z(8Q5^j_pX`ch;R5R%;~j$nxs^HbJ|&6t2>mC60-+cO!0B5>6#6m@kurujPb9p<=~l
zu`;zw#4Fp|W@OsakTxJw13Z}WcMF3oeQZ+?>maC6SSXr%Za25<hQb>6^7Fx8LwnLB
z?y5f`CARf2QxhJ(Q>l&6@si1pg4iCI@n(^Kn)6mQZpq{qew`YQm@W%&4o119#8#6m
z?$SDWwaIAusCHkv!MM3ccIC-=`2vb=jG3qhy@hL56nx4Y&Es)QvKzt#bUhbjqgzNJ
zsFyDf4i9wEtwVO1dHG739Pg#BV#i&#)1|lnUQS4NdmYIc2N<Tf{mp;42~f<(Kwtvt
z(`+wNKFP2^fghqAFGV{5b8ct3+1cx7L}h+$J2FhfOYU47!m*dmlC4G0_k)@UTv!0B
z!dRYn8hzX5%w3qx-jJCbf{to~faU=NCC^Ije$aaeLksSVl0gimihW5$#*Cm94Mhp0
zF@+MRu0MQa;nE7`Em!RA6`bW;=5RTa4b2qyYi2Pdh^fr5W3*Fi-|1Rp^kc3fxMB{{
zUN10|r4eJ6%Cgd7chc8S8(=f|7G`$bCRv3jPt1UbQ7oqQ-ko~24M_qoO2?x9n;)L$
z#gu|i{+#SND%o!1tj4k2Bq9a|no%}}`^emd3zgg<D3Co13u6wK@q(HAVu~IIO%Tso
z%egq1Hu;Qxi+O+zrX-e)c!ET$)vcv?9^5D0QH#xGM8_|~vWV7*uO#U%XdUM*rbx+w
z!d*OW^u5TXU=EpYS=cA!=P;5<N1Nz}GJh>sJdmyhG}c#b)0fqNsExT29mYpIi^kkT
zBaELlDhkI~Z#=O?RC(Vfck0ua(7s0Kn1Jb$%95vgo3CE<s!#%a0ltz>4TH4-%K8ee
zlO7ys>lshjN>CXHexti{ouK=Kn@F8@9qHB7izs(1QK8HG%<JbU9ah}O5qu2f!)@eR
zHr6;TS)&9Nvlh%Q>@Ti*y_5nW7)>e5sozqW9&v3aGJ+#0qG8MrDHNmIvf2~(mm!y;
zjUL2J+9tDqrxj>k0gd0+NUY!kET7b=Mzw6%3)SGi9{(2XfmWXdF}{E%s<vl^t9El-
zDB1_+Ne|+swxsA+$l(oDNyOf!R?>pJx3LuKp7$CLWem^d?>cNZa_f^}!(6R5G(R8s
zYic)55D;2Lw3Ko5_Y`RbYfvbu&RMUVwt?F~p~elEZ@q}AKY9#YTp3k7Cyz%b+b_%;
z5o{vL%l^54U<EeDRt3>ZLdvF+Y7toSymb-}Am1Uwt7P^1Mx=7@tg45;cg8A(R?96Q
zB)|quVs6FTUksg$2as;a5?k<X)gtw@pU~lJ-+Zhbu6KT4J#y4}m<{$TNIxJsI*<kq
zAfML+EK$Py;6f|kB;Wo|n6&zm0dIW&ysYs_5a*g?IZS?(Ne-c(=uCruY{BuvS);4Z
z#rDHPs4#YRuxoI}fM_p`$E=eCsC0v%q;QMp2oUcjSp0Z_>+}ZqqD`05qZyDyoKTMD
z$bKj_dkvF332oZ}&+u<R3<y@fM{3aGhApE`>W_?zQ6(xuAD>Q~(x(%L0O6NM-u)*i
zkFMA)SN0nknZkV*6wfLkfDU?nC|PlTS*9#$7GHUCVflOU)|=_|Y5grCB_q9&4F@ma
zMRMKL%n3I!W@x$SjHh+_5LC6}N|{V_thqz7cEMc0k!f^w0@Pc6-LB*=MO+GY4W)y3
zI^9y@bV2oq>ty#y$Z16|wV5@A`wRy}3*!J%fcrT7_;?fyjK}V1Hj#P)4{pT&OD0IB
zg%Dc7>3MGy>jas(mA`#V)|)J9wBnuS;%1TVO)LSw?<|;;W|=(IVEm<y7|W4LXviq@
zV}lDq&HhYEMXoJ8S?|<JT{`l2q4X%@mO^G|Xs&<0MTAgq$In&UK0apdY3tB_PP0@i
z!;&p=qAWa6l=8C;FcpkuC*m56P5j(3Y^Lwg@;S2_9t~s%;b-!$c(ts9HIokpM2|W9
z^AvDsRLiIpV#Q&&&w!>;2lb3LsnQN3X+VYfy~%Ps{4rr?I=CK)V3*{x9b0BTxG9oU
zG(V)~E}}ga^@wLT+C<7xjqD-{5!abT^4<8X-c`UdGA3Cg8%)`7DV%6P<1KvbYq<9)
zsu$_*I}COuMjyz=*fH68iXEaGle?p?kD~v8>O()$5yj1`2u80C)^C<IEqbN?xWe*&
zg`6^=qOZEmh6JSg`U$kan4pG&$%frVhue}S?oX^pWM}CqMDh=sE&TI@2(-*8W4n>`
zljK+^6}*e|0CmKi3VsExo+X(jX3@OES$>%@HPmCPWuCjttHQEgHjzyA2drhamg<LO
zgN7-t{Jv4Iv<Q}!8uZuezwT^0L`RJYtDx?lhnRCVVt6SXkiJcpom5NDu;}oZf3RwO
z(eLcCRf$o=?jX}>SshjK+45QT5g1{oU_=J2e>*}_j3{uDK(Gwlwkyu>WCaHlW-2HY
zXh6y4L$mSenI5wB)|`E<FVr$xV5a84dZUDv3O%{pyBBhxa52Om&v8L`K%RvxV(}?Y
zx<v|Z+F;!ai3KDn_62FJW0D?(M--7~m#&X8l;TR{fRr-ETuf1GhZA{?C02-q0pri*
z8ue+*ZJe+;cwI+48o32b5yu52t;zH`2NF!8=G8ClZ-*8DXYMpw86+MW@rX!#0Z#Z$
z2ZBp@7@|!b$WDFH>dveT6~1t(?K`rdszp-Hit)o^$xt7GkdDoWl?VKqsG|opo7NC0
zcpV9xahuywNw6Z|6%vOeggqjL(+cPos(u;=1{2)c?j%bQpThDlpUgLx4vpK5!XK|>
zR%}fKsE~S^cU>vBD6-3GvSh<47#bP)#=1<IsBO;{U5Z>T0p-Kd&oJ#L$y9&cy4+(2
zAhQ3QuIq76J$E-gu`WapeiA!|{<-yN^KoYWr`k~rj?IE-*1cNPtR||_-l<?_P<%{(
zOkhT~NKNKef!YNvC+wF4sz{@Chi+i~U4&6T^VTfUn0&*|HzJ=!EQX+!|CV%t`K(1d
z4H#2PoBVBIv|d01V@rX6CZuw>YqBJpHjO{apc8P|CcS>AHa-96<Gkg4SCujL?`i>8
zIjHCs;5-zcyXY4ymjA)5I@T|??4#hyT@;<BjRdHo+Bm>0wu4zmW5b|{hG{g^aFUNU
zH8XLiI7Q>AP4H(#Go0t=b?LGPt&ZX2oWe*!a<%+jd4gWa<eL<Y2f(aE$l`&&Rwxjk
zCczZyG&eLk<7Yw?GdV*~s|@@{s0uD73ABIZUC@r&@;Ia8iFcEEbFLmKO9Jw}Wz1;=
z>$J(oS+~n@oMw$u8OBEK;pARBdcgJcw;4vA#srY)#{!b%SEH9n$>i<g>SO1xjcj5B
zLKWIBPkl7xgzlT!&jz2o9X7c2T4^gq0#<{-R5HuVkV=zwV>~Q+^eSG=X(!z;QGU73
zJ%3eh=D1DsYl>}OId--9bt^bo_X(BB?o$75>0*44#N7E1@4Ru}>>LqpSUg{jT{LLy
zrgabgtH;|wy2R|&apL=N=2e;N1x|f<Jplr@{PYFemgAh=e(tde<Yi@fLZEiqpvswt
zcCkifw||*o4XgQ^A1HTD_<K=_j)l9`X!8psV_A-JDCx*P=e2Owl$F7KMd-@`9({KM
z^6)QRM6*WkIg{9+?-|^Ep4y7#<V;)Y&nZdu*T;?1&39Iwo3zO|6i&R6#L|MJFPxbU
z|5aKxe7T~TX=^F~vhli7hD%ZZsVjqKAMWn?NDxTjRR0?7|64SUrQu#f=VA75?B{)R
z?J)OkSH*Ugh^^U8v-*94d|Q8fJ(fjK6qWBmYgw=4cPxaq;-ggR-3rPQe1;h{j9Jt5
ziQ@XpPLY3hFaEWHft3Urz%WveWtY<D7?8!Wv5XnKJ5&LM%ONAkrNwN6Y0)P~U+9N_
zpW%wH&qsHI${R&ThY8q9$A|Fi5NEL?>EP~Q2#>lrq}ry|ABZm6gfNWW>EjB2`Qq;>
zae0Y=L~F!YNiR#3mshp#g&qu{M!`3z?u4s26}itEXPrnRl?5V689neM>q+V`dqupB
zM@7&|o7gt8&w0MP-ojK<4<X%kUs%Q`4~#<hLL+r)OcGNsN_Hn0A%>gj1+&v0`{8%m
z(A4kqC`I9I0U?mV%v{KnGfnJg@<ukX?z)sTe@EuuPIkZ~+wKGk88Qrz+fD=!t)3%j
z<<hGy{4##tW1@ePz_0JGurAF01+sLHzhpZxwR4|l>}nL>_pN354V$XslKQ*$yQ9hk
zG+l)(&Mv;#pKm+NO8Dn1Ryt=>LClygrMjsvS>xDB=pJR7rE|5Z8ATD^74CAnn4AW(
z-HR@70%lq#p+r=G9oS|ZMbF>_RS*#eJlm|ppXYdNsL1tq1cde4zP%;qZ6Gvyuf;^Y
z<doAd;?nq~i<DC6L^I8uZLxo+p#Khpwga0KYfJzd65y)o6(VNm1f11kaS4gXb1}_#
zwT#<;6o5b`dwxU>(5LCavrO;*Y=x+x<-fDoNADQ>PBt7KZ2iZOl(cv=x5~%1Fd5#<
zMQGi|83W7&V%@WskHA6oR^0lBG6lZCde{3uql`<|zg?1AD3P=hcWhy9^QXt}yM%<T
zGOOXP?)k?R|JghF&r1dd8f>D_$iA<g#6y^vL^t>#a-K&?ay96xA&=5~&C8&nlcB}p
zunxwv&<d9nAgjxdd|C0}E6HZSE>LP;x8Ag5m+kd!_IKbJUzGBs<w)2WVC7!RaXK{1
zz0zRx>_O?Zwb8@GyoS61f8Rw=HD;fa{;&c7OKBGxVne)VybVD%A-p+d8-YF+=UJ_F
z`Qf%4U5dgx2$gOC3M|U#lJzw^eS*Jzwo7wNWldK*CEwYIEGt))5vAYHI=fqr?#Xk&
zIPbc!ql6irxH?LDNP`V|04uQrAF}poXh)$#R#G)AS0d!+V5BVj$AU$cy~>|CBfeia
z0TFVQK$YDxEhAR{8H=Wp=#%v#HG7VEG23)W<HQPx!Ax{tbMm6)(xkF`2HsIw##;Cq
z>y?Myh&Koh9oe-(^*ci3mE~+qVD7oD=$Mm6ix`}044_wiWf*D<jEAZuV9&omR~(KZ
za=+ey(<9Emo(0qZ+V6|eZ^wAtAG(3wv?Xf?A?@R+4a0S%&2z^dUczFbwv|M|HBa3n
zJyE66M#5-+cl;NXJ2+mU0%|XwR{sl|!QKZ7JAE_8&)cFUo>iv6LY}N6riZ(JsU-Sz
zH24IP`p@D9_Fk2`o4e#1#rl+XZ2-}b_&)7}#Ky%tUV+eO?wPKSgaOZqSv{7d)=C4d
zwBG}6J#MG4ob-lgN0?q>*Ka~8!rrNoX>Y(}KaM-UZc$d5_5En$_u2Si;V_Pc53VUu
zhiJskHUr~>M3$2bm}9|^s?_W2GDJM*^$d(QVogG;7}J{-C`!zJ<#blXx-|uuEh28p
zjhVg+_l)|McCL*9(;u>PM|QFac%itk_74l--BawW>R1uBQOdHzihB!=a%><<Ts4e!
zp)7XiEj-3wdhf~k>XxD|p#07&a0Qat7}FWx7C*lSq_=hN`_|<#(5SCC%ytQt1WmkL
zZ0xeU-Azw2oMCfi3@MNMnIBK;{e0J*+)&_Srb{C(5G+-DjzS<z>~X;(Xscvw{i)a2
z(MTy{?{}d_j>B>kRH19lbJaXeuzL>;SVu>o(qh{A|IB(Bg%Mxupr{Xpq1;J7n*yif
zN|$zV%xBpk1HlSJzA0iPUq@1`L_YZ*RfIg?lV(A1S+v54l%%T@MQ?uwtkeLCY>%g$
zMf+&Sp`m^*(?5N~zeR{TS*d{*@Rl5&B|R)u4)VMctfL62Yy$I2;nmT>;O2Pfv<^Jn
zFddz#z7JDYDGVw|bv>U%acPH_nu-!dSYg0aK26D<IK3*sv)V0xuo9fCJVje%x+r5j
z)#r^Dl7B30#D+y1jBHeS)c$S6Uh;ma#~Zy~f!GFr;*y+}_IuN%=77+cy@+P_0eWLT
zRb#939bOw5_JMSdOfhP|Y+SCi9NqmNU?7g)IQ7X4_sJoCsFu8L`R$XibX=8m+(M4M
z5a<~+H8fh-ba58lQS!8PJCq$CrA?+;(nyc*F`nH7nL{fQs2Nb8Y+<n2w7pBtT`8Yv
z5Lqs;)d03i{5t8uvg)&+6T(X}&E}wM1I`m;3bps8kG>(Wk>`juwD~Qaw#NRIHh>hz
z-?%s{fA9eO79_1d)Tg7$G4GQX#&~-WdvHiNn%=HBwe<!vp9Vfw(%i^-EqSnDW6BY|
zN+g#-p78-{8@g(Tf?pB3dJ_rIDR|hMzgtOTc1sULKaKUkI^~Bg4vMR)A-r>?hEBK#
zx9<TeNM-sHb>sw?q6;9r4v>DU@bIJ!H9_?ZOkc_q%GqJlUW5av2Ab#(cJ)$7R+UNq
zdG(tnNL65iN@@MD_M@jFYaZdKm3=G%fHR0Y{E*P%>J7aLp-Q=oBhp_Q63O(vlhVHl
zt=dH9*zs-(+wwT<biBuFV~wt|nr1L!;jI~oyu=Q)jp2*-Off$aA+{V}mW0Q^i9(h}
z5tcZvOCq72xfwsxDzB@(XVHZKMz3S!$LM1H>eXjXZKMK^ECInvONSr7)hG!rT-5-k
zK=351bTN6HV(~1vuzZ9DI{amw78kHIZwCKVJd2k|lOm%vFQ0Wcrdlwvr}RS+R_>DP
z!4%x^_jtnSsMTbT)u-Y?4W0#HyF;64b&3b2?;v`exQy=FE+>g1I#@7y&whXNh2x)o
zw1Y@cNwrc}z-sxUFpiTIxC!T6W@a1BULMBDbMpyH<xWce^{>=Sjf&1>e5;%nZ7-cB
zD0?Iexlg6c*hgqX(cjbviWak?kouaR^M)>D1Y;B{Gj)LyE;O`t13*>UnwWOhxGf8J
zPl3~bTCTmhxlbsx8Oe;c8`%Y?wi|euuh@Wgd)(^8mLV^#8N8v$z^vZSg8ikM#j5)g
zkwcL>U~VYyQkM24YqgZxAHbxj=fYH>^Tx{_Jp9niInyJ_M>#5p%jL(n09EN?tOVYn
zOs)=1S|(}0VZ@9TST7A21xF?zJCs!=KvC~pmbtt%2Q>R{0Z;GYB_<2@gFtURcIJ}8
zz0&Qp9m#Qx9Ae`+wpWTf3g*z`W!Ytc#m6XtBc$_}z}oGL>-}4}wW5U~(kGA*yUw3a
z|D}+l<5~2D%e0YZ)E3EQ#)e3XVvH0=GvP9yYwx9VP^b-x`DOuHX1)aCtoHi6V(-TP
zdeR@Fahs>viqeVJ!f>Z?`jF{g!uatPz%VHsuoQ*pRy5sDQd*qw5KCte9+M5Z(N>-J
ztYoxoef~^CZPsa-`Fgq@0bLvrnnj}XvC6Y}cGO(StA6X1Z{DC+v`hUfqP@wyw>HRK
zF`Oah(e{v9!}xtJCQd0L<?zTESYDTsSiKN2nC=y_ColWz*qdaI1vsS=TQRQble?G<
z66n6Q7C~a>5ZCO}iFdgp(d?nDKjZvEz%*ED&`;Pg-Ve|2L)xG`ES#mLKFvy~mQ=*=
z>L)loi!xLxmhUk!E9o-u4`$nbZ9h%}vJp(7USGf(Ef|j;%r}yi33b;s(xPs~K}Rf1
zuL3R8dL@$EkSOW<z*e|U(weagcyaOi-K}u1X+)$9!!qQ*hSXw);P)YKq>&p`@(t)#
z82|;WoLtC5C{_rbjA#TNA~9Yn%}ugoG**9PJCf9v%Zx-K;DY%z@ix%I0uFa)^%ELl
zTV_FWg`v%1N89XB%ix7;MpNk_@TONKya>UiwZMS*(d=stK`P?lG~%c}<VF#ng2Ti7
z!#Ytz$jPMFWV+~D&>=4?QXtRKQ}e_c4NX%@8o<~6R}wzkNn+aFHuwV0z&<J?>d9cA
z<Q=0CJ%6Y`ae^F%{Cm0vXc(YE>*9{^%J+La*w_Dg<D)?@ZNsrE^%&VW=3%YfqAQX$
zSIHQ_vttyU3Axk>Zd6=R3cxy{su!9p1(iZV<3j3-TnjwHe#-i2;nWv9SYz0rhpXkP
z!mCIXSDxAdLJn`}1f<FaVRjiEiF296P=^un2u9NtoV}?uq;i2})g^ZT-?oE~NGazD
z8;#e4zPyY>x<eX5fh-;1jv@X|;8|zp1!Ar4CO<f2fE+V;RH?EWr->%Z19-B<NbP5-
z(Bn^_j<BFR*X5V+<d^(IM364PPBDDULaAQsY%Mkg+(kZ?03Pr#1eL1xZlIO%H-bg)
z5HoDw83L0f_o5vXd_%pk*D6^u7(^LjiT@69vsUTW<3G#m|JRouAv%yF;;|_^=hLdX
zg+inCHU;M82Mi_5FYa|qb=AnyNHQ9)gHO|)iK{DN%y`nZGILzD;Ng4rP?BeHz3f}q
z80;cUGP3gx?7yrE;*f=5Y<Pcb>0y3`KP;!)NP9xO4T(-jdO$L&uT83a(;^H<!v<f^
z^j`Sqc}m%UsRyOVhy^|2Vguzw;19JoS%}4mRd4l>(J1h$YKxKU&=6@!Edt6$V=bu`
znQvhmn#1gkGT(Z2Fm@FGl)fo<LcmLYZbNY-zVN*#*jL%K(3wWplA}6asox{jlq%K>
zaprlDPktitv9S9OHuKxqK}EkJ2_2FOOuPa3huGhr6Rx0y+BbF>b38Iee`fg}0~KjN
zr8f$W)TtZX^aQ^+oc0e_4oOxWXJHse7+#2N*daxRk%u{#m=&0CFmH9Mbe0f{pwi1n
z#CmF@2rNZ?>D*xMUTPcR9g%CaRQ$A3_glrsVhuP}0Qz?b=N1iDO~8p$hmE1$>vB}!
znUilsO|JpR3BgsZdPG}fk$?plevn^>zU2-dKyys+Oi}FOA@&k2p?OTysZqKBt`%FQ
zzaI%fP{GR<A0zEycSLZ#J>C-Wcr;xSYQ318){au3_-+t;;4uLRSjs8k)VhZ3LRxjc
z+Qg?Vo4p8n{#0u6>ASaZB5=>6!$ulCmW7J9=e|J&<84t!#J0%v?BMF7mKnif^3{6<
z%H$K<WKCS(?j4pnen-?jC(f>G9=*ZuBVVkG%rgG<tvio8DU?B|@uivTj0D_YtL6!P
z@Olkoen&ObI__O=toQq~V-9SVTZLMHc>9l-df-pAOW7Xu;K!+RzyIqM8T`-b|FpI{
zOBEsq8gPFRt^$)cEM-qdE)~an$!91pS3#B_o_H@se2Q*p+rAl!8GrMsqpvyp&m<sb
zy_N^D)f>)+wj%o2cOoM70LCr4pTOhqN*N{&`6PC!n9tFMiBDSjC8jF6Nv28__ZYD?
zi1o4h-#@tL105i5pDV0(@Bg!S{a*_O|9@|sy8^Ej@M}#pQR)Az-HIytLZ~U`rege(
z0ylVtwh*!a5ZoHJUy!v_0>$U#_$W)Y)6<fI)IFMoDITVEYvbU_1;EWBPXs76SCX&r
z>GZAZtUgsqTP+c^<Wb$xO{J{{{Hz=91MYr@wD4P6VudH(TP5nF?xy#DW_gTsPyKX!
z$(E!0Kho}hD>DDt=KJ3-q=HEATB7FL#4PZgfSw4(b$dRG=6}DkAuy)h8t}C38_yl;
zWY7l?Z~K6#zCV{K{eQ=EL<RBQQUQSNZWY0xTMyvRDuft@-wL4f;$LOwhWlRw*syS}
z!1FjufC7VQpW~I*wy1VJe0L2{;sq9G4n+-3{>6o;_)CoEV0>zIeh%<akMTga6sKP+
zF#nN}N~xZH9bel$miPb9zs@@(MywDnfJOF%dqe~3)(v*-R<$Sk;?J{ax<oX<0J0li
zG9mf&j(4B<jORn-qJoYMT)xwBeU$)JuNL6?L;%pWYL7`=MruS(Zzl814}ftAXjd}S
z087w2-j9)Gga1P<{ofzy|0+Os&@6<V*E@ZFRFvf&DIWr%Vr0oLdvdb_pIR$MEQHlJ
z`}zV;*oMgT&K(9zfXss(2k?x;OU5L{D1bK*)jyK=ufL~7DJ^hOX1<2!zWpD~(|-oJ
z|L-p$ED3<Cpv?2!XX~FS@Y6pg6-9cgW}qF%QFZ;_t#jMZ9ss-<`V&vFAJhD%@5$#o
zX#!65NPqc%D0>U2DA%_AduR}ZL0|ysuAvo>?i3MJWavhuK~fqC>5`C=6a^_kLP~N7
z0TJntmhP_a8h1SVdEWQ=zw29zwcN6isr$apI*#8dSa&S3_up=WKc)76{xJXr17{)3
z0i!>AJmB34=n96&$2<6}>1D`s1MlMRs|0nqh>aAieF0`$noPhwDOvweGZ&6RqHWa)
zt#<)(3;885HoU;+sWBdXtn0a1-E?7B1BRt?k>w;t3Es~bDd+sRWP2IU_oS{|MnMN4
z)Wc>~eE0Lr1&=szcqodWtr=ZCEs@APD7Fe-me}${%UO;v+|L;nl*_0GM7Fx_gczVh
z6*2)#ikqv??M61TB}F@krobW@fF8ByqZXR+-s8U<Ux@b2G5@-t*#6)#d4Q0)`jo@H
zSIca&hZ<zI;u-(vn{pQimEqE;e%t%Yl3)fs{!!{nMhh5T-RF{fsr|QXw0NJ7^|as{
zZU=TRvwn|4LDgna&@vsfsN1?K!yvQ&eNJox8~!9x0UCKvbS1BBSl6cN#+7;W0&qFw
zqYekQmoXBDIsoe{(SllSQmege9H#)o(~vlLh(a40E?0H*U;a{(x`spTD18hJOr}oP
z(?Dsv6Jr9d*Q&;*QPYUv5%BqExJQf3g>BPa?a<AVHytZ@926PRpj^0p4exxnkLP1T
zQTq@8+}nsB{ijRjUV0fI1cqY4behBbfwNw26hL6~`j-HRYYN3;#K8^43}e@6e7^k!
zrxg@NqC-dHB@=Q!&m!qW0q0Czn;|P&;&AkTf2)I+*;EfTux!Pn{_nS%9Q9Tks1EMU
z{&9-)F+q>}ug?3o1Bkqv!!(-!n<uJr4-hv$C%6|^NnD-u6X~c;c+5w&VpG}jRR({!
z$2{-t2#gOQXPcjS(*Yy%gZBwZ*0KDkO=&|*`W+Ya1)dhb6}<^SoCvYZzN@nMLsZ8r
zp&eLR;KqoxNWs2e`b92f3-M|pQsCF4MR?Cs9#4N2o|$BT5T?a|0_;}(UdGlfEvki2
zq?bp2SHkDOwJIQ76g#Em&xfLWg>seDFd7K-1YI2%U6D!dg4I+kNjfckV#05?(S}JG
zX5&W2K3b(!H33X$H*O|$x6BI_hC;z%Tymg)30`i@D=S9GG;o;B0_5DapIJSnzI67V
z?T^leJ*IM_WwbAU;z&_|B}2%a)AhXbKKZWKDtIAmPt7qw)yU-M9p5MtSK${gJJ<Dp
zMipaS)CpycD9+^g<Ae%W35O&NfF|DSZ9x5-nMHxqP%TFt&?^)j3%(z8TLRa$X1DD<
z=|XiEH4+0hQieUOs0?O*k?LezVRj3j;e`65C>9|T2?QYDo-sja{myqRzQs&bnd}7|
zynUpNU_l|<a8WA@*nDQdiPah<JpE*Yp<3h!yp!XZ*_ULkY2f?YsN-TB3PBNbLm$`A
zk(r!rR14qWe=(X)fajUAcdkpE@Tu==W9ITJTzr@T0&9}6&0ud7PVGY-fT8_9n=XpO
znZT!Lt~y@x7U+6jJN4K2Io>6HA~zgqjo!(y+r@m=$?*9S@BciUE#auZR<5u``B&e&
z8`pco#t0t;&qLQPFWXJVnZ!grQ6jJl<26G3tk+7C<Z>nJlEcqjiw3|O`_Otq?m=vu
zbP6z3u00I-iUMZ59|s<CF@NniSts{1;PmgrEzAhbl1#D3l%Bj~<k|%PY8$ZW%2TGC
zBpiYg1#L4ZMj}r-QR6->hJldc)Q6Ne8Tb=a4r5}y{r(x@0-ThLFV9a7zk;x6KiRZF
z_w0X$<2QKVCD{gMHvP+}21UQX_2*F0!EL)(BX8UeZq$a9v*O}ii}2JD*8{N2R=}^D
zD+ALCA`6+VF19QJBJpi$R9l16QSH3gN}^}_`Hvz)i_V(Zb`~dEN8#7Fesre;q3Ar1
zd21Kq&*S>|VN+$6BDRH$&tEmfe4b>>^k}jF09c$$JgN1@{5X#T`=3wmwe%PsR-leX
zgkkrb@`$708Gw0;or6<Pz#Z&ORogO);;Ci2*S~j`+;q<34cM@_HNVtm!GT=Wklgou
zOO8mw+#<2hDHYLndrGazRYj{NkXyG^571XzLDE8Xyg}|*9+5ABZeQ3Xc&ydboUq+O
z-pCg#RowpruXSnu_i+i*5c|!%cTF4*f8ec76ZvijIf6dzTP3k3fQvOX>BbR}F1$$g
zJ0UNk${^<?EJLaogr=ySHKQ|#+%HWw@;%|Ns!6UEEdELZ$kS+WZ5@-~k|D{X+Wso%
z6|0Mw-3u^qwty*ibqZ$g5G0WB0$O7$xYu>L=<H&hLq3ogQ2tPv@~)gxj3Y)SiNnGE
zk$@(dY;{kCo10BnzBXw(rkMT7jr;kQnn{vBi;OPb*M_pHBPkcu)3RR{cYWVEYPy<Q
z;E8xP<lYx2RkP&_u2mZ!DuGa6Yk1284GA(7rWmJqk8v4EZY_Nn#Ac10ii=UWsg<6)
z-@Y=Xxu;p2Fyvs)QnB!=)H~nqp)r9Y(8}x1)C|bt2Lo=w*9EJ1;eDz~HQCh>6bx2J
z9F-y$H{356(wxq>4ZWV*{aW*M&*b~V<yy^7E4%JX(G_k<v8N539rMe7Mu7(0Nsu83
zZ7RE`G?5Eq7d%(@8<Uowk$odI4Kjf=#XE~LSKyPkp(@(-6gOM)CKOE->{8U~h{mjE
zy}OTHUfvR3js$LAOE2=qFXv}K_4Bp4KU+yI%LSO83A{h8qAg*mgzkZ8cpJ5BLWs>I
z6H@LN8y&Hg%l3T#o057=v&d_a0WN+cGyU~{G|m4?zylb;4FYGJ`|J5eH<L;mQviZ4
zrZNlN*=T1AK*7x<a3yAgNk{^RJP9R{^LGHmw_V5tZZIiZu+Ur4_?PF&3V!Tj3D{ZU
z2<)w?TDKe2CfS*@F4nDZKxxF(i?JdF{M=E;*t3yX!5}Iu$2^NhsU|}K<fOty^4FX2
z^+$)(0BTCpc-+qWI&A`6-M7Tgw_9h`j85Yvu5n-oqI-^molPx(KZa0Q@R#=hVB>K{
zh`cHc{TM@oueo6vtYRNoEJ=-#CKUViCMZ(8>y8eM&;B=54mV%PC9RE7jRM#zK75hr
z4p2$15)Z`Rsl_v1fvs|p#5!U43b!p@DySrWy+A+MK=FFvA=2lgAj4_}(5QD14k7`p
zYqnwZsC_;yz<mreygD5@iYQq#%<wLgRxps2$&`JR;a@N78DPOL4Pm}Q!LlN6;Sgg9
zP*|u+mNnM`=GVS~w6O`H%?qf0p_*|Qug-wvB?z5ARu6UJ63U7_C!IPGo)yg)c(L^`
z12eh06TqD{b*&1+SdP8#C~<GX|6B<&V+Has#Zk{Sy_u(;CJd%2e4HF?F`w^cC~rl@
z$9$S#->f9jOnSy(RVH5x3Wj`h?aNOrrOmUb`leQWf?2vOu<+C2xJ@XaF?UY(Jm^Ol
zdo7F2SX&q09mN%XjeNa-N7)Q%us|YjsQeDeVb1GJe#gI891wvc9n+?5_L46#j^+Tt
zJ*?|aV^lgsY?VDp8I{rL<=O0gZm&_?h<G5LyM)?jQ?8D^;10%igf7c{w<`qG@aS=x
zwtTkmXRa=fT0x87#Fgd*&4i*B^W{y1a?_VN)<qR+D!a<KE-F42!&7l{<_dBF$@v?&
zgfFF2cJCjGg%?v+N%5Xt7LL4*?jMi8$Ssg+5PsUAx^?!akwW@5x`WsEG+V+hkD%ts
z!16gAOd(=^D=Wj!j_``F56pI^x~SRGE>Qz+!t%&kK9^Q7pm{35_BlbQ!_;=bI2xBS
zYG?f!L9Z*==V&xh)>=UGOaPqSbIrv9Z;iz}Ix;nXn?|Yh%iAIw13Y0~EFGn74IrV~
z&oJUwY(xPNr^QDjk<=J-^Z}kCRM&Xxw_66)3v%;DW}N;rG6k#zp-LSMgx9@4??g)|
zI-p)`fgHK4i#6>97_W-jHeDPV{CXoTpBff;bC9w>`#5+#;06VkB=p_rm*_xJ8~b*D
z0Wwo0h!7S`$c8zxQoa@j{6qFO1al3DiIMkZz4jopP!_>(bzg?Z4WP8>qFbRhGl6I|
zmt~`FIZ4$=aOO;qK8B#JEa}A>-!GHJgs>Hxl%G*4*zrpnnwMi?!528gOmut$i<=qo
z@`^so$C^LBKmkp7)@*dt&jD@7UcM29#zY$H*_H>#QHRe2__q`8g5&pxvK6YY&tkbU
zRmaS+OQeahi;cGSqI%k^h*+3@Xdi!pOzRt))QJicS|C~;<LYol#jz4XPOEmkqQ`a?
z*O@+i%|sTMN9T(lunm5OLKyOm(c<0EI1;=E+?uh{>D>&0_cuS>eY_ojBCNq0z!pZ<
zllB|UD4?{<OjJAHIDLe|Q1T}X?QiTvFdig+irlT|U22Ad2OP-T;97jHhUcHR1M2ZI
zH&vRB97<SCSi-67REIA4j_g7JsR(<dY&S;-qWWIsYt*gv;x?`(N4n=ys^Wm;e5BxD
ze)o*f0$BwtmGI;_sLlj$z+l{QW=J1geRyOX?SkY6WYC`GhMNU;1(mwFdN}r2B~XCr
zOvr8yN-|Ve)EpJUE$BZZyyx%i^21PLm}{u>a$`bg_^w^hA$AurI;q87D!QV9Jl=cO
zH*eE<X5__4-)mgtcHkFaP9vgtd>;lq{MGZ$4=O;Ju&sWIwIFEGamU=eI9s}Y3b;3-
zZfO`rwySREH5EAA0n^-8(Y4})c%u`&$HTW)`ngCn_Y|q-KMI-qeKk*(lmsy=s+adq
zUCT2D$TwOWOIv1!0l!9o_FeNT_&~Cc9%(nq*QOJ?mr6Uj-ysqfMfU<_jIZY<5G=?d
z7&SqQ{p$FOk3OHjDI*%~1|8Mzz@xHnah`;ASuz3q9A4brwf4qvcTi@fH9{mH^RvI}
zg5Y216b=<*NtF+=gd#UU;!;p#cJC7?FVcL&qxkDJ8qDs6Hn#smRQ7kf%K-O@*4#e=
z8gTsXEq!gZvQ3=(#J0$|B+o&R3!BY_VdsRwB>IsDW{Hga;mxE9hGC4}t`baxYKehT
zsymblyAfL6Zd1$hpC@yb)ij4aawgLbC&{~b!}m!Q3fF3rjYh#{kDRcOlYGTa#vw|f
zachZ(;?91y{XKSCiMMmvgP)E(cYc0TVLgmd=$U~A8zY5$TwD$uR}qn|nk?4DC?p?u
zn1eLZ)JD%K?snbG$Ec-?W|Yr+a7KpU2^Sv>{-8{(iOCgFx;7h5!i}kbuq~_+JFdW`
z4v8vGybf4FY<b8}?(!7jN8nT%d9L+PiDNn!MhSN?5I#%4^;Ky~R7T4@qkpwp|C-XO
zGi|fni`v4eBTdZy^CX(@OHcGFVu>BVh7$H3ZG--S=}JeKMJ-dfF-6tSL03Qcy@gX`
zw?a~ONVFk7tG3~g0iq;=arapR*KTwdeYZYy1yVQ!FL$WOv>u46s7_Mgc9rky<5|8H
z2InA4mE#P`!p$w5jWD^2xGH?d!0*#gd`H5Vp-V+JCZR>X_Xg`T``pvSN6qX4_P#|w
z%!rqA!xD(cv<WK7i7j)BGhhx?Uek(3lgxCCO9&rg%}S;{4kE<X#y*xcwX7a#oJe@t
zKu`c2eSD9xRf*WbtbE3Vno$;!=R4SR-v}qT4)n+`z3;!qo^mkrPVzG&9m{H(S*THI
zc)DxgH-hSf1*G~zTyPQbbR)$AQaN--0};|0wJKW^B<{hU)MhoQQ2Dlq3T%;w+KXQp
zyH)4z_rvH0Ly`-l_%*47a^a;%y$>je$d`HAMF_zVRyiX+e~%r1ynTW=BMaw$sinH1
z&EYTq@jp^bahcKSBPLKl<qH{S|K7JWUlqH*^USmNJj?G^K&p*5&@B#r6j)|w1?`Yg
z_Qx#w=G@5Z6}@IN@fLMjrfKjNgcqk&4pl-hqz3F_c(oKz*~3y0XBmDHyO_`36RNln
zU)4Of(RT-vuNEp)A8sx{8^|6YE6U+~670dD_F`w;rgWn%cWO;BHbu<cep$CEr~seX
zwTmP#SOF6^4$9UOSlmu&9a?wC76@}R@%+cF=@|51VIl68Rn&mafR}ww&Z{aagxxy0
z3t}x)9<2AUIQx3ujaFg^KzivW9<;J^*+?*Vh{qXM0u>=)1HFi9Lw8fBQ-<g_MM+jQ
zDs`aCULea8LHONDHDG?s;?2O`qnq(#qw5|dk1ps1VHa%Bi(fh=$zlu_@3}a-4=eL+
z#Nn@fmb0xUzZd$FN!cHOex%%9@&+1oSD)5=<HxF|Q*K~!=1mjeLIM-T#JBKoqz<lq
z2)G45Y}d7EN6jlcy4bo%N@IrhpgL*?ak-&s?sw^{*Xg?UNN?uA<BxV}394P{{K__?
zw#!_#mIN}vdwt)<cG#kIrZ7Ogvf_=z>e^-IKO6nOQ~6$F%t<WAWC?kxzk_)JHhLEp
zpFL3m)uSk@UUfWs1+dDVOD|3yeNzHl{rmJb+9n`RC1gxa?MXiO^^GhdC4L@=<XyTU
z&=i`cgGk0l5zcF?+i(8@37UP0N*+Y`NV<llCI2q}bx&~=urKyeg}>fUAz1_W(i$>-
zpw$0rVE!si4bdYYT-bHnUX`sx{4v6U%w<aixp;#o&Pgn2Ef0f8oKL}|sivJwW<Z~d
z8&p-ZAV8K2v{XgF;jf=F(6$9{gC0@u2Mf-llCW!Ru2Xq1Ln7BHw%tpy=^etJ&|(U*
z+>b7+@09apV2drFBX>Dy6;7VElxv^y^Q?w|jFL0VT<dOhM#K;$=&3CyCb5;&IO2z&
zP08>$DID^R$kx=8zRTa&v(s;LW%bPY@a-}bFY9vu?bol>$n<>s<~@<=qsF6DE?a%?
zxsWO&ihM(|`JxAG<H$~P>s<8(G68LEjv!o@rB4lE)}qOaee^}HGMLgX+^<Sv7Pr`4
z=K!ewHX>Yq8uS&u8VK`jyz?6(I;A_M)kj7@k|y-$;u&&$5TV8cvPcYTRL==xCGpi{
zxoTi7rNcP6a91Y^yuXIy0J1CkqJj)>fyOsd(yV-}Dl(XoYkH2Q88{tK@rZ~o3gia2
zfMq07Y|CPC6G26~Ub<#En|H6_X^|8!bM~;c-u%b&++5`4`2}mI`s$;Et<%qcTq&5C
zj%dB-dy_8xG@_r!&0*%XBE4BsGac~sd0Eox#{JguC#GiY?`y?e@Hz}?AO%;+t{<hF
z;OKs_x3}p9nxZ;SS#S-oW~??_b)?g@S{NhXrt`v{Sh(Z*k55Eh1oeM}KeX@XTZ(_>
zdnS3LH7(My$T>fDtDrfb^WsK;6P_rc-QADke3!=5IIzeWZ9Ti&t!=clyEA+)Bpngs
zx1R6|&@nX_D@z&z1UdFIv6>8vb@sI`Az^0jpWTA(1DHLe<Yu?KnZUM7{uSkbTCs!N
zr-nu9q)Ft1y*ep71oIAY5A6uwUAN$PKR6?_E7Rtk<#cD>e)XM@B;sbXHtfXKo}Mcy
zpT}PX<Mg%o{Zdd78Wb6w98{R?f~97=P-8EFXV6{dZwDeSWJ_iii(TRpwbY6+ogXcD
z^1n%CI}}}0?+!XG{Q4lH@SFOM(wmG}#ceE&aT|WWm3GbV(p32$Zaj&pkt&#CMzp%_
z8ynXYFJT0sANbc|?S||#Iizs0dbZ!O;O%G0*Nl1{Gl)FT`Hy><?;h#^NcTVN|2vtm
zV~oZ1k6pfhNphbt=gF%oh6gutW4Ru{yKk3pv@ul4+@v5i+<QGTy5}r4xW|6Y53z<<
z0U6b_+=Hr`CO&1qi%5y>Y0Py09^54hNd(zNT;8+Arnl+Ns{WP!==?-b>jlPP(pY%<
zK0f^68`2~NcBb9Me%84!KOP&=RN_meyAC5}!EA~E(CAHKxr|%M2{d&+K54EpY2m{s
zzL<L_m#Msy$jzAdd?oo5Ei)klk^-yIs@pR{ro!1@5sM-QPQV(KCyU8V58NVUZ{~k2
z_Hcz2c4yjn*^%DB-3>VWI;a05P(yVDG<a?E1dwJG`l&q0mtb}qP7Fl2b_-?|m4Jgu
ztzZcDnYgkVVaG{xIBaLEV@j`L9&7?Ocomiu%<<P=VG)udYzmZVnxkMKp?cP0n(_X5
zA9@DdAA1~z9}WG9_?b8{+iXhw!mv96YY5mA;x}1+oW6TkszCD22Xl;?c8$NeOxRUl
z2Mbqxn>wGyjZUogpRqn^{wmaIwekSmO#Ai;GX)3lyQ(h?e#cl#baWtJfZ71Xb9Zz4
zC8#_Cq8`&vm*2q`=wBcEh#`u-lnssSmK~>WryzDFd2w?|9-2?FC2gF*vf_XF1`cX|
zKU}bo#Zt+=xtf6?UP##tRI~StnaJj=E1H+&5$>OyV9Zj}7SzbmS6?4fe@}Lf+4)H*
zwhl!BdBr5b4(&+Xw(3##E@ATzQ$o{mZiCo6arLNuL7(t6V51?la)g4ycYqk#b`p7{
z98`Sjj(nQPM)f|$1uSMEQ+yb|_L!<)z##-PCA2`NpeEeKoe#y_k9^iH*z3-a6X%BO
z+U{>3=!L~Wy3X<7vN_HR^m5^8e#Mo8s_l<A>OXjwS;u^3v!?DkVh9J1P9Jfo$XWz;
zFI!rDWIbWKA388IVEBYB2z`&M`{hv<7*Ij*pawclX4DH0_Hm3K5d2aM3#HdUJM`I?
zX>VSdd>>V$-Xk|H-vW9srdj6p2PN(eT*y~Wv5({?`0yJOc{gxx(56u`tSn}$+@uA%
zUTHlGYCI<w;DRH}Ivsk8ZkKgJ<{Ws)RQ*H7kQ6|U{lugch02Lz)m7V+p639<T(|Wn
zbJ{-%b0<Y^AitA{JHh&ojLDc9r34WP-1uXLh?-T7%hD4=7Ik^wq?C}{SI()IK=q8w
zg?<?%eibL0&4a&zn2(5v1vA`vlWry~VP1{e+FJ0;Dh?8%Y+=Gl0%-!hqnjKWn_kZo
zcP%^@Z+5FDMtMpTclAAmap;iBe2X9p^t(FpBTBQaczKainWU!6MD6U8!q+P+-{hgl
z@RW?4xO*;Z6H_|z0zO?9CTJ^yY9v+OTHkZcq6)@z22&v%_>2tC&C2ba)PC;{(tVxC
zH_=D<s07*Db=B>U1Q`!4Huw&_l8n2GtX%y?$ve;u+JHV`n5<B4Md$N@YCrr<Sd``2
zjmKK~gt24#imXQvX$!BEMwf3a=Tk6x4*GhS%H@De#_W)L*si`8x2jfd7-fnU>q|5?
z7US59P7AXD8(@9~Fz<?16nwa>SCO!J?yab+7<Zw*8#`^MTdF1g7nkjRw6p;_{es9+
zf=->Oer0j7mCQX;!mC0dM*bu)ujq`D$N5$Ycg*;zw%=;&WlSJ19x0Q8A*~-|>k6hz
z;}@pXIx0-pbq-qw>9%AC0+p5P_uFk5+X4m5E=-NnaZa+-b-iQ6v&p|OeF4Bo^R(EM
z&~m(>ngJE*ig_w1tZNO#P;(e$?5_W#GVlYpr0<zAvCKm|#`iO98^u6X8kMMXS8417
zgeIX9s4z1ZuJO;Kr0V*1T;<=*N9P`HD~kZ%&~la<X(XzV$~|@Bo@Sal$JsZ<urM0{
znH<2-T#B?i8ndFCAbmYOpL{DFw6;E(5KO*%{1f{mdPju021qM=>J|7bc|O8M6eTHa
z=)eX@ZQq|SU6r0`L>85q6It7pHXL3<jWkxYnwuy_d*v?FsU#m<1Y$l`9Oi0_ktA5V
zg7y^l#10=@#@)i}A|2F6q<d%GxsK|qeexv0tT%mQFHO%KBGet^mMD~#lN@NSR6Dpy
z-;<q|L!D|PL)xar@ZuZk>w*9;x#=(Q{89&ao?qPK$e|DwI?-#mVL$bkD1)5mY1<wZ
zw|pOt)%zJu{bigfPZMy`D$m40l}DRQ_c!nJ&&Kk6v_I?0Uy=mWumm_;fL69pMvfG%
z%&ogB;ZC?MtSU`N^~>0i4@XxnsXbUWsIl&OPxT|aJGI0}Eq%LwB7Qczjhp7Rqe2?s
z0(Wu(Z$Ws4wvK1hW>gH=D-;st_J-(ya|Kh4n3^k#93jyC{z1PF_f2_@w%jmWiKA6R
zFo@;ODW;_;VANBH>&$WER*F;2A+6WchUFGBDR78|M(g5JX%6>?_%<HDfiV(Laomrn
z=IjT4sb<bC7rBLKcKk%2tS$#3R3ZDN+Jc6@aVN^5H{qKrJTS&2m*(7H<W5|~a$$Rs
z@>wc;7TB~t?sva$9#Dzjomx$YJ;H8u0SyJExU3w~1B@&hkVrCp0EuLGr}-0*NWKb7
zXN|GIW9c50A=VOm>j&NGU%l2ePqsRYH=YG|mM0ceOJLVB9%6i5C5Oo7_9^I6F#ia%
z&V?o16?)vxqE@0Aq4lU*uXZ|**Pe+^m+#{#Bf9ZpUvsC{5*wflse}1{(68fJm5=Bh
z?tjYP<WaOH-sNJ)_p8p(P3x^;)mT<?zD0_ST~z<BLlooL-jUKB`&`2)P@Q)EQJuc<
zN5VC!!b18gkjk2E#U6kyF{&@&7w7Z?k1gH!qD<)iLUmnqDjjhs{HK!L;aXj%w`N-b
z`>p}$PSKj7Z;O7Oh|o_-Gtsz6+jZ%g+w5Tr4*m9^6;&`{ya^Z1&KOaq>yND#=bJX(
z<Q+6hT&&YcUlvFvR^bcxAhodI4kt}%GV~N*pJa|oOW}Qc2UbUdaQ*mfM?EfnT`Uqm
zB_{mM!_z6Dvfky*g+~iD?C1tV1=;KZIvNIDB<PKdl8SL#ImN0=)$1v>+dxC|)XXHV
zjk9w@eV1YIDXGf0?gx40nY%E&Tba)WRWx59iqgQa+Cf9Khi8aV4mRUpw$5_&W9Nf|
zj}8B7<Fef=rBhrBHHrMot{kt#s`Wl<EpurbDFq*MJ?l4ygZHe7-eu%H<L^tXQvCiP
z>{l9(`$hLhMbZ<2@S2;cTx3DTXu4ubBelETTmY%H{v_A_NPeCCuOY{u5|jWZX@3K+
zf4^r2UO<$!No3y5$clyEu4Q=Xe0HM;HeNO1{xb(d)s@uUIYH-TnSQwZ#?ps+qa#H|
z9AK}4T1enCIBN{%roao*tPqKXjMLHri7Z$TEM;TA-*b(SEiun=E2H8F#Y*NJwAgu`
zPV^b7dB`)OsOh~Nf%|CZo;y0dsvjbb&-96t=DRRFc@`LJhiGI4l%dB^lczmb-|4{M
zJl+D!27~<<pm_b!e{s`0Ng=SS#6yEpzNmjyMXpz7GRs5+kL^B<=rUV4G-rkQf^lB7
zGcWnchimZH6>mO9;%?dDAV|v<*kGhZfyI70SA4^m%^)A6I#I?Svjd1__gGF!7i6F+
zd1a)E`E9mt=eW4VxQrlK8pL-?KS_3ji_L0&$)0&Sg~R2o80SgFhMS9@FQb2_(}H*b
zKy8ck>}6rtL85CDJ%r9K8dh&%vo1AMC_7MHcrgxOExij`x;%LvDZ!AzZ<Iq<aP`v%
z5%-mjOCsb++ehfhXzcALeXke;Ah?9&h8`w$LwjnvDM50UiLc5s&n(D8GSc##NLKnW
z($0|@ara%pg1uMv>ByNykB9K@lQb2y^Sa%SVK_{u1-@t*A;ls$HScTJaM1;`e*&7B
z@yQzHo7~Ue*-KKaW$XQVoY|(5+3a3-1JaozPTyS4|7e<raqmJV`|GXyyeGat+`nZF
zL|X(@upttUY>dvv*+<$md4ivBzQwO}K{T&3=zdC#+2{V3Gx^U8tAhk+h@#oXFS-B9
zZr(7Uv%edC)h8=<?QZdKmxRJ{Y4=U7osa7kAruUcxj%jU{G<v$XIx3@RFmln$mVD@
zON_%cQ>rpi+6i1Hg$f${Bj}@rIkxVvDk$xb*R%#XkTZKQQFcq*jhQpFCyBzPPnPg`
zNz1|G3zTVH$&=WWNzTNBdp>jXlwo{25X?2xW>s`dlG?PFP|k=v%wo8#wCuZnLi5@L
z4n>n?IM4k8_Nfsd*y$tm?7l{D(gAjvAynwizcH2KQ=0|bUtJt+)f!r!#0hXUUV-F!
zbg(V(zHe!<B$Tb+I2bJ=np&>8OYRg~^%UrLK3RqOUfTMKuLr7L-iyYGkTn+byF9-q
z`9#ekKuCVPypH`y)J?xFBAB4~9Wh&8nQ5xr!|Y-P-R#7I^N5+;uf_owYp?BeV%L+&
zcbOjRCK<wBiS%EE%cqf&FNWT}UfW%c#ZY{I$>JK}dR?bzfIeEP!l!-vQ+10>ztVdR
zjG6WD2jagoFM1l1DaDD{$3yL4EIG5#G--@TEpNtA39G9EGxjk?BD-hBxc+uScwym{
zQpJ=NvAV^rmAtT;I2s*$Td@Hv(;Y4QT3Zv2D*W*~5hIoO@D2LUaTr5-u-p#DvRH|y
zMw7SbkfO!T_hC9OyDTg!Ah?wDvvQZ^5L}r$kxhC!opeCG{y|x*Aw<2ByP~()!c_vc
zqiSw8cHAAV{E;Qhu`JYK^^Pug8U%v0ye#*$;i7_>qodz4w-h|wAr^>fTvlU}=S>%)
zum`;wln?k}iECdo8*9UuFq@-<#tS`aUHVco_+?Uq_xEEbuTP|)#m{iOO*f??EGLNB
z3^+M{bM!#`B_u5KjR@i0{=|#GFTj99mfE?Gufy@;e8}OW(&c5@-Vb?*ZJkq~Lv;X)
zJl8$}R6cdD>%t+CYFykVzcX@+-k5*7ozuNZs=L;xR;UYS_^Ryw<AN+CN#D|qK|TXu
z_fAr3xsFb|rzmv$I?4E5jDz@3x@0LX*;!nBoH7@|^%0qt_iMmFP`&!K)QF|zuE)hF
zD%AZ$0!YD&3HvGc8shn{ZwGMr`y;7jHsOK#8JLJ4_!YfmdE4uG`poI{g9gj>GhcwG
zkT%bQNGl1ub_Y-E92@Z(pK%JSwhI?vSk-_uWg6NizSm0+fNYx9agsnV#3Uou6r}$h
zlY0^eQHIY5!6c>+gxv_ly(hB}@kRmid4%L8Pnb9q>~Y08YlK|{O>$y$Bm}`?gWT4!
zYoXV)n)O5%yz=1TB{sU4xLWc!#L|Ib<OWB(B#5SKs1>OLo3r%pav=fw>6pZ-fk=hJ
zLR_j@uubtA7^PW3`RS32fU~4Tzs13qh^|zdM!<OI1*p8m!`jFfPDjhIG`N4^5SIq(
z;{^-$i7+Y0y*AuBk2kC>z|)Pyb^bysA7A#ojU%UER(g`r$8A~#$M~LT)vtJ-u`STG
zYXSO+n|Tylbv)<_v3k8FHd?;easG8BFN>dyn0wMAe>g2?B4uQ{_vrH-mv3&}2t{Jc
zXhQ;vwXkYS@E8@{GF3!WDXG8)BMY6KbQqOre<IB`4T48-y?c|qRE2?B-m*fptpV&%
zZgvC;xmMAc*MNufV`A<zV>#x=EusoT>qQg@24{n|`t-}kMB#%F@$US%_VO>-RG1v^
zuCI}1kuJ(4F%|m;Vfa;G!WKI2AO{s<Uj_5IZ%)!X=bOmWJQec>Zr1CG3<3MGYzSjq
z=g{v59Nd_?r=NKweHe4h-khs&9c0!idp@zM&%Xd$iG@-b9={#lL>ke)9xf7E6<V==
zO*EH+6dF&eXQX2k(7g)K&$-v#UG&4IF@T!<jxaqF;hlw8`asUo%A2o;@owZQJV<EI
zBd_H?TG8f>Q{T6CmV*_h>7JLnz3$2WB*3I*hZ3MwT&le|eG@!GA0ayYY31sd#vO7e
zA*6y^Yy?L_M9_l4)c);_h_nSc`~w;M_rswAwAn;6K$ii>2Xl72*hf)suq?QWhWwbw
ztDw{ibzI1&g|Fy%k-#l@+aQ67@hmmn{O%VIe*p3|tCumlzux506y^0p+wY#&>lX#T
z=Wq%8d-e%lKlUds8~=!~XlP4O`=PyRA=e|L<lhrH^V{qxc5Fmuq%8AWk=F7K7?F4G
zariDJMt)8G{tC;Epv;NI#qTu06pF`>R4+OhW86yd_20R%XE4V?Q*mfh7gTYycbp<-
zotqYQ7G^fu>=AU{HGbLrm}0^X_*aDbDEGI^_8#ruDJrfAkFuZ?&X4U-|0Pkm!fWv9
zBo7%aRIsi;B%`Mzplx<}&&M=6d^5!9W?i<g43(X&Z*RH`mFB6$(?zZ|KW#&m&N#y-
zuZa-;l_&USS-xlzcy$a*EzT~JTbiG7mYnr$H}W<|Geqy|nUv&5WneTtI)r$M5I@2Y
zAtD*)j5*ia#e}+not$sJK0jIPo)`Kq-7vKDHJT@PJr{|;ME2{~uD^GV9m86<SJ`>(
z`+1=~&4cSroBtt*k7NPD7HYL;683+`W(&;w4{gZnI5c~5lGNls8@ZKwTvzy9Op1kA
z!K*m7Qgw%oYfj2NjLpx0oZw-(BK(oqc(^UE;rpYE68*%yBNjAWxG$hLlDQe61tJVK
zT^13<m6c}`F!aiaX<UwqO!I4wxU?0O=9m%thCjB-Fr`T++5Igz-ja~1$|p9k>)}CF
z=nHpjuK`a`0<q-aohCdr(j#_GM6<t#?DTQ)eOxCr1!ljW`Y?05Z($bfQ!W<e#W%6T
ztSc%lt$7;pr}Id(#`yx;X^@Vd{@ZxO&M6HH3wY_~o;f5~7VP*pGF%9Sz~t9)KnBOP
zM012*D36}&zV>0P_-U{S`9m%0L>zqC*!B15_?ABn2;ey~Jlz=P?M%@kEj$545oqvq
zaNq9Mg=J69pFz#d(FBn<5?+bn-^(KkWRGAOvcs);*SF{-BvHUU352K^xdG|8D@B~r
z@fPKTH@KLoVe5tSLk_FA7w>*HMk}ZIeW$vqK1cvNFj(J5ec~uZ;=7Ur3LQ|y3*_C$
zT^Pa|ziqcWAV)+)f^Pg>ROU49Bb;8=H)w#p=(hIN6TNdysYT9VSUIBRgD#S_L=$6F
zrXs{#%_u-T0qdro{=&q*kF%z&4}69~d-{jE#`ne=6-jf2K_Kk=3U3L-Lo=m6!h<(5
z2;<qBN@qf+f!2r$FLquYELM$ZGwvLH;d|XvrQwBb16a26D#?oKApF^VKMJbb8~uP4
z8av3hdR)vsCiV2xv-ID_iGj`Rn9^l&i@Wsrh~cIR%>aJVwA0b`j!7z0nuXj7R4vie
zJej7h(razuI^L0GeJ)=&dfut?Ib1sB?K{vJaykH#wTr=8wNvagV#?=8T~nej>=FrA
zgknZ^L6A|_B<P0}k}@>n^D*=8rMp6<``>5zv`x+OCR?!)PkXWp#Vx1BfYq+=LTLAk
zqQtoD7;9Hw=(A=Wnhb&i8_^q(3HBIj-e_R5nYb%Q0z5d8Uf9GW5(-lTmXfQ!6IuU~
zy8jW8$EAVk#qP~7r@tHF^DiI^Oo<(_1l{PWb>5vGV}k8A>=%nmEydSlwM^fN!Pf70
zClh-Oo{8g|5DHLepO%%iGomlh&l!W(TuTwso`cr6iyZ%Q;d&r3K+m+$g)hr|?%t~j
zVS((blV06Hi>;UT=<Q<E(dz}KH2gqwgnyxQIy9;ookg(iCelDhXg!o|r<}fJ^BJ5k
zHi`6Da)q9!j2mAkem<=g6Hd&^6}sZNN_rgB^G2GI{AG5cUYz6OySzFNMzoCM@bC?T
zTFUS_J;jGQeP1ZUu64uZ0~vIdvQvo03H-o4^FX^XcA*TSc%sP*Xumjf%-0v}iqT<k
z!~hkkaw|PL<H?(-3O-E9g;#E=FoXAv-zp7>9lqCPytPB^ask(Iba1_@XP|9j4;9Qf
zPOy{CuO4Bo)?alnp_oC>J%m3_Y%Xn1zLC)`F@Ad`^u9I=??gBL?#&!3>+FVa(N9Kn
zFki^0_Jo@(8cbDIeIt)~O$6&i)T^VdcS$icCHD;IcbI+=A#Qd-Hx7@PF}>*BTI{s@
zQABF>{ZhRB`-1z>4kE3hN##`O!(;Wvm95`u4Za7^H%&Cd==4<QWlLWbD=SW}CApOo
zdVpqS+ABAB%d22kOlhGAl=ufv=cIj3a__?Q%~i|zxDE*<^}8{DNtSC_-UH+GWW6ii
zf;$;RVn!Cd<3m#Wu!z4{0O-bPI$*DoZ!`{WhrR7|4RBk06&0uYIbf2A1t-gSYtjnU
z-ciD{TL@BrqC?y^Cq;F?2>pPrl+1%aB^t9?s5>{zzySOb>nE}hbHfnHgj(As_2~BV
zf<2+VxVe3!zmDfW?JCCl!~>@B|Cpl#!rOK9iX@yM?(SXPM1{J-sh^rJf~*soQe%mU
zV{>2@#pmpDh>$Q{{}epLk+>*2Npv$1m0JlsOd4es=g#y`4e~GUYy)c;eV5EfT~_~!
z6St(LhgWjoABaHnU{%rcTG+WZ7V&1T5TCrATag!Y_naQ-%lEsDxZXk`qyeSNl(qX&
z&OfLW6xVAzu@D#06^(i)zB??4+-j?;lU$FYtvmDApJ6c`D8!xR2tl7oXq=kgfj@P5
ztPgBPjC+G7fkGJsw~c&2yK3XCLS%;QL-FOfYg|lSNza_?P)9?Fjh$&N0Qo2&me$%z
zqe6EIbE7L9y0u&qtB`#S=hA8KizVxKUzJ9pbRfP(%PCeCKgPO6n#L3*OMo4PUBLPr
zEPRE%YqbavkJIQjz-+Ei@+u*w>$ee<>u-=D2ckux=Z|^(aU^s-jkL#d;TGD{ywhN{
zPq<U~&`k@x3Xj^|`meW0*DS`A6fC1rie@i+O`=NTG4o8Jtpk9(f5P;j9w@VNl*=-D
z?3Qv#-qHPiYW&rb41;JJivRPwe0EU(dPW<$Pa+J-bWWf(72*6%#BNuZL?DpSt?0w2
z@I*Sl@h>!sr(Ue)Q+vN~?^2+{qq{OnxWl?%*(9nR;;BXWc`TujZmEx#GR^+DR06&c
z7s^4*<;cug5z`USf$%(GgmVno(<}F)?Gr$2>@%xsPv*i2JAvR{#2$!W+|Gf<{sZ7N
zUaV+n!bUg5*s{ExiJUe;KQuCer>-#SwuMvK9#5VOQrLOLCXZ&&6D)+A5Ednp@PVkr
zKHGL*O0^oVHs2_6;hy4J;*AO234sqxcszy?&tsA6Ll<`iQOe^8AZ)Wq3~s3b`bs@r
zN#LqjE<Kc>4T0d+LMx~}>sJcWOruR0A%P0_%!43EOA*BKns*aeFoy482i_ULb(3s4
zK>0Jey{djtBDgDUGBCw`0Ik4@l2~zOH>sM!O`(GMWZnQ*8^w{2xmOV7UG^=(IWfQG
zxJ$!VKoX2gGR$Qc&^7jPCzD;fYSv0V>l*Hgwm7{PTRB4QX|TR{&zN2y)N)qw#(HZS
zfMFzdmS#4X_g65+6;2!Lol<DI0Q=?r7Ky_5U3$W9KA#BD>neiaTjh5#Ttl|*^FB86
zaTV*Ez_7l%h=q0^cF8D^M}e~298WGwj;u}r&iKTS;X-nxc-U)TV#r!~O@FbJeI34c
zeEvVmKn<X#D6^dUXEW}U-an!b7~%4{<g`>n$+UH{*IsAcHN<pegELwUw8*o?)r3-a
zTxcz0Nuk6&s+f+gUH~rgOyAdrp-ul6%1MGI>=;Vztc+TJ#5r?B6)jDLdw;8vo!XJC
zsQuEvs2q5>rgA!>s5YBmh90v*l8gU}eBZAHi1PSb*zq8}jG#N*$Z@?`i_iW&x|3)T
zy(B;xPnUxK6D+4gPSy&z_XQ!J&g#vaYS(~%V-QlrWI4l~5asmnKq=~DtQnZnpKsO*
zCAHDOOSFYYjWJx9KCiwn8W#jB*Zn-eGI+8i9aB6Pg-9F3T+0-zz0(QF4o?{F6YIs4
zTg!V=h+`s@OqN^HZX2I*`$cYyWL@Wn@?yW#=L_OQD=ByKi-&_jccEvJ*aqXSj%|5+
zOCUba6Eowb>{5?*E?e=yh((iDJ}`in;t2*Mhx4-eXtA4w*7e+Lp+cLUV1J>_$uO(p
z**<VTk>9(snCRFGkm6t*IM`i!9q!~*tnQBH-d|&Y-7l^W6qZXok6*yKgu8*AJW*_s
zh09wzDb1m=w$_krr+XFn5e|yzxjba#YqkfI8*imMbx@jI9E@yvgR2voiqw^1ZldQ3
z(q3ja8f-pNSSK(7!f9?^#whH#@j#)Q*u*kH%9cqm;<jQQ*Bb>v51Y)dm)pQq)4oLU
zEoR{8tFjS%Q~IWT2&5EmqTZdD$>a&CnT;r9gza0Flq%SN)T=RpnFyhrO6y*vBV$99
z+whQr#U0@M!db<dcR^X9iZkX##fgmR*J6e2`%7IWZ(Xmn?hf)g0XIJ;@qS=2Lx?`>
z6abuPFEQRs>v3MC>DP>TARQy0CgY|_SNORXSQjF#bQ+q~%fgzL4t1&SwW0(uWkXtO
z`6E=zxg>G_;}|6I2hv_P$g``w{}3LajCCZz%^JSDBzlQ~9ty_giuPl+nn?=fAF8M8
ztoU7X23Kb?8ga{-k#mKUJ!369==gAB{OtvuBqKXwtzoMIUPz|kU-8md9&;56S?<rG
zGgs+DKC3vS22MqzQ4u0^TEE!(g<T<O(Fq`m$GC8!r9YBk>}^F^Wr>xjO_C{;XdRo;
z6xn;SJZH|PM;NwuB8#1|5H|_(LN94r7+z3^2}7UJPi|E7@MK))L=g-|!9F!nO_arH
z7I>B{m+M#QHS~5;<stRJC({~4BuTLV8w)dwjiQh0s>q2Wf2?O$MN1v-w73j3%^G4G
zx0oOWN9Ilw%G&921PhY00M77nwu^ry(L#(c>KPVZ*>yMrOtzUxnm__f=hsCpfc9z6
zOXv}-C&=oZ6!aZEpH|5bOZO{&+9qcs+ZU?J=I1h%ceA6s^J<p9rdg$_n(y!;8TI|y
zJC7KAWE6wT6lp97<&kT+X-W(h@tHCKB8$G`t16f9wpu`3<ZpWDMUbklz2>!9e#Zth
zJge3DC4@iX13C)chu(^wj8@h~G&d;i%UmOR08^okm9YS`a2_D%rjN)s_$E7$@OiKA
zHlpViBAy2+Yxz?mtlrZf()}^^n)S@P&93h3A3YYy2131=5))xnL*GrG#K)NG`~Hst
zl?_#(YE)k6{TV+0`Nv&8ba=N3F-8$DNBvJMUcN!HBm{`8^TGc5#36K7$DI;cX2cGl
zM&791{KuGsp5g?%g?gdh1RYjdriiVAZ}S{#N`WJA*~)pdO~khceSvwpAH!*}UygHu
zmnimHIV=h1^aAki*r9B&&-CYVZ3y3!#VG&w2Kl!5=VROF%Ti5yb@~K)bz;E(Lr)6h
zuFO{g>1kaDuQ^?qdTxjbuD|pAghD*#nS4(7*5c^WXprVT=cj=zNkrwxipC^RZKAzq
zRtEUkVzv(LiNq?&FeSrioF(p~g`kRNFXM9Bg)N~jC0e3ks&`U4Ld+LGv1;Q)u}>nd
zvpGMqD}@6?6%D?d&Qz{?o>y`wLOi6buA8%nl$UB{8%Ww#Y7Oh9CAzr3E*@C}uh14b
z{qpXGjH_$nE(E9B5sC&Scru}o<H$din-?Ge@5A3cY!+Uyzay=E7fpbFQJ<{PC`R(M
zu!|3lA-kV{52Y~45WRoEGosX`)O39QpAta6VU%pBnE3GcpXTI$wK4zE<V1xh7}Q_n
zQ^!r46fv!w)<Mb=9|_>?4~L9ieVG0Xt#Xt|Q#RA-Z0)ADxBu5pfPY>AsG|aPj$HG+
z`qTXXZ}5SC|55O_`xpTu4V$i(U*U+@R*MaA@;z<i6Qr{Pd!2#PHW(!eG}estSEv9{
z`twl3TCe=L%f?^1@W173#HiQFc>l=Z-!X^({F#4Psc#eit_Xr1ysjq~p_ZG89?TE~
zfax*aEI2;|tIOVI`PL!SCPBKf@e&{=#edr(@p<bV4?BL!>Ax}riX$3ydIxD9eTRkv
z?51^%|MEa)sFib_xH>Vz|It$ZFCPIbHX5!!2*Xy#V2VMJ3GiQK&FJ_qK*lz3<0Hzz
zssR@A&LaA~hTE9?n*J^e{{7J_!@z^;VSdm3KRu?u%a?$zfaX>rDQcAQ?1<U2yd8!Q
zvM~F8*QS5GWB>f<?kb?Hz+`G)zoVT0|NKAyKmW!zkJ8gbmZB5>zb}*j@;v_aXw47h
zK~vyC`H=Y6zn)I>Erp-LO(*Kd2c>>AhQ}@BA>ecLe!b*uX<-8lm`SrJ1E!_Lq|4tX
z%YXg4|JSo=7NG~3y%S_i>Q5gu>dCb-I0C5=@U_=4{5HLVJ<W8R`LN77Fg<Qf(2yJ>
z`yS3!LZNNsb5LR<c(MS_0XzUKF~$kx{xr&n^1x;6y$`xLoZtnY!4)pKzN5?!Cf_My
zc7>lPr+SL8gW>WtDioqD_I<T<cMULOtAR#aBaGk80)4Crc>i1ts20s1dfx7>rUec%
zM-U*~X;GGnRzSR-ME#R{m>m`O?)FhXfB(<X#(%jEEpLExK3ye$!}D*jF(B5x&GdRI
zvXo-hqQaIzigJ}ZscoS0u}yG$_8p;gS7pF5_5yg6-n-Zp&`I}+gQUN8%kS!J8yMmT
z!w{{p<@9)xz?MFX`sUqGM_I73SSC*oIPqB*&Q@|~jyr@~y17dABiIj`F^E#ckHOnP
z;R{;t!Ca;F0w@6+M>pOH4GSGf{D8{kw=LhJ&$HXDL_`2#XvCclEb&{tr(DeQK*3Y#
zw5s&#ajQ24C^Xozs4urerCtDM6r=Ynz7cK@Z__zn>8T(|ShLRXOVdxR5tG~nJ^gHi
zXr&}1p$&dx|IfQMg#&fi6eLzIfB);}QqXa|7ntbyTl!HpBS-K7N9%<%Jo5&5>zel#
zU`pZZl!yTXj_8QxDVlaEYVk9ZD#{8pyDW9(sbl(*1-qr<FJt>GsQfmXtu6u%iQg2X
zR)RTBdoEYeoy79XZhDPenQg&;Zg7@VOGuYPSZdT_-vlszL1&RxaeCs!R-;!o_|uL-
zCqULInUt+PP>-RVCY&@`mLp%`^4NO~VYFP<-?*Iq+&s(1bNLA0YHdB$YPfi8!*@$p
z&z2bFTQ38r%!ub>$LtkQiBh1RBHz=Y!e(&;;4)urZ2E<twxRTG;}`zH1xP6g%{n4E
z@F1vVO?+Pazx+o(8a4THyO$&U9`y&ID4zZ)>p*33%_`kVQMxy*=25;C$&<J03(u!X
z&g~o3$F~g8D_(9cmcEVf#|4|u|Cg`kV`*vh<{gv7A4<efG}PBG7ED^tg*eCW>cS|$
zD|hVX$Ismag1yiXB03&4FzrRz09t_RMz0zIorR%Gw;6F7Xcz<hr&G3;kdMpb>N$XX
zzz6dn2UOMDiZb|u?Bur@6_kZjJP_L|y6f)V;&o^YQz+|Z3WP7)X2>Z2Z1qGTy>8lT
zO$DMuR80F!UCfTr^&{Ky)DZ0AkN~`pOn$<r!T;Nf`2EWEEzoa34i~>Ynvs{r{Nw92
zBR!h2tE~Y;X5r<t4C};ZpjxHL^gUe}$UtvK0ToPuLvQT>y;=M0+pp0mjS)47h-8%8
z(P0(Wa9QS0_s_`{nVD?#X-JZ9jsNv=!=%4qGDG063s*^gzkkpG9_H<Sze_=JP<v-`
z?tRygx(<+-h!o4$)-AI5+zS)(hEyVs#$RE?KUPncvxw;Y&Yh0INK5wW@}d<$7$~r>
z-&}D0Q9pLinW!ZN3{H9LuP)DMzq}N>3;ebwT`bA_K3Z0T1*(g{9D6-4s5c293%IsR
zGj~BH8w{q6Z1o4jdbF>it$aYiJ=-aD>6BvkX{-e``z=T@4flG_Y4H6Ky7VBSA80&X
zPJ-?Tyg6_u#W)bJ1uze`H2W!9QQ2QTa4(wVykxzGYKOQizE%x(l3oX={RNjX2{>~-
zQvkfM71(~MK_`T^^z)$43PnFt6-z&p9WOP&$AZo4KmjZGj5X6=>d2}GSn4QHUhrlI
ziC|GfzYEvucOcP1+C4!TZ=bjEyks4TZt-T!u;K)cxZ*vp*<kX?Pvs8R4*5ayACh&x
zz$VGMa;n;7<h`Q+D3omVypf$?T`^OZkxMVXl8$+tf+!7<E#+tbPSD7i`y>Gl!RMzI
zwrRC$Vku|u91U)qUjRVN&WR`Z#RcCjnnv5fhDjs0;zF8nX+7ms)oVY^qVKSt@4uDW
zxPS5Rf%>9Pq2>26T>pKtvz&YDB(PDIcfo9_=JABCRz$aK9xEI$O(bbo>m7U+Z_=E}
zs^Z!NPpmydY9bMUr8uGOw>yrnh(`UmqTE-^EpriQ#zmK$-(Q3Xe<X(NjDXXzqEumY
z2~_@Dd(<;89kzOpzC?RQ&jbq|vCZHwgBgCs?BNd=jJcnQnHg1CpDcTT^>ek6izW@^
zE<TO<7v~Owmz+jQ?xiLtheMyUGUyaerWBW-G}wJ`odOaJm`*i23gOfwXZpP|_;&2k
z?Q@<r`Yk5GueTnwuQmev!}hP-80_h9?8iq8av{h1KlOD-Ml&sv5G_DmxxB^oiqqy9
z&7#+-jLKH-G3#{oFA^KE4a>ek!s+DEP9rCus=;3#m94D08FW8R3+5R?m;AqcHDvi?
z`ZxFy{J{8qyaendLRy-><tl|f-)a!{24DdmUm5+jPHoBV!?h7!`r<o)Map?&t_~II
zBWc7wF~el@{KG=LwDp}7G#tW^g}Sq~z##JoSYpNW<^7vq#?*#YLD30lM2Do9ne2~2
zOY465O|8}fqALAuaCf2H<O1v9r^!nD>FPlS&<eC5bcUXbWaMw?C9@51<&LF6+r%am
zpg6mNqfVX|ZMeUJYy&7HUiFk)TJ@{SPLx;6FHCVVNP7Ko)9E7*AcAelPc~fD{~7UQ
z9dMc)iH#dCA~84;XS+JxI#P_<m`B^d{l8bXS+|z-vc1BU9$kSQ`+;Ow>kZ;-ZZBek
zE0MShPxB#)aQ9!?p{K_Wgb$^~0O|==9kIU0Y3Fx1W;F{M5d>h<Sh`>F+v}q@==N7|
z=r+q^06-*vaD$r$#iH?PY^I|UMR_;@aEH9wVRSF%kWUGU7klZF;X3?|(_b3y2z*8%
zAiPa%<vWFmfZ%LO&Tu|Q-n7jx<MH6#z%qcfw2U!4G&U5deuA-=b`mVIxsagY7GU*0
z$P7eChYQg*=P+o^CoO<^GMHrL%&<8a1#}ZCD3t&0&Vw*v;<ZKm7W{JzpZ5>rA+e+_
zVNqLkYdl$kqcE(q7mI+nW%Q#%x4bClRP&v~S8AIdTS9AR#1XU4x0jDczTXr3aF;Ix
z_j#7jp_8V^>!Zj=68V&9#>AKJGe150%!&s=9g9b({%H~ZN~!~2hX#EC+s5s+`&FLN
z>@PhDxlY^8m+02`T})AKDp0^X=>Txn``h^|^G$$(A|cB@|MGYXOxAo525cG6;!D-2
z%?ZnS8~~iaVk9ra^5Ril)=Xwuh{eQrq+6VC_1K+p<>y=mJvToCTU~9Ig}5iL$p;IX
zc^Z7NFNDx^v#cN-qC;25{8!qm^p;mYgNRx{=CFB*?$<2ZM=dkdKdG`8>{9#^gu&FN
zqVf(4HNiFZ#xP%c`$eB<DQpwpFR$E60Ug-{G$#qb>G-MXYVLcA%jjVkF_ZVE<;1|c
z2dOnl<=Y=8aT^Jr6#j}}e}}Ta|KpSv4G(fQo$_Akk6UgY4NX$p9>lR(@X_F-Xtzr8
zOs2GSRxjya{8|QKg|G+yF%tj9uJZA~2}p+SP<wBG;mJh!9Q<hYN5c^BhH4Ikg|&2Q
zVTu93Po<Jq_eenD@@@|=QP^l0T*;hV-V_&h+^j|gmHbBA$3#%{5GqOUtcJBS0`gH}
z6jk$C;SyL+pSebt@ba$wYt1%!?<b1vZGyxfHQDCgc&R<cgLFnx&Z9Sj1L=G17sn4(
z!@ERFVf_>d!qXm1+fEtCTk<n9>s6o?j(wWJJ=!9JL$feP)n+KK&%8^g*$*|D^1HgA
zWr*Vs#iIN2F*xQF*NZ|U=inQRaj`l6my5SY2tGPbn<(giNO4}9p>V*A2>Zn@51Esb
zI@D7C`N|GwE)|5o4kiNPl^7?5p9(XS4>}Tl#fN=)q7jGX#1yE_D0m6zJI>|!7p&;y
z;$d3yo*-^!upL%`(M#=5eyV^No1ZQ~X;>=)ympTdloPvzOkU7O7pmeug>usSBgt#N
z1eWWA9!UmPiyWh&DS5BSc#EqI*ciUz#yHFnuF(61cO7c_kaLubF~*Hl-sc!yUVT(B
zP=)b$5q}+6O19p|i6%gx3}m0~J|~tVu8n=7>`33u`2TVDo>5JHTib7FQdA&Dg3<y(
z=~WPsE+m2!kt$UJD1<Ij1f<u{M3k<829VyX^bR7$pddv+K<Nlbuf8k)viE+U{p@|t
z`Fh51I5Hq33Axw3)|}V;&1({Pp+<Uqg(G@e$2p&iZpKtDJvQDr-U}cO4RKfl^6vU|
ze<xLnCv%c=WYYq2-1`Bp8kJKwDg4p<`L?UM^B(6Qz%u2;F=&s!-3vvTB{E>B4CPz}
z*NTLLm-L%`ix6}o)usRX&&SF0xcKz}svMkAXmN8G3mD4O*!f`1$&J~p@0Eq%JZs>N
z)oEW5H2QKL>lzvE6n{Lpee&|N3(Rw6qfg*J9vdfTL|0w+leOyvTke|OnY9b`WS8Wn
z?MFm*>yES_S`fqEdt|3p+&<@M3H1*QhpowjL;Wjd<}Sqhx?fEzB7NPrr|y(Xpd0J&
z7Q_vDQ;~S9FIsc7jZ}UehnMYy*LsOQYDD|AV>&aOB(8>@fS4}^ZtK5isZZ$ryK8bQ
zuCSp;$sKImd^1lL?*0}2{o|hb=Y!XDu#(?OI_STj@j4MIa^r@0fP(l9jS3Y9RwE|_
zYesk|p8A2*t66`vKm?a<1eZRJ%L2zW$WzxZ_w^<$2z|K6yqaNC!Nd1pepJ+4LHvyp
z(F{8y{8;K$HP(h8$F6*s1fF3RwmC#Q(lrOH(kS(%C#Gwt;Thl`x?|~KKT4Z`ym0gI
z<M^~mAz5_Q8-3H%)bf~x?;oNPTE*pn$%*efpWqmHa$k&bdm~oyoV=qJ7EgTxPtErs
z?}v<~%}kyL(D~4jUv=EmHm;g0AB_(9!Zk84u3CDJdb;-mr4-v*i&ZRDd07@RZ<N`M
zwJ62nrJWLek_PT+yNw5%`8H4|=G}WCZywu;4<G&&mVu^;;Nn?7z-rb-Q1={N<Whgg
zmS9!gmVv>>QcW~!+`sG2b`MS=ngS6s3jTm&wK_d(H<N?aYIp^_*ODYZ&cCCEBX{pd
zhWlhtpUFT|L~)(zPU9Pwa8?rMa00fhlcE8!o8km3ofx)jj<2R-R#WZ>4DQ-b*Q0w~
z{gPzUrW}Wgj8>H!W&2jDmmWv1AKS>3Oxt!K@6<ZzOQ*|E=A>g2n|iAAhk;HGFYWcY
zF<E@Ynz-#?!5@GoEs)eKwf4>nr^-O@udGYL4m&{<H0%=|PBlU4UX$Zb24@}ql3=8l
zta3ets$}VM;a+}PeW>2*!)RFH-UM<E8~t|3#}<Uvu1$6p%UHi0=%$_IT1fqDGDPH8
zngzZ&&&_haetrZ|fP$smjC5qt@V;A{_ISA7N!WsQ;{FSb8E|s4DL-W~^ir-+2_<OT
zDo<`={fvy0^D79!-?Q9HwLW`PSisP|F=e-Hl>c6S|Gv$?);z~P6dv(+wxJ-*>HoSL
zUsCg}Lpw|VJcYs53vBEUzMbAd@tctz=L8kaIyPCp&+;QQ1f$I>76cxA;b^{L5Qd`{
zwWw?j)01m9^6Ixubd{Wp*n0|)s4@Ixe3F6q4MsCpk2}$1g%xjZRx)T_CSgZ3mHz;J
zD?2d+&)8`6x(mTz$0%}ScJ{u_L0Oozz43kK-wO^}%RT)Y)024a+B;>>*M&^&3F{`f
zR^?yrG>}i5dmY1A#0=RAbqn{vZ=fn9H57bYy|hwFg(l$7=@pJ6DTsT6yir7ZnVkS%
z`sp(QUS^b{Ry&HpA!L-a1$_19X~-?~=Ww2DW&#JPr_V#F_4o^Mw0T3L3mf_&F1H@K
zcXNcf((`9m860$D3CK^5*-d>Tdilo-6<Vn#cY>!JhIT)-i?ex7#X0y&Q~0lL-5r&4
ztx7#2C*4(@X>gUV7x2m{4q3d=1b=_c=(7T&l`+$|mE1B1@(I8+1{Lp9WsPY?&_k+b
zioJd@4WC)lKXQb+R3uBxuib!jA6OE2H?vg_3*INdWsnKQ+j7a|{y6IW+2{31CA#KT
z892ANUCrTzddXOMeDmCXMxgF`alNrY`xn=4zeuJSeS7@SzSc0u>i?30|DBg_8Ir1!
z*#8zT{`=~0NswmhRq%MGsS0K>u1+=iv0r@|{zWGBqRnG&kgYw_)UZX^w}Eb(x!jLM
zjR&?(Z;gK6%CI8M!VngEw0}L1hzH}C$tqN=ha7ijxqR2Ris3#*b8d^5Y)r%G5chWK
zU5f^b!@)eQVU=-PgeMs<!{Y9F9rbBmi(3qb0K=k)j5Ao=iRfZ-Q0h=!l2P$O5B9~{
zN_@bTmd1QZYm687&K>D;*KASqYS0xpmzsTpYr-^3w`@_$ch%BeBsMeqw#FilPK2H|
z6vTx^qaQw0m9lD|f3Xt-z?r_~f{`#kO+OpN(XuLux(|G{Q%B{+Wc<eWdyzA<xg@vj
z)ctPVtp^;X+s@wG{_M|bnRAyMtP4p8KsMU#Y2k8ykm`lwvx<sLx*|P+bI-FOmo<bJ
zeab27vSNB4YmYy-c<I`05*B@>D|xE-(DnhbQxU%3Kd(}0H|E3Qcf-qZ4r$}fEcjmT
zR)9*+jB3c%m&~Gqs{_WB`TB=ixf9XEcojZ<Xye302ZMNd{ZjSWn^^N@T-O_Ml+A+5
z$EHou`sryJf4cf!^Vzw?t(r=jlV)=5w6by3wl8n)t&QW%`)l_14B|}9?8blWs&Moh
z{Lq0|AEN3mezN8DH{Q4_>v{IWm7K?QhE2wwSqiIDD~rnIf35G*9nmg3RkqgohZhGO
zx8Q!~yQWg_1RaZp9xi|ac9jy}it6t9(x;BKi*M76lH$IP0&z34W(|pU=G*zb=egvd
z-T0#HrS^9ZaXVN=#?bx^R=)p~n}MJE5Ysaszo;RlKhrZgRsWbGc=BZf@@5PnNVVZR
zXG7Mbv-|}mfsr%<vc`)n%*3#~1d9?fP(Y0s8UkVM=zv(BcjZNrwA&JC2L;JH_ia$#
zVb4~9HsE@LX@(8;EMZL0!00oE5P!s3jvhG;AI*9z!ZUSpbL?4DHQ^XZ?oTBpC}pVU
ztXrkw-CIo+-hx+2kXwpjla&I6Z1^;2i8$@ORjfzvZKT>$MGX;DHCO1yB?UX*mw^r&
zx+*i~V5X?($1exJS-UmTwgyaXbWSTP2$TTrB$4`R^4(qyR7xuK6%dBP1OzQgwgy^|
zwC*18`DE*nYl<rEg`cS@QD@*jA_ftJJjF?U6eTOur|$v8U1wcS?V%NcBq+<JKx!be
z1gs6hrN%*r!h;E<w|sQ2wageV#6G);ucgS6b3dGzv=)FM|LU58x{O6T6(bGzjcWAH
zA06B!pI1{OayR^_CZJlxsh3yNhXC?M?k`hq=r{D@ONo_-#PZ|Fl#ITL>wa`=qy^XN
z8C~`F+Fk(8pXWY>+vt<$koEEbxy@~vCq-lK&}D2sX*04b!|<iQX8vXdlO-$TpX$^9
z_Wy!0P^eanXLBe2+<v00$#Dk14wj24)kv0F8Ie1NJ=%;k^c(j<uOCeqtGmG{<z2Nm
zOjaO~#PBkaX_ZUyC!FdpZUP~?YT~(YyM>Q$gxZ-T><tz@pv=*@wY{9E3l`naID~wS
zT?uGYa-1X$0pTMH#A#C#J^m{OjT6xo9DYt2hiEDY^vOAm2<)S6ILhuJ-o)>CTL=#H
zbmg`|HDS|o8oWHmPjEacpd1sHuCP~t4s`SNZ!L~bP&J1jpERw(@@C6g=-;J+3`noZ
zIB5#9_X^_^3GVn!EgTb;_f;wZjAt`i-oe<QOl5Q<Wdk+H)2Mu)3|9usJb%PD%K-0{
zO$-nniuYcDtIWV}KZD6{edk4PT}keb9tDR9meh1q@XxDzHt!!-mu&L=?Q@VpHdK9O
z571`LfsZ0@mj8n1H+Vpo5c)YIq^y5fXC1fzq7(7t{uTKb|GcCI*Zz4)Nfq#Xn~qv)
z-Tep)`cyc=L{RF!7`TMP1bq;&KUFT5XsM}?lK*9y@AH=s`==cHfBhdIr}}z;>zaPo
zUlA=aoOSglXiw^|yvY)XBu?lcg18yf&++I+zPf%Sbq7%!G-}6Vd4Kox|9Gr!Z~+t<
zmaw_LA8sdiix?e@cmU2Lj#b#A3;7-uc{FIF2OJz9u!vw3#5c?;cyM~f*J3RzDoXVR
z?n-GnG4oA~FNj9oxv+Xw%!uY&nX%w;Dx&&x$%8J9^$UWnmZPF-u;s1`#a0nM-(uj4
z+(vPEDb>ch9oSezgDDrJz|%pBh*2GG!yANH^A7ARG^+U{^B6Bqsdi|~J@En0aIJl2
z$<?!IZ7?>0<;CR>zn2k-#f!ba=L$NYB{d_L5@k)C5B<q#aqxFnQgHZ{dTN6V^hA^y
z5EMr@DlCyNiSo=rjq$-p*P6C9;B&!1d6wz(w8vdGkq!K2#rb9p+2KLl;W3<~Wa*d^
z;r(#P^In{3<ycH!bnzxC-?y4-N>gX~BHi$r$Zv+$0~Yq}niZeSh9)p|BKy>lD|a^~
zV8Y63^K6E9%Y6H^Un#@Pm0`Ds>YMm)r_hTM%?|YtA*)GL0!DeW5nn2nh_QrzX<I6b
z{qP$6s<O1rJ=1KBxFwen<j3Lq$>3i<(^q3|=iReLdNy?;A^Htmes~Z`cy>u!G#V_e
z)__bej&*J#`LXDvyf*J6nUqTEvB9Mym19Ej-x1>9(ZfFj3xz2z`UCgf&(eRbqlK5`
z0QsoKLciL&{{&^S^_b)YWObX{qNPP4T^$vJK#JNc;1FPW(k6sJ3p$urY&8?`7H&E6
zPnz*g=*Uogy^^GcgjKJO$MED1w0<QDS8>){$&3;5lvP^7C{Bj)xhq$Nl4R&J9%K3)
z!-(sQ<%X#DAhKEoM9?V38}Ca_VG#Pd3fHCw$Ygsm>^$(dSaqLZAXTvc4T9#Uo=`4_
zz!zCc?mvnC4^_zoyi_NcT_g$=u>=B=CqDJVPZ`jq?0m96A49C0?vyymgPldLX|GT}
zDU6sHzWPAkFx0a?-8Ku5Ak$UXM3yc}j2NS-tCr3{zVbtTH3&F{!6kkxmHsZ5s%boL
zE&KY2;H=H!J+Z<=bJ$C?%s41)NFNW;H7t1ZQ$v#5D(4nK$%WZE#pZ2IpTu~aUc@)t
zcKD$*?6k`&a8rgaK9TMLK9KV#Z?z|IRlvnYz5{gUmgGT7^!~W>Rn?`Wha;v0EuOj$
zH+uDRLhh$=zMf^J=di0<f9e4)MhBqAc!YqvaO~qq+dp(K;)bqppz}wTP}m=1JB$Bs
zvZ_xc@!YnP8DjoRRt<(Atf1%F74`t;^ueo7*$4?iN|bE9-T)T-JkkUGpbD|t%Av*&
zCmMgz61wP_=qV_mpTGu&;72<Eq(V)m^&7|=S%{7BW!OL*1C|jX2<7hw8ve-}$tn7F
zTa>|v6VBQRAp2;!M{Vw7Ls06RkH+Ajt|cgFk77}p9>tLfHdp}5O4igC@lWcr+t4f?
zGb0<N#-f|M`~FYYgJM>ql&WHQN!erBbf$9Bsr+%rWzgB2Fz46lsEygS>nvp<8^kPM
zHpMavr;}*vo*ctHyA3>0;(S)Dk4mowL%=VKs}p%lJWDyM5?R%%G$p6Wo6@j%M_uCW
zBXZ1fKZ*9sdyCoW2`K|LEiHpowd?2aW!#MZb)d?~P!k(0iyAftGT{s`FhvOVWk<5(
z6uU;~WwD@y@stQqhF$OWFqE1bDK?Xjd44OnXXV1V4*J6i@h)eeux+aN?ayUP7%;7=
z#C6y21#Z*>h1W!+u3u{TyOZ!L0#wcV&qK9+>jea>cBx4$1xa*r%eb!5hhoa9km1_N
zl^XVGr6I!5^j{Rx>{48ow^8i{rVj;8O8M0o7k55LY;qV!l;!Im#5!p5kN<fH&6ue<
zth%uCACzu=cKnjy{;9){3UV>V+h$1fjv4fXZ>q>^_eCE^M4Qk?_|~Ylcp`cQEVVK-
zhsq%nldxx3&1$TLVCzMouIuZN!2RWl{Q5j@VPhkKPsR+*akwWstfRJOk{9E-1q8n?
zFav^c%<jW_{n7n`il<dz^8zC(m%JmH|4fl1`*z1Y2UsK@zAipQom_s-w+xpEG&MF_
zuIat<VJ{`hsh_bu&!(>{v?TrL31oiL&Nwf7$Sv;irD22i@O?KhaXsOAwY|oDQ+#wn
zW4--ZURxQX(s1OQ@=TuDe>w6R)i9*BR=<_A|4xf=3c^T?V!=A7W0S+!$koOfS2!J#
z9`*1Bg;NBEPX$t+4GeDxqz(#1t8TjN|4e>%Dd{y|5<R~?Sotj|sup}P1&o41O_W7o
z%-8^^c0oQhXW=&l{GOE1J#Rt2rhpYuL}JbhJl7Iw!H5vTzL+Dtf_wyMcc2nym?Z=Z
z#9Kswq8hjKrVwr_n8cIo8WsKGwiGWBGBA(6I?N;O2&D|7F6((}ct3$yv@IF6dPxx(
z3>Waih(J|EVq?W#%dfnI@Eha}VHTr;sJ{kMm)EBu$L<WjKJ}Og&SGNST8V=?)ma!;
zIG>CFNV^^$>B=bJr6FP-T5jsnclCG=AbZdivdZ%6ZfA3F;O)%9S4AH7lZTjtO>@%^
zzAs{6JB86&c=cX(_bcoIBkbkg$YJ#F!x*1}UD4XP6uVa7Ey)CM2{(kM+6`uP%tZko
zyOOvMjxJHL)J3htYA@)648vIH7Iy@{Zb7Aw+6GbUA0?r&rxSjbx%Lm<3I(P>L(3|Z
zARXAi1l}&DgUbw{AT7vs3RugYJdDL-y!2Yt{sX!5hK}_cofv2->9ILm5=1b`zt4gB
zWdOIyT_c+;jH3K}oL^&jCaF*8^;NOb!d-^`r~tI)rl$pOUJ$ijL!&*b%!@|R>Wh{z
z(Y@n`mi;LrUCSl9U#9`Rv@O~<nO;>&IY+rEEn2B^PfqyaUDy*TvpLVQQdg(gCa!yr
zm|Odk^G9f1rh=nd$l*U-*2Ny)3UNtbk^moxxdoXV3Nk8*UZ6`mUfW~p0X5Bw>M|F=
zJuS3$dNx2d1t+Noi>S!AI?%FJAz>~&FirWGw|4DK)S-npqc7z5z^f9FsgOI6Fh5Ac
zJlQ)wVsRMK{1}EjV%QefDdAp;yG}GDJ_Wl@dGt0sF$gWU<wo1oAr(AzzgNJ`QAAWb
ztZPHQ!!S%Rwfa!%=d~8^B-^}P@3e9@F?n?l7GP{#^$@tRgaKdNQYG0!Hq1#9j&OWt
znbx-h*>MUsUyCq#5U<w?Brs<Ed!7&0EZtdujG}?ART5g+BRVmaH|5~=^sRfvuR4IG
z?tiuNxl)=w;5;d2D6f8JzVYSbJCXKZrYEB2fB9Pa@qE)zStHr$qWp;{Tg}^YOzgi_
z_tc^e7#KYrP=gJff(;(lg;4)&u$>ewb)LK?Y3=9His<y3QZHflsBZ22B&`OWCFE&<
z`1PF5am2reB!B;8%amB7p8(P1p8@QF2nh;8#V*i#rvxi-LU{fYDfO)e{mXeH4NY<e
zXUoS>gnn&J=nX_4xwM=@xI_8b<CIJuK()cGo$FeMP<qvTjB6yyrP|Va-+CzZ(qvs1
z?)*;b@vZuJTdw4hnJfiKD=4t^2%#SaUWktQfWy<<7>XxjV%Gz+vQBSJAzykjzl+`U
zK0n7Pk%g1+Kx>@LgIb{r38r=e-{<K!qHNOUNn_Cs?xIiCf=jw33wIkGz-?8yHsVMB
zYMn5Fmpc_%FgqJ?C%W|Joj?&9(Tjlt1sB9724ueucpxwRAiHkC7AzQb*+Jn>__vV2
zk<HIKnr_Y)4{<-}#fCR`K?|J?N!v>KK!4%Ani#rO_^41lD9*MZ4ZN5=XjjoH!Blb1
z8UrJ?`GqFm5|(&ou2Dmvtb=kWjS5N*CIM&)Et7b6he&U{Ed`X+GQ@*Yl=)cULWn=j
zG<mDR7~QP*v?qL(&=CZOih`;Sr7!EiJYpL&i`$VH!)DNuw(24VTIZZ<7rC6|tDFeq
zuMpmX5pTxH6{!w}iUKjo?;<dCO&u&961;o-sQXkpQ1R@*JC8uK$?o;kr{>pir1MFZ
zwDRP}?U7qz4DC-x^!d<ZV1A^^D$i0iLz5bFNxHs|yd2|CP!i8OqzrD|yNjwD5%-<I
z4|^tVx>V@N$4Y`zMjO-+kV};+`Ild|&6oDG-|V6+KUG>KqP($)3x`f~JsFS}j;6#r
z&B^YX&yR>V6{PN+Ca*uTOcknKSD)k;D0^8gUm1UujDV11<hbzV#}}<J?YO3;pH~Zb
ziohv)A68hOUOAR{3GzL4HrYDOnR(tc_Fdm0xg96c;{?M#s<xA?%;tID12`{~h4DFB
znt?`VOot1Ly%lSRFh^0Buof2K#~yov34DBU*kxF87&a+vz%4O?2^wT_g3=w!>@te7
z6^@awV%>c2g#F```g_g)YhZpE@F-KP#@p(D2*4nFQ`l3uf3BSbwT@J!X%!&9iSwXu
zrq|12djn}j;*W^+Y9XdsUnKomHklf{NbWg(HLWFt4s?Z`{rjwlB0bV4OT{$)=d9P5
zsV2vP);(0W4+^Rzplq6u@=^Xfl6Gbkm%4!X%|n%j%SjMKS8y1cagL*uQSC`I_?%1j
zJ!vn2$(-7ZfCs^fhVb!ulIT#2vK5l;N}9LieA_idb#72?c)fo9I`q{c)b*jfgW{_W
z`CtU<WSWKI5c&M}!CuTd1;OOp{E!!Im63)?ge>9IfojFg6=z2n`{2~)#;r5<4GC1m
z`pkK$A7~(@8M+v?T<d!e)^5n$sF3YETZ+J^0BW#ae0Xu}NcfAG<9aB6>*?U4DC*7g
z0XJpwvIc}xOA0vLLV)Lc`I^DBBA$^gz?Cm-#(Jua+mInqEBwO2!PlUf^sHIWtC@>h
z?hTGSWc4NWixmdhe8zEibw<vy)v=?OZ+;_!<g8X<QnrY-(2Ut=|3U5%i(rrD$C&SQ
zDI0ggl`@(<Un+g?dKJj6MFI`CC?An<IsaQv)7;l`ZEvDIZa?O?m2BTYjMtL=cc!tZ
z*9p!KqMlD>`R&PgU-2ofc>3!V{)k~<q3CJ>^j@l|F%YBOK;++PG+$2IyyAwN_Sju1
zSicvc@mRlPt$$LjB>W~QZ=F*#p;_bE<C^0Q!Fm<k*a~`_q?_h<;>majm1QD0G*(El
zp+`{po19bJtN)ST@~7HjCGM%IrMjB`EORN4)3D#sfEXwY<thGEYw(f@uqAC~n|gTl
z&VQ~xD3EvRfLD8(O;+squWzZ#1s3iDAt}B;I1dqD`e%@bM;ENOwg2-=y<P)PDp`xl
z@<RR#9P*Jg02IF^zCNDvNwZxC>DpON<);E!U8X>uX?uMm$SB9O`mggtVc@cY*Jz}0
z<H)Dkr%YFSJ{iy<<i0Y0ZSwp5hZ*`UuH?*lcLl+}sH^`?_o5HjNCJ7%?8dkJ<$o1H
z<>_mP;3$|g!~nv2W_d|4fGd=mE2;T6D=5~PgUZ_Rw1q?XpLehV0_hKV2vjjDu#MU7
zfvUgvswHb?FG!637!gBcyAVsIwg38O{u>PA6GqI}tObT({}Sm2z^MQ;#AyDt##zX_
z`))aCqD;h-%njI>{q?mTkOlw~91buHMVhtt%8xjmP+|jAn=|YGB?|r9U%qgWDED2Q
zeXak`z31O|qyO>~1tstT_Ed>N@PB*l|NhQ@{Wc)&27n6xJFfTNKbklSyg=x~MT>v=
zy8gd>mj9oA1p$*czX5(}my7O#$m{B>rmL&L&0z@N3vwWL6#-K#g2TDfYF^~K8Ig~v
zDKx%q@YXxr)A$AgTWX?gPXzXq0A$$ZbYZFCKkG;SRNtvoAW39+(|>;XJkOf}Abc(j
zINfW<6TLE%fPa_T;TTpv06-Fg-T;T524Rfj@3yUAq07X$m2z%`IEPo{le(DOV)<cD
z8-1ZHka4nps)M`D9Eio{K;;o&UCw7F_P^}`GR1JXA%KiB^hVIBc-H*K*>&E1#Hm8W
z;UOUxEOT=Uw=#6!d(;HT%B(k&JrHOq0q!Ef;an+LnM;E%wcbQN9LPicT^6gAUw)Q_
zzUQhvQ^|JeU-q8=aZ`RI1t5*Ff41>IVVl=tq~0Yb$7_BFIRI9z6O<7l&a<5%Ku%~Z
zA9rn3Af*y&0vpFVXd8wb<mW9>lY^GM@O2Q=g(Tkd;$`<(TBJa%1D+!UFqH*e;@e;d
zAuT{DbCk7E-`bmqfZ-W%katy-nx4H5runvlktVu%C!o=b28<{4k=*+hq}IhWx*%k>
z3Ba9D5Os<X5xL2)R)-tlr9y$tOn*iJAb0MX17*tDoY;)TjH?<gHfxd-KH~%Sk~R`b
z7J9(<*;(&{hO{r9?B9vzpUk3Rj!g**Ik+cN095c-Ie+;5*fM|uP6asn^uvB&x}P^b
zZs<e?kaL4wy}j&MPUYbz#vi_HUw(<b7|Qow?)imM;#O0tlo9&ZZVApK3R0)~jhlY}
zdiNg@(2OC(zK;mq;SMtC&0evv--k`VT}V)j@<uUsVgX=7YvHs<lA=BiemX7pZ8{7=
zOEhMUu_2<d2zMYRJp_Pg2vFSz$FJItl!yQtZ#PMzDxU)dIU1ij(a-a3$3X*ld{GJZ
zZ(S3MZ+yLK)tmGjQj+XTw1%7pBEcC=x&6%39bBMfDl&Tih3A$o@=GH?{zytkfTiol
zGb#661^^7~WDQEX#JMFGWW8MX00?Ue88wzt{46!M1JcAf2ktU%py8uvFNm-U0FNQ_
zm61}S)#J~ri>}0GHbw{=sOn6=4x(Kud$YhVL3q%-#s1G5=zo5P0gfEtNUg?vQToU7
z>%2yqX!7XR2oP&8Lsmiq5La#EPD8+8m2O9mTDH;Kq*RNOfP^X;kUI<pm6)Q`7{1J(
zM6xNv@xiA+Ms4TQ@s~)x0sny&=#)`T-j^w<wliYE(`{kb1R*QDjc*CQ><8nZAJ+6z
z1AweOX}%L%#@H%%z%qA(UeuzvHf~=|9QCw&;Ov-529WYl<-02yqzcH%=kQ~o)Y1Y5
z)oTxsQcPAasmlUSLm&!}Tc{hl4iYiJYH?kFIT4fD8L?zb-UHrDw3`F+ew-?`t79Ox
zzVsC|N|*!k?ljm$9#vOnTpL4Gw+sgT<4^zB$mP$+wNc=}Gz~SPJNNgte<7V(1>o5+
zjR!)cztjIlxou*MMs$)?AbV{p;(V_Tr~A_?hp_T02ydW@go1e^SM9+G4}8(KaSMT&
zrSeNT8Ao41Mjd-Qnz5-tYh@c0!NQ1ev3Uo>-7V3+o646-*=>rU-p_x&tIpJ1yVW_&
z?11Oq&@uk^kNA(Dj46S`YUiO|;a@SvM|V=dmiv(*Jg1(l$;J#SxiSsbX#n<?t-%^}
z8aO;*_Gktp+F>UryG>7|aqVbnEE%CD0bKj3?84K66?ejRo@uYAKAz-dO$QCcq`!GS
zhuoP1O|psE6U^e+JALQVKW~EtN{KfhygfNS3<GjG!SG|6H^<W1MGu{1B9{&HofLPq
zrgYzf+-#IOrAmf!)4}#~BJI8~uhlK31F*e#oxhCd?gJ;sWIB=4{>%wE0%~v@s4*_j
z9)Ok(@*QG3isnW2CG<g;vd`cZeg41MYi}9sA(6eF|KrR-!woMz8x}SD7S$a@T20*f
zR?Z<$nOo;RV~Ugeahewv4o7cs&<rqLxv%1XabA1s571bKVj8TewBpWP<5K>W=+bWZ
zm_n_AuN}i4^kO*aAxI0p>I7Suy}wt&&`CmD*#k}q#67@%H>IBsBo3=h0+|b$Wht1X
zTf3?wOhG#`go0`4=J(Y06MPSDa9mz&4j^j-@ar<KO!D1|kVcMkt#d>@KX2Q8o~2J>
z2{cd>hw&)x+zDF+!yjK;nbnX;a~-hxhU}9(?sNf^I4#(%G&}ZV{XT*tc46~X?MT}F
z=eNF2P1hw^qH<f!G(<bj`|`V7M;~$WF`oPa)$ZQA%ZC+~pukn^QDjQV423)wN|e)C
zhmG(PW|Npl^nN`75NXk3cFk{aD?Lj09|(qSUY;1JS6jrrlZXdY_@K`j<XhdO$ghdQ
zxR(ZW+P7RdR_`7-=t_Rk@z!V*j@}8{z8Ff?#<$Pj874|p0?BL)_!*AmmaoT>mQs$C
zglY*B<_+2gU#aX`A3JzCVNNiXQD49*vFwtvow6O|8QFxyn;Lz&<rbN)5%R>~$vWen
z^obTG@OQ^OzHQNo^xqEpD|*4j@2j&WeYCH`3O4}A>|y}A@%Se@C7-ffAH2R%q+KY!
zc{{c^Un@PiA>FyF`o7wZspymXF^MXJ?}v}dCA12Cq=f$SxeI7$VrD=Uk*D-mm|L6F
z43Yx)ZJ7;J%p{<e+UdQq^Gbq98aaO_pl6!L>*?FreW14_jyjlrD*f{z$;CjXc2g>-
z&A`VuN(~xRvc&EqSM2FUX{W}0EBU7nA4UpRt#3twA$|USH8;CaQwxlU)mxk4BA1Nm
zS-*lAYsB6Aa_m{8FzPyP3g@q+s?<(7`UqUPFj33S8Uy+_)Pc7KfBXm`L3IJ!FeQ_2
zm8rKWwt4GdK5IXPT&^67i^_}GWW>nx(F#Rp@*&JkTNQaHC!RD(?wsKw=zOEbv0|lJ
za!;3X2!Q7HCNE~dGx6an_tyP@d{S2fYUp&WV(sUN_SCJ!@0fA_!P;-oT(N4QhKj;@
z;dbH#0ky1M1CIP7`l@O`yN4E5jqyDRck%;5jyZUd$L4eeB>~1x4MP-pfTYcEYKw5G
zpblugExr2vxL1ET&y?dx5{!~P-v9hzvjjHMTA_cVzKjSr%f5g+gLeGgxWuiq(pJ7L
zHej!$1x2aU$P4^KU7KJ6CR4RWr`YJFR_XTAuaOUb02jMQT4yZab1LU5b@R&8EGuGm
zIIpz5Rf)$TnuUlky_2*h4i_iEv1IN`^eAWOaGJQTTz*V+HfeM6J2ak8m+hBu5A8cz
zy4&>Vus>b%aF2ITdFC!XmFUf*)-xQJT*yk*1p@>(lioe$fv0!K&s}h17cv4DoXG31
zIcUg*fs{Y_1JIy!MAP}>P2ESc@-W$h+h`U$6a?qgJRnNB1L8<Vx@jk#5`XT-$3bT$
zsEFfSDO>K^3NOP?eiqgJd=$%OzZa%4v!=544k*BssMOKG99z&`<(}T^>@9F|@r`pM
zdsVRPPX7fCh#Y9d`;{2KdX4YZGIitI0<z!ZHWi2h9kNqr9?B1l>EseetO{Ydv+T11
zO)-ai7rigfzOQfB#|{ujtemjLdh!|ieYi?RQA$0>RptL1?dk!@{=<W`r3c3sMJ<BI
z$A~TRU^z>geeh(nJ0xQKe!|nK!G0AKb?r?vYr~)jaJ!>5H@Xkk5V3xBalh~Aihe<1
zJ>&EjzIb3<BTe*^Xw=u=^z3WYxu5<eYM^)04|K*Qp!Qk12G)H)JW-L&VqN>tH-eIN
z7`pne3Gn}X{E>zj3%|*J^4E#{`Xv=RJE&VY$V4!<R=yKFcTtfFIl+E3tt5lY00q!t
z`TFXNWbfdHqhBA|Kx{S>TTYK*M=I3cs1OPkqEf>BMBOh`CDhUXgvv$)zJ_Xj2KA*i
zJWsnC)matOu-A|`-#}RduA-MjgV+v#=mM7Xli%;ypk#q^?Ru!EH`Eoxi--;v>M|-m
zR7K#aRYU``cNZW5=Jz1o%8&57T6D-7GFr)I3ZGlSU0n~5i_Ihf)GYMl9ID8fGm4Pl
zv#8$L4AIzFKD3DvPgl~paOzW0ycQHlz6NmNCpG`#vlZ_gWZlyiM>Z=22k+tLKhk!7
z1JV07&qHr<W%f{rFoU){5^Et7UZ7-F{zKF6K!>;W9G=Ml$<)i($$EpCaEK#UF@}DN
zVv`*Uc}V2!#R5eoRCT42kP&KGRCV~Kt#J1{+QvE3h<CtD;26182hOWj&`mLiR3PWA
zpc}VPr}r{12aU6(k=o`r$=?x~>_p;lWw}#}_u93#l05@>VmE$B;YB!Zw^T*aLy!xt
zj`WW)k*GK&Ne~Z<FzRyLzKlQTa_%rY2GltpO+Tzc*gpm*Tona}5b|d5jeSK|?zt(@
zu=yQ1e&o5cR<|aXK0iV}BRzL<wzFa6<al>#%>`yw2BgNiJsvZ<*ejr>8cdv%TqjH4
z5qmv!0r&1{A*b$rwvWRa5~J<;><5U=bK}#D^3ivyv^mNy;`@pOqNyOIckNG^^SJk!
zWHj_Ttk9Sv<SuH#@q(zv{nmq!XX{}4pR=ETQ`;%1x9D#iB&?}5hDbp@h&3~htfA&W
z%qWn*!~sv7xzm;k=bHh%D<M$Ye(q;=zO&}81h@ko&0dV9{&uK9JXx(wJhbWDww6CM
zM9T*s&$R8`&Hd&??=|4y``k^sP`>`fH@o%yg3aDy-LjK}>;1Bs>{gCMGY=sQ0M&lT
zc=6)p`#B44wns*n(sGK-(m`>z-y3x>b=7`AoGxL5IV7|myqC!qthZx7SIp(DwmMs_
zReBfN$aQ#RRkfkld-enpSjptqb=1RnWOI_rZThK48l9J6n4bJ|p7D)HRV<A}1+hza
zHb!2KHE)0X>F$r44tOgvxBnuRDUe4XF$f;Le}t4?93+P&Od2YKZ+>tA2)z|4Z&9%z
z%-YZB)+|?{9O=OT_MUmHMHB6i(dSfbpg(pd7mL`Gel>s)X;G}mu_d$z5|AhDAR!hq
zsz5aLw)+$0NR<yg%-mm6JL7%6$G`>oQr6j1YO+`A!_oNgO?>!kd^in0>z$zVfS7tO
zy$1aqDaXLqSj7-5GHy^*VkO8$+*w~7YB=f)%Rs@5Z^9DdhT70O&J$hDDB0K(QLY`W
zhLw{FOMD0xv5*nYhNl)BH2+;mrAlb`At1%jFvhr4%dq*RB#xBVyhatTuz@XQb~$wJ
zaIFvHMtwp8+IPeW8lw&StT5V}Fp@dC$0OBr-zUu?$>AEh7X+n@E3k@L%j)MD-i0<k
zP$RgbC*aVY@#l!T6_?HN5{r6u6jbB0p8FWNSwHi5i~g?;@_Yi4_a@A2&*O^R=oQDW
zHCX&*gi~6qe0oDwyn`C=ijT;fxPf%<-^p~nS}-$Dd5!Z1o6ZxW%61Y9e%u<U@5kaK
zn2rw`nLp%O@wi&Emh}6lecMpQgkVhqs+T~XG5V>|Ay;`8G^JnAc*v@8baBIJw3Vfe
zTdXLG*WDyH3CEkl{%$bWQ96_^qYs{1UL;|__o!)7yFeLs?<Q;{Zo-<kJA=9zA70$^
zG@Er!w(35VJOkaO9;OCLm(g5fE`C<{k?xcP;of2}C81)agxbYN?uveH()!hz@|@x{
z3yV=~>x7q3p@+uZ!<2~Vk{}QsI%85>B{W3lBpek7x(!mW?}QK!@P}CZA<nOtarhD}
zOBuHLLvz0|{6R+6>{G5#p@|s8Rnewh3i<e9G%ULAI9a5rKnRw0;0&wS_lioLJMA)#
z@7jM79`3xE&HwY-mAB|U@Lt$;_5uMu{Gk|Z`hW@RM}JQ6iiNnSDdq3@;zr!#2)wsd
zqeJwV#4h7;<5!%<_wnNY!nBb<v5X{x(>ebT1h*cM;&SNM*a6TZjF^GC9&N<<*{cnB
zupu^~Q-(B6c>}Kny+^7)m+KYZ;-(Ka@X}{o;*iS8HUOOx&;UGIH{a_zq72Gl5L`)X
zAvpF_vK@(u|Ec`(0igh&MJ+%Ll?<1JzA>+AW_fcOAF686>spBBd?b3>Qd<L}Vg|NY
zRa8xqFqI+wMdZ`>jFI-=6{#cHFKo`dU8$r(o}y(j(EI*{mV|vq43*@$VOV8qQ^yLH
zW1n|{G6bigUyeqJMkS)JsPzdS4^HvEO~Sv0Fluoqp0dZMk}8a0V~158J~)Xe$ABUA
zIPBYDnVH6)Ng4Y)`eSSAknSU^6Ju13YIWjMLzE~27D=lwl^`L|niLrVtsml}vKJV$
zt`MJN5MEXzRfz01(t4%b1r6mq>1RYu;CP_MMitFRi-U5$z1mdDk$e}Rr{^QHRG-3E
zuNnpG$ae@MoxFDeePvqvRa@?mMM+#=)8?j0Tw!&-kH*gE=&IF0%aQVeBb3#ru$(Gp
zgE&F7BJRV9G6n8%NpMe`ypw9_h5w6MXtX|GJ-0!Rn_tZPw0SyGgG6oCSC*;GPDbQY
zD_Nq{kW!K(mr9FR(qqE}riQW054xByenNU{jJ6q%Odo#-gyt{Nit}r+Cqyj88N{92
z@;oKjG`$dibiK~`A7YzSDW$9Stmt~-YZkCkYdhorTx3unAG@W%>%Y1j@ORm5>n3R?
z+oe9CT-jUBR;2kDsH==AZ?2$VPjvR%%avw?kT~B=vr7yDHe0p~k?wO2oQ0S|i&jYS
zi_dwPXMPqWJp@x>cr+QXfiZbD)VX2eL0sOaxtMs>n3Y+%q^NexzkX7EB)CX+EBuy`
z%-C()TZ6&R@z2i}X1-7liA&EczT_|i;>;A<$9q4n?@UkVSU=Z~5f9xbnwnB{Mx4(A
zu~SvH4p*TVuepZD4|SEhDk**59LF3sFpUEJggHq+)<bysz$d5Yq2lR|37=bK<kLwd
zjp=f2HAM79DIV@f!P31_<P1{2y|zthS&>~W;a_s%GDR#J+t4lE<7aqsW4EFK?p#WC
zI`aHHulX$<8`65!>O;{;*F|0-4FR3a+B^o=M}guWqr{Gzuj%rST(XhAh3;*#&61wO
zR#x4QdEuULT~Clgr19D3%uKibw~hV7u~oOo99u8%=8t|($~<$~L2@y$oHzfI&*GP0
z)t~YnR$1?ALAyESpv7}|P4@XY9^U~Z#C@GQeoTdC2uK6%I>*ngDzrKDBzfO&aE~hT
zX%1FBKQ-L`U{r^-%5ciSOr<o<tjmB~oqeIo0eVwbLuQ#|F-fC-HMcSs^ptW1{@N<u
z*PUQx!&0tSwuR9N8*}rS-d!~Sd!FqowhED1(IsJ49XFhrHh&!W_-yA{KgFd7!@a5<
zsF%LNxgHwcb8D9~PfLhaxKvWP2DtSj+uUj<@YlV(Twxy*Qrr5~>*sQRe4_Ri_8lVk
zZ{*6J?RD-<+ZqHiualI?Mxy7_rq_NI#rih))g#xhR*nt!UXRSYAkdl-!<##82uo_?
z>fFAX;(4DC@*z)XCKb<<Za6G5Zqr#6>1*cL&h=~c%YShW@TaOEa(t`Ee6PO)?`U!X
z?f6K;y`=syW3?Cah*Ib|jv9r`2aAV%G~5wbIH@SP&mWY%xIXHJN${N&&{v<6FV)Tm
zCi<x<U@dr=r5minFDI-o6I16hyM0Uydl#Ph-FlZc`+=Lc_<Y+Qs1YnEHUr_6<Bv=j
zWP-vI@<dPxNH^rc_i&uH48a08%6w|byD!LJk~(twoC_$K08ID>iYnj^lyIS|JmkbU
zt{M|dCe0wZaYpR}vjC}SiOQ`}TW^6#I0VVFsE&zcBGA_1e_F{6b3#z$mb{!gn^H&m
zFeRbd1-y`CCH5)n&-ZL4I0KS5_d&DH2+hu@VQG%4c}BDZA8hRV+G5f56enfhCNUEB
zR*~Q*2{Yn;U_N#Gb1Gk``~&n0+_NNRoUbvj&n*a+ux(Oc65a)@N~9TS<|+JpsmaR{
zp1#A=OBpk6l8Q2Il5?0<UE+9?Wkd@v6eE(ZM3MHFXq9gBG=OJZ2vz8q(<78!OT2x!
z$g*ws%HsEO(sdKI^4g}jc1-jhgE^}mr5wNT1u!TWs6q!%D3|YoBFb??@Ny2Bs32Y}
zvT@j*gk4cVts}Me&IaSgYqW%`=vROhnQWxEmc)t*wtmT+uUbyrZ$9t?Io<w~0U$s^
z1x$TLXCpkad3vl9uJ*b!Rj#`KvEl9S*!^Ap`$shRuMfu1Kx}l6)ob`$Y~&*KHqnfY
z5vVLu&LNibWyj*OZn$nlWL0fbsl_9?83+YnF$@3EA-_Y%EKz{tQ_7e@s=2T%>=ILR
zMqk%q+~q`*!D8i#=Yk16f-JxXj>HP}gi{+q$infdd}hHfaTEc-Ca4W;UvshhgU4UY
z{2G)?6)#!w27a23Xd*UTli`z$NEIAW-oXapF<VpzxO=z@&^Z(Y<$dFGi#^8})Zwk8
zTE;8cA@lLUOSQOC?U&<Fk`gH3ZdYg^;P7-pLvZT>7B-mU>M=i@7dkTww3UUohxljf
zf0u5T2D0Q6Nub0?p~QNt8tUe-RcGU?q+t-`WC#gCcv%n^MDOu7a1YJ_!!PUi><K|C
z&M%W_u|d@~-fy4x=3myRqPL0d_9M1uM^-=|TP*J>tC=o0C7hErkg{TAc~qY^m{Y@J
zW(~HLqs>pgoX7smTKfX2KD7;P!d04|JAn?{`Q-+3#_F^;)Q0xrm3DLz?lqL;N*Mid
z={?ZHJ}M+*_n5pHRB8vNV#`)PI3WN)*C@0uht8ZZ%KBpwe703<n8&5v%Iqtk`!Mz4
zkhWF6GhWV04Z3z7z29qJH{#e{HNO4-Dqvy`aGzt;pZG@_-vKL$Ss>sSQIQy!;DmVx
zr2$_fR+x@J?$0n#?X?bkLAE6fq?+~f`6iqYZW+`Ys)wl#Nm#xV$R#6GBUbCJ?L9}E
zHGI};R6<!zn;TcKi$1(h#HgoM5?CCe)lOdk+T<@R;nNHeC)CJc0M?e1-D2e(D70@h
z`eqN?&k~23a4Zyde4T4>`Qaq_IRX-P6ISUe*+l}Q=u{JK_&jc9BPMO)0w{hDX@NY=
zFqeHY5ef5~;)J^=y5i=yiPD}lizekl6ZLyM+EhI(?_`Yf(ONMzBp)=Xf~{ocBG<32
zDg)h)BhLJv#=520){MOQ>srwvXw8U(d)JLVm*(xh%=xsM=$TTTpLgx5M=QBL0r?PZ
zU#BnH617>n^|DeZa^fn8?t5;hs0@!g@yhbwvt6|G`T319MGy!KFL$Tf)%x_OcL6%9
zPryf2=_>D7sR^@bm@qJxuC5&0XKz&J<n8M)?4|=oW8T4HDPsOQw$;A6;fP{SA9fiH
zdpW0hwcbf<a)wxy6Z~N>KJ6kNkX;<kB^L=?ZH+2;{C_fPumElnPAzuD@1>yWs$E`X
zd91Gic?}VIeq8(1i0=X7F)T4&HevS}3E3$W?DzG}vKPs$ksmJzo^_e)rhCa%vvcSt
z`RjMJdOQ3vuw^J~8<FvS@%Xj9sB;~{4CKbM!2RU8tX@^0l$ggl#_Yw1+q$EFo?Ut=
zTD#`MCX+DrJ7$|~!>p5$0(q8U%YSY`bIf0G4%N#(@1QTctIZDov9SJ^2Oot<-<UGw
z-}}3^6|h79jjTLNw~+5d$=-^ve5B-2NAB+}S#|y?VQ}fDTC!nz20Y+0&{#6H^2Pat
z@I_mt4GGQE&VlHu$b^0|{d1BVr7SnT0xO`4f_YjMIT4fUC^URS062j>6*?)`ZDLKK
zvs2rjw;)wx9G~`<!Ysm$m89_@s-y~P@+$)Sh5DFAVi^D{;MWImkp+iJ*G>Zd`VY;-
z9l;dySAHk^v!MOZSSn${0GP?uvyNKYFyBsa6uS*{4hI;ds;vm?Cx%!yhC+<uYOn-e
zPCUF_d*1@e#yLZ^@@9fmVM#eDXw^9QW7%7gYXW*6CfGM2CMqExEA+uBWt6Y+YFQm3
z>Ao^T^wJtZTc=Y0!1ychm<ecSoE{@Osip*KBoU2uH^931aCiC;1Tir#1G0}hT%y@f
zl2PEkeQ^`exPr3I8G3u$ro0eTeaOcP@}HA2$##NEw(SG@e7G}UZfFm7amX~vF1P5o
zjA+pRMJ4jr+(}+}CAaZo&KdlQk4+tj<GmbDs$+duDn8v7>M3|2KT?U-buHSzdWMA0
zV&BM*zWE91>?CNt81_=5W}O1nM45zcspy7pbp5~*mUf<>75hh=6F>H`Tl9$0cM5NL
z^W;LN(T(o#hwh-x$|}VBM~su~WWrfBQ3R9?d*qbwf1{Id3UvS`7!fMV`@1WD4CT+`
z?)~lN{a-{(KP~VA`J14u85d_Zguik$lb+bQ+lzBt5<;LQ0h_xi^N2n$+y1;<Ff;1K
zjHRR^w_4kRV#bjr{WyJYn4kCOAC?YdnoX6+Zdp*U7m6sac(MT!@&)=K+6^kLK>A;X
zDRl;gLTZRbhv3opyY~!j-)7~yaxiLr&I{*-9yAB|%7MW!50Lz6%DPB$0~CFxFaFdV
zT7~x!R51#mH*X4PnaDm6v4V@xZ9X&le&}%?aNrVZW&g#{rG0}OlY7ndE7k_ce;BF6
z;|Wm3FgcUV5XgR1qT3B%2hFI{0ww;t=HI(xp-(<A220s|tw}pMx53Y<*5nE1N2Ten
zM$~?C8C(O+BX>##<5x(e=j8B6Xm0eXfZA9}+70q%Re=(}IbNiKJ$7q>z!v>po3u{y
zi?&DI!}E3iT8$?HG5S^E5Xpz%6sqNscg~%qmd1EP`;b<~R>=>(n$ZW8X)p4+RA`S6
zmiCkTkBC6~aXl&(x7!7yw5mk9e%5c^MpUa37U{3I(3{;9@2{_+8n<?wbrLAFst~Uj
zT~Zqd#oNroHQ?1#9fV-7WmnV)zE#=YoO=bSJ~Ljz5&RCN<j|SG1ch9&E<vQ-iu+tI
zoZ(;+v`5?Y{^&g;@<rbRg0bL4?YQ|pj~<r$QS?v698orfx&~x6l$L??(&#%6OiNrB
zPoe;p`7Y1@Srs=T);And2jnwX$u!d`ZR6(%mEB+CLycUOmw%7rRA^$QSxOP!y1wcg
z)5Oa&ZghD=l5ihN;C$`}01#D)ZWhbQZ!R9EEL&NEb=;4?`3CQzTzVTnedI<$jK)^V
zB(#KA(OW%VZ(lCVKs8wwj)7ZXmRMnMSpqE~F;s&%E}TZmuwI16=u#4V+D_Xq<@oX-
zuXo80K#7YrrX9|b&rz@Op(L!nu0PoiMA&Sn;;GU{mskX6eBSB3>@FYre+c)R$#w02
zQ<nanu7X|$4x-jddH9hH_=CO01jb=<cLRM2&T^?xbET~NaV=4Ek)MG(bC>iLhey+_
zzVTV0h2P>%S{E`9=k^N?*9r`m3#2v-3_lg<vq!h%EdNK3!uiHOdKAXiK}DAmR!e@E
zwALlifH45LS%YR)q<tGozeR40a;ZpHU1ZMZD&f`$sEGwps|BJ}wj>OKUt$Qmtw6Cy
zs6T@X{L;_j3k2^91`)oJZ$nm5z%=*VK*7mUM2p-{1iBLBSyB<e`t#a!%<gK6T@iJH
z*OjF8uI190m6mq;X;2H<7QI4(c^yN{s7h39TTmFeP+j6!LrHm<9fg`SvpUuEmyeGR
z9+1|H0CWSBo8AWceVNrIKfPgv$MYIoDtQQ;koF8-@(eD&_KBj{6%$Gb6-~OpY0&<b
z*6k|*8d^{!C*G0w<_F5LiY>ueEvPhN$&i{jl*s^ag+W<{XSlr1a7hUaSq8qwAbF;)
z5pIi+U`0gZW6T$k<0T8REhwf@rL0wOJG%e+YJbZiASMeTm>ReSa`~lKgExL=(RQu^
z3(qY%AwYgS=$*cge2am0=aq>#hnELROE|{%9F2~IgohJt@tDCFR!)T2TrqWxHjL)b
z)HXNE+wsaI*Uo7;>)?1xxjGhx<#e~vTI74GteKKg`C8=mUK_Eg{hfa8F!C8i<QYbR
zB@>LyzBQrSo2|J6B_{VciN9C)$huKvPMOtB7c_WY3#3K`p}(KudQHuxP0i&_%{96G
z&K}AT6z=IG^kezkK0NuuvF@*_C`#XRegWw%iL=GSCN6zPs}Bx)-PA@+r9#bG=)`T+
z_RQ_cpNItL=QbXlk^sZqvrl;wTTgzR!2Qr>8us54UzYG2Y*jv+qw<w+I{BtKZKtSe
zR=0kcYvwdprn_HQ`W2{|2tmIjr(hZ^Vmhw8-fLas(f~vwhe;mR5Pf8P&u&uxrc_(1
z((PdV`G9utcBhc>Iw4Y^Oa)AvyIK~;e^8J%W7orU(8OvV;CZ&LKLXu!w!n;wmTsqu
zqGgaf40znhCSin=B*l_041J-4iBrNRC}6#ha|YWFTY~!?$9qo?-#KF5v;`DpBUic8
z%ofIGAt)=&ZK~(1G-zNr+Lt+PJaZgcb?3tAVcDmbrQ3W<oX2beQN@%^{ay}Q_7i^P
zKg)}ZGUffa&Fekk%af}7h2QObvmQt2cg0CCPTS>Vk6Hv`)34QU60GM)?MNDFXkS_T
z&O|2da#`$dyq7Nqil@G}6)=my?fyX8BM0>Vd;CNT2-~WlyruJhk}Uxe=f2He!(WS}
zViYKg^vj$lyWT2I?v4|rziU`4$5=#&1BKrn`Ci5efnX-MUKhn(+{5SV!F2mniu7O4
z5`E7GxdT*e$?#Xm4RKd(84j!#he;J<R6`D-OoOzMd-I4I8%o71s8dwzf`NBRR21VK
zcP+ioRCi;9z9~@w5S#^d9!|nV984{graE8N1SjyY`zpzV(&HpJg_P7u$jIS1gn(4s
zRHp$)=8Xasa~5dvddno<XrI#SXSUPPtdb-iNe(7HRcN6%*tXewOL4(MuWt0VxzsR{
zFewTLCykK0-Q_ch#I)nvJkwvO&^t5W<?U%&AYV8N6lwxW1mihlWUd%6^;Z~gC)Uye
zzAtYcKq0{h+>0dcREEw)V2b##RIN5a&hS!h#d6cbad051b(LJ5!|=r$JbYT(Rz)Am
z@zzr20%hdwAIS6VH>s|?6Y5LVwT3nUfgtx~8`!*O71`#fTo_6jkcf5>s=J>ItkcI|
zfHcx{f7OfJ|7*~Na8FyUeo`0s+!pi6dUHRtg5y$uMEs}C65e)ps#JO!PIpTo4uy<n
zSOPrj6S`}$aYfh#%B&O@`s~@hrH$moJ*bJH>lG)m!0bz?{>k)-&g^gL^uC{H$wS})
zB2R9aj&IXIpfC=To9!tyt^9L$W<oF*%|*Vv)wVefxb8bXI8^t$En5TuMDJw^mk_y)
z;<2NW?jpwls!swm7_a*!IiW(gtGgdf&JBkDc3_Wsn>un^mSUqkOI5J0f_p!uEOhiX
zCotpmquQi5ME2ZI!z@QZUtydGcgXM$QIF$$KsW9p>Erg{5r|x@8|((J9ittkqM|qV
z<$ZOO)n4;`onGbO+Fm6ZJMPWk>#a{9?l;HjzYWUG;m{J>o3|j{-WVaIoTbB$6Ia1w
z%<=TZYi%1hb)D40gA`p&v8KHY;ur;=AoTLqZov19B<m<%-%$CeZ&JYlL#p{}=L`3G
zk#jr)PTM~3cwXjo8g%^^_6zlz2iziXwD$I2H30=in1U`CB#vT7nR5o7vnU4`YM1;o
zm!~o5=Yy?=<UAM&_G7Ke?s#jCdL?zG`-(^sy@;|lF(@Uy;8VKIN*X*bIe9#}>gXZP
zHV8xghGB*+m_&rb*6eDFaV?^Sw3P?Nje6AsU&<_`dSc69(8kh2g6M<F(O-=9!?EJh
zI<C_M(jb{19Fj{T1)R^4o(krzi1b=GZBf)ue&chr|EW?&LVM1OMgTI_T3m%*-pvYB
z@b5N`;y%9Z9)gW_|D>E2U=WTA84XOzV7qY^NlVU`8gI2q<NJQcPL>Re15$$CMoxCa
zJ1>7=E2VLdRlN*fG&u;lIGf7qLS9031N!_yO!gC_<31Wlxtr)by`h`F*C0#b(D2=R
zo*fQQ4%keCQHXPM=mxA;r^YfSN^~=Wtqygem&MYTV^6~+>Rtb<h2AO5%dau17EZXA
z>o1Pl!o<v$1wRKZL`c!+{y4r1>@xWwnG<^}mxl3<E4b@%m1ZMW6#kXNh$9NOEAWB`
zdzj8ZQS-BlQI`qBxS2#`N7l(N@3GkQx<#3$D8xWPcJozX3v1skl$nwrhzHB8IO#c&
zTyn2)<u+$LotoLgRmVrA$!9UV1*!aU(W3WnZ?O4)jK7-V7gsQF{|Nd-tAuloe&_$N
z_m)vr?%lia0znWYmPkvdC><&t3qcVP0TrbY0RhFJ8;M0OPzh;ir5i!%5CxGgky0Az
zt}}1X-cOw8|9IXp&ewO0{XyM(%UbvSi#e}(#g$7FLGi*(Z&+&@ap(ow*SuV3P(NP1
z5t=&PgGmKnnLA8f*7`7kvUUrxHXZ7h4W=|GOkbur+AkcOb*y@|{WVL@d#!7RMY2Ub
z7h-lgKkYsamk1wB`GdR^rcXq`eHc>EvHLgOnmidkriHWTEWS{~u?1d$h8G-fIP^LJ
ze&GYy$AGT@)>nstBaFZk)HLryRIw*poNDiwmdwuPa(dRpVI3}QUFmK_SSq<EYxD=c
z*vn?cpz`C}wf~9P{12@)FqFS)A?=dZMa6}GWsFX*5?~W9a&ljHe6isK-NYEo6KvF~
zcPmho=%2uHq=s-qN?bmYU-6VdIF~=k#VHM}b8<4ux$iS*3I~+hgT+QvXa2^@uMpZc
zp7A*YmxZ5b#~Bc$It*I!Yzj^gqUKUTDe+k~EAfB+V1KixBjEGh3J47S2j3Ujqw;tJ
zO`%MT20%@|uAVx`X|@0qX=cCU^e3)D0*w$5>7u@rRVY+6{g!hi|CV!nA_qA~7Gi7E
z5B^zooG-3)&YD*29RPqKA`M_;+SG}e;I$l#3<udFSOT6b1)_PUlTEjVr&t!SC!Y+E
zVKU-?@AWCvGGpol>vP`I`In|*N$`zauNVrZmN!y_r6Um`b{(ddQ$9c>i$4&{*X;;q
z6aU0~hc82L3}r^}Z_nT>H$L%RF8I1!U!b}oq!0}T!bjfRcK~j(kWr0i=R0-@O4Kj#
zpcata19b>~J4f0XGtkW+Pj!j6E^G;O)Q;-)3R?lmD%l#&8x`ixx&}3tU0r(YgWI#_
zQ4&`uK*BP810OUo4(EZOS_j{5VxeYaY;(&2tmFKk!Y#Wql;Qucf}DR0OaJ)``I`u>
z?hNl7{l9+r2`(>SPRdcf=8FK_|9mTW;~4*XH$@JTWlKrD=|%9|AGP#bjX;13V_)hU
zA6b4;?;;2n-MI&Pmr$xH88P?yXvM!s3|W6jrnsL9?|PX4`!?{Q{8F4Cyv_0;pVj6P
z!9UUc;N!T3Ao3R*Ob^T7sL|i7-hW-?|Mk~i%ii$+Pc#Z;z4+Jf_W%9Y|NqbbKc4?|
zBLNZ(wY1#K;%}J|FR`{bQcb&wQK$79KZYn!4Q9X_G%!z3Jg~;`s{C9##Zgwn)gz>T
zYoh*}c1ZpqE3n!h(s7k&{x!69ruSN`+7$pRUy0vRh84KFc@Sz({Kh!Aag%|%Lm!ls
zWOdLe4$rJtBu!DttFFK~_L4yoQDuQUu@BIj3!Y}@J@^LA`dH%Op1JNOjNuWRY;F7O
z@A3aCK*`;VRLtE4LM0W*xN|8@&-V0Pmb)DBv~U?wWq_C?BKI~THAB3g2Ny{h5uSEE
zfN@iGQ{c`jc|~$2)wr;Qn^f*#8Fxgn-F}TW))n^PrzWb%tzi(GWC~;}583Ne*sLnw
z_Haji)32KJr?JQqq~xvZTO%Z&efFor<tZWDG?m_k32c9D7eY#T*Hy5Rjl|gP@2r!8
z&6Z+M40>e$r=Z8__(Wj$)w`~?-5h*4)dV41)JXd25?FL>RU;W@r@%ESUiSB?0tyq0
z-?0H7!#SirR8<NhQdWG;l(*CvX>f6mCv~E%TK8K4$7eqXx)8b|-lOd+wh4lg4dHmu
zr0D7S)L@sHq#bS9);E7#4h3Su65KBlF<bD0eBmIMMJ!{}Ohc?$$(8iL8g)eCF^EHe
zS08H(o%z}A1n1plP-{dT+nTlf2+qm$z<<3^y#^fer$et3kmCnwF5Esn_zsE;TbQ9$
zByz9JfYThB8w|CDY6+;yDQzL`j1>4;W-|Q@6_ddJK2CJXR$ks#Rl(=uY9cHeAy}?`
zIyXB?_4|~DTs0rw@fZw9sdQDB+D)yfOpr%crO(~T;rX{a{Qp>`@{63XI2TWyE4lr9
zaT-Z;E<pk_&^`M(_7)y`kXmzmtw6l(k<ndv6Ck#<$h9ixpK1DGy_XchFB+|aWU;Cs
zzIZ^w;=1HT>PU$~BvJ9QlK{A`&QZwiE&Wbfb#`+w^t`o<So0$c{Gh`0SNCGBzrc2H
zq_FINZ1%@G)ccW<!dJdC<ErsZp12rBuRcf<boRqqr?1#Dkozf+*p%}#sW0%y1xU_P
zrf#e&z7(6Z;^Vj&OpNK3JlLA(dhO@_YVL5;<B%5nu`j}RDuiA%ofDN^#)9Z?`eEtl
zcl(v|QxntyZ{Ig#`{H~giDEUv^+yuC++jJ$k{~$FZ$s-$%@llT#FOEvQ{I8{*AQY{
z6*=^l%5^YT4kDF)<xd=R6W*fd;@OHl2qxHAzJgzaSa<{VQL=``*hwbMVkmd&)k0Nz
zS&rPoj}HB+^mw=H!}=OY#D-n#IlWYWtoI#Gz}4jt8he}n_s$SvJ|Vy6Ke`CQDl$x{
zqWlwZ-6~Jw#T{9u4^L4Yx<pb6cj7r~cq=yO7e5l?M|k!ZVb8Bt5O{)I21`nWL`}(8
zv?mxK`8`Wu0xhEl#$*Da?6byvSXQ(U!dn$E+G!MYM}q`Nh*l=e%6V`RarT*n>dqDv
zWPU)$jq()}JcbGoilkPhx$Y#Gb=Qyn9x<Tz4{xKe#J~W;Z~Jw2edF1*1`=HJyNCfj
zkimVCa8Wo*oN+6!Le4vYVeqy%_I}Y%@XfbE+(vd^qLx5-<KJEatgh~TeDD5jx)Us4
z+&6ATUJ4)jt-5iv#-O4-+&@T!SLg98Y?iKP#Lqhi=2@Dv=s5II!AbMfEo&X~r<-(|
z36W(nO{CuVTTB~(@(u@orSBZ56PJRf)l*~IA^||H?K*<^0gZu|FtNv0J&WFG_=ohY
zQcy0~>%<1Se_^o79@C^p>)j(|uYt&(!RArKB=iyDNm3ZKmTA)voEi{`%9p9{q{Unh
zdsD^&6@%DUh<~UgD2d@c*&$M8<4TEoo)B;0w-u~&Q@^AkT}mR5@@Kad)o_QLj1~bV
z$^7@x8Jv(w{io~v>nc(fuQp-$|I!0_#Sj2ft9uaEU_D}DgpG;7^<KQ+0%!R-xNLfT
zJ)veNCPpT+ONiLz)1~CAY}|41TN#s^mj`ZVDSX?7eeQz<n+SL#bMt*Lm`WV!x@6Cd
z`mvoF@cHjl1@Di)U`-r@(a{aow=^@N)%zQF7Ku(fMBTpsCQM!Kwnlu2ftETddiJ$G
zMj@4nLT2q6LLA|B#U^2+Rl1x166gpYXAsNGC3eQR0t10ZW{34n08gy7NcF?lzBDT~
z>>tTaozI^i2xd900=XXg8hEZq{$3(jOPTfMD-F&d8YMDit}O12*=z5ZSoMD)nv@JM
zUn&slz)(3Ej{1ji4xN&A%->AA&NOpNv~(XtSbpH>Ef$s_PQ4RL+y5BKzJyib%YES@
zqV3}GYSkfusrf(O(><s=#<O;z?P1P!&d+;KW*3M~*Pz*695C*xr^Tr6A(;u-R~X5*
zNNWQ5rX$?_d~X7N>w*}se3}IbHmA#-#syH#<^9NuX2Gz=f%Zt|I$PD-c=TqpX{$I%
z$0UO1phz)g1J~Q7Iq>H}yMvyl*k>I+?`{%$VUieu0+sf86<=!Qso&jgu}?Y3oxU(Q
zzjMTcUH5vv0=u5b)HKCGptrtp9>!rldvpW>=N?r@sRl3I-M<Z1POfA~tgVcU*!rmX
z(mCbG8?Nf5OxYiV+<@2OrVcb@&|SUAYs|J0GS<JZGIMP4$Q#49?8cez_jbZdE+sEq
zg-nfK=C`y8qJWm&+w7Iu5R6{54kop7laLg&SJ!`8oICcgU;%Eikv@~AlZ^UskbDU3
z{`JK@c3pgUrc7$?#g{)-=JFbd&`*Q@SjpdVmsw6|8-c&?LML(;8$^J3w+q2J?@Z>J
z$)(Tj9gWy~E1O5a4E1Eyc;Fdnd}<R>jK-eQt!{9o(x|Nj0(&Lu_3bXi>dF?hpiw@Q
z--4#`@JLU)mJQrvq`lY;hN_ZXoT;N1uq;@DQqYdvaIqeEbUy6r8qC0sVXLN(u%mq*
zxF&J0^n+TYySKYC{us(q=#623j~{Q+^|5D%Waqw(Cbw~(@a9S8ZjgbX025?(s)x~>
zaSvtNW0Y|&R`K;wsEDfAzSdqYA^~W=o-Q?VLps6-ikCrip{?5Brh&El>=x=ya)Ps(
zc)`uIT?i!_A!>S2Fa(v{k?^wx!X)-y=YwQzK+6kZg7%qw^VYqkXnNLcE;~`e8<5(G
zZ5Ndb+omuP_pyYn{EC$NCPavM+7{lh)aFz^MD`kO@%2TZqb#r|AvUC9BUv(eypa*J
z4^vq%<&2U~Ges}k=@cC!(B5Ldbx-XC>cp!P%c6EB97_PZr}kf@jNdcbI4E|5FR*p0
zVZ=zSmDC+AD=N37m8$kI>U@AjlFF$^J~$qC6q?noE%2emndDd`x?omX39Gd8X!7W|
z9Er&QbXNU~zJzWH(s~+}H(tKTsYJ()R^8jz?|4!?uufFF4Ry~HyLQ8`xx8xka=3er
zB;V|1vnF9QW8$8jgT#%fMwr+;o_Iq$^7+F_f*Zt5QM&hD<pU{T_x5Es><ZHE8ghK*
zvwW06ZTNyY#3I;%WRHqnw3mO9!%D>JsJrJo<WQ*`l^DMH<DrHlOOjk$ZexJ5=>Eq+
z&Q&zt>8j`Nnkoq-5a%a9VA2fvew=Z$1X2PCR1-@j)Rm37%SFLc;4(vVL)VOY^W1w=
z<7e4UTdK2|;C7c|9arYi8s{dgDuy0r2>;1>5X8Sv5i^Q4^!44pI7_T5tYBGF5NBI=
z3>kxZQKX(W8G^cqfn0WL)D|WgJ1V;DdOrz0G0BRoPoZlrM-kStvIK1>1$p8blb8@(
z*Fto)Ozhensc_`=E8oTl+_uUok>*2t6Z-ID(v#h=Ta>YByhiVGw*<VsK!^S2=`8dQ
z$003C<vf&0YR&g;C}+Z|zSw9+VpDxpLo$_IYc6)z_&=S+oa)Z)e1!_5n8EfUwW<8=
zZdV1?WzeDzG?NK;L50kG^f=w935DZ{9De}ZS3_lYENFEm=SEFs>%w5Uw<n4tnUp7$
zW&5ep!ZR(2mC-L*>ps%VAL(Em&jQcAVYz9lw9|%L(>VFuaBN^tgTZ#i0E_6oE9DEN
zL((Rvj$53=Cm6f}18z54b_N=DKX4n&Rua$YmpBfUzBKI$+KE)SADZ?FcK^L+ltyU6
zB?L!MJ6qsGDh+_a=AN0^E*#|3he#qxIO`kiv=0)Q5yU1%WlU3Bn7#Ze{^D-<yPk85
z1;-w!f|xg+UAvK+E)iz$-#JE&skTnB6W!AyGf({d{`_>(vp6nAPIeRa8{8}Oc8J)R
z*^O)@Ag_j&MjspK$1AnXLtwh4nw)-*bk!R6;~QQSXT>+L(fFBcz9)*l`l7wwU}*?j
z^M;m<BDv<vWzpm~*Z7cr%$dK=^m@qv3*0j239dPwWSKo0E3_gX#wgXjIZC9r|4tR9
zCRt+W7UQGd=V}5o;X_NNTl>%g@vt_S%W!^TwIWMabkzz^gAFj~^-yuG{o8mvJ~{SI
zz}MX-#)svL_sD)o?sSB9Qhg`7ts6x3;`j0X{ml~Td#bbLquyiiIM?I^<gtOHC;ti$
z@VIHJuJ1VJJN#g|J(%7`J~(b~KHs?h+hQTj>i6bd>N)XJx%=I`IHBl4!-X<U!!D{l
z`;w~Z@;|kBYv=Jd@Gkx{UVs*j)P>0bvhN#dmbnI-`z(@cERvsFBzsxB(MnN{2&BOi
zh}jC#OYa~sFq0|9<#!M;b|hyj2X$-NhcZFXfO`K)^?XVYy|-J0mww7Ez-@KNj;c_N
zVZb`*Ng-`CitDymvBeB;igJ2j<*y2>ji+rkyZxv5#qYIQ74oRF(yH^0?YA+T4`GPD
ze3EsZ(x9$**io7s@QAPX((V-tdVYqRVg>gat<J8Siak;(&J5^i2=2Ik7^!cb9A$y7
zO?C^$Z79gajHb>@OZ9sPL?$aw1Vj$z)jI*bTs<1CF5NSxCuyX}TSgmH97$fXXXP!K
z*VW)_*JE~?ecxHloI&Hl&y!^{^C8@8f!u4sH2Weh<;BBqZM4u&uooiuxhDg-x%o}4
zIzA>J%XZDZo0n5MoM|&PC;TYAsK5e?(5>#6=@wj<>c20?P!a$Qyi+y5xYJ$NWvhWY
zCs(pk!#C08EmM)l3dW%&pj8jMV^6=@Mo_wp$b){+a~GeL8VO}mYyag|nCWSE^I`b=
zWM!YH)z;y2->S_XN_6Oov6+h1-YQ;83^(uAVkpVFY$pLO{acj@A}Yph3|8LnEt0oH
z#IkFF(tq*Eh^GCDhZM~vu^3yS6|3<tg=$ys{&Y_iPcT0PY(KN9p?6)u>e18PeXL+A
zYAt*xx`<mZgoauuCi9HT+ph}=OS0D*ZhyNh=crN)o#G~$b(l%VZn84&>4;5TRp(Z8
zO}EYGhAyV2+J;Q`&`oKoPfdFj^)AGA#)<uM8-u!*4Y|(ABfqjeB>#xF4PWwnDg95=
z?+;k{8z|KxemARP>{+x|l=72DweiYitIRqlP`97PSagi(&6}Wm-fo-<4^GqN?}`5q
zhASsdq%74rBh84WuJZ*RW>LEeYrTqP(5%i0{@Y@;*y!wdLI@4`%mPZaPX7e@jlaDB
z<S>*-I#V&$q$7Fzol)WQ{^$0(#d#JTj>&0j*@IVh{5-#2AH8w%1*?4nIi-T`-HPjn
zMbx(&-a3K<=RJxDn@zAVfbQZ(0fkuqb(0E<^z7wQEgu`LDKJYNH}%VmtP`wL3&POx
z|14pdkc;dbSYSjegCTpkcG?aDTEo`k1WKa5uyeJm0Ia&RQC~ZS9VnaAC3M9;TA-Ud
z9}MIv7q2*(YBGKfopLx!5=naZEW}l6kL|XBGqnN7cmb~!6DD}zsK~G1p#+z|7<{%*
zv5%-u+`nj@OzhF8HOUJzwl(4xXQ&&Nh>Q5gANY4=htuqLxY--%MF?DUeTY6i+`je{
zk9Fd*A$oxQb~`OW7Y$d>VGXNCF4OMtI<f-des)+OYoNHBiNFaFng7t|+9hI&_>S%>
za>_84w)!x|wpLs`dGK9#Zl9}X0Gl@x;4ZbURQ<;eiRu_HVg8#F+J9ZVN$1cX(^2We
zscc;Pd2Va#+L5Uqu8=n+y`G@#acui!O{yP952K~Pr8l}+VG_OUbc8{&wBU-7ci8;h
z$y4pRw#P1t?>T4YkGOgPg~wODf3PNb{@_*MOvz7GmVX=;r);Ddc8W{wC?O#p;ob05
z0yLo}LC)g)sfX=-U0tFMiX$JM^cl*ucgirm&H65B_m*YHq5X~i<+qBoMX!R$@1#HD
z%K1ieCfi5=6<nNKLUMcl>GR9a@2lpiNx4H_`o-Il*Lj>9(ky80oWv?e$26C3OP7>s
zrHXR?6rNcx9M^1YmCJ3eMrg}=+9%&w7p)FNYKJqo@pZ1fD?dA!5P#;urRAwQ<|oT)
zN`ZG1royRgTBC)-D_(UcG201UpQ<>|)4Bf+#X8H=XraqyBVzf|SR%~k<uC-obv?~X
zrA9b@?D}_$#G;t_&vYhrPPx9RD}942ju&tb5$gotCe~>!m5yWb1?gE19pS!k=3S7r
z1a^Nhn~{0PZ75l?u=io!Ax5@<HnP!t%42g7ch!bn4XW~wXM3EgWaeLdk8=JzxqYr`
zIYgD}YDKTt_3`J<Q7>}^ZiwIvZo6JFtG*N;FtSAKX?$lvd)|9ob>>oqX=iOI-^l(i
z2aYx=@NlRfeEanNO*6X^D=MAsO7~YO*_B)Bxcj&KeC0|-X>DRvtc#V%FQoor=;iM`
zQfVa6S=txhy1(?%yP&ijCDchcb(L)J!8wUq7ox>uTV~fExrnKAq{(dfyX7C`Z-gn@
z=U<k4PoZ5vv)M`U9qmMcprErtMY*}<<AL^2rgev!CJL;G;!fBwFTAslWBSgv#L|Cm
zF}#h$vgOj|yhjq{VT^wB^40mbEc^CZRWosax;2lB12F3EOt9wq7pO5ECa;P5Nc0WK
zpO;S7#*SUYG21ZP3Ebxg#<v7=r7Aj(YEIpj9p-5FDgTl(?c0C4vR|vtm$I#^=br!N
zGUgs$*#nOBm5)!-`c6`mp5Yvm^^rtmwruIkPam;4t`h4`S{`jnsrB_FLZ4K8FJgRc
zd0_b9;>TyGaEe}C1q+O!wD_Bx!cTna$`lC<efOe`k1BQhc|J@k>^BK&Tn8GET78-;
zcGE(Jjbwc~v}Bd|nlx9Y)BS164X004gYWjP${j<=R2X1YR~9~R;4Vzkt!65*iM50l
z<iJAWp*jJo`G`q$)jVRX;zd6$szG?l)p)w&<?%14tTLXO44JLlu_ufkTtW4deJ-3b
zZ+2c#DZ`i7c*uw!na#AFqe3L6p`t6<j_!AU`Z;~P)>Wnz5660+xiKqkn#)7u-0m<j
zYMG+-Aud-UDQp>BPAD|#$Axiv1BC(cgF3oq3xgP)gIq&6&T0xwn`1IU1nN&TFI8`4
zC8NG_Z1{Q5s}g*2tS=iO^n6NFSqTCkQ<;1FD#7iyiH{IiWaU=8T_z$pl~=L$c^9=m
zYsUYpghhu0MbC3kZk$xdLT?S}6n5ivg)WlI;``HOpLbGC|9Ex@B5+ky_%*4&4=;6m
zWjxk0w%}#4M7CRO&WxI$dCl-^68GoL+4NRkrca*!K3?H<P}Fcj%?Hb089J>+JVZ%T
z!Uv$N0&zQ5;yfLGP0<(0Gd=;3MC*!mxfo1`W|wa`iamfb-$n|`DKd1c5h*HQ=LrlT
zO6oM7LnWLURHb&NQ<U!BaMt|AYTC5k*3Ut#9JT^SqH<yYgQ^&7vN{@V0n~SLwiZ}{
zky-19_WqTqe3BtQpA6OD3qLH)7LdkcXa~Z0P5OrbJVW2nsJ{;%#0qP9;rms~o18t_
zxmO#gy4mnX55qBAfkwfZjth~HKX?YCwqP(aV+U_dX4nOjfzcBK{%D!&4`{@VAXdVd
zxNB)7=W>R<&W)RurTP)ob5R!tgHU1X<M6Wzhrx2YfPdms0Z`rqk~heS@K_g=<%33P
z9*K=B_p&Buo~V66?74*m&_}AB;oo184YPYYR8rpOu6-JHcevi6G%zmEDBJ>v%@1Pl
z3sx>&Xy{u@8TqpFI_FQ@c8v?~kfrONl}0_*la%~<#=R`7f*`g|z@HMNcdiaUqOIlf
zz`K<2J;(l=)*Wuzk6eeh?nw2{gsN1iC9LpoH;+Y@{ia8|YT$iaso1-O<9s3AsdI2`
zrYeZB>(rChaJ{`KX|l&N;-}WSlUllpBxDR5kKqWSz}&wi_w5RAPzzyfOQU3qvJd~9
zt)^iNI6=Qqak4J|(k0G=Q|z~mJCa~8Dg*ayOUeM3MoYA`_>&nsfkEu@W3=VW_5*SQ
z+gd`+43&fXbM}J)Z51^IGwTA8SD1emjJZ;m#_J%7v}w>?Nk&)n^SyICYy3n&L>;k0
z>W@=VxttIvR%upaQ&~Ekoh<t0JV(Fd-JE+--As$VX8Ov$EFU&fv|=5>1tVod;Hg0I
zkHf>Y%nbFZcNtvj%y^>09y@7{4ZYJZR6MLIFCm8Yb(4BurAK9U7ppy1PL8%TUN~Yf
z^hQ5Ir$1L+`{$yCl3c5%>s!*ZpToLC-#PL*`ScAgS~vT~jp6iP#!H*n*q1*b6{J1z
z&B<rFnWhyxs$ZYd$^BjVQ(zE&7Z~=WMEfTgRLwt62wm^ml^3*nM7=-C6uy~kq6?3p
z3&@VQ9ikncjZ2x>dmR=iWEU}a2uo>KwC%k+^ir7F-mCrJr+Jm&;2*L^D81Y{jP`q3
zOa2jZl_c&31kfU>Pk?AGDd$VHORpa1<s2JCT6KlE8h=-erfaKm(zP3uMC_rX6Dvlr
z2cBzVjYB18GYgxWg>e?NZVTpktYmGnPs?_`h6>fB%%vjGV}zT<CUICxN~E}1Z2(|Y
z>`nDu+))NomM<svpfb4bTowE5)p;X5VpLZCLN*S2TqgdhBrpr8@R>E9(Pc<Sn6FF+
zuzh+3*hFlgK#}vxkiGKtZI_kZcq~u327b+9ihutvi1*gRCJ>#_VAo#3zH*v>RyYy@
zM`UdpGB#zadB9cW4r#P?)d2bL*pz(Gs^rTmrW(qv8|m<Kk52gjr|?b0+bss7=IhTv
z8sDy+g>JRXb5rbsuF9@1Y6~WjJQ^B_{Ss&1pJ@dWKAYo1d-byB-5Ix}sFLzPWQR8V
zqNG~+vJCUIZZe8RPUaRden_oni)?lv*Z$uJEYWP@A3?~PYcOE3<(-)u`0E({%7&MG
z!3J!wEy18o9UM>$_Rdg^qSdjh7`pKzzl57lQFVMSY<A+S+9Iw7e|t`hc_T#`CRxc^
z@o5pzTDW8YSsBPFSTA;O(W1RbPz9jn+2sz&Ggt;J#$O0W?1m5w2EocKMW&JBBlc0D
z``tV2b_>AlC2J1qST<7d$wHD@5n_fo8+&_(Oai&Y0u0eww8;Hxym!#r+PaxI=K$hn
zm6dD?>!2%lWTJSDh&iNx>Q}&Imm%4nt&TI0W1K}JBTl;Hv63BT?iEBLDGFUbIaJFe
z9G#Iv#te|HNB#b(q&DI-N;)(@S){I)+=MgCcDUKYOYTUbQJ{DOZ!+shY^0e9f35<t
z5`p}ATD82=_?{@`yfv|EOBhw=DwmpM=f4?{zCQtPYJ@ek?bl*s>6>T5pNFV+dE#sp
z^VvF$5A<HiXxEC&sXZDB=s#0=t{wSraqdWdZ7=S9j@f3u^2wj4uTMC$2U9Cw-;%rK
z===?v-XvT1(sk^d`|;O3NT^Iz&X)9KD~uZGJH%ZALSsB0JW8Cl7$`@ZQH@;DuAjeE
z%D>Z+&AfA&^Rg@7LBz)lr^7?A%kvvC4T%PvgoFK#{ln6Qp4?0xxk%mEq4l9}hQ8G)
z39sW)wv=|NIxaG%fvsPYyZc!S!zKp2Wf=Hf?qlg4Ky;vchcF`BrxRs2V0W*(0VZ@U
zzXy`UPWc3$x>7|2cZUM<Ugr^07nmp0JI^j3EudSS1FH9>EpqmAH}rSL+*s$d;55nG
z|M(ppU~p$G0)zSn@fdTxI(tR~QtWQ8Spdp>_<m@q$^nKDejdgllOL90xTmhhU!wgU
zHy#_5cC&JpwpxEa%KeX6di9BDPrvQMH68Td{E_J=_{<TX5Ui|nXEv)XMebG5P;%q%
z0!7fUDVW|=;H=Og$_f+(+3)1rq#0g1-DOx^tKP%&gm5F>$?&~aWzteuQ8A?PtzGpo
zr2g|TNGy7%TCW#pf=mvWz~6q7-FJ_p)%V6E`^hJdVw_HLu426($!+csnE=Mv+{$r?
z0KgUp9V6fgePos5)1-GmW;5YQdw^tPgD%a^$^g1=4P%}{(7R{-Di30A(<16+T2;M%
z(GWp)pWx|BH#Y`Mj*CA*`#PmcOq7wDwQ8BfWVDkV6-auMl`+I3^fWxp>8{L^jx;B+
zjN3X?>SE#GNoYT*STI;{LkbR3%iv>s=fK>s9Jf*5rA0Cbb8TzblGaOWYoEz<-k~}S
z9LES_LB!bFg^u+oKZi3ix3yQeyJYM2;@wuAo+tGb1K;uyftMn*Jg<UK&t_=Bn2L^r
z*49V;D^0@UO#0^s56|1fcJRx2TuqwK@Wq@+FWU&xqGF7vfMb1CXsJ79()+}dN1A$*
z-k<Yd0TOj)zMX?TMx{t>Cn18*I<~@M@oZ76iNM=J7;|^R&^6zYV71XV#(1QaQ9~zA
zrlJ(;V#pq0dxE}}bJboC%h)VEY|}b1pGq2Rk}G`U<=b2L)@jS6iS9HD7HjF<H4(Tc
zLYvX>6TBVo+QI?5e>OHKE4%P=zfF~FQw_((0Vw>enM*qLRIMT5)O@X{F~6ij@8$h|
z0@-H5$CHY;o}eF9@vT}UO?EK0bxjS5d_GHcC@qrdba{9wV=EXoRI{F*r9g1#yIVIE
z2<4Cv13IO*$ue6ZOb*{V?!`T7su<M2Mk8DDYt+MI^dX(cVBoN6P%zPff2pY8GJV31
z^d!^%=O}}Dh+EIMnVsnw(QLl<MRSt->~aJ$O3XgfPhI*VqE*8wRq+nA0;ch1WM8ah
zE5G;#8;@ka>rYj23ISbtFpI--Qo+FjfcBF!BSk}eLmT(0#2S<<#)2#b-G>M*jH9-=
zdy8SRD{KA2F5Ka{bjDjrk?zGKx^m9?L6r+ci@pO=$wAwu3u|&=LbOLzx*YSo<(doa
zBGx-O1|;5dJg$d{F7u+@d#j3m!9C8!lX^HCkweE3BYxhh*he1J_B|Btu~nJ>q?!F6
zqzd?LjV+LVMs5Fv^b?0-QAFFZu6?9!cnC!r)N<H{Ul6EMVTxUn_6~BI{eC_a?(H7~
zFcLOn4mjh(<a+92c0El8HH(AHirU^p%EJQI!c|_c$?&n`k7(?k2#&{rX1W6yFgJ%_
z!mHHX_yNmGety(5)6bpNPhPs?s=^Pf5wD@NSq>W=hR*k6O-i5dH2U?cj5JZ4qj|#f
zukI@^5*uvdfLzZ~kj1-3_}WZY!Epi|uh@FazRK||BgJ4~=84y2tuiE6iD_jzL!H)`
zS9xU{?@XLCIUCJ>OUd(92<`ggZz)`Uo|N20QN+YWm(l3Dli@UWwIWn#O(I2&21e}h
zq0<XM@)l(ETj8}VxsT35$RQHo!r36)L>D_Ym~FJ%>gC#UYr|xyd#Zz`URK6C-RWe_
zAcyqZ*G{PyS5>V#CI<r!e`y;J5zC?#OIOPzTQ8h%lU{A@@aEH{*=<_9O6VC(<SmV{
zZzI7!;W$O=T?({onaQX2aaHUjn7B&{slgxVsw1c|WlOxUal^UVIE-RcE(Bfhw&{{>
zXeTB@3$oR;S?AtT5v+biilrvvBW|H|6F-B!E3<NcjR_s9Qh(e7?}+OL9Pjj2_~Th7
zGmW{KStLX03D%+ZbBmeJM6{p0v};KDzHi8JuxnWE!mu$YyyN#(_)JbOi#3&(gcSq+
zRm^MRvV6ixajRY!QA$CJX<*}huuSfy%COAn-2`V|SH4`+E01jR>=wij!v?y=5^*k5
ztw#II5u!rbwF(-CA&Q3D_Z;`Ec&o@dL}f$jD(Ot_fA3m)y>-5q_JH@9FkcnNJ$%HI
zxD(>c-Z(~$5BoSs>~Yz*xO|7ehs(`HB$dg-N(*I{&;B8``p=c?Nqn3L=2hhWn%5cs
z;-`#O6HuD76@R8}&?6>Y@tqcQoo=H|Zeb%oO(Q#b@XMi^G@Q9(h4O$+c61^?V<`l;
z)#$QPjZUI8#^HJ~s~n27nx0r*Eb#(DVPd1-bo|s!Npx?MV`=Qw!bx`Yz3|bMzSUMp
z7+_5{to5_!AZKT1O+MyByqDrQ+iP*>;J{$RQp%0@jH@fp_*L$PGnMJ{bT-P@wh{YZ
z)d1W|{+f0ee>1)UmG=dL>lj6)<MO+dM4Cj&yveF1UCityGEm>vl2v`fWxiNeLfzs%
z%Qr{bJV=dV$JCIEF5_=@(h}fap~6jVTS!qewBbyjFXxqpOh+D=hrcK^UveF+mG)BC
zo#`nJ_#RMR%!J2k8b|9#e)O{|ixf5LvEWAk<@wKX{1>s9uYT&R%%3h7<mpYER+%I4
z+&Z?WtUKnsM?K*mMm|($E)-epRYi8AR&K1Gh_Bd7qJ(elh-PSp90f~59HyomGAp|q
z!p$}}O>XMCeHgUAK{51TU;jq)wKmP(56MX-(-cc)TPxYcl?SX%b(?0s>!;+NG@w=b
zYl&Fmm~yS}_^N-Uo5w$j#aq^Ib-)a@S`FfovR_qazZF&5mJM5f9avI6<UT5hp1Hnc
z^FGMTV=cUm+B!&_{|q!ASBOzTi&!T-BYXa#WVgxopPN~QXJ$`bu8v=PvGGTm65>GH
z-=30E`cLw6jSx>5aD?7p>}DmvVoLa^o{+ERhL*MQ40D{W78PdKqz(8r@PK4=nI>40
z8-KzRW$70XqrzDVbfCG3(Ew+7^M@yE`s4q&O0n{XR6vkAY*yv}OEd)KkC(9;YMI1R
zIN&2__L3<=m}M#=BW?Tb=RGpVFt2ud%c`hnGG`<-+w8$AEYkA+q5o257P}BH0hUvR
zZv&jC#ggA4Lr-*z-8JU)%gRaeNxB3zb!I!K|48Qj%7b7vFTAda{)N>*hOv<=r~F_o
z2u$APT7fxuXdBoB*fp=EcpMA+wFWCU7uW@-$HM>-sPbUUi43R_VwbPjK<tN0cw<wg
zxqOk!!3s0_4Ft5-pv7mrD05xQp963Hm%rTJ>u~QaP=Xghztvj`_LbA0w6)aD40i|n
z=0y{?TW#lK2G9QK6HmwhxY)eQc7y0Y-|bFB4UFU;AXGczJPVGQQiPHd3yp*|*hl6A
z_VQJBe@0`|YPBbK*$0I$KfeUqx)@<GNC0cGAG}xSJ|Ns7B<i%d-F>5I;?lz&vOh7i
z=&FIpblyEA!<pkhKD1CUG3$G0aB!RlXUNPFIM*^HFgLL%9gLA48ZnQA-o?Pm@f$#f
z4?gpk>HXh>04R`YeDnD~bF2Pl_WutE-B&a`JE^*~p#OYy|9tzf0RG!QNpc8ni{<|z
zi9iKeZT}BR1P2Hq*(3Pb?Y}v<|HHTZ!vkQ&XF|iy<It<P_20kcfBW{o|9s~eP_Ore
z$?yCA2}=7zef1QHLJ0l0C<HoY0-!U?xPS|+4Ty#d-hel)quA{n>q|!9Yx~F@cvTio
z@mK#uhW(eza1Tl9Sn@E>?$P>d<yu@Cl{?(V?SnscS(Kss05AdEVl9#btO4mL1sFn_
z;4TnHES##ZZqWXAphYZWhb&S;tbwA;WFF!IB;f_nQ<uHGJ}hcUB#~yavFLF)1am!!
zT_9J`pPz)D0jO(@M=?e4r}V39FyCfcx~&i1!1!P`9L07#-E}$0usGtz?Kh4A?i5=1
z@Oj$FMq&k_k<1G@bwR`<Gix6qFe11A;0Mrt8@y-iCSS|1mwW_2d<e|%H#5y(rdSHd
zU4=h`-CQ~A*98U%fx&e1KRjz}%^@GG_`{nj29OZ?`)PQc!ow&demBD)kwye1zK(MU
zCZy+O5ybl#&|m{B-jg!RGTgCK&<)xH7i|a#5oX}U$zaak=#H4u?)Ft5P`4v)^)%tc
z=&&myEu0hCO#6FDIN6<#RoiV30b>3@a|jX(QD9cBd8%Iw3tAdS;>h5U2%RTD7;mv)
zV__3`I-M?#s919jyw**D<m<=(eyPqzZ~sv6!l>tw`J2@{5g&k+T{@i{Gzi~&=TL*Z
z)<7=*l=lOQPaHr_wsnC4;C;Jl09{J!$));1R--M|yFhNTqILBG*MSgf$OWLv1yaK;
z#j>8B;pWO_(6camxO%GR6*Mu}!+F%J4+LhFq#^O_*o3EZixp0hzy8sR{}1vFudyqz
zc*mrj?^ph{`n55m&|CW2z$wpmU-xdfvNc#Di4%At$UP>Q{W?8<8h{~%$vkgkYPoF@
zLuS+xNNb9YP)OV6v)s%&6!h3u!(I}_8%3@R)2846VhsdpTBtbu$%JtQGM^T40i`87
zN*W?Dh%h2x3o!&_eu~{W!$Zl@l8Pi}e5@`qy^(_a1#2+M$~?fc>2+-3is}bG^k8J$
zk1PlY)-+fqLzwf+E}56B^H04!xVf*#jyl*VtIpA>ggb~Ht%Q?1*cvQ&;D9A@1@f3r
z4du{oSK`6RjV>Fv?OiO&5f12n<Zb~9W-&hmecD47+@xTKJIje_d$PIoGs8QT4-(Q`
z{_D{RhlkO5xb<z%;}Qn{w$r9Y_rn!A4@M$E*lP$YPbpPSc7xB&N;$t-Qyd5J4cJyB
z)eq??n>(}^S)RZa7zAgP(|AT4G3OO`M)F*Lp0Npd2e!lIz@L@ZqTeJiaGv2ppn>@e
zx&1e8>sf|On`B#yRKO*S#r}xW_&j*YwD-0Z|1K!J{k2j3f@h`1B;s<p8drgLYZ!nM
znmdQ)*jDo5BlOfAwuTwoqgY6A;6$}5(I4H7&o&-@nsL*kuT9hr_ooB!xC&B;H*0Fe
z{&m$PncyQ7Y!0(Cnm*c?q|5NHOVcm(j|cQ>^bK4krj&UIP!jWFyipJ-*dPW}Hs?uE
zyS`>T6LN965%AbTdY2x|8RZ#&*r^;-znSw#$a(`Q>}1dA|6TgHsf>c$znCY%vF^L$
z1epuOn=EK0?z~i&S}2wPpWl8(QvSP$7OYExgSF*)z;ixZ*qgi|G{s;<Ge}#t<eR>}
z=IpH<X4)JYB8%krpie(Q4o_gd4PFdAMUO6XcDpFG30&f#4Y<x3Y&?GgV`adi9-s6q
z<!-70z<^G7Zh(~~Jjy0yI)cpy)rMB1<5!BEuB%RP-px~ZY$+`n9YPHh>;^e0$m`;7
zdil}48^M9ZwKoBRWIXM->btwuca^Pk0*;t_dGA*)W1LK^5$TUAkqkt$DEp)rPL;lO
z(L08^<Z(>!6Vx%*`9{*wM=%lWC7h;Ph;N}G`|VIV;kv4|G(qur9u~JX&F*@~_rUqx
z04)yBtyXAHFTS0JYvH|4-^8g+_U@h4#KTh(!V|%w44Ylt*V)TXXw)3S;iNMx$_<Dq
z;=&tjBW0F3NxM1&oa8L=M~DIW7s$29-UqMGV2?OVCcz7^EPh(UoAP0+GMjC<^+^+M
znMeKfX-}NwEbkr5w1`4Jt2?f|`xZYPtY{s|>sLi<M>eZAKdAn6ff$siUXif&FDj_J
z!O=VzyZZDc@z!J3#X2*+xjSBF1%?<UOxQ=sKpQ0Lg~B7gVD#L+8NbDiisk)n@#qr#
z>Igx>sx6b^(^3*}hzx=IdCb1`yM>D=tVhGhXp!T&t#4DZwDvVSwNIMO2;_T>n|v-b
z$F=399r=LM9xJ<3XY*J@PUW4CxC!VMvh~#~Z&hD%iHi`CdmeAPV8%4DkFD5NcS~L5
zzRkYKu*LXYCx+Tu@{WgN9;jh1S;K`c>`GvzAyxG@N<3S)@~C8j`9N0n=LDU`o$#Kt
zKTqU1;P%>mq}|Q^_f7r`<^2pPwC$l*@HYy%Ln-dI)0J3?(ElW0X{+*o@?R}{{E^Kc
zbMo{Tq<Qq)%9=M_gIUmhgsiGdEyJ=43Z;1%Y!;6I8=UR?c>Q^nciY#3`XMn62n=L*
zz`D!D{YDlE<pBs#nxK?7_<hMIOEtU{QsDHvM*657ftNe)mm``i<igM4?o_z)%n|2&
z<CZbu5U9V#Q`{^G4`r})M4A+J6d0a8(<{c{X#h@KMX+vrfp$6FS)od{8)Uebyg~fS
ze*|dDaC6<CLTFyyhp=ktObDBfFcT|LzRU*ju@rWkKV6EB%=VLs%P2><xCRkHstwn|
zGPEXIRKsjoAW~FYRqqZj&a24nXxb_t_fp+hs;8i+)v9Wn=iq)1%xiXaKJTlymT-?6
znN_pi!~PmT<XG=ZHe%m7fv%&c$;Sn4Xr%T{<GvDxz3n=RBN8#mGrf9k204%NTVSC#
zQnrw5`#p9a(~1_vX63(z5n(;pd*f9CDZ9?FmL;JT-B8eqgZa8(1Hak+EWm{3Wvi|*
zkjZVk<Vt}5O@xi6E{<hyl*pLH%x(H-KcJ^4R+;Z%izz>513Mv2hx|to8fBt}qzoZs
zMmA^D;CziqNs|F4`~F%)AJDnuQjq0t^lI_s{y~d|`w9DiX|hTmvelpx;-+ksEv2T?
zTh7b-0}RKXG<bl^)Mfp3BzRy;n7r<kTvgAFJ3abc^C}tm@~)?>D6CK>X7k7KSgS~c
zd1rid6PuJtj9kA1soZi)Zl|q$$)G;kC^mJ(z$Ad{a;1Uqh(Jzx$TMO|T@48FSc3y8
zkIIfc`w*^CCRyRtpHS)Dv~24?HAhi}XfJS^#5?Wf`3FW(TRu?yTxf7Lix{<nbjtKw
z9+W%AS=LU8qaXb2{2eN9;MW=S<rmtQJjuT7(Y*ThmfWCaS?(VTk=LO88ok!pSM>Mw
zqK1#Lk_7W7rEFi`U<)Cv2IX1G9Kvvt@F$eVA<0U1dHMYoug^8&HeO^d>q@{~p;N$0
zg1IJ6M~Eh6=S?1L74Or#=)*xkgfV8{p)esx7TSj#a5B6>qhc@IIy`t^facsvVeq0X
zg>`Hoa=pb~`VOiyn>T5ZffT<QC6*Kpz^f|%8M!3yU{w=?RX@zS(H&!0Jq!Vh><R}1
z1LP19R|VQFF)r7DL6lMT-cP|0$w?6B(rFR!VNm1GwxDETK$D=|-50D3tul|+qS(u_
zS#!UEefnh%se64t$^aOy8V17{Ve(YTHtWCw`QTc|Ydqsg_Mgj7pbR|r*V%p-N=mE+
zz)Bwio-vil@doqY76d&<>I-810^tTKh@R;Sai1c!51^ZZ%Ph#k1$!0iiIty)F+(u#
z)mGuj??B(Du{2)YENZ{C0I4hJ@Nj!dndj31{REm)==&ma=oTB$RbH$NBHT4iU{6kw
zhWVjATw{KoX!J*UGu|m$6E+_m%sou2k|8>k)$y)1P;UGz4q(uJKb5(;3XF9sDWz!$
zPwQ+|8cbfO6J&qR_Pj2V^2`aRtXmeFpl8=%h%a)q$*2w(_uZmhCciW907sr+&OO$|
z5LDQOko+Lt$Y7!>T0_4O<0r)0g5WRK$s6^dUq>##HHj^(3cf$tvQWp@Q54lA%|yte
z3aU~1hqOeD;D?<9YqE5*><h}mbgO<&7&a1yIiloR`^!A|@)&z;&#3%U&GSnt1n)KS
zH-sL(q3Aes5V!qt@S5G2%u$x0yo=dE@h9Gd%9$8R-{L?!iEbjL*U_;%)b5p0vbD2F
zDQD;Bz_-URr+&kcXX3mbTIq@G>cW8d0F7LGC0dh5*CRTSI4$7oXUQfhqN*2jquOWR
zx1R#yrKX&T!mNVpRZg|<yN(hQA6-c2E_@403lz`3WnNjGwdSCc&9YE>r|+L&)xR`-
ze>FihVJDU-X#b7Eq4MYwGaJu#^S2-9k>?)AbFDs*VxD_^Q)iAp`_56oNxPzA)eI(#
z_g?)#ANhQhY)AMoGf}U#bLkW=>;(|RPCfov(&O((GAKQ&>Ga%V@2bG;CFbqSF@ZzS
zT1+@SM&GT!KwfSn_%8GGJ0ouR8@Gw}Aw|7BOiwRd?^?K?MYME`=CG6dk{+!W7p+1%
z3uL>rq+ILKuaB;4;YcN<p!MsHf4EVv#gz{E4@m<veiH&$)dX%`5s<V#?;rn8p2PS|
z{7Kp?%CY*^T`e~H`g8d4j*%+<mjw*4l15YLj`4C8smlVB0TZnpv`bvH8)DL*l?7S|
z1zQMRTT%qY;~dZzB_YPjlpN7Z9lO&;OM8!#){GNJ9<4|DT<=PZT)ZCfbLkN4m5)9x
z9Mb7X$erC(?EbXh&ui_1v~LMqc4oXwepsGite8qJ`Vr3eg4F9p(9rCcn?Dzw<EO+v
z`HFVX(sK5dJkmQ{;<ardY}A+Q<?!%rub&znS@Ct6)%;W^Afs=2pgruK2h*3c%PGvd
z;z-h4OQXBp{sfVGe`Bq2l)%Ra=M8-qGp;usG3<WpNbvp#d(Dlg?)bTeO+z4aEa%qj
zHTX}GuHqbsX(RkJ?#kFJ=o~!%v>~(pv_C#bs0+SZtX}qwo9E_U1pRnHivJ?WSL>1b
zOwYqUnQVU+q*vTJ_Psua8iupdlD0)%EreadDxh`v(!dZYn9XxQ#QLb3bblqoBvCwc
z6HR+L9U2I3C$ZWJD@T<_1(rK2QPpuqA&2~twYTd7+byRY1P^lGuBXJ?&ocZG9=nf^
zalZDClbr4%!IsdRX6$nnf6J}>*u|rYj5tY&q<l{ozw4;zhoUT{hwy_M{M}@oS8G{n
zYr$6`qB^_N&BrETR*1keA+TJbwXt6MzQzi%O<KG1*@=#@=PH4GGVKboiZ}*wKBtPV
zADU|nxvQ#{En2HwxSDQT0gsxLqg=SUm%o6lu{V|uWDxnlh_Ck=0<l6OUa=}vS3v(`
zv)mb3U5U;q$<A+Vxq`w{avD<Vn5ZZ#_^HT|jHw;m#+{^9ywZS9xGLiD_5q7#BW!xQ
zQmvDk9p(g5%nJFx680#phf^pnRvy(t-Ey5QPcVE1q!J6iCE<~Ty($%wD+n!k<>|<F
z{Tn`yzB%a(&3)F}DNG-TP0Cqn>D^PW1h+Ufd2f%OHr~EqG}H8uLR<}EY3Y|G#=qY4
zwfmTo6G`$TYiVEFftS#8>j-XxVG(7fxECCxf_J=j6^L(dSYDfIu0@tSEBV;<H3K2j
zIGya7w|!gpu#a!<Q3eSfYiT13H|yvGjyUiq$jylr#7)*rmW#hH=Z=MY2W8B<yToiX
zNa34&GNI2F&)$l&ci$Xee0v)%;*9^xMa-RavgRN{;ctzE9~EBb{pW>|V#%pp);d*C
zkX_`{>QsI%o2$##c<S_pebXmJYDX8&&rn|~nPw>tIq9Biwuajcr-^=P$4q~=lgfSW
z&IrRxi7@YW7$`;yo3f{!u_V59VLDlU6*mz+8@7fM_!=bEvVE~!wU!3P;HrHtP`ovY
zZ?A5<U2g^}@{|to)E`{yr!D>TPg3oF8HpNkSbnzMyl<qqdz3Fe)hktiwuJu8rs~ja
zT^Jq3j{5BFS$67G0MPF;7uY7L^nH0D(wXw3B)+uN-jr4<^)sg#=jk*0H8-=o3fRm<
zRz15TbuU0fIY%1T3|EmNiP<bG|2T9*@VIAaD%UOI$Tr>q?>bh6{CH%()eSvQWAFj+
zeuv5zBeY$vdT@(;eUCj&?DUgMZ;+mgwo-d@NLeu4p2jfjQP!uCb&>X>i_ayZy8Bz7
zd0BH*nhTXeB$QpMl3A|3mMVRqAlnBU-CBy8#Nzz6y$5;r;@xMRs#K<fiz`~kZZs0=
zk5#Y8rtyumKcT0XFN(%LKf+CmV>|$JvYOOL^L-w|3+|b^ud7o<cnBAb(oycvsLlJ*
z8emJxbnmD(wpO@6I3vC6v+KSi4>sOk#hG8ZR1~X$7ncXY;GqG_cfWlmnLGOgw+arL
z4^q<;Oaf1j%Y>FhJDj*48ufu_>5I<@7gGUg$LF_sOwSTY_b2v0rxf>wQuH$1aFHjH
zg4XYCFOEDWOYKkav(ZNcQNKIAZuHl|G98LC5yzdqhmAbCQ|<yX(4mWrHg^-qd!+Gi
zb`g<BpeXT9vb@>uUzk~H<zVz%YDj5s^nZw<ASS<kOf$y*?K3K32BTRtsx76e=hu(z
zgo^{<XYh`4uT`w;O39P?_n`Y?i(A7gv$1DO>8_X-G|~k9He0VK!V+wK#n~o(>I?v1
z9#54$lJ8g-N(B*S*mpNY8t{k(r62b?@`E4x+#AidE*`(Fq8>-13~au%NHSFG#wBWK
z8nz9LPsdwUd*$f4Z~`J_R6oV)87fra`pMoL+kOi7r6WVo5Y76;=m2~)<<vX<nkL6=
zVOp6|iRs9(FVcjL*!<Y-VZQnMnY!$yN?#qI!a9lzwy7L?>*^sjQ7it83x(OL`WbmA
zMx{%%VKIk9@sfs{Y$m?Ej~~LM$ht@89#zuDk=eU<4+@l@N>b-_eNkb}%LjgLh0H}a
zSevHeDel|?a(rt_8iS%t=oIO=>Opi(m&6oB)GtvAnNN96KHbr)1Io`z{LI!J3Ktqj
zbl$l??YyUd=0C<A{{p)!DF96PIw{xuZ@D@J0XF&PY4r(?J!m|?6>@7-TLKIAHNrY8
z1=6qB2YX&%6Lw$RY9N}1<EbABPmF!kM9-1j2L3@OSkh#~O)(}`+!)Kj*8*1^Z{1^*
zvd?z9{DHC@U9bp#pA=qE8Q;Y<q%08c41N)M+|tR$j!6UoMC8hNi92b6Zl`mr!RcW(
z_otF9yOd9uPUKa%QSM#9SR<TbVxoU?il^t`Hf1#FT)u#xlwl#E(L%2&L7!(&^27>k
zC_(dZ_X|`@je!$-m7~>CzbVRB0wh!ka=k9|pi#SK?q>KkqypuvlP+1ZBGuCzHavsW
zY4-s#lWrfB?oKYrMvnc=6k0qZ9c24Lf>Tv=5pioD7&;!e8Tywu`yP%FrEjudrSNT7
zcUZ&cj*8;5j?1!VmGZr3p=^oR^n*-JIf0zIhh@(9c&6T-K&@P9aKdhH)e{^{KifRP
z&S_GO0Xp#HKEL7CGeUOu6L=NF-b)OOpFbJp80+rGTe~bf=;X1)7vJgo(`cdQJdaTK
z1?>9YWPVDoFT{xbLJLO~CvNuKXvlANtLlT6l**F#_OHw$FC7Sf_$jWF_*7VTCG&Ah
ziaQtXyw2?p8Oc`;9>3H>-Eyu9d?p~Fny^w@%!j-95+PL-LpRPv%gKdv)p3XqqVBXV
zYKzpn`%2o|aQNLg1uacFfXE`qNYs%Jgq9Cl8l78w?Z~E_FN~fJexXuBpHH&Y4kX8x
zMvs;hR*4Val|x&)M9#?=+K+?RAXa#d_@Y_H8JS5m$%JkKetE8e#u2(^8A99PN4UW#
z={nP4VGX%hJ>u}H)%Zo=`u_0GFuH;`)USwr;(xe$GP^@S<HNv8#r$2Kv+#c35ntGu
zwmkbHW9v&rXTDv5&11CfN~X2l)Y}z@53Q@&3IYuQRgNU1^IyZP%Leafi9P=DF*k7m
zzpog6BmQB8u)3pNr20;TKAE`Me8~D#W=iP&Pp^9d-nz=Q(^*JEb?dIiY+0)FV)jeP
z+^#+^%_3Doa^<fBco3P*-=OrvGg^MiRjqFuFP&y%l4Gso4yFeg&{7J;JB|mGe15w$
zRPxFB+w#i}VGn6S+pZW#ilym_w5eAf%v|T4Vt#e53Xoip+v&iWEUak%AnS9S9X)@J
zve}RE3DHq2vd=Qf3&RF``Z{weLm@rBf10le`4CtWC+4X9gX8y9VHCEG8d6b>AF(;(
z)dB7R;Q(1D6oUs5#n#K$l&F<(XJ{O)z+Bwc7Vd3v>g417ZzKfeU8g!8-ZpT9*j9{`
zwDN;`yMbnVjh;SR0V&U#FDX1}ziS$?+Id@>AbB$38US8nC|$(PbU(9b?czRHnflAI
zlARKFkD5K~h6x?6pIR|uSs#U+rcxk(V$K@Q-&eKPsfJYl4fM?MqHO(E5(%yPK54}F
zk17EKj-1IRyKvGx<_i=WlK}JYA2Z5Idq`^Nq-OftJ^j4ZxQk9}m2`P(0`F3DLnd~b
z5t9o1t_O)yVWZg*sC>RkhWnANY8tEY()=!UWhH#YJt@B=h#P+wZ7_Vs+P^HRP4P(w
zhKmTn!)I)$1_!54{gfjj6tNL7Xum)(|BN4%>W(8A)ZW{GQ5$ga22T&xNnO1GZ&O$Q
z!vjm&{iTytX*!58_A!O05Tr3xpjO&OvLSW2x0*4qM?o3*6LtovVIbCxbxg)Yu_3=v
z<d)T~<qHdYzu?p^aR)Z8{(las|Mk>XM^0^KJej`(D<nB<{0%i0BDv$Ac47@M4W;it
zVl$l*$OYMV5J~|0Qu;jFRh5UG_7Dpw+MrdRyLmI3<@gOYeDY*g0-Iv4GguAm$Ei09
zib=KpYXFhunyux}e}7PlVTysF{0YT0Kvc?dNi@Xi&?p4({xP%B2?2J74hbw6j&Y^;
z8N+Ibp_EdwGJy%^0;YK8je!=u7J(y17Map>6l!R4`l>!tRqrmUBRP{Q-g_(MHIwUf
zdlUcAiQxw@KCw7&5PU<`N0nU}FzU#bE`h%JW156E^$C>vHAa9UhVIa1n0VdXn74KT
z*9;#zZ({WU;Zn0xf$MfFjE^|IW#)Y!ecTidLSglPPLZHd0q1ps!)9ivm2MRvBx;{{
zEa!@E1)7GHWDv9%&okOv6R#$`6>fTw8Hg1(j;4<8ByWg}GiQDX<FVSK?z%G(X4S2E
zS5T^{Zvi<HM39x1u24V9NMhaKHo$0Oq|?lI08SegBFciZ<ja#LLa7=9N<%9Bpn_Hs
ziQ4kcbIDi9?)}ahcYu-c2Qqw?WGuVLMOEFClxWT1cV$4D&44NE>5bJ@ABiG&%CQiq
zA8*<it*iMeXNe*aW+}$rjND~sT{5w@zmS<VT~Rvl;l|TiyZYPEeJai~AI`lowOXHO
zPk;d>_HpxjV`-RtMts_PAXt@1G@gRBRg+_Z!RX6Pr7g=zp$_9IV3t`BS#`+Q72Z}E
z>~9%dv@|!3hj_pXRN>q~bz;dE3lmtan{hveT1=hShit@3a@3;bh-;Ruwo9Jbd6lfe
zAkP_lNQn|*3F2CP1(VPx>~C!f68nDw9!GTwra>RZ7nW71`^Bup!<lrmR;mISGbex`
z9E@!+>v)iLtDb?O_ovr`eI{Ji{r%~}nF{qkP4=5kdmVF(@woPHJ?Y+&BBrNb73{(N
zD%vCZ9BtyOj4*N9z*nR>tsL#qqzbKCr7@&sty)L>*1yO}#%4H(9p;~E*3UN)Xe#0_
z?N9zCd5E<^0$1Chq_k_!+FAw`Pp6*kJ%K7w+Ui$9uUqAMV}eFn<!&`sLpb{<Jz@*y
zf=~VCY93i<yEg&tjt*jM?J^mJcH9b((s%EKn)J%e627$WDZShF-!N)=XYz>cQgK(~
ziP9GC*l3lX!ng-?rlW6E+ERE107<Y)DEuldZoRy*weY{#JMVa^`~UwViX`VG5n0D5
z)UhRd91bDbTOylOM)t}$G7cdtBO`l{jL2RYk?c)2nc4ha@9T42*YEfFe!o}W-}k@Y
zAHUn}`onc9j`Mzv=XgBs&lv(CPy}5A(wy=T{*g94HFM-eE($bCNRQAkab)0{mLaa)
zW3Ng~vIeQm?xN*$o3-Sh@9g{J%D+ky4+9WGl$bT`L5E(L`peE#Z7BV+1~cWcX4P#C
zdiaJ&`-|sy0w$3Hd`fq061#`=IE|2oIp1dyRCe_WYj=f^p+Uw&IT-_a3EPox=fi(C
zJhi&uI#dZtab12&kjpTJxU}Izq5f<1o6-XJHLH2a_(BGX>XuE=oErlo2g>b5yZLZ#
z8_>qZ%zmOZeo0HfYTcL07fbBbDb3wN$@5X~*cfRTRQddPvo)TWXLe#$EI1?RxRW4x
zKK!(k@nXx#*oq{b-Fw9yx_GZLG%gQCR`ZZ5x&XShp|WZ3-NQ!VI)8nX8PX&0-h4Q$
zQ7jJ_Am<xQ?cQ7a7XrEcq_Q&k_B$6nE#pKy0qfv<uP04HtCCr)eGa;Bv|b?9c>dn1
zQ}tM3UPy#FyE5#%ULAG0A6;`fw{FHAWH2B0fhlb-?<&yBf8?jo_v@JvcJu@p^e|Y&
zCP1=o4XX15N6pxW*`+z}O_vbWF-OlHdlc6idJeH{N0#*y=+--Q@eT4~XDoR-Ii;Sp
zovQl%{0x$<h9nLE5}VEt<}0e)@6vln<zWc!&N2}HO==i8n>14VIC)B5o3wZS&=K%j
zMuB3olVHBSo*?zH(vZ*!j}?<u2RJQg96y6IxaBaRXwh_~fp@#@V~Y@xyB;VnHL}Et
z0g*C|70EX7sNHkDec5Av51vQh>Nz;98qZM5EyzU|Hk5DJ(aQB{QD_}e-ks+f3;M~o
zFB`gvY>4yKpE9e!?zZ~9pMF$GHlIcOojPt=`A*1gh?$e`!u~?9YvW+wwE>3-x}xXn
zW;N_@LABq=ef)HNVA7-~$LOSOxliMFoAxjk(%c9OTE_p`axSI!60c+2^Cb+#&c-JD
zV|&~((S?PexzUWCHR2<z6z0}@Wti(kLl+}5^S}**OkCN$O9!AQPE6E)Bo6lXW}SD=
zg%i-#9bndb_atK)wiik9%Gii!?%=~{57>OQ2`z46;P*laNjGWN%vmO8b$k)DUAOTc
zxs7Ge;<I2VU>t~YiLtM_b?3sbB0fP~UEw~}V4AfQsC~-NEq+zPb<58YaZ#e;+XIP*
z(3s}$`mY6WpI`71L7fNP9*oP?_b{41PwcI0?sGETNuWQ>@V1jpa>@?}TYE{3as`@C
zaLD(yZ0#M8zI>+h30l>`#*v>)-<7=7tqPNZikU?}v-pfJZ#Jg#`4Ix>yk{|&v@ph<
z`$K8$LGwT=dyBA?SM%r2r^QFZ2-@9S-VAe(h2i}#I1r4!>cSD_oYAa=Zjk1B=;xoe
z5+;*^T1E{|$v2-<riljxt&2j3WlcUfIN~^RC<!&YbnT^P(9eSQ2aK3UDG@MgHm-MK
z6%0cP#8+P>pXb^%B)1N$d<m2m1ZY(I&^6;wx?S_v3^ybV6DZ^u)}{0!Ih^BXJyzk5
z!Z-+dc%~hqpn@p;aQY)0IV)D;yc#}g6gs=DscjOqLDG3()P9WOsHL0qad#>0-KdNe
zVF)#I+o&NtsLWq{2?DDfB3xR8r2xHKs}h62Ea^EzFa3>)x+5Gv2P#)d-x2xxx6TyK
zU_Yc%9V6U#+Z>7}=rv}kQ01K*SPg41TSxPo10yxo;MkGlq<!6lM%rsdnfMih(fyXo
z49q{dhN!I0rSBWO^4wW0<ysJyEQX{wT-SC&Jcn>S3B4+yUo-C@X?1w`S~#8Ya(Ixj
zTak@*kRYtY_z~fygH*qiLcW~TXujJyse?8jPmyP127_2P{r|>m#|-3cpF@L3nEy<#
zZD3Q4ix3f5*#$G#aKTk{`O0?-P_4JnK1UZTvitOl)xK7+IS7ai$q}P)^~=w3d*?8v
z{!?||aQu_~>@W{*VMufWeXA@O8!jNe^`cXX;dlJY>7R)UH}V9=L=#Erh`b`*McGJp
zir0)OdBW~+@~>S@apL+Ad{Z*mS^<6y4$ssU(Nn4(WkuQFCM%&Y7Qc<wCmetg7@|>v
zQcqB4`)J^<cHIJ$AISpiPg@JGk2Y+WF}DrGzS>s&@RKJjDZUpbb{B{T22jcyDFs(o
z^KMr$N|Go!Uq^`K2T(iIo##TAY_^RA2~bdfu-W)=?^z?VkQE;h+eS96aSgYC?X8r^
z=PjVsyv4qcyf3uP=Bf1YG{3_+R0L{ob5T&L&flAD+2>NaP#l;mR%0D&TxeMP*q9K<
z;@sIrQtr9tHy`&_^*ltM-4<&tv4T;PAE`y@@1B!Cl=C!jRAxKHndZl1oxjT;-_;(B
z9xm6M#AXIzQcArmlk#2>#>O)4!{!xYC-WNizTLEby|D4r%kU!IYmsfzO{$;7?^3_A
zq->iQH46;=@Z92WJ?xGPDh$8nh^m%q^WVqLq8h)T-`M<WGdOW9aS^)Lt#zdase9L4
ztTu^{NsQv2fm^|2h1mEF)6j2pz6ZfXr=~#bP>=`4nD};ae)tr@`5~CXio&5p^osVg
z<USyGdsvh3?q|y+`O429?v+~qjk!t;f-usdrEUI1?ER^K+w>9#YAdI&d=3)7J9GaQ
zG9CIrtXZy0mcIX{_X+j{O3oW+EovhF_U-@jas3DJ*dhm{Tpe`E{P(ftoU25_AfIsr
z;3Mg{5$HMt!KS*rd^J$CjRN{<fIHpyfB267@drS{|6h%_GD-d|VdbQSSAhE7S{oy_
z6ezp|T=)Qlb_AbH0d@wCJ{(=pF>Otq$Y%eyfAoK(ME~|Wr#<v53;uE`{`XHJ8ejBy
zL+THy9ndMhlsdZxgm5~%b^Z~8|Mac?!|VTFm;3*uF84pIvETnM2q6AzarP_*BCDDK
zY;wLLupsgS-B(Cz#xmEePhBgWr*%Q?VYVVra6xw2&>-~kb;Q?5FE>S1Ucui>oF6g(
zZ4kR}R`%aJ75c)E9uJBapdwibAgCriUjnI^MG(kR#6K>%`QbkWUbSj~jk5-mIg23@
zc|)A~_lIQrfR}3pbRuu=l=Z7XY8z4qJO3Zst)VnlO2V6{yfv~10`KpDp@C;)J{^E{
zI10LH9`dZ;4|^`x8l?oXWD5;+W8ZJ_LHNxzkjQ4Qv_i=YBRbSNW4MY1oopH3rk<z&
zyH;UlFgRzT-m6Lfz2_RE>i$n*Y-1U*<sK(uqZds#eE0K#BDlc^XC_VKZA??qSXvno
z%2P#v{xx0$@}~0CP{9sxl<*x3rP1~i>OsPYff1viJFBnR^D}9FE+l|Yehf%2JmNtQ
z<1i%_f*f^XL0Oj>WlK02!w$+kPac)pa#C*sh43(_$JheN#OE0<Pr$N4=s*{;NrVEo
zTq~BA?a?+cHPk~&;<&HJKsr+d33V<DE+&T_0`V}NFJ<KJMdfKw_O^bs-#}pv=#V8K
zb{q8o!pvdd)tgXGXk44gwlS*MeY8gY&xak*h=aq2cJ^-sEHFDEyutfyq6G>*AxIs&
z$(pT$e~QI{;{cHjiz`Unq@KLOd02k;X&z)?CjAAFj4?DoX)yb1+GQm9is4fm@NQZ@
zRqf9NUD(xDR|~DP1~S)DXd2yHEu%oT?A6Lpz{33sAfOqTG+8#24v_3yK!wp&Y|q&N
zgvG$f{%`4S5m1Hw<Ok+h$FW_o7BIVu=xY{^KoeO4kltOPKww*bBt9o(Jq>DS*i$m1
zE4cK%aL`%03{5vf#H|4ZWaFk_l*@Pi-d{M2lD}Ili(X){(A4#%|GijdZ+hF553{5U
z!KTsD_$|!1eP<p&*dK-CJkyihGlLceLxUceXJ7+#yR@xJgR|!&a9F(pkI|?>0PeNt
zq*Iq02T9f&kbmj@0?Ob1+QaV!8Y*T5QBUydK$&Xr1CXX0<WLXnLG)_4T!AhK)b`^z
zJT`;Mf$H-kXai-Q*8Ftu0TU9Q=z>{D*6C3i#Q+oCE^rDS^9WFYS^yI+$113AK{B2R
zs4V(eq&tDPRx^HaPP^*28PEt67gnd)C;WBu)H%r;f-JKflJc;C>s+YGeH4u9K5{hg
z=Y&(<#P<N67BrhWaLN#^hthG^Vrm4*aY6w|_Zm<?>4M1s&`3dI>aHm`?#BB7+Sh8o
zJ_F(U>dVhZ30+{*4KHna2-OZXIf*@7PX6m>{wKM~F&3C9HaxBCzW-}ydrvERnX}V*
z>_4_Q0#a`*X+EGRRKX<{v7X1N;UBDle3Dmm2NItG;&xTv<%p4W0IbK#BK06>_z;*-
zW}EhMb_{S9ZQPDcJOQ2=!)pueGYkFF1>ngp1bAykM7`UxYEm&|9|Ftfnl5NwrK}%^
z=qZWaLR~vNUJ#?}tpJFRJE=-wPN{uqJche;7I?tzj#&;Ba%54Q5Sc6(BTWZH_-U!B
zG~kH3T?6!a6FEeum=~`IQ?;ww09E}f(EM`DRreO2f#XKn^=u<RjMCu@m>r6G+}G{+
zK|6mG5p>%~o2uELnq63&O|B)R$O*Uy5GHg>hg0V-q$q^sG&nW*loy`G_PDXMxRei@
z5?9l+U9pJ+GbEPdngnV=!L-a+uSe`6f9ON9czSZiJy3rdE3*m`N*KY<x)Mu)p)U5p
z@p*px8j5sCN2B;4=A^rvQ(WtQe9;$x(C4!DF_+tU3}kSiZmkjzO5UGiXFDOEqOf%$
zKkzOkb<qwNuGxpGNO*W`P*cxWYS&L(Zxm*qH!o6KvYKiYZvIe6fOugdIw$|=Q03V4
z@@+t&Xf|!KJth+6ezf%~N-S}*=jb+NozE@1p+N{;TwxEytVBhf;;bs0#Nf^9jVb9u
zhN=L?DGsotPYd0b)2Hl~22o^+S3kQRkY))xKh@YM**^gycYdvgh=vjG5F^#DM4?DK
zUMtUXleGhIRu?Cc=pNI#+o%_2SchKey0jjA(<AACD8u{y-P2Gr?jEg1Ib64XWwdl;
z*-XiY7^ByeSjQ3NPrwtBfWpar;qi6iL8~$6v16j{U84ez4ZZ`>!pSV0jaISK=V^Vl
z)Be=f&_rwviqCSk!j}D+8;qY;t`hg+Y4<z=qWX=feX{}10kssOVt?{U+^t`c2QLTW
z9!$b>Tpl5R1x}6+;qN(Ej*g`we%J0?gCcExm8u7S2GssBI1S+dO??xe(~X|l;YW5(
zzNK55JLo2?0_`rEkO79rFsJ*Xo!JJmr2!{&kwhItwH^WXjHehU*y4u7FHra-=^X;S
zP|71ryg-1s2{_I_f<t5&raDI~2}|^Z1i<@PJB&=l1&Mxu%)GIr2xfa^kJz{oexQTy
z;n`QbblyLqu?dWQ>yY<tz`*qnR+3bt1U39RG#3~0-}!~*YQ+BB@ILrk%)6LMi*JSj
z<cD(?M0pp!0Ewy*a6Q=fpX59`Ve_SgZ26~eMoz&sKeBU9fMWA;Tx7(~xu>Pko~7r)
zDOhFh0^<wlX&hKOXw<wSoCjAQudhXEzyeSH(8Y4_0`=s+%#gzO9^^p4S3)jk3p}{R
zCTVpt4JHU6fcUDxzIf&NT(4X7$hsP&+^P51r6{FF8bFtE5{B4NPO9zYK;54N<_)Q^
zC|jwI{PnE|5OHZ9cbDPQLHEmAvY8>blT}8rGp8O*{?Y2+m@6<gfJPGY)wBQdD#W-H
z5veFKvRJ*vixD-_gxn)(T1|<uuXgCF8w@PYu~Na=qoSV9-`b@w6$VO=M)Z^}%q2{P
zkG<1)h_RzlsqV%R$J=Z-5=0V&@pgp#B0t45o9(8(6B_Xh_JbgOFW5*w!9|+Ff9*zH
zZ44PO8mKPq4%dI`k*o;BR)LaeNbRQZDwPa`eudZ%3^Cp7UAB+yLA@g2v*gsf16GtD
z;rmL8_Y;7-OqN2~#IfUNQ}6>s6s5p>b;M;*C=7DyRqqS5go%Z7bdG+LbI$vnKzLlB
zh7tw)tH=9SF-u)>t|2L=4)_E*ioZR-<O;O6@eN4W5SNVgbyd_(r(%;*9)3Bnu7LBe
zdY%zF24TECKsHKZBfT0#!@1C^zJ$Wj`hJ{jJ06Vw?or)VL0vTX@F>oRZhstXGQgR~
z?1@g~#auA?@m=%L9|cJW1l)ITt0d{<_JGTTa>>hk$E2rkenB$s&EV`wc<g8O8jk4Q
zYU=pKmhvx&dA}!#K|x_x+uxAZo$I)w=3B~%=4@`yqz%z(f}$P;mGf~7NXG1o%ITP8
zdteqr`$j>nCI-AKz0<H*`U!@a>oa1$DBvw3dfuTTe}~+M#aGd-j#;u7(vDN$2DfPf
zZp;(6G3l><VZ`t8jBcbPGMG<^Hu1S(q_4_K@Y(^U4w%iXf6ZdH!VHAcm_<!ct}`P%
zfK-#1e7{BPHpUd@!C=%*oB;EpzbFL2yOL^Ue@lRW!TMl+7nHAr&j9S5{>`h5hgT51
z6~)YW*HYSwC7<x%N1ON&J^)(n*N{!5-WvQYl68#hB73-%A#N%#cIBEhRTUqOipU7T
zA{kf^eUm`YOslX+q5O>!jgm==oohTYB=<E2_!^%Rpr7Myd6j7z%s_V%_yCh20vKov
z#ZtnDWd-he`K*|GNuX>e;GYavzN%29i7(PG*Dr=oKuGX%)aEY4Vz6@9F(T_EMv<as
zBy$5d08P(m#OVDR<__dWf;@hbX)zz^4l6)9)8eKF<{lw*Zia^z%BQ@2+|x;d(vG(-
z4rIt?yEe~kd><u5$@u|OvVjZ6jlUQY#k-WR3sBFmKcbt`!R{jG(&$u9Y;-$0=<%dA
znptJ&v<c;zgui)C>#1^EeV@wLg>d}=7-WWNlPOh6QDnfRc;J7++}RXDSvQ_VkvB!A
zWp`fuRR-$nLm-MsqG21KcfZzPg8I_QV;#;_A7EG_c05fg6?tD5?lMp8=nN@`a4+0|
zZ8KC;6kWe<A?72~61{@S_M;>n&6mW%$dcePhZDn_9=j5%z>*sxWI?Seu*jFeu0Uak
zC(^F8Es`&sKbm~t{4t_jPXip)qCy*W=Ng~F!?=4W@1{gD82F!Cxv}yEz))a5T}C1`
z`$8!9A&!*i_^}uyoR5NdM#{InS=OR?CHr_h5U23<fy8Jv4VLZ1@p|LaCIhJ@tB%9E
zB0z)~;<uvrL_G&4j;7<Onh#pAuC0764JXYX*!vq1xnS-Z;wRHS<%Il3V`o9`+!~SR
zc$bh-)Xy*4)mx(vY*KVbyTw$0JF=3K9`}qg{qHM{vy;$CNA^V|3k56l+1Gb^^s0mB
zD6*K&HsbU$-yENN`a;=o{NR}MEbO|2O=+)8`?tp*etPUxx2O7lM?QbAH$#OTQW2V`
zbBiFBdK1L_23;kGypo+@+$vVfLZQTjb&1xqe=F>x07SqQ8|=>6&2y5`b2EhkKU4DO
zvR*_GQ>UKvOFiFwT)NpO&k-^??5X$UtzO+5;W(WwY-c-x>~Pjl6(_y(T=ASy`y8SX
zKe~8EvT+VkU#xnL_Em_dmJ;&UecHh6X%`ABW{N-pc8YB#iUVFf)$h1xhDmxGgAZwX
z>~ZxuZ&L^PQ&$O3Sv-vh8h%3ziwJuE22Gxm6S$ccg3JyouYhH(yKP?36Uo+jp{%);
z8MXu=k!YoFA?>&fheRFUa;Oja0>&K_ac_TTp>SfRU}rjvA*TQmyYiNO)7GIXde%yM
zj`@tB#pU_S3;G*TdQW16*Q{PlqYlB?h4yLx8ARTkPtaV$P>kvUaSRr@E>BC&Sy@5v
z$@VKJCRJ?*_*Gi9$nP!CEr#=^(i|9@9F6Z&E!RH~e(~gw{&LU>Rd!4F#!F;}JW{|$
zU~(15qlLh^>%rX{tx3}_H?xo%xk6ad#+y~i9II*TX=imGA8R5%0NtM37D0C>aZ9|}
z)t-fdHIqr{=gMM^3qIfHEWS}^+bKJ3wPEVNHTRM%Y+vxV44ohSYCYLp-|mM_&(Ob3
zwc<~_-cT^5wB=XU)I7Tcvc4BDiY$EOiU7e*&zjPBW@dGU(SP+L@%vOo<OAz!lA2B|
zTT80u=keTBWcQ_O&3n0~gV8i}(X}5|4{ofKlH?8|ACeyJ-TF#&c_@>(HiCGYYGAK;
zp=wIxfX=9YVsmS7V`9-$;LS0q=vIK$k@_b_6>VsF&k7a=vRMnc=9mAS_5BxLe0`0R
zmG~DmR~+9Tx-k$}H#6-Nb-P=2m8P%bF=CP5cd8U3U&X+R^n~at;Q|c(+lx=KZoO+0
zBYfTt3K#Lhhx(__R=1xIezV}mkilJ@t11QG-!Qs~;-cls%BNM5h!Yl6u>3CO%L^!U
zaT-~i0T-(QltV^p)nB&h+xkG+owIW?5QbojFoa0rN5w5psr_~*BVsp9z%&pw^O*y4
zj4^}PmaNU4ByX=eRectvmC-H1Ch5c(zNUCuzJecR!|iO=4XUCyu}rbTRr+dQuHj~(
zTND!e?z*aN^}nyb>TOv|e_5Pz$kKyb6m#i%geji*VS_Mw9}`oi1*1pINIucIt=_4;
zeB%<X;mz9f4=^a$p-XnJ(Os+0MbF({XV4VTP@89{N?gLbX<Zz92eJC(lQ6=(F?BPK
zwmwhoQi6VPW*C16F;EUGxW<!XuTq{4f9t$k%@j)d{f5nO=2@aOGtLg-%OVyhrk=Oz
z&i%ubWLW>9XZq@=Pu;HII0}hnQ({5gyAkJXPyD7)fvH2V{3BI$>db(b7+~;ag^a0+
zkR_#o&IvVI3JO;KIOC~N<;?hL$cg5C#!J{jZP(A+gt_hC?Cq-Ft8IqLrmNR-n1!r1
zv1AvntyYfwPO~P?j&~n-_0iB@>%g!APCx$i?a24kMLrwRC1SZ689}+zC~^v4X(|Pk
z(*yMQhpe%!J%B7Otx4l%_Ir{u11DgVMy_>b6x(mArqwdJo7dhh{VEx585voiy%`-V
z`&qW5>GpUIu}3H4DeAR0g~|0S-gt7Aj!Ias9{t0(bvvjD+&3P6ts+9wMNs8RL~XZN
z*vC&%yy<D*>>0w`fy~*`+2-R-IqwcTSz^Uyd5W{^za*+uRCHa`3MAY`S*7H@$R8>s
z4(Em~Uv*6tNy~exp+^bJa+1rtZ#n3FU!A{xTd(kXR)h$fX&QCQfUL1X+yj+T71^cL
z8H}t%oYKK^@5QcE34f8Y`=fIAt@_xjAPw>c5ISEim$U@Ve*L?B6A!qWuL6v$Ihlei
zsZLAb>wEX<<c1XvANRel!ts?N`2lplTolyNA8@)KjfE62s533a$W3=0Td3>hQu6=w
z8PkaV-d+C;L|q)aL}fuQgpjgC@$@wg+Vx7|<%6jyG98`Ad}+?KD_Q1pC##71<E$!o
z>(*5JxF8c91I?GJ7sO;+14$yv-9YI1)H#Fr;TjkgK;+ri85&?2T5i4Xsr2Se{Laub
zj%#kDw`J`T(la;=B2N6zWhHIq9H?ce9|&2$+sRBbqcf*zAK@L+J?61UkyCp2=%cR4
z$LIs)W~ZIykMX-*XR?S1AuT02p$Wz7;26?A610)3rNtUjSy9;qUfucB6W%a*Mu?c8
z)8-JNvN?S_v?u0)aR}?5MY06%U77g4akKP7yjjAwc^t{!ufxz7b*&#qFEq?<@+<q<
znDu=BZRth_0#1<AB!HoiPyyLnl)3JLqG$?&<$9*m-1SlWZ;E#~GQUhl1ik=)^Q-+^
z@nVD4liy6^(06&T_xY2Z@QNZcDEP5YO=ummHB1WVUcy;AIU6tif@)3XJ>n*R>y`cG
z!4c&0gkNfC)`F>v5zgNqFXoke!TUY>6VbHRsUG<erX5zY!Ic1d+V6ozX#qhm6$1s#
zq}ezyZ^|A?UJYppJ>Ofq0E1$lQ<ZIvAaitC*51yvg~)fKAk#2s&ntc7%w~AhW$R&_
zXfP_QlI}d$D)82p-iHrAgD_Z!9G@9u1yh2?GNngzd&0OF*5_R2MSPfB$cl2&Pq<OW
z(zw>-c9G{v{Ba7&9?e>4kx0;863%<Un);N~kqsj;<%`qut)AbVY1W5wD@^KgR<&sp
z1`ftr1ZxnSc`4CagX3t4H9(iI8K(CW&{G(GZSpRs4_^L$<BpBOv>`h#y^n!!N9U6F
zMLZF#j%i{sFaG!8T>#YHl*1WXJI*u`1{$Q{ifbPeFg36DOh31#7qlm8k2#L`&zu}g
z=DL3RwrQ}$>ahmO#DrX~r&OpC$N>N8Wub>6@M~`xG#n8%NHmpV2t_(pl8JFZm7x1@
zFi;TyzJ~3}ep9`Ld24|3t+h0I;m>M)vJw$J4pZDkx=*^&95d%u!R#vQNszr#;j?NF
zR>5(u+<R><greNJny0OvJ_@wb_55Sc8ev_~sKqiX7x19(>p}wQH8A$=@^g*685W_$
z^T_<Y(mWDPdv=`T;U14`2+kA+6;C08^H9bv6kYC4LN{1~aY)4b4&>GCAibCtcDKQ3
z3cFjP*lu(Z`eJ(b3$cVnuluqpGTGwkB^koc_yNyQ2#|}Vn3~*-^K$_;qG)$#aNZ(V
z3a4Y!je%_*f&iM7zx{lfVGI<<4k)9+HI$Eb^VK5iB}mal=v)b30?aTT1kRRQ%FYjY
z@azx$b#P?qVmUCHkmA{m!gb=O8thvT4rfr}qD6sjIrV4RotFA*I1t8nnn#Ck9(dOI
ze3APU2&K1C@8eQUY*=Znm~r%gqR;4rNz$g8fQt#QCl>@GOgOwb|M-B<rx1}%3)pTq
z3kLe%zp9%;-^lt%D;o2r8l>W_RIdrC;Pkr;hsHIc046OZD=-BApFG^R_&Pr>4BY#x
zV(op(G-XsWr>yn{-E285PT|bt371VY{kY_C%D|KVoJ-}K)$5zcs>PH!xcVEk7i*&B
zehnbW-aK23*9+Eq+1<?BE_P{<W;Vd{8ITixT(8J*4B>Y{Wsf|kXk((VV5Xp8!3t^$
z8I;emuDr=eQ%sa8U9|_=;=Df4GnMZt6I^6vi16%ufGo6j7*0;w#xAy}E-~#p#f|hW
z00gYzB3@nJw)%}RP<MC}bm@&}XF*C;Ch?KQM>PvFTADj|fYu05U|A?CStu?N=u&W{
zf!tmMfbC+=A$@Jk6ElwGp&r1CUWvxaV{f2g6l!Tw#aSS2ttDo9gU(SvT3QIS;!+TN
zpgUPaw-flbfAi<pQ*FMHuA-DfZL@Q=j-;pUGH1aC`drjgw=Q5yr!98}1wIlr_tpgw
zLCY=KK%2>hUit`PCtZV|KC5%ea@6_L_h{F+oblWJ?S^qF9Q>6U&5XaiQq~KjnO$N{
zkH{dhJtVBC4ij+RY5p+y%;!K<&92m>@tElp<E*72Oyb%}GvNPJUj4MYxrU1+@p9Sd
zhyL+am!+BPIk?nH=*ule*Owca**22{U!XI>+?>OsS{vOXFdIynWZ)Dk|L9vTkjZ=R
zNeqxv7~ai6hR$X<ecZa$<1n~M%iVgzu1Sd2%A{l_+~FG#0IIrF4Zdv3?PHsaE)f(e
z3d{x}Yx(a()?hC_q4%?Y2eT*gUZim#Ui-?hzJ1HqA>_ww4!ZL`@$Eu!Obm>ek9{2T
z0x&=-=fab&NM>jtG62^C(FFuU79hp7b#epL#S+q-Jm0Xw4f}T%Z{%1k*nGzT87%&y
zG!<2%kk{9SZ*?+s7p_RXSOX>UNMT8Xe1A*FX-<eUL4LN`323Wq)~1{UUbL0BeGQ@M
zaPqih#y<O9s@Fc^f6c1EbYvG3VQ<F@hrIX$znp?p`bcB5yOUo(lA4+h>Co|gnjg@2
zO7UW3lVihzoL?FGJL|FF?3BK4|E6&hh?X6C02$eWpQ7O!bRr1GHkS6P!YSzBpDpN&
zitDf7^r4lCXl@Iu=vYt}zoN}TGPI#ehaW7VU{dZigztpeOt&!UM4$J-p~=fj3a9kp
z*4i$J9p@bzn)U|nJ+YnsfhEv#FqR?#ybhA;J;IbF2!tV!B*qCUW0Rr9CmMKlz<_+G
zSFCCD98tBF7#wvCtd7goCY_Lxc!p8%G~`)~3)IE42Xp(Fuv5d&X~sD73B1&UFRZOM
z+uC_*FQHb|9mQH((hIUuNQQ*?mz1B8c?y5cb&9OhS^*ud&sFlU<SjHJ&v$g+ZX&4#
z_-<;|Hf!HG6qChgc9fTxS^d^R(*=>Xf`dc<)YgmBk%{970K;a>1w{le9Jl*hZ%eC3
z(=)=tw_t7|tQ#hss|E9gp9Fd)&VU}vl`f>vTbrPUhFsbvRun28C_HD+alj3>e8@#9
zztJk8<LAK&!aR99{^hN?*E`0B5l~)D5+jG>h%nnG+&Z^;gYFrem!9@Z2L!f*i95e8
zvj-6)AL=^l8Cp%a>2+cXk}v3m6JYrHif;gIVV*0pharvzA@9p+VsslZM_;D{)rJtp
z%GPrBl1*BK5ssj&wF8(lW?co=^7#DtBADeZVfZV+)>LzQOw1{~UNNg9bGn?a?6%WN
zE5i2_mBIcxs8C2rm&SO;;#@?=HyfuL$J|F!yxz$iM+r9ZA&gP#CS=<7qKCyB4a9?p
zP36xS;~`vNKI@V^K|<n(y8<v=qHspNVxK!aEQE{#JvaWqS7Q$j-V&8T480vtMdQb~
z1$IK3OW#7@OwH!Ll^b2l+q{zu8}w{u6Xi>cc4Mjo{`;h0{EO=CnnSAPYTzP>$Vd)3
zcol}E+ayaY6o2A96|^aNtalspiTkL%9Hj)g^{GM^Rx)Tk>`0aKF;_c0CMKVQmH50V
z1yH2!ch<PQPv}%dc_nfS1HcALik0*IDcFL`#h-;dC^jGiJ~^mz8SYtqy=VYR{*OjL
z@9`Umbf`r`Z(%VblvdWD1%2vRct^D05$NRTH&5o_d*w3j8>et50#g2T>h)B&`uuf}
z#8Q4NzEz;UFO+yesR={494H%Od!anwvtcl1`QAf-+g_pJ_Z5#NtQ;RwxrtZDHjgHs
z#f>*O=HD~h7TAVcD+x^d_PtCqUHSU(i0V*TLlDkj6fgu6?fdpEqX>Din*#Ao6m(Um
z&jL3(h1)46HU%Ib+6PSo=XBFoL;TzRWyZk7@VYGW9^t~Lp4?@caQ@xH+OZr(HIq^2
z+TY9IxbeuIeLY*q|C^!+2!n;7OC66M;5qZMME>HzvY685bsoR^T#3`ES7B(^#TV`T
zM9A(vTB-*Lij8&*8aG}_#8o=A=uH_S4YH~S7;U2_dZdw;xUt+%z!i_JG;h}PHKQfz
z(AzLZvhZRlO_7N_N}Jc#cFZ}IKce3j0TDz`e(j;9-Sn%<y7ptnqRNMy-5ekVoL{n@
zrCc~Z$Lm=LbKmaIneWB#@}rs0dO7PtaUqs?tW&IIlqKF8h{JiX{uDjG7I_`PgB*o)
z+DLm|r3=^huAa^I9(6K~H`O2wc_>Vqw~rZ14_d9ZNyq1n$}^3Dt|7lVp?uQ~duWO*
z<;41%A|<&gZcE1-PNi_!Jr99Y=8lj8I%slvd2zQDAhf<r@r~Pm(;n~`k&M@Oqde1t
zQLXJ@gjz&(Tk@7?d}RL;h4WJ=#G*5KDDYx}5S?Pj@j=6@z{<>teQ@3QCtk4*nSuMR
zy50NqN?H0);Q5d%2V757k|o8#WyfU%v)z07(`PoHFUA{Q?gD(}d82{F9*7gM+eoh4
z%0jsUdb(M^#*Q!F(3>+uO7dZEP;8UO*U<{IX^#C2KWq4)7xS3fy3o7wAw!HZX)wT1
zAI4=Q@KJneIZpXGbO_X6baug&F93o&j`-m=7JKq^!MqmHf=4%l&>95_npC=dY7o=!
zg#W6*X^WAfwJ532&)(oeCUat&%#m9m3ypq7FMA(vG)-0{H^meH1G{OdG3X&@I=uSj
zTv}aNKO!+%U%PP^l%O?E3&~GtPka3W=2>;t@;**nJOgu)zw~oKVr!oOG<^vWa@?~%
zC-wucM`wJfAJ9zbX3uN7@q2VO^VWa$OP+7&*vJ{We)?(o1(<p0pVv$0#^^XdQ({PT
zv*p^O8UUM#3;IMouO-&i>pipzBH(!+C%a9RrF2^FPT>@@A>?ek7*^)mNx-Y1WFz;+
z+L%ZC2LR1G`D5h;m(fsk5Kg06@!J`mqU!^s1Ka}~1K&0LcAY@OE=ONdWZ(ON4Ut&q
z5kb#OZ;ia#;)HBy>-OeMni_mfa%-OmcbaUU5E8)tYz$}=U_Y6$D*iS1u{7eP<b=d|
zHqe$c{Zz0KTBm2&HCq!=_D=K#ZBoEw>jS-=8$b0AT=YNLBJD$8vQM1TTG+7dWLLUF
zH@H1|=G7z|QsK-ty^$we(DAa#U^3dZj3@@AJ<jF--A=xO^H#Z+>(9^_;N6EWbp46>
zW!-MGXW*55=hJ2mdh(ZXTY+v}vh(3;DEbqRl7)`&*akJj7f`YP`b=2JE2f9>onJ)=
z=}*EJ`vSog1h@#JcX{O_*$=c9vyf6(F^hgaL-_FXh?^bYVgsXbl-^q=ns&jq=!>Nk
z{BQYV+h{kx3Mfre&d_1?$S*mG8eKq1xjmFwore?H)DutyuY5U`ay^&|j#)9G$Y<z$
z?~%CDJJFf?So#RYwsmaOJ~CzD2&N@a!%USqbWy<@;_q5NKWL_GULg1U_5fu|3r8j@
z#XBQsamDaeF%_WOAHPVp$AiSs<{(S`Nax-5_tY}xmEDbw;eE=5imQF81Ad%5XKq+~
zbuzIZSc<(R>T+-Kg4`#wy+Mi(3(z5#>Z8$#yjVOf*GP2hMXoq>5ifR3L&#uEVs0v&
zF2tiAH$G(zGYHdSFlw5JMMoaHNtOC8!f^f3Wf$uSe8>)4#H=sDnSJ_!x~0@UQPu3O
zUF6Xg0q!L{o%wup+@*S|6fWQ5A-G3?hstY5^{4^~fp#ZE_q6ly4M|kzvE+@i4z?U$
z@tDa2pIg_tX!pugHGoD6_&UnuH%Vyk75s{)aT|p&EkZJ*F|V|}N%yAZxBQBv1YW_e
zanaaaot*03Z(xBjBjV<(a<K<TuH!vmfb*-^Ai|+SzGK>+rq9lijfK+3W<@)rH|~^k
zoVn8D(hnkJB-56|DZ5g`2<!%!v<lxMvQ2Fy^%5zfOnbogFmWh^y6BQ&%kX1K``|?+
zSr=oK<OY_Ii(~!EWYQXArsQ1n{tSnft?<RCj#X<m(8acZE|`_L;JP`AG3o)m)Dgkq
zi<Gs;Aa)FPj?^Dis%5cKEnh?7w7GO0*T043ZeO@GOw~OiCFaaAgJrL{Onv7XYG$8j
zfE2Z|kHUVS@KG5+PN1D$HosKiP6{+6r_=f_WK{6;OU($XCH>-P_kZJM=m2q@mW*`k
zUv+>UqB$}1`S{o7-q^l#<f4V;jQ$cEVc4gt8XSnbC=9BDQRPa?ITRV{sTIpt#34Oh
z7%S!m=EYm0LrkRP9OI#ceSR)IOR6VoPl{9ZHY!cxoloX@h6|F{5Sw>EDC>aNG~+&O
z>w9D8+%&$C%*d;UL7teP#c$#SA9hH3$t5rO0v|4qFC!-1kSB&f8QyE}#2?<eC}xuc
z99&Xg>+T)OF{l{D)@5Y>f`^bK4&jHNaaW7CH8@2ViQm8xYpGP1;PvyOCu`%aF4v52
zv*rrAA1e|-C_QxDSeJ#cE8jTi4w&D8@7SgF)wAWiH$_U7f|hqOp6i#3t4`<f?yO!p
zl9crpz%0yMIPW6r;J||};eVnF%zmyNt5iqt>4pSa&Dqn<HPu1v)!Je~+#@v?BpXyI
z?VBYml;1^>pOFA4+z#{7zQGJLA#SXpB1YuLN*b`~h79$`dWF-UXvj?d5IWVHpE~8Z
z2;PgPz-hS6Uh2mkwpchhZ$5BZe5ZGZ+uLpUV3pF{ER;KAbNrdvnsKmp<9VHPSsB!x
z(IR4_Dy@LuOKJ4TjlhmNzJ+7z?iV*xM_Uj1B)EQ&KZ9v=qR^cLFq0;ct|k=C;X!@W
zGHAN;g#Br*l!>R-S8T1nhdlyk&gHQFhM|Z|W&r6NohaTJ#?$H-nq^F>J~WV4ULP|R
ztJ%rwMml3YV1BB+C0g>YGd8C%a54ETLlZ2j-rv?3|MQ~aJxU&4=+fB!*D{NP(Z?|;
zPr%d|>~943Dc}fD_E2>Q!oA5%BFz0Y^`19B<Tue06(YmALIV%Klft3GwBT3NSZsVI
z72*L{iHr?qET&9??8;`Na=5~xkNb|c_Icx$gZ@z%h5_|=z;{eZwfi%t{hEQlBT=Gs
zL8c2*i+f)tf{BE+>7;G2tZjqAC96ONv|jl37m^OlCqu0uUQdF8gAHWmPkSu0pVoKX
z8&V&I&&0{vpsE-VKNovb8EIaa@hvUCM~JJ<+k#p;oHdFc_TC<v=5@xcW<&LQVcGrQ
ztmq}})C=|RlLpB!Ns^lAhmypKE~IySi1foIa_E!8yu2jDUC2>_eXnjYm--qo?@Tbs
zBGh%n17G{B+tz$Ji~ZR-7F+q!w7aY3?%v0w1#X^<)snsHy<s!^iQMnx{SJzMReb)l
zZqqv-kctc0GgspN`UvoGRRZ*Z2qjzv8jKUv?*&qLvwmm@_$WxuZGgtoCdD-2UtL+w
zO2S}}8>onx4Kfn!R6L(u1-DueNCQTJEM`^qC~l2oChBS;#PvyQ0_7TcDZGxpIA~@O
zoV5V6qR=fUtLOrhjxv_Lt5s=!ECj2N{%|MAjM+j#or*Dxt20U3z5~*VeU|%K&4r5y
zL^QlmdO~nLl+W}&$6>8-I=1le=09tm9)@_~*IBBS{^hl-V!-wIItGF?sT&e*F_zg`
zh&yzO`aqf&9O({xGWvI|E902{_;z7%@r`PbC=_Fh-a?lBa`5Iy!S7^za<s>>#=hJI
z$gqY$o%;zI_^!;%s_*Q-ao2i?a}==pf~xwT<OI~Ng3p2faf5HMRX1SlLi`BSP6?ox
zR9Gj80uGkK0sfA=a|fOA?!$jJI3(Y}CGxU~HUDc3+w0;L)cn#(d!1`N1&{&eP>}e;
z)cs%3;6>9v6#k5eM;af1q7$R?5yZt>Nuut>U4FG!Or8Nqv%j!`Hbg)Cl7g$TzOtS)
zP1utgmJOwud|D60PIiV(Ye2thhZ_&1Kwr-5jsK=r%t_=GhyQmi=zsRb*K>e;y6XMv
z1Iyo@{6EQH|4F?053l2kcm3zZ{D1e#qBp2<|EiNvjrbkk1&GxD@jtxe`TyJZ`5(`o
z9dU5txK1T!wg1mPvH$JFz<p@<_Zv$dg@JUF!+NGE5G=70?av-3Rf^U9Ve>@6bYh|p
z)+axEZbvODz+>=;%xmKDE~x7-YvAu3FKINsZ~2E0_RlwLj{plVflyQZ?+3&yW4N=}
z`$vul^lv6o{+BFP=dOi%K;U@b24>B9T<3*bKK9~El+8*PT}ueH8PF-K5Y>NuFO{d1
zB%0n+HUZ(|egW^vAhLFUMkm1Bm6Y)r=%|fADmj%LV9`eE0yv|5QbzyXTsWXqXQoGX
zRJFP6=GnUESnHxi?Z3OAM-!c9h#)fVm$+eTu?{wig+?G%w#8s0LD2LK@?kr6FKSn7
zuYjXiY}l8GY!z@*f}QIH?*Sc}f4}Insk#SC4|947Qx+UE?27~M?;cy=#Xb0&{q<kE
zCl<q@CDZm;HTBOWGv$fTq8RrCh)drAF`DZyfZFEs&Am4t(+DA9blo<a*T7CM1nF62
z>y+S+NYz!)04WEUiQZ74@opbj2hy)@0NCE_+XZ`%2BPb%0`t#po0Z_ZgaY7zbjh-u
z3j+%7jGfA@E(oFv0bHyTfy5mWM3w-7T>e(!gM;8)Xn4JE{}v#tjtmtUhg1$2NO%CH
z(y8&J+qR->fMd`JG<!nsVfH{WP;w8OCb>H@rc8q=So&*WPs>7zjJj_Fzl2|z6-W^I
zm!+ptIBjF8Rb8eHL;$IJ6*3ChlM}tcq+^QQkAB%y9O;)$^nQhSqL8)I|7a7ap+|x2
zvj5>DO|8d%eR|jVt+j-sGmQZuqZCjStKI`}UwxY=C@B>~rOW{bl=1-aMW0*YHc$yO
z3^!;U{jZ17dty~yoA~S4qC81Z={K7X`XeSB%gcYRUvW}q@uo!ZjK|h%kDeYKO`g$m
zUlDU;q@ll=2l_z#rk#}4%h@?D3WDRHRKkH@2WONS+L&&<IobvjSpve<TlznNpye*&
zLINnsrQ0NI`pa5^nxIBoeE(;NRt{w6I=d>5JWtBjKA)X3yj1`lHhxgy^tUeqW^{Ft
z+Qkr5X-W;|4OCJ(`N82?1U{JR@h|kez}Iqid$(wutZE+zU#&?y<t4)%z+=cwzs9id
z2FhcX<?KKWrM&*th0){fo<}2-G?SbffQOU@5mo*8Y(c(P3UI{z(D9r}2Dgv5DQC?+
zUDxRaB4%12O%@!{ov%{-zyPb|nE~pF{_*4LD>NiOL_NTbn$i`rbB-(imdC-7&wb2t
zK;*jxul*n<XK;N-Ax@qn;aiJBD;JV7B7KA*vec*oPuCM!Y;G5!7v^vS8vC*e+N*^?
zIGDbH{CIeY-Q_98L_G`pszpx{OLf+_e?A-kcye{%0bTj~0Wo=L0jmCS10<Clg@)+n
zi=&K=AhS6T2PZKw(sTWRj|Ugz4#q%q7tEjsQq>`do&~XhqswC3OraPggV>8u07QY(
z7`HTK=HeTyGR}FogKsK8#7<Z!>^Z3NVPUxJ*tO0c)T-o*rA!>m@hhBt@DM)n#H(wK
ze?D2Wm3k=oPHG6k*57xq^MoISx-LVaf18@!0iZ-b$<-w;2Mi2d!jOZVuJp@%2j_8N
z(?6p>ADcw*DO}Ic2faQBwSV^xj$`z0ez*q^(28;T*56l-J#qteV?LEvW<NmBDafoJ
zOc}glF<!Fv2>?Vs`1cDl6L|q1Ja!E*_d_!M)0$Y;Y-<ZWcq7;ZAED8c;1jlb^1!}S
zQUAG~lIy8G0i(X^Z?}M92MF7qbx@*rCI9$)$<c61F%P#iB}nQh*}yPvtAW9hwZGZK
zjL(Bg$31ucN%igsE?!;-8BGOh38un<1cfAqZGBLX{1rcW+yX{i4fdU`c7X`b48w~`
zgeJ^MqQ3l6>>iXMsB0fCM--^bKa`wt2F_2ZF{Y|{sMKVvIad~x8&rvMWyZNCx9jPk
z$ofW&@kAWQ4Rq_*z=?!Br=xh0=I&EHVON(R99c{26sapA1v&Cy!^_O0H22g8e}_l^
z<2L7%#_4DJ^Gx<ml=Zr0k``10R5%dtp<skS<@Ockbr5Hj5M0wrUgok!%{u{QvJJQo
z;o(`2K6}XUR)HWcUoo!LG78ol7#RwVQEzuJZ2(RZv=fFZI)a4QNpj~ZL3pPrG;ZK=
zV8VD}SbG}`IB0o<z9>2Hhb8XQ_5)P<tWMff=3Rs`9PINSn7G04s$yfEsGG$uf?IJ!
zUTIgDb>OVLa%J&K>FtzxK<o?W3ySnl^I*B6U)CT%?+SMhA`AmZXc;KE++mXHD}N(e
zVZfk6$NKCHm_g7Z1_Iokl%y&;<u`E}lD4-^0QRLE4DMs^Q(O2r=%Pd;o4@1d@n$A&
zXP}dP{s-`gX{_`&ln-9PDItd8ZdOm8Jng*WJF4>oDE9+^VA!sgqUF`qhOtM4swf33
zTj7{5V~iX0A-9s`815gc&GYZA1MPr06!T^g-H0agL|y!Gin^Vr5vC+JJt~{U;;Jr`
zR7+O*MlxP5=zg}%JkRm1PQ{VWAfRkoa5=LueM*+cRSia@)p(z~P+ob-cpIEx(d9g5
z+Ml=f9$PG&)mX^9u{5c9H4b9r8=8+i4t_Ak-LU!KnfrdYrV5<LZ@U6I95p?jrqQKI
zdxG116y=%Af76_O558OC@k9Rf)QF^W9)GOF3|+IJp#iSf-GT}+t%8`nw54tV&Ple^
z4U+}(-Pd~1kPE(SMjPQJ&z$topKa~~dh%x<asFfhESCsOW@scm4(SX$c5t3F7km8)
zt?PVxwbFo#G<I;TWGo)Q=tBTb*sW-t16T?-69^t$P3+C(i(nUgqNjHD`=}shhVT~N
z)YZFb%VB>+P2#rn&%h}#4C<)RhzKI0ES)4^8P<ooF^~WvpY0WW_c)+#D5F`386GJB
zu{{GTT>lx+Y7@3+>YP|JN-slBiUz7H_*Y2aVPR#ksPiOLPmVJ~(<IVLE&cinW0{&f
zo)kd-3NV)^$ui*MFhqkvJrPb(d`^R)??a;anJ^4e{`l-oEun8UeguXZ1k7^dZh%JL
z-Ii6LR0-s5YctWl2kga?mr<8chfj`kh`d&B`f&@xzfc+*gu}xUGszzFas-(SPLUH`
zRlf9r(6<ovLBdj^h2akOT3$3B<#gn|)_v6&($ZOAt$6JWr(QWW>r0b!1jSLAo%{HR
z5pn%a#=|0j_-Q>J0+jXT{=UQfPswUI{7!X8AX3b7A@sRWybcm5*xvaI49@hg=wG~%
zjN!`SySuavAbE7FVNb#?RDLw{iD^061-1-SoH~ALZs8g<>wY$j3!W-at}_f4jn)08
z@1bV8e8V{-MY5G?(k3?aBnQArVF6%F3GO?+=;sC8iDMa%tOTX32fS|C>SUvF38Mgl
zSzz6tsmySf@f8F1jbM#RIfKeF7@tDELTdN+o+pqI1w7NqQSOpe$;*8S4VoN_=RATS
z<}dEr1vON)*K{A&0E)*jXef?9-mDDw;QY`j&-B49i~lC&qXG7LgC92*PJMV-OW|?h
zt=b=*&-aS?)&uj$7Y!YY^kb8bi4xIgo?$XEWkE+iV#`yf^d^U3WUU$G<{WmG_7Iql
zF-A17IGj$CdSc3e6?VYNF3@F98{RY1L0taFRq<r-X*r$rL~5R}=WCsUsSObJh}J$T
zdTKVA6P6sVy9cg9CsT-i_Z*8T-#)=;lK*oF;E!|G{4L0oP1~1*|EhpX;-P3H=Op{2
z#ObSO#ke)Ed0_Id>I<Bg!{eF-k?N&hlL$rrqSKQjbScvlmJ~Gxb(jswL9p!oD1vGH
z*ahG-<wJZ4kGm#!X%p`r$tK}Fzcz){g@6M~0`GGNEbMKMF3{LN{C?~HFxP=56MC*v
zTG)p&D=aYBpHrrvCqJ-LCZUaq<_>gW)m(1<K_HHJ$xutb0=h?v46nIp3)4Mi?NJW7
zAO-@?7k;<>4Ojx~X$%=kh!N2TA`5a0*1)SRF^wrNYUwB7d~Y_Hbm#_ADzBJ(7hr``
zl2^s&V9P!ZCXYYKnJ7Gwy_h8lthqEkDmCVo+9GY1PzZnumKsQzY||o?3}&4TMB&nn
zeA7-B!G#fB?7e6a!IgV^;~t<wN<Dz)^GKz2&9em4x-7oOr1>wtW(v5fr9E|?xaEyA
zE-@1vVelY^lVSaFe$VsLyX=HDY5LN2Gvtnt>c3>%q@e9SwN0t=aQQD$wTPPRgjb6%
zLFhaL?lkHv^zeJ(hFZ3Z#A}Q`G+B{>cx^5X@A7Wvb*=&LBhOqMq-j|vZJ7enBbyRr
zL-^v`Tvt-j{bZWwO>l-!^UWO{j(m)AWyKQ=+<jQ!kmaQ0#!^;;m3s%J?<)X8@zNlU
zT7i@%jq=Ko1=B-iZ6zT5a-`jkI*K(rZ)95DvT1G9Wk@111%VE7WHyZx#4(@d-pCu(
z%$yo}6y<aMgytYsv{vjvRLUOz)e-61z6S%E7wu>&T|igW(k@Wvk7dhGu}zXE+DMf^
zmba7I=2{xHpafqbzfqwSKtg@3r`0-7s+;)*%VZ$zk#_67(1m+_>%c&C8@F%-HNgMd
zh3g$kws97r^B2_@l^g(n_M9aFysP#ffpb}25$LaSNK`s7U)-)#h!(!`Xq;9Ac#zxE
zYv+8VUpONxbJZO`5xcdU6wcvkjevRf5$FxJ(VhUHWMgo?UtJ1}XUx$jBdZxC$OUE!
zk2mNb%aYFn=atZL$F18s)uYxgC6JYcF~Tlt@$N`DW~GQY;bN_L_g#M85q}qlvX4E?
z&2y7uMmGco7>$@Ie!L;R{XAQqxt+Ot4x#Dh=*>}yL-DiZtYQadymUpq3fRaZzE}+B
zJDIrYTi>U){q(%^I)!z|ukHHuhRi|fB2@fVvvl8FFH3~{?e^w3<^{x052P$>&7Ss&
zQOH=)s#Z1<49@lG{FtAcL2YHqDbuYjb<Ttzy!j%f&Qbr)Zu6G2TFD$9I^KBdhcYbU
zi#0%FGS5G*eRd`aU|j8~5d_;h?$7UWR(_2HwvpPVdj~VheQ${U+f(saDCV}RJ0^ZY
zJaFlWVw*e<!|_eoD4KkK5AF+kIu9U<>#+0K$HEs9ER|#0<ycbbPMdY2<(?iZ3gcOq
zVfVNCLqBHcUOWhxx8kuAs($cmWKn09HR)A)M)(8d3Jb+_arVsjZ;Cf5-bWBSn2fHt
z-%6ni=_Xp|nEwSB0r7fTcdTfw{N&?`*40Cm9FvJHi>^5{HO(mx6nCX-MnCxmjk4ZL
z{-R#w80qNPN1ez+`@Q7-NRZPt&1%Ec?d;-EnFUnN$>M`ObM}=eG~z*GqR@d@H^w2}
zOrjPH%;qPd5#*g0wKDo~dR*kf@y`{-Kk6_cFnYYOznlKzq)Zzi9L^}|7G$?6>EgEl
zv9x016+ZW$u9NbKiF{4;SfnZl=kqsYwc1mA`&aLYDKOr3wvXdCy1WH9qPg-uqbo=I
zLSG-aq$k6!_P(wVj}IpgcMWU4MJUq79TKTA^0m!Dpj-t9W%|B`Hl3q|Gz-6oBSH||
z+=-GrE$!dB25LT`&&%(0slwjUwCYJx&FFk@s$Mn3;!#SH%t)GD^cvoEGRf~1O{1SD
z+oBmf+b!@+5X;@zBfJlZIr{*BR_9QvX*+T-kZS`I+nySu9)%T-)7#24ZBN}R3+LJ<
zk)EgGG!$3t4LiCjck$kN){n(*AQ2*ZOg5aQv4jM`*IO@i%4WigQj%zU^SahG7Wm1F
zWt@GRQTo~v{5x&V?cd%-w>Kt+3OpHEt#By_xv%nl+0s6*9ls950fnV4>H|TMAx`7H
zDk!lypzet?PZ9gx0d@d<uQG$lwq#Vb<|g(sm1GH8#JWUm)QfN}D=Z>|%a-xFY3tR4
zkzkym<&p}GBkBars9j#cn>uUgmsMyuPP~4>egGK+DI?sD0;p8kufbUSx?qgK?<;?=
z=^<G7*{7VpQdlLHDY#i@dXen6`<4&;+^f_~k#Z<DeG2-{RNtto#Vb+HzEs<pZofbn
z-Yu)BnG<%Uhk$t$geYHBw%$}d2V=g1cj{<sat4-)#u+D2hniSX&Ake1v(zkNx#s&F
zV3d8QG#4WullvCkG$RQj%LuZ4gLA5<SibaV#GHC0udo~U;A$Q1^;@B1KxxD1uma91
zGuml3mF=nxvZGMJ<xGB7FJQ}CkbR|){oOL_$#(QhWyZduDq)TLuMVLg>b15?5bd)%
z+=Iq16Fe{VxI=i;%2aTo6Q<~>6q(x((w;N<@!C8Qmb*Lt(RNanPX4vlp?uX0q>YS<
zW?$uDy$^r_*X^=0)NWdY=*tF+_3>59USHCbzi>QB9gW|`Xck9otNzVw0eiNj<VlsK
zP-!bi9Lv$Ed)Vd>aOwJ$B^9}V&gY%VDpnCM`Ux_QElhqwgIE|%1omW=(|hd;khth7
zP3#qoGNMy4O9aUWw0v;n{J|@ogwg)yPx@fAODFSE1t<$Fn^54*(i(;s3Cl$*Qfc=+
zci9oRoTD2TI~qmKUo}uz&H=`rg)c|*ZhY^6um`eYU7wa@S^qpvRNlY2Ql|4*-;AWd
zF)4yOvA;8o`z3De2&nwzG76_=e^(XEOun1OkO)4@F2C{1-?f@bDhB+h{#aPtH(xJ8
z3rknKKsd$3jbe6KLU38f1nhYp?6Ku3_S!P7=8Bumf0NXDLqo;b0~bsFEVb@D!=r__
z3pWHiXRB@ZS2%)(LVQLl`>rL(Vi{QJ1D&$R`+n)zLb0O&<Ys20nOS;5^4=CyWBuEd
zpID&uHq1Te;VF``QdHZKx%1(Jl1?bEo&HJ&DTK`~6Yf{losmCK<QD~+?&6DJhbUvf
z$w8O^=7GO;em&tBfagrs+y1RjX8axJi#gd*fGm)7+HMPVSP}mq5xRUE_iO4^Dg@&e
z^$go=AffLq@2_)kR^v}Xx{Hqu7^4u$CIppl4gyzs&A%qc&Cu*cWf1aXM+Wo5mcj*X
za`MVwyACk+nMAYI6ek@8AA}GFj@=iU@*?!__g7GnpIbxXP3>>ahB)ize5LapG!DJ@
zVKq!+!BOM-EQYW>uB~d?BHUPCDp}@@D1aG|z@O!n4HSa_?>kg!UhR)72$5AOW3zZt
zlNzdk^9fW{B`@Y_`~*|MZrrY}#*sa8daK5sNF_LsVYzqDhbAQoEpsO?==-uE^~(9J
z9JEgvs0Q{6=7z0>T=bF_4zl9=)a$ISt;A?oo$2wvI6D(?D%bA+BZ?*zndd1K88d7{
zY__Om9zrGaP$4rZl5O0EWC~@744KCeWu9l*C}mERIb;9zoO9~^o%i>?o&WW_u5+EU
z?MtU;@B4Y~d#&&Ktna!L<4HxBKfGo$UVDED$BmDYkHaxH=n`F<s_(P>eN^%s${Tj{
ziI@Hwn|uEyhrBiiJ1bj9KOcHTVTVKYnB25wgH|RUszsKFpUn{#S=5Np%0S=K#wAi7
z19wm;6z^4nFlYa5yl!|}!GwX5<~t$N3k144x*5vyA%fte?<(dh3=I1ep6wAypd!jy
zf115{0frW`t(e`01y)f*H?YSWVcOUN7&YVNkoIwOuR`}SRi<;ce7C}x&xZ*qT8_{i
zI`04-Kw+77?K<qE-k_?m33y~R$My}>Kdn4+>oyP3HJNiSOq>x<7NjazmkC<_+R;#z
zo6Qz$?>Ow(-PkwSsgK1ScL?1MC%rd%pcylus!l_2C2@rAxVp6shSA}HkdHn?pjito
zB}p`C)~KyYLel$TWCSp_G&b?dd<o_GEM2D19M^s}!tVCIvk4iqVJ8z%$x@p7L*Xvc
zMWTEJCzejV97tB4(J_|^ZC^;)>M>q~LUS$B8g$v?^LN!W^JJfIek`;RrqqP_hl7#o
z+Hd|TR(Og<y~$GiDX?8zri>M})?6+QXqif`q*-Hm+z0Ifk*G@jVF$abAjL+h0Un-F
z=pD_1O26Ij02aKD7Cb<+h{6o10n=<cyv{BQ_$^}#M`p>DXbe^rYJ}<nKhhCFczQ`d
zl3ShOijz+X6O+ASi4()vw1WEbSI0Rx?FFR={L}6L*2#%OYnP6Zq>I9yA7j5FT2K(W
zWb?cn6^rsUl7a;EbJC&)brj}q$a*D#{>D4I3^gW)DokFR_<)zQfynd#gdfSYA8Z=5
z{FJ=~2y{n`--*+N^um~@ciP^FV)|`W<;g<QS4~4lYcUw5Q+UJp+C^q(5%>81!iC`j
zF+?}+o}tkvw~01(8Rh~8&G1zMzNVJw5GRd_wAUh}rZGmfS)NrGu?x)6#T`PqoeMb%
zG^cL7cN1-xd;w00OH_SxV9O9(-~ll~l$xY9H2bLKL@@EfY8kGHRSs5x7|=sbQ)sIb
zXS8^US*B=uua!w@HMK{w=lyQOj-_#qOR~sK0e7C5XZ){v)01V4-5eb%-rAr1WMDLJ
z8rqrN>OmiOTxl)!)o5yNj-Hq664k;SeCRy1;ntUB7q0_1?z-N|<xR{v`wvyM%YOF;
z-o5Pj^;=_)2$q@T7(LC(9~{=n1}1v4r>Xa20L%O7@C#Yz9=_0_$VcI?d>$#g<pGLl
z>(L2}t@`U&l+5jAD|_>8J~=n-vdM_j$PBIAQJpm>Oalo{it5uP8M25{w)j;vVLl(@
z2mOl#j3O2nmQT8~)fe;N*}{YZmPrb_v~r*8=oyy)zb&^2Ho=G49&HZM5wWiH_U#Tn
z`eBuUvCHP&KQ)5?yuD7G?y&|X^LcBScdMGzE9?zp%f!f)mW|+0>;*rL_tUiMIt&{$
zc_jxoJ|fNm!sl~uAWl!k$h>p>SpT}Rd7kcrc-U44&(&L_+kg|1=m4~xOQS87w^4rq
zBx|-G?tEXqJaW0p5Rqo1GuX2xAbpkSnkqA!*r4-b;yUM*ur+L0U`c9`S}GCig7un#
zQMUS@>ykJE3;E}|Bvs0~_v0ruh>vgBm@lckoP*#l#4?t+R#j^2x5Kj8yAMLSF}J&~
z;;)tICLzWyH<CA@_bP#!9wEJgKI{pKD0@8hn>H9rIrKal8I_1tOfnHc1U35<9NPs7
z|097dTBJO8WAjew!8B4Zfr;ydZ6@Y>k2K3giaAp_Ps!W^SYFW>T0NVb^X1Q0X|}oQ
zwWKu6vI_*gUu&tug_^s3?56a)itm<+lhReBj2I-l`_SBJ%<2*Jl<t{+nVQk*(c;>J
z1GAQ#YQ!fRSIS=R+OJkTt_>*1u%N2#n(uC69s1u&P`G1H-X*oV-3%gZg}i8!O+?X4
z(0^#v8-q6|cwN;pME1Z-;|lr;rvm)~m(4ZvE0P(GD5Q}?A}P~&(|cf=%3e{zd;I=^
z-k7U*4Ju_P9?E}yaZ?dBwFPOlVZR;{q!<Oe+=M>&o)2CLbbet}^(ghGC(oe{gpEA^
zpw%mlr`~v)r{~QH_=MP1Jp;k2xfu{96R^kPSjfZ7Q9xKuKbRG{Guz$Trv|1tE6E4?
zkZFD$ZtjL(!aczXVOd*U^pwuJ&n+oF0>`d5xRYsUwn^`#eFNrB4@vJKKEVl|#n_6(
z5+6{gf=f=m`cSxl^b)H`qdT%sI0$^`-0`w9yQziAwmq`Z4z*9p72jt6=na;k@z~_u
zFP4;q;iWEVqI{kjJQJ~&_gCwUI!Xs3``l$64sXy*)NQwxBsx)s(D9_u>Aw1kz4s~~
zQ@EzxNfuGPO#9i42`y`B&sk|?Nj3N)r;D^7YGVd-PNmg&!<dqEtffQt3g4>(MeED>
z*-?QMmNRp$aH(rSGXh;giw6dyBRDY1#TM#hjLVf7<>bRj%?<5eTi4Dc(0Kd?c=)3&
zQkm>R1KFOB?9bEyU<nE$aLGpbMsJcgPxmymbIBrOa68N;R#@^wp)~FDx<cPVpf*lT
zY+7RgzuHM6>-4PE=u$p(H4Pd)$4(d>6vi=IF)wrs8aZ8*h9IYd16v~bp@C5vEHOGr
z6joF&OjGu?`=4_PY|MP?FX+_LVx(OJMMD@0RqQ^KSf)*Gk7yblQ~tf;@1KC-d?`%o
z#e;Q~Vd)sh(fACh(<9Rw0qs85&f*?gF_U!|JuFn0+wC`sEH1dwr0nzk3(;fuEI=lk
zr+pJDVyV}tdL9<;7yDkv#k?{b<N~H5N-X2E8VsqqRTfj=fGw(yvlXDJe^>dG=8KHu
z6L*^diS|?`L_49GSd7RFk;8deCE?FblU-nj6p@xD#=`VV(U$ufsezF4Yj9q(hj0q<
z2d{Ul%)cR^j|cG6EXi@$h#*ABk+KIF(e3oC4r%+yY*o7KG$(R3x5DlC2+Hx9JHE45
zmZs**y?sZqQiY+&+}4V2x;6K#kL5{+jw$~zo(_DCXj}MBuL+sF(L+kqi)1mlFI)3@
z<lM~X^0Z4bs>Rh=7&cU`Z-k!hKbL&{(ssG}M-t1$P>odO2JcV2SrfA6PnM@Z)eWsV
zLVJd_5ve8NRz5b$Y{M?A%v3!0Nb8bNW3l0Nrt!V;z0O#~_QT<-@XL0t?~ci@s<d;1
z#79nmQFS>Dpzc>ohhshfzkKP8t1Kvm<r}dvo27-i747B%9>Z`^w}7DFeT8|EHvo@N
zOZo+AD16>9@x1d=U+~?`Tv>-!$_ja^Z$PloFmL)0Y~9(mp}`jWZ{gn4&6C)QZ`7!{
zPWRp^?wcCW*AjW2Q^8arCs004w9<?%IALjf<Hdy7d$4#GnU1uv7PP;y+uG>TNM{+c
zJ;D+7IPBw*u0uF4x(%9*b;u`dVDOUy%I{ql?PZ!3fA<CJa(kqQ2G@F`izMvXZN`nP
zjPqwdUw>b2byFa{X}3<KIbGBfsE<$X?%<w=OmAFy9d&3cFW7PVO=g8WjhM&hTTb_f
zjdu865!p)4@9$c$NsmUJ$e{INsj?#3l3rX+*vl$RuwkB2XDPIHi4FpoFnS0D!Md;s
z2v43~>K`|2N6lq)e_wxrZdD}Cj(*ztSH^X}<w9*L$1pOYSn<NL)nzt`{LnYJHd1By
zBGj8b83SnbbHBY6@qG_f7Axi@!Op1wt9n_q<Rsr@i1}lK=lHpTBQ3rwn;#jkb8NGX
zqzd^z@|BRV>Sk5cZsUy35yl6GG`TbJBHaP*QWJ%WGS#=6tyC;jG@7}aX_?8o9>d%M
z@Yf$}=SWaukc`-k0TyJ3wQB}SA~GE((x`}|VK4T7s1iZr(1>GU(PUldh=CnK?w!R*
z_)hpi8GJZf#u;X;d5D<l;S}axisj2DjqRN?jGdM!nV=N57`7*DC1)P1D3C@oGx1$I
zw~ih>Cy>e-$}ZwE`61@=*-sIes9>i}@|hl2oSo$=-3i7fXDdPB<Ht@;#2ep;F?JyX
z$At62w;qD%nsU9w=31cq!>9NN-<+uk>Kh>u3bO|`PHK{}bD5r62yife_Jcz?gw(y8
zgZ<&VHir-{pZ>Ru82mQk@;zZ>a3Hzjgwq~0W#p3`=dK^csQ9vY8Suv*$HtHrGx0J}
z6I9$ZIhJ~J%M0vb9tphNs*ANuUuNzoEt4vHaX_e(!|dUBtAqu~`M9-5CHt0+U(YCn
zd*h0+ZFqHreFV19SG!*3(O6r~{hSw!3DIGrEC437w}9`@*=aqO+#UXT8e76+jy3rv
ztYy=OKJjcY`MbABy~QU9Vy{Q1+*&`pQaoN@a$7y$<jjowRNPzdhu>STu6wl(jkJ0W
zD&TZSluh$;qDu2E*jIG<4dmk8Pn1dJS*d3`NzZHXgU|ac92d=N2T!avq?X~~j%2Wv
zb3#;{!_T69X7BBEI4E_sg{1$3isV59F-^5?)xecMlEwSH4+tphQ4K~rZXx3j8Q-$V
zadOD8(o;0#3TNq;EFKr??#ph(^^;yxkw<H?AJi2SdSS@4(;RAqenmkV(SbdI(Uwgi
z8suZdp?e|}R?}1i!&k7ZDpWF`*@;dFAND;qq&)e|XGen~iH|FU9ZwN`c8LXjzI^IX
zAyY&#UmL)gDL$JE!2*?53<hMI{@4JO1Ef~%gt2F?LOrC?gM=_(g|b_MB`$LPTfh?~
zPT_Tp*jvrFhf#`8*)^|n&kVdeDr|JR;?X9WA(jD)ZpRR#w-QPOU+rIJdt;;FZ!64`
zMGaon43y%uIGEqTn{rZ9aD$4STce8>=ixs=ch>DKwZVy&V4;t)T+5c2G#UM#{h#I0
zXoAU_N%EHl#7!RdEnnv|37SjvW0w8*CPzZL8@HmOZ8oe*VTU{-ZDu(b(Q0K^e1Eg8
z(jb>gsO*Y?Ev=&ZD`Rc~_Qww7<{K;v(=}a(?!i6?0wUd^MJ<zT>S;+|8}k<~K02L&
zJMoE$>c%BWm{FSi>(wcD8)KvlHr$R5NC{@Yiq>1aI9TuDzjXZxZJtN^u%!b(Y*z4h
ztnAv9vh?>vZ&>xr>%iRQ2a@5h)dTke4!5~J-dX>`h`^G+$uzO@semept6pgOMbRO@
zEicG@kDweHx-mg|XIPI#rs${7P_T}Pf1>p#y`4sjdESX(UXZNTP*Zw3Lm|Jzqa*%?
z?O5(Fz4^cVl~m{CiLOyn=>+`kidK&ir5)y8a63^Kp6Q-etUV`5+zID`CZCCBg%79<
zh=HF{<=3RXLnYe89!CjN5|2?vQ*ETV${`-LP_$l@WV-y@L3r}Sm11zCk9oNDOZ-h0
znKdDP2~#Un2I5MB9%Vzt%4uCAPv%}NWFJ;(xULvP_|@v0n$An#U_X<7oEDNKl@bHF
zi04hf4kEp8xezkDK%IL8E;wpoF#B&G?0<jl#c22f)V|;OyY2O_-`nvaBE5^KAl7fS
zwJw@FwF?Cx9pt4DN9h09+g00MRsN|6{NgMB{=yL4P4jhprGCyG`Zov)EN!hzRy4#j
z;w<P9UqQPFr4t3R-dotHf4FceRk-kH5;}4J@O!^L%D=t1zzl*S54olBuQJ%*ulwJw
z^MC(%5cy!2pJZM5hcDoN|H(h!;U<S0AS$5f<u9+Uf8nbCP)hkT-}47&O8t|Ubh!V&
ze*0JU{_90l093DA-WurrL%{Ig0O9)z5J;S&%HaFU{p7#>-hX@lev6_!(fH)YoS$id
ze8#bV2Xm#?46}@oFW5thFgH%dn|P$uXXpD;L4nVYJ$qN1pM>|nt?mE(@j@N!CYS2`
zvS0o=TX~h42rQ*b;a(dq#oO>R2%D*Voh$b+Jb53=aew$NtMNt;v{h{lmeuH#=9YPX
z2TMj<pRK-RyEhNW)?xV0ejMDN?*JsH?Omw7zt58)l6maIOG6PB_eshL*|rz}$KE-%
zosRj?pZx*HaJJLLV=n!e-!sJk`VB+RDah_c7yF=uTwbkMa+MfQQ4a19M}%U9L9xO^
zRz5UMYyTf{+JC)ye-XL&U76@g)c(5n=vi-r|L!}mv$Y3pErkh4v?EVmC@PKrIPM0@
z_yU$AEFX*|KNKw8^4Y$EevEFM8!7wxLoPq2T9J=l%;pi8INR^%AZ`KLRq`aO@-g=n
z#H)ucKjs+}Uly{&R>~*BiV$+YegQa|0tCwlLoGM=mcHNHFS!CX<YYbTVC`E1$q^n#
zqxT4$;H*Ost#!bJiam3V&N31~`c#ZKkc$I7<lIRjb=kJao>ts0wi>jGb;S1@VXF`c
z;=s#6_W;omT&Hb>&7X|^zdrik@1@yJSX-Bw9*vQ|9<rx?#2^b*Q4nB4Xs^EwApbgU
zAa#MbmCOsoAGS(!qH$yrEVu#}BRQ2gOect884Z?|fk?L$WaEJTLKrpnPT|mLDx&eU
z^HJ=k{ct_zi1i}U6I+c{?OL?a%Z)=5cmemT`t2xmXmB`=0kda<9~}8WVZH)c&=CzN
z`{xxS1K`DBKQ0S4hZ%?kh*W^^&Lv|voaL8@DH>hz);!yQ9N$Dhn<biA`Ow_r`=s~K
z?&kK_YL+UpRrz;kv&82se?6`L@rAJ(+z3O<Lj!-6=AKdzsmrWA8iTwN8D@`UOjlJB
zv(UsymLm-oM4r_I-zQjpyhDT)ENRJN<$P)2DyF~SeG+>`#@OjG^A65|Sx6&!3G-Wt
z*@+lF*<(m*`z}eJCYC1_jc!pNUnu)f+Hm4ZI}n=OBjC@@?XzzuNU8WbKyf9AA+#Dv
zE&Emv*S@F0zdz;g(82C!{LqBvUn_in5oXkl|7f&?FE^~X8GTEO@;*CkF*z5DGzjxu
zF5*QbNhxG*0ZjQ-lQ17Ee&+!E`|gD>A+Rp^?cQC9?6hN+r}~fmb<|p6Yk_^XOS9{U
z0pR@I_jxOT2yTUVAo<0M3!4|{wR%6~g0Jd2AiemX$R8%#?z@EMn8Pl>BDJz`?_kA;
zGSpd+VRiN}fG!xTa6}JZPX-p8fif_*uje7CLBd-X7$Coyb5plzfo(xMH;xox_mirD
zs;0~hlC5(D{;L!;Svmbq@O`rH<f5KM*uN1Fur_HAvHw80X`~6}9rzW}QRV{t?ywDP
zE~Ccu{-camNFDkU?Ili`1q74L^rMO`UV_vJr=o)RbhlGBQuSb+q$5>D_*Q<1@)PXA
zIJTgg?rBf~-|PYWv_BXC&{~ye(g68e#LV{V=?e$&TzW!FnYgt!ZwZvEUORtdsR@Fw
zs=Yv;?~#HhS0}W$6uPP3!9%W*m-T!@Ogb;V=33(MD;08uHdpA#KeO`Dm>#wzaZ2d#
z51Bx`OgD&=D>lk4v!iu~+B<1+Er7ii3~CFTGs5@6g_LVU#ENcxs6kpW=r-z8f&~l>
zQ~A2uWQ(5Hz|K%m8c`av7AfXyNo!p?a-4w9wzDe+21Mcq4Q)?PlyIbHC&`Vf|H1wT
z0y*#T%>#-(rZ+`i^S0w)Q3aFUE-2AWNqz%iw_&%=4I#+G)@~w@IGDbcjxAZOZxcA(
zb55E$%6ku)h5%KvHC6An_xTT(y?tk{^rSb7Pm;(tVx5CeK5BnqmwP>FkiVldn|)g&
z&l~@nCMr0~&rE?kn8Z}*3y(j$qq@~d!>1`W`T3g#VUH?81mxl8m3hrXLsxyOK|C;C
zhnw)>se5iQnuV~ZyhkWTxTFV(;JC$hr$0BRB{7QoVj+0J^Z`s8P4Tf@{1}nFBE4eJ
zQ+3F$&qe9YzO%;qAln;glFc<+pZOD@^}+AJ8}eR8R5%SHykt?ohkVclCIP&I+~)9R
z=-tcYzK4jK!r2t<fVvbTF7}dyU26eQmaqwx4?EDj_wa(>?1SPZ#5RBLdy^a<8QoWS
zT=M{X6QW+dIUb^-9!y0d^7-BXoGl|(hlT=#h<KLYAimgODeE86Sjurs8K&DRaIAcW
zVGHC-M^EfHcuPd4cmOe0#1&&Cj7#a>2pNG+eH#P-Jr~$oaaa81d`{7&+wzXNI>_ue
zMWQqSzkJSo&fzZk0u2SFu(U&2&(jxRDIN<LPEG25#j!W~ZO?&Mb}`V$XX`I4`XdKk
zV|+urQ1cM~BH`j|Fd=xYD%fslEFyga#T^6mgP|7LBL~(6R2S>uvBp@vLC(Q;#QfCZ
zYJX%?grm<&!Ux(5l%xh66a)sA5!ouiAznIyBxedo;k06!P~Qfc)6bHAC8fGIx{GCy
zDv2j&t$pF@?q@v&s&Ck_xVduR_~0o{y}<=I{0FJZBl`6>SfHCnBBqAp@Ff@*@{qO$
zOr4uzwPpQ;m2<sVB);Uo{Q}=Z5O{fBuJ<PHZ1wf(tZs37I;&~MKtT|vCbtddA<#TR
z&K4!~iSCUEwU7^VwiUuq$QL$e4de8b)>CaJiZy;4kzNae!1Z<zlcDpUk?S>8o_l8g
zjE||&m3~L1GHo6eeu>?ghc-;mWL$!+40fJYLq+6)l|1JwP%YlUyB5f}69+~)tgAAc
z4TU|c!WfX%C+y`3*tRG&85~I|T@iktVooac2&p27+o>{u8<v8~s3eV*Zy?`A=X1R2
zM|u{a&}~Q>F@V1fqssAcvM?|3kg%uh(>&%aulbB6guGY2CWKy$kGK}W@Y1k=_l?wS
zRDboae?Zau+UE!Dp$9UPBIT@Rdq_vY%RoE7r?s(5bUl7nchJ-HD7V`xE-GT^pfslD
zhiFa|G-fKiY%fd4Z!5zCjl?Vkz>adJ+!o3!XZ}<V?Ko!0*~x~rEC2XV+_WC6LSwK{
znQNyQe|?eJeKLaWxIGNFt&7O?Lyxf^WOQVYYy~?c3QDqUWAeFUFZ>;WKzx8ce<!>3
zO*yNHi?X$g3H9*fKW@x@6^H@52&XuEn?o{dAlorYu}7X%OM%<byX|W>=>uX#B8@j3
zp_cPf%snY58Uhz`_ajS@*kJS?$7i~Y!!HU0V;g0f<C}Jq1xYob57s1vr)P^+eNbTV
zLP{Mw%^Q-y5HW0JmXd&ZRfT%Q>`92V?x6Bw1(_n+E0;I@MzP(hW{n7*SMR-b@qxBm
zJ8VEbJV|xm%r1Q`Xkumq-X;1^G*iLSUdgUA>D(F94lXH{9})!D8F}*yi<i9#Uf{MV
zUPR5DC}fY=<2$q<I~lH+V<<Y>dV^ut$c*88&D){`v4=DJC+aph28E|-osLHALLihw
z-%%Rvu45)gZ=N~vdw|G8g7Al2q0X<f{5b)Uz}@ugZy!3B^9bzYq!H<bB36tMC%<&p
zRN17$xn(Nf7IR#ox027dleR4lrQ!n~#7RK!(s2FAibh3VN!V4g|5jXGSC{TT8Ig(C
zp6A_+JS5P<l_$Pkctqea7fo9vhOAXQ+Qj(z1p%`^Cd?L}{?XVFHao$WofC<Sz<bx=
z=4Mv$Zn0-7QjM`-EcFT(4JuOY(k9=h>I{;w37J9@#~$7aVl9`DjCU6V&T!EzA9Jpx
z(l33DKM*6C(%D*>#ndJq;~W_*J;x)rd44%F(*CHEhEO7y`Lviny90$^p2qUoWjF-6
zW2_<%(YK_xPdZ*6<}%VgFTlb@6BZeqOeU~E-Pp-tRpP4uar+89;$qJSwo|c7x6kpp
z)A#>mE$?)srZ<ebL`WbJw*C4LzI!%)m~NuSios?rons^zh6M+gml=Ii1!~`Ex!J)y
zX8NvEow$9~Qez8A*5n)Q;k&X%;NsNMu(<k`W$#SSbFb^6H>zC=#0QTJ`p}Fn2lRV=
znDtOnGTz;BFpwP7#|_e;Q0DFldv)C$AGxQ?F!85H$OUn}w6<B8@(ZY2%l?*uhaR+~
zzTCfJ*wxEkLtef^=k0J^?!$wYts1Pj+iL6gOZm`}!kzSE^G(q=fN?K|#2i}KG&cWw
zB~cHGqdDicexDDYX*fQ9weV-|2(7_{#PrnXzf&?0QUvbf%_?-Gs-QR)UdpX^4u!rj
zfpMuPdB8v%jyOCg4iY~g!$rnW5XewF=<s%M)N<i^hn|VEnz2T^i{@w=Gsg)bqhfTr
zy3!aYE4(V8^&;hKGXs#FB?^<xw<ctWSSRLsHuzklW#9Jq9SEozr|eT5v8}d7XD+EY
zw4_Qr{61lQ&sK--O*do&Ozr$rLGe6d`F`#TJ$ahddqT=c9dLOyz9m(p^opk**}ct&
zVp4g_Go{Nlu(ktB*DlIGRr-up&e4e-0kbi0#T1KFTZ=usQYePszPT)e<0~vE>XX&z
z*`%ALi(E1d_9?fROJ07F7<9!&(kt8%TauQ3m-^Aae{h)Snv0Cdt52IzWcQe_i@QIn
z@p5W?@0nIx3_&HQ`jQD$yNJ1QnlY%zn+$X!KCGr6?-A}(VC}lq^`2(Q`>C{wtx_=|
zz)b4TZG^e2Xb(-NXXJN&8}PQbFrn_jRfhIn`kQ?9@0r?t-mrU`1lvD6^M0xvT15Ai
zwaz2Oo%K|f-IKjMN9lw)!Ake8>!5*U=xT6UFQ+*RubFgFcuF``Qd%yTb5fhnAVSw2
zx?p%7=S4NgZf$Ex)r;|%L^DV_nx#GG@-~GQ-jojl`Di{ReI#P8OEqLb9eiBTeqxUC
ze3_b@M3Qzu$RXa4j|-IhN1j~bYlgZ7cdv;@e@^e-huuZjX$7scT`n05U8C{f;@C8!
zb{c$Q`lW|ga}_F8dg&?&Xnab-Zg_I=-aC)ED5tVn(IGz4pDo@vlB!EB`%>WkFf!bR
znWS^PiMF3Fjnjn<=7pNRvMvHlUiDIpxvkC+@5modZe}#{<g!v{U`j&nbhO8Lm_v7Y
z!QIKPsUA&dD=l0a-4FtL*ZS5~O4}pzLUWyPxZ|?`OSHfJqTT3WSjia2^M;p@;lC+Q
zR`I-BOL$<CT+p*?l#NNsy3ID7KtQ@qPPEX(e8y#M<-p^#b}r{F@Njx8mi?I@bzZuy
zNLVdZjsCjkJ2?pP#QNeaqj#a*H+25nlX1ypK_xGiaTDhXktEjznSIMc0?Xc%_N*lN
zn2gZkPXqWyDFUBf{h=+B-p^HhDJ%o{m-hcLBmMQJmH;GOT8r<pEdP9~Iz(QCy#)**
z>;S=Q#0_9wNwttZB6l;<E3ol&q#l7DFMD0RG#ip4OyI!?;?3H=BMBJcAeKRyN?mMV
z@jk+|_v8z?Msm_Lb@d!GaTo$ooJoxR(W6OOy^sv#hT~D*H?6O(@2_j%AXJk6h()Q$
z&EhZ(BK0)5DsyI%%sx2wzk<Ng60f^LU9bcRP>p-gRD4r@D&@2NlAT)I@frJ%SQnZ!
zgb2P#gHks_xk(p4^FGdV>9EFEn%CczR1y~wt9n)W8<r3Xb4>A-0)d^g$-u&8Kk0;O
zxgc6{bsk$%?;I_xq|#&M!S+sioHD-K=0h+A0i~UcR2;$$CivA)MqKVlRe9LoL53Yf
zMu+5RskE&fe3OxJ_^T*kzBI~5m@s?@i5TBrho03MCdXzrb4`h5sV;#Bx0A-!4t8yW
zH&QEVpvt|zPCTG=>#wKuzVrU2xRMM*qz~0p7FJ5L{{OsEMeJn0`xOp+VmL+^(SvlQ
zBW4uns+HAOlm#!Wo-{iq9UT1r%o7<>1w76G<SjMWstwS39!PxRmE%Y;%4NX^+eJ0G
zmo7~-C1jwEYLFjIiWEX410CAe$6T*M9D7$csOph1gRF`jbZWb(2<<y?^e`EPJLOzg
zAMFUZS1|S}Sz-t{w^$>-e-32DUI2^CdrZ##8ZJPMnT=&`;jsj>FYa_`@hFS(sbh>v
zYFa}<O9?tcFPiD-)R%m{gHZ~`mDzp&_HkiQQCboyyJdRX>5<;c?aZ!ZvN5$3*}zav
z)JfGdo+D<4VDb3rJa1JFTMNrj42ON@SX9^ouL1%3p+zVg@PLPU>F3#?De@*+9F%X9
z#TS2g_^-`PMyW3X?f42=vTs272hUuWVMh7rM_Ehvc<`c+tlnAr6aBk%R{gBqm75jm
z12;ajZropF&86;n2rypCv<~HNrZipgKQ*p$oN3?^(+&73bHyC{mU*)!mC!kJL*O|V
z&f_tv!+2oAm`DX|cP}o_y0&O->c~7F2o-4H!iB}4vNV>lQaafy+MX@1y}4)zvKIwS
zQ!`jD-Kttr9>_u+G53I)ecvC*TDc1IaK;x2{20X7lgn3^XYZSDFS8gat0i?xyneCz
zK6Rfm3n&=xftql$X?nLfN8eWZ%8qtIv$RL9y!|oi^R;Ca>*!alvahU;tyT8x16&e-
zq(lxmn1CGk&4i5f3j&**G>6(s)jP?V4bGf~u2CowPEtPkY=T$n5^XzGl!go>Ty*$?
zbe~14Wi2r4^%E+!XQ$wMTX!|;-O3*@%-6%jFn#S9)wMf!I=-|o%~~<2hTi;mk<vCa
z0~NX!u$>Y2X&J60y+!N&pv+im1~+bpb{>a$n&W9#S`@LC1xses8Z3*PJ?@{s=kVK`
zRFE4*l5=k+{z?92-relb`M?4#K`k8p+%dIn>ojZLwLjR-tu|=Aj~S4Yl%wj|Pcz&H
zw4i>`@VBhB>a<LE9kiT{i@fx&0QNt=c&<VuuC%V2_}66B`z9(+DI+F9R=$+9C%R;q
z&HD8egm7%y15(eU7qFZNhJSIBoNyARw;6Y^;ICA;?2SSZ=#2Y=texlLUqSiUvI@x9
zIz2K&*{M*I$;0E$nf&1L8A)RKG-5OKq%1B)e^8f`8<+AJ5u%ELDdgeoT33ZBy@i~+
z2n?s`DSWW02FBb43JFx2`D)5g9y`Hod}?iB<g9+l(8HF~1Sf>@@bg-jjWS1{+o8Hm
ze$LH(s!bjZ)!KP(tGSas-8KiSqi0-xE(G7m`}P@j;##2_jcJISjQ+4*e{GPqzQpTe
z(uU|8wj%fXft{Cy8^DTNl2+zeC$d3b_>4L?LE#BR>3E^}egDBHC??0-08_QKNw{k_
ziNal-4*8;_JMp_h?@e-KI$2H!>^bgtovH*sZxd{=OgKaPerZMHB>pm*!H|+5I38li
z8;(Ll&i`={e!W?F0Fef7F!mq(B@GgMBI)s6_Ce1hqkdXT3tk&?7GIG1R$NNPo2Gdp
zLBhclX*oLX*a{vU3^rFV<lM{jy>DSj!fW7yd#%w!Z0yMyqlcFcJR1iuplk-F7k3<w
z-@-mQ3#TqmZ1-2M`-dnAU?fkOC^Y7!L*6O+TAH1gQUefj7uBSf5^Ve$6r<q}QKxU>
zVYb?*QDmmXmi&)GU+lwE7#7f{1`7*X*$LjTXUD8y39$FH=o$j=9tQM=Sf1UIG7vj2
z2d3VX)QIGD5y`O6aBbZ%f0nnoS-%Yy6dZWvJFj<Y&PQ?UBryXsYD=2>Vf+2Yr}*ed
zHYdvKB<F2msH0j^P@#=Xt{qr%&?p+MX{Xl}M;`ygroj_o0TaCPw6g`<TXMS4@9}m;
z9jeJi3niWQ>AO$V5+KUSHIehV%fE48jY^0aQS;H`W}PUA_RKfPHt07_o6^IC7ZBJ}
zY-`A?ZkczSvOyVIE1LVz?;7<RD$0cHo2S#*DNm;J^SX{g`l+xl4NkMqB8`0$Ep8xS
zX_<GiEb<PTqNhQHJ~haavR}9LVeU)4v8l}{nc%?USI}wq9GTqJ(#IJPy8}&^XwG*C
zE`oR{&mld3D<PkT$LC`EU}|PEMNc3)!N-I`DQ1R%@&eewFJX5QSoq&t+FPw{6U@tW
zahsvp(bF>|wsKv<Tgv@!I{84R$2R9IAnk{Suoj=Zr#5)p*wdEV*R@yd(Oa;igK~eW
zG)wQdBFt0I<FP^0n?LU7{)~&+yA?^lllM+X__q6F3r{5f9Z5^IjB6t?A3=@fxIbcW
zy56p9U{1Y)y)BLujlePVhQPIV@H|+9VTiDLk1aB8s-j!R<t`;llJ~lSlS8LbjR?(P
zyVhKM@S$nS@b>h`?@re{jEF-@my^U_(Q4LDsX1T%0E%F{`SF!vpE@(Zo<t>H)mwtc
zvRI5kyOPR^c70<Fd_8htouM)UQDr*^cf4BEIvj@ahbSCgVkASl1T11Ru8FEk+X~dg
z9%4Y;5#liOMF3AU8^x0zg_;yb`)d!mJ@_1)jgf_T{b|EG;^=}5)m*_|p6+>Ky7R68
zswWAU4<a;2JIv))tS<KqB4V$Rg-DEJ3Jx)}NZ7316wr-R)2=#AO|xn`5`kg1PI_Nf
zoapWd@o@bUUi;OvUb!&u#QRQ2dPa#TO<{W<PgB(v<k;>0XPgQekjb#pAm7_3IuEw!
zX*7G%G^Q}MPl2vAc*4Z(`VGeNnv7yGoPY(Gj6P2UHyOkzwYcKB8mrKt09}JDjc&WS
z&Io9qZD9+288{)6)%4lk)+)Zj)4A_iiqi*{ZqJ>Xex+6@)MRkLXk1j?o$1DB->iyB
zGl9>;<8^)=snU2b@V)Zpa((>pXE)#mH3Q<^v?dKFB4|UMo2SfIW@DFTC@*zrZ8uiF
z#czNDOF!|Z$WM&zD(c|1kRGT;dNN4deu{`uLP-MG;)469dCex=YvbFUkVb~!5(_w=
z!)8dFODavAV`jQQSGx&9tkP<LbD(9B+8v>?IFX!*+r|*Wb|}3be`ML>Nz)r5Mn658
zCm?q~P(nnVgrw`I7r=g0o541U+8f1^DR?z#M|RxNDuks=Bq_ivZ~WCkP(AMseo*r2
zNWYk<6K}t7J{Y4aiu=ycl?VU#0kF*ScYD~an{*2)hAm9s<ifrxd0o9`1igu`zArWi
zDRpLBY@d_<ImhL{)NnjlV2iH4)jn|Jk1hH>^QC(%Yn093DJwb5kr9hPS5|$wtj=r2
zd|tqe5Lau-0yA@)eMsGQ*qGh2^cAK4ps$D&3`j`4#+&XyW1*;3aTCWE|J&Gpzbd~;
zzw+6hjZXot0u^lnuO3+G)lRADus*Bn=h%+Lw23^BJjnQNG2)%ViBu-+q38y!HUBi<
zQvtUhR^6AFrY!Vjn{f|PTEX6R;M@Mgnr8)3*(nH$q-2P+`x`9l<=3^srJ!Oo?&kCM
z$k|v-chCki9UrH>tQisO_Eo41mp{k63U>FsR|c&E4|49^*5bW)TXva@-4wHDPii^~
z<4P%<9dFjuLet6p2nTZkXkqXR(!*)fv9tH`_|SdzV#=-Q!V@^rcJF`br}FS#;Dw;k
z0;tevS5pyXVEXJhhIN@Qyhf>-s-#A58}5#_R@toKH&?Dy%VlN{EImC+%-wkWd~}Im
z!dYhCwpgvv1i$@9g0}B>ZEF6!KfRLay{}FhzAcGBa67&%wsv9W9R3io`6H%h9VLlw
z0&a)zkbKhR#%|~&$aaDQFZ?*ibu?h~Dpk^<X1(UBN4i7Vi_1GL5RteJTH#XY=cmD8
z%9l%F6RPBvh+9pLanB~-k0tH4B_sKl%B6lZ^UU)C2vXOjPOJBr{WUpjABVjx6xEh4
zOi9Md+;V%&{({4wy~|nGy7Wl9pjorm{q6kgZ%vOoJ#=p|3>U=44a;3P>1ZN0g7X?#
zHTnHcyg?C8@XoW*Q07mHIo6kyV&H&e3Wxu!S+v688I2}rH=-#dTqe{!Lc}2V4m^?`
zE#Ze7v?v&QmS)K{mawz|1ckFH!4E?kYc8lN?#Yky6|$yKj2$Ix=+jUZ<-+^QpZih4
ztwAO7F%w@Bc>6*A0WI|<hBs|6V3C$G97~J6*1|&46VV(@#r}B8oamjC$&_p(FT){x
zqw~o#t?b99NowD1`usT8{Or~Ob(+%jxyH)(OZhLjy3A*Ul2OhXY|pzbrdfC8_p{?~
zh*5G$mu2e3dg9M}f=`g5rZ3w{o7PsTr*|4HJI^3L_Vn#p(hnYM{rI6TRarV$MMi?l
z<qGFZg_#YQa%h@0M)#2>He0mwq=ZsZ=j(oSo4$@w%@r#eSUQ<<gvr{c<VKK@ZQ)i#
zn|^Y|>+-G&mh#sXT?S3qJ>9(g!P~MM^fN_!Y}loMlO4NFvIpk{DQ}9z<z%j9WhcwI
z6NXDZp?;_NS;n>SMv-3UxjpV6cBfFW6O>UWU51^wq)V5hCxW+RYmK%+fTZX`Ez%7d
zFJ4r?!FWSja&XTAAMR6X;kDdlQaY79H7vAqBd}GLS}ZSq4l^k54@B9j0Ae3E<d$9b
zFLmPNeM4{|!5cgOflzp$;KysYud!0U*Bp_mo&l7kpX8J~|Fk^@$qVR)V+7Ix**3^6
zbib1wz@Vb$ER)|VWALYpfnHdDxukpRmoEOCl_<@t3gpHR(O{%!A<Y1SIGNOd-L17E
z*vDmi8Pv}i>*^ul>(i#=Zktmv3($8H_NW@*!50Yi5%Dr6EEiG&q|tBfyKL6JU!VBD
zy@<jg6i}0)-d_m`@(C1Re$65zc?jn=4Cn$&0=PcfK`Tj^Eixy|1^!-Swvpbf*kAaS
zye%=a+c=>##CZ$wK!*kCml?d?WB|GYnL?gxEN}TYmhV5oc&mKKA`RTmF8Xz)TJ-ma
zOguY4Lai1bi-&<Y_Rz;@Q)JBWXZT+ExE9LZ<P6d-cG!{V(*`j3TD6COe`D@vkLdrz
zfd1{ZHV3Q`J+=M+6<xY{3BZf9EUaSxWV-e%)A+X+Sx_6+4-G2)6H5BGtNZ64<<IiK
z2S^|fkN<rQ`oG`K1K?Ghqpb9kocS+@@4sF3zkYn0a2&SHjsq$k|JSe32JU{{)Rcda
z5r5qb|Nf%81D?Y#)BDn<|KnG<iXwt&-v8ngi5`#@#pj3eu6E7eGyI2!Dh~KkzEHkR
zQ~nEHsePn24gqEX;xCuuw{QR9e8d>b?uy7StH!y0Q0ff9E>t!7io3}5H^iR6UaILF
zW=6fUfL^Lz%j&r#VkxIfce<Ffb>K5pN7o_`6oZIG2n?*(yMGPpRV7IMw(Qev<Xs%I
z4E`p#xR1cRAG_voHsjv?#@fF<zOJ>xjd1&a8lqN;-T&#6sLAl*Z$62(>qq3O5c6jQ
zW5a{}oejkIC1AjqI{dk{Q_%>82xj0)v<@>@N5)(~*f&%2jU~uib_TX4iOfI{v+oLc
zQt`XaISFz-mF!VFP!n9a3qwTOj!KLZz*D;M=@wGcN93De!#-(2&wq!~s*n46oc=AY
z*gc0JT~emTzi!B<%!KGl7!IF$ry2ITn+KTPp8jb%*ipl#`T=$+2H%fh8EK7lkJy2~
z;?2^^!j<<vKfD#WJ&~#0aLncdVKIixQYpN#`fQtWwp~(V_{#Js*eY4mK*&EO;Juhm
zEXw0CSL>=#7D(WRIO#!NPIZi)=;>f;i!?1#a?NKEGJOCo$6m`OMtZ-$>1V}|ojv}U
zF6JkfWl0eWU%Ndh7pKq!?H*pR^dpeonq(5h4tPUCG;-pK9UI}9eZ7h%(oUDmNI{D$
zqd>1%XBK&TEF>b=HdF2zU~>E!;+F6dHf$6v6sxsWVj&NWKsNlS&Ciujev87&-(rI;
zqb#I&i}$Y;ka+-GKeO;<nX0%bMS@WvP>>1pqGb1Eyb7t_d36L5UC+vT2#<svY0S}o
zaaw<8@cyd!+-FvT9T7(_AokPD8mK!!57#r~c3Y4;XB+$?7a)0g)d9UP;=uikxYn#B
ze&2y+zz#&2T0ZWS4E@dap77NmR&>hV$e{<8aBZNvR%G(KT2<|WYisL+=hV9Tbrx~v
zk9)5}E$?**_AGCDZz(byd$p$eEVMiz+#yFVKKaLv<u(vcHjwxiozAZ%Dn*(iL3=Pu
z<n^55AP|6QYY1_Jha1;2jdTyLjp3Aa8&b>BQkQWaTYS3+Pl@;zp=D>+7SRly<&%QN
z&5lLzLt^k>%E;YZhQqOev90uUVKaN3*tXH;7U<t-IfNnG9T#zWu-XQr5FFdfU<7;*
z^p#b%YN;j1^<kZoz1QQs;RJ=6)>UZ08@7XwB7}bY;(2`59lMgV#$r6n8%>A@L%AG6
zplI8gu9^b_lUXVFx}e8X)aPc=SHx6eyGd@3GNTJVocOicd$E1R{w32uxZs4Q?lMW-
zzShx?{{dYmh@qgkW`rhf8AvG7izMqpAL4kbl0kd34;3547KHin!)OEm{wiXF+XE4R
z$HC`x`M59A4Obd>oKaow<Zxv(eib$Z25&>yu~NrIi;Dd!8X3*2KYce$hY&bKvgh-z
zM)obHu}&v{hg2qLrnE-cZmrH8A9Zs%Rr=19DK~3s0$_y;Fy|dKF7c%MIX_@v?Gy11
zK;F!UOc@Qbi{32lXv9wBz%{@H%XY%n-B6@BQY9|M0~k|wx!`iw`9bvHb}PbG)~`s8
zB#q&QXTDJHAC7M+@`4Z}c!sr$YpXknK)=57;uU)Hh}UuwuZO=<DZLacP91h{sL0{T
z0Z*UgtD6g(8=GRmmfWY_yG?{%+X9<PuA@_u%|;(3jq<+jegrEu+=_J@Xe!5@eAgUb
zq(2U-=+f74t65H5y7FO5x0<iy_ecIy2IOE2J%G9R*B)`-gP4N!CMfN%!1Gq|y#n^*
zf!myi?>4iDT3seZshC5J=Waq4g(Fy&>B6ya;ry)PWyK`<9FYB-*-K8p(XC-sVsssK
zP_9<dMfRb%GwZ;(A`Ywc!fXp4rJz~TrzWWKD+4tFB@JZ`;xHR7M!^?;5;=duSyXu1
zapQ@JMBl;sE_6PGSw_n2NZS6bqv{V@;pS_62reVO#$p7``(cmt&QDQa^^<EPcK7SQ
z4?9pIGG$=k(k<%|Xw*MmcxWPU#}%A&9}OV2+F`^ve*`h~g@+lp0Ob$!CNmZa+Sw*Z
z{;7pIodPA#U!6U$vJ6svKCo`m#N}VAcKDsPpb$gqe(3t)uqt*@)`W;gN5qwqm%?T0
z$<te4>!7*10Z=9>VWg}cKMbz&Sxv>umWp}e2D9=~M?UTVy(%%O)ELztjdYalu9v+o
zahPaq@a{s%kDJ|n2{wrZy2w0wd(J8_LvFtN{VhwQN1aD4zFx*x^dTx?Oz?~1<zW!8
z5Z*0RGU}FK%E2KYXc|@^HD@V2IKzrYd`*P}DfElp7^3CEuNdfoVzgKOdR+At(g*>V
zV#gQQh{D9e#X{X%M-eNviT53!Ka~tGopij^9j5!0{Iga1nRUw~_vdWS3vSkhIGDdY
z<4VB~58X_xNUl)^$4p^q(H;jcN9{^e)l(+Y+d$?#8-rxnL0wn2O`z(V!YklICxoAr
z0eP#mwFyU2|4lep^?ha;b`{D$RXhxj7VPLGGiy*hD#^LH@AHH^)WtCVj0t(k8`%1G
zw(`C*aKUW5gj%XEd|NIENZW_#2YS<IQ)t(gzp$=}+UG?Rm1CV>8Jtk^&Nyk4OK1nJ
zJ1YpI9MkCA)h@V5mXx@rdbdMi)GV|T7~znb0$HIv_e*(&t$d|W!h3MWI9Imu)5hx`
z-d;_Lw=HD9856$<_hEfgfzYqUz~HbovZGdH3X;AXElA;i-=NU#jRak?H=aHIE9{jI
zAp3ShSngml>`Iga0a!zHoMOI0&N-9<DJwgLh7SxTDJiUspr>F&5OfM+@nN&$^I+PI
zj-9|<PZ?repcI<}AKsHu%-F3vK!f5DKWm_P^D)fs^%OpZ3X{^M91l<c=@lefp_xvH
zlHMbwB*osA54yKcm}2O8B38H`Lsff7pu86QRPjL`yV?oDZwS6~hHIJn4aqW@ztBA{
zxi4UfTIcfadO%5Fq2DR&9akt~q2-_*xg$LKdruO}Uf{JgkemJ6g-z>8vuOge;%X4%
zShnsn8CXqiUc2{Tfyaivd{+3N6Onbk;kj7_akkH#yR52|nuHOhu_EoSz_RFUeI7F6
z<svoIU-J>W6XM=;0-YM<?1w0u!o6Z|g>X2z0wZv+fbHsey)H8L+iZ+%Hs4e^g^0^-
z{8gNqod;C;H>^sw=!ca@kmhvSBBXP4k(ZsC{qzxLlmw?X=am%Ig3?hStAwtqLp3`5
z_41IZjDqm>Bdo_*SN6qOPc}!eW#mAsgwPCSLb#iOT=>Nm4JEp-Am_=tvEN~HbYO8x
zcJgt!b(`7o$Qx+S^!E%X!LA_=!8e{1K_sfYnhk8d;G`w{QFa>kT@9Yz!y5a3d<&3!
zM3$8sg3)mDl`Zk`xs1v7UkmblB4k=U!Hf~T@$z`YV+XV_aD)k_yY_1yC7l>Xgi}7e
z8E;9e(Lt*?={D#yJq#fnT^D+?2a@BO96sXju$ju1vWc=Q6<7#erTml`RdQmETH_mH
zsUU{RPK%bT^L7BYQUB(E{?Kj|^znyV7i|^k;lu3QkyurgKbI=f^zOxHIFwI0%WGf_
zm5MK=FfE_rG38!~^PB(&@DfBQreNV$%bs`PKNYlfNu%CyUySYywKHp7(X^TWrgQhX
z1H*8NCD~_f#a}I=h>3Vy(G?p!YMIl~K8%cCOw&#pMzO5;7tliE#SsSaQ;Pv#j5s|8
z&(|=p8?6{km&1@b@k#lbWEFz;sD97isR*;ziFbam{;T5`el}`W6^KoTa-Lwtm!aot
z%tv4debi3C#H1!VVOQV+CvN{EBW*Q-e6AR$*03d(-l$;b7W2x<Hy4&!Qrgy6t^<)W
z`Te4RS;=sW{Wle>7LHnT7U|W`DH_WxFQxC~9HMGbikcKgJ5s**!ic*Ug{r*n*uqyq
zOl$8Pk@+Q-UADaYgtO<PSCf=?0hBWz*dkwV@qa!U(HMy#x*+iF(ZY;hTVRiJj6@`6
z-n!_+RK?^nW|?Ix>W515PSdt{6IDT7sv;Ln&S%{zM;%A9{AnS9cupFcs6)DeH8a5&
z0aXFVh9{Ygb>=T)qnPda$3UEG;d-(&C!}tC=>lT@=lHSYi0#qEqfYF$#cJj)T=*|H
z6!VluQ;!(uH`XPc5S}hmQ*B9qJh|*J9FcP*M@%4#GbT1tO~rhBg(Y?-hwh4fiH>8B
zC_=_u6!;K*)ImdFk!1S3ZaEH)liAp9W*z<tbo@e)-ZZUkIHoD2<eq0uVfpuHRD-!=
z4_5`_iC4y-Rg*NO3*L9{*}Zhk>s3}~FwwoJG11fd<m9Kllk?h#Sd1c71$Iha+%IV%
zsl29de+vJkdvd>eM}m=sngA<z%(G)%RZ#x2PE~X6Igu0x#0M!Z>#<QRt|We~iQ{Dh
zZ;oR7ZmXp+rPCXZm(N+iPuu*WQypAZ-0Tfqqne93)u;o@4d$OG*_}V_-RTjV`pOjM
z^pW4axOUR|YLr^(9GC$#C!2VDXsx`+rIdeiSz1ZN_Q>u@4-*ZZ^oXxwj9ng$QuXnl
zeR&d8{l!AWdi-l*dD5AFY&<Lpb)V9BG9lScA1`=&x@KImLyT`@1~>5Ao0Q^ylH$Eb
z7x{i=@G@8exo@7tITYVfSw<a?IpkSu+-%y#o80)`PDVsVVtlQO1V?IdZ%9Rw#9w~K
zwJYuECmTl%h3#DqvzqLQi^P1(;!81e_MoZ_5BUD<l?aJEy3J<UWbmk42XL)I+{5A~
z5f{Ss0SHR8xT`np$@vJfL}~ZQk_RUT=+a3E!z=vt@5%eA7#dI*P>4s+^9N%RgEyxs
z^8o1cUZ1rSQb(EU4X-Z>^AswVuzONF;4Bubrl<+SmsV#ViDh5UT`C@DqaC=SCCq!k
zfUK@ulRL+SkeCN8gOeCiChcJP0R+x-q}_vB-{62uq6LE_<#4cMGL4PFRRew$lSR#z
zcE4M(FppuiT&tA!_<>l5uR=-fBsdh!VHn7HmBuZ4IST^DMvAAXE!)koAa!PcaVULH
zT6^=aKk`8npg$?QZDhJnN!xnynO<_E|2Em<8igqe$^NFwfD*P?epHjr@2R?aE$C^y
zAY$9Aa5%@+M_5+nVf#F&oc#0R3%tHc5g&Nt4QyvIJu5*{yiB_$E3Z?3@DK>|*mfw@
zUm7&WF+NGu^f0$yBMo07XH{k3f6q$Y%T0EV`r7Jm@p6wdgy{t9KZdUSY^(G*MyPoz
zQaD^RhneL0&A#0O&K$=1ZRY6{M%`y{M~}}PGd^>s_ssxbTB_5FMk7tDsLHU_pfoj3
zE?ZfW+chs?gC90J9JHFexv=(r8}~lqQ+~C!`|#R%%UMmu4!$`-Ys&Lls(c!n=o1Eo
zRJCdKTrxMS6}B{bzeW%ZHMB}R61@*Z$%V9{F_3{iCo;cCq;u+J{xOS@%ks1nkv`v|
zmNnQ)Rl8Yfb@qlKKgl;Z)|l;6lfVrcZDZ5x7ikz?z{Rx&Vy9(Vc<w3Yb3NndTL({t
z+D-`#Thd$#p$be(=kh+4{;EANClsPi+q5W75<8CcOwZy7+d-+vcU~FT=;2cI=JU(T
zR5Z>@o!`r%rg_7DgnP~(j`xk;avf>$k!gi6hY#Vc{i%;F60&PN&~+;57d=;AoHD)}
z)E|?^*%bq=D&m6znl%;;KU|H|qFo5{A1#(NFIl-<>aGF6JO`&#73mSDy)x>PMm#O~
zc8ArNm-jqSlDS&N$#<&LrnpW^4CUykIJbJG$_xxX`GUfA(s$`-X2b2R7ZiWosH<U7
zCGm@f5Xab=F3VopTi`?$krG`RzO|cE)EQ)nk)6uMJQkzr3Ke@;{JP{IZ>|ix5sE)M
zo#3YVm_U$H>0O$o#FV&_{Vqyf*>)(Yt!fxperLdIXr(!5_<5n;Aa7CwGj&NvQLJd!
zgba!Ji)7<tr=$gc+^?9>km~qRfEk-uo|&!wy?U{MwA*fng&O{hCUwYIE!kE>K{2B!
zr9Tajo+u7YQy8Ctx8GO=L5(*InfKTlc6zyK%d3HD_n8+&GNlkuH{eE9RM!snKFPTB
zMVs&&Wbq|1cc9QjMB$PY#PW*qORKFwE47N;&EF_4TWAg3xs022fJ5D%LUY6P;+Inj
z1e1kOXb;*9<a|K;z$~acAphQ-lDhp2w9Dtxy@I~`c28I*jM&RK*29RFvn!AC*_a&^
zT;q-#Oe92C2UTK<T#A=ls~aTPqKfrUDUZCNO{J{ra*P~Llp|zQ+IqUJ$#3bWdOF%u
z&zJK13x7pb=eytGUm>+gw=tw92xi1t%n&D9ukD>Av|c(QOcyE88=K*F7S~NX(X#d6
zdz*LI?Xfr_)?Ue>Pm<c`a|CEr8Hk&tw2d{K*pJuBC>6Nl^`s$Xvj;M+V7~|7HBUfi
z+WTx1Fj=BF3R}B2nmr+cNZC(!s^3V3PzBq871!jb#n3{EKGo(QtGr*Si~n=lomJ0{
zA7SE(up4;-5?6H5-o&=$J3xFPl69`Z9{yh`Z$}d7<09ApqdfaF)bBzxshN;yL%zQ>
zsdM8X@|B+R+wXr($nD$s!xwHI%0M=3egPlzgFR2H528q{7{Bt8wjKSousa|S39(Nv
z?b}4K{k)V8&y)em(0maNf?npyRF8%}9Xt|@uB?`^Ip6*QHvh*R4`_5=8;elziQA_X
zlR_+y_kR-59#HICr6k${$t2xM3{3PmfDUhuLDDD+Q|#nm4^Wr^k5YJDVk9F<stty-
zGd)FZG5mR?_mbX=HKv<rE2jtvfXL(miC9UXO0@B7I^G#5$BU$-Iehg98qCSc%0t|%
zRLY}m6|#x0MNIU3dGhovAOkwj#L6jOoDJ5WnVYg7(9sK>+1Im%^0pt{NQ#$bk9T=a
zfJJ|q>r+&)%Ri(;eoLF+l+7iIXW`90(qHoZ5I%uz&*8O0o*nQ2OP}ZRI{Ih>PIj()
zvGGlV@@W;_HM1jBg{ef)svAZ&WcL_{+AEEzg4thdi53lZ|44zrRdbKU(~{6a&v|!2
zDV&)9oS$O&({n)w4qZM;kyEbtjVg%hbc2ip{HJVu@nszuJ?Eows(=)qgHLW%&rgib
z7$sCVa*Mc5mnB=4V>X7j#~c^mdKbCmez4v9G0;JF&l+uUbzQ3C<tLsMc1m3?0^Ryl
zzhAMye|XXDL2T%@X-aMUm+<Yuc^?YstV>`N&4EJ|KHJ*HRR<nA5QxcEL-}eZxR+->
z{=OU(D0tS$B~eiD$@jC#^)d7O(lN~AvN($9#lh7%jrwD_t%1c7fgmpCxKolaSqNV>
z$Q4n~!L4jy%x<%SSH4tvhsxd4?6vqUS4r-rOj)wEl_{n}LtDq<12N0zjh^r^&gUG#
z2}kEZSEvq)`UeNHQ%mV#{3a_vA8<1B3Ol?=|KuK{7Kx-+4wPID?nPhduMSFd_I#y3
zZ=9~4X8r-v`zRRu@cfmS&$O7b?cHk(G?kC&Z@l9OD|OI`d;;Qt&{40rv8*9vNgYDl
zRCMxqWgMVUv0h>$^KN=Z>pG04OHPAM$-p+O(hRLD&<b`vLM~o&3-vK)1*I0-;*{yU
z$;RI-p+-`BWY1lXXJYX`C|2MAf=&JQ0oh^qQwdX`6|o#PCTM5wboP9ge?3r5rv);m
zB(<7Jr5`WEriOVCklGAHFI6PTuIC!6Cz-uAA2AxKqievmbn>3iexnDc`I7|gpM38f
z2~_hlpD8@xHMeOMg?J!(N|YufmaUjZ%(w<APt#wU9Lv*uS8_M+eBOptcwq3}+|7nA
zo3Exda_+@V^{Gm_RX9I^iKY$tnzuc7RsX|%@IPu-x$3osuMcM3|LcfYJw>#mKi}5C
zT_bwyQ`^&019#luBi19VC(N|?B8$1!SY6#FtIR$H@?7FOFN4s6q$D+V*Tm=Fti$>|
zI!q^=_Lf+_UiRuSktI^?K*4A|x0`c~kk_3!#vymdHH|<5A@%wz4F7Ouyt})_#oX~?
z_SgfCdg;Ym>W-iAW$!)spDi*)fC`p4s~5ejj9)hAi!!P9(4i)HA-7w2`sgYYUJ}nv
z`t{C?!izx6u|ssdF**N_v+oY1y6^u+x}s$zR8BV8J6mx$I7vh{mA#UYEhFVv9fV^=
zSxMQF-I8oMkx-Ozj7av@^ZwR#UH9+$-S>T6&-46ONauXdXT0C9_iG3$npfb@;P$6!
z|6CJC<UH-0Wv8wyX8J~V`(+r;noyo<I!M8~NXFqLEvB>PUfhCk59k?Y^@OB2#t}cr
z2#g(;_U=G@dNQ6_72!?ykESt_wTKq4*yrg;$}6=#m%K`L@XlE|2ZMFIm=CB}Jgdki
z<Jv|(jp!N-9!j3fY<6w1c1wFwXh_CYUbhj38PyJ88E+&M(7YHR4Bjq7{P<sN0O{zq
z@6#p4PBp4tjT?69Zt4aHqmPD(H@R_xxvOuQ``8PPuDL@o?NU9QbJx`}a@<akZ4f_3
zV0>=}UHuLi?bZ(p>z%K0zN(%%%HMfmuz^YYyHEJ-OBk3`ULAcn`Rk_?3C}?o8h=J<
zeXQlI{A|K^D_80G<MAJWp%=#%McdItIJo~bFZL?}6jHYl63!syav9E8!xrtMQr8b)
zS-ciqZR`{+4enlI;`M;)zDB|m?Ev3uT1gz)8U#?<9Jl<SzA$38f}gyKlx^K&opxW7
zZG(Hr_e`%tm(SRunUFFWfWyMn#%}Aac39O)W0~aKiqd*_x;y*pc}}^mdz}r9F1nNo
zS(n&N=v$~|Z}VYtBm8X_^P`b;YO}QQmM1-|bbC|oNUw6A#g;eQ1mP7|a8!;BhpgC@
zDN|vh#W_EI;tuR~-f2H5FSyY${MrtZjjY}u#aI;mJazrA3T-bSMr<sh`G&a>cDl-0
zOmzKma_=(BBNUMFJWk8_1n}5bBAjLWl|VYGa!+kXaMl|wteP(mWEp|IcVJx+7bMH@
zWHK-#V8)<{TolJf)09-6sXsSyZ{b$e;@PlrP@%%UIT?s6Rq1J4GTJcaKb*HN+KIzS
zw{YD-_ngnVZ*!BeiD`U(-2M(sV+1(P<*hDgN}lSZyEF?Ib4x}fRd55BtM1_2VZLk;
z77MQG@m^HNQiMVihA#LwFWhwGq1*a|@s@z#dn9bJx@GVe$PvrFd(`Jl?bEfga=QM+
z)uZ$u0pis6q}jf`p0WDVJ7dkgMs1@1Qik^C@zBHcrZt(@FZiwIvW)lg5+?(Cc0W{%
zxu><$VCGkM6ftj%9632>+GGQUG^*}l*wo3k3hM4q*)HY*sQI&Q_s_qdsq0u=xY#}8
z9ceppWI7J>%#q}OV9sx22n;uvx~=*<JG~R)Do17L$OP)69K4%O(0_)1niS0}#?e4n
zi+|2Ou|Yr7-L-b5QAVgE(~j@&YMNlG3RK8EU)c}5CHEiLO~_%6&m1`43N{blE`{kt
zUrJdI<1VvG38tv_-A6r6G?`s51$k_TLVPAN6H$}$jCgDIUX<yWGZ5TFnclW@%D>+3
zQvOlrJEmX?qp$W{)<y2$qQ2R@39^>9Z0a?;Tn1&0MC+HhTkY8gQ}?-$l+UL9NOnQ>
zLU1+9_MQn9rF8lAu{#|>f+EMPMxm4{LzMp)ECZLbMp4V683e^i<u%L@uq_JL-6Kfx
zn5$PqUJfBCJb?bTzuw~^E3%NAj+$!O+H-Z$KA`{c^w4#fb9ZDPPi=Gaf}*7#q_7tg
zUX3#^Hiy3$3^F=t*(u$6I&Ga;O%m7ZFJF^WH?3;E{PESc@QviQm~N$thx-nb+Jt^4
zs3Eo@)jfJiI^}y{B&y^Q(QTF_j!(61=&A5cQ1pmsG?PRo<Sme96oi<BFxz;}6V3Qf
zD6GKTDxzWXW%ByM%{ZF6seMZlMlYfRV4E3vd?~%^TJuF$sDsl|?U^y~>|&n5gsnB8
z+Acfj#bU@F6i%sg`)XTW=L|!?dx`e&Ftm79&eb^#f!&;ZO;tHTH5HD+!mpW4L4%;>
z`qcJ~z45IB$BdjyUte~fgSw7XW?2z}5s;YvLXlh@%QD)=IhYuugYt&KURspVyfu?y
z$7^d3H;D~&rA%Uzx8KlfHpJRK`%{Xlnb~c(Go})K-4B>KVf34+dv;@QL<4}`6HQf~
z{cUDvp1>gSquHTouq9biyY1FFZnfj4-_vo^e9=Bs{rD<(&FF^0Kz6@U_skkDQgZ${
zNzq3vC@(NoHtPD!%PU*ev`0qcWiw}q)z2c{FnwzEC&B3T>C@rsgJ#L%ctOVoO3%9U
zW@F-n*cjQC?DOiI;!ks@;5d;WfU>bkY)Br|)>nH@)Zfw}2@^h!*2gPmT$$+RWmbjs
zv#9F!Bg?WN{pIklBKD>5`tW{)Ei`2nrS#=<d}ei(D{4xlNnq|^Q&?1u+cV5gly8$7
zW>KS<blFvuGh2b=vfz-URz7@UA=xP(Be7l3KPT0WuX9@e2Ju0eja#qeMK_Da6t!6t
z-K~4Q#R4O}r93#zQp!`_tX)B~m{Up-Fjf}1nPQm&|F8q4fiqh8={o<frt%aG&vX1{
zgPA#%)YiH5o#3d+(B{yJlh9wp8hsGa-kOZ!(m>&MZSkE!;T+GO5mUn*MX|OaV;`dF
zzD+7cs?dfUGo0&I)eN@k)?@{6K$>Vipq>a9*&S!$*KU??mP)8k3Ja)&0UyD9I_|29
z_(-@%4ow&;=KxxPc?S5d^QtF9QPi3XAPCrZ;*nsv+(EYcz6mv<r`%y1rHSO)qg;5o
zr0_(m9=alwtVqtj>C9HmT)fucD?%6IINB80bMyGzdd>klmbfcx*uh%KOxu+c0`B)Q
zUJ|XDNsRq_rn(3s2VX>fwYMANZw!^xy)wqgC4_ghK|&EKc3&ICZ^pXCev<xp^_9xb
z)b<o_{BiMcjG*`cp+@iU!YSwb3RJgnk8&>y_<CN*)M(4RPV^F0`FL6{G!fIWzonhi
zKLxWK>BpwePD;qD(2<l-$1XnhR%RqqB<9wGX;b%`6cOg96|W(A-5IP^0;sASf?AYb
zaSo-3IHW4-yQS?s0n>mB3U5wZE8)K}(+j%q8%^G!v})R&CE>B(N?E%ds}TZmP_5x+
zTEpyOC*KNJM2Or!5=c!>o3ekUA_Aijy=az+@s}Lr<-?SZ-cUHA{r&XPf<}gzPB%U!
z)1Ib@qwM%)8IRq4@mdo*>Y3(#$JaTrViOyB4@kOH+1qqe8cW)ZSMkzE#Qj&U=h)`x
z4OA4OKTRH`@=Mf11%?JF60by`z$-YQc9(q5M=~$bL5*&W%Z@FaZAa!zrg!hzkhGaN
z{+p;$9P`v%G^d_d-MEzN#EYzyxkJ7kQXNZ=Y@hXgZ}^UjR_kd0h?og6-JVyvxIgb9
zu{PPGVow<xhu}*AtAy(KInE{O977VO=qqlH!C`&r4>HDoE{{a7z;Nu?$@Jwzzh5Pj
zq7++IQH?5Gf@6v~$Fhd#4n!RPF^JVL(b%<hDq7~|lFCvwcal+-M&83Dnl7FK#%}A_
z9vqhjrA(!)dbP$ej9@!Yr<=A=M6-7yTK(Jj9)HUtl3|A^_gJb(XR*<-#vOkXp?_=a
z)JXUhjv#~h#F5Y=KW;Sd@>kM?Ks8Tt9ga^ikX=p5JwO(*H}45oBR@x;E<WL*$T*Yx
zf!S2XbjOzBA*a{^oTNrA2f5xAwAS<<d71NejmHP7CA;$yn1;}l*-Gdi+1it)CHY*n
ziKJH&K8LsI4_7dS()n_Z(i)c(_OJwI>G|^Q8963M+cyc1Yv;2GtPD7UbKQIW1O>=!
z=T09vE$S#jS6_U%g;42ItUbk0!sW-Y9bHPM%<<HKp@JE!A<MNv8<-@`MyHLZ%IUcw
z%dXvQ-Gj&FQ%U4L>|bflMyh44uljJ`93Q<S8_9QXN&$x49*0O))0zd6t9_ewl|{Tn
zq3uwm5Xy{|eWaj)T4H+3NA=-*cL%jJK(aYoFOoXT;9z+sTQARsWBP;#+oFa4BG2MJ
z7nMU68&WQ(w;hg>u7rKizl~Pvysq5rG?e$TqwMpicg^0B3Q&b~YdjlKGLqh?za@?S
z-jPSF#MqPMO;Zc%quwnBh+Do5FW+V+PYT)}BATk0QLi)jpBKm<O;RCvsa^b9k*M_t
zkn-!tM-xs-kzY+etU~jPjnEuJ7MEpwf%yw<ilOG;63gx_1HaCBTU!6Xzes-w?Nk++
z*3v$1{Ygt}k|odDLw_ljyN*V7j2010&A_Pb*u?nTU4L>~|MJiOV0A43YfUsLul283
zvPlwIo0<i1KT4w~p-QxR2okGs^}vmlStl6nlxj<MqlHYB8GjkVQG@{4FOhWp?AKQM
zw_qGo3TmOjBy%rgq|%8QznAzQ|KmTuXp#x!r}wU^y??x<U*B=(1eZyeo`uT*5$M?m
z;E34O`z=RqA>p=2QH`WBS1anU{JMX}bjV)Jfd-Jzy4_5vL-O4RB$@)jWtQiLpItqB
z^dCMm@VmO`^fXiJuhB`>49rme3o;s~&O`32E%5RA6(0xw-2WlM>YpKZs0^X|o-2Tw
z>#Z&HW@&)!6nf7M-5eeG-}?K%a<%LOfn#g>?0~_pKMK>oe*f?PHEjq=170wVv!{Xa
z-!7fMKG8p)IQqXVssHxl|Mqo1Hn`OlOS%ng{{LS7dxSH)Ja|s(AKvGGc__a}psO1u
z{{7ni*Bk!pJNUO1SHu0&a}czbrfJ|v;0zfVGfCI<b4|jZLn?K42_#~;-c@;&!||4b
zQnJA$=l(Qb-ay4cqnf{`gr1^I<$^b=8C>%Rf`-0lMRg2vvyoh9`RV-Z#}M+$x)^&U
z<14fcR$zq}wC_nec=k9$^(r#s-4_G?F+JdULx+ve$&Un;cSelz&YIdA$4?McoxBY_
zsTZJ1J$_i>K1I+Pu0ZbGq4C}`ijm^em*f$uLifoshv&iM(pNt}pJ8|LrU>agVmZw)
zds2Gt3$*a1SHA$vWK-|IDm-I2IQdw*@B)hUpYngcoq@IIeZ{<2`}ehHlKdRZ!W@yd
zs2*}2Zsok|eFqkVE^v7lYXfj#o|hj4<R8#|DU~C%v(}2wq5B3P@{DNH+NZnQ7vUKK
z7q1lh#}lX!%5|VVhBb=A+c~*p6o&Y(y}k~vfw#W7F69BC^#ufFf6JRb&DAlH;ophS
zOE=*$s1Bt7%gPyX_(OW*rhYI&dT&o>N@@V3!5;P7$Bg;r;8=1(xKv0G_D!q8vhGg<
z73EEacFWW`7Nm!gVkhBYFHK$Pp8(6HS%4FxM;Wvw;}GW6`v*FiWnbk+o&zL6Hx~ma
zIL{C-UU{cVEU1DY_egtCfYA95y!un#$WoB6ESUb~LB>Y$Dfs!l(>GK;KvbpPbSw1>
zBop{4M0U3$nFqq>^^<fmAWz5{;#G0t>@JUiT56sbz-kZ+^bsOr>l8``Hi9$*I6BT5
zcxUCBJSjQ%-*8O8<b;O(tyIG48z<Qx*Mkt&&6{(XZR8Vp+qoiU2|dpYi!Ip6=vm{&
zqv7;&Kp0N5K%=T`2;7u5YJFc7WA6$~7L)rWK*xMamm6)LU6@c~mFD$+ZThj&Y0vk7
zX+0FpbcR=eFp%_TZo$~(hcgi34qmzn;7>V=Z8CdU9RPAqMwWe#BVo!h3*t-&7ik5U
zpbzr$>+bgE`Sw|^|9$6xJ0*bVf%X>Wi2u$3horiz2;B`K1nV^dHuNc%x(?{3Ked;@
zneEG^l&3ru_6$Z#h1kWflP`a220`K`#Z7h)q3oRE>gv>5et!dZQ}BKu^f`lwH5%B*
zhr)$%hOXuNPm9dC`aW1iESzP1aMN7^(TBI4LE-rftaUqFft3klAIt|U1foUczEn*!
zXw);AT|JEp5bWrbzirbdQ2&)VTcD^aOB)Oi#KKuU-=3}nXPsMAGQxX00Y<wFlr`-R
zii?I0@7!86y8fl}_>bq@AM%y2f24}QlDYL9@yo>N@sXD}Njub!nDsrrX<Wa|EbsLg
zVZS0`lbe*OF*%x`K=wnJfOMBi43Y*q%UzlhXA<pF;K36GyPV8csFTPtf_BzbNk7wl
zU+xpB04s}v$LD}}bO*@<{Tadl7u=Dy?@8v82M_iLklnt)F-7<H_b7b#8OSN}i~cBv
zgwVknr-osVCM1t9|F)m`A`)Ot7joO)S1?fS^`Z0VL)oCKE14D#gspc+5yYT$)Y0um
z7ig=4$4|?y60ssIS=57qC!59ZLjs4JnqI;Luqw;pWE16x&=!3`sgCB%>Owfhao(?w
z2FWAb%6{0sgEfud=a^A&?Mu^-&1XNN#Gk<`UjewWA7uKD&o)2JIU|0Li033vnsb@-
z;sjTlAjg6(jZvA|48$a`(4}05W&K78yWI&!O*iY1TF3|G@|^xTaM!HS#sX&F2b;ik
z0>yDS{f$k56a|}$#xO^B(8VUbhf4xU{#@GNexn->FGq1@a6^Z4_B3V9lJQu_;^eE`
z2>WX=wd6mR*F{uv5t=nEmcfuJq}p`s7`iSyI4TWlm2ARq9*Mts*rd4$x<)>UWvl4C
zK#vISq!KW^tw=QJuJ-%t24D{hHfjh8ol1~`GmR&HOTWG`rH&SBVjIoX3*Eg)Zgpe6
z^~2<+TJLB)3_R4A+K8K4ETRwIW-{+WA29>?iX)4Io@4Z|dc{eXl-qaeNVUYW+|8I~
zi<qU^C12kFQg2?u%@rS*O3r{yE^ZyqeJ`cf;XJ8fMUT_NHq}mIRG{OAJ#OyVkA`|^
zw8Xq)C*Tq|HX5ZtfK$r|-GMDG9xsnm(#19;7HRH2mBh<yg7RR-&e~3h58z7607gGL
z7m&1QyZ?e#w9b$wwE4buQNDH(cn+4bLjiI{{GZyyl4R^UtrV^UgZ-wjV`aasMAFeK
zI%MtH+DB8I3VJbb{-Ff8#|sPc!yOaTYrj;5_6`&pE-+Ft)NrISMQln6D~MXRXvJ;Y
zlh@{Vz~kw_eX7&c)%KuCE(O+k!%X-OIOsp@0F^ylCVVVh67<I2VMbTMxUo@Mk7|M^
z{p!nDoYzOUH?h~tXyX`sc0l8P{3ZDWt#8#H<1acv8+W$n@j;S}@s3k~tW?0BlFO+l
zI`Q?vueAr7_b^8$adC#z#N6#QqZcp{ZZ@k}OYB?>)_6gTK(y@|VTs=+n7gXrVW~Ep
z3OFU6{Y~R+!X*e+W4GY=fv|A-tjpmlJc515fmVbofxVMWD<KbtC9eaGOS6|!Z+|E@
zAVc4~@2cxP_82)Pg`aJ?XJ?+X+d#QD!@-a0E}&I;l&d}n!|Bcv;K!dpbA@Rq>Dx}3
zY@C;-d-x4VN>ro@I`?sedpuL#XY=$c<3tq``!|Hlw_=xR$A0V+1Qpqg1~2w|BIG>~
ztgyE4fFwZOBASjqa5tS@Rs>HOoYY;FkVu=#oeQfMD7Y-62!Hyq+;z~EDI!VejMoae
zIK>~N`;0`;amQBLIE(Ql-<V{=3Oz}10VaePlMPEcWaD~4K=$rkqyiA2ur+e!h;49N
zrB-q+u$*D`pn3&{oAw&PzMbCv?(jMj>yOM1_S;F<-o>O5@!+!3G|5C|#YMtya8aD3
z<Mj1rduxD==}d9B-Vt3aNz=p}H}CTUkc^Wjm^>I=<gJ@yCx5uq2NCOtU4ENrKTFQc
zhh)w21h*<!?ubvOiR1;m%m^-a7u_y8Uxy!B6l@dKOb$uwH>F8;+`^37{F$k%h|CXW
z>gFO&QJtq$G2-Btc*gNtxxVzcixc%{+rf(N>O8uY8i*tKYl-^5{#tVsamOfBGdTXM
z`?!J)KE4pKquR(B#!6WWNl`pm?r?xRvRBRDP^4>GV@Ivu;?qDMRw1_VP=+F$Z)*V#
za*iKJk*Lna(HM)keRAAow`8ZJ{c<oQeDbOTM`}L|{81E5Ff=+=MwNBcot7?)cBIM`
zvh%2__n#=yQ=w4dKpXLCqj#xr)qu`cMS)LIr$X8K5fW_Cip}Agj&2XFTY-yNO_pyM
za{K7pmXb&z0bw&u9=Lck7dmW@wW16i9kp+ioExgrLFKoxBV*K*(m5<iW<uH;OwFvw
zxNPd-!B4PY!+POjxCp5q#u~w*IfaJi*9_7=aqe`l;%dCX3j?$LS<U((vFD8cM^Nan
zYZ6^oVejJBs-PrEpbo}~qq9`ja&*4ITW5bjJ6>Q?V3^K%eBh|;flDz7w;49D{v5m<
zoXav`ZW_w&%l<;Fh&dZ)-K+qRb@E*pHL-taQXN>=sb7WwJtmcn(~}!e<e<6*SdQjK
zlp6Az<n_JH1BcpcOM7z8Z0ypb)fJCOZrKtATa2S^mYh1dC)&s?q91{EodjB_Nekv7
zvp|eJc-r?S;|1Int<uSU>pRKA3+Z!2@x<%zIsK*HJ;i1YwqGxIiTt28{1cyMWf`s5
zO`HX7*pWDW7E_I=cg25r0iemzN}NaR9UN^t&n!CH2ITsFN1-#ab;|0lZS35)6xVro
zGS2WR6h2EyUdmsm7-ntXckVVOr@fT9%2bB3idf~IeNlD)I_?d6`Y{)o($ZU1F*|X?
z175~=`JS|Q)hW~UfNSv=T&zagt1Qohr|LNA`Q}b>q%WLlu>3B}yTI$tZWcS3a>Hl!
zRgI+7U|vk8RPOlOhQDPELC};2#SUj|{2xDcDvIvAjX1-Pk@z)YRRrc4)~uPYAOz!%
zHM^~0Hm0R8tuvx`v3q^y_K$nVeSIa%8VMU%1#!cXxb3|JpRoKeCxrsNFUk8<BQO}O
znYhM5==B@bGG59`n+<82$!RDoSd6%z8=-5C?Ch%<y=;=uvHN-C({sX14D|T2q&507
zo4zW+lw%;&)$Ih!b)k8_D>i**VCLHbhhXvypu-Vu*KH;g#f?im*@aVp6wub`@tiw~
z??Kd2#xC#or<qt7m}s1?E6l1r%j)v6ZpzwJD7#QmjY-rpdwfEU@J(_Mh3Gg`c7*40
zjCnxf1bkw_V^&shc4-Hs5X%}YJ-c1yh11pQwccpvIz3{2bdtu>L@HwnTfigyo=f=Y
zA)DDYTV7!yZsEa0tgcQEhQvh=;=#i;?D}UcIqyduFjwftFfh3?Y<LN^<d!$yd{%~K
z(GjM{C<?^54aDt?#c}T2ju4*(iwn2(>m=g2v9|P)tSck&F8v;`@XAtZPu)cthkjLq
zr)jf;1`<V!=yDgtor6bsnH$9-<>o8w=I}V5Qdp8#_y)IdFgFrX_Ododa$L+ViDc|L
zOuEiAMsH~%cj*ehcHJF?&`TrTT3DSY)s{M$_7d4g&gPBY@M2>Ha(+ZxnWkJ{b;8}M
z<TRgzS2A-HmUTK-6HkALUbn|Gbuo~1@a-vztg~NL>=we6x7bPKOes;F<)eC+#=h!?
zM3F2y#e`V~g{PWoljdhA*d7dM$5c3EduZ3*?B4wk1u2MAOX~3dbJd-~&GRWmU>Y8o
zxTEPT&q`2I;4d03%QYtH^Pso#)#4ZUoFg#NS_<MuDOY0yY3X|q8^)OyRPnW1A@a9`
zfHMjl#<&A0-9pDCD!PVt$C0=`dg~@<6-PAvY23Wc)$#|Cm6xjLn~w4}<hXUudp?k-
z4yH<Mc1VoJct&5J&}C{nJJF4TkyV410?~#&$XXSjPVn)3&C$K*;r{vDq#hhjFV(f)
zH;3XgI4ST)J{ZG$MjDwo6+GQP7PaM1)a2zxoEk>jLU<dTeMaKe=q+1;J&_TCnSNwU
z$)()1-&b5mR5-RIS$PL+?cHnmazhzB&Pe-mwN&lVlu6^KKAYa<(a@$;e4$?e-*7kZ
zO8H<%#Y0T*83jL6;(K}WskZD6?gp-_DKlw(>Bg=1;u`LD+z{s!;Mo5l>twtr#TCVe
z@0E0QGJ1=&-(dL3I7-i3;E7CD^uBbZB8=ArVrWQO11;H;7yQp0TeZ)ozBUL4bW~Bc
z2s_8(W36~szfuhO{<wzh-OrbN1KvyYS<k(z{NqsgPjYl1s4bx>Oej7Cy*~GcRo4a1
z&qR|;)SVz`s&bq+F2IlPGM0;drq%MA)kBvfiOK6k{<Pk`e*fZz4addFbEGVz*W4WC
zeL(0m$|G>ADqc2S(XSm8Ol;M=k(fkgJ9RNB&_+|0kT%dBWfP!y`I4Z=8pzRduL@z+
z{0_Y~nvP#|$%WzQM>9Whe2%^s%X$eVH}+(uVNzL|{P$CZSh5FdS6jrAt}XLjn!2B*
z^P|jJI4v0y7|m@{&bw$&!_<n%VWkN@mmgvm?5G`*8f+s>27+_NjR)dwA{7g=eZ$^3
zi}7En4L;mQygn;5Y$~=(*~k?oy!wG+pCD~pp<hmIWIqa|MWxYRHSJelM~UzGDx^w6
zRBF-H7Y||m%A#dP<HTz|9v&U$Hcq=3N?rE!P^Iy}9kC_2(~q=yzKrt?x5+Yg^n|*c
z@2Kg=bYZtv`mBC8(82UE2gZn(^w_7G86E5td~{bEsht9x%)d$fTpSjI%atcao$lW}
zERMH!gk?MU*yK>sZg1Z~1(R%6fU<6WT}YMK$Fft|tugFXJ!m#viC6`gSw}T!-ZgT(
zTWO<&l6B8PLO*lyk7_KPGV@bUv+@$>Ux+t36+XG}u^N;qshBN^rcg{}jN3T9TRP+`
z?`nmQWyVdp?P<C5RnrLWUi0{9w{+q4ZHNMIB)$?fYyi@ge*N8sSM)=IE*37-pNVNl
z9p0%=q!R6%{!75kpCy&w5d^&~d~IO)dogvIjQzLnFVHVKJXJz1zk$)5>SC^(fJNHg
zV5nDJa14wh{IZj6wyBuolIXiDIlJ(pwheV9>+xlzxQ|-yj{Z-PlR>`Xfb@sw3MzUU
zp)TWN-S<@vo%U3K0qtJG1`)^3^D_zpig>2meYM?pDWAhEF;#@Yn30@<&AoLPB%)bP
z3-&bPX#05^whqF|eq$R*8kByuUn1=JtsF0)PyvhpS#Gs(vLFF>Lj0KAyXxo!V!(6<
zCnFgcU|m2GR2;lYgTR)NvdxX1Ea`A?>FZ=`i{K|B+*hDqF`Kd>#l2DyR>@JJU24u0
z(-F<%JGo7MqB%R<<U$d5w*J*tRkC)chPDeOa5NOd(cpz~e%hA@aFkj!cjI}|_Z1=l
zeMkD>q0HXIF8=!1b?4SptZPBhn?1)KgLu_#@B|)u$oL{gY*Kg3M2pP&o4HM*fr`F*
znc7>`P+kLiGCHv2DO{(oe-`(6=Y{`X-vI*VU<g3~NeTfTi*Eqov%i`twI=FIO*Z9o
zTT*IeYoW(K9rFyxFoe4BvD`1L$WxwT>zqy_C&&3G+@IMs+aZF9{b0gg(-R5ml+OWb
zeHXyeMN96t`t7c*!Tcg<j}L0@D}6z5DfH2Pa>Io>xIywU(zQivgf*b;p9n_YUfhDn
zN~YV2P%D+<pH~bxK4OH=5h9-j@hic#BD)<C>E*inb9FA2th3#HI3smR`K+@XnH%oO
zKQpR+Tq&-fPK@FoH*`grpC;pSJgzGqw^YE*%(+9Sc}LEc;W8tUdwY=Z?Dk|7c*A9{
zyAP}^b}5ISLCd?jV)h05c6j%1^zxIDdY_heUGc0<kzO88jaflLHeJ9b&gm&t`y(@a
zV`^R<sa^8fQu)yQYyW*I{-Cfru<iof#r@Q=4Y!vt=3TNA>9J;}Cn(tBR-j*alsq8~
z-gmTN=hYJ_PoeD__Ru;!eP4l0O-Ynl2k4(vCap>gv&SaMgTbm-(L=~7Ir~qoicm;K
z)wj~=KMWoU&pv-0-N?zxjjAr{C_w0y5kV1}D5zcv`l@~C`g|y#7xX1$fYf;(J-#{7
zhq9D--eyzSgBx}Cgd0ogjG6zQ0G3w1G!b$$uZJR-)T)`dszbV7Melb3mUMa~E-N14
zqP{ofx0s+y&Ps0>vwpo8bIdb>+l{;|&UD{u)mBQ?jP3_lSF7w8X)UvO=S6X*vkZ8)
z3trl&+0%K#S(Mm~6(9zrodLfnYF~hx$Ga!MLnQ=8KGK>H=-Jtr{WdW%n@CRB%c)~o
zQ%Rgkz1^Gk1~ldLU7o%LjyuAaU&|Evoo<QnYkltHRm$sj=@A!U#GH9UuOR|6C8+ep
zU-;bXHArmOiQ8Hm#oJ8Cer_vKS2QORl%J!wPG6)7R+|W*o|pNCC1YaszArFPJ9%$z
zOm^62ab+?;0`syBRZw-;aJVShmvv8?%lQF6KtZqyj<zG+M+XDSL>O4NM<7;+uSU37
zww&+$@WThL@0d1M=Q69d2z7w}<{C4un7f||h?stPUFhhOl?ams$GKF`fHnFCHA2l;
zx*2eCA~04$a^Hv^{cGRz)7+w09J#6$m=~uJjGi=$G7y`$sk8c6L-v96WU9q)prXH;
zCooUxpOBXa^d%W4ZoQ^|j0sZ6RQ6v!PC@COi5cZDI2am+d4h$l*YK0(8qohNYq9^X
zI^`+L0f`jAx9Kg5q~@E=`e!05=~$3JRcla?3d>o1J!?vF52K}}Kyh(wI0M)G1SKSV
z`f@W*7_+5{%zUme1KvGJ(V%jcE|MOPCe#5m#@N-3N<_adi*k|?b?nJ+B$(`%o!VOL
z(mBG|*OW_^jPVctt%drMC$Pe?h*Z0gvun5521Vc%r1M?xA{0Ij8T;`p2St}_lDACi
z?x$eOIvhyM#@);Tk42QO@H}5e3qTkUQ0d+cj4l&{9V-Q2>n5MGzwcYE_x%3jKliP*
ze6h4zfeRKjvHhich~6@^@(ilWsZ+2pNO80+d-G<+`=qLlfX>bPjJu!fNl#tpCjSly
z2~?bRxbL&|ylbda$6I-tN<=u*1;I1H=H;c88tC$b!p=2a!cYZwbV~I@^vgf&Pd^Ui
z_ABi*u={oUCGiITBqh?4MYO6e9ka|6_l3(wWkv>i**geKFjfhjb^YUpt;rt%puQf4
zqKlb=Mz@G)HD&kEDA_iQas+#j=gt{%O>`IMUWPYBw7zKTjXLOkJ-8EFV=V^X+nG|Z
zt<rsRmd4wUw>7HQW7RNOX6ynS-yjL%ES-7Lk}nc9Dh|`)H|vmXd|0hXWosRfDSECb
z2p!-q@04@GB|?*H#<f38gG0>CGL>%^@t(+&WLnwMq_MW5Tc1OnL!P@srhT8gX}{Ey
zy}=1_(!6^=LME7%{9uOl8OzT{ILJ@{$5Yeo#h!<4pMiwNmgMaqu)-HHq_G*n#;f<#
zX}x4&${RnkDVhmcQ!1~OEQF?Q%6(%+tl}0Js!QUNUrFsb9&okiruh5#>A0N9aE0Uj
zfFN}8-|+|N^}!f8Bu3gtQW1cD<Z4WZ+qp%l{sRf$-8<pnzfmFb>WIhwE*7N#pC$Yf
z2bXxGS?4DhYpad&$B=>9?{!%FO}Y*hZ*3>!ThT|YHz@~8`T?(NyKNYGn_xJ6c3pBI
z1(Swk@w7lwm$pVM?H0~`OqUz6hI9Ukl@E_KHF<A4(Qf;O$tJIhgM5>|=cZC)UDUDW
z+C92eFfDIEfbCgRhX%(=odoxXcFOWf$)l6_L=yGF*!IhHxE9Nu17d0Bq^C6xDGzz~
zuhOh89<&OO$PiF#uLm275!Pn;+p-}mnhc9QFe*7V3`rNQK0J;W9gNdl-oZs(@p$9V
ze;j4}`NfcoGZz*3=YDWn{>gc`casY7rXa;Fw{!3D=<ib4o2d;#^7oGnMLcxhZ0!kJ
zJpR>X^6h<6TX0fudoI`F@%hmzk!nZbs+1^llDlE|6BKLl$hO&-Zg4v%eLvX0UPJ+D
zldlv;jqlBbB?;8CVER!8!1TuemL|RjA}nSJ-a_&Xlts&5`??g$GF-|#q-~ylmulkX
zS{$eEP~(6)3Rj(DC%lvrwQoLv$zld-XlXo;9YT);fs(u0dex~*i2F@-&Y?4kjtSLU
zmm3=0KkAs9@re=6IzQ>XAc2pzLHvSsQ8QIxml`NbswwS-KTz%?aAgVc-N2ZXBXfbO
zgimJf7a^-8TKqw{xt9b!HsRa*Em;uBp*QbBT#(C<>Spa0^LTi56kFvI5t|m3coQyW
z;K@+82K1X`2b7TH2&eY-jdi1}N$w9}H0joT5T)6rpl7P^iqI-)*^m(Q{IYOCGNzDU
zLH^S{R1a?&QR!^ZvQ75B2a|KPGiBHcgD>}7CGYGRr5Qpwv~?<NIuBmFaiE=GPzk@S
z+e{|vO;xltZ)QMQ?wP6cOY%*+rY+gmQZee18wN@Kx&CGuF-kfa?KIIkqytP3+IOEE
zRp|286lr>Zy-Vuei7fL&n5o>4m~RLlNfW6(GTKh~CdCvp(xo=ad?O)qwgS~IhPo@s
zEXq?gOz^zDrRaDMiNQdoijB)ok4x2$DMPt6n{0JEVt!Ir^~O=&)Vp)}B6VMqw85Ps
z{-=`+6x4P!rn$Z%#SP|5_vE}6?9Zd7C@d>jJh-;saZlU(lblE=@GG|PRHLhpp*$`)
zh^5_?NMo^dFixgHDo)fFagr$s2QixM5Gf8HvBLDLhPEkE+uJk|4|7S&@orvX8Eib(
z<2@m()tHg!vV|0$rqq2qBt6S$Lz3c<VgLM>?t?L<uZ(PmNL{bpM(?(L2&bNN7_w=f
zs*@;>@Rs&?3zjcV;9_vE5KJ+|R%^JNg^im0LrS~6_cldG;l%7}@=+eUd;v=9XqW1*
zvH!;-oF6&DpNt#0{(gk>#O@U{wDA2&09=8gSqkQg_?XbxN&J>A!U5b$w$!Wxfb7)_
z(&G@wv-)6!Zn`YT#fwKBDmy}CCmy@JyNsfY?Bz#La2fL)zR*!JhW4jmBXI7pPqH({
zu*yqk0PkQF!?K31$r7Z}*kadA;sHB<t-~YCU1y5to3U=nBST=Ss#>GLT|`u9ie_PO
z4-*-7LXmeXaGa-NHVgN>`8gy2F_k3@nN;FsNZ22#nDWR<bj|nN^tK|Wxl@G{pJBG=
zgc^UyCqHlnJQGEZ-Jlscm#Uzb=5~C<L3+F!g>yi4@NbB4TuhA>jWbakNI-@rHeDpS
zG43^X=O={8;~LkZ3C@l@G+{>@na5Hj*;yCT^^d95sbq`|NQ#}M4H?i{rV3tqq5NaL
z>qT2OT@&*-bA(vrk_1d1!zW)~Tk0lKst#-cp#WbxPS?BNHriu4bSC^FGG=D-d9Zx%
z6D`?0$u5P(#<h!?qYkwk!|YP=$z91h8bE>ah&~Cro4I^}I+npFIS5fYKy$1NyEr*$
zlk(n;4KF?J=Ag*@uI}U{O}j5H5a*6d2aS7Thw_gt18iS=dVl+x*1j=p874^On~j3l
zaWN6mIb{0S&Fd?ohtZn~NHhdO==nKbak<NW(qawFuEx~_?;{kP;~jZ3YoBG}kL!Jb
zTKF<bNOa(ix3WuzHwD@cZ5`I-iE1UiQT&Kk=<pAByA8G+r&K9yqB->aBzvo3WXZZ&
zd=q+{)1)Q((Bw<M1`K~5*#Gv2$uEeb*7t+=rhgG0DtJO3)7~j0S6*3s73d!^8Th1G
z6i0zxb`oIUWM`*8fumB@*wy?SFquo(?@8T>oep?7iaQfl_4Y!R5?Ttah1NgSvp9ZP
z)r-wf*TM6ZmM@|s1^cPl{5|6PM<`e6yYczKv)q(OV1Z!PBRZGpeEpB%_BY9luq-~m
zt+R!nWBR6g<T00frD2B2?o(shl=04?=0|+-qf{;p)=ppOtr-XMOdRtRY5i#L*sF4!
zv}Le4wZyJxklQGm^{$r#@GJfJ^wgFc6GWPF3O1|w3G2b{z`+)PKGh+%_~&{Vd*zeE
zn!*%&-`;8`w*De^^M#mckye+T8LeiF+s$-zf_Baa@NVOj@em%r`~HSfbWS2q<6}>y
z=8U!#iG>I4TnlH%iEyhga9S=Db9YY^L^OOHJ~Lc2zV2dv$1rVSKgVA2jL9gfi_zW#
zNzJKGIt^mlQ(zz^c$8#ht(Qz|aA%b<FnT?;L3MkL(;&5Y0inwGmc^KL*7Bsq&L+=c
z8Rkw;6SVv1<4_ak?M69@zTTommtHd8!1~X*H;Jxtvbrl<B`tgpKc(q;zB4Xa6_=6H
z$?E&Wfzw0VBq;EAe#L)5-*BgNc!0XX`?O(N_@_4XZ5d?&wV06Ueg56gpliGAU_Qp3
zC3m71LpEgvI8mg{m1+i-(m0LKWkmzS3C-YKF}g$zbpK`9w?HfGEZ=|kuAxrKoLpLY
z#$V}n|GCG1n~ErsY-=C({QACP`YCUw%edQ1P{9w9?>R=R8;-bUrE=;me=pM24@?TM
z?td$}aHM<4=&aGhYgEWI(F>$!=n!MCWrVT^J4zyn<b2me(vbK~X;=z_@%{(y&d&$4
z<cZi~`q7-*{?+%G+#+k+GzVG$-L?}5c<DNjB1LxxSTqBWedi3+A>FnfUZF_{TDjc+
z(KFMA%D7+=yrRkwO4rJut1=SWNqo2Q_y!r>A>*?rwFRAOz)3x)cEg#Rf=1!{DZ$}1
zp=p4CYwBqS%_nM|<^Qp?|AlZZumPi!CbCua_eP+JoIETU@hr^>+6F2m#6^r67hBYi
zv_M_`=q`{z5O$N>x%hzNZb0H0fD|&P9()*?h9Z(Z2@qWpVsL%akF;q$4vO!$2GpK?
z>qU<iE>!;ssQt|!Ob7_M(?ve<Csh=EZx20CDl6(CY1c+76O^*fZ||M9d1LzJIf(G>
zU?T2wnk=UstOlytr8<F>Q+RH7A}~|$?V|F0vtG1I^aS?jjrVV-(qDLlU`&m$Bj?Bp
z{%I!lyZHHkE&NkXkmvj%bv94$|7KnM`|nt~imVi$J?EDHp+o*JBVUs{$TM?3nP~D4
zt=xZp?)SnG-?LZp!3F;}+v^Xe&c8O_|K4!_`p&=q?0@=-VmiEh)DXG;FQ)9@e^&qg
z-H`VPQBG=5JNI8kham@fMZIP|_^87mResCZ4@Lr#6rh<(41`gzbg%w-<Kc!tf0N~?
z=||OhRN-F#`N90p$9)fRBL4a^H|o!is78()9S=N;qI2z_AMFS3loLehntw6!{^SEm
z^RvKc6e<{kVURQ6vj35R83-H|gu!7P&vkPv^&y7+&Jk%J;md~|1{@wd0k6dSw`xr$
z;t?~z71)R2K1fcp@!U{7!emC=Bs*ypA$&dcqkw8=z*rn6a^Oz*V-UDXL*do8R9Cm@
zrA;&YIqF;0=)qc`S>AXBQK52c4(9TkovBVW(u|2k78JhiQp-x^21Rqx)z1t*<fs4L
zXZMyB!AuVp=F9zB!b{8)K)|VvIU-fy7zpj}OB|0B2%iR_YcyW2De@da_?y7IdVAQ!
z{=Qk{H5eUy0Qr_Zh>@_(YE<e#jLH!LSfYAtI8j0tDt8y~dAP{|rDkm<j9B|2a@11h
zLf3EZAArGl8Hyec#ivq%vrsr$fu&m7Hb7%>hK0*((BjO1%VF`?bpYT}kdXVE{zy`h
zKa3C%@A^u$!^Vxmz{-4CePP%KyiR8j`Wlb|B`Ke4^4{&?Nwwzy>fXEx<wuZ`xxmDu
zxjq#pK?O+8>#eIADrokFPrzrgM;K+=U2V0leM@f-HcUcPXiRnF`qaG}2d6(jxHdZ@
z?^HB3ps!!!M?EIi*R$Dvp{k*rv~%HN_8)}A|8;Ks`s*og#7UvHB4?WY*KNs31;IM|
zydYvQoOv1g=CBG05Of}zK3h?FA}{$RE~&DW|Ir6n5tbfdg(^IwKBEA#mqFJ>>;!{O
zvkW9Mdl2T^7m=oi-8=4rsf!dsG`Fgi{}ImT@RO?<`ID<@{1u*|$gxLuRscpkg&F<}
zT=Pj`hOS!h?%GRnv2fW3$e2jGyYJzFY9Z|(t3f~P5OSVidE`tIPuL6KZ1xJn^IkfK
zMs0zooC}Z-YM@INS=q;H>T|<BQp2a`8W4}@W?gTw*iyY<V8`c&!(buZrJ`HCx)0&$
zN3WdO{Twiox9ZSaeedeDbWxO<YxwlOys1}qgSl)0SRt3;MGRN@ITHfT+I2{o)8E?L
zD;lCeO1*V_NX>Poz4}=chi>WL?cjT+$oVXkW{~)69ynP<i3|IweFR2^4J5TxBF2Hz
zP^BL#h&rb#sTd}gw7^wxXu;I*Y)=Tz3B9Y0GnaymuE~UZ?`70-QNkYVhl8wg6_?1b
z!bQ$bh5|FW#IoadZyXEvbY~(d6@viNxBwAEI01<*D@F1<n^dJoJ^~)b+V}Q{);sVS
zx`NPPYhuXOfaLlHVB@oc03fSysN*_XW1C;i*x?h!TbzD?EGwD{4?Re3SJ4xlgrUB~
zl}x|s6SdG^sJ^#=Zl+eRyUEwMrqxWMQ=n*%7a#Iu4hnsiX|5CK*dD!C;}`aK!_p-q
z<mz#L_h{tT&M!Rk88LkXPJ0tBQZ3<`^W~J!faZMy=f^6Xj~6hG2Z640w>}5!;tIHh
zy0t1qAfCi}$L8GAb5hhJr+w%M=EuA5_2!&Qstp7ppvoyBv#mjOZw4~s6+-n0Z0>Tf
z{H?C28TC}AIq~4VlGs^+yl`OldY(WIBl7~41Bde9l^-}{`Sjm!%K!P59z@a*bI&v7
z_xI<?O#y=#2iXYP=7nN}xSXa-7Qus~RBKe=$bdOhAD}vNkiT(Jn{Ek0z|S8A4>#Bq
zc1>jvt-2I}z-|p6Y^jZtxH@@~j%we0mVH~U)yRDr4vTgRy$?KDN4xhsY4=(eYps@>
zeS_^r{{T7IjD%BaXvM;FDTBbPu^&;<VFgRkJ-K?QDY2xKK*-0PfdHh4as7z=q#+&2
z8c*4gB|jyYgJKZ#l^E^~Vg;m>GzLyx-&ltYfjOGHDVSp1Dy2xc{$4<{^pLZIWF2z9
z0QEb%tCAi&L+^lZ6z}O_X;>2RiO)-Q<j=m2&V0n%Yl263hteM|-Y`0wx|5dY7#P~H
z3YFytM2~M_vev#g7<bb_@pJIGO;~tJ8|q(f@gGHx<rH(0#FcQTsP=3uLo|5zeY(}M
z(P(=QIE9N}TrR36@$=6jywzFag(c;IRp_{*sIq&kLaBLhzL%4XZA(Fqe6PrFjD5Uk
zfH#tVT7LOG=bTvUu@@iXwc+GQoA0w|^=Qc24}ghtv^Iy>OOL!1+elnl-7XZ6X^c>*
z;$mH*!(%C_DNbeMYd|`;8e#=s`1aZvi612at4KcaoHXg!5SBF#CM`_<0NTb(*0p=|
zM2U;8zPclBP2ilviRbuAF9@HVbE!b(2FCAnn6;RRgA>FXW&O|c&zs{oM?cBHr8tP>
zHZ0o74SyfHXckXg9e{I(c0B-SF=7_>=QsFPiIq?CU?i+QH1Q<qVaC@Z+t-sOK3eMd
zVvAiWEE7rDe_x=4tRT9p%($Z8i^~vU%)P_pwCv;$#}J|<c&k2nd~i$De>CRg04xdP
z*!kErxO=i<{Q*RgH3glL%0<4~^{2a<-8t2xOW?(cT30SvERe!CfUDma8--I()=u~_
znT<a<f8a~>@ke`TQ0etKB}gRwAY?n4!^|o>b;e+-;rv<I*NE!Dh_nJ8-;a<E`5R~$
zst%p_?R1D7&=CD9K`3R&C>u*vTs0DjE4rl-@%GJeg-g#DG7t)Bw|j>IrK~eRDm1?F
zfkv!XpOYi}?lRMa*xt`8a2nsr$0c|}$#K~He9ie{F12mwKc3SXHGhQ*U@tWe7mp}J
zIvVUVcaY(JKM<Q(N8Uh?vJ<E?s@ccc&$FR9LIoSvMi9FD002<to<XMa%RM)!yqCDq
zcO(tIGBgWRFbQHADA<;jfupLLs)|1STekMYJ#;caRFgLabpX~?Xl59wgr$uFq5A?O
z++OesG6Va0^mw+bcS1Vax&xFPk^>B~RMfFGNy0HPo*uB_FqT4xl*&WGUJ^&64~!|(
z2OxzDt=A5i2CA9UzH!uy(UR@20}k9&U<^29h_$*8JmkYxC)?rB)u6S=dHr}Jr7vK2
zIG}TEB$p;0>KA);UBE!J0ua~}NrVTPoWia!Zqscl*zNQgE#aZPA(JCUGAHq(Epqfp
zCrB3XXL1tKvB-uEE;T49`v#TNttE?DUCtQoZ)YgU#?wc50%Lf8sZ!TO5~+8xR;J@Y
zw;w`doyYIbp97<$%Sf}=+__~%G4}b{_N?m)pihIj1dAhqch`4aO7^9WNUe@t{p|-;
z@Cp|m@zWw5O^z3tQy^)OEUW_>2(%rNj8yOB^f#QXiyW2$EvD9EP);kG#ugqy+2cBM
zP8F89t6%><*6^MeTpLbyxdOje!}r=~^VP`eV2pwIAU%OmClMg|J6UC%CrTv0Lm_vZ
z$kGSOI%R7bG<((>n;xCoEQ}y=_E*$c)((zPU2cZY>QLxM7@o4cWLN&cu$!iI4rZ+T
z?uRYdGSGF$@~q8g>Wmd2FkC-&?;tXb(ICIMYt~u+PCfr8FDSfk)*)Jt-2$yK!NHWO
z;OR#uLwDdSk`=BEKCFUA$J83r5#A|itVKDL*#+31v757Nl+8>-g`sySFeX0&v1rBx
zKn1JQd=X)qXdKL?QF+d>_cL_3M~VHQXnZ7&ohvkq&)UHqX+Fx8#V`ciZuTo|Jr#-h
z0(#xx^>&z^ZGUX!q-nW-qw>HJ&2gAX-M05=4ne<s#NBI(Cn_KdtDY)Y;{-Yw9jelr
zgNkOYdQM%s2o+ob^QSnsQlLug#-h16F6&uxgw;b_U1`8cx>w9MaA8Y?ItJZtdm4^-
zU8ZEt)DD@^Xf?b_Q9pQ&^1}dY9%MXtH?8A5SuAu_&3ATGhIvew);Si~c2^RP#JOuN
zK7OhrX@akN!}s!tihe5MFicwkH^$o1x!x`Xk^sqs<T*Ac_1=h+OOV4v94|J<d6iw@
zJSxja{9vrt*SL~L7uY<qQmT16ZdE&gu<Ul}>$<iz^c8N7VJHw-JyxNjWb$BMp2RI7
zxdK<J$CG@d<D}k|zVK0XM+axcu#B=Mrus5HuyL$F^((c=hx)ohM-V-?XKQp<gW=o5
zRQ`2sQ$bWI0lZ&Q>X?!<9VAxb;v&|fd}@Jqa8t6YHW`=rohwUhF6qmw@IgJSP~zcd
z({}Y9+7fe%1vm2-(Wv3Bh2+hzf?luNj?Nk~O1wM$x4nNzCH2~mix>VB)!2(sz~hi{
zYT|*O<*IRudAp{<OZ^e$m%OMDUbpTnu!Y(XbF1cb3pwHD%kK8&(c}?=1wEcBX$?Og
zfxRAqxh5djbsH0aH8astIHEJsmX>tayG6up>V(_<q8YQ5=WxoWRM{sh_ViTlyY`}q
zv$i(@FA{XDZ?o)81V%#3jEt|Qi=e1&HltzY%qN%et?FWIRdj+6<%`DbbGhOBGKHnN
zglD;hgSeslF&)tz-#bA(V0m4}jUANTI)rh00w(f9&7ruei;U}x=4Li;)FK-Rfrajs
z+pmPLQXa+Yz2ABL%>2v3xsRhS&n>`}e)!WwTq!+XG}&h?FW<DgMcOO15K|u*n&so!
zFge_ib5+H%lgFn@l;Cryu5owQ%)<}1?P8SyQ{7{5;OHon8FWiUpJZ~rqrCNnCSu>M
zh<!KJCgd-_;S5x-^|8)REnnr+@3vSn$D0wVn$LR{x5<m`=_U!saVume^=Lff?|7P_
z6o*l)X=QftYFN<7iioPF&J<QVeCb*XKPiyuaTa7$wq-4{R=?B8ygDV>l;8G^I#4Xl
z46S2krmet#aYX#eihj|ca67N?S)O*6+=%&+m(C;IXR$itHJNW}t5W=`l9{$&t98Fj
z*YWpR8hTmFwvBpe<QXS3Ki<{G?1O$@HDt5asMepeRR3y~mZ_C~gLjT;!^!{Q(DlSO
zn|1Z#PrQ}zb9Vi9lH#j&bFQ|vDc5`3J+2l1Z9R>qK#S@A&ou@7fBYa`VF5-UvTTdK
zlC`l5w#k=nHFCa7jG)tJ^ljl}<Zj5Z>rGu$-9EjCHDQguiqS@B>^FQnL)SvHC*x~r
zJQdUU!wp*HZUf5aFy1zJyvt^lTTC_eL}56_LQ7%c63z@IRI=oG={R{if%c$9t~d)5
zyOvK*4yEs%M&6NJ496KL<5Fdsg21T$JuR1lmf;s8)1%88dWj1U4KvtitVP2#8e6nl
zvR~h<P#u0OC1r|N-~Ak#JE_xE1qlJ`pY5X%IDV!AQxWMls(ZAp(}u>!x+Sb;J-)gB
z6>Hfi9OX#&J9_-}WS@j5;t^t|8q5bE)Uq*SX5MiyX6X^JZhDHrw<r5ibhkd4vvq??
z<l-b=%#P4%aKe((27iGmu=dfu;JP=lZ{wWf-eVbFR$!8}V@uwsjogjMA?YgZn~R6y
zN$RxlTysKSt&EUW{jvri_tnk|_Ql<JtsnP8F^UH+$@mKmQUzB)a+;8$!3T`s()bP4
zhAUq;v3{zG9@-Xji%b16*Gco+f}1nyq@9^<UzfhY;sw8=1-gI7l>YomNQTNOEZI|5
zHeI_vt~xBe27%1B|CJ|;!H<Q@g3sFMvjwx?=v4|wFt_>J-uV%H1=Z0?g1f|{x?SOj
z>56B!bX!f$2P@%cjq$PlcLW)}c8@zZ;^<22gs?YOfd7`%c}<o#No9#oIWEY7rEDD;
z2$OVMqHRJJHI{`}TgyTrK>3s)ld($*7N5k}b}tV*nm2BJq6>;;<s@&jpzY;%uNak$
z>HD*XQ!p1Qkjy@{z7oNreW9gizJlq!Eokf`q!;q_NOXr|Yz}J*vIhVwn5E4mZ`L9~
zx8-boTTar)E6c)}&D(&3xb?isW=^za+FUwM(CXl9?MJwVk$nj$=StM_3Ux+T{C#27
z6f95f>xe*toD2M>!VeipOrhYu)`i+xp@^?#7{QeGcmhhdWTbpSrkR-16H(f<R?KY<
zDHNscPutQN3FKX*dB_hDN_uG<p_6pUML>#JthM;)fzFsNzDQg>Eu}cwN|8UeT9m1+
zk|3_Vs_6c}w@1!{Tr_FlS-kg{9cSQW0E@&3;vzH<1f29WBVeH=`$xC3&yVllE>ZX2
z)@t8%^%}cO(nA|+<DrM`2NJ$O^(oOhrRy$VKvmlAQ$9ChJ3S9wc$5Ekm<EX99Zm9i
zhedpji}HTI%cU*+{dMil@h9aGbF^k{3oQunv!V~5qi@%@HpNZlbGT<&cev%3mmbv(
zO>nE_JDn_sb0S>&|LL57zWef{M-VxNpgux6Y>Z@a69Fm1`7}iGtQ(l4KpeJeczi{z
zx{J^xg+FMo;i9o?Hdl?wT(SuGdxLGl<KaH|_f{zKw+a*%Ulc+KBGisxI3V6AV_8;w
zwa7F35mhkCT!bs0xbTzV$xKvvNQToG#Wvp3AEjdy_+%nO>ULE*G><#SzQA~PO`RWw
z$Qej<OEYjcApJoX#6a#|%niOFvOmyx;B$MaiJoG-e(u(E94~Q`PO0fOt5d^c7r*?O
zJ3HiEMZ#pnp$2_d<8|5K^oSfi%}I2LKfndq+x$c6vbdShim$qL0|^%5PbDk*ag;B&
zp0BC*IOtfT^~`~ay5F~*)3DOGNqxh>;nwoftip7+L6bc{nY6g2IcV?1o>lOM5Qo0c
zPE)W2hF<rE`Z!rc%n8*iI^?@#Kw@tEh73>N8(GCf&b=fm8O$|cHYM<6j?qFHc(OS<
zi4FoQ@G+QFL}qU$564lkjb9;|e{@z{I7GINf_ks3V(`!w=Zym=KfVP@!;Exsp<PkS
z*T5m$kOA40iPiEY-I(_6srrrp>A~Vu-^y>lgn@q<A^eX61A!%UoUir%a$2}IPMt$>
zPo&TenX8*GuS0fi2O;FAs*kID<%&Abm|xf(tq9x|D$9oU%NhqEdfdiB)yp56QWvBR
zhzrAMB<cz53x;`zw}x+nnX3jNaoW(&8G!7k2*s3@hj=XR2aqz)%Y#-+_1GqW2(yT8
z8sQGiPqme#ifA5v$^OPk^XGy82$P{PM<BCjZ{UcgV=Vqj-3+hL|GK<v(TAffLv(?0
zbFoptbp1r)1l_u=DL^TzTRHfriASakbJu&{JzdPl7_wwTTu<|3w7oz~&xKZL#u^it
z!^}`sP(9U{Y>ukCfy>%}2n99DQzUHGIO%RPA^$PrdfDC&ebYsa;O20RpgO0Y+|tYP
zCS*BRVj7~-3NZtgTAK!(bY#ABV&0Y9$2gNZfQRb^;jk3~aeWefW6Mged70ze+tFok
zC7(PWzBFXJX^@Kl2V$Wk*JtHJ|1y57U=Bf88}o*NBNP`uaj$B4Mzq;Rw8=$aWSkbS
z6E@-jWaW?rhGc8eBDV{J<x0tn3@}P-#{jp|R7f=+!HV-BMrE(R%2a`$6c4JfWP06C
z2O+9LQ8(4&DKH+EyYIF^vA41KeEwORiNOB%jW+m3wXiaeWt0(g7V)`BCj7}Y&WTkD
zNj?eHjk)oS1dp={t(Ql-<LH^n96JZHxwui=k4ay#A_^d~42{9O=T|himMQH`{&?E=
zY}Lm&AIetg;+BlxkL20z!*}4-$(lxIxiAI5kCO??b>0@4Pxqm051W5yKm@pfWqf(k
zRf=f}Fc6&5gzol0Jpi98nhF#qBkpHw`=sIgl_-Ss*8pE6ik!b@W6|p$a%UPjdlETf
z0*7dXchoGuc$nq-ZBmYQWfOUXZ7&FBVMAJw#&h&-_S*|}c_}yAJHYuu$80C6y4u77
z)gm=NKJe&n=eUo=+=#;T7d^)%CML3|=~1lf8#l}MPbaIece;hu_5{B9QF^-0PBF`B
z>S5Nlr_yTB#dDr@#IuUq{9UWQxn(`>g2?%5C-w7uwU)<U?``(DhX30-%$fqd_Ye68
zIUUsYuwM107!w9I?%zFrH8A3RBSBW1fBUVH^M{hg?4&1}-@u=aiu^f2^ReG2X!Y&@
z<EkF{Tyxx?SJ9HuvC_RpIHfs?nvqV!V4PMkav=V`$nlh8K>THIUKhvdUbEysIX@Y}
zKE5>lf4F<guqfBIZFoSCGG;&lMH&eul@O#vkQPA^q(Mq)>24TWkdo8^R6-g=KoCSg
zBpjrX5R_0-y1(OEb+7k+?t4Ak`~Lj4ZT*;yHRD`!#d#jbe(cekpXYQZ>?X2W?)*en
z%d;qUvc_P(7J3s)&tGh~IQe#!W#<74R!Xe=x{{aTvMCH56hN)ki?{(g3LNeO6W6?c
zQ#HZvqSB2reYLU9d<K2vNMTmg1rmX=>)?Y+=C}fL>6<|(b30i@o4q1Wj3f}h1$zdr
z<?pjplKE#%^AH=;#~r~NwU6ZQQ;4oZo_0Y>1>Yk5VrVoqD%G84T0KUbSbCdY<LsCy
z;+Mfv>BRxY<i8o0mWn%Cc;;rjW{30d`ThP+FX*jGw?zK0&OhvL4}Rt^{IS2u`!T)<
zE0)Vn7rZ|L;o=F|zP&i+gxm3)Mpz7NTS~v#fWFYj-ai1|v&{Z)Mxb`}dzFF_*uKw2
zhV|%cN@kaRs^o7in^`>qQ9Bz_QV;IU`U5D#&Grrf%eWG9yat=l1f4J{5|)b*<dgh~
zLFFd#JwL;ZB&d{5KVs}>YD5#DJ99oHJwDQ43hk!~`kZnO5ssUM3G>?Sw-k5aWp;C$
zz`S`e=E3HL`VSC0b_;7Qr`v!IGXwmn=p4F{cb;UHLwyE`$u6BtV3F!JpY@)d22UTf
z8OsF?kH{{iJ%4dgK-qrE$;ww^>`S@44`Wcr#b%xmeSML5D`qZXD(U-`c**9EU=~BM
zX2h+j6v^Cnz2`m5{Fp11-Ms<hB*XW#QgIcOcVILvtz`27(K|a-zv>?O7}{(RoYfTO
zO6kk$8<}8~&fBKkb7Q^CHwUMhLNhPV)&^i9+Z87ka<fX+uj3!h*?qS^XU6?z09u-}
z@htORszs)jT0KjqweGX69%MNu`93ozkkaDyUauVdAl~o#f;e3+l{uc346l`>Z4%a$
zBzT<cHE1iuw~25plQEVdq?S3}$;-|#zYOJ6`qo@awP?&@0P%-YA3z|eOG^{!_>H6a
zJ<Oocjv_0ly~@6=!qHMn(IUL7L)n^9;Z>jKJe$0+>E}O_`pJ@acG5vI!b9FDY9$G<
zd_co?shPg{>qA4VtKecEB)SRR0msuvFvZJ$H<bbv*IK3V`nGsoX3nC(wmz$)Z>4yY
zwa9SMY4>Lf|6I#oHxfr}9u)s8Fx8TPkk@Vimd<-H7R+$JC9%?~CyaQBDJ@V&zjo3Z
zu(O(sQ|2u}0+yyU!fc8g4o?L<cFz`dobB{dyaJ8*FcJb+0tF&oT*mOuZHAl{pt9Ws
zLGbe}joc}zFRAjy7im!K#H5b7O?vhhlK9$*3odCThB4Z^zeSW3ZgmGUc69|f_w36n
zj1N2BlY2r~E3|Qu4+=_}PbZXeicOV1dF^J*>z5osBR5@onPIODe02p^W~^LbctLW=
zu(}VSkw#2t&<OulxdCwBrFVos25ZT%-1WjUW%C@dyrFHk-h*O@Mo3onno?#Y(RJHC
zzCAUOfcTiW>@K&YQe2vy4_0ft47A5&Wla~DU#BWlIz)@>_Z&I)ZC?2#&4Qwc(RES%
zc%|d-r(tMn3N4A>OGtQ=YtetgZ%AD>GXOAq$Jr;iyS(R{OTf}#40v0OqBNV<Ajw9k
zwUzdfo*&+s9YU-s+OsqU0ra)uy+&QAC|WIO0UjtO#S>yQ?d1KphQ>u1UULs<iWz#e
zpFQLFm}?t7eJ$$r$uH}`nQtsRe4OFT9>C&B+U=}EkL492*1sT;n=ovP-fyTV7ugFD
zLk0`6El>y9F>LW`Ki_g+<y)DF!q<O&n<-Ct(Xe??JSAi6*V}clBS`+3(14YUcddq<
zJW`lVo#970RKCN2*K4nG8U|c$Gjv*dMNdN?wf?+d)pMI={;P89wLv@ZvPr(^mHV02
zwTCY?WnRmD)u!#UwM1*y#THA0m6L1su>zxv?aHf)F~3glrC9LC)TCLpW1x~v+WNKl
zCC&;6)i-F9#_#ezt8E{y^YU0#o?;)p2Q%mpya*I0yR^>--8qZe+_Ht~1WaUW>-AoU
z&?F$^=!lYzotO=-9p26qzE~T?@UlH(zQc7bv1riM;o_RT!>G89VeQo|$;QiGU|ke-
z4pOazDnSA^3?~I+lTcUxm?lPaigv~F)s*_xti38@@JYBB^Nel<`_V^0?N@rh;-OX5
zX2vlQH=6jm`YNdypM?IMv|oacQeYU;0xYzx*RKEvnJaUy|Hu7T7MUylt-n!zn-nY*
z({$2ABS&d74(m|Y{kr=u){G&iRa_~?y{R6(vH5g_bvx+Ac&y8jt$VN^@x2jR#|%F7
zL-&cxsO>6rk^TB!B`;6zk7h23lTJ1dFK`BpWXO{>phksq4%E0VqBX>SrMvgkpozr0
zHAtyvy?Z6Cps6Z;_*7iuI`et9SCF{)8NpsIg(w6(X|A|81kUbPHx)LvRPL~dLm_4q
z<|Z?;Da$FVCmlV0yb}btx<^Wmh&E=SsCk!PgoGqd#rEOagQ@Cm-ljfa$?1)|t#j|}
z)sT<uL29zCIuQEq9Pk;s3BA?;;nf#<!QR5i<bA~qN6|84g*|%j7v~N9x&nhujt8;y
z&^uPf5bRAJLn2DIB=&ET-*DD4x4%G*KZ*7N`KyR0fYGH%$?|RIDX)r#eG6rp#PD~p
z3dq&4wyH}KsmBbR^rj;anRk?XEd2VNcGcGqx}gqYQOs<<j!Yavy_YeOYh1QW#<$Fl
zty&Fz4~ld6(+l9SgTI$%%5V4Ve7Q<T4{JwOS+ixv%1P&^b}W{kV`AfDWiYL?6p>cV
z-wYw9^B4w|I!_g!=}xV9dh6)ZWgt{U@3#z09CzqR3Kz;QgRY6cNq6Dn#fD=aPJc*r
zC0=<f&vIlr&0=L?<!tyQ{YOWu^u-@McYo`+zpArcMX7ughfzBb)jH;osG5w7#VFfy
z4DBC`SOq`p=irHfq(~aIxKfQ;wI{7AoBl*Re@r~5@I{;?d!dWhI_kPU<m2NLpW+uS
z+&lK75;J#FX0<xRwKo2VnJBG>1d%+ito`wOknN{9qK7e9LagZxE*KFwF}G`qY0cA4
z1o7c%)QMx`u_m%amkr5Fai1Pdy{jN~Y&DXdWo@L!M?VZ~4a5e>#^$lt<ZLY|zS(TZ
ztY)TVMrqLOKP18A+*buXiTDb{QSwouaw!_`3WuM<Qv+3WjV118PS6YK4D7XH>D{XZ
zR~A@Gjn7^=mwH@%2s<I)BiR<mo}nh%N>Q0W8T-x6oF_~n=wAH0RgdixyI;Eva8b(d
zZwwvH<xExj7-6kws7w{WfvFu*UL*JWChu#ljAa`#$svv94>A)@=nE8hXDH0Pd+WX-
zn<$GP-e;9!n;MD|u0zg0U&s<E<D4p!IInsVvaA_JQYKJI_9QqcI?4V}=Mfq%YJD|W
zm(HV`wWhE#Qy?&eS<%$tFwL6tnpD<l5uf~6CLdq9T3su-GD8&~?9*^b=CJ94NiFMy
zK;dH!Oj}+@l6!CP)@giHDgP*~d?X1AA&a{CbhyrRjmO7CKxOeB=v*57dDUXI>A<<y
zb7xZ0?}izrk&oXY;O~g(>fV1Wku=hoZ<ymQyd$Z$a&M{(xx}4&9r5*eKwmHZFX-zn
zWYxL<F@N;;s*~lDDs<mWvHjT7L4FFi@-@D`Oub#Ndn@8Q!<`4B379Zm`D{)WeJ+H4
zz)g_2%zmFCb^sY%MSCYdpz$9gy+ws0R`}mYbx#m)#$k)>^M7$rbg-yj<R3dJ{lgTT
zWh=#?mo&-krOvw+!+TDBYX4VKeba*;Kc@TRz^WLSx=<l;I{6$Qp(vUu?hBAV=MW2j
zOB}*rkbTg>Y-PqwIFImGcv<xS6MOEDU;5ks(u4tYLHIR}e;r6HO4(@?78E%-L)Zvx
zka51)v-Cn`zwbf70~!iIDJaHX1?$bhLS4%Q6(TtBkHrSgy_&(CsoYWRanB*X;|a#!
z5#C!0DGQas_p@=8ECkylV7X__ee$x`rak!o%vvK%)`Y(^S<l~AHTxGG?=N)Z?=Sov
zl%xsAFTH<>us{Az?7&J-QzA}v*1-EYz8oG`{AC?fa`%x?VduU(Eq(f~`l<GfWS3t(
z^ze8EVl)LQu(T4HhtYBgY;;{7Yd=n|WxzY<f$;kZS_LQ1V|h>h&u{p*FZ3f7R0h<h
zjOc%EYUOBxrhJYJw?^=K9G1y_p7ucLYPAgEElPg}IiuAX{5muWNfNfecrAnIn*{p)
zzyHMl_}c#g+%6v<&tU(TL;62X$=?px$sIml<3*M0_xDTxfARVT!gdv<qW`B_`9B`{
z?|<~?RfEqb*|D6w@xMJ8e<x7>?R6X6rrZ`d*8huR{`;N(={tOW?w4vd;XkeRe>-FU
zekP?wK@Rg>#Yy)6@%XF7;q&@+VYj6J7%<L=5<w|^eh0?A)_J+lO5sUZPYZXQOo;{|
z%>Dj)-ve7cWz9eBFaLTv|Jw_(P$0y1el2|d*YCT<WDj8A0PHDth^OaK$U$j5N$)rV
zLbFPshnI7%Kxo|n$QL7`*S@6-+-)8JzM{5tF@(Y(gi(r}Ug40Vbmm|QnU9zi_448!
zFfSCw!Lt%kFxA@If~u?>T%n#77e)1*ee1OYXn}Sdn_Pm?0#u{R5z#`WpJ%)@KbHNL
z^9_9mMN%QC!G%GANwZ#o7b0^q&74UQx}E#qe$wxMheSEhA>~$RYW;cQCWxW|O1lg(
zpY!mQ<$=e<EYlL!0XnQebe$O+!0jA@<;wzz6aK2LHYVLE4jDjmK-DS%LJ@aUfsQFR
zTf?80pBd4`H-D(<GI#Tof09Z6wY&NhIXOtQpmUhUZ^2*Vi1Q_qE?ihOL9S~OcD9@Q
zGC0n8I4NDJ*}Mu2Va;KH#pZy<5*`P<+k>c?B86rI{{Ng~=Tw*Xy@7NRh@<65@cTNr
zvkkyp%FMA{K%VE-kCf|dpJKgn<vRLiqdYuG>z8Bvwq>VF`XBwp82<N@^W-^VSb0-N
zXYCIaq(3#2Jzxx>OOCC~p#z|45$RNl{Im?*o|74`fLuWr`!gCQ_KhVObys@wLBzlz
z`yqn2x)r+91MPaBy`lFPK$kQO^J^nR-#zD={vQFmWS?%UNp59>J>8ux3xm7St@?2C
zUeg0*u@c<h``wa)Ej@seIRG9pw(mlZi!{$d<7Bw9=WaM)De;R}-nDfRpzCs2=5v`)
z5C<P945_$R*V9bn6jKUKvQ{*Cj_<=_edO^RjH7-lu^*m_)lH}t_K%*U$`M*5U$MRS
zxP0t2Q4fb$QhmnnijcoQ&VRm^@Q1rT<5OE+)*lOFIjR5i439P>U0W9%AtJv^UOul1
zev`sZGdCd4@boy+nL`rxGjjs80|fzkOZnaxssX^}uHI_n9TP89UWSCYL12K~0}rM2
zmvHktOYQ%`ZhZEc_^tU*b=UWvpN?Xkkb$}_p;36@=+?P6MTu(Pdf;ry6B_XpSAP<s
zALQu1?-J}T!V@7`bC_T)tPp?fT=0KgyqzqN<=OFPp6y@D+fi^7)TSS<ddQI{rEl@$
zY%wH;48la!D4h$n4_v%T0B7>Q|8_uj!)w_;&!X0|`xd@bM@P77YbCp0PAbhXP?Gk>
z#bek1>%AmR^b15c{C2puKSR!&nf(#v2+}$$oB6;F39=2w)CX-`Hc0Ba4NDQsXrFXY
zVtRi7{!zkLx7mo?@CY&92HM*)!~oL3W^|6@HHp9!vQWTN!^v_<Jp?q%Lw|T(T4+2|
z0)nH#g~Y)2XM^Uc9h>=Q5y=R$DGtMvMyEdO)(1jb>)#ACf%jJmf|Yjmbf<2e(sOzj
z3yGu?>GuH-n0QUWQ~t0aV4W6PV8r?+`h`XLR{#{|0k|#!Zuc0qo_)i3n@|Gfyx-g!
z+-P*(CLlVmM1nyOzHKTP-?A(!ZRE7LF(~W)5>K2Do`g>9%)9`ot#v$q5$(Q~N`=C}
z44IHgS}A@L0Cm93O#r~}tUeum<0<c(ZAhtF$VxdG_FepQd5?xNaS=XCUViME@1q$;
z#9sq(OU1L#t|dqUk<Jr|{J`;+w-8aOpeL>VWI>LsKjHh#H@*xkU%9ailW;DREwgRm
zyGalD*FQHo=PtQI!rc4KUEjHMl3IPawMvqQ$4-Z@QY)dDna-B`5j|8O$#Ero%E}`s
z;3O2p&ZM+>w)=|Cj4IY?RwJ7t0LwX^DLLlRH+^P(&^6OhZE$*OYASt<FHPT~OT5gy
zqi=7t0)alAs(;YbBttk+esn^`RXR%wuA7PkUf}T(gyhZY=YebT?3?2i)?#!759>?}
zd&0aM32WFYmQLJM!RIW@`QM(;EATNL?p|=I@kyE%`MFc(@_tX5_jUG)Md~j=#ffoM
z&9nyx1Uq(kH?8aD!jNoO`LpPPs-?$8Is&xs9&Cc!Izcvw0jFfwa{A`_f*0dl^7N+u
zB}d<1Z|pOmUQDqyWEEjLc71m6&dxonO2l+xp{Bb8*iK#3x<g$Dl@mq3oFf;7xW_WC
zZIrEKan}#eov|r@xmJkG#SZ<XHUo;?imQTkN@elkUv(j4X7t~@O@aQ!Q?LLgD8~_O
zc$_{ez?V_nt+<Q0Gsih<9#K#i6gQIZ*J<3kQ}&ay|MPJr561T*%XM;@+Qaum91~rZ
z>Y60?!+%;On73q63F8-bU-XLnle=~D5)>krzQ#QD_<arIkEN8}6aiyUdytO>ArIes
zK5WG1K%4v^;<(6EAPULl6l4y78|%}+_*54|!K@)IdsPDXa;48jWKD17FMR<hjO7a~
z{k$yF<4rAkO**6mAxBP4Fvej-cBe$@xe7vhif6C#(7?c(!x)NzW3XuR)k^v<A#^z)
zTiUq66RGDHuo>jSeB|q)HJE$Txs1A7&<Ia0zN{F#N%$7w9wON<qqn;mKz%Z@&wSe9
zIAKZV-kb%XXa=N+&$pvLAFhNo`ck8u-`4uYW7#)CI2ead-SxpdGNO{&#>X@Q1<3F$
z^~XMc(i2Z<RkoOpDa&#WU}UJfSy&(DClS(?d;#g&k4d$WfkX3EaA8MyOS{ew0SPze
z9Ridij>!kwYIMbCC1*Y#m;7><G5=+y4u0N--gDfa#H;QtBb^wSJv1^NY|6i~8(4{1
z=a)y5J=}K$21`DdpES2U7@mhh#?i;c90%+rU}Ha^Vcvz&!MeZ|x0|<SR)CB*HHSXO
z&-Ylj+<&WMSLTc>Fcn_Ez1Z7OVhAmm550+~EUBLEEkN<LUWMTt=NX0pemO+QG9~Ck
zzRYg&#_NikzwZ+@xwB*_MdrwXO)%D26%DjeFILkANlmnnO#yw)xqL=BZ>#UxH9oI2
z$6~aPj0H-y)Mo%!Q!_w7b^x*XiD>*>Yf*5T>m@1b<C0cX`O7{bMbsupM_wD&0_Lcv
z(B%~<;6o@Naj)o3==}~yS;cloZ9K=6$bojuV=4$8zdrz?ly^Wv*URVPVFdo8?*c~V
zlv<N-J)|n-{}6+ruebgLlOxvI9c(FbtJ&wve-w4^?XHIiaSvyhSAJZrdD)3KAJTQ=
zv6_SXYT74p`P-YIibUr~SME8Y%hLAI-p5W0w)QtbQeaL*@*bG69t8k3bcbS>;O8Yr
zB6TT8jt-j65_VgF3ik_}PbcxbEXdkq+4iqW*bv>P4ca2s8hQf=R0ycnFQ>u6v%epq
zcDK`b7+B0puTyQV=1;0WQ~i|TGA&=JHuMi&;fyCJH|IR><}v)<5vmXoZE6FLP&Wh_
zdRT*IgaH$Ub@xg;8-T`TBF3}a;I6=nZe#s=b((R1DbyY`Gc|7JFxNQ(bI1)&;0&{}
zkTWV{=FWh95W_5&Aqg%633_BQVJ$rqtB($2)*e5^XCu8S>v<`Lzu*Ovm-M$p@`)mt
z!d*NA{s^--M#cbj9hPFJdW;p!!iH^rioAifU@_0S-gG4@F@q9i0PGuHlsJ@K>2yc>
z?^w_u7)5VJ3b=#!3?XBGn99{v`X&RYxZIzN4F>R>0ZS3yx+`;XO>!&;n6)FAK6w77
zF_fnqwU;pBx|=Uzo`f#5=`h+bYaHValWYmKsA3wIDOiDx`OS^ZW;a%CSy_2zLNw!c
zNs<m>k@aPCmo>Q1Fc@PMd?U{j@@;n}U&xS6l09<GvyB@|hbgCgyo`J{LVrdCQ#F1i
zo}~?rWmdeT2nxUk;+R2FFTF#Vh|uAEZhlc+wLDIxl8lweK~_`dM&Jzs0@=E^N2w#f
z!)SDd8_htuOj%=e>@#(tE)22Hy`m~Prp|40%*Qwso%O;<3o9<Bj$Zppffh3EB4rO7
zPvR>chU;c^?S~4XbkMDV3-fYE<i5$^m>c{#BV?b;T58@JD#<V1R_`e^kE!_SI=sBM
z=@0u%lg%Qym8^*R*(niwPF<2b^p#tZW6R%3_2rm3v7gyizRvu;Z*G(*gwbvrxOu+G
zLzX0mzw3@Xh80e_0Rb}?;kb1JFJ=TOc^6zCPXleVi}VFnC#2|Y=o#=St!p-0v~#61
z>_{`=nWhg|H>dF7my<i2N56OodpJTVfB#%>uWny)PlmbgM<6t{<IRWD>NA8Uzc_ZS
z35B>G86}9zq!)8~v7^7w>Bs3;;JZt3XkE->)skk~;^(`KOo!%E3x~QrQFUojdS7GI
z+@!wi2Z}aSM<0yGFkyL<mGXabI+7R$)lcljRV2Ai!CZ?C25$8Gp?tDuWc9E!F)QFQ
zlW?$1<Vz}T`sLQpshCphfXedAQX0Oi=YNxk{aG^qRbe78K82DE96Zeh!0nIS$RERh
z<%*!cwBQxhas?ZG9Xm$)tgsLqnOSw(q@NsjIQ#u@_S;x?6D)i8i|S6vFW0C7<#smd
z=CkgVbDq<j9uzQ^7JPd33#_ELu)XTESC;qil1%bD$?y)Vq?L+e11(c9{3huW8~I*E
zMx6!!$-V#api@=@&q8}`=hN_ay*GzCDxVtYkEEBy%zXP8nb;_Zc*pv@_6~}ne%ukG
z+f-*MLn&peZiyqif+IVZpZ8t-?!AdTB>Yk~;>%6vjAbEXd0?0N(yC^kn!oGS>yD&X
zq**prTgsFuq?QIx>twfDnv_`LF54w}CwCcWI1UFH`n(}ax^M7E$~d+e7uNSHG)s;=
zoSk!N+fwa$I9c27DRvcmhD7sMAG3@sV@XHqR86&<84Re21yWEKb%H)(>CGu7ttMVr
zs=rBde_5A&>C^OUGHR9$(}*zdEV3h6a>SABM_M!ThJ+oi3NqXEmSl_y_{XjY;?0#~
z7({H-#$x7EZg6*RSbwdRYj^TZmdF^h?Jpdfm#>p}Irr+P>&K<Nd*a1kb>6ewsJ7`7
z%aj@|$ef9<Z+n$L>hp_=x}@Fv%4pFA3&`ABu!J8VN64M_eBh#FuWS$==jGy_7Un$(
zAmNv!Wb7wzqofF>OGbEOZ<}1Lgs}LmFqD-+cXa)a4lC`BWBsD|=rNwV(S7%H-aSu*
z;D(JF_R8q>(9;6-Au9&0movyvtczUi<OO#`Tsl3Y_ZpN0pWZ8<X~s9XCj4}_=-64m
z(1-639_GFdK4%WVwo7o8jWDI<$#MB65wrfhBH^_XovMPtJsLW*DEZ66>dm_^e#HNi
zl6ISoKqLLdUto-i5fS>tofFhFwc2?-GU|SEQeo703X3fgp9$zRJ+j1XlXp2jj4ARG
ztPl*r<H~O7hoVS~=~-l&uodKtx3^GMv!Yh|-QV3VPh1{qm%kU`p=E4>C(`%Af4|~<
zto(C`STA*k5}~xDkHtr*YGxR-R#@mF#;0*5a~0GjEiE6qSIv&Q4pz1b@8gGYeVXZ$
z@7~C%IXyLvUE<ZZ#xLn%8|8mpCr5o=*?iw;$vAOnj;w*B>tJNjC+ST=&pg{)38@?&
zN7-92%tG7@&ikA?;Z1o~+~+bWj&OSz(g>KZ%LD}<I(*~>SFLdd%c@*Y7^jFK_jWqP
z2@F~9xq0ld`@Ms6%Pzr=iuNwh?K@%4Kd;o@Znea-e71d6|8hXHW^blaeMIj@$e2z4
z#^)wM2KTXe8y^m0zw}kDK}n}q>AOUP8!XEP)r<zz5<UHmPLYNY-{^SmKcOh&IcmRt
zt5-N+0YoVu`n`fQ(mBmYQNznxZ`_Ys1-oAMc@-%-q?Ed{!Gtp1E18Mkx|!hJXR1uW
zfce<(nF-5vQ_!x1qJfHbE`OPNvsl{>hn15-S}UWv(aNs+m2;a~qdeRbR$|}kf4lNX
zq5R|FKVErmf*e%kL6Vl-ABY>fAYry<Ij4Ju5>F1ze1DN4lTd0{zj*qod#)UhP2(P_
zsorO_en+Xy{WbSfVeK!a`e(Nf6tm0;bBypac^bds!V^xUWgjY6u&A)zm}K`^S}mgX
zE?jV)WxJ;zdX(<=TCG#zOn(!`U(hPsrXr@+KR$fb6GYW(CrYxzy`e}5lvCoaezT-q
zX2V&K-4K|q#Wr##*Ssu}mTt#wepXlUT~fHW^~Q<X?0#kN*>yi{+C6eICr-70*^03&
zFpN#Y`M#3;V~v8ia5va}KN=WRM9n|MC71hVPqEq<xW1R%_1sq886UerZuFzfQY1qr
zWwkJpV?*ZzquNmgox9u0)f1VZI2w_t%UR*7;K&tK9KKWZb_J+e_89|bf(0A1>DYL9
zW&?koj0pP1C9U{b*<;MQ!FHB)CB3IIdw`w(LceiVyTg>oO5N>2fX4#m>Q+QGex_V!
zY%zS8k@{NU^q9GFshS2;Txr#dhx;i#32RgCH7>ojQsF}*?IkXIwx{Yc*Pku(6Fhm&
z!S~}`k<VJbUs0nA9aqO@*y`@}&HQUaDTRtY0r6GYEibi}NZ23#iXUOm+fY_i4PWFq
z>*>1Eh@0Ei8Rn{&CwFK*mq~bOm}96Txlf7U%kRri;+3r7m)j$yQ<hot1+M9Pv7z<8
z<F4AiRwqY^RlMluj$hj~#5=PO9A9stUC4rj%N^d_;)n&VyYwRf=1WxYbA2c?gL>0(
zOfG!l*4)AJO3!e%7l)ZhPW!TDFT16D%}Suyt$zp9{)RDcOA?TwcJ^uWLx0~6z2i}g
z`aBnui%I<&a3q=@ALKohzn?}1xHKoVJu2Hog%NDa2#?h(_QMv}lmY~>wFV3;$L|`l
zgxFq0&f;hPq`f<G^BD6m01%4-f=v=wCY}o2;_?>_xbs``<`Km-+@;F;#y9HrI?9=S
z?aFrGp~E=oIF+VMys%E)Ot!;N2Fk@2c=a$|G*=k_)<PrLi=5t^F)U)*ERT<;r$Xt(
zH~Kxq2#`&-y0H|L%|!FwXsE+5xhRod9eY$9ItHTOB8F%`Z6jc-6jI*5P?VgC6|A8$
zMT*ddZkEH!2WPk^jFK!IGTcMFS%YN1D5cv=HyPnc0&{3za(%Orclwkgz{6mvbt#F+
zaE-`<nx0=<tR-co_DIeBSByLp`Cs}9$9Udtd``aBob`zC?e*<^^0~vpahX*RCff*U
zY1$jdnwp@jw|{C+>7;AgRSW|c66KN6m&vCOLO;=CdI69&*tqjDjr%I9)lqlt_XWoy
z-{)1q?+eZn?^};|o>yFhrNm*(rYdv$M$#J&G8BNCcMZGBrtjM%l|)ybM{zrf>r{|!
zLd|nupwHpI%Xs|j<fIAT)(K_&i24nXdn<d*4|^7IaV|dbgx8DNBr}KG`6&vc1|9IU
z{eW?(rnh~?i9F+I(J9hIoB&!VzsBU=`0h5)-b)^E#a1DmwoC!!_S}w6D#1r?M#r>a
znSpUKH}MK4USH(=uvzFmd_O`Yc(1v}68L?1M1}6xhH#Bx2nt=ofzQUW_gh{mKg`hv
zmFQ0sJXjpCaf5N1b3W5H)kMa1;5mIW&*eQNb6%N_{_xS$-zw#UYLI-U;E2QrL#*a&
zL8v7FESJ}N4mIwwc$~7~5MvB};h)$zcuEyWe`~hK6R-8Wzq&TSIv_h9_K8NFd1V*y
zH}dwYET{{|Z|ga{Z=5b?72_-@4i20rpFWLA>!m^^nrY3iXAwJTzAFE%{1AK*yF8PP
zDt@$|NA4zaSrMh+iC*qIkMz(3K)LgJc^$WlxI}87)`+2Q6B|}&kSbeG)Ni9ibr;^=
z)wDBsr(-;lH^yTXUD-XgD?y41%WJQZlVHB=i@oBc<7NMH<{{GUYReCwL?wzcyc{B$
z>|WSyjd}S`=;b!2zhvqcZTjEC4$=OSDsQ_9t|`0g4qR2fLk&RPU=^4WT%%%>i}2EP
zPM4z$2i2=xIaUs@v9Fo_aWT1=-)H*rSXknZS3lyzj%4M;L;;UWrFZ#)3wNAKc`M(@
z?X}tGna%rqJMvTqauEUAD;`v$*OcjWSG@3H*M@I6*_E|$%NUpUw_9@g<=a-Qy83+4
zcz2s#q~UHkX^dtIw`u|7-C7;4b$=<H>`Y7bydR^s>L;`0d{$i^qly<QX1@W+X{uj$
zwc6>$R8g4FNFjF~$SotnbG+8KRZ!Ya)$w=P+g2>Jhej;9uvUE^#Q9@!7e5#{!JR3p
z7V#rVLa)jjsLrwxdZ$!n)tyI)3h+4FW<XDCf4B_^#H!W;T<`sb$J<iu%Xe33^}`O4
z9GiPk%W*u}4ezc)KAy;6b+2j9CjzCtvLAR{vT914?;}!ob~KB);q>cuLl&RsMn#fX
zg~%;g7h4DjPjZOO7M->&)=%CIl3UOBObtuQ%5jV-IDe<!oU1b{tT>D8`H#^tDh^dF
zEig9MQeqliREVf~dL6zB&2Q(A)`pyPsY(s}eU%sKu$n0k)9^l~OEgs7pL%0-uWS#m
z*zr?^ZLwdQ=xs{>H?{bZ=nIQ(^6fOrZ|GsT-Jkca_VNmE^N##EvuOEz;VZgB>5REj
z+y6kWpBW~S8C%Hk;P|7(@qY9c=0glT0lFidFv2$57x=94&E++(w@Sf=G0oEeIma+y
z4?4H|4t&pgk;4<2$Z$#LRsrvt#c}P`mzMZX_sOZt`b$S%Y<9AgRkh}2kj7RrThII?
zM69;2MZ`_Z#4QB5khY*H%_tfrA(GZaID+%Bgl2^|$W31}#m9V0N25%V5*J=6u|2Xu
z&DVlg-AJLQ3<^7+>C~+#z}J|qYG3Nq+PMdvV~6lXom(xJfM3@H`fD*AqMSU(!E}U(
z&I_|D2LMaE7}k;Sj-hFq`U>(D`&c>Zu$7c2uWmzdpLQBB@$aWE#7=Iatf!z%u$*;|
zBJKP^hWc2V@3Mo}n7$~I@YQvAZBLM-#!Q@N#Oj&zn6*h6*5dM6)LeG`_A{ChyR!sL
zj#Z=LZ-C-I0T!gEk-U#QCCr%;Ne5q76XQa}t%1dRQFag<9lwm7=#8@c3DALe8vGP1
za@6o|Yx#MT?vWCgUE5QR<u&U<4;ToTR?ZO6FZz+*DGPRO0yhmIUI=<|0BkO%R=4Ne
zU&}yuW@&R4eX6CaPtR$Ze}YtJbP)(hcQSYM_dY>|jcAiU;t7hN5g5KVdaWe8DGAUe
z{y8L38)Lq{&<}#}N0H{(9i0+4xTT!KBtt&=adKjlk$ffX6BXY<SNn@J5V5X=SP7DR
z92v{)C;DFHvIUobg0vYr;dK?Ak1^)iiJ5T$Oy8ncpmRFwB&9%9>pgJV(=0D9#yA`a
zwqQM0z8dZTULT!C;A671jn#<fn=fZJ)GFKYl9u0WkauZj9&R~7@K7Np*RDs6NV>^q
z5DPl0OP|zmo6BF%92;zOLt1=>juOs^9`IwBnEc=-0l|Bkj60`%t&cLl1>64d`LpC&
z2G-blvz9o-wq~@5V@PvdFIE*JbOb*FgT|9mVBW2?a!TZq_US7$8I4vzKQJ)-j-wQb
zD-Lb-fPs?z7bK`WXUhpgq&`DSha-XZ&DsGL&)Ewkr}&gx&VXIe>6kdhptd<$%J=A<
za={h{@cLd=@ZFkvF06feQ6avyb~9kPb}Ge6kG}>8%-`*b<`~_-#de;+X!z7}^Hgq-
z5!{s^xS3N*ErJd)DkEhH(|BB=YMoIM%!kSRMSPh_jQ#F)*`@-0dmg=#^U5@{>mP+z
zn04FWKHZMBQM)`mzddrnOKQSU*=F96>HK!dYL9JyqRZ#^zNDyAF<$rj)ovJ`+CMi|
zcx~fR2Xpv4-sP~>K@CAIxjEAe)>1>rbl1}heJEsDWnn<IR0xh_J~oLQ{THN`rm-{2
z%Zca24du9YHU;`_Dlh8g#<;J-G@=;^f!t&ArWbu+%B>@h+{tD~^PTECqJF=VJEnH#
zYb}n^^1to(W;vi#OmL>4YbfM->8uA;pO<R0$TWpXZ~4^Ss{?BL5`HUN6+FG47sT2w
zQ(AFPH0@SjA(B@9pj*LS%8S3IY-mss5--FsKOlQ*h1H?=AGD&<V`urv^ve#iZvNR6
z$r=%|7|9eY0W+4Eih)RZm}ZXECK<*h0b#@eVPgSgM+3s%1ds&=$SJKjZY?~1$d>%%
zTr&E+HF%&l$dWeTiioL*iIoM;a1ZHfgELH^chynqE8M=G#mD0sSe_756%t1no#2jB
z=Wj6L$A|VCGkOGib01sB2p^#r6Vw)L;0M53>&MfV)cQQhoVgDoqhRnYfOzd1zpUuX
z6SKvT1d^5JMc*+^<cA?gG0gIaB7&&7J&gD<O8l|07oSMYrJ~Mj=k*%RMh2372_P%0
z&A2dhqxT8BE8-Gl6zkl0$|4P;a^~b5t0xp=^tIFxPBh#es{F<wWr)seZxbVb1~Yf4
zqU=?xgE4r~uT4_V^V@e41sg*S(vmMV(0T8V8L7{teLy0^XlbZYHwSj24t8R^3pNBm
z@08yN*PEeF&ry|8P5BC)iqR=<Z>!??kG77Hv1Emft}urzhO|dwsw@v4cA}%~+wHgM
zREj5H<JT>AN=k4{ynX@6d^eC6Lm+x$Dkl%`;&FF&LQ-mfzGTwogU91<hoj$=bnvY_
zZGZ6Ys0dL6_T<d*mA6Ee4Mf(q!vjaezgB}e6A{(6==0qbRko8e7u)Y;hS|I+vNs1<
zAf<~+S)E?x#g8?!pW}~Lqh5E=reYfwL+Kp*f<z5Mb0&@fN#U13$7p6{3CR@6%4LBH
z1p`UGQ(9D>Vun-hr6o>wkL%eD9?&&*KF%LF<~SA<*+4|S;J9?rUNP7)@w71FUf#g+
zI0SL;Ylm5&t;pu-Hn@ry6>-{y+Lh~z6qEiEDcDPzRBh$?x3|^rDUi|4Z|kRgDqIwK
z8o6WQMdO1y=t?Ul4Qn8>A)qtMF!J2o+dX5`ul6bQ!tTkna*bk+xp+}GSl~&hL`im^
zg(U^bNv=8{tA8&RG-le)?QF|0ppn$FEcO0QC{OyU9kB(G2Crn8IC^K4F<g{VaXW(;
z<lDny7z|VA*UA#}3`YCOSp43RpT~UOv%Eh{HC@f9XH(JJ2=hAn+u*jnc-Mvg`=Hro
z{%CPX#YcgWvBEJM*XU<1<tsJq7d~!&G55YulG)`jdhw2w%8lu|RM#63*8Sf{BBy`(
z-1X%gS5;UfSZ^U&U@V<K=kZE;T327bWCWItkJC9}_XEjh_eL6tHVlLolLsy>(xsJ2
ziGMwQ_l+I<z|mS=<H_0doXBO%es-u#!g787VfMbwi*%a{19=vIbeqX278DVh+L%*Y
zkov$<(G^2)|Al3%%B>5BIxH(B7i@)@xg%I+ggMYFZV1(HA}9Th*18iEZxpK$Zo6?8
zOI5{)?+5T)>mCWC-qw@0vQT6Mcyq9edguem?FFlBCTUtAhWF->UPoswJDgHK!NL>4
zPQ^`WP+achzprb0(T-K*K}JlYSJSh2{NbV(?+=jZDqfVzE8TL@#p63zt4nHEj=Y9>
z;Z~(wIyv=T+bbUO^s+Vjo#O#ilGVCBE%3B@iKFub`bd?~*O=L_E<HYZ_u0<t(4A9i
zLD^nR^6i1e!6n&8NzY+wZ=gLhfLFlFdzZ1~j7jsg{kouP9v-7)tHRe+y%ZbEPe)jW
zgQ0{J(k4>2JJj{`4QUwK|3cyHDTp_XqWGdgZ??-$z=XRx6#61;ok=$EtjuR%A5_zZ
zyw=EPk})bkFEW{DF_LDXZ!o7x(<<*6r}2q*;UucvKC!Tbz|gM0XmV>ei^2K~eiXtv
zyXB;le@xyP*o_hyotbZ0=kZ8g2`rKma7vXXy;3ixiRQkjg_-;MDmiO_o!B=Rd&Kx$
zz|<|Y(gN6gNIw58Fn3se916DiP@@R9Sf6JB1DIi@9F>~;IgGjVW_y*BuJ;bDe=b~X
znQq*JyhoP<zZ<j%;g&S1%|*IBVv7%aqAP!LP`0@2Y34m2G?aZqeXqjAaZ|kSRMX6+
zQ6Ezu^%SQhy?_AA7t=*ZxLKB7?WmoWoI8^`l~h6H>+hH3TkzS8CgX!=u$;@v`t!<C
zbEfTg(=0@j)w^O!x^2I-COwaykFcuo5gC-dyl64M4&zDpX*Up!d}(&qGPPZiY#73l
zW=PR0tjBpz#mY0BLXtD)o;KRd{RF)Ow|Gm#?LPHg$_bBM%HZ{763W@*>~2ZM5*9qB
z5?}WMoJ}C~BClYXdihhAqwlFne4*;6XqRac=YQe0e*w6Ey;#);%FFsliDuXDfXp^O
zLC|qo_A6(kiTH5YOzkp8L0A%kBkG!`G2~3#ENxxL3<@W(z(I%yS1a;hMLBIAxm;fr
zx#B>-hcWL+uV7abSe#Few5*la#LBE@)nXs+U|B-6qpP`vp8B@RooOn=T*gc7tE0s+
z>O8t_mRBq5%qzp9u~>nVvg(Qh@e(4?``v1~vxMt7fqQyZ;BH=EOuMYOsrZ(CXQ-W|
z?M)JAD)*NLUs*40u+EJ83<X_^@FkVP?!FTxUzcM`UtMPWN`ZUL)|tO?#=0W0>Vzlm
z=xTyx)WD<{yL5JauJ`Pz0S^g`!5iTov80nI6Zzb?iQTQWZKHv10hXSTbS6@=LNLfO
zH7odBv1M3j@n!s|RuB!T{wb;{F2bm5vYy4bqdop;4F1qcoUeV3goJwWEZD=pon~E0
zto(df7CWG2KuTuEqQESsbk<HpFYrxv$yc#mqJ0)asH4nxV(sMb>_FeOy6fUCMR*#g
zBcNv7sn=9rn!a{(OH?N@^9kbz#`1|G%20FP72(j<7*S$193czkd-D>TDbD@;*v8%F
z(dC*jt`)|E4v&;f#;BgbcA@f9S9m7dXwTLwA1t|Vnr<7ENB>GI-96bwYiOl?v9`LV
z@<{#g!kPKPsI^WDV!y$k=y%HtgZf|JoJd_LG#VRulVn8IQJB?!HmK^vsQQv>`%QBE
zs&?QE$s&KQI8+wA>#QDvxM8n*uHtRQVK5A?Qycn+=<u1#NGEXlZf?e3_Zhs1mwu9n
zDh|ri{F8vS>OojR<oBlT>-~SE_^M>Xzm}#zVleXe`7<X_ri=t==zs4ylP`h4<t2y6
zG7qjlXK4R((Dum{-ZbFatReAt`mAN(Tkz(jUwE@&peE!7{K!zCFW&B+EQFfG80<f7
zdJ1*_^pDt}WO_4)OeAI<<YO4-?-c-IJ#};eK=1+ZYG_1gIG&fEB>p=o`tN+!x3Una
z{x-~Qsn_{$*qFZ^=@(#@e0`A>O%4KzU`C{Md&l-jW#Zr%*mD(w2dJ-ON#!4)>67Ca
zX;c*yL%~3t7WINm^}6+vV;V>bvOOZK0cus=@ilGzzyHjCm%RJeT}9xxi-|wN{>F!s
zrXxaFZ6OYm&99f34Lg9QcPH#{veffG?bBu`h!6+nC7|ay&$a=*fd+8c9Nre^zu~w4
z0or}y3UtY0*T_Jf-;FZ==f(No{-O#*ykaiBYVH3=;;#n@Z1bKimtLIwW5)Wwec`{q
z?b8YP2U7p;BKseH=Rcf<|G%CBm<J(-t0J#`5B8~aKKG8=k|JheU!e^?2BuxE$-s81
zd9CwY?nxQZH!HD^S`}(vh$NwbhNhk|2$~9k)t~)#6CqDW|H*+_I@*uOPn2osUl5c3
z1+rR5%J>SL#=Kjvq2CRL1f$Z@qNq-)0bpjlLyYS>!GoYvN=0o(oe((KdPdQE6EyXt
zP(oV8NY)gIAHfg7Q5wv4f=1yqB2I*u?8LT`x@rd`n;EgA(;T7WH<G3W=gHoZYp%^n
zMO%QY&_g-iiN0GPSds2?>)c~fk>XI`IM;`tr4=5s%;HDD@gP|bwIn+jJlMlLd{ess
z*iI6C{DZPFPx}J@;kFq9J8vVpC8zv<J(5oHe#iR|zR)`$FP|6%kbvLzOBFv@U%0xa
zb|xvH8oon$cCrq5c$8L}ns5Byg1y$s-y*DO4BVKR+<wdxGcF-QyR(VfxPHMki2OYQ
z!O1V=)n&(<>VbU8e&n>^i~jPGn@17nghs$0wR88O03!w8(7TslvU0v9WsugeZVD*u
z$B{JcO)$cE^ZMobB-BvH0E>x)Ee)bX8;@LUA8#CjKT4u)93fk2cBS#tbf&N58So(d
z3fw-|M1%sBMk2Eh*R1-KBkf{8CH+Z<MZlg;qQguwYkj%FJ=Yje4X3f!&>%!^xCjsV
zsb^h#UDPNQbw;!lKRi@d$TuiE%zw-OUuWp=dkpegx(KvlMn&|eg;f4rE1vM7-GE@y
z2))$;%Ok}9Z+CCeS1Y@efnfl+uRObB2YZ<mOdgwf54}*b0!RsGFs8yV1tg9dED;=+
zSiNxxNJ5^fH-Q`Yh2Lv)$sUR515CSfSjcJS6aSdk%36xBOCGZ<YWJhzzDBMOcm(-#
z`p70)=S1=u23Gw9%3TpDN_VyelsMFo$^B~y!VRl#SjQ={ligGmN|0AO`(PfZgp}wA
zBq!<RlgCC2Tac3do%<0AF}WcF8z8bwuU}ylOdRpi=!)pq-oPL97!I2YLp|h-1;*5I
z#8;0N+&8PhfS4oMsT%m#c@!w%lP}qSx<axxaAWmk>WqMq$HV`A=H9~nLNM;uP#F4`
zdO^jS2!REQK|!O8y2NnU2Y@T!O#NzQZF%>3t@S#kw893k>w*B}=V%8-a=O<oE@wN0
z*W`RYmAW=QeE0<$%!oF}55;WuUDKr6eL1$b;Q9`%?sce9>(qvyt)<X=AjP)jjT;5<
zQk?h~SPR+2=pZVFDs1G70K}cB3Z8S0O$Xa&hY-(1iRy_UHWb?g&po`F*lcz&0A3*o
zJ=PfbC94WmvhKBzGKq}me#C|$V#MN4G896ArnUMzV6ztr<01h>o^1?fhxB&y$!Hss
z+J^4He`pXrB4O!~2+=wImo->&GD02|h^gD=B|OM{3rA0fV3{#G1G>oysLc)q!SiW5
zQ=Ov~z1C6+jBp4vp@<29HLM2=T)b=8nSksPsUI)0b`gS3;&s&%^;EclpFll6iug6X
zgC}(EZc0?N5|MPb0qZ8)$VyP0<QtJw!$Fo*Z^AN%{-O!}&n@iJDJbV%Lh|qbu}=C&
zQ*Czn?o(~H?!Xe@8P9<G{BYC;0YYaa5NE2E;kSpt$}?gK?zP_GD{Q<&SLm?K6>z?H
zByIG;sI#U#G(G?cocz8Ciyqyk{8Qeb3kSQ?QWEo{pjr9(Y%{wMR2RbN_u;~bdo(<@
z+?@gI-ze1W6k1Z-K#WaG4HuzYcWV#KPsPugh)JK8*cJludiK?Q<?uV*59^?|VUP=(
zQ3R1&(uM`FOUa^0QSgX-T!M9GrwUkKq5gY|UelK|E+qFvtj5G_XPXsh{#K>{2LGic
zu+sOZ?%0FhHXCj0lJ|u7=vRn4vErQeh)5)lm)QjEl1QF=W%g#>!wa#N>avWur2??u
zAxU=NUHgq(JaQ9_5M{RRd)3p|G>+0nSU)|Q9r68@fj?txEN8JHSQtpiajqxK5$hVC
z8NX5F1(5$&7$k4(C^BCH9Ze__2*OVt!Ci_SmR!>q_tQsGt=96aXU@z&XFQ=12LL70
z?!2Av%888j^FuOYXfop!WEbl2Xqi6KA^Zb0_s3yv(ObWfG!Nxvm91eF!DaSs2A|+9
z0@vmh&#@hYWC(Adn3mtag18-YIsP#54Ci`g?;E;k5{=0r_Z5O=WBX<M_Y=!8FZ4D)
z!fGpJ_;T@cG?av|m-L<=1N#v+eO~Wp;~d3WxE2AC;{0zLFvPsoujX(7*KbOaAd`%B
zO1V1IJ*gYxXmbvj=D9nxvMRa!G_Uy#(IO=b*>O`cLwwsdw=ebUda2g&b*u-joeCj+
z7w(0ge3x7ZibQ|G0y*OPE|2Tk(sKmD1;#zftD=3yf*jp@cYISv>5`=eXeTS+y6&Wz
zzcJr48l5P@S|;^5Ym7`--2BSW$FNJhAMJZ0dEB=o2`Bk%XQlpS_?xyRp{+{6YPE9~
zIH7%DlaV5M$X{(DCqZG5PPKThLh1bCiLL6Cwa0@yr(C_1D^15vmbF;f+>J1+<9l_V
zs$%H*<iA&Pe}8Ol9EPgy<C7O+f0ovvWL{vZ@HL3)`qB)Km$uxIhENyD9v0mAp-rfH
z=|L7=5nXvip!uYh9>%tQ&xPh8^NTBAg3+<-M^0U68V>q2DN?_snpF9)ZUs!Me@&8-
zyc)WHUaj5@j6rrnt%6imqkNX}8G@OlS4)ZU4S&(e2bq4!#0PFhkzUFcl=XVZ1BNC#
zRq$x&b7MadHL%V8`qhZSO`dts%FJt;Z3i8T_hxUA&v(MiTZ}AJL=v##rc#`+yoe(y
z{Qgdi>nw(|N0~r{F|5D|)FUKK$I9uWdx8hhiHM;{Z4@>roZN&*a(4{Ap4WEb^mgFV
z5rOwHQ1L8fG48d>`3&wLwiz`8lo(bV15U#1BN7A%w#2qlLGwo$CB>oJ5gt_b+WC`k
zOyOx1d=v~wczAChbYd^M+i6*<q$d#a`l3Z;smnb1*A|3S-u2udBQ*Uj^|Lkdi3HeF
zy5=93W5Htm>GRrby8(d}itC8g6oN17HEG{s?5Y7=_>1tVF4IS(i|CdIe2obp=y)e5
zfW4E<s@S7(%YwDVl5t+%G(<J_97f7dAE^jOU7FDtE*qJice~o4JtBW^H|ZE{zOw_$
zv|xlj-YMmoW;1srb!)K8XKFoNDsz4LW_C<d5fj=Zax?gpjKb7mlgA!}x>Ev<^y=J_
z5s<O_Lt3vK(#Jxrir(}Mr+6KMUjJee<{8T~6G@r`81Ga(GkQL=eIZ(i71bm%KAn6B
zsRtuMmqbozzoi<h!C#{b20B-6#Q6D&Bxed=S8xcZRyZ!a`;L9tHG2bzE;xkiw7zny
z0t)9EwXc^XsShGQT{1-LHZpixM50lQpi|rWnCBs)M2bR*Whu&DxwW2pvUBaQ;muzE
zgL%Q><9)d>9k0o>nCCh-o#q%#F{H?)d#jkjno%mS$W;7s(g6AV9hlsXjImx&JNwh_
z_2H*X!VO=VsF&)oNC~H6ICk?*ffPay(M4GNjR2_qv=Z&QYjZHkcrdegw)W9dOr=rM
zrq=EV$PfBoYG0}-Ia#80s*YWvKEbd#ydq(PWmiYt9qzVw0RPTgnek-y|F|r-?*4MK
zeQpoP6sB_VT@JfP7GKqd3TlWg*L?|B73{vP&-&l{btb}jq-&)2sMh+UkZYzQAYub=
zjCA>BnJVE-IhL!WSF-7Xq`a{5rcTfX?}Tn#7km^`3}#CAkcC;|0X5+fSt6_mwlDY-
zu`ww=9b2Gz;tj=D27DGL>1mAIOR&)94VS46`sCN+OiGR>N36H~Zw2Qrsb`Xkz}&KU
zBl9OlfNAEcp4>OuCL-BMF&O0KHwE^Dj{ku5;3x^b)58b?H0>0zpBEd42jd!301|1f
z_bEx%%}b9x?${Z0@>_xudK?=|ClitMfOM0%<Y}r?xttTHWkj3f6;$|*R50h!>a##P
zS@G%qou5q^(8jkGhr<vYW?^fR7Y}veKMdAVG4lSCS)zmjL~gkcw=q|!CKXNTw^MlP
z5B3(%V-nH5Zz=Icb^HsYL)cMyHtC^oltp{80B?{kOv2JIktOQ5?uHZUdgsgrPVVKX
zl?TQR$&75ZM$NvSrFl9AhEp^Uvq4Eiz@qTfdw}|Gz$c2-H&99H$CF=VX*Hv839DKK
z402UKVS^<a*VRrT8y3mej|nDoAg|O10?ic~m%yq+x@@>Jjk&<VseVH4a1N7@{AfVR
z`wWBtQG}N58F_kj;;nVj6D%gOwDEkht|t5X_6@|mr5%|T2jCC0<{queRChN+!1hbT
z{1{-vk|N7zp$Ex|i21dK=2`*$$#F1mOCCO?e&+FAY;^L^0Wlf5DFT}jSdD7QaR&l?
zCG<YapGnOk$YVRuHtVMMuA#7ROfHgMEwP$9J#S`vSBrFK=>xVDZNvPlgQkx!C<u((
z6L${n(hFn9Bevv;F72N5nS&~^D14V(w@H<B3r)f@Aln`cLrdC2I8^j$p>y07JsJzI
ztQ^!fgwLnZjn%PNTs|1KmhyphwK~&W&v)}3BlR}(91UZ0JuaaB%8W$jmSW$0`To-N
zfTD)!vd)Y38XvZGr!H+g3`l<M3|(oJpA9k~o5{iMUy3O{jo}k+7K?lvDJ9D!cGl{o
z)tR5XzT9s;RLbv=|KwOnt6VDdSEx-~;;D!kGk<SjwwW@wqC42EB_OhSh_vtiF8#`R
z!=>k2t{G0MmlsduM&I@}ug)|KY8<$BZr~F%5~q+f2&Wedb{wNgsTE<>M|%|i21ET1
z={RSEpZAknf8j{UD6)ajUQ3xcULFx+bwLdE7A8y1clJSD!J+l15;{%!Rj_=^QtdeB
z_Td&}sfc8$)Z})O$G&^*`;k%O<TNALsUxYEISioG!{hcg&Dak|u466@wd~{#t6-_-
zS$)eb)$u_j_$CR>X;p4Y)ifGaqLZq(&^#SBz3v&tCLxvqer*m^L~>N;WT~FsiY_r`
zOv?%@!-cixMW@o9O&Pe5Qk<Q~z*`#gQ}QtTkFeDkfs&XXnj6pQsd!{}q#0FL(W;iQ
z$99V3ZiKThL{LBGN~wDC<IH98mZl?kB2_%@yDZg_+xl0%i_5S&qkTFq#<2moFx$L^
zj>Tkq{uEMv->Q#w-b@?&I&8(3xJFC-AWn`dPtOAPqE%=W(Co_dKPs^|MiX}j%raBQ
zLQYPav4mUV3iCd*Z$20uI93_Q$F3bkttgPP+1T=E`9mVWY;sgQTW%|nyIXS=^VYgb
z#*2YuIbQqUcDlTbqEj~8cJATcR>#mWI+oi8?z?u&%8sg0p{NaPS$H-u)N(z0<oj&8
zOLXiWQS4sf>?YCdv>h4iHp&vGMIiahGT^7>L>JcgBlQ%m<YNajm6A!E3ClCffN!{)
zlk>Eak;1x8ogqU;xvhI1$|Ve{Zl_d@ZaQb?FvYN=BG~Kl#4=oxHwNP5(2>;i7wHRk
zQ{047EW7p|hncWi>D2Mp3q#+V_l~`NZL3u~$E2|z+l8yU79M7nMdnKPx_l`mV_hgX
zsyGk|MF*qwMqyR{DPdcgu2$_7%h$||X1E<o+z!^aou-eCDo37b<-6Ke9eg%tGMYW4
z@JkG%kU;$gu~b5z992}!-edmyJYK4dZ3n9IEsx0bDGJA7T+5bcSeV1gbBc}BQ=cz*
zsT#AUh?Aa=2%p(+m#wNhJj56z8xcvXs?z`7_j^9yWIB#J^Rb<7S!KtbSv{Y~U8=2h
zE5knlg8ziuiOKmnT{`zy@kn)vK=u~a6%AC9q~pt^Vm^==;+$<E9uu>LUE~fn?6CMm
zEjg$AV=$#Zn~GG1TOyxg4NWho!EIt#j+Mr<<n`939UsJ!b2sYHlu+BqpE1D{pMx1j
zspV}5qP!hMnkJiu{lu?p6#a=<{c0@jMSUw^w6-@U_W*g1hR6>~iQ%21kfD|h!QTD#
zo`vr9jYbyI=GrPQPo4lQSP1l-*m4H`(&iD2xq(-rjn>dYkiNR>c}6hOGkH{DjZr@`
zX%^eAv2wnC)fM>EFVDY~yM=v7kP39dhr&A96-vXu=-Tn18)RWg$rXvIvh!-kdwE8z
z*HED@)IUC;#uU!1w+TdWPTe)ZH=Jwf4w0jEEa4A$<w5KpCASlq*j&3*TE~Ve#}))n
zh>Jz@pILUTWBPGTHf*39M?9sbwZ42eQ3gm{4Os0<hTNJD%E17dC2gC_2jf0jQv?L8
zw(p6*2B&FpCD)YBPvETINX%Q6c6Hjm>OMXtCF2?~Gh>x5=N2io^+64nqqi<~PWnr~
zWy5o3{%bcfiPCzaEX!XkWf*bw?@6$GW=N45=_@(gM>`YN-@|y7B{j35nhq~@RfNlt
zuUg!>zp9gTi{8#$<>Qi&=5f2bp&O477r2&B8nyD<HIT&A1VGW~!IQT*_r4Q^iObj3
z6T8}3E#Fs`Imbic5~*sM`O@Z}mdzP9gfw47^5@;d&rHd*#&pj`ab7X}B9%6A#FNQc
zB}JQi>bBxSJh^*qyxFU)dn~y|idkAk+T|DR=<X5!JV|xfEj|18Su=ar_d&`HIBEAh
zX)3QC6=em9>{EhYb@}lMnnRa<@Z+x(AD_&)9cQJ;sGy|MJ%M-8P>y4dKar4ac{bhj
zT=p)T_t|??u23~ORej{^DmniDvG?BbRR8_|_#r})bCQ*P$OzdgI~^e_l)YDkRA%-{
zM^;KDLLC&cXUNPbk&Mg`N@n(6zsKu6uIqhWpX>Afe7?8OzrX)Z<Hk9!^Bj-I{c#`j
z$x;uCz0*WmpCHbQ!g^`EP{F#(CW$&EsZu!B^Iui0Q0#P?pq!#sO|-Wv$r5kgoarl5
z@lE!BD4w?IR+L6JvAKDyw%=Rjieqp=J-nJx_nmgwvTz6vNpLt=wC+})n^E(~%`eR*
zT{g`}@Z?(XF9#pNF)MmHI=)Z3^$%Wll6WbVZHq*>P4S)55YSz(NT79d4Ho|#DYjdC
zy8C1wmxHvbQfrNoEi6GJ=MStra{{hedWS50IWkpMXP#;JCFO8BKRoy~RPC33QKpD0
z<Hdm+oupTF^<Hgv=21NQ8nDnxD6ari>0qbT1G+b5$0m6F+p#E*MG7Tnt~=9^OEZz_
zuL`6-6<W=Uw{}!Z@3%ZG&PfZBxM{d{`H6GS=^lEW?4P|NJrp$L70$+?+FCU2-U?E7
zG%h~ZoKJGy$q#=nv+F2##q$Z-0Fmz;fLU2R;e?rMEHe(~y02Ms)<WXe`X*jK?o8-j
zervvvTlZ#EOQotc!SJ$r@znZ*d%X=;_skS4ENIS4>qjWi8~hXyHF+{9n2`EP`81L9
zm$S324UWl1a!Ix8WBMAXNyl4KpO~-CstQ$svgqZzJ2kDYJEXg*eybP7tU(!<+51@U
zlOb251%}m6U6ramp_W-vBsZ_4l{4a$$rX3T`D3=5tyy9Hn*61vYumTPBx7A_{*KOz
zd#;O&?~WOM`BSmR!etSg1xHb{o7`8U1etw&)P3?gQmmlbh(7N#-c&>K2K34hG=D@L
z28^UPF^UC%s#m|5M!#+aOJ<@M;ptAZYsp}o+mcrUa^lyyKSE@KnIo=$uV+&Wn(STd
zct$-AQ7*xhMEWpGx&dUsJK#)~e`_D2RqCPPrbT;mL}%!u3)xw-PoTb=MZ6zZ-!zDN
zPP0f7e4>elgG5mYXU*JBx8<Te9ISL?R_|Mbn1G;5KDk>R_Jr;0#R36i6#jXdbdb_D
z9M9~%3gb<K7!(>4nI5U$HnU``cDNi6u*z@UgtH#yJU+EpE7H&hNykuD{C!tF8wn@x
z0GN|@(f(o0Ro}J_U+JaoAqo^GjN@z3u~(bMm9w0WaSIU}eEbQx(1x!HD}$O-YQzc=
zxj(IbU|DOnzT}zkKBGs8@}S=vNlhqnxRsuoAhi;opir!KQdLw`D`6Hex_V|R?O{Cx
z2z<A8$_#u54E1q>a|A1WK6SK&juD35c4p4O{Gp$RN9B-zS_L}1?)V~qy;RQ;Up+J7
zF#9#IN?Kn((v#=EyS3><(W^~nWrVv*=a(}M(#^of_oUrlEYL4LSCVuS`3i-yo|lGH
zUu*p1QO8bK>D7<E7Q6CbQ@m646G9LvALmw2-?V+J8}~!(^>I7twgC^MOe_fZ-7fjt
ze90s%Anvc;XPNzRC{Ch@<Z@rVSAJ~N%h}x~(#*{2+|tQj;nk=!=pC9)c&}<JJ_8j$
z)%mc)i!r<B?;a6MZTX!2=6xqK+14%~=$&`ih9s;&48Jym$+3w@zeP#Ld_y6Sv11tD
zt?b4}ICNCvZiY90<_HKs762~wvZ;34M`-)n+30Y?d*9sNEUf>O-@mPJOjir_J_;X$
z=7+f4b8H=y7|!}<<bil0yi~j(PMXlx?d%X&^s2(iH~~j5LRV5w1D1Ihlt+Co>T^r;
zmIp~7IUDcegFOL`yLU76aYoZ{S*n<jdrd1<pMNXXYvOw6{E@S1YU1yl2Ag}CK{07x
zDNV#aX3BuxE$j$BLmqVXB;Exh%29FsmHI5P0#sS(ADDt;DPL9U$_j^VC}Ziucglc_
z_eO>3CQoT}Kyi1;o+DpJNV>I09Y-^0dhSImxvv%HeUO)*JKaV72J@)Bv;Wm#<&M=h
z&iO5XBfczL*z2XW^sy5V$ZZ-L4oxk)D|rXo%qk#%Q}sof$|xBC1a?^X2CH}009Toq
zA)eR84dm!2{8ltom5O^uPd>#Bv~K0^RK1;N-?28CvM!DL{mv_mn6q=L)Sds>)`zR!
z{kxGrIw%@5Y2$DC<#&HRfY>g{5pCmH1rs!j$t62FS?J_ybrzL;ZsWO80x~1N>JTJ<
z_@EkD-}Nhxw&zl>i*uOS)>Kl_XeiMH;v(MD3$|Py`Y1SOT3npep+o0c{wegvrxpU^
z1vNwZt1w%#cOV)^?5hR6joyN5YAQv^81Qh>x9v*fGv4PFyjSUeXIiQugAueM$T#D>
zO+O_nV2y0*?Fo-h{XAnqsPzI+-e*w2+3txWj4xzd?kcTL-8vP3YT|Cib8oez+~aNy
z2UWHAlkq9k?r9Fg;*5x?OKs-c3cq$Gh%DM1mw2Qs^=01Sl85Ocvf=}5+WaC;(rI{I
zUY6mxR**lb+R@3c93iE*>`&ufu`8;z>~UrJCAU_5?MWk&T)cBHKX?31xySxR!?ri!
zP^pM2e8~M^b<Ox<d%c>VUv@joj>2nOi11q9OZ3Y8a&o5{ta7-G;yOyF-w!)YHKvBY
zwh}8%s!Ev?-=n)C5s-CbA21j9VeXfwWZKWs*j3K`5UMRp;N?r`6A#Fm@{3ra1ERpD
zJ>S(-ij-6e89xp2MMD$d;}VpJ+-mH{Cc|b*RpMhe2d_;b7g$o;$~$arBhd2G2i?1M
z__p69zR8|0czbsRQlJtx+MnttwayxZHrYGGiz!$4i`SKm?6<}4*Vtl9-+-yH9!t~2
zV&dBawh`fjyBmmy1<at{&1Pmf_dErkwao+*cS4iq?^jnxWEgeP#8R?F@K(io{;vfy
z8*P?K6a6M>B||x}kvISA#ixlOPs%b_{m0nKEeypLF9?2)i)>b65oSNx>VZxF{ar!&
zjmXD(r;S=#D_Ba%1rUHG$~)cgCdNHIW^R>46Hg-m+!?pz0u}J@xJ7aknRXUuT<9R2
z^cKU=8a5Q20oOeGeZ#DgQ*C!k(sx`%g3=@AZppFauY~WtVChhdMSbq1mR9q3q>UE>
z*O^ntKbkMY`RrpXm~^w)UUReSkza%z13hF9BG|5ewC!mK{X5{Iyk;;E_Aq@g@}Z`b
zLo-QWQa=B{$nWP5-HB61ZY}r>(t^v?2BFRPGsHfV#V7!PCM>q;IEu6QTB+1%S?NvP
zwCQy?#B=nUdYN;2xj(t{pqOD8xU*JPGe;=;6?#ipfjv%$F0+Hw84(Ma)^V%P!&l=4
z@n_#(FYH1#O(R^*c?%{*WSAqI-o=w?9fXJrYFb~NX#M@P4xunE9O6yX?mPw)MiCXq
z7mJigV@DD8hhSAj#7)63CFULt80C%Q4%)+C&E3n^Y-X-sFo~-Ya$YaaQ*CeCm+XoG
zJD`50r>nDMm8JN5pCoF+I24ow%;%|zy|gU8g7V_Mkqh68aq#<k*nHIRRRR4@9pn`b
z{_yPxvTG_nY_8iKW*0qLv#<@l=u$I#$st9}&?3MqCvU0#Ug7wTClt(SR{iffJD|R3
z#$KuVqq+OcL5wRHKX4lR5R8&CmX0;Bld}lWrbH}npk%yPCM#lZoIG8Yf3ri0UP&$K
zXKV;s#x`|<<12f@F}8OYDZ}Fz=#T$k3+W3}k=+?jJ;mnO<-e6ff9@BUjJat)OCO2=
z4=d9r86J3L{L#?+?VVp6job~lR9rY)kxB4HF+q)V%5pqGPfb1iGue@qWf0=!9yMH)
zAspqX#Rjvl(ntskDPD|?KwHTG40y%;GlKzwpnSq-=x+w!h@RB%QWuMg5XSp97F<RX
z8BA-?pEgQRj**Q2WE-&Ff0d2?yxkpka(MyTTQV@$7z00&Sw99K)He(-k-pG8emeJ6
z4Q}URbEk_L{&CJL(@Mgxm1iqHFa@D6O@GgtZ~cq&<k8rHzE6c|Ia7={lz>&xu9y~v
z$m%={r5EkhA0O>&EPq@-cqg@wG;(S6>h0np4U$^uNZ;&Yf*#nb^bw6m*wcv&LV@^m
zFPJ<|DZQ+<gHW1o4Rut3CH5KSQLKFu-k2fP$ufEflooQ=l{%)-XY0`B7sNlkeM3HY
z_4>GrK#pB8q>oJV4<gaS+w;IddKJOOUcs)Y5vsb{P@5@=RUaG7=Ljvwop)}0oPa@z
zUd?AqR*ma^E0pSN60&im=EKehdhEkC#|~q`p^8CEGp`nAn@DuD*mazP;K}Tu;t-K@
z_#|vb;VFX(w}2+3RMJ&Y8GIO{QiBT|0wzl(b9=5!rr-zaG@nhRARBnF?|gyl&U=>V
zK5Ds>@-H_QVCpvFoTDP6DQ<{c!G9ltB-wjZqRsKAH<**_;)TgZ<4on>u#L)gFc-~>
zuY{8aK9;P^(&$$6!BmNs>8j4~cFY^RUW1fggsiajtOCk?Y_K?i!ipx_T?r#J$b!He
z(^{`<a8h@uOMSs^u36tT;90HAsOg>eO_5CgLgovcbJe<Wqhf<eskU#0${cBKm;U{=
zt%->by*IbduJ=!nPk~$p%#P~hiti#hQt|9345q6r*c?MBbb`aVcT&Q=gDHqYl)h3p
zEjwDJ>iF2RRfH&YQgFYeq-*<-Vrdm;L5Q(Xd~87ovS8Z_@%u^*^&J)69U-I{cqI82
zj+wg9Q+09^+ahX043AB2sm|;Ezzzz>UdfMH3#y2?N73{)k6VfYi8B(mR0Tf!@WF^1
z$ie)ZFoL2pSZVlV7Pjy1uN$)k&0AsuR>s1Mq2`_lrJqJ=fsCe3E}l${sqcqB&7Ri?
z!s4^Y<O!h&c+0x|o>l!YCKnW@Y@9TxgIjOl`5uf&qf;lDuvWPPLgvb4u>!sP;-b1x
zv_F;KD|7BzN?;NB<#)GRc4uhUg^HODSZY5FRwAYZbh<32Bph3HWT4sLhWxmd14Br-
zNU)OAv`#Y;*r9&xTYK50`<_e$_c3*Y<LciTk-^~&gk5+=-&DdNq0xfSY(aQ;QI?zh
z5O?sBAI%&IP2fQq-Ax9)Q-xF!tc>buv9g!@!G6M?r@ifQH{FK8R!0lp#xEcneBxvI
zZZ>W46)I@i`SC;i*_RrbHMy-0(FJ95Q_?&+!~e#5oG!C9CCW5Dp4SNyhelz(ye?wB
zmk`J=@uGMHNeKSQJ??B(W6{A;#Uk@DR5D1?>IvPW+#539UCcRTcW&0RnE^Easbtb9
zLV*st)BFtApvCVgN}S8LQS>xz8(%GRg=-d4HtXZr+`F|$q4cKYMpMznip2+V$E%!(
z?>iCi)%&wh>GEOjiPiGBypW&9HmBBgKLm)eZm8~%`X4lpt}%k!b-7G`POMkZ*7FC0
zKdJoZY{AjH1(?}75~D|3OH6&O!00-<Kymfi&9H6b9a=w$f>`tG^|=-uy^0B(Aa2+S
zk-w8i!RI$u50)B2@%IOIS^QgYzI8a`IznL>us}}5XNZn4yUYV8oJIZH?D#F<5^PJf
z^6an)WDe@kdX_&qt0K)<=~SmI^oLZYh4s{mVO3-AWr>vl=|1A+;~qn_5&ldDHR{Z)
zvoVyx=~_qN=+zglGZVgPCoGbc$Hr}cgu8;0JAPg#J$yZcB5A1fDSh^fjnflf-rOAb
z)vz0Bn5ygiOr|C~Q~jvi6O7I_mv7URHf&QlzYwO&_F-1@2-~$7FYY1<H>37XS{)$t
zRqWdhT{SIH3E0!D_gWU{sRDPW#@-!*=RJ?+ILRlYjv24MSOwdsoPN99adm>&k*NC6
zk!M%^$0Fj_xvkb0Kgj1r!hiL(1EpW&`VUbjY`fm}KVZ!N@ZvKCLK+Fqn*P(u^2pI|
z<EIIyBgjD%9}$rD^T649X%Y{-Am|(_<mhud;-DODl5c%7BoKO>?n*PMKK{mog;45E
zaz*r4@HcIc@8k(mxpsgcN@9qYn5Nybo9ZD}j3LzTl2bX{RoYpm{#ty1gAsdx<UZXw
zUIVTdguz_~ws(S}@st~A;+Mg^WSBcjffNuv@HFC8XZdWNlK3I~2s(y?gwTfnh}Nw|
z1(A-CMI8B2CZb2~facd>P;TJ&%`eN)+Y3tGo?_bHKu2egpZg4g0v+DrRMyb5bg`fB
zh~aO09a+aJoS(ddrZIm0i%?%eenCL$+4|XV5{y2_*1MPJ!#7y!4}~(359*u+f>TKP
zsCuPDKqkVy&%RyFQmMERbSC_Er>W&L|Lwt%ioO%3YAFH_E)T4lznQCAAAV%rcidI@
z5ti<m^=!Uis3m@ifx4-Vpj%E(T7U^w&sB5f`C{1Qauj7WJ+S~u+~xQ0o~<`ga=P=L
zOyqF=7xA9{pKCAACOGT+S7M0R*V!*vB^$*R)jqp(b~j93Cf}9BnLND6>c@(V(;*RZ
zMV`=l{7}3a`DfmR&~&9Kw~m_?Lfaor8k*4Q99%Yg2FEqa^hxyTl3E7?zwuGT7Oh_`
zTjbl|2^}duaPEZ?d#hRJXOC23g$5k?6HLL&;Dvi<<GeTwE-<u~WFE_r7jAYjuJB$D
z*f=)9>APWDWs&=tXYWe_8(sgGIY!F_B_j2h!*pLhH%`$Rei}46yI4aN+EGk1=#a+l
zUwS$7%9Ay2m`n=~A_KS$YD-qE8t!Q;jZ>0HFwLys@oNglq9=1R-a-X@>V~pY6xBBx
z!_%0!LXQFucCH3bod@kWn=op1$wpsX#z&d7d=F8rc+R(E7FUa@h)&H}M;qAa%BPS7
zwvtU?$HaQ{erfe~QunY8Q{B5*X`L|b)Z3fpZM50QHU~58UrYi<uWG2d*cxT5)fdHf
zEJ=Df4rgB(@Ox$7eD$%2@@@ah`H!~!IKd{V|K8mcwu~*pJ{EY8rToXhz&jpAz{(A9
z5_244!o;By+d#0j7xc{4cPEWq{Pk*GWu#pfom?S}EafPR)p0D%QHl)SmLu&rW`*K2
za;?4m_?Kw(F^agtb<CH8bw?HX8)V53SSWT^8wHoS$p@2$YU4HWXC4W5s~Th0y7}|N
z&ITWlOW^cn7sc+QrO~nxuE$7dagr@-6dpK(vEw9%$#w1NS3lD?m<&+_XD|^_SqTxs
zwR!xM`*3cJM%atL)JTI+v1JTeTS>ErI*!l1jFCA|a1=vXL@J`p8GIbaOn$k7KWBh;
z=fPaB&%x41^kLX%xVN!_k!p`+t!QVlu&Ad@w(c;FC$YGa<ddJnJg|*<<ukC`b_iAE
zF7gv?=&p}k9KAg-c>myx;h-6-^1gGkHf)imRrif`Or*$iDy9<0(8_G>&#|H58d?77
z+2p4Yi^Q)nNmefS#D#}D@9v73GYNKwOx}}1Xa3mL=Uf#(iAo#VnC0prT%X0CjVm$l
z0}HX$9=07)cTu`*+YO7RU=d5Ispu0OJ@~OF*v9m|CEtPAl*%cYnn<>Gr?grLA*-wY
zD;Lbq2S7DgXvc}>#B$3$!xuRmzU@-W8h&wbNSb8af-gIXUoJ|-Mstxh@be>ygn+p0
zcAFiHk9I!i1bR(fbvz<Tq$!iVwlM}XNB@HN1V8Hg(@f!?d?mQFDJg#_$(&4K2<zn6
z=43tCCl`q}D}1>~eXvg?(vhNOx%~y(+*{pC=C9?7hR!Bzb$;5u^;9x;Al3YDhX*7;
zNs>Ny#r+@M>~E}A7Gm57t}g<_!VMHjyr7#%^6J^hUItPgB}k+rC{59nuf>(wDNvxL
z@nVWa^F3G&E}OXa<%_H;D(Ea`{=vf2j0m71JD%qY^eIeA^3@Y0vfxR{toR$|Bf^P#
zUWRZm$|6Gt{>t3{{F4GL8>OP2w!l5N3{<RlK<*ic^<Q@<V#l^1NzbGA&mUPNPlQxn
zuQJTI6tOpP-U>uEt{v;eX>gw8?&yh@+nT(1yochg4~^1t_y&y6_qsV3IxVqr2)p{1
zIrkw{k17R<=W(o_{NKODd!Xttz$rEV3$OY)39)r^G!^)B+)C%xaeh#^289!42pRPe
zo+LM&c(}k${u~@)HyuaR91hyQsX1&BHOH4VU@<y@N2w+7>+IJ>Wc~LM5hg3xrM0+*
zyIt7QBP7R$tv?SratN}M{eHZaJkFz#@CET$o<2y>J@w{W@RjD`E*}cCYTU8Dv}Fxa
zB4N3gdGtD8$H-K4eV_uYoD@m#K#cm|0fBN`ig@);*WZ8n0DqZC29pGe693DJ0eqFj
zA2AF~Uqce3T7l*sbh!lFM5%!ZocE01`T)^4NE-T1cn8${lM6CB6~Bo?6e@ZVX-K5<
zsyz!ipE9@$PqL8`vF|EO79srkw$meTJKI^E<5{_hvBG8)|Ds4KI}mVpWPBuujFq7W
z45}9g<YZ0$Zh8tI#yc1x&Ym-D|Eu;So|gi?U+Ez7E1BwYo8N!4a{<zjc|52PyGC^a
zQha`$)e?J_=LSZ`l>9lxI(rKw1^duShr+P@3Q3^2H?%#T+T$QaJ$r8DTolwSyeTN{
zD~nX*$;-9?XSl`Q*n9K0^9^KG=OG7rUR1``-|~<CL$UNbc~DpzcEt?>gPpt1|L_q0
z^OySn|Neh}{xrftz*QGUtR5_z7_T0N$cV&1AHvHBg-sGO@OIBD27Sp7{oLb97FeF=
zt3R0z+|Sz#yYr_c>Mz^nU!J&yB1kH}g$iH1_NO&Dc-Mj9|7&4zkzN2-8SjrEQP|4|
z&nWl72;fScKvI`1K}mTCpv>!Ky)(hXS}2YWh)%>|@_(rl@c{YA0;nbK!YHSx!}wzP
zFW_~1zJX<bKC(c^TzWU-a%Wo;620CO#iU<;4XEf;@Lyelm*51{>U|1CR6&^nHN~<p
zSmq;ou9Ln{avmX<mF_TiS^|siyyIF1W1c*TPK9ClkDUjpWQ*U?JM3;AT-aEW7%5gT
z_6t&AtbQqU;}YAyJdS^ROwV}`MI22WxS9UGM|p!#V~L<=e2&dZv`cWPzsR3+xyY`;
z6Vbx(|NIH!&-`z}lzMB@<%@tg-x=~46nqQLR(mdBt3m|=6}Pe&W{GQ;&wy3rI~e=W
zT!?1wpIaT@4u*ddM@wYsoof&Sw5t3*+2~;CC01~!$bXNG2!-#VI&Q)YGHm5u0@6|z
z5rJk9sRqOElj(cICLa)I0ajQU7ikbkB_A5QFvKP2R`ye>21~pn!WFl>?zlYw&ARL?
z^rEyQUH0;DxI8>bWX?(T6l@skp-cYuL=FshuS<8s_b?v{M5S!4{x<kTjNrWOyWWM9
z1B#}@JJ9f9xS!e_r6)!_a}k&3>m<gyy)y7!j0PS!LQZ~!50AdR2j(=iFTpK1Z2IJ2
zDU-MTU7>`}Jt>-Fpp$v|&`)PbR_pm;^RDHb9-6<qye#Uukkxti`n8hZR2j86m2Z&#
zK#usCY``4sXy1kB?f^U90g+11{U&HKn2wzNj+VchVI4vxNe+V#m(dbiHe8z9c+~0N
z<kX}x<S1~jUb)WZ3cBL0)w^u|US}qshg4Mg`<e$DEP@!obpx3ACUFw)4vSq*z@?SC
z_{|v0$>H8INx9j_!>P*u(`Wpk;`5x+fs@Se1tDGzg6ZL4J#(Ew&e!V<Ji`NHrw$3A
z)7DlM<HRQ%PXY={j_Z*EXXVZpFa4j+?5_vx)8pUwKD_hO6x_{di7Zlgf0N%lHoiVZ
zkZf?yWJmAK2kg%{RAVaBgFm|<oPWf*n#~(X(tzW9MVonz-Vp}q71a*C?+H`6bNkrI
z3!%#9h?X8Z3QKQ@G`aPBXd_5wZq<4|HU)iDz@`^B(Kit2&_IeG&9|8Rx3M~<&Zn1J
zHO<9d<j5y|7peBu;Mq>A@JU9t#!urMo_{H@{wmUh*R`Rl01{KUe(k>Y+ujHYzmFow
zlLR0sGVHgG5hs8ZcR#7pZDxEso>^%?QzsuuDX;{Z#T-7KYkzh}ZYm&PC6!Hr3*`%_
zgXx(!O~)NcI{ntxZXx#9h@xv0F%7(VOq*Kw97%ZK6u6YR5p^LXY{c!Xe3H+!YtY{u
z_C!8d%!I-)N^0jjk33)cZE*bxEyz|VjBDY^zgWFJG#>`=eTe`HM{*a14w~2yh*}Go
z)Ee0yUaz-HbfyD{?Zd@~;m;6XWAC_jr-pLyDX-5vcR5iIVT0F$LvY_KlE8rd?Ge9R
zBw*kdJ-WfJ@&3k*wW8)i^_H)Dpj?>2u)<t)B4s2)<NN%K#x~8sR+p*=03~jPF<F&A
zDdI*Qy{o@?6aIT~Q6RZG10v)VZc@$p6KFU@M8XMKAlb?&t_O4-i7FC_f*^x<_KFPH
zYf<kIQmX%Q;>ep?P9F^|=hK|1*-XhtyU$8`j6SA{V&AmV;R2C-jF>v4h#Sv`I8hs+
zsVjqcMpYUpZm))=3*pn);V5{3x^Mp+InGJ132O5aY<&kC*il??mL6VYM238K+BW)l
z&&PFTXhg>BO+AHMzbXV-Bc3a`oWIsC|Nc5?SQ*yH@M-RAv44E-?h}D1^ot%q=2#FL
zBZcm4-(pD}wPcV-S<dzqoFV}N15N=E(A|Z@YDib}sQV$2QMht`U@CQ5dc3K|hC#P=
zatA^|+%R96U0X-jg}`GfS+fwv%I!QFCZ(+~AqfhII)?K=R5|dNP7`aFsHDJ}GWV8O
zDMySoLY=lcE=)VEB^VLcUZ$SI9>UIf!L`~oT5nER;0VTAv1+qzug=gTYDeK4-)h6K
zlbFRtVE8XghNN*PU(3qesfMuf?y_+9(-yUXSy#pU)0lS$NA@Q80wI7fej$Wy=lS$;
z4;ATMP|?+7X4sT8n3<j|JCOF#57a3;-C+KS0q{k1o^mHY)+Q0o@ulixUt*(6tOh|>
z;^6f?R+)IpQ1?a6?I6LQu=yF;Qr%ei#FkCE%2x)nRSp?{iWr5Y37wR#ogRMM!dCfL
z9sc);nK7bMH3cJ=uk|+w`7ct&#H*%tIeoZ*vXQX}&UuiqAw9Ag{fdg)b`_5B*s7cx
zkn=K<yTLy}h9L(8@j}dHeBQK8gX{!@Ym{Pf`2lJf;acJj=TdRec06hu$i-w^=jByS
zN2kQY>H7{iLOqw|ThKqmuhI1T7Y=3D@cTrUR_xWLs60OAbf5f5>)EZ48iIv>AAbi+
zCI3cQGpkSQ{E4HSGKMvDW?q)&2cTmh(xTRo-C*3&c$&^){?M7sQAiz$sWI2>yW0*n
zUek^uj!fzXPo%WH-)$>Q>+?I_tSEcrZ{z&HzEZ4B{NXlZ&*fQL_-5B{vY!;OZjfdg
zuQy7G`V?WROMfEAal_bY8?ffZl6aBE3vhhi%;PMntQ^tvYui{E99mWV`FA(Ze{Jii
zj;avXBHn?AS!Mc^h#^%^7pcjs4$5^uI7M@=6|ue|6k&<j+3pBc&Wgq;-03^qL9>sh
z(}4Gy9P9}dfIE?5&PyaAVH7wS+2kB==sOM&)9Gwn2X)G+$c!h%@oq0wRq9;P@@NVt
z;8B*a8oM(eNu%)trUx-WXe&^K<OLIC4a=0m740d$dfk&);kf#=KK2F2A>Na|3pHS6
zX;S<Yr?`OY4t;YKOwW!mIpqgM@ZS6NlcBx_vXi_nt1zVbWF}ItM)F&`!hAsQN&%q!
zVKDyg@2Al|TF?WnlCC?IG@dm7#I!?k_fa3(q5T>2eJSDO$RVh!*zh*tQQG;G4TdWY
zND&B`HM$y>S6`P#y+>IJW+w9GOMSn8-ZyzOs5JonnnN8~IHnnIIxq5*hCW=qScK)<
z3YdHW{P#!RP`<g+6Cs|ck2cL_Zrc_JWYs+NM~_~Af1l$M@0TkdxzU$KBP^Z@!g+3U
zv7x;?JgPRN{!FYD(>b@LkpAtiHAJrJiVg@76cO4`+GyRW)rvh;r8~w&2x7h2fRRhy
zAi@Q%Nbpy`$iZJn1Z)Y*hdGW+H9uv-@AKkqkvVS-J>3fF;tkI9E83c+zwDfXw~bak
zZQDEadf7;R!Ujo+xP+;6fegw~aIT`ln;G8C=W~j4kI%?M<fL@^UnIv>)69H@)^*SZ
z91P|lC8$4DIXmXax608n0s}q2sO+q?wj>W<4e*B$8RlEi!nYcf#m8iS$Y$IinplFk
z+3&%FZst#Gsz&_+_lhg%c<6FgI2|9hDwCvbT%0&Qf3$BBGLjIbwmLob)p|O@bckdS
zsqfXBrK@ae7fp7|F1TO-u0mf&7NYM^6;6zDX3~hr&7Z)DIpbu8?&UF#_)ABH|DD4R
zU>GzAh7ogr;7|L(v&5<>JpKnFDd1F&KJ3e`ff+=ixseotyP!v^lW#dXdQu3*b(EI)
z1!6fw%1eH5)1Z~i6v5iioc9x!KqfGM^u;}HS*UeSbz*OINX=xAYlou+*qt~;Y3I&2
z-0c%kBKqjw=!rQDvS!Lhy1;Q$?1S?59Sk}3&+55^sN7v~my0OGaS7tC;aJ8FND>ap
zrRnmFJ=c8nizLujMY(uwBGKhwCt-q;I9wjg)a9<@pMpbH+vl2q0D7O=hxW3*@0H^{
zAy}m{L)4s%Z5b2x6|9H$vPtO-YEyZa>XLb*a2n@%?>#5MTqw0*!w;j)kt%oZTnewo
zX1#I%$*nfr<?It{Exm1f(6F!}flV<CWCsSKP(hM3cc_71j}N<pw}g!INgWO8QmPOr
z5S#D#;)WFumjY%`{+?r>ox}h~gjD?o*D}*hzLmpC_+9ArfQufN(mq5JS;dg3D&cs3
zOUyXjpG01C`0<sf_F3`4nAV)?{z#1*=Z~U}P91x3QnGFX0Ww&Ls1s{{qR$E^UOTgG
zGSP_e;sh{Q42GsaG`2ls83Z~IpuC~!OIn>jt9V%)*Z)DDgW;X>Q3(m-_^Tj9*Ls1v
zKY2@;#XT+}d9fXLMnjI;r_P0w_?20x44$}>KL>wn$z9T#W^SXh@TdLj8R{7>f^5@6
zmALE_{Mv$Xwi0<JIw*FF#k|G9+GSG_C*dnK=F(sY*%QUe*WbmAuET-yD-fNsdY?BW
zTLOP*QN#8zY=CQI)f64_t`VMx&ahuUp*Rm(v5D$LiTg8?$~+fcyx8oo3TmnwO)2+q
z-+omuw4nK0;{#%hcl&Qp94CWyaRJY2fY7yL8YQs`g3shZ!M&#<sZ{vbhkqm%^jHmV
zd(~J~HLW3N73QGsw&SkkjH7qRNbZ-vHD*KynJL%G)n@!dD6g(S0_K86_5^B|0wv$@
zwMB^pUg3lct^|8_qv>vY?R0`tU<CIb_KAadkp30wIZh|>x(mg;jvd(5%SqPFa(fqO
zB$a3c8bI^OZ<Lmk<v7dfUrfExfjiSdg5G;JOUkVfPDdqZB-LUcn>lv9a_+ILspp`a
zq_SF+;d;WXD$RQ(nu}fmMa}(^9YIBfF~gg>>A#GNbL_32)k`^~Pza_fjo;($5qR6A
zBY%Mgtm_jPG+G%1S6#lib9;x=VFU)6H|*I~QYkdEJ{5%s#@q85i6`4<B0&ck!5yKD
z9SypspVAL-hu-ojP;s2PG+4u(u`H!w(s#-9F}5?W%<bhOBv*MPej2t{%(GW)x9<=t
zy30Q}Hy}wXayMSL-^wc4-m3FtDP5F6LT$^>i)#Z$gTRHXW~I3*#S{@1FMs9D$f=L@
z7w0T0ckO$)p^5*wTiX3ScCj(fmh$JKM4!L5$jyiCf;3i*;G@Nzz+N^jI*~}>G)@ij
zp!Z_x_J;1&G}4cghqi6BiI{SSvA2~X$fx}KvAyL1ABUX?MfQqK_Q3`AQ}Uec_G#jd
z`bx-pwcV&37FD)vJMI|#)}F9&QarmZ@`=&~*AZ>^r2uI<Rq=!y!b^^0pR&}{ZGQNt
zNF-lnhNQSFKc&mN?rFqNcK5Q=Bq}+H7{?cJn?%vk!WYi(lhr*pydZa}$>^)5yq9t@
z5Vh8+obFq7Y!j4X+(w-1t6MhWYrCC4HguUb)iq@JN3YA!v^4Ehyz?P<w@m4iUmGxW
zXv+0Y9r=|TAoaHt(eZ~UzT5wF{pMyTHY8$iD_zOyz!u_2%;?%0Nj>Q#T752qgkVy<
zgYd4ruMsp8vuCl4YDGqw(mot)H!*S?POq<cNn%fiQUXEOE+s}GET)t0%WJGrv3)Q`
z#E^#)LIWNHfnxVj0G=Ik0`J*l8VO3vu5~uCeGfbgi}UO|ya?$l+5PHUVS%4@zR}%N
zMN3<e;hzhLnx^mf9}={>g5<#z-OVcXd5F3J6g;8X15~AP@uw0puf3g%vmER$hel#6
zo+B%LA!4BuivgeGx#o?SbT^Of3%CvyhYMU+E}M&uQT);(uGKT3la*(2EMv<#gc4+N
zrN8NP?V(agB;8GcIlWyY9DmAbx{`uvFaKtOMY%$gPU(c!qw0}+DD<cVmNlZVN2wv+
znSERiO7H~2_t_Gs(~kO^+x;$0+${^9RkqPs6iz)xxIBYBOgY{e<5~I4E~<Dk=2y}3
zSeBc}m`{Ytl>f^zN))61*1h+19p2UK@uj?(dt?VUU6cCQ%;yi-W;YtLwylMD;?&fJ
zl<uDfghP4wN@2if?Um^G<Th!lal?68>KCulCDzwJdCv=FezNn`a9J75Wcxd#Uv$v0
z%C~(Y6l#Azz@gp;g;GEva#bWZ?S#@|Ddi>059|+~v_9~lc;I@J>!-k3o)7&ecs{7y
zkNmE5?pB<@gA<gz>=9g~7;Ust?cQ`|jcfH;*LqpYU&F4wWdSum)MuysD=NI*hK(gO
zm#qF2Sg;d^gw=xH`Wm8!oC7lh_u@^^VMhr<$WZAA?!@0B65-s=l?C~xWaBa2K4`#&
zm_%*xNzj^=0=Ut%`diQ)I69Xg;t!zZg}mJL2REfz=qrnm5}+D>w0^Fs-|Ck+uxcm)
zca=Ks1bafu;#d=BjEoi714qsQr4F$xr`75m+V3f=wnen6Nnn8E-cC>L57<A%dhD&0
zts!m;a^RcJ1dR#7w0jlWJsGc^A3_0q!$t#p;4V)$UeIt<?eg>vcw)|`?C)INF$OaR
zXZUd?z!uKFB{zuJI#>hYsHP1d4+z5QRT?aEZ5#n>-4P@%6tT^vVoUY}E$;wijyR=1
z(kp~6xsMx)!bI?vaz?UFA=#`65Q5T-fM5!4UOkrWzAw;nVp_Kfqbg{-&m!c&KF|x=
zv-z!P_t_+=WgV#ks11>Q^}~a;f*a1`OS#ieNYp9<yN!WK*J@;gctXcActm;@z3c>z
zRHZPc5NyJ47K5y&4<-!EF|yz?GYcjsMZmZjeEIBFi?8nV^)ko7#*3feqr^AwLll@(
zJq3?Ia&G^8n5~9w`GKLeRMSK+V5B+2TPg<k8eU)1T)szbkU9E3qik4ae|Ww6YT2;&
zYf~7jtN|8ObxW%%3gEIqc-l4nP!e5*M%k=N?c!Mv*wak@<KW&vJ$Nm>6G=`h1xnH&
z;IGz*BSx3;5%7DF22T7eKrz-S%ekhNS?k~QoCyK}z5K3m?({XA(7yM+v)a*`bMYN#
z@_=N(ero@PkSs?7d3uu@=PE$}?}TKXNj(|44Ms7MB|o56EJK`j0<uO*r3)yArM<Lj
zh4LUeQWAd80N8vTUAya-zn8z3&%LTYb!R<Dv}f%4R-H6<J_wNzWC<7LLfKvDvA6kC
zKiWD40MP4Sq^&$2c#`jst+n~sq%OUZ-~9mnJRKA8`gR=MSege5UYo59wr#z1{cndD
zFv<}AOwtsKy7Bz3Li-<nZb570$3V+mI|Lo@Y;NT*QMu!sXv765wbPqeq2VAD$Cl*O
z=sObpNGpq<0j3!B`PsNf@L~8v?eW(VXhScOV=%g_fJMFdqy@P+=hc_*%I|QHxWF8(
zz;Djcya<fZWk_W3&(ULFXbws^`bhsoFR?3p4{_H+ALsC9fPAJLNc*Ewq9z}QJuP*<
zJ1j6SuMlip8do9MOSvuG7heMvHI?=x;w*hO2K6IxWap(ID04Zso>_)5p8UDQRK<mQ
z3ko1b+yfW$Qa@NuePF0)$cYkoS9<#bHy7fz>I`e56r2=>(D59_a12}x5ju<<0tc@~
zm*^PLs!C+NW?3Y0w*juGf#kIV&6RV?uZU|A3jq#`%-&L9{x=}rI2FsxL4srUO?i3s
z%o|{yb{!1A?Q}cT%vV?7nkQo43)<WTO6MA%4KD3o#^FT++0mQTescp?_J*A*uO#w*
z99eB)RzjVD6e=IUY(<Yr)NGjVT+v2(fz8O>fGIVS7O!_@y`+I|%$1eU7+cHQ?J7sM
z8<8I%u?HlNut-M2EL6GDvW?oEwiEnS%4T~A#rSGAMk67LKENg)%m=Q7Q_h%<MCVN@
z)tKgKNx^jQuND~p`npOHAT2XSijwnx47QB8E*(RGZEZXP_LK1A-f~`JOixBF(AR7z
zkw3w6tpYkIojWrl5B4W2_`bp!yxg69DogF6B|!DV79#CKb~#Wxi!lMr^$oz-muiqj
z=9CzAs{^<I#AIR)P*8G=dH^LVe9ILEClc?p*O{H4f$!`b(w@Z{4n^D|V-(AEX<$9a
z8yx33e4D4?5{`=NIJ)B{o*d2*={xt+%)U3(okG6bbAU+gOHX`;W6chbv}=8I9VZIP
z5D?iP&IiXtIPw(LKQ8KBe2vY#;9kO+{QL+7YCndnUZ2Do*XUS+j1M@BuB~o)<jvqv
zZvdRcxm|2fg7J@km-5kbu5i#&Ssvf3G<4|Ao&NGA%lXNVd2icj4^EU#)y5~!YHMKV
zYVGulw-o4`-9ABBJ93bSeNIUs;u{@CYD}qzose;~?jd>Lh8ouUv8neDKlT~@SB^K!
zc6oJC?<?-kJJFq`t;2S(Vl3&Eav39L%UI02BK)Oi?*4s$%R_-Cy7W&k>IL+Tx^qk2
zPFJWrL~s=_?0Tp?k}35I`{uh4`v+Z=ezecEkKc*Qzq2{=HxCRgsCa(Rtv<1T=Icri
z_*Td#&+`Q)nmv)9inV#5T+zaoRXI1-Bo1M$-1Pe=V!ZU%k00^=Uiy8?kSSScJXASY
zE!m?iHA|+&>Q<fItpd$k7|SjxlkO(~02Y4NW_8S6c5C~Pitb_l8Te{QU8rhBXUX2<
zgRUsIR-4JX6uonUS2@1;*QVwXGcZchex<y3ZOUUHyse_PV087=V|CUa1rxrErAb#R
zAP_qpFmjprioTQOUILb`r$jXn{J-y+vhu~`d*YSp?IU{5mD)<6chSsWd~XzWDpP8h
z>SNZuL(;!rAb<xEMozh%D*nDd@7>)Mw*0m00e4K`(|GwR9T_Xfl@F;kC12B!kU>jV
z)-QSOK1a@c3;fPk`!K?F>uo9A1NxS)CAA;dq*7H0D=%iA%Fn)=cB{2f_XFo!y|p4L
z`deZ09@-vD_>%t7EOW2&1tF(9BO+G@N+^aUWKKGdnWmZD)V8RAr~zwMZ|_a-CvNW9
zPlm}V3?n{*(<c5cJ?%#rjVD4H!V?HsU^+H}DWcqop|VO^#YgwQsHk}=zb-GF;UgE2
z&r&(9RFgCxQTn$KUoW{qo`lNWKZw2dn$MF_`5i2)fGsR2<x9eITI1sy9RRI(>V;>u
zcYSf%zO72nFXgcx!7<l0|JXyLyty*b?-2XpRaUlU8G`9b$rI8*aq;cXL$l|l>rpYe
z4<UBt!L*<KI{Ez%sgEUW>cWwZr|Df6cW>3FF5c!%WX^_syLMxNA{@vLEt12EuDEqh
zL0P3n?j=KI^=#>K@`qHOFy5<xm^<@Uv`&B6FFpC7W#r+=b!C}4zoV!<HrWObL>hrl
z-ae{@y}X2fdV6-aa3~!0nX@h(%;cmvYOixw`$M@=Uy}!2sbEPJ@@+XtgEz(R-&Q3C
zHI@eykV<hV{z6@l<l?N)h5kNV&FI0R=(R%B0xT`{<Yc+U!Hk7mVc5lYzg2ZeB(ct?
zuLoW4857;oeu+X?o>zY_-3(x&p@huahhAJ=;xMTVHDDd-*LP^{T_KG8di_7NK?jhj
zQczmn)j!M*6%tAC6UD%UdqqlbBJ2kx&SG9ND#l*%8BaZ$?S|Hw(NMhEW7$u-i;c<~
z5<Lc-98&!LZ%!O#v7#X)&cSk-f{)*FJ<42b=825ECGH#Im~)gZrzkihNX3nO8UOh-
zaQ<H+sM>(;PGBSh&y4b#T3W8`T3_mfmxvUr0R9G2Ql=r*6BhS<v$(md?j?)oSCQK-
zb!dZ2^1=5+DOa_HFrynP$*Vyl*e<*F6Neb;$+d1jfw}6-gbY)>aQGO0qq&QZM=Qrf
zm!-oS!lDz<k;fW6ZYUp~aBvK1h6rjCS9vmfF-1O-3plk9TfJ}XZ;%mBxQb2AV-VS`
z%s@~yb>azvt!x2ZFMB8lG5fQKbmPHK(;o@?-U%b((3|LEjYahOx?5RGZX)t^Otam(
zr}bcWUHp%^@Hxa@tq(mV9FUtV8&@YfqqbPaQlRhQlH3dwcG_JAS#P&=w%f)Ze~r>>
z$GuXCnh6)yJEl6iZ!tP`lQZG?Ai3_+9lsq0#n)gj#OZ6ic)Rr}6pg0|jvh2D68m4r
zBs)3^Z*sP9S3k`N>o1rso8&3-p5%|^UCi9j$EHl|A|6_F{xK3(kWPfhJM~xfyOen-
z$!-?34t&iTnb;HEVOc%=ow<>%a~;OzLQxDN*wSU|bK9al<ap(1W<$?i%Oq2%-|F}G
z7w*bOig1pd7~^q5M!y;`p}q|!%4(?i5-}z<|G`Cx71|frHFfdxTC!%Gbk_buH{MZE
zjGX)4RH!EOmfFVe_adNQ8r$I*QT>2C5J|)MQgM$hdp%mke6ktT%40)e>s7zGAW}Mt
z&qsmhnw=3Cr;Yg?8!-_W8n}|TF=tg%{I~6V)kO>D;}iR@^6vdP@l%MQS;)BT0Wrh_
zfF!!G?tU=&N+)OrVFpnFo1)jcd0j1oHpvX6VSe03dTpZPc`IjE{Hy&R)Tr5J`MrYM
zT=K3R!8uSyPAKv$ujEhEb8N!|-0D=PF#W4Ou$Ba}nr~sb3dcUV>x)O_<c%iJ>{v-Z
z$Pg^PwpZ;1^&tj)%kKj`pn^&WSSf@2(B_?x!g34j^i85WNl;BpSUt${){>qRhX=ap
zRPsgdwbU<y@^W|3m1da-aUH#ox6x{%K;M4zCZXJ`%>-x3n%UOp`6BZ6MfK&C^1515
z@hhut6$j4K8zw0Ch|!xSD7$7|mVi~<566n*sbjC19ZTWE>NmvK2EYtEa@$^BMmGKV
z5ajljet)2LP)n=EME4OT8VNcrLMGTUECri)eU`~R4qm0l9i*l|^E5)JmDJ#7?#rX{
z-xT>=Z$2m|7moJ_Z@YUUw}_PcNSsGEA%S~%PNEsCdLHCo`Q=HZ*S0wu7(qL$EWZ_N
z`4gBd{=y602@}CbU2kfeOay<-e+nD{q1kp$Y=Ux;_*U{GQK!TT`}xl=&bKLD@c`U-
z5aCwDSX95cBsQCPdpz@Ak+Q!3$NV|qYjm5CT+Vz*ML*UgqUFi7d`5bTI5plek~Ur7
zvEgo`d2&e3IK2_yDNzx6v8NHbV@rq);G4Oc{k_L+jkJ0O9J7+<Y}Vf}sk6}aiN_c@
zoZ<pjU83ODv{m0l(^D5EoQPA#iO;%nWS8U5m_NELerb6!YGTUOIsC+9d;lwBw@mu=
z4PD<R7}Qyu$`z--&}@>O^*l5QNvdhOpJh+&Z+(}=BOjrTWO>lX-O<XC+8ANkp&HjR
zF7R=RU98>z>@jgd?wxK*^V`&b&CV!OZ7s%94cVbB>1l@dXIs98o~T{|8VTN=w6fl#
z@WCcu%w_wJLN&hGMcH*v+phMO-xrxpl%P*s;ms?!AQrPYeu`DRrv|KLdu4WgJd7P`
zh6YHeZpL2Uuo{RM@gOqlJ5e1jfj|RPgEpq$lepS5v!f5MY%BD{T<;yz+^+MnT#KHx
z+B1;erRHjFtnCmVe*LYz0O+GtD=MWD^Z2j%(9<;gu*@f%TBDTzAWhn9o!BZreA<BW
z-n`vc^DxuZ;E|U?|3p0gLlG~`0DPV^3>Q?L|Cq27@zmbEgkvF>T_Rl^1p38O`D54u
z#MWw>0V@;p%__O#bTp^U67OvUL3pY>cF2;9?Jj0!A!71gOeO2)iM_>+G6Qt~?Kp+y
z_cy}ML{ya1-~0qhaYux6ItKzhj_ENo92yuy7MIV<qtCM^+o!YN!U@L-B^UOf4`U-0
zZN7EPD6Zf?w{~?tRemd`rD`v~-QPYY`Dr=mSxlepv8~|Gg=Xix#xY5Xtp@YHiw2Uc
z#kATd6q(|Ejy*%4@=jw}o|K)?_PtluBxfZMk2N?S_%w=s_14$a6Jdd9?xJoC+U^k+
z8_I_joDR0-#uAQ%CgV?|8B3vqLt)Jn4*$^ZJoyEy2hf76GhA2^){!ET?@A2o)}9{N
zd{8jkRGQhIDj4SHO`@^P8~8}s>+<<!m8gl>9zxqAn*AZ(ccmC#Y^c)@7x#GA8wrpP
zoJwk2D6kf_>p>bANfs=F^N2b`eAMeFWpF8&;mJ<8c|5De9>P6SJ(m{wo}uOuEv{^V
zmo9le_Cy<}MS}9#h_k`YU2+ZW<YX8jj@mY-Tm$9P+pVMx_xxx>ZSd8{X>o<Dw<Wfs
z$(vIx!)gMF<a~e~79~Nnc%{J2TEF{Bn6s{yq_Xs?XZJbAjI%9LW7_k3d6@lOp|7%3
zK0g@z&QHcavYVkU4Od8%pd#WSTnaS?8B2nMBj|u;=!q=KB+u=8-BsX<+Mqf&4&}Z}
zAKQ4Ve2-YqcUZYYh=FlS!h-_7anI)?WQB5Wm2{p1)D3lSIqe*ItEoD~<f7^;8LX}i
zcP+fwIMYMF#~UJYvYOM5xE8U-Ls&arPsWBqJtcQeczo~~UKo8h6MdD@L#dK%z@_UT
z?Z1#%|BX$-a2VJj%!gCcB>$l0)CN$ne^Ru6B7`qT(|xDxFyHu`D1>`>o>@o0w{;i7
zT<Nj0gdZF`=Y4CNEwx6O*cKz!zprDSqb*}%dT`5}iuBrncwRjD99&1#9E|H;E!}7`
zyLHZm(Wg1buo%18Z%e()t>(Ma`k38AlEq9n(NHOCu;ae5HR_gK7+hNSEYcX9I$u@a
z?1}wlqxjalw_EyxG3IhL_!e^LjHSv5$kN($6)1cTwKx)VEcP?{z-PIJbKOC__jq-3
zitnm8$H*+H+z1g`=bX0LL8E$T=)TNQHY2MNS9?Tkk$4?kePW)XVy$m{phQj4$Fj^k
zy6()>A-LO)<*{x5wpko<n>f)MsvvG}p0Q3nlEBEDW()&t0?U3*`l$C+eb!<HLawC@
z97oSFNd|fG@;6rBih9MIa}+hkvp~OksCPV%K|a*2k4`QGvOZRFY7XsDunp_KpeoKZ
ze8GmlC9`{bgMEUFxCv&3&#}?jo&l#l9)0P}alX2`Fqm2Mahd^^SB3tuV!B<T>ci&A
zk6>opV6ai0ou52@HCk$#R~nGf!guMrk1c7$4^{>eSt<C^FSB$=E+?#U@;ZJO`Z(RH
zP!076ueIQ>Vzv!6T*Wb8rg^J=Ha{Cz(m^YYk1fz?9V4zCleRKL#hZR~>5yCgxio0o
zPwRdpZRFFs>fnDk;r;<ueEx`-l>F$>9{eZ4(c(VMdy+t$8Rs=LF^*vLg~s*BQf_of
z8_y6YUyYauR+~QjSO2v`-wNm=0R*uQL^*_oCn$53Y5>{cu#9#LI%q$In4wblA05-b
zEz#fqIO&H(X6!L!-udVI6Sk*dH{%fG#xRBtUs!noSl`PBk8gbTm<`An#Xa5LTd!f8
z%9xVWZiC_LtCtSf_U|rayvBYOBEfShpIn3bScXtO3|9;0Ly0--%Zp0S6`pI8)YjYn
z#qa#Tc(cK~nZzJ(<-I}%<6qv2<WcsyZ-8N&eJO#4;oU^Mb##{Xz?3O*nsgO38vMix
zM)%Ix0OjJ>(1q8wSYbT*{oGx^4;@alaPw)YzS#a2LVX-_25?gIK?tZN(~jxxu2Vr(
zRs!}Lb*usS^A9wDY)}e#kdi)vCl7+*`7G1OMw7*C!hpcvflClSMuBSH;?N`euZ^qs
zquR>(YDZXA=gITzIj2f_66~=E3DpV0;0u6M(pwFjx%F$yUvdjVQN#+f2*b7v`0Xt~
z7iMMxU@+AaVa=F=Td!mt2~FgQsw=1exi<cd`3irum<898Z+-s=ApI)=u$CPgxG)V(
zNg0qSi#w<JUdGp>O*x1!V}eAKz5&;&6mY0l6E_h^6TuxN{>K1>(+Ja+Y5cj~Kc=w%
zZ!Z*(bUi1^C#U`zg7F{M=YRgopkc&Y-sS(0z~GEfwgvwW2@D}12dEM`a_!IjkN>v^
z5J+B(1jd%$n6mo2YvezF`rkIsBo(q-n*I+F5OlD72md7k!r}&c17ONIn12+4TD24Q
z!%|&`eMuo-P)DkA^Cq;PB~$KFf0Ih`j}r}EEYu*O9Q#QHdRl*Mkmr0sh}-rW;L5|p
zzRMch09+uUaYe8;N4S%{#?be{kN54=u2jD<2`FbY|C~M)Nq{>|q{J^66BQwGg^zWz
zyt(IUUtVqn_B($i3^8ZT_qX)FAyOEp7kB8UY1&{=Z`axdec|(VNB1D|V-#OO<{Ty*
zzm((R_YsQ6^05;GY-uF^n;*@OSD!XV#!Nu{ea|*+zm4=XFc5ojYu?B)hj#wpqXVW+
z`I=whL|cO?3g@+_P&he*<;pza!jKd<xm!h8RCwvsUrw#RK84?3FVtK_p2e3~{hB|X
z1-mjzy91gDCm8BQQ~`{}F`NLStnouQV_a+{6wl`m$5(13%Bp_@bNymqgSJM?t{K2Y
z<U4~S-<m4~wDuw6L9Yp5<FDg!_&^(AVp(UiRP8*6i3Bo4)As>q-m#G;CKo3pfi^|s
z!u^PaF=z^AmyJw6%aZ`%n=cXQfzIIDKb1mOd}#*w)%J)ZHzEKOG}+x3I=fJN04a=%
zkVzaJG7cKM$G^A#fv9hq^znhubN)c7R%UeoqY3-;UjVv%EFJOpk^U`ZL25`r-5(CK
z2DXD1K8Vw)hQyc4NqYM3Z1b#eG=d`okz^G_)b)Y2kD+^%hj)y^D@Ow!*{SXk_a%X$
zzg4n?JcZ2{a(kuQ<B!e58(y0iu<JMoK)GUfAux>l)WzCt$Wg!@@4_=ErHPVUe5fI5
zsH$~YVproEuoVvxeS==n5)5xI1<2>>?8EBmML&ef588>(7eGsdN#=fbqrSy`|05HA
z1J+V0+KQtShw_4X=xVqP+<PSm+z<1(-5gk7=Qc^uMF_?IxA_@1LO^HlfKm1$^cF=t
z$zFsLi68V;NfZcg$Cv<4I&K~C{9;}{bH?72T@x${8S3R+uH4_-X2P5Y2JrWm0?pI-
zL%zTr4y{EGZiANW*jkr^-{v+R<pmb2eaOWEUh!`Qp>Otp27E%O!-M+hq#M;t8h_GO
zrDk%v&)?n@!oVN9AsmxD@TW-h1f~!6#$49x+<W*-0C75Lq{+*zo%C~5&2PRW0dL6$
z->PSJ+t%2B^GEY%T}L)om?RH?|0*Amg<!%N#jTNf@Q)M>FV}<w12hV&pv$;S0$2i7
zBFv%YKmt<koL_g)x|(R;Zw#>=kKI{vFdTGfh0(Ab*y%S09zyxtv6cRMu&X4^YZxB(
zMGVSrm)nXffcMf%l<@1s5oFj4d3~A`sO3us>pY}kp$tv+z(P^~9~Me1vQTbzcia8B
zP@dL$Rxc;1^j|#2L5tyi2>;yXuvX?p*z9sI1yN#qB2+sEO;B&X=o$CJK-wk55Y@*o
zFvz%H=avIM%Axq6r}4HA)lAO_)%A*NRe$0|f4y%1d|j*wTfWC|>f#^2d-5zXB%nT(
zg|w43;8gbm1Y~6nIu9p6c}l@%cmT+iS7C&z`A(9P{ew>UgOfL4XJ#@9+9Pv(MC)cL
zJCw{$`+c$_HI<spf5Q@Yk11eh(HXRT`54X%xQ3kTDIG1lzs;cllaUxeeWc)yXCDv-
zT(?0oQy&z8J;f&pMi7Ciw$`D{rWBe3QZ$*S5dY69dM&W>EfB-sV6s7|eooIKQ^+=;
zu!+3Ht)*6nb0=iHtN>zehe}<d(AVY~y+g6A;Iz!*1ELaa3yt+M)Dah?CK>JGIk4hi
ze6AN&G{>RtuRK{GAo-66DVsvhDqQB%Q4<YAPK$*TgdfmP%>fVF<K68g7+@>QMmEE$
zc0y8Bf*T=f!9J}8ZiMheciB%jg%opYfKgyc0|U;goAS=s6Su?`Np?V9SDtk>5_o5&
zAbDb}$%zX}FmHVY!>aVo24uQ&22s<j0U{@|ck}3rqv~*k)ybtsdW_E(hPjEhGs-um
zh^MEa<Y4aW40#qGv$A)_&QB?e)fkt*X`@O`q1ZzlhE;xSKTO$4yDsRrIqOTJ(1gFO
zc=@!`A@g_7YA+-{(wxhpH=x^A633B*XQ%c}$nP!2Ec7x`u1yI!o=fcvpKs5r%ckAP
z@x2Wb&{U270z<o}jZg``bSj&2;UQMo;oLiz(pXuC$gw<Xx9VJ*ZAcV)ESo~lHa~02
zOZTgai5o3TtarX4Dz97~Dq&Ulf%~<IMx{ZGs?W1W-u_(D%>y5$nzk|SMOuDo^V~+*
z`QsZ*!&i>eI!PG^c;ztyEd+^d4~wywb7<6eFA9^mrf=T7EKs@X*<aLRAYWLq@~1W8
zKkM$l++bBIu#C-^X-ic9oT0J@e)|R`cIFUJlQ;~<>2Mmz@#%~licrPM>qIP#zBSvd
zqxMH@r=3}c>=TBuJn#RHz4r=hGVRyC1A+t*NdyVf2}%>`MJ4n=P!UABfHdh%M0yjX
zgc?CXihu&rrGp?<q$)_2s&qq_jv(#3?s;ZrJ<qq+n&aJkH{P8&9FG}9a^KhWFX#C?
z*)rTmEKmjjld+`TyIle*(E*_H)!OfJ@}r%;EEBmIlT1^Ae~X~@UG<(Q1Azo3OIG6f
zu&D_k_*nix!B*~nxMa-nUQv~j7HbBtK>RuYv*iK<>Y?v|+~fwJMB3PEcpJp3MmGrU
zFb4}hHJ|b#gAe{R=vMiGCu=8qDUTU6_1I$plre4@+==B7*3r=@Gvp8`EQ~zKetmQT
zhid_td#*<qAD|cg!O42Mgmi|4CWr%m14?oo1G(x7k-Y_}2i3wCVc^XWT>+`z(Uq|~
zyTC}Aga3(WX7!qXy;C!B_@xjaU?;;FNMn|v<io&hpX3o`RSyuKgiX>SODUdYn+cbG
zs^bam?7@&Y-^0!#t=bbzt+ZQ%yphHxH@~UdDq_)5cVL|HP-kkUc6$N3?Tb4gjU}RE
zNBiVMzbVZGL41DkYe>ZEx1Dc10@;O*P93e?k7kdy@GAh$5)Q}Qp(4ja7FV}x-2k4H
z_UUCaj=k|Mce6r*3xsd+9avJ7Y-0t7NWw26#mrziOF-|=F|aU}VIpC#d<3jMV?Zf?
zHeIq6zJI@vGaj1%I)M5r&j9q-U#a*vQ_?4(30RW=vd;8Hv8i7P{fC5YxPSXF{Y}Yw
zS%W90(YveQUCzln0=sQ0*;e_4nKn@He9p6DTq(Z<Yi=vqDL%=r)A|Jgo6bi(iB(J3
zwo%Q5wa!5)ydlRK6OmM_;*pe<*MF}l(d%bA4t-u5bG_s^#U7|ONNXP~EDb*BJZuVl
z31khdu7xEEHUM4J;$UP~$<gEz&~$wDAiO$`q*|hJ3sZ+;j}$v~i_ZG2K*2>zGzzlz
zamM4^9z~|#r@j{^J}w>;FSu88_o_}-!;Ggz0o&VhgQ0&a>l@|Hf}50mVQJ*g7pI4V
zBwd^Z#9VR=D$ps5F*I|->G6YsD#7hn;U{;%L~YmOo)4V@PmtwvXfG>6O8Y%e4?&L3
zStJDF$&qv#4wh-YZk$cXz!T|0$U_QYQjG*ZP-cX}-pY<Pw*3a>;y3V!ae%4ub~+0d
zGkrpaA36378p;owMYK@-GGp3h_y2tP-7t7zm>rBkV1zwjv_#E6zzU!PY1sCV)GQ!;
zfn%3%@Po{B0N69Nqyh^|Ml>k#q+00$P9Y!fvLAtZ6C0!ulm=Gb*Kdi#Pa>)<$en+%
zdM|58mYTcWI{F2&%uD|mTnA`nC*BgK#|tyiEMl1lY|m?wHiCu3RkFvR^1KJX5D$_I
z3;>x=Au?b$QW%O;Z~*II>erVfLOjfX#X|gdis0}?vzH(-Tqu2o;~FsKljed_Q#SpI
za9e^mj@cQT8Qe9|En!|U;kU1ZRUVb!Qc34Fqs6XZb2<*N7v2GN661;h{W8Hb93Q;_
zI3yzwu!vL{%kCQ&@X{C`jL<<QB86#um7WGp#*{$r$@D#^hQrDjFXsJGFc45&)1%9L
zKDGKjaSq;qR|x@r9fbmw*|%n|zk?@G_xt?I7Q3<U+~KHc<7TM?vj(Vi{cihvV+0OM
z?1$Ij%{n|Lz1J;<86-mGwn4p@&-9~|sRmTX=f%kSK$ckr2P+pNBeO}<H?!LHL-?zN
z(3fq>n{Nyz(%yb34-n7hIcYvFMBu^7n;*)F4d1ZA!SL;8N*I~(0;edB@Q;5{4zdz2
zZ*zFHq=~ZWGkp@MZJRV-|B81l(L{e;$DNL(u)BbEB8~%Vf)=OGrH@YT{oHzxZmi_X
znp&{2d0oe;)%(n$m&~lcSk+6{aP7BrZ#rTJt`D*ixfN{CTcy$KUjkNVuU$DuwJHi#
z!vh4pr(TFO+K$*?GO1lxHpZL0&ipdtHK)EbII`%McvKj<o$=Le>Dr$)&Hvb-GTbIT
zno{^L1qTAT#v{<dIeXzaAsI7mutE5s{nkA|WrI-Dk@4&zs{110hOTS<_!r=`iouwf
z)M+02b;>Ssqsj&_ZJe{2pbSeh^t`<NIxCQ##zQN0maXAggjXz1Ca?OlESCG&Bh`>h
z-zEBl!0=50+ZzH2*9B5<9?3T>UPB37K<QpUX<a}uauF=GTvJ(6m(azt==MiI)W=Ow
z&~~LZKgCUWN{=yQxZ_&i8=ZYa6&n6$^ozF{y^J?3PB46==W<5MB|y1JXOveUAzY$n
zI4foB+WSHBYwP@kGjWEs2-n6hoarsj1h?mgowRL<IDyHV0?WCc@ezixRQF=3c4KkI
zc~@7_k&l3Qu9X?Eo1|$glLQ4iRRlh%u(AkMvtOuA5c9uQjkw^+I5HbU`3&cw3Xv_o
z?kP*96f@87{ZNQAQI|C;j>kAy-1=qsC?0di;vvU_ht39H_eu|p-fg~Au71GvdpdA7
zPa=AOg+Thk@b<TLbLkD;U5w9}qszt#*{pB)m;3tTi0ax7?smF$jb~*SW3>B$armX-
zvhG(|x3!M~XU^0-l6ULc6Y-Ia6Jgy=w`}2@yoe%^sL5P3l&U`l%FSqZaw4~km~n!C
ztzuP<m=AiXgeq_2#q-LQ-0Y7(4TBOPTEr|sY`@n}GrW*yh)6Kxe;ICf<DFm4qN&j`
zL}8oa_MoxT!h}0*%r{1<aidh*n1QU(o7PURzBqN4?Q!#=z}DxqaEHMw<2XZ}L`Pud
zne8RZFPvg&X~|jumN>OQ@uPQozwM|m$z;bJIvyD5FX1k&42b`ah0CA62~>gjtyh1?
zjDZQ^n(1JM#6`jx%}rCu3F;sTR@VY`Q<7h&Ghsp#mNNFopdg5%+4tEZK9LQ4!Uhza
zd!KK7R~{zjo>7F&=i=>}4KbY=I~NE&I=`y%78u4m8!8&E-7ZfFJ-~jWn)*}Blddjl
z33KP}#(+-;r_k4zpV`)O<U2Bi-hc-CihtFVn-uz~*~RknC1`Y{>Ljt3tRI|QMz!S&
ze%$F5S?46KCITpP;4D@8jpZDk_O*E&hfjy`Qp~ghKI}?U|8?kd<jjL|g__*L(R{xn
zP1-o&I^Um7d>EAeZBm{uOa8dUDtcirjuIvvXbETAT<LJ~&TJF|5*aS_Un#OLKW(^w
z|Kh^<l`36)v^R=9WFPr*3bz<0e6?Fo$QW~xzs5IS?O;c)EVB!qgmmIO<DjJ%iW#*v
z3h+5!pEh?U6Nd1g?7cU`@+uV}J2R~TQ(XHtK`OV1o7bbOvsf{?ydkz_1Evuj89A1r
zn&<s$?&!Z@vVg-%E*}Yzkg)k<Z}WpG!##K&G6NNny&dgNO`tQxXjiP-jMA>c@w^dl
z8U~A}UG9#owob%V_?7pm{d2I+iCc2Kj}I?B`D=TBnu|2^>jjI;HrVUgBL&Q$CX4y>
z{83CM!A?s>7<WVMc=+3Vte=#cm(rcfzbPCzgV}7(AI|t?l5<(4;7&$t1wkW~{WsK^
zlbbxA61D?5DhoFl74DpQ&(m`-<B@so!eIv07W&r1dZfUeV$izC_HEw-(0S)uM?QYZ
z10|UO+t!c7c|EWD-bgY<wO|`IfBkr8bYODiESu0%Uw=c&hCTPQl7sT1bYJCrp$nk>
zdiFm4y-LdgaIz=i_?MS@<2ExRw|G1+XTN>>ZTQ2<1(ECLuYbCMTGO^Qu~$j##IRRb
z=tP6qLKU0qUL6Hb6%wzhL{xn^$#g$->B#IIYgAdQ47hq+H`@a~oOO!|rxLOU=2&4A
zZ15I|VQmp@{0Zp>72aE)4Dax*mUYXt2xza=%wO#LGEF7oXt7u0Nc0?3=_AH6CVTe0
zaJc_#t8EqHmM`2+%uk`a6Eb~M#mS&(l44T*?y$1oJA)cV>v+rr=Og92M#KkY(}9I+
z#-3?+;#cRYBPn+DRY_htJ&S0g_SbehWpd|p+0nB=X4!$99^*TS&jR1cYQh@UTbY1-
zwWwuuEa!5jndNtB5dL;LO`i9olcd+Znl{m_%#p5a*q0#}AbV~05c#=%YKh;4unCk|
zmYG?p_hYY5OGj5ExSzcfJ>*R@nL^oc^O_kdeJH>F;b}jCEy6EBLf_r1;~+Zv2Z@9I
zqnXuDcV$Pd3kCiwd-cEXA_7P_!7%;t>!H7r6kmyw;=f`_6FBZJF3tRergos~P4&E{
zwFx(*X1j8mRF3SnPbSg)UG_&&f;~eMEkOpwR*#fszhcdWwHeNpP-C}#oQs2r5q|4b
zKb!Mrv@ihnGKDo!2wU)_WF0i)!4LX4>4Wjbi}zDfg6Z$_6gB{1blw;A*-1s&rT~$f
zQlxTN80-2a56++RWE~zt0<1<I2nzPvwRU~;3IKpF$W|6tPh*(wwwwt_K{su*;~|Kx
zmVlC8&@e4hxh!z_oM|EM>@wJz3U(H$!rF_niycEgev#R?9|P;oyr3itFgu#_9GnlJ
ze`~QauZ&P+@umENA@{%+GwppBS+;$pkFHegf#SYS5+WV)8!YyYbXuR-{k{pqulWK`
zoe&|*3O|v2E+{tjcdaq`V>mS!cZ^te2^|R7CF5?t-R%-NaO)4PbS^zj!cLeuq~5<T
zZTVI%vRjI+5=tFS#N8AR@?!2rHOCwX2V^^ZVJocZ!ff$05RrGxzgVHRhZy6(PJD5a
zJSyzm&Pb71y8Dks2pr7E&ycxFYyR1`zGs)CyR2@*$TkJW3<p`_p1*GyfVURhUa6%}
zZ!uMIf&(_2A<@JwBUc3I2bJ^b5u-pH&xUk>@@RK?R&vjCo;Bd)xbg9f>w~6`B4c0<
z)1|o>+!%%T1ziV73dU>~dXCCVfdTcUm=2GiY@TQd@JFq?L6N0<15W<?W6p=5_0Vy>
z%(j6X;Rxa{E9Zo5iGAiUGw@n?qj&{MV?~glCFHhn#)9BBMLdh{Lc*k#$~<p;!IyS;
zRNP50fUK-P?x0=Huv{AmpE0bSY+!<H=y26A2^F$JaJnF2-WGlqq>TW*afRqzCL_`?
z&*{KDdhU(A|4;{}dlb`?UQxjE9d4GB!9-%B?<U^E&jyajgfe)iE~p$&JAD4PPZqM|
zhg8)Sxs#2`ZONmDwyntIntic44muvP?6H?tsUcRg;oXo4$DK~MFL3}%orvE=Y_5k=
z9`sTGTdPX5Ul`B=ckLV0uTfB8u@^`m<&#f|OT@<=-Theldu(q79PxF2<^7IE|2Tv&
zf7vKE$3D3E_oi)>^8r};aS?W7F?q&mlB^(=w}Z3#BH`!Y)+opZM(?$4Weq*|?Q6N;
zk@4vw;ddNnTI>6w^f`--7a6mpjqm3^Uzs<1GRy0<r1PT}5B$P)&n3XfayBq?Hl`@A
z(4|Bn$&;A)EQ0?TZY}?s?l)&s(<J-Sjy}MF(Zs}o#^K$pS>0^Y`y`j|%xiS59h{R}
z`s4>7fopJfx?xrW!E|pG-8O5J$dp>-J`M0rVZvK*<!!YWUY(qDtz{VU2ikS1CxxQD
zPfg~{uvQ3q2DRmxr}{}$BhV}$6;ZowY-F7-(k5Qt@_MRr^@2lcp!n#ITuGp0PO%TQ
zDC!hMap2gbqbxS)CiHb}{fz9B4ar_&_zb>QgOX5k6Jz(3$~j#ymGp{~7<&U)c33b9
zn_=6!-2&{O`3K|pYT(4Ym2`tayb-n7&E=@=xnXmHL;LHIuu>n{?s+?mM?B8&CaX=Q
z{<FL<6^r{uJoKxysp&O~N7K<%1LlsUtTnJ=kN3}gQHI>=HpzOIlH;Ik<R~zmSOBHW
zTSJ|Yi9)TxA0gS1(R(wzkC_IWy&vO--g>9*&1nQa%U&q-9#;xj0LJ>riqD!cLxwZC
zPi72_ne7}qS>}tx`G4W#C2Q8*E#9(SmefK0BiRIYwwT1HlJ##=e-B9jBvyx(x?wcs
zU1Z8&_^Fx%W+eUKlKDQTc@n_6L9GMEnXh+gK0uL`L$~pLR@rxJVe2Wl1{<$!uxp}S
zS);2KAF^N@lh!uws$PbXN)b|@L+W5ml8IZ@MuZ%-a3fe;BqF@U_y(R>clL8hzRz6r
z$VVdxmgmq<EzKAR!{-#@N{nFmd0P21F4=-Hmxt=XRYz&ig=#*}nj*?EY$ul5X%zGZ
zZu?si%$PSX(ZbC3v(x;%_`-#zBmgdEc0oz;zy)}OW;L?@Ec89MY8Z+j5prIXxtA5#
zN>ByZWOuLJd@!gA)BEGIB)-YsQ%jAPo*GL!H@c|dZWlCdvFn!#<p5!&7Arnv1Z9EK
zUV_XC_!yeqJwFX2CPZkM;sCx$nr!*-GAz@NKsI{&U<V9_Ga=#ZXp@wh&Zo68eWUM3
zeFDFyczu?1wm~<*Mb;Zd_V{I+z@CFv@dPwK=vwxFe48JfsJgp%u*<%bt4LaO$mMwO
zo`-kaSTr#C>kj_zy&>fq^-nYN(_&yLN#RAd<@4v+Zv70#<|kcYko2I5lG6)`9NR#1
zrV~A{)#B$R!Bz`<A<>I4sdx=py^EWD3q(J4rD-fUpC032_!WF0qP#c0fL(?Iot6!L
zQYYE^porzH-jAi;QhOH{dqBeTGXx(P*Wk3ZJ@P)_VO2oR%B}wH1uKZ@&D&f?MB!=H
zQ{4oi38Q!~Jra-n0c;xCUbnUr(2blMBVflw7vmd_=EX8%ZPQZ$iJ0qc=_fw%l}SZ^
z3C9t^-WAaE5RztEtb?)m3wLJe4eVnuD-T;z5!yWx%apz3r?V2w!gC1DeDod`-Cz9F
z!I&1Ih<ME_)#@aDrsenO2#z_-gvQ&HdBe984Kvf-t5zVBL#v(Ryzk}Jg3y7IpD9hb
z(P2KX9fQd6fsk2mc0qJA`&qVa<obm?wAn2NOY*Gp6D1et65D`B`c`p(vCeyokv|@`
zv$?i~)x;Zsn_#?Ap1~1xQGwpP3_M#G)E5$={Y27Lz4(SDw^UJqVUnG|-K6=ukrq0|
zl&x{~NY^~s&l)(EbM>2G&QyThMJUn52rGO_e~ibQb=Jk{-n$A}6-!dmxXTl_pTM0u
zIn^ERneSVSXbTqCf(;_N^peYPf06Fq>DKsdnDDu24WF`XszSy=(0!gW`?YJV&SH?u
z=D!0Ni^|WXh}1K}4`<u8x>wz1_C48U!r||>Ho^0$>M0wPW_gB|#SM%y32w#o+ZXxs
zA7utd6&$9zNep&C`b?xvDUT|r?S&=ECbt8X`V>r-LBn?iO@sGAdrx4@4e`cOdMLe{
zKrz#gIOc%em!ZE&6omrCuv%kgvL&B^KaArFG)Qe3n<4NO+YH*GrAl>v_@0xz-3IZi
zkYP(oHJ8n)^KJA(DXwk1V+sqlGd^Z5N&I7Fl0nuw1*iq`|1lw<>2w3QhA~JWdSp6o
z=wq3xFlfkQDNrs#z?ch=39FCT3Vah$KX3R|>UiEhb-efB<3ncm58>4e-G4;aEgf)H
zI)D>F0xBt41kLAwmQ(`uGBIMcv&D&M#<NMNAWlne(hROUfjy-tiZlsVAX2&aron=K
z5lp&<I$NlTHo^|W#ikNK)SOIZ-WhB_TsY{IeV*)ZO=J1;5<MywD`muQ3VtT9J+)Xj
z=R<@~rg{xvyken6B$1b@OY<YW6TufHI1I{J2cS#NdJ9&<#B>E04Se+#^yriVY;&Me
z0-VPQR3|!>$l|=L^KrTnc9$r80Wn&w3>X)?A+Y#?>1MkwF|Z)doVDu{*u`1^Q>Q*=
zOo^U|hPk}f6nZ7Q4DyXI73H4S%w0a<s~^9;Lx`~MBo~eIl`v#RwH52Czcsiz$9uVp
zb0YY=&Kb87XOO0vOqI3_!Yd?E*1@+-Yv>3UY`}aoth$HGzP+W;`2cJw%4yZEe8BvE
z>6xi&a6d+A`btsTa0*WTRPvVfEg~tqOH8D2Yv*OuaRSw_nd3{~gC#a~R!Q#D%q*Z~
zT}W~~Gi5jNfzGnO-uJW_TGJl3aha@WG<GW43hxZ`UjBS9(HQb<ZABdk(G;7adJ`jz
z&7wUoG_R>c=XrJUh#^lhN5O&H(rQjlq^Fv+K!LD23A02ibIYm-&geF+#0MX%=$R+`
z`BgpQL|#9x&6L!$Uk%%B1tzJ-nh0hFiEsPqQLRZhZc&0Yqgx%=UnWKrHd#5d)*SZ|
z{O+^)tOcZib=nlodDJpWR)M#_@MO}Ph-I^%AA|Qgeg__!CU<HM-a*-u4*{n8S4?<5
zq$B&lv9Er<{um%4sQjU1_0Y79foAX+xyak>27nz+FR%=vp|voz@$+9b;G!94njeD%
z^S~7+iAtcLKt+Z~o30jg7sM3sLNtrvL&!_J<ldV8%ovrGh^rSLzdk<$oF7aQ9S5de
zEkE__5{!paBW9X#)9==k_wIyDd>I;AknlD0wU7=xR~ekPsS?}zzHzjzN|-Qp&!^}E
z?qQ@ma`3`3sq9DJ0$G96)W?A-Lh432_xiO6wl)P$n2(&8Hd(?rQ+<48apC3>KFURQ
z$*AN}O~|>WrAG~#{x779BDXp7Z#MkX5b3`O^j=Ytkdd-hb7NI}n7hfueSC<D(t*L?
z*aH?*RK9%skC-W2GV_+D-Nt34`?TVtb;Ec2J_R_;RE)j-)_gCq_ebPhFAbL72R6s~
z9ydW{da64^7C!KaAO8*^=P{d)u;j>m&8|Fq1&wL;F3*VZAlM<pJ;cfoKI_hDUWf$t
zdvdk-;gima8KgAWJNUJGUJDmDRn&#u$|FR;4msq?2@SH94%BEE580I3C35)=fCDmi
z<((#LW`XY@&a?L^urF!^DLhg+mQL5(ieb9y54;W`dLsGo02C?acb~3l0T3X3o7XiS
zs(g3^EU%_;fl@nZTb->?-ynvU+3X?EDARmQ)2@}-tSv)47%rr?>t8CYgu(lYWR)lG
zwG`oWBRxZ45&nta@BiA2@EFJ-3wav4j=lZ44j+loSyvrxGR$IMgCr*AH3so@A`K!W
z?CJOH88j#{$oEYQ$ZkBy>t5}QQ+~tbk4v%WbXk<=@5Fe5!~Ua{kgQ06cs*$eO(0`)
z^PzUJJaHV_=`bTNnj1u3^&80G4S4_$aJ_dSxQfPEqkh&QdcOqcBVW(n*%>rsF$(S>
z?0o2ia#iGUC$R+Q7;!ba+t@VtT$7~GT*Spq+V8u5K45{{tNd}!awq6Y>iUOY=-IjV
z=@XYf{_5u@OEk7;h5g9_@WAIxJzisx4W5jYJ^h%(?SuD;)f04sHp`SqTS;;QqR}i+
zFGkz0MQnZL9BT1;q8DX{7!hf}+%18;6VyXd_L%qUW8b&9YhJk0krd)CpANzEo|g)G
zUfX73S3v!I+MD{QH>1Dx<tDsi2YiTadOeayrW6Mpu<7wAiP4;C|57Wf2K)hoCc%(-
zCc?qoU?3oYWnaOho$WGR@LPSB?**pdA4hJBs~g=bZinqdZ~i&$Q5({gEw<+v{rNuh
zkdnwm)&oHj@=d6UfKTyUhJZ#&$QhQ=@{=+m-izO}hSo|LdKt|gAMs%;IrEsONz*%!
zHq;*ivPn+MQ_Ue#ZX=k)XenL$l!IunSs4VZ1?Sz<4)T+aJwmL`3JcPPh<Tr6TMQcw
zN(k_(PCH`1Yhr(w7rSnzdR{$<VI%>~k4E3l@Tvm7%scpbJ@wAOcG41NR}MWzy2mWY
zh0`&}!x=X6*GTZNlRbggm9BqEyX5%sT_Ss_hp+MegU?hi5SA7$?Ea2`v|EHvpHwpD
z9X6<&kA{n5epukQs#qNdNSK<Lqgpxb1g8z{6DmKvm}{i2Y0N_4@1xa&4D`urOZCod
z9ekpKm1Go5*6_UT%s-F@IvbI>t2<lZ+GsX3QB9nSWSVKitbFDwJ$khHnfVF%CN+~H
zOoeGT^!PptPke2i)!fErfcf6lRHG-uIm6kd%Rn^Hr~Z30E8@;N1^gXSMYy44#9qx3
zvv}yX1eW<ojpq_Vzsg(@9HK5{Y)vn7=iEQ99d1!3PhYKNVozuaHS0c5tZm&Vo#QDF
zoA*i76F%d%n^xOZQeQ?y+g}K0>HUc)<v5zT(z~`1bDv4N6x2OA<8INS@z|ozgVdvh
zRV(U_Zl!3{gdKw>k*hXwW{bj9#yqe=AIRP>E$Nj>n2JX7u3m`booN|W`d&c5(l6%&
zrQ5c)lhmBlx_is-8c$<^!8gADRO!qsLC?~3PR_0WI9rA25!OH*PxU@a;Gfp3f9YuO
za_SJZn>$24>Oa;^0VM<=-!Y#^$iw}|m;X2I!oUB19VECMXQZi6{&OJoFOk$q=>SO8
z0&VUA(HeNvMgRf)l8p`UMS3y`n>D{$&A9fze8qqN>fqOPa*$AhfxJTgKmLGLG$~kV
zHb86P9iU0U7PYdt8$zs%X^1~<4Z5)%y#WK)zl*{DuI>G|-#Hl#{>mrX?y3KWa5kVM
zwh<VXEFqa4*m>)U9~zhhkn{Dz?9>+j{we<RFaDn*_y0;o?!SHX|D@^r*MIivf0JcL
za*%>KWC_&iN&p1&D3IOCgrw%RMN*Oi{atV!@PI%?{}>qENOQ5~TH)msNjJo)d*_4x
zsaMC#g#i^eKQb%t?7xrD1M1;E;2DG@IU^7lw(1dZ%9VkUc23dQqk1eUq&0_Jo)B?u
zB+0H&->fUo_7x=*?p6W#2VI~`7KWrYC*NOvvay$eBUA!QXI9@1;LHsHLBYZ?rTh;F
zBI6CzDVN`9EyyzMhJ(RM(W2fZsS`i!{$Qj4sjlu<0szU5fEwbiO=V#Rl4p$Tp|;&$
z!{Gn%hj{}`{Y`9mwEyCt;u}pat^wG^D1=-ATj*nxmB3MHJlqG6ODcipcofh}X^+9!
z&|C`m5>9H+ob=aJ$E;%^#m+M5{3~KOac7Rs<Wig?m7MT#!B>Dr&83EzWXIR$4cv)f
z?hNK8V}Q>bn`%u2gDBLd`Fzi74ll*sRzrP2$nj41JsiP<X;urc?!)$fH5{`L_JDSp
zwO7H|V~QLAe^kZ@tUSgLk>tY;sK>&44nWIQ3L$@>Fwx525EDFv;raF{jVYk(cGA2h
z9tSi?h7w?M0<rt7fbimhf$N#dh|iT_9|rv^bpPYz`i~zH(1ruC%Y(>gcmI}u_YlYc
z$VJ2(Y_^v{3BS~qonwpKl_kag0V`PsIZZS%)TDAF;|UeffH`z&m<@m)p<#H`Na&Ky
z;f=@D6`)5yzq!aQyLY`oCO`5MGQr{BMYPIsMnleja(ftJ=&rv})d$iB2QU>4TPfL-
zg%n}bNE<E~#94O$p;nMQ`y-^=t*nu`U4OEK^{^e?^%IylA+P)h7#Vzf$cFa`y$ngY
zd;QV!XxN2o(?AVudy~Hei!FnAKy8a!ixy+p&nI3{{XJ{`kDtSzuNtJjuOzwo;_umH
zWC|JdMB~jAgV@d?8J>};90U4_pDt;mFy;5OZScy4z+C|3vl8ee{Z7jo5th_RGhtjC
zY)B(X;8Q#D(c*Q5c3~2rO&Wv1)?+45sBh$!jolC-nFI#b`<kl4O7Ff>FYZHiIX-hJ
z7gnt6)Ag=?`!e{;t=y%Dp|xZIwR#z>EFeTk3(&Y<O$Pj!$QiKBc>f(d!#e;CItm2F
zwBF$Tbog?0j(K3_o7zlGMVMRa8@qq}>q13g<&7IdVSn8skriZ>qqdj4!K-kKw6Vku
z?_FsQ?)v4P98<KAGW-CTga}%BXJOIEZw|8e?UCLUmtf4&Ugej;TB8zReTNBtTN%Z@
zr&*CO0A%V_u{Z{L(^e>kegt^BOZ&|aSBK4LfhZ>T$^Y+~{jYyrlT`*c?r{l;uKT~f
z`IC|)>2G~NgU}lW4iE{MD$uyrD3iU0B$>)dKscG)x!WTJQZFGV&Z1kwfSWpyAYva*
z1p-BX8>Cj^tYex3=nkcs8No3iP%_H&gx<36@2$Hbq!P5X#s=<<4%J(Nld^&utY84r
zvOY)3pL3|6=2T2@{a%-ovREOF%x>d+KbVrTu5hKFUikxlzK>0_p$~(f1k8|FWXU|(
z&fp(CDmQP&BVAV^?&_sL+4OJhP1er^9{wr!<(r5~T7l?7qx;Ycp0f|4NC>yU1oIEh
z5(tPj(PwN)V!DZu<GbD(Rk91XOq>>M3%@5%QlzU!<u-XV1y&XxMmhO#wt4}W#F-j)
zzOxR1@fDVr)O@8Luy^9j4ZCGeNkreKO=dsG`Kuf3Hsb6em{vx~a?d8B@JywFus~;i
zdMwTGU7gasgcYj>AvwJp$t9n`sUl+^XnlR_wIInP1dvDs)9(FFXW|ZsV`P?4XW~vU
z30#g5Z&>-F!PnkRvr}T4ntX0KMYyxfESmmiV*N?KPEGq$2JC2EqGCPB&TW%%F#+gg
zUJ1^fb?)7EpN=uzBM(2R)iM(pReGnXF@*M>?gN$-k!BI!H!3ZHKwfzX=z#Zr4>hn#
z1T?&aD91awF9N&Y7(;BVW%o_CJDwZg3MQ@A_o3L7;vnweiW5vNS&Y)|>~q$GN6RlN
z=75;E%gxSj-G2)>Xj4nC-VhV55>4IpC33&~&6lz(mD+(Zr{*SK1JO%N8iV28NnqJ*
zhPi8lTOU{ZH5LZbBObh}TZzW`@2+6psWsoLo3z{THK{QonN$bgu*w@>NQnow4n_q6
zJJ+2sV1~KIS-V6>)k~`1J>j+Vb}eq#YEWoUE1j}5<lF?w#?JFS_rWB$3FK!lV}YHp
zLp4CS9j$Qqsqs%1z5pFc;KkzOHTsJNdqNd9)eatfCQ=5Or?KxJ0vErSOkX|NvG0Im
zMKlNj4?t{O2B@b>2x0~#YX+Pq9spJE06cP%SPcBpE<0o#39()VTKy=8=36fnUzLJ$
zdIV&!+{g27kaNJh!I8SJc+hqg%lYjIyDiW)6SlHIP2u@NLWZoL0Hq<X%O?>UA=e!q
zY4Zz!<m>>N{Jjx~;p6!9DheN2`R(D&F(4mSRME-pl-zCf0WaH#k^6UT<dz0aplCB5
zTNH4DEot)y<XQttU71CQUZ;RB0oez$g1Ppq;QOYQee8l{%esE`y6mK~_a06P;TdBF
zWe5Brsl8)0!h|=0B*uyf%HAF_GvRMdr0^#7zv@<lyPh8|oB6_{5Sp=Eh#30gVye-A
zDr<-;2XY#PcTsIWz;#vz?L_isWrBn=FGJ9KYEVRNFEAY{WR?TXQYibSi4Zl1K*A;@
z6YM;(sOaI+&Jh}PMZ#Ei46j_mQPuwrgv9faVfL01P*7&7(ptDQJ3kn31P=7$I~d=<
zvIxUp!HNaROJ8C|WIt>xEY=;B3s-Kwv3-Nv<+&H)E|e%S#kqDFF0GQAbdD&zjQlah
zZ4!Nmx~X4(GN!xRZ%4ES!2X)t*$1Yp*BZZ=k_aBO0F|$iz49`YWEfX63EOFTISeYN
zrF&^Co%{Sj$-6n=gBYD>>i$@0S`N}GhQZay6u7WFejpfG2)5p9y6~5xvcWkB4zXY8
z0V#kgQKWma8(hCO<^XwpmZ3>dB#OTqTps&hi~P=EOgm(kPc?LGt|i*Nt<W(-?#X@W
z#ZZiXPJ(^*BKMXf^8;g!K`EBD;%o@Vy-XUwV%{uh9*$3~4%^L5Y^>Due+Rf!T0XOY
zUceAY1hkdhthadIk`u&(1ilqG>~j*^*4t<6JY!C%^X}*!Mr&iu<rDiWr(l4_JMFPL
z^F7@2c2O;;VBMhJ4QQft+@lYsQ4RQtRQG{pqcg`(8ZxCJu?Qw(*&R!C(qH|&{rh{^
ztw&~K+o^BO|LZpS?@eA09yD7~?ekTCR>4~gBr?_;8mZQNJ`a>ka9Y=7{E9##Cx-xM
z-YJbY_{B%ydB+9C0k%&r7wOsFDMTXsILGw)Y1trbhH}&?=TCq_M<2))bjzplqCzj2
z^xr;564|Z`C5`xeQ|bwRO7)u5j5$4(!+KJtS=<M&M5iJz<^tB-qhKU^&Ss{{3Mgm^
z^Sa>ibAs>1>zRe&3xnqgn|MovABR~7RSM!U`zgqMrKnW_IjcZr!`Qon=tBfkXq=6i
z1~iuhzSj)BFA%Rk#2KSaIKm5b@%AB;rS~`j`_EgwGzFeFQy~N3IDH~D&HDSsETdlI
zJf1V!nz5wt-r-x;3>Nq&1TAvG3DjSdTw4)2Hj$iY&OX(Rsk(+!O(lP+iOn&1+g65+
z-n99k;Snfp3N6)~0cnY~H4!c#HMPw7v^-}zPl8zy+$-l<jz9bq;uI7F3UMoS7Ux^J
z5>R+nd~>iK?Gz1~r@6dEVGL;iIYBA`GcQ<0*hzNF8QA|L_Mi+3Y`Gem#q<UC^2byF
z3NMxT_CQflOPjy*X6tjjn~;er(rW4ia2j?M`;svjz8kg8Kj%ci-%RP$;4>>o$iRf|
z{RUVTk>=(R#e^|tt9s50Gnk`Hy-#c6@NLAm6rcb=#za}a!6qvpln&t@NP}s2{BX!|
zzAiX9@<c8T4*^xQDTF3>%lYXvZpC4!T;7TKNPnKz2QJ|aM5duUah2$;PSo-Ss1qu|
zCoV)(bQ3fa)$8*ki;5H+LTXb9vjVIW3=zXfdIOC@<3H{oGTG&Vxnu6}4%3=bDiZHW
zn|6a&3avSa9W^W}mY779{P6k#63R*eFOp}Oo`iQZ(wXzd`Du$Q#a~M5p@dY9VgAAX
zs0PJ>&_gsM?rGTkSq+yGb{TAfVdQh2+QQC504cKQq!}Gw@OZxPZb^6<yDa{+KI<Oa
zmBB(7V$70ymcRoLkZjEdlXY2oh^~!7y`r_kPi%3M-1_gz+_B@=*Mw_dOok>J>vxRN
z9=(6&-YhclqRj1w|5!zsnuFA_s>7c^sK2Vk{}}VgrI8(&KmSXthi_!(oV4NS&`2`O
ze14$l@-RSEo5nCR_TZ!4!`C-9XsSAt0<`BgrH(j67w-bydu;YcRlv-&-pDbHyJ^Vt
zqN<g%QY(6oVq&)G>8x4UY>Z=UL=fNH@EW10qlLC3WBej2<syMy6R3Yteizj%5)9{K
z4VmfZj0WN`+KvxR&CAogcm!l3B_Oy@N!nqXl4S=ue@8~{#pCvuhL4}8gd3wyULY98
z8}cU_-cOjkQ&iO>RIO4xo-0H6&?;NNjq<)oz-qhHMO4H_GzZ2-c*aKf*cS3GJ@zkF
z7ue0ujkbFKSb$-WDI$sK8fMbQ@ma)3EEVEe#Oqkxx%|Aa^~@-NyokywMDC)GGese*
z_T}oFOP*Toj{tV(0s&3;qMAX-o=xiW3laM!&hKx)Rz}R&;bDO8!xDvbg1+;*!$VWM
zNGG>QW7LI<sM(f-M@XZu-==D*>SH5LKg;&LCGb)O$cZn7+J#4y$5IKzQkk2p{SY*f
zd0!>ysns|;9e9vCEMOjer`MbyO>eF}`_lgvt}oJUmhg4#XY<f)8n@{f&cCF>n&NID
z4K5n@9)7Ut+Eb=j6BTH;d-R@o<>+)ngt}{h)mKj+2AN)RUsG;vZeNOt5Toh@uGVMk
z9jGg!f(%Zn8kn?(%47#ri_+n63YE^Y>{~^-UzDagI5|(Mqi{nnyV(GxA<0QG!vRIX
zjbi;Oy<*m@Cu#Tq<wx!xD_C$*fK`2B?~8kE#4TFY@HhPab7mhIXi-)uHer)}fuLxK
z>MJEbBp!*%eqtav9=c^jzrP}?M>kQ%Xk@&)o3JImo!VXEHrD91Lb_e-PasX1d>r+D
zQmR_0$kMmymm#W_`&cpZlhnN<#&M=Y8eHu02i(yEi<(GxEZ?O~->Y_pQW6>=4p!Y?
zi~eaSBE~6~$Ne`NBtTyIF-;O&3A1SUF@AG>P}3u%u|4Cac&z~K4|Xmr|IjC#Bbgd}
z8*@#c?-ch2je?<w{c!PY@9IXVc`~6URjo|<C4TsGf0s7p4}=s)_&WZ&_O{2VzS&;W
zJ7q^4nezANd&?J3x#SdUlCYbJH9O+5m=#Cuu?;DqQE(N;;;dESd~Mi(^G)ilv<8!k
znA1@9)vvJ`=_Wn+E6g2(ek*qNHLVvN{l6vZ&N#<pdr+@u;kxX+9NbDzC$By4?%=%6
zMQ}uY-H3QRz-i8<zK|@<=egKp5-ulh4Ah6G?ImQ-vtdIbuXB>H`<kY5`d~Zn^1~5G
zXAY6e@ghVX*PT#UF^lmDZY&z^EB)2FBstF4o;hvWrtMPO?@?=A#FIwKWc&LOaH;cx
z1o+)eYoC1FgZxV&Uj&r3RFYeRU<iY(x~Op6snf^HHbf?)vuzZ%(Gm^={^}C6Ft4B*
zJz7kxRev64(LOQ^r&v&3$nP&H0Rkse*eoJ(KzVNY_}KdufTv|#Yt1a;&8HstDAf#F
z!+3~@T)tfTnd=|`SAGkUM!<Wd@7ml{kXxKKij-|ygD1!Rk|H!FVQxug+ub<7ir2bJ
z+<!RO?q;8UdhAPT{y&CZS}Q|>WUA!9&bZSOq?uz~cWS)o>O#7>eK~T6RJc1DyjsXj
zV@^JM?ti55&Xn)nL0x9-1qZ=g)$1H11s>QBDls>`H*@)o4(M_Pud^iyg1g^1lkwsY
zJDmdyu0Fw75#gJ^76aKf@v<>ygG&0!d|k@H&mRh4#Zd5pTfeHo0{HyGPZg?fY8=um
zykXak=h)hl$8RLv>9cu{Xov96l695+&obqHT+IyT+I|ype0rbVh&T7Lbe?X5Oxp9;
znLjT2a&_BlsknyM^|JQqNb(7s#%tAP>Rha_Gk(LLJn1Pg968XzdQP;D+T1QvnClid
z?_J)^M{dBh!`_wcvuzzr>}CI$n)GJn!W;30b4D;dN%A#-7GSJYu>B6EP@;X~Y9bvM
zgb83<uBt+FmtcC{<+)63Y2wz$Ohs1`c&P0sO752?0uqJEq*wf%x$8r;VlvISf+aVz
z^_@%1hHI%FCL8HorT7TQxP^u4Ifl>29?0JqCR=#Bbnbx8##@@tHnX$SSTSj?H#Izd
zpj+T<XIkE9s~hG<cC}vCf~&S<MTz1yHT#6}AJA&LU&`pQEr@B>-+iX+hPmxhZu9Zj
zJNEgv%(f8kl1Xn9`J^R&Y2WoEd$u!~eyv4B&6M~btHsWJ8L_E#MXnd@H2sgK)2`Fr
z@|r$)|9!dodnZApH3QnZI33~6u8)84x-q)0_6eLrQKLhSOdH!CWc_;m2{$ue&G|;t
z)gR#y>qqTg3%%Bx@0u%**A)w%WU{&oo7TQqD-4l+Ej^4`dWq{VsD3x*-dxZGUANc9
zO2S0!rMwIOnWlOk2&$5Md$0bvr<vCv_34@A<THz{567jJfDK~%M&SDc=|mo=9gZTD
z?zu!&N;_nEJ`&x`x$hg!jkxLQWx6@9L)6IT({|f;MwvDNIg}OfO1|t(t^cGzb(wBx
z0(2PSFWmTGTkZBC#clm6NjP*tuB1Oqo`zJB;6d9|a!@T&iG0RkIc0v0ZCYBvJd(Rx
zNiIZa0T?KM>XNXN3a18J7eoeD!8H~k@WdTg5M#`>yfsip)R_2Vq6NXRp^4U}R^tvd
zDZ<oMdS>3(Y$rJLPsowk!PSrq;+Nnz5JT*#S5|8`7w>&KkZ%sEdbX7K3m(s)-dxHW
z(}9ROM_z2vslhnFM%w7+6P0Z$PUb7LR{qKLysKBSk^E^=ZZe_NGnG@E=TXg1Wp_W&
z<Q1L=DRIMOG8y_GWGA2P8^G`mUZ18cZwZqPL46<I7Zc_6-42{dgAS@wCZZ(lf%k$w
zbuQGOB9$iGwM;D)Bl|kC4K(#7*I~m82R!DO_uFG5RrLO2Nbhs2T{~@=MIEX72M+J3
zTlP2i3%qOl*Y7Rx`mKPmIHj=n5hJ!tcCSNyy6(xg-9`sD;n&%5b;KETmJ)Ve<K8}a
zp$5T`)bM|LB#|~+K6<3~=Y-emIgC^;J$LOxjs6x|rt!;_nMgO#zYf6floq;zx#&Z>
zUN^kZJn!AIyv3t&5ROv`;)+77En}4Ea}E37!g@4EA#t=5&{F&0%lOVo4zGzlaIVXC
zcw3#WNiqpCts)QsNa~p12E0OgV6%R=y8MUTZdIyM!6sb}B&}994m&-!V5e9+y`af?
z$JAe$9UIvi>Wjr&Ja=nRvKS#lo(em72hAP8pqvrSL1c0Pc1E2rV|9Or)YdCxuYgcI
z-3(GeFOJ+jx)$_c;(#b=ZpcUpt7ctCnq9=A@Qz{D*RffkSZv3`+Bd`cUI>B@f04-=
zTvhk$+3!I%xlN;GuNotHRY}rDey{ZGI>%AR19UGF_@F%i;3Kr#dN2X-jJ_e>Pa5Mf
z`bDP)KrVFrbO``OK+XN2I(ne;E<zE#)_Ft!NR3pk?2lJICCcs(V(0p}y}#=3q~*kx
z3AZshF;g`2oI|}Hckk^mvzv7Hul%4|I;-&zgiJ5!DI(=LNi$`z{slu2bRAB)FGXqs
zI#vDS(f^~DoHxL7FTnM$&ldYilCo-;-;*wGR-+_IqDP6f!<Z~Kd&I#b#lJn$&M<Od
zEonLoa-e!*<7Sa2;n}u3G!Pt^&G%UZGXD#x+V(=H^30caDduK#z48HJdB{2gmzl@x
z9-#U#M(3fk=|+xA+|0HF#=>ahd3i@aeo!g}jC4MWnKk3z&n2;JJ%8FEYsrn$lc*_}
z=-$I^&GtSz>w~%F@O{pcuij+-UjpnZFcG+jIuC4t#>%c5sGWXW-hM93wYzr#6XIF6
zfU3a5TLN^aDh13`ATsg>&tqW<L%2adMt$!El=e28{v^p|?*v$``bUNVhX`A=14`ur
zK_}60;0|O5Jf#X&Cl(J$OgGA3iw`{)#^nz5FDf%hlRxMZEj`hB=|IEy4XyDeID$eP
z*Z>vSS9@Z|lMtPRxnuVEm#0|!Ea`aGPz}43r&eE*okujzTa{J5lxXPIsFX6*zcw=b
z(4P)<zs<wz#=@*k6O#<-Pk{?+tiB~v;xA3=X$F`bMulX<WX)IyFQtY^6VKQwW<w?G
z^f&{pdg!Q)X5LTpx<}618~E3d*t=t#1(Rp{`JF3Nz0xp_Mol;l^a9Z`cWX9Ns(Ig#
z*dk8KVbb@hYYX{sR9?i)F7NnT%qN$xuC<|^AgXCSy6x^rj>rA)Vqm>DDE)m!%1p;5
z5j;OBE5lg*$`omXe^xi)AB$gmJ@fZ0r0yoDTJcy!5690<9w+UnA2a#5&Ktn+u8vPY
zP>J)U;2%Qqrn50=XBDz7(b-@@$JKRp?xhWw^%Q#goM&?{qsi|IdRn~p=p|$XqYi!w
z1oK9UuT^e+bfG}p^<W&HH#as@Df9)?x>#KM=dZWFa4|#q6Cu;GsgMsoxN9pL@Q_~w
zBbGZwyY2Z*%v2E#o}Z?(V58?y8v6PYQ!y8xWoxJbXr`XRz>kH>jJK)d;(FeX%EiDA
zBt))wm@>!{e-cS4cJ`sL6QWayumsF-Z_PH!zlKdH2fqiihM!_(RHhMcBCp;^=`<Y$
zzP<<o-gh)Kp`ZaEe0w$RDX0j~b>2~r$P=_)Anv0_?cbDh;$PQ;dL5KJp3Qwh6W(An
zW1AETKP`X0U~j11(bmy^Q;rMP%;sggv;|>cvqAtiDfGKHS??pT#TtMboOaOQ;ObVu
z0RnYpy|T??zLYh3vPdlbP5q-YH5Q#9j;?f?b95YcMO7LgIUZaC#N$}pntja~M}~gV
zMv>MhZtgShsvGXv;Fh&l{i36;8*3^v6}#K_WlVA%dA!nC=Lskogvh1ZFWBXN5U@#<
z@7CvPWuqv;vBx{(dv)nPXk_9+BQyBVjSS$4nQ)3+`wJ2ytqLQVgmAl0xR6PB*xZ8k
zjJd4Y6*4{G{LzD3;n6z$@Fx6Gv#h!^Bb!&4<qMY2ei2Pg1+2}ySZp#-Z_k_%MBdv7
zJSW4@&#I0~hDdJs`|;mD3--dU;({swX6AxF9~$O$W1LUp<n&dUNwl>bCjJj!0om<T
zv&(;|ur*F1zah`jl%!|p+>A<RW(8AD4g|FR%TN96g0~E9Kg*#iEliJmh%k8z)X~V8
z5}wKn*Er1hgRX;Rt&0~xt&|d9?A#=iAd|)~ZeWm4*h4n7d`zi?FbkQ_l9Z1Yvow6R
z+E}H<9kJP$!Xt{M)D2k7r)^F=%kF>c$$TKPuP7GqN^q>(k12{XF3y5h|B=jZkw1XW
z>lzK&-=l=n02FWO)HUf_P7l~g&!TZv_sI<`RT0V$Qm>W+Dle98bMrE-E%3OhEr3m6
zA#(3*;!+5cBixiyqbU5%*wv)3b2jyglk!?$Y90D<#H#E9>UB0!xz=pG_uJwF?qHgK
zL0z6{lY+xYW||vg8qiMq0W}l6TdjxiwAguecvjjOs6cYArWU1n4bQQr8zOn1uhcfo
zFt$d|;FWOE%!vu!dkRvC*?5?=g$E-DI>QcaJa6RqvQ0N`YK7NI$Zq+)>FQdRGIFlB
z4y>vR%YpO$4Kx<_Vb)+pRtOIcL*j52`-twPxT|;Vb)6d$VNYGj{8hDD|G-U)nLigC
z{tAhsXLJNV*_}U%p8G&6j1KfdpCyh3tV{_qaHf0hXyb)Cyue8PHkX1UcjZ7dtT=K{
zqt&A^i>aaOa0K(khpYk0rv2`osl3!9x6k-nqo|xyq4`V+hfij9wt9vU-Ah4imM<uZ
zJ|5FfC3?mR*3LNlfJadU3KDHOS>iyXwrd0M768j~L45H{(;t*VJ1p+_RvUnfipou&
zioXpb|9$SuPPz?`{3w4c3z#B*2{29~10V<yrZjbrLd&Hj%~MQx)7pZ_cF;v`LTn<3
zNJm|qV~JGKwq9;vyP8Z@GF?U0*V!UUB&Bd!zwxz#)ieL+q2X*3`zDX#PU8uhgb1BZ
zB@c)kO)g6`Q`p3t5~r_j(mD025l+&&dfz_}F~C-SFBM6mWD=-ecFLYjrXrs?fEYok
zUnHWPzsSmMQc(3us@K`F-s&+NHcj=CEOd8%THV-nU|#D0tkXm@%%-)+x;;V79E8)<
z{GrlTz->Ny>a4yl%Uais>AUr9iB!VHR0q>7tB~wpD{<W8AYd7BO$UH8^2ThKx_{+3
z5f!L5t*wdZ3oak-lvAD*U=&3S{TIM`_ibLZqdQ<oEKG`biS_FZ`i&|i8~HFC8zLB2
zw(wG|5)F|CBQ4}Hf6$r1s?<gxL4<JIN`OoVrIg^mD~T$Ds3qpCDI%@o<oDyapg}%S
z;igNB95V>3?6(gfr^m}^3v5s?4=CLrD!e$D`s8+A;sY}yh*hah5dZB0JdW7I2Q7mo
zl&!zP5IG+s*+4>p+DnYM-No6~;?u%aOhNc`?w9v$LoqMPav0o|<^o!9%YhS|t?j|}
zL>sHFmQvw{d8x`HU^ZxFHZOb$Ce1|BfrA<In*OUvMfj4&&?@*K*eR#YQyjHs#Gm)g
zloo)A5TD{k3rg}ue4?m$Iyr@tB7A1{9+)m0Jz~JDOA7D^6PEx^i*Y3Z04Mmqzw%{W
z5wI_Q*dtjic+dr~SMjp;N)4g@kEULGhU-9x%BEv}kddU_xeE9alT*lcdoVnv8EaIp
zZY;YaDFLC=TdF>sON(N25xyn#h$E-ESmc1uEkstZ<{MZfn%ZqqOyE<hclq_2!b)0n
zp*w_#1`dwFn=a*o65wmrRI`s8)@uE_Ly)s%?G?3uu|AuI<BT}^_(AD*Nxi9Yr{IB4
z2{6ZQxK=X=FMn2gJSD1sb3JkR^b&>L`6bknZCR67EyxAc7I@8iAXifv#6~;}oU4(`
zzhhh%RZ7I2pXq%&VR^47&8K5L#$%#mT<Qv8^<(JAE0^k~?MioM{cX5cgKD2_X^(@Y
zpNH`jOF(sFQkyro-BsUNj0{Z!A*E{m>Uhn@_Ir;B!+GXaomz_GFH;i<x~X?%3auuN
zA2(y7syz9o*4|=@I^CNcCtlQ4P$qzYCeO;z|B%A{yAlcfI=~A^lnQW4ZGXPVui7Z@
zvxx!8-Cb{>cLvb*;2ya1DRc%{e+ZK?Ls>4uiX~F4>t8T$Y}Oxi$tc8wp?xvY!wn{!
zhPy)L%)~a1oLzHfaHb-~HeZ2^b#cW#nnoc!h+Zo45=!)GtlD$%K#Mk;rz(8KbR}`j
z%HI!7IIVG-Ek=6!wR38daI-9HH6O#8PxBdk$nLAi;V#Zoqmcf6XlS2i<;N3IvqDLu
zst}>j-=2jyZ&5-B2;a<U-;B6u_au9Sx`fg}E<vg80BsgTF^xH@xYWucQ&gocg{<?$
zR(Z%c5P|3V0HMk;Jqn(R4E1l_RqS5;H9hn)37b<Bnkh-R?XY#63hskM$xXstR8!}{
zrgWdS(!q1td*8C0hEI~#d0k{8INAq#N!#9s*JCy^J?R=hWnix9;CnIQZ#-u!K|a_?
zGzH<ndW-7JFr5RgHl*<sl5^(^CVh5}Dg7cg6fT6x(}oKmUP3Bmb36|+)3Ll)*>&!!
ztD6X}TF@BL$o|UWr&neNqJV=@ktc#>%6*>ujkT`D9%In6NbYIo)E8YG*yc{bW(~;c
z3J)6TMqxYf>s3W@rNB9qlwDq681A&z?g`8uYYtEk{)556{EQ{J%(qa%4pi%zsnZI2
z&5t@Ye4gNDsFLNL<E@jHN@a+vt8P-XF!X6l0kp}-!8LCT?J-f~%t!k5w3a?Qm-6BI
z13zO@xk9DLAeFD+b;fZ}5eu#tn_O1X>KYBI!U*zprGk~p=f%liK($RJpNCtc>J?6}
zQ57T1pZXw+Zmz8o^!KlfdUYxWH51WbQn$`iHL?&7`g~aYRoV9}G}nfsCY0$Qxs2{$
zhh^j14^V+Rc4jQUzqcW-pIkZtRx$-coox1$AFlZ~K7Bh?AieeM7y76waH+W0O#VW&
zERe#KN2_mHdvWkstW5g`x$F9HU))TD>#PS@)N#7)*MEC^>%QTq;*|n{!@osI|FL)Z
z*D^*<8c2BtGt&RseD=(cZLo=L-cl-W#DTd5o1OvkBfCdC?p>~~AX*ZO$A03xb2j}1
zmKN)1ccBOUo1N_`FP<6yFrM-^{<*>|=WqE(g$3+f8UEN`ib9(F*i8ACfjK8R+K$av
zF^xa+@*3*oID#0Tjw@1L-Wrz*q%`tcWry>Si3PFC=ZVQ@QubwRrS?d@*5y5WA%^er
zt5@l-_6=4y{E1Sc?v_7O5?x%zoMgK#EY(J=R^0)uV7^wLDkfOP2kHnDAM1JE+c_We
z#E)H#U9^uCd->>T#Jr|AdDa&7!YSklWQV3-$dy2T^5if4DoWV?sZlIz#&NkX09c)R
z9pJn5Od+$`Hc(w-g_Ds9lSqp_`b;E8Q_UDfkw1hTFwA+8+nfqur@e2>GPd}#ooGfM
zCsK0;8J=Jj=^)M_CKatqJo_z%BVj0#{LsoCQTudJq=fW!;9mT>VP|Tx%fySDR57Gi
z@k^y=%H4ezZHcU1qfhJ5c;pIiGWei>e_*y^a(JWlB_BP+xM&sKPhS{18<E*krxcWn
zk&5m@MQ5k5@M0G`Zw5tjY$(@jMN5~ST2j|}`d~`4NJyY_>Hx#_X;{#I?=wPqb8P8r
ze|vo`3|>~yLx01&&(ysJSGJ?;eKF}CSG(?5nnT9ww~TN71>YacjkGarp0-;(uOi)5
z;6Aclrn~3?f`k-yEZ$CzTdI`_6g%n|h?co*7Hyqh2)AMoxwKi`+lPngZ5{S5dX$$s
ziZ<xmmCfCkcEp6{9exb_SvA;0P$rX$&Sv{|0z*J<>z&v$k23MS@0-6(P`#4YgyHhy
zcjfrtV(hG5vcrzsj3g3bjYT64qK6lM{Xi43k9v;#7r%X3?9KnDt%?><W)?G^bD;g#
z@gCr)_G*wO>3i*nZR#KYn;bPMR`kLrCWUk9=2p+|lA2f<=3iREl5&6I>OBob(-qp^
zAk?~X!OpTEJLnf8=sG-wD{*cGG~&M@Ar?F>u~X%dPWl(O3*--+nle{E*V%}DMg8y6
z_CKzb!Hf*wKU?OWTa@z@U5?j;Y~!KtM@;4Md;oS39V}^+hKpn{1^0?maeOq(_!zld
zx+}6}6_j=Fz`jkQPl!_{S4?>lGyr(IA*?ND9}ChG2k@f{Pnq(s<J`}Yq>KEKp`wyo
z9sqv>c{K|JjlO0TRu$F@x|6#a{ezzM)g=IFUOu(>KI_k)teE`c0aHlUo~b3ZJ)p@c
zK6%f3<(xP*gHR)cZA}KvQF4xc_P@>yRRL1-w&<vV?vZU35^28!j~XDQa{%Jeav+2g
z?W?!kkMOt{Jf}z_O~>mDNim*2E-fblQLH2O)1~Qa&&tl+d3FzkKgHY?s}ZnyeVxP`
zOpF;0r}eKhYR0v~@1NeZLZ&<f)5H#^zJ164kF)m<q`L3_$0Zd?kqX(zRvjUd9fxC;
z>=7z^mzh0MIp!goP$-+ScakJ~XE(?m*&+Hp-(A(`bKm!McYVIUf4WYt({bMK_iH>~
zkLP13`N*Z<o()s#T)b}-GCeQzKE{2UINSR-An~6sdg0=7WMzha?d!BhKOAU<BAlb1
zqTU}2u2Egfoju^q@?LJ>JXEHRucEW%|N0b~Jwz`OY%937sjW7DazF3RLy07K>qBi<
zP0HvnR4C36T{QIWnm4`1gRUquV((Y?$qK>sqbfu3Qi<DCGaP_yC^;^luADes*kTxj
zt}w?)AdLH_^=Ut1oa7?MBN##6!M)Od7(pdan3R~LocT3^Y)Gmg+bo>BpuiU~xkOH{
zQ3-20Kj7L&2^7HUI1-HW(229m%0wxQK+K%^&emG9Ak~&a<V(A;2AX9T6CJqUXM*-?
zm*-b{-)9jml*$@}q2GM{0#wqb32WShsE#+pQ``28w6~o8!1S$UC9pjDH#^7QHuL}O
zhmuEFVWu8993THj>-7J8p1*%`ltAFX?^D<R+lxmBz>lwg?}1B~{!|kFTZQ;9e<ksW
z-~0QU{J*{E*b6+oIbXl0?0@1u{M+sNm%kE8w*LMyzP8dIfN@?bAHhC=jIq7@4^0Wq
zt4hanoV`(ZSIG54t#7MV3ct)Fs-qol?+zZ&(4WyHTKzVy`RZ}QpC6qgY7dD|BC(dn
z@2`1I_IgV!uQNb#gCM^(k-mS(Vs89IC}`mQ=Ye{M&c4y{=wb<Y#>UvpHZwNh8E+!F
zcO)+%cPIDS3$G{b#iMkN9T(TZHR&B1!r^uO&E1c%+ya_%KavltnhnwvXT<nZG1+~V
zwhxD8aIIHWFwr-z-hxzFjgnjzMsU(iwE^n<UBC_8O<JuX_A#nx<s+~d`RlPw)`hL6
zh@q#5*=o*6juS!#wf<BefT(HetF9*^>59^pjG`{L(j7B;&9plmD%FGsFi+fe&+WPN
zb2<BO_heNLsv-CBXPKQppY|6C!fRfePoRc(SO;Wy@P{#UfoY<X=2ZxlvW5lgIg*vD
zsb}|^!EY8=-6bDi`41s|N^lR4wVoG8DA+(+<k4RQ8h!~}&rqoN)>KzXn~R^%3Xnmg
zvSbg(0kW%@eCj<i<9}%SCZNYLsHtOe93duiADNA8Z&Uy;5ly)>;J_XbLVLa~)@uIv
zc+3y?bXbUrZkw0(e+OhVlIV#g84~0&5AJB86rH!4xYBc+<w^$|w0X^drxjN>SPYq&
zG;63^DzT+4A)^LmT1b+|4>5th%pF!kE<@|@yF0WCURt@g&#giZ=DiVM$CGW00XkK3
zMB>}C$2CY=>yvVXnc#v}c>}#5Gc7$k8EGe{Nx$&#w&kBsRDy4Ukv{gAN%bQp+|hw`
zFo-0jc{b%$v~}o|n=RGI3)9(Ii}qyG?aY;bn$d)hCVowd)hmRUIgpksBFU`<i2W%u
z%aYOgu08f$h;uY-PsLznUiBueTMV4dqzLFP*r(A0AQ)j|j7wUDEjA|aHoyba;eNne
zEkd4vG;1+}QH@`kK{}T65ZagRmI>_mK_nWHToK$DNE3i6C{+>fy}OysaBWo<rQ2zM
zc?$~d76?jx(GP*boeL)9ClQn4=&k+*5`7N>LheteJ1fIZ00=&z8bt1a@R|M7VBxhI
zgw#kw1U0wtzU}c!TWoVY0tvmP>Z?Me?k9B3nTJeb^iwWlI>5I6>9eg$D;aQ1(jhj8
zBS4n>LGTwc+DQaFk=E83c{-8XUm<f{x~M$di^-JoHIMhTD<mpAVGma5$i2f^fy&;a
znkLc4CdJ@sw)4T^CRgJt=xX@4o_%%b{%Yf3Faa*(NHP*JfoSR>Tte-mzU>-~=ISuI
z%I}h9e|M6^C1<tt-wgl%GG*imdRTwT@O&BM2%djJ{JjyTTnUI5{Plcs_3I#Z>0SV;
zytJ&Op&KkK)&O73Q&f8Z_&T>fbhDJ~{BTDfkbTis{Nb&L_t#}A0*YH`u%#y7luK1G
zzMvYstgAz<9IFHpbY?o{PjjG?VJ6gM7>Y{{=)X7KrLu>7^T3`jW)j;vW`fzzD!jjM
zzxRK1SIGaF14Kke&rIy1mgzToz`74R1P9@H(BjPZ-I<5^%pheX?&**zeW~Um0p{yA
zgj5e0lU0uu0UN&k@P+ZPr~7Ib3PuIG@bu38FUj|xk}+9WflgnO99dEnSQej4Qh~3e
zmgdx6vI6l-2`5z@ZXPA@UQJ~jKq#6jWby%*2H$(!*fD3<aI6Upt+}~I)7SWL$~>Xs
zQ(4mbw`o$@-{sif-(Sf^NVa%(J2=%1%uoTEgdJNc_AGN3M;>D5`!Q_c{3#3pYX#+}
zqjzXhx*lEhD;RN`C-Bw$9B^1K#gRCZDn`%#ZP+IG4}-JEO>Zl3gt%(qy>1Ew@AH_G
zsCFbWoH@m0n|;zMmGHy3_0{lqS!|rlOI4umXSO}6zXCF%kmskC2W9$R>!nm5B?wB}
zdQssP#M&z-K~4ay^)v1uq;2-mjF5VRXJl+oszDKeQ`}S-W^8t*Ylc53{6c51XYbmd
z)FeMJRV4}n0g_+U8^=*51)6C?1iKvqgfrXO_gD}n;D<~F#1Oe|hiJnuK*#BL!Z{07
zjPc2{*L^5LY@z#N8!_=*mkncwdh#Rv2}}+yVqT%dSXJgUq3Qfp>0~2#L8S~%uP-<A
z&`uv%F#$aB?5?mXc>(0p=(|Hb&B9Ip4&`(zlPVyv4Dbcno~qEED0*NNh4*%czf&S$
zjpFPKlzS|nHR8d2slgneU3`X4+Bw!W{_v@+;|+TTf~KxweCFQ2c+m>J)66svN}^U@
zu90?b$@y->%(&3CUKhUk$|bC;Kx_~B?!#mrpvgW#io5Wq!Y``(+-D^1pYW_k9DA2m
zYHGmC1AEIy-SiVB)zJ_l1PkH&orT**w#>tQ+(?i(G7o}k&N8HtPwYy?q6~H2$6o49
zF!&y@z-%g!s&u9Em6qR*L3BT`jVN6DP^R>iMU``d!_cDmj)wy2dszj6f$YJ})x^fA
zV}nY%>Z&&U$se&dY`b=myiW0V+sgb73FaXK#Tf~+WDKHZttQk7;OkUVlz5?eE9%7&
zMUn4|m-rv3pPHqggH0{9YzwZLvUA*=f5(Wdb)grWU&^h;)k`Xu9awgl&TN6^=pFl(
zQSpU~SIhAk$!n*sEpJd+T^Ho|kR&EDBUV~a0s27i*E=uWURQBky4~QrsY!E>M<hQ(
zYkUxq4yw>wjLCn$ddA5EyHKc^{CZQoc<Ydr@#XRS=55(+m2t7brj_)@wLsU#s)?k}
zqYt4rwg#C%sFOD|@9~SLZ>4+*HqDLG>tr5F*~0(nfQggLXL&^IQ*A5P%lxBg_pOoq
zaQ?++2yEhBg3fdOVPRQoo8T?{9L|gV_m40dXzaiEuoBksCr2)2{_+Ay4OKcJRTB#b
zDQQ(y#RpI_Qan&8-Wre%>_JSI?h3FHvz|mTF<UWnBi1Ml_!+sA7=>(2IkE(KsX6TW
zAZ!QLkQ$xq?2=#ukqoyn4J?Dl-oZ*RFJ&x!hZ>|v&!nv9%q=vMrly>P(3Au>E9V7`
zS|H@TL(VT)FFHykRrVeojvBZ6_Nx{p!GJQUi3=*4x1kEf!X&1`QD<&&qB-y7siT}J
z)Jhb@Hv*LWdhP^E>lzPU<=J<(j>{o*Dg7*?`5U-SniZ5_-)1`;7cCG^Doj3xDTS=(
zwP*@sc3y>L3m!~9UR*wtyV!J4q1e*zK$XPq*Rd>rCKskz1mU&2d5iqP0X||B2iIMi
z2wjH=@oNsP0xu8C?{^JzKgf==7`}!GCDU|d&Nh(_T^3!c7dlKs$s1^t9kutc(4(gq
zy?K`sjfm<j`k+WfUG={GU12xDpC9S041tF#^#=FYTh9u>qh*f02BKT8uE6u$f%2W~
zYdcL)k3@$hgue{K$|o9d4Uu!LA?m^$i58uSZ{U)Atj;1=R)9smxL_2#+H~x0)**qN
zc8wz~XGmw+H-u+5kQx}_gZaOnl-+t6%f}NQyZmJE;dKw4{8;xWhlkq-+tNB@&1D(P
zMkZ)5-^uqQbbQ1EWhI8oy?v{8=i<p5edvXQyBU7lev#AA_)a_n_S`0u$2UJh&oh8H
zp=T2|q}RHopuZ_9$5e0U;LPkk@4M=GUx1IJ!0P-E9<|u1nIs{Tt97dkGOot9kIy=<
zSCf*QEO&yUI@qiMW-Ivyr=iGuyo<)4v_Wv3@HK^FiFNeDwqMW*npLug8G~jOBO*$f
zU1D)C`RG5T(0R`&jFp@psZ`1J5kV`h(5>qjMJZF1!g1Aooqb{LxHNSg^2=eah+ia2
z0bQm=cs)5!ST{L7Obypk@8uPnuSqTxcC+BlHz(-EYK5dk(nom9*kh8ecCg^WN#h#L
ziH$H~WATmFJpoExon2X?ot?9{^bCM$MD^)l!6?)O<SEo+d#-NzW3cLPED1jVv|x;-
zcT&e)CO+Rnl9EV4F%6Y(a8V?N)Trf#%7ts>i@1|-U5@7Btx%aSA51wAYu{ofWwIxB
zdRb9$b~KHTosadN`eUPpV3LcO2Iq=%7>tXNP(xG$vsLF!<I_aU4UtQj6rUsW(>Yy9
z1URC5=J$%;Qfr?0N(Y{e%;~m}S;E|9>Od(Aa=z;rRcS|=2hyZDr*HY~GkRO)v+n^h
z1rNvNhtwNq-%Sb)g6*Y2S03v65@{vlw+{yf77DbCpTI>uv%c!inDy46x9#F;C7hv<
zloe8vFx*4AvpMHi@2yk33{`K!>@gO*(WV)#IkAoig6vzNI6A&@z~~lu{1)p_eUx)l
z?5hT6>RQC~mE2&((8Q!NK0HlVXOWy{`VF8QjECZ|!6V$+IY(2_QgpMzs6l8}QmYuA
z8q%A0;NQu}Nl$e^iQqSzc3SxN^73CDa}j)Q`s_Z_7kz*XG*Z7lzs!U&0LAfh13T}t
zFSOsq-bRFx!$!mI;Ak9dCha_rx@N&Kd+J_!s9rVHN~PtP+#ka>A!zfyX7Y`YwgAno
zpX)Jhjr8zvIv^%NlSU<d%#%kiL@Z)(`8i{BZ&v6mHo~M2hZAVqpsW)9ydJ+H`TAWx
z_qT0x$1CD8p@_@ikT$-ISyr!o%QD4EmC^-t(s;e6iVOkyw)iA~f5ug>*}p`Ui^k~n
zaf^)Zs~P9hw%>Ca>9)URafzCre1#;{jC0L*fs;zI-nvy`vS8W67`?CZ3}>=hcXeq+
z=`EHS+pXqp3rF^qkJlM4PpR0fm$OERUb0+nG_+ao9qKTro$2iz6^}U<;q4G)_dy$v
zi_(i!L_e#~L`Sj42t@?xEe*ggCe*8H-yQ3c4t}DjJ`H&jO6YWx%1=>CPS3e)pd-R-
zruU3%sQo|;X{rI$>!UVJR8;a2*aRi?b5<(FlxlR@=OdY$N#*k2^{8`Ql|CP)(u+{E
zib{wG0!ahAjAv0ge-L&;5q(6Xn#iH;b?UfE>w|-IHle;T*o#W&LN@%@=ve>Fx}9wh
zg>v98pJ=gKd;Pg~TuQA1Rx8TM@gS_cBAS{VPa3Ni*1X8={sNFYb}HBAlHx62r#CDb
z&SfXJ-qPK&xPTm5OeXE#*EQ{n8@D$b5EnnjTUySMAt8~k9~?4kvZ3RABL*MQe1Seb
zzqMt4PI$m7)hZ<2<8XA6tZ;cwg5F(-tk+ULpQv~DGP)<qCV%5G&g+~oN7;+mA)tSO
zdZ9SSHRm|mqM<_Tg7|<hI^(tOcsO|kR;O7o#z6nUB}82_?1$ERJ*YA`Ab-=P%|*Ta
zz9r8m9liNGA#K@B!Y30cbJtTvMpAQgv={<}T-_GNzYDSS*(B}T<g*<vJLfpn5xboe
z5VgJXX6BE1Jy4LJlky>!{GU)-u`t9|BN`z`BHWEMX#I$|D)g*YFM9I0y7iI!iB>3G
zq98A?W4XbGmsKmA{B|~Ec)ZhM5O<*wi*u5@n<SWTN-w7fN<vR*Rm`bRgf_C8n1zcc
zQkhnnn$M}~T1d0J``E9rhdl1yx?5xowM5#{`WmH~9CiXX(pG8?<?~lJmKkVHLzn-e
zwlkdri3QF|gL1LJPmJ_Bu>kI1A{T+_yRoQp+3m19vC56N^}fG*vFo(^tnNJvXYR8)
z(rE}E;o4#T<k~b_Pwi6QGOp5+)r4iZgmzUg9Piv7Y>eH!k8&V8c`oCrd820_>peQZ
z$g4WY27dQhz?^!vBh?;qyJHt}b(b2oJC*TMeWtgw9TG@&Qbp?wQ=scZ>o#6BZqe2}
zenS~c$7kcLSN%?wEz{CTprX&1@WaIe^y<1}+_N%GvQ-P_Tq%sL#6Ap{&UGeReunA+
zzdfw>pmbn9b2I{bfb8)=DnrRYvW?{3b@j%#-La+NhYF1XJ#1tNtk-n(vUiXt?e_ag
z6Z^F0X?=SY04Ff?-A#q317@t|4OqhKvY7logLX3HS4gja_a6Pl>v+UL`31*6(WRYc
z!6v{tEw0baAH#}@$~#Ctp;D)y?-<Y|Diz|YE)HC++29_>Vp6Upy(<MsiE^5aCzF;6
zFd?}kbDfozi3~$xpDv|&uT42P;hUWW<150{bJ8HgzFzMbzp8W>$9D?xE~W?E>pxQ%
z&sAb7^FvA=kC4vNg(>fu3-|rh2-F82)lvq?ew?7lLkZ*`lz6^~)Gi6fY5{;zOuNY5
zYgw#bu-ROhK{2hZ&v5F=MNhC^K)pVlDRV^hOYOIhCW~<*=s4MvUAkO8&6mF%33>Wp
z+@nkAfmFO8Us77a!uuEAX?Ql7(>rYXOMoQ2YHB{d9(U{sIPwLT_o!t>6g8k(8yKhK
z^}&8T`^nzj#Sq0i6XE==_e!&-3s>HJUrm<?QGYsIns%>+YT}INcm-gE0gjnz(i1`x
zoSDV)7iR9L?G_bmz@alCobRD~T1J<7g;tCY%SbksBzutPJEK7O{_a#ZS%sBv7jxTe
zW(-HcKrUtCf#a}<HOpGTL0`Kg&t!R%(wf;rPEQMoxzYLeo`F@RiIi5~N_!|q*D_OX
zP4VPc%g=DMa!pG%9d%4Q)`^#$B-)_Lydoc<oBJ%U@V-`MyF87urKxIMCBe1%)h@w4
ziX@fZ87_MD4%#mAc2FKn)Y<x5o`xl4R_d>T6-Ozv`fCa4<f{o<nq#7O%0*Pqaab>b
z@7RM*_-v<|lHZW&e>uCe0kj&0S^W#Nmn3o%+Jna#Gu&tp)yQ(S)I*#<TDm)<+ypGT
z(!!2~VMKdvE!=a@BQf1{-XOV-KD^{uaKGccX$!UVIA_5*6xyXoVue_RIgL5QRY7y5
z(TZ6VCFwRZ9L5}8;`V@Gip$LgRAT3p_tAZ#y_1BsKGf1nBzd<D!CypWZdOYBRs~Fr
zTUeypUZw`?g?XSEFju)6zo~Fy>Sn?^MH5V^kdlQ{Y?9eUEn$eHA|=h2agn^EiDy>N
z=7%d#bQIjtgOC+xM0IBwczakY3DqxZd970FDR!nPw`7oeK?g4vqo1|f<h8&*aUR8L
ze`?-mFN#`~mQVgncxffE4<)<O*$JmyDQ|KEJcv&Vq;(UAbWA`s%4&t0y`L@<>O|&y
zQhp0~<mWN9NW80U&8F!P0E~rdRyBes`QEIw@U>83zCB+6Hafypq+1Pnt*X!F76_T<
z$8=P4T`Z@xle0!9iJftp3^c_>Xw7Utn;hE4d<K#}*-$#2*=JHE=eY?^ES=8l{UST5
zVImgNGM}>5slNbuw5IQ}U-}(*BN1<|OlTtRcI+xVJ4)r`KLZ!>?10W#I0~!$q|sM|
z5~zgWeD0o8%O<xuxP<4F{}!8kT=}sjxT};Rev3#Tph<#NjoZl{A%>5vr;doNULn2$
zF1ax_;72hqA!R^ILG^VHeWnrg)(&lkFuo5b_;?5L{{|#IEE}G(+&R|zGpNS-vawL+
z72{jkBtCMAnAfOyJ!~_mKy~9K5JU>oCDpNfEW;^VvP0_Z%%MHxZ^>v5%G-fUeQ1Vt
ziw0`g6XZk8Xt%7!*zVr?M-^l@Z*dS_7lXps3C>ex`fz-CnE_z_+Tjp6+eXv1eimz#
z3u*&({Ae7dUC4GAvCr^<Mr^NwG8Mtq#333=WeZDy@zZ#{wP^#4-%?ZTiz3b%wvtOo
z8XW)=lw4d%Y~t0~VhZ^zP5O|W+da&CATs@YLi$uFXM$+5u&VaEFlUj%b3Y$V^1nQq
zPEDd(WTU;}HuF^>%_RAD|BVAIsyQfv=o_5}{(D0WIqPT=9}CB7JN=%WO5zBZZ?atE
z|Mqio1=Sj3bT@eRMv7Kw57=2>gNEG$MeN&s=>?uqjG|9&_3%Q}LxDbG^07SSHhq`!
zGUt#%y+4GW3b+y#J7+Hwt5`ta<?ZJetDu5^Jh6nnz^+Vt)xo=%p58{f*nxIrLP~jg
z(S90uXs1Ma@1;3{-3vP&qh2zy@5yT7qDQ)QLV7@mO|TaK>(4L8DaO0tqct`h@maB=
z%SJ`-_SIIrF%{9D!HtQz>6c_qWH-n2=C?s7qW-Jqsic$VW*qiB8(gR>DZd_x$k(3E
zAm<6cbnGYV4Ud?v2a4HmQ@f(*X9=yAGZNl*r$yVei<cdvUhylhb-R*12ObP>Q%8CS
zM|##>Clk+?OAdZ?D1{1aoKSOQOe_l+{b@F#7EZQi$1ASWtTd!@u`7?zZ3LyqUPhuM
zf_s4n6{goDk2nSyhRn%X7~8(f%I_?m0L&-At1L`EyJ_71_VQGmz=qZWX9^>n2-X*P
zCM0GJi?<)zIlL7qKcqhph2EK)u8Rn2rni-*X#b>3W#wp~(bwJhkTZ*M<THbfN%g17
zpmcL(Fg@{L>*_L#DIF7`Ua3sAx-@NE{Z9wolSDZU&Njb0e^XD1!WO@N>}Pq!W}Kn>
zI+chWFCHB#^a&^EZ>&njr<8lg=X2+%D)c_O;SizgOf8}7zV&*1+BX7gb!td!^JQuU
z8@?z?G|}>NURJUQH`=A=G5PHod0Y`$XBr2okH*KZ4RA{Be{K|keS47ksaWS>yWk6v
zJUGuJOMzVQC$YKv>YE*G=2+s#EiWM7EG{N#q4eoDtcaN5)l0mb@B1|0mvV6SW`Bpl
zt_U$4?#q$U)9~MHSI!5M&_RhuY<&gaxMU5My>Q9ZC_4f+bxUFLmh^@14BVdp1n{hH
z;VCR{?!~B{BH_})!>G$Fx(r|*Z-F?e1n4qc;%%@5Fo|rlUG0WhdeQbEj#VSOAT;mx
zz~)f`wPoVW=Xnl`MH~b6Iao)f9Qs84lg9}xl(XTvcv}8=E>pwkl%LOKxo7QdY9Biu
z6<eltDhO+T3CW5QeZ@+k+0|+GPm>Y=h<izGKLa;>(M*z&=e>m3m){}ndD(uHVLzDZ
z9Pi8_H-M(Ws9-LrKd7YXmRIAaK{Rhm>3QhJ9ZqXGr6eh#rB@y%<$%(59pY~ZTYvBU
zy&gZhX?{nJDuVrzLZisHelcKTb0|~k>BvZFZiCxz^X}9;u&CB62{Bq;Q=_2ho?@&E
zO<y}L)Aez%(Km^mk9o}{um{&DO^=DnI-OhexFAnff<*`Q>}(R!Z%S=1CK9g~+xjqV
zBE8;>4Nhff@NJaFEJ;iHNdpYqvM=R=X3#^B5ox`4;WKDqR4%_NoVfG?x?wDXD2gs~
z9+kbPhMF!!jcp>4f}tFiT3*+RFUSm$UtR=${wnk@BlFtJn8Vz%hRBu6cM+k}TGV0g
zDYADdiD0vQWA*%d(}qVZ`*_>CdD9U`qZH$#kNFs<N$g@a<%^y{Y6UB5N2BlH&8>1^
zX|ttZCf@iN(X=CMZRV{*^uk5Ut6h`HJ9B0)SK%&L^BW|c5E&m;M3=MnQj#e~?YvJ2
zCG{CD{K7)Yr9$kQJSpik@Q|7C19+wl4yVer%;)SNZPy%`ft_fZ_wf6r2AM4lig6B`
z&J&Z;EDj06y82dSW{5XKEENWwQ}w-U{~J2!IZ^BhD{)Rc$L+4dF&1m)yIQZ;TGIQV
z+g04|&<54m%k8G(L<fps8jchi&B{fbPo)pGU`@S^B)oc=dd8R@BW`BHTA^oltS>XC
zowx^af1Y+IsUFmgDSiS-S@sHd<$=QW<-s4r^Xzn!dl7Cea1!0lyKtZ{3IrF}B2#4&
zy5)-WvLgdYjjvl4o7T=IP`Ujn@CDgCWJ#1>lI(hmXaC+lR>=t&o;Qi!iP}8UIMMmK
zg-!e_cm%YX2(2(a3)um)kK}S~Q*>>dqQHbo@8D|NK5>V~O?sD~ffH??-c#I|9w{H5
z!JJ{f6$sT%B9K1V%7QDt@cuO&t++>7xuiKmz*|wn`0}&n=uT><vM(m+3voe~CE!vO
zi=`|=Vj<O!*dS@Kt)K68364I5E-w;uI`A~XR<~q?<#2qu@R^|r75^6R>*w+J%^65q
z_3jsdjG|ueK@l)QWW2x55k|XaX{<yxPTWZ-k3GJ2tn+@seu1wq*}%O!(D~&)Dh%U?
zDHYn(+$z)0V83!!8E=KAs_(COL^UU7|IpR4SEwbGeh^xvRAUR_8Nm{G5A&qK^AjXW
z3@#shsi|U24d)8CT-S(ocx8bsQw=SHek4SLTA9kZWuGDv-rb^xPB`epD~ksG40Q3)
z=a3dEg+@8CCE9Z$TY&1OVFDI?PI*XYaI!9j9zInCTe1XPYXK_ZRB>eix_1s2>!w~b
zi!POdp$TRv?q=0wl>6h1I<b$-;ra-qT+cEUm9<{<ZerSK%z(=#XX(E16n~2a;|5pN
zhn8wPaA2zU{KS<tCT{X@`7?-q0&3v6%h})pb2en<V}sZ>Kr<NnqG{;l3pNz#-k|@m
z`(d{&76F><uJPryuzj^puB=ke4HO)L)u^V73K)0SN(N%y*XmureD?N_F33h~%w|b}
zW_gPV+HC;wI)Y;nS0y$xzqdTG`0W!V5fEh=8n`>=Y4h|?PM5Z-rK}uDz2hIGH^BAe
zo#v!-wpB=w>HeGeQJUF(R)QNfink}a9E-X?_iC&0Kg}tpFPDI*Hqn(v-18G=Rz{g)
z0;5Ry=)?m{3mzMzou&q7=*Z!>hgc$_LqC$WAH;jmY*1~igDIRTG_I{h{rcOfuTZVr
zbDg&+vjK&cAin8D`@Yb}x9B#q(&f+3+!!r0w?#jBv-6U_@yR*FR&mVL6+G|<ApLyp
zMa03a{2=>@x{Pv3Dgn2dn-0DMdf&J!39T;7?r)#Bd=iDs$iQ`E2NEyt!smqMRbhj4
z^Yz+FkN}!Kv!sFHgQD9m(R>E5Fkn5DKW^3zn}})t({7HqT3JUuYNLXwiOkNoO0}sh
z14q2R?3GqpHow9w^em(f-%KnEObW9`cRGJ+?c6V~`0tFuW@^Q-#MAcK-@INNb77o&
zJQxy83kIj3lfAvOviXtj2Fo_{P%7`e$U8!1P^!|T=(1-TM|EB{7HylA)A8O}Ywf0q
zp`V$)CoL<@Cmkpyc(LJ&oJ&dlQN@1fAKNkvoE~fpn0o;V)=vjq#i=^?&sdZ&Jz>fK
zNbU4FOkfp`9|$^Ogcdhz=pd8krQp#0aewux&ijWh+*^`7ZInY{M7|aVM%c@Irq1G%
zx5US$RgRIBANqJG^WJmJQK^j8sVjD*1sny3?P>N|9E*F-^1bG61qEI-UhiSEw-CO@
z-B={G{n-bhJZbmvxG+hYcXKBJov<RdeWIZ2!4Dh_a<XiVy^2J9M7^QMjL+b6sEvY;
zNlBDp!==c%D2yy5f=?aXVZ2~(+~~(Z!CP+-YIJ%tm0?1tYZUw?cc>?uguk{F$F3lX
zkxLJScqhcAj(+>lto!nzWEJMoyxepKL8H^k18S#kg!@<INYi(uLG)DAhIvVKZ}}xL
zoE-x*Kao}g?<u*aob()eTr#mp)RW%wkr`dTxe)6hOyu!2)WYLA15+~NOP~959LIOf
zWeJQgBiJqPivp|9*K}TJ9}Ejyxwl4G4i*y<<%^sxihVSF>mC;mWgqh_G>c8&hhMwn
zerN5@h4xeJbU+Z&+AWw=s~OcCEj-~?G$?q~-6XN-o&aTBsqWjY&X&TmURN1+<lvdC
zJB`)EYK^AK<Y_&7#KTbC3%%t;?d`ar9}>3QZ_)?GZ-$<7GcY}kaOM6^7xeRo9i<4h
z^6FUb=wD3h)uV(M@|9x_OJlruXzjscBP@8;$6WGQX;X$_Z-sKbq;mdA2~`#!<~<bh
z0i4>s^=Xz)Us%rDQ_M>-VFX!p=~Omxq}tiHM32f~NjoFB*`(T7?l7sWW&{yyH=ZR{
z(4!6a-(A99bjgZ58)z=AK3eZh&h{~yr_mYHm34aT4IaPyxs<1<L(0aOZJY|lfTCx5
z#Cza+ormv1m8qHIM09Of$uSROg~3;h_9wy{`>{7_m|vN!9P$m6^Nf3_wJ6+pt9V>>
zg3{)?Ez?n6Mhf!jn|(ENCma2Vvt^Fwy?vOWImNMKgXzC63S#9axD>hSVKrPVM$AZb
z=6I;t@CPz2V;xqWhP&I5jmh-k>(^yyGxkzI5wQJ0qRf<nZnl{1MSdzF(>p``CaSH7
zoBUF=z+}!ENgb%soD`f3OVq(~r#|o8KfyC=TC7?>Y-^Hx<B-_Aa~uIqyFn~+#JebA
zTbzf$`LD5Zx=h(MS2K2!I4kPO;Eu)xj%6!nXK|-qHwL?FwfENow`i`-7>|ryP@k>J
zS<o~nKd(bMd%VHl&ai4G$~5_PZp*1bwff<<jqCxhD?dg<z7`GT0>y;nt~lEkBj4%w
zI?F?a4y%720ZfD;2-lF4m!9WeMv7-F0pSOvDb8i^(nn4%&@d`V+Q)GSXLJlz8*WG!
zC(P(tG`T8!=D2T?G{I7``T()NT+#mk&VN7Xt`GsPBy&Zb{$Kug2bMUDFSck*he$eQ
zBD_oW$s{?FaIBc=BL8_8%DsNru4txf+%NO=s8UQ_Ai_B-VANHIBSt;Jb5anJ#*(`C
z6s!NlG5(&G_2?=kTx704Hsojb@aJN@AoBqdvqIDpb+YBkD`1l`W?UQGOAHv~UfHd$
z9v7w_`3Vc?+wSVR@eo!g*1{P{I;(Xn;M~9Sh`T2s8OyrV;Ln%~<`El^^sd&WNBsT!
zK76N?{89;6umbH51{bt-Z35V$p8<{S-TuRL+Joum|K-O0=fv*dh8vuFv}g10-}l9v
zNG>}OUvQyUU2OpPF*XO$WslZo-zh)*``h+QGl+R~1-|&f*O^WK@V?*w;NM<cUk1-z
zAy=W<e;OhGc}ai$<O(NT*u78cCVzin|Lvds{`px~Sm+*#K6Lu+X!O?z_&@*28R<h*
zD`l6(_s28hfB(2xQJ_g*W@xMUWu^1GIR8JsIHn*;c>S@EM%*8}jlU>||M{Ux4Dbj!
zj+p8H@@E&0ss&1nEu1(DfL{4}{~_v*%y+7vksI96O=8|UH_g&tvIEE`E@bb4!`78*
zSO59pKcYhBS<nG3{lDhfln7BnFpbgqQI}Mk;$BTb<HoR>Zo9RjD?e<!#g-dcqE)%G
zV_esAZ+d21U%j^d1>xTA0z5xZA3!+|g%>dcu)_-xpVqV?Ksh%ASj8${=og!cLpEW9
zXEcxBezJlmF$TgfGL>I2N_bTP_fC;*zqJ2hfraArCg|K452qt`fbRa5FNlknZvl=#
zV|dmCaq@=gmEOub_<M}}mlq>Z@Z=YS`7C38$qyty{2N`?{DTrbH{ti>S|G=mB5>jk
zm|KU2KuqF$Q~@jBWx!^>fA{sAqgU;OU=>4P%w)eLzG&Y-V|IuX7mgXNf+9?6|D2Jz
zoAvEV%U01})LT1**kuDaUI`#O6-BxEFD|x;!bw)DXa~N5^>CS;ApET)XrKHS#8MZ5
zM&}u__X_*h0`uMm?p_(fnM`!>J=`;BO?#h=))c_lJDyM&BOVt6NbZeqw$mtc>w4;n
zZd&}`uS&nBas@w3!4X%-o?n?zQ%t^yu>(;pxPuQuidOH-LdRI*NXoBsh3!&~%|x}J
z`tOj%GuR-kMjNQZ&q$OV@kAVatwGsMoM{3P>cs$^ND$S6vn?M=7IFNClv>UNi>Gm0
z<WWf@y?MI)0R&$W2J8oVu-uj%uK;89T;xVDH%$PbNSQJYOe0g|C}Rz-kmgj)unt=w
z<l^~!?;WekH#0t;NSK1bN2hJQFnOm%6G8(&@pSl}ylSzRf*@eIX|UE~%t0$kbE{Ls
zrqAggZa_2)guKuB*+07{lFUR&ZZlk!(C$LI7e!!vUK%E>5kUM9oZ%4<npM|5U$7Ee
z;fOyA1%)M?SZpb$@fk**Vlwv1l~*g8<K?qDtJen=39m<yV2A|khXBP$Pja&)_8~Us
zj!3SBY<d&GDWCmtnTNiu((eAR$AipY!NJ+`n)i-V45<&adldIeQvDednBs%EUwNn+
z{lByfBMERd-$UAEX>E>Kga?d1MG?wCC7_*Pno4+mKEBAzo}9^T6cu#3>j~j?3Y)3J
z@aIk}=!&iWb8Y$GKXGJ4epkD82HVf2x`W&o$X5e$1MxfuPe1cnde^xO^cU(igbk`1
zw;XrwY1#_}j+h4k_W9@Nt8GugIyF>t;EP>a#n=|$OEri(p28*%P<=r-E)KztO;2&}
zWL{AKrTFewBSearN(PXk%xat<B0gS9>UjXmLY6K+#fRVQNZsR}cRmO6x0%ByC_j4$
zYUaU*irgJ|@h#30p28MB-V>PJIKx0LgyLHa@q1z@T`^#Sts~UOlokXKK!}(^_bKI-
zx51p9Wrk@VyNw4-P1MmY80X{;cX?&beHso;>k4lACecdj^Bo+WQ+SFt5%B}0Fu!<d
z2t~Sp>eOY4oz($0;kDz@EgwM?A%jWZvv+?#Oc{oiz~{MI;Mx6Jm0Lx%s10VWMrsiv
zc1I?8Uv}2CXKP#bQy%Tfdd&rV&LBkA3gepYQNPlCIm|8~F!mLId5#4Vafd~_cFj|{
zd2xxn3=xV1rwwy6jhlq3U|n2?W%Q(#DXd0_^E~fD0Fq@;B&(4W<7s-y6J*4OHY_$g
zvusvG?Pt5JOF0Pm`u-Qsv%w0hpe5#AmJDL6VpU2cV)ersqBS2j^-_yvgg()2QU9SJ
zz9xqc2)lFJZ|N%}A<)l`AWDD5GsWtaW1VvwdnO2GMce6B36jA{12Jxn+={NgLH2o@
zsJ%2rG2Eu{pieNx><JalIJE)YSE9LE@rhb8>!}9YMaqA!@HQx5=h<#h7yB#2VTy%F
z&H?u=3bY154SKx}gehya5iMh1pv5;VRIOZg2R^sd3CSu&t^pv`SRpaUVh|nS49nB~
z4wcdZ_S?`(ia`*GA;#S={^TFa1dkHhNS~Are55>^|H3OmYpUM@=Z(ZVAnRPs*32nH
zc;<JY?noD%Xc!d&yks4R@i*d(SVDhY8Y9~b@w#&4&OyRAVa(Eqy(vFYk;esK=*@#H
zAaG3X{Wow^xU1whb+>4#J9)Qf7l^WhFezvXC{+)j-XQ2z9c8eba?r7T{4+kii?FPk
zAAIM}OUSWY2LYtk(8IqM-cZ7FRF<ERNs47<22>{pmkyu!X6GsNAjJ)O*@dr9Ts@6X
z>DUM!f+=VX!UM940Zrvzf_aGL@hwUFTn$)7sig_ikdS-X#NM=Mpa|3=_`Pm#QBn1P
zYj2lK_QT<s9hG{QTN3&(5S)R?cQ^KJ^%wX$L)wMf{Cz8Hf$l!ri%?$Xb{#C#eAm)O
zjepBVP*TySPG~|WC(_Q3E56G+5hV~Huu}Pc$99;ic-$^+7*W{lS0j?@BPNpU(;4UP
zLa|)yRooMN`aD$l)ZT+{JRZ@DcE?i9N9B3!K70B_^hU0k?e`h27X@HDd~V{rLxM!?
zGgIdB0^1{i7eB2HNoxQ){m#al@>Tj9x3V&y^3{I<t}56kF4IKvT7KTHg+L5j2=@a&
zWtaP?1(sWyC&nEXE+y@ii9rqd5H#uu!#$XPpok`6j^O=7Q^$Xt*BjZ2K~7R*d0saT
z8J3n~MZX%Te_tFudWW2@to$|Ue-=dZL}58$uL|zS-TuKYRE-99Aw9oo@H{*$iZels
zMPQrtlAH`@eBSO829cx81BaD2bnL)V8z+V=g$$E%6>^PI$%vUS5>l4ZhnYXMA&f(m
zzFE<p8|LYW&IO1~7$h2Gn;)Ra`1&}R*clUB0SVmJ;5kOZT%b#fuK8*RV__&|CruLL
zD5;QgM2>saZ|UO6Mj%|5vk}$8iY1YT19n0850CNsyX1VJD5O4kS8SZWjLCQb)FurA
z;Wp&QsQJZN+{i;h8}$%ZWI1O!0Zd>TedB!$t0>4arSn=!pA2rx?vrd-0eNBps0%q$
zZt>=qF?_0rR;@V@^CFpX)<8}a1Vec_DxLX?Tx;q08w(5&qrgmQQ?=CaGCac%GlC+e
zvqwE>%1}u~p>7n<GV7y?uOJ$WFwOv54IOH#`sOxHSQ$}6T|dSp(|K_H=)GSrYkXmW
zl9d;OO1W>3M~U0LC`c|h$StZcb#8D`LhpIjTg=`bQM{ay8cOO)w)6lOgxO`8q1iZ;
zSr8E0Al4ZFW+$14SQP|0mxDPvx2To-(0xiNFHux=`vQUma1~iHXp^dP<?ty^kW)B{
z=Ew5ThqqKLZ~6`6Iuwi>5WYL}LibHnTAdN3lLm?}9lFoq;0Ckh(8uShS$ErkKf`!b
z*y@y_!}_5+6jtcY&986v5Rb|EMm>pI+$-lf>@SxnS{imdU{y=Ud;=Qfmc1z-K{+xt
zi5jZipidgf;T1hkeWZr*c{J~vSO?r3;`rEWUIM9JsO?Tp>3=)V|NFVA;}y^n!~Yj8
z@gQ;8*C;3v`#vU>9SYNHkwe(P_Tf2mqA~9u9NhO-jhH<~tZDuD&dbysHmadP^;rs+
zE@<9`eaqHdmb0~4rz(Egy}tiI%!vos6ic<-^l?RW8EZ_{;9Rp{nKqaXD>ATg4yn62
z#qpJgg(LBpT_xXWsYFunr*ex_-C&D1q8j@c{@Jh+8>5J}t9QoRM(KT#LtprujoU~%
zAr!jN==N47ef+4>0B@2Ze|c@BUZY~qwCeSNT}wv237qh$*(~SNLwbGk=!G;8%)EJ<
zs!(#FMdr05d^AVQy)I^dx(C!_Db3KCOwdb_N1L<8@I^FVG>N&T!dC7PsVAg>_F^w9
z(xMi@Kk6O5ac}XaiNW#SSj|N|s^QhX%yEk=-9NdO-2(XN6AnW%pNc?hQaW(WN;=ib
z)GQdAYLn3$6Kc(4n0w>Qr&0Ttr$!)k{4L9yDE;{D)_p@jmZW8)%6M>4&18G?3wS@W
zm-i<k_SCF{kvuteIUDG{2Hm=?#XDDb!{Qfd_46|_vXdpyr{0{L?cBQKpMLsgw*<8t
z>u~+Ib-6;5?T?>T$c-+o5yu2fUaI+W+Rpc^YftZWgIOKj`i(s?5~)1sxygjHI`0A5
z#T(GPCn*_m9;#oFxaM+#!MXoNz>@uAu@~8Mp4lr|IIl<nImW+Vz<yQ#Zen|O*(!e@
zUL|>$Ix^(y@cqO-a&&SkQLj@Q?}hmk36zt5Vm3S}QNV&`8fNnO1~t$^5j&HTOJvnu
z0&phFR<K~$gLaLqIb6Jld2eP*>W!$G<;GwOaHYw2^J4$Z)c&~{)>j5k>}VquKI=FN
z>qWarK|rI7uhA;cUZr`OBrY_*4<D;JXfS(;MJfavsL1e|x1LdXuU>W&`F+IP=DGow
zR*z_=FMW#@-YaTm4bO`thC>tfZ7P4sRd+2CpUwLM!uiXSCCfFv+o`=v*Dpyvl9|EC
z=4b$s6X7?en8M@S9h3>dp0a5yO=npQ(zX))WIlX8$+l#HP3*Nz%1wsiyg8x1+xQ13
zr5pv8L`3>iKi=!<8z8*yEM`#oX)}Vvhv9}GT&9OZ)2Mq|Z4sRB*p-(IpxFi8d({}x
z2ez~{{N7B4A3^sd#ngVYeUxg+^O1<0l}yo|JsqjaPc8Y_tEdk`2KKYIFTI{NrUu75
z@f9Imm;Z=9_!?j(!LIjvwbsw3gzu>2qfYe)c!}MzoiF|p!K(p3cy71T-Xl6UmK{{)
zIr#xB=|J1*G)~rg92QpciIdC8Ym4tCx(FjQyI2@yD&z(SY3{9tGhFC|lN^1fg?zHm
z^;(~)(u1dgSakx4^7)d{w;zAU1YIa2jP085Iugp!%RDAUYUJX6_oZNZ1S#0?#{?I}
zFbHH9eTSYBKUF+xo@bi3@Dt&+wAZYk4Qo0UPCM~^vpn7<^PsI`U)k4eC%Pe8`8`U)
zmRlE_+7U&eK{4+rLvhv@J13GBs9b4QIB)HjXn3T_%=Y{s??m{Cp-TyuxQ$fpD^6d{
zuDCof+Em4zbuo9CQ6~66rjS^cNt%^uThFPnhLE5&7H`I0k(y3{LjH!-PQjj8W%uns
zZ7XfO{B1Ajq-k7YOStK$Xfna)m^077d&U_l2^)4WgDb!qIqU5*;68D9x<me|!Kr<>
z21s@vhp%yql)@x^#%hAj>;{qPmmD39YaitYb&{fVPU?3NN7U^?F`OOOCYn8$F0o^w
z=RZl*zUjgBX`5gpt2uG03LGrLUO0y@2AS{W@(~lZNjsk3C)l^=YeYju-^K&YNezmf
zj6UaIei8o#Uz(~0aA|$?xkme6lX#2T*A(*rcJtQ=`tnUX#Kr%t0APM`SlL2E1SmQf
z2DjDJNhVA4tb1ezxqR3W(UE7atf^mbjHFuKi<J{0mz>$@7;RcBbf8cYwNG*-l3bZ`
z8kqR-A(PuCM4*p2v$cCNdfntExt{3JD@Bx{jee_syDsXY(jH5VS0_|_g!KuGzX5$C
zF2_RP8U-!FK}bl~&T~B^a+1O_Y)sT7#nrMSjD1$4<ts|b%@{)qeVl4(+QIX{C?sF=
z+FeQZ?->|<`<qeNCsuTds1|O=$YFu9!h!8n+|n0>V)#b*UaXq2g>ma#&XRc=S?+o@
z@gDM*eW?*~=~FhP7vz)}NFV|jf{z~hm?M0Dcz=;zK9h%A)0fMX1eLE=I4B8R>4hE5
z#ps+ta5ab!(b(^?<gd(05l;^2=YOes4-(xLwbleIowb8`_WJ`BenpxXT5U7(WF7ac
zC{UqtPdMZ*vdO@<yQPvlKgj+sF#3ONyY7g%zS<}|=`VTeDlcJcAS>}nVtU+7IU<ZK
zEkIbHfs&?JYU^H<t$&(ghiZ1^)*ka;fZBCX6Kp^eAR&llc@(ITl%H_v--2=W#a`w-
z1(z5m2iTQjuO}@FVZ?-76dK5Tp^j341#JgOa62Po(3D5U$XbHfRj0)QlIBl@tHP~L
zS_IES)X|~oltLDQqn=@M9%~qK%u{G5z4Ka*NbZLur3k0g87mgTN9kPx%8AsS8KU%d
zy<6Pp^NP9@keo0LTiY%EoH%KYB$pIKv^?J@EplNOXpZ^r)>_0qxoz1M^p=KZ{2N$`
zv6sU2bQg(6jUz6_o4d}doz&qMu@XILAw8euXst{^$i#Qf>QLvnux?nN_7N_G4cOQN
z#Z{5r_jjjqlo7^cI<e2lXJmPO0r^y)T($8B+8O(d2QL>HkgDtz)Au+#^Dh5db$4qw
z*${XvTdeh$F8^FsA0OZIEW=d;BQbM%|NOE4GLQa}15bGpx#_DKIl=SzvHNXt?!nJQ
zi=MpV@3v^DyPRS6wtHAZY?RICFIt>ejA+!gUL)O+DOJ7YDVMjYC)w~BYQ@6J#WCN8
zwCwjho7qJPx@~<#gxp%WP2E?Xf}1NFP%E6@-U}KJ_iPGH9RYw<3H_Z7&!Anx^Zlhs
z(fD-mN`k_onDF)0CFu2Y*^GsxI|t7v<vq8qY`)KnHlf||YqoxydYB!oJ<GG_3WL~H
z@)bi&hO#}sB4IXf+#DM%E1OGPb(_W?DBC}OX1zWB?L+v!ZV1~7)3cXnc$vBIM1K`H
zgy+-{>%@_<{&F}nu*IkkiRESN4UE;2DSi@#%(7b~=lcR6PyBHPI%;NmDt;httg}-v
z^5jTe0;=z%vg^s&BOA&;Xn^4g5$RRA?iE<&OAMR5eT;ZXosE8LGnQG6+pfE1kyM^^
zp!!?xS<=xOtg>e;V`WrY76Jc-t`E~HJIe{#H4=aZM_@B9p?zcKd1khKzDwVk3QDiy
z$@aGQ;!Gh!7o?L3_CGQ${HQVOF0X=AEj@ES&i@;F`p?&^3WR;9+@D<jsRfs0WbMGo
zErW)ZL)kkXxM*RwPXKU4o@Ghj-Q<jU0Iiq2r{DYu?7_IT_uJ#@D67eKG3Wt3z-Bbv
zv|QN*7s&u;m#p^%o(vyPj|aKgKfo@_6Yq<VCcr32%3{D{HqgXI#&DU_(gON6a)`iC
zYS}vf9IA&&v8!E%O8;CjN9wIPfc2uF5>;g9?4@MB@v-wYLf-S!aJPva29Rwft6JR2
z(ficKfp_LkqT2qVxsEwn!-{Kx*!KzyWVuf}uDaS{;Z#kB4=s6IFyHEd2FCyqXKTbf
zL3-6j`H*SIMZIRK1y^MjI=%#~UMr|#yuq{P?xW`TJ{rR7@HC7GnXEeF%#z}`@7;V<
z`?j0;zy@%s(t;zfPD%)Y50P&^!1|mSiuwKp&hf7Y5uN|!`8V4sHAilJ=QTi%8VOcP
zd1Nw0zvmS@m4iO^&k&6}G=ciX8ZPGklHK1YW-=f@!}%RsR0FJ?GHcrHY#FKMw7tq!
z31L#)CeHyM`{CHhoQ`n79+x<+UPt9Kiy_7~(|~SsCdxX}qn2HpTFJf1Ng*Mtr=l}l
zf#dwr{YF=Z7GCTflASiHGV(pQfZ?PJgxO1O5d5@U$@1VRMU*q3apTj0lm$oky(F_}
z?Edr=kec&tNL`eAWR>R>&b-lnbqSa<Xxi-4giNOk+%DjWeTapkR31*fL8>;;`<)2K
zmF*(IKq`A6`Dv~Qm_0;lDH56|w&IhOtG>VH@uIL{skr9#9!ZvSb@Nzv%w-PtT6BLO
zWTmNazeVQman@YOmXvViZpvCKDgd_NwL^BjYEk>U6)xlaF^QP?0T3Rm<<^|#C={o3
zwJK~CJmuL|@=mxkYTW^CV{2ki?6&i56AGOkwR#sY@;FZ68*yO;GWcRfIEJl;OK}#B
zfpD7l#^r5_f2oNQD9l(ge?k_uW3lR>Vnw!g_Hd3FW@*gJ^w9FXr7}40u~e8A;~oLB
zrxDCC21j`1D8T#DwSy?)5IMKDH}0f17i~#}fm6(+^7?7^t_8?}iuQsa4i4oAI>Ti!
z7^e%rju>#CgJbj4SBFJ%TKNOb{L+ms18i?2h2c1Av+H{Cn10e;%vpWoa2eF{O|Kl6
z<l@J@i+gi6L=(a&0?6Z%sd9Eio2O`14vJ(Kb?rx?3QKN+38)3ATmv?u-Y0(u2y5aW
zQfn=xm=229%3Vz@?9Vdzb6vb}={jlr<o5$5|8ssh!u&x(E&)r*M}=zA^?*)HzpCOH
z6R=aV7;wfXnV_j6ZB`t?wBNSn9`99)Cs8ddbpr_T3rQX=_qlPiYU%z6@cSpY+~yI+
z6;(x38_%a#FnO%dZE?-FC+gx|x%D0FUWyVvC|5t~jn0V+^6AtZ?KY*74lMJ?JUoN!
z^ySuuN$WhZN&pG(H;eg-G;}8Tg35_4k^=ktV|W;r-XQH0k%h8U*LsYDMHdoz>97aY
zw#MndH1PV+%CX=dzMoVMH(uq{Y+0_Ddm>!8SVhX0Hvnh|o)O!6Yy&vsL>G2-XkSsk
z*Rb820gw(GGf*^%#;XN0ih%oz%*vGeMBU@^IWq<_?m=N-c*QqcPpDpAwF7fWOMXV7
z_piAf0s>2)e7a*`EdpBk*6!`-5`N{w&?JxLe-g5LKR+rsCLJI8e6=WgLMqyIYdlsf
z*=sAaoZJ*$(hmeLBS59%bHVjtlbh6au}-_lue=KIr$Hq6F?jF0jmQ$<Bw9o7rxdzY
zBPRg=5`5=K!rCy%mfAmA7g(NS39ct<1J9um$lnlDb4(w>-b$YIZu^xK2{WJ*Uwhn^
z7n~59^*Q)(f-B!hN&kvK%)=HLu_354T2e&@^&DC?>aNts56rj#Z&1}g-bHO(cPyMb
z#HsPK$;wph(j>*@R<-ShidTahfWT6Z)fM^)FkcfTy&l{N`mURP2ac44oF5fx=-zkQ
zA$(Bup0(%^?FiRAdN3Ax-UrCN_)^uQJk5963(Mci;liweg@fCDhF-FE4nR)`I^1`n
z%Y(*XN%<7j!{yVii&F8=WnVqq?)^OZVV_xpkJ$yQ<n$E$<leQ<isA1M=p7DO8gp$}
z30pkwq4ja|)nNNp<)i%aqRc{tLYwQJOL74d6y9T-#lUzUfb2lpc5?6zn2&k!?Fo{K
zyi|=_JQ&>UJY+-lzFO#Mqie6Vgt`CkVgI-7VafN1Dcxd7x!YguA%KQ%wS+_x5J(cp
zV=k(HNfjrgk&wEbmIJAJ&fr56Nk7B2ua)KPz0)5kzRtm<KeTyd4_=~vukvv#Ij}h0
z-H?{#%zor^<Xdbg^?3OHkad^4)#Qi6-NJ5%d5}|yZ5e#~(#8hLk$2?tq%apqS<l6~
z88fb!yWT&G_1!xpcJykaVbNZ@#|17~HDT&upUK<8&Xd0pmC`&rE~@<8oS~EFvdMSY
zO^?}+LdM2X!t0vpSf9X(uff!fUZkp%0Q4MP^o9V|IBZAY1_KQ@tE4#Wc-z;2@>BK{
zTN&hf$LBn>1n=kg+*0H6xh1|#f-=PJ+7KH~LETH`WN%wj3!)R>a61#UT@}_ZPz$e4
zgF<z2c+)I;(C5~)4-z;~ckIvele_N|pcFX#9GrbLtB(<;W4o<c1~eIzUWh6f%B4na
z>+TFURaveLZmwLdmdMENU3zqwh`oVQKB|}};XDIZbFA8M0`wgWhL3o(#i06HG~cA+
z1jn~~rnbRKF)sw6+o`CK-H2Ygu74v(=O%e&cr(SV6+Y8G+qRD$4V0bn?j+BtJENZ)
zSC5I0@1}gwldO@NcvAG&`piy%sxFjuKh~>#9uaq+rf7nI28*k3+l}S}+q5?E>AInl
zg!Tdj^i|p$+P5iyQs?SbS5|Or(~}^M!D0m(vv@P^_Lp<6uL_UJ({mkj&HFI|y_lsZ
z4i5X1(Vj?%<p)OF9gmAGExR|Q!dJ2eyvB!~Zv6X^=U?0Ep8n8wtl{sQ$^6kxm*hc7
zsxyw`LXV|HXpr(g#L^>uq{jV?2jm2Zt-oEzJNjH{uri}kzG#hklPZCE&0<JL*ob7c
z@SWV|<09MK>zmzp2PfqyLUc?SOVb%?evokA&SX4BrR8p_<vaE|iaG`*vDY_=f20m!
zBvFP;ci6m-VX!D3Rh)Ko49WF&*QjmkbnDaRwC^6*at6u8QV|$Ue&0F3>|ax?6?P)-
z0L`r1n^#YxMPm}*cr5oM-j9_^i1S<BOM^4V_hKUFhEoh~ELd<LFg=Mp$^8}{<>{e?
zwQxh+L+yN~lG?{C9<#k_R@VwwvNSzgt$9A}Hs%dAxfnY=Gho8n9bs}rCGaNW1ZCB0
zb?=SE`O|Y1{~QO7X*<24T=uzdXRv4^&xzen;*(w<b|=Km=vlU|R@dpi+8nle*uLe@
z)WDc>Zi8`~Gv@`Stkt+6r7t_rYj=E~<ist_>x!8h!*4@d4K5tmOyCS`0c9FSDW%_}
zJ!-DMNh=hpYyyXQ2!#zYW7s$}i{5ye&KvqXk<QAS-=<OGD!=%q8PCvi`|=$1?sJSu
za$vkx>ne*?%s;T4woU*~z|*R#_v)`r+1EhfQ8YonVr8~1Qix09Ij$%Y%q`KY{j)zF
z`ziRah~uT`|8=ZRlm;fu<SmU{WTbWI0QBK}LZ|oCAJ5%Pgq}gjz!Mmq$^Av3LB?Lw
z8-N&w<qBQE(n&V{H8kT-xdFgMFyys8&r)s2GgDCUL%A=OR10skMnrFEbJYO*+5q<~
zJTE3R7dWekrQF6>@vMJ#+>5}&mw5X|SL=U*=6$6|m<SeN6Iq91bI^Ntg)sqXLxcTU
zuvLr1_ixTfZ1}luCABGmpWGiwD<X&j_Y#5j^gMJTLj$!-bzW@J3R=h7)ND@V>Pwpa
zgS^v+c$Ev9^xHkRvI~FD*6NU22voIPtl73ru)PEoNeQG<77gT){q+ug#SZ|g`-6iU
zx_-VX`ZVICX6_TU{aYjY2lzM(TC>_Afh9jrVSl>`{o#j_=Q&`1N;nc0|0@~z_m}=Z
zzW6tf&0m=IAHMRx{I5p|2&<{}fL!apum}GLBK$n!{PiAROBke@sVO%9$M5^^Z|i79
zP@{>xX%qi{Z=nD@?`)OMU(}s{`F8(@>zJBAX4%Rf)xWpBb7SxYrQ0z3rFdhYQUrk!
z>!V30$uqwtY=8UfrYI255_dzD<tOdt(Z*4iUsjE?AK;8-4aQjLp}rRXrLi|dkn*E0
z>@?N!xG(?Oi>^t{QKg4Moe0}n^*D4N0}!IqlXkamF%rj#hf_&nn%VFOx43-~E)dur
zX>S8lf$$}49L5Nyc<ps3+YSS?U6Bq=2^jATKuf+NjjST$*W&QE;}Hx|VI*$gpSyAX
zTx~|;YoRZ11_--LW(Z(yCG(*CLNWqkN(8p=Gmf?k&z|0f{PZ~pm3|IdmI3I54q7{z
zOfsVpwb42>I|iYDZH(w>nYA(;ZxSY6;NW&|d*}|WYa7T0${c>u?iuMbfnlWoik?{q
zGbtVsV5iOlpCRN06q{|}f!8aOJDm`N7+lzV{qoVdj)bEhluujiL~FW#K9F}K!^*L~
z;`haZr#8+Jd<5O5!&pLv>JAG}F(K&%3PbUp(vdZQJaQ1d0+#ImcnBhwf>>SNS3(lg
z`=Nd^gJVPd8Sv3Hc{3%w!O{`ukWx_EQV#AHU2U+OHR*2+V@b>}uXg$_rr1*c5JZ;(
zaDJBc9Fb%$Hsbc#Hpt((E@3||Co+?f%D(v8N#NH$^0x!kR~*?6l4&fReo=tk_WOd~
z{f8f)6j5!?oBxlq_YP<(-PVT@aYPw~fJ#vVp-8VHic}%=CP+6Fr8f(p5}KkYL3%Gz
zlqOX=(nUoG5J0+sf&$VJDbmYtZD*W&&OP_e-0%A@M1o27yWe-MXFUZx9v>p7+{1(S
zj6MPE_Qa|1-7#9S(=1YBrKcybkIVrcpuQlQ{OJ7P^S)XYS(lXS!5()e&!FsIZGM_p
zB}*VZSb$#mp>GA~fCgZMArY7VqUG=J#O0&N2HUKhll<qs{9Ybm+GF<iynP}rPEiN5
zxK^P0C`6Ebc+08g)7qE7MU#<j4>`?sk8Vtv|LrH<gPPPvZk^08Iu^fforF5jEjb~E
zqO*V~yac14x*Z@L4I-ICNYtp~!=w6yc7%}V2*?zdjPU9d0Ln#xM7~4r+U~P(<{{A^
zPH@WwS33JO48jdvt!`@!wmMEg$l^J<0))C2f?I?RkJDvv^i-W6^ymsCTSJoAS@X8R
zMzRbsIujcWJdLCPI3QH7R=D21aYY8w<exzXYLE2HHDJE1`1tcrLR5Ge7{ZE(Awf@b
z(zB+ZZ%$>tF}{^6cmv7dm`NI*)nq?Yt_|FectM1xK`L|{1iN4cTyc9;K2TCLmMz}+
z9_XN!KCPtusv6Xk4W2Oa`05yMmysn|x*R7SbZ!+GPXiz~y|TTz`VwKeMg#aMZ;24S
zR2%g@y7XLF@7~>lEh1!d>E)|!r_xi3W?+%po4vVs{CJJh(uCON4&Y>kNZ9Dj4TM_=
z#IncY8kY?nkN{|gOW*ncO7Bz8+eYXw@V4#Pj&eV|c^L#<qo(G}ONsb5n}~BUGVW%W
zIafX+Q!;^&qbi6AKgm(6eZR<2PsCp@EI$a(3&_=A_F4|>cn&s<X9ilO)FfgZNZP%;
z!TEYjCqvp|>GFn{OE-Eu<|GXm&JHXzaVFOnX8<$qb&yRX+<_lEvW=yK7c5)T_^XqF
zjx@?h+-R>hC)ssD)}g^|{)Vv9^Sme8nIDR`RT2|o!IE=y3sGN6l(agex{N>t@OYDT
zH^EZFvL>Goj5U}?))SoCzctRftQ>sc*g9q;4SuqXE};{&Ubb4mE&G(gA5=sfW=<HM
zz4I8lmVu7h>JeNoT`$qIsXc#dWovUQ{@rQ1W5}K&wB2I1L+iJA!t%xU6E6(bcD{pS
z=DYP0D<J1Q7Juzt0cBO=N1I$ONwf8~HG~mt8a43uNB;XDWCv?_kec!5?FbtfB?ZWe
z^FBczfw_*U|GxJe?Qb$;`M|lwd}oPDggwkDTaH)(fTq*zheD2L)E5rn$qTIkpLpC|
zMtOPt{*jvilYq~xQ6jKgt=@cq5sDL%&PDVs@tgSwW_Ig5)3w)i$*?$VK_x6_`Ci#f
z*_7BTj~L~=SrUb0?%%n6@H4pWh`l&%O^OXY6UsCxA&KIF1i<6R0RgAH-9yH5959J9
zuOCZyQ139%jUCe2e+~8}@Ky`Zlc-5|YCPehP80~_TpUJp_4>n*>2U>oQDhxA0l9Vp
z26HLcYL(8$&h91OyIl@bE%`^50>YeARNLmJ(!V%tWHE^`cdax*PCEzRL*VN$1K-~)
zPj?!EGsXCseQ#bQ&@Io^(`g0n)IvNDMz(Xp{_DNHhY8-ggN<d@^Bu-NZn;V%Acmk1
ztBG#mmCoBg`d%Wvy_ombTKRn3e#!(=JCM`Fdmn>Rtt!ZO73fq0AZvSIC}>D+_@rDZ
zv?0`{c=MqbB@Y5%A@+m9Hj8r#x?eVspgYQEv-0|1PWWaboUHME<PQ3V0h9Zkd1smd
zF``?g$JwnS6CBXY#+d-&sqnX=3}LOfL8TKnu^+39Osd5s&<bq_20Es<FPzLf(y)VA
zZ`6zS^2Z9lARQe)eDy1M{nEN*M9sL4;d*cP-etF<aNCKcFWI~2+Ji%41&D9@x29&B
zerR*ABieS{qa_3qR+iZ%vW0wrHwA+YuT(Qi1Ia#`qQ&%nai3hb$!C5IUb0@2`N(L@
z_r#O@<W)jF2~*zn60!WyJom-7=xw<mkohXZzc+6w9!WJ#I+>P~0c=S!`oe2xTT(LT
zk0|7wx#whMdUL1Ud!@M3@O#|)JE_;j#imX9MuaE1+F(&o+Htm*bNo#ERFuAP(+a@v
zG#NQPjfOZ-^evpEejT@7cL$Xwx8p-O;SJ{Q2;k?-5b&L}TJJvMQaWS%4{l_u7^q-2
z10NdyG9){BsxG8yi^~(Z6H2<wtoA6Y!><n({<c7gxkr%2ZsP%$YnfKGm%VWgus?a4
z4^S{Ufq4Y#N-*<7f71sgO5JJT)flBhjrFRW&dx#ntZ&01G61-cBSFRpvv_h41xCKh
z9UrtrQ-Y%v!PkY70(*`{p>7W&!J$hN%Q($;@H0|4VBcfj7=&>Ikgbg}@1BdLT|fAR
zeBJjALtQxY3bptcjRa8s2eJJZj$tdSfKUMjj`_9jMAa?ZcIsVhfjjW&<gFROI=LqR
z+oL`V#<9&bwy~G6+}TIO5H#o^&4o`yF4BRt*Y^9KQT6*~Ml3(zQc&K99k|*iG><6g
zXkLjyHk+89Bn)5?yOW2ql}Q0E-4BRz^}_7T<HtvdA3p&gFYib+=<IR1Pl=;cA~03c
zs{Fp^g)ye{gtJL$XS0-XUaewT#|olNZ<vM3FWBR;@<U~{pA5N7<uYv4Ri_H@D7#t<
z-0>{T-LLuY+<h(IFv+sZj_()g0;gnoO(Vz$z1qemhFK``W_dE6`dDNLNhke|os=6J
zksg<x*clo6EIG#HE0ZRBaP_bQYNiJvJe{)vp!M;j0Ycm&wqs9V1+W{bnrVzFzexj=
z(dxcE@2GZn!c#3cLApEauhfj7Z2}GF%NtRxu`}2-s}o^?*8pcXynDIEG;ayj2*Iaj
z42x&lg;L^$9}yRFcS*%i#EB&N6H^hIjFBGYcf{VpQQj8&_-gCbTGo1PP6z%@-W0?y
z>xhG!z6!6){ml{Z9@QXtDu1GE1yRPKII$cVkMucWH2kR|{#RRz81C`DBOAmezs&`Y
z;Z0NQ4Eq44-I>)|B)Q@&P-u?~0+e}Zp|Gf=^4iNcKX0JiGdJvRnQHH+m2}Y5k2M4G
zZ2@8}-Jg-2zkY`5CZk>4W?%);t6_kYP_wnzBCq;2u60S+j3EHwLjb|;*sA<(0a~dU
z_TfXQ7kB}ek5DVi9Uj~Cr>K)usrgf;`#+bOXvMl+>Z^ZAh<xDZhq+_-jRs&K7184D
zWIW$N77-xPboZR5fI><C{2T0|t!$G*=@|KCXYuE=47(Vq3d0z%`4b?q?_w$)Wtba+
zjw=NZkICsj@#_CA#5Y=!9Jf*^pD)0~vHPV`zOhm^3@550AkR)8*4(y0d9Eq-_^FuO
za5Weu(EB$BbB{cra?*aORH26c!dWq<tDm=Po{Y0qMceV=+3Uv0wfeoK%M%!!T>aVb
zB}rnNCP``nciDHJ$*hMgH{Gqi?xK!5E_q(p`E~X+Jplub7%~U+Xo`ySP2mdjs7a4n
zx3QElRs)j1{R!OoWAUkuvQB`(V{vbqlF>`uU+e$ivzZPSjDP+n#95?Rr5lY)X_XdR
zHm;W%c)!o##H=00!n-Oh5@)X>z#Mb3X`s${G)2|fpS!|Ww@hZ&2j3Q72)!2vy+)T7
zl&)ioCQ~%@Uq?*MlPn)@x<M-=yegHg@9Zx+cr>3R+{+RY%;>OZ)zj+c0x$k?%UFt<
zvpC=Fs_rPs^xIc!o^AKzzM7^7C5ct%#-}u-Mu}_JZ)^8v9g{9+kHW1vT-xgpLFM|o
zV8F0M{8`hXr1-NHq-dP1x_~@K5`O)STc={obPB)Z>6xQRD6}4bg%GpDnVAqJT)-tm
zzu3*F_Da@qUnQJ|#B71QQ5WvJs(|*Xjo70x*XbMt+kHH5#+|QFPAr!(an)G&@EgUg
zsiOIX@B@8g;ZEnFU0Zq{GxW0P+9eef-{URK4wA9F%;O}3ZWX&Rk3)^cNri_CGb%qf
zW$0WvWTD~uOmJMqu5q{Kij##Y{>X&e;o?W*R0$P=ZkjGZ&D-l1KTQ9=tDB%<v5NQ$
z@|UG%t)t#T@8PA84&gB<{z^$h5KLk!ZWT^y7Q&~cjfI&b49)MULT>_}2jy(AxrE8|
z+VqW;rpiRJG2vY^#!%JSkGYHz(ffqPFy52j)c4SI6vbD^l<qZ!8pa?{r7XyRosRuj
znwZw}xM95&^Z_gIepf-``7x@u6uITdEU#gXvAkyq*FT0m=Bts`&^)3&akUm-pVMq&
zl`M7sGKntagI-xp70EwWJCA6p@G;d$!WgeWW+^Fk-m29exOe<+l!`xkBh{NP=*tPc
zh&+&ELiYNAb~{^JDN1gs@zr88uN?$C=<U<*y`c5HUr`gga<y&Zp>eN4??ElUxP0lO
zXV7^he;{%e9X`6Y$NHm3Ag9Y~w)TB*kxYbrrs>t$h^a9JSICZWZR$vb(lZV89x=S}
z@MEd3%qX(3a@3HNz3F-zQ}W|jx}@JE@ynmaI`~F;sylxR35=6WzdqHRL@O1`nmOK*
z@BJf(X79|W<Us0h%u*$Qp3RjS9*@nn$L>6~?uGEj<?0koRi#&bTl15<2kw!Yj9pNL
z5x4r;_S-y<G&Q;`MQ*vZm(bbp_^k575*2dueGFcI$5rt`tN$eY{rKfV;FE;CcO)R~
zclJX$N<LYGuDcgLDbvq78%>>~7bT^>#a3eJe-A9VQF?G{5|=F;b7Uax1zQd46s<~A
zyw;wC{F$@}O%eJBtlKu;9yO~wsdv)OIST5T-uCKbtn9v6th%9QjO~mJSB*RprV!1y
zPl)@bO33B>Q18fPAUWPD6gkYH@fis7j8l$C-nIL4C3~BOgpx*?tuNFw=s}XE&SH3W
zsmQ(?wE_dJvAwE$wqR7PG?-}v$+}_j@yk!sj#)aYs7OJ~4Fk6&xbs{Tbqg;#qc&=c
zS6}IWNXMqYYG<dytTrX!!s;*zwh+gr*j4rnDsxYxoyOuzgE>B3iyr7no$NW!_l;ji
zH(WFPmHoSi1<DUp+`mYzd`K0&))U-XRuO9P@obvNkIeM`@$7uGwvdip*N`WVT1+zQ
z4-W38C9ACT=*Oi2k@-doyC(E@=_X`+=^p&flJSXjCHXy@Y%><9{4~>3J(8=;+^~fS
z$tWABs_G037|k4+<1`Vp!>drWopCs-lPEH5t5tf2Q=&b!xWOf=J3Ci<E-Gc5qR*vJ
zLWk2sm!y1_v=k@W0J05*G~0CkG;bMbo=5t_wAQD?tXki>SOj6_uII_AI)qTtw>R5!
z-@9z|GT>@h-mrq_b6e{?ACrzHp5r0G-uX?zoF-?6e@GTxEz6pXy7kzmP)WgQiSJbF
z5`~tJ#<WcgMN3=DUGr~lk(x_)XSG+$x5Pou{NF*wf1T63_d@!rT8q@^UwxGxInjvH
z3p|L0ee)Ds(o`ZyY#k4C$Fpt^N_l}!CS(-R*cDj>Y!ynI-Bh~E#r&sF-vZm&aGjd_
zq7zwk5_)6^kHiVAOmh1ILn9Cmt59pn8s`%;VT*yxvr=<TRnDw3-E8h#GAnK5123jG
zssb)5hekfVaG&D9v0S$TGLC)sj~>$ee&q2H%E*x;2WZnSrj_sWd;0p*CyvO^NA7Un
zom}t5F;4`$%+>HtaLkWyuP}soxF*+3M7!+2V?76`3%_ru7;|r3lO%;KTEcyKaxz(2
zj&MOagVKwVO4$GyJJ}fK$mZt+K}xAwN0+K}rC>yU*c#(|Xn}D2yjrp{)o|u)R3C#z
zO9zS5BH*^%DLX+tmU|G&Vt(Fkt9h051Ox`uFyi*0c8OFvdphJ)|BEB4ua>?(Y)!o!
zr*x6mvMZM?*2Y70fHPCQcr=wPt?7`Z2)jv|u-V8bNjZxW&Lz4L!O;_6D%tkQrnw}&
z^WX~cB_l+-L$Ti%X=3WddbG~V|LT&JnbR0C43WRb?)Bm}_W<@tYYbST=-Qz%z2~Ae
zwk*ilz3QjRXqSX^_=>J@><RTap7?U<YS_e`$BN@j3}c`2i4nRptMR*cOBu;oXorEw
z5{Bm4%sn-_3IyN{iedf%RnL%n5fpW0TvG{uhl`dRh9ON}mp;!QGrfD16h6=ww}WJz
z;r=_oai=pzT22oj5mi5Juz#Dg`mA=Z;YTgSm5y}$j$v|Zc`tKwe$}`0Y**1Tp@OB|
zI{kdjW&Zj~qHlr&$&~S-Ap@$lxO4<s9d8+a)cv{?nn@4hNv!yTTm1eTo%<Ix@dxfj
zQi)T?QQPgq2iS&#vgyo*Ac_n{O<;-BfXpEwUk28So_MH(L*zu?wRLiK7J=c|*E5P;
z&Z0~HWRopbid7{?qrz+hy*8E$fBK@b>TPT(rX|2gsD%*T@ZkrCaaS&n_t1^EhcI8F
z+YW%l4_~H~>UIC%SqW(#xtt{yo$E}6Z80$~5jpY@HMVc*tAoYI&CX?1M+sa?_N!*n
zQtb8Tb1p6{y6>y2-k?Ao^xrybNu-g0U~D;vx)s*%HM9s?TmA~mP57{O!e~EMmjsOV
zPpZT1{SF@#D-!`0OEt+pa1Yl<tWKm4kbHGXRHX0XZuRC-e7OC#>C$fgHBB;gHj&Mu
zKD*mTW#Z`l3wZVvu%b`??IF_3_Wn%^ea_@Bq>|%6_3qB5zaT#T1b!^+P|xt)Sq<Dd
zWl3X`5me*=>8(O(eT;#+XT_JI8BhF|`d91hwK%i>ie<7@yl<xK%M7g$)@Zlx{M-4r
zwD}f@mQBDo=^if?@MlL03pY>GYCq`{bM8z}B@VJ}mg1{gg#DF%sAMJ!&Ks8Z#=KXT
z4A&YN(f#tlci=nMr6wTX>i<|+J{eO-Yc^dzat0XI@*`oabTP&}Np~vP5y*&+UYj=N
zY|D!UtDw;n+b_bpjW)2xD~PS95m1aeKgG)|F{Pf7DF~u6Q4JEFMJWxI^Qa&oGsFH`
z(zTu$M37nDV3n5L7nyVpcCo7o9=fkOL8Ad{_<CoKL4LheeTqw9)lq%JXOj#Pwad3h
zTAt9KT>Yw^8dytl<BMC;T!(G)Q=!)25V?o(AI^bTb7XsO@~mWZ>*D78E9yrAWY*LZ
zl~UknQ=gta29v%P4ztCS-R}6sSvvS-&7#?&mQz3^wjT+17$J8p_@R8Q38ocEZb#}E
z_6od-KH455rs+8~SD*g6$fbX8?XqV{6MziyQYQvm+XnH4#3cy@HkU&}i{A{mXN_O=
zrS*A;66+OqU0;sal->~@PL>?780lc;9csJ~voPwxGW^9>hU0G$W}jY8^E|VCJ--HJ
zmB;pY=l=M5mCh{w?1I(B!Q5&k2i4HuXsC*~kE#rWyN60hI-82TJ93FVDlG|87S8*U
zn~4cMxpid|+}o9Zqeg$cLe6qES<pYujUu}~rkPjrrRyk+I}iJZF;`0HkR<@TOFNC@
zN4jg3`Jo!Rka1nt18-0xCs$u`ir<_A1CY3jtHH_$Sz<Xy70>3hE)u2k^4h2DgZH|V
zz59>t-<StB5zXR-DHmY8N4T*UhGM*$nmfA`Hi^E(g8rXtBIB97-=l<xZ+%N~jSP7c
z_?{O~WX`g@7p7(@vZh<|6<m$yEB|~SBitd-qr;}t2FaY^Zmwk%9a;QM{_6zc67Wur
zmZ~+{6Z-p-eTel27IyF7)%WiIECkegy{KJ=2MBY{kLn(}f^Uc)tyiky<G<s-HLiB-
zrO}}KK3S2L94$A|q52qYP0y?inq04nIj$KXoHOMDT>m-tj@s}j&i7)Fs2fP-^}g1X
zU*NJhVL(gvB=IG;5}I!j`UWCSUJ9=Am<m%e`BTBxM7PL^wf*}M678;_?D#8m+*z#B
z#@@Ry+qhA}{dWgEzdFoJsLm!m7WnyF?{6KzfoZ-<dy-TH@EF9uG!8kk%di(he=m9A
z7$!bQ@Fc^1&FEghROxA>1-au8Fv<#597-a5_jM}<0X%0E>upRI22FE;mv?*3Ip1H<
z;mO*fGsc!?&vzKOk)&%2g2vue%rbX;frkENibX<uuW&0_*51S2OGwCAcm&t!j&+(C
zP~Rr4J$piVbI#w&ZfMSHF;)Dl+~ZLt*Q5)HUjBnAly2C!awY8Hbj~NVgH`EPNmJU4
z0Qb(7i?u{JWJ$0d5#e<Drj`(G9oDh~#A!xUw<mXloHwN?0zi+nTsno&A-jJ(eQ;Qn
zz@!)7X-h!3We6jfzI1<IryvZk20}3ecfEMGV?Ws3QDG;uPH&4pA&!D^Szu;lRLUt$
zqI4KP=OBUzQ<e~u)3<|(R+P7p5L|^TA|b-Y#z~4RD6-8TGxVbb<#_Bp3(qd4$v#TM
z7QSM2<tFT1J`Kt%Q)?O2!ukaFIJBCK+PgQR%ngjFj*JB7W`et_Xi09to3#|l+7nJ9
zup22-G%uS<`~OB1XU#?Y9r`b!_Pmc7<F`M)aolQ|<81)@%j$vDIO(v)H7@e^b%(W0
z!rp{U7fHrZ({y`I^-NL;odEe$iOb#)yVYZ6x5Zei7QN-~BX0Tu-<}N%4=<q+?LD2D
z=F3S*2~x4$uKatst7@xV?W*&&qN&vzSf<2~N|w^+DBY&v7`2c)Z(*W@)7~2QYAK%`
ze0g0t_=XCdSDYBeSZIP#9`V8N7M2)_Pkf0tMf1gPYj(^cOLo$r)ieK#iPrF-`GZ5O
z?tFv&>%k(Ow@-j`Z~7hKxwdl(<R4)W5&=NsyH#2<+H+$#vo{hOjKe_;&(qhJe*dT4
zicRQ%!piPK4*Rb)|6Yy)M<E<?%u--JPN_yDgvCZ+Ca&f7_|>oqa|?lYu`PmPw(nV%
zl)c5^vK;Ao&mrex^cHucAV_S0h?6}XGJ9lI2If2CVM|DU`PrD;q!<Ri%9xY8Vf~(r
z1}^?hNZ1#bpAwTu$a!0VlDOS-PCFK5c-<;DX{%-C1giBL*Pu8Xmx-niF1!iW<^wRA
zaRlYLb;r#RcZT*B^fL1n+dSZx1D+LFmnODz<+4}cVh1Bmfa4<%DiY||bu>ekEqC^o
z<3(#7^>1C50Xpt@2X*V$b5$_unuW&cZ1WfAK>XO-&vP=lPXuj_e|_5OIfZj!%$haw
zJekUmvlRaJ#XIH#>1%^6^Fa4Ls->^di0G#nw7_jRd+wI!{d<wAb49(afLffA#%VMA
z0m<g13hY&Z_yNfFuty9G5Ye!|NTr*dK-3X55`2m^bG<_HTz^B{KeW7WgE7}!u@<Hk
zffe5k%g=uLo*6wJmY*`r_wu82!!Qh-Y-&Ficf9(H2Qn&_dBVHgs>^qluHU&pW0)Bx
zA%Xlct|@}<6z>>fmbGH^Ti>NKSRgv|skz=+2L6@;pS=L0S_aNRl1K+`B|z-Et(Bq_
zj_fqOh=nKKvZ`pIX%u9@IxZ;Qq}h?|idaX1Yl_LlW6wL+G~25fEaB|ETO#2?2xt)w
zwsROizj4IE5-=r2@hM*Rjc-s_ogD475_A#nfw>Eni?&eP>Gt>Z;tk?4A?uCW+Gpu+
zx27Io^7<Gv89cLO8)Cf)+dUgkXZ_vS)z9y?;#Ta&`8Eovsum{XYy;KCJ<HT4c21$q
zTzTi?0Qvq1!We#+E6tKcnd`Lg7c_5%wczB$#@RzO8RuF)MrXhjsSj5hw93%vHqQtD
zzxMNGCh52;oO8&k>WW)sqEfQ?@2xOl92aruo`9ec|Jg69W;RKm6K{Z*Kza!OhVKJ4
z@9~jZeA+99{AvjXI)UkiZS=a}I3xYZR<@e>fST8Vfs3&(o1PmTYxRCkw0Q9h-MXA(
z{i;t$X7R<>susYLPb^W-*UeUiuZmhN6T|d|-z{?w^^k&YShZQBN2^d)uBdJtItU--
z-QfA+uN&)Mo5a5!zaM~cblL6AtmQvSNRwDHnQxoUw=mt`<!^566-c3CrRp52^VutZ
z^~x*Jg;c?C7Cg&gfSm9w*%}~0!CZ55<-2VA?4jy){#mwV7RrrqXH4zFXp!I-rj#z2
zogS4j_vo%HPx7+N6vfQ5^j%yrrd{lZ@8{C(ETuOGnh4BoVbV12p<*-(<^@o;HBx_~
zbZ2L8f`~GXwA>=n^tpS!<J!{)$(!JKv8#<t!Ivb4c1x;;Xe*UN@<XUV=3O4nC}|!i
zYBT9BfMn@cwduX9CnChKme|TKf==_Dp7H(9bBPbp{jFSgjrA>=(`3wQ&r$HcNKFjO
zp`q%;R)iUZmPlqtAspO}W8FbA-{c!E+(x=Ws`clq&y$tCi6_v7L|e0IRU_iw!iS42
z46HArALoq;KZ9y@XE~zm0czxMIBl+^y|uD6y1zv-+n{^zvxCCVMbo`f;)IRkZE7~N
z$R@)C<`810berev@^t=<<tYn9Jxy6|cehFUE!$9>pk2I_?#(AVr7-u?lWe?TDIM~S
zftJsiYBY9-!lGec%4wJ1Eo3xEYRWz&HTGsdoa!(}`0RUu5`V#GOkM{dzvK6n5`QJh
zz$IL=l85LZCzn$3=mSY6jqewh7nPHQwK_uLl5MFQSc{mio^#pP7p^+7qnK>s&b-d1
zCOomB^L$*JGIbj@rM#?Be<MVUNyKBtJIy*xXP_hx`)TqxrF)bP+UvfjROhuo?hr{k
z^uB`cnTUxH_HO;n`XBa9Y%NUl#?$-uZ-`n=F#jeb9mX&<AIPS2yZY;y>l3e2;^vRw
zThIKllE2q}Cj=_Al0KF-qCJiTFG!pJo{`mA73)&6zlepE?+vd-WJUNq+wy@lgK?!7
zGlY3MyLI5b{T=Kf<P9W^6EX=uj{I33PYRK5l?T+zvGa{drJYC_LLKsF-W^m^Hdfxd
zB^!9|;j-Lv1+;sW${AVF)a|FTk9Jswb`#i@DbAIND^@6<P~sz<YP+ey?@zoFg;o6a
zb-#xhlc?_ziv4DCVks=BlVQw_{<@FGWJmn2vEJ1Sj~cnpwEKX#*F#!mD>K-2j~45v
zppz0}6dVI&{s%Kk+3R^&Gqgh@vjo4=J0F^hKbm0Qlt1JUnoST@u43Jza}BFHy<g(o
zC7XKYfeO)%j7a(+EJdm`b~jaH@^V2YdsWo<+tY5xwrP)*(BDVl*+yxM3i3oT8`$*4
zi-+sQ4Ow1vC{5CBy)KPbj%*F?Q2m<y!%&}$H1t69zInGPtGgM0xS-RZ>61=QbCqT6
z2BXCuZ|t?4Q?|m<q0Y+7m<>KBS=M+oN;zx4i(GtYQzkP?8)dC}sdz3-DcSb6XJl{I
zkk`0$1E<RoZP_mTRnS`skQ%!qbnU+7ui~9dw9+nb?DHZ&AUQ`*+AX=0E*?=4RC__<
zWmITJX!M|m8LY<6+E~SO^%g9gg|g!{E8`mD^aH+J)yob8_cyx4NS`PRvrhPXCTx3$
z_El?CJh6IR^*jugOxv06)G+UNUImWJ9*3L=pPy)?JB7R_-6<2*^Pqn|v2I!Wl-g!f
zn4E8yf&7Z7honbx+231186zy_UOV2XU!0>q_<{|*KH#M8#_x;%gRUBSd=INaajwuC
zj7Wa?-ytXj*%^dlxZ#w^^lu_ISwBQFg)xl!iL3<PA;(g)zhO~xV+MEh+wU*z@Auh{
zJ~!0zZ2uqjo&WaQzerv8D95y*+}}};O8vL@P{Ow-Isn2Ms4N&eP;83zI?x{-4%%3r
z?rDKJ=%pu~i%plyym`uz7rE{u2kg&3J-Ms0h+-Lh5a&SQ`4U><*AP!($G1Q4A5Mr3
z<b?1(p8eau-G1=>P}D&{DW*OQ;tk%NinFAh>CL-5CIdCa2jIy#NVzW)=kU#cC^1>?
zl2dy3ncOkKSd1?+TF1TysmO&vpx@<PAFh4qGVnjL4F7u4)(IF{m#emQ-T$riB_<-o
z-V<sH(bkA*$>6;`dsV3&%zjQ-o<^Y7pirKJCYCGG9#SrffsfGx;U@w!Lp2^?2=s3(
z(O=E-Ur!n*Ba7s^{9kSKU;pxddhnDAv5why+(hs1d6a*B!vFTb)(iQ-uMX+${kISN
z-+t?V{n1B285Rw_qpRMp6qx_hyZ-B2c=`h%WXEkW-9!K1Uia${RI&i5Xj0IjXZ8Q@
z?e`*7%^!mo&;0obef-1gpz3jigE0>&t;G649u#>?H{IB;uKea`N`cqric_LoP8#cb
zRtnKt<FsefTp0r;hYc(L{T~0%PZN(M#jD(_H2F1_dul;>0yrf<X{TTWc&)}`^;@i;
z`dGyfEY=l}6w%v@gB-D6l{ZiKL$cvnMAqd2dB1!r;VC>wTA$cdTEwxFIk#*+6kJ&W
z?R-1<WaS!PZ~9sZT+wbM)+(hsK1)ek(Zq2AJW0-OD2joLXZtY#yk;>@T4dyO_MTZ<
zCBq-8{gNR7`prXtnSyTtuu@>=WFc2-SGqX)^RE8i_6vAkodkERCMhqc^4A1_iq7z=
zKgTQ-&OrF_Xbp8e@8J%ujUyz{79>e|+>TE{7>6%Gu-$D97Vlh_K-4-11Ep!vuQQqV
zfsMgi2ITSq`1d0rviG+j)Ik0!Ld0Wo=X(RU;MP@O3%ba!o^|g6K;Sxp@Lu}C5<ZSN
z8G`Ln5@HUBnI{3wR}8Wd24UK8eAE%R@s1?~ewSgy9|u^ZH=b1^Z4@GUD+3IZKDe~G
zp}sc*z7CB{X`k*E$*5D~2*e=%iGR}*I=eZLof1ZYn_IyTRK#8|^D3G-KT?O#f98V#
zlsYxU2_fSI;kzX<r*xl`83-i)9X|j_13vy?aMSx2>gjR*xol%8DSu}WV)5l?mt{|H
zOQ0bk?FLMd-#WsaX<AglYJCiLx<SN{k&CXg;nDGHAMBd@wD+w8cf@gdvSZ(#??WE*
zrO+DpL)TyfC9Rcwh(g;s(5>X2I&eMs?Yf*H0{)MFQ=Gm~Uj-0NA56x|GVcOB1|rq-
ztSyX!(2t2h7E=}M^?3&QriY@7&Rk;~7=-@Qm}?g91q&=U+kh=^x__YtXZKqCMX&pl
zxuEC2dPSO1#F8)v@9${Um)-5*-F}2s_2x5sg!)zZzSJmx%52}wEJ?5RUcS!w)k>i9
z^L?sW@S@bx(8hD>J{dQ;`}OY`d@SU7varwamV`I|;#8Z2)**~HX79Bt$|$(}Ue8Z`
zN3;=$l&}xyqVb$|mWB+qgBYx7w&tf&-9bFsz|wpulI+~odIAkum?Woe_DeLD&+@<=
z;!Vlqkb4EdydrQ=bVN0M2NFYp#C&yElINldTTvfKD6RppGVGf@K!Fzs6ZSiNazKat
zbNxU@;uX)11C0jBveQzqs^rEj>~0U;<hyJ-WzxuC0vsL3pTxyk0F#P<i7kMwTM)ki
zF<`6+XT{+dnS|PtU@b-W=zlwO|MQkp=YoBj^9Wwb;Me|Po`ON_)Fe`lsFKYlXc*_8
z>l1glqJLcqS`!<@)iVzgo-sbwY<9GJLJ6jbYdV&}F)F=m$??eohsReV>r%TU(Uv8q
zWUu=i_vnpnpM|a?u3r*x<2V6^D!|(faF!z+rHM$m-U>vtH;7bQ(FghM{c-T#=rd(^
z-+y?|gt~m#9<q7={4}yuC>GlHK<i`Xu*TPTZ*TNN^L-ur6k=S|k~N{jL%qwqySt+o
zGqnkX+*yClB!aTJWkb<@;yym=<_ZuOzoyanU|n$``amM=2t45_l8Env7f`xTDziv>
zF{*<w>4HS|jF~Jnm{G$Z<T(<B{qmMSLNeByc@qQ<BX?6t$m|W7NB-P3y@Bp*oGk6-
z2z>8i$Rw$3-4bsS2_SfO?~Uf8M~Ll;*bKy|tFT??6<8sgWAecfS2{`Qcmc(CNE~F;
z?mbAqRoPK&(@s<PSyK8^9*WFPe&d+`&fe3}s_ABhAKC06JU!q%30yS_hkNfb*-^8;
zPj6?vo}|Pqpn=WtfW79R-&Apl)@8NLUf#Aaai`YBuKxOB!}$32GTK?NuN*D6L>wn!
zMk3m*zRi$oH;e<BVxAJ{ey#LeUkte;vhIe?l(QdiJV3NY3ixyhun5gYX$J}u-LfLh
zUwhVlFs~L7tGqLh9<>NlIOabYlfD`yVy^j2M))^^XgP48l&}iy*arSU+v@`&Cpq!0
zcharYR*#3(Fij0kEiJB*ax}Q9<`%$LvmZV`dWl(0_Vk^u43gJey5cxVwB5SFAbx)b
z&9KM3ln11{UH!(=4#$dk1AUkP64FkHk1;iCv0?K{iT30*i8y$VPfN2za$G2Spta4a
z1Z80bfncIT>wFWf47rgE$<-0C;}pBmF9FV=AY#eA=8Ds+jIrRm*_P1gwgrQagG=2A
z)0*$>$0V{?r6*z{eY!az!T+FXk~Lt3Q<%53Z2Aws^Y@(;p#v+hbB1=)lp(Ptm%)Vf
z5XKTXzYC+=aJn3qL*1u!?tqLh8>EHqr`|<)*V|KCcMsk?+(d0R0z2BF!q1SZ5$Kdg
z)20FS0tN+3MiaAbiDBjZzEw3}E8)Ah4YPQIy66q?@02#*GVhoZIC?rQLZqybNvE<$
zdf&wt%y+jsb^uFLlun=MS1PH5Y0oT-Zw`h%BwBC1CJVctPIne+kx)wBPwm1<Hz8{4
zk*CkvJxj$;VF_Gs5!9p_yR#12&e}w4fTu;Gbrc9l_^Eutg-9+9s4apwi>@k7ojF-(
zmP1C6l^;&|Ml^y=t=0q<lA_GBwq>cYP;mX>Nzudj>Z=8nH9`Q6H5_$^Q3tL>a6}nZ
zM(;#wV6!DY-`jg%+{qV*2MlBX=IpKhIrn$s-xLUl-XwLqf5M1T2@Vpq{&Sx_#S&dR
zRc?bh@&4jhTVDmk=q-G)W*6$?pk;7E=(SMOKeJZJ8ghpVDc(>H28ZfB3rny^|4c*A
zbl_kN6aNaiWCSjs5`K#S;nydyuM82X0dNBXE4W&I8Y}~v%+DZX@RP}X3KP^J$k<iF
z5!P=0CRUz78Yr6b<Rq8cmT2B8ll<8p+u=6p;{G|wfk<yppeif9l4W2MHtXa>mHRU^
zRqge8VQlZ(roQB0?U{}jY9ItQZie6=*Q!*+=GY>>(n`D6XSO58qr7{e$=qaGSKlW)
zR1x?VSR-%2DmEjf&>e)&zDOWBQEd~3GUDwu%EctX65-@PwtC~Cx5{-<4d<*TytF7F
z&YtECt#;h{QQ_G1BJS&#)yU2cW(2jJ2`cE5`nT@qFZ}GU2SSCQlsi{gnmNt(6Q>aB
zp#ALUnX*f>KzU{V>uZk~tqGYpP#_$99mG8AFVsN2Ks<}Rz*<BTt|^A0NdWH(%w^O8
zn)Ue(CFV1%D9j^7V1b(Yef=Qj9_1ZMNvxk5T?gqf3{h!>5J^I)0?nt;2Lu*hnno`m
zX<HDjz<!XHmWCGdM$V3viH5ZU4E|_FE==K5wlS9zK_8-R%F3Z_V|i?>6l!V7<_qhG
z#m7PoGhT{PkFi;vO_GmfXT`8evdp~pCnysycl$3DXwZqU^s#8M6ku8l#_a9nDrLXL
zLcG1I)~!kJJaU#?BoU?x6x4F$z8I$(&>Viip%+4GmAY$SyBC_?FUYDEm0>;bb6ltr
z?kxd?2f|u8%v+)l+#`r!Q{>jtv|wS?VxH4{%{I<9NMK_gI4*wZN>F4l-G-V6D<>=4
z4~V@zeIJ9zycT}JkV>#5N`j6v6l5Mom=6u|1M6BfKimL}klVO0xiL7rbF71x@i^2%
z20X?YeeW{{9g8)V_ou0E?A4*s7V(d5+7g0vg5xb_Y+C8uEo7#FIZzKvVEG$Vhikp&
zpmaQV0S`BE@D`2qS$J*i@M6MzrwFVpHjgu1FV&I<N#4ayk3T334?-QEdXjOpWMAhT
z&|zuAbQv#W_h6;jm~8Frtzog+Ba8Jr8k>}=RaAXzW9ONfh*sCxFay}VHkYGs%gjvC
z%~|)h?K9d#b0v9|z6fpMzsfQ9y8M|pp%FWs%T6`~6{E0?h`}LOqeQM!QhT)wsv8OH
zB7`bGd7I*TShkvtmTO*?HDOV`*6h0g=OU*w1{XRQ_rPuSZKdd$pUXC#YWd<`vOm25
zzP_*hhswZJg?xuC^e?bOHURbAN`!8ABz&#2++XRH8ji@7?W5$Zh6`54jj2jbYmMlb
zcdX9_?>=S5<3)?=V>Z+zMGQv5whzX*`DXe$NM`GNiKUYFS5i{Li6|e2!ns}{>5aG;
zZb$`pURiR{bj11M2#WZ)t*>(Qs&QoNzPiURFAC7p(S{U7uhAA+^_BC*K~j^jb{i6z
zzQA?Ost*UozD>YJiJyfg!oS7aYC=lHsK6B?U<5&m&227q906R<1CAeU*UEe#AaY6o
z-Mi&B5a6wlyg&?8SjAiB*}8=AM3oy}auxVwS0?oWMuxsyCPOO4>bR@^6BQ(*a}5np
zf=923q+5AGzK+=)+_?bvsNe|IK79%O>*(4Nd&#aCT@75W3ho+r^FU0iCeBzDSH^vQ
zs8_Sb%KQWx0)`H3*Ts-)mM2SZKy`LXVCdBHoN>6CjX%|0s~C|a9DA_4KGSFk+jbM1
z(y<LXE5xy9<lHtX{TVSv{GGz5dXB-Oi81kEtJ8OU&i6xAROFX^?d;&o9BbT!zxCiU
zR70w`Y51z`M;37-orc;6aJHJb-_&qqYKYsf?;RDpW{Wny*5aYs)@=bIM{SAgn~%Ej
zZk%|ExAmz}(Mk3$uacxYtQF5Bs5kDPlRdxo`2k&wRfOvc7jB80!e<w_{BA;Oxm$8W
zTmJ5qt@qIaNt~o)blho=^8<Yk+r}WG$BVRAK!iTgrZN2uu`iXJzO^R70w+>t(iwDz
zf3$y}tY+yGCc7@hiTK-8sZ{^jb^qHl=B%7`i_)rpZLj=_WSEB3HFos+3;QONf~R;Z
z4W;CLlb#ux%Q|s13#rFcno|on(84?mM02Y*7gjDPUj%m%szOeW>fX6B3R$y7;RS*v
zM$8Rcnr+02T@%UiS29%#Wh_3Bg;JrwG{n4aP2w4N^c9TQ<JZC)6dJn{{Z+`YnOM!7
zNXZn7Wr>hV!`tUWR-PV`T+|y06QEjWLn5Z@Q+bd1ODkIu1WVHtv5}pgGp%e#$-qlN
zUt5ynZrY{#PUA~8RxD;?2pLPbAE-ujW``vgtlrXv)W@vpB>5|SSC`Z=Yh`6w7q%uw
z;P2n`@qM?o02;iAE(PwR39%MMO>8O+Qp#`obyaSRcqT2Gtl&$VP4UZfvo23QUCYpr
z%C;Wq7z{5PI-qQZb`n_N@M*Cet{x}p{1&u4=ZKDEG(ufDceEzV7j<JsrF6yLeKGg8
z(|Ku~Jv9D_i|!EY9YG8rcx7c?w(}G+=+s0jyaaAZxzcfsOr+^Ca6RdD02ODW7w*{=
z0MXur<&*qm8SSIRNfoF55mUuJ<)wX3Terje=-2L1!n7x*_n_JhP2qP(^a9#-y!e=j
zojzN7ssi6)1857Y3)C*?#gMlSdBAoXujD5}|Bgoh6<5>CcCbODhL3R?HCXg~;FEQh
zZZuPF35Af_{^D6)HKxjs-PWdjqcbQfi@3MYs<uCLFyiS<<!aE0Nx067WKxWGM&Q}g
zi&}_&99%|~g*8QypFaBRw>Kd_$A4cfxG)zopw+04dTqp5x%AgdPFwJCKaq-4@+iM(
z4Si&CMcOenlW-X|<>KWtYTcc`@kKRtM)O^o)rQHOq;-JyO|arAk6v-OzCQGraqYGH
z(v6D7V~D-dwH~%ldZZQT{5<$l69rD=7E(yB-CIH{dK!XHItVbl3s2%>YjaBr6nOb=
z?T$*X_tKML%j?{Z#U-2;lSt}a^pr2#(%}|tarY!>r;NnP%Ee7aA~|4iISvC|kqss*
z^Q;chih>gfM*0#@;8-eJ@2&jkfcBEu{8(K!<8-N^57B9&o4RfHhLjnVa`91pTP(wu
zC_-&8da&>78JlAwTenifYUi9^R&+Z7WhuLYc+4x}d|~%_rxs6;=THxV@J)ZCJbM$?
zPBvrrP1cjld=}LJU)3i<zZHS6m$Be;vm9hF5I{9nAT&G21Lh6r)nZSHmfv1R8R~rE
zXkuub3Lgevep@~%lJi%jx$PO8lHmK&+Bx16O(sxN5hmv3`hwf5Y~}AUdhF92#1g*8
z8296kbc9$gN|t(-^OB&Gp6hK)-T45uy<@Ge%{zxnd%lCD<<QNP5e1I@Uj6DzlMfzc
zb|txfMm3OP=*7%*dL_dzt7tmSmvNA7IQx}>k#C%m@<Ek?yabR>5LgMpekYujzv!4T
z3D>8AyIW5$==<I*xSu_BYR?VC)TsrluW*lykgy0M)4a)e$$<yIql`Tnm6y_`<n<An
znj_en9`HG&G6zL5CF(}qy*;;~iX8;%shMuDszeZhD}BBDi4qUl5RBEu+`fPAoV~g`
znHq99dQ_f_kDm-`=%#8i2Q$@=VCnR(vxUUw_Ymi+{E%fbOvWO~TwVJblrN!5)TOyN
zHN~vYk$z;ceV#FS=6N6ihRZ)2{NY-@26WO(3~Geuq%U*n9MOiXFJ3_$!@-#GKd<}H
z!b9S12wPAZmS>vj$uU^g*x=IPy?@-|eJZgul8pGVYGXRbFEh|G;He*lRyI2stE)SO
z@e+soT@It|2dhUGO9VqskP8|FGo5J$5G-I#;PvX`GTLy-H@B1y+-oJBaf-S;@o(Gu
zFD&@Ko_xxS3`C9^8{PVY(il5Q`CVQL6+%$&iN5E@-tk;3nP$K0!@MJ^Y~-JT7i3i%
zXhR_TRCPzIIC~8bDILCP4?lcvDbugc7<i3r^h~{A?k|I7mosB6F+Qfyh~I&Zi>eW+
zC_gBfGl|+>{Nh$~oR+M0(U!V=Q|BpyVa5yYcf_)?Pap9LB3NMy*rwSRvjqL@5RLNk
z{K`O%X+@@v_mA-oMhe-K6%+@ZO1|-Z;SuIA(JTuwj6TMs^Oe{7451cb3dRdQcS3Ma
zBDREX^>)c0SH&!k-2CTNQGEHt?vmMye%n{6!RMPFQ8N!+L&X|Yp#9Hi;9KMDyr*nx
zMgvxliL4&v&81nq41g<G_vf>1Yxfnmo#$om(|=5+oL2KwVOfEZf8@)OAc9;Lxh1K4
z%2lNq2q#{Mlg24Dv$V82dHt|Y$ww5qd9jN|-mSh{f1RcO^8;v2CLhnBbK3s==l9bT
z;2AF2sXY=lGmzU1^vZ~}nAkCh@5*&g44}Umq(qkMt}*2gYE{7n0?uEFhF40jY#vIz
zjyT#FR7+dwow}H3fko@IbyhJO&T0NFXn!@7)r7s&cR;u~(cC=ZuroW<q>tP}`*1f;
zscGs<4ihiL7}C<S9Iq(75z>+jFL}7VB8C9_^&CxNf-YY~7bve<^^J9)EvH#k3(~G%
zM@%<CDViL*3R7DquWvvRQvp!0*b=O2RU{9Z7>V<oypbiexSa05oA_eNum+FhW-O@3
zJkK7VXVniJ&8Ck`$b9~LpbfnTEq$4)GxyNLXM$g!xJ9;&(A7?m>jgqK$S89Y_(2%n
zPTE`hyF+)^zPh3%feQ`dA!LTGZ80<<qd*$%xJh4`QWJ1^Llcf@vFjnh@owzZv(Fw6
z;(yFNj_bQw&@Yyr2OfHD^{LjU&^3jvoSPyY9fAmBK4oSx{&axIk3{zkF|R_HeEGNB
zZ30lxFkbJ<Fh^s3Os2qWUy-&$awbRq$_tf7Kfy1TU9;5Mn!sH!!A-O`WmG^aCWtzu
zlCjY?UopJpfpC5jH1keF-?V`S)*#W4M|IE1^hr|gbB!m74xKjBmwv&L|9Ln5e5R6T
z&rJ4FgPLDf1W^ZImtDcgvaI4oU!+X4D7)!aS%OMcJsN^I>SMNf&nJJf!;&qhOGQA?
z3-`iTW4aJhRO<`d<6&1oo2=Fg;^vgm(r+akWbSEc^=xE{J9GX7+)T#^E&C%!Il=;m
zFP+)6$O4y5|0SzI^fRsnW!EMI%3e4zr>&yPS5^LKZ@c7@`iL6;qXbvz>rMqdb{M@x
zMoN{EeK6)LglM)o!AkM+NZ7=E1}`7!m%xa6v^CM9-z&SgYo-%I(BbcLUy6NC&XRJ>
zt234GFwG#(6s;|=z?IbWlXG&wxkLAohMwMzg(Tjx!A(<%>K<Tf{i?mdDDQb$+<whc
zB411)+er;QvHhiEh}P_WnL4`Np^Y!^uGDBt>gL0e-stjH&$i8HNB7m~h)sEO?72aT
z1s?R+@2J|ERf^%glk)xb3@cW?5eT_J{T#{k4pbI?G|lH`DfblgSDd=6@AQv&7&4XO
zEvuaV)6{K|g61bHk(@;vPe>YV(S;cC4O_xV=scZ^NFpu1A8!Nr64?{OybZF<mbIU(
z^v{@PJ$JrszN1uKfuXXVO|LBxU!qAo_R5H1JmH`sKJ?`{5JuB<7&{#XHyqW-#-vl8
zv`(S&JmWrO-pQI&j^wpesINgdr70~@((_PWsJgGrGEM-v|0RWJNRR;_5r5xO7|79<
z5IsU!SR_iF6yWPaqr!gbn;M;!EYouOF*;Vx-f++U9vWOIKaJrxNXjt2##@!4zb90r
za~$h)8P3nE5`#&W>SmvrSjo_??=Kuc9aOu7p1Y6)u)lLb#tvA3D0~qH8ZmxEEnKm?
zTJ;Shea2VyO`&8&So==elz-Z)7#}KV5w=KIQ4k*c_6*2MZ*)Ht=?$L_OPlnQWbU7V
z@F1=T4+ME0g8@~D?PJOq$a%6cLJ@WzQ7G<i!ieJFgq_DmiQro%m*3kNfp)nZvdqdZ
zQjcph0luulC{GZXGP>VekwUgX+pgrj^@Kz-_PFIH#b7}{*a0^O8$1q<F&Nfd7hQ}~
zdV&}&G$D4qYn_E10_nM^bpJJi<BXbwU+87yUh4^xbtAo1Mmy`CP1^2<()HB=Cx})1
zwaWoGZUT_<_}pc?TKh7sNarjY842mw(aDgg@dJxv+i%vPUCnXy45gkIn^rj@Gjyka
zm3npQu!ZO8Bwhtl4InWiCy5fl;yx=XbW7bxw#YD~T{OF$vNyLgdg9b&Jsw|@yYY8|
z0%Q#KLfxE}XMUH>Oy$?^@$?}jA7mMghy5UN?04yJmpquP0gkOO?mz3I-MF>??h^Bi
z6WIHVx7>^FA*Hh~F)xgk3s%?*l*a|^8Fx15C`F&lAKAu@Y2PEU(e^|5mj&AYJ&vmo
zW<?!10nwNPPB1M)5&+@^-!gn%=6DdW_`<3&rmhrb^-q*~i{UFSn>|+BQ0{A4i%Y3c
zwgg{Cv6yAjLur?rn56kP0iOVm8SfaIEguaohG?`xE@pbJSulhYjfuRUr4%F*V@8TG
z4BpHNpyC+s^oR6iqNGT4D`aFU;0s1+AHm5rPW!aNoQ@9{PAGf&JKC-lr~)rO>)du}
z^3l^b;%H7l-|lafl+&rayflFXKF!^MyQu_EJvYB-5y?C*aQ_21irNV1ANy7X3ULw#
z*OHY((BkX(85}u3E<(v8G_EE6_H1FJqJ$f})RMB(!tZ80U7qVYJ9`2uGOr>A`LnJM
z+%G0Os&1a!?uiK!NJx1#4}LWr`VSw}cBrERW!yZ{_$43#%PBQTOvFsTA9T~4E?@Z2
zqaxu!C&W8-GM2eFPL5{gjdh%FWn*L7=4q+vloDuHOy<-Y-`~Kw^eEsGByitMny5;n
zAMHxlT(Z5ZU<6n3(s;OD;05}i))<qZo8xM$@d1hp6Dc<X!N32uG4(h@h$hqT40Ipo
z)y^yJpwGNV=p}<NsT|z^S5V$yX7|Zwh-Xrx+j8c!*FkixZFjoqhi<fFKo{Hrt<e?4
zPo>x|ud^C%4*~a!i4yn{?hscW!GS>W@SQ2zTP0FUXO<$BJc_|iX$W^XL_DV?wzVy;
zH88(~kK~uC7$g#MgK>%9r5>~i_5vycY$1}gKduaabzQQWY>lZ9h|gb~q_Z<(S|<fl
zPuFdJWV@+ow60EP(Yo)hOO{{*?-F`GXLjwotzW~3D}1M~FHMY;FSJ?BzV%%dofjE*
z9g6F(mv|yR3A2l`o$bv*m^Y$a^3~pq-?yr!XV~<7ZcZvEFt}s{Obee4H7Ir(AmJkN
z0{($w1nmpL|IoW^`r{A&hpioX7L5zcwg!R2ejN$&!;hVkACmK4B3>`+RONjavL01m
z1*W?s7;HA(W@K?kJU8_6<((HNFJN*oCs|uD`YgOGR4gPUi_z-pm7d?#G)p+ioW)0&
z@1u1(fCT~36`JSmB7L;dc$@Fvt7i_kw3G8($h5=+FOZg*LaL9fvVA*r;+#-gfbIxS
zs|5^GFiLBSHRC9nRyL7`_g>#pHlRck>2vbRjHmLfHVGEB%lUZG99y+CDzoN_*~*o}
z-8^K7Y1g=R(&M7kIYf{3MtF_oaC#G4DdQ?8$r^TTpV;w^+>~8b@{9uq=(2iJH6SO@
zM1RHcg}mu~<~RBt{yRRpjc!dD){C;`7XmLNsnB&|+Ew0Z*Yl34ywIQ<r9fro>4uOg
zGjBnP){$ySrYnKk!C?%k$}*?hC0z)nI3;F&4WYDFgzL|68Li!F-f-@@0MVPW0fBmb
z?if=K9NHtKfhYw6saM34%6Jz~>ah8*2d?{;fcaIun?)7y*DK#VDKt?thN)6(3|62@
zG^Exf+V153McYg0nBE7%tsW0{>Jt1u_;L9WD<H;c5Uj^^OI=$@KUbw}ExX4u=a^{&
z-4beveCZovYvdKa%6f}Es?7lZ!@orGhc{~~TZRqYXuv{rln4GfMo4GKu{oGJm~9<%
zR2n8S;~nIwQux)bUQbThG$pL!n)<{uNAHhsw~1ZVcNiupJJEYY*)KM5$)zh@3L1>v
zOOqI;q|#buciKhci0tk^rIrh5BIKWWvg|+bUReiP;z8klrtcOqG9O>Y5QYo_et*Cv
ztNW^qwuqUQHS9h?$XE29V*zw^jO50r4`j8)oDU{q2~<uT)AG9VX;|z&tdyC6nC1Qy
z`Y6S-xsiRq_;9>l^FzOTFXL}F?|b-9D(Tj((B2<_Ebb#6J%}MD&o}uJIkMdJ-z6V-
zhRSDBtzr)8(~1xBYQfZC@f5j=*lU5H>c*SCVrE69Hx3jXbh^ZUrVC9eMNlchgdH#T
z7h&0;$;#FlIngm?4=Qd$9u>{jvDYqBejwlxZgMGdT7xF<7z`0ai3U#T?i8|%S0p=w
zNmRs(CBk=!rUmKSpJVwA(kD&}5YigT<uqBFPwC8p9m~lC@t8%;nG4K{ovL(K_0Ka~
zwp(p7u3xgI#XonpD_%Oe9wxZn7e49IgEjkhJ>z=c;#S#<rW_6YZ>W8YTFVL)LJbm5
zYL>0W6WK6bh1qKKAs0RoZgoO-)8=F822YLQXWm<3Jgo-!6@SMH`+h}%36ls7;%(Au
zy8kN6kCu$tHRX^c0c*a_Bw&8`30(<%onCWC+#+@}>~KZ0uo`|%{waGU_k=PDgKbxz
z2(BI|39C4j_e%Po;1XVhf*5t#?A9L~7h5Cp9^csLyo*uylcCTlED>0w)lQ6@X1ljs
z2=*%tA}>Gs2Yygyf4z~EowLMV*BMJtWlwP|JYmbPMmAye=s==@LEQW)Q5&g0*mVC(
z`|xLP|MUz(7i+zJDeI57S$WRK--LiU--%di?>})v7JJeiN-u|C<(_v{irPPn7Q?LM
zyNC{04-4@t>Ae_86Xg5=N&(b9>;4{K0=*vs?zLfj@h$rfoC@d4`U~;%m6w0jB7gH%
zp5`FmLXXG2=|2?G??osKhvq@P5jbharWi=Dy@rq^7gzt4B$**?cOe;uZ-!!LKiM$K
zMNd})Nbu9~@5@`i!Lq??Pm+3VPErt_5}hmUt{Qbg2CT@Z9vdneWQ=uneaKN8AYic$
zszSOM&lo4Ae=6<&Aa=^4fq^l80gCxQey($4xkpnB+dp<W|6j?Jl6a`A3lLfmsGWt;
zsz8ZGLSYcqhPDhXIm?g9PgO7T051S-$djYfaq|&TU?C)H!y#1Ib|*3FWi!X$Ql|fy
zUB0hG_?_1Vv&Q~_1AxMV+~C05GBs*N5>1Z_nalPs=%4!zQ<#1@esyTbxLrVQb|~v6
z{J#zlc`^gxl_zD*uO`f&CyOZhhx!sqOmM*)f&cXO!r`aW|IiEl3lsKV={F^(5ZQA7
zHJ;Lcm}&j%i~j4eEIq<747fw4_kUU6|GDJ;`-=ZBkn2Bx_kVh(@*F~ttlpcs@joLp
ze@&JDmj`(&u*Ex+JyGTVKV0Dd{2TAzAr=86CuMv7ffD!gTKy-@)K&wbrxt(Gwf`@E
z=j07?B&5k^d2aCK+vN291X%;N?}$Yv&{D2}@+I-dbWcu>-mA>oc!$R~Z2vxP;^%iZ
zjj%!laFvF?ZvCglbh7Mt*h#3~&GzO!LHJK%z{NZotdV;h2u!b#tXYMgnZEpF{vsl)
zLZkgBMljYz?EQ-29?9vw>|F5;z=k5grCY+;xySm{r4Gh*7rb^s6)T@^JC~XVM8`89
zd(OuoEQA%{@I7){0S=qd>bFk_sb>Y|%e^#GHeeU`kyk$16Ih>*2O|7_rXT|8vv;|=
zdh@mh_1x!xYj8MLfz0f|+t05bNUYf<N^BlD)8~4PhsUIbVlmcX+RHizCxaQ^(Z7=F
z{CD*4x_ThkeXf81IQzlRbuJr&`3_xDKNLfkN^2uD#Eh^J(_PO)GytI2%b!u_l-^#x
zv<Xv3Bnr1R-4H~rQ77F#`-lN+#}DUg?g}9q1HctZRNF}ukAU;K8x7689BLRp@&7n`
z^LVJ+wtswx8YxT)rN+KwjZn%KV~I*c%D!)rM6xH5u~T-TY=!JgcEwmiMD{g1W#8BQ
zj!)g!^?bk2{oL2{`u_g9y5^N;obx%)^ElqedpXk&ZgbY)T*`ks!|Z_QEaYC~LW5rt
z-yc)C4%KK8?EXQ-wX7kd%h?rdH`l>ztz&H$rjdOxu_&n)2TDnTIK0t1Fc^Aa_3qQ^
z*sF|aXV{bjYJ-5!{V?dDBmNHfiwQ#8ye{~em5G?*q#ywyP!42lK@e&W00j5YKCHOo
z%){aUX^Vg`x#PbU7LLQHsEnutFgJZo;jTH=;x7F~I4;&tQG@@f*kp`zLhs>j<##1j
zCwhLLN&j(koy$cs?Q;u~e=z;tKwsI=V$icY0BOyKZVe$67MurfOPU`y3W^}mx(**v
z1o(JALi)Ay%9Lgwa~gxPF9*ac%`HGs(2jQntg0{Q^1L|`rNs0&^x)A`#)!p^m2&ew
zlVTqNalIS;&DQL?gbZZrfAmJdtuqH<n&+6816v6zkQSjsg}Rqn4>Jf81np3;Zm_o4
zhKKjw@RNNvo_xRA2zIYeumCl{2OM3!iNo%~B-R=sl&#GVUS22X&~kic?x*0`@d)vF
zo3la25}&A<g5UXYXB8(}0|&mOdl}G1{Q5Q0kD%7>2K!l)#nu)*Yw?Q<m4jXRb={)d
zba~6*!EadOL&=~QJQvm)eG^EVUT>s@uQ3Lfq>lJJZl~uM9~izG`tHtuTwOSD&Z$5H
zbWuelgX_-~bT0Q8<FP*I%nbl(Peqbiih$AC14LrIcYM-E+g7Q5+-ZFIPPM@%PE_R(
za~9?(PjJ{_4;<aSa8A5aN0};?$KbO4^w;O4A3^Tmg>vjM@KJAcFF%t}h{ZhtmZY~8
zoH2!vk^BB;CBS0sh?5^3WiQ?iY*J0MW%jwRgM>2`Xs~<3I2P0|h=$IE4^fW*$RT))
zVZI>i^0}Q}qm*mQec&nZjiuU1oLGM*|K1#R%EqJAJ%E}95He9np$0g-wWt*==|;2&
zb_Z<u{d{^x5m_1;@{3r(g==>Ej_<Otf1av;KX`w?qJw0S^YyQ$|KV{<aZ@`&18t3n
zRTB*{6a2Av*f(KE<pLE=ggG9`4;O(?WX3<E)(_7{A*S~x2tvLH3Mi=FOe@5uq3~+L
zFjOM`YN^WEG*}6kH`8kxzF&`w45?9f*CR_tVJY&!ste*%#~w*io_*UTA)w2h>I2+x
zdoE(9X5qMW>7&O}&wtrFdR*_rf8CM))6Ls~=n4W~+eiQF*0pCRwkGqi5cQ>MSk4B3
zUMp4w<;#vIktK*R$HRf&WEx|^j$nM2ey=yc@3aRFE+ZshB$vlF7%M(n2^^(!MR%nm
z*nIPSC**F8EWq<#mOK6fgmMWL$vM|?7ZOi)#iW1M4WBnF831Vh;tk%?>BL-Z!ADGn
z6iQ?eJX}>M;kMlahuWD(X<_ot0Vbw;jw7Tla2gqjsxzmjY)}XC=>%P@h;K#MeBj(n
zX5?Ebgt#2r>XjzA!+-tT<Ypw4Od}p%Z%s*9Q&i<4)jD-FTkFGp5pdQW%g-bnE5r4%
z<Q@@z^qhZ>JU$tsrdr-Hv`ziETG%y~6fBVk(qqnum#O=yX_^VI7b<>j|8#)!Hzkrr
zA|JdH3^XY(GIso$e_Z8XsZ#u=@gPQY<|n=+fD*`NPu&3wvwLh#sF=OT(l-D{uvtzN
zRP6qW{lHu`k1R^uZ>t18tv!w|1&aiy8OBE;R=*JVot1%gQV@k%ZKB`yjf%AfmjFTe
zb03VqA^gkT!pP-U*_<`El*LaNy;KNTd<F7h=0u(a4@em1jPOal)_lJCf>SC0f>xsL
za5P6ft_SJl3g1q%B+oDcB-WJ$iu!D*$|ocBC{QuQSuXLfoL);j(!95re57}x;d0^;
zi}=8D|8FkznerA1$xqH}2u3zTwA~WriA{<{m?zhl*|uNhC<IicemA=1Liq){$mG?E
zbr=cBOCSXC0NTDViN=Zo>&&#alGM-*r%aRdsZzPBgbC4&?ps6c_vq&|Qnt+hxk7u5
zC}sN1lDYrtMZRI)$)a!yQe2`y__Ts&k~`yoWV0@SxIm+i@(l195P~7H&p<v%vHO5Q
zN=q!$`=tsi6+tg-8HSixFy*Igytp9Qp$r`L=h`@-YOBQE-`+PiXtUir;t;1i%jP|Y
z1d}7vj%s%o`cC6)#f`sMmGy!jo@_Pg%|hu@gwRa+Wc{w(e0_u7D(C4Oq%W(>Qm|G7
z62X(0!KBK(x^zgHE~k8oJ=RLKK>Ix96^ee?_0sXi47Z_g@<1w_SsFHD;bJ#<oSC?o
zC(eMYi)?q;d<XYcm3<`Fl7wa%3ZevQV&+6+s8f{k`TO2hA#OYCof>-Ho4N2T_QN2;
zQo*U=I64&12x0DFNPZo1<(?QE?t<hk#%f@Gw6R|A+$LE*JV)Bp?uAm=Lu@TVCaF|H
zKAGS8afPG>DuAH~5RtTNk?$a$t2|47-=*~owHL|A>*|q^_38dIY#f-oD_><O1J`jB
z(!|MGDN05xgRj6UvrU8CyVZ*EP0s2?qXS#PB*UNCV|l#dB!4XhtI!g1HC2;TZ-^{7
zZ{wWxSda%f@09~rqblFzNoEzeEQ2frGTbMH^*)UN=gZ$IHutJ@j<lMVX|#zIFq(rV
zCT6DaO~{QImTURx)WtSH=p-mC0-PeZ$U<q)hCBI{L^?$}d5GG9te<Hhbz@@d^Ukx0
z+WMd~^ed-ZP2)YzxP>VsBVH2tH7AJ=LfLY|wH6YAU1Jaq74*bd$_<QNAC*D7C+#rF
zK+Sk)e=YmTZy6184^RJWW<b|~Cby&6yZpkYJgR`h={=aK3JiY?b9I^$kQdxw9$TsK
zGEYsDit2jPZpLDFJ!xS@vDIevPX2J~a?B1r?-OUeq&=<*y}$g%|HHje2;J?S%iKDD
z>0slYOYstahxv@fz4y7#b$}W7yA2FKbM<M;Q>cgcLt2WnV93bi!gv@{dnnY8(-N`Z
z#o2K;etb+NPbGs4%BjW?O$`;r)~9$*N4y*CUpq8g*o9a;{<+Yn;C)wS4TM(RsH@te
z3(QEF!<?th&5~kf)n|&@lH`5LA3*vR33p?lhX;ua2?q{!5pxtK<1KkGu?{nNR*sLd
z161Ef+nY{qoWq!O{6aD}$<+uQAT`VQC1zuUm*B7>3$=j*i1MTrkUGgezgfM4zJntY
zn~XA&zC;h<p6aGE6>uhZFsK}GdWN&FhAU8*aMUZN=-4A!uF^l`QzBK2+daPEO?fSr
zG;MGk##mE`hA21|COiLBi7^v*V+Lzbs<Ync5QUox(H+ln571s@a?pM7iS$I$WkEie
z&0~2mp?LgNH(#oCAeIo82XRM?jxZl<*IT}UlT8xGE6C~S2bqh&uSO$&xWp7Z5#Dp6
zRO`SkK&<YpTNK_D_nldJo7^qdTxZ<B;o0#XT}xc!SzzEvw3&-9W0TPL&?**}5U;4O
z67WQ~%cx7PL}f?ai~bf`)vb9iGb9IZ{#mT9JQzGluQ}mIKVOvU%apIq(tO(N3um+t
z{bVW};OuoYD|}<rCE<8i1b%BC1gPSEvq47^OFNP+P_|S@557}f*Xj?B9sP_U@)I!)
zc&uhQsq-o#t<BkZhjjrx`5EE6Rx&IGQ*P>B(PErk;(NQ#WV{^bWwxxV@1FXlM_!J7
zlz(&%Ny&?u(z!-ZbUfA;^|%RqR;>Waz3#8KmOoUI8olAvDc#}39K2xW)V%kIrfhT~
za6Q8>dgk>%w&PO$B)d<}asNw|=7BDtdn%80@)9<B37V6t_v3t%tXKoCC592|;Eykh
z%~c!%`S+rV2SSTKh835G6fdgV2P*OV%9AMV)^8Hnst8G?CxC;<Qpv{0%Z3f)Sem%r
z)KWGcS-cund{e`L&qW`=Y(;IMW~`-UzjvwrSm>Z-V|}D#&`-DCF+$QwHa^tx!KtQr
zc0LDIK9S>RlpkxKQ_z-E6!X*{Re9H5<E>0fe=KtSSY*|U-vrcm554VXp7Tq-Dy9mZ
zuY23{(6@<2)=I}uK|9B9*N-nNZ;V-Xw5js8X#Xz*+KZln@@MbukLqPVaN?Y|HQ;oD
z>5t$1O2Zvrc{<4Dz15L9ZA^dpOIOm6#1$2>OM0s|da4)S@S9PKX-6rE9o8N7G+S4*
zIzLILRk>>ZaIUbM9ow5eeX1^&yy!DjNjn768iIFq;>wxau~3O59Wg<*FTF0u)6My}
zv*fdsw8zhuc2C4bJu|n9Y&aILqGXw@7e1+_oY0VjjWY=lxvb~J+D51y@KTOs_RlFb
zKCx0se!ERWY}=%5D`>zXS-IH$lu?JthF4|Ob-qLnzASd^iF39mFSu}ZThmTb({fNx
z7PS)`Z4E@0%-^QGeK^=UrZaF&T)Wx-*Jz@-LCaflF=v|*oe+0Q1D;ZaSDVXICP&1W
z^|coST}KnGRJVslT=3*3WP2|+$oC2>)NHC6$J~Z=qx&u=)Nka^o<9}6lfQ8xfACtB
z<zz|e-|DLW5ldkvN{r(CKc6Wz)Oh{eA*>_guaMIrTVgRf3QmaPZ8UVIk$_J3D<2Q_
zJEqs3Z~0~G%`e1F$dzosFOTNoGx<qhGbiIxKf_pgFaKpVH+;f05ApXLgBq@`r+GUs
zqL{vfb=9!&vthT+;rN1S?l_m`1fwRE(1)^BqXYb@*u+#-J<G3-QZKc$VAv5qsncVU
zND?mk7&ay2!TZcWXVx^utqyMGkVArW(yXJ#DdTOU(tU4ix?*B7<&%iN1KBy0QCnTH
z()B?>EtJ4|4t*?|wZ%*C0!^m!4vFO%OAoqU9iJVc=)I34UYfS>m6yCG)~Fdxh~i}p
z*aq7oL*;eH!a2EZbqn3KRbR0mZFlR;xlk}JWU6&qt&hrAnn(}h_so;e@UI*Mn(mO1
z!n8t>!p{^kgFcofc7>iXmZcV)>I!Fq_v6n94Vp3Xv2o*1I6!iToRa~%*A{Uq=$@8O
z*}wTIDgODnR}~S~BvO6fl4c^}xn@~twwFIK>3vw+W2<-rtWhuq*q)V>T-kJwaC2a>
zd$Rruh>CL>AeP(ur!GTEgXuM!M?Xc#yT1aCzMdsZRzl^^8k7d4KXDR>7&h@$iD)xA
zs}+e0m=MqZ{6y+Pu&D`>T2Z2HiVfs-DK(EZ(b1U|h}1K@+Aw=H!gp=@p@?P2yE{V>
zDOX(9ew<L6Dt$$*#N-^7&`?!yq}LU~kngR$16A8cr)1|y7n8f6KsSCd!vB3ybBcVE
zpUKT{DYoAp=7mkcB<Oh4WyEWHX2dUn$9J}Ep_zrdF9D)z67Sui=_dVjC-o5+)D`3!
z)yvU4shwRx1KRG-8bO7a`v;oveskb&d}sW2$V64?{s-qs<?&zFFANI(7*PpO%=dTO
zPE$&m#7xZEJlWBzy{AYiq804CL(*#Y`DWysX115@ql8MOI1$9M%&HYZq9aI6wg%TD
z2;f2+an1CzQZVP8rMph2jc2a?Alxz9?S6qcso$CGH)-)}@-SQ3{HeEpO35Ifz|oXb
z6n<(3%InT=9Lw+I%}yqLY1U}Y#fG<QNBNudD@CX7U`xJ<S;m#1=$c61)?Z%E50zL#
zihAwFaxrK(?_`ykwZ^yYPcxK1aoou9vXW<;Bq}bYI^EQf=f$(2HhHk|0cgiyn9o^q
zM&0X<X?gEs+w*IVK=&OLcLX=0D0yMtj+YCx<&;s>n>Svt({7FqRxkJU5+0+oKv#P<
zHHz6K-8^=pd}?_jevg5Zpx5^K1rVpybr)7%Fn1KjcEdR#L;wGDP5|bb+o^8$8?<~H
z!4Y!UqrtAS*{0*r465x@($Q%7b^*QH9})+ql*VG@4+$lUBtKnmKc>Wk=^lpZzDeY=
zOu76|fnbz5mMHrH_ES?NNg>gJgLH<k_<J=jYLvF!C6NlwqCIxiVyj0I9QTh}yn(2R
z<Rvykne-*RU-*6A(<{>#nEV`Z@2Iv8bGWjcpyaFCeYk>l7J<kddI{HJ1c97XB08i;
zx1Cq;DcT4s805(4WKmAuU~vtgztG}(w}16!FUQ`2_&6O@a)n#kP~Nf~N|gF)%8ReQ
z`WsNa;C9r@MeV41NiDjQ91G7@7~i)r&&QB?Z9QEI1Y&FhKDqSML3+C7DETa_{D8pK
zjP@fxXZoEwlJW!E)KFH&m8Af68M#|16Q@*+^usB!O|mx+f;wS6!!EIOgYuQ>jAjA+
zAY!+Q1cRO!_D*q2A94#`H^UNj&}oS=5Dl@I#_(UZ!d~a|((NM`r~(Jg#=+*Y;$G+R
zv#4e9NdN+@KAhc@KS*y=o{L~X*1B4)+Aq6z%D)g7+6lHUn9U@?1@O%!cFU?rPk6;@
z@E-%>bP(BgR?m<0{683J6SV@Vl%8hmZB2`m(jP1EZc+RPEPjjnz(zLUTEiNeBm5X<
zkgC!(x;ErA{?v3+`SV#sN!xeh^4xJ|GE68}%554HC@<DSmR(DRLio5lvsZtI%tF{V
z&!ub0qRjh80k=eABdF@33n1SCKnW?s<y^M~-%l#FKY!4$q^T<YLM>(YM<}S$*if4I
zY_t)sHbWSRGh*WHbat!g7G{1r3k8|n_~r4$c^KtvHPGKL{?w)SWRFU6X3RPTImq4D
zR_Fd4<mVw|og34q$Wj=dcilVdHjH5|6Z2N{9=o*6yug_F4p9tYh(fkrn3p(lEVFU_
zM3|*^4!@bDI0@YLn2t-=T{G|F{cXgIkM6DN3|NV`|5EMCKIMyQa0xES=N+NTeXpD#
z;qZ~@Pyu&n_xbMeH3^&4go(YqfIK0=jSIJYR)<V?)MJAGmvN*e>vzfH|JRKSRUYe-
zhTABy{vg|ig1v8(sc+LIU-^soR@@`EUjV|&E;@>5Y0O$-v!>Qt&zZe=$n|8|IP|TG
zaC?s`39=*MeYFt6gILaEGjo1yIV&Ub#7}iBGadHa6J9CgLF(C$?+Pxxy8%kR-Y$Yk
zFhSNU&u$HS2Lg{?00}`Ps?;jHSS-RZtb~yAf)HtUL#)wov@P@)ZJuRIUlJ=DcJKY<
zPbGdafDNE@>*5kuf0Q8dnCSfvryEd$I+8SsWPgsgD#EU<lYwg}%}e+wyjlT?*0iqh
z*HJu|T7vkOUOAF9do6vt+a4uglvs}4%ba9}{>-UuN-SNkY-1MsEnj&lCG8fAjXvbN
zO@&Lk{72i#I`i`@;r_1#7|O!p9KA|alXF{lKGjxDbGrs*tJx&z?>edV5o(+LWbMzM
zLu^xxPCFm(^XnF}3n?l2!y*;9pwZOCQo>%T?Na!y<g^VG{qo}4ih;ji$yMCIfb`;g
z(>BeS#}<NpDbuePHYG*y8}H@qdBoKV5?ZEyQ6BLN(&$Tjt2ePFk*yY?88orB$FuWZ
zZt~acUZB-fTlOiZtoeW99RO*}LxJf36!E-1NfeTv>Zz(FS+H^-i@z;qQ}FH86ox(+
zmx9A=p2%lFP)wLmn}WXT*+`!!2=)N#;(80dY7}$~K;L9Z>;YTB$O4Y5zdc0ns+V68
zT%BNf`5W{Q)jKGVKCd*vjujV3eszDaR6U%J+csVPMY{bH@nxKI70E}|E5(Zw#WF{%
zmE&=0%lPQCl@vQxP&GB^`tqNRw<?*4@RgrT*DOLTPwPGc(@JYCX4V(Jm38+b<UH(5
zIv>zxN+LDMIN1-PTM>i%zkns^m&4rY#9AK}9GZAJAGHayW!6eDw`#LOuWN-LiOE^*
zN*CiFFDqCx)nfsvQO{qWs2JP(+D5E@eqmJXN9bFWLX}MV_~m+V=hP-dEU;p~@ZE(z
zR3^e~-ip5ycHEYPO}4gmT<h4*U6K9&r}YP9R^z@~=N~7N*AwW)1Hzahe8Ee(OeA{f
z<!ZL3P;$H9G14Ofh9FV+3|1z3G==0UbBO%8K@y1tSq6|N!W7fy?7T65rdlD*s-AKK
z&sAw<-42T5aj+}mERz&}0LZ(yu_KI!*;Deu_+#mqKO!C&snem8wPxZllU|+(do}@E
zS99HUdF&L9ju};5BD5v=k}2vvkaozVP13@o!emKUnK$MD&-n~IF_jqESx812RH4wD
zyl@xHOvytt&AI8AbDP07LY-8)CuNUukM-~%RHU~Z%2mhZ`*<q&_rySepL7i;rx?6+
zH%&5xV@jk){~3O|L4^2*NJ=XsF%o4G#(26DSfCZ{yLn>x_Ik&Y7w&7n!3Lx#`U>EB
zeM^hM*reX!Y&PgBC75DPIWc>O7l|ewwhlCZ1jwcAPNYPzfE{2=a`s=um=%iDX@9^>
zryJ+wd#cfWJzK5d>Pq4s9tSw#Eq443+?(d2v!`8qo5z7iZuHy#6IZl4w0hopWp?Gq
zj0Pn9H3=m+Z#!VuUZ19R4ww|s6QD)hAO!XqKs4B9ik?uAUhRaHyR^jiMMip_ljvNq
z#M5^Tn=CO*#F`cirmi|JbA?QP$4Uei2rg0hNW^`988o=Z>qy5=Z~5Bh0vWn);ER}4
zRpz70j1TW|Eq_k=sGlLj7UY#m##4<##EhYy`6~y(UOQPS$y+rT3?IBv(@N?bm97sT
zjGy^Zwg#xz4Hz9o8-p>)LeB3pzQQ`HLIQgI8J7U}?uLN`US*d4FY6g{@o6sqH2DTT
zMidF^m*i~t-+xQf!=s7b0U^S{S$Get-YO;-`2`?w?-ylWMtp-O*8BU4fuJ%A1@+q!
zHA8x9;J@gbVl7&<OV8`6euV%mQ!N0x5UI_%;IA4H;qo)NP-_477pYWB-=f@@#LPC6
zqU?yOwYdanUo}sV+YxcMgqw^W(#GZhDn(e?lYg>ea^efiVD!F+72oZiv0%yjm0tQ`
zMo`V9hsm}HM3R1~g`Xm5mvdf<99){#Z~&@jl)PR(3@^^pZMGLqO8b^*ACD}}F<_Ao
zDnO>m+DFa><IS;4t?yi7Hn~XU_3pQ8Kg<DbdaRP`q!Pm+Eyr!j?a9$_#Kpt`BFdu4
zFV4()D>s6ACqGt6-2tLAW0j8C28ro(9HAJWQ#wIAfAxa?4N;viMatSK@Z2y5L3K_h
z#L>yt>%2avN0UF@gM>8k5uG?=cUK^24urvzN@r+WN<Nyw=gIg2LrHI=0VOOSeGg)s
zgbi<miIS+DK;DDvjcqBSTHt`<d7`^S|8{X|`@z<w29r-&?OW|ed?Qg!*V=q8Zh%uP
z$p;uTyj^_%^uQsxD&C52HkuW@!z89XQ5^Ev?!mKHoA~4b@7(pOH$h0~*?2DS7^CGS
zj&xbLWA~cGoQ+m(E-p@mer%4k6$wr?s}##Y+yEvYnw(3doQS($z9g4)^lF&TB?J}!
zgRC`1eh*rvJ+u3$URFM7?=1ITe5Ha52!U1B0c)&(HKG((SDfVD+hJAJm}GM+-t^}c
zEcL?xK!#N)X-|&reJaN#ta4_zWjh^QDQ+sMJDm$o3$gtBExaYEraL6wwqz2MhK)_-
zb4_pMtBmiD#;t=yi{8x+wtAp#Oq|bhdVsuTKH_7$)8y@Jr_u|<^lx<i<@=5ac?+ZO
zw~qCSK)MB?yl2Q&!)7bSjrPJVO4_L}No%#2!xmV|Yd(PK!Mke6{xgAuvNXhbr|tI4
zPOQ@sXuU*&9qwU;h5CZ&krcxf^HPD};#t}i*5R|85>ms)yGw4*9$%T+@!wgQ)^Mt5
zJ*#{^t@k&!_9ZOAC3Z&Ri-gw3KQU#bH2#}pAdkfL?wu6{Z6<d0bT58ix+(5BZt>@L
zCRO~JlUAeTcgbyo8T!N%mGOtM7WE+sG%!MK*D_)VJxYc)_`<SE2wr`~s2Aj7pG<|u
z9OEYQvPf;v1bv_bS`&dD@tz>dqGYONft>aV@+%)%c)j&QhcxE$Zxd0Gk)Ld2nK+&&
z{qQoZpEVpMzD9Ufw54`zDq`&J6>Oi-8pc-rOg@@C0?*MAsO%+8ARR~l`mTM}si2ms
zDTGDru>Q!gU`unn1=fPMQTyW3&x7S>IaD=yEjraTgpG@>q&IP1?<a<z-{iQsHuLRm
z2&wh7O3C3Jt0Tbcc8#8{7T7*zC=(-gSgk<(g}hdpE>Xz(VuKIUs9F>4Q<9++vJ%P?
z&e)I>8*u4eBuPMs4MQ+_n`N5}UG^i~yurN1)Zj>RZAIZ)+dTSV80D3YBnR1HxsTU(
zv{h2C&!3|~hRY6iS8m|8(<4HR`GFcAawqhG*eTOh+wBW`F%G8MjCx>vu%iz0Ox|1P
z*A!NwZW;--Zuymdy?`m0saI%Jc!M5aC9M-STIPSUrR0P5__9vc*ed4F8o2xX>n0E=
zWcQ7%mV7S$P^vxE-V0sb=~3Q|iD-Qv<3Qu4GS97M7R+<UmVBp;tz)%IK@%l~EsxhC
za(}nji<~c*P^l*IU>^32YzP1J6$jIs^ENLwKM`7A|K$ZBf25fs8455qR;i*}L?J+z
z-FM9yD<fMFo>baq%X<Vp19y=bcF*tO*C(6zJ7#xYm7lt9aq`rjdqnrZu=!R=cC>qE
zv)!ZYe}a4M)nPPOX4#Z6^oJ?Hdlj{X=k!oG7~X#E{Fmd^+9`IOK}<I`kGWygsjm8R
z%HeWWQW&`pi(5!&`=9eUA{QpXY=xOai{i!boC?R&Ec3KtC=)vT8Ftg?Ju%S3j8HmJ
zsib$QWuO|)>%rB?OcYYeC}x@zg|QakMyc{Mi>fQ)50CqrlaLc638%*SnCWjHOU~P(
zN(tn2$IT`2Sd4CkJuBDOI*Yj)WY+ZpkBgEMy_Wq}`4a=*99v(p+L4sO5Wl>%g7Sul
z)j}aA$77!|D%7%33ht}*MP3Stytn<R$Yk!A-m(y<JHKl}U9Ft5F8hLom~KwK>a|T9
zL&HnKn;(J{aXGAtW-h*h1ssg3ly#gcq9ZaoCnIt4$=FWS{5?O{*t!1RJYmgXp<ps$
zxyy!?ia))q!s2XPul|fOb9GwBw@2tkcAJ6qYTj3VI{5~VW}Se?II%k66;|oHg<FEz
zrRVfmN>yk8{Kqdz;iOm_XSG#+!V%TG>_~e-lahIj**S%5ea1CJJTguu-;M_paEnNc
z8<o1Xbhl43e<}}zYZ+h^G(2S?@x@hSdyl8Jq$+y(LSJ@_Q|kO5`R%_`z>rs-W+@^?
zr?pe(U+heFRpK<1!nknaK_<pyo#|?hu?iWIDr)`{*t-u!!!f>m^2wacI%g5;0T@^=
zv0Gw8dk_jnkXzJ4+NXMCYSbvwKZcS26Vm(Fi1)vZe5FV<@u=m)O)bZs`YJra6<vZ>
z(lD))CS5L)k4t6p7evI?KkV>ewEPb3eaEC=q0Xr&E`DNBD#jNB2p<8F3*o3Mzy^%@
z#XX9<+~gSZNI(=EFf+veMQ{5bSKuA;`>P9G`kU_Y)sN&(;tKQ$d|cb$GHwluNt-r-
zIeT4V9u8M9!J{RSQuI5WSACe<en>jQm>Imz?>t}g<Idv5^%`3kM~ESL%z@}vFoyns
zOlf)5SWXn4Ap5$&SH381l-2{>mmb(V#t4r!{r@ui%(dkD1AqKaw2eP-Yk&sq$9nGj
z;@AEn)cv){>^#W4aM%G#d#Y?svL<U~;ZC)#vd1<sY+JJeGTa9k*jiQ7s46E+L*jiq
z<*?(}(H)p@TfMl-o(t`lY{gWR5yfSoG9W}bYVkwrH~;&4{9OY6`9lmFko)d>mj2sx
zAHK`cP9RlSf>Da~n~rZcE@v|7FcES2e%xeUghr+x)<$E!7@?gVJVJNVLHp2u`<MUy
ze;a}A$YYks@=t`N|1fh!e&p*6!j@dUo=p5tRK5Sa?szFtV0%tnv&a1}Yx3`({Ld?w
z#Np@TDt*<Y{+H+Re|$*cMNWk6`pP%xpJ4ZY`uuN^p9lAv!++WQ{lBi{1h9Q0&bMLy
ze}4YyRfH|P&8|xHmr(U5)h0X(b8vBeK-f6Wz%_crC4uR-2gSZhS*6>~4ML$>=YLwz
z|NgpuH9`k8DE^uD_b+w=Jpkko2N=8c9i+F?2fbO=1KIl1rQl3C2js=GjpY#BT-3nq
zfBekP#><ddR0urRN)LS)V&?*sN4uV3eS}`58C*5@dyypIxnpN&pZbvitKxtQkT4HV
z-M(vbA{K&&5AXDu>kv18fRJDhwyUaYWRt(ZP_AJ54$I}h7L7tN@cbLfLr8uPW?(4$
z9zb_*H*5PF#AkXrE|~?YbN%CzKFz`*C$yz?r~1F1y-T7`je<n$C?JyfGKjM0VE%Q=
zCPf@9x#qx?l22>_@Ty#d?6D36g6iC?%BQcKpg`^gVgN$ivcTUhY`CM~2{KdvJz($Y
zoEafWleG3$_yAqojgYCgly3j>6@8}^<_y#VLkUExF$>KKqQ7a5&~=^>Y9d?iRS;)a
zxN><*M$c*DL_YwRbAZG6y>N%K2D5kKv!mbHRP)l+-wQxzu=_44C=bxsF7Vtl0)NbY
zD_~n@P2AQwPef*KoWPlR+bsIAQd7zvnY@&nVRdfz%l{~sc%6ZG5U6WvuKkOp{*;FC
z!H=}J&ll|)n0$H`M{4S|d@+}F!LhV@1*i)QEDs04_*LU#SLX%1^COsBcyT&`AiC$t
zFVH+f5nOi1G~UT{2i9&DNS>A(nmQqpM14O1<IM+3G0T4Qy=F|wJ%8RJ=yEPMM`&86
zom$AFx(<2ja!|2!TM_V>x{Ep%DdCotk_+TOr5c)uqFdhiHv`d=c~fDxZuHupppVP8
zbXr6_8XJsTC3fB;!BWqg@dN1{C4Uyf$qeFXnv(a(l#s<(?*8_l7vsMfGEhCGVOoTA
zihdYL8N+W=nmaJ%rHC-2kDj@tMy&6B5b4dWAOj}zE($jwjl%s>J+)7QuuiDL^O`OM
zFQ_6RaU!7L?g3XtBLK4v27{mci`<y-9sCK_-j$8&OdrIRA_MgcMCW8R7F@%JG@3Xb
z{($YqL6k#tCvG|>XqI|ExFX-<Lq_q?dNmmdn&Q(jRw<bWM^mmS9`<IG^)I9SAnkVx
zecRl^HZp9qmiNlx<BvI2%Wrbt^xIq7%QSNve7iI2(o!P7UQ)8$vhcW90_+^4g=3pt
zx)d{d&&=M}q_HkLM6~8R-$udZ{U3|iRa3Yz*C)Hw2md}XVt|yc|Hw**DgNCr2OW#B
zkI+|JD2uO!T~{Fb4jJ2zW1;SLDM#gWIt%qVxXrkep>|a7Dfg5*)Sm7-Gdu5<^8UjV
z**oQN-|4@3?SEavf4>RtwK=>o&2_b{e~q=JXpe1nSN$?k7F)u3HSU59f;A#P1>)Bb
zVn5SUzXHiOGL{kO?AyRid4(hytOJX=2OO~xcamI#n&+~*!VK&FV$aSYqVHVTaTmlc
zsys&oxX_V)zQ}IRZjv4bRthIsP8*ASfx4(pPkaVIntTu<_74FRyQo-SR=KyX6$`>s
zYebg<zJ!gbkc8}<$2Is(omi@x0#JyZr6E~boeupgF4u(Ga7BQLdUc=gTLkbh=faR<
z4Pq0H@IgJfH`Ejh4j5)5a-V5xFmTF8T)`oNoS_G?2X@kM<`b{qL0f;`LHCo_gC)iJ
zJ|o!mMJ)vC#i=c;7{^gxYe)pCD~iWqK@6z5ng)zD#_g}iZ(YOeg}VD*C~;t|qaG3a
zRscjd(rm;VbX#OE7Pn7#@xg0^{Mp7nvlm{Z=ep!A4nU#`ZJ!Z>11!OFYuEerjm%|h
z%5A|tVk~e*ZCB^uj^V53|MuZzS=h7O1W>n=K0z>#fL3rGWvtLHO3`Z;`gYA^%J0(a
zBVpblX<jG^nE%bmOKz+9%)SMygm*H95^-xN1)Uka@vHk@q<b&ER>L*o<Wru{5i_SD
z+sUF8aeG6nu&lgFr9WX#L#ztqc%)C(uj;w2g^$z<UwyIvq<UW~twM?R{N*o)QCfu>
z(ab{Sv1=%Ng*!xc8}cRExC41#pd~l1XT8OWVU}*HN9W>-tEJ1xK-H7|=S>=|2k_6n
zEzUOYg$c(ZT&^6QnUX8d4%Rv<1+Vm*&Z<iUlZpFGG~HUTb#Ogy@|gZi(pp4emm4?@
zyc4hqHH9?y&F9I=w`eA%D!X&anxDA_H-oeFiToWvQ6MD%D^S^{(6a#)Z-F}E?sliZ
zJrF7kMgV<uQ#UQ_Zqf?p#w|+KZE-*AQ5XM90v~SOuj85c?&51SdtyTaoGs-qZ{sq?
zk~9bH{~;px^%QX6a;84KvHVl7eU&F;+@-(l@O}{B<8>%VpKp@dV*E3C8gD=a`#fh7
zPP1$U{0=nim&!oNsbt86JB&APnDUX_HSoHCGgjkW8j=Ktn*(yP<55qk{Y&tLRK)LL
z4x1YiP(1RuYDcHm3Vk$|uj642qGSP{VF`j*omgm1D2MQ@IVkfBK`S9j!iiVrq^+i+
zc?X0pUc}m7rYGu2QJ<{_84At^HRRq(eD5WWm2`LI!LiI=VO($0`_-=Nx+IHOD0-5+
zVgk>A7=wwpIWob4(MwS5uGQPXUS{?;%DdwnlDU9Oh5Od#IWZE;avr=zLoo0~d%Jsu
zc;_^r$1S~4ay*aW0>kaRys-E5Ari>c#-F#4QLvAwMWCKQt_mxju3HDUkCAV8Jo9$h
zuII{luR607(Sk_l1$4hFGh70=W)6WPR&9D#<elhP@FeK{5|FWIoRDIq0ic@O3aYCm
z9&Z_fDEPE^qYdD;4i>rMc<Zdn6iCLzdYIJo!3B=<z*y{(_r-a*ugIR7#T=Rn0<)T8
z%G}e;B=ebJ4v^M9O&y+9=zJe$2qGl70E^k<qt^+}TV9LAJ5q66DQYMMv;m<G0**8S
zP9l!IWKFPkS(Xg<QuI!!2P;0hH>sWB0J&_zcaHQw6R)pCNZ`EG0Zi3HkOeG#JU+;!
zFFn2^0F{GwQCIFwAAETjg&^!X{)h!Cdgi3fxw;#SsT{MyD>n;mgE)O{DY$<SHYbC(
zq67pC3Fv<Fq(CdU<KN#7OKnB;?ITQK>$u&G*x}t&o^05)qLXYl1fNGmet<oi=hFVn
zbWVL+tbhKTk4e+F2KP{h54{3?3QpXEO{dFX*}dXY#JBhivP`gJ^#1PBl2Q+D>}vyg
z)ex%<*Zw7ZFXGJ$cj{oZu0CZ{{Ye)5_Q}gLjw7e+M!}-))Uai<ZEr`jm>+K<7O%Ax
zKUU8ez+bF7|EOE18?36XUj|a*=nidcT~w0I0-V(X4*4eLMdE(5)M0NsmqIgN{*%>~
zGXoJB%0(|WUG7ie>*j&-4@Z2}=D=Mi7d8yPSmKC4aP3_&G}CqooPr1m@aWT<STgcB
zsDKiX^6~IeGl=K!LxoJiFXMQnhUKt-3fdv)(vNqGt6{ANR9a#9YKre!<nb}1#TlY$
z?tD)<f==M1?8ZD-S41;Vm86g@bwiDKpX1d7Y3KCGbr|gW)*6dZV@!_=V939qy&X9W
zMV4`=)ogxn5q*cb6$9}{q_!&lwD?wX&w?k4MIz}s5WF85*`8??avZ)gTL|qc-C7I#
zb~W5PbffeHT}N%g3WIJdl>x%=9$^X<@gep9-b+Z7@Wl2}+qi|1DvRb2c21MGKDhGH
z`_2@MAIxM&=;s3UF_<%Clu|sa-}KVS%JU|q#PVV8>+QoNw%3-0E`N8~TG3@x=a>N9
z#92Y6`?#-uW_1XwnaSbJkmGmHL%Mzn2wb|D{KOhuYnU9HoxhJ^H<4&grW-mSKd%Az
z#|s8tj1<|5XiF+CMvn%L;!mDM+6LYyJV~o%#B|&F>1x8fcaU(=_A$rzGROU+p7G97
z8LBC+a^tXb3i$0DQhdyGE1G@jd!%v>V$6`?{Noy_2Rd!KDU4R&cs%*ibCN^lJfm^P
z-l(T^C{}T$!!V$8y5#oUeeNQL%Mal4kQ+wa#f^^W$F}31Op1vd7~v7&C_FGD-dPlQ
z80DX*(b;&bbxRFZ417~L27)eASIuSEH7#O(bQZzl>*u_Z(H{A}?!MFW;EX|0)^t(+
zy_cJuly|qF6{*NL9G3o~d^@fgJ3W?l$L&OiseoRb^DIq6t-NnZ+`~La)v`-B#pzL`
zC=ai4Qh(y?eQu+IHo>U{vzFhHo<Sq@&ulz8T7piC3eNIL9{xzqxWXW*|GFyZ<au_&
zR_|`ko*9(jJszu<QCcmO8k<XG!*5;honw#c&Sls|6EApLSUO-$X7&AZG+gxrH<rJY
zf0ZWhg(~Qzr0b_K#!+e96YZ)W=^3N-)Voxu)?y3G6mh>F)L|*GeY)he{&R!%RQKA)
z{VTWr%O=GK^We&rQR1frQej^g%bz!$_3W={3jrB5wMO_xaeX2|_PVeCiDR?^tbFa9
zWv#ULscE0FRc^*^CR&W@sx@ZWeXsT)cvE}^oGPriC%fj+MnuFW@xxL75?}sOhhp71
zFMbgYL{MGWbf%?DU%>X0;izuzC2hM<wNdO5YQ5>X2bo>Ia!oh6vs<#Zm@h2}Qd1hl
zh(U8_TgG3_F(1v_QdVC{@Mg6gs`$#vM{-sB=G_S1ZVM$^4K~D`e>C2uG3#3;->iZ=
zIAL+{wO?};sGPlY2)OU}N2^N--iBl*jqF1|<C&>x{n_|R{U2^syRJv1@9^#V?dTM%
zt;O$>ZU{vboAW-oWUG9|m;5lZWaGE2anC$QA989>AD`a2MJ~Ehj_T=QzC7+S3jPhx
zsAAN4TgphXy>qeVdbILv^PSvL8X5ln2aH~o=`)Q)TRqIR86~bER1mZC9cj?z>VIr2
zGb%llXQ$)#ojB?HOyZYR-95wve`QsN){m`2Jhpfqd%gy1+B&tNHK&o;#iAE)8U%%v
zY|pO(@#&h*7cQ>@D(}uY|8h-oNYq*9(#z7UPm6mHY+^vZ><}(+*?o7DARfpzPnI)!
zF-6komJTkeII@)@+b-AUAx!SPd2KNUJs&A9oOn=OQByad%FRuCmV=M@y|WVcdiv#7
z19figF==jx4|oC1B=bjlnh!s-)y$XnTv}qVWZe>dwIE%&q>YMNj2lh-nC|qgb*p8G
z;XNy#(Di{b+tJao_C449NoyC($~0j9uU3C{^Pd!Vwx6MI6F<q0RlPiNIe7W49k)@q
zU};FX8lQuoW$^879A+%(z3Qd^1pdJ5)8QoOM~Q#O{CurJGjcxU=V5YOx|w!DmS}#D
z)wcEFiF=a<r>Hj^UljXP`gMzb9_98=#nzp(UEnd<VETCFjt`znz_5@|%|)JrjnD7n
zJ=5}mxiYDlN3pEC>=^Y1JKd;11%JA2Vz^*OL(wSQ;tivZMRZ`ApGy}<+gp67iYm}#
z;Zd-7+YM|gze{ONrAv{otj&g|hC`sY%SJq%^k>EDvQ^WK=d{wtX+HO#MO_=NkEoQ}
zqGI@(K;^pp3rV8y?LR!I>@7b3hSN41r@rE#{PefYk|Umw=Q6;phdwgkqIXEZbY0HN
zTju$ZG_G=kSIn!j9lo3w5(aj3(8n=k>Pne-iu8%z>8;z|c4^;d;nA6AynUl-bcBU^
zgC-j5P->m3SEw0$gS_+eQQ4ls+t!k@*ZNs(V)JTk-lB5i*15ENTGnq}e)_TEg(JTk
zR(D=pXtH|swARLWY;pHTDpMD_@3Sd+ZX|;UqoR2WozfTTM};af%r0Fy-F0N)_UwI(
zQ9-oeRKgWh!|vG7*_BAe`sUAA8nG)pdOuwRPoS9D$Lrag1PspBG~d9ia1JLlw>jA|
zOD4W<T}kp3GSJ&9dJvpFlh#AiiOpQ~y<IPV*W3P|(93&doFqB>+<%8ytqKu)soV2A
zVHA0f@?Qdl8;ek^e>d~1lLn7DV$7MFeAw4WwY1=SV>#XzuAzH!LoGwnphNq$@qAzw
z4=ewhJ7k}(UX<Ffzdg5fabeU4z5ioq%5o@Dm-YQmuuITduUyqk$@FIlICsy>dQDcB
zC7F2*n#42~9M2m=2FLG2cFMlt*lfqFHbJ%K<d&CS8mt4S-<akHnA7W(2Q6lLjVQh3
zFOo;=R{u=-EZ!zdXIwp~ewZzGsJ@Ff|BLJk%q8KY^3l)lupM85K$9006;=tAGI+!4
zUznKS9H%%oz0Kgo$ja|}8fg7xJDUp_>6pc=0*=;F+644bg*!$+l!{07c1LOBDk{@{
zT9_+K+Z`*)zwJ2%d|x`T89UZ$H1++EoB}V(tndM9RWB6VHU1GdxZIS^oLulKVk3x*
zgp>qJG}4*5o}bT10S<2az=}>lsO3h%j@Y{A;y{3G^`=Nn`>(5LoHUJwqE{8Wy-B9+
zkY$x+A>6xz4amKVDN0yQof{$kh*s83qX6kf_E9a_;Y<!;Lyq>14>h`-x4Ks2J8n=N
zi(D)&JhuLkR27rgFS};T{pM=!o|o~0m3DFDVAQ>SlMOV}4Xp_ol-W;QiN!ZdCJ(J#
zjX@`hCc$@)tL~O*b<kygcvU#R!r(;MIOO27^+9t~>tk*Aq$HhhmEuN;{_t)+ok0>z
z^M6vj$>A2iMGz~6!U<<?ApOGZ8}z_64w%G?*MIT5W1M3{7QANJPxNAuIc3|AO-9Op
z$aQ@>1es+YhlcCY-~3@@^0R$R%y2Ok*g%=ji@#G-qL~p7%Q(s)A%-Kx#%J0G#zB+W
zFhH+@rlG<(L?AM=;(cU@GU%s(%I$+iHbfutyB3IFpvuz$wiJ_s%)m2ryG&WMkF4-U
z)^Y#ld4tx0e5LCKH!F7o!<hE1lImsAD2#pS<Gl3IcUIANnjcg8PU^BM`EMggEwC_}
zv2Lp>Mr~EMr59(I&i1pT?D@T)hhPNA<`i6*-`gz&G2MLgghk3KlInTs+iGkPG!_oL
zr=Q=^2Y7{t+EOr!E*39znc5i96;2vpzSKy@x}2U}<jXqWGChoD!hCqaahgh?c;m_n
zRg$`QIh;9{j;DU6&OAJ&`bmTLQ?$z=A<C&ck6zKXWou^2zuZ&|F0_2?pR;#7HGbXv
z_c@bJQWg}~aYGy0pdtvb?s6p-Sb`xnt@Mc6*SUCcB&y*Y<={|S)TtexuV*lw20mwQ
zujqWy=8*<AXKW$z9Nc?umhbT4oj~8U=#|z*-jA*|kFS{dGQ?W}eLdnthe^U!QjY|;
z^}Eeez4toT{iEcyPM~Ohf5n>xmn6~!x}5%@_8F2zzT?jL^uRygF$0T7{R<>qN{qxD
zt+&`S#iS&|=~4d@Nq}`<%g_twEp|q$(*yeAgqN$iciZf9&Bxk3z=VuuZj0F>#AhbL
zDba_<Gp-o0%w4#nCU+zkTF&R2){u)rhG87##6kEoXv+@Us`Iw$Z#}%A`g}J(5KGI1
z!p-_#I9vX?GnOjzldnCXs?YDN0w9qKRH{-_8pkKn1K%hpJ@5l!9Vtu&%Q<aDV3~*;
zuBjP>c8R`PmQwT#HosIFh2#jxIav}JxQ<i=<c-Azdv!}uj+_oijWYDp4371|99_D}
zyN(vYIKrKx)^PC&(3k>k@rR`9J^|DA=8`yk%Uq^CgwJrZtf5;h$x!}3pa4lCvo{GY
z`19?)`}^BH;>d9*4HdG?IGPy_8ca3BeAzJThU^x6rZOkG_C=aTF)ih4iSrvYael{8
zs#$>0JdUr4F&KP5$ZHuOa%*z;JQ}r>-uyw9m+7)3;p1gqk>Z%g4W=X>vdu}EP_|n^
z?k&k&`;X0r51B##5d)SYaX$N0mww>R3F<#)A4pMKL&T0<E;2kSZhUtcT$9*<hNUv<
znM@GZ3m=?{_j&=8Z4TRDSzn)B?y?BU2L=X$gZAIpus=S`zVJUnP00{@8W2d6DgTQP
zlm&fV+-HO`AR|ohq?qG_(a|bG1>^0CVc8&GpTJD#igs`3rkgXpa;+sk)db1&)RaWp
z>+gK3{8zB2ie}5nvD6!jcN|)32z8=;8PVh9yS^^lR|k~Tztj(A7>>p>Bx&=Z-aFYi
zF99SPyff&c<DCAZt2U-Bdmuwsdgrzxc3E`Ujz~YH$@`U^if(yL%{)8Sv)D2|-~jE}
zgD+>V1Ek}=XrJxg`|ZRABAFzJYq0Yra+b{#cM&S?vjY?RQSa93M_Ca_E7LwUH$Nit
z?c13mgm@<iwwDxBWy@avw9nr9pzi-D#L9`qKc@6hPmr8`{VY9bC2?ClW9^;h<|Y1r
zefx#+>-sGZDB~Y>@mxKXOFdVh@rbs^=GvT>%X-FElz${O<y8E+Y~#5YpNq=cUU9__
z(kEARY}>r?pTPoLb4`5}Mo6&A#iQbxeHrs&KI-ouaB##zX&cyf!%w5zgf^v)&VjN(
zc;WFI*B-O|wyT{x+N=y4#fLtoZ;bOF7YG)d-;`_Spsa{P(i2SHZaGY91uO&${p(nV
zM@8XhQs#d|!|_rJ?D`dD&gpUg`YWq9h=B-mZxlOVd3zqlaly6MC`0>4D(wYwyQ=x5
zNjD~6AQkLq{@Fav9|q*ql$5WEHQrNfnJE`<WD+XgsGs05T`;NAU5axtB@0#O#W0)5
z^nC3n^*{(_qq@t|3hmKfDMsj|6Ro7-Jeibkp2V`{M>ek8fD;3iSDwNU1rg(3d(B>i
zuW=sf>sN@?!|Km0DpdOhBHm`@uTH-9VEIAAr#<*}n`qA4W(R$SRFxN2U{&kua+9kp
zLfjDyO<t1H++jQ^Y^k<xKPw3aivpEBx0M2FylFH!qw7g4I@=K!D4-U->zm*c4L<t1
z6_J=;ugJ!Jy30$Xud<g8P`~<Cv&^4rvG&bydpvet_*UuuT)P>jwMYrobHin=O4GhF
z@-XUTO<~;(%s%1}?DR9_xTr^*^<<gkh+nwI(_8_r?3U7l{h>h9kIT6bzARU;2H8OZ
z(ExU5{KkCYP|qMF=i6oW3^Lc5n@wa)T_TpM6D7V>NrcHWr5?ruXFw|tqT)Ef4_3SS
zgKQxJgW;VD)QK3^&ZdiN&QNCt1zXM~uKYdd{P#ap_<N9eRAh*&)BisHpGru3c;6)<
zMN=V|cAjRwg8v~Y!a*!ns08o;Xj&ILZYg^|XsGdHLjpbzEf}GXE_!<6EdcRU;&QS^
z2<tU92=^jP47~45hDb2nH~DCIG2AUc{LH=iiXUw#EcV`HfARhYFp7zc8*?7gV#0n~
ze1!~_*swGrS5C`}4&o1pmmB865RK#+j6EclYm`(Z3RR!G>IP!-J%SdW$vX5gY1MZe
zV2sCpD-D&zqA)uq;&q{vb5i0<qxIFIr}KQ*i-C_(6Klj4d@EAd*`k1BZVePok-Y2Q
zeys|7p}?kBFN~I7+bNgHjlVx)HUXC1R&ppSubXS_Z25DhQhHzyFT<Nh_$ilt`c#8{
zzIa?0Nmkueyl|rU<zbh__a7iaj~#vqn^6M^8Iq^8m--GaTtWPb^WbKDME9&9(#PmV
zYK&$2;tOA&Ml1T!+K^u}20<@D>nAC;H{qWU@u|j<!_l|dpWnO{3`^Ka@{!8p*HZfu
zM*cMB8ir}2zp2JhK91^K*RgZqg7+RCGC;f0Tw&Ga%6EGyq1<0h8?3yAll*v@-8x8~
zU{&>9x41o9Oxt)>Z~Y`?ubNtwWeT=;xgc*ieUP_$o#ofs*DisBd5H{^%4hcaYVJSZ
zq$FaT3suhSBU|RK)}3flU$E_)qN;7<&-Lx-l@W?`nl*mKlBWkIl+xNhA;JlcrdZm=
zU#o8=16}gPN+Oq@ip%4I12=AtZ1EfxtqT%JaCV$hbq`2_^*$>KCP~#4s@*{5<OrW^
zyFI7=IJ!nL+{X6pi83t$optW!(#7Fu=-Ygq7+R|wRZWHxgTH2rzkiemv0HVo0<2p&
z2<bDglo=BD$Eyw$$0zi+yK2VCZ^;=3$-at8xUh(tW^!tp#7(j_PLDJ-O@<`)C520<
z5H=M`EdN13^wovz8m@Df;nIJxl%1)FXbpQ27c@SxbC8Qrsc>&^XX64bd6fTfv-iPC
zLw-Ahkdn6$1(6+3ezloYeNvjIU;o+BW?A&)P<ONJl}V=OUn(kkp)Am^d~k363eSG=
zmS=LwAtAewn36**cFJ;-o>%PJ<i|(ID<=JGC;f8}1(=<UCt}nTx%}q+JXCUFYM1I;
z5UmujeF}xU?j+5YyPOuHH4cxnYvn-#O8X&C;EVX1XctN$+BZ@D0iHx_BR!;dCg0rt
zQgbqve4yF;Zm2TGgp>Q`vqO<O4$&oMXAlF~FSRO@{<Dbx4vuerre(k-u;d&|PZp4i
zZGXXjxS;o%Vtrn;1sT;}To|GVyvN5`0HYO&Ujr*|oXqv?`0YO#?`|&1l@<4}_DNuO
z7oNE0O7u;`9hE0bbC&>tZENPrlvnQY<p@y@TVfxIZlAUu_vDU6gkE4{ae+2Tj*pK-
znAXiSUzqkiL3HD*DRw^3{!7i(J|v>jDVYP%dI7sDX5)v*ho+cOc9gD@%u$<mpRR<(
z{i>7Ds*?$wfvC-M8+elrAKUN5LUu%2tJd2`#70UH3Nrbg=dE9Rd@lrl9I<PhGVFiH
z{mSCNMm)_JK}Ic>T%D!L3S0iNRG`Shh*p5G^1!IM^c=_X;>`hsmn|D>nxMQfLm5C>
z_qee`+^M!W8FZv4T~AI~FeoMPKKwbFGA*-eSI6b%le-`-XZKXL=k)9m?RJ7N<<WcD
zRliD0IBxaBAln*-9D(l7GMU6d^*K3u1@Qz<<SRI|*CtD_5V>t!Wtoai@jRaMEqyQr
zo!RG3+vvqx=U{LsXFq^!BgCA~{QNfOm{+CGU~4e!6Ou=q)w-d|=Y4<SBu$R!La5**
zvl(QYIra{$7&P1U@J=$#vRhBSvHMc1a7MR&-|sx7%f*G9M`8<}PO;#?WBBSk!+8m9
zM=Quu@`(et#dS!#%aM`DjMv{S#tY0IS)p*K6|ar)x2vJZ%4_6u00C<cDE<?JdY!@^
z(;8%#)|~F*s$boL-ZjHWh*zGwAIYSv6Iprbzh@Y*68Cv`#IrPK(DhlVSsBH!X?f%V
z{r(%N<4k)Liwq22ZVeF`?H_cUE8t(NbOBODeX+etW=V)dC9n1ahPx|)R~e>Up56OZ
zL+wsS|H$rB){?a>gL2hfKl9+*J1`SuF-iZ&q6C;^`U}=toquR;uW6~JNIX#0;zy!g
zjC^I4(*37BR^~r*5M1U-zfyIoDGvBcwN^w&yX=yGI!IoIKS9;Vl=YDE>}0$umpFEc
ziephMF&lA{Q=_2KzI$_G5V0ZWNqzY0de}?S+wmD3Oj13-MQOY*`@oYfu+{{pf-e||
z4JhFdTgkDP+Z)Wk8y0$~q$Tm3e@`-VBh^XatprdQNRW8UjLlo~{iqQbJ(cwmyf=U$
zlDnY$wz)#Dw$kMdd6Dh+{VNZu6U$uOgf`w&&po{vSR{AT^W0&9lb)^b>CU^wn>og>
z;}jZ+MGdiDyaj9~Qa63Sdwcb)fMYnh?a6xl1l9Sjm7QbD>p31G5I?YN&Khs@W>CDp
z7j6*K`A>q4A6+>4rI~TB3+`<P@fP(kliat!%39qbVoa22^m*sI%!qr&CjAYl0^W4b
zb5(hnq+e%JEijrUVbMucH<?~Z(J5^iyizW2tq_0i_*I;_4<;$J8%yrL8hj*z%xxf_
zW0sNe+RB*D>K5oa1sauyA1_#cqO6>+J-1CFjsDS@<)t5;>!(jyR!f&+^CCwB-$0eh
zY<`1U`tzVeKX0AAbCCNp&IfP~`Q6HqEmsi#=G7Bg8J^0#qYez}CY6q&6J<&#rzK&#
zSn&-2Zzw#6678Lvjdz&&*7ImXlUBV2*(|umTy>Z5$#2s%D%mcYylbLX*p(xO_EzWY
zsO-p%j>iT?(P8fQ4X-RH^&e=vFcv>GJ+*34)Ew7CPEx)rpiCR*$T56kzL&Q#AtGG#
zKpTSyEfd|w#FR>-YFI_gSGzkZ+5je7B$!X^yzeYcCYGucjRywK^q4ERV9{2*{(T%t
zE9JrVUoYII;KnvUeEQebYTdTGi4L8<xpaynHd;Pc3I8)!iM6qwlC_=T+9Gd_Ym0V+
zYxM@{3IUD|jhL~)F$3+XE#|(YPKhg`nr9fs-~0{T{yWeI*OvmS$EE+a@Sg^5x|W1W
z;(IiSEecz{_`>rX`BMxtmgJnHdYU|^M-G}_@;86K&JjzAYvRejCTpO|_8yxkL~{!R
zU!Sop9TPgfQdL<eJbui*QAfFYM@1LcVCRUD;gQ$H#+;*hh&i-$pKzliSn`-BhgiDF
zEsROSWOswa5C$WVN~WU5W!0z)O}_rsSET%n*Q)6?xlFIE=!c#R_9@^XpW<=NFTRc)
z51cC0^p$-OFaPz^N*I|f%@efb;%<%bu|0C-OEaUA)wDRfw2RqYF4GyEgp^f|Mk$@8
zE`=wj$BA)2Vk7w74!KnzOqY}CI!%XBGx8RmylL^a;cf?)NQTzUFA&`EmNd|zscOC5
znEWsRr5b$}nxCq)oju?d$*DE`y1wkmh67B@iHGXgo;*kq(+Qpa2K)U<fP5XV$=f4%
zsu?{eCf4#LM(=6P>s#(M;;WGb*2R>UpH?)yg_bU9tsye$-I(dz{q0`>4ild$*A%#6
z&~k@RnV0bYvG?ZTRPTNJ_abGXvW$z23mM9osLb;aLS@QK#s~>Z=2}=-NMt4%B9eKY
zhmd(nri>Xglezpp-+k?UU3=gAy1MV@{^vQK=lK0+uiat!4xj1$KF`-#a_2zcUQ|*F
z<6Cj-7NZRM3t#4s=vgFND8)93u}V=ADwCEtv$|zvoWr50>otq^{Z?<mSM#lk;_kwa
zNG%jJRI2(F-hRpO9u~8^;*xhYd@DJ1g=#%k!&#TLTBNEBUNyE%v-%EK^4{vS@5m9N
zX>)Z%Z>B0r;)`CYkK3-H%q#21j27Ie=8K!5PwmYTUu?W-rI!=m4^SQeeL<v|WVIq|
zWml&5SQ=IW)bQFqE4TiA_y5mNY678SKqikC#h<D`e)w>p(WZW0Quuqj!SA7ZN<1uo
z3~6xsSKL=W2_TiOml)m1F8$N_Qpy6zoV4dte+NByJp<q9m6oKX@~3b7F9=e<30bJ_
zG!UH~^ygnhLI6pZAV?F-;1o~`nFGfnuYW6pfEfnhdEfbuf6w~;pFcLn$HdeC9UF_-
z8gS{d;J$<79o=b0eiL8&>dh8m00(oMB+k13&-?Iie&`K^##|tsvDTmYsryWRPO~i|
zpz>q`V))b&6CsUKY7l#H9SF;+2e8HObqOVZBx=`SVfgis&Jsv*sKAL~-E7UAkqxrK
zK>~>pNb473`!4%eie0I76twc$sC@dfusG#3{%@it5bB5WY_lo?8ikkDc=LHnf5^?A
zobm%zj|s@}a$aj`vaSvkVym0nlYY@Zf?7xrvUr7YHRblN{-j@th-JYqu!F}u{09;L
z*WdPU{;%JfG<fj8*6aQHPrC}Vu=N3sOP2rn?fNe-+h70l|EIUWFIiy<KnOxuNEyY-
zGpP)Lxw0OTmx9W09B{iDK$%P~kIHyMq&$e)hKghI9<7fNKuX*r;A^u3xwiR$fw5yx
zlU?(t0r|HV;VWDT7JDjH_Fwm&^TltRX$Wq})dBv1F{sYkU*I~bZUX8ECBTzA0kY@b
z?bfL}EZ)Wk;*h5TEkJfilK`?`t9HKB%voTE7=4a6Z@aLAxO;uDD~7ygTuI+@MS+PT
z?+0L=T*{dMvPe*<GqqS)7LU~^od?>M@Gn3*6V0alHY;~9ElQbb{I%pbkV9H<H{^p(
z$Ov1cYCmDn<5!m;<cBL{_tk#~Fc*f}{NICkWpk1L1P%TDWh(}--g@e(CCz_b<RT76
zG6ZoOR{%)31V|L?89g_o69Dl*#6$Hn37hZK7NoHR8_LppgO^%v%=yjaD#JiyWW!gx
zpNc{+uMyF00E_Q1NZ%Q?4}JphDWibTN^V3+c74+g$UMhDhRJ|^`|$gx^guke3nEXV
z3VlE&!yZ&$ik_Vp$$I=gB8kZ20YIvsgV=&A;8cs70bChSEc*bZIcyX{q`(nSKV@nh
z1(|W(8>Ke@VCMT1(?#GpHG#lfUO?y7um%1oB1ZVZ(f4nkmE2;d_GTEUoYp~;(;h3g
z^vctRyL{@YFp!Z1t{T|B&0t60_{LKYL~zJb;-jm8UjBsX!9nfBAjiFc*JuBwbN!q0
zwUP%2lB(<nvP}Lt?bi6!gGB6y5Off$3i)6l)&`flh2eNUsH`MFNLOPxbLIZ0Cx!_`
zXF*d*%#H<LuDAaSh49qtK*mByzdT0y9h3lzK(uwkEs2Bs-zMvPq8<Pops;ms%ELP?
zu2cl1zCIvv{^kfIp$b4sim3=kvPb1gkGp_tk0J2w8l2r7b0`BK-9|tV4MGH~0d=b|
zFn$N@-4**>3y3KJx_01j>u^Rb+<hL>0p|k|4tTA~LK#GdB*1+^NbN=iPnCofcmb`K
z{K)1?5DdW=0aR0^;A}nd8!;LzN>5a;f0+b&HWW;3CBF$uuvsa5v&iF>+Tpt-|Ib_U
z_a7)QVCkeFs51ITf;!|{`3yJk-26}qsUBh1@h*6Slq&FBfj5f1e(Za@E4h@?9$0op
zp`bl*i_ArhwUj16uCx~VOMAd+KDJrvE1*3Da@94_o)1ttj3<l06Kn@8uJ!|98a8`y
zA(t?<4zdu+^#cXkEsRt$&;>U=ScURgSguqPL$q8#c!>GF@A@zuzwYG6dQ+8|Iyr1+
zGdh1UGx1SyF1TyEX-q%b0O~Jn0}~R{D!N`h0Qz!$8>+`po;?Uq>V>}`a$Sd<&Y)40
z2JEA@!e-4_BJHr`VGed0?wUW=w14wPrF4K9XV@a^JIf!V(XUn&jL+Kf!|W>fu@63A
zR%bx*H`68gRs0dK4IiGqEXOq;b=SNag>TaPQ!2a_&7f#`_2{6-hcS7k{{o<y@$UkX
z$$f(=5VAF(;$=2Q@xcg)kK-XdF?6QpAHDjGfr``@Q=_n<3<3hkC2B-nO2OY+Ng?@C
z{y)2aQpBOPkyK~ykEdtPu>lG{G$GaN)!VnwOAVsuroX%a_KS@PL9n$bsX1Fc1%#*t
z&|TF$FXp6qzTNlQd(fe{62|onFe%1gi(x1|!Rx59un=d(I2$f!1h<d?8K3Z~YhX5v
zM@fAA2JFd#)w-H*z~ns)isIzloaz$O-vb=0ZQ_9|#SS#xlu1>_!0dcghUE<hJWa6%
zs$1^dwIdQxq3H*YgSDD-THFSd3UG&=VB4Y{4ugU?OMn)Oee^px@(PD#0v|M-R&`N+
zxiwY7lr|>=%xR6#lx<DgE&+;xX97XZsUzS#gDu9TljAz;7IT9=;O6pFoer1YDY{B_
zAxdP90!6V&?Z1a2U%v-t_;Z#9D&JauLIF%DIGBlIdf3DCmvnZnjj|+go|-lr8M?XA
zBuGxQ3k(;!7<h(H#t;8m8PWRJ+J^u;y9%28*aafgb+Dq1fT+mOQe6O!PF;`WBRap6
zqs$(ct{v@Sn07q)OirmEumXl5DIa|LV|gm~;JFBdmp7=(=f<yN=+j1IyZ|bE7Z_ly
z)Ln*+K~MU@kU!yKD|)i&J-@`%Esvj5gp&F03rS@&`)zfLKlUcWoo*Ses7oAtHC@j)
z#0ga2HO+&x^XRC8#3#p1myx?$04UyOMb{c-{wB^c8=Y&&K}QB7=Zz7G%%|AVscsbz
zUetN}1)`qXj&dcR0Lf(|yNRG_Gg|qT)p*7<w&I!^&{|sz)5xjXm~n3zNT42?NhsYB
zq3)l(Ii}$6{N=;Y*KhsS;<2wxvHb5|oJ;kx9ssNR0KDlQm+4+y&nBgkkMB8Wnu8gy
zup8VA6ieHB9tqyp_`Mg+i$8;<5NRHN5P$H@DWs+`%}^wdVwqYrfj!xDt-;7-eV?^~
z;7G;i?OjkKIxf{K&><^jH;#T1PBe!rYk}mX5PU5-ye|yw%Fo~O%(<)5<$34pw(X#$
z<*Upu1w12BKeRiO^Gel?xwkwjMkF+IJw^Tz;qmwL?r)#S0~b#|Jh)x<_c2u&TMzoD
zmK;fsZ=)SdO{oM3NLINASyD~#fD-yz8A$NT$h|0w+XQ;ij}TE7@f?*UJ?@W)*n*Va
zn(P?Hx}z`aB2oSuNf59vQSo{?=sn2qe*imkKI->mS2V9Je@GQKl^_eaPm%fzRP(yQ
zj6L+C?MwtRmA=^lU5`pf%{;UFO+pRy@T3LIG86=nn2rvTQyXE<5LW(qM5UTw*VSFH
zEfFD*u<^#KB6c9&GfE?+6U$g0<L@GKlSLzo3O`n4MsEqIAB3CFM=^wb^&dt=`+&_%
z8V-xObg}s!_i;GgHQr{maak`!GU|;k5Fo5_NFFF-j|+bjE2B!~nz31g4(cP1K^(%^
z{^sl~rI$k>gKIui^YKtwsRO$saOA1-&nML0@dLBqMS|>W<BqZ`O~DVMi2ND>#h&JS
zr{$1soy@)y(3CeHeT>=bQ#T<QxJR_v4jA?nKjQMS;1)c@RHr}OF>gU8f>N=~7;p&4
zH0yGKR65|&m!C`*ACL&zIPb*W!(wm)RGnohpT+tMya>K`X-==_M!VtxSzKQx{3Vtl
z#kd+wr<$&#2r@1UnZc%gT_B6T6g<iPm8NTG4_d(?5c<AVPm#K0TF>!TOdalLQe85m
zc_y4G$WWSDj;_F;Rz|7ij9J`0eEk*ya|TUr;Sk9d8h2)1DB=lH=!X`*5e{@x8KLcf
z-yyprkbF4<$UiHxO)~p`$R|FKgJJFKy9Lhdg2o(z)YL&K-<PA*4Lsi>a6qelHg%bY
zZ<~5|4VVasBN*-W)6~t@;p-J3NSHRY+fL?g2{?5Is7;FrQy<}aufIa*HzxnQMTh#$
z0<vB9lHuB=#Bb+I-@L;=5w|X8?7RahrA2$j?{Fn~tW}__InwPi<myt^&8_lZ7IVzL
zsFiUa0!Y~WJ#6PdR7u(RGmYmG4a4btCO%{gpmnS#P?vdd<zxlyQ~Exy)tz{Y0{e<N
z?}At7D#V?hslqsr|Kue3L4FimqYo@*qEsZA7Iq3NT9{nr$;r;>QnQD1RYwmgufbqa
zFl9OEyYMP)E^w{-P}R5DK(A|F=twQX%UDoY@aPznur$f%oEb%%_IM^y-_u<^`K}Fc
zEOX`j8V02eExp9z;(v5qD8OO&1q6SnMxzYqz9zRJWTuwROp}sqVsY=mHb58cpeUe+
zsR4bS70^sP*P{pLN02d3)axT<kQfx&t+7b^;<_PxZ-Y@@rWea;>rbT(*QA`eh&h9K
zFRS6J;RsF)LNZiug5<r##_2(bq^8fBVzX}wsS{PQIf7y;^c%1cUZNm<!Spl-y!u_h
z{#cJV#mwJH0@Gt_^<X++emdrJ@<X;aCU`^Fpe6D=JtDRquYe(yO{|-lKx){xRg5}x
z6L<IH>;;({qm=5Z{3&W=m!~wBfEqpg$!+R2ee+PE+XNXr2jr<_cxPDHniZ=9AN8IA
z8$dU-pLrj>9}Rx>RkA}eX^Pb#D~9EGH7IIUg)tiOsh+&gL;`RoMmHt`Ii%WyVOCHw
z8%7`Oot}~5meSA}&$v>{uTu5pW%0SB(x(@5e@)sOkFCA-g=GhlZLfl@(m68br<CoA
zlc{QmCw*3Y0FaN!Xgo{$Xhg0~&fp2axyjL`_ppSecH>Cof{OX}ru~VWh`p-dc+5|E
z56{h<KV)H?L)-!>f>#K%uy;1Inq$*X?z9S+w_qHv>yz$sfRPw|8pX?^fRR4HxN)>h
zLTWjdv16C8b+C<~S=b?JcEbsdhFUEhfyu=Ir)KvNB+4Ua(mW;5?P8z>0x|DGe)=L6
zTWC^QTSUAVUbD>l76JxLOT_7k6PlPEhSFghOyNuN<SvB0tA(u`Ovwi@oX%z#3l4Ge
znJ0Mr=aVMjI7#}bStdQre#$7JA*{4`{X`!jnWw<d4On1JV;I=on_46rn}(DVp_pc*
z2okUrh{AaadtYQd03&rnva<ZKPvx0C!j}>gdP~`P+`PxE#l1H}j4?@X450_2Q!Kk<
zY)9ksl6?2f>pkYO$j-5)<ECpwrDqa)#CrB`-Wi&>(&8R~XvXD&y{aIA9XL#RqWR=^
zT3s-ou;?Z8&L(*0J_>(=D=sH?HrDN+0JbeKVX55t3g_J3SaybG)3&j!*wc=V%7e=E
zLE*ZCC6oS#;VD8y=ci%O799yw_3cTqCGlL(JL&FM9|#UCjIdC?aYzr6C<g1$7!gO}
z^W^Lb`_{cjX<n(8crKd%c}FQl5X>?~67?@2oA`5IzIW`rY)cxCkkl7>8(AdGW+se9
z4EUBYe8n7Vo0moa<cu77<kB`&v&O~?S27TLxo_qj`0ap#S668~=iZE2L$ml#w%=sp
z6W%kz$W!dK@WlNYuMu<JW@XO^Nhb%*AQDn?<X)7aTY=KdOm(xoMo<VbIjOV0Ym!@e
zEXqk*YN?v!DLL{I8?Ovkl6fmKkpgLTk(cGN=(<@{^M~jl5|#F?X?A0SaA8MHb(tU5
z6vlV?JS4;&`($QFc7j3}$&vTcGtIv-MuBg+xE9Cd<Rm>JbRHcZLQ9UkdU1`1!>Pe!
zMEonE=w23;MeY#a_#KUz)PzE-Z9-mpRRe-p$uHBx$hNfFH6m^&P1Bi)5%N9uyOlli
znnZgVWp<Y;jU&oDFu{di$9Rt>jwRk%^C30ac$Y6#tvt<}0?}@HUn5DhDfZH1%lBDE
zBX>W(0ptjNr@q_8*VfyNdnU%(YXd36M9MWTwW!zr#1Xb+h2*6s^Af*1s3a0lctE{a
zhcuL<{Qelv<inU@I+bu^4K~2AA^ARqYm<Cl5Xmdx3*g|T=`srQA{CE^QiXy@-culb
z*oVvP%y6cWHaeqG-UeJ1SS}~&M;R&<EA+y+D^+gYmF-in+h)B<3zEw>t~mLrjW{zt
z6?4MVoteog++M<)O|JdQNI<a{7MILxZ&)2)Zq2t67h*;+U+ih-35CK~WE#fO%|?{o
zj3hCpu%kZs6luVnQ0`cktbci8;8W=vGt}|@wC^blv6+hryfh&aYcf5U(#L&|429d}
z_ku#!C|1lR1UM=<#hnC<f8M6A(iPrx91jY4t<qzzXxPc*(r_zd8$`~fomBAGaq}UQ
zTilVwA3RkhyEs}(J;FoQ$&s`3d^%ze)73)qg{VmSrj&Inbei+6cpF1+^!gxPlv_17
ze|XHWq9bv{Qp*)w7<W)8G+O3fySP55a41)MEb^;@qMs`nGkw^LKPppFq#V9j;?48;
zOw`_bSjFB=-~)WkBQ)#9J<nRZjqui*l-AQ7-s>&WT36;{hHq}XLk-G4)=+;G?iJ)g
zaC1EE-ls9WHif%Wj6)*W5>C+>#GbF#6GrJ6kD{AmswT?Vdimxc__-pYdAK1$=q|bW
zo4v|wsfeOp_D=-9gQr2OO%4-sfO&p80yPus`KU($73E0!RM}`ess<_D3~E)`!N;RO
z(C8Te7vz)Zh`#*AL9vcy-IDqn8LXcp=exX|Kyi%66_pU<(c6;%fl>JehSWsKoH>*e
zJ?$ISC!Py9Uz4UdDAb84HSXRlc7eT_rYokb-^mK-gs#m_iZ^OKsearo9xR_w?v~$Q
z@winS=dVlCpr4DWZ>jNOXKbmSv9D2vEi&p|4&KhX-sspod4S2HBfDs(vPT_TXCTrN
z<_@23t|4Z}`fq=fy~we)D&$0(=@i7n;$XK2^x}GzUt}d(X=JZztJ^9VM0J(r*sJY+
z8Tnzk+_0~>xW>p}{@$51{9}*zZc(!=Z)-R)V((Kzz6~5lHkgTjYfItkJ+959Db(j#
zQFf0cCIWuG40p#+cl@d)M$e)$b|9IQ@b!8hJ$}cex<fWirz&H`lhytnCYQ^@orKWy
zRU(K_?F(R>5r16^**YTuJVC@;V`@SyhZrwp{|#rm<viKdfwCfZ%FUW_rB`jW`bU2+
zru^%Z-dp&@cq=)Cf6Sx36hw$8&-tq`yKK#IB=0)K7v;)t%jj3eB*4{Ftm{%KP(R>x
zN}Rfli=S?;t#Y?3=r-K@=#Lk{@mh<Bc3kh#^ot5)7MyDFX?V6N(Z@jkXS|+*$j~VN
zP*zeTLo6RoQCE5Gq9-#-fG9+9gosBuRzv14dFo5Y7d9+kIi@cYC6oh*LVBjCoqIT6
zwc?^`<1$$b|0_nIst39~%}TIul)rA#C_+I#<ge3!arDQGbaFH@kqh^Iu(dD^+un0p
zN#18UW;a#Zp0*_7={(v-`4$}@vWHHaQvntEy7Q(DX6ac5$@%O6+D?gqFmL&vU6^mI
zo|orP;}0^vGA)voeir9-HJMJHVPWGoy=3UCv$T1AA(@x8IU~fvc0xz!*)3kgnRwSX
zjKcx8K@!Ttw`9n+b$HZ(877x|WE+!Yp{hlKhvQ+p%4M2V^g6G|J^cuS=Qe$N_>-Nr
zs!Ra6aH3<w$n#H2&Pzl7H%%qcRv87ApeT94RDg(N+>qRyvTA`5COV9{X~BkB1`>7F
zK|WK3%NJ$*DPGWfyimhzvDsZVH8C`Zt{F6abr5>skN9@rj&pB3{j{&_OOrJw$V!5l
zz9^)OwOE*zTxM8agY2geQ+8Do$q#~3uw^OZN#MBjMQ7DOXDr@!&bpKlQ=I8gJCDjG
z67ShNJ$a#+ZWq(r(HZyhMTqdMp3G+DK6kAO9Jyk)IJL~Uh;#22Z*oD8*W6G*KSuUO
z8>YRgK=a6SWFqrxhP9Bemc+zlE8M^?CI$rR)C>$zmEoLMPJd7HEV=#LJN#BF+VXt`
zhddY`eQz)UR<}&fR?%FWw{D5>K=O7}VjY50?JV^gj#J`l4R`h)v3ju>do%o!3ReYf
z#CZ$b{{ZNZD%Fc?`N#TycuWm2Kt|kVtZXuleo7rpds>vYC^fkW%;1L)E9R)d=#`T(
zz-X*AQ>eeYEZTetBzab$u5LaLO_C@A73$_6P*BK74fil>o71`i>Ua}@bz%yc=f=Hy
z+^>uxpa(2#5RUVkay>ek2^h3Ne1&B|FNu>}E|~o`O@#26dny%^P1?<bp5AJvfJa-a
zV-RpEvQQ_9gseIbcfmdS;kWfcZfIR2XVS;>$u9~=<n*a|U~(G$arG=ba-xS+?$?TR
z!(j+ATp&(0<g=;1+Sv+#W5G-f*4%JKvwJj(o6*XAbOjCK%R4>YFFdpgK^}U(s&%A~
zrA%m_=Z&q_BqA7D0BU?2k$IpfO3xZCw-K_>bva%R(>yG-Gj?-X4PR>SJ<@FMrkFQ4
ze+Oz@=Y!o>Wq$tMv6ITbcchvIiS47|fFBfcQIS8%vT$}tf;f@GkmJ;T<+blBa9*G5
z11p^1At6YL>cWN;9=Mj|eykO(nzR+~uSNtlT-dzaC^CQEPTPn-z<L+df3WN4=kq{8
zuGx)fFM`nQo}AU1SMDml6vme^STtAW1J8`~o*j)DRGY}15nuYKFj@@PkE|^-)NeVp
zeFAb^XeY&i7rny0{IKvd)ux(eR2TQ<Sf;C^Wn=%C?qZC8)$wKN<purN(vWvszf-^b
zgvm^t#J_ufeeKdD`vp|Yn#Q66rO$2#2bP{ARe34KL8bW)oMBtFK@(GVg$bzl$@K1}
z&9IO<+YKIpc<6J}idO!*zyUYwEq8h3>+5klfGD^YJH2{N-3uMln8R%3ypbJPz6M;x
zg&<@{NNhW!2LicGe;|&C4TRrP*BNI7{_6nFH_@=?gk(%8ZL<hQX{HGhGS)_Vpy=n@
z5Q$Yoi>7vdT{Vp=iV<1yh;2p)XkT&`6@v-0!0Z(3@TCMpzUKljh{SWTvvzhlKM3=5
z(D!UJVF+~tINH{|R<>FYh-F3_2${wk5=={1C~R?-9U3-uf_Tg`hBQ+@3MVkv;lRa?
zm}I#x^Z3in<!|^>qy-PLTN}E;YVSEr_2V20iPS^TKOr%mY<g&UWNnz&PH#!U_vefA
z)3LISZ3j<or_H2#CL7v6*jiT)r+Z5{zR|ugy)w`$YPWQRwyy@(Q!~b9)`!9k$cTBs
z0;6G3mFvaAXl1cKWAl1I3k;E5GInYToV4gBl}5Ja_I!Yy;zM=)9%ZuYgN)zafjc?e
z-;J6)*pr2egt<?+Cq>~W);i0vYVw|UC@lnw=FI9MYQMbyA8nXQ$*iqBP3Qg@EUF=f
zXDp{ao^Wf1(T<$dO?UG&V=&)lcwqBsU#h&aCtDCcII5uuuQXn3$=!HR%>b(-vS!yR
ze<z4UihQpvrab0te3{2^RdbUoJ7ZyalNT?BM0>4^fAhJ^N1R}HELykzYheek%DBh9
zE0Fin|K3MXvy0j2wtJ+hIX;^lQJ5*ZQiW!53J=<3L@v3Mui*zZi&L=k+AcYUQt%S1
z@&f!KXF_1tienFu(K5|B=1DNTpHr_N_NNUBsojbpN6xbWSiNoF_7c(Zx=dkn4ey)G
z?1x@{V&GZ+ri?F6Gu}LpbXU+D&oTS?QY-vRj_yojTK4A(4d6^vzbRJzXvU~DTVg%6
zd$A5T;8~Y_sPPPGVDhq@@I9v1@N$I-`#yTV>t5MK@bV0CM;Skku#1|MeWG3csj0PP
zxZ(WO1Xs{=%&H5CiSJ@b9y1H!A>X?)dYf@1P`nyFGn1VjQJdtI{G<5CmmJ%lw?E8+
zdJ-?XDz<XQ<yIxX=%LZiGj=&4FUXMz^L#>{D@GG9z>3Jb`Me0687PqGmc+%GqWd5f
zUS*ubWLO@l_&KS}qfwQnuItVFQlU@n0JSuJ%AgQ{)LcZupK#7uyCGv<)<myA2Y6Zz
z-k&7%7Dz$Dwbufh{GD-vCu1%RE@CHjf+s#b@^&9V=h475ULyR<S31rFgK@?MD6Hr!
zDDHOyC$3d8paff(-Qum%m`jNe=4l+d%cbxoGtT+omZP*+7r+qh^FGQ8V_3U9laR`p
z5XWc0>Q!Yp*%7)E<ayx2{j22KRTh)gcHM)2^45w2VI-Z5w@b43NekjU@<RMKT!~<M
z*LM-XMe<2a?7n4Ja3x_h0jRu$-g~R*MCP6mvRXoL3rjJ6WHd7^7I_n&gSvyP87e~r
zc?jJE&zX`!Gr~hm<u~NriF?-M`4535fyX2Qet{vYj^g6e(-h{f8Tw={;-v&KosYK5
zxfsh=+$)2+&@s6U8Z2^p;)zZ(6Ny_fDy`sM)-&-t{Q?4XbX&MV+I>#NU4u$e4$`tX
zRZ#t~r+}5F8EgIoyLf#vSY=Fka}D4{-$SU(MS7rB5sJM`qdYJ+AIyy%uRX*)ZOWE_
zKw@<W>w~zBx|;iv9oYD^*x$Rb!>U1*E)rsq%ULB6&6)U0alRVAV-6n${-MG6r)sfb
zpfFqk@*lVLnm7OOEX<TBsI$;3>?}?wQ|$4^qtP=8McAXN+b}Ge<h6inmSDq`y~@v@
z2t@U{P*+w!Uvy^rlPvDr>-V#zP`cH;(%u&JZgx=xM}iycOzMjBk~i2I?bICDvwk0j
z7$;GBIVZh&U((DRyLwFF8e|=x52ay~M8zIDa9G?oakj@>pvgF2z}7jigIs9<9g)w|
zPckS%i*3kC-`u9Xa3k<X<64zlCx+?ST^0BZvhFod*JOCe^}l#yfEK2z?UM2D-9Bpy
zyz1eC`p~wopz+RAb6Nl{LrD4oIU`2x+lr&TGeB``X`0;?N^onIUGQIECQ#So^Mqg9
zVE<rtU510SiAE;v_yJ;U>TRGTwTs?6weSIZ9@0k;$OStiwqo=M3}}}P+dwn{pEUeH
ztkdu|4<9KLtp#QT0B|%fy$O7K7KJn#%nsrrJFNG60`=Z3ST0Wl>aY;#q7iI9)o14A
zmEJC}U}zxpsvaZdnP{l8!r%B!O9GYHbObu^YE&#|#f+VJ)vz$J9ZM{oc&9m<PBqV>
z3@ng3SX7Y=0>i{ed;?O|zkp&im-#`P9rH6<9q!WaI_te0(wetpP}xP&S9nl7n%5@k
zrCtfCkuOJQ-1bRf+NL*(*cfc_j=pH2Ht9!Lf0KLh1b0Wycjq<0X$M?mCnchvF(930
zXw*DBjoUh~axX3nC#(leF}ahbQx_+mFA^5>rRp`?f%AynFo~Fk_RTLAIPFXNPu_!-
z(~E}G3T@G>&oM(i%9%i84k8e&iuEt9KgK4SV0oJ8tc(ogCavtu9mL*Q&X`&kjl}GR
zU!TF|PeS!lLQ!8gX-qsk(l*F%pmT#i$Ucn})%k(WR?zD$86LBFz|SVp%@^@y7Ez%d
zh)L6Hw2}uHt3vm2kDNgLaPL@T@|80-teDwiI-`4+FV0J~OD8Zl-xANd!iO&3LATl!
zb!lr!DUIBri+H#Rs{(a3x9hKxCi@H1U75Y2?ZL}k%ne>{7B0Cp$Mh4Ev@QGKb9KPg
zk(#GP9zLr^O(xEA0QSB`gM_r^I6AxY>nUn9&m7yknZ&|37#+e!94;(l2&G6947J&}
zGODpgv|~0m7c{FU1$0({(BmV}xzL#}gYCZ>B7Is76!r~tKF(U?Kqpw`Zv!jT>X{p_
z`Cd7-(0+VS!NkFTYhSyeNMJSKO9#h_QqCl${BWmst2NnVNA7NrnErpa_g9r``AORR
z$Jvz`UmQ~pS}B&by$_){><Z=&VmrwPX1U-8gR5jdyvqVe{CJOXf!&eH;jG>Zhcaaq
z91PCWNfp;J(SaeFo$^fcGStp}z6XtU2Q#&xTHHTQ@~R6ogfgx`Gt;~ba{I&a^@Zuv
z_5~lPLquLa++slL#qo^-nAc5EII65%YzBz_Q_iMA6@ks1&pgTn_G+yzzsrAG!Q3^z
zt)0Q$iL6`&z@~7w@>XHzO}TM5pec?GVcBoM96YUX2hfF{stt~1SI}OH3rxO>{3$0<
z+xCz*ARnOWL`K_A`zkO%)n{jQ+j*9JYuPK+pz8bdRY*x^Q+GTGYKC$7R*iwL)n{X$
z8uvvjQP2kL^cQ4kpEio?6!#keeX_J%LrPkA4T;gG3pPm7CTrdn{=MNrR5<Xsbr?`0
z*4@@!HYN7h@AGmCA>Bs-KFJL19K%9s1RBLUQ!5tx3?A#Yx1N@sXUV$m+80&<8#GX*
zve&Kr=r-<=UZKS5kzG7#TbG(xeJ!ECu)eh^Jes$`4cIb>ZY?P%UnCt}g}THdAP{KF
zqI3Ey3%fJveJ#E6ayH$DB#E`;_L(m`BuI)85X5881SCYow6mX28`zK$kMWia-WCwH
zaX=o$c?9AfZSW#TX)Mx>D|IS^SNoD$ta&(#DRzM+jPLCGmuoJh_ve$RJ8QaWS3kb;
zU=4ACA(36p!=~@}Yx7?n$R#iVG!2ABPau)&5`Pp=V%;f!Y2FO!y!jAkv6p4dJ0DsQ
z)$?qx%Ibaoowq;Ui04NRldPOArS~YAkAE=F#3yR`&ZBLu|HcZjYf4`&hT%<kFOxxv
zRy$vv>?<gc^=yJ8fI<e;d@$0~LDGF#<z2b5S8KUM)3CO%z_h~UX{bc&j4W-I_nenm
z&UYK;LT<)|ly=q>j)LO132$~MlA~P|Oq6jXiD`!CMOX6612-{0o?UeJ$eKCdbL9*d
zhkcCN@ZxLPz|jj2-bIKT_F1|h_3vo+o-p<zX+GzF<F44#H~;y_sr(?Eq_c@*46uf5
zc!vd%R`DU4=06>!iz*ho4frQfD#-N&lB6^TLEc!|0tZ|UROy|+A7+^NkTGCFY{za`
zyJZ3#@EFz?40kTpI&hK%m6}Y<8jSWVO71wdFFkE;BdqH7Ts5+5@)led%&W|-+h>1W
z<Br^(02$(x&fXag!P;X{SCFO;+A>GJK3$}_=DDqD)q?MgS4w=|z{=Asgl3P|EPVH4
zT^>00Yui}@XashBj^$7;`Ily1xfBLSN<yeL_eV;oJzM?EwbW~SaI3YEi_}Yc%!=s1
z`tt>+XD*R`Mm7os7lO>}=0?xH4>DQPMlLdr{%{Ve<oxI1%mfWzKp;DP$j)0SHl9<d
zR|xd5Lbex|=ck9xzE;d=-iDv!3iLP*IsLHR@A=sXJba}+=KE{PS^cw4VILg#&Qjw`
zV0>+4qMl(J#p{QG#))y}?an>dZ;YX5H|b^sFj7&pPPAskG|c%@$5RupX9PxxQRmF$
z$yc#x|I;3Owp1iLV(MS^%~AB4WKUC|vwJE!+!HZhuGom^W}crG7Yl)%dySM)J;Vp|
z%Xo-Mb4`<c(wzwAnM-m$J6r^(#sl;5r_VH&US)7<CYv|CKG2k{za=u|dn((=A;TDT
zES1=NeTlp879^sPU3oE~k~AK9_X&-DIYjQyaNZGWF^RuaFm=IRaZ-jRAU13_g<)J&
z3#);kL$!UaeY}$HPF<jO+Ly6%1Tp|1w)hoja<I9G1Lgqf^7vAh-I^~p;3)ZWGEWiU
z$*~bMnQL>MG)4TG6B~hqKj@|DWM<y!-tc1bA_iL%4ooBV3yliAS7`&2;jwhK&*{*>
zwL>>O9CcGxscnCgKe5l4%%H-pkn_~j%)8h`FNvyfwFU{)@f&(L@t0GkyU~@q-yn2|
zOgRoKqz2EUk>%ne=zu>1s*soSOtI!=;s##Bl6J(iWD<69d^DE_gX(N&o5ZUWd=z5m
zjbH;h@g72Fohm(|L$1W|9=j2mUY$!{AB;pudSne2>Q_bPazS+j7MR>|AU|ou#Z=W6
zBCjOjrxYt+=(177ijQ@;YC!rZ%2Clyjj5jfiUSL$#hTfgloEE8Ekl?hf>xM+^$t!E
zQ;Cx$cjlPI2sFCEK&TX|KEo47wr6+YGjO5I`W0aJaTPBY20Y#pZj(-)d3ig>K7M{A
zJ6Czz#xg{F6hfp4mzQ<#Gb-6r&f<hWbOS-tnk<bjPsyQLUvL+<8E)&Rc3n)!GQ}sF
zZ`FU){Qo9J1;R~A2wsq?bcyNr-QuDdJ~mjlSdri<Qdo==C7b|uEarZR54gQwAmX+3
zH+WNrqx|JvtF2JgvZ~LlWW?~awhEg>$q1)-!u${J5%p2xKfQ^cvEqy_k<Q2T5<8JL
z!eXaN^5Um>yciqs8#(1T(24ErTj}rM9<~>Co9G+xLgwL28@M(mGs0>sr%+4^eviV4
zNW*zjmGNi3c&5a2EKJK!V{-(h5$>!h(kuSg1h&%G;0YV=b=BlPI@`!F`!4W|Mn1{>
zLELv??oC+rsRD9Fk|H7jMe(N><mlmdMO3<WFYjk~)eVU*>bskAHYzvMjmTBqA7Ryg
zZJhp=JY?ryr9bUzlD#uQK3tmYg+)(hPjMw6OBT5c%p>$%qALf$v114#8y>RAus3^@
zkgrNHwcO&rHGRt`J(QN|`;H2x0VmLcd^-6UCon`yku*Yg$lf!czT|*S$d5cWZKBQl
z&b}r6Se_9(xV4*;?72(@yZzA4_0kZF!y%bD=a`Ra0u^j1-X|aVRau`u#`KvD$tdPV
zx?nU~zbV9cnTt1<Z#|R#pdi7>O66#?`w6+hHj8VEqFb@hU?$iIIt$|qh?$u+ZC+0(
zMNOXDQLOHF-90BQiiUri3RX)yM|;PUkomDi*nD_YNnEj{86Wu%tbm*kd&a7YWLoF|
z28I+3C^N>SroV$dcY6~gpq|anqPA{6W}9HSeW`Ki{`Q0WuWHpM1IZdy=CA%Vk&+B>
z5}J1VDL$C?(t&vE!H{LCh{8Z+@m7t&{Z~pIzw*t4pj*h<u$q!B_7@IA${D=s`{cIN
zcr-H@*K}L_de`a)GV|^r4S5U*4=H7k&5buh$&^m}q7~(;LzS?IkM0fn<`v6p@7a9o
z>BHULZqVk1fJU^lA<DO^FmcsWM+nbEbn<QWU%usU8e6%PDo8$6Rg7u>C%7IGjD91U
z5<jP|irHXa*<S2RTYWBj-bSUyqYBhm=O9Hg#K}hwdjiM2=y#gW)V<<j1JKS(#$Z<H
zTwW|%UvxPu?WU=&2m1S=+~(%~VTkcDi`d3(Q?V#Wa%F0KNC3-A2L_Lf3@!CPG?{-@
z*#9p3_qz$!uy=Cu3xAPyroiw47a*qwWHx0a!b_CfMuFV$i?)H2yeQD&+JXA=WkUy-
zK*vy|;bTpVJcFR6D&{3{mf5q8K^jNE)3O6s;hvMXw6y@N-9%LYR;#l9VSsj8fAfg~
z%>)P*>jgQW-}+rv2i1M~b08tu$SHJYA^`e!bT6eGk=I9ZqVkCr64(jqhhry_;Z7k8
z%&M!vZ9<XSdKI?|Wh4Wsc@2Z#@*ggbe|3`nDr95FOA+z=%LG*%1jDZbPQB8oBRLAS
z`#>ocNtW#j$+ke5Cgwd=!84d65Ok9EO@lL%Rdx9D_vv7n8iVhNfQzTz%;xPng_;8g
za;w2`4D9(6<{dq7Otg<w6XyC<q(bs02FTV`HSlL!LW_Vvw;uoy5;zHW&sVih0$-X*
znnGxN%G^N5fA`*8hS~{KQQy!1>GAnVC{F?1l#tln3>(l*E&_2uqX1NH2MFFu>c~nj
z@HVq{I!X-u{Bbk}@Y>s%9?hE&NEK+|O_X6uuOT`vLa1Aym0mpc{H1g_R%(-ABRkk$
zY9mn{QgLE2+Y7%g7J-{w6AF2ia^<!Dn9%=?f#l!5YXA7qPeK~32NHdq+2H&C-Forg
zd?@9>3aSqd4>Uh1{z{VZm)H5v!!ku38j$K&i!S}KYx_?x&cFY>cMN(kWB2ZU|5Zlc
zf4%{W6kx!uP;|Y!`Ok{me{;ouy-7R5pl`5SO0DJfA20FGm;1M$4+((k|42?S=f%Ij
zzP}EhzyF*6?>~O)9N5M)Qi*)vdZ|_Sy?gP$8Hs<pgTLQ*=8OEG#4`pWLZTQ;EylF*
zra+XC{VUHcWqsoJs{hMhno>o+2y9{yoY<bnNdL5N2w!bE6jWlL>wD$T2DtwRgU)YF
z!K8)Osu-wtOLITif~yCb?ov<-tInm;nE(3|{vUnP`vneoGd+;&pp-f%7r(wV+92*%
z{n5X^#eaF_rI^^kkY;-5sr8RQ#s752|I_Cw3*=x(FO(+T<o#8l`#)aSzqc6v&+qVJ
z1vI29>@#2d?{2`3DKw;~TXa7C@7|gxybxLV;n>@+|M3!kIjX&4Aunk8rk2Z})|&r3
zgZ<YV<Cm-gX7uDS*Gm=uyBYoeKK}pv`2XJ@{dGxN>Mxo>4&j{yvtX{7yrlKpzlW)@
zXt4Q(Oa2@NOeR(+M#c_gy2@MTR<k4g0BVS50?KG7x{F_<P4I*dZ=`c#tW><lvhas^
z&1o&~UydlhC&1gR%lB5|5mRanXv%;32{LRzL|2z(Qy{5B);wfY6bJ2$UC5OP*p++<
zw%Kp(KvAz0s=5I&vVElL6=gkdh=u~V@8%iKP)H^)JidASj^X!jUbMvrDE1ITkfsXl
zxJN)!@smZK)kHM_N}VP164nemK8~xIh}o?>S`&r1{i~2`(Y6PibxMKy<Z<v{9HO_@
zKb>QF&H-t_X<4q3PnJj?_Jwbi?JUg_SqL6oeXLc>i(mM|#k61T*HclAuT%^P>}31Z
z$uP*D8Z~tiwYXv3%bWoA{{v0IV;sroMyQVy{F(ig=Oaj;4P|h=L)Ty<Pxf0SSAZs(
z)1m~Dpy<AT%u==u%t3+7VFL$SS~(r}Bh@Tf6>x2HgUPbgeZa`Ig!oYdxLCVj)oT%W
z`i$b%_19}L6TqbpRkue#CD;Mv;uffht`_eC4?+Mk1KR8h1I^8EZnV$+s(_tuBnoct
zDp2MW0erzE2*(vR7cl+@j5<ZwZ;jt)kk_=Q<_W;si~!51nIe-{aVVpBWyht)vJb!w
z<p9C$s<*+JDCkDijY<$_WT+jscAN?zWbNSMkh6q@$M(g23a*2iiZ1u=i;=T3eyx(T
zX+)qwev#d4f9m)Y$sOP_C!I;~1g<T6cq4R_02ZfoXm3you5RoD(s;agTYeHGjR*58
z#!q4j<Ccy}sQhdR?jPku^!)-a31)h~35Apn05b19FS3}@KeG$f=@1{(K(DjG6-251
zmQKoUMg0|2!yC#MPJXVRG?H{eHr7`7iT8HG<S+;iz9Sf}%{u{(MXtaLLuaPY4~Y7<
zQ2*=$ciq7}-5BUc%JN_DlcN}cEM;@D2*vgxe{(c*asrSs*#jZIu;Y}}tump8E#Ofs
zfQa{G@7=)RsR(iafflPMct8WoPp&1OQVhL76_O2%s<k|z9<S-T9G<cP%BRJkS0uPx
z3<BSxw&YZUJ|n2Ld`^xg2|1_(f5%PPQY@a|R$L|D9nY!zvFacTLxU8y5uhR(wxF7(
zn*z0XJ}*`%-a^2^8=R%4HOo4P`$LT}*mZECJ&U4dQ6uoRN_0}lyx5W>Fo|0#)6XPb
z43HsWw)_GH?FyDP3?AO=Z#f}li8MmLhJv14zQ@^>qG8RV!lkC5TP+mEao@I-arM$R
zcs#4w;ajWX;PH5<>w|j8FPG>NqQ@1m4RH1=0ZqVn#pPa$uLDP+>|-chHZ>i6GKQWF
zeGq-Y>4MRBC2;Z-6zgFC?FA=X+d~j)^)Scry!05P*pR6OI)V?DiB2u>kyhbX8?IOm
zNrEu$saa)f5IRDHV&1d-CbL*~M_v4rHh|`5Q9Qlg{A~3WhOQd)x|_CVfS$q<Esru-
zKSO7yQ3l%pgO@l@9wRpfMx6-y(ghH>0sjq6Cko(NqIV&7R6g+0DB{<jzyDYUI<Q^(
z?HSCy_<mf3DO9`<Y^FIban~kp4(^`^DkkxnBh&1+Z?Ye&ypjlPoq7f0I9!4%T5z-!
z_)-McHx1RNfPq@b+vE{=E5{&R>wCU>pGhdH{zLe!jqDNCd%zrmV=7a%#xu$JZvAGh
zBzcKfxJaAUCD(!c`RaPY>ic`Y+N4Ptv-Qs5*NfdJ@W8}fEY=nznr7-@egCapF1D8G
zc#04;4+V&<0R31#K;;(7k^v(wM-|BJu>U}%Q(SjE$EVQt-j;}L0yvl*px)(b?a`Vr
zuw(_+M>y86sw%%2MjCsPv3%GO2NSXmckj?Qyypv9znvm1`%FWwKX-kyQqU@55DCja
zMgAKn!mpA@HR~X44S%`i4LEx0(&@9}*N@&<vaJF&q;RGKaH-6b%TO#5BzFfF?dEaW
z=f5sH)=~9PHt}`Z*x%Q$P>J_P&~6i0ciW6yXl-ji6r>40#q*kk`g19tL!>AO4vS&W
z`5%EL$QASE;xW(k$3F)WVfjmKtY^f=Kz>N9kr>HTTwrg#9;rB56gAE>Ap@6u48R&%
zR4^zerxbLnE*Qa14nNl^C>dcwF<ecTjdd~sJhh~KAF{&(!=u%Js-PNLrE?NgDS7kj
z@pr-6CO?K?p%F&k18t)lO`<5&i{A(FPf2LL<#&S`>;kQ1<lcWtXX4mdhSHgYX=Qf6
z1&hgijBHJhymg`U)q(Y5@9P%QGa@R`xeuzKM-U!CMJn`C67bQokXGM?=q0Kf$@2%m
zeNBTx7X>Zs=Wvv4scG_@yeEhKTy^jfL`Q7kpCzd^HR(?5EIVtz4@~0(lfcc|Os#jH
zOdqK4a!N_IB@TKSkJVXzM6F+*e^mS6aBtl>cQj*LEJWiQ(0QEAdvJ>IJZ^!DVp6|^
zz}Is?9AHo~s)whKOhL#}0S~0`VX(h9r)7~ztL|y_CTN{w6o36Qx_>^rnN;4S5KQ)T
z_O~xn3Dm~ZE)+AZ_UzH4e2EOZNoBTz)@7G9d#?9Lj*SSctN$Z639~g9DEg9Ia_#MR
z^XPYkQlnU>#uM4zvJHan*p%?kH+Q<AJ<JNFWb$;Yz#UZq`C-)+ie&Q387$v4xlR;%
zN^Aw)nNB9<B{ovL>D57@j?B{iZANq<B%C@t?R+zWh;$W4M1B(9axBM`Ax>-Dc{=Ku
zKtL|*igWH@aLqpxi&oAB3DlFeF5WRvImnj;v%?smm`*kfJh@kD*~0mB`07=WkawMU
zBRgOwmorxDIS2BE=Rq>cPQIOfRF~ZwMjPq)k{^)mw)m7Ju7kZ+`V1+ffym2M(WQ^Y
zq#3PCyTE41)n)4W8RDQ)7ZnJ08n5QNEmf8LD^}I-349!=feWZ3YGNGBf7YcZH@neA
z_-{>WfJxFQ^CbKp*9kCbegIso3$q7+M3K|a<26rz63|Gx1cjc=>c2QySU90OD5I<$
zJ2;`t*d^mkbwEZa3JRt7+-Gpa0aw@1*Z7YrJ$E^nJ+zWS-(#iv2#i38dVk&Oi#=eD
zWN<d?Ti5h{Nz^u>1+jQ~lZbX~@TS1NgK^>@R!p~i+DH4K)rY!5Z)4i$D&R1*2G?Cz
z0W0%ouy^BuB1c#am|rjUKL#5G|LX|TDJVS!$^r$<QnKm-YcxnV54<KvU9G$Efr1&o
z2~duSVFz;)P*OIy=iv?a>wfHh(?|k|0Vj#|(M0|Wx@YeT5RnnGNHw<f@g7o{5Mo3?
zr5c1isV*Lznr1TF2T}^rd5smJ8F9O7C(J<_Bu+h>txU!x{s=<@U=yGI9_=&Cwi%v)
zuD39S=oR4_c_ns|&V;Fq&BSoqeOA(+eUuS{fLM6J`6tkX(czzXn!?X@G<wIQy&=Oe
zu#%y~wVcP=ugRC-s&2y2AR-XzKT%Ig{v*Lj`El!BJ(QK?vH1acF1*$8btuAOtE;h0
zMFPbR`B_<2FM+5Nonw^sYe^iC2ndf_v+4t%_;z@7SZgSq6~wB2tA?iLgY8^Q<{-Ev
z(+KeGWE~8T+E6G?mOIUYJIdAE+u`%Tb6IR7zFq^0)#lX*+vxRvpl1@VR@xRP0w(j1
zy0X?)fRfTI>3Nre3mSe53*cpPvr#o;)m6%7V@zFvjLq%dN_R+eo~%s=3gJ9rQN(j&
zj{r#I{r)1@!`p%4Z#FtK5)NBZsXP3XHxnq1{b0>{$Z>MwtLq9;L@q|_rkXKAIG7}`
zD}uJH0gKUBP)}rW0l70{OjX+_!yXi9kKmc(Ds!{57edR=T*JVnrC7-BQlF;8&O9Z4
zO7~O~@0_ia+%w;&?JW2YBediQ8mKf7WN>DMC=*8&T<TMDp)6EL{HNn1uiDdI6B1sQ
z-B$({CY;yvmbWYRI!6b#Rrx+?sq3tRFd5dp&z|8C&7$;+W0qUp#T0|ehXRM^mf2%_
zZ*7F47r?%`FZI1!o7pVSRA(%K)dlyIu2T+&%q<vC#afDV+1hVVVxP+jkvdw2)ilS!
z!wty({dH=$UJ6rZ*l9A-5K5o7@QiAaOH5PuoY+pbnjI))_;ycma=+!;y}POtbX$`k
zPwOfMvg4U};S)CaOKi&XJ?ZNevdL6P82K)Ko3XO=$Mf$-+wq@iXvyE*@hJf!qd-7f
zX?>Bo()*_G%^Q)9`_hNfZCw?;MgZ0HVJJ{~C+%c9=4x-~VwLp++erOwvHeokDfo5#
zu=;idSQ6F=uhNoNRoiq%mAvKbyX&&XktF6kEsuD^>aP>f3<&n|eI)hSm!<2e22ld)
zfh|AH0&jqgxjnAgU$v$FOg#ZNyJY7j5Lkx`o$bqoWD|N0#zLRd1dTw-!@N4a`RFP3
zyDY;M`Jstm>u^Wyoq<pa_h!f-=a^9c;?^j_qri{f+q^osCY(s=K_!2S<+#qRC*OiY
z%5)4PwvXujM#$=g18Rs9uDcvmetS<R>gTb`vClAz*w}qYFXG)zrmIRYWQ{)s$hi;0
zs90$8xoNu<6M3&=92uvY>y-mp^C&6Nmg}>z4FZ@ZQgaEFgy}@n0Mb4>CAwDXXG2Z|
z_0Vw)>y4OdBRM5*xq$d_Mh0J#Gy3)+IgEBu)~wkN-)0J6%7qda(-Y2W1i{d~(u5;Z
z(Tg0{_fhuDQ;y(WzgOG>LAVn6&Y}(J@f7RHbO*?EyMLbAItItRV@H7b89r-2`jBN_
znl0r^fSohIePdcBb_*1~S|4i5^@xC|)X+pi<?t-^x03AVlw{>BpFW1!Jd-G;WT9aI
zL<`|2m&_zrVgO2y$nYz|wB3HLgw5)Jk-cbT^jOhkH<hD2%wgL(dA<?UA{x5W$2Hn-
zd<ma9z~byvj#r+AWDdQ?(2<KZT}p!WkW3JZsz=nE>!(Oh1Yixpjfo1W_o;cUG-}1-
zJFKafWZ(EM%1|oQa!nqB-93L2N8Ky?TB<{sW_cSS7*cvE`eYZ$Yp`3ew~ql1y2vxe
zM#Ghvdj=tk&G#>W;j?~CwYkqY`8a+d)!ch0jp>}6r5NNRHbg~Ap}&qHPM<Z!Ck`-t
ztPQ_}U{txH9D4XEGg)>yZDN`rm0Mc7lOdXjbgS!rd}-U*QYCNQVmi{m(Qj$!__jC7
zK!|9LPDVFD3U)tHDoT=tkIj%N@T>*yO5o2Y&`CqYJd5eMPTw~hK3Wbb+l%Vy9mCI*
zp5i!SaZxH=du;=&mFL{^)Dx7w2uN5y^xyJ5jnPgm8kpdB55CgvGG(MfV@~!mX+fhE
z6*sHYN>)EiVs)ciXq0K0+Md;(hN6(!dj57=O`3P&qjs(#?F`po2@<m(;8<?`gTtO{
zma(1i2a0Xg({4C3?6xJUTTnag#IpJ&i9N!eT!ui6-C6pA-HTd#%U63!2TP;7+YFpY
zlZM!&P4T{ix3f_s1ZJlfI4|$S3?AuJ&roRxYzd!ES_Cc90te!Vmsi3fsROSs4LlqW
zp4DybmO5FSfkv+{7{pO<{xDOqn^#Q;#EDGpy5w+da$6aaGJMke7M-1#UEzN3%FjLw
zcst}RNy@Rs>hAGF=bv`t#PN<K8+Dn)U89}co(R{id+$}*8-(P-ylmB|>=?lmUlvML
zaEd=DOn?}}yL`QrcW28}HUEv7P~P5q&C&<yeRDr(1h0g(u!k5%Y|^{-3mN7Qm^j?$
z){(8Q$@WXloQA6g!dzwSW>xR8DCC;`!2(F(HNRt%0GBij_Wr~TM|)f&x#ykBn$HYx
zbF6vCLX&#tWzsftnqm}d*}#uSY>KqEQ_rZe?&Zh^v7qF5impBiJM{=&AfZ9&)YW~N
z96Ki)W<LfI)zj#OG{Y3eV*DiAUt5jMNy;Oz7dqu{l%*!#em3x^O-`N|8_dXmUHf61
zI~A3`$+Bd`m3FoZdvn9MN6o^?3G*rRpEbLyITc?_v$g6)X}<ohedAz7M7{_c>`6Wc
zqpPGg__`AOs98=PANT0p*`z%NFu)xNhU5qicbJ4W`d2nMTU{vMa@fWVMX<^120ijr
zB(qiMc9QDe;q1R1WiHnLYG7|(W=wtZwYAewAUMIj>QWyQ?x7;$4pya*+;1RN=pvx2
z47e(Xz!9Hw_<?IXE4wh}DhChB^Z4>@^j0|idfFWvo9hA|*&FVjE6HVsw^z~q4@v=g
zkMmh<28nAh*3w-gJ(bhUpJAaYs2(Zw4E^=s3a<iLOe)+{Q?o{)JFU!mDAvoF>@bb~
zwCrc>hWX)hrR$qe5?;mxi>&3{tG-zu3Eo(()}8#6-78yt-0+BTvY5w5tu^u?MuO{+
z@?djtL`QE-Z^He3rp3gi)C=E3iVugJ`w>eY+9R;`E4&G9L>H+1^S8IDN;*cxM(Ib(
z46-tVnJwq8gc5FT+a&BNVd{!S_cM60aLemLB`(iMO8WN1w~B4;dqypTn;+;x*kRdO
zrY+eTq=49^OH&3e-trFxt8E<QtZ^3lmjj<rzU8?R_WE4ZDJ%H<+SM-_W}_(amY926
zoh=%{wYKKfj~Oh&9<;j_C@^H$4ZIJhh#*Q;^39Feu@`($tr%CKH|8C2luG|@h&1TA
zF1yvyjz3G70-3I2xAQHwDb7^q!7UduaUNvEjZ~#*P|$J1;mSHdNhH@n^dIBw-3UB|
zE*Q;7b$YU8FnFHwU;tB!GLeH{`N5o|Gbr>egncPeHfTSL)sX|gXnE(NW6`qF+>c?F
zjRLHaZt}gfT`CumEgs6dX3Z_95yU113eyE}zOgd9?ejwYv8O}s-yz3yj2_coErH5M
zd>$G^xH6rKEVb+KP9{80$>A<JSmH_3G0`c4>KZz!X}YlbU}@jDZylF9lwqC5^c-Bb
z7`r-|uBQnN^tV|axY&r^Xs5tKa9nG?+BH+=N82oY_9O_)9|p_HxX`JW)kebQex!w6
z%Yvg7JZVIn7S5kP-PIWaHWKTDKJosaqXydwY6+y2g05##9`M_aOk`OHm4==wSMK+f
zD~h;4bZUvwe>XJPlY**-i($+7AHzYHpmLX~1okXH6IhsGgD;|={S;JZlBu@TnGZZO
z_PmBtz9VrlBhu3`LYP^Ci`anm243Th`|dpE36oa}&uL%(4p3L45zcCahcvZTpSa?a
zdkQkHo@cKKs1=R5n$N3ztoG#GeIjfug09ze6yxxtKc;rD{g!g-OcE-N^Qp<$s)D!X
zjx^myPb4hXDrM+e*Fa)%(qmTn11y4ROpc6h2Gnm5CAg*Di$cptqTSqu^n!S}Z!xRd
zebtFbhlXK=j))>rP3A|^m!6@|OjF}i?bWS&t<!m|+cLvsqA^nEm2dWaR+vDtP&ex+
zHEol_eZQI-cr>~7^;-*T5A=V1_HNL6{Y=o8iU_zbf?TM|74uxz=+*9n!M+Ow^-)aT
zRmE{a6RbU_#taYpu2oN;)6`UEo$Y4VysDJ2pJ_p>c<XYo$>7!s>sQ>C%$j_U#?H`I
z)cWsl-*tRhIKOG)N4yHDpVEPM+47Ga<s?mCh<UvXVeLKbC*g`n{w%+@{G#WRH)*7G
z34W8UPZ!m9Z=rVA(5ses0DzPTLlfGDSS!ZOc^sC2g#*E}|BM1w!}i-M!cXf*RY1V6
zvSkg+cJ;ceEBriv=TTy4a`iTRx{pfQ?7rUB8!6A|d#IO_pO0fSkU{m)Oy(mz(UUCc
z?@~!Z)72JNiV5;liAUU{bxdpoZUosai8Q8?3o}KqzIGupxslfdw5;|WA}>3<3dWm4
zuT5B^syZ^9-L2#U@-Jqupza)-FI|f6O=9zkfprfcZ(Q+QF=ry-J_Y49j1{)ek-OLs
znM6mKxxaO4d=&Ypex%UkQ9iC}u;&IeHuJN|7Ds&s@%=4=C@=Ozqm?`MF3I@usL9l+
zak|O}71fs5?R`Aek=XyH=-uUxcpSx{Xmqs2+s@)+vd=|3TpeeN-c6Lxsq4N7nEOG=
z)3tSNO)oJ`()*)|(UtH0*S!7@V_zK?RokvRz(@#4ch>;Y-3&-e2`DMuC8<cKbf<*G
zkiN8p<dBj|ODc_oA~n=dXN~Xs?Y+OV&)Mhv#g88_!>skJ`?;U%zQQ}Uxg#3SDp2(1
zrg3z(xm08qsJ+<beCR8yo=CS$4NRZR5sk1Kh1-eA)H+&@HX?QE*z1A^Vz(bianxsF
zT%CK*YN?<1rN-^x@7Ql-hLia+y8y?YP)YtT$8@H3d4pnf9Xq5nYGSZ&T<G2NQi^3C
z*a!`C1lRP7gN`QPr+LDL%V$JMuOoJ6(?&4o#?^&4r0Q(Mw<_c6NO@dw{2bDr*mZYt
z&s{cg>pZ^PHW4^04C2y_Ycl61c}yenJ7{X}irzQqPn{JUfc@5k=btr}FMJr`j?%c&
zGDgrLXt-aB@L3LDHeL&^6lxk8N3g|;)DPdb)F}HK$ZGd%o9rdQW7)b4&wQxg>!@%p
zmFV%J<v|NPC|rY=h`R%qK2sm0Y-jEBx<W@+C%c~T8X03=$q~C?=1sH8nCRouDDcjh
zFTPN!YuchX@!AsI7abLyBG<U89u*E)0k%<~UA8!7DRj6B-J<O}&ZxyRBbQOE9GDk8
zp3j{&Op#Sei3^7<V`;vneqJD|NZ82sW>RHRL1ePWn@1D<<53G_yHtJqtTN_&4KTH8
zzD=rUZL8A{7odBR(iRR$Z<3{vc|PeC4xyeGK}-U3ri8U_iA;5fYqN;2w1X#ncR>>;
zjLaI{<u<3oZLTOzC|^@FzU<10VY<Jy`<xN#Qtak<uc(mhWiKSxg_zVxIz*j%V{7~%
z+%{heE>o#hurm6J;(OddWMT&H!T1WRyc>p?4d5Fo^zJ_NK2pk2Ftaegt#vDtC8$Wy
zwWd7ZgKvwNyFxoBkEG7aGF;VGo8)QKFBV&4(=)1a7!C798YPVg&nFEk*ORo;?T7pd
z6D3V3URG|R_AOaU?}+XTZS}cw^WW3psz&D(#z3qA&CL`^g<OSJn|P@9t{!xVVuR#Y
z!VKS@M4OcO>NzZE%kQ@2d`Gu=RJgrlvOJ$Xm-tLplUl@tW+hAZ%^3rC1;Y%TW;|De
z)Q7ITt=Kk|WipZFJ&u-WJhzv5ek1en_wPz-)z26L66ec=Exk6si}P!K-<vkuBG?bJ
z{c^@Tu3oU8aZ7JH`i$Z8;_v;%<*zddcCJE3iS(Ts!s$KEoE*BvblRaTRoag!)~yTq
zc$7#=zu3jTP%L<h^guFe4vEs=I}=3iHM50?v@VL1?PNFJ%#-h4H9pR8J&7-~-}c?r
zZ9WOg3(9I1p$ITQsHoKm9cz^y{xS~qnDbQ1M>hJ1ELCpJj*lE{TqSltmbX~L2@&3m
z-Lp=_#FQ=7PyjteP&=;*<ggZ<$St%34p1(In&I(9;h`W*qWHtuNRl1AGB(iEf@5SR
zewaIR;I6tkwf7@eu$*KOkZ}JEj}L{o>z!eC;Ykl;ZMH(z={J>c%gBa^HfW!7zRLm2
zO@Z~4xP;kbVf^<<L!VEDY@(zPsNA9;CmR|SamPucA2tywxHqU~T^zlVG8BHJ!sD=0
zPA0?2pR68fFr^g<AaTo!Ll%@ak3*%0Iy>TE@Qm`PGnCI#;SiQQ(_lC=41df~zmZCK
z*UNc?JMsW7sbm6gr^H&C%JJ^NM0=9b5tgakBnx*X%s&oIxyM3FLYD{#p1kMdf8zB{
z;tt?o`eGi$vWionuXez9TYMglvwe?V5nd?ifb18u!vR=^p4n-(!kqI)7N|G+Bxvrm
zYWFuN{9e(75|k75GY+}MOHle-l$6OP^X@~e)0(fQ&kgaxa?5XC-2kw4Q$6KNYAn1!
zH1XABou{K9&e!CF8odrV=6T?fKtN=p(Lu{jLz3O7(O_y)+{wiNY`gLMIe8RYr<Xx1
zl8ad&!^$teuIMEoE<XDzO&ztGyoELE2u+wRML4s{JUFtfUJBDJzD+w|on)G)x~nlx
zyoU7B{jVcm=SM-k|CEqtEZwhT)Qo@=>5?3~<XQwtFQB=0xgzWb+GO0m5P>5g49T_)
z($&hZe|H!Kb+HLnl^1uu3`d}4tEkUMZOfjwD%_X_1t++A=lOJO&|xg0jYoBd_E8Dm
zPn%_$F2)eWq2zefxbW8FwHIF->lt`J!{y{fo<cba#G!awMbN`f0qHaa@Rl+R7%nXC
zM4es~NG&TI8d{?oe?6esYnX4B<vOr>0hyqicyrmuATRsYrJGe99pAy%p5^1__)eT{
zzCK(gj*hj9&(&wLbNTv*7{Ttws+`2@g;BH2RM(qNDu=D(fn(QVSajaWmrtf_s)gYe
zE}=X)OY*si2@R%|1`z;Zc5#G)pg*ejbiSYwsVfKV*Y2lGjH%kJFXJD^JO-r>MQTcJ
z0lt9tTMfGl>9zq}#^?cZez3%Fv<+>@hgAk^+8I`j8k;a4^dP?1J-9AbRr*udu@L)}
z7(exQpFZ|QWU8h4=H?51sh2$sM_iWe{7%QhX2d>?YeowuZj*jVSFkEPRqB$pJC>uo
z;+OC8_=2KGs3|3Ror5bKS2aXq%H~*)hCJ>`O~rO&6acNKebpIZI-}TQ-6y{+@8nwv
z4N4JB|Ct><+zyOk&bG?yZ#-UdeV-`rWKNJZoOL>wYDZj}@*GSyJwNTK*<F_1H|6<W
zp`IHPOm8CWM~NPcK@Uz7dK?;}9#~>p%rjt)$o$x0yCdi=j$nr4=Hcz}08i|jDvd(S
zI%*Y=4RMIReB*7NDD!D>x{<fEx7ieG=pB{+?iDh9e<o>~*OCS{_9N^z2sJima+e$L
zfj@m*4R6tIW!4T{Y&YKj%DiL0b$%46Vo}OicL<i_k+8=IrtJItl+%_Poe{%M5*zl5
zt*0lfFMBMyhcp3tpP3FFUzqy{HEoQ>oCXl4!i{8WJ0fMk)Q#1V1w{Pr4H`=@UVl!6
zJb43QEda`onoo-~C4Gf}QlWPFp+YtUq{l#&0W0OpUy6qk`mKgyQRBj6o(TRngKqzU
z^qk}^Qji3gHlG{qUl9jF!nN(vA037vG;@Oipgf7vB*qC-jUg)7F`ciAWI!2vJ$XMZ
zJlxs+ft=u=nz^2xK*0G|&c$PyN{>38caDid95&6)1=t~Y69R=+cym)Z)6%nW+ZO?S
zuj3ztDcI%lKY<9MPhxHpp}Cu_cz=`1*iL=n4N&jiHo&9{4)nS>_&z6Vkk~oR)Lv7W
z{*-VON8_LW*s}{J7pIAS6lzUiO+Ka`E%({3T%(o~EH{FAkT6YBF08D8B#1`>U~;Lt
zDL38xHv<Lt0D-i04#^GGo#Z7-b-{e~$pfcGpDn4w8u3fVmBqZYEt$)pf<MIs7)zhf
zOVFc=$s{LgSQ)3Ajn}tBy&9oIUGAcn&Kf_T&JH8j=A!&uFW>SHjv|h(i9JC;^oDZk
z(Vl<w{MbBAxgq^KiVwJF4g>2u*=j8mDzs;Y$cNrszupsuPjoU8nenWXoO!uyoy6`b
z$ujE5S20hm^pRag)T_r1#Rq+&Od|88Ov(zY8M48Hj9E~Gd=3+$QsFCMk2daLNVJ_o
z$D*yie2lj9Hj36Wy0k|;-GP7pTbfoZ1Nv1zVB+aFJz5?U`|O8TPGGbdFxKl(%2U{$
zcPLKajs}f6xo~w*q%ru~!?|=Wy>|b%%E4F$gR+IP#%g^dz2P-Y3(9`d4}R!^zB=^!
z?J8dSy#b?mfjYLL_O<Pzp!7$VC|Z3Lp1>ux3Fa1Id`yr6RT!;$P@O>`fmK=<tFd16
z-f#<cX#9{NHo6(-Mg8kR-H9S!gAv}-`_h0|P{vv8(Cdg-@e1O}aImEKt(aauU+&@o
zYVOiRF(aiKT%vZ>{Td&}m~uZCT|eD7_hpycP2qzLq5ko7eJ;z?hZ`?Ou=nZqF*b1n
zG&TmpI1fxmUbTtc{HuB-TXWJhwkLP$-GRu{aefg7sJN&$7PN9dZbf;aChTZ-$x~^p
zkX+sG(Q`QAY?#;c?BUgv=yn_BJEddxO>6vGeZ45V|ID1*xzXpc-TS*4&q2?X508Oq
zy>nZwo4<jA*xsVuv5U(r;pJO#wXPxXt6%R*3r7j1*HMg&5Y^<%<F;rzJQEIhu(!l(
zn&c+OkCSnS28G3h0!hv3(pf!NbqlL$a&zI<a=9v-au@WCKFJ}w?Gr>34UczCn)RST
zK<ekB*C?52^hCnly--3m^E{V$Uo$7fr@&FN%iZbauMV~3O3=}+1ORE^x{oH}H2M&o
z%L()~2Mhm43<Hgbs$+*T_a?{etBqFgCqmf8%u^-bD|)<VKpcOO#O!%v4vykefno8W
zEz6}WE62-JW*Kj@&8~__*}wFtC2M*NW>!6R-p+68&{@rNb2%}Rm(GaYI%Ymj*JFGr
z`Dx<j<XcJ_*9eu)DUt3Fz7RgO%G6iJ{xQxED?_c<A(Gb|#VSupA6z9~O02y4)2eIA
zTj(GO@bh5U$b`Dl7GkCAM8R}rDu!ab8kV`?3M!Fl*^xRK&l>4MY%Cqc8xH1y2~H~7
zV$2Yo{=TA(K%X6ZQh8S^K@s3s71eRxdC&yc5vpC{eg8^&SA_ZrK`AHf{g!-qP0@xY
zUc!z@&ZRvvR#i}8ZtO!b4v`LKCx`CX{E~3o(s3zU$e>x4)>zn<beH4ExNw)6*p!Cd
zG{MsElr_nb7UC^z+O`tq8|EI=32Jkw?tsmnzI_yuW8~=PGlv(wO1&lL&}pj7=C4gv
ztz47R$2szro~*k5SyYBz>FKR>5Bwmd{F$@=qG;N7=(N4;Np{5!cS2GY#_WJz{5#Tw
z!lN8&tgUZxIU!_JZiysR!u##f6S6H;%Kl=izsB&X1onYbY-ahu@41TNv=LFiC|YYB
z<&l(*8K-~XsKwiBGa}q|U_wS~%bnO*AB~MRJE#-2q&RA9vcjROyu|h{QO|y}`?q)P
z%I4o-Z`}u?WRgilxbW|RAnr#dGvGpwDMM0A%fqnwIkE)F5j08h0Zd6uYr7P=bszZ2
zyCSb&6K=yy&Z6!qa>g$kQ)r}UlNAIrt<_WI`TEV`3FIb6A!T_8h{WsaF7iv+SVD;w
z?GgfiH{=AX=dE&zIgB|=Ny#g5XSsFaZq&$m<KeG$W~v|Mm9Ez9-P%lp-ZK$H6+1Qx
zj@IgE>;agoOi@P@myLS3qv`Zr4LgX-K*#pV-x<HReP*RCe9+mI1ml<=#K*ro`Wnq|
z6I6ID;Yf3nDfy`u``&Kdja8O~3wv*sOt{}iQrK2yF(ly&f<wtSj``GxYS3KCcT7nU
z-=Bi$Aq<<CrMc<axY<!i(L3IlSSps%@g<~*XNHalLj%ft@Uwm8V-Wuf34n%FEz{>#
z;jpA&1Pdyfv#zx1kkIs3KnD<!EMer#iH+V=#LwvJYM1HmbyTw4f%*-Rqa8NaEArCk
zljmy;cE617uNbCXF`wGyeeO#XRQG>|F$%zr;sIqcAB~s8&#QH6%05-^w!!yJv+XFp
zS19McckuB5;+ttl5{EW*`oA#+<G&LA^>`<*)WWA&lbn-(y?E%2^pt+_$&}dScGiu^
zT-;|`5f7G!Q~MKMdd54EITu7qFzu+wAm?;f;yfGkjU3S{>TS>6DT+B>15Ri|qtCbL
zkDbPfH{6zMUBFRCE`<ND(TBpb>YAVu>nL|dJHMap3i`!G`5B=3ObY=0v#+o#PWXbg
zYnf-s_!pI$>#zY-mJ5wgAsxol_yc2|9$z$3TXISl!1Q$JO@{No6;m1Oe7>!qDz+sC
za7Wa1kUG2@hG0lOfTVm9xIZ<=6>&9mK3T;sG-VY{t#Z+8p?+7GZNuUy_awYKv?vSB
zGS`ANOt@@IY>AgjQg*+vEK*AVvFcJD=fU3|SoVb)EEgZO1Nd{Rm1xy`l25}M;%v}>
zb&C(Te+<+fxj$e$@>Lo;<m9?+SEl?@vyDir14UAt)Opna$+P&TTo3>-+o0(#2}q@|
zK%kRYb=mNL2N10Q@Yx~qJ&{Jcokc2<JuMocRWhtigt}^&;HCU%JC&9uxL0~SaD_#?
za`vpz6*AuE4si2wlfGWQzY}koBSUjGLq|*Y94_G>3XC`KELwS@NF+okScGR~l}f#D
z`k3nmV!dAl?xuc8iov;3qBca3Xpi?ke*5wJuXeMqlLXxYkvd=6L~p{7s=iObnHAVg
zM8z+2Ez-+mBC(+{9ooK+H+!2A;)!dt7!qm34%_UdQoKBWw`u@(bxasi5e5&_3Y&mV
zvU7Q=MO2kngy`s8|H^}vEkyV%wVp6YLKKA{xCtz23OhQDPCp0TlX>41N(nU=yD5Bp
zRK99TX7qY}BW<}PT1eR`r!2UH!8WWDHW9B`FdVnj92+Eh)>gw_Z@HS)9VwtSp-r|6
zak4G+$XaNXHT)4X4x8RSFWr}IGk9!0E6s`})Mm2;X5m?LhG!t+5x6g6BkgQLn*|kf
zVjtSsdZplm7$?pJK;}Ca!SD2ml&b0FK2x$MKdAd6N8S(l_;l)H&^I7rY?gOj>Y*C^
zQtHVI(Qod3^(5EFKb_O};f|k;$I~=pJYJGqN_DOPDJc|lM)q{LZ7GSRD@xo3>DG`a
zq5Q`$^KS`=H^hxzpj+T`Up4w{)MamdbMj3(e%GE^DfrIl?5f#ktD$X2S!7Xzr@$3y
z|Ka_S`bv?izumUR<%|<_Hp`4{-Gg^0jPR7{NK}(pZH~AR1jlT%0-gv#{xiy7Cg7mQ
z`wDFZ=8#b?C-)$SI@4m8w0YJgfw|9*KzU7kHKsK~eCx;sI`NC(KW#lXSdO%jbU$$b
zsp;l7Z2_gz?gCt?2CL=ASwQRYYTE)e)1b1OGIKkwEh3PwtWE^~df)wWFO}5#P_*Pn
zAu4Z-lTgaL(uA;g5!)N{kONBm*LF*cL`vx=_lsQLX~#bAug&z}FHXp$bb9wvsm(jY
zII~+)$Ido$>^Qx)rmG`3h9=yce<*32GJU)kt=7FMMeD7<XVLUPQBq$7m9?+r?t*wm
zH%T#R!@aj2={^&9^U-5yQViRq7$g+`8dZ_MzYCE@<D@CpRYSM0i?T#7rnzxUo6M07
z$wn$3zeyio<vtw_Fj)*j+Stho5wco6<rK$px3i!uujRywMPwUm4ZvQYQeSk`Gt<=!
z8S>A4T!en&Vd(mH<SBhOC|Au!t35U{CLTg64`E)m8Vxh;Qn_}unuRNH2CHBuAjS!A
zPTUm0a+qJ$YSkR4gKN2Z$ACCatD9F^Z`Y`&BHt>PV8kD8;n~=gp2)<^;2EVI<bq`q
z0-orLYqu@{6G{AB8=KWdxZy0ajB~{_IITVt+d;fJdcKHFJ-#A>CD|=J<qg#}CFh8w
zVR2B(IcnSIz)GhWg716?7-dbzfpZ;1GTG%n0)wzKG&ne4`q-(i+w=*%O_CZSkmP(x
zD3eZA?y`Lv;p|Y$FQ;Q3y=<Bp$g^)TKERZwaw*66<Hp0AI3eOHN$R(_KjyBe+T6Lr
z<H}Kiy3r-p=5kjdU%;`-n10#r?8j-=NxCiUgT-^^<Kd2o4++y)=#2Xhbf|Z|+$)5B
zcXP(|q#jJo+#y^KUh^R<B1$v4om@^N%&pR`b&C_HBKc*1d{3JYTjgLjZ<J5hq>a2Z
zOmuc`$_`v9c3Ub~mb#gmOJ!$%F9!w{sHKGV#KZAv!ZM8;Z+YDgRCA950ZyjkVuT~}
zk4FO`te{)U_Cbl4ddoL+|3X?BLxj>~<wj=~)Qj${%}b$>5sK}JTtWoo*I;X*{Ugf$
zfjC*A-v>bUQZoXgsXRPg@G|$@@^j4Z(UUqhp`H-zcFx5$;c;!P7vAR{e!BV4YfzO|
zq3!U%I!IZ=DO^L)b@9>*;wluWG_EvA!@d)<`ZTe1TBfN^WA{j^(o3u>Yab3~pL#1o
z5JT8US*%&d#P88a3ac!#7a*r&-x-=|Qep!Mo~;1n+Rt~-_VTy<1?+5HWYuclfTHme
z7^&1b1Ak%xek{_kIPKI;18!GN&GWAuOQushdgIeAxvyTOI{3OcpLlj{FlTVN5ME$F
z3x~Hfl0QLWw6nVfX^ZciaW_+!10`@mn(BT{FSF4iuWK^TRVimU3^=QKy8@TmkF!m^
zi>X~9X`MLN3jS{_RDis|E*vU8oIsA|fM2z(Pd>IbC%nz7DZQ4j*;s_2GXi<-COdA8
zju+1+3Vc<Vr8j(fUzsd6s&?rwK&ltP-9}gIi(9cikDf!@F|h6qu&YSHLE-U#^*k5U
zY0P3Q=j&JSFT=F31fDHi&oYr8@;ZsV0CAjRGa45NIC*bc+|yaTFqTNmCtniAXs@_&
zLz^xn5G_Nm2%&vm%t4!3BN<j`#4n#<VDZIIf-PJ!uMnHAxmJtGVWFoGJ3wA&I83sl
zoBFpWAtsL|VT-wPC?he<bkL1Bg)od(O&+7R+wTA{tqFXRPHxg=rWH}xWBa84U15*%
z6D%r}uN))0LfiM*k69axLuwA^aU7IGA8GmJBc$5x-93`<l{Rs!hVw_;mrrXLKPXLn
z@Cxsfzzh%fdBK7H^hrNPpU#8E!ikGWDqnk3|N5)_x9P`)wtJW#%k9-UvG=0GC2g;?
zJjWIrM6Nh0PMQv2wi{u259K3^7zKwI@mKyH4rt0Vo)*PT1xE`Lp+bKck6%rEcq^|M
zTZj#o?|;j<o_SjC>4`;U6&8Q-s^(gsy6&FRFusB5%Z$?`6SvNn$wdM~JEV#M4Yz6-
zeeOWu1XzM6EWW{(xD1yTpo`06d?{oLsWQ(hmUUosyDY3Flb7N6>`%DH8tovf8s0JT
z+Gg&-1E-A$5-n`O!pL&zEzZj|MrKt)Cg&YPhIh<tiVWJ&oVX#6S9k%W?=9RPH}q=|
zfXb;;e~K_W+(QED@wyVS6mEdvOfQeIQ;6n5n6i>MIzhf!C(x;!LKO}%nfE|QNpOY@
zy`2p%R*BA|+H}BS)&s*!<A2G6n?ubCL#w{$-WDwtG3YLn?T#b3e{Y*nEo&A$A|U4c
zgm^$jq7l5l#3E8xb)-3c<$aT(z<2nfV;l(Uv9dLcL^2M(`6{4ESdTo6V3%f!R;967
zrSyH80OU;OrwOE3p|{c6;(gCOR;SCT|2`6dhmaqIChe54@O|9~fgHa5&71mHh8K}q
zcpiqlR#Tyxynge<3a82YsuOVu&Ttq#SV_3avlAfG-b)D!gHRg1;5gqIi~Wn3^!@0f
z0vvZM>lIjjJ7ZcPT3F__s3h;Wg24}M8=*?*j7an4y*u4DxJtHG)(!Bi?F$?z7sxLa
z?fB`_J1jY{Q&RAYS_XjgnMr+y_xk-CHIke75IoFR_x4BwDQ&|9@NCm=HHj!x76XZE
zCzmI0V@qE6YP!l=E7T2@zCTdgcAn80_zRH`fV&t_XW-q&I2F<)5Af0ylGkPi_8WgX
zKF>RTQjEf!MHX)nvb*O%$ZdnqS5Lr7EehXaBSB27oF;gS6(5oEJPAJxDp6e3Sr%{U
zCbum=SmZKopGe}uhR9Z%DDPhS^iN&=A_tP<I5QlGVD^Ymk=fI+tgLPpT#|g{;xdI;
zo3EJEo}{X4XD5=F@YYW(j!Go))>FE}&B3NntfBOAEccr46&W7&#vKseF`ZxW<&}h3
zn<+^OFZGnb6K~T3U{a|A!Z5Zy^q@#l2E}~Ln^p+oZ2}t2OHIqX?kmCO{f8p;pHI3e
z@xN1C1yT@QGI9u(CY&4=d8Zhigmgv;P{5<UC8a=VmhK^`TDHS#qXlS0?~kbDKA9WJ
z8p4%N6JO-5e3d7es2zcnO5lI7<`)2AKGW<_>LL5(mttyC`w7lS9D3<w%$L&_13wKv
zSBcur&9MGUusvR)o~7uc&@DZ?$jDihBcghHxGX{jqiaLF2kUZi4n8};kz?bO#rwFm
z*Ni2#;FQ4Fl-c5-^sbPh=hf%=j8}3c^bIu4o$Rk@_atwg^JF)7{`JPT;H+jqsy6c+
z6lc8k7eNiGA}{=1kwtlavEDzn4ty9yel5OZVm!&%T_3_1LPyxnIp`UB7utycv=`X(
zy8}4a_kTsqx91{^6Q{8bHS#mSbb2IN7(wjUlAG*V#<sb_M{GrI7WU30$KhsSFQ8g&
zh)+$U&qh`n_f6a2+Wt)yNFzV7yO&^nbaZtsP|2EY-O~pxkkB_k;doGF*&ibsxB8@8
zA_r_sv@M%R9)C~#sYoTCLo=KROj<vr8Wme(_0c^)5o0S=-u*?yHoCy#p9wg)2Z6^k
zYF8phI{BQ^aCL}BU;=rj@PK#B3vbYZH+GDoKeARC6yXkn$E0UNxmLymhLRc*^zwVa
z1GuzG9YQv${G~#D6R5NXEXjiB?m_(BA*?YLbz#5M%9FezuGDb38D-b8qr4XsLBrFh
zu77$zeHL`rz#Y$121HcjWR8sOKC}b}j3GNHFk;bCqWxxm8j_6X4xacZxcV5(j>H_I
z5MLJ2Yxw$F<?hW!0bHSh>B~-dQBnYt@$52$NKTOz+M28#jdxkd4>f<G5Z29+sSZ@Z
z(r`052J!cNKm6iN5KA2f_&SPKEFvY1J8qLHghLPVR4BWxKC4j9oO~fvD#A_%r5)ao
zDRIfsD^kD+*vA93gQ(n3Oqr;Uvqm&gjk)z|%cJt{=qRn8PP~9gw$~8fqF;=Jtf@uq
zn3*E*_qJ7oT{cy3C<i`9QsU%)jNZ9Fpkda7Hq^dB>XHUOt<;x^;?`*Goly9$xUB`G
z!`~ZIAL37r*=9?ZNBHI`tVWw@FTj?IMy+=ulOxw~0@CDuhc{-boBe1hI96>Pep9c)
zeCUC1eJ@WuKqX%#yR5U%D4Yq8g>>K63CrPp!e4zCNpCA@dgk1i-2v6It<w>1D75`q
z6<GfKL?su^l7<+v5%pB7ZPVwQObkfXKawOw0K4BBiJ>|;Beh5aDQem(a`Qj`A=b4J
zpSN*5+}hERF*jRLY@sC?8{n`}i-1!;^%dPBhOy3xB;WvV0<?b0OXgPUMj`Gs*_)!|
zZ)o%CIBUP0Qw=`)Z3BbBt_-1rf$0<@<SkRPa?EA7NhIDF5IMp6FhMn``#RJAQKRHT
z&I&w8NM@KWCkI*SQZ|g?UsUnu<Cs1PWIx2&5JYJ3d=eGjm44@jEKOPjW<<GKFJ&WZ
z3M-n?tmx9(vT3z~(q!mA(2;mH&g^}|4iOLZ(A~ZbRpaZA@cFeUaxWKMQbaVKSC?i~
znxH|_=fr;@KV$Xj^7w6)@6LwBfP^-kRn?$dE<YsdTU`ZM4xzZ0#ym;03)`2O@<4y$
zF0-*Ny8B4dLESlXBk*UG{+gf(F>A8)6BRvrQl{}}j|(iv^8vMl4>Ic1B743cI0a&)
zHQ&1Wt3dkcUayA!NK)^AtHf!LoRfM3dPKD01>nSA=5dGF<qa7<*MSq>7At7B_k*_o
zT;wSCpuwpB;aJniSMH09xii;;5kwmy2)NMm>Z>mbbFGAm@&zwbt{vDQSD(rhLqIN!
zwBv6?@B8lvy9z`|sNMk6;i*q$vEPZ>@m$eGRa`daZw|c(X~S-%3*>YwT~4mIAfQ=Z
zgAS7adWzfx%Z2o0tC#b~5i+5xP(9iX^{^)I5y-)g*p-DPPOJ)D(c2Pjv}fwnRj`VK
zcPTFvUW@w5P2tI*Fn|U$kLQ_T4QY*DMnE7}^$DjyH&wom<Ts4Vq%n4iJWNII_iU>n
z4O|j5OH*G29NIClu!wZK6O2eXHdaR%{|XSB>JE)rAK^pa4e6meC%M76b{7`d=~s(N
zUH!u(?aqmp(_?6ReuL#Ii%Fr%t=MXN_=(#Dn<v?R$A>9IjHqM8O%>WLhgS@}5d_Q9
zZi<WV<dzd5>84hLkYnkXH$s80mwuVou&+#GNAhF4lzM;V)RiP$7I>(=7uQN3z^=sj
zcvtHL_XQ<In%w8dZ@0Jk3E6c?Ux(6!nT0Puz2A>R%Z>M;mHwM5Mkh$7Y~R1cn=$;6
zADi3isMkf}4thE#OpG;^y`RVXUI}?-V3(s`Qcc&ep?WK$m04<ukfvVkv47iaZO&=l
zaX<Gfg!Eyl^wIZ=pqXhkTjoyf0`ome!w@c8yv{V~LoA(EM@&bv<>`hKWVgU_OxJV9
z`iz5!W%X8wCd}$dzAx`qZifeN6Gmf^>*g3Qz-_4>f0ZDbCfrR}OuG_x7N%y@2u;UD
z+|Dc`hJUvBd!btd(XsL%n~X}F{Z+b#JJTXwkyx9g&HPmy-^#eY!tBYt@9;1;e%!&0
zBexOiYTXNvfLG2dzMf@KvUlQMUp|<=Zx~AT3KuDl9r8`PwLHo7D&H2_m>WPbRroG~
zRR$lYUSsf_4#g^W2kY~qYgnmsmGb+B2$0|4^gez_xVXa=T{UJyp?alO$5rf155!ZE
z3$W3d@0lNe>Q_>Qh@0g`OU#c?CY8s$TKY7MpPdxFtQhhboK-pyx*dv<qB$WDO)OnG
zGawjzTQ&Cm+N{~0Nj&4c?8fM1B8sr@0{`azR9L3HE2}?jTIYaqmiHlAvb11+|E=;u
zuiXq}CtXp)#)|7;8WNwB#e#0Jf|M^29~uC^e@pr`Nv#CXjz%uzF|s73QQP`w3Ok{o
z*L4$x5l51NE{!l7Px^@L8CkwZtzp~=rf*~kQq7GmivlQMo5ws986ZR4QDDFuW0ZCC
z^_X5{^ce;1v<%P>q-!DfVKiTzgmCae_#%g|FFYrta5lkwASo`A%Itnrj^YA%I^8~K
z4b9>l;>j4%FB+;XUt+GG>%<I#tytFfMn1bQ`qU)}u?<#SPl{nubINd?%2fny8f|)+
z5rLeuDkD;U+T-A$Isj~Rr;F!*IQ@T)BdD8zqAZ0HPPiZ{))do2twKGwj6A^l1n98=
zL`^o!d%jG;B!6z!pT9E9z;=k4s!$6)8rBZ-d<xzw@_FjLUlD_=Qv#)ta=`gsbmA$f
ztOSZ@mQYesp_VlG?OqLuR*@JO)$2TODsh|XKdfL&pqXCJj%9Q^U_`gr5V@2ma{Ohi
z$v8}q^gXjdEOwJ*<uFoAQS&*ae>LPzg{XAEZ2GiBpC*>sK`Di)I9jZFI3xpGLMP=s
zs^)!yZGH=6RWSv2tRUnXr)a+dpuHQ!??ecU@Ac-tcX(Aj{7^98(#%XQdoFcQCoLH^
zRCGkqb@6F}jk#2LV2gQjQAqJEjU=@8-5`)(sgojT|2W^XNG(f;sMF-2ScB6~m$XRy
zA&2tMmwCy>(1&|F^Ugm!KDga<2QeR81+3wkQ<5o`h{WBuu-u2Ka#hc5vOA^k8%<T{
zps>L+^U2>Qb5x6{#M@$QNgblL$i#+*Z(ENeU=_pKI^b_99i<qt+|D<7*k5;_Z1BB&
z*Y;e`&au|XVxhdV@^ra!O~xs+Qnm$E)9rmG6P%i|5e@er5H=~g%_#0vYz6T6Gq+~n
zs@%PgDb-117;(v6Zbc<ADn%s%%GHyhOK-VKSw+p%=37H>xA*JAO0s+y#2@c0S9IrR
zeN)_pxM!R40z{|UiW2qfDF}&><-eY(;>CT0pH?T#ZDO_m-A9gd@6(+l9Bav+LDdZ}
zYFX-t2^o7M;{Gsz{;>A{89H+lq7NJX1b5#<a^z<`dlNzeGWfcU$nrp~us=^b4|?n}
zDH+vxUVdo<;W}*urqYaPlwF$jX^wME3@Rh>n{Ae9JT17Dhe8RhSxqH&k@y@?OjoeY
zA5TAPPLkZ?V4;(D7**3JtQu;7%_=JKHTZ}Fc4;tXj3ntSpDpp=SUcOfhkqef4JFGf
zx>?RWLB}KqSPMS^0qF;yfgZpfH$p`zGiH+lwki!?56#R?h;62YB8mu<lpJaO@DAQ{
zBpt~d_-a3WI#{uF5;->4oeE~!W&1qE2NvXEk@3tE)V(iWel=CgtScInB?_}!kxs5*
z?pzRR5`M)}pXSU+FW1IeP|ClIX|Kc(3#eIKu0Pw$lGFkODs0PiPjn`EbqAUv%Y3lh
zo>Fg$z3l0GZMt9O8?NHVTCo8(2ZhnfM*!a{K9*Sq@mI1xr(W_#<4ki3VI=Rw16Kt<
z^2DGaLPzLL;;2HlS=J5w6hpTf1~bmZanvAcjf!qIR$_K?^ky-XT_;MaJ23sxpuF%q
zSGmCN0cHT7DumFKF<Iy+0WL?>h*^tEPN>Y3rJ!H*E$hR#tys@0vH=(~sd6=?&_!nB
z_jEEi8~C--^oQnm=Vj)O#7egxg0Oq7)pYeY<ybz!0c!y_HuW>tTBMaxai?>y?X30R
zE7j;|wDw?WIYr+bH9=C7oj7oCn{Z`3DQ%8CesrTqpk%<-rP*tPRUJUdbmW-v{_RiC
z^gx$Qjj$q_4T&8!TbY3Ytn6&($!$+28Ei+wcf;d=19Wad^FG{GitQq2u47=42a(3Q
zQyA;*fUIODU6i5)Ea?;WpRK@Klw0?f{+T|juS^8fg=R!{f8a)|!d(2PVf?4mp}yP)
z3kyjN@jH*spv*|+F++PkGy=M|cRblozOA<dOPZez-IH`al~vI0o<x@GuDud^UHg_H
zw*Rksm5nKLJ;*`UAFl2n(OC$~6FRg^O=~q_B`uuTb^FoxTLeVe>>R0TPguJdj}@MP
zGp~iy?l4AsV_e8Lcbc}QHw*DS8!;tzT!s}TWL&8%GY0~JH0>9oI0S4{UfW#em$5to
z!-DS2U2gz0kMZMP;!mE33k;(#Q?-@28%w1wGa7sjZHPKtF0CI#mpF`mj47cYb+mKG
zr{NAq=50&c?-(xXlyI1M;?t7Zez`S_G$1MqgS}Ze&H`YhXN2|q(fp2=YUi`pwjDlz
zIYlHY)KvPX(fEhbg9)Wcovjvp?vvM-TUkp}S=`y&$78EUB6Xxi<C%*azO8xgy*SPN
zkkPrVsfb;mjw)SYUcTIIucO^|yE}6u&^k&uTM54FKK4*U{*aH5FwsqBbJdpEO0NTQ
zB#pj<_=u?X1-QAsq=o^C&eDW>vYW9H1GAcz<WlYT3NsiZ80s(d++Gg==UxYEPk!;k
zZjA6~qg}GUZgky@kjCJ9>VQ0Eyiq#(wKxJV)9yRqhh>~q2Y6~XRDag3?NNw`D;&zz
zA4CZ=F_pCMb90V))-gXIC=CQCiI^z`=o@SiAG!q-XXc(?WCAqQf`sDZ4UvymGX%#y
zQ_1P_dx9%Dw)C9-%=w1j;Xu4NUYf!KiT+HD^4smHk1lEYE5NAL>-QF&k^Tt_!(}uN
z+4$6iBj+VBv>RMB&%C<nXO|^X)WBiR!cuzo*^<(Q&!?4`NF)z%M{Ry8HX0F|M;!!s
zVwm{zOu;fd&XmsZVvVUtnYg1BZPZ5y%pTD+UU8pmcB<YG-uyLWA-~znIHgml4GW46
z$z7pFHXCr(*BH*a9O#+>7;ofx#j_gZ>QE=pKTx;ntca$a6Rz<}1C<P)nLIbhtDi{4
zeSVR8(PEDR;Ex;E((&C*%1*0H>$?J&qkI}eRe|M#`5wC~Sz9r7$SkyoV!EbbQ>u;a
zk$>>Crr3grw%=4*)txeX*)7au(yT7vaJt~u`=^(BV772urt5Fbu(8nPvz9@bF_-wA
z?VX1?hgN}ofK-yO?%a9<D-BRheWf_(Gywj&oc*t!KkavOk*IBh$Q^z4@wHRF{eSio
zK>HE;ECv*Ak|&SXLWGw}S#mk;sUfdEQ#EpamgNb~L`S1Fro=|+a&wqc{gmhw+#@y+
zQ!cBeL$T1lzq^+|Pj={g!Y7T)EUbmat2=MN>9}K0v7h7(dMvUM-jQ;0I%Iz@-1_QL
z6{<k|J=z>wPy88^(qR^&xU8{_HeBP4-f3FFb%{(t$MJI|Tij$F#w+<nJsWZsv!i&5
z&K)J?S=DhpL|Gx**3^rid=H)tXu6{B-lHYE`GlS3*l|xenc0KY4X|-nE@XB)Av_FW
z1ix1zazs}YiGC(NmJ%hs$@Bv>sD^LJl|^m1?l8D^4O~rO?Y5UR_D#$>58g#-D7Uge
z(<0oeOL6u)r<wv!K$aw(^qz9Ws!`JHfnw3_9qe-u&e3O*TCE|Jhv5v@-9wfHeiPwg
z%>7hq=?7>!t3W1D4lWwuaVj$(HJb2N`7Q<U>`$uVYi{ew%pC{I6Wvwi>%o~`%g2V<
zv;zp|8$h=3dh=B@PVx^`>%3$v9$RO{B#*gMiMjHilI=&$PZT*>L=`lOs(S_=<*Jx`
zJQ<TRq0u9bZtZegs7XQXDMvb!Tv#?*dz>Vij&2;=gqrVzSBq4i%8x!XHZdV)BFP*z
zt*J-@LOCe)%A3;LJHuhE)M`0F;pfHQSn{BN5sdQCVSs<)wEZJ^nbvecm`GqMEX`=C
z)*)P3P~%|SZRPA?=~uGR#eLHe?YWh&hxAP&&kXRU1A#SV+@@0?D%6QfqC3Hr2v|B5
za)j)<i*LjWg)vW455Em+*JSJ~F5pD3B|6)2b0}uwgFY`gMG0Dmx)9;m`;uukjHk>R
z7DY8Mbx!26Tvlm^XI>2X&eKM!k~vn>(O3E8u3Wx45zm|}b(dm;ViHGA$XaI>S3msc
zs$K+E^`8Bu4**J8lrpV?yED!{Nq7C%UI^eulcV90eSX}(^W?tJ;q>Q$<2;o7rP~{u
zQ0E?oE~hx?ZT7GK&^@B)3Cv7|rQPvWARV+1%AADEj906yftxH0M)~xMg94Sd)7->3
zB`ndwgzp$Nh>S{O@PyLk#_16JFySK?9%mO4e)ry*u3MyX?<%d9Iyj2Aq?A>$B0I8f
zpnWst$JEUwpCVztQVCh!vbcRK0e7^8(C)2twd|5&J`(Wtc%d^Kt%Y$Le~!u=VDU}3
zq7STN9aogtl+s{PjMUIMojOg9j$VEXH;`OL#HGPwyl2@7q_aNck${+R5A!x%*!coB
z@<4`}m1Wn0<m}a~GxD%R`9^VoxiY=>R~MmIOpcJ&YrbW^ag*s&fYqF<M7{o<>1nbL
zM9R@|ZB}yZ96Sz@QoO~3v+ZjndIZ6LyedG-%T{d9@WE7cW_53RYaYqC0f(-3W$MjZ
zb(!tX)+F<YpzzypPx+^E%r^2*COs$C1d@E?Gw|Epd%L!{4HUU<dMV%*D@47*`ywmJ
z(IN30!!*IzHnPk<qB#4Qwqeb<RvaxM7{LgojbE;Ti<N?W4OSB4XEmRfxUaZwWSlZ8
zjrg70o1Og)OvLu4?8<zYM+D!#09;!YZb$?9Z-^wxZzjcys)`FH*<l6p2qpG)f)BJe
ztlNaxzoofORS-TJ;KC%<b;kL^?FqM|=g45#rIygJ?d%6rIEm}8ZRnlv^vW*5PPn7B
zLz1uTfhvt@vq96<dd=2f+WunWXHF7`hT|n9oTW~e5LGuXb@R3Vt#1C^RR7ak|8HMH
z+2lh?u>kYib9cZ7iGXLYgO%Vs>OaV#e^zQd<0^6in>DDLh>{Ul!f2%MwZEBJ|Fkv#
z)0a93i=1!~ja!EREnsv>uCxP$QqgdWv_Ll24}SlbSLya-0^Y=v>0sPc4O<+|oh^`O
zyz;*t9RB_P`p@s=$iRk>Ltw8RvQV?m{9!%3i(-Ul(gJzZ7=6IDueW8CAN_By)@=iR
z4$X~EF<-+LMm?91r`K=)?HB&mE&ThF`mgW0|59F6fg*KYqf`??Ky288{{Us5%x{mR
z!)`2De(96!@jp_CmY=2vZGQ<IW_kVVWCL7%ixfil8<!M#<hCu_pNfEJlCJ&NEB{aL
z`uESC5dz>!793G}IBI_DnY@g08lgX{i=x7Dcb^dZ_qz*3H_eS_g6)Eb9x8U@gj60D
zBf0)-Kl(r2xc|$~s)E3IpcZ6TeOJ-{_vQLOeyHyM)4wF=1HL^qzZ3Zeps;IzG(o!s
zph|w1Vd|Rx&kbB(7#-!dv=xj>SD!ps%dU+I$%^qa^n3jOxQI^((AYt<eN%d#|GFEQ
zWTAiqp7ixq*WDE(0m18Tj9!xb!-$?+r|%aF<3Z##fHITNn%3`=z06OeHL>6mv8Y7N
z2Cv=HtVr#k<FwGL0zgxy>80M}4BQXCHw-sBDhj(V1Cxpr0|to<6^Jg5l7eZ0eEYVZ
zB{!g!tpRfL6^j+rO|l>jy*t0VHNRLshf@I2keP>ot#>bSs{Q-^i#4C%bDsptuHCZS
z#O20#m70#@5s)uKD{z`($iL`k*B|}DWp)$`qA$<)b7~TnDH_HO0NO=@9iJi8V#r67
zuxr8{V`>fs>lNj$tp9yU{I~BcAixb3CwU%J|F8R5*Mj{Ea28(!3Qwy<*D%`yWSW+w
zqleIQPWu()>_?Qqg!Xnj_;ye?0;~q^stvyRX<6Mgn{&wtcwuOuDtVvD@^Pj+V7T@r
zAh>`g5qf)pQgGCa_<v4KqIuC`DVOiic3^hiX3&CCnns6sXSIJT-6e+p!2WzM)Gc6x
zW(}3I#kmhf7DK5U!+{Q@`0KX$&7Q#;JM>cR0?@J0Hh=RR&7L}mqYD+mK^uC3GAF;7
zDVHVNF=v{a#$rs~i%IlfbS$T$suShi;ZJ@g258{2Q-i+$G^0T2evD0yc2uB}d4PY2
z(0Ax>tLUV=^WZxQR3c?kyUW_x%TJy<W}5)-N$+BvqRW9*&N~U0x5Yx|1G&Cawjm`*
zhUNdU*!U?>mzBa$Mh7UxKRK#TKp;7gnxg#EzhSgpqSreL4AirLS>PeC`ApDjrB`lC
zZ%e2bq_738cU_NmQ5rE3fdJFMJ$Jxg>mD}L2k30=$G)_$+`M4HLuGHQ0WJyIm$$NF
ztG{l6nTn$tA~4D57;TN{7@U&NC_(7Tj$@O&f~;P-53KGk2EfQYw+`seqObtv{chu(
zB)XRph?sY>^H^{gBj)UMsq1D*;U#VA1Mn&k@biKY4UhP9IS}@H2zX&FC6YXI(u|Im
zoJ(4dsdWM~MgTsE%WL`P9rhs_S9MV`pRq$whG^9Q;U@UTKpddf_a59VA3Pget7;g#
zZ+?56)i=yuAg#C9NTkSJau>9U!QkLJpn((L`WAiwy*q*S0q@?G(CT!gg}U;~5zR=3
z+ck!G7bH#B5zU(0?GHC{Hs>;e>VCm`evqH?e_O@><;;$Hx_D7r2yY#r@A;?pbWgE>
z&<J~1P=t-r;0tt_Bfh(1zCOTOaUvV^8n7~o0^FCyr7wrIA8=E&J)b(4u^T>Dkv%42
z(UzkO`Bo431QEQ5dU`KrPio~dF{L8^=9hgG;PCY=GnDrECLl}CBb)~un=owV=Q@6f
zC}@XlejJA6MxZWW_{QD!_#Lg5^pRp5>8GiM*1C=}K$Jz62YOq)zLeO_fsRH0+Cm%%
zJ>rg~w=v4R`SM0KR5j!z8HGM7E3=2WQTUGN<rCim3-lv^K{sQ5n9E(c{s9P1oBfei
ztu9o2^c?_#6p`x&_btK>;cC(#oj}(7>uLVK)p0>o9Uqac=t}(iH@kD<037b%;!IsF
zFW27DB=Ot<`0N)Kz&M+H4DftB>>&u^@7R_(zyVIt<I?6U)33mIXxJeucem#%Lkybx
z-!;X)lLEv?1$L0%f(O`-u?|>An9O^RGEJQp03u*jtr08ws;>2upT2?hRT)X}U2bNy
zx&x`pn}2RnG4?g09Ez_<qdeFJY@CvQp$q`LOo*-sI5?gy^#%U+Vki)A4TCSi?7xnF
zg#|N>n=^pldi#8A|IYJmi`Qp2Eq;Xy{8lci3J5ef`+4`zBAJy&ebOI+UcuIXcSWJk
zV&zDf%66aKZ9s2X>iQXaiG#S01_hmyU2b;^Zif@CD*^xm7drae!>Fa(mS6zxO~M%S
zgSpj5`~oahZ6n8kGtqn|tKDw1ZZsq-|C2=iN1#?`W&yMF<<}dl;*lZH+YKna@tK$V
z@mK<He=Dk_g_cHm83Eb77{MDQqM4TsZ5=8>!CCvF=oXLT+|Jm8kJ;Dg&s$V1TGyX5
zVwbTI1C`M~>Wk<SH<S(zoM|yf9utp{tE*=|lDHlWz8>sZyOshR6QKg0*tQ8<?-))i
zRNkmauM@r6M+r9%tK9;f-B&h%SCm`0LTzu|f!9pufY)bVI6Z=qYh_xoo*4>kGr{(l
z0nF-0FHn2muZ67-jp0%DG@EgP8l*~lg9q1R2lJlINdcQIt~4wZB~7B22Y?}{+~qZ}
z7A!{HKCqs2{(f0|m>ebDzxRUbT_q1);>pcxhToTc&}#*unuPWsqOnfEeUkuX9jnZo
zE865E{(!S@h7NYKm8t?NakU<Ho+|x3^+1>WBOq5a(HHNk{Z?}+GtxM?KW#U?%*d_w
zIGgD%Kzl@VpW;CNvW&-;YDTvsF*EvGeV6LXMv16ty3&B_{gvD5riMll?(bSAN~TQ4
zc6#P{x?`gQz%F^;<Tk6#;QkHr=C@DRv3qSGtx?S3`RqsdSJ>pADx~2xb#fvbe;b07
zZ;$%zzH+rZhrn7K^XL^^ZzO1Sze8z#(u81cts@1{{a2dYz7WGz>B56Df~&8!LAE!T
zfegT|eBHyK@p%#02%UfSsYYGbYnv$1oW32D9zYr_+8_rYj?ADBwEThg#F^pj)xJc^
z`kvXUl$W)C9=y0jkRMi>@5H?v#Jtk@-Qy948zjIH78??*q@S4uj1_8&XtK=w94~;I
zO%?t4uCZLk^&#m59m+KV=uO)4h0E$q-1QQKBk5(g#sH^EV!(5&7Ow`_HQu>O3y81M
zM1q?0P^$^8gR=I>lu6pMpl58xhU81pPu-g1tH329!>3@t=C=6|a3qET^;%MNAs}zx
zN@v9~<@k+e)Y*?2J#Etx7hf}SFdAatCY(jRs!GB9sXDplxQ(X}Qipebrrxm?ILA@}
z{Z-A6+MfqM{t<2{GfgN)JEE^fL9Z<>VxZA{G%j`W^T4O;1AKRBDBTraSSCkXlL_MK
zEYkzbVlg)BK_IBj(}b78^FD?*^QR+qYRX#1IGGKqaY~P0aG)2&2R8Z?p082$xa@-J
zKADkU+=8!`*6`h4hfBPRQS~18UlyVbL`+&|Ab9&FGSAdYRL$_%&vDH!dJjKjbyl5z
zm(U>@J(f3JF4nM~y~&fFe0891c1(QrV{3Ic!7ZRT<%B~j`A$o6x1@+wOMETD9VLJi
zfJuey$C)8uMb7gQfP-Tnie)x<`2ZypMIY~{ynU}7I07e8_88(##x$fRCkAwEr{+<B
z#p&z_=_sY%0Q~R`A9#dg;Mr{p0SvVH2BoVJ9dh;J*U9ATIP5(Hdd*k-{E^_^8$##i
z@2^bMF{b2m<*or==9Av9#=V)+rT%KUNVJ#~K}~or_|X#TdYX)-Vxw;sK#zMm#hvr7
zzHL)y*Y4h4Q=aSjP;M%~=TWgvC%_JUo%-fmS>jQ(*rG#m(X~jEPoq?N3SdGQ#5(ne
z85FcmQovt)m0@7!xKF!(zt_jY_b2pb8tN(%Fmp|>z$auGbq@P7U<4xP)`kPX9XRv%
zfV;xGVlDZ?A{Q5+Z`Btyf{Tdd`O(qc^Gn&=arj*ZMW0MKJ~RkEHJF|CV87vv;J_|U
zcSTy$s%xf>U~W68I~Jq5*u>{;D?wLA1+t#7LA1ReOUq}X&P$BT^2Egt#D+_@r=xVr
zjmqe;&^HYi&Gzn{g01*p0s+QiZjYA}<*A`SlbKL4fHsAZG#>CnsM=K_+i=D6S((^=
zkS4dI-ttkRPlOxbRvg-#1jjn2M|n}P`Xg8uVWMlsNd7wOb*w=4NZTnG!G=_tJpwxP
zPE(f{G>mEoEJzGm^3~WIR}KQUr4qiaq${sjbG2$90)4vvxHg2AJ_><T$%Gu;wG8X1
zXB4Q7hf*emH{>Bici}nP?LW|D?wLl_i?$I5iflw}rPdtOKfiqUqZQ>6XL${1a@kOW
zi0N-uX;i4BmT8&RP#o;|HL9-S0}U39klow?;c?E|p4LR!q}%A_*FfaLg3ao2n}{#N
z4HXT?*?l+(lAlD1Xuwq?Ru+K9;_>Dq^|Cybj5YFm^25den24Jhz2Fv6{za@OLf<7s
zWLQ=G6dtcvU&`+hkXHquK3;ANJp?qEzMdZ7$6&^^jfNiWum=Mr$vGtQ!1Ek1ox3bd
z+8-v0*}e)aVk9y3iVs+y3|$NK{8_=Y`M}?&FbhrP!WhEo2rNZtrl$<hxWt1mR<WS-
z*z$mHv)=0#*ayqEkBsc!8(`}vV+bf`cmh6f5h#Ql^P{*&jKjc7%7-y$kwnQuQXDfJ
zv*Lh`@X(6N2rs+@4F{LmK<!T;@885Rjv;GGA*S#qiCDERT#pCEVq=l4;+=QD#&5ZW
zmE<1J9h%^dN`Gnmz9lo&X&TH<1oHSHgz@emRb^?zmjO8u^{ZXaEP4^E=V8{#I5gOu
zAcZbba<2lkTSBVplManI0O62L4RDs{S@mO#aO&JK%IYM(UBIRaxkfbrBD$dQFbnKU
z0?$>^B^PTUHMO)3JyDOHG|{mn!DGS!L%fD`)A$nUv{Dkr%0~gL%nS#EE&|Kp_UY)^
zXP4u_L52AtRICQ>p%IUSp+J{-SDzGEMZ`FA#ni*kIEbyW$;n?>qnpIxgUZe%!Yu&K
zZEsi-#jOYd;sg8s$=LHaBJ83ENA$j(8gn%fD~?O;2;Lab1u(c79N}b@gfUuecKZ&V
zL;R@sHTBXX7gt!eoO8?<ZH9n=QfZV*bRy$e)BD~s#x|M`h;y#7GCc$6$i`G9cd5)&
zNQix|d5!7l#=ax`5e+Kzz9l??nsia=8U?ZSS<RT}f|JaSBuuHIN^0u1;pH%f3CJTj
z?>$|{m*r)vGaW#3tH${%g$e?gL5&QkX!ywnQYsq3t|!f=&x+?J$~G?6Wtg-IfPqK#
zm<azyn@!8MEMAHDFjKW=DtyaH7r8Mu*PkkS@QE*|em4qXzN#8LbdI($4)l2B&xmwI
z5_2GEVpDR^5&{-{?2AU*j~vH1`@DgIb4U9*C9I%z!^rL=EmcK_-{H@@0CW2X$jdTs
zB(2G?Eo|KXs`F1j$<2ly@3S*V$x@1apniF6RPcuPXp1j+0aE+L&#n3=@TA)F+2H<S
zh^L@r7xZvwE&1usZ4gxM`x|w_Y7pYr8eA4h;Li+a%!E`F^$BIW8`GuR&!}4GbY#x)
z&&m}P-&zg!it1^ta{VvH-a0PIc6<8<L>Q0;=^R?RyJzSIk&u$^?hfe~Lb@C24oN{I
zq!9@z0SQH5C`sRI?!E8(xA*ft@B2J|`58iHuIoJ4wa&HH@jZ?W<}b5ddjL9nu4LkU
z%u^8QQ0Z!k51$>zDnXjSYMDycd&!B;B&BIpxVIZBF+MW*GXgj!820p?hh3@huz03l
z38zJ&^@kjWCAzAOgck-%=!5Gq{pJPl88jWI?cNmlNlwh0_<_vg2Tk9PtM8jF5v_*P
zd-}bD0+31%gKPFN2h&@_nwqr0-~)fe>;^l^VdQ_zZkRo#rH%Fp1nN{07r89%qz>^A
z=zct}3GN#6whj@`vuFOg>1jUulw$5t@*=aPpl~gfTnP3|bbZRUcpU9h1e#W{Z%3L&
zbQ<R>r4Pgw;lZoEd<JnQr?ilZ*k-A8OF#A4_nl7y_b&&ic{t7AQxbMD^#PfogP4SS
z4{ip1?3!QENI+}(n)gfmk`fR@VkI;%9$X5A!s2p1tovf4e=JXNCO~G4L(TFw3V}Y0
z;nkL(ge7PM9qQD5ig{`NMdv%@YXbQw_8OOLSWrlcQmG<?*2EP~zD+c|Y80w9@|orR
z*r*?eDWO$*8!zXcVww0}$)IEUsla#@0Tdq3`*x?&$%t>fWYkLw@|xw*%&4D8XNe-d
zmSs+g2PJGv9R2{XB{X$+%2FI4MKL+Z)X?rku(Z4grhNSvZVVEMz>|f>;`NB?(a>u~
zHv53ZOLac-qqQzvM=LA}f77XWUE2+Z{o#DEwtj_qz6ueh_wl{aiXnOr6W&S9EU9Kh
z$TKV6B4P5W7oz^?4Dosj3X%&%=v71KpIaZfA416BQNIZM7FHB*SUAEcp3Zvi2hJyz
zJjg?_97P4LEaEUJptZm5D1jun9-Jq?vh=cwEyot`fI^vyzQ6u!K^xaEqx$hZpS9z7
z#_7*<Gg^`TDeN;FG3hQ&vz@$*bJan~Z1tUBZDNmNY+^ToXV1)tv?QejTGf`JxYD-M
zM2y1Ec+(FubGW%vf6YvKtQb@S?(?p`&3|npU71ZFbu`sQ-53>iU>r3GIQpz&Q+=8g
zbOswy*!As}U$<5&Jr4K;uPyH;z9ru^VCb7l&U?OL>#v!}pEKtqmf^?)Mrt>19-#(~
z`F}iG_@424>n@bzQ}X*c?CARysdUW^Q_U{}m{--QE2^;9GDxeO-pY;{yPa<4ydre;
zz!13=5w_RKN8ik=xw5vwrt+$ILo23dhs%+aw+8}L$d4Zy;jAc|FN^evBw7Ny41$7p
z?*Sbrpf<_EKK_rsNexn%0R|7y&m(0$e7?QvK-?hW{PeZsPEUyzbkZ!|(A{<CWw7+t
z@#JW^qQkZM>WC5!enM-iw4g48pMFbPRyF*L<oy=E_Ry%v53?fN45wq=ZHbjRXE7W<
zl<<^{RjRwK&KH)wRkprbS|^`eXCaCzc8g3T=1Ad|U-Z=HJ?33_MRk%`hMY>4O|?|a
z6gW(U-7y&OCxhb+)jf7ziXxo?ER2hyScWI~3Vhvl><$<o<iiWVPGs41<=Me>BG101
z;#0Ops*8%l<HHWhxq_-CDjD9Tljs!E62gOM?`3U`#&{h9GO=(CwF4t(vR!9VN44<Z
z&i&f-q2927{Z61*&;Y5!+*VN|SVtsd)YYt(e(Y0?au`wpgLm6WZ2pG2W$&t39Oxl;
zcakwPu!<GX`gI#yK2nt}dE&6kdWtvAqdRoTld!utADyRH5N%rvKHlvyfbw>if_F)~
zDN|OKXs(T<|82*X@J4%ujVX{HUz8~5bIc9i$|?gsQkVS1KD6%`0DY6$&1b403Td1@
zcU^xivs!sn`>PgX?!(+@en~?>1G#avAHJsCX`>fZdXR5lHk<Ce+GH8;hO)^bt9qwf
zHaSb^AzAX-K-I<+b8a@z^h28I2hdA^LZhdCs#VKmjy0K;jj;oed`l1|cIYj*fJUK+
z6QvKe(ckP}+CW5^NPw36M`{*(%?Ph^ocH4I8*LGGRS6Ynx=PzK{~QjAA>Q|D6Byqy
znp?Td_YAE0OuxOr*`Y<z!$W2RTor)PMHQmv1;#I9v{XPfQpHF+YnR>N`*Zwhco{_p
z)UnxNx2(XvtFHoL>DgEwRf&zhjy_Sb=TZBwNSIfCuAc)zg$**YN3fK>zCV@F7tFMB
zSdHZGtbB<g++VPZ3(+KJeQ_<|D7A@<+qkP{kYtGeV|+gGmfdC;9?D)=W3ZU`r7R*{
z?fsgL;JiCl`UFre_>@t5z4TGiT1B($$AoFdx2M%iKY?VQegYa1R-(AlEMaG$4%KM2
zGSTmQb8^#@R$`XK?lS1E%04?Wx6E!Jrmk-tZIxhv^Zj{)|L0B>Cj+t+{3Abh3bE>6
zNotPS8dEFl0D{0EHQNf38@Zcp8v<t(>vT5RC}qV>OKR$IFbwwqYQgrsd5o0$&N>Xw
zZzs6UfuKhoudi@+%s(kOQP8p=Jn=$^16ojdN)mEPvdF0&tRK~;rvKIpP&H$Ko!X3f
zJGSGH2#LI=GLU}4@x}!wDq?@H_`Xw|UtH9fH>p6iK8n9huQ~rj-XGM&fXg1FLL)~r
zHV55*-YUkfR#BCPZ0t&z!OkkzkIKaE#qh12BS8LIZn^~5(vLYq%7O5lEu$x7v8BLD
zrHHNBQyB0Tcr-!4m07#M+w{`++5?z4EfPaXnAm@Zm*HXYbx=DOQQ}ggvFy@Vv0PaI
zs$8vJ*PJ)<8LeMXnN;FWKLW`iA!M#S_>5xXw|pm{v38=M3T@gHmPp%kaF^)>YFBlM
zHVH$WYFdFG4@t%ItOi?ALU&7#XtvYcR8s8R)oG{Cb<Q;sHuhEaZDykRMM^jh2HkF$
z9O=&8_T|sr&Nbq)992HGOwVPrTNFmxG^z5G*j{?~Gd(Svm+Md1B|ZC`k>jVhO`jn@
zKaR7wr_<NCrdh9FlFlec_O1iNN}1ulXe_@yk}``mH9H8`Q8>YPPFPH%^n1Y0!Bq2P
zYR$rvpC(EJ#Y$Y}SZaP6sfQLTdzrjS`Jb!+865VVSBk+!Vg-=;l(Me1TZ|D85ezKd
z@VhOHqwmJ~D~riUd{~#g<x}ro^vJlLXHZtj*)uch0X!G#0^vd0Mqy44bG)U-XR!5n
zH|z@xbGw-U031c{O+S_Bz}20>w({(dY^epck!g;9MP`(6w;<SjT{b|aud#y#v*dy=
zfz|hqCB?+qjIM%6GKXSX8hYmUAS=-&;H8iQwYLA=+|}$l<Jb?Od(cL%&pV({=6rA7
zmnRWa%XD0BbhIYW<uNK>&=LQ#O7UPZ%3k|?v#&MGITSne3R$JTJUDC;z}wB;NRU(q
zm-%eHNSmn(<Q*Zj6Od{v>0lk&*V!KVbny6gU+=s(Vb!BJaSc#;Gq<-1Oh6v)4BF4y
z3G!3h#Yo0ctt2qo)jiV%XDy))SF<A3G_4o=zWbhz+uPEW1(}|o6Gt^Bj|T&^IcG8t
z;<f)MtUeA#IP`I-nFe5NMmk_iq_y5$eqfb4@G&ejUGFH9ET0f&mjL+1Vde5b-%cg5
zQ)>F}23Qs$N(;d&*SD`C2lkhloFE)%7FYu-G5mi1q>E$;&z?=i&?E2hJd#K-#FBR<
z-}LMRN|WK>+}8-WmprQF%-izy_drW3hk32j7<iN&nbBGS4>Q(8c*jUUx}Ew;wD_P|
z+HopsDygq#UpsI_myhu6&e&MlDFI6UVw3=s`<i~lnYom&UnAsZ{-en)P-9@WU3T=i
z;m(rd{;7lJb|}4>%K#LkyNnykU&?b@;ngWRAxF#b%6i~mVD1s@l84hwrM#qFx5ZH5
z8&2{FcEgQ&oKR)RBt{8zHT-yM{DJ(6Q%KzQdl0=~9F8q&S}Ibf;>IIp@)18Q{4~_c
zZ2;Fw-Kx7cQ*NU0Cq!7!pQ=2uByq4&Bz7hMa7-eLRaQ|*WOQrrM4`pG7eu*XNBga-
z3@DudSf9Uj;syYCg}UOX!b69n<uG_3;{@JNa#zqG5P3|Zd@Mddt*x3YpH62cSWB-9
z(2}P~7u+mVwu@HJG!Qind*bD_pZ@?npFC}B{IMNh*k@Or{wjz9DtYWwC@?D=@e^tK
z?5+J5=PV9<Ps27w#)U6}uP01%=n0v*EQy(9iuW=gI7}S9K#7hYo!n9yWY{{nkDHLa
zWQ{c}vSJJUfCIQ6OvovT7U7zy*6(hhLxc;|a@9?nj@kj4^~{S^R0+t$HW^t}PgLOS
z3qCs;{F;gR_USBTmB)sr8U=JNrwWp^#<rhBi}srYV{3VksdmUzOF9APgh{k8k1U_P
zps}G#^0ROe9EFj1B~lHTPeHWY(`K|Uf*6FuL|(lIx(yZrh}7=9Wyx`-kmZG0m^79@
zKyG$Qn;UP2naHM`3_T`!X=X*xGk_MTgie7%h{U2IeC|<wP3_m+;wZ;7KrAgG?Q1#`
z0kcpU6qGKg_P`q_{U}#8zB`#CoSQs56VO=0J=PZQIV&{22Lu&WkFUYaAGpUz3C!$`
zs@Pg-TA#ww8^bgO$K6xre(6kSgOt0<u<ReB(o<HB$taO9(x`F{eqVxUGSURWVe1jG
z3rk}7o(|l$&BaN7MHHr?S1r5wK<CW9rA-jux>0C$YBqRE$JXWP?&KjTh)DlL#S}d(
zUl5oKVz$y&n2&T{DPNef*fx8``~JR%ZHINm>GHY)_G4Uhg%S4>CiW*k5I(k*s>0bM
z;rd$tIyJLrc+?>-*zQ=KLQ#_TdFmCvQnK4qON#XfLDXFOGHwt|Y5zqQi`vRLbOCqV
zL*0AFaQ&%qsJq&q*D$7gA#b%6rxVAwrt!K`?AtyMr9h3~@_9^R4YZ<3YGI}>OMh_K
zbsSq(gx>h)g44av`%b-V#RIQvH7CX+!3QRI`UGH*fLGsr_)H8UrP}{?fAE?y!U?po
zm+w>17O+|9se4jXS6o-x|AlufBEa4S5quQtj>D2ZF+7EbC+4<epwJ%U;1{;X2a^|0
z^k9Sr0ngDOx9gzF)9cTa(Ql@&0H#|tVva<c?r8d-iR|wQ1GSM^l4Sx|1&|N#g{A7u
zURrIwiO*ob0PL$9CrLjlc201<LoMmW_e)%Uk99{UWP5GE=gzH|%^;AFH;;t|4?G{_
z)nMMJ2kIIA^Nt#FSUKra{C1i+6^TeoZ;%vqm9V<EcTR?II~3zTMOok_aO<4N;QGTv
zd}2UCO5E)35+k3O%wgEG8QJf#y4?nNEU@O*N%~G6d&Iz!zn*+TqKcqO=~WCNrvtE{
zkL+`9U#tkUwEK(=LL=i1Fhb+cC0wgy!;9~!6n=+mss+Rn)+B5mdU<xXHtdMTwM0k~
zMyxRmajIh|p2HsOxK@TSLWf2@kK*5@*K9Fgt^SZhD$^+Y6Nwf22tym=o)XO>4^#D!
zb`Dp|zB1N&;6j*|XZ@RHC5>eXI=+D$+>UHZy32v!yasXZLa41TRTGx*bY`e%>a+h$
z&;cDfc67Z5RiGWUpM;de4A$X`5(Wx3lhxBZI3RqW?AwC>)i!hXiQAp*xV82yz~v?Z
zIyoP)C9>KxG3f9*XCKk-7Ex@|zMfI=^clIr1I+}G)oj`_V!kA}5%9zq*-gkr@Zabe
zrSWX;h9f0Gma6**-Wgh1v}ec-H2QUCD1AH8O+12@w8aj!wZYQIWB^l^E}@Td1)(2f
zu}1d2j%j)kzb||4#U#Hwo2hMx$2KK#^)dhlzmhhfm~R^ojTuA=d$#{Xq)>s)@(yh|
z(PfAWzmQf>5QBk6F}JPUntJXxh?n@D|C;gU)iawwm!CW^Jz56=3@gGpk|17RIVRZ`
z($Pp_<AiCVvt;i74j*T|O=Qvp-+Jk2cxI7$D~ZP0({A)rP{J{j&fu$Zs^x*qTfNz3
zLPu{24*JelpKmB$of*+5iuWq5#v*}Eli+gvT#8-$DWHouZxo5tr;?--GrxA`Ituu%
zBQNoqiI0e;%N*yYX01z-Q-wf&!2HD9gE!Y_!0>x?`g*kFIbDs>&qBZN*%eo^<ZM<i
z0|45h%wcJQ9LnG?Oc~1X=dAnRmh3$Obp6yef;@vA9nNKzYGd9%KIKTnZ=B^WC=V&F
zD^gP!mAa=rBJ_IGJvL0Y;&dYu*%#>nXtF(So|SOOMo#>&{UW|3_v=m}1fvU79DGbq
z00stm*~+BGDUN8jDPzKSBHWW^D&d^xQjp~jmuj~#V07-%t-wXSYxmtx1gYMF+V#~-
z2u->kxzZ8m)eQPvv>faefs7rWVcCuOz}IWsT-G4YqsPaMhL!}#jTz1#Ix2gMljxMb
z$n9lAe${C)-2^u*=xIO1s=rX*N0?M3){c14?v&bOc&ktG;3JEPUQne91>qgYu%ps@
ziXSrFBa6=Tl+`#xg6#uw6;qe&W~yg1?fC2nXQ!(iV8y2aGMEY)SwhH-H3bGWT%7%9
z{F41S+l}yR;2h3QhH_~6zZ9c}ywbzW4uxhv{bm7@$sbniG2|~3UzReim4LILA`J3V
zlkU<H+KWc9?y?Xv4R$p$=MuCatrnTm1O^0ZDuKKeO6FamY&gQ9qVs62y56N*QU17U
zR9UT(Jo#d<nMa1I_LSPmu7ir?fPFk>!Q21scG}#gSo@nAI5K%UROEVM$l}WCQ`<2Y
zGjzeM>B5lgLkU)l)&vHQvD||tZ47uE__I+{Z^Y6=L`F&yaCfU&hGcd5Kcs9^&Pemx
zFV`w|r99#o|1IuG<t5E9#sXiQIqee@e}U7zROx&^DOY!E<8<6w*7@cMS*B47ZG=ZU
zviHtiPv6JUhzem^SzF!6rw)7{jXCK%dm9uh=rjGOdLN}iYfI^A9Nr^~q}Dd*^5<i`
zkatjd`j|G3QuNYLgrP*y)S@B#i_+w`BruMrda(9##@3g^*9pZsQWlE{FlUaaj?T*^
zbQKY~(!3_Eq#>3{==S{14tYGIKHE2G-Se=K+uf6R-t*jb<_Onq0m`z%GSwNOi7>KV
z@q~R__bRB!nBj5mUhEq~rth(_l}A7e?+l7`pUl&usH`;{if>%dXFN5+!uygamNkxO
zDThwvhMhx1-sw{*Mv&Z5b&NrZgY8F;S)c%-D6chj&5}AjDy#%aRQ&S{-6H+WtE4Zb
zGQS=At)(XxDGLYtK9y@E8cYIJBc>Gnh-1Nr->O<ja!}HQn@T7Wmp}xFc}yu9-o&!x
zq?(VBx|AjgKGVRxyghG;fcPUWFgXoRz5UJiyf0-nEZ;O8)X9wE?v|^#Y#+x|jj7<T
zx#t11Fm84l9_q7EpT4J?vZZPe`SI8T=yGlg*`K88O?H-gx!T#^Yj5QJq;3V~Y?V8X
z^qW7}zS7n`ABySaBKe8`Vr+z&;Ds*3086TT@JhSGf`=nVYqq!?Y7G86SWKN3q>e><
zc=Sv}lOEUVecOnBO9orY9@Z16*|U~bhFkEOem@=Dd|@0jg$Kg9^c8Z+^N!L<%fPxb
zqno)zc>x00plJl(n&s}z{OS4qfs2|_g+c%hn^W#Thc3-S4O`%!-j(;9CY^sSKT;YV
z7!}ign2;1KS20!29qtq?&z$0+)=n4tu}wtAJ2&!O_WoOTLSf?<Hba+S|Hj^!i-`6N
z(RiGe9Lv;ubCd4RwN4=$yjnkU)tdKC^4uO!QzlA5!xbGn_deq^qV=Frt+EqOTYv%M
zRob`lD1h53RB=WtjVB7I_(ZnL413I1SNnjGaEx^vDjNz!lKZlJa@%O?L~v?e*HvzV
zR3ngD!2i9!x!LF{aP1NY!>y;uW$XyDK|t<X!!%o42BJfLB!LqT5gquQ+@#(@oOoAh
z(5%)WM4jtZw3<Fz8xba{FDOxHA+TP`an!y&kJY~VOYNf@o~>Ts3f0YSz_zEk^g4u{
zs#{+y2o&m2*QO^9oZd^peCcB&z=2~K%O>4?@;#MEA*dctXBAaD(wb7fU>?#Op2r;v
zi`Rk;wu8yBCD4YbHeN+w3&y0yNW6T4QNoXd!&pUyr6D{|+hun>T(z9E3lw=nfI|ow
zwKBcq!<-Z9fL~zpT~~`E^U$Ly%<+6{SDZEVF5TfreD<t2DrNT^AAm|)47(u}NB=zk
ztzq-fjWVU<cHX^km#szxPt_N9b+xoVaB`?lY#gl)b%B3~HjUsI(#ClafXI>DeS2_i
z41}#zNlh(j^8D+aXaGW(Q}@T7UPl&9nq;#>jayQ#ULP18S>@zG*ON4O?XGT=C}T8H
z4$aF^;Xou40_qRqKr&GkjdVk^%?#~xP$LR@j=z{VPzUm^`?95Kafh4@6AYZwmt-s}
z$0{ws`ru|aUrz&cA_Y@0;0v|9Nl1ztLuJxTF0A1%9Om%2h(0VjJ{9)&kiiXwp{3w6
zyXA(KgJ*CY3Dv%E2-ziyU;MQN4{Dvh<(B+Xugg8Xrz_=sP$Wbm+8X!#%!D=H1aKth
z4sPNlix~ycx1Wu?fayP_8$HiYuLnOEt#v7*w*T+8AVQ@okm5NXlrF&$W`Ds>wcQ~v
z!JH2%kmk~`O4;s`Mv2-tQmzwEKUgHUm^7@<$VH0jO&qI>D>0DeZJjRDUr&tPQj--c
zw|$+`JkJmpO!4%<(bApc%i_4)!%N3Ec*b^Yqz)Qv2^O6Vb;*|DG=lO>+Hc$e(T_ZI
zO(4}msq9lj`JGGWR#aWuJCdeg&0`QssIi5>7{l#Qk7td+>Vu?J$@i-q%jqqHZBQue
zRnwWpneM{+$Yr5ug>A7wi-ceuC!jev4(mk?7YkBPlVY&a(?UvA?IA)P;Pn6AbMtCF
zWHGq)yD<CKF~Bm^MvPOJOD`AxGf&Nqj51H~iu)G<S)#!o$aC{tWF<~g36ZG*OnS+P
z#h0|%X52=AQ%Y&ZRxAvUJAF^KjF@B9dUkdM%4@NR-y%Zhxg$@9#rkQAdF6R-gvf4M
zPeuQBvef$As|;Jw5w4vUron?0(~p8lRtcYXPv^Kz9!*q;yE!f*2N;RYQ|{fu>pz7@
z=%@Gl5Qf>l%pv8b_4<y=H~s|8j3do_imA^_rweRckXbNYgawve(dc*DNWD#)&5n@;
zh-7%hyF3Z(kFS4}(%e_2pk-01ny^-u2Oc??@@a?5?kaGc@emB6yr)V@p{=%$O&w|e
zfP~LXcFkHS(`6++{nIib3y_;QWk1fv#$}aNiO_HDfQxBfR{0n{%pbm*S3MBdDj|iL
zHhv6Kj2voS-}}O~A_(MRg{CV}nZWE}gvB;i5d=QaIrz>((Ml>ay*-1rIVPb@BYX1o
zKB_>!Lct~bYC2m&*MbgZ5#;c_uWC?4qjKb<yyu@9U3kgQ*c(cHX!~rvJTu&3_2oYa
zy3lQ+Gr0KZZA==fNz!pgFG!JB$5Fk&pSi%L1aD}MDmUDYoX0gbOu=NECo|1b^pU96
zu5$h#!gNx{U4N!%+;Dqhwog2evu7=<ik-A~^wy<N8p&1e8`a;u%KCL2$9P5fIpY{<
zjNXN7_Gye`4P!uEBC<-J)26ZVVZ2>swiix&*9;c{YecZg#L;f0W3FJ9=j?M26k2AH
z2Lxf}Mq{p$%W-^2v4jezwt{zd>*Agm2m$(dsQX5=6AAM^W6zQ#V;=9LI;Da_&nvA~
zzk1_#e?XWyAdLA^ctZSegE*$X7?d1T@{iyn9RTodZW+8F>#sCv`TM2H6eti<@YfP_
z)l{zF(|EyP+jq`gN)9_UF4RCq$%<*@@6u<X6M61QZG?FcC{4JM6<ArGnHcld%H@bO
z=&th?@E#gh`C`g6rp2kv&+(HuVU|_S?|cX`G3oiMqDoe6y5RHR%L=U_YF+w7UeoQ%
zw1<z7F(s5PqFYq;2L^#d5j#dW>a`u=ijreu`NFtd_OBdeBuieQ8WUTCiKZh#CsD#h
z7DBZk!W&6%Lq}V+_C3T<-cR~nQrNL38}7BxYotz~W+KnwQ$SPDnO}nRYLD&cC!P;&
zwZ$ImkrqQjZ(cAg<}3}BLi|SBE`FyET1&I@Bo_^TLN89*_ha8z-S2Z^CyBY|_*2t5
z9XV}B$2v@N?-qX_sUQt=M)sz8M@pjo!NYH^7K4z#CW_%SUhF0Jo1WjVp}4rFR_e%w
zxR>393GI5U<0HM>&ptbhzgBNKv=R7d?k92zXo1#2cWAhpj6VMyuqnfyCWU6$V1OlF
z`}F=Ww%CBo-iBkX$f!u8P8Q(iJf*Nqz~q(2`ZYhmq9}vQn-mRvI8IxU=kL@X!3PC+
zJyD@#pg2wItF)+ucQP%P;aDUxDkqUO_1({6Vr=Xt%*8u}Ht_uDC;?@v%%j#K8Fd)g
zxw5TQ6Hh62408*x!^Q<O?sO~*0+{F>FgTrR{?m9-c145^m@f*jO_v(g<P)w4^M@6S
z+PVq~kBlIS5IZ{{!&k5NI~4_@+(ZFnLAuEU+n%bXC1k=VDIi#_&Z#dEEl>vkARMbv
zCM5z*X@ZoR|A|n{9lx-kb3}8gRc=H{@?)qv=xYNACKDmV=7S(r%^>wOy|j~efbl&W
zEwV}86l0N<gCRTarv$1@E_CqndCH2VQ;;1V8Dze6SHe4G8^w5P`4v0RWt_?C9N|4#
z^i%6s&JHl4in9rxQ-QTitefJegkwF(pmj*E7&)~)$E)UHdG%S50HBnSU4?_6qjJ<x
zLRK(y>Ky!L+>vA=;I!e6Fdc?b5vx5ogyl-yRD}Tz`%@iG91De|c$;_%F8<R9#YCA|
zzKA?_8MIOH*kY$c@U^0>#Afb9Xci8~`7yT8S3JuBnw=DHb+m6X?;f`=xb%5(^7^8Y
z^@mf6$|O}c0;2-x9P_WfRJd@tTjDA4YkJ!`u#kXV&{9Itl|DaEM?~6ug~=?f8%<5q
zHW$9JS<gRaDN)}HV)a-szlmTv1sW>PfG%=w-)H`hRUglN7TngAhOK_=*-jjU>^e4i
ztEg0>jbk5~a)Q+jC>RvtsTLFVY2h#MFr@m~N5jm5kdG#NP=jQn3>{qTuAU}-7;jJb
zA(P@?8yxr5?5JKbF2XUUii*f@@P>P30vI!#xWQR*j=WLH4Vx;pTCL?M!__=D`SC+X
zHpmNe?!c%X5Sl)Ce@NZ4XDy*s$Zm@`5DR5nnz*(O0h1fiFqXQR<(jTbU<7e}+`olv
z^5usHM?d5SKcDhq%iRYBroBXG#P&qy_+r)lOcbx%nA3gHNKCz7-Gr6RiW>goopS>a
z-JU+`M-;#_XpqmPZd}f{l-}$617`%>$~<@F?eYY!88^FlvQ<_3{Y)?|8q^gR90_2&
z@Xe0=@kN<Ta!sYB5^>ZilBWvd6}JVMEX1b-wj(7~X8-(@fGSeFjKfjwuTbKn-@Y4!
z>NpoUa1(EfOfUwW6=6UG3#`OrO%@oyDN9TFO*X9iHVF+xL0@ki)lgb_IJTY7HE(>?
zP(bdjxtxaUg_pN|3d)r~jynLyd0M#UjPm0OidQtUE{kQU1HxL7$P_N6Hk$Gxf{8^}
z+)NgE>rnfp2>UV~mGv|Zv^7ejb>Ugaw;~KGL6g}X)pxBx=0i75btw*Y;ullak%SNl
z1~7G7a<;)?m1BbK^eb9($MnFd<7ayg$Xe7ZH>gM>m}#wM*D+%Bi(Vm<v5JYn>1Riw
z1O*lR>*=4a`+GS*#e_NBHSp*}V-lY7pE|jUJ1u;YQzkSbxK9VkAIvQDx7IAVH?rEJ
ztZw!Jv+Q3+e#Espd%quy6%*PId5R%yn{7s+^JecX$Pp<=(dyEF#J-FD)je&<&I>kT
zROsns%e%(agpQ_cZuO@kZ9Iiw_Z+9kxeZJmxW;@MYxaxn{TkV%{?!Fyzs+pZX1}bU
zO5ygg;7?L;4NA&O6dzYitG^ElWd3Q@`ueQd;qy%5O<*!0SD5}upFl<`l|abB>fBDs
zom*y<tWtUOl19SAB#6f^|F46H!_auL7|NM6&?-5T{ygvbkGuyt%nl<IE7Bc|ZLkK6
z!$xGVQSFcqVW3tq!0drZOeId(9k(`1xIric3FGk|s=QB(G8V1t-M7e(OP?9AesVt`
zM9}i6s$aNm-<x5wBEK@ccB~&V3-@@m_rNfSG)b$ULV%w6Dk};p8y6FunL?1Z!MsdW
zn{XoC`X=9Npy1aJcsw0OWpLVw1fA_G>3n(8T8o{LCcw_vl5iZwQaiyLrrBA~{sM+d
z*Q^;lg0VrMO=8uTZt3|w==UO?flJ@DH6J*YQ=JFe*V<z{mI=Qcl>G9{J%bGY>FG3F
z<$5G^D2Q~{Rd18smVmn`(9YZ3Y$i9xyKKUZl7TdAaC-|IRd!C*ZvbgjER^gRM<&%Y
zfdtm;YI&MjFq=-35V^qY8WuWu5%+4z^FWqbhv6AtDwDfywh{>3`WOcD2tz727}g`6
zKW)C-JP#kuuEgGYBbuc$2_WLdCcME&--_OERVmtr!IdcFb5*iGhA9N(9GYbJeVG-*
zGiVhx46Zb|q%$Z4Ox@){Bd|Qsbh_ON=*Bk>yl}5=OF@g04*3liJ;cVr0U4hWz*YL=
zT7D6ycpqHRQO|v3BTyL-iBU)1VpAig<$~SP=IzRcq$y&4#Vw|mHQ<B=dCVD_=7+aB
zq}PZi8+W{WHFh@egbJ^eEeLn%?4I?-zPs=MPtJS@<9$5*u^%7Cl}tC1>52=~s~N5!
zHcoRigAfZMr4W`VFPoSzM90Kves>+&D<M0oc=Y|_xcJUq_x~&KF!9Lf+8<8n;E+j`
zK9S|L^+@ohL*n&Qn(c6MryZ_RDp4f!mbkxW50{+CYKlb~?M!$;w-Gzgpx?(@1x#Hw
zwMSn6l+y>C7w9L{4C#-9&SpOHOq{W*R5wJkfBYT6!>R7A7?Mr_WfMYCCJck|!}^Zh
z6rQd9VS5Erlaxs|R~>z)x#8QN@4Ye>JnNU05N#a4JwIBD^PP8+{IZDLF8!LkqtKB@
zwLan#pE|8Hf(|C=CXCNxzu-}@94)TJhP(T&p&C_sRPfw5;a$a3$N6i^U!F9>QilwX
zXO8Khb^%4Xgs`>**V464Yib~<;~~*OUKvHr0Q!fJuOsj9Xb!eXeauyMJBaG^@mo=1
z)OM^A#MK+G<=(%}lQ+$p059UgKU%)@Bt1`jLmMc@MD!}zg4ZyOdl^NHs{O@J{H{-b
zXHsyV!*|9JNQ%^rT4u=uHm@JTI8buQ1a<44;kU*~YwB)PGr#`?HxckGmE4GUATd(Y
zS>xJ>XhO044P`)@s^LP5Ft}(zR2V6K_tA4}00IwT4Nr-Gf&UPL9T@mo8AvhakusW`
zSdAVwqVk%f`bylAts?hsynHQj`%YL@uHU}VGO58MUrDHw;_;b9eM-j)Iq-)g#bQ~w
zr@7+D>bOe_8K)&}bvlC`Wbnd~coJd#f!3x@>}TSiM9rp;^8r3EdNk8kfb0W>0w>$?
zP!n!7CJOon$U=lc$oe1^XZN+cP6k{32VsziS>6c;9Zq^A@-HQMoAD3w6i)n#QG0YP
zJ?LL7Sc)jM`{=r~HGFBn+DRS*ZXEI|iiD)iq?!?Ye$3b)Ytp-lG2AI`*W%z5G%MH*
zRAOj`pOdCXjv#@G6q%d&l`H!Xqi$#Xkn3PlnK>W!GZ0#Otac~$PnI;L@%AFtql!zG
zHPasfW#`t{lNh1bK?`K#E3MzW`1ONKT17>Vt{l3P+*PTVwW*Ipyik}$wRCew=lsY?
z6=G>c)NNf3<ZM`Nw&xtg6hfu~W}@al&Y5hPZ6yas`aCPI#$~%8UvYnMFP<D-ohBEB
zPPR);I{=n=`roIEC(HZMOcfWaxtQag*?Djtn4a4)Ru{OB`uzU0Bu4mqwF3T!s~W<p
zJDyGlG0sUJ&9W`RqLYm(8Rc^Gx=8ECI_S1kjDWD{L^Sb2x=Emna?hT{fw38$GMe8@
z9q-xXUoP7mdrVPJ`2;ThxVZlnblG@mSl9-PbzIa=`x9Tp!y(=!XB>c+*di`?y1Cot
za#giw{+QrI^Kps*1#Rs&Mh1$wIasghUf|o$v7)5E#eg{^?gtPp*8v6D-j2kAjZXBH
zF0mm`S2km(O`-X#=C|n5o>-P4V74Na%*P{a2~-CBA9w8E%p@blkQkna0%IQ2J#Tbl
zg(Ot^A*S8{a5pjnN0i05-iy*@jy%>mCyEJh78TmpIzNNmjT1j<?mSM{QO};CMj9k&
z(1G#Lekl^unbuE0azk57`@qBwj?BX(bIsE93KyZVp|~s({5D!yvvF)0UwkiD!Tuz{
z*kEP|jPIBh`)NU5Y%f=Bl+dZ1uJGzW87cD9$8cxy3c)t1;>;E?>`SWA&`}!V(w5FQ
zYjSYZ8eKG}3c3%=kw}V^co!*b?1HZ<aft81Mknjf<qJM+KU+VAt@$+svl<_a$_iU@
zJ=ui6vs4|vOqr9&P8z~PI`vPc%6r`NGd&HvbfNC;_ydy{IJdH5F9H;5HX<&YjV`@|
zb=5*GtCQ1cWR@n4ZYjk4K#Sf%wlbHPlaB6{)P+Z^DND}z<kKyk#&WhTm;g!PGf9NP
z^ZVLKH+=)Y-(Cu43?Ri5zrS>n{{nWfhKA~mL!F9=+vcVpEQzvEx^hemVDXpg*LLwn
zU&-aw+%>rH$iHXY;5w5p2_=k~`k*pVqriO$tc3G?F=}97m(r&LaD{)P<Ph;;g+BD*
z<8S%0J{EUVB8-wJgnlf^aC;KyXq-cRmvuUGasBNwO%73eCuDTWgxZ%gSw-uKWr;Z9
z8%Y(>9W@5~AMz?T2N8@ykcE!cSj*36kxW6QJ@g;m=ozxwul5$)E<X{kc^suh*t9&M
zRaVMEV6i~_L5l%iA8<_rR)eGrmJo;Ux-7f+=_P|1h}S7<;??F|fngi6b;HgVIhOdp
zOT=rhzLFSl+Pf)=-<=Ev{3?H9rkee<x3FMqiJxY<^RR#hn3d2fs4Q(CiZ`SrXqv!-
z@WvR%;6beuf&apgCa$jrilfDlS7Ib*Jf8$q4b^_iI1TgYzA6l4Nj$TygzK8#9QOE~
zmBhT!ima@+S+U)N{02PRzc!|k%+>#Hkz(;k$!^K~*;B{AL;sg*woSt6!F_k*`H!c~
z2@j=;^?DD!i6oHbSa1;(e;sW_@ZcQ;e0=-ViOXrXg5$YbMZ6+;+Y(3Irtyu*s@<j>
z7&Z)?Bm2*DBb*W9ZhVTq(U16k-BY4JAV)hhB=nAXVCoSn?<G3>Cfx&O=blsfA8fJj
zSb&8W_JG6$z9(YzlQki<n_tOLtSUc9s~oR-oUT<?``$*r-01n6d$mJ|`G~zqAOeph
zYB>92zYeTzXnF@cxO(<es;CM|K>e?;=AihgbzdP>YMF*&?d22#thk=0P65=rt+e)Z
z)^_U<$t1%8JN+0W{hOC*6l&+6v@SmERN70ed>5|8D0rx+*L;es=NRz6EX=a3S<}^J
z$V02WALfp|zm?1?o8hunaC)U)tGpAPSg*8g)~cnaH;S1Y?fnHtETFMuwruh(`L{Un
zuYX9%Q1(OMBQB`Yexh;kPU1EJJ<CCwNO7lfvR&-aQ76PvS3@)lD?QrqHthvx?#~GJ
zl=^au0{yH+MU&?x;&Bi^l~_FP)-~75cKlzhAM#TOy@4W+{u?tz%X<K>_=UL74N#J_
zL6djHG(6Lz9gI-X++30+E@R+<Qx3*5Mqd@2KH6OE$~=RsW2y>{-?~P!zh`Lfc}(8K
zyEaX|D>D+4gqICR<!}<RK~?m#{HYQ?KJsF(M~QsrS_ENWtw&PDcQ=JLn!51BFPW!D
zIJzhWWw+rdn0|-CuTT(-b2R|Na6PQ-(USN377M>~aI-VoJoo?7q$Jr5wm*?8*vEP@
zo)_!kH`26Jb+r@be+5XQ8>4k|B$cQXk65VaK8>$hPhU9Zs~pc?&IKIAIDx+D_0?sx
z)d_x6JgD|py?E@7W|y&NZPY5CtENxD+|IG`3wX#{GT@FPU(dAWN8mQ6c;*3W1c{@x
z<5R3Mbx`PLKCaq4>|3dEIXp7q6cFZ=pNScBv1xp>HEf7^B!YpAJD1hNVMwtn8rBUO
zGcb}L0j%pT{&2S6(R780%i~Z2Q}IltDTK?RU*C|t^#!!<WLn$-Tp>6n=dthwvbNoZ
zv0@p>YWo~gM)g-(nDn}G|9q3F-}f_=LGtU=m@_y_W=Po&p(9lzRs}z6A$VI)laD1A
z?pz<N7f<cQOScdDWw@}u*g47O?6f^O`UnT3lbEmhSjZwNE5z~K%~l!)p;Fxx{zlos
zpQD^V>t1$fF|Ryp0Oaf+S3DPsA0LsRFy7RN-)1{zlI-c1>CjTpX7X``z8phy4cA9M
z&GY_}+?gHJFz!dutwpix_C4Ds=T{VG4OMLSA<bz7ZhYK+%>mo)^R~u@+SIGN7PSl2
zZw{qg&ZRU?qBs*4+x=8wO|k<WNEwl5|LBWVAER0}Hr0LBHd%VFEnjdxpPYsi*5$E>
zR+s)uxf;#U*oeq`?VN@RbwDjbZk77SPMSu$Q~sP@%8!QrKBwh*za<UIFr~;)tetkM
zH&`U*Q+XooPv44L>ErO95+hV`tIqx<He;*%5H326!heaD5$bML2<no*GIg6{fN9D-
z%4*q!kpWDhxd@}ZNJ)M58P&ZjfFwU03s>ZBw*m`6;#!=hmx475rn(EP;x|gJ=H<ZC
zkK6GUcg%f1iml^A;+maSpuSHKgb63BQNc{d`!Vm{=W_iVp|L7yjF<O(Q^NEyAakS4
zrt$_*PzKXn@z^<SvUmIE9W4c_r#sK9*~oGfao})b$SZ!;-Qp=vhjxMQLG2jCVPIPV
zAzLjIw$Z4uzgYlyJoM~M!vw}|CvqmBAz;tikT%}$z$ds|b_WB%!;jTE(b~BlJxK}n
zUw7G-UE|j@-JxRn3g{<gs9oO!7Fv!*;((<0ScqM56(PQcU}7R@X}i!e!d1-7qf#Bb
z>ksWUsOasvgmg?_L;l>im)FD?>k0klNL7Ya*3xS5A+JV_vaw={Elw)DV89LujqQnf
zoFe$XIGUOq2au=5r}&Q$hov(c?T(ecoX&)lG}m2{)>u6<N^KT!4_^IM;yvPB;i!;>
zU0v6l2nSeI=TjUj99AEQy_aX?DVKl0&D(Gt!rm&0=frfB_zehV1!8(gm#B5trLl!p
z@bsJ4juD0rw$boSB8cwcwQ8DolDMVO^LFr`yFDov_c*6K1ehi?D4ljL<_(D##Vx)Q
zr3h?Iuvdv(I|>4N0@o&ih&HJ#ISg6=ls1jzrIe#&A0W~d9J#Gly*Q8qZ<765ingm+
zpgLr`pw5oHlegB1Dh0vO?2WwV7I!gbA9l$2N8_t5Y!;q47LS@9*;}~xnTjIhnpx~J
z@`TK*>-_-O<LFA+HYe)cai3Vs0#)NAp6!MsNz&ak+DfIj$HC~_&)&OhDxa_TfcAZe
z79M~nvv5Q0M>#3Ac#&Htnu4(j&4)^A6*x*@2B*^nE!P^fsM;WH6ESJuD%vS4tgyOA
zatB;Gu7X9@a;p?vE2tZTT0r&fZ^T70yh3l_Z4E(_=O_gRF1;7ZEwj2E_Tbyqw}j={
zHNKC!KIx6Yt)0Bp{WSRfI7^zHZq)(?*OL<qB}E!^W|(%A<~(1uFA^iqpIhxz6EmZF
z61MlOIkrryN|vg0Q045)JH0`NhxR1~`;H`#nFI8H!R+Do{{5=+A>iqWWJqOQW3~e_
zN~rWpXX<P%w-G#wUp-XFO&ZEEJ2)0aQ3Gbg%m~l#jDfDhe>@T~!b0_qa+?Z}slzaM
z5{&Qcpp9e}Ob}TK{3K~8DmSl1%%JqAVGOUF8?}Lf7&=2XeYO$N%j5tGs(?@T@qcLn
z{=)ijU*jthYQXEQ;`8-&i#ePhMT(5NtjKfQ{*cbynNYv1%$pc8oXMy-usKM<061$_
z$Pz^ElxZu0fI+t?t0ncQF<h1u7=o${gCsQnMh6UvxXQyYe4L!D?aabSgR<!vXqfov
ztWGjLzyB1<Ze_;BtN(57FTbAu^L`f+xoj4J?Rb}^D7SnivKIi;q>Yp*vmhV?<EFjY
z--Q#C(DIMU72t2fk2_A^Ig``I{si!oRIdo^ki`_kV4ENq3mq3ck%%_X{>i~PpRI#d
zu5eTp8mn#rj9onjZzV}|bRvnv8Px&hYLH!O%S4DKKeE*mJpD$GM)EXP1Jw`t=dk}c
zYbB{UQ8T$kbD&uX6#<HuT^3T9`S(xR#>w_<lJvD&A!v`s<J^nZY~(21EVpaiambX`
z*F-NK!>QbrO?nf~(J^rHU}V77YMm^_w~gO<x>2_}MW>)652Mr$6~{1%9aS+d@9`CU
z4XoYSxwK%9+T182^qm)=7Qa`v-Gxc%%#lhl&w!2Gt9&XG6BdcuM3Tzp#BV+wE(`jO
zcTD>!;h8i%B7w8P2-%&%J(E9_@1C@b(2{8fuB&HdS6n9(UY=L2G5R*Wu-T_`odFvU
zM!TM}9!$rjaw{<*v4eRZBkoOhYBzKCHF|UWY}77B5O2xOXjJ<~Dkf#&Kzu9&+?E_D
zIZL%OJA0!6TEd%PmMaRH|5!|+#krs~9zpL!&YRo*>R0X?28Y8#JvGaaT(1%rrsy6e
z3G34+@eetYDJueSWU8mVvg@zw`bZ9ldg6aVrpDsU6#5@o)YBj9SILU+`<H^$JWB;9
z9*M+%f_`~@7vm_)Bn@Fc28@?FI97`O&}q8R&6NHSz;-&wE~9Tt7WEG?y0@Ko)7KHd
z4?-wP)RAiA^ZNgCp#2AP#BGje7CLR;2acd*^H|mYAZX6}10&lTgfI!SZk)2o!{FBo
z(umr0^5jiVt(^Uz?ZZTlvK(9KxI%ixT`a`1$zx0EPEZ!#^P;Jw^HdL&3<Xz6mHXRu
zB9zsadz=s_Zk_8Ln$l|w5Ah6Xlmn8{<aYKJ3?JY!USt3XBYo25XUi<lB2&3_ixL6?
zdIqBP4|0p&;<?t3&%AM)7ZIcYK=m(3Z!V3rBk_>+k}CexCZAuL09lM_MaS%$&<~k-
zn}KKxOtYQ^ZI$vu@5sB*D-Q1A7IH5T0Uar?dU5@ULMUlAv&b^`eN9^OxwDaicw`Dq
z&!yGsQ$qNH*CB~lFAFT>aElum>ugvX<i1d94hp?)R^D)=Pa=Bp*`lvcdb~tJ^V0Wm
znUl_jEyXC`&Ij&ddogsWL+=bEMFZ^)!j#Bk?Sga%Ogw5@9+KB@?M6qCmB^CS?@<oG
zLCd0<1uljS+TxU_>X?IZyu}4NnXI^@<9JBXDcQ@v5A$?1PC%Fsu<}tc-iY?sUr8(e
zC~cJEmgM=_;F)vx4mCNG+nO%afdecAgIAbQ_Z@>K)~5?}aC|}KPMdGoHIk+zbETMR
z&ZT_MN(wZC&qbCb2~))WOhFuIme1+_?5mvMy#vG^FW!SGrhW5uDh(QQiJ3Hs_aWDk
zEv7fNxn7<%OwA!gUL~f~(On7wp<m;I<=6MQW~%pJ(Bja`RsDgAJ{DyybJ<}n2$&5y
z$hFHH6}OZy@tuF4Y(A&6m)+g1wYn=4xC!lre3yNYUq3?+OMQq9K;@SCLniNy{XP<I
zr!uX&@phG@V>VfkD!N?!_bfiC1ELnT7;}5|p9hTpiO>Iyf(t_irK^c!`#R07S;wok
zza91<Ma9JZI|vWxLXoCJ0B%y^(SB%${@-6G#0@*(#mV&@4T#GD1M>?Npqtyjf90Po
z@K09wAB?sMkm^M&bNhNPaMgwoY@IWCci^d$6f5*LBYv#(A4Xb~Z0H*P5+=oT0wSC7
z+a^DRhK&AelbEUEXB}=g@;siG60ATlepI(ElmD-XW*<O}L5l!3jdAs@8zQrOz72>V
z>~Xo8N5SU6S*guwDsH2o^My<I0n=&}Z#nE}Y^mS~aLB|&jG|=;IOr__smi>Qlqkx-
zKWiE_vf?jCw_KFrD?r-LP}{a>`2DdfpmcBLu$T|}`{xp9P;|T~1%qR5GCw{7@EA$x
zhYxE1^~C(|JMzyf+Gr#~jbp#*fcJ>hf*O!ra0BG+*9ae`%LRZ%0bb=j0SwmVf7QkQ
zIRlt_Pg-oIadwiCkYRy6n~%9bxlZi{zI$RD%VT@P(}|Tce+L=T1Zk~}%$vdMy+WxT
zfC}<N48US<#@u22>rI8F&~rckXikqlCB<Egm;#S{hNF@2Kgst0+|K{L)dQaLn*b^W
z`|7KQ1}^Hf|6<E~fJL_E5ybn_7qH(R0&><bbV#Y+Uz3;!OWVJ+uP$ejBNk8md)-Jg
zqQ>3yIidpnYI}X_Y^&zXyRZMg<COSWaode<k@Hd#IEXt=DSf5!&rSII_y7CN{GVT`
z0f1RE9aJEM|7Y^@&&B`0|DqWs;vxx${wn;Z>x8H;Er7&-Z$@l|p;Lq6IokF0EmNTK
z_aez5NGOEG;!LvSXahSPRAEYq1NlDdb_b@x*SpMKlH#X-paT=<{rak}`Hs;mWm)~H
z{|}qE2m~xC;^J>#{`2Ruc0e!v56ZXh&ZcpV9Gjp^p7UzVe$Gq)VMx|bHIiTTA@!V6
z7-bjq6M(R6#WELwnDafgujPVmwCT91t^9$E`!!GnV;UxX*H|J|81iuau3qg6O|sO3
zZzmv<c~74#{tB<Q@jd3&n})z$Y2h9kN>GZ_|7OGg*KJ560^X|nT=<zKz&8EYcdj-<
zLOOv~_|RcZ!!KZGqavuJjpNU1B7s%X>6T<YT|fm0Y6e_^1r!a`Qmfnosb%<34>O{_
z#9DXW0_ZG#aN_{V6Vw<bpq8<uAH3KvAd5Ous_Fv3Y#IS#S2Up0S#ZAuut#g+H|y>b
z9Oi?MNhr;onKc#wMPL8WEg<Z31?GvDHV_g%fJ)pLMpm^-f}5~J^Zykw|5+ktc?d?a
zFiB|0`rlitDt_RgvXDCX_Yh4K!ok&<=wTgIHWcFlG0=gawP85aHpy?N=L2YLc_hRF
z83GOi<RAC&`Q?h#{e1Inqq_Q!KnVgUT!3G|Q2)b4!wJwzcB+*++MvKk+19E6NPDJ3
zP!i(>I;5#wA+9{2v$nMrO@s4<qje_6U9_v@we2lHeqy^j&<No*Mn~D?zH!a(^L3}&
zuJ-!7Df*vF<$pe9*g+gnnW}Uc3+VfInUqNa<rv@<4owPeP@aIc&EfPX0CkKBF|CRN
zh)qnE>}%@CZ2*YgO^RBQOY5w>IY3c!p)6FG!3q=I-SR3S`~J{Q&%Frxl+E`M6pGg+
zp^A;lB%2M;sS2?X=@lW8%3Gq<KYa&kF641Y`juRq*R`g9nQj07Z9g+vgrFDWIb{Lq
z-+R~07Zs7{p8#ZLglt1ju(<H<q9KGz52&Inf<g&JTM9ah@lHf}h*Te7e}ys~a$qrD
za;-l@9eCim{^tZ^<_<pDjKGZocx<%n?MtuBQ=r7zdD$nr8cL|Ydd}1*z5qvaDaNzC
z?0*Q7lwnOW172}WgUkKA0*Fwsz@jky`d{&KfL{h_$Ql0o3aCPgOXG<GycSH4KXaLV
z{Ng}hXY0xAjmc~(1dh7a+z-<s4;E<KA4p8aYui`s%>fzPq~yxVb#lnv_6hL4`EM2h
zy#S;~<PKQ3xMjPzA;0C!Zr>q@fm;aDNEVNUJOCpy(foY`ta<*IfLTsh-6ja&<(T>c
zYlfx(o?f}M=m*f~G#*7fDlmE#%lQ5W0AOqcKpvOFAM|R(eE_0g)a`NqM$b=nU{hs;
zxDN1baBl&B+v7Jt0CYHY1*kGo65>}~)7XI0$I*9r!i4xlTrLp7-w~VDDoAI$wt2D%
z40wqn@N4<E2bNo}NpKo`Kd`YU9~eF;1`l#l{1yy(E}I`{x%~&&lT?aN^kE&$1-@YM
z+vRB-6|P=SUqC8H4K#3UtY}<z?@|<~o4WBVeh#CX1Pak0){Q}s#uptb)&AADFT5r!
zN)}>G@4k_QbxZ39U3?w_;vEgbDDedF=a@Y-cs7q9@ogcVk>L#|ioHEsQ4~Lo+ndUq
zxY`!9w-~wnV#o51L;l?6dnbq&sjlESwVk~^uvb~P>W{B2DFA4OAs7IYN=mh!XDj+^
zG*$#VJEqfTr*C}#Z=_a|mQU%CSE!@~fQM@HXn$YM-hw>mze1fkrgZpR5a4A{F-gt0
z?PeHp_UuEsl%?J!jot_R>SDeFC}b`E&}+%dC5y%H)kX~fJocTX)0c^C8ZUT?$zHDS
zw@I{{Lf|yHdQ<%K;-_Cq&$kEO33kHfG>hf#wDswU8qZ$me8uAqH`G51$Kcy*HB)sO
zb-yA9+RWRrC!P@iXm|Dy`Qi{DQrSy0BU9YWHh1T%g``Q=I%iw=Z?xBvv7yi|>Fj}8
zt<Fi#iBHPb>$<+@=w%>C__cW{Qt2F`D%pKYQIesYDLYjDhk)$O?q*n-t<KU2j0;9u
zJc`%JZfA7R4h;V5d1uy47h3NLBt4P*Q?Gg?FpYznw|L*4uLBi+s$ATSC7{}X(jFVU
zHK)Rby;m3#y!0xb<9$W0^UCpT-_?xyoPM_3qYbEw#n))yOx!ozNkj43*k{dZ`_V}9
z@ZfKnHJa6==!4e-05ZVtvIs#TU9r3yXKUaWBPAR8rDeI1D`=)#lwbnfe(Rx8CX9f`
zf-4nar|$uC5Mt6jLQ(+J{-(i^{HV;2CP38UH&J735gT`si<sF}=dbP`qO;B0lSO^i
zXD39G*9fBB8Rjy33E>QGfGv8M=#7Dx>7_ma%Dh&kqTf|1qWI5@qzC;}=<|0Z9l=LN
z^HiX8izM5iqv9Uy8$_cZs@O-+GJIE8*Y$jp<4IXKdroBew7Hv<&0sioi5OQv4Vt+J
zAmVi>1-x6HlS+6Gc-l3PMP)l$$FBf$W84Tx3pQ2-ZK60pxU#lol7Pi4e(e5Hzsh&|
z2Y^YkA_@&*7}i(|Fl_E%ImK%P_;ar|a&LXU;Ig%G&3pAVV1v>-5jxc&39u}4igT5m
znFM2^TT}FydVH45(N!#@K!2V3EM(2XHiu~*Iejy0z5)1Ya-BdZsJ)|=Ejua34Pq(t
z`97IzGg1kD|0L}v0x0FwEY_PV)qG@`)3ZlyM|*?-%AF-=Jk;kV6VZ_t0EK7dQs$)y
z=8*JV*(V&9tN1Q0lY;Vh%Ezs4!=Q8E>duT8{H9e61~}zBr8<(qqb;qB;|09o{|{B?
z;ZODdzklP{^F&70afl=P5ZUus*&;F`dmP?Ma%|cA$R_h3WTzvt701ZVI-zuAL~@Km
z^m}=KZr{)M`}+@Y^Ljp?kLz(=_p6S4A08{WebxV?czU9_a2oI&{RgO&ur2%qSSI@G
z+g7e!(hsj0&Wb18!qKG|Tt0~nhOL-G05yJA{k*FF3WP!<CJb;Ha{=E}dT-9R)-a)X
zAh#Uyvh5fsYG3l&u*{sP>b$+$y_O{E6XLg|=o%@ufosdV<1A|E5&<5rt?ys}9|Cu2
zGyFZS=#*hv^P0!^);{2w1HX<>yJY%)cqH!1aFVW=!*B_>%~N{PU|&9lwg-Mx=i`iV
zg~u%U20a>Gn&W7;HMjEEVG}}_nk4M;LB<u0!pmlg`iali!WcO09u8l(KL=1rJQ;TZ
zgl2jMKT3Fb2>1;ZfmfBpF~@qMSm}AiMS|Qx?cJ{0+YbBxWGXuGd4J*a)IQF9Lj7P9
zZ1~$;Vm6+k;w9t>@Zhl)sAUU#)>%#mafkc`qo)p6nHerQ6falLM_N7ktlA7M!gct)
z2a`ys^V!bnpZP)o>W0OO8Vrz>Kn0E3u3X9MMfe%+rpYNj`ev9(PD7Q3+$5ikz2y1w
zEXM_?2kLZ7Eo!ay^*;@_C7B1le1|3UO%aLzj5bo7;8MS3KvPEg441#oG*MkVJsQr-
z9Y(wDaBku@H5YC|6zQSzW5p`?0Q14tOT13K&q~;@Ww(h-=CS+NphM$;&_T}4fCcj=
z@hpmf`DBR>7{JL~rJm&s=xya-3QM~<y9D>?v61Ig4mYN&k<Ic1===C&iZ?9;Wm{TF
zcfB%LQv36T2@jPXToDUDCFb_ntv`+k)TR#Iapt4Dbdw7Q4ypo=LEBrOBgqx#w~M7D
zg8!ba`q8o4R3v;jXc}OA$+cV&lD>Z}5X`pAzjuOB>n~mn?2Q{}Ytmm+C*gVr>z)bh
z(B)a3(&a-Q+jf)Z^MxooA47$y0fmcKX8v8d-yreu?*awnp{9qcSXvjTU5-bK=iEne
zO?E`t#I5K-)|;SDY3g7=Clb?q-}v8Am4%g*(vO)}s^Ota6CG&-tZ9Vxp}kozN3s)}
z7FA|7aEv$CZ^nfV&_;AtN^OlA=%uh3Hej#non2O`$fQW2dMi{Zk<{QqpO!@MfE4Rj
z>6EDT%g5pYtDO5dx4>7|nJA4vz(O~)SAc#{fzp|o9#W<58)FA*r~69I^2X3wT_P$p
zi%if!2;Cm{qGz4pPixql^YMFUJp1!Y+Z|>1R@V-=(TX0s{|{vj2Mzy@hK|#9LC@1K
zOc5TGp*RMhrv<t3STBqa-FE`;3^@gq*cht#57hq|4QATWCK71k(l5EcMWf7IYCofx
zVlZinwU6po^)o}6C9UB$QBHiH=I?oniz=mnIMnt9K{WJm0zj7bvE}`B+ax>0F$fSq
z^D<}}A2OHz%rSMLE~o$L)2mIRlLA~2uV3=KHfz)iStjPe9zUhwURnjl7>}cCl7=|e
zrK?ERXy%8Yy)b6YJk}AaB``eZnA#siyt13v@$`9_$^-7oG^qQ^pKtQ<H|MM|k|>40
zsZQZ-Q{(X`#uxc=>pcGVAAhuTnya%BI*?;7qxB0htaADstv6O1nDQ7Gn6lW~<>P_z
zb!LsJ(7y+XfXeT~Y4_6<pZ5I(#5HX2)1A>cU`aGEjk>I3Q#zDc{0YoS?}g~wDa~>_
z$UU22lM(8NI&QDR6ydpv*9mfAR4rVb)QpHSeO2>NLay=@P+_PRBYkb!)~1wulS)K}
z!1gfdxbFt)RD+Dg^%KUgf??Y4)l#;q41Q0(p2gL(F1Y9-w>wKQap#5IPud|6AbpcX
z*Ev4O-|*NtQSVX#ZW5!mvh+w;$q4K7Ev3x?{#Z@(Xx2O?qcf?sm$p?R;u!QzX9P(_
zNnBmn<t9*q^rH1~K*1#e5$gyt8DYN3fA0VH)1!%-4|&f?c=aqiwD{Eho`$BW{=Ous
zDkqhbm%UhXBD*4cL|VSDw6f3WExdKM=)By0>;2T7xr(4ao0mY)UKa$3!q}V0i|5^^
zwyURB0+Nq1uf{hU$lrSPhP-?Pn4Nq(RKk3FfbmF}ga4_toO5g~sS3+8;b3uR`p)y_
zXtGn-SNhY(Yzc5*QM$v?%bfJ0Imdi+b%wM+vOo|Yp1LD!j?IEC_kM{YwN#+f^SsVp
zlbq*fdy8e!v4RW2wV=T_Df*tdt;>J%+6?S2d_t%lMV)e!S*?5D{SLMc@eFR<OsK2y
zb5W*HV6YkH2=4&-aHWh|Pv!@P_BBKqmF`B0Qz$WG{l3w_?s8Ka{)!xrEGmrLjX`B-
z1lEYTE!=Cuo2T1EMy4hsNX8ISFP?VMq9_+L_it*~ZLw`{y3a6R%r>61Y_nnp-+z4q
zq6JTqOho^$xMHXYl0CVGn0%T25d5(`rNwN;qMUNvjQnjwkZfrh+zKHwXO`JxHrUZ{
zhCFBHOgO)Xp|&+6$5r=v-klA-%hG6uSI_}>lne27{BV!(c=~DW3jg(_lq;8Lb9t|4
z9R~@x%nWyz&>g$m!cYcnaQ5`XxLI+2IdvRcBlgjz-9)hmU1xRT2JDLAA7D&jy=gkW
zk_II~tg&-`SOfP}eh)w8hbAnuPf6K(g|aroy^i)x_ggcVetz8Ut)@AHF`25YUIhy#
z1+XKmlYkNL31}4?-I?_MGG=Z+<rV9}$_aZm#HwAJ2t`{5ProjA&J4~@Z!|#t{1G%y
zGhHgKRPW&R-d*|P_g9RN-<wY~P$%t-#`1*QMc#`n8E?vkP*7cPXiHn2ih#SvD(*i7
z{rPijgskTVvZAu4lSO*i1xwv;W*dq?*?l))duW$Qw+~=loo55_W-hG_XD57*zhkXN
zbis5O<yV2PYgpA<tOT22Egl`SkFwetV9)MO9PLmn2_IF7C4g8F(9{CM9h7!lJB-YR
z)uv{{p<_}<)!c3icBSp(dg%R}9~<d7exo!U@LD-_;tBq^@Zk$dw)TdNaQx%)4f3lJ
zxX*q^n%dIef3-AD?<GkwuV0l)YRc<!-6)VsEPyq4Yh?Ol5JxFb%{Y%SLgwYv@0zRC
zd@)R;;W9;c#VuC6ZRck7wG1MPFl-{)DB1)fWY+UOv6b^?KFik7<S2~gsop26jrGNf
z?-oW!q#1~M`%=Cn^aAFs=Tqp&OYN-O#A=#ESUQ*e>fSGm*Zog-<_>Qsul}pEI>PzU
zgA@ha#fHPht)B0(y&NKHaW_WLdIxej(-jd@yHWE^u^#&dv;DVTl~shRb|p0A^yK`g
z#3<h_EQa_^s>p$hr!<QWtNC7l$?jEy*HOsnb!(8ScZAW-qk$BZFprc4tj}@?G~i|r
z+BR^`boyF^oACTn{+eCfhwqikh*Cosi&9j)FQ|5jX_A!i55((NIjr&TK1<1O?b-5W
zSWC$V24G+|_OHygO1j74k+-qizkI*s6d+!>TkxeefB53^QDl(q`riV9pApob^?U#S
z@2G&8<n@u`>_~le{YQgGX&Jf0_v?SdiFdZ58GO|fqV%BY{*}#+9)pfJkYCTC5S%Xa
zJvu4wx+6i2-i)-IfA0V(6fWHv@#9g+sG3(#&++R<MEiUBhj+e^n>fr`De4@)`Bvaf
z%g5!{a(5<F-60A@bD~%Lh4ui4(7U}PAveOM9s`>VoSd5nN81)Sex^>ebof3Nl+iu(
z?&>-%s_eeUONaf`TFSxw29^p1qeD;E01+n(+uzrQ(>Gp4VPi;X%B^ieVn8>TkXCFM
zA(C%1R+UBWtMwr(cUC+tUk^K@nsy37sI`_3EHI28hBf#u$2~`!RocDof79fEOSYa(
z`Zc}BB*O_$AjwO8enH#+M)_vIw7}jA@S2y62jG2v^YYp~)P37*-TK-zj;7?NR<H3f
zV(SG=$oa(3KbGN`yOSka#;#xuRrSEEjecsdG9M&FLn4}xCyaF~LGacCofNTzsj4vR
zW2MCdMc0C`o?7FKF!>^RBdsL&1p$yc{ht#AyJ%gLX2q`C@EIY!`%ba9(o|a+6WZNH
zUF{t0r}cZ|3?oPq2(QXiY!N!^F8Nm&l*$L|eI)~gawCznb&!l4?WqMD%SQ&WiwyRv
zj>fZ)wB8uaQU=>GDd!cM>@ClnyxQY-r!N2(u1W7^$T&^p`ftl-KB?~5pF?EodXY;;
z7~u!hPl;KauLq*HXi-WU6Lc1T`mML!Ew~LPih-jyg_(4f%o=0+Kidu9meN)iZF<@{
z2EtM_T+#NTBR+Gp?fBARPk22&zu2vwxvNo8YlTJNA`-+eWYMj=(y|L<t3Up@d8=a?
znfa$K6y@Q<(9OlbHXoJVIhi-rd~!|ee0Zytk5K$fLLI9f6Mm+t>l>gpI^v%I#Q=Xy
zOtPSh3xT<C=HumY9hPs{bpik8tH!EL2x`1_%g3KUS3q`;{r9%c{l#PKfRE>tET2zL
z?{S1rn*!|I@dK8_C(%5{vwlqV<QWh*0%^Up@VR}X5n;8u_)stAI=(wecab-7S7hPY
zGq?nu(YB@P_>J|xw_ojr^~h~&x_4Y9U+pR_TD8NPEJg8Al}OVW8JrOI6&pVen`AV9
zpo>Mh*y@8xkhax-&avV0mnUsmO{Z31(?=Toqa=<Qd3~t!qTFp<E8NDq&DUa_DCT9G
zR8fDA-Npr+=?yQ(V<Avlo7Mh6engbGW8#a|Ks_IMDDlY_TQ4-p=I@Z4DqOu_r%KH^
z*>vezOBT|Uz}Wz!IX<sC!Rd;byP}`W7}(aM!?AbpfR2qUZ1X8G?0zE*tw$D#G6mb0
z_`Ed^KsfAESqn^v+$BOKv#=E0coGZVS(<Gvv%kW`V&u(HY=aNyyd>^zBl2eY5A{zX
zvbzK3$|uf6aOA%StSX;nc=8>0ytwxl&3W%b3ZF*730%?l7RyDeuRiuB%fhW?Oil06
z$~3lIFN|_2u)I~N2_&};<!mj_)~Z^n#CF4%FD2lpaB4`%f_b)jgfP>aFPHT|lbSDk
zkM9CndexR4&H$w?7ZzW?2y98Z8ZM;6v{*Vd1ph#oaM(&QyscMR*3l?&kZ023Y}j^9
z|0@|b9@8l7D5?I#og6<jqUC*TKCf^p@)Z;yi_Lfb^dTsv*=3slt%Ju26{-h#uOoLR
zte&QvFOD!ie?6c%S&q%=vF0vH*l~+i_j3%n6CijfJlNz@BD&Y%Hs8ypx%iEIJ*^>+
zp>uRqiuA2+%XPE1$>ior$4p!tpuH~)2H5c2Tpr>@EYbK>4!=TRsYY`AHElf1g<HQz
zf81q#!bX0GnAWq;jMoxf4ZWBSx!3E6OeB(U?YPw#PMXYMbGA+bv~yqBUuk+JeLT$w
znj43eunu`U99)};oe6Q(U4t?>_A59;m()%)*)et1FI;zD`>KviTO29w(uJ7Fd1RPc
z12JbK_c4cw68Cjp?5%&O?vLaJs;gT$I`^vHN>VQ@{j<oJDIU4*zeMo5usvTw1j{LX
z>jF}diYFvYBX{UP^&ehrjHVea>DoMxZg5tcNh7<JzNJk21L3Ry7|b@>#vdJ;Eo2--
z5nX{M2K0Ff^0ni{DcyEpMhZWpPfbN^cuMc1rKtXGmS5;tT)_BZ2RGha?W_e+9syiy
zTPjQL+XEbDa4X$FZ_?~x&Qs|OwRqmbiXh>GjP7Xh$Qr6e>C;kW&%UnV+oun<LC0h(
z^J!NCjn!=K<R<(AI#@KU0e*oGo#VF9!}!44B{6?W`CQL@T|8_&YPmk{M&RV({1gF-
z+c$&R&e<YB+pweLKiP{%=G=?USI?r@o>+$=c*~wR;zem&CB+wH&m_Wbz@=ibc-Wg~
z65A}u6j^@J&9kI!_KttH1@u_X-=%QMm=hIsw>i!q(c&HuQ^U~=Mn^JhnnT7diK?-y
zv$qcw0(^<O?Kz8viI}vpv2EASaSqQI#_ai{ElMRT>j_%R9YyK!t6BU9T%oxgsVhbb
zRpIE>O~%%<P>*tp)|!$smT9*{mSx65Wn|`lnzC~NFRdHslyBQITdr`T?QbAx=`x^&
zEZy|~?-KEAq`EyBgdCOvMF&a|8EJuilISi9xeni)8u`U}D1|f1<T{7;I(7n0Pfv`L
z-Ahbda^iV|t7kb1!c>c7Q1oS#L^4Vk{aX$TX^@m&pY>%~&D*THL6=(-PN5!G1`BsS
z%JQK*Td%e-I3^3hS{5S3DP>TbaHzwv!_0xU@}$+M1|AUykox7u3T}?3VCfH~jp4(w
z8jR7wfj{pb?sgp2!W#Z_PZP>K$|XzW>)CFWv+Y8-E^$br<n+uajMlM2N|>XJ0h!}W
zV6l<k@<A1mkzRUEs#ytU`a~)VH)>zA|5kr_@Qu)f-CXf>47;ZR%N@g?>Y%FntFb#S
zdQ)Ky;Jcip%$OgZFN)k+ILVdc#9id!&-Eqg#b@p5pSXh5yMaUhT?T_Cef{DusR{>x
zb2`T~oQ&Ma=~SUaoiQ*IF6)rSj{>Fxi}d#AR2%}P5IT1%0JF$ck242GEPr+p@Zpia
zlW1@!@1Q=Sv5`vnG$IddIFd>Ht&8oIrhpSs5xj0(o3jOTG5RP=XfZu;w;*%mIwjeS
z9Ol)VIM;;pYW>Nbp9)V8{9(D3jDL2sDE%z)-_b=2N@C^*EDc{L`ViO@uhI!~=t?`)
zh%eQdsBBea`q;%@keiWrT{@TMez~bV2E@I1ph~HiO}EmWKd4j`tQOOKsaXcB<-SQ8
zq##!L(nY;Vzj$9U56p;jkj5UnZ5qH2i-=Um{pJgc{PT~?gVf=BgC1gv+w%NNES|A+
z3$&M7XJCM}l(Tt`zSVTEpfa7<q>T<|({Smixx{8|1^EL8mThMJ)~k(1vMo*Eu-6`;
zVL(75V9{)oMPErU#fi*WHULRv;v{{*sTdmmkO$60_lw9v2>p7=G63(~(xoa_P9D4=
z1{1e6oNG0n{EN|L=_Q$F`#t=!O#A_+AxhG9%teW=l1OLFF>zTcH8%+3D&9!P>U0Be
z_b{$`x1)fw(hFdf?{AgXT6vijSxO9SXu|9B>9FfWAnuC@{MaYU9G#^b4DB4{+fC{i
z)|N?b^ccF7m%-FK0{Ia}#EitMDUpXy-xtq{lD%Wzra0D4z@BSb{m`h=9QWih5V$7L
z)=T=F|7fQ?_6>8_664ZHNAzF7<Dc@6xprEqQ42{`wS9%G^tbDO7Zf0R8lmbx%o}l3
zgzIdR4<3or8R2zpI-?u3X3_FeTqXNrE&9dZQ-g9N*zxdVQ-bKxRRuDCI2S}m3R{U(
z@8|z1=B(yI?`%(q0H08oD}>QLCQ+`jy@V$MZM|e0t+qeFtCxn>^TpCrD_27_QS;PB
zuf%5|>$;1;PYq+K2Hoykq*DA=;qTZ&#uYX|OPsr83W+LIs3=u+JlS33i4BK7(3<j-
z<&0hN%5GAy`e*y?tH`=K3E7XS+sA5$xdb|{8>rl&?%p+}C8gbHwNr>xnTeb9nOh;!
zs+x{9%Hn<arLWSv@pHW4Hv8sXn*fvUSAm7tC75dju$hVj-pzHd)T88bwkhfU_LAYf
z<RbY3p+DbA;4~S(k)fp*dqYVdrgUcqYx49JGc-%Z${oraG6c|`s^Ai5y|*E*23;$F
zOVH5mnGTyCYitKxX3MpeVkdSJwiqb@3V1>W=PJr*)>56V$DagMuP(^c;1{f_@s3Qq
zxQ9(7k_4qNFICwok)`R0>fA9uQ+bhzJHq?{nh!o9PN{=s>0GDwJzf#6H=1l5d;{Fp
zW&*(0(@bDz|G(RF0nJhgQszq5ulzb%31-n`FRg-jwZiaG8eF3895Jp`=_h+qdcDl!
z>ZyH(X4P75Q5BJ3HfPc}QY+WYyD5A0+K@@dn0Jn*-|EwbNhxnIQy0eU3x(YLKo!T5
zVz=ZwZ>7i(O%Z2H8G@vwd{&n>v^hjzWMQ6_-X0p{^7GM6dC1|<FZQtOezis8$l6OU
zNA5jE4b>!}Q{suJ0LcUO>6}Tq$qj3^yv^QC=^MaNvfC-PYuO;5{e)vWBRfc)OU|)%
z$l8jvU=Epbx2|=lke(%%<J{Vxp`HEf^-Nrr=ak@G3o?l3P_nS-U(c@`3Hf4>Mvjin
zjXdUf!MSu2Y6;+j7<!#lcJoZ)oQg=p6g=E()I(;>0g+~ANAY;RHFlY~^Xqwdrv7aq
z2>+p(<j;4(mt8I{s;;^6Vlq>4{P{FL;9o-xCgi*K84Bs&(SgGBXpPRASWPahD-Ct8
zd)HR*ms}|J6kZ|0Pg3RGa=v*4S!Ti9)VX^on=(5Pb0f^Zu*EH}^+e`hJA)u-Qw>;2
z+_9~WyP#T(@n>@o0#uK^1p<t~E<~OG=BsL^1^UV)7m+1NxFobCqc%751(R~=zVcW5
zL9Wd+obRA9Zl=XlR^oGE%G<n2mwl?QQp!R9i1STSG^t5xjB=niAjX2rUjVO!oIRjs
zGd2ZC_00wL($P6<2<_%UD4B#J`M*U{+?s4uhn4nlAWHZ8vH$$pAL25j3}YjsUyt(r
zASklx)|td;szc)PFuwzcXbh3Pg))V1<?m$pH8aCMk|$Uhf}EL#DRFt|Z%GNp?$7<>
zw!)N6!q#o5gH9s58^oU;<Ql2@+!M}28zrCiMkZPZ7zsVdzoRo$HI7NwWPNAhM7(GC
zw3@~6$}0|v<$@>tpUnSy#4`(QB$kq$r0wgjL_;db*;2~me~KS!5Jxnx?ig&o+iYyT
zJt&MoMhB?llutJ?1HdB8aes0SAi@`0SfTAX@PLW2VhK1+Q1vVY-!Lp@11v<7F1=*(
zsYQMdjW@7bt;Gq~bzV93kti%}L#riqWY*2El(^@!HZFRFQTicLZuFEjpcb0d5vpxP
z=iTyV-{72P?uQh%Le2_HiYJbLXe2&^Ja-OfOeDF#r@j~iMOzpvyid?zzR&+tQ@!7<
zfN55jZAnw`=@>Ps{B1JbxMpV%;bUvG`gU}+WUJ&3HI6S~4~3WVEK{*~VIEelqRxvs
zpPBP`iw;A7c}LL~+c6_~hm(bwX_x7l<S`=QhnQ9QpJtWMi|C77WvSw!xm!C675A?R
zk9O7ak~BXqoiQEa*PY?Rs5y?78Gru3gfdz*pF;q3fKw_>YLF*2FK;nL_9lI*wEO9m
z+sU@^)$`vMV^fr7Qhy;w@_dc_;dx3_2FooZq{*3~1ANLJ#zok6V$-|R%m?|}%ak5)
z5?g@kMBC#$19AW<JN1gZYaC;OL)@n!jBHD$T%5d$&lvjX(hnm6ReMaaXmmftI4V@b
zZ<vLoFY{Dg8J7S=xmF35)A{)6iOqw_n^TXbM9^pGzNVs;O2iT&v-KzUxwfioM!voz
zCZfV9W8SB<+w3njx~;FpqqyF8c|1Hhyd<D>Z;dm%!zdy_LcOT0NUOH$!=sYmVkV=2
zB;csszP{h&ksq9W2wtKMW7u-%O(L!<1WF7@q!w?1dP6-=_oQ{zuP*%ms*lTN`nBJ8
zjN7dmv>C|Ct!8vMO;5WU`}DJ>1ijQJf3<hWBP3Gp(OkW`xy2PrZ?dBVM%)kyGbXnS
z(ILl9yA$PIB2%w1>oBll=wAYIo~@2^@?D+N{xdj4mou$gPbl_Rd{;GTTbOPp!$3Fu
zrvj&HIi(1B$=9f=6Li#dFEk5cntMA06RH<C4BKNc8_a^i^;w-pM(zrh!7!#-{&24*
zHrHFcc`1rVwPMpZ)-|{<$MEPBbt_3k|7Uk6=KO^v|BI>=HnX@w7}Po|N-_c;E|9K$
z<(1z*+z1a$QCmhs715Z`R82FSVCO3U+%52FYy1s^Xd&0Saokx|c9&2`5^pk;PM7m9
z-JX9&F#kYAmPYLcr^{>lD6>HU?@<*dWt2u-Pc=<h3%)rOXm*<y%=qZa*+FbOy{M90
zamh5Q^UDDb7iX8vyWBXpu<>|noE7Fi`IvJXV<oT4*-gv$@g0z3nBH!rKUA6U=wEPY
z5wZ;au`Y?z<3b##6tM1ed~9X<o+NgFG00Cdn7gB|CA9V0<Yt-At|(s={Z}*I&<=^2
zZ&MR3N39HD>bxoC^y_9UKK84lf6;v(yBozL584bOm;h7jb#DWfwaczlpmvyy)TO0!
z8({={Jopj{M;ImO5b}i@1zPf?p8T-OOf2}clW3Fq!YuPe(XAFH4>;Vb%z(8_Dx`fw
zSxsV+5HbsLXLV94e{!9tY6D?E336bax}K!RvMwSA9~`F);5RbY5x<sHz?}j`x%jmO
zOO#u(erXgoKIF(Sq}M`~Y_+e>QTT9e9zt>2MuCBdWX7>JM4k*6d6<}Uty6!~xviO9
zXr^D3(X#saRQX8ReLh4{uLOnyYQ{2b=J2Avq0%4r*tlXiA+3lZB{9h4T`_DzV$n8m
zeBt&xA+Ea~nr)`5Mo2ANRvn`N-HiQ?H89Qfq+8b2_xB-KL2z8To|GE~9sj)Hm*>|k
zhCs#<eU{($zfPufRD+?&FPc0mIn9D6tXpE=hNw2(p?wPr8ZM1&=GNQ@E-c4jC-m99
zMXwCI`FIY=q|yh+dKrdqf9yN6?qW6swCvDSHP!Dm>`jtlIQvw+$q|3u{}|g2Q@UJR
zfB(kL;kLlJk(_CpXJ(edLx1#+ksN)-pwCW9y@~d3<E+nPGASlhu{YorE=2&GhP6*D
z>DDrf+IdMQ@4bkG)b^1D7gI&yfpeEv?EZn(1h2BL*9;jG9ouHX`43r9BRg$?DJTfY
zyyibwcZc10^aE4!Gqq(h&NpkVVdifFwHvB*yp&}!mq32X?Cku_O}qTVs>C0A#E)tD
zb$pOhg_c7dUI!7L4plE`fK3#u99TkA5Z_$2Ra2np#|zPu;>`!{M>lp%akZT(A0H&b
z9$JmzcY(*2_Gas8J%thvk4j&>&lUb9pG|0Q)o78kE&LJsLpD>+qMcGAidbR1{LM1p
zs`!mWfR_}%c>iF({JT_InXmVh)<xi#y&I1UsSCE<Q`$GF*2Ca{L)}CU;y(Fd^ZS%7
z@z<+~)Amc>ThrRR|NnGqt(_Lq{Vu4IL3s+^ldAK)iZ;4+D6sPJBs`c(lb#l?&DdS`
zabqioM?c!?pT%B2P4?)v7Bpo(aG!lHy;>N{!XHZI+>CwFB%1>!dF?ep>B|v(CUQ!M
zP4r4ZAV|sitUZqWrgrbm*K)KLd?O#=AvZ*RoOZ7GAYAM&J$>Cd^dfv_m3cakbzl+~
zX(j3}Zh-W6S!b9^>Ei_TX_Ogv-AI0AfMOoH+{K_NM;`;%sLVs!0oqLQlSF!YK8{WZ
z%Ji2mg2V??6pfgfytCp&pUX}ou=tI^@}!kt^vtNA7`oiSFlDg9NG#boH9}cbE%n!a
z4?c|oTdj~zPc#U?xwXK^)&!D-47LGY%HtM@37t^-DVPM0)4e#eOf_5l^}2uV^vdbA
z(<>+w>FAr8BCch-ie|?}8uR?Drp4T&WT^h?Hoz`%tXzH-mqBAek3XlrZ&*A%B*l5T
z&TI+|-FlJ@=gF?Cc@~%TLujZ9)GK#LN{wOuJsgtni?G*cd1ooK7zZSop4~JW>0kZV
z6-knBxSG4i$A3%uc;wmMRQ*&>OT`Wpl<|^zYl?m5%rI`6{TH359fXM;8_5I*g}v=;
zINCfp2_o>|*yL~)cP^~v{n*kWFV?rJzMocyM{+Kd+s;JrpY@sYI_eVjyWhJgxwJ#C
zJUFdZCgpUBYc;e4T1>V<nX{WrX!1!eDvBL2`Uj+V#Nz(eS2We@M=`1IiP{nl2FU1r
zgqOt$E|0|h*;`oY5WKsuYF5%wCes~PjpTUC>n+M>pMgEZr12()>)PaG^MAD)Od@_-
zFuxIk@k}PsLz$^1ksVjJNFTtKbj_|&+B@$vCGS0}&k@+YUA#DERj@MTpzQ!W^okjn
zH<~_ZuuZV^yqbv0P~JYW7sMS5$B}~5{SP2-^^Unt4xENSow_Q6SeQxKiHBm(mI=f>
zN-MirU-+qhsChOCHJBC87Rb-$!9HxnMl+K9K;#BMK!F~@Hz<j36c0P~?U!NKo4o!G
zgd~#ag>1X`XjzrgdCf;3ws#NRc*wqNq69xoZ5;W~fn?QeRjym?)L;As6JUGyITT_-
z5`V{l3lUh=E8}=u#o1Vhf{nJW5h1cTSrxA1u!(_O*R&g>@(E>DO=f%2c3lQ9L}E2n
zkCP;#-y_h9B84rHKR#HyFHi_tZp*Fu)Rs(5Nck<KKOfhmIAw8($lt9M`wIxT9M=Jh
z2KV(2-nReQ$4Y5}tW^@_Nszgd&WDcIg$6Y!RS_)2I^IF+M&uP}MRcby56;GrTZ(yB
zPQ977rokYG|D7WtiX>s`>Yf7q<WC_I1u!`*`ld3)XL}f*R6t8dftnDnDV9^O(t9f}
z#q#9Wz7KUA^Fa$ikn#q7lk}eh<XVx8yYYm9@0#<o$X4eC=YB;$oN($=!O%~a_vlZt
z8<|c$`H_7jWTa0uAkn3)Qf_Eot1~B%2rWmqdH77m)mFoA6{}m;F5I31e02`dpUQTn
z>i!U8jJoNj$R|QLdUe+q`B9*9L+hIC6#dx{WUiy9%SVW-1u%w*uKC>KeJ8#0_-UYt
zZN<eXG9vvsf2uOMKa)xEY-4C?XU#7h7c%z2E{7IRv^)LrlI9>ij>Ajm-}7MOz;fe>
z1uDKTDUiJE%1*DGsIVT3XoF3*UU71#Kh~^mm?};oCHv&+3ho}g7{_l<?~~!07%ocb
z&NpT~=8%Y>W%V5A$rme5pR{=0$<MFuX`%<vS@dF_(>ba!SKC7Q>Z)^3ZY4B&s^Yw#
z)ofKG-_o9yIh~po^BZyC^8N^=M=QU<nhCco%&C~fd3T1jtSPQ;n43t2KVWr!?4xQL
z7vwi-1<~1YxdO=+oq*eDw?ANYneNo%^T059->X963aUJD&bCMLyL8WO0x~`(Cr~j(
zUX~f&y|4ZjQ|~4ETLKeo5k{h=&~cR|P3(DTsKw{PO>J~!yGhp~5DckT`S#r*v-^#>
zP3H8j?!U(5ZOCL+R#A7NvNn~PQL;=L;83rwHxx+5wA$C%V<LXE{qr<vTJYK_EHfCb
zXx(Wgi#1#|(mj+@j@II>;M@B(Bq$%4&}nGme^W(MxtFrvy5u4mkuIj#chQ#@A71aX
ze}BeLxL;GvoP1TC=>ak4e~uLp{a~>e<u##F6Dj?Pg|hoa1m4G$b-oqYC}TUeNm~vX
zc8WueLqhp}z%gp@;Vi!MyrDkSrRU=u>f>;5HytV3-^@mx9W+}0{(<Y>ovsx&(k4D9
zs&+f87dlg_wGVm^a6z4!*C)bK<~iLnf7H3(^ikL(*fd6VGHn&g#r-<hAU&fH%<N3%
zKXZVxJAL<E(!Ik^q8St)q~>9l&co{i)Lx6fUU}rxx)8JZoqx+ow~vM6h{CUTfxL;7
zb<yA4r$$YjSe^s)Sx%d@d1}cUtky35{prwxq1SSA14}%jjNXRqTV=dm805#izPjCN
z7Eqkqwd49t{OZYyWGPFO){2!5cdHrhROBDy5be0@l97icn~kUt&sr1Yru)Ld=D@=1
z<ULc~v1rqkP+dIi*Wp)@56Iy$EKQ_kfGAOukM_qc4-VD_fP07Q&vnuSCX`S>DG!u#
zYJN1QYLjJrO1-^Qx{<VU<ek=A&77B>U@5KgUqw5uS5Qk`)bJ_gh(^rIQvK;t{Yl7x
z!K9;50xMYd4Gk1Sks<OFtF{I7@levjc_|rqqp@7gL9MmdagHo5&uQ52hx%VG26n<w
zo7<?KRJwv&)bO8FoOaQ;a1GIhr<4zJn2eOEIK!V>xlp0Nbmko)sv_iJ%<bb1$%bs&
zH)yZfQDh``6kknj{2A7s#{{pV$gol5FtCyTq}@RCCo$6Fmgi5G4Q_<>PSSnKcnJFs
z4V~uFde~E(u?8U(;;cDY@tBz(>LJC_8mqyq0rjx-m_OT!7qWPYC1He;-s6Ttthql!
zhy~enu4Rdthe<p)x30)$I8V6waX%W9)`;FLjnX@Q2phn7s+x+{SU@!?|M>(g?xOXA
zG%{7JCaP@-TDGAvl{xorc=U^SFfV3f<-#L#vBA`E4x4t5r=#W9ji6ok@DSv^O1t0f
zRY$e;M2P=aOJR?Q5}=0VsZ|*zGS(HY`BJd{BFWn8%`k`ikuhb5`tlUP2Ij>Fb)htp
zySF&{n@zjJ6>R>Q3ra$j(OG8<G{$smfC~EIo)4Sr1A#DVC;rX~I9}-0Qz371;{9T+
z9VWwd>3+fs+<?=MPwg7on-FNjYu%OGWuD7Iyo3_0hSXvmGPIsP13w!=D5AfoA^5H|
zkx)8pk-4>Xwxw63!I<R2D%;HC!qL)EHT)9t)nrmzA5D7evT{{OmtZ8|p~>IHOYjuG
z%xuh2LH%4inV|GacI-j_!vdfnPW;*${VhJ=Kbh04u_C9bnpSlJDLPbGM4y(CGv0!D
zSU%hU44*=zWbm&0XkF6VFR<iT_#8-+d8>9+H1)goDh>$0{?)<1P!SM@-6a}yPf5#~
zxtz1D*CEU5e7|9G_~j$Hj&fTxtXSfO6zf%I_NL-Jk~L;Hs}BH3#{W47&DZ6L#fm{c
z+BYb{lSV5Tfy4mMsZtZuI6JM}GD-CuOo%U#_s}TjtiLe-1&CuX6Qdj|aRZtHYPm3Q
z*wbJLvFT{NRrJu#B~pk#ntrz`KI608>6)C2Q8xUi`@i`ToIqrEOsI}9j?#q)*qgnB
zgFA{{S`vITY_qFI-T2hmbKTD&Okd@U5devJ@YKdd>8oWxh>MTSXU^_q<3Vf2E)$0c
z!yi|4drA>D8ktcVw)o?au^(l7(@fG>w@;I9lNCP9&7Tg9h)j%|sVOX5@|y#_G(0(~
zfQReCK<x(8k(}Hndr#+L;Yv(eOxFWbb-#Y!evY?ZkckC)j+YizkIAr@>p9j?kEtQk
z#`JRI1j8%ZRdmoy`fSSSBysvP93-VifOx^)+6xyv(>&Gl;#y#Dl3}F)QzN-`RYs|n
z;PX>{+c8+8*#<s(1<d)50LC`&1U0ONq%Z-s&f~ocwPKT2ob1lg90+yBGIHk<(G~>b
zfzcA{`p7~X%tJbN_ze^#lXPsB6{j)Z)CT9xjt`$1YE}XkF*)9%j$D2-d{S&#>_%^M
zQY{2|xb92IP2fFWP56%oF@@Z~GVU}XNJ9jWXlbFAxBAsknb+!5gB?o+t$V^X8~0*3
zw3TV0SB}C0Z!EL&&kSxZJQ)5f71f>EEG);hZA{yW9L!%J_XwG7ERXE0Sncahp%<qe
zV{*KEV;Y&K^zWfu4;;$8M~Tv6x;yUYWF19%k#hI>ryY89wXU|mq^Q=ww^AxNZE|fi
z$$xJ(+h#b8)>|gcmxd*vT_=*CuO7i3iEK(gjB)B{qz*dfKEAyq7)ZVoJ1Z^@qPesu
zZ=7R~WC)yj<xd3Io9^?I*Q{<%FDU1~9L2w0dX^u3$_H2VQM>oTcV<ueE)C_0LJi~Q
zP2<{GWVv5$$%j1Y0hhm!<(%-uj`W6vu(4<XDez@9M;k_A^52Qfl$^9asihCDqm%NL
zqw)2<X(9X9P#GzR3zU~!h6QsHX?`PJQ3Id4K)4Ynl65D=6n$Ol4Q^&sov5L@a-)ZZ
zpVIlt;2la&!$bIQ_h(PGDs!<>%Ct|yP2W8sP)wz4M0e99R(LQE3({!ov7+cxJ`WNo
zGD+!Gr5`x`gO(Q^$llv|^EFLJaX36PD6iGMxB4SO4%OewNOmnM6ir2W$N?K6i#FE_
z;g}9$AJ89>kI}?ELkB~aUC2Q+IB^?{{&D!l^C)1cCUcN0o^EuIv7L=ImNTrm>u3<R
zp1@CKE=;a*p>tE}*@nRYxmHNvm1zANl&8dhTqq-XnFGo^yS!d1>*0ew(%{~3=(WPz
z$iai8l=xN#GM&5IJQG!>v#M{Xr0L8#2ANX)IP#>bwH{!?AU($b$S{MM1n;#WlKwc|
zXx;u)!djD%`J@XX37(5Xv$U!Zgfj2q!i8R$+f+U&oI|cTnP(`brSxwWEJ>|yD7Mfl
z#DB7Dh*3_ngYc$k+-qOWcvQiI`tX){0A>6#q`nc8R-vw<X$2S2Rv~b%X}D>&ehWKg
zIhEHm<2E`V)#rUEcI0iUemoUe{r#K_&aKI^o2E}mpH;)O`O5()l960-+M9}igK7Pl
z`Wvle3;i<{`}o%N+8sY%X76twfcm2JQ7OMUqPCJzxfjvl<PQYlRX)UO&-<TR4e{uk
zMl(`A`Bzn_q@!uN1b@|p7}h~efIg>5`Kg<)iAw}Jlv{b)L!J37VfVKa2X<N$KIKe~
zrSo1VBnli4XjiLLoMPO{b@9(^I;xdo>bmE3?thA_YmPZl`}alY1)ZcMMbp&GnrJOK
z+9h?z`KJ5C;^P0r1(?a|7e6Z&n&Ox#cZd6@ON?mV8nimZyG9LLB?uCK+aBvpJQnKE
zceW+XAOB7HfN3jR_#s!OpvskLemW6OT{iH;CB1m%)50)P9=W4N*L_W&2VzQ=G?AEU
z7E9rpDT}5PkJ}cu+NM@)i{G>Glw3aOuGyfcoM-gGbl5z_vii{3ljG7SwZ?NSB`hS-
z;5&o1Zsu$$bk+gmYmWx24*3@%Ht)8+P2}MWdG#|lJnH*mX+6RyS2wL}=ocL^%{AtE
zI5Fe*P&5g(eiCBPH8}xjAT)A3sNnM_bJVQ-gR{P3OLk0u$h1S9@;#5w@e6tKIKjRJ
zy|b7_vXJAO_3-ZNzhAnvN!}T!hEGdIk`r0&Dc|1jF-81n`6m)VvQS+3%uC;*NG^#t
zDb;#qZH#wH((jY=6yL-KUO<kPjRZ5=d@J*W#fN4=z&fiIxo-5t^^3?O6PvS7Fkoe8
z;3Kwh0@1MO>C2Bz8&iw23{R^T&sSIhsS^_#w3qwS=B=vD+fLlBy_K~o)lgnoJ);=(
zUMOjfy3U2_**Gs2kYc-x9gCSB`~CNoC!iVG2Pbz{{wQserV*<l9KkF}S^q%DhF(3-
zx(i73e2AO?Nept)ueoOKAYFxTCkziJ_K$(~N+ZFZ9MfEa99B<qC@4X7iZ1GdwJt=a
z@P~YU1jZjxH%Y&;gNDm{X*Ea4_iX7M`_kc&Z{po6oPG*2DzlhM)Lp?S!%V1)-dV6w
ztqPtGfUHa(WZj@c0*yg^j*0tm0sAVHn^_bqH%5tOIaRgu`3s~ygl1S*dh@dm<9kU7
z&C2sSU0!dM2Rn>SS4DcXL8R<Uf-ies#RxWyT_HJp&9nv~;N^9|BV-S+6-($`1;R6r
zXXft!kvKYj+IG-ItMo_=7SS~*WiPb1RE-DDKF-@=y)$C|(}VekmE1MtK~o$qhmi{u
zQ{N@PNsztk?STFE$uQr^JE^!oV^mA>I!;U1GUo{;onOJsEf9E?+P~kmM!{By#VXq5
z{;~W5J&}-G{6ej2*({FWxnSC6OwaYVbOT1U%|{vfPZ^UV8OU7d=;`Twp+~R$Gnr1!
zQn2sxPp4$Jo`qp|I(_-(tn%eDZXzqo=nFOGGtLavS<s@bU9vX0r<O~qVcxW2q3R7X
zP9$bBd<EO&n{7C9AtjGrn{E|!`B8Pf)x~UfrJ`jrmFhfRwOlu|Z;FWjC+1elNM=M6
zi3`yCeK>{w(?V1#<Q3A&kK~Ob!GYL|ZTNY<bQM0C{K_AZl|jOPPz3D%zvWks6`IJY
zU~a`XRU?1+Y>?_v)I^Mi(-mC_>8+-YG*rjDTtg)5ue3x;BnT_8nU;AoInURueYo*=
zdH*NLtt7=LM$`A?5G6OqUrzkj9BRMERk4iC-d2{9VeEYy@^R8iwKn`o-o{Jdn|%R%
zv#8T+e);($?2@Ys^8C|MLYS@!geFcXFs*lH?$TIKw?7@iz28LHOw(eOf4ApQIT^bZ
zKY7CM#vZFkC@&9sV})>Qp=K!LPJYclnM={R_{y;m5eDEMZ6YVeIb4qmTlNfFzgw!1
ztW`2_F<?tVd%K^ikaj)gVgDU11?+67Baz!WBKDSNd@Hv>>Zc)WCpjN(N>#Z|@%&m0
zb26=lKCy{x2UhOZXO+cFw1*Z8e@2vL=RGkCksM#0bt@#rX0GdV9h(H$Jin%%G)CDe
zDwF$O(-~Rm{kugfFHnEQx^4vOP^`r_x+DhdoIGYinc}Gs2fVFMI?2n)TY|unuNEhB
z#<;af46$=g%3E_a_*t8so2|})=boZ>YVe`2YQhJ}Y)-FlGakKyIi{C)ZoPd@VjY9N
z&m|9<EernNAT$u(w_~2Ht*_Qk^*_Nz%mU6nAxbj%&)$YqB1lpCifs9vH=SJICf@r0
z(TOgT?-s12V5f5BonwMA9bI{6X`ubQOYzq1l2F4V`-)VU=9Af+?IQ0J7H}t~6W}GL
z6Wcn%lYb)jN0W6nI)uu_!=P@usDX57!{Z;WCbCg{h|PqhG|43EqOOmR8s2M|v2?1;
z7dlMlbep*xRSz%rS8G!z5U)6+SV^0H=2x+w#C6W-8ky>f7r;`Dd805wr{dyJk{s3C
zuCiD*Gf=@!W!Z!X-=H(pmHyoLLIV|;POQu8u{#Z6eXJ>10NYcmTpe32{b2g*^(1_R
z-0dQ8-mf@&s8o;ho0b_KrzS9=V^$IO$MsCyynb2fUw3{wUC9^dW1ukAEG8A8M~>*;
ztq?y)>k7TJja~{negx&vMco^7*j8X~hI+ll6@N$CPil8p+E!qf%*@-gDMiEX$2gzq
zWOQ;O&R|BkNUm9X@uuW^b)0?n1kMnxci_w#=of9UXL*8#igY6GDuOq0sX)f-i^s%y
zfJSk99Z!-^p2SsN#D&yY>tx-}@a$wumv5Zz&%uam>9ZRxat0UGZ*k^e5-K5tOFwL_
zc^C=aq&!X~7vLVE(ndFjdViTSt(i22nhE~wo=76`_D^EA{QX?{Jw8nQhIr=$LfjWN
zi(HybogR95tkx;qn43eDu~p7Ssh!bo5l|%1L55`FNeYG=0{_=jZQ4yK(2oezs#vN6
zU68<ONkRa4XH;`pE-6iDh}4YG)lyy0l=Xx_?QoC&(*gHG4bXCR5nVZ3VL-qYae6t0
z3VWS$HZh&utd9r2wx{{j@n*KHx@oSYoaOH!X4)ggy1<Kf>E6=x)B&kIiSW)IZrAB_
z3IM(<CrP<`IzFG0oLo(EbXbTdtiCjFL$%d}3r&ofC09egO0SJjW&DK_>Dwpn+x$gG
zOhhP=IY*K3(kcJ<A_J@oQWDg+_}%Orf2Avuv5=9{R~R}<)zoD7m=%8(((cg7rp|qW
z(VYAGfB6B9+x(DptQ-nR4twjM=(k27pk9DVU6U+DLX$8ojO4=ew}YuN%M1fY-uk*f
zNOI37$z|`s#Z)WrYom5kiD+*AN&@*l4xjeE<T?rQdFSk+;A)a+1IK!wkR}Okh}7Q`
z$Tv3mT{0FkXpE=P44t}m#2X&keLG!p<B7TY&Vzh5%KD=3uA=IjLnFUDS9nC9EZ1F_
zPEO3h?!BVXLd~QRFYu@seH&`hl=Qp`A#vh3-f~=-K0m2t<U-oNp<eoJgMW$sSviS~
zOqV2+?6G(P>()I*x0&2@D(;y_=xeLQoVI}Q7vEjAi?G|En5jc?Y85o~^j?72ZSs>T
zxyN!ejrC-yLF%E4B;m7ng<jMQpq_&KUtl%PC7~Vr^nxPPkrg6*9?3*rmUVGqP1%u5
zb%>BD6XaLJt8qm$+s1DgkLT&d)ZTWb4T57c`<s|_Y{zv!b^BjQBcM8zNpVBD^2j18
z9Ig^Z4lU@UJ;D`S8<q%-4JP?QxmJJ`-hFnrf+fF9T;OZXi{hn&a4%afWUJY1PJ2#z
zs8ZIiZR+-1ImSMT`+G}Xd@6}A{Ymrn6>CbE^H`?1`lZ6C1ldquBm{@d!<SA~Mqbr|
zr`A8dongvg{il@gsj?v-5LC@QtU#ZZY-eKmCF)E*?Fqju7%@b#w#n5l8>U3wE;Nej
zPB|r+Dn27v7VC+BIG+}rNYI(2vZen0S=mO*k_>{CMy0o2NSIs$$t1`D|GnF}*p;2A
z*zR}=YDe9o;o~i3!!C?5ZnWU2?fFcVxyNmJucg0&WS;@EyQ`cTYF@UTtUdLD<KeAj
z6_5LuY1ys2B24FLmip$7dP*xmtb7ds*zm0{wzH^MLtl0SY5gYwUCJj7sd7omQLI#)
z`~7s1DOqN6>b)<RVodAAv8nHJx=O|~+Hp6w<2p&IM*RU`pgffQ{2%aE0Tq67!aRRu
zCLrrqJ8!#o7?tzk)t}?>axOyt=JzN6PMMgOGT>&Mw9_6$Q(;=dTRoS*70z-V+6M?H
z1T2O#x6`npm-#2n+=x0w&%B0aSA|a|nv;O)xo{mtPBnG13dZIInl5nryK;WQt#fFw
zQjmE#FLzYp(?~zkI;*X|OY``DxFUm$q#_g8$1(I$8gY;Dwa9t<C$B+G*<|>zrmhMq
zV2H?5q;8@Ru$A&)cFEUdY!W8`_CBA{OdwV}k1fhmRJiDQ{-(3g>-zS|{}#&tH`*a}
z(|b<axt6O1LSM*i=aSyBYonXeU!8N>wmeD=fkxyPe+YDg7J;xnvoQrg6>S1-K@%-N
zMJht7z0}vm^V(PBKqMuwu}U3^`1;Y10;KfNTc6n5lNB#hSErU26^%gZ_~JR=Z9ZEP
zZ=#|{%M$jfJN^My`omz~E1&Cso72O@a5MMNZMV+8Dqx?__m`>%fAhl2Vi{OCMVnIX
z3IQ~OuT<wk%d`3p?b)XKjs;aqkmQ+jvO%}#ggw^&x~|do>-?0`L?VPq&I&8zHbxD7
zly9V6Q2=>Xg{0D5(k7t?lG!OpmZ4*`4@Q6KW|WsLwX?}EWpSlL)2^*tlPl+XCX<f}
z3|?qo<+I?M<=12%AlLL-*j9BTTId1trHj$zk~p}`=eD{LI#6ZAYSc&4y7lPMjG{k)
zf1<(`yNn@7$g!}>&{t8&-`eF;&tJ_exQ80#CkMF$pbd_jU>;dd9$A#nQQ_{5npMJ=
zYD^j}^oxxwv}ob@MeE%qO1$IfBl{OdL^O4kl>>maa-FPb&JHu{T3fkfZMMS3vB54c
zJ$9V4Ug31GsXVr7EtT@f;=mjRz~z?uVz^!CH}dVc5BIFJ(N8KY=fu(pj0LM->HDJh
z(YwB!1RR$Obqg5^t=Vr@p@3J-AdZX{9Qw>)^yqTfGDwU9pIMD)RZ!r^|9mU-!-`id
zj!ScCJJ1k4+>r(@k&qVb$J^UfQ7sSfc0}CVj1T8q0FH1^3B&MWvNfzCsDO8P?-D_w
zJ@-m6_-7xLZ5AQ5-Ui`<RgLN{gEgfptfz<S784n*p>!F)KI^LWTO~s)Y)jCo@%3hi
z$k(@PREG<FCXNeSTN=dk^|`JNPn>E4EbaJ$=a5|=T{^~D&kS#}^E%~!D8S2gD((lF
z<MZ>!CAe%>o!3X9NfB8?wYDS{Oujwmf!{Z({}Q3x1lQ?Pc>+&IA{z(<uBa>d;y^pA
z3;twiHKJ6CJR3i5cAT)P1Yy0{G~^+YF!l`7`oFCD!0N7GTL*cl=1J#BC|3TvblmL|
z@qz8ltZjfmfFOY!rFZh(rJPW)Emw0`b_qm6-@WCXJi81qJjStWI%*a7uQY!+)ghEo
zH`wDh<d7H<E@+BPgyCv(C#%R*z?~%yB3sHF5NP&xu4sfs&H(TdPwPZIFJ#6>Z!o}N
zkw3Xz<oNP)s7GxHnD1ORO!@J(F!kgh!<DIx3`y)Yoo{6W0)Jnf+Zvcy<m)TXh8=>s
zCzQEXweEs#7Ec#1(Gbsp8tT|{53a|BvI+li5W3)#!cfV+SVhD0DnsC!kEF^_836jd
z9-K~Hxh>ZBlC5Bq6}jz><8CVd<ROX^IOIGGToT!h3e;57?a5odr?~z;J*_wBvVm@2
zeYt1rGNF6PWioUIndV4~jT&J^2I<{0AEuOD9f-2ZFm{en{+`x5t3{$kzN>0<$LVm6
z&CV@m;7traWi;*%oy1<%y~N<~|4SF*P$5gn=GT&H*LEDEN2LvGKgG(wqOrkQ8pg;t
z#f6Jc*vJjWh4#8g{CfR52aJIjOP$~d@CtF%9~pQ?=MD&I$wu%TT>GX<7?AzKC~dkj
z>hmm_icx`pHdg%25M^AflfvEa2k7K~-(E{03~4tjyMw4*7D#biEN;vVkT)5+LgBA6
zBe%|lbbmTMv?;#Ze@WYtqY~22PaxPD8fs>dGZkO=c&Q{{EO+8N^_IZH4N=wzgz%)q
z63V2+Ug#r7a)-51ip}%-0x|qT1No(KKS3yuyk5aQKz0k^egA_@<AG^p<aH)#mGlf%
z9II#O{6TLfJC5tb5#9+?|5IlxoQE@JX+ZeD@UDmoBulJ;nf}d-$&qJ&$z`}nWOP<Y
zWA@>9(G(*{E;d5kr3Q_5v)<&h#)W2VOCr;VYm2N&+gBf_4g2Cc&}<M!b76=*IL9G)
zi^PdllOB{%ZYs}*@c-COsXz{O@lrcqBf0wz^u7q>2_)Zzyn+OIb*Vk(c{$GuQz7qZ
z;BP=}fHvS~F=NCT60VXy6pEq5aWUt$0FFr-K`uy$XR_~G(ElUtEug9jx4rL8cW$Jm
z-GnqqcOxYtB?y}aX^|2(jdTgpp|paubc1wvcXxNZYkTfF_l)m5_q*>HpJNY)3Icnr
z^~`6^`TXbaFIM4b>cl6Z-wlpTsfe<fQyE-$>@|%P=|goF-jQJ(w3j=GMiaNjN~5R&
z?@0o1)&Tx4qL3vSiY#08fr>D?6nt1@<<o&bQ@1=}ju&=%+_3_*x!n}`F_xBDXQg9i
z3T)98G~^nUYYDI7AqT@+(QRi46-YberfJ=zKM6s!uQk{!o_ED6=F+uQtKLct3{#IX
znj>X@hkH<+C4Nlf?&@(1{)*4_!;pE%o6hMqFe>wV7)Tpv!tnEvqLqi$4hWZVS%=nX
zmNlZ%Pg6Iuv++c<o`DEyKZG2PqQ#&RXAG$_<2B-3r^+&JAuHnM2Jiw4IejdMc2n0=
z%20XUM?sEUCmZ4WTZ59(=c4k#_9wiy_s?XSiD;6RJrVB{Q1BExRW3(ML%ifDp=gTa
zy6na2lCyXz`t<g9F^nZ~oogk)sLCZYX=GT&oDgmCdzzJ!S+5fd&zTJ{l!Mj!VsQO-
zm~Y3W2Tex22Mi-=c<~$HWY#{}_W=AzO9GYd20lZCWK(X)tTOlV^-gw3X%t5Iw(8at
zhfT##QXDA;b41KfS2q9Sx^+T+mB?kfj)cUi<qt1teSZXkKJ#OHDEnmlAUqx6Kn?yu
zslTAIK5}0FFAN1d|0hR2a%*pY?i(VE{w5j+-R-lV+n;RL2@|4rT1bYDWn23jOpyY%
z*}hrSY?Lm==<i3|9i*ukFCWlZJ9=Y(Ct@ozFpBK2F{&oz_wQWGI%ZdJvirR0LM8{N
zu0@<JxkvJviTdTZ&WL2DoWjs;+yqTUT9Fy9jS*8lLYk3{zV&2#eEpbf&!vlv8y-PJ
zblhpmx}a1o59N;epbD<Wr%`z1L@0K?R@7X`qLOeRCt`|{G4B;ux>2xbl1NS}%qxly
ztabJa@tnbMZZw2u#$#Lt7*3ZftAeT55gM1|8W<kk%=%(unWWRzZ2jH38L#`j!fMl>
zfi4QPqlxTl)`_hP@cpkKt7bbW``j6yGwa{<%d1m57pYU>>%P9-CP6d%!ssZ0-8+lX
zLQSZQ35^hW`8}ghagbN=H5I^*yHpXQlUKwt<*@CxGJE}oYV7HZCEwa<%-5Q2!5vF=
zPv=?r(5eWN9j9pN=}P5>T(8YjZQxW&a=v4f`6|0MVse1{1N-q3jx}j3&wfSWqouTY
zu86wYHX}{6I^&kGWy;yJuJ`D7A=OHv%%?MQ^M3iC>%YUK8Txx7&KkNM3IV$hQxx6g
z+5mn&U_Q=Y|62qt{qdHLO!jvwT31^3oah8b5oyCgE|fL*1+Xo6K0o4tQ3)hNtwfts
z$Mm?J<kcyJ+)PAmOU~guW`9Go!?7kaBUY=Z4#<MdS7pWMHe=Yvp6np+V->?>xw&rv
zMvIV@x;vP09W|-Z&@wZFfa3%;DL7IlBlwQ)-In{>IzajU{q#1+*JI3Ih@0n+5`|=k
z1k<xc#l^r1xp}F`;|Nc2;BBZ{Dag;zZ-Xe%(^zjLi!&Dzx)ny$wev^=z2Z+0&AWCA
z>&>|^*StUO?wy}SQa|-Qh7XUe`qp}n{&%|k@0-oPa_Il?kBI^3K#W+>r@fG1Z`IHz
zxb`G8w>MQ}rw2`3F9BQY5G~cJAP&b6KBex~5<DH33j!@*AEROmQf6kM$wM>Yn`Fpq
z!gSofef<CM9scL<4)vpVSa9E*)|z)aZKa@2zlY1JOFGD{m5<`0>`!JGKxM!P<EK<x
z?Z=}a@kIU}6cT+0Po8bS!z7mp(T$<Q`WUEm>}SQOKMqMQHzVPtX4ahBsK_A$m{&5T
z{?WaE?%V%f)Bk_N7{ieeZ3rSJYiye7CxEIuv0-gfjqV7bxAW=);FHRR-EPJVN*dhK
zabY;p4-dfS&%~<XdIQ`eJDLTXDZm;I%`PbgwlD~s<bI>{1-g2D3ct58DnB&=b4_jQ
z35HY#k2jM+@S(br{ks33*AJ+9UZZ<n)-B=ygMXG+a8BN}aqk*{kR5yw1K>YFRSo;L
z8Z1reWb#lJ8c@rj9~S$TX&~Qi{iY~KVeQV(<%UC;y$(Rw6u!J`CrCtxH}mP@{PO;-
z?E1gIp}CZZZn*IDxD{@y8(Rf;o_%WXBg%j82c%Bn6o@tjY3F6v6?vj;tTk(2jB3{7
zv9JK1UUg2d_#S#399_t5dRGg4?*IMm`VXJU=?jNXBC}JfCI9=Eihb$}Fi}R!mWqZL
zJrqCP0QU#VP4oXo8e%J6@Twvr*Bhw#Nu4nKFEG{S3OQiqjn-a?p8m}b|Nc0y0-B#_
z>k~WxG(uN9w#KU!BS6W!f0zD~5su4+vk|1iOP`0q@uP`<0irm!CIDy?TMg&{2>?@p
zpJ6#)H@%MhsRQ64^qu32Zh38)hc@`q_QeQo&PgyJ9}s`M^QngqeKT6Ji0a1T_#Xe%
z);5GWOYnDphNBSSpu}k<uY>>XQZ`q}!{0z7rU2-lZ=m^#&f)R?G6-Ol@E11T?5w6l
z_f0Ndb?~E(oCD4Si$w#8T=eknQxaqD#<p$MJn_Y%_u~=JrFn}=E4iY$6R&570Sqxh
ze<^bSu*Bu#PHH?jk?YsxZsW(ZVfmn5j;4ASF&ca^fW3nx-qecu9Qh&lLt?P)N0Flo
zp<~0bJf}1to$$iG&v2mcvS#ro&4kVd04PQecXycXYy}u<L<YArlb`d%&VRg{oW0}n
zxY%=8@<sa$CoEC|u|QfL2CU8?O)cGLto28&(oL&)s?^9k69u{Guz%sNk{9m}T#c&c
zU9iw79V9lBGycQF^#6Po1D22_ic$V~selBb2tC#7W_vkE{LV-Kuv_nZ@pL+hqr8t>
zZfC|Q=;_|cY5HY-$+(uD8FkW^&Hn*W%D3_|O}Pp1Rlo*9LsRSG!$p-)@uU}8YxAJj
z1&>FQs+ic>cf3Hu`yK7HL-CU?`uHd~+!CQo4fF8RWL+!3g-i5O?Ov6!$y!CIm<)N$
z6ieGhrZSEaIq8A}GI`((Nxs8;{F_Lj9VYsrXK@2!w|fo%&1hzjK^(YUHAgjrFm_hJ
z4{W=1B8oB9!H)ohCw8^OX+2T+eI;lwf>n4eTDq0ox{zy3_+jWw>YHlAdn8&B(8CHz
z)0WH&r;R=z3!|A7co*h>j`n{)=KBc8Vm}%$Tm0Ka#6_<MICi*jUlI2til9LfFG56`
zfF?EAjz-{RUhf~^GL#zu3{E>4P%4i_B*7+4e7JUgGG3*G*luAtV*2OT&9}4sP1pTC
z{Fwj<aC^JV7DsfvDkweJwn77B@^QsElVWRIpOJ=``qS3;WdPQM6CFdWXTwF0;@=HW
z9H$X>0G$n%)8@eM#Z*IJfOt8WJ`x?6vDfxd5czoo%(*go?&>XK!uyCi06sM6rqAB%
zH020*_X4S|3+|(SdtyXP53fiNop&^~k=gE>c}@X>_6~A${Lz`^{}Mr=JMbO!hemul
z_@9>#`kJm}R8=@I;|@q2L8(4D+Fu3u5z3*8@9C)lqbl`dvbds<DLJ$pch&=sv_zc`
z9;`hguvMU~hYj4!oR~>CTr96wF8(h%RWG;v-W?PFkw$ve%=zJZQ!Rn9?|8E5!C41z
zCrSee7`<!<zy1smZhNDTcmMmYETLC~HT&a<ky5!WA?O`dbkT)XP}Fp8z(Y!NEP8d}
z@y~y)j*2kgzt8>|HTf5(TrRX3;Io>{x<p5n{N4eul$2foTn?@N64;i!IX@tz2PV>_
z{TG;(lT)<*7VfWw?SW6ix%W3@{vOw&c%K3%Kytt(a?205hj?(}Xe&?pF}8>OKhUDy
z&(fxwmmYv8xKxQ6_$bnvKXBRDZWT7~ayVdCP^^Vg?6yxJQ!Is^0MG%x<kU}GW%w7F
zajutcSb0*6ZgVfdbO(@*apCaSHhb^eS&$qMh68N5YR*QK^yc8K?oXM}l$kR&yA#nK
zF={4?9%huvQrLKb!MLmSq7X7;<(Kv>6TYKg0DH2|dUtN`{v)o7_rqz%{w6SuXeUE9
zqWnNJP%;nE0f4K;rL|udC_Wa(MtP=cu2_D-n;8Y-=fw40vUL#|(0*!k*!@-YmLe|=
zY`N-2k!81zAv=BXr+lopu#;<tY#|jGX=??f8VRcVi4j@-e4*jA-eX{qNH(}l&SILf
zc*jpwA4Q5X>P+XEPzpgcir&wEqb3l^)0DajNgB3f(-sXFg<b^!2tBTPfNF{hkL#_o
z3m26GIF*+u(rfeZx$=Q?K#xnW@6`X|ooz}=l-P}B^~23>8?dX}4USA00zoRJaE<D|
z1>P|j9M%F*xszlyCE(*&Mv<WKzRv()l<#PgV+c>9Xp6uG!4poXu4i4khr=^Wakh6!
zcFv6MR|-DiQ6AQ8Qk??kPrj+{d!?O)hdKcIys$6*y3ax@qH^NY`D4~y5HL#8rFZsx
zhXK|JM0cB;X{U@^{oTiF9JwuP+3$7@FC8N?MF8FG(0LZH-7C8y7geud#x7%^bsT1$
zy!2ZCs-4`}Tov6WTPyXl^P^y<hQE@FNMJt_MUN@w-yPGU`!mb+(z}mythnvJ{}Ah)
zg{zk@T7DY1Yy?oGDR$EOQ|P4p9+o7&0ECw}71*$~(o^het?FXb%BDzTxlol6C`eZM
zOzme8=AYu7FfWb63T-pDXTjTDMU>%s<wj!wzvmZZ(_L3_z9si?`4QhzX#%iY>O>9h
zozl>W>!w7hTm$k#FQ{-U>fqHi5I4%Macq9I`?66{yp8!a-}^5-ePZz6{472+ult{2
zmPx|LcdPOCz}jCu*uv(p3TPI8hI4b50WngQtrigOXes85!~Gb}FRGg6A&8-n{iMV{
zVZ*>ZnxI`hrkYIwwoo-3Ac{sd;B|jzl^77(Om%7nI_Ged`G|W$H5n;K#gLR7r2w$+
z)MYk1_{%e3S4ID!p|L4xL47v}sIfg3Z)?$sw{MuD+6Y0@bn)Cqh)@7bg$X>*HMUQ0
z;bc~cfNy;?5=EsQ<VFeaT)0l;bZcoF;Zgv;7@tx^o2xA4W`eB0pmBqfF9{!IkCO&H
zA^HHQ!8gP!oG|()Tp&*6f6L*0-Fp-h<=Gy-Mk+<~(Q^fyu_dgi9=H@oh@!p^<o~=Q
zD3e#j=gUA$=W0{@RJ7#=;D!T%gkz2q?lK#Gb}@hGHAxkiFCm)iWxVDw=&=C){OrS6
zx4ak(z96IE%wyjRDff|FrbRf#)mKqWA@xq3qZv+}{>@iAZJy#B1Ec*^s>$UXed7wi
z_&&xo<X6EDVPN!M@xVIMoIm+6jduEY2C;!3){s>N$Ss<)UwB4prdcXoc%NlazpyVD
z!|3uJ8?_PGNHh`Bn;1!`zic?fffG|V35vl_nZon2ThcGCeepb-?S}Yb2JwF-(wI;z
z5W%sGdUa-*+!R~Srw9`(U+M)Eem<s<64i@zN)Mk@^OzZJ1qOjhTmaw%m2MC5Zb-Q5
zT~{w)9N5ysc?%dZ1p#>k=s0$ByECa`;EUMk>DoC^*?cr?b(<Gsg=Iz1-**BP7F38J
zk?wg+)|fjwRvgw<#Ed_DK*eYXII?Mk!@hJEeIuprvm?2DB}Jt-r4o=Cdn%d6erYoo
zsmBF@LEu`rF2r{`vYBG>N^X=SIt;mE!F|4nEgW8w%qBi<2>c@t-=JngKaF&>`870L
zF3g@`bGKZ*hv#D9wm*>-P`8Y?)Td;I)R{Cyc%-$UrPRn->f7w$WW_+YOZ7{^hJWBY
zfXCgmO)7(UjX3{R76MN=85b(mNU6U|=LO;d0nSkXIE<ft=*td|`XgIVSQa|%k!%jV
z*qb(=HHsOMM0+~MTdR2g0N)7P5cO!YX&#zU>a1sdOOg3rbXbP7m4YKzgl+RY#*UxD
z&W6naC_!5nd57QBBq<3&NkQ4GV%8(d$|@oq{18?<dInv^Bq?K&>$O;0kqNmu&DC#S
zMp3<JdpJ{Rw<tEaXPUCUj*x6QVs>IS1CCFO@2sUK=758#4ILluZ0c5n@OdI(ZmYI<
zzldzsp<<|Br@5lLx;G6jq2(cK<s)GD_e#*-Z1MD8yK^%WyvAV~KCk{O7=~8{!~{^z
zn?D?0!}YUJRw%Te<&{>DU0F9oP6X-@N<yQcgO79M)>i3%JwAkON;YQv?KAr479u1b
z|BqUbm2(s+d{dt`T;V>E;Pvv>>|-E&P#4wiu+W=IinPt*gZD|>X)oTvGb6xr%hpHW
z3l3>OQZ}O+CD(hSW;RSWQlCA6cq1J|npm2=^iMwtk`tk5_9DN089V4sf*jYC3~#=u
z#KTPn&3U^vQIIgfe57Dbg<pU#2r<c;c%yF>Ul*CZTN_0yKslW=3qUl195Ns3mYb6i
zV1FP%4Yg?|Z8gPN7Xk)adBzTbTnzy)%<Ma63-cez=?upXGCB5Da#TwNR?=F|>T@ho
zy{=9;^OTg3#7ymJrX3HLRKuTO*AYd=B!oOe&uxP=&urJx(wCGfC%)e?Mo<lqn*bCx
zsV~|IwZWavvs94{^>jCzZ@&)dJ}{#iYgH2Jq~Uf)Vm#Yn90b~RP=_@k2K;aU%k7Ye
zll4BUA_C3=O;d~MgrgkO7Gt+f3wLDn$}(KKk?B96>!vZnUK5XrBN|Z0+()Bo90|ol
zqszUQn~;9Uih4E_Z)u_2yb8lDjNMcbDRjgm%fP|E7kj)O)VozG`|?0&EQ$;>rG_D%
zP;x)9CAm&pQE}vAM(!34RP#ECjjsg2u`KUcL`GbrQL)#e$#i*}+r7kS^u(@O@Z!G|
zrJNaE+rPMuutRqI@{rI?9vs#s2nmY$3PtBr9*Sb&9ga}YptvEsXWLH=IeMa{({h6=
zvO0vOv9M(ky}NBvgU9zDcUaP5y0%eMJ{Mo0`SG5i<QvaOLRo%r<BMYH$h9o6$m@=W
z@N2?SQ$T-f1g~rTKNiQaoq;=`;kQFgWMSV=^NzVc^#Ms685|RI%5op+T2J3FvlGM1
zd7%9Ly;zB;FH#sb`V{MsNf6m>p8bSiCP!&A{(NF!&XgKyuks~`8MAnSbp=ZBepDi*
zbHE1&Q*ZPm?v$}EO=o6}j9n{kZ*|swG_gNdqe-EnaQ}?c%)^3d2woZY!aO>#hSykD
zolPk>ry5H`;@ub8D*H375RuJ0I#?5}fpr`baWEu}_xq3E>bfR)<^4n(gH_u_I6Y{M
zUpOKTWevdvi#Xhm@en_{`^>05%<d$YzoA>9Kz0B<bj6Ov`by07qm2QD<#&oxiYZlH
zoy*5*VXSJs(32=U(tvMJ^%0x|IiY6^D3Z~*F{m-|jUN*-1485UYc<{?!+#M)hxstk
zqUxacJzk4>Z&q6j9{iZ+KeQLixD?&A=^OB5baCC_ACzwDazl~B)p=t`jL&b^a6XaF
z(P5}d+AYBAzV3N&86LCVY233i>Oz2z0DqCbMs#s{^N&Bn|7wKX=lr};I7^ozluwev
z*5@}f8XwJ5+m#o|G;GNpt(F{TD!0_`YcZA_6~kJWvMH=le54~4i>RrY4L<S|q#=-#
zu=WJJN+FRM8p#UD7|Q~*)4Hq>*~}2COpNNc7iRp4fj=yz+V*m3AML>F!|k~5^4d@d
z>P4XUfPn;GnuR~|FsfoEj;9Qmns;AGTv>t0_O)>Gc%5az-RMR~UJY+}B^~!nXq#m`
zY9>wz|Hbj2xiYsqzL3Q!JAX<qQ7|4~<?$cX=1kp4>mNKSXh5$5LrTQ(`bP7k&2*)~
zDyQq1mrD6Ze3g29+iy9{@15!hZ2=`0X_1vj9d$$QccvnPRiJnT3kjitA7sFtax1-P
z6mDUB+a$IimgH!@CgGWH>77PM9c;HFs41Kwe3l7b#9kBAa{VuJ^FkiN9@<bxu!;o<
z`EXs*d=$qUGX!in`H~Y8i*s7Eyyd|N@dw&z!NY}(a(c20P#JMGYTHbq;#EGugIgOv
z{+T7W3g3&4Qdz@lydn0!Rx4g#a}fYpl3C+pOW7m()kN8JaZ2mrj|^zI@H}qcZraXR
zw&?<B=}eaur{HKF8qMoxrI^?tNLQd>$K2Y9O_?h3hII_jF>Shz-_-4tt3MsI;6tFY
zJ$aVAyKQOQ6n~fWLc&{aZNZ4x`0Yi4?bdGt92xhs<j-aGmUZaCKIq{e5gQtdCt?=u
z(rI_A2Nc4yCU0YP!C{qMZJkcaP{Jl4H@OzR2j=)T2YtdvkLCtF$Zye}pCw0m)_#JC
z_hc*%+6zqkm3gT?b>2*TRW-r`qe3^)3!?m>z+0;6HHmzA3jp1G%YO%I_{U$`bvU82
zakcG9&7ZvpPkC(@&Y$vR+;U>6zL>2i6ywj`e7d9VLO0p$3tyMv?_Vo(9HmZ)Uooe%
zF!H!`3D9`Y8Fq23N{5Tith86EpQc-3b>CBV<Z6g`az8*3g_?@t8%m%_oLP+azI9)|
z1TR8EUIZ5rn4a%l=t;l5wNSHHv!BPekY4)3-s|S-NI&=mfL{Y5#Qf7IG7`xDc*&6>
zpb8QUCcgMS5_kH+n*H6~vjx@$-lrDCE1AC*qOSL+{oN$__^HJdUNh*MaL;SC)qsYy
zj4Fdif_n?YWy?3>BP!QD(xkq`ciqkn?FL{XN}z_~=RHBuOFCFL(|UO8lJfhT<pemZ
z#xc}<xya0W23mBJS<~RWfi78w(1PHXR+t#d&9kd+Ag0R|@C(^>+2Z1^>~J3Y=hw89
z%1d&2a(o^v%s}T*H-Z;5`dfU!-OP0Jh{$>`_5~Epo&6ApqSwDC#o&at@IiqesRMfq
zCl@rtDsqF2(+VxupJH%$-`&i)$T{VVNU3@2wX$V?;=+mIL;qWqQ^9URPJ+C~kdEEt
zTgpRXi$(dD<=d@EvN!xnB@cCbhJ+xxJ$WBPS-1J!HQOf$(5IIr+z`!54G61&1a^kZ
zDf$vY0-kC9jm?|$YiwU8=NMV4u|H+aJSzYeiL$@;)if>Jtj90yY4jbIC>E=(T=_*=
z;G#pb!CU9B-;sANr`Wok&T__-4RaZQtwoKc*44e2?L_`tV%HLZQC4-?a`uhEXb}o*
zt*vL&vQ|6i_xVhb233lnD{*7Z^zBEIu*N3@(V&vxo*~=aoZ{s)3!vR`83+wyEmK%V
zbnQ4KHQ`k{UN|hvx2KUJXQX_I0Od0@=Njz1>sJcGy6k&tusSd3p?@K~${%jOMv{E&
z)s0vJH#r#A*Q%2Q5pUwi!9AHoRjdfi#V1X}M_6qyoPG`v{Ef2szDDvA>^}QevXcLN
zj>r*G=A5f)s#jWGNa{A@(6TctcWru8T5znx&6cpu&!0{8oDK~zh#y#N0xf`&)kxvW
z>hhLAzh6gABnZ{H&r~XZj;O57Hka0+wVsI|Gfrm9ZUZnoHh3*g<-j4RFIC0vqw&dB
zqI;e$X+6E-$M$GC$p06gvo@=4LNaB59^Ty1%9oPZS;*YQ@&|-VbYRraW0gGahi(=o
z1QCmmb9>fN%`b&eUw&nw1r<7Xj$jvRn`6!y<`;3=+n7l6C+gQqYu==QgRG^$)4|T9
z);uu;=Q6vx{~N9g_>+pXdTK+5`U~pn()M-qqVKlP-m8BhpLMvlnV_3{detd*QQG06
zAF_h`Bds^lBvKHP_^O(%w|17vj?wX5<g5dDm3KsVq!AkkqiSCp#gE^aM7*f4pRKXa
zYRWWNyDcCJ+L1x8e$^>jrEBS|{Sz|hl<=y$ywkZDBOTGs)B!09i@iLRpf6cTYU;!)
z-f&Kq;J)%#s=edt!Uw4ooQ7J(bzMYlCv$|7RTz`^w9tXkwZQGYt(3cWd~-jzDBNMZ
zjR*anPPvP^D68+iV^hMGo55F$ETQy4CZ!#^gAg|ZOA<G}r3X7K{F$c{Q<K&W<iRIf
zFK{B-cf<Zh4Q_bk`~sUBm@)Mv75A!Nq`$WHL-V%ZBpVIDFakJg>@e3sG3k}ciNZbP
z4hn~-SAg>jOCtXBM9xw+EB#|f*aC`!Q#Z`=iy{QW{JU7<SBSSbQy7T9yqao_EUZ~{
zp|JJ#d^+Kz!&Uz=X?(Fm2V>B5E&;vT+cX^MvUIJ5746yyiM4tV;8%7eUA4S>*j;~U
zuS;XYK6X8KIeY|;8#3zD*XHkmVXEiK817}V!2_eL#5KZl3&r`=^%_+ayUd{_37$l0
zi@Uc3j&FW|Q`P3pSdmGZ$C%Rtne@P!S64(L`Fi2%&X{9pGrqd+r<cz<sz$EIhRd4W
z>%rO3iFzaeze%$CPQ-g<C8^?QE)8$YdZh$~a-nERisBE-GZTR|p4*}f(Tq~wbdPR+
zb|1&%8ld5Cig$ciy_X$PRX|cfpPp4to0742;j(yZqzD5{ODv<Rm)C02tGBGt%^l}Y
zn*}zZs0+iSCRuhB1z6X-n%l1BtAcS9tG;nu>|M+LVXqf>++8`aGy$U&+)1yhgvJim
zfKN?O0|U>wCzHF^jU<<)8?ub>b<<K<jtU0d=vby(k3X<PfRjO-3?$9rb*HLnIMpRN
z)EroI(pHye&(H4ru!i_Z^;B1n%r<lP>phA9Iq!n@>q%0JKM@VOtEZI-k-gqC#!gNA
z>eN<QvcAX49o5}gWb90ZS(>-xH}$+x&~$rGFF%*$L&Mo~kF)*?Mn5(w*UKi(F0|^(
zx}Y;XD$N;fv4~2}(F}i5`#_xa-h2N|?jGhE_Q4E!YU?m97>|=|A1&u+*^J%>nm;Y6
zQgt<HL>!JeT!xLK;>Y1#uz*d%p3qz9^nKx?XZG%{#^0c`^oXGV%1@?2r2G4#j|TaB
zW6$KYl)C6HJv)DHc>R)7<H3s*Neesp^h1<Su|~<0FJARYE%t(OBvJ+&{fN-jz!?7#
zEGi^M>_SecF7XcWm$-L%3HuGI`7|I?I)99ivR9CNuFD1^fJAmCsa)Qp<5FvHaTJMw
zTW(B*3-~GIn)E*@#AD?NZUv!ojK!7flA@ana;AP9QDjHXR)$|lXpv6J0LRU6@3*(Z
z*EYK^A~O+6J?G|wC(1T~t!%Gm+oRYK5E>AIPHF~(cC$^1ytoCrp^#WCWy_rCr<|2v
z3$2_+E%;f3Oek!7+VSs6_k<+*3eN5^G+a*jY<s=q&$YJ&#zU51N5I+B;r)KK<7%s_
z@c!C8&&qR|%d;BnH6G?(Sais9)RSH*LQo@s2+r~UZnn8mbt&YoCkxV<6lKi_GVpd<
z<grF-VI{os;Q?1-e@F=&ZKmNynm+3_r?r<r@yV`R{2MLsFX*=5W~WHD4Ff|;4dnSJ
zThDNuH{Lmh0LewO1O3qMElX>17)PVPk?P|fK)|^tSbmB%j{ode#mGxv*C04g4Fy^D
zdcCpL^mh%i^D+4$mATadirbu_u3LM6_$r#LuLWg#)W{UVq#UL6Ok%zPAVNkutYCfr
z95a|f$Fc>M_R)d;Qdq{Yp658tgZZ{E*?9_pRkZ?D{NW<pb&|x#fvx;xb+Q`hYTv(0
zs)*%-R1)xTclS%NE@aGL<dj+MT|BfqCn8%QJX@C(>P0+!$fh_jH>+LeOlRx%cXxI`
zKu0^Omrkt*Cx100u{u&7v)$sOHakAQ@p|)HH9C?3WV(z*Cy{;`oM1tq^J?%%ibxcF
zJPT%tc{uX18e!Zk5;w*b-H}3L_q3(AvMu!n@D}v2EA4yPfKFrD@|Wva6l7IdjT(>m
zA4+{VI8r6wm=p;%sJ`wLOuvx5&(_Bud6f~?16Kc``{*Re78Xg0B}EyJ7tYV+B9X#@
zWzrdGOK!m+UZK=(KOextN+AAvlUZ0MEnN2===S4Qz!Hbln*Be_0r*S(wON(q9f4eG
z1n9UIf5c`<aLDD0^wVx}7$t%cct6!E;gbQ;&#~t#6h-%T9559^P%GlF!0KqlWIML{
zGZ^yu4q(`R;(AO6BtQ3>!vf2I<H|Z#>aq%TWMY(9#_c<GC;=U!)XkFv^$I~TD{Azq
z{<ro}E@#G4;z308OWmvPInav5L0dRK&wx5{Edl5YWCTy`<_+=VhV*zN>M@norY@+3
zzm>bA?iBIeK@e*u$IhkB(Mu6znak`t1(68#fHIUF&7Q2qi?2I&#-T%z7MM5a>HgK;
zN1d>6&vN{65t3`EmXE48ro)0}FBi)Ir=M%lxJ^vNdvWJ5kL8=);zK7avsTea3rBPl
z?rpF3>oKdH4FK<VI4Cx^*$0kTtP^CU__;<;5!D#>H#?p(7OW$HQfdD)t&xEu6F`?@
zzUl;H&ESypICQ)(wukNN;V7xrlb){p!AhuM+Do)XMHWa$_?+3Cg}lPA^}PHhnnVZn
zm0S1Y^q|*<f1EO5Tmu8HJgs6em_-xW2f!|js_>t;B+zwS+hQyKGL7Y(hO%=4+)gsy
zKZLYM0VvGnAdmz|QAYBOoDOkw4?U7_@GE@lgQmv^cA0I{G~a3uvNyfC&u$wg1wv8V
z%Z`s5D2Cr&q#BGyi}J$`Uqt@ARzQ?s6{1_RA+a>fj?#?BM$C&w4G&5|X^v}*c@PAW
zVFS&Zhju;0W|_<xM%!3LfdlV3B{I&<#!e7pEDW16!y;a=Mtk(I46`on&q53fsBo7`
z>ld$-4|S^>#2Grch6c{Y*5XO*FCUz?$k->wK^;Ma_Q;D5$~lW<B=t&oi%Fnc6(MPS
zcz?YoOT_tVeOQV7eRl;NJbUytdiibgQnTGr&-S-wB5<PkZi_QachLAts_PX%y%?5#
z$<cK8K@UN2Fgo~^Ta=9HplE_y^=2vlfIh|z1J9tiY`nP}F&&OhkrZ}wO?@2=1A{?2
za|xmURN#XcHU3R$K7n<eew4^@W1AFqyPgiz{3OGIYsFX@(FkQyqF&md^eBMTNA`=(
z-m*dW3J&YsQD28J<>f16I&>O*>5t&r2JvInfm4xI(GGkelyn@o^zNwZk||Kwm2x(t
zJ$7&%zSNOwFnahM{fyMBui{O6weuTUdv^wlQ9mazMiA6wa9Jx_-V>Y?S*ti?4bU8X
zQzlqY&)w@q-8lq~^fw20$G!q*<o&CexNR9I@xd(k<3>h;9CP&SJJgd9nh*0G?N;ca
z`NhG_!Qg^7tOGJ`0XfX9!sbZMv0R`Tl+N6fY};2HKStXlrk+#dKabH4pbp2Pg+6P#
z_Mr(5^fxncqKpb=u#O0~rbA~b3;i0^#G=$e2=u1(NpQI06@{~#e5m#DW15p3*-wH5
z@Cz*b2>|ex*1qeg8}5oG@{8bv?c&#uvJE+ET-qx*)*sx7;kgFR-&_NN0V3#2b6W{%
z)WI9pBK(vF$N?4}th~{i+$}f=K@-D<Pa?nuz)35+j(onXE<NF<K}4`9wcDi0p6@M<
zNh_s1MKd?8b3m#10baQHkoFLNJW7Y=HkeF2cknyRQEiP1@%fB_y29b(J}c-0m7mDL
z_q8b5!|PW$#kA}lM);&37PB(4#h21M?_~Lg4ry!IfqiohkslF;*GacDfVP9V&q4h&
zUcPDnxgOt`l8k5g<tcW1Lr`OES)N)Br5j{P*Z25KePFwFal^!64Epr8HM93g56_c9
zIUw7$##i7@fBlAIxwRqwLKa3xRj=wuE#0N6JDg$k?u!VCD3WHse51{o9z6SixFKFZ
zi~gB=&~dh+Co*x&yj-(HdTFa+hlmpJFdX0(v1%i$SAh^{SzwigL(W@C0xOj~vE+y!
zMitU?neFxB)i`FGorWTz4Y%VDYbDB3$V6OZ>bOx-Q~^Y8OaC7^w;uCG;QL{NS019j
zw%t<3g84XSDt^u5omgeoy9CKle2g-8E?p;0aSMjL3Wki>Oi%WrrXU}e<*H8kI8}7V
zD$Mbr6WabUsOXNeL^2$P=+QZUOM&`#Qsw=c+1)s5HTe|b|4BOSMT?b@rw#A%&&kIT
zqz}SCv2*b8aoU<v^VA55b)ZG-MxFwl`Q;b;@{lx3EtX7O&-a2IbHbk}`Ov4zIBdV!
zjwV!KwUtIxDjh=iDS!l}@#oonNZ2p*D88!%uS(Am@i-t$fsTie_Dd;b6?0n`3`4ZR
z>6UhGnK&+TU>EuFE&^LnWohwVn2NERr@8L8z38H<+3Z&pns@K0?fn<LY*QQ#Z&E)(
z-(8jv8H#7{<x|>@@F(NbCTo5lVX`S>m_b(Deh_kb<IeYU^*Y?c4}&_h#;br{SjF+9
zUNLH!NUe&rM(epW{Km=A`Wh^1+rAc=F?Fa1bZYnSM8g6uEx@tY!oDiz7OpQ~n`B=E
zZ0N*iGJ|qq^&B(Eb&LKsWi?iY7eOlc^58o&{%hBA%O@(V>x1+nCeE1KHQlTuS5qAb
zXq5Q<bOu13*U7cUM&TB}@orRxceovm>oMIrGt1JC%-e;>T6|c>cY0K*$<@P~yVTEc
zv2ZP}W84aD5X*djAj7K9?=>%i#VB+5pQDFPQWO)!`6Fe5k10_fJs-*^FX;!0-!d|3
zq{A%15Y=EvfYr<RkgLoP!b}WASSo|z21F|uvJ+-b%bX0#Hw*7;#~dz{mjy@Ol#4oN
zzvCpxiNZ;u7_dBek)_)kY2GlP#>Mq*vM%$|X~fF7t;^|+*sBi6$MB5_p{4==o@a5B
z8_X^K=?B=kH2Yvg)yf}2+MNv%qTUfV(TB*qpW%-dADQ|Y!dlm;{5QFHRTXx_bCE(S
zlDt&TtKvLXKy>Y}<La?(t!%TuALKdZXLYzpM)6`fG3s~H>?pweDb}0g$@hSOl&9@(
z3TF)0J26y<o~V!>u6Nb39DXos3D8|oOS&4reR30P1NB^N?-AouNHYqrW_4h{Pshkj
zNykve-?S~5cPqsBnMQEfu#ad$2jOE+Rbvt=PH2R9dy%Jn_q@S5D>hdz-yPj$#bckb
zzg-bI5H@e%m|1y@KjSHO0#4#@c>FO2=?C?$qjz#@>Z9_~T;BGQR$xSAVW7e4pLw%4
z+4|;)Q*bXr5pAjQdZ{{{Fw9IAZ0ipae=qFWJsQZ1y`>vRg?FA4eK1O1s{ME35I~@S
zG^y-De_*miM?uQFmNTb?B&X*%$&FqMC_QeW4#Ehuj){VImIc%)$N}3zmG5h^6e1;^
zQPuj8J%xU)kLqX1==IN*HDusal1~}h5Qt4-&c`F%+pX?7OXY-uqDhP(8hw8hdzZ|`
z8Yp|)<;W!l4&-@TzwvRL1*?oS2mZPWt8eGI>guijW!{%QapA$?|K5z?9B8SzS@JLO
z(qe$Dyf;D%95QY0u8zg7O$xXh8aI1=B*|woer@#~o-8VJ(s1rEfG8tvD>$Y7>m8YP
zT*e!zWt8y~tt+M8o&kP>Ts3+OI;xC6RAoqNMjKrKUr~FDYYZm6hC^@+_Rm?Cy?~-K
z9-J%OKu~mx^@%aD$IDUe%B&&}9o@Tm8w=ww#b%dG{&LKv%Nbu{kQR*Z=Z)uQCLcxG
z83A`Y5++yFtzKfMwZVOt*XA_6S+l%rqeYAGVvlnM{p_YOR3K_oxLpA~`2<0-sb)7!
znJ58eLAU=`zBtm8r)~GyqjlaFA_2tcd-$fuA}#m+r~RQe7bUHS+cg-p)GLMc`2|M|
zvt8x9lzp#Nr)krPot|%Hay$8Jf6{l6uexMBk)xVrJL>7sJDX>HAy=-tnHf_LV(<_-
z90$ZA+OvV;)2r7DwqhrnIg1UA)H$ak7UIry4b;a6Qlc!F=u~|ls-kR@%nEX)Eh$*%
zth6UY44--)wKO{LJoYUTCEZH|8y%3nW9_<i6ZhSv<(276Y$rDl9PD(S`NKbk*j%f_
z-H6)B?G>DD*=c_M^J(PU%X4~A;mjcffhH`lm)Jh#MNv_w#_ca{jJ1EPvob`9lU790
zjv@Te6lbBsU4ljeAq>#{x5tOwyr!3{Y2m|MRD(Lcx-Y}bt;%J2zcvsrlC7E_SL>?v
z9=^ZnNA$e;K>RrYWrF};wHzu!p`&b1e5NmvKA(QS@Cy{}gr9$6Gpv&T;gJjz0h+ds
zkl{5$;)E1_T251+MvI*Oq_12#YhkmOeq5=8dqkAc8++A&G?)n!kU%6wb<$ATkmS3H
z$R~G(=eF(i$ZKTZ(89}humA!Ei17&X+ulFY?{H;A><`?DVZg&JswKpO-@PM=O5nE5
z?X)k3;R$5TzfHe<)yC#7=^38;dbMPHK_g>c=q4jzZii$XL*vWK`6@K0?2bhTIiy~q
z?8~t7R#p9+S@EBr&N8azUmDGp(lt%%X-<qIbbe?o1rig0{CAX3swmoRNwllLEZD7F
zZc*z9g`_8GlD};f3SO4x|M|w{FLUja6PRR?E|@%P03@bAos0axf*?EY+w6MXR(2S7
zX32LJXn`eCe`s)}I3Nn4QZE_Z6}$`M7}+5KBYa|EJ4lXN5&mvJ+Vo?6^D{#zBe@c)
z(n&@}RJP0<IRA8l6!`|tSi79W#$uvO5%38Jz<Pxnyd-K1R3)U%pk6@kxD}61sxwd}
zIbvEQ?s?+7TIL+8D1L*bC91)AW!?EfpmJE`wB<^4%-K&wYyURH|He-zu`(Xf&t40Q
z=SUfrg#)w*QnQ2z8u~ep1J1=!3r=l=L(C53=NOX`nK7|H=6az5WhdKh*i8}<^}T5*
zJ<NDNP_gw)7#6QWIV{eJQ3llex9#38Cvi@RLwS!}RRY^(DjCl6ZOL@|%RnJ8W`%s9
z<E(JG%mDe;B#k_H8eAj%kzMw3g^FGGljZo|zCj;OqMqS~J6zgkpUfG6aft;{Bwexq
zr7Na-=D#lfM~i$@zUgm^yn<ls0`_8^LK1Yzw4h56d*sz#_w0}|SXqa6qPvU?u^F!V
zm7P9OZ0avpp1y@JLc7C{(E=PcyoN|9>E?|0d@q|+H}k@BLq2q1%!o4Ptd25bySZ31
zrmpG~649-^ud1f`(``}J8fdy+GA{$;TgW`D-#o>h>AoS{d_A$2we_x)>A>z({UiJA
zjr(EI-%S^ZAi5GOT#eSP8<WV*9IlO;|Db+xUByJT@XOucz|@Rlc?RgZO*+!bJ&;#W
z&(MWg=z=VZ$XAUZGA}0U8gY6R(}>Hkb3i4k@#rAt$ubFDuFu1EEpMl5?(GicK{~$|
zRS=t9;&p#S?eEAwJ!j#fQ2kq#8-Ti|xDNxuW0MLrs43K7ri?&2|G<Lr%T2BAC!?_)
z@$%P(JjM&plcKPor|tgXr&YHmzJjaQW|#6O0|FvI*W<k0oAbNsvB5i{PXr)t-2Q4A
z{_j2a7EH)K&hcoTH&%vm6oW|N&J_iCd#gzWdt}eYmC`YK5eKEu>ST=ol^#1KxR<*3
zie%9Hdb7XwB&I3d3%*>N1C*h3-1bkK{&}<%+>A~KMM7yY*z#{!L!oHKuKXZ2)ht;;
z(gZcX@W(Mg)<Yrgd#S;Rpl&o^<0mi`rp;OUnv-9w1=+Ht?-UI@k3yl0%fH}e!dW2O
zB7@PwdiJ~BXr=84(6C-R_obO~N@nJ@ptAenG@?=zVs02SO-Y!QD1sIz2}5i-6R29+
z3A|un9#p5*&Fd19Rwxo~i#|FEh0o>3M69owOysj#`i_5~EBPJvMb;A^$(!o#jSGm<
zW&tY7!UYj9ILlW<)n{v5B6$am-ENsfMFyX|7~mnnZb9Ai)BS~&J;+*g?FjfAyRVW`
zbuwY;Ig{HgjYX9zUPJwb4aDL?LJ&pl&F49$XdFr;*E_Da!)nAZJD<Ahm-wbbIgVOT
zUwJ0flwoX~Bn46rjOqg?dE^;u{&%MS7&x$5U|CSx<XD*`gpjoX-8&*1hq}@T4$YNs
z3$3$SPG~8l$3j+iU;TfL;X%xgcc-BGB`rtMiOGRIP_Vjh?G*cz#7oZJ;1*PI>3EN6
zK?gdviY%fY9S7TyXsLJC-}?Gqr|{&kEdY9{76_#uPBrojsqI3$tPx<rgBK@kmX62A
zd2l`@Q_-R&=#kOlZhqdq%doDXU(w<M?>qKyI)C%V3lfr~+TFUf5{Eosn?L27g1VK2
zaN4`_ZEc+QLtOl8#GdaLD+KKX^xNr)OD@W!Wd53w5V0h0fh@Hqa+&zyd_Z;D1V@TL
zTy->_{ejfB0it4x$D|w5lzg0uejCm(yTHp2GaN8Fm>wT|z_kU!<F8B9Wru5a>=3OY
zb=A-(<I_Tn?)Io!f95~u?Z&>OOOv5C5Gecfl5=H;-=UbA-!4^LE!ykA8{&d%-rbcU
z@7B0nRfCQT=+)Fp#&_8+SG<bTM2jOWvVIO)K(=r;$k`H`TW?_p>T~o!M@hse^8uRK
z)$B9@Xju>Ln?h*Gr}MK6G})qrMuxhg8{wy?%}fkK`?v4;cIVn`OTRT?IYe!~^RvKR
zICZmIB8$GXtt3H7f4-IB9?yz7=dJu|>F`1Ja0&7mV*_zAgTIE6%d6!+>C$lJ*;5EB
zlc>0dO1iM}o2A2^_F>y4-Q>HZC)JU8N36$R0hi!K72&6S9$Y3I^?6lTmET%3wryx5
z#lxzED{V3`y!_aBk2jO=cuXEBd*+Cr-;~#};rAcfUC-N!y;Z$C{A{pTx0?@+V?J!y
zz{zCQZA|+l_uz&nblS%QqZq6aK}w2dZT_vniPh^>AHw_+Z#}8E9lwPbKb|Ufv4tlI
z<FWqXw8+j+&YrpgyztD;(Bo!14{4QFr1vXI`2%I2N|yL&LIhHLnqFnYtdKgR(t>(t
z6g9P*I1APp2_*V#bE?c_XU@*pgxdRA283!?XouyV{T|t*EZmX%K!eE5b#1;O-(XC$
zc*HqKqiI&jP&y^gHrp%1RU2%7t&<nh(ncUpNkDqakbRpNG6KWqT2C7kx_Lgsb*b0!
zjBM6yZ>~LOL2Yzk7R5=ElGLOmaB&~^6t*@=e2oHTy@3gOARe}#nN=!qCT=F~u|L`s
z)i*RT8#g1=%b$34U-{G!*a?c;F5R3J16tuK9@W`iz!}^+bYOGZ*Mg{nZD~yH+pidd
z^0>M2ZRfA@wL0SaVHMmNZ30njA}Z=nyguOZ79>3EnIfL%n!Ik{RTB}T9<DfT5?7Yk
zg0T-;UrRi;yWA9d0qbsSb4E!1Z^`)_$a3VZyIBnXaZ-_$EJ5g-SRm)^TmB@xrSPH#
z9cMUE7x@T*t{u$`>8k$jc*QQh{Z8Lt0w14ysij>2sU48In<d&eUSsNVSNB61kRbJI
zj_kf}(*xh0aSWNGaQ3~Xp;3<E*69WDaU^4_auPrb#@@YxPJC(wtWzoQ{kSAzk!`o<
z?igFSP&Y_rd=co-W!M`&+cCaTsQMLc9h%;KW)FC-N7x>jaf_qK=%>nymD7Ih1hl#b
z{%LV&ND@cO;O`ijsxn^yeZaIpVmKE+6T6TkdLkl{x3a?2&LGq-Nw28iF}QE36RMLZ
z@rl&I-;wW42ogbrt^@Xl-8X_~{<Zpy7-0MYy0Wz?qRJ^_EE3P`McIr7Y}^lG*ohUH
zVlJ5I&k2}wUugqJ`$MeaLKZUF*nY8)9ZcI6R<~+}u#30Eq`dO&y%MzG4@Wt;bQOV5
zTQ$ocmh%!y;Ux-d6Vh*(z!=7fmi(gHqSRwQGAB%`R-O|YtdJx|;V9Vo^<4WW_|b3_
zzn?Z&iq|4uE9ED3c;a1_8mCfiAYx;wNhcqKE?ITp{+oU%<XO$+2;)d(9ftBA56F-R
z|IBI7^d#<92(0(R&OzK_7-7W)`BC<Jq=$}r{k)w?S=`y^repE9gV#;nodDHy_dMC+
z*+F+eS?&D(c~4})>NHd7-sN?81s6uwHTvxj>g^=o5Ylu;H+DU%oUg5b%5N5bI!|O=
zy%=a;J+)G}9?tuF>?(&i-&nkHIso&zL_EL=2KAj}aO|wU^LMp}d~*-tzy$iqb0lt)
z(s4dZJ}6e^WciLwkaw&Pt4TjnWIgYd<6;@pbKLn%5afi9DGvwhcj=0ZzQ4&E`sLuu
zMM11ehdT=l@d?Z7Rc8dgt8Yq@6{6;67TIc|AS+mcechz*cq0@Dq(Q`L9EbDH;}|`0
z7L4so`-p>q;cV)U_{s#N{Fzi`I<Y9iX@0snr>kUjELivGpdW+i+&T(=)j^IT=}lyK
zl4tTl{kuU}f?)v$%xu!kpu*U7OG40VzWa+hO%xyJ@q#x^pZ~^zSQ5||YMHO~@;q`K
zes;v|;2t1DIoDRS{3GQapcjrq-!6Q2)Yf1xb*4w$=O49LrvsX^h1w*0F;KY4o8I{~
zmF*^(HAA&Puw{pbJ3AB~*LmlJsHY`cF~@hR|IB#4%CC&5-M%zy(IL9bvqKz23`a7R
z{+V!(2_mytxqq4kx(}#586N!Y9QpSKXdx1UA}79RHvX4f=7nz|Izx&&s+jOE@;<6T
zSJeB`yR(yCj@xFWR=UDLE~Z7hSE-VisEE~6S~sLV=7hco$SY7#pr)nOh@kiUo_+g^
zLK*43_RzP;KG>-zoea)-Uu~G#QFq9?T0$v}Q35a(o1-JQH_B%*o{WS#f46~x1fRc=
zaX?ulO4KnLYGUOiYYRYj_hAH|w}+tP4r|qMOP)1nBW{)?^HL}v`7dBYNKh_XnkW!x
z%=8(se-WZhK3mj9pm7e{HSLZ5l>7wrdT<6~smq$f-#-LrT_Wv($nrV%KpwE_fC92m
zH^(PGcu1x0<t7bzyRqI+HO?&D*J@LreL{vJ*Bz|0nC<<q1%Xndeq#KR{vfM3+iuA2
zN9Wn^jfcryVy%&U>;0SOz-c6SL%C=D*7prZ5>#Hf6PX)n>so?Zx?zWC>yZqf1GB<#
z#*yFm7ID9k)dBBM8Nwin4(f_Q2Q0Ds^}&z*199d77NF55-;|L}7M92PU4N<LK2?lp
zSoC@&YN*KW(gQY7a5|m8xLIy<L)Mp<))s)lMjUd)iU97ZWF7mjCB)R5QTaDZRbT}6
zY?kZnM+>@iSyS1|53GqMIhaOdVYv9xM1d@b5;C}Tdu3rB9MEH;!`tiT28sd&BsP%g
zoUbd}i3ahP-yG@Bg>+LGBkcmQ!Y50D!#25yACAMv*^g{KFIiVkn_E+U&Jk#j3-%Y%
zJvTwXn8?8|F1h-y@%PN|>}27ccG}W?u!UW^`-ut7^vBl+It|99a&Ph5ZgFkMs%lq4
zpTGRW+1_@A9bYGK87+<5EjNFbwpn1n3ZhGfmSUQS1>aT6sW(fJnF;Cx%cYDG2eDb`
zW}XH3eOL2m)IEBfKmy-eSNKM%0US+=1c)WN$`oD$v<FaOLDNgA=Q3^^J2NEfCcXgC
zW`Y=v4wDtUPP<_u$**-VbMDN{A)8U5A+tOOJ^<oYs|DhKKS!nv^1C=gK88{)VJx3;
z-x3`{>mTStx_t}GZ)zT~13@K`Ctcjzd9Y=+5XX8zrT&}}a7UTm`gD`NW7dQyk$zYm
z#C`aR_h6iv%XvL()Vad&Z_Atsk<T&f-&^KWJw=xOr1Y%g^o%u-iD2_>rkPOp$8z7Q
zm{_-*F^oiyD7ou>K+uu+A{-(|%JnhdBYRpVi5}Aba;q8qYEk?s+!^J+bf0jo=K->t
z;qUIvqAym&4p|r(bQwSs(Q3GTd6*da>r=x1b;Al2mz-Ob=FLM6$da{SM`3}y(>R5S
z{%~0o4n!*i(g7r8-d{q%cRG5YgvId-{=t!Ylcao(<Z6x?sl?n>#S9FLOpLNjcfw%v
zngT2DX*@k%0u>wW$~vn$l@obznhY2w1HO8(-}}JnD+dfD(yvl^?8VIQ9R@+T!^|6)
zXIiUsVbj#`DWeJ$%HhI$bN;IP1tqaiI-%8Yo>Y5&c$f`&lMCqyt4u6m(jAs58fHzE
z*ApOJ{(_U;hMY89m}SAw56d%1eAZ<-(OSR2yWe-uNuV@mkpcUl*Xy0UUT=v-)lk?Q
z$=&O<@Sv=t?h<UDH}I-y5SRz9>^v_kPY_ZkQ3A`!mcJh7OM;k3Zq(N7b2%UI9_N(*
zmTi6~t0&(CgfhIV7q-g=+b>i|$-{PJ)6QAmc?z7@HJrVGFhxL&m8LCklMYNa9xAu1
zYi|kw5qwYYSb$c<m3`elUlZdr-MoO^T=j@oA8VaF_=9}8@~opxtVcpfJjJAfp)%{f
zql3K_NjmI8_mHFiwQaA_9R0S1GH4383;MH#g>v9Cpvz7|rBP7|Z{v#DD8z=)pVrkr
zG>pAgk}b!dJwJAilfN*&uiyZGylL?DXkfb;kefV7GwFx=h}*Bb5|8)ZahWsoCkwo2
zn)XilHhu{(>8xTlyUUsz#scz=fkW;0LfJ*0ys|&3|8^67!OcBk|5n7((-3qu6O;N&
zW8b6*8$Onge?;<-{9_=D;8ZZiB;GTsH72hY`1aNrl|>2zgciEj6{!t->|z(sF@Qhj
z|9Ane4`CqA25n9Pdo6>vGWO_iff0kBU2AnU&ukQ_GFi$UD71CWzFHnA-q-1>y%&nS
zWG%PZ10N|9ZU#C9&*Q6c9fy%VM`<TE=Yr6M@;tZkdVmW&CKWof)nxp#{hMM{g}S3&
zSeHiFS{WLQB@ArFT3E>tql5bCjCsSXhYQJMiSKAZCMPq6I{<+ZXjP0=E_x8KM?bPv
z&5bOo5f~+|??YP9Rs?0<19R*JSTFF4%#dPtHxNhuJlVe63lKJwOvmT$<Laa=Wu?|<
z<WkB&U#_;hNfSF=SNXJ*dXcBL&_ZCe)|;b{ulrjy`wTugMIf22dvbcx7iNH8AI;tq
z?3m_xFTv?|H9jZU0e8QajSGPya^X*Ml8l!!ag1HI`|Q#Xzib+{7<PA%L4Hq#l3CLg
zzJXdtn=XOx8yr@ajhlGBBXB;Z;Za|ak|`mCk4}sGLhVV6!0T_=R@U%|HW5JaF}jgE
z*v^}iG-$9RqZywaV_gKW_J>)`ZZ2)+A*DYSXh9-ZOa3NgVY%VsM_Jl%F+_qFq7mtS
z2w&<xe>bi^)GS^vOZV#HP}kS_-TC|Ki@JGC(6WSK*Zs5cUC%T>iyn*GsHcFdfr{Vx
z&-*8nG;DL$OrwRMXjs*vS6yC=*sT7B+(GYkt;%(>rsyG>Dq#Oe;%_l4aO^<k7-JCN
zH{E<51&lmdF34o+URZDMj+CAu1^4!=*zooqFC;VvNhG@&(BPqKfHlPW^};rzJg)jG
z6aNo)ZyrwN{=JWHQ?_JEhC-%nDmIm&BxHy}noOC=lsU7^<Cc(<5E(*LWS-}#C}hk$
z3q|I6?7N;$=bX-Yf8Krip6@@u>vvsU=WN^F+q0k7wAQ`u`(8TW84!qWxF_+U-0yUI
z4oR0_!)sp?J?--5=45~SM~gaI4@QAV*|Q##AF(o^)ZlHK<Xo4x>6%_N8GHAM(1m@D
z!EeRfTQjmVx3`y?*EaQO59Xk^W5cI7>_W2N64SF}%b;X$E_vRiX}&RWkX_4h#J81u
zQ-DcAv}r74NbUt4_&wZgcXI=FlAKTPw{08X+9FgwkbD0i+mjG}c`e#a8eKA;LwDO@
zARoGJJGWV%zZbm^Y#MKH>JTJUX<0E)jU?3Vxyn$nVz5qK*wC-h^Kv@b2RUt?VX;x<
zjMdCZJ$pp^rSH9pA`^T*VC7ksQ1r3dAgtO)THN)kkJ8|~S4b%K`#)Q@mOxS+>QKU*
zK|1+$+Qyg4d=2)q*|ZLorAtxIm>ux8pHjHiYu5K1xzj!l1ts|h6Y0bXr>W#iY2|Il
z)AycfnNY3<$mg(@`;ML1P#U2h<T!%^3>b%GkdVvrc|NS@8oo#Guw3igClnpID%RgK
ziXW1@7}K1#r5PPbF6jBiq~=+?+7JoG%Gjx*CT1F3v|GabmkCt-43Ba~#=je&imtEH
z6pvzIi=<?Ubgb5-Ir;t?t<1gSTRo}M68#@z?z}tdnSCNva#qi}_;WHT<^~$mzt^y7
za6qJ52b?o(w0fo=%W@3w4|mX_f9%-Q(1Gdk4wx$yu|E2?=;H_OMgv6(40?XLPs+W0
z74WDY&6bG)cDfU<1&$azAbX)9t8Q<0uQ(sX!A^(T7nG(5=VUOQ`!G`I(o@Z#I@jW~
zHZW&XZuVB&T3Y-ljCr#cbBq)-MS^LOyJ+m9WmQF_wPs9FILyCri7)WcY&f-01Omya
zI)8rVO~T(lpf@J9rI!%l(O}#r<-t1R?Ho(S$Jb0K8XPh3-hq1Go}6tO!Jb<Wb@ZeZ
z$3PYN*{&+B(&^^9xn8SIClP^@z8OJ8&xS{zLcilFH))Lyg?EpOgQl-6dN;_b8b~tP
zd9A<9OcK4CAuF~+;*{iiYbyyBPEDPhyzv)o1MquhuF>{vEA7Pgd)2i1L^36&w$0rk
zqb~FX=E#ZRMTUsB!k5WpBYIKu-TEO!)HZt@MZkHXN~&tGBFdv6G+%h?gVl_K_0^I-
zixQ^+>}_K4N^_y~yVeSqv=y~wRh`Bk?KHb@uQAVr+@*5mOTVEof4I^4Wf#w8_OTtB
z2tL-Vkjd*A&BD`Z4;=C?xi_7wld^N8s5U>5eq(8-Kh4t*WM$luc|w*9xfH>PA)F!o
zVuid`%<%(OicMN^!rmn+voZZk*2HGjj$(f9ABSpv-NceCS`WXFfjq3BORPUEC&=0J
zxU)t?O67`m9l6Zx!6(FZF3(dlia?b-P{y}VL$p@E!Jid!;zmEuB*7X*j_MaDL*^w^
z{uNd;m`w`)(6#)CZnSy+QQMM!OQ$5#j@EL=J&8{g94DKVmIE-gq?moHMj}(Ru3~ni
zU9I4MAi+l!%jr3JICjYRc!%7qy9j5mS7@aOx=Jc;xAeiLR*qlaqRSiW`}$*oQEq`-
zJZsDoFAKj$inmNN?XY3z0^)>Y$&|pH@OOT@IN@%b2cgGHlHZ{YHa-b7UwnN7H*d$j
z@FlNiaRJYWvwM=vl;+WIgMK@T&vr$sGEHwip2S;^u=liijF`>waqkDxlO}ziIO5CQ
zm&0G(!F=a;!Uyh9ev(F=cH3UJ`t;Pl_{x2UAea$N(l+;x&ez=~rp`9pz(C`8?S-{q
zv#3Z$!!|(%)yUcs63>h3KJ*iA{C9Dr*M9Q|(Wl(=cpRK3=?lbNbUJV0R}1-A;Fv+)
zbvz^l_YIhkdL~;J%rsqHxLoq(RCNCKiHj&lutHt;I!>$2+9)1Yb*o5C*>+(RfX$ny
zKaRcIomE2B7om$IGb?~HUnbuj{>Dj1ZxHPg$~hXGQ0Y@v)Vg%hl~6J>iQCB`5~RR8
zF7#|74a7cPXt{a^A-t-_!Mk$<wp<CP2?@f?Q^eKSyFD%U!qRn;`sA}7>@j=*+8EHh
zJHWcPjzvV=Lnl1OtEkbU*zFQ=oRyH=-Ku_iXIBsQxe>#Wx3BHd*Qm<b_#269pP&y1
zBx-u#C186kL|g1B$=Twie2kU1HNw1F?&n`DVF7MfrJIk!#3f;7C8WuV2$oRKkgV<X
z?fV>*I-V$6T}p`ov*(GIDYa?BKayk_3bv4Ql^oC!*Zxi7EbZq=VbQxZDCN%%RF*T@
z4OlYW{hyn=m}^DQ*IrB>Wzl#R^X!UD*&$#S3gL^_s^Q{a^!0dvarFrX9VZ}b8wnLr
z(p53wXTrJ)Wg8SAy*((i#E_1DbxnMq<?LWtQ?j{%9YZK{^@&*~Tz6X}t9i&)H-EeQ
zm^J&vsFp;%wwR{3b2v?R+geu8@+|_1ic`%Z?Y`dV?R)rZESvP37i%9bYf3*&FZUA+
zh|jt%e&lWSl*{npM%N>}KF!QBeW#9Rm*dzfLWXw2WbUN|uS#pY7%JOdY}h#%c<0E#
zGKCA<#8?qZSJ>SI8*rCnN#0Q51jpOgTe%rG911xXN-MLDevE7GzRYl*@m^#nw^*^&
zgGq{6Ny6E$BI>MdGFP+@Y=V23M(?-DOD&94IhPN`Z-1|Ha^a_SjE$dUreD|pPym5v
zWj&^H7@fL(aJ`<Gn}jbig>Qrk(wmV-MR&OWn*9cZP`=qu?6n^=PxXw?DP%tIq9)-5
zRVhh&m0JgcY#tw(+%{uL|FVGFyV0-*^L1(TRICXV8zrZ5v}*n7KsEN7Vw#6_U&bDK
zL=wpv9o1P$XQ(>#uoJy_PeK_LInZ1}_weJ6pBtfvM&Mf~LUpv=y8gA;i+3V*tf5-E
z%^B;+7CEiT%kH>TUVm3Vw1u02+2GO{Tk6b1VQF~J#<C7bYhD_E1QaK>eMk2<Tbotb
zv8hi!=j@>K;lR0N-hNwnQ13v#{>n+y&`ho%Gxp1bu4q!K&MS2<64B@-yECuxh`F!R
zmQ|$aulwMcvjr?<Z><NT$Bo3c9!Id#Mm(<zV^t2LpKF+)!{ebn*WgvOxeuZI$4Qoc
zBxtaP!|oL+s$ub39(4O|r;zhW+gNz-qV0Xf_vEFg)Pc4dADXp@JTx+f+#>81VkBxG
zN|cjX?=$yrK46@f`@!t_%?v-&ioPHeCHZnjZi-pFyR=3G*@WN+o(O{iU7`c6Hw2ga
zw5f<!1W8^MPL30m;#!M%Qviy%KHJax8EoU)Gg^#Vv_1*loAvCOPBDETw)mI~_cgOS
zoD5Bx)vvxR@*p@-`)S6}VtZkXYiDcMmGm%607-v~<Z<V2mQt-UrPFEHNgX&PAvRtm
zFm0^ele=_ZYUq4SeA*}K=hPqL78}-f`YX<eevlhwi~A^aHzOg|fXHJ50FXn4U*|O>
z4qgu_#58l@nu{#Qzq3G@Ka05D`xZ5AhrovXRsX5w_bvI)4E!Gt_{>Q7sw#%PNQ>n}
z_XiXbhdz|Ji*JBmxoVU*E6wLoQgXzt<`DU;ruC~7cqiW``UoCNco(H4Yv=21RW16n
zn|HL%$=I!w6pl#WkgCK#+POh_I=l##9((8Y4}XdMAzTf6PJh_#zcP17-v9Z}ljJ@s
zogj-bHm!6-nT#%B?+i%?hg8+*AawAW#p)HYg+Iz4a@&#+@}9Y#R0iLS-4iJbe@@3`
z)#C(2({3?Y*HS51Xvj#2`mqHeSM^z8_S4tp6tSY{-pM$3QF(|`Btv;Lf7cRxOWXW-
zV4eD#w{q$JFL!EQ3?E`_i-)P=X;;k4HzvNG6(2f8Xw~EZ60DME&k;+oR~!~4Gx_Dr
zbXkZzUu9O8fvT5I;}JrRB*SCbqAcPD(5a_pSra5-VU(45{K}Krx*+M>3u%Uo{loIF
z5tZW0B>-TmI4-{BjHX9gMQIitFc-ck_GgX>@1%Z!vI?Wi%;x-SoA@8Eo%#@-4;L6Y
zYgS(C05bt*Oq4O3!w`x8W5cMU!@V{j7msPl<f=g4%cn!wF-J~+Q~HOkhY6Gx-Bh9E
z{~V?N7&m$o!dM`M&Vva~?vcCqjwYiUxO3zi?9G3F=kp8VI~D?N!rv&AzkiQS4i-F}
zijTIjf5rHJ2K&GAU>_?oXq4qxp8w?*{Eznqp4{0FDmv{u!is++@A3Bdh@C~Xx@8%S
z=*C+`zZc0-FIU%Tp8jx!;+BNVE}aa#Oh5gr-U{OQl~tJZX+#}?)DH>8hoNISMM3`^
zxpgpO7%z%3&+n;b3*>G)F$uA`P2tqb)M}uuGj~Fy8(cSyqLL-l0w;p`diIc{INg;G
zDiolgFc>Q+-zVw`fiXMFAy$`(Q*k86(=w}vIrR=`Er11yiQ`z1P}Cj4E)5%BB3jjW
z6ZIa7_uxEnSMq+=XUfe13fE7P3VnN~>xh|MDBZUDhW7L>fMv+c=g3>XU93KEU4BGw
z{)GOUeahD`_0t_!Ij3-m%<EHITTeT;&Nz!s$2@-jvw7fIb%X_PSf{Ns@B2j6!DpQZ
zuaUD~pcFL%YjmL^NU)cS5!NS@jXN;#{0JyS_!vQhJpz*HS;tcH=W&%_f5Ql#U9nW0
zv62Vu5GOStigT7i5IoYUZb^n#lyRmfA720;AwPft1h@-_z0QsFI}EwjeDh7|HO=q^
z3!c*&kj|Wo$?y&&VFYs8H}=UF7iiH{-BrGtm-%&fODI2YYY_}gd>|;}ECpVkTQ^ve
z2^ExX)>13_yv0PP4ZI-8rdSvP%OeJ`t0@-mrNp>xk1?;AvdY%7eK|AvWDH7+7d_bQ
zW|gLEuY5>t?OjuO6%NEwgRT!tY6lOR)BVRH`Rnd^j+}=XGIEvtlJED)b^jQ?DykH$
zZ8X-s<0<fl;U4sy2tjPpG*ONa_4>glXN)Zf$7|uEJq7#Ba`^1ObZ4%Q26i!Qpz2n^
z-$_*GDY~l1p<+vuE|=vN<6GUrusTpQ(;Wu(S$eNkxDYKWYwAKttYMT1eg(}-YVrBy
zY3;AYRILz$GTDPgr4=~6@FVwSQtWFFkpslOaMJo)kG=7&__{Nv3)_*U!<5IDEimAq
z${%z#rDH;{AKm1~hIjB-^t8BT^)AxlZE%V%1mG2wvynXw7QQOOxD)o;_{-MT$OEyI
zA-A*4ikq({bEG8V{@bENp63&RUFWn&gY=92zu!X7tCFbWX4>1MbS3*KKz${NWz;Zu
zq*M*wm8Wmh-w(SF_IISZt35^(h_QONjyKtbOWDzFPKZ4ouj^;IL<!Vsjd%KhYuwT0
zpdH;LF-^<5fH+CA<LQXa7oR3xdk1Nuc^7`AOBmYF0k%t4Y81r9u_FK(-XRSIZylGw
zXJUlolH*aQn{P{OXv)MqXAjnL3`s7ph;WHuQqLQDUza+WBSN)9`_oaEMn+xf@TH(X
zMqL=c42_b}oRCrV%j^^c4}19L^6$)H=n(IS!|vM)u%`H4&==jhZ;LnlGBtf_7MErA
zoKFPZ!8(`<*I*1Aw0!FAiUG!Gx9_{nACSpEcF}zfP|_k_+kb8I`}!QNAlzLDaVON^
z*q<EhJT;F9H+4eCSGu$*Ag(f31p(GT(?m|QC_^j|Gv;A~dF!5$S-sb7+da`_bA7Gi
z>rCD5tV4y3N+7e#yl08lcK@U(oz?5puMb@3XVPBwwxQZ%eJ;&KL2PH)a|tY$E<14c
zTm<aNnkkYD!gibCjA#$L=P+g*rJ&gwLA=x|&8MI`Ke-X9>vynPFOY{8^tB;&)JlqV
zr3-HuEDW$QgkJ+NQCuQB%k5yYw(k8ki^7^*o)%E7^fXo7l^JM<vDdO}C4A}0?UygA
zA&l$+e@4=<aQ#aJaY@rzb%#lznr`R_%^zx#9$ucBYyyCMcPH-awoxP?mDwqd8zZ-=
zRqhp{L;=PpSivwSKe=<MnRtV-Ui@G%vzTb<=2U7dg~mf9-HxR=t@A9C`<66I$N64@
z74Ook6}lI9kiDk{b`cF)1^v4j<9KyxSAy-s>@+V$Cl0)!XXvWWnb>}c4KWi%j0zpb
zx4?PtdFaHK2w>asx@@+`y|=i%8JmY%huw)VvQ2qJQ7r#cLZI_w{+Ju{-kT3L4^6Ft
z?u`~Xq4uOJik)g%Imr<q8sQeQJimi{)}*(nDIy79d=^W)a#m@|zSwQo=Ez<YRoL!I
zhWkp|aCv0;vwC0alwxVZvi#QCkfl!y*XjoyKGj@gUlZtOyB>>6BtCiLg-TMOk&g_s
zYsWOl%vcTcZswZ9srRQ7yBf55Wtc|G+(w_biTV}KDcU6lJAat7QHa?Q+<39+5^m5o
zZ9``9-h`TPwo4UnyvOe@ar1q|xo)R?VWZi7FwVR!Z_i>*T)E-t34NA>M=ky%2K;jg
zc|K)_>EDvo^$6hy!oPS<e00CnBb+flAlUiHLCAOY{sTjB2O)*gGdvBPAI&4)E$Glz
za6Y739lUEv_u>|BeeKk1I_)pkEP`XIyExJ+0<5D@uW=$DV8}*_rkie)F%Eu+VJEBn
zM-9R=xHG4j>1@&Qy+x?QXYI|#T-ap3<iX(hXMnY(Lf5gag@<w5M>JelK8(N?L3*?q
zF|@>&5IX$|Dq^}Yu_(}=r;sE?`}Az#m_}0F%KOt}_vdrscDVb2iG?kd^uSZBJ&zW&
zmMa12XCYS(7gN}nK$iwJm{EuOdF+dQ67oPF9xk1x<PRsp)YSzs(9MGXnRrxNQSCXy
zMtWYAcYd};HHaDHJV3l>@4lD*a<X0ilmRG?;eQJ2?;Xm=7IMF&nbg4SroJrvIcYvs
zd*>bJS6y5@wKXZes84uD2%N}8^Y<<i!aNTt6kmcP)&2xWHTrZVAr%WY_yvQ(O2W{h
zIKOSTuD_FJ5;+nt*e^(2;+}B?r_*`oB?!o!qm1i%_!!@f!HMB?dh=6a8Tr@6hzi^D
zqnD1q3R5r~+X5adeJ98V-^^{(5qf-FfjTFN7+7R!GRo~Fxz!lMnM|$?Hm67|^qW}T
zLbSadXl@;~Js$)tjyx1TKad>4N`}@VKFOjZg?`*ix<gZapv|*ojZ}NxYfUDzq2R+y
zyAp0~_NypIu)8X%3>ImPxEAZg%DvV5;gwq-*TVkyB1?@{5EpGayrWt;l^zBbn}b$h
zR_mmh`1R@B`AUV7aW_HAr<cmu&Y>S-%a+?J&pp0N;px1Ac(Bo-T_7S{6};9tv_bON
zeFwL8GG(g^tW|oyP5q6GM6H!Ajq64g3T`h&<CIJip(b|QgJ2B@Q=x%Q<w^xfuF9tS
z?BsMtg<Cfy278IR+^opHUXT;LE&i12`#${p)c*T_{Z1g!t@Q|($@i1;0mnQN+yGoC
zgm}h{z~Bdwz7-s>ZRcwiS@MA}ZopSxjwdgS6pwO;wG$~XYt(vF;>%q+%HpSYX9e00
zlR!v*pVW%-09*Ztuqu}%+Ane(Y=C%jU6vsL+f3pPS4H|)6ddz^eVn^R*B5Q-i(99R
zHZL6U!{*C&HM_A%91%OpvfB%CHquw;pXq1EA?CdV)L+31HZbL(3p;&{>#@%DNO8<%
z!V7kL2pNC`u2xyxz!9M_>Li`yBYvyLAc73&+%ej-fWDI#<{#mKNJioGd?kUp)Fqre
z<&!Un*fCbaJnutwTF(0^TKbUyn!{3h=%wJ(qc{nwy`wTyHRsm)$EZhD-_|71Je*WL
zE?Mi_Z4kf_j>ZvY<6UK@C7|+RyCs!|8|Gl9RB;_+7@}_pu%E(IUrn0`C4NWK%0^qn
zb^RT6sx=%u&qzg<Oyt=p2?_W($ZMbDa?qI`91^7i_I*4o`I1}@$&}s^H2CJWBKpL>
z;|3J~6bU8N?Pr~d6uzeuwxjyK(Qo9+Vbxwq)xgXk#q~MB!+LHBmb3TYg!tr8#gaB9
z>YBZ(R_#l!8Sgv3OGtE!EuEL(O$0f1M#~hJ9b%bna~VgOWf;&xZ>o!4(7`Ej^-_RF
z*Ng>!^gXqr$O#tgv!J%Mwjmb7B#;v<2j+6rsv^1N^^-3<XOinrid{$mN0gK==v?AW
zT0+N-E6N2gb6ai62)l5}B>n84;;0g1!ITWUnRq{I)4%XZIklYc@<BWQmn-**-NCo=
z?%DC!pn3iQ;?Hx15$$tUVun<B36+<hsw`+-bph*(JlB18Lvi%LuYelUi<}{hm{M%>
z%luZ*(g`9>Rd>J0swZJ>;zkbcnRe&a8#&y(#6zl$)B<i0=D7ujyL9Z84*f}JvQH6T
zF#g4#NCBT0I`a9EsObhrXf-249qaf`A)Zn}mAZ+{C^ultIDu)%4Z0$Y81W+|3xzYe
z0UO2%p~h@4jIr0ER~Ksf+1?<&<5BgPsA1NqmGoqN`3_oz4(cXh9saEA{Gn>?Zqy7?
zHtY;dD?0qvO8q7&F;S}pTjvk24mb?i`}5D7^3>q>)bN(JPYK9O2w+VhVb6&+-Lar<
zc^~N#C4MX_x+HvE;p2g(+nW4W)%i~-HMmh9kftWTK>bph`h*m<j{uXtSAuc)$Alv1
zU@(y~B+51s6O(VDRaifuP)gTiZUArCBSn2#I$GSaCqgiirEW=Yv^dv9R#btKTEVLG
zj``Hn02`4Y!=qUSVhq=uA{%s!$hXYm&B00OrXjzaM{BJ?&%kb3@S-EQL5sv-E}I63
zne!dDZcRIz9LEz^^Grnk3ZsHc?{DQ6LXmc&8POQR;EozIKDo=S+^J4ty47O{IWmF5
zk`5yqq@ro3#BXY*P}ia?-*OOs`e2M-6UjT5?0tMYULn+tI_%9-vzS=aqrBwq=PDiQ
zIg6jiEfR*4cRtHJDvlYgcVF!_nrIek%}9BiO@B}Ps~PQ-bFnRxIse`Bv;JTQ;K;Ai
zHYjwZOa8`;|CLF3G)-ULxS2+oOUBKN#*=FNSBr*Hn{KLqjEx}ckZUvVbV(#h(O0W@
ze9nK)t|Q?05JPS#&Y;J#?B!Td^6iS4CYjAaAiT&hj62;kQl0&P=(hMnjW68;r;3Nh
zYy-v#i^(%~yYxI8&mRVieor9&3@CnhKp#t(S^D(HX~%*>db2004VqpOs3!6eRLeSC
zErL;ON>CkBl_JBbQZ_~sJ$kdcO(#V$q0AHY>=A7%IrWyIG*=W+)wW5{kj54qvM;sW
zS%^G>NPFpODRE<sCsn<@)u@->j(xGVb5l$p$hkyKB^pkYbKujsVY;b0`FWH9ZFE9W
zfhkOd2SQSD6N}g=n$p~&g@AeQYzdU{tPR74C8LgZW1BMnnoo~{!jOO9Fn2W96qOy5
z^?p5CAF7Veh=*rS&7q;|PLX!`64s`|l%BC_CiinwT%IS296@a}D{XSv9M$9s9_Q-a
zV{kj&?yEzad&T1hos1>Js~HB`_TFlqckpu6tQ#!SXG+wz?oZrlljfRX6G#n?Rk303
zo8=qoG9ADMX3|}~e3fwrop*2avxB-A@Q|R>MHxDel8im`!VP}ar?<-_?uV=or6uR%
z^mClVWg(`q0pdpFfW~GSB!dFMT77j^pSN<ZOg~rI5$)QS2ZSQQC_)cI*PQvsdpe;n
zf;Vq^{hf@ZE#tQ`F(B*u0eJaGQd2p9goUnBu<!XF1IQlW5ir#s!#pdQqYKfmgGyo{
zuYA>Ev3@2DR{9~-UuQ<i=~W&xm}Q<9&Yh<&$SO<|n6}Y{FwK%tzq8#-jeK8Wbj^&V
zTBSfA$T{fbQL0n;r`^*PBK^U>INy}65_4nOcK9BhLv0}y|I^@<z(OmUyWk;DoRb??
zaf7GgVc&U;V`G=|H8TS~dPKJaI+0h^BxR16AlW)3fcI@tN*OnK|EXvk`xl@#@wiyA
zn}$=@fNw{?-6;!}l&IUsKzxbM>Of+`b`^!m2>HVcAcWZ=S{hI2Rbi=*T1xR%TQ|)<
zs}-36Z|ZTA(Fz@9-T@xIbnT`nJ-Hb9uFK=r5C-`oj)~^j?OKEE04#w;l!EH#x(=no
zh2F1YxModqupT)iXTX-V8qC`}NAXWHor2`Jfu~@f;p2urwM*)}afK*B-RvB!aEPG1
z*l%-2r^B<8<*ifu-`X9WX&|iBvRt%?2~6*65bUoR-9uZDSD1n&{n#!Q7;DjgV?q-4
zg^=Srvw6+(lh2y9H1m>|rZ$sehEEQp&DzlzXu{TTB8QK*F~OW~I@r_gPNu{MXI}>A
z%@fo-w8ehS6Js6(<mmnstI2b~|1fgMbIC}&yi6_gX!V4~<}P=m+q4o>vw+gpjG1WL
zn#ERBaqp*gMKLDFd;x1CRe(*epNtYMY97OnHi|E<foU|LaUkyJ(+F>Po}2F3rSB-<
zC6XI7D7I`qmGCN!+r?aI%M$rD+|7lb>e4>>N>ZH5wdRr_@iCS!EFeV}P#%ipsKTQ{
zv8|p+)%_ixBBm^TERdvQq2#S5ybtFQDK01{L4Za1_!fON^JizN-F_ofK7zl`W^%f&
z92x~--E{Xe;Ay91g7NtJg3y5N^@gYJfHHZj8SFE}<BPPbd*fDXKCyd(8$vAXmuSd}
z4{c>@`*{!VYvt18m5L5y9}hTvzqp=I%;{-qV6EOqjsS7QekOC!dD<YSYmw;$QevF(
z$|`P<#IO81asegV^Eh?zt9hf#DU}KAJomBdwOP1XWmR3@#)Gf5UsM?GoQf2X56M`6
zzMIcBdW0g(wBqUQ0c?pyLBf;lr7f!4k6wRhUWKBH-Rk3`u=oK8nzkAZ;7nHVUR2A|
zvTlxV77|)uB0@Yl%n={?8pK|`QO~?@`@>aEXE1QvqQ|<r)0&B`yiA{rdU^oW`QgGt
z)I2l*L_<VjPisDQmy^|XtWvc6ez?nWs9g-(f5&(E@DZd4pHu;WCnSrE%tS^nYO7lZ
zduHHv3!pc>G>wq@iKKxGO^`m_P2;D0K)Bn3G3fIJT(QpANR@%VD`$q>{L8d-qn?G1
zW@O=)?1o}-<HpDnSbV7?exI8AU;6<B(nmx=V2(GBxcreaF&sR)ey5(*pHPYlN0S2h
zfS4EmWdJ!wpg#mTB!JpgYE98xr-FI)Lq60kjMe!=r1@GpM3s!cs}}pQ_V!1_c~({?
zbo|XzT2wVr7T4!>$eI`tN7TDoSxJz#&J`6-Ll2BQ(~qR}PRMdl)qhaYp3#dibZ)8u
znNV&fR3U^@kK9X3{kFePBV^W8Ri4+*Y6W<q=GmXRXU@a~pXwUG9=j0PtN1+8B0{qu
zLNo__<a}>RTi@vEOhsj;3q~o}h)KNqva#a^r*Wom;0KA?p^553^OKngeq5U){><e4
z?R~z&Q&POv9j&5%a1M|6Dwe&RE;8hT;m>P@i-+j$)^yT$50!u5O~3zmUk|^g{Z!U9
zjz326$^{+B#)2`9rak5su8Bo{=DilP4GHJoNacuhbz4{{B?jn^7L>2NdXukJ5vl(E
zG-NGOHbs4_H*7Mdx`gN|%|jYy`5p3F(+(L3(YOhv>RNZI8QIYqE~L-X3r0Q1M)h^K
z?I<}!KCfOf?leg#O63UxGmTb4b89fvHy3`x-S_&NtdoyD$~P<7b%PnS^FR_f47l`$
zW-BndY`pukW5_zZA(uO5k=uHq)yKb;n}_DGQFx~a?*x9K)mL##{wzcO)F3VGqeBTg
z=~U-}FevU`COCi%Dynx(5=r*5n+I%$mnmzGF9501QMif7s`yp|=eV&lxq5yInWX07
zqfAI|6b&6uqPFH(F$F6|q&G%btdZV8+ev}0vDP$|3sbvS{J?(xOjHXt>Oo<~t&Xb&
z4cLD3I5qoS=yW`?-K1yQ!%9%>`r=vmJ<R&d<{=o9FGNTC5J<bk<Vv~9D;1<##UHWO
znzm^t&%1Hzcu^~Nqn${8uyCs!qg`&uNZXpx$m-3EDVOvYy_QBxq2&&JT^%1TI9ZGw
z?4%V*I+E+sI~tI`)*4Rj+|74$u~jnM?)gu4Q#6M1f>rM$+wX`@B+IwWbdThNQ-J>2
zA0V^(Ldsi8*%V_zJq-`|qVk@?#zmbZjc3i`bqZri3;|ELn#Q`-$50ts&B21l9A}!R
zb<ilzXtQj`cSlod5#)g9^Fc8;7ZFGq6V<W<k!RFbIYZl<gogUTlR}){3JyT&bp7z{
zwuv{K68m#pI@dLq$43KsR~PvT-b$Aws_8E70OF%Hvg<jBCeD%XK2qPH_0ZW7L~(Zy
zoh3oN7*Z77xV5^RJO^b3hbI7+k3v)054kbSc42jiA)-^C`3s_X7IHjuSr3_Wq8O(Z
zWl4qr*wLJic3az2lSVnhS?XAFKC}N~yy2`nyF{NS8c{<>JW>ndo>?ly&KwxYDax|A
zbb-mP<Y;wch=uf-$Je9ESJ^sl3uh%Wm+eIrnLz{{y$!3i$0iktm1CVI9vdr9uqABU
zd%0|~Z`MKzUXc2+6L!>q#NgFXLJ!G-H6y&WABS?St*fK;g8xmFY`Kh_=(U4WKU6&b
zi9F~mLov+iFV+p;UJ{)mV!4ZQ*rHo5u%kua!4ZkuY;+#^?R=#_Y7UGJP4+7d02pCy
zm?K;OsDZSU5!xslPtF!i+L)))50OPe(Z0snU;tZ_)w65xnI-5judWYAsU}E9*6l)n
zDG|id*Mf;6)wIVWo;cO1M=Z^C2IMdbecf>d!NjC#2cbe{*xFw$D0N07ZaFe_Y#q?c
za|o7YJ@q1Gbv=_`*+Dh|y5@DWfW>mZHMpkVn{extc2;$l9QHJL1BspBEa1HhuL=fl
zUqsq99iD`BZ#;V$!P?sOo{~^ulo_d<BhQW-r$163<?E`%8nd6<Fs+OUu&uS2M@%r)
z-y0FEphDU7_H(CKog`Ynp8k<6^*RAnCeFW-3QsBD?Yf0Vp9cFIr0c`<GQ!g}h-&ks
zeB_KYP+s2!i==Z=C0}GU-670u*QFx$>exWq^<cB>*{I}PH?GJyMao^KdwXrQXH~lI
zsl{)CweH)Yvk811<DqT*_Fq(a>7*VyQ5~jTpiE*jYa+#tTprXZ18c$cHJ7u=B0ADy
zz8ops$J0aNgGF(IE^u3%E5cqyFiN@ts?axZ(vYShxmJnx)Ah{3Gk^<96Sg=#<sH3Y
zqx5CQ79p7;M9=Ft03jAkIOd93Xra4;>8Sw7p>;!_NwB2sdEDFNXYU+(9g1nguR<u_
z^ky!%Y+z;DII$zomd8f{?b<fC+=B%YSls(pH#0u1HJ-~KGAf$XBNICQF2%KVXuh8h
zDjK!WN#dwx@q!u@O%5S(T>s|ov6YwaC!^f9W^SB#xhkO4$7MbT5Mt4qf*4dJ)Mw8n
zJ=90qe8SEX7Uw1g(|5C};yv{b+1p>Gv~N$+pCQDsy$~Ep{v#b=<7*^|G)T32e`eC>
zczPjQH=#{8{5kJ{AuolgyYGMqu&XnB-IZZWUe>=wOkQN6wu<m{757)doc(l-I~kTj
zCzSYD<uczc>4evNk*IqW`I8AgJtW^jZPs=w(u1dwuV&wf5A|g<=C%pjt?HLLUfw#R
zxjNyyUTOmiOJ}+;<~0y39%x+DMw(rD1st!AC|`c2!<Lr`mCU;$7{a}VCwRM$*jz8=
zjB~le!iyB7+U=>`h^Tu6%uiKapck^xK-}%TNIq=~V>0(*v`8^u_8KZ+=li6ns`{Dt
z*u+imlMm}Yp!kTnOWVC{at7%ggYY#Afr{%D@X3p$1mB52RlZ7tA0fJmMqXi$#AVFi
zf!=cG<K>4dpwPI7+Sqw>#AMd>+$@9|y-rbE<Kd?pd2A6zm>0>ASCKjyVt5g$ckieO
z8Rn$*Va#uPp@+t0zG<iO437PMI(n%(SP$<2RLl!CzQvO#LmzdKv{kP(Wox#6ew2R>
zU}Y`$rjy!ySxS6(FS-<|)rqK+Hiw`!?Z}hHmm0%o<37gNTx>_mO9<+!lYp^2j^SF~
zJO_}blST*GqEU6WRmW8sBI-8>l8bITMani!rv{8w1MsO}$=pGdGeu{7mU*LIrqry_
z&WNueILPHA^{FseITD9V^gG_(z%C!hD3f5ySMo0(Z0%I2e{`7?V@PJ`#2q<)(Z~{i
zJhMfD)M;meWitg5h;!dc%)lL8ah5y#RO~5*Y!$cDH&*8Z@*LVpcf$|ueKhh%+o2MC
z9L7KLdC#<raKU@UcQf-f6coJW>j`W6%V#hWtE=Z35ZP?m4x+pm6pOCUvEDdlcUJnE
zOx#5=17(`)VY?>ymP7eagWlR<ok<FA^KU2aI4G|%31*FNL~c?aAnyH~9Bt8SdobrD
z18ji|6~qEF>vJK;%$E_&6myd-_d5O>(-daqEQ5{?+hCj9*}G;HL#Z9d(@pSu#d})!
zd?Hnloi#8_xM4}X9M~@d!DfX{f}^ESPcEDnFW+2!=DIJXydv0Hsmn&8Vn-0_H|2hg
zo92la%GJ9B&1FV0cHu*(XtiCe=Tw~XIl8M+?tZ1<+PfXwJ1aWHd1GDZ=HgX6&&xDF
zS&n#wU>}3$n{?!NCK3)s80)_46jv#fdXt>+>cIqhj5G<*WTHv0Qz}GGmw616c#?UY
zm2$()BVD^H-SGPwPhR1Rc&Upx)QkjrP@X4@lh0<uNK#fYd&Rkk-!f9h_f|7ndaF7*
zp&?~~cA~ue0E@9Efu*k34afG**-=8m0GxX<$?=TRJKa!Hj3$?s_q8IxBdtE;<AA%=
zcvDr}N1t}iOO1mR+*g|Q#n>+SSljs?K}p50LZh(voCzZv2?0L)*MLY-xfG1b;!av*
zd4)Un*u(#fgur{rFMb~HxX;mE@RX%aHe7g16#1eQ7cql=L`hjqh^MV4E`@HdJAAis
z%!|V=NSqOUj!KR7WZKy)+K-JZfa_6uIVw*gLlJlMMKf-P((^asU=Fja*R<HnlG}Q;
zqldf>OA&;f!@o+X*Fk!^@&mU(u`%^k^>^x`yg>I~e;$Dw1z+<g#w8u+eS(srixsz6
zlU&wA8Y`Z1hlK6hB<yX1o|eNaQLB>bDf_tCb1XjZP<q`0n-9u0R$P;x8FS^vGDYr~
znK)PAgQE;wk~#P^jm`$*;um(3zi@H-G~h0LCZ%e=rm~0aftJtZj-xrxKf4~rKhewR
zxhe+bC%T@t&TLPctyg_VV}&>6rQYR|wY48@iExJ|jVms8!ub(9xa_8Prh)5cnTfk(
z>H{N-xNevv_UBMGQOT5uA1aJP3Su+lsks@K$}Z7O95(HA)XWW0JxkJJQqo_C$UElr
zpJ%}F`2E0mvbiffrKH&`0H<0<i16m2(a!}zY+nYUpt2o%QG}2-qlNj|Zj6y=EbAFT
z-|Q{p-l4fk$o(rex}1JOG+_5e7-@-5!Y+M}SFP|T^x8NLH$2ygbfxw$>vE8qI-_Z`
zJ98*s@;L<vi1?s3JkSA-RFo`@;_E_Cf!;cDT;ds9mAD{z?H<)`{kEv-W}s%oV!8LS
zwVmG%BRq-r&8D5t?0zj!LZZqhAFiEUXQPmoMny-f8_9Sf&C?F+a|z;>sIS)Kde2Ut
zxG4ButK10qil>M$LnIe*5%~_AN6K*M%p*Q6B<E?kn-mQDvWHbM^+(;)c=c%Z=A+N!
z2&T<duYSRtQ?&=%>#ZjYE@$HIByLlp>|avHEOwiSg?Bu^p2Tb#^wFJ`n&(;p<-JT^
zlE`{G{ELn@TBE`BA^CWUY4~hOlG`$(NVB+*oa@5fL%<&|r1H`WIGV@PCdZV71#-Ha
z?1t>5ftY`>h%9MooZ~1-LA7kOKM(gtq8a7zxTK)F3A3x5@{vgMxX<*pCDL~Hv!8u2
zVOUmK_RMt{O={{ge}R3*`SqexrB`;%BC`p{*IY8_s55gk*lo&QIHLKbD@H=K_T7t}
zTVwCtQ|7p+87D(#R)pA0(p>UL!+5(=6iw+iO%kn!%eyY3&;(QLCI>%Vd8QRA*h{>!
zKlMq+!yCL=X5E)7RS{+e`C^OW7X4*ccg=4YgWIP0JI^b585>_!cfY!4n1|W6i?(~w
z={;qR7pu>go+VPl(<#|97Cg7(#>}(vN$Bfel(4@S<&daZ#}T34IYxwB`;&SHp+*=K
z>zuM5iF}_(|A<chhaWw!iUSqo>J{9(zw~D<VC!!6seAg}Pv-33e#Q3`xfY2Gymu9U
zBxT~-ga}!~3Va_68(t%wo)B`CMd5i(ARyM@lXADHlnP(`gYovCU++JjDMn~|1xdC~
z{`d<X=v(N1;1!J^2Ol67?k>|nc3?z^g<ty+A{GSk(PwUIf3Mj8#|Qp1^?(<|y8$z$
zv-!q}@27<aS|0U{n&Lxz*pa0aj!K>)P2;D&hdNQ}_P4$Nx|079arp7c4k@_HP_;Sq
z%LDWGLycY@;mlynHRb*<ZrP9D@HQLVJ_Qz<KL|fRUfi!d-iLuYF!96u^gr>FKO@{<
zdmw=H3cx(Lci~@tfjtOq*ZZP+*H20*|HJ(L^EP@$p=D93)Uwj-e*i$gcHMtYbnlBW
zgjGqeOn)pP?i2O|Ty0ALVwKMG75P}MU;S_b<kLKpQVai_o?z&Sn52TaDd!<ZK$2qk
z-X3|#?>5~~N!<<8yawS)B0-5~MdV+$pa0u;@Ro%?bKd8b+@BXM??`9?jA1KC()>U@
zc(xQ6B_p21htYAn%PE{9l#-G9gb{NUU)>ympy0<xQ=I#&)4W3w{?_K>V;BCK<HnSZ
z3$N*FkiNBudoa>}+84>JbfZj?O}=zRcyjX1<hJ`EEgVM>R!gO}yvol8@z2qO2cG8<
znkP$j=d(ZOIYt8?$t${{7|elU^g3p>y--(zvij1UC23`+dF@wc-kxBXY6Nx4Cw^B;
zAj?A>Iikr9Z2klT{>M;Up@s!xivPm+`_<`Wl5Eddx|5F8g-~RaZ{BuXY7(kJXbZ?N
z`v6-^7`ij#BjiEmup>K=gpyQ1OT=?b?&~$oB1}mm*W)&Cr~=_)N5}-QI4AYCCFnrv
zgHQ*uT!E-czD<RjM!NsH19PkT<3q(+rzq$Gaj0&hg0kzn;yD_C$6tv@B}d(UIS;PT
zoy~wS%!5CAu|Rq54)BO>(VsGu-|SgOK$aRPTz9UYgATiQ%Vn<%g3TRtmo`|3Zhn*L
zhRE>wH@nW-#Ks}qZCiGf3rv)1&;foP--Oq3%x_W0t^1Vh(}F-!#2w%@t4Oe;&xx?u
z+y!DCy9lszv<@fjAVe;kNN6Wk1I!_W4xW%awt+YK@i3hQ(q69pM6=k}(z*!2_a_^;
zJw?WAAgW+E?6y8K4+-WLHa!pE@ja1QyyHQ0z~j(Fnpsvuo*EwV#$DxwN{7%pHU`-m
zc<Y$STZ?9Q{~=zt1=68?TfVe6xwe5OMc%>NFGT(Pco+d=uP=1pWdU-|DYspq-%TPt
z50A(rRKe^>qgbcXS!!N9^4ti7vpkc~A^O+k{`o{TMmAh(@}2Lm@(evc4C6+^8Do2G
z9Xiv~fc$fXQr|X<ew0H>27$M_U6z_^fVM6Ye-JKkmj(;`d@G=U7J2NeIyj2;!Qr};
zZ<;(4kJ{atbN8QvHfTI1XW*)9f`kS?^-=LzK#ryhuO|b%$g&VRgKOZV2XO>dAHV`&
z#4F8A3fF`bWUK%b^Z_wVwFB$w<|i4ksnLS*aCNN9XF2V?=ug=r_fNe?L&@Y-AW=%B
z-d;NSSWt`d;HM2iJH!#_R~k-w@NHk%^~(o?&rHzM(I&4`GVQ{r>=gqRphl)_6cy?Y
zH=FpP4`o^msm>y@`u!2uh2^qTA+gM;gc-U4xyPRT?}2UpxlJJs_ZCUItpEqz^Ulj7
ze&0wB;uloDJ<jV7Ll%e)ZjAj#s6taxv5!{3L>4GHi`a9G02g}EyrT@+HFN`c@{NW4
zm?46_*$vG$OG|a7nF?WF(~Nfp>g|?;YE}rpIke|PNO5Z}hU}>TWP(Tl>0-UP5)d(c
zRZ)ku-F^hnhrQ4UC#7o=Sk<S}3I?4=>cn=lCWD;ix4hZ`xjl@e!E`5RU)QZ5jZZV8
z!;D@&#O<Z3yE6+!7%CvqrOjuBxHkHh<c6|k-M+g0yZq|UBjhI!C?BA%to#w9TwuXV
zs{!hfvIP^loVzNUq?z9bk+cO?Dal*5&|}rF87TJzzey6_MJFOv7(U=G6+IKGwZMJI
zl*JJV`WUO$>Wd|$xri4=q)oFLOC87qT5Ac{-t+ypQ(w_$ZUW!1x#$XV)5#(ERcAj`
zYmFva8*+DV7OO+FrFiF=M~`v8AF!@t!gih=#j!`#dZQ_*GzMS~(M5<t$M2%|Ezx#s
zxosE}49bY$hLLTHZkdDXuva0{*!Mjisf}u>q?!LZ8EFuX{U`zVgexRvcE|4^v_N)>
z&!iG(-W`w&-~(@RZ(UZHCECj_ztIM(LsObIlOV;kb|zTrBVdF^uDxDOr!ik|CrfN#
zh0wr}EfYH^IqZkx^+m9Io$-*xT6}h=CO=q2wM6&L&Q3i~#OJ0Wn;;kujwhRTRmDH@
z3P%gv#Yu@mj&RlLSMybo!k3(N?r~wErbM&Bp_t_TpiQFRsUhfd@n>qPF(6YiV2eMo
z(c!%ZbS+kNiq^mj^+5P8qOIYgF;pdN+%v^7+6EmZyQY<d=>l;dAKQC!g)WX4LWQxL
zD7C|!1eJ4S1%x$!jOWiD#Ixf?$N$hT&Xcgw+?2mcTkf5L>H!5Gni*D>D5cWy9XoTr
ztE&7C$=3c%_BhO{X+S4;%eT0?wN|sWKPLZ6vn5}s27l`9)><#?CV{T|-xxK4rf!h|
zc!_Ftfk@Nk=*xGZqLEN!9E9K#{G>LmgxU&Q3J+)%p3o^gpo6k+yRn5u^r{URwShD>
zr(k~cR=4qJ?mCe31o9UQfbJB8d4CVXX4<>!&AiPL)qOH*yd>H2bUi4n4}t(>Z|L?_
z45q_WSZs$c)wwmT_BB90=c4MNbCY7uX*z{|x~5|~rBDlsT3zs-j(lBWERlN}<O9ZD
zRmu7|3VUez9h#DtrWI*dX*^n$Y_lXLf0|x_FmkAWv9C5bb_f{ZVo}fe4qH{?wQ?U2
z3_fKV6slhjPV1T(xW2S=&NmkgBFZdw&!%x@KwUxznyPy%lr%{^T}qz4qOw>;nA*yH
z^g<3CR2MgH)qv=OvP;c~iPthcbSUne^0)3-zcgY@cxtFzHI+Y22tE^9J$}b*VF~4-
zlMa%BABqqn3|^{}9dWv^sb*TkNUz<SgGSt?KIeQ2qgA-G9(SkDN9Ha75!Z}wK@*V)
z^xpxyPVV3AAGtxN^PzpJDK*s^k4yBq^P!!r?l*i_sNc$4okpHY?<xi9J=IshR4t!~
z&g%A;7ED_o89ZHde`kAA44gu<k|OZK>3eoRE4ugc%voQ)%6C<JeM+U`kd=GjvvM|4
zkB>xwK7~O7Rq^Upjc<JIWYXqFdTCQ_dpsVAa}!qwmtG;hhp$X;O>qC&o6<*O!FAaz
z21>cSuVt9LjY5ekZTH1>oTFCogEi78Rsdp!;Lf!@30k4F+`M|484RQzjEdbL{vj|3
z@~uxOUWWzjTvm7PB(N>R5=jmZ;B#Ul0=-D&Yk)#J(|J+>^$F`07~C%S%ZI8AgqZfW
zxqnEdfSua^WYmz3b-x6Etpd8jD&?MB8&ZMOY)6R7d6W1;HRuv6qCl2cntY~4vH40W
znt21dcz7(<qc>?Hs$%e~MRT%g8;f-cfl)~?tJHvRq4`I5uwqI+CK*OxVx<HUH~ki8
zDJbVM;xD%Sv;-eBgWjgZuoX*6*@Jv2xqYCFXVw|muZ@|YR}jdbtgwoK-gphKZ-ICb
z#BO3&l+p8)eynXZuTEigscfomf!vJ@VWKa!mRxD&X7mGgd+ysN>juw!Jl$T11*c{c
zmSzp142f=ls!{RlygOHW;Wk^xxIrhJx-@xO4gnSm>3t@&bn3Ma7%U2(mbmXC?P!&|
zjl!|sYWn;3qiESWe+a~X7OQw-Av5YGI>Piva9a~p37Z=o?12d$#*O{een*odt6#OV
z9+43D<0EM>nAq)<Hn2-!CbdRyLM~xga7aaF8;zp6hjeDpl=c@W1K{}mtI-*V*_fKE
z?lII2L05T&I<_!=hj$}TbK5t;b{dIx-JTR#lc~Ns3#AEsiAsZTH4M@RT^9;+rcXk~
ztC|G+xLKBOLL7OhFR+{?OiDx*gZlUc*cqLV*=D6$0-m>O%sM;07yG)6y&=UPPIvaV
zuP|J7uijd_lvD~d@>uc}<m&<>*I3w=h0p^OjKN(^$)u0A>2c1lpKa2h5FLdGIJ$AX
zAk)H1cT(2fKRD*qep#6}>jCbW6bVX>%GCeYvLA6`$j|LjNdw=Hc)or#3%O*ZpWVs+
zBE6>;0K_{qSQUPJeRBXBn}5Wb6JJ6(Lbm7gNcrYlX}{Era%Uc#GYzU7v?Z}~P}>#$
z0=XGednO1jHWv!<-ur>#fgQqi(f9jIEmfy#_OdDy#esoQ!At)_3m|_{@RrVdsZYpF
zYJ(hL)}Ai38^V=XK{~h5lWRzk^f>CH)VsNj$vBFPirr$;%}gyEJ=KINN{VmQ{YwmS
z=@IFog1{#nI2PG%Y4MY6^}i|450Yw0<B{w|UkZP&CsB0#4EoZQxiS{DEAp~eJc&q1
z(a$|tSoXAN;+@aFM4MEx=2W4suc)||lzwKi@yyrFo%fF)lYjW2IR4VT%=bf8sXpn=
z>idO!7HUt#;=b-e@ZTbR-$3`UIb?8cG;^YD_?0iNrwA`1v0_O1_}M;*oTgkt4<2LN
zU)ho2HSu!D0FYdY_@3xis<L{}n<IAyVZtN8yBv`=lt=U;!8n=N5buGxe}uZlnW0_V
z-dvN5Ez8GHx70QBUqfr1p+m^wU*Mxk^FRz>B|K8+xXq3fXbU~>o7|8&=#SfP>Nm7A
z>rfNow#;6z#}N^CY1;R2XaMPJ1X}wz9Ct_KgB{lDo1JY#(osl^&ahOc-htIF=Au3X
z>X76csP#5EUZPlkrPn7(-I_(W($tAR9z^=wLnmRkmgVPA&+d|b3X-Abn`xT@n#N+8
zwDtCN=!=1UBo5FciIGb0@rRO88_HvPBk6Rc9^SGlKBh@T9_VYj^1ZCC;b){`YlT3s
zrE`##=_j?yANPb$evsb|Txk>dWB&S~UB2iulOkd%M}P`0y>`bnWiIA}y~sX!6pr#f
zk7dt2+65+w(qsMt)dx=*`W(HU>{C!W&Z89J3#2SZ!axYlS_&jDlqyd^m^f|)63P<U
zIZm;Si^u^Hjnz*bfCc~NwVUd)zPd)f#cO=#K0Hf^d9m-=w}&=63!-Ly1xC-O&Fa22
zZ%ewUFgZq%&t#aYlTVNB!(Rtx5*hREynhW$Y36Rx{KhIq$PeB_6Kes@#lE54PG3*(
zesQWy7(ukHwxYX*Ky)Wio}}dY?ZUm_nILB8Ppk#bsv^XkX`G9PWti-4R3_-2?FvCz
zQGUM17o3ZFz>^3oa4QoOn1!l$;Id35d{^Fs=he|C;N|6}fnSLc`Q1YPCqe!9U$qbb
z6mZw$ncR1*M~9nIywM)834aj#^uIQOd^et$^!OcB>Y6EfWqhRNHX?!afeDs~bT+(v
zco|+&$x{kg$z@&W+rE_92pP)m6Yhp;Z4mv;2WoFu^K?;lf|l*nfdbMu*-Qz%{UA!D
zpATGXIJ=FIdlvl<4FN81L6^$|4kP7c^Jj-P-eXejPZ_<py}Z6LxK`dp6l&n6t3+{?
zNl!Wr<u**fJa5EeudjZy)<bpU?5B9|L(Re3lti;|Td*uhKXn3ivSx7Say!(@xNu&C
zWm|H6rDSO5J(0hv@qO}CZ!}rfCxklhkLY_wh`W#s8=HU_#Z`l2(XjH~u`-uxO1b;m
z8}C%A5#7$vA!mdbrFvY7f4x0!w}Z$C@{|x8M@SrE)~YM&UiqXb#?5*ufQ^U`sn_|@
z2lGqZgdVB{jb{l%MM(XLMDms2ywu*+JR=hobbmd@!LFS~PdJTyf@$Q<qV5T91Cv1W
z+yJ3A{)N|q<)l2hSG+lP-x!sSkXg>ZdK$<r`xPhx_tk@sm{S#)wFj-pw&A(-K+6<F
zrKg5!lnMVCx$H*-bs2$rW34{xeGh8rUGW|It^L(?)q#xRO4|DNA!mulr&bweGLTJK
zZc`k$SJk=ORWdkEJG0JxvpjVRgC<ABrB)zG@<YytIA0nuDJ4}-DKk6JUzdD1>*J0~
z3cq>##T)%Beh0|Bgy1enso$N++le_u*F%?ID_x5lG9XWNDm|W%0cChBH|RyM^hfsL
zSg4pSKt(gspuKzNnXjmn4Web;IamRLyc#$#W<w+*{gp%rHE{>CR#j}#A036bS@p^&
zj=T8KoM5azlRwQ7JP1HQ$p;}Rdvx6@q><u`qnYxujI7%rR$tL@sS_wYJNH_-2H_b6
zPEfj+PoA4Qx~In109q8|)Q=xY2fZ7+JNXDZwvZWPrdR?<L($5c1Wp8TAU&34k3;Vd
zJ>18suR9O2QxY^cRXgi#ga)m-9!a;x>L<m5gUqjG#wKn{qapu|Ss<jZeYDJT71PPb
ztKZrzVf~0iVe@ntJK%!o-X;%F+~IYjv_y+hIj|Ksp7N)zZUJm7|F_prX`e|O4i=|w
zWpBo4mM*?KKElqtA&XcYXo_=@dW4pX{t3+fEPE3tbVlkD8MU@$-vfpFC-G{Haf98-
zw4@HIx{}X*=gB>pBu>^UnH_D&D?|<ogmwK^GC2PJD@!T`*6`};gR@W{x$C$ytv%;e
zf&@LKkU%UAZJ=0=Nj~she-Se_(A{Ydk|mz9`&(3p_JfJ}LI^gUjS>Ae(*;L=2l6tq
z*XF8Av8<nj$HT5`RQgl$3Fm)#{gPREr2F>io9g!`$2dF5sVb^Q0ytym*Tppihg_D<
zctZx%@>)xKvn^tqc66_)D~`sxb_^I(6Ko3Hdqh{e^|4W71M8B|7E>&3H7k!>oj2l4
ztuulIA(#_AW&8r;th9$8ii#Cu;+gsAOgaJ?)*vPrbBuA`SEyQ<P4+IZbd}B8W!T6)
zbi&9mT1yo|>9nELLfB7R^<(NCyEYoHW*CdI(ACt~D(g1dK7t$~TU*s;Zi|q8Liy3b
z!MPU>-%r0!G>JRg)(Lygr227c=2`-EUBwvT4@6JVCyTHXF`rU?b+Dx38IUkR@?zKb
z6<bMv#N+uVI8E$y4=Q)VoI67o{R$FUtlOx@0TdfQL#<#X6{)T+sU~+5YpK5J^|ZlP
z+3$`$3g3dRX*}c$*U$PJ?_E$okf@{?XiR>UHJpg*qwZ@8jOtA4{!;0R%xr&0%GS~7
z!R2JfZn1@jnR){mG2Q}=R7<S2>Hy{roMh{yqQYejoS3x2l_^8;mh9lg)K`0lES!<*
z#(wPfh$bRh^fRco9})jj*IBR1T&p<9gvZ=TlPVSF^cCvpEJsA0<|5N6Ur|{Mh8nSm
zyw=kl%qrs<nf*xlQg0;l;k1xu6>sv6eKU(Wjqg)5Co=#m$$hv8W18QnRTtPbrBvRq
z#RkUlv6D@<pTJl3i8qUWB6fK0cl+lb_qG4_*cJggIVFU7|5o}4AkpD?XzokJ$YvY%
zi!lhLryG)pPzVhCy+q?mS?SC|yQil*+B_~rA(|HHdn)SE#`C`)yuU;8lwO1$mx%uM
zYCabH_7p`53Xys7nl(BtN7jD#uRGu1CPdJ7dgJ568it|uH*cI=zqgzIvGDLeKK9f?
zuAP)e^Z4Ix!#PUvVF%#g@y|vgMjHr*s0`YsOxW*y;UExt|M^>Y<NtPVe|&!V5}Yeq
z0iT`!mMnkztrF6OLx@Shn{R!A5TCr7q>B%$;k<1>ozDBW@BNP&#g9L>P(YBf^)&hJ
z&$qk(`dghagaZGLD!%Tos>wgJGC%)OTnxB6U;LwHfAyCCV;nrMVxS(xCF61U?=$gN
zZvf)%1|>@7+r{5J^Z)A?{qG6Mg~t#?Co<wWm;PGP_#gLUSrUfW?fK`!e;eW-<Mf|D
zrZ<4yE%*-Mr-fht(tms9{~sm+-xa)J2!yG@z$WjioAkLKJaPZ8#`HpXB_!O3X*F$I
zJw<Kf@hgFPh7|!}KY=Tr6ZUVb{BQU0$3^uYkM)DV1cI9~Ss_}mMThJLst_p4z7<gI
z_<#IRWMcsxe7WZyQ@Bz<p?DBu+W*H7MGAA|r-wa$|4>gqjo5#=v?~dbrAZj<Qda!0
zRz~GK!U}gByUhACFPnet2S5B6SqWRqlgiKxm4EksdM#L2<d*$MKmPhi|M*JkNZ1&t
zPcN=a{rK#!KdzC3ePJZ%tgpte)fE1*Md{!p^c;Mda;;yiN&NHuj>y(tYMt`n-;KY#
z4&1)W!}VRM|L*<!c;RUB7n(hA=U?rMo+R>c`$!l!obCSI`$r(eAD({`Bk}K6J)b9X
z`#jmTGydiKOA!+5VBz)pUrb5<vCU!l;P!D^eVqRA@80hh0kgM;!C`OFzgxWjA6fj`
zD^qIWz5qW6f}xGbu8N&z${3)g5|L2ytpUMx&*y6~zi$up#Du$0g7-^_a}(wUv_v;l
zn)V>EWs(X#DPdUp8wu_tvYN)nrv_>uf=gLZE&r2S^JBR7&&RL@A$#Ec<(DcDf&14W
z)osL84XMSmH3EbUK!l^DD!{ijP-Ue9S#0o2_2}TSI;5UD4;8o~<_2tuS>#EBOT#&z
z44a~8z|<x9%?Q9bY}{af6*TmnoWdtUm<|c86;gVKm=j|L3X2+QD_>iAYWi2U((0dp
z)V-Jce*M`thXV72ZP&4dEd=M}bKP8?2Uv1r8p(fjfzR}L{UxFybezfSss=*BN^sW{
zK2<%mJz!PTH*=|2?KLHkFlThFb0{5S`agu7WmHsg_wR=W=>}=g0i?T2N~8n?L278w
z8A7^6kPeX?k&y0oK)MkaQV=AiON61Dd&cLv|F!OYaqlY@Yr&d0=j`9!-~HL&ZNmXk
zW{jpf8AZbgftRx3Kouxe4&VkC${d%OZ|(qxBxP2Xy5Ap~l2s$ndY1sB3DE0R6P5K#
z|9{<8v7DG3{Ge+<(W7w&8qcW%llY#18feu^{UTODQvq#Ccra<8o08SNGLN>c+W;7g
zi6d^3TYv{7LF+5`Erg!k0JE+&sOIx^pe5w4$J8IQT;`Vsyf+@SVX<lz?&%eJ=#$pm
zLk|rNbfZswi?gQb_W%Gq#U};W)xTL*X1RPe15EN<n;EjEv)&Ep22r#frc{NY(dypd
zs+Pm2=&`Nt$r6KZB|N4{iUut$^h%0<#yklMw?1fJJ<O#3H~)$G71&%3+(mgaV2c{S
zI4uG_WQcOe`2%!lgYT1Vt}U1@9;WIGdEI`JUIPC4pV@FYx+$QQAztyi3g9X=syL@#
zzOJa;AAe67q1e7%%Fj0IhX%O10cn8rYzNR2nj2!!Eigf#(w{mBGMm?Ba9IjGBj#ma
zb^$ub_-M(q-&^@W=|aWSz}W*I?FeL*VrWF)^j>21>%Jh(O5T#-KW9Hc)=oRSbBsZ6
z9L!59?jJWWm{f>h3G!(KxRLwM(XMXQz}FC_Od^%<27S=ZdJBLnLtW@A&Z7B6dng@?
zgpKuQ-l*g%5iM4{ZR0WT|0@d931F+4fzHuFB3|aI?GEs$uI4tP*>^Y^Tl)u*4$)e5
z0&TIK=@#JI;}81}71h-a7!y83D_yXgVt0zfIlxhpicDn1Yu>H@0B{#;0oHNTvD|E_
z2Y&DeFb+brMNI@CcJklxUx(vi$|8nuP8-^pQZ4+PNJuAZE^mPzZgz~oB7hr20L|p-
zY|R<8W6uEq-zW%LDf)=dyLJy&3+mMXO;FzebZf>6P>BhlSV)xxwA&p5@NPjvcH%Ie
z*mQ~^d9C|5qj%Q{?$*NkI{!}boD|UdA$n&sP`fcJ(C^BJ?V|;fRoop;UZ`q}Yydm&
z6418j1~{OJM*fqus-TlHevs^+>h6DTVwJ5$@#nx@&Wt8N{*N2MHj>&p!?Q1zev!rx
zU={)smCf(^yKf@T(YArj2Y4s}%ns*6Awmbo?ta9v4@5|7|82S_0E3+RVeu?=ul3n1
zS@HI$X9_|v$jY3Pr_gZH?*=ouPb~tHAdn(RLq+L1Cif1@<!|HlgueinmfS$B3k)v#
zII1X^U|*u`id6g2Bh(OCYz$UH45y;nMjVU_09TXEU7qOvzfPzB<5x%v8J!!(Z*)Zq
zJQFTrV1v8e_FJCo{@`nMpG}#iZLONyvk=mN9KI!>k0uH+Bn9ryX`J$DCcwS%mLZV|
zYbcazge_VI=v$b5Z|I@W3`0k*HdV%K)qX{Z)N3+HVXIix2P$vyG2AY^t;$%i-`t(L
z+->ZKv67&E=PaOgK^o_o%58vefW#PP0Deh}s_}rf`L$}W>K>`l8~h`n(%!u#qm=|F
zr@I2KkAwk_9f3JwXVi+i!lm8}0N6YLLD}-K&`<h-046@Drd-)qQ;woHptmXMas!YR
z)RaL!z*>t&&w0T4?`YY4-*38IR_i*B?>9Nt%~kuIzpfD!Jx<z8#yIrq4g0ePV=hgP
zE)C8kV4u;7-JCmMgc5AN-?wmM>{=MGtp@S|ngxv@K>;&nZ^ga?!W-ajCjzz`W^|JV
z9bj4Jh=hl}q_0O!deaf264{CFF0${oX(Atj%C!)I*-!NDs^!oYTFh#ipzrLX=>J=c
zxAO)eZiZ#B*Nrj%KkCg^LPsPaD5g{8K_9EADc;E{*{z#`bwr|vadUSUleK&pEOOBc
zxJIe05Q6=Wn*0QSG#Z}>BPnCVS^*AThep63YQBpy^cbJ9W>(Zuy>aGo?I0UT7>dbz
zUeAxfRJd-P64;1f=B}-oPVOl<h8`%403NpH-UO6r)b!yS>3=m=6v4Aaoh;d3yJ=fu
zltsoK1s21mx}ARXshGd+RnnFPj4vgg=igqV!8;(ABAE44Ysc2d7rYES6`Y@r0lCsG
zI><(lzlP!W1K=n_jV#RyhbkTLmu7xX_3mO3pbx-~B8?g?uK!`}fman|#IBxK0NVcN
z@rK>x*6#tn6f%PkA!p_g1AsJRW7|a4aK7_<E)1A-b!g|z01Vc?<TXFeoa=F!x|*PV
zAcc^NMD7npb>!e{IbyI{M*&fFrlH{mQ0G@tcPs)=440xev|ap?T;4M@fce!FsX(=S
z{6<>U*xf%nd}9QS^&vxMVC9(|qGi(*&;j5OO1K|%){JEY3O=pN#wizLgW77UAimRU
zp8*TPB7hFvxtAQ;+rjA>;P6I6D}{#O3XBNrc->>3Y4M|vpI0aj0gxGbBOF?iPGCS>
zyd#baMR(@Q0it2Y6*2Y$sH;Mv)Ic#M$qBxCr@#x!(B{|0U2n>teNziO1~MA#oq9B#
z=$t85YfW*Am~$qad@mUm8aeeX+Z`&RWs|&b%rFl?FE4p}-mpy97@9Bg0Z}`o-79`X
zlj>S;Q%`}lP{1Ak?_AX6w}__b>Sz7GB|Py=hxELkCD8B<C(9L$)JVgkrAPUi(-1wt
zfxitSQTc*4vWiheLr-@jpuYe|Y3B8Ml@6AUZji}?lpQiL=yq1n1GhJ?%w{%*fHB;^
zD5d*rS3nPA&nQWQ0o{Y|sAaiL`TF%z`-PBXXk-EvW-MKsS2pw)z5wGMvoRUVn%?Y_
zdcG*-S&MvcUT{TC-~DBni+_ft!G?sJVvtQIu9G!zMjeHS9ny5kUH@J4F**(v4fzT?
zq_xVunbQG9ify4+-E5O5ivg&U-t?aIb&l4Gxub5K;QJ_i?2&v9-dW<V#B9eib@dE7
z3uBJ{TB%^k9x*|soP~ktDWIcD76jP9T{e@kps|$ADgbu;d88(R8rPNQyyapkQJBFz
zM|(f;9&$GUsO$HoybM>W{k>D~1=&xv`(wHRtS=KxjSb{KzBu{|Fy9B~as6QJPiv-G
zeM<rH@6Vsmhqr(}$SQwRWyxrXF9;aq8jRK9SM{CRreFA6We`!oU!^~(kjDl>ZYBZG
z;b=eQATOEMtId-nbsNSc68fVk{w1}7YjEsj+@#Iq!^ph#|3>#5OU!yVbob75ZWS*u
zA8-LHvfTGrdJ7D9f4U`)N9da@Id&*H`sm)p`_gVaJ9Gy2M{LiuBQ3KJuYiWl#*s-Z
zfi<?xhnw!?<y1@dF+z-#xz@4!$gcqZY}W{60a)t`Xz@Nh;`_87TL;}4aQLsu80-ND
zb{<8zRq&SJCMnI^iM9cmjeI(PdfYTpg(K2Gi|{TX`M&ve^btX|I6z9y^-%B&&Reu&
zj;U@FFzsP3sfdTRzc|z|lnxXXFHRmbVp73iP15C>Q2^{fFAxzI71RlUdnPR&I}s4E
zgXz!+=u-^VcM@RL>aj2}74QNaBK!L-)_lUm3`0n0c}pWebYPmE9>_!2O*dwK!YjDo
z`Suh|$eM*CG-SPREQ0#%?M6W=L<lXJ^byZ=y}ulZOU4WZ8ePl+c5WZfbpLT`5LwC`
z=kwdMb{4wW7%W$h!kPQ&OZa7gHL;xf+hWE}GHW*Zywyd(>v?HcK_v+PH@83Fbr0@;
zlkRSui{?68byQrFyrIhjUT5QMgBN2F@>D=VU*%JLoaD*`WUlg!QG$_&L3P98JgujI
zo2R4C($(S%zxYtC2}@v6PusApC*Fxva3o3k7(fkb16;r#G9YVhcz&sOrjjYHy5O+6
zuYTUY5(cmVJ{jScly2mn+$SQkWM=`W7)#{R&+`$_Gz-5H-IuIHot;9LbwHyX8@qV7
zve|hxt5#rg9bcNf^8@%?nM!pNH<1chSo<Ieckp^g!o0Vdh+Q6t-PY%`fq+|V$ueL@
zdw9w<=VS>G)%-e&9ubPM6)kzcX)E!+lRcjZ&*wAos8{~=DI?(Ng(BmThS>+OWAn#$
zvS=eIUVVd&9$*9#V#n_kb=@nzhz~16fYf-sfoSa~)h)vH0w2$)D#q@q30;=|Unv)5
zpi$JCwKTAlY!dm{hjbR1EQ5}8L(Wy6e*oLc(`(G&OM<(brMo?&*5!NtxA>&KVV$7E
zet+P*t6ux4nJP~!G>A4vZa5rnl{~~<=OqkBCzvw*Ua*CrqkopgYFYWQkPbVbXnRe-
zX?2I@{MJ>622}8NHpZU^Mg*;f1`&6$^)yD_=}4_7gM}vt*322V)1(vI^cUmMk1#G^
zZ@DKBuuLeg7ugsstL%%xIm-%MQS~{E1FChyEVuTh-ASuZuKzyDu@_-JIlKk)0vX0G
z(<8eQ34M(Yrk7|iAz0smJ;<KN7Cb|r_r)TQs7PL^1D#t0OwvO7s&@ew2Z%Lg4m9^q
zKL4-`O`o2HsyagolLME{QGA;~Al|pchaFu#K+kTvcN2e%J^~_Z?hK`Xkc6YS`ue}W
zKu^9G<0PMQ+W6HMx$$yY1;n_)S7ZmD;Cz&y2JYj%mG8jBOLbY>)sy!X8JlBmp*)1Y
z$yf!&X=sHjffigGYSQ_7LTokd?cm;c!W`Q2zMse>Um7-BKTi7L@rT$BAUJMxjR33H
z1~#2P`pPOM2&OABb*7&gyLNw8NH2rv2usgW#1b+C4mzl@^xN3wB4WMyxC)&PIN_K5
zzWMOa_(ahiVKgAL=LP?p4t6Y_R`?(GZmai^QBT2zWaWCS)J&OpMx{T<{jTPg{!-r8
z%B@YIq4c4`H`ecR0ZTYF)>DF*f!;s=u`)>AI2}0;_z2I}5{6D(>@mgw2E7^a4>47w
zKWhnc{9R>&&y?Pm?YhT~b?u>5)A)SDCm1`$_yf0&c%Md#ZqQ+pJXr!YEjW2QdATvd
z(DykRFp&m<)RrcBrql{<{i~s4@h~GBieHcos}6ZVEBZhplm((X+(~Zu+GP7ezp00w
z$8#Kazv7|9^?;IHvkfZChJ&C0yyU%Wmi%=xVm=}Li!lpZOI~GUydPzVHT->=4{`RS
z7=B0b{<#X*Th<z+kV!0ADYEnJ<u))RaB#pb*s;R)4hz?fT$$ZN$rT_~BZ!f;n!v;V
z%m2hZR*6S?5>2goSC(kjc!VCj^Ggnafjgs2SDLocZLNhr>(FJ+qghQEiF~11vB>L^
z7;<8vC_TF`bb3g@rQl)1vnUpiPfj*MzVC4)vf|#`Ot5YrVX>Sm3USGEjp;T;GZX($
zAP|;(2O=BCe5XB|wC_9WZ&n_29!s)WZc^6;l;q}5loq9cp58(i=YFPlJ`Um*ob6(d
zX<DwxK5rw8;RfdURXwKq`N!hJB`PL2)hi;+?6A-jKV#SfZAfpt*P|8iU;)4xjC7kH
zhaP(aK59OalC(pNJ#;NLoa*xGwofjUn+gN9@2@X7H5jN_S;cxuvZp1-i{G8Kowat9
z8uFGLRz@a2x$7bRN#@Fn|Bf>-aserg`K~{=nBu28vu}QyZ?Z<DrOh3dTC!pCSJ}{B
ziDT~QN*#IiN%gk8R-KoR%hR(J<44)&d7U5CjnG$>T$xBD6mH2(Mv+d4_(ld%tn=F0
zEG8N~k&!H)*SGKUld_NiGFUr0GGL!=&7pcL`Bm2>-<ogwcspPQaC9G*_Wj)`5DOOD
zgW9Jnx^eX^3W|e)vV`u4z|9SCs_=nFf~vMkXMH^n9RqM-FI9d>&K4(+)>AuSiEwMl
zyhqW}yq{3Beki_5{7=ykoW*u*|4~&)8gBE2Aij+;rXaVQUC561gfw1v_Cstt4J}KL
z|FM)Lrq)Kyd2|i_B^EwtdwL*F8?%1dsU4_mEl+|>YP1zTekp7pHU`q%2?^VP$~pNX
z)~;x~``8t{toaMi|16QLz)P-ekzpu|%PnJn`#*^a!121o?lKX)ZLefqDmai`qoKHd
zy46o;C+G=YJ&m;*jAOy9k4jI=MOMsSIPtdft**zKjQx!76H}830HQ!~D7|h136{aJ
zsq`>I->&+VNO-od`t0LaSfMK&2-3#?mp7?P_ks<R4;!i&^gI)Xs}8y7O;^r30$r)D
zx`M3;*M%e_!TF|t8+0DqC%u8gd}rRoV~Kx@kUK?!v#(oBqz_!v$2x-NK;HJbQAicy
z+c~h2NJjecesA#4kU=PgP@RstJyT+z6NBJr`LluYk5ERQAfXy{d+UVCyYgq;e&dI8
zU?~v>s!^!$(9(Dq6`b0GZ+e~77>ln?qwD%h>OT09>m9_x1NGUnNx1I!M-!!44+1kS
zpO3%92jC`{Na(xb?Uo?48rh-34ta+qpj&~lSEU_j-<1y7we=Xmwe!VYpY_kaT1KjH
z#M$%;h}ZnNg_i^hTn%@nPxGLpDR=G6erPjU(+~)osxrCk?fd|Pg(*`!Jz*&JF1w^z
zI6<;-U7@LLrk?KX(-NeLE?j%dUOavFjdnDa=PVU>Rssw7<EU0g3cQ%J4r7GIB}$$0
ztsS@!D%Fo+8uojSoYK+fst=s`sZwP5nSJI6Qkol1KrFJGFPgcDH!4PMY|mct2VFMX
z(~Nwo{30*mrfN?!Av+9zQn(#3m0<B>igl$M+kn}g;ceAD``Xo-Os7y{IKjcZ<qSOK
z0cohopCK*5jec0Qu!xX}{hX$lT=r2<0YQuK+*6LcR=9UhmA0!bn0Z92`p3lS%1g?x
zzE_jHk+x{e#dHuvS!RcLEK!Hy;c_;+rg%%9vUFvo(InJq8I!%{MEORNcFwtMJ)jh;
zrsC<<V$UF)FVzY4>6B;BcyM(FFOrsAw?2(KW~+T{`n$|Z+9`rK+Bf^lj`_}pdzMp|
z*rpvN`vq(4)>)IUXrFp}V8nBjm63ql-Pcfuqyu;qh)00_U&&sbLy}9917xp*+I8g{
z>neW)Hb;xNst#{YuH+1vQkz<Kijcg*TVI*^7A?;tRv+0f*S~hrUmw55auU$5(zg_@
zR9v8&>z_<S6c?;EsxK-8qJ{1dC5oK5*Mub^b@%oHCF?$Cp&ST{KQX^fb!e?(9{q7v
zMEYzh720H?fcvzA#@TM+Gy(TMneSqvq6F+RKI=gntLjyi^9M%;p*z*=nA@Fd$v*Dy
z{`J8y3GV$E6h8u=?+UQtPlJDnK@$1ff?XMB=$5Oi^(8Y1iszj-g!J3#&TMtM!$vty
zp2#_b;m>p#Cz#F|+J|Ks?+YQgAd~%?F)o5xhy9KTn$*9%t}<%;`abGQiR6BOW28M_
z9AE*m^$su2yd5<IP<pT%ZaVzC_<QfXWfvYkb7*Qie^3HtogOH13uN`lL>D#SjN1rz
zjiX%J!+wh}4Al*1-TE#wHz6qUT3enxU4gk><OwW#8wyan+n1w8@x60A3glGm#e%!~
ziyFDgJ>KB(QbucX+=5xDI{x((2e-$O{woidyHbQP)Q``AY-JcZ@znn3yA`&7D~~$D
zINm-9yZliecF~;SNtx{-G4zW6jmVG;yh8HKebSByQtuADx9@l%FPK6M6o@YXtY2%E
zT~xV4CDUZRSO7hrh2)(?L{gOgEpcy|7FW%5Sl?HD)+56Xs;|&z*n?k-@jeqI#J*6D
z+m94M-UsgMR`Gc1_5<g70%-FOWO*gEt^+SS1oaMPGxvr+_$?yA1ZxCJ*mk`<xPOyh
zwq;U!PB(yYJ~Q&bWrHdA>0ywYm70mn5g=op`aII1*tl73hVgA+t=dzU_&i$8Bp3Mg
zv?0|U1?u_@XiA~XmCl@vJGE<Yd$!J@M__icq|Wu=yR%@M>W{&Fh*A^tf%Ip^Z}970
z0P~7S1c!4Iw$MYnIul_ixy^=vim{^+4G%Cta3nSQl<jmd;@jV|phL&lgrF;ul9jxS
zfv%_rpnNj{%KndP-x<~nOiva*+Uh62;LAF7gOGU|M<13t!wI6LPaIjeOr5t4LFtbY
zHn4yREdnTwP&&@sHTQq+ZXwOLIIFyqLo#!%eREy*dji2Ts>8e=8_ZvRir?+uBKz>c
z)}!(4XwrH1_~zVnGmC1kU3zrWB36GasEAmMmqeL`x228=D`45eEh`FzwK!C+cx9G_
zt=>&h$3dx+GRM(vqDvl9^Au7ZmlkiRa})l$zQ5p-*!vZ*O=x*65GYbZel^bAKg`<@
z%!vA?TSxPp9v4ws`|iuJ?3|;U!p2zH=ZbJ5ppJKlfxn@|KBoywXYEBw&pjYxMl{#P
z$7K*W1iwLs(V7PlLh7tni+)*zz=*WJ?V1)?eWG?Rlr%Me{tc&EJ2f1T1D6^ER=@f7
zdyzs-x)FiBd)~edl#9MIM`!g{@^*vCS%~@?9eqL1Ac%kQUc=I2!!WnVnKO@Qz?GEN
znh|XHHP4^G*OCZY`>VB<m^q&sF`7BsukKJ6#}(|QSHsx&d;C*UXwjZ74SVL%n~$)N
z0h(wiwT4`m*HQ8>v^FQW3keaP=B+sQ-s#+hT`fLx$faMh_zwQvtuB9<IQXVz=l0rd
zrv`AgZsjys1T>J>kRdtB$j#QB^GrVL65W}mNuv`)YnET~9(HjCsq|J3V*1p*RwpAj
zdoyLBTU@WcXa)>+7I?LC4arq>QLmXQHfcp6bZYn;l*Xz7hT_7jb@Hk{qLgutzJ|Em
zB=2>U1LCI*q1>)u1C%!NECB|4;$qJ?+?}TI;<R<zLoD)KP$;Y6i({~79Sru%-@~65
zA@b{0>RxLehx3duyuw9@mW2tX+ohf%B|S6ICOi~aU^>vZdoGG9%E$}`hUh*WwwpXT
zeDl8Lm$Qs9C3(2E_^+O!TxUe%PQ=4?=8JT}342avWLH#?2tsTl5HF?OifXPDDs>yB
z8}Lp!g(M5={?wR50zqC#JcJl9oO4hoby_-NT#<4<qZyqV*c>NCe6t?XM5ydM_v-W4
z5B@ovCnSHlCzBsdiil&^nT64+1;3YlIgcqR%=F(Q0$P-9v=QlksB9vyt!I%US2t}_
zvQs^|Zm3{qk6Q4rP>(sAz8OX|L*Y7C^$cmiLz8VsC?3k^MF1{5JAHQh`5)liVf+vh
z)BKK4p1{7Mm0abr6I$n->K)i3b*k;?Kg-YGSzeB3n)9PkKBxxAqpd*jglf;}^6fG8
z@cqUHU0BW27SRh%n|*}i=uN@Wz>Lu90`v^<D<_ID*?2|YE`ST}_3wz5C$YIj0D1!T
zF*luMWZ=e%drbQ>-A#_1W6XJG0Uz$+E&FYHHSab}2l$rAo&Y?9`AN;yhuIUlDAe|G
z$tYF|hZ3<S;1p04r^bQ<u7;4jmrn9W!;=#=AuV^YQe;KG{XU>u)&BDoKB_1MhHU#V
z)|pD;xFrOG-fvQM%g9YK^?bKEv92q@xERWyfrjnv7)t&LurJcf<DQHE+?=eljGVA)
zQ}pN-;>ICn8Ph60{u@i5J5=wBNEk;vg4Jn~$Vl|$Vl~h%8vzlSA>rtxS-hmv{Uv6z
zm(CY^S`i4_jOn21Gjxw$<3Qv;*qhBAP-~YnoGrBMF$xi?qTmE5!<r)Qk($;kKm=~R
zQ@Q=xD{MhR9L>2dU`OMWynn2xF}Egh_|3%WOqQl!z*<tjO})x4E=nrXD(Ldlv1Xv2
z2LfLjy?Z9EJ2p+o*~Ei;5O&CTLc8jc43Di6cD9kJJoh#Q8D0JfFy^vJ=Xi*mWJ%6T
zZccXZ6=MCHH!_j{7b&!r(G_jUw|GEuRQ|o$j^FOvoJ$kC*&4>qUCYok#>HyyKKC3|
zSco`{o&0KSa!P%{mIQ@Btao34)~CUQUYhQyz5V3%Q^$?*m8*$V2ubPSFY6XXrm2wv
z$JN9ly4Bp4Jdy_%121aiHzlNT51}4w0*CZ4X9Nhb;M{T>v$<LnHPp%slW?J^syH&?
z4DQ*Su#fd5-tTi6B;JUCo@ok%av8+9tDtU`0#0T4r(VNNi2SQgbWuSY*2Q&4rl6zD
zLwUfDsd1M5z^53c;b)`6vF~Vuc1m63|IQ*1q-(M`ewH^(C#Dycf?$IMnL_qqZ;V&d
z-M`$J7MyX18?^T`hX@IP@>j%ogypRsFTeCyzz5O)E_mxO_CQ*?^N*Cv&!iTVrrn4l
zUZ&Y2dI~Y7hr+H6n2N%(nS{~>wj|PwRE3R?Fj$Fu0Y$T)hc@iKs`_fED{Dtx1z}84
z>too95*nABx17*cNuO#78A`Z)O+QcL$_?6U?;O|!xW^P(8c)6dZ)Tilaiv4$bB_p*
zk=1TR1Ha$WaRz~2moiRPZs*k6xBn#YWMmMsu*3QCAm5wsAPmZ~N$*J>B9O>$mpSW8
zMfo`25yNj>pnQ35lGEo42A(b1;Q{TK9bCMdH`als)cn%B9vfsoS%5m-Fzv*%#W^Fl
zD#B=9rFo$ID-Le^3N~R{U#FZj4So<E^1BRfW^o4ja)W6y`Tm!^t`mISBj=sT5_un0
z2CjnLhvvkO<eQk@>Y^68vELHeilFqxw$eyBu{b=`hMUX`WD+q1a2p+qzPh^!_VU8K
z#5$*VG;!z2glh#O5YjNO!AFN;STgx8v&VK_xn9eE*Ug}1hxD{}&lgzcT8Kn@5sxmS
zTQum+#pynr?E<ST2<^9fInR(QklW<oeTi)K1pUa?T+5q)!teh(DkfapT2T{DY^sn0
z6>nM3hkEm^h}88TK#fa%XAxgijIMD{UfQR&x(l2*{<`r@IC%upN>M*rw%3n)&yPmn
z4<r6FvE3(%aaqNSX-#hQ87d15CI%3tE@_?5r9QtUIqu|gaS}O#dmj_S2(_w;wy)SI
zekTlY_SL}VAhg|Mw=7{cywkkjuOzR~b51O~@iDktng{&#a7U~kMaB2io}id{YlR6R
z%YYT$+2%yi?+dMHbQght)<(~+8Mxd!Ex++d;_>m}&2I}#Zcb~`CovVn?&TBo^|&3!
zpgY-+=11rI=Lx+DvB83U-HdpI&2$X{(%=g+1(ZC#iL04z@cSSlcj^|bBd~9O7fqNx
zGGi>BBj_kz^gitu3S62>n(pr>yv#NjZ1El)`7(r0$+O_SE}NRtb}OlFEHp1YUuvyq
zotz8s+je$f5K?dVk9$3|@KrygB-EwHD|y2`VYs;OJQrLq2jh|2r=_&7W86BD2#b(h
zCk<b#l2lL&M*N_=Z^ahBJmSyshYq*@@l*_d)Au8v8b;pmZ29|txH7&sG(LCA5Jo%J
zpZTg`OXCYby?I$ka|DLG2dzgEm%jJY(JHYf%?h>k$a1YJuZi4#-%R$^x~_Xmmj_0q
zw8YRJj_RH;-z1AAyCWY8hri1Uu#2Z=vl|ym)nK9<lI@cxKUg!5e#5ZL5>x4bOkcLc
zzK9ljv_rl_x?|9#(e%jOte>v$c>4##JNZHQ7AfItc(<C$NAES&37(_+WR2qzZvQ-g
zU~Z{3AYVb}(~6Yg&EfNP=S&aP#spyXiyG6Q%fv%MtP%m~Z_i=(CJOeFS5X9}j7NO+
zN^~)GnP)BX5mG07F@6L#=4rL_%)_#uHWlT%Kg;!T4>F!e-nh|}JO{RD%I+x9Q0n6e
zpFUxFl9dB3U|fXO4Yz131&DYtFcG%e&}-b<S5`omF+DlY5~uODch*R|a9n0@#v%CY
zR78=ST_kveS{_WQHrk;9RTe<r$tQ9I^mlHO+OzjnF<6}lT1g&DGumV(-#EgHU&N5a
zk;2^ncK5#dbiVz3zv$6EMB^$_Ix^t?F7u#lPfasikF;k1$!UN)VgI1wh+E<_c3%r<
z3?WH~c%SO<0Gn8KIu7(fjkk=xrk0k;8V?&JN(ZMa2vKuL?W3WVCdf2=T#mbA1bUK3
zPmaaJet%DNFt>~Xq)$KY_&ecoBXWG(jtnsRbUrvpF&EjgDSr||PMPMR7EMCPZj8^%
zU>iL>0V$%4n{AOq(Q~4<pn@|$M)V^8?F&@nzehK_P~?torweA`KdZ8$k5w|mRfIA4
zR{8s=$ItzAk~db*OTt)L=}^z=t-c4M&$ki$L?X$f-{hthKQ^GTy@TM^565ZWLe@Nr
zN;$5#q9cOfFi?W<FG~99XGU~O$NsWa2>w0;TyYRKAAS#?XnkH|f{9Cy*W>9oSx85Q
zMkhGq{WfUvzK6kzm1T_A`N^yy9&sPGyv1c|jAInL=_4N`{QapzqDJB*4anrn@B`Ld
zF>p@xy~ndQQxZ%LVKGErhkt~v^oVR0MrVxV0P9j&*}ZFy;6w|287z<_ZW5U_$Kvkf
zH;39793fu*Xv(HuprbaL$CAef(n?gf2{@kfenQPc0qT2BJwKragmkL)srJVccMVOR
zGdY80R42MXP`kq}+qcC`eUG*9@?T_Q`v?-$RYo_9!rz0~-#Ps#yfA*0G%9&{b$pUO
zHYgyh_`TOhG*Dr7Xk3V7XpP*aY4Th0Kx~mPO4*W}oGmFcHe19fzF=!I9*3WI=1UdC
z)WuH*u6hi}B$W;mY2t!OE3y4Y%&KWCa&)Fe?V1_y-7w1A1vhlE-$hw&usK%M&3wGY
ziH(l`hN%)8iCR3$Iy-;Iefjga1Wu_)75O%#5ddfQns!*Iv-M6gTqzRJbAYmciv{u^
z4}wW=NV*7I7LzpOR1SRm{e`O@`+sUwckSvC59FN{%_o?k*zb}I2;yZSOg^w?yceUb
z`h|e<`q;@0-yK+HSFHbfM(Z(qac3ZUXO`<nd3b>B;ivk?erN|bbFS|?ns&Gp^P%_e
z9Dn>!xBeU#&=)ZNpzh_9>==3a{TM}}ReT|q4VE-^^Yxom|AsvUNs;!R2cq*)0%4@b
z>%Iw&VIe|~CnS=_^6|jAb5UjlSgTKsYu6SuNNpB`zQsX8y&Z`1c<(lrFu=wwO7)NL
z_ZB>zi~RYdKUplF4LdlJk1QeaFZUH!0{My%Lc>B?bVw!OX@qNRuK*oAn*yqB&0R&f
zN5Cq#oV8_=Et2t&zt{iR^ydMz&ysFb?Q^^!1L~2@V^#GT6Z}B5zZHa<O$ZaYrJp<(
z>nVJz736SBF_fMhjC^QdZ@oaEB;rOmotr27B`Yt;>T~3IhcJ)by%W6DCN52K=0s+V
z{&DgbywRC<F<#{5iD@P+G)&tOtI12TS8uZFUu@_nHv?0Y%a1?zq}zE+VjwN6=PWtX
za<xOo7)A~hAB;*$tbVE&izGbcFnUZt>ZT;udCs_9U8!l}oM6v9?9Lh}a;E$3<;Ih%
z>W6%3=&+}fU#~ev`pjDoPrr)8CzR5cicRn$<!CZuoXun9e%LSkZSocNw*t~T+I=HO
z?vHw3>t9zUDfq~HlEmmaZcbF@YKMM^xy4$?M=6=PF**4II>D;G=OFT2{^q9w>h|1(
z3M5SW`2LsAIe*@=9dSp7+cfhm4skjG`_*~VY?|MS&S_|U%54EgKdZ-5#X%>ePVaD@
zO|q;Jszl|JSLW>xT#ap{f&^KTlusAf0l{~9u$-cTD3n@}e%iRe)Gv<vi^vdb!RCsR
z2v3Gdgv>3nJS4+p^-<kD*1B-3RUcFgF}|%h{xjJ>?G<khTz{<@{P5k~a!5@SDf|8@
z-n5&)4p-%e&~CErPsibMy)+XJfx4%a$D`m?ks&CYm{oK^;iE!1<)8i6OlcGSFY^TI
zbPVl@4=0X(I*meEJD0xx!7Ou~58ic6$#iC{Q-j*ShxJ*Ydh-5xB5FOyM98VwQ%=D>
zKZr2!elM^2si@TYD)b4QILqPZhaqU$@4j!f0^-$McGY4QLpo<3v4_9zC&?|?@O1L*
zMtCv*diAV|Mca1iZ*54H=6K-~YQ^2}_TpQnJDGg?-?fn0zkaLQV`*oJAJR^MYeLG9
zD(Y}gwZo#9ZKe)FbxH$B+@~)`e!UH_&-{iAi{RVbVcZfoCl;26f;x_(JV$m~p1wHi
z^C~2smoRt1x?;v<D1IgFOzFwDr!}`Tn?P=PBW;@V(Ii5!ZlD5TVXQvhZ8w+L(8{0n
z=Y4@T$j{e4R44_eFbh8hHsa8+BwmXk4AEPRU8-Kly{(<W%?CtvSW^$KO{**-WV3M9
zPC;uX^H*0s(x?paxe|z`4&o-8El><ySyAiklK+vLx>dH|+UZ7zdC;!|6G*T;PZj;}
zd1S~_@^BqqsCB+@b|F{QO7rAJxoHG(Y0FLlvfS^^d3U_@YL~fa;T8Me%2o{^izLkt
ze0=ff9~;!I(g7B3y^Xz*?Mv8>`SmUd8O8|bd*#wXvI?>fVHZh!yh>-ewQqp?$&pVa
zhE#j=Tq(Fyg;E~ZnNkj}4f<-9v;w-(oBad+J_3D%D9|2GZ;EOSXleW7>$%hoUJ|Ah
zwiEx70cVxVBTe)^4%H4=-t);21pbykc(#4c*2z9803M|$Olf0V?{XO52C`X2L=IjF
z0^;pqkjGJ`FYd-Yt>-4bQmcyjcy*#vZ-`;fDc|~j7HP0f>eQYV3<6Mn76piW9+K2~
zYw~vWlMc+PvNKUl+br={et^9-Q9@pNY7|cf(V5`QUD=P@O}X@G$nAF;9dTcsyCq&y
zzkpB1h;2p%{roZ--95#sz4jReYu>RiUHMys<1|rimTcjku<YI+eC_npY@?rk7p%`4
z%R(Mo@4Gkb`crqqQWJzlUg^nx7KQZ-y`YS{Xw2Z7Kx%a?_<)__Uq`QNtiF?PCkxJZ
zqAaxo-y1<v7$xbhxo;e&XFtf&R-E-{JI~8Ld!qL<y`UNC_08#CJVAsArH*)<oUodf
zuX13jcgY~k)MsZsEK2@U7DKkG?57#9h?~Fc<gpVw5*cP|K`K;a#^^7GN=$HqpG+ac
z!V6ZF03Wcje(H7i6dV3uM8nTBouwCbzhy|9_kvDlEi2{<b3W(r0$Yb33(?&U&~dHC
zAdI_we{9?j(=>NZFk;>dC$`~>6?h@lWzkUal-`~YB1~yz7n@1LWXB5QSjiOkfNKhg
z8*R8O*tBktniBn9p8_bQB)y-TOrCLkAk>>K?BNRoth>b_|4BX(0KFy0KXR(qb4w;g
zbb#xal3GR~>?<^VsJ7Bki#Z?9AsN#rNqT9sxnE*MF)~{6oWJlXnx>yBq5^JT2C^_?
zmfPMpRJXqGguxp65TCuY=M_llKy9<?1qOr~DU89I5>@dA%0O~$;~M^Sz-Z3U*K$`f
z^hqJK>KJRBiI`3L167k<{NN8>Y<n~>Rbol?NK|(Fck;KH86lp86l>kTbyN0m?qm^z
zsx;p`n}12075><PEm=ee4)&gJzaqY!cwFGlFR#$9d0LssyoaFr6gW&K<idL>xCl$x
zTzg?$<9YX2qUv6ZwH5`>wjxJ^N#CGb3+W$-dwjvg>oTDm`2=&fdFy6Wkm~c?LAJ#j
zKD_|;PNNirS-4`{WfnoES0>K2l4xJPhD7BY$qMUg-t>rx#J-`YHjuw1iPnpdfA^x`
z`}8mhOzuWR8}kzRGE-Onc`qlE<sip|VFRUEYxM*3*}=i5C>GyY#M0qrLKx_ShUgqS
z8&9;LmALzcR`!$7=8Awgx|<(H=a;sel>B?xDOv&gg@)R-lf#?X;Tn%thIoxd-@82s
z${)=0M624NoxyVyWmHU_kFY)v<<w_{-3Ou!hAt3gC_8|T1ovRsIlrg;?oX@GM^%2@
zB4vy1bAI(Zw@v;2)S$w_ybD*6FTjv(RwP=YBf`P}UmOMcf#Hs(>7HMRYM5g%hNH9F
zqfJIoIQti8xfRGpj2Hv5{^wHIM)sa_SEQ;K&!^;CnG=#;)H{@^glo+T<63xdFNU5<
z<x16286LN7OAvCGNDM!H`NJ*f87GIJ5KjQ#<_gBFY^0&zHcPL6^NlS#X(IdMZ^Gkb
zn8dTi_RD!5MA%A1P7$b+Try!6+?7~@ky0Fi%-P`csTk0QHqI{EOBIf7AJRI;7mq%1
zFd{R_vh7y6fb#KI^Xcqu%ne7KywPk!!su`FEtlGy#w3m<p$Ih=FFM9|q$>mEe>|DY
zE*B!-#qitB#@kflN6^=8iOL%gX|}g-&yfYkj(&BI-AlZM(bjsUiS@<%`zBD?#ld#}
zcrv&rOUiR|KG}`MQFSn4k-G)lZZ3VkD;T8A;TO>FS`mTrgzZk~V$#XNlcmCXkgtVl
zUar%5iwtR6ykK+XZBoJq`SW=j6qTLFx1PS(t>-S5Ade<bGrUMV7t`#>D#Qgcazp1c
zgP-|;@<1{dh@b=&I<Arz(Xl?ZjWKQ*6jBHa`BK^&wW<nw;;S4Sf+G+a6GPG|8iTV#
zN6dBkYk1${(04&t1o&8(iauX>a!2wjbm8*}yI<p86t!>!)XSxtmb@J67DHB0DAebb
zZoF;OarVKFjRujurWv1QiW)%hI9|v1uNLiT>=E$0V0_S6^~uuEzR!*F{>3>s--4r<
zm0%bLFp<A_LetZ^ui?qLZtoqtkl*a>7bk8;ybFEh;UU!`jvXRp_wQ*Hzgg5e^v4o<
zdHDMW0`uVU3}S%Cd|XMm%Mwg5Y}PK}!3`-UGmr+qh9mn2iL8cYz!LClyLIcSoX0JT
z2$lnlJ4zkNFHY7;8qo5Bu?{;{g?-4V2%IsLGgP1IGuP-*oFgs=*#lM&D^T9jJFz?H
ze>yW(;I~s8yubUVUR`*9$$}VCm*H}o=6!v;B9`2KZKe8Q*YUdF1|>pLy|y2>RFb#i
ziAtJU){^a6Le(|a3SD+`@{EvAu^#gOEFv-!q3g|g)y7A*U({y2$UFV(r;LT}v;7UG
z(SpBLUwM)U?8?rMrq3>00kvy-^K0{>R=mIYP=|}&jIQj23FK||srIi9W}aXZ42QW7
zwpzTsfXNpa^J#gtRZWP}2jNbxJuYzC7XEk;S9Jc93`B3}Rd*=|_yGWCG(DweIf!<W
z2%1H37hL4WDx)&}E;3ylC`d0_x)n{4KWc8GmyAKStimh~=gOao_+Fa^`O4>1WO(OO
z(l|keYmgtoq1Zi~sdeP9DcZ-EY3&G^EJqBOe<Aco_$-|#?KEb%Sm+)MR~Z+<Mvh=I
zCM?wvgs^8TugUmziqSZrGYX6Fk+F78fVFjMqZQ=Sn>3|pYvu=S#`svA=|Ygt3Cgog
z$4YczU55kIFU<6StxQK!S^7F3Ez-Y#aY@vr&6j9AbSAPW^71@#ENxRRJ!h@lej!Z2
z!sMHPV>Vf#75_n;zGw7|bM<RLynR5A>m&hIp!gxvjc7R;O3u<5MKmE%wqq5d`p()s
zGq)vrUw^aCnJG25rPS2^@YZu@T>3itOy9m=nrf&xmJ@yLN=dX+)HPnCOo;5+Ct}-I
z9SrHB_w$zi9TNAk$rzsntCii>OVH;Pn1|r7jR<{<=zfE>+V89098aLw_sz)?aRn^e
zUY&1O;wy~;Vp&-k<KCL^kSY+#d_m>Qb3ct>QXTjqR5yzO?cstpH>8A~nL8wx_&wSU
zg!!dD^)m4_7#k|lMyuZ<P^6d75KML@v`B{!{DZt}R4$S30=>7y(*;x-Y?BVLpgbZD
zV7;{0JIPt29tV@fCyAAViw)BQ%QKb*`%42guF1EMB9B;wV@q-ZqR<#SB0|e%Pkdkz
zgHnOkoL+KwqkWCmw)p28v;K3UJ<HjaJ;qg!*s`_$q`}JlO*#E!F<vv<MF0BdPc1Xt
zv=1MkXqnjgCU{oA`=*u77cObuWjA#qN=E58@|`!BYM6)UNN*NvgA@W8h>8T&Gg*QM
zB1D3Umb7kp=gb|*08@aWg~{zlZQcT1tbd~}TC>G*R+l1eWK160eU9)Dcf^2{zVWXI
zQr>Q_Z1iC>>w|yB#YyL~vo>av{57-Dr3D{kg6ujMYSx{I&(l_#`*8hOs%zzoj$IIE
z{LSZm@$4clbmD<xA_ocIJz-LmQ(jPhxjlU{+gE0=NUpXaJT}$wH8Za*AGzycE~DV#
z{D&+(F2^`5KD};W6@o+Dr5s%7`KKG0JG&39%@HL^je1uxOCY|@g6#TiX%lC+B$B}*
zj$(mQ%$?=)pLCGSt{GNSZ+7;C#c4u+Pv3m)ds(d(RQ<01%%?%48iJFvzdVw)|I?Ka
z@l-awgWdSy`juN<LW}2GC@g}ZaS0daooC~Xw0vHj#y(^Yg!6h5=QTF22qg=$rEOo%
z1;@1-H+)|N{d16e2I1dk>C>_#jota^4uZ8A)NZI1;yS0IUyXs?&S_z~nIyNXeX&af
z{V0*{WD{(e^*T2SQs!`-Zt{vq+|}}eYjHX(o|y$q<jVgu1Yewc$NvjZ45g1%ZwJ+J
zZTUx=cFP%uFqSX>LD_~<b*hB6SvFCNw@I|Z3}~QlxYBXrAW3=Jw_E>A=CnjH!A+;a
z)3~tEsEN?1g$%N1B-y}-gvE;)h;H@Ng7#*~kXGb+zZTWz3FsuTV#K*=#Hb7R^#gIH
z)d=mscy;fnp{(`W^XR!J=ra!E(3)+Z6a_Pt^Ag<(Hm2eTEiEb^f!#=K)356%oSv~!
z-0!rRB`!4myKqb2MGa$~;CMzvjlLuPgU5g*5<j3Ux7qh3KVk5M6FZAu=$)zmm+6(z
z{LsmZflc`X&-4omw4kkD#?3Al*JUYvhG(YdS#`*6s%VHMz;?DBg=G?J`F@_c$F}L1
zgw+Wq>jDJ8EjKInAJ?l4OHLVZ?h8n$dnZB?<&{ccD_cQEVoFC=p2q}z9a{}J7~@gi
zO8nJMtkmzpgYnO*`O+RmY1d=dF^)^BW_>ZYr=*Fq@oZb0gM-HoWvza4&TE=3FhUEg
zt4SQHF}8ubrB}2Q{HG<;iD!mkJw!VTdnB>nk?_djXGJKn2aYLZWv>hv?U7GEmq?HN
zx!OkOwUTry4fEMO@Vkn?R$Z*^6dcWNQMgDtt@nJ_DyKQWx>K89;h6NOljfu7Al{+0
z9p>KiTj<yF2qtjAX43VCiC@@d?F{?M@~p&jInZi?rvjwQ1k-?}9WLc14g;05Zwv7n
z?gS~4AUg^sw=iY_CV>i1d<z1`G>FADc5kWWG!jQqKYG4|AjF(v+sL-l4(GwAEyxG$
zN&N%{uU>E_6#kLPk0ibHNwzX5)IzmrWRD<%2*uxSJ1v1JLbdPt1gBkcpK``%sf);4
z8($KK5~PHhI1z}CZ=4(ePRaM7Q(vdq&b8&VGkVKOaOUC%b~z6T)(Q6B1G>?R@ha(T
z&z)+ZSpuXcs&X$QEL+kLoG{`(QP<dkv3`lyOel&w+v_mtc=4CxQ^R?P6&XRV9v`B3
z@kdTc2>V67gp8cR`uP#dqW%8{b2>abtxw_)Y)ZuwH*bXce$qGCYkflC+e{|A9XCtS
zW=3JP>TF2vZF|LwBTViH5jg@gn0a1lr=Iaatey(1$oiN;#=!fR0+SBDfW8RbE$fg~
zo?=wF^{v;`Qu!PBa;IvGE~zM6dnqR~&m2Zr^UlPO$ey@#p&&4O)!gwl#XizEowe^K
zmt+w@ZUeM>dSc}r^a3IYG~byB9g99^#Cl|*8T_`5De&p#OW*!RQa9C%#=&uKuuhCS
zX7BE{0=&9s6E}Jj_on%W{|8bHyn5Ky5YSZcNI4S-l3-I8<jG!CuN&C#X*?7;eQ{xn
zEXu5VpBZw&dO3-BTd`2Ud=@M8A+^8BhB3xN#aH9gZl4gpU24b63D+Xgkfnt}SRXI<
z3j&ta*Bt9x?uiQY4I0ar=p2^VW*TAWE{Mgx7-7r%Nr-ar%^XJmrpvabAc;zLl8rCW
zP$&$3{A=K|&}ejgJ%n&2Fg8Ny*m^14Ewv!^<>026)x-e&S$9<t(WuPIM<J69)2Jok
z7lsL`=5zn4SY3s{b~4x(je)X1|JLbbypv8s;cpOfbpt0UZ<ifr2PEdElz;ciXXoF6
zApsi0O%KTybl*Ux4jI&=93#d?H|cw&N>cF<Qi@ljG<H`$PQL<-GfeUPN~f=tAdc;w
z-CoxR2yC05^%q=KjtM`)ZL{K5Io6v3flp^2eaZR8ia%V)!!&IDh`YXy>ot7nPyc_9
z5&3$<W6STa_ZVT+gXEWC$l*a^i4byH(1c`e7q8cvjMW3>9;6CZR_BQYW0IX#pQ`$o
zevn|&H_<kDhmGr}5h>!}x(}<l%5@vd4X#;zOqln2utaMQ<8{Q7HZDl$Q|MQez$d`3
zB9sxTxX<5dwW;;{#*fnv$w)sWVixR7?h*<a>LDh60xrWBz(O752tz(N0^`qB(B~W7
zZx29~T79#nblC?X+$jX_)oYTvvANj$CAmi5Qd{e)WJd}SU@eE>#Cd{K$fIIy3zCEX
zi`dmXh5&<@{Ec63l?0P%b12ntr??HIvS~Z|4>G1&U4a+q&rZpXJRZX3<jNoyUc-}p
zBv(T@iDmz_`j}z|v}4>QJEUSW`L%^K7o%(R?E_486-A&Nh*v8y8MzmkSL|ORQp|x$
z)5WzS>m4Uw6Y#`B0MHCjOa}dsXS0CzmP~J><d(bLRD6i2#`~0J3^v#`XBF_<AW;f#
zD^VE1JxoMxhEglU7+9X)DxTd#Tu#ZPhY>oMQv3FYj!-NQG*fLqQVkf>;kY~C&t!Cp
z^?J0Q(glC-Xz56ClY65)cds9lYGjpe4DuM%uUKrdb+jKzN0*Zje@bPCO=zDEOlml(
zlTYJcq~kvbW;?f4tVPyHhDWK)M)h!BB%Rq;f9O=m(l92IUp`z{Gf(ZofI+28)?5a$
zv*Hfo4}`xD<q&O4_e}#Muq%7Lwcguj#*FECahhfv6zsCeCe`^+>&L0y4j&PNoso^X
z0-bNwsL1p!^mK$w)X&SnEBb5hT{Lrlh~kqZoT~ay#q4mB297n!D1nss&~c|tdEq=K
z_JI!F^SmVr`j1sl2=g_2KAWZO>(ao{32*f=18z?szV#8mCGH0SZkEWimOK{=svSTP
zJRIHHlv&<uAkoybm_my(c{i>+i`}*aH+L(K6IVQwH+eB_1#6L019_7sS|9_p2KTT|
zBghrRxp~*)-}J9q3<#!sSraS7aheMoo#BdK#m|EG*tewDruE8<kLwn;#)9T}LwdSl
z3fl}&U|tzkwY_PZIZ#Hm-?_pN_Tb_oL}x$bNpuZr>-Xpyiw*4_qdRe|a<!oQGfc&-
zQFfiV#UGpW$oqA8QEhDO_h@EC;w?6eJ9O_zgf!iFOMF^QV%+j@kYA{1vkIv+PB>J!
z!@n*qWiIArf&^8Y9uaz1C3V-d`=J-wQtb*!MBLSn(o;Ha3OXv$t6T^Z#rw(~a=~^=
z<vjyPFOMQwjmIL3dH1CY2r3JpTWa&O30#~ksX^@`B$`Wa+qCaZkNW3xy-v5WwMh3%
zcwdBnXZ%*f2l|5iE`Mp*+O_YiWR`cy5bQD_pXss^S@!869}A(f;%GV(Vdj9%h3_*O
z<)jFtG(3j(Xqo3qCvf<rx5=ItAA~N`Zx`-`uABgvu^w-qB<^1EqyBT2)um7xP~oEn
zb)Ezx)DVjcCrE3&&n7k%u?x6AudkL~IGh~@$`XKlU1;j6Az>FXeU4mJbXTFYtGoH|
zm3b|J7Qa@+TtHSXQcLfy=S&$CsYgs?0V&7M<C)Yr9xmOOS!uP)ZbhJWlD%aW?)AYL
zUTqpb@B7SZ<LG8lvv49Ufgp%&1d%Zk^-Q?#0=$h@NLXG-|8F&c_^3uM*4Q&4I_*57
z7zQnS;ixBg7(8^5q+CQ7-Kam3Tud)OYQg8W)O3LVy1G^LlCOg)jt^GN_qQ7D0o@<`
z$?o)K)|r?@$~g+WwC>rohIsewP$UHlesQ-XdeQ7)kjQ$RU%MvXD9HASMKJ12<d?>B
zFK4`lGcGeemJ!Bse;9oz2$-;{lvrDBy`=g#$f?EpQ#Bc$oT=LXRs@qDw}lSGBaQ3V
zOuny2IwmRLKAD|-u^?%rAPlltr&F}54jU26rCW@t%~rTSky%dtTVTKYt$b2%Sj?sn
zq)X<Ri`V_@vU{cy{Z}vpk5%u6>1PQB)53|vtbA@nc-U%@Nsf4IJNtzgx^i0nai=u5
zZxlkEj4}VFa=-{yl06=qhF?Xh&*IdVlwNBjR_^rSJ!YO@->yvxb&*lBExL-?Tu!AO
zWJTr%Mh|6kz|@s^QA84qd&t~g-C2|R;Yi&>Jobl&I6=R?4Vk@3@Ag?PC=~D+0(?nO
zPnuY@g>uMLvm(6)nj4?<Itd&GET-T{$fnhXxRU99jzOuubm8o}YJah_&7o8Lfq#of
zg?6rbVXf?1ZjBh1II<yKVx_~QrSgx6ojj+URm$gUozEY$A|+N50$!%54|PrL)&w1G
zv0jp#|8fa9SUff)^B55p;@Xxg)ZVe5z;Q})Si4HHYf%HTGoMQW<+kgn+qEIi_{6^R
zA(>gQofDCfV3~o%diIZfZ{uTw2vm#ox?war8JiVoxDM2cttnHKQ4dsMdOBAyc#@%R
zrO{lS1p46j`Iru;#c+Ad?H`@VhpB}W`Wrjd!U1Jtv19P$BOw=iCqBz=+~i=z6hoS8
zV+I#LLX#;CJq!s6eGdP^Nk}?rOmF1Fn4j>b_0Gpkv!lGZR`V@`HIDD!i{i)Na)V-x
zK--&j2T9*>!!+w_svTK2^`n<HZ!-@6w-<oz3Pp@_q(-#SP4sjvE0r7OjUr9`nmK3U
zb10TK$H!b2;aD1wL4KL+kMq|5l+d3QADB(C6lO-c-5a5$MfEt49&sk2l0)%dkcU2(
zngGjq*33|l^u9JI+zV2XG6Vo2$#)EcIDMxx1i6*FKmna1#{0(3jC@&m6+G#b%!4!)
z9{HXty1@eMx|FpHOADTUzAo;GXZPwCove*sEVZ7x!RcXyq@?r&gUySY^1?g4b-Fs`
z^g(vNFb7{W+Mo(DF0=WZo3vXrubo+GP1j6~+q(<Me{>lTBq0VVVQ1+^MU<`&$xBdr
zDVpiHzgn<UhYpnt2~BY|k}FWaMB1=(%O^k<D5dKaWvJX<|4L`D5jDnBVeMZr=bej7
z-u+J*J*7~s1bSjV=qwdWZT&gKoWJ_PQ<-lsT6L;mz%^%D?tjG(y8xcikzKZUYNo*d
zMc7+MRT*uI+jK~~k&p)6fOI1uAuS~#Aq}ztVFM!FB`Mt{NC--Yba$s9Aq~>fUB9)x
z=bU@bz4!a>82q7oz^?aQYt1#+oX>M_40qQP8!5~o@VJ^!dG+6534bX~Pq3!YP~MIV
zU^ucAj~F9vgMa<VskA^7xpjFzC4Z<MC|u`%))HWm9ForixpCE<V9nr7ug0z%U`2xQ
zkETzP6f3QViq%Y#5W!$v-&~&b+E(E28S{q8#NQ@6<_rW7v8}5&1YvV?LrTqHd>j>>
zOl-qFG^OZ?$swGFLDyI4q<T@)CY=LlsL4tZ-_zE8$k}^>c`osRj=T}NCQ;#(n#}aL
zLcVdJL-cm8Ft6GZ&GR9|>kZB2NRUXwE7H`@NGih6Cg#QtR$N%sE3?g~x^XGVd~E|H
z9d7`ZS8<c?!k?cxE!l;|+oGc>R#vf8=R$<%&*FilE>g8#EWQ4Bm$#U5ra7OLE&1jt
zt_q+eLjx)LFuzcjZwHoPJlv2*S5ZNLZlmZfXi5sE<ypba^?EhOGt<p%ZXIn3o7`y>
zVZkAA{VC<0jNQ#$0ZSlimQLh`&q_6SKXx^Fl&*7daQLGzKr=xrAjsv*b!qy&8H^He
z()4QbrNGp4EtO`dMX$9i1wlaw5)Y4fOtN@?|3ipHWrO55`&p;fUd8>{W$#;`kb!5#
zN-i;CI}7lu{Q;71=2h<E{k1MWHKn^_JcniE#UViy8F9UFIM8`Z%6{*%^a*ZAZ2V;Q
z#rE=<S05tB6WB?T1B2#-hzWa5!qt;<e+`i6nRg^(XQh0SO%_jni^G&Aun97Rj<wHH
zTq_|ohXf-1dZGN(YjxPxjY=l;P37IZ5y>sgJY4TIH4W?{L|`V%T_>hX0Q=)_?^8vd
z@e1Fds^*z40CpF!llL4L%xBvqCx5#|M%#6RqzNRAOy=DvCEL`~n?XNWrEULK_{Uw#
zrO=YhM%}O2leDBYO5}COWSP<YO;q_n^oDvCtu&7AzhV>q@*3!0O3>mD$RlYZ%LV;c
zDiy}_mO>;m7_u-=m_FO{Cmn8z`6=k;8Kumr1C{V54?hA4+&0t>xU0FQpGP!K>}POI
zRbaY}zf~A7BKA93NnObXg>=5=hgc=Okwvo5EYba>d-jY_bKgtxJOhY~Zif_w*3)&K
z4O@qm3=}ffzU1_d*n?dN@@$hO>s;EEebmW$tLJC9FJL9%^+}K?%;}aL7GL#Fe~SaF
z(mG6&RAidOkp2Gdqlcy@!>qDyb@UA|&sxj!Vb*%{s>g<>pdVXI7x)b;fL<kwcM4&6
z;He1Mct^qEQ?a`hJFS_XtkZWWC}1U$OV2V!|89^bwoUQ0YvJI1R2FkZ(#Ei)p4DvV
zrSqZWd~2JN*8Z7zfX09r^Vow;h9=u<-tAAfiWP5vI24D<?v#MynW~{V)*Q1$J^PTb
z;DP5BffhmXZ~N{ohgZ(`dV-GBA;Mlt0Ov{85!-gSNP4OwlH4ggyjLYGqdv3J*nD@h
zXAVmrJYv3R9GFm=jT_`*@5NEfCtZm-@@J)B%3vxPc%m7IRm^nn@kf{{KSDT?kc>V(
zP&E#hP(4FT!C1t`J@p7I1i34zY@HNRV?&6x+-yMghFbh=6<bM|Os`evwWJ9Bx6-2E
zXMnF`!bL`{i6)TKh(NdSu&62hq-|kjCp4cuJ<!oH$YXsTplmRdDBZwdRs~<!aI|>!
ztPgcNp&opEM~wlR0|?z9pyiC%H)unT@JqkWelfC+bWll4;^Qb;M>yPZRR!Q^v&f9I
z0Mo>cmIq?MN#^SA7u1gpW@?>K#Qzk`cC|->UJ0>^IY{G<?0%ezaOEgq#BdmYW@Tni
zX(c$s43pNTHJ<e_zi^_z9UaQH;oG*=6D8CV@a{PEfy;hMUaLuP!@y@H{ycXAfa%V(
zdp>!N3;u_V%E`%Dx1Ymh$;}dCQ_*}HXu3DERY9Q`_XksHt?KZ^tOf?E@oEV)xmua4
z2oYI*>C~P=Kli!5=R9C)=MrpamNmmMQV{<K9SV^{`HvsYFW6I|y+-LK3-^uzwCuR(
z(Q?9S>Wn6~H4nqT4=I28mdQZtakGRUemVz7Yy?Nc+@Hlctt6Y$>y$*$D`Mir2AA(z
zs_dXh2W;t6ZzJGi)-k+OIcf;DzX!nY6Tz&0A%yGI5SzoCE`q9uD^ydH`@m*E0E}Yk
z&HhRW=&Ae88G|#*oCRh}02FXiRjng_4;IfhC8e>_3sUe!a|0%E?-5)N6mRSBUjBX9
zAj8y~65G@+O?zmPgH1Tn^#bW7hBdZ@3EskG7#z(QTeixaw>8@K#wshrD7K)LH3S_J
zab8s*-K=)QUl8cKp3K){q}d>%mwwGpC(pV4__oOoR{x?3#=|ACcr#5M;cPO?Srs3s
zI!a^+6~sp5#db$sL=UVXtF6!le;qBH<Ifuc(SNMRo`H6ZMpw|(@2B$_ho5<~c{(({
z=-IRv>)uq;PzSqj(GzuH$P!vb{+h1jtkk<Hz<YXzo1;o;8Buw4z7lywGGo1MO|P)t
zPm(@5p`=yHt@`|UNK7esyE{1h`ZR9e$jWm0$=)(a%ef<<fL&EpSx*mWW<y`^A1hJb
z_o~9%3uK%vm<QB_+a%DZE*doy*h*xW!<f2=4<$J|9FQt+g&rb^a-O<+6%W@p`=nZX
z$N4`_1t#y#yyIeDlD%G|(rrsQXY^*;bnYu)h!o)-Na5@6r-i7WpRPd%_Sqb#HKW*~
z0Q_V^JXz6esGfK)I;)VY^#!nzx1!bFiUQ(S5@xK2!nAa)0=X#(?5Ne&Lq*j_T{&e@
zC`sC2Vc8RC#ZE61t~*aVpwu`}COl|9HjJ$@mmSIpeR=5ip)S$BD<I6bQckCTGG=uS
zXq`v;-$Tv>+Gv_z8pIe32xV5|<a_7WPnAbIKk_7^p1EgsV)4zsTDgwi<HuP}YXifu
zT5<V})BJ-Yxo`aF!$!i2Gb+}5in=YV9B#G4Ji&%lF>4o4zf*Yp(b9{9XVSILG!HEp
zsduZ8e$Rutu|Ah7r#u@*+mVQz;%w?ik)u^yp9^i5>N-huU8r?{{O`4&a$`$X4HN}8
zF8N1wj#eMt*ge<yYk6Vj?**x1SOjK5d&&>LQ0Ikr3BMG5c{)XpM<5L6-*TF%|EeE)
zMyu%9KTvl<u&dFhHHNyYwRO}WJY1EYr_^=z2O#E}9@v%LspW+ymNzimj6oUa)DlMw
zb385m+z6W4Uf+X4&zg`OS#FeTmk^v&yY7tV{|!M5c+<F8+(ydg)UudKA&^^Z_ZXvD
zW4w*tL(P99dOlC$G7Stv9%u^)doggS0wF;oz}gnH)otNV4--r3F&-N)6sbW9tYU+B
zD<U0)5E7`OF0<UKDPPhmbN1gCZ6$7R4~T`oB3<rafy4y2*YMZ<6F7g6=l#WnK*~!o
zDRm;u$B4UX?|_1iCd@O5Q#y^P{U<TjmG4M9E3|F|?|}XtiYzpRNSjVuc)tZzAGf;^
zYjHyuw~vc1Tn>2&3(}1BqV8Sr2_Y{1?Hd!$&`vp0Pzc!*5|m}Ze5oY#obzqMV}Bh%
zORTR(J;JGctfU0?R3??cUz1H2D)A_|2Jg=<r5?P1m-Mc(zBuE3@YQ0x^F__`C~*<)
zi&ST<jl*S)p$U1dV6UEbk&;+{Ojk?^-Fa!c5S`3}_gD^u^>o*MTqYVMB?n$GC0=i<
zHD|PF;1aAzWWy>By(l~E;<EGOMmHu{%hPFkUllAVw}K_qvk?5gps!BIGWDkzu-W;+
z)n}unL>nC&Y9evu)Po5KB~w1siMvXB$;bsUQ2njQf1Q2JRHw10T3tw|7d%Jg{Fmdf
z^U)YlzN@!yN}|wM9MT<@;n)xF6G9s+mE55k$G-?$f#FT`8}&WwkT8p|`WiM!pBL66
z4~~Cl9vK#cN(6Q(17P**vpF-xdP(Rpn-I>fE!iM(@ixJ-u_2J?1X8RW`Q?f8b`FcS
z>H=iQhVr?6MfpSxuURUJ3UfV?joJ{yeI6X0R@Q|}6lAd1C9lXAdWme5Ztu>zHY)Qd
z-edC-(rQAI*jDAM5XnQ*EVnW3fjrps@NC`cAkYd7blb1c$C=m>UDC@n0$kFw=G3(f
znX5pLh#JCC$}kkQ`nn7M{g-Fz=D;$dcvaR-ePmD5bglY__7YH~7O1L#52dR<*CUa5
zPr}6{C0`vj`PEySmzeZ9ZN5}PnsZ*dQwmP?QFkb)mghIvI>i6PIxle@Nt?z`wCh<h
zwR=N4_6*0n1_1l7>o>3L5R&PdWrfcV{>zpH>|My{P;mcnzf@eW-Y9)ZRD?`2yjlHk
zgvcD)mYm1*)KW0T5n`(6q=ln&TSm`I(oMOE3p5l1jn{m8RiO1mb#&DE_ual9Rg6Ck
zKoNmNKX(XDA{1OL>f<7cCq*(L|N0_d14M>3(^b`?7w9-`$f(*zX5TFI*#d>{LX7am
z2=R=_VILl7+F7w^YrPr6eJ4XaJvAApucOz0wb1W8`)3ecN&7`vQKMV!gmfm*O*<o8
zCv!shmOi5Ro>bZn_KJ@fXdi{sqpaQ}8!;BOA<OK`WcICEI_N3XF>RZ7TAjbKk8wy>
zVYxVRY>LG!#fo#zLGG|0<$5<{I8XGcykX3?&JP_W$_CA6p9zAw2J}9^vKSO=yrA~j
zkc6&Mw~CWVU4f3&(Y><U-g?vY=)>q^Y}CUn`0i##dfS5SOqXFHz9WNO*00Sk4V%Q1
z%IolYorMXX<B0EdNy9|tQ5^ka@3+oqMYrPkIv$&`X)fnWW@XQ3-ZK53a}mAM$@?Y4
z8R3#{Zu%}t@Cz=TiW=;tIbFveXo>0<+AfA)=lwTv={bYnYI>S6H_2JN%c$3<I~W|@
zn)FyEv%mJwd5u5VjIwiDDOF4(eT$c+MAR<f8K<IK*&Lk8pg0m{R{Mipx;DMM@N25<
z>_KY^^+krhrciAiT;y7IQA(7qh_2Mk7jsu{tJ{WkF{X;BC02{sSq}4;EY<VcX(iF!
z<4>>Fw^t)K7`<LX=`nKHgP3p7J3cn_19i@J#=)h8#}}7Kvf?^3ShJ<@Bu9XMCtvW|
zVML|GT%5DnnoTpjK_bDK^D{i%*06#zsFRD8q=#m>IiraH0l^PqEGAa_&u#R-_S66E
zA9m4nn!U+WouNdrzw9BA;_*>$mZL`~5Qt;(hbY!pAs_x?(AZ{8^IY1bn5&6uO{mCO
zDuHdl_@2X|B$^a07c$zgc9j%Uq>rY}If4;<{VreNzrD);{BM#t8bUr=ed`Rgzn4yt
zkd29xZ%<Jm(sD8GW}b-A<m*c)UfOk9f>3AvQd>e~JxQq5#jG0=XSLHh`VOT@^;cOD
zJkIaxYK-(3j3|8AMxWD*Mi6?e1tB+(%Rnu2dCikl*GK@s&8Nsp)&Cu4`#;F_!7Xxx
zM(%B3RvE7T9XebOU`K6A0&~RpV%PgEnjykFrT(_+B*WY9Js?KHofI}Tt<{=RXnh@Q
zx`HeGfJj$e!`p<tv5UxWWO}1mOJ4)oaCI5F@T^I>{?<owo;HB|qreA3$5PNWaB}Ap
zBiwCTt3N4fyscj)+X&o!jiN%pljxy4bN`XhBhV`H9PNur1~s+IqHou@0RUFnfqtVP
zP_rmx-Eh#g1pLUSMa2}P8~q~)vbaUIGV=dP1}ED?GjLst|I0c>c%}><ls!){Gmd!r
z8%u$;OiB<SC))ty@|R!<$4Y7o2q8%$L=JnzXIch<wA&b@@Q<1i-d5k;&;OvJ^?`5Q
z21wv=5vF59anCw{hOdcoD=4bE2Y{1~NkA3WXZz=9!-ciFe}-{U2YB$>Ua%Aq<Ae5S
zUv1SLo_h|dSc=*xSl8Mj#8mJuAW;=5tlRo#T+(=YFP2dyBJcR=DgqsSh^4XrUpLwR
zGm8Ws=@Re-a4P}umKkz6o&ritAUM6QJ1;m!p`vsKphJ7mm55Qu@n8-F|8u1n+9G*>
zD|cv7=xDR<K<^9*AY|2IM1XsXmlYIdqq2x~Th)e}wIkGqt<fK_0dG|!iPusM;a{WZ
zK9L6WsAH@s+1`RtIy(d4taz3&1X!;JrJ~))U<`IkJlcBjzrN4^^Fb&z0(WhYwjKU|
ze&+x6xw`(B*EcG>pY-(o7fkr!YOg6@@V!LDiu8ys^N=08i)*u&rKI6LwcEz$Qh);u
z1ddq~#lUxkm@h6&3%jG*{(YnNpPM^GL(>4_^8B7s2hzR2aUhUr`~r3~gr>*hUMGMU
zQceOR;uq6hublnt^gFZla#kD!;?M<fU2l$o-Wu{nF<H&&YsAE(Qm{II09p|?58x3h
z*^IRTH?)Q9BCsu+Cy9Exxd5Bgb@>BLZMwpWA)#HH(b^6)jrqqILV*oeKub=cEVP+Z
zdV9GY1dw^|xF4)t{IagOMBe2_h5pa6{?A8nX%Mp|#}M(qR98G`hyjmKO!KxKE&3A#
zK0BRapag)eaMScA@Dav&Lf46McAM{R0W9xa(gifqw*hb(wequ*AIoNt!!v{_UQ@>`
zzI5a{&STd*K;B}cRdfT|tJrf!6>z!wup7W&H%5?*3B;~T?{-Vu@t5+mBc3#C%VItk
zUz-yzjlfpn00iPv5b*>6_!qMt82zU7in}NWVAQ)^!HT|5zIC1s{-l0Oruysuk3QXk
z#(*GphuJk2|04j&dE|2<TDUBA_Ep96iP6$p2Snv#kdW7x{p=OVd!$Cl`@p$|g^<~|
zgW!n*K?vSY-}A_`o9MW@IBULpwgdJOOTi2gSYRv}+()wYSfOYS<7gtg1?1t!2P&Vy
zO%(?bx42rNxf2ut{-q^={l>!rsnOla!!SqSu+*5xL)%W;k8Ai2WE5^qLI4!fEtNEb
zz6_oi2SM@f)*uIXQTDF?IA#1_?{Dzn7Kpt{sm12l{+VcqV%QH9XrS8x!02b#Nbjq!
zeiC~T-&miy19B@KOHpMEnh(Y%u(zAK06i@wAb`*WfGbr+RaHEk4pepQNKzUIG%6=I
z+SKhC0%)6b|L)^JDxd>v%!jkzNr<L*;%ncwXU&7gSFs-1#|Kxy0BKhP^4x8J36b8>
zO#F$-J13J#?EmW>BOcF8104D_JqL|{ZkaL(P?$U1nXfk<Z`?<^4I|N|F9A4Nd32w<
zL;VHt7c!yX4;Z9h4+Bhxyb;2Z)qFWo(iMkfXC0%K|DP{K%jOFRkR?l-c^UZsB)5*a
zu_K=1Uq3P*d}ds-{qzZT{lSxZijV(OKg6kth0uo~k4}Pp`F(27<I#QK%?LuwgAL$G
z2@pJiYk}FiVlXquOLT3`0%N%qO=gD}22-N8Nl5AM6stPJc!7j8IlB4n34*Fd4+I?^
zn54;W4-=cw=FcXy4Dm|b=eFAr7Ac<=TVHe%ChXliu{csBQ2;d?iY@ZZ0Nzz73AmG7
za%m2vnpy7vr7r+1lx+L#HPf(EHpH4|3`B7RY4t$9+y*iYH&<#3Y!CQ>dl&%-@9gFm
zb#*Rgc(%`;69mRw_e%76N1b7E`*T7nviVdtpGeWmjr|DchU~R5fE8tG0Bq)y{qJel
z&8Pv$1(Kjc&l)t{--=$W>^3#;x8W=Slx1#nKVr@P4Fr-Zi3oB%0&TC*-tr9i1MA)E
z!EXftVC`&mKGZ;tLJ(ugHAp@dda);QM?tVRYFqbWuu9og_F-mPunkD55L_=Y2@~~+
zocvHX22<ZzIRTw@{5BsjiJmpv=7?QC!Yc-C)NJ#yLtx<=>UwvGw<c)(qz>V{X5XP3
zu+u?Ek-&7hGO1YG1P;!)xf57cJ-uSh6ZGumIe5N8dKf|IVtX(ezHpNR7wq0``D#`A
zCBYd8qI}jbTt0f#{j%@x%m^RWFoRN0%4<3H)yJ((0LC>5!q@ELdIqv{0e~pqKaRGx
z_N1V2*)qc+Ec*|w{vZ$}-qsI^ZOMP4n2TdpZ?J9x7Df@n=xVzk+c~OZ8WZnjc_O>^
zyH&P6flI@FYkiSpu%<uzbjx^4^N`7v;?L^Mgr@G`5#xym;~B_Bf!Z^#15bbqKRu-a
zfVD-S5y0UnF`)=*pk%wHdOyH5|LN;Yagv?Gc=0B2mlFt*30T<=hXnhu$MDt|9G_aI
z@l%`41D^hA)tW~A{^sA-F<c(xK7=Cm=hc2?SbnNSwW=t9DZ4J3U2|3_5S;>A#b`Iq
zp;r;O(G57(BVT#u;W>^g34CDfPhki57yQZW3nkLi3;%VHfO;jW>_hq^H+_cJQ}pfg
z1t)~B9^9FNgc-U3IvfC#5c|N@58@}+)FghYwFy5Q_Kg*|+08OG9_vAW9+Q>YdGz`B
z)pRV}fR@sDErb^^at+m=E3y3DwXaW19vmCDeBjNz#pUtl5`xUi$$452psSU5k;-2l
zXYw1o{&WbpL#(Qlx)-efnBGtyi<{m72B*O};Px;3-Ge5rEQ*;#{M8F{{^j=nY&!j(
zP&#5+b3CTdxI)PIktmI`_2bmrSS5aJUHJF-Fp1YT#eyRRE+qr01~)aATLtZp6ZP)@
zX}<IF*24I1d<6VEZkS2PKO668+}Dw9<W-Ay58aT7RHYhx{BZs5KZs6y6C5K!*z4zm
zCSf17h$~^QhRc~l=kNC>is`98t;E>G(u?~VQs~lBKmr&u<Gm+3C+u+c>FN6Z#8KLK
zqp*fnqyE=|rq?GYC5b}~$|x-evTGu;OifFR82tk&Vh3gR??2R<9{tG*V}DKjhV;*b
z@dbkUj96l*z2@omA_?wEc>r2&?jDbS+WP=YfGj^1B&u4)Is!}8HmU?}N;|T*9IAuK
zNLe$uq+sW(O*c0-^9SIYk(coNTEWHZ#AFYZ5(xH}Q;_z{StY+V>aBPekX-2mWb=ak
z4*Gd)&!Z409!tRGS?hfNJsVYLhi8Hg3>!gL@NgZ!wsw=s1d5{iX5j=!1Hf%>M_}?7
z6-0edH~r3Pz+GM&Rjw|$^@#%SAV0Qv>#%>`u(|=^hHe8>;I2W;5K&#Z-cwXtExmUi
zaAbRVR!OB^AaKe!auvq;uVJ@~m||FohF-NegJc`x!S0|pmVo@1J4kT10l@;>aL_6R
zS~VbSpGWyh9d?5G`N#cwH5&%j<;%51L($S>#^s;ok9|TI6p~{Yi5S)9)e{K45BVLt
zpPl_$F_wCyvHI=HE-JoYOJ~BWq|`IUG`r`iXWdlJdI)F44Vk4aNR;CtoU^lh*A`g=
z9T|0--;`Z{YPd?*Av6VzqU)Yt+$bd&W09G*H+)3E%&m_*{0O=4U6tb;=+__zn*}+Q
z+0sWv3m{KC&;D7djb<~{34J(QsSNOqYDakQc9@f8D~#zHK%AYo9=>x>sd8%^)A51S
zAPFCDdW^juC?-TA>5o^a%5w2dqGB6r0npc(xR|a`sv&U;*2lIbaV@S`3wY~pv+E#i
zvid&WQ+*A^$2uTFi^&N41(xPLV<w@WKgxtI{L6j8^wLS~zC*MhR!)G4;}TO$ro8}$
zx{&g*#S_uPzlyX#9A7YJ4Q*p>=`|4%`Kd$I_{YAi=tl2}*0vIniFtV-7Wmf$nflcr
zRS^V8u+@%*<M=1nz|6I_(5Dxa8S5<HtT-ywtn792KEes@2H*1jCYU|%H(kx`ovw7|
zHzr^wE2b@_ek&FRX}boMfRK0u_EdjQX6GhHpvnJY9y&NKBc1axTfbjmhdD28*`$St
zqSu&uAKI4!y(aseL@YwIzqiBu%HUOz_vyU5?cJX|?~h&sU}e&^wI?Gadz0Kd)dkqa
zi=b_CwFxi^0)fWj4OxfVzC_C+`tDEVh`mg(uWDF7$W!);PgkcX`S>w{KmehGsusxF
z|Hr;qoB)mNT53wloUOvMl%#tH-a!%EMRsnMN!-tsX1>U~z2TA#iNJp=B)ECrsfM&n
z9Dt%7XaXhITse;tR~!r5)@smNHE;Gx12|*$7z)#btkRhea(D;iH0T|v$ZrDPIrNY|
zMf1Ue28&IaY3joHsz>~5a1*o(O%g<C#(~NsE~Ec)98KYqzaXlDhQ~R>ttsstm|5*q
zF=;roN3vt%&W*kjMB}1Fu99uYp0{(M$?PxAnfPLp{0bsl%@e!%io-DOX-(u$h=D;x
zXh%Cn+vFPbDg@+?b63QLE&l*J&OwcxcPN2nRjeK7Gn{Qv1ezCzf>3bZYfQLvPggNh
z?wSaL?o^UEi87tqJnfx~AP7zM(Sp(CT%wwmBrtjR&cVLp@H{$b!+|=0f(<dkQ<h&>
zH!x6QIm9|X>+>&CB>qsE=n<3|#QXfhH;U}@VubJ?GN*ahbxq=})%Vv7k*X00%f_cq
zL}W1;6%nm^LOV?Ey$iVa=+y>7_y07W8IZOcAzOv>O=4+3G#*K`rzkMfZh2%&+c$$a
zl*;<nknQogu!Wlyw}&FQVnf7jl9lJB03k_nw#Yy|NDY@GOF&!+SYRD|+k~%1bpqS{
z$WI8eS{tmg2CV7zCz$tAm`dal9n{7T6rLu7q%TrjY$YmT?4eg}1hO0&O3p46ocyss
z>5YoQon>bvuVqqwov=x)-sLc|uDxW8JtGWRFH%tM6-iMt<a`FmP0t3VS1dEGC2Hl?
zO{43<2zi#Fp25(>kfA!vS*kJ<gz6f6O1gW0h6XgitN<GjHi77QX0tj*wrg`Rt1J61
zY9GoWvdKk>#~RBY^ekkk)XSe`xwqy~%{|6cM!E5tPlP6>5JUO+MY24(OkEd8n}U0=
zuPIb$pPlftauR22Cwn-$9yLN<)WQiBe<n>4&Hvy8ccJu0b@1R<tNiSwo^GGaUyE2T
zGsbV_BYIIi#3(^@yZN%44eq)nG-D~)^zPzx?<BQ0d!0%-#w!U;N;p~`)dR;N<;O^C
zfvRI@LpxfTKfX{8BhF;_FR-QE*qwW~(&buTTI|hm;?<;t=-=K;T)y&=|5CY<v>~k{
zJ)5xCXx%(17rt$W=L|7Zf_UL-iGtnsBNX*-4d2x7Mlxi)#T@Z(&@XPk6AQo@IeaA&
zk}J=WtzUT4-@L6n3%m4WwEf69Jf3tukfbs>ni0vwePYY~N6O1+Y$r2q#>JnW#hZfB
zh{o$b&srHi`Z$isSPgU`Ovx!zwsISTTgtq~dX*8C7O35lL$jIomH0IllUZ3hNn!ji
z=bXavp2%8_$W&C32u#0;wV2$W8A`k<U0cn<Ch7xS^$mt%is$#0O38|j?1Jy0un;J1
zSk;?9QSUc>*glXc1w)j?$>D@gXS$t*)^x9{ea|iYS}fW=Sfn4q*Hd8^U}F^J=#!=M
zJp>vRg77c#TIm*<&meiPKCB2G7fNta<Ryf+h}~Gj;YM^RBfU3#>IX$D7gP|&b3W+(
zfsAk0vuGdwn19lZ4mzcy*hS?}(WZaB?!J0smLBpj3$IG*swEr>`)KxVNd$MWO^r8N
zY7$3<QKTJ>dsHeKzMgW)^?Etl<%g11Z^pOTx;6dVS}1n}44pegP^h@+a%(ifJap1{
zF;Qn-f_K~H(@^=@#NfJ!Kd(7TenK?>TuA74RXBc3M{K-buXje3cS&iCqPe~U4DAr8
zJ(U{9l(+#e3uH9G^Mrd4_Mz$`)hkJkC1rzY9;7{@MdBUfSfDN8um_weQOxdrXQ?k{
zVaRMLEzZGr)2Prt;ZT$T>L<TSUc>m)i?Q^?*7QRa(+~U53PxF(Whgez&U&wU>p+Mh
zoYMKsdm%35;%M4xdnc8!zh7@l4U#6U``j(q8Hy1J3-6hI3{|GE7j-v!u{yeQB)mJj
z=*2r~z4REWcH|U~z0s)mS)^BJ=Jg~q4MH(s`@ZGTo^|h{#qylR^IHp(6FGPG?VgIb
z-Uh9C=FOU(<h22dSiw+br83i--vnbs?|8$)MxhN-6ihZa^&y|Kv8%H~lx8Eg|2F!3
zgrRtBPcLd#%G2M5C)xI^{#=g?hf%jbwz)YqUTvSVrAlsY?z6p=;GeO`eo`G#=dLKe
z1dsS~DY4;yj&KwyO+DSxK7r8bTGGb|C^`t`D20jfzpJL%bAY3wr7&5g>}UJt_uvp&
z#I?jdDM`uqcH1<xB2Ez54r|HmQ%pr8?!WZBy0?f`o7D6^wPEbk9HN`8#BM{yv+Xeo
z{e{mf^L7P~xJ1M@3h7JrKMf9GH%*9)_Aq#caDQ5jCzmd7WudT5N|)WShY#lYWM-$-
zV$wG-O#zckgwsl5v=IG-HSXK&!C=MqxkM5X3w@!)aiFAP`{KLg?3GkxSOC<_|M9t5
zKbEX+1mjF$3x7!pEV+X<*n`mcls?T0XGB;`j*)+RV2{xNi#%jK((5=otN&&yjs%I`
zj+-TJv@o`_!H%B7LJApnQK%svkcokez3f9D^zt5Z92=LmL)~fF=>14~!dGLjqQx4i
zt+E>W#lk9$`%!9YYFWcq_r9hMGZzq5CYzNd{0zPuNO{E@mI_CKb%X?097QNIT$h@~
z0I)^#Ql~hzYhxipa81-L(PR44p?Xq7k5Oqo7!jxU6aM%r_sz^5Eu<-x<fp+7=!H+H
z+#UN^wy-LYbe;r46&;jSOJbm)WBAcC^`?ZU6>$)_>#2ppXrlgdcb*jWE?MS4uQH1G
zp#U6O(U(|jx>2lCbmZKY<JZ6w7RpW{vp?B8T^>1Fm@lQ<)v#Ya`Uok&_3ii?L%$Rm
zw7%!>v|axJX-(8o$j`oZNitKJ2BR#0EU=Km9UJLRtX1OJ|Jr+Dq0t@I_>>*Yw0j0f
z?VP`r;`o%#xo}6B<o4_lbL!iy+s$0>UHx0PJd0z@SFom8G#_b>QO!v^hWpa%4vJ{=
zyJ(eYr6(zWUl6fW$kF?_Izxoc)yR{s{;UpMMshHiPQqR%CX=xyb!N82*dsh@MqgXF
zgXG^OB>nQb+I4?D)P=F@+1qE3*u?B+)9hvRCl!b;{4ja8Ssy=7(<Xs5YagD1E~R_0
zS9u%nnmP(~rjByOidLHTAL5rg(ES2|K#dW~!)6;oR;5f*#mWr@!g9SAUdbb4NHf?8
zJ*F4tkbObS1VZl7ERL^j_XA*_y-J_G!N#2RIIyylJD*b*m?3aVHqSD#A9~ixw~fKE
zjS^yb!(yA_m=!u4>|x|wNj}4gf{I_}$4^*;<n**f%X+3UeK`Dk?~6(Ihmp*U4<qg;
z;`O@CCl~5wy44LVZeOtd8gb2-EfVDRey8-TpBuO38MWov%~<ALFzB$y=gRc1#7%%~
zQWD2p*VqK>=vyArSwGn1y>^6RYK>}5H`!EPF~N%m*=R|PFfJ$#nN_t8<bo~_4_HZ+
zMY9G1u=^{@Mbzi{;;VFCC#^fb&I-tV+Z{=>z!#~>%!3(SY&4<bnm9GdT%!HkHHHnl
z-=0req4$e<l+%F8Hx7cS%{m9qX5ETMI||+E(Aj~Dh-!6jgvr1_B|dkBIa*d2L+E|X
zV@Dloia)NhR5&;Zp9PSuZXITA1*Zbg*Q1Kob?B4L+TpaBEA|%9;G;u-bCR_Qglw=h
zj&nDAY58)Hrm~(y+X&fk>nH0aiHv`4+u+-;iHsjeH!>0F(vK)89t&Rptk4jxqGrz#
zX#uVy>poOC;&k1#UjGxS{@1S8uUuhIlkYtE*4pZ#urR|i!>yb3r<<;ysfTSUx~8qS
z-4N@c`&^Hx!u?cquRDCDGl51kW|t<_^FL-8PT4xa1Tcw(ZG1tZyz(jxPb_|-o;?Ap
z`qn~2Va$r0-81Jc2If$OW=}F#vCATc+6*&P%gB}%=!r{nq>}yNncT0z6uVn4cr#q>
zt9M2cpK)`>DVGB3Y)5O95^4ISTujib@dszqIoYe>&0_O(LxuWox`AGlmqfwpqJw!9
z1?9Ac(apgE3P`DZCq|QRVycVa9Ijy^;@%pQM2W;1I|s^Fue4)6T~N1_rLlW5A4e`F
z>a-oK9Am1Af)lMSZ-OnFU6q*{!Gz90ebu`~gl#u;Of%L@sriBu7RM-fdFztLnIp2O
z`{{87oId%CL+y4X(9o_G-9N-qIHCTJai{mIai{5(Z*hFC%_;i~d*&t0d2gl-^BUd?
zMXLQ#_fi#A3mV;1pk_22%^3fRUjT{Rap%VQR8*H?AGZ)E#wx9F^IX@TJE$pFqR#8c
zm7mqIzo5|Y=8u5R-J7eQpFmPkb6RDA0M5|Iq<fNYMYx?)Q2Z%TuSIRYl4CPZCBi}d
zNB6;$X{75FpT2p4=0(!0%k-*O<5ICFD)(?^0ioqYv2Xo2qr8x9FZwdriO6-XBjdwO
z)AV}e?;BgsZQWK)aTcm0g?A+>(6O6Wvo1=BH_Re4q2e4Mf3x!;ruZKsa9?97q*gRP
zQH;!sp6#Eb(<*r){$FsZ9o_>UQmLVBzmr4#k4}avN>yV_!=I@OwV4@1?}bPhmD9+;
zoFYVnl$kxl+T)7Z#J0IfWsgfcF8@;=DOhm*a}f0wnN&sE>H*L8>!Qd4iBd(AgrA1q
zjSekYOqs!6gjBSDjFC|Baa{;)M2X3XePeY#+8?My(>PqJ$RpEkjuxg~AV~=vbb#~$
zwRQ5U(PWXIPWsz0e7}bnk{eP??;AtI`u^-WcvN=y3-U)73zcm-72CLKt%~`l1zW)_
z_p-(%BlzNniMB>s<+_p0Nd2|rI;%cfO@Uo@8zL$5yagn<g!G)I&5S`B9*%?4W{WnB
z^-<_)M^*5}{!vb>@{@+RYe}l-LHB=-ODKL~Fr=-K$Pcz)U3m9PV`j+dB?-m|{%MMA
zL0q!GoK{Eni*AGT!%$?#u>hU-tnC}!MDgt*mqt-2;$vUflxDeD7eF=@VH;fbuXC6p
zsb!9r4Qq=Lo0LcD5vOZw%<^4n5c^S$yD_bH21miJh*@?#1*wtP=G>BnY<VP?iC0>+
z+O=3MOF=Y$o*U0R{SPt3YwQuPH)?49+rbrt)uRRjDt=i8Yq}*CV-Mg+Gl_~=Mw~>l
zsXnJ93lWOcTm?TwEet!NLwtYudYJNvztFZ&pcl!$HP7JZS+`KP0ZHsu)(uQFOC@29
z>P^(lp<Jqdn&*^Fo8*-Gpy;+8d|MB#lS|vqX}vhD8e26b?UvA%R=nEqBy}ZPF*P=I
z!9n?WkSf?{!{sjTojVzEQcB>Wt~Wq19xjyvRp9xm7yy#8V<RPNgknW`L9W+?(McMF
z(Nq+1d2!Z0%!8QNqk!`Ox!YYTK-N1(r|uXVDR?HB;@~-93tH&4RI3ncQccX*C|mWf
z<ppz?v8$XOQRPS=Tc7Tx36vB{WDrWQ;H`MM)j45oIr+OCIQRbS;Tt0@L<9Ar0qsJJ
zxjf`J<bb-mycde)fVq}rB%k~to5R=T!T0dwH+sCS*<Z`Qybu+@`898j!BT-i0(_XT
zJ=vV}5Wyt*e3fS>8i*rkkw*H4<=Nyza$lwKV&xfj9ED~xrD!pwbJbPbYE$*Pof35e
zzGPY3nx(*&Gom%T8@^Sps@3<Gn1pE2)qMfZj?2;WL}rH<>lrj;j7Bj}cCof_`>gtU
z%S^j5n#>U@Q_{y@u2SYTd)KEqf3TI5d!5uTyt$FQd$`M?>o(6AOq(gIEh`*(__59d
zoO0oR+>5IPnXrKcn{FbiF)=Ck`#~0zs-&@shY{~s#~zq5?DEgFVVzYJE2mg|yX_4O
zx_M#-g=>6t%5g&9yC2QrVcnTz#KP$6E{D?c2c<>GlD@J38@vJ9e@jdPm(J8e6ezrS
z@e>ga{oTmY*@0KR?B+Eo24xT5_g->Yh<6K8EhFm0e_n0&EV~Jj)z#KfEu1ez(qKL8
z3a41733PZeHZ39h43*qt_5M$62((775S$U<Ls-`x?H%&-kda}eOBk)?p-JdO1ST|S
z=(yTO@&JSU@o8Ilk5441N(ia|zfM?K&ZW>eYcvXI8%tEP!PkeV_Z`vr(#OP0)@2k#
zQeJ2^+OyRCodL|7m$SYc*0Nm@f)pLcaF}+U9+2URG>(6j%wyJ+BW#(uk0EL*ASRjE
z%8o%@7vJd+$km^CZ2(0Iz*$2<HNbs1HZ5aLY9f_1uA`_UC2KqhqJ+FkBs;>*d&;kb
zPPgK;f;;VSW1Q@<ewNM4=?lmq(@wH|4HJ4Ql)EH#@qFDZI#tcJl8N=gG#$;yL7a!F
zXVIu6V#sEOwU_JVusEB^tq{mD3mrR@Aih>y4irUelPuB&M~^+a9SMz&GlR)dFG<>X
zQf4mN;c5t6aclmSrq1K3a$`{-zhRG6PoivHjYDM8f?Gn(M>PLtJ9CE4Nd3exj5%r+
zQAkO+>B8Kikf#xTO}1}zpIRTjjKw!5igmG2fDmZ%iKlgn3#4H|ARSm7^kOqQI^E9D
z1XCg`dSWTG?witFxE~g-mEFGd%MXObq(4<?WTL%qs_4q9Iu<!R%8muP3hNGR$d4Jr
z_wmoETgt_ZS?oK~o#-rmd!s6c%c!Nj?_C=yWkA~Mvm=Dw1_x({IDBB@Q|gc!2^I|B
zBN%&#0-5K65MRhhFa=9LWz99$ry>cy3}#t=)igoadVisT!FCkJQo+=F5^QXN36=Fo
z3Q&J@!>%7&<*0K=<=jPX?()>v2NHIegLfXyPl(Z)+(fc0OYRqCM2P>6av@2u4NW*8
z-BtFEioPDCmkPh83@1O5O<AW%u_s2px3E(=C{ME*Zw^1iUoZ;DxDIRyu`T)`n-r;%
zzTi0KH2waH%_{WTvMrL{TbPOh-^@Gx@Y`3B7h)JU=fP^-DLn{J?omV9r4wio;NMfp
zZORkN-<WMy78w)H!V{<JIAX3~HcyK4k=8oIf=3`SxU=_uu=(4)`u7l^C8;h$RO%!0
zOQ!?nd|itQA{8l9^O+a8X~XqUfB_yh@K*ml9vKi7DL)*nm8{WMBQrl*Y`TnEkV^}D
z(}r}z(>tLAQTidq%sLHkHiPa?fm(p}qI>F?T$_ofm~f}da3b_E<!7MY1hv4=Ve4!#
z(3JO&a|M#NfYKuvA{vOlVWtGpfFWAp{V2#kiv-Ce&<Z7mUU$1yMrObG%#SQyX(pF7
zcrRkoa%7a?$|Z909T_5Kb==HhTOt~Ku5qbjIa@$8*-KM|%03eWuOw(0GuK&{s~j2}
zJA&GMgvNz+e2=Zb&$CEp92K(HO~mud^&~Sn6HN)uJ4v$<hT0$V8dG5ru}cpLjm#NB
zkVw)`O(#{ie@=-@Z1UZFDveM(8;2b`H%<rzXAkOypO(#@yoRVrQ=|n?wmw^(0fo|>
zDNJ@f2Rq0V10nqGobLH$H()FAW$YDIsbhbAmia8L_M@BQt(3m?$%EgntwIXfcs9}$
zabf(kd^1QN^V}&Oh4Z@*!U;JJ2O4;A?vDx`WU;Zui%p!)f`0n<*4T?xE7zRH4wX7H
zqcBVY;MCa`jF^17Va#JiBUx<u1U}0;Vc}mad98vO>|Nqo`2w+5m7cm{-c!@tj??;4
zJvmAZ7w2#lcw>e7#b8~%ggCc=PC!G@i0b;&Z)^EY3)EM+qAhbm_SrgI?2UmC^(T=|
zxIcs$;fvEp#-rb>Dy!DhZFWPl+{d}D8E^S5>8QR{&UqXSW@A5nGAEp$3bqMV)^T|(
zkY|Y)z0eA&3PDvj14hKLx=|$&Cyti^{~>&)fTv8aGUU-HhIh8kHpLjG5$dOCTdCBY
zzX`Usy@H(zwb}l{GIxshfW6T|t4ek*9RFx53upe#d<uBx_-r5aO(@K>V9*`%RJvxW
zrHqHWzzM!?iJAOec#ncpfqGQx9(=uap#u_KMZW8xeYWMc+s<@UvLJNdJGcNlD$qq<
zcmd5}C{vws1M{<CFB&R1!e~M*AHqn<4q>}VO|?zDKJysg3_T@S*ssdb=T%r8Wz~~~
z7`R4l^q013d5cYRez~b*#TX?RQ%SK1eQcS?-;B)^c>HxlMd~uIxf3l5drrI>W(ho5
z6Me(^YLh<(*J0S><S``b&BSK@I;_@Zd}XEGA?6243Mo`^1*ug>OW)KcaoZvrGSa;W
zy33p2y1rtJjQNw_t<<r-#;|5GIofP%;&nIqMKAB}Yihwm0tk)k-8qo2ll9W*$j;e5
zcfXV2G1fWNFuig-a#a=IO3;$_I<+qhxeU7H>teBB7{%xp+OkQdTs~>ki5mKJcq5Th
zKu55-<Wuxz@#gci!sdEbQ{QR$NYD$z&ml+^iEf5kvG9f`Cn~x=++vr^&Psw5N+aY-
zxa9D|;ixKJ&$k~jXIAaJ^Q3Q@l4T<=^&!HE5f^w_X>#IMmMia6(#yr3sp}GLBqBYr
z`>PT`)HV7NN%>%~(p>Y;GSA?aM9{5EMhqPQHh%eT9j{zCVeY`FgvIaumCA8tiap@E
zFYlj;IGaVXslQ04X}pvO;!RIX4~W`dlB`KZic(gqc%ktm)Y#4!AHO4tfA$;Olv0Hv
zP4W#E61M*!2FruH(RVhQT@CH!U}--_!AX5EMba%Kc6l-K{qeIb9j{QVbfJ(m7tE^4
zaA#}M58(s_%*;rXYzG!k^dB^48$|M{&aY)iGn$KmP9j_SjMzM)qLhhH3r+IZ{&f5d
zBjeA^aF@j@qM``O?_`q96S(x>U?kGZHpx@rNK;9`mO8ibtIY752U9``Hh!0YVVUZP
zM<6@z?bRNFtSiLnPPF+Sn|dsFu{ElnD_{YKJ!Vfi$*`D&)0=#4$mSuJAYT*mHrs@H
z8#XtCk8tXz)5=8o@e{9Q41o}RjTksk5EeCgu{Ijv>$d)e5#SwnI)-6%1mur~uLfy_
z)2m(06CQe&b|lblpEp@#)1Bt3*Yr$VpR0W~89j+z(QdwwL}}?g&^*Qk*CkpeGZZPx
z*RaBV{X|B`!;*m#)KQKWjyXNR%&j8zkM+>~{lVC5INv=@C}z)lt4BGa2}ie!FXh8(
zRX>>qU%$X;AqgTu=T3JLsiV6osPowU*;OM*Ztf{GUloC)Cd;rzzdfXRbQ9dm7Zz8^
z(t8fQQEVvU#h)9~N$qmNh#)g+L-UMpNX*BK7eK1l;_jfxAB%5cee86RUF=(_#dDu%
z0!=Z?Ean6ct(LNnG;o!sGG)P~01rFN!}{XPFE4b|1v0>wFgzr5`Kp&AIG_Um0NHP;
zcj;Tsu1C#Q0^j-10GbOs$IEAw{kPxNUZsol3Gg~r>PTz{u*{6j8ZgrR`t$o+mp^w{
zIofCbSgb?*XmFCfEi@|XZCKL0{rf0u9mKw&*E?(4kQ0Ix>Pq*}%K$7?)8}!95JqLn
z`#Jv@O-S@sA4QU+Va+LQyT}!syDQ|qI8e;{iQt_0Db*o-+%oIM%{WQsK%`fnVzKn1
zMM3);4~vCdcB4Z<%Fs*C*%IoxW+7~(RAq=x%uVG%O^Nb!hG+4qbS4cV?v+&d`H$+n
z7EU@lP~Rh~yZqRudd`UmS$r~vCcU~UAY@qEJToOyT(>KWZFPWNdeON5Sc>OP>`j6g
z&v_X%h^#*WkEv2dcAY!HtK8ecW15>0-YuAcXUI?Mc-?YHVe+tpyQ%?hl_}!r!lQsI
z5zh8>nIr}alSaNb*W47(*M-|EF?sI-*;TYF`9gVRUa_=VR3=tTsZ=U6QP9LVDpF8L
zajdLTiKcCH+A{2eOaN&xa(Mz_FbB=0!!^UhPD9<M?i-~uST)vtLJr!Nc0W}oS|4Z7
z4Wy^maZfv%!GpQb5~n`voEV#u+@;MgfUSDDO_@~~hR%ZfW8WZI+C~An_Hl)^AQr=L
z=R@>_$f@D^;qS~JJznzk#4AB`nzwc$6I^)Xi?>Qw=j+y}yrQbqQyy_JY$!C3S1~Es
za}<YoM2J;sL?!mFhFEvAQR!mUmcDrf9m_R+&~NDwN_S~&#PRg-$pu7$cfD5<6hY*6
zbLrYV`b>iAifi*`c!|^zGrfkh&F#fJ=%MTt6Z**93{S3d=s_Q-A}-g}F%u`3F>|f7
zN$!>0{!*ZkkaOH#m2p@cau|B1Ec$piD$ts|sfX1aqk+hC*1jP+`G<Dj2lFNo>(-Lm
z*s%wi46e>U;(t8AZQ+sY%EY6p(<WJ-iyb<&saYNMs5xvF#zrDq8)a{Q0wO3ZSj7pG
z7m+jT<NbNUd%vkJQaqGr@kVXu%Yj%ZA@PWR26!|m;Fi@5q_-6PGL<;i%<Xh-xUi8U
zn~j~lM;PqKHQoM>N0@6|Yp9zfEMF_94<oFS8YF7vZ(|X`m#;IbFR+yTxWQp9_0j#%
z`%~W4ph_{p8^@^vH^oNxpDcil3=HY7vWM{Ah76_KFK`8l@4Cq<`n_ki^J*e+TQnR{
zv}yO#NL7|WhUIMq@tk5!-8+^#4ytW)JHa`BbjeHm;B7<#Ay$CsXf#rCPM+%ZslG1_
z7LsM8c{6j&!Ze#73M)G7Ln26MV5wTnI;5_q;OWkk^k-os%iL3#IZc74yy->a+50}A
ziZNodWft=)ndXMFgeCv$-$fVXDcbP_Dy8H1$PYD|;G#CaYJBe@ZGYLVlADTrVm3&7
z-E%d)d1l9YRgj?Nc*x|r@DX)NZDO9ZuMfo8ib-EG>UF8~`KW_&nXNXQbDJLN)5YW(
zE`qImgrfc6&(`$$cg^cti0NVCKc@$j7Eny{`S_iOLgII9jOgMQsAEfLfmjw+kG2Pn
z@wh)p)+CFkQ7+saJHv`l?29*fRda9UjR$P6n4QaRyU#W9ZY{6##4!1@b&%v1V9EJ>
zTz?)RSqmxUn3wN<4^RB_RsEDvtnz`c7j{DsMPcT@#md%EPm?T<jHsGgMR)u?chIkD
z2qu$BcNmzGTv(G{KAFOF)&?b8{yK*^peoo(7gy<;u&_dKN>-Ya=iqN?J4||LuUh%Q
z4wT>q0mjsmBwDtq@>uv;iXB~2e1;KCG`u?!k)B7zZ57p?iAX(XP2$Yl)rNQ_1v4fs
z(jh7jThYQ>A1<~^akq9@wF*&*h=O)b9nLVrR@QTLiUxvHBJx61;i*LlK~UY-j>v*4
zVv!XxJWkr^Txt0}xkO|AG{GpHtX0aaRcP{t98Oe5<EXuy=s1E&G2(0sPYyIf6>b$$
z^$YoNt$nH05&b@3RCLB)#|aBci3LTvHyKS?_?|lnsrAl!GOgKm?B12N5^+p}4P2>#
z5Jb;L^y21ksh;0{GB&2q7oVYh*79Ue@#B{9Fty4-S!B6U#Vg&s;8f3ijO$?`3YU9&
zRuvZ_9#FLtD{I>BalYZl;j?2yt!STG3&ksKf+&n>uS*&ZLz}I(JFu+^%Vf46>fUFV
zl6_NZhT0Cwv1qJ&Uo;O15x)M(TgLNndGt-)xAsaKSM!QnKfIS16#M%|FO{PgdRcU2
zQ+T6e5_3U;K(mc#Hw~mF5EPe=Wf)VD7aevanBr8i3ddIXvGo&KC;vE}s$#SJDdp8F
zHPb3Y*l=B~Y%~mkCLFl;YQ@dQ1x>1w6q-)<bBL7t;tpTICLSfl4<NE#F1Pf5exOhG
z{&t-en)hS-I-20DCcc%kyNY=+0xP^c8g$cNbcHtILrsoFtkV+>_!HMg>&&1_`KiTQ
z3C>FKT5*FvOyk0H-c?wym=)<*WB;h&H(EZSUwd-mhv$*^b7U&vSmCulHXL&-Ii$By
z={eZ|eR$#oC=E|rN1kEP5lB0<;MbGBg7wlSovd<Nj^$*Hw_+5vpC7FFvyqeJd!O^L
zbWnQ>j);Et%1)P=lM-g#>sha|Rx2H{9j>pmwf33cHlfHN;OE@@xDM7+BNKyR1y`U7
zfHTZEebsNN&(pVvg*FsshKuVM_wTIPmfX}U-<F2pE`dtGPl?!70iUZ?nM;BTlz0zH
zlU){OKYvxuOV^z&{z^se`nUbMJ=jNS=g1b`)UUX@!1tySvVqCjs>12}`sX%09=Um5
z3O)Cyn)7syn62=bH)k#7F4oM0O4kCxWm}S04f=6+Lqk`x^`Rg>Qg2GHf7)a@cxfYe
zxshG3|Jr39BgJ0_`>d`haZe3mh?YvY3g@`=0bvfi#dd4(p3PV&HWI2QRN5w5n&hA=
zo2^XCq0SLkEQAJr7OO1O8%aQ~KPs9t_ZIT@_3$6FJZzH61#7?0XAG-XMQ+%VqkN5D
zLZCW_`qC8fqDkDcGnqv%W-Uh|Zl(HYQjIh>RbTa`)Lo?A@?2tGeCj<Vom%~dWH>X-
z|Mlww8toYGO=s&FzUZnW<E`q#-i6^k=rz+}_0H74CPl}lkWN3PCv1`vLdwZ7$j-I<
z81%#Uh;`?0U8vsqj2G-x@oyUbWZ5Rv#d8O~L%wu49~wBF{+(fX;ymYJaQv%o?NxL}
zBc)Z=tubz3U54!nz3c+%W@f$;XRyO7;y{~jGRWth?-Sw;C9zzLmpAfzmFi?a%i=Yz
zQ!{RFJ!W(NPLRxDrebDal=AV8e~xue_Ht31Lsa~gF5%w?h4jVl5C1iPG`T}O6EHH$
zcrGQ$8T;zLj7v@48O*wDeiAbCK9of4MnEa|&HeF{f418zVvW>%*U4wESzckp+d7IK
zMq;Kf0-4W1P^nzcXOw~`L{%G%c7V!-O%rzB8br4&AXHa9WrA~u)IZ-XYQh@eAW}`*
z9drj;VWyodAzmY+K1o6N${j3N_|}2h-e_E$HvLy<6*epHR+rN9#xs;1G1v&H7B^l#
zSA>zcU%O#!5~=M=bY=>hq#^ot2l!{Ly*zIXsS?i365PNTjXiVfvF+gWXj!GoqU(@%
zC>&b~Yax=r^G9~5;bu~n<rf+I6^T%4cOttsgTp>@^S<s%onP-0zP4kKU?~%^6@vQs
z3tj#5RwGZht|%OCOe`9jR%t~d6s}C;glxS4c@epc28b^Ggmh7Wn3!FAy>wx+BbJe?
zb8qy#alGCO2yP1Ru8f;VilUa|^d$2hVf^C*^*p;MC>#b$@;sf!;V4KQK8VAA+LsK7
zFe4v)HYI}X$#`?Fh3>A8JPA3{`)!E?W_R9__S5-sZ9n?7u-TByD%sJ4M88Rp$R@x<
zGw{MU%xciYqBME>Q)DoO_bN}Hve4U2Yk5~On!s<C!WZrD$!K%ilLYhDX-%LFaj)Wf
z^|az-w-b_?DILciJg*j7(~n(yuoqGJf%BtVICpG$%#h?5+R3BzJiWx3dm$2#*+)ok
zT0U}9AyZi5AyW)XeSABKd_6bT)Kk9{CwpYvfA<H9wCX116|6CMKU;;@=cy$n36_)m
zI@U~*M|4J7>rGrE)<)8z&&wnaWl!1kkN#X{SP&Ajd?x!Ji8)-l^xX@nj+AETUgI)G
zi!q#>u0Chul?Ab47XG8r;lX@QF=8xu9vM_#SOnR_?ag*m91L_3t%hWHXqOL==qUvh
zT178Hs>B7lJBW(&ZkgUI;fE$X%ia+5ymw6ZQdaXFpnjxQe@-5Fo<7xk@c-ER?r^H#
zKK^55WE~wdn^4(%&upPc2q6*KB75)12*(O#CL<gUNtuzIk-cX|_MX4{)A!lG=Xt)*
zzrXAHT~}Avp*Va#pZk5^@7Mb^G}`8AkE!4BeER5%i@9^YCm?#80dMbG76vM6!-u>U
zH9#rv2Y=*e*!I@3R7Q#hgV7GbcIfI{@xxkVlM~0l37n#IgRg6pLH*Q|VVhK|R|}3b
zrX_zpEj1~^hQS~yzr(jZ_4M<y8^ZTmo#zNKfFsmA6pdEr@3JYgyn3W<{UQ3xoiEw#
zMnBxIUA%LEv+@i3;~~#kj}w+jrqkjjjrKpMp1vzDPV0%xFsQ1%xs6IP2e_aswm*3P
z&&?8kL{sOtA3naq5>G~YzE!MswOp6PTV3XDal-W3b=sa9ux=BN*=aJ?d}S=y;(sj1
zuMZHnIvn4kI>|0hVtev9Q|#*g@=jK<+J#$6({O9blMI9FjSk`(kDI>R(C|#?<gaP@
zT5({W0i|E5$)lQwe-U+rQe)tpW>(5QFKT0)G<T#XaMX{QG5?Y_hn3#orhuj2Y|fZW
za#O*<BS3eMCu3-pQ_s?xwUA5WBtWI-cBa=U^|dFC$uX?6J^mtFb0o^R*a8KPBXf9H
z4!=K^4Ty0lGUfA+nSME^-?lPfJxl*(?g=@PP=1#=@0z{$;{>rjOw_D9PdKed44zz<
zcP+{BGwz6;aJ@J<Rvjq|-x!tBCaXp)^5>y4OdrWjMf;{nP%4}AHLdcS#<ef}VeT_=
z%Jf?e1btU~6a?d>LnGJC_a!&DR;dWvUeNr!H%-Tyq=8u0k00xWLr^2as1T@&)y37Q
zr&s1x;=`w2Q5CN}*j7yRj`nluW30}T%rWW|KfPAu;_UJ)$8}o(3v2%Fwx|V-2Ynf7
zdZ{X>!k4bWOY`j8moT}uIV?zK9_?TE2q8w=X51i9PpnAaBXDA1Hd*1BO>gH=uxSoB
z+=|MQ9KDy4(x@W@F(+|!A2vh@zSd~PGc9|kM3*9!F>B(MMh??HU{lt2oKOgBx)GCC
znq*fNtv=B_AE+5!9Up3K$dnu+;VSA8BdnHRBYC85+==>zA|%fDjP*=DeR!WT|D+)D
zW2TFm=dCqagL(!#td#*du}SM6r*hRdP$*1U?9mG9@?}c20fuS*^R73uH~b)xovAb8
zryfkjM?r#AeH*ZO%KU;7LsE_}7fRh8Uwv$s-6q#tUP?x%9IQoaC17q`Wg_A<t3cUp
zCRduK3*zPhug!XLP)UENYfl26eDEQ&!3_Jn)Kn3xXpSO++Ipv#;m+e<>{4mMGE;;V
zK|52d%U97Atfho)^XY;G&)@(;RIQ_cr9J8mhOu-+ojnA@a?Y?Z*fV1}jl}Xid+Gc$
zaav9opcd^cmuHr<`8ga<1T(aKc!+K)Tl=8>kaS4ggN%<MUa<@|UeLK7VqKG`CQ0R1
zpCv`XANfVOY%$tZwprx%`r{B}o-x>L$uj=F*%q+CXccN!SSJ;SOzQ)ENw7_J&o;oE
z*}e=JLA7iAJ^yR7HnvGP2I`Lvq%D;F)09?9TaHqTId6M?^1T!ml1Iq_<ffh=V$q>3
z?ep%fM5$-a3>%8DZ^8HV@477+_ggcs`T<8M7YOb<dRvUQt5Gkm_jd-_-^5hlg%ULe
zB2Lk8T`INO)s#<wL#$VvIm@+%XpOq5!5opMxEdUlW?SCoh!@@XhZoK!_cg~~l|!EC
zHkkey<__xhN@?C#y0)E{rG$B@>{P#7sW&U88R>b_vD-5CH<=<-#2*grs-ouU?HV-&
zXACPMq;e+_lqeUvt}O2{p$eVbYjBNDhJyZ%aUh|Iw0d-6h1F}3L?~t-|0F8{CyaJu
z&)<N_El;>Cf0e6N%Wz%fN)Pjs_^a7mdMoECu+pDbWGn~i&#of8ln@h?51#~m?JaUv
zP7|e?dF85+&p~thrDYrINqEVLMp>z<Zg-=?CF9JI)k|kN5qbLdVEEb_e1$H^Dlpwj
zJ<zQ-uOMEX`{QH5j=03ng})nBw(z|QQ*~HXNZNMT2Av^h^{;A!p@Z$pU#3_xscK_K
zZ+3MC=QlL!(l%L+esnw+MV{%lbId+qkI}wJW0V?wl~iVEY=%6_ia7o>TYQBi4UbZT
zjD7?uwqlh=MYROj#O3Nw={|spO#krj6&byvCCI10JZG!zpJbwF7YIPCK7SW6r(v@&
zMS*$v!Ia;Him?R;>k%D#pr*>p>cjASpA|wb^l-o>^_}(wA^v~_?(q$Q9<im@RtO2^
zgM>;~>n3gF9d>ZN*2;?K*kU7<X`O<rbhH3mS+bzK+H#}TJpR>es!a*CYMo*?p{#=8
z(X~{nik}q+TEt1u{nE?pc%8+sV@kp6<s*M9mHtAw|6|$<4ScUN)lYxuVv7MTaZx0T
z{VC$tMMf|ATNd-D=C%W}ohV4Q{Obc~>P#y8+5&le*2B4c=$~nD*Pc#&sq^@$SuC7@
zAMVgm`xVI4IQhjGS3AXBPl2**V^-8qjgtcZ#7Zw`=?(+`Pb-S2hOK=c%XuGX>bXMv
zN0ca8*Hf1q?rs7nyljKmy=qRNJncnB*?5s*6`)-n33j+|0S)I%%YrI9#~(#cTA_L8
zCG}@LQf{??>@~WsB?Et$bC}14*#T9SZ%(6d58P}QlhMA19FHFCaubWq!yRH6odKFg
zi<Lw8BJD@26$QgY9hh6TXC{`Ir8PPXx}K(|g?tk=YYb+3M2aH*`Aa1j1cR|JPT3)r
zFUP4ps(%KT%AK+{Y-aqL&mB2WG9#-WB$|L>0Z50Mijz!6JfC;_+`-JUBKnEjm3}fy
zG<Zg9IPqC}j2u4iWf>@MpVc}t=1M4#L(+?7e^23uy8Ei}5kbW_ujqly<-rY?+J1>*
z!@+^JrSPX4fQH{Ak&rlYnuhJ46T)c-{!aJ)_#ni1KZC9Bx#v=Z#F%gU)yDo{h7ot&
z5eDwx)5{1+wc~Lxt;{MOhii$>U{h8-?zlOe`r`nQf<=`2NaP!s_1NhTlfImFVcIG1
zrhTBn`>nsa(LeOSrZ|gnK>9v=<`V|+D7PQ$y4Go)#gB#o)v9&{>$A^AT}Z>=2=vS1
zH!}-YedHy1R>36RK&EiLzvL~Kg2V<Y^UT)98PheVm~b{w+i>QZw${8c!aB?T>{jv$
zG6kI#U*&3N!ISw-{wRs|C|x{Um~+XxCw1Ym-^P)5^19h4AR|*SRBaeZy=0epEnZ5-
z=m029C#(ve95yqI=8V)NO#$*jOj6jukYY)MU_hPgsJlkOnW9raAL-8!e<TcjRq?ka
z@IO{Zs6gJtNc;G!qdIpJg}n=u@d#T+iR4T5jzHFQ!oR-kjW9$u0M|1^AM0uzEh(X4
z(?>Q2BMWvP8IVt?t-HnXeN?HD-Rj-qU$NA8jDYm=OH>?D-ih{?<y!P#+Z`25h}&EZ
zLJ`kM2b4K6tb#mltuU_`qj>L9B><^~MCt1ke1}FH8e4ehO)Xvy=b7IUV4zR^9BqCi
zfjBtg-}2DvhCDYj(&vIho`*#cgmPZAYBzz$-1mIH#kYT5^l&;T2!AIiB0UhF#wS9M
z=Jk%noPPHpr`siz_gcpsl-6w5gO=MY1-i)N?U0I?+hoffZ%5uZ@XBTL7~&yl?h#rZ
zmDnwN002YML1(MB2*6Eq-bas-GaWU|^gu_JwA1l0DftSzAU~tV^~D0Y!Q{)_mRC@O
zE0%gj87;j!?<k|=o3lrODkhj+Mbn2@T5BCpm*-DxBX~Nx%3VtF12FAGkm438Qb-%Y
zx1p^KMQ4`~j@TLI@+3=;7#A1uEQhfI?b`PPtXs;?D=)+FNz7FA!*vE5OaVckgi|-g
z%~gN?jNj}fkFGVdtd<_{l2lV;Gj4_!@cy1FDnAg8W83^dtiH{%{32J_1Tv8T@@*WH
zCHbJf`OEuv9BP)?9IDb_&0{D_-PRB01wZp>mtR=qYlKffOf}_C*6lSNM2*oml>Y!)
zMBN$wJiy}B-4W{%CU^%qSrF%7^)b&hr|4>!d;zu-ZhPJqjY=}ozNd$>=e&`+0Sg=k
z@5}Lr6h=!72jMYnr&fWj3EU^ENIi{(yQ8oiF~oaMdiKn;H?6B6$B^Q7{n{SQ9`#YC
zC*$f$3s&7b^8y`dpriHjG<g)kS{t2Jl=pHA7{hsq9RUwKk>np`Xy@xR$eI&CAMmtC
zEa+|*uA`rxnFBiPpvPxp6$w*7g5C5?cuwDq&HTlddk=FXkkvfN&|b`eBh|I?eklOP
zk;+1D?;JEX$$xfa6+gkWIf$$7eF%8`Cz!57fX1VHm{Vg7wzB*P{`l=~fON6_Jp9C8
zIAC_XoPQvE0txhJt!Vlz^X<?;3W|+dS^HJV%)8q)()@Pt#)@^CWACw)kz1#;2gqf}
zXtB=j%geK~Y@C4MroEf{WrmOpD@Wg+5WHLUa_HWOjL}qFqfjm6#!@>KBmeKr8|d&E
zHw8Mb72DW`esII#hjqh*gPJrX(%$}}L%F<|u@$3W@SYw)6K0SE3Y~SzohTc19>)<7
zl=nO`4T;thS0mN<8<~2Z*<a<RVQZ2UB@^t?DRb-ejd;w?7l|a-wWsbnusADkRs!kZ
zNj+a+DYM&mIowNzFN%NO8{@;Fw#EW}T*+P2E1=DW)BBxn`I|EIVg!BtE%)ZEo(|-;
zy!Bmw-qxVmv;*tJ^m<Six~P=k1l~HL{m2YZ^&q9<5!{Kzmx)Zl=&PwGk3CR);sIsh
zld;^}*JVu=jB3QN$<T!9tsKT3KH*<G8?_<x4&Fl1fYtC;Amv(VBveR7Uwg1h-TVX;
z>E~tX1~&=AWgYP_tk_e93hJAQv9CGG9tt-R3SgtADFBHv3B8o>SXvxk()r?LB;Qtc
zJc~V=EF%(gd+3K*+(Bf7Ksem4KvGM~(xQ+R7INvAEVdBd^>V0W`nl?70P{}n!OqY~
zm(EY>(Y0DNs@+9{m)21W6#Z{U4N*p9DkdHQwv*<Q1qNBqFz%ik=~!qadiw%@i5eJh
zh&8}Mwgej-f&*T+GVM+P-Koz7uxi=e4raj9CB+b)NmUth?ErvmFoVWc@+%9!iS74o
zk}tlIyzTo{HMRa|eV0g+I#31ANmTxxg6ROhfu0Ecq2lKtUV^(~XBa$nr{zOQl9@82
zl)G|_ug-vMfBdF0YR2kmotm^(k9}6Hn(vQ^`^A1=4nA*uWE40#wUvl+4lZv;Qg=VR
z(s{MKu5PVCeEH_8)`ZXTBP(;JB4G}Ht_@A#`rtp^i%dLF<t;{hOc(#X1Pb_fz_fw*
zNxJKO#-6@GMNzGZw#Rs*_5&idhO?0!Uk+%mI6T|tG8@GHETwFSJF~9hSkY#+llH@&
zQnqg~Kg#S%yur!-`Ny?uDOsXFR1vO#XoLLAZloZKS+D8xZ>=Lj#W0G#@H(c^(wQ@y
z47ZNIAW07iawJ6e^QkKPO`0`?G=#<doW|>!|7m5h`8dkJ=SG^r>yjUTb+=1ptFLDW
zYMP+MGCgWIKb0aJ5J(P(5w42p6PM)<8vRrpd)sLymrdDA)3YHA*V~WsjUIDOvk#!+
zP$smx@_dvoyrNWX;0CuOZrm7Tv)~KtO=}kFTx5Px7WKF|OCn}udDekzhI?tQ=@|D{
z@vHZbsggGq;d_Y@{AJI-`Cm7C>(l6!Q~XH%xMu3LdXMQgro0xdwZ9B)t7t2Xox;C#
zJQxqwK+{Ui*!|$zZrbH@T~$5DQr0wrvA7ZCdKe!*;&!`{E``U1ik{n~NB)jnv(mX~
z=>(E?w?Y?LvXU2JEBd}v;aj^`hd3aP+^4ZCQ0X~Yz8ai8pO{BC%UHDI;#{lVCuFx)
zYg1mdERwl%j*_QVcvB$xEu*8H^<wMXY5p_YrNl!6b~}-~gMu@-TUPBF1?6@29zR^@
zS?zRMbdp%?nY8X+yb@MLS?oDZxaS<Gq%IlEyR8U};_6?p^xRqJEdR!Ax1d$y03U<c
z+=Rc`+^z?O12QAq>{2p@(oUR)OZhhMbRt(!?}3nEtQm>M+vB0QjB>?Bkg04O9g<6-
zo~Bx5om-c8=8Q9D>hLrKi-&rq?bA1!7~)h3VT3+ci_3j{7>h22sTjyfq(l&=kKR+f
zoj=dV#&Neh)?_(9`&T>mlhJn-q+U+C{n)-k)^kF}`v+I_j9D-v0N5H4{MW<BA5Go=
z{x@$R9w^Cy50Qa4|7u$Dzkl)X|M~yl|NmeA{}-M=EY?k6u?5LvggQm%qRc|6EvbH@
z=8wKHz7gc(pX~p{mk`gUv26vl*&8iCIl3=l-+M?d+sg!o^i0_I{0QZ;u)qmFJ^-w@
zc#Lv-raxOR@WvhxMCI7VniLHB37#V0V<OLtsfJL)s>Fk=3ueHEV}Dcze3!yfEK>{i
zK6Q6P((LoI4ZkqJmLAwrI*4Y*eBTth(Q!IdcTQgNd4F^hxH@?<${e+6g#v>D_d}fO
zN8fxePNu@f9O|T+sV2fKkOpj>z>$UAr{8V@PjaVI0D?5FjnTmDAjCQ^C2#MOZGOb@
z9FO0Ty8BvLtM_atFhyJF@tHcAq?A5vAi)V<o>&oI`CxK!JTx=%B$LPw_+B+ZE*}ST
zSPX~$T8#!kD$^18VGdjjmIQDySa!eO82(G|^Pg%5Zyz45Q!}X<aEl`eqxZBD1rDgn
z4>rjs-SEc!F3!w=9GwUVxP0u`)DV@V(N{^{=M;kf_pKvAWYV3b7TAZ(fC)AU<RZj!
z8}7k)(-U+EG{D*^${X|MhtLg)(^i>t(&qqCfn0)4ra&ed6=?EfXK1EhFnP^Jur~s8
z1oG=Uzb)&4bwwzcKHcBdsB&fT|9KFM&~(I@U^w5zrF`v%n4EbsaXj~NDT4q|jtJex
zox%uU`DykwykLwTc2K({wSLe-<3I~$NT|MPmh(^G{xtz=I;n~Po)RhXEFAMu#7TmV
zCVn^u1du2}&WFl9pYGlMmp=4gKQz@xh;B6_jEe*Sd7`vO`h;+`^rqui>{1ofa$Z^}
z(0gC!tofPsm2%JNo^=+9e)eJC`{m1WEs1&$sWX7Tf#bkk<q*i52y%&g8Qb`N5YnA&
zLZ%_47xQi3*xzhFJ^XU3x#qXe8Swu3oE`w%1Y(z~Uoc9j%KxzRX?#D{fXlGvgj8I3
z!Ru8`)GadZc<Z;F6hQx}A-B>H@Z#fYy6Xx_vuC&@SSz%YZjXLBK!G-L;;yXfagaw?
zfZ)SL$Re<G0cm)w@Dj2%@}5OxI$}(N26p|x{01UmElhg%EkBQo5rB@dA+JE}FM^j0
z{_YF)ub1byWt0+gz{&cj9^68`2iCdu+Ud?LG4&3>DTsH-=vheS`EI`1+1}6xI*i<i
zQ9R`kT-W{|poc*qA3QBk0ydwQNny;-6A>7A?`p47FZ(a})g6B2iB`if^~8ews~Ry0
zI2kcaG5M!|fye$=V9moz{@~Zlkdv^<ekptzIQvvJ0UP1*ZIc(wgJ-4K-%L`*fcHid
zcv5KaM6wQRb7inm`j`oT)k9?rIpr8{=#>3MY6XZs|LIRa0O{f02S2(+Fx+gx_QxW4
zCrN)7dnK<BWacW9$zB3Sy*G;B07-HAm!QHM6JSf*TE51U^%*>oH=wa&V(kOa_%=a{
zIO|Z2{wEv6|5OEEKzcMvX7!)n7yLW{y(<0|pXoj3<L(dWqd7iF4#}KvkyObxZ=1IH
z%n+EUe7c2ps5^V^zT2zZl%pE&-NxfP$Iy|)1r9_rTdfQvnJ42*jVGX8WjYXc^<P3g
zHXekGnF8uG*B$&$3{CL`Xw5>u#%oDamy(6Ll8r|RbpM=b6>pz+yf|Hw!IIAfg95(a
ziyJU(@xTHMYf_fpj@vxy@tgERNsa-&^AH6sj@btA12XyY5}|KrV3}gk=yx{ccLcNZ
zR5ohmcYpxiWPtIFovO&V3tP`aP|D$|$;HZ017vAM5>78}5Zpl0sO(wKfYf3o?!PVY
zvUN@%^C-$NUA3p(-h*v<2b9uHR(}$y-S2l2_VY4*J^k3z8}ABf!0G$yfd}EN+{xM#
znN<o~;6=!!g~&mX0H-#{d6EFCM?)@Mdys|16hRoIr=Ee1%gIkNJQ@iFI^mGZ^`OMb
zZoUkie%}Y{p?mbKCb5|S-uodd&U$*WBXfap&(G;C>R2pJ@GY~&(VMM?I3(1+k$@w)
z2icv|K1*k6=i0(s4=*+@&U^atDPHt_V%%6PCdoa45H}1)i_s<$nKkRg1+N=2`scB%
zh54W!?EYjj+{QVup!0A64u$)*sZtqJJ^aOn4;$8jW5i~K=GFA4?+Y4RrT5Et0ge_1
zdN&bCFid_(d<CZ+zda+5OU#V!#nLIz#qXVB(DdEhNRX62<C}2luQdC{Z^dk^i+@X4
zHX$Z~XoA*>xhvr#V0$84KM84E%hN!P9jc1OugJAaY~wDR=S%MuYy#AGYPFG#3uZaB
z;HWajd%Ya?ZM&5-EW?$PL4D-z;tKjk`{}Pe_X+RG)n_%vVU>FAeti^sO7H8e3bp-Q
z5U{}l89YPunU+s2@KFo{Tg3#s(c`Acf|q+8Cg<gqkI%r?`Lp8!@`)Q6^U2(wy)Dos
zIPWfghh5`4#oKii?y<@IiQ81xI<Z>TnY?dvzUkWO2pSLGvpdKa)qSa4WAcB~7XIU2
z2297JUNXo?^O`oB&V#E-1s7utqcR5&k!t|SBR(9>GPw3URjD+%oJ}Be9s%DzJ@C7z
zBKQm^;olT~zBcS{wFoT7+_v=N7&K_jk2<f-H^~IwPqRZC3_l;Ulkn6sGH8qn(8t+>
z5Kl+2H(tW2o(WRHQ404&k3Qv~>~h^LU2FsH+ybZb>@xVENC$=CJK=A@wrCrv+#6q)
z`hX*WgV$BF^A?dWGfUNBt9~F6^CN5nIA8Ai1RK7-wT9Yip$P+)d8_MggDe?70J;#<
z!hT-shc49z+q+L&6GY9pcesDN`XS6bMoJj6P2}EZ=7Li+zGT3wa~I(#DwW?a`%(wC
zoREvNsf)b>H^4zUtmg`v^$5EYbr|0Q2G;;%fDE0gfdS^(1T+-m4m<>ldp(BPt^q4<
z$aA%r5-g)MCgxl+Hk<-rof{=<<o-ldK|DN_!o+*4bmI9~$OqrO!gJbY8!)ztK&c*x
zl-V)Iyutx`%5rFKg>G&NA2gNCFabzNllW@UTwpQTqV1$wmU27k>FmSp(Urmv^lZ12
zr+pj4D$0SSR2bM2o`r`K&oEjL`Nl?nM9Q2p)0K`NKq;z??PQ@9b~lY#F&Nc#So&Bw
z6G-N7&}K#O_$;fN0sR~Os}I`x@PRepvzfsz+VYwAEg@^Bo3lY4@F&P{+xkj*B9S|;
zx$&wZ;(K!WI&h}+5e0r!QpL<l%yo(Q0umt~`^C10Aw4BTBI#kM)oPbbeq{-~NHut}
z;i+s%7T*<W-d_nNqd<dyGH#Aj41C&7FdIU(s4R1QcYhFl_P13QtC2W%sO;g_t4LWn
zN<L4jusQG4ezo0Bp;hBD47K4b3hegC@L>RrT66@6?X#N1^&x$8q8?md-h&VJY0OK;
zxtT!0kQ=sWpEa2(>XcQ*1<$|B%?P`u!Nhq6LP-|0)_~x$?^18Yv+WYivjCHVGp)2J
zetJ^I*&XkZXq8Ndpe*4*kQ_37K5dd(e@F1HIRx5PkVhI#2s-ttvWEay;L*={QJ`dK
zq)!37haR$5?y1_$Jqy}Qb-221Gx)<=N6=t!!1@4@DZce>`}-HX-CJPi^0iui@LxAB
zu){K9aH7%yt_B@ZU7h-XYvy8yf3ET)AEMqXSizj6EgnTWWMbHsgu&)7?F({|m(4rX
z=&>U77wi3SX2Dh~MwwX*){^2wL1#j<JA=$bm=%z>3Yh_Gn>283cM8E}mMzj`3Iv|1
zx{};@gQe>(;EZ#`$UG~n00RikquC3$x)b9n&fq<1oDgdNuP^BDo5NX6#c(*h39+Bh
z4KWVS7@Oh8x&@goLyl2zNdg6MJW1d4QgwS-GG-pKxq;^?6i|Cw&!}u3-J0as`np2c
zDA|XJBt4jLGoch?dhtzyvon@$)2-y=qH{E%qj}<`Ua}sg$HC#3<8BQy@iWsCCR_k-
zHpjaLaX&bVRM!C!Ik!>bv?!IW`T&q3NkH*1HjPcGfbAsSf6%7rMK~-ltu|`8NrOT2
zsJ)iuOBk(lI7<0v=p~_9B8DQL2Kt7p)(HXD+qm-m<rq7)7oX2d(v+zc*h^}kO|nRq
z?gyKaVqiGQ;e{pA&8FdupgbaLBa$%!SOyn>gjk6Z@hId<+_fy3ZWgs$-w5=ub=n%D
zFb;Gy#lH6QS`c|8?zTRI^D&k`rfvdeMMP!${r76rFzfe*Os00QtTEN2^~D-fNneAR
zu}L*Kd$WyVbcIW|*iuFNK!RS|h2o>gA_jvz-42dP_GOw5EnY1Px~%){*zHsb5eaif
zZqos{HB`@l;Fq5HN?T}g`#Bg|T!pj8lzCziXat9OhB*grzbp?dr<JpCHo$T1C-=RN
ze+!7|6WCP7@HWedndU&c4Fz10<(f1Q&Y|+h_5o4ZfPQ08FiuwZuKnB)u<u>nAm~WO
z8!PR6VU@NWeX2qIbslqa2~2UO_Y^MQO1IlLCJw96FnWRQpU^+xk)W3lqHmI2eCHmC
zaBtqbCcz0qAf51gZ+`;|KG6Fmn@;J3IF4?C&dwRzsUj?Sa}yX8mEMEt=btmK6`W<Z
z8;riru_!XMq_g8xc&6xjWsmZg9PO_sC5rw@ZsC{(EIR<fZ3qs2?2#yr*(?4FbF@DE
z{N2zy9e5H8S+h2itG=`A7@0<*kn7l)>s*{WAnN!aN?UUN?X>rAV_#_U%ggKto<LNK
zgWz2~>vzDzt<;L*i(1cEy_q1Xrl6dbV42aFE1Pe)C2Pw@6-FKUS<tyz;A<<)qzndg
zdCn(hvxQbc7j_3JuF+N$pR({DXW?s)b_*TN0bn>v7>}mwrjcOT6#PTiRnrQ&HLJ!N
zE9Uo1ODJ4h*vOnY#Yz@Yck9{y7*Fc=_pT9KHdG%&-4b3CaVn#1$ut}(c|Oj+vI=^)
zoZz6LPo?pb9L(7ktQwoG8gtP*BMg=nY^gAAuWtt$`eb(ggH4u$^P#V`aBUq&KJdkv
zir@1Zb4V@CwU;79H``Un0b)Y2of$8aN(IKE71ohmYk$m!_maeU$#AL_Kiz&YzIs*!
zMpne|DK&KTAL|x5%C@JkT!_q$IF__a*5`OdT;7eHMo)<lP(a8jBA6_+S|QJ{Ue0Vq
z#4E+^E8RE?+4y)Y8^MC<?ZgO0aS}tEvM!Ro_Yo{v5%01LifklWNf5NHEYYIOsro{r
zjDFYB$$_N%W@1Xth4=Z4$LlH!-fV+goSxn@e25y)*BW?t+`s%ZP2aw}Z@DFu`a_AQ
z3S7f6mhdJpt6f+S-@64!oL`0flHc4+^{+_eE}u;CmmIxP->Z-{(o}rL)<>T410U-^
z2bsFiq9WZUz8w*9!MxC_L}N#$ftM*Y;aWGft!6zdqq2$|t7S)+D<k%m(x|@QSKh`(
zv?&>;PVRh+Se^sV!}mSoeXj;b)v;iSEy2QidLl%v65`P7)UMor=X#_`yxoiPC%c}t
z_R<@BVirSpDTcBAtp??@BW{jyXGi5qJq{H6uyoyR$1W62!!PrGfMNN0MjOiNgRW~r
zkL4bEX$&`x&5CAtKEeuNo~41_+%GgB+vn|#7INB)-FVG5N_NBHo*m1T`kG(oY_q*P
z9sME9u}uE>I)9VD{l`wrOo1`NU<D7>B_B?p2#;dIZkIs4Sis7_b79LbWX>>!lit;f
zGyl%;Hrs%SL)`d;N?x*=8?V@wx?cMNuEs?CQ+T@l4I|d4a4LA}9y7N%(1vAmEnoFJ
zonuIn&P0IJuMk5^r(7VCz2<alQ;7;s25T$03)yr#iEw0){3+4acjDbNw!Y~SgXkON
z2~2|uu_<oN)O_snrTqKh1D)k5D5@M-sXhIIFRGKGi{D5(wRj4T3Vs;pyUlFSS7@NL
zBlx|h;W<@N)q=p-aO2AgGm4W3N&K-6$ZN*#Pmz>{tveEgyJOCsy_|wI*E#$$rd<1V
zZwk}_S79Pj`MjkR@**pWUwu6#V>dC0n;(K=+Kz>?TnvmauV)HmtJ~xTsg#}4nM;iH
zLN?jOf@b%PjAfHPbN?EGOv~2LQ@oBJxE83}SQ2d7?slgn3R0S|utC^LUqMFI8B$LT
zaUo}zKi-17J$Fw>gS>0hYWY*vYxx1ybE*Vbsb-1++2&1a8ow)AD0hmk)S^h16o1qE
zL(K74#6QYa-Vm2UB|XzM*qwFj6{_${=JeMuv3jIX(^>?p2~T$sdT<G_8W95_;iW15
zcLqbXqDF*&<f(XMtI@<c-iZcB2oZ9UMrL-vM*M!I{=Rm{UyKX>v9kT+>v~7#SS0sj
zgCT%Wl<-;s`%-tvSkvZX`>7%red`@AiqtJH8En5|wJ~5zSYWZc)1CC*!h;v7k^VaR
ziQg|9lhqn*$II~%s@@~2sM!SKN?;q;oYWxs^G#4A6sPvQB!Yp{qfI-D^X=0?90^Zr
z<{Z<ka#Q4?+6Murbe;g)pkgy{Uf#9vQ|IwrY6jjUiycnowB`7W;t%ZuDNXoOEVK-?
zMK|>GqFE!wdahUm4;9rPmDh9L=tSug$5iOdPd9AaaUwcFN`19&zrus#SPSD0<bu$u
zZ$pt0aT|jrd1q3Tu$e6{M^tXZlLA#_gAm4>x|2Qvo7eB+l{LPRbv}5_Gv)=(kC%`U
zT+cX7+O$&ui0EVZMc{1|4_AlttU(nBPH_L&^vb6#u19|`iPfv$*@rWv&AFZvYmAF4
zcJZ5xg>|8QjhY3UI#qeCS)4h(-F7;D(+f=!G0R8BF&rK_Rv*7Ii3|sbFOaJvHOpDq
z4O++sEut+7fq<O1>~`xa@NroXRTp}tm*KtLR_|PxK%T>gG-R!Nbpd!woM3`Y_C?%G
z#dmWZ@0l~F(`NwwcztdKrmCRr7XPt>{-a=7ttN~e%GQW*WM^=rHUuwb_4CGytZ)!8
zVKla*XN#lSSVVrr1lBK>c5{TPC!PJ^Tb)>fDh>3F%M9OJK|+`ZLyXNQ!5=j2=`OGG
zlhE-g3<Fi1Vm`ndCXpNC^~Z6y!U2Xhk_>aq1yc&x=M1gZlacbZbug>FYPGsMixbDz
ziJ4iI*(1Knd9nDDB0BGPg|cS3(gG?}-fZv7r0z&rQh)+RzWbT{+fCyiAH;uteOnIG
z`c8Q?EOt80BH0vahMC0^4963N$EFCC`hR-@X33j!dz!Ov^Z4$OYB9bjv#?VSlJ|z+
zB9@0)sjt?&A&DmZHt4DfI|OZ86~2!6QRMdy&YZ-{Y2^p}j@HPUK(7Af#xHAP!KC4M
zr0b0E(%gqqgm@9Q2Dh~+2_=&vDmDvC?r_j#R7^Rfssw#P@0!VuKbUXFF1I}6NLDN<
z!CJKx>JQNjB7<F7fA6si{Db!ze1fCVInPN0(dNH>gZzMPaYghHIf{`}w!%_f2N$I<
zyj^7#CK&prle0AhI{SUhv9DRl?G#+~Z$JuhlNA%;$I`uHo|awHt*m_jx=-VGBO0X`
zx7y?6@L^q}Cdr^Xi8OvEVVP`(!^3f1ptC%VAhIpUlC>b#V#4L?r*|}=SqG|g<)9P2
z)cz3l>UYpVR!0r}YE`Xv>LE#=_=2!BW3%!%+m3WMZ{-Tt<e#SZmBVsjr3Dq~?#!BK
z*3}?`8bb8zVuG>A<_iNWjmjJ4zV}BQUMMC$dDa6l0iY-qO5?PzRne1-rd11T%EiDF
z#2OQW5n&8+N|K{(``i9@+mXL-dt5a{jfMkiG{b|Mzm9AFEvN}T$%Ym1u$j{WGv9j?
z9>heCVim{AVk-)e0_Z)4x`94SK_xIZwrzne-#sfrw8469xQsj00O-qXybKg!khk_H
z*GzC|`mnUL0h$2rpXwrv^K$Tj@9oGN&pY(cWF>};5MFL9Xr)g3NvNuq6LuTx6FSF4
zS^Xti@SR~5O$99WK!-Jkd4Kwh4uI$w9nUAm0dfq)4j5bWE0-&6RcB1qFX0+a@=W{j
z*n^s}cbZUlAG7{@`9AcGm$oX-7C)Z>EP^^2n806-TZBB}v&@4d!eOKjSdLkKX(aP9
znX*2+LjhY2eM^*l#wf#*@HLAG#7))fPRtiIy}C3Vm%(Ogte1cS2^Arrsgru#QJUoz
zUcbY_a1=gdq5s^b4@AP_tLs=xY#G@UR~m)esk3~_)#RMgZgnfi99Brn!Y|P%dv_HH
zVlW4ahnR`;>{sTKP31q0Az|w`-CHwnVs7uA-Y~-ecP<Oys5*8p5=7nul_L5i?S--$
z$!|-HV`BmRAt-9-zy2Xa`g5_;6r*HPFV*+qj=ZYq#c?O-NmF4-&HZpVXy`zmA<}SH
z61WCs5xr72tU+z2%F(?B!<Q_4KF|Hh;H$kjmJ$yg?h|Dj$gjF<{D|kN-IN!!5r-&^
z#Vd+qKEl`6HU@#WLaPMuKXpGdmI6ppI`^u3H?{;liD9D<XzJ5KhDNo$>4kxFM#j>c
z2-UNzhL{z*slHm^gaLRfP9#A@>qzPLV5g9yCY+6It`IsXia{A=dEdRx;C>|A>CTf<
zy5PbzW&AnXXPu-F&}218BY_j=vh$Y2=eoA#<8%YINVe^E`fVS>L6K4C2D?@}e1#>s
zW7TeCTr}I|NJ|4Ay7#rkX(gUF1VF<uWz}JHK&xb^**|AX@zwK7zNUxbM=NB<>@5<h
zd&iOphZC=SUq&LqU)TT0E})E&+Se~QU=!!EFc2{oA`P{z*EppGs-H-!z!4QOl?UD8
zsNv+Zk-r~HL6tXEPiWyYZ&5IORWm1X3?WHJgW&GE8z;s$-LPZNV8hvemu%I!^9#VP
zG~lW>M(@cZ>|WeFEk6)U??L$<)HtYDDxv)802?wmy6XFVeSn=y8TW?%{Hj|Z^QsQr
z{7I$*3DXQap5z9tMsM&NsVR|gL9<88)xa@-k2dsgt9<y6;w~U+d>xY-lL_^~_)icE
zcz~JgG-cgk2+8@#_M4Em9spF&=JI)P-EkJDd+b%_7p~oUVrF`?H&mp(^Dgl^lD!JG
z(~s3(i)*2@2Yc0n*&u3)Acs!wkS-~VpX#)M#P@s1{fTCka0%+#P6-@0Xhx;I#eg*q
zj3VRmKLHEO+3>@kAV8s0CTn{G91+qRjGJXlK;W#Al`BBTTcHQSZbm8D?JK!n_4i<R
zVl~*X0=)*-K3dY%khWX~Zjii?CHkwPzXdP@S-frzlB4STIAJYXJ}WxCMGcil=@$)^
zbw^zeYKlo<)M5pPjrNUmYtMa-ZP_7?n}-0|BZoVfvZ(c?z+}z9C+%{RUMb1t_BBtL
z2_w=CzPz*~5@J>m1O0Yb_^epW@DI+pJi{9M6Rr>7jIOMn3VWLFoOY5@!P_O+uMn(j
zF{{@cu*v2}sElD6Z~ZAe-vI4O`i3i!*3M&$X_+p%8ngx%TLs4#YzFs;F8!i6JQn6g
z4?l9!#%0=sjP98|`+iuF!-$zRb~#|S?=|iwh-_Qu*uKIW`e6r7k1^iKdxqGaOkf1W
zli>35Pb$D|B}_tI{_ih&G34}|(ZLQCxw%<fx0YsQIALFa_tdhBO{9NBfLhV(5pDfm
zq`O_Z`4adUUV=<vRfTV+9j%q=Hru$5gc_61npUO(BWd7y)Q$&-<>}%fgo(=Rr1Tv6
z`hL%JcOs&Ifv}AFaCaN{I<V<}Ug$JtS7oda8%M-}<+J&Clgz1`U3x1boHe7!JXVZP
z6u`7+6Q)ep=+wtfz0`aeXJ<$BahtSXay=1ienD<l4Pf7icEdPD;i5Vw%qDT|`@1*-
z*pX<mOaeMp5y<g8j7v)96RQ-nQu9>=av|K$*|CAg@IFt5Qw#O7;}}{qwVs8Oee}vm
z_a<0lk&1yTC34kzNSW!XwepS{7{y)qAU|pB7V%6ZRXwtD&qe4uhAF~>Y*jYg-ER4d
zJD{t5h4OUlib%)J%6gD9kzjCDiBhL@<QfbB0Z$R=-FgpB*dso!4&=e@(Nb3Um_iQv
zUPfd%Krdg0!}$y!&QMxoPU4xXHlYWY!@{b?mtX8gZH8=aaif;gaPQj3wgL!Uq^ET)
zKEIqhk@W^oxzu8^`Q@FbaeLit?Hp9`T4nc?k8jmunHZk``1%nYTotKXI;vu{vtTAH
zt$UzKi5_<t2ashp7yhItj@hl!TW2WXV69{*J=wt`YSA$POqwQmbE@O>KwcQ*2lcwG
zmk!NY{0Cq#{~T$7IVir6*r(u$=t>x?@GET$4QU7D)}elLPt4``8r+$iOVCud6+%{b
zd<;IPuK#kCc-<##rn+Z3*-rfn#>UkN@aZ`5S7<rmm@8A@kp|{wpGpdXJd@x4)eFG;
z;iBJ%R#kQVOL=F{L1I&7Wzvn+g2EfmmQroFc{ob1CKuCqXigks$&?PKuP|_O>a>_&
zk^~Qg=5k66kFm0ZB@mrEoN2a3uJB|`-AjEluF*3~6BKcI*=-aV65u#sr!(!=PNO>(
z9-M9esm@*!FX%PiJ*G{1`*KW$e@ChQr(=z2GQ6~5%~X8{2Q^H7yW&b`<9*%^;fA8e
zmuT)?W^*TC<HloqfM+?ukt~#Fij`3OJ&md*CWSoNv||YeafFR(o;etq)^`UK)F1;&
zLUe_lD3mN3;r8G_xboIvw9NUTDORrCLnt@Y3`F+efmpW(V@@=s`=a?lY%Q1B&;-Kp
ztDR~(r41vd>Zz3>dwFU0&h(aTESz<x+YLp`$F9PMnYGrymMQx;8-cBS+YQ43p?e9D
z90JClF%7}24Z(a3I2BFr3rkiu9s|_ZPHGH?DTwImi339R(yS%3lpjNJq%C_7Bk*0d
z580Y)J&_2*^0N%!7~Xcx5WXEndAoDgY{n<`4Ud->lwKJW;n@i4Fzsyx(%aWaZ_AQF
zxooDGTo;DkTVKzlXIp)xjaf-FzAbEgRq$`Xw#q=e_pk}#v8nyGbtuifD)lngYj{ut
zAzV!daUCzA9H}%xMp0=44vqV<@LL;RvjUm+VB(B*?a}M)3`SP+-*|z#iiC$&gEpc!
zs$Srm2;23&gEG)Q;UX+PPHCLm+>9AztgI8*7Lo*uBia;6FJ-(<-|2i;A4L8J;%l9y
z!*LJO0XK981h5bb>oZI%kMcHzE61Ixj?|YmQwU$|nOu<H2UGY`Q0=^=Q44<Q)Ye@^
z&pbStNmB8Ilix7I*|u@58-odd&qE=7c2EQl!QK0+dO2b!Jc<mTeFMKtwqyKEFu@8k
zxY(9#53*IAFwGi}P2~+x9tLA)Nrq(jM{?r!tJ@ZXxwTK#adBr;X%~3Tb@JGt^bc>4
zZFa9mVgPVb>Typ;*oe1IZ=K%8BgaD=CAsxg2s9}1ZkY`Op@WoPT6j~YDz&-1ftE>d
z0}_uIU29&tkSis1M_DMb<b5cU;l2)Anvkg0oWw@$GhU26_&tywXvAxi_i)H^^$vHD
z;lmpmYgFzGV6a)?cYB`l_W0Tc5>0ABv=9~HE9n<6a$`fK3gp*n9)hVhgDrDItlLyi
z02@d?woa=~dowzwI1t1J+eK@=AmOx<(Oyio;u;Pnb>R3O+P)Z5WA|dWR(!06b45~O
z!UiE&+oj|Kc*X>w!1)ij8?_rD<h%^S_OEbS*Y&aGuZhT^hgc4Es?O_Uz+{^L&eF8I
zh~?YiOtU~o8RYp$FqPH`SjMT<+{t^3rkHXvkK=i3jm;lzyF@tHIo}<AT3B2wZ7*Ra
zvyzA3-Ves8Qg_^DrYpgj+as-?q;@Ape1J`~a^>fm*yvSVfFC1&<Hxgc3vYe?FIo40
zoR3oD@xs3`qPb3f2DuX|r7j3j+S`F6tMp+9BX9InxX`OKKzXJTv40(jNh&_F9m*|y
zi?@-XF%+C?d6HdYHy3#PMid_}Pv)Ts-}6sLxqqOnYP`%53<@L4^@jWvRBhjJ7xApP
zVRxD98oSJqsw^y#*1MXd_jZ4@oNGqdyC6~486wP^%}?+qTcYV!XWv~@Qa9t}xHl7k
zORplMUKLISX2=_TJ9v(8M5<zGy(y2GEbN<vB6dJENdrLRieRK9)KS5S2EK$jB&g{H
zW*&toqsU>B-g>>dW8FHW*Y@%Y^J*~SZ^3;|djAZ>{5>-xnyF9O<E2z(mNx>Psb|13
zLqvm-_tE$C+F05~PQ&{j$xRfV<VciglyK5-3fic07cHFee~-o^c3QASbuBJif#Wb$
zIA}>F+dnD)R2WsDlAmNV=5>F`OTd(GutQ+^gJ))w-y+jz-`Z(@f{B~;*EC$6>D{gC
z{?*aDa>sHCO*`<x+sV8QJ<ohD2XQg|vX10sWK_VeN1pn-W6s?0yRqTuFbLSU;SOCj
zhU&Ziev<KyfMv9{E+tRZ-gixHno(Z0Ps16Hm9!t;V^4hXF-h>15s-XKl5(W(4f?+c
zl)PLA#Rfw3qKpG~KO?;s{*c0Uinn5oL@SKjh=rS3SFl|MXc9=#!Ik-9r&<do1|>uq
zxrOh|z9G*NT*FO0o$z!}9>HF~GYvq61(9joRitnG_W2_HFd9S_$ULsxl|I5UBekH)
zk}|6D{K2!b9rJt`z|43KUHrvKd)R6FiZ?$Uy?o~fnw7npl&R_jWz<p7oa-prxXF-E
zT+eD~z=^e1FFMQT2~tx^ib>o2Q9YH-SX>5NDot14kJf@ep7fovqw6ZksjO(&ozH36
zK8@HRwJhG@qdW2T+e2}B;kXF87e?uVav8U0cbV(@%e?8*qx0>42rKqmXrw=vaVoeP
z(=2C5`c589DQ^zA#QfPQ5eGBev2?q&o}T$C4h?mMd8T=|i6gD@TD<G)Ti=7f;}JN7
z55k`@&dS!KzA;B03VUhwp9BXu?$4GIBgwnX@FN4h6|qPMONj-W5{j~gy(_l8BVerK
z7I{7FKz<x<z*O5LPKexuGm?n~^L@r+B7I;{{DS-_Bpe?D*2suW99~sTB8-<&Gof|(
zG4mi$h_R@Wum0r^!Dg-3>df!5ToN0DE1N*k>;IZdD7y17{DKqJIg2Ta(-R3@gzD0q
zWO5jPY?LEAKHP)Nwj8k7w^Ym?zoDpmb!%AlHC)g1%n4os5&DyAWyeVc;Lj(Exv)f=
z`>r_!uwb&&u6+~1vv$<Sm`!iQvhi<|IzOC~e)BR!4%-?e(6yTMS=>aaB;{yrs~Dty
zJMMR=VV2pvtq|76)cx(n$Ldg*Odw<)yt8;Mtp24lf9{z}tyS7VrVY?MlUhz{3Hz{=
z%6bPL-teluUZfkdbzUZ`esw{t_EtM(yO@GPj>x&=h;vn&i>?sijc&z_Yrr8vwI^q_
zwovCGxl~<*xNS~ugj~W{X>N(UtdjWK^DM&b&Rb-WrX)_{(c%R~_T6<Gnss41qj^PI
zdyGy!K;jY9KdH{0rrcjv!72p?*Vn)(4;Yrc@9COS77xL)vs~dFZP!npC&6GHTo4Uo
zd3jePZ@bF=EFd!h6+Zn*Y}7rWRWW*#v|BTpS4YNjOyX#@Fb`ieR?AAu)p_7;cpIXw
z+~&kihcK&%$yRstY(#XACT8vPC?)V8Qh^Qg@EP;`8Q~NZWZw7qrMrTId=R2wBJXrh
z`F<P?A2vMikud^;EDKwV^iJxXtAFF7`2+T*8t$J^H~%Q}rr4z!t$i(=W#QzXK>Ack
zWoITVqCSDyzmkVD3pkN?d_rN*zv0x;ED&A!cn03)XG~HF#UG=XfL=|oUV6_%$z}em
zNrqTyOxW<lv%^*yd#;oAayQN&&mmq3zbDWwc|rSE=y*OBGp`jU6k*3>o|@6*oX1Jx
z0f>LUxv#W9v-73GTKvPWpX`9wdpuS4==xu-n=4?t>xX5m|4T6X->$|#zMzW3hNz;y
z!%hJMwW*20oDd}Q87G4h2+>Rsg8(q${c%s{<+th^-&xXNm~{YlsvwVm2%=_rDVCRa
zx<pJMHct}>c^VPVPj|xwv_!X7ult;T$)tUA6JWxHuljxdm+Q}hgi&08bi#~H2-%(8
z1Tn9ozbTcZrmJtBE12>lpdi}D>=@}$&ztfCF+OYNW}V4=W~tVB)^tPNpThk6nm;WW
zbL9Qy&F+MNTd(_Wnf`ycA^&ckmfQ#!j;|YeWMlp`-}xUe>)&4o+yvjCNM7UMzawq`
z@n7r7*cdUU$$T+9YyYGO|EF(p4{G(vpVs`2T>GaM<{#oo|I6p|c)|V4-z4}x_m>;`
zzkS)?zvG+B&;xL%U+PaD0A6bRT5!sq0RRG#K)0xQGN7)r&BgCs|L1!b@DTrYs$d&W
z`m=5^ug{7@8aSB1GfT}Ktc)gr9MI>bK7HS|sURhf&Bg=Xn4QN#O5;!Xw#RInK*q~y
z<lM$;1~jn*AUWG|>L8|>#=d4b&wrEjpf|=>&>JvCsUmu*T+JZnI)i=P**@IdHj>7A
z=ktb-!&sTG-zlJDUIxq6&K)d@vq3H2?Rc}95X$u-8e^FdJV~%E%(Ss*+DiFAd`)QG
ztX4h8QD@8)Ndk0F-VRO_;r`ogA5-jypk0Ev300>izDEm9fZ;cATuev$PHh^bDH0@5
zR)g&0CNqF$6|MXT|Kbq)tCP-cobU8x>Vg=-cAKg<O385gcj79PDhkc|%|OySy<M-4
zw*sRI?481YI?(?0^G&boNtJ0y_kmvD@#ovHP%IKj$K)`ylskA^22A--i0=Q3z2<Zj
zM1D2_PA2U`@MCU%?D64#4TwI*>))5#<fZ{%5YOvZrJg(B|C~Dc9d($Di!VUht31Fz
zdLRet-oN9qKky}PO@m0YVUSupJe)to6<Xe)B52Cseko2>-oY3HG`&W^244;%4FXJB
z8A3>@ea?SI4-ZHXnF5Uvh=%0}hBVDlbP~^HwN}^dbe7+r#D`vx0|x@XWT3xl0{YXb
zXQm8*=KA5lBNHgtwM8Kk<D*Ef!JV`0J^9?iV~6Lbhg&ls0m7M-kz$1#h}@4eH02oH
z=S9e&1@eKAp*|-RMA3R(e$|`3gPnY@vFa;$PjX1=x}-(&X-c<$zcex&Q*Z>*9gW;s
zmHU&Y|IW*LaLInk`-90*2)K;-sq8vGOMq}&U>scL89-oP;lS3nJc2mX;S%@754l7Z
z@K2Yu{4RscN)L6IJJ>yf3An2n8MQqCE7F9+pAcyRW+&Xo3Y#xrrBg0C9I*IhOPmY{
z!=|{E3`J&}fya9X_|CWAGy|W_y!{sc?Nj<L9Qf8E;D}FqAm<>`aWya+7_YgvFaC=p
zTDCn6oTY6cQMMG1hH7M)MowWvZ>qO6%d1I%B{j+6(veXZyYyx-L~UZpaV>nkt4v2;
zI*JDAw|m|fR-f-Izdui7<uWVYgOmds9z9-|m_Opy{<x@RUycElP>S&|iAEY539y6u
zU~&cj_-9#)x|krS<j2zTmpCo3!Ny$>lVkmJP57769dK838bnb)<HH@S(8qDS_7UP%
zN`G(hHzcnbeOed-MQmSz-476<UT<h00|NB%a2PL*5%S_}ma<Ot<|Tjq9dQr@`~b5a
z1a>kOf!JCzFp_k{CRFoV;4o>MrTXbRA*nVENVs>a2HIYQZ2;NG_#<{q&5vzb)1Y@s
zjN~Kj*Hi_F;PTV$ew*_M2t95DN$ljl$Nib1fD2Z)fjbTWGc&;0dj;F1b5_&yVSn34
z^X>_oMylW}AmsNqGmfT!%o7;%X(G5B$@k=GY(YOj#BlUYIaI4&O<r5XF3G*HDLWZ#
zaM_0oH=q8=^TnHsZ(H`2;&-&N;rP2NNIAmJ`}kq&H^Ba4%9rFD{U~7hXLbxRr?M!k
z5%}$rK~}?IDR6i9+-yTZc<wLOxB1wSy_=mg%rFpW_%ie$fYngp^A(uzj&Nd0>P$FK
zwjyM{{w{sQDD{<A*Ohl^=rp32-Zc1-NC4kK)r{<4Zqq-1=srYRpGBcAZaD!BY7STD
zp;Yq_PzLIGj=qOUlTkddVlr7F;!vmq>7J*D5XXM71_<Q)+X~y4fa;|ZIlUn<brj7*
zAnr8VEPt_h;kWkrQmxU~qYD|)hRX@ea%zqQ+T%eKZX_AgWG{5^u2fW0uvbfvx$)eo
z75$nk`*up+>PFv@%*N8x$kfQQp7n`|XRpT7ychQ_d+mRFopR7ADUfuu28NYd$$(RS
z`GXVxC%zL%YjrOMP=3cgyPdA+)m;|o<h|9j)E}ZHebn5s3&LGO0XenhtE%JlnS0u;
zU|rW^pobbqX-1VH_W)y(iVYBcM-N;!LFP59H$W_xZ5X}B!6mqRl0j@W1%*1_VOZe<
zo&}+R>n9FLy4Lp!4WvvDdX*V+i)wjt&x4i8f>Sq{5*#kh&kQfe3>LsPLA>vWN?dyS
zll@<-zeEA?{p3XX35kCt&f%0CpwiE{l~M9qtj&bPNVim*o@(*8T|3e{_G#mIGQy{}
zSXff?v>HGMXP?6_Ht&zx9|H}<XL|RXEPC(OP*O~9u%{72HkD&sl7#mYm71w?ZKH(R
zJCgJK!0$XStx2t?bDK}GX;zO<DH7WswDKuoue!ceo7DN*i<B?5)9YEE=opAwwCzM)
zS|n;r=g*7=M^Ag3(xhiyLP=w{w%PZV``LLj!5tPompaMRTqJ3sl$1U{G$+CE28!%Z
zz-8f1D`wCVU(T8QKr^ylge$v9BCq#~{jDN31E9mlhf%Oaxw{p_d&Vw-lE3kc+H6!c
z!4(`Fi94H&n$CIc#WKLoE{QEFS*RveY~rfTb3DD23FzA0G8d=aQD1P?7o}<?XZaur
zfuo_O%xiE5Yqco$h(-gCPC>N#W+mO*yt0(xbCCExX=izvW9nV4BBpF3hE8<CPrYH2
z7R|dWkI!$k=TAJ--H*6An6mlA))Fxn6pQjkr+xvFI7qyOEANkLG5=!Toeb!`XUXUr
z4`|S-e!V`YUG<}{5(YDdsgBugZrQS<7olm|f$HMCQpA(@51zczW8b-M5ybG!3?&Fg
zugAw(BU0(3owgZKUV~5Ozw()1-UuqH0~FCoz@Qw&v97&8QiWRhzLhKoYA0WIx~}?v
zubmH-{7cPLzbW^E1I`$)<8(c*`{s0>kePt{htK=YPO!9gs4ohmQG_)raRuTD!zm3O
z$CZN7!LgE0KcPf*xaka3NARYG;T8fm_5K(kVqmtWX$DB_+_f#L!j`1uw9em{W*`B5
zmx_Xf?K^D};V!V1o|FQ3PN);h?8*%9Oq-H8&60OxAklOAF742pA&O+7U0<LC2odI^
z=<jp&SR^hp*i$HcQi}0V5{7k5GU1n`QI8R6Z-GQg14y!Pz0H$t!m)>wpY%TmC&Pmp
zM&X-k!P1|=?{s$M%o;Fj@Sp+HZ^=(_)#Qq|m4I*{X)2tV2*Ut15CMzbHUa#o%5x1v
z=9XX*rV@OQ7BA%?c)dr)?pIAISb2)~LZY5$xNWm#1;d5ui&#WP@T^6)MB01W>~N(m
z#}?66C1Y@$3}Jr1cCypDE4Bhsncaecl%YSQ7;&37U?{|skoQC`WlJ#Js_Ol{J*uzh
zaUcVN4jk_z-dGa|pOUU}qdNJ`(p?9g<TUQmy{F1n#H<Ioxx9pQ;sYiTU9zt$Tdvv!
zgZ`}Y9Ti<>VYx{nSW-{`RTZw!%}OkJie}1Azi>N1E~ce?$XG5m7bQ@GCn%9F1F7Gk
zv&-1hI;1_kL?6Dljn7B4k()`$n#|ubBL{Z~1nZ`*$EP#$+8P@=(<7Fz|GMh~l5O)d
zSrap&1~1Z7$OJ(eScQ+-`9_`1T;1sf%q|K@mzcL;eXl!fN2<VtJn=)~GoaCVS`7gC
z&%fX*0&HdJc)_GO=j(2D&*IiCl8hI9aJDFC-!u_RbSnBExj|dv447kF2F^jqDuU;k
zm87f8!4HBj@`$b$<9_4ef#n#xP5HP{b~@8Z$X#Tuh41bZ^dGTjYOzlht|4@b1j;_)
z(z(RO5Un)n*W>Wz3(+Kv)UcdF(bkyzf#};iI7k17y!VW1y8GG&Ls39P0#S-6Rf8a1
z4AKz-C{;uhklvBdL`o>q0*0=F(m_B$dN0y@4Na6PARwXF(3^8|&s*kw*35JNzt1;y
zv0_NhZ=b!deU+l8ZAt^I4S!EJN~1nVSY5|WUmRt?n;zjW+|K)Lr+&n68sf^cVvbq_
zN$ATiqkw0!va)#6lN;C4J^s3T=ZCiD;WWn&!PLctO-o#HA1bDM6c@uy`sy|01D<qv
z3fB}9CR`3evXAcJ9pRXY&%DQOyl-SWt^vdKNOAl8g#Y2V{{O+d*x@9_ni8{2^+PnJ
zwpxn-*XP@#uWlv3{UVM`TY`gn1RH)S`gsHZzngB}TuglL99yCWN`&+K{Xee6_4Yvy
zdv2uW%po?ycB+k=6?v6PeG^bFpEI$PqSR)4FKSwOTu*03xy@w<_r64G09~4MT2?8b
z1>PfyJ?gt6v6O<m*xt?I=HOqTS)ZM9xpeP0gz-^Plb%j9Sr_04{g#&@*Tr50#!39s
za=i^?rffzKBE^C64kxZhq{WMil+~lev{Q+rnGPYc#n&7x%=#RPEs1~rC!fNU-d;0Z
zmc#G9ZU7`W!uW_%UI5`kcY(<07dYP;d-u4`kc2(41oXR=UNQImFx)J)DeSFHr5v2<
z%Z=xW(iNsrHCEur2;BMHvyxt_+yh<OuO6s<mq!+@ImITgLc^zqYo3c~uB_K{uVqla
ze+vAGHllVC%R(kKYB=EX2VDoBS;L*l^*rc3fz_13rz7ANUOhmW_(&(`Ww=>1^t+iV
z)K(~Ev*gWWy6@agnx`H|;$Sa#p{-mee2|lbUH%?YkMaRx)P!tbirlVGCq8HSnWf(L
z2{E(Fw7fXl_RKubl$(x&YalJkdHOAff&nV<AjE~|L%b|>cd!B4o(zB$fdtdOUT(}?
z%C_kJtdrdvkk6BGrE!Asnwv>sejjAstP!PxhC3g8Q#LG@n+QJw(WJ#!2?BE;UUEE5
z?<X$%9Ri^h>}r9+q+-LYI-gU%7k2*$dL0aTD<0w8sUIv3)w+2~_S`>TLAR|M>k124
zP#2pNP4YcEG&JfJZ;c&Q!LOCEn$%%vK3$JnpbJK*puuk@NZws`eM+df96R9l%UOAU
zHLf=ccM{#3H8huVM7|+d(|3b**fdX>nd`Ht2Gsj5{r*wJ-(%WS6EMl{7gtYtt_0~>
zC;wa%C6(I7cZlK&!|%2m>4dmS$QGCue9_!8e{b=8b2Ti#DffmEWa-(C=`b@Wi3Yl0
z*sZyov3l+4@Lm?_);?_IMJwj8^@vUDwyzji86XaO&VUKm4Os1qRHtStq7G(PoU*Vq
zT)Xv+huW9xdN6Zip4)Tpo>)%rUEEC-4)deoy!qeWaG>q<XVqv#{Fw#eq9ZSPdRG{o
ztusRU((Q$QN?Rvt4oZDM!W57bHwtQ=?#6;|RwUu~mpN3gq_XIJZD4R$re0j_jWWNR
z5L<1N9%y5hRc+^8tUhEj+h~&>W@Gv=BjnfedU3?K4Nwcrez75Kv0<rj9mzd1n>kCj
zpB=HZmY1@2Wc7QoppK(cuVPdu*Olz>w>R<higyxDu<o?s^sip2@Mlr)WU_T3psw^D
z>rU>b3*tF$B`|i{q`A!jrPsPQifr65D4Ob#bbq>a>4I9-@6+&#MS*y~+v7@JWi<0P
zq(5y~CS0T9Y;GiswAf6&tyKqe??UxT#tLJ5HX4Ge@r2ihmt|1#W}xq_Y}eW^dKsrr
z#})A;{Lb!?PP!$jv{ADm!)88u`N)jB3z;g@*L6FU;<UB3a8L@kt;B5HICe<okyJwa
z+ixoq43-cb7;(De_v#g0Y(#TD8`3Zv`6wF>gVNZVYwi+C{1)j^4+);j=!8!8gsSfy
zA41Z&=1>-&MNV`0RAIu(<3FSXl#VVVL=%YwSsOFdhD&(EiR*g)i!-6Yl*%{g%^rrd
zdGl@+=J*UI0=lNl6_rU<d4Z6Z!^-2*Er(Y7e3>6(DJrNxSUZ&2BcsXqq0~*GD^P2!
zP=!G{!LNHvct6vXd9P23;}|_`SH>73V>@e~nh`c-mp*HgKHKRjkgCfS2bB~kZlym|
zb)D9`^N8U+Rc(3*EF)ywN8LicK%nB@F-W-5MD%M!_3Thh=~*R@eZ{TiMRc~*y4OgX
zU9%nu;_)4BfR9JQAsSh{e?<qg_7#S0@<<3P5_mosl@bhcG|jLRZ<QS_o|m`g_%!8J
z?`!s%T~QJf37vhtpx)0khc#`Oq`}utS9&w7CncHo;m{p4rI#m|2}TSxjl1{OVxou_
z$%(wi!+!y^ke}Yb7Lkxnd?NZhZyqxz6yu7&pSS6$!`<UxR4i;~3gb%un7#ur)?=lV
zpwZ=*W3>rLf)T8@R4LYO3`UIU!xW-hdr}HO74RXf!+M{mE_88|vJ?K$^IvvYlcsbV
zrSt|Xf4hZya`&u_S`AvG8HKGJF?5Oh*z?t@J2cd)3-dNVl&gCJSJqH3T?9#$3A!Uf
zcCGx_n=@eGR8XTUhO_iwpYdNbZ(&&y3_MR0if_sa^Fv5KaSKo2{)(;Ys(q5lH*tpS
z43n1s8umya721)YjKh9)Mt&|aj}g(LAik{{gBDKCWno7bhuCVEF*s?_kO8SkMZyn!
z`g4NR<z5zDMmqVHfTvDrY7M+S^l4wvKkRsemHdJxJBER5FgHOqhaRGTa3GNW{f8R9
zOE`hMYNvd?G{^I^S~LXu4w^*ZOnWtrJ-YVcGzXRZ9s&&>!^mM@W8caq;u61x9hEMG
zMlLmAOL3)o>WlWQ8FQdNFnr~;W<U=ydqy{1I&5~P_+*?<NBpm-muBurH#j+3#FG<F
zJUL{8aDphQ9O1m7T?gt6KRKb8Fy^Ae$_24tb*NBMsIoH6K2^Tji<jvB*+ndmmCYHF
z0Ln!Wz|6VJrWc>pl&R=R^tSjV4R{AZ`ybXOg5sC|*r>~K_}L?L=>O08`hR26|M$Q9
zr+^yxoz&E{S6f$A@n+2BwBMA{Ix(pI7KNI4<!?kaCFA7du5Um-*xh16_dO>XX@iiz
zOJ7oaE55Nyfja@paRSr>*)^8G8~s^1qntVwIh;$yCNeYLsKGyTnqJIR!#qdOeFRUh
z+BUQXpWNNzN9PbV;CG|BEalbrhyIVUFgzMv2zqJJYo@HdExHnh2ewmjQha<fWTk0*
z1hGV+Mui$J1c5ZJ;n%i1OvEG31m^#W1>Yw{bDr>IVW#(LP|cq<c;`j@x7%@$M;v~C
zN9ZxwJH?Ol4JjSP`lSdS9?J582vy7_(GBl3oA6D8G2!vm$$TTCmc#pmUd9<vxOd)8
zV+Oy_njbG_U{r2ip20q_?D!hONot+|^@T6fU=r+ODErkG?O@jt<qOG?TFs&WE~;g%
zfE{ZZ3k$0>vL-OLDYfOCcxRpFT=<Bg^&OGtqksI1oS$YTSY0G4n-wnNYpg?{AP8~j
ztKV3eiDhBPY&zGBq-RKc8-oevSHXM}8&nZ0)HW1&rV|yj1+LI5Y6kcXE^n)!1WQlN
zjDAqWu71e=NdVJV!o(WrjYe8Bvlfb&^Ty&TshSerQ{|Jo@a@T_>K+#rl(60R@OUb}
zZRQ^Lyf}VGZ1sg!>I3wahI||B<;TrJ%uDW7p&9GMFT4w>RvWgv`dL1l^Tt-m@!Org
z6{Uz-J-iHId(E_bP5D}r)V*K;R{vU%!oQa95TX6vjtk=A+|oE=N91az%Fq~_n*9|j
z4N8H(lK6Y|i`zmGFUmKg#G-#93Z7)00d3S7RXe`_?hOi#@+y6_c;XYdhEG;pNi6Y^
zeC~KXikS!r-5J*_|L-B8#d&fGfPncG6;;sKIZSPlvf!i*A)W1CgO>U8-UKxeBDy)9
zVFA|=(^o^_CCnu=wo@(vXj9wjt$3fzpL1n;4jlCX=wDkjZsaU6J@O<J1i&x(tS=-e
z>L2m2&k~l{jDOdr2$`f}rh}ElJ5J>XpqX%1+U~e^zS#(v#=W$4?;7Lru|J3?EdlS6
z&9CalfpO3o#`K5O4@Y;V24^@~jwros7t&j;YS8btoI`I#TBK}~a&l8UuCa_M_J47~
zU+)g$#JWM_@!FbWCYc)v9n~f?`=mY)fgShN%K;T2*oA|Abk_T4_A1jUxYwHjh{_l#
zvHUf7EsmccPtE^le;N#qV?(YT$&nzzJ;^3L<X3_2*8!j&V5~A^iZ{y#A<<X(K!Alz
z<=yaTs}N!TPZ7fJ86i*a`aZb{5#4Yy1yn@-yht>~E0SmLMsq=BW8^0Jy*blji2B$U
zl}9{Y-NVJaB%jd7)cD<dzkt<;gCJdKe5?uML4vysqoo-IF2kXcR1OosAV02%VPKF1
z;FPdG8?UIUPM>V=)X;Ld#xY2S05;hxxESTL(&Tb2&yU|y%31B?hXZ(oDb4_#Ppisb
z9utPu6O6=SAa4(^%)D}^(wZe|^1yhS6kJ&$`9{a|n1uoElJ-xDgFRPIm2WhJgOX-B
zm2d6E4K_Jm)-~Bb{(F?uoC*Bg1S6zGFSLlF&?7&Sh!~tx&Yzm@_VG9{QJ8ENJx|@T
z)Z9SSvI$5LMb9TLU_o(-;dQ^FAthhQQf0%beWL<%Vhclsm{)es7QhCsf*q(K<7)xn
z24Yx{1zU5#WiD)|f?I8_4w~j4B++NBy?@=5Dsd9Xit5Pyl+Mr4&WcG%_X5&MUw1S3
z#(&5^q@bR7)=@afc+}!YLRM|C{*VQSb)s7<xgRxhyLKMzoSv_rUMy4&r>RZTN$!r+
zLb~}J@6jl;`Mi?*HDQ?ETh%b8yZ9;`PQq@+icF_e*}Q~!wKz}&{05>#1^trua(Q*(
zX-PpksPW4T>!QC+%fSlx=ZSF6IX|7BT>h*1fW~p8V>Ym4twSm|4iVn}t}3{;LPcH(
zbPFJ)Dx(0vj0Jz8pB`LJ=#HXjB%06Iw*ai+ce?yA5BQs7+?Z7DT@~<sM@$X&9ctLC
zdn#+vK_0W#x)A2BQ}?XSs^c3$*D$Jh!XSWHEo!+m^H$*!Gsi6O@6A)!6B9cwrXu~p
zY!%1BF6FRCmJaI0ySqv?FvwHWJ9?!Y?QPW7@lProW6_l<YhTEfRv;ZY|0duKhm+$b
zltgJw`A6ENdKF<OqIJ`VgLHJ>3s23lFp#umxjNn8+c6gRtkz~`;Hr#c;^bh3O*$H~
z`Z6krJQ98*tm57H{CKB~saaALfaH{6&Wf-tMc7Rxf)SJboJ>L4ca=e@dL*HiUVU-o
zi%m-~4Z#lov#_W9gleKH3Qmr#xS!bYWPg(U%R8?gI5q$BNw4|yN<BSnsZDx-4GT_v
zqui-KB`XNx;AvcKV>*{^f34$b_uG`Ehp9Wp1juO71N@bKbS<z|D)-`h;woCKTt;&+
z{!N3nJ91ECIx)rkVj_2~4QK7Eob@53{}z~$3B{<H$dv+}!!J0dyCiy|?;J(9>_rP@
z)x(0MOWyizVYwo0JdLx=cffXR>&&bf(LY^_QG{Jkq}IIFOKn>@Z>RqiA>cM%m}Amq
zU~7cTKx^CUn!!GLCJLW<Sj*@THMKt{TBP#ljR(K)gCNE4sbU;XEyuL?e*1x6fVz-)
zLwyUWuakoR(8y9+UiiBD%<}e&Sp#cp4!@NWv%K%L{3kCa(XhjBlRYUGlLLjuvTQ?n
zELU&BIK@_BOU%O|dHsuj@jm?97wvk92_^M}KFOsDfmCp?3`I76Bf))CmxGhIGym={
z%G$oQ6wo^Y3TX+4o5^=Cxh3z32sUm3ELyNSe@7gasZ0%H({mA3#cK^8=TgxpPBKg+
zx8xL^s#EVYgu!!mwx3h}Zl1_3iVx>lb|n4P>|W+sf|_>Nm9BQ&;qVL9{g0Yx@Z<n!
zvro}Dd6vIa*bA&++PdJ0IK{`4AuUkm_}C8`VPk79cxFC*Y`q7tSXIQtGMM$I7DI8)
zpv#EF58l+L>~5wpAzZ!O8?Azr8&Q)2z8w9E1Slie6t6k-46V1LGA>Sx{ZVQSI(TOt
zN-kC#cvRdK!IXZ5gnd>R;Y`=4yVz_qGko<*c}L0y5`ty-E57gL&V6y-sP!5?BJ=J2
zF=72UTNZ9Qg#LqHaaZo)*|HkgO%yD|(jUhQtHi&z`4sO#x)q653-vkL3DF7TrjNY=
zZiaxRCrBp@r4tZ#VToQ@3dNRd{ueHF@8p*7^{zY?r41XDO8VvmH@=QUQ;ITqo!l-_
z*6C$q_+Fhodc(NlWFPRPa+cq9-vvb`qr~{;`Q+D6IUvZ@zNmANQrn%uZ9;|7GP0ud
zL75Z;*~_M9mICu{V8D}9+al>Qu>naxnUH>#skvDovVv2xw70fLvW{{Q<R))P%K8Fq
zwVJ<pvtNB<1t4N_s~j%p2t%O0=*C_lFa~S`e#nIHKy&psy=5TZ8Qv4Styw*s3@N&h
z81j`-cpmIY^zSu%E@_Nmt1|$r>7^%=)?@XCvGMicbi~GkD?#tWzce1|<j`Eb&h~0w
zROdQay1On^ac|g%ZX_84Wb#)jbqCs+EYUMb8H=#!e2c}tKz;#h@Lz2Lz*^}MBdxth
zyGisGz_(X9&ze}6EL$a~UiR=WbR1Tn0fs;*1H*_IAKk_uF|IM)32?hl!hY~%CDGG;
zFLq8=_oyp|daOs}UT}OdR$JQ9m75%N7(wWl^;P#H0FJT}wf$9J+vZH*;!yGEv*PiL
zuv2;nd|TbO_RM19@+F4%g8Cl)_kd;FH?i{7njv9tlWtk^FQ~H2jiu7~XIVcSzt?%E
z02rl@ElO^Tpm?bJV5Dlh`*zJfCqj}=<TJA7_jNigvZ;u#E5P?{4viN}8@l5?J45Y!
zBXOAW)>h+3%jmjF?(1`D!dvO9l=p}BPZq#7b7QxM`VZ^t1JeyLkXjb(wc54s+fMDn
zTKGOZ5YR_6q4%@Q%Y;fVY3Pr>C`=7Pgi~5<uk|w30g*;s|Lhk><QEYz(FuQ~<Iz88
zB$6lGx0*vJ319)VdG=x^-+Yh>Kn+XJuqI2|V0ks)(#H>&!wy_TpOyWCpD=shD}2S=
z9=b?!8`BcQolcM9w~brXD+OU-s(dLKK5*`-A|fN_Fc6`Ljr3#Lh#C{e3WrYC0J}NP
z6eGeCurhYuf?Kw#!lz3^UYZN?snvR&pNo5Gt$k%zPAX~jeNeHx6*{qh_hhv9V@A%L
z$X-sU@VKqC4MppC+08wUnv{pEkR|G|dMt(9k;ZU_V28)yB%l(c2d(8#YBrc@Sn`7v
ziiz>F-{+#;x}sdVo6*A|XwAW-Y3o#<6EsT1g>qp#I|snnj@;9dYeX2E$8oHj4$WSl
zXq8~gN#JlXLcEi0UI)F%_{2F$1Sh?v+>J;>i+n57&vmlGS?nlf8~_Ew4y6L5c~vaI
z-75S)?pEe7A&cJrDC6xZ>EdjL7h@yjfMud)0NS2Vr*@5FH&ZJKaV02InEp62<Ho<y
z?gq#xM>Cy>PCtPmkQEgdqW2BaPAd=i%eXG6576=PS*<0Z3n=lYw{)wTHV+q4DG0pu
zM74_vXWF<lk$q9rJ7JRkt%&Fa2^^z_Htr1Stfk~!LbXVufI$**Ic-_gTVyg@U?QJy
zk~rFgC~N`b?dvFWzPPXKdG$xj<s;+g(LzyfO;Rc36Hr>Bp}DS!|G8$Ot<io_o%6VD
z7Wz8Cqeo3l16shH$%4y>Z~?Fpcaq6$j-2@KMp;5A3>T;&<S5oe&O8HL058I=oRo;7
zGtaW`niB*6P>+O)jhG}}*%U?i^ru$Gn{<i~T!N4Dco$M#4+KuoCFA%!S@S8K0DDf)
zF~@K7A58+fO*l>h?9aSTOPGCt<$Dw6V-u3}b;23B=J<_7waZs=1Wh3=!@V-Lp~6Mu
zF^hB&S>-Xl8}@7ThjA1Qa?GV6udO*7Ow4XM-X@qZ@_kN@okYYZll2d>seNVw%8X=_
z#P3a*3uCAk9c9<;Q|UJXulVF;!IvG<rIKEG(9Ja66HT)-eb-B^EmUaVMDEuluQakg
zgFec&7dU20`#Wk?N)UC9zf<13W0IyoWWTWS;hc&h@Z*G~PeZ~9DT2qgQ%eD8lM8Su
zO1Kgg1MkE!jStqd7BJQQ&!qeA&5^fz4uAaM<GyzYu$NlEhI#A=>yF58UXOpH=}1kz
z#YCh-75#ay>iM_AX>@GlO^Lba?(Nkx(A>{eU}WCW@A-cVI1>Wmfnb0||FGw9L28zq
zX4hKkzNc7213r}3bklk~#nVbHpq0Rt!i6ruRorSquM3nggF6S_xGO#eBqPZ5sfPGD
z9xn5k`uo10QnOm`5;z$E#l6miwV}2aY|FwXep;leL$0eyF~tqVtE%5(RYP80?I7i0
z`OEj578ZUUmg`~Eg-<k*5mGAUX<+R#%3ZF?tK*$0l*p&JxJE%MgIJ=#)q$-`zO{Aw
z50fe07T}?@R3Vffq+1hbJz<g(7>M3;J?<ofPwk8QmtQ#z`S!jZ=h-f$RNXsMbE?a}
z3h0|n@{uF?Te{S7)+!s(;XdP{i30TpW*gSwFxb_jY8sZ_ajq{p_|zEV_<k@s>8#**
zhn0%Ti~cwy>J168nU&g*!J^SM30TQEN|~hBM@z5sFP5fT0~^$t#)R;@pF>W)Bh;9S
zBH%J^UTwcjjh2?r)*qlhB)2@4;qnX@Q8wdVTELMhW&_Y%bGRmNzj!9klsOkZwcD^^
zqkY_@U~+2CdL=N#-bbS^_RCOxa=-HQf4rU|C8ex}S`^*?siF@kF#%}1>`70eZySOA
z@o`@hP!e6`E7Ju<AkjHZMZ%QTinPCY@R_T;$NeeFiBa;o>kQ|I+cfcq0&ydf(?6`D
z<JDJVp!>zbAJws+_a+==W!vyXd-C<f(J?QmoP}8SxAx3?i5vL4H{-Zjrg5P8StQ$H
zX7Q(tg08BGO}-ji<&{Ph8c!9V784m8^`{sOobH=&jY}PzBdF;}rpgvU+nS^nzWGr&
zV}lq8y{1g1JgSn!C)N<j{Kq^c=!JZ&cm3D~HV>N2{uIaEs!AF5-ihB=`Z(E}?}-JG
zfLM}s%H&T<dav6eC8So?Nj9AUnM4V-eZH*PT~U+Tc|!4yEO8M3rE?|XYjbU$<@zG}
z4OWEjPidv0<d;D1fD%+XZDL23pWh$Ay4sa8eVF=tNa~H}v>)}~P^1Q!=~3d{*R&nT
zi^HE2ni+ADQI$7c=h#Ro5uJ2xC=hL7*qU!SmF!!w%A~&LAEEeGdCb&-HOK&CJ{4_%
zlL8pmA7ew$aQyKH^ralCU=L<_J8~uLTkYkL3fT|&7kf~w*g2>^yGE0h*=@iT(5Spi
znMS8hC`d29FQbEFqxAA%43IuZw64&XRe}aAo;S67iVMHxLnIpEPjBmVbi5jt;ld;N
zdYLq+Sp4%lvqW}33F?g=d4Dvl7c2c~=&-(t)br9&Y*_xhNL;BJLgG@s(0=EBKuPvZ
zV^*0J=lhxHj<(<(8P7;Z@*cVY7J|~P?cwG}RHql7+osGK2t{58<~-uf&EPH^F60oM
zQ|2Cf1)`{Hbj7_Q%g+V0pj7yJaI#}364L8snor96^33cpa;dwrd;V++H@wKr`^>qa
zQ9~8jAZ4Tqsoc>dpeO(F!J@+(lyz^HrkF>x)6(fGENEX@QQKBK&51^f8fT_&<WoY>
z1>wDa(8JwWg+r1zXs4ZBf%GL&iWb;Mtc2}E`VZS_fo|1>_KU=|E%REQ{=cqmQ72re
z7dL0dsNZR-P_~Sgb;GFxb-4>#$mm&Ic{p0lVIIhvh^A3B<Y&c0D1>uwmW?Hza&tT6
zsD&eOdspsOYXi!RAi41YyMjvRczd5bpeVdowe|I%kWv1PY=J~;oeq!7O)IdJYjoCM
zFx>Lp9}11DQ%*m#7PWeDdVjrrvWb6oV_dbL^HO6Vj3u|1o|Ku1!G8Tch5aaqV)q7z
zZE13cY@CdPOl=dT$krNoon_jiPBrl}78%`<yW_nFW2NL+d`4-;43m}?K4W^bcH=<k
zX&61rc$uJ2VsW7QC)vS{t!h0{c*)@_5v9`pSm{hbZKGDd9pUA`qI37Dn@uoOdj3vh
z`*%I-LxokQMsJSCIY%tUl?+t^q4r>~y;d7j@80e9Cq_Ra;>Zekc99dQ-5+HnP2*E~
zUz<r5+v!aNFFZ!dQNAl(DD8JPXgl#Ando5i{dIgE5hdeq*iz2{!9{6Jb;T(m6CO_$
zyri8~lF>F>6)W*i9wsZy3GZ5`-?Os7ebE6R*O899+IQFS)I))rsOJf#r6VOPeyb9<
zWK)(DC@G4QcZ^4)*^Jtm_J2hAem=RUX?Lk$Chf8o3h_>?LMRdXarR-+tq;k*q1&i^
z18yJA9<e&Xo>UjT$fcAMiPF$=Is&;-j2_EuZ?v%TAroQr!oi&B9V#+yy_^#qFeN-J
zaljSmf&a5Z2WL8VqV$7F@olf?GFY}0qObJVi+TOZ?#1Do4y=;9b=wHW`jzJUJjzph
z)~WMV(V!YD{KwV!e>1)tnKX%(!)}FM*mrb;JVaBV#yGW4?;uXE%tcV6MRK}(fF(NS
zyD3urQ*)QOlo(H9RoU?3KapDe!wNPhCy{cTy-9S0-WeED2);-KcoqXpTJAcJiL0oq
zbWD@y{IK=LKuh2R$TvjF8=N@kPI^w;wW=&Ae##=ICuUWIci;TCHuHb7kU0m0{cF~%
zdlW*mE>iMqP<8^E6gMB>;Sw`z_JnGes%8EO<Wu92!%hIA>J8}LVbxqDq391F_9Q>^
z-R|P1C3<PTfv?+3rTCtqX}_g(t1}tO^U9uVwy*}f6Fs`Yq@-S*r%A_TPo(LNi+<td
z{Ey$?TT1p#;t=tDL<r>Y1Z8I?bSubSVwJkLvoccXrv8uKL<mkN2f1f6z|Q#aFKD|Q
z*g9CfP|8BUoHq!#$Ox$cdxpA8RLvVi>(NKD^rUjj!oURpQ$DPJ8wJR@96w7(bTrN}
z_`|{6iXv5x{okzme`Dc$a{Gf#tmr@ZQgaj_*rhp6*At20mjM0f{#xxN;!ByR-Pi=M
zv#0Kk_at|`#rvWkk5%V2_d_>4M8O`G+m2N3g7nho=<~=3(1rUHehs|9Lz+x9dx5?1
zRQk6a<p0MzR1yXlt+F%O;{S}z{a;`BfB1(1D)0i5X_?ml`G@g;{G`cc81Uh~EaqqQ
zA0Pk!f;#{I>m$erZ6x09LQUi_4B)wXJ-Kp!gQzX?H@g4(@7c!RW)L-SB{S(|w-6W&
z_d`W;-x5ErfkdiWfE*j0ygohJ#lQHv$|>}i_>4UkTDkdOKF<G_zYdgAdaGQzCIf-U
zK^`LSKLH&9kudm=Ht8R=5pYTPS$bhq$#xw$R&D|eVk9LFxWf}izURkAKbru0Nl)Yw
z=(_>RlsA#D`(nD|M<CJueP%97wjlS6Xg~pC>yoNnj=qcgr@lVo`32IiW<dW=Py_DU
zABk*4*OasrRAM9GQZfVY&vv<KBS2A|c>Vg7Cq%0ckky{$(sOPDpIZqNuUw6inl_{|
zMa^*=qcH{`_oA&#E4*WUCtICdQRLfQnZQ4ZYiDCx)+=Xw4v=dFr2s3p+#4WoKZU>}
zHeA2*=_)BO`&B*I@cqlu2SmUxM4kw63SA{0R#687oqrRtfHQuTt_pya%>epfA^MAA
zfL(W^Gzp@*GAoss8rA|l8Qcr0Et5fYZU&IkV{~yUAO$%A*fIzR(GlLjj9>;>CtnN`
zsJ5BkSh8m(LPaxR0@^a;YHd{=2ffwhqn)Lpr#jFlwSY-E1Ar-7kLN8tS5IBQ_)hKd
z7Y)E0LlaTBr}1&W?-|Su%m$l)r+nA|jL+{8lM(E<e)i5&Uh`TlmN|<8Frg3t&|b#x
zZf6zNtV>u66FG_ljKzWmN*L*wN&a;{ol(QpZcr+rL^k-h$@w>sQ3Fg8WjLK100VMO
z!5WND0NGUvmZSQkfEw6B<ok-diuhgwT*Mhb-~IXlvx@_x1smY8`Hg?0bN1WZB+^q`
z-q5<c5xInNanhOG!)91FKma~ddG!0sRgx1T()jRoH<Db*n%MvK<>yxai_adoaOKEI
zJ+$vXY7PENG@KEd!2lbB(uGtdOmq8jqd!?)k;nXmNgxfc;60WJN^YW1#p#!G$2oEc
zSM!j`{5iHJX3qKpVDdfdhz6pYW{_Jp&)F3MPx$&pthAW6$F8sr!A^My_-eoAG(A~K
zeuI93t)SuTsEEh{r4jp8^u=2_D-kMpvK8ff(PJHn359GD^-#$<qjM`S!B;DJ3@nO7
zi9A2)`dY3MKaZ&EQXWqCz~O7Y5v?{`*i1GK1VaXgc9!m8?>?5-c-Ys0`G2wi&{D-5
z+(9emZA{I+NX!Ck5gk&4LA^NLl(B%$PE_(ZfSD=*QWSj{_-zh6jD=V10qvX0BSI30
zp6nKsT1ieeFNFGS1L0C&zae;rMz9__h22C%^O7d;H?SlO0g5ooKan6raYkKe;Dy_y
z839CXLqPQu&%fWv8F3KKeEy2jX0hEdgSB1`Kd?7G0BG}&(ghuPL88eKJ%nLEiIQjz
zQF#-B9DUQ#4uZG7D{^BTEr!?}A7T0Fs@Wek<+6iB<&DU5I!nb_(3-QOk%j^_lR@*_
z<1H8Vzy)5T`{K)orx!4PS);Y-A+u{Pp*)#PJhv>|S<D?q)@Qe9qdtxr)gr*WKl=}u
zB?rN}PXGuM-pIniJ}V(rat0js?F{@v3(B5HrH{-uIe|cKmKoB<9f#MKi$MC^1IU3a
zJJBm9j5EG4_rb|%z};o=of|2-bt{l7*%6Q9aBpt+e{dXy{fSvAd7zbGU5|R3TPOYy
zGU}pVXh=fA;Y+2cUiBN4m?7y2%Q~&u>rfK*GzO|WIB6nXw?BLH)~yim$wR7*^ppW<
zT;f{BiK?U<5a*m0dKKNYAOLgJ-zk1zXSiN1!49njjyMK4u(2{{5#a+XYd<my8si1&
zAyy{g0WE+$!b{A65g27VS2%K8(IlKGXJnc@$@%Z)jB=Vc{T6b;^C?E`L;N@00X;n5
z>qX0u2GBUP08c`R$Dxr2c%ZJ>EBQ?lyy4-`!S@SX5Z8cF?uQ~^D={1tHD8~%5Ot10
z_fww>j(Dj2xW6!&=U)OkDTB4DMTr0S@kYaj)V+4qUQnuB_I2+7GO6clr+AOqkTBN$
z0XJeG%qhBdq7nA-&sMJSX?H`5R0CB?Z+OKgBM+Vy7UqC2poofqXGV(kb%i9YXL1p)
zK6Os8{(zI1Ffhe^i%gzp`TjTYcGU}0L;4SaB+tfA(ua!fsiQ_RC8VcJc^G!(G!^QH
zA-jRRK4(r;(}wYtVE2cV45*flK-mle1)QC|;?rNPOt5s~Qv=L8nfj}{*4rm%#~YKX
zu*0{Dz*e*F6%#0@r}|ebe*s8jDCp)#-R5Jy{WD-2wpY}DZO%4IM&WQ8y>Na13^N+&
z$^JPG)PL_<7suP#h9n&zz+!@!-9t{U*xQT%nxDD#-AlB7#2bGlRQ{W3Jbg;ufwEaG
zmy*dZ`o~O?_aDqCh13?=c_6qh3#Z3|UCVgt?d?~40N+4MTC6eq1i0b2K|yg;3Gq2f
zktufV8h#X9?{f@dh~hYKU?*<$W}V#6^2m*sEOo8qYxiF4(!|)d%f$wfid5%TkIqcg
z-|m!tR@l>IXC*T}VcOY;G@0S<^aKi@FEvHqC@Ux<dI>R>;4&u2_)l{Ex#g4U{Sc>$
zv-NObd6%tCViGT&$31Fz&)v-KtC)QpJ?nnaSYiE9&cv<N%T$Ry9$Cg^L&0m{Qek05
zE^)^idD<$Nvduvq9YBYiG132!J6ZHf!+NQqq}IOEx~k-Er?md@cmc>CHb~lqU?X0G
zT38n)lqKR8FIzMXuD^!<{_+5ZYbIhuuXW?xl6Dqg*mH_*m#~{uTEXR@tg?K)EbRXR
z9o#%(A5hA9znX5|B>>-89gm3U82`IEh~ff0U`A6tAM2B;{c&_`CxdA`+_|%wr6%Ik
z(jiNFD=Gf8SblF(g27*ln2(~RX!K9__1}=Y{LhmEO$4h3Advg%V$^S8Zc<8&;>_;6
zi<t8#qX-71leGJnekS34{)uk|QKH@o(Gi_g@VLgFz_(UFvBAo@PMoo5>UN&CBHmng
z)0JC%!@dT}-+n~@tLsfLqgtp{k!=@#1;~t21>6rUhLR$X$QP9R2OYuJ_(pFT18a~L
zfb;%gZHC^&!$^~fmlsjSC@^5A{bBzxW7T8YWfe3r2jOJyEZO=mP=i)~3Z&8P#&KK9
zvA0<r0fyV8)14yik7Lb6Bo9ZYfKK7oBW~o|GAhfoi)B<SJ80)8h^STJHB6^TS`foF
z&BAK=l-n=xMWMWIaCC#~@Hflkj`5MiX(&=M58RTbz`k++_0v|Equb6z<&<a7G3#8C
z+f6p|i$Ls~>1F-mkLwbMtfdAMXRd6I*HRBLrY}r^?uJwNXt*6^Fj$A(>5rNN(NsM>
zG!)J8W~#uh{;DALM#uT$;@(Wj#9tUO1CY(Y5=vXLYp2}U<OS6h4lceMldqMYs=fa6
z>2v<P;`XbqmB~QcNfPz(#wuA_J{3twC!l^f?mAN&FRP|@oUwOy!?zp2t=j@*9Xc=}
z?Z2XZAN8!57v5G1QwHvjn$s?5ye>Nh{w|AVIMmrSuJBJAj~r1<0q1L-EQf$j7Yg(v
z)d?QqBJ7vBjGOy7JJVJp<RP(Z6ER=38Fa?ZG>#Dn=|r$3$0V!C%WP-*3jZZetj9S5
z9XOYrXUHaOz~)OdP!f<Yw;QzlYSLA>RSR0mYPm{PxLW|=oZt3=nQARZV=vGZ5%^|H
zfCjtc+$Q!(d`D<&j^d%aX90}FWQ_O-PHNcGkXqrbhNX9Uh$>cMutvRrsd}?4>hfN2
zMGJm5Etc(h+57S08q8VSx{xDXcfDS7Fy}z#6#DpAcCyW0z1#BOtyH&jewGWfHz3|%
z1Tg#)k2gX(3#F^mk#IU$8##(*aecO6wYv04_H7*Z??~OGyXV{*77uw*R`CteOo_lN
zfR<}2W1S3TZx{ozbGVHK84c&IrH_J?sm>8mZ)+Md<uit^36BECFCu)J;`5jH5#c_>
zg+iTF85PxW|9C3G5Nv##Kiv<!dbauXtAPYJDOUpMO1Kh=w*SQuz#l<X<Q&`?Nb|b3
zv=;=F37j=5Tuccc&9j(e+oEGh=He|5c__Xq&;b{oH!;7uAqE5qDb#i{Cz}e4S2?;L
zQuPM`%;wwl9f03#f*GR9<CbB^?F$QZe8nVvu86iU`hfR<qsCEGTn(fkRh3+b)-kyS
zX_SNv2=Ef9H=TPGt&Yg<3GS6ZGGZ>a*)gzw-~8pTd#OJZ^`xZuO$UEL=~kR3t?Y@@
zsPEs?x^R?DGui!1Mk>&ok6%j#UUO<B5jc?l^iDImMD38BX$*9MWR|8w?_T>mh0>^W
zYe+-oK)YqucSCw)754fEv(}&joOiVsTq!1Fdk^F<BX*X^u#0%$3gP#Rh1?X|w>cZ=
zeiLyL_8sg6*`Sw%3N{}oH(03&wJ*I1cfmAbDZ}Vb83$?27eMmXmh*#e&=5|BbU}6@
z1vxpR@ESly8nCCDz1}WtDX06kyACLNJf%0C0<&1Uoq=|_M!6Z(vA(trMa!-f@c5B%
zk-I9IizQl1C@o?RI0TVt_1LfnLJF6IdIM+!?x85c>7I*d5L0FSxSL|2`~z0urWm(h
z(}l(jLR~katEw&tI$<q0Sv^|mlz)U!*uDYE>c(?-cStael<rUsJZu*^8Udl0;pbFD
zCh?f|K(95jzT-+^-XeWiXBq^hnPHYs2SsQ=QG97^M-MQYOtn6&ZIn6QvCe!Fw8kxn
z$#w);65YecBOtb%f#~)&{w<x1TpZ_u3&KLSmLDE2J`YEe`|aqp>R=zoXT?1$38|<8
z!pCNySy<TPkPd?cZQ*_e-3vy?9|vD#o^8H-<faMuRKjm|$!jLE^Jm^%Jb{vIn1R{6
zJj(pFu3ucNE0vM{!@O)=Z4<9zbILUREs^2KNkxOs+8}PZqsOI4w$FlhG%d85CExXO
z`~K39-tN!bSi1n}>1%rk!LC`cn)<X?hq8TrtD7P0%YMNMHM{!}EuN^Wa0%$qd@uX4
zr=W%NsXS8dUNvVujw-q*O99ATJtu4a9r~{a{0!%!8=jT8Jg+G5NU`euaZAg%Su?AF
z`uoxO#J=&mnkOG`R>Zi{rawao%%|0YH_5Wq!Z*&WwE?IeE3yTP1!~m-q+Ahx%+~*b
zwnM2A<2Z;IsE+AI9A|{+W0P5WwRK`wbW(~U+HK4#Jb;&=;>3^Ko^g!zAmVlg`coKO
zCAa2dy%fMNj5H>28d=re2_4Y>jGl9nrW}$PAqMI8nM$ySNCNh;JBwvNbcp?lbVat2
zd$M3ssro1Y9VL}uU9R6WQYsEv)Z^&*W6WY#Dz+Hyxx^CWxb&H7ZOqcEy-<N7<LwRx
zpg+x^qRy~?GEC(3my|`AJ5o<nxeM@2@7N|54q`$E5j;cL4ze?<vLa27EBE(ySWFy^
zM)d`(z+>f;xSwFcA;MY`vI18LFMF0@kzQcaTw-H3EKnKOh)962>yJ(ejD%`DIzRrs
zHEj@+wna>D5v9FX{(Ol|alQ=t_AH5NB%LZb-M(lfWu_6MDkGy-`{%S3W9vE+Rv%}5
zyZ=t=v%xxe<hGakHSJg5GF;$iSiBB@Le<>@FmcN`sC)lG3%|~`o`T0W-b;XXbX13k
zbIwU5su|GrM%#+gsfltFxO5pUjLM&D2Q5I^dKpniA-K|>M}^D?5laWuf%UsHnxY)8
zpZR~k-_Rz$e0k6DhY;miz^pdd_da@D;rOhHGa^cA;faz=;@H`$Px9VcNrmk5;ML*~
z%M7aRlB@()@c6V(brhGvH}h&@iycV|Z8-ThxLq07!l<6s8OQHGOnuW!{dAJw$?D8J
z#p-0CaAeY8ixF0LB-z_zqbmMP_|2EsKebmKxGUN*3Vjj6DAH`ewG%keRgQ~&XdD|C
zm(#PG4g-xMOu@ZaM-h#Xd=`{^tGglD^RjgRH@6O(v2XXvvq6%tTTdaGUT(s{)B+cQ
zeAKcRSJieY4stJ{W3cliRb48IzwP5_ly9}+vHOcVXzMG!UMMgs54LWP@y19vWNAKW
zd+!{g{-th}0y)_F&ec`u{%ebC@g9}ywe{G>uW&yWjsM`lt3fy@n_0;3qR311y4rsL
zzQA31ReMtrLtVd38WJ@b4hvJV38|2I!NdJBz3FVq*4p_hp@7~>tFBtgc3@?}qyM$K
zdg6C5_JS$xY@iZVCSF`f-=iFysjM=awEA@P`tnPDuVyL#gUnL&YFA3Z^;pZU;371u
z^Zj@<G>q~-p{rJ1k)Zotc5+iP{T7CslE%~P#WRR*h4ka;ibsWpt!5!#f__Edmu~ym
zA9ZEkHIpRp>e)Rfh~N|Z)!9y_+sbC5JT9i`-=39ZRfJ$Y7J=ug-NDuNH$gj*luvPU
zx-G}wgWM1+#ipYrCwR~9+-OatIHfRjx%lC<O1f-Z<(JIej0jvy!UP%8=g`8GCId|q
zj&B<j+<M52vwKwbw!VCVKc;Gg`The$6OE}!X+>iuhX_WG3tU$9hUdaWWhBZ^^#qc4
z)&PyY;^NqxTbm-*Z819Hh_hLv{ZmjfuD(5rxd%^DlKq=Pi^DQqb3j@7De*UKyY}%&
zN3*ylc{Je$LyN$`Uc`I!g>VK534M<}rOF@4jvecqFF^jIYk%eOmkW|wJAH1;T02q;
z<JU9Byd~npv=g_v=&6U2dkfa%&g@cHed^-t{0=67J0hWKa$Xreb`1Rz|7_HTm9#Vq
zNU6;3)L58tOPo%aKe+(OHH#<r!}C{Fz<ch@zEM2!c<_+ilb$39Qt|Wc`+so@+7Tti
zByr~5o<UOeSO%mFWNEU|ay*EWgV~>hQDCNlvV$vc5lB!m3z|#<&2b5x-7V(y`U8l1
zX>Y!?CCU(;n8#HzF#jJxkiID|oNxy@jPa+uk7Xg<S)@}kCZ{LZ4L8*y%mJmcy%3Gk
zqx_=sKttY*Ua3&8Btp*`m2$G|z@^z-L)R|M_A-B6>DY{AmUa0ZqUVkK?Vk4t_KP=W
z47EM($)`Owo><*`<cuNQOftSbL#jPSC3wMlLP#UabKbg!!>^}8d3ge30Z%g2;zWs_
zewX5BFx(p`!}N09xWQq}yimqw-+aK?DmIVW@kVIFzMpG;zg=%%F_x+Nw`VF*8jdr$
zz#r*j8$+3GZKgyktP(e96;h9kQ$Rg+pE;sTwU$TFW-BM>K7c!HIcwVJv?JZS$RnF-
z_t75Cu;sfnArEh<HE$=37OGO}PKkjn5A6)fh)MdF&>?+6(%VGLl^fTA7?tSVvi&G^
z>}?@FUntJw#sK#sSp>Z)MK4#93m4K<=EKs2PVs}e4DE^3d~I%@nhO)D{0e(<uT>WD
z>++B+Yf4p3G^M$5jM}~5re3{-XGHgv6$cI+p@tzRd7vLopI7toRDN1Gs<Ek6RehsJ
zMthIxx%m`45Xk&^emPxn(OrHf-L_s~XJ?Yq+?McrNTCT~M=^XEh<}R<$r8uNw{1Gs
z=^f%Wca!^$b;d%{t1Gpm(ig->)|qaqdB=k~$1}J8H6!xH6FynTbSK(L^{>{)&LOE>
zqM{2_1dRuapo|=!Ke|M4%MhAQ9>wkS@t*Aw9n1pI^sh2`rbSZMc0zJ==>xm~8oc|c
z0bf^MbWnEJG_lxTg>yg~ECoij_2uXPt3K(!47mP(>EwR<|GvoaCTNrf&)j;B1GQc{
z*qS*5fCuU3bAFd82D^zk{pe-nK%g8I$2Ch67hLTSc`HRyf7U-$&jNINyGQRexx3^E
z=gECyrseJ2-Um?!LfH}YXv;f_e@t0_1)wEf7F+FI?Xc9{mg3|iTw=hHeWK31r<~5u
z$7euvjrG=^ZZ1Fj%cy@)7!*!}a^N&rCN{n7uKYu+`8)Q6$5;4jU#0o=*PaFe7`?MY
z*qmCJdoF3qA%)qnvEzIh`Ztfa%=A;+7T<%-8TB^v!v!cr#Ep*0H5QuS02m)l@^8iJ
z%9DFNsU4Sm7*fIn>ph0NpVwh;QPBON;!Ya)&P5pqZ7qtA>a3_K11pNdQvf(9Pv#T*
zNn#vlKu`(|ewRN9;ASLrw<=oXNsg;5QA8kfZ>#4UN<J^V3M%}t6mwnapec@@9<q(f
z@~Pov$PbUQn%zL1N`k{UdWK>y=~xR513ZP<(xOjesO8Aw8Oj*~X(Jd-`;Rf`nfw~r
zzvs3DlRQg64nk~li~iCxuAzU9*xOb!Z~X%+jk{{rNzO_|1A78{C7W|J8*}oKZjA5r
zrt5ut?0<+3ooCD~fhmstkCWENly4@WVebjhl{a7sQ>j&aA~jn_OpQ}a&Q_0i*PV#Z
zT66HV39z@Gv`Uc;M8D(ne#VW64$nFIe^kQa5nqT7dG$~6yhXiam_Hp6E}>I=didXS
zH?VRP<WRDkf{z*ABZ$ev^;%(BV$?iKq<5&TSt<bT_>L3u`2uzv+C=p2s=pt?m<RBK
z+rNV-99>WdBC>4wVxrs1r<6vtH3*vqm3Mo8YZ|CL3@pmQkOEFv*fMnQx9gXmt-;a(
zk?B8pdSV!-(xK71qn3=pgA5Ex1Je1rxBdXlgd|4clK5v!x>2xfT}!!6N4A3}I#C?C
zjf0c5W+i{Q{bI^*V2otTh~?P1PyV3c{*JY2J=xUO=pV-pG81l6xmw!q9f{~F<PCp{
zN6RscJoCrulBAIU%IG+!f!zh>YK7iUulqPYiy^Z+`U?QYwSc;HqZg*pr>`+2V;Jto
zQh)Up7n}fb-Lt0oV}%h>9y25_b)Nbe>*i|vAxb_*LAGxpvF?reI8&3LiyRjT`@ZY<
z%3QO-?gJaEV;)}>C;8vQ8q)>Mds+JnvJ-TSTPS7{c21A?L6!UcN`H%s21xp!0SVWG
z!ycSz{!Z>YTlb0byFX+@+W0UpE1SLWFIRb1ngIwG)4`JIwkp8NcSuLIQnfBc%}`gw
zE2$OZF1dTlGsR^se2^DE`y-QboNnQ+Dgdo>`d#y8zDM2VyXMju$VaM2Z9Ki|y#L@$
zC-nXnXBgjO4@Yi9X?KIP5=cYg{Rmb8Zv4rsG~}L-rOZAk{>vh}{S_(vD1<=<k+2YJ
zlg35utXoxL;{_d*hcFry&VCMgn~tsug}>mz^|jt4$~3~@$PfmVf=yQH+?9DzTECzX
zaoZVm23q=dZGDeG4RWG00`A9KJ2m+b)-K)yToy`dp0gNLLVJvqC-Uq)G~F3<Ol)U9
zWdw!tdw&R;3nRMQi58j<B1s2pt1eH8Nu-K8>x<4<s_NvWOY1L9P+W78D}xpqcDGio
zj}#h<ZW)U@MiP}uc<F-8!&lPBw&_%CcJTMzDeHTqttgo@oT}B3mwMjpVPa?5l8*r~
zJpI&&@ML7I4*nt%Y&G4>-c}6CUv=|p2znTaw=G@r84U-ru*xOru`qZ-DEt>tS77m`
zcY!`i*sZboUCre3%R!Ae<wBd3i#~Y?Z@ChjM%SvSq~=s^=l@x=XJyO?k*=0<m}$=l
z(u;&&T?M`5EP$|_>KP=&y~d+@_SSNA%S2=n#?DY}6923I;HmB;#AdH%XLgKQ!j?tr
zI$A3Sr5?F`d$S)`*<mxqS!1Bet=2k!8he*yO|0Z`2^F5E;L9^G0GEF+RLY#R9~p)$
z6S1|e8em|iw5peZWUFBFLlVq>Eq<&ffX;r{A@9y1>L!5n?B2OrtALv?#(1wZ+0!Dm
zhqY2dDK`N$X9?B~DB<dBcc$an?>(+xJ$gM_Wcm)>ME=(>r7P9Mp_Gjp37cOK{8<k=
zv)q#jbFf?pKRw<r{E)SJU(zR?_hVjmaW=fuQP=x;P_CuH_^QljYEGd5s!t`){zaZk
zLbnJELX+Z^KUViPp$2hWMz!nJE8gZWkQX3CTR@vM0j|^t`ou<}X;K(%=sFsh<kcoI
zl;kazEcC)XxPYQhblVOu{3W}{KjH~em;v!^T{^}QY)p4NUvXa*+29uF?h0E4FIr$T
z#h8J<5;{qh097opP1V#XLS7t8Zd?Pq;8#-&`Uqo8^M;dx)c4TM9HBtgS0&eh3VBUS
z32QbL?OoasFZAK;MVgs1;F+mAv|_Qx&|4NTpH?ZwZx3iqS$t}1e)L&MqvI>nZDyhj
zpNy#eXk-c+1916R8D%o|<@<rkcqlys{#0mhX{bZ*$`#W45La;!FUFZJK%Ku@sps*m
zas@wlI}&iZgUpOt?_tl+j6CQJx(B__Z2uH^Yo>!~F+Fdszhn)GdmOaZ|B=J%S5DO}
zX>BEl&6HyXI_uRSG$6)Xq`&svH~+K72BlN@UxO>t6IcELOGIlXLh>J)39n*<s05RI
zdwU8epu?<yYip|7`!jOVGPwj$V%mr__XO^5!G~N}B}mXG>b<ciLCKx?*vN@&Wl#9u
zv1O+G7vTPq04sel!#lD&0}9T-ku(HPRZ-AM$=91`fh3<&gG3t6Uk6CI+a%#o5cEmB
zfA@rFqXn$`V{?Z#{%MgwFo~_*(l;%2BF;Za>e=uwO}qu=?<*k@rsVe1-URiv3n|Ia
zERK)&f3#cu1+tQnAAxFXGnX!X56%pViD0;p8zw|qC`(*(c!@Nv=xCKVSqa#T0U^mU
zM!^?22F`(fAdDZCMtuT>Frb@FT^KgR7(|AofBp%2f#u+TmF=b48W<New)!|6?3{ko
z&8`99gHBj+xP>u=@Rfp7ww0=uqF5WjaWtb>D?~Kaf7h#+(gWYF+NeJ^Xv91cO3g6|
z$wD9^yKZkdQw=gCen9a)e}K1{;Up;DI(ZGEnH3Q)_;+3aLl8#ljU|5W2|CC}d)oGS
zVkW6MR4{(b`GO$w-i|f3;(||<`H_|i_Sfd`(l&<YNa_Vc*xEqhA|X7YDUibJ`S%Tv
zl&w_F9p8eje)&F6G_WO$9{=`=UvRkxuh5WS9w}}B>C|nZhHg|oN~9CMHah`c1!sTB
z)&6FAsh;NR&C<920<YwMzj@Z0x1fcoq-5i}m>1@38m&nQ)AphBWw&}TJ!E}n1H0qf
zio=>Y0M~bcsSkWliYJ(nR~bCT5Z$-Y3oRdUz)q~(^R28oAr^1z1f&TOWri}QjJ9O8
z3Y=D%f}fGojb|-yR$%I)TEBpQQf=he9{)ro+V~1z_xKs_k(=HGE%lMnvg{J9uyUtu
zN7~0~fgxpHa)j^3SFe`Ydx8TQYxkmbUv<9aFl-lJ>q~&p{RZ(Iti@vtQ=|4$vgAyn
zHgEUukk^76SJ&;*3XfZ5Uj9S#%l%ydwznqkV={swUA2U#-XG48*}+%I%@$6)HK%!1
z*2WE2#rLqgEOy`WvMi5y%;SVxJ&6pD6;98BW{uBhsqN5IhD-HM0(KTIJ>i2TF3t%n
zV9WMGkN{S*%PacrHg1<%yf_IwMka@*>PBi0Wjw;#3bMq7>~4Fcpou2N`xUnm!g|Ju
z55YY^$2;9%B9A&`>l^4lC?;thPG@8K0E93_p9_o6H)k4wb&}z7<G~y&`Dw)2Yo4na
zL>NA>QMa~LqVV4aU921}JCO%*9(Mj5%<CL1kR0YkJ~40s{+b#tL-2)zy%+gKJO`f>
zE{h7n%U?&8BxMYvI0w95amUYyDiO?uCPaNR5D78kN{x$zUe1h~N6`S8lTz${q6fz)
z+`Xu`-o)&yNs4L!T8s+cb$j%D_JSp6h>2-`0D71=k7A$Zx`Ze(&b<k9xd|!J)T-Tt
zxDrTZ5fB0+vn&@%Z)y!~LI%0c3(e>VIT|z}rO#gL#$&lhGYyL(I!ywyO;~c95CQ+h
zbj-&A-@4^(AxBh`lm}Wo7D#IvWvSP$@>wK~bTkJ96H#(P0GWNXI{55-dY>B?(cp`K
z8P)mmHmQhn2Te_gn7|Ndpgo>~aB*p*#PXIzrs0y~jq!ojFvZH<P#561m$6?yq#>C<
zvHj7O7G_Qk)A#Td)eLJw>Y9Q@6cs+A_xOrnrBiYf`SlsFgr)@?!&g8i!`2VKn)9KT
z{6$l%lqoRnA;{RLmz$6l$r-2f_fmiL6%#fndh6TN+;<+9eI7>nW4kC~$%It*uY*xi
z0F_=H4Rgkr1?eU0XzAEUV(n#Pgy&Gh_XuZd4I+ZBP2^(!N8cCJZjB2?#XLI_-B2Lf
zmF^2C@H<u;?byy%m`oLU^WCYx#W*+63h8T=S~>?ys){HY=yYJzmSF?BG$G;pjZ<$m
zfMH+}$a<_gRomaBVuTI<80YbdFo)F~jJDO=SIq=BwF*yN8l|E(zWP7qEazd*&r|<Q
z^!?S)5>{^$KvVe~Y$Cs=b_6F@4IBrerD^awxtr+KWGcA;s!M_BNJ2$%(Ue23+S}j#
zLnF7ye5`<d8x=h7<-~(vu=1{uv($?zKh@a*a3<S?N~%Rc7yKv|twBu;ciDT@8Fi-}
z_z<;##Q{|2?}-ar!_L~OK}f`>-d-0AU2x}<`EbehEolR$Qhk4=lchItZ_q@p<nXun
z{^%uYpB1}v?dzQn!o^q5&)-RVSy{+;=gn4Riivw}(OqEGmKgQ?=sY4w*e>PFTU|cY
zJfd~nj5rd13*1!!RDc>)W059zp{;}oMiiA>M9Dy^u1mbGa9o^toV7&i%C=8#Ua}v_
z+>ONV21>q}sacrQ_Pydu>cF|+W~#{lRo-{UQ{Df6A5tAD95SNJLPAC;JG;;jvS&hw
zB71j|tm?{4NM)1}W$zKPlN@_zk7Ms~zdn7h-*x?5<6D2;kNdv<I@iJJbKdhcp2LoB
zk{wr;^6-7$$aBNNN?X|B`!_I;ADfQwT1hU5IH00##1-EsUQ?yI70Z^p$XBoZTSmFt
zc|Dc(QA?%X3f)he)w!=9-?*j~r8V5gkTR~-#rrU!#AHk~<>RyhnZ)Bc;X@1UfK-i1
z_3f(XclW8UO~27oA@a5GhA+~BY$HqwUE3)C)Tghpao)G-P>qN1`bK=3@j443>m#IH
zN1Un8`V$aQDtW)-hSj_03SqNALD{!JH_ReRWNF35Vko0#*4yhhKcuDYB-CeRd16Nc
zKhWcMz1{E3dyN5T8ru;Zii+Gc8KP=F7Uf89oq099;qxMckX^_3k5?WPpY%8PK6Dcs
z6hd$G=|s5h*K~$&PgRURsg4Zuq|a5!^Aghk=IlWNnQtsfANlCv6GAEt$-y>f7K$?;
zS;x4!$vjrU6YSYXM3Su0)n7MSv5$v7Wkm0EL^xd&G|V-bzMN05VPh_jEA&b~5C$Nv
zLK~uMHdv{*q=m7EjAfsGi}g6~u6<$T9yg*t-CJ#LDP-kpc}oGMnIKLchy;H;%1SBZ
zWh|HC4jx|^{O&G6{1vnd&k_cOz?;X7-I&grC^I?8C|2aYI@TZ>yrL5`unytSb@r(<
zr$q^jbRCI}nxQ9k-ARW6#X{%|BAc7heoeIOi<VV1kg;m&+g^!?i?+J%_uKq=dt)P8
z_dp{X#YLOmx(Ml)1%PVFT@@0qU>__y?g6k5#?h{mp3#oX`>HfbQS{YkWiq`{lo^61
zH)EVfY9zH+dc=rCO5pZpi=Zj(6eNk9kiPMWaU?I*LBi7ubvR0~DBY2ox_uOdmCD@9
zB2N^CKCB2=Y76LJCi%!6=`xV(OOP<Nl60WrV=##e%|zT*5UFK{Q$th)>O##AxkFs1
z54#7gNyNXp(<s=vCxh6>QPw<VKl#Y>{pawM?G)Ja-Qgp?A)|ilR#5!C`QuJKp!1B#
zYxCuv4S!U~oq+z}WtsM_5>HRYw)SX^YwQl;OH_C>cex{)l+8I~8--DMevEIJbh$t5
zm`)QBD#~hDaCfDowcAXfKZbzt!~97B8djV@(-*<HbN7sTec3JLN9=1q>y~TE-KY^s
z_B&x(mJqda-9U!PC7#<UEDJbQ8)f<DPS1~aR4AqGH^2XGd--KyyCimI(Neaq-lXt;
ziA8-u{QK~6oy8R^&MtOU{pb&}UUi%GeViwmW#jSjS>l#LBl~7lEMzy2h0Ws<(U~$3
zema<s>;FVWjrIaj`&jh`%`0ed4LsOuMAKW<E<2>%c}7n6TciA)WOc)q9%-`mD&DA8
zG;{l^^qWr!#|GmzNA#}tTA>Cp*!=$U_vZ)8f=wFd7*s#H^q(X0rdS!A3qEv3*3pUT
z`1@Cfn%%SvFKT_>bCryc9qZ>rWmQ>68ui>(!K-Y<n1ZbgJ$D@^pt9X#s>LF5gYq{O
z+rrv`0U`%0@|Jz+6?@c02gy1Q;{zApBNe4fx?!h1FYok%=WM=xC&MB^;Y`NKSPJVW
zR1fZuxd%0cj7=gC_w|WbC+ho>rsu~hM}X@j`N1d{%|krUekL*2jMDuj)g6j8p@}Fh
z0z)Ai?6JkHORDVBA&U$V1LG7|_w9@|p(WDY(H#C1EF_&+eqp6dN}n+K!GJ}sddxa+
z;}OTe(^GiB0CoNrU1q{z9T1+&CG&~!`=g?<#@j60rMObB!t{+y$zd;IpNZo>^2$%M
z450I+wlig>MIHSkAk1cn?)1<!Hq*Hfi2u&)zhNOGE7cKV<0{Lxm=)q*<(wZTafEeo
zz)Dgc70|{xrMT?oa#PkV`xFXpf~%wdxe2rgn6f3}&lqG1JWMoO<1%r~P;ttr=uIy!
zho5H{`KcBtMRyWAeIUJH*>MwJ)AVt?m*3>faPzpXa+K+)%LoVd_N*7X-4V;ZPW)}g
zB?U^~YqV)DJAdO&y*GC&%HzhB5CwIiXX|QnIs2_Iikll)G3x^BcEk>cmwW3k@pOI~
zdt$f>cVn+t{ZPu>hwRWiF|a9IUQyzFEFouj1IPuoH?H7BcOmxwjY&XDAXGJ%E$=5q
z5mk>()wp#D7%%B-*Yw^~NRVd`5|c?_!(`Q(Z=w6s&P@aYLiZHWY^S`Yk*tuPqr3*N
z#l(hK5)^WM6LVA0cW;Eo3l}BA-#-)iJmVm((uA;UO2?F7FEr;^s|);OGeGs4mK9Cz
zapbXijSMNXC2_#pZ!e@kZKd3;4+8CN{_-?nawcY__cbWJOa{bzG<m5E6~jKHk12&-
zVV3d50w^WvS&E3u2ymHaZL45XMCtWE1fLy5lTCMA^)>k{qkaHo>RA2c((7ADNmD8S
zA%$Q<Zd3lHRQ^x@I~Gagdx(By;O({XS!qE~^<7ktJwk}BEC9}Kw+tdOG}=p%({)?4
z#N-d+OjM)T*XaS!JElbh{_{5wxI~!r2u}RXY4`#Uh~W0%p=nJBK*xaJD_DtP5hOfP
zc=opa!g>lbL1z;SMaA-tn^D}Z7pr$zz1-b^8!gT?0SFMhPs8DT_tu8P$)H?f6BRmj
zyX*5Jd?_z<AP&)JNR7H4`YYdy+<Qj>MbHlb9bIJrs7XEmkX2oU<K1d-O*o96J_j)D
z?UMcWKyK%)CH@az=hm=c_a(trHN;hb^!|P_QXnWK>$RM3Vx0w9VQietLtls)%SaSH
z-{^ee3Aix>O;8ti)sItW`LPJR^^wA|eyc8L%1`q1{}8l7SbuPAR=>FMJX7YsfHV>2
z6PNUWeRG(SDa8R-Hv(92n)T&rdgPizXcmEkU&ohvFr_9#&n`<GA=evi0X=^5--Jsy
z1>Ju~9eHaYZaJJMd)R*x*8Bsx;s?)0+z;X+@gzRA!#`iy|Ja%_4}}j%{q5^7`MUhW
zQ+&CPbbjvB$;e^;X~F*Ezq#wp|KEmyc8#dleMQgHqQqlMe!Ff-{;w0qaURws#q46z
zPnKmM7a?S$Hi&t;<@=~VSu?|BfahccM$dNB$IZ&q9+)tYQR4iF4|U#Jh}}78`-l%$
zD_MQi4!gaz(N?_j{k{T&^KqnFqyOn-LIg`n?x4STUM_2q)h-%Ka9_jfCh!_45W)r`
zO#cv&rc4O)&FDi78wr{u5e!=roENqc+Qb0VdPtCJ7A<s-#2#N@i#)Cj#Tz3iuvo4a
z^8BTf6Bvmc6#`v)S-(3l{mVicGbI8g;NjUH_7b02nfHkdhg_3a<OW_y$q%;>Fty;~
z0!@(t(8W5&j2~BtEZ*7PK(b6S01HLaY``IihF^9E63HGqjUiwVks_3nX%TB7;27YF
zN6GYo7cmQpd8^}~P$bC(&^R8Zy1urW73k}50TP#zhfp@q0lby&{iZT*27JV|SwOEl
z)R@5qK!R|T!WTLy#i|$NEC45Z;M%2g@e6WC1Dw*Lv6}eASdRmc>Y0c>Zk^ehtstwA
zqtQFe;5x-egF-;Q-a>D5zZbzV`vBm*xS1wKrtJa*Dh7HX6Yl%sPL`jixEFUD6yf!y
z4krCg=0giO1vf3mG=I_sapCX=vJ_uEIGzh)Fuk8Z1NlBiRa>lQ;L#KuyktOW400`|
zXFfe47dc2k?yW(Ant|Mi;QbwhIZ~nBpEU=4@B?B>(nQ(79H8#q0m&RJc)`?E8u(R#
z=(h-sQzP(79<1cJQjmGQFWJry`22YXz<?IQ0T*W^!9KYHPD8D_NyeR31qAV40rf}v
zwET&8Uvg~vI<q#*pjS#udhR<pP5Yd(#=(b9WX(ONJ=c_IQq!Qu&-1))w0TUlydJ4_
zW@X>9H)wYL<br6M<pO`992$1rU5wKTk+TQ<gd0r5wAid*V&oog<RLb`Mp4$;#Lq5?
zSRh&>gww$&TWH>Hi*3z-<Q7_Tqoav#B#z`SXPgNe&=jld9j>STci$wJBLqZV=xs1Z
z&U~sdj{t?4lx4`1)Gnhg2O<*n*TAm9@FCpY(h`MVw-7BoAQB{b;zp?n33<i(nT<DK
z8CQVRbU6qWO-h$;7MT{?z|KM%<Ht-1OQCXgu7CWAtl}W})3jD#5cVtBYOUT6V4`m&
zdmn%Z=AVTYP;?7ycddd@k9D}7A41QN{fGRAC2~UsMrnLSLIlE6n6$#suF09TuD)$*
zq`Hv~4A^|BJ3uUTX!0CqE8Jkt|8DR)6U424h?N&vQ*?@+y8ZpgC+xI~pw%Kz^j~1J
zv_%Lj&?PJJ)_+gXd7mQ;VlzRjQ-o3%mC^m6qg`F;=H$n!Jf)#y-5Spi=H0I!4(oyC
ztau>TK#YPcT`OU!RWy0wA-Xdhz_dDM1T~O_Ssd6*OEDd_{g?@b{XDZ;H`*b5$&Lh5
zzVeG2%J<H!+GiEo4P;%>rq{vX@}``PQxJJy2;=3Ti+`D3mM^4;)?i$`>wlkKxaBb*
z#k&eehA+aw7B9y5hTyT}CsOdg^Hr$G!bKoVp~i7e5|A^~_qNVD$gipwc5MBRo5;Y&
z@#D-_=FYCb@j@hP?UWN`=tEq7_?j|ieFcw?Y57{TcA4lZ_Fd)15n{4i=<iOFZK^$I
z*}hLuC~;wcuQ#+ZcolE3vmVY!*?SM<zR--Gn;C5b5G@_T#yr?74Toz~Q{-v`_jV!x
z83wH9_=THgDfde6S%{?21?!g@lufM*n0|;P+l#x08AZ->P@u{qW^+yk!J*VENk=>f
z?3$gk8^FEmt7sV<2)4~y+*{r@Qq&gj_AyrMBb@sMG^IbNmjENVYvT`;uaMEFMZ%ul
zil~U0`;@&q3?L?K3qp({UH07WcbkCl_oR58eapXaZHk942E2eyluU!!1}Q{iu{l86
zZqQiaV0H(&k1}bzC-%&xfjDCJh9WuUc+3-u`GTvQEewlA$TkTRWDZ$VFn)362ia@m
z)R3S-$)aJRsbL_$#)UwIb+Wj3&r%0A!HsDA;9N?(UODCQfw~Og5JuPgsTY?0@tIp7
z)}|Msp18zOugtz{rVWo{AXho=jkEkouJZDrH-OmDmTAz(T}hKy{Dw4V9ys#J(<~gM
z1@qEE5$I0z$mm_|Ab-Q0Ci?w}jBYpf<p>Z_?Ys~B@X1;t7mucS+yj3GGN3JzvBx;Q
zay4ZYJB63WDt3^2aRb@R2sC&Rk0K7Y?cZ4pmQo=y=bV-SQ)z#v+d8Lbq}srJDbzqF
z1<LqQ4~{vsmx+JIs8TVL<H6lHbn_VYu{(^v?v#87xdh9ywtcf4;{P|K-~If7MFaH*
zSQx`Ut{%8{Fx$Y_@P!4{mNhl2`tU5d5tC0A0Hj|ApwhYO7C#iC$P?h{3j)$|_Y<6u
zJ|n%bJl!QMuCejv*!Xp)g8SmuwUTF9ZXFK<{V$uN2t>S7KyLry1)?{=HHioqC4c>u
zakg`Rz|FU!1WZj(30-K6Nbf-OOTwqnGHyUub&efm*&>NM*^9;`IfRo}U3FN+)Y4*@
zhNZ2-RL~-b1i46=S-uBo7#EZV<AKkaTjB}0OQmqjSVV#j3Fs~RKJGW|8iLjUDR^NN
zm_0gs8(cwVzu>U%5>B%;zIH#NfIfm~_3;ok@S8HF2$)bYe@6(Fh$W&eLXoJF7-A6h
zP6hiDIcSNq^vM%<*@Mf;`dO9|U|3akg)YB|PIvxx-wh$U@N|T{RS2)Qm)>>H%CjJ3
zcowUoi@V<b`-&1d@g)sK9@h;W?dg~_@`(X)k~<W=#lTe%r&(cA9Fo9#$UR!{8=IZ-
zy)Imce@f=VR9J<%MOD@YbY$JJr9-DJLI|4lfjE~e8IXIBnR%@jI{h<|laOrMUUpmK
z@8=_*qt-8SVl<V@^b;;ZZjB#7Og_t}Q_ceeWbbVX_ndb5wzRp><oMpj=}N_^cEmV*
z@QfI4U@tjyJ9673#Co{U#4t-+p18+?k~LS<>A&+tgjHKsEG(Brf}a0|VjQwS)DmHV
z7%>kj<T8JL-7tqzR5UK+yh&+#7J{;JC@qe5za{A=-(hZ-CR+I-L*`7|4ZPLNxSX@|
zh>7=$6<FYtE6j_ptY<m>#sVc;II>t!-#Y|xniX>4PX6`my)rxTHcwS|q=~|p8Y4V;
z+^dsMFb877n!_=v1NV4OKbO7XZ=S|+xpxOHPS-FH_Ov_CNm%G6eb~=WF!r4}Srx|k
z0)=r8Y#Q7K#}H{oH)nFn;Pic4>NuvD1DL=Bwd{k7^b@2?vaXBloiQuMmSZ#__}`Rx
z?hvrNm_tWg&ZKW6PiurgNg^R#$F@4SfBNa{*G$p2BP$(h*$f?p_FTJiHr8mbt*2>!
zi_89Y-&7DsM4WiOk{hOEVf@CtJmm+DowE6xxEVv%;Fma=*Uozb5jsat))dWx>Zr37
zbTpdFqi>#;qwJQY6q2LtlcV%#)vNR1yKK=+<k@k}!-}Ep)uX^y)`6c8+Ed-q<|E0?
z0WuFfLXGE^wb<;zd(dzzn@<bXR*Qe*j#B<M(iEwXZyqyQ*xfP7CgRUV5y&RyKVr)L
zoP*ajZy<yC!0QL@)=uJjym3!U`ovJ{Cl?ZERYKHEZVs}WIYI2ZwuBg}Y3F`jbm}pq
z;%X2DsLHLF)At0)Y@5qjt3kK)+;iF#C4(AjR56=dAQVm<&yaa^D0$fUNIRoY^eu=M
zfR#_36R4icQ<}?BK9cLq<%mq@(oSkKH?GKEiDcq=zQ4+xfg*q*Ia*-A=nbQNc)d*Q
z$HtXMuSOfi7L;w+*Y6cP+E$WGa2HV->EJ;a)Yjtq>aK6Xzh*BUqWmmJ(sUFl%9!`g
zZ}MALtvZ9_)Ys{4E%hT~;k-8vYSWJ8J4Tq5nD^2Z*+hKFnbBUq3j+NLlm_yYY4VgC
z@hw@G7xs+}b$@zr2N+`SqA%|U1>WJr%e#^_+DM|+(3oFnU9uy){tg47l&shYL9)_x
zJ-vSBY(A!0<24p#3xrXB>X@wAFfje;krM@(ovQ;&n-{Q}evROneDMt4aeEW6;Grd1
zb&f<LXe{}>mSQW9bTJRvCZ32;>TE<?zI=|87I^^k%`>Ns4&m$c$J9`MwzntVCI$mE
zhvlO5s%f11R>Wo$G3p~8A7V;@`Z@zS-<upf#OrjiQ0cK+)(gh?MA9hG9MDes#?#h9
zj5YqD<!_V`V1A}&0eYq9EEn3tF0HYzu?)TnY*0-G!F$4S_XI8H7FU-jRC?{Q;z4qh
zc5ZhrL?P9X8Sq5kUez)^Vs6*lH&!&#(^8$`{BlZ_p{g7a)O@8%a<Y5aMZ~*-k1RqD
z8>Wt9EmDfzwN-fei8+YUxB3Ww2Am%oC+a*zu=$Dd@(;W-!=OE)2hUrKtvYcqo!DCJ
zAltwN;s;$E7*bqeuJc&^D_1l|1QNOZl;771{AZZIvkeFnmncQ3NZ&)CD-~`OwRUg~
ztHfS*ey+$n1<^P|rz4U$V7IJKMDiw4-;G!ix(IaJKC$l{%uZ`<dOic-pG1Rn@e*>1
z0fuA*^&jF90$zl)?STiDHF^Dz0dKPruZ@eor*!gfq++wB<>aNh$v!8Ty@*>a85YIF
zJ~(fafY`-{hvp;4g1*w*6b>x$4j4mnc8IQLWYofDp&gg5hj@S3LTkpQ{pQ)5l-A0h
zug|}1qKVg#K~Zi;Ua7)ru@TvSyg_)4%vx!O>p)xAl$NxjfJXfu!wASzqk1}L=>|fG
z^Q?S&d{=IqR_kCmX+W9a;$&4mO~IB^pk@kN1FOn*-2Uo^hHGRhhydb4NSJeHsAB|-
zY|ny!PGMiGUcEbCTyhuFy8A_#wBfgZ+Ype+M`kYZ?Aj;0>ODV_`>`id1%y)ic8J?0
z;^_Ur1Rcdyc#9Uctga%$;6~`J%$iA;B@NWLWK|&9k;#QrK7eOGNyE5tu{>cZq@!Yp
zDl7rts|WO?P?38?vd?pUoW03$bFP}qFOg0>*4=^vMan4Az3^Gz(jgG_#0647&)W`B
z1~SL9o8{^ElBje~7nu@yf&QAZCPYuV=Dr+rmp#|7MSx!CW^@o(4rj=k--AYOZf;{C
zUp~iNgh?<lwML#rG$uidW6ZGkfTgMRB5K-_IZ`hjk$b@P-0bD3Km^~7uaG$FI0l_Y
zmvUL%g}6aL_etCEKGEiPD01de#32+fJ=MvzBt&4SKK<0x2Tq4+fCWDh@0rc5<^OUT
z$`w7O?dBorG=4W1KslGvN5bluP;!r&sd%S*&djdl%bOEYFU`Gj!_2$xqg}h{N3D=G
zLJGj8{)?dO@DO4pQBB%!UaJ#Gpb6wi4JVlIJ;3_7=frsGB8-Yt^2aObGp!_XZP4qW
z<0?#cJqjZWQoPh@0{vHS%aWil`2)iwb0)7Bfq36ibeyHMjwH}O<|HdB^U50|=!G8u
z4Rb^fl;_q(JmeZ0H575J@V#h!NsrO>+YCe%9fZ97y{nY+@MljXj46|r-cLeC&y{j_
z`|iMs9JSVQ2&`YubY|@sxH?i>wk=i@xMh)Xs8jK2Xn%Y@;myR&>_b#`pE<$r9g7?#
z#V0Pl_&2_?S28UgIfs!mk1#W1BPwaFL!UDcVcu>Tl!U&Ba)<Kva_@3CZY{Ry;>3%O
zO`Wdqx3ng^Y5F#CGGw9e9$#JKla5OTfNfFOE0G&1&1iSQoH8PmNl>K|4UcZbe0E(!
z_}D^T3!&Ha^T#5euI%xeK3Y)=zYzUNB!G7gj8oM--<&V98MQk&;VXh+lJY_k5E7xu
z&2viMb=$NTp~?SvYu5{JD#~`CW+q1pn2}C=hs!M41yb>Mh3+<W+vg5z_Q2@0?xVN5
zGCZxTv;64(IVtsN5CGZoU|gx$>3da6P%Pm-QXhG3*xDsow<fhVlyJk1asNDZ`BqGy
zn>49;XmK8);llMj9*ss>ockNz=^nEx^#EhM>v`LtHzr|NQWc)3ZFmok7eR^TjpxZ7
zp_3!rUam_u110?X^Cq<34<V9_k>TrQ4%%Rs+T6-hFr_-o>D~(U9qnj1T-3#sdVO-&
zp%gipGrc5XGoU-ns~@_@zbKe&Sf*RR=DzMsY|e1%eApN69X%+q4eRIKGO4Q@Exi?*
z!+U#8Z>S$?XSzYhM6+wV424M#t-8R;5aL`L%D;cAZK4iT1+!m7<)CgaExine*j!(o
zw3hUpjs2}^w=;cQ54Dqz5?M=h<3}BWnQ^A1fy=J%S6TKuC4&RzyUVea0W{$yf(S`I
z7QfH#Wpoo?#KySH?-Ro$^|FWXZ4ZsTI5JJU=H$o3!Oyn{Wr$H8ushUyZ)W+oZEIsQ
z*e<=$98NXZnEaqhP>R-1g?N0+cV`=rG>FOvPuW`yg7T5`smdr`Y5!{p%5Pb^!%tN#
z<=bN^?S|8DL3)(53>gV$`P2Irr|3~rc!pG1P?=Vw+QSom>O(P5D0B%;mdie>ou;jc
ztAHy=2k{o*-Z(=l--o8GI@+eA>k5X-t@A$GdBz4=VA`dQ<W*aXF)MW>D0%YJ*L{-j
zR@QLZob0mem9i1xMzsg8nJ?4cLZwHWnCpi(m;^L!?L`gQX!SnN-)z_bNQQQFrybYz
zmK16(%`4DMHxsY!TstaGv*i1q1q>ro%6T|_uWf-6XFkL4L-WTN{{Dh?pLd)U!HeVu
zL`PdpC00Uwm|VRRsIjVPRc!+A+9`yuD=H;MJ80L2R;w<)&w+)*K0M(z4zy_Hxoyy(
zxOlF<&9V-syY=2`2tgIR(aqo<exj`nOVyR<nT-B8>`WB|tJJaWxH#54TyQj3DI<Ow
zDuxl@tQx!s6OgF&@n-x{38iz3t#%%THB<+$b+c=yeF-;?>eNj=l?o-qVa*}QJJah6
zg2C03@N8jsTT2h?EC<sGat^<|+6fZhbsC?7QoX@BtN8AFkQHD%233uvmJN5w2yQPs
z6wffPc5PYtV*h=zFm4MI`oWR8Y84Mqp$oaMIPQNcO>tqQ;gn9>NV;yGrzn>Rmk}ai
zmnauN9e*@%(mA_*`|J#+tG^2O34)~Co!}I*QC|(}n$Hnf6R78ihORZn)#}?Zbj?&b
zw1A$IgUe8maP4}4wb$!Pjc-u=(gAiufGVkRnB|$6_hw9MDk%jKEA2uexC>VMkK!_@
zY`1sJD#xJ*rkv+unh&?8>!k+f^hKDGkd4uaFo+-YIW2b}3{z%R2%Z|Jv3X^a^2ntC
zH2klnjkTM(R|-{TEJx`K2~$)fLV)u+#IYBiu`NxAeCZ72`=t4*QegOs#d|aVp{*Xt
zVx7n}zlGO=IK9F?fIikzH)T*ySf8yXw`9An$2H0SaHZC*y6@@w3&D|7xbvjRVchv9
z8}lv!=Om$2S+{BuR2??NH)Ec0`)*dv2OcC8ofkeiQc$dCE6O!FG#B&d;(K?P9HkrD
zIV1PiMfdxj)wp8FK~Rcds;~_#t26{D=$CR%Y~^Mcky<L8Dv=S(85-sygqYEm6a&&Q
zAvsAZn81Kp6>Paf&%HK^yZ*tLq&$dSTfW9C)vY;4^=CpiA;eY>Tk)z0uBmaT{9T5y
z?2ZVu4la)Iq-EWx`E3zuSz(?~!&b<L?OCTe<lcMttXxt`Goog;2=&+S#~95zoz#TF
zz1Ggb^07`0ttG0xwYeTiqwYi~yViVrh|wVi&p#<Qc3N5s!|OkLt#_ZWHEec^iEnSW
z^k0t}X9?dYMpdn{PB}o)ED;0-vja^az~wq$<z$`7Thp}}hBCy~ut?FI&x004O1*(7
z*Jxw34&gmSr-qmbx5tYwcx_=c5y1o!NtpDq9Pbcg<iqnBV6y%5!~o*lvN|`UgyzMy
zf{v0Z;bTk!o$zhld}s=Pg{nmKUBS|j<g|Ds+8XA5T6eUnQxpP0)!tWmWlvZRI>6El
zb0&TWhNvHj4na(@!Jn;}gOp~873CuYl<Bz}gG&41k9Ci-O3@Cq6&|vEt<+?xD~wYu
z62x66s%NokH=m0l4v)aQ1bnfo;PtVb(}@(a)~;3GKiQ+&dpmEF$K_Z(Bg>PpQwCvX
z!FZO%BTe>D1zwt<n><cvJ<!puVJ|_~rj=`J<bMxS)?)tAtcj;&VYLXSLn|xg_&P}F
z9U~9^y2C*}GCsx(o+12&UjKaTZ<C(SjJd%N{L~5aLVz`g-J{u>INyoRsj^|)cBsUQ
z<8B8EOM7MTu%6w7b2=VfAU}%-kt-hNwyHg;Q+8+4hC=^bnG@u&GLKEnT=<uj>C3?z
zg6fAMdIq=V2a1G%i>^o)h`2${{M5GG(ucxf>$eV$->xdeV_4-E=~S>AEV7z5n8HT~
zESg!k>_9g>v{b$5vYL)1(dw%vD9}Vh@h;==&?A36%(}nSCAIJXuS?kFSFdyzgB*bL
zvwYy5-qYlPBG}MNI+<<;j5>A?<x#voCX8;g=WDxI!lksaL9f1Q_G;&0X^@mcg0!LS
zohJj_R)n?~-{c%ybhwMtNw-(|Mrz~9IcPmObo#L{D0({+n^rBc(z-xE{LM!F-GW?J
zm5XJ}<5eEqWumw9%MGUbCft>U@_t<UulWs#I9X+>cUa8VfS(i(Nb}5;g7qb{`iSOJ
zZgg$_CK$hz9CWMPQYo}B->Y#3@@3w-`t~#q_r2zsQf@lKhm@7aqk&g$n)I}mSb8uv
z9S+5qM1RvE8fE#CTS!eMbQ<_t5<MQk<WkMxdDnO7+u$~2&+d^QHMTTvz=Ejh>JRgG
z*>V<IucV!)t~t5AFYDGR0pBD-Ix6`|ZbBCUv&N}&u=vSB`;>>I)Pe&S=+x!tt69fr
z+*#FVOYrhkWl0i8ChtNiQ4xU-tf^Fvj%4#UHM?=g68{k<q+BE*C{=$ut^Y5fgW=x2
zz7(iH9ofoc^d-4mF3uS_CboB(sSw|pCO4z2D#Ko<YGa=bZn^|mwr(gS@ou;3U=O_3
z)+~~%%t3oq3bHOzQw_2BeS~&WC`!Yv4Q;Fv!5i0>{V^lnGSJnTTiT$3s*`T?+a9kV
zlQ}x0ShP^8IX%pyj{<B(>-Rdcp*cxcaNUSkb<A`yecywt;@>m*!>+FJau&-4ywt@m
zPH_8}U#&^&eLnhnTd!V8%}lq@=}u%9TCI}qQJ&v~cf%x8`OS<ho1#kE=Q(Xbvwd*;
zs^K+!p5JNEWX}kD-D!|aU6!8f=~-4%b?%ObZeqp^UA0b;yx~1?1)W+@w^BJ~mt9Cb
zdA#b|yGk2@cUMTivAm#PG3eDkHsSV3yn=be)|sG`*RRjP?n<t6_bGngq|zfgs%<(8
zea|(8Gd$?Hr86<c)9)+s$2}b9gt<8P$bS*WHGO(SsiVPXRov(=rm!IB#`ZsPHd8w1
zULw@>_-xldWA9x-oj28CKUDvl3(smD-rRZA-pil8+`5l?>@@1^td@I8&`S*~nND<&
z+`Yo2FEr}gXLbjL>G?$P;o*fF6uUn2k7&r789i0JF*DvIxii(_E0CQME$+HZp7teL
z4`?cz07v}_wx@cW{ts@b9gmW5cd47!E4bEqjq-ngTInwwCpNu>A0F5YWwzg4!ctx`
zcU#n^T!pSpi_cyXRVzzTOI(U7_Z_%E;;Y*;Oj`J=k3Ius+n~-Pb)NJ(p%T!76(=Y?
z%Ub>bLQ}2n8kaxT^N$zaMik_zqimYtC;k}Jzr4_REJcJ$REY{J1_OvptBwz6j6eCE
zx%>Syf5hBSbz+{~XaZC|ESC;8>ixM_kk6K)hso*k8(008{o+;XDbo+#;R^xO+!F#3
z&94;_Yo4*MoSXUMu>H@AF<IDmmtF{({5kJ`dBnej^4@7Q<fy6j6R3YVqW|*iFrgw>
zr9d9ntRZ6eH~7ng{I7|7ISJpbz=Ig|zux<=|C;wDh-juhX$joiy~zIkn~b@@*G@Aa
zME(77|M{Jll<-WVcW{A`|Nh*&pD|w_zV>7|3iZp+^#90I(Be@2n`;X#&{J)G#8aib
zdszN?cz^tkhNj`Sq^sl5dej<N*l#zOz7NH&9PuQKON&3^&e)%6=FdKG+lEM}+qNRp
zX>QQSJ;vw|>NZqP2)!yq+<7kj>~H_)@pk0A(I_6he0$A@LMa5LM+9!ZLGA%f#?s1P
zUB<8lRSY4uv}k;Bu0kXQo&!m3EIK}zR;;6b{Yi+nEeRpJP#&rZqQ@B>vP9WxD$ET2
z+qVDP>U<f1Y$m69agjgH)}OqvPxC?<Lg4#!*J1j%5B_r`rD#adD2KW4uJ*sY)OBI|
zHAaiFvj6)Q-L?6>d8lEB57w-V{puJ0l1E7B6)oy0|I1UZ$HjzduY87Y{?!S7X@{H%
zjx&KT{>$L(`cb|noCyK!eXoCcXh#pg(9U!pIro!E`{UxV&w$yZxm>L9%R}2p1w-4N
zCvf+dXU`aBk0-|M@h=Z;^eGtHJUg@RzdU=DFnfApYYfr5#_xZ2=RbUm$aNT6yV1ta
izdU>YcWw;t9fFEM+1aFzHqj{f?}FSV*)(ZGpZ^10FIEWv

literal 0
HcmV?d00001

diff --git a/_images/IT_sample6.png b/_images/IT_sample6.png
new file mode 100644
index 0000000000000000000000000000000000000000..bdb5354639b0a6292417d2ecec9c7cba4ecc5c4e
GIT binary patch
literal 365253
zcmc$_bzD^8);0`+AWDcxHz*)Agn%%VA|ObLbjN^n4-F#SAYDUBhlDf?okL5<&<#W9
z0PlFtdCqgb@BQa_{O6tD-m~wy_r80rbzN)SdqR{Hr12h8JVrx9!;|^&{u3G+&LtWe
zrrIMcRE~AmTMRU`$HnFn5=t@>5_C!qwx;G*AT+cOA@PaWN-CP9UlAUFbpM2>p>ZRr
zbm?fH-hIjXh)qe4^Het2H`7R(MBR=fIQT2WR#EUbOLUXcU?W52pHq01!7RH*ZGrvw
zU+L>^hVSoBGLm^WJvP%l4(eSFyc}uJo`cFX($Lv3qF?pLyub9|U0ER_yF~wr#+Uiv
z16t2IC`@yCxeYD;)s2gr54QN~dESI;nUBxCzGMKPcLN<QASbPJcmtH}Ni{(#T=4=M
z%|u{FKqrlnW)mD|{;XO&4iVwl|AHXGaSYSA^}XC@OiB<18YWL7n+*C_nD}kDj^5*x
zpuK>~h+x3;EOx0GMqb{qsAIyo!4b0G(lo=va@U-S{cs1SK1YZ8R}K)J0sp6-{xd7M
z2V?h+z`I%#(+~L@TyNvkSaE{IOIaJtF2<_9t+C5K#!EG#_hIn<rt$-QR0TF*$}Gp7
zu}k!VHkFV<xU27-SfsEbKC?yW!<m*C752W+=;3&F!>YU_wq%HU3Ju3a{^K{tNVh`e
z3Y9draGdm&z0=;zZL*qqK#WxHE@?2bIe+skwp@hwrt$2t;Bkdup)&gMFUNe;Esb%(
zs@1h$M-N3?#<@q2UF{=lJvr^A>V);8mjM@Mn^OHf8=2S+?pAf41mktS%PP3jbWcn&
zS<!KyhKpXIMKfxA=WTfkPH-)JpY%PY!yrpQu5~Hcb&{7H%nv?C3x`kjtsW+b+7r-U
zPng5Co}nFNS5#D}jJISCqft#^e|?d9o_vm$84S4@(02VKbC7Gh!f7~a`}J8Sx_ssn
zTwn5^55TQBb?AVlClB6|1wS}4Kwt9x937(8LdK45DNfIVmD<LygkID}V~0kCCH<E7
z&BGzTk4qGF=r(>q;#k*hAGvTB{gfYlJAC>u`&~*jjR~&iJMdGY<FB4s1}qOkgPvtQ
z|9~lzwLYR%K*s9Jnn_WBfBjGhCq1Z4d|||^fRw_2Q9Q#Cld0pi1~$L1S({waL+m!?
zL(ex4^)NTv9v({7QI!P{w8<a3T~c^bCbR*T2{Gv&KcXI$E_uq-F4Z25`GP!#jxB~E
z7q?K7B3dmMqdufJntqK4D=5o|jEpUuk_n9{V%d<yP|b+nfRziIOFjt<la~XSlAYjv
zY_l-ts3_CG=f_X|rW?-DnP7NfsAvE%vNgV`K-j;bqBaY4?0&ekVsBmts>7>GualY=
z)Db8n5ct-!%6F9D>f%WT2^(%BU2$0=SRP*SIJ~`#!hHCRrY&+w`U73?qbq-&_P6ci
z?+6UPa>;&Xq9D=@F?=s-|H_`o9#9eQnm(;a^|CNp4z(^UpD4`n3e$dzH+*n^?_Qjl
zX^^>;%_aS|;HOrW&T?=&C_pbj(L}ZKVX{0#I^ny(_lwx!7(^`m6U%d=P|}ya-z8=~
z6z1q;3TJtK&SlkNb9jCgTKt@iIQ6;5bNyF@@!|0rvQOfU<ND(wU+Jn-d}3lvWvPm<
z?`Ml=i&tcYtI{ZE6-^iU7n&4VDG3%hNUqnj>P@_#P>(Oy%50E6`({B)oJH|z<}-~m
zhcn=e;a5grPV(5B>epJ_`&^}A{3Urs8<`|IHSbG5?N_ISM~=vEQMqGz5#6%f;@?&^
zL<)V>&1}(Ye%ma4>(`Bo2f*`{ijm6cwvHH1!Ijq^bQ<JH8BB@##id1}<ycK(xj)uW
z!1vmZzrz~KR%$dnhnMC{$1KS#;p{;5#TOTgsI9G?%iUmS@RrU9`Ph$L_wBsxgdNBz
z_3xY=w=JAA*|VXYW`hE5jc{Bk!JKsNlJZj8Ou$6>C}ND3kBcvuzgOVPYgeleJSG%O
zlnE3y<oJ}^<N;(H0!0Gu26C@WDe@_71rE6G42&$ESym3^&aBlJ2EOWfJzTwb8bjU%
z8@GzlES=$SOcNXzUhvGn!h15^KFTnWJ+{*D(rVk}RrS32j0h2M9F+IPbRX8JX2O=B
zTBh2pTF$p1G%W;gn7N){&(XxGs)6k!oeg%6GIf}i4@zjO&WX=?&Z*4_fJwl@jk1l}
z=g&RFZdA`X&daIUsWI^zsSX4`QA<!w3DyXDJ3D~Czu|j>Cqya8ETk?72aP2y!|-4O
zoEoIIiK?9WI->PL^&53G&bI@D{TvJXnR6#Ic~uXqE`|;q1?)B(rs@WS$!J1o8bwSX
z*ASgyt>JvVKt0cLFFg!)M>UGwr@QgB5w)E+{ukIaY#tA9(g;>o5h`or>%}{i``XJl
zox`nwuJT3OY51s5@_jv;mAF!C!q-MBYWygb9&&sqd>z7Rn)mpjxJ8fLDXa+DacuB>
zB_2rB1PVXq!zT{l5Ax+wn4bLEvz)ga*6x$;mr5j>?=V#LXfAl2kUHjfZ$Jz-BV5`|
z+B(#o%$kHF!X`{PG?`R|^n`wrx#oo$i2$38<oqw!QddLIj}JuYnGE&Ydv)OTa~tQw
zj%n&|5JR5hS06X@H8X$F8Mkc*$Udom67|xE$e4(L$$L4_-m;!iCM~xp*N`T)HvQr|
zq6@H2@Ob{=%Wu9W7i-_wHeVgQ(pJGz5m6cV6y$GZ7;xyUC9dT>8+7{f^mf~ryn;L;
z*&o&ki)Bz%ZYU7>C}1h0Ry+Mkhb4i-LATm^^yiBYl>o|qNXEi<2?@#^g~H@hY;)T(
zX&cZ%PZxJs6o;}VgVtF6Z=>z&9~f@lZeoBA_Eio`?J}t6;PE1H2{9jAR=f<`oEmu<
ztHuO$*5}NZFPCrI(tDt`576gR%MZ4;^i^ov`PM><^eF&*W{-QmsJmHHb}bEyYNu_S
zoGXV{+-Va+w=62;?KO`Q@tW||eJXDYk@*!7eUy#-g4VOY$ZIvLuZBp5Ik06v%O1;K
z*92Nwvx4GC;#=lw?B?Qmf+~0RbN9#gE033s6R4$t{D{j>F<SPGV~v#+vW*sA6G7?7
zbdkm;FyH6h&!p-;iyZS_<=xP{!pYnkD*_wE1BjW2@i`a?L)P15Ot!3e>0gAM58%J2
z`cm##RtzlgvQrV!0hH1TzebEibchCNQ<P9xP;5tZNw*Si$B;r_+&OH185ylHh8g4a
zL`12lXF^ykzNIAa;|b*R4`{7uHz;C3tSp@^MUINk%5N06Gm3;u!S3VaW69rmjm=Jm
zBGZU6D1|Z{TFR{r=Q^I)liK_mHb*p&IUOCpgnlFy(x5g2H0p~wcb)b36~l@Lr_##!
zoQ1%U%fX{sB3%Vt`Kk2s+IhSB(OHqk8@VX6;j2_p0l>n_d;=JG3cg6+0$v;pHL^AK
zKrfKq-M3Q&4z%H-I=7E*3brbaS1Xxk6)k)O?wgf2UH}(%$LFqIA8&Vv>cI!ZD(^4H
z^bf<2ORU6gGY{$)PE`E7{RH%2UZeM)kPc1-hl=rH#d^ujLU)N%`sp3^9k=B#%X=2B
zT#K*3_rM?*YsO_p#YM9g+FPYl<&#xkMVKOWhOrnyIOpBB`%iZhNXy;YOxvh#o}TVb
z9UPZzVsWK_*jRMeDq1vIVYJ_rU+%eFDl;HYzZ}yVd~5Yt>4SK@OGEo2KIp35DT>`n
z_*q^pr(T)bHYB?V&B_)qRk-Mqtw>8sb=^}rXO?zg8^P%7(`!O-ue)$xZ6I8XahZQh
zdaI80wPEssI*Pr9f;41I<>k>>QRzo$nCKK}4^b&})G3Ni`R}w8Itv<#<f8H)prHks
zqhbE*83ok!&nFsn{wed%E5`RgG;GwbXQ<OP^TB^TjdPia@n307HPk({w<;1cGN`MH
zu>%NX<M_qa>9a<fH!1_i?t`Wy8XC#VKPS4(rx%B)`lro5YdC4h%Ly6VT5}qj*cyR2
zU9Ii@)PpAMDuhZ}gPaWMT&=Ba9EDs(82<5u5GwsA8^A#Kk4Ky=MHn>XmFOgF9YA#a
zoLrn-45E+e=;(wUOiYD7y_fn|anvsnhA&P|c0vGvi;D}V3ooawgBgHZP*4!S#RK5s
z;XpmX;pk@LWa!Fa<H-2WO8%>!_aH}O2Xi|ob6Xp_KlK_K**ZIkFfjaS=-;1z_7mi4
z{@*RxIR0x`r~v~0+yQWNasmEbH>#-cpIjj&b61d+=6iE%)SRK(5as6P<`Vu#f&c5)
ze>eFbMK%7rC=VYW&wm#Ek6Zu$qN<J{2MJqiRHII!|Gi-UD*T@}|5Z>J@MrA*F&F=w
z=zrv*W?J;IFyP;-Ci*ziNlgN^kW}XH6+fe{C^7r<dBBGHV)^G4m3}Z{xMJTIhlVDO
zCiDL7XIJ#SbnGUoUlX+;u8BH>Rs$6)A%D|%%-o01=uT<gAr7C(zVpF&gZ}O-<~aRR
z*@7(0N9iLTIL%W^R;>m?SaRIfq9QBfQb3@#sgy@^%LU-R%nTeV<^DS|GBQ^$#92xK
z^S$_A9ho_gPQUx=OaBOZ!1#cH?r)CBueiuK!oO`F1E;sO9D_*m?K5=3r+;&hU%ny;
z{`%-|4sq;f(Z2fFOrhUi{LR2o)5UYBLi1O>CVc9P|AGz!iZA#4q0C>coH(Y*QoitC
ztr)7~)E~sLys48|9)JI<nfv;vw^h;Y??g@<Qz#lGGD2L+&tLrAk`qSzW*q<Bl7D^V
zM2Aj(>1>eA^5O5qoq-NRZ2fNp_0L$h#8D!%+xlKP`Y-hbwQBtsf7$U7U(9TZORtLs
z7bcnS_~ChfE%o~^M;cua1+>8yOBW*&1j!*p%u;xn!bE>JE4~WMToa#Fl8Mzc=q<Q%
zC@%5Hc{;->1vo9x)|3BII-P`kHBFPdnl2VSl>J^|KnvddWmV#eFGh~Aycwm!xe(BM
zZ$&xEF3rDE*Gz08vNv{_ZcDz{+0TL-zQN5hyi8l}?`{gv<IMhcLANpUl8hdY*V@7h
z+QKcCNcl{T<^HO7;+XDSD783S>Qw(N`B!Q!l=w%d1+1U*{FP4o(#WAiTjb3VvGd=R
zA9{(hrbF_S0e?3EsO3RH0fB<2r0!o$V5Z|A9CI9(qW?QB*lI(W^C>?fePz<W(xNS$
zKZa$#TDkY{49oxf<<Bg|@`n4|H!lOCb{865AU$VRL4-+MhToP6rpoojcyCX~T{Ipw
zO51&|X^<AG0h07H?=!LA|L>$Vvl(;NVoAUhM%JNEBj9;y$7|5~Rk(a>2A}?LIfyoe
zt-NV$-UHD*^wV+w*A)M5IkY9&Fpym6h5rIG8LRdJGshe`(RwU+EFS*9)7cU42iAOt
zz^k>`!2;zR7^%Leq3GSEaqJ5j)9}1B2U%XbC2`*U>hY9V1))Bx>QSlWVdwTPMX}r8
z5B(;#-Loy9-^VaYr!YndSAGM9zRKg;OmmE!v1#0AopO2>^tE4&Dso{V@gPaDfnkth
z!cq6g%S+;aC&<FSH58Z4CquxE?feXW0x{%I!PRhH*mQ;QtxIua7-vJljCJh-j3JEO
zh(z0pn|EA}S9JjqN1NeuS09EwH}q@~vO&?6Q7UehYR!LgcY{zjB<D8ruiu1t6qRjD
zyFFmlUMD}tq;1d^-41<i`FYs6>*5UWf2Zt^aH=O-0mzZUe2B&QtY!J(r*&k_oDHwo
z&CD<>6nM1Oh-l$W)N?PKF-^6G6&06RJaSZY7+kquPt<QHjfdV{@3OKFaMY%`9kyW?
zmpNb1mV6Fr%W)}Dno-EFoKR?3=t1;}WH?F6zAoNuV8GqBou2KyyW#ZGgrek3i;gQ`
z+uD7`$Kijv_FRHQTRa%hAR6~xb|~^VxuiRa*7o`Wk>OUZ!YO(J<28-oo8&494&XLY
z<UO=}d54m}%6b$i;7Re+p$0ii?o8vcSCRfbT=!Wd3d@F@j#05yTaNm*@1_mgdBw;`
zuk1?V)GrsOwhqJ0x-qP!za4c6BM&KwVsSQ}@QOC&I--as2@Y(;(v&woxuz@sEa`uW
zs7_42oV<q~=QCyt=Z%_VEms?q<0#ZG(WbVyp0Bevts3O#MNx^2CyIZ%&UZ)l^!9oR
z%8zKdDKngC2q4kwO>-O)G@aBm+YffyD-ut0p42p*(se2IJP07wvtwr^WymP7lEj*o
zZ%$1+kx_pkQ+%eQ?*9_mo!G~`;dr9#Dzk~2XO3rEbC%J`-KXY9<|0i0Q<|PGgF+Me
zvGv;vxBxbdYo;Kg>4a7B`~?QI>F9p1Y8V&G`u6&a_X%fR=JrJKrga1}22@tr-1WPF
zE#Ik}K<XdA!0UZ_8YyT=U2MTAU(mN4de`kKa%9PS_F835)j8kd?wx?gJ1p-7_O~A5
zPAzxWmb~^W?|BDbkJrWj^bn6DmL%Nz^ni#-j!WiyQj6COc-!TS`@f{!_drsdNAf~e
zu}kWYk~Vz%>LjZBRuld&SHkx|F>@6c*>b(ZcG5+vZ}QzNUJSRV<@Q|r<fw->by)0v
zKHbr4);ymN)n06}MbS0JG;Y!n4ZM{=E!h~YR53&JKtv5f`T9%r%`IvQDAk;Vj(Zsg
zQHrfc42f6M>eF$aP>r|5z{2GczDMP%l?5=X$Z-cOlT*zR`W*K$2dX~7*=_a3SU|%F
zMaOJn??J;Ht~}#FOgdUQ%BAfHE;5IEhFzXe`pn2C9tO<yZiI*-j{pO1Et(f~QYq1k
zxi~(!b<qj_0Z#N*=2Jt$-|y67;OSONQ=H&^cI~)~5kl;K^!2Vy<6}A>QGjBRXKa3W
zfc^32Z}Z=Thq<`V2>LOc#Wh54PuwC)DK5K)81=ocek=n92%+SrNo^DBzXx+)IisCn
z9p$ws=7V4jJc=UH)X76(n)aby@>t##A?NJ(5br(mlPy2R6k0&7KY7dPiU`{=e+>#H
z?PWw~q%SLaXm|-kFEcf5>epdo9|0mWJ<^0pPMC{so77Xak0r$Q?Yqg%Z~<p^C(ygI
za)s_i6my=Vak;LOV|ZKBHON0_rO;x16Lf}R_4Jq{Q=_)sqv{9dK<U$HlY_A@EoYP3
z3hD#WqtZh>1MIrjUS3vXPsJ7QMy2l;_zZ^xH?0b^FS7PzliPh%dJZJ$;wDCY@o9wX
zj2VLV`XAtL>61f^j<C8@n--h%FPMdxvY3>`Lno7CS?Z`SG`Bs4IAxe*zP9tXVvAo>
zD6RxCT92YEbvhn~IvH)(odl;S`dRTRZ~DAfoT%gRhqg%GLu1N$N}D?R*u{}7i9ek;
zhIk&!I39J8rb^Seb^c+>3ft`#_H(w-X3D1nDKSxpZo?uMNj}I`TAKhtAp}_zOypwT
zw2N5XslRnK5;%ms8usD0EUV9RTa6Haq3o1l8Op$>{Cpb8Q8S(GkWPHpd+2jDpwIdk
z*)Q+GgTn9BKM*#h@3WxeG%7iy$SNO))VBT=KV#oRJs2Ud_~q$g<@`pftxbZ45l-#d
zejjtj9@fiB$Q%lHa193>$cCFPSHcc}_jec2L2O#jrEm)_uC;Vj$!X*ovlw^PuxR6M
z-FnEq=gFX%p4)nS$#4qo{bkPvDfCiuPz2ZV18@a4*-~<Su~S^$5V`u`3A+J_u9Gq;
z4VD0o@z!j-bh~YwAi@L#3jS&7GIJb?5x%+Tc_M$Zl@ry#<9|#nA=b|Wf*ik|(s4}2
z6PO=ar#*J~5<L-F7j*T)mi!64>t{sW0*XCsWZWb3$wl0&dG%f>#FbB|>n|XO?~zf(
zyP;*FGey4oRthc)&WZI`8>t;zYD0y!D6^=W2uaihC%d6&hYmH@J@TZ{y0pG_uX3QF
zVHWCyn)Ua^Y$v}Rf1ii~u75gpRmAwhJIk3;j%X(YuFscu`6L0nCz;7b{eIA?S3Oap
zm(!=a@TpTskxM?N`S$=urZA!?(FH2EUY3(Nu|3Ok%KFV0VdJ^icDNW~Ys5@nT`}-h
zzUh1;blBor5x?4}z8F6agXZTv`psx}Ix3y&M<6;)W#36yhEPHX>XQ1JVn%C6CpQ~E
z!?_*OzrUWk`xGV2^Hnw@dvUFrokh;uuu@h}Fk1A0K8R{%7Hd%q1$Bdm%N<psS(Yf9
z{_xQriWY2KZDzQvJNCUye4U>T)<V}m4Sj?Z|5}5-!0t|d+PF90?vLnOAhzycx^UgZ
z5hJ=(6kLBkrDy!wb6j5F*Y7Ac29-6OM93i1*SpgW0<n^vj37mxkfbAsf{UJ7et0&9
z2x=CA482<--Qnb`BsJpF5ku6B?sr@z;t21Nx7hU$8HE^HAHPo|V$m`KyNLr&{X}#(
zJCa@pT1Aj-Z;#}LQLw|YTosw)Tk>MZ{muzJ2&^70RgFqfCd2Q}YiTrp8@t3LhILEa
z?i9=vYP&LdZLGzFQ=2UX#q=P&FMwuC&+y~E5g7xpUS1l<7gy(0`Nk}kH6Cd3%|&WT
z``paIVJMb_cz#Vbh)c;2!y7VEIzFGXRgXiGq1n{Tm|__|fhLu%nLWHg7#gf%TXWO7
zBaeF09|T_<Mj-+sH5CBOr{8BliMp_KmqqeLhfQ@ow;yenSU<$nw#_%Oueh$S)&V}-
zxDU;z$0CQ8P&_HMW$*cCvUF+`rBJw%39>5uk>J2vuB5}|8qx9d2nCB^DQ~wG=z4;7
zN^yD1YyjJ>eS7Rd?r}4U!wrQ~@b!~hWPdO=W#9goVq6!pOv2`}k0M-H9g;%Pi$4pg
z-Cj^1@IZkxtpNVUG8Sz|7<bS>4w>Jl8ty~Mm>QJo0#^mXS?dDDLm4wlP)tUl4Lb7f
ziL*@0Tb#7o@GId!hsUd{_qXtS=QC_x2JWpc@go!`=Z9@&;m?@or_X;CD-I{I|0f=h
z>e%^St<v6WaBJaq4AC*02k^vSr#A3Y*-3I=1c@Qrx)#{B3J8oR;@x%xcifJR)ElL+
zycLCy;{;c89j&h}4MAk>=Y#Kojgu6o0f7!)*K)RZuI(Sj%>qSrsgVgS7p0{a1R_gc
zs#Xu?lH5p>8h^Mn>`GK<rU~t^TQ0eT3m*RTf{9|+`uxgP_?SqB<=xBJ%Rr?JW=tJ9
z0)2YFrR2b?jGi3mp60s&enr!c*lF@06zA`5MI3ezxbz$l1H1CraJUNhP$E$7c|7kp
z%;&NHD^YMG$(VJ@eFZ1>+m>zirPai!kJ&dW5O0(*LmvwI5lr|zPXn98j5>pb^|mxo
z+ztzK<S2{L;1g#zVJ+O-!!O_YBPIor_>eT9t#fMt0G%hl)OFwf5uu-gdq9FxCBFj6
zCk)r`Pu+5Fsa@umo$N|W9%;SL3ntm3*&}>{JEDHvAmLOLJ~)Ei!(CIc0G@LerUw+m
z9R@gTp5YkNr!OVV_g<1b;pRH$Ls6=fWHZPN`L%tN3|n=rr1N<CQ|W*#D&K5p)B197
zMy*HXP>}Z}IyRMohJ40DJ@?8J>#t^9uTJ^WAvDz|%xQMtC{>U*@cRR%k~|l!Q)E9)
zV_1{UYnmPP#KE@<2(N8k=GfQ2WWzb_hlNjYNggHp)NmI*_T#JjNTTPuoML7j#84qc
z(-%!I@r(`g1ybS097r3IZvJ^aqcroGCEY4y{R^dkC;!ufWR#c!E9MMd*6?l}Xk&=E
zEU+#SSw2miCf>4WF~nP&7dgmq-zf~)muqm@6MtLTfe-vGJS;G!Wm(!eB=^B8sA{8y
zZ{8j;r=a^X&9JHDZCK;5&*U4<iaE63mu)AL(PhVJVioD-CMY^zAgoBX#~N@@!<Q$s
z7ar0tf;=?rqx1o8#Y}Us7a`llESEO@J_%|W_pC{aT`R9^UQWm3uw9g9s@Dh6f<LbL
z3lla>yS+dez<NH>qQ}tv=ZR;P%QtTQBs5(nQ>vhWvdKh^AMXO#u6m{@<3flOg)8#*
z{&*K5e3)fMq`U`<tJxqJ)RverM3qKu#Z>**#}eg?MtNPy<R4UrX4oWb&WS$I2^9Q4
zOC@sVajuDycCGGC3E`V`a*pClDhw_687k{mA|BLm1}(amu|Cy3T@6F@Fe7~KwKm|T
z1p4VE^=&(SJG)55b=#ss4`zvtq|)P54@%n#J)Y~>eBRrFO&4`NpS_7j-NiQ-q2ogQ
z;ksRn{9z~M*Ory-{R2$-={`D9CO4`)Sxz#)iZNb*yA6z(1M563@Gic8<gCx_)?Z@=
z`gZokR*Oy6m2Z&iK1Jcpo_}&DMGN!fo0gfJ2Ec>*M<cd*Q4`|}Sd>=LB@H`8^;{zv
zx>glk4$VZOa$@AkB=st&O%Xa#UlbN^)(<!%m3n&)-lK{9OxIc9$o^b|w@dO=D;!hE
z8RbSsS)OYVCTAySG{DnZeeO;&Oi&0TkMdloFQiijd54AL;X%}34M#<xEu6+(Wg}AI
zt)!<jGLs<-5v>s}Sx;nu<zb5YQwWGOoo)Ke$s<rt9rLc|Ui+J}^k0qYppNY}b>@H;
zQwLds#wW%`r05pf_}iSv4l}L=`U^Ge{4?(CUZ!Tu)WtDQAp^!b@4oKI>Xx-#V9`+R
z_v<Ls+o(5v+_TUPNhd^<1q|ESa5S>nSF#m;Cf_zpGpp4unBT<o<89=Ly@3cEo)B0y
zpH1>D|J3HYoa2*jY_=Ob@yB7L#+}#$xAIAOgV)PK#mHFlR*5Jsk1}=nN;k+CWhgHV
z?LKsnUiU2d3)pl58i{Qtynq-3<b)0ciOhEc>{W8{`kskWFP*@Wot-v(T4z*y9r)S0
zcOq?D%(%6>BKbxiwr|7YrJH;zBGYY_q=k)t?)F!L%ql*wV;_evJgfa~LN+t2N>>b_
zzgMCe=aian&HxQ0=@Rz8%5HC>NpK+pRQ58&yBbDYniXFUZwe_C2~A^<4Gk4_cJw3)
zEew^(?>!Da;BH)SG*Txmb;GBJxYIjv@AeCy>5v7My|Fb}N(4Fa+ij4?ik!nil{_2n
zYwJMWlz~>nulMYg5t>!U?rc*=WS3VCHtNS8_KLD`oRiak9RSoLS)j@=W7zyorqVhn
z{m&MXcq)$E;{f~AoZp~DM&|z3@&UVD35Mv@uUHHYjN?z_A332s#Y$@pqR$IEHQ}mE
z!)x!MU(^dA!Wgy^3+7nfTVH^EOLmp=(=<=*CXz+yP>eT*n`6g{{Bq)3`t$3QzE>i%
z)=N{mYwMLqO3;QHtMgMfluNQlHDa%QIxDqU4@YhYbeHXH_%tvMz{M_?{JP5)*S)Lo
zD6|{3Br=LNr+3E695%~l(A7H<n!LBcmS4=K8G0z@$dM<J`<6cUH*m^?9*~#|0>vOG
zJD&Z_s>1T#Kw<wAfa=hG=vUKtRkcFoIkOeBJEHl5!m`zff6>*r76dRfn(N~odec#Y
z;_wHff}3eorm1>0xm1D<xflJ`ggwV6=4KCfi#Xh?+S|{2tlYNKOHXyvZf)B;67sFy
z1`QDTG*Y}9B<W-i^Fet~g3!yTQcjH>5vR`E6w5LkK>faY>Dy21Dd{2`HOZBsHT}Od
z+;Ue8b7K`2v!|rFu^P+RwKbQ2lvIjTyI38;NJO%Lry~*(>CNPNJeuCbh1!SJR^&ib
z&};s@$`Ea4=WFE}H7MBHrg;t(N1q4l-?<d`sMT(r-F2<`XIQD4s}r0pHs<J_jZXwv
zN$;iWk{7e;-a2c{5P!N=B&sTdsSK+=7?EU-sM#8P-2Tlkfk)d4S}3@e<X35<UgmYN
z@3N6YU_EEvIH@8p5>L+c6MA(TI}F>*@Up3$zcid{FwaY#)wKm#!b9}sU(@=$xeKyR
zfL=|?8P>)_X0PXL#l;%CPM?G4#t`ywZ{E)>!+n=gqTF!L+Xy#f7^UeTu>c=4<d(Ur
zd-i_t=FYH&5@Mf+erGlqV+|KWwsv>fs?HB2C1Ve+)ogAoGSm7PF?pT`Ec9OjYl%<g
z{ID9V%a67!Z@dl~`FQqi-?Vl#BCPAgkV`rJUdFc&&}8!MK8%rFG~+Pf>NpIrPx6&$
zFs`N<GFzGE9k?y;GqB1?IeEu>UTj70$6TTD25RLwT_ot_*%);@u*x*ubhv$~I5<TF
z1<#3hfKYJ;?py5h;@(H*_ePs+h<kN@#b-3f@1WjZ)^{oGa0}P)3Q5~b^qtHwK+CiS
zD6WDH>vPT5+Suo?7-<^w=$A2Wd!Y}OE%`QkeX6spY*%ZazZ=R@NJF{TmXOVFd=B7Y
zhsf}5uwa%%OEr-OYvPxw!*booH*+}52dUm3&L%3rw;f&daTVi+nL4u}Lj3w(2pdgf
z+uAMy=`#AMRe?<__|Ul+_}6B&z=44MhFiOMl4~RvE*V#*?9i~M;Fx5~{&;NlOOE`E
z#e|EyTV1VYl(QRO=MeEGU?@neshi{sIpKKfYw)EpSv4vbzDf?qJ|=j^58ZEub`vhH
z@dC+vDmrbi<wf-!AnVgDDNW&w`rjijC1#_c&9rKUngY~SNDhYxE}y%r&4b~@axNNh
zwz}#9rFTwaCu6vZt3pPy5c@DmGw6gL4)_GWvQr_5hAakL0oiZjJa&dkJms6ky>SO*
zH8x$L<I=~RpxBG)gxf#f(cJluj8y@6-eTeAh<?e>_Qy71dX4Rts#lq!`Qgf<W4AX>
zkW*(ycorwUeBqM<W0u;^u%FSxbK9YLubkyd8lZB>{@+BPmeey&ZOL(Oe`_+gZ~Wn&
z^4F&8mI(Dx_c^G3f@C<)PRD~M$K0S)M}Ugb!SdThdY>B(vRe<TnaG$*1+f%|)peqN
zOlux%6(ER#mP8+Zc^>t1Y{NbID>&qE`GbBx>rptzYQUK>%}jwDxjghERh0>Zt4x?^
zau*r7-oIfD+#a#Mfsr)FDFV!51zzR!I;?#>R_9gE9l{kPj=3?eF&tb;-1JU%7;*{y
ziQ6HaE}I@^Mc7}kCb9GA)Mg95XTlNtOR^lWx&}aK#k_LmU&uS5+1}d?6Fzppy@CZn
zfmDVeo&z;ek~%Y<T>|DVd-yzMEt#UZ^VjD=CsL8z-$!$#Q|-bQooG3&cqy(_PxUl5
zOBxRv6gVZzW59zZi#+Vc{m-}UHm~FbFB~vI5pUd*{VxT^9M`TC*oGAjoSE#|*@|o0
z3ukzL&$Q4=E+7V2tOYB{K%|;64kSVu8E){}F1?}CYl7>qO&@_{y~Lwe(RcoM(jwA6
zF^p4RfSzPccYV7??RK4qjHgj23=Vna#(%uh>n<zo22><FzZu++1HJ$mPw-WofC;_i
zN3U7I?$7&(wmb7_w+3*myl6vPxV%^!Kb2Q>n=x&WH19_zdNnVf#Z=SADVM9u_W-*Q
zNk_pG<&nCgp?0iV7eR1V%BXBm9M|D~(sflfkEF7fX|M9;J|mYV`Q?oDC~P%KOivO+
zLIij7ompa^0Jz@R1;>vT*XLbOtKagl`)=u`Kcl~6<6s^yvh5M*@^Cr6=Y+z0y3{3n
zL96(<MNDulS^|a&uSj2vmyvzpdLTh=TRY50HseM*{sk1qQNzWO`_2Uw&1i7Fv7GNQ
zc(ja_6&Fbr_-%1L#D^C{jCEo{O}o_8ZY|RfpGTV7Iqp6eXd00I_EorN(*Tb+>vt5L
zD5WC&y0ygTKn4j2eR$l73hPX5#fN_Txh&yqv%xpto=UdYP-!(&`cpT_eA>>bq0Usj
zzzKR~i^4d2ESLBv=bdHMCU?h)Q2)g2&IOO8c*T>r3ih1$W8*f;2V%=u#1hLBWpU;-
z-j~b$w`z*V8?j_=FK{W>k%~UoDz1Ig`@w@KlcwQpy*FK9x;2ue5R7A+JI-UOo171M
z((#*kbD!tX!)4xMV>k%1^?c58SO8{K-vNNF`#}4Jj}bkCmVhWQ#w=N-_36ObpnGbG
zxoOIe*(8>5U?3`%cxHPN^`!i@vA)k71uR0q>w0JjcbRK?cM4k0`)tne;=abY$!BFW
zKXvq6JHwm5P3L|%6&6~|ZWf4_%6mGSKFr@^P<7ucQjCje58@iRS*6c`P3hle7Q9`y
zM<z?p__o*%THcdebKcxcY#d2Lc=SE*ek8m28k*KNyS(m5)SG{`t>?Y0>Nsm}R*(Mq
z`Q6Pd0@m0nOpFk7y_ax2?Olb{P6E{LHK2-uE}z%Euh+(pGLjo@8i$3CTa7ow04+wV
z0^M?OGk4zI=(ig^1Gu%qQ`}}5?!Gf#(DcEdZb52#QC$xR?%mtruqVS!pO0>|Da(KH
zYb=yB9;L)vMkstK?rFJ8s9(PZ!Xl;?EEcIRTI8W2rju<|DfNMaHvGtSchf>p;!ml>
zUx8n^PvQv}u>$WKu1Oq+_b*a4O;fvgbf(y=dmr3^JB+O`#IQV~fqlj<^Y%eU-F;4A
zHGMB<fwwPS_YOU#I-ZPG6g5{$GUjThBTBzD><A>sA}Wm9WBsMzOV%Q0@tftO{RyD1
zNtV*Li9!e|vDliW)Y?0uNQ09RX0L#Kh6>v_SRaflidpZmmM9SVjm66FdCrx+H^3sG
zNxKq^<L~Wq;(VokSY1rp*yh%_{ykMg0@$;Zs!KT9S6H3MBK6L!Hr8d^4i&CRu3dB=
zsCa)RK^A?)-JUBV!)AesD?(x+c0po>(lenjGj+H#58u`L6!h*_7@O>?HLPRVt-bzL
zrnKMhuC42q18v?ZUFx_h(0&V%fzV$ZX1uYj>|;7m?&rCso>kou@IHZ64sdF}6g%R&
zqhmH`z5Wxv^d?+xts5&lx2~DBVDk$fH1L{D_d1`kd4><CE6b%6*7G`3O%zc2%p+(S
zC30@({WK6<|0hyAkMjICe6eY`!2FfFmV^RW%iHjMRP~&9ni$p#FKN!EH&n@bdgml}
zfwb7yg}Ka;$Pcs`Ar8O0To&E(v-V=4%6H!pg<{*AL6M%Vm_*0SkwQTY{y5;Xa5c9B
zR1}amP9BmUuA&|ki{{o1JSN8SJWEp#CM0cfTMi&?xXy4@>i!cBo-(T_##Eb7H-zNk
zwqrn92EdY@>exyI&!Ef@*LE)&scHgTv|k<ZGUFP~J~OWIh0OilK+g<eN$t>i@uVd`
zy~jDvb{uxJPw`KED|M3H=Wf*hAfK$p7htmZ$AyTJlj8wE<6BfHA+Szp&L)26`tD}G
zHX|7|3a>*pX|boiG=b-vs9=qYHP1dC54tmk!&@LWuK5j(`xd==i@9R*>yKX5sBl*O
zc~D3Qt_1^vE|IRB%9NSGi2bQbHya~m{Hq+?e&q{;M?rqKs37;jX?nbWCG&6fEs4VM
znER2SPtYzf=%hyAP=Y{$^6NZmBX=%3zc+|%{ZDM$Tmw#0$~&+Tl4Pw>QvFJMD6H~y
zIbZZgX>O!=VhFpuen5>&{d$69<;9=9ia;!HNzmi=r#0K4HPD;5eGgZrh4l}*pMJNN
zTeZAn%4w-bZG_@AEZMJi*Vm*r@xjg!h1lC3d=@@qs5lmc{rGEm*vDG({SN~An@a-r
z^thT0{!%=)nh6yqp*)`bJA(b5z*`%a0$$_ZB6ChRA4Xd`m!Gar!d@G&zZDwCmay&6
zU?g%KRm=){LcLDz^?1Jrc`}TEbkVM<nyB`RIRMwwT3BxP-OGZ9u$yrm9<9fSbt}XG
z_75Cw*aME9w~Oog4Vm^(ySxU+R~6y>;6^@L{rq`I7$wNmyld}Wc>|?2#v@;6-VKaR
z<z{eP1EqAwBb(CI^;Pm(LX&AEPu?ea$NkzZYaHUh^|vAQ<8vzN{ZcKt#Xv3*ZoNdR
zt;acsA1HU6))Our?(Gchm!_c+*3LV)_=A`{D8!80PQ7_^X-uA+T_8H{quq1<ENKr9
zsdQY$puhDFL+p7z3zj+ASY%+k&D8eu0v7X1)3XgeWM)n}XvwVh@nBkVG`<RZ4)ID@
z<<lqEBlP}(7*9Wcex$~dNLcb`)4FKpgWC!m)*AIQv-cE(T~w#U)vuD@$sYrwY=rx=
zEwj$0l!A_LfN>G37FqH&_$j=%8C6V`7m*Y*W)(;p`gTR9bDeK~g_3*P88q{*!uJ!+
zOR9d#aMrTAmB&-`60YyE!au!8JNh6zE@fD;5-A26AftY7>Pm0tkMElhndd^*72r(e
zC*t(s1V4InkeT~9I0~bO9c4d-$vS0#JDiwNBc;D4wXgz;T`G2Dkpn8jR|uDh3SBI3
zV?B?K@cGS0f40?<e8JAb;Yf%#QnTvxu^4|Cd&^J5^@c`79bEW!y2><cpP$=Dc-h5<
zxWTDOnTN`LgnI)PJIw2dOKr4~?qgHcYxG1j=Lly*(9-05V@qIA+cuI{OwXpS{tB+H
z*4(mrev?{Z+qAES>xs-_9@KFhHr$}nDpgtzAiqB6_7S{)Q+ur&k9~iS?Y&n!*Jiq+
zuAgde8&QQ7TArE@cWs^b*u>aSgo-UrlB*<v)Q}Y%yecU-2c9ExQ)WVx3B_s4%e;!h
zW|41P<E{>GqjZqvjW?gFkij6f47aN;^}#le`?k{8Z_XoXbzI?BLXCI^#4RU7b-iis
zd&#C-^Nz`qi5k<m>1LcKUH-#^<BB4-dhX|93Nfd5jkGCM0Ff@MpEE-7h)yHlAZZ$h
z65ijZ5HY9woDCnJDv5Db>e~}zJ+F?vn!!cWQ?~NdE0UBcc8l>?Vr+P2T1HJ%ck^x8
zj788OY!Mq_V@ma=6}k~wajd%a>dGNvU4W+qKUmwlb8KU_@ostc+hKjl$uyR?Cq}Y4
z%4oUx#jtD_43G!+C}EI~P~L(9g2+9=18(ZH*-eJ3MyB**5Q8O2kwF7AjOV?dv)<03
z%=^}CT|UIU|CNBo4<;uvk1HGi3#aBS`AZOxD%ObXQ4V$cSNx7wSRBz&WN(b4{e)6w
z$jJbC3zW7^h8otC<j>df5QZ;;+wdJo_K~4%Wn0Kiieo|Q{*gM|Md<ZJIh7-pCbxHx
zh{FRz@sN+eHgTclj^XK{d}ecA)LvSB8|Cxwhi_@JN=?6``(1bXdSh3@T5b-siJy<*
zT+IV<iOv^25P@u^ZP9IS!yon%_2Q9@8e%Rpi_z)RX=h))6h@k!aXp{G<#Ct3%kcuW
z2HLehTPNQRle0`*b<x3L_GTOiibSVZ+uESK^riC4P9+}h+q|?#ZCf^+cDLfJ>oigs
zcDnXco?aaX7^S||G@a+gvFZ#a<a@itc~lTGeGiS6Hl$*-^$?4`Kb~|E#s3|!MoqQ;
zfKmDSphQYh?_FfBPjD8uto)9M+T0UrXEspWHP+@XRk%GDZpkJW|2wRFcxXwe4#jy^
z$3;>}?YIIkAhBqu`5i6^E{l@Bk?=d}07EW#c4W3*{+^pOEjgo9KgoKeH-;a*|Lvr<
zt#kLIHES&J7hW#mth-HM5UqFJ1=DPWRZ4fkjq&?DIT6Cun@eA$oKU*cm`oWpXe64N
zc5CW;hGt(7IHo2684CvOS|ju|Nx1u2h*-Gme&9>dN~X#$$luc}5L6QO+s~?Dz1}{_
zr^X<>#E>?jpC#YHdZVIdAloRrFxvZ5VBfYF`iipJuezRW$_T~{!>@T}vqRUSs5REY
z^QDh48H?OE1h>MZsj^ZznWz6NZv*KX4hm|uw?O0WEvWYo;C;fqB)xLI9KW3_bZ466
zsUsqjaG3Jb96~=TmeIqN__?31_8JKJ-S<+PtLF`|r!pJW7;d`~U+LDQB^t`Oz6z_Z
zsGjO=$Ynd`a(?CT`V2QPuX<ojv;UcSIv9m?$PjuncYM$d&9<xj0W>Bnl<d+C0y0rG
z0UQ(G;;wy~bvRA_ozJf1yNq$x*@jK4!;919s7QH9$a(tY)GvnoC^5d<Gj$mr=633g
z@BPds$@FYyr>}`yqZ{>)Wu8~Xn=Ch8_uhZ-uCE9pGs<}TX6^kcc*`o9V6@(zS_Fqj
zz>DoVO_)^Zl&m+ZcO_187Ox-M>m4#H<HOuHxv!B=Zw;DN7h*mh?|4R=$LRJYR*UsA
z_R#t?-IQeni;G2vbo9cu7Lz`-nJA0(@deUCWXJ_$C+rJJ3)zyM`oW?&x%bJ$K*uTx
zI`Q5$TZl?$`Nm@~jTfFhKt;bxHj2lsJ)hR=k<*(kmyv)>X6D;bww&G83Z-Wb6i<i(
z!ZKaOLTQZq2e@b36oRE^yqz2SrS<qhJKinCj5MHr7DJYCflV(&fB$pIv*f!Qp_=5c
z#F!18ns5F2_IU5D%o>?+!6QW*nsfC9ZbL)4<!(c)--{qsp@*Gd1q;Z@az$usCnsLo
zG}ezD5%1}tkAoiI)}jqTpElMBRD1_mQ?*HJ?piHdmU6LgAUOe4Q`S7eh5uwjs6wjJ
zVKB43OtbZ5HHG6&c-af36L~hS&!Wa`lPNd;bKp;G!<R&j5A+7BY9vxMeCNf!cfuHn
z6;snpenbsRgCfC_iA?%sK0w=<_I!<hxPs_WofHi(!NAQ<{J?r15Oxd^Pzj>345&y?
zOc6GYhw&{Lbv0X&=P2~{^MVsT83A8A-|wO68=2Mo?kt`OXu5hq!c5Bv%z-kss41`8
zs#9Atmv3lV)+wFIscEu*QeQEL{9{Yx@ZVb^13nmL5bhb{dtEJW?_2$neGP(ikD)bW
z@txM$;%<gWYDv{HMaVYjyEbi0!w!UCJPaX{GQIyP_D2FGM40y+&NqxanS4TINtM8U
z68ppAz)g(C?Mi_((6RhY=>hNEjA@TphI`$KWjTh6yy=#J#PPVQ`iBEg<aH=ieewLl
zq?p<#Cj4$Kxt|Z&wC&3_?=jh0zT<w`>~~-1!vjLR;zdBu*2DT1vG;ADGt4u_(Dc}Q
z&qLP<M}*gfg~Qb#>5CJX7|d(4<fB4a0|H7E8*ss^2yyMp+hi@if1PaIo4FReD$Qi;
zb!MdcCc}O8&g5jDUm-8uX4V7>-;WPZz5F!cbw5~n)qm4*lx;ce;0BvY^tu_2ZaNtp
zNQUFVSCPdg@3dtL)7Mj#)W?}-O5KPoUwofdHKy=>GBZeF@nTY(vEuPfC;qvb$go$V
zyVkqr54Td3vFm%`U7ZvTCJRM6Ren<;x_k}CjnvF)SKl={nkdTmGi_EtZ;D$>>N8rt
z;(nODgWuokzq^r|f2TRYDo14bCU!cxC0>0$)2DL3W{wFAO6wdFE^X+M=nb8$8?mo0
zre>;9(RCBp2;n=I+oLlyV*dU@?Mi$7m#WjtHKO8N$D^viFx3mn+n?R&k%l>DNRc)A
zb~MIfgY(4V$ivG1EzD$&ff*Um5Wc5Q7uG$^8-&G|3VWg9A*yQ3;A8m>mlFH<uzXck
zX0`hCr}rb80remCi=&?aq{V6<M|YN0#%jFuihzUB1S&9ULv%R`HXzUu-Ie84BAS!6
z;z~E5nipwcMx_a!2@%k4r=K4LSD<l>QWBef8=H5=MGzZQ=vaCD;j;OVN^o?Dh5)cG
zR2#^CkXlVVDPEB`R2iF~BZz-9c!vRQ?QZ^ft-1Qyg}h}87#zfBrx1N)KN&1g5!flC
zWeOJfct;A95Bq)lGDpd+*-AwDsGt+ao%y<Ka*M4(DxXhurY*3O;?axA4@Da^q(;Bz
zE{S3#V|YSg1`#wmHwJjnXF%U*168Ukd^Zd)>Z=8cUbTY&=3c3-8&~{#dZ)=Z0?RFY
zT8bq{KCW!DSMln4+t&wR2MHr=jbh{vYbmK*>htSRC(3Po7RJ@Cgow}t<+X>W;9mSQ
zs;kt~pMK6~G_Smxtrn%A{=yOAAm+Fi;4vxHq`gdIBlSJ|TlCqSY6YNLx+at2C0e3&
zN0XRj6llN=H=(z7GLx}_NcWlLo)R(f><|iZIVzrDywD?1Ee}^~{^GM1|Ej+~8MI#I
zblEsyJBw}kGlorw`<U3eUN?DRo!W9*g-`FTK?&i{T~cM{**@10YW5*q#<M&=A?J&&
z9xXrAq=W)}Le!f@_43RJIX@|oSz1nX86@AS5hgS2OOL`UPN^iJaNQl4Mx}*v<ef-2
ze>CAJ!H<DwULSzuycI;2=}p^k8QNDlzj-HFO2=qVhUt8i^)3U`<JJKJA{*POa}#eD
zF0KO8AMIYFZxe8dN{J;`N@$~?{5JTd*>wfiY?XqY<W35BeP7tEy^2j+KVBw^%<X~L
z-ix751y9+M3a|EQ+pVp}tc#!B^<@YHW~!jt97tkJ;@l`cNrX<a>)~MK?G89Ts!6zi
z^0s0{xhDMDm2^KfG5*!1H@R0H0W@--u{9jQX&*aQ=kNOt=vR}iB*F6Wp8mxQaj^x(
zGuE!tDoI?VR20p^vrP>tw$Jaj?M`m~yubB|>Ml$8$kuep%uLZN5QiSPzX#iZnJNEv
zBb`I*yP4bSZKeC>lOuNgsTTr|fNavFg*7yR!cbg!DjxL2FyyT}1#$RwQ+7rh0q|sj
zT*H4}&^F&!e}<<U%}J;US;O0=a~&<E=>wLsH{w7znkAy_wi?Y5`k?VrPpe4^p0n$g
z(M0^^KcaLa#GAT?O9F${<1Z2y?G*Cbe)|FDH*k@6PW^_-7YLsN5`oK#6Ph`ua?ewf
z<Ad7r-@<0KIr88|$GMc<MKZQL<y3*g-8rajW!HHsQ+b3>(SghDg<t~TVdHYkfDUBg
zz^C>L>b>a_!kmRlfvX!nGmt20Y(SJg4F(i`XQ*%4Bp`28OFx85>usbHw9G8!?ILC`
zvEfK|8Q-Tax<|$Rnz{wH(zi5YCU2VIXji$wnQG?E>)Df0JJ1^SmUfmJ>l;^ZO>xwV
zYUPMDT|pRm?u@FoML4ki;)^zyQ&kmF$3XlE8HN|}Fhe>QSGR#b?R=+(uMo1_=hku@
zaH+uhm>PFw={L+!BV_M4EO@l+G}gc_P>y_#&<xo4x<Wha1sQ>fpwJlj$D5vPl#Az?
zD8sPUp(KUh;rA93V_Q@9bKJaKCWF}A1X{;b(U7ut`JXp;46W3&3mSR|#0+Has(IBd
z4;+2%va9!4rIStQBE-sJwQRUdU`a^B9|GMQYEb-mmFyoTts7$<*v%aW#Fav=xZjZ{
zwjg()FRh=%vdJ$ld@^qp{P7{m%OA`YT!Jy~la#kt7hK;f+2k%WMk6c6t3NEza`5;{
zOYVd$oweuZvUN*<e|_t;Y5r80^Nv8~V<Ur(YhQX#QSCTQB1kzmd^TQCSJZ}IV>LpH
zEDh#REz+ydtE3r;uHOH#s*g+)!QUvA-e>ZF>C3dL|E}mzE1Oh{e-7kGiJTlzWA(W@
zToifO!3iPb?O@uEjtFLUt``_n5F@&AiD`=Pr8Vzr6+@T_0ygRWelWXt7V!?O;m~eW
zS|?rPZ&g?c0tUW=6Yof|yftHEJ)6F$C&R`H!mrJ4xy$90rG~=kGp_7hY8lNFrv5}^
zzR_2bcYA>6HXA<32!#8%a$2Ju=pSQw%ZVDo0{bY*0?f`9kidQ64Zg#$HXiB>1|G@T
zC^D!FG}}Eg8;-ztHf=PSV2~Rv9t|>iz_AaW<u-iQQ71)2#FDcWzsK6OJ+)xF2-=I{
z&r+STfK%N!+0*Q6<n*zwQJ-3*+BBHUbM;>nejH*qR#GD>d_;OY$+r)ib31;o_B-si
zI%teh3UcVpHK{qA753gtMjKu-tT7y=R`yRzQuA!1#H_y)<G%m2curHaa%FZ)E4cxx
zd&Kfd%8M%Od~l%uxhI_Rl6chY`fztz^g*7Np8Ab&VZ_Iu36e6#iLTD?NF&}|6Bx2D
zaOZrI6B7_5yCj4=)m=r5F+>dz-J4>0gLfNXTmftZ*FP)1e^k3nM&H!6!gztO&0(yn
z?30eTOHzFrNNxN>YS<AV@@1R_ciY2lfCSNFGDQq}@`F`YrSV6`?q|0aFtBCqX9vC`
z8!gx2UipnN5D(d(P~nbM4xVdrdM169E|7s@m+$GKd};gHGXn1}$20D#L;Y}(mFi~$
z#;^dw=Ybkw9Q7Y`{a&d-j3*lWBBuc-<Z5NwhU=hMZ;b1gP!Z3JwR7n*85;Ma$h8V}
zgbc)M_pRBfz<C&{g&OPxOM}Xo;+Go+bm3y5{$R&px%Spg69Y=^0-q&sYCardOin|(
zbbB9pR-P%Oz67+~eA#g1*1WW>Nn!r53-YOSI*wIM`1s}mVMa~sb!H>L74g<!^0xAd
zb$>HSF{45%#*9gQ!CUzq<RiTNq7kLapQTDB+n1UaS_-oa<I3y4@J92FvkhF{EZc9U
zyMWmOO<vI%<DLe<S>1<VvT6WBXH&HuiJ87oGH@7MKGlx{QPjUxQr!{2LKdf;Qj6$w
ze{u-!T2A;oPOH-%3pKLfDathYn}r?Q@a@WLyzL9yhHa)P$glBM)L#G&95KFY28QE)
zD5)NkwV@Lk`1yzt<N0FsXv7-_1nOlsXZyaCq*ZKoO;A0gYHr|-M2Gu@u})9#bBrfg
zcA6DY4na);_SjR=jG9*!kljBvMF0O|LyDQEIilMK>W5G8yH9*8Wj~caHU4q$j&EF6
z(zK0-?Cr@imekmpIW1Z&c{+}d0i8D2$~_=h)!%4BQJnw7-CIXR*|vMb0t$kJib#iW
z3(_TBA_5{MBHa!p-8jS`AR#HzEg`6s<VbgS#{ffjcMR~JqtEladw=h{_WkVreSdvx
zxm<$Rz;(va$MO3uAujbMT+i}Xvc6Gj<pp4foK=#kKn?;5A7fFxCC!;`LB_EKMX5|U
zhR`JBmE5{zoZ*J;1YDZ*H6ir?b2X0d%d*w$CE*If=3j|y;i3++<^iCk7hU_5J`LxF
zjL<z6mSm(Z;la(M*#~mcjaXUXI{G01Z3w^^CxvpcQ#StqnbQjs1Q+dSdTY^Q6`Onk
zjwL^Lprimk#&_1W=>EsPRXoz<LcIPSF@#>Or-})DygTyBPgoKQR&<IwI9M|U*3AaW
zG6cS5A7}R%50F6^$d<YMKbbxJIkp-HbP8K%?ha`La}kQ!fjuTAqYVzt78#1?HW^N*
zx^O>;U!hb{k0+Bx;)&gLW^s7vEHx5lbLklt)t{@kh~gc>XyfmmAiolARX<-6T@T$~
zS>|)xaFRTgB*fFxl6cb|Z-U>^aMC2)t20`{YrEV|EPiRSp2M?xhe6pe<VDr$IlsdS
z`P+4o;kp9qO+yBn(5>j`^UmM;fudPxT|d4;!c?>)Muirh?zvyYa|6kl(R(mETIhMb
zqS~AGK0Mb&+Kk$SS?4`hWm!s^ftw4A220@c_v!o98M5Iv%p?%<ydAN7ijQHlNb&QH
z!B>JPsox3JUR!Bm|A2i^rJ3@|cj)55eyT|*vRFv)iG`LE>aY}Vrp|4)Q<>bunmeF-
zntXr7Q}7MCQAxTrIFG17kmr{&PIF~G{|>88izYKuPRJ|S(#Az}dh|J2crK5gZLt1_
zbFMrg<pNWZoczB+pljt^N_zIwB&;U{25yD^*s9gpYkH$~M2$ZJ3~G3_OhWIp>o%Gy
zV-utZ1(e9v{1Xcx-nPDcDy9mw%cbz1L4qeonM5^?DY-)@s=R4)YKD2wIx6%>>TH{S
zurda76x!w6nu6}_Tiaea*f86%#|b0(ODA3dn^Uhumr7TwwqzuFJ??{!q|=Uo+OEQ6
zY*>^R{+X6@K1l#im{zTpjQi>_XZgZi4V*0a?qWIc6IQnuFS?(I77_YG$R>*EDV&Rt
z6G7`k2q->^%)@q=d!-}hZLWXcM)+eN_eT<CXUQ_Xh3S~~x$wKbUZm=xqe@s3HGcTl
zuMY5P$qe4Svm20OI;B&iu1=4&rcW$1cGuG^BC(h_#pndHbK;&kdx@L55sQdLF2iTL
zbqj=yc!nOgi1!$ar()4P{v`G>ccO_@T|WkdE5EZ+L+999e{47GWO%gR%+Ku<A?rQv
zV37lPvc?#<uy$)LXt63j>P_}{jh9Q`&({k(9mQnCB9`4lQ%s)tu{Ux*%?^7b%^Pw<
zbMsz?gnCEs#0z2dYU)G~uYd-@oQjI_JK5@6e37A_FtnrPwNLd=@QKBD&Ou)-5vN(o
zAZ(A#7h5Y|OY|KMLqsmd!WcT6wWhHPxlJgR>$)IRrfuf@!zJc|3`&`D=tLqeP8FZz
z!_81x|AFIA1hNzLuLwjdW|}Sqm%E$lY){7I@dZpjK0PFfTuNL%ptcgl*iAuxQmFmN
z*;z0?*QzGnGt0o~WHtc)A%Y(Ua*{nB_varl>T)YdQ-`PG-&AIa{Ug!u?xQ8T-JLxv
zg<%_hu|Jj86Y?;}jnO20h={Bwz4DD&{%m6alVw45*&wAZ>~zs49g<=DQs#nx;xHwL
zDRtPrX1GRZ+dYLUQRqN?=;Nv$=|AoM2-}wQ)^eVuXlv0~aqiEhcH%19j2f0m8t%R8
ziw>p3>YWWL&IDpiZrw_JJC?E{nm6cL!;2ieifd#uKn0~!mhgJCY%n^CxRVib)^_jz
z5k>TE5}@LSd%kG@GQlji(pvN(j4|+`oehwH1tiy0`O|K4?MT&A3Y+?$?D)6ZJruq7
zmbZ&bs@pNw<I~qigcn4;`%Sg+-n5tW&jXB>L~|H=Tpq734<PP0^FP6x*ITN}COdIf
z8>g=wjDN8lU|lwz#S%|8sU$JjL-O0iZ}a~2-8GJxKLd$V&12Yz$mPVso0)(wf`>z@
zqnQsI99D|(Q9(ohJ79?P^`^zv%sp{A&lqLQ*0b~A+53hSX}!t+gEhCril<fAHxdKu
zoGIJjod?g>`js(-Z}Ve|#Mjs`d5+3wMFYL+T_M2zFO6eEx$5xkxUU_af7Vw$co8Oa
zxDzv1IH??~6a#;|QA<qY9^_4n6_tB)BnYxdjz7CLyq!i1pE8BFl<nQoDbLm`?&0rU
zR^!O3?Oze_B@o#xiHziq)wW<}QZ}%wzhQGj!t`0O4}px}hgx9nb30lMS&GGUE`qBG
zl4$$PZdXR^BS_!~5JpIZEGvOezp(qn%XN<&jD<x~vk3SQ-<!xJ!Rt`h^1mS1R*T1w
z-<BO`@Z<_PTd|}QXz+1*<1ANZnN=&9(#?~;huq@W-Py$r&tnOJENp5X>XBVmbsX<Z
z?TBt0if+fKeU4EHbPn=1*m%tT;|u;^Q5!VGExpZky;@2lR*jjnYNufV)S&@5*F6Sk
z$l&*!OEAPUsxTI;X4`5CoOO=MjUu^>Deg6Uc>`NUN<NCf=jYMK0y^^jVd9_(QqkbB
z@^wjL9ojQ53mG8fCHKwSuFBweCL^{OrHcAmMg;*Ga&`?}`{t}!sr**M_b&8Kg=<}=
z&_rAQ$n(xJBcrOIq*(T46RlUZuVmgTCklm_LCf?(w^FJ`)*}Y9mQTF{v)dfCa>cp!
zH)tiwm>S=6P>}+I8}}is5X0GoHH=!$99c|t_)xeH<db^xcv>`jw9JzE+Cxx#6erOp
z@y(fm3K5ISDVBwYRL~k~aRl}6*uU-ThEQ<Wq+KIHepk^oj71wz)^^D;9T?&NaSSaZ
z#z)OebfE__Y^=_~kj*!@A<Eo)hYT@ES}k94S)N9~7V<17;^8{Oqqh$=57pui0wb9E
zzT1N4PDNE>9RhXF?bl#jsl-<?Z3y}$cM_Sq-4B|Id-hbADik?Mzn?>z7}UuR@32h>
zaI&iIT=1MJ(88z@vR^v!QJQ`qlr%;R%wH~NP;|cQWSuOEd76C(!iOhpBCeYhPAf5u
z!!WqRFv;?n{LYE5!~FoB_Hz=N#xUdN{g6oLQhfH(#lZ{m+{?RXo4tdGry*qA^!w%V
zoYf^VDhe;3<21<KSfCE|IY!d1H`kJ2NvO~6U7pN?VSpJ44IvvaPRiH&Rj8?hAv#<}
zc+~<7W|{<>w-|y3_h6HRc=FZ#V2Lmaz8|08L`mS!qr|+dvb||(Idxi4EVANNR4RR@
z)@HtYal>?4vz*n`8aJN#MuH@O3=OVtWT_9Wk4MF&S!vp*0JDZ`o-1!gzz|3|7d|l7
ztJpA3ubpRmYpu#A6wvDPP1bfkSN+dyB`cWiA-n`Xb&jV~PIVPB%3LT;3Kh8#Bmw?b
zmnc#l>$ol<r|0>bkc-M$ZeWdVfi>Rt*e~L3nRLrGVBCfw-UaKcpC>~5>WtyYjZf_v
z)Yveq?(?2ePBU0<E1##;F&+$<SS8WjXK)Kv#00<Nqs5ndX4cZ9xy6d^V^&`Z#*Znd
zYUhGjbw#<JeiJatR0)vSbd6AAL4+AD=xv{`+Lp`<?C~UVw|z4!k<&KBmlYYY0lS(b
zDcSvEA*B-x?5r(h%jBf=2c&He263A*gqdKoQJx~!#6rlm(!xE^PXCB!-ZabGOTVqV
zM6{_pV?P=MWv0W_kbbkhIUYnvg5Tb>jn>1r<bIk!r#34M(mQ*_g-Q27hKZ$Xl3@Y3
zI%z%3RmzDDU|_9$^K9%D+Z1PriY?q1nAjKN6v9sC@;Q_NAp^_Z#u=gxaueI8wX)$f
zn+q5Rp>}!8V>aXKtZM}H0>QEjzhi8WaZ*F|wbuEGUoj?H&6R2QPPkRjz^g_fi`Muk
zbAf6FHBPD(6E3P3olqdmpw1CcZbi9t!R*GS)6-eh6iA_knhWN%A-l0QZ2<bbw#MB|
z9&bwiZ)1zR%9?zstdt*iU~N3(imvxLnG6)Wn(UKwaW$E}tkpN@^9pwRfU5z9_TaUz
zoe%tFPTI+cbMg!qTGf{}b2)P`@1WW&XEb6=c2+yyf*FB5hx*7dw=kc2yNfWMB9-xj
zKoy`n)PnbJAu|*T+vKZxg6>d3NZ5`Vo-!Qjf=#?w)2MyXcMc}^_;RDVGFwXaL30K=
zh4)$DTXFoC30w|ay}_QUsUJ1iHic_{yy#B`<LNajHH<a-ur(mwMTZ>rA&VUy7u{zx
zp=ZOJY_!mmsJwrO_AUnyZ3NN$xyeDiI7+xI{7EcuE~vx_-i;Wzsblm{IrfTKrhIKs
z7QsIlfDH(~TMqrCxknN_RWh%Ab$P)G>-!lHM*jNkm6o?3*FFc{`-ic*MQ<`2zPvhv
z!Uk`b&K0RYn5Tp--YF|GG&s(+&4jNRX1$3qx|egp-53`}|CA$onkm)e^Cawi)5b2O
zr@$3DmrCNo646)ajGH{eg>uqYQFoJ#dtR^;Pi~gP4Lx0qy<pfPV-0@|rI?k~pgdV*
z*Lj9+0B6yGJ<Q;eu<OBtE$DP$^EMXa*Jx_tKH%QyY7VUIWdv;uEoG}}&puMhHxJ?-
z?q<|L*S=?e2Cy;p!y2HeDu?K4XDd(wj7*sECNr%Ylxy?`mntYTFc~Z_MpBV`mSz5@
zaSx;f!I*+;AMKH17-&!+2+i-QI55^oO|@I<kM9*bS|jq`HPd%+Sh^_bSI<C76QFFM
zD*>CO`M(x@y11(LW+U5`Ds5Fq>$X=S37vD07odJKY2km^5ZFKOXe$vVy0@nnkNnu)
z5Rj&ZC6RJ<dL|+7-^H!7_f3)b3U+qVBT(>{nXX4VWmm<{8ZU2Z@Y|Vy*;y&hsR&)W
zVS~u^?Q%Bg?vyNVZ9E}Utw?)acr@0hc0!SjpZ*=U*4))qbxh+;cyh=laIgc5Ck2om
z^Nh({eQ`0k#VMQ)q3mseveg*M6FMPx%saKYDOd;OE)dAeCG0`Yik0*(Tvj8Hj#~5M
z^WfMP+PuHoR2I~B@{@?#`o8aFmxZ6S+l+t*zh{bUA!$t~9i%|N`ZJjDZ7Iz`57Fz0
zF<DdHt&wsoOlG%l=Xpesx|k&QBu!y)+uaY5Igw5ZA;%}7w9i5|(&;i4%#gCui0d^#
za!y(PPbI_Av94ZbfnQ6U68kT=;ni_~LU0ni*kh4^sKpTf`eTP&(HH0SEaGXw|4=bO
zNl&H}y^zN^jla$u=RzKkbR4m1n9|XA){2ha6ll5KM5{*{X+KB^QNrhYu6(&hP81!n
z`=(O{l)bU$$i9EF2^T;6mRRM`fXkG*`|N_6)fO@%tF7nI{xIHw$Xn*9@mrvr(2BVf
zX-*GYB$LOxjxwX^)PF?i2YzJowm<8!)gs3ScO{WsFZx0|%+He1aXM589j8mwGV>}F
zHIMm-OI0XG9x7$)F?6FngKiBjO|Of?sRjBF&UK-)4?_$Gy{j1}<t4kft*_@Gitt!i
zKf9Jj80cfwlCcfy2DROLE3tPOmbcsniq0N?nygY};~7}3^UIeK2uS6=*c>8`N6ytN
zLQKVCT|ZmrcskZBhVFkO9$Rb-zY{R~M0vh7uE>f0ccI{w!7DT35hXk5dV<Rms1(qv
zd#W*ty;=Auaj~kb6jn1hp*pmFizk1*6%X-tMngv526*=+9#J20kZgkmHfE=7jTu%`
zv<x(b9pA!T`)i<D@jlzJ|4imt3+!Y(p);oPLha^bWZzM~D&$cW)WQEs)>)jDO37vT
zq#<Q}n%i(0r>DV8%c$oZ=$>D_FYNL4PqBS}8f6L?x7|Kw$8P?1U8>bWD-C&>L_!zm
zDa~Y^>Um^eFhh&@c4F4R(7BpT@Va69#1YTygl3tHjt8+bZkFndrynvk(7TgJa-d1x
zLS(!r22e(`R+qW=q{w#xDAmKqPeBKnu1~F%rS$T2rTTpM?1LQt%-Z4>3d^CJDOTZB
z6f~dnXM>DylG+`@3JG-F*1O}NO|5(*g*hM^h+>~33)a(!&dVj4YNN#OE9$tDjESA)
z3BwIvxJtCGDK8?C%Kdd-^{OmnR~$QE%Opza(II08-C92AEiHE#G=E4#o4=#^S8CJJ
zvv*ZoG+9K;WbCitMt$$VQ2h|uHPy>rk?kt37h9sHdgB&i8pMGdzxvwh!maI*mup{t
zNQ9KTAlzGjSkCjPi|udcSGFg#&&hMlr7UZk|FxL9DyFu*P}g{*#8rZ~?O1aXPAcc$
zBPP4I5O@9=UgK7{Cca&}*B|N9IgLd@;ah!h2^rDMb}j0yuVl)_Cxu~2=ohnIpSx-t
zX=ANg?&VI{rwS8ZJ<jKNx0u1iGK#8pLZQy7xXlyays)HlIPh-&_3>)zMV#7v3i<h$
z;H%fF%8U4@9ZrwfeI~0D&3ap-N^P|IN~t<M-B;_q-7?xDv027F{Vw?M^TSxJs<oQq
z0=8QN6Ro4P#e3I}i{2T|j1_nxjeWaE?7LbObHmo-lvCtv+!rj&u0OAVo;EW5nDb6O
z(c?FO(om(A=-~7US-HgKw_#Uql<Jv|HM*jppxbvrHtN<TQKJLCKWbaXf##U(`V;0>
zI=Y-eH)zz459e0PFZ9L%ZV*$3i5fT1oxX@QfyXU#diV`s?nxi+eA9Uy3>Qy-o{T>X
zl^Oj6JwMu9S0}xR^Zq6l0MoE3DC%J0s3vRjQ=I7MfSNk-<8q3k_`L8>Kfvf5-z>Dz
zQSm$-VR?R1k9Si+|83bD*%j9mNBm`6EML0Z8ATD9iV$iMexpT-3HuNq3i6ctKCpO|
zjr#GcqgFRjkGNF^BfDP{7CA!Y3Al1(H3E(YdSvg-7WEra^tECSe7GZ-$&Be!(ijMp
z#ZNQkWzR&6YIb_<3A9qIKYUcSlJyG1aG7x2lWoI`_b+OWc^?IKqv1&kfu8F}9D8~W
zapVaY?^NTrZRD5nl+{0&l|4;S<0#roRMyz@_CnGs3$JE&31>%0)J$j8Ccwf_hg+4~
z6?=Eqm_5h>v7C7isjzrfk3GpEwpWRXl~M(m-4%9b^%!S%5tW@Bm0fz-+q6r^NC3l-
zubj%!dRd*`O7>Q}Y>Sh^<w`%qtpR{b2kzcJBO3^ROhmmZm)M3^c!-bLF@h-6jAGAd
z^7^J%AO5kbgnJc3eh+_-AgBH3qKDg_Z?7uwVJimx>Ur3R>>h5<;Vj-v*4QYj7kKG5
zcv<wB4N=Se?Md~MWN(|`WSl;sJ^a~kW<X=L1V5mJrr5F3r`FRu&9cBY`Xsf@+O1|q
zOnN()tk=xvdap}4#VGiEp*a#(rxnMy>C-b5^q)-3M?i++J#rLMc8u5u{h<;aRGl&V
zWNbuY2QLMLRzi4t>>$^S#3UzFOu1*igiDbxx(X{o1FG&^E=Dl#6y#Rvs*88NkY!AT
zdj55IuhL(Eki*LBd2vRV@DCR22B(Uxign!zf7X-cHVuU;(<=X6);(y2XIe_#uXvJ{
z@2Ue1^hR_^xi$&YmxGk2{k3UCQd8~4n+^|@57yguA4v5KnluDF#Fj_lHNJedbv(=L
zcHsO__ybpd6VYYfick<}ZZ4=CiZ?MC{v_A$Xgkz0hWD0i<l;n5f=~!;D1I>eP97hM
zn?&&_2t7ZaO|7phNFfYEMR6!Y&z8h#R%pQhy`%u4jJV`TW<HrUQ%nl%Y*nepSRqL?
z30-1-V0M?0@o{62-o!u=37Z_#oJ{Q(y?iS@Vg}pTZsLDBs$X=Bt0MQL6WrYiQ;!yK
zTAU-`{SUW5dwBd}pM=5EO8ou)(<Td?&JW01oyRmaw3=lG)EOQpzwU{ho#mTP^}2Sp
zG-gYY$2Vl-f|~hn<8oH?O}yCrY_-}0jnpN}AZg}bPiOl~ww3h);C1bMsLorp;{NG$
zIIK|_ByvGjOHTFlU0Q}y{SDIB<UXg%t5Nf`$=Q83ac?lkOg}P5u?>Jpi&4J-cl-vi
z-6mqS>slW%v52=_-(?bq+$Hi3^6|%HwM^8W)&g)LQ*?jrS}yElq+rlu-bN;Y_V_70
zY9md->@>(rBI!uiCNqGR;(9)mol4|I>~&rHmtTJLNk*ojsSxY~%t1+;>nBBz;JgP2
zGhB4^PnO3keDj)~$FjpGi{^+hIay2nl37j3TDlmFJRnJ~e%=NVMa@qn6<l)lmj?@y
zO@bXl=$31v#a}VLjDaL2)U?{}(G?3pTEt#=MX^K3#@|pSmVP!Mqs2xFOqgdR5WdV1
z)dA)u-`Dd5q_4gWKBvV-aJM}BNZb<dZnd6Fvr*q02l5t&REMZcyHF7Bw{H|B_Q+q_
z&zDEn)?AoAADJ{P?Xca_b07VxCSyq^Vbs@B`Oo~;)V^{^4ZK71RtDd6wuf!aW=nhb
zRZWp;@N4|xtaz6M<7hrug09B}@^iqIquJiC!+_|OEw?=bTvpVbtmYV$%3dlFO~D?L
z(#T%dRJxL2;7j~x)m{<)@~d&3&#eZPt*6edkkN+$)Q&HUP%k<KB^q2&r=Go&x*xa0
zDzAR#q5L$GRmzu6Z}VR%odxdBp>Fji;-^4H7E<9CQ~p;e3&>t%%uKH^<wTXxdKZVs
z&-aM^d7Av)b<Snl*e>CBF~;7s-LXl~UAcBp?oSbw5zRHCXT44%zGmp^HB#^U`NHdX
zJ>l3hE~uz3aOceZQiXUsR}_O-y?GyrAMQ^5%aUi14J&d{mMu~jwtu=M^iFhVGOwLG
zn<fQ6#dE|s<uriNRm@(h`y4dolF<y%<&vHeti-;xqNthgg@A$TDnEzWtVA0JuM#mQ
z*o7^l0v*waW1FB=Mv#@SXz*n4(+CoVcd^1W^Wyk<D{Us_AWDN5`Q6-}WnxVmu>VT^
zYzflP;o@?`B0m81{dqwPR1EIQU{7q96tkVy5M7~%+Ks($+_&o>9|+l$n2DxoYx#!Y
zML@#EbkFEVPc1J4YY(X|Zei<1=yCfPO5-m)f5*eseC-^((;Bv3jGoK5+jLY74B!y`
z5*+L`qPq@_*)H<r#l96Tx);cs(wQKwNHM5#w0kIfQq&Q6wHcoGNQm1`ii4i2jf9K_
zWG`NEzy3;K`jt%m)tt_wAfY#!^<CQUimP7>A`*JUb>!T}W4C!v0*fX^qVz<IFOCqE
zM}M&@pO|%0t@_@<8EA@S-=lyA!*F~pKhTb2W*MD=MuGPRxgKhaw!~pYhtznGz4YEI
z{4cM{P_|x~G%r|vPQ?iudw%vTCA&E*F?jVc(FF`LeN|8y-8lv+07iU@?UF&I9?e@1
z4ObrFkR2AjqI9MwjvPSlZsPei;?Lcc>7WYp*yy|u8jHq7Y-d^pR3@MS-l+zPcv7)(
z`M+CfJ6p09#8i$Dv*%<#j6|D^pHHw(RB}UFe(~e$k+7or?{kew((-S13?ZfqirC&n
zsWkW~5kZ57RpZ8F%J(*jF2p(rj!zl90>b!%cpBA>SMCWZ!=uTN(1%xHB3p5WpY9wj
zN?mm%soszGP!RdxEg^!-1&x!trhsh-^4RoiJ|(&fwL0o8tgMu!86j^T`=xxAz^ATk
zAn8U#u`-5z3jYLsUU`Q>MlhamiWEB78BgZ&K-Dmcc%<G4)iDx(qQ<X0Biqj(;tDC}
zSxn-JGLiF?)~1)|Qi*blaY?F(b%~dQlPUG}sLE(Pq^frN7Qn)w+7!jx*<rT&`0LhR
zJ(uCa@Ih(_o<dwk?w;(}ciW5Yg9!d;vrjosHTVr?NrNR`Zb#PMB(}bNc%7B2cmi);
zV82(QmUj!kwp*yexS?X;1*-`KX}_|u!(HJ=XKyywq|$1i=pRm!QA#Q8)?VBp7qtE~
zKmR($ge|@yAiIs*ACx9ti>FsX2LrwqU$A0|W!qSkC!_d4Z2DHV2fd@}Rpe5XM2iu3
zrlLs7*U4RcH>M8EZGS@Y_v=xEY}C3THRZ7o=IjINPZE7n^g8JfNqwWq5C$HnnBZAi
zEup~xOEg-Qy&%8=N~Q_tiliayr)O<Q6n`nA8&Z*FH60eg^k(EwhXWmLU$2x`P*;aV
zLzI|P)k<8sETyv0(V469uQ5<l)-9tBAxTX~4_I>Oh`+IWq;pK{U$h&jA_F<>4(_;T
zav0A&@Cx>rsV`VI_vhe!GlKJwDf(QK+RgV-g}#s$!luU_t@)-VxqDrj53*(*)!*@|
zKj(GfsvA>^E`0G2qCxz)>;Yey3YYqOyKRIueduNvrDdk+MxA+dY8e+L>lgBzn~%#F
zARAA<@3*oThmvfCSbqrN>#>JvVKklUElREhBp({(_Qh3bUktj4hpH`6iVTj8g~e31
zAD<Nas<sf~QCc3nnK2P@Zy}s1m$3Y6ey(&>-<`Z+F1gfSpltAMu1~`l#%;*0>_R+j
zh@Ucqz$%1#g~x?IuaH^>LU8B8a%=1b%KD*EE@tLh^+7nm+Cq#;Hg=eZjG1lV`Lwva
z3*Ahr<hp}B(1O?5LZL=62O-I5u_3Y8cp|;1m$Sg*k|n$7IwF@M`t+>DX(2<;b*gmk
zE44X)nXm3C^?c^@@gm<V4x-jKX3?cBJ-02~$9+gm;vJ;ozu<p9W`;-McJQAf2&>3N
z>XteLp-JvhM3x5@xtJqkJqn0D#dsbWRPKz$Blw8sNVG}^v&n}&xoJ47+pj%Hv~|AK
zjuLDBbMcRQIOluLXy7MdMtpek*ZDP%O6tJEWHt?=mz*k>rwEXv59RA|I@CEykqB4Q
zD0|hb(yNSjY_j+9bkZa0bk$4WAx~%y?$z>AV!x(V5ELHb5^1VB@DrTxd$oi4meg(%
zQa#nv?sj3yi%BZyfwX8#CK}HQq3r@v?JKWeLvSwx$Em3+E?z>amrzqM!&toN;B$JC
zv1@wJ2?^o*z>9@_BtTr7WIhc&*VD-6{I%k<^i$#@Rm=FJAM5s4(3E>3x1X5oWjXDB
zjFq_Uo>;z=GU##EZf06_2zT9P?pE+M<rcZLU@cxGtbJZ<;c~f{vJ~MZq3<Z>X*8T#
zV5KCuzH+eGu}xV}J{&T6e3N%K=c8#@nq!=+^-1mNk`Aw73v6i=XKSg!vEv)><UWYE
zs>k==UybbR{zaXC*?PDXrDB%g+FES}+J7k^p{HH4v>wLdXLeQHgM;yD`&}k36A;Sc
z%d9JFB36u(D<)|zkJIK@Q}o12YS`7|O_QbV)V%T{q0269njtaLB>Rs->qGn9Od|yt
zSZ4_^{%#~}-dakFoPYbJ<IbJ)coGf${1&9<Da*mT+>5OW2L4i`)2`u*TbwnY%BPna
z)pe2jzFm*Alv%^e4nnBkxoj#{^T<~ERz06m?;Wp~goJ(x;X^ryMy!A0%!>%HY|S-_
z&0&q$ETVFFuvx%IDB!|XUrOvE6KA=<@xvv*=NWcpUWu=!(veh3&|;tnPfYUdxe1+y
z@C{Rf+q#tsly3+sLX%Goz9;6^r@OZX)aSF9ZCv|trFCqM)0S+dzl|Yc_h#cqUkgk0
zB!l`cB-cKpke5R*?h~=0&m;s#eGRI5COrG3r_@)*zhp+orEZjNU&mL8$qX^>t^^~J
zFvTIDW)flLtI|F800USK7rqrlW0VKh8+B#D5=Gh6<_-!J)s13%!F-~24nnh%TXbF#
z`6IU*N*i~+ZkvQ+=+XKd8xlaKj|t10S;|?vpve3Y%fJHHRnggkSRru$z{la?lJ2vf
zqvfjs!GR>lj$9iYBR_0mIn<*nCyNOORe?DYUISomK458N)f8jLwNO;6twqp!<X)tT
zd+Sa0XwH90NIc-pd-dx(ZB?KINYh$PsI+P|K7o+<De0H87L-?vQuLMA2s?+Wg(L#7
z+JR5oRmCs#A#hDcXMm>Qy0LXbDYl+Z<WA3Uyiui3SUrVjVjg^YVyyo4)sMm#wxk*y
zx9^MC_fJWU8fhpvFCQGEgq%~oTBbsF4B{V$svXCEl!<ps?{Xd79*{^kkiRg;DGA}b
z=2!s+?@JgKga#5(mP}Du2ClWiBZ%qZKUVtbG#qh9<kYVsZts~pgy}~)f1vn-3Sp<G
z3x|FEUP|kPQsWrq(MLF0(+`+l%9v+BL`2%wCtiurEuI~h@Q%D|X}=z2)*<rNmH?%j
zkX=hFg&yYw7FruiUPFSO^Ye0FZy5sYT4pRYmFub$S;X6ZBf9gKO8wn*lxgQ_A@tjb
z&2}CHEp020Gb602jaV9V{sfS<@+?kz)w0Qp7!7{SP8OY6P~ZE@<9k4R><Xn29Q*xN
zWC-mSsROq!JUr4?LdCfrOYshqwzU`d!k$jwhTsP{eT~jV#l!)+dM~I`TO*EK4s(8J
zbS$wtF>x_morvG_w5`O;jJ(>&<P_kDp|N~$+pa89SofXz`lg|s+Sm%`xc6Ct<v}w<
zxS^thCFk|wauG^aBbcy`;>u(yWx>JA^oxUTNZNASZn=NBOMr0X&6sF4%BL2yS$yfv
z7jg#-E2}ddWOvW-*4)=jJ6IvP2Z<}$euU(8N;5Y($wln@Lb*?aLo~Sg{gJ<3w$^Qv
zc{+-ub+!K$!l^wkPRcmvUnNET%-xh=IF#>|>+z}0MDq@oL}y#PxPl92IAv{yo=jP~
z?dG|3Np(}dk|v)E3FPWU2_*MYKE8^YLAfcnpd(8*MARM_c(Rt0Dt#(w;@(FFTLxe5
z8hL|Tqd_)9VsZ1eWP1);@$|BG>*CsvX}HgaO*&+!0vCU&lx-cSvJi)pnLJiUCjGqE
z504r)a|;tLi+!R420g8~$D&)u$#Ik-Mpvjsq(+%9H}LL+((t;h6Z_6xpnC{?zdj~(
zx=#LDn%Q^9(9BeLkr_|?dOBOc&ksC=i}ylrC41e=f3&>8!nQ?%zqC>ITB|@qPwG<%
zZ{?4ClZz)inU1oO>2FMbVr0lvRs?(B=};7Y|L9AgFX_u-)EvNA&|*?Lk)q3#EGceC
z-n%F1n(RDckM&7wTOe4X;f@S6l**R|Up<qWfEJ5vf#f=-@XDiK<Yaob)rUzu0_)WH
zX%yw6(58I;!M2>q<BEp5g6HEt%x}qqsXOekOSti^&Bp05+G?XXh5y2z{On5JmZ4xM
zxJJ_aSlvY?vapyiX|5%I0<ytIB+w_XA57x97UNXItlARDfLX(t;WV&fHSj$xLkj2f
zuLWo6J-H3xF~<&`oyL>*XMx^IA|k`XSY%YHu+xlu1=pX0_@EEFd#(JywQ|js0ZU?U
z()U>0Dw+|=hp&_>J{WJJuKQyR%vigPR-K?a)kc?e>}Ysfb){HFYtHsq_xmowwWXVh
zV$Y5?Au-yp_6|=aVER1GG(V`&U^^64?^Qt8L*Ypc#%)PbUE%i)`f|_qqgh($)Ht(Y
zRkoq$`Q;nfyi3EQ*5|1B*^9KsrJvX)j`3jd#%AM%_jdu<iO_`j)tRNO>GUnm9(dj4
zh@I`x&a3WIM`dwan_p+@ydy0}J5;{i=_7GgZz-4S9lJ~ms;%{*t9-{j=_8u``GT7J
z;-rDC)vlt3snpifRQp+c2Z<wHLRLzY0S~))SW#mNV%<aW+kT|MMehzJjuH#54Eye$
z#X8s2HS8xm;14^TT+cv0CYm}Ags%F<cDrwT7@ERY3;WV<#g&d2<k@I9b!e<|O?e_b
zm$EEvc?S_yRn14WTnoeMx*!GGLFG)6T*VfkqAyrazfPV%ON52*!XHyuOsq$qo)0g!
zXO*EnoL&^h;!?=1)5JZE^ky=Cie0Eu#jt)?W=(aW$wiPcbc}>ENi_6s<WZcY)r72Y
zcu6y?&+lh{3e?O@r^4Izs&YEw$hVy~l!^@(hv%<N(IQ-Imn!+Vglc5lHvRET*yjNe
z7(=n)fjJy!KbBQ?futbwhi<OpQ-d%11`j=MlvBj}{5t0stU|6(K~pXXt$95k)+|FO
zw1)Vwk2CpN5<GfE=6dPwzn!Jtdv0zR(L;P$UL;$oU8<`4GC%Igv$56Uej-U|IL2R|
zth$=w74o8@wJr-eRZv90S94F<DVgcH%1|n-$A=5js4qQ7&MLfKMc4)xA`5rctWk}2
zQF6~3bM0}Q1)OPHOBlAAyoC`W4V16|vw=-K8K1T`e78QP9YQI!H9p;U;bzrF4;Eu8
zZU?YY57r@|7FVnZS={R|+8FxQ^uuQnw!`iL2as3S4j1`Gb?4U}Vvn`cW<HEm+FV;=
zZ|{Y!Z*nP^Xf{2{Fw!*i*?4lHn8bCE5yEYB?Ud1S4y)^(xRVid#!Yj3Ke6niFhi%r
zR;$ASJ9#EPwajCXo_WXbA0kW6PLDP$uL4wLCpLML+3xL`FGKsQx|`$$XX5UKg@lIq
z62`alwaKoqGH0xqwr6+i_6rZ5Q!*&ud)tUQhR_5@?eCnMXf=2j)z{HlKWu3l;zp$*
z%-#g+Nw#@@-jjFVJXA3;4sz*;mtN%UjjngX*POw?sT_D4jlZ3`UXroBy?{$<jZ54(
zm~U0nATdF)AhdQf<k#EIRbRtL3ht?kWCl-Z-%DkO_J})?;lDcmkgw<Y$Z0Vol@GLh
znoBg+*qoiFvq3rmmd@g-o=k7VvPhn&h|p*}h)~;pq)ABktdeFBr%jtt#s7Rcs(O{d
zX(T#yoH1}YxGdFsPdMa5<E|p(@=IEk-ub80zFq}ZNYTeVQay)Zd;S?#{EpnH9pp@<
z^n${bxa?k;Uc_{}Zb)w=cH0G+^x*yB904hbprqD(%b=yrB4|Ua7^rpegIX7E2jwZA
z&fSG~PUH(WjP3UX{R0<BsLF0@ossr_UE%hx(k-E8qgcv44r<DvzF13NWMEZ-7DRtP
zA4CEp!DI5+1u0sfAgRxh(W~G%w0(5jJjdI>-G|LU`b#nHSYBrrm`>Ki{fqkMbt0iV
zEA7`oci*8gNB7(`V1yX21Ym_?$xp1XYEybtcYBrJtC5rC8kq#xIWMEcR|)Q{uwix}
zlA~$xLIL>PGO>AcUBN!2-|Z&1LzCtTfP-~y=}`D93Xn>eb7HicZ7#AR(ToRQa&U)u
zdpe23$Y2YnDJdlX<cu(v04yG*sf9ItjW@Z7AL(kRnjC9Q0eUDm9&a$=uRdZV<EQHh
zs3LNDWn7<qAqP+MF=U=^(${qvF4Ut_rPklubH&u%BZ$4*li%+hEu{G1)!q!O%cSoc
z1Uba&SUEx~cL9fmHaR~%B#AeT<dZ$E@aLw|ufRLu+zm8JB5e$(3Dj4)|L2Xo!x#ZA
zN0<`Ljo%Da6*%5lM528FO7;f}kCkm>j3*2c`6Kac_9mbcBA2`#geKvzcX2Cr#<WUq
z`FsJ>xgTa#=?-wU5_N3Kn>VWVdjLRc*xM;7F=kBar%#}jQg@%$1Jv|5E8L&fM;~yB
zbx+2_;^l0u<^|3B=-|ydm1T767KUx&U}`1EXjpvvlP3pghh@4|*es}1@D<K})VuZw
zlfF+F>}_80KqMO2pR4??VIs&onIZ<nj>a0<8F>Ku=&zM7fSGM$xov6>atp<_D>eVb
z(XY#V0G24%sAqsa1#cQ@!}UBkKPkYKR!dUZDfU5vc6U8zJ&8;CnVbdQ{jcO#$Xmkb
z^NfFzs?Qt$z59S&q5wzw%S~@UT+p5iYP|l@ymn|N3|`$-n~K#g7lgGLmtCr9!rP#C
zd9Ke4Ki_>hosxNSXb*my%@ae=io+)`Y3Cb^yNBU*;C+6$ZkAuM#!`%i*7en-K%>&{
zgHeVd#p(S!do8dg*PWSq8|is8Jq~h21soe4So@9oRX#L0eO$;*U60*nrPzaR0pPa*
z#O0_gjRO6x?Qa=jJP4ag>%p<6os3o4gMYCv)&2E*`96BIzn7mPIRjDyyz?!gyq0;T
zF=#kz3;JsC$#Oz8o_uMvGXXWEh8I~;8=_MIShxB<)J2)H@P55uV#3_~YiG(88IK0w
z^dIk^=FV^{Y`)J6^DnK<&$y|ebUg~rls)uH6Wo8{h&lkwWWEIk4HCW>9h3nmsraoT
zAKE*{e&meXCgtLuXK^6!{)+Y3tiGS`PMaLKrAmfwx@fR{Mk&lx*xC1T)Lv(fy6jQE
zrsv_X&H!KYPNSEq?7sl|djQQ-ZO;oLzdr!|BpB+9f&h?^H}u@&86fYl_f0ckE$&5$
z2nYRZC;pv(JW-|tKR4x5M^Fa({wUC60w8dW?vW}jNiCCTwMiCitz##}{~mFN8w;Ab
zXDCXe3auIb`+tNJ(9Au;(0hXaT>U?_$v@}_{_8KjQ(mB%d-xVfZ2xDU|IaGozrBDj
z*=Vu`WkqPS&i`cf&%R`!nR_xKWxfB0%st=H%sr~A`n&(odH?lx{U86kVvc6+Sy@n-
z`EM`%_gVk{?)ZP;dYr9ho1#v3m%t1GKhpC!IUeZ2NdwTX);*-EY(4FD<>{IX2T_6L
z*}*DA{A`8cSakHA=}et>k~$R(2C8(?$jj;uzaiXS7=TPSooo6C#!_pzRh5Lk#sfXt
z^&SIjQb09a;{pXh&jG5Zmwq{aBR((LrCnO>37YmM0d{ewmXeHSAL^}~u69oZA8S8Y
z8}2l4*!{~&?(;88G+l-}_SM3F<IoHp7=ZA%vp3(<|1BxrG;-H`OMmkr03Ebcv7XNh
zFtx*Y19c3&Z*)Dq#cn*J?`~ZTvg+{wMpl?XA_O@ChysdtTPPD%GGyXO_21Q#U$TSN
z>LqyXg*8bdAn)k~)<$e=0$_n#@=rOb76YIs4Y7HzDo}x00!0X}-y8$LKy`s$_?3_y
zK$F8?Jngm~56D?s@u`J}$S+P^8|J^p>RSL>jQg%1ZLLAi6GuqXa*xQi(`6X7$cTc(
z#er$0B0o36s#vj^ZnuNA;4#P~=bZv@tYj5rZgGzVA*i97HLcX{0@$ZSfHOEvPSb!7
zX-0Pp_?s>MzuPgxt~hQ}fN~=M`e>4PS183jei;qdfG)T}(W5blO|&>Zf>3bg8vN<>
zU3oSmnGiCg%bfs;IJDa#3OG=f0ILoU8f3a0>_`sTn`^G19i+n273m2VnhRv;0)z#x
z4)G<p8qS9cV2dFbplMdWxQnMV4zwIip-D29J~E|np^^9F0lQ5%z(0R`)8JQ8%JBy9
zOSlE5-({RbuP%-g(VoKwI8lxyvP+}f=VQR+yd~9WC;~|O70ZCY#iQ|@Y!^**$5uA>
zX35s;Jn@$A@f8?F5x}8#`gDtkm6QWOdS8WCpTCbxe||1_`5us}%Dq;g!%R24j^$(4
z^QI*v-Asw>ld9xy7w-Ibvm+%TN0~*nkic{wvfJ0AAJMUp-EQyExvI!XGs|*yHHEv5
z@2vOAR%Szn;qI&XYuPy);7AY7@z=Rzu3=oi{p7#?>&uUuw`VledhV@0LqGVf5OeM6
zaT`PUfwzZ+r?RBW;rghuyF#k?6#UgOK+-$ea#t4Fd|Y|HRXNz%*aRHpO`Wn%!kt^4
zeCnh%79^Y^b;?s~DF6bN-)ZF~?Xz!nqNQP^*v{mlpd~Dou-glquyr@zD1I=cRXhKh
zXsN<CL($D*TVzd5r;+ob`^DiX0+cc>WNAH}+}v-lYyr>ErHi}uM6|`ZIw#+E*ey*P
z^Yx6;5KA=!<KM0A_ku7nub-%SM?SlI2fY&(Wp$vGipC&IvaLq{@Zjc`?~^N9slj!N
zmVKXEfvp}}ZgGc8<b*SU&UwlmXc?Va*oBAu`+k%0LA)JRLxW0}%?*Iv!*_#jiI)l>
zOdbJ%x$_w@XO$$eN5AX<$lgx?K-#GUhAcOlQUBHvbfAyGM2Q4f`u7`u$HaT3n&TeH
zs8}TIymqgh3!oSxi~^`PfR0qGx*_2lc@7RJuA(jvcyJMK`!H{;eQ&O9k7R0f8KmS_
ze{YZ>6S@J&f671tR=A79o=STc5U;_vVFD9DshHG&8^`ecY`tI;%+rAhAf3FXi$YAy
zn{@B5{moy|YVWVrcDR{7C;ju_74g`cVD?<aME)#}cfpkR*|ircyTw*BBkvOJtPgw*
zR!kD?KLToN?(<A+C^>Hn*ejfKfPNdiNuz-Qv)J5jp|oYQPaS9dMp6qXE}G$UqkvE8
zVD1>{g>;#C%Tx*GF+5@$R&OZ${kz?ydwJu=&7B+{rqI6_&?^~Y^9GLMfOe^cNkPq-
zMIm{rgjY?ed4IztAhTkmnHep2G)v?$2|^QHG-o}i*a1om&Ie)(VOB^58wM<mMB>KH
zX+U6;rH+c>(6hipodcz}34GxFJS<bHPqhEWhTVM4je&{!F^0=f{`+Y@<tKUvH3tLo
z;&zIX_V!XN(R=Ku@9q--#3JuC25cgr0!a;KcnHY2=;pcbGTryP0sat7&e>fj=9<Jq
zrFsuMH!2aA+)~TYo_hzQ8L~u3`WXP|9T{=?8OEkxHHL$GfAoff+_|C9-<;&{w~@Pb
zdnUHz(YX}*(mkOw@&<t0BW}lA)&MZ<2T_$Ot7pH_c#+8waNWlxai|w`@>yBxK%V-j
z*ud*z7)XBg$`VZbJdWQw!MbX<<><KfN{p#)8^ACfc>AJh=7@4xIc_Vluddd9#>D8P
zm?72P6`B1&Yz5*Jmr489aUkOh1OUwQFiEtfTeHTaTrov-91y8(>qxx5)G4zfhqLVe
zTOw`;c9|t935)*s6@D342dL{-u~0)nNFW7Ul{0|WtdO})iM^MZYa0mbv9|@UY1gBo
z;Az0I&<}r?&0!^VI}vm8KAYx@4vr`=)m8^9{j1j`oZTt0^NN8D90z6yg0@T6fShe4
zQ!eu7Sa>X()*Hv=o>E+5#<3(J#|;ojb!g_(;-OUKy(xfx5EB7rYZF+C{o2MWi_!ee
z{rfOd0d-bbqd!40So#nSjXq!o$?{kLMF~!?(_qD9;bHiUD;Q^;0D@~#YBAj5N}3z+
zcTx$?@<}-^P#`uX8)U!rz4yZPdt9yngL49KK3QAl7WzAHO^(9_vt!POr1P{pD}k-b
z1yo|<x3@uiV2{u*ZKUct42|0L{g%#+LQqapJ_0r$0p#aRcLHBAXU$0i{6qGW1`fRn
z<LeV>Q4SN`d#*~C8bfPWhoKJe!fR=7Q2cmUh33Q4_quRK-kkw%NL&Q=4O}qm(=nv|
z#h)QeumNlckqFQaJR_<C^=LjSq-%>$94%VZBw5T=>VttQ3Dx3-jojj_n$Z!*moLpz
zNUskyB8U8~83T8iMH{H6p!>`3{zzujt0$6H%A&g;2x!G8(Tr=r@*q9vrebt#-lCh{
z!8oDmNSQT)9t(Bv!)T5Mr|L|y&+OVro@KFV_nx;CpsIueUafk#X}0>&kT4SH3Cr$8
zf!^nN6-Cs5@YQxz`4-SFi&!Yr-V|#Kf}fJUpp6*FZrudjHnlL%=X_{R-@Kj^VzdyT
zz2gr0+B{jkBgUnbewawce><$Xyf`N<*PrxR9n97VrxDwTHGpwby8@9vJ`K}}h5lK@
zf35glmt4buk+Apd{*jXB&!4c*eovfUd8zw8bwsDcJOb^9fJtgz<lR&sO!(NFTk9iv
ziMFLm?AKZ9l=ZIVkW~MY^!5Ptg&5{BWp95A+?Vwp)WS|bm6C<Ez@zj0`>m>$-V<K-
zjcCe6k*!}DM88;V#MmS2`(<f2bsoH4_Z-Lg_|Upw5swgL7fAC!k95tACP3UJnA)w0
z^Y`AzHE#z}y1IhNk3#>r<|IrnU%fYI{+jtF?6J+HAOiTY8-T;D6z$o)A?#53xAXxW
zS|8j@X#BH#FYjRTG1n>Q9iu_?&^Vp(+S@viP(Bu)ILLp!zyJ8W5!!<w<HLFSM^Qq(
z{ce#?41HG}6qyD-d7k_67C*2XmIIj|^NM?fS54GB-%M9Q`OSKgocNb|k}bgyAYd68
zGpaePiD3un%GePj%)j57FT!ZAfVzhs`rCs8j+ueSuHWIt_#iMxYXn0-^zyVztpFl+
zAzHN1b9WYyeZWD40NEJ!=ohD=XrTG5tH<C)TLCkrt>^jCVeKjHzhfDOC4nTZ+1+o)
zLO=Z^lmxAu#Q<moNL$t}l39V@HIk<TPdZg|ZUHW@6-c9nF!$O^foq(rnyq9X9QSW&
zt>}B;=wGDp-wpjYblkrlc{zo)^IK+ds0V-RvHS-idM$@G#S3p4`v1c>|Ni)g_0U@Z
z^KbwAjk9@mYVc1ifdBAjPyRny=4*TJ*CfhbiZ5V!2~k>mLI2v0|L~js^Y>n2ps(#U
zo`63Az?a9Eo!0ReZ{_8SnGw0yy(iR1)eD7LXWOXm{u_6(0uS6py<WD{-!kxR5A&94
z?F43`#^Sz;Do<aVw&sO)&S=!eXXU?F)ZTcv!Leh061x89Dt^5KzI<%KYD<ZU;@LC<
zx-^)S^N1I4TpEHzqxupa-_9uT2yKJ!vv;&Xfm58TRcuPWf8zb0m-~O&hwI?PeXjq*
zcCf@hvB!V~ld_i?EL~{739=l{H?hneZvd8AZ>~ZO>Iap$u*wm5YvsI_U&Y}NiMMkO
zoCmA+&kv<j!YZ8CwE?YY9th6WbJX+bN98TN2Qug$Q^P;ag1ES7cfJ+T;CIV9C$Fps
zIR81Hx^osQF3ykPfSw?Gzs#^6Qv{^9Hvn&{TrcGBWj_$?HtDb-#UE?_;xC}jkz?T!
z#qKulZI>wCXr<kOW_d*n&we_%KY>PZjt#jv0VSV8fyNYL47;9%FGtmgl4u6pddR3T
zI;J`k6O~`J_kxxTZZ*#13cD&%)6nB%P-fRDD>>RkAkh4dlvY2z^8kQ2d1oZ5aA+KG
zaQV3aj!ew}LTbE;Ll9yc0#bv*G^v2IgBp;exM6KTtwF<M92iO^#FJc#w)#Ed=T?=o
z{*(jzaeotV0wtObZl*L}+@IL!%U#@iq%+5>E4|h`!>$t!q}T8J!wqPS*hh66g{=$a
z;!Hv{Y>j=dprw2klaIs-Dijq`c^WIMN+QJ$4)n^d9=I2jY&;~HaRu{-=jzkwq|rG@
zO;~{dUujuR(u>!urv&(e<3pR9UnG4ufQIbF#Z<6;+47NDHA@pwX1xhw{X!{1q0HCr
z2;kHzdiz{IS=?jR-2XtVL|4Zd>o4?o8e>NS+_Yc!{@A8(nV8KtPMRM==lLqfR6Gfb
zJ8gdn+Sr#S3mhWemMNa={ZM>wAgCgJqlH;>>u}Tno|9kE(Gp5&hOT5BuJkil#jKnF
zJzMIsIdNwu_S;%e@|_uYgAeoHiB=%6q@2@!e?pC$xlZ1Dcsn{ZR^{D^=f&~sQ}C@H
z&UtV58IKSJ`_}cS8)bI0^r_o!hw6v>^A$(Q$Ai83i*<{z>a(2T<Gkw}C`j7HW$k7x
zhyL8LP615Ja=-~pM1FyVEpqO|O25j-oSi-(=#O5yI;vzpo3DM_kMa-IC?3{zZR~%r
z0XZ95EjuO*7|%O4FmvCkv~(ho$m0l3EHzKyt^GzGoaDWizNv0dT{x`&)kP)8v`hKN
zjoP1s>6oYnhZ{)dpSD1d_9S6OH5G>!KngvMYmD@P<a-2r&B9yum%57|%Ex4y@mUPD
z9T+%`0}BVNkxkZ?3!p7V9#j~m+e}rAq)P|y@VZcSw*rD+i~Z%^V|L<^A8MUzqXo9W
zjRS%szx7l*kF+OQWK=KXmS=?}aM9tTWx(ML7IPV61Qt}C&^h-L_NJ^<wcDWw#nWks
z>uV)mdCD24MnGEIe4kT4X{$v@)6{j9D)wOJRp8frX6tOpsiz>^T)PxyVV?*eNQbae
z1E03Le0px6Ui9w{@R1g<HJ?K@vHr+?H2q6`449+omEkB+#W0SX^;mu=wdYe#_qG1z
zjd>#R!gd2??i*62x|ReYCxKYG%W3X9!7<%~r=vxb9JSr1-Ek(JF`V#CcPq#_)vthR
z?hWemnJJ>pb-76EkprzV^_)_xYJ!nny(JeU)V;R$Kxwi`!lr7pCS@m-{mQYtbbGAl
zOy9}PsZi`_XOMqwDjX$xgXB=uiWYJHUW{j?{BWYuI!J!ocHax0){9?E?K0gnQKBUQ
zp}x>`uT~$iozW6s;W9HC33<pNxEsA}Q0$5%d&@p*vx<KN4@(v{&Lg=hPRBw;1)(#E
z{sfe!j!Ru~Olv^d!vQOK8<;y5kSGAIsLMix>_m$DR=mC}F-WRbqw@+l*34r{o#8@D
zWVJn@-_;U?q@~D!ie)u{df>DhwH%!lg6!uw@P&dm=>@N!pgmzg2Vz7CXs;9#9=Av3
zgHIvQ(XN>L5gY%@my*Wl;t>)CveV*y%HGD-6D5Tp-Z}An#6DhcU2N8SG)2aGk_)`3
z8j(qK_=Zkar%LGcqN`!YLZF1Ch~Y{ibdO>t`#=dCxj!Mb1qhtdnY}u6fXS%x>)A&W
zb^iJqmXn`h))i;eIfbmolqVA1dE&KM4bo|s!U^J@70R-}@Ah=<sW3I}{J)sA|Lm}s
z5ChX`1+)6&KY=JPO5;6psD03S2C6Z!q5{b)w;WfJgr$1%^i;SYCWNhPK3<fgl--xR
zBaTkv>|KU-118kqx)wPV_b(nl`n5b)%n6q>@U%<~yPPaT*i-6&Uf#`SGyT#s8;L~Q
z;r0Z$m61eO3M3V__f;{KXX?n_8@k4?Dd}r;;(ldmy|Kc%(b>zWxtzyey6jJc8;m-c
z=(H&$LY|K)*IxAmxb%}OBTlh;x&TLy_{D|@slqpP9EsB%RfJ*Vq1DC6x=C&ZZ7796
ziO3b-7@RS#n9I_i>S*$E#_5(HVtZ^@-{Txt^_s4nlr+u>^Um(C`LJR%<pz$M-@(Pn
zd^p--ov2@cG?X!j-Hz)u(UVY`&EcSMx{Ngiz(%lWzqvNuksVVMm|BQswtI=8V-ddz
z2xbrJJ@F<xXeb11tk4X^p261Roej1(PQI5uxiJ<>#@zrH{MCG4=Oku%Xm|(I6l_5J
zoUaEu)Uxhl6ytoIO$(qY3g&bf{dB7i0qAw|tPy46PYyvHb7$%vYu%<lrOi*!6+Jvu
zWI)EKoU(B&2J(5d2U%PG5~mBqNS3sLFYBiuWcmpv5;RlAJvTsvy32m%x7q2gYyLh}
z+Zvt!-UMZiLXf4OKr>9}a1NEfsxj?OFaaDH>F6=%8nuRvgV>Q*N+5?*rWnUn55POE
z(eYZXDVYCyi}Wk*`M8(<UP?<To~PNsQJCrasY&V$*zZi)f4)=y(bm0;1M#IFw2=Rg
zr$)kvR!9%9yu{|%+;_ig<R1F3BS-{@etHt=_BSACZ(aW6bxXD7tldedzMS1JGa2n}
zG+o`b(Q~h;tK-bGkCbs4h;JPqzfg9bI^<X4Og*0$ty~R4ShYzhBkY-I-}>g)KFTJk
zZafy61~E<y@l5RW$3%0r0$;s%TF9hpwLg1?Y_rGtH(d(@QffNkWMoURrl^TNf!yQz
z>@zx9>w+rJi$P5tP;Y%PsV9wn!fa;?2<^*|u=9A#6+oZiwU~)!X+37I4h8iNz>hbd
zw3v`K!*A^c3<pAvDLJY*4720jy=-+9Az(pDT0HIQyEJg$<f{b^#D1#yO-`%qEe8_K
zZ%!V-*%(QaGPV3@A*3W}jHitNJhlq#x6t&yMvlmd1)$Qle6fZdX2CG<yzkWj>;Gc!
zJ)@#bx2;h?F@TB*5l|5WNLCRP5Ks}xL2@pXMv`O-BosjeR1gpeO3pb*h9VS*fD$F=
zoU2HZbKUjo?(Vb4`M%xf+#mPX{nLZd7L2#vC#<#Rnscrl^9N_-1Xly8l$4G~X+%+0
zy(Fh}qI8AwsT2c)m^%T~`<&cg>#=_$#k)VwPpDTw8?rE=*WA5NN1%ye033o8MCg_{
zVXS_G?)s;|1Py7W>oHi25FtG`-G#VUUNY#KK%z))g4pWV*0CgQ1Ye4qIYPMcPY|@*
z&i+G-_04bc|CPU3xV=5JxQz=L{OlrVT{u_J+{CxJZ9Do$pL5&59JY_E5QilxpS@4&
zjPf~eo)qc$Qk(CE?%Gr0Hhh;nMZb9yE<u5Y!Htcpggnt&pH)^(*C;Z2s{U2OFm4lH
z^TTHAWp+7X^TkY;8%1}o`H={dN3m3J99dEdNylfZRBGWp*I_<*9$>$m=>Zo&fL>c}
zqMg5C3v`Hy{Vc}96gFY}=v0<N2;MSdTJ=x?pyaFa+I6hrT#oYI7SOzUmqLjm|15fN
zLQZ?1d0xvSp+oW<2WpN|%nA&XbS2~w@T%I2&!{LC9cL0kW}*F)4}w5u$Lxitw$S~`
zhEr43beB6f;ahmnsqn|*u0xP7>vw5czoV*AqV%?O*XSj5q=hW-t4<mxTt6bA4p2&0
zz5P^#f%`$KV<pXJ3H%+!O1jBS=C{t%vMv1%=KrC@l0>TDhnG}-r_H((hm;<=4h5Pg
z_Jrx_ylsru*ij(Z?;ebx?{kPG+bNH@RbA6~ZUS3rdb)(FvHeED&>y!M_P?m(!`$pl
zdrs;s+5d8grrXgRqYD*5o~hk*z<lJL%Of|0>n;Q0^8@g{@|if6VYz{9S%`v28?t1n
zj*VL*PAY2ycbL4O4A>Jw7uibCchMnfoTkK|elFe6Sqm*oAx0><4h3A5ngs%hkD%vL
zd-CJUCD3cb-2U7W)b6nWV5WYR7ge+2CKp!rBVsW*mtA`yx7d1W(k7SfAlf7?cxv>3
zHo;hb9x&KK(tQWMW+-qjFCAS}+-m2@)ENJuXQ}-t%~31A`vpu>+Em$j@a02}l513^
zF#4dDQbmF<E%pQ;4`FjqMjigH>+?_bA30L9UOb!j`;wD?Oh$cT_%^2Qh~sV?&D#bA
z%YiXixhnd&v3>jF1Wa)?Dz+6d+ug(Ior|n`uBclB9?~IVu0zR><I(}<o%y3ZO|@C0
z(9AS~ADuJiE6{ZMqDud9jh}5;ig8Nqqin<Tz#q;<3PHGnpB<&=^%;<Xttl14xyrFv
zeg5IV(X`cyA`%?z+8!7&uM15;DcK)7Yvt|A3y;Ol>%$FW|LAtPh8(Fgh0_PI$L?f1
zHl=lEhu2}8fEv`kvml0Bb8{4(U~E|E&W^)9YJaJ7HUN1)o)9-)Y18njgrXx5Qy9_A
zDzbk}HFY4>9L1>p#t|wcPbE{3I@bqt<`&kJ&z9tJP#DhyBoxf>_w}~aGCXr(2=C#i
zBXt%{`7~FyKB%IltJR|UjPtQYA6r)GsOC?Tq4F|Raz!O>XM!uSv!jWhklt|g5>CYy
zqSpjHzfE#uIA3nJiL?abg7+Nm+I5b8WY*MsEu~!&94+@VwJ=^&Cz>w?;g*kp_UW9Q
zkL%4h3Bz(86Hy4JSZqV5n8l)fVh@nNF91*+r~!(9sFsVG);s;v*HmM?=Q>0|nfR}p
zO^HmIvBLaIk;Gc%Mpp*9@95jSZNu@}f+1hAZhxBjTtUbA;A|ZUQfGXqO~e*`pGU~*
zyE!H`fv~N)TYZVUoNGVY?0WXRW+SqRt$S)YFW#8XnFaBujhA^Z0nqz~1os2_YZ>j5
zgGm2D6RT*KZ(NpDWTv8O>pEl#_HR&<Mj|LbZ+G;xi8Qkj(g^u}HKZHv_mH-nC(7oP
z=9^0s1(TsBz0j^6)vRLU{4ATXi;M=dg&2VV@&V<}$e|c2vK)K<<<(&Xz+TzQWq#y)
z&PdQ<<ufptS{&AP%uS+t2#fUp;nw-}MkLR`)$_;Y&VOA!35S$CNN~-<o2|$AorFu)
z*6K4v>n9PwBd=p=!ljpLh-jE}4&#g2?I>I;r`-R1Fjn26n8~-FxIc_YSXf(nopf`O
zP@AJ)KU-qNZMyASqDic8=C}zR;UjiEhRh}F*?n`vm)Dwm<QTb6i}RQ+9SOVR93W+4
zZBl5d8I|R<_E;Ut0bBB-vMoF~X7ZrzN?>4jHx$X{kaC%2t1A{3P>d?97mA&;e3(r*
zlm9tB_>eatb{$5M6v9aJMo#SXr6Iu#Y|A@(1b79+KZ;6c2khyrR8sJy&UHUvB90+N
zofB+93<*k>Y2b)94hhPLPum(WMtZ@;%wLCtMT&fb<vjMn>FcRxvE0P@kUpqM7Tm>g
z&_!T4l$_~6`+GzE-;s`J8-O^1n%-sn{@^t`w^6d4_8e$5zdIJu)i-!dculut7K=+T
z=(5r_?EG%*!TH>=?gqvbOZkj`tuj|7$t=R>V#MdB1MdWiE3mmv76Ytle-u0`)a;7n
zt<H~x#;;&s$<(|pzL;69?U_tQW?$FIE{=t8^`XATu6}(6pN($Ly={Bj(NAS<GZd$$
z!n=nC&#o1eE%&4qJgO%iAl;oI+egE`QB~t2RCyrMB?!a6<!$4NvZc$5vbT0AXd*l}
zFwlPSHp1@vr%v7c15efyop$#677+A6CS%F8yIjUWz5i<iZEpIMtXQ{C3O9z*nK{xj
zT^mEtr{%TmGtz<EkD|5JWqSdKv(|c=Rko(F<xdfx3YGenUHPdNm7U4PFarjsMK|-Z
zEQt?H)$<SBtnttCkyHj<3VmsgvhwFZ7r4C*>u>sVwE7H|c$6(U;U2jWD+3)BJ#Ryj
zpupMZl`qK?ZIa2$p)o6P*~Lme4UKnRhw*c{(BG-aqF;-dfU5GRcGn=fOWPKR2i<`7
z&;|@Q<23Fonb>|{mg7asu;V_LsVE8M$wK4&IgpKbbt_0*=Gz1$3?Cf0`Q{i59Lrpt
zALxTwf!OKU@LR?siz8KY$aN3%yXwtKe?D^2iS7Ebv$<^A8ZTiBLd4BzfloXp1(%`=
z(t{b@fHhF>yH%n}67%vQ`dor&7Xcv5S$tYn0cd`^sa3V!TMB&<I{OdFxa++=*i#ec
z)uF#rbfshj(6n|5H;D2(Kjr@6L98tQ=yO7O{hI9;quN*MozeIA^(FRo3b}GMgimUe
zhV6ixMK$?Q<469g14GvD(9_*bohD^wYSraoLiJ2C8@;}R#47}JgPZM_ncChn+Pt}j
zA^8-GCViRj=JFivN@nW%GM>bF7D!CJ!+6_q<F*HD-tD$7WHlSr_7>AbMGo58jQ;eB
zDxj2QFszMmijS~rIxMtkJSb;97q?#9J5V^du<rn~(~TD2X^h2lx-aExbCw@vHlKxR
zJ6BZ~D0ek#`=c;4N(KGSQN)t6?V4@g=hGIN*InGo_?{VdU$MPA>(P|~nSNqO#-I&4
zY-Q|YDOq^{#Uj0+Lmta#S+{4DG@N-*Ch<jFAb)HM<N=|x9D>}<hx4B%UF188FqweM
z!j;8QVz*#6<h1;bij~ltpT_b{uF{y|n&R;ua5x-K>PXeJ*#L3`Mm}OGX80y~b_!30
z;l*gOdElFgF<i49eZ#z^GIaXhnWNH**FRxjbU481YTxrti<O<9pNQa^8JssNfz!qm
zbqsw!ozp08pevR9L4pXW0e%uVMhp2+!f8N--zE_O#lE?|EC)&IL<wSS9OwyLkN7SY
z!Yi`v_Lq?_Y2)0RzJqz@s2^=a)ZFRdl&e(-8JvDhjUe4HUen$Un^fwrzslb`_#Gfn
za1?Z3dLDcsr26LULtcI_$n#35h81<kp@BB(PJsXpdbPf*X>~s#k2n_g&qaX(l08Y@
z;r-1Hh%P06OS$McnN2fS=bhPnXwh4$nKiAw;7MF$E>D;7CmpAW>aM;v5gl=2=U3;0
z3;JFCyw?O5I!zV|yoj5x`-m<EJZ7!kIu28ZO*%`B93qYn+YOD`bqiDt<(}a5pI?@%
z?Q59f&XCAu;xx3lEU+@YFQb-o;HHMO)Ix<!vU1s}W5R7h02Oq-L|HSu-q=m&L)y-L
z1&%cIHoKy>t|pJoQ%o|ha~^$y(i7fl--@((zk7X(xhp{JJc2G@(lEG&?dn3^?x&5x
zwxct(-7U;36k%U|sk@7azFs|9xQ2DNr(w?rTRgE&Q`}dmhm$ysI!<c#M+Z-N9-(~-
zMS))b=+mi`4k2`734$yE4iI_bp=!S;erXLxGNhTa$dN*&Ule1WFULX3ddLj!nmlCg
zy6l~iS3U#DJmv5kLVA%{LK)bo1CmyTCUiCyZk$n3*gG9>Hyz{UsqUPfse_n0D@wiU
zm9>}u98R`#S~qdtg!*2x%0#R?S;GY=6{p#O^#R#LaAcBhEiALZZee`q{*I$&fQz;f
z0@0&>!h4?)C)0L8E1@dr{*Jky$n8?roe*5_AJb%SFAwE4#kZwg6fJ%ZwUKRyvT6@l
z$Oe`fZP^)dR2_Q&94J$jw3if1O7c;y(G?9cyggqFKPFdH8%n)-;#l)v@qzynK=t1-
z187ETJf!Cty-c%FP};q6>-n_CVqc6@4gXMcxE|^qkzY@z<XvB{1AY33$U?)BOWPLX
z)6}HS@8~r`<_aabx*p!D;KJ&DBFyLcQtq#}Q!6#hro7`cScBoM48~?+ZV#l#p?hQF
z&}a^-RH?!D<}Sm_J$IOnw&?9lzueY&v7{e{#~$75Kz}c3+9IeCcnRuZ?|46R_L&;3
z;tdYCh!(z+WmB{4_O3GJF|bpL(qRg7@`RpAJ4NWr&!Y$Iuf1T)OdBm)jaxq5UN{k~
zxur7Leo~9k9b1W6`p%5$O2-G-0c&kNqn0IF+RcRGTf>FZi#(R2Uwy#-ORmrJKF4ed
zy5jn^i_(?eJ#G#fnI<}<`aq*!-tM?T{k#Yg8;vB{h%cNsYmiO_6c7dteJ{q|nYth4
z6>nB#dvjULSNh-vk6C{=0%WW##WS7G;{$ZE)E2Ha@*8J(zR740%L3$6PTX@e?*&@Z
z(8;I$#oq?yjM%G$Vo9B!MAqOZ@AT?h*wD89<LWs;CqvS_Hu(1k1JjTgGvP3N*G@;t
z6%`q#TG9bGE*sU^{GCKji;mt<vR#cl!^aoAo<#B7i%urN(fOluUjOIHtU?kxe&z|E
zt>u5G@4vIqxVpSY`o|Wxp*G9_C>{yH^mi<D^`BnmP?m;bh1K?pIOl(CUf&Bss04LU
z4{rY2m-5==NFB!-l#?<foeNNS-{CUZLbPU<iR!@0kZU$Jtuy~9uU0eC0MwR~$lgo^
z))&7ReoZA~%>z1FW}j;>I5t34%cE%!fp%KQqVesAy=gDYr{7x%<)8grl@14(_r;<v
zt`W!)_!7y?6qohtFWJZcbJp#)AJ7&llK0yGA{SgB3%XJ9EA2=5>F`?DurrJ&Id!Gv
zzHN3U6lBc3QN57o`Hn95pjg6?f&Sq|&C(ay&?Kq?@Gpp$veKCXZ_<$(gyhSekaw?G
zF&)1qdk^3&Ag@4O6~9kR#MKabT)V4KuE&kqR^I%llIdCkV15a&^tgY2uue1jCID&8
z3%KPNB=6_zG4CyXbH3XI7rx1({|KJw!@8z<nLc!9-X@rLe~vj1R}@b{LbclIXTjGN
zfr;9!ioyit>cH)U&=vWXKmB9l_B#+CLVPYe-v7<Fuym`V8xDE#DDC?(Hbqwy593w(
zf7-^tSwQwe4T_B9*K1Z%aT84$nx)!K$YJuc(La1E@0m)L%2;~w&M(sI|2>cY^#su?
zNHRa@%khgrdi<KIn|?ueaN?O@!i{cwD>Af6#xwNuj&a*>k^fW{z@IZNh4;&|car4S
zpZ=fzC;BE5j%k}IjDC|){)fMp{1cwMV?&CB<e!%PzrF&ux@-4bQTlh;;t&7lz5MV0
z8eIc~Le()YQj-6{Yxw&kta0!{f_|p^GySigk&X+V@f~hI$$x$Fe}AC=!%Ir`ffu4&
zllvg_?+5rF{Ph2N(XP-9uS5CL2WOL2Pe|g_f4so{+qF5t^cfWVNEiO(fD7E@pqeOe
zj1&}T*`M>T<M}^$_Fo@iExA`8%yG&uO6pdHIY<Sv8C-U<a|8l}{x|QObpJ_e!_M?i
z&=R@Sf?<L~X;{}y`@j22JZf;4(>X1D=l|dQ-n|KGIL=Fv^a`H;?Irz(yFlI@p$b(p
zN2UK$ZS`+|RY?beAnQUgqyB&R+y6cH`TxJ#zr95uj{SeUYHn6Xf$-o|oS=|$(i#Ox
z)yb~!I`_9b_PpthJ5u%2(sLv&)2?8j1F5-sUe4%oXu#RGxf$9e1vA12kaOUc$|2v6
zTeaj%o8H#ZqO;-BDzb>-0A6)4yZY#Y7=-(L=u#s@pfTPQmF2yiWw|yNk0S~8$5{U$
zrfcqz;fQ-BGr~s^w@ZXgGvV@sg8zX;b76waBYoU_+A<%<9G{-RUq{X@h-(p!bT6F6
zCkWN)(khN8<%pa;W&(v{yG-6u{9MewY^$R(aOHNa^&WJTG~f*H3P#7<#rL9C&S&7C
zvMvog5n8Z}v(J2uoy+MLQ|{WW=fb@p#t*JOT5i7e=Mj?+xL15&18ECRHhfZiPPfwP
z0_Ab;1Jn|NMy{YN5r)G~IUC+>72w_6Q3_>55&D$N2?42z7a2l`tDTz70Lju6zWz;2
zRTjS(wQYDyfFc^9J>rr)qp^uHVWEj^Pvyzq%*#bNExkd=-0=UNVQC2E?HQW4kzzXo
z5EN>S_ev)=eJ_YKGi_!0KdI*^ycNOhnaTo;Wi7|G9s^UD&n#4A4eGG`eaG%p(D4{6
z@o)zQ;X(N%U?SV&!a8ghW?I0q?OSI=)v+5Me2Tv0mu%u!9JppcN-o*&wA{!GJ=PM)
z)}7Q#*uBz?^k38w^ouo8Ckr~t;H~b;rJ13xSMMeugx>OvQOXifSj_>ualLXd)$u5)
z6P;L6Re9`kVXrZprgbAvRjd^QTO$I|57w%qJCrf!QwC|$Do?*t#wGe0eMlY(7FC~*
zJH;%!(zia>*VUV6^r3ruu+&ZM#J?hvaVfF_5iWb1YvUi}dVxj%3@9mn)tw@rE=Yp7
zXsmm}l6zIT4mu_#P`|zCe@YzgiWQ|K{#Xb8%a)4!Q&s5~i5GlVR~zJncgHn3>jrlr
zuTqv5$}RNU6Pn9wn|&{K;V0G4md#KhAh7R<JX>@6-QHE=ARQ2kdNF#Ks2^SagUeW6
zRLuRrQ73PEXe#9JbgEUkNF+yNMW_VZfw#P{k5HGZtcfXm0a~X%1ez+JV*4z2-wOdu
z<hB5hcna8Qlqpn>qWTGS&8p;m(7B#ofY#rRHTEfofF3^$ywi0dfMdfDRHP?ODRjue
zw=KFZFZ51eAkl5#!QKv4R77Hn*HsXrM4yKS-;BL7`T=#6B|}~E7v;(ljSCu_4NZE$
z!hSx}nvm%7AxGG?$EHmNbKPQ%H)z5REtC{g{n6&{w)JVw=$nc7AIik^oOKYmX>7>u
z6^g3YEnW?JFBy?+HTao7xJy(;xmoOFQb~NN>Qv78iB+P`{f?AliOIOb5rm#xgQR`i
zN<;is<KiePN6-dsIH-|7r{PK7dxZ8+t+yX`JNO*F5Swww*C`xA)zF@Y?rnOJg97=|
zjGp|_R4D~a&fw2f-x3&#P|z&2-3{e~A<XNs{p9ePg@937cmWW<9m9(kP;GCLDE8d^
z>*1-d%4d=~D<(>3TaMNA0Is;}-4v@s!s?2cOxBfbYhX;1Xx#B+*UhH;i*vy*M!hpd
zk$LE;y#0Y&5B;)KyxX6)AZh{g^37*9@1tY^Y<Kg9R%ry)&Nrdsq|QB;eo2s06J6)y
zw(5QYFuF_}WRqudl#g}-u7ULZNvtsIXPvR@$+0Hz#aB?a_`S0<>8Cn5>eW!fIxvi8
zM*G^Z{EW&~FOAPfMVcc8=R9b1zHLV`6r{0^E#QcQQMq8g{teUy7vRk-JpGVq9ixj8
zOnVNU@bXqx+v=dZiqsZ$Uf6JD2a{fTg}sjI53Dq+FFE+9&1Hi=Dr4jmQdwQAq9vbT
zW-_KM^3QNYfpp*wd(&BhM_i_hAQ9<Px{#bA+g(vmSY{iOA80nxzr`%`cWB-`JROJq
zAScw5)n8z$iCQ~CdSY|#Nwr1-M^%W@BMr+}b+?rV>If$@>!#w;RmCOjv*z^$>IiI^
z(RhV}xtGM)5>NLhIShg@v<F-d&GF<9VN7XrC{I1krK3bVA5XKsr#$d&9`2vL%k98>
z-rfpt9lWg)&q`$B*@@+Eh4b3oCBr6v!%RMl3t@jQ(26j;HNj-0;q;EyiG^@#JLzr%
z%W_26#;;FKVmYA5FST5gW8*s!^mKX9u<Pp7Fv$ow)@b!-WVlF6y6&pRSOra>04o;X
z-Dbg^crC?7`y{;zSJLe5!`m*<Nm?3IwNKvwk>=OL=CW_nD;B~#iz0PpHjRA2l#5a;
zcV>?ucaQD_MNzPcBKTyBG~w2<3&r<7`Ft&d>pTT0hg0c<m){gJUtZB0h<7fM>`;PC
z^QP|%c4(!92%`mpZw!&YO+a{V?J;bt&i$%+wDwfZEOaG)46NwhpgS}^K@QNCQp`Nb
zF;PCBu`(mW@3uKUl%Gz!&0YZQn5z-tK{>(`Tv~;kAMZ?*f0Qi)(dtZI5y|+<(r2Bs
z^>XfxGyYr6E_)wZafh+NdHwmUp@U-jVafATc{%nUF8ujANPwZB<lB@<W5KaERcP6X
ztn8HHqdI=2gkANHdcw;L*{Qf6lvkrlRo9WTO}#*uCUsuAf5>gy;$}YF?AzEtz+DrW
zq(2@n)Tmn>KB*}@<}fB+e4+MfVaoI;Fj)|JPNyul0syW4l_|JFBr|Y(TYLmIVWc@k
zT<6krzw3O(o&f`_kG2`W0>>2QtA_3>RIvWJ3-9%ix96*P9Lc1nB3@4naUZEh**Jj!
zbu#E!SG!Fe&;1CiffF_JwfWNHs4J8CxmFr8Fcj#ujAkpe?90gnVtqrIDzd_5*?+|{
zl~`AkIo_!u$K9|6Y<jbVGK_0@+J#1KO{Ir@^?`ezXX*vT{#FZ+3O!?26!l?cH2DG|
z!DOFOv4yTofp|%|=wXg7{)lD~nInt!w}~#B^(ckR{OOhA&RUvGo_a|a=Vgg#!u+)e
zF6EB|JDYFJ6<q1U09n~gB$=3AX~+x&3q@JBt9Rb~U2R@_T#@e@xTnh$$f0c_mI4IK
zz#`K%3S=oXAJqxz=<jI5Qr!Ht6ed@O{=8-nY$zx3f1JhbE2C6i{J7U>moc9(2+Y1&
z<yOd+)^SM$^DlxA#t@}qSyNr)wfQ0K2_DBHAQ|f|U-ACLZSX8hXLRApLL7fLck|s3
zR)XCdw}7ptKoM#1SK$X3#TQ_XcF}4+&lZ)u|Dwf9S|&}Is%usE&E{?wZf~Mc5T6`p
zwHC8`4FDx7QNiS3P<r0Y<p4OPN0p>j3<|{nnj&C6Snri};fT<2cWSECvfGzUz)Q*H
zANaUp0RZ3BIdh1K>t2=vfr&_fx6)os)F^oV2`tw?1PD1L3=Bc@_5Ik%`5xnr^qW+f
zVQ-W`579c4Wdt0maa1!VlO7*Pz+XwI$j7V?+@8QXr#2+rUdY=8S>Ya?yvHV@n99{=
z1RdZ8*k3Gfx8V*4*(Wzm<=rg#!29=EctP4F`b3gp)??|9vuVB<P)+}&r{*rarN8l@
zPQ3aYwv|1#ZUW7yQn>v1o5y4E!8Ge0-zA6DH1{`q=aUB)fm|j*EO7xz?L(R=X|MIH
zk-g(J#=BWLM~pGo<Jla~(G1@P<w|(P+UEgK%Bsj01eE*XhkKI<d)-qK>v{<i%V+<r
zI^C8|Cockl=h|1EjP!%K0oYDu65aLr0gS345|3u9qR9@atU(LDPH5jb2Dpxroyib_
z6TQ3q$cgi5IWc$3nHuIjRcns`@^%VlK+0-Ea+QLB+W8hEU+{KTZ@)cO#ykPemT)|@
z{XbY9?T9DAwT`1uO>88xK|viUH|4Z_rWe$;oobJI9N0*ikVLm-<lNI|NG0%P%+AMC
z{e=izRK2Gkh*0JNTOO#R7+R>e7*R3$8?Axwt*pV2TB&0Rdpd1w^V>T9n#^FD$=~k`
zp?Z~f2<>WS4K~1?NuwRktql!$h;@;<^L3#=fe!O^ur{!^h(tI+F=N$ZTieD7+?P)W
zzKdUo8yPmH{tS^t=pEi(w)V;2TUIeT*mnMcPaVdhkwp{y59T1;=WH%dg_+#spZb`o
zEaV6`Ya{u^arQPJXg&{VncVD5kfITu(>V2Ay_tNYr@z?RdQ!{)k$q~dyH1iF*mb#N
zLG@%ZRFQYR!q9U2omrNV>DUGSU6*+iI%{pJ?K$9B7()*zA7pSsj;E|SHa-y)ZW}m%
z9C^vvP_}U^p}Fkb+KfCFXppn(=R3W93;V64wxKB=o7<2-Yyb*K>zcdHzu`1f(t+F1
zt&?TaqI6z^)iG#rf$3+n(eB6UJ|;0>n`Z_S9915RFs@(9NOKIH7eG!rw=E<ej-lID
zEzN1FS&=y0v-Dx-z)9y)P$gIg<@S^;;Z}E|w?7{kkPGUv@atT5n@{QyVccvSEQlY<
zTx!(PRhl)HR0aA?YyFCJtigw(`-I{%Z)NQ5P=;Zx=|+uR)(;@1eJhw`)oIJ`SztZz
zUE>Zj_DnxbHo~abBIvEn_Sk|9FO0ILLHsSDvj!-|#VdFD2<i#!Q;6vJc2|UwLsZM*
z2)2aAvgD<y8(4P!fq02iQBl*S(|HA|bZGVoZCpM|kibH<6?VvdNl|<cUH(?mW%1~C
z(}j@9Uan@7D*~92LdPe%5dwC+G>b-=L4MZ-1Z}yhpM$VZj_51<%7lYou5F%RY0bg;
zar+{Z<9i}K$9O5TsX{*wtzeP_!)^LYTI<-F@nqu4LfCUH*ZAVPT^|?Wyt2K_|H3**
zO{$2n8#Dr7TIKnq5|b4JojTAoo`#4SoYC;)%JVs`niOUJNwc|Bp9S@3`BSrvGFuiQ
zLX>ezME^jL`KYH8LG~>h!_~*nl1d{{5wiV!?WJ`(I;wOZ*ac|p1XT)`uzP}c!3TF1
zxUDDSvc}$Ze&770PNZBw<vt6eT+9U;wxFYJic3hM5Zk%gf2kEbqG<IxMpEaw9n}!T
zTw8n1sSBN5JUEzRK(i^3vTfK;y3*t@{_VZ<`@#Cu$Z*m>59c*5x1f{5cHbDiHbUHb
zJfzG1Q1*C57-eV1u~x9Ye)7$q(H3H9?_uU}|K+Dr^wc!NNthz8YYk{^-OM0r7j{7<
z{8dNDU3Ms?yB<WN=XyI9n3+!y!$ve1a#9qNdVw2Y!UbGxmmB4fYW1jSnockjtiF_S
zJK7`inAF*U@2+0W5&9cXUkNTZaP#WdohDu~2Y;hAs!~L{hcJ~UvAlVRlzcXr1b5`v
zuT#*W675fdt8_nz_ybt6gG6X@-?;UeBHDyv8ic6128CNF83mf?dyL*R@BJ!xj5N3I
z{)E#a$HiH{`vOYwM9EJ@3#U$R+_V_J2d6XCMdvJ#rLy+7?zW@#p8&V_DgTepw>D)A
zE;he;pmP>+sX)vIYUkCPMavx5*E)%t%`US546HSCJ8=H++DB2|YgRJ%bl=(4mtWBz
zp7X2nivjMwO#KXLOd^bh#3>qWE{efkD5vHcxSMYk>r+X(3w=H0F4QZ(_*lKE8ksg1
z_5F)b+!$i=3KkZWL?1NpXtrr+TPB`tffc(q#;o0noCn_{Rl7OtPBohZZ1BqW4k}8{
zD|b_;%7JFlC5)1Z5-N+{K}Z@}l(vj4w7&v8W><_RaH5wh%j8OE`K(cdo}gn|jumh&
zzw10tIJ^joFaKf{LA`zZC<PZJ!meH8i?R+~v5fn?C2s`0igKSA7u0I&BQ8<uA0;sB
zXM-x@v_)FQ;B@r8`l5@I)BMtZgM3PYq~waz_FBy;tjl=4g4q$of~AWUwKf|*<PkhV
z$#J<@<v?34dtO6?lDcq&>q(|m4SNjiv4RIpUijnf{twjgvdG{JE4Exk9bzSpz>*@H
zR8(J-6Ves5cBQ_FC#5uVOQ3^|<(@(ncn0c0=+BPg#JfzKhmmqy>Hd=s_U+gZ6@Ns{
zGxvy;uM3)njAL#IoZlKKG)E~8Bpm37NNJK<H~9rSA*Uc(Yrhs_)&tKU5A#ean;DwR
zDwZ41lt}CQ3`pU+SBgSb%lV$IU*b=gFY8xjr|t9DEea{f?knN?qJV$YKy9YTNz!ai
z(D6wE{3I$cG{qJruw|bq8qzLGqaA2P>XfaWUhGu~_7Q?XR+VJ5wW{0=^D&LLYiG$$
za0h$j6^AtN8MW>8)7(k{KG~LOe|<@yNz8S01?j<fznB#)m#spkM@0DCK>b3vUNP50
zt4_TG2)1;u<4jRM+LB?Oxg~p^uceaQB0+0WtO<*6ngYWUzs%4`&bc!2bt{@b0$R(s
zLz!gYfp>3m3_&s6R?AA*eF6SGECbtG7p0nRpm9oeRGim@3_i7vlPzQTCBuenn3lq?
z#n&^e$6lQf;uSAGgSBd%7rF~SoK(!LE)LhP3)#MR2{xSFra_!xtCj@{k7PH(>(f2e
z`Paj(IFb{1yLH_IBl*7AxBb-ih!HlJub*1k`P`4u^%shAaJFJ#im!9v%q$vx8;H$1
zRQuW&G+2^RL13kk1m$a3pvjZ#wu0;VX1D5RjMhglqk%B&KtHG26|=qZ1?7$9t;+6~
z3TTI~GgZiTi@5K=5%J=Cl`G|6gm`RGCzIMvGSxZEh1Bw#Y*p#aj?9SL^pz#33z~^9
z!F?aFz$dd+R4JidTr}LXg4?%MV(sS0LK7Ws<<>fbn@G0)Xv=WeW8lAdxFkO!<GfcR
zY;6wM9ZSci=M-*QO<!XDvG|)}&b%WiLt2(A(YN&HbAq|5vX{@R<}YEFl)r6SFl+@C
z&tLhTqq$NluhZ<rprRBZ+n#=_jjg0K^A@wZZKlKox-80PkA6YuE62~1hZazAeMbCm
zI34OWqNq=z3~9>d<7$t1`QI+{29uN*-KKBGARV_n=(t#Ukh^t8=#`)5yT8=x_b!ob
zy-j=Ix#&tuwgWVYjBZy~Agjb{(UkKUDGWzl7;XZK99;Sj5@F?&MZrd`R2mdy7AWO-
zKWj)J5<_P0Q21I5cqN2*9TvR9Dt_s%v~l|Iy)NolmrsPyk8Agj!X?DdS+M}1lQ@gv
zjr6hdUP%n3<$D(MY$Yp{LL9j904Xb-!BfLDO-B&rXS4(5Y0w6O>?2;wTTbYZ7WLI0
zLa~{a0sybhlGM4P<o$tbI`OV22fDOx=gn-Z6-CVqChBiB=ktdM@k#Xi4bY^sbOo5h
z1~pe5E`K>5TWP{5<FZRY{k=uUJ=|<lrjD&A7|51urD?OQR~y%%rAFLIRt3gcd{*7u
zg=AkJLxev37sYuUG7oiCfGFUDRmNUIl)BPPDh&xgYMyN*e3$JWXLz&A&KLvB`f9qc
zl6rGBKuE!|df&Wj7%SusTA~go?ZJgV881Nqh2f0*A@XI9-5Cc9j?-H!cu~cT<jd?-
zU7|)y3mOMj7@}WcJS??bQ75~cTf1J1u~ZQym%rTelKOk^IWRhcVkSOIQ(J1Q#5QAd
zlRG{{m!jclm)co2@y`5jpDtBo9ih1y_S;nh`W-<^_wG|~CCf25CriiQOnm9~C0EyG
z7HiIag9a{DDaA($mn1bOIZZZm9=d*`P|hgLjHhuB3RslRnN7Saue0D^S12o{qfg5!
zVy`_Tmouw%s*aFaSf3f9IL3VS=*X_gN_q6m6|$Xx<Yt8}%EOD++G~ljf~)P}+Yv&&
zbAZxeopY~*Kvr#8Yb}`ZN4zz7k$C_6=40cikfTyDPT|~kQ^du&k<=w<L_Z#mV!f`l
zyeTvWwMTKLtF^{*Oa1jO9&@_K92fKp)#7x&ERt!m;bRoIe#9E7kbA$}R}~#%(L;E(
zn!9f&5BI|grN+GVK=To#o{|i5T=A>N&z%Pbl<{>$ccp}Ko2@9b7@7uBHWi7gRMuQy
zq1+I6o|c;o;xb{*QgH3lm3dLn$QssIqGy4rDm>*cG2oZ4*A-#fbVDSPI(`8g)p&SG
zv0DBlj2KWm!A*gA5<M_|*VCH5#-UUGc!I`7n}LQzQ20M1zGP;n(#!9pIUp$VkZnJ8
z^c2?ou=lPSwM`7UAo+&JNVlF}qeK<t0B6oj`A9i(e(^p1_4Ly5nwBZXj3htdFFDh9
zz@Ulix>jKsEuoHcKd&(gGSw!v%uZ73hi5$>EN7iL5NFhCW9y0v=yF!*68~ndUXuXN
zqsL@F_<{HPR5&Y@#D&}Q3*rpNMGtm)*f!-O`G3-ia!K9IBY&bHBxD^@x{z&`uR7#U
zB8`i!22tT>Z-Ye;_XQt^+0iRY993SJ(+ZDRO+E?C7prY5?LM;9-}Ucws#VS}&D^23
zLbgrn>?3<YB7cNa*4Sug8v6t~;l}qIYcy9D+eHP28BdoX>K}>x<pG|cbCt>wN`nFn
z)1J&P$KxEkRR?)llCb2lGh45<%jk<ZeWXUkxwx{n4Yx?l9tgU>x@DZs-|f_-Zo}|*
z*04#Do(DENi(q}yY%kun&mZgSeV`K&4hmmPiM@0Lb`Ro#Kue{@?}@EwypW@LQ9-d8
zgwYaNi6P%w4)VyPc~{JyIoR=3c5P0gDC<uz0EexJ^TRZSk>wUX#~kjT8&I`fF9H2?
z{xAJ=YSul-KvVnWkC$rLDUv_P+kTuMyAG~z90^=nRrRjVV_@@amY);HJEz}8%-(3%
zYb~;r?1CL)qB;Lio=FN`fOM8;Xs!$^E`ABu<0-n!*+}UQ7G})MD2L7sNEh3@TDawH
zO}^CO$j$=Jnr+DD=|CIHDY(+|&;|iob`PG8*?!8wHm>j1$u(E}qk!&#|C2jjbOxZZ
zBjLi{m4+Xx6GXLFA00&7(G_&5b-mVDqZEnR^dp?)FColc$=Z$2eD6P$<5R!%<>L3{
z$UF(5DJ3(B3x(k7Mcb8-*$ANjv+zjJR^K*=0M9*x^eU>q@eKUsXp)11G>?QYDzG`V
z4ExC!wA>Ypn&KYU^rYHmjrURNylJQPN!Fe{C7q!)bfG9UxGM<vph@gxq(sobXzK%e
zjHq2%%l~1Q7#y5PmAxJnQwldqdNV6$h1@CwX!7M&XWE$aVT#+cc1*I4>bunB+aZ86
zO;)|vNh+O>o7q=%xI@It(emu_iDyc$jsEqZ;tIM62R3FZla$j`_^5;_WDmKa9Y#{K
zmX#z<U^c_eU88gkEkKsZS6j3(>TP`vrY(I7UtRfe)^mQKWQu(Vpp-A8D=&bB!$+8@
z<P@mj$^_}Id~%k+7Z=E&9cs!?ryu{fTLr|>YlDvFF3n1Ick!M|L4&Cqho~v&vatBn
z6abBHCuA|^dLf&m^qlw-Y>s5_9>WV&E8u|Gs@{%Se@V$H&O$PYbcL2bd^6u7xhj2S
z^~*g}(laWA%qY9pp`mvcRDuz?TBv~`u%b;tvYvgrEBfhJ`wIfL&%5)YL&$rR^QJDH
zyPg6+q55<CLDP8l*hDROoPDI)P_jsuj^(P)d%(ygmFFxGR4;+rwxk|QD1H>UmUAp<
z!sC7`)76N!;ox!0F%8B;+EQL)DM8DN=GtcqO}aSQpZzW1k$<=U45!v~{rZ*#0pJwr
zDd~)8{q;kqB9AKj!N4L35h4AJX(0-h%R%`K+MP@Kw+5<ZQ~|JAQu?+j{)Ic`rT~Dx
zs((nJ`Pn+zca`*6ve*tv1V^Mf+6xMZYcQBkaR-XIqFPtg5n+BA^Re{;gde%C<?knA
z6R}k4_sB2KuA4FANxFmj-x>Mc{VK+;Q8(j*J-&#@**LXIFmF=40b8pdb|Oh3s;L!+
zM9N8>DbKLK-c9`p;0;BIq+c@#U&8aQxyrg<Xl>t;x%Dn|zAFpmuzmf#oNLt5NmF&^
zS)@TF$<jQ61NisO*hhTg?cV(Bq|WyvZa_jUUq=3xVzJ(gMVeGI^_FsB=E&g58VboM
z0Xs7~))dFbLR<#}unb;bX_O^0xU{|L0_w~kZ#PGxBDs#_L{ZuJ#(@;BQqC=6q+hHS
zg4L3nbE=x6Dm|bmT$FjHMopqCA0^7zf8=V|TBI%Q!aXNO6{0{s1Zu<3*4#;?EqBcJ
z2pe@MF@QDXL;2cr##yczy2G0sQ1_Hx7w?Uax29_dxSPqE9_Nx2+H`Eu+eF6c9Z%<>
z?@#`1E!*q8BM;y1`Qt+%zB10e?nY0u?^XuYN3ehMnoiAb+YB6uqo34#BpLhfiy64E
z`2#EQf(Zd7GIO(^CPQL6!T<Zl2N7mIEwMbUHrUB?3HWnZ<hsW}g?uR=u$^b)g8fsc
z$Is67Tw$QfP>ft}F{B$ER8E@^`M2eCa2@*k=9C_)hSguDkt0pwnc%mXN?!i2g10>0
z`b;~VVv>H1xdxVi0&!`ZGhjh=-C*FrQM!#`7?O|(INVRAm;k}?Q^Tt5Rp~l4X+3kX
zdy2QMKp8H}5I};vMScpYi1N>TyImo72=O*Gk6~AO^{}>A?rZ>-W@6duV#>CqxZ~rX
z_BzGKA_|)v#>Adw)k(^I3Hs2@@jbe7q)P;5m107wdj@XPEalv*Woaz_Vf#dURNrlF
z6+pWhp)^rV<|DQ2W?F`OQmI<9TfK5QPfPE3IR%4HL|=lfmdV@A($Y8E(-H%1551rh
z#WqDZ1UYzI+OJIIch6D&=>GOxDZ@GG<~Mje*VYiI)y`~>6xBH(0N)!jWN`$(7y1M~
z{8W-YQE!O#4$Dl1bzm7<jzND|l#XqXwbwS#tM-?pwXE(<2unr6ve0)2p8&k>OO=4y
z^lwX8Jn4h7h?=eKdINFR3a%CQL_H?$DCKs+2b)Uf=Ei{uI%N*Dn^b4;m6PCk_1)`B
zmf|3+*n{@gC*lTYhrn{LMfqReAOOstJf$`R%AN1vKo_`75`QoTu?9GZWl1)z@u8*t
z)O5RI0SH!p1M6_0K4uH=o(0nAEEeCo2ye_++ZLdUsafVI@N&yIeqk+k44ff0$^0lT
zS`t&z?ki8fawEZAtVC?nk_QoJTH;dXC8>3K@*jjWhoce$-b?m;{_VTn%>BDr$&$w8
z<1{>1f`($(q6A&^ao6Q_H37AGs(JHR^9AFXY>0L#fZ!&5Kz*wxGgS(;elPSf?APk4
zw&(T|cZZ0gRH~ykF`HZ|R(vxi!t>@OdH40NJ`Mc*K;Qm}f1ZTI)UD4~?E6QmytZAU
zs)iD4VAHD_g9FW;MQXT8+J}}25DCl9?7RBct|oEAUyXKH{K0R(Fj%Gt5@Dkz7cj7>
zf5CgC`HgY*F0>}Eow*Wq)5+%GX7KxfYt|w8PP8(1mTgFub$7y1D9;SSXkZUZ7U6mN
z*+YxDN=AemsgCx$0D0AK6qdzmE-FgdkBIcd6LKX%#)4S|4NT~JqJ?>zTG_c*`~lQ1
zGCvh3FbwZ5@IX}tVBy=$FMrWa{6F4HIJ8L5`&{k$df~{3j7R4z;}8iB%H`W@Zt~h%
z`c#&v(Q8K?&MMJ~IK+;%9@wB>wQz!J&VV_sduaa#b$!(Xq=f4|1g;7+U9E~mF7FtG
z2TQJ?x$34S?kTK~Sv-(Zl!#Dr7NL`*%bKM*AUT9oObV1JFx>!hgE0A~;Lc*z%<O46
zPV_;OTJcnqOaU*{qjy=dM(F%}8D0##O^-;1bKW9eRpw;I%rVm{sa$$=m<{-+fNWC?
z6X5!mWZ?B66oRo?3aEAB5#H?<800CYc2m{k+f&9$Ov3KkTOBrdQ~h`Lt1L4dYLfZ0
z8H2Y{ZFUviO)~hh=ydix)p`EBZuTifPrvz#LpK#p{;JjEW;o5TNKTXF)a=zM^M(5j
zL^E{6JL-L;&dO{RS(oYwz_>uw;0X8v-YCX5XLg+74YdQ%@d=pA5yF2!Si1_<V07xX
zArPB9?-${~7v<V2^fD)^-m@+MOXmcLP#Yx{fEjd7+vCrK=HA5{ZjvX4lXS}(dHdL>
z7{#f~@Hr~vEyxlHqXi0qcsM?bbo84B6U=X1zsv5K%_<2|`Xf+{v!IuT22Y{7@ZIsY
zmqXycqKC>ufeiuQHk;W2Qs?TK&q(t*S%?gJGDyyxz0NUE@&&{5cMRj!2$jUgFVVNk
zx4zIXzb-!!r?Jzq9PrY8#i=D*ul5ooBInL{KF~P>*nYUT1^c@X$g(A`J|buHG$^``
za9r~4EX4fS2{}C+vXT$+H6cg1UE)`a;!KS9OT($o<%)WrnIWigS^M1?eccXbqq8MX
zOnyY)m!#eSezY70;t?Z@&u>yM(wRwp;|flQJu(?QX@i@9-f}MCK>ToJxBqONFda}G
z_E^Zzo#q*b^)m}Ir)7C_(@woUq{W8jwzhITp-!=UN`$~P^6BBHh$U=O&0DxxQb9(-
zt@?Okb=LFI$_ZT|k&~p(BA!QnuAg3Iie{YIN_PU0Z$HS(ns79}gI!<pG@N%JD(7pw
z7ZL3ji#v$=tb4zy>;;%n{`-vxO*b69dFc(_pLT=x2D4H<nc5|ZD*^nDNbKqR)w`xr
za~N*C0Ip24;j8>M)5EHY{-^;g+B%qURcp;3URP@tm+TjJqq78kZ7pZ?kwzrRsIem3
z$ywd}Ro2S60fe~+^6&9}J)5k@7O&^8(jjqd+v?(*ZOc0)2`TZitT@J_SKmPlbBE%5
zn=?E%!!bp46texTdod2HczrE~t@GI8YfxAmk+kNFwYB7;M6kAL)|8@oP8&h7ePwU?
zafn@uWRU6RF^@^%DBaIUHQ`>jgnpyRHFIW1(t0o`GhSwiFKNP3Yxk&Z_?;VQa0GxR
z5>%2A3hIA=vECvn2JjSXJc6tbK#@VN-DWxE`xhhyxP{`qjJ@RaI4cDS?YTj#dJftN
ztUNBAs&h46r`_M2R_J+@Ic-YvO?AfA0eudX=dF-izhWNSUolUoJ^#M_K&wj?Qqs$+
zKNUezOc9CT9!N2@wZ}bO!2S5MU2+!N{OC0k*P&k&t%bH5+v^KaLP`Ov>YGqHXs6?a
zV~@2MY0Fg)FtCjZg&wdvMe)ifQ4XCRTKgB9G|=mw2l<A!%kJtP;}o1U-$t*OM{~k-
zRM^;!1NhB;h?uU#ut?WpTCAXhJD%<csM+M~8rT6hsG9>hWM&mqWP!P*z(d+guy(k5
z`zQE?Or|HTcyOMm!Xi8Xt4_t^F*>vQR21`J;DT=MCdM^a1ypMPUxIoib2kvEPxN!!
znDYV6!RltFZZf#ja@b-93S&i1uE!RI?1PYsTDk7bpR%28tD<xjeJh%Snbhpj$<23t
z=KZCktcu>L;Zmus%LvzKN;os&iE?s^=t1R^o>R;+M5srF=wTnkcD>Ws-!5bmVRUV-
zc#*z@m~Yz08{z98IlUZ$SXVx7;@4OQE7>C@KgZ77_1~~9gRcDD0&>cTgWB9cos(Ng
z`mr;mSR6i{AJxti&}2c&RO5S|o{(Xk`{L*5)AjIid%4~#QP!yqSM9a5z0FI1dQw#c
zMPa9Qfo~#`V3DnQg7VLkj|5SbbtRRyWnHZSw5?gk+sB~K(Tn;HGup>}{JzHZO+qc&
z_ZzN2vLgS<fEpx%I{l4PV5*hXPjGA~2QOi*BL3(p#)fpk8g_+u5>_hBBt%{iQ@6!d
za9~eW7B8cS4=qIDoM+~gXij)uA>^(Ma)%a7uK<Vc+_9Nj1W?GxuA$2&q}B3vIM(n4
zDm;et=qa;IXyE%Mq<M?|`1d9L*o}g6u>>rIcSU^3j=iO80+pA~K}NJNwGOX=`G;NQ
zhcC{yS4HTMiFv`dxzbT;zwCoS3e_2!HVsuO_b@Z8Q4K%j3{n_)+!^hZ)ynC5i}Q*c
zjG)=)E^7j&V(`!F=UNkfLX$5-NA8qEPx!z~-_AaW;))K2T~(#!4_mspqcCggh_W|@
zNhUNNae&1!l&PtPfC$S_gPiGCzJk2vy4_iqh1C*72SU3Gu|)x)y*6z@w3bP!?w^+L
zbbP-?t$HJ6`}rEAU25AHFtoQ?Vy42(X!QF(boTCZvFU*mP80ShL2-Sp2=U&l>q!KA
z2EH`U5TO>6Sg;SASp^AOr0HlmUB_KP#aDBy??unHJE(nRs4<q1HtMv+b1G>+eRZ+A
zHPYLnG{Ux*W$Fe0qYXM$>UW4lxrtx05Bxr;ZO3Y%gcnbR-aQ{7RhcCJbNbXR!R3@y
z5<vr&Xh{aCQiy{yyQf2mMJogkDI4*nUO7D7o5ZZVHiv?vUE*sSK<Bt-TWD#PfHbqd
zCU9(rvsz1BP$k-m5}7&%La}$kxD9-XeQmY!TXI2zTVdB&TyxZ%&isY89#WDd4Wb7n
zV`K((zkhSImF`+@ku%#Omqyn40o8K-jn6WS`tdoz#q}EHM~pa)w2&aA$~blTaC6bp
zW1~8=z%Es-psO7->1w$via`mbndh#k+xnJVNYk`0gN%Tdn297LUnxfxT!@Y}yd&k&
z(S+Co9eycIh@+9@Oj&Bd<2W`r2~y?(e_%S{{GAJXbY69pp~s!h8)3of6&gWE!NbOA
z9K50on{?Vu#M`a%Zi(I7Y3$#)nm+ZlbrRvSdl9$Ib@2Wo$9_^hbx==mEh-98Fs4J1
zT94eH1`15`_o$b>?Tx)a!Hd%2N@u8{aBY(s`H|ye8_PKp%jHBpe%$o1Tt?w5FSR)X
zuq~I;7k)kCq0vucg1{%nz{uGXCYSw6K4%mKzE5sc{gn10@5nCZxF5)$hg>KE3(g^m
z2yj7snhX0)Xm=~%k^53&)~z@lMj}KJHOWRb?%(VN3@OGwxq@|o<_m7Q1-U>Y6}*j&
z-K0sS+LGk_Ov~il=CL%-Tn5yF9pBraagXf;T5cI(!5K{G_|@$rS1g59E+9h9a6Nnx
za6!|l2J3{(Hncl;Ja<puR@t%<vSDbpJ&Z%aVES;!d?9)H^{^)q*E_krBNM(vPesRS
zP#;?-9)}k9wL40cDkH_>Gm6oJfGmvpCDfG|O8JLqSxLx(Odq#9XU>9wNsb9nvNkg(
zu}@tZPl6(ZuU0YoL-|>igjfyhPBEIw%lw@oeGUgGwHNU_^Wb`x-K;{fq}bH~QspPc
z(2Y+4taN)fwd6feAcU{f76zXaeDu+`9!Hh|gOPH>^XA|ppCPIKJa#2>_e$1#|IAjO
zDjw%HncEZNf-h$1nw>BBU0_Q=V88OITQg%iF!+Wq7$K13q2aj7KsB?Vz@f>Syb77k
z`u!AE-09UDO6{S4U120{bsW)CCP6*zqDhfS%8wNJP|M!*D=k5FEF-jc9n-CJ)|Q{f
z=)X9R^S}92y{=y(kjs0yHAg-;1*uCJhnC+58Yt-Bt|Ra}vY=cZD?~K8bg>T*t9W#1
z>j-uC1N5gt6?9ZFLG$}YhF0o1OVYo2-*-`UFt=}iESAtk)n^3b!5pYkw`j{k+Bf~<
z=J|}jE1{3IT;yM6IzEW<bPvuKVk>O}Uc94TTe6MNYP;f|`Q>(Pd^Z+F-LaHZ(BdeM
zIh1n03nMxpwo7`M=v@$*yk56?U|P(!^2(nKeE&j(|D8h3Uv{sf>vD_NLuHlev;+A{
zF$c4zKf@;3@VZN$DT+JY!tb!5BxQ^6W%M*NDXB7z-%NlTk~E9%^;r~0!L@JKd()R`
z4Z=roOY*m^ks%3NBMO9-`2m3fwnDG|E_Uz)u6t3T;u`sqer46aJjcg`N5V)k@)-Yg
zv=8Tt)c|!aPj3o}Blf@+d~eRkQd*%xBKFv06HL=KNDfYX`bEiz##1CeZ~+=VG&1R4
zU6?=mB%Lb>M9cGmtP21X7Ow0{uU{EB#oKFmnc?)wj4N+akrrAonKJ3AqtR4Gt!i#d
z*bughf=Ber7}5e}0A7-;Mt_gd!|SB$awKFV$K(O{;nK26QFBhv!3GZ{rR2dMqB37r
zps1OdTS<?7dq-J(W{f-{>$9v4yA)rYT=RX)E_V&XADi;mb#GP35mL$Z1x02)iOsZ$
zo$kd2r|{^a2TmTq(56$QO#iIPi?m+}PvCO;aEFFJvB}^usS`*yRx2GK(Kf5eRTcNk
zv_#)tg7D)xzFm54k!Sw$>(3E2b=wE%pAR>-(wDq(TF%1Bku#h@bDoYqi#fwa$d6j#
zIoW8<*j86qj_t(pL0Hn6MPEW;EHkGpwe2i3#F;DysULnEACf7b%Dda-v3K52Cu&og
zkl?RzmMc~C$Acy|!rwQ0`dw0T^>A>VqHT7^$A~sJr0#3Fu);VDW6qqoxR7MF#<pcj
zJV-ec3?>4Fj<m>!-YhMsFZ78=^8_XWXJK;*$MTx!g1(l;X0RDfRTjcxG9V!@9Xorb
zZZ_31x>Uz1LiRJ4jn*eP6;ptOCP{w^P(3dlP<w=-PsPSw+XaW0xV+<-=_g+0*fh-(
z<I3CNh8L-b1^W#ua$*X)1Ipb^{?i@$Gm&xY;Uj0C^aJuR&A4Nuj?X<Uy#cjRB4Veu
z=~r20nNy2*dq^<UE<1S4eQ)>sd;|wJL+Dj|{$mIT{+ZKvw#?KQqW_{UA(X*{h?0Y`
zR<tl7;}2^$C(-%7vb=oIJ%??xUMXw(HU8^ro@s6)?;7x<<EKoX0{LT-jHW@b;aLM0
z-fL|^C%g|b<9Ss}RPINU;%Y56$H@zlje^$@V};#4ty4cHb)JPes@+=ecw79es*(I^
z%{<H<isGfwGMM)$E#^WsD5rj;{K36`si$8gyEv<N7N{^QI3BtBJg&LEvS=XCnL>GG
z=`fTTP~E>Tzt~IF8kMj?3{*)#^%Vy@HlX`VI4!D@9@AqxNgDI#Lk7kAzal2NoVqv7
zPW&EOkm~AclHZ}_FMpo}O)8g2w4t)p9tfjMEkuMCS_BXvCEZ%TG?0ZJ;xv=z*IK_Z
zN!H$)7Bt~J%WB$y=6XDu>$>Cwg#38FPlAs_cb+X=o-d@FU0&ZT(zjgJst7L*-q%3B
z7?u|JR`aT0bN*KY&gfc%AsI<?dqijWBjrc#^{Z(v^FLOHHBPTQ%Y6CckL3H!*E02=
zT2nhtvUazXSh+>-g%_WBkHM~l$6<=<9$1rFPO|nlV8VyjKYT6z;f7EltWis$CLXQ*
z-shvvoR=dOU%ctEd(OJy1{0UyZG4X;ai&MozSTOg_2dCnTI0;Cl-AUBd&QE?PMK&*
zYp$Z5Ax}!rm1ft>j$79EW1eseH{B-kGe80Sy_<L2%fzTrxXNQwzj$pxjOmTe<*gF-
z#*3<5`3i$!*sP{q>)gn9xn3}T=h%7vk6Lf(DK<+iNx$1iDAEjvZWNydu8=kea}rJB
zf83uS)2`nV(Ep2Ec|6&H43d(v;(O)YuhPs^(WNbwn0tD_fp_Q15}l&q$=EIFoX12s
z8r3ZVEEZeK^_vG2>Nz`um)B8^T;gNM%yxZ<mwov}z22+^X)g%5@z|#GN5DUK@!4r7
zGpX1j?1n}(j?U!Fv(#gW*uXCI`Q*rZXsgT+N^1tWEr)b})pyEK$#$L+$5N$`&YiWb
zdhb>`h+Zn7Hl833)b$uO9O>9j5o%rCAEJ}WKY@4*&J6XO$_x^=E^}1+hPk!fof)KM
z%}^9NZ=Y&%4o1`z51!3Y9lQNkSow&Am6?RpZR6zbuXwi`TQ?=C(h)m$7t8dJ?w0#p
z2HI@rThy=0mT)myk*y+Q5Ue>ULrw1gICwTyepX|Edma52@s!8Sa)OP^8t^KvcTxBj
zXE8LQMGc`fTmZ0YH_k`JVhq}h`k6uWpv{kPVaz;O3U*u+dqc$CsA4Xe9hZ5+MTU<t
zu16*qw__}z%f;J4e)Y`l<XADT(cfo3HSwMhu~W=CarkN%SvW3fraW<&S=nb?;Ispv
zc|uuw1RX>(FmWaB*18_K(M<J233oEQa;0LwyrDBOs*KyM#59#ijQw`{MDbOT+p3c+
zB7fBR7auwL4Eu^x^(Lz2sv?a<;17!v6x0-=e;gBiEzkNu`@yG#_C1%OA)@(tV6v|B
z8AO#eFNaUr3lUZwBCVS)bm|bwx=MuzCB2)}%`HN@miKAJw%y%szS^5Rx9`#GB^Z?z
zY#Ty;K=9KdA5h4(R&p!IC|?0$!T7Nwzv3SMCH`KfJk8?|E=AY#e{nAgbi316S4Hlw
z)qF8moqw&K;6(G{8%d=4r9;p3e^kFdGTCw_-c%Hu*eG7|_B+d$W&utOA2%2FLvKvk
za?M8EG3;BqH$2r{kAR6yF;%&xX(##=(QL`!|6%VfgW}w_uF>GG9V|EmNpN=>5;Rx{
z2@>2xXe_w91$UPKfg}(-jk`<mpb1WJcfX6h&pG?l`Obacs{QBQA72%z>T1$-_w%eZ
z*PLUHF^8&lQBvigm}5RcyIq4IYQln6XH2K7xodH+dr2C`_OA8xL_DQqQ&<y-$u01;
zlSUNR!XkHoZefun9V<+P{hNOwHXDMJb0ybdx>d=VXZDfMg>ZNlPyVln+`%fqj66sD
zq-=bAmNXH(n;#XAqmUx*{Oh_8rg<jz`eL*F<LKoY+TdePf!1`VjgkTjUT<?nbKawg
z&{r9rPQUTxPq!(IIdeZY4LtdEH#4MbrF7U&GPcBHg+<B!GpW@2LpN9#%Pz%|E$ar`
zJJYTTQva)U74O%}x;y(|*-duQF3HbYh}a1)WwNGIy?5gc3RRiy1G5Joz>@CkD~bYn
z0{lt)QNsm0;+>i2A3qI*YlY1v6B6s<fB3Hbn*|>!3o?wHmbyBN*CH0nj=;jWg?lr(
z-LyRN)#C6p!-Rp-2t2SyuXyfXf9l^K5TRphcxRv_ZuR>)3KL%$Rhj@da_p^m8i6Zh
z&Z|~BlYqkzg{kkjmSOMydfAW)BH`I_T=CU^x&`|3BbVygf!h%rr!3r~;bY7YXXSKI
z#9vC6A-N1{uge7h7F%dqowwk5%@YgFh{_cFS0}Y+_@p~G!Z~Fi6%{c!_WtEKfdgg_
zDfg*2Fy!PaSkmo*Q!L=_c$w=u+h6irt=Won;1hc`3+yQW@go62z_?1aT?_u_6aV}T
zHYEaTP!|vc%{)6C1EGq;<G4`Yiu@iNqGu8#?LnR@^gEgU3UH53wx?!h^ZsyC{pUUS
z*A>u(kOtgbtR(4E-DR75VCq^B7<9;am**G)m}*BDx3*Lq5!QDJ6p%r{d56nm+fx59
zFy)TKtO<aC4fhZa!hhXDaMd2}St7~)AJ-x~3;E`#02=r)@%7`YRVc7lyK|3S=+nP>
z=Q1-hOKg^6nwT#O^gEjipx25C;;q3Q9k~@?1rOjLXxjqQSpnSw>VJLie}C`;9s8O0
zH6Vs#fXQIEpdj+r1WlMWh9n)r*@MKYmtpBU>`L=M0f(LiK13U6nNlX|xvl$D+ki;H
zvK?jpm$Rm?9){ff1dGrwroYzfpKry7Z%~*4n(*}XfDdB`L_8;KE`V-GuAgw5FVN6a
zwPR{YqyBylaC8^a=C1Pf>snCB&o-`sfox&W^KPvFT<Ujk>C5XYxn2T9n1_a-5?*kh
zyVR=&1#es3-1UwF=dmX``cELb0Ulyb=y#P_GPmioWFXN9n*Zz;@YiCfmy|$k63|>r
z`|GJcR}Qui0(^SP`}VpGNLN;gFG_z0XWIbDpOb=D$dw*&e%S*$v;e?8{iX5VxF*Wi
z)T=M_pKI%{<>!MwreZJzX7C)gxls^buLLDFgUMHvX8p+oiD07o2hY+>#=oxBt*ith
zlmX>Gzb&a5qI_%spi?pb1W4`zFzrJx8d%-^ZsTRI3EUMQc*AbXTQOO%?cW@`{O4=I
zl_Dia9@D2L-_<I+|JPEJWTNq%%Wk3#RQX#*qXvqE(akoBjZV_9|K96|z)BHVvyof)
zdq$<CabT;E$Cc>D{AcL-*H`=>ACNkL$c1x3{^IPv#-0E5$<Bu$taL70YUKSrpIX!m
zA7n58o1pvuM-=~SH~kM^z?VxNEL*PgX(EgNa>M`kKjHiT{p}xnym#3Jy&(W)|1FMN
zI?>;|5C{aH5}LswKyl}Z`ue)))x0wh5d{M`(GTPIdU^iV#Q{6U0THmt%+D4{nlvpr
zEs4(9>bA-kz`6W|&*wn#Z!LHUWYmmifca5V0fqQ$f33piT_87gasf&!OPp-czqbqU
z!F}ZHqOcDD4xPE<2IG(S3i(X__6O|byj#<IeHvqbb4@4Rt?5ha``mx)viou=-mU4Y
z10u)2_xctF5Ypxw$pio19Wua%fRKjW<RPv5x9+fk?OjN#-j$*KdpDD`?JlIXap<d(
z{0}?-|NU2_I(H#$r%tu|e~n)M^*{XI+y8%W|NqZ+zZ}E`$WNZcK3w=}<7s(O+6d}8
zY;<2xKg=xvR3FR6wNLvl$rYyg@PhNdF6#ey^5X8;^ctaj-UOUa*7rbX)*1U(+U!w&
z>0T?AfbJ<!PV@bdU>o1X|GVTWbzt*fr9*Q!!=3&-==OTo+Ylgh^a6yK=MoU$SwPYX
z1-t<nQVYwG-|?G~&A|JD$Qn@fOF_ph_?5cW$GS9!1oUujL}0>c`+;RR6%+_<fUYXo
zYxHUD-?PR4@>qh$w-M{AlzSiOQfv8=f6Lz9>H`=^BN<3J50&{MxiHS|hH>tTT$_Rq
z=M<v_WkS7cwCn*JT0Sa(tR7xSJz0|YjWzKOFVglqNSGr+`*N!BqrmhHQ;&_oL4RCf
z=AQuBv0Cym#b;Ew17KxnZZ79;gF+!pL7)frbEyiX;{jj-e6gLL{-2k0gUM%d1DuF&
zzsRnFqn8u3P58vH_nqz=*!vyoFfk#Tq8ku}U)%tq8i##aF_CXEhK}3%#|6M6Iz4!Q
z@_W$hgeT#Z?kE;;M^#1#jYAhfU>DX(AWNaTI52Ft4wUZ@@O3gd`&cW&BheE=0;iE|
znx{bvwhh3J_<F!(ycC#;2ZC{<OA&Dw|CU7mdu{Wt$GOoGb0}VSLX$RN<M=C?S!ORd
zc^hQV6-z%&;*?qeAU{pK4eT)PdmX0Do$)COACgq%J)v>gD(gCN{Bid+sPC915|egW
z7h?sj#M-chxr*Cl7a<^fYy;Qd5_)`6)%2_FG}5`g&AVW!72zt~B0ohZq=g53FWjc$
ztlM#5*t1svy?%6OCKwn+JAlq}*Ypr5HtZ8;G@*7|`(O(uqa!1G&4H)MSC6ItIEJ#l
z5J8vRLvY>)2{9|aZOmH{=Dp`n4Rf1}f239qD~QaCG;O^EI!DV|8K_h?1OfhAn)!8Q
zY0l`iFb84=`kjaRA2n4KllOj@egIRJM|sKineqae#p=2O&0el|$ExPLcqK4tw^W-{
z*C>>s^Tn80@msw5{ocEF8(NW;9`JoJ1e4eAB!%uvd5!<I&iQNQHZwmQa`v4A0`{#q
zm7?M<U9cDpfya>H=Z}d8bHMuRZ8C7P@aDJ(HrN3y;)Bt%;5W9Qw{|=if^tVc7#ft%
z1b9;m)9_h49pQ|j;S?^5fL0eV984>&#5%>lytCd_T2SZ;?6GI0pG@7lD~JHC(sdG4
zM-_VGH|}=~nNr~PDmG!$3RbgaFrW!O`U$$|H52t<^PcwH$GCd8zm|Ph^Nq>e>ni~q
z7lUlvyU8B9)pwfQ06>;5UMIVg!cLT-S7>NkeJcfZbj`<t{IH{^$in!OZwtS+0^FXt
zObSg|ST~~?m|?7jc|6=>>59E+)o46OC<R2aU?5no`T3pxYD1RFwf!;%dbNPp20Dy9
zt;TBNXK(Q;xsA(cT2(NvUQexs{AU6B*98S-W%C*xxTgBVB;y<SHHMxaNP@p%?s$Xb
z(Y>Z&)g9!Nk|aBKGxNoLeFE!%Jg}+*T7+>BY#ZY~Df6Y<@S90)1V1MFcIiL>=@OWR
z?1U98eL7}GVTlH#tY3W3^4l#~&aSM37z<tQJ$yc)nS&$)_dR%9i8=JEc{HdkRyjz3
z70wQlntql`b3krGdi7lvP)IsBs^|SYvBE^}iZI_VAb%-*X1nz(3gw`KL|-)gm9M_X
z(px+!LRz``Q&cR^f6t3ZXOKC^n+K@3`e~RD@ks-CTX4+)86lONBDAi2eLJL?20yz#
zT!a?qiOc<<1tv~()8Bh)&(T5fkwNt3V%W=Vl_Bk>CPuN)c?V0>GQ+rw*9IcLsvT`l
zR-}FBK>v2LRRAtizm^l0THLkx9fbPbMZqGbGRNmJ%~7j6z?h1ixSfu6hRti&ea?O}
zVAo?+q#}xXWDZ6o3bePjpx+e)unw`)DGOg}90z2EJFhyMb=jvNberSZxX;YIMrN+b
zT>oPcg-B>1Ptx6h=8c0L(o7M3YL{UHsw6+L)gbVy-YbjpEc)s_rg7({j8|&#E#VzG
z);A!I&+;CeRrCZ9)6M9n;(%#ME{~$8?5`kto$LFg>YoQOsP|Pk?k8Io*C2WeZg4L?
z`xQj$HS4;H19o&(R^N?3W$QCx?|OfSG{4I?`bRVSig2rC&x6cTm?)?TT<Xj+U4QSp
z5b9G{QA`VS{jv?Wg#@4N2zH)3*Dp~jP1uQUm+y*ZDpV|!t}ov4i3_gd_Exx*DwuTk
zx?Xc{hvp!>z48KI7EIzX9$^#>v9F|xpr(l0t$xJGcT4kAG_5*12%UO2fHRu${i3Qp
zC!z&UkbfyYZfi5Oh0k0b$s}=2H$OZj6m13B%4~6@Cq~9aV3{wIXlAA!?P^VCenKI2
z?S66jY7col#@<7!vi`iUSPBPJ>tRQ!PI+y~s6r}@bR`4@yivf^wP|Ix66z6U`r^Wc
zzvScKKBdA!08UqhD{dAAJvJAtUhTjHJ9lo+`*!49;J5Zp#CaGm)h6Buq2hFnUY}xh
z%a&?|(Vc?Yh1wEy*(_b}TZB7v6aTdRhGsxMhc;(i(|5XN)Wms{|C4_3wCy8`SJ@U*
zMC`b6UH)iA&r(A;fyt8m6*E(hrN`pP`4HYDv{gp<d!)!wDm16*;!juwz$~AtCQ*Xm
z{E==>pQfRrX-)=Pw=W8YQS+5^m4=k4KWu5|uzP|-GxehItPCtnw~W{0O~Y<t>j-^o
zSgfTIJITu7&akF%ashU#2PcDf(RX${3gLsw@Zf&lS^C1wmh7}Rkn1-z_0|tOuLUHK
zT8r=qb3A7%QyAk(u7&xictKIlpj=JQWVfbmTLR+<jEnyr;I3VY(u?Hwgi0hzR5@za
zb3`_bFn$;?z^)57W(|X*IMOZYVX;+4Ct;0=Ra#BgNEXRjBoRcdg&^4(Dw<S^OpA{d
zSaf;Fl~|1|Jmw)gOy@F|QFE_)CORT|1WF|X2SQRc#HKqVlsi-x<L-g3To-@P{ToF&
z!|ywzSvQq7mfhJ><6vK^cpX4U8V$wP0p4g==(Yb^is;ODZ&#z~;?>!BRpXDBWu2R6
zdH(yer;P$zW^_$z%0eAq*{n)iP}v90w#=bvjSg~6Jen22n=AmhEI{u7TK06?aKt7X
zk0o*{ry_!Pc!_+Z4{5Bk&f1v}pU5!^E&rLYe1T+-mIIp>7xTDxNx00BzJ#o;)?g|W
zd+H-MLQ_%41u*%IzzZY@?p?mhY^?_!{qI<2Zs?v5D&s##d$Bvb@Ow9eJ`!DlwM=S>
z3;VZSI4(bux!^{-SIy&NAwwf~*J&g{DMjk*HCfL(*)yXLB$l;`sWfk<0DHibT(O<Q
zLiW<v4X<Sd90}`4NM{|{t$=V&4ou-?=af?hG@B@P<J^)^5TTb-gB}uDoQEFMe6Trr
z?Z7kcXn9XjFo>D)14<Y+m@YV;O!FQo>kdKI_NBA>#Nn}kH5>VUH{F~@G>|NEZ@*p;
zp~I$_Hsa2<MoE)&j7h1DN}(0K4Ig17|L%nRrH>j?LHAg=Nz~s#)UXw~>YXcqhuyVz
z7lNrzI5|H_PBZE#x*YYz+}M<UtpqH>&OLxs@QdD@EzQp6>dK4a&)!|8)2s0k&Up^k
zBS4(OtkeR$a)c`82%1&hUALqpvb&jJho+oi+f9cQ{w8^p2V7oz;-sWG>Df4P({87A
zbGm}yK&%F5<h2*7y*x)e4;YG=j<-=xbOfkqKb*p5^+u)iJlw=&GH*b(O)aNZuYua{
zGPrEd+Cy93n)(x<;I6ANUbxryi2L@$!5Bkm)}9ZOYhtD3>$^Pjk0J1;D6YBPWemFj
zz?~3Osy3LkPJ7+UFulI4auW%?Ig`**f(9P7P`k!DKTF{Py_UVq)A|D1k)@x&oM^?6
z0T%%DiAf3Zd;-55;_O{PCV15lD}d4GW)@NZhP(J#ifbzd6Y>EM(&xS3LIHf6a&cqQ
zuAVhp-{`MTL13A80{SV}nkTiIZ$E!*1vvD0K%9f}gO(WeN0;IUF;Tp_Y7<e+dv)`E
zSoQU^gVzgq-g~t{?~=;j3qPDGmWMV?@S|YeGLwK+C#p!Ps24_eqJMia;fr8&0GZ$_
z>o!>gH*6;^lZ)cjf$Y)KtUq&IWB+@OsKz~&B?O;%P@}(wT}t}ZhIbI(=_KyZaoxSa
zJYny#iZIr4xjRIBM%;vBAxB?2E6@oo5-#Y8GyTBWaIsd~>_}_AJ!cR!0m}FIrxl^s
z>nd#^+ZX0sI?lRUW@;EDm}4>~>rV)Gnuw}Y5{auPoGEzPv6pNdxnIVcDS+-8U%n5H
zSS&!?3LUg*36O@e#^yFXn@cGBQ8pjMqLeo^l2vf~RQN*^I}@T7;;%8oW#ivtWF&6D
z_tFFV93+}p)E7g+h!#cKT}B$6H&YDJ=Qh2p7rP-UWJtz(BJ7iWhxYiJ!+_xb69@Wy
zfOf|ihR7H`7wOc+9uV#=BYQrWj=facPz@rMd|1|bzqzOw;m()+HHMv9SrD^K2tdq&
zGR`$#{@Ue7_5{qwQzE06_Y`d_Hg3Vb!~&WMr#@{(tgGbk+=?1N2M#Rp+_uzP;Aw76
zY~bbGX*ku(I<jWkGhQ?Xr}|MS6n^;8*Y6nmctP8`#&nV~Klz^7^Pm>GMfK%36r2`M
z+E(uhpB~A`VR@sG*d8766jqbPhK#`E9P_p6KQccF7KC=3_+bZid{55mh)?f(yijX$
z*Gi5D=35Cj*f&H`JwH2Z4hP6B2{fH6Nj!gA^SxQ&^-$3B;1;)bTAeQY5l4n61nuiZ
z{H+`}Yc!{QuRsC;$#6maW*35hKl%KR<lkX|8m6vkh(cm<z2n?;D9OyaEh8&Rz(l)I
zFq=>1J$E^~O8G|E1p6WP!V^J)@{Z6O7jNAx^uR>}!G#sik(gD_y-fV&=LJ07D&&C(
zhwGyS0vP@HIKyQ5l>Se3f}?{XA3h1B>k`SY=zoZ}O@*vd8HKKLf0yf%sOy0V9v&T<
zmp5>vv^BhMmA58W=U^nSft}YJ=YaimgbGOvsLyPhTRFE&j@~|Vxsx8SzcV$$7$&mo
z>LpQpqt7+}v@XEoY37HDS-fBa+`caCJQ^3eq5zD{b|iuxP)9syJj~cVvf{?<RJ;%N
zDpjVw#%l|P!A*1}i=^SsaAHC^@{rT=$QJey3|A_GY0$l)FiJ}~@+&1OqdI>Bn7}q<
zXLP%5>E%IJlG-t+Pa0{z+Co=IcH`xlXT8<$t1~E>EUB#sY<nY1#aHanDG@kF#to9b
zd|2{4(^hwy*3amPZ>5Ts?|iIFc{W;cp6(9)F0J5iT_T}gLSklV*&6BERm%H9jCzko
z<&x?JrZ*MiW0>Al(1)1jB=Y(s#~mP)akNSXRO-myFccppFdq^Om5H(L#yLoe;!njk
z8Te|Za+=PW|KNtGJ~qX*IE)n8v9+UiW!)@o_c84*8`&j|Jh|8ZnMZD7*?ffUTx=WP
zJMP97eLlIH7U#4rW_PVAHq(`Hfi{co?KL<wcUEX!1M8ogysDbu@3=dDsYS9q%&~oh
zlP;(h^zYv%yan>z@jhdhGC{wKh){pQ&zgMRNB)IZ|44Rp5GzmDM~6;~2r#$b5<4Ev
z4CkBB3Rpeu#9XT?^9grLwyzw01Bw5ZGy3RfCExnsA~e#8%`;W``iJsOWd7k%ljc}6
zG88tgv^J{zp+ru97xYf8H?;9cAQyp0<m=jtXG&)>b(^x&`A{jhkR_3)9noGKeq=vs
zmQ_cDJhi5m<K^Qu;xkw$jnhNR!^Qjlk?)X^8>Xjl2dfAX$U)ZNxVgXl1vA_t_mwVY
z!z+x+;u~G7A_FnU{KKY<G<p#)R>SElTr%yT#F?uR0@Al=WmKk2l8fz=S`>YEswxlI
z!yxxcFS=@za9-igibuCI5lNkvhvLD7{dv|oMR}fKg%6_YtJjl3ea1|001<h*fvs;8
zXgN4o*|OLE_;cn?Rp^KIqPB<Xa`2xvNIRj9G;w&|XKie&f(CK=TO!Gm%jsEjrb)Vl
zX{<{9k0WjyL9Gf&vsDJ>B)qqtz#oNiJalh&%6t5S9PAxOSx;~~BlUv<s~6v|OXbTu
z%eF{bt<_)E-YjXErfVJVaZXAZ&&<GHaFof%whPaPD)lmrOTMk0eV`Gl)hFNYCu3BI
zLTLGkFJl7l4}jfw79;rwGl%|HwbaS+dOeQU!TDc<83z>ld-aXf>j~5F<rGUt92y)W
zArJ`$I3l#nx44}#J#FByEx=NbBo)7KMMqS^MRESV_sL4dGtW$%XpEPH6+3asg6!{N
zPiFo(3xI3?{e%|0Axqy3auTX^#iRu{`^Z)%+1g$t$6zMLwP|Y9n$^zwje+r@f+A~H
z^7H(a)GuhTwC|mlyk$DEUw4%P;*!ysxIRQSRam!4xLE8ie@i%Mg(b$;v)$F$B%Pyh
z%)QcPmQ6v~v|h&0bg}-?NQbw6g5CmVj~<ux<Q}beI=euuD>f1)jUqbj8uzd(U<=#i
zkdM3>g_mgCOpv2*E|y4JJwSP}Wyt$Mz3;phcQX5>(UYWhsaSbE?E%((21fPcgSLPo
zwy2N5YF;b!D@;XEFg;nbbp63V+K=7<_Z|*LZY$wIyn}$yc|wRiBV)CNu!^wQQaDdN
zML7b`s-aP1zU9U<h*IXsS({W*rnQ{^l&I9}Umz<W)SgP~OO;KPm|S|ZD8!*(VaMmI
zZ{!~vsx`<?^dwF3_-O3AlG)cEZ4yvZ;IE^|+*BTUgJVjwLTfl6Jbk^)aUmWb3TOVZ
zPDmbd;2=%?z5ONIv4XB?e(0ze1tZMT?MCy4OJ^G)`7V<SRceQKsCW+}TU(WUeizFb
z4S#g1_;P5O==kBPMO^ap`gNTP*&qCYIO$Bwb{lPU=IU4VyXhM26A94wnO~RwOfhnk
z?s-yc>+n4!;4Q14a6Tdd-h{uNM_eXumzcE)J@B3}Lb5S;c$-Av+$p@KF}lM6wlyF9
zMZ!^CXKf+9)r?<Z<o2PLeK9@NtJZE8L{FHuM_jZ(`L#>u!;wR~SDAc-JCpzqAkAPC
zR)%rS%1PvN;&S;~Z*q3hS=o6S)m1!%m$r22DWz+CxiJ6*48%Bfs!V%?>Dmnv{Edws
z73^X?A1)DX2d;J-n>{KGvwB{RBmd`SgvDR>yTj5V0eltCj;Lu(`a)KkM+dRVLCT_=
zUT>a*|LA+IjFI~KND^P8If|E=SSyPMaxk`0^0%<|qF21ALem&m$>~NpE%&EzZqJR>
zZFTZg$3A~VGAiWpG>Y#Tw!SHcR4C!~pO*nTi96Gry3Fr%&!y+-YsN+upVZn-;98e*
zipQDB39$m!;{tzf!Dq<$p~*y~Bt6eNQ-4&pmJ5n^v#wRU46=52K4r@q<3ay`f1V;;
zka*}%x2J$s1oO;nOxneU(fb)k-z!sMl1*!2MHcej*rp3U3r1kQ5(u-9PCEAZ$x7(1
zjW{e3^YmWtJhD~E<oKgneyUwl&O|uUI)02)W(Av{SdnWEMQez}m$-WYDCq4S@WXER
z0IbJCHi<;`gY>mh$WN|<dSkNinLB&fD<ao+bAE{w67px76C--!Hy14vxk<k+wTjjC
zNoojbVQ)`FN3JmHJNkzHf|OSvjCQ?^N`FUsK8cc5LH|vmg6=(q>Hhb#U<0hL9$#df
zsfyE><msk4qPyXD#xx%2+&`n`e*9KD&YQIQ(NZt=k(4j>ZeB)rk9vKr&mdgQ3IlW7
z>tGNV6%M^)r=rnSB^mCLpP+V3Od&jzp0@G8)o}O0$9~@&RnDwZ{fv~rQEk8^@FKm>
zzM=9_a&-1Dlf-cQ86(Ae$<h<;MeCIz`<wmLuKUAWf8K&&B?D$~xA_FP`hK?IRwmtS
zl%mx}A!L_WB#t)VC_}%H2jly|#jp<0TIA<XZ|7Ry$j`J<;t^Af+%y5`43*ZE1siM(
zEA;r!ZEGUtC@|Aayo$!v<2fz!j0Ez>0SE9=)mFN3C+ndNRg9}y7j4r{g}G|Wy~c<{
z7i=o6KsjU?JYV0RIV3!b{PLDo`|81u%=-&v)INUIKo+R6qI@-{_t<lmQM?U$d#*C~
zegbvc6APi(n4IZ`^E^u4PCGQY@q1GR3WD}vSwo`M)Sj`~_Fxrbsr;&ex`js?*9p9f
z(rMDL<*~ZJ^)}jz+ex|*sN!wvaqL(M{#_YzHXaI-H6K??9UwRX+@+zmIftC@copMg
z$Lctow4yNTt`qfbir+U=@Mx=5&J6u!&XoR>r^y)nSP<!?^?&59@`(H{n=>EE#c;KW
z%{JpIoyTGO+eA~D1FTc<UqLGbJg$RI6)+7$qc4>LjX00Bv&^d2l=L0$WR2rn<MJxh
z#mDXm7oz0&@XhQ-(;0b4nTbl(mDwe2|7jW@|J-!dir*MQ#m=T9wRZWqE}D_s95PO$
zC)<n#vCwQ1j_rk08e<+yZIRv8rv9y)fr!zx)^L&cl1U4R;zGqXv*P4?uI)*;XgK=M
zN#SwU%Y2o~3dN!|s~lzR)p8>G1tUwfX&Z|!I3`R?Y3A`@44Mo74Bd%kp<JeAVTWhh
z<LQ@`GyQ%qo9m(4hG}~}RegMxgN?cpdV+-%JRhQ}dL|<YD*$_PJ)<-xawL+CO{o$h
zS7#TNcXWQ_49_sd!kYfR|EuhJ2Nd6{+ArWsFeLlN2<{mV!kweU;=Cr|5h>xK2CoAB
z=9A%ejAB@cyzpXFBj#?Rvo#V&O_agqjA^1)XH{QLMAm9I&k<H{d<rAQ3HQe|Hm0?@
zJy@^9h3YoNjBBYC=Zs*SUT*F$8zUL0a(WnrYc$)y6lOA8y&c#pFol7WW|&ePkUrl>
zG47|N2I*q-qd57hDK@uJ4xf5%05LQs)CKsu1;Ldva|f~`Z8Ubqhx0iVnj0hdjZvq*
zVw3<bZ7G-$=jxsOb#b(K;40=sWA0C`fD9VKbjr~q*l@HBROL1t3aN6}wgQ;{tP{s{
z&P2fGtUwF(2JAvA9HW%-@inBe${yCFAQooKn$EuqAkw9#(c3+i6SR@8%l(F=o5u^;
zjRkY9>P_vTja=L3-8pHciO2ruI2SsC7bSa3amGg^OYpV;|I^+{Wq8)iBb}orMHe!~
zt@Z0|<=y+YFmsx@@_6&<NST%*2WL0M9oLa0HhlZcJbCTCXm{NL|9+Ms_^w%)bS6oS
zv|gT?%YmNpMUZ2i9bxO3!cD2P?M)FW(}>tQbhFzgAWSrPsi9mKRvVSjP>@ua;i>FF
zygjUJ@mV=>z50kS>gc+`WA&GsdMnUUFjYu9%*q<Fb8O2|*9SXAUP)3#tXOYd#?7AT
zEz7CE5c-*LZ}!D(!}k^h=6YDtH{)f(GJov2B=8Oo%$2Vi{Yq$z^3rl@<nhehElZCu
zcClsp+RFah^MJBT<UHZHZp$V!CDUIy7#evVqwEqDj}-5(B`dn;-=ukK%bNIVyEjny
zka@5jI^tA>^cm?B_=irt-=V5ZPz)n`GW}6!^P5tPY1^*eMrkF~bOUdPXA{<#!D6i7
z>PT(;5&kqh%5i$fVO^HbGnh!*8}FIegQ~Ta>*XKv>&+7>#cuMDa5OdbSVp@%o&$Tm
zO9t)8W)Y68k4xZyjlcX7pK{Gw(%VV@UE*#@bfWtDLe7SV&Zh>wej}l|2D%A=avRIS
zHl^phQ#)^=HID-l1m}ws(0t=D{Gs5*2URg30+pdJw_CdP?17w3J)_z6nSO>)65sz?
z+Q1e?<KbTJuz8}lM|qfK`3|Vvuz`bw@j^RM0xywvRiw+ev=D{MwmkNn4r|8m$!oc^
zSV>Nin1uTe<C+UH-QLzII4b|yO?!wS<%wR1AN|cfe!^h@h(s3YqMRn);rgQH+~4r8
ze9FxK6S3*bv8%ASM?t34*oZtn(QcSg%3UX5(y43Tc8AH4=E-N!f*e9osUbOb87c}o
zH$ZHSbg~8wVfLefucI8k)Pwqi#KJTe=C+Jt$p-4w{#Fr$FzvOVc_vD|X$b4v?Caqk
zKxNf!QCD>nVXnhR_N88YIqf4|Jog0B>I9lPXcce@QdU19Bx7usp?roWx=1S~-ba}3
z0$o-jI{GgqQ#t!ae*GMIhx3F&${72?F*T2;FI?O<(iVb1fh$R|>xV)`4f;0PS1Qz;
zmbRJf<%mtSBT^SyuNu|(Ib(WgNL@q9<LjFyjlu%)JlCGK!0B+0Q0|W5Ve?Y9s6%#)
zhLD^-?4x$>KGt&-DY<P5i_xO!n&?5+8$R}xjQMW=3<{5aOoLLQVJ{00FG|yVWUr2+
z;94;WIaKL+SzYai+$KgD7?nBis*3qh=)@0_v~qERDc;E~NpsCxCdPQSW*qon#8(#b
zc;{XJaQ^cc2J?cr-fmz_Pnco+WLjR&9o;e8M>Xza?Kiel&TNhnj&4k`vLM;1KUAg;
z&ezMi1r4b_)QXqeUdu|hvTl|BOVu3PS{Hn-Vd|uZrs)REIvk6%jEiIy1xhNO;X_e3
z54T;XYd|-=qH1*})Y)wxX{VmNdSABg7RY39NijLG=5M-?1s3Kr6w?KV1Kmsm-j#MW
z-!tB+T&i;fop$rH-vUkR#_|p9GI_UyqFCz?f8-;{z5ubHZ{ZUD)#c66jr>^e@jveA
ztux(0rM_N-ImMt=)HUiHlU+P<WjLPI3YsjtyO1U_5#tZW>ga8g$h&Dyk9Voq@4DYJ
zfi(?zR>Yb>r}^ErInI38{3A!AKnZcmp%zJBWtPV-1C<uE_a;C60yN>a_L;=EPi3{F
z=Xkf}ELlwkjGyXV0X8rp^X8g}8Y)keY*Vu6`^he+CvWz7E(Sd)!zkpn>(NHq0bO<5
z!l&b1IOKXmqv?RRueD{+bH|>2VB0L6T|75!@UcY>IdGy4;%Ww^Rf(O@F`lVZL7<zh
zxen$D7#6e_7*Tq;!r{Gc99hpq9C9EID^wpmj0hgP3gqG#A&lIk$ECjtH&n&d0R-7Q
zqN;5SiQxx!$xrc4d5^AWuIkA<apdXylTs6$vv?SESAIYpS5%W#Wy&RzT%B8^p1*es
z=ES$&kXux!36e;4j^3#|d2N=jwd}e34L8gy;WM;4xXfY45LbLKCMpzq?e_aQXjgoZ
z?L)uXc&5>ik8#iphJQ`rZ^fRFQ4e628x9KMTEtz+raaTO2a|5CPhZz7X6ufA_j@Dy
z<%QWvMg3!*of+Z~Wzgyje~pO5w3l1v-yt9zS|r|AYmMxsNXRpsNG!_vp*Mh@2o++d
z*gfS`Kl`B|PT4pNm!GoKn(b^jrJbYo8B&z8o`W<9aQZ38vgpa%R~_zJJt}8jN4eCF
zd}P$Sg%|RxNs&U7i%v+QqQMCGs882kR@pneA=WY>h{F3Pad=Gjo`MW^-gIJ?mDooe
zbBPTA8RDTg3JcBX{o15PUbPTF|7U38VA%^RJjXBOWFCqz9`Oe5X6wI4ic7Q%1BvRt
zs_`-th`sk`xF$XuHY?i}YB-GO;Gq~a$wps5K`Nf|WiK|A2!)f3N7OoMk_{+RU{^Iq
z5fy2=<g)chk&j1{Rhk;0BVZIlQrcTyJH!m-&76_w2Cwi5j*LsaoC>3F;LQG>?NZ=t
zHp{Ytmss(pYLV1sC4<K<QSVJ0Ng^PL44_R^{1I?wdECW1FO4~8;VUOj-2S<W$wupu
zjrTl#Dv4Up(j^JcxP-#@iZhH~1ju8|db@E4I6>pou#3JakmNZsN|e3`(ytcoQ^dG>
zKR5fl3JtyI4gT4WANkphxw<}r)9@`T)>7jaVOMeHL}WpZzwxf({_HQ4QX=E~8vb~o
zde0M!^*cSP!PbLeQd~P?Ix*HwQxvbtp5oe#$0WF%l1&?53TvT?KaFuRq%{H5v16Q5
z+5G@B=eXMEz4T$YF+FR)@C+UQ7EpmswRxWFs*48D`U1@JQc5iDW2uQEmCV>^<+?=^
zH$gbcVLjZs#A}G7?S;V7m1x*LDXZ?Nb+)IE3(5kDC+hT6cm@6V6XekdQ;x;vWRcxi
zamPAc%5o0>8xg<zzj=ViQ2Ew+v6nscMyhL3hB!OX(G>YXi1PYGH2ZoTo!>|5PJ7p;
zJe!LSbi8h;>-x86H_g2?*RGjuS48#-N4^z)Tux$ix16}Gby|QLgj6SPrxL@G2Hy{M
z=&%pri1ea3e2Gp?uvcmCSlWNsl;pB}uJ)UB+ooMGq?I$Wg-$9!AwJC_m;n2C!NKo_
zHOm)?v?mhPq4*&kg?I9Ca-0vlQRpNe%w{KvC%}+!y&LvLcLP>pT2dq8UX>Aq70xc^
zG&1AwSPH`ZrcE015Ywz5J8l~!zg=^UabvhDlyk@rh$mWtxG*ki#-Ql#q-Y<nOjzG{
z1c!^zJlv`j<O}++#bo$Vw1vRyw=#jFYOz{lv@$wmJwkMF2Yc+71sd9j&i$i~3t*^I
z+4<x}uR-TRMRBp{vuU&X#3MNu`I$V~PG-ZtGg?QCrGz`@@(Jm&G`nH*1a2Pkf}`EX
zj+#dWir12!3Ym;qGwO@OVTmdul<HT)-@jqM5PK~%uOVCeC6<wyg-IoIYy0JNR-WvE
z=ypue(>J}zIEQLuRpGAm&ta>*m`ndma&jpl8DO!Ienc)ZB7b2{3bCXh5+8k_G-;+(
zWbnP=_ov**XGQ+8?(9^REywM`Df{L|n7i!{I6^~j2rVT9BF&p90NW>&Z>S>py2=qQ
zrZa8_6IO9i>QB;?tzQ@P(lX9oPY(S>ID^Sn7HMwe&mF4t22^0}E=Cy_?GJR7PEQYm
z9g!iBmyGb*&xKu+K5|BJ<DrhEGkP|UPufu#2sgVc8<d&;OwD^Coc4R6I5dv+F>cWp
z5yIW5Pp9=%YOHgr+J!aPCZ?o(CIZK6-U^Z6O+Kz_M=e?$d=1_FG+2R9j;r<#7`vO|
zk=(!$lG{<QA>?Ch^=8b<lgQ7E{MkY=qA;n=k)%T%<F7Lu-FAl|<jZ@TB0qNbFU~BG
zK8Iq|5>_w|zagKbiLEl9-sfA$ts3kV=MWe0nx0oM(uQ<o)nxInWk!o%E-2>3Oy1Kn
z^7o85GtV-HJRN}RTXd-hx|722A78E&d#A0So%%byWit2o4rNQ1hdudh_Zg?$jpF+Y
zxh2h6l95~-&WmrA!K1~+yfG&gNxSg^l^tORxUNy+J~)g_{DKLW=;X`Cf~jw~Pesv`
z)--GT^Kh&a=`gPqvp6@As8BCRmPhGisd%~G8O24gB@OMY>c$n!2vz)C8(d5o{-I^I
zNOCPz!A{;#awSjQ-5!>3O=gW31o9n^??%s!APa%21-mnY@2b)`D_NX3t<{o`No}Yn
zkt2CNY;g`t8N~^OYB_bUyfnY*_VoFKR|C!)K}5JMGZ%rRQBG7lGD|4iGn@DRT2K`j
zA#Yiai_Oy)6DAo}s_UL^+6I})a&+#(W%AbnH;`@33vcP|Uq=z}PYhgzmFrL24jL4_
z6W}3Qn$KWKS?xs@OpL8>FnLzjtCHo~F7sTiehFWNY};LgG49E`K()F&kUX6`nbw{<
zEQR1%886D#<%jba=FVA$jr9H;clfH6@eYcPWM?TbH|k*g8|{$KEP}(n^2~h8n9JD+
zkvdCTUjMsJL?|jDq$wXypi$#Hwb}F5B}=;Epe#lwxihQ9pPgKd$)x><DQt)q!kMSJ
ziqiviS;nT<pL)4DC`{X;S7j57%|7&xezb!A`X*3c6~RBv%%yN0iqZM{;(It&^f>H;
zg%OrB;jFWWIh`yX(xHEz^o1hh6WC89#BTvyVlxLdc18O7L6@HGE(v69^Hdwmy~Mv+
zl<tu<?5k|ZZ_xKrLN56;HYf0U%{mpB3m#P{mZnWw3*89L=3<=3VzF0euk16LJ%g#j
zb_8iH#*p8Jk`-Fia8P|pm3`!HG|g}LQbD&xGe4Bs^M-iVb=&ugsad$fNyRft;_Jt1
z-qALpOG%AKA6?vcn1m6vDbn28J#C{mRA<@jczo@JLjvat^9zMX0@$AIaUD6(eBX<m
z_!vGaEqjyG@XU2Of<=4W<a+C4|CX%lD?X|zm-VVUY8QR;voc!8m%Q7;9@qG1s$868
zbIK+%(!P5iikeH%^)j0W6|ZWUBQfAkJD}^cC(o-kT2)sxZf};*PL2TPbL{r&GhVnP
znN15-Ed(<uChixf;8(t4(>*T?`;+@n%QUndATCZfpTA^T7=}sRoXxo++zOj{Te(uJ
zsWF+j13D5CvavlOUKm#!cUvFi7sjRjEqKI`yg(%_?<E^XlLJ(_#dAF<S%6^+lD}Lf
zM|q@EgckZ<ZR$ns4Cm2EOvK=CJ=94#5ygPXX9w--3`~koh=jCF*AVehTZKbmq~VVU
z5pk=U2jNfOaT%+g&EuiXN;e^UE$b7<Bp)|NB6VPg(6RO?^G<#-i9*AAJ004BUloJa
z|3GfE(9|amg<VnDjER1fB=r>n9*gKd3L4i24|AGv)_4Y1^+)gW>}O=?wBCFOe8&;$
zZ{|<4N6*1sN`dxa6ydk-loPA?o5TtI_uF1lDQ3Ryht3yXBUFft+KxMMu@(z<8K#WK
zBo;xX>W%?qs;qG+X1X5>59&8~E?sXy<`EM1HB%YJvh(weYNlG}Q07`vX)2mCAsHNU
z_TyK6_LYETPpDRsoJ+CRxvd>)>n@#!`3<vMP~@K~)Nocr(0%_LWq3DwFA$-i#5}R5
zUN@fTOts1ucOg+L%X%|dYhxcy?x@ZrK#1bS>SCOpC63kqo&xeC#I~Q#sE@(%i~u}8
zWH}*_&k<BwL2ihE@DWdGhLr1xmTX<19eL-LMb*(_|GT^}n~)@}R2U&=!hT7(1d_Kg
zb(oWZ>Ll=SbI<bc3CfMU@+Cn3okAKHiu(c6(}1{lWH4la6^qeKp8W=Psb$K_=%zCs
z^_X>ury>h%)`q5yT|HzkGuQ7O?yQrK-qmYJnTA}I2Z@)T0*Ll}3jSV^SvIH-=kzjU
zpPfbBFdUDdL@PH!kJpp#rTJCN8BU<_R&pboDdpirgJ5SA>BMqWIl6hG9Hc@~eNFd=
zXzwCV?}ZbVuZ44S>czt#7SvCX+&mWv*F`UarJH)ksKf0DWr_K#?86<W)%&;E_wD^(
zMx8v~BQJKl?-R~Jm&~9)e^$vh*1-Nz>^|j^V|`W8kU-|3S0bBb`ROT0@N?_GJcxqT
zA8CV*{}BNxp)Yj=@#7)QP%7ZS7%{i+a%Ix-0L5wAj~ZvqB~za1!3Y>G$HdY-LVTqe
z_%t}-m73z4&S(A;p%2oD)D``B_NBY1p`#gG54ejy(dh8H2<r1<P%gdy_JHSnMsv8$
z?!il5QLT?{yQ0BO0T@_<4)-jWO~}<<PhW|ll0SHtf$9j*81vGQO#-3jAlrOOBqmK4
zqu3;(T<Q9e)3v%kSpe>cWBTiEgNm1Rf?`w(6S`Vi_$yr1CfX^rL{l&^sTwP7<ZAS2
zZgHC*B#W4+yqepIjnMIrU@A)GL=`p_etxa)j|<VD;8IyL>=BP0`z*_=$gY_m+HjPx
zl34QuA<%2a%9Lz~u*KZ(1~XNLNy6$y{hdQiX;A511%|ozfCI`R9XG0QX{X#JLkU}q
zl1lq8c>4Y(0geo@2YiX3Cf1DfXO)$<?)?rLBFYn|NGCExf&dG+r)=1KxVx+PHe^~w
ztIw7TxAIc1N+->F)y}&fRpCY6u+w?Krn_Ap?aWR`TZ&mng&Y=qJFqI)g~XMm)!o%>
zRt!4`Uj_9cezw`1h=*JS{9{tpLV4m4I){<LbUW(+4P~C|yG$pZ#LZ;nK5|7c>Gw<U
z-6*VgSg7llhW@sZEqE?1{&N;)uSTV)lJp`wLIr1J`sNIUsm^3;WE<Puo@E}(Ew(p9
z?JB>Dq%M-Swei6l!tkmOtPdzAY}v27BvkZj5^(5mI$WOXbCN~~;$lUhA2ahXb0nwJ
z?@o`i)_5WydyP@YGqHvkedJ#E;gxZFa`ioc`1(^luc}5$sBQF7M5)lsDb3C@-=fUa
zOi#=XzVW`-rwQi4!!ySN^XdV}xI-%G#>u^Xv-0=v#Mzt*!VM6^BsC9EJT6stUcd>w
ziwGu(7jDEZZ;Uk2{jJD;z0jEWkuO((z$(;!BW_F;uKoO`%HDA$%4u*&=pUHL(|dL#
zRt@#iI_sLHpWX|zv+U)1%FYCx2&)%NUn^Bf&z|{8mn&O8#JpM?Y<H_syLz)Uf(CG8
zClFC)mNaiRmP^eBNTqANUpnAjKbzIB@@p=F9vN3tBgmKF8m3vLY<j-eyJn^znkQ4;
zi>+ej3!*(KTiiX<lD)}}kf%}#vrW<PW$;X3E;3CIxl!eh>MS*Zx)Tt+JkG`J+p1XF
z6<#E`-dL$9YK{^|Hh$5sZ+Z!tS{=PzrQ|`K7?Ad*ws<I_H@Va3l}^5MO0e`&y{I*r
zz}Q1N(z#>O9!e7TF-xf?oKMo}EAtX_-#SM{cY@H%jr*OeWuo0&PAebMncuxSWKN{h
z0wyFs22{J$f=s*cH<`g$DeZ5#ZFAhmXiI4AF=f6^k-BUn(|%49xX;9<3!81KH&m0k
zZzZTA4^>#3!<X9ErDM0K+8o2hoR)eM`Lv+!k9aOs^<G8D^6Oo&A<}VPpdabZ2OIbV
zVMT0CHs@D7@HU59(5MaPqj)8kMos`LTII%Pk8cJpO%5JwWAN;$JPnt`y2_UvSiawD
zbN}3qF#Q=abgGrcDZl0hs5nk+hb5Oq<A|q%2p~$lLxL(FsMsb{&ztYr|9mN<q38PX
zR$04>#$uS$)RJY#e1r-Yn1~gtU^;9Vyi`gpI?<j`w}#sX`1Uxr2A4(!Q6~z&c}S;c
zvVyHc4S3)s*(t~-Mg?hBv(sW4tAC#VH&6srl;XyVw;aKb!U#+_YrZ?*Xf8+vf9p2X
zO{|SBPjkZ{KwjZzi6--D#ry$A@?dJ@zSu3pVR2+MD%AX~mQ)@~6hmwZ1;ZQTe0^$l
z6E4zAX{c%eTY<cP5%av8(+kYMI}R*t`HYux&xe6Hd=D=%D*1Vk3aSUb<O_|~TwF^v
zR|*ULp`ZmaO@>JTeTwOft=40yM?oU%B_CWp&5l!6X{W>r*BKOvU_FC%(NprH!nMAO
zTZ}xcLUme-5Xe_q^^(=ZWI>g|m*w>7IFaM>9JV$Z@pJB=i>zFFuviM1ooCkpe7K*w
zHs1u>raNHdI*@bdsU1VGtk7pm@84r#)!@>U%}fWDQ`$oYwuEracRLz4&4lFV<m!qK
zqlq!F(&sT1;E(BwDu)kP<!r^y(_4_I8_IrBEIrj*^+HyiYJm#Ib{UbD!n$b~cjGP%
zpSqKBjSPtFZi#As>&uC*G_3*P*+yOXY1{kLK%Y<tdC_^f3p)0C`w>bkH%;bvC+tOI
zjE5)G-6WO!qmZEZ0-ZK>sI)1g$GUyqA>G_UR5rW2sLCWU)BUp9u5pfdk}>k4`>Rux
zt6I@*Os1NdPv?U*Ner2_yQ9U|3ZLuIps$d6qo*F|fZSy!l>qHRF3+O1|58FK@YsG<
zzAn^<RJ>Pk*5>LdYsU6Whb7v+ef10yXNq->Pqj4Pu1~rD0g)=}y6=xk#V@D1KN-Hp
z-H>^ht5VEWtlt9So+Y{tlXCjn^^u?p#hk?te(577uaO-?zaTim3Wd6aNR9BJY=QZD
z{U3)`^E{+km=`NwSZIUyLRV}5IqcHVH^|Ao4tjyYI%~Es#U!gu&=fS2km4f!T=r_j
zYA<K6A`W5u39_YyMlkh$#@<v<2a0nB@%$3rwPI_yL>>vrWr4H^@6n<;$7wvln|bg1
z^+!ewVvS*6&i&z3*Mt<tw#(WVgZ(HSlqp7n>s1a*y?NVox4jnxZIf7iwXM&WRZ6<n
zNuoK{nw*!C2XoU6R9^^0x^^j#H{>)&mBlPLnT|>|wQ7TS^YiWOlU|22d~Mv1&3OLo
zrWB?~o-aCO$qr4@6Rolz9g+^3izk*v064JWw`zDd49wx9^#3l%aLy>wsiBdQLgMqq
zk<=^kGn91K@M;mbgAHIsQs>%4X04Ib+WEh@%U3oTbbll;YBKk`5G*YT5{n<C)l2Gy
zt}vM*EKXbNPL(rm+HjbfL3;}UKcQ2h36$+J)V?Vpv9D%5?fLD9V9%kNE1DZ*{905+
zEnBCW^baSIO+R0L%_ZswkUG_7Hd<OcCO;XKq6rCXgm$tWo>dj}s#M0KG|0OYwVJvj
zbl4y+qUC{R3nLBEzMRO_yjz-qB?UhVrElr$r=RS?Iryi@*=yHwh*8xFlyMmG-{@E(
z;kKYE`hEUE-gRc=7iJLW-tyq9QrG5%q>EJ!don|`K}5|Ho0`bIsrEp9V~Cq!Z~UWy
zAB>Ko9*j=taN#ks<X7PaGeszy%fuI-A%C()enCnlR}JXI+RtkQrm|m_woFN!r>z$8
zY&zzKs`_aOV=}(ka~ohKbEP!(+cxm0Z$UP4t-9pW{<U(9;y9iY#{K{{FO*Cpp4{Eh
zW|rvgXEnGpk$R6=v#VyPL{ItU=Rw!ZP&_{jHB7Qj&M6c5QY#Y7P0mF{5r3zrCi0YW
zl-6-<EvPttj?Wq*^NIx!RE}!LRc0SKvRM&BCf#*merP2nR$(6uSm>@Pp9ez<D(ScQ
zvqxf3PHjtY;@#Y7w`EW2LaqET80#rZf!`7;Bo=iu+nmGl+=_10V0_gNqwb1gt9lSF
z>)6zf`J(jIj~2!(ztHF8r=z%Kl0Ka_jgtnJrJ*sMTo7}Vis$1Y=Cu(M)t?qg4jx+k
zDk4FKOrenDAK1oGp^I2CR^$-C=splcdM)GSNNOiJv27{6Ay&*LhUL&E4-Kab0Qa>!
z4|!}f3R|CG9*zz88~S8sw#hF>9ONNUI38#aB9t8gJ1x%rPX&kdm*0ics|YLx<20)0
zPzA(no~u7cjZsm&w^lV;442F?@Q=dTbyI@i;63&#Ly^%_kIsCg;o4Q%V<<l<pLPW4
zHqYyGxLKrFPpFpd@?8Fq?6uzpKfq>$g(n5TQs*fwg1Ga^&1ZPPoAT?v9Um6ERu!L|
zFIy#9l+k*2-crXk{PEOTFl&`r4ZSKf0cm#(c@_csNyH`Vh#s&K&;~E1&l`%lM^O>{
zzdpgs<GPUHA0qkWJTEUyhs*Cn)nBWl_wqp(dYu-l1H4`YlItooU2V|Zg#E<bN7M=_
zWdYQ~=wsgXc+<$0(sH+zD2lf0=h-{&oDY_JTBK5yY<k)qTBJUCg_ofnkSv5oR?26*
z!9rM~9Ykx0B;Xb8@%W%(r42t<ox%J=l$XNt+Iu_!`r31x;2#N=OtaKIUpUE9_cnzi
z+A)jq&XHqIE^dxjK2Fc1vp{5{uaa3+L%dspc&b*uwYzSU%1AI6c<b3<!wn=jm7eKr
zZ!q{>x0i0$;cl}^l{QM6WH8n6DvKq)KN7GbBCCbhN5ZG0tXQsdenME<0dek1qHo40
ze?fms!0cX#{`|pY#W2fsEpp+(n76*s=%phi{QYEMMl9HC5j5wPcHy4@zaR?nkbV7z
z)C+ucQUT-i7S<<d7X!;$o$4-K;n$Ok#kN2+gfbEpx`t>>pP`ddsk^M91^rT5#DXn-
zbL953ozB+_;n3-5Us&jO8F2;I%T8u6$nhOKdHC`Mp8Rd^#Y_+Vq*1j2VuC@UBvuLe
zd&nQbQ_4F=+>L@`=ZK=yicCMx<MF%jvgkpwsP!u2bl;0{<a%A`d%M?*l7cB}rm)a`
zp^Lt7a&8$}T@}%a!<K#8SBIZM(LMVM)F>sjOzF6&yA&sq91~Kw#McM^0Bm8`lH^#7
zq9TJ_s9hdV<L_o6%)7CjWuV)WLSnGvYsb~OCS3D){Yx=#OFiL=ABMR<TRu8G<l!XU
zx$U$?0SeE-@{(Akjf;7uq?K%z^6<f-X<uqlK?CShGnahnq$lUN2C6kuPvi^sJtQ&|
zd$)rx<l*<H6=@ceAM#kb|N1#z-T88sEy87HwwW9d@E#_cN@jg}?KEZT|F!x6L->K3
znQQYkr-Ik)-XdBCU%R_S7tMRE7<uZ0T}5H8o_HsRgXEWe+}VfeUu_9`Q%s|ew)KWn
zO#SkW9dBv8UM9x4#@w3KJWM7chAR2ThiWcL8>jV9E|N{DJ>vPy_U8bd=XeNGuhAPT
z{Nsn6rA!s`o!^Z)pFK^9c$0?9bxce55cL5#a<Li53FK&S4Tr*8FdOZ|-H2riYUDUa
z*2%U9PG4s&G5Il7$Y^(SiYFl?3)+%It$4mLpLwrn-qTEjjo->z)=t=>en5noeQ@=~
zE&RyWhgOW#n0OM=yIsUH-P|tRVdy&@wo4_%E5D6u#uHs)VdyLUmDvIt=TI|^kX<M%
zS#wuYk@2Gc=4d1eof%H4-5O|>AH(0=@64ZZi3odp(Rz={j)rYW{o&8bg?_|TtoCHi
z_w|<vP@|XVSpL&yKYGy;Q)uXv=;{sc)x|A&()`ZBQO?YVNhq4C&x`71>;wz2;`{B3
zr$`v(#6Yz}*N~FU9KSb6CQD!R<`3(QNEO&72;qs0aM~Hf23-v_v)Iv!aes&xuzU9N
zk@<*~ce_eE7n_f9ITM>iuL2~%*LTjE;)zIac4Zjkef;TqD8uApEdg}_ikC&x=dfOO
zi7&bRQa_#dJ4oEK0=!6eeRw0}GLC*IiOGn{&7mu#I7luNktBUw68tph>g_*7W$=8w
z$kt8kX~vVeXa`08%t}XpwpjX%G2NZ~nFsrnF^!|=T)(PT9~@8}3tN4xEF8&~yM#2w
zi4zPh;tx8-buBhHE=R2;iT{H|!-#OPJZ8x1P=KetytrJXwPy*MWM&GNU;?WO4*KHe
zJ9^dmt{jel)3K8E8$&8cfFrA|QQ7^tFT+0;WMNSipE6Hh9O<o#z^Z;Z%|b)4=8Oan
z#Mc!#BcGBB4wx_NCOKmh+`E1{{b;Y}g!W*kM?@wq9<Eoly(<cR(}a`ITvB{Rfb!w!
zTobv^ov-iI*ZF%ZbQT9ofxY4AMb^YdzSv~gCz`&BBI+^<Npx*>))Y^tD_wr<{l;4H
zKFlk*ws7rORIT<j$~d{;Fwx#*^spJ1CTV5FCUtJpj>p&d{gpvqF6luNE{~;Cl@wZ@
zG1v-p=Si2G#E2DgU-c28Qu-RYjtNOB(m8g&d)ApogBgCVi$1^0vd-^<LJ91;47hi>
zXLQxe%+~Xn3LEFcwSA)qI#!oGabJiD$0Ifgk0P%}{y&_(2RIe{|Nmc7h%!!Avg=Uc
z7=<|Y&aQ~;o$S5I$_!<PLlh-@9HV2;%usg6-s9jH$L#+;clZ7I+`r%N`~Ur~>wjG?
zsT>aHJzneic(z@SzRj1bP%lL4-Q+Gq%wSs*NE!q#wYFNjA$)NNgG?4FU|!Jr>1jUq
z5zj-*iF>Ik`R(7jW}wxFBDd@kMhT?B<cP5?RrrKeN@KCrIj1NzLhSqNKS9l(_Xd=_
zcgN~`^SBd~Y)v#a#cYqQvd{G#RoM>M#AEBJv$(w#dqng$PgjC1<vv*`*Y6(Ue$ni{
zuIBc(n@M%3!&7+=1@rlM#`>+mh<@Nm{+FV6Vu?TSujoYf+g2LiTTrSOURN<q1s2Eq
zfj~-Z-hGI=fQ;)MisY{TwkLo=c=pJWA^cTFM}8j19(<Hgje9#Ct1JD`R!YZIx?y*v
zr~z`R+#J2P2fe&qa!^)jY;f2T!q|uAx(Kx$dqJQ>-as$Eb+2Q%5BVKta(Bw!05(Bj
zdCIe<K%F#Z)iGt!LV@|RR=!SC2OkOfS<lUZDef$-<IwX;(-Gj)v>LGRpCn{M$j{Nl
z+rc)m-)8)o`MfQv($+hYlKHsmeBX^iZ<ou#7IH`y-nuW;kgQQbruW>U49k&8DKGEv
z-3^Q2wzY{6G2vZHL>O8>){xINFtD&3@5j~Qi|(9y0VUKz+aJ<9MP^@ULm`l}TTu5N
zg#DqI;@+}`{eYg;JL9kB$WMZAzAVvBD1Cc?0S(HC4Ut5k=bhTN^BL=_(tC<5_;xg*
zoH`b}lm7x<wms+o<jMmI=WN*i1+xRIAmm(z-6iXyDkauS&bAB@np6UEGK8{OT!F&x
zIRlM!oofyMONM!cc&SUE-V>>0np6cTpHJsKk8lT4AbnGse5&%+_!R2(R^(5Y?Pt%(
zJb}E911S>~mD|H+(A->Hk<$<^fpbX8r9z}GRg?3v1`tnEG__7`mp|2fipG|8gg^FG
zJGpn4NBlvo1}oE?0PF^_GqFH(El>itw_UMTOBagT;z_FDKBR)Wl^0!=F!h<_8my)x
zW6jiER_j@HFcrOX)3TJW)Gi59PK79_9DWv|i|f0TvsJZ`I}NwB%4oLPQQJaOK6pZZ
zS5i3Q<;4J<C!;n2Ibe|Gz&0uQBo%4ky6LlYeWI-qOJvQyw%WyPSt-It8p%X6XyJ;y
zVr8m*&*0p1J$QSIYNpnZAVO;|=W3Gbn<GHL4`<&wx$P3o&z(d>ETI%XAK}U6pNkC`
z|DE~!WGP~!F)oD~+MhPYAHRLSJb#5vvdfYXp__a)*p6?-q%-MpNDe2N?r-lygZrwx
zJ3S~LMZGs(C$a^mG+KHwR$i^bw!jG1kbIFZYJ|mJzC`Z9FY@G~3zWEp(9G(^2NRTi
z-RH{5(pYv?I))0SeSQU(#2k?XLbiL_b7J5RYi}B`CuPH>TOZaz)|dwGSd{2;?%wO(
z)5WK8d&yH4t0g$J*}GYn_n|ZoZEZGkT@x2l{EIdS<3?k=BXqg!o)I>QS%%&(%^-Tq
zjT~k3Q=U2ad-z20soO`Aa_6;`nA!wE4O?s(9RXkE2F1+Am?!+0KR@9<E-^W`t3qJz
z6a*hPx4P@?@WDqX72R8I??TPLZ~H%V!S6%_wvTNt#9Ya=+v1!A&QLRe9k^de4CLos
zCBSD-sBbL_-!$TAMCRBw`PbRSqX0~s5NTkVMmkyTG#7N`1v1n)Lw;(t`;tgYhxk6<
z)Z3K8Zb#d@?ZdJwg?i)Ox&I5x&;TL~5FrE;`=uEO)y^=$4-`p>V;v>Tt6CYo>Waq(
zK%bEc9@1ZhZ#D7)3WU5~totNT@2=>Dz0<071b)dlno68W{QIVYwu_Jo!ZeC~{|~BL
zVL<5{uR0@OBE!4p7N|Z6Na)PAvlA`x^>N&r+PW6hQ>NU0s2B{?l@!llZzs8qF1>hX
zBF_K>SgI4v^$Xk%B$gH>D|AAoiIV`5O|N20+4E3;6x<iVBzJrnq{|(b{Vb7szwD}h
z`F8iB_U*JJW@+jKO&EKm&5Ug%Az_E^uBcc%G!K5V)pB;VS2~5+LAJUj*?K596u6st
zPLA@s=B8S%Ix1{=bP0x9`gK;t-M{O?i23%lF+`0i*)wLFXwX!5WP>KEgsle(gc!X^
zT_G8}9mgH1!ngh+@uj=;ZQJpUy`vfieMMt-9k+RMxLNlCV3!OHrR(zA+Fs~&Pn(yx
zacJ9eMQ9hZ65#e><BtE2Hh#uDTwU_UmASe)|1!Dzhu1px7cC39G>S?Myv9p2+y+MJ
zEQ)laE-UlYnDc(IL6ED;GO9!`_`N6}US604qRd(AI~%a6G5ZAWdMCMsYfIyyfmkj5
z#Cw_xH7h=<TeF9|fJv#wyk97!da`;5^$X&=yjr}hvbDZ216I_M-e&|DO*>PgF>oF5
zRN{T3%_~*vaJLS~9BVRd!@7_2yD85AO<0VVDJUoA1KOf{qT+78Pqifw$sLG9JHM+<
zsSUG3C>8&}S|;{%o2|MJCPr-e>KV;yS}|Glq#VVTQO1<2O4Z4nmIEooF&(AXOa6O$
z<ilE)xixJVi&frL+tq&8w3i1B<1fB=on&WP$875yVev)@Wj??eZoiZt&1z;9ot7#$
zWjtHi%$=AcQStiHXQmAx46Ys2GO!M+s7bNwRTj@tv+!}b(Qum<xDRMf@uya#xoLG@
zKl$<uKixG*Q#U5`7w=RTps_3>mO-}_KrQ}yi)JCf<1Ufjb(E$eHDnqrjSm(Q9ur)^
z;BN%8s7AoMq@$}gKSyr$#mKKYR!IWN0aj)*Qj%`;6c91!da{+xG(@ka2oO8ySrL|#
z4GjZt_J6^-mS4ZL`?6#z0<Exi;TsS#YbQG(8#u>yJw(Qmq%YR4b|v>btE_9qdElSO
z)zDY4SL1k2!giw3csLK$@O=1@{^%>?K&0t7YnF~f-QIP(Z@V?S9-A^LFFoJeE8Ivu
zQCDB26}uo6WLHC*^z6mwBxMR-mP-OrhJp2Ex_2*!3$abag6T6JR1{k&4f(nxwM2F!
zxm2XClJj;W$WIw<NhKYuAWZ1nrWg8>orr0BcV0*FZ)JK+ek-N>j!f8Vz~gx3dIna`
zJel~I*{BIMkS;ItstS?gcMzLL<$8h4E173X>r-3jXqdLqIg;TOs*@B0vSB4{$n((>
zr{yos8X-q|VDo*7tx8F3fA!lkb)YSGH;*DG#jU8l2KO<h!k{fyM7LeIO})7iaQ3~K
z*bT@+=T6^Q^u)wws~iJ)YMKvY5LK5-c|MJ4>bI{sTpuPSQ#^pM^A#u`3my0q8WU@X
z*5J*mSO_-4zPyF5UMBbI=c*TJu|Me<HXu%-)cN=euvQ>F`Py^eNjTCedRDD-?0MIs
zW`J0<tJmgr+CzV-NvkBx^?uIUjM}hLN?8|O1^R}xhz?4B{v>JLuq+ZQZS`PXW-?%e
zP&_hpL%Abo6aMf>p70XK>Se;;;cS%cnCiJ09!C!51(FkDJw^-YnYlr#h%KO!n*gIU
z#wn$DX8T?z^63itz251WmV7|pKmLsPr!~aYIaoW^{qEKm0eorjxkPj3J|WxmGQ|rQ
zTVN26csjd{)2%FM$^_E~65V%6PG-}F^8k*85)k?#JO5S;|5?Z92nuRjuS>eBsFXbN
z0?gj^s9T}98-8G<k&Nejfj9qHm!$o0=??rEwi}`RyB>tke*6M{(HIq1U!daeXBq~v
z(0|r#;B^uf!p7_6b_}#7eLd#zW|D$^)`FI!eqd38QE$CNO^PbJy2b&ri!}A+a>omN
z^roCgQCx&J94}<jTjza56@-Tkee7sSE4~&byQnm?OUmdEePi9dLxpyPWrqSy<#6F`
z6FWPAeswWqKHCaw0ac8lZ$1)JUzh=AWw<H3MM7U&pT6JYIO6B6kT|Z*a_*ZG=ym}+
zJwptKzUzS8ds-<&-*W^?A;RIzO35nC9%fR!Xl9A8)Hv2%S;<5i4C|wry1dM&TlWig
zWWz2tu+5k!o&=^)*f`-O@18*7Ov}8KsnZ~P@-mC7|K}Uz7u*1O5+9<4b*tsh?Vtj+
ztIC?1WY1i=X;L{oD5O|LSpC?tD>v%(xI%JI8-ud!0^W%y9A}gX*}djD$YvtfR)UC$
z^B)Kn*vfBvZC?Bw9A5T0IXegKGaX1YtN#0Dz?=T;mRUwGuE>T`o+yn%H6rCMh8-tS
zg<mp^DV5%|nD?8x1s~BS(XVzDyq|&LJwN@>-B)rF`XWrHJvl!vB&4qJP9PTpC1x|k
zndT@Zz9gg$#I9RaxzRQ2bxo?-1>NgdFQ0Xwn{;fsf{w|ylNFOPcw3Y+S2@-XI3D2F
zvd3<qr3yiw)>JuthJ7B&k%<ncwnCFEI3m708lE=#$zPtikVxq7<)GgInni<v;iEkC
z{m)GHf0o7n^uZ^^EXC=k0~75V#Io9kUKUfPl-n|*U##_`vMAqh&L-y1Hp{cz@im!c
zvPlS6b_|DoEtS=oc~e_yNuYD-b#pg>*o;(Fw>`A{TA|}YF`ofecPt?VpJ^b~7uSJi
z<5<HzSG`n`f;^%hzC$d-Ihxa9KNXRj@{LMj<Mxyay*l#-as=}-vI_$y6(+zfkQ^AJ
zn!-P*;}`RDHkLpm$ko&TuIvBl2f+Vt;0wYd?k#p#+vIL!rtQ-xi%)D}Zk3=3K3ELE
z>~>K+41;qz_smR1?VTbll#tSi3=as^L{@HAy^;&3BrfeH90umu3=c3Gtc*T{B_UxH
zv?e$}BHjFY4}=B}S_rZyL~L~klD(F}Phcxn56%&v(^xGft{K*Mfd0E6@P4FU>Q<xs
z7T#a1<prnq!vip(?Fgd2D=*@xdBF2CfdZx_Z^JpUe`OwjdV2r+dt+A$V;lK`(vn{O
z_=>0cPI7b-{wM`|5E7bz@?Rl$wyi<9qXQ5r1aiZoZs^4tdi-*mAvMBDPuS5#ZDKmM
z0g<h|Ud(7WZ9@cjXyR~>GudCB(>_*wY8^KHWH@I~7QkL!C7+wsLz2X7f-uqfiyxjP
zw)FFEF^Zq79yhza8|Uw?7k0%u>oSvI_#WrPB*J~%T9p+6%p!GwcT!r+O2;%64!}ef
zqZm4q2k@tW35e&TB=&CO8OW+&?b$TH|DP$<|IWABzB2^ky=reb18utv7l>GpUMH-^
z6F^Drjn(n^JB^oAA|S5Ek6_>5Yz|^(n>;`*iOnQDRwL0-Py90}3|0UgJs!GAk5PL6
z@5}ptTsU-&1Ue6ELG8bJt+)YAH3%lxKAH3uFMPQAcTgeOOE@ZWOxu8yYkAu0XZxle
zr~_Y!Zd{`Fqx`e}0^dkU5;VNoMqGRNuQT`mZEgOa{~pLE4<66A-%M@r|L^ht=l2g3
zr^EpOFQXwqemCwD0+ganCW6i3PdVy;Ut*H~&3^y;Us-T)Cv^!7G5+Tj$_f8J_=gtP
z+Eg^@FB9MmHavkbiOk!ZXCPN;4+Uy6J5HuTEAT6f)+7M*H**|dF#f!D?6sK6(F5cf
zsvWyIDUH$e0{WBv)aCu+K01y~7o%Xq^#32f0l>@#G*m3<_qo4+mMlF2mBj|YV|>?$
zq0vbm$#85YnF0QZ<e*Tm0?abmYf*PY@bwp#PboleS6i<@TQ7C)O9V)+))|QSrvOdM
zvMOH~;ODQVTC^FWSAazyrTFi|{xFc(N+2ZgpH~82aZA%2Jc^}m9;VsN;fQC+;YBA-
zJ0KvxigzI&1@>~+JKDRF4TpY$$#MoJ0M3C`0aV3D$98(FXlnCFaMAx0rTz~eG!owg
zM*^|0_2sfZ&DYVJjvy`%1B~<Dw&!N?k^!cE06g<x2zU_H+91Pgu8QNPc>&R^q?Jl}
zm%}(`as)^_U@E{Z76pLuJTX2)Y3OUXs2J{eT0Ps+5)hGxCO$PBE4<r#Z%Ad^btAK`
zeyD@Vhb!H=>vmv2VDR0W1ekUcFb7^F=DVXoD9;~gh#gUkagk{8Z@NeuqUTUoxg8GX
z<imiwdKg4K%R7ku9~J<-)XA31_w(DLdOTT>3-@obAOmUVHHH7{s~Sl85UhX7quFb=
ze|=tHse<u{fyWilaPh!xI0T69vrTo-;SbiCyGDl(M*;Vz1vKa<f5tO;&A;LGKf!Ia
zFrV?-0s@%9ubilNB$EK|jt|uEXlLj5U6=0`P@n-BMTo~6fZk7TXrBoJNAH&(+vI2s
zDxB&!VOmqSC*Z}vsE`Wx^VD>H>Jy8`{eah(iT$Em2}rpp#P>d{zkU#C#JCc9bn$%_
zp}(I>*3SQ(kDkR3cF!I51fsup5AQuficGwtCs@2!Hj(OwfSz@Y#>m%+<YfHO>ZZA<
z$t185pj4AM-EBlxG1ygT;y|dpHnH!%<Wln`zkbMlo0FExfSVvER`Tb`{iE<73^}C$
z3(anzH2u`!S%Sl#?~{|<2wuo*Z*P(~_^s(FINbaiKGp!ph+Dv7)T_aDSl4P}_NoE@
zGGPLe==91JKU@Oz6`}^g`Cn8$+jmCr%xmX9eEIt*FxDaF^*q6|?1%=zeLS_*c!4~`
z=wz0*#v~P3wtl6Or~}oD^J~w^9{vUn5Ts-e`$2SW0#KM(Oby~G90%ZeA^72!8PbVj
zDs(|hWFKIvlE=c9M&Is%7uUxVfH~KbsnB4Lwzn|$H_S0&w$_A+?#_1LFHJE=p1O@M
zKBm8`V{m!D^j&(XaW+fD44C-=1!gxJJKOClaMV;zf^f6kvszEh{3?aoiTPGswSY-;
z9k{CVurodqjQu`9e0Pe$cQC3yw!~RZ6+61JM~lwK$q?w>#=CUaTR4!BPJ+$iDxET<
zcbJQ%80m#$d;C>ZXx-|DT;bwdf#f)!TfJD{9XgCFsQVo(b!^f2?Y~vX3^+S%6m5Nz
zY$7>nN@rTjM#i@(*N<7i0S6Np>}hqf750YsZ5v`4G6suY;5RftY=$lmmssJibw#%p
zrvw0sUME0#GQ#W*%}JjZNVf0x9&?cK#a2$nJ&=4riRmb&I|FBC$J8309xCtE%GZ^#
zyc$VlalxO`qgh5cT!Bjg3)L-OVNH4+^4gJ&_!zth48#&%;T{*2chobAicS)*K878#
z`vv^OB)oaUahT?u4)6OK#%zs%kKWGwNgoSFAW#mg>dl$%D<^GEfH3xv(Mu4PdF3gi
zwpHyz(vwahzT!6=rW4^i+40KBs`I*N39nr{KuS!Zm3B4F!~R)%?7@uy>$<XmP>P4X
zjy~r>`EXnTockZDUVo)t!>b|#>!*O#6%(itCGsGsP~dC>7s5n%#g|pbfa8fNR5AT&
zHg_$cX0>PL9L0Oc{}$L-o|}Ko_p$i)k+{C~G0^kbuO3i|yK`qTjz{GfDd>rdoUp2X
z&t=@OulPW6lnfZW&sWP6l_us%-M@A2ALf7fGLYh4%Y8hjvkxvMdG<?fO8?w+#vyp0
zgOGaRi(rNn+<(Tl6u4g)|D}lz{s_DuV9j@P0E+B;eFC_y*0dQ5h>BpcEdam^Ai;-V
zIN&9`9xNX&hd%7{cm<esT=>Qe4<cW0TFMh|TZ%uqk3-I}6@Z!!{|!Kf=f03^Tjc?C
z@n1(@3kxXX{hEc4mU2(F0kQ{8<no~Z<*nggDJs9gIY!Y3P`+DVj840Vl#+E!BerV#
z{rZT0QMP<w)Z0OWGx4=dI6V>^=P?r-Wvc4{$9jWq-oyvJH>?K(p%BIz2WcqzxekcC
z1KPab8W|>j7y^><3^*`%rG%RUIh{bc;|l)9(y{)Ss1Fo_g~#!`jsQUtVsy5XJQEvm
zk`v`eOdAvez<GRexxg!_61n~9dTs@;J-8eC_O43ZO79n*i{47@_sw<c1@#Da^=>|H
zUe8`+gEil!>`}n>`^cIlOSTsZ^L@{Es5v6~diy1ZsYt<<9otLqyz*?Ndd%^_x~H5c
zpWiAp>=im3YIl%5*#y(f2%<iq8h;u31y3jDD8&1b;4K9`Ya_GAF7-qjDYi`E+J)e;
zSn<vd6&-xui1+v0k(gD`l{rEPJ7qTvRj}|}ngNjG`aEU-cn<NSfmm;J$gPD=KrX#e
zve`L53Z|ge^laEab%}(C989{Xzr27Uc+V8i5HbURbPZDqhTQLZ=KQ-1HuHR?szVq0
zK+piS1q@Y_47XvY;dy#bJOJ9>njcJEdukwmf&6R+_;I_g6{3{8q?&tJu%I?ux&>Cq
zWcM~sUwmjKuq>n`PTq?M4?csF^WGM=8^1Sol;Jb&J8BYEBOE{FR$aS8Yx4Q6#A-OH
ze;gAYgY9F`B5FeL+C!F!!>&Y?$L+%26YuWrUF<O>e!GT0xP7cxJg<3U#U0>A4}4o4
zn&s&qo@H~*2#sivuLD4!{-9%vZrIP%qU=Tt;8Aa({K0be_UAy)^A&L54PNZpisx_9
z7-<Odd8z9Lv%Ed7p>0>mccbzXoL&WW=)VFLGM4ipYZc_r$*+9OsS9oV$Q(8Ge1s89
zyZ=?X=8LqSl@kr&i!_UQ(0+XKYf{F?Da{VKEq=x!z8z6S<L9<>5TAMS;Q*WM8FeOl
ziY}ejIj8trK!hf2VCg;cWE4YDh3fU}fGGCYyy6PK1?q>&oN7*Xx?Av-%sJNR!~w*u
zy1|3Lc6S6|Wb}(K%6T^?s!rhfnRO~T1?@lI6iw!6?iWv#Ws1+L2OJGK0tn}@h#G&F
z;~kN|FJ2Y`8NB@f>!qTOfAR-<f)P;5zzrtX&tF&po;q2iD#JQvuk~RdtI5Nohn5g?
zEYi$HQk|iV_Tb_;pzZ@|IUO9JN<&hz<Xu~g-4Hst-{*ML@YX_C@-p0?8JAqvB#TFf
zZ<ht{5?`lJVJ`|l9oBbIWsaL?x?%DQ)JOEQ8gGi2iW0)eBJn{B4{na>690o&-$^w|
zAq3W8I08*QSCwTv9&M%-q8=UQECKSkr{oug8%v;J%h5{51v0YkAmOXg;PHu>D(I>M
zB1e5m1v=1I{kc_tP}kpbCFieD(LDsBDTi4201xP9Y6%E=uK8Wa|HPA=A$Z{fq<|h@
zhJUZ<`E=jQYed21#)UDtepWFSH8Tq_jD$!?4oaX^592-ZRUaOv)fP(gjMH$|+6N4Q
z@t8KRLWcM1fh*V7>6oZh?>s&xI4K{swC<%r;K^Y^8x@yKZ{Dz3yLMtAM38dy=vJ)h
z%+_0F9umBfmzcrnJcoo6<s9CXa#nnh`a*oTGX*OfQ{m7rUu^G8l+j5xsTO5_StUP2
zg^a5O&!py|nKfXoH5(*wqHi8ASPu1gsN*!RDjNYL+j+<8qqiF<RQNqrQ=3|eY1D>d
z;39rUyN$<^Clxzgcf(TZNtkD!wf4eoz%lHQY94j)n2Ecrla0qwx}@!eC!N56qR2Qv
zG@;U^cEjS9US7ERoDgj0EqFn&My}>Cuam-f_WD<_5DY5nR&vu`;Ss{B5Gj>MY=UeB
z3CudyGI2eGO&}z4SyB^MXp}oq@*y6cb_0}Yp;-;UXgZ8?eDaN1YmO3=$0beCKF_cc
z<f<q#D=e+O%1syVU&GVC$IkSN_~esB>aHxWA5bvYa=vj+{5HdhHf#0=BTh`TgI61F
z3sAz?uD`agUMmq9qQ*qj2J%NbN0+@hh$McCl71j_{BnnT!ZMJk*(_w~hY7|H<5MS|
zcU4+LiOFIo-1}u4@x0OM`9A~YARmq(T%Kd2c`w2DBxb3UH!k!M*Yua686jzZS<R|R
zXXOz#&Le~9)bEyy<qQgGk)eiMKTOs`>ut0Yj)O`G$9IJNhmx2{uMS)}ZCzeI7<l#Z
z_s=MXm<m<ll?H`CX1^zctbP%Og>ZhFGLTZ>=oEH?w;|)moL{e#?nyBR@xO9?31q9K
zXB|4{=IE<M1`~bmMV=+zKBjlMsov&9wYMOzDXsD3pNR7&q(q7FpJz@HNGaLa>J@S2
z^1-!7<sTAlD+c5vuH59XtmENxl8%>vzE#E>xdMgUEz22>(r?GPSho}2^FYbS8bN}+
zj^4iE52>#x4+PpJO%6-GyMEOxUn6Wrt#}{oewTqjN^T04WQ>}7&s~Q+x&`%A=@~bi
z(;1+hP`g(e5u825-JyE2*e|H8%Dq(kz?i0nsvMV)G7aT3(dp)fUQ_D1lPrK6AE{4D
zQOsX>Cvr7qnw{^9OvDu?q^A`@Tq~61?H?}OD%{vq-pYD5oW?=b_)SXOozkYGRw#^$
zA>Fp7ol6iZ=$VmH+0+VEd)DXD2z~z+Z#D+Lhv|iknvo}dV6%m^wLa%w*(8#{1$H!r
z+^*>_W{-V0+N=A{vkzNNqW%tHk<H$J*V3<O#WO>qN90pjAj;l!BBTQ01{GIPF-}k6
zyv^FPabqOKv9DMt(7=?27>H1Z99SLF6z`Wdoj_rVLQs2Ul^;nd3k9u&BHT;8&bfI+
zC-<7PhP?das+3$Tt=ajU%!j<H-+dG$m#~P7h)U~QI$gfSG{6kG4}~sANibx7u<7D%
zHV#^7-^R7ICiutf^oG>>(+OIB?vG8Q`}`)aQxy70MP-ZHp?D8;67tyteo(SPI=#f?
z{gSwom13bdl^#>uX3=h!(@2F>k)8rg@9|v!o_5w(W5|PtDnFLop1_mNZX}2lyr^YU
z1Foi}4UZwXTNMn0F@nTPs>W$48Zf`Eh8Alen^&^c3`tfU+rMPTJ$QMc=aDjZVv2E9
z8#&|OC7{4J(k}&X`fa{<t!S}c+aJf+(`vV0RY4hvs8xIcx#zD|n%zol{X#l-(w5-G
zIqr&YUqr@fD^wTWeo)a9jvaCQuwg4qcfeY>^t#u)nEOcedJDN8>XGZWln3J12MJ+c
zUcgrspXP86B!ETklg-@W8FS>E!+ncW{GWu-nZ&qK>JPjGAAb^3d}7-YAbrY%E-!?V
znngCYAb;<z^)zeGxYZ#fZG*PXzklpX>_De-e@$3*maAMODdl6Dr|zcJHC(EjIPFAA
zWzFPT?E@Y6?*);c&Pj#p5<l!AkTOk5xEvg&@iOAVrMEq`rZaAC8OCqVH{A-*+zou*
zSHL~I7p@<W&D+Ptag#W2mU&i&&<|<xmJ~;k)$V8^CQTMm!lVAOsS<ey&GEP==+XXg
zJLV4Nnz|{RSXo<=fjMY2`T()C+KI2?vV2tRd_K{6#NuQ~w<=5n$Xh=Pw*0nklvl^a
zdDumbA!BcyriC`Na_K{fuYGk&xJYc@!*3b&o!?J_?N)0OGUM@4NwOTgA4$b?jh?+o
zASzxwaImnk*cSedU@t&|Gf&9~xjGrk5@$)(QJ{y^X9Ckkg%twbSN+wmowm6UF}iA)
z4xgU`BIhrQ#UeWjjFfDJTTGOpZ!eWy-M%raB|Gj_WVFqx7p8eN<>T8BXTknd>PIjE
zEOdzI^VKi!io!(~ZY@Y6=u)wk-@ap&wH146Kq~!X;y%7cB0-vAr-fiaiGzC5<>CR7
zf65tw+5~Pt_6qBv=p;+ui<A<3hZ@31R=XE<T@QgCMBD!Zp}zwQ*949KTwl|nb{K^+
z@T5a8Jy(RZ=Ki%n95fgYr|1-J>+Q!E>oj(mKK{?C2aaq)0)SYuY@Pqlu}EQjOo#kw
z_O+V7@A{4Uz1w=-sg+3NM2nc|1ZkO8LE#;8K}v%~KE~qvnlBxCZD@LYyNT_Mqekaq
zCD2uMX5HW9>_{IXm5oWK)O}|mGYnHW`S%T0=}xa4epwUnPPGd;tf89I1UOV=ul2A!
z9nq^TX=Br)d52?72`!Fx*a#?FfKD8g{-n^0MY`CoJ**z}kMFO|hzRvfV?oYVCVs95
zLp^FG7O()@jV2_Y;Cao19EP*DLc8Al^ac;Ll+o{1V&*n75CN|vv9*G70==dA2gtYP
z$C^9D6S=unzW19<d1vw+th<IwoW8s(kkCTk_HXkn>9wAGByvoKet_ZcE-!+2?N`}S
zcMCpdKI~K|i#?cv^6nHuzge9++Ft1?zbzwTxA#!DQ9IX`ypN{BtB=dzXuk|fwr}T-
zE&9e}rl$(oAly-(EW6{QdSb^myj6$s>z|6yy|UM0d#tK^?-$aWEn039<f;31I%7=f
z8=@06hCu2haMz?hl+!E0uC6CPWhoSxLX6auTJY&*m2+#2w-px7n9y)B<aG`u6x&Y8
z=ra8J@%gdX?~+r;Uquhk^6!Fkj$^cx@vdG$soh<@wqB97g0l|fP=Y23mKJX9r7tn3
ziM(p2EP%x1s${Duih|}SZf)=U#+BJ*_uwaCT1rp()NBWo)2*2o9w0kRqt;OH_Tgz;
zmws)Z(vwbXuo;}FVU#DsTl=5mq>S)+@PI_ifEC~Q+JgaLYA82l|HhwkkKgX<(F;pI
ziM7z*xv)?p-`P(d&!cv}@r}phzImKX>(~CC7aEf|BDFPBXMomL7;Wo-8RhP3xewf^
zZL2xj4c%3PqfJ|wa@uXon0~h#l7*)L6wLw)6@-9IOp?M(EVqXB1N2Jc>A`P_{k{vJ
zZ*?>$qe&<gd6{5P>O)QlTB??%&f^*{1qIu!T^ea(vG8k@Z_g1>p1&wfDdX_y+7roy
zJC73YL<Sv8Ow9%G_#AAYJq|YO5A5Gg`HEwH6_yK3JrLDftXewZA-jH$ldGhW;LQJH
zl?MrLaDa$YK%-HrEU|8sBn1!+6*z)MUD8-(2Qp-S!2l1)`DJF{F7<O?aVhNAckxNA
zDf_5`tvW(|M`yev3cb(AyqPvCm{ukjRyR;D*B|NhyX>HOY%Wec0BDR|_LC3jCiR>f
zbawmLzN)*6zgqt2w}Z>AVt$@BeadqBWZh_5-RO;##26N~AG>m7o_9P1U9R%#kTF=!
z`Ap<%Hl;dN;+2%VA<#9<-u*t(K+_*B_Wi&K-LXc#*D20!8QbkU-(30fY*HMzzeej9
zfbbh9a`lJrw=q?VZz<DPySLh09+9+t>0DB;a%bWoUg!qbQ6O`?$Ikxy7pXEY^!_=I
zfNdV$fUuHX>($@)%y%I|5o+8vTS#(t%d0Cc1F8-mWw?SAiMOBq`UX!q?51zMh??e8
zcb@?eBXZOhqqwgpFZm<Z$b|N2dI;C9fe9rL*3fpNSn}U?)!+NVb|;+R6TUDoI&#b%
z)f;6c`!JlYS4N}TuDd8iBAc+4ghp-9=ow#iv$BvOlK*?~InMfoyJ-PmCqT9lM1QGN
zaoYk764K>t;#Q@)x@wscP;(Z#NLoe7^}}>-i>CHo3WqE+jt=hWXy)af{V?tA(}J+j
z>+Vvff&q)<5+_@Z19G%`fEVKy)9$kLiou>Z-NuVVid&k;_6zJ$g#&h;maD?PTZSdY
z#F#WN)9a-B7mcW_xzLLrpC<RbNIx?NTcL<IRb9;xT|r&~*g!GcxZU#1Q!c#x=e*Q1
zC?m#u5F5!)-N(D{xnK0RSZk6Jz7*8k&GVJZI43=g%yBy?qlr8ud{iHSd+tmLo{0Z7
zLGy3aqqCrOwi+zBo2+lN(!B)Af+cu^nndkqw^I}nnQHbe#JUVyI47S$Bjt7aSrkBk
zj(YRzr?wt<Iw<coHP`QbQh;|59bK1i%HN-u58R45``(P@pJ3<v9)e@V+yMCT1p;YT
zQj00GhV>x4?M%PbB~JQ;M^u0JpzfUyVFf1Z`qMIy&4X&1w7g?SP#ZX|V;W(=qSg+2
zCV#W8&pzv`doJiyI3pucWHKqRbHf7BbLwq{esx;E1L`Qwtdz`mZ@7M!;&Gp=hi3Hn
z7}7m({*WBDFyAKV>zT;Wtc0;)iY4ej`4M_dBC9+p;k1z(Gx-(YBRp$jzz~)Z@L&_>
z4{oD4rt1jZ<f4;RCtbEpOhrxOck6xUW#}iU$DNGK(F+W^ThS;0txrB1jU~V1HP!6L
z&irV1zhF?<xKEZo(CFuWcvwenwFln8rWtQH#2nHXWF?68XX=@#Y?%9Y+1Rtn%y945
ztLHX!Tkn+s6G|Lj?CkKW&WYT;l~MsGpw1<Ob{O-S39Q&a?UWZpQrunCm7f;+F4*M4
z(?6hD(e?n8<cNG4DAphWHRMZUH}qO8DNZ-PV(OjTHXQ=S2{FB%M1d4B9-6-C-B?jf
zwkJ9}Ag>q=d}mL^E?C$IB^D|OPN}U&?$iT+vA~0esFC;nKVEwX?i&0s3#Hb6e!EEi
znnKzD28+0S{${`AdcVLmOw~%mke`s>Wo9pzO{{{^7AP7KVqY=!$khMV6x_*b7z3U2
z(^`$2?)_!P^|n*Y*iq0fI5jXd^7Cbk%`y7#7JnWFw)lp|px{4S{Ni6*oFIdiB}XCl
zY9XOoX(WyNpwZ-9M8Rf=;_Ku+)Q6diS77dR_Zsg~&6a@_nj@M~bl9j?r&=>$Tkpv$
zC9UZ&4t~<P>l6xGkLoi*H1ZQiOEchQ6D*gduXT{#C%jGkcx$l@Z0YuG=~!B8lzG2*
zdiSE1MLna5zUZL>%s>27iQ9Ta;Q7_gX%eGx<k7R*e2oto-oTG(RYc*d(saiIyBQBX
z!C#<=yBQoyV~N~aaO5D6FdX0Ef9G3%A#sd=byZ3-@_?uB{M*o#ohFe>;@mwB?#b=&
zW?7Weeqtl}HoxJlF!^>t?SMQ(O^0y3xk6S{$Au$JxZPqzER;6T$Y4n!XXZu9OR@dH
zr}SM4znhFX1f>GHSS;Ne>tJgG-x|rKhk^Z5pB+(g$jwVLH6zogM*pqIIMrXyS90{V
zj_r0p1a4*=VPlR^VBP6co}5dd59#Wb>HZUcK_v+EgSXpz_GsT+=${W~uHKw|a0alX
zwNbMa&seT69R;6@i+xshaD(_f=vy$;;+!O+8$*sM2(>4x*y2FZTTA~*2fujIoG$ck
zm-KgcKmqp_r-=2coCA0THczH$`qw@#;uKlwo)nAq7h{p0JMz5k9hTuIgArCDtAV&B
zz_E6vBx-KAZT8_G(S~0~rd7uxKnKRale`1vTkgf0B3l4~%*t&J%@_%C*X<7S(9E|~
zfZjQBee?AOK&j;;iHJfOn12i)M#=h>;5jU#Ibw&3gab4M6h>{a2cxvGV0du}BYM*d
z^YYxCyVn*NDFi#bBqxmXP!e}`;?&U}z?FExdgAw8r~cm=#HD3536!`i6{?Q@LrW%)
zM=vj3T)KApOpp~ZcS?=nEM?rRqvS+BJhg_5-nYnOUYLU&*^1}Ng?ADK?_%Ia10G;J
zZ1ap&>+3AKY_F4lrM1Yjy^I%mo|Vf~fBY4B^K*gKHUZeINNbf>f7I@TtjsjKOIQD1
zRIJB}jPQY{y*!uVZ>82yT;C$txci;Np3zpK=(P-|cKW_pkB)E-N<b+k-Dbh?cx6V9
ziu6NlljJ!30(#i}2_yaRLlRD<{{L-<8ygdks=j&qPlvYgrevdG$<0NRd{?E7yCbcJ
zwKu2E_NyR1r9Yeld~`EYvY@5OkGZSXTt9l{u-B`b!a5+Ns_~=Bg{>&!o6udQ^^a<%
zFlWdyKYwYmB*}t(;<{Yok_B6ec3Nd+AE7{mpA>o+X=@IvJsDk8&D!vreYLJ^uyN5t
z{|Ba_q>}b^#Fgu}1V26Y^c^RjSm`#uaK3NQa8z47nuG%a*}f$hzTlO5j)U*Kndybg
zb&CccRllLYX4@`UyHR<b64V)x%}4L@dRi};TpoF9U}EIyL$AV{F!~f7IyL}G0QXC2
zD?=wrg6bRLKFCSv*Fm_`O+m-h6vg{r^E*%xbCdG6y!oYlhTk~^M-xVkADLuY8zpc+
zXLoAHC9Tb-vLn=1O)k_Lk{`)>F5L3*+O+6gQ(teZDlVs}Q~io}@T;G^6rPr(r?b#{
z>E^qnpZ6axoaC>ae0oOPWfD9WDrl13&?<}aHz;q)X9`>E_D-mKr77Ik5BexQn7i42
zW5#Nx7jAL)*VtFTlB?EQzq4pEgMu*VM3#&B>t&chQ3q?!=SR$}xc|xJuM(%wHDi9N
zI3}m)+$q(FRjqNHdhs3(A=O;&-zeOynG+D`XN6{O+aO4~uOFr^-y8!4St@M8h>rF>
zC(sf0t{Jb}Y_o!Fa;yA!ivptrs=8^7{oT8G8$oL?@7x$0KM{%CjX78`QUB}7Ot-fp
zH3yL$rLkEjIJ`Zb{*M7z|IQ*~>4x7I0A===oWVr)q?wvZ=Q4FvlE+dZI=ni93(cp*
zQkU)qAo<=uH#34v-t3tm%6e^~lH%Bj<%?Zc?T)g{uORZe8e(U)_6SKfcA6?QkT#JN
zy4<Vx%WO<?g76}LU}3`r71};=UbnBj=N--X{H`J8jE2=#9n!K?MkO$!y%Wz(*-COv
zchJsJ+{i7L#*lv1wP(G%9%90@BcOZV3je{eSX0p<%7_SC>E8(c7;-?A-m=o_LxmRl
z-tg=R3C5v(t@pxn(`KoFnAt;tv{}(x=#`XCq4$_?P{?ZzD17%oZq!Ni^TR2JY*`0A
ztoKCU`-<29?3tIuySLBRdu#k1kesJXAtdIF>rswV8wA53i!gPHqHDJd?SF$(u+n;W
zN2|0njh(jj1K4#+ecex?t@ZnA>rvYMvCiAGnY6O|)IO$0^ZVZE!a95u1QK1{QKF=n
zF`EbF3|rX~^~<U~!A5Uss4c3Zu#2k29o^g2MMFl0DQx_<w^U{?-pT&1q%<9t6&qLt
z#*|?*7A0O7<sKi4SgBIDz1DcbNj~w(XhV$4z4|Q7B@iUKDfZkoG&NEBZ1}i2_ko9}
z)Ku4{vCj~bbgv&<XRXcYd!DQrJj6j<eKJ__H-_n3kKQfpYk3ho#uAQIAnJF3C!I1^
zLe0U2x}Sj=qMyS;Vf)_Ku4E4;>r@NEWU5F{@)1UE>CCFHN1LL(xD<OP^xjg-=ax6S
zb4}`-A{d9F7>4|`V%3QJ=@iYULu|w?gko=#Eb+K&CDhJbr|H)IS=5;q+1QWzP)(O|
z{>WB>vYRe|{^X8G@|@ri_21J(#4KBLN47I}NCO?k1l3VzGcjKLcP4qn1=wjSO5WIs
z(NbGzYrnjfsaieEmeQ{*x?UpF?5Co+N*c&KVux5boGu*tu~J9dA%kk)p7470T&rH~
zqJKsO(Y~c24WyNOv8lF12PLBe<`_@Mz-@EiuA@T%6+F$Jn?f;}qQFI)pWBwSxH2$B
z+P2|+TsX)yFGhjPD11D|Z+>7^|2S|;a*WU-ms#dhTRCxr5$V?r_a+xVmBnAZ{b>b`
zJVpKBY*lsi1%2p|d&MDRvp6`L&XN0g?XMqGd&0no(*R}l-i<V>vnEo9@{Q5{Q7}t}
zM>xvN`qmAFim>I$X72-GwQIO-@!X#-xA&VWuMD{27iquC4d$*L@vqC?!)o;ZUZ+2#
zKu1V~^Ufc;jPsPsL<E18k654U>&G!F%J-O?kWDfeqDN{y9cSB!f;dpmEE>w@+oD1V
zoGDpg_Sl_IUm{V8RVD6c&O95sz0GDvq-KUGTaRwTwoU2DFP^YIpJ~H^9rsg*W~41S
z*vqSt$DTI8%lgukFmpiEjuc%;7z}>LoM?S44Awgn?a_=fpLu=~rCneB8_-?{bHqMz
zc$GASvbh&)dL6pfMi)EJ^h^^Fd~+2my;YxG<w(b~JH2oNJ_wr>6|;AjSm&+()aD~Q
z-v=WTQ9?agfJ>DGz3h?pu1Hi(6jU9#psBZi5(0B;hGc}sb}i`c-B?IL=`W<0Y_W=U
zog+j(09LIhZ8sNIqOKXWKO?50-DaDg)6WWMw&`$zMjr$Js^!@le5Ug4>WK!_Q(jM!
zF4@E0l0I)OsF3Z0^e1!H22`x#gz#Fa6RL3g(F4rIuc4&sT+uqFGujZ+i}On?LHd19
zLp~7KQRr7ub!itWtYb|QieL+Q0VP|)Buw|7W7kLa(j{n;#z$rtZ~2MKk^5IDzau5E
zA*~zAk$70X;ab~u*v9t-!JLsKT4z*YcXSPxH&)8%#}OnzdYFVS1=QJ?)y8-mlWw1c
zH&_&#luGKa(F>RI{`7Y+VJ-HJ<1T_7F)rKaa6P5|I-CBIndwgTZ)1^(+;khVM@}&_
z)+sI`8P)y(AexVUI-N@QwG+CtaAWa$Sk>iQf5+X+ug?)kpqMME6yk^syAOq7rz#n}
zqL)uzU1c1f6G{4otU4?s_`tbkv$t*RD|lPP-H7<4+F@}VUBr2_*j7tc$g^w?2d~mr
z%D8+YVoi>iE4h!z+?xKmiUD1+?IX>8*V+2Z_M@gMn(Lik6FG0p`$u_D&@2K+i8Y7B
zeHLtB$%gI6t|Zp45?JsRE_FdU1x{(3KjW^xjn(wZY<UpGah$S+>UW37R>dDpKsg~x
z_b`v6sQRz}nE4Q|c4z!U$Ie*+(Zbx=D-W`<rEJ?K&&>r?Z~fPWQ&Kj0grB#vuSSn3
z10rYWwzvn@A21}twon2O8(_D`^Im|RB<}vLQuRO66TtNTEyzJGC>eF&9YVodAQd6-
z7k(Bn&;T;!>gf=TODEgZoHHcEg)*GPm(W<M_gtRarQmA)x_oh)4rYOS`;PXv=Z$CP
zZ{)mv15=ZqCR(6AEqo+4LbCBjtbmcB-Gu@f!Z*ppQGaEz%!cEmNifg3G8}_Ug(cQl
zl37Z3lIM(VXQq1_zoiJh=aoMZH`7CM<7wf<$-+F@X_`6mw`>YZ<!8FZ43oC`qxKv1
zSl>rIpz+!|?qrX>nV0Sy>2owkm*IK2XIM-$HY;!rJ9~;J8|%G<!SO9IS|4Xy!=mP3
zUib#;UN=+Ta|CTQ>V!T0&W57!kIoFEu0f?p!;_L?{B%y6!N%f{QrZqbm8MYv5exws
zE2Fzwu;vnwwzpkG65svIY?rQ@fyf~X+Pdfp52NFh%1ZLr?0R{5f<i5{iM}iTN_-->
zItO;!f=dPX3HwPXP<#hn$Ej^I_7rw9u7*(`LDrNR-*6;FLuU+jTc}$&jXXFogn1!U
z_(Iq~`vZl;!8BOk0}SO@Zkxl3udvY8BPq}8gu%f?<okh*pp)Av*A+5svUR`zSe|jO
z5yEcKT28X}QLXfZT_I+-jPGURp^W0ZYucl!=j7>nE7;qYPpY*)92*;EnPsLYkMZhS
z^JJ)bIQxH${kd}ZrZ1;!_q7muxqz~%>Du{QBN~_3t!c-j{F&d}Bv{FQ@{!5Su*LLn
z8Ag$BIu)KxS+Sq=*G)G1*}$prqXespFd{g$&k(!abN#piOv?;mRA;Bn;Nr{*(}~IF
znb(7k&u{jYyarRq{7v>06yK8zOPVh`37rD%1AZ_;k`*AkPAcOmq(92>j`@N5l+RH9
z=Yn(KIp=%QV!)MlFfb{B0YjyX?4Q|crBvFsDF>hY*?6&@5+TVhh@5Aql+LMEPU5e6
zplClwjQlJ+G2E!-trZ3OeO&ghEyTxPPd<tXw@Z$?Bke7^<9=lxg8T4-H3ZclMbTc)
z6=C&c@sK!zeA9F))HCKFV<YzrNzx@pA0@sUxgdXSq>w;D+;DMH3;@H;ob=A95<@-0
zR$4_b?kF}yJuFsNhe*Aur%*?%Tsvm)_faO6{l<=I8Cb8^IzGc~eq8o*+xL7&`}fTb
z_EtT9?%n8(7Wnh8K|KwD^MoR(kVQEj!$!}VMq@r$4Mi4<h}4M!T@xxqI4b!SDr81Z
zm+VbcNo=x}=aBTdA0rVZ#6L_xsm)I0nkjqpXSS}{FTYSBMAG{k8|2&ihO?p6=nFm*
z5=9CLPZIF9!~LXhZk%{YsiRdC$koP*hljjayQB$&Xn6_QPuS|`ZwX;Rbty9{3qPzR
zzKsYIzQ~Y$QXzp1;oedZzS11Yq8-jftnlmy&x&`Y=xa`DHNM~Lu;5D>&=3n^ZO_1`
zP`ANcVxO`-V{z_J&5NaLu8eby^0mHTi=cWwFe@&_64h1cgd$}MlF*AlB(KU`pR|JG
zr!gVzm=C?1Eq&=WNnK0URW?QV(ZVgl`?}gI{2kVG0=S_&nwQqY{e)qqPV01OVIv1%
zG{Vpb?m`A%nVE@pfw1x#eP)k_Qsa4;L~M89`WjDV&6K$-6OseZ((0+4*z7VFsV5RN
zgx*j$c;1MPIG<j9^%CFAdoG@iZqr|%<FR)QUa9rRx-x0yQaZ5D&(@v`IC|b^2ESoE
zF-kq6Mfg!JTQkP!uMNOP8uI){D6*r^2Rlr0z<|Gf6!P3Ff1e!-t&5HO9iamSKlF#M
z;TH6j!xEz`vNE%ZO~eA$JI)C&@r*oKw*(VqHmmV-_5Z|pfn4V|(`qm_ZW2x6#2+Yd
zm6sRp1{x*V&yF|~vci*H-{psgWyjgXs|v2{Zq+j%K;h&SY1xiN5WW%I_MCxi%jp$n
zo(gO-$M?AhFJ94GZgCR(Zpn*o^~7*SqH5GZ5ijag7hzT7{y1@enyyr}Y=Ay8#wgT6
z69G0rvU<Sflhra3K=@bj13HF77vlbj3JyfES6<8UMRe%SNn4o1pIvbOGwF$EGMP&e
z!^TZCD>hJ1WE<Xg+=DJMT5k!<*gfxi<DJ@e29@Bj-s~?b=Fc-@byck{`gwj5dovfF
z?=Q)y|8wPndxcJs;IWhAS3b>uCl!uk$Yho#)c1>|eDqxDQYEwoO|5sD;NNi$4nKK|
zb?Y1Ic&|Y7Nd@T`RZ?hrWVA|P+&&X^JKk7-$hw7nNg&49G(7jDr(C-vxweTAqS`$d
z?bSvCl@BiuZGf6zhU#=0IArG;8-8&fv7miNRALP4E01EpLLXgTymD8Ahjea<%bFab
zcgaWU7mYgh?3xPMwddrdZgU+1>XCsWDd%9L-TCd?tB8`V_8B1gzpO5C>RLjf5jbZ`
z3VlW*sB^7^8vAZK?7q5x6RPN^HlZ!OC^GQ#BwRIX@rKJ&q|S3#KnvjC&~fRTmcRn*
zn79QorTMmL<5A)gNq+AqJG<c&ApjLr&et#agL)z+0^28>;e=eNT7Pg^dG#2SiSF%`
z!dmLc_V;|uQm@DoLPG8@a_hU7BqzK+6oIN#ulSMdrx!7kDK3hYESRX=n|D(x@0vK@
z4h_GuPaxvpW=7xfvHxDv$B74iH%>oBNCrooUZ1EGkV>7t4#6&(z;r(4I|994d)t|w
zl^7WLDKcg{K?U%_j1BFNd>Bp^%nFTRt-%pxC)K5@rDp3AP>8VLYMP+}gGtoQBW5h|
z+R)<9g58=q(fys`^~1zugE1^SELJVq=`_p1AAI2;uWw|#Q&96O(6{?GF;{*FdhWYN
ztF37+-RDYrV*@)S8c|(+3cWH9{#p+CT6Ufb12JVf$?Ydv92${aTv1=~*eUAXy`P>3
zJ=?>E+s##`*XJ%HLVOedZltoPzC7mLG^oj3O#Q^^bSfAq$IB;M61(Iq1Sa~iL0<Lb
zix<1680YPOEJ{QC2qB$B5UGn0>%Cgq5!s?ZC@1M8`@+X~b@4GwPSob9SD)Z}-|EG8
zHbGvIUESHc{D=t8#f4Nc@wKGhR`GRW;5DyM5^K}7d0i!ec3q~y7(v@|kyc;B`*-z)
zgDu`t^8DHCqz0|=VE1XJ0n~69tIVqkxZ0OQ8l=D|(XBI-6@pAm0t;vId<DnPE8Cfw
z+zXZ~#>>VZ7WV^rEqEmSQZv$QY|defzx8ym0UARFExC+xDJO3o{NkMVwF$~ecQCSh
zd*8%7WeS})7FO3p7>oX02&{bl5hj>X@2~qKQ-V@roI?HN&~s*jLfcHi9z=^X{2qSp
zCIQ32OY~^+f#UB%mEI%G<?I+O&x7^{3=lhWX2l~HevGf)t>Z@C6rn?7ljI9~(YNa#
zmvLT1D>QP!G+%90O`)fpQlIXjRNp0OtsFnd<Dv>ZP=fzPdaF;mYV`hB1tDofQ6k0q
zuZ<7(ehL>c#k+(!mX<K`Os-Xd&suE$c6Jxu#y=ZSPEP)wzVDkf+Km}JenIr#$~bsb
zC;LkxiA}z$34754?~J{oD51L$+3Euih+M_4eh3G2C(?um8#A(5R>st>qB7I5*=nZG
z1wp*J{TqlQ=xMdi+RRc&<f)`B-bsGQ0r3P;%LLr57F78~McgMpONil8w>reA<}5IH
zXLUKw6a4g+vzTY9V*6NHNjarvkjir_#0&|>nwdXwzeooJ5oeO4f*q=PWfqDU2GAD_
zHZl;D8dIHIK`X*)m)Q<jgn(CbF%_jsxS_N;S@S+sUlmAu1*K}BZsh(yn#51s$rsb(
z<sxh-X52RBdPCyoR?M4SB$Br~!;{CdLpsEcAlYH&kA3sZnJd0}G@$(7iN9JSKp)45
zA1&L`)L^xa{ExGSkL`{OY>occ0A70c>P<h!Kl9IjBtTh$1dE4FEi6>dI1t`_;x@AQ
zAs&e1cXy`2ZD+sxUcF!GY09)m)ogJ0-$<2kN#M+YbBa|GmTYExC)Ci2I`~?0*@T&m
zwpoV?L93}D$SOYVpV$D#@jK*57HjHCxr(|k6Ac}f;IrahfyefS@L&r_w?7a9;P&xG
z+3?*LXhDv2M`?b5;5;fVTT4|S45H{q8F=Z8><T7qgN3EmBN|}J6f^Y2I|B>}f2_0J
z2zmB_=Z(~f9NgS6-<!*!EkT-<m#?^<=OJJJSY1;2&r24B{%#c=pdh~Z{s{?bG~>ID
zo!nUyj|w6?69hX1dlg+F-j~v`-3sRv39!AB3veT%e6T>Cb2k$lH1o%qewj{;9qm;$
z7N^-Z!v$-_jA>EbGjLYQT53vq)^3tHhGeTTPz{$&1Ebl6R8h?{r_xAR_{8^2MnENI
zau?Q%MIG`>w|Pd7s&DJ%PA%f_YFnd#9)|Y^g=5%Bwt$<y%Zk5c3@A5pA+r*)LE%`u
zjd_TT`n!v3H3;2nA+6c0^M);;D2Fv(lQs_V-Ni0eij{KSi||rX@yE3=*A`MhQ#=GV
za{2uNMAVP&g^U3ao}PDU?!U_1lEj>{#lKQ-|NHe*@g^jWmwMiwLY!EG_FfAutU`~=
z_)#pj7>2k1Jlc>U2=Rf=AIOqwD$(&uJGI4`8XXNLk6)(z5!SXQY$FYW1hpZ4z{&kd
zw#6)aQjB2*VuX)4-q(om#JT@le3w!{LXn-!iwWVse_o&0M42}*DvbBuz!Oh!<fG0)
zU^fJ9b;xN#SZ~g=MreeEoT0$g66B?5GoD0I8Pd?J3{BbMZQ5o^v5VYcb3J`{x9~^j
zRI4CU(9aM<rdLT6(O@9ViSFKW4fdY8h^A(cRzLvVXpk2uB^~iCW{7|2gT<5_1t)C-
zH!bpyqAz}M_6KVqpj0G_O}8m_5}o-@fmQAPXw~zk&Nu8Cj5E$8Fl%&^v(SB$$dS;J
z))iS!HOPmH<187TXYV&6-$qykd2M0iK7bC4)r?crzB<f7NY=q%BaoXyJLP-rqT5~k
z+<zcNI#lQKjbOnne-N&x5Obzf-%FwGa(&}>-4Zm3>nFO0)Fn2&N}5qsUS}%t<olVb
zZWRhbWA3&7r}WuX6SmESMojOLvDc4TDug4v`Zy3BD@4lzW!71CK|3kaTtK8E(3Gv{
zsoN`Rg#*?<*HwBZI5(qU^~)wBBv0$HeWE^MD{HFLlcP|FY?W~I9b|?$G6ECirTYIc
z_8#C^@BjaJA(8DaBe{+0rm~Zf+g_Djh-{ML77Dq|WJfe)kD@5b-g{+dgzT9R;kI}E
zpYL-{XM8^2^ZEU+>s+VXxz0K6_kF*{^Z9r_#_IW~dssV`=B84*kqz2w@t!V7teaix
zhGop-5#L9JVYekuY%v(%#yKXwZ2&4P23xUgaRjti<v&0A6NPfPs*N$LgSmpKxq~$I
zcx<*2N?F1VSBoWg-O91#j<G9T)L69RY{;r6W!cQXPZ&3B5C7u=M)+>*<QK+7AFn$7
zuK4-`L<l7pV0Q%9q;qvw%Z9#&)loFc(?|**1I9V)bSCQ{mp$m<#q5lWo_<$xW6Ltn
z3Sq2*Tg+xB+Nm>w^1N-)xW9K~$%*g5j2-Z)dSqo5mbXz_pZ@z=N6xAwz)~6f5U~TF
zuS|D<%yp1OQ2}DE*9XjVBhv^jRe5d8aS}J)LmNebY0LEH$9-vUd7F$l^L!aWRKV+Z
z1b;$OSrHSL6k}Ga+{Cts)^2^kHIU2R$?6Dxx9;25Cg)V&U!RHRp#=G5q*tF`6wdnq
z0~u5o>g+TLTdiGY2*IHbl9-R~tcrLO>dO2MkZ?`mhdw;R?`TR<9A|DlYcw*8ZV?%{
znS8&9i=RVRh{mLC9K<MZDNL6OZ0_lo0)jc{BT^1}8AIViD2l|{@0%T=>tn@@*sKG-
z5>p<Z63KJd#*3G-YN;uv_Rd75X4G3t4asv~ie;Y>q<Q|x$ss9peFifax;9ZLeZX`0
z2CAE_lbek)6ehzvA`wqo-BS4nVI=K?uh<!_>Hg_mK>qMS#{kwa<<gNF#JC0M0T%s%
zbjKgy<84;oPQ5-i1R|ee=aIf9Ofw7{`<%C<R(0a0VI;j?5w;Sy^2COAQJyv+(ecwo
zGbHeK@w4zFG8%_M{KYko)-NV9+>fegI!o{`otBJ8O2fGc$#yAT0ld9ch!jLIaWWrB
zJ-$F+|IjPSS`-=Wyz|3R1R?L6WER~RhE?hLQ6;lC?{C8hQP8n%#}<YDZn^hI=S;D`
ztlkJvY$>wGkuUw`$W(Z;6-G|;rbq8HKDzlJHnO_k_v(k4kNl+>gVW5WHT7BfQG^SD
zjL$FTt7oHsbz;A*))^@xSQI6Wf$6kGYY~KmRm1?gKtId=jT$=TlLWzmLe774N!!)P
z`RJ1IwHa@25Pe@OQ!ehu@Ls6vVn?T%*Z8oRhnD$n&H>%bxm`1GhqiG1Bl$~mqOx~@
z2xD_v2$%f=mqO0AkulsBEiDlY<C4R=6h5Oh;go7pVX*F3v~>}xLou@{=v<sXw7Z=o
zNqFP@b8n+mj6U;eH}euk#Lu2zBM&wAkd^@HSq|+U!F(k*4Zl|+t`}}I*KFSB$3&!B
zr4=X__B+2bfc~!JBgWL!%w#V*wm@U!ZXlibJwjl>$LZJJdF0JL2WzBv$7$3QYYUd_
znV<PFQup0Etkw=vQofsy=7K$u`EsB0$kLK(qtm{ZtP?7m$3qHe%WT(*%T#1vsqZX}
zpV$*7vMq~(+Ak6UtQG{Hu)VbgtDDjGY%vU*^K!-<(kp>KiFoj@LLE6LWRP>>#io<}
zKdL+ga{T9CWG^p_afz(a%akpzWqL^xmZ%?`l~Z5Wk?YPK_1wX4UX!~t(UfT|heOqh
zJ!70;u9La+jCwzoO^M@sl%NICxaQKG&u@8wD)(VQoKC~a#YkQ=SS^p<Q2l&fO2|{r
zNhMXruACOWIkc*V<XU&p&77aSAQshG0{ZG8r_v;`hVL9GCY*X?xBjeZRr^7M9-A|0
zlK&T$^UpW^`!mcCfuwoO<Txek`R>}oD07$RI=bI-iH{Fd*Rm#+fl@1(;R)HuawFni
zxN+x;cvx06ymxN^IEy7U&C9AJ;STVoO#)cB{t*k~`tUuILQAa81o@&bN}UWitgEkx
zEf4XZGP5VY+E%PqGsiYDHeMT`A$WcJ+StKH=N(V14QJWOI1kpE@267ebhLhbwBAxD
z0e4EA(3$!Jclvz~{rBG=YwV-31Nlv24dfLZE*NpOp^0|`HTX_FnG>0f7?>PRFTQ)t
zigE!v1fw!R7s8l~%W%b|yg4mp)*}VO&ITJt^%@+&h)Ra$*3lJC)rBUm#A8-#66Ca1
zHls)rp?%p`A>MW9c<Vf<!@9ypA9KjpEVuBF=Bb_?R=J<`mCR`r6u(LwkRACY3r2pT
zQ(V14g(I#l<&5f(`lt9vbjV&F5tx^kxLgAnm%dH9YpI7c$3t=DF!3oSBOjaO%#c4X
zr!<fTW}cRv=l`23_P-yRorslobMs4BEiBukP=cOh+<Ni>4CCj-IqahZ%;chgh#=9Z
zU;4*#kl8>LA#^67Sq%`3FNhh|sICOK;DKsB|Krjd-@iWL4jNcBa|avG`~G#7{kf$6
z{$i4@2(I_qD;knL7CoSSW&qoTSaeJN@1?v935Jmf7#6p|F2Y@YG9jZE9JYE1Q<26Q
zw|;%W&|}df+v{XO$t=(R<6-}LgMa@hI@XXkyN$WL|F0TThGaXoYG<*M4OGZyKMG&*
zV!PHbcbO<Bh|iYkNV?GeerHSHXGWx;X05D+ghVEA?I|IA)9`Z5Nl!ejop2wQggraY
z@vm>k3Hf%^W13ZdNjcdgt6KOgkNsFl{JB|zm|Y-xBJ&03d7fQRm>vNKgma80uL~3H
zDz<90uc$pY7XG026V3=K%Bp(dz16M&C<~9bsTWjwn4zE~^V~!Pw4^lXq0AvzPLBwP
zAo5$34$kdTKKVXK%1z%tcp(*}uHvo0GX17#t3ea4X0Kie!Xq^g<7ijY2L6go^W=oH
ztpf6!HH;M<7X5HDjPzk7mcWofl2(W-1zE)N!ljggn#O6#l>P2>ugifA_vpjw&((X0
zTdh_{TQI%k{mz~(&6wG_MMN(4o9G&$T_UqCHWxOY_XDF}kKDRcPW{=#PFPb!HWvO2
zw*tjv&&Qme?7Uw@;6K;&pD!MgfxC->Qp=y;mo*1rsevs3>gBU#tDjWHeQ2b}AZzV5
z6EHy=@L(A7#cZbIrMLS<vV%)PcVWG%(ba_Nj)96&V*1ztJ{b4l#J<o`5WoH91rT!f
z+PskdyE%$jnK!<VJU|t-*E|E7HcAwi`l`iev%xEAYWoNs@-$Sy=d@4Wqg7vkZ8MIj
zuz1d#71R`2g3rr;(Q~7e>WO2&YsVrCmIkQ2#}M(by4X003Bz1v{ZSiVa5dv?Ti=uf
z=Y|?+>OSj9Ie+i72mfqeSPo^`IK&|Z$fCBbsa&(%@<8Coto-3x8Nq&3c}Ns$S%13Y
zO1@ts4ERdFT^HpaY_q%Eui&xAI{gWRvvfrmR;O-AZrCxs)$@Zd^TL%`r>q6HWbYjE
zfjsyVD{(9T-R+3?p|73(?|lJ#L{kBV%$!2k_<pTqO^W?BaiM6*^H?|#&7j!kBb36T
zk!WMl&?VFQ0YYF`v<7xbp$JkThGGvVyiJ)DZpW^W)bRmXKGee0voFi+ndH-oEONOO
z&a4*K`*SvsVWKs5HNaNmt%$4yhg2!t>?~@|O(!FL0Y&nKlOaF*U2C08e;qiRpp_mS
z?IC>ED?q!n?+i^#_4ct<uYnudObhME6QHVQ5|Y{W-q0RwP}XMX0v8rq@cB#h1@`3-
zBS)46^Lj)ie2OBY&{~MOgIt>$=`902eu6s&DnkE8^!Jo!!(}*$%8Qn$wY7V3D`dN&
z(mz~&e&jO#`6&%pj6o^gbEd-^O2n3^yj$=m-uaWY{xFXLtIFJ;Yx6%}ya<FLqvG{E
z!(Utdg)@<hE;2KeJK}9ppeemsO;ZB~g#6x#-e$CJh;_PEqK7j>*Wq&}0ctp~AQ%TJ
zECuL_ckXAWC!=$MG-AR!#t=lVfvEC!7P)oTMM=bFjZiwe*Jc(6mFYOl1}PDHzD*DV
zFSZcU9!8~dBL$-w=x5h^T*yuyo&{il1<qO)6D2YGw#Zv4(J9=)n)#IdA<0p}gz9*I
zqqZsWw{7>#QQu0`n+xJI0&o+kqdK-kc7b4VkP_^54+Pj6LJQMx)7D-Tl~Di4>`WWE
z;ux%^{a%?BGel+iCK=rf_N2eg@qfH9{RFe1(N{gjr{Di^J*XnqPIH4(UF_>WH`V!w
zGo8BSMV6YxuLfvNJ{uyvo%YPtea>L?5Tc+h?{FCz1QKg45ONqimx4_;g%o;yvw1H4
zrf=Qutz1YauMJlIjKmJnQEz|--P&cji7VK7<Eu`~<Bx{s85)6(=BUK)!;)8Y<~U+E
zU-72r?4C&Vj!#QQBTqs#SCq=6XET%Fs>;Vx`chJd^sl?-9|fEDI$S)xQfF*`a|nV>
zw5LAyQFN4&&13ZmTO_Toul|WwoC#0Kw06b=*!&S~@BXk$xvfGfRl%2yIkBAhB^A!N
z2{<SgDtX-CtHEldH2F9wCPZ<!15r;@`Um2AApjd~XNjC@FqZLkUR795B}h7r*=k-~
zrU_M(e4|nNwo6t-?6O_6EVu5g#IR%6))M)LXf>ouJI+IZ8}+kky)$?|=DPK4!$nU#
z{*`e1fQ5@Y+%)YsdaPFFsm&WiPPhvVy`@<Tc2?YPR=o_-OWD24=Atg0Nh4V&Mm0QU
zub}yuNL8(_;C71QzI^Sc*?J<1e>!%k>sD&K$il#vLR4%cnhSj<flHG5>9xoCRLvKn
zvjy9pG)^EEPUHCJ%^8CI7bTBM{QRj<DpEb=^f`GD_h9?^k%x}SYL1s_*2Oo+e1=&V
zyPy>^)q|MW8rg%ahoM#CVemsdGI=U@m!;x2!Yb%@UK@!*H;G^&#gA&kEfKh+2wdY<
zoNam1ulXw}{gl>HYCYFc@G;-p#=e9t+3VvR%NM6ppGA##zPOJ7E82OuH_Xuk6b$A3
zU6#0{1>o(a&WOa`H~GA-S1&1Jx3S+Tcd|%OC!c)z*2a0gdy>rqB$o39ZppXvg47ou
z73n%8s+1Cx-R>?nv28qWz1v=Y_k!KqXF?r6e|};6Nsvw$7t`JLEG~2#)Wal1yHKl0
zS3(0*jE{I}H;&Kdc2FHS!VtcnZ;3fXs0@lc&z~l%$MnK8BYux#`+I5h`|DLnqytRI
z4*gZJu+o|Cn*w7$3aE*n(tLggCJOs$E{Kl0-z|b0uLdv!R-^$CfE&dhF64Ik+zjy8
zH47D=nsLCA-XUvhd_hYrPWG6uj3CEou?G@=N{|{xQ(&1~wwte3Ow@$Q`j8g8@W(Uz
z$X0HBw5tJ0z)>*6liu8iGB4YjP8^jKr>ma_icb#B;ny3%U0P!sVEE86_nX^8trBtv
zP$S@4jT{B<VJM6Tvydp;s1pmlvxU!O3lJ@}w}lB1EXTohXx&wSNr_nm)@@c3CRVI#
zb?h_Z4QUU6w5!2)xKhzKHoXs`3Dw02k_px$g5K`Rp~sjB&WcIp)F7oOnERBLF#pcx
z9_<cb-{hckdq+b}FBF~ix~~cvYQ_Zo6murZ**7v1{2`nDv=kc}QpolYFG63a+r-M?
z(vZ6FP#~1{Mb<WGV5c}69nlc>P7e3HqnpCJfP-1`)Z%rvXUm5HMqpG252!*N93vg~
zONH#Mr0yWKT`2mN3oJ~->#}JxjiEX(=8Rj-?kaFVyl2OFr^%c@ZUxJjLDxeAjYDGY
z&6Db?(}Nc!Kkb|y&QQy;HCmfdihbc}>Bl*I6BXa><AFGD2wZQZm97L5st%(=7+kid
znSJ8ZyhIhcL2H4BvMiR5bCAlkUK)X~SUr8%=9K*QKFc~oV%tHo$}j-T{VyC7Jn%qy
z+Yk5Pc&~mP#xNb6YrbYd9eqR8CT3A%Fk@xAQ=2^YltI|{AFv17TpjOdblQsSSCLzm
z0-_q#yFT{uSYVH|)aR51o!7syl4*^a>|Jomf`8K-6`nu1@TjB{bG}$~VuZRq*X8}!
ztpj1XhNUn*$`AlEiU9b|KW5q)QUp-5Ub>Q}8XrCmS2P*4GU=l>cfDlUs`lt>%And2
zLC7HvOOm|-Np<Hok+ckphES@>h80$}lMd}|xfE>2ICfb8WV+1|*M*h7)9wZKt0jcE
zFy<mF6R-o>!;itQT(VM1nEo3o_s=V*Bp9yW;U4tqUldPaG65#Pn-65UeXpcvR+I!!
zdqU5*%}zN^&%AkEW*I;Ra<Z4S;qk7sZzUBxf9JuoU6AYW52P5UBRZpSFpvy!2;O*D
zFI2X~VQle&8&jrcc&F*48I(4jhM|=JW|eFlB=x1kbBIiCLfNBE<G{YfPLNj!aG`a$
zc$id3KoS4uU?pM#K+&;X6HGc0ffR~-0t%Xey{Abxm~P5PHB>SBRFknY+xt*`gI#b}
zb}Z9t&t@aUTbYIAc_61<>AVOVNoqXpjmutjMwAKFb8xnAV7IqTN`3mkYXrABvQaT@
z8zK`RD_LX9c>-!1OT#*(b#-$?BBP~d!UCS*YE}*=lre?Z-7vklsZN4B_65q2dl`W$
zbZ=fsK8ynsUOqDO(sQ5nYUdy9Z+NGij+pn1L(RLkvX7-fT{(nleCK)#YFdKX<%7RC
zwa@!kE(skKE*;j-p-MYaAI8R(MWrWjKSV7PnbKuwHjtw*E+do+oGv&nr^{iuiunPO
zogS^Q+xg_PS9Q6;v##@^WZ%J^+7`6O`3L*9&tVrlV4C9Su)oDs<xECV5=Avwwoo#9
z=Sogv_t2cQ+u^!zLxo<g@E@N#dUJ5<XzN<skEfpY-(E*L8yn{4k}sX(^vqy0<i90K
z^6>p-IQxxy!^*b9LG}_RB%*ay^8D{d|H)o!-yecYTEx@mHk{sHHsV<c)#|}C<_^!9
z!tU(GaN86Pvp6NIg_n!Ub+5w*9bgAuKjSc5BDe`iogQ~>Jlg=9_6ysL{ZApdcuV#@
zF1Y!;Il{ZlR2+bN{}=pH@(xsrgn-n3IHB7TH<m*`25@1om>vhc+`;pPDw~@Rw;jS%
zJ&JT!C?g|XQV_dLBa#b7#naRMkG_(zH~H)sXY}GyPyDLFf87Ot)a@OXBskh@f6m2T
zw2((dg>JqIqEb33A`{-nu6ZR$EiU3a*m4AsKcQy{QEPRR@7i7sp}K@O4q$Z#8-Aa<
zx|>1dgGXbdvd;X*E2ltx*1O?zDF|Qso1Q)(u#WWU(YYwoa=g6?u8K!6j3y{(Fy?au
z^Q;+BMqQn0P|$aJ3Vo)lTbcOe3#~SW<|Ak5Q4S}kbdqj)jD+ByE^qhU*B+Tov1E`F
z=?91vMe%56D|N2X-+l*fpzlt4U`|87GTE6|iQgkwM02Irk&e5wI*C5<=NGLuHBVwU
zxlR1AiC%^0NmM{e3U>Pr>H5@@^m%#2F3ou10E6BOt;C|U?+W_NgX?@{g2;O(CQ?jR
zIG;xKhSm9o22o)|-dog@QlLW3CaQd$g*vUfM6xhmsIGGngZt9k2ib8J%&4jS_Lb}o
zGqskL8_j(RBR_b~#0HSl#lObqsj421?%U^e<+F(1>~t)7T1-S?%qau^va;p@g&tSO
zbhRq~mCt1gQ)bpS`u*!~nseT0#&_5Hp7G4cfK%@gPEB*H1qg9_^*OqzDY{1Hm<##Y
z%zI*A1)tot(C<`WwY)HGn<Oed-AZxA)nVMRXZ?-!;NeyU=B5$1#bIrwmBvYrgrH!S
zmST}kyF2l>xtyRZcW)UG+0iZcd=<u)n*KTY9a4GsxE%@|eGik_<^pBW0g^RtH`aHT
z!_MQ9VizBkRAZXgudeS`F*h%&Bl&i5d)v?U@r)1Nf1=-KI?DNSztH2<AXn51I$=^W
zb9cS4QcX25Ca(m2*94dF0fi!*8x{rj<#J@D>J0#&>dIa;C`x+bmBuPqSXxG&-5u_S
zxdUotHn!!h@{Mc%u|WU)v15^hAe8*)I_~&NLR&A$Zo#p4NTpfkBUoTWMs4KP)$AO(
z&3M2w@Bpu{4NbP`NGRDMx(|Y45l23}E>zy<eY2w`CxuZoC0X;y{u90qc=9Jc_%o_J
zliCm1UlnTKcsvh%;F@>(GxhUtEepU}M(2U%9E6@*5AYq@RmU#XrF@+_*XW1QT4lWI
z2`)t%t@6fk^DCq3q}gv%p;qN(A(H$~UEl9p9?ex!CwHE!UHYfHvy%nd4T-Nut<*fz
zYb>e1gBfd*Ny5Lz@hD>-n$o{K&Dj*y%3^GeKJo?EG785_xK=+l$>pXGl?agKEpHb9
zTiu5ObLZ+5508O=9M2^RmbC8VSX&|yOFwe`1j|7rlQO$&CI0i5-IaEY)+<`)(@)sx
zcagqjZo9hD-FtnsTd<TTUnQ)m?8{f)xonHJ2RY^$r{}c-Y_A;1=$vwR>#+Cbe4~@U
zRD0R$bJlopxtlh=a+Kvd=6rrol$!qF4QU0<cw^}ahulh?Ijzr_w9m=2k_Q?Vbz@)P
z;J35bK5$)~F4tle>HN5%G;E5p+&;5#9E$DZg_bMAq2iO6?ad>)dK~)Tw{@r2iZSAp
zytB#Fx1FsQe$t_8LNpygc-ysaWnO&;*e-G?31^l*+WR!PJY+Uv-W42^f|rs%(HNk2
zOA9|VgFsOx$wfb}{z-@WM^gn%b}4|8olIiiDEvC#eD>9dl}+c<_hvuy8(H&ap&0Ex
z9#+Ku8*zD-Sj7RKZh|#tU#%%Mv1^huEe@zC^FBW7eU-CJFYi(yOHx3P?qGICD?8FP
zP7^6$uj(_UM|io6=F-+saga8c+f!*g)VI@d5xrjQr*b1#>HJibMTzlPz=ce1!F@rd
z!Wy(e0q04{5#?^gv2k~RCdAmV{;yEPnUo9`!Ghy&X(A}imB!hLUy!MS;BV;NS3Bld
z?&dc*v=-J8+<O=;qvOEPvn|)&r}bfuW#jjSTuVPHo!QQhBKt#fYJ7RxbLO@DEZ6(K
zZKbVUm_Bz(Bl5^yom;SRSw0)ae!G&~D~na;xRKhf;<ldqm{2i{jQt8?-lVC8=i46g
zi%RdBQ`UM7U+Dj@LtyplP-NFMr~4f5BQ|NN;+dpi*TL4_$W!vIw3(AGhST~~VIEk6
zC9HQOE@B?|r=FQCh3!9<4L=+_3%zUbsK&LMqmv+kTn~G>d7reCjV&tARd*8aM-xUU
z`>1>o+e=WbyUT28@$=omd7QZLzX8w!_R#XPh^?Rh-6`}wS)STHvtwTKCww(;y_9iI
z<R%;s(JLi?c{IUwS;4Iziy%2gdK@hEk@IilCB`#p9qDdwf{kPAjZuq=R(6&KF08;y
zE4Q6km}Qr@-Lk_yORlo(cp)npmBPrL*Mi#I11HoCXN<O<9S^xm@E)!4`B)LaMz6!5
zV(EkL1<Wf!;zr*~Ea$Al`NL}B1TE^X-)oQOw?Gv$_toU!>x4$&R#w!<AWWX?B%ble
z=6#@a>Wpfm$aT>^1=h93&o-Ci`Oi$H-3k3{IY=Y9`Z%~3Ov)G<JOL`%+6B+$=vKjo
zV;9T#jtg))x|Y7Jw2i%^Rj;1Ornr`NEi6YwD<7@1cuLH(5IW8?gJ|dnx|31fCaDER
z0cx64x64ImNn^S1@>W0*v1@#!{h?EekmXp>(C7$^4ai?EAQo1Oi_<YSe1&S;KM}*X
zXei+f4(&D`6hxc3Z|w`(gGdD}nx^1@n^K&3AS~RG-<NBrnwsF6HfjUuFhghtrp3NK
z7EAfb#sKk4{{{*UUo<`a&=O7jSIJ!xRRvCXjIAs3SskU-axn}M-M$Xhr0zAKX+$G#
zphNRu+*f9bR(e2A{vc@hPGJ*YUY1rhi8Rsz-p0GHpXK-Z&B<f4C)4u?n6%HQyn7Qs
zUjD$36L&2kUa;Mvn#8%{C-@)cq{mLFWdW;DXR|^22c7ESzC1pOT<|^i2f&f9Pe#{S
zXMDFW+oE>0OJi{^(S5nelqk;>Em{`pz{NsK0p?=10va6EH8vkUjKk=fDtr+D8ywGE
zOAs*Nxp3eqSB)jgcVcR$q**~<E-ELVFGFM6YzFFUW=UCEX)pweV}<c~h@TeN>*U|M
zf}%knSdtYcdY%({RZ2?GnwnapiG2JHQ|0X!xk(gnO4OOM_qT!~NZJSQvAS{{7_Lnp
z*C9Q{IU=rQ{uP`GkX)sv{T5s(NN_x=j#*ig)ue?hnrkKRaBkXW{b>R^bDgIbFW!px
zQ-X*YktoqJ0q8B(udTuz5NR8GVPw9e?BW4b+d-7_&Kdz$8##RBcLl;Od3p%#zo3Wz
z`n(pf_fDJgz8Alp8=?ANx2u2P-&+5MaJ;!3mbP*DSfzpOgPiBK5V9_AY<wCj9Bi}k
z!SwEA>ev_5!>)GP1(Io5tu`tV&%b42RHMc`((t4->F*1!X+^w_j66jg9ox~db9CV?
zFJ+eWD50$CY$vF{VO@^<+4tyN-I6TEJKO4`jmKji^V_SP<<G33*=V0y8%UflO3<5G
zG)<tOSQ#&_tNTz9M!x;P?>iMkjaU?y41c1BCQ8-OxTsS1zwwdZ{|P0n_d+*on)aV-
z_7;%~Np*5GRn;rTjVKYpaJRIoK~1e3*;At!*F&cV_{w*Yd~SYuk$umN<0RB>$~Iqm
zlZs11yZjUoG<MeM!gA!3{d7gl(0JQpPkm9Wt+QO<O)mU0%^F}S5;gUGn^S?GqyN(g
zQE+pIHQvkhZFO*9kQ$2R0#4_Zaqk|KNscLS1)JOfcSJndG?>MP1(9n@dZBbmlrOPp
ztDAPKpSD0LZax2){P12@Ni%Sh^<&{nxLrWr7rurK|JqL_qNOaEvlgR0-QsF9M>^=A
zAEY`&j54EL;<ib;Qi++UwgZ`_Dp`w4;I@#5jn|+EAPIK1mbOXOk^sfpyI+P1@at)t
ztG}Hi-<`59NzUd3{dtN^c|QHI3YA8~VQUr3=Bn~D=<Tf9AQTrvgtwb%Sr+3-xe{=Q
z94t;ph9{HIn_o~ho#k*LAC6TECth^N#YU>GMjrFrggh;&>y3}g)bdpS4eh;8&aJSx
zPxL<Zud865R))!Hl<36N6AnDMani5Y_vW$qhy)Vp0xU$rao9&#@hu7@jYBK6>F|tk
z0nw+MW53YnIDwC0`|yd+J^~er#2~!m;c!39Sq&1kbE1i>H9|hPsbMaT$?tHcB-qF4
z1_@N_bC{Zt?=t1dk2wQb;{;{%nr=bTosDUBk$L4P!2t=4)@a^jJa?1rU2w^=wH=wq
zc1PjgFB>I#tm_ww*T^>z%=)pxQ!RZ8MtE&HnjRhhGIuEXpSwEds#T@D6X1Y7Youcq
zE$Qr9VJ-Oyuh!gU`N2?UAenE_e}g07*aT-N&=ndLTc^yLW6j|IdJ+2Yh)FoPvMDZ;
z4G-K-vqo9MEucdud!#2F*@@6jXqTr39s_)Z+n|3P!G_S5nLawRT9ImwbLCoa(?tJx
zIDNdX3$a&Th6C1|+AoMEaZ~f<1%=3wY*pr>hH0NDT{d1Vk(CRBm&fi+1aV>9?R!o~
zGs<`wPi{X|LneDU<NhodcLkkrmkMExujJ!ye`72B2YwQU;3vJ9(@*|@>Q~wC3*4TH
zwo<<VQwp|yIRcl0{92sMj?_r`?i=0yT<C5ZD4U9tiz5K$z0GFqO-oS$rml}H9gK9v
zm{!d0BOTjaJNdwcKGAaCxl=R=Y*HJ=?I-(%*!8|Jnd}Vo8s_;4d@rorv!&V3N^Hnb
z-(0SNq=ZTO-ad9V;@ppGVgp0llR=YAWd>l|?~z3#Yc#@+-ntf;RwqZr#ef#%w0Aga
z>CgIrGOFHjt>@~_9qh4`Vs`c1jwj^i)w<;4Cw#mnC3_9;o^*LXcOqXRkX8IP_@<0t
zy2Xudhq4yR(&EA4cG|VQ^m|UrQ^3Ln#mDXw8=A(v=L#_mIe3JEYCb`JdDhPgRTir0
zHE<R~u@vY2^^A$bes<zm{Hchu1H_e<xChwuDew3Cn>hZT$TS!2+#=XF1yqI$JlA6e
z1pv>mBz+vbY~70i39qZ8k6K;HE87RRe8~|BXPDFpo1O2!_i=}K`mFA#0=hSReGm*k
z-)|gQ#RCP^s`hTWU;e%Jd_?uJk*<g~dfOs1;EWYbGnQV)<)qn0W(S?0)%u#iX@qc|
znlFFIm^F=_dXkU<NL~*hd7-Y|mZ)&I7tj~>V46o*kV)YW2b!HF*-P6$Sld=pu__!7
zzij!2*?u}Tpja}>dp9wtP*>{OdGHRS|L(JEo<lqQV^`R9^`=)nP=Ff7i%-Y8Fc_M8
zAD~Y-*N2I3d}fsk7vrYW<n~;qM&|mhveLVW9^fIrS&lC`hLh3XEV=<kplYO4N4qV&
zeu?1!i&(<sLUcuo#I5OXE#n6g;B6IWHF5^~+PQyWsaBAyR?w+d*jHV)1`a%j(Qe-b
z`S<Qs?hFgy`m)gH1ltv>Ni*HI_+e6HYP?$m*aM@~N_4xhZE1%b#Q~Nf>uU^FJXrfa
zu<-Gr(msH4^h4s)^jxY<i7l&1TOasMQDGiZfUAm|7*JGv$UnPZy?qXR@x4ta%CkTR
zbCBg773L@fCW!)rqnMiK@2k3+Rh?!!Q7z!UNig+MsV42)6-6hStP46EXX6%TVC=Di
zd?kuA@AIHDms)N-ZU(3v4MvHAAT*w2XTeccMk!Ksgs*t=si~PuEfOt_D4C=g8p`39
z)dC>Eire>5ABje-xDE4HGq8=M!rVSIrI@X?09-r~#0ejE_iZxF?bQ;|LkTga@#J+!
zRZPVlXFBTLOHuF;8i;YR4;8jx=qt#+DPTSulIsb6;I&AuRK<2rCNen4%ud8_lVOA@
zV{thb`@k$L{)qEUqWpT1JYp=4y>oC0(!|$Gx++8jj<wn(-WBfjE36Ds@xi~ueP`pk
zd5|J;YTgu0$+C;f>RGXp7$DwhPXQp}61uyqE~#1}#8{W5iV7LuWL0<Gr+O-;%;WA{
z1f%;}DDrC5M28RK#-yUIM&q1fro=VGty-U5k23RTj`^G~(UqartrG@=PnNuO5fJ7$
zgvQ+LwiPdT2Y>(7ALq*7K@B7HY0*Hh*EB)BOEF<0<>M7yfiUH2tkD6~Wya5ygSf#u
zf*Cw74Q9;VKzJ!PG9I6?vf6}E*pNDI@YcJ)w)sl$+HnKXsq_J7m{(=brYWax91iD5
z@h%8TbJk&OX*|AX00?3FZwTR=S`7Dw%m|B9E`ksWX~mC5B9b|<W_2bT1#9gG9>QPS
zPhnoIOKk*5@0w-Nggo%jrf+R2{P~IW)#(l5pDi4|z~*U4G=b{z+eQmk+5>!js<EX0
z*%l3?JDM$tD%>>fho?TCyS`X_6~<!(;34<|EPEW`lo~576y7?sP^ebLh@h@65xKPO
z)pZY+2tQiNxi(gjL5#cCf{s<pEzES4)6*Thg)?HL?J+}M|74W@E=v6E?~aU!cYeC0
z_R&Aao{cAI4|pQx_cR0SujIP}$SBNe&4YOvoQE8g88g|nTQCYbfsb8P7)nmDr=7k8
z)G}?Es#S2688{f;dgftp80{v<d13wsd^dw@Fw3$-z8wpQi9L?4n?p{4EU+=8`YpHi
zbpu}-%1;J?Z_po1HN&d2^Qtl%lzvw3^B#u+{W~saQ18vKDO&P3C#Ce(;LG=8{oXGx
zBh3Y0mKf|JmKrvN+X#*M_?eZ8!>!K}+ed--Vm9irvGxt&f*%NNjfxDZ7sh|Eb+7G~
zu5+g(L~l?s-Z6rajTxu_HEcUat^;^ZP!5Sw*EeJ5R3gP5w;`2h-fg9365_sqzs0ih
zdD`b`UaH@TyW>;^`%DH{1lxc)_)O}ZMr>-GJz|4R_H=)96wk5zL2763eL`-yg7jKU
z+NAeK_Ihu5RWjXd$-46=NG>qA_8q+YMI`V#FONms$Gz8vdnXC$xoBcr6lILT%+10E
z!>njOhLU#0jBg)?tx@6G)YU2_8P8@NOU*NG=i0g)$4$pvs0^DC#$=#)erLC;KMH24
zgaOduS8a33$>uNx?I{?{4=Mr2AW6l71>##Mo%u8Xo7#vS?6d>aC?}zk7*yceP;#GC
zNNf&Ja3Aw&eB^d+#eaI^C$^SAhhj@3;GU`LEF3hK3Zy8eNjc`u?W4x|{bcm?ZT?t1
z`G(uIh_AzU7lA-C9)E9S$?Z}?;^qb49diZmK-%J~Q~kcvC!{!=FbOPQ0;t`EYxm0i
zCY-2;Qt!M!Q5o+%aLS(dA)B@3N@v->(8H>#=)l>2tZ%OI7X&2Yoma(Qj*>e{a;N-c
z+2VL~MxpQY!#u{qvUz5inzJP6dBXaAMi~m`BM8t+D>-1f@ue@^yY9u&k%P=w1@6EA
zEcSz-W<DG-rB*T|&iEkNVmQ<XKltJ2x>}fu3;^3hkgB&0bwMa%pDZ#)x=oJ82m99=
z@5tcl_Zg{DHh$Pn=^ZBz^Om3=InorXz%oVNtrs9eU(GtmsVX=0gsc1dpzhmYSsdw+
zxA)GEzo%^v5)?Fbfk~zaiKok!%p3i1!g;>{`GKddXJwA!CteEhZnsWVS8<~4#H7An
z;A*}fZ_qfkI#KV)Dx!6ghV%kmyLr2ygSxrfH<7wv3~`pA;HEW4;TPP39hXmEf5<?7
zP{o@D@T8z5ylJfRkRbPVbe2IHxNYqtviY^wFZZPw30^Ji7s1g*8i*+__I~QMCt)x&
z=<0ofGoXBBdnKZx$g}ReZvMQpZ4&>i)_MZxXYn7@uhcaCPdoQ~6p<Ys0A`j^Sel4n
z2OxshD4m5zo0<nWSe<HfN2jHK;(r$T(wy6@;^x;W2CXZENFY64F3R|0v9i+R!sgTI
zn=AX#CzfWEn@gvs^nI-OGMHW3ZWu_dPT&u+Ofy>pG}^fL7!%>+x)SHpN7MdkD0NAe
znr(tG<P#vh{L2&WziNURbp#%?Dt+^-+v`w(u&>^{CpS`Dh$x#s2((q1veQ=ketIQ;
z-}^H+!?CkBRK>PulFtOpJXcdaKc9H@GD$`!@##4jd-rESQFTgT@n^t{qKlfCJW2lQ
zz4vu>c=vqV7+JUNpw<{yEK`xld)LVlkSffKys=KF-2LG+%0lJ-NSo+iXd(OQxHW(_
zwcMH)@T+E)K!*6NTPN@`Wp@O`(6uaI#23NQAm+rJ&}#<Q?l)TF&|wi~zo-QY;Oxql
z`~z+;_y*yUP}alwG-y|{Z>>OiP5%n;C8IwL+lbMD`RPv~ETV0-;8I0KqvUe#Jh)#B
z1#3w&DhA9wmKiSKjh)HjD7je(?>g<<`94LeAv3Mn3enjtrUA8oY16<CAsn$ocxZC)
z52mlTIg#&?6T6@O;=?*3d|2v0X0l)jV5Klxy*PK}m^U#Ewxd?6{vpbF72u4}-0gT0
zva$Td^+7wk)88PRraSwpP`?7I`Qjb_jy5tUaxSJWaE&`w0E8wn3B@1upH=~^xRti^
z4?gUR+k+8mj&=rx208EyUuMtu@$D@PX8>(&;r-i7n130Bu@ij&Q1Q%-<a2)<8v-v#
zq~I<_7zqB6#;#3@o`-LMKmM8!Z_(HT#G%~`O44H;LbGFkA_xC}UU(?O_gi6+9`^?U
z@r95#fGuYcPMqX&G(?BhKzdA<MKQy_;T8YqKUbB4Hy`UhaQWOHkzD_^^#21W|BoM|
zLVy-2%6u**jq_hWpa1oA0$a3j=q}$<=~4LCHS!N0*FQT&1ydn})^)u*D(&IF9P59p
zGX9TGplVVcKGvpGULDlG?yUcK<bV9=-Gor>o@rU6o&H~s`R6d|zg|3}fo+OOEYMc_
zUyu18PyM$aF~{L#Qd4^LHto-+3+EyL6Ia3=qEgI|vOn?1YHEws3(4(}Duf6_fZ%B^
z_x0RFD&RRfC9*0r2g`WTYMOMhg*#FHzdwYiUE~|z%71J9Yoz9(GzPU5Rba-aThOCZ
zBavyYohQ!i0VB}&Alilf9q_l!02P0x-EVMY9Xyq(^Lil^?!$mzFvK=^x<Dnw5|9Yg
zgxXXHjH-7xLLsbk0Pq-V0Kz?%mD?nF@*=%4q*w`nTdp7Tji{G4X*)nsKF6*ed<%{T
z8ldPX$Hn8kzRe-AZ8D{=x*#2aF(t-2Z8fPN<zP-G{>vlL8Uzqax#N5-EJf8i3K>?_
zBIE)kb)-9qx)B2mdJl2$`<qa?-FdPBSt2Hk^coWLc_WEP?ib(gLd%yki*00(I(=s3
z#`odgy-r93W(Dp`G6BhJ$*o!MGWKxrN?A!xJgO1!zqXUN2z-RSq$g~D?U`Y@`?6C=
zItO33jJ*%p#M#6x)lHz9PJ0EOKoZq9^c?`(?;sFrX+)Qm1#*DF>qks2_9FY8G=?^k
zV>3-&Tg1FB&>lQIeIX4?@ECN~ds@4m;`InM3!HY2s>6d20O-#DYynEe_hyrgPZo@5
z=(^#HW1Rxe)oWq0HxMAC#ty`!GDa24&ojFdyoQ-;CSbK#H9WXlcNjY0*NImUG`rV@
zcj$KMd<{q>COxQHBj8R$qT0G2WzH?T0g$nGOpZq-$%C+tu=LOAg`ZlLSwFXG2lS?Q
zuhEvXLu~~}jtG+^fDi~2{++YedHsD3Rc|Jc-=#xfK!n}-p6l&gZ!iNIo*5$Dwhz#G
zBsX*9*|E{}!QcU?Uxb5;cE##(P<(sbInlNSl8!COUcz|o&LoqQ<+Y8y=(ezuYf67f
zO8(p&f4z{AgrhG#Ao<R(FhVJ=`z~NEi9W;#@nSB3VGa2lMf+Dkcu@cn5rhMk-1sLI
zek!n6xigO*H2=y696^LNR^M;`PwH?c7n8j-h|X)qs(ozeJOo|BsumHl%xDFlm)9gs
zGJIes4ZZf(#vwMG9N4#p?*kO=?}4G7KSDM|?uc<va=bHAxZu6?2FZu~>~{=FYkluD
z6WmIx$-z{UaQ-5hOXM}{QybjtpuY4ko{C9-hO-_$7-mhD2(moZ_Sh}6TK^u;xZ5qX
z{AOg=>1?0Q+!Ifvt#F($b70>&s(=FmbUxat5XCtHcp*1(kBncf+<xwco~NOC0(7*c
zImV<-BuI=WuS0I(E!F-lgbYpFkA~gmboJ_1kIelYcC7it*XK61pfxd;&ydU=V?xAy
zo1Y6SzRGnZ7&x{+Em>@L^mNo1Ogr(v&c}Ul2wwj)@{OF9U11Wy6ADIKI%>vDITuU$
zMn7dq(w9p!0rx7ANJ5@6Le1XgGc}b9*$wEu(*!s?rfa|*nJJ@q>ga~EZhlhWSQ|(@
zky1;1G;PbN_6OMwUFc)8K!&gpP4IR#e+H2|`ibPb6)W+JMUo~E@4|O|p0wb7a~3Rd
zKg7F!Z0miOue~L-b52jjfN9XH#oebKW|YVFK&mV{9_Cz%v8r?}UIZUD5^R9eX~JAY
zMT)iLc3O5rC@YIt0NJ^LUPw2l-9HbDW`2Zt3!!(h=+wrtt;GA9-@cv{B*j$~Ks9u5
z&9924dM%@LVRqxF=&<7zJ@DAvDU)(Szi39*@p%I|{6CbK|8b-J_c}f)GcycDkZ>!w
zGLJtzlR*Z&e)lgffj`NfY(rL8ucX(Sl~~tl;>JgZ!sDuJjuF+Dka(V?8C{D#eo4j=
zrZ!hg8eYQ2ENgEb>Lsd_I^)7-8PcnOrPig}9DjQk;AC<*xGUU3;3Mqcev>zEb;f)2
zr=C-~K*nCg_XXG0)T%{r)1U;r>8acOL6|!=b9S#A$-Uq|;JNx`$cE!bo$)844!*aZ
zE#}(A(p_7f^zU{5@F@C^hx`4<s{2TlAd@xs;*Uk78M<%8f|dj7QNHhGD{aLRY_2me
z$6>s(ED+#QbmkIlN@Ik1_M1-T2ZUO=ijB6rPo3S*(Fsz}Z%M;<A`2z%huo_Ms>~U7
zr4yE6euXD$IP2G3)S6xG<t8sk0Rf{dptZKEAL_=TWW&Vm?o=WUMuU^ha(i93;IO0I
zeSR@MOW}S%s^aV+J3V+465eIz+#N8cYm1EJrLp{pq@umFWz;j>&lppkT-be4vhgf0
zLH_<W0O%$6a)EbTH5Xt+H*h5L*?X*6Cqv%FbYZxP0$|KtVUNNwv&b?9D06LI-f-D<
z7JDpCB(eO0TJ`ws3(1=lBn5q^)}kSGshsE5V14Nzavsv0jk&71emrx8o+Fb+;=6!!
zA`O;MH6p@q7$(_S(7kszamNt3<qj3v<|i3>n+zAIvDDptF7kSpkjzS^kr0_ByB-`g
zKc1jA8BqRe=h*h6zKs6)=tYGw#+Jmk%;5@uPmXo+RaBgOc=Wc}z0QhCxc0)W2i0ev
z@Oh<nt*)$kxExf8WJAZL&V!^eZu;BZHH2P?a8TSl4S*gTn?}+*lEI-S;V9rUY{n9o
z7=Fe$_}dpTk=~mFWj*49O<Q!q%}ciq{>Tz+e!OtjO@fh|--Dy_1n!wY!)^}G=SR0@
z>q?SJvZ#)|q48Qh9#^=lMLl(jZQ4WR^#RYN^Ow-!$47ci4J=U&b?N*g7W~DN^c$1C
z253Y6EV~S=?8=Fw8pUrV)-|LJ!^bom&hbo*F6`+KRm8T1>0y3ZWc{DZ?>?Ek98`%K
zdN(iqQKXhcRLyiI*X}=fj23}qk(kq|h!v12c`ylGzrap;Jz>+YdIzSE)shSd2Y?1#
zM;_8Bw1a@@HxLjihw=a?-s8BN_x5eUQBRqg=vhKzf40#BMBILvfbBnWEd!2uMTo2e
zY4Js9Bi{b3_d2c=La={N=0X@L{Cv&kfC8?6^(((u|BCfV-4iDk;o|S?dtU919j5co
zi`Vu$IfRH4tSSMuE-I8%Ac2A>5x9tSqMy)^Os8|-AzDFikZo^9**2&)t=mUP(VLBD
zd=EU`-;Gy?I;VJ4tmkLqS*Jl*Dz_gZ%8_Bcf8S{%ljaM?Fk!u?XPe*iDg9^{*%?}q
ziQ%V@dos({V}j*qpYs5WYzEn0%Xir6anEUcc408NT2OU^{N5umU(Ezx><H8jWpVNW
z)KnC_d5^6=-~%Mh3gOI)q`)Fvx}n+*5#$zGC|Q-O(L_CnU7_c{OlH)KQNTbs5l<0S
zvHF@H-;X5x<@9f^gMxQ9ZyqUZ`NHcYzQIxV-H@(9GZOIJ?gQ<8+PcERV6GtGFri>a
zQ|D@rc)=JFo@HP|kZfJXi4Nz$$DxpRE0X~8KQqJU75*!m0C+p?r2bS6QLGGP>_U3D
z9n!j62$g#9-Wdn;67~4*<K<plU46%=#lLdB+C}cgqII79?#rHP$v+hvM@}o)^{ilx
z18&;kW<ML)t<iqzw?VJBol7&)ve!o&+7a?pDu3e@Ylhme?Hw7pwN%x7ZUO3L9!j;5
zZ(DRN&E@;nrb|}ftZEu0t_bWzN-P-@#l)+Qtcq32TYU!;$Yv2Ty>#La;3T*Xqg`h>
z3iUJAV3hNiqCMA`Ff*9wI(rI4TcIhBXp1|i5r+3|n!u@<u5b5p!S4-9mpg^LZH%Gq
z#WE(|S9P|7aV%CVO>K8W9p+7?2hv2Juv;n!KfG|HfG{o<Vk6FGWj{ZD3=!Y2Yf72)
z6s!o@Zv5a`804K+<_Y_o9rWKilRx{J7(}tsDu4Hnv%BLt2`qqu`!~s69zvHh@!XJ+
zjHRVM9E4@cs^4Mr*$Ol!lL)X%hayZ63&r*R>ME#`zHu^Q59U7`V3kqo6NDkl{l{Sz
z9KP3p(a#cKdOy*1T-JY?5HQ9OPwSa=%ue1nOdcD-Rd1#lFT0MllNOZrB`0GsmphEa
zmhg{>zlZJd6=+6t4RY^-T=vNxRHgN#i#bw_)G-U3Q4rQT{J~=WP}!I$Lao&!-hTz1
z=(q=)#678sW(3xDOmG(M>iEYnU$IBh-$@pqB8=;d;ZF=mjzE|Q7fqQU$6+ulrMixC
zQeyIDKR`#yaYS5ddjC75;hC0=T?81=4;@bX+z^3+0gJh95NJs74HB&!-JKHUW3R)|
z#Tde<5t+2CC9#uXzlL8a8>YR$PR~QX<nUwsNLqZ|R*?PEy8$`jYJG>zO}9ztr&-LC
z$~=<`%Lb2Z_E-*PR&1vQ97|1<xo|O-b6xf@DGiBeIJqnx?vvAbD6ebbDEeZH=mCow
z4|3?}zVj16Wv1paPXmVbj-Hk-S3_%7f$8e&%xZ=PU{!nW&6$*X+^yhXx&}LSFYdvH
zFFDHax}RR~-OPBn+6I~mZ?d$pYzu!+yz2<P-pHcE0{y3>4t$)31BdAt=M&e)t~KD%
znet@s#YVryW+i^le_%s4a>ort!q`(-__gm?jc~K9ms(|5-;=L~6xqskae2IhmRFvy
z9G~(ty4=7$lbq=zNOWs~j^H04Bjynwy0hGp?*ij(he3IsAf}_8rzQp<;CZ0^V%w>0
zvxv?G!<A2}-(qk2AiQZ$yZd{0>}o_No!}JP^dSHG07X=Vdq86wr1t}E<Mq4y5(24<
z?Kr(M<`MIpFe&!ftF8%BzHamjK?H`O(+)ffb~A28Q<oXG?aJuyxDPzr?qI8BkDUc?
z(S+7QH!p^b`G}R{5z3YStH@<1VkVk8qWE{tbl<)^<`ckku0XN8QCHo3hW)v_5)|}^
zo+VHNcfZJzN>wlN9h4|R_zxFbleNuj9;D3M62t`s_05gnJ%#wll-}KmHin}@f}Lh@
z^J~3zx^**Zszq0d=Ajirp$;6*5MA<~HbaLVDm+!1q$)XUUi?vH&_z=9O;pt5q_>Tj
z(<-VR(NXrN1RXTGN_5X+Fa#{`q^0>pkZvr-sk;^{yS=1Z_9TicqmJZ)T0`CKkIC1}
z6K3dVZzVlWS(|S){n8ggzI#VhH1lE*zT#4AvIYF7=)ovcjCX#}3DgRGRMh>P29YVr
zQ-X`OLHIzG3J{lZO+9*hiv>8`X?4CV#u0GT=iKs$8$aeJ&w2~d3R*;4rG@4s^R|}8
zPw$uB4jsWI&B#n0IV(tWM={-vC)0ee?*|miF;R_Tg4IUmXo=Y6)WvOJbjN9?AWrBz
zRVh?B|Ft*?4A%NxQrLBLsRq{o>!5(y4OfQ+;dNpal*!i&J1ev33jbZzz?p_T`{WVj
z*rN1r0vMvZrF&DNmHl~TJ*up0H=lbsvrfMz?|X`bkqFZ1c3I<=q0n2gc@~shmeGGc
zN1k;Wzg-17s9d%d68~=>_Jlqb`+AtC_WvMrL(-8^D~{71>IE`jVm*1$WEW(Cr+XvJ
z%*|V_14FYNEp=~%B)RQV&BIdOn8y-=*?JZRsCyU;Zoz>smaKnA8%3D<{M@szCV^`T
zoYuxoj#sYMAxV%t-(tnp&pW8+ChU--4y0aPJ8wJUUak&?q~DzkwhwP#1LpLaxPvmv
zn`>QAF$aLidPTKLr$Oh|GMou&<I6pEFk%!J=KY_D9f5qm+{?f0j#P<AXeFb1vq<Pj
z9-JcW<i*`GPv~-gk+9!wB|O~V8B%<Fa8u+w<t%s8`wiujL_AN+N3PPW@B5Y~*$L^v
zCpozlz*~hcxM!L!g@Z^or0cdErT@MrEYn(v3hP=Rk#wEKlukAwDp1fW%8ZoJ=J0Lp
ztRzUUa0tPBT;d^lMPW>lV{g_3+MDPo<>Wf7&U#-fR_wg!3Zd7+$aVjkOzVa6mCNx&
z(!N%Cukp3b655$9v_u?ng6;SCKEKSI5q9Dv%ykz)k#dbuG#&)2yumichxc}~HPa*u
z&T5z>G$2~Z`tx;}2%&_2Wvb_dqp+Pz5Z=KC*K3_bZD!anb@tFM?dj-^PeT*)U=BmR
zRS-M-M!B>>?r{(+M9}fu`ISaT3Nv0aU(rWGHSpVG#HUP!?ELu1&eEijPfXx)Xmf+`
z_f?8WuU}!ERu<79X(cn=3VtBJ5*VWwIQTuKctvZh!Ktc-Vn_L4F-r==oyu$^vtu~s
zU_*Dz)Wv3tmk*4N4U|oJ-Hl@;%cW}q1JC6?ZT**yhws*w7AkmIpyl?qZ@k2_P<HKb
z9Akq@)b)wRhM#SNPx!pegy$~x<kl0ufWNs~{a?4W6bAsRTs*=j|5zQ*NcQ_`GaTaU
zw=EoY9~+d@+h!L>x6*b9>LseTei8Ry>v#G#A64);OiJ_7W_rA~^FI7##?784dT+Lg
zyrjBp0=7H0<5aoy2s7Q;vAy>?Y3XRKvEEpL(3~+@dNb?8kMmBOxEh`DW|L#ba<Kxr
zi`<yoV9yA|O>Lx^`u%wT*%xO$x5&Rl_Kqj+y7^1<yQ1k=#+1cxseFIcb)AFLC?qg7
z1IwXRy5;=aH++kNM8ahj*OTzhz&z;WP)g3;(Fu-i48_L48}q}eoORVZr^sbd2}Fb0
zCIy-DOjirqLx?J&f=l3>#FUj3EW5@GzYTT0zQPmJFv#OPWFV~+%RzE|@Lg4A#tPP`
z-utkC5mX&PepVYcLL&xrfo6+=tKSS}ta1`C7X5Q#(v87k`^6ZNRIuygFT3RiN*eSc
z^f!&9MJF5#_LRKYrdg)7CBA`d&>>cF^Rq#YKu;62>fxo#M3pzR-eG@v0U*l5d#n_6
z=Pj-<+$pq=Ejto;q(Afdroe*o=)?O%HLRrpmK1ltqlt?X<upTQ!bCp09=H5iku|TE
zk>jB@6gKLOmHza#<8QV;e*s^A_Bd5sBq`>fjE(;?rBh8JfnuxJ)cD2&D$|1$ACKZp
zMWWbKwPK-LWsiztr(_Y~(c2uZUrr*Y@MBaJL$w;{cVZrWoucE7YDhp*ghe9qW3n_K
znE{AE&5$Je4tyVqV-tWA`y2HYBLfzp#$h$!Q~mvGu7GRtV7O7=ZOPf|w?>$VZLQG$
zhk|WhlE;4lMVx+-n(KjiQ!{#>ptCg~0ZlaS5jMIi$wP_j13K3Ozjstn8yR&8=A6w-
z#~V;EI1B0()qJ0=!J}UKBE}R^8%?H3pHt>1mUxBDvj|;kb2(@Gp7~v(#K@6jM69pb
z8V{8jP8H|e!uLlbcwCeoCaSO~r*yq<r0!WYw(S6F+8>e2HEjPdG#x)rJRL(GJa`^=
zWY8D_r5^ZoQ8J+OdMzY@2C#oeV85C`n77)XT}~6GTF#MlrHx6|@vy4>%wS9-MvsYL
zT9ag2b9!-(Cj>aQRcbraAEO?h(rfH6<#6@EziJ(P)oVGweoio1ub$<05r<}KS+_xA
z^l<9$%TUxow+fwQhAY-N50Yh{pm3j{gx$EX!Ou<!4ZEA=;KWb^)SH><bbO0S=4|9Q
zz(Vnf>wuh-P&CZ--rN>I$sR<#J3wJI<j5N4m-c0M%sq=xWCkt|)xI9pzEv=_G^R$u
zw~a@u-e**G{I+L~b?(gc)1$^<DKRM&Q{#Czb5p)Dq=_S$b)JhkK-3X^f99+_%cnvE
zmyw}iRd>eLGufkT(Fe<iY?m4Cc*b(Ki$`(QBIcD2CPLs^V>-g<gMUwnxlE18q5!1C
zE=zL)@l>Xox>cZD*~wVIN}*ej8MLz`(J`%D!V~j{H~PQKqBhwOw2=IdQcHlGb`t;|
zbgM)NO%>yj4Zv$0WjlZfNv{;b)Cre_QSnXF)tOGIeZzWM0^S7b>S|yQ=x2T43T~lh
z*O&!6FnwJ*S&vUdAFl)oeRMPF0}coAx@I^qa1L{^(+kpgC<d&2o$@}V7u9E)kOr<@
zxMEvZ%QN1Iex_7Y<9wj9^!=<Zck{`SyV*Zg&wiXFs&ZcD9!OF5HaTp@F)KH8?(tjj
z2?KA0R}?1M<IY_`6*sDTHiD+(+IxYQqhOd7W)C?ke&@Q3y?+#WHS>->_a{^1Wuls?
zk;^y~`L65nwWe)w3*oUhX4h7j=|0U-ra%%0HNI0qYS=IpQ3})_XR(mUTaCR$x_w>2
zd`gfns#lb8lLM<b6sHcg2if=HAMEVP7Xn|<(y?&r*Lm_o>zGB`=ZWTQT;PR^F;>9r
zfk`UuHprE&%UP4`J%GUq_711)C<Yp^Dhz1p@&}6;Hek6Gu2ljY-WFC7u&%h<sLVt*
z{&DbzMIoF!QTYCy+#<G0_;ueFFQmfg#c}=r@jcG??nCeXjnN;u@8v3)thG-aFU7^E
zODiD}QOfx#^JE?um@df$uQQ-~e9r{o7cYV__2Rj&TEbUENDL)7T*<%oogO;zl5#$Q
zOHj~2EBg|^T_M^j=fvvl3t69Q5zp<D%KZvdtz^Vx`)CV_k?TiLArnuHa}uMWvQ21z
z3V<D7b$#<IT@Kgm1E}N(>dGidsh3U&?s3Ts*Oc2vTRvsmJoriBN&iGd443-}j|<Pc
zF9wdN+YZ)6k~f^O;p*RxE>xPn2RQM`sh&@^?PWG65jbbC6|8Qft3E3TkWb4?ybbLr
zfIXg#h2I{8{gQ;P#Dv)NTiv(^rWFX^_hU;^Awrhd$_wJ39=JZiH5!ri`kQtyKB4XI
z3%SSUHkhe1Wel-1?U@^^&sX>NYp?B&ZMN{_h9p|RTm=LvQPXkb6))qj*Icy|tr;(f
z6=PIeeIAwC?|iRmIFz;==};HWGteK6v@!3eEjLW;Z!bH>FvWT`icCn*?j_hDGnQ6-
z36-En-Eh^rv7$bZSHX{yK)?8{U@V;7lD%mzC*4q)^tkJrh4r=nIjenf&bz}k|MQ=K
zE~W5((o4)Y3kZI^e%zEVuIUxm$xW?u3nfV_oJDlQVNMSFiGNCOHEGbV+<uPCMI0Pj
zMstW8r>Gj{?!YPkb;>3Ioch*!FTayx9WygCySp-T-7MO*`$G;tiDw{l>zOML1UhO6
z1qwi42a`hV*vUL2Eg?Il|Bt=5j*7bN+J;FffuTc`kZw>)=@5_{a_9j>lnw_(x}?RR
z8x(06Qb1xzkyNBhLO{Ah@;&Feuj{(+=X;;a=l$NbzW=_p_+yp>!koYJJonzmK8|C5
z1}%Z4V$d5=N~Pp?X4_Vfw46(!c;TFJLAV#r(|cFH6nSL_PhL*Z65l3@{8QzZc!Hd0
zU@X@u*|J!|-PkKXRG*F(ediw8;Pu-o-Uj%SnU#6P#Juj4tP9FxGXQEjHKQjPm9;7o
zi5PR@wBL}R;q`H>00Jj=YWk@n<x9(qZgL{S3|_$OjqoDmE%b<8MBp!U?~72vL7McH
z)F}M8y;XxF&0lE+SR<+Snxs$q7|a@i=PvD`kEwg7O!ym`JZ)vZ+`z~+5i9)&sQR^*
zK#;zXswX7_dHZ$rI&cb-+I>O98(r-WEdQ}&PB<pK-^PgE7dI~oad{~%rlmM)h$VCE
zlIi;6m2wStO4(f?xPJrMv_*>^OJra7ZfVtMOdf<fUNU;yVtEYLHtpcOqKtpZqS5`5
zJzm|W`{jdtNV=w;W~8v<YZcYU5H|*;+w?bu7q?X835nz0D6)q2KH#so7&sT-syRFv
znfX5BUD9Of**qFwJ@Rp8=ScH34dFkqz}@9h=K9Mere6Bl{!hzuT!pii&NwdJvyQzC
z9F>57{$j$zqC8&2h%x%}o3c0wWEnzZ-$BM@nP@0gXR(FaeIdoNU&tSF5d_soQt>=n
zO&HKU^1cxwbP+%bJS_!k?xcu1(sPdlQ6XcliLsxz;r&40r(7hDbt2v|l?DcE11qij
zcG$(yzQD9)=1aCh3F$FYv|B3{FM}NFIq8-hk3~KW3&uX#{Cgy%nh|Q06Z6vqba|wZ
zBnyj!S($JwA;tIEAn=M>pq;O(jB@tds+~uQjSLd&KOB?-`y|p<hF)cwFHXG{KI>s!
z95h8yD>#1n%4Px2sJQjEPY%{>5xC2u6TuRJzp_eM`<^A(XMVJ5LIp7v^D>)hGeiiB
zI*iS>(2t{0n}VKui6SeJ)fm(e%qt~JmM`84aLT09lH<zfM;xLe6YHO<7vB7ahaX`#
zTxhcpEP^LZ`@vNHJ(i4^Ku)h?xEFvs@hL7*C+11{c=lIXm+Qbh&`r<MSft)_p^ghL
zaNz15X_>u#cCa@3q4COm8glj|sD4T&F0i5rs52B9W^g}E#M2<k<Sszu`L^0_OjLyD
z3TC9<A|sx7ts1V7pQI|q-v_8Q4g(qDt~u%Y7hd6q)}`{>(bd>ZPe~3%ULal|);r0m
z$)Gfcsx-@`7l(Dy?FdB9PuGRc@hOs#P+zR&A83ChyE&276O=RVnI`b3eHm79cJR^t
z7?6C9x)TMQsn1ZX>t{QaA+mQI%$UlW&zKYtpB(nE_Q<dt<5oz!V)w9f#PRFSm&1xI
z%-SZOkrWYabT-A*w7A)BJ8X|qH5iW8v4?~}q{m9(6TrM{tjO$R&H1M6-80PdPkz~(
zt-2RGiVW`8)^HK0mo}%Ty+TTK&=EZGV7%G|#zV`r$5&Zpv3~RQ72hFv{lvdP_$3Cx
zzrNzMvAC(ptAE_I+qkF4%C=ggrI*EW^2&UFp*$bs-h0h-?l<fj<W!wqz&wgKo?D;$
z%QeIv588DyKU4_1b~|UD5z3tYQ`t9J%sC@Jmj_A|And-R{*|;s^%pcY(oqo?p}NCO
zTUE2aIenMQyrf#S@!7%a`Ry#mm${%BoWr7IU+ITKeV)X-TGO8a#q&$ak7ouNG~43q
zKRpGrGVQ)9gW<I|R|$tEN*MP_OsUShKnQz$q?X99(M@kClY=}13<RYT{fPQuO2kKA
zyX?MBMO?FpnnIM)P-uRY{ix>9Mp1xs<-<OQ2J#4<sc3qU`=w1Mc2K!Q;OxPd;{?%t
zNBOy$kz0=f(Z~@+%|}kP6<&Xbs+y9+F1dSE-|(yG3-04euTWPdv(-Z(8xx&I?YXc%
z-)aoj>3Rgu%d3012SvlG(7E*dTs5d`biTNb-adR!wj;VmTCa<n`yLqr*Eov-(eY?Q
zVX|CW&xeZ}St*R{M7s<^UxhDz1U?yxp^3QN7{R4xAG?9jf_atPM32_?@fVV4iJ4~L
zR-Th1yfT=p7I?;wwV_YsH(038lPu%!HzD-IE68wL9N6Abd>y0EW8mRFRBN|+ru+6D
z@yeI%;CvLHV?ymx4Ix*q`*MGdJXz`~cLiB>y)6UMA|K;EyH*D8?7NX)KR{ha<VZ6T
zQ;$*IFlH?%0MQeC;C|(Z^Q}gzXsrezHxnho%W-e45!IBlsp81+^#+Zg7KzlpcG0H}
zrD=OXTcnd7#LiJ)E+M~KZgNYT5$o_9U`Z<>MFsu8Z6ueP)#i|{3Qr!oum3W+GdrcN
z=2+qL%S{5H*zdfLHgj@J?>6h9Y}PQnnGX^B3MFEMOALpHZj0SUd-Jem@mt`=vR<@&
zQJ^w)h)w0q&a6$doD)>>uD14u;JvX?V6&nHdOK<k>iT1xcKQQ{^u4+WFMb|_VhEMf
zggEZni+bsy+>$=WG^(*%`n??7b~U#U6IG9M{8ixgT1rNT-WcA>+h@?8*Foh9D%Jek
z{MXf}XpJw01oXuD#taWg$)a&rJJV`Gf#A5+L2GGVTi)sy>5E|<j>z)A&D8S!x_GAE
z7tS#m!DVf{lR&8^WLAzO)D=0~3#)U>?fazXMI!M7a;<En#AGrlCq~>Ki8uS<MdY20
z4}`G%co?)NTvbUaeG+;^KbllKg12;Kj2T|0gZcK!>d<HXHl9BZ>EEaIzr2urfy24o
zL&i|~>R*Rh7D|S~dx((ZhCG#J3W?o<*Z^%SP;Kh=c(R}Ac)npbC7@`}E`gv5aM_@1
zlI(=-%HBSYo+n*p4By%DI^Ldh1C2hapY+lfZ;2syY{;3G0$libWuLxF=r7)YRLiMg
z3lT}zD`32*_Eet@Xn+)$fwnwuBqnw>AqOM{j3EvQ>dH&_8jU-)zw%9a$jnt|ruKv^
zl&#(Kcy@zSZYR$8RLBxr`s}9cK1QICHtnoH+qA-$fY>#Nlq_p4u=Dz8h+A5Bh|L12
zZi<7*$T1c2pDRvNPSnabV7XFvdH|lPkkGYu?v?iExb#QO;jK7+D$jof*AdT=FL5r3
zo)BkqF5xI%87|bn<w_RfX~`mnG!TmvJz^>hMdEugMJh<Z+#xSzY2)t~V=S#&8E5#N
zJPv${yGOo9VU1-;-z^Dl5{)SWOkEf%DIZsIp}^qXJKxho%g=-{ZS`n{N(^4Rg5CJn
zCGBiF1@Z^n+fnXL3VOj3?-P_tlxT^18Ka5CgV}H_A;%#R5wBEBk|#2-%Gnh~`QhIK
zm`E4%LmR#zLhEoa<4l72oyVwrgf2<NX9L>T^1X>qA}%bQK4L({R$b3KrVK{g*!$fc
z>ue-%N|h>s{(L(8#V^c2iW8h2o)MgrfJVk;@=fMv3IWFq)>q_eiJ+qr{p5&^%}6yb
zDbStLc$6e87edpMz(>Im9ac-ZM}*}U&36*(`bnA;Q*cRi$p98!A)WSx981cj^Q3bN
zKV)^dDD^2FmQyc92yzdP<uSfC5fQ{!K|7T@ix{FN0}h$T!`XEVqUu7swd?ZQ>wyrb
zyDhi7TN&@n9=((~UTbxxFD6ct<A+}c{Jef@&yKm#DudtDhBIAy6@OrQixi)9=h{_k
z{8ucipQM9t(VePp$8wm`L6<y+8au0*OAj!a&p?@B*)bmr6&z^mH#HO}`%Mr7ZzYpu
zp#*Mesw@2&>^kma7^K0et}q5#VqRrD8b;*#-VPPVV60%iufld^TkM8G4dCk8Q24jB
z(o1hHEL2v$UesD^R=&3VRq72!cGRV{k07a{$gwyn8Tp``4^9j5CK(>jJ^W)@l5%rc
z032t8(DD@l<LK8|<7&HPZ89sx=1@NW3-QxeL~8h4f_@hmJP-QoY3G@*!jNz|c~cxB
zt=Tt^1F9v%35JN@MQ{=IBs$74!2)I>R=07&N8q@G)bVf;0_tI=nU<$$r4oU?bZGU>
zbtxPsS`(P_9kAkUuUY5eJCEmDLq$IT?WuLD7DT~vLp;hSqMgf0ungHLhzGq9^57@l
z66Hv8Wkv$a#~3<Ab(l4+y^EY*_&HN+hlp9guy_^KNDn-nFH#U|BS3}FF<|j_Z>1^L
zF`iPs<zl<k2cZF`8;H~EICq>Ruibvc<X4nvxe*8_9O*~a1o?hCLEm+^cxAI9g+jFR
zkr;A1NrHE}&9UcbaP%NswHW1`t2;zCmV7+|R`);H*1JV2%!bu}<%s4`bP{i7nrN?Y
zl|gXW$s5}us9JCBi>#~nAky1X=nwN+u||q0pwtPXiAS4~czDIG`*<u?+o-hXp)$pF
zt@3dPm_!K3ZR^Z7MzXBt<Ld#94D+|s2#`OYLp&~Kr#k#BV5k|Bu{%n1;#y4HIM>48
zT9(kBGAWkQ#=^HRHC?vg>0<+Z-|^nH)CcMhwH~?{-WZ|6c@Oclws`(Uh<COZwMAU4
zb%u#uo9#PB^X0}_lLRO~ZcX&KgRA(Eel$;f1aCw<Bmv?Obgg}};1KrzdP9Oen)@6h
zgcy&}2U!;wf+Ap?P_qx{5UynpWg^1*=tBZgMqx2cIm!StSUQuGi9vNHoxp1*FiXXN
zB(`*$F23m-37Jxu_+4o*2>-NY{MAx-wY`~_Qn*V@f$R9DU`WPNC!M$O)S2)*4)7Q*
ze&Vobz`T0ftsb6`|4R7h!UfbcC~$TLUd_h;`<r&dWiKzz`$Sl9+>qUR6IjJXEm+q~
z=UeM3cv)=hyb7ughO23OpR@7duc9sWqIsq38ccRqi^I@769Zyi`<8Sug%xJOhrP6^
zx{ri-`C8*SO_g<L4WD(s<l}pWh3xbMQ>Ky;3b41<Dq8T(sXF#6WqJj#jbuahRj!aT
zVgB=1;8T3<SsGI11o?lbwQP|n0BqPA)Sv9H+;mAceT^OB<Kq64)TQsPb>~%gA#Wgn
z>jG4m^Rm6&8~m|&q&IkN(-8?qa9*zT29PV@@JpEm@N58CX05Wh8YU)eBlx(JaTC~q
zHPG~(oc4H=cZH&E#!6ZRjR9x0JBjNi`ZsDhHzz-o8EpmUCk6g{;r^Qvkx3d<h8>??
zv;7&A{|L=LWZ}C%KC|Boi899V<TETy15AGjVIV{#x9;N)+Y}=DGtU2z3IsXVXYgCO
z`HZVx?k@H$`}&W|{qyo*l5^e%zmJN|apbS75PflK2VfMpp+iB1*WOb5M{fPXB$M{@
zqwg|gpRWDxKl{%oF*X5K{zG+)_^H~zm;U#+@_%_(n;ZPRA@}o$7}noB-+zAV|Ns5}
zpR7N)J<b5_cit-lO;Q}Ud0<E0<&Kpq>i8PK5ckI`Y;1sDvukgr!u!8G&Ht{d%uUC@
z^~?3e8Tb45p6YsqI{NM}0Okl*t^-Z2L;}RW28>M8V`_5EHm!5hskYNUO8PJgq-pxn
zmuK!2KYracs(h#c%+4Lppmg80k_6>|PK9*kTb@>@*SyLx@G$5GKOd*37w5-H6*t!g
z%KMv1-}&{54EiQ0qk~aAyaAK-h&A#uz&zwY_X7#rH$c`Kal0i|yc<8^^T8Y&^Jj!P
zpHYP(HJ>pnpGi&JQeP@Zd|fW%{nnGQLEi_z{>KlxLk_YH^9<gYKTp0D2SzqwyG#pm
zqh>OJ)6xLxYdP@qlR)zsOFWMuiFaQuFMfN=F1SzLNVqR`0q(xtd@gU{cPmviviJ+E
zz762yRNYCr{g`E;GiK82AU-ZAQNV)#{PZZXFICLs4G=hqRSF&gjxHMvi0Q&dwPyfH
zC&O_>WyQI6?g}Wi%^l9Wa)lN?#U|3-ddOO#nJTJrv2}T&CAe&sB<)+j4s4P)g;ogX
z)_~K_8;EI*-SGYA#Md6W@>{Nrl-R6~m2Z}#L|zC7g}V8+S4soHEm!arV)EYMmS8+Q
zA?L5luaVO8KW7aseA%dY4B`L<N&3ZnO}8-9!%LuJr#Ovf!~!E+EoK-yy#M{||MT|v
z^L1cJ64)aqJF}H*|61kTU=*0p@Boyv5uNkFg1pb<Fn|)>lYmz(J^|zFKx-(G1`KLE
zVX8Aphu2e}n|FU8LuwJ2g(qYyge^I6@oU}_u^SSEwt+`%-!V_O(Q}0KJAXzIC5I7(
zwCW&4^Mrd>O-t<=O1+pVuTtPUt7hD??N`S?oByfI@7U2K0Y?QWjsSO16%e~H>BL>K
zIxI<*7?uoS0SArfmYZb+5>%x%-&sgnr9y5<J?na>`lhwQ#FW7XhiLgE+}0QFCp#PF
zOFEi!mCe#uaYNuR9<W6Qi-SkXEM)9lO&l-4{{Oy|MPa!R>5F};o2Qxn4piSJeAhhE
zejhlJ!AgCWJXZYkwq<*PrgP#NG(eb!lJ9z!M(Ch9!`<uSmARlOVJqz`pdTtR5i_cB
zNJbgcDcr_#MTR65gTsWSrtaK)Uh~dfjJ~8gdHH!j0iZRaA#0lqE>hhYKml<Bc&#$|
zak}5h$}WI)#xkI$;eV`;KQMaHU<z)4C|rVFx(kefJ_7*waYxI+tuY5M;fgEDc?_9d
zLc|LE<ZDfVQq=8>9NuUDi#^63cc5mW<|c`Hd(|JiPWs{<8Nz&Sy#X`=)!&U*+Z)Qv
zvoyv|#IiURmsJ4&FzwZ$f}1egu`)Who7;feIDTV3Wct{e1hzj3wgc{Fo(0I(ja7Py
z`P~0Aj{cYO7erRs8FU@6^*ru6%O8QH%N7f?k%Ab&j4olX-3N7NyFH6QiI4ysZeg>f
zA8zB|a{;9ZAK{r5ZE-DZ1~}Qep>qW=_lA6=&+3tRzmhg;pQC<|`;36Z$MpbPejbg%
zrC^IS6d9BVtK~zgSkUf~yP%Od=397~2JW0)AVMFrhg}AXlmO|HKIkf|MtK9>IxfHa
zw@~r!?ZknD9gdj!6vK=1@P($TL+id2kqTfzt20aUiS^im4mso5O2Cn{P3vFl3J5rp
z8MtBx{{x7}Dhpcn|0?-02;#`GGL}CBTLRmW+Sw7}kw&IFExy&@xH|~U>Kp@3FI5$z
zqdCALWb<jMxeZHsZF&-jy=`1M7eGI#!XNoGxPt5hfKXFSMnRgOk}PbavlQ0B|HCs|
zK9n%XAK&F;ga<J1{961ho6qGF9kuo{xJP~fs$iw4ESx~BaIq&b5#Za#*Yk-mH{?Ff
zx(zVSKHmb3Dxn@L7tfztc&qm^gP1iFN*$^you=zY{K5skUYKH?V6#8ES=%CS%+Guc
zlVa69Ar`)aAKgM}9pO>yRSwh+xz$6@J(h%TfLS2EU_@OvYN}%l7rT>Yqm}k%R(fdP
zuxW@)%QI6mWG`)+{|Mx}|NMg$9s{?PFd6^SBUlfuXSDrfjpHKNzU4w=zyO(V6NoYs
zz-PYWxC4Q<bzf_>FKc8EV{i*M1u_daK=Lx40*niOkar<Z^)4g8J|0glB@GXS6I^w7
z0x?lA3QlWX?D^rLphdaU)Qz8I98WVz*V;C=E}DJ~VEFrNOMf2;YxYkfO`;hSDEG{(
zw2Sb)>b*8mZNCVJITZNq<q1KUHkPqOS9)*4qC19h(<EB<T89?~k3SLoN>?*U``GVT
zlR7EhCbVo>SE-YGAGh1Xrrw_4eR45okI>!^sMfN8wm?(!&I;?}soL`W)nT4;JR4)0
zli*+U2fY1N((iBnb2bI;V@`T>rbui%@>eSTia<6EgPB^<?-=9iheou)Q1tb?2585P
zu0`hptl55Y7CN6NfNF9_Vj=pLd{Ho)MEDh*Q1s;qdOZ?ez*I(K?6`x*B5Bgfr#&qG
z+K(`NJjSCoEM6yO87v$uWGU`Z5MV<l_oj*s6f?IT2q|J;zw1E>`y7nX$;gxxc)$|Q
z1Kv04iQ0FWju2%U0NaKupJ-k>Jo@=nI;(jZLlRTIHnvTWxe9=mNdnM6Oif@jQ+I>#
z5p~;fqAFmFAM7l?+6!Pg{u1m-lTDzpkN|we^D^P4FmX#zAb{pcrV2=Ja5gQ2b7Is9
zd-PBt73>-*tVGnjFF~+XYxX-=he_CUQ(nuAmAQa1Ed{~nf!m01MnQXK-HT5{1-gzi
zaHqEpuu$}X2cqy5urAc(%qRt87dp@YsDrC_u+msEv^TxI7bARwlLjH@bH{bQUArZ}
zd9xOF($ntCe~zE<YVFp@jl*lvmq1LV#d9U+W_Z}i;RX)~;94fPcbTrYzZ$!Vlm_><
zN2>j7BqHk0;2)Pq=E@1I*{^ZqF68cTk>S9|K{VYhhpl$w4rv3BNmbqtZzQp2)9*?=
z2y}nc5lI!{G=CK(9Ul=j?qkY`5%_Bd+~SjrB<M`+;j0SZXHIahCH2)!Wx^%W_CA*Z
z?<!+5fH`{zCa;6h3TRRUmaUMf4Jm{8m))BbhZ?ZXki|RxDCtAMN1dn(KL|S#rx9{4
zGwksg`(D1q2u7KByjbAv+#>p6^ua0p{42KUl$FERd%Xw9mupibJUn&`ejHOk^;q%j
zz^=f&vHpP0;GU%~rN89M$Kh-Lmpir=hnX5j14$nmI~hYSQbF2`o}C&<y7G!Ax*z~d
zd@*;bUtzZwkTz&rHRC4xj-DT|^7XB9cS;#hVNtx2kh4g?gXq0atmp<2Yq2A4CoD0p
zCgigQWDH^3eg<#67m$m96{KaG`Y~Q0b@39fMUyGh`;Wj#rgt@~Sup%`Hh}^Q`&|dE
zY{0B(fV>M5q;yH!yTDdem`C2FV7-{PwUDGZLM}Qy_K;Ss783ep4up`Gp+q!<AyGfq
zfU!){10_MtgbQg92YPT~Ad*2A{Fp4vaSK=zgqU_U%(|Eq-PdnDjP6a7&;@-v;st`y
z0*G6dIer3)Vu6?pFC-nJaYZ3+rxxU%>a5S78Gmr%gLKI%QdKpD*WHGrFx&}<%D)58
zDSk*dlUCu&C_$2WwxQq)7Fh>^a=c;#`H&R;_7>sdGt(0?15JKckPuVC#)FyY6OJIr
z8Ki~mIs*7czvrWN#~b+!x4epFU+fM$zYDnDy~4EfghZ}!b7eQb9Rx~5T!5>xi?^U9
zeHHX3065+hx@Qea`-%sj$thxjm6HyJe2f{#beTy!F>Ntu=S(?ye=MBHW(sj2HPe(N
z)6(~r2R_u@=zM4L!HGR9cvuEj9a2jwUk^un?AQQ^zDDzGtLbc44b%d$>49T;f3ygH
z9&CO8o~d41J5P9xCoK#@G~e67cndxw|0L==IqcITff*S9)Gi6rI@OgQDa=OJ#hg<w
z(|3w<?nViJctd}!4I4H~inD#KJ-y`^8{>?<wT_V4?mcs9rTDj#q!+(U(6Ykue_FKM
z`a1`SbHO^<nb!UKI<NJrR*4&%agl+OBvty;6Xv^QhOeFXt*b}DK1xf?Pw<jNY^@c*
z3uNh5%|~>1qVf{_2pfo^w8;WSn|!^saGzxm;z&}NdaX_jzb-_uXmv25Co?9->32lT
zfN?%s1&Hffe0RKPY5cWc1?I`qYnd^$(iG)c;au8(^ar_|R5i|0fj+ZVhZycmk@lCn
zK(Qr7HTl%;oTuvQgUIedi*jR%szZ>VerHS{ZZy`c68_@ct4gK!ApSdBCoEq?`Uv(@
zzr)?2tloNms6IC@VdV90KB`-n=0!$i#q0TLg`pkoqOcJQs&92HQdBRVJ~a?hV<8c<
zs~+g2+RH}<l;iav>&#*1N79qRITcS1PRhn>S~A5}8wn#MtC7nEwYQ*pTQiM(n?T#_
z$-+Mz6ZylT#H#OOmb_I|&y!3LYYoLp$4g1q*^-pG+jsTEe+YEz+S_Fq>sJ3NVi|wu
zxyhV4insLKc<PB6-+qeo8uoPq=#%+|R!OLkeV{wv)vF_zgGE%$$kEVFOY=fDUD9Q<
z!Yme7t<bFu<3wVO94l#i?1|_-W84eX6+O-^G)E#MGF6Jo?U`MP;EL~1UAi8u9j}^c
zy+B`^w3*oywFNmdK+zMQ^34t6mbs>Pw7C6J|1Aol-GDQ@^!#sVC@X{l_34SvPXiu#
zRX%XTmiRXG--E>K$&$&mbKMXALMG3B=FZqW%$<>EFeJQ6h2saAFF24SWkIS3)n`?^
zqGyhE{{UL4B)oFEcG6NaexZ!v$TU3W>z(}is4(B^i{<Q?Owhl0R3UC8ulW2|ZfT5E
zbuXo{h2<-tKhBE?AJWMdGOKYgvNV$aZp#CA%SlCxux%!R+>3z%fy)zZEoo3{Y9yPK
zP0b`#O>dJ!-i%MlVpcs!>2U9-J40)&Lr-|Ca1{l8vg0+K_%~{jo%i^@vGoz{3(+Q)
zIus+SU#7&d^73=Z%`)@)!c3V=ep=V8Vas!`E8B_Ou9CzrX6ot?$5m)5i#5gdt6LY5
ztJ$m*AH|!%c$H^dy$_~M2vBqtDVdB`RgLCj$y66VZZ^}-^pAZKU#N7KG|g4|?p&ky
zaw+84auY4G=PyK2xx-2>$A2AynVsx2pM*Y8geo<(I4AE7$b8o&3KqR_G0&v)95N`t
zbkt1xF?NlYgeTSY_mzEzasXEIy!Ake>c5yCOx9SSQW6oOFCJ<+NxaOg|8+acP3CM{
zp>vVSur)siP5bQ&*1v(cyBdhg>)t|6Q>?{YYFK7IgQosNkC-xo^@|*e84FM0Wgykf
z4#M%z1!G$t<)cVP7MZviK(U#}@AvIO_d}Dx72fvH(0txtn?e)IB2uh;Q|u7ORpHT7
z(93q`W3eZ)(To%IK3tDO6cRn8c#1>OE_4HuC(HScpPizkO@|RDW3^C>`9}zZ%3XJw
zpJOohV}!!-8%H`Q?ER`*jiZV_$-*V;%P16S&b%JxJWIHD7%8JUlj`4319$d9AcwV>
z^vnaTDet|*AA(2@0ncFwbBC*Vtr7I%a0`7v9B?p~$4}*!oi)ss$0ZFqeuCLd?6O&?
zcA!`xzT}aE3>RBno$P-OatQYGnzpS5FtBN)|24?sB#MmZ#iJ?<Uq<}7w*SDTsSyu!
z)DHFwc<?uNz?qFRRFal2tAK`wpq8*e18LCT7WTjJff{-Z0u%siOHC)l4@FJ26yfL7
zUvI?y@^!9SVx;j`3aZ11NODD)iKzN&it5grtKFL5$`jzOeGO)X)qmN2)dV6<jsY2w
z#u|D4T?=SnW`4Xh4b?OKqPsZ;Pm0YRWfHBDKsAAD*W>Yu-27Avn(NQ>h2G{z$wTLo
z@{ixkrf~8SQNAbR1|y8?Ik=PJL>(tW*tlMcBPt#CU>&q{!+8c1??95ZJEW_n`q5)U
zelMC<NFHReEbvuQ*BHE2eWB;PFr=)YR=K%;I5)TCE8zPEMTauijkG(6!ZR?X>PYP7
z{soYzLJ$I6c4kmL9Ez0p$S#K-7suXEo5Gq0lh5@Q1uwL)#!F)XSxQaMg@-#xGeVos
zbK*o>EKl|EuCH%?m&uS@O8(uQcYb-#W>1czdQ(K~6{o){pG*fz6qK$y%IIakO<&H(
z^nL`u35S99txy0oU+R}{l>+qO@Ewb<g~PEUpn4Sd0v}Z4pxPymr;g(t0AlL%i^|WH
zbM<v&o4qWo0}>A23*gm2RjmI%VE83(&>+xqZRO~nAW%4Z$e#ma%NjrdR$uT}EkZ7@
z0|mdZ{Rl5bLn!-+#XF11%)*8zzuq=PvE0{<lEx}0ly)f!m0T;UU!X$BJhnm)cu_$s
z3@IudxB@KzjHbl*h(IxZC8(Y-LG=wg733IAgU&ICQlG`k)@08Hm`QdJl9M<1ppZ%c
zI$F<uKlo=3K*w8Hl@8NN0=Oc7vX|}zV7LPlid_IYM%GJ;2zBN|-zt{Sr94tHYamo)
zNTRW0(>eqNsk~Pwb97HlG$*evf^$;z9e3D{KY5pjuOiv;;^_%MB#!i{!ce5u)_#Sj
zk0=sHLFP_h_2?@Lw5MIzX-b?zBF~Pk-38)Uv^HJU?9D{m{>uY+Kt>q1R?g3-OcqQe
zJeIr=p9A$ObvWg4tNkO<v#&uEE7Z&53=`t_UjJ43|DA~Z=L^c~n6Pc@f1jpgoLT8|
z+kp&o&=r@WLVCII&>WU|Zp~y6PWLX_*Jt<utCFP&>P^QTEnL-=*Pkx<4L?-41Q*WV
zasPhDd8Q$5Lq|=_sP3nJj~!PJp*JQ?BRAzXdTt_el=_2Z+Z8Ro^JEjL^QW0{Bu>P`
zIhkU9XUFf3O>el_&>18WvXnHsSmTej)Ln`G2H?@x_8vWnB}6*Q1_mYlxy6PhKSS$(
z#@I^^4d(3V5~<x-f&=`y5fnP!E=A!xeqhK=ga6qA<Bci#U?@8?$!Pj`0vR|3(l-A-
z1*Eld`cQ3S9~ZXkOF3fSN0m$YpSh3Q>k!MT8Zo-`6<O8k_3=U&!=lV7PXKb6nrdGC
z%-;KL<W2(^jr^?Iew2UcfZue#aQ3@yxn<ztubX;s$J1L=kdKhjDUZd)XG)XPCq$=?
zVdAxu`aI7wjGy$?rik~?NAp<A{~q{cVP;a&A>v{Ft0eD=Kmkm4O)dlj)1rPGOlBAl
zb3%xv@+ad3rd-ekp7ZFwj~27(j;mT(@GB@)m8v#RK$JtW<Tw$)8Tfr~(WbV>)dB+-
zo%cJ<0b|@IftkkVLPA0H75jpsEdioa05K-$lnZ3c7%s`q<9{ZFn4|ACpEa6>c((|f
z=9n7QpIz{TJWis4kmeQ+WJ<5!unut>(yuxKmEA)&>Yc<y^bz3XE8RiQ#CA9@;DLi$
z^l*;^i1yb8K@$_+ItlF7`|StcMdT%<oSA8OGlp$*fOHS3KAh|@`Ld9q^HP;*@_<Zf
zHgRh$7)xeA5dpc0_Ds$&ChGqQ2BfynJioO792lY-ZVM^%=;hDKJ^Ao-$7)gyc*7^i
z_W-{MNA1y?3}d0gZ9K9Ij8~Irx0^EkA_Ml)L!!Nfl*x+|x4LFfh&2FikeB^vsYdEK
zm<Y5tWaXNjqfALYYLr$Z=cD@jEbRwi%<#DVD1P~8rZ0RGLroO)8jglyO2s@kO-TnV
zF%3N7zIlL}#?AuHMy9-yaTjTkxaKTg)Kh3lNoX~P*%TPrw6p=bCo_N}NraF3g6H!t
zYz!la%ReAj4z`L$1&)BD5W8=eauX1Aw}&8*ZrFj}!8~6DzT7~}nK?!rzelm&9-~hT
zqL~P_6NAii@5f8}B!zXg7)?WIo;^$^#Y&Sp9BUK5=_}ZqB-GY^_59@E){dNu>8oXc
za@-=h1k?R=jE8#SZIDt=?@1-m045rdcRF<HVAR5yMLkZp@fo7hI1$<@o8y3^PRoNB
zc;sk!G=#fE8F?=Se6(a-nP^wRc99z}EwnnjdJ*^+s_DHgt!xCJX!5KhjerR9zV_F0
zDJM2BGb_Ti_I*kwpehC1z`k>C6TN-yZTfz#tq%{TU_Vh+(3hQs(VZKJTMmNr^UdU6
z71Dyu=WF;!kmJyFOgT;35+5^0&~(^CGe5nFd28oQ3D^8xI^}u8ys*pf`next?=D;o
z9=1$d4#jPzYVJ?J>n&nGQgs%(7>p{j>WK81X&u}FK&E3eE>lJrl5|m?HtajjSXdBq
zY29^Ru{XBiq&p_>LEN1|HwD~ptse!MyoKn@OQMZ)bS|i$9CRY6Kkhzd`|BwFeLMc=
z3r;$m8o68hm%I=E{#Q;poFTVVUgIhiG^XGM(xC^3>*G0$-;`t?se$-N&etJSYVb`_
zhO!61Cp-YwMyjXZ(brq&uoF_(V@Ogd8mUvnRXHr?S5%U5QUk%3d@E}HUI~usXsOvE
zun~fsGYN1Tl-ub`G6QpvjM;NAsgsF2<TbxpJ*$YSOi+{`DiIKPRRJkcvBPH<q`FbY
z=dT`rdmrJ~OGDKY5lL;c)SH~cBKdfvI5|I8o<~^;zf(`9c(M|F8`yy}6LV$w4rr<#
zH1X5L&-TVh0wpC#j?<-LD8AclAl}4S);R#SUxQ|}^3dRms|@1W`@xO*i!F-gQ}Es)
zv!WQT6Ib9h-UE2A);0myofn{h`=Fk3TXl#f3#3qTCc(^vYW_xW50FGoNInB}GsTkL
zUlz7LWh^Yvax<25AS=oB2jiBO)t&8z@|mySc><xT&F#Z7NIK<a63XDdsH!QttJ6Zu
z5+Kv5%BLS0X@f>=$*ATd>KioUu8%&=kJm^BLXf^|)$m8wk@rE9-pQVEQ>YYxe=UwW
z+N0y5AI@cZz<dMMIv^Bq&^TVUD*Q#48(j-yN38mj{tT{9pe+8wYLk~>p@Qk<%87PM
z`9{HNhVe+L84udvA^$*cveFd6y-AK><lC62ra=7SOKQ82qFC9r^vzf4J|ygw^Vcq~
z$;TKy*@!LE&3iFaWGFxuKd^?~%!l?}RkUz*DtxM?x1JcThHwXVS7X155`uA0`A5Y1
z4bv=7lnH{QuKvMS)mLP>VI8=KhjzNk89`8N=}f~*t<sQrh2bVcgjwf@(4D6IA}@4H
z$EKqJ9c76qDMqLGwOy`|iSvAI+MNE_1HcguZU;)LG{S!9rSp|@YkZ?}tI+e2mMxJp
z5_&;+K*_i6c%HBk%26W*WAU56ProB$Kw{SMJcH`Zzmf`BJ1ju#&&d}N2^4H$1exn=
zRu$bf1p7VRM2Cw96~MSCqJHw$*CcOO;E%>3ouV*=Q4VEi+%dZlFNK(zJ^8A%3+7Fv
zazgnCjjs*MXXY!hK@AaZ!1Rqskqc6B+y)ZO4TW0cC84i3W)vL4vZxmn`LsU+=X#qQ
z&h`QuMhfM^gBy}5wWKUq&EehVI2XaOYwk<ELUZP``txKgXz|;p?l9Ra^J?M5Ojkn$
zq0>a&K(|Q8y{*6uiB<>W!6yn6UZ_XoL$NxcB{V6r0Nt}2vA}0;>*ONrQB>Z)H+?s%
zKCaq7>6V{spp2R25oAy9QG^6RHr~lH%?rwe?~Ma`D4aQL9Cr9hkg-=BsAxo)`!Ut_
ziRHCvnEKntjujXdTrba$dh`v=Y_MWDGM=(P8;o<SyaIQE^4t18DPlDS;*+!vw=S{o
z!hODc0+whf3&nTB^AD1TIDJ8GsKPqo{(cQRh&{aaoS8E%^yMn(CaL#+oXuF-O*lLV
zJScPeitFvjp6xhU!yf+R@t&8j><<97D$`$E=R-rD?tKSR(^NSppgD<uD4rH5J|N~W
z2DhaRIk5X|fTqNt=U&%atxtny>c)E5Pexl`?wyooCXQ|0HIH2|>L4{Ak|VKC@T*)I
zt*d08-@hymZjKgTs%dUe-*Oe@H>zOel6cAoLo%(z61lvXKhoZIKr#B`T;jJ<QI8s}
zcQ5=0cj&*Nz+^puWTDC)L}Q!C{TYpM@n;pvERsTSkgqAuwuyUOkzan0LZVuNGKr4G
zeFjxIK7&@A?K}_u`U~lkM@RK4WsiuYI~CNg1m57?DTtD#z{~1`5WFK?E!7)RN=M1l
z7f-f4BP}MEr^KJ5EH-G#VGOFr619xyAxc<<@AL`ayKkKthPQ)fA7B23xRoy)@2piE
zM$sxWv=VzX8tff?Erb|C&^ixkC;s`S{)tF-Ql%nI?{dalq$1s;yCLt8Pw`9Z@OMp3
zhWK&_BH<knD|#6FeYQduuh!CJ!lW;r(RuzcvJcRFS+-%Ko_M6wG8XT#oC+CrPBdP7
z2*d$JzW_3_L_Nt7;U%*|kua~I0|@0lJJoJ8UauB_P;%nR*<kgD(Ai~`cK5y~zHn!=
zIyW1g+@!>$On%e^*AjatBXg~7{gX}Olu<lh-@+c)Xnd6+ep?#Ch_(0VAaIX1=RsNr
zlAUI%d9Wp@mNX%F6!4_yrF*r~mJw?s)$S-~-`i#qkb}81Mj@*Zqh^i7k;XluGgF4r
zF?(0)2NXHI8xE_kkD2L1m8|?x&b0fV8Nbk^5sMd{W7@9se_WrBy+8E{LCrz<gZTF(
z<On9&K#K_4E22Sv=2SJjnBjBqVbW5cxlJl+XWE*pm_9G@tYMw*)K@;JYkXA;WkR<i
zmPQ~}b%JVvU+q7a;@~T$^(g%ak;Mtpe?F*yk-(Y0P0rtj7RG}D;qh^;v=x?7bcc&C
zYg?F-DSqqR2#tI*XjigwsO-eupG{<uc>DW{P9;l^0_;gWQpRgqf9}a-GE|s*_<Qib
zJQ#t)4_+{vV<DRkfj40?)d!D_33TI|OZUBc#D~{OYQYGR@G^AlSpfm9H+R{9p_oRV
z)eD4BV@yEUil1u>TQ;C|ry>dj<C40fRv$88vKYeXLV<GD>0$W2fVG?S&97gkuHE#a
zgxgS!M4)>SbilV}x97GN<#BFco8j+~Dlarr)o@M6BmB=bCL~Xi@~+w6m>lKpkuya>
z(lJ8RkkLP}if^{seFuLdZYg5!iWk}?u1xI*Rp}+uMZ;$O+F(67e|M(^kg;PK_VT0j
z*{4K><8ELfC&IiINg=POm)s^YGmhqPYNcE2;@*#yc=@8HcYyz)0CHlv6_61Y!35fu
zV7`!Mp<$U5zh6%r*Md2R+4--#h*egQvc&=5XM7Ol>jk2yTCY*Ukc;$xmZtyBYgt9m
zl2pR!TL1IoYt!R9OjN1P230FXQR_7M{2Ki7gZ%qFe0JNul=w#+;E17c0$YGHK2E|i
z?cX=g1uN2Z;hqF2_ZPwA(_nJ7k_kU4)GuNOkwbMQ!FO+|&kcm-u1}2qX%K=tLU7Ol
z;&l!sq$0Q3d;VL<KLBSp6hz2h!mroy{yGJ+ugC)?zXBUAW++FyuS@3mJ5`E$77gvW
zzs|{-5`3n?3-r8Jn^}B3kFi6*-6mE-)%0s`xd;4JK6Muj(}3i#*n4Y$7czkV+cFZ9
zyz398SsP~s>F>A5KfU|EZJocD?f>Wn6C+@uC6d1F0*|s1BD?jM7r<Yu@mnzB|MA3s
zT*o9l+jUYIXZ{~P^WPcxe|Zt;gT85`qghV>J2Kioeg1#{`%6!>T;e0MP1yg@C;#6N
z%iTcqRuGI_$Nr10DLV_c6y?C)1AtWYGkab<7_WN##$l=!+MU4113cywE}le7aT#Ex
z8FoZw#)BM-QV8?nJ*U-dZq}Xt2pvHc2bO<cN7({q@No<1xJBds+@tesf|ji~zd&OD
zPAgM-huX9~cMf|OaJ1s{5y18E71&(e06I(xSM#g{O*o^fN3W_J#&uGFFpK|jgf|}9
zIH>M4@>l2$%dM0OOzSGVdspe>Z4Bb$k?qYFCq+JB6oe}8*LDCG5-#c0WY;d~Pr0*9
z`rVQuc!?TO4QS@Ne_88)IBpwM`Gqm(>^*0xM2q_unV<%af3B>$Xut_y>`|#;{MUo<
zQkz9L`L|n!5n#&Yhj?IXWd@4RyEO4TRMcyt&NIA#U*nCY{)NYW0#>GK{VY}RYYzu9
z_{3(OzKsKvvk(|SNnU{__I)7f3Q|?beB}YWo!r`p#kzpAslqph0u_9UogE)wjW$pa
z2|OZXpB-)T0Wc$V@e<VY?|^V@u*{-4tvjk)IPboi%6q_IQ3Ld6M~y61IAPBVFiu>j
z(+cWu7-?_bA;od#%iVoidlZr5;H_HBvudtOeyV5c$kOAo+i{S}vbi&sGSGsFeWh!@
zBZ>l>;Xq8uwJ*X)0*?Z_<+n^*ZLln^^nV@@Spp6;ciO1_NkKJvkmQ+r=>q_eFo6ks
zVo`{34{X3?vm!LM|6G&i@NFPfOpBV|^ifW|!`}1UW^fz71MuD;V1kW<DfGOmW#*0V
zfMn6~Gf{gW5F&$yMoftYE;m60#wnVo#><K3sj(`}K`nfCLbJ|~@nZ!6Q)&f}o1<??
z)o=GQbLCWA<*FD`Y$2ZqfkeY6@?IINTpkS0rm|jJ(ictVD)pi|O6z(*;QBtsJ2Sk}
zF}I{x{4iy3WSMW+yh2^0M%+)=A>xm90c`{=YF^k&+8_DNT#?qASv$YnNZ79giT(Rf
z{_jUQ3!Ku0+kc+?KypnCD?S*R0(yMF55NOStZ|&2xCH`1TwIjwIw&?08U2psmip7}
zZK#I*nc7XzbT}(xh8ExGu-rCN({HWD1AHH*SfrlqzXWZH&adyHyN^MWC08^3&Mu&a
z-3PO|g}ZfOEHHPp&(%$$GdCL}*y;Umr*}7N75JNu=hHxiBvKX>{kA^>?xl3Va3l@Y
z+u)BlaENGp&w5zNIiZ+Pz(w=PIg(H42Gn2o9lPGijG~kb__!g^7Zb5aVLIjCVNyyO
zAhhSO_n-WEDKTMT!jgk$2aFt|wDrnuLc=!!f0SDYG^()5ANJ9Y1Lix|Hcu%g=ewDF
ztY-O0fr`;w{V@pms~(`#;7TS|Wn6*Nysz0`*}Q`H*S-dS6y@=GXmeEtqyNF|z(=1I
zWu%1<`#&Gpv3fM^7!f1p{V&?Lb~gclwOit^^HzFgBxWIjS63I9EyP^=kbZXX2r#A+
z0B<-Y>N}t#^qXEWEYwqVA*E>r%5pUwS2U{_z=$KO&-JSSGqxMR8!zKJrBH+x0Oy^y
z&K@<D?tIJls96RzEp(Af1Wx|PgX>r4*wGfv@x<3eli1n;53r*mXCquopSlx65?!+c
zDppd~Ioj_y&<-G~-UWav6?puF3@T$~+a^pxebICVY8+AJvpqsyp|aT}0b0QUbSd5L
z)5&nf(}n2y?9;YICzlsj_I|Kz6O9B5N7<R<PHttq?I$qQEFA@qW!nh;l4MdMk0HMe
z)ODRApL7q)kcQi!7?x%Jfy9hnOmplz;P&2)>P(ved}9ZHMTVt(lVz}9-qzuKO5<_3
z?DcT^D5}(0FWp%_I(rt-VYrWLPr1fO{h<6_xoubpbo632DIoE#=QhdO6`O?{d)Qkp
z2_d*cF?)Cm0iw08C|YZ+=k}Unhn5R3gx%?4#f?r*Lu=-*etzJ4ycZvG%Gu2I#r8wO
z6g`%8G%*a$C}7dlP|}R7L*{m3n^sETy*+G`d67i)a}@BI9yOP2-)(b6(G%UBYyvl{
z^-{O%+E=NM_AbT1X5chOJz1Dv^{2?>-c1-O;X&Eam-`nqbR0c!sqw>`Ntpw`JE+>^
zXo5fay+Fsm4(ruFwo3IDXq%`9gspB1(P71a7SeM+aT0s-5GpjZpbm)2z@Gdj#Y5RK
zD4;Ms9xY9m)yBlUmnrT0y5bUy(-0D?&8OIW4($g@*jHe;pxp@BecT5^%_@J<ZgmH#
z%>YJntujR<D&Ewx)NK^B_!ORBEtehvU6!f2{v;s<bKe7f_snS0UHI2JcUzFMzny*S
zCK$~iF*rEiCONft1DF%&fc_j@cpR99+HB9Zup4u23o9uO#g2_5qk}`&Iss46E@%Kv
z8<qa`gCdm?61)L8RD*!6k!k`mU3Kzv*)Deg%(E-9K#oj7gBlH|)mu@F?ks%j0Risu
z!qsxjka&<KTF(29Y4C&Nn`UU~D{DF=@v3l2ePPbP2UO~!Xc&vExKfX$S^SV7-1-U$
z11HP^AA*ESWq>)ksv4c<#{i&bU&5L3Tp3iN7BJ^UH}^MU$I5_@3^O<zxoD4-3cF$M
z(=dd2kpZ8?&#Yjys<y^Y_kb(0E`8_=KB?=52xJ5PJyXWN$;kX1Fv1+SKBctLZwwW@
zNdrP_9RQJh8IebMg2{mR?Qh_bB2U5bOAdd0usR$hwGR+wv7RayF@G?;RTX5?8$P=<
z_+(7|v7k7@-<5kpz>`l*fBU7bri2ojiUfe_%4dR`%W3W}&@9{w7S_?7@F*J{>wV_b
z1F~L|sR5{;E*e1uaRJ_}Ecg(4KiV}^NxD~H-DIpsp#CSzV?h1EWS&T%Cl%~lVu+p~
z6?+c5h9)#m0hjd}JYSHc)QHKgNBUmzrVyVMOOQchu7J{SyGH2l2Y{JGw0U2^PukMx
z@<dBMYhtTrdH~4JnrhHgfLqyJvwb}FiOtUSe@?tWnnnW>)yOExwLc-_Kn8qp2g-dB
z1@tNknJhc}0w#V`7y@>ihj;evHfVwJ0d91vFXeMo34m)KdOiNNyBI{;bu9S3!bT0G
zVNe}cKwWH@0;AOqg)V{9kC5dRMFYz@XaCbwKs>v-uK&nXlySif6Ot+$w~Q5X1?{a{
zC*$F2If`~<9?bSf`eD0%1~e-T!I4^5>$UGwb?!EwU$eP^wqbHU88zZ88}zm-oTl$k
zyl({D3U(60(=L)Z){r<PJn?bYp|uZk$^q8df^~pS-v&Q=I#X-{^0=XK)gxvwoFLI?
z-WbHq2qeI0ErYVq?8p!z_}B+$3$zfwSD0amrE?qfT8KP)6q6xW<g#$&$a&f2lV0(I
zE0m|-E51+wx~y2}b}J_A3j*l1BYun#fkd>nz4DU%g)f$sB!0Az08@$tn3?CGxeo|j
z7V)O89aJ@Vyr#8Z+N%J@R)c1qq7|BSq7F&2@L;Ju;HZ#@wqBTU0@ybEdoYUBKm~vm
zxDw&G%%)e3>3a}P93r<{6->Rs6=)F_fbUH;!h9%`#t3^0@&4dAX@I7JE|570kJy2f
z$U$<dF1Qk_l{Nec^YLAv^DXgdZ*jYJ2y0Fv7?}Y|cn8}t0CL|%^MPZoO&kKlrljF}
z#p9rK8{Jjd$8!}_Z01*_i>-y#*VNjJuC*<4agBm|7;Q}e^aWMr!TFbfXnNaK^6<*K
zk+=nb<*nV9z?E^!HMRM>iJhdjvgk5=$S+G8#@p_POOkXdj(s%gEMr1C8z4m(<dG%-
zd4F0mFn>ggv#x>{$kQ)||1fyuf+?eOOXlpN)mfxiES&WfhgsL*8E3x)nXSxVe23?6
zFJ}jweQc_~DwUVAT+A+jUGB}@8I@ao3Cy48ibiz<E$%HAxOD1bM&-$qYAMm<?&gbs
z(iL;dAhaWF*i+6A7_bLye;Mw<M#Ac1#?=Y_Ww#;XygUPE;r_Cb!eY(phGq1CH`QN5
zK}mTd1X5HbPq`&kpDp`qtBDyrH{RJ(R<0kYVQlGkv!n~fg`K164MR$A^Q^G7N=n&8
zR7>QZantV(rDS`eF!7#pFR4C@#iL^Ix&-dQujhy5-}OPMp0dqDYa;m?i%RRc$r3l&
zL4(o2pk@Jdfnn$2JvGDy-JeCwLt&x=<)d&7kigz~o0GFi0WUBpl&r2#tdL$e;x^tx
zEBmk>XFQ!Juos&swxZSI8>3Qro5QLVnwt>`m!$f_+8Ar?#6d+n$NZx4o{~WZ;R#_C
zmi<+4j!r|RZ@4pfH}yhnAD5Hp=l{6wy<VOXi9_*Ok7=U-^$}$r;H%T2m0@gjpw}Cx
z5_^qm(+i-lO)=OL**Pumj-H0wOU}broJ~vD+AQv=)(v<Jw5PN@El?TJm1~H3+bqDn
zT`7?>LU;aE^~g0Ak8t1+ec`S1*%b`v+Banon!Gs)!&|o;8*uRpEU(Ygx*vRt{50;;
zGdc3g23_;ynOw=Ck%Zm(AY=Nj^y)7tgF8#{E$Atk0sF%5O5Ch`0u&bRf9?(KG(f?E
zEf(n0Zp}vt91bm71;+Nu<{OktRu?8DJk*=hL~c9CbmHwTNKD2yACHkhgN&yMc>1Ng
z7J~0eJLM6Y$S1vYlnHX|3ALf5M$^{voN^igDu$Q(dcSxR8uOu^S2OZcu(A2ASMkP4
zRi3ph>BhTzL&DZ9jm8ge38`woOjd!^el|#gbUDS2zfh)3uspcQkET{|j6OwpuF2sS
zZJJ0`-;ZZqx^MV!$XDL1DTS7PegKd$8fEn%@6!Q|Y2-ba8B&Mx=Ga#FI2`XBc*Kw8
zuFluzwfcTUX<Xz5;P}Bq!fKD`w?mE%rhSczn0SM<?ZH@!(3(MlGt||b29R-5rr;e|
zZsr&12*e8#EQ|xA4SVGzltAfDc&bh5+FR{M9WR+CrM?j?AW4Shtsg}V_f>3IW($Z2
zZ|=xSjb{WyPcX|f>SqpyM5@ssqTclFmNlsIbsH7~#I5u=lIA{_7ic4ah^N(s4z^jw
z)fy9*L-IL!)Z!E?gdZcG`B8pK!(Y=K@(YRoEdD!?gwqbq<ri@s`Kz`d!+*fC9sH9;
z{R?@q=p<<wM0*yyvOfA?C^Q${Wv6IDrI1UDE9{5G?#uBwv^;xD=oz$Vb1EdA#5geS
zcNPjEVqDb44ey21@us^ts^L>qY$zko>yXi{e~{scG~n6%cL8-RG1VO83z8f^+#L;R
zLOUmnTtW_{5zY?sjLkMOh-SMLv`DGG_;DE^C?Q)buE%onwyi3gqA6*baZkfFhOXxw
z0dn#~Q+0+@oTshxkCSAz%tR-BRM_sj1Ix?OkELHk+X&f;k0n~J3AYP<ASw}W!*N{~
z0W`eeywayqh@Hm?CrnPy9~}ZsnyB0@9KTiWsxQ<MftmH{hj1I-H(C6Cee&bL!)cHm
z9qV)n&*tf4#bH<OzUn!_Pk@qVR8HSY7jy!RT&~pE@ATqox}l&(`q!gddxMOYbeTm}
zEi?9i`uv7t^6Ytsv_EP~YJCr;%a77w9AUlSY$L!cpMmokPM_632sMA3!Pp_h(F*Kh
z7%gpte@h^aOhJj}YhMmn{LPjM)RBe%{F0r+&#yGS`re_WJ~W<?iaWgiwnp+50li1e
zb}<9pI%Mh-Rhx?dlR<+%XeK?BNHd3@T1h6Y8{L*YsCNEtK13e#j;9Q8B6fkZdqgC?
zY5A*CU?C8@zdo+3H_NulBA!>Fmv+-mqkHV34$8<josHXo9H~j=F=mNJC;T8BE#=Bq
zMY#dMm<l1qwdx^!TchR`C;?!8{Jd|GkI?Pq<NKQ8b0KhDLP-Ijr38gB5tO+g|De6y
zJ36pFhMMXKCV_VIhrd8)$~jiDn6`)wu19&1I!d?&5e+L&ZbpnPyMJtmV8q8|hsoJp
zsv{AQ9k?q@n@19ED~7V?$kVSS5Sl6SZAKV3RW#!7-J~5lFHHjzN^<07TAC90^N|`r
zok9-HuFUU{QKSDfrD<p@x|Opj0`uf<eGP<=W6yMotaEaBOf0yWUV|lzBxJ{B7z(nV
zcn_??FUhGUBkFsr5y_%Em*)K`B5HR$);XP<e!ag9NLe=2Rh59DkVAc@7;34y0ya%P
z?;HAfThj5(WSBqqR6<_Hv-?2ILv!D{J)F$rMd4G?{qqFcScicY&U2U3tNln|q7ioZ
zRVgU2o{ENQH{xO^AgYGvxB#IBjmil+*SN%J=~p`sCaMD?6qHJMCjIbcfH<&PN~j=!
zk`y`+JD)^hX%_K+0~iFV1NyMq2H+s%7?=-^#BGEb8q!{fqRCW;<*<_M<S&kZ{gq>2
zrCx>kDy&{Z1T(y6p}CWJ3ebKwK0Wm}_srb*^j$mz_ODyq^m1B+CIJe|@6kw=0d;nT
zy?pj4ARVaIxh;~f3{||#`VJAyuNFazg4BNmc;y?+Dw^#Qc84YH0-@6z^t_~3V$&Mv
zkoOQBs&kvoFMVqqx0m_)Piqee1oiFRYi!YKmU=WN?(uYeHF|`fV!!42lh$k1l8>xV
z$lLszfQBT>TvlHQcov1QCk*bB@r(!bHf4bjt!bag`p?veH$0$0Dncd}G7TXI=C_7H
z=N5bqoP@wdFe*Y;Rd1F6<i79PF#{LD_$?~n)nYFRxJu*Ol3uSiGU$|OmtQ(Y>~w*6
zJ4lKs;<0-?q__%{r7qL=y+dbdb8P$rW}W|}0!};&$wB}-{GfB9)+MyDk&A~ps7Z6E
z@oYf3ab&CI#dvVW`9gwhBsu>apXbeQS;yKxa3UujFs=P9)#^%@W#@l3Odg##@`A90
z)nxRsi*F&dm3)tWl;%v7K-15H@K)VGwR|=8L?M#V)KXs4N<`lBz}jXd;`^1+OrT)$
z99JlFiTO}n`?D_|*EF?|c33wmF8k`O7)cOmq^N41aXW1f(WmB((3YrIFD<5np6`vR
zT_-SpoURUnKs@{?Z7O|PExr`r>^EgQ`5b;Bo-rp+zWK3|z;%OyoA8!cqn+1~X=0lO
z+u<j00P+wVo2L7Ro70oJ_PJcCFY1bdu4&<ls6GbNMb*%yV@O9#0pUxB4b(KE5D{5E
zd`Xu2^+cy7HdlWR-q{#<?df_RH+<J+Bm}p1)1c>pL5=Y?&r4PiVOuL?-(JvwYg$F$
zz_ru%qwSs18oL$&o66x6(*oD8scJHUnKjcMlhGrZ_f<~h`A)~D{K0ca=Z4)pCN;%*
z2?<hEj3&QaN@7Hc<;5oPzQr&4)BN#h=A8c*d2bn()z-CvD#E4|5R^wjx>IQokdO`)
zrKJ(+1_3E40TGdwmJsQZ?v@sj4rx%3Qt3Klxplv_U%&6W&cAd1v7hbr#9DLBImSKi
zsH+MY#aF}ot*y+hc=R-W7UPM(>gtiRj5oX{&+v^;O+Y@GDQS*s;uo5409hw+lKB#w
z3DeK-Tj)lpdQnKQ5VQz^5$?zX0i(SPib8w+i;`b5qQVr)Q>`3@9YtOg={ld(S7r8k
z4`hYUE0xQ6<O^iYgfO2fMCL^EDp|_ycg&;cZ+j23e0ZLFg(_#<;w=qMXTjdt=pXcZ
zb0Oabf`kkfGomA6-k!Wf{%vd<m9lyAX6~i*kvrQX9|7Z>$-=3*>P3N`ib!C$fh>fQ
z*+$g8dwwtI8JUC3txy@<{sD}x!Ik7&{d5mHN-!G|Do09d6mstqk5+2I9<nA@p0^qz
z9MI@(kA)`S7_*wWuVgN5!`hos-U{F53V#J0>P1fk*~f&r@Vs=#XHP!<ad3KVrVbie
zdVqhRxIcmGjTdd30D%;`D{7OtnwfsMV_u?4yrXxR<#;zAn7N6XZ`g|QJiRj(rER;g
zJ@+Tv`#s9-e#_4CngPkHu5I66x90iDlL06_F`Q#cty|k^fySt{I&!-<Dh$AdDW`Ax
z?1O%^yO{v!`j&lFFP$5Ckc@8LY&8n^@1d`pHp)}nB2MATFX8pX!4UP>vG4F-34J{M
z6It8f?i@FN(XF6|_TD}R`5|sm{af8)mJ1;syFRm6xOy=4E%~0=S3eU`seHqfO<MbI
zZ21Zp!81`Nip3N}VGR~EhMNZLofRaFvMqRYKA+|pf1}SU;eh%^RtW=z{>^lRZFqBH
z`EklER3dM*z8E1jr7;^`BWu@N(AUP+ceY^S+PXu4K<RR7T35E-WC{X5k6gPRar!-7
za}I2u+j_Fh;RK7<rpsoA0OrcS-yNH5-#y9Bu$~PZCG7i@%`Fd`kG?}#nhQnN%P&AO
zDc=19H|#~~5l~yVqGOibBWhZMPwg#=klL{8*8u0NUl1HaEZwMLn@Mo~zzQ25^~NHw
zr*dAMa<{p>8J)^5Wtqeea8nHoGF>S=4+=5G&gIw}xZMOnD$z5;JU?ut)M8{kJha`e
z$it7WE6-EfR_J<#7&0z2f-*g9A!?bZ5A3?TSa<xoS<-@Bg>K!4w)oN^SOPpwxQSp#
zyQev-MDz~za;M9XV93jH`$yZWwEzU%m%O-)yBLl8_LBWmwK@-Xq7yUtJhw5vx8&Y&
z>i%AKwu6-kx*yLLcf7|Xfxq2?pR;*cJ{LQ_srQdMj8IM#*mh6kavb-l)RpIGAqifg
zV=$|QPANG&!@Na;<h*ekFc8n^ds44Ln}4n(offdw@(jOdiE7PM!t+YBEz_@XFpWTh
z*;nakE=IQ3psw6Zzo(?upY=3@QVM|yfR%CH5yfDp#HN&f{Kux_%BV^Y|Lp0{BX3@3
z1gW~_Qbf<g3I1j)<BI(E>%|VA`&yECzU#1YX$UZ5MIe-}blXsqY@BupwuTDEszgXi
zPoZd#oUo;D$u5_M+9NvPj#5a8(!?Jfh~yX)DTF0bjLaA0(hzNCMu055%CL05ze9OD
z;dZ)0j0A~O>I|<^wn3}n#CQipG14-}10m0Z=;_(vtbkS{Syz>9*OXOLBLJvOKfHPv
zbEdOgofRrSWd}NDvn#4l4rO%<J7%&la}tz@(uOnKi~_>$WJ**Z;wQ+`djQN=354?b
z<;r7b?*eYUz>Ow#uaYOeICB8BZYuS`c@wU*=v$YgwW@-}GTfwfLGVEi0uBkMSGU&O
z)C<lg=wG@+RfHf#Ov|mfIl^6%UnC)mE__oAgPZqVff_U-9b0H!`JjxhpYg$H<tXzp
zTG)$U*7*i=p`gl&(nAGpH#wk8eYvD$n~=-!<)YdAtths;_m{WfAE0b*zqd-sG@DjX
z4lsU~*+W*{Jd!-vMY!wDH;j3On0b)sOEkY=FJ6Ccqth5eK}a^ElJW)OE_T~s={#0k
zZW$UoZ7BKA8)-!d6D?yd7T^q`+nd28G9_x>LSbY7l$fL?Zyj7{YM>XFO$s%*%<qvd
zI=}2vc-K(A=QU|G_0sbK6&twxfD}RtIu(k2cfYIaq$d`WaPa4$h(euj?|_GQr822}
z<_ltmoseXv@}Z+%zQn?OewKW`*!066WwKOD7aupu^~K99#a}zOEpWjCKk0W&7|I{x
zSn*(&pZLH_ENsy$C-ud7Pf~O;+=zeChDJ3B=zSQvM_{xUX~`p6khkVsvZ+5QYYiqV
z^rXBN1X!Obn|&UgqP|x0<UYlN4KeaY@qQotjZCyYg#U*&jpa)MwE9x~5d@ybuB7be
z2e=J``f7kI4sMLcQ)z*v`*->WQZsns8=sVH|ES@T<lW#P$CZdW0-IY*C3As5h$-(c
zQuB6%2WtLRv<}XBp{#i((DC|X^HW8Ehx`TadLT$&2MpaXrA7k7)GEcJx#zJ_JS%A0
zn0zqti5-*7YVNSOp#e|cg*zD&d;u?+tz40{`7C`>-&a#L9O~H^^$#`I$(tS5K5F-;
zqy*ill+vc4v((uYL^<Br>Fk&%H@LR0wj58d*HY^%oLc}RwnCd}@i8?@oV&GdF3ARZ
z4C_AXsAYm>a&gW%Kbv#&mjE21S8d%_rUI}b>x;#V{e(H?wcDs+#1NH;Bzf`zsUw?t
z%GCf$OuS1b`&mTKv*|cHXGbwP2+_6uFiIMY4bmJw*As~|1awm|de;*gh=Adqag?UN
zBGdZy3frf!@S64_m-nz(vKePnzQha<6((J;UhuB$E}bO1aVuo<;?n)W9*byt{(LRB
z_*pMbE@8~sFPQ3j*63;n*OZ37(#thnJ%x)!VauKfgM=;2GiEV^jv}@#6rM=16Fo@_
zg;x0^5rPH^OT7tk3$m@ZD5ZB-`c=Xfe3z@Lh_?kkjW+z&l2QVpES!7T^~*6zQuG8x
zsJ{E<xDy-NAde^1&?8#`aSgAzYhc?szMWnV#BkC%D5SaIJ0=wgYj^$EOF08n!oJQK
zYhbGpQGoxhp^*~ZD^jEwdvHkaWLA`>o=<%)6cG>3EXZ!ge}3y6O`+=3KalRK;Of_Q
z)|(jBtBNIYa8h+%&+;t>7C8!r&jdSSHr^r2nP}zLU?CR443%ixi%z{h_Iz`I$MTcR
z4x5cXM~7C#1Qdu<!A|2<5-5gT>u=xl`QuDj4sGSpUOoY4*4RxYr(|B~3BfWCJyVWI
zJ?>{ecCls8eu3$Z_DvjIbAq?}3LHsP#m&R_31`L`D7%_IavC&a5v841t<#<Iub4$!
zU=pX_j~BTaQL)ky_%OI=NHp!~wG4ADFO!SrghB(3RV$7K4p=ODSOm`dk|D!mdVy~W
z9+j+?J=Roz7?5H^Uh!xVOln$Mo!XvLu~*jfEa~{5r51eWW9-;lkG|8tt+b_(V~%x=
zirN0;Y``ShQO(IH3aS|gDp(Lgc_2aS==gqVobd2biAF#u6>F<z(;B`LUgK!$8=HU!
zf1L7tv|~tcK;E!M2#i3&9Q^o@Oz{Y=(?(In0@G{8wn;umrR%@-#kQk^`=2-WBIFmV
z>ZJsM4aF?gl-B?$7Wz5|--Q1<;rsOqBwqMi1)7U!=zIlO7X~EP2HW~LSXYM17bQ#=
zy|$WW$bJ26T!SmLK^Gxj6)88RW+Fw;$cqMw{a6at<c4{sQ`mFjdge!pp}6^BFZ>tv
zWG|gB4gB??|H0gmp=5@EP72|@zl^H94Ct=7?>SU}xo-C%eGYabVJ~R4!&om6o1ZGv
zn=g2YRpz(zBKQQ0u#M;G?bZt4+!!PFYB_^M_=>V$3`;G&>8@baP32w8d&z};ZpTPG
z^In?Q{FQRf8nB!(>WLRvAe^vnD7{x4mdnV7?u-`|6)HDzBhe>wWc?RS?BB>y$bX7{
zMy}syGQrZntb#<b>v@)UHfF=sf$WFOYeLUOhv3`j^1JQW?1DC&XhLvzE*$aq4P3Oa
zD!m^yV#C~SI-5vDSCOx8=`*HKj_*Hk4W;+~>pyY#*h~5`9)Q!!_;MP{xKI-_{rG!v
z)_p{^h%n>2ffdXI<N6zN@BCBf)?vhtGycLL{;Z8(=O#EZjqU(;#$}9~E;&XtsCpZF
z{Ss>?O}I-{>6ZN!``Ze?5#Zd4MFz&PQbCsgCdH7E)dK;a9dMj+(8<~P3n?Vr5ffJy
zYnxdz;#EkeyLd%Uw2^rM7(>QZb<=yUfWUl(U`t_E(%B~g^&p91VMuTYt2d))gu0Xp
zs{h;w`2KKWK@J2@x$$M{@tpsO=qK~i7hx^?o<@)R#ajN^2Lib=XDwH!=4$avps#lV
zx@DeHP=Jm=&tTHP)sFU3-U$Mx8FF?j_;rG!>R}c+|6ZEnyt<~t$*A*iWVRLBivG60
z`N76$6GWI&6VHpT|J8r~pZ?et1tN*t1<9?yMCkwd9Q@`L$~A=O=XAXT?{|hv|DEnP
zcM2i*>7N_H{2gxnFND5-CZTv$)4<aIc!o*vm;crO>BrMP$4AJ0{z~^KemD8}@4x@+
zGa&c5@TGkr_}yUmZ=TsduZZY1SUi_0q(a^Q@e92!{Ik;@M?@?2SFieX#khO0)^CF{
ze?;G(>}#?-&<UHR9=R^)GPciLCa(Sq82ZN>{mXSK#|A%4KBcH%z&~83R}u9Pd^iUk
z$@6ROdl7x)2O}0(iU{XLf}(WfK53?AN!FM0*=K>}ZZ*4K-n^Rw`#r@a9jL8dfs{_x
zHs%^q*Z-3{o^dRj-Bm1HpjKFW-%(SK@F?djsxV;7xm(5_Y6MA>VSvKqijD&fC{%{1
z8lrcs=b&J{DWR6FstBH9uhDD?Zl&XZQ^;kYJlAe2X}R8)Ri`)-@Z0BCKY+pRSHSLL
zGcGJ!BMd-JqmDLs&ARcM!!qqK`L&-?l>fHUM6V&1UrAEo?mwlKUTg{{8k3>eq%$t;
zCAK0D3XwT669~9iCs^i@7p8qFdMlj&PXt08%2glHdw}irK}-6DL>1=_{)H=|z^#Zk
z5oo@rUFi_&akxh}Q&e%X3|Q*zfc{WILD@rZo>olZw(bQs_ID|F!~#Y`v`D@I$kO@4
zVDI8z277tcVD;Sz%()3eFJ$93g5EE12g=R%<HO10cEr~sAnvC_U*O`D;w7V<&bhq+
zYxi?Qi5i51{+>6cuQB81Y=C1-M^_9N#$l6&uW&VZanbj#f~UJ)4aa|7gnwSL`av1k
z2y$)vM~r_yiQvgt?egu*DIs)`)4S3qfK8Zzqxop8PdM0BX96bNeQ%6!ejdz?t6w4!
z|4ARbTglR*bIsQfOp;`Z4Z*v#56A!)g&#Ulv&Y2&$NfdUH4H8ubkkc$6#pO+S_OCz
zq-@bdm)3g^gL)v$<ofpN1k!~T@dq3eK)41K#0bb~2<@GC{|XnuVIT+k6sOIUXpr}b
z^R@``DvP5v%h;sLI3(@1oLaus1rCG<_Bh*08o=)u*vbMJ7%RsBuI*6sLP6`3BcuVg
z10FR-_B_}B@W@)6hrMBp`Eg9bPbX3^l@+~GVll$00foE}Mo|Qz<kUY=3Gb$@JaN?#
zI7y@xx3CmjDs$4Jq2s}%yAxt8Wly}*FggZkn}K*K@CmINqS7y4H52KJPTaGExQTgK
z*`??}+>AEQtB}IG`iXM)o#&CEZNa&i8|w!M-w;%QtZZ5mO~R{&A;R>K8=u|+XKl~m
zIC8kX)Bj;SR-r}V`99!k{uA?e2icgOA0=THSFHu=lt<6Y050syP@<^-%eEDq?P~V=
zSc*id0HeR(E&kd$ymCWff4e)I@#Yt$>p9IyEVn<+@a*Q^T1=zQB&waea@bNqi{*J_
zrnu_VG7$3NhM?k)u{l5Mn&axY2I7t20)|>_tDxhJfUH_Uqqmd+*8uoj5guR9(=3sT
zV%LtfLDd`|?RN^URfltm=rk)mu0zWUsWUk@RK1stu`wRG<xve36*~}Tc7S6Z9f$>j
zFmdGn5=#I5FYywhhVy~^41Lv~?qvm=1c*0i*U|@M?^vb&py7#jzC6}9PEhNt9tI}+
zx{c&GumjEKr^$(5t@lJ$n0}ylU2{p)0_&F|yLma6X+Km>{FJ92Tdu`OQNt&+O&T%S
z5Odi*4b-)+yR$22+!9;IXE&@vh<;LWPGgpQ=7KX*%tS&2dQ`aV8lW5;5@$Q4q}Zzg
z?SZ{~&{4_?j4C+Qd{p)RI6nCXX36SR(>+87a`KM^b*{V}uAE9DWoQ4g2ODK4?7x;S
zPB42j#KS8%@u~Hc5C`#vCcOYd9gI$gcOzIYQns1~KIe_}u&Adujyf2gIMRDuLU!7!
zw$i!Fv8%#?3N?MRfIMPt(4ew%=Kkk=mS~TiZ;KB-L8=*taUERPK0m*X7YcDQ3RX3n
zsw`iD!xfE8Ls-dK3w8CAOvXxcg}@`IbfK4`IRRqu5YgW2t`HRB?L-;3>BX}$_-0J4
z&{!|g@Qh(1;`6!pspqt(_L1v%l#6}~fh+z(b!W&M8jGIj;!UM^*YqT#^;-Tx>2t&;
zb;qZj!I5Yl430UL3jvwYyS;to@>FT`|5TNf-iPwXvOD!%?Vkrggd7?{8bUAGqH=?P
zT<&Ehpqa^xp|B*TGzrB80$3@j0Y<M<xK{`x(HOuFMKjR&Tp%8X@1#7?)Vk@&wUinq
z8NSwh?~|urx{4UsEhAAIb1<V~S;zB!=5-3Q1FSl&*E%AY45JE=;bvMfG^t_0uvMcx
zCr-Li3smtU7@k)^_(Sd)leb_KIR{`(1bDi%u^Kcm6E{<X(IV$|%tVz_G%^`EH|7qf
zAN_4N5IoL9N=e~0F-DBUm9a8DTycao7;q_^<f1b|WxfZ<g(^5|SfX3u$OpBp0kau(
z#XsPWlNQr|fO%YCvH_{*&`)Y3hu1xbOzAqqy6_$`hSu}m$bbVq*dDTeFwo9<!m8Qr
z{5EM!qxWPK-?OHoD;s#0;(0&=^9{f;kPYbYrbhL1c^o*qfM%0u`IPwz1RZ|oORjfk
zZ=}fC1~MgWN8`ihYB*i4Q?q}ezY*2&)gD;FPqh+{Ve^uktaY~(*@Ln{;ovkX1;J=b
zz9QgwIh*zf^wy_c-%FyfJFY*{!#3cH2)1Qog(p6nCT^3%w2xM!K+kLC>?f|C0ALHY
z!9aEn#0#a4!4*$xc`Qej3(EX4;)QJ--C&M3=7DGRLX9B2X`W-tWzK(nj6!a7kFiJi
z!#g@>-Al_fIJm_WQ7ka6y)WowgNH;C+$ZB=T*y{ReSDf+sJxepsm=G--u9oQ@;_gB
zeiJDYU%$Vn`)83@ii2jeHVhm+BI%b+fF_0kvi-V5y$pDCp8-(CK3JzBLt_o#LfD01
zqVpL~<iLi=m=cdT2A-yS20_2<=7^l9e1viq`~r7+qv{S<tou@>@yF>g0yYqQYA~Df
zJ7l@jpyyrQ-(E|KSbf5hcg1TM#F<905O_Bbmi;_zWl^BZ$CjWXAQ#i)f_b$09gsJm
zqj6z%=xGz6F{*#{u{o2H_ELo@zyb6OmIf`*-NbI@!flqPr1fG^z&cIE_~=9tU?Zj#
zDH{a&U@yn=9)JDnYWXbTJG$LJSTRpu(EEnA;$VM@nq1vq9tGR>sS0>FEE7ea7lI*a
z?6Oi<FnAP<4R<97sztGCP=l%^F;L=gkZksaTIOhU@FONO5iL@?1U2<I0+)s&Qi*iH
zRyp(3DH=|Q(Mgo5uMwm+?vbLae-?+872rkH?{E_B^3zQaj%h1@;$e@{_gm}>DX=RF
zj^i?ZX8U;kw7BV)$bxdG4yM;HN^UlvCoa&vE*dP1J9tc>w~60q6u{I_T0kIuG4Gkh
z;9JBs7+7^*ftiPeK5nCUdYXNU|0#CsrKbh?*I#0U>RhS6O}fDMb=E>`aysvVf#}>@
zkB~#uH#~2ZI+c#F(rZub-<`JG6hXb+3ur{MeCy&0kYk+8C4(W*e~iW8vQ!=wz;hlK
zaE<Zk`|D|ty!^Kc^vCOS!Y2<e^!&BZZau-usBgUUp2RI{ROIl}N+w;}Q&$_X)KsrU
z#*=rb{G31>|8Or^I8;|sBk{?1p_0<*0DC^>sc$5_7N5wElWO?9>g>9Xk0%M`UNth_
zr7H4e<=MjKlDfm)Mi6ZI0*l9Inx%1A&M1a6$Vh={OfRkQ=+;@LbXJZB?O_3-uL|vt
zC-av@wm)B6wC~xC-;e*MvG={2JY;;^ePU>@I)C}9FnLY&$ISe?otDegPGv#kw)uM6
zC=W^Yg8*c&ATc_Ehy8J3RU5sXn8BR>p;^RP3Y4rQyK}%e9s6>PAhwh=xy!w|WcpN)
z-VDA-t)h{F5c5fD6iVu@RfW-<cVo`pF1gS><Au8>TI}8(Y8Dl3_D+RYP^g_qAszQS
zxspx_PscSlvlE@G#tdoM!JhD8v|EOQHr-fXP0+{27h2q5`&m==bl);sEA8I4nE;fo
zF>60m)+QMp8uXfTcQl$k;%Tv`yzJ@$G2&eO>KvOoy(MX60V{Vvu_0ssjZrt$OUYGE
z)Cv?PF<Glx#9zntxx>}Iw6=7IwyvR-D{!DL`d5RO$rX4poy4+v<@E%`Rq2wq+Wce{
z>U?$_=HtrcmGzeoqRKyUu!wl{3a}x1bK|+j4SqrUfi<ter-LOuluqZmtAngbmXAy_
zxVRPwbxF-Un-7;Y@q)^aU>Bzt9EOi4HMaFTflhhiB$nt2G=fL}`6s2As1+aWum2lG
zgGNYHaq9M5o_(lb_W%LL0B@xF6odX0^cF-&{9B1idv(5Yz&yKZs9=*H$08A(Tw5_b
zvr?<8=MX}4jK^%8`cxT6Jf&8PQN}~%m9y$;yJT+b4H|1eZoTA!H45o}>>VBN+}C^S
z#JXnmPulAaIB)efh>fRnn2yU9-V1IF7D~P3_1IT{@TmN{+Q$*uY8tPkQT7OCZMBM(
z^z&CTJ$~Gpt1todxiZj1>;L&v*5=|3kI+H;`0;K|EPcC{h>iVma~@w2)3FR+qt%_+
z#bU#NWtG{711f5gKnhe>y|eqZ2#1z@T!Y*0b?ql$H?+;kaYx5RyTmLaH>5j#f}*hg
zrm?*po~>&oIk1z&3SM-5sA+z<x`cB1v~AMhjZu-$^^P!a3(Rk^-U#<Q(u>rZ=Io~?
z_jJ9Gce89eJR6dHiP%;f-Gc)=@9!@BuiN2ufDQ%ZqB#4Xcj=`P(GDDO)PWPwphwNG
zs&u?G^h)x78!Y5tz@)-ylttAXwj)hl=Y}SXFG4cDPn*~tZQwNtSWw)WqiSOO-rUS~
zH{ayFuP_Wvo8qR!WXD(1`hF8ml9XWHQN|}2YB5q@$DK_Y+M3j*tgAPq63aSostTZ;
zaleV}j6=5(PGiWSWW+P5QgmF5Mn8~yk9<H~&G7BBGw}vPVinsiCLNRu@>lm=+Lmsu
z0)JfgM#Yr-uM3?Vm_~Lz`9a&wHJgjL2QYEEO_1b&3xAoj?9uqq9KS12xQT_W9mzu|
zM6TsHRb(urw^oXfEOm9wf4HzEEx<bCLb)C!YxCO$JI7EERut!U$$%c~=b9FMT%UA&
zU~DfIRzdU<${^-&-Hz_F(wFqf#ouZ5s>*Ft8wn|AyJSe&pG>Tvtk*g?+fgyxJ8t`K
zJd^o*sI>*T#}KFyjoH7yfu}~qdf#>dWW8Vb>UM)GNFyq$blMQ3@@_xBg=C&t1B`qC
zI7?cBP-9_q@?B(0I8o+um$eQ|5q6SGfR6py`?7Zg=bm5h@(WI1(`3x8T*^WmwMJlm
zmQW^xT-IsCy$Cqz{Pu<Mz?i$Xn>FSBV?7fh<{<4_p9lvRgZVz1%=8r+O52B~)g0|D
z>(`mZBjywMWOr@Y+fnRxJWfm8_+$xPHyo2ngXPqTNz_L<&YAhUy(!+mZd`<$xB_C0
z(pvZZ1xbE`=755?+P4xmuU&gnQaS_<7jWtEvzDnwyIU3)yr*)8f*n1%`(t7--cpb>
z9WnY#3WXeRA0H)cLQLMlmUY!isV%SCuMOhwcakIN6J8YZC;!}F3cSynn0O`G$e);_
z((Y~=8HGdz;@Wz!i(ioz*$dSMjf&X4sI@oE24|f*aSiI-jAp8lz)7qQL``U(TcX+q
z-C(ZIg<PN++7@c&LxO8D3`++!^O(*tO21S(!!h)geLcIi^6PrPSlYL$$-|!U<UTy1
zb<uCV)%3|(f|p|b=Xi3*6!|9l!BHDk1tGR_nsx&OM`@O+B<9pkcBXHi-%VOHLSMQw
z<pk=?IKiubY-IUI=<)_oI}uvTpGZ{rc}?!6?za|*bm*Ubcgjmw!JFlGJL$_4UI(on
zbK*ZA?A~GQ@+m5o1B=YrO;2BZeS(RG{C?&by8)b6dtj*c_(}~pb{KqzHvRFf2FP0o
z;RTc3Paqrb{vPd0<o}5bRQcaQ(J^1!BUE)tdyYD^En$GurBZ45;~Cyz%D#Ow`tE7i
zxH9gCpXpN-pdvp1tVBA5T=H#I2Xi(nit1r!f^^B?04xOZimH3xp-+}r)w@{nO-DT&
z%x?mqf?Su;cTv+cebpxEDe1zXt4iZwJj$C@NFQe3>psdcH@N6>(W{s$XO^I5b*d~U
zG`<Qxjo&&z$L~54-kHS{_rYPIkYhFE@l`IaTSrd6o?(I%KKh39e|y;cyfn{~9?id_
zNB#UKUafMX21P(;qqCFF{{(n-u+USBtYlT{_3-DWL2RWm1G>qLgeOa&77jbneEpde
zc`Hxl%o7%z^Z2{4N6)+lNJ|n3*Dk?d5esjJPPQ_M?A%gKhaQ2&do@mrSuzG(_@0{C
z{C(H|h^9q902qV_Y3JUH9BqWOH60Ivu;UTQ$aMj<V6%XVtcpBe?%YSRj?6L0<pZ5A
zc`iicz6UA+se}^3D}78%26`eTkZq+PCAQcUgO^2SCP`ybv88fjQEzqMv*cTXbFU{U
zBC!sD?_brX%|L_y{%)UjWRLYgdo;=HkMlbErr|Kfd_lmXSN-Mx@p7F&t2@wY`?)W2
zYob|0GhO^%iWCD?3p6ePlB*6T-;u_REFYc7uKEK-g=a;fb@9FSFa@21ywV3Z1m+_$
zfG6}mMKZ_?Yh;db=V*Tob7NYX!d=-x=wXb*1}Tg+`XTA`6LdxFnwN=pEFr7?J;15t
zzHUgT^JyCxAwfub>!-_}yYj`z`EKX>5H81?(Qxx=tEv^^_Ops9)ozD@eoKl8S9!i(
z$<WyRCVX%N9;gPa9NO@WQ&75Y9J+;4(XXw^Jhyj%@;Y2?J`4P;Opc@S+ISRzB>TOp
zt9yZ+GRiJcN??|?lALjo;u(q@y-Bkk^j5uHD61q7DI#`U+a?Ph8>-0!YLpB$zjN96
z<pod%=`EB`g1^~oED5$ua@NZ2(9~|&yLBI`+TM*URyq^4!EcuzELI;Om_ENL(0$bV
z&>7C8#>#4NY%+`AH+ad3(y|e=S^u!tp~yFGb6ve`FE<SLRr81Tv`L197Q#GNtldxW
z#NMdw8Rz?c6fope?eAa-)z_(cPrl6>5oV)geTHKlYA2b)$49{PW!}x{FskSJ?XJN@
zq%IrILQoTJ@oSGIq4I(zb@(LarRNu81YJi!{Aa>`qGN9UWBIA@iUR6;lVIAg!0xet
z_h=CAXif@1#T-s|J38dAW%M8iA7ao?20>Q~(^gI9C}$))ie%J*BTO5@nt6d$es7?9
z8_^e4n9x*{3fQF`t~%9q-evFUAKr(9aA`J7o}B}gs1CSkH%3T0M>X_j`j<f0A2R_K
z>Qo|mq`HeBb7p7cM#a4F-#fDs=;MEl2W2sQu=~yOU}s|sIBu~)SSx`Hi2>OQHI?OT
zPiZ_5)3Q;}g%gYLj%Ej4Sb@fU==e+MMZ@HhX{&%zXn2fsQK&nA43fuwKoogHTK2&e
zRjC!XxIQEXDZ%#`1l;%AdXHC`zhWuU26uc)A6@Ps+icu(NX|qK((L?kf3xmXl4UbU
zs?+bTlsDF!H+X$TOW0g7=nbwilEu@R@VH{CRjb&H+~ewU!qzZC$Q_lhwF$K_;fhf+
zB^@w`xpc?v*oWevUg%z7FsOKXW;E%&Zg2bdDCaVvK8`?W`j?zHJw-6xc;R2fn}99H
z-iNbn3)&4{02!(vjJ7G}3-wahCx_OIT92xSTVG5OdXgh#Xlm_{_E<Wt;4bFk;&)#!
zjtlBt>AjBM-5+92^3anzor~)}q3*J)r*en*?1OC`SI|f`opbO=bl)189&=h5<w;a6
zAKcmw%V`w)FO~eCn`<oNNpugz?@gR+KbDh8B|2@aozqqOw>0L!owtP3oY%1RX$n!4
zy101o=xs2=?r!oSW~JwNE6b2@525N;n8#RFZ?uY~ZNQm<ZmR|9jh3Lp<O!_h9k*%n
zUTMAz1i$kg@{Ii8%Bk6+4tsD)Az<Kd$1iq3kmcuFYSfG+Z{Sj3o=y3Pl0!J4XDF_X
z3vt^p&OGxJn_BGg6)Q+_k&&Tm@bXn@32SzP1D@OIN--RE6Of0y#Cr06Vda_K8`SSW
z!)SzkfD<ge+;yFCH0rKZ3=^lta6!FD_%t747g&+1LNA}9=%n2B{XNL>3si+qTc$s?
z%BsnegEnE!tOr(u92%oj>S=Si#%g|3HnW7oCAB(Er|VDOaH-*|S(5+36OS9yMa?I&
zKl09-kxL`3k2kBGQ-FP9sO%9_*>U=Jt)-)0mkj$`;&uEkjGH#Z&5J1##ULsVGF@qI
zPuS~f`rH~aoyF_GyEOjZI@<U1UOM^Ac>iN`pR4|04Eu?)hmBU<y6y1&i=Hml@}pam
z`>D5yN$LjM0Jskmj3jx7^mPOwgU(`q{BD=f!=QcyNzZ%33V$A~v3Onx_9vv3ixj+A
z)N|ZwtnGBtJq{&w0*ka{!(St#NyGPkgy=kzBWaRaeaTQb3AjEgZ+B_FGd$y_Bgl63
z0rS6KOReD|>y<P<Gehh3;x?6%j263#N&-0D$4O||9fkWONw667=H4yS6lk_&)2Zgi
znt&gtN~o8a+nPA1mtoB~R5}_aAq==+2Xngh<p=F)j*P~6aIG}r1nloiO)X9*_z>wG
zTyx1?sAN{DPdmyM4v`y*lz2^kZH(3|v`HhLwlr)uWzd~!BPV`P=$@n2t4}Hh-NRm7
z8U<mzS>2ohd3(fH==QD5wqLE{ugGFIh0h4*$|hj6bKSrzv~vgw*HXd7&Y2fy(l3)3
zC4a(~)37rhNow}x^o6oWSLnmu5BGG@#_fHih%D+Io*(B2FY7t(uOFA!P=znKtxs}N
z{rBGZ1RBc;c$!%1rSCuavxW0&l@WdBO4x)Pas2DKC4-tJcyFH9=MO$Pr=E<}j7E<w
zGk=brBMkjYNua+fqfD?{k)2z3xPCk>`n9cLL+tmYr!uEHjxtW@&}4QTnb1;r>;Lv?
zUJ8Vx{~?+p->(kx9?|pVJ0ax4;up01I@#!`)>-@rjX)-Qk0IVh9P5~MTdd-lHfRE<
z;&~jR@X>H2^KRwaI;#taF$<M9dWJL8)E{L$)iuoMJ~BD2>&5Z=4-bJ4-!oLa`SZ7*
zk9s67T#JZ$hids@a~!pT%D@%m<`y-ZgIs>o=3q8-aCJHB`766`w`H^g-4LIuv3uc)
z;xS-xtGU0m@}}ckfdSF23-5qNYb167^gf%w;9v>yZ<QCN(nU1acPkxC?>~3n`(6lE
zCifC6m_knMrwUmGXA<wHa=4#qMf<Nn25f&Bu4q^XONku9KYy00Gn%gZX5<P~Vf0Gk
z@dkxS67Gnw3wZPHR8P%dQPx0f@CE!Y+sa*XtG{+oeaRBfe|CYR@fj!<?%A&Xc5VE+
zCGcU2-f*SpH!jAY|7_wnh@S6%1o?NE(@Q{ei-7z-4w!S)_8}tY2Z{FGs8f1cHc+m%
zm+!W?X#KB0#|u6uef@mQso!lDKdbD2UDw<KaBX788H~jL>3je0{_H)6MY(Q{qZ=yl
z+vW0it)XZ*&7=DKcK6%#=)b#he}hZGMg9N6k0p5~dQNf+7Muo*f=I0hzH9NILF8Zw
zq7y0|HOn5!%7$s>15+%tU@yd$*&Oi<$Sv-Z3)bgn`F?@lT=mntcXPj4A1{gFzM7qE
zHvhAF6g8y!kzy9D>k-2f4`<?!$L?#?A|`ed_dAM%DNILZ`aPJktpH=73?57NcjY7l
z!*pn_F!@llKr#mIdLC}(LbosoF1im<?^v}fxt~(~LA&v+73m-Nv)gsD-EROgKLk_;
zOTm@ma0FNXST6hJ4Tz}1^|)^+u2cstwv-%U1L}u4ozq8R4$3OT;QC;-7*4AJJ=M84
zQl1p=W0J;Gy+JXsSk!ay_^LVg{apWJ=oc!x)9e;QO&^i|s)zq}VV%dOT+gOfi1q#x
z((I@6;_BgV0$e`)v-t4^VVLKWLRUm@?Na9g!^joz40t1dxu(B#*dU1HL+B^4_t?F8
zM$8LG?%+-q|14nn#n7s~J6iV_83s#i?NkcRzOy?ZLAF5LF&=(5i38ILcbHq1wvw2m
ziw&=N9=k3>_7e=#ShZ8nM<j9@Jwd`e^o3={8Tc9inH~j<M_nwtQ+mKGiD>4*poZ}X
zWbT;2L_VJnOH&${9C+rk3A`=e@|0f?-Zl;4q2qAMw9gN+!$q+dL*e7cCEv)Nph#+2
z918Qsw!Q5&3g7luTUFFaJg}5->xKvySU!(O_qqbD!zeJ5bSdZkuw(6KK84Ia7hkCz
z{NL>mFFOEpRJq95E%xVLvGnATgekZOba9Koc#f3wo(zGC5oh!)>0#;~(ohg82*K&)
z(xMJNh_MNtN4x0-Pvj~QkBbJjPf+cb!e$*IF$K2fWzb3WfNq3%lt#!<b{K~^R>;L}
ziYc^<%KsAK_<rKUC8jE%>Ewdo@Bt*9{h{{S`;78#3p6ySR(s&!jzGd9xhCS)oeSn6
zDyx|rJZdZU2H-s{Huc(_Az9b`6@YJXd(X5>Vn<z~AE)|C!*nW3?Z~vXOej*j=w2XV
z-@gU?hEeFB^neHLkHFRsV1g8TX$~oR5Y(mJd+z*hPUwDe^A2GWvoKV>(`9#Ed9OKr
z$O%SPVk;2$*l5<X9mq`=6f}Oj4m06OM#??0`*}u6Nr~!P&?!GR*SsSG+Ks1&u%J$R
z2LW4gA)ycH!z9=XUy4$53hW`Mf|w}%`&Q=coUk%0&O4j)3+cCt5PObX2vlev8h=gs
z@_987lM-@ar+5G*_(*!3RcxlPX<ypZlTBzZM_Zqf-BlfKRWM|-GI-w&?R;@#Rietn
z8!|SoE$7vSq2nP%FK4=TT_s5MYDgAddLn7~txbS5cm>i!nQ3ewqH>h!b-V#w937jq
zr!5nEY|o+otJqU~u3iz*tnub)U((Tc287q?!&r97OV?1qoH^kwC-wNxb@O*m<_YvH
zKm&C*)%ys49)>#D+;j~=@jJAtND8BO__O#(Gx@A<ACivxK{w0=_Uea<$$E4<;lnu*
zwEB;IjqNP10x3W~Q#?ipo&H9=3G&0Vffg>-6@7Q^t;mrr0Rz)P(V&Gptpr~uBVVL_
zN@kbgNHo*ojh=Ln39zEo^@Yx}Wv&;3Ql^Hfze>Ul)D3OSe3ql!%M(>)OkJCjDD8;N
zyxvpB-JG0?k{5=Q7Ye_0+F<_XjQY3J2z=W+&Q>$V=W2BW)5HuXkW^C1rrI3HqBd7q
z<~&{#JT5d(9RE7^Fzo@vhkT=qx+Ppv2&L0l2TOkZo~w)nB!S_N0iNO$@T!+T>`Z63
zHDFdnFw_?c;5_G{qOU&Bt11_Er2r=XR1q(mxj>%4_KL*F4(k9sGw_pB;SkxZK8?h*
z?YCU^INP`xMJTSZtS~8Mt{2QR$Y?J7>g<FJBQr{7#Si>}dCJAiYRpe>#6}Ay54wW~
zy8V6KZ%VG>`l597edHfJZXJS0fCjJt%#;zIAMfzAJ+;1%w{>^7M8~bh1GI{oGt`+E
z<J(rGSIk#Sy6Z++!@n^bleh1g!qA~(Kv9j(9VOttlqJN`NAlhsI>x3dabw+4p5!B1
z@jDcnC_714pt&fZ6}x3Z{Dpq-eC>yexep>{N)lLUwiScfZfGW5eQO`SaM53Guxd_I
zrbqGHkb0PT5&MwDD!;jF6z&Z8{}VL3f<eoJ;F-NB!hoO=*v^$_-D^+x#2P7goNeW4
zeUl#V!kSu7_S>D$3Tl~b(!64vKks#hJ(?-TW#H@?*|1eYF(`ha5lKy@t$l7{1p8Ad
za5MgJN_7)20|{8ZSTh|*ACMD#h7F<Pr&k(+^+gsxLR%{MIW1hA6qtTDc{~L^XNHIz
zOyh6GO&+ADRpJyngJ01I9P0+~emXYaT!ED0bA={JhE6vXSp7v1{y2EFRaS?A+2CfB
zmGF9v0Qaq&WfcUVXv|aqlae3HQ!Y2(j?hQeM0KSY%*{q0lUkX;mUZ)EI^ERs^*hhh
zf$r6Lm%b;0aTT<6+<qpjU^GSc@KZ8y)XH`o+lRPQ;=sk!&n98X6|}av%Tw?U1z;FC
zPD>qGp9RKz^tJ}zI~Y_+fZ&y4zK-xsM}ZdLV{^-Zp<9wY9PZhfWcY~%$lya@F+N;_
zh|nMb<nUtEv4Yc3%*Tjpm`}=?lmd=FjLi`L{4VRVz=}X}2Vwg)_@*%r`jfPH@UMP)
zsi0FEXI9i@Sr_jzR~ovYv8+mcWkn6Uu6Z96i`XfO2?CllAlJIbj|=cBNn@>CT24)@
z!h8<)z%j6_$!dM}5;+W|)a4O6PtoVDnx31ve&gKWSok3(!`>B;-<)aXEG{iemd~Ut
z5|$9CXcvdwlZ6kXf%JHZ8ckf_IV373U7INWI0^RUHmxQhZML_lRbEwU5heSaJ4T$_
z8_5TqmPg(ge~cY<wWEp5;HbFyjd)7|R@uG(SE~&DvK?(Dpd#00Gy^ATt%d<gQxvZT
zp2rxlKNZAjur+Bf$X&f+N=^6#Vc>1$dqQZME$sRj@53AO%`TbE!p*@oxPUEo?vKL$
zv_im?n7$uF;evir+0v1&GW0_q@m-6v==#i;r)Ln5Has9Tf}k#o2=^MTdT2q;wLT<B
z9-;wkvhwB#h&@r*@8S4}JL?-c+6(yYeUaVvn{sSAw*>+x=Z5K4k+>7wK@Awkcn2l<
za<znsv#voAC1#Okha9Sb(akERw@Kay53YiD)LlfrY8oGBRcVT~$=VR1MMd{4Y{~{Q
z_#PDdQ%6zKCEdDh@ES>UZitAdDtS8yyH`Qp{H1h)7D913+Vm3Acb~yru!!pVW8cF}
zcZP<v)>NLDI9jtvN_hoZk}n`!BWPWSlE3y^fMMq0RZLo&ryqcl+Vnyygq+NHzD703
zsJ68;M!NQaT=xg3P@0S$l0Y81t%w&I-n`hHn}-j8BWLrN`Q~EQ;H}$~+N-g>7bqaH
zr)n-tzHu*H=4Q%AH|lGV14n{T<v7Mgo-<Q5F1;k@NN-NY@Bv}j1CnFX))!j|B8?_%
z(W0lTn-h+;ET(KSZEUvRF;H51SJxcz%=1CBa~X?h;F%U?<dU7DFW&6~Y1eJIXk39~
zC5c~O!5+XFFifY=PoB+YJIG3-{On2|M<UOBm*GW>)xSSVf4=G{QGcu-^{a)dx09*Y
zexCxPqMEeOs+X)V!8QS6+EXpx+o{j|X`Tu_HJfXE)#j%e8Wd|5VJpqlLJ$mkjk6WK
zOm-V`suxj@B-n2~0k&dY&cP_T);Yq_&4Q57DwzRQp=u{F#km_Aq5huT5!*K5`p;JF
zBApImyaP$`&s^GjfSD&@HSzJ;B>_%5F@+@&_Hvw4_Mu9}0N%dN^8~7yn!5pe&*=lL
z0Oev!)Um`Wo@((^ok|;SyP=%k^ugGmS8_T;?X44MIE$yL<ibZJ*Gi5CUi$|=)A|yl
zGI(86T2qaE>50~r$jux04?RJIA+@SW{CVGujXpexlg26VDethJPg_DiHR=MLuKzyd
zb4attc64m_HiQOsQlnywz7CXbeMZ>LMqi&yqzN$Kl!PPWQC0KPexpnblNciUmH`z#
z<uJ<-w0nO)oe9+=$zIU8w>qFsWQ)l@6$BjhxE+p7M{plGW5}-h{<^sIYLP~cjPqh+
zdD{dd4JzgFX3plS{=+80Tt$?HBsjqu20vIbQ0)Jd7WEXWHWs@=hxfNZf{HgeO2!%#
z6{Kr7*i4(U?u40;Oy1+7alfOg)n@S1nrLJX&3rKv-fs#Hf{e&Vjn6`VtqXNL?1(-5
zX~N|S*E_Ul$GkCcae9<zp6b**lq9rqR&OT(r=+B2YI1|pR#&4-Kv)QEjii@7o#Lqw
zjg=EIN5ztA_)LKQYeA-%dSs40q-6YY(qBU96)pDJ@s=xJ3o<YkmE{CYJDqplGImo0
z`i04Q7jkc^Y{7U&mRBj#HTe2u&vh1*<vZ<NYW&X{a9hn1SKUw*-?ZnfoOsZ!OL*7o
z(B_|5(dsz9?yR1WL1%$)Ls21{LWzCDoJa<7+`>m2LEOoC&${r>P{#J{srISNkGEg8
z8!A?Cani!=vX=~qK@;G6l}IEoP(=Qrl!`;Gg7wYS8$v)TNHJ?E1Dx6WPJ%F!u#s!2
zYN*(pz|HHGBvHP4f>gF&+bk=HN6Bo`RsjDjInrx-bh!@1JEY4x>&VUIXh^|)uDp1^
zntf1(XQA*Zj??du53eNx$eAO1R{Rre%5Xw^b#%Qvev7t^AL>$?in{arKx5pYw0b_K
zZuJDi1I*dm-6LJLo(GFviEd|{Zf&A}d`;fg`HsDO6f5=(Q|U{#5_9@2Z;lY-z~o#w
zBy)TdkB_Oz$@%-=A1u+xUScw?;P&KxcaP8qE;{dLv}lX$uzmF@HPSVh<8UlZ)_KZm
z%Rb$K(Ir<buchXQA-kFb4V6VQZ9_uy%UuaK@*-~jD#UM3ayxoU=NH)nYbVUzWq7`~
zwl*?sd>`Bli3acYjjZ*_Up_Mrq=bZfo8H@SxiLuBQKn<hKc#C$`L%{V`0=cW5bwA+
z56R?3zDhI82`o=)rC$&4eC40>%wMgqgndv#mW%<Xz2PR_)4~AW4Q6{5loIyA(r82e
z_@Pp;ql%ibB9E`F>9&h@PPSPW&6}6yUjNBF`Ugy<99S=B_V)O!&p&oJsrrPiu8A}Y
z(0ofuuX<xF59KFS1jWK3@ldj}vNKK?$Uh6>#P(EoR7b$uvOz2U9)Nmfn=4~{%i0H5
zK<fys?{5k#5wYAJ`0l1q>{KHiM2Tt~zikScW53eEGvkHE!kn*r<}d}R^)KdeTIhV3
zv<6C;2UqaROu_B&<>4HINV{5EmP?4x*MJIVHkCMN;406M3k_?<8?g6tIdyTNaMH~T
zVG3FXV;TmPgi)L-!8AbRJ09baY1)u2#gzy23FTyJL#<xIqY=3cr+LDU!JL7BIj~t}
zEfKGX9~OW9Q*7!w%d*R3P0#N~zJKV7U>2`M4oy`(YB5n#BjD-WEg8GO?i544q~_W$
znUzfDRLYGi)CB5nH-_vMbl#YQB#ZF^_7+D6yD^n5R&v4&StC+(thm9cB;OeOA6V{s
z9(_yd_ETBF3!f(KSJhd|E~KCFD8_>clZL%o1wYF`38b-2D{S3r#8^JBIaYCAsNa>}
z^X=1%;rm&i?6q?OQ^*!vGACa{M$g;v+|ns02it-}h=0P{@AP}>qS@X9&sl4eU%><D
zx8nq7V?Ew<$o^UasqqzeFn>o0xb+l}O#=t#pDmb4GrHU%P;ag4CdTLK1N?=)n96dE
z!qc<TMibJ-d{+&v4k=I;R^10*-=wGl@9CnwF8Ajz73OKfG{Hj3Ue|vSdLv$%{v~L%
zK62Wo``!WM*NQ+(R!`;bZn+NK5I`&=AXd7~H=XwcsaszNmieuH&hy9|eg0f2rO@m(
zLLU*TF8VCEAdj#M2<RU=D^x)R@a=r&Lml@Zt5RU86f6|;$}19Pm4bxhj#i`M%I8TK
zraoA$SRD1jBsU?C9ReP!@#@rl@-V&}_l?1;<<MhrItEQdJiMX&Cb*JdLfMeKWkLOm
zwg6fkJZR1eSO*4#k=oYf(~l`*BX8Nk)U|!=MGuoxJGT>IH{sL*+n1Plp3XMq>ldOQ
zBn`D_+zE!Br}6haj~x2kl=M%aOjK`|=9U?wpx+rwAaTrzCqJRT=>N$Pv(9@3Q6ywu
z|Bm#lRI~=s?Akb`q;)o^HztuPs@VFs&p+g!q(E6qtTLl2!f$EiPO0eMI?Pw5@-1w9
zJ*Z~KPyh!9s@>>}VzTXrel-o$wfiMXxtLD5L6uO-&p+d^nv^v2Dr^*qjy-o{SesIw
zO(#{Wg6g66ltyaqJ*n!gFqcXv4vXT3KS(83wIcVK2i{b%O3m!T5Ym_D3tmEUHy+tl
z37CaDrxjGQ<QCt~s?k~!P_lSwI(BQg%T9xsL>l2hy^eQ79r*I(xGS4w4aVN`bR<hr
zqcX?MT^Zl4Ji%gj*39Zb?mNdqdb1Vo*N=LrSCZVQKB*`hr&W<jB?vn8WNPe$zPW!f
z&dj$|vFhEjfwqbvd1uCNkpY4Xyp!2CxPF}$k3`*x7&j>zCpOu^61c6&Q;M9Vcu>F6
zUMcNrMP?ql2e}#V*Y)ROdbee0i9=e*rl_5SgJ9xK0CnsU^!6b_#inz~7rxboa0x%P
zCRmk;kF@Tw6#!F{cmUqKlKrD{m1Gp?ZhznlUoR@9_3DPQlCa{g_IJR%qKb9Cm1s<+
zkUDMPEDw<pnxCilfI%{b-y@hiJe=!^_>;y`0XvrNi$wqYXP^~8`N|S93IoKmo1Vaz
z-)~QCJ%4PJZkft1!DssRmc=#G$w)xL24j2G>4?V-g4p`-1n+!Ipi2ei9ox6~>d@yv
z)ImrnSg*s^zD`_zhbDquS@I0TLFZ*@6XJxr9_+A@@Z0FEW_!Pbh1>goTDVscn9`pn
z_vhSUDdLgD1wvZd<4+WgH-XcVBUBSjddKQ(udfn78mWL+|AD#^_%uvboE3kUq*;ah
z8E1#%n>)vdi3)+;1km_gBo&zL8!%Aln00P5f%iNswgKt0G#P7}Td2i(N9fTmoDKH1
z??1Y(kuAjEeEzy3dkr|nbj90<#<wHEyy#T73Z92FnWHji-2ORqgr}AP&9M?}=5Nql
z_U=O&uGJ63O%rhWXUUW-DFKp>pEY$u=%9J+rq*3UQwhF}*w1jB7lUfbe=~f{SWAlC
zn2_7dsy(fQ>0~6-7$gHa9g3RKz^S$3h90l2X?dX*?;TQ7+HbGth5A2xM(d%16afT!
zF6!?KLS_%vIX9Bz-eL2o$dYormQ$xs4|j}H-3eEzc#4^DdGG!n&GwxHBB(^#=UUe6
zhUCUoN(OyLKM3TzgX>ueiE;ioJ1B?`Z5|-wV%u#EwLl9x=NmZHC67g5RZylyj-@O_
zTU%yRB_waS(XDrGSEV|JUmlo`Z5z_HbRns_WJ%#X<#pwCd39~BmM)!?2k^^hzG&vN
zENKvV4moRo`oZ+sPwO2u$uT~IR;L*)jPa|g{hPth%ad3+Y1a-Z9AtjgXxz8aj_^M(
zq)>fZPtu?f0(PYZ;Lt_TP8!c!15XL5+KuKvFM0q`o7br09CYum5pZHK^A3Z3L)flJ
zQC!v5h!6fX6L!@rPBmPp;^Aa}a+1+r*Uhyav*ci?+dIe4c4(b5Kxku(vj!LJm3$Rn
zTxme&s?f3LgE7a1)_MfbGjV*7<rE>Yg}9Yp5cPx;Qz!G%jJPX^ZC2Fk_9Dr^&>Qh)
zwr>cjpzPGj31wPFgLjJoW-7vgUN8_91pSW%(DAuZ$wnQYTP0aa;Z8rUs3>6WV#+>&
zPe(*Bi}3W3h<OB6YOGveZ=XJb0U|f*#VF|EHOyT#lQ1vY0H9Gd7_(G0d^udLBe?4I
z2&`^ncIT{(3H#+bF!tL5AU)PUn%C)ZF$e%#$Jx4t6pYg|aPe1KP`cbGj#z*BGp;wx
zvmkgjh7&lgXO?d-%efZm)_EYDqJk*x1A}ji@5;bM+~2(v5c@F*?GDW(okWb1wWdti
zYjG^gPw-Utz1b>mT@-ID7@E1dQ9`mP@yb#;$Vlk*+<>3af;bN;2sJsYdR8Y2`-N!j
zOpZZ@$?3Gg<rH;#l6xf6pP0`{M)AJUDwHDqFZKpLUIi&;Q{{uV?|c1w?&oUPIGxR?
zXv(bE*GfO$ky`u;$Pz)1bwc4)-I9EVyyfbFkqr9}325dymWQidiNb<8vL&x4{g6`Z
zDUFq`+9B~5h{ojE>7S34+0tFAaW<tePdb{v?1{ZESEtLaRd$v}%6HNi0g3w;EF^`@
z(ciXoY=LF}vL|s@0dw)quergkZi6QIRl{5d3Fa`dd)F26`>E7*LeZ(A=ifh*{#onX
zLOU5aAXJ@K!J>}+s@r4hO%g`vZ7cmiF@)wqsjPID@TM&HtUj&}f!s0HTAsd`D1)ss
zYgU-ajTGvEkZ}Kg9{l#$vxIUoB~YNC>uYOCi4)=rkcbf1c))0FA<u|D+0R#*M2%DS
zJ6)jyCo(tS1)eagSl|Lfvg&Pzmsg{;>}1bO2cfHyyg72_A7)L~E23C6XP_*6T3g~9
zCI|3Nl<d@EUg!g$XT|y}<kg(LOhq+;kVXe9edAWYm~h?iV$8HEx&=^=?<+z|`VV55
zWcOC3BHKv>?CueqYFW+#WLjT~{0840DZXD^9~i?CeJx3%P46S`8-3tZP!Qb%NPpMI
zcC&g4+WPZVV8#E(N$Ue0#|VbdY{A@?kr;EHo<O;=4EoNl<|C&&lW>TP&_2p$0}k?-
zP?}wX0d(k^f`mOEK_tA;^QaUSIHqlBk7A&7N!|}|XUh^QCZ5(5PVx@!0J#>WMl0-M
zTG4?K_V;rk9^X@KS{L&y1NUj^LxwzpTaTlY50|K3S;FJq!F)T~oH{%89RAgP;V+KE
z#xqoBQ+uuy_F4yckjg8v%xKW?5;n-b&>gG1rICB)&|M!u#(qPY;N2XtDU>JO@2LH~
zFOa#pg04^D&+`k1;JNEZ=%_WOEFY8}ofUrI+~M)<Sfm}p0!lG4nZ;!1>lzh4;K#+o
zKhu)uW~+U8s@9i%4|A7ama;EfEuZzZuz-#%Hz9+-j;ybm88RgC(VP0(d6^GC^@WNe
z8dC)ZU3jyKosv9#AOYQAI3;}y?Q<=|U>;75him#SfE`<L;m^j+lSDH$D>9=Q>_Pm<
z7^nkHHP!ZxG->Rb<in95>wOrGngVSB$KX!I#4ANOv+TOVu~=gIUteOMoujh#dBIKm
z#<{$IN|m_orcR4KOh}Wv!6=^TOQ^G_OTcGV<NA9wJ32F>_$V{i%Gu;CjrbX1%EVm>
zEoQY~`;;H7p2VuWO}2G}<>s?&V0U_V)zON6g8W-#rh{UJVy^JYnFQ;ZoI|QI>A*wy
z)>Ss$Lr(FF$!;sipcaMzya6LZ<MzX;g#7Ydxi#6}xSTBIchbL!Fs(luQ_!K?`v{U|
zGptw3c8aw7H(m3&<o1;tZ39U*NvYnd1=4*;JHMIpzOq*{f=Jg=*8F-MMnz^}A`jMq
zUER5NzfPUM;X3qQKyK(i^JesL=F6M#1l6GpgAi}_>$91c19NHZLY?Uzu*`9BpFu19
zbRSqT4Z-s44+#V2pQy#PTEs1|(lY+w1Q{TB<1(9s_4s8qtaCK=8h4)deEoLgM(f6L
zGh&jCK<(E9KF%+G<GB6bhmb7j2uG>e_<q>W86>3DB1FqZZRYqS@f(9GVb?>BPU<G}
zfap(GZKBue)Y%u@reO(?>!5$8U0CjhGeUaB+8q<Uo-!7JelAZ#zt(V(-)~m+g{I#W
zz1Yh_f~g!+R^5~$+ou<~TFD=v<moDIAf^uK1jCNIA|$>&|HHId0w@Y{=#m`2e#25?
zJv8K(e4Stxo5Iv@qfO^WjKt0(hM+v+a)T;K#dkf?&n-L{FSpeo!@J1Y3`EM9$hX&L
z>0a7QqO=`C*?^S5)dnog;YTM=alRWDeVL4HGsyiNm2u<PZso%yc9`t(sS|nmvee_(
z|A907`z;myx9$3088iR>um2Unxb+AcRy4h~$N&2B|K=^YY2k*Qkxi=o4Xxw9eQ69>
zfh;E;|HT#lO(FGfZ}I=@AA5`5E5Gh&6c#PZUCWXI3($eFy*Z!Nf?j*Lb@#5ze5B>8
zNYy8A$`SE~i32+SKIsI5zUXhZ0^~lw>_Ipv_RS7IZ+gZxFCepwg3!ehtVxO>6*Xc&
z4A?)Q{TYG8j1}1Vv1fP##~}udT*3#$*q5^>_VUs;5ac|@imq=NTs2A%4vpHU1*ldF
zZi*cCBl2FD;+4X9b_CyJ{S-1`PG*jA8vsD{IL>8I5t!I#vG1RqK8Z%$^TKZX)7KuD
zaeaIP8TXv#BycpB*u5Jvi$B!xVcHOD6i*1>n+Ui=_bp&->gNFTZz}Mg3+I*~EYlLM
zlwX{-_h&BvS3Cijz7JrQJGwsIbSLB{bub1l@$g)G_zDO&N?=aX2KgNqk8*$+iIm^q
zIrJj8sDn9kj9pdiaPEn=axDxt2MZ0~8mQpb#{DGCh($2iCZRu>c?07&;+#~}PX^4$
z_s<^`4TOXBQ-gVtoK6jxQ6j`7$>pG$C1zPlCg~P%+K!-QWgT6|*|}{FY@J%hyWfk^
zs9E^uTic4pg*9!$I^PHVA_iS7lxR4GiB}AL;{C6E<H`@WUVmoMcmRMIuk-qp%^*`<
zp9}4n&c{_c)yVR~(YlaMQp{G|-$C?`L=VpbRP34a8T)6!$Abp!YHYMRU{{G~ec(L+
zp~V<bUks4MA%yWU_(~pi7lKZFdT*}Lmd3)ShZJq5aqOBpvmQ&CuQ`q8aOjj0$G7qz
znDZ{M#JJYvYAQf(f!OA}Nqu&e`OYx}=afMJGs?mMf>$1NalA#ho&8U;OVKZ~3&G)e
zvpG0KJFyO+e-MD_-COTZui^#HEg4F-N8hr8+QTS!{=A=3<~NI{teqpNuX@S~qpP6A
zvV=M=M*kUw$VUiZ{(rc8%ebtzuWeKj6fn2}5v5B+Kte%UTDrRhMM~ORN<bP6x*L>k
zrI8LLq@<)9r3Gn_GuGbje?0Gg-0wLb&o|_UcdRwnoMT+$3ZvH&L7lT{6<I%)6+Uac
zT?R-}&h<ln8cZ_2Y;QbfW#2)xGH#TB_EhqpPbm6J2le+mVj&r!&}12zJHPLrGY}@0
zlRQRxV6gcFKHe~JjzUPhM;GXfsz7_N0!9gEX=)$PA~w1<4Knj;Gy18QDn_rjn#s~M
zfGtipfZ61q<F}!W?Hg;Frmi>@NwU2GOp9U&18Rnu`1{Fc3?V6@@i7-bYd%@N1vUN|
ziCX2;KLBEFm4|sVrdwKFo66nk4!}u<L0u5Fa6`)so8xZe+;R_xy?$aFBW8J1^BnFW
zqrgE<b9S$jss-8BSSP9Vd&9U#Zv&2ID31%hF7Rg7RAmVMb;|wy)~v1whoO2mURUuS
zyZ^mr`t^57N6h|TfM4LwdD$PtOkWT;eLGLytv(e?02GQrvR9I6p<h#<PK0+od=;@7
zvBL}ZS6KCEePfGM$QQLYK~pPT!{``PS3ngTsxSElu}OeLoB;LjBCkS>hXu5s`nK<k
zq|>~#VBD0fiq&^NLXSWQ5dLl&mW0#qKQSXv&kJats{T6Z`Y2EOL_~1kh=GtL%Y7B;
z(8dHYOKDKi8DO{a88&#Y#~%jiuB*c~`#`;a?KQ*O*Fkr9xp$G_ME%hpX6j=FRjZX*
z`!Y-vp(>;Q!Rx3&V!@!;TfxXdVuxA|o?gK|+)%onm{F3+nO8!!(}*Dbbom`o<~m{K
zqA8+V5>sjYr5oy6#ze3$Mj&)-aq{b1pXSChb3)0;i4wIn=FH*|QCPV(SkOC`!I~Jn
zoV~A_xquKWUZ%iaME$6^r5kvXY<Egt_klSPTmF#1O)J?WvjW72_~Hc)9z5IbcsmtN
zt@+X$i6rm#^`M?+o_KG+G<zrG%j@ZkI}sP}=ss<lfhQhMcE&n8cslK+_}2706@E2|
z1s$L<<ieI!AzA`$K1(i^t%mY3iNu{%+bUaH+^!pUo0WMQ%17P$+^(C{#%msLCLMco
z7B6P!_&WI))R=T=GIP`_yGu-0MBEB0x8fNzUs2jhcp{f6e|S)FuO4Fr?4o5YDnTdd
z91&40|9E^>#+KBX*gfb$1son6FK={y-tYSo68Gb1TCvqRO80DQg<0NGhh%;$5FRRK
zX0QAXl0V_Gh-_{Gk;T8pGMOqExUC1H_FcyN-GY6T;Kh2?4~Oj5!<W@SluY>vBv<k#
zX|UyJFD9NiL<R#GzPlAH`2@L$g>c}LkO}aXplI0s38^|4sJ+}b<0jrL-hAaKd42-0
zi-B5qcNdr(<Uxj{(?RI#NnkAzd9$VY`!DKDUOTD9EY!y5!5>3F+1w!KBN2#o$Laa}
zr%+8fFg{5`?0%^`ioN-8N0_O50eFut+Z!I3h4a=7s2|S|v7azqrQ`Q$6oxhpO8}?Z
zv7g|fA*z3NNoYyQNWr$pk#4baKGwpsZWWNsBFIXn&Ls{oM{w%~z8#3gYPVFL5M9+L
z$Q}ovY#Y!)EZYglw;*q!GN}d;MT=F%TqM<Mn2;QGY8VE^7OO@pP$8~+2OtbdxdUAz
z(UMyryPCWcb3hL2l2!@NY1or)VcYUOSvCb7M<zHj-_UyV=77WH2hcupK7pAZ_6p39
zNW5hUb^V!3k<c+I&C-5#oiCu$xu_}0pv6#)P<1#MO}fFK%a@bBC~Q8}W)iT-5*Uk{
z;9!+J0r_Km$IoDt%!0>$ha6bFot;pfkhHklP(UcpO<krrU=K^ruInd44dP5+D+d(K
zr@KzOefU~I&AvIeDDUI1_fDNYhi#JDqHYdkue8#H0$tAS%;6g*a+%{FLugOk3Ru6G
zs1f)Mw!tEO<1|E;>g_sM*(&$lx=`#50H<jQ5Zk3CUfel@+SXRVB~Du7(l+t$<hvSi
z03Ht%KPi%LnLzhgkE3C@hMr$4lQ#(xi<d@M(2JZo{O!2tga<D*>`L;5y$rudhWnUw
zLZNK5@&xsVK&C2zsm}*3@d8{PQomVOI9ENX#8b@eKO?*tr0KGMm$T1I-3W3n&tM4}
zJ{$_f&Y$XuGaNp+xJ4W1X<%7yU5(@c-mvU+mWXF0fb6|=ETNZ0jFN|yU2zZT2Cncu
zd35)g>d4Ua4v4DakYOQC5Hcwg0CU&xIidM>`C*RNznxZ7ZpQFI+Uh^pCS_6Us;S~Q
zZ#L3~2i|PAypam?WQxOdg+oNj#s%C+<WYyMC9)zV;3-UNu(=J3><f5nQewIhcuF18
zvBxysByf>>UjnnN(}sx~bZtPqc@jCPgu#I`sagPxSwn%GL|v+q3B)CAsV_?fRfHH6
zc>Ah5E)EqbvpfYj(;7va?Yi;;Oal0S@VNhc8VAAzsi)}`^KO3TZ)Guh%N;<cM@6Gt
zvun0us&A$oE5L}~2l0}ZIvH_ptyM0ugkj5J+*{EoDu*qtp)B;{^@%I#MN-QgP()Eh
zhem7igR9h0ogzyj6rpbsTt@g~7AIksl!YO9#ZKv%<}K;=@-X46S%nIzcrvs1@K8%|
z;aq%?X6J1?mw059R%DUarY4_ayd_uCReTq{3IVD8kgH1CG{8%GAK#{k=c^viW@g?e
zBN$;ywCfR)3_`FGQ6~=0>m&$Yg6EK%gR2#74dz8t%x$j3$ETUdDhBd~{!c28n?hty
zwcBRNneZPwCj#`gT%3{7z|;i7qpDeh!$5Z0poBI`^$&X3sh_8QUkZ+)Y|8L|ab8bg
zv{RpSIPpSW@aV_eyEQ<NkywpXS09sRs+dunHA?u1MITa?W*ke(X|6g;6Zf&1KIXl$
zm9FAz&Dan>LFP2a$G{gCXHjR;egc+vDRK_7%-Sq&d%%5V4|nh@<Id^KXQ6uW<4CtD
zWU&iUj1sSeRVk%qlkUu6e3m8ce{XxzpcwK|!^hsdaHqjrBb}RBGm8D_quZn4zW+IN
z*rDj*ZshTf`!R7v9n&yGPtp>58!uKeT(dj*XjWc7hM)M_-Gk+K3dJzs5Z4aUo@spy
z2fPGJ{_|1b?yNLq&o0@ldA5H{*r#zg{X#cbZKnDx)@y^qY2uKxg*16(Pc1W*)0o4^
ztIAn!JzQaTxRA?U_FzwB$80C_MJPrFG&N$Y8&@jWm~zaBxqPf=Qd-_^%$&iqCL?#6
z4GF1=6xUvAwNAUq68Z{ZMvAYNUaAPwo;8{%H=kde&ig2?5@Rw}-u_?~qx)mRrCWRE
zIul|`Y2m7^735_}rRVcVT?HNDU+8`%;Udg8N}Iq)sAf6Q<*czxy>j_^NyN4uk4rI$
zg~-egwhfQ8s@*j5C+Ub#o3t}R_HH66-WDlhAMP~%z@(!4rX5$?N1seX0gGyv=y63!
zi_gs|J>bp|t=F3L2kpoE-0zF0?7KoK5GeMm%VPG!WW?9xcvGi<frFQGr9wsH(iZej
zVid}3b_?$jHdgO*KZHH+^mQQhle7?>@Jl`_q8r4uxe6Xh`{6?0z|lo!YXJ_n06)@v
zvn3fS6^*ReRE~hVW9L~91H^4sF^MZ0wWS(^nnelTAx!v?s^fm`uCY^%vD<~6Snbjr
zjP<btYW(@-g^6Q5OdIR9A=|>hA_=TI7AM$p%<OsJoZYHCF`BSO<4|*!rd79IXZ+-B
zt8A47n2>lYu0}^@G&Arn7dFx)c<hC0x=4Q0#!^O(-$t%0cb+=FdN}%u-9Oz3|7|`a
zV#rdJZroNiqrID6(NH1l5g#Yrxl*20Y$@51R<dv*#Fh|dO%hkvKl7-YOY`|EkK9Pp
z{qxq`w|EH%u;19qt=={DtSsQ0WVxzb_<gbR0~5Pv>9|e$*PNR<2$K;fBop-t_q;QV
zTlx<Ai0&%7Uu3+=)no{~P!UNrpRPw^m1wE^GK4=>&NGgtoz)VN9O7m+v64WTBo$1d
z6<$e}plqusu;8znr1KVfe1^w9#OT#df2Sh1&as6-_CwBm?!$Dx;Is?hmprp_gNw6x
z3)0pR!33^OR&eMXokPdFt1^+%BHt@r)Ua!Ft(QbP*B21_jiJGqN8Z3Zl7HwI9ECfx
zY^6CbuU#_0114D_AMfYs5RY7BE^y7T$zu6>>-)>F263jRxqNoMH8Y1#HGF2!Qz;dv
z^vdTWv%-z7Tv#^FraUj9BueMj7p<6kE|4+yNAL<16`#O^#fX=1<EjsNKhJbSKS=>m
zpel77x~1>-_;|HT+;*Nw6n&Q$5O{qQ$8lebLaMC2pW=w=>sK8<Mbu=ApvWcPuU>xJ
zzT9!Ei-Y)~ueIiVHl8aH@gue>W+U7%iEF4}q}U7X425c~-woEm!g-vy0(o$cuy{TC
z-fr8JbHKYLM{yJvy?nv#Nw@Hs2u`d#p1hG!RCR014}psq+O-<Zp;d0a6VeViCFD<r
zQZ##V9bi2FmUr@(oygz!k0(5`Vbr3b7SD0|&s{lF5F-LR3fSF$c>yq!NQ2baeqP>K
z_}Nt@S*uDQi7bamP)J=(lw_hHVK?$l04B@4mFWPa)sIqpn>di&a&SGHN{MF?Qzm=!
z-5o49f`y|gJa@?(-6LR@`l<1wui=X#zhXK0lmQnU%rkftk0?q@2i}^URi%-(FRP)N
z4NYEx;u9N#IbfMCIV@NnBN}R}7_nz&aD0~%X;W5gZmvfo2!MT5;i!gI7Xp7C!8|I9
zdVN!|T=-)t*j&c=?S0A)JD@TQ76XrPXY8)gL3NyVl2>ZiX;&y-!DkxY29DDjLvhg6
zXKmNkN1$|;Gm^n;j|U0KXHm~Mv8M$7mF;@-5q>$Gu0coyR`HCX{3#s=PY#F8vm&f{
zcsOauc&78Y>>Ldjawp%JjdWk0aGW{+IUYAn8OzxA=ufjGlCy`s%RT0~IIpVNaSsD|
z7a8m&MpaFhjO5;sobdTQb}1<i5)ek+i@$nC4hpzAf83BM=093mx!Ygp+HWvPsaZV)
zFhs+KShh0$YCGN9=jVP_2=%2eL}&E&$fK}FjP`4WPv!Eh?UtmUyWPaHvjWSO9q8zz
z{WQ*9@aAb`*k^FK77!kO_Z5jTA-yd2j@eJzLGHW+^9*TFG(GIVreVSf=l}6olaIjg
zTh+INY8YkCr)YZBAxsQ#ODuz;wT@2lMkF0cgd@jby~3)_emX%oQnfV4etaL+2rN{J
z7^gsMM#C#*N-Gx7NoUn5<i`;dZ-GX7HcK|1y9Y_Ez0OLV*E#e*$v?Bp3x>g;L^pi5
zLE9r>`&o|5c=1wI*6e8Dh}_6sJ^zZ{KxzsX&ipy-L-1YUc!KHHzw23i^7nOyHUJ2{
z1h4sj6}};;OtE`CC=5X#4>E`sZp~YUTo~rdyqf@WAmiJSKU7C5B~$WD6wcrdE@$;5
zJ3}#}LTWk(I0#+(K>9&FxwYe3v#@`0H_N?ZV9uVUovV~v!L#hauBOjbxiGp?S={8_
zzP!9g6{#GqOg#2d<=4&#21smPtq=zXg%b4qWCjXAM;B}CZ|?`RE#^o?Y+jDS1V-fY
zv@S3-MfIwlobP-I+1RZ*wS5Rx$HJg9PMZ((60GXElU2no#mXK@WS?dguM^iwi#%L%
z$*S=#!tG*8iEnU`Nd}(R@eo4v#*6_UsuJZ|ovRB^&T(6kVjp&7m(VYYPG(F<`$V#E
zim_gB``-C`VAHe#^N<qtp9k^)e1)$dh-9^xU44A1f>uon-JUl~%|`p`OXK@XmcW!v
z^dEh7V<Xs0yuF!q4FASPZMHc01}nv@{a7?AWD7(nxGcKaN(17;DaW7}8F18KXcQMT
zJjW8tr0RW0gE<<&G7Q9-=iwiZr6+oQ{6hS6I}PgW1uojpV!Stnuxn;b`n`tc*eM<7
z-L3~udR!bPU+7|AN`OIj$bgeo@8x`uCU@r#WS>2J`~2&d1YiuUzu?z>G*I;vZ&g<F
z^{aB-3g(&p)*L_1Kz+cnKaMTZI7~2`J!;+m?m8^64@faz;RdSOmXI?aJXmaJJsU~(
z+H8aki@_;KjHLtKEUrv8{+rR4%<h{0y=?T%ba!DV%z$eYR{(1aLP$atuZ=;J$|TOB
z85ozdr45x9S7PaT2fvqKsEn`C<wC@Yw->*oSJP2^|Ks4@3MatL%<}?3-QZW^1QXM6
zzQd2dI78O1d4q%cE;sCiEu|(Jn-fE-Walk&G#>)N(CFy<KOFLH*qX-uS?7N}AVRLF
zZoz$-myw6G9?c+R3ysJC@y@a9IJMMAg2AZZOkgv_EO6J}W|tlJmHDVTv08Yki4CNO
zuYSa{X1U|Ny5jV8)D*h6vkh^&?5bAP#naUN#9TOy6hLUHqq0(&&V}hHk6lLPSCG=Z
zcxO>fvz|kw)Pu240deU8P{yi?2B&RJ8pe5bGix*<R5eWYc?h*6k_wK}J;*MznQEZQ
zMId{Wxmkitab#R^%e25GwWilI2@C4Z7X_``VU~ayzQO7l1a0VLDa;>$E%*phFy2;<
zqiBF$IRPy)vNuo+q#r=r$j_7Wa%`}5aBllAK7}XQ!^@B1)wtWi$}H;U&fMD$j^J`9
zQ0~w;<e$ACIVpLcsfkJ{uf;M}Ap6#2j7cJnD`I5EIp9U25p$NMoC^E!hoxQEfG4IZ
z;s-Tabjzmw#%pKj>8##t{s-%5W_>-s2MlgWT|u}+Ju!CD>x9fgc#T$k(ujY{m4bxb
z*zM8$H|JfMwnl2*k4>RZz}u;qU}XW%zN(D*F*x*k?Oo3vGLF6O*CwQN+ciW5Nq59C
zog<Q2QNlY*v2Aw8s?C03=wqY53C=?_3!>H(mSkd2_pbA)k=Q8Hg(v}cyza`@sOdF5
zVhb7r_u1cy52+N%F}%N8HmtssJ^x3+7<>BU@fjx-%`c<D*|=5Zu=Ha1rj6xPr)arG
z9=*%uC&nBCDT%15M)mHBwYvpN=1pQ+;|LXObz)-Givu@vgxzh**b#k;o?Yuz@J>00
zOf4#+n~=m(8Q)u5wZq;{m8iziEA5EIaTU0=eG=!GHxjhZ$Q?~@v>#qi%6_w6izul0
ze{Mw8V(PO{u@hhX0v5iHKxV+{VB9`pP}8x0p(oDzT)TZDUsH@odu7g>>!W!urWzjW
zz9FcI7-yQz9?0QidMJm6;Z7--y^bU2oq)?W2l3_?_@~vXr!knQoSRB6`Kjk^jAX!9
zY-(dK#hv3?3bdV1t?bShkz8`gR`B7JPX}Bxns4RmyPG>K^c=)>F=s{dfd7MlOFroQ
zl5HqYzW@vRA<H7W&)ETOm5(zIdiOQ3EmrV8%|I|m33TH=x0{3yzQStfAa3`usbwK4
z7R3|b_##$b;B9PmpQIgeE^58RN6c;P<J|e4*-mAVqMsi<P+@WJ-c*B}ec3x3`s)v}
zLWcsr0|8e=^P$h()$!`8ifC2~O!y=vASU(}$iLN#^=Zmwkqq({E3j{QV-SA^#A0-m
zha;zk;)k5x9~7j8+_OxgR|J+`>Vds)MKIgTYh`wvoh#BKHnwI%f~!w1w>ZQ08qH_y
zyLBaKP&}TRtfI5g0B(a}BCboqD5dR}Sx(F7$p>S0KtYlK781u3SNjCk)thm9)qSzu
z6zF<WY|(-Pu<&7AQ!RO-0M5*XE$D})a^?nRD+VyF7yyqw)au9>x5ZPuBLSh>j70@0
zlRCif2|GVMFd1fD?bDe#m($J}vYh1d4~?P>R7_~6UXowO`XVNsy4?o>dz0kQ;h$f+
z+#V9x(S?bT!Gt_RN=SvdLPi56EOk#c?+3!+K!TkKa_Bw-r$6edw9SVSD#`W2oK~Y!
zSdDRhXU>ac<PsmdFsSi|>u7jV;FXxI7C$7&k>sYPj&35=FmOeqJm&_k|9DwffQsqe
zxH=5hkzva)EMmvu2Z_tfdi7WetA7z4Kgg)`<V4b7<ieKI1I)26F!UqGbSAj^t=d~l
z8up&VinEadfE)-HXS@ec{d7u}pML*+NrF~MT@{^0(B3+g4|q1yRrWZC2X6rp?VNqE
zrnCqt4&l}D?}Y_USZ?O32Pzf{OHv1Vp1gT-D>x3U905;8rC2&NL?l-RM^$m!M@t`D
zEknC*$V@;7vw~7X0AI~rSkOV0UiXLGTkYE#(Ieo6mV|%fMI-qC6Ntsz))GhsSXep3
z)bi2p`qXWc&iHldWx8}_!Ra^Xw4+t4M={&>Bi&k=w?vGTaI$CK@U|WF7q}4(XvR|-
z<k(ti47WfePe%gv47JY^LVtoX<kRdcGM-*KyU04#>k~geUX&m0E2}HeMu3F|(VHW1
zTv)6!lWdoFGC53~bx}15=0gudC1JYS7e!UNtd60A7eXqQHqB69RV8(rJ#k<LJQNUn
zL?v!KyIQ{{SAv}n6Bb%IUmS)DD3p1zUxitNqKxifisJW8@FLO_=_MBb)f5q8*cERS
zx?e4+2IcibvhGkK{^c|&HZ^7`w&D*IE|~SgMRsNrtpbWh5f4IDlH;F;N?Dnl9VKFl
zp?enWWsstp?Z{x@if3;Q3Op_zJ}F6R>NK(`x<jcj)NWczL~Sl66P~~iMU=gC^0jMs
zU}^HdUMsLjnVovP0#!#9l0j5ICckP<q0Z@#adkG#B7&?ed{*~*z1~g4*RsCnMhq}r
zAud{L)nu=&$%!CXN;dNG;%4F7bU}hp`pp^M?`kgh5Ox$@CMTnW=qITom&|^uA1~VH
ztJE8AmI=6qA+m-k7fH$MWmUyleFE9GGNv*7OmuD#M5uA^$_^737f$Ll9n`DY8&_S=
ztk~jI&A!RBCV00D>e##qm1&_oR-8ASk=#;r3=5Jx({s<|XKwa$mob@%9!PG=)4K}t
zh_*;|M)D@zuh*<O!jF>EU(%V>_#~kHEf#eq;9&M<j;+*a250i7G)vMxc`N!@_U-3t
z%oRi$hN4{J4_3(=C~C^tCy!*)7a1H~OL;p?D~12aYQ#B`O}%!mVnJ3fKj7&M72ZGG
z*#DL!Jrg>G@82wHXBRdTgUL^j)I=??WUgta_QFRJ_v%wX{xON`pn0#(uY5B$7yS&A
z$BsJ8{=qU^Ihn$>{%%t9pmY*CpBDt(`w@sYQDwx7gwgqb6~+H6@db_SQDYlv=)bnI
z#ppUfu$R3F<*}_S?QVp&N9y)olrLua3?A8@?V<a@_heSR>?Fk|HKO{WYD<Y?wWpHi
z2)>kAPoQG$K|<X=A;x&wR&=EY*H^Zm3<GyCSMg(!eZ21RMRgs+cT#`BX8&(K&@&O!
zKvFK2oG8$eYIY?GyxYh=q8h9L1BQp3(5nm%RM_8B_v7c_#UOA&5@KR(5iU{rwx091
zG($@mZV8lvMC0Lj0XEn95(y2=(k)I2AF&!REtXy4k^f~()D-m3mOYF9KoH`^ka_nX
z--3zaWPr>XO!7OL<|Z?2dXT{h;)hxaotVkE?e78RUqqwi+lb){ZPS09C*hJpDX|$~
z7<~PPt-)Zv9^&&x`_$eS_g}_Q|LZkGJrnwG4~yS+pZ|mH<l#&QV;Bm<&qRM7TmOS|
z^gsUkzh6076`_S)73cYDHU0g-{O^7%B!KL5L51INfBV$@_xt@tYheUhOIm0u+5d-6
zOo0%7B!(Vi{p~CNFG$s&f6EX7%RyO6>GuuJ-@dM55Xd0<X&+VoCE-c_hQAAnuNCO*
z-$L2;NgbqZ>0I>}Z)sOP&JtI80L_UMM7{4}`_BNBX##zA*G?=ya+%er>#Q}^*m|*l
z_vMr2k*}%pp5vE1HIu~yJV0BZs(H7bVcZ^B+R3{V-tXb9iB$9HtI|w)51@l6M5sy|
zv5TKzoM3_&Dh&J>F154(np0OB)0QD<!Ed{4KN<mMVlLR!rm5$Y$BSKgap?xGXJvVj
z^gMo<QtK9&A-#pHq;Yxv?@e#xdPof3K9EAmW#@t6AXdl-7^M(+dl4wJ;C+KxGpFlo
zZnIkRp#Dg@ORg|yT>(Axd6&mPo5b5x*IW5K+zSTl_l|)l_3k-D##Fqi)QWI?CPYhL
zXMMTW@Rg&}O^)9!JfVK%BJe7cc>lSGRimEizNd2xokYx;3lWwEOxG+C9Rhg$MD~9P
zzpk<j1f){+JZ-G0Al-((0mT1D-KAY6?O*((w@r%w!9VgJflkpFC=DrKkH|0A|Gcd;
zNpRvcz={M-*VTA~o}b>VZ6la4tA-A*%$PHSrz4iL9V(vNpAY*Y2TVcn;>Eq1>~-NS
z<<VCO;}X5hP%%d&C*!SmsI3+beLpx+fWc%5m`d@{#*V>%Tn45v>u?`H0ES2sER0C~
z_LfHSi&{v8x%Uqyj>jt>e<Q#QY9{I7RV9_qX04wDzUR#u2KBP*)nWb`{Ov@mx36C6
zkA8tUHH=3veA~C_k1J_Ne)IHiucRwNBQSXL?bk$Ck^W@B{vSq501;in6eT!Bg1hRk
zS}qS28J9RijroU`0Yo<8c|h1pc?F)d<CP9;`hQ$Z*fM(_cE-nC+1UL)V+jLAVbzFM
zSzCbmqN&#TK)Tk*Dc=A5Ig9@Lp+1M@0rhYEj3gb$j7%H_V#XC00@6xdVTJfv7XeO%
z<sgkSOm``~Tyj$YpzLMX{BM+0gtj46m2_+37X;`C`{dCL;0)bE{BfmLW-krfG@ke3
zwt+6}LctGyu%a9_()vT2b6^J~HC+1$mkCG`GKQe(cZlaF(h-J{8WhplK44yl(tV+!
zB88J+ZPN(N<eK4jF};%AzphEhNGZTtDgwVd@y^kY&yCjduzxdI<Vu(9ouA>#m6i#3
z_;*|2zqZzPEZEQ5n@x;@|2Sw0X$XVMB{1HAQg+yFBP_{V5NR2<AkCPOXcyEe5=`$K
z`w=D~-+c7_woALMQi~vDDTG@w0Gbp3(>mD^;F@_#aPrwHkzG?B*>pJqKSXs~ia3t|
zJK{<{ZWrmbioI_Y;cO+g3Cj5hSpenHx*{j#$@|OINOlL&qFvPtOI8T<N?TRpdE7Rq
zz~T={MNyVuvJx|;5X;ee`aH=4#8gGpCGF-ecmX{C3w1k`$ADj*HB<&3kQF`l2+r&L
zeoMAn^}OXFj(fl_D}|<`UwDa0J&Yg*k^XE4xgXNzersU^vGfP)Xst98L3p+e#c%z7
zMcTs2^L%sa#;;zh9_ORnG9W#z1k-6mcysa?woNx_BA(Mmmh=@mv>)&z$x|cHTde>l
zhPbNV#+I$vIRf(7d75TuH6|`z((n|-S?f(Nc6kg_=O=KLM-ci-)6<HhBZTeZSpt4+
zJlM{56ALc+)*eaR)i4o$0?vuP>wnY$AXv{|Ozn7T|LIfM5W*J~pNu-2{&E2Z=>pjD
z_`4A&3n;Rg*G$}BW@dHJCgT0pj$Mco*}3TDhX1lgBlIYqP8l^!!{3$qK_~H!;kn4J
z98a8-%`Uu=*0MAGAp0C#Y~HdQk}eRsR@~5f+k~*YQJrWoh`Nw(On`B{$-*ksK2D@K
z+x==qR93-LL!p@`Gq-28(}P=p!IbI=45pfS{gJ~Q@artFn-9q~=&TlTXcK1odD~oc
z0${9x>r#DNkpARXlDrQdB!w{QF{IzA1!pTp3-0{^+!38YDJ!?d1K6U<U^3+w+>fhw
z!>^_I`U=GnLKP8_kdf_&j*?_-47S7V7Y2tk-b*()AK3u3H2VWw-lMQW(g%EXZ$O~&
z<@hU`D=Q>nxoor*iL<)9kTNs`8Pyx5!*K6xR%czm#b}4wMf6c7*qiG(X}>_E4ogL_
zvV5}r*t&Sj=pZy;_3YfYb3IqG55RcdSK#$KHm)p%gfhtVw7u-5%Z$!9JEwh(BxzWw
z*Rfc^j+~<3Z~Gfid>dop;{CX1`DFSN7fr_@vaS%x$o)~(10v<>Hfgh+3`g^h?2JvG
ziTCe#KW+xkbIBe~PcR-m@ei6$pP>1~`ZL`wf!Ub(&{u73?NqD62&&ew^=F0OzaIAg
zs3rgUk>epOW|GZ8!$*Jq#sdQrhRAktQnZ1Z5Yh@64Nk(rh%vU0O(}R<dm(Hndg1UU
z8aT7r0s=4I);yLp_>)a!WB+gCAB-R{2aF~AJUa>%^8#~7Gw?lEAWN$in*+)XH9jCG
zTKL!XEg)G504}oQe50EVW|)IkfPo`E*dQ=d$g=<@ZY-bk_f;IVeIDG11m;s&60D7p
zyTc2B!hJ#@ECNGDDM`AMz0j-ZDRP7_&mfq9CtLq#XFy=j<N|)n+4NL<4NUV`vpzjI
zFk36>ysGE|%hHoU&dQBs20WlHVyc-9%T!4a#cHBqhfM_Y2^9UwUU2w&@TV^l*Si`l
zH{0W5V5B_)AGxWQ!6-|J3@imUsgFHq-7eV{Wg8%M!Ws7TfzNQ{HSE!F)jyS*3Pj(z
zD2=$EYX<Qy6nCO8xPmIbK15O8C;PaO@+X^r;$}(88XBHgKVFic&ex!`qsLw@u6gWx
z2ZWQt$%7>p8hmr%DoDMfW%vqRD!rb?wgL!sEix<-9?h^>A@S11MURi!wcmh$Zl!QC
z&)fD!BWiObv)~b=_ih7$*l$ybMz@6a<UBAxlQdL$?PfpPv@ks_9zEFp(Y<bG+#yH$
zs8Dz)i8szdhTsWeY$W_K$Nm@&e@>sr@^QU~zDK|<JwL6LprZYPKwN&^rq}2IUdwNS
z8OH^Q38_nswh5akTs(56+KJ<8vesO^?D$(<1BW`Y=9Z!ib$_innm^Z^G;Lrs_#_@7
z3@orvIy2P|j&&prk<w(Jt6g@eRzsJDM=R__{guFD=xbCQo%|IDIY5mg{6NGi8v=ga
zP@Vj63Z{3zj*SvyBPeKMQ3dPx$n&59S%s@MxlKnM-a`A~-6);y3vU6STS=nMw^>S0
z2O4FODS3GF5&Ud2gYl-JEU=XW-)-pUG~YCsnd_i6;4|Bxv*QQxrKi8t7F}=HR}~DB
zikJP7KgO=Ahkca@)J9AdC-kiXTj_SnAiHXgi27=~njI^3_`uj7zDknjGO3vecK8J7
z1ehmHVz)mpE8$ekzDch^7>$>f@2)<w`>t|%LRnc9=aA#}Cn+QGgfj%uh?D*kU{ql#
zB{UiP=*~#TXr$Z(&?Fealc!`u*;Mi!e8ip_-SDdyk#f<skhf4;5}An_i4(IUIBX@7
z|C0-5O~_j|`!f?ww01V2HC-Wz0<)cW0#UwEb*LnYIq}{PTYW{4EOjEonb!inot220
zflH1s1a%<gaTy@p5|~oOG=<H42r2^%kL#t>On=37_SE6Q7{rdz?+*GB#B?nk-(Qoh
zOJm3{b3f-fje!w>eOGLh_@dX@Nl$`TxEPG9r$_y1o|Ae`-W5BErORe})iV`EL@O*F
zifx4Hps3`Ps$?r3&fqWevS+{3WdnU&<x;uuM4?M>et%$7H>0YuvT{+QUu~0{b**TN
z*wqKcyZFQz=4-h7CwKU~yM4>?FqTal4gyf0dv<PXUsy2R0SMBJMl{}wa=^m6T0%cI
z+!)Vm=rJ$`F{Z9-@`D8k{cZ6fi`=+-?9f_}GWoxxc6=!_+SsHSF>@e9HZeTZI6Yj?
z|0{g-@8cX#5@d#9{`kN}|JPr6;L;`sO35eqOIwvZA4jbXWMwj<$r+j7x~R@hsSXiT
zDu@d5qTB2kdu<p~vGp2p@^_X7yUZNLGV&GW$x}p(RVU{_I6T{306_ucS?#J{dF53#
zr{%#ZRk6nR!V|_sTgKGyn@VN4=f+^@^L2rUPjQfm>%qrhBk}C_ktKor%wZ&9CFiBK
zW?MCEV?W^&?HcFbC2w0E2lF+Pnldu3%#UqZ2g#om%iYk?GK_cMRphN2xVgo8F|TFf
z7Uixn$}?$~Jbyn5gKMo~*>$7D@7M!j3zygE2jXB$=Otp`MVme*xji`kJy-T!u_)Hl
zd>Bq>S>Z&e8OD{|vSPU(nEmW?R^$DziEjvrw1;<%<dpFfpxcO3vCT2i#p;!B_Dw-6
zH3zKLox*zz+LYUFiv){NA5P9^J<m(<4KIPjC-WhTxZ1_{3#$B=D=DurUVN<iqEO!`
zzp3BDj?uK(B=ZHOy!}<8!9B|siZ`BjiC1)D&pa4w>>KOZPo|M{=#{$0L$u^W@ft&r
zloTC%ZqDV}Xd^$t@j-5Zi)MKg|33v2SBjokC6WDCYBSMYsuzw7`4(Eu_?f2!zlxHj
z1S;C9#cE<=2x5z;nt^>PePZu=YL86HCbX`Di78W>qI=v**O)RcsLMC%4RbzCjpyCt
zL3eMqcR_>ZO{RpOYCvVDc)?J2fOn(tCC7x|r{UAbEjKA<@x%Mm>+tNqLB7iq67%7v
zOkp}`_#VF^<+@F6##Wm>!@S0TnUCT_3}M7`)D~`K7$(1n#L@4#MN#kEW_ymEecT<J
zo|-}B8fS)8&|D^OY2bT|+IwlylqkdLF2eehBNC(An#|va2zP}$e4Wx)-2;rP&n{AY
z<xtI4DZ9&>GfB9_IPjo@yg*N%IzERm;q!W}6)@lRpVVIy%U%4r&?oT8Z1;=j(np4j
zIy?~`G{NJ=adVp1OjEwnlk72CJ8TM*4#3jkLDRj=>OtJ_It|HR!4uV>y0}DfY4vK?
zKV(`buOiXqt1q#1{tQPI!pOpTfRER#7OO_)!MpBt2a{F`hvqW2^0e4fAWo-O&<Zo8
zPSL;kxdNbAKFF17g1cPCVJejC)`do`x{{FeHr)^o@%POswnH+H9V~jf?67Cv4utp_
zE1SG6rOry^-*dt=F1fyb0DOl}dT*^KYA$q|)e9ErL4t%FJ%`q~lES1$|E)>S!3>3@
zzLk%+D6%L+-%f%Z_bBv|hO1_%wEB=+R(X<IdJ)YVxCj%FZwDw8(b)Kbt|XQH0{dTI
z&nw~@rut5l@TeGz7C*bT{u$LwPatcLM|Q*!?i8{51j6c{6YbWgTv&C3nuDow+(=-c
zutxoiuIh|350+adrTr8#E%R$iJ}@7B7q;<s)iANn8qO&xwMxV)cpP8jnTuw$wCh-7
zL-p*He@Oi6Bge2vxxTnx_pg9*MPW=1{x_LnQJG)!-l#)iv(gK6dZnUx4zMRyBS-vG
zK?%>7O6^G4&Sl;w3DO{Hgb9%O5;yDt6t1b%H$?YruK^lEKs4b0JYUR5$PJ%ggS<lw
zOGCX2fDUmityC86)M*i9!2AF@XyRgfNZHtcD&k=05})D%qmsrAgb`h2Bwk>gM~m0~
z#`l4xsIq|5hT2CusQ)8PHC43WB)_S)HCX#o4e6<s*dsDk?x(wKUP8>1<AJ)H$`ayP
zK_a`GotvU|<2WlLk)Y!{$Mtr;Y%7b~4xGzRruoJAz{QG|WL#|(z7L07F#k-*;>qi{
zS<hrRW=pHfGz<FJ<yFJ!!O}7UVNi>oPa~=yepc}V+6t3=p@jWe*l|(l9zvq+0Nh@&
zX1i4Cc&z(rEZ%RQ?%DNpzfBdHVO*+((n}1>D0LNmyHjmE8EgGQW}QMGPrj)wIyxMv
zmtTf`ww<Izr=I(YmU5~z9y)(1N@9E%&SO9CkBbr)kHRu8EH&o_N=|j@+Qo~RJ}+}!
zh9E3Vta)YwW_2=P-fyJC5026!wRN<y<$gW^eS-qbqcVyAkajOLMlA&B`<1h$-k5(r
zh025<X(*oOaO7}fplA1l-9dDC)Ak;uw#>v5#+~Vj@VyI`r3{$(Bj8X-6{YtihRkhu
zDmBGv4(vZs=p};ltxTDeBH^GHXovRb(B%brIJgS^aRBXy{&@g((K6)R@Sw!hiRMB@
z-nJg+FFionIZY2B<}anP9JiMEo;y_`=1KmH+Qgl;Kld%boH*bd<sIGp^6BU0G+0hf
zEw7W`zPxhpbjm7_SD6-N85W&QFmeEA*MO4mI!Pzr{a0U}FW~PAd~YHw9T)DvVV{PY
z*IL8~JCJ(Y1)OTc8v{n-0fVHR_pB6Z1GCKMPjp4GUQ~V|!aiKBJ$6%y;w0QPWJ#C0
z90hslL>JTQ6uezD9ku^2FDp6Ci8F%#KK<Khg0CK$=ytvJ<q{*uZq1lZFJC--51iZ<
z-!au5f6=)9FP~U-+UkL*vC@atjl}HAkL}0vu4yIQ&?$Y47>$W)>JweyAv#LSuOUmZ
zeD}v*_2I9*N~FO##RmUE34Pe}sCX`G`~i+t1-;2fHCcRyi|NL$66^)gtC41i)C`$q
zM)x;l(!`q&hnw*C-l5f&#|pbNybW>y2gU|8I2zj^37Vt8%%87;8qV~)IUg&hSOG3!
z!t<p^)gUs}H#+F+XSes)N3YA|vO1rncGf8|pHt7*y;41!1o7&_S)UCa`RjZ4t4Y+#
zTBI1*&L)piJ1Yr@Y|@;w%euZqG4JzL?M!WxL?8(#BfA*uP9a<Aex;v}OY5z_ilu+W
zdjIzinS7XURyllb{3}XZQidwWkwM#9-4=7TYtyTS_jIhkbmOA>^~LV>=m3#A7}i8*
zJ_PK9UOxt?7U&fqWH}(#=W(D^$3b+96{b;yUG3B_+qESqBBjV%hkf5TZ_F;hi5pp`
zFt#4gD_#lEG|`GPB-$kSdOaaADmCR<Ga7I}o_(&AOWb0FV)m$asR=03bUCLM+Lc?L
za~!#)pSw$~>YkmHC%-xd63L~zfn?wM=2$dy?*>wT90ttvrr=piAGTGfpyl2Z+Y5=t
zYvj}^knhI(Ry-kD9rdu_NI747Ft+xn^J^`+eKc*ow>{eDG0DIvW>md}?T1BKO+`Nq
z`No71duUB5SaW6-uo)1lWkz6*73fa0hA-_D=DKL=dr<Bat~2~!uGs%$BQSz|;6}Vv
zFfuuRgaB+-lXV^`{Mgre)0wBS*Y;p&WebYC905Y7`q>B%LK1K3p3=Gsa{#aHY=&Jx
z9Wx3PSkJN%_WfNL=vP3$&wIW=$A%o#wH<6J0lqRE4oqG6ppBuh2P=9Tfb<=Zjx$x|
zGTG}?X6lz9X4=Hvc<uUwBs}9nkP1blK3{4Q=Z*EveS%pY!v;)JO6*Y2Jt=uT6t&ab
zWG+@pFNo#V7Mun{TWH#-+Ii*~=H~z{{|Z6E54&DIf1{W*sNf&hedqkAMJkazMq{E~
zq+!knwO&57($qbK!+Rj^EJu`|h|RH@!0pG?=~!++WGLa8u!M(U73j*3QNI1^4SeSx
zwnwqDqKFy=tRPjLo&$y$T_SSce3`!UXv9h2Q_$pSqbp1@QgD_gsj{gPEXTzL`BX2*
zkJwyn9u24=j-IIB_c~vHL~^>{yB(F{nYO@^HI@1)2G9P{%qy<a%9;69m_}GGO#zx_
zxtiYN^und^lBkUXkUcA)*B2bZJig50sX0P@`(<Te>QSa><0Tz|pK>^7q8AW(_U-$h
z?v8|Brc*WiqP5MIE67^?BUE4$Xv7TBc$p|;s=InHwX~Fd-8E0KEl6{jsbgfn&`<nB
z^uoXD+rR%oF$yQwr{x=-{J#rfba)+LtTc@B&lo1GQ+$)c;fNGfC7W8v@=AETJfx(W
zJh^a{Jd-pe^HfA;xwxuoxN66Snk)@KAnB&NB1%!WdKBYQJ2g{jD30$CS$+iPYI;U_
zg5nhjP+>4_2PkEB=zb6pQOwX5bepZf8Dk<*)=*`mV!Rj98g{K6grJOzhgZR&9e^E&
z<Y^KmTmQ`DmPJLS=Xwo$Aa$}_QK3#@)-0^6YCcvq^vo<*R85{<m=Y4_czgWwK*zQI
z?8<P-J6<m*rzHmB_Iq*f@1tc2(N19?CwJEB-_J6Vbpn$f(ajC8cM65#PN^@Uy=B0n
zj&~`S7D_bSxv5y+pBK-ESTwT<stVCD`VyR1TN*P@LX>Kmq*@HDYdGUsP!}mdVBSR@
z;J`ex%9d1CZDL%UoYyG8U{~@;F#W2_v+<Euy!qM9$6poWCYr|+KxZS1*flBjyf)BK
zJCnHsIaMKmG(N|j#T-@bGycH%mVoF{D->lKYMejOFMWK8N_h&bY~0+?<o90oSfhuF
zx3Ve|!_xITgl+oR>PrW*E_Yg~P)FsNl;=0xQ~Dks4^|y1jO;DZBaE8iu-NoeJ4|MR
z_ISbTO36=A=Y#4!W*F#1m6vFc&xnco>QqauVX<6_8_`@VY`HJc=v)OBCH}Z8xkuoE
zP)P>CaH;vBkfX(>RHSb@o@%vx1uBQwj(DEW`?f##7G%H%KKFu+;@}+}Yc=L;i`b@X
z;}HMrrE`{GK<$S+cygqs3}bELEV00yD@_jKRzLH>)5YtGX*<xy=1n}v=S!a-9lqx{
zW46QggK(+l`Q=wWxLj!<P8&G$-azu+nHKm2exm?Gz=aI*_cgmj^u{l?K|uD+ofX5B
z`Z=6`Dt~P;F7Eb!KWVi2s~I_w`iGN~x2mlw_@Xf?@PGHt%uyYkQ`FbC#k3;1^Ryi2
zg(s1ij|@ojLy!W_`o6Lb5-SeoNCAkSRW_m?#yO-)_53Iz8jR~rhjJD!B#xpg_xOq<
zLW;fl-rFVkOnqb}w)2;t*D$q7@|F86+q#&Z*nFh+i;ia3q_-xYdvN6uaiE3v<eBf6
zF%^ZU%!=j4c`)rsoYetmbYX43Lyf6o@?KGY6i3VwH6DIT_yK0UJX;oNEI&Yn)YQ{3
zVlkf{FcmC-z*r1J&BA<TOgB`mX8^C7wi+Dug%$bv;Wr_hqlc}&UP9GL_cMmzdrD4f
zy4<%j+9#H9#6lqjg8*$1cB6Et!P|1N+r-l-mlJXUi;9{xvBG?XZ}(iobVkX}@HRb>
zMEoSSjK5De(?1l~A9X&HLgsC(mIjU!njEu^M(qAy3-#|?`uUSUe_TPG?#=ccp5%8t
zgxA(Z(@^Dx7w)s%Tk%T1VBzCSbK8SU%u3Ub+Ro=IpOontr5UK%85f_LTAt_v_Kgnt
zJ#PmIJ!82C@k3XtPq6Po&17j=Ux}of>r6Z!BXqd2Lg8DD<$}xYC`C?UDX_798GDS+
zVf7J>e8D;I2uI@A0WO+*HF#oQ27XXna=p_f_3T&a`}b{VG7BMepDaNaM($$#xhqPa
zNIi!gGD)GqY{o53=z5W9Mo)$*P{;{ehsqqAo_o0-r%lG@ttVR>p)irfS#Jga0pIaX
ziC~=(#J({(oIa&2+WAcXs9wSU6lY(W&=0Jl7g1ER&z3HYV!iexFF2<g&k_E{LHh@x
z5l7b+lO~%2`P{=Rk2K_7&{Exv6Lfcr*d(y@PELpEPW}8aJ6UZa$YDJXO;78R3b4!n
z(9T*Ie~L4XXUp8AQjckf+7ECdl|{Wkk_4lPcv-|J{czXgp`aYqo#uFu%@Y7m7J5MS
zGe&LhO3vzrW)Sp`_h<YZ@Z_IH%b+sQU%Tt?gbhFAyxYj$4zq|)V;43RM7`n|P@9<Y
z4{eG?pX;nm%HnABW)lolCzF~NZA6o5vU*(zBT-jh_uFBQrpz;msVqQdkn{vi?;D2a
z^Qvs}evB@s?qGFMBW&=44m9EPoetY|@3Yk7XSm-&#T^DS_U@Kb4I<0ev{Z%#&e@+3
zk?~ik_I5qBc2UzAC-8;M)hz<DLn>_qmTTK_rTSQwzEe2)&Pn#y4f^}TO9%s<a01H>
z?dm;r=U1tjc^V^;8$({r=CNn@Nx}RKd?!^=dre~;Oe%h_u#_k&P0vPkIiIgviS1pE
z7$`w<P^Z(^h~W4qtp}<(c;1S;_OKF`-5eBhc}1w|{xPlrO8RG<C*(b_O;&l%`O5T7
z+*&n^5G);SWOP6G<<xHrT#r~<XvIDi&4>M3hsm8Nq4LkbTa6|q8ttDn#Uy6%mU)nU
z4g(t<z!r*<qN*3hr<Rg<MWS|vJaZw>`xa@-LogaHOl-Ki0+meewtf&PS0~wt?^OOZ
z<4-r`gf)jN33$tfPaa}ZoZuRS5o=q#p39G~)W}eL6j<DXo{7f>#1(!=lrN+OrYY33
zXg@p&JMp`Xv+wjt+(UCO-Xy7CSH48{grODUGpoo1qgXn)zb*_x*U(WX8l*(nwn3`D
z|7sanxn$rM@{<S8Jg5%j<o8qFR9~W^TLiLV^=I<~Z;J^8&fijq=MGtor>2QOW2DT(
z`8mZ!vc|7_&=N#xJ;MtK#253Nyu&3y-&4pHZu|dqh5z;7QxTr@F95^IlgAX?yVUXf
zfX!3GG%2r+JT&09o5P>4ko=PXwuqmBH(dVOB7Utv#VJhG!#yZ*H(gHJQi=#UeC_r~
ziE?G+|KDKafB)WNU*;JE<?)p+{}XZ({I|DA2n*S^JkJElUHk7g>fdki+uvu_Vlu4f
z)ZTCZCkWwh-~9xRDjO^fM#uG`$G=_m|J!Hy|M%DT*cY9IRGy0q@sHqH*~(B^{(XHS
z!~)gHT*HUBf1g#!b~uMXuZbjLQjZR%8bXx!x<Eh4yQjW*-=6@;t+XhHrvP-QD?lF1
z1K>PF!e~?q4s_+gW=~FuNf?VoB3@8QXA!TNTxjL`d!MQ0sJDXf)?n|Og1h#Ot!rxg
z(7Lq(8J6<c4!j=Hpc($@)O&KCWA~o^j5Q*`NjjXpCe_@Y7<8M}pk|bkls_3*Oi!cQ
z<o7?QP|yDWXJrOpibZV==oB!Me~7KMo0G0(S*CWZ&9)%_PN{n4!kwSH_v_tAWSz!)
zpkfvWFzL({aP}>1GZ&oKR$1}iJMp{q!w>^I`qkGnuYNsn>oc^^C<|-%=1Fm77F)Y(
zfhv5{asA#NfIoQ>s99GaJ#91*DumOuJu`g(81_w&Uc|s~@hQw(G4P3gLZSsV5Zyu`
zQ$Pkl*Cm|#{m-w!8c?03X4i<&c$&Nx7%Oi82{c}#njeBO^(@fkLc#UBc_7G%kf=SG
zN(_ynh>Q*7nZN7={IR2oS3i+nE1+Mvu-JLa+=J>n5Kbb2CEYRp3RJw3Vf1nYYfL*n
zK&aB`<eoW4O4@h+3B+!C#)D96&P}ys;LLYoF@>kcoA*1bh-U=&r~PRkQ6P@6#`w7~
z%Kbe44G0hu-eV|q(3M+v!4)^Uwh{f_7wo9oAZ+JIecn}KP|JjTB#PQQzg5*>AU5_v
zi^=a$K_NMO*f`Q9H+lZr#PF~ogQUxw1v*<Lfv@MQPG}T@1TfB&kOPEaL|UMd!10&1
zMSGG`HPIkhZakv(-_EgEq8?3=c;5%@MFgU4UthZH0sVGYlhv-b5Ft$$e7%h2JY66Y
zJYF5ua^6|J0o_JW9wkN{;>S~aTH|`TM?KEDhDPOidPdOHe8QZ5Z8cV@HOB2X$#GpS
zmP3k4dt(@*=zM%7Tara@zy?rlzKe&LE`e>BF?jagge34c>4OzdvD*-zslW&D8gdC1
z25FoOYx*xM5LNOXfcuti2up>_6JQQKNW)`A)U(HdNaFBYKp<x!*HeQHy{PI00-~db
za4qp;iPvP~yT7<ga+3A=m86pt7E1qafo2jS+bCgU-Q8c?w2(2X4sQo>?yB2+mGu+9
z94xWsApMz9YtjMY)jq;0D97Ol$8pR@R!N*xf^uH=emx8%^$lk|TXfRz`x=B69K!P%
zDna2S=C;>v(Bk>b=6sigXB}3mqW^dBqB=!5iX@<J*JR&RI-}c8FB`3>rc^bKNQ8S9
zhf8wEpp1F@_?xo*ryMg6gW@72rg#>nOZf~Q6M(wVBK>gDhs$hpw2Fb%mp$J@34)Dz
zfW*?vTxeh2Y9ie2-#(Zs$+#`JS@Rkyj%c4#ABHvJu3HP5zgu~F`?m!H7t&^x;TyaE
zRpgN1{8$;$xWuY2>tVooGyhZeChp+{o>`Re<xbEb)!eSCH(8)kOn|n*3uvbfUx0{!
zMeTuaSC;KvSdL_pR}3K8fp1|x(F%g<sKs&9F)b`NgQwqpctV^~`zmSfWsCKE0=uiL
z;2uSU<<=vmHSNT+ZRBq<mG7t6XJc&%UYY@UdlQ!ExucK$0XN5V1J+D+2bW|7r}tZa
zJLk>hfN!HV>|a#<=c6JhjRASs>>8W}n*e2bOyYfmUAC0X1wLCIhiP}6hy;}3O-v;F
zhv2Tk{GQIOTVj`MA)e#WP<Mkm%`9EuYn&-|{1!apsnya4tppxl=~+?&!5S}PQn~Pg
zu6J-UY&32a2$$8{(D5`ugDoTk^wzGDCm*4LrwDQ8P^&~qA4U1<aX-Ltu+-w-bQ&~F
z2UDg!r$^6}QL~AW3S7cwB7NUgRXdvWwP5feKjq44;^JyxtGtB>&HHM(H|oECh)!j0
zxd*&M0`I-{P-*;K56a=d$6r(Yv?a&+R?8bZy^f_~N$YPq1-91{9QGtJuSpbsnhe_W
zW3NBJ{=n-$#76Pez36-0yUmW3dwhyfgBOaiR)k2`{hWO4UtXD)<iNsmLtv9?%IG+3
z+ZQ{uueez?xe}cKFl<90FCzi&J+%n7^lJ%^LwR<Oz`Lq=YXJy(uP4jlbW$7Bw-Lyx
zW7QVcKS?N}dSAD&^WH0#ZGDL(a{dCko>ygem~~1!;;HKk0U4-Y9b%8ZXLU&?8V^bA
zJ@U9Z&$jU%O}gV^)KTtSN3r9Q)#rL~Z5s%KtRpgP1^B!qhOd#hx+SiL_-S%65sHCL
zQ%2)FVfW3|=AI{YH#QqaSKTX5oWRSH5))QE%Nh>1S08ai`Z_1Opfw(3x7wUm`0}9B
zV+}J}rU!HWvOTL|ZdlEIC*!v@uZQ^7^Y7hpxxRbU(*ELr`C7Z_k=6lzb<#gzZgWUi
zu&4R<uVF&wZ44w&{Sah@@Ajz~zJmP8C-x8&wZ^cF=S_XqBX*}Mz`F~>At{p-Zxli$
zw$x&7TgCz;p{Sm#X(&R8Jqyf{YdlllgbyG(gi~0gkoFk3CbQ6x->S!GcqlSt1KyEm
zUhICT-TP!M+cafzc<cq$upHtZc-A3+a)7mBmH@b4mWM`3Zu?ntQxaARm+`L=G~%s5
zonGX#BF1xg!VNMeV>G|R*3b668wN8bG#C2Z!BVdk@bCE-AqZ9XBW&4bE8GB{=8+|y
za>xQ;(_sDnHkh|jA!FPk2Kr0iNe!k?w%}6>Qad7KVykJEACxmEGnl+5#TZn6?0~1R
z)3ORAqSF%UH1sTkp?PtU$NIf6x!d^O>?=L5{T=~so8=>xhJ6T;8P_fwyn%vbsHH%o
zKp$&GE|wGvOXefPnRev~yn>3$RY+IdLLiThR){V@H5>9nUb7_6V&-cCA2X+g(!mSw
z4G^g74~9rf^N<jCZzbB|3;5k(r3ie#Fmpu7?^x&%cl-_Kuy9;x*$%F)mVl-O8`sn>
zCj`fuft&xLlTX|CK*IzN?6aE`BK6q>(T~96i+-aDxWP9bu{P|3D`E{DyRs$jAr)#L
zmi{9n@eV(TSi1WJ4uot}lRm8RPauPtg=+6&Ikl01*QpukRg}6}b8Ux+JsZ|cQ(%)E
z+^z?4VVhVYs&}f|N3jIT6;IBGK?2*7xv$9%7{ok>&J<bS>+--RUE*vZz*vB7bFhK?
zwcd~*Dx2&=c)jYPwn|_%_s|70+Q%D^8g~sKWlj@)|7S9CT9|Hg!!Klv11jvn+k4;B
zy@QY2+vqqb9qi47^_{<6J*e?sBZJI~((aZXyZX)CH3qSE1VVS^5*FBXJB7amm=p;U
z@3}iUy3Af^=MP#3Q)3?#es_XWd2Aq^%%Q9n?XQz1JmG}y6C@{iOC<U9Nnk-G(9=o;
za@3!4OZHno-9Y1i0_KFc_Ej4s7z;^oN;YAlPy<x5^Dv9C!0<*YEjZuJ>hZ==VdpQi
zQO(LjNG(|f_~R}6nfUuqnbzTr+J1w9a!JKw@dCt#?&i~78;{^M81u4uG=XAahS-Y=
zXI0pA0Z}Emqh!Si^An(ApkI8Je7*r6JB!M$DQIi=do{#t4zSLq6rh;v#c^<qiox7L
z2V2Wun99>rg-T3xmztEh?n>%O?0722+<eSPfISL%PL-H_ZfU{GfAH}P9yaUbZJ+IA
zT@GOzbjuZ!%QF7SZ0RT)L$Z2O8b<6`s%PH2SdX%$VReJ8+oIoW^UySp?P%!yT{TN-
zdjZyV{x!eBJJ|Ev5bGG{f>QJ3X?hg>`ZAt>7^=as_l?y2?%ddd>wLmU<{1%*+>=KK
zJ7o^%XP%Xu+@xG<g)A`4RK*{}_7|+;duCx4R3n1&xdVuP&pZL9oF*Py3!<v!q;q!f
z<q^B-_yM9bE_$FC<UIXxm&gYAJFNYQ@yChRIQNxh&t1}Rhb6skb<geI%izMF!b5TM
z)i1ZxLdCdH;YL=Cws?jy)T@mL_;NO7@i&d7b93HkTsk1c%t%Zw0GEf`QPi4#Lb&n|
zQA($GF0JW5ogb6v@dNDt>;V6JK@_1#w7&zTVe3>~0Ot?H#gqrPhthih73g%m3t7iQ
z782`(YZpa+7$r1Rs+qTef#nDNc)zRjvtZP=?1C@${hkfX(S`*R{kPql{P9b>rd@rw
zi7$ks`iW$;CY07cyl-&8qgy*TUvp}b((0nj+`Sl8_W?J${>G=jU$-|=-E3_Co|F)Y
z3Ax*#8A)vRy^2+AJyiVYG7P!%&Mbz`s$x#1%O{X5hVZ`NK~LB%CeEtX5mx#V5{2`i
ziR~B1XAlz{KBC@>;ljSpi{p76n)PMEvP49hlI$0sg=;?T%T$t22qGwOAy$<%M4LGZ
zX9#vzL9xt3#NZf}fEyg|i;2ERAj|ly_<$of_49g?ypLUdyF3<JJ*kr2XQSOU^@68-
zvw%F70T$>Whyu9L>$Z#>%#Ovr!Dy1{zATy7uiT4+7B4;AGIDzaAtdrx3|QzU<9o~O
zcNiJ{t8|SMeIKJ2S2c|*?1(=$5?EdzJn$$|7B&#xQ7xWAs_#eruDO%W9IyK-@@Q_B
zK{Lol)Y#lfie9xRc{q~A)|s`=y06K%Gp<Aw!A&$s^j;9*v#NL`s7$FZA4$bP)S35|
zfnIj#bMFl`>;+ZFCDrofzE5qnz6amOnIVO@OUC8d&Pn^UzBj29OJ2M*5A#Kuo8(0|
zhL0RSs6PJI{Xsd@SF&_ZwXM1Y9fZ4{yS6y3VMnMMKF7iTba}tn@w-t5hWt*B!4ad{
zQQGqB|A)Qz4y3wo<Ht*66~`V4#|~N9+i~o9q>PL*ib7c-D=WvAL_%^<l!PR+QZ`vB
zGbx*lY%+e==Xvh?exC31-1q%_f4{$f|4EK<KA+F^zOL8x8cqt6%9Ewhb?TjU%7mh)
zRBLttXRfhVEo3HNdDc?R<Ue2cxGKgnJ?HDOP{$s&OMxoO3k?(J3+zsxn+n~1FJC~U
z5a~Un*KhxGm+Y%dPj1bk)8|@Q(Y9RWiFqxP+V;N(WSR|COTWHT_ydf-eCGf$F)BL$
zP*zm;lgpI#H)h8tB*DJ9+{MDqedp#E-~-bM<hPl*RMrrPj1)}`LpZ`EP-pY~PrIb&
zq>WI2BR%ewcv|bxw4*{_aWs{Ri`k`~#AfMu)+$LJo}~u`Zq9t)!#5@FgVyE3_cutK
z<c$+vK_EtvCu5OXbrevU;dz?RX=&t2zO!5`JoSra{NPW<@RV7s^8hMVb<-120OFVi
zc7LKb!k))EPe-bzzhaG6D7SGSyqQ&;6g%tQ;;)6?)c7@)bStQQjc38qggrcRT4mDh
zKCv2e`0H6i2Ell#Bj4Yfgx>(dRhR(NJ3a3KNi=KyoxDf7^k0(IN14V+x*y9|3}o?Z
zo0*+`Om(HsE~?3516O0)?<v(@cA9CZ{|1`u^u}9kh^<`T{ThGw;zP)+YL^~1@kCza
z%&1wbjg?3WtKiVOg%LK^=tnB>hOf`E&HDJky1Y^QIM;5C9jV{f?m~Ze@4)9RLkV`V
zzHxtx`*9uvbgT5Io%3%MAm5t^#piVIeJ|D7YmA|s*ZPYu7{2#n+)Y>XE8WE(HLI*h
zHZvWKiflcxbPU>rD8=uC$5N+uQOXdkQPGz2I6L8VB^lq@#P9FmFrJd!DX0|0NIv{#
zavUpmt)NG&noqemv9UzT8Gq4}1-td6(eypG3WN){th(%kKD8GTLY<UeLW<7&(u~C>
zAfaX4`tHpt(O*|!^xjT^Or5hL%$SZy=IEJY4QH<{knfbX&rh_a4rSDihEh_@#$0r|
zcUi{$d%aV6rvGC8t(@4}czwk~3%4z3>!YWG#eZHXehpyN@aL1VFVu_~c7~wEum<Yp
zao<ZiaK+RWx<?Se-fB8@;k?E6&C<I9RVO0fQrlS+#$9WvueoVI2|L<ZNyVZG=gsxr
z6^S|JXm6H|+pN-6)7KEAsuDl;sZq0zmwbyhY(I$C{cF-uYm+*86g#>DF>MVVpxY^T
zyHS*JZSmu^?nZ8bcgIzNYVQ7mfOr#h>#z8WUt*rl`5TM`<%Tw3enTu&|9lTnKvo!G
zvaPb3_Q$g&7=H8CAj8pw6SFRNpER(tnn1!7FCuNq7kD(SB=|7#AN;Er?3Zs8t>BPl
zW@?TG`$><wmy#^3?B?=53&Mg+bl&pOwT=)eC;|C^YUiT!z@=5kbeeBAmX}q!#Jsyp
zilHYedSo0%K1OdYO6wKFp7J^&+gbibV-jw{LKxwjuLeo5w5DM59gzLlnu_rY>I&w+
z20H*`M(@6sOwvbSVQn=vrRPS3H2LYas~*!<5sn13m4^CibX9IdQJ!2~fBEH4x`@M!
zWJ8<fBh!?L8!B(jBLqDF0gS1Vh?dunlJAO=$LIj4rJ=4iSYsBFb%a$i-|4Gf+K?a*
zJ;UoL`SBK5wVo~8vt#;LlOxE;$UqhyeIfb`L)ERV-1syC28&C5+4XWdN}e>jDn9yF
z3ZtUXeb7+y3aZi{O%r4lk-8!G(@MR<(vM+{C*{QTZ$RTq-P`>Y6CIKZB#VaVvvX1m
zkE5zmTKh}vpRUPG6tXSWgQN3ZzxEpG{Bv1OWS1+!ffO4NNU+nveXDww?%P)^YUft)
z(dLO?nm0$oeI%97586)bdrkijf}8d7U8RnCxR9)TV@0o8t1CwZKU6c39(zPMjcbx0
z03LxuT_{PNURwN%*BpN9-#to9-SY!wOfEc{f6V%*zuzwI-C&$w%pwy)WJIPTNl8yZ
zzJFE$_LK(s<X0RGvy)PPf5f{0y3^JellL1g`|vZl0|Ort$Vuo>T<4Z17I8KKxp*E(
zPDnIXqA7atkv-9Kx-@Fp>YvuseGe8Uvj~|LC4;2JP;xat|5z3|Mel^Zq<$&+4sa0-
z>0*hWeiXaC4Ax}Yz`zR|&`5Z!pT-5ly@(|rQj{&At?EiKF;5xWz1h49QbUHLQNpZ^
zrzK5`oU~nV8{8pc<ntz^DmW&A0+EofNQi*4>hvqX;j|+8*8J}Kl+lOS{#u_WSz-0G
zR{zB4O1)hSV1*7COk=G-+O=VJNNI$_%}`B`bil^T;N0lf>%1if?<KCJ)ex!R-=Rw{
z_vYVz7Q^h1FXpto)o+w}YJT^9)jdt$``#Q8*BI@3o=l2<c0Kx@O$eGU%Q5#(I6T)z
zKc3q_`$jI+_L|-ydIOh*-yj5UV68I&h}#_);jAPjpFjIJn<Ow>bY)JkzM%grWS^x_
z7};FNkl}M~SpT$E%Xd}9q|Ofg_t(ahc#q@2fdl*dwldpnzZ>}d$RzJSGd9?CRma&R
ziX!sxtOMQzy%L}sc}JIU$156-#q!G2JEOiaduY?fA8~ArJj@b(8kf;*8lG5^BtBQY
zzv1(F58S<;)ou)4ncw*GX}_|l&t~k!gQ<-pS8Ugx|DuWfV`Fsw4p%X-o&OP+G~$TZ
zHfK^W2+GQQ@j5-S5pKN+BgQ9ZnA-biGMc73vlj9AQ&&f7x~^yeU(B@7Tzs|0|2xP2
zX5_@npxqlA#x_}q5ckJxpxa2SZX>#&=Mv!;6UXYo$a%8R@@4Blg}eReT?Fbl1(^3^
zkBL@82sb!YBnJYO3?x0i14MF(?jju_fV4OC)8AhN-;wc*faMnV=<^$Uz>e#H0r$R{
zvbzPJxH3EFurEQFkq45p)Fr(I6u~%G<+0+#*SoN{wPbOFMzE$L!aBK=88LIr0AYf%
zaLn7W#~B{cpMqPHPlQZ;3461@R}Hu>$7A5`biji)9spdBkYKpsi(VkwPg#$}4{K54
zNP6CQmGJ70%Knzo)AyisrMqEJM2FkHZf>LSqd4r8Q`f`dkJ(KszkZy$1o;F7Z=|k#
z4!22PC67JTzKaNU#tX$8H-@gNq^}@8gg1hdgILy-wxQ<A1MVAl3hvV`zhi+u;Cp+v
zQ>dtdD<?yzUCQl|`_~3H<;{A9!>(alLWa@u%HL~Z?l|}X+$9faKuy?B<7bvqj{z=_
zPOztvhb?v{1j*M=eh!#)OSR3m@2yptt(K+I+elN(a-X9~=1=%LWc>g+y{0Nfb)3&1
zUzE%gl5a=EQwp4^_i_+ZmWd>6Tq@}Ut4={NRbc=VEjkq6iTJ|$MxVzQrW7L45Lr(^
zQ6_l%7`W8FWjAc26^k)wQWe;uJce1R0}M*ki>|RXbHUOf^#wrK4pyFkM5MV$fOI&6
zYZ!h2tzhQXsdy0XE+z<Vs*Q)_O9yUyO@T(P{pp)V$F%xQX0h`(TP|L(<9Sd54Cc9C
zTiRR^tFFaii9G<CUjlS^?~%?>Bt50tB<y$f6yh`O$b(Q1IB5{m36{=P_D!h#$F+G;
zo27X$`hO}gEl4^AXuL~mUZ#*y-SfoPnP+`F4YPX;fK~bIXctvt{vz-gjfxt+gmXqt
z0e~HLn`0HW8tJ+0JRd64vOCp}d@xgt+h%VFi1nEU&5Oe#>dnpAb?Xwlh!;@H&MAep
z-mkYB`z*6i-T#3!D{-hx8jPz}+2d=hQ$lT%gwbt;ufqTwf9?d4x3l-+(5vUCpPoHA
z1^2OZ9rah+WrK)#oXmdy;&;2i=TYaVUn%eK{MBOeAD_n?-3+x2xYZu=Tay@Vcp&L>
zr^>!dFp5s3YMRjd+J9J-&6g6%{FVa@_SN(myk^)=4Y^8+`g(|!H}KtWX*Rs`ywaP0
z!S4RPA6_5vEiz}nKvTr&=K}Vy9YADCJ$rcg7i^2<&8)re>L?@{8H)yE<i2cUl0%aZ
zpqrH@gD{O`x&kYR;9vt06oh!N6;*=9*5D?&pt|vETmquXJbMdvHOG8{{pt=C0+u85
z7#v(#2GqOcL7c8XeFIdFL>ecPkJD7cQ9gm3Tr7U0X(FN<h?7c){;uo%hxQ>E^xF)t
z9vzmgi`8G#nH*|?xNOPNWrTA5(Ans(R6bUUWp{6iR1kSPbk0m$h1o2IN%T4bo;~dY
zHs$$%C9A8^L$7Z4f!~q-PB<G;8k64*x<QsP4lc@E+h9qlGTwUK*ry=vd;E!Ht0M3)
z4Ho^HQ>9ltD_kHF&lA1(O*GYC<v{{jF#Z-kXfEsV@Z4Kwd*W{FxiY5<L|4ooEGTWp
z3T|e*E%d+TCvb-zB~aW*@Xw+tx7Pkj)%FxMb66N0u^lF~j-zO(alCDUXu%U)qiRd5
zF~!qn%>4L%yicMEx|G$S1D0m;_kPuQhW=Wp0P^$6ZC32F;vn+FZq*FS?19jExO7A`
z?;y%wf7-Dl_7uxZxvpeO(^G1p9J35QMeYL?o>V;Uiy2`n=ac5Fe@uoc!y@WX4C602
zz$+ptw_~|Bv)iHM=#&Hi@72W$Sfk2(*Q&;dJUK_y?_FBjZBbf?46CZxZQZSBKDzy2
za=zM4d7)-@B&Jhf@jH)smE+G~Mv1=7l~LutTe2joN)VT0q_?@L>3;t*)hJqH77PT!
zatfxKlN%|mnIfjZ*c^tuyXL8$3=GuR)Mxq%@yBZxytkCn#O-eZcVF``6`O!yLKFoP
zL4_(tbDl2l3IFz?S|=VJoS)$mTX0YF<(?cZ>TXB#`7Ocre%|IWAk@-j^l>ye0*umA
z>+iEaPoRyTPMBC%v~D#vm8IDgKPOq0cHY)eO6aMkOA^GQSejyJ;&fA5wQ_p#t@P#l
zu$#qe%D=d@Jy!!d!x#$vf?vfU5?sLy+OZL>TpsVp`SBE9MAx+ivUJ%9x1V7(a^6bx
zM~Lr-RRL2;ZMN^ZDC_Qr*6i=O`OJ0`#0lc|jm^N#J0W<ib!#ds>t{WzI&@&+hs|T?
zcDh9Pp8T@571wA=qLv1X@RxIN*lKm-FFE_=$MPWp(Vy3i!o`oyo%3CX`Nzdkl^yZb
zc*yP)eD#S*UbK6W<;y1<*BYg_FF{h<>83MWbPQow-k4tTeN9^7vG@c4VRgMuXk9rS
z`@-7i%cqJbJNCOe+#=B;y8;N1{E}hzp2OGcY56-2fzF_p@c<;V1>8G$;cP1l%JYaW
zw5l{gtU0=QOY>l>PN3nOr`xb+vkMrtDI0^`b1p1IoIRtX&z>diCAGJn(8b|&A+9F)
z6<Kzc;P+?>N$EGpHiX$vw2>{`&v3$4h+~^*l7HM9<eR;DPD06a-k+K+mqF4yFdX8i
z;i#gvPp#*a3SN9st*0)e<{sDbgYcAJy(NIgYr56^)byW(3UaCGr(*EAM{Y|l^gXYX
zUOjdVtrX$D?1>$W5>~jUP{%<&W-%dv=+;vdl2r0cN}a~FyYpiTwhk;xmDN&YhHjA#
z3D#d!+yOU-@~z7-H?32>_Z=+05H{fhphGmpb!qDCETeR312Lb6`%i@3{CIz|YiQm(
z-L2KVzqIgH0X4>Ee{6-Vg9Gx`2_C5*zU)QUXI>fJ`TjcQOTf}dA6bw`|AojX-586e
z1b?G`{z;MmQCywv<-Vfh*DdT8zvr6Is)!^)2^;hguF+fLkTW0XmH|b8-3s5f$m?U9
zcUt|VIkJQOPJ?5{GY(FbtCxo>N+xEM7^keQek@8wv`dXtxu^d<Iv__xzmsI&lJWbJ
z6Y?a|Mw{$*?q!}-qxzUPflc*3T608&7|AH%Yj7^(q2pW{JLw?5!%01(p30GYs#Eb}
zP%d@S6VA6Yv{gje<e`U%m9%A1!JK4j*JR;<oVb6bs0HOaj8XFn*?EY&h2<tC4;t1S
z;3T#J+XB8`RX6n>GB-loROPGtUSyU0W<A=v1YB#6$VY9hP>z|O!ZOaL0wHB#96xWr
z%?5a%FTSF0vu%_;VTGKfp5Xwe`9oLl6E<po#)p&7aO2W=3TU3|cTwD6$DW8A<U^9Q
zK02w8X;pBzMy#u5-i|z{CBxBtJJHc&O~@+YHHjfILi<+-UN)6Lc+DU0J+fZ<dVG~v
zTP9KO1@E{Iv{w|{{0}OD{dfPz-T4Q(=8Y^I=kY<&@)+aL)kAyC=bR2Tr-qSIGsa!J
z_kDiR40V}ue#k}{bzX}0ks|Di_!QN+G~V<~JAIasq`ix%y!f`NLPZuLeZ3*MiY%N#
zCi^I-@oZ_%b!~1nOQo8?!vry&&H&oh2$eXui&-Q7ko+J7PM7-r${o~*PtL{0nuE7l
zTx!Y3v~s8i1ICpWzQl8FhN!otoOD&Zu|hVu{>rgSDq?m1VZgOQ)XXFpB{9X-I9Xs+
z-`CJdHbp9(iW*B?@=813J?-(?&^K81VSNTE&2zwDd!E~>++o`74D*flq1%=z(m01{
z#|%O<@=OKs#2C3Md$@#wOK1|8n>&)zOt~WB;7iu;8+^en;zwVl=cRORv|J>4P=!i|
zRoJiFNxn?#=hf51UtI||NN3s;-FUZ`{(Llbyjqq#CI94z0c+Z6&KNtOCNCPY{TlkZ
zmm+@Y8>{xCdwoAUTKQ*aX9;@f$JVS<))_o)Wtd%`$BC7E0vzmrFMa*5UPt!;u7`8a
zx1sDIDdxKxJ-fEt04>4yLzapy*v7hAyu5t3MH@wr>ZZiR9Qz<5Ox>&mol#ATBuW^i
zL*9Lr^9pAuqp~Nj_5Dt+n`$>g9EfjI_E43(AEEh?h+?7aiA+Z*3~CNlR^;9xiqi%H
zGP2)E{bbY(5$a11ff3!qkb6{S3%OZ;MQMi$@Dod64`ZrTEkko7OS#{1@T6Fg6PhyQ
zJoV+fc^RLHxcOt+mp!kosJ{tCO)yjqd+r&np1ek(>!e5aV1O3OwQ8``akuLgiiK*I
z0z*xpp48PTsrURIOHHQXL>VhYra77*9wOj19gKIISzh0GDjTSPr*0+sM()Bn#zn#2
zC=+fJYZi=AdAE5}b&4;zdW?0IWzi~J`AJ(+6{s5PD=@iASP$%~w%9TsK6K<~lqc%(
zPWM~XN&5M50~GG9{JChZO*BWW0XDJy?v5+V^+0r>g240qo}=6BxtFxF&Qq)X(q`~%
z<5pWx@6}lKX%2LvTh+o^M+i(LG|rynCaCVs+3uHa9OC+5_dwkgqdkGzLKeLQ!G#cq
zNT2BRq@aYVEjwY~DXAaaRovk=%>@G{8I^~(8UvZ^kJuT$Wqz+9P_&fxeAFz~$K5I=
zalHGDm`Ke$c?ea8SW`r8f_M|=VF$lj(nj>B3jU7j3TENjHN1SE@8ZPYQaut?ONj+G
zdE4$({xR<SlL8Ztdb|9_oQRF43&rqMn2tCTY=Wd;dI>s6kO}8K;EY38-Am_`;G%tS
zY^GO<eVv?ziiQe(gUe(b$mGn|Pb4()+|2ukZbw^+q|AwYA{IZ=?JG`&kT+w-nO+qZ
zjYc}fm~*joJBYvb*v$#4=$zPXp}{Nk@tDz0{-JICQ-pAoU$w5&DE!xdpokjiQ4x~Z
z<|}v}vlMcgQ|RJ~wtL(+1|X}bt(1&Ji4tc}azv~(K3qF8@kKHH4P|GT+W4d~RTyH;
z;!C<I2MEPzqPe#rha6!tEmsud`j~PVV~ZiiYCkt;S8!WZ?)8LW*#l?<=2#8vK9;8j
zC}o3Ot0{t{DMFd5!<BqWG7JCl7XAn2lSc=}ktE49rr+DWe{wQa&l17FUwl683n<|n
zYDH%5K=M+tqwJS}64S|-eKvCoDL?pAq;KKzWC3pSzz4{MnEC~`&z&IA(gx(<<c+TY
z)xEHROGtj6<%%$QjDWB9W<BjQ$c-;|O#HPvbH~T4Vc1Bhg8{IB^drZl^&#ErvL2rQ
zC%v#|%yh;`cu?OXgJX+mw^Xr^3Qq%avp9{vD>fkp4-SyLF?daG@n6#y8Q;xmkLNYZ
z(NSl?`I_=IUTcM~7^YL_zM{;VLOsBV5n&KCO%y!yL>mA%gpoZ_TITn0JmNnt=70Zl
z3-xd`m73Q^{l7rt;s!dP_~j_V5$4>@IS0)pLS+OZ!PTDuWPfmn|LOxyKcoq`)XMfR
z&gNfz>7Ra{Q~Dr9PzjC1-?PU5>y7$@-m8tgm*Z&?T7Qd@{q^hd25?ikgt!0cFZX}(
z$s~B1MExGM-~QWH>94<$9o@rGreo++fBZrJ?>-0_CpTWwdvL<@?<e`c{H7uOP`w&7
zn~(h7!~KB$SDWZEhz#{6TJPqW<mo!v-u&<*3c2->ksUEjbiF8L?+q`|jW{qhZ{kdz
z6869Uc<GaHpGZGjw{!mOMtMsHD@|==Rzv7*Wza)#3dWB$gO1Swa`2Xcb98_BaLY5h
zhG@Q{pNp)k9|p8-^Cj;?)zDR9Oah>7%>lSK2FEFf%wJrLE3~dYd-l1_NkAb|Eni-4
z0ZF6)@&r}$db4fpfx@#O&;__@H<Qk@(k!@85d+wFDkNj_5j;u#%$^%EeVxGZ(Rsc3
zowCkq25^!Uo=MlMas(I8*U^}mzNEikfB&mMsXEg<95u6?uj~1p0BJ)EOk`d}O;5wE
z*$kk-a#fQ*P59lrLT0US6!0P3$bgj(GDOa#!2nv_23fcPERneoAaA3*f5?|AmnHa#
zadv3$bK9@Mvd?#+Ts2GeS=pPgk1LZ!kA4K*Y2ijNXs9JX5TADHQoL%A8VjqmivSRL
z%MrFIf?BTy6*tvtqj+d=?f}91MMYSg8X-2q6oVA};`d!gIxJ0fb6)c@n3LayF)MHT
zI`|kk>}>ulnX{Z_5w@-}MHs<=xaKfOEyyM)Ty6T+ur<3s{cU#zoCRASaTPZ?dyeS+
zS55hkgQpm&KCiaQc(44^+M<|*hFRfyA$bsj`2=b{lP8!!a{C#iu^IrapChbuQo-e{
z4y%$8c0e3d5Cu@OdkbIKBzTD0)W~`tC4{QAs+ZM9(edlTRX#nn{G|exD!D)i>v?gh
zvv(I<S?+=_tf*uRpvGrm-R0;3o2KIvkcN5(il_~^0EysBy#a94g+L?yZOS$WFh@91
zw+M`*WC+k~;o4>}i7dXv<Id(mJ4i=E5p$bU3G}<ozEZ%OW=`n0TNXZ(f(V1j%aFy&
zOOkc;fvXGTx0QgCcxjB(J5HfvtYgUC_p0E^nGE?!Ptn{T5a__PI?~P`{3kW|Ki)4@
zq6SXYy64QMY5c#Pb@XWm2%o=zgJMs$&(#^QpB#UTzE8&^-7JLY4|Re@^x=Jzi3y)l
zN~vE5>XdNn@nsATuXcE11Q=@3sbwOzpv?*HBq~_ab{Rp$4_+9N@PggIFc&9A0*hHK
z19SUkAX85Occ~N~v_(ISpz;~OI_2fvMC)>68{wEW-UG9*HW=T}BugW}q@>%hIlJFD
z%?ub_C|CFzkcF5yMu|t4XNq0GXt%tIeed4nH|(cyqSbxwwy9hzHwS(c0R=eax&Hud
z{`%Q+5rpF8DtNNN<F^M;HI#Vp)zzyK$LgOi>%xUjfxUwv7zQCuG0EHsiYAa1m|cil
z32kLPv-IAi3S0%~vDD%9k<%d~;U3@=Xed&<YQh3h%p0cD*9*u`!G4!cSS<i1@>Zpd
zVf*er;1Pa0a{E*nhQ=^psgAT1DD1{dF>i%`^q4s6)Gh8?7WOv!nX9yV2Y41}72P<T
z8HsuW99D`TwQIhy|Emf<bbw+lH1Oy`-EY;FGxGqUM9px8Pq_O2V$W0jHPV8kGo8VE
zFHW{lZnb5hNS?!XWAA;`0(aFg7{xi$?2JG}M@-NL*awl7BUyvmoza4u=o=pq+lS>5
zP#!(_SqL#6*1+(;PzVTmX6%X@sjkMoI+Gf6J7l@Rggt>x#tN$(v7qtRm8jfxhAQ7v
z?s`s50a41bB=iH}F&D5AmZn<LV#$-a^+%6Mn{#Lpx4=%vTm>6kh|9UPv0Ua_#>6H1
zosPVlxui@|+5O1aVL$t7-MfO<-2n78Y<vHuZgU%^M^~M*$p@GDQ>;LM{>b!+sm^OS
zU3xvQ9ogEVn;nfm(@WX;&KC;VStq3=hnbVq+lOZ3rTYO5Dij`B*>al_+6-aYybut^
zl(OI0(<gAKHI$f9_CaUO8L~FGnwBaabGN2QG}f_0Hx*i=!}mgiEWt!&<8y2$rm=E}
zsU~eg;c8fB+tGGo&14oUvD6&vr)QpC^z<{&T37D-B9hC#P#Y1zC#QLvi5X<oHeI)z
zR8kKOvJCAJpLHU&iqU=wgE1X`Snb4*-Kluo@YuTR;_hBhq#UK|n31~${ZpgOwwA#E
z)p&&TaKW=&HW-%oo25gDF_C%F20VfKnUVS*L4aM8s&O33VGUHQ>-z{>lmIqPnaQo4
zU*4J&p~|DvIL{%j{T`2WgAfxkNB;!SU7VBf1@nd5?8nb6`GaX@9`Mwb2d_NW0cI~z
zRs=p22!Pd5I$QWtjwjnUDUWuO*nQO#t>pn*%v#u@>B90tr4mSg6Tl_guzk?HmygRC
zN03eYN%IFxG|Ye{C*R)S*jpk!cbLUQ-xxChY3V|uHA|rypD+X6D%S$xly+?yhm>eu
z)1CU4C@J(Q%!3eODfjQ#JHaD?8`cQ!QTPl3+DV-U%L-0@7>1)gR{M5h7i2{9Z{)8=
z|3d77V)QFuw~>zW2QqFOn5_U|llWHNGn$Z|bmWYti$q1y($J%;)e6;>ig&yssgCDD
zoH+}@+C;e4F{DrymO6Kwe(r#C>>1-!jLGZJudo|(9k`-c4Muc5Pt8Qf_CY0RZX0s`
z%yS#18L;Att2^#9EM6uXQJMj?7054X+9nz-fb&qosf%-tX(aN3oP+S`_`S%WLl&N3
zcKBI&WUgi?B6dY~WgS`E`GJ$6Bglk4Z*)a=tI*8Qj1lhuvsF|MoE(G&ou%c(-IQuJ
zw7wW#j<t*Bo}C4{!_~HY`;PbiD9j@KP23QjB}0{XwRQ%+Tf#}_gzJsWO%;fixi&So
z{QXGq!Y?}464)KKj@E^p-Z+0~KD6exF`w`n`)W{J8^JvA9Yx3fzdDb<D4)r~&{y*g
zNeVOnc7o*6-K;0ZH5#B)L~52W&`iYo3(WBoDk}As5yRhbCgC%4M!~ytws}5_BS_6r
zzlCs?4QxUjNS*Cq!}AiamDu~vq4*&N%do>atQru;M$5;H-ahUiLZ+9P<B<nMJuAOS
zPj(2Xqx#N28{z1SHpFa(++?KxAea@fQuzqtiI<@SoeLLNWb)t`mkP%42njR9l3iCL
zpT7p4m6qd@!wK49_}s&OsBT6T;5*kP^&oPHMvvBJYR73<Xhr!X&R*jf=hwC|vvQAE
zTY+J@f4NB<`^ggYIT{n5ZD9h==7OyD4v^z==Ma^wDkmSoh56L^SZPF@J0FDiK&^9z
zR?R?vn+kO8J|2#;gWGG6bzz6wH%<V$5G7WwIWnIci<RMR==ngM#XL4jF+fIsn+=V-
zk#&WCdD94PF`!pFdD3nYj+S_xYCU*?`Q3@_Iq0Yyx(OP58-&io8kBMG5k`*(7+H|F
z5}lwMV@i4)EUdbX@mIfb>qn$AQ8(>Icpj$FFT?Mc$pCC23aK%Q{@~}QpBR25u9DIl
zfSpHcyU@vf9xpF0^-o|jp<dj@y}HAkgMYUHD{M;H^GEChc7G-Wm$@IMTAAv6RdRS_
zD#J5A>ZO|;VuFd}-WTp?*P<71>HN5`sIsveH4*r81go<C2*__H5M8FB5biDwWFESN
z+P?D&A~AMr<iT>4M*pjqGqxj^Ogj9>>HU^Z`#gwau1QLqQR7vaei+6UciyzC*8Qr<
zw`QDL1MXW>H(s~k2V{`V=Ld}E@7$43>Aa;W`HnR`UL^Zo=R~~bjEAJ9+2<=9{y)Uq
zPB&B*_Eqv<t(;)2w&8ZEjEvqUcVmrc?G^}+*HrT|edHneBsr(_agLrW<J!PpySIho
z(`wD1D_O1nwi7&B4K=ZIQg*3p?q$zgpRr8slA2nTmLgDbyk>`Iyz;h_)<jjWqu_+B
z;B$FaZ!zy9#{{Z%)eT`$z;-vHr=!4S(uqf2D?+gCTwYz-N`LnRCKE9r?{+jjez#s?
zxetFmo+3J4H|(>aAl1e3kV87fUZ>8ow0BS6IeT&GWBx^|MjJHw#My6Bi|3@i6O_#I
zjI?sd9(^G_d(U6&^O*<w3Xz#S=Alx{X5UWTeM4ba{sy*@z2YwCm5p1va5F;lzFmg2
zV~Vpi&$tLBy%Pl2!W@+wcPN5B_jwO^FMcZ=(&^=%J5-t>>=k=C*+`3fO^)?MFD^uK
zjJztNDl90n9J_1E_o3o2sp+UEco?z!ie!aBW^I;>P0RIXt+^e5<-D-Ao1$>U#VX%^
zQC#!aQ|a&1pcT1xrE(_qA5>0(d#a|m`d<&1>32BAXH1gfHMy1<TP!L)@=m|J{7IXO
zMglLmi5QM{J9#*Ft}m+I521ev{I30ix4Wf`7df}4X>aqnI)^Jx5n@C<BrjW^ABYYV
z(>Za}F(Ujh+A!7dB|*Ihb8b4RMegZ$ic^ULoD&n=3g7)$$=kbmmDPDg@wZqHcDB;+
zY^sqxGtR~s7EIahZ>Ju@T>&a$h__A>FIFzjW73IUO^bU;j#d2BQrZ@qFm_2@N$R`h
z`SrvS&e03Qo2O}oQ%|0|IR6vo!pEge=6h|nd*7XUKT3Afl$L0k>RYF}a!0wddyiaq
z6u~>PI!hG$vOfE$239USv$q#t?DR26Xrkwwdlj`(`^z`NRg8Q%zAqrrr4mIVKYy?r
z{u0)x6KxHMXr<3F%2oSL^?%}3|6Q<ETae+=!{8m!ZvdET5HTT|Ys9MU^inkt_JvFD
ziFi9<7`QG~23ILQdI4g51?XH?7-i<Jo7Pd&RHMFpIHuIKe*mizW_tDs&5l>gIjvNc
zm-(J}ClbFI5dE#0<>;^9N$lwwsiJ?L({{-+R~dhyYB+BNoMYq?>$(`Leq8R@u2@bK
zsr$iccTQO+wOw*RKQ*r7#uhAT=ma-slA1q1eoI7fELX7}>2rp6`~1FMXx=N@(QRlT
z6rdNs&Uu3IJ_XGVr$Yp9tdjO?HC#v1d2(mE74Gejbg_l*1eEnr331wOGJ?o#Qqh8=
zg~}b&7@=dRs9g%YMY>9@q4RD$wMz6gif8s)o5bQ5Ort>fMQl!ZZ-%SOw_(UiDa@p4
z5Er;h7-e#NjmeDe<_|Uh`TObim*=7C462_(qno4p-f#ExO=4+rqAG`cr(}t4M+@%M
z8rTb!LU?Ek&^KAPMr3829HenSwY;KCa2DL$Pm%4Wjqf*-Zr;-xsd^t|IlM-Tsuz-!
z1Gn)a-9pfWQdRNH_XEr*#2R5BIe2Hfw9t8K`1sw!h}x0Xy6B8o8gAp-^kZEi58wpF
zz-w3|#8#_#jD&4pTS-z76hY>c@=RBkpl51Xee07@%T%>rAWR;F(hIrLo{J4r;C_BW
zP_zXIv;56^Io?sNwglRH`LVn(jV6Ncb`(Vr;#au%jAaO%=7BB8$66~==DI6$;Ru2O
zKmyg#ZpWP6$ESY@TmsCba53Nbu6%qPL@kgHj=`lfQeZF3(K9}md8U~@&Lj1@j#Y|t
z8gE4o%>h?M!+T=r=T^Xy6BIgIh_ViqS>1*fzh&v$`w75^jH>xgq<|4NiXn0W1`kGt
zEte6sV|K%~o*$&Zo;MY*xlP9);s{(Mf}Zf>yWCFv|9qy}HbI0*F-xB;_BbR@v-XI5
zrn_*|4F`w&Qz4kfe2P8~XntD|6DPr&mjY>-6(N^^(36p`df(j@{Qs~M9=H?oV^+$f
z5UqGA!@F{^tL^=E`Ti}8IcI|1DtYig#s5CSEMcy^d)B{L@;96`Q6?mxfR`CtrrnVg
zOn0i6bB1N;G;QVxS6(@K|2V8*`8Z}v9~X>WIEEup!8;@>a{5Ywqo3G0>qav~a>aV^
zr20FGs%*sw7O{wF$S%Y{XiogJN391iAnwPWt@N09_?_>DaD4J%8lzNw2J5P`UTIb!
zma3hW$ROZ4^jm?@@*K;(X>pETZ|NiN1p_1;v=5tFcGNLPFDP2X=a|6e%<^l#vP{I!
zXyccW6TM~ufOCW2WQ9LU==weuCV7)<j#*%ZwjDZ=#%)x69iRk=IUzQut_A7*oetq^
zSZNG^H=T0F@*l^H^^r88s2k(q3?6D|lXQ`nF3S?cDyt0vJ&?A=EZ}Og)E3G9{0|?x
zzrqj&<cRX<^b>ji_(D8}bJiqbq^8Y2%0{@Y;PyIOT<FEN61#zhhi{OctGoGPIepHn
z_6tL})ag%nn{u8j{j~AsBd2pD<yhrJe%xkhYV&N-qd)8z1S`}G$j5_2u2~NDVr;^v
z3Qg|k&^(i9O&<@DZEJl}Veje=qILz=U9n<wpoit0tYmSFvSA=Sv_c`Dk7IQmID5B_
zEsmYpE9&Iqr?`<qHr8G7G1j=f$@45j#qtirC=X7mwnSl9=aGuQhl#v50BAaXmPa8z
z&m$~l+8{bl)BVLdLLbTil36=$;FBL<Q<(rxj^Nz;Q^GXgy!Wl!Y~kk>@5t#GTJR`j
ztrIQ7v);aT=Nv(%mARopHGSd~X*gzkG0IM~J7&Ym{OfwEfrj2(4(|We?V4K_{r-1k
z@c{X_9_JH*?;l<%602Az^{S<zmmKxL=j2I_9-BPtU6CQj0#6*t8^BoxTMv>OrXDez
z7;~F8dAh!{@TIys(j$luy*D19&CaAM7DoI6rZjmtA!>8<^5j_y+PeiV`VCw6^P<LA
zZHr~8!E1E<$l)W*=Or%PV<!DZH6OCGa@(JR<|5~~lY^L6YN51yqv(6JJ(#;Aa`Y-K
z<wA*7;<$hg40w<y#oa0^mwYnj$UF|^)}z_3U_KqM>F4F;XSvVox0sqYno6%{T{_fQ
zB4Rt|<fW~ZlVTZI$4vASqB3U6oVwoup~BBo+b%;Z^HmJmlNXmzlf5M(>=0|m-sSk#
zHDmH)g4FWm{6fkQ(;O+NO#33+ZRkigTHiyHRLyu+w1lzE+E-ID%5i9SzKI^&_@5gJ
z-5A&<^jxM+x%C^8Lqx91b~AQzH%x%gR`Tfhdp|NgHFCS#{wm>*?ik9Sc)7^F(Pp5w
zMelK>6Y*iyxRMSJSJ>{u@Tq<1(c?ujuPh&}e_ot_d0`jo#xSWHo@@=+P^^b!M78mV
z?mI{C@qOkg_Rckn2k-Qa2T&~TFmUoUaAH`n@@4MgNg_E{JcbZd2VImc8U(i<V+l=Z
zno9O@1ly%?MtQI<oU8FI$<dQfH}`peREGHT!QJ=sZxRon4-t-kv0u2<U!j`Dsm4{x
zQ{RUe4iL03Jljxy*{rMjxa2Vum@IPI9aIQq5B0KYy=?NAEMqFHDRv+DtiN_2yQ)y>
z1>b_)hLVrP7u3fNewz-Z6d}tzwzILlvBI$PmU->RaOI1Sz)l}p_w;$Lmi*#rQI#O~
z4I2|-TqKf@+|3&tR|kf~CZBqE-yXN77$klRSa@!{zs#^l)h=mPGk1P)#D{eEDc9ls
z#$#9ePeC+rx2hj~pB3~1jCItG?fiGw0`Ar`f6Dum31j^^mBxzIH|k4AVinp?RUNtJ
zL~QQUVNrX-`>;`szgLuR+o4b=WB9t%(*Wv9USM4L1Olt8Tqn8A2nob%uH#Ip8YSCU
zV{5<vjnsLOxkM1A6Hx!swejkzYS7`FwN0<<31w?uqGLFN<^G;`{3@IjlHfWuf3Foq
z(ECWCd>a|BaV5H@ZaUS54bnI@wI^qv-M_Bb5wCeViG-4|890yp;FWa;MsR$q65`IM
z2TE}hi!mYc*@Do7>E%j-UPrV@jG*AJD9ynnB<92WnU6YmP?X57kf4J$FjH$EV_tpr
zUsRgu$`1Qk0Sx2V)MzFGZJblR=oZAm@Kgo3Oz-Z|u}0~b9x0R7P0HMt!)-vg!SQZw
zJOM}J2#y+rQrlkP<tm#A>F(w;>%y=0^H;f?{22zZsg&ie4C_71h_~f9+{LfTyCzH=
zfn3|5qD*sl4x6W-HgxvsSp%EkUGKRmz^&--mSZ)lOqE$b=$dYp1S*ocX5;3tIc0f8
zb&Wth6=5Y2p1<dF36z+hf!kzEdUvpp%~!gSq{F_ZfRjhi)bHxc`QHd@g+>?jpQv7O
zRDE;ir>>+&#>7t``$uUezmRTe#!}O__GB?NviMdOK=(o*%Fh2W#oqdhZoa4BG(WEH
z<eU=gjImg`SBp6^k3hIM&NP<ZdO_YhD>@P}TRxXQ=TY#v2og(@q8Wr973oCt4kGG}
zSJ%6=6HABOss~OXY6P`lb&Nma+*Cflo6vf189HQ(V6TM8HBsw5GgB&qIZwD>87f_r
zO)q?Cee$lzX;pSE-ItwqQaJVw5Q4N!AEp4~Pu@PD!f_)g<sY|iL(NTT^*3%GAH^ME
zlkFO$Tcj<1XBAd%`t2MnwuA(KwSx8A?6Y!LdpEGfh(ws`>mzSvwJ_|{)v_w!&hIaC
znchrcYM0tP%d<o7EaTma^R?mb5k9H9&FfS*V27l^O>%gBURF4IfWO9j{O|b+Fn*}c
zAoIoz*LSqP&l{{E0Q2HChw0)BldmhfJcj)V-)G-W^(WohPJH6N3BO*U_x01IKM+&w
zS7J60BoJst^kkwVlAsx;6CSo_Rchg3vBvZ%Ib>D;D2<t>&EX??l)1TRhz{j&i&*{q
z>Z)zGlMml(wTIyc$m?cdJ%=L0Z9RgSaQS4a|M77qvdjiu)(JzzA%%6fb+7<DAP6+i
z^sB-v&b)H7o&h%yr`4xrT&jaC5Y)aVmrPT7G?#bvptg77I1>A{%hAWB_M}e9Ewuhn
zC7&74;4)Qzw5C7;&Bag@@o_R6lq33=Ggf{88j!dP_eLWF-1lK8=Kx0B>&({ceZ-t`
zp!>vRcD<1tyl^XgF6F6l%EImff8!4LVP0OjrO8fjT|Pj3ZX+PMGl1j`Fb(jUmU?Z2
zqKCoYSC{is)jR-*212PKCo-?w(h5tN5D?gb&`mr6fS1ukLJ$R-I?#Z7JZgWS0%Z@u
z-`xcskKCyG4}1mIVdFO5Uw)w`PC4u~Q9+wQl~<)w6GH`BYOd#f=%vaQ73JZvk<QWw
z&y!elAgY&TqUjgPQ!>}Fo$36+pEUM}VQZO>USfN+v<g6w#2(@fHtlB;iM?4o>D<LN
zkt6($Y4_qTdFcY=eA_KjD<0dhE4pbMyUq$AAmVHN*-b4_fhxvsojh3S{{jH|&+Zwe
zeE^pM?gTk+J@>*;w}<nwBc3P=PH_i46c@N06dR|B4Al;Fpl>`h!_>V5S`f_!O3gV>
zb8neXn;t*PltRI<llT?30K%omxo)X7-qiL3Ks()valPi@7@*h0ce85})b3+?8{ltw
zJFeU;UWF~x=s08+j8(eoM>9Y}bGVz`DH6czp+PnrJ316iND;zwg>r<e$$ay}$EOO-
zBo#W}9`~@OP9s~6PcME0yVkU8WO7o%YMdBr&U|zz7uTj*^Sc$C=vB$;V#V*ptCNUp
zPI~_>0es>Elw5IQ^7GLRAgNo{)^@%<^`+TQb`SQSmXbKLr7&NL0TF=)NAej;`(vFO
zeaAg%2kF0i8i)R%Ea{*3QR}N{;v6~dUQe^&HhZe0=BP5ttQI(!+z-a$E=J?8q1uPX
zIXD*F7cW_Kn&~9qYPcsmv##Fe6u&eYI2AA6oZ%L(el@dp=zL5`$ht;W!{%Enk^$;P
z)_{w8iQNEP3F@T?i=SY_U93A^$~%&sIN1Fvbo)9RW5W-+f!Q|C|0aU`$7d!Gj{nxH
zgL;3gAW>mAg*GABvzJR=EhvQkp58M4r*<_1F2`!?upT**+|2t5dv#cTd2gTp0Pcv1
zAQ!%}t$J^N;DbM9C`U?(cPihl!YJ{^vKiE9z{{C320h*Xg(<2M1pzx?2m7zltyqWV
z9^z~d^{%{-i_00^%7En&*0F4ty#B*HrZ@Z&Cj~u$nhRqG>))6y^GWMQVpPjfkKW%2
zNEt3Azs|CliBFB80#P(A2tHrUi0c85>a-PQ8)JE5S=+mY=Zbm$<K6g^f1>U}Q<z`x
ztMogHqpCw=UU4mZ7%VZI9cRD3c>^;R-@JK`0)9{iq%37F@DO$ow5as`Ieln(Z|2FA
z>^!ax5LPaclpfp$qvN}P9W=v=mNszmQR2YOgNMeU{Wf4=&?6u-UqPT-vhu+Wv%IgG
z!sy?B<bOj(zkeniDmjzpt)_qPZOa9Sz_&T|`urd{VI92-t3i`&3<*q&(W^fOP9qyo
zii~rn3FjUE{Y8JEBmeoq!yRO4xi5X5{NMja_vTxednX_#MJML0*_)t1P<U0sGPr_R
zx0&lN!I<BF>wkWMi!-v`bw1g#{lEDY|K}@zdp#}kdfE^5nE#Kj|NnpfU%vk?jKm;w
zV%h>dK_2X&P2leNKrk#M{$F>bsywWnW)vYs@wZ1QG2$%@IwC5&tEUGUfbyRVu+8J+
zF{LXoL$|_?@(S3w{Kjg7uB=sh$c0<#a<7;(S~sYs+=Z+Vto8?w7(IoXxA&>pa{=lR
z;G#VIibS3`3O+}cwn*HbX=eH0Wud2IV11G<&sZL!TnlbzzYe9t#_^W*{Z02~Wv63J
z?k0Xs$8!o|D@qz@&8x3Jg9J)MN|8ktCOal{n<&|y&o?sYX9JV7J$-|7*TazW52Efv
zE-Md2(Ip<DRy4K*WMFOGyzjZi3r#pQBg+&io@rU~4gZ0A{1rf=%?%W>qoh86JU^v0
z6{|gZj?jA)=q<eu0?q3LOjg?Pu7vyog{0`mM+SEgt_7zjY{m_qm_2usI0L1A9N_AN
zB;MyIxxw%y_%(<g^n-SOYB>*-<_+JNd#2&Hygbo10YR!M?qiLJ2cpD-gW2=&K6fNK
zUPISQuRg}~Kt$U(km^f*p}(lg9|y3g3@ow46jrVAPf*#wr$r;I3TS0oa9YP5H%LZI
z!^4=h5FMkc1U6bhGsd{SrZk*5;x&U7d&ivGxn_;TGve%=hD=G|>>&sSGsz&+!a+V^
zB&Jlmntygvx7v^Hs|7^sesI2bDpBk#O$cZ|{gM(?e>_*7%tP$^@HE^G04MK#)cDeL
zY6@(5JMrfd2#jUTX`ZhICC*>`do+)eey9Img=v}ZK+716%ZAn}Fp|wcZHB*>&s(^)
z5$y8j_vF22&68gsCgA@d%VZ!<t-7Fpn1Bs&>_6XFBxjS;qZgL&i20s)+Q}I$b!On(
zM5-Q!kn2>Ui7Ldtsi<mJa8AZod7GL`4bwW;lNaL!c>F`V>>$VsxcxxMA7Le(OaLSE
zR`^Lw_c(Hun~emfpv7^3aOpb<%G=u8YjgZ`eEMWHkC{ou-5TSMr$;|w@QG#2iQO6}
zcP3jFt>=B@ODq<TM12O9fG&o_@m3}4*ab*!l^&splAH>>*86j!*5~wB&2R&<HkA2o
z#QZj0hI6dS(;!*iuEsv&JiGARPa?Um3_IUHbL(d6&0E5;@|jDR`tKfMmms952D!Gf
z{yrO|(Y!4!0O(jydnv3P@t<yky}b|;(;8^!XCs5IqFEn{=5Y(!5vB@(Gsl8h5c^b5
zz7`DsJl-B8Z3eYHqaf-vEcy9fR+P3w6yiz3R~Wml>e&F@E`~(oq$f!(*yQs)7x17~
zT)Igv4O`~~WB^<D8NaQi3&Td6>C(z8wG8pu&Zoi~=(Df)w&y~fF6sXyQ&o0_@f%^T
zhUVu-@WhH@pJRF60uC;;UiuN&WQKMeF<sbdQrSb}@s|0cqzAVlz9kVdGSgAkHf!<c
z5|#yhXgenY?h<QEs{LNE{%C~${e#3f=mK24eB*!T_;3n^$S!vQ8&ME39?`ft*OjgQ
zd1}dFhE^bDG*UYtQ42Dc)Xu%D1m2k(dh`Lbi#DI1o#RKG7k7X|#Y7ei-sH`G&-X*~
ziD(Ij5#6-21hQMWj5%YWxwO*9-NC#zbtdH(ez@Wv3?p*YeQ<+t0D~($?`AeLGaV8Q
zWCbk)SbvV1^Tx=(wo@vszdB-7U-2Q2J;IBr!z{vhOzBTrnt%JlaAZJl`WM4dRgP{G
z+FW7z0Yv7B`&d45;|d~d?R|1auQ)E>$DCDCSPg^0mafh94eiE{nR~ss{OL*By)rQJ
z5C6jItN2856+~psP}qfhAvoG|qU@vLi)I)2EzK<V1V~fnPhJ$fnsWXo-e6PROSt%F
zIq9rziCu#+abo|q=jl-DL^jnCm6W@tirSlekotZOI><OpS3y05&oHm_aBQOvq{LMx
zY8m_0gmMteBvWp2E`NuzFNDvH++&bJ>zc!Zm<PinpWnns8JZ8&fzH5Y4E4AmXD$3R
zX8J7dG%b+{TU{r^;cii}M^8-)XkSuX59}4U>{6$#YEe+(B!FFNo5T8#fso1^6+*&A
zJ@fmY=M$JIQL0~F$Nd5&E6ojs&77}ebmY60z*$|7vcK1qC{x}oEgM-fX~g`InF**8
znQ`ul9BsvR`n`n~9~;G%)!uEstDugyoIU<ApXouwwtc{@Plr5}?iGRb4GH=V9U<q>
z7!@Vd9$HSjiZc|>iC<QKjoqR&z%?=rA{D%Ri)K-w{ltRRnC1d8C7aLGTbGIGR@%ig
zt0ZN-w1cFb6!-6RiFWl36Q3OzX{jjEPNT5<-3tIE-m%X%J{CxNUPXZSL(i{R%<MN1
zbYC~Ir|;CT!M!=gD{fn9Pw#l`mvs4;vg`Af{#jMOnCH|MPse_c{YonKV7Mgwb)0g1
z>4n0F52ojQ7htAC$#zywjQvhd{PTWK^(Wk0LhR=zfA=eoqP;nK;KTwr+24f<*i1~f
z(|-;)Icmg;&5S5YD2Pm(R`_#?kC0#@3+S)|;4tQ^QPr<2B>-h66=r$OBxypcI#4>x
ztsn(%-F_R87C|Hv;~%<R36A>w3-51{P|%0mjqxp}Hzyp$?VH*7X&|JzE!bEY6onOs
z;XU!m5fjcobApXA4hq8w#Li=)!CNsJF-Qm+f{Y<bLC7!*u#=!liP8v1?_66nLM1!^
zPK6NZAZZd?P;=9Ai|-y{*Z*qRBgP7TO}>rIMIVwku=7D^gxsJ$FGYl%d_tdqyZa!<
z>BbfrcBBTk4;I5&9*}&}3SEB<V0e+#9B#XmomyLPqk7+%%Jd=1!*|85V(AZ_5YCbf
z<rusGnu$>NP~Kpgx?)s|=q&Tmr+OA_`#=%2m*8_qhy7n#uQ!;sk}9+A9_q0)4<(M=
zT1E`q)V$yXlD=f8m~DG~rbhz_<rn}5IieRgRlR|~G;|N^$5{gyS$DnD>T~#2jbQeW
zx^o%HfIw)W<OQhBVPf==?v>(z71#iCGpD0oi+#0=wb!>}xg6`*9GG~iD{_~p6BsHN
z3EYj6+lMAawe9ms8yP3}ks}K)lcO0|`tyhmS=$PGfu5%Ny|u@PH-q%oTOXuG=HQyS
zRKig*0RUk+S^9`4t~@99Igq!HG1Yv%4=Tk@lHm4CX8zIGjq@O6b<uHI-6}RE(`)|#
zv>y%Kn?#R84w2s1{}J3_AoIn+HCm1(@NnQNOp8BlQ_5}e%%sj+2)}M3Z0bUM^GjmS
zB;nK4`-__BwWTQ<M-?EbPBCnM5s2El47us+@uAo%XmKuL8|xJqJkhj+%qxtqPi>tZ
zUOeR6_x($pht;(@66xpP2JXEP^Rd%cdw{?!Km6?ye=Lg9{g16e5-;?A77yBg_gRX@
z90+YmM5`0)1!Bf;ju*rB#ge!nPcO9<y5ApK{Yaw;ocKiCVeEN7;C36ZhAH-dxMXXL
zW8cZ71u>civX4yN5e%h6xD`$0Fe&zKBm%`gCB+`aMyd^<t5h$q@Zn<;)4yZ!%ac=I
zL>N-e$H`Y=Y=N?N(u8VU4+Q4<r+_vm_oMiIO@omuVyKK52K?H*z^+oG{`w!<`W={L
z(=>_?+2hH$HDgg&jTOE!gg1vyJi@v6@Fc7WR>FLV(ewm*hc1LP)&u@ri}U42>!`OL
z8X@QU`j`sWRLBazZI@gFzC|!uR)%ufI6CB$cG>!XK7a1c^w9K5B`wd<WRA7NGCa+N
z$ak3lA8x5FUNiQQ6&x|oV&KR;JOsJ=$wLN-VhHEaM`p1z-(GStHs%_7))x|7(;SGz
z>6V@E)kj{RhZ^B@qU(}42L`LQulDt(0|olX3W|6L9F5ZRZ)xAzTSd5Mt6Hl<+TIn^
z1j=!XW}C!UNSa<7N0ZodyM?t5i$r_Q=I_d}<n03$)bCED#4AHi>a#2e7%)UqYGQol
zAfioj24yk|VaZpLs0mXS7DgOXtzBFmvct&Dw0a8RjxGC~25}C_nyQ0{a8$fFyJU*k
zR}9-m=X*Eh5u;U|RRgky1=*DcojwY!o9`4TJjoCu*|6AV_SKt0xrm8S3lawesGg*b
zq(GH@ws2i>@c!O9RYNt+Jbw9Rh>BWQcp^y&JrN%lALc=83I)yB(%#L{X}+36#(XtQ
zrFpsLhm~>?ZxB!1SuEjcbjn!N)M+;ZR(k<rrCyG|w@J!4-8A9_Tg^d#hNiB^OYB-h
zzmVH~lS<5WP*R9;d<NQ-cexfdo=a3}Z|XLVwnuJ8k_ZU>!TS2MaXe9#g{})v`fnQ(
z!}Ab~8!yr_onQPt-z5oWxtl%E&C8#uP9?`P#zy7DBlHgrQ(X&Ahii<L!H_NP;zU0E
zH>Wabt%#`DE1^_Q^(QWYw}9GX5B87vv*wFk-FfVuHtFLedGAanh_zBVyuB=HV`oF6
zbQR(?AGXOdd8r29^pQ+T&G{ac(_O(9H#{-jbK`n}=Awrw*)_>`HiOT$)oV{d`jA!x
z!&H)<h`Xeq`Jf-eD%&plE6%-hP=}j1FVn|o5Ax~d1Q`ar1^Fu?=TDQIB%L}_>S2o2
zRY-N!%ES6U5|Nv6>Ls|OKI$lNm~dK^oB|6L!{h2-x)-G=tVRL{Yc2BFgu+^57P6bO
zoG9aN(QhdaKkTNP-(uz-eM+RIwpGuPL!u;>o&FnBOGswbmSII=S@mmE4N1A6j$_6K
zhp~H05n?&xQo{yovG`W+g0`;3GtypNUx&kA2ZK|p9IH-`n(gw2{px9kPcAlEa-ow}
z46grVtTm@gKdRjtlM~ccVAbhttkK-}Fum->k+oj%uK2EP&Su_mQ3tI&+cv!%Mn(_+
z*$0?LrV-rvJOUxd<AR{5%T0zxxyERU(?DfeakqHFHj|IdOp%#1sJYYqcsTs}yK4Gf
z-mGvRky(aSk>;r&D?0CtrVA`*fgeSa;!`vh5sMp7>-EX0SYa5sbI+`lmti~;vN|tT
z*LN^12Of}_RUs}8%76DFhftV<aT$DJy7Xde5e_xaLHC6}Zzg`f;gXLY6n3uvL*Quq
zofz7J$<Ze3xVMwjrCuQtWIQM1B$9D-cy3nLhLFS(054)!77l8Yvud=~CA_bS&Gwg_
zyJTAMHr>LHJSeZ3Rw7fAwp7JzwOc?wUi0QM7v&etbhQJY@?)uj!aMn$>-m3)+j2`^
ztYtGv)a3VM58nCtc8-%zl{Q>}CYf_c)#fMVEw#R?rfVk`x~HucvJ10D`+#+<l{zEY
z-Q-^)VlhA6QyyYhxFNrP9Zi&^k{}!U6=DZ3a@`R@Uzwj}Xx-^-Dd)Hpd$H|R{}fQ(
z_LkL(!%&ZDPR!Hi3a^#`H25Xhx$(6SwX>J<+$C1)Q)8_A2je=N?XHyXAe2c`iKDIG
z;{C5G%ThZ_rS|4@OP%Df{ygmOe_27Nj+&M=+}irl=kDg41(&d21RMg?bPtKfX&&2@
z#m}O4IPY%(6;6sZN2mAR)x;(8N!FImx`0`}IETxwv#F>rD|?+h^%ePJim(UX`mzhW
ztVxEo7m4Ec$B|2WrYF?4ck$5ZJVlUzUVp#xeA!2RzPLf<ng_qnNPo&N<WvTP=qHN=
z>2JJw0r`Vi($Ij%dIIuQ#7}eDDlD;**3usw6t#ml)jo54BPUU(XWZ#m0Iw&l)adr@
z$8GBO@7>Hd-pb8436bdp%8qUXFE$5yt{$#fv>gW_ZYHUmF^~e+w7oum<#u&V`@3@V
zCi=-;5gf~Pokn--p1%55#rd`iN2kqTS}!TBYO2XfJQy!OE3lYd=8F9s`WBE~nY`nj
z<({xn?^l8(sFv#%vuxWcgp9T6?)Pfrl8aeGokft?*}JyoLeIsywGzj@i5A65^g7aI
zLQ=)%iw=L44(1QP&>9tDMPgxGkQ&v|o#%VpdZCYNP3=fnK!ku-V%_q5YLTKBG`Z>F
z=|<i(-A1XsAMnFdZ#Eu=QI&cSlLs;9Bwj*KPu#DD;rZ%gXl?m2x3X^Mjz-paZ@CDa
z3V(<>jUJ;kVRcw5CjLgXd?68)M%+eD$7jChW@?08{5TO87a>!cB~t0q7YI$_1q}m*
z2jR?2$g`S;coWgCEjchb|K|n02^SQ?3&|W47PFY`|5eWabdYp3IxnPX5<G0j^}EDU
zz?}$58+MYEK&Ubr+_<P{4uZ;TOe-Uy?Nqc#w;*et5wa0Kdk4-KjG_DtPxnNiMrw?{
z!qivXjJ{W;E$+64Y0^?wN55H#zNE+f>E&#eDUaiu;0wlq=wS1Q`KroXirVOuY)6Ep
zEVG1dYRn~DQ}g@z1-YVh?-9DI{vY<<JRHk)?E@|pQamUm(?cX<B+5J!A(D9rnWvCh
zMCO??geXmj%=0{#F_L+nWu9lg^KPy7+WXymulN1u`{z5p<FGpP<XE2PzOVbb&hvNv
z25|}M4nsr3&7VJ#6&uCrBNOZ!H(_4+0;CqxW!>eYu?C^pWk{6d_Fa)A6S0j;+8?-9
z+%GUm(9Hs!E}4)4Gjo+%+@o`fRf1ls1IOg(`^9~@1(`o|=`3!4QpxCSoA|`GR%4?b
zdY@}et$#P`6)3@El`YOkIg7_s;+L|_+h`l91KM<%>9k?_=h!O~!5#Y1%&Nmo=KX`O
zBMm;wVzz(cXJ$(Kg0T9?mdR9^KKRQ#Nba~R5u+IeMg7zdb}(j9Q53nOta2g5=;99l
zu}F(4(@ub7=z5aOKx4w^4)P2)GBC&&{3NRx#P(xRfTAeFHuH5GQ)!@Nv25w-&7P^J
z5^rrZJnrzRSGB<ifSp`XNm)7?l!paRDa6lQ)P>&qLX@_|TQ6Th;tAKCm9gaJ$dB63
zv1Iv4mldOYGvv5BQ(h;Ty}LiJIvRZBX`ANF&NTWl3T+`Z#nmxemfS&jeRxD~dnZha
za>hn&FKKb<iXs6|W=fJzaL47!OZBfZZOf&q>C&zD)sa2l+2pZ-qM}-&+uCh}wIHN9
zPzggBC=Fk=|6@1*CtTvM#dfy<A?vYP3;r&l^$DM*V`XCq)tNn}c}@w&7_*(F3`xSt
zUXxf6^>T<5+$4{Rn3#rQ>NyQwqD{laTkIiB7o4uMC{z<nF)Bo(f@su6v0Xz(>48do
znb7}M_#xNN*x8);?uvS%1DZ6GjjP1G<%>l2d@-?i{YSA0Z@=CC3jH|^;wRKiD>7JP
zk4oSQ)E&)7(Q9+0fYFp^&kg|zvJLchof0~(#O$0>IK0o&?@pn2nsykvF%tvnn2czo
z4m~)FCbwM#yD<7#o*SJS9YEmH(ctnjXgm5OYZD}yi+BD2*lg7Eg|j2O3|~H@%*wCA
zF7EY??UE%Y+Pl!L@p&PCFWW=6`iRcR2LZ)rm&;_?ip4*dt;yU1{fI=&biUv_!a?zv
z$^2p*PQ!r$%PF|^+p2tzgm5|75D&NJDw)~p-={Q=`)zlg!0Wnitcr)<7WsNnGhX`>
z^!t6>{>x!pMRG^jqsZW&pLDe|sOHeq^el@v9hbDEPMrpnnz!I_<~A=&drHI`TGoi4
z<w64vU4|#ior`5Z>O1$}ygkLLnRs)AzbT%+>C;-+vnfr$e2FzILU>6{19YO~MS-{H
zNq_1_;LB(9K@;z^WbyO^bPv-gQ;+p<?+-qsm$##jilGSkV$^J_|F!<L5R3h%h<u8W
z$SCI0u3gb+l-$t4NJx?FhUDVF(TqXmH;zDK1B((56Ji!m4V9~Hkugk5H<{>o)wu7!
zO+H0(OQurU=Ib6^bbK3+9?%BID57x1tdSfY%8(6XX9KG3D^cl|i8!C09q(lwUXGcq
zKM2SUFdWIYc=RSUeAUO$@IGV9p0p#|itdZ^%?MMm!#}VMD3%uT9XJKujX*}4NdV7u
z{OBo(k)WK*&P1{UlE0ppzLNdhUA3*uKaJ4_Twj+ZZKj#KZuGdGx3dO{V6IZ%_kcZk
zXyAjGVF`Ieryhb4qOa`62a`vCtF>+dABLX4ZsLoZo<O*|$usO^(5+-v$3x7HWPj()
z(}bFcqowEyZk2c&^cTOT45699`pO{EBN=t3%=t2U2O<v*RVhPujP&2KU7C3!KI-6{
zEB+F#r31rX&#BsKpgr@J@LK?<4Tq7@kL_#1{XtRPZ&O~BUr-`fV`5Pf?`+o_1;{hL
zc!L2Ecaa~aP)8zpHgF505w}DSNOPgkmigXW(wOt)*K-p@w{xMh@UXqn{f)r}Af*LI
zCtu=ukmI&{Zm6o@imJdY<BZE7fF=|nz89Ic-zdy?7OG^riKmNBqNROHv`XH}*7KHF
zSV;<8YXB(Oh#7O|bZasBR9nMSwo~(yaRXeOdL~coQ`KDCKrj@m7@C58Z6vvDc0k0i
zji*wyL6zbv>(yvTkjinsxUAli1cQt`rg__Oc_0fVfcxJ7RwX1uu<D+dC(_sIvZ3-b
zRI86nky0Bj4khK3?C^NLXLa7VLUf2r=A#|tT4JdX8&EInO6P-L|Jkwz*Kn_-nA3%S
zH*I~ejux)AUN@EP{v)9L!F7f#AcpaUkl(O<Jf|`P^B(z6S7}RyX)eOd=I{hRQ076P
zf=KKVV2$ySYCZClAoDsqF1PdM5+MdaklIo)C3wLt-qbHwEtaX@c?GeRFfNj{`AO{6
z*bVLz(w|QDXIqcuole9n9nmzhI0i^s;kGp7;^eLaIKR*SwU^yG+W!E{*%BCu__SWm
z^MNKVNDfATkpIn(vTfw?G}&h?XV+>7_CyxL<<1~pA41~p;-&i%ym=RBgXcO6+W>Hy
zMuy`d?+2nADIF8u%AVC!FSD<%XXIHE=F<fTFSK&xJ?Oe@=;GgaV9`&WbBCybLOWo8
z46G|Mc!<8N!nGxCI!VZ#WGj9ld~l}7(yLwDp|2k3K)PwK9<ijb9vU{dBStbwDdS=B
zewrTF#5~FY%Ne|9xE64fE-M|c^>ewA><W^TvpIlF8Q$_Ql_x#fQja6U-in=pK%NB(
z=gEj}YOA8L)8^D33|7Zb{$pTax8x3Q)fZH@m+p?bYoGq{j$Jw>d|_}l5}5E={OGSg
zGBkVE@PC%hj)c3TDDm~R@19qEadI;O_@8fOYf8ct_QlVOh#z}d%!ZxzU!&7)!H&hY
zr7xqdN(K>JF<Yz4@S?8B;!I@+r6$BkIA|1Dm|Bj`Ea^OR(Xsz=|KbG8)o1|vV}DSu
zGthF0N=n8-vUj;hJdd^2vc|TPns(Wt3F2IM*gUd4R0eLI+VVDPB~vRN(fOmKj={+X
zx7~O|1(y;qFAk4b6;CsRtS%h&t^63L_K63A&X5!nO|Yd{PaGAEVBejF{h@Ym4D=6i
zirD;^zuA|EG5@f5{f~wGYYSWppT@ak7NfsT1U(FFibqe&z*W!|CZs`<A0xrkHZbsP
zvQT%B<lu8K8`XkQ7jwzGR1<M-MGlq>Zl|Gn+o<HVq?8nTM8pluoYu8()RwGaa2Ms3
zOHd&!^h|kQ0BntG!faQp7FWi<O<C)Eh6Oy)5`b=J+IWe_iUrBZ;`5~p5v416mqw{-
zb4U4Rw(%RmVQ&|chm^~2sw|@f#64cfzk;7lK=9hs!`C5eMgg3u;S42{slikN{rzHw
zf{N8`JA<$QlRJ{cJ+ern#1T(`<cK$j0i6=)ia!zc>w(vDGRD)zw;9TL#36ge`%4~7
zs-Ibq5s(Sj_3gX;GcT^*6OX}&hADTD<oWX}bpo@*Oov~}oY%$c`AcUIO-K};(sm`3
zlgXWt`ELSAYyrj!s2oz1yg)wn?zwy<5<r!l{Q3uGdvjEk5sdqimC{nY=}Q(2`&mu4
zLB2b*Ca~Ue{XG!XlR4vmh|rn8|9pR}rU8EkfZqCET^Li|YJNjl8T|2#bi<V>y)V9j
z7rRNzbu;YyB!SyVIXG!8yIpjg8(y*4j=9g|5!gh>38wh3!FQIcCdTAEFJG9Tcg53N
zAJaQG=Dxp&&~RJm9W?n;&KieFj);%cI!C36Pu-e>DVj!hqRY>#9aBp0!RCT{c@d&@
zcBJIIxRFUImSqr!-_f(;PcZ*VZZ--Ee4Og`xLBrRO2zUR<OQe@lu`Re*n11RsHT$c
zEvFAnc{#AfUKD$A{K|1Sd-b(a*jXxmwQIg*huy3=)3-M)Td*0K8;F{ouD*^838@Bg
zFYdH;NDd>ydv`VO{}|}ST{$Z7rg<<>k~D7@bD1Ct6T%w`n!-%CogMF<4(T;r#<(yf
z{x;NEL8?ELvqITFkGp3|M1sz~sV;z6;w>Nfz3n&THcHAX1>B8An9R!w0rOB|aa;Cg
zy~(b<uI!XY`+%*1k6;SZJkR;A(#7x7?#ujy-IZR`%SDLW-svCb48&tZ3YT$bb?fDC
z2X_>WgbeaFyvz*uJJW^O9`(BUY@cc<3@qmEUOB4eBOW<<L+OF!Na+5P6Lw4$BdMe2
z?L8iD55V*fYmbI#`pSms3*wPWNJLxkt?z7;zmvrN<D>W#D^L<>Tpiyy4gTz^r9+=e
z$`XZmGlYlG_z8=9)?uGKTXVwQ?T5c8L$HW4uFMYMHjpr{DV|y*5vThiRB}FLDu~+l
z25p3Jv=`%t=IZx3o)qt858^hWcw-IUB79K29rli?a}so^y9CAy=8swT32#k!FQ0fQ
zh+{*JZuGv;MUW$3pghc?SO9P~xiOA-_!bR~2!%+ev^PU1VglEHrqQfB-P9|J*-A9w
znbusC_@h@sB4nNPm*rU2l4=E)Kwrt=A>#f0y;TN*FpF^I(|0FL2|7<t-n2$&R=5ho
z2EctL)-LJ97ZEodD9G4U*d~|83%|hS^1_u7f`n=Ixc0c~C^zm~i}H!}t}=x9G5{=K
z8~4p$@Fd5V_A{c%x@Xs?FGvX2JPA+tC7@1RrLa!y-70*4Tvn<pJ>syu=}tnsqsK3D
zLh{mU1>ub#Iy>j)=gZvvhfp;Ue{msyBFE*G`qhS!xQ_e9JPczC3hH-O8S-vSlYhw$
z@20_~71}Tz#J-KML^;h>s4M)7kIIENN3XiJwfou}?f!;cBwSp`_cHG4+0M&*)D4)4
zL2IXQ-M`o&8I2Gsj!eA-FxI&mh#FXy#9q-viNxX)eX=2bHwI5zkfdN9=hKh-^3@dH
z*>5`-<WA@NVI5$v(KK)!aiyu5IAdD1-aK#UCbIEE+T7In@z1W?Tx;4tV<oeqfZWXb
zBT;XH1s8nJ0PiVxr70@KJy;d=YFUVSGqk?rjfL&px9B)*cVxMNt=*&pSvsKLfQ5Zo
z{ll5fk&qmc10bK;R9q@PJKm!D1=EiFomntPOJD4jl_7ji_6BY&%L7b;7e7V{C)}y4
z_`Qs2ta(hkROVfm+`N_($+l-btk&9JF4?&MJyYwvA57WD9~plBO&(%ADSdJ04XMx`
z?HOv?F|<CRySL9^a`Q`#hxC~*NL5aVlf^--<yQu1$~kvW360YRCSIy$oz)R{>@cKw
zCJ+CGk4g@%C3;&tqkd~gFD`BCq1NWWN^2`}Ak4+`g%`z4Q}IyteHGEv_bEZS&zco(
z#S(JjPtbWH8eU)gq-9NvFKefDdD9GZ9MeRSOwuhAtS#MIzOvT{tcK`&btj3%#l+OV
z3DKN=bFS(<m}D!ul{vpX9&@o4Rc6h!n99KXEH?0(Uu4&L-KG$&C*pe%LLFv_{Fn6h
zy<a?K^dH8u3jbJ!8%#qdQnkX+WeMbvbt;*GP+FQXN$Fe^KAPJ|_Nd@&X+d}e<qK7y
zQty#?6TAc4fOZlzS5MIy7l%QE@Ykq_^NZLgNffugv7<lcgKkDJR{&FK2qffK>lE+E
zwA=cYz8jXvWr-#^4&do?UFP9IcQVo{Ps_ye5WZqh@)pDi)rnAuDLa3lcgHnsf4#Om
z1rO^+i{qEZSA~k1ZcR^Zj$kCJe*XTKCuE;S0~o!@2##jGhdD=g!{4IiM9;L*E4eP8
zLpIfKTPcM+XbtpbWgZa+zrp!HyS@f>t!NZ0hliN>wq*X}najugv(bb}CdM1ag(J@G
zyM-Jc)WD5BswFYa_~vNS$x`maY+lVi$DuLnNEahAxO+x_6Qo3^Y=!m$*tRVc*H-Ww
z97~JsX8QR?el0fPVvZTtDS-|14T0sdF#mvX{GrLD>q8U&i?uZ)KA@2GV@+EBT$Sq+
zJ*C09h}Rq|n#devNIgVb3d{^v+J1U*>@T*vO#xPdhKIOs2s`Qru?X=a29jAiT#*?{
zXCdN!J*Ao5a_!z4);AS@QQs78E3Ed(q{us=WI-rgG^NPtCMkyYtw}k1H^Wcgqn+>-
zIkY(}KpI)l#TOmS+Fa}+JK4tj_bJpC&ImKd*|LrNub-=85IZ6JL%&a#<$l1duVK^&
z$W1_800f95@=zT(5#1=(+!w^O%=&ZUJSB9=aeAqmEcX(uMPEn^A}F<{4)@5uleRk+
zs{9<$`FB6?DS3N46F`xw;La2C{l0&j17vmfs<jvgENaycDmSWrcCj>_0V5^zh^f2Q
z(rMtElgPR#S8XjjkDJc^ceJ8#;cSFP%B?s3&wQ@C*G{rL!OKp8;0UkxNzF`p7uz6K
z<m#dggmlGlTlT-BWX`8#d2`ctQd%82rfDWfWGqmvSo4$2F=fMOlfd?>e=Vc<^UUr7
z7L8zrb}6?#VgG5xec+}V6QYx#I_U$%OKC>d2D7;d`f^#+5XXR0&2IFQdz8An-PW?j
z_=@}A*voK59a5!uH=|<){sNx<BN~-;6>+k<nn%<u{V#kWL-A4Y`G^6rR7bP{>su+S
z?l4a>pu~rL`00=1j(QEc%*){2f{++KwAuV^5C0FE-~rt4-GE0){}WCpqJ3&83vI-!
zrgOHsz!iKC>Y1ANckZbb{XK{2j}L?rg-QX3jEdnmVEf-_i~sp`&$Y|oX2U<X-uJ&;
zssDKS;Kl#{_y50q{Uy*VFg5xLx+PYKg1xV}X8hsrZq;291YbR6DENC(auZ`92|Nd&
z{W8r5Uz`tJpSAfG)n2@UhW&oU;g%*s3Z4QQWDHnBN!JHnFpRe>h=2ty4OUWbA2ymd
z`&R)bk<q$Rg!ohfb1-mi62e^0><vN1$r#0SpQ-7(7pHcmd%4@@57pFrD9pQ!FcRao
zXR*M1;xbfXzlF^WsOq^=uGI;Mwhnr2Nzr(FiqbLo1Lt@HiR=R?0(10gPp9fu9+5P^
ze!-NuQA)OcYMQ0R6$lz5BK|e-SE~d^x2rY-gH2u>#fZs<TiEK$cD^(e{?sPh5lAqx
zmT&M#>{guQjxAl02a<6KFx;9eWR*Xuw6=aWNpd44J;!?ezdj<KtkF=Yuoldz+x%H?
z`^%4l8E2Fa9_a@wB=+_l1hd*g+{e36iN2~splrkdh4UAEW9Q^Wp|Wv5wJ*w``iq1h
z7=D@PvVve7#Aq$YqK^|X=|U37ZNW&3Wa=xz<PT!6P(xJaFimfEI_)lg49;5jyq;}i
z_*?oMg$qHrcky(b`GWfwHk3AEItZxMnjmd{NClqa1<q<9mLPmKPV;USB<6FWqRU4D
z<Ym0D&C7LGza0<lpe5zua`ms9)jS_0N+`j5QvJfj$??Z%)1O_boOaW9ih;^$2p%tt
zm+^+5(|BGUsSn>*`K)yn0Vv7<OjrY|O>Ew|h%^9y5gi@EZ#lEXhpQ5=zKc@^^lV#%
zapV@ZYdc;S(3vcOH*){$LUd2*&TrSjI*Bf8VCmgr-URMp6KLQN8Ww}dICU6g2H%BX
z7K=F$njT_CCcLaGdSK^F+9+J5`OgRUKeo(2z9C`&(v9BhmfvJeZexsaejVVmD_^(*
zW^u^j0+#v6t%(H(s3C%Y6N+sPU^7&$WEQRn4rQ`)5Ff4OJQbywGno%NJ7V|%d+gVL
zW(CtlDuWvVR?ZMe>(4Xxlml94`{TZ<dS&%`hX)FmNJ%G#O&f8VJ#_(ZbN>6sFhqL+
zb1ySzBnTQHw#uplnoSXR(OYOb3ot8M1M16dkgYyMV4dyr6EqF2;7TjEAa-QaXu{DH
ztB@GZNplPbG*?oyvm6NqnG^z7R`deC#-i$*RHGHLdZGhw;@rfiGU5i_cN*f`gPlr|
z%i~}(Su>oEQ09<G!!0V`I@}I|8JbWL6}D#ScbXcmBpM@OLPxz1dE7+5CGc1$QUH6~
z$B5KL*SR7H*wCuPi$K;hVW!^VATX4TB}KdxYs{bo2>A+I06V1K6fUzu=bq8(7k(5N
z#U0~llf83I_s_@TKb#?+v#{St7&rVC{dXjLLjRPzI4jt+Jj{mfUwv!HQMVW{q{Z+1
z2F$d{dX?wJiMX9^{$dwz(BRPA{s@Sb82-)C6Z<xTEXRPQv|qj^%_t3l90mUFsQctx
zCJ%&GNQv@Wz^JNtl2~F1>?E$OngZ?@0p&pV5t7;wbY3~yE2;9_c(9^}j$tuD_8Mfd
z#6asBJ!QbtY%KB*YB9-${BRk4H?#}=<UC@rarq&8@F5ZeYXvSvuBzviGFzt-Z%(*D
z8U)y7D=I1QEpk0SDAS@*qV>9ZZ;nU}oT>7_WW>Y^xuoa~be!FbZN`dwI*&UfkjT}G
z+!ks-T`y0w`ycy@hd7@lJZxgBzxdYhzNDbv|2+EtTIjFyka}uyOK<Ma*IvKFt=p|8
zl5M7Z{7r=9<Y$0n=(QC{*b;=Zv4o`QfAf(qDmvNH58559Ha3GgViJ-lUQ`13%W!ZZ
z=b;{$$tYqHqQ5X>lF%|trWtgt+ccPT8Phup(DJaU=XiVRV%!x1+j69+f;1e)(+zcC
z1!>6H3r=OkjOt}9Su8u?IcVI$G{hLcYCHINIrMtl05~>|Pb0rXj<kwIXHEZ%`_v&@
z=pNM}A})cQ4w#vhv2}$YWB#xi3i&8_6@?F8&I4cM!z<=#`2-G)7F!PMzgbXnNY^O$
zF+)x1&mRg0oce3>^D*o@EH4WnhEw*vOw>Ej3>W+d1~X1_JivRcb?;Il+LDhR1W;>b
zKPFG)gK<_6iT9tx<p@Q5mI2Kcz&5Hu@V4qVtM*{RlAudSv+J_7MNy9&l>ej$Xh4dp
zZ@x`>9VX1%+|6Ev@dRUA?@|OYB%fLAzOhFCJhz6!6b5?1QD0j$wYW|l!Tg_`<r4XS
z;VgIH{o*WdVG2fnug@v*U%PNa;=+N~S{M_9)A}PHkqx{&Z8${BKCTak@=C6^72%M&
z%=m$!nJyHU4B;GJM0<35CI#NN);#+72%Pu=e}+-N+y1aITq?{+dmj}=Q{MAUmS%$_
zQ2HFo#gEbZu5#5k&`dr?iNz$-;9}vNROa<i9z5Qq8fqSX;<{YASnf!@%#);0RGOQc
zdwt~A^<3($?Y7&C5yemM(&t+)CBcZ5c^KbgAyJoT>yE#Y`&R?Xw^pGSFMa0dSqr$;
zTDAka6@$=qw>p!S0C2IqHGOR}JXW`5pEmP-->p}l4>yu_)l;d{H?KkDp_|io&4>N%
z<s@tUg_HT|g?Z1f{mi3YrC(hB`WQTA$Q){TdtYG)HE>aS?C4`I)b^gFxM<#8PU|Y|
zY@?vCRG(`3Einn`TawblW3_%CTbU2LIooA?F7bcUIk-_Ixjmj<6(*GMz)k&nmV(9e
zO<xY9!&IKl4IQmz!N8dwFrK(o=L(|=Ex#qsZPa1yG-5^ZNFYhXx%*Mmc2xPzcHtRX
z8jSz;wJ0KtQMpBioe#d}!pIjd{YqiwA*;9zkSMcY)1t#Vjl(ycU?*RTNu`e`&H|7(
zOL`DNqgzV(nC~T8{ywld5?%4S=987ox0rX&Ng!1!PchSJ7KADVAceewMTqR=LjY12
z%zcP=K&`Ke#AKP}`<I;I8@O%6m-X=M+%Y_xMw#VXaiE8FMSm8{s%<ppHzZ7q4qKPN
zZFd+{5G~4sRC5WkQgGJmX$Zb(0PpK9I6G^i-TKTvWt<H&@t0r2<!D;=INmqGo_HpW
z&v9bxx$_Z(qL8Am!F+uMwLQ74aLOI9=qEVhR8KY(_rXAmNdG-lA<T(EJmnf#;(aA^
z$L#K_czodAvj|98n@*b3otuISK{Mao%XT}i2#)tKb1s;AQcpYrpBwUZ@9;{eYZl^@
zs>>7^IE9zkkh%ihAbxihkO!Yw=OESsMmNqa5$PrwT*5(2Hd=SEND`yr)vNJVyZ%z4
zmtK#&u+QjHJWK@eJ_%R(-1N48zC}|~y4J)jMi%k}a%t?h!1lto$RdC~+W%2Fj#4fL
zx%UI7b~?9pFun{f4nT!$M&D{nxV%AcchQC#gTmsb^(UbD%HvSJ{N9T>Pbx~hS3RhI
zc!k1V=PvidFX6H6(CgVaEz>>Z^6GPogVB<Hc}8CM!{%j3JK_c`4>9hlsWsZtaHQV$
zt_sWFLilaLLIplJ=QRze>t$(mq;P$L7pw9a^J{hZAJ2YMyJR{7r{6{HHk|~8SFO2j
zyDshdq#knVFUxOU*jbgYawT@f#xU)suprLnTlR%b)6LI*e4gYyHL`v)tE?x#L;yfN
zzn=E#vqo=fEk0ih8<fu|)0v*ugL6-8*xRmuE5`AE`d_|;T~XmxnF0+W8K{$l$d_jt
z7Ah7%_+GZ6Q7@D&3(SoQYR)j%HGYL@<S^1<<EW|VKCzz=vlbk4o&+X0eh1k&p`Z&}
z@nj^L$@y@zXAx>y0?&)d-rxq3nO0b6F|--zw~{U@grXhS^DUrzEYcM%heNt6tR2>S
z&DA4+!a)z$d@uIQ9*O$Hh!qqPr-nKh<CunCCLb)eO0)P7&C3vwifzs*Qodm#96`(y
z&<C1ksqe?%3I-F(zp`z*I=0B9dC}kzKo?~C1l~fyrxvT?scY!4-uSDco*hl|RAwXH
zhN~+D-O5`ki#4d$@go@ZF>a2)1HsYYM`0_S_erIiVXLGH1J7Gdf51P6SJvOiMCj1>
z5-!I)gj`v;tz^>IS#MnwU|kV=EnNED8^`Y1Ic{kM^P>#M&~7ao((~`aj1z2p#{$ml
zF?hwu&<Zx#w+n2i_nxtc-b9U2Y(2B2;gm?P=Jl?M{`L;%#~PX<bpxvCv&&(<7m|6G
z3_>LMT_FX^dvqBXyY5{IRP#*c)()+rgMPx%`?197eK@BG6A(wsf^H3SpI1=W^c8^N
zqa@#FkVA5IpfO$n8X<%g=G1PmnCLj{Y+&_`pw0;A0gAe_F@3x8B<5*xp%s*j^y6u#
z4lVt1iJOVKenF!j#xG*k&jB&1htgR9y;94jzTR7__LfXlJUKNycJoY_S9<G<=8q12
zLAVNWzLWJN*_UeKrtcZAgO^#N-D17&cQXfe`BUy?x@a<^3`i2tO+SYdsSY=TpJ9DT
z)!fXlyhz8Bn#y(e8P?W`t;%GmfJplwo$(>$*qwdzSn(xSCi_yo>BuJYf8AnfzZqc-
z=6Xl|Ud6)k#7-ccBHreVEP6n1lYQH-d#`bzPHc`I7!yW*<`0Xme5xxiy1Z0<b(Nr+
zvZ-kFvW>M$fS=nPa5f*WJH<`ebl$M2F?I3_T^LJAmuaGFp3Ff$<Lce8SJ9S+o(_6*
z7|C_DVGL}i=e`NV?no`x34F6JQh6~(olU$w8CCR@(I!(dz;#K6pI`U9{jzpI5H9g{
zma4-iyqHMXiMeeQ9()CBz=)zN?{ntaDx?Q5-5swfC%Zy=?uE93g8RV)^%XV>e_J*x
z>D-B)QyREE+7F|OP^k+V`NX934c7%GAB|Jxrd?pwwu`twKKwlra=waRNR4E>ZeytC
z%6;T1-L5iuhR5ESUgB{PkCd?M2{ljAcAb-;9aaU~tVK$q(X@#efklTL8;i`~fbzY3
zy-K#0`McN!?Iou)w_o>-xd_(!JXG8Qq$+eb2ZVK3*gi8~s9n;wh<#vfTC>G7_^7r{
z;5I|A(r{?@z3m;6dEP}v4Lfa<_yw5u1-K3e-+GmiG*k42#l};=W%qrI4zG=f?A}g}
z?ly{i(Xrjua5p5OXqM-vZKhAi;P8q7o0I(DC;dJlQU)o_!CF0a!Ku#8%rR`0qKMrV
zH*Z(>7P2J7p0^uqk;#}pp|gP2HV9!cvsRrpWj|scQdy`!k|O6I9TYPe7d_6&_y3qR
zT!eGWreoQIT4TavK#%>iBlT2yL-~-(;uve;AW<23_xYqv74`OPkISD-{coEeye?cw
zm}TV}Np<=6())JpBtd?3(RS`6xNV<Bi|MEd*Zb#zxTS0C;tTtBeq66$$Q*PmN;IO8
zf5fhYtIvT&od3?KPE3G)#at@de{S%e{DEv^ckMVml~-=Go<E1!iJytiMB5bizY8pF
zvI{;cZEQ%JB0PmYTLk+fUNd@B?6Uu(_l(6cDNgb_57d;oTi>!IU)c_gGP-~0ljPTM
zhhNLyi;*dwpGD7Yft{$qsCg>PRb$k#&oG95$%fTt`2ojgbrJQ7j2~_BFW>(7x%ksO
z1GQr~;9rF=lHmF-*ka-2rS)qeefmC)sa5Ue>prj0-198Oj%}y{l~2?W0j>nW4)4|l
zHQ9RzGhnit+P8q;xTJmk=W5n8<>!bF8Bc1jhghXlZ9F#ZPkq|W?HLSas0t&&J27K{
zI@ze<TP=Ca)V(%=?t+qwib;Ms1d1PAjqjk0rk=aX;=Q?%yz6GxQ(p8D%e_D7*c_eF
z8+6Z;+9Rc!SM>$|3&%2S675efdhlIs$}xnt^<#^8I%Tv2-G-m5@u99cMDOkC3|^$x
zCjH}R`cHq(3Syk;m_IgFz5QqH)&J_m*a>ZE!==7fP^?{>hY=#7N<{ca@0iEd^sBNr
z@2}+sTr>23{5b4BVZ#kqR*ySmJft!by)y26vaR072Sdy5r0BDA#BZ2One*{vNi$+y
z<2XfEOA1qQu1Mhs)I3}Z%X$SVx6g}>%jG+YVtDPW9jA3%EMO$Cl*fNci-mk*Az5MK
zNam{q%XEz*AMDn6wL^ryk=NXpLb9^X)ZFhZz+*4F_VX&si)O^-a1m;qd@$eVC7Wb(
zyh;Cpc}!Nj>VfxgGlM8;X9H6HP0Md-<WD5xcgPM5mfrP#pluj+-{OrANm}<{U{Bo3
zw-=ZfeeuA5M>gHjJJyPe&@Ap-Obh!oxkJs~?65pt9@|)gd3G}(7F^x(HG(CMEozQ@
z=HEg3Wyt0dvTC<#SwHSg(;%sRV8}Mj%-l#pAIyVR9x65K%Rj*7ZquTh<eiu1CS>X@
zKNS~bZ`sUR*rH=L8+)zN`x?W>HfJ--_q&vvgLS>dvF(EsnthhRGf=3Kj#aNeabORc
zu1UyQ%ebj0$M7)6HHJ$xtn|^a)&kjaZ%{V-LYq%MB~{O7R}OCJ*f9&~75tE8ud|Ew
z$3Fdm<u}iDATj9uyz}YzZfcZa5=E9pI|k-?IrIk2pX->X3ThQlY;;M|CGB({g61>(
zLslaj2=h!G>PITdNL9U)@%~%64zCZ%L&a_-^%*<xkF7!201IVW#KUPvAkyGxvmHl5
z>vf7mt%{%Dz%5<eVieuqMD+;zgWeDER%TSgJ^We^k!yn?BJPf(K4`9dWb3#7#}4q8
zSfG_igd&-;j*qIT1Ib_z!}6}eena3@3aJ(3v{f2{4flq{mS6g%FI<z1;^gjq@GNB?
z>_IY=-+gfNmgo)fyUNNECX701#wwbYhe04^PKEyiK}PHdOpwn8#6I!k!T~Ej<eXK%
zbzJ%~CCH7Skz6z(2jhZ#zsRWS8&i8jp6gQcE_Z|Qc$injcI+WAr4Q`KNY`IEPbJQ@
zlneL@6GZd_ZBhBT_6H0hZ_TCFoDYe#8j_EH>g=RN8_P-gt6W~GH4@0ncTQPKvwyS-
zM4U6(m-KIX8})UWc~aBrZ!?+#%{%FE#ZRaO!oWh+rBG%@#C97gZ$oo{ivGBFTj``*
zf!n80bz@LFQjJ@5FTjTs*isQM656RIXb~8H=}I}jQsgic<SH|YO|w{U^vBL%-Vbi}
zL)BYfZcfJ<fLDXwjq-c%F67lBrEZvz<V^^2YAIJz)yf<a%PfwWM2VMv9Ly`#-W$-|
z9%u7OZj$)bfc?>i{f|F?($aY9Y8NT1{kc=PM4!;7_r5^Y-!|+r&x(+QoEe2{@h*)M
zi!1vlpmN^fW?E4duwuEpF=o(@v6_Mz+G3u`b!rw@rBSIymq>c*F+lJJQ1R%~BIn1)
zoWx?E^)83-3|}KCy*M#&A43Q$$^QZBG57Nl(NVvT9A(>MscV2KruFlCE4~mL<$}nU
z3(*pcU$s^rRr~UkX+&bVAHSkgSBWY^8#d?tEV}4PZH^xP5i3Sj(9hj(@^&T9SM^3+
zKG5t%(&L@Jrb{^W`&YHs%KClzbbVjqp;zxa=}|x9s|zIipKqWYB<{w`Wt`tZht~~U
z!f0}tzE>cS!MpYCD-@^gKF)fspAr>QsVrte@)}%D>WL0M2JzENzd{1f5hQJfinDBC
z3%NN;ziv+Ja=kVjZ+{eW$yQqgWtHQmxs1+aBP)?D`zkspZRLs!&aGwa*QIh&`L@Rw
z?Ce>7AE^J{H-D?eI`AN23iy32=iNF<>P$}+nw1sBcJ?;^`G@2B;dKJI^h>L$>4Ok;
zA6*plrp|}Oa5v+oP4%5|y%F*>FhW#x%f7=eZY{R`k*C;+c^3>YI={<%xW>^h+OH~%
zQzb?M*sE-j>6#_0m8G)kvgPG*6Or@zTc~_&1j3`;xK0bvcJL3=KC@H&j~|bZU1KKE
zvml#sS!nm+v<Y(+{-kE@Qtfjiz%NaHaN^+_G&&?l>jZ}Zt~#>(=k;AQKUT7p1adq1
z_sBP^FjbLJ1O67p<`}1&X-Rs&ca~zRucdW^+F*SAbF$Rq+f-X~<H5c^9nkm3G5jAw
z;eK1|;J%uafWw?<PSz<>e3I__FF;)A%E0C=tA(TT_|_UfzvFn~QL<I{s+(<ueL8W}
zUO2n%T!O&UTeJGFWF*Ymj1S+jZQqP5%F3ADFG+swHx^vL)%>x$faA)nU5A<?w!xxK
zClSHN1Shq1{2vALIt#Sx522x+{Wk1!{ik+j9_ew+Bj&AfAm{N<dmXO|*Xw<nc}1J?
zj&f_<WgB4WCtTeg1X2kOZx=}UHiho^Zztv8-IxEFznGlnGOwf6MLPA58~E?t<;K~Q
zNBTOy&t&n}t+)x$qJ0JEPd7hmDT7IH#RaDz$JU>aMAIU>PN0B($sUT|jsRD#RK^Ei
zVyriw(QoF#V<TKu``P!=DZRITfvr5*wayB;?^<!#1iZEU3f^C&(qQX@<x!$M>Ngtd
zP=sBXR>Qi`SbOS<|GhvZRh$EquH20u%ILxMtOSWL)+vs}rEC@(_UQLz$<M&J0KN^t
zwJ^U8-JE${LD1fhS&AoMs#hzBvRiEz7^C;6C}BohAw4F!%V0G&bwZKqhrn?`i?~fM
z_aC(Umg1v+U%r($(IdFgxSk!~1}0zGY;sWYxxBa1T+=$24e$}Q`m4{DOE0tZJrLTf
zQ(Gp;UGW!Fc%E%WZ_$=qHM1+<ym7Ch^ptxGuD3|KUlb8-uVJ)0!%OJh_|AK(@-5Xv
zC3ho72dd!AfINNnB_>Z@zJ>bDGY44`Tjj8)^B@9uz80}=_hEJi2!_G$3FIWtUQBjz
zIVd<&_}dHM|8l!<4_tgp6Pf;eYV8TtU=Z>gR17y)zhzJgs?jnE3v&VhvXb@~juZQo
z0kLTM>n+#BoQT)IRp9D}TZlS6^{c8pm|_p`d&{!X;_Va~_P)=<EsGztw9RWpI$yt&
zw%=Z^1ogKV<G3D`tn*yH%(n4B%PIgQ>ybF7iO$ovkztKn1pbUb>l!q&VPVaX!(-tk
zm1Uv3`e7S}Cw?o~VRlX{&u&L#1#;YR9;?nYp-97}%6@-p=oxh)^~LS7@X+j<+E~8|
zImC?SAxvgi4PJ&$f&E9UlmD2BYytcE4?|s^G_#-?pL^hFcwt&zsokhoN-^1LhZQQQ
zF%c9=-qj=%;vN)F2flZf^L{%pAImW+H*U?vt^|EC9<B0xZ$q1D!V;7?;%`a%<9kX9
z&7x2XN(>T}%2M{&nrjmUydV3gt~s>uL+r9ruVsW^{e})fj0QvSGlBWR?+QY%(qWvS
zM*y;aonVCuy3{Avjc^`~R`JGD>V~d!Cw12zQu*?|gk?bFeaC-6iQqXXxPy2JKSTB_
zg`#7`qW%cd;*P)35&NPwSs%sytsNYnuRzBphV)<j&0|wU*?PkRl#3sAx-q*w+lclt
zS-{Hon|cRZrM%`!SM2gVRRJW$KCB`MfU{~Ap{-u}i84SvE>oQ;%euIjLu$NDP<ITJ
z3{&iZw)WB*{6qHKh)o&SRgbY7rKH!R0StW9I}LY8w_tkJ`>}ORGy6QDxn#yZYJ2+^
zCKbk=i@kz<7LZ8mzP&c`&?~w}jH6&0_QvF6D&YWIT3+E&y~Gp6#NXZg?}Om4eMGn&
z6&u{RCq+*G3zR}|DQ<?WezQ+~P&+@EPtOf4c4X4j51?;3b!nQ~s#Rd=B_4qVaMq{M
za;*Y6FY#)dTxuPb$Cz5{J*Guj!H%%DMFL0B&!aAk7j|+%JH@bK<EwTeL4XN4yytZa
zb+zIoReG^@^U=wm;-Y2<nm*1skRVa7BLNP14!`^$UV^jqk1vj6vBjg86D2%y0^Ex}
zzO_&pTa{tXc!>BK%=JToV+l%qr@6A(*B{k883;_xsrK61_}5#=2uRmo;w^Z!gX_@(
zoVZ-aCe}*=jaDUd4C`TQ-4LQ1uOE0dSvI(qSb<;fC_=F9A~R|KOV`7o^$j5^rp`gW
z=fym^T%Q)Up*XrhFkqb9#`~#QO61v<?FYncdmIGY#%wJ?u3K=u`R3j|m)Wj_@$0mH
zG?^IbmA;F{33|IWA4AwoxJbi#sazNI*v<#LuavbBSs)ggRO8)R-iB<GS4(|tJf>e7
zCl4cJ?1B+@*L8x~4(#JBU8ecr0{x{bN0?0GZ%I+wD}P*kv}{`a{nW$ikIH&JnA@Qj
z3u2$-Z+#2$E${H-nrh^1Z9nGLAK2x+>A3oF5p07Ty48&@Xn&3^gl#w0^-a)xFRGv_
z?z{FJg76{Ce*VWM!Ow#D$2)o%3L5Q6b7O;MS_C+vZa})*QVswE9o*n(ueLp%<l%ml
z@|BE^)Sk?sIic2Ft(1Gj&2BxbZ(6iONdU>#D<<P@TP~Www*wfu<v=T~K&4oXry<d}
zB({4nSM%DtLx|awX~EuRGNtN<(n`Hl?othJfMqV#H<f83%Ljr*cdty3{!rDmOdnY?
z>^Oq@_FZB0VaeTV-K1N%q)9JbNrN|^9oE)+N>bl)ST*Q-(ns7sO2ucuUNjZ0nGS;e
zzUCo!e%r|}YL#Tnq;DK{*FmhYly$Ws_(|+hDb9_CbES{lA8@qyeo_$JeQ1YP^y3B~
zxaFn`+4rog<W3+X5f_S;?VF@ciOhJo9u#^>9ZqTwxtxyj*c(qBuM>z-PkiA+EuZOo
zk|YqDA8H?F|9o3w>lCS)`_NM+DS>j*Tz803iFN=Ig0iXME`7IJw}-`HPBUB3wjz8V
z3>%O@clKis=E~d&b-IF~?C9ZAQqoNc=fCNe1jL|QN*w$A@preRjWwtlS$u^{%%aXu
z)|=6P2lcdg`vc0a=t==97_EF&1_$-IXgm;=Ml&pGQYWQDAVjiByATJv&AIIJqS^2S
zlqsJZXWU*RPR>*^ih35p-TwvDY)$y|$cWZX!HjZk{o;~#^Ummyx@e0ZO`E_Kp9eli
z=<d{LmpFRM2?PEFLKM1x0+U;xCeoGmEE>UvNev{1!99Tj<CAhxz$1JcLzO#{#crb4
z9s>mHoyF)8Vz=ZN=75$F!SQ>IdFfG@t{$oXQ`90_S%e}+4yNj7vW2<ZYgOa?{6ks9
z-<YRoOZFOToK_?PfFHb=Thn%e7*EDezb$h$I;+`AVw6BAWcb`w3~x+E#oI?N0S;eE
zHMDdKUTu#359qoc+C!|5gOhxLeCWEt%gV$=9{qv2^1cyCEcccxSXIF%xjj+IrS)3D
z`zNs}9{6t#K7Un={T<cV!643>H;=%-!K~|7;<L=V$?LNHyHVIm)aI1i=twc;#<=J<
zw0P{`a|ZET9AjQt^>}usARJOnQJMfsy2v}HZuU~Id(L!L(0WqaSAgN10}Oo!fK)I+
z*rGU1WEZ_ZON=3i^<ARI%u1B23~J6eSHWOzus6M@_AHif$&$0~^=i1+F%8v2`So@I
z-}@HU)fw86r!-?}jDA86Fw=#6x|BMHL0MyeyuSF01l=YLQQ0Y6iLmrsQ}hhN-c$Y{
zT0YT&K4N^G+x5leeff@gTF6rUK}X%`OBN`Pg@|t7A+)6ZazVsYqJTlUY!5G|`Us(O
zcchC8yOsE;RHa(xuXeXRb@ix<T9RhRAY-!pspF`I5?z6|vbFV4P|#j7|L;@b-ytE-
zL^QN{?T>oy{n-k>X2bC3M6sgrtMRC*?RTZadT0gRU2yVOq$4j(-FK<lS)KSbu*m)j
zp5>wkUW{1mE~(gs$*ux!Z=D&zm78)cTE}HBg~#J$_PO!AwoXwzR%OMM$a2y!EvmI^
zWqRoDYFVw}@0tbtjf+*04o)dWsGhM++{}RBlV057vi{=y!N$SY4OfYxE@ZDnkgqBl
zW1L$sfce*FMy@wjM0ZwvB{j1Mu*8B--XUD4IcmAG5VqEr*QF%uBk;kwPcXc(_giA}
zUZ-nLrGM56)g+puRcy(S-6!@myMG@RPu;yfEV?>roHscE6|G#~@AReTOt_qK_ctP>
zptVSkUNcZnwLq?%DqS*$1s);R+()59Qzy5d7CILMO8fY+t~y;dJU~^A+Y^+EClrha
zGB;S~8`Gp+0+ZnEHgvl9O3f3e&&gI6f|Tsu9sYxz-X9hDl^vqhcvjm~Wou54#a?h9
zn+x%1nX_=O5L|lC?ti#)BqcCcB3Ef<in)FO2JdIt)BN{)`9=dsMW@A7$TT<Hu+(uX
zP^TOj-nES!%iS`je%L~|KlU<WlW2@FS+?Osrk-=P>H7*hYlX#nCt-=A`>d;i)h^mW
zGi{RH%!~Dub3LV`xolZFbWQTa`$qss$0s%nH<HIa(XEo~%QTl0C?@^h?IN)A{!o9>
zc<7<E5y693wa+-0Z{~lHv?xtk8ZLt!PwMY>ydeeH@s^A~U-@&#`&FQH37_z!(Pt6W
zU-P62^BRg$@}wRnS>1T{I*?!^KajZe)2rPXn(-<F#W+;nSaC)os$6Klubp;d@$B^$
z)R-pj`_wKMjNl827^_@SE-My?_cZR&*%6}0pn|9NO4g{G%y7;zBRoksYl;D%fNo`f
zjKG6T)Z`|OxDmV6_Pg?;At~ua;(uT9f4t&3ECug*n8(O0`>(QU<06K4Bi+I>3_EK|
z^&IUJP#yb=g}VI2VbV_7HQ&jHIR|cysYg7@O)X@oxkmIe`dWQHLz#ok2-HYaV}|DC
z3R7luPkV%GBxp72CkDW#)1cOmAaik=ipd`4^1YysM%iUYqXq=J?g*%0iN6lQ+33;<
zZoIRVi}W!rz8~ixqAx$lrLV*+pR_Ray&2Jed;cd1ci2zj9z4YT=Yz;n_UX4moS@VA
z6eiy8OP>)J;L?>Eyk;sM2<_7Z$z1bV|KcQ!PjOdk!4yoxOaqd=E#V`wzjAcQ0$Atb
zCP*3t2rKe)!Sw#siQ!$G+E)w>r!${`gO!7mcNr8I7TNVAvh0OZ>}WwyJy|+Arjy7N
zv0loQg!Yz8?)As5aU^Z-2r}1R^b3u+t@i<2O^4UI2XnE0k(bjf=C7;$SNZyDy=>@U
z5bbOZXWwi77f|Z;6c*yTwGV&=EroeA^YXI`v}Y-A**;92aay74TCc%lw}4&w^H_w;
z@BTRlE05IJboMuJ^*?Z}Uw_3A>if4U`Z1v=XwH9tvD%T+qW?Y)Uokx1{a|yU+SKFu
zU*GmWe~R;jFdX&|T>DSI@lO=(@BZnvAFlLhPc7!_#-M+DbN~En{_FR`gs>z6Qy3Ng
z55Md`0e>#BPqEwu_)oZ${hvPWzrT<FPhJ3Fb6htt8H>Ig(0X$sI3q(Xe^UeF_+Un8
z5!eNJfc5r&h%-ym86db(W{>rdIKv<{?(gpMuM1xi4Od#bSvCJ3ciIgTmxjs}o<+-o
zKBHXVt$d~z5Pz`+NXm<!aTYmWEkGTYXZqpghZoy&0#1M?FCHH)A2*fofXs0K6q6d!
zs`Fog0b>Qa7Y_?)yz_@UBW-bf8RQel#%uCwjP}s+Ca>i9H@og_0iUfGl9a4IrRT`-
z@@pRrn8m-Y1|CPKqIKMdR}TGZ%$@7jKiSRefN0Y+P5ML|tldI3bB4R_af|RlulEGZ
zN86j?22F&Tm{M2Y{a`V7OAq`Tc<Yaa!X*tm?&<0|?mw%MSt|^;@MCzc$w>t49bE@@
zom+DbdMb|$HLZ%@gHMip0g&DL3oAA=K9ss5pgbt0ji4ZZ;(z}`tp1rxoDcoMU^8>B
zWT4#ZZ76ROXl6Cd7-~p+sQ_DiU*>|p(zyk#NgqP0bFnfnSpbVjtNrE5N<Hd#M*F|6
z#<iGGfh{$@sTcPaAmz*2CEYyZ^$+f9J7i)I58=$W4@!|e0H8)q^XD#>#R#VZAGkfS
z1CPC&eN`L$s9aL&k`voDU&wuvTzJ$H%h54tw6Ca;^gSoT=w<gO-Tz*A{do+1M#`_A
zNB1-S+{eG!V)&m#{F6|rd1Y&YDtfsrgrTg!P}vIeIH(1oqiH;3A;dB40GX(B3k=&Y
zU3(z>m3b7pCi9WQCCB*Br9;d`Frr=rOp^7+I^5<0h=T1y!l;`*%{={9=5K_ubue}T
zRSE(C(6?)~Zi#Vs@WwyTLhZbI+<AMb#J8ZIA!`P}M<MSjxMuFLJTuR*36|vVEk_uS
zQLe+i$A80j4o1eL3xHAt>~{NK`H`4|+wTsJKv7ukty|Cbe#&JSsV}eJ*ZJReqSu13
z>;Gt!egC`sY{a2CaDm49Y&)M!E1}6L**aC^4EFbN2U|#<^R%1;LCP(g+m*-V?XIQ=
zZH^TNU_Ibki%|*8kU?aEbV<PFF^NqfHO~M+UtNz_H;Zf;8ew-7z$wuW8K!IY^+Rr5
z7D6U$UV{AoI<T~d2KTmH2I*t8^s%X&9>D*~Av?fmzavo5)Q>wanL17EfB6s<saTg+
z_CJKC@6rFl16M-mo-$6sqIq7Gd5Uh}Oth0<ZRs!^xa5iNAUL|<g9ZafPtT49+H0?*
zAd%0kfI252v+=Wfv~EvVu2jaXA51nVYpnwrGbr~u@GWQiUZ#Hh$3F4zn=bN-uq!bf
zJ>`v38GlqfvlCcHun$O*^<KVRbFTx2nGb=8?17Ggb@1Z@jIC?TA1XD%sZMeX1P{Eo
zR%;&samLw3!fXf5-@Bu@>dv2R;}K^Rt9N>W8VQ08a{#n4h8ql2{IE-28pT~@)!1sj
zj>OOI19S21xsko@LfXrVpog#qD@;57XT}jG1dGVd)%0@*vEp<p5z~@l?6wdZ6;4`9
zjbiy;f{R8nnFnB@Sz}YXJ&GIpW-Xx4w!`6j+1qEu6MI5f=sU;ZR$(ROgkpwBNY(+Y
z<pn4%aF)7S*G;7e9c#!8NKI*97|{<%H+M06)tsIcatLSW=W3*|dz|%Z!(~slXbKUE
z+bk>Jle=*El$|@4(@3YIvTOLeE8l)bA@%N_OA~{Ml!JV2jPp4^b)%Q{IGza#i$=tG
z?AET6hCQrpK}viBSc@6=M|*y{=f%r$(aM-h#hbjN?mcPFGk<=fMyMhFR+G?{Tw5PO
zZ0@V~lKoXK11)?K{S0+$>D3Fwa~ic8cS_YA)il!<FXlb7@i}$qgSQRb=e%l9$8+7*
zKb8Kb!23ln8<s)JR9;)N!e2+Sa1zZ5?3j{jw*}RN04iQ3bVFEw3;W;~Yl$@BV>dVv
z&nCnSYcac#Di=uSt?+LG6Sch59f&RH#L*8GO;*8R()(-ky2rdrGvqa%2)th-4qHem
zw^0jODJU5hVZweUdss!&@`B#A-IPQ*H~(Y!akI+3ADI>~Oll~{sfdPs+5|IkGZBm;
zCe;1+Q7NyS#skp8cjn);B7v$4fFoytn^ZrXBFO^&#VN)gUpeGNsc2lS^O_SlD2O!F
zST3@eut<F$o#R8I!N|`{n{LAOgKm1Sp50v!;nU+#<MYnGMyUf@%-;(+i`A%T*4!3O
z2XM^B3X-X5l~mW~1F@Uv?Gt{n<5{)%LB`Yj9<VJn#Oh!7U&>Mxy`(C;d1fMO23F*C
zr@QRyr^aS4sA?%P^0o-Ng3LnCz+V<YEOE}cfxkdKzcVOp#CCW<8`uwkuXXPY%TG6c
z41~Fr3+NH&jUjb2n!|2yv2Y~R9^123%dXWh-XbT!vTU`XVAKmQ8rP`5m0=be?W_w?
znD}&h&%CV}$}ID{+PFnZ%`fca=TI-25Qq48_IEIl&tWGQg6nSm5`b@FJIBz}_<eB(
zP(K@n^F26XVJ4x6NbqW=(UMI@1UDDxnV*?XfbZ!8;K=$rp7(aa@mIG~H%1-iauZP8
z%*rBYL4vZ|8{!Uu1>n0v5ka$z2`fV_eFk{~kR*Uy&<bQ`;T!&QkM<un6Jcu_IFYQ$
z=<^kSZF{fViFUyD#2T1-k~lFtPs6Un$<S87wdh0S&w=SXRZpNdkH&fu#F8WJR4Q`W
zPxwrxAtC|DDBtS$u^-J?N_1TcF!ygpg)IW>(*b-E{8k9wTtb_JfNB@@{TCV0{+i~+
z@d`Na+fZ*59eggR;B=6R`oNai57a$Y%?F;R7U0+^fDum5hnGg=r2yDkE)~y&nys|i
zK`rD*Mi}M+xoT;w2&?k6+izCo80`}Y@#xLH?=sPSf*si6;_dwFcl^)i_WZo{S*qm*
zSX`DurL_P@t3j6VB($qQP{cV&&O*kF<>t>z$UlAWIt%nLa+kPF?gpO<)_VkiqxbCf
zV+yP||9gbGG$@2GgT9*{{qsUY^)XEI1CNy0ley}gonl0VHCU?y3xrndh{qOj6>$o_
z@YhNbb|N+!h#v&hGee?WDxbo0hutR!`i8ix*9nJ^)`P<cOlVA6IIzTUcCu{hxNuq!
zs4vS=JpP?SE^dDv@6*?}<kB^RZ<wq9wX)Kv7tQ`$Sw9!ClexG0;MzBXqw}HKNS_B`
zZFZp`*uSGY4MU8DBu@0Irxl*->RI4BCAv`k40`|q<xyUGSltWsPUCMqaAxP)VW*{<
zw^O8cV&D6!;7&v<yw?=hx{tA2rg|59|HM9e>c#(3;lA!6dZfTB^N;4V{OMD?ok%E8
zE5@QrPFz`CsR-~`Si0K?-DXUY$>rUSF!!!84MktP7$V>c?lkVym5*N*^M&!Y_=EF@
zP@54SuOuM@Lc_w6n;d<QYR?gGUwcth#cUHH+P4XGzgspo_s8`%{BvQ#o3_AvWohn;
zO*{WrNPLJXqWM&7fBZVhBDa-0052;NgknWm_*Q1ohFjH|*5&qjJR>6(FEv8-L@Ncl
zELiJ_x24IZw6R_mt+3b-%J)5CGm+MEi#dRSKWnpsf<|>`c@aI^+MQ3qy4zpiEwI_x
z=v}3be{hFC!RBgazo)C!9e`!Om`SAR7%b?iKb()Us<SFGhHXh+YgR+0MNH(%(yi+@
z9Gc_FO*!U2<l`2l90#xt2Vrd!Z<k%Pi?eBF*nStK1?@V=4EWK`)s;UpWN)-7?Z*oF
z<j4=3$1Jw_XV<04tPfpxg!=m!$Y0JlKUjP~yt!3->PtY-7!hgF6}Hhf$E$kh1Xrps
ziMP|^idHgfH<N>1281bFEAAED<k)g0U8zruu#O2@dk!VR4Rwg4U5}BOX~FC^?0$d1
zhdJK^GDP5)^@z&9mwEi}q=bKcrTpm$y~e-Rgv&`!mB7wHJNfb9C*!(%ii2Vs3`=?$
z3`8(w9xvW5!$PaGel5plpkm@KK3HpJCaHvJ01=kC1cp%mT;gmbeY6@am;kud^X{Z9
z4<8ft&+5_RHRnf8!HIThaWg#Mr#<aPt|CVdJh&5+a6cK{S`feW<PQh+D%p>%gVAes
ziP8s186DtfEkCHKF_(=@;jT5|&B&ax#oi3GvMuy0$|dxqE|{UBUH?&kYh2O$GLD{^
zT<g|bK1(k_j$>490+9Rem+r)m9575@HF%`}k62#r@TxFbj}^Cw5INz_js7r)f&fWz
z)0O5A4p{*%d!E<a--d9X(cHSb+4uPcWo|<zB(8kjuGlWmIgq)snfE3oRC4(0VD1A>
z=s0}II>go;P_wHEbfGtCZLG5;!h*}7`MNqRG06AmWeR`!7+9{$E-0pt{Pi6D<AM5@
zZ@$GNz(Kr}g5@97zu9|OaTnf$RrIk8e+$e$Wr_K>#&&1mfX54aIJpP~^H}jGk`(`$
zmB-V1vgTK*4Efz2jiZJ5oR*qZpVGVmVvHZJxDV7(Kp)vpJ6g}GP{h`G5~!5Rl54+q
zwm*n0<z-h&>^wvYeA-bXNTNt?a`RNl-uM=%=<(AY)vgG6@CGYiT(;K3n-!O!lxX64
zzfz{tAy*M8S;rJijnOHCdlC=hM_ZkB-+nx-|9-5+t*n8u%?>dK)mOSLmEac5t#0zh
zN3`qZq8%)J%k)1Mz7!osrKDx)E6d~vXZ%VFGwk@3kJLbm4z|Y`(=`5CTXdz?_uPEC
znpWn=bEZD9)JLZ-a2&DZwnS#BhEVcxucAf+{G#z~1e1m8=A9Li=hk<T2#)6n*{W=h
zmPU%}%bt&;$gpD(wSwpWdesMT??_@Fyv_S(nsqO{^%fuJEwIV&8B*Hr{(sne>!>RC
zt$$QX79uVh6zNiF5JX@B5~89aqQs&>x<y(*5Trp2QV>BbK)Sn8O2Q?b3MkE@^UlY;
z&)Mg_=RN!U#vS91asRk~ZNx3{Jl~k}Q`3S4*EtlHT$<B~^d`=^FG`CAoz>)~j2+qo
ziKa&U+LW<LIJGU%n)ktM3~7Nx`X$GzDMC5ts2f(A<BKq<YaU^vT)V2y4m&k7BZiiQ
zAiNkFpJ>joU$#^uQ}L1+=Il)QXC5<4FpkfjgTm$9hPDOE#V;%n*|qy4p%Y1x8~80>
zj1fbt_Wkr0yd%t@(+^j;70jV=GTwrodic7`veH8piNetIpp1xRM#xX|aJRq<Hy+mA
zAhr?=7Tfr62mZH8Bz$9Pg2H+HV$pHBD1CWrSWD$eE;t>!)?4L{fN{nJy*rEUvVl%e
zzc4l3rod^a#OdP8Gu&v8<1!dxmvj!Ztv|Ndqm9{boq9$zraHK2e;S631aiwZWyz_L
z@hjW|UwJHzDkr-lVVlXkD6<2CZp!4!b?IyEt7*SqeR9y(V-o6WazjiXvVhyojk2%W
zn%Y`8d*48f$^BBs$mu@LKKyV`vIoi$pR+QRs)`g4ZS@ZZc@k|+77B}hNp*O&z<};~
zNb?GHJHfEqDD6-Fc%D>l({GEFA}-8?$vlnef1r^6j+Uh`qhQY+OG}aXvxt(XAWoo4
z$3Rieo~Mw~&}=X3$yulHFCsU>w;!oDy*J87BckRI&*;|?&v!=7oo39$KFFMyQk5*q
zXY7vex!I5P!CfFtYxl^kD(@DfU(2yv{j~Seje1!oKUv}w&i)vAJ!c`e6K@Ad?9?ld
zc|btVZalGEQ2(^OvFHFrUtCsh(+O9I<?Ri4i)^~?dsn%ZTAzw!aBaW-ez#SC!RC;u
zrp*ZvlYnc`iKI`!!FD#spxpW*qZZ*4>378xrGt|N(j-3fwlooKc$rSm<}^Or)MOv1
z_b|C{I97FSvI<v*tu`x?C6*p*R%tA4Jc7C^oE_^Xl%0Fn<B~oT1h{wuWS0BVea68V
z{ef<018}T9a0VaWC<*GKjP*j9dV!*2B54{Kvvhv?MgGKI+5fy!{>K6L`>*WOQi{vj
zZteU7DafX@Je!&6^+Q!pkHMI^IsCW=)i(e=^tFOVunNTHtQUhvpa+*;Mlbiz<(Y2O
zJS}-!javSdOw#(1%C!@-%`M2n*Xr>7wR((lCzxSx1U)WL{vp+BbM|CL*bzQWZ_o!R
zOE$y7mUl1u^2Epv$I(=;S=AxD!9-nILnG}c<btG(am#~X9h;$F6p8VRhMeO(WX)xt
z&t!bhI3xWMBwSv~!6O%K{Ugg5C2(X-+3~%wrhR`EX1ErRZNr$^k$*eP7JdridH`3H
z8H{)nHEX2^iy8HEAKW+eMM--H;LyFKXLN$yooqN!t>B^K_?BB>!O6uF$Vn{<$R6)p
z&wo9&63Mh3>?;NY|CBCyha6G}OOL^RSP;ZhF&hs7M{#3$1kBEo%S$=>XO^6^okD_C
zugO$cO&DXO3j!tdz>mTngRg!(v(%LwwYfkhxJ+3Y5_EXp(O0$}L*x36Q=sQm>+1fT
z(1gSQ68&PHKP$U%-OrFu2RV_9&%qH*F71rg=-Pl_-*bp`=Ls?rnfUaA{(EyCxSLn#
zzHw?^UbIsY(Sn=d<WSfh|KKh~0bUl|d1Z~wS#~MK7M**!%y#~+I2$c!saDheu~?lq
z%`~L@lrVepy^q{I;V!<?Czipz;S<X?^^q&JbwpQ<3A=XP>GSEgTma?y6dh-d9L{%~
z%bpY3TAfOC;E~(<r1k`q$5TSjJ21EI=e3Epa)AWP;j|j*zpuyt@tH47$Q8__a?OtZ
zH?lYLl=v8&1$UZ`TZBjTdn`u;HD2YpmhfzBcp5B&UdLYo93!{<rQ#EmG>J>KFVQ1b
z6ICz}NIJm9-4HOM{wEO1C3->Wu^P&=q)85K6Za)~Rc5cfm`|59RTS;Om?|VY2jSYY
zpWp;$Isl^?1Iq~IZ`ITCh3Xc*PB=nh+l1?negfwZZSuIKhJ+K^<glb3#LkTm^Lk-0
z_Rq{beYbZS84S6|1xt+FQ7e!Aty~qW*MdvOcA?oi{1(i`^r5+@oRN-=RIZ8lCAM(|
zn*I>iF)K23gfROs^cSDNFqq5__@s!rl}{COfmq?oRfeu>(cWfu=dyw-R;yMwU}{^i
z1UuI?{H8!&=juKp574nnf1Y}O|Gd@l1_$1xdpE~-FAYB3%hHm0RvA0FP(+3c0#&UT
zu@ZzQXulFhMY61m9tT)&f;|?0EfSwxM0TPpTt9ts7<SXBHFVagtn(%`R)%@_(qxa3
z4hi+Bl1QJEQ2ntt{E%%}*s7$y=zd+9Ht5to;xjFN4t0#~+Vx!UJ@SG+%5eo)*M>4(
z>lD^67jcU)wYEX_F7#(mbT31%&oa9U#|V=R-Nqf7#d8%F{5ds*NuCz%BX3Rr3%YBm
zLin`nT~QQg=HY)f-F;9ES%Y*g?Dy$Cvfif5a-lJhe;qomCKra=@nO@pfRdcyEP_`%
z2p(NtX~6gGK$$1bfbWMjYrinud@-ESctO8ikKaPDE)R5PpN8-WfP{AfDjxc-S<!U>
zkwppahT)NZS^~UDmMZHPSdMf;9lF>LocGgqLFV$fFM2O5qGVQrONBdzKp$p#W?+%<
z{2)Yj*?ugY5=CcN2x5gdWVFj-i41g+xMh6QURM<2m)xE7?j<1I7_dMj?WMw>hsvWp
zBn{KDwtA1Y+Um=+VmA9hcV!1-Xj42+4b1&0iyYR_T)OA?rX9=)bc^XP1*=3EL=KZP
z*dz2~yQ1TX-OD)-my!Dz{#_xg=RujP`K8luZ&qfRe2Nn+_@ZFUgy8n2IiksaS%}3_
z!e`eVRJ#RlSC<Z$OV~KK{k_eel3)0JgR2`#2;{YJs9Q*MHkum`_G=ADFO<|HcFQ;#
z9bNuBHHO$ndmfULmKk0qXE&Z5Uyqk<#whp0nQIH>P_Bpjw567-&+8v6)q(-yXdPQH
zQ)=$tyf!^-M|1WUuswOOCS-9$n+|0^--x`-wPQFIPi{U3?=RA$==&4Z0aj2!9rONa
zm1v3k2;JEO44761VcS><I?mR}CU$-BSuMEs<*oU&2SAH9U*Sl%%zU`;Lqa+ruvxGy
zT>Jp=&?t;E;PXX@4}4v9-4Qp`Dp=me;yJ_qTf>LB+vI@N$>)Fnw%_f@FKxTRuXCV0
z&4DVc83whCS#mBCh0}SJl5Zjae-Vgm2aqENFDE}%bkoGp^x9OC-%da2UKE=@(a!6j
zwTQx4i;_MJW2`;yztNnlq6kNuN1);S$0fWiNT8&PgqZ?-rr$*BvvHrMPGaK?O<f{z
zHMlAuUYHcPiP`!CD2+fkD7Nv7CKdCMvyQzT0J$lHPjj%b$kk~*f=MVzU$gd7ZLpVp
z2Kq8q4YeYpWP{pR`PS_mV2Fx+$J#~9x=N+WepD&WeZ*rrK4_Dlc2k{{bs-ycQ}7~Z
z$wfF6&Y+nrgC{ASN_k6XgcL4qaKy@%tqvY-Euh{WXQ@I^%c=1oyQkMAE(?jT>Sn~X
z)>Az)I(E$JE;^#?7_keNWG0kN>=Rd8*)jucBnthx*Vs?EKrczJ92d)jZa(P&)Am&Q
zv=d(LbBL}zZf%mpJ(1y0XRm4HX%InaN#|c%sFgjQ4d%YGTa6I)&|Y~wrsI4|*xRA7
zTPX8xi|^Aa{q6Bj7pON2yLl}45lDW4yA4Pyrb=U3_NB@5ZOph6N{QhP*3R+v&Bkg2
zO8Kr|J0u8(up~ZNb1G^B#NPI<$;J8Cdij<fl6sxrSqJYrlZl>^{T$H(h0xRoud)aX
z8P)ym=0ig&)-l{J02*FkCOUjH2cXnb>wNo{2neJWPX^OcvFG7REc&!ZsoqWZz>A<+
z<eobpJg@T-L+jvZguYXmDm%ibow4F8{5V_f?~MG&1^A@{xDT7@F7?4b5g|5}2k1Wc
zJP$~=RZI!um92QozOD<2*puhnRDbr?5BEFx;1iduAG%2goWC!lLffL^Sb-VuiV9#j
zN$h4-SI%lMnU|Wkid|>$BN&;p-h13ZdOtw1f5m-ZoVBxU%Hn0ch(YLYf{Y9hVRogX
zt7K8iEY^kJ+Xa0MJx7kJ{;J{R=&W^S8CeMp2>6+Ay#?sKO8IeeVm>Pcu2S&-9&SCm
z2)EwRXeG}``MP@6w{Oj=+w+uAkG$b*_GIt7%BNyPz$-KHL;0=MaTzL(D^QZRwT9i!
zp_w8wcqygY$NM6W>=oZcx^0~pb258Rz}R<$YYRc)@tgv&2lIk{hFX2Z&d$IF)*fC=
zXzRJ>gNUWgSdQWMrYd;t{HM>EYC&zinx*Myn6cCKRHAE6#=#=d669RImry^brcD=7
z2f)p&_r+3<t=QjqhOK?#ndtF$Y##z$1Kn<{3stPF?E4kp!dcng2dp%Q^<cC4_J-zm
zdva7?ly_X^(OkLII~Bx}<JsLKicEjuGXFbhmr{&)+Ka=V4V?Z*P59stp@PCM>UFU-
z;l9Y>i#Gz#DJu{h{L-pEqUO_HBu2M6emsQo6Y2wHaJsQuVwT~C8-`$xLG6*1!o6~!
z*(8`#wIrgc=e?`5ipWrG+S`9fb2`Wr+p({Dy-;ils>%c3^qeD-9P7x$gJ$M2mA6#?
zOTQ&gj8z6IL^Jvp=AVTZ1x?M)06ZvjWaxnV5Ck^Z*}{wlif`Y(p5qv(O+2oU2#j9J
zhoD#3!MlJ`4<K_!S?HFH0<<_~j!!gRdhbSAvqSspaW0zL;DFMfTB1CfE739l>owd1
zl(?Hae|=L7&_Xl($x84ode?uIU^SlCK%;KDm_ljds?b{niAEB`_j+zusxkc6UtpUD
z<QdyjB^lrT{0+XO5mTO}<dM7<K;K|93LpeC%PcHVK`?nVcDLGX?7+D_7%8{!J_xZA
zc2G*&U*17kc><Ie^HuzxlAe3rRm5;8CQ)LIJFSomSUbot`e~(KoflH(SZFq`IY^u&
zvpM#VQ^tD0xK0ydUw;04@8d;4P_Z7tf+@>Phl2cRGftF43e;48#v1&G4=zEG8~fx-
z_CMqyU<Z~jkpTH_H-2eM<<tONLwy~X4ZV8K=l=vn{tu)qt1ig3;zEAo;Lm56LW-h{
zT!H}uLKw)IFy=*wtH4wI>Rtcsg8GZlkT>K79>cKs(3;_Yp>O}gt`I;9`k4AuMA-D#
zm;FEf_}_k~e}@qM_A>)dKp|pxp5OY}|C8??{QtRMKP8H0b6W1e%Ft&k{Tl2giLbXb
z)~rt)Y=5h^4L$uP=Mzt2rq)K<?TFuV8{zPazjpq|GYQaW7JR^d<s`&h97`-PSYuSe
z{GG1r&sXhv9Cn&#|AY6X|JZ3yiANyrf2aCjr#g?P4NL|S;EIU(vG(Ta+hotr$?g+m
zy54JF?VjJ({CVwKR1q@b8qwbL0R?4^q{*~HLn2XsRICrIFjipZw$s?#0)B3O^kBc(
zduMRe{zGzVq7|RNQa8eYk8X9ig@%eh2Nrg*9~r)ga1MJjDst8zK90cD2UcHOAiq5N
zE|eD>A8feT6L5%bXO29zRWmytQ)s}>|3cC!mHYCOV`+sw15v|JM)?XA+U;*m7b9Bm
zdn30=mo}mAGTAf3)E9XOD%cQDR(d$hoiU%kDYPD5^0L9;Q$jD@Pnf$aiX&JraXsMl
z*UZ*RVAVHP|Df@ocSA}xHCzPQkrmqlzb}F>d}M4z+k>O~^ah_O_1L)Nxx{*|iZ#j+
z9|IoGD|T%jv9AKb8}qAKdf>YB{2fSmzi~QS`WPiSyI`%DZoU-+v%AjaHV>SB`@#f%
z>HGO#U1s-Ky;q}QWW#5edSRYjz3&+;L>0$oe;$A{#a`UI8sgl~aKP5%;0qSStdrN)
zBOVN72(-umNTQ9hd=66}M4>I1&x)B{`)Xk(!K0Y<R~ct)a7D<MLoT2>fL|A>jm<gt
z%ug~J2E6Fwo#35<EXBPO?>}k{_xuvie~aV8#eyfNn(OoY`@-t(u~7iVTp?*gd56+2
zSFi)Lo6kdwxmcIGLGTQ$+3r$3yRg!!Cb{)-#Oe%IkfA5uAGj$rxy^`=$t=lxL7Mx~
z=v~C#0-TLi|C(mfE(kD+Ai8h}$>u|haoF2-vAsQ?*RJ5xR3BY>{?DH9-%poSW8^>_
zuhshY$ARd{Kp-=X0qw;)BKtwo8kT~^C&R@C?&I4nAZ6{<-gjQPT*z`71hwUenSlu<
z8Uwki+>Rh<ETlMbwxt7FubT#1tZXvUKm^)>JglgJTJ`Lnq>?Ra7sgAy-n1*Lv;S~N
zozOo9!29|by}OqO;W|~gfi%Z`pslg}P3Z@7M&<x$TE?zs<$grhG(sF9)yRe1v*zeo
z_pNsLvMZ=vK#IGT?K}KgMM)n3jdEyDG>ag*VGy!u23NLq^^N&#cM#YQC`vxCaG&)r
zOTM0|Q(#r|fB?Blm7Nhm`_6(}^<dAn*_G*H&-S_ZYeyX7-j6X@Vsh8Dm@?tYmk8*{
zJdH?N+^^1_iL$kl25{vL#42W00-?mtgMd_ux5B<8+hl^>E2`-HR2*lea<!e(1D6o*
z1xTkkDwG|5JA=k%8x%p*CU_MVkJt0c?OA%+XXUVV*FOA>f-)r=k=kVIKG+uebC;ml
zg3Bf&r6&|cSo*=ble9!!-)t)LYdQKvl#ly{7l;awW`8!uhw839^vsKzexr!)AzGO}
zcXWTfxCqEEiD9*hFhJygtCS9pzTumDb7Y}v$9@pxr*>eOEPw`(zalL?swtm!r08Pg
zNvrCe;YC|$yRHd+T@8EN0^v>KjQEqp?)uM{CN83eKT>Y8U&MgAAaxH>NNBmdGFvOE
zE25&SrB!<nEHijP@7a-EN?Gw=>Un05y-HHezqPsQWCIteMNtp7D}_r}-gSMX-yC0~
zyMRgVU!qRfU~G{JExwYz&QrsxvLyJG4xI3BGqQ>j5Sm~ArmxmhJM&~4pL=UF?|lAq
z72#KP5sJ7fiHTo-)DZI7Tn~f~3)z9HLp9)pK5(9@d-A((F#Y#Y$}E(%K~%>+(k0Jd
zLBG5ErLB4N%va6IRgegZ6NhI}Z{GISW-=Ut)7chEZT6*xr>3$<!X!IkV?8Zljb}Hg
zmK^%}j__p~`EJ&;&gqVBYhsGO-7exI@o^D<O>w07a}ibVJakd)vQl85_pGo0d>%uL
z$u;-cTg5RK+Q^k@j1_lXWD0zYsP9Z?Yd*ZO=lz0M)suG#h$q&N80NZ0H?Qf0`$l4T
zJ$%~b%`*M}e7?#P--TwWm7weWpI3STCqczKL_R`geh=VqLE8y+6=pL8nGC=**6XLQ
z!g1dR9yv?o(%ggFGsLk9PT*_qtROgje`Z)S&TzKbu!5XJRw?{aU61;@j-mJ3Z4`I+
zYye%mm!ZP<OBje#j{1@t{0OLl(O3E!*UzW7%kIoEujC&u2;8^m0j<5!ZU(P--IJ(&
zckJtI;Rhrv8}G;$ZC(S?A+?fGIQ^L~y8wwStJu74jmO$#(Z!-Dv~5zo-Nh=PROL9z
zo4jj%IU*-z1Ty19wt%UhhsS<S;by%B;O7*^d=emopcT6Myc;eo`#zh(hWj-gNG8*}
zFp;|CH&Bc&$uv7zgh&}(VBI8DYkKP2$35LAVlsTTu)FB}(`g#PXO1OcNqW%^clOV;
zU;B+b@1v?Z0mVAp?5Tkvl$%nJLdP*>k}Uag4|OkZf?P~rL$^dIX|a!V-hzM3kSkmO
z8)m7})%SMojm#KDn)Q2}xR|cW{&~F&vr6l)*r#@?o<o=9y1f#&%<jj{LgAqkkEwa+
zfeow4Hb`tKI+v93eGj<*h{qM6<?o~qF_pP(Vxb?e&1qa)7C#_i!&4NIRR19XBT+t*
zD=T_z*R(@aF5iY{R$m<bD3fm}@lED6i{{{}ntZIGQEgef_estxL2@p2L{V>g)q9gO
zYlMcw-Uk)(dY`K+`gP6y#_zNesPH9}@X9uqimvyGd>bchUEK@EP{vdg=hENSeV9LV
zGx1%~h}lcSdwPp94sW;Ads+X^T^=m&z=LRdSTKcutj?59$`XKu=8zGJ$P8NuC|j%Z
zsfIK;3Fs;~15BeG1EZQ9DwZE1Ss|{VzZ`_|ABv2^kcK-5ex5rJ%<3KRWcKRGSa0_%
zR$o}2X40UVQgD1QGX_gQ2rn93e12t<tzE_LromYeFhZl`!Veu?^afL;!F|cjqM%7o
zBMJTbd<2CzLw(u-3hP-ORh=!+c;Z33c8|D+)MS<$jMOZ#GP`VV0C#Cl_3e!)iC*r5
z>^&*E^ygC1(hq11HkYO@xldzh489{;G)meriaar6MKOL$jWbL2tz!_`@muv65R=Y~
zVx;;o9_R>V+)<Hh={`szn#4%sOXZMdWlC3264C2N$|$j;M3HJ@DO3_B7*kGCP0+JI
z5%M&vp!>@0#=Cv%$n{nMvtzjH7pP>wJSQUoqUnn>H;B*odIR1OJJ5Y)@H*cxTu(?8
zhby4Aff4CjYr7mL%OE+Bvw+Jq7u=%#K<MhK(Ww`)_>Cgzf1CxP{5p8bL%H-uXsC07
zw{n1X=mVb{<)k`S-({vvxUK>)<KwFUJRTFF<zU|i$|*6x(i2RJ{4{M9P-I?|oW6CZ
zMd~iNxQ}<Xy#-MF3#vB6oBc2ruIQtk314q!*_anaUi_BvQP1YYPn9CIh?AyKh#0Dm
z$oS`RzV&R7=E$xq;~X275X9&TyC+j_h^=nWG3FEa1ex`l>wF67(=;q+bAvu!$>c;s
z!Gk){|BdQV?qa^%QuCo;&v9`uW!Lc(m{d-dB1YypeMncthLf&wyhG5R`mj^uR$xYQ
zFPikBcD96~Aq*3jTQ4nuUq_UNuSC7Zk&BqH_<j&&)!zO_b(Cw)6r{9e<T~UotlY<C
zK>;m^(_4Rf|H>bcihcC6ore1>v6Mu4noD}PFwc<)_rnr3o=vO<C+{!SZZ*dw_4uyq
zj*zgxpD4LjWE=ML`%J<bPjq;p*{404PhV*ta8*xQ>%hClv;N=A)%tlaJN}Uc@NfOz
zsw+~OtyBGjhukLi1Knh2QJ)DpQd&(aim)zHJ^N0>8p20rK-n_37lmF{6)3zFVAz<}
z4nid>HMzZlWhZx_HgX{PtZ0eTb6@Ahx9w#H^(H_KDcD&%6<P{G*&X{qiMEl{u>$vn
zWq;X8I+4=uan#bIoLu+D3oo9k1)p4+2xv*UIl+0}zwGStg@eb6$8v!S=WGA44VWjS
zPGXV+wP|V(u`U|>Gkz~0-Lpp}UIfDqN*l0*bBaS6*}SKe=uuOQssR-4-!sJ7u_pXE
zBmp`}>E_xu&JoHf(Yi}1w1oCOlL<J3DjTa2n!muPbM|Aft)&=YWop1C-Kz;GVnPLj
zhS`rQJic%~AUdGITu{USm_zqqtsEv|J>}%rabE6LfazD(bx;wBcE)8CunL8?iP+xv
zBcT^2eqd#tZuizD^5l6@;#5$j{}g|cqe|Sk8qx4V`Tos<^;;R$EPkwnRCPMk>(7?u
zUxW(S=F}d}9T8?hW-qw^R5(dIof>e-5`yib3J<HJO+#C#)fs20OcDS=A6*IvVLUxg
zV{=r&YR=Dz=c>33(Mdd+*Bya}WD?;j>cp6T-xF&r`K&pDc|js7Izrmg@oMhx?<{cE
znO&voq$k@wQ3(5C`B~zl?`A$m=aeNbNzdxI<#*1^#jQo|ts)#3hVqr9Yv}Yj92@RD
z?&Or<&Q;mk%>K1VsoYk01Vvmk_0GaBQJ_K@5nFoHJn(dQHs8*;+D!0i-$SZ-j;j|l
z+lS2e?x?+1%N&3E;%+Fj`<P$<83lQ7B%*Ql@FH=W*?`|UxrX9ClA?$2w$D{>T?=k(
zJ+nkZ?1Rc3--Kqg*?mP%62H)h*CDdhvM|`}F6(hgUVF5B%2Vrd>q?k#v@6@^JHBIE
z!CIfD_kyJ;rR&4)ZuK4+?D+P{qY_`iJf)&8_l);X4Dg>I+rNG=iXvQaQT~U|3dt$i
z44P^nAqoi#q&{eQQL<6mQ2CsDze{R4Y9_?@>c;^h1{^o^C&m*jcGL{v-1}6enH6CK
z)C^C#S&j&9^>_N<th|4=FVG1%bs2jO{LCg#5AI4VIrYY*A9`#7C*#|Vvw}CmrG^r(
zHHxkofY>L?q!iT3N^*M79+#j}Gd~GlzohpizQmo&N#t^WVaIj7L-93Dfsr9WkFQE*
z^CHrbu8(j~=l2X^YfS(_yRLUOy2O&#>4N6OE85oNS%XyS4fD7XhIb8^VQxX(sd9r>
zvb8eD92518fafJ|UpG`L@CXL~ywO2IPd52Op|yeJX^2<s?2V>I*${FiYKAA=g4cgK
z4rA=$zYucZm|pK9inSUPv<-2(hmHiD8h1~lOGwbeaL>5nOgX{zj32l>Q9;izDLo~P
z7{i|pjWX&Z^`=c_-zT2ur3&7?taqPbUEg)wm$!`-Es*3c*_v~F!(2x*e6171#Ir!g
zrvv=0DYF3I7_VT<Wj)(lG3|_HyfqCZ1?l&HH3>$_ZqM8_D17LNOSrPutE-}wDZW<p
z>ZYM}yM;m7*y*j2jjGrZRK}cy?Ft0(N9;v7)obD<dX6NX@_P633q9`V!u8DZHdThD
zK+M$3#gRE`hDF|mwX?pbJ9KkL)T6_UX*NngOqf~ktamTIq)+rO-R39ijVG}sH$<km
zZ7PEjBU0N$7-&HAmp!LB^(w5-Q~;8MIH{*BB$h`Sux(}G*Gige=6-a8OK^9x&*t<r
zNB2!x(gkMWaj8Wv!6Sm@n)40He#=+URC+xLKS!>lQDmxa?FH^bQ~|tNX2Budvbv(4
zDsiz+FKeIIcHjO9NdDV#4ZZdK9#r4CEX7|@e^%Z1HOSbsmu{q!5gu;1g7{ldNP7O<
zW=lu&`jrV@;IJkzXcrwSGzVG3k>K4S{I$lr{p4#tyL&vSU2G$D7X={*G@mJBr7qrj
zRD-^tlSL--GRX4!keE@h+K{PUR&mVdbdzAj#q><;X9^uZv|?5AH4oU*PO*P?6uQBV
zy_bDQS^2}%ZJJH?H;!mMG6hMcd<e4%0}}1rh2SUCpNIM_(;}#+i0aC&gjjgAv@j;7
ziX!P$Wfz|L3+EX`TMS49aH8mk@%q6oEB?g=dKT3Y(9Yp*S??e^$cnJ!h<T1j=Rpw?
zCK$Uil%~~o0gX2JtcsS&v<Qy4nfN{S?!qS_CkIojT*^e_YL969_woxVBjRFiUN3O}
zST^ri3#Zsy8jewIdTf3TEiQfALa7S}A=a2?i*)1JgHM&dk4~Ma%?X;CXG5hWp+<&i
z%&I=A7xk}wR7_8<-5o<yY{nkx9U`)rNEQeSb!#!zS1y=I)}SG-DCC-AhrFfR8T2Yk
zkq>r3vL?jeEcMWcaokj0;ZCRX3SoU}o)wE(X)|UyUFWv%@XnppQOraPS#VFH%pJ!}
zl?sOM8d`m>+u`O@9%}7T1@qf<{{gZ7ZK7~E0dQcI73YWjKcJTTlVtAh0}YCJZYXw8
zxsQWh9(_eV(oFd!-TxUa@wEHOxaU3=%d@8}6L?NMK)3L<r`-O!E23cGD7#>(etMao
z7@HLGtb2#*9{U}mt4nje!IY6@OWluzUnSg#xSM6-SA#}dqHbkySGcmfH`J?FzE7w@
zkMq|(F^xRWQ~T7c&=-Iimz+<<#^6E_(G0ehQf2XxKr-%)b2X3bLN<_R`1%A@cGq`f
zIYq{YZ$D;EmwEoOf#fFb!kcwZt+>@*viuSwmR`z~NY>MJ?Ke{tPBLWDF?}z$CKc=^
zTwx-5E>A&)828Kz3hp!Fh5Xec;`wt^2T5=yWPCf3pzgpFx~*=*!eTQ%(vJ=PLeI`-
zhG+TxYS-7FJqYm#?(3_69;9N34q+Pkp^c|v<=j?e3_ErHB{p{?-DQQ8gay3%vF>yA
zy2(KW6xsPG5dbzumr~<nve(0E&<;f`Pb#swc{bsveIGyIowO#SXOY4TU~e7s8%6sp
zJxo=N4*GIDab;<ZuJyan*y(t$B5)t?rq|(tf)@$c0@9HFV#BBH#wwKQ_JirB)Ly!x
zB-yv|9wQofPP=lSzoS~l5-%EwN_{`T>-=?3#ZoguUR^0E13s#AYJBWB7yOD>wQ^b5
zZHi^sc4@43YTQkA>-jO6`|nC;8uj=!-(NaCD))I1jNSH#000qo2k8`u<FxB#l(I`z
zit17U87@N%FH;kKb+ZfOO-MAC<;$R_0{j{d=2?4|+^QQ*mf=ZWq#ye9Hmd0PWn_#S
zMfDsB{7X;O44FO^@qrAYho#!7zc9}QuR}uSlnJ48S3>DV`jy7a+x**WvkWc>mPPln
zkz}rxL}1F`6@qGJV<mi@0KS7i`n&BX1idN*c)3@P){3Q0%C~gxg@pU2sx*L=?Tr`H
zZ3JZtKor0`H4fS)gU{I;DPN1`CoX$|yRJ?7+<HChe$L2LL1u-QrWiwCT-c8`#UNK~
zc9^aSMHz^m>;Qxj<TtcSs(B;x;*E#<mpZTfc)?Sm3}52HQrj@rF@hpOxnwwvUrtea
zMF3AVyl@mr5z$<rYl~|Z%n@bKg5WaO8?)G+&r;Y@-0OL3Wx4J$sT~(}TSuSo(JF0?
zzix-ILhjVm#2IJy#C8w?rsrklgxoYckTw!>&h|%WrQLKvx#oTCZ$tTsr9q0$Xn08c
z&g$77eO7qK;wgj65o<Duup-r8di?@vt35Jd1C+LUC|kj25*sh97HT_XzI)8E3znT_
zXf{v0;eOTQAdQ4c>YnBJT521GL?7EP*Qk2sSeuR=#OAOQ7BnqT(M<AuN}<lZ*(559
zd-?iz<MMXpNLR4FeY>l=I$X&=2vKjeCp{cY)$T^MPZ26P*Xs3`>tn6pIKsM#;VXEj
zRR(<a{s!9(t~nP1)XwPZu3LtGP)}94#FR~J_7jsH>GvQtF+M1&fHO*_`_a+kH_uEN
ze3&3wR%Xzr^?%SEc!U(S=#ETh)_&FStskzW`FWs3^K75bG*zDMhqyEw=S8Qf@k5G^
zerzdS1+aV)y%zI%U*7K?V$y81#x@c<5!RhS7i%x;=*iy-bSHU5l={T;rnuiIg77Yk
z^Ec=LdgFWTD@QUCfvIXng<fAA9i`mPD#T&acRkVA(@PH-CO8{xg1>6&m_7GQ?8#dh
z7FdG2?uSC#r?}n(&?ZrFBp1k}vf$25(bZX^4Ot{YAA172b?kz`8mGj?7XNoeF}D^I
zbZ%LI^};&qQ&IYe#GS>gjZH|VAa~-fQ){sK+KpI!S+0~mtjVU6i1xKtBahpZ#t>IH
zI-#E!<5}vy%JjMiSLNPDi`y+2=k=~?X1d*abcwZGAs})o+~xQMNt^Q~)sWa<FGK8F
z@IC`7WXItQ^WcxF%(i1Ge8fJxegR&KVcvdj`Pd$?9#VP<nYOUBCfE4PB_Dr?2dYPd
zM=yQkW6pd*WqTXmd|G9H@y75&hXMY-eX;VRTn|X){)*sQ=ay%bsSDaB9sLl&)$rP@
z`N|NJj?VcydW2n|O6|bRKGSb~Njdcu^OFk+RNucj?ol~DXie6o%Gh+})j2DrjQY5n
z7QfGuI`gub&}OZx0jeMAW1idy2a>~n!XSZzIG>#G4DR+U$-obX#*na?4<p@Y=mTZ~
z<%+Zfs1C>8C_~WSLX*I69|C6@ll8t}wKf}f6^Ft!1JJH&*S1t$l8U>F$WFIP)@ACH
zvvHl8y^;35i?VW|HhY3ozvZ@5lH;=3guzwm!uO%i4;KBy#A!B07+#4C>l#i#S$2Kn
zXzU!+WK=P+IbCZ33z;ri?{P(*vzCq6rw(eBb}gfZo#%~pbIlmkH7g_b+DYuxC0J97
zZ%ahm$*Tpjnc4dSY|{VcS)?%-RnT05+o6xM7e3|Pi+BicG;>doC+a*n665l=^LZC)
zKYpaFEDVtB#4Rce-T~~xR`Q-ZmVw<9dfmO%f_kBGr`&4Ksq^d$r53>B%-=ASB}&fh
zdOp&rx5}{iARx~T?Y}kn6kh5oJMvPeNUx%HUhF^QT`c)q&Fq(0GHt_!itT?0W-I+3
zc-e=b8(BPs$I~p*SAnGt`a({1MV5VAacRPs3$fAk?qwM6oguyeq9@(c)dLqQS$&83
z^i~@cAJJ>hCLei9bWAcO)iEe>hKRLe*hb%&IpPkkCrUHdR`XuH$^oFX+!x#nVAcf>
zYHa0`;sy_3+^(|azs?g%ZF*{tAxYgw=ux}wEJa$E#~zflDYEb^2%K4Ve4JFoXPA;u
z6J=4xP}ER)fqP+I(AOr&dE@%mS4nNmN`_2!1?m$lbkajl+l{VL_S2HxgI)y@Svs1H
zXiX&8X?2P0DOAjizogl?u1)hY)w4`@Y?_6LrXiO_vslA8HRwdvDdKw$@7QmgfLbM3
zeptiwUT3{M&w{#mDNJ%u==)TrY4gD%>pil$&z9q83)4m<Xuqo4BnJ_pa}&JYU5oR5
z-VMCu3H;+VE}<ox=hyDM91eeQT34;IIdg~QO~o&}1&q*?ToI(PYA;eUa)f^5DKckX
z%^3nIvZAAkw2ZCi?PUgUbKg6|0$23emlSDRd>W+!_B2&txP<wH0WtZ;(yDBOT`Pb7
zRZ&3a-sq%{BKvp+Dn&6o5l^^)bG)yrVCkK&8>0o4|D|jcMasrcX&F_2NC570qTFXH
zpWGm2zicp8(`?h=&2mvx_7sE1d4(HD3<EFAO`+v49%QL8wiBkikRhc=T+)4oj~g|=
zLKgis*3d_1Ju0diG6O7iSj^6NU@z1JsN@3MmX=B3j#3PE?Y_*b%nCXRj53)q6KgwF
z%8$ybWxW86RVw3~>fjP@&wCxQY6qOl6(juyJ^@8J<0lNR9MtvFPl6P$Lb4m4GXI9!
zK5HF+Q$_N3_LE@wov7>xFmBJe)J$-Al_;8d#B;?o)Xr0~g>n-~RZKck6uOPQHm2f%
zCP?_FR@k6_1y6LWk50k`C{23L27FG6aV&%3eJ9lT${Zupl&C+`_AmC6(q{{G20Jt_
zNmE@s$|%g1sQ*obgB-naa!uG_T*Jy<E29L!8`otJ5;uvdM4r@<9C>yg2uwxP(Qo%n
zd@_MjU|(05(dK&dUACbID7ud`#~1z(7gMy?GxG+!UP9QtOjDK)sX3nZ&SK8O<++MX
zw|3h%k@uLb%(FgioW?j@+<b~t?_7jY*^KzvP`;1BdxADW!gZQdqMj`{MzM}hY8n}~
z{)Oz!YD%@4Pk(xBCwkj&?`rOIgd!ZO>C4IZeqsih@?*zEU#DtU16H#^(O`q5XcB&#
z(|+x>h1|On&K0CRO}i`?6Z82ow6^2rO5!ndc=sh!pFD}_z<Zgsz-O!blePdC&A$8?
zuX8K6x=Vt1g~dd2Aby@a(CVB^PE0c7lAA3Gm_4U^k(pC>#OY<wC83gS7aSrl{;|~5
zihb@tbk)Ocg>o&>z*pC7Fz33R<X)Cr<>yE%F+At6*jQv11*`0?jgi#f`NjQYuNaAL
z&mT<Q`m6H2MpiO?+&NodilxpLRJ`ZVUFQ2D1G{N&w}OoD=sH&b#N&J^#Tg3wAd1;?
zLH~oIgG^g?Y4d4>ux`UF#*w?V1>T!tjj{ud%@sTyi9w!)h~`*k*Pwq-ra4e_Y9b@=
zK0}nUuAcnunCcdeus$%T8^kn+_&T0rXp9S^O=}mtDSYtQmgS;whsWV;o(1+6pf(pQ
zN%uN<w||bf|IVQ_woe=0XP5ouQX7f*oScxsim}53AAHW%-vtZNRPviP#qfk6{UX5N
zMVlG|kE9osNj5?Ps@b6XWJG&C<%#<{P%=*{zd}+lrdWL2>JQ?>j@beUB`}<baqN_x
z8j0|?kf<KW4O46*p_!_aLMVG%5LIMr4xKQrWkj61%i&<L{-UR3Bxrt?L9ZVfM&rzW
zvF}{bILnUz^Dw41Wv3>GN-L!P9{F~3T{&bO*0=0bGT&tXf`C}EVxrzEWzlz&rZ`ha
zr%v^$SMseN?nVc$K?#)%at~*b5QJ@<O2s-0DNPaPFwP>fUcF39RsP(7Qn%7gfs6*V
z6NU4(7{NT@jVpA{IA!auk8^9-sPS>a?wd|HItKQHu)QlDg(r=I7TR0&B;lD>ObFzd
zQ#j^c4-X;_PMvRaLaQC$$bAG3RL&`XG+wxy?i(4yS|8$P&oSCBp={N=WX*E(9PT7T
zTOShK7@px3lqa`4j%FVCkvgf(tE2m_Uiqo|BWI}JgeH}}R^HRuT+4SUFo7M%hjF_Q
zQ(CO;+?imd8M~hcW}db<L$HJ;B3T1dsY884`jR#aUq7B@=_A316_O8L{?=@6N2u!w
zIS=)3W-^2F&N0w7<>x0&d;F|=*6#3i8TcG?p^7Dc#K+|d<BFQ4>ac;8ImKm0{OcFj
zNpZdm{m(IFbhAl81>)#LYsfw^&x~KVjMFn>*jWLhy6u3GT{3rl`Fia@^l&|++WMg>
z>X5bGK-b*scbHl$+|(RJ-14`6ikQxuY-nq_XFJupXVPVIS1{#{<f%N-eyT70a^%F0
z@a%(wT=|N`O;#4|{oz1~vJ?><+BUK<$>)}Cb2n*?Inp84Swapr-emrAwwbTTH{<M=
z5G-pV)31`lw|;E9c83ZdF5+gKpKCFtkzX`ChgI~#o|n_hwApVy+;y$#rBg%G^Z#9G
z_J<5CMF^1<2t`%>S!jIWL$RH@t9I&9Un3^3F~~W<-MLLfB^Xcnimex6RN5;a22$p{
zGrfPwZBb<^*=3^s9P7~5QdC@cRCzGLbr3}Ag2_yJs3XX0poSEWrB*l^V53ad^%!DO
zi#V+ndmB}3W#jO#Mk>L~&3T`3LpYxK!v&cfC@hh{hKV#Gg$oOc^2FrOBq=<2axLqZ
ziWvXa*doiby!Jd88q$pnJi3|eWhmNUgc7-gYTT?zIwFp!7l9nyyYeH0S-1m5^y_jk
z+h&jv2Vd~14{22N!3CR^i^gZut~6*lUMP9E8I9gT1@l~uqpTxuco*Ctqhzd3Bh@~8
zbxeh%j}1ThL{;!~J9^+jA8omU<%GddYDDnYP$hd*#8}`ds$3#}ihz;@t1}JEs3#Kr
zcW4ylXI?pz4-UVxv*1(AENVJo(;$b=Vp51kc90Ge1PNq~`SmqmZU!>rJbkerHT_sk
z%!!@Dl@rF!K|uNwqqy7@hqfPQZ)15W-<4S?J1@Rgf3~bR*6f^fAwXLO?WY#a7N1C&
zX!15VhEnrwh*yi^s>WFrY%cDst_mM6p`dzebb)?z`998TJHV2H$@myM;q}0@IlGgb
zo}9T1Y_5v8lHZuy`fnjgYtQ>Funz07g#_wMMLl$%D(}XuOXF0PmgXpefA*>6+9?g7
z+Ez^V3NY;--B9s5**f^K@VF;#RNX*>(>T{*@|rxa82F$SC7W4Muf0y1g;LKRo0z*|
zBe&rtIpJ<$WFYo7ei{|6Q|_UqP!M#=9&jK*FJfQ*TIt&o>@TJ<p$%SCHWXKT8n*qk
zJ1b4z=p1y8#MkJHEI92(p_HFwG*WZxQa^b-XVd@rW8&*z>ex$XI8Hx%9QHW9F+o6A
z@QiNgph6LUw0M8MQc>x_=%{z(d3=Y?R9nU3`Nf|b_hlW#pdjj8@h&p<nLg-r4-lr@
zd$9k&h}D^gzb)}xr?!!|-x#A^U6fSLy0J`}{>*4clJp>!gvXe@qc5@?mm+()9lbk4
ztl+bx=&LY8+QmuqcZ0JtVO-{tZkNiT!9NCP-w;66q|#;jjHgwU1)XiRCa32Yj;4|&
zp5#0mKpHQ*yUj$Fah(#PGQ<M-HibbU(NXtYF6-c=%Yz%IU*<qSiW#l&xQ9!v6)SG}
z0wV_nDu`{f?)-#N?qPxn;^4J9!UmQc%@W6P1sdrTW=j7H?@%HfnJ$5WT!ZQ;)nK+I
z_7PU23vCYZ#}p{EL+?(IYU2r58NRrp_*cj%Ta9g%$SBXs(-JumXP&+ABAQRhjG*=M
zj5IN7{058C;a?qJGT$5DS19q%?E-P#GrEs@Yp8Pc1s38bGdp?&n*HV-FRbuHgo@#p
zgR6xPUbwu|^GM+j*1f^Pzf-%ylN=a#)by;@se87ZhQXKF4*A!h>KLj?s`P$58V-s(
z!f{%A#y^GBRNT5nYb<E_7`YE4si4uW!o8!$5#+%5Urd<c8I&Sm$+;1FK1iHqt7PAH
zqylaxi?ZE4m0%*{uhh;h4i}E3GPO}?D>V+#&_B_(nfM;vfTh6+Hqa@K1>F_7_~j*~
z@+&#^sXEUnIc>tgcda`1flEhCc}s+vj??s&v&np>XIrxQ_ECSp{ZowWFH9!7Z&a9t
zJuXOGcP^1Kz?k2pJ%FTBnCfm$rsOR16bX)T?%O6i;S}8nw<|Jh^|i5@{4k}z7bEeP
zxLF?&haCHoi-|R&eL=MaM*eP`y(D)2!bfccnn*^XUXa{4aNc*LU|@aGxi7&+fkXH5
z^2@~r9iYgIDX|}R^mmLO9wZ-}#Gmv!+9b~JpGv{l$~dZ>W;8C;BQ@E{3yU!yS0cqF
zA}1`Eqs4(x%S8v9-+Z;`qyBDObWSQ&N{xf^tW#0!U&BdP*bZ$xbyOz*)|qotQ=ce+
z3wYz~5ISSBk3&Snrzt;P5GH%2B+K#gv-t$E4`q5?Dp$h^%HSV-Pu|aMiyRCQvo+kM
zKNo$1-qcvR@Ppp!)JJxW$MQd^YCJ2zl@%AvY=4hE!J_5*7~%&m{+8n|?egoEpQ`oY
z|2z~F8E(EuRffH=*}B1G_4v@gY_XaHvYM4{phZmS<Hx8=%1BDI0=OkhDgx=)N?ty)
z8M=#h6gm3sv!hsi6w{{<V<N_N#LlNFS`BhqNLo!HxI;wM^tQTeYp~iOT*ugvejUAs
zfn6h&Uaeq{IU(mfB~NH}HN|1Hs)$9=C{p|}yEK9JrH1Y9EZuW{k}v%JdS-u(a((cR
zPmT!@(r!9&{p(K~KwSc{(cb}qeh%=J+^WxerHPF*lwAAJo4-F(+b%McQHSABGIKN|
zV+j7iGUmgi8D&IrI###!gNdKd1My3~Gd*vk3hajmSR3)wJco*hDL>WyC?KC7ERF?^
z9;d$obL=S<#=x2;*5#6IP<aQMup^giPstIKC^mbFnSG|U8Yp;wK;QW{^Oi3e1@g%I
zjUE}^vz7VlLErKvM21%v*$sZ4XbjVc*qV%zYa_8uyBP5ugqN=iOK6WL|M{!RTTwsI
zW+n74iMRM)CI<h@p!&_kLnJ=F`D4fb`ZkZ{#kqy;2F~V7%zQ9RyGUV$%Zx9u?Y{}b
zt0z-Exc=oI|J|4NvsQjUSZ)1PKa>AozVyE?o4^0KCR73bJze_S{_B6gDF4IX{{P?q
z|BdJGM;8jSg7!}(Pk(|EsZgOXF@oDqkJx8#O-rZhera!}jt1m_73zKRc=?Y#_|MJw
zZy$tSfj68wVKM&K@pwaq5Hcex`E`q~A~ySqE5uiR50{@{V!&yeQow}zs&f9pN$@*!
zsK)S%Tu82tj$k?$13_V*Sf6if#fg}H@02@uW>D$*WA7?88F-CVw?O9Hofp~@SrEQb
zu?g%3@#J72=*h83{!Wzcfh3MQFskonF;36<nvO%|+L?wduu8u_Tl<u21vT|)YdoF(
z^gX@ycrg{2de3b)P!Nqw(-Y|YTy-jNOTFLj&i-FNy*%On`tc{~lo}zO<hF=E9|<+@
zp;t#5<Z5%kR2=(Em`}Sp$EIKsq?|D@K>E?JxgIg$<(vvo3E%nr?DSob1Sh(mR#<)>
zvl?d_84XYW_mEINQ|P6hs;6DNjrB->vb?*6Plq#d1(fsJ&%wHME?*eQ+)CG0%HnIV
zX;}n-Bx>6-b2V1)ZphV_vY1>~+MDqpcWwo-o<>w8tDFHuCLW-gD#$@r_E?~(uTK9B
zGyDBvP@~-RZ*`p?KxWYMcW@7Z58n*tqAujsiG;}b$KUng5|ev`W)%c>lrsD)#7B~d
zX#H|I-4?}lViS7C%Slo$$gq%W9OZPP=-__DhTbbR8FZ+q(E96&%JPf9T_#2(zmLJ8
zEnBJIhlu<do3Kd(Dqh;n{zKj-+tDxsN_mihUn6Y2`ITFKRm^$nDvjo?SgBZR%u}QD
z_Cu0CyV8PQKy=t$@MtjDo))}(5n#>ekQ2x}hgSM;?+i<W8p9Y8wT}{N2LK~<4WJ&9
zd?~G&lR-8HD3W^Oy62yqyI)^E?Po|mf9mx|?Q;Me#CooRQvX@`LYXQOqT!^iNENh9
zM{nfq0JgE~Ivg@EC&IQ;`DB0>YPj4j7V5G`7<;G7->KSO;5p;B{FVv)SaFz>8KTeX
zc7q2EGWnp5`lv%X4Ep>@^Wr+$1H=fzc#p&OWUjrDw-hwzKl)6g^o{#-{$->5Z%geb
z6t=3<wS0e`h^NjIf)^^VCX|6NKuY87b#e92!0vE7<yDJOJa+!|_g^pGE9jTIKKH;+
zk!3y?Y2bWJtQ!0YTWHOl)3CZdo2@G0mbPr2c+f*B?DDJkGB~{xc3IEVKLgq+=>-_W
za*2IFZ(ITdBD%<&mRucqLJP3Ys>$vw@i%;t+uuw>tYmk4)m3y{9run(AQ7@nbGikV
z#&|-qN5`1Yz_8|L5MOW+CQwj#O@Yd!{q!+a4zw@HdAa>mDmOjj#R@=*ufFb2S7nf!
zID?5Aj69Q-nVkc6aruO^Rvd8_+S$jCa4rG!I&fklV0cr6lz%%5UJT(z4;%wB0$lWu
zV$`Dl=Ax1BhvF-HuIw2F$M1XXlm%hWn`=+Of}mz!J}+?|A}-B9QTz<X?rDvEES4fh
zoi+g3^pRcanatPPBj5wM3yBA<ImY!jaqZOa3~F_5)T}a;{W`Nt=6XMOfPj4&G!{l*
z{g_5b$m!#_Z=tYMMMHO0J)p&!lj)HQVs?#_fd}ZyggY?0HVfklpM$%gVFHo!`?U_)
zvWLj*ZaMqxDyL~reja7G{6g!SE5BaJ4d8^tj_f{RBtGaQ!tKCF6<->b$si^M{d>__
zI*uBh|J&;L*P{uFqND$)6vWxbJd$4kIc*+6Za-`^OrA>f;v8sF@8}Z5){Xk?WTouC
z!-E*{H07B`fp9=wX-^QZx7RtjvIz(49Gc$Y@wo|2eV5zLv)F6{%zHSInx>6&$vew$
zl)Kv1vJwv2rzS($axTN*9r=*cBa&yO_kP~of>)M#j8eNU!a56E`1*R1l;+{jpkKQo
zR5FAoEg9c|bGj{$afo?)&@Zp+jnSni>8(s_woo{IWt~v%t$zGq{Q=o8lzq-)%}c=?
zMhYGsD^&a}ej;(HLh9&qf-wA~GBal2MyJB|B%}eTmD(t?FqY4h8_6G}8qAmy_lkYf
z7Es!x;ctFU(i&d7Y(clY_moT3u;}I`>D^8S&&bknAC+n(8+ZjI%DSvDRQ4dJWPN_H
z6-vAqaj)Qa#y=>=p+0>AAzYWiFvDL8`cr`gP{zjKvUCSM-uOOgsekEKg(os$*Cwow
zgFEqS8wcYyBYwF3c(_&@Q}N;LR<h*S`d!D_@@l(%Obsimj8O5`xC3hwQ4<MHHFLN>
z6Kld;KTu}b#R20;^@$0c+R}~W4YViy2_{kWrcU<=yskS|JN_s+yYEwn??@L|*PoUc
zjIf&<XT;|WmBJ5ot7Y)V&&a@VyA9hXD%$u#2)Hu?XZzhTKhi6$F#_opoyqUTPmSh%
zVF~|M;^kEJLxlNnOEvkgl(1Q>CG%<jJYj;b9hx9U<+82RlNy!GBU<J=a5dZo%|KH}
zb)M)0Y1QC6OI^oZ9?iTqZH|1^So`Fp5>b2|oEoQKM9XVrBsq`mE<nI-b|SJ9^`9w{
zgHWWb8})?z_TwJJC}Jah@&yvPgI(u%S?@9(#eiAU;KMLy4(0b;24%EM;D20O$iHL-
z5d9pBXqJ!C>ZCGwuppp5`<I|O(jcucakKLWkyfi?2ofOld9FX-?9&r&O;%9xs7-ey
znSoY4o2^LoNqPo(B%-1mf)n^I?9400_QSiyE3!e7Fb$n-1(x||z@YgY(M+^<-=FGK
z5%PQ&hq(NB5z|L70EG1}<hI1+tN?0SypNa~NK6D#Jf0g1b~0iia^&+-PLoajCAEDQ
zM{<t1fqGn%vZpqQo1aVJK~TY%s8P)!=i0cC%j|WM!QN2T+@18huTJ0a76@t6cSgO2
zt~dGjJSHk0yAtD1t)6g3&r)n3AU;AfqAsciMeps^Xt=~J+D~+{F*ihOa|+8LQfelf
zu~3gN(Y!#em8I#Ii@Vwt?zvggl+xlhxVEV^xbf}g8}u5bG^>5Q^REOK9&_f_I2F6)
z=RE_qD^+nCuS2-^_x0LhCU*~^!lQtLYVnc$l5&A4|5AU22c9-<>=Wc~kW7+Jn3qfs
zYh<C2bc@<MdB8z_PPy>#%v_`6^Wkb=UoRH4@o<3IRqc;>(o9pt7&#vn=UWoxO*OgB
zQD_^q#yq<ZhphRI<hn_P;aEIG548BK9%3k-(e0%&#qiTg71ghVG6ih}0FHkXT+7T>
zz3-w6k855u_NJ9%X*tJL$S|$T((`xc`ysMh6a+_hKRgmsW&fjb@#`aI)AOe!_vg`$
zsB4~wXic2#ZWT;Epp$*wEGYcyD3Y69I+xd)S+%K6k>M7vbRWrYPxsnbIIhYf>6jAI
z@**W0%=PB!hdIs%CmIzVHVK6KC~yjZR;`w2r#0qSBo#J^-XM+bI4dzWCqc}2|Gr<`
z0w9zk84vf{*cKk@UPm2$Oz(Y&z~7nRNNp{2Z?<)<nDxH;OjK=pUVa`bLn?>oun33j
zK$_6F0<A(2T&sqF<(#GzOw9rls&xWgEtk&r;3_cTaVb>(DocFexD2?jC$G;gA8}%Z
zPM94TXW9bb?01MX&t49gy@fh|Ly%Z?EM)&tFj3w0V4jEU*L@+DR-A`g@jNuE<??Jq
zRJXrub2Qxq?_(P`e}fNPq#5$39$JV7Dg-!D3cWg6z(8Q48jLY2q~_?gWhaX<-eTyE
z%(;Ig#HQy>M9pUw;;x*Epa)xz)(B0IaeRh4`Hg*kE~46u4{J`~Pih`O;)qfH(MKhE
z$h~IvU0P0<r>jaQ-#ngMyhI~O+U;XjrFL<S@6{x!+f!}vrm0R#sQhH?bv%*HnTBye
zBlIE8Ir374^Bw4d<EDi82Azibku4xq2DH(p-Gw$<G+%0ev1VcwhDh1<g)>2gA_~zj
ziM;z>h$U+E*Voo{`AiMF!6I=#gV8E)h$8}jZf;c_NTS)+^>*V3Ch1dkD^!hG(yB^)
zN4+=VTa8M7G!3Br_?jeex^p%&lPN7i9evj`YzY~R{8&+A%nkFkn*@hOLUOVdf)gAr
zSqpU(%e>+I+Xnd^2jMoK`#*6OKWg3-)=><dfXA7(rQS1IABwDhG--5&yk*VPa7w?Q
z%9>G0FMy@hQ!Z?qeD7tbMxsi$a$_SeIn+?5a8+a6KszAxPz6+T4H|yQZwU$1NVJm6
zTnk@YC$(sZ1iOJ5@HOVcgc9t%b$oAG<XieAnbk14cP%v<>=ADTkBx_Y)e!MNC`yt8
z05%y*i0}RI`0YgEk&WA@W|&=i)ssW*WD57`5=dD>Te}T&urvkRNP6D>%%y4T^Ia?Q
z<sv+{Z&3a8h-LJCh1(o5>qrZo&eU^ho_X)0swR+?j!7frQ`499JoEWVm>YxBj$#%I
z*6a4w>Acljk`v?DM*jDbFW!BXoz)6Y(<AUk#~7PpzU43Y^vX=+-TSc}r9M06E3r^|
zBt)h#XZn4K*z>Q5)W-=4a5y8_6B7E}_M=Hhd5VHbSgTvoWe&8KM5OvODULdvlg`BB
z%*hptAMeG*)w13XY~GHx9l*G$ci@Rk6$R|e-b~5F4DQN5yiJ=lkS%=uRvXsXkeukD
z$AeYtMxi;^4#{xU!aKi=t6OPraV@-G8rGQ2dv*C%n}v&fQm$b}jhsNzksp;v5~7Av
z{#%9XYFTt|hNGm5dPKuKUeO0&3Xl96XRafu2pZETa6UVyUzbkK@^m2V;nqsR-*!7G
zCB$P>kN$~hr8JO@nM%pt%b<HzkRXUHXpuEgp;AgWyV1ySIkcPYUclocz6+DZ*hYMB
zDApj$zXq)!Q<yw`fjCC(215Z3ZQ2z6DQ*{x>8orK4>xPSwsJQWVx#eFuN6nlj%?;+
z;|^qcmw}QbzZS!#8r&jW)mxbRHYQ%7wUy2T0s-TKFA30{=z;FwhbPyn&n<$dG{Ywy
zYY2|d`V2WW$G_v#Hlp5JhB{4?()=CCCAF-J@$VXh2m>!^Wt}C-pz=%#5Eis%angPy
zp}?7jW7m)kY)EtH71bXok!O17)W6d7f3f%8@mTNw|9C`JaY+{{xrmUJk?c*95m}d2
zMif#;vNxGU+1Zzr9kMlSG7{M$BO@YvkMHAkPM!Dpyx-^D=l8qazJGrIIp;d+*13AU
zp5yVjKkmaohb+@Vv1Y9EC;@j(XFxq6J=vR35lMz;H7?JzWo4Irt4MteRy24bw|Cey
zM|7I>sBX;Wg$PTk5e15MXQh``A37N;b&YGFBFjv?0lkrFrsC=-?Yt+@iO!;)6SHa9
z`|h_spCk%Vrv<Js&aTWLRR}dM-n+Ry2=~d8isZUJRWs^kf**WC6%NZ?H7TnBlA?i;
zjD9O*O}p#0AK;DEE}7H)z9L#ARvXa~TjPxT`C^w?Y@pMOr#+%QW3=0{|Kg5B8}2jM
z$%Z?6)L*96iG(8IopXM3+8qV)d@nO@?3mB$*}B#5{hyEl+_EKooqfI~x_&FQf@1CI
z1?v;+mlkLceq{jZj*%fbu4Zp{+jVA4a5lMOwOGuEk(;y5?0Boj5UIX1N(y~~;kHIh
z`7m(Q)NWBf=86<nANcw{f<HQwJ62_cQaN1CQz`*JHM}BL*^ENT9p#jjJscTMC8HUt
z;!v4zIcAQTWb_Jbz5Ld$=nRLZNQ&|O(`tcL6u2>Eys^`Pt^wqy^cWS(+rwlCJ^XV7
z&?9@2xMPz<!>8yyHjH0AA~QCWV9cwcFq3j4opnJ-NdBiG1XK8A464SQ6t&TvXE^A~
zAfaDXiMM9xkePHj`p@StXXrjS{4rfArpQtB^^`aa)=OdxARrYYdQ8qe(qy2D_)Yr;
z*Sm;!l$wu+*NwlhNH%r+8EsX~6h;7uDA5nJGA5vAXMr5aoQ!O-Z(W%h{@dxtQZf1R
z{?0f|GLw!H;Z;7zc>{G07hH{my1LJ_f~$JzeO!-|oMHcaz**%9b(U6puTQmNzE$3z
zW4A=}-qeec8rlpd-I2>(DbM(rrZ+@cj5A%8*)>pe478h04Y(YC!CG!;r(*hJTJUjs
z!(+jujdJyZ=*~K0T}J+gY6X57-Qw!i&%VR8Ypj^XtUAOwD5TAM@!CXY)CCI|p@@MV
zzcW=f;NwlesaPOSgPLbG;Mvx}i7Z-Kptv#8zKDIvE{8@ZH9-r)4r0njWiF3S3aY9K
zjS_eY$h5z#oEK4Fe}d~l%gghmTG+{7Y^&`qKt;#R94#h~TQc6cPJsIm2HJc*O&bn&
zpAUg*T#f@fYJx8;ESlXtGMe<e?4D+bE*9OtSuwJ+5#L(lPbNRq(dhD{DppfxwS+$F
zW5=3(q?f=-6H{1`0M!pNz;d=KjlSq+Yv+EyEWh=(Z{d^6>Wpp@RIzDiE?M3Vh@s}X
zzq7Hp#*K?URA2ml!fRj49x6KB9mJ41Hv913o74aN%)tURoOY|n&Fnuh^DIhlK+A?D
zd;~2z<+@H8-fCmq8w8~4D;T8lRO{cu?JH~*@5U{)thpt)+R9m2b|^4tR}<YS4!=34
z<@g#m&#Ll$9egsI#n(TZnnX8)%@*L>T|UuLnsHni*+$1j@82f$h7w?i)v~w=1}rmQ
zsmO)DhO0?q>7uLcx`SgftPH8rrSBC0NOvegv)@4zPlK*Q<?1dT-Xzq<6JQW`%)!~x
zBosavbUjTC((bCyma6l%P1Rny(ZW^Q`ktS}*x=92mhx+}p&UHwC^w?Ecc95y(2eq|
zqDVJ&%i6w06(b0BJNtR79!~X^IzG!THY(wpS!e2OP}bSALpeXhQHM=DMPcK}bB*fW
zSA009?9d`eFR6O#%iA|n&oZd5$zp25s1q)g#=6?ynxIwuXOQyeQ&)2!Fscsw*Bx}8
zq{`hSj2;}1S7%+>jh08WxxfE_yq#suX>;!DuZq9I1;XE3Vske{63uL;l(g|iW~jO5
zNLKJ{tNZNAw@>^%N?b#7b{*N#-F%bi`v<vQwAe<#Jw?a3l<Us7&h#X`E&fuMtwjm0
zQ=d;{$xGf{neMPy^fx%x%$LK7meX?|Z9wJuk{y|T0VXKr9~<Y&Gq^7$>gyJZ#1t45
z<7j4Zf|+}2N+AtbxHz3@eWAlyK{^GBJLrxtfaXR!f`6xz4cU>4n|{m&ju+dQ$Xk@H
zgY^QGH+lBX;X}FbIVFpYjYFdB3AufT{`_*98qFIx-Qq-2G$)A+mb0+j8Uh@3tfM%Z
zxVJ!&o_wy8J3KIuVIuYe>@5}$4~m35$5EUP;eJx^5st2QOe(&We)8ypKL2M8B`q&M
zGR@3SUhk_sH8yfbCzsa6vP0m9$&<d2W$sXt-_3u#2(tXVG!@Ts`u(aHQW)?NFouW{
zpTMoZ262SLRCEvBF1bNZV7bMQzn!g~_P`N3+M6oLD=H2uGnuP}P5m-nhf=%1vOj!L
z*CD{t<WjLu1n<NBLv`BfgdUcwjdjE`&7HDe2oBEJ0Cf4BmWa@!OuO_!QpGf|VyqSx
zNqc7=^pQ=D9z{2IivN(5iG5&;x2X1fY3Ez*hYp1%1w(mP<e+CR2$X!6f&W|xSM+Ra
z9H63t%@=uIZx9f95Ri);$fn+ZOL7;eJ2H0xv|jGI8u2buw><*2C-=wuco_7r_n|=N
zeZ9P$f^<iWs#`jA2M{r5OD>_8?j%f<Rwvg&;lvbO>#C#8HLNTl>KKBS-kR$TV1#y)
zWZsyWlss$kVw3NqM3j369uET8IaX}d`=PGe!`S1h>EQKij8h2wT?%Xurx0qc=B*8I
zu$;`SA4|X~+<4M;EumQE_5ZRvCIDj1w)@H%{_DPHI-qm`5c7quhk1f9L#607#?y8*
za(^KoLd|j-C~(O#@_J76+O`Q*6x54jdirG9S4zr&ztn;lBfkdhSqJ5ndSv4gMFAdx
z#@+!X-~29qzyy8(j<38O9ySH`2O3m8)y+o;JOD(aD2_aIs4PH|ct;K6IQOVOhgSMn
z3OC^!#Llrcz&VgEq5G6W>doj)hALgp-PMQuokX?J?yI!gAQ;64L*GlMNqB_(<iLY=
zzZl};UwF8OsF#gs90wzI<I?5eN|5?1`vtk35U{+VJ-g(>6V4XwMdF@5e>``7CaIz(
zR;K3?|BWWptC0W+@&|AwEn9#AZ9xUnNM@RJIh2%wR)27#K9+<G^}K9x>QU1{PDW-2
zTOsok$+pXve9W+eYzt&*SDg&rF4sOJ!paW9tit**A)y-g5zatEx5fHYEA#SJiqO0J
zVl~199>%Ux;Z!c!22+h7rTA(MV)l}rDj$xu^!zm<1<pD<T0vChcw#MPL2Qi&d#b@B
ziDM&o;}n<xT!U!uxlMz1=@Fe5s;36;8h_h>OT5$(FK~+SyX;$3!)W4xSv4MJffC=%
zN&6#<jKgM61@F!yoVyDbYKLh))A@%V-L8fUB4W|-V^n<CweL&gkAH2({}^Q<v*(oa
z#5sPJ)BBv}|Efh8@h!_$kik)m;r%!<&?56WFWu+}`-a05rNfXa^Qp;6ooovnCU^oM
z%P0h826tXo_BE<vS{8P4<#erC0~|hcr8W%?hbf(a>sEQ_WiVCoQD(!6dygNT5)JkU
zr$J|93)qvw9lyBi*mw>US>R;YzI`sY*Na+VDw`&9LsApc9cFuQCK9?Ox8pLpF%XyV
zdb-g~y&xgumO6OhFtgW0Xy@t$>9H5NY5Ex{uhv!=W_M%yVQ=Tq@A*`j1A0d7;QT#I
zH8DF@O_P9eX6=e*AOm2m#j8hV-WgTn3xj!#9M{|>@Mx8n><iJOR~8SUT8XQ$DwdEJ
zMJ$c_fq2P9VI%cL5_E=q0G_Fbc;V*V2px_*5CVY9yB~?_9iN6~<ybM_)Vsv>^58(Z
z`umwHObU-8-@0!<&s!EpbdJwpj)B83%?a~DH@qdi6#61q5*Fqo&5r@ZBVis_m7;Ja
zy3PQ~$@@FuxL;~Gvvy-F%`_T9kDHEP<f7F}#)ZT2)ISqozcLGuUdX4Sl)P8gx3{mi
zV4S~FB8HE*RW}+<o-ZC}4SR-tE-<rCbIZ3P`dI5b#ABF~Wc)O$$<ZTYq_0QQ;R+X=
zAy0tbs@Ua{m8+X4gub|w?EA|Lz%i&IESmh3NMOYdb*S<F)1Z43Jd&EsW>$m^DUumj
zeN8W`xwS^5et9Zj6&=hzwdw5`0%IkyZtd^!?88*(?esCe#3Jj#p0(@6bevd6-ggD>
zOej8K&Jg$TB(x`r3TjYiI;{|{Vf?_#DMYVVWvJzzPeHIp&C{&Ychot&*5=d3+|_lK
z`R)#jxgHzeLiAwaDpyS_Wg492j9$v^f|C=oFj6)6Kg9WHR7=cw-u#G>ZGtk&{D+9s
z^;%aoz=3Wj0h}#Z^z7N%*vHo6|I6Wc9>^BxBjRK%KasXp0thKOu5M+CZ|oy^m|jds
zCH#5Mta9jc?*5(kHy}O-<xqMP$Llx_pslj|`G5>OlQcJpvC63kAsL(QO2@m-VCJD5
znL$mM`^Yl&N*C@)Mz#o|H6p|vBi7w9-rLz+V>IAej=e>(&R8O=7GjH|&iD#LM*;O?
zV_k(|K}_8)6x=GGi736V%b5w{?~jac0Y0Z(MpG75;TJdx6!L-{#&GYAqb3K54>Bri
zuKU;lK>L7nC!B<U38`JXwYW~j(Nf|FR?_s&o%hxQ_ru&bD47SV8g$W_52k}EU|*cI
z@}%YWu=SQS%(D;NWGJeO_4{~mZkbHMPW;FG6o-+|0v<k9{yphjv1agEetdMw>&(C*
z#tbslP2Alb#-P12=<g<8FMr_oZh-fn(0ncTDzwO+(p{qsrtIEHQUpY!mU#t1H$O_=
z8i(2XMEhxNdc{KJ?8s|sBQ(rMe~5h9#FM6y`@Dxb_Ce;719a}s8VqgP?zL;LZ+dUO
zywgEwWi2rZ_APwU&y@550y*7|#L)K{S@$DOk=O05+wc2jyajyIM}XXwVQm?Du>tx+
z#&7frqu_{A1TehN%yb=LlDSIC?f1>{2leh87xqrpD~&)xm*o6WZK*0w^j2R13%`3>
zbwBLXjx#fF$?oQa>!UZe9}<Ku-d9_}7POG$u965|PaPTd!gvyR5{&XGS)RMUxd?oj
z-^{DjPQaDR@zBGF>nFb_1s~6CdX+39{D8Kd59Y10Fgf+p9Xu0W?1DWWCIGP;afXqc
ziW{N#5QIs6yQA}s(xJN1DeyhRN8d9N4wSdX4s;zq?X4_56ct>NL502K>mZ8#;jJYT
zh|ied)6I@ukyX>FV#%>{uD*JT^X(>~@qE%g@nbABe@oOA(q)8;Ue{@ftYkK1<My09
zg@qK}N%tLi%)m)5{n_h6rjia-u&~m->Z#8TbkR)*9Rk9SQBzbGqok9vLn`FYOH=01
z$|xUfk&BZ<&&TDP!YrZa)b6)rCR7}(b$W}58oM(&gXIUvJWyBt{DpS$co_*!<~Nkp
zQ|mu73o8$`jj{M(oM0qDK={TYZf75VoXYy}10iS4LrOZ5q&sv$Hy&+%pvcKZ^euX+
zv@x{F+*Jc-8(6k5jAiB(_vW#QQ$(EZSuNP!$Q%XY+5`t(W=>coyR^6)-|O>PXbj#J
zR>SZl+Lwmg;MSN#=#}qND)ASJ-K*}ppVyZ_ts3uRF5HgQeb*M<Gqg;$V?g%^YgIra
zM`GJ>Khn-XFYu!A<37v+x$lL|g9AQq@LY=}jlv9F=ZbgLGne5A(R?8G6NUdT7%3YG
z93fG1g84^(wIk&O3=e#@B+HikD|N81jWg)*Hy9`y@<XRM4{NI`F53Gu6aqpu*@H4E
z=TY)rR0@hdNO1bY#OWkIDFPp@2Q#t<qS+D`j&j$YD;vSDm{!Ty7Lq)QuUmIpNV0RZ
z_B1!asLd={>h(7+tDRIg%9_D)D?Fj&ck2<Bs#-1ypgy^kVNt9<Pr4oAky>Fn4X}xP
zr0eV5E&AXFN!8W>xVPNU{w#8DnnB@|L*!?ubly`xvAL6<cWAL2G?bhQbhdIe<uuW!
za%jM97+in!4*@*%=I|+zqWsnkO?Bc>T3(qXb-{6BRC!G^8<?{@@XLzfCtwygqG~=R
zV7|RMz>%G=DrPexbsWetTWy>=$2QEoYxRHs{mm4Bc2hWe^+Mp!a)$kAteA5lG;58V
zEz$;6sN@5FO8KGF_&H29mS374GZE%Fj2?0ZA0GQsr6uBCPn0L3h|x79(J`Aqe1W_<
z%xOm0sqUOSo<)I&ovSTdA_}}@U!Jh`V#-+v??adkclX)*@<|vQ#~q+HkmRhD+aRM7
zu%SDHi~-MEb~PF7^kk^l4ylT6B})R~AsuFBcHqFFccSdv7vOg&nV(b27CQG248Olu
zVhjjOV1oa3{MX_+MZ5z%27H>FT>a8$AU0?Pu*bNaVvKcz3Fy^TKSD|3SOy*>zou%5
z?k~=8)pdgCbJw{MrGgv4hG&^ozZ0onhm+q7*f__B3isid>nyR>M@D@br=?jB-R^uA
zB(4v~gXZx6vb-6*?HHw?zgb}(l>T+-sihN)+`k89TP7dlyzh1^*i}BtyQHD%B=hrk
zcv8i}^b;-!QBQ@94+{tVfb}PljeZN$R&8od?1<~k%TI78>O2ye$sk<}{j2u;=MUa?
z4DbzvS56Q9=FI)gFR9%}j^j#hGvEK^W&Hic{OxnIH}DN7K8D=>>v{hc-upXq6Z!n_
ze2stpOa7PB1%5Ld5puxDo+~`^`#tjCe)T{0&i}uskV1#x#pG|AF8upn?El+Sh&%-=
zyg5&)pW){OzA5Z)biAl92hE<&Kx-E@(GsHsGz~Mb)8zQ1U8t+2;x(R?74%B%D1*1j
z(GgL3-fuDH{d836{pWJf{~h6$T4jdCvie=H=hxuEgo>8mJah+4lRC3?UiLYo`475D
z?LnZyRV^&!bUgFe-6{ko#EY@Q#MXbK%;|SNsU)K82SimR;~P8Dm1XLe=2eE?3vJ=M
zYJ4Vo14NIH&nQJdE09B<6uiSptD!Di(4f_6dfKuwE{WihQ9W~&WAMzE9|#a+bZ#X;
z=*O}86I7pxcEy|Q{|gBe7Q(-8Ikgf37-7YeVhjvTAym_Bvr%%75tvEAA<f&wAr24(
z60ieQU}wn)IT<%GkbEVRQKu5l@<88KzS60b4vi8^T^4{X5HxB4X9`;Rgv-MsV84Zf
zQdI8HkrT(BJ}*4Ys9e{$tfgaEoN^kH72Sm)qWUWsoW7d!-1^1}-Px-I+lJ$W6Erbq
z(*Ds+vddHLG0^1<BuGV79g98%xOw^OdfeXgV^#)d#94N%vznKm^#*0-xSowhJLaCP
zzktqGy>VRiE%wJT#r4ME*A#2By;scJlM-h}%KmsN6NMDdMSNKQ+cHLeo&p<roag5T
zf;r?*VqISMQ6I4#VVy#NPND&Wep+9V(dK1!Eil%nM-MkiC9n$@g&B-zS+YX5ZezDV
z&M)t>GR=rE$)HL4^kn6GzlL7P?c>p}W#~3!MjT-t!m9m()Zb6eYZo#ctf-Gcg4J^P
zvGOz|O;~z&k&=EC!r>w$+_#z_aP{%_X!tQ(-3K>xt#L*qw!kN0k&sWn?m8#n3{Hok
zxhHkhPiW-WMy<pKc1|*XVtJ#=tXn!0t#i~$#lgzntUHF2KZ2`4RNia_c`n1C59EVE
zm?mqZ`wO;)SgDB?GmNqW!Ij_bTPe8S8;yS*f@f3@m~de!Wsw~934jM8e91+c@{Fq~
zaL;52Onx!o_yGKA$<EsIM%^8L?TjB{p-8aKOCL>6W~)MGC9Y%W5OwdzFf{Gop2$Tl
zHIIWWLpXHaFlhR64j7WnvvCGNMr{T9rFRNox>DGx^={_gKA0RtAVF$I-q&Fxgxh}R
zrEGnfl^1c1H7jd49jj`VBaq(rAuEGQ*ZjFbsAF*RbQrw-I1;UQqPFW{CZRz*&9;M-
z(X8XY`^%<eA;nC0oi6>)bD)ml0EazlM>xcqw~&1x4b|}mHuM>$1rF*!e=?e8F#5d+
z-$#T&q1=|xL%8r|;b6B`U-*iMx>=P{eo8ae>T-UZ*3~nGIY(QaX|5jMmG=JfWg~lb
zZ_$x#PlO}pKEYtU)o->dIs`5V4sFj@#eXi|?gS}1Z;U~IHDCOc^I1}k+@dErKqB=q
zeT&EK4j-%#-f{T`VujgiwgA?}OhN%<qd;U4<*Ed+zt-E`HGPhg+8x6X%^G*XT2JqU
zBOZqv;KcW0gv0#ZLKO_vJUzWCO?S-5-tF&I7N0si%Hj-OxlzytJv~zghfP1d{19dG
z*>fv+PMms0m%u@XMtXO!^zQZ4KG@+}Gu_!bj1u~CpTqjpu`Ja!U!u)gqWftV!#MKm
z%6#JA8C3~wBM0R8aZ}pQ)m*ebyGLRwt9ehjJx6(8&!_vUt2NG!^>jSts9feoDvlgD
z7Zi9VtKA+7^fbk49i1g;t}UzT@E3sv#@5>=8`H2#)l&)*h7UO2=Kvas>tiSbcxNUz
z;s0@myP9{>E;m`du_RmLbDEXuQL7_s6j(QTeI5ngePd~A`(sx3`{*22Db`4w%H4_9
zd4Rn4fmgzY!c`F~)!*m&ks`x)mS}~y=SJD&J$FpckEU1Qd%(SWXPCjU7Vb6~uwQ;v
zC~R5P`$A>;*GUv@+TAMrStnrd-QM$m=-)g;Ve_z%qQN@g9fp_sV;;tHr(-yGA!$^|
z=#l@NHA3v>wcjk;oX#8K?bC_%b$XWp=cv+Am(U7Nf)?rLdz<unP5Vp{H;!|%T8jF#
zFaNrA{&D?YMk<N)k7PN&Dv7U(43DZP^8R2^QH`VuxhX%88V734;awX#?K8w22pEAt
zUvA(d!P4rLp9Yr|OxcLMQ}FRaq0>zl--F*Bd(t41#Id&N1u)huEI!wRt?um_LC3^d
z+6mLwEo9uz$Wi2IAwyaAq-BELadE7k&uPKP%Jh|aTkENL09z*@FXa6CduUUrK^>k$
zXIn?<6V_bK<F;X=6hg6V(~g*qH$RWRb3&N~vEx1rmfLb1=sV_tf!ma#q%F+YGIONx
zWZ6L6Iw#-&C|<<l#rhsP@)hxIEa(O+#q%8$equBRHqw`w@on2741x;*<615_(22Vg
zK?E>IPdn<Q6xJCsh@Q^}Bpy4xz%f}yYISgI>`{>N=Hy9=(lG$7^F2TX<4(BD8PRku
z$N_&W7RJkDlQj4QoUviRn)K9*Dxik#GnmQiy$$S~=>AtXTaE~RHkxJX$u_Qi@E$Jg
zvjAy^abokU_ITO_L05^yYJ7bbp<1{ZpQ+*>I#(3jP<F{O*ZUUH>>f;ttfF;*7Hq1&
z%{#@T0D)4Mgl96+p244$TI}z27o^hoi_Ebm<ik+%vzQ=8Fa%c9NZ3-e#ps>AoL3*7
z0QbIQjn(r33eRM}489s+yYxL0j*PZ=!5GBp=KdFVM`7?f&!Jn{hLfKa<{`<FWM_*c
zpP0mHAAb|bC+X8py)E)Vv>qH6%m9;Elkk!7)iPP(;9N3xGajnZ8wS{^A38M-;tuv~
z$*S|s611hK@+f!(uX*k|ltU42eMGPDMo_K_s?sE}dVql3sc|U)x|R2ED_UPOZoLaG
zf(VGDjXG5u#oNQzYSt>9ql1v+zi0Q$bnT|anx1}t|3*v5`adq}pFd2&B2_lEbF|XW
zCV&tX$h6>_l6(D~2aAVP2kNS{ffvPW{;g<^k+cOkGR`3hDelJ}R9Qn9nLvvY6cjXW
zQVju{7pw=Ga&%vH7P`VPVglS`UxjTVb1@QrhC))ZU(7b+Qkij_Td3o4soVJKNo1%K
zJCFy<x8BIj{e1(9A$^e|ik`a++`zlfBWZxk-C{0Y#32i*@Q^G}DuEi@BguPn16p7y
z$VB!04b(|2<fPk8<fF`fk@-e3cLTS0C1UR$i&o5++C<8?g!=;)ET0h_6bPyGF7z24
zd{rj^KqZy2by3Gf7UBWm1zxQo<oOO#E3v(ear5Kkrj+hZU|CBTZ-!GQ_GM7>VIZNV
zVr*@qCZ*g!+Z&i!Dt+dCeD;gmIO9Ecq`iAzK#l0_Gp+U&;S8;FLeZKUNS*%#joV2o
z_wBV+nls4UJQt7u=7;mVbG(goObFBrW94h$rF0tAXVmr@j8zf7?1fT&;KXCs85Vre
zR{`RXCO45Lrtyt#AU!o`fqrQFxJ&OKHMW<L0o3TufvXhTM3ZcF!;o(f{uvmz4NO?G
z*6t*6S3S=%cOq9pKSk6xnT68Xyh^>jK;MZAIRzfPR5@<|lCv%eK>WQ6!Oc?lUFsbq
z$)U}XY4SS)tF_Cn5ZUs`z2mO?2EcShMVV>j=4jH{DANST;GaaloF5D`o?w=xOcz#+
zXGOghb5`SdU+f=P*yqW+j%KSw&4I|_Ow{aKFO#D);#CkPZUYIR`I`sGv1tsq8u##x
zE52-T4mUCwDTojo6ugt>Ia#zuw0M4ae__wxQJ_^Yka<zL1m&>LHW_=V@KpcOnD~Fc
z$f4IdfwwDF|Er8*R#PEHoc5D~7msGn0V(nB#>}ebNQ{mLd%~h-)mY}oT8+mS{cma}
zjAxz{zCtK~p5J8BtePsm)lq{%&l>pa^^iMQ>)1%lJou!ME$A%yvL`D?xfm!T5C&n<
zo=M1iC6~&DU2jp->B;?e&(!EUoD{`lQ(=!B=!;KP*N_??4<N)M8A3@Fw^lT7k7RZi
z7qMpwT)HAJo219T0^!RXSL($q)@@ULn0`cRs49l&kUhR|QEMbBSYepGWCe_sUuzna
zL}V;8alXvhr3^5`ZU%-RgUe$lUdnKuoWff1u;xL;*z3aJN>#6Pz09SqnQIM8`R19o
z+A5*r;CP1QT;&dDyb3lJD>l+$d>H&h3?1#l?z5;V6>(oOs30-gI5dR0=LA~UnSDB2
zjjWY7Cvo~ZT1BS}6RJj;lwL2Unfjy@E(KTiqsnJAyo+Ao{~)!(>rq5YI$N3nl~03R
zTvfAd3WhyxTX?H#9YzE02F%?U#rzeWypNls?P;1H((>YB`KQmJ-@C8}mU^s|$`g8A
zxY#9<CMhx}_H9pUQhmL;Y_4hm7riA`JUN@0FOchAR(0aR>fvn~+HBXYqVs)B1*)50
zS~Sn!>UFxiOjJB@b2DRA$@5;ra@n>5Z-X4xbgYDl^Gd_-Sh-SKR9=Ln{~|$J%A~~N
zOBQ1%GPP)bfSJd7*My0GSgTZqc7a%NR&<oP!Kf<Fat3DHw|!h>rt!Ml?Ep6E+<8B<
zo0zYQ)u%9}{Qk`-7CCE${5$^np|xW$Mq24oDH6Ol{n1FPGp}1bYf(F<-eR##t;LF^
z@G4oveF}wV8cH5XdgjXMIxV+ZSV&0uG)VAS)%{7l8|U~eJ?o5*YfFsb{-CWjWLN4_
z$Y345K*V>}d0w3ovGF{rnWgT!k0?(>HHr@Q;hFO~NS-gkrC*X$<IcJu`=&eUuKzYS
zzNs=-oAb=`9{GOWpRe~>?}-J!4ecR!y<7%PA2lm*N)fjDM&My|NPW07UCCa;cZuhI
zmsVzsVyxfZ?cTh^ZYJfC7)?*+w!)8_kE!%1kIEVHC-RUlt8txw(1Qi$&<Sc64W@{B
zibyI4Oz0=_z}2qRycc<yZ(ss#X8D}t+IpWU1yw;-{I{ZePJ*=(E<kHZAM4I4ob$Hq
zzF@tpK^ByxH=>;0;&9Y)A37vrD~-?h{U+WE{cdY#SOScqC14w^@biA+CH@Nc)AVWO
zk4HH#%mnBVvn_b*8aBAh)O5y+yOxk#H&P+NBrtB(Dr}{oON`2G{Y9&Ha2}x)5@E9$
z7*EE!8ih?|ywm*HLoF-@*@ul~y6zLqUMFr5;zWq_)K9Bq4DXxoYQ)43LoD-{Fk^Cd
zWd+;$o)11uIqR9Orc_%ST-%x3T1x9su9YdgqEGt}mfwn7Q=p4Bo?zm(JjFLb#Y%5N
z<<R)e9gIVqZ${LK;3efdonZN>5z}^oo-e2~q3Q>Sp<8utVpILllB`a(2Ngr-5(uZ_
z7=nr?q03w7Lce^$Ezx!rQ`d%SdkZo)mRk?1t{D1H<~*j_$ZWYmq1~`VGBei8894;?
zgKh-k5p$f-LvdB%D5vj3wi_P6;P;=+uzF6_ppqmeY`VIb=l$lqMn>pfU<E~k1yRnG
z8a_K1puU>-?85+0;=;)<`F&FqPG4$8!h|RAIaiHU#}tmM3I%Q+vMQgM>^;-s99rpl
z%v9M=fc=hW&C-qEQWZ4?fG3x-*M9wPyH08Zq!-i<wr6Lw1x>BHxpKPl4E4LxEgH_3
zrOJ@0Te(HY1RE7TcMybW02?6WPatzdq01*CBSgM~;zc2Zq3b(?7hPb+nG^4$ug7p3
zBC|I%TA=K<=_#m#s1fm9MwL>Ryp6PTrjSC{)5wB0R^LBpXF1G9iDz@*&_#;p<3ct@
zw~pq2hI|+~!z;lRiHB9F1o)r>*rci9BM(7F)4?xqfOR*8c6!Zfg0THpV2z>BPlENX
zC=V%MG6rv{+lEcp1KXLn0jZPa>b>wk%nWF#T|uV@lU~(mHy1sNJpPaYj4dp%`@KLc
zPU2YOg__q?cIa9#>4Ka;3r^;QRWzxg9M^o1-rK@r;&CS0uH8I3m+4fTjAV61G*?Ex
z$_D_u)l*+()JrNot?T=sZ64`r+k2u_6m9$-5miWJcJr!tM75nmqZEODAWs;9Pi|Q)
zVAY)mQA6qli5V>h@iDqo`}>mzDa056{r5vk_0B@Gme8=2BuHhYDIQ^EKdht#`@POA
zPCJvYGO)tjWG~Z3pB{XpzCqLx+e0Xp`z5l*`af<*bztDM7Min}J-n8G@!<*ok`8$v
z?+P>xdMRknsY-s3a;K{zJwsl-dh5BQRQ~Q&Pa8(a8_LXZFw_50<qA$oB`nX)!beZf
z@2Q9KQuaia%@Mam*!~VPQtO7gG~05FGwkQ+zy#H#FhW$&_2A?kHp_`KR*LG9uPa&?
z2EAY?)8A-P{J>nM6qIj)dO^G$hvevIE_4$x%pc$9B^iALlp)OcZhHls>z)xP6Rih4
zjFzGX+e5%pw8dt$sGO7pNxh?1lQSLkY)IT@<~>W@s~_(W({>F!z4yJZLynv_z>-k4
z>&#0A&%m>e(3T_h`5$llp2x;=DDt+qT2_-4a((yvO_ErPRlPFx;$I15gW{m)8T{it
zh|st}J<s>p9xIx-aFL7RDJ>7@!;ngND=JOdd6o!8gF3OU@BU_^c!RDLL!E|TGVV^H
z{Lt5CNJ_qO`VMg=BQu~QF9~Mf={4n?_fD?p6`(AtFhoRPH|y}cCm@9i#lE1FB?+`z
z>>^(|r9)dNQ>c9tdzvI}-q&^@^^Su)*Tbim+#!G3uC3X{Z-f?S=IJ$Z7orK5KutE_
z=9jX%)N+m{=Q`@bJ5Z_kyjg~D6zXCjHpDY}e=&6GPpBuqq0U+Xn-uyx^?^4#t6De9
z!M~86aH&y(C0n588Hp9&k9-xLk@sn2OOvb4Z!()Y@io{dHE2-!zrc@Ci;11DW;VE6
z@!BSIMzwuEM>K5a?U6i5$IrG3>rT^@d5hY|fHha>Wu<{mbFqBP7is+VF)}}SQKNmO
z{z#IZ>2lYJw291~*RNrAG3={940`Re=V)+wtr^R;;!~2;rUsiWjm}*n1j$#%PH;1i
zz7@TE$udz*J+}OtR!iDs1>GI?0*N~yGtwOj5~X<qsmD)W^jzlGK4q#yD40ZQ+1Uw1
zo=<#S3#T8Q)Ic-zfA+4NRbD>kp1n^Q-@9|QIxJml)Mz-@`=nFO3wEQW{9Zr8vU%;6
z8_}hD3)$h!nR|h?g)-@T81<%|rbr#$aj23$6zA$Kv_N34gpRer3Uw!W=!+<wYr-$f
zBPgE?tF?GsSiA1FC=Vu2AC@h9a=vAzn_a1IjFvR8G9}Oc*w#~h+j;Xx{g7lZFU!uV
z<CN{-=+fwK=|jPaQo*`HEkC&vct-;-LQV7zFy*>RSYfZUvAJL$1#J7b!loj`TwN~8
z4cqE>HFdl(avZ37M3t=`pTO%5AQV(UO!qaF6j?s2msTWIU|RdQKYQ_TL{v?K9(=XO
z(9w`$oZKV`x&!Prao4pKNTE7)C0F<9B@&aDZnB!jGT683+dz}dU!tTz*urPe)7KaU
zs&=7^Nj&XWUvaR$cfdF3wk{{)eOO_!*5zK}2=eZ7oJ^SN2vPNPk;SLifPZOSS_d*i
z-;RGQ3mdzV<!v3dBCkLptA`czYh7pVH<S}qCsy>A6f5VNPv#gh;!6guGZ^TsH`2=)
zSMw5hCsZh}ckO0$GxgDAXkJBXZQ2BxDFy7c`bQxZ<}8d0N)?h-JTnrsFHXfuS{n8=
zlvc!k==aCk*U04ow)`;HYNoSKLn%H<FT40P7+UG%4!S^JocN;Wwte*I71SSv&O+~(
z#766YmH>s{RXS|QOfgt7+3Y6`U5cK7;hGoW*&XS!lzBysUhZMf&!9?S$1eCv)g&L9
zLi+F^Mg%n}a+O~6%N|4YzuGS)b{1rnevH}5?KKfmM|Yh{5QKTc{#?=W|D*z1W>jk3
z1&yM)^C3*mGKsomLz=rZ^~Oes3*mSNx2FxAG2U&!G)h<EI;+MaD>Ogc;!-fwe_Dow
zw4v+xo2Yt@R-e#zS~_AYX`)~pM?i*Mv>YtSB~o%ldHd@I$|Yyas@w%1kEDv;Bl4--
zV%D6sKFo4bAN1(_bZNu5^on@AV?rp*#|FlJ>}M>eX*Ou(TiZ6u@x06EMV*noN5X3b
z$E6tdNi!>Rx$57TcCw96-(6E3>3yb1Mb#=57_u3-`VHJw)+N-W`w_%w{0KEO@5Nol
z%!0Z0S09&QA<PmLwZ9WIx?bak=)Ot;64Tq4IUgi^>@bSCyqu7dp+UyxOZD!E-b}lK
zVd#KJ0UYhF2#J3k+F@6eT(t=6);g%)&W{Q;=*3o@L$+||7kbBQ(>pfpG_$mhZFN`|
zm>}Qg6a%|h3)7Zj5({YTO{jNj8ql3*hA>V1%@VP4qAaQjx3)hWznNe>e%$1~XVp%%
zL2YFfPzCRre)0U0SbDp^Gl$L8!+77$7?hPu;GAhsbzm&uL>&IW*kq>Y<n|puF(or)
zzv7pD+eX+#+Fu3nKi;D1BsqRWE1vlLe@*`)+3<FWx!KtZ2EGLtsXvZDz>ecq+0-#|
z{;Ie<w1$Od0wV+!f@x`H87RZ$x}H7}gMJrBeeLR^<-^OGa*wG4;@RmU5$^&7{E|Mt
zRgKlGW7$&I>B<VtV7Le#%ZxzbKs(T|TB>tJz1Sz1aQN0KFLM7oo4Csg%Av!ubalM&
zHP{PjW6<V79)ZzduC<LhBCgY-934iApMGbkBGtlxc_vGSPhH7|yZI3iF^i@ZY}6eS
zEEd`49|O5DeN3~<)Iqb<yAZU><=>vi%jHrAlRmg=fxtoPlo8N|Nq4nnOk~cBrbC^Q
zd4685-WsP}9~xg#j?9cSG=)sXv_7&-Gv#YhK6GWCRUi5hRPiE;Bd_eVRfV!(TE22{
zJ(PWN@;W%&AyWrXSPb-O#8lu;A5<rkvxA-eUgL`R8Oyj!*C6qNw0j>~NHfUp=Dj~H
zi<ZWo$d}tseOaX^FN02!m;1b_ZV6iYrZUr-WF5vSc0Y8EF?T4P@YSugxmj+0dYG>8
zG-?rBPGq8VKY-DAsM+|TN&1G<JvhXGaUcc%#L1i3Y#6*7S>ZK+6~d6&V)fw&+QDiu
zvNzoKauLd!+>pVb%X4lCGq}_X-R0^XE+prhXGF=y3EoX=`vT{pL6xF>WoA~EKfgL-
zlHtt(M;j{F06l&saT>8|5wEvu1xZG4>V~~tQf{N)4vKg?7{~|VYVt?MpJ;j>Yc7=W
zKacykNg1u9)ljd?&+C~_k!_^H^Zd|ZS_Vtq4_n86V2If*UoASoLDAwzLiIh7RVdHE
zHR3zP*SH@Nv6`X_*A9_y)e>6_RTL+<uk|B!{NOgaxk~iFnk1UO8KkD?*q=l}`apw#
zq(j}Xpm{6ppk#N}^9k>G%l@9FM&YG?fQ@aRQjg6WH|0b&$mxx*;9&LVOT7N;_WFmp
zytfQ8`z98Ukp0;+YF#?uLqbf2BXHw<2j4*>OsdHdywhKh>|7N*$ZCnFh$OA{gXv@E
zJc2Doa~rawI;8`N@QCExEYWQe-^YB#{QTL7Fk}3L&t``zOQnSHQyTzSx5N}8#xGR&
zq%#TtXCsxnhl==0y6F+BATeXE(bb)nA&_9uKd_4o^5opO7ZLmvrMo5jLN!q?cj6h2
zf&OaCwX;27dhr;}C2q}yjOZ87fFH*);)gxTNoPuokI<*Tb!W!}tz9L&AsqB4Zm;_q
zP#!)AMN?nT>=7xhVMT3ZMk!Fa7%5BDrS_12B8c%ghPs$xCV&hff&_-d#eB;mR7DiK
zaJUdv(S}4!`h~omhasXYJI-i3a$d|vwu35FzKLPY(`XLnH>#}(!k>E3fi;uu$;ldr
zzk@1e)`Z?<SoYF`n~Zew62KiU4T`dr)VO4F`)0G~Ww%tTBXtTh`1(lLr3_QiN%yoo
zGZjrD${WyQS{pjx+ZD^1x0h{6hZ0EZ<fUkDD2GOJopO?yCMXalkVj!<zQ54cvTnaR
zn-5KFlVH#uTcjDbk~9%x?5t&GkRU)cE7D^h*lVCx+P^kDO0eqGt|s`Z<YwjhYTJpi
zUyJRAXcfK{Q`y+AoMB5m+mYqNoD6PtZ^3JVb%5eT7hkTG$4nr`GXeTamJ{_9+y+$?
zatmokTJ-~SYdzd39VujW)6cUQ3=S3aZKq@t{nkbe6#(NP+v!E!pSO7iKK@4F;aOL8
z{5nU-`m~;*#45PLbggaR8arPtiA2#4;M<|+$C!9IcclXlDe>ov5m`B0qF|(N79H_A
z$ow4=nD~jMjV34@d{zapUITJm`O+gLNG2QahVt`SH_3*|DPw_4hSn7=ISPSvjJIs@
z9Q>u~Fh*MBJ@*Z~Mr>icaMl|`z`9YEbk=JYCdV9PG~DNHF<2kM99`+q>C(K|vlKL#
zb*B3OjKA{QFxp!A)<DzK0XTEt!RXWRa9Wt3=_u~?^b=AW@(8XVE9RY<FW)?<6cR^}
zrq~*G^`uVTu5F`V%ZrapoH^2T0TT624?Z4skH2!sF42pJ!V0k^Q4WvJc6(PvVcJL-
zs^RC>BfQUV3&IrEdEO6t2Oskdq~zHW;>=nV7c_h4N_wh08FRcCrAFpx#7(Sy&wMbz
zvmd&9L_o(S5SNcB_nJjwS+^X);FP<fhId}i_wYDCi)<fu7N#`1r)+zaTdjyC7}@Rd
zb_3VR_XiHNgrRIJP;@%n^o<}DaGt27(7q^YWKC^ygE)k@cVTaIO&aTZkD6S)-N*}t
zk~~zsO+2i$gZG0u7So5}_ZM=dUE(a}9ld+AU|rhN&`I^@!SRoGt3Q}5y;QSvd-RK=
zpTbOB7Pg35sG2!KfWHwNKPT&?s43=sdBj&dr>Fct^$|||28WKy(}njAXK(X<<5eGW
zPT9*ZBQj?Ee9iQBZ??|4o3HptcI+gY2;`X#*iz+eRv$fKki-`*#Ymw2)iGGcDQb-W
zGm%n(4wm)5SCHOi&@qQ7g}47DV>}jB1*h`YE{!V!>+$(4#7a`&6rlLZ2-`)fZ8Dml
zLR{*<5&Q4}hq#UB;q?Z8KXjZK<Lk?danBm%&X(|6CK{0(t~pm`^?+mA4WH0sQ;q%M
z`Fy%VI?sU*R-zooS8ssZ=IV>S$fX+sUaZ?j4&O3cSHDWc#vS?bSJm;KA2hEcfZmt4
zrfq+zk|GZnym~}0F7i-5S;GApV9FM@6Ha*R{;~n&Q83x%?S75;p};1JuXtovc=}an
zO=jdL7IBu#<+;WVzJtnT6an25G)5gud6n}|K#Ow~G$qXO8D!t#-(`aq3T4g@w3N%z
z{IP2t3YaK3rLa$dJ#z`fxSu+gah4<?TR4X=_<Z^%0o|ybJD?Rk@1SB~>+`YbF7Zo`
zV#k~JO&{Rj?6s2rLbS|UANh}~;?Ey;2f^9Uomq<2`JXS!JM=HApVY{x57oAGS8u+t
z?5m|ig2Tq6fcQtkh&VmRB(JcUeRAz&-n#&hV<I65eWsz=r}04_Q9=jFYxvpj#NaXY
zA_w4{LJ${S0=vrGAKC=#-?jYW>;Lm%|JzR@iJ`@~{EP{E<lnyC27!_L4+#1tY0Cf%
z%noD?@ob!&)yM@M!J=~c)q&|wFa`HGYvsZHs4?rz0%(T*Ae7`6P2+QNl-%m@aUb_u
z_&vQE#xwtq()T~<u)iD3{*B4_YgGGxo6?7{6hr^U>eIp=a?ASu2a~nXZPPAO_tkR*
zI5q2)00QqAYUJsrpm^a%JYoC)^1T@>V2htoiQN3xWCk&Mgz9b+TD5eYm)E+B2tBn<
z%0#?;yT#u905BTGnckP@Ac4V!J<s5Z1o-ekd3L??=}+$79X*4$H1?CNj{!<k1jX|W
z0AO3!3;Mn>a2kN&42i0tx$VH{QlMR(!3JWn7ls)57lNO$R-itJ(=;r6SdnJk4S?Ou
zX%IR$17BK`ksQBS8L<a(w)bEblngF5kFMES2s@j=T0I79Jp%Szk6eAMCrJ1UVKys^
zI8RX}-Hd0g-umkLaMcc)XA7$%N!Fjs3XFcR1t5EZc<lis$6=z_yBa765IvnPXzdAS
z&-~(z{M&Z?`JYmDaBwR<I%EH*F&Kk)A*&a~C}KubZIW6@q}g_oc#**G0C0v}qe_^+
z*$VufAo$Ts>INWqompDh?AAKGq_e#*lUtL-MYYb?-9VTW5ahswrkNo4j1=vNCJ{&=
zY|Uu}oo^wN+75i^NeCo~Z3opanMHHyvjBl!=kms<JVVjrd^ZB?%Xg4Y@3^qtv&tMB
zgz%x9AjFH71g{eak>6uNP&xKHU|dv)hz9a}SrxyIwnr>8m2ax&2)sHr8Q`frD#lIL
zx){fQvl$Fx2<B?9`3bxR3mcY?^Ie${MsLrYf*SK249Sa^!omC<%jFJ)hUSaM_Q1VB
z|73wSl9ZTA%*ae0T>BVxx&=;cKMAT(0?`$!wFE%-2a=9F9~JoTtWt!=YmD$4LQf|D
zdbsE(RY0JaJ19mVd3ZwMK%VoWx5RfQPqUH>-)tQ~FntLm$j?Bdc$Py3Fsn#-D%$4!
zmrvg%ZUjC2<E$32@nJ#CkdT%U9_)>}KuVT{O7NGOqn+p%vVgc^GU<NQE(3t1j1Jy}
z1qeS(D*&m!0aauP@!g<Eg#sUhX?>ynX-80o6$^+w9AT*MZm*ZJVTolwgLB~5Cz?Vm
zj{_t%K7@UpR4C|ay&8+z<Mm%I;{?^-O)D*!kq~I}!d9>}vB#RqyE4*N0%Yr3-|+T9
zx*GA9g>k~aOa|lECye~bU0q;mLQr$9h$G_Tso7Rg?P-V^pE){9#ref#s3m~VgYF|v
zp6pdk;A^-vBb&p1OX!rsN8BiA!Z69d1X$?c0|Xk&lWkO>oM4m0{~K4ai{09wbG(@(
z*64ADT5pjSY06PDU!?|U_R*XRzHDE!vh(;DPs*LHCxumb>kXr}^X_Wz+#>Kno2$GJ
znt|r?G3cXe$BfSM#gA;l5Hto!q3V=jJGfp3?tCe9;2uMb2yOck^nsc8UF3T*;ziZ6
zxhVLsY&;8VHTi{WW0vy{%y`gtrveE5UZZ@i0W!7&9r6HW^tgeSfTV$rf6Hg+-&9p4
zcUKB2H{Jj&kB}osIK#~c2O04Lx@8CgcOFyMjDw1xuFe}7Ma@3SQOG$54oO+LA>f9m
z9<`#|L%PWFyTuquf^e#zG4&t*)j5#jqc&ax=n54P;dgXK=(9#Np|U4M(pA*nj$wRF
zdvNz7a8Xu9d@r(vIDa5=Q^fV586JU&Ewsj^bYQ=K{Ny3&6$JcSAAR6h?L%mOrbV~I
zr?M5YR5-?$%O1<o@8Jtc?mbt}v0ZgM@<cd$e$EL(CT;-fDH<RaO^}U*ACAPgKg;Tk
z;hd2kdFViz7sgoA%BmcHK?OFV<f{Z1V2QPW#_22)qI`|pk&)c5*dQ>@_=qj=7u#N3
zdj5N{O_Ma}K<K=j;(u1mmmeQ6cwgN)0Po&%{`4zdPnq3O%cFPZZREVrbY)ewl(vWf
zeImdf8(AYB!}_^s^zD?hdG{npS&q5w93jh()&>$BX{pZylQeKzb2vRsum-Zu+{jdM
zA%LO*2G5D|X&=B~EvfsItxeQ7gV|~s>d#H1zINP7l1}7vdI*y<UEWoF9nmuz9H|ib
zDuT0Cis3GHsZC_v(26<G>u&VutwXis1p|XV${KG##s-oM!CEg=SjfnrV(-9}6~ZkZ
z+9$L+)nY>q$M9g1UB)-WJ5%Ok#@?#ZOEf}Z9y?e)(vTg}h;58*0qW2m@DBIxaGP^*
zSVZp?l3E1pZ_2bY-ti`sB?t^RA^}0qhbM3W4vZn`AqZlU7_fUV8v8O|mF}lG;9!ER
zE2{UcyDNhH)X=NcpLwjn&MzLsC&}|-$nE|td(Yue`&Q%7Prwz|Jvo##dalLu(b||U
zv1k68QETN?furBU&8zX4dISDlPT#5eSeRO+?Cpt<#NAAK4E7QqavH4PtoB&6)s}e;
z4rh`2lkVX1c5huV%ZqEm^e7;weLv9U<^qlz+aLb6gSWg?M|Gv>t<#S;>)+o{jF}@|
zjvtfHj|=DEB|)EEq6##~xZ|G|5B+XWgiGzg0o{^})_C%#FhNQY<;5_U8bDtd_VUPy
z^E|WaK)Gwz6C{=!fwTzWKu(6!FrA54zOe{Za}x|h31!e_?+3w9Po>Hyti~pYNGB3i
zHX>Yh*Eb^@@dBJkgj98)qg85_5lJfmBrySu!%|lP%Pt-z>>uRA0mM6F*kjvzd-|cs
zm<M9mxAwi?4)l%KQ)*#<=O60U`Dk&vMGoJHP-%6aRAf!Ofhpn-ngc^}eyX!Ir`tfK
zXld=EMk+S?fSlUz1f)A26@IS8XR`?Nzh>ygIFn&~(#!ev2D{+Gt&Vex)%2~5IOzlW
zE1QsWL(5*WsN?jL=1Fzw0$6e65E#?Ffx88tWjYLOG4v%d!12V#3zlT&&D+`BY)>PO
z9V@B5eur^<1_ssPWZY#8XK+5)qqxG9Cc=qOdU3sxJCrz7D@Ch7nM^CP3XR^@E<GV*
zqliuhn(tSbr6?92rdGQo>f?OO4rpNe`Tz}>zD9}<D#=>^X#}peN4yMK?zBV;5_n(*
z;qu~YqQLVx6F3wuxFw)-3EZ$QMsD){v?<J`yv4s32;wqBY8d51_g2tm6{=Na4G<ws
z&YUrxD@EN!SR5~Mr$!#w67mWg(3>VUE~7lQWi#8tO!DG9xT}{<TEm1m)jY9z$k5vQ
zOSQxf3?iR8w5G&T+%X|Kaw5|H3Mt>&l?q@rGg?W`=)_sEEH#QApX4EUzezf`@V2kW
zLW64a;R%7};cYm$Fi^nU7<LAed?NhWXHKl~lAa#Z6q0K1o*nd9@pcSj!;^7duRWHd
zq53i5F1^!D(mef5iI#i6>GGIS{?O*B-{R%^xwIn>c$eExA!a<vblx*XR=ui8fOqCl
z&my?=xBgdvG%hfn!siU+e|{zT4tN~-{}CB*lYyK%4q$CF3|I0`FF@rHZ;GW2uM;d|
zvW9z5WnEsM9|8cBp<xz4K;4%$L|tIYPHJ!+I$}YtAwZ2y@e7dh3dZ8M#^l{JJpgzz
zYYJz?S10yUK=qB(^7~oXIFNWBmzb%kU?5P-|4KA^LQYXS`zq=i!i`+(IDsiJcDF(_
z(!`1*q*VB_#H0*fpMe%xXeU3xc+C}|IL;k7=XUrdIUXmN65kxhUv)4&c$(gA<pVV$
zIoa%ew`Vp^;1dcF*sTU(Cbz4^F<a&^`{BF!fsTq)>inbxj_C_?gVyscc}kvo>ws+8
z)NxXKs?+DN_kEl0AvDswSWNW+=`(UAu|H60HCrd^PWH<%)?fm?8N2S`;h$R($l)_1
zQ!F^lkrob&Au548AdqV~j!#Z}Jzu+KY#3Nrw|gWgV>pc7Jn%YUMV+(HXf)()b9@>9
zwmmpg+L+13VVEw4^7JMlR7Vv~g5EmRyvak^<dpVm;#m;_J=8bCgPg_2ZnC%i-Vj;`
zRGI|hEsFKH=DeeIMCEMdUX32uVVit<{Oim+?uMPfT6~8_4+P<hzeH!ELtLFdRTG>i
ziP!+L;9jy*YH503PqmU9z{0wm96}ry?a6#ZcWyBzAHe%@_C<n8&9zpo?-PA7JdW4F
z1lpTeJlX3e&&7}J)4k2)8;dP4r=ZeJ30PO(5ju$H%8{dZkz_+!!Dc9UN$k_lL-Ow$
zu3NDre!si`{`$GpFu{tc&|iIrl)&i}e6lO@Ry^t^o?9J*8cLU(OQj^Q4cgC0v}rK<
z|3UTJ!n?5gI=D~_{mt3Z`W$*n^(QCN@S+4gV$tVuc(rTnTEk7Pjfu>K+P@JE=h+Jv
zHN&UB1@w_xR4GVuy0SYhbX_(Wa4%+_{*wp!bOnsNucbcR^6sky^{F_T#sz$8qh9#b
z6iwA~cbVPHI;VCHE$x3Mo7&dWuyp{3|L*l{h(tYBlg2AaFIw4^{B`$08WA@;>T>1O
zyFk&r9p_O4`McVz=!XP%&0atH^oY43D)^ijTI*T&mE5<^>=i4TVo|}3uQgMj1#HG&
zQERz742q!mE1jVhxJj7uKEEuNEBo9H*N}anQG>%C4*$6r7fr4h-2^>am3lK#b=y|~
zTiFM=qm<<vsMi5OO9!19^W%UP$(fxhkznDx8jCB2jxOxC>~#CV;ie=eW15$1*w85P
zP%RjFe9{;#3sF5)fH7({GB7=IRa$)}N%U+t(iVNPwQV$MqtKCf<ezWTjxe__G$20D
zG+ITbk>GE$mX;b-V4h{v8Jp$0`uYpyB6sGP;T8F<B)t`tRqs8cnl)SW_ZMS!iFD3i
zA8YRcOZ_p2#p(4`1&(3355H$lR9*6fPSh>S?KRCsclvKysN`nf-xl*~x&%gZfAa!7
zL0Ls)R-84K*)4p9t4#|<IAIw~wINO02lfni!AGf_UU4XlqwXq)c9PzMS0VvHM;UL3
z0FEtIJpUDFiJV$_qPF<&pv!i#v}l*t5z4CPrI@QFGWT?X16bVQ#pIDwB3&56Q3J#M
zr?K)w_}XPOp=!O=EO%rR)$YE-^2L^Py-XuMX`s)b?6si4b+0_5h^bB#(#n4*GtQTh
zPM*}N@|zkjVhXZ#Xyt8;<>?M$x2f9zjwOA=D>BQrPB|)R-t<b07z^W0&{Lr0Rc}?!
zW;ocipRuI_LE($&?9M{c5GAFNO0^bTlHSa6*QzNyHLl#qV8*;j@u+jz?X(XRtP8aR
z-|zaiP&6WS5kbhFP=ajE3+-JohuwLYU9&-o$&t|+de`s9HMbocp9VX!ri&m5f(Z9K
zwsmuKMva(k{k8*o<4nPtue4cC9I8h;maO2~Tf8x|ohlhF$LHcc<spuO4xbXKljlsT
zyzsQ&R;08?b9T6HJZ=CJKh(D$06{2a%XUpIs($#3*!)i)w2<05E2Wk1Cn&>UjK6V!
zq+k%}5uG`%6B?mF8FV$cVpuk@e2HD<s0^oDQ$xTQk2ml&io;q^?S6!;3rO!+d*?-c
z^;JyH#-`-(+WgQUr^+bPg`77xwynS>uJk=i{4?DGMUE;*JSAa(1edywB%LeSN~Avh
zQOZ-MD&nkE#-VDd3+iS^$D6N%gCkYa!$0<91@(E=4>JrvowT}4bqS$Pwpji_o!p2|
zSQPuq1~B|jXR<W?vU~XN-bb+Btb4)TWo>YXs%ikk`2oUyC1+hiiZ?*K_iB-xBdc01
zYdH;TWgdP7-2Y{;W%q%da{Z{4B&}w{QxZI00*`8ph|r{QBh*BFlN_99JwQ-r;fYw!
z_GYLvQMLPvpA4-PySRBvUyINoZr+&U;nGxTI2oRO<ju#_3ADe1w^G=_dPio(^EKaJ
zU``GL0Yba$2<yaYD?9`I^s(L$*vNNKyu<mFTx#E>%FfnS*LjG;7gLw^%GjvAlijDb
zB3qP^&d$M<b7#gwhg^I`O+7W4&aR1Z?>0vQ01$eA@c*pTa6PNvJ;3Tt&fg!GC&2NH
z;$jAe$R=<p<s@lC^fJhXxm&Ezp3PwX#65%4(e#VcISJmzx73Z(Qm*s+Lt>Q5Aqsj<
zFuy;n@}#b8>89IHfJ095trEcByQO?djTQ<=$4PfU7S6<euV68=Cx*ZP0x=lapG1NP
zZ@^7n9L&49W<)NS>hB0@er31J)aY_3k`(E5{=-yGv*6EHCyxU+i1+MHzdA}(uuWgS
z@r6b$wZ!{ha{Flq*~u?g(!8#EHG?zMbmL|xR;@Jr^Ch03dxPX#5R>&jLPR2nXZXPI
zI}oQ=OylU%V4oLb9_EA!P+7gLu!{k8L-5|M)2vFx9<1y=F(Bd%B(>vs<_R6qR}d#f
zK|OyR(XWkq@ZkTju=Lg*D6!_S#1n85U9>{aya(1O7kg(&<T{i$EHgesNz&UZ0sN1X
z8Y`bv7{nk@O{cxG^6!Q}>Y!nyiK6}Vi=-pKs&?BOvoLB}QgILMaFYYH(EK=PPLCF&
znEoLA>C&3BQmb1zvm83CG(H9;?PaP+cTmV&1q=I-VGuDXhxR~E7sEMa3$}#_sQ*=2
zhnxflC!lKtibc~AR`xh4<a^x%jsTrHSyX3%se*d%Yy8wKJ)vU`?>T5o0V9b68)<Gk
z=;$lBn_QG-HmC%`m>cBs)z$wGdv6_8W!v?ODx!cWsbGMVgs=dmMM?yb5(`*}bW2Hh
zHwcK}LRtw$T0pu%T9B6R?rzxge&YSk9`F0O_x{cw=Z~|;!+3@+8L-xUU-xy*`Kvik
z<QUuILYge)22JrG<FVGVz{ny~qmY0ljR+IgNo*{)pLndavi={4sGlU_-+d%So%;Y^
zJ=yK&DJEHZqR0<;9|Jrei85M#!TK2P`q6KUacNt~!(u76)S*St!Me3{dT2dvczGyC
z$J%--q8I)<F4H0_DY5Osgqa42y<@HE$E_I8P>KuANZ}ke`eouhop6nehH_ATL{V&z
z(Bpn*1n%mNzPt4)GgXNejLSBa8Y-um!{k$huGD7Gjwxs>%*N}2Z8%#6EK=fL-3-GK
ztMv7_y4akOo@v+|ulh+j9iU!<yMiW==4Z0->#OsfCgB`~nf^&sPkkn?uj@a4X*2x1
ztcztuNKAC{k1w||ysR{|<8JTsW%bYu=U%Ry5yg3WdKYolB*@4N-qW#%Skfzp)56EA
z(x;cRstMX)iUcK8`FctF^t@FhrMj&29EO|?9Z60(BIFSdu5--uCutdF87)ILe~j<C
zIDdOZyUE~SS56IW({g(0R-b&RP`^=*RRF#HysX6I$wm~OxULFEq>z)nHW`X8r)0Wm
zjy3>A0#AYoanoAl3N1&Qn*7jes5ojcg8MaXqnN;XBfY?TYxaxz#Pvm<R_OR1F@9sW
z>v=F{v9pujUWs7PQzW9%h<2vkgj0x0MP3i{;lMQEa%`|Y+SAgqhu%IQW&i49Jt1(`
zm1Xv1s+GGIj%6zO1vbwaiz%-|;VOnokJ*wKZbDtzpUh)9vsF$C;S}rJ!R)t&+qmso
zJJ=IVqb7rolm~PQ$;u)6TIayrIN+ZzxvxkyZh^N~nydJq!zD$quZX%f)PuS2Hb;PJ
zh|&u`qszTA+YNLpS{x2u7ZvI)%cWu?egf`PNQX_GH=(cHRB#`35+?M^N48K9IA`m0
zK4DaVW_5iA9LxbzFtug|vNG85TTzPQzCW6uOZ$vHQ06FLpc#Rhh(Zu%goG$H`5EmF
z6I8PXJPKXxr2EMv`u>r2A(nx{Nz$GPNw^Pl5Vz<{hE{i*fk$dRJ{fx1m}K$18ci<X
z5X5!q>+uhW>bvLsiH@|-(-AgF%{E@~ks6(`-I}Hgu2`PWEDvg))7F;W=<ZZ_>|!As
zX;?d*kUH0VJx<qFXikH;y-nq`vv}8GU^3_KjH~lSr`RAdeNNrx7222Bj$H$Q?2T2!
zG)2x-yY*?J)c&5&ZIHa$_fgI6uXXz8YW=tW&%AfVcBbTf`}ca*XU0OJc{;&7ENsRP
zbEm~1t@oAZ=DU{!mDm{Bo$5XE&<vRwZ>`b%gjWC_X6XZ3;|HNBrPiW6sIz@*#;T4x
z+vUfl&n*uM3oe$cSohLDhvB=l%Fom1tnQBJN!N=5Y*IqLreW;8uW395LbhNw*s}Fu
z%6VYc)&?LR7%!i1@8I=e<u7-`!mWkTuLgkypjAHeZDB7rhV&s{Wx!YUqF<T;;z(d)
zWd5f+_f>ry3CqDnFUfk<07F~7iB^4~70B&~ZdfBdpuWim<#Oq_S}r1S3`<oj?FznG
zb7)X}V;3MR0xrs1ZQ-Np7fEihwC(}0I4Ayf)b=g??u9ju1KI;O(_e(;4JBtaCh-z*
zF*Mg2hINT<(vNqlZ0872h4}!C+ZZmbjLA_cB|LQyZP_k4<W4ooGMfSORn$xjQnN(;
zZ26b;<B|i2Gx|x^E~a}}ZLSkH+Sq#UET-^twgfM>I18<1Q@5--<8{**+@k`ZHgcZ1
zs(ya$6P@WG)l%H&!!Nb!NJxfR`b6-w8Ia7X0uRd>cOAXwZChri+X&Lj_U4XG`ATWN
z=wsTmbktvkQ{J0;g9VG0`k%kvcvQ%|F!Y!;|JOGSdR71?HVAu$vVXeLmN|@p4&glc
zXclc=jk<Z8RAHwYG7)k(S)5x?C-{|LrZ=auPEIxVJ?Z_rL(qc+L&3?IXOR~jl{0B_
zXb`nym07apPuNlOU<oi7Ybc={4%#~xVHQb4gno709Y~uPV0YPVdgd}pZAK+vn>Bt7
zFt8%dQNOD<)m=F&6X*zdLKeoq4wyF;Lf%_96=9F}@Rz5b`AIMEN7Y#Crq}x5;3)5P
zMyOZAj%c_9OuP;JIJIa$+!&=ayzoy)kp(~M=%NJxA02?a+iH`Jqe$PjUJ_@=g~$F{
zN!c2q8%qwu`>INIw{9ir8B1Ff8FYNQozp(vT$GR_dHBSZ;l34vttHZJ=1j&2nkw`3
z{s8Mfy|B>*m5k<;x9o8YTvotJy7ZR5y(6BK;A_4Kebf$zp#r}k3+bKRnQXZT?;N~m
zgAnKTw&4}S1v7@{R(S`Oy|-h(%JTHq_>c>0&?-^s9-xPA`fFYee)>yQ|D#tltI|RX
z-}viS56Z1bk~eohJWP^_SZV3yh%y1b$OJ}Qv|zHWJMs8PcXrnJ06Np7rLh4ZWS74e
zvxoW~#P2<CP<)1fVunEY;8p&_TWNmMqdoinw5{E}Zcm>UjJ^$ns14WzEoS$!+PYdL
zIJ+X;BzG1Urs>%6$Berj1_4Q)3l%v;km9|P<>7Ld2bkl<WVI(#x5pn6&(=TP=FM)w
z{yx)|FbICsahVtM&7#@M(>27a3xY5ts7%-js)b#>uxWFc#BB+%qIv^M>rE>+yelzE
zZ>>O+R$4sf&l2r-2w#0((#H*`okCFG{v3q;W3eS1tK0h@&t2q57WsjFeI@nsI|2Vc
z4+{|sA`vc~?%8N10dqL%b2#x?gZz455`4uD&7mwr(p`JkK3puNnc@S5?O=NT&>W*Z
z_)84f+mXN`^12^@gP8zyZQ)ynL2LT?k9v|43U0jF4t}%(*!JwB*B7y!eQU{7Z8h{T
z9|uC^gl03vvBmiyQSl2ojd|Hy1j|9@drun~s4ss<9qr=j=g7>QW=T9nbJ6{jr+gQM
zW*N}H?GZXxc|mqv2wM3>8CRvnz6=4M`p|~sf{M&1j!b%ap7F>GWNB5R^A*BNdBe^Q
zdSvNRK19Zm;@{f^!Cto5l)`#Xd9C_>+hhuyH`{C+l;UA5*pGAr+rw{o+EqR#yP=1u
z>o^U4M_<kMVAL)1&ZT6MOW~v%837Zj8g_a6E$?eurP}fD)_VBw-)=8U?tc@a>h$UK
zz&&)&p%->iA(|5x%LOVcBVhsC=Y2h<<9zW^_|4?2Uu^V?i9UsX&Z{Y;JBRJjSq)Tg
zIQg^!Nt(^%dYg=u?&TN#vwyB1xywjA<y%REd_*kiu&Q57dYXcXmQpNXsHi-|ieWY0
zRF_P{>8f<OpE-#QV#~h@WHGsKU2rhP?RH_G(MyL{#f}6r&GG@cE{2MHc>7ef*`B6u
z$j=^9gdZP-p;Dv|XBDx3H+h;G=Ojp?Dms$>+S-1~W5c^*(`)J~0<Vl6%pRCZxt%tl
z9+5%kGZ|fQI*;oZx2ixmv<Ur-mcZ$U`!!8rRN;y;w9s|jP>7oY(`D%s{{6HwIsl)*
z?NY{Drt3Qa$L2pExc;z40CSu*;-i214BTaN9|}=4mGjO=@*?6hSS;9R&#nameD4Vw
zg?@f>UwfWD{R9rv3Bh>pf&>c92o^r&PpyKmEa~0iSeF;Xu|^#a^6SmzQ(gV`$zy^_
zCeOBgVU1fP`lc><gRe#b{UUKdk)&1A;0Gp+ApF^(c-m#ut6IS+yaavBs;yO?>ka4S
z*gTc_b>cMZ@82E5B<I|1DSf=qsE?<8<h!J>4AUObF5EZTfH;m_j!SbYX&=utxzeOf
zm?{cU{ZP=W?X}!m4+%eQM%X<*Hv6FelM69+lEL&k1q&_Z=+S29VPG2Py^<MankixM
zq(`&%Tq6i~Hb@w6<zvihkH~cnc&8gXT#zu(<`ZzI??$XnRYji>##u>#;<sg;D{dGv
zB|O?^>t8pN-|vKiDp9p;Bv*s-NB}#I;*GQKu$|u>%-b50;9@agC!){Zd+kZ|k#6;o
zVf)LLXIiW0rM)X@XrYI5$`vy`@pfFXRf51)*8EFsFhhf{3DJ4XN!^A)Nbcdokk+=p
zN}z~WPZP))*n*yIo^_XdPW9|>m(%k=D38)H<5GY31YCG0(}Xley^<u9*z(C3-PBma
zCiLJ`1~dHjc9XD{6H%%bV~pinwU+0WMK~uhw0d89vII~HpNN984)R|JVc83<tr~i#
zZaQ+^W3!CogN&vDsF8G>dj|mE%+w(MTvpphtDviNq}aWG#ryfm-_*sq)Uec_-|Tp&
z{Z}(GV}yma+U*}cn#$@U9{>(!l(l5QZaT)~>vu8>R5dRghTa`I<tQb8Zgxzk_=E+t
z^#wW@rP}>UW?Azah1mh$I?S=rwd)YY$}YVP;>-cy(wl+|KHK4CP0RxZJqC$L0Z*<2
znKA)yI=sD%XO8Z}bb^q{56gG4!McrsQ^8G0s&$a?ku8{;AJwfb1e$NG?x<^lBWg|I
z)Kx1guius?=?_G($`zWel9c{>%K0NI6xEz2kpu=GCYl23t$UsM5CL5T<@F$Y!j+af
zm(0ScD>e`-fJw8{xah5;xO)|p2>lkVg5yKqSpT^O`nU2(&>ilzAnnz6xPM(veP=B6
zumdQHp3?P!*fJkPx`l8q+}=EZ)iz~PFzU^L?^-zymM7KhDe|EI#m4k^@ABtbx5t8e
zXD`aeeEwf0*#FTV1cQ(_%O?Gf`tL^c-%9)675<Rg3fDRIO@NfgZ%^ERYRW`{>8Rb*
zf}5VuKltr#{6Bx_e?8`69~g<?$!DC1Z~wR3@V|8#evacl80i_$euhW#fAqQk{^l7x
zs3hj`3X{2h)9U}`pEmfgNOm?0Q1b47ex~PM6)s2XRJ_nI6-4MpevyAWkkGLUIXeYR
z{sYjKbtH6rsrh#8WFxdSE{nP`R^8pwj-;Q^RzG+n)_^|kM+Od<1l^u`1&JUIq*Kqf
z{olr}FTGRqPyg-x5!t4N`>@@}{$1g(`!E11r9otULQf{h3#TkXxU{n7{uIps0bR_)
zX-L$u^1bdJWzlkb5Wxwo9Tt|$k2cx2feo|0t11v8&HC-xpnDSnf*wLAX9aTFK%Cwc
z<Hio*qt)Q-edu<TMEVR2uzWw;K1|L;s7mC-b~wpmC8nTZp4!cQ;R7f@haXSZs4Nea
z4#H+aCH!nXZ!XgGk!%cq7g8yjz!flq!ruDF!Pqt8Z%$B15wuMl9}XYyng_02yTuY;
zX}`|o9NhAY;Yc8%a@BDy+Ie1i_^c1}(Ue`wtnklpH?3E8jn=sL%KsPZ<!1-7Qaa3^
z%@Y4{^Jm(yk?{+>_8a<$!;O6-3e*NFFl}oP;<o3lA;9Hf{!t!yxr#vtlX@Ks4R;3N
zOFI(QU26oP!8QTe+z|8SecU@lw+BE=Lr+H0I+jK2^^Oz8Y5nQW?nJ1Q-B_%6Vgl#5
z>P9J10ff@P1W|v2qRFa#K!IkPx{!@o6*8LC@}^>JRM`{xq3(PEQPvcjis>2h^H+8d
zj%Ep5Wb7YB-xP!KsTh)j3M^4g-f#)^p@%X+x`816m4fl0CikEF;$e`X!^I0j%WdEk
zDu(2eWrC19E>M1dP&RL%M@sdlwoX+}Z{6i+Yt}nI%f&U%j=1}gG8Ta;smq1QwSdr#
z*0CGLC8vx|I<<VcO29XALQVKpLWxfCOKpc8$wutmHWjg?AF-5kgj@~;H=O_Rn*X$i
z=gP2DK+*qCtfuIO+htDOg1Ey4J=F(U2sn$B`v{J?1UQi;nif%w<sV_hxkAamlSrdR
zPoG4pYeIh~?DK+y8fG&fZ&?eSP27Y7-s<Tmnbk2*!0l2qCUUsgv5L7y@2i~bwrzu-
z`G{oaVo1;x(|*e#b(b1qi5<OUm*XQ@^717z8ZkNH<K5QM6L!cM)V+_Gc5Fb!Wk1>h
zPvdgd1*KHOuA(zSaLjx|SZk?dHkfh<qLiCHw%OM{7foG#2~Oir<U2873pwItkxqO1
zBoAR=^DD6MZQJJZ%MXp&H^wP6Q(@*_p%=L-pW#pWLsHY&q>(IlA0|RW2gK8!`uY6c
z;Ii{{To27X@34$vkckeso;+1${KRElXd0;-=>#kNjfS#gt~8L{uJ@sWPUMk_PN6ta
zSG3F~&ddAs!T)&Gsj*^^WAmYBy0YJ2ozXQc2;Xu8e5QH;@^MSQeaxz%SGe!I78+-Z
zF9-8tqscob3gs|{#DhjKdd&nvK9tzz&76^$c*Wp4VhXPyY-5lXq{W3GOfmi<<|Vu3
zeMqV;MyiFwSvi+A9ez{WbVotyHO(KS(q;%apbQm(G#}^;r&)uJ#>mT8Ht#3bNnK+$
zO3pAiRsD9sw{#w6U}fxj(<{uk2yVZ~2eV`Wj1){b|0+-!eRK~U{L2CCYMZp>J4tA*
z?<A^iL5_9LA+?&%_)KdZocD{YtHzM=8)14G1SL3WU05&ZV%JM_?pRbg{5<W;8Rxhr
zX{u3xL!T=BHvUK?T)jv3Ux49%ojiY^Nc_~WapJgIKL2aqjC4EERGtAQegbJh-<_;Z
zQX51k*RL7Cdhi69y(x9@>|Q>7(e4>jIpW&QFw4%TGit!UvWtaQWL~qM>_H|%nScS)
zoLtTEEBX4nIIATb;T<>f_!7vPBDgEZ`p`=DYEmq<`y8It56E?&CHf`2^-Ie*qN71M
z_Yhpuh1F|=An9D<fw`JKR?TS`$9du%?y=j;csj+Q0V1>3zRwV`2i}2{*xc|Vmk-X?
zMNj_u?EPSdJcty^UZ}q+ATQzy=B7x)j9H>T<baD+h-zn^RhQ-jjI<!{kCD4pA>6b-
z<^r3e6&wwE?b5RQN@BbN!SwiMC_UdvwzShuPzniMgF6tVwiHfJpckQoN!eyRda)1Z
z+>VZM0+acueH!7{7)$ikG1W>971WO>#L_v$JQ`!Z)9(0e=q{3{f|~CpoT^IEEnn$U
zb;Ywc{0@f8?c0gV%VLew$h;>n7dnHGcgw{nR1GB?Sn-bIuJdEnju(PD?yu?%V?6+(
zDTE;59vVv|O}be5(I+Mx$+~Yfx5CI9w;)SPb!`itXb0j)d~!o<+4ZzZ>lz+AbYx@X
zKIM<d_-X~!zw@-ROpmonCNzrqs$qtyj(6&uC6&ZTN*%UMY-ZuF68S*f?oHZ5RztY;
zbO!+}-GsOD*hxj70#%ab`hvoGZm*6T^nWZH8Gu%5Q9rXbh_b(h#p<H53>~G_S<&aH
zox?>_mvsegC9J4LadH)|j`TloW9#!gPWo&d`=ESbG%=#Kp^R_kQk0HJ%=i%VaWQG|
z`&oL;wx4R6cFR(1rdP?Tl3sMgIzdc)neo&S>dqk4pTn^~)u@B`P}=QmpEo1M4sqJW
z9VGdo`V>;UK{>jckuOZt{^lTQSvq=F;-`;a{-9)eANk(}&wnTe-J}pSMeKgvUr)&g
zKit<;!b6L|Q7nPs#R}*@nZV244q5NNU)Pif^+6CltuM`>!m5J<O*e@C>OAqT4)Nf@
zR(|<rJQcPaAjMhaRy40gu#rB~PFfX-!n^}LZt1rwDe0@(*J%{aA;o$*Y8Ae9gz+P>
zKsxwVOn_zG?qT#2M`nRXLC!b%1LCc-ENLS!NY~ezv>NrJHakED7w1-&FEQP~6KU3W
zjWHgZxZ^|`AXMwCJXKfvzP=2wt|zDchdPdrE`H~@#?Zf+NO$>u_aPt}ErZYj9dXoP
z``$-y?LwUC_O2H<GbUEsJx=cgi@5L1SjG^?-AhPf!?>Nu3IENqz#mxM2uVXYWa=_~
zXLwawtM0o)Gz5y155O_BxE@j>0MkTXfFw}5M=nlikP3g&kJP}O+Ec>UMiO7=WMvgV
z?{iI~fO9V(NbvCioe8xgW**M`Z_=2eEAk9dcnw)!`(CE*(%*xqyY9}^2P{=g*iCg5
z7BVSu6L-6~jm5J4KRoNn{i-E7gxIDmUHN^sg84s_*}&-HJ-ut?`2*$wE($n=*9MlJ
zU_RWxIq;y)qBB!Xn0@UIbm#}!rRX~Ch7p*%dHsJ{fE=R{MLdp{Z}{ZR!g=_I@<lw!
zYWgp-OGSTRWsy2S6j@#DoU&1s**_2o9yC0QBsK@S`3W47SFPaGC;*hHNBLx;+WpH3
zKjtC&Zofg!dCaT#T}^^r|4mIi={-+e`Q?Iff?+~ea;G~Or_YBjknl)xcxKNm7m2Ui
z7%R7p2aHgK-7@i5M3|fncm4s6G};sp?lQUX1(xYW!kTfSJYp#rxE`qL+jI!4O)nqh
zd_Em}7k{^wBk>-LC~`8}UHJ2x+%o7~nKYnMBe->ybi`l@wx5-%LQo;M%4vSu#j7Kd
zqPcQz34C8`O}gHP&s}l=D!32v_2->$-(!eG7!cEjb9b=}?4eE=01MO7>4&&w=(l{c
zSqc__x8BaEi^SKg1Fy-D+=lt#Jwm|=BtL;Ytk326uqmxH@G@s@3>dDCgoclgwgu_L
zo=O_J$h?OUE+MKlZoiOwzD|2H<IBVxaV-tUsF3fuS2Ch>T=y!zds=fVVmIZk%u7l<
z=Ti!bfe>b);le2x8@GbhiikjCfMihJ@gcGR^@r*!k)dc%sSZ7nFOZ0ZMGQVYGkEi#
z$n6Cnqh56&oMEbFj8OWC<{{%!R)*h$^EAUS;bT;s3}QVs8t}=uiTjD<ma3^l{S(J`
zK+81UbdL!i7e$nG4N|66S0t3gIxB899wUZuk|?eend{@dZ9-V*QR0UipM|Z+0T>lu
zMypi$Q^ij_kDiO<kx~+Wlw#QTizXfkv699~wm>ur5`xbH5?TnP*VantYGe&lnT}y-
zy&UV^?nRuBH?8OmKCp_Zzn%#y(g1ay4o2)wk)(o~T0-o4noe<orTm@XH|OsvlhyeU
zeKd7@72qtc5HbRl$g*%+z4@?|^K5a9^WLn`GI10gU)pulVzNwo`3t=mE61z%uKIom
zvX6AeTO7D*S~7~p>><3=es_LZ7n>9;S1hpspA$iO!PaqW&7Ak|G_L2tyv_pp@E7d?
zXx=`3Ix0fU=b!+BT|N*g5v&;B!a~Oe!bvx<eO?kry1FBl`U7L2*}RM0{;9<ezfO*w
zSFEYw+rtSf-Elg^J0$(xVhzImHBu7&cVj0KN_rFDJVX|y7&r-x6#6L-$1MqYm*|5f
zoAUX?v0;YGbfmnNwz82pIWNf~xXW^%eyNlBFu80PS{=g5dZ|~;C_<QYM_gBjK0Ar9
zDaeO1>UD7Q6E$sxZdy`auj(YVDC{5^t_u-X(e*XrA$ZrMrhjGm8N{iQK6)1vwZg19
zjQt!4=Zff8MpY#XwtWQK?n!>@?GJk_QHGc7ls<_w8&b4a)mt>zP=+Y=_!&JBjyH<n
zj;Y7=uHTLHDy^@%-3Iaq73+W!9a?N%g%-zakr&HfF?D46Ngr<NihmiiLdHRBW0m?(
zSL2aB-^+qA08GaMMxe$H!6U7jt;bBfmJzs6d41{1V}`NOp6nm%H96|>p(#VMCBzZp
z$tlibjPgCXZ8uZpLzJkD@4PW%nYS47HGjuVa5)s6vn_xs-!5Gi*JVwga=19#RwEkX
zH%=EzmoubrwlA$}kH-WQrAb=fjNYGqbUe^UlJw0s_{wJVbVG0??X`8WeNC*jW)}vc
z&5!SKS>w-XpV!JCv<~<lUm#1mBA%L2^fKeb;<$f3w$ox&79p`?%)P;D?ZYik7eoY&
zFFB=yu|VLt(bw;^jdVH!OGXCU`cO9j4<5^&+Hgea45pp{eb_c%H#fq1z25(TRA<kb
zda=QuCF4>=@-63pcI%-dksH+gS;0zDX5Mj1u14W`1unk@O-1OjP#(05f8$y|C$O7E
z(b|~mlv)xqiclY9+6Ang5`f5PmA$z*QW-;(Dp*Jft!&Mj@0G?!Q60`mgeaBzd3EL$
z!r^#dW{r_Dv{3iPjqFLu1gNj*<-)fjesVXl5o_L3{*-PR9SDJ=LfQ0Nv<&_p9^tI`
zS0gTd&X!cHA=MqEw@uws_C(VgUb<`f6c}y4287Y>bchzvT8rh7v|Fmb_?V$eaP8p?
zFO0Y)8qpKpSK0v0k+;6Wp|My*vVC<}C@xd74-jW1VL8S=&GC=5o8t2*XVj&kR{l^z
zo1E5aN7pFd&6k^0_A6ZgRs3W`!pRH$3uS{j`O%mQKP$p>8~cs?NKf*#fShxfTe8_S
zP;$Z92Q<_Hclfjmua)aoazkQIzc}u}Sjb&<i9J!rT_X3Q!L*Fp(j*fcH#H285rDU4
zGK%<dgi1Rip0<g22Yhl=tr!=%Kt;YCAK`;a9}5^J>~&A493J%53MNin@tYhh?jcG9
zubby|4|4N<(_@Hz&;^*N>$TLize}unTy<H3CTlZk!G>Sz^q!KfP%xUYR`Dov@Mi1i
zhu-ZfQOlgs9jo!XxU3<a^Xd8R>OQKmOt7r+SD8QjCaj$OBd0=BZQzzhplb!@4jZL6
z>wwXhp|`Jy=EcyiUz2vEj#7*-&?q!qxlc6MS)Kjld3No`Cw!aNBu0Ezauw#Eiii0T
z^^}nA1rqH%et09`ThaY6zvs)82d77ojAlxLq@xUwBZshFjJ0Q5`~W*Tvz_=OrMUI`
zVX0|m`qEBcmF>{hNfpcB;MiCkh2-bO^k&UN)IqQL+R)i6n%U1)O0ml0<JJXs9~(aR
z5EkO7(A2<`CyEaP3ZPvS>!5KylQ;v{jY_8$GZtu6sAl?Q87Sv~B`_(@fY3G|>y3=s
zEC5E!H}~$2geRS^W=?1QqP+D{yf4spi5c&J0EQ<GD}EBp?3S1qfT3Izo1ka^rWTVg
z{N^<;<bTb6ngIcU3Hpd}`HRy(BcpDy66y?ZwNEAf7XdCSqYiG(C6IyG=*F~c&ULYH
zf8x2j?7SMFRsh*GKO!D5cgeNk{J7%B$?^i}m$^?HlWj!HrW8TU*>`aGD17O8(;Y%5
z!Jl*Gsr3-{YZC||XhFH$R<254cwz2ixQwt+p-nBQ!myaBG9Jn0YqldN^~P@KDV!Vx
z-`h-YA4$qmc=<sR{%Q$AlCo|qSUdFv$?78gczrKPX-gVolHKeIy8EUm#k_{n4GyL8
zM>nGFgZtN-OhCEclJ=zQg4kmn_b4O|zXfw_ktdeP-oY(SJ=K7DZH;hyIrR2}Ao#y@
zrz214i&6gWj_y2f`EDTRM|aL9;frzG6Wj~D@iK{npZjj*q)dz(N!#2}-n-Mql1YUy
zmpMs3l&$DL`dE6>D*l`WbcktG>qh$D4{74(>&I;ZD7&E_XEUw}J3AivYU5Ul<(3U*
zPm0Ij4fU(ofw8IxDoEa^DqF?Fhb@O7+p}R1u|R0_;+vs3e{Rd#6nFZ3Lhv&9*A%Q{
zw+`X#eYI8T#rZLoQ#z@|(_P1zzpq!_<9sx$cz;NK^E473s^>1JM`fyAhQXh~>BTv<
z1Tv1bMH)LJ1=<)V^w`2F-2Bcb(z{zfu$@gnTqwW%SnewLlT&Y#?%d9)QsiC$(;$D2
z=%R0&(mcz$VU^i7RR4ukM=L%$bsg3V_psu0Ne%dO&Z4L2O<g{qZZ7@?81@yp3tbfp
zZle32a3hi1eG>JmcMZx6CNN^5t5dm`qihL4Epy!)iG$iYbEi*JEXcG4vZl^kgHMWm
zWxVWL0d!ZNVnhflhat9W3XC61&9s49qPCz^Ye~DzpF=eBNhabCMIbwdt4ZDj6i+iN
z3gW)<+1_oW;=D>A6MBco!ME&LhV5(|j3?S~^3)kyx5xc$dFkxs2krRU$vF>&b|>;k
z<@2D!sBB<yf1>jATvG<gPJ}pjK2W!RfIL6AI%JIf{ET64F}RjWc5dEq_qSs)*1fo6
zrtt^tQXKZ-3XlGFT@a<RN2@uz?&V(rg<!uk9%dZ%a6_3O0suhNtT56q*P>wid5=s=
zwaT+hS9ac=97F(+z3t`Y=J@zyqA}Rt)(+|S-UpKGwCTQ2Qt%P^7CC;8Oh{jnbO-4W
zVQClN^*AGsAHP6xSPEb)tUWk5mDKzGy2$^r*m8@<-4ICdlnVK)(Ap5d(lutw_B&uI
zoG^NDL%J2A#X^+Ocs~3GlUFdwP2+PaJRNB*@+Ij?ZPF_7;q)Cc$&WhLUR^wuPpaG7
z%D%ic;z4*HCS^V2f=*2gE~a_S6A#bUc1yr$O|?Zm2*UmGw{K_`Hl%5I>Tu3$Tpo)#
z_;JN&V$*J0!-VXDA;G6)j>_}vYz>M{B&7^&vaJ0L*ik2%!XJ~WT$hs{H2vdDJj?jQ
z{)zV3n|e>jFG9(!3n~mn0~m-1{*qxp_s-^W$KQ|J-)EwxpK8u0vWTQo{Sa}oRp~F;
zGIo7&N)Z9g%n@pza>+E6!e{84?+Z-ZttDVA84e%S>F!E*8=xd(C+!c~w_kcT<%nha
zO*&D9mIW)(J`B8eC_cA{lHPbb^qf+9BT~G%zQ*$+`z{1l(dBGRk85C+TOT!#nTshu
zzHxq$E%_C-sS7nL8l%fI-)`{PQs$o#DG}_Kv?DLq=;i(eW2#7vK-yx^`<)?}n@P(M
z7TSWj9-tfNw51P-fLyH9Afo`bNgjnk&9}{m{MMglyc=HRT)F@|VF)XIcx1-T-5l*p
zNwO#UNjziUv4&rXNQ$gaySdwBWT3_FyVI5Fd=b$`g($g#wa=;;_p4O*P5by6hSZA8
zWRAO2Z;yS+q5msNu~V6yfCV_0dEBIS^mfk6T(4;<&-2eC4h4OFkTt5FQ(2XVBS11l
z>AJB^S^G)o&$ZW({`^99HVd|MUUc#@KJ`fI)HN8<Io#4~KHXLFuBD4^#jwUAHs2_m
zq|^PXZ-4XHE0r6<Fd@P5`19MFLH1E)nsld~ea{)=P0ak#3JzqNmfVIjSRBD^gHNqK
zm>-UyZ>-0V1~rRn`W!-Ixj%f-%M~d?3g~7g=-xt4bBkj>ihEOSyvI=ach?=E8nN&W
z{}#=JlJWs8T;f#-Wmk^OM4i|SM1@YA=FBTGlKsL5T!bT9<}F*^FU_HsDzfAZ0s!$)
z(>g*Me=xDIvpW8iQEk<tU6Rp`_*vweHVkvn6bLP6;K>LwvMce$>x0;`9!%4E=$G={
zAEwOD*|U>IXq6j()ak&PnM7?QdC%1D+}A`_%IEyohAZ4j%mR*y&!8Ftc2=dT8cH-S
zRC2_A%8HTSA<bx7m%esJ7x#c^T0{6iVXWRMfSDA1Ui6Y{0rd@Y>T)JU(~?zhc4dQE
zaUykT0SVBL*((*+yljP!MM~rS@dcvS5^`#|$y7{Jihdx)A{WibW{~k%hjvkwlG0)Q
z2X6wzY6-(Rklt8x4wn*f-u)qF%!2@kAom-0hW)M1pQ!J7#8$;lo18m+JsknES?=2@
zkvu6W6EzaDT^X_7eVShOGC+2VbxF`<fX~#xxIIx)XH!OiYe}v&2c|yTf92m)ioC?^
zEGvjr4i6#8&UXv?ejY0<t9b-U$#4<3%hHK*+V}b;FkAHvW~P|*e(UVbb&`Xj1=>xb
z`_{7%iD~@c>ix@ua2UiRRNi*E{!<UDM&J<#l?f6l(SS=i3A8tIk)48<oJ>dG2BUo)
z+8+nMeHi=EFr$#cu5TGZRZdObp6$=i4CZvg-pe$wvO2i(`!?f>TLhmzsqg=zKzcX=
z@uFr7@6C7`mO-&e3lLP(?Whq;28@MnU<>pW!sfaslRiB|l5ksUtm(_43&&g)a<nxl
z?+&ELpsS35X=uSVk!QbE>6X)=pr;@c&3bYg^8ZGgMPro#9E)KMna`wWWq4!>23#u!
z%~#NiMRC%^<|N@tYjhZo?jf9&+bNtP03-%j3n$*Tcq#=syzdgyBkPRzyBMwHV8#$;
z{+GT@qdRy2+Vuq^ynA{I29B4ai$Uw|xMM`x0^<2v-H04N4PZxGS3|^Rcc713FjA7&
zBYW=neiYk)lSNLajxU_?bv)XeuhvD<!w(7=(P*W1zqpt6JljZ0evBWk<IY4Eps!0t
zc#YZn@?*ZA;~TLxB5qdnwmY_@sqSXuof^Ae$Z^Y+F0Pm7>#3n8ipd^q?xO>kAT^2s
z5t^PTmRi}b&j+N6Ik6GX^M24KaJHoo<~*?*(*`auS}7Y<Y(v*1)*HyNfbfAwm;1GX
z1*4VNHd+h)xDWFD7^CqxWAN^S+U?@}6;ihy^J)tc8wPhXhWpl?T4wpp98EjcG|n=>
zFmw5<OZFQ!N$>N#e}u3~dhX16nk#(fdfhjqHzj)gDDHIL>3qjPZ4KLQoarfp?d}&A
zviT$K_!!YoK_4@2eGJ_PvS1l8t4JK)8`Q=a49al+c(!(R_-2z{*%0??IH#G(!LzQ#
z_&5tvyE{26vxtT<WxK0k3wy@p(Xa{JMj>|DbJK!tLO)$54uFAPY)VyfV8b5)oY)G<
z;a%2huZu5BLPdPj9%fH)+?eZX5(Fl-YzUbE+rQbmma9R-!QgGf;JUv30_4O4QY`5c
z`!+SlaEC2n7f(gN$U7jocREH8Qi5=dGi2q>A(N1fqn}7b;`VG;GVSL!Fi=>6Y`k!J
zJmR9BUN@U`Pc84-2F}>%Pc<*aR}AWwxo!+w_&&c^&vFxkbb&!;?fPFm_>a4;e)SNi
z+Da3qK}#`;6Mu?<mc4L-q^?GTmEhJeJT$Y-DFNi>X2y<K*_fLvPg7@VQH?u%9BmZQ
zN}LvXo_dmR+_Ne$O;(5$er1f&J{W-ScA=}DcO8inBCN>bbkWq&wC8Hn1K5g^`++q-
zxNg&TJw?gb44K+_Z^pJmmiJv1FFvp88y__O$q6ic)FXa!o`Pl2X*OXykG@z7Z_sOp
z!4E8$m`2k3A*d!=ojWnCEBCV8(Reg>xp5G3ZBTCO^Daxg3{eJa&OG9Nmbfh<;PGG?
zdb~~W7%bn;p_uR{pqx9+WU#7EJB*^)ojz-ZHO4&Vt?s>cAAPsL-Iqo+kC>OXdA8md
z3M*CLE89Kw9K_{v7~?tmc7&aFxlDU~jy0!ZmgoWyFTyP(FQmJ<JH2%2TY4<MRrHxR
zu?_-o{(MRv)-0dDx&fwOpq4=@Fkee2>p1^jlm!^;ZHevscI|@(LEVG&JOm##2LdT>
zXUh=DMg=8DZ{1uXQ*$)aYx3B$XLS}Z>S0-7*}ENH!9MX|OKp$U_Y3hn%}U~7JY^Ij
z|IDKiPyr-%;LGROx%rMAN_~$nnK$x%C5drJu1)&R6E)DJBnX{dv<fjs3$Xdl-3Z22
zLRjB8Z|IPl!VNwBbnpJGd$*H>cYr$&Cy5P&3q?CX1eVj!4LstVHU)D^gZN=1G6!j=
zpSBdgTKcJQM<1ISWmTRjyT(|Snw}Rar_7hB_Rme~N#f*pAoBV4_O0+=-=t51g-pgl
zIFf29Nzhgtdd@?d<QyZ)FpyS3DeHrm_yq<dK)ILwF-|F7N~smWcp5d*W~<joTdx}D
zKJIn;dcuQ=nGdx?lOko&@k!PKk-Ns7@Wvx;M{_R>mzR?-oo;e9=%K~-r0ND~BZ&=K
zU$nyQ4c!kKeG#UR3(pgG#cLj131ySJ6sUfAtfv=CnC}rAKS^_i8Qh#Sz6h$ZxX&`4
zH;J}2po3Oa+jJLbv;}QN>7+pH#WW!W3*38O!)g+Rblx^NV)GhVoC^pk(C(dbMw89e
zKQ1FCtW|*>??6Z?pZE#2j3`qjr&OffH3hVWSK8e!)QTZ?fWRcPoXV-|X|a9nKe<;$
zAIRJexC7M~*IM@V>#?TTPrE3vgy&fL3TP>fpO&1bH%c2}U|rH!e9dvy26sAZM6scZ
zWrePUHqHX?fO+zfaov|=L>xLSQF-N>!nS@n(SFGBfr5qa+>F{2wT^Ox(>=68`(HOg
zo$xM`gN2a9_c)Je>bsXCqVXFKE`n8d`ApASvKr?ND?G*xVoAkpN?FxbXXs9FLj^{)
zNY*@p<=!5)jL7{~r}7Aq8FtOg{1+xHkL4SOuEjJcP>eA$%uE&$o<wb6x8X|PMv(cK
zJTcSlkBRTCxuFzQi8-~qoLR&mTz%h*2-_I|z}NZy0hj#qN+KVX`4&(w)n@Cx^8J|=
ze+3oeJq|1i?jzMK>i}*;?AQNwxc{pr|F1v%MCd`oqydHJe}h-*SZGqMDyyV-3Treu
z6N?}Re6pf);RPLDG)QQM8n0}tS5eH%p_74SF1%u=?8S-|ft#Gs;F)?uFg3%V(P%XU
zNw(+*isX<a11<A?kGjeU^@gWiqbn>))#IXK+J)?9R8xJUp9r;UR|TT@9a>vM28tuC
zF)t(^y?XxJtN4=ydB=Xx_W%8kZ($SMo*?1{W5|;c6yfccoynkBjEp;>un%c|&&b*!
z-6L%j6V+&vOi4WB>N*!maKg%?BgHV1Hx#b3nGMji;o^uP(FEeFb?W?w?BIwzlsJh5
ztKXP`mb4JGn@PEuf!%Z!b!VV+QMxf2Mwqq*>L#T!Q*#9G&<R(`vsAtMU)S~D1#dsk
zBJbPlg~FBpj3s_(z@uiP1VrTq2EC$HID;5_Vt}J3Fl_t%9ej!W+;0?$_W$s6@$itO
zjHRXGYuXF)e`f(Kvwc5jM%9!H8EbPU8HH2Kls&&y<>!(hD%2bK?-YKwmHh|)`7fr<
z<Q_;@OhV=Tlm8|}{z-KBA72+?L5|~h=E?@Y;|KpEK=8+lGdysS=6RHoeE;f?{^M`?
z=NJF)$N&HL-FFi`?YvxmT!yHll>Oq`ng3>H{{1=s?JFi2C0#QJh`SLc_7STbG-3q+
znP0ea_cX(Q^9TNT_`zm*$OPyOulqTV+=fBkHwh$lwJ2S3#lOp;{<mKF(2Ix4s=!)8
zeB)e^kn_YK7rSt@ndx%z|JD0cM|7_s`!o31hI%%lsNa0y{eNpKaK+8NiR?vu>+dZ8
zdwbCoJ&3~Arvf$qdl&s@6|&)uHY>LO=MDF7?b$zF?8tB^10c4_KjOdj`9<&%JB1gg
zn&j_fKmWWH>Z<}<1w#&*bbohR{qrhjm4Tg{ti_RD<#%Y*KYe(y4|bs8=qa&Fc>kA(
z@t-%{{|LbU-<$t`Z~p(+Pd+>U?D4kC@w9G)4jC_q#mOmitNtYgnZd?_!qF=rzWxWe
z6=PoU4vJ!nQM0%q(C!jAFU=v8(I*OE$Q)I+<3At4%XrDAVZ`rCe&Yh-@`O#}d}W^#
zmU!4U^r8FZh+3VsWTh&C_Bhz(pxSApO(N*-GRRqbHX)$p2l(DiV8Fb^@X=QOA3zS-
zwWAfP0q_kk0Wy}CmHMxq)Q#=>D-eekgNCWxWaj5bl}q@8MU137AhjGogiTLK!qJ*|
z$8Ha1U3pSS-owCE2Y|)~0e2nm2C-X;vlS>*KkSn}iir>=yuAMYmX<fUL`U7IM~#wR
zOjq!|mAj|hXJ2OY?%j9vU4_KRVk9n6B8m%DFyh|z_JhT5wjn`Rk!tYdMeG1!Tgv^O
zOBE1-{3uY1bv%3*bPI1M)8#NVmugu3mGwt2vX?V8BLMr%0_zppZcLm-<DELO(!|Y&
z>Z7q7Cs(||nwS7Oz9G4U<6A_{CwN0o!}};gg!hQ=d;po}_YfS*k){xRyIg!{si680
zB#%}g_Kj@SIuqrvIX4Ljvl9q6oe!Dd;eH28CF>6nhd~|n9Xor-B`q#<+V6Ih`$L`P
zra)IAHUR?2N8Lee$Z4uPyr*#4f41ZEei1jtguu~+7&ZP$hQe7!phu4z%p(r`!S~7_
z^L~V_U<6A|O~Ws!EO9O+a&@<C4-rJdJi&z^cvF4s3p7Hzk9-OJpRb;x*~GDbK-2R6
zbxRtjo$xW@I}rZWbm_;Z%@!O#r0Fjn^}c%W{<ITCEz2dJ=h`9S191J9%fLFI80eC-
z8Cd@reAP`(a65xGAiCCi39>9QBG1J9rLX%>hsewk_7%33eZ+nSGk0s+#csKmuYBj)
z0VV;TSG)dA%5_AXvJKJ!Tj6u{Yvh<jP4z99W4h{*BY&#F%zSO4VF`#=s~XDfoVdeg
zBvbMW4acf^?Pa|Jm{Z*xH!b~gj2t!oy~7=+s-}@r^xV=daMdPxD^J`NbAk+$*eDJi
ze+Ou~N-EYM7jdQA9tIpd?MNq@S%ldJOMsM^!<ge&NbCu4_LB#St1El!amNemo|63_
z)y=&yCCy$s!3LQE+b}Wh4~JPPBpe|GD^tv0O_0cKHwc05q;%dH4N?>9%ymJg=qf{4
zbd$g}cznxDk^*a(8?G3peQYS>5bgWobwWYi9k-n9c#RHA2~S|N^XU@ALs+nv?Zh9~
zdlKF!QmXH|?`ngG?OdX^+AkGUGz&2jE6XSd^)*dO+|NxB{%NkXL^=>*#jw2UJ8J)A
z(ER&F#R3SIw8OV~-LNs=(;6&GLw27!A9804)eh$r<?$8a{_Wt7Bg3?)3{|bV<Pn9f
z4H7on#Wj=o&!vUPHsjA%Zo}YZ9mDb6C84nSqkjd#e>z}?&Mwdbn@rS)I2X4IV_P6l
zv9PSI?-^hLq|1ZRD+s3@7(D?Z4u^y5SITi$(F7nlSdvm{oP>aQGeinZzC^RN-xlor
zjI7-D`OeyvCOedDEeQ)L<3mdbG?Ar0=S5i#W_~Ao1tpz}?9KP0{~)6&t$$T^F|TOd
zRz)4rSPB3?Md2}I25<n{rs5<{0cbC|x|l~01Ru#Pa#N#-{6msM?R5g9=0ixer0?-b
zHfqYt0frML_Cq(`=!nDxZ5yLP-whuw**KNPerGlWD%n5@3-qq(`=CB?I!@G~VH%AQ
z9RZ2S2G#C^k?`}p%lNo=5L`4L)}nTCLDc>f9sLm9F$f{;Qk4e<Zbu|!9*O8`3(mQ>
zKt#t1NVnrGkp&;D?qgL9LD$Xcpg9;aN0q;(I$bAmYhk@ohr6fQiEd7;HX3x;7!Uq`
ziMsx+#1iyCb~{O#bFsAM)qA3X1cn!1w5l8H@uPmw4BR{C!xWs^50b?M>&F$(Ax@mU
z(3`!Q<u;~v5aLY-L8ZEs&PkK4)zG)|!6M0UokGIa_cSi%3IRa~O;blF^U64X+Oe4m
zeSQo7{7#Y|6(n9-t3n~`BZh+%Yaf{15aZX7`vMSsvn>!>YAwCNvD%7F)s#u+A1<Dj
z!uQ8MAAo1+jZO*(P3|X`Ie|X{*oslp*#oSHn>bj0-Fu6#`AOSW08Jj$!ghwi>c0C!
z+hx=67?R~f-qdD279IErn-%D#;1@RFWnF1ST%W=Rge>Q{?ZPe!BK7<X2UZGa@j0iC
zp#r7G8t_?+DoRUT$uxNW${!jAk^Ji)iasgpDtAZ}C1E?GjD2su5k{SD4;lFdV>_qp
zVV4d!)=;YEB8Pumb)USBxqnbibu<!w_uSD}+iJy8Waf5vsPlI5u>6G`vteZa3poAt
zLA`kZx?Af`S)ws+aa;K&?m^0TzsynMd$g|{VL9~M1>zrI)ST_uAJC|k!AzexMzqv*
zIU*YC4RNR93adX#etV{q7~k@#NW+o>H^0V5!r4Zn2<XgOt9f^mQFvEU9xGQ7+$(-m
zVfQzFeBjz#0t9~~Ka{*fwdMV->YmnU>Ai%O&t;#DtBEu3pFOvpJ6#jJQ73%iJtW*v
zy72YUv^TmQLjKWM|3}ce_rysbc~DkK@y|}&Es?vsFMb@~7UV$N$0*#uC`9rez&XdO
zeK15m?qPdCLU0bk^E95X^W`<bwX3_e7Ob-%x;ut$r(YL{4$hglH0ho+F&4CwbSp52
zQFSE_hAs`jqUvdR@(MRNOZaF<i^`?(mJaE)#M|g#hI2ta0t7ugW8t<IrtPAaX`LtV
z%s$9ic7Gt&;jHl}AR-*Apwe-5YQ#5UDXiexh>(;haZ#8pdKJwfI<%5zmicp-fof*F
zi&o(apve2=n@u@)V@}Am6VEjxHBHV$Z-V1v)$t(nOO4dicCohI51vNXf&FUHx~H~U
z<4&jK>Ue`5t8<Vs%8QhjVtY#H!2C6zXF<qkfl!*DBlqLO0O9h{!pXiB$H%qz_U|gW
zOL8lk#^CY3m*r&1Lj+Rel!lP8v1lTGIl(dCeLm&rgwZ|cEXp~RJdg2vry3<+bvg*F
zix)i-UUl^5oA|On&l<JL6*9Gw*4^j*J(E4Y<+YFEJWngxsyDqBNvrU2iJFk$qJ7`3
zzK|daD{QOtR$|GVshga4&ayY{+S+G1m%=aIyG1V4{kr(&rg*6nT@`!w5~IN0`;RXw
z-xl$8)@{sRU-5Vwjhjn$a?Vh}?c-oYTZ^^Av5&*kI-hN|J<NvNmtg&11IFi-e>*FK
z2>X>LB{Cd{J`AimlOk2!&<!n~QRl(caQia>`}nN6Z(M4xzB@BVrtp0%Cd~OEXMwie
z$1<(xP$6SEo7eP6+^x>IXQJJgBuonbL$sf?Vpl{7c?qxzI<xB0xh(z6-x7wy-exe9
zHgMQOb-2IX1^OulIRiCpfs?h|ms~VuzJ;*8PS}KsCM3V%0#7R6li)LT?@j5Z_p6*D
zN5jhqjWq@)4TOtQG7M_1Z?-(^-^I-ht5}93`xO%LTOCg!%8$7<L8Bx35?rx1<vZWm
zOKt|=^-ILZ681D86j*W1kKKT5O}kX*O7{4(RU$jvP@NPZ^{ftiP$gCQx%MqM7Z-JC
z%K;^=Q&!%qBy=A6ei`D|9}F0NHX^^VqD39&=XRdwSH)Lt?MV3w>Xmy3ILoW!b>V5J
zSJn!^v9+(m6j3rZNy93~fBLlvErP+win+E-f`r^j_=EdJ^qK2s=h(D}t%1E<kB^Xg
ztV`{dM<mU_J7^h7HS?k6K6<F9W)`*Iv{u%Dht(&NcIJujIz(aD6&@Y#hnsLZ?OD3K
z#(T^_WtQs~Xl~bMF=h#o=(|TJ`b!cIfgIaXFd7Z;QcauXFsQ1aZR*i#^LDE(NH|`B
zQ{Ap|5_B_5IOVwb8UxI!X=BK=h2fyKUDGf+;jjANaj@VQ`B2%?w+zHNpI`lyDQX{K
zIjUm`hqsO>{@%R=iC{Kaq7WC{LN?&gQ#=MYK00rvs_i;ItG8HkTe+aiH&%C8g?(Xv
z=i}=|P-g0&Z4DA-s$<<$JT-hz*nJ&@Ho>uA1LrL*^O^YqoJm{A7+F_f%LJ>-aE}9w
zG<2k}-AyziWk2)j>Lk<g_WSoEKuK+O2WMHzxC9HGn-#Gw2AdVn!99Z5HrU47+Ir%J
zYu5v)eRR%#;Y7O=m!12SeP9!*H5QazZ6R}MM~^znUNik$_nOZ9o#EYw-;XZuR&oTt
zXsw`}^s?85;hvkM8YmlEFlNZB#ODujgkWOrgI8tB;XVd&h?jBDhzYGnYe@z<GdF#Q
z@XfCT*Uai1PAs3Br~}XQf>+SWNJNs+BQoM>WL8dD#z=4S1o|!SQRTT+cbe{emfLgm
z7^8{Z2~!Y=1{FnaUueW}5eS{QVO0wsYjJ=0Zem>_l1f$gj_YUU=ZwURN-5NfA;)#u
zxJ-el8Xc{i`>F<6jB+sz=Ia8j=WHvZ_E$3ASWcW=e9BV7Yboz?VhZt>W-2m~!*Gar
zgvD;kb~k@`x$?2XmwhN|UQ97AKMb*QiOs&;9+F?i9N%)ztYL5e53|m`tl-_U$HN_Q
zG8V6`sI${2l6mK5^%Tx##*Tk%m9n3ayLohPS9;6%^6`hV2m6;2hRTRnS+kcuM|WU`
z!R{i&daebD%PwQ3mo-fL>OHO{$=aqYC)vhL`_l>Ur^zjh%yA^4L2^#0u^HVH3Ch+w
zGExSaS77fhcPalcTu)`S@K%~_7682Wo%Y{|S7#e<&PD}Yx>bdQr<bwLrWL%|c1-Me
zHx||+Wk|PlFV6KV;n1{t!>?1zqUP`LR$A)bS3)m1iEw<4*DPXb<ZR=#DU894^>SxU
z;hUHZNTHAJy%c&TI#9RI-zqdtBRGhdu!V;0hO0w_OEOn79tBr?g`6}*qf7a!;-t_~
zq6beXN9~S`BHy<J<Ids~ncVLqQLI9$3DKM;G>ll1Se;I<T^`%0qZ2}wpw733#MENY
zjI3W_V@TtsL_|?kqy(En7P9(E&nq=10?al?*76_*uZK(d&L!Kq{7YD@^eb9xFu=u(
z{w|f;9nJ@h9*xv&YRbdlSuz10)JTGt4Wj6R8W%nbAAUhXQV*qZ*u7D`O*N!VeK%)S
zWJl_1>JYm*;}N<gJP5n{*w)}ZhmO=|Iho0eJGb0^P@S61;s>nuu!-=L@HtUN6`~!D
zQm=3^29l~1+|G6(Bf$-FSH8*e4tDy4pOe@}`sAN4v59zZerWxG9Zz^hiAerdZ>zdA
z(P5I=U9@PwcvRg8#Ug=6wJ#Uqey+pjrX@U+I_}Qn$$ze!j-Zli+b{net~k->j!z_H
zE+yF9jgtnge#p?#QMJfl_gy_Iym@vLLm#e@R%Ts>fG$&?_dB{H*HU>ZLmt*h%2;-o
zbOdo$Ho8(#PWxQ5ua_$zlJhh&=$;cr+c`J1u4IclHprd1)<C$1cb9;aH8*GDu`Nw8
zw8~A`y1w4F1|O+Bv^Xee&a<<1l|?fIU%sIA*2kI9opDd1hS7KX67^*sqxZqBAb**{
zkwG&PYFuzi9APRxSM)ezXXj*Wm=;4ilfyaY4!J^j;GK>GxX$1#LtD@p@FD_DYt$Ze
z;v~aHCJ=rT;<^I70-WE_RA)!^cExY{T*PTS3BSd^1}`@8R9iJ7N_7S-riYv9x%3P5
zhsS<|C|lK8<;Stak+jSS!3`D<FMl$k9+Ku~2xMXuSl_5dgfgZyzJ^TfxXQ%ys}ya*
z_lU082<+edoH4pvk#c&sYBTI&#^)nixd;^XacF#&fDY}o@RI%wgM4U;)mp0pIK=yI
zw^(ced9bhNb@$T2cS;WtUk;}I&WjsPl8jUv74Nc}tOOinctRB4QCQ(oF`bqu6)$8Y
zWWLRy`lW*@e4_s}`HK-8O<RWdVIo9LCpdLZ++y53MXHL6L)<<VJRP39mLV1s>>4zw
z(ID4&dRD~M-ItUM_k`b54&n9g2;V1PESREDg?<IWdZ^hf_3C1A?xwH?!=iO(#SNSh
z@u<6nQ+A&Q)=KIqmn*}BDXP$Yj~ZWvWF!{}oAzFpwVO|+oof~_LbUBRVCvu3i=B>P
z?_}}YN$x?!jQ;tl`Eg3OD)=9I3b7n&)hy=v<<>RWyY0Du{g_L5Yi{u>@nHO%bKk4I
z*j&PmTVE_BPdWMB{OJfTn%yl+Xg{~ym-t`Jk`NQQnLKbk5?bckDBZ#Ja^|M8;_7Sr
zEM|L8xe|SH+1OF!zDm>WJ%bHATlR(7prY7Yq2P0CJGNG^-@O#_^~@4$c+83Mv(4i_
z)kXzhXAeARd%ksvr|06v<6Zp=AFqz%242W}oVj|7+CuvE^op6L;+?cw&Wx&J{YzIt
z8gqE?UPQ4jy={A%GR{nF&WgfGt5HN^-1FU7m%J5&mj6JM?f9i>&DRNid{RkV7tCHP
z>R<g1^^x(d9o+yM(Yd0V46y>4husVnXhto%*H#>5{SWFILvnAw{@zjky48tmt*om`
zcOcfxsrJZhfwWxV6)xLl2SPT}{Wf_k*_>}4Zl9g;Y1K*Z#LS`kLagHF%04T-^-X)K
z(zD5P3w&Mg>-nt~GxG8L>zR{hz_OxIY(bILz}`*y+Q_^mZWxd;x~u1cnwbNK^W9eK
znFnW}#ogXVbj;eJMb3A2TUDRaMhyfQy}ikC_r+|kY|8G<v{5EXX%ixwoRlZ)RNE=v
zdp%XONjJaQ%5G|fc7@2+OE5%kL>pDHBlt=5T)2ns?Z*9_Gql;{1tdMMPQFNc9Izl=
zRL`(VyY}$mA7sGDRV-a@W)6Ju5Wi%)efRqOfbP{a|2l@VbUAC}y&_85q*x)cj+kBE
zRXJ^}<odFTU?X#(Y{}*9^@U&R*_p+%zFkW%_(U9KMlM)iU5olGD0km?s}WTr&JYx!
zkH5%Lh4|5iWOAt@m}sQ3xmk=uWM7EsiZgS}`!!J0ksgWZeqrUHQPMtDAA+aTx7*-Q
z!7w7dYoJ-k)H@V~%g90eP+OETPVZZ-+fMti<UC8#cVv=SkbaYKRk<d=u>mO+xh*~A
z^vu)V>mhoC>jeqX?RC_q%}<_bpVD0INHVr1HQhV6_4WQN-_?O?l=6-c#_vgE%f(VH
z($x^Wy}r|i9^!4p5)`B3`l9L9(*gYI9-|?6`Hpd~3$DjFJtB@!ujrxTi3kx`+)vVN
zpw6CDqa!7dV0g{6e($&$gBM(Y5%}5=0vBUN<~=SO$1Oc=+PLiXu#Vw_Dc-Vc9B!Ct
zjj{I}!s74zvPI?xKAOrL$8x-P^kzzh@>E$HtBxs<JD<xfOD@50+VL<_CRG+C&Tt|~
zCOYJXH&4V|(^FP*FCpGpG^mSc=Q0z=554){?7iV*jKPz3<hR}a_%Js(zOm$&s~fvw
zNkYU*mdgH<XE6q2mQ;kKdG7|x-`NzU2%oUXxkA<}q9%W!OUcSV8TK?~ufc-DbTBU8
zi0rZ2WIE^iinOXBme8?Dtbl5h^Z(%NtfQj*yLK-~N;`B7Ie^4a0@Bh*i*z?gOAg%~
z(j_gRl7f`d-9vYGcbDhpdDi>9zw@rM&ROfMHGl92jLytG``+KZuj_N+Jq&0cCUnMX
zkzcr=m<YjiC!16je`kL(17}~KK*&U|OhituGeoH?WkY&`wwmC!VD%2AC!KRnFxv6q
zn3jv5H{Z*Vq5{SBOqv~=072Nd?!&VtEAX)7rIBn=VceB@*YL(E9@>#K<hY+m8K-$x
zI6Z8;-)(#e`r`STb`(7)B}q1v`PF^N3NE-SRleiP#VF;Lh_wM9dJfP|m<m=#<<ph3
z8XG))na!C%4XQ5c6jR9_wl1iM=hi?zn}&!_%w_&aOg-bK5OGQ=_tnTYA6+}Vm_#;Y
zD)Z{B20ZwFYF$CvZKyqq*o}qP`A5FVyc%M&yBJJGw4ucywSsGpG>$cw@JfAF@x}ds
zM-X^g)Ir9~ev*x*N?nC?{Tt-i<WmF@;obXOd|@(k<nBoBSLg?+_|@fH4;_pN`PKLI
z4CeiI#B;qmT3op{>e)0p{G7&Z$>9bNj0DidO7e<XdbpUT`X~8i4y_*3U;ow?weTY0
zhEKn&*QYZv@RKlqUD}SoUX`=f>1qD9hDK_ZBuF{zejra(z;}@5I8vI$+LN+5*;fAx
zZflxLBHB*{=@{WaEtb=JE1)Z$6+wAGUt%d#J5|4Otm0q0dkOUNj-Bkdew^syOU&mk
zuRI=?!aYz9-B$|8)nBg7y=)aOv&ti14HNxAix9@jygO#xzf%A~h(on{-5>FN3r~;6
zfyO&?!#xX^F{GERDrWmEPdjTPoa~AV!ygZ7ZBAcqtel9f1DU}%j?coiqpn%ckACOP
zSN0Y^kbRIPBoX<oM?ImxLHaO<drkRD<Fn+-H2g*aHAq|I>3zKLE^1Iz7<U4;KXV&j
zpGG<c4Nj2q<%MTDk%*y0mYJiLevf!|hTvQL_q{VI;rjB%{oiZO0$O6Xx>=lS1&FYf
z%Pb?^yTfb2l9OpKuaReMoAoBZR4BmvR9t%q&9m7oEsyxdHTO>Zu~ZPTndUa_Mpc<o
zOatS7xAacDo|CzD13ltkL~0CNKykq0sI-4M3(V31MPDY<F_-m3oYtmYd}6IR_vx>(
zQdt(EiIoQ##Aat<ax=`p)2=@h5Aj0|4bV`>A#OZJiw<fZ9Q6BH%W*EnX;+ss8#5AV
zD~Xbc7@mM1ZTS9OxXZL}qpXLo%=W)=0bQR-IdQfnN{R;8D5e$TD{Q?5A4UpEylt%&
z?v0m?sU6k+9%RKfN9;-wBpZg>^^G`NDSlNE!%!%Jh&?EKahzA~B%Nj!O6@0-%)f#&
zHkPVwRzLl<-9|tT{<DElE@ydQHk)_LE@5b>Wc1pstK{@zD+XXfMaZ33>Js{<lJjzW
zxdK08Q3_5&Gs9-wtFm5V-x&-f+5-p)AEuQ`kecg&wyJ{MYeW5~Ja9KatM&b*wr$bC
z3x9*n?h^W2gMpN)agF)bxL=h~2GVk}EomDE?77*xC~TC4|C;pXB%w`HWf)*5Z8&OH
zvd?67XlYJ$Ldz`;miTwgXPe#~y6EhN7PM#@!#h5hEq+_#P{`8F<2|$jPIFK5Y%IMz
zUL{Y;SmrqvY6;K&$^;hQ$v6{P<mA}b);2y?OCKM#E{95(Z|JpdSVxU&79`YtZe~r%
z$eM9(B%5aXEfs9Hfn}3*GadOuNd=~?V8bt4`Q)m7MtRK$jzi7I&rN9anh>~t%v#Yy
zqb5zPu!^+YRwdG_6Jx6`Qxfxf#^fx>7w&GZmd1njsJ4`ds3myG;kDV1ZHrNRlkegK
z6;=2(STT3>egm^r2847zdU$h?^OMa}t=~x-Y(L)g2{`7HcP<z_o)TFEAW6?)L9W-f
z*5+6Q06G2Ct&dZWK=P8A6$QvfeM27h)uwWQ;UosS(-$8EI6>WY_ZBuA{WkJtLAthW
zttqs9FOOcWv%>74x8G|GZ{`5y@(k2kqM3K_WZ+oeFoDskNHsRiVS0+SnuQf~J3qeD
zAg$)aBGU0mjb%+9h844sDnGCto3QyV#^SSC*%NoZIf(Z$0$4#28Nv*c3pW}i56KYq
zM!?HM<GqP<>n|HmyjHYwm6Xq4Hm*>@b{}c_%olHegz<A#Fs`PUO>26jUGhgV|Hw|;
z$f~}sCP4OjL$3hoD%*(vAfk0A1>5tvvus^3@fPe7jhDmdrnN*wZq`+1HF<Ti_X{BZ
z4SS;v!`Cz0kXD9-ZO)D?8uDjCd78@2i@#krc@{XTl^LPywdD>LHkr%&ZPd7n82w3K
z(_Y+ihzcItq7ZN`S&IEjTXHOnE24kIbB5SZyzG9o)&z_cuSzOx=pG#1*&*>>BA8b-
zY7Tvyxb9atYX>WtSWOM83E0zx^`CK0oK7lMyVvHT^w{CJi_+oDXD>SdG^{H*&Pjv;
zj_={%#kqi`3GtY$ljN58J*8w}po6TvmyO?5JCMcDQ<QMFZgEe+;p6Lm66#C*9)*$i
zGWQ1;LId<{o$1cJCwE^sh+;7|b<b!&(wf*JN3bRB6G2woyNw73hZD{YE~z&(cE6S;
z7Bk#=!bO_-4%?^ko5^Y6Otdx$*h6ZhSJXqTA{)WhhW#!zzY~4!mi2Z#!2TT~x8EQi
z9UwH94{g7G(_}HsL#<~N=SPoDeXOOpo=}oofS~d-H}o|{zHUpb`<c9jtY_U@&Ab$k
z&lbh$%;}PopOP-`KZC7ql1uu<GTyU#Yu&X4B{kr5^XCt(3n%F&J0`v}G8NMQAVAtj
zY?iV0<;|$Oo*JvZWOd-gp-O+fN52HC(bR6JN4j9{A|{*u-w+e4=3fiB3!68q{B5f+
zdL{QoO99K+-S1ALhiR@+{7qq(&j}N;R+TbjN%D0^LR7ZYKDLW#1sUPFh+GwBBk8nZ
zdkb)3uD4r+tBgji5!>pFnw4@TOfg)Ahn9X=ADE6en4M!`WN5s8$kX}^ez?6jdm!y%
zm$4VZ)=$*vqqj9At<r8A%p+ksy`g#()X`cr*zQ-s4#3#p+YenbrD$=IH((M*p}g<%
zt*Qa{-r=fqjlb3!!?Vaa*rXkHMihDV>27ouhy1IHm^I!thn7s}h<ti-m}(lwPJd7?
zYyJiqlvJCPmA|!t5m-`sHsO6&TXqAuoY%eds+@OH6D4YVC>jm`1eII!vY<M9zfEa$
z!DFxV5Ewm`@;OR&N?CkUm;U|gb=L`d<%h|Wy;rtp#y09$3r`b526UU7RL=4hv)Phc
zcmH=Yy&c%bcoL9xIceyJaJlM#fjEWIpXCV$snvLyPB@RY7F2!g_kykHtHE7g`lICG
zOW-?BVFIB_2QWb)n=wV2k2^BB1Z5ZpQ%^g7oBA~ezKfJ0K{2MPmQ`e>?^(NoJrb{>
z-e>|p>VcWC2epuQrnbcC1?Ift-{=EzR7f5PT1@SP`NfDWAcG%D4rW95Sv`^D>yX9@
zyAk-LTBPXGFg=E|8Z2^%3!_pjz|rjGI3AYCMR&~LbIb!j6AeRh(#M5y6e#tdMex<A
zAd2W<Po>eD89=rJC)SqRbNCZ18Zb4S&Q}IJ^1;W_&qBp<rS+d8$4i?ZZztH?%&~@2
zBELQv097tFNfh7nMpcK)qcFRNe3OG+-d!H-@ywWlEadP67LiAFabd?phH{ieFvp9q
zysC!Jq5K%sVXV?wq@fO<D|;ngAc|C76JZOSB6V|3#|6e}Avizf<X({;8c1X^`5Tzs
zW-frKApfI3U`~nHx7a1PCYXQ^P_FH2I9Ion18uJ`-vMUc-N%?mT#(F@nKTrvppGHE
z1+3rF6P}rj|5E!!V2(boG^!#pe6-n)*OiRNaN?+#-hvZHeq;_T!K1+$lG8&HbVtp_
zC6y6eaudQ)9}_~)1M%Ljd>eBwm%XzL#__#&FJkQ0<g{A#xgRm8DOjcePiJh{{O#`G
zd~h5!X6|#Rem{!@Ibs=jzAV$s8n@AX3PVi;yas<BZ}%3S_-hQRZqr7n(mFjI-lHeL
z9<U0Yj13|@(xvcr785>=>GWy=JRDmhqegAlz+m<K2o8X-=h*)2uhxk?+}CkME<a|X
zV>ifg^4Jx#bu=c{L=J_%Zvk>zDH8Jm?B!fD?OE0Up9ok9S^Du)6bJrj#2N)l-Sq=r
zX6xybFaR(toUPG>fqCa0#6^zU`I1Z7miv*6j9I9K318m&tuvS)0hu3A?`6?>wCt_O
zSneH@HHN{Hp`ht_pdh|mWRHva_W6o!@o$~5!uDITAyjt3qOU2k2BYG)Hx&__r@#`y
zV}<$)kZQmVt&1!Q<Ws>WQEC-AlMNF_3*7pGlP~jJuuh}EZSvPImE?hVWwl&AL8V=$
zIjgGz1GqaKoCo~HnHv&EhyqV%1@snTG|+**2Jn*nna+VG(3Zxr>@sOYNm2j;e)<zU
zv*Z)Ps3bayqx%7)_pXd?;|pFa?*}J<j!w<RyxQ56MEX0SVzQIuulz8d9xh->ochh0
z7_{+b#XHe8a_i4~J+%)4sq_n0K6|VnE29a|sMz<?v6KgMG;KML1^Q<9uM{G2n{ixd
zz;$Da(=WW5d>~51ML^-G^|~A6asx1jw>)2ITa^X(S>_~vfq#Lnrcup)k&Rb(WIHg4
zk7q>foQ!CB>9TkH?75ApR9HfZ`Q;rYIp#XDw9)>utIkM42VJQvhl2*!?P5IbWN)md
zSM@d}%6_zqW?tS&vJ!3O%+42>;%smdw%%j%6W>D4{kO{tFXGyTm><kXqVO-g#_t}V
zqzu<SLU3-g;3lLHM$4_$7$1})g4|~G6g*m~53j$pU1-h4nwMHv7SawW&J9ZE>;O86
z<@b)N=qF-H&edr5sdR;^AR@tGW%d;r+D-{_&?jzn=MYO703U@VqTKRak%4Jvl;wIP
z6-4N0=I8g0uI?S82*e-_Ng6G+M7OOP^EYsJXHS6IwJMD?YrWdjJ{tV`2)s*M2Am{(
z5r2+32dU&LRbuW;GEkyJqTk_h1pen3<-U`fZzt!YF4-1I3VnQX?TQ~y#F-c-UWO7i
z)>)(Y8!TQ5W>MBuy|}ON>XKtKCYT-4CJ+3R)O7h{<Ca?lkfI!WQwU>z+iYuL@lMvh
z#eb@kzPBuxUpO_xI=RTNOmoop6wmWVA@AIc3Xfh}4paFmu>O4-*SkUs^;nPBiaA*t
zICeTDGFX%L7+6x>r4#!^1rk&=BkdAfaQdwszLtl7tE7~M8J)(w^^57lxOw-`{uSjx
zog9R*{w6Ty1jlxC^%G$1y_Nm)uc8>|3ipVO^Odh)_m}YQGsFQEV$z+kwFrpLw@K6~
zaiQO*lAC`nOE(#f?ggJ_`<E;ebDJLrzTNn^5l}hwY#ESzZ7>|yHTCeTA~t`fJ@6Iy
zBO;3V`%g=GSY_J3_7*k<q`M+w4|=i4&3U(J7u|vneTRVj09LqIprkUW0;L^T5l`ZF
zRI0Xi0meo-_2^WtVig^!Oh<&8>D@HMCR1Tc4pB**lYK~*mDo8-fH$!A&F!W9J8W{k
z;n4OVqRC;Ba!FjVW*j4F`eGMA@nO32{FQLJ1v%M@>gde-ti@5;Q}@QZ2Dsr$`iY~2
zW|WB`b^4Xuu5^b4+S<1Dn24E8Ho7N-J;C^uu>PtLdUuB)=qQQ<B<z*;Y<(de-LAt~
zb1KApv^C;Q<T6_1RdWo>$96K`sDk0<!kAF{JkmqCbocUAanjpiG^*7+h%m9>!}(aF
z$l~|3`!c#h7Jjgp_RU~M41mZU^>uhR19^zrV?V?vKkD<cl$eqH(hWY)b%-Ky$eLhg
z_wnO+9%QeXCXS0k)RC;2`u7P3wS|-FDcT`HVua0D$Ij4Nhj_F#@Ep07#fK5&o&!CC
z{te3<!<K`4sM&m=GrLABSJyinL5;5jZ=bD6hFz)yz)IJt1YvbSW10*%=w|Y^5;ec%
zx6X%D)s?gjVmOBxBU4X-{4j&`IFPa0hks;39?CS&-qyn9*Q`wr6V?10yobkBSPy`n
zSmk3*fWQ<Xa-)<9l_uhH+9StAt}dPg*tANW7Lt$T&u_K5Dp^37-{zk8t`qV5BX)x{
zXS5imTK1KCLriodsn;5CJ|keWJTW>#&=%5gz}V&RL|rxTed&Aqe1Q4v%jp+(woWc3
z9SOZ89(~mh0-DjbtXsRRpz0vKVsL64!+2wz^56GL#)FmV;w8Oxq~PSS<~&v-a@=-|
zb*zhLF04jl%`$sXHH_v-pooIc%fB&LnS5LOZoQK~OsH}-x@4+L<}G>p{4Q_zk8({f
z&Mf(?9XCY>R##N}`>TBbO4EcTao&CZVs&GAx0I{>BFU${8}214rlZ_T&`Cone8%i`
z{PrPQvyrm<0+1wm%l28Rr$}<W?>LLwdI(Np>g7E}^2R$LTeFp2Ve|IxBY&W~<vXv|
zOiHl+QD1T(T{Q`Gi7L_IL$47KJgE;qSjfhdsH&V~O`;itc`H$=CgA7{UvRZdGCbJG
ze|0fczGK;!n#_&eM`ox5=Bz1sqbKVjY0qM9PT#<$kLGgxV^dzwn{9=chu_e6YOx3?
zYECjHQ*d24n9Ai!P9957%1BK<D|np2FgC@qFon9BP4Rw0<tYq-o)b<IdzxQX&a>%O
z(X7JNm@#ybxX3iaZ5(JTW7PjRIR8t*qMhc4MJL6>I8BW>`*xOhmWoWYY&ZM&q=bDO
z7E1M*1qh+(u%<P~b*l|QC!MR3sG~w>+aB)##wj-WGuj6TW`W>n%=7+)nLdZaHFAGA
zA0uHH_~2z4$QRdmtlp5RMILEm2;b6@zM({{2li!+e@kr3uo$Hw+YtkK65ktnIAV_A
zAJ(kfq?bhcJ(_Wsj;-_hdr+?0?RLu>$0ZZtQhcBY2_8f88B-Oehu*vC<zS1sZ<Qtn
z0=Imr%e=6?^ebD$_RzN?)wI%ohlIa`+blP)Pe2O2NC)4`>}bxSY>Xh~E<ztx>X?7X
zceIySWZZ@{xffW3TEx0NyZy$Kn&p8(#&>MRJz}~G!v^L{PMKD1BRx$LCB)|U65CE^
zKCMpLy(gHnhFYcxg$E7ORF!b+teuZTEWEeppWw^D5g0x4N}@^(F97~v9atWxK)H!s
zRQ!B<|MA(*>&Ulp3N#<&`Hvi<OWf)mvwk=KSG_DrMR57TZscd*U8S%6L4<41Tj125
zS^S#6T_TbKR_Y;ay~Jq0prmKBI`YGX2M_y-!%=bKdJoX}0bp=LFt#K_EK<G)l!pA+
zBm533=Z79jlW`tt!#RFMJ+}L-s&92^=ivESkHlNEsVb8;@KVTVY(Y%%_)@fiCAeC`
z{0*DWx0ShEaKC4=LW|^A<{jdqt<01wkKN9X*dpUUuPw;uA?GwzVtde^1&uTyu*4Ja
zrJCy9R^@qjeV8!DgC>_b?o5y^6FE<h({#LI#yFb*(bD(Y%$U$PHvxP8k0h!;E`3bF
zmy(dxDEaB#`Z!9^!;OW1Jw6EGwjL=aV;auIi{ZJStjnkYNk4nYTGz$#*1Yp$u8cMH
z_DM_4b7m!5C-N5Unr0MF%&c3>?3ci6isF!JyO()wfrtWD+KiX~vT-*BsbJJIMjQ@1
z$zJz9Zqdcw2)U!rHo4DWZ;LfZoBoQgFHHuqJ3sk3@$>$Rw)Ep!a=7XBrAm6UZz<gP
z)Z;XPv+b-4?84+dQ8FZ9|L6fqob<AyBm^Q@K)cr*-`#^0T1eBvm{uEk`T9^UM9j(v
zQ>vj~2+cLd(|X-afuyzDhC8h}m0g-$XqIFZ8tI8m7sVEdCs*$8a(jN#@|9VdxmI`G
z-I>GPd#WL=epmOud5s(@P1$p?MHHpaF8FLeSkws~pdJJOiv-Dp-vK95yGnNzxeGTA
zR7&2`#8G=A*cp5xoGR*QlYcoCZ3r#BVf2&`FQ@=q5zoc^kLX`FKR!o;JNK{gDV1w5
zt&ZnhBW2CC-W|9iU##Xj;)O9sMZvkWSW*QoRUf5Jg~6~TgKdmB{n>_PdG772gXakj
zJ%7_DgqN^qQEAM&nGe=W%hL<WdNX3)c{w*&qaAnZ6!E`G`bhAO7RkriG%ZTW8cl3<
zXPU;V=hZRB6oc;=l8^Z&&nkj(Kxz6~XCk_)s*uK+`gVOu(rW^P`%Wqn68_Xn$-r`D
z7@orKW6o&es@ceR)&$%c4W<z(GTTH<v43y4Pmw-m4x=Wr7OWXuCqNY(!!SKK(E1Eh
zSznWiY}5`0B>BFrfPS6JTzhu$CIJtC2@#BBG0<kIEqCX9_jUn~_;dy<^g6T)gE)Kr
zCmq<}&;U+hcf*$fkCLM!<s5RT1w_BK*FIka)A|^cZAeB`ii@-~g`Gn8<i*rJUDAIc
z@?&?Oi|YwVseC6Us$U4e=(;5kp9?0fqWba;;{Yueux*W%*=Oz_Z~f~|$&MRPM}I?O
zoy_RPf1Oj9PK@bD2KUo?6^6RU&~x1qEZe<qYC2CFDAPeeDeG?lU`c$MJ!;Z=IG9o6
zSnHKfKK0eJw&JB#hkOt*a>yq^^1=L)R^2&DXD-;bp2Rx0ipo2>n1Xv9rxC`PmP51P
zC_R%8x0}Sff2OPJu4|epdnYLvX00#%1+SB{^`pA+#~<*s#j3a|H&<W3U#ZohJ;|$z
z3WOJLTreNPrO=E2yXuI2_*^?AKj_Dg!>Kihq#{%n=3_2&RqCqXMsS^001=_&hMlgv
zUf9Y&ZMkWjb~}J>?YIbV4EP^P8@2_Z`(Szk)Kvmn_@6AT5ZTh_mP>ry9XVRrWuz!3
zr42Y4YP%7SZQ&(tK@NIZOYE)#p|%r?Ot8?m9f9oM(rHW26$V}zcif9prnm$|31b{Z
znrm@f)P%9YQZ_Sx$6EuZs4QoE>f<W2{kcZ*^rW;f&OT|E){`MQf!;cq`QT5IX-3$E
z@#0=BsTd)st89Y#6c>(zFU#RhIuJe|*q;NcVJcxX(re`^fIcGVohCJ%z@yZDhaI5a
zr%@X7lcdgwC7)13VkTxORMzb<`%<6uO2aC{%i>Fbn{yvB*l2PMIJ!^VdEu*+xf?Jv
zx8LzMCyI1gd}9b)dm!qP{prY;eLu*hNZhs?!Ot~+_jg7B$6YpJ*Oo|DQ`DjNh0GAi
z+fTrHPb*-Z4ucs(MpKj%gIN{<9Rm>&F<#}<XBHB=Fe;;7=9jvhjKfQ?p{4Y4dnNdM
z+D?||)?vfN-GTrY#gF2wva%m#9(!M|XKYAUVP@Vub20CU*fkVjg$v~bxSdq5@`blq
z6KAk5T}u})udW`bya$sH_`14{yPZ7~hKlRsT(CqvbpA9ES&58V8GA)t_4BdCibn9~
zEh3)MxQI;UL%Z^7_UoMGiEK$tqN-;DId=!#OZSVkA};qnMN5g{5zbtvX&CBbdCQb?
zCJQm#&9QjiAtK+x9TcIYJ(M~}ZdK`VG;nE>BPj8X7;48CEljk0tLGf1V)c=&ETj<g
zdd|oFiz-X+2__eX5wG(X?@-T~{Dpzx7`5E<Rxm;3O<mc2%-kA7u?-)@wiU|5PA<N+
zcyBx<m~EoWSj2jrJ5(UHj_j}Kv%fBu9uj~OnIZ&f+UCZSzBMZPWHiCMllR`j*Ji)c
zjd_#6`gI^Gj^ee#^4NWr29mv_=cU&wy@F;-Za}Ws)~y-JsS@R=Z^L=&To>isC}!mp
z?-Uw(LQ(~7ma}1AoBS7uIS+5zpgMMeK+egmIn5+2UG^bJ))eA4cC$V!k6$SkEgQ?p
z>hg8$$I+7rhy62J2S4}Z5?T&kQTNgXkcSDwB(7W~+0&hI4f{Pi|4@ha8Qov-<n`(c
zl#Voa`E2mqXnqWT-<*!VM5V1b$(P8QEe@=&LYyWJCIBV56nGn~tIs-&AWh!L^u+~M
z#iK0KN}gR9uL8dVk}-1~meWsIhXwT?-wi~6E_cmm-LMBBOKdc2l9<*}=LVy;ELL=;
z{U8``!lsX4UTaSP+(YmYTJsm=etK$Yr=y2&wmGbai+{D{Jqi`43Yd;e@`ph$b+cs|
z2r*4r+dBn^kMid~vCNH9QwGvA^;tG<(Ru50Eis9oTgh0Y6HL)4Z9*We0Vr2(^K3Bw
zjPT|W6Osn{>;^s>(ZNNxYi0Anf)ot3st4xLNSQ!kj^Z~=s8(6~MMt6E{17C{60kN;
zJI5S=H36Mq)eeAacZg}lac>P=vND(Km*5d9eT)H#oWtrvcf4>JE+}1SCRI;63f9pv
zme%8Z?D=Ck6l}Pz429j>8+$Rx+ndh8&~?~_I!|;5(b-$i))wqlzS(C%9zUB5k3%h{
zbOp8)k7<tcA993q2aP0|HvBlHU}E=!e&|OkFtfZ8e$Wcl2APqQy1d+Cz0!L0@7h|?
z`u10-SD~)aw}P<-_9y34KBMVuAa7u(w<It69$wUWIF2VT!EV*zgs|i~{)gkQ<lbYR
zSyj~SZdiZ|U;tyUF+<C!av^k~yg-3#=!1Vo`IwWyau*`uC~E|XZ%e2<ZBMEu&nWxV
z1!Lgmw^9zw+E*q*$+zJ?SPfOr@5Ws=eWtr*x1BN~8Q}({f=R7vj)3p+oB~1O<N3uY
zp<{t+Yu>Bfx{{uhM+$GU+r4pyi^nI6Q0((aqQ+qJVD!e$sh-E_*r&V|Pa*O+{-ev8
zOJ;AHyPr9s$P0P7Gw7U)zTc*-k#`+-&pq;fD&k!e^h%RRg2itBiv<8oevu(8zxzCD
z{g~h~to{a<GrfO!0m}3V&zlGR!94ZFV6B1oM>ob`C)!^-7qi=p!=mFE6gqu!svcyM
zjj6mwB?vzW6Xw1~x>4M2pHo+=TaRvSd0R9f+BKq^uhp9+9uN~{k4x)jyX#=F80J43
z`jdH2tBVx_8Bp1=oP7k|>br5%8WwTR<N#Tn;-ss^v7w*$WjX<5y>kZaV$MS`SV-;Z
zGqN;TlONSfeLuvN6|aD?#vQeGp}yE}gIZ-}dFkf6gflv4GZTYYD437N@z*g5`e?Uh
z5qBBr5r4w9c|sCvfq=Ni7j;GbqKLArR<u!OL5WOqrU^KP?5&1lT%IB=nT@AavkhG{
z+=NQf=AuP5T$5OA-X_+nKX2o~WRL#&CQn`b>3^IR%eiPk9Q##Y78)NBi9#$FgOR;$
z&Ec~kBf?(EtZm@(PEVB^92Y7q5GXo_?1P4~9i>dt{UR;p@LJ%mm+p0vhW2u38y{tN
zKg+B6Yk{ML-r`>^IO9*KnC9)Xc+q+J9+kIer-8i_eCUcrDSVDbxTwf{J^&Xoi;dLs
za%@_~=<9Qk0;Ww>_B#p{YFJU#00Udpw&0U4@ln8bA;8^zroEbAi^~cJRlbdjN%u&S
zJu^hAiHjZX$37MAD>$ub6xMtr&(+f<7mBZfupbsF1uiW4)^Q6eZb&`uEknx7+jNI`
z^njs6@`UdpxXBf(QA8?ypDD}yKTEv*m(GXk1mU;yLo`(&Lbs+{6-`2Tk2b<yWy3%f
zVuWkBtoW`S3IdEtVt?>eLCM?=G5COzP0Ess#Bf<xAL;v^BU7MEB!4D4@~CX4fh*%^
zRstbp4Zz9Ayx`KGC9aFYT-(w!e@!otOL``jF|drzVpSZ5lia|4(i?<eOO*V$tP)cF
z-TmO4;Q?dTsH`<&=*c@YRvn>bysTfrAMQN7Scv-7gdELzbR)dsEMVYde(Ya0(b&uD
zbX7tmTJm|_eH~z68vWzcJoeZgL8PfWe~c*SOO*`0RhXweB#7i2^*8Bh2V6n>TM3ib
zZG^EVEdd9Ka>NsROC3(>4-yhBR3gl;jU)|Fm|X-DQqQ{sP@XqdbjCaV$-+UgbZ;{1
zjC^Gl_p1fHtfE5=$x6~I!2&&S4M;R18$F7C*h~dyxeQ6;CTWzpcNvpb*!0kl95QUR
zhdi<ye+9yGhG2%&&uW)Ogb4Q|6T{7q^h(!^Iv?^x<m&0b4qv_Ehz{Ru#2^dJ$kL)<
zXi8Enf!*1*=L%7EzAL>!simymmA^h6it{17!S+RRnK>92nZad{N}JcTF0F28%2^P_
zZL8iH<6`PyaM9EkT@HQyEGSp=ZI0J*nV7dfRif?v&oE|aqL#&KRy2RC8(kl_)uzvJ
zQ_zLep=_gUQmXJS-=Y2eS|;tC9Z~E-hZ1Ce^uo>H3f1tFe>h@Z<i|q<7t{;pJk7r6
zleEOYY^57@)E5k-W>~Zhl2j~`)uglr!Y~+jEoSX^?H=8yET|!oN4Qc{Q3Du~cWE$3
zo<Gr*#CuY@_a{+Y9C)>zN-_NylDAT36Axb_`D3<7i&5r@X7{F($I1=CCXBySPW1dd
z3!d)C$LLfaj9Wk0BOst2>CN<hpT26J<$1XD@NFP6w2kNl{dc8IB@*_;P>IrmlTsfY
zvGhP`xdq;K3BFy-%W>9D^&&?DL>)KVbp8gkYuxAb+i<B?{HsVe+92+WtkSwz4(zO-
z+9q-xM98G)=sM``eG%##JAo$b0gkV_OjGBjEvMykvfut?Voog~kR~w`pm4*7ZlS*!
z>#n}@{7<#RIASbsV0gJ*b(Y~nYZCM`$`QUj#!LFIayR!s%3avpcahOGnd(toy?Bd-
zYx7uW6~#c;!v4`?6ZmWGOywXk{srxtf3=tp?^*Sbm5XQb0qX6uE@l!D6Sgq7cbgs+
zp9L~V_KiHfV#|_hfW9wsFuZy4UbO+TvtB3YZ$RQ=-k^91dakNln<brc{zg*s>st8I
z+WlKr9PN`jSKpdf$JQG+Vcw5<N6>R$VqOTl^t-=dzaa-4T*BQ^_1vTplNy=tr|<`*
z^BbIXmvB@4`ZdiaY79Xdttc0PSXsO-SQ#uhknuAoAn<0*5R|BpX^-=Xy`MVCsl*9n
z^OAYm)SW#}uYxGcFtE01iL}fxpvn;B)=Db5WDd%2g^&d@!2+=yP*3BqJ6cb8TGeN4
zu-;RZ@{nlEbP;9jW0@y6M);Rb;y`e}ky~e08&WhH7F)lXyEcxtspxGqgx(6()ZVs3
zOc19I4<?I9Ov{m_GbAt|xac)MxHT=frXvH27W4@P2MZIipUjD|s!$-L+HlhB)l^OR
zqG90vibnIT8r9#)G&CM|N*mFxnRKfi!-W1;eJPKQeDJ>%+0jV*!9vsPRt4CDoH=YY
zL%tORZ8Lh0a~&9qJn6*rsTZ+^0VwucHXDhxDx0g&X}?eG{+fq3HosbH3<>ochH6rT
z%gYsT!Sd^?R$To;tti9~UZ7{e&$0P(%ymomf%@nq1Og+03dc=}$4nFGTjwv4Jn_OZ
zJiOv9Zb{C`XynKkPh5vld;`NN1F-U7OMx|uNP<E`&^^?vaa|;j?9PtlMUCqxbIb$!
ztg?>x!IfVVI_w?ls9j)fW92gWP(dS)krIEzk0VG8I;8Eq>2v+fnU5H)ZC$v0qpzVD
zdMwEGiv@IyGNn-$m24S{=LTOsNW1<Juq6jzX{-C3bxOpxLfl`?$$udj5YwX1zPRO0
zN=UyK%Z{kDd%oQ&;o7PnAhdA4Bl9ObCqQci?J`eWR$3qnCtd*@?ZfKLef#0ObYGGu
zb$BE#-JRA3NL(}m7?Nksn`RRo<gF4`ttFJO1yLUxYB_clm1Qqb3Nk3x;|*M33B?HU
zD&$cR{$gt`(q(xjrz{71^$+^92j|;~OHS}WfD8R^^?nKZfB*v<|F|iFM2lUuLk{mx
zn(?lEkRx<Ki^Tr(vrgeYW>~EMnR|%w4Smc;vJP8n1qbEJO0xj(M3UW9sepP5Y#y1Z
zHD%C-`%S(NyU`aBD9|`2-UfP_<jMIi8TsnxC{i!*pl@D{BQ5&#Sqju;k7JzQ7^&^S
zpv6PKWFrk}@<mD37J48M!~<+WCh7H7NL7GVN`L=5GcqgQ$l#%5PYYx2a%pB%=j#;3
zZJcBIIZ<==DRckJ*t=E%a&Npo5pM)C;>)d^eUI9K-_22#TECiJ<qxss_l#<=uh_Bs
zJ1u8sZAIRFOg3o1E)uG09W_iK6wZooG$(l&u=z6hSjbvfw#!=bd-q4HAj<~|^{@6H
z)EB@Yvof;)R=?6;je-`sfufBH+6qYP2Lf}av(IXnfgTi-1+nIsY52;Ryk0-&A<x@v
zAk!_A_?v)b_KrB`<$}!eLW@MZ3r@jOpFp}+VVS3Hnt+K&vKx!eu^R6#5TiVEw;pvd
zx0*C>vcu38d9t79&!6^aPJXw|D%?fouCe*>Lw(uX|HNaF^%U_8G$1(~dsME>4UXzo
zPEK|DDQQ$cQz0onME$@<fb&aIS~%MzvS=0Zms<`-nOzSmvDtcv_7OhH+BC1N6iQ}Z
zOi`dSn8~{8B;-TpQp5$W6~l%o!&T4blmVgFq^UT$;3T4fy}+yHZO=3Z_M^}GwP|?a
zEuQhLQW%hJ49G2x`E=Xe=D_P2x*GW0IcXe)VH<|;Vh`B6ZBcB5GcG#&5$nnt@c83E
zUiKUwRxL$UaU^$pLXH}ugJ@X#7&Qh;Rd~}rze(j|Ngd{%&Kom~)C(X{K(Faxbuh59
zaK|1Lm!h3H6=t{!;-GfYSr~pZ69~9HS<*xHZzO$r=VE5_<b6>rr$*NlU^|9UAy<ex
z(>rUNu<nY|0oRJj*Q5)JHffgJi{>$r7Yg}i8CO2DWV?{vZw<=jSp5P`JvHjHIJ2Z&
zXv+`s=R^(Svw7!+ZD6o?p^%aMG>f;#0*B3X#XKG1_ahsGjM#-SyR_DT`RkdU>z8X4
zkYlH4(c2c_`0wnsa}=d87IK-~<gSQE$PsfL9uxA&vvzb&qQFh*c5%aSgg5gh*(03s
z)GP7gv8@4eCK_d=5(6tTK8$oj-`w058D+@#d7W!uc=#hu`B&3!;_pIBVN#<4liQf5
zUf47^#?<5VN8J9%iV<R53l)UmBNxI?642`%Kip`9mpTvR1C=>SA4YK5Wd@NbptB7X
zcmwg-N-<}dea}wx`Psva)lEk;?gf&B6XbV9>D#_)CnF>|T_U%t$KUm9-g;5;>v$(E
ze%KaB2ttE+g7gYW#((E%S-dcBnwsI@Xr<$XqGMv4ok@|)lODf>FN+{tHZ;tBbSBbc
zf7@Y-w~DVLy9*uk$$9)YnZ}(WVp+gG2P7EQmnwh6I=qO7V+*?Aqi~%SmynSkxFWO!
zkr0bf`SS%`CchPLs)}|^Hk{P;KF+DVPG4C1=sDzqE10%(la6VkO{{CtQeXQPA4S=g
z;CGcN6#)vGk^7q(0*%+m$|wA0KEOh~Oty$5PSAH<pfPi=GWyf7j<27|ii*;8F9%15
z4;WFIkDF9Rq|H|!4Teq&jiZ0g*pY&T>Fnsm&n{}jyuqAdQ{8YSxz-jU^;CQ_?!~~N
zg35&3%<*=&4mFj-$i|p8l8Xl(!p8Bk7$8yeiNRn}d1n;}-r_KPm(@W8cY6eNE3ME@
zbl*sQS<s*{6USbY3Te%UQ?ylr))$kU+K0Q?ni8vxrB9<Ks^&e!Lr#P1&PbXBf`f^8
z+>l^*Dh4wZh1#zsHmJhC-T|2aJto@Ex(yMZt8=U;p^y6URmJ%Y$M7>!tRfL!czY}(
z)&xm)bX)lw@Y~JOoKDCy0=lux_IZ_?4uT`S`R9Gy33iF)QnMVR@;6eNeSdmlYH(TD
z*sLKU4=yZM<)GW=@-%4#isjaps>Emo<Y#P3Gr#8udsP(s%G~Buu83GOB=?GeE1I*9
zb_cjGc(tZ&1#U&&W?x3Zfv>ZVx&6u1Tx<{JiKxB5pk%Ad?KF*dKIh(JD&!Es_Pf^D
zMR3x1_qdawwV=t)2}73=X}t)Coi{*^bA})+_I{TpsokI`Q4dS7G;b{rZNV|ZdC(wy
zKb4+XN?_4=F!%zr+c~k_G#&dIV>>HbTX5UyDfnu!JO-odv`nS20yvUaazcFnNG^Aa
z1<=L{a=Hx^%KL*?2Rty|TTN;C1^3{@Y2fovzuaida$B!sEpK@JyOh8m#c}M)bO^JZ
zVCKs5@@z3eG*B`-3?C2n`lY+Hnb&8xVm>AHqXy1CSAL&V+CI8Lt2K%p6VjWZ`<LGI
zU8LIR-mJ&W#kw3Zvi0i}(T{9b-7KicXL-1jnkHC8^k<I)dDBvJVZb~A)^L|UG*Fm7
z!)*$O-u{_!P4w(Q^juUOGx-AB5nUgm(IdQ%him8lS6y1V3UubtsWDlyroy<#b*#cg
z@&_G=>kD-8;p92bMr&V4#Qo0pVOz5w(0DRl*4q->vSXZnBRfO#F`GyO*qaU1Xlu7&
z`JNb(nx<=o&8^o$2Sjim*C<|3uhOM+JmN@eXD@Cj<!nUL%j=|z$CsT$QN1Q#gRZE%
z-n$p&z@@>VpJICwh4H_Tp#z_~;&<za&3rs{{z4pdP08>7ge<UU;Z?|fbR!E4_eOpi
zWz*p<kf*kkn?0**YLP4t_y;mW)Dt~5$;OS=Y<01N5M@JuAc#3!3)St)NQ<bECpQIH
zHga${7*C3Pf-Bi`8#%YV0nQhC7?*?ww09;VlO3Dk=s#Y9PSrA{ZXK&XMp8KG)fFgJ
z$d)1;<PlBn*BCNn76|{<AIHi>A8VYS{3KujnhVzgDat1%W$<Q5DA}}XPYCN)nCbPM
z-cRG}7#shj?C$>5$<=o85gF7+2Z?xR6Q>?4E`OagO;gl?^dVd!>uNgXp*1y1E2Piv
z@MI3+EtF}l_PzUZ(Pw&f))e4We%}+rMH%t7w?FB8r~~nElx}exrA2mk!t=~%x<Def
zZ*$LDyiJ)~qu$Zg$gusC><wqtwE9Ll@JG?$&o*nZZkiu8hd3l0U3DS)4-9g$f8v!&
zl(G&ujNED)1tX(FfLFiSIai|7QrS;n^^pe)cEea%I^#M{sPrdNA0+svxL13MC(?Rx
zGlTw@cIbzilo@Pw8ru_~f5M4hu$kd;{5+*pT1RDzr@AnvZ6z|05Rp=>Z42@i166nm
zD=cfA-AZPu%tB@jbp$h8xaX4J$HtOyL4Hb}WyJ^*H9lQOB<mjM;DNkPi-$jKfeVS2
z%P8GoF320}0$#OsHiz_xezN=`%`<Hy!%cCKXVft7u0gMUE<zA&*4R8;o{wXa%nbs?
zK_DfUzta7gpjUOmwZX;VT2+IC;-%65xvU7mwnt!qkE-%d&?T`F+VOkS<cNenHt&1X
zSLthh*b4ThRZ;`eWEOuue|mp|s&(gv|Gv7;g%GNcvQj4sI#}%g$RPhz;7`>^UI!T2
z3Io4GBXb52BbDxpI{g5dp{%jKqt|=vxR*5f&dc6$?zpX~7ZpL)o5(}r@;zYCPjkJB
zQvo}Bmr?d3j~Uya;K-f`S^gj|^BH~)>!ENM&2g`RlhQt@127yL8z=v)QgL52(Z=IR
zef_oYOq9}u{3HRrvY~GD%D>|{!-;3khO4>AtK0=M?<X*=er04V0Cf<mir5b*V{xmh
zS<|g6x{$Djo$yqamUywZew=*&GN@$dlkDx_$I`9ra6k`I@eVbHS7R_7_I%fot((gD
ziNoR{Z=7Hb*;ab9fp0;2W@uOFEx>{|q@3&yG4>OIW1XCQ;dq;cjdp@dl77XQExM)E
zhcc=N<CfzLuYWsuc)90@>5IlFQw00*?Ult$Wqb^yFnB$d+cY+qsbTO`!&v|8Xwl{E
zLW8q=`hbw(pb0G;D1Z%U7Yk1?kN~|iL|k6pi}2$(e*D73*B$6uJkb)<Q7z?|gl37%
zr)KKmf=A}TBYR!bAyn}pp<4czm(+If$}m{?MuJimU5buW)rE3YvFh$XYS#4(xGjX{
zfYAl-fQ`n@iRsWq;x|iGwg;&7?*~<d4dd$6sdEc{tkRmmoonqI{jc)BJ~cKtMI_sv
z8d65zYvoY%GXJ&VDt4~7)znYdo{{n-$cC%W#!jc?M8DheKb6oi&Qgz8)6knZvq~QN
zF}vFH({jOH|9u06<h2J7F6M5K7|ism8SOJ(Mb)=fmhnxtWya#6fj_95Y?I~vXDNqZ
z(^%F>Nq+s?jb8==_^JK5_Ro1&TeroF*IEbB?)%(@`#gL(%4yJTscwoMB8-4LlH5RF
zJ;t+Gz|>t=znjp|VZq|VGke5PsmZol@5M6~liXVQO7-vDF!;c*^p~ZlBck_e-V`Mr
zP&)s^h4SaYZSl@PzkS6qCui5PA#|weFz2-v&5JT$!vvfloN*9z)7AnOf|^=O43Tuh
zmXkx-$K<(!TLV$2f{n`44&!@KZUPj(*2@{sH{m+OhLH6LYACzVPALBK-_S?bbsmvy
zh3qd#rYMLagP_PDUu2M}dp5S4Dlnu*=UF*o`vZ8@ecc2o?>x@e9PGobUsY{`0MwO;
z+mSRmj}^Xyg&pB|0J`SA&`Tm4Ko^P%LPDyd#J{G)1PcqMQoZ_|<=hgS5gz(Uuc08;
zdc)IsBvRl5C<r-26NkrY%Z{Np6y2{4h)UmvH9y4*c;51xXLc>rjAhe*maaSay;ru=
zexEr5Np)ICouSif86<Kncz!r-lgGX+%tuj&YDIKx?%y>QHm9Xqnx<B12%221joC`0
zVwzsJ)&!_h3k2kjtMNj6+T$$o$n)oRmA#Ox(Cr#5LI*>Uk~(PsSY#X3lnAdpH3Dtu
zn==JA-z^;7XSP9v7ZzBXjTCM=8}|X1`1;jckTU}BKw>xGq2@`ge&zP&U46d%%Yi;6
zb~({jq<DQw5>#r+MGg6HL=I+H<v^6iY(Eg9O}+w08<3Y+E^~*8tdcM04xC071#cba
z1Y-X%4#TjW7d!KEXm%)#CTl>4q*8qbgC>K4(D9aRu6bNe;RNiuFfP16`@An}zoUPd
z>g}vRY9}?Zy%@O~=F%h!4JR$Xpu!5}m&J&AJL(GCu~q>^d#0k;!co3d^b1<4C<nH>
zN?L;%W%V?fZn;KN!h1wJbA{tL&xP|7*wXi_r>RjNTb1^fS)mHA*GY%D?s;-&o@G7F
zK@`rAT^a2A6=O!7ALKaWceqCn(vQ}T*G)oqa0!tWz?OVrL48=PCIuU+bg^KVNaJ`o
z8SAh$0-yt@TQqjoTNWe-jCRa`%DDX)2ot+KlX<WY5OEu^HD&}~b<%GDu(fSFhAI<o
zvtNOdBm}G=$#O6p`o8A7s#UhU#E~uP;pgv8R-x|LXus?QHGq)J+&``gwMHCRW?qE0
zBB`{~06wGPA*VcyOnp|k4+KP^=SMw-9q5NDXOCXxu|g8nmhgxU=pF<%ISge56!`0S
z9^@{^rZ`pdY5`0OHn5G@xn6baQgu{=>tfiY83F0Xrf=`ZNq}U4nv~bNJ@31;(Q9OS
z?gmbDH$&YuEP1$h^J+`B0zYY<KLiVoUS<M_!NsCHD8Am83^SeBY?E#K5+GIG>j1Cs
zTRDGy+0LG81*+}?V<rF=ETB~jdKzZ^`!Vc>1ZesQZ7vKq0ve*&Tl}7Nt2?3U@8bhP
z_1}?xS`lcdl<ql7ho7Aw+^MN6vFOZBCu&4!feD!R8InZVn9Iw2y*(WT^|$EeB!zb5
zvrPZ>#K6-JlOj}6u8Ofj_;;K*sJQ^XSFFtw{^cz$NY#8~sJ0Y{d0?ZA)G(<$_~JBl
zvNk{?C%P_1j#f0#meCBQ2@}KjTV`eUdFxyQ<gTdS#&az0Gi>PrT_hXbz#;Rkz>9m_
z=0Wy$%vKi9>#tB6){a;5NbQItWOr37xuKLV&`wg>ONekKhHE8d(y-9_2Rw}YjA*Gn
z!ML(<htAW^V+w&<r{8S&AFtN}wP;GdIsv*_Y{0N#&(52XWO6O`_e-#V{16w((4BCe
zLJ$}@8DjG8xgu~FmG*8^;_;m14|q|v7*{T`4Ac+Dg)8?lBMNjl1F`Q$l71$WQvvnl
zSmv^9sThJ5qLr{BqrBvB2ShkxLS0m#0~C+bacPkBi*I^oJiGI@vTva=um>1tt(|{~
zI}lG8^hv%ffwC%OgGM&U5w^-x*8BCZQ78o=06O9D`Q7r_1g2Q8F^{uYG|*{)Gs4hK
z!z0Nusw3(pBYcL47{<tLmY41|D0nu|L>y`H#aTH7*E2q(<d^|ER2R(|T<7m~e6QkW
zDxPK*PzNg<q)A)??TSb!lp^i2Rc-IUSztWYL63s3$uzhSn>n@cnKU?J(f;85pdpkd
z2R+~Lvwgy8L>jGKo--J$B4wStkv;1735CLkpE4?DCS+aA=~e#TjZU?NRL-h&?4$L4
zLnS#XI1sNRx<eJY9f{SYBM3I%@k$xNE?fhu?+E8e-|Mp>Cg97a^c`+hY;3r~yde^C
z``vAj;v2^38pltp$mafQmRX-;o3XM{H*p&!SYBC?ugsrdLkdR$?`%z+uV*Ue-l#ac
zf`nhFq>3Jb8H`0{pYIX|-B%;|oZPF=t$JJqB0s-R`?HvSVK&2}i78653fCRrov<r9
z3rLOqoz)Z#b3MKC|8*?nZPR!Z{o=mQOE{vb`7^kyeMgBCY(5#oti_G7%=WIRp~{kR
z4!#KtP>L1HtZ#o)l(bBx)ZhY)URF(l@Be7~un^Sus@{vBptuaIyK4e@>~qGl)cVrd
zz_%__&M&DDbDBdX9!rxgoKU~=ZnXJldW+#F9(J@gq>TUVXmW>U7><tjG9gP7)cS)A
zRJLh6f(+Yi8|YO2`i$8Ct*LTFt!nZdNNV~m6B<k=zs?EkyQ2Xazykb;Q4<W%ymyh2
zd^G38EY)3QFzzcnh@OL-f=U=U4r8;!s&Q2c7}nRC&KVNU*K?Dfy@(W2kx3JV9tk9j
zu*dj&_N=?`8;-DQgNy`{Nq#C4F^r}SbiGp7Urv3(Wit4LlQ9caWmK~r3LX80nNdNr
z%xjY}ary`e_i$TUFz^Urn*wvVdw3P3_?;PvCPC3ihXkNyd0i0h%t@3ow#I9tkC*_z
z)J;-Q1}4o%vJ{Ie-iBlC8D^NODb0uue*!rn+i_%RXD0%OfkTcX1t9P9ELiO=QV0y8
zSXZ~Uw#7-C8GER}<`br|2n+P2fp%21p&zDsZLPK+-ts(`C{hYf!&H3646)d8o;HYL
zy`XkWAS3JCJ_`?WQQsVv3foM_dTN%;M?Tg^z@p&K;bOn-q=FDEcE_|g0z5!Q=wg+N
zh%Q?Ijp@CMCU8&L>9`23$SrYA<OB#Bk18U@2MM>7ktyS^tHrc{uIZX=0cZrdJ)#$5
znBUh}dG07tJ%;;RM4o07`kUzC#A##h@UYl1-)gDfz4es2lU({7jGRb&i}#encqN@`
z5q2TU4>p@}v!Xno8g(IT#+x6bG=kNtGiqCGcJ_OJZB`Ho+`^WcQHEfg8rs3U$*%zW
zk2sFpubI8eOR^Jaq}+DO@uJ^10X;=e0g*%d56e?ATlqbA7{)p!0;5wyHGj`*nKX8q
zj0f!RfeoF@C%rMC*<1esKYhyC7FO+K{72>f&+Xx#KMW)KYXdasyxmz<eUF`jh>1Vy
z)a>;!hVr?8(3U-Lb5WDA(imfn`&^Yp$^Bg3mGG_Dk@YJ)P~^4XMCNLWN|LvBxC^N`
z`xa?U9mzh!Lrums5p4+Sx6{>lnm{`XBg|4tENOo2a<MiDU1<*u1@M=Ha}bG72bjO=
z!&vYLXdtXKpP<Z{X!jMLqh6aJwi(a94fYooJy>q3*C_q^0od1n+ado)H~F8|^`AfZ
zk3PL#a;9WChgJ<b8vvUt+^R>1ohiVMzPyP;)LNI@v1;@O5j3nREFteV(W!{bqYHe0
zSDwH>TV!%Bnn_~oe$}$U{%@aI=hP7vc;Qvd2Ytrp-F)${@AfkcOU?WRfH2bb#IKxm
zyURZ`i|YXBp(}*mK)P914PtvK^P!)Wlfc_`Mf0Dy@(=X@uFqrfXC*+f>aIx&q=BaR
z&9;7mxoX?IympzTQ1Jln#Q(QPPCwf6M}lIxx)|}Or_7e1C$Juqp@I#NR;nlU^<v)+
zz^)Pyd6srn0x6=QLYLF|6Jsum&minTz8MiYCIvtDvd?clgxcOv3RlhPr_7#@hJVV&
z<QwYhJ3p!NW0LW(ur!{J?kL<Ywjw$F3@4`leF^yv_>T0afuuElzKM6~XwF3b006aZ
zq2Hu8sbvdjTL_=1i@`uQ%eoJ=%nu)3&VUHd7r^fr6LT0#=-6YHjR1hAAqN0Q8w$wc
zD@vz6a-RZS48t%(tPudtM+c(iqeL#=JVX0NLH>;QyXZ^$W9DZ{MgPx82Yd#Hor08*
z=Qx5D#xCgA`gPswvEiLS;Cs`xB!1w3m*_Cj=9fUw8c`eIdjwU?N?!xPU%JaYmcuy}
z?brhDzRG1IcNc^^tH7s9@eZJ%V>fcVOa8~7|DRCi&KUlH(U<WlWCo#Fe})}E<VL)1
z-WPr#{KpkY%!kwj*zX*GPUxxvKeT#Nl2A+Hq0!%oEa^Qksq&!$fb}AxQ}q9Tz<G%r
zUI6)iAq7;Je_S@CkYfN|8bl}C_QX~UfsgnDFUwn66mGfnb291wc6I&xNpdiOCrx^*
z@+4P!dWfQehy`42u21`+LcxJ3STDGS_)JiX)Md%Gk)Cj($H2uO1o%a<&r7%yAK}7x
zEDeAMlI`O<k}kSD;K;<vEDlU(@N+*0eh;(Gk&22g({<J=BlcS)e7nZtSEPM)CENe2
z9I<Bs_g#Cz`nbM-hncioD=h=ZHneQtM;Ni7yMt=l0uad84zO~zw%C4lyj&UJg4^tR
zQLrEm;-)%}%IbLmpMb$@Cv9o+JK&%a4DdVo<vDxlu!^=;Lc!Po^~&=4$oKCad>RPd
zYJZ|G9^dd+yg?if1a5#P*8r=HKp3zBRrEqKf&y`b>*{5G-*5ZhM-eu<l-A@+mKUG>
z^Aq?#zsEUfK)S*IV(&eJs!F?VQAMICQ3(<yD;WeNBZ7c{h=H7uq+}$=jYLT*B1yo2
zpprpy&XS|#oO90c&eh#-cb~W4^L_Vxe{R*SQ$=;r;NGmg*7M9~&N;>$<7InSAv_M`
ztKA<R+>6rJ;6(SkjVGA+Ed>-1%P)qBpEv<hN>K$#9Wf|Tw<fjrR?@X!ILr3^s>5wr
z-dAX5Vqj@Mv@r(18?^4!g*wx^@meyv7${l~Sg)4u`bDT{d>h}i*Rw+>2#t4H_@)oE
z!f_CK!$Atpn~SMAF+qt<VbovSOv&@pemVRIe#VOT--S{TyFl!$(%y7A|Lgke(H;8q
z{nw_#mHeROEc(Klk<%)~QSzoDgV9as+buM(xU@k8zOmlFx>VT*SwDF?n9^j<C&ltd
z4W4@Tfymb-5bYAgIXV%4ljk|<kp>?CEDYpaK8Sr*H0Ir1`?PFonK<y@9#*ty7EG{_
zV$ZU={P8hZf+}FTj1)fnHJqbHwC_$*w9;ds2!w~|jH7#8V~RTgF_(Mgx8owq;E&8s
z%U&b;+qV6y5%RYm&SAp7u&>ru`NIwzlMRJP5s!nEx})qHh9#K)9tZtoT+fru1iMXM
z@R#J8PYj<~F*F?T9z`soReF%2TdqoUVjoyXY+%B7L;vNqq_55%LzRMw^G%%P&x5>(
z6>q%O-9>Z~E4B@FeXPp%nGj9Sall>IZ&aAPkz-olsz$BmqV-H8E3$C~e4KdJWMpSU
z+z5N2Vf9$>9jDeIT+61CN41q7x$&CPPNm<oHFr=(Eek7niV;c$E>4?M@2%VyH6soy
zcQrnmiao6zTyoGFwB;<o;!cdNv2ad!TMhcBLypQPH^T}F@ZTOgDJ6r&$NeyUouLrR
zNf)g}*iQDiZUsMdKy>K)J!W<|z%7356kN5aw1UCW<=6Nt^lc>jQPx#Blh+Cp-l9~R
z94vj<Eg7iZa<5=`7Sw(57{|vu@}p>TPiueLAG*HoR$t_|r2rc5C-uC7>ovs1wY4K`
zdtVaF8lQT0s}`hqoF4DGc1OR(>3bcI2)uufN8M6ylSye~*{0l$KI2+Y9e&toCr?4J
z$hslv9*H*<TNF}}qqsjCZc#&=;0{LYVfGS2p-a06(QmNB+-2);uB%19XWZ>Z_74`~
z${!YX<|ooLyDFFNZr@WZ?_~T(Y~r$ptU|zgHSWRhajP8z9of0Q50ylJ)fzxinGqxf
zaW8!6J>rhealdicjN)^}>Mm%H762`=8jdRhP-ONjxNn>kZG2vhe*mN$7hgt=fc-Kl
znT}2E<?*!qk&P|dgE6;OZP?z`(Ddx#=O%H!ePaWrw*5fo{D#O*pW!$n%}scouGp7U
z=cYMmMd=4JzQcygu;+LQt4U-C_COi_dSjT!7QJJ`L7V3Z?~r}3Hmal)oJcut_QTu0
z$E^w{BcP;mnfn-zSHpy@atEf6C2u?F*fubZzm}o4AsmM=L$L#<kD1`Y{TQb>iqaXh
zNvbXi3=y1w(y3pTVf}$rr6aVixL|ergG~1Iqcwl=RA=#*GxRgbbF*-@m+pjInWkf$
z@W!Fb15MZD^!7J~VDu`&>5u*1V<#s;khhVkXA4?FAzZMg{Xr9tYbm!^$3uzGK7uL~
zWQx}yB>N${8fm*}z;@6F#mcWm&7he~1moVBG2u6r$7K2?FE}A~2egs4(*0WIS+!N|
zA8%LF1+hKY+-K3qOCcJbz+9;Zw{n^jMD?5c1Z)wC=hj?D4|SkGh(2d|k1%z82~kXa
zta?o&FYwhYYEc(Xcb#6Y*Lm%dxPs*&`~yFmszE<lwO87Txm(q`2oAbx(4L=a6W+!O
z{{#o)=ibQfhYAD___IHz&VG(CdF$#l^;+h@PS#lso@;4(2@p|+E^gy;mJsN`la+9n
zjtkR#7EiTt`P}v<BBg1Mwq5n}HGOsQY8eGsn2S5@e+sstS`c1Xfrs6nP?~z5tFCte
zs$k(OU7wWxirRDhXeJ5ndAo}n2WjUbFpjlGdYx9yAEoXh0eIvXvMh4XHY<kJEoX9Z
zeJzwPJCefw`&avU#DM2xlHDIC*oQD|(7dGr(e``R>zC*#*l3__VOywr+Mwh?1n!)X
zL+<8MWBx{&aQ~Qxpv2>Nb?dp$+5~iH9ec~G{qgZ-XxyKta|NMRMR$j~YN_m6_*B)P
zp3q*}1l1i@5$Kzpt*i&}`2~Y~@B$t3wc;c0n!uA110mjr63qe9H#)t3VvYBLs67!i
z#6bxcg0W+GdM&vxx54k32NAQ5N0SN-s7FL}+XN<3!9ruSRt-e{pG%P0Hgne(Y}Z6R
zkz$90+aLFcEK!EikL;q&7^lZNZrd(oG^=I?6!TBeXjyMc)zpdJcB|$?g!c2vWRDOJ
zUy1iVW7Ds0<E;mJIUVn*-=N`+|CE5J!rxr0I2VUyBh92Hw`8#~0j3Dg&-NlV)Vr5%
zk)nO%>rty_Rxd~%V)YRnqVS0|!m4@7RUgv_#<xpo^yu>aUM#q;RoH{<mB$1OY686W
zh*%xPfys3gEwR^-6HK<%D!jVI6B`1Qptz8nh|vI@YlnzUYEYKmz5T?a2d!XP<-r-7
zxO+HyT84Wjrx(^MAxb~%t!wLM>C_u}AxHGAfI4j8!lnl~3-sxoBYkbu)sC}$*`9TJ
z9Fm#bps}y6&aGKqSd-j_bvGX}VO<mNzu&Xj(a78QWHwX38VF6zsv2frynb#&DsQXW
zGepcJNoc^+Ag4nh|BzGqHO=ycWgN--oz@^5>HqMaVC^7p{6#0O{y?LFa<G)LEWDJp
zR!3o_#pulroRa=f<JH$5f4QJY<P%gFP`THbju&QZggiDIkzP=rx8h?D*3Z;~GxOet
zWBi>7eH7X1TrhFwnbg-_(iS8&nPfH~tdgp6)GJFVbmu>mNl}lK|2(+K&Y&w?co(&u
z@$TW%bT(pQGDfEFvI<YS>8XkFn3e=SY8kv365Uy^9$IGrX$*G*)>dMQ_Iu3(>S2b>
zXpe2dIgpv@j2wxY9;ybNN^~c3yP8d((fA@ex4d#7j=}d&d_ngcNHon~_f6eK94;96
z&nnw1xV*r5M3f^E6*X7ab#J8EK9NR}T9G9xs<wLTnrPYSj6iuMr-g0%Ei6Xk1}=5?
z?O(*i!YZMtoIs8gr)aNh<UCJg4M;kOa@wORIH?q5BbI<W9jKmt7f<|Xt)_;3e)#cd
z(9GxSRcI7rc*Mkee3ya(S&p6Ty5^nv^;N<)6~<=G6@UIRbnO>cFrA+-fC^Ca<v`6B
zkFvlk#KgfW!*;}X7@2ky6z)e+fVzY5F5N}*8#dp137O&*74%f?{nQiX72G1De1mFy
zJRcthf%EGX;$bHKz{seZ9NgSmVbAW1N55j@w)xnr$~3i{l|U2sK)13=YmEuS7!~$3
zBJL5#GtOX&sO$S){=QRtaoI^M#{G_B^Q=*<t7M#M`U}I2qb{jmI!r8wOhg&q+nOc6
zB`6&lPR#mb$6sVx*>;I43KlWB4TpM%!;M&uzZFs_%d1=<NqFGupgR+}JBE~wj=`Ty
zg#r%i#ZPRiT)(oLn%xxAY@b-Y3O24<)w59*ZKHAKR~1%VZ&ee{OZ*ik|5zZXbR=O^
z9rZBM`E4QT(Gs#;SCFXi$7Wa3I*XO_X@)LX<kldZL9tPTzoz;O!_tD9wKxyX6sA4N
zaMfZJCR~i>%c>a~VFI7Ur!6c5Gilk4*Ks5XbG%hIZ#r0CK%>kddPmFfj9&^xQxypX
z&$qjK4_L)Cmscl5?PN3;U)8YDGt7kchw(0|@P!%5bhveglr+S(y!8L%6}VxhvSpi*
zx80-34Bo)nD`naI?-<u^>&09)ue%Y0`;4=E-SNm=ra#n1U{fwy6VwdjatkZdov+XH
z?RGSPL5ZflLPs{$;+334<;#vk6}<Pfy<Xi=8Y7*YYdd)PBgUwxhApIEuSIKf=6ooU
zt}+{6<1|iN$f0VJ$!TLj6R&XwEjQiA2NRf$=){QuH96bhsSqtZS`?mRto+M2F4n5d
zIk-o>30kLlpGAq1ek08OQPoJJt39dq&MUiM+%bH3l*Uo@JyL~arbE?=U#VByF)mCX
z%6`|SLv<O^Y%a-YMia}~($3YT_zyJ`RJXPrtgNAboKjp8*ydnmgLp64_J*c0kb_Q#
z;1Qx=>MF$X*y6Fr&E+3mLwtq@v_nQZ(8XC1BIy;`M(e%GF_bd#=0~bmY<;vzydWj0
zOs}Zq>y#8cxlyUjmEwJ(zNnUiIp<ur%)r4Mdl4qHl9Dm<#R?Y8YAjyaw+;<6*4uI}
za91rea@@5b`jCE@@M+(ae{?%kpyP9~szr@7=lWQ@9rzTTM;zSyOh1R%ey%$vOc%4U
z=uY?2?1!2_xNqupvC`8Wp$=)}AS71Nah@LPx>o#FF2}pq&{Me4Aq@A%-nsm|g8sV4
z0xSB981dv<ev^#?emsqNI1DcWUkz{DvpfFOYLhZtAtF&70&jK|1#eFWNq1>Ug*kGC
zf`#q8z{x6*J|VT<ZqBiKC+V#|x24p`H@GP&93N(!pgc4`VSFWjJhxMB;B&Ff(Avq8
ziFnErtlmM9!EQrDOxDPCgfz!`-m<WeB)u2MNQ_2$u%qP0v)I42K_pzUFS*O4L|h-b
zz^h{o3akEbF7uc}rz;na-Qp1RBMBiUX6g0RRf9toakH?bnGGx5KR4eB()Ms1GMNT{
zDd;)qiO}AE6y>#scQvpuq~%Mc7q54?2#N*`-!6o_2Ye5Q?7ApuKz2uugW`pHs_pt+
z7v0yZ$_xf894`v<(kL$POHKK&wWd1#5+Tv2^nNo^qmWA%fF_PP-*Wskt=h$0fV4t?
zz%2RVj7K*Hr>Mis5tqPM4zJrzDy%?nPjk}jA4zp$IUi*SRkt3=iY=$0?bJ_ez)vq=
zqeh{ST10Ct38sYPV;sc$(}?zRKby~Lm*-m`9y&q9EoeIO;aWy-Eo$OLMnQ|*b$IGQ
zdDP>2y_Huf=pJ$Rb!%&QuL%O~L;CaT9xv<ncuPBO%H}v{F4FPJ)Hj%RRuoF}>;2HZ
z?Kim;kG`-2ZRl9_G2GercK4>+)yfEs-i*4ScIt;R3!~cEPWC^pAMq@lK*?JA<5|LA
zj|PTnp{;@8V}gF*w?#3p3wNu&k}d9V;x$s{orJbXl8p>+daL{6*3>eiFpNq#HyZd4
zI%*mmC-f53b2DDPbX>p}+dH!=9^pP;aGk=g%nh1#@x~5UpG=BO(227NJ*mW9ZQGjN
zOmE4eG<kPe4Xr~`4{(b*Fo<6I65xMgl2e%);s52G#&t-#D9gOwI4xjAbwny*k;rWH
zDsb(VjRWp{zWm@<MoHUXbD@yRe>lJy)*h1oD?p!>{P30QV?c!!wC(WJ!mH69f3DQ(
z>3;v*^hW5GjoO9kB2qg1)L=d$f^c&nmGj}VUk`pJJ+=!;hYw)HqVGKp<~feOMHied
ztC@X?@Gg|WQ}au?3=}ezk~DKG6_IYdvkzJ4Ii6C{X{^BudqcB_&!4k<xZHMAvi!2D
zR@OXABg>$gvebR!m9j;h1RNQ=kY%*tRE+X_Uo$!bO&)Y>c@^e13XHmSEq2Z(c#bf$
z5n|Cl<=xG@9ArOlH@NP0<1pA(e9R)<u@t~0&f_>aH7RGLn{*SBUe@AI68MWO0jCQX
zt-@RFx2rX+N*_n3vO}dZ5ZWVOyz`fbqC^V%94EE&)N9nq76h1OcN@Dv3NYSzS8<3D
zv<yXr{pzz8;P_bDxPg_zMWi|Ri;?hiM}omw8alBWd$5bxIiXaE22cApYd`733Vykh
z#4JI1!Qn6&$4iyjD)O|OMG1c}Fvw8tXGbh;>rjUuz`zx4wb{Hq!<=A5S0PO5jNHy@
zZ(NSKp0!9dgTU_|Ts`aHhF81RAwyH>YrpZ5jFSR!&C7otww4v035_O<&c3hhE8z2C
zpW?cH1)DqTARjtEb&40=d6KHo#Olt)!&|n;I{Nt0{krw<g$OO0HsULsUiho=cVFQO
z31)~9_o3K`5FTpa$-3Xv`KWlGhr9wE{l!|f-$mAGTWMdLG?0|a94}J9<Pv-nCo5-i
z0T)Z*2~@fKC^_Fh`W%fr#<jPX3Al3Ktejjz_Hurem)s34N&2WkU{}P1ehSVGVO3=+
z%Op{?`osZd?WCo!iO=FU(W^1s3W&O>@QN<6CT^`S%di?z?aUa>uCH}_XEvS<`xix{
zau2C0{h=$*8(Y(8#biV1XUx7WjT%L~m$n_#4)3}neo2MfWne><N@Nv{<T^jzd_8sl
z*i7$HGoAf6MDWL5FENbOg01{HtgN57>sN58)XwBUa$m#}>Op_}<e&-rOcr+J<^8MZ
ziJV~-Vba7~>x93+g+zOtw0zGO_XM#uBK?(hQQn13rEy2yhTauq!(2qBSnGV%GV;04
zvQ6a_Ov2;47%)GF|E3kGSvJ>UHV&PH^mQH5%%|Ou$2!=YI@jcjFMgC+PE2p!>}Tb9
zIEeD~Rl8w+Ba;Q=*?<$3m!VEce*u9)r*0U^V|8u3I?i*bXKwaW5XXYSEs`_rWqT7L
z@kx{a;e4mnGuHim><~Ew&Jb|PKMzdXu@w9Qkb(n(U-XKMVz#Z9hDY&z`)=6Gc+VhZ
zD;zN{3=AayzLPA_=jCkh;$&X&i(9^2SPhc^QO;OopPpNWP0Zj&AG-yN9LiTm(4fn3
zlpS_wdTV2B?3=P;(;vrVaM!$wDJw8!DZSAu=lm-zC}ab1eBLZ(l$T|TuF7|~!F1|E
z9&DuQ=o=cCNNZ<*ok{Pza|5`Za6~&z1U_5gG{oBV-tdGYF#pmpG=%47V|_znW@Cku
zYP23bUbd3yPcM=^N5Nvpzy2;2Dz`xOrmwD;MkA>%43l*si#A#v+3r<1%nVFt<ju3W
zxe2wx3MTRG!Xl!I4QDOT$_olN`k`721PYfS5ix^mXT)wT+hrHAW_!DlmBin&w6BUR
zu2)(b^){eXd^Akj7^2RkU`{9Wk%K4FT6Dv}TU>mUDU6u-B{S31B4p~_y&|s0wG3H-
zv($&br)~2-rC^8{sJO*6BPPnd?wVm?c%bF}m-Q@41KnV>3EeRer-{hUad7hfh_A(K
zSp&M_8YyEO_eqVbWf>dP1{|L5A{9PH5!DF5^CVK+VLo@Y4X0)2D$r#~BpHmy1^2<d
zP`!{t7pfDp0lgO&UCfnE$J}2y-Z;=S!pM0)YZK!-?Bik1TB(K0j74cJ!J5ou+%b^R
z*7x#nd*Kf#!VCKb5*Cb2=hgnmD&}O+W`vltE-wV84P+1z^))(@>O?ig)ui6pmSvp4
z#ys}FK>j8btlLgRS>NmH#Ri_kY-^D=@RsGIBG^pLtUAwea;yE-<!x)WUMI+TiJ5RX
z0RGyV|Kcv-D2a1!eP*T}gR%el8`Vxsteo`7s4w2%pV{G=P2+Q)*P2}2Q&*}XbGBtO
zkd2C_Nc?^?JGcv{@OWpCn8`eE$hK(;>W7qxkb*`wfdcIIRPsl|4=1k2ggtlBIKD?s
zpR>2WsCz0DDM$S<FM3c({3PaC`nT)7i9v=Sz-&soXA|OpAte`}7A@8ElDL6+`lYLM
ze<S9Kue#x;8?9dB44<u}!qXnc_v#Kj-=I&RsLEQ7588*b73kFA{%?1mk5SXkNWStL
z;}`B%`C-1<9MkmO#}@yYPyKv_djZrI;bzHkJxxC9=r-~5vuO_(SeZ;((2^r#&HEWk
zyD4ToQj-GmgJ<dAmF~H1^f&XKcip(<ZuCwvT-$M&kNcTNpt|+?#w~R&Qma7qS2pX{
zr<o`|RkWp^G8$j)^mP-F55S0?qvADIFi@c<wp)#j!jQ0%$+mgB?e=3J)VQnK%I<Sy
z6o!iDO8IW-Dx<Lt^D39cpH`~`s3h$6U7q}+29LvK)lHFuIkqxA!b__Ub0shBbwm!r
zm4C1C&TVZP9Qj@m=Ihrl2A(hgQr;tI`gMMpVDD<`B$Y>2ElG-fzfL|W;U1Ok&#NBC
zOP}89aDG4c&q!1h@1?ahKvDXy1$O}*4N4X5iqbJK4<%A~GS1;Dn;$4Kw2XIX^ZA*|
z=PGFHYBmsYI9sD$Xp+ZbV;f&>8fI&SoHxSIx?vMoXP85V2y|14IdIOB91A>v1Tj@5
z3Y2beKdksgS3mKtor%Uf@XpZ{&>g(Rg$gU)uZg(VU+H404@o7yaR&#2!B#Sf4wZ)D
z<h*~Oj@npE0B@~IDceePr%EQV>kSog1kQXm5mtr?I5cr4)3k>MDv4E4`Nb|ebg&Vk
z6X))Y7hHL4rW^e)Er6<F#-$vNE3+}rI2@DTh6)@Hv%T_#W~g}7<2#NKPd+g@hWo^R
zDsFz<3EjuxZNl^r^Np+SPg&?`MCr`e{c(hb?U{EO{WKJ^6|UtM2=Wj{jy{Z?!$y!{
zwWr^{ycb_r)`70ni8bVFegMtg90JZ&E8U{sb)Dk5<Uq9-igXxScVFXaDJiRf9F&%Q
z?*li_i9qJ-k)M{r@pI6}Eo1Hp%`<p9bWZ!5;F2mBtG70t=TK18T1d|`kdkh42u_B#
znLHAEMYC{L>uJBwhvGMN(mmk2yTIYr^5A3f+gzON>%H01jr{ZVEhIKk0d0#pNF#YI
z*v?(+*F;k6EQzB|5L>ovpXcfExo!0UcXMdriXV;R4<Glyu}6uxL(WQxDLkJwdTA@6
z{y81GFT&pHj;TXs+3{g?ZKetrEY7-tT5F4*!Jr=9v)Oy(b;A@!u@c2x(8F`mnC2~-
z`{MrTI7`mdGxXc$&sn&;s}^LvTq$2r%fEHko(jh(;+NYn%IFx3^(V7k|1+uuND!iJ
zl{(3MqVdNa&_L~yj4%SR`mG-Bm@R&ew3R*}dmCae+1b{`>{n11&8JFx*{}|ZO|3Xw
z0#^{o39*GMFXnZ4MM_ru!BZ}Qi?ammc~0(G1ih>R-_#gK*{(sS$V_RPO1Akzit!}+
zOyW7LSiNjU($e5)u9<A`xAZ~Gt0I~f|H-L9aUizpkt0q<zXK+EV}ue*HDgC$_(n!B
z9%aNr^Fc<$=O=*axAfsa#Gh%s(F{0y#05MG{$^?Eh>JT$FsOiIjA|0N%C8QYs~Z{!
zl9;YSYIqUMT-ArB5-xPc`ex|$OdsW|YuTa_)Gt5hCNYy>@QvbPr(GhHtE9Pbk^r~?
zM=I|#ol3S(`)7&^<u4eQw%6(>HVvIAv~JTHQb`XeXiJ^ZuPn%BUky?>IzPk1Ld({O
zDTTr=a=|sLbFQbOq^AL#(j{PDVr&x}dy3CsceG{IQ&WInX~g%grk^s86WPu6-k5;0
zeCY<m)2I=sar99bBx|h7#w@J!4U~2*3-vYG)yIkEG4gG97otjd9AzU$jXb#JpHyl+
z!pLEpX(MTiQqSA6HR|E^R_DS6P<*@Ly^}Um4>FBYZel~>EFJvQhoq8Cqezb(8arUv
z>kr8`ZTv)nG&BJM&YwtBxO6Ma6nf#^S_OdcQh|W6eM^EPx(<b9!AD&4mS4|*{dx(q
z8~wm2pPmrQp_}vOVd`MvMYrQyX%qM6?UdrmuhIuii4)P2#npqSK=b(emd2+C`=N6k
z;dLH;`8Y%w<B<35<@I;lZkL?z(ATr9@sbvCNJ%Zd{2Eqtp73Vd7E}|&&625+VNN<n
z!D!++*0XJ|)F(GL312@@VYxl@5PeC+FjtZsz_(>nofv}6t9PW|wlD~6??dtR-tlYR
z--?%N^f^~TwaiXUy;~O~`3|l+Up;?BCLU$*^4bQTnUiJ;nmEZ^DyTab_UbDS0%f^l
zGLFk#PZ-^!{Ail<MIjp=H11_?0<hE-`E1XX5JeN@)_jm#j++yu_-E?)ckTnSR6i#^
z|6S%4h(>45qD!i|@Hl>`)M7ZG73~QnphGg!lW0XIx?qIdTW_p#kSy&rL)XC`S^I)u
zFYQO?-OKV%ZG;oB88W&2oiLl^DT8RAjX2NR-|B)=;+<HD#b_484xsj!QuJ!CB^iji
zP)3Ywbw=k57;Bu(l+}00Jm`(-PO)>dQF54UbUP$t9oV>icH#~Gg(+Ky-gC?vvM_!_
zzW*~QzuzM9#SeO-W%i02>?(<DcAT9PqN%Yq7mTiBvN6&5y*zvMktpGPk@R0mHSTH;
z#>D7XCIzF1<GIlzo7i#`BEH;-$J#*sWcD-H5yZSVZ!1$G;_v)1B0G*$sl!5T-W52m
zL*Cfo-U_rE=2x42<gAe2g@y}-7R^8>3}>dbhn?OQKXbs}n)DXaZh+p3x7}j-Nood*
zyxLBpmZAM6ZkBEC=@*CPbDx4R?bK(S<z{17jYMgr<2#~~epGN;PKIwNjKJ<f=@vEH
zSz((II#}6cf!ImMUE!EWj+*3v%Vt>Dpk9>g)pnjW{aY((>yB3$=`kA{#%yU=-RAsk
zt*-@$s@e_>O+G{`r@ncF)pLeW+CVhEHayLT4y~N!t<+Y)p7Q^jC;n$leNK)r*zRz5
z1EavDEc9==g(aQD=pw=+$IyeIV$zaoc+m}{j4!D9R2Mnu0KYTlJnm1m`}JrXBx65_
zUIGP8Htg~#AfTm5)3qv{iT^^!{<iMwJb@b!a=NP~`0ozesCU@1W4%QDaSVT<ZR~^}
zR<{8&@}W81s~`W-bq_b7p;N^>%R6{_7m(ipb{s<-YWYED{SQFo#18$r4nUP7wX?Ii
z1I4@E<an2-`B5Q!Yqwrt@=^GXhC_4t+Sp2dzZb?TQndwBiwYqEs4em7lm8^86h#4v
zL)&dL!v5deazBYGLfA<ZfuX_bsP9+2K6h}Dw7S4IW0og$&pM@fir4bp?!Yrn$6F+6
z?aep36!d{|*59;{|5p?0x#(GR4^EJQMHn~LApLjA4-A}-ljNA98OGl20xb&<>CP9d
zmK|0>-MM_6dgBN18|n_N&4r=gg+?+t7pVvYfca{)^)J3J=Hs?7Y>40E0j%lJ2mkp#
z@k!|>5Cf9&w-*hs!N40SJ|dm2N!Bbp10d(Fh#O#0C^2LH5~yod{Prr&;a@;JL&HQ~
z)geqR+Rj6`kJUh{2^TG$S%ou?4UzX9M@nFM_S{D;lYO^5$&%)|z@X3{<XSl119@S-
z#Xj1WcA^S_6%T(ZiEclNf=hOJpYgYr_n$cG{V=$>+jFLWYs~$=9G{b)WvfQ2jd3_M
zzbX*a%c)DbK<mgZ%C_;^;seC^?p%15jjhwCK&B1QCwn6kN=cqK>Ty^!`z-lyH<%t}
zD(c+A*ZX{(!aW_RWHo7Z9RjZ4RZZ0d4F>W%!DPz5*!siJo}V8?tgU?!gQ;;K8R3R2
z@&mMGqb$5TWVhQ#;z>(aqsT>9ME11Sm%s*q+_V~{(N#{*eR+WNPsS}x31?s-x^N^;
zs-d~J9Ilp&YXN_598h??-;mko0+YeKbxn`O=EJksreG;=-6Xh9S9A!^p&D>%a#_;R
zESEl0GQOv8b8a6wK_Ke6Bt*AF&xuXYr4K0H1yDuO7^6-+cRb}HSq%uBXlibchW?ib
zmGf|cN-c{?liiO0zCxij7=X^O7R0di_rZhT@BSEi)N?mL?NUsCqtyhl=Y1WBANG=t
zfDUyw9f86s4bPK+VoiCguD`!6iU9bk`9Bc9)u29t?NXrXSvFX~k&g|cAtAWU8*FRi
z1Ek6tR{BT=18}P4ihckZVdWPLk3y`0J8XNykRqU-6V`=NE#gW~(BbFl$AuYw+FOQ}
z05*I~tv#S?^evyw8a|6fN#Lmh;+#{jGr?Nh2eSEk_9=b2D=F?Zm1u9tjiavqM$Y8P
z??9$j{D9HX=j&H+v+km;4y`*9?a2fzwU8M|#OSzqByB*Sf`HKyTzFc?5rj6y&$-_0
zZvpntFxX88Cn7t~z<)=vbC;rJUz67*s-(2Ecf~E(bq6C@7h$$$fNAvMuQZ9`sa6!6
zX)Q`>5hhz7L5R}*zAOYNeT@>CB0(m0*_uw;cW5A|L}O4|;#IKvk~Q7Bw>IF=7;YTG
z3&BbypsQ%oU~UegFJ6YT(}8djh)YBWuViVZ(Ta_*46NfLMz_moZQmQcQ@6#hq?;g;
zXB$_+&e936fA;s;w0}{uO#NWDwX}C;{j+)azx<>f6Agv~ED9c}$_N@aM`c0_%`e(G
z=9&o1f}HIwUp^l|4S_Ag8jg5hcr?xI`?-r6FIE51pVm8q<YIKs)hdGbKdj6@k!J$b
zSxM+olocz(wka>ZR_DG3VP`ZC*U+7Cw?>T`Ax9dOW>_R%&iZR-|NgjwM`3wWl&@yk
z{_nlyk&o8?5j}Hl2Rs#g0e!rV(3)3-c7cYs(>;Oa2-FyCu(Q?oPx2UJdIm11NIOuP
zOz633MX6;m+U_<`UU4-o3%_VyH1D;)l=1Hu5m2$G2-m-UcyitGf4Pr;hPi)zLv?oy
zdS-a6shN_%+!4^4lHp;FcQQ~y27JgZ;i?eK*1>ckg6%)uFH{+V0_ooG;rnMn{=fW$
z*1rP5%;G&y_P>k5l%ZmuK@j`*yk~#s_x?%Hm2f141H5=8-}X0ql>hnz{D<EH$Ozmn
z3Cp){{uyKb^|}AU%l`aTyEZ(+#q`6pq5qb={O1<^`!^Q|Ad$HEisH+Y|Jm!GlLCXM
zXwic;|G&D*KQH$m-<<P-O?7VS&b8?O+3Sa;VePCGjW(wJV_p6EjQ)#nD$ZPmJ$9y0
z`EJtx==J|Uc>G_^ABmkFZN6e?IMYz{KU{^hXww8ydfqryA8x&93|#WedV77XcqZl<
z8lCumill$Vu7AbWfBs~S2`&7dkXauPuH;*6zoC^Quy!z&a9%h1b)NsUzyJ4d_2-+<
zJxA+O5_fk3ZnnIlF}x=-31Tx@7ZD>N&Gc8g{C~6(<L>Ew|CV$fra?8nHLyJ%Q+?Wa
zDf)l3mY{m%MmE#l4C~_m(Ix!(kK^QFGqjcp-2I<jfCMQq#Ihi(vj3hR|Bro)Ttj;y
z5{rghKfnE}ApbA+-amfze?R{Jefs}*@&A9j_dSFbp-1RhEk&$&i3tkQt(l}UPl%V)
z_2|S3?cL5oYX8bpKk#C^K_z%q?#<e=m1i+8QQ{*=EVgLZ5>Ahp-Jmh6-zmp@arB@k
z-({%_|F?*Z+Q9j0_2|DI2W%IC7#L&1Kk|NkJ~_b;*zKHLdH``*`6Dwl7pO&Uy3GoY
z;vhUV(??%-k=g|*w&7&pSAZ6_$H`Wg5={~`{I08L1wE_#t++-S0w>*n7Pb=NL`a9P
z8VX`!KRl{|uy5^O>P^tL$i?@@Ufd-?xK6sX5BfiF+T$QO`)~8!NJWI;4hz&NaS-dy
zC3r0~Pwc*`4F-6)T$@sn#$P%FJjAoYZfAZtTlu`S-OQby#GnLt@8giO?mP+i2Mj4s
zhvbLSzwY1Px;cNOpTJ9McK}SEX~gr?Js&f?9`{xsOjtzZr(}o7Rls=Lrw!p_-%9nG
zCu#}LRFP7zs=#LQdI!BX#SPQ@(~dj1^|pAZ=U3ka&8I|=Zg4~HYV`T(;3~qn8;8Lc
zoY^=>9Ve)>j!wAl2eZH+MGr3ceyaY~VYs{x|EsAFz3B1u;+6B*$3N@g4^RWmc{j7)
zIQFsnnIh;>964e37bdG}+4$0o<^^9jKpra@)WQL7@ZK&zJ|pRJlzDc8%T%yXR1qKM
zj@^52z)-#9ESZ9{8dGS|mT9el&F=@t@bDrlTSo`YW9?rz=fQnYj1n-j@hHPk9Ar{(
zk~x%FCi{MXwA^rkgzPbDJ&`~lI;0G&LB%0gX>;$;4R2WLt5)xZCx-6_HFsyd=-edN
zOy$k)VBS_T{^93cnH!C)y^5a6+aU!?e-2gs97PZ|Jh8p8BIowMi-Y`I_JBeueON{-
zkJNFwMrq%z91&=M?rBqdW3LA&^T-HaH={FPNq$g%RE6nFyepr3^Agn$e2yM0lf^p^
z^q41dJ3h-`b>8cPr(T6gVg8DOo)|h+)11`^Y5P!JPVvAvN+0tyt`(OefLPuco4cS6
z3aXX*+)Wy^SJ2TAJc5CZ3(!_jQv(h=(nOkM9cTdbYh*zr+WHvGoiur`rF#3<y6+B?
zBH|XAop)Ig-UOEx&@<02(!q%1b6km`RWcoDq*w!G*e6ogLlEJwkHg@ZYZO#BlfOxk
z!KQtJeLrIbxMVuzbTvuGgUDptdDhhS$)8}W<}`elBJlAW#&>Fx;qIc?^D^}4nP|k)
zdm)q=hEgBY(L9eAb*mvaqd@AVHNf5lGw0A>`$`(3J4H&2)gN9?7{2s17VqZR8PD=4
z=t2}<TZ~uT#wuEQ=jwNu-TH*>4ENC9qx>WIM7Y4KvBPx?ZfNG~)e__PRU<<U!LlVs
z=*UIBDP0G5*s5-BcR#F-;x?>ejMO5^8+I79hDY+N<%Pq4QNCPoL$e^=E8Uw25|e1Y
zYh8r^Ni$G0W>%o#?yY`_)DfbzsUbWm?GZ<}L#Am~nZ%@CAZ?G+J0cDc#C#p@Z4AME
z77MInmJ}J%ABGWy!gQ@n(|nk&EwaYX8ijIxnBIN9By-~c=n)qRn5t_Vw|jMI_(2pz
zgY52LsPG7|a9X5aA|7KA5v4GW!L<Z8O&`oJ&az&G;VM<=Oz4(-su2WoxRbMu!egq2
z<wOz>)MG7Nch7^agequof}AjW!SAh=Q+luI-s9a-@pjBopssKX3B!m$wTh<2uNNJ#
zffp>;!Xpr{u^PHH8u@suHrXBl6{py=eW%bfEA^yfp=-Z~2D1mRIX7NcWC%8vaMCXI
zfzDDD%<%LMHJm##tzj#X(E-x9rTsH6ljQlS6!G>O+3O2FWnX#Koq$vJ+Z3KliJrfG
z@v8f(tA2+xseMzHTM{WvHjCe=4drK*B;i6Uskom};piT@EAe@M@$vsH^{d#sjqEnI
z>zGP}h5GIvBfN6Yx0+<;ah`iOL$IR1jHgVUGN_86W(<01Vd6eP?_cz#sZ=~|cy;x5
z6=1GoJ<$Y)W7+(ob6Sf-p9Vp|{30ny&r6eNqjxs>Ri5A{1^t~bS9(5XN(2OOE?$3A
zq8~>{cwQv?7W`4d-+14s1N8Kwh3ok-MR3;m!|+?-_YtFs1e`KBTz*QgX9qk^NZx@p
zw#A7mBS_h@!n5Db#?d?L25{(#21<A8+ld52hjxOyvI*T44`@HuQJ_X)`hxPEwnINm
zTM78Ua_}r+<ASbzWd7r582<Zudkp>zjuU@H(Nj;1x+-$gdSBNqu5Mkk4tr@@MK2S2
zmu*?A-rW_(b{u=--Oh~G!j<|{Ze?K&ZT=t9f1DyQSKA+>iR;)BU~c$J-x4A+kA3n+
zuIkc5g>g$w>t|7<&~7B<4Ct{A+4C1asvM@Ro6K^m{=Noa#Bdt(F))qNuB}ZLL3sQe
zgpnm1P+W^Nq|%6c?1CsVezhv;pcC@Y-E#x<#+I|T@dWl<XRRwTy!wk<jT4z%kK^X(
z%f8CK1|8sityfj<`*3C)#nX~a#luNm(0h5izhL_}6t_+ltw6dQ`p()z6;eh&Zh8+^
z0ovWlUT5HYnf3cN+VU?r%G~b99K_r*TNpM=ie_^3irFE`U;3f#vt;vw;s?2n`25r4
zidTz22I9j#Hrn>Nk>egEj<!tTY*gWaG_oB`p%Q!Pzq98<RdpBBv;ZrSb?2IC`QFG%
z1)t5}&*LevELlBBS$6rK!xP$Zd!viy{vJ7&A6WL?DQdV~;%Gj3GG52=jx*2b_D+2L
zvDrKG&st-5IjCfHsk?5tc4BxkDVA-<t#-e!xVi#DQKMZ-f=T98FGOS(J8CL-oWi=y
zt{n=`&kRiY@<<&NFejpTI#h|3-?l{IbxI1!p7VTr*5jd$o+hhQURIleSm!b55!Eig
zC0x1KcBS`a8|E%bOo!eeLZwlv#6vwngrXZ~je#bsmGXw^tDLF+&u+>(T_{_!CfY;i
zc&DD94`b?q@}O#ODMVdXFLbR`^f2|y?mCtHcw$IJZXsTRIsHX$i=l_{4N-T*$30Iw
z+pOFy1qH-(->&z0+<&L{fb)ljN!3N;jF`)VFeJvY7MeTaviMm0CC{Dx6~A`f<FkEr
zFs`Y#x*>{5a^?O#gAXH<(VV<B2Z6J^NgriFn6zJEHc+JAB)5Z~&D{e8z=g{0ly929
zwp8O`S_kTK;o(-vrdhqIj0<izVK-klVODEuZ`%@knek5Y=8J9QEE}lTQcI1-+g=D#
zIlNP&Ydoyvw#98&=*?(vlNvVPD>sJ4h(S1qht{n)f8bRnxP4}u+`@u>dvi}>0T$X$
zp3z+U$K!W#^)#L$BEV`z)vi13D-)pD&K}sSeNO<sr(JztXMGn|(5l+jN9aeSCxwW&
z&$>Xa;^yJ&SOB18{~|`+^g_tutI$5j1Xy;R4GX=iTb@5<E=-|H+7l>8&rSO|?myQv
z5A$$h{aC)Ba4=<9AG*xlqsBm!HdRX8HlK0WVM$VQxE|VC@R0r^Og9rRrrg;`G5jN`
z`(Tcy&A5ixt5;d%50hlx_5X?|;P*T#`juobpk~E|Tw!wF=0G3!F~b*G^U_p00Fu^%
z%JD>2!1aK0<D;PKNd#v~2{kAE`b{zNx8uGRGQ|6Ffa6$Nn1%55SliUWlyA7$52f-=
z$)6uipb<`Hy}O4ny><@(uTty-5LqNajkOsUnX&lL)!1a_(8&w(Mg&!{d!KU8=y=aY
z8RF#={G9p)|FUPX$$DfvR}qOXbpmJ?wo<`AR7}GE%OlKX><b)*gbsl0*mNkQx$#=d
zY6dWdi^F88Sf6BRSf-l;aq{F2>28@i3f+QM<=nS~{c83@U?pi5HUs^4aWLw+-o|T*
z*uM1piK5Jkkd74YV(qn_srhYObQfEfZOcG$7l!t?a2^c^u?^6A$YCAet7S@C2g<u(
zJ`BXwAF9ITq=Z9X)2g1V;;HJGhe5|n3^bS!48o`7op0Kn?|o%?aD6JWfd*3}^yWkW
zwIuGrZO{kXIJ_%V=&3*w2k@&s#>=gbkWpB#%GA2H@)5*W1xp+hP_tbVIGQ1A>{IeQ
zcqq<v;tKal`~I0jz@(ad^#jh3KN*22E~;~IuKxu6oS)VUTM-z`D~M<Ug_W8^7bG{=
z4M*t#7s$Y?R2#Js7BvO{na!e%ntW2ZjJUrWZOJfnk&m38sa^>n?O{v*kRhEAI!k&B
z&+_~?UhB$$wNRd*!_^Xr(U9zFI|WM0Ha(&38hnqnbDWwv%{=lgC*->?YH8oMRZva)
zx){HfCRq>U8loD30AR1Zm8Ze9vC*}_A3qNagRA$G3NjRUtP^0~jzLEJvSSvT14>3{
zjk-ek-0q}__+d0I)p#YZeE$bFx|{t#eJx2J>;HP$x*=BQ`RyXT0w7KI2-Pk!YVmKl
zZ3w_<f`oR9>-p&y;_-4T*X-+HEiN6evUD1b!nE$Z86$oP#eLNWy+e>Rwobz#w|m%_
zBd4Ll_iQS+W>|Wm8A<KeM0lfct<tNDCHbxKO%paornBquwnmSC(Rd98GRMztpPx2B
zKrk#CH*5U&Z|xefQQ3Eg^TEw+`)1Ny7zpap8S&11Hg$h|Cyu<Ub7^+;WJkd9P*Jqf
z{c~Becr+>5qQnzT*85gJI@?k&>#`0SCPW;yC+HpdZ$vH+wgNdwN0&CZ!4C;C4<X11
zCbO4Abs);A@Rs98yX45bH_Rf}Gu0k&xa2tAFPGLkqOb6dhKp%me4(o2TzeQ9Fsr_H
zYc$A%Y!ZE#fPkkH>tk4lE~;$)sy`(W;2&gG@&~L`MV7NfV^nV(m!WTYtr_Z+LLBLu
zx~eli+#;jF>m8(%$HG>(o>(&Ez4p}%HXL7nd4Sq|+$!VNc&KINU5W6F$3Z3p4;fjw
zLNeQZS6+P(=FVMoYZOHN3&I?Y;)Uzz>UO>K3G2uVKb)Ajz2^2+k@Wl({k9gFb#0=b
z=ZFbY*%%3&-ReCai;FD!^!THx=(}#gU_$FKZn1V*(_E_MnuG7Jj~2-O=%2F0N~nTM
zuUj#s#;o3AD#E<7cax)UY3Y2246+ERC?4C~!##=$Hk>$O7T=e<2_UQ+-TQk)9Inw$
zl3(i*9Ys+z#5AzEi<OS9zB05%Tl6d`7y`bVyAXHn6tCw6?{Fm_=MoWZh<{>US{T**
zTpd4yu7_9WGBqJ1v!45UdP2qL*}(I7q^gO77{|sdW>FMC@)Mb%5<Nt{AwH?($L~ar
z=fr1gE(TPSkA37Yen6&tIsEtn=A8~*YGNP8AKE{hFYt7(tl5UiPQrr1R=7{Up_eJd
zZhHGGuR#qqslPR4uUvivwlqO<R}vAaXUSMT>m{=`IpT8(7xah-#%MD?DvBM<eRiiK
z<A<6nD*v_u#J8}=v1C7)Cm5KD%<|3ZqxFp1U*BTGG$Fk4`ks$@%Z1juXE!z)j*&ub
z35~L|={WIt#=|;+V9M?EaCFz<o!EJ_G|)Hv;9qf}gWz1imLax+AAO>}9%hzaT7{^w
z7h#abmMeishd7<6a!Gw+6O6%%h(fygRjF$1I==oT_e3(5;tNO0>11wx%+q9kZ$&Zd
zqXTF}U4HDA|1`X?MlI%65Tq4d!fd$_dgd(CRt7LY?~z8Y53M!-Xez>Rr%Szg9}nJN
zl|d2GhydyyAkta`WSws@E!aGNJ4^N)H>o>oBd_E`DWffJ=hkc0Z!a?8XQOV3H{!I3
z9CNYp*NgE^W9YL2FNXfW1Y5ETWkNI4%_l;mz-Fymz+~jR`K`Pld5f{|R9#u^HP7pN
z@*S&NLA~lGR2e4Y($RFQ=CgtnyvD)etli9%k$4xZ=^8&QJydPqWlUh3L$Mk~ZaVD{
ze6Y@Vim^aM!cr?yDqT2B-F+UJbLusFE5AZ<Y)f)^U4e_+o0fzd&*%&{^Y%FdU7r`&
zW`d!!ktG}HQgLBkc)OWHN{tx%9}>a>h)J-|$UWTgAW7pT{PJM-KEF+2U_qD(8%3s0
zb+NQ))7ySK%#GEPIsA$s{uj0_n+k=~&PrG#FRpo_`(*w2_;#G<wz|D%N48u%xRHQv
zqIBQLAJgb-Ch@azax)rqVhh`l(SMnl)>t;_3C{GT%4)lNN@gQ$LAPv^r}Lpq&z!~T
z>YWsK`ZwTLqeO9UG<*jo#$?ao{DG5@*&Xy(wXBq2K>CH(2TfoX0mSPiKz?9UqPaFw
zD@=UsL6Wf6`UA2t@c!MCA3L25K8VfZEh|4c6?NYYDT?O~N9^Ck4_6wCIwfc&_Mc=i
z((#RyKMU+6BON2F<HQ!}+G3qAzd>K6f8w`$+HWyVVt>}~T1|uG{Ske2sp#F6OX?5G
zt;JSP-_CD2$%`aqh-dZPi*dpao_v5zte1!F+G`vfxL`*)jfSDJ$QMpsR!`RZO4?4o
z^P0tenhSb<xkW~21al(qq=+WKi##Q1`MNL{b<H9D<gH*7ACZQIjBr-jQ-<<C=cVs5
z>k$jPUtv2JXfLZ-mXs!PH`Q>obDy94ZQs+cHB4zKt@$`FYnZNt105sfDGi1B25z}e
zpknL(3D9@MvMR@FQaksoD9DtSFjxd(-tuVo78cdpzKS5vwv!yvyc(umqF4q6g-5FW
z7U55+{q0{~AI#R3EmnVQwiotnn`FtCC#zwrhRI4aj9)!+2<8>-m2z733`oQls0}%z
zs1q)}Ql&C^_t`B$@rVih=c@Rqu}K#*8Fy@B`WSSoN+14?9qoyu8S`D|#+Am|PE8Bl
zbm+iDElgwuR@I~)F&fvha=(6PK%5{NDzL6(fX0$BXkjasCAUg_xUc`>td3(WLCRCz
zAqf|Su}hwL(>b!P@rnc%@#+Ec3Y3viPK;TPeF6(*)~gkMBt)@XZOG<hwFp#9WiXb$
z&b=7e&q7Rm!eIQ3o%^b~0q$8M;>Jf*51+qKtWaG^4xD~t?rKH%zDDY1AddY}JYNiE
zr@SX~M#kI78PB6h8XSv4#l7C<<TQod9~G<=eKHA+=wDi?XXYgYza&(ZB9rBT5uL@X
zqvMK2^SpfqA&m}Y))#Z`Ou4fWf7$9&-NBCyP*P}Dy_}{f!3dKQUdbIk8jlZ~2K3JD
z2jb+H=*nnhSCTwpU?N*PsP^&fA0FY%#qp|1MKLf@-cum!m?S$eG0x=~*IEe-6m(gd
z#cJ~j93vvOV`AEv&q}a$fvVC)^hx-I)@mkh7s`h!(Q|}mLE;KE3)6l)QSN>$DTtAd
z0a`sn7+>TKROW0)jh#-p@#a&s@*4Dtz(C%ZyC2mjw>G+e%4M}3pY{n87c3@apXw?}
zNxgabChHZ!Cm#ak2}a2zsgnI!oYz0!Ub{VWH~RWRmfMfMj$FW}yY}GP?e8)V0$9k2
z4?W#8G?Jm!a;V^%a9FrHdy>{RTh=D%s^*%us^OZJ+k3FQI3Z{RTE|VM<u6gQ6ax|i
zw>z~iaIgistu5V-o#T|wrJ4xeRnBm_V>hlZbK3bRc79?@)*p9pAYDA#v-#`#ndafA
zYBO}B>3hRM8V)e*=qQ@h+FE4cIb75)<@z|kCAB)kDIIrjqQ9h|G+|URnD?--mA1a1
zw!@S+YVjK}CKbgmi^FfKCmV9j0{aH8&Bs$+tDIs_%Dmp~e{v5lJg$9tWQmN^L^WWy
zGjZu`MdtI6aytggk?YRRzB9>Qa(v>ylhM2Pc{0&YV69}Hv)$;#K(3mbwI~{OX7`%m
zn9g=L26wcpK)|d}kG|SNgL~B&9`1%?g~5UZ@ohny1l>yUx3wvIazpe^f(l9EF;ZqU
z5ANN+jm1;X#}eenA-eL3e5${kZGG}%3xl@(rt-V|zygaq5m=k)sg?XxTJ7&Gpux;t
z|MS49T))rbxk#Ss@r-)?A4s7@IC|y~R}!jnKbUB5nzwNZYXak>(>qSH_vcWj)=6g1
zF3ftQ$<*8?^1hA9W4TOc_v|t`ZnBRh@OhI7#P?#)r|s=es8~xTRvH+n>vh&?`_xvk
z1bsd}x=d=^G-1Tu$IfrG(t6b}E5cur<ltq{(!O1^>5jn9%~+wb98)>x$S+sf`Lpk1
zH+`=_qYm<AyMxFE!q-vk89GA8*A#T*o`;-vDQ4QtOE;U?=X{*n*ng@RBqrR#_VpTM
z7qX`m&_Vau_qjNbFxxgQ93B-gHY9ZSPWdlZE@SH;>B<C;;_L1HJo8FNmfUFznIH3u
ze?^nYzXz?X2pifnMv{?bH$&{N<DayJ-XhHs3~8WT(2pP+)kc@mXp$04Bt$*pJR4Aa
zIi#fEu%(lYc7ry<Qa$Ur_$PVvds9}aXuHH{(shC8`JrS+9cj%Ai6M)Hz7<h7?OB~8
zl$A-<WF@>*FR&5N=1tXxSx!nwakJehVl2yJ`N=s;x5moBHHE(=uyOm0{4m?gk(lqq
z`->M3<BmGSXrD+Wf)}e6VV!b$qa=|tt&i=ZosoadarKj3TK!iI>o;bS6bO7DVdN7L
z7!Wxo57=O%iLcP)ZwzN~T71yyAYBg*Ulorr=Ve@z|8aUK__^*}qFxF1k)Oq3?AKWP
z!4#%}=F=X(gV6=-qZl%ip0-=A7kD+s1k48$79unYE!ys{kh<TPR6bC~IkMH;dW>!?
ziH2QS*b9C1jc6{%#y8l_TIfzYCL*=GgvZYL6*-PEtYZcGQWj(~W5`i)kG1p?-R*qL
zsC7w|{EFN_e^Q<L+hl&VPO`f!om@Y2PFd1NAHC<57mJzT?pg14Ze4t*qLWy42Iaak
z=Q>iT>2|zCVYU2N+^mQD4}`t_2Fm8zT8hRQ`psjtwaM`S=~e?hVf-6xH+l9d2I?YI
zb)xlKd@U>&gh}{<X4i{Z)(k1$y~@<7@+Pw~UWPQ;JhWYLY<8yAFzeL~Jrwh)|4+Qd
zyK3{Sm?L8bs1|MRtz&_?C!7-(M;P=!p2+JSJh&#)>5Cyw;!s@5_)_+j+*eDIqXRjb
z_gPc*biuz*4^@=oXn}T8$1W|$D$|Wg(}4w0$`WO3wSSf`*fTsLUBcE)Qtd>z_iMf_
z$XU{>o}z6yb#|uhmI7+bUyLTjfB2*ozy%ojJo)v!!hm`bmA!uNL?fVJF6eOiwm;!s
z;b1huvxvY|0#2^HTM`Ys1{IPQCvkj)ht;yJpuZZpE}>3!2MkCYACFNo^}FKWq_nO&
z3Qr0+CPWs-R$q{J$HvPt@E^k?Br|%QD!NNyQeIz}(*1K1>WPN<<buKAX6-|J%KE;j
zKzGN&{;Z6GO6?h+?Yfe+5N;ahrP^qhAc9whL1^Vk3sT&r)kk$PS4j%9#2P5-MDXvc
z66S0uRPy4gq<oho+ACcA^8GfC<3;Le6=#Wt&|6IP;T8BzsikOE_FYUfq%k-BXYo^o
z9Eja*Dcc_Y2m_dMcAxxX13xgprE-t$?gV4WR>F<CIJhIAEU=iTs`w*%Irrm8hCZoB
z%oUR!dH31=-<M}tdyK;g400?sKeEJK?+V-DO@1S+s0-w0rZZ>bW9O2|MvG%x3VO*C
zHib`TIb2+4{MIQ9RqM=3mbl#ywgNsc^|R5Cw}q)~-tDCqofiyt*NQE0BJ?92a1o81
znWUZ)=A2<BF)|}(NEfQ-VNOU-5Nn!Q3F_6;it9`WM$5#j*ktF#U9aCX6t}rNNxdat
zfAvdT+Yw*ZSXd-oWBC4;Pqi!0*%!{tzGiMr5URD~n9WMtqkP>u8Sa+6`0(^WKMz(!
z*yn^IDWN&$hgZIimfWxB!&WO(=$&w8|25pVpWxYTxI>6FEukZeyB+`IA~D(Ec*EIA
zoELT6!Cx(ac^5kS>+PyRjPr^~s)TSS`+Wf?yqdM!kF|zdl@9e9OC_~s<GDM_Zi!*(
zcRVS*l63FBZkOo2xJlQ?{iK7hzaO%-J$+^3*($&L`1|QKMWbDb6CpPmBUe6Cv6p6x
zy8^q<8(k;l5APIOXLR*gCv;hDfBy{6!oKmd<#1~lBMgh=WOhgQT@v5ibUDsWD&z96
zksk~QarHGgX1=)Z@wIo~%=6*$<CjW^ATsjBMTnztwL`3(CTA?jZ-(TkdH8LFP8Vro
zGqXX9_T%WF8KF=c8?E=N=W;^b&0gT{G?RX8ZV5_OoOyB0PTqco#HyK;vjq!(2KWBV
zIfBi~;=&Zs&0D_vABK)l#r8S}6S%oMNdlz}T0cn442#P}%>Z9JY?1buA*Pt=x!ycH
zRCldx&FX}UXz(GW+d)O68)4Ove#gxuvh;$9^H66mHm40q4b>it5%73=v0J7eDVAxa
zsuq@;W0?=reMA_)GjvjRy_{73RKVulnOE{0aFZ}viODb&r}$Z}DSd<HMr;uCKq%_7
zxKE7cHh2C`M_$I>i)V$Vw2if+LaZx2R!F@c^#k2)qh)BixX>$=^<2UaF4AvNn!0?X
z&d){P#fg!daU{JgWAV(?#OHFPH(2@93ll^kN9jF|14}?I?mN0wo}1RdX@9tBpfSgc
z=uX(B6xC{fDSvpH+ju=C3U@<3OpZc-jN^PWdu9}FeCNy^vEiptg%xc_X=~~0ig7}}
z?m@qvY>{hb{<7APl_g&<8Eyf_u#SKJPVt#Ww+F*oKaKax2BbgEtRyxMOU7!lSW+dF
zrNVT~4DtL9po78|-;bUXCKZ0YUU$nf>)UbxzT4QjnQ;3$vdrsynqQO!hHFiKUCqr4
z=hrnC$j>t`s~~LFeq<{4`ih~(is}j#>0Wek!#Df1>~L*aQ;wuyp~{<<Q5B_4H;0*t
zT1c6S?o^VsrU}GoKIM1ZCaga|^HwI|IJiN2*->;ftx~#lEl|boqwQiJ#39zI7+q6}
zp^b0*Q`RB^?eA<?51W|tfKMD7WICz-VTp$WJ2>yH38tg7mP5eBF5`oVpu%T8EGeC`
zb`%d|rzOSh9H$CJ<8;5pBJ&t6%*8G2C%JeZ6tIV0l%lz5y*IFwro&V0wffXDRc6~e
z8PO(J+R4wmO0&3rh7C}p<GDp`>-&KY-OHA7dF6pc>iMNlKeeR#h}VQ7Qv~s+X1!QA
ziXpsMg~sCbMa9Kmlbg-BEU)li5aRznxWkK?oUk6LT+9^hd{vh%7lE4VT(t3f01QXr
zNP7v+Aq~5DdaWOqHv5`C^0YSk`;us|%*Z#*v_-Etj;l3m^Wr^ejTJPja|$-hA9%`Y
zmoh?SleA}M=w!JeeJYE!xsMFJZ5?y?y)=nWIGSQ!ADSD%Fo)9jM*bM_<n$#O;)D}#
ze=7H_j&*ZsHxbQsttPb`woyf`95%do6YY{;1GNWd+q6<~pW|B3lH=~kvt>yIk2^y~
zTRE{GZ9AbJuwi}HkGqxT@z{O{A0gWOF6j~dvB>|!*?U04*|qK8i4a5?B2glGLPRG@
zMD&sb(IQ6l7KBlPAnGVlMv3So2!aqTVMgyIdM{ygM(=gBe0%OY&;2~>{lEWbec!j1
zm4%gLxaPX{KKFSZ$M1+#O%~}oU0`>LGU~kap1Os=Zlap!{vhWp^H~g8Pu(Ep;f@(j
zI{fXDdLiv4>PxHmcDGpCKa^j@+;r<U4cqi<{rSL#?;KCz+ovsV18U=z`@Am@!ngW+
z(9OW-N^i|P@vZN@(QH~_Ch5*9O=e;_0_SEOmV#!(iFf6i=Hk$7WE<`|+{CftP(qf}
z)D!gLY<ccZ-3JD{g~`*~;%4g{QrViXt!${~=*KP&N4Lw~;NoX$_B2M-xiuK!g<%y^
zgoB7XiHi>uD@(zu(DHDWQTB4gD++4Wp*)jS{KE@2=HINZ=A3l8sK+DVi?9`TmvE6-
zthSFvh8uKzQzqr?e(}@f`|Ss>dOb^ImUanYjkx~gI`L|Cr$T>s@o=a9O{Yj+w@B~z
ziy$`_Be=V(fmIgndpXP*S>ZBnBo=;O^jLMLCkf@T-ey_6{|SjW7JHP&kX3DNi#P5`
zuSeZebY`WiuUevGvHj{_+JBX6cymvuTQ9L9q9rJ<DVRvFD^0hu3Zz$7O~YkKNiL9n
z*9|t4>7D>QiKgH)H))qc`(iP4VO=Dg)iLV%WyE=wY#thrYq)C!9ps6AO4o32sXsnO
z-;kT5VzU_G8gKDC1t4f@DiMi6>Pyj-68wuaE{N8xX*U20B$%3&^?adjqFKkYcU=v3
z($#p|U*vLAFdZ&n0(_jSUif?|`sK#X53Z~1$~60rPeE(8jly&PO7LOAJtc<}$ehfy
z4oa$LF=*5O$|)}ZNITg$zJmsa40_|G-|)#RPL(^6d<nEnk;PAk<n2Q5sPYTFr2A4A
zDL~@q#m?;tj4WjmBj(?HkNg#KXJ;=E3eW}c+xwG=JbJu=bq7)PP{$8t5kE5)2~Nc7
zc<@I-$O5I=W7Ko2jk_bzB~yWm8h82pj*=S$cyGkk^%$rSDzzl3n3GfG@CElJyx-I|
zx}!6eYEw|o-jc7+7Y?Bmoi{4sRPF%2MM{zn3UDJWm-MxyM!SqtXq~eSwSaD?JUfXj
z(Tp4QG+mmlW?8|{jJwtTDiPTNqOz<3t`m49ZQS#k=U4)Bc@?5K-LR$Sd^xM!H#UKy
zrT9))DJa!vc;QCw#1Z>$zRy&g;YL(0KkPh)<dby`+1`ymNe!p9QwJ|P*EyqU;596N
z{m4;jU_(%9h^LpAO=4Q%x*<j#x6!y0;Xt4)Ei%{MjRw7;10cs-&*?zBrh68ww^~%~
z5`{6E?r;bTMLJ6pS=EIeu{Xe_b?>lO^Y~+n<Hj<3lU^8E;z?Gu64ASa!22*vIHh`V
zc%T=Cq`m)x+HTqu<<R<%9?2;3ZdGf#y}%NgLOqET0YMvTPUdHG*vR;d3IfBoi(;ev
zHCf?7QnFW4mfc@s%0p?aY30~_m-f@OQL8TFN-x}B-g@sH;M|F-RM@6i*HC*zwNGXe
z?BNm||9i)=0WKbo@Ab`oi_|yj0%qNvvz;e3zJ{V5p(PQ$rODXyNg{iTv{^Y%VyZ55
zkEO)4`obk*3wT8r`qF;Xo+j#Rq=bEAoTW+F^&<9SK$L18i}cX3s;-lNJJIi*34gm9
zw-9dWwjyY4tcu@9?XKXtz)0Bsp|3dHt(<pJuodoW^`1XCC)|xv$WhHHwUXmqyRyb+
zLgR|C{m+l&4Nr&5?lhX<EmEmnsoHviCBJdmq))y&SBt}n%N-squ3+$4@^w#RF)c-s
zV0N26*XoB(f0o%5>dWp294X>3%$@N(nccqd$3v21TJ3&VrLkKWq%Iq#JQ6wbkV;<9
z|9zB5{7KB~bQpJXk2Tm=v>07hMGmLH;ON(qrEvqIN}E-3jPmfBRv^;Z<E>0gG5(=#
zA-}EE3u=3+b#2|MvA>nnP*!7)f3X0V@u=SUN|A<HJMbq$BJLJ0LtdMRqY0(Hv>xfv
zCf3GkSMb616b?I??)3LKP)$QTXP%C5FuWK?7&+vvB6^y;w#R9fSPjm%ej!u$X0v)|
zs>YD<<x9=iL^e7->u52DUNS<Rax_w2CK66E`U$J{2xn3s<A{{+J3%x3d9!h&OesI^
zqWsU<H8BXuM5)fWlvH$}OCHHbA|ceRYnXMfoveGjFr1ST-uC1IO@3MY=5?#+<*6lq
z&xVZEnUYf8w;i6cxpQUS`g^lU100GEx?!f5usP$YGy^lyD!Nj<ssrwl)(jhjo~&%J
zB$m{&B^MUeGR52{7dGyAz7{y%HC#xUrEO8*TRfiCD)Vqqw$oiH^qPIxA~bO{+g;Cr
zezBP8OnMxGFhtSz?cYOf^$xh$ju2I>?GZQ9(qQo(TNpYlwvWDg=0b>)4eRoptx<kb
zx%pc85+$Ks`dfmQ;PY9PvRu{H<6GAAeLO6gs$WtEm5gFS%;*Rg<+NN_nHKq{tnEGf
za=v#p@zC{LC^O!9Jm0T;i>8`4iF=|)kw%;x{gQlkl0CfKy=Gc6{|Xw#tJtmotlyD{
z_bsQ*#<JZ-*NH*Lf<)P97+?(5>WQoMjhychG!+yyy)rTmW2FVP$*+}#^!UXx4P<5J
z=cnUS@)O_24Z-fx@cAqMH^gLCOQPxmTeDDx!5;be!2oroTZv7wDx($?8%_1srhQI4
zAr837n4tc(<Bq{Zy-QvC{KI`I{4Fj5x)?rH3%i-x{!iwtD{(4jq%Ha=66klAVO<U<
zO4G4V<<dNJ*lQ^F4Dv?FLXl4eGA(oZ$pR#EP|=s~%iVH$?w3Wo<UQHGiJ3(qy)usL
zC5X<*b_paRnSid3$2B%4gGAu_Zuf)2xgFk@?WY8V{g#G4kC}JlnT15FdL0eN-XHT0
zKg{uOwv-vx%`w~f#rdiKAf%)H8OnfmXy!)J;%k~6^NmtYxA-o9eK6({SX{VFD@bFA
zZ0>>RyJ;FqENV=(&n}6ziI(vT7AzZn5%1T2<LO9`{Twmf@;!Ary+mx1K*9iGN#13&
zxAK<s!5INExX0+(n!D}LMZ0Q&S!SkA)x8&oUqda`sa;mJsa*Vo`7!=G7g1>5C*5us
z75E9FLcwE9jkYA8v*o@@^8L_*Tsz57?8<)febSZBi0+Js${TM7+<BYAg_VXA6Mt`r
z=rr7KL?$l3U3Ti-$<ad%xCD81D=)@-RS+~8uRVX$@#YZyJp1XrKEr*LH1fYY;1qC2
z6<?HB%RT<06_GNgz`Dr#620V~r(})oRf1)TNBlW;g?AeS;i<_AJ+x&-b;CS3sbO(2
z#@^cbf$fr)1nD9ZcPM2s?VSB!vp;S0J&TK3^UYHMTnrZ~bumBv^?q|bcu0MoLlP`d
z`LL|Q#<z3oa@cB&VrKYXC%6#ey~;EQtLriq73y5^<!u3KtBJ11D(@=t`Xc7ge%caM
zl^sP9Ubblu4^ef5)Oj(A&%YOkoLAAVr{PJlyC(8y{Wg`&(Q)9a_fjJ|4CSg#4Y6TQ
zPjIvJ)F2JBL7SS{ye^%3yez|!yfd<(&+hp3LD+3{zbwYS$uf*M*ROVg4v&Hwg1)Lc
zWK#Tzc<mM=HzQPOUL(vVVQOReUQ?L%vcJ?#+LN#+)2(M6%D3zu7A?-+Ks*Z8qou-J
z;&Uh(Z6FbzU>BoAIYP=x{ZE&R@61!<3Hp$TXKd&Ype|gs4pef~$o?v>;j-?H`r+2x
zargX<>jLMXlWTP}PuaBJm)6X-Q4<BmFHy%WGVHR^b&(!X&0?-!eo^vRENobse*V1X
z&Z!!~);@omdFK;k@L|i^`8!V)Pc11;tM6Tzztg+l4_CnNv4vW|hw=>%hEhhVF?V`!
zCkl5*1XW_?EN>U}`h6HFB9G{|()E>+Gu6SaiuLP<W)mVm9wUjUC*Jm`YlsMR-&|=h
zczQ$enilm$nps8#Op0VbI5%50dAM&fYp+j<O)y(ErjAdeFPo_IBmO(<4)o58;6wD8
z_3djdvcc^X*y~0w_)=o<y@_t4T`ZqpdvYz3X<<`ld|u;=#rthBYvaxar#L<AUDd1U
zf+=6xrl9y38QSdfg^IA!1QWa-e2I+Rcnm#S&<GRPb~i)W>lR<=7M{NDD46Taxo6v^
z>OsP=S$2G-1_1e7pCl>xg{!&Q9z~kKV(PTIX6@9<vCH|*y{;|a;LRG}c-_a#Bl7B^
zDzwiATcEzH+aC-;?#~0+kB0DaQo~^jQcZ1<>RAUw1B>Z`#j};?hD7L9^y#^b?%-D)
z)gj`PSjx^UOAARL+Nn#RH%Ylokn)fdmEJ45&gV@&HhiszDXZ(If`_&!DDn4?yKY{F
z<yI;eNW^^4M&}$>AJ?;wsazGbav%IUjtz^HonEwC%-|iskW;M;hOm$2ne>1Am+RlS
zih`C3&IEZY;haer3FJWE_aDI2QlUT1-Y27`TQlXxdeW()A<M46IP6Ilc#gI)tPqZk
zH_T=-LS!YJ_yKlRj~iLVNQ{V1gQIP->Wm{%y^m-vQ6K2=$YxiW=C^cZq9*;!A3#2n
za#MC~S8`MESvTNCR2l^b+7?Wiivbq5AoYzlQTH}NYC5#J6S7mV!yIvxtR&I<JP-&E
zgaV@NL~3~$11>`+GCmf~M?s1flk61lyHz|0fy%#h6irn-`l0@lK8^eBJ?pP*sf>Ri
zzu+PA%%z`8GEP~16+fnUFP;>b)=(@6uxu8nG=|z1Go@F%(p+Z2GjI5z;u&7%_t==4
zad$Dh257t>;pcH}l>LLwVE4`5dw<Xr_V~u}kG%(8&-ZtmW_b4nC)+21RyU)t{fDvC
z8C?Ts6B{5s>^$P$c}>>dZ!l-hJ*_UX_vKVIRq~;{c~wZHuwn}zRx<A1)zJNrDx6e4
za&EbUibS$<`^eSyH?^B!)T*7H+_mN=$!LoSyg5O9R-ys)f=1AEm{YlA^h#K2Xxxk@
zBkiGv{rxx?_I*VChifh+<6a-6?B8z=vb5aG6zU1J(V#(VtG^4p(4O8I+QK}t@`ju0
z=X~^IIsBaF*FliEJm7@yV0p$9QTT|Dm8_*HY%_<@J!Q@EG&tu%#}x|_^^G^|{SJUB
zYJvWN-8J!dh#)p|tM{LBf9bL+FDmK3rraRb(iLMO$<!oQ+3K}um+&NC=>rFQNE_SS
zq$i`6_$Y?A*i<(0J15D_#+p+ms?Qed&jrGp;&)_x>RCx<#klXW4l1t`kO-hFH@b73
zFCy9;k0E1J$2TYAvKZA{bJT)sDP4AMYCq>ldL=|ju31YdLepF_|LHZWeMxmS-hQ{i
zh!VE9urDQkd-Zxx*qGXjT^G}Dt(WGL3B1R!RimFo{1kks3yN}|oG~d>Chr)b)(iDS
zldTHL3x2&6(IwO}dRJ6F`mztxXXoG(8%~)_Ybg>%b6o9g&%Y?c&~!s{EKV|S<M0`(
z-_es?P{dDf0?E|zqC_psX4*jCJUid%uu(v*A%R|iVEJ3S7M1?>Afn*9FgCC&MIwC{
zxl$g!EhP!Fpwruy_{DPzK4s{!J|qI^UH_=DPXFwtK&~_CU6pBU(3|SJ>d#A-Gdu-Y
zeP9@wBKBb685z8@n`Tmmqo(-*mWnMG_DXFV8|7oS@e87x?ppB0WI1OpX6!xq-6npK
z56O>Nqn-r(zE;|_BZ<ykh?!4{<+6R{*(ZRjr&NkPOrPC%@3utuMV400mY*^{8&sU?
z*~i5Vd|!GoUeudn#9_032ix$2$5HcX<7I9-Q!&$u9%VB6%N82hhQit$w0y52*^YvI
zce3wCR#`|%usGc8={pOM8Ba3(*jSR7VO6+-$w(+PtW9h*o<xq1u{-wK^|7maPc#(4
zUJ7)}bYwi8=r;1Wh9`vyw&F`*y1_n}PX&(ijhT9je~-I@>)e?6U&ERF?=T)DynhsZ
zwTdpqn?@?}nX&PEnxk!`ns`J#(eP3K=65yf!cYh=j+|KK(kw^kC=^mZ?T@l>LvZr=
z32m`IRVm4S42zHgS{^n73G?r%3W(XpTO_IygWpr({OPhg-M*jH^lABhYiC2bW?d&M
z2x8$u?)Z)$d(t1?#HIW|uWbl(3O%xzWucL|fPHD2fWn`FThMWfZfDUUgsUfSnSM_c
z6{f)3xy*XWbXK;m?Lr0ga%NNzdqJBl1Uq*zs9AOezu~IBkIDK)a%dp)r)rfh$Ol)3
zziir6i5IB8mVDl$I%WA;_Yfox>wQ+)(I)uTA;<}jy?uOp%yG;I-791rdHwF#ey%9Z
z@?jiEC0i1r)N|xbcQ_`QJXgw8Nba5|TF&gU7av^ic}#-X)-iC=M$l1RRjFftx9pUp
zr^?BjlU3Er!S)({+w1yv{Rd(d$;4aPn+^R;5=liOu?3%gn|sO_bqg0|B7x{!sRG9y
zzC(gmc~5>kKHDU+&D`5I!m$%Upcs*@^4Eg&Gdm=nlQtVPQ-#05fttq!JMm>|`L{{)
zJZS+uq0Us|{R+(-dwcNr+1nAG)3iH9K=Rsx%`81)m-J&oYVn!GFwHH@<?b?2Ui1_F
zMphSEOghK9@ffXb%;HPHFlyT7{_K!2ZnUh{Bv65Ah1D8PrEd~a$<`lwl+{_m+Uk@l
z^sA*zG(1_$yKLk>lgu;S>RCo*>$>kOBvLUPi;eHL`Asu%El4-x<!J!Cvj4%Yll^p%
zv-PyRD7}i7N{g?J@a9(|Adb$S+?Iv#U0Jx4ZiGI~TqN8eX+aX1Yo<M1pw-vXE=}B8
zNM!Tj+hI5-S_vJu_6cPn1_H65O71a|V={ae{p3YF!>pNTk#F4m_amS*CU@*mIZP6H
zy@t$07W6(6mtq3h$2)=Hf@h~r=&DQ7JuSvyt77k`iOzZws>>PS8J+`7JvW;zE1lQe
zPCxd|T}&uiO4NU*b-9G0q%hd@E6|)yFCaBaDpgDWu7k9xU@x^?V&duCBeK1+erK&?
zA}u`)o6I#B1THlE{I2f$QI~o#Z9E6@71l%`ekc~_G=SV|bF}!}J(}|snd;ovBGBEc
z&9@-y^Xd3?%k?!g@<7SE4@uOY<1#UXm@ba-j0v`+pW$P&im$a+*+{|f?->qPf2fO;
zv{4L_X{OA0vYFO^KTPa)zCul*7xASIGKcSvk4dttSk7`AUc`RFGNGk^WXaYxy~XRf
zaYl6N7MFWvZ|YlW3;`)CaAUDsRE{CHJG;EJJPlQQ6rrVa<;ZcSo-VX<mHwbwnhz!^
z_+}8Xf9ROo8x!stV1dywj40+^XiWH=`X0j#ol1%_QKx>4FRW7iD);GDKYfp6=RO53
z-)^#pzsK#;)DnWQ>=QD7q;RuakhPFfpW#rKl7LaQw@^^MkrZ5ni?4zEK@fIIX1w#e
z+hv|eahC@^XBtGO{2Q{#BVDTsl4Q-BE~`y$)`g+i{=Y^Q;K)Hsf~Qp$?ApXeQsZDD
z`PKehTA)|hvl#99fu(z!P;{4|9o~IN;Of-DlJ9i7&a$h(z1!@t8zjH0K)BmBKDr6X
zz)&$^qHklOsf)X46uW5Xy9nJdRA@!X9k`g-B+clHWCIG5%poR5y>Xy^tbS8ZK;z86
zh1%~Tz{@xD>g(}z>^db&9Tur6n{M!Q>;dx+*h!I;@id7sW|rKkgBt_M!$5908u*L`
zuVVEQkpPLjR~1l6CF0{COyigUaqi;y(8+=9nxZB<uYnFzw*p45nVrF96KmVUK*Qa(
zyG~9!isxQ_;@z$!V)erYr%8Og$x;-h%^81@e)#E#%ZSE3w@ZfJQPcINoXCGHABvnU
zI9jLj?Gve_ZAKpl^(hYYl;6%+$~aZE+9kGdf<u?FwtJ%aT|8Ct$?JIb8P#uOs>*-f
z?#IM=tN~+P$gaT!bID&&x)J^dD|?IjJ<WuehL|VsI8}Kyg<Q4GHoS#kuTF0DgbC4#
z-iq7v-<vsaDZt9-VdV|O9)g)e-M|}fEvY+Nu9UQ9)HHf6(E6#~h;4kYoLU7&i@MPO
ziG@qnhmV(1xW$5;mJ1C=qa)MjDiN?wg#-cW*P=cG!%Nu^*C^{Si2fZ)$A*Yq&-6R9
zJ0TggdPnmW<wT_}6JQbS@tiH$Pv)zUwm3h4fqBZ0PM(<3okXU&Qu17Epd*>f#MCoU
znqP@(=}Hw}A(;~j>X_$voitZpEYmih$%xP{3KUp;Y)uF&j;(bz<#i^_;+-tQrRbow
zA7}pHUggE`(|k$PGYv+<Px)J%G@=}95FMS(85%tu-L70}iof_%77b@SjVMAr{E>bl
z!6!`wzK-H^uTr6+Mn%>NqPCx`6x4SSwZ_+A1pA3^Vg+BSkSq5;oyY{<k_<!4Cf|Vo
zHTx!~1LL83Zf#RBNES}g>v<XkEq+=QDK1AppyqJciAirj${EB^?MEKe)1YU)S?o)`
zlrY7FBY5ZBgU&4AB7x48(EII!1e1z4_ZVV0*717CigGKal3xjV%{1won-2=4gnK$R
zExArVe#i>p<rhKm>cqtki}Y_$llb#kESi>BDd`c0?efi;NYp)CM@Y5oGM_Om)F=2B
z6YZxK`;zd^Nm^A*z`n2Br`rv~3X`mal5@U?UDq$O;dKzf5R}#4(LuG`fQg+YB5b>8
zCe0+jxOK6&BsPl?o~atC=nZjI0T0Z*!=heC=%->0Ett=fL6D_mdyZ$Dx49rWcLRS6
z*1z&RI>X0OPnW6g4In_kLsaXUS&;yD)Oz#YT&!U|&xdBu3v;j*`D(w~1gbeG%^MxY
zmFLSDK56z2%2Tne;mzb@l7}~_df0(ve-WzxeeXs|TbSqoT4`CPJiJq8>s#lpw)V6;
zyc92mN0EV6=!aqBv2-om<?4=%xNfcSTF(!5GpQ7~`=t4qzbEyGNWU<xQ27Y66H6Qn
zVE3(^#<OehUAwu{qq%&6hO%Xs#4OgX-sq(uO-Qql`}zIwn%Hfw9_Tl+eJZy%Z{Yep
zaueb;SVTxm(zC+7ScR4<awjE;k*%;HHHXh%>H}=z2~s$8D^1rr6VmmJ*CX6%Q?OsN
zc^|F6IeK(R)}H<pw{8{{C{#|f@;_Kg^9OevhR-vp*2<K(K+vb2d*!l4i-fO7_7Y+y
z4W=s4Z!*@YQKADhRGm77Q=KNg?}`kQ?JtEc3RuQr>?Ur!baY)k<cGanokz$2UDFwe
zU_785FBx;b)}4N$?-jsUoqpMcXXI;fv}kz<JG8j`i!^&r+ISvMs{g04NUhtDS_R$G
zci%_GSEK72m@n49aGbp)Ko3beKlo8@Q4h`85luJykVGKPC_S+wt>40fZEn0PmC*Z=
zKLNt6S#~|~H0!9__nktPr?GTf`irQGkSEQAB0kY*wBV!jJvT82*J{{!(h>Yl56`AT
zjqbDQ&mV(G8TAp%D{mgMM^db?Gz5@9+Em1TTu>+FzQNB@H!lSzz+(?Oh3!!o6iGLk
zQqErU(dxd#^fy5S?n9MErm$ro($_{A+zwUQyuP#<98?oh;~SznrYHYMrTgH@UWm$5
zmiN*(WqmsE`KEp;N+}zs7eC?eU%fGL02xoE;)oXAxPts(Qo$QNY+cL-<2uRgIY5p2
z-#%j@rD!N?7THA+a}lUpI1WuHo=NLRr&h_{>S4lm+b_Pb&0yTyy?R-qf>>(TT5V1e
zKh0rD^u2crm5MP7KO0}jZN!Bp9nz|A_1oAw5>`91qS7`Vffne!YW?@P#p~4)iP_B^
zX#-;w`3ky5Y;q$Dg%j5+p2>^~Db44PNT}TGhIj&Q*nQzyxz#}5qFar(0GEyw5IejF
zyJQMM;n84MTqLVZ;1ppy=KV3?baI|$$9-HjR2qEQgnB;@suc8<s=8nJr}wBvP@BKr
zG(tcep*<gX{4@R8uLcVD>`!fxXOwO4TdOW~PqmQCJbR+0`-z#1J8qwfVyy4-f}G)W
z(2@wj=+$FNrbPnpWlzav#Lp4t&mkhwi*dlK+KYanE#azhdC*#ZWw6luSfF3!OcC1t
zgNV=XOMJ)Www^HYxE*@P+(qi|=qr{b!h4)5J6e(MaoKhTt9)%5YDspkN___-`F})T
z>?>@`)01i5^@%6`Cy0a8^ROPn&l_StbWSG@*!ZU{A`IaYx9eTbN5#!D*vHRf*xP+;
z2J@F&Iuc7-OPtU)XfrI91M~WrGxZuu5~sSk43{DB-cyqB-|eU3%e>w?g4Gv#{$*lX
zfy|IJ?+OS}{6P0Cf&WUqie>Kclwb}t@T~$<(+>(f_kbM>c>EZ8<?ttHNlimL(v0S@
zNbVv`qdvJTb^oVYLO1VeEV<kz5Ot^bYgx{I?pK7naLVKU<9_BSvb(R?HV&&6A8v=W
z-!%9oPTG?x@w;+i-EDF?jxivR-@>!Ltn5ht*E;%CGC}Bk^DvRy$L?724nmwYhkX@)
zpL|B3M=4CWd~?Dyxt#El$4KSmbwA@219z6vsbf(qy=dC39vK3QkD`C|b2Yt>&oZk`
zc@ggRM>+%;KKJcN_lv{wc@1f>m)a8kaY~;mKB-;0Y@0-Vq@m{4mMD3Vay9Mq(@$2p
z)i>^SiT|BC1T0--7gGbC{b}j`<#`8k(IZxuvXAw!2(U=h&EI;jPmXOm=)Ld(G5MLT
zH{_3o<}W3U93}0b0w2vb&0zx%M?rO-U#;7xa)s&6W{E_-<w9|8WF_-OQEncZXafl2
zcDKv~eZ<}S{uR_W13Nmz-JMLDr8n3OM&rxXc~1Z<(7;`nX4rzaj{Yl{dLZM<eea?%
z$B0x0YZfxQYQYsQm*M7@u-2_{g6%}af)MK22(oZ3s8+0(Qt6kY?Jf<8kdJ`iE=`&s
zP0w?N=jC4Ns1o@AF1ww_4>U;Uhw@ck@}73+uR+lh+vCl?l<_1d$K6f>S67I)%M|K6
zjjnd;FX>2R7Ebwp&zWl}X%mN6nv1f2DE}$EL34QJDz+K+!$ZBT8yH?3##i~Q_L|}T
zmti)cl#qd!5$V;r$&sctZ_i$m`+JxH>R#t50r)R%?H&Y5_EsC2^R{<dL2%$b<XJP5
zKQgg-cVP3cW%}vJJeBCgd4r|5*81iiA^*lmxcj!B%Zfahn+ZHT=J`V)q6)b%EdOVM
z{r4RE<yAZtb;06?u(6Va(>EgzgdTUtkbzxQ2A2zYM}UFv-|QlxxC=c_8xyzA(XrOt
zX_XzF!^o1tJJ3g8Yk=WN2*sXK?((UVN5-=1C>A2Yg(D8L;5p~*Ch9e`ZANTj@0B6i
z9$oUwbza}ojeQqE6f`Q0UKU>-;4zBLvVgK^NVTzUIG^o>@S6ZV$J(gjeAEt(4q;J*
zIn6brc*im->n0c~{qWJU(l^bx^Scon6L*?y0GG~(=;5xr1EM=o?<1MXC%tnrIMoOz
z#=mT$Uq5CF*ormpp3l-AO201;JbYvmb!q>@+i+;RLHEmG=meL-X*t16SVgFGS3a!S
zMXZLl8?+res}(JPpwffLGxU&f@e~k9I?x~9Vur!**liS+zJu~W1+TuYJ-D*<c%gm~
zphIpfen>q%u-h&p4DYPwx3JT#Q7w9ewp-wL;u=fzc(g*TV8i6D@J-9|d<g{b5u4C&
zZ>296@g<V6ZrP;cnlCs`zwuJsO(E1VQ@YxPN{%23tx5t+jz_Re2cCOd{KKgem`yNW
zsVG+LvbE3v7T1jlE^mkKkG-n{k_+U8nPR<N#G5N7d<~NFaho6pGQC>XS%-S1B(W*p
zja-jWb#Og~Z_H!t(qYU<-$DO7Howd+E_z6&SZFFy0j2k@ri`v$vQ!Na!L>FYf*!Hy
z$N#ux>JrM9_GSJd%Hz5@bwotdHC;5;F9f6j#q?dr2`)}TAKwV%LcSXp!Bu5#au;+k
zWyYzP0tTsr;>bika+Ll6H^3rbNMsC5jxNE<;&z%@#&Ix$+uPf%OiYW@-|1cL++B4H
z*arTSo&m2Ro}Q#aMFoIZnIW`zM7aZ}J_5s5f8A=xiLpR);zdk(GousNgBjj_1uYB-
zRzOl>=O<fZ?5&M`kQ##ZEZJso&EWM5-UNXTi8If)^rnuV2aR{xSuX`U*OB3xvW?BG
zbN$zH`vn2GWqcg?|04Fj?7lm5hQFJ^?6q;E+9L2q0zfwkxemlSuyIF%MAwUv<HsHw
z<*hg;sDsB_0BcoB<o^DXHGfw>^Sa$F1MmlV2-V?-7#%J6-dsaU@x%K37r($L!LqEX
zE6DC9Ffx6*6A5@7W0L@4lZ3#MiuMeSfO?bz-BQWiqF;5*j@}?(q&*1nz%PU3ku7i)
zi_lyd1GEog03><csr&viP4FnlEj7jwWy(C8uJgG#0smDGM_<Bk2S>x3B(0*=%c_{!
zpVOWI|Nm=P&mJbR=>Vay+)fNUmcN01Z81Mbm3kDw12S<a374JoyfCb~PEop@KO8pe
z^HMJ)g98&^f6(|Ur&iXp6h-$_0PeA0cjfY&5kCS>*^>C`HH+d;yB)M8Pj~phb&D()
z>vgDab7Hdi>BKxWaW}EtEx-TE>(HRrjT3w>6AG_1M69QnQ+Z8=(FECb9|u8~G}imz
z>{lU}-mYkCx%gF-=Tu@qg)iAseIGtmLULHKYNpSw)r0s9I^>=%<DD8v9p1Yls5^9a
zp$K=TX8SSq1i(hDz@(cvX%v>%^*;(iya5d&J4A=(Q}mDKUZ5<=mPpDS=QP@zDxm~{
zzt$xuZ3LntpCe6cNi@0w2aw=zXAM+%3oZuey~boX-VQri{@yDO9)a;^9<h;HQ-6rV
z#_0gVXrCUN(ElG8#`}0+-%iHTZ{_o-r0{=5Tv!CTqWtW-*TJgFP6Nlux2Qc5if)&k
zltPV7qk;(xT>JX36Cq5vCHb+RsPP6^*+}9y?C!9)RVt?As6YRFT%vLijXVgq_sm#y
z-vm4&IU?;h%yTG^N!*1t_*yP3C}+D+0POd4JqV&*oY`2<Ccm`(DMW_KG}X}LY=1=5
zRq5Ai(dPrUyYl?8G)LMB0l_5jK|t6vaQUe|PFYm`V8nP-4IjtVAfq(8LZn0iU1$X1
zo8C@&FY;AEd|C|70iN$!F~nI3PE{FIF9?TD{SrtuO!?v*%s)n270{x)^45W{hvjs>
z{DJ_XAG@EPjSxx_;1HVmWJ%kRo|5QS&m{CREvXK6`zb8q@T2VRw?8*<8KNNjp6dHP
z^5ipZ&Qx&S^!)Q#*^fbKgi#!`*P2{D>acII@%GoB$uKA%5;ico3fN@&gXw!6HC6$?
z?Z-2zdLD*%(}AlWcxSmu(EgnR4*7<kLuJGv#U$<u2}(H9knD6&8ZrEn9;b59bMGC$
zJ8RgJ8Hmp)OXxR=;-X6eOr~l9O&d?{)7pc+Rba;3-)lP4Bj&gR6tlO2F)^Wh9VheP
zOF&fFpgnvtmLj=|6mHT&b=?H|;S7B5y)U2(8*2clB`hl$ka-KG<4_p#7khh-O7QsK
zH`}-IL9(%1^0lB8R-WmkVGGOZdQIYwDX4-0cMplc+gSBTw@~<-R4Kz-(?7WHX`ptu
zMTQI){*4DtumY63WUJd*d;dnxNJm_-*W1amxT2zni@icNT;XcUd^J{}_SYioi^?hf
z`Ju@j`_U}_Qj}q#0|-r`d(IyUZ=T^nj;5FC*8aJR{}Qd;VD=>tXd<Y_i%b6?mb(Jw
z7GxQfu?<)X2=w6IC{wO37~UviUfNL0ONYT0%Z2R23raKKd_?h&9+>7BUMz;jLW@Vr
zs}Iavx(yf(#kmjIZaj{Q8uh(E-l2%aMJfk2mz9tqTtz3{%#=MB?XJn?#2tuk6D#8<
zyTc81s~q&r!iDb=7(#6;0NrK{@U_VER0pM2s2V6zD0A$gdu;Q&@S^vtY<rI`IucvN
z(To`1znZ7RX<XfHV4IctC6!1gn9cqx1YvL*wRw7AiVuBO2`uHa=hMyX?GX|hg_P3*
zC_$^b*USyi43W;LgxG8|pcC|{maQc%0D0_1>5nAE2$9gOwd-Ig>OS`?>N|tt`}l&H
z4K=ymcA;Ah&?Z~#j%w@vF}N{AL*Htno|c_B%}L7msxqiutdFY<2^ai&Rei|XrtIm2
zh*Qk^P}sGjFM`w2BK$_YSTF46Us7_jYc9!Rln?p~HHV=l7B}IekytZHLb8$(sG@wY
z@|ohO0rkr8`ZddFpd;7%4;?wEw<wiWt*QS9F6w{#k_eB;&EYE|Hrw39fxD0SwP%Ok
z<W)OA$KvoU3kkyhN-#$actl9~vUd22QeyV4t;SckhLtFZSt;lh9qzX~-`#kHhYkvE
z*OHb-f^{Xl95LzrZ6@p9ijx=~&Z;@neiMZWeyc>+V&);-etOpqr|||KIXM3EbvoF;
zgO&cbAMWqJON+uU+B*UypBw2FV4vdWjz|Tv8lhX&Wx35R!o0!7c48@)?BpE_AHNj^
zs(Z(%!9uTfn}fbc?|Y*r+E)=aP;<}S<S7Z>{%rX@c3&=iv@%EDXDDa0!Wu4G;bi`D
zM4g4`u-NS_n+q=^?-Y$4gK#>oZ{@RiZZRgn0G82NIQ!>P!tVkOt#(x3n?JwCe;!f(
z=jU)oxeKeFky7VwTn}-n$<TEGu$gP<r<aWut(qarKQqHFfZQ;>U`*E2Pp`^gj6D}!
zKj|2Gl`=T@${0-0MOrXAI3W~@JVybMV=dp_%f>spCejSA%<kdoR%7^yU?7>uXdTFK
z%G>*~t)}10Dslocj4f~>q-Ee$kLKpYi4+`V#=9(B=ZKT!DRJSSZ$Ldf3c`a#`}HGR
z4_;-`J)i%xTd2zbDCZUaG46kFAOGwl|ML@mZzATS3IL3LLEPZS#H&xajbp+i!H5ib
zu#5mDU8mYKAj6}8#b@Fi7l!^b@@5af8$<y(y%D+wBpdw#aWTTX{_5oaEGaMFf-Bp#
zr|^yR-|N-?;4l62WnB}8Z)pw?oMy#RHhBTIMP@#L$cq42J`%Z#d;NJ9+~V~M%$|m#
zW^}4%ug@SW&4NV!3=eTdC;lw})Wzq%cU$-W4!Zk4(RcrR#~8x_{(YxWOZ(^lUl7&*
z_4O5HrU4b8h`YlU;F=f}_CmG(U;k{z(l|ackxI}XcvYP|8{lj~Mu7sAzAeQp;=V1Y
z=u)k8tt6^Jo@LrY!19M}<C2*~KtLH1mTW8a6fpcfKuD48%HUVFo?|VwhZX?S=0J3%
zd=!s_+nCS2O<B>~z`)oUe{eV~VqHF5*TpnF4n6-^)C?ZP_%>&V=F3MuP_~L#w8!i@
zY5K69g4Cmvb?~YX@+y3OfC^K6#ok`^->z1eFkmXSB$XBX3-c{^)tD00!>NqW03-mN
zjWM{QDHnkqCG-uzBS8}V5*JH=8@3gf`uO}^is#PbaUkf$1(4dqlEaK%rg|Mf>wbf0
zLRkF(9CRt=n6a}owPl;T0K)z9!L;us9NP?0;SOG690!s+#ELuo`Sr{Zx@SG+T-zb>
zQy|VT9ywy@QQBm)_qghsBPfd^zb<(ucsqZy8sJz1waY<#?P)VD9}JASkRZe%Vt%Lb
zK#J{g80+=_`<e!XYVzQLb9_eoB=pZUac1L(joriTLbkup4maOGTWbov_w(KW0@2t)
zypbI+7zE%-28LD@BpImUQzigjJl{4IOzm-mo?jqebYM(`uGTn~(D4$<+S%oZDI6ki
zW$J@6@2^d393t@-F~>)WRDcV#4(7xeXKH#z;2CUb$d}Hx9D1EAKJ0oSYE`GE7CRv0
ziRAPBuOLBmeEl809TCeo0MXrNr@Mh<Vu)_SDo~uIg1|o!uVZIia>86-FvvQ$xE0TY
z{7v`R^xD_|`YyA9TQxkRL0$9ju7P|SJsg<mC+#ZNMH<IC`Ehk8WIGOo#O@jH=j!|d
zM6j91d|Tt@(W+}WBZ@SuI`t{Y)O9=o!Qp?Im>$d=jPA@a{T>B%p*Hd9TZbDR6R&Jq
zJ{-)K#T}=%X><$8@z(Br;iHS=D1IjT9Iv+GYzg{0yVj@2W6~_+>e0};Ps@kEg4_RX
z<uiK>-mT)ye4Bqcfywa3VW>}M;D!P);>(S>TDQ0ISO+{#HoXrYcr?;ag6+-z85M@8
zRG2C&hh0VX_Es{#g77kxI2*pV%Sm4`cZmkc>B@nWS)1$#z^whx&Euc%z>9G!6|{f2
z^v~r2k2<gfO05HD-w*DNhV)MGRG8ohO4i&seP|aDC44y<O`w6|&RGy;7zpQxQB4$V
zHr)9TKgwmeeZ^y0+31D4_gMY?%w<4WBSZM&&^`Kx3w%E1GqWy(7pbAcGn?SC6QM0$
zZ|3uM%=7`!`XAt!>5^tn%C^P!8wXQlBP%QPXBqYe_IO4RSM_TRXoC>dAVevO9N7hK
z#V^(krjy_RjGAuS{+QhEzo`x1HGba!)_sdp${h=Nth%V-!T0zta(by{1fD>Y=WjIj
z2r#enj2QM=LCbqyVib4#n!q7CAL(t2WLiJRkXUG5-=RLmNmrEKK9X{b_F0+dGQvLb
z=tCF|QWLl&>~N^NevJk_`eZ>bu-;|dOaV9PPsq)RFSnl-U?VHuuMoI1PsDIwr;oMa
z-+ktzlwP>78<DRmW9J^Ygb1eFCSS?ahN*I>j>R=GDlRjAA<a=nR}R+NSwz(`l`md~
z3EdB>P?acoFetEcef^<Ju3qK1c{OW6%pY!A(X?pT5=rRiujkpcoBXx7aOOLr{ZHs+
zba&#%HiDSE#gJ5@w&ku`9<3r{x7#I{f~nX!Gk5zpXF0e-qLCpxU56$yoa-;Oa|RBa
z9-eCg6VIBJz2;98fo(>CqY6qqrJG^?$5Y9jk*|${OlrrHVRiPS9`}awD$cB)cuXMQ
zx#cjGxR(eQS?D)1nGvmY+yQ29@u2-CakGgg)6p8SPlce=|L4~FKR%(g#|16e-%j8A
zvr|e()D`-i**pM^8MCN;pp}5k^x}fk9?aGnN$o^M2K27f>W$UjhBvz}ezF8NNE~iW
zI!K;6Wv8~Kff`?IeASK%ydlA^Tb>_|;JmC9e~~^9vZc;A`y12w7M(W0`eKC(pqrER
z{hd{}2Jt!aKbt0$Dq=cYQaS)V6cK!V0I^aDe)1e@Lg3#xK6!woFq!XJJ@8ToA@ca`
zmm>J(=l!o&f_#jY^F^QY)?x54j0169xN|Xc4WPx2gJ#J{96mn7`~rxaK;cfp2Viq-
z@;kOM+`Xaa+@}BoxhtWb%-`;iOdYj|`dvDK{`8(YnWW);&E&ZylPm&D(oh5-Js<8)
zdl34eQKO)7<rJh*F2yATHhVVR#8K6C;KU;mD>)f!0H`Gpz*Dy91^B<#fYL)-Io7ah
zcqonoUvhP@5ue`ClIJlx85=QmO!Z_o)$_(Kz`&l13&t!R$}99-;L-@9T(Si{fy<Ct
zzg58Ogb^}0zt=EoV=Ble<{xvN@{;2pjr;k|Tb<7vw49WxrPF+SA&yK7<$#2wKv$Qc
zQWlan>wAejSX}NtbJ=~MZIU~cW56xcz}tm=94sEM|N8N`Q_n*ffb2%>ykf<^vJvNe
zZ8QO{S^YUK7Uv?nEaS@U!v}WNUr%D_7hX`WAQSaB!H*(b5b1G%W2v#!d?xSPz@=Fr
zL1GZ?x}vUCuAyTpE@v0y#>O|Ffg8g0o%DKXp;tam8T?L9b>&13^3WV;&vleEla_qM
zSs5JUUOCn7=i4htTA5MEJ4#h=huf3F1!1D=*IA{;?8cn*BbmS<LNR5gXs=&c>A{Cu
z39pM@c<{G!<+yZptlW#KWg4(W8vO!ghCKkRDf7?WTp7ao;MRZP2y=do!;j1PT*l5p
zjd<%gd9Ve?yLz)#PZ$&|VBm*b1&tOfVhANyf68my=+93jHN{uHt8C9p5newUDpyKg
zQF`nHo+>fgAcsgJaGMCU$hozay*rOTR;ktpl;s<)edFf8<s?`3nm-o)N8jat*2~Oz
zb5S5aVy^gE4Evww?+60VE2wCfJo(l0F6V;dMcMBlxvn&(hfwNB2AAXc<gl0$PB3g5
z2Q?+eH6Cb5N`teZOE?~woWysKl00e>#PIy?hAe3@E;lE5X1~y<x*A+dG84kM6$*8h
z09~ukfwD0Hy}b66If5E;YtkzIM5J6A`h+%u+_G#j(naqqMDcIhd@EL*Du%Fva~&FR
z2CZ@ZH(Vo+r9lquU$g)oQ|n+~3L;!sG)t1P<IU|k4$18h<lgj$NI|x)VC6}Wq!c7v
z0K!{(ybpe)Hk+P-mWLItl^#@jhgBQ|^{#;yvSCvQbk~gOBw-zJ_B6bM7lNaA6s|SR
zVl5?i6N~T?)tW#Bz=)N;B_mxDVg>2&g|01AQ|!He?uL`+vuwzrbxDKez=-{$W}e~A
zWPEUfK^X<_QQTvFm%+<oJI9#D`&ToC`_VQNNG)P94Y<#$^^JyA@Yed^qN>QRTPyhK
zDd5c5IvxHfboJ)jrO%v6Es|Q*%-$;ldXqbG&nE(e{21k-G4#taj}jq8r#`<oe+Sm_
z#a=SyZAwZ#-^=!|$JzVR*^nmMq;H@MHmo`M%bAaol`BJTb8Wb6aJ0W)51z<UOJ&>-
zKY2eFiknHkv53`v9`bteJiTj<aYpjXgu<=U$-|lv+_;7!ntGsDzFO8(@xe~*?~_^<
zPq}%y6WNK32pqt#&#;fYkr)mY7+m)W-6q?m=dymVRwP#vl*tsw2?4M_o4rV)@dFTT
ze@CZ1**ico>L3M)!R6WPeJWAH`d*0ej+K5O=b>z*P0D`^Ov&`J&KoAj8u(hl#K81D
zHLfZ;aV_1Y8!WI=FJ8HqJ{|V~DW6L8@^mcS@@0Yh+VV-9y}x@SG-_#ujCErr$rx4N
ziDQ`=RKeM#Ck?Mw{Lo74%!&#|mj3+PqJVeW)BYfV?(}dIPlcnM_}ZIz&jJdwKfZoK
zSMwPu{OkANIwGju_vwExa+s$PDL9$g-yc-IzKD83qcA4%FPrW?9)dJaEO@s%^gs4e
zN0JZ<9aR2gcKU2MF6mv?fn1YgI#$O}PYa^NqRWSnNHhHl(dBiwoIdMehQkGow(Q2I
z<{>oS%r~;coLF>}N_cC2>eI$mpG<6SjE=zkD}B542m99F4@;s(W6Sy3obsL~k_f&S
z$Ic?w67}}m^`~R8V!Tc!&&SXA8fp54P)qUUb*xS#;dG}v^$J3XBpQOQ^oivTx13ny
zyj@#e3l>=$Iu-mhF2_XaWrfb&;i5s9+m#p3Rn5NR^WH{8Z3rTp%?9jg?zG%@pkpsf
zyWQ<7V}ikJj;L?}hhsu}8d+!gP1abk_#Gz=%Qe^fSD3hC(@GKzA)1lqH#D3UyvWYU
zy)03eCN0-XBhHABL<Ff2$~3W@HkaqgL1k$HNEX$Q<Q1_UH#(ici+=PwTXO$vXEP#^
z1W$-&yo7x2+(p2iFmL#xaBWULR#Y0*KnG)oCF|MQWVCxNuugUcCMA-XjvZ`gw24F?
z8T5`glF?ukD;4zSs8t^v(1?=i73yzYKYNGhd$PL{cOpo0*}_hnf5mOP;<RAkcTBVB
zk_e5rC2v3{`iUWiwWd=+8m3Ve%$X7HD5Hh;;4QEjPxXEe*=V(^=ZSg=3L>{<7Q1kP
z{%O<sPJ#-C)-s2CPKdC?%7@)nk%;{Qw+4P#tsY*NmHosROtBJ5oudbifp#&LQQE(@
zT~K7vky*M~5~Kr=;cBJTJwRU&ZkpjEHs^Kg(nlwj6I2DTNz&2u7gqR+wuERODvdvu
zJI@*TeQ$yWl1T9Bzb2`$vG~#p{gVj;Czej@PUrX~m*5Kx29)LL;SkwTb$0wpuq(;^
z)V%%T(YAp@JwamStuA!4$=JPqHEJiy+U@uNiePHA%YyXCy3TYEZ3EA^`G%r^+%;Wl
z3dORaY-8QqTTZJ_StgxO5f&OvA_WT@33FnuCokM@KA=zneK+nX5fuCI{;1u9AWD%z
zO7{10x3=U4HLr<hm9Qo`i6$*ut_l&KPA5X_qV67X^cjsEwyAjwI;~nRTbiW?2T|0r
z`rZvS+NGJ9oMUv3Z@_jjOhr?W6chMq%rbSPAu&@WoE#fm_=)I|=4onnxo3zwd9;c{
zFZ+2NQp@yu%QE#m!YXi(XzJ+RZI6Vv&HEHBeTo&=+atsVE{_?xhNqO5qDsE5|9JX)
zfV%I!6vZg3u@YM6`gCXxS-p9z@7IVrh@#yY|Ck_(9U2>EC5ICZTa8StzgloyqeZS~
zS1ZixurF~;S`TCq@VQcT_e2HB><V0o<BV?H^Iz7N3@ZQgu5yyQ7Jlk^GVXdTvSN=q
z$yg6y@flq9N!maJb^1O1cZB8tQP5JrqbR_@<reh%zmBB$)Nt?QSDAz&07B4VCne$K
zMtjeY7%hoi?%*Ygh6c1$=%b{*W%5)krPIw;YTyU(Y*34Fzc49=bll(gAS{;_9LweE
zbi*`HreLD54zdb|%B-<>&Q;*RnEn=tbr4>Vz5MnaLAGI75k;_YPP%V?f@l*Z4v~+s
zV@Fh8K7JTh^f`hb-gFzDT3Px{g{wgV-?{lczITj|q_Jvb#VIIF=fR%Z1hl6){)32g
zi%uls_hL}xeKJIVg<r9==T%b{<dQ|Tz^8%v?6D-DflNjXqcR;zgyvMvX33=z8K_9(
z4NF~MsA@{0rpBO!9iS^tG=+(k{$%wGvkkd#DiaD!R<0jXG>C1~8_uL7c-zr|mdhqC
z$Ep29gy^4MO7w{VQQFx9jE;Q3@)H!OgP#HRcl!~^nDhNkp`NhU&2P<7{D6P-?WoU)
z^-%|5Km_iD28G>hO0toRkg0_vMQRkOSTklZee;MMURFxBvJn9dAc}3tgHbbLx#dMv
z?b&)xJa{CRA78Tj_V**1;LPw}!}!>4^9PQ4X*~KT?bP@q)hK8hbP~WY8&W|}mT^*U
zUORpome`a{KQ^cMA`G4sIHOODMd(_8iS6V$OQ@`YjT;^`D!jzZ((DET6TP=Z5NUN+
z^Ha<Q>$;86NfK7J0GnP>Zrp%Y%_>)5aymp{{H{6G`<0LXtSOwE#dA`&e$y$?>ilpK
z^i+z0mbMd}p27n_PqD2ZeuElxN`KlCmkTlOxFJ5Gf6sj5au52<ZoK9@)5h?S{ypmr
zJd~}gE!)(FVSmKKgVA=g<6kp)%uYJVm~?Wm6r2I3#zg1j);m%8V>HasXmK9$S~M03
z^qFG-S!-{H<*&`#pCAyxG<=Q_bg7Rn?=WzJr<O0|4{!6{`;G&R00U2wnaxQ(A(QR=
zHXz?x!)#_rwl!VRxf$g)T?HE6bM#Y1fM_kgVdh$H@OBM1lI9)br<q8q^yhu}GKPL=
z_%xpPFoxu97mu96gU0IBg3>O9PK=+m6T$^k(DvJ-!KX$TtPyc#IW0JWL)nPcqP~MA
zFS!WK?=e;c_tnFvx|IehgZ%sLfdNP$>`&4=`VK0vOj?Vn+SNTNFQd#3c*#f%9d{+c
zaXsk2jq6ib;fjUU&(%Bi#2dMzR#a%=x8qpch#dxsd?diO$TPee*FBOk@*2|CNkn*d
z@hnmwoF7_#LZ*uJH>YDZYGEO_<E#``Hd4BHh{iy;gL|^L%PX6cJGG>L-HBNG5+>VO
z#{7Xuk-lO6mNbJ5zIAHa5iAKM60}U>Cz&R?*IkW1oo6rClug}R75lGeWD;Eaion7^
z?oV^}2D7?lUv$UeRd#2pqjKdJ$&y|b<=Uw=POm||8em+sNp&nnK}nmAkAIEGhGLX?
zTJ}-!XivU=VkL1zNd$PleiBe7>q!Z^w!TO-b<MK^ELP*MfFdUp*gKyY{d7fH0=`6c
zF+;B`uWB-FHaIu<@^ujkH3s0U?;*vLawUP`2-W44gLk-gw6B9TSg{IR_4M;N2GKBz
zJnj?|)Yp7#jv(e3kI9ML8;5)@lM_QX4)3q`g4Pd%-{h$-7&H+{9FCerS`DCu%gXcg
zopgp+ie%PQD$stZptt1srD4tU?UAHP_3QYlz74iLkJbm18e!izr@(?y!`Hu1cl|lE
z-f%Z*gciF@y`)6pxB;MMU+}UFvCJEH;Zr!x?ZypJd~cA(u8uM<Ie?Y)DIR==c;)O@
z9({Q_zWLm~D_`rK*vE6CVu=RXV<?zt2Ktp_lJeZzqD~p_F=ZGh3W_|E8XbK2BAW@8
zo-dObss4Uj3nhK#FtV*-*^XcI|K;tY!ec9i<ox_+?r?)my%k0mQl~I0e&>#5|Ah<p
zV2|Zxo~?e<N@PT<9#3ZsDio5C3ZgV)!(%IJzRS-l_U7{|h-axLKY#Pc_GB<yQhTMl
zbO(Q`59kl(1rKvFBHGa(o?FM1O}Nyn6>&TDDN8ztB07HNifJ}&y_u22u%?!L-J?2u
zcUIQR{qR1@k#;1&MW5vL@_hP8b8>l1iBJCBf$iD!L#Nae{j&Y0@K=S?rr$aVq-SPJ
zSD-Wr@@Mt1gD|4#>U{Y9L(T%tybTSSk-$7{)M!F^^FV|tfb2L^Y7KIs;<SGqys<$c
zLT+Rko`GIz$UPj$uX*Y6u|f84x+U4`gaeZm+CyufgBWwgSV`iSd1oU$3fl>+3q5<!
z=blgGZ@B0OA^V%hzIfTT)_d9Ij2~1^^E*)*m7x|tlx=+RdcEr9m6ev*c)6?f4m9f{
zX+K27<<D}j|0t<koxky+6KV?}-YZtm`qt6jheLnu*(a(${v&q!f7!E(nT^dHDgSxE
z<kKWCQaer<toBmw;~uh5PjKQV>uUfVhWK)V_yTK(4$WB8m;#Q2(Pe6^>`e{!B;{;L
zF#lrtC?_rYXbIF_ZjBRMFYSo$ntiVPiKo%Kw>%)^>Z-%J#b&UPB@hrwbU6>t(F-&+
zd(gR>7`xkf<cM|uP_|rFdil}!v&;}8xru$-v3tFXoy=R6+b#4$DAq&<kOO2ua5cCg
zN6PN3m@C610BXQ?(_zuv-lon4wAwA~@_C%%r(*Z4AZP_=%&_=_2!N?L1=BFb{oE(e
z@S|RX?A9}YU3<{4j^<aXPG^v%T^zn7G&W*)QRc4fZCC9Hpj$uBX8X<}Qij;=oQSXl
zDj~}jFDPlrLdit%PwSpsd~dsE5j`*xKxDVt&ocim`DKA8)!xDLMS*_W&OmwYr@ga{
zUo75$vp(=4$}s7pZ@XchdkUc@<IB0`It?_lmBZQ=)3CGz9Wi|}$Msnoed4%dT^A0b
zLGfm12Y>yAuJkDi%rxFqOSN~?JL~<mM!NNuth)WzHS#tX<S0JYar`rllJ>`^VDALG
z1&bKxl@4e^t03iY93vnrA@jadbT_Pl{#<;y^7xUVew{*7y`N;mI~?6fE;AlNBuXL7
zymtgfxzFxSPv6-1F6cy)iR<>HN>xW~;s#XQ(wkt=M7P|w2tTd#80?%NeaYvwtlTKe
zxXfbIeBr^t!f%SBdePeOO{`6(n;lTqPfr)z7jjLee_lgjgM)yZoDpeI$f|G)Yll!4
zf*1jSLJLg5%d=OQB(lbhmMAXbMotVCPhN=qKka>aIF)_dZiy_({A7rXr7~7B%M_^;
zl@gi9%tOYWSV)G9Nr*C}2o=k;tYsdevJ5F1)*>@yj51H(b!&Lv{eIQc-hb?4AN%<F
zYaK^x-R}GTP1ki^=Xsrb(^#BUbeBSBcMth#&crr>zS%yNRt};X`{m+Vf1RzN7oY>J
zVh|?UmV1sTvbE8an{IMREc>H~tt@%;{%{>|93vQUTA#EyS?11i+c~?W+m5Lk@sQ3<
z92+KlO7W+yABJ96wC`(OX_oFC<{>_sM6gg4dT}f-dFiH9zlG>E-9X<*tZ%$e>T<UU
z2vW`s4G7n@v`)Qx-&Q#-+q0-~rwS?ZxZz$VN>H9{0)o~3XYf$FKHRes(~^FjJv^l}
z$~6BDL2DWeo*t8d%Jg%bLI{<0s^KorUO(qZEl2CIU@@+YxB?%f^mVi{@UBRJ>#EGE
zX2_GKy&pR)@7br)b%XJ^GAq@^8UM6nL^qGbMB(y<{-}1??vZj$Lw9wNZto+x4+_?V
z4C<~Ds1I6Mue(51GqAQ4tc{P{-9bXXInnvXuvmTQxZc4YH`1D^m(H}Ao!>2v&^xV*
zxhS{KE#&HxW3B6H%`h|hC(r7%Z^V%}N_zK0?^7nHtJUPs>;;r&P1I%nUFXyMrUblT
zSFg4m%jc2p%SR5X>L)tR2j`G<;#N<8seS(NJF@TtV)*wD15uTKF%E^&{|kT^xL;XR
zJ}V4mwF$eorzOQbB&{Y7NuC?0PC+W@tjh+wEs_@-F55wTX+IR`EuwOe#W+5cvO&1;
zZYZlA5{>Noj8vrb6)%gN%Y#$21q-N+9ukMvo_S{GP&TVg0zj#OWyl@^ottLeHt|S}
z#>+%5+Zie<38C$FnQ5_kP|L@=IcOPm)nZmhFIMCsEWfU83HX%yq`yb^yF5AJ4l%}E
zJ-(1Hi~R)#fKU#nVqOM`9#>=ZI!M${v2tK|B7DN3@TL9=>tg-XzAGn}U+(4bHtE)X
z&0x`|%m5fO$qtMG4$Ltni=HO|q8O`TyX=}9m*`x_Hf*Zs!jrPtUmT@jREdQgtrLkS
z9d!54^l7INoA#{Ui5IneF(+IEoo^MHW_f*5Ld6>bf{#h>@DJ6yZh_wC2{>@P-eO4c
z9fikwa?i*5oA>#HS(Jq5Er&P5%oz<7$;V2)Je1A}vbD0+_d^Xe4Nq<A(q4F0d-2P;
zy(Xi~A-tzWiVv}f5fO;)c!}P+n1_+g<e|>4+7$fI^WfyX$EOT7nF4|th(`%x0jaM#
z>#z|F-Kx^>=sc;}Em{FE4=>H5&RSXL<Z}x6kqq`hI%a^<{PC;={Y^yo6GbR%k=r5m
z7ltaEpiVqHtX)ZLvR?`!?#zceUJ14@kUb$3-|);eUe7^{MM)LkGF&_)CySt(VD%Sy
z4IxBM><cWmAKr0M-$dkOsOy<#mP=MVhD{+$gVzgiq8Vmgd<C!W+2!4A(Z<o2#KiJL
z;fP~##4yd#$vG0QD6R9l4Pd-7_xk;h@0N685$mF^H@054gVUFiwF7OuoEf|d#G(3Z
z-?cn>yHy|a*Zi*S)<qh!69qW6?tbIJOiz4%lMF$Bg=K<vUV1e-G(S~%pgg_#kp^x@
z-ae+KmezH8mLWCW;=3YC7)`H!rWSepabm{Bg6J;~j27!JHUzvMHZ^+at@+?4R-raV
z9<lIBpjHiEi<*nMl0L;@Va3bkrh6Morp8S3R{e6P6cU=Ak=p3#%=Xpo>0%1w>z2tp
zD~f;dQLKB_K1@vixdSlG?}Lqu;lT*=`>p1z*!2_E86=j+o<DJ?|F=as5RL%k&s6Sk
zemht)xP8yO?Eb(;U5^rB*0ami@lri=LuZ>wCoCIFH4bMLl+wAwh~Odk;TuQ!?)v-q
z!nBU(scYaPwv)R1BdZE^@D8L0XTg5?Dck_9i>=0R+faW|NfwQtzuqM(9wuw!sGSE*
z3>7<=PlR)!=;)90+)zI3*IO(6P&EwU4>dmHkJodfj@rr2@PJ#!Ij+okyQ!i_l#N3;
zKYD&?#9{vgSU8Fv+0Gn%fZm0hk-H*z)ffwaP!Mazy-K>x9Hxv#GOQn!&M3)ri(8?D
z>964LsUiGIne0d$-DW+*`|ZxR92HfD+a^pOl;gN~xRhTRIy-SrBeo2YUONW*B&+Ka
z-HEU0Uvr~U9Ew#1U2Jck)j${lNV#aZ4O)To)q5dwjH{>F@wC-hyIWn<Z}CXIe9VxW
zRMM9Zt}On;%zwSj4AcoKWyGn`_t7Iscm~1CgK+-YRBMat_T>Z=sI%kn0Vj?gsPPU!
zvG7oXf$?wx0buI+fJ*u*^Ml;|N}oCe2dBCXU(l8q*@p*MiBS14=%1zR-jM_Ouie=k
z>(4kN1}&|)kn4@l%Qf!35(i=K6#XLaN~-)>iOFm^;x+P4)D^3@Z;o955ax0Y%0w;0
zZ_0Tml?KB<F)%yq5nI|cD#@!yZhOK@V`sv+&$*;guPa7|fTy`by2myz$SXts5$TB7
zm5hp`7lH!+2ixv`15^h?P0Rb1HNN2VsiR-urfmatayr%@x$+G@oOn!=E}A;ASEN2h
z0*npHW6Eg{R+OxqejJD?0X#TFXg*-2nmo5aPpV;pt+Abd>A*Sdpgkk)6O{LGY9}a7
z?xMf+TwYuDNyPHkx&SO2gcF_BZB-}AkcsM45`+!vyqmEOX1nL;m;xt!X4<pfvWyJ0
zc>FdZ(nKn&nrROTzE@W1$b|UwQB&EQj&=mj@(We)WGJUpUOvdpT0=Z9Z!AvhtiSE;
zV@*10yN{A&3PQ;-R8cx!wbdM<uY^*s2=z+{N%AyYGLKR{cbIR{bv~EpHMeH47*3)e
z+9#nLX0BHCGc)(iRCPBOGC%%IB9!yB5<)N3N0_klZn?p^{}ag|H*jI=6X$NSk7;ti
z0MfVyHL-{KFX{RdOLvrcaiHwz{hbpXHgBdK(&3lqrOgiQO}ogQbm)P}n%r~m$(`+(
zB!67wl4`d~M)0Q0yDXy&S1V0Tw$+X~ZMSY1iT2E_(5I`qybUE6ec(Was8M@2jY+F+
zQ&pZeBRX9wVF6&$hdK$5B5pfQ5<Fm!e)c8CfYL#SIKBhde9SvCYHOjmWx<Y5b;az=
zpE2iyP}I{qEPMkhyksw3aCak}QmHS%w;HLKwYm_m3F)5G&~Z3De~@{FzBT=|YZ<R_
zaE6KhYYtl@of?&{x>dvNr{@)F7sl3(StS->)V#XM(c_OBt55Q9SgA2kRPTBqXIg68
zt;}t&t-D^cxc^KtM)@R3v-3^+qw>GfEmj?w?0t&SOE)&h@<yw^J6jO?@zTfqk6znr
zZWkPWN7sf|V`7r0oKX8bGE)ZRa*oo&O4n)#)L}&*1V7efj<h=72>von{`6IBxUAXb
zaG#nl3bh0p5(~>f%XbC&Ps{PFqR4a+-Sfcp2W73{Fu7A)khsFr!+jBFzhYp<rt}|x
zR>F$^pb!%r<$N{riKoylfgiv4)rT+z1i`2{bT;H0qXcN?`!+w_Y|=+98G{S0&1#1F
z{V-Jm`<@L;Jbo^}S3vvxo6Eu1xhzbz^<w!<Omt3VjM?wzJ3Q6hO$Y&bQ^2F$KL~yB
zmvPq5g}Ji_QC@3M)WiTiP4+osFh9f~cc<f=(vELnDk-YHz&?q<%Gp;YN{9qG=R~NR
zclk66^|h#R(MV$UE1=%A>p33T5PbaI^Tcpg;N}V=hxd;^^YaI<lmg0WkCednJ>Ojk
zRgC-Tae!>vBTV<Fa|~cXN`Y`m%v~A|2H6Am3EM2c+yM@5^3I1I-9av2xqG)0lpaaI
zX)uR4P=Ao~iMv+%sa##l_*y97tII+gBFmiu2oVAariD#pR+)Wn0CPrs=~*P0U5K4V
z$1wcUd(%Y$eK-HkqZyv>gro!M<VikjS2qYxkw@lq#)_$hFuo9ZT?q_T`=&dc*7U`$
zAaV+DnJeYX#pxKiy}KZb)oi8ta`B7aAYQ0+VAm7|I1=f0aoB$}K5szygpG;|#Rb2k
za{c^8Z_qT)gU}%-B<$B_nj7Ex<pnCYBdn+GH*(Z}>Bjl{3n=b@6g2ClRmuP1{r|uD
z0t%QG0Nu2PY55f4c_Hyl;Z&>WKg_5fTg+W^WJ#J+WRHJcE2reipf8vrD*fdmhf9MA
zC{5M}{SO36qgSS1Ou%BZH*kIsz?=uOJ8f3asF(={WSWT=3TwH~^6CY!Kt=FS(7E;l
zN&B&z{QQG4KSG&3Wms9Cv@s@?4M%%{agYmbn2AWv>vyQ|zjHiY%V{~oD#82KW^bbF
zZ9ouQc%H6+m9=AzXRYAYP;*8u@7A+woomZejz~T>4z-i}uDdyb?0N5U67=PVn(HG%
z$)P_XHA67^8X@^<@Dq8zExdu-l?B#s$AR&*ZTc-Y)e59wHae;$D>DQHaK<2oo?CC-
z#vyVA9u5gI7>g2kgmEx%%mrJvGulj)#-8K3FOKa}fR134*<)MuE%iA=_PI`#&0{mQ
z9uuL6AT9DXL**Zr<JWZ(CXDc7V~euCmoASoAzW@msg%Ews@LX%$R_B5G}TIAwdAK0
zz=w`FYYDvIz)7G{VH1H2hj8>efg~&0@rr5X!^gLeN&TUs4zS%zP5@hQoxC~x_P?n#
z|2;?k&(E*P;bsz2GNiuK61Stc-Iw5n^Z50!VjTJz85hKroUhbQm-MVQAtepi)bJq&
z_}_72e_(-r*)UPZLeylxO)ld)#*K*&o8MgvT}#kvJwe=!`{oLTE_^3FT9laTy}#ZA
zbvTb@n4>2DTfE0cm|7n(Q9klOjGlYWq^ZKWu$IbFw(reBU{<nOgGyFyJ3k~^w$_){
zfN9fdw7zul8t{f`n(ot4vN{>&d;+`g7mJ$mi{|&&ffFj#G_>@*C9-lVfW;7_#^rr@
z#_yVTy>#u-oeT1#mE!v~N4Ibse+#S{OXREpGf6Knxm^_=-8yui?k3lWi`?wqcY=2r
zGa~JRP88i5)mme1`smJSp+Q8o_8|xG^ec%`tCL$SWMGVs$w-NCB?R+varFe@6*9J9
zDj)*4huB%X%f=B?wY4_mq5K*Ce_<Q{Ig<aDg~5MC2f=iXe*82>YhzdO1MMzk#%?%4
zRaZASCxcQDJVdzhm74ra6g_3k<d^Uhz2KKU(g~-W1*ml%B2Y8c<}M(O(@&m}Bw9VB
zce_wKjW}+T2ZHb8gH><_2>DVvZXqhxG|~lnj@Qh(<_v-%RayP)?y|hu2rUZz<snxm
z><QMe;5CCs4S_931m2$n3538Cp++mINP;=H)Lv0Oa45JFS$&k0D$%G_`s+yD77IK$
ziqQ75NjDz-F{w^TZinXvhd97TUVE8fv}T%fwb9lNaJ~Es+Pbw8o#sr-z>p!mQx)yc
zGXj*w9hjpeI9pBWj(oYrlW+~dW_zGmJ++5q6OlL&!G}aE_Qo-+a~UXmNl*mSN8AyV
z|GN1a+)_?V)^M#k=|m&+^6i|98zg+F#&HM%k?W>Ztmi(a`YnAuFj*Y+><eRnH)}bU
zb+Jsq=WmjS{>*z=&Cl!~Q3rN+wN{Vc(bAw&;+=niz%GDvVi1afI{*%~T-GV4`tTr=
zC7Ol`GgsHTKH%>^?mR6$zUmB{Aw@poGwWFa%}>G#h6jdaEu4D&D7D)N{nI~$xWAtQ
zya%Qc#r8DejP%!|<L*r=Zqw!Q%L=wSp+{c^VrZz1Ec*PpR#GOQ(gW*`9Ws;vp1LK3
zr`DefTy=T!bKpp6fm`bazz3no3l?(B_xjTjMM}mQm5-{pIC_ui*+Va*?au6`k~d7D
zG;}Kwd~gv(+70o-`sb}3N}K?49F_%W^(c95A+wR!H0FAc6<rGluQ<6&aGCrkKe_?~
z8iE6C^XWhyL-WX7LzGJ+W=b?sY$JSa2uNd>=+;I>Krf<#mq(xENO+oMIs)U7_8`Z|
zfO7n2hLI{VKb|i7;-M`scsHrsIl+NJNPWt7i=be9k2J074Z%!4<XX+(N##&Y(Kstg
zFC+rgxMSYjQ+8^*C4u~2qud9Xa+#D!^42#eycfY@kt6jCxgWsi&8cWcIyYtw6t@R3
z_2{#-b?@#@Ms3kj^g7vfN9jTet=7f;m+L2>z?a9QA;ps&dOZ4X;7e@YJ4?2C1diTa
z)m6I{?Rg{al*`P&0IbOoR1HQtyGj>_?>#i+yCb|+$lR2Tkvis&m-P)_QLlgH2B$WW
ziR>eEDR;fz?+q;YXng3Kh5*fY>KtOfbDwb?+T-#){FxkkE`_cg4oIzTpNq$ncs+JP
z*Bc$uK@oZa!eMI=4zo}fEpr`CeS%#*z3TMD138D2@x8*!gL@U?@k3?UjPBcdb1+f3
zXFeRokF)nI9D9K-lD~@|Yj1c)V_C5>Bq}>sqFOc%5>>OoAxH%cNN6-JMc9Qwe-1Xp
zBhW00gf5bcOZ1v~6|V=~5K>~Iwz8lLchy9Kh+4u$@M^rCP3S@hB@QP-g2ix%<++8!
z%>ypE`qQ_Lp`mUud3Nu<l78`h1F+KgUD|7_3P!I#iuNjYD;-C~+a!}#_97{^_65YP
z1e-&m^C-GEQ8xd+o7Q?_xyhrPsr|7-4k)auu!Wx$j9~KSjW;@?U52&E#fn07yAt1p
zKg?VnfSucZa0rqnOSPSaKj;ZRv_e1qV0?w~7DdGg-of({U(Yyxc|SPFMTDN?_t|GO
z7-!NDoJ1`15W7J>+O8PW6SOSlBI7ossp@A=`60E3pFX=lp+9dL-cai?7vJoaj+k|s
z4cX7WyGZxM9e4TJHsJW)#B;Zl!*LZf*6Ye4a>n^>2!--ff(gSjq26qF#2?4SS@s@T
z?EL%b2#=1{`5vy=?YWmOGV33>to2xfhTy46jmAaSj&j9L>xZ$5gcJHJ!w)z_<|+nR
zC<={<@4_g5;C<=zE=;^GHQ<Q;1i}SAhBr0A2pRLVP`GaWjI4}ZLB^des>voMQ^5Yz
zd9jv9gZo^2#7UAG{ZTIav~JalfM_h}$kGk$lUoxJIUA_UkCc2!4JFQPyxSUzG!vRP
z6ljBPrlLKD3O|o5|Jw_;v_Z0$*4ifxUn8rAQ{)Ns3p{5^03|w^%Pr+XaGwhh^{RqE
zs0eH3-tBe~4w!Og5uH$840g*ZrLuB=b-)w73xbo`v-e!NlS<$~Z4~akP&X92#!IKh
zdDR@(cbc+4q(gYe&g@9%3X*WIJd0n79+e$ck5k2%_Fog9YUri}T%R8i&c4%aV-OT-
zs9(>U?n3R>rvB-e3`23$W%+xwsBWi{Q%O!YL!}(x7zkC+C-VBbj=2sq&kBf!S2hbb
zt9I{`ecAKi<^rws;H~V2nGssPPe@;;acD2wsP=Yx*Yy6$Ktu}?Bs=J;2yq^}n$~;}
zIXYA2_gL${cK=(DYQ`P^@LgwCC+){ijz$t&ooqpZKq?N>@`vYpi{LzEi|u&7I_VLY
zNW@OS7!@LG7xv~zHr`zRnv(ak+4Nnja3@>>M6ltJdRmQF?~k?05#SXfA%1#+wJPLJ
z<mApMB2s|E?ACD;At_AX)a(GgTL;f4b<;fLEbfYNEZP?~qdZsyfOkL!W~}PyBvM<O
zfa*?O#ro>h@T8|b%(76|jZ+?RvMJ$}@F>$~la)iLx2APj{`54}<@aWxm17NR1#{|V
zZH>d`=}`y=r|B1lgn82su}MUvPtsAX6vrgZ{ZZHrt!5x?85Hsuy0igK7?UyHjUubC
zh%=B-(Z@+mNFu^09@g>s1$4yGzaaQlajYSKI%5E|{tc1CbGz>T%77_g)w46xF%&a6
zwW~|_O0X2=cUOb>)P$1L(>SPd3Zif<h^(t$5|pWhTZKc~<ZsyGI5;nOLf|=$koI+s
zfmP<Z6{-=V9h@dhj)wXhwh_qeX>nYlWD3vYa7>fTNQ+i_wsUJJHm&1i6($2#qieT-
z!yJq>ANrG(!w3;QBh_~trBZ6|0X_;ePNKF@RC{F@9+U0ZwyjXzqvP9ErcV*KXXbPA
zdg%mEczDe9pn^%DwYERH9SfpA66_^CS3WQ2C$BbO+K_lE$gQHSDnL}U^wV{FV3Kci
zF+{D<JNIqdU^xT{4k{n!TfNEN2d{(s>8uGpht=t{7g(nR?5Kt?P+CZqIkImaT+jM!
zSkrZ=g{YC>9#O*cTq*?{S|aU<Z_mpbP&oy}mO><DZQSFmH$3~Drt=Rkz*wINh=Teo
zk&NHPhwFQMgHeLENNnpPN&5)_iRj{TbkYT>yY;^$L4j;TV62bl{g436azbjbJuYU*
z@h{~{#JmM&E!}bz9ol&8X?_G$_R<dTLXA5yt$>J`2!x2Gwu1^wEs@(LD*%4VLBf|9
z(*w`#*&T2X80Z?bJR!5F8GSDc8#88D5;Bl9*dKv1=NN9r3EXvBKQ3pH!q#&pO0*_y
zlU5}Gdy_nQbz*ORDo$*-<%(URpF}$OVk^n0?{Fqr#G542&X=Kf5vt++m4OUS+|$&V
z{^HxkO*MkqxN2MwP`|mLCRQ_3GEf>W`Mx6%In=gXd=V;Ij_c$68|pcB>H{eX^iFG#
zc+=TN>j*w-n$5EHHyl_m^2mAM6XC;AnNt|NQ=;*TU4}F^p7-&6PHy#sa-AGRzr7NF
z)o$5;n?beUH0_XIO@qy;+cS#NH|Tj@C52sLCLRu=e7aDPQ(Vzqv5H;SEHCQvGV)lR
zTb-2GpLZ+E9$Oyo9=W2hnT#Ao&#maY>1Q8&Dtvu=dK|}THolX><Ol6JGNGo;G(-@s
z5CRCDiz!b%Ci@0vCeHS4`f2u3F~7WBvhH_6MT#@f;N1Ph0+LErrU-S`eI_`NJLWvP
z(hX!gqbcj1uL==g?A^FIfYt6mxSjo;q~AOOX6Kg}!_+q&tin=IGZ%<@%&d;(REPlP
zr`?_IvAQ%n1l#j2rr~-Wrs0zcTHYeb6TXIZS3jo3#opi$i-^{g$45M*+BW?6QWAIH
z=({I`kR5{cc@~7aFNky^W8`ACzAovcXVz8urpyqJ&yA9CQtJx?k$b%5Ve7AJXa~`*
z)|P1XSfNb6d^WJHo^Fzwu~QD|jF8g%V)@j7^Ypu7SVbw1xL|5c8FO6KY&(zra0K$i
zlS}gupG98vuJ)n2l77zQx;YXc;oRPJMRS{D1<N7)uo1_*A3aF$B)tGjrgKHgNRw64
z;}?a=p^Z|EVT9sxNVW-}PSJ+FH|O^J-aGO5;~AnR%_GS^<xOfOeI}$2qw`Tcnr$bQ
zgEKihX<u0ik%}C1Dxj(BQ{8;xQP%k*o10d&CG3gk8}lQy`Y^gr1~{Fr|8!u#jk%xU
zHq^rGl}oN68{Z0J@q~hm?FZ|%mBk?dTG+_T8S{uZXR}wcb`P&Xb`%4X+Vr(z_8#9D
zrwe3uRHyfBe@BbjbAJ|i2QlXls&@$XmD;J2Xt$xk+}17hE%{<zjP>N{U;)1e<PkFw
z=lwyT)NWIVfhTu-9VAjAH`}T;1Sj4z?-H0D1URsC@(dg~kYCT#u{*ZjGJSs)e$iHW
zae1|KtGFLNY#sz{`=(C~VLNu8i8D4P^9^Md&LX{?)Rz`D$%Zy{7zk$f&L(;8sf@jK
zDXXhi;X#&IlhngUIaHr_xp$ZhK~Avq1_M*u$=OmGc8S(oCe`;k4@qq<4CCu7m*RgN
zf2O^LPC?F9wFs+Jj_833Pnu-SBf2|__oY|GrQ)a4={h48VQ2-D(M#4eFNBTCy-qP@
zO2ieqK@!ovLBbt^Myvb9E;n&>#W|UgC1(bDLSoeB7VY-#yl<FpmOx#B_alK`*xU60
zSnIjkJ$#9&i9l>70;?*L0Q{0O&@JYZOoV9pLh78@n0I4>_^ek+V>S^Pez$1jQ8)pN
z+D|E?sgx{Fn;Te`E+$Q+uJ=S0y1m*Hx7{Hg*qMtk7hg~aQA#c-|9oM89e+hp3i6e|
zrG;fwH@;Ozz1ufdK)=)G@P$#NVbV^N`WfNPC3fma-=y}+3uG3JQ8r#w>xGea?aOel
z?OIg2Y5lrch2m<;*FgDhj33V<Bxuck2COqDrbB+A*MDjKJUgK*^HSVe8XeWXJ*PTo
z=_9PoU#{=wC~}`5?vy87``q+}uP^PqqB^&L5{CSZ2=ILk%7q`~oIkF9p*T$8-8ADN
z^-58bGGxOHtYzfrdKV?&)7dZiZk92S@HyW9nO;#_uZh5>;cM4jq#3QQ!B$W|-kDF?
zID`jisNm_P!mO=KOqKP_S!_p>8cW2o$!VBA+?TsL)7vRhj!0wEsNU>3#oB(p2@&m$
zEt{3VA={3z8g>~;F{@jLGIZp}lUaQ705bZ!AczS9SJ{xH<F`do8K-x1!2PO&siD^B
z!q`o!wG)^_O4V~M<SUH3Po-6T*>9%4NvsOja2o4U4n*V8*}442c@g4qJ%bhbbIw;j
zzDX;*ZBCZFnE3e6<u;YJvvr=J@r$VqWs&Ue8M<5@)5$OMpuM5{zO+l$s)(ce?cc8D
z_sCjyw{Z>oww8~h1Rvx9_?2(>z7(&Wq@I>{!*{qTQBT1GEufTq-prOOv!IQ3bNWub
ztUzQ0A=$U2i5DM9XeM7Y^jV$VX?gOD$UJ|82jCttz!^iIq9_`UuNV;HDLHRX=#W^=
zF*Ur7&OL4{<X77r@HwH}oFjJy3OY9Ina5aPz#OQEc!+oONRAJc+I@;S=*INS<ch_;
zLGsxI0b44|L})HBpnSzz*4qcfj5cW{c25!%FAa2_-`i!t6?i0hB1z1Vtp<9DU08l&
z$8_(cS7vFu;ogaPi`6}^+vE{_=%t$&>9`5!9>8^!DjlI8!T!fdf7`g=(dl$XK-;_P
zQnK6(I&M90Vmqm?^y%2}%a!2;v8tCMM5_vqjvlo(cPyhu;j0_VhbVPPou*P_g0{2G
z9J)PJvFXYd*Hh!UD1UY&e%m=WldX$v-dy3+?)6dl`_~lJjXnBG+k~0W6RF`;M*UlV
zceIkDRw&>C6~4m^-@oREM!7OHjirX8tv*nEzvxegd|e!Vx=sw+N%!q*#^mH!dy+QR
zuJc^c2}7)%?o}|XBLKLitA+-EDi#<<r!!uwvNM%G|9Y1Ha$8_kR;+YH$we}Lj|o=9
z$e<*GV0wD>juorgvInlt3?8z6^JuBav^L}lB*mH?xgx~7pY}L>|H;qeiAs7hGW5~j
zt9QOXQg1RtN4fVk-698r&Rnw>INJLV5Bl4n2Sl(Mt}Vx&NE=c8{7T<0@EY&yYc%p`
z=k8xbFF%Ir$A=nf)B=rT2F3R`{@a&~{OW)H#IJ+HA8+;lrx&I8cyk4u?3pWYf+l*L
z^Z#ES@^|wrrI{?6|Fi%UtUnfnCT?AUjw!EZU=#EpDupxreqnrH55AlH%BjZzA+vyk
zA<;emt#1#Y+xn1{-_wxIIQZw2*28ZLtyGE^6w|+6r6FF75G$-FGm%pFhr_AptJp&U
zX}_U_<NGdaNm-f&VSOCNuDiJKEv?Ph1Ln6&^V^&HVqgq(sv!t<#tQDiOOcs~|MNSj
zM8Rqe|1`)h^N-8)`=7h0SeaAd^6>JXj@p50*pm3<76UbYxgY+=(5>#E|1>t$^?~hA
z?<oueDHKYx`~lN{T<KT$sOSSh*^;c}-^c#+p7?~gdnK!*9Y6f)lK+}uU+?<05gZ?D
zP2ov@Ix+ueDE{*m{$eZmpP~4(Irx8NReYk{d2~H0Sw*sR6Z|`&s-cptY<&H{0M!@1
AdjJ3c

literal 0
HcmV?d00001

diff --git a/_images/IT_sample7.png b/_images/IT_sample7.png
new file mode 100644
index 0000000000000000000000000000000000000000..6cafa761ddb478f1e70a584c3fb9f1c3c01f8bb5
GIT binary patch
literal 578291
zcmeFYbzBth*FTJ+fJ%rU4XdJrG)TvSfJh5ScQ;EnyCNywNViCLcQ*@2NOyO$unW)l
z`F`(sJ->hM+vm^cx38I<YiH)nHCLSXIp=+y{q{*t0_QQ=V>C209LbMjifCw$kZ5Qa
zwpbXznO9;&9%yKf3(Q4DKS_#;zW8JhF)_C?Mnn7fEjAwWlae~|mqQPh6yLa~K`|pq
zFH+DH-<xGfW0Jpo^i(R)JKaEnNY$1pF!0N(t^B~RmiLT`0uA&PN~dtj1L<}R+Wh<P
zzPzlxf#2PorX_G}dTge69MriScsWp^5gHe(CEsH}kD%|56ua`^SXm(;xw`iSjWhlJ
zN3@=G<6!mW<u<ff`WqKF7^d**#g7TsVi@dBSKN=KcjF$KUsiG_e8V`?lVXBcpzJj!
zni2O7w^lMW)uwZd`Ljyln8Q$q{?~Y+4r31tTg7BPKOi?ILwmp;&mej4%b+kSL`&y!
zV!)nXd1xRDVFu%eX=)CR;P4atn4uApaS19oT;`ftzJJl4rq98?j@~|1YryxZr|<L%
z_QBYl1OIJ}k;%v04OWrZ<hPFkg^S+Sn_iApd|hLddW@4~@Dlb4`c*0C-l)>xfC;S(
zTiPzcYfutCnLt<HdvKV5JT9$8(8KBGNF~O;pa^&@qkhGY?+gj4s)<xgm${F5PLOVS
z3S~;k3?YwFR`$+%)3-^g=U5^?^zIS|BAaqI&!bBOIc^%xAM>7+@#ZPqJE?NW1um)m
z;;mR+t2%xt)clKW^u*OJtj3es?nAABPQ)_HrRnB}{+^9=OnY~$T2H)Rwcg80*uP&q
zF-m`X5Bq6|5CScNS}lsB`Kfc9Yo6HmsKgGv3~rg$r9jt74pL_>=L@uu#i_p4qc|Zu
zyqDJ#=8GE7&<--o%F2{}HK)VTD5fyKyiU4ExIjw}OuZS<bXAl*$cC&i>yJXdJWIbP
zoBjmbo3!-4bL*qpdn`*&?u(EF-app6x8(gf;+sk{2_vSZ@Jl+3q&BWk_ww7QY|$t%
zBt$rP9uE6RFOk*Wv+)TK#<*^iX2o9cQNa3o^z>oo`@{$;BW(5e&QA$WzIbNn(cKRU
zc$Pu<@quK<`iRC)lDFP()5(6~UO(h}loC)ZJU`;~lbFnRK{!qS0Zqp{HB2sV(>9s!
z4>8*mjy!oD>O9zNdwBGrmZI1XuTA#I?TXBkJg$vp8UMkH$5@o35`|A`+CQ{MJa|nS
z`GO(xRW^2>I9Y^BHhSH+-iViL1Q-Ds1|%d5A>=e@1fk3NMEWWQTzYR=F<E85I}ggr
zu$YjX;z+ky7&4U=tKo9tCVkZoVd{+2ztor4V=;gj-jp5M@la5j203&;Tw1X+uQjg4
zsZFW<FvqLKU5v;5wP%&{IL_6@lO#15-bTFQvV^w`U-39XU4=h*_?4<HY)Rtdi$E-d
zFMGR4JL!8o{V%LipJ~Vlw7=<#3E9!x5!kVm#k!{amZx}=7a;?T3!NgFIZj^kxN!YP
zcQN;Z^mM)SrA$@{k)Nd+8CuJM?Z$pOe)2}j<qs2NQzhb}xT7wk;gN^Yi;8rF1VO}a
zyrV>?Kjvj=r3++ue$IZY!C+5_2r3|Cc%DS)L8wcQ9~%;@CiNudB&I(mj9y!nOp)eo
z5?w`XT|YxCL#+JUMP(|5jQrpEzIjIZR-br(+KaE(z15izn^27{(MYeCIR9z^dY(b1
zIQ^N*iOGrO{8d$&e^$a6Pvtuewtd#3V6MU+`5Wm(S=C}iiu;v`Az>r3TNLgXUIZvQ
z6fUZwK8#OVJH1(_Nu)^v<<pIg!-C`eA@W03w{<8y5nEPw$Z?1%aVRmoidBP1!=aMM
za(}G;C+9mKt`2Jq$cNF8EDkEOjv1mE{F#BuOEVXX@U5+#tKC2+=Pj)f(y^Rf_w66s
zaXYD_l;c@DZd;GerOt<Un)H6MsfA#F;LS>5D=aAjrL#<wj2@1GI9WLZxq7+H-nm+R
zWH%zCA&(=gCdDP+CiNp>;?C!G*OPf?LY7Mg;XYu!)ibboW?4R*J-t?!=TG1B4qmx%
z7D?JQ_{%C%y=a=rA({7=z`SQJ0_Vx^_R&`pnPV&UZ>+YB=qu;Urv(Z4e;NNUGua<(
zP%&bNQ!Z9+QZC_~=ljjKSU-I|!I-7~sG@ps=ll6k_b5$=Ny(6?rt+-ttmmxCEVna}
zvp|DXgXRUH2lz($g6X1!l9BQOZUe;uuOg)=#T0KfFVxB2If{pq2ZxWGmzGbJchPw4
z`|=>p-~h84F(h7@Iaf=lj<0T`mdXh=Fx1a9zn?yPI{l;KVa4U}fdjYgX8ly{fB*^A
zH>w6fqtxqEEw~0eSI1w+v&2gWozX#sZ1?GIY)xoQ=Z)_rW;KJy!<%HhmDNL~wO{K6
zJLLPC%Qv0yR+g@k1<3EkQCPxV9h#N!r`EVH4HT5P;YvNExQ@75_`j*daD%Y(vE0e5
z@EIT3;CPGP7p?Xec+83W+>a~3n^o@jWNFXxkLBQYSc*>)fl#jfa0S+E;5t5K<an=N
zBqsHugqwtQkUNPr5mTs5utHD*u@dp=%Sqbm*D6HZ3^w9(Rjx&@`kvDFg<jI=>$dl5
zE!NF$T)-WYRe27FJ%1sjH+0q0t6mtkZTm?*se2Os#(=<(0FMT`>~CjTM=hD0ou92w
zl~j{*8FkpjvX1w7?%|uS-bR;eQEQv@2lSds7)pXl1BwB@R{DNNP8z}*PBQ^#rDv#Z
zL(($R&;;MX&cWzc$_n*A1*N$yB~@yED{9fjG1+TZT91~#{#edJ-k+K_A0;YEo+X!;
zaE56PDVDG?KIrLU3l3*eP=BQ{RyS_2eVv2u26Y3obTF<mS!x!ycn+N`JTH9C$&e8%
z$uO%zTKraRqGbkl?#-Faxors@sOe+rgVk_3TU&a|we5Uu1|eZTSw1ty5H4tMRu^AO
z42HLZ8YX8;7Rzom@msbm%4F@-kK=I~aa3XDs61qDS!f@50~fFLOciO3dL?3*2+o8l
z^;zmf>blzB%KEKw3{h<JY_;udEPFut&VKg(*nauR(n%cU2Y#-@E5%3+yN0ob@-nFg
z3$KZQ6l98EL!&e2=iSf5s;~v7Ij@rLmLGYO+0|BfHu49lrXGeD&d5PzolV+g^NN@5
zW$?uSt{8<`i9>M#|4%PlB|$BgB9Oql!;#Pqp#V*?LNW`o?a(fXR{ZTq;+EI9_M2uS
zqt%9khB!T;;i@U=sdN@!6XUpWxO2G%G*&e0<uOvNES)R`j|<LAZsfPq@&$^V-G7me
zC4A*DG(8;-OD0Go=S#D1F0t01?Ra8GY*Pg{KddKlJU)5TB2CPvMrq2@pey9mb>81s
zFjz1&m0Zl}#OIuPHFR7<pe?5@JC#yWGiO^jIwRO{BNJ{4M<fYxv&^r|)jRW_IbWu1
z@n0ScH!w8xv|J*g-KZ%%dr*jw77FX;=T_OtYB}wUyakN=u1R6zHUIqXuUW*qlkE;6
zoy7rg`Q6o+?or4|p_MQs{h)6CRLKYG!>u#uHF~Frw0Hb@Bp(Yd&`D_GyN#dHP3f@f
zK$W~H=~=LHEkHQm@dvnAQ!i7?FPJuiP@m2ePFKC<2jwZ#48eFI%(q|f6mKVxmb*3S
zknnEyp6*VqM=qJqg+KX4N8fX;0HH|<ppBE8-Lblqr=>nMI|1o^ZH2A$rFy(iMl%y0
za@Fh<!feI=EUS`Lr$7n$mf47A1!0-WTX4yg2N6?T_mt0?CLcgTsl8#nMtFDH^LLec
z0tM(-xhP_kD#n-k$@{8+_ZnoZCTSuoi}n^MW1&5`M~3zgDBS~gp?l>2S^jX34h{XE
z^Y_uv0?g4K{QDj`;P~eg0qlQd{&hr;@<+o2emw(r*Yx}UarYx+I{JT<F>HZrXd+6Y
zl9Iqt$<W@|*v7#O;<z*E#Ri;sWcyLw0S%4l&7b|Aq~hx%;Q7DJKdU*a$;$8<Ladqf
zjUWcb%&yk|JO@p{l@BOd8$0U1aJ9Cwao}?meD%*Ad_einX_i+n{<+1`Qt*|U?57u^
z5PRbnT+FP@tgnO~zj*ONz~0D&Pf_f{zr}%Hg0IXR9c}qoSX^9Om|ZxSA@-&$Y`na@
zEUfG-?Ceay9ZU{xHjetPOg0YG|9Z%OJV(sf!O-5^*3led^Wx8Q^$j3Sj)JdV{i*0b
zKmU49V^{P4s>#OT-^~IV$nxh33mY>l%YQx_5Eb}ymhY3ftFe{3n7K93XFwf7Y+Ubn
z1pX2DKQ8@OmH$mt?Y~6XIXSuhyXb$r^#2xBb}+UVg;)cXItu;Qfc;zezhC^fpa9FC
zw*R-j_*X~&a~9}np~nI&|2b+xkDo<3jR6BsVJ;^B88`yd?9T_C2l%4<*AXZ`2xQm1
z5L7@z6GoF16Z!0VZ!ZON?lmW($HDf@`D~a{);;!BQ_vWGM3`fYiKqghxzsV9)N}Et
zgil38#JZw!A7hP0+Ec6x^CU{eqMM(>xvv?gr>DJKoXZazJPuhd-5@5WQG?NtegVl;
z$qz;gyuA2N@BP*0ol3*{9s?$r0)8g<Kuq{=w)^);k$EoyzhM2%CJcHS(WVPBiV`FI
zy9WHJ9Q$u-2HyAoR{npx{eQ1F|G%AZ-Xd&Q0I%bs6=$5ubpO?hzu7v1S_AQ^=f;)f
zQ(h1G7#!6c)bG~E>$uZ@YYh_n8}tD%>)Nl?s$yuvreVMO^$KoK+oI>`&}J&^7V-K%
z!?V9a#e4W$#P=PBc}>e7o?>(@LC&^D*{^ov$KLRY{}n_=;kPb9Ryw~)u8<S@w3He4
z)YmLR`TxH)8R)`&E}*IZG+FnV<nwQT1;Aghunq6CUy*kwlw=406^5VT0T_HPuOq4W
zSAFfRFC&Zr>)<(!b^JT|+m|nVT5kIoB;WqcgnY#s@~C$SLSot+#r#+DBj5){M^j+8
zH<EuhH~iUv?!#FfR2};}`Rjl$*tgfTC9m`TYW!PSX@Kt2Pew@9|CPM(-5&#fw-)>Q
z`Cl1ul5U{;xI;K}C5it^ehYM8)Z{4MSe;&q`a=lax^AjRa4ROrb7=u~hk7v{qa?Te
z<GVr7!ZTa;s1Ht-?EQnxG~(;o!M4w<{&($l{MH&lf^0hJX9#Oyx5@A$vZs~^V<ZrR
zdib__95!ushf(ep*KLpmEHCDQQRaVut$(ZJ2_mb0;^EI|{g}73(fU8$O8&24v*YWZ
zIw@V|><+#ghCS-mt=`eZw^li?m{=MH-+9&yFgFmo?S8Y2!(F2cD^|Xi=i7em@>@S(
zqO`PraXnu5pnBRQX@)Z;`(=84tpDm7)xesX90oO9qsqbO*i%sSe+Rh!olA$)rfE(t
z3vRhLztV2khIuzBE;o`OL>rxR4)A?kp)0)_pnn>aRn)tNIH$DyZvC%c6-;V-9(UoQ
zcWJxKJN@h;J=?^Kxpmu2nQA&}f8u0cRI}iI@FU44zu^aH;?YkVWm6ssgYR?weR^xt
z!uxZ2y(3p4g~w6FfQwftq9;ZUT@(MGzO{SwpWf@DlM_N1T~3rhIh#N8-=Pj+hksym
z=1@nVuY6G3W54PkIt`gUYukY{?=+=76>Q)v`@?%IoNZHvV`%M&xFJ?Wqi@~;E_j@y
zZ%5_ODqNW)=Twm7NSjuB)0B+;K@NL-7{?zQVcNL6;CyJ*)qUdDat)ZrPx5**_#6%N
z^PkU&&!zvT?|kjs8bOXkt+T-FEQ{+*UKYpeLZ^rLb_@M3-Pc!wS(-^yKt^4Kp<FX&
zze$l*(YMBN>fM4Tz0{mpVC?yeIR|-QbT-eWxNw6gqG@F9OPa4O_SFm3c<tAEXAZ|{
z^ELCn$)z5jd+?HmXqo4xkYn?%m^Q{XZ7evl^u#*?5I92Po$z6B{f8Y#<+kR+|1=~>
zuff8uYg_ce6WWT?w5fal_OSWJMinRXF-T)&h5xW|wR+B8+69;tzMndVmP+Q&Dx407
z`49J4mpYm1)}n0oewW!aBQ~D21Ps8eberyOZ>FMx@n196uYazo7k54hs!1CFri56-
zpDFPSEQ$eR!$}Di<^@a_LyvUx{}4Leg8{7M8w|f(@PK=*Gw9morZ{Ub`7awCsdAx8
znl2I>fY+Y;sgN~yGQe_J-;`p_q~yLOh5_3*{xDrWp;LG=5du6RG19I$ejpPlY$hFG
z%Y3oUN33cKP%QQUcC<F6Hvr4gr_~nXQQuZ6mE0fx_KdJP{*er`v4_2+1WWw$KAJyv
zhRqvK=;{SB*&v91Wx&V1xcW|Xt|<ZplFH)vw;3dXJ~(5hg1Rl>(Z@B<;nlq+#Y_@j
zBtae<l*LO^J&T@DlH)c0-Uz)zUHX9EHp&`RSX&f4vPK&Gj)c74S|SMZ=+vi#s!$1B
zZ)MFN_keV1`~Asnwu+mtcV;--pVG2F=Z#MCxLSs<(Fh`|q_{7?WleAd&LZt}*03P(
z<vAl(1miqW_54|q6%n)0GgG03D`=z)kw3NFW=hUWSju7?(aL^y2w|7{g$4^b(kYm-
zu|c`!1j&R$VQ0r%jRfV}D}7Fl>{_!>CjR)uy!@hV?D-TX(h;crilbQ^Ugr2hv*Elz
z*)xtynp+}y0+q>Cr0lt|xW*ubJ~as1+^=?+pNk{sP2?p-a#CmQ?mu<)oQpn)!~{7>
zGoRwuTqy=u6Bw1BQ2Vs>z2%SnF7%4kW{xLdp=^+Spf)E?BforYivpRDWRaiu_y)_l
zMN`5N!x5XJzW>uUhVid>KCp%$NHA2$yDp=j!2=4WxT(@Qzdzf?I6E(MxE2*=!~T}s
zWlr@?;dx&XSd**S=313}M7BQE?lMSjlrEtdo5q`=xc1eS<E%}Cbxx0px#jo;EKo=p
z>PeC{{4cekp}pUf&9d1e5ONB5Yu345ZsQHLE)_+@*DRuMe9qPvnBt?Xwy@4Z+ZIz?
z7PkK=z<b40IJjlju7`4-11_-2nt%NVXaQ%|jT<g->N~N9%e(fp=4vGv(=g2Q!Ml%m
z0!LqN`sAU{RMI_`{fOrnD+Zs@WU7^g*R~voLTV2^H~DR%Wx4E1T5jua_DsRM%z(Ce
zezqy=B6@PU*A2bd-&}CtWpWu6C!Wr!U&ZDaT*Bp9Lc6?}g)}^2u6@xdNk*imch6<X
z8?%1DdX|~wsnFSosMGRZPp;58ea%*8a6JsU3OX<a-y~?;@Enx&P}u-x!K!xn;h20P
zOAFH81$92vV4CJN*>v1Z{&GG$ptPW*x!zC_LzM=V?Q<1GLEx2SQ*S<6q-S<|)Pcue
zeZLvde$m_X#-hqGse>#K>L;~soSMNmf8?GDDLrAiPjHjf(G-sJe6i-Z5(PnFOgNym
z{Novq2xxq-?{*4G>UlT5Q^y$Z^c^%D9tfR8hSxSm&4*U@bY)pl26sepRmbEoBhPZd
zPFCo$N$=g`bllBr&L=e+I49Mz#yagAFkpGS{~$16nk<v&W_YdPM4DFkGueO<m<k1f
zg=}98F%*`5oz)MZB%5)pek4Dk5@)_?wpqkD`HU5dwg~&S950bPaZ{X}ccK@WlX{K1
z^*C&x|G~(0azHPFDe^srQBH6<;Gh<gy4DQci=WSspb)?syPTv}Bt;${e53r(L2}bM
z9I~)aM$nHkz&2vvrMsVpm2BJoC`0#r9|D`DwClu=%|Zc2Z5?{#1A3tSG-bP=p~SIb
zn9p+EbwErQY%`_p;&gXC)zXehwq+7P<!+Wh=RXL^MTj&I!CsHb-TxCVMQmPpcPE&p
zl3s7``%;EMuj{nV-$R`=s+D>dBvED;a~U!c90}Km>ukwr3IY|+Y<{~HG47#vziI)Y
z=^^e#PkqKihqN0lAFB9QO(}7Yx+y_kXQM}9nr`c{^YppNQST|bMty}@wg_d#eyxkV
z7H()Ip*pWtQmrc5-mC_BQE1}{viV6zg6P*iQ2?k2dbahs1i^}tV$R5>O}h2QD`9@0
z<jMQ+)?>M~!$iwsyI*qr*7ag|_A>6oc<e{d!gu{{2R`+Uz_0DUF~+Il(@C_pIQoyH
zfIc$h^=${2AjgePyA=~)Y$njm%C-`2*FI#rHrK3G1?L>K4MZ&-brRn)=$TtV1fsoN
zPeLI#!Ej`!CDvx>OgZ4AaBCVxinBuqKVKF~voyW)72Nrm_4V!*uPc$=^rDiyU~Hri
z1Um{bY~y34b*edUvQV0{Ol9Y0{NkpaxH;mbDn2z>oyh3K=Su?jRzVK+;S@PK1`Kg`
zdzyA<l$+{qv1C|&Gr`5)vTief+PJL05gH<u3j-o-T#D0^qZ^Jigp5w|(JVQO3E<G$
zh7vQzYdcR5l3Pp^yS<_2-_DL^2jeReOK|spoJW6qdvzQR)9RTZhbV`XIUPCg5|w3}
zaFaL<C?1ANRfK!(8G19FN7a`CeyPa<U1hPHE#!2F8;<lXe2CLCZXQm3m>s6>I>P{I
zjr(Jft18{7HwN)8re+?@M0DWt7iKW4QS4L9GSTMora-T^WoI2vPX9Qi+ZWv&PS$Mu
z0cqbKS=WG!e!gKd)>N-fDS5;eNE<EW%6Zd+DNl5(?Xnp=J?sJHn>7#Lw}A0OO=f56
zv=?t$=PPm-T}xTROziMB&f^zgulV=2SWGTuYI97DDHy5p)G6$c`WZTIK#<aRvA?$%
zYsqk(ltwa)48UK805-e7QQuLt5h}iwm?n5R$8oc8**vu*K`f?M`}7E%J!8L2NM^!u
z5db<;6ycr7rx8-5gp6_)Bjp0@d!*OCrS1SSKyCW$mpo|3Z!8j9;fve}I*O1o6lz0y
zd9@mD%D-Q!G%`B6j>L2~{))%ng3>B>e_6ptKTE%MPzinKaq%f9D?tF|nQ~kx4?T(2
z_xxhlwH~L*JnOm|TJc;}i!B7iYH|g*$8Y2fV%hCZ`e*}O27sR_h<UN7fgf1AkIufz
zDX>IBaUS05P9hCB7|$`Yysq271R0o`Cfj|;y&`K@9t!uzauRtUmj2B9t-EXvru+wQ
z&baSf93GOow>OJ-sfch9pWET9h@r9;#72@ifnaahwzfcTtF%DzKek2`ygTrwTunzj
zo$y}aV8D2lrPbGn#&<m-qy#jM2Gr~rt*mHmtiw9O$6muIDP7WTqq2s*@&OAB5fXI7
z^H=N~>}Lhy##s{K#exIax3^b4V3Pn4)Rp!90N6+5{lRgM7*V$n4<B+C(&EL${3M_Z
zcgl?m<~$#EbI<_6x=UI_{#d4huJo}J9^Bou+%cS~8Xfh#1~NGWgEtL%d^;j+x=Bq7
z=R4q=;5&qEiPbfvcG;iI(1WPwl~0SRj@ujT+C;Ds-^Fjk4hIC}u<5cm`rEzP#`c){
zka4QirR;c|+mAX@&OXXDbQ~mk8VPMY<2$TQ_dA#xOnY#+ey6tyGaJto{dCmmVQkcO
zV>vCr8DHC^Sp6`kZknPzW%B7#5<+W3H@nnB<-vGb@N;s^%&jx>8ho=?VVY!Jt!@?^
zJoB+-%tAX#sVdfZwvO*5#<m25Ps`%n_L!BQa@;_ix_;iwA|Z8duSv37_YCBqsO-RE
zdX;5ysscmT1#IU)N4}}+tOD=nQ3H&Nda_g7Wrbh&U|Ndn^=L<w^<$rwKNQ$aC#_H<
z0Io6_$=9*j6%}xyhKz^lIHwF@55ds&)%1aPu451JNoXE0Hy*W*#7pzyWAR<{ZQ+XQ
z9}%0*w?AQ-Q^kO#to2083z4ebOf0Wv*LWSMy5cSyA5C$U7cTZI-_Y--!sYs%ADib-
zhBso1#)dUgM0Y@hNOHh^BxHPA3!^TF{?ZCGBiw*U=p21mg{ota!M0=7>;n~b<JiGN
zp3!4N#Y#@p+y%#ExQNqRRZg&CUxq8-?%Ybh@YCU#Q@3C&<A0M*d~V!QAVPA}3)giC
z5WX~+XsGqDQtFy}f{Jg8%Sbz4|BNVh6?-9{UV?r5l#06cH(lp2+g6r>+i@3h1;lGt
z(+|qJ6->#1uHmZkFjzIXJ#BHPhAYw%14aa`nRhPmtKvbbR5&**Uo|YEirc`;B#vjZ
z4*q>5jUv)kgxnV6KmCu&+CX?~Pw=DP2(`a)Cc*GR?dq1C4&?jWU8^gwg!#mF82}zg
z_09(h-DIv6WAR#~RDi<}rz#O%;~#t;RNH3=!^ChC&5bdBE$1+Cs>6Ln?UnZ%M`gZ!
ziGdl9<Uvvf)AsTlHj@DidShWTRrlGPb}Zj1q~46-CSeY0d7RidPCSFtAVGGDF~kW+
zPmW=`qRx*=4h(-@&PaJ2IYC4t^ZiTwAOudKq7!=PF#I;m4!X(>VCRpN)oGt?4N~ND
z(u^Ulp2BGn-RY8&kcI=#z3xw3Mn{)MG(=qm>O`^U&c!*bR~6S#Amn>E@5_}~&3nlD
zUKa>yeeP8(3NgbpTY$jm*;A4|?;Mj?c$giGDm$W>(QUdFiV?Thwsfl7c{s8m%DX)y
z9|Z+c`!PuXuckenmP<TgtLxFA2qteJ-rn;RgQwAj5^r9!ezulvfBiWqh^;aP0~YJ6
zpojm|d+mYZ7H6N|&M&M#c(m%Spqr|NeLJ=hW3Y?it+Ul~_7!_xuYgB9Jq;^FSPg6j
zHX1H5Sobt7JwiaE#t}JZy9Wqs@|4BL<p+tn6L3V1XbSh!M^h-@f>VauI>~672?YOE
zx=%7})ZY@)c<g^#Ai<mJeX4oI*Rt~E)aT1^Nh5Lb@nWO|)pi*F0na8*@EkzCY|os}
zyDZLYJC2H1EhfAAfC1)VYC^jL1UM`enlRM)zUlr}sPi1aNUS%C1Z$e6IL!fWcR0RD
z#4;eBH)|ZP{dMsnvDOG;qRC>J=eXPEAdoujr#5vmcVGdknqcuaA4Ja-JnwZmzp0{O
zo-+fHYrdW+;cJmn+n=ePwb?DMT?v|jTs7Wa@8V9@vTtANQa6x88q0`?X+8GJ0;s(<
zkQZIU3?>&G6o`G*nf!(g9M8F{<5kNk<fQsAIU_HmRIMQ{KV<xsjN?j;nA6Q+3yg`Y
zC8y-{gLc(oo_zl#Q)=4RdA&)d7F5Nlf1}e@qh0YIf7D_MNmP5V;Bi=l+UP8y*lyIf
zov$J^ufZyqWCMTOLGGjxY4a#b2;ClsPZcA)-)+{sE<s&%(_|FoYnz>BTk%TvJ;HA~
zF1tB!Y0;!*+7DUGpF=3CSWbCK{N4u;scO;rKDs4sU0it)<FDZa0Usw%29KepZ5^8%
zJL$o5g__gU*@mpU!=pHT*YG1;CmjwFwywY<=9QWn1E&QQ0pIiPZJYi`C8B1?Nu;R|
z_A<Cv2+PhM?7Jz*-+p(bAjf8m^4M*G56mga-_946jdvf+NyBZEAWC=8pIxi#X6uc|
zR{M1VER*yLutTrQW_e<&A0%wz9|_8r4RDb(P0rzqy{Ltop;cU{{asr-IeGwir3asO
zTjJ0aKERO!4o{1Y^6PpY+*J{L=Kov%Hu+sEvm81*H*3)SZULnjhTs`j|4sSh7ItH9
zO$t*ow)mv(p^<ipI!%DpFdk90saq$ix!x(5x*u)R5|wZ>7T>%!EO@E4S%WH?XW(O#
z%O1Fj0wcTX%_+FXOFo5pX}>)pUZkZsju|?=azjs{u33K~;qWpXS&@{}fmASM+V}mo
z%XrBs*U(eeyUdby;cQG7RFYzOA%mQ$;7Oz?+glS!uJ%OQY0U+n4jtH4ZdR@YwFs!T
zAA5$mnXZqVv>o)Y+cJgyLRKna94Cbjv~%+(%4GXFtZk@YxU@Ya_HFjlU4|oK7n8&Z
zjH-E#jE-N(;|F_+xy4<0hWX}#-`}kVfRXw_W<4!T`*9$F+=NDz$vypEzs29{DBF@0
zWT`DwIbZ~a04S;LKXf9`j%}D{Q9qZL^PPUgDFc!hf~{QuvTLq>ym-|5n7CK|CKy3h
z<J>Ri*E99?2c7DqkaHrg(Tng$pdNkWP)kJlXK7(agLU64Th>Z;eLLbVf*RV5#muTo
zb<d$0W{+yWRi01N)sC8TaXP^aa+vFx>eKJ;ly@=h)wGh6$mu9IScG1Aq!yH|tF84?
zr@}{fy!!|$zMx-Xb~r`hMh_TmavNITQmb4R=%V^<rox8$KM#pw(q8Nl8n+)>;de8k
zB!a6>MkS$7yVuHJD-hA?B5~q4NLSveO<PUKefLz6dj#hVJ25k!L~`<Pbevi}ftOLH
zWU8uhNgFmyF;RA>*s9R**bC}%tvGjmCkRL3`<D{oK{4znhY5%z)Pj>vowah;gG`@I
zrdq=xlMBRs2Jj7?Tx(g10Ku$Z)!^YLE9H$Zgw!0Yx+mK0-_fA2OAOk=1uq?;#}8b9
zqZ<287<yHWofc8VW_%ys6A7}F6Da{A;+|K;n?8cSieHj1_s0xnmRqwwg%mW>ML~WR
zJGDPke<Z;-)MY$#VCVbj@=AK05xP=w#j63eUp$D+jaytcIB4Nvwo#T0h_zc;qv$y;
z((-#U3Zr3(kBD6CCgiRCAfPgNWIDy$({<jO+HtxM@2|FNmtRPrpg$uEo;Bja6-Wb1
z({3+auI(v?zKM#r6Y611VyhJJD~S#M?&O)aYgZSJI<{hnc7r*GQ`<kqgw5MF{bb=V
z?VHO1rXO3t{9u=xsoOn{8_V1<uHc5~U4d|cs(ML^HQfT|@-5!DyG}zpb>5}kDw`P0
z6;4EGZmg+M&@ZL&h4o(s+oh#M%`$GqWC1zfTzLXO<+F==XS#~L5vH0KHS2~HVV)#(
zAX(?It2ZHC#s~{z&-B_A(_O6x?xC$l`bPoB8qFMalNy$F08vvEb~&jD*+<5OFDJ@!
zby6YUdbha-_C2vIP4hV5tyvA@uLr!RI-6%Tg)gz9$##osbda^S(5;LB|5}7d?)S|j
znU?wD4&hNPDe5@X#Vm`AfaV+p+d(j`HLKhz;+^L^X8`8yQjWC~&LF@2+dz*G4q<mU
zYOgnINNiXiv_~+2Ki*8kp_8aB32?q26Z^-VYyjk~R$r?OCc51LM9vC8zkHdo#(vJG
zXSKffaJ*l>!owJ+E|B(NLtCLHNsvCPNm%(#&4}$dgTrn~OLNB(mop}ntDaAWUQp#M
zNB(44=0z}4%Ot^g4?m;hDLMZ$48!R1;(vf$9YwTt&`BU)NRZ^tW;0h?Zwakx;p@<8
z_$B~Eqxq<qHr-G$lq5ia+_@g79gGtUKeoPm6&z~k0jh1dSr=EWwY;;<YPc)L^vrFq
z>x$Fr+&{0IpQF*VTI2T^;@=_hn|HZH9VBxw+lC7vYm=@99c(72P3)eq@O4|a>`vN@
z7gp3x>A1~XZ^8{-hPS~E@<Lo~O}8_<x6>_l<TmxG@SD-Hcw3>n_osE}&6V=3?<{M#
zSsIV8$q&{H;kJj!O-e{wd|jkY0{D6j@}TQ>B3{>tIN9M_!)-Ws1FYrvho;-aIzXpD
z7lyd+1YYLTcFWs>5p55f9fr`fHiRy_n&!WYgJYH_RCOE{`7a#7=_~xLw;=2D<6KJR
z*B@%T8_rT&77lHiFNj>GC@<?<=5HdB+~N}EGNl<;CKt`DYfg1$Jr2NUjV=zSxAGYG
ze;9~pZKgOydhA8^@Z#XsaW7b**Bt-Yy*uo0#Z<dJu9>QnW=2+d*dwED#*!cpnWb53
zd~XJ!{0(p7bQ)Q~ig66lQF1esHFMj=xAW)T)71@D>l@!4oEzp(zRxU$<FQ}qbwZjB
z4r*7!W`x7;u6aH6+B)H&+oOOu_R%)ajRey<fW{{hGFVQ&mjNvKEF4qK_K*u0Vh?#R
zJ9|D@t&zZsM^?8V04}Yk=d6CrDG*}SIS-Zmm^<<o-#6(ZIC)^W53ERIP$Dw|sO+H)
z&o*lWoqIP_199qUK#_#JLX2%M7E9*HY_fv0h?QD&pa|mt8>u-@wbB`j0>5y3Wc3k_
z!Z913`2_<5zfkcJ1A~9KXM{JN+z4;58kgATllvfIy=0q%t*mfdbT-eWlc|>5iFtrr
zS>^`u=^y!~;v{9ZJ-B9&*-M&r{3V;B%l2C4YSDeSXcz*b9Q&~Nu-G%^K}n;}Is|qz
z%j!0S$$|auqzr%kv9|GcPP`3C<tc+j03OzewP8&kP0|Br=Y1e-#{<2chJ=A`e)r53
zHyn`Z+5;}0N(5)~M~#n4{9*RuR41d8INWnfBbY=QW*Mv*{Q*x&e*amgV#2xoshb_l
z-qxco^ZAq9RVi8)$Jr^HVK*WBJJMr$L>7f&@l|QanPSVbyt=Qog-?qo^pnnr*GhPb
z%!mGm?kROK2z~2NU4;ci5y$)6$c08~2&=sf7l{?h7oa)mD-*WLJsl-dBQVLWN=2@o
zhL)#54o~3uCqNi~>vHuX)ZO`V-lYLh$NnaOlN1ThmKljp_jFXkSS!&eR*bRU;lu7-
z1%hv$#$Xd5HvN=}XzeX|eV_vnZvOR@r@5&s0oB0*8}P%%C1iDAM#}Ph=V5Ldl&=xK
zQkpk&D>sw)dEYtz`d+-;82;&6V{~<z?b_%g+T^F#_tq%AOoZ&4cJd3_CGyueK#XH1
zney)nHalb6of2_9s@I>ikcwtDvsTBuG)j!FD)M<aPjC?=Sly2f*H0J}$rUU44sh*r
ziqcHa9l=6EHnAYl9P%!if=U7+`P}PrKlu!$v-{H0^09`w)?n)ghLb#J*gJ%Wjx&$0
zMb4y&bX@go0Menp>2iVp%3Os@sj(wJj^PaNTDR8#*Ns%wilzopku~80b(xvWHoGBp
zNVl%Nn6d2e`0nx#?cJS1`fGEQ<0QPg=_lSqo{d=QX(YK@TLtQX+6lByM*GLC(x3Qj
zI$qzVhvvZjJ3ji{J~v+_+ZKb1%NtRUQJ<K&zhX4ILu`fDi`@A&qhIr%5^VLK;qzWo
z=C+@@pSQ@VN2PxDnvwgRJ$9`4BP-g41jW3=8HQl#C-hRJ2zH`y7-Y*x8_2{sb0FRP
z4Y0+cu*ZV;;xp%~otc5)n0F!^t+pLT-~EigC@U^EP}i8zj&9d2q?!19-5KXraeVZ<
z_{1sU;fmb(V17A)kZ~H9--0eD?#y7mVP?|{(V)it&$@EzG>X}xc7eiESXUT^AGN3V
zU!(MnVWx{6*L=|G8IYH0-b<VXz2des?yCaCX{>-%uJjcWbX=zV6LiKhmm2!9-{yP^
zCyXq$!vO|_#8XBVPUWrlpS<&hL$-JMadDuo7z25Ur!2Ep04HO0w29+4D9v*nX0{O<
zW%Y+JY(}oLm_pH74xkBVH{X){HxZAfG(2_KhCFHO+rmM5d5AIsFz?I~WI|cBYIgZM
z5X6Y{wc;6_6XY`8A$|2y_MrBqIk1(}=pb}~I>dgtQ}B1iQu4w2_2(m{#wCZHCP+8K
z;Jt6OsaTpj?&367mx-HlDM#p9I7^DzDIbA2rad~&vtbRBp*?Q>Xg%Dz%n)p~D^31s
zG8W(_c@mNl`(2U*v!uklE%uh$iO0J*+&hgwEEv93%J8ewcD_JvO~y6X2zpvq&q)m!
z{erV<R}Y>@o6PQ&igVfLM@~zq>eA01MhLP)BI1}>-Gy5GwgdNq13=%K&Ta{0YaR)m
zbe8yC?v4MFOJS;LSDZe$*^98Qn*TBVRj4ucI964+>?P72q{~m<)cbWB`DB>j#G`&Y
z7pJ9;=<em_6>Q<x(}ilsrqi;d@;7nn0=@}O7YZ+nX?2^x4SR`CzG4HcvU2t0b%E)w
zO`}SL>0ss)WBEpVSRiDtDBc*P?p|iKfY`)chPs@-ae~)vFBFt_#cRCl6tt*TXK$(3
zp7)COUVU;X6z6r<Lu*8xOO-Grv}u*Vtv4ui)74Zye{t&=fk)&3=Qnj7u8U7rAeMhi
z)LVAb%s+XYWYCYFQ|*>|G_zRFDabO$2UuYfO{<$et4ow~|KzmUkkt=$lfm9x2-Rim
z9>S%2D#zP2mr=F*0uH+|q)VE_EpF+X!GUwOAPoDNyo&16fD|l!%f%9aWfgJBX1a_n
z(ND62d^A{i91gXo`9m<4Z~sVQMi$j=BqpRq^dWb0^mpz119f>>vzmta&*;H+PumfA
zj3+f6)!TEMcq)sCVHYxFcKs42Lj_>;xK4cr6SxU`8<$l_UN+tPyg{4vwYZ1tAiJiw
zOfZDo3a|WHxz?u<D$ILIAd$e`*w>h$w-E~B0P8>BDdgsQtl8XS*K|6}pKwR_Mqog`
z-DuXjMms_uSQG9?t{HC79<W1Rq#>SHt?Ee*qr^NfBEvmjuhx}bbjQ79>J2U4!Cc@t
z31)iOrv(0)=+@EM?oxBN^~U$j;z(rEvBJj!U3xy`hV~}r6*}?3&9|Ta?_8-9TZ7--
z^yRgFJ&KV&=^&3WSukS})eWLVl=i<ZFoK@YXZs+<aKTvWrx>Pp8K0QEpydpw-5R8`
z-`6Rx3u8s3!*$gK4vUVs-EpqNxrH7h4sit0dw{h=Ug|r^q_nkOqy(pkN=v)0s6gj3
zPQQhntKk4P?iszFNWywJpG*^zJ?jZUj5n(BbU0}gzQ(<7=TB;awLUL@9|I%iI3^5v
zRG=JiN$-Xu3i+aRg>|NRpmZC-U<`F#?SEQxhsD$>kN9>vF}y8tx2t>iHYPlH?cs@=
z;X-;19LV+{^>PF+s~60UyFeJ{-h~lz_ri95aM$Cke_341yW4uLl{@H`wN4mrK_)R_
zADd}KoqYG@Z4ArKy?p`-q||DL7n@h8%O{Qiipso7hpSxdk;%y1HKq`A2_E;=Y=1Mw
zN|e})q-J#fyqq$aPgVlE>Q{9dPa~T{FL(wILXzqxni`S22KEx(d364ZoXQmHglNB1
zWt5IReY;({sfb%-o#}ZxT)KFC?6G*0`cTqYMxACPi{#f40qP~#mDV+D;D~uOrIY{n
z;pFH$u)iWVd%E)Mqm{f&76K~L{2juk)~L#C+muRFW6cOS58No~rkegsp=JbTeoCLa
z1=>2%y8NIt4KU#IFYJR{)Pn5Saq28kcw@$1PwygSd<KV@v&}5|+yc5z(mU9pGR^57
zH+Yuq9=~Dlh{k$u5-#pj4ju+Q!z@a1$tTwata*kmI(UQJEP92DO9=F{Nj4IsyZ|~|
zx0Z=44;AVKp){@yL%{80C3FD$PhzomLJKu$o>YAaZjAKYivNY@jen?c(oSx(?sGAI
z_?A)fV~~(_uCU^^Yu?!CHh&1414DGyhtS9C;X-sMr+)lvG();Ao65E`<DWiUxp6=z
z=S|5lpJ)9lPr&QpdEhqPr)Ch#XK~2f{jwnB4X&|ZxcWtA^W1P;bKZ{GVCIfnkc%Cy
zyNPlglV`q=_2Z@%3yP)@nB+^Jb=e8S+-v)Mb_HdqAF;O$!P1k$mrkPQR$t%CU=~b+
zE>TkzI0X7=43=19gG=Fk8P3V{mt}=1B*@X1p{u2*0Q1#C$z}oV&xXD50h4^EIL6PR
zJ8V9QEqQm9qeLK2mCN<9f8f(u108Io($Rg`*Ce;tYJP<mU}{a~+eunsehH`*;ufrE
zS!^2ih+qZu4+P@f+4JwN3N(6HvFYEkIR2KDE&$)<IBl;eRwrk_rKT|>^#Latg(0q3
zdYJNU8S{fvDr3b|exwu-7(9hK4lKd3v#PJpOfo2jSy#`d)rmb@h9fN6E%A&BN-FYo
z!E?qicRj0l!6$cD1DD3ohTm3w#^9*=n{g%_C^K>>+~=NFWLsL7-KHTo2I+9iUZMZd
zM(+Ho3tWJv(`nqLB~Er72StSHZJ!1~8XAcX#Z^<Itt+RmIeoyvm*F+qMD6Q1w|hc8
z5YJ5pS)5zeK!J-j2nG>9bVF&nHvS-AFuHW^Zu&X6pOq|IEA2G+*Y3frgJO9pk7<(O
zWHruh>K#05+GN%$N|s~)-88WD93|_y8_CCe2wmB*Ldh4dV&;7?>PfjS`r)6hJP_MA
zjKA7j%hbwR)%%+j*Kl*5^(t$I<6diLoUT6KDFa@B2!Bh}<`<^>e6PAWNF-THpYiie
zc6uvtV<;o=SO-~rG{t=!oztkJ34l&tCB5BUcu!3#l-U;OD{#snxm=aAnni$Q_a5qe
z?(azQIN`Kl(yy;mRo82R`Jr}%r1(PgR<N7O7G`d;y%cL$=g%ColQJT8KLVqQ^1cVe
zUf&u;P4zPa`$k_;SN#S@i*5Os$cWA$A5*kFG4FI8f~}Ka_K|#Ue}RXNc~=D!)j;Td
zW$$P$*PX~7`Pf=QNPJ*e`D-u(cB6d%3AovUsJmp2h$bLuKD>mQ_V9dkp>++&CkTW1
z-A)4;8N_3O-t>dPY6g{l3Zx){Ibem)$?nKU=&3*9min6XLS#2}7|*L1hJIFe(>>(l
z{g;r4NWQq4C3vd1x(H@4Wpu+D=e{qmQ1+|Ema8t7nVh&Dc7%u+iD}z4kZuGag;XS&
z97#;iS`-wKg*aQcqx*Q<@CiM=o<D@fBc6+rQ?06wVnsI~!llRrRB9tKwUn7=1Leuz
zCw&vrerGrITFBa#!3W$vdf8F>`;crz_~ELokYiRD=grTa`%YsQn!k}{GShMsHmXG+
z#Ji&g89o3~1)`CesfIC2s&UJTLu_Sc)+VC?Se)U{<rH(MiP}oUh(fAXEq&}#Gi8pN
zX~0)kq2*-!E%)uepKYR)J_ndvx{Gj2jmmwVh%W#TvYmbSP7bbAi6Ti=WOfTk58gPp
z!EQ5`c3nKR`%qZ$!sQ=aj5Yl|yyht+gXCA%RQj<Wz~t|^_7H?|^Zddf^1jEl8A)-Z
zDWmPDHN~wDAsL8?!QjlpEsbg9$K%^{tM+f@yQ(4mta3q=pq;kFlleh?ijNd$g68}N
z5_c0b=Q><h(rxT=F0@`VJf69md5!`e(A@2xC%C5!wgD%dYSG)q;~qjfS;-loIUw0j
zL@XqLg!^rNfby(0RmA&}1d5%2*nrPHa^26ACPrfnzokru6H*ju@4xY?+v8^xr<84D
z2T4pjW{+^3vOAtMNd589(_V!ED;=L>Gol+Uxn)Sb5_k$?zBwC5f$>SC?sovg58))B
zL468KLb0*#*Q{A=rN?qUFgH2k)^GQ%J^{z#2QO;WvKB$753AQ$<&DR{5p5ub8p31K
zs<1}GvGU$SATqp6oEpcFtP0djCR~?~Og1hltC!C$E__K0m+Z1(-6gXBm0|o8LllF9
z%LvW?kvxX^$y|f^k!n>KV*eA_nmAOYQ!m!f1U4?&vQN$<Rad_LTA7sX<p2p%eW4ri
z4PS3BXa`7VOC9VdanUFA)1SnGG@eK6rmCC%HpWnb)x)gla_=sE44)iU+^0H(v%pr<
zOWRZYg2VqT<L@c)TPu)I!RSNh2A3dG5Z~U%9^oyXkYAhK>?)jB4TxLt1368b%f2}H
zO&eSe*Lp&pnQ0>>$}guxCV>aQF5eT|xjYFjSxg4f;^vFD(~eN5UjCj%wXG71I`(N^
z16RJyxe5>Tq{y~FhhIftovT5fkQA4JrSRtmYR~UyVPYh(whqYKbnQRiBuN<+>0)IB
zD-hjunT1?xa8Q4b@o_wMC`gWWzM-aSk+u7Y2b~?gm&KWdH#YeAdC_k*BhN1QjdyIG
zCa+3Mx>x)6Qu1M%=d3H8B!vm0mnI{r-|AGtM3mNYWZ2rq!wkppP?(0zCb))dEpL?7
zaMGtSKNgO(TYMS1Qeuu|(vfYPKHYO~KxdDk43S#u?bPJB1QJfcy}}E`q)4{oAUorV
zch6Y14HX|`)i;WO5MpzMBfkgP>?E3cbZfA0S(l971pJPKZ^7(%JJfJ^6N?s}1#C9X
z10=jMts7t-WBHOu=gYr}jY`?lh;oP7X^JBY13RR6=I!bGl_$z&XMmKa=1;I@#1Mo_
zdpT+Xr1qw&w=FdGGuvY;<G~?4ZvWY>lOiozYgolqf6P8W1x#Ek#Kf_0f1J2TiVRy!
zEdQ82vN1Be7CQ7q(8ZQbnQuiRm%aVls@0}|c!F?GR@ied;V;;K5~Iv6gF{^ZQ~i~;
z00*f&g?%b4E#DdpFO}R@ac8s!W6+ai3jz6uPMOMH0XcN4PWMrV64R3Vo)mdAiR{*3
z_ZQae49wN`o^-NzUzB@i?N~S9TKx6+?&gshkzEQGYFuE&M)048cTJ;y_Ga81i70UN
zSS+2Ls?oEXnB5{ugYd5VnVa>NF;~BTHD)yZ?}|^m{qq`^!w0OQ8a5veWT}0+JOvf8
zQMK28UQo8K`Nr}mBp(kKn9XnDCzX~_mQV_N<E#kIFRwTPsOxy(eSc<kQ?Hj<@hNU4
z(65z@<;rGUs)nMYGIqY3vGhi~VoB;!_@GM@_SQHv0lS|0ev_D<8aXl&-*dhD^;3c%
z$B5wD@4YL#7r$fKT&cB6NRZl26Uu((*{izUv66SdpB}iDBTaQ*6AWl>qK^3&m|3bs
z6#2E(p0~waL!6D?w5vJO*d!sUcK3^i6<!;vi!!f#?6WD|Aws8%c@p=6yBVAA<PE<r
zvKSbRs$J)nBn+7HE|9N5W9o_AmWOG?48@iJAu;^`R(m|#m0>2UNjookepz(eVzl1}
zdE<}<)}VIQU?ztl;kPg&W4>x<LF8-z(Y2fN81*zjcP0<<oQ6#InR@Y)--S1z@@-ek
zzd^u>f_G`hZCFpi)qXZYP-yNud!U0uuxXnGi$86=BF!g5dJRM7pX6fFE-aiR$lu*$
zSgj6m_fokZguRE+tE=%1#5}6r*LJIVo>)9*YInOHx%+jwHIf$E#7YS`8-hJyR->L5
zaQQyy?W_Ywf3!Ae>o=1Rc;*#$+6cM&^W0eP$SO^%Z7IzQfG}HN<`g`^N7H=8?iS}l
zkGXwf7mo<FDrD^rf)?@(@cW0et3~39z8!PF-p<7%$szU9Gkfp3&!f^TTC>9Ixj(lP
z{*A7BeBZx9@?kX>Y43h%2X0$$q1mh6c6PJ=7Z?QB`+PL23E39US?of|xtU{Bt7qx#
zJ!3@Os+qhEz@0<+#3bUJ9CJaaSvNrCS9`+o)*p&ytKfEspV{`SEG*%P3ee1k310k;
zTlX&is{h0e3FMcRcu;7%Dimv2D@mhaGhr<PtvkU&O;&%;3eyki<5*QDnO;Z`2)_B?
z^r#<Y#tmjvn~<d9xXOae4JyQMKXLk{07d8pmJb9Ok3S5-S9=~{IUko{Q75HPy)4aZ
zu*=fVQ_y#E{B%2aK1b(M%TdQ^5XCKhycIf6#HwL)X7y{_Dalz-KsbEh-|CM)0NFGi
zi_G6QeUYpjki@%#XKh)df+H6jrgr=J1CL$uesV4y!Io=C9;d^WTH}+TvjUxwP{m3D
z<@0qyK5>>(D;11>osyZ0QvsSBHZki9g}^@FV?uYy#y>#;5C<8vclVS>&)BIAD^+nC
z&p^+FXV$!(k-`CU!2h&5hUwhu5Gn7u$p9*q@DPvVY5<PjHMphO+L*w9Ip@&fC8TAz
z&WVjF+w%S{8=ibv^Dfx(XK}6`>A-?y8W-;3w%c1uc<Q1gUw(53fD_Gbd#nrVD0onQ
z89$H}pG1aqnz7MD(w`;vWk5;7V|vOBzX1_Te_5y8j>KF4KZ-s<pTDJ&$3xVm#6?Me
zHu#FSU#H;Tx}F7OPthCEB|(gl@uTtIYNQhAsN<1At)K8oCNW^xdVtk^6@_YH>$Nbs
zx*QFMRx*e8lQiO&s$$<3M*a|edRJzzLkeV>ppzm;`;5`(;sQ=HpX*&E(jSJG9;>?K
z;Y9V@ny*haH?kw9`J0pe81(XN18y~^V9!4c_-_9LCm_|Ar8q+y5`l9ItnHX+YH_5G
z<EZ`<sKmvwW57H`EYEww3SCL~Ns&1mSDJRQ>l$RA%KR(<y~HlI@!CqXI?f@&DJWTX
zS$McHEOcEK3$Q@bFG=nRUBErci5EQ@w>cEa>Jr)(e?bI(pSg*DFi^(1oxHi4nT{G@
z)Y(Z8x6d}xKd!Mn3h-WY{3TCbeNt6g6qZG-&N^!nF!bbCNSp42{lw1n{=)8$Hqbsk
zLwL0Y4x>%Qcg7yIulFLoDM0T>E}n1W0Sbx{l{M`}!$R_D(I{HBTp{a&pCs;BH>msd
zpZ_{Uw3xl;1Lcx>{hq8s+=#cRqp(G#L$)xklK>v6<1z8REsMKKfhAX7tKAEioxEHO
znj)Gll6S`GKeYoga0^zd3T7+njc&Ecz?NOlRj~`E)^S*@?{l{V*-%t(HO}87RpH2P
ztT;IqK8O&-eFQxvFeMp#hG`5*XK#D#Ug-W+=NW&<qbXe&>Pm5$mjEtMpa~$ff|0+y
zRZ9QqPu&<(M+k*iC6HfN@~JyBcz+|nuZXYD5Cc{iI6rv&Ql1;<cI&QUg|0h021e6d
zShNzz#;R3MFZvJHK@OwAS63M2NtZS(Z6%r{G1J@AFfx;K)l@H>ma@GFaSjKFg`pK@
zrD~<BS>;^`1&tY5iT>&|?;~VTrrY#S1|aH^P6$)m_%Dq_-*(+%Z^ZjQ)iHOlYByJE
z7S8cKW}}3%N6N(N#*(=U@LucK-Bt_%<IBUi30Pm}S!ilse=mAbdhq|l*;_|N*|vNC
zw}OPUC?PQj3ew#O2vSliQqnCg-Gg*WNVhc7Ie?UOH%NnY*U&J)?;P*v-uv0_-tSub
z{jD{BuoPzIx~}s)kNO^;qe+D&#)=E5dcMy?j1MNZeCk0Vv^cvNQaEg~nFMdlVht^B
zho$tH&pOUnF17}%xNt<Dy(<eHeFIY@3n;Qpecj1PDj*{@?xgnGr%7SPQr7c+gMZCs
zEtDk0ihPDbR!&_CeS}lSq|Vio?u6!1r`HJk!%JKBl#F1jm|%mqLPn!6$(2w3w_6Ve
zR{h(qr!zp3``%y>x?&OIzo;|+*O#J~mehWK8DY;KEHI*ZX6YrG=3t_DW8a^KFq&|a
zbxGaSGe+HbVc!HMDHF*P38uQkyDvJ8b<SIeA0yc53F?4{Z8NJp5D?P5p!=(+;dHtn
zUFcvbY)+R9p<HCp-%@d+&PZIXScF*JKM84uphQSmszUX((L5UrK$ZCs<DHwuSRJE<
z7t8TG-^9;)h0Dezr@Rf;@FYE<1S|ir06u3LCq26w=N2*lVFrrG)#)7P1~F|oBDGfO
zhZ=PqrsuNVEvYlb%?opiv*goLN>F!qjZHPTR_=0SYI$h?6Z^Ra&^0AlH6&U4Orc>J
zq&Oy9l8QL)Orq_h;8+oSn|t}gziX!EGp-A$TBG^U3(Rhe7IOI$bV(3vMVHNWQ9`SP
z?dvlzHxV=f?3X{K(S07#PETKG7<}Pw>G8QRQN6Z|y5Dp)w~rRuqhjC&1bIo9%z__Z
zd!h|kXw8NhY;TUM>NI>;y&4hHz6f*TAgm@K>hSoU6ZwjFnj%7vn{A{yt3@<yH{mde
z_gvh#cy!P-SME2Rjxecx?gD3o52i<#6)B>O0n6dsmrwyCqjZT5QRHdUio@%RH<gDD
ziuAjpOTYH2lOROjww@}ok|6ZB`g@`)iAj7oOlA3V8#3nE?zCrFfc=&qTkz0m1|Chb
zO1zATcErv$4?^d1P^?3h{G=;EfLNg@Sa{QE)pUJ|C<=o*jhynjy-6rf$)79!AQC(>
zVkPhnb}i12pUOeOt~l)kiy*Uz%aFs)TCPg`xpQysxrE?bWgP`%(W-ZBwR#i=;g63t
zrnaN%<Qy_u*5Y|r-})}PfO<+WVJvJ}X8c#>%;}MZilNc?F2jFrr|;M|U3izYK;CJ}
z7O?>O1byaBL*&1efV4vh%cEXCOQWkwn@I*_q#BhTW&a*{om}VJm2<G;+#z$OVrUVB
z){{4EH|$wXVJ6CELRQ7x-<B&3CEMWn@oKq%#G^TeN1YkYJvpsQc^M`j$d$vCFDvPf
zZ}M%9#crSZVJpKweJeFOHd@MmM~n7V;!NTjF$T7VkBG*jJDE-O;*rJ@B4{PcO`y-L
zcCD)GH}kx)5A!Fr;yF!apC&(|4|xhiEBZlulk8KEd(GD`ibPe_^!&s;Es8X_A_lny
z(#u}zCLC8A98?2o5O>%6O20UHB8_Rj@_eZspj6a(473XvF+iy{&iv4UR$#gOr%R(k
zN!{KL9OzQBB%>hxh?Nc4ToQ{3(H5{2a!bOFhyAb_j@sZbqh6WU3SCN-Ef9^6)tL-%
z>#CnX*!$LGqb9R07<()2%CLQGgg1Ruv2gPH;u$?H0DP)NnY+(*ajp{qYiKVxE6ZPa
zrkWB5#Fyk#<aYhBFAn-F{Y+&{S9so|d4AG#F7r!#%>8)ZD~w)!vSON<aAozFQz35m
z_TmNBQTNia^!&h!{p2ZtTgo1?9=ClY(mTM;usR5>w%a+q@41URGOSW6PiLuKar(ka
zQOy*y>4I6~fOF_K5p<d&tJPZFWM;TxYY)gl-jdXh|D_=;N+$^^d{+&+9w#;nE}YSq
ztK}IEpuJ7KOY51c`yeGhH6qJOp<~6zD)_Mn{jWmWW+@I7e#4&!-(w;lUz4hY8rc54
zyldeWjN5U4(?|<=oQ3XQ+7o7i%x*A`zrOOGS9C`L72`sEtnj<)4Psc2EVU7q){13&
z`Eje=9TgCK=mz>7LT215%$}TlLC~_Q8Ab5uf3s9t{y5W*ho$83)g%w+rb`M|=f9w=
zs%6w|#^>l3)11G`ATdQdTuIz7{5Yd=6gCL@l1K|Ikm6{ZWQ%tQpgXpxndtUYn7%O{
zr2jt*nCEl!7Ns1TwS*oFy0D6rqlZTNR1_sK&~dCEZ-X5(@*9)Ia;5vfo)KvD%;<1l
z*QVI2{8Ai;rLD8m_oFAEe0ixS?GJ7H6}9$fvoS`|2|EgWx3tMJGoQA+a$&JN)em9j
z`BrSku@+A6keSYPE_-rQT0Xvc-}AswAunNBoHg`zXewmw#nLD7afK{>(%ftd;5BNj
z3dMw(aM*th%im`+=Tx%ToH4Dl+$_nT$SQ=;5=t$tmMm!183fU9Z-~0m$^i?ulLgJ{
zsrc`9D^hHb>1*ar6@$n0Ls!pnKG*cyuVVh?7b+c|AQR*aGC>@+-6a^<kG!7np4S1h
z`Z$!21pD`X)r7YB?{|h4Eh<VboiI6~aEb(9E)z)^U*@eqcD=S&pZ!*a96>DawwrpP
zXG9O<Tn?8ZjvcEmsY)b8AcK)G*k-MzkMZD=H2N^@JfHCSZyy|vMLh9ulZRh(3`jfM
z`6YQ2MXv3c6&@OC4|R35VW8Zs^<*`5{SGfl((mjwSm}y%;IRC$nF9nj3ea}j3J=DV
z!^vz08k=PL0n2;1QB5CZ8y^mq@d2%AOSUd>O0iIs4M-sgAG2rQ(UAyg*>*o8Lr{GB
zc?XB0`JpNADF{9bLLQF4rFAqdVe+WtCJ8DPz;j4Xk+F#y{?~4P17!ciC5l3azCk4L
z;768A?rX`Gle+kI(UwLDjx4cyYy)m{zxQjD8e>vyZe*ChltSUXoZ_j_eW}hzmz*@!
zN6-_WmaAP7$Aq*noSPf?0`s$y`okqLQa1l9I4^Eb4mrQk;x$d!t=XF7IWztKmH47Z
zRQU4IEoqv(ySZk1Q#8{~!tP_lqUV<rm0qF`&R#p+s{5?bcp|I~w}y@vRmA?V?PQ%~
z?G0Ao{zyUx>n}bz&gpnY&7B<=O&#qEshrEnTD$K&^JxO&E8)W6&ezsOXwOZzCsm@A
zbA!nX1h{4#qn+9D#Rj)JPTjhWvROUo<u0od3k}60V1iBFAKtnGtIM|<tt56~4tt=C
zAdip7rO;8dv5LA+qQ}msb4*@52NKIq6ycEK6{8Kqq3H21>`<mHS#<=PXq?zW&B|BV
z9oxl2q{zj^WM;dZ`ti&Hd{|U{V(&Bd3_oXS{dRF35l7`Zd>92SD7vV=8Ir7?68huq
z*g^&0j#{*;MV%k{aYZ7(+BAa?$jFPMX9|@*pxA6cz<Yh9*{D-;d@8Q3>xx0n>D=Ng
z-EFhq9AtxU49zo|xW}phQ>CXlt_iVPTnv(nn%U21RvtlKuZrOIt*6IS19`BZC*#KA
z#xAlCq}TPesA0$Fc&HpFh&UIRiaCq~mFu*spbxXNLF&2cTKolmDgy=k;D5Tb%Cu~5
zFL}iRil&hnWsp+(t`0v$-KUvfN0e>LKrJ#;IDQ5F$uAj#h=ufu<ty1-;NhnpR*wj0
z?(K*Br4Lv01GWAwy<}jtadfue>8YLUzp&AIxP<B~{hdtj04X!!5vm@=GCHDZzsk&S
zh7_b(gSG-ap&!aW5;&Jy3sgWiNYF#3<w}Q<+%K(8uE4{sSa(7m7p1^WPJzwD-sHQ=
z_gs3()tl4GCq^UnOjF1_ZOd6?_dH%FkaI}O<_jd+GL=PLMl$FCVLZzfZb61d_LAB|
z#Xkw;kd^0dMSl{=I^4(va#lpTMPtmN4#+6YgN#zBCY<rExP-1W2rE`Ne_EiBU(onJ
zKW7hkP`#e&#&d4sO+%?mg)KY1zC80W^NZQxlrIK!gS>Omg|cZmCS&p!t)qoTY<Sv&
z+$lMboTpb%kX5Op(?B|4I>E~<0-ZIi5XWEYCO8`RID>j&df?f)`9NAMKwqc><1u@e
zcBS_dDEp<so3!TmAVNBS%gI0|a8~qge2X{dy<U+Ru3&2nJ&JW!cVMPGpw_v#Y#}FX
zFMCS`O*jwM>fw0MnY852hI6B<kz*Cv_<jeiz>0k{xr7k0CNaoBDCfx0tg#YMM7Y7m
zSsKH!`-J0GXIwz?JVW_L8Vg0r!mH(17+tEfeg}ML?A~<8AN8WLq&|NL_Kh^1WGIOR
z2UXA^B&CCoRmLuISh@#RKIh1yFsPe?XLw*zkUB-MT#SM@$6M%YJi*r?`NnbJL^u9q
zcr~#&**`5(=}RB7=Mw9hsR_<8r4JHa^6qY$iyqjk3j9~jx5dXZ3Oww%ZYzF*t#!wF
z`_LDK<R<W`$VaGYZdb!OlD6j+m8s4h!|4d2lgkCOiJjDTRr`96*^_47{S^o_mxaQ&
zU3C9;K!_*FHSxPoGh_m8GkY;9KNZT^A5rvMUHE!GATh_Ma<8mOXVd)R8QBumKt2;w
z(P5cfLd|bp6SvGI`ac0W+hA@?%r!D`X5lq_WOUTWt9rU2vS%fYt_S?y_<d07>mgEi
zx9?KLX2E;06ya~uakG461?r~4(-0qf3n#NCejdAxBspw7@{Sd;d3^8EN}s|>KfSZ+
zrt967yvg|}JVpIX+Li|kzL<n9r!OnheXXA<>aAF6wnac=)r~PPjmew3fBs}a3F`VZ
zZJNs+bLSQ@_4xssuVa5$f7@b8vmn7TE1KI86Q&e7+hY-^dy4i9(||&8(!k(-=Z*3D
zl7xe&B^S%&T0xkUJ@?DkQnF)c_C@BU4!H#f11c8W{}OeXa$4{pgQJSn->TEw0?LjT
zF=K0$?QCvnMpI^z(y!?$fxeZ{@C{Q7(#JZ>^<5P12hEEw4v_t1=_QgRYVi5$8x+%?
zPQN<@XW;@{c@bw9S3+^+Ew7qOfVOETMpgSe;c7QsTAy*M%UUS&w@tg%Rze~J(Q~$`
z9>-rUA`S#9=Z7IW2<ovXmnvPpeS9iZi&KRyfi~2(MEol~w=vy#OIhJv6&puuSGKbO
z=oeqp<klz$VXVFvB*lS*6S#sqrMea-<qvpbTAx=YAcVqCy&BmrWOORYI3i-fX=FAf
ziA($FV{RAcNTSmUorhXHHz#(^^}6uv$7T3H4TpB~>2&mRJ?8)j|J#ffmkmI8k})S4
z#&AL-fh^Cl8B*1c_vD<X7j4EWL{d|@q;7n0e@GrjAlPYLtr^UrM;PCg<fOT*H6+)x
z^=K~7OfXVRr#N3J%NW^772r;@|4}@og9m<*vgH&E<Ps!lt9T7zhacwCR~|{(bA&b}
zp6CGA3h3;dKM3kd3Aa!y#{aG~ceLnO*t6ToDhg_yaY=9@LE~e3CtCLfcT!Fy^M#nA
z7W86u`|H<ty!x7H!cUAoSGI*MM2D+%+{Qu$WBbGN_wiEAL+HO%9|-Q(bV%7pUuRt9
zC~1Cbz9)Z`xk7mT(-?9z#TihZlGpQ<>u2YUz~<=V00ybv@bsM|RgZuKf-RsXQ3~*m
zQ8Fz_1r+fg+dCK?n%a%-DwfqQ(#j1Ta2kngQOZz#QWS3#2e`(<B~jc}k2=MonlTf*
z1jW7g<7I7XH|hE7T|-!xwte15<-jsHa7ipGMuz1ZU@MQle4Kx5s<d%UzqH|ZBdbiB
zzn1Mc7$*w_Ilh8e%qVdwf(@g);`#>x8NF=;1rORJ9iE#ULMV)A?`?b>RbMg3fkg#f
zPx8+sv%it=j2&knd0*=0m^CWnj;+}#Df8SH*&C^L&nJwoeAtDC9EH`_fZKag%XKrm
zs{2`ph%nrN_C7BTW>hlCT>IH>vhvP>)a3_$TC9+fS3(vi-H|U<LwJS1qD7~ImXqvL
zjFHx-{_+v+H_LV{4{m{-QN`+&5!^JX5*6q=__Dl0o<m|0NxT-XtKo4nzKy83yz9~e
z@!==Jvw?!f{E}r!P5R!z?wBn;5@5K-m}{~Pr5eJHzH2yyxPEvK+nFN&PROJR(W$WO
z;a+7lnhxsK92e+#_1qy^;3Eey#rM{+B$Wb&1Ubi@tmG^!ANg2nACVC`Hz_HCWjWY`
zkkNq`*mqrcSG2$(oZ`qLQu;M6SEEUT6oJ$f?7#wwyRAG9wx$=DGK~jd3q}Q?H*$`1
zanM21g5=eUi!GEVJN{{!ykr#{-p#iQ*9>N5Wn-w;TvE=Vg{LBh;XBPi)x2#L2&p(s
zBc~{4-X5#df|)W~E#M|z4etc0?hQ7s<80)l>HR3Z-bQL4LPVgyTIRo*7rU(WvDB@3
z>$33<2hbIF-Z)r0t|=-GfVAd^sdxz@CTx*;vJ_GuR*$3KZ~Tj!j(VoK>v$Kd(C@d>
zp!)|^4DxHhe#c*MhTb~gR76pB2d<CSc`#JItQ0T(d~tN^Qu(BqP*FsHL`uK!tS*Ta
z6Rsh^wgRe~jl_JqqO}uu>Uey4DV5b`w~wYr%sVqUnj4G@&UecvPK6tkjJK`J9Z%x|
zN{X_V6L=xr17`(n1Jyr;rsOyW<F)LrO2|ec@-<1Z!q&Af7rIHz$v{sjT2s%$xxY!L
z-i}aJGeqFN<O+5hy!y8{7Fz4`tLr|_K`v}qhb4+0`#dSP*?WIsSvM;*PK)#|^rw|>
z$PVU3R9qXqS5h<Gzl%5J<>qy%6v)gVeh2Bko4IA!%7dn?k&)y5Y$H6R=guqcO(|RP
z>;`56LW~-bYxR4*ZgcTS-az`mJSdr0nLCERj55a6iVJ_g1{8ifbRrir@yFiJJW2A)
z;P+;lGRPtn=RWZS_J0;))cj^`QQl3f#}RGmIPt-}Xvk5QD_~J?emHLNM|K>Zd&CrU
zhQKF11M@`)o;cq$U21>+$O<Mo8@8?kPpV?=GD?Bt>`QcthnF}T3ei<SrH)GyYH1~}
zE%<IgJ#qj#Q}&Tt44(MX8orI}MEBq>Aj?X-c@)L#j{Q>O)AAQy{~`wimg9E7p6UJR
zeMe;SF;-Y%w46x59U15Jl68uxgN<nSH>W_5UPYAJEzrYzf;Dig#Qyw=yoA7O;ra5W
zMN?NjgdV8_W(p=6$u+l~@=@!O(K|m!zi}=-U6+;sE)meoAqM%P9B&{c%=W}mGDHqO
zq|)Xh0UhGe&3m2N%HCvVKyH$|UYIAQlfq%K%X}dr@^w7rdxcQi4FiXLX8)_xKk;=3
z-)9a9igb_4A1_n&>kk3+;(1}8pBD$AnMBx+UKMXP@seBxb)jgyrPb?pL2qX`qfgGq
ztAg0w{9VVs?5In<#ey6k=lsg+m1hq4K33D(Jx!!UNugt;#iQDevblMQxGfq<K7|Oq
z-{k`W+rD;n9X2h#m=3!U*_^9oqa4lWBpeyMLksIdx5ckI>w$wQb~R|$8!mkL(Pz<;
zbwd&;;ozBnN9)~R&uUH#*!*+mD460hc1_+T7WJGz$D{~~$(A7$5V8~&AegxLnBA4_
z9H7jsTA%q`)h0>T1K-~J+?BIIP(mXrW4AN!ulY)c(4Z-|cp5Z;4Xt{gL!Y<W^VuU`
z&d(FsU(*K2s4{f|8915Bn;g83Vp^9tYLUQ-(K9L+8Q~;_pH?#N@$C+FYXQO(t~jf7
z3%N#=zCYw4d!VZx=knpn%<(E{J7?`1#!tOR^Px7~f-a&!Db`nhJoi~K!vNea?`kcj
zu{B<)lIu<CQurt+8nwUD_xQs8<k+_3FOpkc)v7s|v%t7HMp`)(KaIWgjDDHSCh>5z
zH;jh^iWfs6Tlh}ui)>6;n|9uMLk4TEb3k@}W^8o|xl$I5U+PrQ4iT;L-^%U3-TG#D
zPET2IV{*RS`$!!<30*b}?4afa;FXZ0jck5|9r5l~j{1A~cp>0^?7_>THzYM3zr&j?
z$%agRezl~Hk2lDOz-5?-XEW5N+kf*>jU8X){22?Dj}f~M3LkAuU`dQz*U-S0rv3Av
z0EH)m%i(s0x?clI_FRZ24gzckt)nGE>5d5P<jM=zs}bdV%oMRi#ZfoTj5^Ch3YY9d
zG+`>9E8;sF&yToqT@-#uV=<;gvUbD;z2HO*ZB2FfrOuq@M{d+3P3{zY)Xa9ion^&-
z5cQQJ>N+~zYJwT=+bj;}W~o!TA=S;|>@y=QgD^^wgt8w1ufCGGtl(W|V88(XD&#^Y
zZrp9;r!|S^mQosn(T-GoN;Dx_VnpV+4&6xs;O|xH*aYr|EZb`|MjE5VE{}`IDDAwf
zirKBSJG4VTs*eF7+ORyg!*>1si;Zg`kjQgDd*|lvxb81BuUBokg&?=zfI2{pQ#^iI
z%!Zsux`AdQZP<tzBdRz*_7o7)#ar~Bs1B-paJaCDRW~_YlrlzYsl}GrID3|ySHCoM
z87CS^-U`Q$fC{LFLw`)elO?e!u2BWTkwEtO>L$-uDegqjU7U%$s0#I|Irk*RhyK;h
zQ0GN@=%c3Zh}RyY0bHO}J5To~Zzivm;9_Ma2W+>hrfH@6tm>1ap^+erSyDx<kA1>o
ziov`vntA052K&ROC6aEL?V|(NR^n<QJ_fm<()(%5g3UTp&RImmJwG13BQ)UZ5k+pA
z&E)}9d#Gfs8otG7%o(0YpA|BG%~HEn-@=u}pvsP?qUby4InzpyF!MCHpGFZ-b_C+}
z%Iiw`ev*nMu}D4DS!1??CS<z~-;fmhv>sIqUl0=HX*IPsOglaC%~#yIymZ4=PWzjP
z4bJjYv@j(BS`oE&6m=I7=7!^uX0^Yx!pti-<>K2yhgS7F-S`_9c9PGu89NQXmKpwx
zYo>M9uNP0==p-?zn$D<FMxe}9+;N#-{3Hx;DJ@yJl89(OBod=_qYRK4B+ZL@qHLP+
zq7eP$0^IvXDhxCb=W2bn&&?Dz2X)e&)<h6V)KVHI(?GhW;1=D7Vu|weUSTBX#QdS!
z?7~lkt{?Z*j^n`3@(YUZGYV=T&dZ?j?VYG(gl45bA*RUY_tq6IzzTW#oRO{*Vy)&J
zMiTMCbpt{w+hPU!SM|R7VY)?Ss=F@n3G(OabaF~N?Znrw3JAQ3l#PM3^-F1t(e*rv
zN<da5^Y)Jye!pyr-j&twe1&!viYEo_G$Qd?uZM(&y%W4VJ)Ve6vmcUt!y6f7E4t7O
zn_d0DJ1r-xG;XV0WpQ-)WcQimR8|Z}-~G*k;vZ&R8pXq@c!ngQ!()J%M+pzN{k-_s
zZn4ek;{vGzmY%wNOV8kPt*W`O(J`58D)(S}e8m`88dS)I@>UO(Oi%?Lyzf*1eJ_I!
z*Vnb!FI1aH@EcY1@ocb%I1UOj6OJX+*OSB~WtN!j`%dZ%^UR4Ckv*}CacmjU{8Poq
z(hb5jf^d(-HS6lJX$8X;XtN`el>JWu(eI$~n?l(%&kq7ib4HJ{Ex}L|ReDwT4ah`_
z%K8RfbW?o+=(sVvT671$j7|}1&qmp^HCDe+B+}7IvOysMsvW+R{lUMxMAPt5UiIAH
zzh6Ae!Wq}1W;&nUg$D-PT))(bfxgnj-H7Y-kQ9cN-uPbho|&qD^mhH(`7oO7QhP&r
zLlsCU(XVYo*GxN;YSXRg@=Kpu03|V}@UC`A{&KdQ6x}`>QC(?yM<g_aUwEm)4cSMp
zg$wDKa!@hi1wA&j#R?~nTXOOtZGM?JIjK`Zn9HcwEBQdvGd<rPnjZ=(D3CHaqX9Pw
z9fd<&S!Jo-gWfCo$gqd5#@fXkO-iT8h(LK684=tk?OD;Aw^sJxH%jR0%~_xxl0}uI
z3cbCr|0V@8=>%;LdpCHTqP$Mm|H1!Ir%_<R)D~Lh6?@~JJK-V@Sy#>f1eHKvNB@wV
zc3^wM$hVG;m3+(b^TKF6MKvRZx6STGw!P&XK}-Vx00W*f%ZRwD0Y!4YTYX1pM*D>_
ziWrH==S9S8bic*r5CBcNP(ELh^;?HhwO>4&OG$fW-)OK=&;G_ahquFq4esx;Xzhc;
zgJDXgy^zil8WPgBsA^~j)$h9~*`G5Af3b#W`*MU5Fn~8LDDraKeZ7(mA-Z|?S|_lC
z?ckQ>GOY24kRL}oqTTw*&K^MatqZ<-sUT9jhFD2Hiki3Teip}sy-Y{*xVV$JCz?Pc
zffT*SeXmK^9vag0L1;zqt$-P&0N-YoRdaN`T>Ai7Hox_oUnf8AQTCKTh8&?)`0GUo
z@hL5v_D3l?UhOWKZ0qPTcO@pTb_bFnc{VErywFcKAxi6lKhtM(Ukh5wHqO5PZh*0`
zN^S!UseL1LO_h99pvxyhx`_~d`!Veo05(K)aNIh)Q+QY1+|ga~*CqnhE#-U+Ybzk_
zlLHb&mZBP(k{01GrgQU=(Zi&XMear_A7=Z_k>z%_`SM!jG|_FeLGzriXBNk^eyZVW
z-p%i9cPW%};P<b&kIYx3VS~aihF^SI5O<h})pnFw|FUB?LEFIJJ?DO1<j#g)+%I$_
zXSIv-D{pRn<syi+W-}*^XRNH3<;Td2%Ps?}mMBlB(aW>p4L@Rg#-bm|Gx~M%DG=0R
zEY;&>zjbYU;uN><WFmW*w@~Q#7s`rn8ap&tb?tGQV@0($owK0(QR89X7ta-&v-5L0
zto?T4xDUpd)d%{S%d$A@XR|hGb;m6_0bER6u6NFc$5oC^Z{jsR_1)e3PIt3M3H_vr
z(Rxz5o>K5ch*-LFys1I>++*o5x4w}hMZ><#Z|=i($=<K2Px9+uNU&Pli3N=gyA~)<
z%0(!o#*f{nR^x23E`-e1C_Rs5oJP4l9q}%0IfiphL(l43dU|Vj9p`?SY6ENIlF)H;
z`oU>xu*9CTpRZx&7RybrDOi7Ac5(k-*vN<T+bh%#VXz*>PY=<g@Q%=A{mzy`e9uru
zUY;E?Qt^wt{f)>s7$C#-JNvLxy)lb7#&6w5Ilm75J?LV9wT{D2{)5P&pA6$3X+vbR
zxZ;k#56a-7B*kJ%D9bKRW->3veM|paw%lKiRAl>{c9mE2Y$H&f7p~$D)>Wg-T+_%(
zUsbProc%<>^L6s){a`bX&#}>u;%7~I@fWYmvWkY~qj`g}R4M*uoUy%OCsFjZ>(QPI
zuU*)D!2FxRi$^%#!n>-P_Rw-xwu6G53jsl`0x7Op00k1fZMPnp)8miGBv&Guwt<S3
znx=)E;keH3RxeN<^IXzHN0<=9>lxu2sw%kQUL|RUEMtTl>px?mXdc2Mm|M>kv+hxZ
z2~6}{jQ9GG9~)dV7{8=@iQ7|~R*&T&a~a8Z`E|qWd&;dCpU@0mhCt6}{>zMLr(;HV
zO;>gg-qI;r!GNb4VGqA)PLrG~{SjMt9P`%Goofa(?%{|u`*=n(*>0N4y6+Sh>mnVt
zY=?|fezu~*R`rJ+>57UCq`Z4g`=?g0ge*ph+O0~9QCc>rSQ?=BvE0Adq+1l%qEfs@
ztvX`-h&ySeesh5Pt?90+$n)*`Lz6rWX!*zmCf|{YlicvBD)Wr;%<d1=3y9HnfL}=>
zZb*Ouou!CTWLezguyDsRcRfm8FXR;EN3N26B5qWg6WTJgh}`4#Wvq?cbdP85nE*ZE
zm_!Gwew_94GVH!*X#08i<NFIH;@`>IO*gVq0$Wx8B>t7v-wl#fw4{5^0vXhk{Uv+H
zBO!^ILBS};IZtP8ItGu3@{{kQPHKqX3O$bm&VV98%Y1O@9KP<7@~ly(sz_$)boAFy
zTGsAo#(ifv=3F1yBryypHun7w0*BB;zDr2xs4GhCH<}E(5!NhEjg@z+0qt6+F9K(e
z9T2Zgi6n-?hm=h<`q_myzo6`zY*=@Ce6Whht(-VkjL*yn-~Mv1RTFFfeMdpk<~Gsc
zn`%N@^v-f7%%~rVvdJkm`2yNo_R$^zZ*vnWIHg45Yo#>ZS%2Yecq*aUgr;-q1j$ZW
zIb}?sK0$cWZw0isNAiS2L7DP-UpGhb5NmPXEm`mUo|-<L0~b^w7s%DNrI?iJ{s}b-
zGk4*=SF-<Ux6VupwywLh|EVF!39ntN496+O1=U*s3~hs$t~B<I>$AhJ;;3j*pKrg8
zxR}GIX-Get@(jH_$|59)@58L*I8W+}(x!I-;6et8!k^S^*ce$n=VWMI(QOY<6XhAA
z1I4ueCaj`GexBZa>02Zf^clk!kCGI<`FAv4w%|peT^>76QZ%Jm?Xf#Sye7tho_$j_
zVM7G*$f0oGy{Z{V`A`DJG4-ACVeP(0-A|DK)N65|fW3}V{aQJQS$ZjrI4M}5q$`Au
zu6KVur+yP88rcl-FNNj5&hDaH{t3QeD5t8-wg6ydJ5U^!LO;_!@^?68JOpp#AuYP=
z2r*84MdT$+`}6e-@0@+qkp@>CpmS!hDrLxQHumvg&Pk!t1~_IAf0FIIVj~{2^6U~b
zp&klBeJ|z4uKF|ziGu#G9P{`c40q5T>$Gsa=d?Z2?nq*RM>Ua*guhk{QxK2@f}hZ+
zv#wI`bp?rFi$=1`;203L2>Hm!U!k=os&U~6m-8d|L$uyV2KQ`WiarYroLEPV{a{9a
z1`;;DmXG=O={i-cEjHU#+CrFN@pV8M;B#)_((}v57F?Do#<87Xq`zpDTGp~hm*Vbe
z0FyoK-3Dn<he~2Q2@dc80!42LknHW%gr`K{lf1h9e>Pt2ZBOnAIFd*LL>Ly*-wk2i
z7NR~VcAII4k`FH4cn=_QT6ago;=AQHbIc~{5J93EsDC)l?X1vYWH)5<ZC4}>1wbOF
z*!Iv{0l2I+Bc;oOKUYFRe%GuTJ)k0$8i_aly%+#-L^6g83d>(#f1$iR0oWn_&<I+E
zWD;T9mz@#|?Rp5OUgl~fEV8y!>B-@%Zz34S=7PhF_SwtVymVt$XRm{z$*JGSfoqBn
zDGmmg^cf?Xj>@MA-uF=K4>I{1^_@o83KnfFBkOcyToDwcBsNN^zp6;&V`w!0y(WJs
z{093U2W<y?a&-<#_lVWB?M4_~Bk#!_0unLsodfV2DFKpsC=L}#>85o4?S(eA9&dnK
zJt%y|r8#l$&oAD;3l?iuvE!9EkpQ8W_S0_y(;>Cg2}qRdw_A*^-zB$jxd%u@zUTts
z`ALcAZW!MZ#u>mG@W^B7)I)5$o>+kzqWP8-@BbE>{P!#YT|V#>7e4h^W-B8v+12+a
zpgW+SwL)$_iJJ1v*()L8tGFAAl>e`93&89PpSp~)-~Mmm)BoiIzU>1pe#w5t8_fUr
z5gC8(CoU4~PHwAi`u}gT%Z3|ZJ8xexy}$ckzSjSG{(?UWlB*w}sq*;mWdh#iG8RdO
z$2Uv^(B(@WoCliV$=0EkGua`&kg!uD9L`LGwf-0PO7}T(V!-pw?J2NL|Mg<~>kkc>
z1a1Maswn4UyUnx*H&^?DH+}Nl<L^0O93%p`3|=<N9kt*a3^7;qaaeuNj(t34vaLtW
z7LPg)NMZ*HD+wem;pF_bY!DHP)b>s^06NrWZiu?NvgC^?aG(D`tN7w%%rQUn{|>JI
zZxe02wCR5Kypy9+;9reyU@@CmHSi|KEv$@bk>~8^YR+Rybyg%OS;|xRt=e=MbJdIo
z#Sc}?LjlOe>?Dx;DUd28p>@0aeWs64fv?$PnQY<H!p&**9Dx)$`_xe{9h*a4Qn*f&
zo6k>B@dqxVgX$S5yfJQ&Ygji_2?9{Cn|KXnOvWf|4M5SbEHQsjc$%cT<Y<8UksDXc
zc-v#E)h%#jVV!gct9=VSK=Q$BKw~%Tv7Fyt;OBWg3OyyNf6=i2k3~pwid<bgA)o&O
zM+WjG>Tnmx#lsD3fe;3-m{J|m`O}vkjTadulD%-X{~FD9(UZkgFxYiM<R!RQ)0$#G
zBD42BH`NwYAglkNlR^4$PDAW<V(Qa~AN@jyZG=9}3jk*<xPyK3V|ee>Fg^0k;*0Yl
zYmZ*Q0G`(Z)s{U#Xbrtu#PfX4!Hi)){R{a=>C7&Bwd=EKH^;dL)gaee>HO|?UaaY{
z!s-?{25mrTs3wWa^j#Mh{L%}Z@{@JrMNSgHsb2BYT)nq#XuafGehT}apYQQsU_qaV
z(n_QL>y~Ud-6g4+1Hw4HmePgMcM>;%fLtR%q~3<C&#{aX)q|`<zT3&TDl!gKk>K*%
zrS?KHok?7JewTzq&{gCqRVNT@*}r8|)2pgjG<Kf*;^@S|Xcc8->pTHnB1{A4$6vJ?
zRYkRcbTs$aypUa;4AG7eU{I@{tv^r3aULSu@;halg2%b=<mvzXnBr%_!D(&V`!-wZ
zUmMH+^G_s}d9UgAM$h`7lYt(XKDBo~O;5TMUX(utR~UmzyRTqr{6p-UBQHfSG}x5y
zv9ZHf5#^c3!1RkXG_`;?{CFh0Uz)yJlKVeTED;qU1a?tmJ=O4^nG%Yhi@qTxQqyTK
zN!j&%lYnV!P16bqw;+_|n|-JxH~gEmF&whglE$AJ0to{(M-#SFgY6n>UfUJv03j6q
z5W-r%&ci<HiyGqIF3$6Q2(C-5q|l%JS>*V_x#v1yfq^N_9WGN1<T*cdZs2iN0M63g
zzIPYJ&}=o-D;v^OO7@)m1}?N@-X<l~Vfzj7)3c8m66;25&}eZm8$B>$wE9T>yaFMC
zYxr>ugSeM0VBMWc3PFpq?ptQtT}I&}Vpz(z<I?6cTV%-FL>~r#T)n)qD1*Eg&bYSR
zjMi)6nYCUcdI-J`7o0>y;V+`DP-+#xpD-6*)9#+i)pgo@<67*z38m9x@<?9S>eCxU
zwaj4;O(V>uYn<OowvR4?999$+PI8a%<A<ogE8PZDK~iNkAy(v|ROp!qv*JulLB5E&
zlRTOl9NlC@j<R!=aX>H%6sNP>mp)dzKkU}E=^V6*`IT?8(wn_{Ps(s3wmB_^am3g(
zY?O3i-8I#e7fu6(3a+cf@~c>y>-#hT4P4zRC5;vQj^r;i?57&z9PDopyR(XKi+gft
zbc`bi9oczTT6A^}qDuB17Z7$_CO4Q$9S&v(Cac(_e4Xkqn7=j~cBFOp44<ogsQFc>
znV_>&$}?PZC|CR>WR>7%M_^rIV5NM0xf3Ae(a)H7$%|6hrkE&~L@D=(`H#}JC{&I^
zRVJ9`&dZK%xoVf}xq8F-ZfbTbDrc_wM0O}UsZA_ggxC2}t&>AqR!Rtd-7w*%N<rZP
zLgY!^qH4?Ff){r`6lAtHA02Vv6Hg?@u!c<3v@c?s8#O6EX2PZT55x2C7L3*!*tMkk
zBh~MOe;gFxfV15>0B*z&<Xj*if9`()4$i&oIBi>SL`P`=DA61^BgwD_b5dQm&>jGo
zkE>;p8UhL^r6or`!_#TEi1q0zX4oXPMVblmr2Z*LXJ8N#O=!CnU3H4wEILjZ5!Y!s
zy$XAzqL#k*;Rc<M2B73aL0NE=oUkt$hy;(VE>`25fS|2$27uaU0fAPDm){QH<bO2X
zoRmDtAB3J({Hd-m)%=Y8(MfI>dIcy_q_7|L046PV{-PnhAD9ni-tqTJ<3SPLtL7yQ
zYZ0cS^1KcqEI&SKArNXG)BR@S8jrKt-j04c@+6Kw_qij~$6jVPB12b&9iW3fxCivF
z_@TQ~@fg(MHImRnciJ7~XV<x`CFF)>zX+ijyqQ_JSyOm%^4<xM&1=`wJcU<UK}m7d
zaoUuwdhDHSEr3+_U4!q1*<91nFcu~R7z(Ph*3hG*H?+@*SsTyw)g){HF$&ixb^$4j
zd=8Nql<Gj5Gp~L?-fWF^SixB}16pyrHk1WpTvFi1ZVZ5Qj}uU)^TD)zmcbB?Wh6H<
zK)?oQE2?Kl0H|pJ$6*uP<1)9ENGUOQq=+gn)1tZtWV?Byca7opt|>*A&mnP|wn1))
z!_GM*UjLV-ojmB++pY!4EbcCJlUYr>{zqF~m7@w8XEh3=FDX+1=3NgS{fG^){Y5*8
zDpj={l%MOcb-X$KjDq{AlCub&sYJ^3B<t{UyQDfTy%_1`hdEm}oTxZ~{A_?wqCptH
z*wqDS0sNsScDitTQ(Xth`{F`>3?1m3nFYpql?cc&-ZV|BI04ZO&IzMKk_!-n-BULY
zI;6|;JI&RaFuj%rl6FRuND;Sy?2k@@F1HW6->*JC`CNHGnQ1ljOD>U%o&<(O^e@gG
zf*A!|<m<_H1Jjv#8USLbd463M?$Ngkpk>I}F86I_LHm6Au8#Af-&}bw%l7HKn{_~%
zu>&@4J*R1bvk6={!|F|{i_;q^fTSLMLub}<9`(@XK<+pZ4B?`Sw&A1V-DPpGy}4M&
z+n=<u*fd`a=5oj&%kB%0H%~Z<w){ehc$))~>cRmwlgPvO4bO8dBwx!}XUG`Fk^22Z
zF;KoHEab1|kCtB}^*$RLWtw1prXYL7=D>a-VL58<yI4<yxY=y%>0I=v&Vu!*;gn+k
zEiHG1UVt6cz546H)DWc3PgS;eSPDXMiC<~H0pG~4^Z_A`=A`81Hy_;l*9nulEZOtn
zR4L_h=@V)iQA-pmJ|C8zN<b%<wLPyLoOa4Oe_h@?76CJ9R&=w@K;M4=EScx}3-A1=
zf+E<5=U0`>E)z1fJ)K_Zl&e9F^t;5H$za6G>;qY*MV@(QgKDs?@1;D{s+9Iu=zYb6
z?(e!0uyWXnCqjN%S<%m<;cN}2G+0?7m3`M_$NSV1--X{ij4MV)i~sE<w}y%k^-lih
ze?-$BID-YmX&oKqB3{!+Hev`}qPhZn{NxLngFZzoLukvhtJ!toxQp&^|A(&PnSCV@
zQ5Q0u7GQ%|)F|@p9j%Y2?+7kjD5Rt@meg|`MJ+Birj_;~URn1y`Bf~RHJ&y6D7}Dm
zIZYYG-BXX~B%?NhiqRgS`6i@&>#n!ahFxwIn9O4ovE=b;_JXMsUQ`QU6pOcrHF<nd
zc@INE8dYDtjTj7!_b{BVp`QVnA#u&F%`fX)ON&yEV&@MSv*ySUsOt6U`cJT|D<eVl
zerx9J=-#a8jF#{vg%=yk=0?PA{SJ$iCCpL}dr~*F85N^Hr+O%4vw@6_IU1iTU3>#^
ztSyEvE{&lp7vM`o^>EbE)jB7^j7pa+1>ZH{wjo9_X=*CrMV(;9`p(qR#*fp!P{~YN
zSsK&uR1YKXBInYvBQOG|rmgkWmq=!o3%uQySJIJ~y+WC!Cdup^k(y@YW9`7~ZMP<T
z;`#teRha<j|BT~_$}4TTj}H<8pJoimb1n6{oG%9CcEog!5?xahK9&T1PL+dLJ0B}(
z<BuI==0Bd1%YTkfjAj;<tVD;KQa5}A5B|tV9W?MDu3o-Kab60FWZugbR;08@AL6X^
zJL@8%ry&fL1l!)hFykepB<OW!B#qRiz$NgM8I~?Um_jr2Zn!Cgw*1)THwAa3p&>On
zBp~RShdLuAOBV`mz0xN(y_CU0(ELRV%#40&QP5{0^m#kDA$8XVfDh5jzEI?sTOoPt
zSsy@K87db6270IPq}v#VTt;FBBLv;RL@J3-igqq={3XtGv(33I&y|LA#L=Kf-S5Pk
zwzd{w5zQoB9-$8cUv;KYEW=@KRY8d!x85TADM2N&i>)$|9<(-zx>fIqF0tUfZaM|M
zN6?4E6Jv>kg`Xr&>b``Ds9%_kE%>%)L;aw!!f`gvwX3m5*UScw_w#~iU&{<R4Qqe*
zoSp{<#NNzb`vTWE%c((&G=$!lH)NiKK2l&ffF*s{rxrI8vEp~0Qa_HbE3K#b?Qn!?
zUkA(e;ZrgS5)q&qEqP?TNE)J_2AzutZV!GB3_A+KN(ow?BljMTH$Np(Yu`qmMCLx1
zIOVRyoK6XHq9#O0&|;kJ2A55?aW=B%-ohYBS;n=2JduxWQc_)Jlv8cvO9z`W`;1!o
z%tgF|<hWEA+AXW+i2A{}q-=h2yANxmhi552B|?3&U)}DvL5F@HDfbFRNj=`>v~Id~
z)RDZT<(tF{dF?@L_hL=8-m}!}(E_{Nq#6(1bKCi}03(8<#EgzoLyupsTCY5OD|Z5(
zX795sx`kkp(h$yFTs&MR^l;wb{5aFsuv9$k_$b_`tlM`9Dr87r1ra^p$2rYpPZODb
zw0wVeAIU%w*%~~m401LE9G2i)?Y?*!H)A;i_nZ>!>%Z)NGfi~4Q!%n!{EzN<CGQ>R
zOGdhXBIThknDOoT?`?wAli^s{OlI*hUVRU!apedR?GRlXYWnk#1zz)ek?JdfZsQKn
z1&=d{z!Lm>Pn<&piPVwvOeBqQnpqE$+N;lZyO`+C{m6W=jW1dkmPt|XM>lxh-bAW@
zKe}N$r+(1YW3C;`<Q%znGNGdr+bKkR!qP^~MU59Ij;tdEN@UiEs|7{06h3WlkX;{>
zz}E^R<1Jnp#LSTPrNKy)Uhbj^XR6a#x##_=EfM3y3QQyj&LbdxgNB+MIQcb9LU2Z9
z3hRBB$-CeBu?6*PL<+`l%Pd~<w@Hwo>M{+@0fW~(12L6F7NtB87;k<@K17r`2i2g>
z8)wtKhH1gM+|NG^Cf#c`r0jiQap3JveVML6VDZ#XAkfLVx{Jn$YEVY?J%1kSh$z$J
z`SA#%CC-ZZ)oo7too8}yT`)Fn4B1HzjqzjNNf|c`qNf+*3nn~jIH$Qm08<i(oFvr6
zH2!{;vw-8~T2xb`e|490bcqK`gB$B6*jWUA%57@&v5nQbDaiz3`m{-XkV<&<`kSJ3
zV*myXNkMJe!@vB_!OR^@&ork5SkjI){+lD*k@Y9<jgvzKw+PvZuQxY_GncCo7XWgo
z>MGT*uiow<1n+oR8xC?}3y>!fqHGVG3_8P3_RaC-sRAuxdj)xWG$0Z&jMh4YsGgNc
zV1QYkSXL`Z%M0IZQx@wa>5vNTChBQrf>qItLAB&$#<t$py7qcn%lXkzn~2Ko6qB`G
z&i<|P{eq1_d`22aIRU<d<%exvGTIV-S0VZdCfbBO$xhf<hg`0G8jx=Le*LTx<b;-S
ztbP3Qk#JNgSNAyS`!{eHV+R*hw}YfB{6inyU3vskJ9t)qK+(aZ9XH3CFVA-R6}=)q
z>2jF?bqa&30P_YufZlUI2kU=|ZoOc%RLwdXswMaTUfV;X7^U$9!qIcE_LC46n7k2f
zaDB-1t7Maxq>+STkBF|n-t}|NQO7sD^Q<<Hoam+$o5%?^p1|t%F8;FvccG&F%3Z0j
znh$C`2E)bkeriP7>aW@#r&#*Le~@FEd$J=f&=X8D0;>tOz$#v`^!yU>@Cl_Um+XO*
za2E}h{&-tH(?D5-G!(4NCaN{rd8~=R1&gHD*4+0)-r7h};(vSc%Fi@b@TiaTdPzv=
zYsiS2VsjdgYQNqW-+NFZsSrmDp`fdkD8+%$X_yIf%$*$M>hqq&!1`*6%Z^t+^2$+S
z^+_s=x5y;beSf9QFjxy6A5*vP%bdjCQ4`~Kjqu;B`&Tb1;m6$SU}cFYHYGXzy|Q$>
zaTf(tU`xC7zMdeYsOKYI)Jnv<%jt`&9(G2?E<}BU%r-H_TPAGZ<LPc>qYPwEWWP(G
z@GexUrru_fr==0^=ZI>o_)~BE<i-p^jGM<r;f3&t?X>C=Om7h;JENeHviGq@ro7fk
zdDZo%iy<FXC8Aql#3x4&m~OK9=M_g5rBG#Tgcp<8aUWQ}IO?Om?4x2cfVS79t>*Dw
zwyg-hZ>w*59_zBkoj&W*QHn2>YA}B(okM}@>2A{6qFuG`xHtt?X1ut*Oi7}_K<^m}
zlU#?^TgxS&oH((QElkU8=*$XwCX4!0on;>!gWCU>V-Ru&^Jcn4L7Do`V)CMt{Cuwk
zjGJQ-3*^>oL2cS)p0UyY9(>P*sQnfUr!soCY1Ryxafs{$D&S|F;frV*i1IJSn$Zu5
zZYFsc_SCwbXcs<ir@$&w_in0ZN-F!PIw;1picL+<ByCs=8qEe+Gidh>xbg0c0iQLl
zUfEH8w$w8A_x^IY$f@^3*w>w|vJ|UnqW-`S_%CuWO_pr*+aowCu2p6mfE(5rN@1zm
zr%$OOggm$O*m2oO_n#1r=X>A}mwIRo<NB-7w~gp<2RO+0;Coh!ooDnkxA1Km@^1yV
zUvrsfAeRb{x5z77ycfrBX6st*;FyX;d1Z3aWMV>o+@y#uMk2%f+Is5b%9?Eg#Hxpg
zw1}ckRP3$fHT<~Fr87<PDEV~Ok63$SM&vGL;Ry9)&I|U&(RpXyC=WRyN>tClEWrn=
zZ>*19)CO>H2`Kg{4aBRU!EP9in~cTv@ScMj5n!=r21iYxVBU_A4IOJ~-z&-wS_cy!
zJ)ZU^tqc4)gwz&#Y}Gxq2`s4@`&|)7umbAdW;{$1w-MN&2XMeV>IM=kcn!#X4gOg(
z$Vn6eg&n{JYQYL!wyULHyOt>H<1E4S8|3u&<AakQ{gUF?r8*ZM>o`{qVa(7f>*rf0
zS@0|1pKnMP4^VLzVak_3@Nfr*^N|7RGSh--i`oE5&ly_FP>Eq}{gGFp^D65<X7WEW
z%ixD0C@O8b^y#22|3BxDmzDHSBAjH){dJ47fhm=*1FxV1L~suoiy1Z(v8N_jfdfQ%
zbJ}jZF(1TcPRA4Ce<BB1em^^`lrVQX9+na$6v_6V0{&czI)$yV){_$`x8e-Z_|&QQ
zfG=mUVw&){x4;bASVA@)GC~oYaU2R*bqB))>}k*KjJ<M>>w_L7Gd~m{MDSNgzK7?M
z!$?*w?&Pd=$`CMF7ATs;swZ1+ZiCXrX!$u=162u(xv=6MgUPQq0-+s5VkzqNcC_E@
zVc%jN-vAi-@rg`BXNu*M^&`t)^vRa&WKGJB!eGqU;25gs04lZ7S#Zs_{^HV>`(XlH
zvJ?(drdVbJu<s7AXDad+NkFBQY`MK9_v+p25P_$=p!K^33?I#2Npma*(OMmd9y=3A
zI;V*2v#Hul_1R1`;$jfx3Uc;qGYluQ!)_6eq!L9=KK|mQaM0-A1zDB>-_(Iu5(7+W
z(f641$?L-b)5Qs9>7<T03@L2P$rEZD<zS2(Aj(R?lw6)y9=zBxoH2A$F@7;N#WR8U
zoPQi@NxCQZX;m)yw_N3WxMJ$cCTEn5pE+S<4T!_f2ACw!Zni9A|6u`?j2RH@qZ6!e
zn!n@x_01#|%DaC)svFA*5nPE7N1Pi{V|J2Q{ZhG(*Z0<~Zbl<RPBrXFaA+J`2N?yK
zeQTAsO8cEJ<RiA^@|OB8OU@brG1>#s%i2`W>GetIV>|zDCo1(oo4{UHlmf#XD>ZJ9
zX^S$i>+#0BdQ-Z34PVaYG|rb2A0lR<F6)vmTQp8<1_Ud+=6wt);LKT_y*^vmmn$R+
zx$w-b)Tnr`F}(h0nD!>_^|A@@UL#j^nRu`|8s`$*x89|bNLhrqH~~o^a<KKxRJg|L
zvlbc7j+aZtCaLPL(p&&ISa61N`izFmsz&d*_SWG}#h@cl>6UO3#Bui43d<23-DIe(
zB~19<T_<f7BX<dN*``T=shc{U8W7&5*yn%Gc|EVtz>=Yb<T5m-#B<M%aS6b|eu}`g
zj6O^Lu{yU_cgKKnp_i2GBQFXXc3!-7=TW?gFn+ky`hQkw{%uq3%8Pn8ofhlFDEvuG
zfBS+-^<1yn<NRPTtlf=>NKIOF)?e+fw?udAk>OYNJx>fPd@}0f&8~!d7|0>FZjm%0
zhc(nhH^KT-p%QhjMEZ-7QoeQ|+b%ex-xI=1tZqn<A?xtcVwkBJqW@K|)t}!NP23+_
zreB;}@s&~FEI!o^$oY(WU%DCZ_F*x|u-1TNS=ZFB!U3idK2)!9a&Pwe_N?E7FtSsu
zGh>&6>E&#<KnH^2Uq3`YxR&+=SypVE1K;vC??I2tTY5`o`2%3E;Y^{#F#&wOlBk{-
zd;p^8z-sZ*-7hazLbI*LxTw9Fc$)Rvy5K(22mBqMTM{%OK7S;1Z8d*LXFWffn!6j%
z&26Hw{R7O&84d9#7)$b_T41H-k|0U455&OVRv{dN%i%+)&X-d<ku0l$!f*WZoNI(t
zd=)hsr{dn!OQl7w4L={-cyddT;=g)gY;M%7lX>#CXy;^p`*pJpfn~-#JjZWvrZ|`h
z-GE6%?Pbjs2m=LYPRC59nC#qa_Jyx@qUgn}JNTrRPWm`QHaLBLhSG?wnT0{P4m0ns
zsfJF~{Ku)gI`D6&F1TQcO2)NmjP3I$nl#0EJOm@g$sAbU_mD*wkQ<BT>|MNZ<6jzw
zUSIMz+8tigAl|SF5sUB$Kmt3{?fyAe|72wURci9$q0_zfjrcoD3;u1x9~H1Ec=}#T
z(1r##f_B-l2c<^%dgU<m<@DnDk+`DwQ`$pq?B9(7-O9Ni=q=A2CvLRGu@A;JN|af2
zz%)-Rs?Z8b`;DuZzxIFiUpQqLh_!<~^Rt%HxSU@4W-?#M>>*AncO2QLFF#=sXB1)@
z$(-y&l4{A(Zzk9ND?-n_!AyfDpsPPWaXoU@$_`|mLf1gs!<eOQYem0sInJp?MZb!J
ztd4Vn5upy6^HZ(2tGe5Qgx$|FS?hQRUTuNz0n9SV3`4n}Mo}2^y(!v6%l7^yr8NWO
zbU|6`O^vpzmy4}#)6zOom?d6hE|+ul%Zeyd@*cjhb`?+c0ix?n3dhOjvqANl(2DGR
zP&Rq%kR-aqTNdg$F?A|j!xp9ulGV)xdj<qHY!LQE{Q!-!B<Gg6^zrIr%Sz!bf1OpY
z4A?O@mS@6u!xI;Y%Amei+clBZT934FD=>CuST(+*b)LdPJu7#aCXik~sF)EAo4m7X
zOp3o(k-sl+L{Q;MlJIDs<7v7u{2#afevApo?5yyw6@?p4(^8$Bn5(QuRdbX8KXA-X
zUA<&Fhl5j>H`t7@uk1XFf0}CnoQf`ym})dvWf~Ht)N3e@A`{UM;gSFGod3Pj_>uK4
zqDANLBLb-dAbT(gVsFF}G<GtaI-#AuY2%&zfk`Yp0{|1aWB$TIfwkum2Ou-J-}Md*
zUrrTb8PAtEwiaGV7~iAQzf(Ca=4&akePsHprze9^f3f6wtllulp%*ju8l;ARGem|i
z90UEvxPJhv&7yZX_!47R4|L=n?dGd{i)>97&Pu?ttpfD<i3a=N%I!;umaW_Ejymp#
z6131BJCKfcBlu#uu0a&$_nK&-e@l5>$Q_W+hJo5-$K~lgPh*eK&i)MhDwurBtRmq|
zj8VQ97OGIa4sy)kJ(>)Iw<Apc7l?|#(qaQLdLm8q>Vdr@_{DghS*(}vtcKd^e7EWL
zh8yz+Bp{ZJ0@G%Qa)&_)ljN*NXz&V(Z<E?c%V~f!+cmnsM&bFm%W8`T0zKC=RB@%5
znHD~Q9#rm5H;D3Uc7|izI0yWH)V+07lwJEae5;@!NQi_=2q;Q-#~>*pAShju1Jd0f
zNT-O>p@gWUfI~BMcQXSFoze{he0#j>37_wM-rx8A@xE)hmVhkqx~_fgeV)g81SX~W
zxi7m{xaRpD-T7p-hiov;>x_v+69}YPjfA?nA!x^zg)S0<=^s?W?imHcW*!MzB46qc
zn}RN`J7RlKR{EoZ>s&%wbgt&-?cCeMXmB{$->BX-n{-xn6pSz3sY&>*0omV~R05L@
zJ=^E)j_sHe>Ft-FC(=zUQ#8kGK@raw0(f7B?8ycVK;zWM*8bXE0^0v=cR?CX_LO(g
zS#!ygj)XgSx&&9IBN{=1ASuQQVe2z<H%Oz@LKT3p-Xqe3(0IZ|dXr8x0O{$A&u%B&
z>l-|JC{oq^%b8FGIp_^4)V&x>|2V6j(iMW!QFh=7cZ7cJV%kxiU1V?&_9e1h;@^_J
z_Qw+Z=viKk#lABc4v?S(HA+;hzPB4q3kZ2my9qK3e@ySxUwkaWJ{MwvrDLAXvhH95
zAlE&zF`_qj^XN83wh*u{(N<n$AQ=}gJqEx{-0M^Z)xkx`VL6jCSt_UJJhoG{MV@3$
z*Dqj`Z4=RFvc6E^!y<2!{^WX@(FRQd5<$enP8?Dl4{6_h7}uEYxcb_aq)ry$C16iT
z{Mm|oVeFag-~|>y#2PTGeRV-B!W8%H0lCvfd9tQQkmXikxAbJ#wK(#;riK>@QTvA4
z$7)sd{u$ndHw4Xk@34<bPQ^Qd5o~r(1HdvgqObI4vwf=wz`<W=B!{-sor&yp&dx%k
zS;t^<?4Z#*4%OIXK`1M(^uv<(VsE!`iyYZ@d(@wZr#uTRan_a+d~hFcb|}|{_;R#~
zmi{v->Weo0)32vCz%XTVv5uW`VCBL?=<rFGLy0VcneK8U<<Ly~0F<GFxc}j1P*c4v
zRiwE|8(7~ilQj&i`Voio%Pww&+u@LiIsUbTJ7s0SOMvSHxA6d{C?@@Yh~^^qq>)(3
zC9Lxjx%SrRGu$n8gur9UV4kIpNT8;oNflBQI_B`d1e12%1^weao5e`rkBACCKG>jP
zEgiQ%Yq{Je@<wbe(k0x*4h*{X)~e~~v0dj<ZqMgMT@0PdXdM`-g=kcMxBB!<ljQQM
zBc_8{V`2{j9ctfvZ3UIlIwnED-R6aE7q1az=GpIiXy93-TN|+3Bd29h=vDw&)@63<
zPd!FwH(W6A$qU>{k)6b+pG1r_g6VrWZ2($0U3VYn5i#sC+!o&;riBhcS}(`K!QEqA
z^(3i>E3h`{n^Ckh8DobGv6Uar@`?FKiFtd9<J`0#9^0mo#&(O0|Fu03??$AwlMyAh
zjgnrniKpS&snizHP5C88=tuxw0HL5_mES3IM85U7I)Gr|Jxz_rk5fz(z2ydG?df4%
zUq`0uUY|NYM8Bq4WzcAR(Fa*3x>RHvNyOIZTR0CAkax8SXfM-60`V9Uucn`jebk-w
z<8K!kVf6IWWI_UB0REVqAw`Bp0=`c|6`a9_Wt;x45mJvVq<+WJrDVbxc~oTmeMrp&
zPaybKLwzggK{Xh}as1$bWif(xL!$m&jMe_=@EdFf9c7jfJYRBcUW1_s?Chrmxvwz-
z-###rB<&JhOvcKy^$)@hOPF?OB$jnO(=f&HrqryBWH(i2Kg6cia5ReM(bK2J*Pfs|
zZhc9jQC?e5O$OEIY)rp7bEUm0{p=E<$P_ktmkv)o7mc=qvN}t6@w|hXx^2DRl{H`c
z!h0LQ=|{hggp)c<8%$ea1sj_pex$h^bdFgQz;Dusv}dMnh9<0QZhF)xDx&q-XC)1r
z!n(_PkAh6tCW&+Vde3I}wgLPv<8`th0kWcE&43+D7T-e=%W?(5(d+R%uCj^Ktpjgp
zB}MRHgH2ilYD*yJ@ZLsCteb|+NBpg-mYuiRl(rdpwBmvH*W!eNg(uz6UkY+-7?h|V
zg%#1};%^YI-QJ%TqKw&46G_w1NV03bnE?#{-XAkImQ|&JF}D@Z(Ok?)g;cnLXRz_?
z#*1KPn>4wUKy)De>3uTB1}6bi&2r1#Q=HRB#9j(h(FgI_C@V*FThij;jzN+m;;zw6
zL(${lUX%zhBAVo)ImIBZ<US>ik*%*-afw^#lTC!}niZRgjvvDa8)>2~5;OGp$89Cg
z3Hmp7^8LkP=)E?9BidX5epNOO+1vbJfx<^m#K^&6x&*dS_IY3LwDM~X*N+|Bi^25t
zWM8jmIJ!*)s=N5}P0aNiT1A=ZOX|{nz$JS@oW?N{9z|S&H%a5a7jCdCC?{VEsvGx?
zWHK?EN~Q|RPAh}vO1^XSjz_eGR!xkzTT30-#ookKd6`lOANu0PTHoY%%O+;xhUgju
zka|KIDI>fetcNBZkU*(prMoO7#<s%^Xd`8!K$Di)7G+wWOz&yq2Cxlz@P_QHu*8+>
z60^^!HWM~_;>gV!QAj|JIcvCyE^|8e&zqCbvS9^z7MO~dMAB^6x^|CZ;zEx$H0CaG
z>h+Uyg8KJn4Iy{SV#7h|<|#k5;;;LqiI`VdJ`=^0?=1ePl{e|G=k66g4a9i?fPr1Z
zTVyj^NZ`g)H5Dcuu_=cPz22u>DAnLc5@zFnKT_3F`*|a~#&m5|fHN8LErA7gL*TU?
zgH}8|)<!wkR3l}F2;Sk%u%<R>1Fv5o7<e`Z$yx?-grNPa?mVU|unTgHJeh8wlefjG
z@H+?D^fXn~w#<h{A%raw*2F9!7M<~&lL*?qSFV@vLV}Zzcb8`q3{~+T>Lxa9>$Pgj
z%C1~7t)pF+)FAUAh}_B}*3_uyQW1Hgm2^kZKQS0xm?~RU2!{Dy^5)l-xUqfa45~Ea
zDTN$AFJ6W;s+HiexIVp%Zl$;4>7uo5(xP<B61$wh3-PzX$#6|w1xuNAt?w;O*f!1(
z4&fDqlUW=~;sh8@68nYLzT*gn&k)3ZwA#UEVZ4&gkve<&;+wFZ`qpBYKu{CVfd0rk
z?PDYEz^J5&>2M}d14@wUuTxPXJX!=&ZN5qZbdg~0aoivpsp+MV;dX}=E1`;i%dx*3
zcLzjPw-l=j$-eTxCyF51lhqB}Z=GQ!JFDN19%_%y?!iYs{4_7xMVOOpp5J$nu>?*p
zo4sRjddcNPZ)H!4sNqGKupe_{T`Ykde+Fz51iWCku7tnYps~=dPF1A)4o3qa9tc$}
z%Sa<9Vn;?6z|W5Lh`L?l=85j!d4NBQ<uw<|oWIocx`cIGMmU@O<F)K^x*2iGh_YlJ
zV)-ENq+P}<eg<X!`>8|C1Ka93`jh3##CysXdN~0WB?S*XFKyHc4Ej{r%w4(F#?k2l
z-ax+S@Z#vzs<p36_IFyZc^*y6<+1e$#DjSH9{p4?Zk*1;>1!y)LZ%8pt2=Z_-EstQ
z^8Tz@zXbm>IqwFciTOK|=$%{JZ^0N~-h3`TTTH^;Qi%#-|DArP`z4!FC7oU~Kdf!O
zlI-Y3epr)+H&%<lW1ps|{3X01oeV3wMhq+CNDyNF_DVAoixu}L?(^oTjUSp2J|;sk
zqV0O(X}<Y|AU99hMd9$jhUWh`R2M%XiuX?+EkMux#rLUT^aaewiG-!rgr&AI2|w-&
zp~X#hPzi4T({2{*L;5D}zUbnn<IuN=dKkv;HicD8fW%jTg#M_Iq2B=}Fg05$Z3H#E
zR|>?Nsv=QW%oAE`6PgI1!`aEWmy=oT4+)RHtwpG$g?YRUWEBxa&U7X;@j)f9&y!gf
zgp2n{2v>TnIC~SyzRuZ@^(V6ysgD;oNkZ{NCTSEKBe<CXI3#?Q43RFpOTBHPmYB?H
zDumqcO~|^}Q<qyNpuKCyaFt#opt#A{rG?7n-AWFfcxMBdO)@Lfd(SB!&~FHnNW4Q^
zDz^o3%9GrKXt&xkj56v%zlW&&Cj3$1jh=0$iJ=6By)0DhNnVxYoC&7f^Q&EriU6xj
zW}On$K6HZF4jEJwRBeaa#KGPybzb~rgI7XQ92l!fzV(Q415`va)Ly&F&|x9UwQ%=2
zqv6S9R*gzoYPUv6{fVL94Je+))&oi_WMe>g<8|vFqu<VISPunBMvwU{hVelb>P|hD
zWF#MPPX?}QGVYM4*z8%GCyz=b2BfMJTYpfBEBRQ@)TwnWwmA!?pI2ahm+)@miS#Yg
zvCMbY2wba)#w#YCGt-^@3EZouHKK{%$)FO?Q5*Rnu&4&|u^RZWbwgD0DBX^q9CI8O
zX(i(pdeOWig-sr0BzP&PQCn=;l%THJjR>FlFNO-KWQVKn-ZP~gPB3AH(pjNUoB8p3
zHT?pyu?$9#(~nlf8qUIFK5-MBWv*-AN6?iLKG1p#XgxNn+s$U#<7Seuz1!>u)CUwV
zvvQ<M<W{0soPDBgL&6>dq~<Lla$z*;8`b=m$rHB)8R=o!>LbNKo6Le~uMcpZ)#q7D
zZ+8fquQx;MFUx9c0{hSglQB-4qz7TSb(q2t00m+aC(UC~w>>NKziYTP*U2m<v{msU
zpxf2P84-Nj4#o>jb$zL}of$<seLLR$=Kdyq;7yBm!KaUu(MdhVS>Eyr;>@9M@%)l;
zWSz0bl*-3)BxVlDw5uv$h(DO6Hwq3R5vFXo1r8&xNsInj;Zag`ooQsbrtF^iwpXve
zPdCeJb*n|U#Y?p8(>QqBhJIR5H6E%RiN6r<e?n&V?|{jh$6&@|+P~rQ6q(vw>B9i-
zVuD3Wk5NT=bzI({-wcV0dFCvWw4lf$-uuK&OaM!Neo^yc@h66_t=zbu#kl&GWC-uS
zxyP|Otx=JC_%(`6vcUhBEt=FFeo&&?(?7@h8&2%y#<_#vu+jufYL+d9S!_bC%9Gkp
z#J#feS;$MfqgdyD9}kX^a!f?M!ZOWMg?^11{z^Fi$qDDti(w(Bhl<Z>@>FTsBmN}d
z{NrBo3TIFf0F>@#jJbdR^DfcF;FHmdyaM@SZ?EQf1L|%x5`g#<P2~EN)~ln}a0=gB
zjdDf(C=%TU1|I0)&NT1iFQ}q;tH61%BUKPyI5t0q!Vo4ojQ}Gaho)LpX2%<bu0MOj
z*Y;&4LxD-9=|S>uJA@0gU?<E$f_7m7Fxplzys9(;WSSbXy>zqlfdqu+xz}gff2R}u
zWMLm*W}*lAka@koiS&sQQ-E6K0>Z`pN|$6WYWU3#<fyZkf`hXhgsqP4Ee~Wfnltw=
z5Q%5Ho$igq(Y0vNr+)z5(F9<}nCNjlulPa&<SnuHYlW*oQe7E9_IvQmUDG@c*4K6-
z4mawq&^Vr+9M~M$m&jjMrCC#nVUp#>38&z!Ujq(`>mWhspl?9E54v?z%(D!W?VT8%
zJJWgR!*T0x^ftdj{s7h|gLU_CRpxzG?(d?MpVtN@uDy;&h6c>77?roh-9kA=N%i2u
z;B(nA>ylI8Avl5tPM)y=htFp(N57m_GO!GwjqlM;KrYkkg%f^@w>^SN7)IBCIpmR`
zTxCYj_vu{x8nv*k`ePyv^!YyPGLe*;$Fy=)cR_tLFwRAgNz$!Klk%IU+<&;=F}RO5
zM~rJj|KWE3b@==@AB1AwB;p&SL4U8ZU;n0mUYvh@4^ILp3l%QsF8%tY`CqCvKHyxu
zjwXGmZ1>ks{NMXP{QtBY@ZlyF9XT((R?`^P#h5a=M53JZ|6c#H6oOfHA>Z@4e(#bU
zEiR&D^eb$}7bhz4MHzRVRhUoQh}m70lUCD|{%{XRDq1dMjl1nGL*@4<P?Cs$!}2~m
zg3=fi=v9YCXngyf^_O}5DLg4Oi&p8IB~B9c?@F$J#5lpnX0+fbe1B^4fca1S3D4Ii
zoqb$=eL%~P=?Nau(&FU`v6x*O-{cY*GVjCbE9~`j{J~Q02kR5JZd!cQxubHjoSmkl
z5sYuh8yTZ0^rg{)cX+7GZgw1xy=n?%tkRZ(H+S&xou3o=#VUGFza&uc?pk9}q&QYh
z2j2GdsR2oV&MKi3kekPQG(PBW&)=*gtdUuiS)sjhaU@TZzUEyW+!${D68qFp6Yt(1
z%DR8O3KPs#*sndX`Qs|yNPI|~Vn05DOMHCS!c^t1G<+tJlje<X+e<7_j|&w?sBQ+m
z_(;4p1~Ot-K!PeM`*lsjkbC2s(+$iBQ>FbEbtgbs<^-C;ZytalJpl$P^8l#X0F+iC
ze$aMSf~ih<6Pb3d6)`U&zyDP-MpFVpBph03e5!<Jg6aGHO1|qyvuG0!Vn~00m<m6v
z!HWK&z&|nfDa9G_$@u-D8~^go+?ni)P{vQ2=^bUS)_h!?Vr4Mey8Tn)wU8db?uM4O
z(uHMWuib#lAf0pGXFPks82#xw1@Fsu!oWw=^H$Yubsf8av;!XAK5L5d<B1t21(q<Q
zFY&*8|0@E)3#;3p`{j>~O}qi?EACmU$BG;OHoJQKZ!>V@k7%5*@;V8PM~>}ZJ~_&!
z*%!`^=11Ct8z*R;v&QdId@ZrOBXleha6#zO5#A?j6CWPFUNQ6ld7))N)b6dcWVPD}
z(Sz}4BRgx<hoD<?(dekRAA5D-Qk98NE}desKwtIw*@-m>TC@jzb7yVC)&b1j1Agp!
zn=XJVyeCkn-^Vg===>dpaWGI6UgN>Uc9&yPalJO{u2_TLWce7D>V0gEg07V>rYeNM
zD`gPbN;D|-KH0V*nJEAqZl3l5U&Eh6ObByciHa2=Kkf4jtIDDFJ}jQlq8Qw8D&gTC
zRzNV*#fbQhX+%3A=jfE=uwf;|Q)@e?yRl8*o<I5fu>XhIp)TfV&Y#YI{>P^G!QkQ$
zmQR%Ep}(-_Imvhw1Z`7U7N_soqySV`*1eR)@B4p1x(3&`Z?oGD;G3cHOi?$_*-Phi
za>~LAgh#9jmn*W0T=wOJubwqW(QGUvJ1Dw<gV4iHi~b3~0Cj^NN%_Ig`(C$T_~XFJ
z&=ey}DFr;(L#rssVCT|NHL1$0xdMh%OMtWHc!kpT)vCEfUb<#;ISg?G&bx)CmaZFU
zT5E5R7ES4Qqx_s$17KAXnDYj31iqiUt>iLX#7T?=Q(rc6^t=@^PbD((1M}0;JGNkO
zTM4*YkNn`_2o-qJg-;i3zFGd3Nc{&+_m5Q>*hUyJFW_^h2i$+$wq8!(mwxb)?8PJl
zxzfE2K9`nccQeEqY|mpgqI9Bv2Y6iTlf0zp5CiOlQY;_rVy~Hy`&F$Qfw!$~F*)jW
z-%7n5rAu;#*-aMb3h#gWfnSFL^yXDia0vpuF#Bjf__>z!gSb+w(#L}CQW?v%0q>>?
zf}WRzGX7l{{?j9TlLVH3IWg14KhFInor_)9(ciG|B7-Yh%jMh^W*AZ&ReY#U(Zwf6
z&>fzZ;o6w})RN8>jR&Vnkh8f`!u@IYy`n{rF$a+opB`;AneS>-%gnlo`CY1IS=zkw
z4oQ*D9>^(w&D2Oo|B{J$lWy>A9zV*igSGHe<>rdRcyGi=$d#?dk1VA8Tp|*hXoqTs
z??uK~Ph80jrlU%{CN##yyk=X7;cwXAHW5-+dmJdv&A{U{f-_QIXG9Jh5<_jbIB2Ue
z6WVWG2w;>L!}NrnM$-*vtC<xY;3`=$ePldYU%X{xRw{D4z1hLD)guwF#}<EjdybYh
zLhE$KtsEExkN19*Jl)R(o3pLv4PUu&?zx-rr;)iQhb^mI{$VY|D=df6rM;hAV|TQ4
zT*dU#XwMtl?(6Z>dIGGuk+cV@RX%$#&@EZ*Sn5pLu8-AGi+y>k=5+=Bo~A_-jvcT4
zlVNQ0K$m2rJU>V~Q9%!pk{Xg=*&Yp&>=v86EqxLfTwPuoDBDoanBQByQ&r%PyJfE^
zCgD9}@{068LWGv`tmL+se~S-5MdVqPM8x{*16<EM-k<W9C`kBMU4g%0b8Gt=W!yPv
z;wm;y&o0uFeUmLguhoVt?{4-R_Q<-X0DGYRHvGb3N9Oy?m0^`)B@JqqNwQ6*#4SS8
zqdy6oct4Lbk84G%^X&W`4f(nq{$6kHK5;)itJwG9pdPbs$d=GGCmIB^Tb;5u;`xi1
zCtUt5kVaxc5_ptPvICb_|I3mBxvOUq6(1?*SXq6xs+xONy&(0)nX-twrjFa|c9zbS
zp~ctSaGh@ETL=b<kkHWRv<l0jL%sPWZr{m!3Z0|(d=yh6DTWg>FLBmzDzox>_w&G0
z?-{zamtI!ro7=L7PxCif&KhseG(#*XuN@K2Pg79#DCeiI(l!?SQUEG`Li}c`;B@8r
zz2A>`GY8Dd_Tt$xav7FeJZ=gVT%`WtJZhxv(nZQ&R%v~EE!D$5mo-v={{Zb!f!f&D
zF5SOuF*@_;$WN$8F?sXjGB$nbZjpSMfi{5?VT9Y@gwtyE`j?h82b&tcNf>GBL>Iib
z$~uY48wjqQw?eUFiG3;>>#OHy<1JDJC%Qds^wpC1^&G-He)!(6rdfUyH8i4G&e9nS
z(Om%^DdgucG-^pr--Z<F?fg`eg3UC+%e3^wTiE@AQB#-yl`v7uIM0@@gEIYYt{CJl
z|LyW+A5IQprqd#vIvQ;`J<!BvVMPiqP{1{ufJ;poIC94gZbpjGT{jA=g`lSit%1~q
zO?&VZ4B#e!USqB1bRVfAg0a+;G$V2W=MKyR?6LBCfI{@?HlJlY{0yjA=1Q&6AaG}V
zK2CofoH(NVBRUT6>jPsj%?J?qdz0?nUU$IvgJVD)JZB6YK%b1P6=)SMlw$Zsyu%+-
zvckwiUP~8Yvb7^c=!K?|*K#mikWvt<UP0wD-lxFXbi>;&a<fhWoP!t@YE7Qe7uzV&
z{Y@ks08n}B02SP2A?5*YJnPoL+2?a&5D_$^1(S$*I?wL2Wdi7YF5{5%Bc;+IHSKax
zcGa-7KYV@J?bDWY?Nb`8WT3DWDL$0M;i3UL_qXIGU_xoX#carkAdIxK2{{KUoh@(6
z0Xf}%83MsMB7QJqR4rtYw!U9d2h{ML81wNOvCkwL4Xzkm_h|MQT=%{cN$R!v2kE*g
zU@qr<yOxN*1(2#yr9)-<-*?u>^FxhsGXRybyHpN<CVM#rHQO5^8>8LA8$VXRBZRdg
z99R#`o`DHY>d8*p`HyuVYi?0%DrtMKrq!1BeORkTR3ETOR7-cW#2{yfMlRqNmx0a=
zNgHT<VmZl(;M+I?V;^R+QS<XEytxaU>#{Unb3oXI$~5)B6cHGe*@~#h?JmjSO^n_N
zfR?kKUQ;a|nBcDyu`|$bInm36+k(JbdwuVttkUc5gUYhynx+X&aG#F_NnjZ5<%O|3
znADmE>Gku#)|Z3dxRvPO0DeBvvIQt`m7u#DZ%Mg96~p#p$%awMcJLhNl!px<n#TH)
zr+uZ$*>eeeT((o<D8eG?HE?{zBMyQ@_}$SM(Hb=<nEPk(SWDi3fa!_HBh&t3>tH6&
zVwVUio{94lpYxtmP#WSt^rUHX6*cXofJX?=*3w6b%-kHHR@b)T#XxAs-+m*0lxR_+
zOF9O7JypQmP8`p%03^$z)BOhryC)?vn93^!jGk*!IP$mi&j3?C-xVcKxF~Sc<8wAE
z6GgqYvpx7ssT}n79bVvA$H>NZysjZT1~!a`ppls!lY26YyawnyG<)5QGdo|aycCGg
z!2L}=bpjN@Yp4T^a2+p?OxuzZa=F*#U=7J11I^fw{T!rbt`4{&Psy-emdLG8<afqp
zgZdbfdcI!@cN$O>&P#(^0fowRv_xb$goNSArX6eLkD|y8s$J5~QW}pRoaZTq;6=8D
z9}g9bgIhm^6T_SRAYJU1fu=E%;A!Pce7xn=2fsScdu6aJFG&6aJA?S}c(ErV%G<km
zp$8LHbvk{%Hg0ejq+h|T<+>G+&X~P&o3*4B-Pf#bNwk-m>ryiJrYR%k03n7jh(gH}
z6C?_9b$IsG1g#T$9S+>i5&)}d$ZUyWRqFLZGHascc3XM`pEf&Upr6O$RUbaL0!^N<
zWtUl{E-Z3b!)M40;d`m-T+wHg&3z-uE=%CSfEv<jWp-(xRH<ZUrQ`j??Ad<hSu^TB
zRZ^sM5VeoEbiC@YR$owp2`s@gzUJ9JqiBP}w4|o~>EUwd;=ribCVDLu+k9KoX^qsv
zJ+iFk=qPMFuS3c%n)QPWL)rFn(i>)sLXMt<SPHPz9Cs^Gr)*r&lE!{9tym7xtb;NP
zfgH2pkhp$t;NwA0LauSTw{16iQQHwj1I4Zfksv48XQ=0sUw-{Ur;USON=Gt!=xDWt
zK=Iq9z!})LqHtx>mBfxRcXbdUxIJ_vYf>P%C8uWxh|!40W~@p&*eazMLW3<IiU{ou
zKI>>FifYL<lfqQ2#$nZG#egCu3qL>IpeLD=0t+rJYQUeR56mrTd^*#T#P8BYz3cIp
zk7gB*txxLZ;Hon=4%<$Bk88-a^OV@`5;X&L$P@g<qk!WSFgomTt&c?~7G|?xuOxeL
zXn@gh$GV+?#5)y{!A0Er$xn$Fz%Vn?G{s3xKaGJc#PY(Zk7XY}mz{)h{mq_0k%Zu^
znWL$P+8Ggj)aI|};?N13++1~L#uo=$8{^K*{JJ;eA!K9uPyE%GxGkShkb-zVT^Ad5
z=^~zJP2>8pRx%Vl$(S@ybSL;g6oPMKyc38w&O`2}QjFC*<+-k2l}+OMR%{cs>(s$g
z9KNv%0~>j(f3uiAAY)kafMOlx4FT%&(8G`hspnY&OP7|7Tw54r)5N^ci{QA4d_ttW
zkH&O`Cn^YKxc8}NTB93c5$F;hc<^~g-zTJ{CNYp@VB0ZID(s3I!77TzCHB(Crdesn
zLY$qx?g527HNc|t2S>K_R0xho0HHu+1qFvgpp+jX73L_%)MiqHLa<MFXisf<>}(ln
zY;oK;Y0Yrny0LsM_^JPDW}LJ-eu9_>S}>Y~r-&%scP0U(YMZCZrlz;53?$kyG>%lr
z{Cz4CU{+%qdxxIn$+5F6s^i4Sud~RuM1DQNmOBw>bu(_7#@8LZ-%a>PeKSZ5z1dK-
zSTCLGtoj_uz^<zqSvvznl1$ZcC3wwj2*`#IV0k53zub|ZZTpVv){qfEy{zs^esg0<
zzl_iKV_M8q&*3q+J!FZd!MVY13aNt9(Owa!n|1Kppi7i30aE*8u`Vg^vudc~WBn$N
z?kY~J$SW}-8?rADOlwR^@Z0T%5;l-P&V?;qU*|S~TH(4;as9^R@<%P>uMQ$>-tH^8
zceZ>MKDJ&6^t{4hmro<VSxS>Cwcn6AXn8}KCZdp~$m&^PK4pVSUHh}Y@yx&b>JQo^
zwT8Zb4THq_>3Uh}^oc60c31aK>wBJ`ehTld{?W;0fIMVxmXTD`LhS3?_njKazMsJB
zVJ~WwICDc<^CO%W^U_BkUY}UT!$_#Z9$lIqd)V?rn}HJFE*IAQ_=_LhHjUn}k*7`{
zXQ$C3!%K{<W_=ul`kcB)lAoPY(vagBdOo@xZNzGKgZ2=wi<`yo?G>`#PA0uFRiQ-e
zbR0XfFmv;Z6#iH^6GTuG!D9Qx)%)^|euw@dIzFd+n#zt#QBzR@L@7`6%OpolhqKXi
z#<8tGK>4xbNAGXKG6h5i)tGXIzF6aWLRMyLZg-PSTz6M{RL><kiee3o?M19_Bz{u<
zL7I5&CWJR^eG5q#Fd|^xZ*^P>o8(dc*b_W;GzLU>-mq<CYU+h|YN?@>*>k1tUMOIZ
z6{US7F=`Jaq|dP=OX_zvjlYR?k@n@k-Aa_suE6qH>(AdVGJYr&438AqZVknDEq+VV
zw{FI5x|PS6uF)_4`SqO()3W2VcMky-cB#qSw5z=oRdN;yi+B^t2aPSAQ9g8-{Jxnw
z;9@E$#eh?G%MOqrGl34m(~Ve)K3yC?a*0Xc_!-s>oK!!|_!?a)3xA9X%*fjj+kW^!
z0Up9_;VX8gb?j1HwbU?7cSyfp<#o+Z2a9%B)|9g#-A48S7tnFB`^gNxy+=vQ9wF3B
z7N0QGs&b=nHbpK}pEz?n@@^ACgu>r`90g)J+E>C(f}fTL-C^h=9m?M>vDzMDn#T$E
z1|bAe!G5m~lbsD?571T%7lVK6x^9W1x3q<kb9D;o8prZ6&+uAw>TL!E*Cgcs{x$$&
zMEojUa;8{}7xzYz=GUG2$H#D<N$F~ukNFivx&b2oM%AZmYpG%bxl37RwwEq&7|Hed
z!Qbdd26HDs-x5yN%~d6D+%3^P7CG+Xpl+HCP9s;P*(+p8w<@itAY1(WYhLWS5`aI}
z^@Esve}cy1g!H{P{orz)p{Q1ivt%Z-%JXvNbJE-B`Le;2hnuBMu%ygM*}ME(Eh8E4
zhczbg7OAORl%&hlNYJ#7=%(;X3a^Hd+8jDIqgteH-Pg}~B{zdi@_-qFDa;cI>(}lb
z1$8!4{kYHhsrSQc%(sC6;LsrYQr!!UyGk-|)t^XtvOFlRjc3rBtFdZWL?&4AwwWYI
zHu}E7>oc<N<Dh<1rm53;{fU256JW~looOsnHnTkClc#+8c<EfRLTfV!n-_U3n84~2
zLCWg?%si5>!E0pz7A}uBbE>gqY)5VIA`xgg?#<@OZjLAds+D|W4NDdMLWjLZV==2;
z^{Y`BPS@4pHxR3<wA5Zy^${#k<R|wn(hc)OK;FKxdvWRIQ3m0|PT91ep(>s@*&C|Q
zsU>RsKvZcLOtk=UB@jx)bmxsLn5MptgBZ3E_Bo=fWr=Wlv~oWT(Z}AYyK@M}<+xsV
z$xgotN`(b)6G@Y;6$;^|VLF)`yfe=ck5oSUZE~6S@rc~;w-4Sd_7Bnc(dl0E+4+68
z)w5VNM%k&y!rALboEOwyRrdEqZQ)vjVeYOPZ-Wzeq;#jnHsDk4^FG{^rGX;hriK+@
zS49g1?Db|cubkp+!Bi^0mZAXV5PB+X2UIcY8b#OU8!G74RFdZXrr6g!<Q~Nc(#3Kf
z`2zA))%B`pQPM-QqAXz)vf2jt^4LsDznZW3TFZkKWUcMgUiSM6Qo09L5Cm4)b-nQ0
zdv@(TQ_T^{%@$ZqfzPp}DP`knFP@9Lyl-}zhs4ii6OH1B_^=6`w@>%IKd<6&KUj@p
z-cD}H!TgNxE3Mx+0$$y#8Kq2p_~MmnC66yY9lyVr<@x*h?a+|mIP7`QHnB3}vhztj
z_a%+XG^(7*oT9tQpkq{im6?;}E&~<bY_M(-<&RNAmJTkxVTXw0pn%%^qRpW9aV!{g
z(v8gg;G1L9MOGxa^W3QS9$@jyUjqo-V;P;Ose%-QDo~wP(6>Kch0KJHtq|BLd_SY3
zt{L9Jba3!loEzXV25GRSLmurzHOgjBv(^l~@(@@0*zI(HW5t&?UZzkcE<#!(z&;eG
zw-O#BV)Hf@d2uq9X(YUrc_&dQNE$FyBKCPTEF@b4IEuBr0O#^P5=p*h<`B+kc|c7$
z%wQ0>bcEsEP1<Ij8{{1Eko((fXm#4Zi?ezWli%t*WQHVgd->kpAAJYQPZ)A*$x=4O
zXrB1O{d&z~S%3KwTy)V&ajPd8XF(hDVFz*qt4^QSPUA)3lE_gi`H{E3;~#sAC>?Jq
zh(onO!}@nqG#;3A%Xk>0KNMst(c_ON&J&^gAZjPk1m^Ttp^6X1a>~<JG60z=JR^@I
zP<8gJ$3MYxe}bzMQMGrC$Ib*<|CS3Y;CPaFP1o3`4JRXyOf~wgPAj`6IJ2p>2DpW-
z%d<V8X!y8p6zbj0v$$uSs{|r1X-dzheLmw9H>=!shkK+}mCxw)8?F0>l^Y}#Y;+5p
zHx_AY=ayEV0Em6!nv1^X0uB6g?3McJy=J8KvYVO%EUN(dV!&xO%6Y{N!r!}>km*Bf
zj!?7fhMhUUC{`@B%f8wrny{6EhPMLaxP&pBFfmXVodXejs!+AFak={`bCz~pD5#d%
zML{Up#S~2zdxn<}JlJ$bVNTt!K1bMm2Dinly2|_vD=G^o^y^AsxUZ5{-1KOq)J7lb
z2f>&;ybs{C3*eT2+)Xy93+lx7^?U4=IeO5r=5>HgIK%RwV^-Q#4nOrt_iPh<(Dh^w
ze!?72TyGp$x0hRKuS^+&p9CfcgUK?oEFqt@1ohmP??KfLk~MSfC4&eQ6b>-zE6UHs
znI7Q9PGQ{?CQ&HIl#Xtd3#)(?z{RP3!<nJX&UBgCEoY_zENtpnhTm*qR50PGSC^7Q
z=9vWUi7q5sutAYfq2lI*s-aC@FFdQZT;Xzz2J58=CbE&+E)l4Viq|^j4KNNRKUGZN
z0w%|XYww+xkhDL6zchS_y3f#n@l89c=6UG2tv0kuw?lNDD=eeC@q0H*p%5}v)Ky&+
zD3W5kVJpsrVuvp@1tvX;GnW9sQ1uov`zoYH^chmD6sW!oXC!4Qko_YGyv-$dOuXGr
zZXpsvJ|}B-Gl%i7)nn;*!$Nl5R?U(hdDJAPg)F3mRE`6xXb0hgln~9IqpBww>z~sA
z4YVu3WfPEDZNYDFkah&FgP*~EJD;pMvFG%ZxK;#%oIr=KPTWuPRB^8S{gt};^OcIf
zrv)|CfYxtM4Kd0hE^JV2ytR6`+L`#t<~HqC-RmpDB*I6|F}XcNpMs5mLSMuol<qi@
zW2AZ}lzE&FY59y4&uEadg>v>u?!uGag$6S4zBN{U)eYF4ZE-)*O*J8~jE(n#OKF?r
z8td@XRcQn2In|meD9>v)qOw&71<j{AuQ~=PC$^3y*bZUvfCPiEuTkzAWPLDg6Z{UW
zoZG*C@ge~(K9S%4&fw>U?VGCj0Z46$@Sc<4s|2wKuyPEjFdW`qQ9P0_{|tr^yK2;A
z4+ZIrzbtgV43NZmjT0E~Ewec9BF1TKq4T=NvY5nGvU8#VRgyAnT1j=>GmTE!2d#eY
zvIKUs;Gp;+6LD+4$#H8_nkBqtw!>pm7wD_90<Ya(Ao9k)$(3n<4?ut8nLV5{#24Lp
zgAklySjWCZ1{j$!hLp@8axb_mlkVxis{{tN$yPqgC-<=ehA8W){oUp2FwvFsO?0q`
z5=rr=VUnB4Hz;;=GSWYTCAE!;!S)%P$5_>2JNwFhr#IZk^s}Z0#^1|k4nTeR$OZVO
zxLqLMxMgUMCK9F3yeaBllCj?9uJ`GcaMaU3woVbPEJ|94G_J3N=XdG^NM6nHie7R#
za3FHq67gg5FAVOBx8PI*Q?73kwVD>UzWXy9>J2dY!L_hZu)G5Dy`+26vgfauw@-08
z|9r4gFSiyCCr!u%Q>^6ILGQ>4jdi4=bM~e%R)haUUnqJSqF&APEQrc(wzxFBgcQ<7
z+>w#mloc%2QQlJ~?4ZN)$>Rz%v3n2y)f``(|L%+Pr++J40J}lRNFLP`t5voV`@Q=9
zs;-kfsx3QS#C^7SMf+H6yI<EEuPR#oyqqpiZM;8@)nlwXUTJv0+Wf3*w0|V`?D@pY
zm^BLwUoLp<R&lZ>>9+O(wfoXXF*aB_cU=rhtT4l>dfS@(ajFo4D{`&L@?}az7CSPi
zXg*G}t&K^qEzPP9$09cMvG=rp-iD=tn9e8av9GXf;Os+orgC$tE%v=-RN}&@Uql6$
zMW}hH#XhD2Bguivqnux=yJ-|v(`b8d0CI|4GU>=JujxY`t}s*5gKgLNp+~4sU#(7W
z%|JVzBlPX%^@|x6Pl5T<3{lCnQ3f)VEFtKP8_9!1`-A}a)X;RFlsN16ESn6k2wb*w
z!G(Bwr?7nj+8Y$LEWt-Yt^hqejqMBPG@WMW`c_4wjsnnYD-bH27))mKki+UtSi0>o
z+9`~*3YG(gv#ky}@906Wo!A&(F*nuDeV8~j2a1r1?fBZ0yBTpv&RMW*<}FiG1Snd&
zybR&H8{}TPvfYi7-jN4jQ;hY)yzP%F({%$%Vp`l`ALliXKjeA`bE3h_E6fzcS+`0#
zMxVZ6NYJ4jv0N5zy7A1h%&Jh6Zr8~NI43?=DE)l=8Tfu!Rcabhco8`<{&k4{dAwp?
z{oWm%O}xi{@v3~SrGCEY`j(Xgv3(~fxE5{X$!0cx#zBqOJ&D1Q%7e_l4jK_<V$<(V
z>Y=gvPLBe7h$gCc(#SL9*2W~GAa{82r&C?6O7wXumJo2`Ksl17tRvd(lM!`xX%~pM
z(e4i<p=P;Wj$pSb*idE*KP14)UZa40M7023-9ug|^1A!OdI0F&pd~WAFandw*$;-3
zH~vJCf((i;I+lFU0TQd>)2h$)u&rDs&yqDLJ+eYT>xD12O0jZdz3a=yYwnX^3SHBM
zh^=vIdNJ(kDL6oFNZboU?yk|wmr-DOM1<N-S!##Lz_2Y|CWSAC?VBTcv)RPR`|@YZ
z?ezYv+F%j+<Z52P+e<Dp%K{N=b-va_21~JNg9no|TJFliQjCW9_5j{40}^nHeQe(P
zK@f?W`#T$fR%1v5SpV_BCQ<r>V((jQ7GjuXW~M}jzD6T9UdmsLQ2)LWq{@v=grPR{
zTrqlM1z(3`nyWRVM~&6)SO3<TYcid(vW+tzYN;LpTYLH$RPC)w$`T)l^jT$(X^Zv9
z5=`bgrv}FWK9j~Y)#>@!_co?<?_>-mlE$v1*(qiDc*`sv8Sm6(2J~Xg^r~|w#B}SS
z&ZPZ{)T_i_wsJM;BRa1eawQg76f3T@C?jP`JK>-4fW-lS$yyivLb#l)&naL)AJJUe
zn!R4^pMbC8toz(+u%RdzHWAJl?$QV&iRygm(dcqtqy3j=QoEF3_Pdu{p(gkHP#obe
zF6#$>5x_>7o#c%$i@M~@q?wI(rdbb*@UHeQ8GVZf)he2GV?k5dkDq-pAk#Gb_*s2J
zPw&NU&?RR+d7mEX4WEOr<ZX7IcJ)Sm1*JF16okIu?z2Recv|i!UCT&-ru0G7D{Lg{
zLLgPtqI(NoeOw*v%jA(o+qK*{xZf%qb{T`DNPIl2=~4(;@^KkNJOL(4dhpm>OC&k!
zWV6kz+wJlZ5AiK(+eSSqF_IaxK+vuQhk<&w2B$-u<;^$O=Aa!QSFY0@FS6pG&&5}D
zsB1?V1W}Q#0SmJ}z-FK$%NMq1X{_4^u(|^2Y~i7l9+7jFAXx@zjI0c<!U1C%u>%yX
z#mr;{4^}b0XL)aE)#TZ5c0?}HNRgKm11p9ay4(H+2IH`sIl&Mt2K;pT)Ws^z$g-nR
zSVK)Gn8)P83C^gptdMVHU``*{9a`yoYE96GTYX?Au`#Q_d}X@uNs~mb6={l=V*MR5
zV`4sLb}%`qv>{=+1!Y}Oxu0pEoha>df<#HE$zx<ZM`1ohXoEGz^*Gf)%vP#ts2Szb
z0^8l&uz!2g?`p0o!7ZLvymcaR(zX3@2PM=ejDvhApS1Uqk@Kj*750*kr*G4Y#fV#I
z8wHP-GEPsTtCUjRl$LCOF`kc9n%BPOOHQCai82Gbm0ZMY0drwZiQ)8ZaUoAX@e5&U
z7`+u&PxYPWQzGky7v)hj2aA-cKF0%T?P#}1k1aw!{(8FJ@SoDM`8A8P(+wY4AlZqs
zaR|3*tAWf#*EGX&!il!{Iyf(Ldp|uRQ4;y~t`ifR%?kXkCliN&F|G1;R9Y>a#<wa^
zpEC)N#Byk`OO~_?rj;>&)txuTP#bE4=p{Z}A`tk{$l}f3UC`>#*sSS5*PkuQ#kXw~
zk$cQS2gy`ztoqa^h`qeHW)b{)E9fA!Ch3e#FdOFy@l!5#Ig(=1E$=sqhmO*fSFPyg
zXen2hT|zE>qdFxuVz9eQb-IL2t}}(S?VCA$L+17&|B&N;!&`w3Y^RRst=n66!^tVG
z5=Vl3TIvia-aownXto3d3f|u{*nwHK_l9%+zzm4GUYmkD`bpFL?m31=`tN`8WI-Q2
zoI^7I=nr)A>-EnJhaMYAaT!7SRow1N0m6^d%rEao@tt?cWLd0>ik|si{Wh7D7PZ%t
zU$frcH&t%BGmYZRaei3VY=?%Y9h?PEI*(bLECn1;=LKJ>Lp5vTM??zl5wXYK;%Uk4
zBSTtWUvoK3?PJikB2s`fh;P7K&V=7LQ7J7p7DsO5DP5r~k0YO>ZTHwZZOPclr=afQ
z7})JAxpd{QsWw-{zifdcdJ>=el&)gEyCc1{i|I}*cF5{?rPPo!&$i`0CW}0xYpD*f
zPEUVVwyI-z`em{^N4u*K)vtT?L}K2Sfm0ix=HDIu!~ncj%>YY!KMZOKFR#DKeAvX(
zs2@!A(oW+D*l*Bj#|qH0QcICrUI@&(mz#^_OXF7uCL}ySp7lK)$g_%Pl(Be&cUxD1
z!U^rxA3BOLn+(poZcq)(z}#}=+OYLmDEBu|sIVJjF!C+=)=S!N!R*et=83k|ch!Ig
z(YvnH%zQZXZG?VlW%Ocb&E#>=?PP#<Fq7?FbDH^_$mShZG95;4?F2ij^TJ2*^F-2g
zd;%a3scH#ln)sqW089Eu`f7bIc~!O9gG@oN?a0(6b_1KZY3cOCE9CXPHD>g-Ms|G5
zFNA<Dz#|$@Td{b8>*#}`8I1;OCR^}p))yds@5uNDFei3yYQQvCH@OdK4Vxb8B;Q+X
z<0I!9U|S2_o-=6M30^ik-{rI{CSQqh!Cp<><*s_I@!mWzqie^z=-M6aEEVW9RrZ86
z(1#JvEOzlBJ4-kEOvm2wy`tb0Kw!8yOC7#Mf!>_&UemAbg2;$%L!9<@;rNBo{99YN
zbW;7nz!{gj$F)NjV?fAb*ccqz64rXjue~>LIeYnjDj0IU)h~<tasLIU^|%0P!~h$w
zBg!tC(hqWYClzl)RCY%82}x;Ut4HSvasLd(_BLos-yJ=Jw7Ea{THprgUKurBCo;e>
zTOLeOTKf+rk+l`XU~nliVEsA9w7l>=&)UWOI)ge~Z_obRPxWz|!Kt_oK<_GTQ?(^?
zlN~{xOFXVZpwY7GR}yNb-Z)@}w0>{j+wEE)S(yyXL%mLeR&Afa<l6NzwSi4(I(P(q
zcn1xhShOq~&7knQ&8eTkMobWsoxXTOy4&Jb!ZYAK#{uhICtlsMe!&}m#5nwQ`H~y^
zA%|=x1S)0+X0@NxjDQ3{6`=|oJzhPG@1%^bNj>8X*WXRR!K_~qo2-QLn`H2$X(IHu
z6jn&%N5!$k6gMFw90y4lNi)qiYXCTk-7?XXLPo}YS>nWJUNGpKc5P?OZj2cy|7C%n
zL;|=UUCbIyzpL2IAQ(bg_W7Q_Sy9{loiGNLWMglZ&QTvx9R^3F_edUdz0DRfqND!B
zH+Jg9l1teU%?h4HEKi8gwC<HZns+i#>zk`)lh_J4gV0d3PYKXq%y@GtG)d+;4)!cv
zxeMG|kd7LQLSs1TTzw}|wNB1!ELPtnVd)G1df2hz*vR5NRzlkRB$;#8ME68*vYyO7
zo`L<SSUqoC67KHxor~GN@OU<%vrhW<ZJN{wnlx9daHXVxQ@Q(VCT_RJr@5z(FvTLu
z695yFFG*v!3n{q4gwZ2MLSxm4wp=a9iIyywXnoKq3#0t<o=y}p^rd394|c#PzsO<V
zYJ4|omkTLLY$TJxB52Nf!IDec|0qzdc*^fy)lYSqiToJGOeHMYHMVIluXGZDC;mA4
z9MuM(W;J36-nLy3_FGqKqOQ6svmmp(sfxo#fSM7Pz(W`JLKHiiTs!LW<($#sTpTJQ
znQ3zND|9se+*>@;!KC?98t5h_WsxVtumFrOJCJ~CWo&?eNc`#j5{1;3sN?$(P2jc{
zQ0sSaC$;D0Vv==*xf)rzvzhvheC5~+`1Ojp5-R6ai?2HqY3OvN-WARtgEfd(mt-$h
z6~+;_YnD^ufDjnj)x=Qr$yov_81OG9l}QfT0$}nzg7ZiZL@iZMy;@viCKAFBDoqT_
zTq%R+Fs=1~v~fLO{#?L8hyc-cp%-&~9#0%UAs}m1DBXX@y;|1Bko4|e6sdYUIbjs(
z=oA?+2KR{)BG?d>NZ7>J@e3wmPTO+*nzAFZp=>LCz#GgWiBb|4@0HMh9UtncOoO8;
znIIt+Dnhq@mo>MmnM#{E(2hev<n;%}&1?AOGIP|Fbo<5>%_Er|$o$}fKG+1)JOwaj
z$Ihwc^?t&oSk&op3Nr0aSm&0R&4WtXIQF-w0U_>4v-O!LOLr1q54m80xv`?Ml>Tc9
zmdSNu<^h)-v7L`76Fhwi5^FGXX2pF!K(XKk(aO>VBDYH<3sJjqudfj9uMskLSjL6(
zsw`}j$Tgz6u5ZLd0S6Px-?eQ@F*Wvp6&X3K@8fkoOVXUiqb}ve(+<jBs^-3@FuImP
zOaRDP%9JqE#$~#{4VeCs_~@G|`Qm!$%do1y59ta2TK+!rb>;)}m(~g_^Xt{8Vh0UT
zkWZ6@si(&Z^J;oV0k+Fg8#Sk?w@-eAnRw%!z4V4&EPsm{(f3+guUw?{=9oG?K1f64
zl_Bi2PLgxK4F+0wmtUzQ_JL$~q8jPD$(GvDUv+^-qjY`7f!#$(^elOH3`9vavetB_
z$8OB0;IfVDmQJpAB*X@}R?$3(U&=4re0i`^#y<|YPwSVjrFM0RZctjM9xd8;6jbD{
zB1@G-To;Z$`%KQ~S8g-#JHZ(z*iL>-k?9zpbrx>jJ{=byfxbPnD+uhd5`JFVCv#g6
z_&;gyvNTCn&rYcnc$}=2rm$WVDsEp;6k5)#pyoIDh>0}u^6peg;SQtfCcEx6Y023)
zx}O)^gT>h`MPYIVJN*cN(0g#;0M|4oVY7^2i4h>N5zwY!ISL>lVHPNUx(Gb48KG*k
zo)pN6&V&ygte};r)fkKBN}|Z>?2)qX#v;83DaM+y6mHef-q--~PT%`a(-k`_Lb>})
zg8CBnM4%TdI&ER%8{J?c3B`k*d2w!q6=Ul`>F%P~o>0ucFqw&05p)I5;=~%N=Ja?A
zQD0EpWN>9?%F>`e;l+KOV^@XRrxl8+<DkXtZX~P`dlJ0(1icK%vVu@Gnv{_5Z|j)z
zZ=0)2Nt`Cd7t+=65Rwoa?k74E!q+GBCCYD^+)Q0;Juy?0J{2}HWH`@9%v5uCgnZd>
zMUT;Sn`fIcc8P*Qx;vLdkOWKkm|V<Mxg&vSL=+eki!mOHlMo1KALc(g2K@1z4IaCA
zfuRGBs%KRvB|Rl)_o#7?I6nG84b;w(L*_hvWYeG$k7VHk(H1;}yGN!?xfx4dG!ONO
zT>v$$U{<toOE1Z_cQM5RNTNF1Kh9PMY1=Z)Ihv{tB|K<L5a{<va&5+lsI$a4{S}0{
zh6kdsM#5_XUEX;nKJ5m=mEn={5H(M)ez6!z3B6FEMDa<CQF#r`M@WcpvFGZE0f~4^
z^{L>IsU;_~c9w^nUik8EhW|C|!iWKeUi>(d78#^!aAq{-#s8pNAG5F3RX6H=CmO;3
z+#^5Lqe|ZN-e8n;_}VTz1KPn<ZftCSyN3OmL(-v!j?yhngRs;Q9<J<G<Vd{haOnw0
zc2)sbUuGI%;#5b^dsmxy*?3if9=<Tyad`BfCoji~aTtx)q4yb(sct$rFU*9ybs`lG
z7dus~B~i2zx2UMx2tV^9xyo3~@=({pK1YeL?7-GIBAfB_2PsUEw8F)q%PkKVq)l&D
zTN2PHJ^QsTdJR*8@8>*w|9c7kp$qpp3HrN4Uok-B-jwFIN<x8SAuD+p8m5|9%<4)N
z<^#)8%TQka3xN4^CwV~s=H4MQUgj^oCc!uPRsgAwIzN}_pN)VnW{8$QuWz%G$8uP_
z##P-E0Z8`)V`?w=$P}jkRgZHAqu)+2bCmoQk3k&E9*{@b$;Cuj{^&ANls>3p`=ERd
zs!=PzE}Y-Np_8{Y#p5V}nm`*zQA~PG`^Z_WZ{d@M|5Vgk7mx>7X#z^J1N^(`sMy;)
z3X6&MxV7BE6>>`CVJaw7y*=1rW+ok#hidt>$&vrs$UJ5C-K;jknD!!b0Y7j4Ou5G5
z37|=Sm&vgG*YExF>Td$2Ovm5)Ei=`Nc@GXxfOLOi8!vx((#IzKWGg5wIjOOHviklW
z&SJ7hVdcE9#rWW9wzX=eLYF$RADoz8rVIqHjCcWdlvn|vTJP73kaY$U(yRmBSsX#f
zCZIl6V3TRo0;=Y>8?o+^zXhAw#%h|{0cZACN#>st9v@(|U4Q<Ue*a*PXiE4pN*H7z
ziJoWNt0~?*#pVYUd?kiHx3__Ur#>K6`}H%By7Le`Q7j_eKcyKtBxcfeG@a`=d7G;R
zoOQ5kd<T>kVPke-SI<*tl>MhLCJ-+4vXPD~srM!d4I0LQnO54aVs#DhpC1onuZedr
z(1&|lgVC}}wfo-VYW9Dd#(oZ{1s2=mqu-2}{S+}l0AMtnyBYVN)L`OX;I8LH#Yy~O
z;vpVKVy0TBANSh44<B$(Ec((Jwg2UI{J(sf@&sJ2;^q9gKL|B<*nz@w`}&XZ#e0XV
zI+KqT{=;qhH&Gb^BSKr#sbos}eVnX){gCZA>ZJL!lC@(vGGIlxjmP>bzp5~qg2^xC
zPyeDL`u)=-V*qG5aoOJcC5rmb&-|Ac89@YUa^jcnGXL#m{LL%+d4)olt4hJwY5OJK
z_P>_NiPM9JW#ONg`b(|GFR%HZzUZrn(Q@9ur2cnV^ULe}k6-M?0nhW+$4}6Ixv>BC
z(+@wjeQmYx{~}oZ2NB|bx)(eWV1dbeUcUJ&62hPR?mu1gKY#K6hb6FdqYgy~K~JSW
zczM_7f4mebSbgOf5*iiw#*72z;Lospv@O2{VBC7IwiVL-Pap5P#0xMsTnGGk_PdoS
zbceTj%-);;8s-1DpE*SiKMe2yTMk~EbvDi!wGY}3e1al<Yasn^TjhUTIsg22O$5|}
zOI;7dzp3)bSy*~g{gEPR$gkSq|6+OUzQDlLQ|;<~|9U|F+phA<WeP;pz^O|~mofXl
z{LlZ(w~5Asms+%JTh;M@v?hJmG19a3XX@=={zo@JoDlO)6Xtu8{x?Bq|9(q8oM1M!
zlfn1)KuZ75PwBt<wD&beul$->U;58vrvJU$uq1{N$ci~-CI79<{x6>TUmL{#{g3~H
zJeH60jUg$sw{i?$7<IUwvTpI;a^cF;$`@`|igx=Fs}{XHsABzxLe`AH_obJYTBxq*
z<u_0<pl>XOyj9xfS<Z|(Hd3oWwqX<}D?%3c_jOgt5GNoOXD%i??_gypC#OiowPFvp
zsKwp8vR0F_S~4u8BJdB-%6FK)_JCmFYm5S)CT+s1Icj?K)O%nBrG3>L%Zge~*rzn7
zRCGjZC~tBdS*|Gjm3{>c6grNW8tWXIS9;A+dIF|Jy%V1CE?{w*{U46l6gkWl0_VVw
zCqTFG?0YiW!4OR4^R%%x!HBd1!^C(w?sdFt2ZnGCLmbkRo~wDKXoybHOtKYh?CJ@$
zvoVAOqttz$I0%A_W-k*^#=9$}{=PaVg2Y*8&8m2wDGFxY=6Lan=GoB|g7l2;8;C8!
zC=8be%pnJ>fxGKEkhi!1Z!9N(AUK0ymPD}OHjRT_h`Prfnc@TJaFI%)*5I!ZNa)Ii
zVMBu>*;Py-1=28<7kx2c@&uqcjPs?#K~-FAJF|PK^IUuAyGII)$z@*t9}0pGMi+mk
z0|C@<99WeuAJW)Tj||i_te&{14)=?uSsiRRAJ?1}Ta=#HJ@*(`D#nCWxYXm`F9WlW
zzVoQ}-mEw)0?v-zx8}hVN|!Z3Vgn=YSO?y{1%ishjzAuR=szD!+&W(qnO(vJGqF`|
zO1OYT4!6j)-Pfo@X1NQ0?SvKn;&C`<m^qJ26RG-PlM+LQc_cQ(-yV*hXCJW6KISx^
z0CG}vBNOXyG2&PRPY4p8?5~cnvuaFxW3sAFfG5!@yJdZ&P--hi`;9T2J9v4Bc|y)W
zNTu+#))wR|?VRtL3)qYPuIT8!j#V$vPSIYVUQBteGu1Hq`i(bG_wLTFfY6C`7z#8F
zWexBI_{`lXj7>ZepzISsk3C)dIB{+jOuBYPK+cbE8u^^=yVjZBt7n+Fg$a<_r|q;Z
zDb}EajdAq)cG6|zJ#ckQsY^1xfwOcaBtK(24Isbk7&7|DkBVSU=Uxg-zT(gotGHfQ
z=jDVDHcLuH#|#L(o**TWNHEOltv;Lyvnz-dvRM@Rf7pBLs3_OAeRxYqDIuk_BHbZf
z1|m`-(hXA5-HafJN{mP<rKC!Cw=k5TbobDm0}R9Oy4}y-+vj=T?_2Np`|n%pU5mAy
zwYE4j_kCa2b)Lt0ggwWP&hX!MUpYgZ`g4r0DDW;k0#+22fR->0JJ06|(wUK(o|Qa_
zq)^JrR2|<rm3Urb>re4hOoG!FtdIVqgY)u_9w&^-Wu9{&7SIXgCq+tv2>}}Z@;K@M
zsJ4(G?P}ly2Jad-76!1Jod9VoFYd{wBVZWHn`$({X|9KbhI63LU%TFe33c<oIA=T>
zvn|*V0eu+trLV!~3Yo3_QN~lw`#>S>3JDYD5#vBRmZA^DR%d!{3M;Gi|9p)FmLDuZ
z4G>2Wys5@!8?B=oB;?4_nFSuR&R|`*eYZe^1%Q?nE#{sOAcN?*lYpGRIizDAp_=Ff
zOwxsj)fuDLA!S{flsq8@_)rRXwqFme1g_jqGMJl;aj$qo8?ju)&F8VTY$;yNP)=ba
z?Co}MZN_{m9sWw3p@%M_;1Y1kZm0n^tK}5ou((`cw6S@m1gjA>!u?D33=~DNX>60=
zQ6AzcY{VjVR&xTNPb?U5VyJk>H&Gy6@I;~p=*uTxkueMA?cC^uR8o^rfD*{w{gO4L
z!qASsyZJRIe|Ih)Y#FR-PbPOq^*VduS~^Z)HGsOt;vPvw(of%0^zylrsMrZDMo|g`
zrlF2Lz+yn=+WDo5u1GM{h11Fg)Y3|N<bp@?s55}Hg~TL*y5g%XctA84JLCVkvajS2
zIH;W?02r-c_kb-Jqab{n>E}*A@;c2Kv2mAY#%_$>($~rQg1hCC!wOSz>Pn;I$2H0I
z1?L60WnJec811xOq1*+?o_rIGIuB&G>w#s$1ejb+0Cw8^urbJxw1mMuZ+z#eT6nif
z0z<t>1Twv|O|E2{SvX_JG%z-cJdXyN>8i-$3~?Mcv{SNREfD7f_)|uiJE3Kt0jY;m
zODM_FxstuxI+qz;9q_cQy-hRK1lhAd9x7v1EKg$oY1_Xuo7@QGbwA7@dqVgn70BpV
z1lMG5E%oBKf7uB}fjf8jVgA|uM!=Y3xdkgP)U|f8tm7|!?I5qmisBNcmuLM|JNzia
zS*PxbB52_}@okenb%*ILI@BEuwR78Q2H#TGeHwIJTqtk&wgGrBR&uNChDicBNf$c%
z!?UmL_wLBEO!n=J@O5-`GAwe?z2_>eE`b%x1A1;~s_h5~yPkAS$WlSv?m5VVQ%nvU
zg{u<ws%hV|K7YkiX1bUZ#$Eit4dj2t7Mq#X_nmp2E>2)}R$6xI&q1zlgdNfJV;%BK
zadXyX(H;f;TJw46zY7TPV&MMsEjB;Cu-+QlGe)oU=~>(3$S3+X)iO^!&+CbsJOJBH
z{~nKp<DWZsLs^^0F{ARBubgyf*&uMa;1#^fY}0h(RKRN#McVcTnr)G@sVe(tl}Otw
z#0*Hi8-~13xg`(HcRs_;F_g1-m#&gZ1NBh-IhXIpmti*$KLKVh5yazN!RV&Is+xNp
zkMOt3I)c2B^HID+$=ni~tPR#JR(a*O&zzN$AgAc|fF^Z;QKIOPe5xxzOu_ucn@evS
zuK<vmDsJ!)V4m415;8+~W-{#gnO<R9gB6I6v8BY-1tJlRmNAbSO4G1y$b5&ZPv3LW
z-G87w(`i(?go<iMZ8&8H5;+zZhk6qtKowDnBuf~qe)k}UjS%yUr$M)noBT{*_zaBm
zr*{wZl`b}br|&Us3Xra)LmY?oB0U_~Y#@~BcV}Dalr3M&W5XajuiMc3&vDkDEm}Kk
z<9TfKxwgVg;PX#L!^@s1y$0sN#MFHJ>B)wImfYmk^eK@?wtp=iZNSz)f1y~9hkkoS
z;M%!@sq{XWq$pkFFdQd-mU%w??fkqF@CW0X-3}DX0*1LSXg%F}#T@y7Ct8=?ZdqYN
zJ~I93kVij9o<<2KBEjfym7yHY2l<a(xUeS9X^`9(E1{?lXxp|Ycp>&A>#4n0g!o6-
zo<W8ZS#d>xkIuTjLm;&TfCP7<M<-Ab@<=L>DM{_{7=QKMCV*=SuF`tt>z*$to?1u|
zk%(~<D+7C>snw76xHR@4aj0XxeB%~Y_EZ0mHLY_bTU<O~25&sZ_<oovNZExmNJ{{k
z#d|-=>^yVa-tt2GFpr2vxJZ}m4-liLp`PV~1C#Ot(~T3<MG^)BGHovJQrG12D^^JP
z-zyLgEZ~p?NR+xLqt|~d0H8TX`rSFvXuBuMr=EE*`4g|#M@g75o^}l&_N#nMDX!S_
zl=L@Foj7OAN(Nz)FcMN%2(-%#fPvK;Hk~g<vQ}W~hp#ZcS$V@w+q)J>NHceX(IMs^
zW*nP?3Et1z>(5t*sV|5<oE!KFv10bxz5GCrIGoJ(H?|3`RM{2lYfaPY3iOnw9w1$8
z4#Dj@@ae7cM<_qBVL749p4T$3W=wH=c<|Q8eW--_c!oc~jH&^nbPa?q=J>~@H6Ngf
zwU+k*+6I-vK`gU<U`)dS!c7Xs3S0yq4j59Ck}_|C)7rv)EE)$E5pxdUw4)%g#o`E^
zk#_4lhMQ*qPCWJC2k2e)Vy00OV;s+@iyx59Dn%5;N{yGP>c1}9{cY*pctm@lUV(&8
ztZ-dY50&b}EQ`)RlJsMqj(WzWI~=l_KRtYXC=ys8Dp_Z&Z=6i3P7fnc8}{QXF#esR
z=Zx>nCW7lz&X;7)_WO6nDVKDEWQR2LTk9|<YhUKCT7ZD8sI5YKhpT(dMN>ru%g)05
zkc*0v76#>Pt9H`hY4Y)JUjVr!;6(V;PH@1YCqR6eSYnyU6L1-8z(Fva3}T%ry+^^K
zp9FQyqgh5NVA?KU0R}Zt)+VA3yy(gO9IT(2PJo$!=*`utk{3e!$+j7Ljz==8_64|D
zZQz=y_pdDxp-bK`#q7xXr7@`ao`6wxvMFwK6}bEb&9Yu^vq_Dmdfrw5^@0IZI)fva
z`p89O)ZOxSX15ykzsm8%Mo_A)Der5m<CknN*%1W_!pK`COER<<5h1!Gu7~i@fgK^7
zd@F83^A*U}`hNITvb1dk2vY%LrnTOZHi5QkQoJR<J35gz^8%(<x%%m5uZ22y2+*86
z@nP(wkrA@mUE;2ki##?x7{!e><_RLYW_R?}PRdSIot~YgyLB$zF2QB?a%iEQB`otz
zGQ8o1wX3#Ppg1wpb%REuH6TgnVL&j+F^6}X>YbFG&aYw5EVA-nrR&jrqi=RjS@e?0
zXc1?eyL_D9?X0bKe7AOIsYQH;b)NW0EDOmc)e?n1gB|wZPSt-xpZCk(b5IYy5ql+8
z37=B2nj=1#mmT@|fv9|Y!Gq&38GVD&Z#wfHkX_-7awAz$+I$_8wtrfI>GfS5wFK49
zl<mf?CEr8wC82_0aodeD%kN0(lA&&3Br9Yg$}Q7NGTThsU}K~(80BJsIR(z0_`q@P
zju(*0)5)BU_MRZ8ezJB1<z;Q1sj7X}*4&x^fu`j6kjCAARXWAj+ehZ26#6LWwCnp*
z9WNu9o2Cg3bQbEGvbCyjBRTT}8Y184|L)z~ZGdHo<X8t0pRw0;01F*5>%OWikg*%!
z^YgUGzHQsS?SYL!fND~>`ttnA)z;G?y2;D5*j|jZ4es+wB-ty@tdYYV`u)2m-GAh8
z4##8<$A9JbQx?gc@aj3!3~4H1z*|B`iKrRt$$Hh>OkF$osnp!@Im@X=lznwO7NS!$
z9=#mayLY$J)uxs>J;~r@2of1B+2bO<SMxQ%E18L2ylw4U6Ww%_0>iavlcBR9;EW%e
zG<7Q#wqc0cdd-0j0n5F?&wG7Z^Vk#Pz7?wA0sJ8F^dGIto<?SEQv;!UQN1l%j@6<Q
zjy(*#A5OPp3;Zl*d?rH-Hmo&K?Nw0Jkk=YGSP3ANK&+_YKc`ioDh}(zSrV4uY}vCF
z>=!*CmHQ^}q=4bGZ)l1KlPQa&bsv^j)a5y*O`O3Wqs&dLv(jJKD^XD*y9E{LxFLf{
zQ$TMbwrpbML~6QsOA}wD23w>t3it%joi97_f)iA|W6Qn3*AP09TOg9olrKUeL{_2$
zij7q!e@NSK{N1q`^nr|ar}OPLAZ8mMB>V<g1Y~Xxx8wjry$j%s#u;Nr`)|6Y5;=Nz
z>0k8vahN7$jBK+Bv#Xi<{2T_-c`vi47F4f(mA-&yPl=T^5HZdGF2qJ4>MPwD{_YtZ
zhGtFzd6Tm-8ma`q?p=d|_-ClsV2;tCMl%gtgQhca&5$M5EEvGygFejJ!f9UsoWV5-
z)D&Fy2iBD!sx-)+nKm}>5cDmTSJS?31@XVHC&pqck)fWc03BB-B-J|f#y|Dn@--F%
zp*Uu;mVkv7Hql9Nn+YN9$=&y{pf8jA5mRms93p0|<3Rzfv%%!5+tzrI?6oqB0`O~<
zJ-EB70Z<Bn6EK$ehRe49vt8(Tb`*r=@Nt6<V&@p@*9G7MiN!gcvPgGu-Q()ornlvp
z^pS0Ory80HU;CDayKh*p`;H&>G?n8Vb=A#_5yBYJGkEi!=9BKj8zJ*-_FdAV(T;xe
zv=N=cp0z2kNcM*iB{JOnQV9)%9Z{l59Bp$t!@=k6N`y48)KOLMWj>(#Y%ym%9oaAP
zDp;lcv7{*bJLG6dccCpGKp=Z9pyj4QlxRLQXzv0$_)(O%t>jBb$@^xxEBPA<lEXUF
z^Nj9EHuLUnPwhKj90x<rgjeorjCO#wPtwBNBVstQ!;O<@yMda`zgAVyH8Odr&x`&A
zM^0|>!<d?;9nb~833C-MaNWB*>uBH|mtB?D%s5R@uU#Hdf%jAL23y7Cf=@#z&@?IJ
z%zZjLLodlPI$oC)vJIC$c8!Jo{?r%Y>ttK%B`3y+5WxiKMF{k4>W{%U)Ul8l#kxd3
zBr>Y$<PdEd+e-o@I;;_4U|cEWz+y>Cq`#q=3uF`_?kT>*P%?3|1R{ik=t)*f%-I_y
z-l~kkSQh`Can2O$>TBfvS>Yp|@Jf_WNJAh$K`(1fShn<r%}<iq5uAXcAw644kqh(%
zqYH+nLx3all)gQ%a_mD1!-E*!KWlhpx(g-MBHMM@_5zui^`(6>^$RR~*S3}P%{tBe
z1I`eY_E8)_dKgpQHD0~{K(Z@UMb+!c&Pb-T<v4xfz2&lT8^47p9;OjhvAWGA&`leK
z!zR#&7HLgz5q%tW5#aLiNWHZ$21BF~$+bN1Wz?5W(D0@p;2;NEzvF{Rt{5<nG`}{J
z8W6&9i4u?iT}-v!oz6k_2}I;fBAjZS`(P=vSI6OwU-UI*`%bb1=?iZ90eU}0){o(?
zyQ8Tb*jjPH1lU9+tcHy*tsazce4?_9$cKO82&BrYyxL-&cn$?iyKrZ}p0@N)7c?|n
z#XK2^>bzH8IpU#Y9Mq$`S@m9+r+;Ec!i7AAI+Wf+$GA=p*nx#n*UsSXNPj+&@j8>7
z4;PZU`fvH=mSh*UT!m5ua&G(2U)v)3U~wF@GtEBiacSBkd|?MJw74lzzh9ZiU^V8I
z>{rG?@)Y0Df=wLZU{hsXi%G4z0fVhQmEb6G?l~Ku-N{gR8mh_b`F@E`vNv<5nkQ!c
zbZGp}(AT-un#H{W8*$eZTcv5<U2Zk&aUOqrt5-6_DWWAebrP^X<4Nu=Z{3OXDCHE?
z_nMAZ6%x@vsto_LZa$J<P*zdP*(1*8LMMa%!k_&6ly1p}Pr9J7c#-8GpT~!({VdT|
zeS1m_20X~rmZqPFcg&i@BTH(E8*wZMfc>%KQ6xS#1Qi&GWzgbc3ymApF3a<AK0@u)
z#=tcRVdWnRb$}Mkr`O4W%gf9!^|;&W<Z2UXpyF!Ac6{HDUj)5wBP@497(3NZ)-?8@
zxjyI=FiN~58BB_U<PG+4$EtH4zftl~e;phvA@<tS)NDI9bQT$4_%ahnYk0>J5P}Jj
zvTX<q+sx-snA_?1C4@Wgt$ADf*a-#7C0HUWb!;+YFYcBB3pk4MJozV)=B8Mkwn&+a
z4{zg7KRyCYY~2zy1C->E%Z4h71?Ep&`6EPlA+jZhq+MaHq{wYDIbZLz@X+TB!)4rd
z<4b{+uf{1cuU^!{kC?kwzCK0?<#dGTzF7-Vn$i^uG{21(rqJi415^iW=hQ}Crf_E(
z%h+b)$m-OZ$Zz4EsHHpo!dn4GM|Ib$laMKiiv5D$>G{Y@p}xX+(g37>6V+>r=@S@l
zr$`#hrU`e6PL*fB#_X&8)`pmPkOA7bv>D@D+|_am6S{d{V}9vAI6ZPKau#uoIqIb|
zV!{v+50Ja5RIu0eO_fWzY<G)$Fq6;}h{mrp=7c2%??|llihHCG&AUG};YO=8Y_WGn
zxVXEN&=JDK7j!ckBAOWmGocu2f4Za&*Oa{^y=$J7*-6l8UM_jpSVGr;`!^)Iia<H3
znK4?A{m7e*M>_3Nf`o)8O5SFcP4-x^wzX{knruPwwt|6$t!(*aEjkk)sdCMs=Z(P~
zi%*(}E#7cuUW~z=y#B-F@l<~_Ll+@zp^4|TN_JeU_r)77A$$F4BOS1_Ei^u@r@;7B
zy8UV+N<PRXH#1Uo6-I{LiMD4RvlX(|wiZ>tf4(T%pxec&pdZZCGA<z7w;5-cf0*FZ
z5siptP~S}3xaXXoyT8;KWjH?NQokBdku1kV7jvZ_jS_z&!(IV#dRjqHkIo8SNc})2
z)}7o1^`S3!QDFy9=j3L4CqYd1N#M3Is+sy^BXGXbV)g=0(YGtuJDG$qREtY`$pg4W
zi0p~|>1K^%sxdcBmN%l<mF8kxtj4g&IY8k&jJ<*lwNPEUN3KIr!qf9faLGV9199j$
zWVLw6;L_IhO<s>r95&izbYECno04%0Qm`8F$Ni&6gGv!Y96O@n%7jQ<hH|8RH{Baz
zWB_x6?tV;XQD~I4#LFZ8vGAzs&a2gteA3_3BZb^+RqMxM6o^x*+8k&KDxz1yzqX`r
zs&#rLyJ<yO96LlXuYK5<KLGkZTN7>arw`L7o^YU%F;<r~Q0CgAi^j(Q)jE9x8bTYB
zdMa_zy=s7b9<bBZooWn59^NP<kJ0P*crW~$)J4t#i#&T<;UWmQeN`X%*7Hrm#6w&P
z0CsxpBfgbM2IFnYN^5bTO`By7T8=@)ycx+?Pj0n8%r&uXX*R!Vm0ou;y&~bRv&e-&
zY!x>%#(8#P{AzBK{3_v6Q=eQXm3@qIeJlZuwHnu1A7yVZQK1WWpOmA^0al{WWexbA
zw!Ovr?;l5&e_nZd8{B`XLg_nImyx2K0l8AiqK-yIfChi>`nB+xmcw%KU42WPBi|$f
zoZRoRdm)+)shIujI_qWZp9&nO=B-~Par9O^s`j;Ygrhj?XQ`35QH3mpH6qb?QbG>8
zUVbqe;(X9e53!I%^Kzk^zN|X}A;Qk*Ie{`vsPTx&aj)Sxx@7X1c+alZxcFK?zQo+|
zhReEO;M<Z3S0S6nSx#No9Cg)WB{VmW;wxS^_43xN_hsTA{N`o-su73WL{j<0FWhtj
ziY&(t8{d8*H3LpRC&#;R6~s2fgG7PH!vdbeJDJ(u1I}5XS<p`0ZEq499#NS;I3MU_
z6--t?sOUITyjBoBlw`g%Dr1Cr$2RSDZjOjxb@)nwap^31$$=Ko_5;(7R{Cds^olbx
z#f}~pH2~{RoLN48|J9?EIxv|j^l*1~9>{<beSZGNu6ASPk0OM*1y2-*dj{8H1H@~%
zRavM>Nr9M4dPPgqo&wW3boQ#H!b7Q}9ynJqEGK|+)<K$2r$Vv6;~3z?=qZ0$jzp(J
zt4=}~7*aIJdDYv;hawlVq{A<ZPW*6eO{c(|I4%F;+H~td=Y`#OC%`+R#6Pncf8?h?
zI=69Afue1sZK}dl<vP+j?1L)R!ZWzyIpjqauh6?yGj)(}bQo}0l1eGLUU>KR>31ze
zCEf9cJpbIyVVvd<w)x_u8k#a7u{|uMff2JEitVQ(9&w9*HINH4>C(m1>Y92w5mX=c
z&wa0ACBZLvHi&{`2<ol#)b-mTpD%UzHjDf~$}`#TP2U?yD%sQ~=tO*B_J7od7T_Sq
z2gL-3##>~#CjftFy{yPQvNySmk2{|;=EA#vC$4nMVCiV;StYxtDJQjVZ;oV95^c&{
zRE68YM!wymZ}HUu?V=_S*zIoS&nv0zeV$jQFNiJ>K}Nl>$``0<TO%3j{CadVc13F0
z6<q?|@)v*Ja;vq2{hFx(y?@rX5;afu@lulbPhbI=cF@HiyJ7v|>OwA3m^nJ`w+<}{
zhxPF+{1Rldhn;*~LSf4bA!&wDe|LA(5#-@;kXl#Te%)5E3hS6~30>Q6AvyZ!gD7^P
z?`6_yB4ea7>71?IOWs<S-eOasz~C2rCGUDDTE4a?TOO=5|5Sf^;N%kpCaL-tX3k$j
zFu;M1<#{~`24bzOiFSR%9$u0`+i&y8CXmm<J}5)CReHny&KfYPo2<J;=Zf@Xn)yJa
zAs+as{ISBd>}JywG9*t#9m7XD4@d<NpgCC1WY0>{ej;WOb))N+x7@MJsydwJ^J%=1
zj}49C1WoV;9I_>#-k;Zb{;s<Gn32V<$M7hTuoIvI!oBk;YMnTD99Yvk2fUHt6<^hi
z845m*Y-|-VUZZXQkDJCW`eVtwHYReugX4|=6r(bnd-vpY`dsG9{b&I{ng{&?k0YTq
zYVl#ybP3v(Q$EDu1-Nc^3s??&CHHy#_@22VQPl%ZWw09;&rN~H?#QI|nu}II&-g{E
zH}Ex=E@+X``>q!`H&tg#N*BMl?9x;2z&ZX?n8$CrZ{I~mSteN~=Qk%B>xU+0l$Zo|
zZv7i+cfT!ATEd~R*NE;I7(ATp>J7f0tJ;mL*CFKgEJtm`1@y^53~mxs>r;{FStydM
zQM#GKQCz_b+_w?4Pn?FzJ?S6C*5P^ACN))vtBW&Ef2#kgOH3BYt-%wZYGeMDBGuXx
z@*iRdH?UWa?g%p2C}OKuwcel#nl$$5j-3|nHeKo$50tBh_{W{~FOOs^&+{DpZB2GR
ze_yz;6RtrewZw~9xL{h87rMkFU2dp3E^?k8{Jm-|K=pQ2d{d<Y2l}!@lc9xuQO8qN
z%_;_`#tZ)Pt{|zGWWagK{CBQs;HW}l-X~}%<9uDB*y-FnT%ASDiq5TD01+Pi^XaA3
zi+yh4G3MQq?tbsHd`FI<Wt5$T44kHuat$B7Wt1(<;HJ{{isXc0%kFX8T1Ne|TFK23
z@0d$%HpuW2&fG6F1(2}Xe?DNUfz^Q%?4rMJ5M({caX1Y-A8YL@)h`XSpLkN9ZX?0}
z<6Y-<aW&-D7vtZ8y&5ZVShc{~(sb`zp)egZ1Cm-k2C)B7?=5~`>+;nIX#mg*>|=>0
z8(M6;5u57_1E3N1dSDf&J$1UkCOaC>8z8KXI-NSI*S<CdCUtstpzAM#U0Q~cu@B~A
znvF?9RH35ZM2p-yeH-tU1W`5nK#gX_-?xX_CHY77gt5~(SWKdqT}O`H)R=xSdM^O0
z+|g;L+Z05XKkFN@wW}p|V58xG-=4vP!TQt5EuSY5k@kDWgK!ld$Hn|p&7ehRTx7)a
zOFP29HoS074TX0?Q=rebZR9)#bG=ckQD_Lr;FSta32n8=e<o!B=PU{SL#y4NY#Pkp
zbp7VYQ?H`*pMA-^c=9rQ4|J<}^*e@??^w-I`&C)y8Fc1g_v;d#Z0WID^-7ky#Y2K#
zoOy5FoxdZIx$`l)*_gd8*1Y?0=LEVm^^TH58y_kS2w~8>?jU3_0{7qRr`&ot3#Xq-
z*$KDWM@}Q->2z`4Zw82jcRi^{x0wL#ONQoeqYRfT3DzMEkV&Hl@*;hecKvQFHunck
zv_MCq2jN)~Pf=&SgCvd?TMBRhik*F;OK6w29V$<+rsN|9K;NzoniG9O*xjAwENI->
zO@js6sz(ph#^d)V@@D?t+Un#GWO_@T1=*$p&ed$B52gXpL6;i{y0G8ev+<W(A+ccg
z1iP7+Wm55WrPyME6cg|$70)j2(LOldHeMQAgvAt!drtO#d8&o@xZ)1#xRUMAKAtye
zI>(1cu0Q3edGbLv(Htmo**9zVdgch!yzXo<Z1E6IK%|5Ixp%U^6IZU?`NLbt|IW!v
zBJM)D4MzVvTasA_P>MgbR!t0_?f>ILWRy#IRZS#czZ;2U>X&)<-GNf6?Zg*Wk<c*|
z)AmA@5GJHF>lhF@EDK=OL;vBZm0Ql@6qxoUsHHH=jMaLhz8Gf}@fgM22W2sPePd?$
ziD(?JfSlqh0o0W^^GdQ-#X|A}YudB?6WVFi{Z;wj$mKuYH{Rny1HZL%PAvoH)=pKr
zpfCWRd;TG($@YfWz*}h=I9Sv@e!mUX7;*{7L7`v;Gz1Q+v*DnAgvtn$s*;B^NVtwb
z02N*{t;cObSj+Nq%CB3*0NtW`fOlz^=akw*_ytaM4$m_|-M2Z?hgawotdqGKQeLVf
zs_U^W<fgMbkxv%W3chh_a8qu9QBCWOFz&u=<$DrnM}U%TvSfqZjU+;Ys;+(gCfMF)
zukY`nGZeVRDxB^2WFOA*ft#kQFy{1Smn)>fj}W$vtSJO8yb5cJMxpAvzVo!xIwrVD
z&II}ure{fOwZ8vGG%X4HgOIw{Z&7G|WRZT$OTif^oCYzb#~r5;D^Zi?FLiG}6MU)m
zMTDLidDYdE-Z*Q8!9zv^v9|EV#(PE-ua^!&zBnTAk|{8jam|b@q77|p7A?$5+(qAB
zP+GO?5{JlYk4*u$TXW?qq%JrNUR`bBb|>6`KvDqQq0RAM-_WZ{UiA=G`6G^?UnIeG
z@f8%K@~<Bzw!qg~DCTsiuELX$cMI3jwTSJs#|MjtOeDvolRh6Fu%AA$>Z&>ZRN!Ol
zr+oyDzIc*n!|9~I>2~_-X2um`07v3AkApchUb7P4w!y!{2hd8g<Ezj`XYuUiU2`du
zS79S^DMc7H&*SI0B8U6=SH+S6dJi<(6T2N<A4)NRG{FbLU@AVNw5gJKN`Kr5IT9FC
zW+2@ZvuJYNbOEol43CEnSfnw#MfMnQ;E)CMRP%*2JW2v-wf?PfiHgyJEElTgEgl~_
z!cO)1405aZjUb>S&1@ZTnGEc}I1)Dq4K}(8cwoTq1MzNREU50F>q=j^J?)fm+K4Cf
zNPIJ!(R)5ZbVz2*?b6l-VS47lh>o%L8nCYZ?%)2d6bf=^X$oX3NnHU*{jt5cN>TTt
z!b_LPWUEuJGb`Cs-DyvH@%*l}mzt6J08|k+n<N<(*vr@i^*yvzQP+TQtOMr<787{%
zZjX>jMcLi~rRfaR3gx#|`$M}aEfjw{Kid^e%qTFA)^l8xAMgRqxIaC!gd*cP3XxNb
zP<0r4y}S37l8?`KwCiH?Bw)DUHCxc~kRaetIBc(REM+riU@J+$_KQF#{H=<HGkZR6
z)vqxJ{MssbMV*sCO3;dsIV=iN^z+m-Ej-cqbE5^whIP(%=$+~cixp>Jcf5)XddyP$
zf0gqcjg@!SK+%pZwt#R`Wl92Or$uPlo!4rYGPZ{sFz$i!4IsT%RTf+OmiKeHLHhBq
ztQ;Ez4-pC9WY$3cKDg`CjtFKgUbNA#E%_9!8Ow(meVNhOj5~}0*1v}YgyV}1uM5Sk
zCVM&9w_b?0!HbzBcoHSJ;uutVTzbD}D87FOPuh7*eT@sn>Bed!3$SU$2kVsg75>>Z
z6R>fX$<31C0SEe1yKzzC^<c_(ntM0<Y{#>1&9JOj6cq4({3_s!MFd~|*8;vsUvu(@
zm;WjShSj20HTnCW<sk5|?!_M3=xeRr&UG=Qz`U)Z2OcugRkU#ylb}@})D1*RZpSyE
zThqf4UA#5i4pyn<wgEl)coPLKdpYzf^C_U~h11*I%V+*~*rujAwLpw6E4<A^bINAE
zeaKbj*x=$FJpO~VCAY)nflOjtsr(woNIw4+B@MnEKM*~*yth{vDxHfOEC?9f^K86;
zw?D~7*Vq1r+|w;`bWt>-tCz0<?Q%o~H%8A<<4J6xfJKx&5o^D0)m<;;gftvTJozmT
z1%l~vjl_Dr+Jr97`2a@JR87IJK?`Z%=<Scs#=XbFhwBf^u&<#(1<k;7$NO|%VXSOd
zYhQdU%~phF=r1pTuN^_ub<6VDE1iQ3<W8WF$u%%my%&j&13}U%HCyXlh+?L$md{s$
z`eZq3pww8Vb+=oUT<9Hc|2M_vnJYk}Jx27x)f4Ipx2m%fDbMy5Q0BfST4QPKp|lm;
zQuUWbMogIM0b8*uJ6OSZ8C+!KbH={5JFf=(ff+;oy1w+m7qti~UVI`2<EQ|t0}Igo
z@{E*!18u)HK_cY7O2W(FI&@~>p1~Jrj*)9zb<gm-_dvQE$$d1T{2_+JyX)E(=yD%%
z14@!~it@{y$YWmx{{z`oXYX2}@UqQ+>oP83?GxytZcl#@xJQl{shOZ&Is(ixBlL<I
z?CYhpm8(;X4np@=C4NE7!f=V6fgOqCXBrIRUW+s}HnkX&JUW(s?;lGjt7t(ZHau}e
zkE>}drZihRZ>t96;Vi*(@!ehDb#oBAw$lL>;Dp%LvV8}zi5m31nJu*juvr`9dMTMm
zD7=YJRra=XUG3f1vU{Kot?K1}BP-M}rVfWQBj;D<+T*_cc0|2RY%JwK9|v=ir@DD0
z)SAszJ$ttbcFOf}ks}pH9}N~f)~X3^*RiPGJ!q0Fs<?ik_8`x3gXr`pJiyHvXpT-d
zM#Pn4u>@kj%iVNmpUfn8{xw~rW?M*z0pqHE!t&NV-<6%WcZpZJB%?jtxyDIrAyIED
znAx7qF*6=N*on!OV#zwc?C5ouPm!G_%_*U+Njx_Pb;vOACquq5{CeXN-K<iqT#6$G
z8u*S=jolxk92)_fmFGu$xlu>(C!;lEO}YHH(`M>=Cngz+X8f9UaMT1JM>zVKGpTw(
z04@;S=O6PdWp9ca6v}8jq@RgK6QLC8EmS180?-dqLnLM*mjE;4`vq9ci_jH9Sa6^_
za^Jgo*0cKIIT)K};u$kN+N<nm&Iyr4HIdvm)&KfK1Rzbp>_?e9PO7l5*90-z(7=*P
ziKlGfIV0L0Y%%^+C+yRlYauB4l`UK2Q$2uVdfvM|2kiAITKu6(9!!dGQ<AifLp?S9
z*o}R3y&A<=19~LpR%nhMV1T7~LZ?fhrzXAHe5dALLk|Y;g(z%p(J||EcJ@X3`&E=(
zOLoHnJ>pCESajMLG^a;nO8zaP+w$UtHtPc4$u+Z7Qe3>YJLz=(k8{)e&+|{rmp-__
zqA(7L0#Y{CLig8yJ)U%};ivowplIxy&VumiTOR$O3qDmlqh;wohWGBALkK0^0vLP5
z?)Z4e$M4Q<lv*~%d)fl7Wu6^gg7!VFOOohagt$F%o7N*$eeeZ+d(+%fHdk*O00{fr
zYvuf}d2=+o2I!1PkHTDZf=q7(hyDDbw&aJbs6FOLv9x%;#n`o6lDO3V-=fW|@0#dQ
zV3r>g3(kK|k6c|7HuGSJp3@8>G8VXxScU*?V}_M*GqO<oQNNm;(iclhI?q07xPzYN
zFa^=s4y5nHRx5CO|CQQZga`pf>J2P6^5&TZV#uq$ukS2ctNgI_851Gpm${Fpvkwr*
zU>9`{TzTfHY$+6FT=BRym5sP>%lpnK{IzUDDg&5!BH9#(N=w#uR{kUb^Sk2zAU=b7
zz(=KN{)!C;E8<kckG3UtzLAG%*UnK&!%BHd6-#AxMl$L1T;|K=GvKi)?8=!dJA1tV
z{wm5-n3>DB^;KpV>BN%fgR@EZQB8x-9BW^hZ90KI*=vbhpyVRt{^J4VYM-zRCTU5E
zt@syTG(@9x6l5pK-=;M>>~hpen|QGu7Vt>ZzcC0Kd0Xd8Go$<SR)YOc%X@ESsIHUp
zx(6=>pjsFYi`_xC#39c+_G~Kv6}=93jlqp7uKdI)`>kt8jjqd|PmO#9wi9MdM;+vK
zv7N}D>qK?4V70v@7`9%7I7#0b5k4P)#gOHk`17BEMsAqRYhCZibvUYCxdL^&ehc(4
zI*1h1H?|?cDW{gp<}h>tuwQ=O%<&V+?%CXaJO##gU)aL{cQq@{Lh}qs6Mc^@$>|Qu
zG?hGO!Jo{UKmXgW{6}V3@f-soW4QlHdK|sE09CQE>-SsogC1d1u&22$0AM(Z5yj8h
zxSCR#7#Ss%JRe@=8o?)hSrakJZ^UM-()mLCeSR^yRf75Y%bkl!ZEN43O7zrkIKKm(
zkatg!kYEw3U_z4f6)J}S_%Q7Jbo!`6uV22}CV3|D<)xUAKqz1)PuUc7?_6Nm@vbuQ
z+I{UFV2L!cC<UhDj9xG4TRZc^q=>{9!M;qMH&Y}wF`m5}MgOocCe(Ng+tUyBb-@86
zMZ|1FvZO7!y7CbI<*_go`)_r)2`9dCSo1VEj;YzK-)H2Vv`8d#6?$huly!v$<tuUY
z<5i!YJfhSZyd_cPMmi}lBMu;kbX*yU*UG%+rj7@L+VJRRgZXRPIADaLko$gy<Xo9m
z$h13ZTA3fV8=V?HjGyWz@IJh_b)Mm<(hP7!;GK~+3D9O`Cb8Wm6ue}*o45o8w907Z
zj(q8K_1OS!D9Y*o!jc;=F+Z2D2N31{#5g^UsLrYR6_Xg+Y@X(m#?Qh(jfzjC+zrR(
zp*hM^<LBsXD(C*nGruJCNl8m2<Jq!;d5!d&uxSJ*|C<lcpz(gO6kvj;JN2jLNd^Gk
zMp&m#{*U{N!7_wryeCronYGi!U|c{g3oh<yaWFq|7>VXt5(xK|;MCcX%P(fi1u(bd
zqU;8qvnmb4UIKz5JIUptvgmNTsP_%V0p}RD(SV!o!@vunC`wnMg|F8O_We}bfNg6<
zs~+Rmk1o;4$iIY^coY9+?J@Mg3LDz@Brc5a^+pv-tQWR$rZ4Ml0`}@syP);nlyC>F
zg#@L=1)|k#qP%fB5}+2B3G`yIGyFXdFF^+j;vv{KP}!bac!KstK4;mc!FQyJizyti
z93<{P>>u^$PId)g7PEBfWzdcHXGpp1txI?)A=YUMw0mQZCBv1i42GX7ukAP`b?!)5
zq`0oH3#JI4A@@dwji$kb#zzp#qe*ThfBmgz%r&>O<jv)%Hqm^kHBctK*(*}%4@uew
zqBVIMcQs2L>Wf5y;7J>c=RXbX%hbPq@9<S~&2RccW!%6}($6Y%-?fi69AT0P7HOZH
z>vtj?FAb@$9T!EK^GBb|j`jw4=m4W3pJDN$SejWS<CDkJW$Lm(agc3-9WUc(X~;h^
zWZ%S%e{mcmI>IDAlHuqD?j^0!-iKkFwT}nK1t-pfn%cNRuNmz!AA`1wHYiV(pgdE*
zkl4A9VA+H!E=WZVucVVY1q^jUW3`RM6UyShew@2ivP}+&culk(2C3BImDH0f-{#j*
zxRua%#|seGr;l>oV_xVIuGeT`OrsBG71ewm9~TGw$?57Q#Zsw1u+dU3;z$UoECi#Z
zU?3vqO-OsfZ+dstV%dAV{IjB@T5zca$mQCRo0u;}l&6U%$mawWc2Ywyg$Y*bF0y{w
zesBE9={v~@Ub$ZRqQim32_q{LVL9%zve0v*vR2Y23icOyUi7QIRMa@*<sx7^WN6AA
z4gm}okpwgSLd_!6sao%*;9yhm{Ugq(POe&J+A~S@ncrD7(SH^U%u+ZK(u~*R-@lbj
z2x(ZBwbCV>0YV8%yIF<8*Plj2OXRN!sGevI*q&&jN1tjc%UT$0&Szfx1ha~xGM|1^
zeZQe2IHM^Br=<&^<Aq>e^O7Y_=Z1fV=hV$}m8st9`VlerR4bPuGX1j2F2_XjlP~+J
z<6^5~XS1BcnP9T@>3lK+_Pv7HQ+EOY2SpA-`B$w*+I%}JjDN$_?wKr-nb6&1Ea@Ee
z2ro)am@?0G1=s~{Nl=VYzO*rz+&5b$+EZ1B&nAXPbrM($E%r#)b7zN4mw?{L=zc8X
z$@aRS2lMwY{a(_UKVgx|RwCiU9hyA?H9Yv2Eu9M(wN|JF6GjD%MDMaXM-EUtHRt{Q
zK#?D}!Gs9cw@L(?qbaxiFknf{R5h+@RAN{wCFD|H`EDLj{%$o#btrJX`RWN5%uLR8
zZI|9@F^=sTjf`ut;!PBH=xhxyzbF256#lPg0_ySQh_U+9w&R&+3IB_6fThp{R$kH}
z=8^IZU##eBnuO!**t-ZE-LF?)Z6}5({Z=XeCKNozR|azd+<0!EIxv6@+886FZZCeN
z=^%*zh?m>J;=)pNDtf`XJtU;<x8LWd#L3terF=s93ugQsdT?WUJ~su7oB_k?@bkCO
z1J`i}?h&d9>`%pJ^jyyobgs%M&;w7^z$y1ZGks7uE%D!~WtZ=pR3p|vyx=nvYN1?@
zVtw49-4Rn|2OoluI43DbwQO|G-^gaw;XY3$G%WxZCu!;)hsR&-aiFe>93iZv+AH=q
zoB^V&2flysoX24#h(5=^vZ8Loeio$j(S4vy@opwhA+5-5y27M?QXKpH?()x9lViY&
z<~NcBY@rrlNkqkSW)J3Tvl-E<gb>3h<q~Ji5YEWsGpGj;TmJZ>8H!|41!;qYqn3$f
z#0n(8{Y$i1k5RUFl>+{jdHoB)GLa!g0bvw5G_#nZsssDBl>rL|w87j`3)cbv^uiD>
zWWW~+$v7OYy?8eRCRbM=0gDXm<hR`Ae+rHt5sew)!?68y#*pFK#aqrrP*xvp;V!&e
z&H-};gi+x&0}?M8lqp}iybH8qBvAFX<amSPPu)H=`cAsZ6w)BByWUy)n}7!T9~tC2
z7O|SmTLDKkpZ7ysp)@Z(hw&Gb4FgasJ_`uOml;ofgf@eSEi!P&mO_MjrG&V_?DLE0
zZ<7h|Fa7oNl!>2X-aYqnal|?l<dR=^I93Qum@V6rgRJk)yi?*OfV`PT0Nb$z0(A-j
zjlII#;J2%2Vh1iPEp0Nx@8{XC`~HR53G~Y$P_I7=1SsC6GodW96ChcNk^vm)3i6GB
zmH@JU`CzZL&Vgz8U||cfWH1m(OLG^u-_VHX5BbZ5<DYBGgyovhm=Q4yV2ItqHX7St
zK;CvDj_=%m3NC~rT#ITL?1L+c&&??RlCUR~K#RsI%;?&GNNp3`u-K13ZG+IOr4Zn6
zd3^%3f`OE12oBc)z8AhQ{5a5!Oe*$9_WcH{{QXf%f}Xdnm84ybCHQLr0QN=^^j_v+
zss#V>UeM+EAN(XJLBAjWP~ZFSkN-dUwZH$)|GxkK@2~%2ShKM@MWF0|?4kdIhv(0y
zAi)-g0jz(^j$J=>R1H|j&DbPTZ&{iM@BM|?{|(*$&t6y(1_J9Brd+UAb>B%EQ(&}o
zOhcSNg=(Q#i1pu|B7Z-Q|Bo*M-`|-LfO{)ff<;?$)3cr(T?eQm#E=1M@lBn;zxSs9
zZ!a2^uNQ*Uj+cNvAg$!G1TJJ;3@7RTXyN_WPnysIJxZZO#=8HPOIaNHeb>&19RHU~
z`G4R4f3N@lK7aqa{(jqk0;}%LfFwrG?#;WP#;c0ic>O)6)7&iVKEc`plk$O$(Rc93
z)-Q<vEO7s0pEsGqe=zGf_vY!zm3`Rx5v+;L{eI>g8Qkk$YCIHg_KE$CSA^nzC<u@B
zKRWoW-N{O91cW%UQ!ahSI`_B?bi?`4ITmaaq`n4jzYUf82J4BRznNh=5x&R^H(mJ%
zS27-+bc=~aNRvkjRT-&ZO&{x4WD7QwurO+Ld{!9=kN2+d>)zPFMTAPPo9@+}Sx0*u
z|G#gR3EWuyS=*t7lF4pd=C^GweLCVt?BzezLXu<(QKxAG#>(Z}{t_KZKB2nd97IPN
zpD=~B{MW^0CobU8kwU=Rr|y|LAa;ETm_0S|H+RdRf)jk3TdGF2+aY|Hpuy=@{Vzt1
z98T8DKRSoL<U9-Q+#uRWiz(K4CXuRy1IaQ*JZXpU>|5p`Fdp(UHiN~;^g;c1oPkS+
z$ZfYnNdlXzqU7^2cu6ZS;52M8{rfK%(M&&vk-8X}vlX>aE1A9b7LF%p##qpWxpO^I
zqrx@Acj73WcXOpv!O+KY*S%g6NL=m3K!R(tBZzq22BQvXsEnV5h9+K6=LGQ?p%S=z
zftz#V@3Qp2eX$n>jc5#kfO+=EU6KfLhlThKf+Vixy8TSP%Rq=^U#oQliGJ~+QKWAL
zK})Iv#|G$mU{F4~?~!Y&f!|K+EU+cr%Jh#n^dJD`vZ=i=Gz0u4%GrW~B?SDz$l?jw
zFqYZ@D&J={QcngR$v0n>dkN$7Y9g9jXt_vh_lTCJ_8jO9LZQHsCA>{WR3o05E;Lw~
zW#RKy6hp>n`26K1_cpZ=FtUfY0WvHKxhx1cQS?lo3{|`YqwrDxs2zW3>{)xC_Bh;1
ziuo4_aW9?m2ER5wHyHY>&lcozOFCenKEVmN=Y~QRJ{ruj2!PbUSI6?bMcZYE+$GlQ
zm(5uw7-+Jk@`1YR`R){8CyjFIwK1@xi{y}CIvIfWo!Uzo<_xb|@May(QDo0{UiFA{
zgP#Trv`QNue^%;MuE#(Kn2BLrs|z4i+|vQB64>@Xq5cbWtmWH)vo1~trY%C^VGnEm
zX}kWL(x%bxCcix)49b7SB6NY6EZv(>XINQ%28bAb=<hgmZyZ_xC4Bn0$1{zZRfjB2
zc1^eUO@q3h7|?p9nY9%1S5HU+|BPk;EtAJCymAEE{>%8ghAaG+X<Nf7wG9NE7G$X8
zRi$>m6#@aUf}@R*&e=I&Y2V-m9Dt}^d1^xn5x@lzuH+C`=y<0{U+(bLbpg<X9WOy9
z7@6@v6)h=+<WP%hHfEv!1z-mH;Xq^~YeUM`c)*De4e*6%hM}NwD&u<x%~NN=Mq>J8
z$jCH=@)zC%&7)*wLFeom(g+RW08E-P7SLTux7-g#eX0OuS+Yq97WUHnz<}T&5rdd+
z8~XeG!ZK*kGs+SIS_)7K(A-IUeel(EC~II`%w19;sJfmJDXgq5)47tc(xATXyKc@%
z3kYddBY2QR|CDOq)Lo<uBip~1v;Vtx^S|B%W$8d04HTEvWtPhGv+45)LN}zZN`QIO
ztEEvIs{o1ed2djoG9+B80+sa~&eK6HDRw@P<2+>w2n}zvOaKrmJIEUClz=mHje@Yq
z?bYtw6Ukf>!Jpe;*pe;C$$e^CW(hK`@pW$Ix;#eZq@~$H3zpv<XM~10hab`whyh^x
zl=bcFSjDFmGjC$GSQ8<XfA>lPPUdID2X~FW4u0AIjdEbfoiDLl&r%1qGq<(BXKt<<
ze9?A4PPG}Xsb>`8X%J$mlc~2dodg#_T;pAy+kaydCC7|)wYbTYPT^nljAsOer&(Km
zdKRzjo|$VeN8bl%&N&a61TzLf6f}?f$yLy<td8p!Mb)&R?vSv+XE0|@q+zA2wOrb6
zbnml*G$$kwdXsY!%7Om$=m_-n;uC=9Itz_YQ9~8=l)cnIOM0H=;X}-p?DlvB%rUX4
z@SquF=e-O<nA6QICly%<;M>u{mB*xY@pIUpdl@qU7=0DIqO#-zDa?hysObQ`w;}J1
zhR9H2z%tf6v9Ctgx%r3&x2cpqLptZw;nZ{LEdDHZ%U6d|-Ku86zN$hRXCgxy!Pxu5
zp(_2%^Lohi`~$Cv`GTG%=Op2Qkl_U%J&!5a^~)Fq@3$Um4NG~P>%!b<jq*(?NA#gD
z-5T9ad<*Si(h+^%_C$j*b9ZTOWn`*LYMue!r11wv64#NRE!T~f!0u_AMuD_hT<+<Y
z%H^$^rnA{D9HZ+(&bsH&z08I)_w2R(G#rJW(rPlSxH@{Z(54kL>c?wU?RtJ*XVXOa
z0`Ob%hc*i=y-$PKHEDX`b@dF4;iaqXV>a}AUIi;92{E>Sf?>KPg-O;kjN$dmVtq}F
z7PslS8^(WXtfZFg*LsObbul_W33+k8?;FIp8@cC87aAMky1(x`k-qvabPFTxkC}3w
z-WnYIF1(q)r9h(<K7#QUM{Q*(^i3^}t5v2r_9Q18Guvn2688|mG&b5g@N6FHLe8ZZ
z!n*9vFkq`q=j)i?-5|R6mq^nD=R)rHK9Skg6AI>_9~*Hvi&%T3aSB^Uv0*1rQE}X-
z(~$j6TYLioR@Noqu|KY19;x<N<VUh<CvR`Nt*btFW`=Dp?n=&FT0Dxvb=bcJ#RebK
zAB0EnEge0RxZWM-r_6S@byDr)1ZFjHsyEcQ;cUaPg%5VzRS3pS^7I+E8LI{3K~>ZH
z@D$=~Ju$192_i}(#Q+rr{?RlHX5wqgiZy+Z@7cQHMb;Ld%!T#`a`5ms-geO_h{Vke
zP6_D(H#Y;aUtFUn)t@y`rvS0KmOA+?C$AYyMOr^M?i9z@am$qK^++ByCq+hUgME6|
zcly)_lJEME^CR6Vo2Nl$JzYLfa5@32E?kO7BWYg9SZ!k4{3^@Rdm)tfQonJ8ewzOC
zvEIOke4(6`d~=T={~u>1uREy8tpMtVnqF|^n9<(ielnH{32D`<<^iueg32{@KZC86
zXW0{IBdoudqbJ*bxk6Chz&`~CknhYjPo0-%GkCIB+0a#O-<pBmd!?d;v8HtC<{}ux
zUE0T;C>oTu=YOG<I}-rLAS9)yP*%2Gmo>Gz!~^;lWMe2kzgePLke%(YPQHx>Qn{<A
z6p$Xe`5{-H@U{U>TjdhO@@oM$@idGGy)9_gIKBZWs%%4HYxy<2QV?4UP&|A-ss*OH
zyx-l1j&B`@0RlR2G}yXny0(BPQRG=PAaRO1utdnuLG%s(vlP@1*m;CT0pQ#g;FNz=
zodjTR`&E)KHYavp(@8;he1p-U6KA-u!Iq%E;3^$*E>a5EqRqy7d&;hmnNo)F|0=1#
zgq6QLQIt<=mOu@^R-1dl>(}7<*`@KmcnPNtmiTWzx9M-x0@$C^q4m2f`v!02Oi&rt
zF`dnWy@ADA1BOdCAW?B36BP8a4bIzIgoKs%BTsdsm)LKF>})V&!}T6amkk={R!UM1
zayxwu+<-8s)KBR6WzmMXdx$4_*k_%QMd^ao)8UH~a~F>*K_np77Mc4Qcu+K<RNkw=
zN&cq}TRXN^W0udA>k4(9>J{g^@eSXKh^INH3c%i9AnG7z3c0_-->kvCyUm|*Fpr#r
zk3zSGXVG2W3!H4dyjC?*X+!{f|MPZ_=q}*auQeWVA}U}Lj+a-S&L-5HAE%zbK1h;^
zX5?;BS7m=)M0ty-<uhAp_NGfc<NRVv{~W@({!A(?2>3W`fRnfAY<6;q=6GSe3q_aH
z2H0imF?Hzz7Y-dK{I>nZ6i!+gF{GzDfN1NFF84ywtuQ|7?K*os0RM{|(nvWo{`RB&
zj?6E3{xo%;Ys5dxVJS{zw#!jE5ON`#{Uy!^&&M3f8t@ul?GYWQb$3ga<QWOXH%Yza
zy1Y~WCk7irrT1Cmx{7<UiJdp3R1&^%-sptwzf`rzE}5IcO#7mR2#E8Xe%!{3h%n_`
z`!@2LX0NFl*U9{%s59Ju!`3)#mmcQ$9@xlruw>s%{&xGU?`l3HpNq=W?bQzkLRis5
zk&*fpY&?wLm%D~<l0Ry&bX#Fs3Yb2fqW1KMq8#tbGBK7UKKE~AYrckTE?=OR<Pi~n
z3{^(v(vIY8Au69KMZ6!6Fnye<P${JpxlK_Zd-4;`Yj-1G&`x8<-2161AuL2?hH9vT
zYYGt457Byf+5EO%Au8-tB86c`c0m=M4Su_jD7CPobq*>nYb}z~n<+?$5})wqlItGI
zOb3jA(_b*B!V7J)xsz@Ng4Q|gT<QvmjZC96Lv)4MJpi2E%sNh(=*4CC+#mt#!R@Jq
zKc^VGr2NS2=HTO^<WC#3v`UHP->xW8WVOo(U(wj+gi5+*icf^(_*7XL3=QK_y`G7O
z2cQ$+Q7%g7qtkf8VW)rE95Z3Azic1xE7+uCO65SLf#S7RB7kuk#vdlR)h$VHy|!Vg
zD_vq;=7B1<_MmCVRl_ugf+#TwexyS1iBGb$KTC{a?_oo8jL=+322oro46UwJH7?ir
zvNXzLE{jJbGqZgn<_2M+3A^34VFH%<ZQ0;rev3CFqAOj&XZ-R%n`5fYbWoY8=7i!Q
z9D1#_<T3{#9+O(ylvxYc#A92oRT#a;G7V2Gd_5jRL8LLm37VrwBMQs2N@yfv_I@VR
zM^(a(e`K%aHOK4yT`bz(UeGNc$-vxY*)=rUS7N_#h*4SI&hh$gYeYW(VjfRsJFuC`
z?^`L9nudK_8t@@P2$IZ`KXH3COo0-cDhoxncmFEX?3fa9UQ9vWi4=Zvabo)h@`ABg
z%iGcqSGpVfvx0RK9zT8>VqDKs)!Bkm)~b=R8*{aR!@sRw(B{ht;P}Xhye;3y+G1Z3
zVi_8dg+Zn1G>Jy#%An1Zvqqy5W0#M^!?7KGIpgCFwE!2>=)fY?gk5X&;%=QlW@nJ&
z@sldLgRi!etd3deG2eLQPX#03+%TUy4tP9NY8#_n#hYn>N*TNJ1+~F^Nn?nYy;rnu
zMP=t~N#E>Eqd<~|n)EGy;S|raL-O{@=?~y|9N*Dzs|fT1QheixgAZHlP!ZlBnngj^
z5V-K-s4*t=v@Jw@8t%Lob%*|FT<vOjx{a;rZQA?OfwEI%V^1|ABR6^kySOzH-Z|6Q
zAMWuUGYL#7%t<1VO4VJ{Jo=>C34;EI`&@FrfY-oL`9`WQ6#E2M!!lq7IW-P(7QUp*
zr(<)wLFLxKG;_spxMP6yW2$n)#l?1w<?-J8qT{QvJbkauR68RYpgy_Cc<f04e{Oho
zwJLhuN;b@eNq^*8f49`2fm^t%N)ulk<F2wk{y0I)lcCqD1h&Wg25$``-jP&X@<^*4
zZ*E|B6nb%UhyL@sUgSgm4O?B-B(C-Mbi{)BU*gVspUP<3q?UTPO;k(U9Dg>;vlR+^
zesU|ltBV3NN7n&N45h|O{JMRxavy)d4}Aoyrocc7i_e9gfDuMq$lpci{4TH~gfpHK
z?VL*Bt}lLtt?l@G27$LEu;@UEKbeI`itg~6ey@`3?~dTh(<Rte!d76#8AoT%lHh1L
zUkJIs+Z5c#AKJ`F{zUmnn%7&^o;|vUIP7w11J*<ElU;FcgULUYdRJARiv$PWitYwV
z#ig{gZ%_;TiR6Zxu+n4o<Ap|S61qo!2-(WwK4fP%{Mc<c7*L6&qC3kMfr5ad7?KlE
z1=#3`E7;&<@>1|VD(3HU4$&nIpXzooeX?-cZ6(uxbTth75rfZIe~OlY+fRA_BUF<d
zE^fe>T*(I!C)r`#F|#qdF*AxyMaR`g+(K)?MapC-&DbxeR%RWi)}>qQqv}(e-JDOK
z+_)zo=q|oWN$6}jT5y#$;{~(S_YW>K{u@=gD`CeroQ2t#biW|15zqR)<a#Id;tZqo
zZ?F5Pk~(cN9bcNUfNGXA9yjcEclG^lwucdD-kN+H_E6Ct$9iTbk+s-5#23*RS6#9M
ziPodE*H}v9yG*;#JX^h}F#c^E8>a3?M1dcY3ok}fy&cIwSgf(q{nY>&bfZL&fto=*
zwNG(u!e&LUz-7RGH~0#*H&U9Zdtk@#)7fkQ(3>BSvZzfJfxf+ne+_O!x^sHbHl_`K
z30S(U)EHv~dMX;|E;fu0+1F=gz-1|#A3fuhtoKyzOPA8Z3{_6&Z-s`=Bg&84cQ&$l
zLa?`Hm5g@ZeWqAlnQg}s?@aWt`)UyBmtk|aC|c?#$=iSFF9V(Np}gNB`={M+F#=i^
zMeaTkJM0^(gI8bI$5LS0I0AQ4O)rewX`);?OUjJOHi#tQ>WDA7<TCjQIfO6{->2q=
zl$gG4vDB{;>XB?mj&0|qM8N7)Z}TG?Z!Nj1*!9IShwdcirgjFMZO&n}8Hkb-13xq#
zEMUc9&7|MZt6*bcR7vRS?PmHIpyDqMnBc~ob`>``J`OW~(D2OI)>fZTrpt_;)(LIo
zycf&Gg&sXLQU9YQgVNL_!X!$bYDGn+sJfu1nKAhh&aK=#50mYUDOw=3roXGsb*^c(
z2joOfp$dB5DD(a3BcLMbKzt@}(H1P^**v+L)g%!0TAp1?w@gKWCh}hu#R-AQtv4bU
zyYxV<phK~cy`kH%b&kHH9;E0VbG%A76m3UTH$wb_H8z>+hjK-;vAE%a^!M|^!&)@=
zLq>M7CNg9YYh>&IyZIA=-6eI@9robni&}eG&EERmOm#h<bY2YJKA|r!vrevP2FjBB
z#%Ef(rg@u9bO-;3w6_k6>V4aO4UkegrKKAYX&AbsyE{Y}5Rfiu=`KM!l<sa2kW?Be
zDd`$YV1RwkH=f_~?)@Ie8~ZO0WHGbW%v$$-#d&?s@TsOrMb=?We%L+d#Ql$Y3U)LZ
z1qa#LVNxLiug3C+>Z_t$kXLMUv$y0abWTZsS1L*SuyFW(iwiM5=4lH}l7P7Z4(<ho
zm`S9Qs;&(_OT5mx-RjT8(9jvpoNpc4I-g6`DrqUMsuWLtZSOdpIwbj3*f(z%4$Rl%
z9OByT*YNLX&NT0b^*B8W--&!KbmtqhhDdQ5Q;U^UcM$D!i(t(O)6x3ysAW9XttZZF
z-q$7Fw)zc*@4sGz2u<<NGmF5AuiM>xAcM#CBQntr9V~EDRRWe1P0Uwc7-yA}hT>wl
zp||amQ)`nI|Hg0G;rGz}bl8{$2^oMg!(q2&IKYt!eny=(O{xB6;bj@gt7lndcyQ=-
zH(Gp7k-9E}w*hr<M{aEQa^Ms;4Vl^>uUMIxnuWE4^6eidFS0uGYx4|u&-2&Kf6t(9
zbfSGWUZZ`7@4td%1!fheXk)D_z9qK{852C3sZ{FDr_{fIpEq<wf!piDucw5%mFTJ)
zU`YIrsZ)SIDPE3F4@f?i#=YL9c4sLQ<3aFdwkg928Q*FszBY#je`8Xh=e_B<GE$ex
zVC3ZX2=T{HIMLC4`ddZCBGwUsY&t3Y$6TEatuh|xD|Nj%S)u#yuG~|p7R@P<X3FSV
z#UL*bX%?icHT``vrJT2BF#XhIr858mfDQemjw;FEadiD@bCTDy%Lp06>?4GYTau|u
zo2<UVoq3`Yg*-vxHYZDL`B6gxlt8mC%GOZNd6K1DdL_F0v)<eHbg!;(fs%(|H~#tC
zwy4e}hQS>|1NUF230i=&keUn#2tKlj(y9k!S`3Y$&lu-j3fX0zjvw_2GsHx&H@`jx
zkd6u>r0FLs%7)G*Z(0Lec@Zh0*KNqBS|%*4qLL$UXCctODEx29rY|UugnlFAYk3su
zpgIHCFd@iWIne7dy5kwo3ZRSj36XzA66dEBKBqs9=flL0foP6|pq1FKE996YWAHEu
zG89+lb$Q<XHg$~@267Ycb+f$s<=bXHu<j9=RJ9#3_6I@a*EIKCXD;%{+~&R4|D3mU
zFLMsTYU>iOL8NjiJ3$}Zkr?vN{nba^82|$eX}@1Tn1dnsg0C62jS~$8a9&YnIimV-
z9QysZc(b!r^27R7)~22&Wr0m9^d@ep_+2dT!spY*$J8Sz3h+z%?PK3I=cCO}5_lWs
z#X^ztUJ12z#}Xp4UV{P7I&U3_6AaGYWWbKBWd;347f+1Zv1SU(p~ka4BdQESrS&qD
z+Kt_m8;??)ZtQT&XVCnT^Jc>6$2N~7p7!}b)Pf-uC`KLpT2m>iHWUp?*awZ5#7Tvw
z7j5vk?PumD_cJ<JwyQ?GRLhsws>zZMOfQND<4+UV4e&7h)+V7bMD`UR_sApTv`o!h
zYX~E4P;$Q#RvHs@ylN}HcsUa2H$UOl-Lqok*~SjBanDRHbosVRKZX+>n$-77$Hg(_
zvhecUfzT9ZOrwb#Ydw#heN2Dwb4G>1KBEUzzheB<rEGp84U5CN0%LT3_vOWn(bV<I
z*rpJ!@jPq*$4{@&h@-8&@K9!uK8n|8Y59lg#n7S8mC!M+@|C?;>%yyw6N5QL`Hm27
zuTQI69#n0<4!)Bt9R1C;iI-D1=iGvK-1`eO6isPTzo=<v1iU5=?7I0=H4`cUj#Q)Y
ztI?^+ydLrk46-AAS1UtW5r+i+4gI!d-i5akm1QddOciJ4aW@EE)a-O<SG^p%6*(H+
zz>?34P!hXtA#4PIX{zObfyW))s*HU$xxo|=xnok40a)2&{@$bcbN19&x>U<>lX<3I
zFGT6HGOWx*-jhO72zL`e8%QS1t`{#B*HBD<YH0z=z$k{5i^Z>Y#ahfD{$ET@CWIJ0
zK<yQ0@CEYHGSmb;w2Rd|h}?BHXDM&7YO1g+kv@<q?eUp4Wu6J%udezg49!dOt4jO7
z3!#tcOwd=m=u3J)4U%_~a@lM?L}WHOyo-J;^lGCJ%1>z*F>j*(?ELcOE4waw$^a%f
zvikR(mQ4sIQ12pAP-N13e+o+b%lAF_pIh-W?t#kn3V<VUEEmIlz!XKVZzt$>VB3us
zjK>%UwLH-ZU~&yYR5pG2=i};$WjbsUYmlFG)k{6Lk}DlAf$EbX%wkRCWxH5!w62(p
zhczouXX<m%t-dj(s+m=?X?dxtE}hnG5iJ6}Wmn=4&x?>xAc5NR`fkkqO0TE&(fkf=
z>Qtr4bb608DU^G%LU#%k^k0esnCQ>WUU(QQR6ZTvS^U1)B<DB8N{KRVHIF27$3M==
zwHiV({LE+%%NVp^sw(WdPiFh;#7wd?XL&Tc2)&VN8i@h>f+NTftcMl@(7~REc^q%F
zssnVzPF|HoE4FK&&W!AsSuS^ITD;7JekZ#=<I%<T1zCUp#ZW(Qq^5#DcdEE8r{#1d
z=w?j?aoukGwE`)rV>#>uppmTVKrc5^V$KE<P1f9!j<a-W6e+hqIRLb1SK^JC;hy2%
z+wt>xV<OG@<`JFo`qXG>4~x-pc(cixOU(f3((QRezxRv6SB5heG*2Q<crFLd_8X90
zOzWn!FsxL$Q{%1p<&G~j&Y7GxXcQZvXq0wdtAF-Af4|5|TN?ital-wqbM^ofrYK_3
z5OQ&>dQrjekhe0%6NEcK2$lHv3VHH<9CIgi=@_RaP;b$%g2)>Wu^OQCy&S+*Fy-_q
zJtQOHcY<EQt$g@q11^YvBPwSb{q;w}RLZ<8$39~lw_SWe?X_&oSf|G$%pR4C7Tt*E
zSm*3lLh7%_=lY<jbssZ7*gMs7e;|63-#SuOzYX+UQ0A$rX1lKjJy6FmSmS11xSB+F
zmg1q*h*NGni>z6RA^S?-xt2Peb+VRUnLn!1Pwe-dC=j{&!MHt!eD1=BOB{8fv-}Hk
z=cy9M!5+dFUqx709ZwVF8U)9qS|fy3hgHOuuUBu2qoG&bx85@XD=U#+r(nK+;KHB3
z0L+2Y8i3wkT-Q_EQk-3eLnd=&%FyIB8cLylFQrAR+hJFHw0ha+gM-91k)Du2)B2kz
zvK=5KM+b5|vC^#J{kedqRF2=pv6isKFGCNjpO4!{KzpXNjmhoARM$-+_PfD^7}@5k
zop#B+1k>eK?!~pqux;-#kD|1q8mZu}`df_p@8yR#N3GN{^<Ur7@N-(pWPTH(KBR!L
zEb+Fd54=xhfDVqY**m|R<EX~BPzOi_jC^ukZs$U8Tg1#mzkpuyn1|5XG?ecdxI1bE
zhx<P|C8Er`<r@iurrAzQ^6i!$!ziGJ{xNb-O=NMBfrZ%NkZ>H9u?;-Q4?h3Ea|r~R
zANHT^GdRyArm*eBtZA&XpOR05b}{3{g=-YQ?Qe&;u|)EH$PL!rty30ECRwDILM61k
zaW$Nn*10zytKA^*<;I{_bn5`2&~*DI5~qDy-+&-oBJ1AVC&6{%0qG&kraF@AZO7{1
zi7vt)AsU$pCFJBqe#tfV%GM$o*NumvkgVO=x+lBw(WizoO3KEwksQdsE9|OV;L$lX
z;>P#dl3~|B>l~8$me2HGFx`BV`=UMEb>yQ=zU=qYzp&i4qr*lB#+!wvgJStC!_v!y
zw1(SMXU#;Ae|NSh#cNZM6`o@_>gxNYhKR;~Cc*_bJ)fgetD70S^sut+DdFYpcu7<+
zWY<Iz!!Lj45ya-*^VVWEAD@z1Q^G6MG5Ut<?sQ5><*%+}KYE>~B;+hb3XELiK!YsG
z!**$t$mfADner!5=nP^DW|D~-zMuTs#?R*PV<YhI$*=Ukd#~M)`tlRuZFBwRhJxuv
z*m=d59z9A{C^lY0nq|qHsvnef;-w<W?du3dl@~A^yaCBw51y@(Gu!UrRF2Z<mo7~^
z1jlptXzk}-8Am$1(X)p6B=A`F)<Ct)DV>L(KR?tNAl{e_Pl0jhKkG50^chxl={hv8
zqp|!%^CWM4>GT$%3SbKzA3K{`L7{LrF?|rwjsI+|Sf7=*Cuh4p2`qOLb5jAj35NtX
zcg^h`Mdk}MCF^!(&-8lxo<3WCgZHMd3I#qZKbaUs)Boj4di}Gu;jU{M<zcj<G(jkt
zlWY~}74$o%F_H8iQ)>0;I_hjF{4q!st=gH^nfd8^z5C2Yzrp8gF<SLbz@A}I`)t0J
z>4Vy(myqo%Uw1xVyM!2nW#5LQ6npA}?PD5qS3KFPvC@8nOIbWR3_qqP|HygQX;my&
z9CZE+^i}05QR3@zn9H~p&biDO=6Pp`JTfQk3)hU^JuTX1c`<4V@2;trE$tL9XEO-*
ztN&B!0M{bWO0kW65@#O!=*mGSq7U}M$uIqFj7Lq_>1<J9jIQhHHv|=wsPG8?E)}m+
zo<A81);y7zy@xCdF42y0OlrNi%#Kj7;vVxlIbK!qbUk%5<lN}*#=rL{APqm6Hu4i0
zqu1bAqK27>-LZb+qPjBin&!6vA$7xD%@kNCtb?fxztDhWLO1zUTCR4+{Bz_A9f$c-
z>iN@P(;sz8d}Hh&ewz(QXnexAVT}$AW_!#`%);p#djHVZ@TMM_^9RIPRfn)aD)p-%
zXp5Tx^UWA%BcTp3ED~LtxkN`PG~^cSma+KgPy122TeMr^u6HQ0;3PjxR+uL^zmbJ-
z&7rjYWQ42WS2uo;`KNwQcV>rTjF?^MZtuH}%|aw?l?%no7wwdVlVpC*Y&Yfx&_1OQ
zOW}jazKbXruVtwYhl$)=XM82pRLdJH<PrC8MIEltoqep9sn(Sz4j)b;2D;=@NP?LR
zmQ<6+aG)v8QOKU_RBz3-$=lPIdQQojls*lqGR%^6uB{16_4SUvNJQz_CZ&=wy-ivk
zz<xUbW3>Ir*#2Za@I2xoXi<)D8Nhv>y=o@_L6@bs%%65#-?=m=BRgkdwqdUcqts`w
z=XN_C*6s80?yQg)lUXbyq|9MGtpYX{oBS4?F6FP%ri15BuR!hjwmWjCSE%#-p;^-2
z4nWI<w=~B>U_;mD5y)4$+drdSzi2n@@Q{2Q_2&ETxQyi|GE-$sI!#ADoN(D<k3RPm
zG=FAx=YOtd&St`VOd6zJPjYVa%WapLHyf<fu3jdnRVXyIU)*fA^zddnye6HgTVAGT
zzx?Ef^nLvks*FUaXwK-5d#gqERO6B>G5$;s@!h<8w}`o3bB<Vb2)!la+19Qr;J8o4
zZBo4h8{#;`h^Z{bzqLEl-7|Uq#g3(r`EG0PyQ6)s`wWpyC?%)3hrgH#UcYF#67j<`
z)9A8<{!8p980r^)7sqcUvcU}F1GYyO^)NqLf%e_U@DijcMf}%=PEH~+UPh0;Yi;*j
zKUA<&Xvkx;x-XW*Nv}v62)WX-+g9D|Q~fifSs|o|S}4o#P{{3ZH8y;b4aw1O!<=-5
zHBf(*yuK_ybZi~lcj$rCqe;6J?qcKhmBbkSaTQGGh2_(vZ!`hSNUNy4BjGOipPx~g
z2<p{H69osU)3Gw_8xnl{=v@(VcGJJHIhIC{G~fuT9*jUwVAQeTW|3~Jql}<SdQQ~=
zT5S6ws`V}zIyAEAvn>KBS-`9Rg@QpHjV09fdcPu~W~3p6)8M^j$qWh-)P~IamCn7s
ze5<70E8oPPzgCjlZ2%`sib*WMzg2rG@VBTHP$0-a7b{I+uV{^BJg7A-*gUdkhknMU
zXylob#?xjXm3ma&-A##)zTJ_wk?w!Z&;I>TBJDvF4D5|MsYwOxCQS}ne7UfD9(CmC
zR|h9+U|v0%%avY2RY#V5EIs(L+f{;5W%>n%<MzJ0QdT6gYW{s>+M3L%&HmCjA}gGh
zB2H~OnKYzi+K=!!tQsyO-Gj(b)$*r+&0JvQ!d?>p;9nL18y4(`MaGO4`&1!EMxIh^
zO!;U(!LZx&dZv!hNr%3)1?$lwA$mGWD=5n}8lg_|E=0eb989aIsuFIf_$nvWkRzcJ
zN+wOndRr~NXkW85C$-UwjZ$^{vdjocB2svc)xRp$u5I=0?H1o~4^*AO-AZlA88#^R
zX8Kl`r7Svf-taOaIBUbn`Uifv$UulnWq#&{Yp+;(LoU;o?C&215}_U)s(R9MA6t_*
zocVWMQykq61iHMUBj59co~K_-6XBlR<|4f&%uzG=iWmpk6^4qO)bh3Wv|_+>;0Kl;
zoO~vlUyNRh%}$aMI&pDi3jCh+@QVXEb%v!Lfp3ECDUpd)@|ow7X}_REn*urdm-V_g
zzF8M2gcFO9UA2a9yZP%qO))3Fe`*Y44CgD$kAXoo=2`e9qz3kri1x7kVs7`;zWKb&
z$=jB`%J{{nv*HTueUu0iK)&ACx=5MGUQQ4*QDTArs2uN{R-{sueD_qRE9l%O=>+YT
zewx_59dNwM{1Vr7Y+ep~6b;o=Qv7Dy6{dHx!QLu;=?Ewt4*CL<o<%)9@}vUd^_s?I
zkj1}M)x<HP>cJ3knZ{soHC@q9eZy|)hXTv{%%)9HKRn=Wb7pg;a|k%3KlQZqm4b(U
zWTs3=^|wWx@m*n@66ov;DENO}>$=n6cZkqERwrb~X4sHh4iw|FR3r2(+<q#v60->k
zF3Kp~hO=g-@wG3gEMdsOdG+N{M`-s+y157q9<y-lECU@1(XxNFD2^E4VHEV@2VU}2
zL0W6l8Ytg2|M_Vo<1S^S(rMvB<<|(B_4MaO(4pwY*pR73MZ<L15hsqwbCQ?kw$Dma
z<%1!tG1;;DFKt#l#C20{Pc;CB39;|<c-$a*yhP?r@8?d8Ma&A2gHM>P=X3i#=X#52
zmw&^0CHaRzDFt?N+X7v5y-z64*O#*4aDT;V|5>`uW=oV4^q+?{J%()AmtOSq-Gt=|
z)+2scTs5ej$xHl0M&d_{%sUEw@Hj%HI0K4IJik)x2)V6Nh<-d@?^og}WMuyATm5ZB
za#zSVk~a^D9Uoc>T%qgGKK*Ui(#fa9#PK!{GO$jp1p;>k8+T6eAK7m%kk)DP@9(O)
zM2iS(c^?5e5ic66oFL&5@g{Z67yZV%1YN5BX8lhR^_=#gWlfP^@{P}dnc&uGGEATy
zT7KOa(5%mpN7K3%0}*Kl1J>%BH(I}YLiiKkf9CCZ?EcuxH^XZ{;g55u)L66QGhYLq
zkffGVe!onbh<zP};NWH;KV)mu^jLEB=4Lwd&<=(qmKvs2;6@s*Uz_SzO7A-M{QS_o
z8ERMrS5fF9Mqm3XU_D>h=lRX8fi-JLx1ohmYNH%)fYY5w6<<QEJWK*DbkiO{6GRSm
zU=lchuX|`HWAsqD0e7}V)p7j%=BNrH?J>?bk^THvk9Z(+N{vEEw}aD&+rg_1BE}*$
z8I_|~9m4hdR<(Tk1okPeiAPRTCG0Ng?-b-kWHIbBF4iE*zweX{Wp{f%^RUXGtNdA;
zABj%Z9KD{jEHJs42f_H7{);dWiWh4Gbdl5Pb(`-NpDt3N=E@l4E!j4=i8NPRa%RiR
zNR0`;Ha}2F_=;)lS>|n2e<V{hhC;s~N04c$_5q6;k3)gm%iRXYvnVJ=Y0Bqf_o7lZ
zFc{*y##gNt4X@$)T5;{{yAC$?VDu<A{<Z^-i#TZAYwXIOT*Gl9x5Tn5lqb`v_c@9Z
zq9NoI?q-C|VQRgV?k<0$hJQhj;DW<A!4R#Zf|(t@ACIU}Ei?t>O@tBz+SgFqbi~5c
z)SJylZgP$!yu!sjuY@+eH&vfXbRp9PpSDAm&r<V*8J^^i+-)`11i!yiigDSd%c}K1
zl^ftiBt%r=kzfSFXKP*jvWcg|Cc<=tQa(a~JC-8xj!!M~@h_%}KrE3)uGeD~7`7@(
zFk9bJZvbU?o-E<^E;ah0eFfT}xVv~6Rcbtg9(p%h(+^bL5eK11B<)W^StS#b7(Alg
z9${GIhn{vN?;c=B{$N;L%;C^{=^x-XxB;{>Y(U%bz^~SA?_-FOD2dXnfHk0WMSED5
z5W1DpTB*fcki8;u8Px^u3?A6*`YJS$6`uX5DNbhNpQG#a8_a<YY$)3HbY}PJ*76SO
zS7Z5fo5Rl`EQLD{sejOFXF{Val3@gduhGr>;-vKEyMqyBCa))Er&|V}JNmnz@6h=U
znPX*A+u^6DmAXhxakI`|FY_(?cEdJrX6x+*G3Ytn$7aJL1;l-R$2b$pKhHsqX|B>u
zxuB5|Y4*Kd*~A#tQ2}t=wWsEjUv|{7jc2^wKembdu}4OohJU17toRqJ$BF6hh<|Hz
z6hJg7{gp4w_w`gaUSCpeht72&hsYk5hrkK~l2Ydw<rRp`oxcV$SMsMwJ`(kZtSrX}
zk;HT8_PV#Bs1AQ_h3<iQHRI!ra8-SiTt2kW>h9(IO4?;c;H*ar(P0+<-Qym9%V4Qs
z{1lSkH2S;?^3O`JoBvL8$$(<q!fUYHC71cuvHY^}!EQ<(*`V$Ki(2U8u;%yG2bFGM
z0k}6fR%GfTT-33*=GB*Fhw}lBQ<6EUIzHIgHhL*Upp_iLPxu;^nfY#9uBY{@7j&nW
zwrvyiwsdC0)CeuPhE=5pzNc=EY{#akO7}c&z@}VgA^fn~06dy&>|;vsUlO=HAbEsN
z$+*Ys!njg-V>g!r*VA$;8eZW~Lida8*d_D9{0aZ}X^#?+f%rpJ7jq0SU*NASqCY2N
zVsQQN#mVsJkBqZ73(fl;9}~UZJ^MT$@VWB-a;VPC9}fz!FuCuN*{t7)_i1hX5BsQr
z(%(<|kj?kQZ&U~gFL#g~U1)a(%gFQN^^PH#MAX-`U<(w09>m;yPqRB))|8=lA7}Gn
zdcZ1O$%R8LlrZ<$3di1LlrqALdr-lM>$yxSD&8N|)z_sr=7Ep6FVeh^uYV-mNHBJq
z1YN&|;!$`out%p-^anitISIF-Qg^j0=saPx==AqT#S2j1Pwm`rNTgZLk5Qt=vsaMJ
z$r$LZ!$4Dc%LnlKYvGTP)E;yHBO7VP`z075Mb+K|7Y4O2j#>6`zhwYodk9s3%7Xst
z-2v~sUCk_J!7@|}1vVo6OM9tssV7^pC~wG>x7MxJ;T=T<@cx)h#;w1LQ$h-CLTE*5
zziZKmI$7z&nqwby+<X}K)SW%_$hj#}WT<%j8Fft8$%5+_sC7nlzk^p=4ruTNebnjb
zS$%qg!OHh+PYNeO^RhH@T>w2w6Z3w@W{UNen@Y0pNDQh$XjAza^OJPVqSLHjg8}}a
zsQ_s+^oUjw@!^nd*bMc}+zd1~Y>Jn;;_e+8R0vv{bQnyq1|EPXjlY6!=`z^4cJ=LE
zWK72a(HpnDTaUahD_H`2_t6E?kj3Q7)nAKjEo7k8%zxmp+?*=n_A{pEC+X#8h5afk
z(1hC8Q&IW$J*iY$ey%lb7IrUMhfJ3Ikio|470q~9JW0S^Z7FJmfP`N<EZaUYxYs6i
zEn|X6dM;TGvdW-zJKMa@H;<cf$au(TBWc2?xjg(=5mY)S!W-5m3>2(uw7DImoMQaw
zrqr$2uxWmzG4BcD-5kdLn%m6WEg9WB#!yQ9Rkf70`K&zq>r@!1g-77wM~c}b;Ftg<
z!jdYna*LRk!o1ohXlS5v$cUg?FE~IB8J9J!{RzTgvp;V-gFK&W9?GP6AF*Oah%JqN
zUs9i`;#{AB^A)q*D8Ihq_{YU|3JIh;Jx-pV%*m9~H4BbpAX#33*(|nz87tS?45|0c
z9~<p1Tfnq6jJ)f2QFt@XL0geWb|VXmY!z`~Of*^aOEk9|Ik61xkTc!hE>iybfb=T6
zGRMgD?R!TY`S+R;S2w!Rr*r*f+Wdoo-0wam(nnU?<}FOsDJ24`NW(=zI%5U(nkYX@
zUGF!-EaHyOUJVnw1AOjd>8-DzClVc$y7cHqI%}uRayRQ(QWEq{fZz8V@^bLGs|DPc
zf#y|~gE|BKw(w!@5eg+?VO5d?o{UFS4NZS#Wt`#(8@W~8cvk;Y5#OQej0_hqgM!ar
zKI09n#gW!Jw)Q9FAEBfO@D6j?j&hYnbbNEcH!}pa=ov%-bQ6>h#zR)&<fZ8tTjSBc
zOSM5=ag&?r^5``3uZTOE|7Rup@pa8`-cjab=h`nX#8w>QuU<}QRum6b0UHt1cl#IG
z1rjrbS%^W5AE-%!J4>dNt`=WA1X<6Yoann$EX=_XN<S17h3xFf(CVj8FNKDBu~~4_
z>`p2xM>oS=6}tbpK)CoA9Z)*o*DzNoglHJ{pT65sNnCrl>i8!PO-x`xeM9A!jzydl
zdi@}x4chAzRTM!01(neV^e5|B!SY)75JdYQSiWhoNq;L41|;qk{3$-~?HH=H?ZHP<
z>FgON6&L>5Loo7@5?bolaB0}E5wrPTi5E)ccsx>6BL`<h=h79`U)?6Ddp6Ny=e#$b
ziH8-g*F)_2^Vk`{DyHb8TuHv->%boG>AZ{@C-3mcy!|R80*jN6(5k!N(71x`C)CX>
zM@X6-+wol-<LmiHHYDr^FaDG|q3#HG^h6`vYo^xPY(X=egHuq<=Ep?f_$1>d9?I=-
zCCx^FlWLO09Cn1Q(A|KH2$$Xq>BeD)oeTcW47{Z8Lf-Ifl~jQ%;Z~yC$2W?P7EXnm
z`dTD9xb`;-$2Mgpo-Q?RGX97&qW4n+pYIt)bi3!vwlR-#Shiod_=|1P6%;y-9U-50
z)?t-pt>IZ+tlK>?cgKQ0Svb8(H`zyN6UR%`qBIoaV=Z!I-d;^1zX<3UltJO|k5zmT
zc%WY`_WgWz_DISGrj*_zPD8J(4M3O!L1DoN<bh9k+6L_XHt6DCF1e!x%OwVou<&cs
zldCj>WoP-8%w?Y8cdCL_sYk4CxRlBu65&4c^b6vI=xU1>2iNyg$Swg8vEydat3AtR
z(VF$6FVSVY`}NLk;m!>)FC}MZ4%NO5nX5MtZ(L=eoErZ#>Dx!e2_d+xoDRiBzTxhB
zsCZk9kJ3M2wM?VKZ0gNZZf+2ktA^Kah6{_#c|Tdz`9b7{Q3jH9tDw8$pr%u}%}}QE
zK|BX-BMw2st%4{*K8Y~(p&Y4iG%Fl!GGdK$FQWSy?n)BOAa?!SFf)9N(a+1Q3ofCh
zfm{wS-?dOjqP`bSf_+j0qp}$CeT+8<z@}q4W)qS=PF}u%{7OZb8M}>)>&3^JRx1l0
zkcTRcdmRi)BdmG{gp!3T86!N^j5Ku5H#_2(o~e&+6AvXkf;}Is8sYW@J`()9=Z*5J
z>Z%NVszPaO;OC)t8f{Nqv%YQ&Fr_l|57M1sA?p&BI@JD<uG+U)3fg78D*vDE@a+9$
zP@R-NH*<4f;8gDB<94yOQTetbD2!wtjvp$o&0MZyyQ9igg;?ojJi!lSs;LW~=iUfZ
zEbp_nOH!_sYt|Y|uhy*$%Wnri>v4`;nM^+nOzU`3FV8tEX*UbD)4(>?u1CUMPj~)2
zc1(Db;qo+4S1^3@c8tC_;N|LVjax0W$85F@!_};WF+ef?2`A^*_a+I+UuH}Q?*+KL
zzAjIDW?1R3xm>pZ8OVB)={daH9g3ClmYy4pCyd;#l4zLKuM*3#GFsv7IZrGW{oZE`
z`I*0G0QGP=F{D<2Bdlv>*zF*LVj5$rXe-RfGw!famk7P^559swLu!B|r_@K2@P#h=
z74?dw%~}#GL&nrxMNX**6Lh7jk@XB)9vjN6wPv&f=AJDl$A$9@K1EfIRL!3zfhRd~
zoTlcLUAC2)!xA^3-RtsP=x2w=$!{r&Qx3Z+H%|Z8oOCHUV{T1ylzl-IUG-Qy7=4LE
z54I^c>{OmAh`mc}m^H4R8mGB&DbuF7?=sH%q%W)fr8=h+MTe$GzCs5;%U3oQ04p^d
z_QH2@MmvHGkM&D&0Z^9mt4I{uNg!JE8s*J%_3vmrUD#RA>B>#ylODPRT}<Fxn=en5
z3iMwW?h))F9nLW(^N)W|`SRR1H8|WPL9`2BCy?pqS+HnFrhxREM$^KE>0s%xm!;Ni
z0NQkbqEH_Mj}>J;{tXq{(Iz?Xu0xyGhEv5=GVI248q}`5=x5ND?bHdM&<kN%t*}Q6
z%eE+5^$OTdA9}jD6}P9Y)5)~$s5OxIrE%$d`+b)1_EPx5aul8R%%bwEUKQJ$?#0BF
z0L8ao>q8tWnK1ess1J)31vGr!oV~A|C!vG$^qLQsjq%Vgs$YHTJvOx2VVtNxL^0hQ
zkonjLjG)JFz?49PgPmQt)A76=%bJ(i3tyg%&ZWaI-jEOZP!S>10fN~3kxt&%+m!!8
z>^u38wqE>2P66l@yHh@=`+cpa5hY|kmVLK{pNWR$)v}1Lm?Di8<QNfU>#Y`Eh+k^1
zcw*6ngTghc;)KZ{RTnGPqduNOR_@d1`UWJ!;_tGvnt(m-IlMb=6Ph#goqIM~Se4KO
z-LH#~=qxh}RQ_M<_)XA1Tuc2KNr*I>7AkgN>}t5984aTnj+~0C1I<}?Fj=RqWCr)y
z$_ag>uk%9fU=54XWm2_^bqJ*<h`gWBw{2&(PDvw}fAl4#cC72p0jIURbC+ufs+FTa
z^p|7X7I7qf71GTd>2GZz!HXLdYY%?6x+}-pQX7A9YV9u!B8)}W;<qp#h;#P~-T3Ny
z6K?J_&qk#}8QR3-Viv#TY*8I4sIbCOPtpU^Y47I^;8)gHAQJ&o42cdGH@^i<!?(H!
zZO(m8u7skFA9|2h<MuU0rib#kYuBm|FFvbq2D!)l=v!5u4|Ww%>EQyV7!*B9g8TT(
z8-+Gvqunj0Ej*{f-HnNd4A3mWC7C74d0lE>tQ*g%s*&rilX_0x=-{}9>2J89J`K*v
zHk!U4jU=XDUD?ol!`OdZ&A!B2lm9eUA6E9VVB;+q2GkyOYp%`X-<@i7nb%d67ZDO(
zYo)y1YwS0G&Ndzwjs38;G77(a^GSk7=~La;+J{E}$x+D9-ekYM_3BE*3S{G>nDaZ{
zbUwWnW+i8ceDCA^PbTnprQ8b5f3ka-^;>-PUu2t_kMH$YImPx%M~k@ri-8iG(h09U
zvc{AA@iCdL@B^Wb%m&h(`pT^EdnjzDbUokEtZF8ac;ZzslVBfIR?#D8+Ke2_<V$F)
znN;bXH*fYNuHzh5@{5IWjk&+l%#p%QbuvHr@y7bajTH)Q`gpij-L*aZLWU?p1Vo7z
zv-$57)(~z*#k76ogq4q@kdF$>qh2<EjML_$^yP^ovdavi)s!bMUEgb;@G=H<w_Y5X
z-V57QDVYmp@~rTS1$?trL4o0T-s6t}OR^Y#lluA^>v9x)@hGMJ$BO*G<Ll8u3P<f1
zyoA%tmS9MIA9=rNLFA`b<NH}dE1kI0mE2MzuzjVDs%K$S?_3E<wKhKs?y*ZZrLe&_
zO6PUN%t;S{QNvCzUvi;UL~uuj<<ELn^}G6eRMTEp1bgD$l%s@LeGX|%*Ix~dXQT2*
zU2J5Wr%N1n>o;xHU@O}sn>R<+{ldTB^z=nRe9B9LJIZ}Z$d*6K?ae62w8;;>6n3y(
z0SrG#l<{VQ#Jv*C|3-vDVD;Yvu(zBTGbtD=at?WqQ3-Z2>u>{-e<rE@tI)DFR{R*@
zX>{#N4&^ia2bYp|Ns#h6B_UPSZq`Ry%spvmnS375WYOGlV@*Cr5*&nP(+CcDQRS~j
zXJ=Fmbb%lkqh?7nyH=wBF7_0mx;FzP!(GKzuOLQP6#>)y;as%9LwVJ8G$}Fqt&u?|
z$D=Nk>%2A^eS@FuqK<;<<0`*J;fxP)Q3g5k^Y*N9b>gYlQLL!jIBmxs+N_IsRDFtj
zljJ_2o#L64CnU=dPEg4+q~UXU1Pf}|_r)Q&9klE^;VUolV|2*o3#wN2Y}>gcb6JIa
zA-;obb78!GK4KNeVn6nHd2%K6Dyo`nFvikSx#nEEQjuO>-AdEz*9{dbR!^y&H1bbw
zIN2uS8`WgS;pAn97*aZoC64;jji-JGa9$Tu%=P}#dze8|i-ilQfn~faYOWN?l=G8l
z-5(P#G7e42)yL5=+V5K;+Z?QX*7n9)hX3Z(_GMu#LX<G=qZAg!?lU*MuuXrCWu56^
zr7@t?Q{*k%p?eWwIqLPwv+;o4dXZ;0?JEN5JBNwOa7fO&qUXxL6Eq_e&6wG`Yk0WR
zhO*t}v;hCz&a(106L?qa?=|n1`z#ziRIY)~yDj#}ui2jX9+B<Snx-RH_*V+UB{2E(
z&s#p#D5ztClrwsqgBoI$R7n)H&!Yp}t*dTKcbAz_BG~sCO@DvZ4w@*iSC$tU7nnQ0
zc`v3t1B3-nJ0Msn__v=^GPL>2^Ar?~8wQnA+6h)%Zl<yO@=tALBkmMs+H&DSx2GSv
zm%(V*fxnmL#z<ZH&295r7oRIMKoYnb5@q@iKgE9(37Z~)3`b_EZ2%B1_t8TEf5S~w
zg1>oFq6`ld?j3#V8-Eb6f}jCk?3aHe+rm$yApDnblX{G_#XC#ruB!UP^5uMaW#xXg
zX9uu$|2>E%h<|v5=<E^#ZKIJ8hyEKNpG`-fk7hrYTRM(*lSD3Nw#-3*QXOHR<RtNZ
zi5;4AQRQFuQ+U~FP)VerN>zP%%SgZ}7Dhlm-(AP?gs~oB+W8mHax3)PZgn6vx>voT
zrK$ctVSRx$2Noy5)sg2pML>!8vT4^%bKs&+5p3V?deQxY`^cTZ61Q9jcvysw@XVK1
zLOa3%f|`p<#&MWD!6l(HKFkE)Ulv0PuT6+)0zkzWn7^}ux%fEXHL$e(EvDb#j9JFx
zyPMSmSA<i7OUmg<y8&>}4u+}{@3OkYLsnT>+*IwiTfSZgI6d-tH;-O%5$#A$S5^J&
zsc7?bm`Ed=-=Bv21z6*e3Lmb*#5o7G6IuLPH(Og69rgj<AgOAWCo34bZ#a8C^5cr!
z?$@@$7lwKt0}YcZ=vS7WEzckIgi*^4U>2|8!WVX{LYP<t6(e4g*$(Hbf_ix^i{KM9
zg}`vnt`T@d)A>2+4atgSM!GD|oiE2Xl8#TMR;xkdntnr=!GTfa=kz-AG-;c{ivLUI
z`8T!hKO&4YN<n4F@^RZT<HYgpk)e5x-eCMOE7@ha5g2xCvv$hNUXaqwZM2=wm$nU@
zlpF67L|7&hAwa3e#oBE<cQtMlzmoB#mN}3OMdbo=Wu?C(%gM)5$%%gJl8nu$c`u2U
zmHj>*i2w)f0Y<n*Qe{3JgP4gJ-G>(0*vH||+Nn;aD!g}FoqvxGqrKQ%ay}^Y&GgzF
zR#0|3>u^5nC@I5K_{h!lBTplJ=Z&G|H7KdPFr%iGTz$*hxCHKA>CXlBTX7Q-RG;<e
zUwdtNF|ENzz@Us4<;_xs2iW38fNy*lMa!FJ^FbfIL(xw;9NC!E4ACr5blR2Lsb1!@
zK+d=^UXe_Y@+dlbw*EmFqm;D`+BO_(YiHtW5y+&YD#b=l?4xO+dW>dKTnN&JH(#l8
zmYFO3fX*GHywmtXDZy3L9mwpX6a1Pqi>t3_RAt77*M?~8Vm!KvW6Wj-VfIt#J9*<G
zSYMo1CT2AL9c%5QFO&8u?^54&Qq$vU2XLx$9+{gqNsb+%I5W4BRH%M8X&ls=RM@CU
z5Bx>{`UgtWDgGE;A+cIvQe_X84spdLd;*(~rV@Je_YsrP_ac6=%73>%L0AUqwv5yu
z)Xa7LMECzW0K8JC3C;Tr;ScP?7qDBZj!NQ0mvr#FV&BDnU#N;j=BFmLn!X~srPe`E
z5UdO_?tm)Bp?X}NWzI2Xfriqe1=t<dhspkOoN}}@O|Qa5;#dvRXaM+D(z%G|1YDv9
zE1yT7^-D(pIOcSF==XX6pd+R3n?Ml$NK8w&yfX1brn~+;a>BQeCtUjA2aLU|zD@-b
z88!OvB0Q$>4Im~>z-@#S6MI7z1?EJ?sue~RfAknvmLjZZ)JapuS0R<NYd}=zJ*`8P
zX8=4lmTJwBjQav^858j}Jh8*ph0ZUIcTAHF&%?C-B=VFxWeD?2s3$sqIAowor|RS?
z&f0~G94nu$g7R1YgXb<mJRy9jc&plTRnyx?0yj^>lHn3YtF*qbq(&R#%A?JG#*@bZ
z``|t|NuH|Y+~li76|o$|r=bqSqO}>95(c&ka<-WzB?_k!!VErSZcma;<QEa}gh<qG
zGJLz&%G3-5{Ib<L{@vBjOuwIdrK2P-7z5;lRMrfPOq>Uaa8+VCtl-&)$OUyJL~$0W
z8~5S>CI0o#Db;%FE`)gnA<DY$4pFrVW?0Hq*KwJ$(6OYzPTwK3<%4JD&E;NtHQI2|
zRb@d`*qp<Jj4lVL7|}yK_FU3$r#o^E#9DhbzffYMf9t3nKJB4kSdrkAiLTAEYl(sw
zkb7J$$2Q+#HR0)2QDwM}37tf4dvtQX`q*#$>E?Bk(*~EDe*HBax#mu&xk`WCyhv$u
zBbqtjiP#hiCMpiVY(g~~-`&7`x%Wet7bDAdUfWjI?nAMhqSqS>DF+FV6tkT3Ps+C3
zTt`<f^(+?-O?&GgIPOGdOf5nN|NWl!@4QEoEtq8qTmL4ZmMyO~2gdtCg<??tx#ejh
z-s>si7~<n=7GvT_5@AacZG`;ix8NX7R~(b!w*YVX0r%8N>^t&1Rou~7H+cbULs<vU
za?&12Imak#PrH;YGrcvVvUXh^`A0TFukPI8$Nd{%;vBo~oKF5T4qarJ9(SCwFDxN<
zU6ARDom_fq`OVa;N%7c%Cm!!6f*3C)5-aHu<I=B4Iui~LTMk7$Kq55UK{3S8EKC2U
zse<ho3>(O)mI&u2uS>F#h3l!-{z=NFl@G@;JjJi`HaCr;gcv2N@X{{C^{RbC4M}Gw
z4rkv9*E_%+#JJ?WgP%m;-bu1qg~jWI3>dmK71KZxQyJy|RB@}&CoRyGl+}Ms8mny;
z(uT>?D9R_shT$=ok>wqiZiTXoMzE(6wj~P;N}B#c=BS1~+%QIS(kgPv(F3hjE7e_d
zEy$!(fr^Mef49bi6h@nYcZzSC<P+gGE(YHebJLadkyW>a&*XSzbr&s`B(DP&Y)h*`
zvNz=$bmh;hz3Ih&K3`f%{J<iwfnSHo2pk<t7`I1$veI4Y&S}|Z=1S6>q~B;;r%~<2
zm2PJ8{2YEU%Ga~QzmMyZ!4u}euutzJ%Ysr51jS{H+nTQ-TQZDT^5u5B*IFwLp339}
zBW+;Vn3W@`)V`Xg16M#QEXxV(xq6(n`tt3BgZ79AAx<LZBR6Z}YYus(7pz_CVX@-I
z<AW~h^o%K|ZOz9m-0nY_gy~&ne*(m>wCd^ypH~oNeg8eax@7Ac;+y-8O?l*0B6LV$
zxbS_)q^tX`V7{U4Gcrev_=f~ng0uyuIORmLt%m8$qWS$kj2l&%%)*}C(*`}YBrr$4
zxRZ!0nl7SU@-a6^R~J-F(lebC4pQJ7vN2=LO!GJK(Zy#yC)v@q4L&;LjxR8CqLc*t
zHs9=59Ltuy)6*CCT^ql7-7|4_yJcmkSy!eCvzy4UrtN=uqXtI8K5tCeov$xVVE*X0
zHcWE#<0)-C<Uk5InlZk_8UL>}|DWgJ|M;IKRHPLSqGWgC|7;)s$FK3%ADj|B5c{)R
z0Y|m@cN_Y@dmV5^2H@&aFkT5g{{MS*R~Y_O@H1`SAnyO`bNPEm^%rFaS2q+)NAaH>
z^55V553AJw`af@iMX_(MSRmpb|JTR<FV30>T-`=DF4n{U*Kt=kvhk9Cyk=UKEd9><
zzYFyG#^%%|_N>uV5>sT*B>(Tf?-e#`x3Q0owCulc&?!oN1X|7{$EZ07BG3?+kPuXX
zVSTIwh$+7yh})sZn3r_q(FRwq%O4!~^*EVzz%Nb)1R{;T7|P|kdx-^r%vz)x`bl)a
z<)cj@(~|~v5_uYu*cAV-?=3Vwko<4ogx^I$;PZrWRp#GMpEDKJBnN#}nI&GC$3^h4
zwMGc^;G~JcxP%BD0!%agYGLL>CpZ&Y#C-o_2iP7^^v43eS6v_Rv(I9pzf#Sk<3Q#m
z87;8G$bb4>q6POdigzg%&=VBsl`<F+8($<TuaebA8~QYW$JhYY@2Q|x#xA|ehTe)v
z65f4qc49x@@~r6*NklI9*Ar2b?>L`ZGO+|GsQvyu$^RCk{ridj*Ow711+GX}a#iHt
zgNQki01tTi8yH@t1`d)kGVxg;@N@!V*!cz>(iV_130#Q?xrKBBxGo!T56MvY1T1M0
zI}VuTw*W+@E!Nm0;UiQs@W;}B1VL=u@(kVv;_+%QhcE<MtqxlJM0&5OS!4i%CqC|u
z>Y)g*y@SQYc;{o66m#J}5_-(h3_T2&B(-1-AIBei=r1Odec&mw<sJn8|D655C;i{Q
zKtgbPTH=IV|9goho!GO$xnkq?`Y`2N0)rdfCGcQHUU6Thg1aCa_fviX36*#z@=AsI
zmh@K?JbHy#Kyp(61mKQnx#oLs!A?8`#NAZbkE+HmyRdvK(TH}q`lV8uin$Vl7kAw>
z=UYm>10$t)*c>TQL3whgd@~O@MfSh$kpI~!{`<xJ0;}Ly+y~o#f9A|awQJf)hW!e}
zdbHvo3gbZadl;|k!~7eh?zYKr3c!07Nvdox(}W&+vKR(YfXjBF!ai%o#HWe;ugi7E
zb_aVDYc9eUtq0Gv7x%w^I8@&+P<x514r>^Z5X+ou09X~I<)o##{#iZy2p`DB5BS}l
zP_8tCRFrrwhUN&i0G>CNL(iEOn9*EKZ7`(Ug6lUPSNqzvYvP6X+On60H)Mo+Vm1r1
zdp5PPi7wJuO9rIH_21Nk!pzfeGfz)s^z@qcE_&CNouMg5p&2$LLSz<<@oh<VG{+(w
z<O9l0xvdrxfqrAW{X*$Q-eZ)@t3^pe30*z*L_?*Id$+l$P#i5LR-G(&k;mqf$UC+V
z5iWs)2pCuskd?Ymtq<;a3Ov61j7I_|EvFz%{UrqXkOfp^l_v%`(UzAhF5$a81oMIx
zPL~rfmq272xs(}hOsNfKa$fg>4Qd~6?GPzFD_@$2rZ6@?zHTaE0}ri4qTfx*w=RJN
z5t$<o`;#t=E-s%?Sh^rVKJoQuI*g}!QJgF-#}(N$?MeMMHr%EfGknB>*L%3Zj0XiC
z4{>o0mp!Vk?$UfVtT#;_<9;qCD^AX3d6_inLwkI&%0`&7c{6y*G6w%yL$jTc05OG&
zl*O+48~rCgGULB?p;)o^>KNYerBNZ~6OZM^D+6AZSpW)J+RH0H@Cwt`a#cz&5qU{y
z_=(Tj(M4#N8IQuq;F{9kPFM6!yoFcKAQDIBm6B^_w7~rEsCf*>1OKW}@QwEwWy*yA
z^VZV$AH+az8-cGOgp+O-X#E)gfiKq%6iJ$x=1oJ%EyC!M2YmV^7~qD$=0sY$1t`4N
z0otz0T&zEak^{K7rFa9_6bl^-=ZB87Y5ffVAx}X}KUw+Pf#ItJfMOYlu0ZetEf*Rc
zg|#17y^q;5aRLmuIo19bV?s6FV8kos<+P4O{i3Tr0J@o5fP!EG!s$<f(=_*HQeF8Q
zKq1*JpLm6kSj7UPW&^<Y)<gv2uO|G}V8Hk?4uKMK0?v4YiA?GinI6A!Wc7gbvd|l#
ziBd4w8?#SkjV;`_3sN2r0py#}Er15j#|?qhv7)qgT)rBFc!L0gF>^snQTYcnxah4j
z977QGlaQ~#n1LFwrrxvGO{!IcImQ9N(0kER7*jy!2kbj_htd&diZI~KqY1{Z=`q&8
zI!HKp3qKj-K%};oU>+Y2j*nxr(_Q_hNdPEj%8c?c7C`P#gD@w#Qe8PzSK|LV9)<G3
zHy^{Gy1cYpJITLsU;-GYYqA4#v=Yk6s?97vBfv_J`38I<&7jB^$|9HfQ%R-Xz+Vqy
zF`h5y>iq}6#u?O$l#&p{O)g+xs6Pij#X%1(e*<f;Pz4v0<3qnYeFURJH~kXGWxj#V
zv9d0(K`JYo1x2YRF|AH$D;L0euS~24)VFkm3)wbFw)5MvGb!JF7n9x_HjC@S=fdMt
z69FGHL}&iYAIrOQ{SSZ`(F4EMl8qTceY?2Z(nDXj2s*S&Fb`nguaEZ#^8r!6Q67~K
zfL5&oKd>(bK4(*~tKEtruz9hBQu?~v&pS}M{W(M|%ZQgbp1QmCyBpg<-G6$6?IHh3
z*RJZT1(@hd0sD-h74Yyg*nk>tZxv#JUG_x52)FzBO6{CUi22Gb7<x-SlOvN-%91q3
zq;>!%U=C!IMks_X(<k=Vl&{{wG5!<KMPr-Tb$W+Y*))n~>mw0&JWsY1f+Ullp|a}n
zzh$Ng#vaYKw>MRSZRN#0#Y9X!t^<F7H^?%^9h>hSy(6zp-~*G);LFo01SEU;>JG54
zb$j1!q!`y<)qt&0&mif6qR(*a5h!n#3+MPMfpWk*LM9ILRR)hx0?&6zzurf~0UUEn
zV84wO&k$p3BQ$VfIzK^3ZCf@4d64kz$<2?~j?bmZ*!bsk1$OLT2u28dqdYLj1AAK3
zh<A(z%HQksW4;I2iU^OC)b1@%p5UxTqx~#|^34QLr$)Ri$Ee_9y#bXEJ?=Yo1F^ET
zDPU)HByVPVjnZfIrpSo+@1fveJ|pcd0~E&}vVx4_RIkN8(zF65)GR7%aDaw)m&NK{
z%@~gwGHnMiy3s?lv52XW%y9EF=B%>}Z^5e#f_G&Ep5)pe%`D7Dz^DlypR5fwGIhLY
zvA=t0H4Yb&MNO%MY5$KwMqer(Hh@bFVOo~BeUF4}YF6I#o2#|c1%wK((L*>pBhB^E
z)+VxUYd26@UIRN}oF+i>9=Su<g(lQwC9SOAf$=`O?GSfK5~_U@5F%merag4V|Bxs~
zhEDjwj!`Jn_sV9@B=AY(LTErrE6wB5dDX&BH1QR%oSGp~CZgJC6D4a75pW@pkkU@h
zmy8&b-wu5ay>36BqBlJfXTvn;LNn8uPMSI^=6cUyAp#46=U|Ek1)E4OzW;RvxjTV$
zM_a&19_USry$yUXWhB;~>#GKlC0(CIK*!G0>lycV><&>Im<eP1J6C`pKLLnH)RT3s
zNc?}nd-0S9fUL?EX;A`8coJ1rcbrnP9auFl5Ap_m0n6V+Jo;rqwW1*M<x>M2UKnhy
zDhG~D;XbZK-4AXvE6EpwXwVlfA?Fpwh6Jo~c&Rm??>i=*zYoSNH$l0deC``epv9?n
zG{U=V=RTe}A*11f$DqU=&tf6`^912mo=jP57`sPg_)b-SAg5i0lIqPw|G13R^|-+J
z4=AN8=&j6mKXDoZ3?80LQRyoxR+O4wu0w1M0~&N&;u%F8xfyD=#6-V#I&()omlW^R
z>x6nQ{*A&qb)@cfqdU|0Wv)E%x5qJA@BrZ;s&ik}-fVVIJI#J-g!LxYus;ugGS)Wq
zIlvM9*!mxdPh0|jEAg!$P57u%q^d8z?RgTI97ON&P5HOMmZ9VD4ZU)v-3dDnv9@Dv
z%N9vm)^mn=0M0}5F*U^GXL>bE@PnnYi&S`2$9a;8r`sQ;nd4n5f=$pIl9OVYYn#*j
z=k53Tr^+zd!k(cA0(3u{5v`Sa`d`&gu3q+i2+Ryz{ATQyKKBX#Sd)sg5BqyK5&M5a
zv`57E>(&z)<tldZKTq-3V==rK;a1PZI8a>q9Gov1<{XAQic81zvTaS8;5v-i0*@|}
zuK6hI^H~Kwq<NlOwOWho=5&HV{=IDDVX%6xM6SG8f&_s<n4XYbc*w<@Bvw`2r~*xb
z7m9~Xs$8<UgY}YOzF~S+Z#>5?in(7UHYPLLn9B+rm*y8qhAD+HYv5_tzVxY(Z(IlS
zXiB+wi;`tIX^#cc9AG^oU`fAT2Sgr~T=ZH=wwxy(6&6jBWyti!dy0P%`8<}DvEHH7
z6ePflGV(vg$K!Ii$^UuP@dmEtRN$jLrxq}cJ)AU<;C(q{tn^OXD>u9$NJzd#pt3a2
zpuiHgEN`TcxVO<jZa9@|iU{1@P8BmZl`OMwOR9xcAL!-w&QgmpAJ4)9tK}|7oQr3b
z5+ySjzu6Ah(!JKb(Xh>ySSp~@fF6#UULGn{s;HUgVz6kXXjX~jE9Q}B+o`ipscpqN
z$z>!e{xpDp&gD}r$z~5XOd^9qNj6#reoU+JCc3B1^B6mRB?yb>z@vi`AU|In+wJ7~
z^a0`E6;guxgB4Nte2XZ@@U(vK_+mK{bY7$+)40yy_oj<mm8D4{pr}nNz%4g_l(u4D
z+3!14p!n2OUY7eZDf{E+uMoV@b7i!>^NeHvc-?*e8<HE7rN%sV*5c`t;u~cF$g11F
zF*!DBBF+2u_x$!?$3TCu<p}Fb(d;Ixbs+KE{eoX8DB;(am8oF5=J<Q{HnK#oj6;FK
zS?xDHAjiq&WQn<VwR!?KG)g2@!FW|=q$>Ci^#=l1<w@3yNSRw9vsz2do`)wwba0om
zo7Y#Uh6VN-<uB#hnDOe+)9_|r+t^-Flb17SbQ@!i2VOIRyEom<MQ$mAr-ps*S%#<y
zH}(4||F{1>j-c4siK<~pEED$VIc-e-J=Qmxqlu~>wET6FmWPvix=aCYriNtSU_bX<
zd6!b~?YMx$=L@&Hu+oGq&zrC>-zX&|Wg)RItMk9s%5Vl)<PpY<$3#s&#KG#ON;rMy
zwi?Fp6qC{TM3yqPi7I&OM~IYIsyAadb@XSt5^+D8n+Ix7*Krb4b-5_2h2cj%(%C`1
zshb%`p{pGOVzfcA52!U`ylr4wC}1VlOc4E&u7qm!Wo&O#7W;OfgTwbYVp{35OKqua
zlo#w_JV6A`K3}(D%9A?$kKVwPlIf=QjaLehJHzGus@ZSe9;U|%Y@GC-tPU{H$>(Tb
zZ+WI%Mn>N}1mPt<3W4jdPrk{hqeDFQZ0$mR2*7wX=<WS(+ljDtF-cSUfVc5FZ_3Qf
zMnChme8<5Uo_$j^!sz5lR$doo34OZknh;&=lZtnX$-f>;f2<6-dAF-ygXDbWUH4{g
zJte)l^0VGp$vH#<mk{fJEIE0Gi%uOID+iNyb+%AH0Bml*>gV1dz2aPx&pn`5eidNb
zir#JfFg){i3)!PU+cLLE^!2a2g0k(<ca52EYjJKp4Ik0KZ)LKHpjLN90<@2jLMXw$
z{{-?I$Nb;BQHAA!CP5zdEy7?(Jnd8mqE49UG{__YH<`j3f?$tTPHUW_$U$?037_Ih
zlVbsj5G9S;)^p#2E-{gbH7C5%o`O#lOmA%Ltj!QPIRhw%nkP&N2#fTzhkC33rg~r+
z#}acCI>m(QBtzk0!4z%c4b0$mi9NYr>%OD8Qi>6&KNW55;==Exmu2z{vJD7k95D6B
zmm&IyMlCtX^?=@2HQYoK6C?8AubB7$VeGBrqU_qeZ$Tsk>6CV)qy!P9yGua21tcV-
z89JqVXsMyQK{}-sX(^FzX=!-ZT=#ul`+oMb_vd~71?FVtT(iz~#P9b#9H7<HV5Mj5
zr2P-h#^l+_id3VqzJ<M5HHa=$L<}vW9f96!jBY8Q*yJb-rFvu~p~&KJ6!8az`IA{W
zV$EH^ns*Yy4N?n}N$4$e^bn;9%A;8#sMZp*L+$-qG)blCO2rdW!}mT+SF5qp?6c^z
zkFO_ezLv{m*k(-oneiDg42nACrjxO4JZ-p-{VK~0<_>F#ts5Ft=GiKo4dz|iM<myB
z_gCF6ORO`Cq$3mXZ;Rim)MaSy`J{@3r2Ugu2jUrB?;czCy1iOg6)+!;r@DXCGMv}F
z>^ci|3zYSgTp^+!Z&uK&G0y8=bG;-V^ZjHwqi_2&FOy{8*HZJa(XUqrrIW6=UX+^3
zX+x0|M$1rjVzg?$C~Yi3o3f)EmTQgun9&sm%9&bRR|dEI<H=(%IrFG{r5nec*olr!
z`Hv2vR0$XM<&{#1NJi$cBw6>1#F6EGulmwcd<c8r7GD@v#+lhgYCV8n%LlLQNikmt
zmMd)+_63(-e``ANxUOZ!Tylyl*mA2m9Wryo3XVV6*gCuZpgnd(d_zxperDGg#{+#u
z5~Q-m>t?pLFY%EJN~c@jWK&TB3ytj<J1i+F8(5a{c2do6V#RUJ&#`KB5Hxnz@7j5&
zb#BOSqiE1PCu{hVHN$fQx^?7W(=a0Iv$N;75!*mAQRVWyFb?%D(M{^1yfV+ZU%1K2
z4e4~XfTGX7mV49Z{6zxk$7<2x8|N$fd<JJj%_oj+Lype%DGTdpobgn5$6PcO?K`>D
z{qp9`BOw)YPHq%zH7l8k>8?=d@5Nr{lP3pkarp$k|M@KZS1ARqF($!^>70NpsH*Ce
z#e|^E{66EeAQX(#gdH#-=5xhonk8yWIr+1@E4&>)ZXvHDWR!c1OSce8Ws|m%w_^^5
z4kj|rC{I}FX&P_ZH^ck7!s23d2J`zU!*ne01qa1WHdVuwU6sOhNMIl*kVVX_S0w{Y
zLUu&8h(^wZpl*HozyqW~gP;n<07#eqo1-0(8&F{<|0i;scy#Lh9DNk=8+q&*lpx)L
zuM&-7ge1@kMoFD~W>7gQ$WvFm3fI{fN;pNt$by>T4vqWcI0;Yb-LH#4jwxK#=WMK9
z2M%q8yIYXMGrGTtF8nfdvQqDj;MK!54pzawvl;S<El@HAY8!sjFGQMzbg4SL><?>p
zdDXB5n>2vXj7#OcF=BjJ&0*nTC>7o_x!;89xcx)y?elIKH!5|lLm&IRnzdf<pJ>^Q
zu82uuK%t^91reikgSb`8a<~OG9TEzOBvd3b7?B#c(ibTJ6v_cer?lpFN2)bvY3fkG
z`>HmU^$?rO#xxVVQE^Ba1~Dg`7^5aL>fB3TE1v+~rxO~dDk-}SpCTHsaKF@zE=mL=
zwL4}_1>>5y*aX+L+W<U3*GE$Do8v-mjuo|(whqvzOJMj^gTVz{J@9X8QJc+^#nvml
zX7Tm<q4|;K*M*qjeB023+X5yPdxF2qU=h!ehtHCCmYt5HH&y((_f5Y!0!duBPB00b
z`w!9&AZLFe++=SWGBNB-*3HS&U8HWx3o4X0rRN|I3DH?g0(~|e1D_k`(MN1>6d&W;
zL}&Mt^ym4;^S`6coyMbAzl3Lkr2Xf5Sra;)72bR=WfMp53F$D#<X|NW7>LbnD}~)l
z4+b2-2KweXc0$tZLv$d-b#RvcKqV6}K#ZuamK8D5;CVYuWUEvEAt->5QGw=8(v3#R
zNcH)0_z<wzKeOu8>y0F5Xu`7yA$4Bsi}B@m%VMDEM<A=O-U2H8^Oo=G2N4c0K>?XG
zBcg;9b3Ly+Cr2+khelxm*V3JOUEHV&d>eE!%!knIajB2Rdb6AG6kz8YqTKqfKa9~v
zbF9-cB7?|#=T{%Oa}8>wh?}G3H_>W-*WFr|r^Ts`@|^q{8Fcz`r#;Q%h>+I_67^j}
zF@;moiJ)+^=&?pa=pEonrVMm?3XC+Ct11&$Ipi_e_JfJECVaVvD}9(vFBc><Qlo5l
z3Qa=wn(Swm<%SAP<wg}XBc(TwU|BytvtNzMX1KqqwW}TZDdbOg<Cnqpw(0q-)zM=z
zC-E#Fjk(DU=qBfShKFtAK^G?L4JaC=F;<Cyu^F=QLG>gz_A0ufd>xVXX)aWhu{u#a
z5ui8t-dIR~;@wm%>0n+4_FFCZhH6{_K=NY3)0z}fXHz1%XJwp@?Twu_br=bnkaSO0
zAB;wRLdFJTR$u3M5@08r&b7crnpp`TL3z{Be6y*xI~I6}=>iqZl+Lgc0YTb5N*m~G
zLfy-o=iJ-xA~v_=qlvWnf7mntN>3ohRB$0J5;=1vwmAQN?7b;=oc4HT_b;;X^fNkS
zJNTe}L$vV5W{-MB;iM~qH+;59X?YNsIXu_R(kl7)y2mbF_a(PRao1SQv;jRs+LRy}
zbycElW{*4Fj_ir76|RUyySB;p7YcZkgJh<B%F}2|p5)+4wq8;Z9_6{F2RqDwmoI5B
z8fKnDhO`&a_28e_1$4MyPvR@NPg_%AX@!ney@uZ(pslF-5lqBg8dY~(V48;#?&r*}
zk0tYl?1ytDMOER$>iV{JFv=7QN4@^#waFw_#4FUSgcU85>vv^y#%0kP6rr=&iRS}D
z@fwFXXC7lkgB_~~;W!-gbtwMhy+XC9g?ZC>MQa|d429i$<6kmcG!;jj#*zyCU#Uon
ztU&d(G*~qs`Ed&KRWT-AObnv-scSf=_xsZE4d%1Gb*3y+`n31iU>0-G7atY}nj(Fr
z>JF=CerQz5g=1|Mtk*@$gC9hGy3B8vRAg*+t>IQc`CV%=I;0acXZpl{Gu;`o-y{B;
z1whgKE$Rcy(`Sb)qhA!?CsieeeBr)6Nh`q|a%y@bYkM&sE<{hM{HH<pd|WbSw}ker
zDa$gETHTL(HC=d<o>KklxjH_Xi?jB5NtooMeWb~Z&E3jk-^;&KN>WP{C)&Yv1-ZDY
z!*p9t`9FXlPtNcRiFpR(z41W4#XBm=T+=pmt}-C{j+cp<-%)+RDGoOgpQUuaPgVG)
zolQsQTuxOUc5OG+>b%4_ty3G<`FJU}yDQZ@w)-KjBl7~x*cg-{_8S_Jww8MCEuhCz
zcph{nvgr>i!al@q(u=0$n6qlUsBZ=JPMe<{^T?D4VKK}ok^*zgvu_a;bQaVwNwmIy
z22U0Ox`rRrjuZ<MW{KOCJ!Rx`)@)f%WAwDP2czdZ<~;i;6nL<-wX^g%e10jDU=ZjJ
zE5STXw!TM74#X=9X_G#inq%UJ3Dd>XH8Bl<DoQeRJ%xa-n64??G9-a|J!KQdC=J8V
zf7P%!Z+-_#0D)cS(#PRSEIwt8>!}~p@T*yrI|aibXA_bOrqgOaVEqZ-Kfb2_KF!Dr
z0i1gM(zdwiCZ?}oD)D4Ba(2lZv9vn-2?i$#xs%Fj$uv=@XuAg2VwtLRg6F#*D(xxL
z=9HH9)}sko`&QG=aRq_0!dWPRD3p)4q^NrYoujzbLpeh61r>38^;$m&ZKS`le+yrI
z^qvIvJ#ka2EU=M$j~l8;MN|jX(&!T2%DRK6{{T?;ctxHCu1-GM#T}_}+Nrf2g=kxW
zx?54%ppWD>l$@`pEJ|V?VKMPbMsba~G93!iQaHW;^i*6};!gI}iRh=<@8t=3N<%#|
zCfb_1YI#w_Ysw5GJ95&LYtNLG-N0wm=9ks&lNGP!m2_?~?3G%Gz!Mz#PJX4O!h`v=
zsWXowQPsa8@|d(@%x6<7wyS=zyTrjvVStoio^0rYa`Vu6)J8-?e~m$7@k$L)d!?JW
z92CvIn^LLg4eGhFo`v>)9%i>_Xf$UTeoV9$?%~wX`oS5-tb`9&%*q^%e6GQ^6(5sc
zg>MqqQWs*7%h0vUj$$Mk{2%30B&n?JE;phs>Ueg-7iNPoOTHSeKa;ETWk#pvP^noO
za!yvy6;P+PAlT04g^vn{qBfn#v`~x=(&?^wm;8f5`6#rq+|GpSCSFTd+60NtGiHTR
zn3DAS{L*s5;v^7*na@1Puuf(~>4;yknx70Zn{-$^GOb9D*{fa@JE5@c%4E)PGPl(D
zRR5a^haw;xF5!Ac;vFD%KTkHLWcKG;(1NXZR;ii&Ti|5;hMB)bX7gn5T)^zpN{Uc%
z&X!55rR2GI;*&KF`;yc9YH*)Taca5OLW#Ul_YPeA-)k?hViv@^Ivb}%)sVyWF0X8T
zu9ky!gqYM{#jyi?x;8xVL@sChuRU+~V&}=V`WLvIU`5W|zz|Ir<yMtahTb8|zxKV+
zE=|XE2owiXh2?e8>Bsz@7s6(@U1Ehg73`Er=Wh3_u`rXY`2r+1q7END?b2}MxONqL
z^LnV08_m>&WJPHnQO+AlCU+=h_F{`t<j&PB+nQ^=Ghz8kO5#6|D;s}pzfxR4O)495
zFj4Kc0$mV}Ar~VGA7}YcKxrvrmo?QH85sYyq}I@TI)!?sQg`8MKG30?UM@t(60Inf
zQ+j65ahUQ`7J`ibh&HWy6fK=#Nj&E8`s+t^v+JZs@3|ISRoev?Vq}lEquQyXn8UaF
zKOi8eWa8uHuj$<u!w!O}WjbqREekYcZ2}m>=Oj%h__i&Wy_h&@#VgQ#h7oAP*Chcl
ziZqr5=@}<+n?=ZxLAL@o;<G*t?Oxx)mC%sQ>uvH3YG_l=P8hUrAX~l>O1fQAU#!#m
z?E_&%=!v+bpZ62g<dqvGD7l*hILwsF7>J~l4HepDb!b9Nd5%@?PF6{FrYGWzjnk5%
zvf|x;Dr;%jyaxlcEGe3A^j}i;=+@~$4R_teLfWk3@|8rk^U9MNy~)VU%yx6FuNoqF
z3WGNdYuqh*qi{RPR)}K51wTF06qKbihP<z5g(%duSTw!omOL4Qrfj-+ezaD<tmNz6
z?SEbOwY1rKJUBZoZ$I}gU->9<<wS9~>SUEjqGCavRkL4uF^A!5HOfmXDc1)i`+M3v
zxJaOtOH<7h_>Zh43{Qk(P^3b4+UmgY-4ad^R(z>TsZ930?a8-|RF0+&S@U*luVu^=
zq}@MNF=fqks^$99>(S5e!oT2a&vI}W#dZAtUd;uEeF8_H&2lQ;^&zDxLzi9f2Q^}x
z>Mi?MiS+aX*Hz5a7s1_Iv8467(<Js`%ROBvxrG87m^$v)%H{K`1QK+E?EHS?a`jD>
zr8?zDV_#}qFy28IqfwTk9lF2SyF}DIb8RW`&Kt0Rc@Ppxt}2ZUxbBwP1_P)^>ri)X
z{I!TZ<^XSrg>ta`%tX$YQpK?q&+&@g24)+*Se~8$Aq&a#tJ|~o%uPHQXeM{YmVK$l
z*{OwpZ%qre`<v#Aob)U~YPO+lL_sE+9o7V=zbQ{@_d2|9_Oz$5g|Dkdwuwej_rXuI
z@-(M6d)?dgh}|~$IX~)_Ax*f}oW*;mDip&6UTrV-XYMo3-~F9!v5pOCtv{>|5xkWW
zUGs0ESy_&zXp*Zea{F{1(HatZCm+M(zvvO;Xj=@~%5=^O_3#SjSL5Axl<h6hW+opF
zPM&)^THIenlP`@lhnie%?=gB-mhocI7~8KdhQ*%lhBfU&2Pt!e*|$C)=T&KK0qDM`
zZ5L`8*R5r{*0R58%0^|&hG|>NnScpPp7gJoPt}ZzKvNWLz;HSwUKr4_Z0zh|7XQSN
zn#E^5G964FF^2RcZXJ9-t^A-UIOwR=R1n#p<~Xe`t-B>3TapwgeDW~Y9z1#m0y;88
zvwEuOZ`~Eob0Ytes+);Qi+Xgt`&K({lA<W&BxHsS=i$TX-|SvuOuLo#bmxK(Xy=O`
zwvq09fSRR!NPba0iu2xQ>+5pMkYoD9v3FOsoUWQC<=5_-adY*tSdye9CU)vQ81sJk
zC_y#-FA2jb^zZhy4j_~rjIWBX-dns{Cvr-KVjn^F^ebevq4w?>m(6<(i)|kSyKmmK
zx+C_9)b;<dPhQ|;XBxjiO4E@VbJi(8A%(`LpC+6%$@4wZEeU$dvhk7J0|+4OcGsY}
zum1hugwb}W<TK(W<6fGM#rF%Ern4Y1-26nlo3T6WegSJ!$USl~$qC#*Czi1WJZnP-
zw6s7!Ymn|}Av0x@L=Eur%kt_V2fd|Q^7nh5`y95H{85?z`CDWpSw9SQm+!llh#`0i
z#y%4&(s{xSi@O?TT^;3ZELBnav<1eK*lQ$i@RKFzZ1(2H4W2YbY+B~IB=->_fo9d$
z12bMQS*VsDE8*jC@*%B*>DZM1bo;VA7d?3F+DA|K?G?~faF6?JnPJH=Nafl?Z<dEy
zABKmbK}}t!Tn5eZ>uuF5iPY+m!_1AbTgC#d*D}J#i_pcR;l+ZA^udl%UKoUsZRTjt
zK8AhQ*Hqfoy^m*AqfOE>tt_p-$2Cy?Cg5PhVLh)v&SS&D{oH#G1=D3nW&Ov6Vmf_9
zEyN4<6*>tM+FJ%p0+_eWt@eDDfxr46k0d-yB-l;M*?G422be?#R!c#;5fAU@4OZV@
zXwv6_0Rq=muV1@R%=xqzZ}_OQ=sz6m$YjQ!@!J`>-uCveKmCJ^$!;`IwZDpa8Yt}T
zNQk-mZdTxOP++6lB_4aZXe_w#$stgW+u%PMDuRWZXL{9dT>$Zwtu)dT$Jxms$prV`
zoNu;##w?f5z)0{Po6SWK?KrV1pZR}xGEtCQMSKsbSCcsX$qvUVUZD1L>BTzw&rCHj
z_GsmW9aqWoo5&F3@K$r|&`6oe+DE(~4bH{z8O9)kIKFw}7*Z{~c<pd|v0<K5;^_mt
z9I4i3<Zc4^rNr=({%---g42A*5K3GAb}tGo7ab(G0FR6SqrtaME_%tQLwNQuc^0*e
zMj&U)9!q(kh?8x56c#);a2oOATP5fH?DL-<`!6Yp-DNWwRH&=djB1v0O7sAqusWLj
z%=!c#**{mQp-JiNWeFMkMBV!Ix~hm&Rjg!OCiyr(vI4`$EYL9+)q&l8xmNs6l8?qH
zscx<%pfvqPh*D#65hgfDn=7#1sD)t?r34jggLAGOj*;Xvd{I?PF)<M^R!n}LJ)G7G
zie7sREG4xsohL^Vf7LH#%kc@E7q1Ooy>%-3lrP~k@ECdT9rHB1Nsa(7KhxZ;V4yI!
zA0clqd4w!qRjCf4#;D`6$O!6fsTapbW5}akA~z{P>pJ6B#$A_8kH5C80NAp{CN9e6
z3EVX#X5?MeJNK*~liF(5>5bZDyXFg<W%Q*6M3-#}Yh!lcd=C1bwyZDI5%vi8TfS2I
zEa@i4ujaektFKm5V1w95FXzU86GMteBmAB0zV7OB#^3xMQNm4$dz<O?r)uC4K}Ibo
zxXcar2s9soQiTE}JWIg-!9#u*-rmF&?bsa6sGxg<HZd18dgs1L%o4)HceHxH&^Se>
zjF&=?j#+Ak-v+#sf#mw_YtKtpJs1tv=Y$Tr)BDEo$}mVX`sMlDnG5#5YdxK+f%yEO
zj2<(O)A4xHK4#2u(Dzh;K$HG+%+v8LkpWjQVr%9Xl`yV*7{w7jjt_2SjM--j3Hs$V
zbO>Rq0$0f5p^-FxywZW`A<YN}t;7J*K%78#@ng%kNf4-S5K@NY8-)yzG8V6e3FO=6
z-7%Ul+K!JLZBuVOp=8XQXJ1QO!My%ip_Z>;`AptYp?FY43uYb_o8{@K$6z)(L?7-k
z;dm4zke6W=t<jMRNqqIHx^IX0p?XyNEa>0c9L81Bvd7aCR^$rq0BM}SC%#INEvXw_
zP@2S#g1J!8ELZb?0A2O!dX)D@meTGPn4e+-Ut$psZ8-rl{UcQQSvvYXZL~(R2Q+&R
zel%lEaBR<k!0a6{S3f49sw6hFC7niKHfhOe;q4YDeExt>WKKj!>-CYjcb?bpgyZW#
zmf%xa;$<ds#(M+Oms*3EYyRAVBa^XFwpY(x!aW{aRE$4GX1GT1q!oz`X|7caneMhY
z{DRdUU>+2yj_hpuR6|iX7IrYE;X&KGu0SXE?*n@x^xiS#cGiBl`&C?VKW}F#_E3L1
z%escXK%kMdVN2IT=BI(`fBNW|PM!|yZ@aK?z92DEDN7K?@f(gv3)4%#P$85ab?hDL
z*OW3~sTTS+AE3xiu{Iag{p5+`wE#(AyTBXNqyuTI%A?+Vq3wrF+`SXShB%<FvkH0-
z{plO|Di)G&vTjvej?sO<2GiiAVc*&bbQ_aR8@R^DG%TYymaW`;xl}=U^!W0%6Q<=2
zXWwTv;=x&KpJ3aR`sVaMbwq44XT$m_HJpah?0o|1*T<Z4i>hz(1z0YPolMFo_8^Yy
zPtwgm&3;g#-6-^0J;VY0AFil{+wwT<WzcH{IPJ_Zwix&C<y0jg;e(#Ls@C6lOo=mY
zn3a#aG_aKJ@}E?f1{+I8+GFww*)e<;lll0ZTQ9KPw%0r(|Ctx5jYEPWADj5!oA+{d
z$;CN=*{MCEphXY<{@wjoA2Dc<stw`|-c$T`lJbNg<X=+QS0TKAS11Y}Ai3Bfyd2yY
zMylwvA)P1IHDBM2ynbG}!`FNjlfJ%%f!VGS)@<?tU6dWrn?cqQw7cEQ9wyMwsPK(u
z{{V~Zb)nHrv1XLti$1pJ9~hL-jX~YmlM~=_Qt*TQo)f&336;ctl3EY|QINe)z}vNQ
zg)q6j*+Sts08_a9FEYI^w^*sKJw7xZ4Khdei@04cp>?UvW)u3OY=n(MCsj%LO=(*i
zDPr8B1poZd+kpl?AVw&@6jg%YSYy2(_+SlSHfcJOW9K8uSx{9r36tv&W}JNeb{C8o
z4T{GdtYk;6Onc;bq2*1Z$5K!R)JFM*kTS_~HkM{qP29QH_uRZ2nrwmWPCU+kIrMUT
zTla%;u=Ug=Z9p)uq-+q-hm}u0SPtlsrtEzC;OH>I<pI9-)B>=(<>S)sTas8nuK5RF
z5WjkT?ZIc+eiT)<S>Gz2(NyW{8>}w7&A62vC6=212_k$3?ZX!Ro%7A(apTyx+&Rg9
zsKMES;ZK!*ExD{RxK}szX;{#`Z8FRjft)r|+29$^C3Gyp*=q!6CCEnZPjl1n<}YI_
zVAjKA+JVz>wPqA-JHauIJp40k1~MVWEcbF-$czH>0c@`{p`}*9IWH$oC?Y3h%uR)7
zLDcBfRk!2;jVsIY`SFRpyt9!gh59y>@+P`qyn-TIQolfWZLrJh^FPh5&v-`O6IMus
zDaq6v4BzHf*Ey$UpYND*G;A-^PF<2VaQ-o~FG+6~;EY-MzLyYb!P48X@*^T-*xK`%
zgY&wy3WelIy6N|?{?&F=%cS32C`=626`^eCKF~re?xko#<1hPNnCx6)6ltwAZ#dAT
zTtjc~rz^(3mwdwrxt;{^?||>_aJV~2*BRs&EY>W}I_stb+pAYR^?QLIac;Kdr0Qr>
za`aLC_`4M}eOt;vr=(D3KHGlR;VqgSBlW~bZJTf!&T86tUiLj_?RHVad2EdMSEG-z
zrES;vwxq9+<dv0);ClMb6J|vYWpPnWg9&F}Hsxn%qJH{0!?^7086$5vLEu|~2z;Zt
zmT(rR+^uF=E?Kw*5n0!i#`B+k!`MsD(0wvY?$SNJyX2QST+KbRdB;BZM+Wb9n?>{F
zLW?oiThw442$QV6d^zPCd%CP=+*39b^`uFV_n+|U;C25WHCPeX@?1uh5<WO_sRRZD
zF8AR|EvUG#af7x<4LOT(G<T~JpWrXIP;nOCOWmKJr}b0e1WGDzl~+>kzCUnRi3=~3
z6iE?)Ru|$+!ah7{f~Y%vU~jrV*81n9IQOB-m|x!BVIMUKU1I1n2?-)j_``L*nQ`P;
z?6&s(wkwG;Sc=bqnptk4>_zSL{NifRv!S)}=HK{rEp)|MHJ?KbSwDs1i>-d+-><b(
zA7GNxdr?~LN83#0XKl<lDhK-kk*nksP`^-5#isM(GW}AagdEsHu7u_SV;CzYc%(9<
z0XL=Z7H6dmi@Qi2)Y0IkOV?`K8hxr*0i$`!=V$3YwVnSkk;<iE?IDE|ZBGDG>oaXl
z*bm+w95dOnH`ve&ZxJswGbc7~=l+<e9Wc-VkM2PG<~4r$rl~fgxg15cfc%x>-IV@2
zxCYZ6f1Q~5T8gKEN#sijqVhOfI~WbKyQlKyaqqtEEdSkSC6<^Z?xf2<V3MU4G`~!3
z?Gt0wNRG8$>=lqFlGLI6aLW(B4_|G~zbrzrQgJZqj>f-|NKlOQ@W!RKAzv(U(0OA|
zQ@d6B6-G;U8>&M<G@Hy>@YHUHpHbL~?CWd}=UuD&G*coQ+E%`e<(dV*$0V9x+A~cS
z9E6UJOG4#K#f@zOnp^P;%Je_RwKj4BcKwQ(lDxfni+O!$wU!ffpMIa$T@LN@Rnk!C
zO}ZQ8-Cb0m+?^RlZxQQ<G*mtEYSa<1Ev~uSc;WreH>g$I*S>}geXd1>BoGtw_O=`|
z7M|87L89n*vz3oC0~Df?7)0==v3kOLLb=fs<ERp4-gD=*yZr0E5hkZFUwXTU$%(MW
z!i4Q!$6n+J73ah6T^kQf=$)|S81d*<d9Ja~a25_|tax8f6en(W=~$lR$_^xPmyanz
z8@+diqvz4(sEYG&(~`2gzkk(l6p9wQv=~)ng=<8l{(SLsQc)S6cDxFtq2A-oyOC@H
zVLG&U3#HGj{GMLvOl&uA34N!VTw@gfyfhUwK_|TA1j<ha>p{a_V9q-JxXH!wQj<Tw
zBmR0Z6g&L~7^P-?s>LG}8n<@T5^9@Y#~EzEGE$V_k><+GEB^<Co%HN~9p<=9mrUEi
zwiEB;vN!YN1?aqI$=2VSt(^lkF;{vv{;k7~W-QB|&>57fdORb6m##Y`V}jNyeb{OI
zd~%#dAvm|ESntMM+s(OEZLDsK-y-09b{5Y_Q#I$B_g45;Uo!K!G!d_pSreJsA244#
zTOhF!obtj(Tb4y*Qxoda`iak`pKTSEAkO$1iGsp{hanknX*a2zAo;M!t4lL$SQ1(~
z%;wI3Cm4N)#!@h${rG*sUb)Zow%lD_t_itaHC1P-VulG;;i|Cc$O7|*89ePUW#~;F
zP6^*TSF+;Tlvc($D;f#dyv+@4=*P}`{k39+k>OV!%RQm+H0Hp?(vj`;TmuHyu6+iS
z+N_}eDC2_1JgFo)1pwREGpi>k4}f~3Kn#-!TC(mrc|S_vcR!I#i0{h>EYaKZTq<K`
zneub(%Xww9rxmbnh_(@jz$<B%f2)p62$0sLBj#M@?WOsJ*us$$DM%NMV;>GZZ!%`w
z$+NG(tR&W;B!=--{ZyUoqE2+G)XT>F#TtE|uId~p<dEaEL%ZTIK9lf9U5IP2bix|>
zj&oQtT3l__3&}Bk0viUOmHe<oG1Ca?kl_)U;mB<A&0zSb&9QPm#z%3uMbFHyh{U|J
z+MU&;w4GEiTBdGX=JAw5W5oGbduWA`m*XC@oMd@|b1Nq%2=7$r<Oy{gb3QSR&dkmk
zXrCXe;f@s`n;wc`(<y7G{d|w<Q)@1_J{<P_GaAkUN)ijutb!KutF7Gl(%S|T=a7>V
z_}-MFu8TV=gD@fQLbBDX`sxE|LxnVrRhjz*LyWR75D=TNmX{f>sw3cpA!+%mz~Zc3
ztXGnoQ|^#3O~PzHHA}L9mEHmdrU%LHCRM3q_#(an-cMPtTS4%{-8RH;`DN|C9o-Rj
zU{E_K;ZJfjudE#>+0G6d!1iJ*4>UC!$G*Ol{W<VH>ZNAs)R?rrj{4`SAQ6f!{`w<+
zRr0?hYZPRYdq}VkSii@mxo~u<0}gmBc!`g&A0t!riSu{*9{@=zzs2%dWQ(Z7ReLEq
z%Pmg$Inm!Qo`?itd9Tg2y(d8rN}ac@DSf$B?39ceMFJEQPS&JwF#LoKeS^+_ja;W*
z8E<D_9W&?h^ul7$Zc+@@GH3V=1y;RgizeGl0e*{9_?ALA5cBZK9<>7n`P>9NF;DpB
zTYFFp^3u*<b(N>Ccj*SSG4{K;Iw5jexg+>SF`)v+Rc+H3qeZ)xcl{W#=$|J2H3`1`
z#u1K5VRVIR?Z^@&_JZj=o#+MH!Yos&+`O|6U`~$0giZ?Dg3}thRTL*mgEi@VK%>s$
zV}EBgnmdmi2F#}v9$Sv-z}mwyqH8DwDcfp?U#im!m>Pyi#dLHC?=V)x1a#JreA_?)
zLTOuD?^vM_WA`I*S~6l`rSlcnsly09C8|sSK@}fV(06tM<A;<dIP((Zll(q+egMV;
zmw5>1_x1IG0UiP6JEAqqpJ!Iu3*1}wJ*>%QH#BI@$@L)u8_Bb!)Je^A%u6aGvJ5ut
zH`0j`;ndq~Pf2IPSZGDdZnV>LANdBR*}R}+=z5KLi(y*Xv;dA_D;q2`;YhspSTy*v
z)nqtlknrthc&8t?RGG)4B9lPLHc_SEYnjyQZ{0`l8O-3+l_w`}EZ=Vz=dj5qyiag?
z0}9iMya#Qs<l8n1Eqn2pD_R_oE&GQQxLZMa@2K{3gx<ev2Xzmp-zbl7(dk90vcGp7
zzvaL9%5frj!*HmYN&9AWUA0fJEWKwm%CPH{b9C=K!mVkqJEVO(d7<iGZ6<x<a@^Zj
zKLu}weH!9<7S4-G^!)NN#1i>`O0iv13u+Gxei%hfvt5n0L%okswn+&E!Tw9Wh>Knu
zc~6c`FqKK=c;CinoSjOAuSPA6veWQ#ENPeLtW{O&Ouk7xO2p_h-YZ~9CJ;o0jIMnD
z$iJ<EM;%5uOV46n$(SY0&oj&XGm?%(D<W?`s`lT;$p}JZmLnr1N_;ZRbfCWWsh-r(
zrE_1^g-c)4xl1Gnn&UB#h}Qal>ZM+M?rl6CEPTvt>i-07DUdh+N-;Q3y5gl^@WC%r
zNV)q&Wh_(jhOr`ppn?BE!|mk`IbNV74~3>_ph%Zk3V+55i7G~TSB!WvGoK#@axbv@
znPI*?H(n|s!KSE8ER&w4cbBk1hK`xJU2YYy+G*u7*ii)6Qh;(YUyG7Y00g?a$|Cas
z*WT^-x0_wCO{HGZe1)zO2?`A#nUvK{$(@|nNqLL(zb(Zwepvdu?aCI%z5f(}oWPHy
z!bt~aygEEos0T1X=<+55eG`G8m|cIv44V6=x&+EW9G{+aI!3eht)#1h4EN$9DyY8V
zH2YC@g0oV;jC-JyzvRp9vLr=x$i^o}rD`yBnjSG}T{-R%FVg(!tMOG(7<=x(dDx6K
zoJMo3A{-@f)KAN4G!RyA22@1~iMqDDSLt;$7)tJX>0bhZ%A~2ioMfmp%W}|BQ*e*^
z5OqPbxFZ>l9=*q&=4m$E@x~Rcf8<U=rt>em&lc)aJa@3iUfA5K%CjbbV_Yz@N%RMz
zqBtP3kb`41&i<qB88wwl3G=c32Pfv-&CM9e=-1pwY;FhA-ruR>3lNhGY%ASi9)F3j
z-FvdM+m9V&8)-7os6C(0fFToKA5uHF+PW?Uk)(QRiB3TD$S;cejvq&Xy_{(*GxMeT
zZGCHw7mDsBbH5C(#_6Utgux$aE9)zS>d4<Ex{YmTn~?w#X~w6hKAQ^qT{?-++M5-n
zCmp_Mg?Nu&Ewd~|d%FQCQG?4z?pXJ@>8W3KxqVXVL)?1;X2{E|r{R}U>2mzNIs)dN
zniIeCZ(NVbcWS^HEDi5_$<ymnTXfP5MsehAsq{sXo1?rn9T<wXT|Tw*++8z+51yB%
z0mcCbzhz;`*HWE&Sscd)M`6Of^d~R9_oq)<z8u59y{B9YYYXPUa89r9xT3nG=Jp5l
zL(6)qkW*9c_CRu7M6KO=%{7<%@6;b1rS~4Pf<|%(cB(i=bQcyKO}r~%==Hi?)fZy<
zYPp}H@`nB7=qCiojHu-R|BxundKy1`^At0@@5S)*fI9VBO+?;Zr!GMLUYY@g_~s3#
z_}?zy(J)_vTN&!9L_nGIfKWY0?X(&Lz9$pRUY3ZpMlic+VO?{$YVejDjFWB|*&|E*
zIoPJyZeoJbLp2Z3H*Ya>x6Zaqu#>djVqlVlGih;k#$tE<A}1~%laL^;r+K)Pd3IcU
z`Z+;EXFw)q{+MzRVK-Im5ySl4>yv8-I_WRTetc*Ec#pJ6h9u*fuiP)5$Q!`GCNucF
zCuBmK2<5K6q_+HUjS>7stnbIoDy1VD2Ch8+xtX4Vx5$J))`06@B#b<`sQV9~)c7i#
zn+=?2^DLI4g-@`Er-jcNN61tPxM;jE20n8760vpp_j8~}Np1nfLqRNc6!a&Nysa7b
z``=m$0D3+D7jr{I)B}lI>+_PucZ1JB`P#iM#UevMvSyTGmROZzPh#L^93RR{U>YgF
z$wC}jZK;%L(8+__Rqb7^u_mLNz&Ch?J8I$48SBtfHQpoNq%pcuN{b#toS~vog`eaC
zHqX}c^RI`MTJcY)&>aohV*=`_Rk42X0c=UNEO$uxZ(4Hv;9yMqwECpWe(F=$qrP7O
z?Ha_WwEOO97<eQ{$b0l=2>>>7^tOp|Tkhaf^rjk^QCX4lsTLFWjE1OjSE~lDe^ToJ
z8Up%sH>K@!Z^K#Rn5>rY<mGe89dNgfxN+KZUye+w(@Wt4=}d9`OZA<<cSwRmPz&z$
zm@~_qrYgzUK2reH=8<va$&W&-nU4`W&chYByJOgx)3@bV;U_%lw;#0xuXby6=S1ct
z_+J8|nodguYZZW<l%0uv<m)qEI5BpPn4$eWJ^3!qM}6=Q=!but$aW$-TAX;njejyY
zsZu^=;9;}#o~97Ob0bvX$<h0^wC}}bef1;mRyo5!{v-4jOgyJ)v45%^{}xvp_27M{
zbmD{_t<pMdn(8{Kyzf7D`)jXs$k1dpGf0vcNRTiDOH_m>z%i-CGoq)^{JvOxosrS}
z_XUJ$voNFJfdr$F)YKl;>V(ZF>f<#)QAz``T6(gtKK`OkgtiHzWGiKH(?3n&(<{9D
z$hU<DVG2#pC*6F2eKAbuU6n_j60MvY0bR&Us}3`M-7rJbYdSdtOV1q4MYsg!=ocFm
zF~gO{zW^bU(cXl04zzM@xcc`9Ry9?lHt~|-tMq5HHOwZWbzM)uqbO4V_@#&o;zT4c
zc;Lj14_C?{!VmA;tvrCb-JDK`;|D!%>&z5@_{=jENjM$O*h$K=!ZCkFUY+37Rnz}P
z#qblKuu3d^z^hUXW8{^B=_UV3y&_M64O<x(CI~ZAHzhM&Gj+;#)agrS0QQY20BcT&
zWixUVEhtM*P&>n_M+cc?<zVOP+lpn_p*lZm4xA44C!Q4=4+uz-gv<@bkD<#MsFFC1
z=g&ho(%~AYWqT^fMT{!SFjG3KvWY=y16(UFMcC@tNmwPL@^8lXaze+t%r$a^LuJPc
zQz|%m7z7U6MFnAOS(tmQKh$twzR%<+fKkTWQH-u~0QiR1GiU1L(q;%_ehOTy0(=UM
z%24T)L2(`Xkn|-gU0G-xSD<GvfS>$4nzem)eZLv_Mx|h;mf7EJ)e1TcWC<O>;Sg^7
zw6;yQEahQ!twW&fTzSvK@f#)%O<0!E9`~N4%H8<0d6&g_sw0JgR5%Joe99k5*{Z3A
zqM$)3>LlBJnHOu=(hE5hg_G=!z#l6)@OBD^sZ$c>Mr-Az2@^?YR!uj1n_G2)0`v>T
zQz}V0+jVk$O(0}bAm8%hIDLHMI?EKQV<Eh-R`>EYYDXqAvN+LZ_Ej&AWuIG=P<Ofx
zUafs*w79PE(xS7wnL;tSu8m%BTX+@S-W=df8(+l#4%kl2{(Y#;11KOSf0|~8oqjs2
z)iWgx6D2Fdc-L92ytFps-Pei}nR}<220GT|Bkal8Ip4i)>n4r9-3T|lhDR#8CwxS2
z@WBIIhP!n|8REY)g})o`|A-0yE+7JH6&~Z0>qvX8Cl~)2;r-VOI+79O0Tb>W$kPA(
z2mf3Ou?26U5cmqgV6nFU<H!HJ(#Q?Gx5a~H%Bp`#BLDpEZ|}r^9ec>Ec)u8*MV&2&
z=tS`e06*z2d6YgSkFDO&ez*Al@aKPBDnbeHL3sF%U#k9-DE-%``S0%vIQd(Ejgiy)
z+6^2D=Bs55Kk;R}0k)zn5A|pfC_?FhvuxZqpREV`f4j&ta9&qDbU(ED-`>!FzVQ2f
zl=8MqD^UMCiirU+lgobt75!f~<2i@|x)y)*Z^)Sazg+)6?%jX8REIAFH?SRPqyty@
zKmW*E`WH|)a^N9&9cRsjw2P!-G_v?h(K(~%m?Y>Yn8>5={>Uo@QvA28{rCI$uS@zo
z0srE5I4S1YfBlP(NJuh705jkM$v<n^5QNBG4-BsOx1P1`pJv{Nwt`eVLaT0oJ`F4P
zI!Al+4tT!1GMi%y??dRiA<+U)aD=ZbS-?;##0mw|l>2prKyYo=L(E09??EOFh#nFG
zul)b~X8z-L^*#l77V`KRs?dM0BB>L~PoVP8f2shMH?A%NHTKbt`tM4km%w{^X`@Hv
z2M&I5ggDC2qHht*a240~Jp3Y!l?FxziXey+s7OPsG5Ww$&j|==k;!`M5Go$qPkBs=
z<)x3`c^cv%11D$f^`^9Nb-n+Q3y8V<OKZOADF3qg$f?SSsOLEJ(+2meNK=~aiKH!m
zY#>CQu)NHyvZak}5~oHccK&SfVn?Kk$2j7R$7I@sMB)GcM??+?13M4(<%Zbrf8P#{
z@D4EY&6W3{1E8GlE|ZM!joKj``@kvOgb|OJYRU;9WFDwwCLoNA2bIT$ff22CC@_#k
zY#sw!@gO3;^U1u~HJ8Rp1X+k2w$Ql|6dZpFsH<t9etmtMc?TnVmu%d#TKt&9RI>Dm
zS!l6(cjSHv1i`Ky0+#i^-0on?uQK2e@TQnV@fM@ai=u;-t?HwCAJ`7omlNHGay7{#
zXcbl40jL5J!QnDZ9X+tK?|sG(iP;B#SC=b1JFH{0D@y~``}8RXp2dGG?EkX%9UX`-
zqTkmhT!DXm5sgZp8~_IAAjo_ySN#FINuL;?O)h$Q+GzRrYSjpyB6>%NlkEs-kp}_q
zkq%)e`2-#bir&lX?dbD>w@?k_+|jR=*wd)B3<5+rHv`)dh7<#jy_ZnH9;uOyxDpcm
z1=69H%TvHDN8!JeXbjZk4nVw@qW-lhL?<J$>O;`$whm;>e!3*RCnh(aZDpd*7Jb_B
zYIpjt{#x(`V=PyX$+-!M&HtAL|A<&{);_VHe}Dgf`QW_*5t&&qrAtfzHeY`!%NpfY
z^Q=|L0QKFuG=z2-_?`Xqk-%PHzDG|LO@?ym*D{8~8IgijL7S~bbMvE5h&WLA3qJlW
z7}ldjzEtsXH<ZRbLZntx5cQS+KkBXj<)(vo!H762@(y~2{(eVU?RZ*%D6s}Q0kMFX
zstE3X<Ht&hy7K_)YY~L$4ME~HowI4MX&JgY1YL`Mu(?=0gICavZu8S!beh+_y79_9
zg9EHzfyXcxRK9)V!G2AyL<Rw>_d?#y1%G*eceAy?t$xD$)Pn}V7fcc8!Jz61pEnwf
zk{{Z=ZBA&_$Mo51Ic^cbzB_jMOaZ<4RXt@-Cv-G1L|}~X+DzJn6-Bq?rj>zOF0dmB
zrO``TJUlsyP}n-D4ah6#Za>}UXBt`MSS+yrFw+4~IOc+qdUl`PPGdv!toXKLveo%?
zThBZdt{~JpYdJ&TChgBE1q?f3oIjtHI?XSD4({?PpskV(4kipm_$UYa486R`xDv3Q
zAIZGCHoxd$5TZp4t~JYv<nHo_cN9paTAzPapBuFJ%5*3`+LcPUy)(_@_Cx>Bi>^by
zcyHSb&by{BV{7aGX1?Tj9aC`nNyo|uuX)dYVGl=Jyn4|^N6}ZUfEqo0xT>qi$KDU=
zYtp|Em{+REKwa*^%j^razK&g#j5?!|GSCTYJKmVt51p&}#YbV2KZHO1_qrKzK}0f#
zGS5zPc*|nz|Nbcr?xr1z9$IIMPt1aatn{Qni2XZtlobTQhZF^I=%P0ga{fcWdVnqf
zOxD#s`10seSBL^git$IX;f>5&pd<wZ)zzBJn^HYMF??WncXkMr@pc!0&tVom4OC*u
zl+?{Y#We`FEE{+d=db{XDsW8$2uGg11asfReb6vgJQP7j#A)OsfH)<10nUxZ&u7$0
zhTe_yAQALt2Cn>i@X+Nu0>2&r`ZqS=ivi_fxh3#qvn76nxa&!5M?A>ezk&&F5m1nB
zd^lwvt++)B;8>!_hk_zSJI#GeJW1>n*<%1lw&fH)9)7_IX5-9M4N+X6i#h^IN}h3U
zw{6ma>o7-OKj?S9&-4a!reiAr!b404Wa4tnBoul4paM3;AbcS^W81ukB8>ameXV0}
zHC~yUGp<k66FBkp<_^WC8bNa}(BTh2<%njuZ!=wVpb62yz9PoVbMBPZ%-T-^S4fbw
z%G~M2MkdMgAQt}zH483timz^o1b~LxKsUyJvlcwV9<DWE4HcUqKllpdWu|7^pi$Pq
zDVvNysE8;@q7GUxFJqb=g8z~|4@4kcDy85vMc3gVAS{zZ|01v~7H^4ks;YnF<2(nB
zrn;sS?C?JyT?&RLwdn_!EjK8&gaA3q0hp!dMt>vx3~PWe5kqGFX>|56OS@Qy$G*oR
z|I^QWvo<&@z~q$n2@t(LQl4_|Fq*`C8`?%&0~Q_+QT}1OOODG-bxnPBP<XhawV}8F
z2?|NX3*Zr8o$FHBVPAXD(00mN@`7SoAwBzRP>eMBZN^zk-}OcNGp(_5u~LQz@Ntz4
zJP##j024-VGoKkMW~jdkx-GaP-dA&rI^QW26?J{}1jMC|{OXb4cOaAq@>kSR(F`qy
z*UKbbua*H?BD1uz>tQwMUZC5s-jI87{{~L|7-K$gsLy!m-*p^=0!W=3Bf+)GokciV
z^z*4-x1cFidlXAz#{;ND*qm={DvIk<ETR8pLw}z7e}T<}Q{x|7?%*T!sz;1`z<oZr
zPNFY`m-zs>?n#FyN_ryydbU}U#M$)7v~{Z9lt61hd-!`FNQ|wnzUm2v*`?$ePih4K
zW~Q9-ZUT5`cG~)ge)FDeJ8-X3|9&6wf?rx-?q`GUI$v~AgJDFX(+P1Ny4M6vR(#%n
zx-2`;H%Hq-p9#j_#}RlwieUto%UgBKywU?QjGCvw)5NMc1A>J!30b1hX9@&#51u0K
zQ9P(i4YXzc<qk1;iYOAo<$P(E82wnajr<<Ofn6b_4O9-nEsMbQu`B59^b6(*Zavj<
zFd7=g2a)c~XGC!zxMqj$)lDm4|K0$nfCR%#DE=q>-wFLO<Vp#_a=%}Y0jklt#6Wx~
zK3Uu&p84l0)p7`<;8VV;s`G?5YNTK9u1D^kx|ik*6y`qcb^zoQMQ*S@5(6$D4*{bs
z`TDfNhZPHQC5E9xpA`4Jy5jNK>Z5ZoE@w85-D^ue6knG`XqARiE=(3S0b{)Qp25uY
z6bh0>a^r~`@lg6uwjW$a32sdH&9?dn55SlO9+e;qxM`e3C}$}CI~E(*{iA?`iDI)q
zH9EgfNp~eMYC_}m@#(1O<g#jCg(?rw4*-^s?V6vvD~RYoi+nh;!77DDZ!jO9lGP48
zji4nYJ<615<hK_03(ufY&%vI#l_CY$IBWQ7fhj0>$-{EZ)<iK>2LK?+GFb><fyDib
zAe8E6q)*-?;lO(48IvZ+W(l|DnKd<2EQpl(_Y*d4F3Kh9C1in1s|tViG2I)6+PU1&
zKS{SVJKGiw2~Vv^zX1cE#=A|14$doSN_Rl#sd>WnJ9ZyDPwRyOy(L1;LUnQ!VO&WF
zrjy-7V=$F_;o&1;*yndQ$Ts(GDE5$UejwbaUfFR*WjEQkW4Cz3Hlg@YX<bX8{jaHK
zF9|nwl|MuDiNhh`ev1h5ocOYmaN*dyv9obye>Ag{KmNjl4{nO^zdp(VtSE>|oF3n7
z{$d1L|NY+kau)Ts)__GShcE$N84N6QQ_)aea?vf)A+*A&U4`!A9+K3Jz<H|m2}V1}
z;>7_i%AE>y7*WSSY8-XeEf=Wtws@hR`1zl+#smrJ<WU4Wfb$hr#mv)MnhK4aL-d4@
z5kMV<#ccA|_zT}tvueop*amxkZB*K}C^6O;*bc252gK<(@ojnw>VsZO^`z{!#ph0m
zUpt0(?hxVD(rx;gP6dTX)f`>viYa!bN6)l`py>#Q@GO4W;@D{7_q&~<Z{8nDqhIiM
zF5DC&MM*>&YBiedo&ql1WeHxhiyHkU2PFj6NJ3y6&$o!`KLI8bNW>@Fx$VH(83(e>
zuQE6Y6qmf(gHY)*5i)xTR@z6jzC~YXAu%FWsyp&Qauzc5(7<-5Rb0+6bRSC}YQO=D
z3|S4oSv>FMiSq%=xGu?hOFC0TGKj`-9Xo|^<e~>#HqyK~#9r0|TH1S|*g^L=A@65J
z%lSd8G~x#`v`oKRpC4RP0KrePu0r9Zv3<AZE0ax(&&nFp^6|ZK_P7%bkLBy|z(y3J
z5N<{`p!0E5=nVV9&@W)6K1@AF<bV}#5F(*1hED}5?pBRo`wL?1=oM2<5A*LhmR!gX
zfRh5lm)>g8wD(0}pMY~~g;zM4#b<+mv5NcNGO6WP{M&6AX?3#H!14>JGM9SE^J9w=
z7zR#GREdtZJR`Msa5bd^3$%!sv!;Ul%(-^X3hCQ2==aJ$UUE5FSx>X$R^`oQL$|Y(
zo|B-?PZS`iGvdu<RV&FP=DEq3R&4V_m!CBUAW4Zbdz~1VW0w|SRgI}|d24=&^PSkX
zFg}kA;X5C-Oq127SvXNMJntgGuTz>PB=3=%eKTm6aUX*2jLWVq@fAO0>;3>{7Y)^H
zkijeBE}HNfBvIEKoKWO3LCdU#JpExu^+BQ~b(@N9ix<aZL?i!1$e_HgqtVdj#e#oJ
z+Rn*tSEu4wyY9umC1&>pH3{$|f2?C$TR;8^zLlH(`c5ycdnd<VTHp4dc8R)L|9PuM
z<{T@`-N(tZYihEeW%7J>bz05r@r7&#+}y@oeAl9zHjRp#4E}zbq@aK*S;bG;=W<+Q
zquPD$3GWpvrKH2V%;^Q3s5Sd{#B(nDqi1hBkmH#4u0dKi_PapvaM~tiv%%>E<r6v=
z?ztfd-$*2B|4!RUAD|oMp)*^t#ECok`t})p9yYe`AMljC)jFBZx%YJ}h3ZlAP8s%c
z2pZN>bdcxRJyGM&p9pR%*EJ=2yNlIhC9^?D(MvC)HS^3-7!WpCy#NdEY_R3!PiMKy
z4_(w6DZo#LBEu2${8a1v6T3R0eJ)BIfdk*AldpE(j_8Ly54yJr;ztblLg*%9!`Ye`
zc~SuqQnzg*_Vo@U0b_90p!%CL+=mWNCFo0exA9689mY^a(0h<|sd$G4GXdcLBTxK!
z>2e_=M(<qP{ZdEXW!PW_CpjXZC%vLRb7v_}kQc;GpVQNnZq^Svkmkqyqt!8lA3u(4
zLWUHlooKxmD9mD=pH5X^YY95tmT6lMUEaP}=5k7w!)bQ#JGV9f+X><7G43SUse?w{
zCbpdcV%t{_n~{^kMJfJK%r5F=2qYZK%v62Zz-=igk~Ep9-l|dTW326q8q2W%f?n1_
zI@TX?o1b*W`Lvt#N!?#JTCmO9Iy;$UrPj!o=3inTlvv1nXWD&*2#%dSxo~f*Wn79k
zhP<iIZh}P$E~{?kMFsKg-BHruS6|Z+nzszo(|SpVKkxa%R!=va&jL)KpSF@+JWfxm
zV}Uz|l>p#|ZI5Z&Fi~bjLAJJ+Md<y{*A+>`H<TLO3J5a<9jZ#>C5rvwZGiBI0fP30
znFP@|)}!pf!y)OSeu%1Ii&JpyC!FXPMv-{+ik@TS5IgQkAhSzW(v9f1Wzv2Zy@H2a
z=sq!YgXmm6m>1#IV}lP(`8n%X3dWs`@*+%LbrO;px^K||4HP&=xUFg7%fUIB9$Pfm
z#$KiUDy~0HQ_Ned!BJ`x=mkKriHQt@j{_x$!@ddMUDK;jd+c&}VB1#@`!P%r0#UH~
zxJodVS+%Cubdgw%TuR}y_m>_r$`feYorthpgC1b#CEv`_9v#8dlw^IzPMV`!njbFh
zgVf+9na^?BBG*8IYd2my#ZoGL%D+h?mw`D-G9G_6Pr7l0TVf70iR4i>!6%<m4yUk@
z9c6Ql!sx?iOuQYW*rPqqKZXzcFzj1w(83||%zb#PP-!krR!!$xP4&594a?|%P+Wlh
z<BnU%nL#DUSeqGFC2}6lIqD51Macx^sX)DlWR!kNrzW@hdckOZN*F%f5cQiX31Kba
z{DqI3J7y5Y^Aa8UOFGJhb#YG;^xA-q=7U_htN#r19^$^n^nptVjjo)@spUMS9bLIF
zZNK?LK!*goTwC+UWMgz6E(eiaJL$4!G!7vhMn59t0v74l_^A|Nayq;R;sf(g-w}r=
z&Pk9QIWC@UXQ%ch0dAgw6EehVfq^F#6c%JA#4=kzHAMVpKqNlI4AA$<j@Yh51|s05
zgrg?xroi_!)Mnc>y+R}r{gAue3ZQfJe@3RBHS}3gGVME$zo26ei%XIy^L;6}|3$H2
zO9XWRiPdUPHp6WhgS#+9r-^_rHw1q?(8+s^*)Uf7U5ezytPZ<G{qs;0AmD)aMH2U_
zys0lq3<Za_DkA&S2CRIRg&^cj%<Y<IacjQ!v2g{9vzF9ZB}1qn5HQ6cz7z!c2$lze
z_n}hq7|t*G=>3`6OG@YdU)3ke#ZOHTeP_us+j=6;1kO<n6dN~H|9k`r5iX(A2!tjQ
zvfgkMgDcf&(>gwQUvoJAt?7v1+3=&PD3`RRqUMYsfuF!B@36w;UMMXaPGv~%&gC-X
z^4fJJOoz3OTQu6fj>eLjSlbj&r7N@0Y)Xnt&bzMOk6WY6fIDOS?mIS9QR!g>wX>1z
zD@1E~vq}-JA>3HD%V+Z<;U1GaVL79p>J*r@Hh+!~?Ck_^Rn#a&r;^mwGrC9E{!t&l
z0pgQqK*f{FxK#0{K8dq=IgI%sRHk_{vP7qR+X9iKEadXHVf?EUP0Ek{oPA7DI#O7f
z7j&S)tbJ}-yw><H?aGHU4~ANUgABXAe?_W0JGA7zZs^-P(~B*doG-WHBfY{y{bv6G
z>+=yd1;vN}bc_=hJ9%kQ|4-}6_I{)L<>cGY=WUaU%+;e(<aw<>_FL?N-HMlxI1B6#
zXeiN+=tuI#x82-l(Q!5``@1>G=@(6pL-x-h6ul))_OecDMt{b2evjKX&r?0$ueR(`
z&gjlu`L%zjt0BGMli=dL+VVFGfOE+hw_qc<Zv;_Hf%9z~t)M0S+1UTQo>%mn^Ux5Z
z>247W)DY@jXGf4Whfv9_zb6m5Q~6SV8*){lln-A2am3ore?xWWnZ~+dFcm&nBMCdM
z80RfP^TUb?5FTQdzh(1S^SE=QIR821slqpAKW4jfO}11^Af=J0^jq*w@7*1@e!JWM
z$Jkp(Rkf|}<I)?c4JaYq9nvM;k|GTv-AH#xcb9+&Qqm<LB}i{Vy7Q2dlG5OB?sLz%
z_uS7I_d9-nIZ)Xw)|zv!_kG?c(uOH6)~!!{Dnr%<9NlKn)9z!97N+6_9emRn^sk@!
z*IIAr&Qk@PXEE={hc68uwzg9Fb1;}27i_x!zK?MfX3}Oy^DJop<u0jeSk0SUzC68o
zdf3!#%TM-m1;8~N`XTGaxS|0v2LvmZIfLWa*yNU}{+ulBz@AqV`#882)-g5MFqqmK
z)Vk%zA}26+92|~8zjM7D=UcJ1?k9w`h>v6J@?A(ZOUz^LR`lzuB8IoTyG%}=AH16`
z-b7XYksdo@&cgisXS+=X(T=MC{mw1@f%SgvW(ctB*)s0og~H%4Dl7GWgdmulhzT5&
zV%3yJTa-^fvW9+Z07R`6uX>LJ2vzY@@!@8n2paOOT)g`;PYJJ+eg+w~eXkdaRhY9>
zrLDlqk>%G+2_hCXY4T^TpeMLoPlLtLR^tV{4}_G%2ZP@?V}MKf?Vlm!R$Bavj%Ko0
zhFkr4nlqq*Jn5*{#?NKms4O*K1BX}wTf?j*1kKq(ScUB)FK+9w&>ky%jw0jHNo@Uv
z6;r+eiuyS!zjUWqDvYbCtj$*Y<IeF*CVX<)RF;Q$C!qXQsdYwRBEVza;H$r67!mv3
zI)%G*wJ;MR7O|Aj{Y&6F&9QijPZwCr#%{<51eXOL0c%Pa_jZWbv&vVOEr70|pz<;J
zt%^_Fcm>&}g|WCG2PNAZxHe;0$DwJ{*|*OBWxc`F*egwYZv9VOzJE=CVl<K8k-)=$
z`RIcftF#v2&<V6Ygw;l3dFxj~?`%SfTMX_Tku*iz@M}0ykB(0d$%^?2@SQ9QNs-{i
zk)r%F#Fu>KLO)fX{d!3<QrKm3x(Gyl+5S8m9#gJw5nL1&Ek#d>uXeTDZL;#et-@_?
zpD{^HM!k=%`m*34e45-7B*?T*WpsH58lz>NnHRP<4JApT!lb4}z!?JELovQBXu94O
z0pFJd2R>a3G7_N+;5_kRORhbIg|D!>1fA@$=^g*FXV;OX6uoye-2FF99Z1RTq)oV4
zU6@IuNBSrNS~pj~$!y{b+!w0a*}Y$c6{9#MGi6y81YZrpp|l-OFlsXcro*3w26WuO
zfIHo3y%$TjyCPr7lL#iA^{uF_E?6hZ?q~7d9~o}&CVC?5)d0h<<hq)`ROS4QL_MU>
z?%}t`J=djfJS@*sZ<x(n6kB)RWFPgRR)Eub!BuCY1I+F$q~ih7KF&urabD|HCI3an
z=M+F%y81z{(pKyq0_>>A9K_f1epP+dq&O%!XEBUo9pR2i#&Ljcybmmx8a}cwEEl$q
z-2apvnX$h+YMk2_upXyn$+2^lRW1HRPSlW~K07H2#*$lW^38_Y1FdOXgGJgO#>gj@
z@ldnAj#WL{3>_TbN&Oyfr*_)b(QX8W3gw1JK76}1flm*%t448}_cZ4mw8gZv^`*Q^
z)6n}01|D<a-ql^~g~%9+tIn6IOlnrbG_EoD&H?`>w&ChMmjXp2pnK`*$9{O{Ph0lo
zSDm&(cz7UbjOjL-mF;D|q2)fTNdNlrPhBL5mG^F%yT$2cLcOvyyu(WC5?<aQYy{5k
zZi}rinW_6?=L&!nWo5K*|Cv5Ak#$4ObbDK(aIoH%8?fJDW!>MxiC`kb4ZIs-s9c=1
zoPcaLf0^#~WMCdZZVd1+k$r$>@R2jxa<^3ca7&F$R6%<%!#iwergaNJu7y}3{j=qj
zMMo?xS8xZ;NoghOD8XcRWk~mF{%-1&S0L9n3Csc60YhEj$vBeMrevp+XV<~}(*wUh
z0?`P_cgiqOBH1w~{fCS{$*npUR54%ak-L=*=z?S!(f~oy$_ZCB(W^n1l)kdcW@dF1
zgFNIhX>=*CWgD-es68`aD9w%)O~ZH^cUGZCe+zq2Kdt*DVD7mgq|TRJcx$3vo5f>G
z`?1_Ttn$w~QW<f|cVZ`zPIEw-EKlRSq(w1C(Y1z&Uq0@tC!Sf>dF8Ykl#c_64@q&W
z26Cm@iU<a_IRvIj+Dl1K<sxU|ndPphL+K=c4JNjk%hB$wNv<Zgk!F=+s1?dxkJFd^
z5@OM>>v9v0%lx#so98hj@$X8p5}$(NZyuyK^1F-TP|vKxole`l=->TqH$QFHn*l~7
zsD2XkeeU$ydMf~UQ@zY18OAfTQ`)Y2*T>i|I@Ysb24-)nDYS{(d9UF^M%rfiISr9i
zgQRWL?p})8y&&U3fwBTALK6IXMSCY^CNA~pPJ&PDTMLUx$Gpvt==kUqM{AvulO&V)
zSA0J!u=ip#Z)Y<ieD7tSGaZH-+e`Dm&=F~NV)3{wjJ`Pka7?&PLHCoVT7F&fLp}E^
z2fXU8EPqDrjaUcidajtC)?i-P;6ulyn^D-xLd7uFvnl!YWq3dNT6R;c-cW-t<?nt%
zIx^ZJ{$i38;te1a(()Mn`V~}Zz{z#<PMRN4S7=mwdyBO!zd7Q~HT6Iy`xgIqTamXy
z10i@4O8#_lbnrwuw57(piPICA>ytAgPc1~rE($jdg_mj5aQt$-h=D`L5Rq`1#kv$%
zE>&nAQcy&pIpD5LP3vE9*d2+Fzm#ihE~ntHrSeXSd2D_j;_9V{*&mz6cL(4y6`1z@
z*`o;LJ1mz*;B*Z3VC%LIqGn!Ts?k7hQbf63QciP|@W<b|STFK)cYHZn3_f|cG5_jX
zFRIF8XC7NJ`|+UxiPTdiYp|7T<l{u){)_XSH%EgMeoW>D{f%LT4sj|eLn(yPg_d<s
zw+Y95Xpoxqck}VaNLGUq@h%PocO-BmTfpJ#{*`&*V7G7A#ugwg)&|XY6$K-*H>w{z
zDRK*D(<G+VD~VioWIb5oH(HVglI<Jj=;weu5byfQZpDX$MED?6%e~fHpRiU@2!bRy
z+Vb(jI0i-((mEZBRuo*YtnSdnHQeu_xobssmpjOVx2_bQ#d6T_;gzvKP&%Fu{=5O*
zy#7@l*=Q_%1E1Xei=SsN%Uz*{zl$n9s(!r7ji<L;+DIn(;P9aKqE2B5x7I$kDt!kv
z0+_xvV0U?DaILOW^&&PSF!B46YnA+UQV2{w9j1K`x))r;{XYQL;NLWfeB?lP^Z;7z
z_)DHb8pR>%KC?sk977?9{72TFG8=ow8sj0|3b%yC$%_Idr2zio5V86ObE1!!^xw_n
zF1(^Sy;fXH#Z4vqXk#PU6i>@(9F_zvq9r#Xe+~?K@AUh9+8JGX(+?ctKflm@Cy=bQ
z6sxscH8A*Rc@e}Fbdo58Lf=<m@G4_GJGIC_JTT*8x$9*#3ZE0&$Uh+k!_cBW!$R=J
z(yV0ERI7fRcn8S}WjLvgIRc<UY;BOIj_>wliJO?dnuSww`S${{d~CDSNH^&HZf8>6
zXPx{s0^!h)qXD1ni;_2+>24)hEdqas;H$vs^rg;kKwgT&mGX!5dWO(Qo8zEX*$&v6
zPuY>z$eK?Y(W`&g7T+FK{S^<;D}ZR@M3+AFEeuB<=WN2}S22{XxAs!qc1KG794a}P
z6!YD7FoGsDP}_6t`@2>}A>ZX8Nk)uno9{&m-UO{3QOCH^`(EpsMj}Ia_u~ku_Kn|i
z;y-Z!u<gvU>)wZU@bP*x+!>pokf}@eZgQhPIo9`E5EpVtUTm>jP*K`o`3GMx`r*b)
z)N@pUz41tB)On-|sHJGg>I09mbUvF$)v&BfL-e1s)^tOuOTrE1GKx)I(g%a%u*W0u
z#pOv%El8~k?u+T@E|we5pX+PX+wtQ~ZsIW9N?bm8zV#A7e4VY?7b%}xz3$<*J0T)4
zeyI0&+~FK0KbVj`dCIE)^JtHj3$hNT-?Is|pVBgw<m7!mX$?Csw79;%Z%G!kHT<;>
zyba_hZA{oz0L~u2GfMgMv0x5=vX(2gje12tOEcCk=jIn}l?rR!l_1ePSRi4y^GPf0
zSDu~6JfSvV>NDfZT)y;LR+yvz05s%65aLvSY-h*$j||4F5_h4u!?O%!UzBfDfB)nm
zgBnt`Nj+=>@3nD>xi|8LlUNaE38|+BQ$mXG+|Gh^=Y-f6fGgR&J%g1P2wzcN6Q4E5
z3>?n6Ein8UYPyb)<vor~_QX%p+h9-#ck?-$)~B!YZz%T0Eq0I3l-J2c8rf@Lm-2?c
zT5U46(_IC=EyZx(EdWGZ22MwUWO%|uNB(3HU_uZv$M0|eXv5#Giu1?k{*GP3D9n^n
z%%7kAiBs~oEjs(z$$8N~V0HeS(;VpLQC6g8rN;#t_?o3pO0VINI+YVtTu<xK>-};i
zV1?0?ICGS6v{Z_N&-L?joaMf=1y>paPBe|TS=8!Tdi$+81U>NR5<BK>I|liM<cHJh
zyNAT6`)bA4-0fR_Bb_MHqW5aud2i`e9w8g_t41(w2w#n{{NTE-&vLDn;;%BG<~DyG
z4`ZUs(^4j><?gT~<ZC}A9np1#u`IZypwquqDvcLdgABHd2!4}{7gjuD9HxqX2ph^n
z8X!suObM=|5#gL(FJ&bkcw$aVrb@O`%q*q1#oVr-eU4Lz<3@KQ!a;rungK6k_iLs2
z>l(7lx#mE-6qHAfjFtU^T;_gL?j%iAm>LZ_ko<O-(1yftx#pu=b%I~T5PdsY!Kjrw
znrC&o-XV$(KAR7xz?7NqE<=}%?!fz{{)Edj=Rx-|yvmm5YFZKZ9us1Oj0vZsIG^z>
zIkoj?*kU{FD)UNhrekkiKVNqEO&?>9Kl77X_Bp{Zp-Ur!yL%4un{CC5bU!KAZ>IXt
zdOO@Z_x+idWJb`a@}h3|{%a*6UFabDyJO{4XBl73g)*V}A-m@Fe@6nrq~7|E{r0Sq
z6PfqSwUc^es=GyaR_}%F{MgTrJdj}|z;L8lH=-g5o8cC?V`zmcylxHt#j@q&opJ&V
zX!8_!0twjZ_R3g`+H-3%31$MkXjvKTuc*k#w^$6TIA0^lEwhA2h6lVsOx`<Roh4Qm
zZ!;%T?9?AsyLo)ZT2!lM=8|%Km%c?A7Q0sY+F}j7ODPAe<k`!@tMRo3a?}L=!(rO<
zJmsN<WG?daa;!X?iC2pp(|eIL@va?3?47b=Y8&t3_naCU5HM^r4;ocq(r#1n<pv)<
z-5bOc)#KaTiu$-cLnN3S&EFJcb#cy==O8nqI75^1QuKdu2$V(d8@e4yg{;4vBL1x7
zs2ra@D^=PkESs<d(yh+Q7nLw>KK<Nis`r1IWb0+*QtYFw;s`R{=n%0k2z1Uu3^BXE
zxxY_WzE~)YTdRumA6*ARKhRlF$Yv7i+dch^{2ld1q)(QNlUB?}=CVuA>U@&=8fiL|
z=&&W2HI@`G!#|4I82FV(^8!;YcJi9e^reCWgh|a2fK>S_gs-oKw5#-H+}ndN8EOEx
zYlkR_{F_sQ`TzlSK}X2s1&l(j^R?c>q53e_7^5@PkMhB5z4!V2koX?IM-=;ZA+T^0
z9PV`)>_Xfq2;?!y;7J~Vrxe8XI%!<)GY5qvriDjRBMYRb8<vnNuU`1yeg5gbhtlyt
z_Z?5;W*(Ta^BeQ)39<JPnb4-yCosKn_5<vjxu0xt=N&_$X;>XeV)V{9qL;aav7E3^
z<y`?n9FgJKKb4wid-il3**b$mG2$$PXuCR*`!GL8NC=S91dm7kNMQEa+?7Y86&`pb
z?9%alkWMFWKRh9bjxp}W>$r!|R`sD43JfDcBf+Xeqzu#S2k%IUn3p-pN6_*NFc3k}
z22gjSOgmE7v*De*a2v8%0_A7y6Ljm`MdR$Cx<Hy0jW_|72R(REWCQbnMxIon@riFL
z@X!9<uDQ91!EndB`w-xSJ#mixq9b}~#brvaG{D>W5ho+E&URh}HN};w<@KCRVeI;=
z;vXAyeKqQFRPyGAsqVsL1k#r7u_jBC3nCj5C85sq@ka6i>+OlL_n{)+m=W9t2I+-g
zS+~fu*lRZ{hfQu{Fo4t<+2kZrlPA{gU<Q!2%X)|J_SSzm(#*|R!CN3KP395IB!!#v
zsLqiG185a*SdYLD;!EZW^4|B%Z4GuEcJcUW2}m1!BGCiL&MxH?_UMWoU|u`1A$v8%
z&FbU02nB$0*Lumi;xp43AuqR42{c`jA38z!cL<w-&Js5{mVU(xz;JKoPkiA&W6;}x
z^I;@|q&)8%nFK>MwIGvM;!v#w#y=46agkGKt8;^r(kVe)?xTd<=*}c$Oh!mKZGVt|
zr8FIf#9EL%G>udm+7ps{m%Y-#J$>dnq7>fb@pDmQgnMFXYU=T_Q0j`%CZqs27po|~
z<1(p+*PhB=jHKI2&id1WwU>p`n2IU%4t4&{q3-n4y4GgXkfdFzE2o2nx^TRmv9)o`
zx5kt<oB7$bRZ8f6m`3HO?8v*^iLsvd0)f6bbgOTGyUfH%+VD-eC0}A9_5js-mAhEP
zgP8nUr70Qv+j7ZHg+BC7ip^7y=c3qdbF;|drVLu=EExN%d4USDrrFj&$Mm#l5{E{K
z&9-cMkDCw5epU~a;-@Itr5vOCBF>e4WiTH9S^|{I$NE2(38`eJG)3i+*1dWoViqE}
zL+rLtYG_FLY<>{$y6)ogefyEYq)}ug{O+O++h<GlTV~Tj9aZd*4A(N9vR4sdtzjKo
z<%^j~I()Como*&J3n3aZmG6GY&<m6Oc*vBlLFbrNPoJ2&ObtA1koHhW`xpAFw-V`E
ze6{Oi*5t?zDDS(AitV(myWV{+@cryQHmrt|nDdqTRy{rW(@>!?k+3)q{g6%<BS~B&
zhf9#Psh{B^vyBY*Uk-+wt))t5&xJ6hXl@_xSk8a(cSAmbI(8b7b2%JB1Q6ri()<C5
zd4NnL!l|x|Cc4*Wh7;T;(nU)<-!Cv89(m`K_dOmZ077)x$1RMpK#Yo6q#M`IgqrL7
zymdOoG?<D;oUs%8gaV0zJ0W_$XJtGj3#9l=h?X5Zd`d~?FT_fJG2uX|W=<SIrc{9Y
zhTey&QsVt<@lY~aR9LE&sUiu1ChOPi%I+TkcvA8Cc=ccvr|Vl{whJjzppAPSPi=Gx
zdY=lu9~(JgbukUuejb_}?D^}CpAvPdPBEATYWwoq!qhf_j9&!IajCM=0M|-6Ss1mV
zCacaaYn_=Zkb-sl>8d<2oyY(!en*0!mn^|M$a^6-o9Hy2Xrv7$dQ7L0M8p*B0p=0d
z&e_gHYIXPLCIyPeaG1ZlV_akFVMa`((R0&Jd!Xyf3M$c{&lF+56rVHar@zA9iF6kC
zlfH{xvJ|_~H9U)*NauCZDO`@rk4Zz^<?sAtSJ^Oc^MqoTbgc^!S5AAAc=u@;H`VbA
zqfEP#XSCdFPe|lLoUV&4zl*>8;-U2ZUEsdWxfN~qWGJ2g)2oMbA2?y@fFz?eNc;=Y
zee#aAPJCP`i`k7H<2J?mIf=Go%mNXCSouAthc1^y>o&exy4p%oPJ$f&f*m&<c0w<R
zOV(mQn~0X*;nDQ^z-y~j3qp)aZz<*Wy=B?AN1y@wIzxGH5-kT{n)}*hzci<WxZ~ni
zg#Wc31Q017^as|Et}FEJowb+Se^ONyi;-zn2@#v0az_=DLtL#=6e7Yo-s>G@SbG#&
zSJTs#sB%f-&}4ES(PqhjM2v7nuGpDFX1KuQ&?-m1<HAk8cV?Q*m=!v{PI_C?TpUZG
zDKFqlLDK6o<oAd-fvz-VEN|7ksY<Hh^DUc+Wk@?1!nLeh#^_QRia+5sHSJ;5eai1{
zj>K&qqU>PtPSr%TL0=L*6WLy6>=xESK!rc0E=tt?xg@)W_1LWK+q`(?5RLfmPT0{*
zocP=h<GF}P)I8iYy|gJsBq|p7`$G1>ILjQnVpJ*ElT`Rm9o&`)6YV>BJB^5&Ot7P!
z7gUgn3uBct8V|q={{_GQ59_xL0(C9Z>axe}aqgUCBYCxkg;?vz<kdrZUn19KMAs@8
zt_ANKvV3S_NOa>(GtuL(ll&nrzC>H^bh!t(md@P9l(MuBr!|MmfL~BS7bjn*QKEm5
z-=4M{B~IX+CspWTEZB&D??-WDv$1Cp-usl<XG60X{gjX~hEqd3CSmGe(pVxhPCRLi
zmkIsJl=y#0xa*@);diTWADW*q!D_@LAUYBdBnhE233*Zp2(|<aM*<@AL{2vX0+2+l
zE}V>6h;yrbLVr}4;dHFh26Bcxgo`QjS92X%BW~M%rro84j--ICh@~;BbY1nL^C0hH
zH~%FQZ9gS)4<++B@37<mI#*)FyQuTLS+8|+*2>1UhkT|y`zH?LPCOoxzUW;2kIgPx
zbazrkR<k~Gx)mT1rzrGO29HwW=S$GD21Ct*pwEJ!OF_`zK~@DCbGtH`<N>!%Y#22w
z8#&OjCt9^qD&~DBENlx{;m#Ub^d@2(I*+kbc??jfx{=M)A~>0~L5!oCoyo<H&l*#*
z>MEcIqhC_Z_Xe?+Z@&S!Kel8y7T1I65ysDw*!E0s06Hf}I=;hv!umulC?TZz8uXLQ
zwp)!38Jq!wuSzaTv=!o6B~b_366%9l&uqx3<v#H3;_aGHg*G5-HY}_7(9W9&S<O#4
zGwZ#0#80&oR+y%zxLKw_5@gUXDW6~}h3F+Mm`?LGIDtLb%0U5*5TXhx!SXT7IDd%W
zj{lu_KFt_M<uVq|c?(uL-+2bo0wcv>U!WEBY^m8b5(Kx10!K7&<TcVtURdH4kp>e;
zBUyLZ?kdl7>ayC*qF=E>1gnqVGvH&PIZG{sQ#iG-n0=$nq^VblK3G58ZDugqu)_V$
zPqmM#Ylk`MCV4sOUa27f;tvmu{<Wpa7y{gpNfG3};Szm?z^H952ejMp%xGj1WO&Ii
zo{His4+ge%gKDtKRCTPHG~|fTOkDt<m5O>thx+I9LNYD~k^SnfpakaeD(`b8VUEoE
zV7sAh=9to*+(RX!A~J00koyOwbh?*VwE{Db_<J?al)-06(>PBE`v|N>&^#J!y}}08
z#5e{3IiJ6%KMdQQ$8n~U`-M8k_(a;&eZef5uoi(9CfiDx#kJ5}ah=q}HU=WvsrX<P
znU$^iBO~v|dg-!7^K#hq+VpO#r0CxvNpHQ|8{l3_bD4Ya$c#o4YOmt;%6wj%EkEM}
zL3R1_1`OnkofM}l5KOn@2v~%r?=6L{s|rc?%6?Iveum+r?)S^hC7c;)UlCGyg#5l~
z|BV5tlDekk@0w_~JY*((6EvQmh}|EOY&mA^<@xWK730fveQmb9ugosV_l@9duLe*f
z+8Ao7%W$m%E3AY^$b*p~5Sc1uyA|<(%Aq#K?t0pbXRG&8bkMc*%jY&GQwHSklyVt+
z??2gDX>b((2$?K*A7UEiD^lh#Z?V!hD6A}zq931~{AD2Bfc`nkpudDof^V<vw%27?
z*ZL!Q5eK96(}OItrpSY+f?z?xx$aW)oU1afqBt(aMz^})olC;h17`bo@9FV2%pPX@
zSUL!fsHr~XX9{VjB)L0%pW0ygL5&Ymvf>;RR(Wjby;HafAR|a^A|0%sQOI)8fbR6u
znao5{RvWwjM6cOYi(BexDc2as4yW(n8TWyHyAWa%M4qAZAEy&Iq`p0%e47^J;IWP^
z?)(h>&l*AiQ7YsuE9>^KQ5`|E)4vv-BQj~=N0|o|H(X6sPeR8Gvdu>ygp$l0U7!kC
zarRHVgf$Xl4N2nl2h|#f7qN1F1q|*d?(P$@4GAild8Ib2HH|4>&w~v<Qay}aI^$g+
zj?1LsxYq2VIOVijttv%t!Okeob7RRylT0E?d~&~=bihL{lcU^~1AQap&n!(obwoIG
zOxhnYEk^ai_4dPrf+Ns6jPNZEpwdo>G+iwF@PmUZ!B8cByCLTBSddn|ubP-6*8{y3
z!H?7<brU>2{F*#9o1wbXAJ)CcN<&du%lJo!<7In+{7G<T;nje<I-8Z|Ze}T0B8D0k
zM-+^&Lo>0AnA^f?VfyDg>*q;!F=}7FjviTBLr=b<Xm9Ij6Vr_f&3CwNL=U(>#tI<i
zRxw=MkvF0WvGqS!?4lrYe36te{7c7ZSCxv3zV4|NZP5n_A1<fw!g)<uZ4NEoUn-qn
zX~O9o^yz2l+Y`-UVp82zZj|3tc=`W&1b_X3^cHBeQ7Q5NdNgOmh^*o;2m5F2o<KIR
z`+3&O_0KZ_{&+g%UY~6AE2>3DQtrHzdgao&@>^|3`VxtSKv?j}Q+u68g71;uJj`+o
z(Y?iW$<lS9jLUgVWt(IKP<sLIq{Gt~w%5)DD~|8{6IjN`EBXdOrlVgIG=U=TPW||2
zT1?8GE+NMNYj@PSvLO#)g&$debU`0PbY~m9b-|;%cxTp0faSoP?iDgCphA;+q@8Zp
zr<2me-?`GaIcu{UA|mx6BJ4KA^pz>i{N35*51{gCvEPjHj^PCP&lCZb7kLsI{l&-X
z9oBNNP^&m?;xv3I;5C{Posd_W7N7pgTfr$;<vVvu2~aEXUsx%$aRV35eC^ZD3KNcb
z<g9sI`L#ShzQhT1=dTBac7${?nkBZS1Jdl`yseDX3&`3sm6W%mMv3@o&I>4=xtQY6
z!ozXE)N+fks#9*5HHmIGresgyg_JpEYiSer{1%#*xQ$d+pR1I-%olxZ9~=2zHZMfz
zgQ@8O=dWj1bI&_hfchtOG|uHEw{awaSnR`--lwhZLmYj-Eiy`_o0<o_G~w7BQLp)R
z#(9MiS)d5=`=j*mn1z2AX5a~PN<v@f#>%3hMq>^){Rsa`0*-g^C=uj9?&SvLQkFzg
zn7YreOe+h_NK0f<=sz$0RaU?^?kIINkk!z^#ayW&1RSK!BS9<&ycmb9TiJjFm(dMf
ztbIGDlJ&$v$5RxfeR1Cv(@cdV-q%2HHaP;CdPG@Tg~mH%1tUCfA%wNGev|~QO?gzd
zK!~LRl1Z_9s9DHsPzl$}!hCLDKgD4S3c+pX_b{Ns%hN^S(kDQOX$|$9Epze{gH7ef
zi)IdiiBTkJe7STTkJjr4#87*l*Y$v;zqLkes=xK2`igiCDX3y@j(21QuD<BUd(QeX
zB!=OGXI>=4c;yeY+G7wINe6AJRU+e}>I<fKo-7G7H-_@*3}^DP=mNm1t`f%bBTcIk
ziUOD55s-SOTYB;@x+#;@W~a9~9CRR57)}l*G$xesbP(~wkfnKd9|}%?>GxR-ILoKd
zys{*fCGYKRlM~i><tIbx$sl*#P)Btsvm}+>=ZPbU`y{6p)3`@WKB54t^M&1r-~3B9
z3h4WjUByl$DG1vneu56GsFO%_K0DR0)$~9VgFDoCVQQf!NAs{FQ(#cNR0#Fc;&|E@
zZnyph<3~o(bv@lRiiR4R^z+6q&kSP_E@aF|E;Kh4<9fBO($=-O_aN5@aB0gl4UA(m
z)p4RoQRoO`F7Lg1+PK@}!{eM<*{j5Sy<`B`5Zdy(rKq2l*$BVu4g5h}8+%h~Kbxxk
zs^k|$+D<S|%jvV|76_<J2150xiR<q5qPujf@gkBI+cy(4Nv8gfE#yZwc%R7bTu2G2
zk~nf=rGQ9Z!^)wL*?d*Y6=m<W6C?c<(VSMbi*hoCDL{cS4gn&!*a)%zTBJeH*WcjD
z6Aa*jzF^Z67dwp{c2V3rjyn(@LKKVSrv9~NUqhg;ofhVym32%YggCu{Tj{*%pKlMj
zKS1WLb|>omcO8L{2fc@Y4WLq_oRtO_`1A`eA7dL%N}0~zU940^M-yoDhRjOPitj<X
zarMu*`4h6L!?)kcLJ4SfQ%V;cd01Av#R#5_@!>5v<&Y0fr(#dU;3$N(VawG3p-n?S
zupfI559yAgJBMbUV0M$h)L!FU#iMxfJzoS|k$%1fyL^Elx^`FpZhuKUAHYe@$R&<L
z(Wbfl$o2#Hr^v(kR%SpN87)?F&}Os|Wm??f15!NY!Kd@3+B-hL+3@K=Ayk6%&%URF
zxQUmW1wTzkDVN{fQlYUXB&S<gY7<nQ4gr>L@!d9D`q~e?T&H2G$+5uhg<(}*HX_&R
zzMosu9C4fj;|fslt@111#y9~Sv$#BvVP}x}M?yO#t)4f?DDsc}{A}TIo&|wIeI-|B
zxbl~CP4kuo$CE{MLbadD(KsY7iN|oUgJSV#p<AF(d1-<k(hBaP*o5$HV6nur{+h3o
zQ(CGbAlX>Y<M^4Iw$_I~n9QhlsRzJB6kpVT6g$`WQ691JBGANUIOVe9;S2^;YN`0c
znUQlp`emn1Y5?bj;m;sPxMv4vE;*&|u0d~k3LEUWo5P{YYA_<*+CT)DcH>1fIgwa8
z2;38ifbEe%nOe69n%5`u7))OZs<KO-H<Z=&5VSM9VfnvIMUR$CwS7Ftmr=s3UB`0v
zJz?dI?>fo+MGaTnqVoWv+%q7>+S2nd0!s<~iZK0xd5811xLwzXk{rlO-jdTfxS`xn
zgsG3Hi3;2cS47q-64wjN)+-9@$TaNPb&WcNPfO3MLqe<=LusGq-~^HC{i3`*PdY*)
zm}IHDI&=9hzF2@0{j8w9PNHF~)H0+4>>`amikcxRlF8(R{zT&4CDwyvP(g8Vjj8Sw
z=Y#zLFThQ}Ax@aTdhySfbc{H%Kaeia10{|9RkrtT2@NMNx%t0N^1xNew@W)tKF*!!
z4BN*l2XRo+%0}<TzrX#PmBd>V7Wpga6CJC&(Qwg%=q{*0_07JeSlI>&gzh{#UNnT!
z>fAk)5|@N$X->E@A!w9T=hH{A4r)tMl@=eK-0q@ESE5y%_A!56Ku_s$-Jh)~U9}0u
z;!`t$G=AZRJM*wR17THbKXX}m-V-x<oNi)yJ9um$O@$L!{m!(U$d6BwWQ^;oYK~M5
zVEP1Rx;#HacNm(2C@pIdj2!?;Ml%SXD1ic+Jj!Txl9U!(dRY#J^jPg|G39#}S@zR>
z#UKgmxp~LNg9;(mqU65W8b<7$C{d3OCCvB<-<6{2FgAeDC?*Um&353p5-Lv_56y8E
z43`gTDX;5uZ)^erHT@(tZDRZ5##|ZAId&q0XqSP2n0265N>Bo6Ri_lhI_`7kzaU0s
zUBcFttw2*u;*1iAoI(AjAHQ=B_(!BTF5)Hr))W6Jo=>xZ*GbASe_I~2iZ(LDvPE3C
z+;g?^^TQ={g;k@_R~%LLIpOgq5%q>*pGb@in!2Pp+|}8hTCAoKzR$Xg9Z%WAIa-f`
z)=UqZwx$+ztyyf*j7c*nMfmJJY&>=o4odB73hNaOP5oacqMzmu0Q%$%lXil~TuLA*
zW!dDaR_~66QjX{|LHJa16Uv`p`2D3q_*IA7bh&PS7>PsiuZMR~dEZHXt?BCEI$7Z)
zKFtBE`KsE3*P3&cEm%@*9ntp~Ub0{7L5w-iF6<Tl{kK#sRM=y#M_KTM?E@;oPgv4z
zjachGAJ*L|5WVNI?q*mu<ODfr@b6OMny)i-FPpPz9w2@`W?+5we=5np*Bl%r{7)6}
zmS!{@V&H$ZDM7Fw#9D_+6^^1SfGIf-k;F6A2uLuUK2w|weM=C6{ISFnD4m0$oS83Z
zX>%ztT(E@4mhm<~S9ekSjdWViwxai@*+H5nrnjBscgQ1r>6=$w)`mL|_;uxRK>q8i
zt~U<gZ_SQt`zt4ct-2r9LMWt9O`!GxQNoQ*@!3$t$!GzwasMd#Qp&7fK~~ETfu13U
z=X<RHX_}r*31wdgiY~26z>7VXcG%-9mcAjBv2sE<i>n&g^x0nbb5LD{z;ok|sX%a9
zJkINv=?Fe!b~9tbRAR_uZzrd-tpYG5KRm4EB)1rTPrM;8LrJ?No`H`i*LQdfU{we%
z!+Y0p1%xZ<%FybMwO5g<zSIzFoju&O4E7&+nu>b}f(m)|Pna0sk@39tHC@dER{NHz
zH?_5fbV_X~`ke&!^1G?5XJ}MI$W-?zRGO%Q<WD2qX)D5|P1B|~vpc(0$7l(_5g8}`
zjJZ3j8pv2ssoGJ;;^uZ-{5UphDdm)Hd?=ro1Y6-~W<0_wv)?1$05sl8t2&$mL@B;$
zqkuaD2F(GqU!6>!-skP4=g%T-CSDr@_BFaBDrNyr%+ngX3DH*ooREm<iuFVflv!t*
zuX^GxsfaGQW+U>Y><dYpO(T^<T}NI2B>vL7QUH3f#jR^MR7TzvQ{&f&ng1+<xi2|~
z2Gj+D_`By&dvA#dy?NAe9Oyxs1fOw8)|)onWQAw!W|y*L26SY)jZ{XGe@xWU(>7CP
z;Q*;SevlQ}v^UEh$^Jmv80+tWhG0#zmUMfa=6D5vjAMZa_ZiAevd@&oJt{o=)@^DW
zeSQXGoMm^9sGivBL(yb2%#_51&FKxlOteukH1bNVc!0O}=e{Cfm(^OVbfzJ+yKl*d
zu!j@(EdbZCGbcLe|19GFUef>m3R46T2&V#78yQCbRipe>Uca8jO`vf}6=8Z}987v4
zx_^8O-xA+ovft1HXoP=@od2)Q^6yXna|ey0AgWyMA<}xR{nw5H`L9;zQm2c+_<$Hb
z*qxeh1iq0oPr`h*!2ip~L14IzZ4<e8^v@#ezd!gNU*)4B#I>4!Of}sRNK4K4d`h|_
zU@`PgJvf+Q^1pti+%J&P9|ecjVL!m(PMbS*2!bMU;3)b#KKsnC=VgE1utIS2jI6hB
zrT*?S{<<Ik<Bx4i<k2EvzHxa3J}ewq4Jj-51C_}qS>WRYN*f{|!fNwZ3F`a)Usv)v
zgwhd=xWa&b){yracN1>wah>INKG#JEGQc_Ab|*RA_h)gFp2Qb{i#p?m-@_k^$Blx2
zBi8@Vn=ej>3^L<uIbF7f2*iMq!$(f*@d5&$FMqa8|M)!yD33t=wm$FWS5Udej=&{I
zVZe!uiM4+61(tC8pM_~IGx*tMQMm#C9-sX4VEo5J_<vpz{6NJ-0G~vHnD7kp&pSIn
z{kGWyb9eb4UEBXXegDS~iE#jh27mMEipD=K%`fg7ptRy8(V~O=`2Ev&-e;tNClJRW
z1TpS6-V<4!fb?Y%?_Yr%2!B$0k6J_IMF4AadfO+!i+6s5d02li5-!I)a#{bc<@!G^
zK>#s07>+$Ka6F^?&+k8q25AL;ii(xR(d9fALqv3vtA|$HP_i+{ZL`#lkt<Lpa{B(h
zS<$9`a&-?>{;WelvlE<i^pJ^FU(|icZRQB*BbldJDjQ7%Kwu|JcV4>~nY}YFF!XW+
zy$LT@=(P(Vmp%klynfg4n=tIaT65QJlWnU1AFCtgCH$tna2Z}d_|Grs7ycmnCY-|G
za~9O@osCZGHjyi{HKwUY;af#x7C(V0Bd2L^Of`^s#5EMeH=n_+VOHOn#-YOlYruIe
z+zD^P2!M`DJzOdgIh~CHF57PeVbyhZY+ujDSGC)(H4|UUr7VomYcpiKzGwt5BoZ$k
zzw5JVfJ=pg=D6AT2ftLf_#Bsz>(jG@zwsJ4l>2`d?O*>U3tyKcC4xVW|AKts@J^f5
zYfwcPgQf(7(8$-xUhW9)_qPh<IoQA_$8n|b%M5OiKx9K{2Djz85aUI=DpUX{{8Ksi
zLfVIkU)1~h?rP*bghrxe99c5PX*GGn)!ObLMU>$-gA((&-iDZLZQI-NKacu<t_vCn
z`~=G#Iw=<V_m&k8pgDrk#D@R_zE-VHqkvD1UuqUWb=V1p+v_3i%bknbZSh(rS{X{k
zdZO=-v{8R0{))xF594nyO&eQ{{TlxBKaVW#+Y#lre)L=L?R<Aws;sCClkhVn^BrKL
z`#IumGX0~gRF8anVJmCi;FBjz`A0}Q_Jb?izwhg7+J7AZ@qXqn9(75Rx5F3zDEij3
zF!Tw)7tuJJ2&ihmZ*JxlbF}q$%rQTj5t`eonwvwmXAm<9V5{6>ID}-m9@$z(1dw_}
z<E-e+H}YfFj(1;#*o%RL)Siy<+wJ^#_#A=R0EFxs1D{2|_4VzQ&bV?the)J*T|LLq
z?;h6iLlomC{yol#?jxW5x~5LPk6U<|IIa3|D3OK(Yl>I>{-n|G%weW$6aCPOI$s^3
z=Bu{k-%G6w*lAU16d|z&E<NmNPA>5)H7(`0lpH2rBhm3!{ZlblWa0LQwUwH8)wewq
zKhak<l57DNpMYIkNzZ-!Fj70Fq%i9Z7I&rQ2~vAhroacyK{=(%K}EEk!=m>jd<Nf%
zWS0g~{j|?*8eb-s<?U2+HuZ+_^fmUl#W$_{k@HQz%4mOW{j-U~WB~8}8+9E2>dEtT
zlIUpTKJK6LBr~i{)NCLSHjC)_@@Ylrn^hA)kadi|%UV98|GFlZ#uRKW#9lzhLguwe
zbKeFW*A+GR-NVQnl-*%8H^lz0-=)pe{dQVDVRwIX*qCY5I>Ox;z23`zd$Peim{Z=$
zDJD<YuHwWgHuT$J&B39`qCd3<p&dQvhlpvTuCJ4&(;xFpSYG$dU;lio9YDp)xPM{L
zm$5lqRJ4~7BG}@5pHaCZcHz&g1j17BM&Pf}2_)@g0B97;Oj=H^d*$3ur+3ew@_rWo
z_mT!!Fx%J&Z=zeeN>sG-5l$tZn>Oc|op8<=xXjrX+HxE#_AT9brq(iFm>$k-NUn3p
zK85eko{U2UkH1DTn8^3%U)q@zw!nFU0e8me$^zN;)xdryly}B}+jkYi2m;TwZUn6~
zZ_S}+pK#~>%^^i^9NB`rNN^%t|Gf(2iRju!er|8;djGHl1|!C?ywec)Ij937l(`7i
zsD9jT@wcnD1NU@Qw^QI&pnV_^bbfaj(0=Xo4FN60qwO@?o=_Y;xe+fO?&!4&02XZE
z>+xF0B$DndVg$Ug1!9P~=_lgh5S%ek)2)J6^i*T@toMFR->yk7unl<xG#wP7z@dN@
zoGmQIshV2oaB#wN#AkWp+glO>adO`y2{(RlVjh98Siaj;oB#;iV5HX=2+&^O8BISp
z6<)cDBX7F-A@*DFz1y^|F_3uRP+WdP%e?)8AVRPDVNudOf-LAaJZOsY#YtSx5wM0T
zI`F8W*w_Iwl#+Wes$}yP135z>U}nw-Hz~kLrdJ;5J^}70IH0yENFo`1*$A4za4=aG
zhId<VNGA$6MApm%j)@D2IxPw<(eT5dS5BaL1u`2?DqkQ*AGwVY(FFn>u9XZru4CX|
zWiSW>q}E4_92;0V`{22)0{;&+)p6rhmrzniaL6NXY=CDpbE^aVE_lfG-+Xm@7jS>o
z1^2LFt2!r>BZg<9oD;zluUY8fPp}KfPKwMiV?@Di2?zR}5Ril-WuA)C3hz8K0yOU^
z+fw#`63S<Frv2XS4~X4Bcc18aP4&;_y)ae(ljK#nQ;cfd8Qmd#1@R9|iQeo_yp-+8
z5ejMTd`q)As<d=vd{6LoT30mY$O9NV*j`-%D7U1xC*%=^Eife0z67R1V-3D9QzP?s
zMx2ep4N?`*4z_@qN=4X^&LTjozKZ*pZ$;3=K|q69{%rSGKE<7thF2emnOofOI#UZf
z(k>b<*Arc9_yJ6sa;FF7Kxnd!=N%Af;9(1G=T`5QmZ4q?*8Moc)zUFWvHwwmGyzN@
zoD2r)g5A=_T$`KyipT<hz-Lmh_PlAtkKF)V)qL6tc&E-A>AH@|bOFy1d|*W1_cn_M
zeWEk(HxUKMgs=d3#i;)pz|kJ5h+dg}j}Il#V;rRq0ERmK`y6SBEYV2X1A@PJjz1c8
z#kW$&TlJMxdHREPv_2j`!1ls?pviAARbX^P7+_qFpU+prZy)?cso_-+AV=~&wKm;+
z;0&$y)=3y_bud0vBODv9r{TB!_!&UdyE^HIz}2G15qzGdpi2vt_**aZYeAxuq+>o)
zYX09+Z`&~M+JO*Q1fJRZ+h^6I5c--tRvDddnr$;OU;_d{gsFY{&K8XM(tBNv-x`G@
z>(gmNBT4tJ%jrnKy~YnixKWM$8~!=7<Z9q|Gesz#bID`n>;|+Nwxa6GoM(nthCpqm
zyr*ma$6>vaCq<P~<)hz4O6^7aM9jeQ(7*fs>MIRsDyVZE{!=`>4xy%${vr+Yi)5Xo
zm+V&DxNL-bX(^`}`Cm!PlSHy^&`a{YW(lB<Zi!@#qE{6;mZqua>Uh5h<U5fG@px3V
zbdscs09Q$uhDk+2pW#uV?!K9?vyN2S=V|+#1@|%P_gT4l^~)$CWEaODeT*|v!j-xJ
zEw&GY!n}u3m-|gS-+^^p=#(CrARQ6d51N4w`z?oU{a}z^a!l_Bv>D#i1(9iQYM^<h
z)Gh7!x~|G0w8h7XGIP=p(?}>Guf3$}a}er978?(3OM;s>^;f~I<|def!=wokc(J-A
z<mLPISFWc+8vTx=jg-v3&7G0qNNSB&uUxcUSkFpUPFSwt{|?DbfJsXYUets_8A+@o
zCupSWXLWx*J8s#|i%tTHF#P~nAGYv*0>qkK7o7XNcarbr;W{S*u@LD*s2o)+y=27m
z5ZX~kj_4g`!atajH6xY2F~C%pw%~LiYFgEv1GYC_&#C7fGC5<8;W+0!s_GTxN+N}>
zw0e_dTcZ6*if^7T1U<JxOw(u}D6S~@gci9+S{}EbK;Q0Tt%6{oZxTZdE}j;3Hz6BN
zD`7=q6w;<Euo6*$`yZKsBVOb85z<fyXI2J-<{;s5=`LC=bwRGIAx*cwa+j9AeX1IH
ze?A-@g`>zto+5z1{**NJFtoeuR}R!~>NfyB+WJ*9)D%NDiikAAadT!TGbn1>We)dP
zDJJbD8<fnP)drhb@98DCruhCaJoHW<*dLRl$Y#>$ZYfoIr0kP+BeYR!CuHRj5|ak0
zku#5K2)<SzU?frRc{6rC&)70)DVeJp!a3#Gr99Qp*#CK3LS6^m-sMY?UAAt$Zv;WQ
z?1Ge^>{2}D3)NvT2=Wd9(k;Ggq1TFjie4&THZ^O?=dq`CWm~>~nI${OuzF(+BPF6n
znzTNvGE0;i6ycH&K=+xeP6QE(`^xM0EJjyR*P(V3IC*c-w(1H+CeUa_u2DrWjN6YZ
zmQqq#Za#<<{+3$^_Z<cL6MU2GA~`+_+7z+M$#vZ#JHhxnyiV`@uX&4uy=Vu7;~ugW
zJ0xST25AFsbtn8X?!}YQ?%|bfJ~_gibc*hbJEAOK7acr0xU<n0unc}~EUsKY0|87g
zcOxL#IO87N7M^8+ysbL%hM?2v6eEB+^ub-h*=0kD-2GUw>|0Sm7e3G`%og`Z@FngF
zgj_XV_RWSE*;%@{*DYAIvhmS|Z4D45xM5s_VutIJScJvsMwA=NbSjMBzg(4b_@}#J
zfg$;Y9(`;44EEjia_Y`Z3Yk<-O=Va7tXR7T@~-p)i@JZ$1;OzZ2&pa}>}2a^NA<9l
zB&U5gMp}eXjTuuLVUyRxj$09+NYDxw397x$MWhPppN7wDHlGn(2L5Jsjp{89K^dDB
z7d6GzscY+=j`dmVG2wP-1LXiiUPs|^90|ck8Kl4>{Mwhxd3Ao>$e720#fX9r#~&5Q
za_1c=_Y+trJvfwci7Dz^oc{%U1;dl-j`%;6*Ic-VJ_yQ+=4hKOXEu@{(fb&908A4<
z;S;k)NRh{2;k$}mcHq2Zr_3&>;rp`jc;_{OJ`X`={VUx@{tOs37N20{CFT4f-q`ZX
z0YaL2@rNM&O-gS=mWPQs1Hb99vO`lr(ThMiiomC}Y?&!s3RMYG2@^c4Bdy}E`Z4^W
z?9v!nUu_JSvRIcTw&d|QcNvY`E)kZYQk0?KTz?$f=ptoEpE;iL4QVc|u4rQHOd=F*
z7x5M-R7^M`pDzc=Pu)gzm3SK!`a=0a#H!z_$uyhLWU8kQp*GP+mm7vFb+4v=tzy=8
zE0BsKnK;KeQU5?i3b~jZ)Gdl@5Exvc4rv$K9V*>ihr8OH^HaVUpMGsMu^oIA+^8ZP
zhN~RF+`e<(z|Oj7g;tiPu51-5Cf}OkG*1cf-r=LbBikt;Gdmux{*Hl{L_%v%vpi;b
z15AiYhmW&3RfKj5ibnPlSavD-+{72t^YuX$B~ka(Rwx0Nuf0H-Y?D-sN;rtY@5OBl
zmY-so2HXzg2Fc0O9}XMj6C&_kq>@ntK5<LMzrU99FJIXRX`jA*jU99qRUV66QMpuv
zk9bx=Y}w+3TuS+EHml9i8)W#7B%ZUaYxb%PDffwODHHkhxyL;J#$tQBW`8{TLqrjZ
zntEhug<Wd!*pTq4h>)QV6aI4FUWkAFh50qFW}jMjBeqy+;bOZx@(CPn66L$^{kzfK
zwne5KuW1wn7|^2UKw`^om_pdu8ZAXl8EIYusYLo78b%5$vpBt^?W=h()rd88uG77)
z7tOE1F)>b);D#PcWnJ28=Kt*l@JhIAO>`#|HZx4L8js>2=v`?Z*s}EHSeT&gAp9$7
zV%c%Z`kslU@<o^OWiVzxEq-_PS*Jg#y1eNh!4_4lrpRyMaK)Y*m&=w?5B1mToSnuX
z^T)Nw^S@(;(g{)IYn|=tZFuaoQBT?qd&$2$!477C-#WpIO1}}6*LAuU#96ej`I#Gk
z5DUiXE7)L)nrg^hfx~+flEi48+Xmxa0ZjqPFU@xR`++UFSM!6j+DsH7eP)zq6Rdbl
zhntNmY(Jc1nB$d;N|kuLv`#)^H<hxGSv{AswR-gxqjwSs&NjEO*|UoNEW$2%S(Xn<
zGozZfh$Wb@Ej1F!6lVR6@d#82ZBpJ;jeo7KF`{6u3)#!(UodKn&alS32dSrkohMPd
zO%ubL1XKCWGl`a`nwwG1Dilf0rAuKL;~l6t5rNhk+JbT|56l2*9a>cvcsnRd!Z&D^
zQF<51Dz-EX9-b@(qUJ)Hgdrw*WYX(9deHrMFP;hr11vM})sROV&m-ch%DQnywlNDA
zmQQ)zTOY&8$B#?F{(2d9*p|Lgx%ynw1r|Kt=hA89B6_1Ch+G=?5mJvWHqf)&?u;x%
z13LF%q9%;LJNH};+&^Xw=^qP`$qi6@5v=0a)3O?<fKgDqg*3+VWt=rae|W+hn+{kb
z%I02SoF_aMAF0B6Okp8jt69&wrQ+D_En|+*B6mT9WB;RWlMJKRH(0ja@`IOb{8yP&
zktcQ$6hyvY>}!2B%<BCVLCfs*U}9zWF#sq-Wtc9aifH572dRT3U11Z_q2a8q$vL2#
z<9E5MT9-);e1wcDU#uH_;<PBBh=+*OxBsh?@;PZ?OiaGvG3bKxZAhet6Kv(f8~euv
z6nyQdB1|6(6Vp>rZS`7L=KE^?>Y^8-de*@GI*?x4tZ@ZI(4OWSUPrf;y%2>&_kXLO
z)??w<+5OroR0R|)sCs|MkU48=3kisR@ZJEhl(hoFChox(QuVCDj}4U#jdZVT82?@x
z>fse_*Sl;A62c~#x!9F3OQp*){6RodDmAY|i3lQq-ct^2;;MfazIK9f*uNA;TE)o-
z!d$F?Phc|=aCc7Kxk%A-G*S1Me}Foy^xSV&$@%8s3kyOjm$HwF(fj+CEe&Nh0l6C2
zcksp=lO3_nYmz&LjHqkOsNvY@`bj)eEi{rac*p1TnOnw`rLJigYd|%P%4Fd2#jIPD
z)Q5nj**+a;fKsAs_e%1)tHH#ff~wuxMTqSwn)R)x<*cJ}cyP}28?H_xA+m0q76;7Z
zh|}@4iSHNZY%8bH=HHedA6I2vS~S(1>5Ggz^<UEu4zJl=_Bs79<bP*&wJ<z9z~IF9
z8=Zxnru$PCPmh1$=EyB~d+kP)XRou3XVIPFCx-hl9%w*U0?P=|a?iP!6X$QX;f>%;
z6!+}5&xd=Bdh@Ky-892x?#9p=5llAw%jm3E1?a78zPGBMLaU?yApTLdDfg*SpY)xu
z>8%@J0Pw7tIjt*-pe{KMZG$0*UQ&k4rK$frFOA*nx{@+Bda;h>f%Hu#FTRv8PaUu7
zM#?OLconY9G}~KB!*f4=YlXoEzc-t;z@bX(iBQkitj5Pq$&b;-O^9^8Yj4Y+SF)B<
zyA@60hxrKkANwpO7%gQ|?@`y}xWT0NjS_GCch(<Ay6p318+n<JnT3jMrfa@}q50$t
z`}oY1{8S7#HSTd5ZjNtf5e_Tuig#VC9q=M>KlX6?0A+|yPo2oS3!Ht^PJxMRK|?&Q
z{!dhx1`|P$nKFX@3uOz+La#6IqreaYO>MAiUa*yjZDY2mpZBa)erzY0y%Lx?=Nad?
z(U9#>2B%WeraJ<|CO*L?5HB2T^%7l@3j@e9jQ=8$b{dD15fq%-XvLP1ViB6HF-C+H
zgaLOTRxUKbAod9NR7$ye3nBap7YTx}-~`Q1IEJFcubVdRgzb!}pAcx)2x5rOd~_dZ
ze{!~SP6$nfL${P6?MH<ieoTYWZ~zinkeLpKOK1%8)iNBS^d1I2g(dc7T8uGIpSuV~
z@xW?)C3q+{*G$P_PX#6F?vZt0L!2i=boVsn35@}~DiMe5P#aGXBAx5SOj|kIJbeR~
zQ%4lNr_4Nfcop3mm6lD}cZfoj<$imPL~$uDjxCd{EBD2m^lgn5<BO}@JvKew`XMTf
zT48~=7Yobh4Miu`tGDgHFMf9d8ml#10Lh!y8ePj*n^tSppan;!Dnq3rEXfiY(AE@V
z!~s69GL3+g+6BxqGl)Gtrx=b&Ktv?u@uQOGX57ARrs)SMjiL&U^G8f~D7{&j($+ks
z;r^DCW<D5>!a&PYksR==%qT;1LX+MA(alIJTU(Pp6Os8!E!<!#1Di?#P4J;$M8l}~
zomCiLQmu-i>rn?<mfO0Xw*KM~XESb)7O<-(3$p5swbOsISIS+M7C++m)8$VX3Alck
zYS8YOs=eA7eD{&PP*uuLf+Vu!uhV66T<=`*CJ6VtC)rmxS`w5%8f^7+*o`GKvkO=&
z5r(`>-=p_e4>uS)#KW^nh-H|M>%RB!F|rgC`{OMV;vH9p($uwjovKZ{Dqn$Bj$T|E
zx}@4rtX}H9c;%=2PVfRkB}cd>sPm}vtL>+3J?g}eAgJJ)_zKJuQMS-eVt)OWzOhc2
zco1YoHC@BuHgkErj^yV=Ws63|wBdzo7RUsl5`OPIz$8%@gC@A8ev`7^8;g<m>>1~8
z%Dx)ZF~*!dxj90(6U?w2=a-H~AJCcOz?l>02q7^QL{A&&u!r0VCa6ec>&O|fh<a6G
zP`*PHUftLYVZGkA<?Q9c59i1<54KZ;)H*5kmDeGlk-4hr0C!_I115$?+d5gIYrxB#
z#EQkl^p5^I3Mi#-IsuqPVTP1*9t;TpK+eoPj`c@X2N(Pribd|ePLB<=15CdX6hZ?|
zn5=^R7h<V6fZJW%0Dz9CdmVwQVQ&%hp6j>d%7cK=Gl~R`YmBcj{VgX51K$uXZjJ+1
z<Jh6`En=m!qEv3Y6#1YmW;P9Fd}!T7MY>^=X<W=x0ogibwClPUznv`jxI;)3S{oUM
z4$lX4OHNa6Gbay!igbsGDL^9(qgux_ixMo;e&t06+v;@$9wMmM)YeiN(hkbQUaFpF
ztF@h*e%ZCL0bRPP&^F?=K+?BpE(nX<RSiJ=90qqyFdi>ZXxjY$IC~4AD!aDrUj+$i
zP^6m;N_VrR1f@hoy1N@eVACZnxj{leP(X6iNOzZ{h=g=^^R4ZD-_P@W^F9B0-}yVt
zC^HCqb6wY3=XtK<_#N3rmnGHlH7}KC>M4w=O#F{;iA)EdfI*Gc&WLP6e@4yk4<)fI
z%dUHIwJ{!QJ!pR>a#|Ut-RQr=N=%4c45(tcNheYh*%y%6FFK?=Z|KjSUrB1}yqR@P
zeJ?>dK;iI1?*;WR1R2N0_8SpZS#32t7w>xy57F88aTNM70-QbRN+>}w%JzhiXlP+-
z3Cd@lX=?gvkFtqvOET>cZt6;CLvIs>P^Vd2WXyu_ADYCadXHBhOL@ysM{`3``w&|O
zO6e{o+fq>fvZF=n8;#<=0~0v1bB1Y%0q2g-P)0-~eo#fRk=x;LSG1i%tSi9**u|5f
zro~qyoh;-xpczx2Xxj8#URsi~@wCyQ_PN}n$vfgbC+`dD3GVGkLB-uzHBVfdz;~lG
z{w#b+!=C)uRWa?#PrzvL1|`Iom*GDrK&W&yL{pfZ;jY{kGO@S{z1xovM}~t078HIH
z{mN(#Y|ZfCMmMeBfuV^7pqBQw$XtVX{U}K#>{s>Z;JvJHj=N5zl##?8r8gMX3xLqV
zrR9fgn6WT0Wk<T{sumK0{LTb^kX;K_Dnqzgg>8!^-A4|)a^%8Ac*(o{Wx~B}{5sNL
z-*=f%BoxUzpS&>!fj(Q(#AdaQTGNr(>#tY{MTmy^85kO*iCbPBn-b7TNRil$bhzE^
zF_Ab|>*e7NPlUh=?duDWH4(~J`7yruXr*BKI>#9$`Ik)4>2Hl^%c<XemY{=(g5rEo
zKPQ`R7Fgkhzqv6Bispa3PggKd@R&$*{y{~7RMENudgsy0Ya_+#_%&U(DVgZ<YoJh!
zfw$F8FR%{QeFvG0$VtU3@DHX$ptV=)wRoLQTq@P7pI1q76Qhba>84ZA-f@(=`2O{a
zRM8NXq8LNC>@3>MLeYRz)CxjbvS&y<hIN{$%n=fc0|{eof|L#AY1wPh5opG5;YsVA
zegzd#1UKqutS`ML!vY??+CQ_daDdA`LVJev+^6hT(V}jLGva|H^XT=x>Cm;_A?iW0
zq+cAzhk2xpoWbTcO8Ox}=!n{$T+u7VRQ|6qX6bH@QFEBI3Vg$=JNKS35*Q^|`J1wK
zI=*X@3Jno1dYC06FB~YBo+*^f#j_><R-tKcL(=P&&9m$n{e@pdygDB-Gj^M!FK-D<
z{m3Dkt$~6lh$E3@Yrs!dZ4cNTA)Ea==K{hhRtaKHnHU<S`LS1H4ZW04mxjRuG;<%%
z6W{4uZ61!~Xq^qWJjhO+GwEfkQW!F8JF!qsRE#@**88CUH$?ulwI0KfQQzaK^Mft&
z>ixdt@`q0`WFy`2savLfZmzKI293i^NQCY$Sks}(^ph{(HY|C}F2ZOZFpIarx&2u2
z?Ms-H`KL^X*z6_s6aquP983Oq8L6|@#-*1i<Jx|^sx2uFlFUMbbIQR`AuY$7LPcAG
z=z0LVM_M?Kx5wQW_$vXoY)>c82JelSVVF_+PF6Vk^FYo5=($Z9>2tnMPWMo_{X~}%
zifi{<7gU$e0ZCJ>m!`~DM}=cgYLvoM4dYu1?$oOkj6FI53|fx-O`0UhzR_BLZl!N*
zORSW|Jsq72Y~GS~s_^?SQ+smoA*nt2->JtUIT}ul>?qLAWD>zMw)S<JctYi*DqO#6
zCvYs8bLaVNn+Ad6s-%K^pj?3~zG;LT!I|l0Jil5m^R)`P>cprSKJf;FsH*zN>)vHw
zX$%qJo5z8lT<!;d(>Ep{EQ3j=DUsflr)^_wOM(%5<IuubqBNaPmu1yz^50Vz!e2am
zF~{AR&Y#o1esAex*IP*y93M4PI?5C7LF76^^rj&e=WZA?<OQ65ZdE=@g54H*&K)>f
zZIRTaKAnV!7#|*F9WACUa*g=74yF{Zw#dJKr>0@ILnFEk9eKlf@Ll!yU&pbW_wb)Q
z;3HzkHrdAxuphR`O0z?=D~+Wbv>G(=dZ3!H+sPmnH@tS55UEPql3~IBf;BlJVRi&A
zQAOW4i4NNCa?EWz$dq7hboXvy_l<BUZt$}NYt1rLr@_F`&73U~QK!zF7vpWpozZm;
zPjpUANn4&CYhd8_KkHUi7W=-?2f;MybnEq3+azfO9!wV9rgvK#%=&6q28NM-(10bS
zU`+QP$0GB^HZYgxKCvgD5(-iY+69FVE3R6RYnNKu+AoZEZyGOVpO=&!EAZ~FE?{CM
zd*l{@xy_9oCP<uxt8EnIg8@!yVK!x<*3<HRL?|Wd?cvq98hnZi6c7KmNsBep@GS5Z
zOFiewS&BKjW77=<*&ld>dRw&8{Vag27^e1f?+A=tJPe5hV8{j9z=h^Th3&^1*)-^!
z*FgmZ53^P{g1YsNH<e@HTeu^Z2?7vubMLwg4-^GTu1gK$j_fzEV@+F*G$(V+K7C^2
zHzr51A1bOQaC_Li*(dZUFr)^Y5sd*ivQ?p7Ft&t_jEJl1{?|EPfFE05fRS-Dm@ne&
zOy$7S^D-MT9kJh|1bPkYOe{gYM;D`^dJpGl8~1*64~_HWn&4szHR;dDh27u>N)zcb
z*;{$c*+R$f(8COxV<$BCdo}LAd_Efp#mZpu=Gaeh_f->CPWF^=qqXbm#>OEHyDITa
zo$&XSuaCXGG>*NG62_^>09cU~7@AqQQ`<>z{kbuPaOlWP{3c&CBooEO(Z*xM4-`zY
zoq)Xg=^drPaq6db(gEssxS084$YO;|SbB#-`1bc`LnuaGJ}h|_QhEDEMoPCx=h&!$
zhoymjxCk3uTJ0a<wT;-z@wEJ;G@dr%=G*lA^$+{?aOh$_d?suhzsNVe&$@K~ZX^l=
z6l0_`cGL(V7QhLF<$OCc97hjO^-TQTW|rev*~SjWedOc5^x>#Gt5!y<^Fdj0H3J(3
z)0bqx!8vHDC??8IKn@B^3bD2T;n$y$%8w=xkJc*W&Uq2E5scGtps$oU=OfO@H)g*4
z0<7Gwir=q&4d*ACrGacL0A8H__^$zmpNj7-X))nZ?2>W}nMp3oKHpgVw@qC<A`X*B
z<8}n2nF>Dk$KzA11@qC(tZ@q-WCN@eMcAzS3x?Y^bOVsw{)X8E%23vKc}&BL_9O-Y
z#)*|I&gMi03f13D-(edqF;gXAy3Gctr5hSNZOHyc(>o^N(zRdpnG)p@WV4q<NceSF
z&KhRMBnRhv<Ik_I%ODv>Ri-=2g7bmWQMwivLMaTfY(nf`+S3#XOOo%x8Rf4_>z)$-
zn8bYCsVH!!@@Ix+8jDHk_R+k3j;^omR^X$O)d%x<Y$~T5vaFgVit6i%U1T5=5$wu|
z8KztL@pvsJ+$FQpIwJ3#)!@vj%cpgqD%y6*d_bGXfsh$dT4CdlqXI-VTgQmV*@v3U
zUy-9M@q9o4Z!3#(E3qQ2@M86>kTJ)|Tf@}6%)v29;&;96s#i$K&(fC_2)+FLSiOzB
zAKRkWasq}=yhF>AbJ6A+@679x#n&xI8JJ``)YA&_XH~6$#vI38#u=x(+`a$@%w2tp
zhXU7|U@4=g_Ir$_n3FC{y^o(7RTe=ppYIK?FC`#58D<Eo{2+?rh-OP|V`U{k+<VbL
z?n$r671hJdx<AU)OP}{$0-*I2>?bATsF&#0&3O=%ZH!+@2hSB_8239Mr_CDguNhLr
zjOaVL3mPpBt<7mDxaaOT7RJ#kRtK@ZSo!cY&!PP1!?n(Rv1!sU5#t&c{iaJ<6oP3y
zPt(GkyJw7pL;4a0WQ&W)1k^#z0E1w!OuhmlTDA##4;2%ouxod<sqK)r{^GRUfo555
zlPVAWx$?XaJp6+ivM>Y|(?(i<O%K<AqO1I7BJ8orCTr^lE>{eb22m+^Ed5l)T$+00
zjF)*z{fCtmi!u(WQ<fRxldNryIQ8)XZY68;C6ea}jfcx+!ad9fHHczhuJN|JpWG|E
zL9$(j;dAvupTL8ch#MkHWIgV64XjsrsHXg{m*9`R>xK6N++9=5+Y~v#I@*QmSZH-#
z8m-Q&=p<T|nCkuw&tK3u_AvJ@Kh4rtuJ-xy?}Mx(Kx1AI^>H`v9kuHiA6PMigY?mf
zfS3r!tl@3`KjNp`cNFeRP8)TuBeS=35vGRm)L7Nnxm#WMXqB2+pA@vX_MAbd`e=>A
z`bBpe9?>`wGOi8jaTv@bx)E~X-SFI;tZ8rJ+YlrAMWKQ^9CEj}uv@bsApR8TF^ZWk
z!?rus%%k2<2X>?lkxjvlP-E66oGP<wCUMrovbaQ-%9*_$qZ-9r^Z7{qz~(2ygBm(p
zJuSE9CGEIG>T<7Wc1S_rZ;T()@qVXi56t9-pY<b;hqr=)oA2~IXZbo7D6@(0Ly~zQ
zcJhszf<vwWf5k;mG)}KXAusN<Bp{<(!_6c2AmL;QPh_*RTS?i3<7a8dMJ;yRHN`_P
z{b`}u&iml>u19gaB}I|&#lT*Oao#Qq`$m$EQy3w(;I$u6gM)zqB`nk1zpNQoe}Rg8
z|2^+V%CK_`4<`E&4=Nwk<8+w^3leFh7lUF>Pt!2Y1d1E-Q>aZB(~fi#_9foh(y@0m
zw=_q+g%%GKNI_T&A5Y{vsl5^YEc@AKm>A3OeVEU5lv1vhcErt+=6NUG!xYVtW6ixU
z;HiBn9-2*XT^)@hyYrEev2um#EHIFIW+KRxa?2orn=G$!Q)~;*l@z+<+m5>~v~Ex|
zR<58i^4jbO9pj<RR_79H3i_r{!*Y8mZcSG7({DIm=x1RX6wmOkULfXj_!z#c3@BD=
zmh`MU``!)bL9}`h&4Sy+GGaJO;s);YgetqV9cf?17QdrLobn{(dJgE$6R_?ZD_QH0
zMp?*yw$D|7XKi7)+W#En=S33HEx`_#n-;8*Wd1oR%sL$nE{x$rVV}r<s5#O`yKpBL
z$b;RMI1i}la9v?G7BjTV68U|-=;a~4@N4no-fhc`HK^nzg`H~ruIyau9(QtPLNH9o
z`ECSHv)?hfK$l}NwyhXLx*IA_i&;@{1I}I_&!J^(>Q-Qm99cd|D!uiaoZ>V4D@V^t
z>#DL@j-xh!+|E30m%b}6fBx>N60fCM|2%v~_@MQ2^VkaO^043l`-t0BFVWB7o@Pw#
z=d^hZ^QHH3Qe_cD*JUp}TK}~jLZPN-_LG9qePTf3cNVpTb2I?#1(Qy-3Y4btitS@(
zX-YMQMG5D;il-Y%=l5<QnX&Ce{wF|iR&?Mv9?XT73tLBU)jp>#j=H3w<R9$vPG4gL
zPF#hh_L|S%`l%09sf#VhRt1Lnsa%F6B9*7c6g0DnjkUxqwO7j>jg*;G)uH8bjU5ZG
zT(N_ordOW}ioQ%!bFy9vCj$$a{ePB_&+*-XzUWYn5N(QrDZq`x&D}W=$Bm-X2}}yD
z%tfvD-zlASklPp9!FfYon4>%|!cI|!Ub&<RwnIaonVlx4iSdUTPB5F$Z}umXN4++c
zAjL6NsLbt0nyX~{OX#l=qe`Qgw;_CIT#tHj`O$Fk9R)iCZIk;xjrQWXUSo@s(4?m6
zY997;12<JeM|ATI$ezn0iGfA+Btuf}KLlXzS+?*-Y?u8Jzd41R?EYJkK?*i8XqS?n
zQy|+Ywh-7`YI_A)sH!{d!|gv%MbU(9HH)(-pQU3)SAE;E752m48(#25yMvaxbXR0%
zbgo>Z=>-3k1_DyAA#&TX1G*U-0t_?qOm7xM2cN5JiU5exEwnG_n$f0lX=Yu*`@ZA;
ztSEjF?nPBU-(t!VPoLyqM&4EE@%s(}DN~L-*xc}$;Jlwc&~_4|?+4zdZsS2b{7BG+
z*6Y~xx&fIq_1&#2Gv$!8TeimNV;Jvy;vW0NFIo+|<@wa{VkYmn`;?xT^e+)={sa@-
z`1-y9m!2uUtkM`!WzYkkznkPTY(_;fCEKTV%#8ZjFbPDL_vLq&lGFw^x}xd!Y*#96
z&J1q9<Br&L$a!#Lxy7Ok6gp18(P5>>8H@GWGD}enZjn@JH}{=Z-8B!`y^e?8Vr|jI
zU%AN^X$XIj>aA>FM*kgDAK`;QQE=~Y(>$|<z52@2n~2vVgr$HG>*Z+lYB74LkZ<Hi
z;I$x*maL*agxf|m2Q&lal&soM*`WD)nl8(N+H%(l`5Yh&avD7|Lz?;)8_lHDLNE6c
z*4DdjJlXW&hvnvV%SW<(gZWM}>8uT>3)3)%55nH6*79ot!<Fx;&iN6aG|Bi15}msO
zyU0mONTe<lFATRxw1PB#(BQIG1J~}AA0oqMlUC_C8yZ)XH#&D(j_m2ZP<z;UHG4$Z
z+D15X^}0u_@En8RAo;iN4{6bUUkQr8cceIS<G^-)*yn@b*&>d=2-|%F;_a&BgBy%5
zr5gb#FAm0^vh-WRW>6ICOyatdCnAo-{5asz^HchPh&h^{JZ6v^Qu=b*_lmwT8>>=d
zVhs_cSv&2;<Wd>_IBXsr7-lgJ1u1k`i(QHjxUCz39o0M5Pl+3^(q=W&&Uk=%5U4#o
z440-xS0J#huJ9Pn<W@QI^L!hsEjMEfG6GVl2`SU1>Bs$GUD6i)sZrv$Dkm#>tT#B$
zGZq3IfGL{&0V9AeP$`%XrRPP;?gYnCLOZo$17c%LH)6|TFd5yD(x@YX6u$B3m~G<E
zzMq{c_$ycUCgQ|WH3tU_Vb)4|#b<tKK#%r%a_+WN$Hu{}jT@pWgT=qtYL!&d>hbhp
z6%@hD2;e2vuPCMFl{KO3`PTV|=-B9uxa>$N<&XNK_`e0-5rn8TLyBp_YUnizWM#hL
zkKG|_nw-$%Nn=-oRDsfNRQ&Kb{?8OlWP_6u11+2ZoDg}<CyWv9x6^}8kYb<NUs2R;
zjJ(%yzb2FG6mmIyO|&}?>y#Wjn-YUmAlyT{!IaXpp-QDU%XmbL%RY51)h489Y%bH!
zSi&n}uxOaLLYt$WDUXm}q$S}-Ez%Bn$-_!${Pm4HqtqHnemK<YB$nQGqfq)OwrAug
z)N5d`@t)}y1_43iDK4GGcpFn3_1Z3tM0@aE-$@)ThGE}HtbqEDkHFEy9|9255GIj`
ziGMQ_LqDt3)Dvq0ZaEmu4E_h2=3TKtk4{a~Gxd(BS@_aju%Ad@;1j(BDDAQybhj>T
zs<j~qX0jLB3Sh1)FgXg|iBEI0ur}Dw8!eiC{JVEV_d6a&9{yW79#`~mT1Zm-+1h9x
z6TXSNy&X;MxNWk7gj-W|y&+fFZYatTguyags`kS~0<~+-7n`#u0e+tbi&%S7K7RF;
zv)<WEapiSYgO_j?=xoFk2#?+tB0Rez(8IJPFOX}a9WjSr`h|$KSXdXRXOl@k<-3|q
z?fWgioX!>eX~E7{4%C}#BI`~YD82>WQq+u)L-ppe0k;wx(h3*#Z)89&I)uRujEc7Q
ztI?TLiN@+(<)WlP?zSYI9&rtwIH>cYs%k~MZ@@`RKB<36C^TID(Drj@7@pghZ4BOl
zLW0x49W29=Uy>B7vO`g$>7zxXod#S9k@y~0Tj23^sZG{kV#gO;rFYG)d9)RRv@thz
zNo8(I#(hHc7;#RYYjEcM43Pfb&k-(Y;rPXglA%SvZqW;eo}boV>`H-W>;i!@yM$r~
zQAp0rLz_I&oTw!>D!*j6QV-shB{{SndeLRX>HH8#K=WK#abF)MO(nx8P9BkUB0SIa
z4wZ#TThOf$5-V_FOfwncjsn8#^H1pp*(}l%cG$)QJ?eFlTA&XWaC|rWO7mBqy~6Z9
ze|N{c!{Cp{&Ssh&a+I(fdlq4-XQvx7vvVm~1N;_KTg$uwq8Z>3?b9CSSU36X3?~P&
zmGe&x?-mEmp2icFh&Oczjo%LySrl7#Oy20?Q9E~cT|(bmzmNWd6UO8X&I@fd0EmEJ
zDecG9pC?qfU1&orRR$hnY)qT#VyR3M!vP#I-QdLO%hjXh$x{0_KM-fSU~>;u(3k{j
zj#Jg~=-Ty^$5*#on~MX&n`{^q?`8j|9_bdUNi0_vdn!%1PvjVe3I_%)b61OVoOf%p
zk3ALabe^X3H`-bzx~2w><coV{nPet!0Ptv@1}jZ5V<-q|mI$zrx1Q|HWb5+J?Pz8-
z_2a|P9H0uoxAA8k5J+DnUtmTpq`~SmT<(LD>C;MCQkd-Xh6X1AXpvsm3kk#!Z~B3q
zoN_v~moqp>L8qm>3%`#gIgTYMjLkjuz{Iy75=KL&cld6`Z4G~44IhWo&pnL)E6q|j
z4HmrwnzRIQNI*>5ZdQ9&lh_}z5T_P}G%>f~R}I<Mid6=WQ2PEzfrs6)&zNL?pu%Y|
zxVc7`#;E}7XphE3!gGw}j3cl;IL%^IMW~cLCnm=q6ms>&7r_pJpYf$zb@6dtOQ5B^
zV|0Hnkn<%jHi+OV9T7I(K~tBAvd@rG>~xS7moeMGBuaZ5wI>+cUkwtN`T5LG2g8si
z^mQFgoB&*p&bdtP;~Xv2jCmuuey2~@i%q*>{V?D%KFt=%Od76z3Nc0~ykJLKRA_QK
zXvhS3<whWZ@s&rfDXQpBBulU**|C@Ix>Eio6<Veep?&b>&6<;*IqZIzefkXdZ+fMS
zGnO`9wui?G@Yrq&&oN<Lj0v`_Te#~n1y7$a+JnD@M<r(Dbf3*Wqtt4<ZyVlkzwR+z
zWET;WSE1YEtX+w03Sb~>1D%%LDt@)I%P4OAL?hI#l7>tZLG?+AQ_4pILKP6x6kzj;
zbEBgIJJGc&z_iDL{`Ld=)z&Wm7cuk^DPDwk^fK{=yN8%Q>SJ_lVZ*xHV4bzKg*xqq
zqsl-aIfKv#kr#ob$0{<dfw_Ecsjm8;KB#tlX0gl;WX-%%nwliE=w#l=!uj-tN-;zq
zIR}U~KleCf3p^!}-muTK0SYHTNO}5%-`J8mrPPU(KKwwdShdldbT~7Ak)C$_u8>$}
zH>R<6$Fih27P1K*t@{YWfBAkuvuio?N2X<lN%^Z798oqql5oyWTImd$JbFZZCjcWw
z4egXqORRgXyPh+b8aCX%kq-B4O=-Awz~hu-kKccU<GCS#8Ow~ZSAaYv*YfeV*u*Jo
z{x}RJ+^U?0+{4T>Cjao;JTc9N3dbq1FgW1|#=K>*C{ykr2(lvT?p4m7YIIQv57I4#
zHHjT4N}_G_%+W=ydUhMxxL=>G1lZi^zlwKEw6MFE^;H1FgKu>98B^MQX=itQ93P1;
zD!5&3g>*-ZSO-&|A_CwB>h#w!fvzqBZFuuR;OZ26x0c?cfWF0UOG>!iMqhN_t!a&S
z%g@tNO{rv}p7y)I)%8`_O|qZIL_(3Qp{~nu)<@)$I#^|E{GrQ`ayuQktbw>1hL3RZ
zq`S&|@Hs*2sXYUE2TETBi1P{4Wy}V;axl9qkjtYO#^sqC)bWB}7Gy{G)0=*1cv2_G
z*`Y7Jd>*(v!7y|6QeOp^B#+zvgK0987&xn_xM@F+iuqmkA?UG_nk)?ei;`oLV%;_A
z&S<V2`D-=Eh+&fVcBi-=9KpHrv`(rqj6W2{cR+_g$@12kuw2dP341M}RJmULROPR=
ztoF~~p^)13JD=J)N^pQaI>~KIg&+Sm{5AQ~HL>%%Wn>ccBU{V}QB)iCOl(|Py(M(c
zj&y32<S*$t)A+mlD&+AM?)R-v3)-bl+Wr!@0XwOI&7h;HsUlN9%U0CkgN0BlxwuGS
z%IE}oSoKcAm;8BU44?eSC9UCs3{#VNF^N$DGM#b-H8zV=%PYt(L%nG0-jT)2d6J4r
z2P<EEirNv4&iWT%7MHY~Aee5t6K-`%#I5@NQ>-ZN4Uk`tc|p&X;*2+@x7R$oxsA$z
z1EYZd1EO;YWg5;L(v^shGO9#2SeE=17KxzuzW9`m|FYalf;h`oJ2J|GEeqSp#Y{xi
z{Lk=6S1nT9!f*jK=~QmlMX)Q^Mr%}9h9B#b{zkdwDN;erJ@>GbL3Bzq-60Dl?@J6G
z{S@}6;6}Z%E7Ot{6YOblZu<7)o-Y4=-n*)@q`whFi1w*Rw-j$*p4`?X;!YjUv&8WM
z*VZv5+y>QNxPpd-&7x+-B|ud5Q?xp`svNXGEYngR$nN`#<80{{^IobYgcZ6cm2M!f
ztcwE$|L+f#@v2Fuh9*3Dm8Fu$qjAoH-e1U(&1Ye7<6~py5p_3;*Xp`p@;9VNRg)SU
zev8cHKQHPnx+ms@t}~xubD;Y{;|xk?kTuvPD0y~=>GDWc&l7eiEAGyEPIQ}t`YM3~
z7*UHacPIbc6ED#3P~3i;(i-p3*<xwq9?aXE^_JK#;-r$~!!@x_Nr3Nd!)J*5rteK?
zzB$C70^vj=Df%U?&E?sevQJ8ArDvfPUmFoP5>f4qp!b_cBWwbB8tH4nLC=XfWr*}g
zJsSd&p9ptGkY<OEOtuG`(_em?3qxx&0R%J4Jt766pOZhFAFv(2<8v=_fJ5qk1y>`r
z^<>e!a7Dz`n&UQ-o|M2Dm$Y(9OO{MXd{B-(KN%e~2@+R}_iT=JJS`U-HmO>uW%sAc
zbq^LK8f6hTEFW>JEK$V4Z0WhSr0RKU)?nLhIuz)_^?omhHP!p7rB-05DF|<;=nu6q
zBbCo%GYfyZSj5G7(}!1A_97Zvl8RTPZdyR6wBYf~=_uY2YO-@YpDaPylXQ0KoCHz7
z!l8}JTXAK<?viG7H`2!O%PP|xKc1vZY*#Cjuak{b1GR=}BHnUTY(JnGF^)0ei<V)d
zs)13ms#hGp0rp2e#m(n+i<$@%>noYsPyt!#C|;ED<A7ZzEl2V}%`=0wMcstT>Z01Y
z1=Ztww_hP5thKSjSd<|#sQ}heA#uNK7t~2kEfREj1FNw|ZM2ah+|SNFQay1Clbu$C
zj;ZrUI{{}(<(S$pnGer0ED{D8nbZ>sz!b9@nN_LGj5k(n7E12HWj)R+mX++3F#%m9
z|IyiKH^bTRRSE)zmbBa8bgke$?nPq518=*1MMQUsD;*OtqthM#<0PSTfx$Kp&B_eY
zi)L@o>A0(rN`0S~{azdFqP~TvA^cwkBR%*AH0^d?7o{Q!QRu^%<Y>PX9FW8uzu16V
z;N-;IO_}F&;$kTI@6<PT!Nj)*qd&IKNRew|xSy>SUmC?m9-}BeK}v?Qw^q&ezBayG
zDE-?QDM#-Bd=-!8m7-P@s6bjC<7<*;$*6Hvr(_DY<jYlAsMY!X1xn{TNQ*`pu=K2Z
zYt4)T^A)@+R#*5qS=6J<YRBNVAU`X=ZT()y^C4ds^w@~w_Jv&#iUG;XfYlNf(>OjU
z$sffx7aDlZ*HyBsm_qZg1I_>babP1bA2T{B<grt9+O5`=tTM}9S_+3jZCAkPDZR^4
z_Z&$1=tZ@wD-{zgN<O9q?czuG(_H%Dkrg!EYfadrX!{1NxJrxR8LXFv%Gx+W1Z5b~
z<hx6<Fkdd|G$ht03bv>hbzf;2xa*~(@)VH<dq&h4VtKrjq$Ld<-3S@>k>JCakYDZg
zV1HCUIkr(J9xCsmu)OIJr|%_B)cs8|vV^67DjDvrWyhvKoK5TE3ti){LB$Gv6C*N(
z9qtAohTk|Vr|uAvdc1pG0X$v&hO|0S^9oDMA{8O8;HF(Z^3XXN>aLursRtma%<sOx
z7hJn&`Qh}K(S@M?MvQIvsKW#iF`k1jb|7mvK?SBO%D%4+lE0H85YMp>k#E@O!V8y&
zvk1PIeTpUw{3I#s#FxARo;%;F#veF(KL~l$scG$938XoUHjs*aiL34pM5vS{wIJ|(
zaZqbJ3HbS>qr_hKXLY5_Kav%KIOQ0Zf%8E~I9sJg?;A_3<$_rI3j(*=Z_j*2l?M?I
z9fh<(3MG*cZk<X)40r*y>sH|f<5z3$3EsgjQo14I#Yh7Qt$No(X()!xSWUH`P28WO
zYKh#=n+LBoXlp<LM&|fLJWnw-mx+q@WUMjOuJ+5aN}*7K7uf&|$I`b?UNaL%9K6Mf
z{wy2KGb3@5_smD%;~%M>=5&uSs)Lq<{HaAap4j1wek>EehC$%J&2~j+dJ#Kbf2}#i
z&x)3luh0g2@U5!$BlWqv%wqaVI*h@i1MR{2H?tvrE7xj@6YFus3E^`%IBAG|EE4>V
z=oQH0agD4vay>*}I6S>Fpaml0<{o9`zww4m%rdtUMidUGqno#Nocj}yBcX0!e|-Ar
zak`DH!4h55!ldXy->8NSnTT6qqW$JuSJdGynSv9plgm}H4yu+enWkRZn<Yvt)5EJ(
zJjs`tPz!G<H~kg1_b1BtL?M6sq_^;Wbt=0dsTNlQEga#SF%cS1J`KpSP<C30A%h)$
zW;zC)>_rFJ^RRYFV1&;(XN*4%@g&aFba-lvYu6F>`ZM3ShE%trC=Xuqotv>Qs;ONL
z#IF-rI`=*;Vn}fCTf7dR^~E8l8(jvt!`JnbG&LZvoWUY~Q6!X!mDpM5<5}xo0|^FB
z&fsPZ($tDmoTP0h<c-en)<zTVa*R{`1g+e>VdWLtF2l}gY#znDyhv0XYov7Kt>0)_
zOnaRi3%Gk*^wejeR{m|cz)GOuwP0Udvz(G!{Y!j9R+J{-xaCQU7LfFDxAwp51?Xg+
z>{4L=XTgmTJljTE1gTg;(a7ZPPfB!cN?Y_C*2~M&U(Qo(i37{#2q$T8wva87MYK=M
z9czlz2NgL?0!l%dxVi}MH`Dn<Q`rVdhUPKa87=}x!3|8sA)gz$j08N4jCdS5>t7P*
zRp7{<@uFjH(wp+k5SCCQNs&uUy+Uaj3I=~_1>yZ^-_;$HV-kly2WhVpuiJLb#n7Hb
zt>|CxC+#u8PbdPS5;ZR~b-n`SK)x$BDY|m#Sj-b)QaSqSkSCJbPtzk9P`-qnk}t2Z
zkVu#)r3%k1Ja+XF)-%oc(wm&YFxPzR<tz~O#w*y{lAyl8So6qzPRa+}o0}%U*oDk2
z5gf6%y`Dzx&2tgq{&DnRgFl`&G^7pg^lYZxaj}h@6^<3mkR+p2q*|1}j2<X3>Ly>a
zjG^Yhqzf=e3o*&hu-r7S%FAq%yZv<wdIHZ>51-#~6_E#*gEO>kkL@XGY1l_T0e&zF
zNA$3*_0Pzu$U@86;mqLPAJzFDUG_&8CMX7jYGCv|bqWmH_?OE9==C}8l*tQp*`++a
zXOc-`^5TI>47rJ9p5YFX3B<{bkI#lG0b%yXhG^5VhI>g6`)O^?X1&31haKW-pLi_M
z|N5QXxW5Zhxv+T52W-OP@k#wzMXM(3f<R#ms-vbZN#8GCv_wH*JjIRi@k!@J84blL
z*L6~Ml@=ibY7-4ZEWm^NqicrRPj+)vNf1NG2eUphVyLikR>he9H56z=8_p#*>rZ(Q
zXz?IX@2zE8NX*@vA3J52>*L#doz!?`B(qB&Jn)8(CrXGEX~;OV_wfmJGywXTY-f?2
z*5~f<I6+g$DHfO9tP{0y@VA(w!%=;Z_$?u4^%n1`<UOmtn$Yo2G+MX;tPyCmq`dxw
zbc*;dVKMDnG+4RwpyI(Rvnbv<(~}E%_l-3o82CWhL!W-2!TK%18-15{m+!K>&ah98
zM6zd8+&9L;KW2uP1JvzZX1;yOrt>PLbaq=2tjlP_?Du!89xng%+`*B&v)W@#D~&e%
zVPfGmae6=u#(hs(XS89XT#wMp1-&jJLy$X<C&_MgR>_XF|IP~_xeik5@0Y3FYE=pr
z4DV=H8bExg=p6-Qj600flvlJ$lJPdxpwyDQv^_NyXjM_xm~{K(75e(apMiu44m>mc
z{)jUESA*rk-&)`#*j=m1U$h=MPl-t&(^sjAF&`5%3p{-gh;0z96|L$0z2tHS=Sb8R
z{fOB#nBP}e5@89n$GK_d;^E_CG+JGVh%hrQHhl!NPX$xri?vS|QP1O}3=3xL_^nu@
zm@m~-%AMq_me5g%o%$xS5VM)JH=J;lfpjUKlg4|j8?8$&`SG(CQyb1Fv1K2Nsh^`4
zhKIg>Wg8-sT^g+XjWw`_CTH2V-UyiRi@hP<q1VCX4Ow=Q#CCx5<^YV^mfAE7fc?aU
zh^8lFNIkbvL8P9unCYt(^^C!ZD`T!(kd5uC@$fa~u+|3SR7{rbF5tYo7Ks%2JZ;Re
zAvc?!slbJEHr9U2-3XrFTy!O0QQDTCvHgSUC{msn?RxA{56J0eZ&ksAqy%T?34YF0
zhzor;&i8E<FK6`kAU!pbYjG)hc>mvB;s3;I!A3z6h73dBp|Aa=nELa8{Pn-GQIX&8
zeZ;ZJzd0HI)At(-2h5fA_EB%=e{pyJ28H+yfP=MSo~PSLJ@mhQ!2h2Rkw0HzLDS^u
zOi|ULGEPy^%QZ{_?8H1~3ax2#|K%ON5CEh20&<m;!T<FSFfbr+2mbz6Ah>QFRfbZ6
z6`qgrtRL*|DSBN#NYQvQ!1$r@U!`;Y<L&v^>m|`c(pu_IJy-u1$dSLlgdW=u$tLLg
zAXLscj>U#}k<4rPGW?tNe|t9%k?#iU*-7XB`rZ8NrTyQ(-&I=hoi=Ezw=4cy{r;y*
z_+MVk-#5jJ|MHs9?)fo1yVE#jma6E3k|3z8V2KL4k(<>j(2RLnLIJ8NlbXO`YP(a1
z^rs-sI|c0L!olifqAvf>@ALnD(cm*bm?Y38UB#1AJ^bg3Gtp50T%vrqI)2>p2WKTu
zib)o%bAUm4D}N4LqeaAZcC5&tKZiN*lb}yYdcqOZ(1uiHD`zR;h)jd@L`C$uIt7w<
z_BsUy9^H&d7H27fjz;BqVzB(jL^szbjVOPih+cJ!xZ&#@C{V&KftAt9-QHC-QMXkk
zBz5@lOw)TM5jW>Pu8`|$4y&j6HhH7)@5DoBQE2bCGrLetJVyvqW0_|hFRy<7Z!X+t
zq<rLiSnkE&f<*w7x_t*Zze@%|2~sSxK#p-g3b0niLpANQ9e}Vu1lb}>SaIWc15LXr
ziHMucd{eaOewPDCS_%5c<sZ~oyx$DW?MHvO;}6wF{g~K^?D1#Uo1FTRxgHy9F1_q8
zp}GRYia$695KC*(w!IXmjEm%VR0Dqkxf`<0*0B2WiEMB!H6M1lW$%Lps%%*63^F!#
z9w<?8$Zbj_TI~QkZT@q1l$Ge4TGFp&IP7su%D6z&u8u+%YNu^mp*9H`5HdFzyFb6F
z({TMuItuT@VH|(zrPn4_K)<UXLRq~c5gUf2er?un6yXrw$R%t^5O7W_?cW;t^dFP@
z|K|qzXJNU1g1k>+@u>c7z~e{#2`u(s^pWRJdLRI))Bzmfk1zW)zC8qQGZ>&5IWGZu
zWP(|eBw5JW%-4h$06XqS9fJxsVOPMkC!Ut+5(F?>8lE4nRpg*3>p<3HF-rnP-{saJ
zjvxIz8Uct3`Y1s<uPtkRYw6hiy||1^1`3c_iBl&_?aj*y&UqtO4i(j)G!ZX|^}UbO
zj@|-`Hs)xqqN>Q~eii6Pf_Vl6=_y^d4=fUFZwY}|-*|-OuNUS>`Tuh_o;tdT*vB(#
zyf0sa7m>I7McMmtA0QE(pM!XCjDo_l=hJIVVtDTqL;*(JPt+55@p+@cZy4{L@2tm4
znjtvuGz1af6!8om!vEt^O9J3=^Y(4#ADRBOfLJ`ebqBjM5;+9z=Xg3b3XBfkx*yo(
zPes7bOTZ2qZ<^OxDoVjEn>Ag)nJ-3>4@H1X3n%x}`H+yW(h1Kx!HdY7o(tveOuYt1
zgBM${3_w8QqqL9+V4c}M-4VWg5&$?_;-3wX1gG2mLGtW1ChowTF$OZ(>iF1aZlu=1
z9p{NO^gJiJ2Jh?!&re`GV)q1KaDssM1bX%Ml^CACLX2U+`>5E*UVt2tW<_L56!ic5
ze)(|g04?5BPNl+BGLq_hEN&8JQW(HfvlZ5^=ifusf)4M#0cdHosPFAwzfh<#kcf^W
z@ttB^s9?hhC^ScD20%CFn+T0M`*ykC`+XaDH0~jpD|+{}Hgzj+(4VQj9J}!%Yp4@m
z7h$@*;<!PRfXMY{<}CUM@sw?U&fBS3q57U8T4uCe>WbVR4N3@pb-BXI`}tF^(L^fJ
zh5uOt|8<f6ejenI?o(d6KQh5T+f0N5YAs_^iD1)kjt+1E9dmY4IIr?_+U>WNxchc+
z{Cvah72tFH)f-@R$`q!*wbO6$M_F%j?Dg`PhSIN{{t|H9iN6+vC88bl?lSwEe#qrs
z3vFD1Rh#&i0F0w+0w!mRZ-8NZf=_U&a3a*%deAMBs}r}etAG`{mCCJ=+ge&OZP`-3
zyMRdcEN`7t(nyA`S;JlG0mshpV-WGnVldlDLsm>dNa)P-oMtCrZNWjVX2PFySA;K5
zN0J)uH=KOi`O=8ofi(aWQY`>t+42MA{ThPkx6uo`>)Hg{OYY5@VtVNLv&v_xDlf+}
zyZ9}m6$x7A9D1dN@ut5UJL){M*PA6V=8ESg$ZKCPb1a(1#Z?n#Z7`6dZ03PtZ(+IR
zjP}56ildgI-&8k~;O5N1>ghAsiXBJ0cWpO|y_LPH2YI~A_c|4SWR$+qVPmfsAX-@g
z`0ip8Xw-2(ciI9#fX)1U?Egv@AS0+SzyhtW0g?@n6-9yPs<=y#Z$<AiiQ?MUw6XKb
z#JtR}iU~dI_3UAvHKx&>J(hQBwI3!=khtmD*R>s2E~5G66uFCfd2<PbV%XFZkut0+
zi4+|)dxCA7SJULE>jR@d+sefCjp4?dX=*2VT%|)B`Ne%u5REWPGZfk9wix~AplCP<
zRAEomVV+i$e=U+vZ{L|EG)M>hAEdu=-3U1eqjCeAWD{5AXr3}&)f}u&DcI?989?TU
z`KK+i`y@)6yA76rC<_C*6PyBE)5CPY3n~G@2*Her8gvf#e*n=F1z2fDxd~v~H7mg&
zZXj8pR%5T716I0cBM=R}c0UDBsaYB#?p4syHd6dNDtZHtOSC3^IR<LsoV{o<DSA?}
zu4%Y%sX~Y<gwhfk3}Q7n{ayqs=zxnp4-g+XUEKhA*Ft7p_|d7Ni!L0Yuz8fKfNvkw
z2@x|(rV$3D_vu8;A6d=Y0*z;pu#VCp1tTz7o>LQvLWKd0U$F*-L(}c~v>h*V`LmMU
zk%=<yz459BK0u7{!2}pac|mXx@qGq3bvTf~C_kqYFchGneAVl$291JVpOyghR(>Pc
z8zK@7d-!F|jltBerUYz|8{PswvpxK<?+Mgb%riN~%#W;MOi~Y1A4!7nA$VrgCHN_y
zOxPC&Q4c|PMLE<0=a1JUKI+d*I{?Ewe=#$?ke86z>o?c`umIjpA@9y1&UX;izM{Hq
zG9|I2vEZj5UX&^L4_La066`M1MZ?8@*j2!sx<N-jH(Z=$*BT?6aq(lN1>9Bz-Ykgx
zmnTS#ks;hP5fe#hukF@yd4adI1AOE-(3`WtitB~dgT|{Z`}|l}fX=}IU7YhN05`|a
z5dR6}05HG=k5AsDKr-GQQxN;;+$N^^4RQy94|?SeLx_~8SwemA+w{Ra&Oo+FKoqoO
zbL;@PYH9sXy2~op9fT0lKqz>*+u901=06sP@;#MP@A!R7`^eE|&y3;H76S8D$-+E%
z_^ZF_a<WglZZp@1;1mG2zFe8{m&BT<+LS#rzMSGH&D}WzyM)!8HK-G?BT{9{Y5+Ft
zf$@x&q{m_>q_xzz^R?160A!IQKjiZ9IWEsYZVioPjqXihs6gHxe;!oNsoe9U--6fG
znCl-xV<zex7^pm7K_^e7|CMWVF^u<&T+vpe%GIZMqX5?RCc3rji7$T~5pl9=7VHuL
zsR#gkrzDoZBid9QgOk{3lk2{5S`pn+6dL;NwGY*QkF1gP>TSti-*&w1%fJNLRUwvj
zAHq^tBDudT&?pp(xh_roiD;@OrK>dK{0hvpV=;tSVlhvoi3tfwElkeOihZtFVw>gY
z(!NW}$^T&yVSHqo0w^&e%`qhtS`h*^y>3uZ&+K0THbF}LmaV<i10_0CbXTVE8MbZd
z0)>5`O%c=tldiF^3Y7p5HnzZC858!#l_h+!)OE&FQ*^=|{~;{r8S(xMthKkcr0?Em
z1>uov><R@M7GzrXa#9RW9Dsf?NOH@^JR&WGa5on;T8`#2at~R)vuo@Q#6Kn+P8p8D
zqAhb{Q`f&*S0^fwZHT2`O3H|FA*8KllnngN&4gMdE%#MX5aG$pLKOdIfi7+WXD2A<
zqZ=rSTw04}K|y3fr`MVaKn7&4Jpcxp<u)cGqgYiI+330?+t|auMl_<Ne)PbOD}Z6f
zK4&%3ht~3oD4<lzm%1)rIrIinWT$oc>(LR3P5&I)?(%<WrV|uje1vbBTt?USRpCwP
z?-xDN95c-ypNGU0le_YJ6P-%~%u8onwAT{-^uzZ!_|sIrjjZ>XK)sQ(oQB`@z>Ri<
zQ4)L1D{E`UD4Y$UFoKum(<V@sT;L(A?f)&NnMgA<wEU|h2dXU&^_xS_RsG?_k*IrT
z!XK%`z~MPzOFy^%S&urI?ZYe5=`_HA(!BZwKaV>dnOM}sfZ^DGE-)PphTSAq)ys@p
zdS@RV0V1ch*UCR=YpD=l{A&b^xMgnWt`;rBRCNppIl>K#`Asx+|1)%Zt5YMkQZ6c1
zIoWu%q6V-$Q>Gj2TC;CDbUJUY=58!A4nVf_&avtn0R;>i<--eP;@TFp+pNm)-7{E8
zCzWSmM|6!JSQ<p;<}o}D+K7}_>(P@Bj0a*J6=Kh7+@2vV*$Dgkk>@O58p|9@(hekd
zJ_oFJ_AXN%r8*8+@VOe(GAJgX$6<9S%=XnBlV6*H@?Wmg!m73Q>xHMT{*hFJC&g2=
zJAFb}&B`0kmc29zoOuK>Vj9b01wBB;jmD&s{=PxNSC{iYeV@Mg%#yyeQ(40@Or;t_
z54QLf<w_<mOQXFN9ugwKR=ICPy9VfW3(Y=i@NP3AIoc^dN<!I0;^KfnPQ*_mzTz{R
zR{NOGu*=wKA}%M&!66-gnh{3mEWS|B`z*59nA9wWfX5mg^BpIyImXPU{?#{NA5#3e
z7xjiAi<TRVhpiWuR>k6iA?jJf#lO9wz0$*%x3D0Uu`|lv>#AWEjW*ABWFG3~NMo7g
zX{++$E*D#vzx2nH#^otX)!LE{h}W$6_WDRFifVK>Ia$E@=IW$?GAAVq8H?B-4FmZK
zEL!@XO4%Q)tAUMe(0N$U<V+=SxAO#KfT=-3QA|pBRj9dnw`jILl49)P=blWNQ<qzL
zQd+@jxW8GThj3Gl$&O9MY!)g@Q#~nb)k9=KvhHd+=0($`Sh-3`3e(d#Eu>!)Fnn{l
z301^5kePWh2N^KGq}vey-pGFYVdlvZA3~h8ZX}NHN)Itq>6H-`ZShRu^A57eV&oqA
z8_*L<?s%WAb8Ya@15~yH3dXs{RW5%`W=ObEE%Gcvj9QLk?&{IeOb@EP)?|zwy^Dwp
zoNS};yZU*Me!`F?Cg6AL4y`;QCJE!yg>vPr+syMVa`GP*8)>@}V#`;*-Yb5&k~`(G
zS)IeVGe0;2@<2O0mesw)79)5N6ywiMd2ZnJA0wj8TJ(!cKMN0(pW+-<SY2?9?e*WY
zM^qTftY7PWSJyoI<rejX>Q3Jyp?MCxv{=hQQL3wV0M+Df0%<ayoj?wiipx9cSN<7a
z-IvW|zTH9m;kR5wUlV|(e@PoI4EhCRx;k?ohl|tBhNEFQISs6AMou3B&H<L~GHd=W
zr2%pBX!tJ+UNR`mt5vryRTynmhAAW80G^a6JOV063E@f1{yR|=n8``3jL^&qf~;O-
zM+13t#Y!Y|(Ap)<YIg<tN&Dw{Vwn|~{w7tq78HH@Lb-H)f`mKy=o^T%n!(!f`A*PZ
z5()L>bziX7LC}TpWK8R<V21@ZD9Xd<k8vzjYVzNi3VWr8chL4mmjyPKnIiHPV$ec$
zEn8T$MlH}S1Nyrh%cu2_b)4DvW8pDw36r<OxVWam5xXH~i<4b@hwr>oa+Ge+ZM`jx
z1EItL``D=v8p5n+FQOc*ot|Dxp#}COnbQOnm3ENj@0IF$_MYHh4>4dG5@dN00L(3&
zMUD1feTJVB27;cnFNIH@?^+@`1B|I;B|Y;>roR<WH&n>W);V0w?$W&oJmjL_a&6J)
z8cW+%$1W}_@|1xU3J<WPrsvP8*+TP!^*c|pzv8gGJelC)P?M#qq0S};^9~%n>ukgC
zz;l*VeD$6VC>i9H+(rNCQ~gr#nV<U_I;I;7A7E>l54jxl+r!G+G0BWHM1xC$_*2}V
zUAh!_N&KuDgP?qQYg6*=0$FTP8n~S_g38QFQ8XEQfh5yTffD5~D^6Lhq)%Rza*gCU
z{8#mpCD7VGvi*Hq)6b;m34luvSU!jSdHgwFzA98JepI@D>$So7*GFU_c2|^P+wXeZ
z5-wNGryx7;#?{v@kxz7qLWu34%%Z_u`vkZ`+@=UzIL$UiInG*dS2fsapod+r*E_r8
zm;~O3pi#?d&r;oRg3BcB`S)dd1Nt65hm~b8av3_l*Y8z2PjW8WLhYT5Fl#*3-hL#z
z^eDyF7=-|z&`Rq-e>2z<k-P8h^~!S^XAu9oX%8gG_Pm`Px5Fn!OhbnB|IB4M{qZ?*
zn)a&6?|3jtdHS1aHMHVIoCOiXU>1bi9mPt9u5agXx($HjecbI~|86NQ%yD{@>CP{H
zB^EnJ<1Nce-2UGn;6fqJr*!UEkpkvLo!5GKSE`kp#cuCpPi(Ese#>)VgdI20XPKP?
zck42A@x@U(Z_Kibu+e3C%V!ZvOV;KPPV5l6oan++H)C_Bln#IQ-6-qyFif9d#VKy^
z>1lGkNsv8vU@GgIH4r9LzbQD_!pmAi$Zpw+%{b4IMh(%{)G(V$ctovv9bbV`UgjuA
zbe(TxXvr)BZYEr;s=n_gv0lZ~FpX20-p}6ZofSp=g@rm-AT$`&E{gVQ@i+E@S(=Ze
zr1U~VfNp24y<t)b0qwS6d8oFA`GVefX&AlRlq-j$tsK5Au|$cT;p~DLW$*gU0h@MQ
zdV0EtdG3YxoX5)5_mzW&SzYkw=<9lZN|8SF!TJje>pOEx<xxI-?A_d=8T02_)7Nij
zSVo$>rBW%+o%g3-7k%rYUrM_?=;TRp`53WiD0#-vI_~+QLa)78nWgH~V_%|SksT(!
zBrq1zgz8hIc;2~a2xULTa%|7QTlk!)|IOB{OlR>7XGN<#z0+e{m1;?3tja-bE$YD6
zQcDxkOW*6*g}$@nKG=KF9#eG6KNg#~oluAue|s-bw9@6<fB=&pX;-2erw{@D?LQxW
z5ZaP9LiuXg)wxjlJyo?Rq`I*V<WARiQv1}~xThlA0O<QY=nod(wM>pU44@SZu<krK
z%cGoRC@&9lGWSO_mnS5?3xJ=BA8=mJ0@cD`C&mD5tQtrg2Bv6!z)dC0<%_C8o!dXP
z4;?7n6imIvTr0ly3v`ukQo-1JC#e96JK~R6H78mcKn)&XC8{Jzyrrp;SSLYsX3;Cm
z<+iC&)z3du4VV@aaFPO@%KFAD6363(AxhUVm2Hq<#C&h&vuoD)KIdNO#%NwN4_2Zb
z*reEdj0FX+4o9yJOre|p>g~2)EkBcJqC+*UX9r$VMe0;N>d=}gFd3fEDlRKg_$fj$
zkrM2n>m_0}#w_5zYhW(lSch!&V2@p)^5XmXM1Vu@QrTDI33~S*9I!ns<cO!1^39K7
zIK~kCNcjXXL50g*7SXUteqB7%CKKi3cs5vBMD(MP@4UIPFNAIC_uG^0Ers2c)ewIk
zzuH0aMUrv_@_O$X#cx-Dt$h!+Ia&7m`V^!ZvF<C2Ao06tn2fW43%sm6D)9162rlKi
z0)D#NvsnFTDlv-1QlyTd3krK8MkeTxNzP>?{i815dp!-4{exy|;p6%{VZIolz8DCq
z+c5G`^>)g}cF=nDR;zbk1MlT8AN#$Fl({i<D)-Df0O6rDCe^l@xIx?bUXUaLFg2I<
zLKh@o7^3P$k&>E;Dtmo%`T6rbWvzO%y}98=Wcz2=cy-KzS_3A`X`1&qC`KD8J(gl+
zb{_Uw75RiTUEI}~DAJt;UHR6<>+;ZzO_X~X`sZ&is)yhX+4%ZiA$vJ&0u(2Yr}55`
zISu{6(6yan@ix&5!4W%owz^3E>8`wfRJMV(Gnt7FJe#cH8%P7F#D0LCd2>!b<^+Tr
zbT^(D=)8YT6^OUv;J(wW5nDemD-LXf&&(5jXG;uQFDE~}BwJYfVtxaZKTaSg<qep(
zWEgK6Q|Ad@__pt9NAhy!u8n`yejFlt(7dw-j8DB&_lT~AxUen;3LP)rc4+h_Mjjva
zNV^V*SAJ$HVAC^>`<mViTFBMYv^~HV7V3Bl0{W!qH$HmfFCgb5W7QyA4@*|uh7Sk*
zPo6Y9!TzSM>2;k$2!&H<IiA$Z4QbJqx9bm%qTf=arvZ0EU!F*td@X6}d)W_~*+u8y
zjKvxc+y%OJsEFoH=bqG8<>(#MSpwF&Yy2W8A<Us26^!0#U%3h1r}QQsn*pux!EZTT
zewK}G(C9f`x&<_Y;G>06$vPo0dd(ZRC%mH!W7%bQ27v8`$jk^2h`V@oI)8w9b%jo%
zc|7m$b~<>Ymd{EBQ?KX+p5WL)1mw(AXdoM}_hm6z%~o;Vw*`2v_b(yy4^L-IsnLXa
zynfAm1|#}$*R_#PHX^Z~SX(Q$G^sQ@&1vy8t28$rcnb^)N0ns{Mv4$wojKvc`h12S
z-W34J33KMoeW%~W<&GB#qu>SQ358xh1}s~r*S^A~V3|1IquaSzh<XFYdl3wR+P0lj
zvONtq9>7<md$K*7M!K`RBfb_s(h=5{P|?3M%V1ph>-1am4(__hfmeNNRs+wt>5o1Q
zmca&AYY)zeLy*aRzQjX?V0=%x=mxt{Na5{HstEH-^S-qNZgmc*8EYpu=Vx>BeM2-=
z3Rl#q2&}j$t|;$A-AYkRWjJ@G$<HLW58lgSXR{^X2vxW|=I6StBKNGY9sV7R>m}(Y
zsDFK=ggp65TwKx-r}Lci8A`8|CGmC1WXN6eMBc*XLA_1cV-#LeZx;YsmPN*h+_XTv
zwpzHe`PcAmOPbNH!x>X!<gR)e8Om-v#x7*^L>sgk`TJYAVRNUjKg*t9g*f^0Q!&ru
z!hEF{_M0KfqP?(2l?~~yzX=L6+mi~F%!tDij{~^ZbdRgYdE+BLV=9elRh7rf(ofjF
zD;qz4ta({6Ucky4cY@nlkxQ%daBkdszBt{Sw^-QE6nd`dN&36ozx|e{AyJU2i6ZvS
zg~Yjz=QEN)yY?%vYBA*gwQ4E$or|;LVfnB{e*I<2u~B~{Mx^JknudATi_E4>*j??1
z9T3e8Pg@u3az!ZS%4nDtb#}Mti;cNN{#kr*Z?n!#G3d_^LcY@x)mOOlq(&UPzEZp-
zo(g<nLZk=Vl(;dBgAYrfQ{IQ4ws9~GmK`4|fBn7iI;dxM5s2uHz30J{M*n``r|r_h
z-W|n#Bcp1<-cF5jKErru^aCBDb!q<p!`NGZMZI-x<BBLCDy5X*(1;Au5+a?_B1ngn
zNOwqwfTY9#Qc5UDcY|~cEe+Bj9YYN;z_&+_=Q+>!o%jF#ugeSPaL$2W?7i1o_qx}8
zXa8JiOygW&B1N40?AfO5o}nN2%5<k+WW;+~DM<!oh@3mAB{;37`Aa;wW{;cpLFY?O
zSCjV~L?D{Pq`ii#mQkz#w@OY!yKOcNZUb<Bq22G$R{d)x1rfU>irDw6<2P~R^FDv0
z_c$9gL?lSAc<cwo5mMKWEW#+)@U{cgt9(lBEr=Qh00{H*Oy+I_<)F6iabc&amtsmV
zs1CV`BTP~3?%B)?4q+CALL3IFMv$1#^z<Y`xerweHfTpXFw+Im09%PyJ?XZJZfu*S
zI+9aF%SiOH9>`<ZMdAxc7DFmG{f6H4y{bD{doOiww=V{V)`MR@MysU-6YG3@p6`>K
z)*j_G%YGmxva~JsqX|(KOp|tgb}IEqW^3EBDY!%M5MjpMsHd#C$BV7D31HZi4u6t*
z0XcPo@pd+X1;gnVP2L<FmnGL&I<YLT)`CPr+_lYB(nT$c5rtOm&U>A}2<Lp-hR}b1
zEekf|y4l0gcb-0N$`xi%=jpm{ps}V(TX9xaAJ<e&yoqrZ;X$!-ia>Y1CP3(>h2jn>
z<btVBWxFzOH+cS(<L{$m3GqC~YsAeMB6DK<DY%k$x`d8$A&^ZiLrAeVc3g@$plBTE
zxvZbwY@@Ed%kjQ;yRYZIHzbbD#3)I;!3tu+V7+uo+=f@kZSc|Sr}<k<cSsX&v@OjU
zR?f9d4JUYd57(8O0jOsUWO~JeIA^5op<=F5oi1^PYV@7ojhXP;0y4r7X%_+X3&tr@
zDJn?L5su607nVK^9>NPOtjh^l^TiRj-T9I7qIGi~)>%@!EX;)-U{f1gM>sYCfjjmz
zPlRdPBh=GZwPK$iviBN=YFNoVd-O8-)pZQb!3w5#$_almtyRp|cfJ`VB$=jpLke#2
zmhheyx8b=kDyD3BW$hTYLs1g7-xjb2Kqt0f1wfgHaI7z9K{7SYc|!f)=d+fUh(=8f
z5}dqHhz=>Okzgv=Y=ANG=&!`*)pO<Nuwt_ebMw4Ojjr_f>yM0|nZJpJwwq*6S~4Nw
zISNX|CZAY1a0xv%^cZA&p6TubE0vSZR}4oz55&hwd^(G7v|H*hR$80y`&lMLD1COt
zK6hb)!~37!x_pflr9Mc@+r}4!f9uJ9|E#JL92wF%17paiM>2;fF``eyk77hwWf2*v
zpJzU$vf~m-xJr3@_Xo|)=kY1u7QZILt}K0n$N}wu>V>C2GVkkb0C%jV74GOYGWpP)
zYJ10n4x*M(mFs&Li8t6(7v&+?5XjeV^zwrk6+pNiBRe$(os9VMD;KU@YSC+HG_uwN
z4r_Ot;*Bdipx0Z|ccOjezCGK@TeaYR+mU;UEHXz3nk4szB;!-T>^1fpU4rngbux>J
zm3(~Wz+CcdqkTe7nv-l|urhmPp8QrfMvCGUS&1*fP=Xumt5Z?45vC4cYaG(#AUAt_
zt4MA}29_dm5~b>{JJw8bc6n%?rc0~|H5@Q;VmjMO-EBN0;HO&je_?+89BJ96egeAE
zA&kvJ(eAZfu7WO06<UMJ6$DpkZ&3iS0O8P5y!7dZe48F2ub<S2DHry6aiO%VCyR%A
zdLbP|<0L{LWq=D-b`tQjc3Lp>)Hv@&*u>!|!YcF}>plJ%-U8u?Pi4Bb5MWc)@?sIV
z6c<DU_EgVIFTEc{LYV_|58j<}o<Ai*wSJe>OS+A+6|7Ye6((+*2%W=s4LtvmYkr!w
z(uc%100?=g)aDUEG^fc&D{5^{^B)ohz)nron$FMmF$9;fJ=nD(Vc#!k%hg$CI(xe@
z3+vz>!AA`m)~ANLSQ=e^T;B)}3+a+Of8J()XlOhkwoDTZ-Rr_KmgKjpUj1C1R;p@H
ztB!B*QDgt%J3p?(gS8Gd++D{93GHd$_+LH6AVhH+7d=p+p~3)vk`sjUFyf)nd#G1S
zfmewZSZ==M(^DWN^}pec<KKGu+@_o&0%Eb3Nu;-rSjf_R0ms|7vmHSof_aMzp>b~o
z5CbqrUykK<<9huC2{qaL^x<GlIBo5#tadkbKBuzJpR@&fX7&C~`Qhl4|M>q(`6pq%
z4Yg~StL@ddSo<wj$I?iI)y+Q1qXEy_1{NZs|Bi`LE_{iBvk@e&LENsvi4LK`7G#VU
zGzlb)opBZFJj;IX7X2Ls?#&k(H==mEVS8bz561f*APwg=6@DcSA9t~bCZyhWLfh_|
z-7>g+_QG6#%(jn;Ll`9rAv;HIN;d>X2IBB(ZGGK1y#h_7(u6W18aYU#oy>?;MC(hi
zAI%?Au`Ds{6s`D_HF5=B6&VbQFGyOTy`+WsuI1avuQM31%e(#5;&@`w%qw(Q7*on}
zl{9^aBng~mt#Na_RXrc@+Qf#mpx<Kcu{ny;q$@`VQ3jVIjttRWPhZO@2Ost4Vi_D{
zCmB9V#PQVX;^D?7hIWco73tkOcp#9zNsmn5qVjFVxaF9h^cbk~Ke2X+F~{xF(F*gz
zuNt^FPUvmSS_O#qsFG0Q><mK*++2&#Pa3G9(TJcYg}l$rHXuzui}DW0di*3btIb!N
z$%Bny=d=+ZiWuj4&hZsg;K|nL<sE!go;=r_mK{f|*UPu1>Cp!sc%EQEOUq-;*AY~D
zuAzO!h)&IdN-ts`4VN$ets9lN%r11TKZYoCUoh(6XS2gHpx>uJvtof|@8u?|nFz=B
zqiE-+Lk5D<-@3H5s-a6F{8!*W%Rb$r*tG#(OZ!rCnO3XWm;kk_<tIoK;vL)K9tt6=
zh&i3j;znfh-u3gw^a@kH^)vvT*nmHak{^jL-w+~oS?xpIUh))}?bAORQc+~V@w;;g
z)?$EjH+LsfV`jlkce(-e_y<60@#0Zxd<p|)ZLZ<?hgqauP@JB6*Idq~UHECs1wI}j
ziQe7TvvEcHPBse8$Xl>0^FQG_sR72CNyuSD+9z!aOASWpW?(tBq34nU>>KMhzQ?P)
zh>xiRMJoJU$!wcnM0b9v*OTSo-H!U#jC(J4slu*!x)TJ=(ryZvNmmQ3iF5RZyJ>^1
zpm{G$vm*O(S?P-yp3#<z51Ce@G)UTuuxE@et5RAbLLIoJY-S<q@o}%Amqqood!NeJ
zLYqB7%u)KHnGT_Y3`1X#P83C66y{}G1;#Z4=P&D^UhAT7Qpae&Qi*eQs`(}8(YwdG
z_XqR$+If9tOxi?)$j1;IuwWY97WVq5{LpVUx0t!HuSD<eaYXSJx;2r8;xR@&TufzQ
zHP!H>-y%7T9;J`au!<+?$VfDw4>v@#z}rjN^#yVdC4S}K2u3Ayh>TE}YLHzx&@*Q4
zc_~w6{`vrq=1MLZ1PtPyN8mx!HA6DR$+oH76*ap<W*F7q8)O%i)(5-O*vXbu-kI;D
z8KVl1#LKmeu@AkBw4v=`&kj;zO*l6i1D0b05mz-HJX&q446qgBYky`PV`g5Bbn$i;
z&q(MT1f8y9c2W9U8O-K_Na44;_L;{0m3TCKZT5NeV528HlV)6HW)AHmrDca1QnacE
zv&F7Gik`S@SxV3;?^ruSgQpIYwRauLQL31f-D$58Lj>6rNBJ>AGV#l~uX@E|Vo-Zv
z+q?3RJI8SL-Rq-T(pzP9-#-ANx=i;ZtWGsj2dI!aaS~Ob&nE-931tycnHL=kWGw9^
z#1}hBqmzNQ1-XB1pQ^!zfo0l}y!}x7@TxGQ$I7w^Ww*=g;C}l{?1JV@sVvETngw9;
z{2LVe9Dxjq!4oT-Qtz~doHUvj)Rp7sa!<%jGYy#R&53B$fE}ahD%9S7JJ#RMKYeTG
zJeFPCyb|gwBzXQgiJ2@0+sq_b4~jrcE9LLuN6r@P#gy4SKp*im{W93knZ4)Q?+b~l
zpJYl!-1Ahri+Szx9o7!A4MfztxE(o%y(=7B3<^mr^56IDJcqKBbwj`5%AdO=bQ0e2
zyLJ=dTZg6*c$y6e8jw~Z4BuY1u~8>zX&3}s5Yv@9Zw}9vYE*xoA6aB<7Fl|E8{}om
zx5Ep;sGxNVd;xi|!PrpU6Jp$9@2&+8d5d)1MVB~31h9=y(`@5SxRX0+i7c%1;KDZb
z-V(hnm>Sf6Ga3v%xPEP@Z0BW6feS%vTWTu)TjRR~PD+8UI73nJbcIrfh;Y05iE2u9
zQx?2=v+Zj6aU-*W%KC(Lr!wCL$sD2>CRW%=#XJ#HDNe0=Gh{%IZ#2cxm#fS~NPSZr
zf3@RTg%Id^izdoHdDn2`hHt~^c0&VI+oC7-Ig-jJ)0Lq9YCCQ#(D!WBA{MRmE)(pE
zEy7XL!Ka>$z%EFd1n<XPs2VUSkC9?RUFu=1U(GJ!wb2w&ncJcED{q>S0me^ORGMwa
zV0VdHX^qIp+f1OU;!ki_^+T@G*Grqa2M%NWMge`w>6K(5GF?bZH8TMb(vvUc_^1Jr
z!MI&1p{77j)L}iC#R>7`eE*p%@D3S6NtBx1>H@ei2*GEaREa_ncM+(v%<X{I0<K2L
z_R~V$wrBOjyt%Gy`O^Fd&DAG3bLKj3+^!L@a59w!@RM|tB={4QGH(S4wcH%~&UK4=
zEB9D<zxDbR7Cuauo!2alkfWbGPt+){V6E|}IU$4maEzt{GU_6(n^~JZOLVqaq?X@_
zdr|lGvlWQquU!KhdE*4siocS$Py)vyK?{(F&C|TInOTI<Upi(X@p{pbA#kh?NQShS
zy|L%&(}_;zV}%BtS)7ocyP<E7?jIE2j|+Z%S}18ihETS?ct6TXKKR~PtH)k9A8(-^
z>`_U}Q**NdQWXQ+%E3L#UUusydxgg-73Tg3Y}YP+LEPP;njR2`tYtme4Z6bO$guhk
ze^Mh2$4_7so7cB9=C@&*={<{ebQq(DHXwJ}M|g%eyr$UhRX?oY)+TO*H1uTI50^rK
z|5{jqU+tHebpcE|<hp=yUedc?amIlVqgF{gX*6nq-qZ!>ZuymKH||%5`ndD96Ei=O
z3u2@ss&`LTAZ17`Qfi6Ng*MH$<KZ#n7G*r|WmlF{*L~ae1C*6S06(=j4`9*@`clQV
zvUCDFciHWF#%vo{_@1PnMdxns(fTN>XO))$MQ?y*9W;cd+)q=Ni>}~i`=&A@5@a^r
zwMcIno)XcP7kxqZct2AR>o^RobaVDW&F^GAS;yQc9%63sK(lz2lh-E7%m+AyJ#M*+
z)|ss`I}OH4B~zuC*bI+s3c%Cr)}2NU-`QC&SL@`>77VOmE;SkPST+KeN<JtrG<Pn7
zm$#?PIhjh{HJmBRHP?wIXuWvI6803wl3lBfqmg}=wD%?jFPn>$+}70#Btk%BM@l$%
zt1w1qd%Et-<ywe4;E~3XkRV>xCy8wY9|)b`wCwJuqffI7QlB?=+E!H+z%-eJ;e*)8
zVEKM4V4yyX!`zNIxSjjZ_R0MIFJKXBEEfbZ6o9Ih?k~BvCaz@s3i-l1n6Fo9t3+sV
z&L7|`ef1&8?VyUpIb+iDC*4=Jos-veKe+x#R9mBk5PK`(M!p)fVYlt>DZ6l`^Vl+b
zHs$<RXSfkf!#bG?m4zs_)4}iyDRn5V7q%q+x3bT(5$s$`<5B6JJBhU*0%Z7{>R%3=
z*{oFotisJS;JVnw>3fC7Tgf!q%*yOpP1G!Fft#xQ4=edb2!wFYy+sI9!R~r{XammV
z2j7O*{(+!(o2e+}P9~gt@V-3d$(xg-q}p|hRaW^9GKA*#Bam!ipRXA!KinKXGze?r
zSWKTjr!6q$%AkdpfB0Doetd*mY#E+;UU>t|IX`G(^+MCFy!p7^<Mbpps5nMvqAX15
z9<>e{_)Bp^1Aj-B*{lpVwnbeoVjNtrpMO1$_c`6O%JRB656CE$OdKD$>c!%*gDKEv
zvKZ}Lm^W7p4moeq0UPdc)5zU+Fw&|jZxdifh`||&$L#XY-h{r8OES*j6i^3OVwVE+
z`p=wmTFRuS91{s;4Zf)wUAE;nSRvRhLPUFfTDloFg7t<(O}z{$$?_C6Oe&vwPha<p
zFP76PBqE&-Ri6pCbk=)VZf;~|af-y3bS|dt9f}T(w^U35UGNZ8vk#8V6yVLONSp2`
zPEO$2R)G8;mgELt$mBYArd`E7Dvu1#%36)LF5?q~X}!5?yj@MXms%86>t=KveKG3+
zC#?QfLZ%dt<Jxmm)@q$pK@a!3LK4c}mL%i7foVq?S%9aK@mM8s%E|?x#^IOrjyB-i
zzNx{+$v6?$+ZmD^9~nI3RhaBdSIB6cBslw;H<_n=Z&S5`=?ir8&M1%lk8&IuQ(~?u
z##HzD4Y*u!;u?9LLZoy_{Cf9Vpf%%4jxT2pQ@wQ=y?&C00?xF=<Y6umT?Y55$eX$4
zZvNyFoVqlDFO){vlP0ZW397TArKPN)dN#ZEyLS3hm)=Z)r|{EauxmR}y?-3;Wh&c#
zys#5dBv)(U=arc!Pr<MCax7wegBDv4OFB+N#Et=~Sa<tY=uG#T{@U@ZjOK=MIi%!j
zLn%vN2o8w{e|J|ZCqY_Cr|g9(jkqa%G*gE`=y%6SjR75Qo^^MHUdR(~Z%4^wSn~o-
zbCO3;j6f`W7szTYBPm}+D{ldom+NO<`P^+l{q*8Z6t*wackg|SUGi<c+fBE%a`VcQ
zcI)TUIztOzI|G9Wl7pvF$cLVIB|FO&+?P!un3a2dLPDNjjb<1i$sUV~a_-yRqS^zl
ziO&xk@_M3V$DQd-R{P^jWet7Pf^r1(-Sug%nTlQ!S}wvPe^vl(r^f94fTvZ~w7cc)
zMtPA(Igh8+mQx$S)~JUA$7g`U>{`dM2<Zou!fQ)7aC^f&rW40oEM<5Rq8jYsvi2qh
z;}I8fq|4paI?9o?YpUFiH6w1S@W(fZX!wXCPKv$+(}y791-!omZzkcKPoaJ01d>;g
zfHCSkm0CKv=+#;&N4g{#N;+pDV=kn?z}9zllVEich^jN|&S*03xk;9Q$kl+3T>4~v
zVs&Ryx_z$%SPr&c>?OsS*X(`$1sx3PCLEkM@%p<)9@>OGySH0wN|xHXkZmetUN@h@
za?t9}#G`v_^NziS$!6ro>G42(Pl|IuL9CJ>)}V7aWhZwGx0RdPRR}&3stx)2ykg^u
z{>9dJHQoEQoVaEkB^T`Qsa`?P0}ma6WXqz|OaW;KR33Y+`#~k=0-bR*3gD~!r_!zC
z*CD-vQ4j9sPZJ%!&rOBY0Y72XlqJjVj$q8A1h<!vi08F{wZ(d{K)j|<o3C0c&G=aG
z^7@T`K<z=7hzo!xBUM92zC4ncg?&_f++r=a5K3j-o{GbB_JR+pqF$hqoE4|NpdII&
z{3<EQST=ag>5*rcjpMZAV|ApVE6;>dZpLoxXYF=sVOM3VPmtUZgc$7-o7G&w`@zi8
z-$fl4iI3)Khe`2{l*~0|4m2qWvTdLomsu$Y8i~ZX<hb(Ct6Bx{*OX>p(sO5In^}X#
zNX`iI{y0-~lh!vXW7(kfGu<G$z{Zuk4FU%A%m)lM!4`zosw^nwbM=Yr(lkt{A9v+Q
z@s8<y&&t=F{IAmZ@W4muq#(^R&YsgOe(=1r2|}=}Tp)+N_?_3F$$yCJ<yitpJBDLf
zK5>W!7!HVKRfHiwpU2n_2|-PpOSoB31_r?barU)2>%%!OO2NeJ?W@$A>JE!;o|C1E
z50=U@&Ed&?>tr{He;xQ9=mY;7M@k^{OzPh61!KC}jQILO<Fs((H9FJjeY`C&BlV(q
zio7exGwXfoUhWbBqVo|P3B9NI<kWSs8PlYGQ=vv^NxDW6GB!bczA{6K1+&k%no|j7
z@&dZ)w|&bb6U{I-owNpdWL+->9N8C60c!cPpE7eDr&4YDG0fZ1TFqB+y>h9Z<&@xw
zpyE$~j`>$)c%t*h{;>KeY+W0((4Us^SLuMR9>g(Wpsi;;?+T_Of{~8_2{-Y+b22S+
zZO|<sCgYFt77UqHbWB?qO{`rPZ)ZfP+<IQ|*i&X%r^0pISg%16+V}KvH6X-En)u``
z_^pZav8^)?0G_7O3RVxHg~O1tl;eE>N;ziDKYNL;E<cLl>vKFf4pjH07mGeOBcr+D
zD+J3m73!Ibuz5$n1P_p@$=4K`cL(HXRK_G24^fKQdc7e1mUEryy~cVF7=*xS$W75;
zl)EPL#H8wBzU(QlzXOMX$;i;N070{3>1ltqP(6H9VV~Yq!qQX+v{Aw&VM39Ep^sC;
zTumRvVHxzHJ535@2Z3PlFtGKUsL7x(859`O6Kdt!^lob;V+G~L8<_~C-e$KE8l$n{
zgnu*U!+&Tl*gbD&l%p&TjqP|7ZKO+JZi_c4;&ue)gQD0RX+X}no2v0dfkhD`Djc*?
zr@m0jf0L>9n`KSF%5U#e!{0IP4g6$gP&~rC)m&U^!-2~#aWv~0ldVDXuG(SsS&=Tm
zqw{u5ZpS<9dQ+)qT&6kNNG^29ppV|uZf>d7INEDC4T>|()YdlJtyq-OE~K;a)S|hR
zCLi>;>Dv_%=05cgV?=7&iZwAKBlwU5U%NU0T;vckw}0;+^aN+uJ?LyKDkl@q+!|&(
z@*4_u(Oet<LJ{R*%W<zMW*xms!v+r-*Ns^=V>J#!)koqyC)BK&9-GV_qkTna#UH!Y
zNHRqj)vr;e(l2B$#TR#j#C!(LQs4FATOy{g{jRH+%+JliTw3oXeU}xh@l)O){=);U
z67rt~>^bHV=O(fI==tyK9_m>Up-d}6S$gG!(w)2^rEKR0Nx>@uDQZ@cQi)`PO~j2e
z{h!2SCv1Dm%HhxMW!!}Ae@E@stbz=9uXbg0u8kCfnrFeTtF(+lw8?J4WJ@N2qkis4
zYVf@p?3yW5mXg+DO8M4eYwpboH^KR%Vb2A0#cbOp%0%!NEkhd{c#1q6VD&nNH$hk<
zm-vY<>rS`9BR%RD^GCB4hGg??^WilLTnfF#nc|L*?>0zQ@^4q4fmy5A(o}aH=%_6=
z6;dpVY#R%%A!1glz56o!oE#T|ZZXaL<lPL=Nzc?-{r$^cSzwfBRHx1wL@}bqiwg<V
z3Y%=l81mX|Eo74<)H_xN(gX1#zR_)2$gpkq8I1w`azb|Mgr9WxqLs>9)E&S;#xag!
zE=S;gu`x%<J(ApxF^(hMZ!atbfML-tQ8spnzN{W4!+46@8WZch2JmO5(Tb|Luff(R
zD%H^r9HWWG33VihidwE(Pq$Kq(s;>rFjD%`yI7Zxo21A~UorMOz+*@xD3<@gmA&$8
z@Ee_VVYOo~-%poMBDJ-m@xa!2ND=iuWzon+Im7WD7?Wiz&U~H@+WE|Bw;J)4<5&vr
zju*%2`r}+_iM0tD&53LiKp4nQxfKD7GM{{$qb~p<e#r$)MV)jy@J>cKoHr8=it_I{
z1eqmt0GgsDt=7Zp$ZPRzRiWPb=A-!o;IJY`yH>1>sAtm20}_aCDQGL7o-Nv=04YGC
z{C1(2=Ph7T&)O~fu0N=B+U*Wu3mpg1X*sH>0c+J%AS4UMaWi_^;``d2?p;QqxozY^
z#L?<!T}+|!p@j_f7ze7fPkCuuHz0xnz7j5iRd4ZXz~n;sD_}Z}*bRpWO$)V~*#@|)
z*3cJZK#DMZh1&M9KUv}!v(_&0Ekz)N#lwVWfjN|1SPH*_hxZ)=0+j#HB$mx#@E+@w
zV%V@n_nb6R1_U&WR-5$)v4J$>(4?3DdWm2<$mPpzx_@{7vp4(F>(KL#Eu<|lv!jub
zpy<N~rOEqPXf$S%Y~kI=?|fqTv0zBMGTbxe;N4No{?DLY(ZoL9Ek&;r6(iY<6ZWfT
zVs^DCpaw+TW)f<@K>WpTb{qIiAJJ?_@@$FdwUZ_YQqJ8!OJ~8^78?A-s%?^BT&vEt
z4+$l@07D>1_M3(hDXSUq36sIpLDK-iySNiW3pNzxODDTVZp9R%TKDEs=KQo<Z1#Tw
zyC8>B5U|BV<t;R<`Ut!*!}2!qZJq@dUMHWp-gO6_4aODAE?nx~9BpNH>RU8zV|ZSi
z0UrGQ(3+}qbgS9*rJQIAf)p>4K{Vy2tgvsxTNZM9xxbqIe|P<Vb-QTsXP2PBoqoK}
zpOW>)_)k3Zx=xwN@JPl?<=+@TJ2q8DlP{KsEVtL$$pQ0IUfFnt0<lUWXZ6~kz{Uw|
zqBvJROgrGrV?OX?{z%K~>|niqSn{srVgMg>`$eAEP$}TmobI#nl>@r9D=yVyq67PJ
zS36eF@qBn^xsf2RWY8UoXq!$pxiJPx3;}4cex6^EScTP(+Od#efx&OP$<EUWwcU!>
zIO~J<#H8P<CeGr=5LgSyWo9}zJC<1yj2dKy?9;A^?LPv*!|^GNMRSv$y3_m8mi0T)
z6RyiwWM}61L|lAqZoy(&fhq42<Jg<S4GsM$cZJtIZ#Ymui;pZ1xcBMam(r~wfq%^d
zJZaBtGgSOH#p}GmSnJ%o#V;WtqmAsgAxd;)3_V}9>kb}Fg#gQ0-SyPVT~PwcZBvQC
zCM~gO*GFoWPPYKnP9uIer3zXC!T~iB;K5Kpz+}&I;E(f$p-YS(c*ozRuDxzcbWOxS
zP3nBUZyS($Nt1!iRSMQ7hUL|fn5;d2^UjxaX^?!pKH5I+O-kP@g9;P^%wT%JB)Byt
z*LCg$+V@HqT6}2WBlf1`#%fl@1@<T!0|MSj%fo@j*)PI|4u1~o?tZ#tY5Xd61$R{4
z@a7Ca1t)!01f~^7HJxIcc%HeoeAnnhLOnfLPF+WsTY4KGP@MH%XU<i6RlqZ*e}z>|
z%0Xegt(`lU%U2JXeh;%fkfPCR_M1=RWXOb{kO%V!O*1>>K#n`aR~_$j!R!^>H4Rcm
zLRj^SUSl&(g2DFFrpV8WFt%NGo2pWy-GT~%rYC;05r(<BXs)}bK-+5?HlxsvcE%t&
ztx5TKw#EwFP-=YTmn=~;*cn%0?lz6j=9VZ7KlhKY9ezp1F{I7ko@wx6NT*)fV!mi&
zWV<56MPHosR|n&F%KM+M!pWH3)nSPo#!owkAIYq3WWEe;5cq51f7HZ4yQp(AldE3k
z5qZbSE<>HG(dYO`<T<ZNI|FO7Jlh>qcb9)j<}%x@*)zGZ1&g$D8^QHkyVIaz>1BYZ
zq;uHDXn7tlsr0EL=yXri6QvK_SL|~fwatmVdSN>DCLUc0RPFC?oy$R&Equ%99uT>?
zD458n+w?MwB?Z@k^0?Q|dB&rH*22bwr6&&zV9gc2GMCLd;Zl+W{m^hu_`O~R#$@?@
zej@fe0ZZNv5qEid69VFz_!7Mj%-BM>t_me$Iz76q%Ug~kYTA_6WUKqwB(pogm_v$^
zTh^%;dXq14(WyKsHC>=1HTW}XbN=8fH$p_G0VIxqn`NH?hc5FG{W#tqTl~ryBGwDt
zLY6Ac++VMyFgC@wB&89r-<wc&&(v4hG`jNKm1{1QGsK}6^D}(~T@sGSJJhBhfgXvk
zX%ekxOc{=}JBnJ)UCX@gogTa%hbS3m&HY|h!w_Tw^KfBReuO#2HpQ{4f<DgODQYg*
z58)1mBa$>a?HJtlmt+z?>@+6aAAbS_6VEv+Co<|v=<%Zmh8()>;d^lV&`Fntuw)#F
z2j!#7{v}Htlzk2Ig^x?p7S`3cdC)P=eNseF?%Nz6BDbG-(r(BYUi6_->$=J|=Lg0R
zsE0a-*JdY2MnJWO`2|7#&B@%pS3zH3mMI?Ubq8PHPn-)(GLL<bo=HG<HV}Pe;do!+
z)wcS5bK#rjyRXZIVbAZ;rP$>=yQ|qRUpO}iv`aE@*rRp)w}BbOx}5sbsmLj5%h3V@
z|ItD%va&6g^f6HiNt9x$#Q4Je&-t^*hHoaMC=6i{f+rmjiM-_zBq)y_nxc|nxuoD$
zcfD6M-0ksAxkG$J9t5RnIbV7f$7h{=JjOKHtxk4(cxD9bXS8-1oQhKr*tQ%DnSJ#@
zy2meW=2?D5Lms7LBUJdrmd5azX9@egvIra$DWl~EM3L4q4CnR_AO^O&errL_vt%p~
z+6&cb_j}zyL;&(JqC4fqzdIPeTGhWB*8kVn9^r4W;r(c%^4t%P*K<ogP|J4t{bzqU
zyuu()bzGS@|Jo8qgyL3Zh)qyX1ffMi?<(dW9@&3|mVbPc|JLQ)b_%zZPgeUY1Jm=Q
z#0~!`h{69e5#ilty4kMl{E(iiQICn}m<w_l=lnx~TrBM$Ztw3@^8fkziW~!4M5d={
zF;QyfiIW55+Tmdoe1k<s%wBf?ygN%V;f0d&x!jr|NpIs|g4Dy!$x+tqCxFfeX}rPN
zJ&P_2w2ZE^VBAyyw{(q_ieB;GqUAqtiPL8WC?{+Iy6a)Ivkekx-%v((S8A9fv?kj$
z;cS64XSe<2@5&Ed{GDKO37UJfRf}0-g0>!31?Z*LI8VE_dqMp>Slw@zc5}ch!Q5jt
zAT!mJOW)DNB##G;+MC+EPe1(UHT>s{`vs~0{znHV3}8uDR0Ci;T$%>D2mq0H=Yauk
z8EeXrfBa{qgqOqmG;wm%dc<eJtfigGXE~-|LtuyM05r^;(I@0bS|YGmXyZ}_AO5LX
z|L67c&j<Y1=YBJ6>77xtcba5({-1aC7kB9Y^ZP3u=#D{0=9kj{T(>YTX4-u_R1aJ8
zjq`$l*_&^wH4p=Z-mz>19nv2``e)K9)iyu0^e<QQ-<N(<4*hB$Qqar)^=kP!eYP7e
zrg4Wc{D5D;Ko8)`%@vwefbrM#Q;_F9Meu-K07I%S>;m+s8qEPeSpDA40(#)X+EH4h
z2ocZueA;E&6wgf_$mKHbR^thRRAuzvS^wLz#rIie04v=-uE6fk3ZaC312eX(pB0id
z3!hd1IiMOe_pf>R%Y?rW0Bo#Q`af-f;roIHI`zbA(5C(Y^tjY4xdLoJ+VS==kXiqc
z{Uv1$J-h4crL~ddj_5>7RGh0(@<A*czm{!Z_G*da?BP-|pZVI{T2L-&DPKESZkkA@
zC?z#5<t?M-XRPrzSp7e}A|Div^iG_RO6&bUizD)i59nq#wHE_M^NA&3xc=yPPir4k
zRg~@ms);G~c8UdXc$ug6qdD{P2^_6wr!yC56%nbPU?E_EQ8&9CqoQDD-#C+{oIV9Q
zCy(6)U%#F9CXq`K7)L9yS#?BG-;A$5l#iMqFXP@NA4{<`5ul`WgxIoU?D<^5s<!%3
z?Kos{|HuFG`uqFA!ZKf(SDycq1?&+D`+qOc6LLnce_d?Y9!SH+LQ2=}pmg4BmTdvq
z+#R(mlPVj-(BqY?Sc2UiG#9Zoost_bFV?DzTj;$d%(N7&M9qV`vV~R-R*W+suGe`q
zm**kW%KGQ|{Pzm{`x=wqgN>bBG!$$4*ZH<`36rkrI6JcV-3#)}tVDW9SX0S}!wph4
zARgS!Nr=kyR?*rciZOAO1pkgnwM@*#Al3;CJHMYT5Q~8kGX45}{o^?L3C?fHOgj9j
z%_%er`eIykp#O3ul;~rHvT`TJ;jiD0jp>Qnl{y0F+30R@k`M*!mgxsGhKAONvr61i
zOra()2i`r`Z(a4UB}7V~H_h7#e?Eu)W+(4_`|vxPd9x(jHZl)E8P3%?JQ76PF-*Hd
zs`8OGUIU$2Z|gP$5B?<jVFoiYaR2Q>spXUT*hP<^xYCN43m09{2~lkyDHiC$v00A0
z|Gu5|8fQCdPTw1HjIOh;AJ8ICg%4SWE{88T3bs(1UjDLZPc$EFeMKwZ7}8$mV=rJ~
zms<MP9OD9?Atk~loaNzE^d@=}wR%D~6PG*0Dj#mTr+BhSFEb?^K0U%+604urzArW3
z7;*MmN6O5+Lov&b+IndX?*?{a8>4pORQWI<#VxGjx?;PA2<Z*U_Kwa^Jb+NwWrR7{
zbweJOGA(kRhhFo7besoNce6Y449(#?Ctal=FUzAoqwWD}Jk<cWj5`I=Icq@P154pf
z)lM#n>)PK@<Kt_LP}U~Ww25624!btS{U3nd@hK7Pa~&{XaK@WHVM}6$1U2Okuk$?!
zkL3hc#n{V`WA;gCNq1DO>i9yg+5Njk3jY!Tgc+&9al**-%2?%h(dLS?lZFE07Y-!4
zE}u?8^I;XJQnK(|p!RC+J#ni<yP~7@yX;dvPqHTgpC^QB%JXz{3*hA}NS4xm(JfGQ
zI{*z1t;TJjCS-@6MGzrv75|1Ff>?lJ7AL!X$B&?`(#ymC3FJ(Eok6LE-@f((Er=EM
zHm8lTqAlR#wbMk{f`vt4<2?!7NgiBhh%ZHJolF6~<`xh=zj_4l*~)_lMW=tX0Im|N
ze`5ja1mX0`z`9y%G*dQ?#$X7{P%om@Li@Z0*ktU$4>+N}8SPpPFr_K9FYAu3X|m(f
zu1p(**W-hr7xy^_A<pQ==_KGUFX+;rx)#exPOBZ0a+-vmwhNulSvSmho>rkHIJQ6x
zM6^Q*p131zq!Nm{cY)-ByDQDx^sPXg2e5j|CvFbZxS&B*Tp;yA$1yhI6vX(Wb|CbY
zwmw-=@Ob5;-xN@&R8@;_@u%pchpV8?E<C-$`Ic>%^Gc&8ZJKZX01s5>SwT}z19BeM
zxr^ybCD2vqeM5~+&S#g=Fp6$|<yX#MPi+7yGK1(MbPLQXJAi|z67RH&xu7!SVRgrs
zgKe*in$%Dr_3}e>NMf?l5345jg>F?x0r#zr_L=iT9_JqJ(Mm9bJzn&j+wu++H2EN-
zHO||BUTtb+em??e<n^MD(ttGJEfF#41>8N`4Ss?&(*cAut52ILjw2dBti4XPDu8VC
zTPpg=ayQAUqKOZw49$s|d7tFS{Q8&OfhKFXQ~dc`)5oi51lt)X-n?+{jUHW5{}W6f
z`|K29HZYU^+|^2HK+iW_5a$Kp(LC{Npy~AjEn>=r7NExPgYwNrkrfyNvD|GY8)P%?
zx_zOm51fX#zzEQxuaKuT`fARdm1S2uKfHy_kGlR#0=RM;b0fg#@6t~gi$R}tn6@OI
zC_#ypqfx=ADftx``U|giTqjQO#!-3si_;$^Mg);yFxmSl-$9bl&)~KY#`CV}31F=Z
z08v4RZOWT?Y}Vqq6A1&}ES{V61g!$++ZX4o{ELxdF9la?7%W26bR1PvC5diIfCU=T
z$K$%}Mr7zp%+L~IK}cJS^KK}}VvxGnSdo_KxkL!GSQ?lcJ(iH7uoW)_&tDP^=@`Lc
z(*S9N)or`(=v{t=Ow=nN>ayovSq@;F_c*sgn*6U2(VpzZ;K*4%qf2^FgC4BihIjZY
z8#mJ_#@lqw1eWedRm=6mZm8KMk-(#ZdeZX{>=t8%LDg3lfbh`h=vAU&3G>_{NpVWH
zIa4DVrarIJ&58&BaPa9l(MpEirvsF>?Qplljn(zbyJJT2KVL-~3;^{ovIJo^j>}nI
z(=Kgt_~Z#nTxC`XOQ9=3=Ool5)x+t+G}><9qlh03(e&{f`zMYE<EE)BQ$qD(XEb#&
zhAz7DN<<0XeWT`s;^$~b*whz9W8y%6Cm~|Ke9~&qY(cYRYHp8gP2BDcfI3^OCm)<L
z=<b^>X>bfgc(tJ|YFFYKP{wE_DscwiPz8|gboc|~!<mFh@{jQ!zmJ-)(Hl!A8qstS
z>Y8-2bBMe)nzz1yew*|!pr4z+XL%56r-+Wn*0A!oQ}M#zMrGP?h7?nN8a-}iKRW=U
ze<-3$XC!}enyh!;9#Dd0z&bp4W;8TA#qNpN%sH;3Xt=x}d3wM9LXgYW(*dmZ(_Q)o
zQ<qd5Ad}bb_hm~liFY7^{kW&wTCOb;RZgvgRsl1D-2JReBCsbd79^|nZws{>vk*!B
zC&Q!Z7s1)#pkyBf7z_S*+&O)f@QxqVefb@!Sun=dAM&vqq4E_F?64QXW2@<aib<HI
zmE@Yu-ZRVBY}BhfL~E3gkWpE1tcx~IS`(0;p?(S-MbnryJrLJG4<AAe-X~?CyRPYg
zEY^oxj<0H$<RIrl$Ee*+uHhfa_K?HK*RL}Mu+X7o5^J_kXp)S?aUWLdtNJ$B=n%uG
z!xzh0_6%STJ#_R3nTD=*T{0jOjX^69aI8lN1_!}mny;By!=HX0hYlR?h-f>4J;vZC
zZeq)#2~<()cz3|KBrD&=w<~xqbGS4lB{$R^Qn=8lWKj3@FN$Kt(|4|xUn%*v{M{ZL
zP^Ul1mFLqV1U0cG-3Z=aYt)<b-()Kayp&r<(dWamy}Dkn@dE1l)`G166_DRN+9Y_n
zh~y+1%q?GLBpf2c95c!0__$lEpLARHgM@6~P_77Z6Kv894rm`DmOYAeAF}gAGUO6n
z!8sG~azyP<DMv-?12*DN2nwlVJ+he&Y{djQa;E7UgtV%Az}d9f{@&z${j1uZZBkF>
z_+iAmCU?&n#9e>qD;d;V-Nls=r$EIb<W1a$sl1qsKjL5N86SnG=!#f#r>p!o&&D3#
z<z)2>F6*%CamsR`_}kcOyIXMmpvA$Kr`m?Hx7-?~r!X`H@&yL*BUm>^X_3volfY>X
zKkfd+M6sBzzdrm2@v*&kEIsCR_5GYhi{s~gcN<j)sZE;RX-^q&-~4pv=vFT*@!;qL
zu6IsY`@m{a)?bAcZkJ#NcgJjm5K}Jn!d@S45UjRRE`=Y|X5i9LKvi4eY-GK#Ylp-v
zAYMhJsbGr{XuTZ%8WoR6P2HrU<y*OOvM7R&A|MW-h@F?yot*X+O1Zjr`(?dd_#+Fj
z9gapwEY13(^k18Mii#%6y>SF&>)WK%@fo26$kx^<Y~i9o%`kKHyTEEpo%nG7_2+U)
z-v*_cvcF0~9~AbEk0#0x?`8hM5r?6;psv=dPXRPqZ!73*L#2>em?8#zfge+CS3oJD
z;x{(<?c5o>9FJHx!3eGlNvPt#Z8U#$+2|jT5B*&>a^d*^f=wGM?z>iwTtRjm?+$Ec
zJM}lB!ea4UA$mTdg$QTRD!qdv!P|%Hib*DS-5oUD=AzOp@0s0yx8sFXrl23Gked~2
zzj!so8ki7oKlRlg#(N8my<YLuOI=XEGd;sFTN!40Qw@Kvi6c`khyoP!0ke$=5xe6>
zQSTOoH^DY*#tG7%bA>E>1ejjXH+P+(J+iKJamFSILr=j24TDVq=^JzFr3g1-3OI@(
zKx0nZ4$h&jKBUi4kc8_iBwj19P*!<Jj55%%1)0GEGs7rXm)%y9GjZBJmBH-1<6!Jf
zen<L;ubym{es|YtPFpWE`|d8hM&qjji_C8;TLR?2=*%CtZ0Ra26Y$EoWlogoeqk$r
zrH{hq_*vlb!?uns?Yi09GZ+GAsGqp)@8n=*Tx0748i1W%Nh-Z=`>HQ=mbPs{wEQgj
zUW)t~E$qv`U=a?Y=%#a%=nL3yO7n-1OTJg2ywv-`Hf<pnAfhvp3%C@LTn1%E`yC1K
z{%>6nj3_)o<VWmV@O^!c)&1{^tO&(>i|^kWC4py8MIoa(CfkOS3TT+4221z@YO-F?
zWK5zKDO-{j9q8P<8bD`48DSkQe$e+Jm>UKPHpC63^PXDoUWkeo7vA`F&k?l}yy`Wt
zI+MZlaeXZqY5gIpv_`&^n#&w;sVEs3Ik)MdJvmz3L!A9)3zAwx=lffwHkpuONkOsV
zFhpLP0f8-Wb&$FF6&yNH?E5?uuV}rz3$0^S)pmVNa*L_X+pf;tfwAP+S{{eK-OmDV
zM7(fk^oFH;mb2y{&ybzv?`I+Y(Gh#-<jpZU7B=vpMU;zi$T^IHNcLNWUcSRxPA`<_
z`N6fNTH1#d6m<PNbe@PUg6sAVC9J=z#s7Th4+aJJXfUpx#_xFc>K;aI<j>Y8zBhA+
zL3-#k@JyCvNx~bDg}nSTOYB%TKjG)1Zr**Aiyyud<X`CicBuxYJ;Ft|P~eC`8)AbQ
z;(^UT@KkLNOhwP!XR2;0hRo*|ViFQh;`4z?N&t>{_QBp906t~HkK%U1IXNl6vSMJu
zXjd(jP*;1^3#g~PNq8)ObPINOAx#5@l-SS+TOc#C+W?96gS)lq*U#0hV6{bbPTW>t
z0fnQ5#EV59ztOr95A3a{Lt+|$pci&d(2j)Cc+QK0S7;uJWe>Wqa+d14oKOs1U>q8e
z`|_!L`K6w7)Y+7S=+Y4#lDSXP^sgG8N)e;%T(=%bv$<cSlses0{0_p^db2x7PHnfF
z&fWKVNI*BtGhm^V68aMY<WbGOqzMd0k{1d5^=kgE%)p0_M1*LZl(@S1cg@@{fLTaR
z0&S<aZ;cWFFx?kDlNfqeP&;fIMWz|%qIT^BYp2l7`nuLE6$QS```QSqKCYI9eDtH8
z#WR12&iO)M_zSDDlX^huu5OHQ7eY)_MBMSLLmdxIBSA&M)o01UwEM(TwkLSH?Sr~B
zNtO6cZm&OY@8%wC&j_vks5x>_oHW$X?7GXG$ikOfOi(_EsV#K}H6Xcj0}ZIM80_@T
zqT9He;9=4m43Ef3aR8A4A+pCn_bcephw430BuO*};vEr7UmO&8Nq|(N(E2<4fZ)5~
zshOf>j|0tXb2?<A?%mQqc}kef$B`L0=s#L;O<I(wc6WnPVWT}BT9@bml+?UXRXaXf
z2zOpxv&`5u(3|6<Ahs;|{OLXi<-ZoDjOfC&;8Iup?||6O@Qou4R8y+0a#Nr{I&Vb*
zRRA@1?LtJmtPAwjCTI;2=<Rrcpn0bN&Uth9?l8a4dx;(J{C~PO%E`j;Wgr&#E}2zg
zxA^Ez#``>puPz(NG=m)SNz>p<_2sxFuN+74C>e)!X((D=tKXDkc18i^>h0Dd?A-aS
z`qQnwdFI>;W$P}!+)~TRxwoEXTXx_PK0s<21l)NbG4u6jA%9?2c~qGaX&C6b2!)L5
zd3ngVfpsX=WM>wH%F>eiiU@qOni=PwEcU#1MAKXXdZv}9qQ#jUGx1@=w*m0=ZTXcp
ze=8P%C$gyU>uFNh%b{K0)#R{7+KMQi06CT@v{rAjP<woBRP!@uzK}zW?@Fun{D0oU
ze>OcBEDJ-!MR_f;dcR_y&u5y#iPt$2fS{z^0zcvsA%G>TKO;7f{@~4kbA{OpAV!M&
zppZn;TPgCdW*9|AE!2L;w^D**UUQY&L1T+w?5Gr)3KzEkWT;8F4C>7w_1Dftt;8XP
zp|K_eA?ro2y_dxNwatJ#MgCqiRmK(bU?726hUWxAJ(5G+FpS%syzr8GuWc$4;(L!l
zM_rSBtIzX;?Atlg-RrVX82e&t1aaY|yt_H7axrq|EXQcit0yIBr;Pi2UO<(j?gF^t
zXf1lp2Q!{z=2S^d0ck$3y*Z5!2|sTrL;J`vo}7<KwFFC41GEBJP1KOW1|<m7kL02U
z{M>W}>fR4#pA+1Fgz7;KuY`QV&w1a-Y9Tgerv)3HA7A#=CUT~T9<gpXE2j(Fwb3}V
zuGxH)+79aGFXb<=8H=pO^0as^C}>{E@&m9|JS)1CyA%(&qqxn{!6aV*Gr26;xQCvr
zGflxQ3R#a0a>`BICbcXRv&$Kjvo|7*d&umQJPyaRY0M@>CKM*_{*9XYoZt*0-XzpK
zr!6-O@P=V}xkF!c?heQhH|k|_J+Ch4{d-OPXQK$4(t`bA%51yz;x}-UjZJl^|Iz~#
z`(Y`r%Va`6z2f{{pqSEd0836%r^e~PHv`P<l*J$EFIj^``e^>+YZy(><D1m)hZXeo
z*>9|y0|ANv5*_Mq*rGM~Bp6I6&^BJiW_HJ}k4pOj+6$74Z$KImAsI$~hq?<F+Hkyl
zx6ir%qlm6}2@p^5&k8BGuHAjhipL;>P7j;-#MRW#e~o;!Kk38;u25(#m4+<<g|H%R
zw+=@26z`}+zO}I&d)axJ*>8^5B6?&5@iGC{K&6&L=M<{<!dg&0iXZ#n9B2d)Yh?X3
z9Fi3d7(EmRApW|{X-P{3Va-vE?3yGyj=Tb()84!c?2C?d-GO^@T0RY?ynPJgykeKP
z@tf=rk&Pyf$swW^Pr!*Wf9*=+j=OYc4a0Y8mZUsLjk${5u+0T-4T_;){g5$+j0Ld1
z--Ywe2L-u#fI-pc?-;&Up<6_IzSOy5<Sx2J*RDya11=8Jf#7TNcSZg820nN5E`$yM
zl?_eaPCDp0iu#d){cH2jD#`$T4XK?Ve<>#10OUBu)q8GwF!h}R0EbWqNOleYeIjzl
z?P(7;>Q{d^jy{@T44>=XBEW+PPDR{h4&Ksu;>~U}VSQKp6qRBk;r-MqI}`Q}uyQJf
zs5HJul4t*GREkDNrD)>XVwT?+F9l}#l$~sY1ug4vk?v@TsbT}yec}}LaFT}Ux;Jrm
z`ZJWrfMXU)_8g|W7-+6Vd$UDn6SEC~irafKy3I4*m#>H-P{iv_a}<*RZle+<g*jlS
z7D@|~`b%Ml;;$Hx-4qfx2#Q7PeJ*AO7>cX-Qv$8+#t2IEK!Z|l5=e$Ue>UUiX9~5L
zq6rSl&x1PE`71Ncn~yV=0t7UO>AP>eCO{c?Vu?mY3YEtg*aUSoRRN^_+ICRzm)Qu_
zud}|hc(&vROLQeJRe(45rVT9vTanggA5}_9nvbW|#Wci>Svkan$vzcrw*h+o{#jf-
zXhWlb(g3c{md;bQ?WJ&BPq}`y5Q5-C%(u|XA*})_SLf5MN>J-yJVLFKVVlywKz*!b
zo;o_<NB0MatS9wbx|<Y+1L;XBS}JnaY#F8sX~>+xfvkXzamUrF3M+Xa&3g&n^=K0M
zJtqMrmu()`Ua}ChjkIngWQ`sJ3Cq1Aub6Q7cb9Vz`}ye`K5euN^~uaJYd!Mi&Vz9T
z^=-&)2BD9*m&pqDK0V`WcS$|>0&_aqA_!@**DZ+Ke~^g()Q>(2()!mB7J6Hqe;*MJ
z!Wax2Hsw=UoCASqRhwj-=p-DcB>As30FRwW*y#gVzd|pjpz2b%K_fsJHvz-C+c&Ai
z_@-9v!)-4AyB}<N6R3K0Y#OgNGCQ841$~H~-$~{)x9Ekbvmg@!s7>KJh);e;ob0Kb
z28MI)?(P1{z2Z_<h{aATpcU4eYsyC?&p*%Q6Qc!~O!JMhhds|ea>9dpTV#4$v;+3Q
z_Jl=zhfO$`jAzY4X4pf_x4@cV%HZ^TSfw~MMZ1~;4lO+=nTt|GB9CJ?!rrAGju!YK
z=%hKB==3TR4~@gL>yFuW!6gL+tYk~Y+<oZzQ1{l5<5nPNjz@qS8oELoNcIZ7sB=3&
z{5&>=_nn-*PG!bWDUcIhNpVNWc(JA^Yefx@?t&U+xIGRyM@~R3GLj!4fO3*zh-37L
zwd^{7iTX<S*96YFWX|`67vL{NMdge<haJGtKIYn)8uZ)|P?9tW7)*%mDRx%XQVDH?
z7FPvAJjaBZ1>m-(u;C>6%4*=}&A+8H_CiZVcrI0H!uz#$ZBJ+gqjpaD=Xfgj?yj%a
zLIAW<(Ni?DekIET@|t&jZnkp#gn!<&G_BI+_L5B=4&uyu_O4ZB<0g=SXsjt*s%R?k
zJYYToU^{X~K7go*(o<|tE6!G+Av!vI=6Pt`!<r3_3IG{1DAs_yH8jRyFAkPsr2bz`
z-9PmlC%Vm_@KnY15A8A(6F-xk+8RDx4aix{h8E)wO);9GQB?pmAC+psbuS@~WO=6Z
z<1;3H)%alNc}jV0Dd+lIRowFlu205RY6?3q7%FQY-v^yd%F@}FooZS{exG#y6yH7y
zA|KVj#7>CK&9o;alH1);dIVpNX1r#`yDs-aKR@=w%B=b%7I-rA6L281IGMlGiy-Sv
zVLyIk$V4Y;aI^BIN|DdMM^KM|d~|l&^VE`+iz{b|{`g(s=<iJNA8l)Jhg?KxLO?cG
zEXiLN0JLfg=3PaE-1c8t)7#7iko4kf9m-l(QCZ8ds0$f3eS<ykxmWb^6(<G&K5fvY
zW&>p8N~+hntEucOHRHL09zu;dNkWvW_jmCxLn98TrsrDUci~EWfA7jHIq#Wu+ls{~
zS+p`%Q45XdvksDmYv$+Wx%qE^Azkg&Q+(pNshzpzA3$$q8RV8$tS2FQB!@-sLd$fa
zyCqG+NUDfZgJV2%r1~>2(6?x!r=QF<3zBjH$!rw>?kmn|IbHOVZP4KuYv5mRn+1K_
z<*#Dd|GI5(QE?|TUKbwJ{h-`OMWujNW!zoROz?7r=+2U0l+E>m+H0-w5<Z3OZ9Uu8
zJ9A2tw}fXwoZW|ywMDz3^Edi03!Y)+kd&KHy~3$GvT8-Bhv7>l<{DiU1BlQ2F@#E5
zDqR`r$60x+ujoJT|DN0fuV6@Y7UXR9L7Vew4PiXfV`uXu4K^SE+EIT6r+6PraR-fY
z-qh}8wV8U4A3)}B#p?^Ymm{dq@&?EvM{0pn{s%Ck!rjf>yg|;@5k|(@1}f8TbU)y{
zf?%5v2>vr()T{!u+7|7fFp)XU8o}k?0l@K1&@2kDS$V<4yOx!Hy?x@Z1^#Duq2HK~
z3A&%o8JCY>_^aW02fJ{n7qqK%;<@x!Rs9us#pM~3Y<wRB{^^63k190%NtpOBv<m+g
z(1(ubqiTAQtG5$)Cm7%a6?~le+TbKEWddv8@|_kB(eGd7b4wOoYS}@*OhzSw_*i#o
z%T(x8WBel#gWL{;+XLjd`QUQ50CW0RV@)sZ2(V}!(bf34IiC;G29W2rz)9?cPUDCV
z&@vS~KohhYO<_L+a>>~)PZ&!+Q%_ymrG6<fZEWME#!SklFU*^l_b-qO7dLuoRJ+9f
z$v^Kadeb|r_c8{dqNve=xFoRe@+seou(>!dGt_}>ZwVj|fl(G{tus%uoG7AS_@xgy
zPF7wvLA>u@bPHid`W?_Vbb-II#q~0~kB_n}9gsBp@(xu<9K>y(d(2Y!4Z{l5={6v0
zhLP;txeB%a_82qYYo`(G2hjNZp=qACgcc5~08Es)s<ffU=Hx-Upr*42bLd5Yj`!|Y
zR}3_Fbw{cV%&sre{#$!G5#64yLB+WJIc)y<rvKN^J_@2}4E9-n*XMtEZ2w8U|NDUX
z&-aDf(MzfQp-`soAKdt#Fv0)&`C~Mi{~mq2wD@;d{(qnV{IkCMuh0EOXpE-_u9*FI
z&*JYh>Hp7%6*@E)N_RES;_t2dpPT!iZu%eJcd$TXc^e3owZB1<4@uvEs{M*R$c;Ij
zR+1jc3i3NpbE(g(GLq*ClT-2aJY)Po54ovCAA|Y7y)vKM;HPH_zmNO#ZiF9Ufb*;_
zBrOXc=Ve)fX-bvNRPN&-fnTe2qlbSzm)>*O4aohmq!>4No_(#Oi?dmz4N?>Jr(}mv
zZ%#bA=X3aV(luK8%@)v&sFtoESLiHHF*gAOhKm<H2c2x2s*EcI9kNr7E$%Jf3q1bU
z3tUEX(Dq<?c7INiO-X&=DdM2=MZuA?o6S$Lc`c`43v{xQoMaWXejr!Bq&kTO-f{%S
z7+z;A1n+m8L4OZD&7PNU<ghx}+8)P1m&@GrfR!Ii`=EX4+*iVrA|8}X?lj)_!jzs%
zcpZVMf`?!6K3qWi&G4qc;JpaI)7-KF0Fjf(EDum9tpxEKjyA*F)t9gG0=u~8ON!gq
zNS-yRi}SM|m$4|6c5;lr)UCCE+;<uDe;R!CYtom~iATaH9&ML@e!us%^zXd#-<N{$
z6R>7-0zc&ceY9pt>j%@rc76g%wS2?obX^SyGDfZDALbT;xyKUdag!c_+V6COiY5W<
z>Qi8pvWaH)R~H8<;a@fwJbAA3e%NN<AEcfRyXI>WD7&@nFV%n1cFMH)(SlL;To(y=
zj5;E<2m5s$AIG({)68j*0kjycOFjkM6Sw+qfG~k>ju$EQRJ{T2MW9XJc85bM(H6v1
zx64M2C2x}tzY>9`o|4{Ok3z_%PK1AmAXnmX=&zmp@U}OX@ZT_%f9^R)^p3sts#^Td
zhX2TM8G!a8ZvZvu`K_@<Q*>(28s0^N6yrJu&Ce?IO!ALc+_HibH}%`@8nO@#Yyc2<
zF*PT2VAzAtS1h-R`ejcux9E@mL)ljbWSMSjj|d7FGzf?&AxJ6$(jXEN5`uIK(k0!3
zpfu7AN=TP<OG`_)NFyyJ&9@$%nLX$1J+sgE&tZ6ZpS<t2u60EhCHei0AF!e#)+WqB
zwA?0F`430B$6=ZZ*`aLW@dgpj*8RR=cB+s=|6fT=GJ&u%&VBl%`R8$(!P|7~$%OSz
zZNlr#j8|t!(AD4~LwO}%M1C-;zHv|M<8MDe+dg}9&sfeo$NnzH$@_**y0F-5a))ck
z@ki<$+d*B+&#nH?tJP;tgxJT~rd<B3e=&HBg7D-t*VuH;-$y@t4yiuL{1+F7yMUz^
zXmu2B1%?nWDH+;dPnW$aD|>aUqQ(fyv#hzAGI+&D%wLilX0VYJ&r^=KPY$1W@w}eN
zr)KmF*fR8a#4A`n>r%#l5heCbLXjyC^9|>Xtk-&FqN;P033hGn&T<JiR%bEXHu2j$
zZWng2NW2St<tRPMNz_Jbfo>CE)kwhMRTVMzT>YVhcPvL)f+;7FG?w;5tRGEYRwfTr
zILXx>ZUNU|&2eXG?&M&owt~!d_(mL)$pbFTYn*G1?Yi);Z|m?!j7eHC7I?Q1<eJ}F
z{cgOJgEyt|-89Z+Lz@1jMQ(2ifI75t(-&-*_q-gX_#RZ}SZj6-!~}6j=VCcLi?zO?
zUdNw8e>Kh2n4pci??D{nfRDxGUhxgn8=V$DElv9>YH&I!kwvoP9P?PBWFLc^U%4`_
zt)CsKwXekIcyqeZvUB^Ap8BN=xdAPtx#r)9X&$vM3-V$ugy=mOFSB=eW@_y0`9IjD
z|7n|fgfJd&GO6kPVm!Xx$3RqW2ntHCYYph3HNulQL<mP<Be;Sz8$X~@ECYr5=S$ak
zh9T>qc9X)R9<8Kd_7Gv=FhSPQJ+pC+*Ja+zl{U-sgMf|NeLQo%3{?G@vlaGNTaQ&x
z-4Oc+Cm8g&;QzG5EeBRf6?ywGQ~7KNA6dhzOIARS@FrrUT5qwe-Y)})aU4bpkG(%g
zC-Rwrr9AW1*YV(W)E5sG)(PJpvIDG@)*k1mTK5P)coSmEE6BN5i2od082gro4-TaW
zKB^$~y@Nzi4TjL~rBC?y_P(CAYbMsWuaoFb;6<zyCLnfaE~hOAyua4~(Xv4d6e@vH
zJwc$THnqG9%Bdxhz^mUB)GUP)O(Np{Mf)KNQL`C%vIap4uk#eqCD13qJJt^3#>zn<
zl6AeFUhT@*J`gI#5f0Az$)5xqAuBMn9f$C;S%ig?*j1KJZJW`)>5o*fuD2Gt`A^4@
z!uPwg)nv*HF(=SZRI|xjmoqu%iDw-=M}UE6_mlSp1b|cxaneF4qtX%@3KJ^gMdKgd
z^xj<sbJR`KYpXLLew!gw(mGQLG$^hJTtwWZ9?7KMR>Ic02k^_>zW-n-*60C{P&)lS
zd^$g!hiJ=~iQ_wnj$8g@O^w6RYk*2W;|Z;zbUgZxfT2qaT1brC8n|HEJZ%ZIk>>gC
znOBIe_pJdGY_Ty3K_zYHL6L%sRFPJ0M>FITV0bVWbhxv;@Cu83?Qj84ilgpFcfUlV
z1q7R=5OLwXVc!CDerHHV1mVrZ0<5xd+pw2v@)ZNbTbOwfC|DpZ45eiGuJ6MSMRuCb
z-@Ul!!sMK<HhLKZ??yW|hYW#)d>v|JYX}jI;Qro-Hvu4mluTk5;56j#PX620b{8+d
zmHUq(8QcA`_*aN+^R2L3UpqC?;U(d+HrR$GT^Oh_=8fNxcnRT~B;Jx$3qhtV)wF3h
z>HMxiZkqwr_Cb;#38Q~&(=?_?|3z3qkfT~8N|hyxS9nN)>uU{o<33Ei?qvlpOObS1
zKq&>@oxP$r*YG0tj-vK%vRp9hf%ofb^kOcUGt)@$Nm^ic!i|h#+XnrYdG#k$tnZ7*
zHOg#$l2r43qDpwm!Z+1uF^GQbAxxkdHg~1*%IO*(a!Nch!nOg=hfi99&$kyf7C+(B
zaQ@!(3DsE^XoN1&NJw)q23;~Tn~8y!==cO^xOmY@)GRlobqCj+k6G)y^M0fpBU~U!
zhAYt={Z@Cg>=USWoPbB>FUf!xolei99aH<-r-}m4O}PIZt3FDELBpkc+don|AQXq@
z)PKVn_7t9;t5zL2@eqy_?335a6!d{78HtycuEa2K(I`{2|1zT*h1g4G{ODeC0F+bs
z_Oxr*Yvmd7M`st1WArV}N;6xKrJ43Q=yp578+rb&hBqZpCgD%g)J@;bIZ4tDK)0Hh
zv&zgV7&ZBB8f0Dx40Yg5-kF~PQa}Lcx!5!Vw{gN^eP$o`nWAvHCF}<3_cqREOid@;
zWcsmcKc3g?%4<W>sISv0=%hV8_Q})iB${0|f8A}dW${qbgfLpEY~}8?X+Pj4k>Av1
z-geu5ux(Rx@217>H}-<0b*!)OsE>)lY8l7OjrZbkws-Awi(FulWl{Kca#3)<dG4JX
z@0hs{aE~4-k-GqB^?ikKx3I*nKku`fS4}%=d$5m}i}D+h%~D94-m8jvQU2L>VPp`W
zC57z2l%lZ3D)GJfz4ttWV}OG4ESPp_vbL6p?FPy1Wvd<_Ma)_4LjCC&+zXj9kHh*9
zMvDrNDXQ#^j`qF_FL2lcP?3ZrHfU&RMQn(7MQ)`FZ-`z^IV+1%@6T8=Ou^!(Dl9t*
zN*j&0PYE%>IpqGJ^AV7CHXK-&6>o?$U%oFNivK*ii8`^%sF8OEeGU_2Z&E79Qdcj!
ziXSoKHNJ|Ne7ZPXcf9|?12Gh@BKNpkcv}&BJ*tUT(++V_Od5+3=36KCjDsnjeviZD
z^geVkoaQPR{!%a8I=xQEdO?+%g1AtM&=e@CHGEM|ngwITDPT$G?WT>SNV-AZ$`es3
zZM=U9J@WUYp{nV;RE;@V8j}opW?R8Uk;8|TSM0ps0-3bR0GWt*e14vkzQ?wL7{NJ%
zGi8vo%tC0AiO@v<e%T(u_q+A$K!Ng4DP1&zy_qsQRm#)&8<9O9c1lcB)z{V1BJy`w
zPVjg9x}I<uv^yU`sOR=KQGcRdY2xd2H#NQEJ9XYkGaUyjk3Vv`jPo2f;UHj<U%3{i
zpx)Ot!j(azODMt4Nk2l{#j%U;(1ck;$h-6Wxw;+^N{Ts3uax%69I@uSr~l>FFD@cI
z+#+@Q_V0c9$5Zs{Hw7<xas1}4#?}?FyX8TnT2g3YA#_%$<iT~1)C427Y7c(ZY32h+
z#*s63r)FyOF$5jG39%BhB(RpaF?%TI1+CxUi(Hpc0_oa?DZ15uN)54QMdPdYCFPmh
zXOB5^*)A#12#gE6z=*rlFR4C+d~jvj2K7D`MP$lO+x5K$x-Sq6?>p^7KFl)sWoZEH
z)Ny|79x7!&xcU-o8y`^VNoYD<FbGee%*|UMp5wweNumwna^DeM!X!MV;}l9MN<|OP
zwA{fwS#lBC`gC4U0kzmt)FrLckDs|@JX?PkSb)^midI}zG3Nv+JiJ$t5H<~A)y)Ux
zG-&I>?zrc{&J_*=Q+r$;PJwSitbliiXxTxG2-VHdR2kh*3IE0oax}z$1Nd=UX6rC#
zIFrOEszD*ZP~k8Qr4TAM*?;MY1w5d_#<<S)J4o*yDE`|W911T0gv&A+w4R*z6NFSa
z-_n(rHi^&O&~IT&9(|K`%cNU#L~~>Krq~rM)f;Q=FnhLV5QxEZkuDd%CjzX+*UfOB
z1K8Z_0m4*Qut->GQM1ZSjo)4M!SJYcRElVqR9!f03Ch!(KaHO@!z|s}Qoy^F+D7&H
zIM#jR-f4pRD?-X*r@GDz+<pQpkgzD8K*yE4Lwt!yV&g~XS7zh*kD25i2A=J6qZMC+
zK1HYcXZ`Ck+Yt09Dw!R{n%8`nkl$h;Lzqe?lqq0(51vkFwc^~Sx56%b6WDl-T^2X9
zpSq4j-YGIY^l?+2r$j$d+GSr;3XI!)$i&>qvOS0Dwyh}E?)sf$BrWKzWPu5(okOYJ
z{f&E{BV<}0>oQ7t)T$WV%KoFb{|l<{uf9`2pX&D6=n@~)zwF6UAG0SpW0mXxt4*h)
zK?)B9Us46Dvv`bcI*q3kSM&+{;(XG^L8l{ohk+(iNe5#Hv>W;nJt`L^e#7G_<8wLS
zD9m3d0h)WiF;GaY?Rv#%I8}S|7o9cshmk-Bj0fXLpu>5bu54<ci8ysd3sdPwyc<VG
zV@SAeDafxl3agO9EiuVDg)Hi8-6dj{#_06s=+yF}>x>5jB%*{7bGUNGB)!ua)Pabd
z-N=V7vuPMp%gTzHI7r#Sh?el^7JXs;f@IXJ@@qWco>U^PSI*T)efH9BDZ<f6BBBDc
zpF@n0R>ybI#uXvqBj~|R_D9uBYa_{`SfwDPselfJZ%w^ObNMa_)8(D6i0z&3CmZI=
zt1!NcW87Np&{1J3HCyJozaPcX7aiV3*k@jEs4qx=x-uG4LeLOI!q$cq@FOgfHC^%-
zO=v(7?=bsP-%r0aqRqB;uZejaMjA`D<CdcO6au2WF2J#jSAJ#{#%W5?{gkasD5ndx
zrz@LM0k4WmlF^1j&BQ|;af12*^8WLAAvqXR&Zm|zh6|6A*n%ap3esn~8aujB-2z={
zIq*igH$`|^??2zz-G33vcwcklsXoxZf3!&R?dt}+jn*A)b+1d(enZ*~w+gmV?t{*k
zm$^Q{6jc%NBrt5HU5>O<mD?5Jy73>t!N0ccWMK$USoAf_<oD(DA#VB;$!$<OjI2VB
zldIpI5ZHQZG!A!Q{xb9hgfB}{!EMYr4$<eqQP@2rYrSxYG=dustP9{ID#1>TdrdFm
zb=4w}TNI3fAAnp%H>J2a7-$TCY<J)ZhJA9wBxkn6;~h89!^gqwaJ@;&Q&^zC+R0A8
zz0&cHV(K!d{$N2B2dG`I(UTUzCL}~vmk2hQ7^7UCCf*KNme0^S{qj|C>5GfOpv-yr
z!>}0)1$J}>wHsdT*ZkH)Zu%9@U|HA|rbX{m&C2KEl)0_F9|6}4d0B(MuG8S6pJ4MV
z;_~6*2D7@5c<_(fsu>sA;T|~@{nxfI=fOvfz4kxrX4W%jA};7<v>3%>G#ov5N3ybt
zBjw3*zfik$)@zP9<u{s*hSWIBH87w#UIUZ4-S(@6v+tDm{Gj5$7VS*|P4o#t^l%R|
zMjP(OfwB?}%t%%*YL{hgWhQCUc&v!w&I;!&<&^C?L!I}&5}Pq-Er^*KPCHr)nH~>h
z5sAFM=4dd^ZIj5+!J?(Ba9ytZe78zsk41dMcBj6WCf=o=?0F%Fxo=DAy`^B$p648u
z`teh?)`6aja&O$i8uSL+J6{;0M4sL6Vm>xfc$(Eb5Fs)3N~j=Z9TGHgF3XiNXpR#{
zY_ImFN;I<k2#-_7-!`R`ICp-CWHsIg<JhI6UbF1cfGXoRRW>@b|1hBa3+TL<g1T3)
z`OE(pVU1y+_tV`J+}4RWrYM}oA9?P|h?_1<)rZ(HZ{jpEmav-W4jvwn%V;=$qh{WF
zu>Ap{3_WYj5mvY%pn74hEDzO<zXuEWo=)C^eM`))knZtA@Nr|DSZUokQhWDdDCrV-
zo^X`ty-fz#%RxJgaX}S4-7;YHw?wU*gh|Yms!=^ZbbV;0KCON!<{@|R7Jz^#9!;qd
zYa?MCTH+M7z9mVO`7ssnn&j)q)DwzlrGKvxrKfP@yxXyvwN46;7vcIT*sY}MaSup)
z)F$TlKQ4V~5bsEAhrYQITKg$tJHSa4EYEd{KB8Q}-?!9?pm(uvj6vp5tqx1uu$~e9
zEZsT}!Ry_(es<!E9>LKZ6uElx|B1!;SE%AMDpWeg9)j|OzozP3%x7+HG?3Q^!0?bT
z@FoA)^FR!T>>-4&udy9TNZ8I_a`Y6UvF?eMjGLCH`yLtN99c>;e7j+iKc^_?#o<}U
zg-3YO+Z(0GXR_b+DMEPLkQ3yS7aZzwBN&Z|wRyk68pgawWA{+?2<!Xx{rY;@aehXu
z8@ak=D$`#_vl$+6O^>NG(oa=5Zq?b`A8&FU*Y@g;AZ5%v>*@X-0#lS`$x+=90i#ek
zEGaKFRy?p@@q03($K-bm!Rl>ZcgK|(VH}kh04(Nv4(r-V15Mw{>lg1xfu%)K6QOE{
zs*pr;9UGEg5w?dtuVuLjAg2_;yYK!XiNIN6cQ{lF!s6NZXaKYwF;n_alxkvUzlmA7
zKNwa}EyYL_jfXs8<|JUv4Mq|Wa`*MNR$~kNVapsD5rA6@5KLntcv91o52@S5yV)9S
z&4bQovMZF60qgN(e#~pVDmRFTrFLkh414MfR7HtGOa9&Va@{hcWg+jN<$?{-Yi$$f
z-8L-&tc|#YH+^AD#mi556Gvziu}3beo;+7K`IEi{SEBWZVlO#pHi+e8FS$+hLI|TN
zL6Z6?s}hy^wN*09bA<0EKN{(7To{|J4sB5$i_vuk^Phh#e%+rTjHtr!^vz$hAy5=w
zXBQ0fP<jDOrC4SXRs)$Sj=fsIZ!#!u2;AAP0giy9`)^4Y(+~3T8Xth#!;%$-!pu>E
zLKZ;IRBS3$0UK5tm-}%)jdrF8PU^gFGvxTnoH2eV=I80zN=d%)faeuDZkAD<w(JGB
zlQzZ%AKF(%Xlo`o=bv{1QB`ABGgtCMBMyh29`WYY$#YX8oKw;~42L`OFS(B4UWfqU
zN*Hk~+hg(e`+$8U+vQy>NTz*Z28oMs1VrnC7{*oX9<5tO5C9zj=B0O;J>M}E=-#%8
zsSvMbUO<9e$ODa)G4Jw_dutVxwd{DO-h31zk1GtWX5RSkMk)LlaJA7c)=Uu}Lx_74
zu#>n(Arm@zZKCq^*PIVzplowve|3=<w;HMF2yZ*UeRC#eC}_m(0f<ZX`Kd@c9+eT)
zDLwF{U1O}-n?`_nOopj_V0H?D?8e_*6z4}sq>&o9{qiLC@NC{ErdXNT9|8M9OyyJ;
z;W%Hn)YRwYd|$p4Acp-SO6ZaYY09nHrBa@rxHP6rt7u8;YpqtqcHW&E3(74?H>RP<
zZET?a0WYUsYYSA*N{grIqL#8Q*hdqfKNH>;+z>rZ;T7y4x(cYUQ!@!8ifBaMd$n(`
zZNY*G8d$>36Xz>3U?j5cjVDq5Zso!04<Iz+9|0XLD_kw2<o)BS|Hzd5YkAO#%mJU>
z5}SMQi(1!*qjl@Ckfh{_;sIsHY-?!y=KV!*T}cySa1rO3=ujC!LPeWUu3GUbd`3de
z<^7ZlaTMwI9+Y|Q-Ap*iRj+ispmgI4u<r<Z3G0CqU^!*29V^O*O=lSd6!$c2X6QhF
z5sDlLQ)oK1-EXJlr7}u_#`0T;!H33Rns(iKBZ8xWSgh;~>Sb1iJr7k#BH?|yM5oYe
z;cZvX*57xX3In!+JKjc8Ec6i!daW#jKgm(U_E1SUo{{n9H}%`YFjgAJ>%s1B0g!Ih
z>rvR<7NoBpO_kN?riIAV+jXi#l$X4n;N@3mevY|=>rPk?1P5<sgsrW4&f8r1fFu~F
z-LD=62UkL)7M?!pRs&Bi99%Ds;JI?KyE>1u8*Z31IBI`7kJ2sEQg1U?GJtuq>58;$
zNg|s(Anr_B(bm4?)>vS8`36}xu6E}BG8E%ttUH4~*ga5AYd&8;bM~BPpQ~SLrTsyA
zR?&IJa#)m<H{E6KCGaijSfKkh{9=R|>}^b;xn;YagBI@U(;Rb{;RIrf^@(V?TW95z
zY);pUd$cgDlKq$ffZ~GT1w-jE$ZiWnWnTVJ#HL3%fbowj2|y+k;KE8&_~rVa1E9}!
z=usa#M?ylo7>?v6%YEL(VZ_HN=VV=314-rg@6lfLNS#~S|5&2@XGak@f3Ul{EXgYK
z{a{ZFFPH<*z6>N&*VoiPbl=v6D5|b(7J{y5j*q70a}iw((ga76!_$%savbahfp;wx
zr#NatsmY(ke?oMIT?~|*cz2tUZJsH2MqjBnt=<_FKAC#4iG<7TMN!y?PdvUkd*|DS
z#vT%e2L~pty7~$eampW_mk60{FjaPjU===_la!pDd4D<TMPu}E7kTWHyZQ<=WHybB
z^j?RtU%U9v-<-F+sjzcW!c&BF9pJ^OmZwpTsolAhFnO;|r7kZfK2pQ6HM&?pU+|GW
zij4l?b{vLDobp|(n<|e2DAx7+vqL%YBKLWssc&magGti2tx`yhig+YyG`+N)k`NvB
zsqpp%^BR+fXWedkT9-IKH|hwGp_#_Gdp|h&OVpK7MBrZ<bTe>sm29p8+Ns8uisSbr
z+`A%NY@eC-L|-X186LD3DV(R|kt-4{lBk1ApqdcCsJFK4xDjNIuw3;A#b-41ei%v{
zD$X!TfY&lI3E&U<&hag;V^p)wkKNCerKtVphBX+_%T76$li_{lD%@L=l!1=54zlg+
z-*=*z++aG8S|a~RC%gTcw~Ha13y<c&1@8h^A?i|Y?TMq?)sr6crgru{#nHjOFF53@
zIm(hb_{ID!!%3<7*<VWDH^tKb$JX^PKZK$g=ofFX!_j_!1`SwIya_vSqmUR0mtJhr
z7`+c767PHE+<z0KC`6rmpawZ}T?{Ty827TUO=IKfX!eW?Q5QwxUkP9zxd45lE23ye
zM{l&U%%HsQ<KkOlVKtFhB2V{{S*roCU~hfSS*MwtBK_^ZG5|gWK~uC}ATRmnQg}+K
z)gH&ikbxM2d})bjw=PM^+_S#^lNeRfm@9oJq6mUj9n2l_NG`Yqym^FiLh=ruOO!+e
zJmirnhJHISx9Z@q>f0i{-s<9yUH9M4ENq^RaAO(V^4&I%d?AEqtbkAH5v8v6NWb%i
zKANzikfN;EG=(<~9^E<A!EUzmglEe{)%r3~3GgYOp*0emy@HJ|D`Xi=Pz;L3QxYdf
z$9btmr}d6%>G-?Wb~aCT4*3WU68iYRJVsUaiJ?A_dptV=?Cn_DWRWNSR_G^m@3U^>
zqhP=I4fXnOXV3+H`?r6}AmdwV%9KC<XA){4@2*nvs-;V>LLk5kJCRrEUwniCwmAas
zvr5pwcKdm-%P`z{QBY9M)QZ<(Q`c&{N3@cW|CqgS6tnDej<fM6lWTHys6~uxz`%@|
zcSr34m`qdU8qpkLum<zb24_g+yG0oUv+=%Kx6^)`@&p43ZQ+~aFmM`YF9(v4#e2H?
z6x4L8|8&g%I+LYqaP&5L(%;elJh6UN6x;~$vMs1axX!s6Kk?Db1s=06_D26$SA!+G
zz}xF6z7L0U)wv4|dMIeD2f5PNqz$14y^EWi3*-T`VlnN1{?4|<-#ltJ*C`o1|8-G6
zg?`+eYwv_<`(stBS6M)Bo+vUMV~=a%gzN6^d-FD|ly~*J;x3Z2gIyR<E))HoS^nmZ
z{C;-n&Ja6CSBv%i-|rdn8%#&Gm;dkr_|L!b>!(X+;0o=$4YkDi^9uckr{G^#^e+cR
zhn#5J`U9T7e|rAo=irPT1%9j3<XzNN`^z=@zx@^WHkub-^EMB$vHo%-|F?g>y@oBY
zp@_vrOmyRKj_c2d{D1$!x0@J{t#QQHc=n&~{r~<PT;uffsMtpXFQuA=|NN-TH&l=7
zw-0_!Q57>z`U&{y!uf6WX*QtyE{nYcj6Mx-?j2tNbz(f5rXWYQJqyBrxH->p1vVn_
zu<reFQncq$Kmlf}!#qN9M?6_1psg;O&h*WvuOQ11r>7ZPe);xhmz}!-?6@RB&dl?j
zvHJ5}@y4U2=H}qgA8B4J)Ruu&t<`u>$;Z`xj8pJ>;74SlJvajLk;aMKsH(E!w3_V7
zDZAA5{8yJqNsfKRirH>b`A0BM$@{o^t}a~thhx3|h6)}xvECHMKQ0C?O}^0pofXWs
zV!`xd)F&6FiDq3esBTP@BfBI8A&vgmc)6%I5YA&Kl-eBtmCKR^(Py#icgD;ee1o3e
z)^`7sEg)hkhTvRPR}u)ORcgT1o(D66dqnEivUNKW85V#i#(;XGnN7CUHLU)v+LCWA
zs(KS<)CMKDJ9G$FI)~C9aykWUR>>s}O+dpO1Cey}19xKRficZ;_WWhZIfCbpRiEmd
z9PK~*_~J}Pfl`Jmetkt;`(U_DX&muW(!X0ke{&avXNNxQ%G@6}C|wrCchd@(s<8;k
zmj|Y>^H7kpLJq4$z^KehLvsMBK9owvV5FrJ!)81mL(Fax^$iSt3LYs)R(o9_iUAtr
zWVTv)?+C-0mN-tU4(LHO`>GtaJdVpikre;cywMwnjcHGgUn+s8eEOL7iX*XtJv92A
z9np6?fiKhcx#H*%ohUz5WB6BQ5@klWytZUr7Y#cQ`aZksa<DBgYA}K5D)AE1ST1m6
z<lOlVaf_tY^orR^Q`tH4{FnmjgTn2W)pW*MSI+<aOLGpqX0y(G;z9ZSiE!6N2d5?W
z_f`vC1T=f9B_`2^BVZ1TO;;3-`2F}4`4SvWz(ieYFJm#V_||SFy{r7<Z5Ht6w8iP5
zx|TC5ALwem__>Zjvt}GpJ<f;w58Bs}FX!GtkUdVIg!vI3g||ng!^f5y=~{L&A1mwJ
z`}_Nn0*5EzOSSsPdMC3AZ|W27X(K0r48d}vyV)OKE&GzARj9K!Npno0-gVkFA!7Vs
z;!juqreazYCs)yl0M)Jb2GbMTAF2|?P6~=g>FT~s{P8=P_O|_hzp%-uBVR+LJnHpp
zf9U_sUDI{H$+T{fd>~)@LEqc^XYIbNe=BPyF}^a<$IO)Be~Nbht8aSX=wq_@7HZSe
zbVu<361L}}iD|Z*h7C^+YTfz=K|Z2eb;+^jG$Y3u)vbqJUYdU-?J37M&JT*ByrMsj
zw_M2PvizCTj=)&=ZgP%A<w+IxlvqEZ<qrnlnfym5#3O;3Gs}l8nTe)XJrnabYqA4-
z@J$u{KLjvpmR@<Mw#{iI(cQE4JT;&B3;U$P9fpSP%v=0b)s7|2xw$)XxM$~Dr2`b#
zL%Z#~R`^{RmPf5*luzYccJkJjGs`l`x)n?p)x>t351A<2`L&(4C2OCh+Ye*8NsCgS
zN|J%Hj*4ctk1*8omRUkexCBM_*b3(-4fDza@#86%&s!ne4~~xQWyYo)rq;iuFDxgx
z_TZPaOm6KSE}Mw*s;<)SyKXxeDX#9_!Fh%~l+flBQ9+(VdhMPz|IFxPjT_6krOG?v
zI}c3l`Da5P_Y*}DAJB}Rr(-6ciSjaif|p>PUhdx~==EIIssF9XcN3Ot*lp-TJ0%XW
z$4cK#zq#JXcz?99+s3}1ZG0To&)gQ8Sd-YWpZn&RhDU?<{QkOICiCTkxw{Dixl6Cc
zc6RiFx(cZWwsx~5_Z&_V0>2PW{J1rqNE6;uvAjBi{`W6S4IXSxlcK+!JU}*P0b#s7
zAOf0#J90)C7Lc6TUX$5IDOWodJHW!x2?FJ3UtE5u7Xq*u1^F8>K)mi=dug8O!MhJ#
zW2ZmyT{Uh*d$tb{T_+a*kru)vE(G}liRrshaB~?BY6EoiUS9QXjIq>M_R`bu;Fa)o
zvxspPNaNo|ufU3fM0e|qZIpbPhuH6!f~PH9v5L8Bh9?`Z`J-Tr-#vmjb;rVyurEZ^
z00?&40VWhix<i2Bn-{rcFCj+N?=@>(p8-)`5|4>N`9q^WEX^+1=6&x2D)}K#o9O-u
zoMUx4h>j)Gq~be((k^F<j2q^`LML_vd|0Z#DF=C>OV^i}fFgSyWIA_8n{7c>GY|Eq
z9mUHM)3F#}K*sPNf0H(J6}p7o4%@*O#C|DoE=kz015gO_5@^*+sh5B!Z5G377&1tX
zxaF4v87MwoU{x&rw(civnh*XZd6sN;Yvb}I2MDMst=<xkPZ@>;ea9b9?XuReby=lS
zDT^e^$+=1bDo_Qh?AmvT+_`-UQoQja8L01hI{-&~)~}DSv!nT3Dwd^vSb*7jmFD3S
z&Vbb>KceOS&Q!=;`MC=PWWZ>k$#kLuy0%!vARrFVlUzwA?I6z^7l@cx5wy_(Ah!vJ
zt@+N0qfr2Q9zkvhOLMqW+3^AZhIe7ScRMCoT!o%Bh_$O|Cv(1f{?wO%Ua}>StP^w;
z0rOQMgSa{LxNjw6**h83EBdbj$g-xfO`c<g=>1r^dyGfEgQB~Fujl9pj2Ha}=`liz
z`7V-i^^b$&&>HQ#5NWZ4;KtEftvZyU$R)+Z{!=jSZY{4#H+gJan*dLMc`TUdt((Ok
z!u)5#eiTTy9WbLYhAKk83j$O6%VZPQ=ES|SE5#1p?WeOTZpoSyUtXTb_>n%i3KU$r
znPb(7WwUpJ<&T?%Y;r5v=0rJjE4@zC@e)W=v+=}o6}4kIUjM6x{@?YXyC?cEmcpM6
z0A1J%0zp&--Dy_^0X!v3jeO_*&9^WdanDsPHX=~6l+hACzc%XojD-g@v!^7iz`auf
zy3|V;n9+VNCg?=vUdeyiehL#?B2W}@6YqpsBH9xDVSWh|dR9z%1%CRI=Os$o%38@&
zXoxZyA@?_yypyE&x9AS*gbmEl0<8ZljOi&bRf-I7l+LJW@$;QmxQ!DB?{LAY(tX*t
zn+o~3O7FBSz%YaJ@}?o+bC<Gj-zo#g@$o|Evv8{~WDB(LKCsP2?5p;H`DC3BrBK-x
zvfmdQHrdwUh+f5Eh4FX3AXUpqY3QnfunOiDzS4cU5{e9G`VPqv9W-Be0)0+f1&b>e
zVQRVKchG~Gf<iBJF<a*5lY7pa(Dj_NJ>(slRM7p#5eLoy?gVY93gdIm1s>5KF^%g7
zJYcc74b(nf6w+@k0|R9rJQ|8vPU?MeCZa;km&obsw@N=;gOx?3rQY{;VT8|@e2UKg
zy$D_sauWR~sVPd$ylL-`@+n68$5lYqHm>w}uS8J(-nr}-v<8zk)iaYY*Sn`VV5g8h
z*D)B^J5F@3(>%bipzr8&nG6{{m9>k0(_5xVU)z^YvJ{%i$%biV>mW%1o7H%*1CpAZ
z*1rO3dpUcfi&u3^^yz?asm`*obF0Qk@`B-tQjeMznzNZpNWkVg%uZ};RNk2x1tM-;
z?gUvC6*0-1a;uHUxj;A^UZ9CKL@RP%!IPjD!p#0{C`SSz&Z$0N6`e~Yxc#1&U5r0O
z=`cr$dw^!jwN*dxX2GtpB!af)*6y5eZhw9-5yO~3LaAfyl@XYs)ZTo7x$1U+;kfXP
zWiwn?14dUUHpeTaXOw0H*0ri7c20V@AJhV%c{g8nv{y2B&1T5=cuVDOwQQ=t&yt#*
zaa+)HRXe}^%H|)bf&Zr`Ul*PB@i~k?swyr4foX4zYDY9OE^7-EUy#gj=l2}O=}$Uu
z1>jvD_BJF==t|`0M{rLRQf2y?gn--gLKkt@^dxZ4ijPwSgG5^$#YekdjN%4Hv59q^
zh3%4BQV`;o79RXrhT^S7%w7G!Eis*YKEBhaZn)(RcM<~ZBW|dCkoEKEis$AgHMRb5
zvEZjcrH~M=+_$rpy>JZ6n^a68>8%%5OsH5E*1_5C3j5t%C^~%11>zdz>Ti;oe$uZ@
zHYvoxidVO+rz;3Ekf<6sUHV}W619v_COsHQ(HLKbwuxW7>8{NomO{~-;J!AJ{y?hn
zIZ#qP{hskMi_j98X3O-5-eO4Z-W_85*7DQqkC4iC^XYw;>o~#9-a;Ad2g!FIq3~nq
zn9hos8!ge^NNY1zd+3c%Muh9;8xHowSEB~IYj8Tx`#ht56HZGTr`yL^I*wapGSWMO
zZOX5vu)r>!RHzSLyHfrJ$;MoDVa)o;)?7D|8ryre5thwhi$VEPRYblD4_hSbLcJKf
zaZb;1`d4N(CNX|r^}~kU#j4K*>Pw?}cc{-~G$!L}zV3sSa=!Tm>a0$N4ezA<VuflZ
zd1h^o2q_w_S9jDM%7%`SLjAPCRgn1Ko2g;RG@f9z?(`>7T#S-AUQ)6%?hSgbY$tJC
zd2=-K*V7_fS@!B5*2!t8bLj&~i+^uU^+W~8t*E6!(H~M&>V!6OYimbNyzdjPXdg@<
zk{CASI9r*cXL3h)Ec7}B)5t1T{M!aIQ4+hcIWeK+bQfILMvN2u0SO&ypZyQiK7%WR
zzDyJ`NocP^QQf9#g2gy)g(RX%Y1dL!gpyhI;5OLUXryll*;H+a2j@+PpuTZ*jkDhZ
z!MoiL{2pe-5Chq#!X8-$Ar_D8TP$!jT{eaaT|%zN+m6E=l#hO);L6m)qB=og^3;5P
zzA??A6mkvwB=5i}-O@`oE6LM0b}A{7g+qys;_at>dQ>xvF@~t9!qRs^BqcGoEF>sf
zpy*kwrc;ANe#yS0I6p70u|abtX(EABKeorV?Y?~ydJ3*K9j=Cb&UVS@YHs|VtYDtW
zEqM7=4lj_6rDh*b-?Ype95%s`SsmPDLU}hB9fN&DaCo^;HKvlbr00it**cr~Y@Vl!
ztZZf5lc5UL#Teoe<pr-mjDd_mo6?)dw2?NY{k6vZL{su@T*0%B1SLLJe{{b80V|+Q
z07PYaL>BMYt1Y8|x_w8Hya|ZBO&%*D8-hJ`R7=58kAfqp_LM8EGuq`+IE0nkBB~Z2
zjMeH%AnHt%=(5An6Ge(9Xr{T=Z0XElR3$ZwPqP;)Bc>C;@%>Te87GQKDr-_UBY6()
zH(#a4#zp!-BC}Y+dud^rcz|&)^wVwC-YshOifLbZ7T@%|#>XDG!spv|LidEmQba<B
zA8yZ;?u0Fhs8X~qK%WdWgbmsVWK>AUJS29fKG?v?%a@Oe=T?N?_Q*-a&Et)FJX0r9
zsZ*dfsJRK1>Rsb=wpXZ2EFtNn9+W$bX<w2PUcQiq)NQ=xU|u-2GHfN>(7&7A^T}?i
zUSk8?7tM)ou{h^tZav3Nc=#!cleQp<f<YWRdq+-|xj3Tj2M9_i(&)jHX=Rk0Eg#;s
zw$W3t#jGf>1>0Pa&WDwQPO0xv-I}Tzs6>X8K65h-Ic7QGTDC{RYLRHjx0F;dI3YOH
zk*<y0w!FXIwj>wjUw5t8&-kD2S|D=Q%49G6e%EY~yXKK67}EF@YPpL{RU3jt)e=e9
zeX5V7J#M=gbc=38N?f!CrTa62BSd>%hYWgBN07o4PeKUI4R^#KtH+sA;Fh>D@b3E(
z6wN$p4Lqw^pRRfx&`52hIgVl~oyhOV=`sonhrMNvt16QiGo?cYBY~9(8C<?$=~QPv
zf=%dr*VcX-@(NABD}<YX-_CMi;-((zN_C4zHmue;q-1WOq;@K$;QdKZV7LXb8=vcu
z{XGq&uSg8H-1gwFowFBlze+qu!NUpDxN$W+A|I2nVq+qMWh>#QfIGa$bDOhh=u9kI
zjkFqQwatqv=FiO$Y$+i+A}HMDhi%d<t|NwxzAH>V!84Q%V&R*y27+BiBg!(_6lIT!
z<}Z`7Gt4FOXXM(HerD?%n+ZbhK-RB2@Y?2{Ppz>t`N7j$%R=SZm0ySg{{f}OMnyll
zg7)hd+_?nbMg^7Q&Y1vT*4@(;m*HWQ9l*(z4~0f|pCl1tC}+xTlmdHz<GaL@iR+d+
zeuRuHCEM!xTJCsipT0Qvg7NAV>!to%IR@%DM`CX?YWfL7_m5z>mv^16;;}BjPZXkY
zxN=J;7B$?JbOdm<QQ^n^g^t^c!OeGug#UO4G9TDK{&}r4#P?_iW1gyrg7az->o(jx
zoo?q|z&RG?XF%v#NSFs6Wt=p3VoHi4u5W^ev!!wRAg$>WP(JlYm4v0dcn=r0D_5|^
z_Lc`fc>&u{Ny~?x3)Nc$)>%^TR5K!{UnlK|Z4AAC6k9c@meUR{JJlVM5(C9+gcZ-;
z6YDW!lHp})k}aD(I*`2#1&Z5{(BX$L8;aH)IwPW?f4HbJR{^VQvP>rY1);oMLB~J1
zqX;gMFSAzff!r)^fZXQ<uH{g<)dC{x%YM$SVM<1?EzDY<NYQmGbK=d72Okc;nj2(2
z!nm4j+D}lINI_%<gDTEaxgCb*WZc$^*w)K~ou<oK-*52ID`Z>5A>s!ij?n{~j;K53
zF=Ca9{uWd@XPjiy7n8X>Y1qDrhSY}wT8BF7%BmRho065J>bcjUC8ST>6pL3&mJGPH
z!UN3iT{X)KL32~NSIlc%Ql#68OLKyER#>VtA_V_4e#NeSVFs)bzA{o{&3>{^R1f&T
zn#D{qBI|2+cMnpdkq%aEWj+11<*NR<<zh{yiJ7EotW?A{PO|mXFzCfUd|vH4+Qjy^
z@X{G{d{F?oEyvT*el`8H+$dhI9qNF3jcMGDq!mUf2i||73Yyy}iX6@yV6d^_8A9yI
zqCtTKM;X1Klo6$|K7=mIs^pH#fi(?_taXM(CcQ^3j6eqWY2JB6&52j!odig{Jdb4?
z`~?inU0aUo&XZd=fpNM0J;EdQ+R;}m;Cx^~HER1VDQF2hp+ALO3{#qFHdM)KXK6q-
z+bo-wB)+i}z$8p6l?!R0glx+p0MCU7U6!dO#k{3S9`oX<m?uS->`ha_RhI?PjC%lv
zz?6ouI#y~oJ#FGe^;WDcNO0)Hgq?t^L;~l?@OT91nP-LVpWxRnI;I*6SA$w%qD^HZ
zjZNOeKHyMIP^Oz_EGFx4pa6M19l4JSKHqug2>&&hLO%ClE#kl~VEuvHuj{+HYw`Cm
z+HdZ?NL>>^Ui>?*S}ZP*gnruBGLBd@*sN~Z_)CliKE{clO7Zi>mKU8y!>GaMmG(F{
zsZqu_CzeigCz#?+6R(8B=;mX=B-k}8%)m1Tqo_I?^;!>$uXVEir^_SUS|4Bdvjp*M
zdXI?|nori4k#-u8Zc5{{sFa$KnShQ*N^AO9MwxlfZ*a`Ah3w_Y@nIUTwBBr9iJpJN
z57UFE?^Go+d3~s#3-ibmV^G0G-1J|N@Jz|EotuGmgTz~%HKXQW7s#|&RI|PkzzNFR
zlC@1YSIAZS<_m5*tvl`|ufs1b<(Y_eW1LbE;@va7akXp6LCrU!r&x#_(Y1crk<cdA
z58cEh_!u_*nn<1(a8Sc}T9ia3CXFf%MmL}Z)Bz5Xc!vodbgX-)tmm7ggHKq`QGM={
zwR+%j*;F{olHd_tTFyhZ^LldDYOSKh$=*RDqcX7xUi#_j!F=81$=!h(#+6fu5izf{
zO}*IA2b+P+C&cWp+QXlp0CvE&3s_Q+%PW)5Rd1pgBR2}<`FWVhE~p<?pH8t`Yka!;
zWdHGyYk+Zpn9tmNcVZ&@O}3I^#!Vzx)Lv21vzF@U&5PpuFVEG1@&cEesY%sJWO)%p
zR`}pLcdzC`Z(wR}WzR{!%?;!C{yrtLpQ1K)kT+C*L(%u=GqY(IK(3^HxhWUVD64;`
zZLHy+W?3PWFTOH!Miim`dMg2p$1M;mN)75eC^i{k;ee|rH^isuQIY<(n>3SF9_jRv
z>0GH^R+6*Wn;qB%(7^ma@H5Qv1XO#)w32tJ>OY^?r6wCFA^HeZbv7y8`SsWB29@d<
z?)aRd^=}PA?4?jXAhv4uz{`!-5-oYiVU;#PCrj}j&XE<d81evmQ~d}9cf&Vi^`FKG
z?;A|PJx;&m!+<Er-a;=dRkaHy)+w1j?Fh`zq+U_>@G&@^YsYwtdep*vhNqKxpQa$3
zf?S+6nx24SPtIEKdoUyJ1)ox)(S7}g*HfK+bn!3Ijg!}OK(mVU(+tCKbzcGxlJnK{
z+ezs9u{#oY#{%$nvXYFOs7(_YEK`q=*@T>Mza8!+TFs<d6S3o)nnzdqzaz#CJFF%3
z`y<V1Cg_oU<&FHlELS>%L~JE}_2S$#vLhr`uH-6allxLFEJQTRDpAd%xgiEeqgR5z
zN*`x_XP%0c#yv8u)W_Il?Jim^Jb2>r9J<pAhg4^bSB2C&X^n|XXHCbl3exYSIv^zS
z@?F-J0gZ9TWLmSD$s;9uuuhe8IMH|#ymeDD>TBZ@h#$jj9PG9tn?$VHnb~py28-<s
z57|5>EL1TcV8)d`*rz6a1`{I+YTqVr9Lf)TS@=kgFA<$AmV4|gKAI$a;m1jcj3kbJ
z5cgV{R_G&kOF=7Pobb;@TrjZOe`mX$@VSD4d|Tskkc6q13t*Z*0V8@Vm|V96`aK?+
z&>12zm%Q~<&{Te*Le=|g9zK7Y=>+5aI>}_BST<fw&f`J)$#bIrP)9zDfI3pwp_k<M
z{k9n$cEJXWO{)IZolt{yAei|VOH6aQJdBjNpTs9@tqP!px)Y)k^V?V`9Ur($Rn>Ri
zl3MspXg<S{3C6v8)h&2X`hD2y^*)&1V6cp#z{t#QG>Yf4c|J`s;x?@Pnp0aYmfiGm
zx7>D`MaY#$J*`xigZXnzS@|bljrQ+=LZ$wR2T*F5UcM{I+A#!B(xzK3_6xHQw{GYt
zezF=$svxQx>>5UG_qrvL+ASViv%5*RSh#t%i|zFnJQY<`pkUm|LUmkvIuJ@LS^scd
zt}K^7(m-4f(W)1Y%LcoKG2`f$q<9C?g?39bP={dB__Q7=eM$>hRC*YqH_Q}IM6pNT
z6oeQQ1|%MQrom9+jRb41!<DaqElbSNKTng^*hn)L4p@Sb`Sm--eeWNsvp(!(_PM}T
zVnw|ZO0F2QOd{t?drP%H2<CXtlX}b8F9RPm%=F7{*-WtLHDX2H@U(}96P0#$regA?
z?M#)Zo6=VIR}D?btv}yq(oxHN>_G#AdFyeijgW3!0b7_p31{>jq!gzCGAs3ll*W1c
zeb}@puCd!KE7ottn#<uruG49DzRd?9uLxEl%9PlMOd-d05&wEimF@CQ%?LEM@ns4_
zB=nA&YQWPsOnT0L?WkbbrZlX$nZGjZ&O7!@s|iM-pG#ZG8sQdfhuu_z7?TROmcz|o
zQvO4^K_#JeiA_Jb=*=IEW%_Lt1ud<Wi0o#O%{Iew0+J{|Ye*~Sv`jcjT|NzaKbQyn
zmz`ZDyg>HUpFk}NFq+}vpp|)OpYO6`CS(~mJ)M7X_Oivo0iMpm_P1t{cM`?MUYFj3
zvY=VywZxqAo!21L>Mtg^NKEhiQ2jNGU}y23mIkn~Z$<{IAFdPz6^!w6Ic2|TR&q}$
z2<}oNs8oMnfD&wBW5?vQsu0}l8`^=O^g!c$ntlD#@F$CAUon=pF<QfBUuw0Qm6Cd@
zr^2Lbqay8Awx5{keZ>cg>+iQr`1<=3*raSG(<n3=tgtscgAro02&&%~smh4;51es`
z_s!Gf<J1o(vlK%MC*7qMZYh|@lRn1+=DKMFPBENT^JOQc6wJx&O;3T=;in*0xm%OP
z&-Th$rOe`;u@zWx+nTtz?xh#(!&G*WKf<%{Y4<=$0?s3nAoc@IR&F4R)c6OeL{>zy
zw#zr{HkwTunpkrt*W}mdZsZJa=^W?xkOl<2iZbfMbXTjrMd{2ujy0XSaLaJb;DpYQ
zPgOf1n!_@wa`#l}j+n~i!I7G106qO>R%BY^6hUx_gtWXvsVB7nhpSMbSoVBCfOEu7
zP3^m#w9JAGOFb_?vxKqqXJ$1<jGw!Go#T0Hs+Tx?3(9Iisr3xT-(Ae~X6#Fpor<K7
z?kta^kI7||i)A0DcF2yzZ;!5*3$>&q=ryIz$60mVQTV*%H^(zyxu)=%oZK0VNRv&e
zo@{~5Pc(q@ZPjwcYon&i0rL{M@@*BWD=IUqxAY{_0|2iy)m45<d4-f-cX$37Z^X2y
zz3IT&HoXbP?RAooghsA+w$606)tft<f`89D5%>TzQ0n&3!9SkW7XrR7h!T()(eOhp
zDfC-YI!f5lbRA+NxNkmniqu-|PC&KLEOM_ybwly8lpfwu_)F7a5b&X{urEtx8<o1;
zgL29-Hz~s@yYd!3ffI#g^q@LfxJmycc<athvo4#JZ=>4&G3otx`bT;!!qVt;ZKD5U
zfo^;85qPQE2RqC9D}>3-W^6|Iw?3O4E+(ynyE{a^p>YR;#InT@x?EZXXpRH@Fr%eC
zYkA0JbDFRd<moj?^3I`XY@`~*;BMTCNt)o|%%dq-oA=7dzHo~!-*W)>_|~nY*{^jk
z5l%zf=vuyIcBM8++|SwHe}?h|dXY=^C~Wu7OZM>;=ku;V&=3ttf+U#A8&`mpYUrRA
zcJGWZQ6GEBR7y@-)@Z_AwHo%W;?!hMxlwWXu~{h?Uq!z|55;{Oa$V*!Dys;af=lnt
z(f(Gze0-x5V(U~6(~sj-x}faJClqvtV2TkBnKbh-bCMWw1b6Ewq{hu{-|@65RdYI5
z`RCF6+fny30zqj>!EO4|Ul%Pk-E7;{(P&{m!e|JlPAdr|B=I?=@>PN0k?UygDl{9&
zTnRiuX$U$Wjp;8B!8F_oIM^}3)UH@*DlWc^bGFUn1qCj&3c`w7sDJ&_|Km{Hxul`{
zieJ~c`j-|WJqQ&-p^WA`qC=++&XJmE!Lk+_EPW2u=;cT76Ih|xbAietqovenpv^#{
zAI6LCH#yt|sIj#say&KaMgDj;{>L}{tRusbZ=3sZ<J#XpV*h)7_4_DTfcH8Onyy*1
zF#W^(@^2US=hq0#Av5EB7w_}`(8d4zvyL2y3>L!8I3Mrt_y7OVeT2W@N<Mw&Z!iAu
zr}l40-G3H&B<|D9Q2c*?RGtsOi15kR_W${O{^_SXFQBe+{M=ghuj#w^jh|erCE!}o
zrbesesRDPy=FJu4ue|j@?+-7o&E1o||KPmbADo?kzj<Mda6o!vY*)_xI@|O^)a~yF
zunLKDR)Fj<hHX%Z(nh^atjM%@AYGPH?pso^b5yoQ^(gB7Hws0dO%tir=|z}Tqsu+X
zqIU-Taw*AYB{-fWk!v4vB4f`tSZBUyS2ZNfq?CMVlI8wKh<_9z#qA#ccH#FYlR#Ra
zp1}tXlOU`!j<Zh<>^^dHmk??CGsK47vjv<3J0SR14u#(bLo|Ry53#2e8f9J(d3Bl0
z2y~x0VLm!zS&Gbe!Dmfg;|6d#S?}<l9$Ce36uvoViiqcTt04wEwWad#CDDSDNY!ec
z1&8u|cV6stRFtX)``5xPhNjhi$+q%KZ0q^rtDGx;JD)&?6Mo5p*Zz1Q&~Ol<Qb3~<
zl=O7z38?D0M~oU(!#qF*UW4`Rkf2|rq<*G3;I7NzE}MFV^_|WbHmfEeqCMlXU5UyO
zk8ee^EuE0}EO&wgpIrpqDaLiVXtX$(oA!f_x@Q;DASpcw><i0f6LXXb>Id@V`F)?)
zT%;%9?r#4~An746g7=lInVW&@q4Bc%)#BvGc@D3i7<Fs^EtLox0eY?yZ2qw<(0U7i
zSBkn&In)~X^4&zBu#W5mUF;J^k6OL<i0BAL)dSIi_8Aam8%ROXl!WF6qr2Pfk<9mC
zBfg)rnMLo~PuGZhO=F3Nd8$X=OQ4$aE1FwXgMS9Sv77q$<i|9W|KFFDW6Be+V`Zf9
zl1Hsa{(#cdW8rgb&kP8#5R%o@vym@MHVKhAPp_PFXQ6uWb-h&T*H_JyCS+va{h}Q{
zps94>?_TTKo6u6<BVYS{0c#5sz(zH5dO}9ZYM?a;ak*Q4()`5g%T+)+W;z9=Un*Sg
z1fp`jt+jA&1Lg6`sT{X>X7Vm6yU7(*W|I~%n)um9*HI=$rq-Lqk&IrHN2{>=beWCJ
zxP*`yCBKpPRuatK=Ak~lnn+Y~gVZlweF@9Fc8#FEH?n(|!93FK{W_%^nH__?F7}%0
zWP<aCiL=J|9krpcdDC5g7u5>Qhd-%HA0F{7tMF6r_-)RntuI*A(CeOlw|YF3@bc&f
zKYwncpy_;rpiRSSX0Cq5(~=a&FRtMj{N0m%J!46Sd|roz{SIqm4D(l3@rtLmcqF<x
z1BHjN2A+_=Y8PH#*08TS)ZeeD7?jR<-y&pxac@DiMSA#)iC&nd5$?SJveT{ArT}U;
z=C#FzkWsnEE|cdsh5c>ftud`Ok5{*J{Su;0)%MG0R(IGB4l35yGcHZK*6?jyJ`9k5
zlymH;W6!wZLOztX)V)}B*g3sd7_&iXRNI`ddG{zTWMWf{pT8^Fjq-qNv(ai}OHEW$
z+SF)ep+Rt`9?qpyti?oZM>k7)YGL1r%Td*OGVONIsPk(FCyu}0ejfV50-Ex$_wuio
z^zAG3Kgm2tCP3yH0q6^<W7IsWCn3jk9vmxgLzl^dM<c}w3~@Gq40A`o46+k(_bG((
zLSJbgAg@^ByU(H!&=LHIN#}u?z>XMQjeu8~f1@4L0}+OhNBC-3aIEqR*xRoquY*lz
z2gpu4;p-xMYFH4CB%*N|c52)9s71(~2qHHR{?RPV)!QM;0O*exW#x2$tD6?W#6ieX
zy2HUaJ%Ww6f0Cb!z^;33rWmZSju2x!GCx?EqaU%4P-R^@_<97sDA8a7HzAqGH!)J{
zxMS2J&2awm)x{$WH~+y0PZj`^u_mv#$-!Cxf2KQ#(4MKU7a?aMaBkaS1Nn&-5+m7f
zGcNXy;;dyYI5fq1x0KRYHbmWdnu(v|*|Gp|sXh|P2J?U)QLUjgv4G!C1VlpZ!A0nH
z(3hBL(#@~g60`XfA}HtKcx-~M-PU@s5Akc-P|GakI&Cj%Bl_>5pUk7>1%%9sfx#+5
z9V63wRlUl-fLoD#XJw?%z}c?>40fiQ2f#?v6zP(zK(KWsyp<aL0PdPTu=iz?3l3lu
z6$|&U_8H)|Y>3U!x1qd5)$<<)PJjB+mELVW=_@u?v;y>iIG_{APCW*ahVh?|dex4>
z-D73`MH>xg_wkk=X9SmMljO4Re0nc*!c%(=svyrNy{pscUBsEr`Bkpm-zS`2F$Sqy
zvsH@QV&|J91|uy(RuNE(ZW;I+?9h};tJ7Fg+~|Zs_`-ZM!d(S`{wZQhR|wOXSm<8A
z1f(L+8A9e+p6PFf4PtpdYQ7nmGrF_MgM8Jc$bJSFoyo5MNSZ60B^QbAvjXrYMX~}q
zrip`_PQ|W^$2U&5zYIj|Aa43(vCf{dMqzeJn5F^9yOvtpP`?|wd5>Z}f3lfe|L?1%
zwvZRc==VK-uX&(~=!m_?oG<A87D`=r0wOeUTvEvY_CEI9|Dg5o=T9eo#LPmWFGG)E
zHzhHFi&O_nfxA4#6rLY3Sy1SE(Jyl?z7-f!*v4DQZ@A2KYFB%xuH|)=cn8e>f<O9l
zq0o0&un&KtTT>hAnFMxcs4Rb+c9RR{GQN_mjux6*9puO3D_M$bHC=Nj$mHP9S-hRE
z6Jk*ux1f+RH>y@*qTKbo!U3pGXgMmy&*yRs=-+f7DBn=mLJO(`(%hQKFq921p|__h
z;iad-oPE9IlSW{VdVKi9j;HB~%iBrtMj{?Yzh`Kwq%D;XQ5UE>22C#l6A4;V!@m<L
z2rR7mMK-@6FxyvE4pR6;@F~wZlpqro<_FR{Gl)uXSSeUJ{E^a>l998@9ThDvv%q+Q
zcju6y22X#a!ZxmwrO|b>{r54O4Vc_=Gukjdvf70-l@oV@ey++Ki@xtTZUJ(g1UvBr
zLR^a3Z<zAy1zMz^-nWuFVx!5?>K`Z&6dbbC$#SM=y_D@uPuPSgtQYs`^qbP^WJSs5
znj@MAAen)b)%*(&Lc~Jhy~l*urNPv1uQzH?<?Z(j;{cbDEZ3f5g39+XV$T!ONEoM6
z`Z<lz4BZJtD{W+~7Qg~SqEKfrTls#a?dnxfD;8W%drpjplo+b2tZuWf2?EZktS7gc
zsefDBx0z;k+m|SM*gAydx~cYMe`U*`)+p5giE{M)g|7ICvVJA5oqoBqq|x-tg@lYM
z_)1fmGp1-R?gVElxm5Uf%(h%r>P!dO#mi;e${c|>f0jP|xs^&B5=oLe@BI2Q845kh
zWyc?DWJ(2<st>knkifw$jV=4v$Ag7z#k<Qs;U||Rrp<-EXx9etcKuPb{D%sc{){l$
z_4~hf@iOkHh*c#GO9;A{$MsYgJq*r?IqN2`rh1~W_a@@K<^c)!znDfI1oR5J<At)m
zaW5Wul&Xp8;F31cDh9{>sQM~B9SBLYdgyr6A$`0)DVvYHys7)&3kBesE^<Oehe0KT
zk#_!?{YRg7CfkJw8EJ=FKI~Jkh@|A)VJJ{z&I9C{xSR;9!jOI%8TGkPTBbOv%Wxsc
z<y<nH*hgdWMu%zjrjnGqFAzuwkygO&vPF2bUolOUfV(=pXuvl{#-_pR2a5&wsz$C~
z%=QdoTU@AF&um`X=z}}b%Isro+9;@>+la%CtG9CX%^SHnYS)mDT_Pb8R<bBe1-)g3
z!Dsspl+gx$XcQS!Vm|XmF=Z8@$$#uZI>4ChI@++}CG-UFMog-QAzFJN6uxW{Caox0
z(%KBpF`O!$>XBUCPe~cC7xr^zitZQRq}#g?tDS864E@BqR;}67Bp6%8LoOiEL~KTG
zCW)@jrsQCMc5E@Yz{7%+dJv<9ZP#>mIQx;9b0*7>hLh>5+cHV=KId&}D^T57vimLd
zD9yP@_0k&GAd<RuL(#Yfbw*_dLNopR4=stS8eEefhZ@(q8NtF9azGkv7e$V9^&h$%
zBIt7F2FR7p{wn@H#lGO17q3dDm6%F-=<55ML8-8$o`e?l9YK!gic8OCg_7P|^cREP
zvn<^`SjR9j71O;6U6dliM7HZP=PWjnJN;@MCzCIMg!RhsodESc9N+AlvYKiiR6lUt
ze?5W1wpP*}`c}ODp%32*dj#<u)k?^VyKz*5IFnFHr+lDplz00AFK|Yc8ci{E#GBzo
z*sg-u9-Vx8Z8H6#Oq}i8Ya+W_nA#j8t1<BtvOVu9VsEKvvN{wrLW1g*>hKB$R8|ZO
z>W_u0DepIJl1_f;)D)4vyWk_Z8j&mYTZHjZhVS4-3=e9hFz@0w|Bth;fXXu8-W?H?
zP^3OkLFtf20Yy@z5eexQL>dGHL=dEu?h+~K4(S%8B&9n92?;6by8CtXoU{IOX8!ly
zwOF%eox$Vxz3(sfe)bcJP<($?vofe=&l=*%<UUe=;4_t1((G{<baa`-hwPGeo~^8;
zgU?8MaTCsMm+A*EPWJ6zml}Sl_q50zVo53#_<;n!PD_bi*xIW1T-lzz(-y&U8Y?x6
zDg1UW?1EtKk4sPcv@1I*H@ODY@p3=%F5kS&Rla?Y_MFl#rRM+p?I51nmu8F24LAwn
zJcx$TT`JNr4G2agFQ%_d(cD)5$s0mvH@AAj6wZjgK}9CY&>pc-K&@G-THsN_-Vq_@
zcM+d-PdYfA2<!e1Dwt(ZcKXy-BSJ9uYxV;-9PgULfpVXa6c!C7?NQPLnlGs$8o>W4
z=xTMyW-0LSNE;4mijC3&!}MXFM_&=n)*ARQ7IE?xnWybjUsf}RiKH4#y$5by9%jpt
zccknN-v0rtbID3t$^>)^E4}n5cRDos2QRjU)9ey9Ly5tmQZVG0(z!V#hSTz+K3i`t
z-NjGGR0{vFe>-ZeMuH*-GujG&hK4VcB0qC|YLs!Cp&*kgz$8b#>X}u7S1c0DOL$4D
zFT30t&>cf_?WLjtSCF7rc)cOWT+&MB-qKb$usSv{&-Wrt31DK|?Hw3aqOC5at<N?Q
z$hrc@WOw$b;z>azw^}LZ*R)$V6Vag;J4~PYH=gi6577iCWHudN^MT+`$Vli8ddO}R
zQ@3j(yB642S}+N}=u-;gI2i@1dhD%qpeL1!+ydlTJG}lQ`v<*70$@2{-wd^jGtZLk
z56(9LF*j}>Te-8(G(BZ!{=<(wBLj^<3SJ#quSYBahP1K-iOS`J$HHi;MnGaS<{)s=
z7okdZIfvaO>o5~6QLyMXBz+#Gve>=|S&)?fSQKt4AUW#2y*zxwP#)`;CXR(J+;=K3
zif8eNg``fWK)?RrT~gpiS}_MRMeiQ4x|aqlYR4VRv=!zFzDh3Jfm`qyU{}Qve&4fy
zO(kK%pDndUukonX!#6YQ`N~K!PLY2V)wNE*9<b6w-}B^NVcTW0({!kd4^T=$NaF$C
z4Z%vZ@MQ6)uWU2<!jQGwpZp{q&6zsPA);Z@%S6dksbC4Un&0kBdG;62^{?HhkOBgU
ziiX(zdi-4XD7*v9k#1u&S4CLR`v#3sC3sPeb7vzqSi<ML+3*gJM#^DJ_zciLl!nn4
zn)%AxE;iQ1aDi#*AJ?tFO6+_9gSSqsnnUjQFt98l+?MdlSh>9^<3;OCtK_9h^$F8O
z35(nyRz#g(26+3%85Rg{h*TxocP|-;%Iph3=;=(vJl3|pXXEW#*s*DSk+8c{DXbQv
z*mzanKayOG7}x@|f|w>%MyYdCBzOl1-pusc1N({T?|t|(F;fBP2pmN`H75dW$J9w)
z_$BTd`&whYl;xQv+7U46PW0ToYZ~iuw?90D;E*4H8>Q%lq{3GROd2WCM2G9RZOa&<
zyICeHtcKms30$asd>WjUtmJqm9wrL?E>8c7%gfmy@Hd%h+^?y)M3_(pxy_c9vWBM=
z^L2<n%PyHERLkpY(b*FjZ*309C*-;ou_l~=5N6DT<>vvuz(g{FB@WYS<OTBpRT2HL
z2u22L(6E%}Xf*p>3bjALJa!gU(f}7-gut`SqY@$^ohWa9X*rbYHsi)HkcFI-2a?cL
ze=_L-@_o#bM2dObdUs7_Im*i}H*muq>KNWS{{dU96`G!0lXlTj#;?yIU&W%z4)nA+
zvR1am9hSM6F7K5_;|-+=8YK_Yaxzh#B^3$^Zi1rb;j+Uab4H-zcLQ9XLnFAz3D)12
zDP;W#PdrRH_FwoW4S(13M&9-ZU;ln<Q#b&I8Dc2)3pjPXOewe6JO;2%`NF+&4<Z5v
zgjMe#x;qL<7CzgRJs2+2G_D<RPsKm@Fnub+sf6>{nl62-LNc?r!Bs$eRDOWq#J+27
z^w8vB(uq*dTSpQ_6|qLu3Fa6n4<jgKk2XMjmo^L0M|`R-M_aucPbmvLTBWlySmcr<
z(I`dPmd45pM+vwt?=1zo9O~zTq~t6BW#&aDi|Xb`<BF9;cRwboHIVn3(wa&uUtnQc
zP?dtG+VcUxUY7S~z(m^nMG-oqIw6ok!>phJ5$R(W-Ro5@17i4jfy1@CB(%^gvH25A
z|9HOe`Q3eV8V>6cn!1Oqb%;^#B{JJS`8!IZxR84|;jLE4pVoJ%&>TM5qcW}OA6&Mj
z<#6M1`;(bP$ffLR`}%8%5>Oo;wR+gA&J9ug!}jJu^t!m!FVFwkb-AKpNleXPxO967
z`&{h<S$L7eUB($tlA5X#ImNHZfT9{=5-ExWN7CF5^}3_SJ-b&SZ&Q5!8<h7;iz00c
z>cVNjq`)BP6$$;`1xM5H4wPn10(T7%wF=`3W~mkcODkQmyo19D=9g@*$h7v_g%Y4O
zj@Lrl7V~l=M4HumHevhgNozQhI~Z*->_PxVV&IcOgth}iLNK<cx*D#6sEn4_%yZJr
z(NU+^SwEF5fDaTx%BL30t=UN2Dn6gm)CJ1Vcq_Kt^Zk*k!U79?tk-m$H3246YZF;Q
zF={e5USLCezVV5^w|&j67PLcYvyOE-eLCK>oR5DzAeTT-DNo^fm*nB@6O_TC+DH|R
z7{A-+0YT~h#f#=T!dHsBVK?f6*3t%_BFWr?M`A54NA;t_F`qiyb1hf`B}xm$X3Vw$
zi*FFC_qXC8b~PO-dvLILp+jG^(C6*^>~*@tWv8%OaOn7ua|B1dHpLx#+f>AEo=Znh
zsbEubUVWZsZmB25SK}tcq*N4LugNB9^(Pk;SgO3UGzJNsq(-KDixQH4OsnFcY;i}t
z!R1`MeCNuXZ%4{$;NO~U1F$?gk_k#u$H;ZugK*KBRXzVDg%hzK?qT*(^mTvF;t_Jd
zhn1my<jVDHWiZ0ZxB(p;ed-XPFPb80`9a{H2<ZBlH+B`?U2&`uq8wRa!FVDbV@S`$
zMSznMgE00xiMdcgvmnr@8>nu1*L@qD2r4DMejWW`zHsYdJvzM|V3=}i&|2_X#gBxv
ztb*5{b=NpRRBj-MdzDpQLKNKUJ4SODCFtTD{fE3cgGmUwEWvb3K>h{e^RHMpP|_{&
z7&ye}+C^622%KE-AhgD>L^+&YdRu~w=sqL$cLHq5<!v|}(yVr&8+Hfo$ww?<MI?IQ
z0{BS##(P99<`K&BafIj1OJ)O$=lDHsKS|P>EE<I<=DGQAe5@uh=-eA@n%f8|r>nmV
zbKduDlm4v?nf}Nj@oD1lokUTc6PcC!zr6tN%Y%Xe*T*Od5XOyT)t@{S8s&|dJ2MQ-
zodBc3JGtG#hIPy-P|TGlA|N2i++AqcC8WVFJ(X_sVo`+yxpLh7xkp-Nr9C29j5Qt<
z+y<tv)<HC3w9fQCL{_6CIKuR25V+AMVQmFAC9akN;C%*(w^<ySZ{9K}Du8K0QLJ#@
z(c$I`g=MmSn_^{?t1u=}zJEv`F!FpG5%-zK-8#^D78f}h<#k;#OwgY#NA|Kte=Ha+
z3+ynPQ|54K=quC$w1QzjlGD+sqOY+O?&0+?VuIT{%T`BUVe_7z>eg!)d*C0piPnV4
z0yloGpMaN+EBCN0?3OA%`_@FCX7~q{RX><DHv3=ec+k=4L<ktqgg&fCgW#^pTc2w2
zxs)AV&+_VD`v;BlW5gIhZG!LoFY@>GQ=%6_5`?C4PagcBhzEF5B{HwY$;T^&=AnCD
z62o0*oJQ~(<_hwXLL>3sCki%SDIT=jQo4_SOf#_7)g6&SP$Lp5!ig|aW@$QyKvOs-
zt{1MNh4l__APV|(ScBo14WqO1@0P&M<%ZHS6waR*s43NyuIUtbB(iV;v1>MI*{C~F
z)F(DBDd&1)?`XRIm~sn2rbN_qTS<xhAf&|W&~GSQ*`v#aPhx8(!n7LN5X>kDJ{|G1
z0Y3SuPTmsJAs$bCn$oZD9U`peZokBfa}QJuz6Of$e4#ldS5(uG^o54<UfI18D^zaM
zCa{1~cCA_{UndWzIeZC4x0~4hPnM2Js;oBiJ5$jmBuk7!-U4Cih}3{_a69mH#0zP?
z`B6u*Gg3B6Gno(!;wr-aUUU$8SMMBwj~3LIWA!JF*EnvkPG4rSi)$`clF$yZQS7P)
zTGCi_a&azHb{3l6o+XXdU}NSUOf8<e-_x~eO-d;5>=VGa9^<qN5hIDEgtY_Dlpaa>
zljiJQ?p2TE5(rOQ&|55ByST|*&Ed10=|J(WXwitz9{4@x{u8M(q=ONObCN`4DB=Td
zDIQoh(6H(0RFkt6$ZHpimJaCo7Xa=xTrHj?e<$mGti5XgDhHGk*Lx1vSEp%C#QHAI
zUQzD9H!JXauJsq-&~=0||J7AfLEPZq%|ki`8h1rh!8EI3f&Upp;3)~|Hwt~<2ctZ8
z?&z2TVI2KNk!yKBnKJrKK3RUKcPSf617dG)An}|9_z`~o`-a&cJ`H)vH(_V{jidDr
zdAa3Qp+cih<KChG@n53gwd{mXLk!|gVsC7=Rl8qxclm_+jW4omj>T%63P!BX{QS>_
zR~QOvzJ>NEH8<z;N513T88L$w@4J)yP~7`%Lo+~o^z+9yS{{gK5)1NSNxC=85A0<9
zq?D9auUIoYjd-(UVSNm$;~v#_K?T#*eYU$_fmigzxdX_BPGAEgw(XPUfBXnBYSR~G
z{W5)NzUG2~`?KAmCtrm18-tp#B@jN)Okr=pqBv;|3k%z*C+hKKQY+S#5)?}`7$??S
zNmk<9;IA;DfT}6$hq{^^9x<tWF08e%dxwnt#Ly_$Dm?R%tl1-O{iNPYOzAj=&2d1H
zN6lR%?O^ZEMd<?d>dD|2$1l!!HU%p0@vFziy36N!;J~$nGAX@H@jX>dQH{)ai&bKI
z`W|f!cStnp>Gy47mYm8(mo*-caW-@A0Jkg!b?wcW^OwIe&NYd?+vGEJcNohK;4w*}
zyi8i~d1&FOg|mr(sd=xaQXEQ}x9yPc!cL_{SX<QEF6Zi_1L)xVP|W+wCW0pBCH0R=
z=Zs?epA9~*g+ijTKrd-GWq_dTx>?wq-&{8&A3p+S|6?<BK7EXVfV_coq*rb%sgqcw
zMQtWGg*z<|<h(6#6YPa~LD;qD_Se+zJYviXBAj@r)QhxMNee#@G8ka;<6b-U*B0tG
zT+NNp0nRV-{ZbzN;c7xZ)kq*6!OX#J9Y(l1yzs!7s>bA|+u1M?AvZEC{%ZCoFsAJS
zHcpQ_VRnoRbMO?kQD!|R4L=>HAmSNp1BwgYg2ZvD(t0n&5~Sx!fyz2t(XCh~-d=TO
z=M?eZGbPfqkZz<uu959m_smcK1Tgz3=cD~gmErjX8rYRyuk9+eNDT7gk}@vtpyazP
zbYgtIr%U+RIm+HJ=}Y%nr!=-wX}lHiP7Mn_lEA_84H8jn81*~`N~KzSoBtxEoR*v5
z#KZnx9j;-<)%@t>j>6yj@UKW{9~;73P;r@I`85HX!24WMhA?lUYa=zSy@z23_H2|f
z=%5iu$NXl?Typpiwv33zH^z$4k|z^jCN{QR04<7$^h>Es7xa+Jgf&-W<wkcWU=U;>
zt!92BcZL7sXz+nGStxA+i(|1;NeQRG{zB5Y@~HFDb8{S>W#yU)sfmYiBC}WFwOUL*
zL_CE4`<r#OMtA{Zj+h7k@f(ajFH0yb26JQ8ij0B@(IuTyR+7VMjzfcB<TjAO5YJx1
zi6-gq4W)8<K}t^Ri;Yoj=|`#Nc>rfS6e7m^qhwj$AHa?zmSPVAY@f=mX*8ME=l~fN
zMR_lmj+0a43wnrp9wXW5xxc3~F_^*=ZLM%*$NhJApFn_hFZOmn%z5ZP{y5J)^Z|zW
z^wjZw6|xi=92^jI0h7bk1~d(rq_lr<JuL$a0ln4GH)O7}ZIN$*$edC_cA<0(1jfi_
zctSjnm(GADnWKW5<!t)<*(MyiV55{?$d3B=2l;D%{9m8%A{$_%TcxY(|LA!0U;ikh
z7fe*K7n$$<@&Nh!z3@L5qwBsjkPVJj9Qpo18vi$|0h<a|gWGMKe_*xz%?C+129qHb
zmxHTae-9P^-5+&Bj+TVZdc}XSr{ox1DKPz<O|uJL9iY_XH-6rYau;0@8x$oY*rg9q
zZP`zO=g^+7_iN_--(N|>BY1gIViKo+->zY&Phk;)M`YT$2bXVJ#00B!UXncEfX3j4
z{ddjYnir0}aIt0QipN(1MTZKnT`DV|6fsKz@^a2kGb4$uP$8ZG?}f_=amIb=!5L4x
z)di@Tza2k=sIc*vRp>DV0%f!P8pbzhW2DqfP1*wUkN%0ZhwykL;f2u=TL0dH3_&kN
z&Jg(arM@(Vk<lDj&|e2?O=01Lc!?vld@@iT;rvM7ohG>A<4(1R3JuxT|JoA#bEppO
zb~%y(^gfysF$@C>4SSsz4#?mbG|Sgb9W1`e>OzM;2VIJk65b?e!rws0mb#I7kW4mb
zWP6}0xkArD7xKMVGL-HOe3vwIXM!UW8AR^Sh-EHcLf1|H<?K!4o`!1QX_195){^mE
z@q&}*F0xb5I<wj4WUvT^%qeKB(DI9$wq1z(TRsS^gjs|Vv$uss{A+x~81MmRAI6V6
zV~pX}R5UGN(=7WO<8r+Bc(ygX^9&X_aPstz!9*k)45<dawD!x`9wHY0GD`zFzRw@`
zI3r}~`)fC0ejMW_SP-&;49^gUj1HiNn2wZ~YlK};l%hv2rMKUD)6%X3#oP>hT&Ek9
zGTKc9uDyn_3#+o?rlg;D^1)N^zDk8=ITZkoRJFp#0rPMnBhDF2KIR-v$U~Tydc?Z!
zz&WB_cFz|1l=EUQO7t4B2c0rZWz6^Fub7!6btKOlneTqM@edm!fe&$h>T<4Q{B@6s
z;s<?@0V|`+Op#zZdz0sso(%Re9PZT6sgFiMeCubL;&A;8X9k7IMiQOnMDjW67AFu!
z+ylR3;M~m!Aw-A{@SV_1a;vqDN$U41n~r9Ee2->4O?#~r<SV`J4cdJ4qmt_ye~9ny
z{U@pDR%KHcxEXQbvc5Y@$1hHJ-S_eD1^GXZ7&tu=a1rR`>}`CKU%yl=Gvz^epB-Ak
z+#LYZP>2zHtJIpx!Vvr(LqUC8BxC)xL+_@r88VipzzX%WD30K#@l?A9=1)<*<hA{*
z@rB@-w1>yi9s<wpp==uG-(Fg$F#8OU>(+X8`cG4>K1s9@Ze<OO#UUVr7J@BO6d?Ma
ze({NP44imo^T5<0*NiL-IY4&fN0bz<6i>n{9eVzd_x_*WLff`hlH3{9MjLialY(ud
z?~>x)J>UO?8~*7giu6*~!EF4h`Q~;bzfO#jbDbN81?d>V1(@@q4rpOu6l_QgQuHIA
zh+DJSm?c3S?v;Fv54rVK58wGQ$GoH;)-Yq`3!j<q2WsN9f=y4)4PKZRS6hfvB^pU<
z^m+0~B`Pp3Pp4!LwOug8vv`TlHtfPN$>Y`Lu)_?-y<_>KC#23f3UQsG_m1C~JmB6p
z?EeuJWcfTbKyt^J(oJx+^x5wE>gJ=j7g0TFmk0H-E2NbAFW9WjzKAb7BYg6#(kY^}
zO{e(^PHYm`7o`H%X*%x9X1#5-`NU%m`Hg?HFRChWz}=WUQPd&(@!;)VzTEhe%9@(x
zZiif=9Fp&C`Q#?_W7=hcp>av&%d`LD=lwou|Kqg(AJtLcDKw;Rq!tf&O5QtKWUP;5
zBOqrL$-~wEY4_6sWyJVEHtgZbz0t|z$9BL&IQKKe@(qXnE2JZ7huOeGf#Y5MO+a&^
zAZDk28GhS>X=orv!_bLrP}vg%i|z;D<~zb!`1G5C4P+@86y^CLrbv*d#WfRms0t*Q
zwiASl^L%^R80PGkZa}ZQ3EZ8s6+8s_DuigSHxKUfG^XF%O>z`-f9I#?VNGzB8ICAT
z$jz7oHM!k<$J?7X5kAcDFE5^ZFrvAxstJf}=Kz#{1Bfrs)0L5aGY}O}nmY531DmMh
z^Sis7+jok6)@P8c+M}!7Q^4V{vGd%V?yR0#MaIsgr=d`ecYq_N1B6+y$>5K_Me50)
zB~Ibvj_axo#cC4T1p2}Pe9-Y^G+RK7>`wx}%v)}mSrBVDipBKJOR=aIvBLx@ZnRZ#
z^uZ)hq%z(^fto=&58oit0hDJQARL{`<%&mIuT97q=+6e<O9w2uKBIS3Wc_=;IY2i+
z1ilG?6sfsV4o7OiFhqlt;)=|R5kS{{pAl6#;1SRE&VS%q2amse$Kd4Acl&@wo#PFX
zK^Metga`V?kNr{tY>^!Du`krUretU2HiaZLJf&Zz?6__`N*cIbEr_r+xxrp4d&y&N
z5-<n@h(NOg-?^vs(QsuIck_yG6?d8e9+Z3Zgg&zq?1aA!(9B97a)DP<Y0C4mm`*T-
z=Y}gC{F)}HmHY#R=hGV|o?Gs2iCpM)BD?YUUii|B-Pl~rg0xt>=x*iZ=ND=K9k9qT
z9xchyUr!S}%4pqI2p9>C-nhHUpW`p}+sUaN$=xk3DXj(XDosICM~AJ#i6nxL###d|
zVscV?EB5D)CIuU_-_#t#5ZVw1rC$cLW{sVeEE)W7m^<Xu=6s#}4-cjec<0|URp!aI
zwypkyg#Op#cHuPIzsy{)sn{(isy<-vqA|%~FTdW<g|tp=?(wI+Qy{d&1=sT{_Rg?a
z&MKSXsRj$_Qg>h|FAxm{cK>1x>7^t<CG3h~#(<z`xdc<D&p43Fd^Xauk+TklHKFpW
zpnAdrPpkvox8@~Q+(QgE|4kg4MA$^=PYkx4hAjn;c`I{-Xchp-(e*qH%pnZOQK@Mw
ztq;*fw4dgU24lbr!bYJ~A1aQm`0<hYGk{d_EE<!ZO)acRUPMTFnn~{|bHfPsHFtz-
zk@TJNHQ6+H14-`OV<KqRrJ!3ZMO-S9oOAtzkK{Cz|A^RjVp4u$X2ah+S(tcz{v+U?
z_{%`9vG7w1*3m7cJC~i%>VM5Y2Ovo52v5tjD18O}dTh(FDESl9I8}KwUC(S|x@urL
zAUvG07yep63VKs<9rAXc={6R*HoY6k$0vO-Qv@L~@P*L9R}eSbtAD+lFF&#6gU^ee
zxZo8%2xM-l6zqG2ibq#|vN-n=rGO!R#a8iit&7V|9Z<lH3(c%R%prCITZ~j}(wCBu
zTWoeYSc0QIZ%=PU4sfmys;3lx#BR{kNw4LU4ZROIdS;<)jzpWkJ(E!;K|HGY%ukwC
zvBeVqyLeO&1~gsDPmmni<Nx4Ubwt9}{nX#f`@zR>ylmiBccsLO?4Lv{)pv&Kq`;AT
zr{T^?TvTKHqL`{2@n^zb7<*BBbs<(kiScQG`aAv4aQ^jGWZ6f=MRpy`gZNaYx;N+L
z&HQ@%=KC|x)~4#d$7ZjaH$HYn{g|)rYyp1Zf6%K;JKmpQ)eW0_z6}T*(61id+VR})
z#HP|xp()-zaAvXV__4)R<K(hmQ{;C1;duT`>3~sHYb_!l+cRk#%&5)Fv<Y~A@DGGi
z*aCQm9A|#j7AUDHh`Y5SZ{ND3MP3>n2j10@X&1f`WICJ*@fVwEI17>Uy01z%=esVL
zi4=Vh%luSkNWFcF<5dG`d5DKD_lR0970;?9ADxwl2PwhzVN5OKS*WNtryvTikPvrB
z;?A|&9&AxoY*edvtj7a_S;;E=`_s=Lh7O%YqvV`fQO3CgCpB;Mui|8eI;o%a-#IrF
z>v%1E#5tz<qqzH>8{)&0BUzB&-5(>Qm{?Y36{`HM#aE%Z0kLpsY6E#{tZh|GBygH|
zg3V&sn+xUQrQ|pbTEos+Hi#$Ed7u7GzVZCg3~MopU}a=M^^%(ksT%VFjP5p-gvABt
zle$=J1H?v|<|_J%#g^O#FiF>MD2vd&^Y^&bPr18!9Pu$~tpw^kPt4;-?jBRoqy&k!
z-+E|Op@+ZOafLQP*~NGzKswVh$?`USf&cZl@{eG+@9IytIAPpv+d?u>^(RZJuL`RR
zgLI|C`)n5^P(SI;p7(6;E)(-nD{YkIqn$3r-|UI*=bb3@B8*K#;c9_n0QYoip^n=V
z%i4XRX_>v0lOSDc;@e9!YrU16-&bEL3?3&B+zad$8@AmB54@2aw&$yo>5HKQ=-v`D
z$9F}0Z2AO4|4Fq7M!@9yzbR(#AiyM)VhM(2(wjoThcxOM<<<s_I(nyGJSo*+P2G4O
zyKwr2D#OFl0$aCqv)OQ+MRD^J*Avj`weMdPoeF&-6uiA`m#x;eN9u{mO2j2skLg{J
zsCRcVE`Y#E@<U>ayHy4&*74W0ac!PXnMlLZQd?ni(f!PTu!bnPAGF?v@3GdGxpO+t
zk{d8C0+mP-^EQ|=FRb(xk0<tsPPTD5Kk!_Uqy)}Pb6A3uUX7U~8hzzahOe?wX{aBI
z@*<hPc~L{V@L)7F6+G*MfZRsmuXnsnZ1DwNWOnVnLseY;dfxD-@x)BK;31S+ZCUl=
zHsj?Kkr!5<EOM3_-VZlrFB<VjfLK;QRMnSGG(UWSKZet=fsW6RR(p{5vxWRs*66LO
zAfC&E5|pdtI+MO5z=d><54Q^w+^Dk<jf;q1N(9dxDA6g;(N}5}7V<KVK>orFEt-KR
zYO#?O+w{ecpH~8yz1jes@Z2G3-Qaq2@_f2w<p0x48U^~__3Xjz6!|;R9$Y^;Mn=2@
z)xa@gai>k!_{lNy;56@#VhZ1v*C5|Ty<v(Ry5FTD!XV0~SX7;<K<S2K-pnWfHB@G%
z8VQel_bQE$Uup%-Zku=&UGSTd!W`tCtjQlEvsq63tmXc)$PZ$A$!7A{)n(FKSyIt~
z1hg6B@P9-_=B-Jao|&oDq0w>P+<`O_j%6!z;%Jpcb);l0%Mn6dyJz|V%8c0o22slg
zifmZqN_4C}%I``ZGwHzl7>)S(J{~-3G&2U@-23N9tMddDXl{+GOwR60{*kt|7k_eS
zSaj*;1TESQ{_Xygkva8QU|B{pvK3U$i9a=6Rdj8Uq_p8w=+}a*jrFbKtCHFAEXK!P
z!w?w~oFgITEVB>M3?tjzF-5c$`*(w%e>DD+1C?^fAJiB9&3wyE(A&FsfG8v+F3FSc
z?fd^yWdE+W<w}u()98_}{hz4GYb+Yd-O!i1Ks$4#dVH-SR^Yuf-xn7G8%cIKLg4*W
zZAgwNtBf8ihhyVwN(fQL(-LPe!kBi#n*@x0#{Cj)CD-f$L3p%F((!q_m7u#pbiyZs
z<%jwVr8mbr1{VS&R<-vu9q8WTD4tcg_@vq{=sdHE{Kez3&}7!GD$7iC5Z?_g6Y0h%
zMtDUwWsnwDAZn`dETm7J8mzJGRuoJDPRv0Q^;sG%%_BVJ3_Pp?y)jh=(MJQ@QJ8OI
ze!hV=+OQh%<<!S;$)v^~0<*s>3tACNX$;QMSrB!MQ@Pq%t*W8%o(VBJ%GJu5_pxKb
zPY+Qw2h#s{zIYcC1%~;R;3>AtIiDvt#rKpTQKvEYNHX_LTsg_%3&@TUU|QAhvEf`j
z`vC7Bu)ak;K=o{Ee69ZcrA0<p8Zf=fF$XiD`+U@2JR2@S?Ps6hB?XhG^F>-&4P0E5
zyIE2T3C>|g5~CN9kU}LpQ)t~p3qx<w#3!d)5o7|a$(8gBq)f9TorAp2P^RA|-ZLv$
zUi#`7Hbw2t@g?a{k~w)jDLKjGB9vLq_X`q9?EmEUtWbN2dU{%+Hc^Bus`($*`!(C?
zuRx`B2Mkm7d<RO_Xfle$Ix|j?*jgbQO?(LmlO{%LXNxPn>M5A|HF^3E!-CXC^{LFB
z_q<RKh0nFrdwT>XT=YX>MunLWC#*wmHyO+HK3?Pr(z6W$(jG(jY2W-2blJu?iD|SE
z)JyEh?9yImx7iks!^ti;(5R#m!-@qzarN$icaf~yKOG)ngb0PSk8}v_Pgs)SrBDVe
zOuZr{P+`f$@$oE$((Y$r)#)II!I;m3L`rXwB2itX47@tek~>Ptw?)rpF&Z<;gSC%r
z`g2&W^uc$PBsIo;L`zt#K5lVMOqbljjl_zV{t?_4B^M@We)<5h{}M++Q8QLVK6d%V
zIa;ZxGGAm#GTj5sX`axi@qFk8ityU)VRh3SP_JK4ZhdUv-cX5H8JG<xH$E)5@G0aC
ztlg^udKy3EH(U^sk%obJ?;+gvqP{fNTJSiRMgDx8wMSVZA5&6suiU=Xoz~#;JpS8w
zI+vtP8aUrtEE(oJKul8)<dJ809yuM%DRvLSMXs5;gqWtTLR3o}`afAhFwi3~dB)9r
zj9=sBKU@_sT!2SvGfHE^J833K6%8grH`wYXK|&Ii^M-MA%T-5M!25=Bo_4Ifwi~5+
zKe#g)()YE}`>0d8`66nY0=Y15ekP3&b1>?y^?iO(r3vR4u^A~F6`e&*i<g`wqS9V>
zquYBUqnR=<p^1R$y5PRswmqa+GXgUM`8A|-HV?Jo*bWaNBNS+zgpl*8{ZhyrcP)8!
z3{B7Bj9y@zr^j6a24ew{&<&Wmd}*46Y~8EXR~)q@Tn^U*UCtrtk}<)UjW!^`z3;?h
zHQg{|C`G2+3Rfs+S9|inW%IXAP;SpUWhRE2u#`hGm30Ku_G2DDEKA2fAAAQX)~Djm
zLtsx$q+~ajmCxf`0LN2N=dBkfARnpFfZ0Tq<EzOzmC=*Ax7crxOQFz$8lL{63{w{c
zc>06zD2t1~KR@_KrTRF?;qolDnY4c*nL=0SDO>|d4n-Ru_*@?V4ie{==vE2zBo_P^
zdtgLYkL4j2cr}*tk-{Z!cAA8sA<6o42o0%fL=W2VL0j8IgQqO-#xj+%-eb4?aIFPJ
z-1jt=+mUQK-$xYqhu;}JCn)7K8)7dz+-OS*v>%)ha0z=RD*%wg&b85_YNON714=@}
z@{nw#SFxG&yhN0Yk>?4h7(*1v9`HY?zllZCNTspCy{}K{oGzoo<`ni!RP&?5?LfL&
z#0ND%14bre5gB@fz&s%)aLeq<B+$+HesbV)(uKv?PY&UT2HM?&Ar}cBR-$HQ^?x8l
z42@cC>rb>_MhoV($NIs1Mi-2zF=@yYPrSW>$P*<Iyj22;oKqn2>oB5k)g0X0etr3Y
zi&6X5$|$%6ZGDs!X6Oh6u;wiEn{Rv{6W6SENyjGD#Oeh|=5SpmTs&(PXdR>#tuxZ9
zWW)Y;t>EIpN88qi_ntWdObx_>X+c2lOO=h<mdcLhCXkr<0q?Q{0tMsPZq2@E0u`|r
z71pz@Q;_A}%J>-pLt?LhdLzyEYsQ8Y94OZMcQo8hM|HfzW<ugfk5Jde`yUrMb1aph
zyKGPfxVyYM{LIU)@h08H6PrYO{qE}hoOO`8Jc_>L(~@=glXwNZyD{^ut6M33#ex}J
zy@hMjGNS%J9FC*r<l3RU<($KR^jjkFUq(Gb{Isxq;_Ub={<QU;B?$UpI8o(r2Rv6b
zpK^(YhceXDV;$oP9zLNL*dXt*?K^AlT|`r}o&oySx+YRfGGruX?#e*vk8TZa-Rh9H
zNq_5MScy`&oD})yR^3GDx6#j5pbfiU#sQ;~DUi|Gxp5cMJ{*M<^x3#d@VP@rZYmY&
z*rn4hU?&)xaNq5WC?^vB?hYQW47Y&KbK1N+;}EgUQlfXFqI^x2|4HoRfz}?gj#FF!
z4l#2J-mOfQ{=t$wga?*j?zm+&d62Mj#~{!{<vqPhQ3fK!GH1FC2pm28x;aDZF!Y;O
z$gEo^>tK{m$Sr6a=RrPAzO6F9A=#aqxQCDv7l#q8pZ-+Rr7Msc6HuNw0QkO2E?a**
zVwp&D58v0OlZReOa8MGe2WX}RfAkW!Dy1kiixj;epUs2%+JbMs@I@ii=pPapWc*2Q
z;^OlEkfV?5<E})UM+k2eL#)JS?mDZCp<~QaTadby%`Xg(45%L^SkXQVO_W^RyIkp$
z5XBOHTfDINt)(-8segkU!KU$xxBQNFKJKa#G*0k!&j76U1g$+qI&#ib{myKw!M&D+
z<O5Efy7-Ax>f-#qrn%*GD=(-=oi23{=AhJdXcuwl!{v3PZe??c+O!0kU*Qs8Et|om
z-%z~d%`+1+7uDoBK%CPn@XZ0qENF{dVAEi<4Tz|)Fug)dUfoMeHgJ1I)OX@>b>N>Y
ztre~1wgY`iw_1h6&vxIO`HaXslhU5oO~Q7Ior^?w**_3ARaaGwYZ%d4jLxu+bJ5wa
zrKKU>aX8+Em*CJ>0L8u0^4*qyz@J>7BLk}HWxJT4B5hYGI*KnL9^wQa_Tm%$_KKEr
zD=VBtvmT(+(PA+<T@j_;B+me+N+lBWdsJxN`!wllhz-$;@CP1(=I)E}g#fgj-OneN
zAs|Cox#6K@`XD*S`q8aDs{}&C3m|6Gg$#}E%<m&EQ-hv~5PGYGMFLdp#l#4p2tK@j
zk3w)dBvHbFH0!p{$85F2>1eCHE4+43%_VQq1bd-7nLOA+VrEN{{c{do-XMv4`qxrN
zxb1+*Xh(&+m&i>cR1OfQ{#%Cr3*N#+{soL@j-8=3Rj^~&Pa>zTV3Yp_{!f~d87F;d
zBlRlI8DkIxtk)Qyapo>PeeBbatTZF-OhM<v-OuB@y|CUiv(Az|c}aL;bZ<Vi>ZPa~
zk-wuTRc}d;Ube$~7)V#{;rnaOj<Ti*I+5O%Doak+j_#cv91+`G>dOcvw9d?X$`44u
zlW@OHg?S?}cZ|n01p@#RbA3J>18wTqMZ^q$l%ND4DT`ch&jEkU`ew9_P6lmOec@NM
z<7s=?$3&kL>eXrk7SgwRYAqkVMUT9k_vO0PXfrT?ZJQ$dRS`i<?1F}E@xX%CQ`kgl
zSfg8hPeKwjXJH0L)`?5>P-c(dYEuf$oKeQoc1BuUbd%(^)D!aiKA$t4PTICvaR<8{
zey*^fpz+bvq~nM+$;3x4dB8vy;4K3rRAJA@XXyc09A%i!fj4H<`;O78_Iz-kR1g2(
zEic{k4(3Ue|Mn#Uo|8gAfiMQ}oXj3AHyi6?5dNY$T@NWBPxD8$rc}LmqtH)UxX47r
zt}GYas-i~9^Qj6v=bOYnUFRl<EFfR<A4ls~EqLhw@7)jOQoh0M6IGCyp*X(geVm-H
z$FHCC?h+}-r(t~z9x*|t``7K5I-jf;+3<H^FZqqXy1_E(Ty#@;EjWV2{-G%K9LR${
zyFBu$jEzX!&VItc!shf)@G~jXA_wBiqp#=rywi>;;06pH9LtObY?%JIq0FbBE!RFI
z0bDBPr)1f>oy#CeZ1OSCTcZ2&dcRw2(#Kpd+E~=7lY6#p6)KUjp#{&}sHQMfad}NV
zGTH!FGk^bz+5JyqJWus$Mzl*zD`&F{f=`2AgH6l{EC;U3rR544`cxw&O|MWPmze~t
zoH$Fer0lX3dUG^x4hAmr#`94#S!}7P1vYMrYr04KM(Zf&n}A^Oky)~Yej`Vdm}UFz
z)cTdM>*60*h#a*S)v4oGwp%UCD}g(wS*wa)&`s7)R#-1+19tRpoc2>x$y?3u(F~^{
z?h1BGM*2y?mzk71eI&YKAkLsbenG+DM@@Id<+wraimZa^4~mzV1~`?gA}es?kua7y
zSXtM(t&{eUFaeZpg?+GpwlOkm7;OHs(QR`w5qkcM{VRV7lL@Cy8U4m`iyD!oZxww!
z>WdXepXg8Gt8cj0nzWzn-b^{4@$eA~f%@__4?+FO8yuMO>MeMmLd2vZ!ptw8dq3!B
z!0b*ey-i~(Mf56BI7Ca|Z2siPsl#S&%%-cn)^U|*XQy)KSfFdkmUo_So~LWWxu(ko
zw`+dqg+M;Xm2A?HoQdN@hOHVyrw>;t2P=nMBxDtJS>d`xq53eAsJlXoPOtK9r7q29
zq1}7?77xFQWch4W+J~!ms#8XG&UE(19QelGSpTjZ7@B^+^W|#F_R8q|z0kRf9cp-%
zA%9>6{ryiL6B^NBf7bWM`jRivCEy1J#0@NKk%s1KStArdB{D+0r`IzqUSF$7d7>@j
zMQA^}#ZIV1>uOSh1#vk5sw{9%(M>!E47o&Aleyf-XhPIrfM?13a0(1NB1@<8G})8+
zcOs=LRWA@wocgC9iO@eVT#S1jY10Uy{rMw`KZWKZKa{!uj?)}%nq<7^0g`0wDqaSa
z+k9hgbfcETDcwY7w0<+fi{vma!5eqhdP-`!nVdelf?d_Yp?K!hGsOnYv)$xqd>q77
z)~v6UQ+akG?>efU{ihECACm#&6!x*yop6VLJ-e@>e`sI|`+2gE7O+fsU=c3!nBid6
zVck8V4tgb{+oolvvr-RpZZi$UGe8^BzPk=yKb&@RbcSRQYY?Un+r71rHZjvuZ)MS+
zS!~#?pp_#xLmQnAxXi_P`bSGJHLz;!K6%ZE?h%011sE`gBGFTMUqSB73Q<o#^4hj&
z{L5=KN1L7p|NYxUUbN5%I>WB?TZT>Q-_N_n_YXci+vOqo2xc|CX=bm(7)+dLK`&W|
zga}sAzNdHvz306*R1IRC0taR;8f`GL7Nr;xeyNLOHd%vBL4z=<pmUY?W9Qv3XQgLK
zxIF!}A=wl~8Ru<IW=SOal3SjLTR;TO0bj!Odsh}a9o6r@`M<w%S4Ik<Ivvzgs<*#i
zb-0x7VqeB}05-E1k->RSo=$D{GciR$S57m%y^UFGD0}v)YT1jSn$`ep<?WLqmm}z?
zTfu;E+0ysh<mXU`0+Rp)%uSvz?<~3zA;iZGH{7NTY8ul?JFolYUyJI$|BIY2JhLp(
zA!eVSd*IKV@c;5pA6ocE1z&f`{%NoM(>A1n6=#Ca?fbtxy1#p{u6kH<Ybwk<0^+|m
z&wu}(zn}MC|7)?tOSf8ekB&s-4EX=}Nl=Az1SK8N^}npnzgxRPECdNtdBhJ38UB~O
z`QMim?AZu7KN>@3Z2tY-<W8aUbrBvyrnGJ?cDl8i$~ush3ifk~dZY}rx)f9b?37IE
zw{Exu<ehBssvQPmbUD+?v2e??tR!gtYv=!$mnNizd{)+Pg=W7$t7{*o?x{V{DYUlc
z!K_owYL+b6ju(7o*&tUe94ro{fytx)_A0AAIIO1%vAYd`Fi#h)6|#54#&QJZ_ePYC
zDC;9q@4TE@av{p{6x|(nBjz^glUQ^&+J~#UXs7Yyxl%ItfNbM5%{pqUaG!F$IYMwQ
zd?-No$l?A!z4P$<@M5J_?+nEMdatc7{~&fRNDsuAs=$UbrUK%16k&X8nB}%9gm^AV
z0@=$nM;-5q+iAW<6k(4=Gx&l$0&v*64uhlxMxUe_DUiO@3N(oofqiQ*^X~yN@X(K~
zEiq_r+nW*_QfAv}Ryd;bw8C!?4&)qSDlYD#eM+`*DZ9z|Ml_!Vek%kX(jJ<i*QfPd
z#GBduQ|5+KOAzICPzuzVKP%l{?3YG}%*~K*r^3Asu<Iu*`k1K`dC`=TB)oyy819uF
z2cFV!+kYK#e_a(`^spoONGB+NZyCtD1hxSkD8GV2Wd(S;=zC;KAWoBKKE~!zqYo;A
zhd6z{{SEcxi=9}+c;N_*+PZ;?*~1pW9+6;7D$x#k<I)*WTyfM*f>D(X_VEX7TOA;B
zw8!u~0E!qraGTQ0>$jLO8j|qBUsYcT=Y3VwY*b^X^j%Z6c7sYTV}{AUTyTsqyp?J!
z0H|kb`#Y;_AO_7tqGT*CkaFKc^osY@Z3h8mibX<_EC(UTdbaO@EZ3=Ia<vbVTEh*n
z8kiRfAr-I5L?dH`;SXCidYOY4+5`Ruv;VMfk>i2H_tnjR{mKUIAFiZTLqeR&tY5%C
znCq0DIrcFzvnmBBWkcaJ$@YwFa2fMXWgvIJ+z~tZojI6XFasZq9m9p=)fo7#>_M2}
zEOd1KlJarg#(~|!E17&5`O33m2@1_B*XE9;4Dd^?(eENrOep}he;M6c+%K&?%J~6C
z8%A|H<wGLHMM(<&Z165qtaYn@)gOO7dlG60VY}66{4px~YoTAmTNtk_1#jTW37q38
zrF<xq48Y32$nhG48mby3vo&+5U;3&lO_2UZ2tG)(bdDbp(cw{Ed<gdJapZ&XI3;P`
z#R$Fil+72~e7pjZf4hJBMPT!oEXtI4{#hGo<9!%#9a11_4p&x}gmZ;L@aRXV)(qkz
zMI6Y<dSx0EMh4QP7~kNa!QZ)CgZb>t-c7*HHp+GR0aVo26EwR)jFO}ON`jj^;bm`7
z{(Q>o`Wl&Y`q`eZq&{1I^CPpZ_O>=#zKrBC!N=f7RBHZYwS4|6&)1;4??V!==*mkl
zq+ZDtsIvyv7x7_v)n@_pCoPRss*ihYeqglP3P#2ko-Mk@SSHZW-On|KJ4jL%&xF2d
z{cWu9`)#Ll0`7;w6T=mb#OvvV&JuyU)?`U_B|NX{%A8ghchBwjt_$&vd@q~2h;lQi
z<oG&GCO9}~<sdSvVHQ&;NhU9V*2<EM`)pv#h5Z7%sJ54H;kw?^73bU`yIb2WY?3Ys
z{sFF8q$A%qo?3m;FLTnIdHu)ZXIC9Qc7ATst<v&1zB59BV^3;($Jy#=YB3ltW@a1}
z75k@q3H-OjUuxYRefrH!eCJ#MYpm(pnfZN;9ULqeENi}LnH+xUdv{#!tcqaFAd%(?
zFD@59`PY|K!B=09D@9~QX1QlYRtDTAW(>m_JnN9i8X%*1iIr7VFmj?1^|5ExSKqvd
zddMzcv~t|>BL{pzk9gAuyZnoBcdE6>F`6qJaUBx#f>Tpp&b#kTtEteF&7j5I)zZR@
znE(D|VUZxl-i>dcXs=_XbvF3#FUEx1m~j1GIcib<^SbDxMgtyaIM{IxC4K?(Cq}TI
z3ITOq^y_Lsjqks@K+XylPSr@N*>&JHjYiu75*Z9uzxN;$&zD&N;HVrvFkIPyF>(vo
zOfmuj6$Xi#lkdSVss$+LW*T*Xfe#Q<xL?qh?yWtXWh-gf2iUm{iqNm=Ndee&>s<$M
zx@edehr6uY{~Qn~u;0RJ540PeX)o%p0DMP)uTz^ZTaD3ZD#G4+4-R#0pebaNm?6&9
zx@y%7bTd^WEvOJRZTpq3+sxhL1rAWT83dufhxBpDTIVA$?4wg-&aK%u$y$O@CM_f&
z2SGb5%RTXICOfm(xK|9wt|9iML3cP-%E0g_d(s^>1h8?Ac3oMz4gIbo6m{)qzSkUB
zID9Xke+2uZ<O32J4v%4(3Q*w?@WwVM@M<lLy2T7^tnJb3Krlu)cvbLOH{dI@0$M@$
z*6h9ahg6D8Ga%3^Z`X041^Ob8i7t-%P(OG}>2Z&~yL;C3BLWf>3Q~M^<^RB~`SQFr
z#B~>RVYKD_;8+~-(<i93-aDKZJP8H~tpsGy<3i4#J-UX4ezvaP`HSoy<_gm}0)Bv1
z{(A%WoD52j4~Ooz+C?vFMCAc*j1I09yHCpmC<eREr{|t{sG@JO%;EZRwe!~Qby)f}
zax|0aYs5Q4Mxr%{jn0xwZ}Ca;_ACbKHIY#0#=U@GF282D-Y9O4KMn0bd3~^31V6eO
zC+NmYV)wvb%xvP7#9aH0F3yKJ{JWDbA?;kookxxDF-ua0>{Xw1O5HX*v*eBl-i(J)
zw-Qc}AhbIt?khm+VAzeQ<8$2^L`?lQCGu;Y&4D8G{kr!vuiE?kTZ+bmxmxURVkd#X
zdmRvM!K;(>T>gWQIw@&J#S-Cs`ZRrUcR+#+okL*iQLNfB@44OMy)KuL(vrL0mZDB?
z>$nm5g{us(5M>86ZDbCb_JGhhHEMKuax7Q#>ffH{FVQnSUa~%6|5s(Di08VuIq%3g
zwW+LTfRg2Z^DHa3BC`}+2%aGclY`)c&MOGlG#{$t8KcRDBGD`t_$C6Dyj6unKxC!3
z9L|vQqhuTylQ*D(xo}~SF?9oA-W$Fu_&#ZRS=sZ5s9n}mC63iY<4Pa3ckImV!%7%u
z6vho=+?Vyd6hS<jbj;Y4|2#{#8JsU)rwQZAlH9EZyUdT%N((c{pjOR6R1|U2e`5YM
zBY|PjPC4ee*82DA9c@`rv+bpc>dL`VOC4EJzb3S7C5dm<;P0c>^&Jcr25?c5>Tehc
z+V%dRTPOuph#`MGQ{rbk3O3YzgRs152K~2LWcIR+sspdUgsTq`FAqog9P*6|cRrvD
z8749_1=BPCM_)E1pKC&?F`R?V%#TWV#+bZ@D4HyRa=xw%ipdM+w*fro|Dkbf!L3|S
zOi^MkBdBl0^_Z0JS^jVtwjdwM<{4wqmLeHJXML^moj1q?E}q`3fUYt#f@AUS?y4j~
zr7*FM%TaM!y&7p7<L6BTy0)D9;L5$4Aj@IFTC=Nn!RGF*7?7q;fGF5<nL}1ryZ(Y5
z$I}J5)dY9k_X5B*D&DC(OYT7~yHzp7b3+viW$YMrgX3n_#gRG9cGrCAcCLJmA=7Jj
zL+q0K$s7IH*ySIlSzZ9X6X)Z67F6|Unev@Te~l{ob=JCrb5%;|Z#_9bX@<@r7sX%#
z&1yd9{5=(Ndqs}a;Z4{f-xKu$jfYTkn<U!L>gLWmX)Kq`1aNnKm{|9B=XgB{dV|G`
zHSQ!<B;{#&{ES8&@0nF|l&@j36$v9v>A}UU{2NYLeME{a;X9QV9cT-@IaQmA)3`^1
zn;bb;R@-S-+4K39vI`F_v4%({w02pW9q|UAHGiqRxOz%6D}8*`EqQXxS&Z*r_2pkn
z(Dch_YftL`)pLPc6*5yHWEYxxDJtQCm;+AdtDb)sO?W*cO;F=WEx3fdSKo(f*I}{0
z;{b-1@4si9rPbhk9Yv?|67OC$goTA{UTx4<<!nZyCl=^0R1({VT|V?cmY(jbT$hY<
z<#hs2gi_{u1zj9(_F*vp#SiOwE=O5|nMzEu)o2F&T`8Wbf2?>-FZU2VXD+%Tzen=^
z)+)LS-pfo9@>RH2+FJ;i`?A8ZMEaNRCyzhk3z`zH{vzTd(|<YzEBS+Zb#i0GA8+qm
z_R{O8Ru#MLcWFz}Xogm*;Z5+ZC-M?5ioF-lQ>@A26l4J#u#SX0@FEA5bXith_Pus2
z{M6@)eq=LvSCa&TA48431z5!NAO=0&vL5c8+Q{oPu{Sts-^DGKb1KAOK0NCol|p<-
zYc76mKPhNEB@VyiDus~U4CyAJh*mx!=2U-D+8Z<v!Kliai3gJN(biE1z0Y2U>vb}V
ztiA4(GEsTv3>4MtRu#|LN{%laBPq!)-!IfoYTiEVKGpE0?hh-9IgTAIPshcTjs|xi
z<f^k9FFR1$JXWOh^t-?pA7-F<1yAMeio_Cs^bg(}{`RUt^fh@1#BU31YV<b47^_!l
z+A1tq6ZpHt%6S!a;)k36`)V+P4q55xPQk{X9qQJbKZx%X=whyl)cqmrjF*G>z6QQ3
z!$R{)D8)N@ynyaIT!&#k-ZW5k=2eJ+CY_ekQyg%;aWoUFsl$23<`#5%BJXL|;8ae_
zrZ92S7cM-Up|v?eV$w4=;;$R{VA)PNurr3aFHmP{dop+O){&!2R#aOn+7a8sf%0Yz
zTie5g`c~!Bk_?T6XqhGAX9bPC!3%=N>wzttHgoL-eAC4JNmm&gax;zn#0nT1a;Ii{
zz4sC~gvrpkD#lYC+Im)S3PX6-r9cL9eXdkj_}O`np)_vyLK&6EGB$I-7_kuUMKW;J
zt~P02MS3WEB#E?5orf3_aDIqW&a`12Po><r_>@#-k^bd{-(CQugAX}K_&n<G8p_mJ
zF+9r6SSlCHu~)q|*RBWceSpol`k;GSS2|C_7XQR1o{R9%4r1qGbb{H{va9c-`+@A0
z<&~<NnjgqTr$!e+nQtf06X@CgM*b4&y1<&#wI+s*oDRF>t*bKov2U7q&n;_y*x^0C
zb+-PIS3eGxWNkUlHDBX@xVc&*0cDsSy_tRK7tr&S=B4OoaNW7aBm?XhGp4br^Oz*l
z=q_bvSH<zp30?Fu@1saZV#OQlmM}{?+fgB`0(z9hS97erSL6hU^9qEpEgiW$;SI(+
ze&?CFVce4({hq7Kd(D}oINVIjS74e^r)9c|@Nkwur^1?|v%aJ=1zY#r1*x@#9q`2;
zCay=Aw+?l#5~sP(Te@4Fnnro)%%GW(Pt-xkTYVXg+PdeE5|@8BGu~x+FReS-IHv*$
z<wa*--)a}UI5V`tO&0SE<LcUUJ#@4=vNPY)@Z+C$dhqM}s$uf8N5W}u$URs;@I@56
zUOd@*#Esv7zmhMcB#G!qzGc_itri(0aD56u41zk&ubm@32X&Kn_7vMU%?E=vF6yiK
z3hi$iv5$Q&&0o9`)yr`~mw9&o4rYLT3)rrfCcGKqnIk{6szaFTgBJsb?8v^o27Odo
zGr7EO4HA7o9=Vhp#B+l&UbyLO99!c>(VF7nfTD>c$=VyR;ay{K@D5#P5Ywr|4FS5N
zxz0)oI-1}8{$I~y5O@+DhTHDsTmQm{@MJ0cMddIBrmmfjOPw0WP2^>^t|V{JR1L~L
zUq?=?zlk-ZOrjCnPlT;@@%0prc~hdZ%oj)|%oja@s0{<!9Mf)bWzS<$`JZn{Cgv6g
zE3e?yQKq}gc(atBWEH<+iKM5u!GXR4IX*oiWSy*(VT({JY~7V{tIdy0ePSDA$b)m-
zYT);({L5{fO1Gf%qXzm+lTK)JiQpkh#gsvq_myBy8@#e<QHLr0<ak*C<95f}fFJ9N
z3vr(SKd_|w07Zqkk56F&(WPK9Ua4+!&-PJctmB?2)5YiCWtXm;snZ#;d0gODtJlO6
z=gM$l^{f;1?0%guc~t-sc4tb&bBeY^QVe0rGwW2bO^IGg)I`V38_=@X;_aG`cbac6
ztHQoS-jYP5Pqa7in73zlKs0vUUv1g`vHB`|-$xA0<F&I~jp*=}-!&Qj3t;%4|B|qe
zv`8;rv=aY`5cUzG88f>H7nMMn7qx(n^A)cR3CX_7H|YmVz>ju0cqx6art;{5PgS6(
zP|P85PkSV}wyh~~X4W3R7Cm&T2S%<b&I!0PbVz9rCI)6>AKgq0YILCAZ7Yt=W10s?
zS+1XiwzyC8n3fvI^orxmVgO6n7fCETEJ~`9fsYYsd(U>WKIOJ*t*Nl;b7oF%6lZ}6
z&g4F}jg-Jc(7@~<qJ-?^6oR}*s4!}(oMR#UsIIATEE|*uIUphVrC(l=jyU=8ki;zo
zr4mv4+l3_xV!tWv#XC&l8tmH|iyz?42D7ELf!90r&;^>Hse2ij?u&XieHp&6`U>d|
z*LWen>|W7wiLW#0EX~cP1F1j|^kxUI!@kq6Ek9`8l~0O8EZEY#gxcRytM8ItZFJ4q
zAl60R8#(qDH|$;A1jXINf1{9pUHN~3g>COqe4}C@<|=5UD@lvxbFelurP$9p|C8uu
zGFDFjPkK(jttna<p7T)6g}B8jrL5GKEVqKWtF&k!ex+Mna~wwvnqnmKjT=_`F`^YE
zMr1bRB=%^I?qziG>;|nwOY?xoKeKZ9NUC293moo{EsYvG3`<p{vbP5fm6$;&{#jC2
zOeD~x-K$wNqk%~cqfrD3!>KQP9-h9!2M<r+uP7-UP&ak(C1Jy*mvo9%{=K|0k9(qj
zEpO#-Zi0q{cuDl<kwkiY)&`f|`}H?TCd8xK&YZibt6?>xF$GRh9hK?GN<N44A1gV;
zc`3056G=~veKWwjJQ#hr&xYZ0=Q1DtU!Uaf|ApZd0yE3a-1v2b$elqCVRb{~FLQU!
zkM1>Lc*#Ox{-X-r1pqz=O>V9xeM;hK7Uo2K$Hpfwg)lTSS770mXw8FZ^he3}xj1SU
zU8cS1{V=Lp-es4)tN?hxVs49_(Vy_e7?AwsOMQKqlCM}{5dAFFj0|HKux5qOyP>DG
zg!VarX&_Ks0Hkf}z_k<|<>n#L=i>xLl15Y@;Q?g_pipXCpvt_Qq<a<DbEqFokrvIN
z+2<*Bd8LKmiz3ND0aGa}3-MNYkL}|vktKKxZLlLy>oo3Ns}=kBmUgZINFkJl6?zt_
z{Q(MI-q>$eW6E&i@Z3+iJ;}%Mec#2q%4g}%HGoMcWB&WHPQf`|>u?j4`*5%XMu05~
zOX)btfjD5F%N<y6!HjSNB7VR`w&eO4ckR&e%}KeIm`mY?lryKhBK7>JCP&P~dtj}T
zP=2$2e0eor$8=+<RuaQSvZma9Yl`RZ2P{If2rO@X5Ww~4`%7;|r|`4#O?bGj2v>6z
zsE!|F4hA(j+_m{82&q7X1Ae48S=%-*_vjXo;f$|6cE@<yP;U5fp`h_Q__J7adGt<Q
z7A^-ayMVILE6*V@|I5P{G&F?UJvY|L0`M%zY}z@rfdpI)epD2iRa%w%<<t$1Mw^J?
z^WgY6j^Rt|{1x&MAB>Is&*!_|U0wScnuUu;RX-z3^dQXJiBc2Yw`^*0A;7cfv_r+b
zn>cf)yD2;F{Fj$yi0HNqnNpPWGYp}h?s<vfk+4u0?IHba+lMKa{SNJZ66ox>p0j`v
zRM0S?D#ntx4f`ROnN<UOdT`O9UrEIv*~(T_-Akq7={Eug@^3+BM!D-#-Db=9G%t_Y
z)H)rU{+4CzSFwFIaBl>4t3H?A4!KJ7|JXx6?Q!$sy!mL!7IOY?#-XYLf4~T%SilAJ
zK-32~#z=B(VcV1Ax+))jS<}u2r&5^4iCPPd>&^__CbE$XTPqmaubl5O5?4QRb#kk9
zY9<+(Znzi7za3>7eI<5kH7QV_)eFv#1oLE$!GxfSIHYuUmA-8A_{`JlGH^9B8xuBC
ze;{-vB*5lbTH}E!2+g;1vA0W^$o0(tr=I{+boB**DXh0=o<X&tUcHWM3=V`T;%<OP
z4G%SJ=Rb21I=qUKv)EN>iSg$9*hK=RpP530{ED#VJGcf%UFZJ~ckdmKW&g$x%gCx+
zGRq!GwnWMnDG|!b9tqjmWJdNZQe+E}kv+2sQTE=#71^75j<fE&`~H6K``hpJdtT39
z&wtnLmg~CC^K*Va$MHViBeGgXn<~=Xn8nR6iB{qS>fkud<qJ$5qmK5WLQqG9wmn<@
zYmD@(f&24Clr*5it!VDP=syraxwwlS-yV_(*H)YsCe&XYf+G=eMp_2MywF#w!H}5P
zfZe~4e#<%uuB8sh$E#p~GNfvcQkS6L1{B>DWM5lg4qzVNNB{XzDJCYt4A}{Q)ADe(
zU{o=v36&83P+Xn`X!BLSLPpF4kmW@WGPDgDwm=TdBSAy{aT0$KW;dH4P0(5fkJrLC
zoWx;`pE5jg9GzSxX-0`UlRuaQ1=X*@)$2Sd$fzJ3CVdDNSS|y_6l1emRC#Y$&{(x6
z(NEggKLVs?Fl<xzg%mxz<~cA4E;QQ+XCfE&SEUbm8`p=Ae5>7>qPFl>OSV2hGz|S%
znls7f{%2~IKuOBSTYoM+&R~k&pLA6d;KKmPnkfLUHWE7RyC8q9jbwMqZ|SV9FA!IT
z)eSWfqq@#k>A{tgPFa?{rEs9Wkf+a`toPHbVvw{uvLvZVPk^ZU-i)Fgv9b0jgy}VW
z&Ze^tbM@uu(e(;k4Jyi_Ux%6dm=ln0#%PP@t|k~fOthLHvjskX2``>Pc&+Cp-042&
zgh~K>_&#4uwKvVP;4vpTq$>>2bHCov9mB^!?;S(5z5ziLZYT7P<gWw8botHOR<`1z
zRe)0?l`n1$GA~(YKiQJWpC%0Pp6`z$lDMN-IE(E8yv$VYt~x%{f&LAJfq0VhXJ49y
z=MvhBY}bDda`Q{$H@-|>Ygq05u5giHSI=DE($8}hJepRUDL=<S+2&d@@*pU{GUDje
zP=@CvqG{6QQ<dMfYbr7G$)_0hf(WnAwKGpMaYox%utWNqxlq@`uw)vjeIsP@mwAMP
zwra>n9o0d^PzlPqy?!pyuT*5trQsW~!?X2Fzu6~^ka^16{jj1x_DMf%MBhd-C4n*L
zKA_Rzi)3w0O};rXrj5$mAzY>H$^{<g7}T5=t4N4}^ZnZ7i&Ac=<m}?1`|g*Wb{-|R
zMm<_WF%;tA;+;Ue{}<?s2SI9VVMDgC${|JWx|r`5P*T@9HN@F4MXjR@`Bifgr2~bg
zx7W9dfDs>CinZiVu!5dNt9KJq>wtVY!<?g9v&1r*0{fzRyPrRn^HdqAa0jCcBZ$6V
zM4(@RXuUw3(XXQwvGd0)Bxq3tT0lsMxWjCLD`mdYq)vk6_znag?ji^}J$}`i&btT%
z1DfMW-#?}g)KM%+3f%;Qy=$xK3HlSMKoZ}guH>5B1*1DA>x=uOhsVM~S#h_pH9f66
zqmGA(p$f|#CMZW<oR+Me5Q)1|bNbkd6Qvx9(J+vlib6gyJCNNyk#wx`oPnjJZn27$
zI|A+F-}*#R8h_qMc7<3-M)x8NHdEg5h>jjSmO?c#jO!&FwP0lLW@PG$;oBaz5+hna
z=987NjMRvl;gChu`}tTy?dSTsX2vY)UG<-MSt$1fyt)NF-RaR;%!ZV&V<4xVORj?E
zXFSZ0Le(eLHYA}ezGrr18SUk$$-A6a_*qUZY=T#}R#(Hp_bo$w<9^!*{1{D;vm5Ul
zVh$V~<J=)HpNYJ2{+k}7x>ejX5m`7wRykENp?`c8MJeNZ;WM)O<YC|ZYxN!fwfgp9
zDrxl4u<R1sLm!K3*dUyk1(;#${1Efkknz8J08ebTKyj=d)$f;Idi^;}fmDKwImS;o
zMp^kTi4ibP)V9+f&Pi8D;6m44wIfziW}PdM^}Ud+6C&j+pG-LJF5LCnkIES=6Q!zd
z1h}}LBvWXKcl~;<JcAt+>cts5ZCzR0^t+?W?sH;P!nJ3812?5g)JQkY4L6k-f+o_F
zX+LptJbKU9p%hQ>MXDsX-M_z4<x^v4aA^ji_Q*2;VM#q&Q1(JhCw4O@SeUq+k_a=C
z<5Ok;X5eRG>nSMD61<5pQ1rS6Y&XfzvOPd;GM_;^%k~&1*(lb=Ekga4JeAgB+--up
zrzFxK^_ebInK+^M_^Qb^Q<yL#+P5jP_LcI1saq~#?;323{w^|a-EVRG-D8?On98xw
zQdR2o0wN5x*>`ZC#dgrVE$*glqQ9iHni_?#d5bu0v&eD@y<HZlC51yBJqx7*J=??g
zE$(14LP2(X&jxLOE#7Gg|NAcM@XfBD8zeiO(*(9H!Dsx92p0>|O{a5gO{h)DQ<nsI
za?<5vS<v_&J8{;jnruFX_QYmnd~8%8Da02$@B1XxmSt$9(%CjP<?PKklXV}Lx1{a-
zG3?4aXJJZ`kukvUJJe*X1iKPCjCq>SHrvH&lPbNAqkusQd)Xqtk^G*ne{&_C(A5EO
zon1V<N3o}nZyVw|M)T#tS4Vr+&ot3}aSLdaFmUfaBx^bcxs;K$PunnXRuFvt4#3Tj
zaBlnsBQU;5X4Kqos@>NE45^TpLXJ+<ar}0SnfEv&dO=nB=^2TA0<l=T4!QH9vbE<2
zL=Wq6l$Pz1>jw4gW}M-{*+r0^Q|4MpuwfFO*!f)J=#+yYglGk&zKb)eo!@sjw-qa7
z+*J9$hrrB=ya+oeCdWYfPmY`maO4_Pt%3&6|IvG8xMSX2DFLsCPfN4%sadq5d=2Hj
z3Q2D@@s2R~qzW=-gQtTtG(u^!2OS*G;`K!S`Dpz5Nf#q97c`ik6aBe`i+{o{MQL8R
zEPCyaLABA13Gi5qID&Pwh1PYEp<JZWxwvP^0Y_(xlI;<Oa5{Rck!9~n$fS)usH)F9
zEDp4qc=kn(KPT%}$!7iYul|#XBOVXmvyFuG$uAC1sse^)T&>};;y$zS>MF!ef}mF5
zCF<zkW;a|%%!7o$-YocNF!iNd4ueL|f(3P+av#C_h9t=B!mQv>mKKO(=dCBizn5z%
z?7X~_BU1l65*stZrMl()F#gvT!HPLRi?}Mjh2->I7*ZZud<#F-;DN#~1XIb3%3ajo
zzLWUjnVRMHz2R;1|J^F|;3A|)Kb?D2f8J0z#O6x{41b=eR6t<KZ8)Nyk#*wzM`zvQ
zhw+YYJ*nZs?@B|2MkTw*-}n5_5C2a-!*70)I7SRd^O*f5D!Kp3N%+mXGb;ulE44w#
z@a1nX%wG%bpD)lfKtn@oKP#yH-Mjo*2mkYTqW*vP^)~}N=bXKDVH~}vmN3*&{F4`^
zEZ}t?jP7m-u?@dacm0o5@?WrH1}%d6E9kyG^T*Xv%*BU7lzR=-5=Oix$k1Brl1%Do
zC`>@xM#^l{phJK<xq0A~Uz;}6H0MykX(~b&-49k}NX(K8fACKD?<))86o9Qauh;#O
zrIm*Sag^^Un|o=Khe8D?_PIK?X;cYvc3z3QdMBM6Ce9Y?{QY8h{Rkk9-cMoXIDb4w
zJ?D-%<<urSL38^hoAy6AVvp8et^eYP>7ejD|1UXW3s<{DIiq`#aW`^W2YdsemS!0!
zY=b@oc1DOZ9~p9Bj&6wfiz0@X!TZjDTFo^h3x)utz?PMb2MK7~<%Nr2%{+ChFy-_b
zhu?Dh{fQQVU-RUDzECEDb^SV>d+_9+`!ydXD3@M<^WqgG?e20`J*585Km=_BWW=b`
z0AwrjoX<G)A%K8gATZO&q##da$c6%OIu)8@>8R%pLR@7tCrJ)MXhklkPY`7*Zi6zb
z4SWKPKuOh=bj2@uNKu|B?NfT=8X}yV1;6x?d>o<>1>sex{5z|cPPQ<L4|hmLYBT_a
zl>?HCo<Z-f_JZH+-Y%fB2oWq=<J9a6CGDzu*;=eaL2w89|KdxdN+FxD4coz9H9)Zz
ze5T*vqs(rvl<3uB0vX`t2tYGk;HRY+5?CQObixTu#cSZ>EY>a)zPnOS)CR*{x!Q4$
ziY!t$jf^wm7koGs5yBO9_Z;$Fq|XkD{KfWqe?ALRu=i<H*|d7QLT{~}v#*|otsgDG
z=kdG8r3GoNwd|b2{@gz4xiO(Rj)sVxe%D$$VT>a1NXy3daJJW@C&uwOLg}IrDEfgT
z=`%afXYVzT`jRACP;_vNpJm<Ir^`&0Ec-sITw)BLpgc|e;R5mr9-8ei|LwT=$0^8+
z@CF||ZvV4Dy&r{9tbYKv59j|t`cfH0c+(?9yNna3W)Q2dmO1au`R??vfWy354Q26{
z;ktsQE=*%lvuH1-)*9QLB-(3FtXLgv#){gameHTl6e^;EOoyO=wHlrn10N3%tNGnM
zBP~>u9IhxQq~+#4JJZI^v199;<c2p99weRDZqU(PjNdbSxK?D*Gh8I}LDb&)a&$+r
zK1l|Z`+T}J9>KTdHEToB1FfCYv*mN6A{UF3TkTD~WuEzEaeplnwhlJC!V|_k*D}7D
zK24c@<x&Zg#V}X2Z-LkSw~%DKQ^!)%IOr3br5DTZd%>atRZE8s60QBz+rJr6VAJzX
zd0JS#dit>qX=8hDTfW_<eZNvO#P;C~Pr;YbI@YG=dOwRMr<#?@E!6A)^^yCl+vK7M
zrB%R2LgC!$nJ)uJn<*)UI&u~>4D&=gqP5HQghoV+U3289_epnOF0@YmhHnKFk`QFm
z>Ps>IStW`0lBk0iA@~$u6S*f4X>Koo)G3%os84ti%-0>xKish%gz3U8AfS<})6K(u
z*+?*+7L+S*L5+BKbr(>MSx}3-U6MqEWOo2@IRNLs=;Hzd;z-9dOaOGN;yXd#VMd<=
zY%7m^=?R#9X#s@7f_U+$e)FFnfA<a%x!>9T$+nvMH?|ckt6@rrKqNf#S`fnLBypO<
zVU8s5PF%C(7P0|>nL#A%J3}yC`d|cRFi%0oDi5^?5_bCDzy_35n%eL7(Cp87^u3X2
z<+s2yF;*!6H1Pc0mhj49&trk6%|I$MKBUdA0mCe-O@b!rmqNAUK$bBQu!h~N4Hni}
z@VvfATniHtPQ*Rz*y#(`VB?8yWMu1<oa!nJFtDK7lrK<VPzBo}?ip~JNFO9KC!D4n
zqvv(UC89O&heS}XLG_qtBnOCu_W@MqEl<zgo=JPAbDr1=n9fVPh#gY8w*C#h4QSNH
zOp5cNa=T*~!hiRPya2Izl+#BET<IZ)JTBBdy-*CQ0o<lmEBAG(f=$EFOHIXG(n3I2
zo1ZYueEDb^nrTiL@<c%}V+%m)=aA$a1bFB>(awh&5`dIF<LCChp%*NK@PqCbShNo#
zE{ua)z?+hDI_}!+O1R9^t^ydgerinLDcK<bCI1QTotw>Z)?L@n;ujk<yRLwjSH7b;
zXa}-sAmU38Jx#sHRX)!cJ6?;0nR8n~q_;TT@!`lSkf_4TZ0E{vxSs95FlM$+6U}nt
zUBq{h%r_1bRG8~RAcFma!{uE1b+91)hm+(on7pF~gsQL8KNbSZlckrQcBUQtEx7|$
zOv3wsGqu6k6^wy)iMTANr(8)L9&_fdz6`u{2;VdU2sQ74hjtcl{`fTcX`$}(Q)2%t
zx&HczqAybX)!j6D^0Tme9gBg@4Uu1wj$_U;h;&19(RacK*bb4vB9KF<R+<M@kfI^4
zh@FJtBZ=A;r!Uw;XBlH$2A?d8TC$)fb#rMKj&@1vLO37p01a@7x-M1EX+FtsCYY~-
zXOcncBEpsqSoby^*`@owBXj9+qkA-sh9S?hW*NTRGvr8wL&LmN7b=;zNCL5m4S357
zC=o;P-u=>PDu6{xO52RUm#f22Vjn1aD03=5jEI?gu@?6@=pT4r%!Y@zk9?d#{Mcha
z_lAqzeOs^@S^cq<wyiN-IX+%tj<RAcq-k*ko)XK1ossypqu)R&W~-Dh1N1^5+sO-q
zSq}9AI2r{=Go1IvnAQEh_bdP@Jse03o9?MjgcMpNs}El->PFw@d4z|vbzIYLNnD(O
z%2QqhE`mTP!!1pKqvWqqG##z|;iE>4cr>Z&iuVxaD$M5s{?{P&jxwx+?-mcyz?p{E
z2uE`kAvu2_LtT+D56Z`hm39Y+(<*heyi!jLDyo7@xP~ca{W)uhd<D`@%$<{{XRss;
z&tXahqNKdeVZUe1vj5U2keeijCyWadEXF(j*QX@43t?}*S(p66oT?auN_4^#EBkwe
z-Gae17Olk`=nNn#yz+T1d_A?;;b~#<R~95do>f%vpIyD_r%25>2FJ?y7}(D$z(&$^
zu4x?c0YVEl_jSO8rW8pekW!DKqW>6pa&1yObQ!;0d#7&zHWK+5%8Pv{-B^H=)UOt$
z7Z2Hh94KajWLz3(TZLBa56AW0A5j?jW582)=T3%4Ts{c-Uc9V#(5rXCO%>^9-W9-<
z8BhbQ>mt$h6`GQVyASr`A7v0TBJ@OcNEEV^>;ody)fGSDozHdKx4M?wGu-d~Rz;(H
z{D?NQ{;x7SLmabQd^xkU`F5^*yo+K!`Mg19BZCeOc05;)Mx~2lEv+A7D)X%OmUi%_
z@`ag9$Ea-Aa{4csJnwHW>)zS+CPt68-Cp*p&G7C}xn{h1oA+_L0lz|(kCAXLVci%V
z47U}eh2vdHR)?b^SnZI=4E1<6sbix7)!0I0XlJ&Kb>*6;5qQ)pJX4_WZoRFbbZ65x
zKn0JYZ7}WJhPsJR@|xO_wDd8LWJ{yQ1<Z;f3F*}M!1_ugr-<pedWIhl(ROeBU4(9>
zI3s;pAmXvb*`?cLnl5ZH8_Eq6nTZ#I9tRqIDCWJVVjgQxM~q0xH?Iega|&-L@MQQg
z$6_f@%p_%7D=H<X*U11KxRL%&@jl@Eu552Ndo)whbp%DUWN?D7&m?&_wmkSjg_I78
zCwQ7ntH61)irIxttD{w+D$<3GmiHzb`TO!^#?~Yn(*c2@Cr_o*h6N~V$YBP{NhO5N
zBv(464;EhWF3d&S22NbI_&4-P&`}@EiKqF-JsqjXy4l`t_q1s?89FsrI3NkSr?#yl
zngrD%%G1ZL@~x?^#-w|whp%?@is7w5v?v-$t2$*m&2@jD=CKR2{+$y`ibSvP5u!g0
zl;Bjl_CCpKoYqd9a?N^8@Mfuk8BN6zQpOTu#nh}U{Y2G+?Cj7!<~WMGa>cvCV)^3c
z)y2&4kuck{fKY0S3jbEWQ-t#LJ9hNXI@F7R84SO$@1Sv5Qnt#;UIg^@FvZXj<zx6@
zm*eEZSle6L=s-)cQ*Z3@m+WaSrm*W22BXOvz{R3^F2=}UQe$w@M;<-PBk;09$-f1I
z@vQ&UP?6cQGyV^DCo?oNU(8<@S;S5`Tu;S3EtXPqonZ-2INC-dc!Fq(3p~G=5+)jv
zbY9GMB#(#k<&g6^^JGc@@6uV9KVBi?*T2F=eA=9<4i^tA1rw-x-L>D6y<K<<R^JRe
z8IP)3X*no}2;Uneyedw5>=T0zrIJJy$<QIfyrPv7SCS|>x)ln#=nTy>0k($=pV@)2
zXVGgRW7x*-=kT<vm{4D7p*ZAbpT(HS=Gy#1UqZgy2?@FeMO0y?Uw*b-T3!Iqu24na
zr|G`8yk=tT8Gb!jVmTOZFBW2xGprx$Bp3sJ$GTr94@2K-*U({3xJs6|Is)aY=<vi9
zqY*kh()#l`yhVd2&hAi&K6JWjZ+FANkd8dpz+H{I<CLjQgk8?{FT*V=?fBnT-132J
zF?#@6Hl>DL%W}_%!u~qi{)sp!aw3Oabo^bRpNAbF7Z6n$TaxYYrLRIqJ^U!~wYUkQ
z?FojXgcYcx=x8UFKUi4ol$8gw1aJ%US&vJQ_?#-kth^-g6#FL$YtDY(U^-=(rSIt*
z=bJu`LjHwy3;~`janCBaQfa<E2PcqPU9Q!%HV+s)Zh>7WiQ{uL@A24c^~Fy%764Pc
zhp94pwBV{hIZj6Uu(iyqM-$6PHtvjw6k0Xx-n*4b$`h3DR*vE%kqFxZ5sH+oKVc;?
zeRxMhhtk7@>hm$DeaU##5k~o^#;>ejMrs+3B*{HZ(>&?A;;;om(9>9j#Oqq5xN7$<
z6*3(kzUmtJHp6cC@taz#7nWkN-8uRRDEC?cAXdG*bbBT2=e=9up`zePNUADNvr>(`
z(6kGxEsN?0IySY^B`W-ukPyAv)<Ab+LL^(Wb<d{kCh{bfgMBhbTWN96J)yXXWq{i{
zGZ4o!!^0!F&9&Cxh6{wQ$cLI~n-()@G^|KJLt1WT=EDw-RE}3?{^Uvg+MoY=;U!D|
z5hJRnk?3zv9+Pzgu}n+knHvcG{}jI(w|=tb_xH#Eu6f!%H~x_n1zNgfyWL-g_F5LL
zp;Q&lRM?S&{c^WA8F&iHCiD<bzdc=l7CJ=o(aTCbgo-<=I{lE>ns0~3dDFDPCQ7Q^
zPO&)^#rKAbLsL_wlD=HUs-uuX+b&{WwKmyX=F=Mj+gH7B$fX;t&x7Xc6Q?>~7n=NS
z7r3&2dh;X=Mo6c&1FKkpfMJwz_Sym%g=K6PmomFoS}r3}s64qiL8Va9Jrglew3M&F
zc`y`KnheRv;9$G{$>zERCfHY4oL%UK($zokfpp_lL<mgw!2pUTGP2oL#7gwJAHlLz
z$qX4;`Fog}Mn=4X${?9h=aa=rzNr+hp4*8T;Zi*JlADg0>xhsO#CYbqs6K(KqiYwd
zE8&recSobx+|*WH`P=nrWo1?<lgY@&{9R6wkX;TW2;<N!q}#+;XFXB-_Dea<bqgws
z>yKbV-$Ep1kcspZnV!)c-4}eAlPvaWDQ^~-Jdwaj^069z82`#5%TvuR$n+DZn{<hH
zCmS~#=Cm;a#%s39j1~)$S-l6zvb2cBz0~W5anqWZ<0r~}vku0;A12I4e_DgiNv9_(
zXTJD-vgTv3N++pYPI$O>Jyw^Gjo4Occ$r6;3BN7o-ZhJ2EZe8wRv%|fC~S?`D!h|*
zdX{Jrliq8XMn@DYeyHrf)I7<&p!KA$BqHmnOFAi8&G1N7D_d6Sd}SzX^XqIr0sq7>
z|N04xmVk05@v`ytAE$wMxcF&rF+48^a7!auM;gNiHAiq`tT8b!`Iq)!Jq=)v=F--9
z0Ea<9na>H63X)S2D}IejT1mhw+Y~S~V8f#|BS5b<HK0(m68>bzu47-Qeds;sTgbGd
z1<!VEg5?}L<c^u~#Ae%$_VxStlsCSek1o8k{X|MUBfNp7>+KS=8-bS+ltXs`Ykmuw
z7K#w2fC-{Q%%o74sJ5(!c=Pn<14s<@j#aPJ5h8L(B_iE5AZEvWRI6Hg5A7ja`G@rR
zzTs*Vcdq36HSABk4H}?1OB3H9v?0fKM^$B{v({4M)UJa6jZVr0QDd))kfOt_;hRCo
zod-W<N0Kx3P5LCeJ}IG>*=~@2iNDd!a$j@hS(Bh{zJV{k1H%lhfO$;5@BA|2RC<hB
z@S~JFV*i|xuApR<$YSoe_<#mroyX;hE8Ava%rDKl_5>bmNz^7V)&6GWjDepjE>0mp
z`Rm@Wg?n<DoE_?-7{-DP`$PR!A^P08=o;`c(`MJ)695)$(~@}amk<kMKZ~XN1it3)
z{N5bPmriow%|53&e3R0&UK2f8%dB~~iA?L7(9|01y&Jk@rVU1O%~=?2rEV|{bJrTd
zU*x<`HB-7VuvMrYd#xMx=yS!1eVz6SRgz8oe7R2M3t4p|1BC-FB(qW$Mp<=wR?wTX
z`MW7g47{aym*MID5{T64Q$h+NUe0WBA+ph|Cw5ap*b+CU7C9+!)tt0Tur{w53_FQ(
zl1Et%TMl<amrbS~cmmmZOEy#{-8vjHN~&5o-MUZDq0fE1O_WPpyxCZCq?(UsOu79F
zA}lIiYue|w++WRT?cWkQW?vIs-S{^6KGxLz0-BEE?t!k7fo;~IxV{qNHOU%BoX<Aq
zS9DD#qN{yR8oK;hzWj5?g`gaaW%6Ucj!&{{^qCV1C8sxf9=%A{#A+(0m5c}YS%kYb
z!DTq`L;5Q!RE-`nr8nki`33e%*jnhuwnr<OplOVES#Omk31z)&qf;WwPhLKo@Aq~J
z6a9)dv@(=N=qig1<_pA#cFmnP0-gb`?*`Z&z)-6poE1kUYo2z6%VDpT?=HOdY016{
zGIKvYVr8c=+*UmK$IV!j7MXP}e`yusEfmZ@`r=}w^GWzb4SY&0K1}>J&4+?YuUvN5
zdQwup#J8kaUNoInFBuN4OkjQhQehU15Ma1jZOY!CmBFXe9<(1l0m2@O>D@wg<#LO9
z)!u@H5AVy3xIfMuD>5A};@r7m7KeDT-44r(r)nz~sp9Qj8V%V*WY~ssBBP)0(p27o
zjp?a;?Rc(8_|LLr@XD=P-N<D*HQe8cnXV;jAlnS5$a{xB#DDCVsLz76y6O$xnYYg>
zl#I8bw&_FGF+O*B0LPOB8;Cwy(Ji>9^ocdeV#534%3gxo=9F94&%T^_Q@w}Yk`p|h
z!|*VWgpYxnL-tCG-}ouxIQ=@Br(tu}J{Lut98CxEk{m&c?GsdDrY7d-<J9RKX=C%1
zvIApw5ePVas!m3Pz-OaIAIZI1hQ?rEtfZ6Xm44nrQ74_}PnZ+>RsEV!c)BZ>lg|W^
z)*~B1BHRWM$VqQHZlH+AU&BHfew2eATiT<S8MD4XP0Pj?pY0Q!#Zck<HY}~5cW!GT
zQQ}cM3r}<%M{U#ULtuu9j7E1?fb$UX67Gnjw^*DX)|4Il%`XMO_GB{)Myxf*fK@d0
z1~$hrXgFhxhFdy&B}(Qt=~w&-=7~mIyB9CMrp;SILU)H;K>aLj6v29enzvc+0K2XE
zQtLM$w~37I2}VC+ugn?ta$(N55KbTuk}a{w^Yp~C?S;dLX<Y0Y)_x6T84yaU!1fMP
z=~NP_fKtW;k;Acw7nZdz^v7!h8IKDeOwv$NV_kdFC)j-O$ocdpOb~r-u6%=u!`uAh
zxy~6_*yA=w><Pc!U#EF|IcbrZV|qv>!<x6H=H=Wzm&2Go9k5w~(KQvkdom@8o&I9r
zdshiUF{U=I$PRu}H>$ENp~oo83PHzq6U7E;=x6;oPD%f6m`5fLkayr$QPrRG&P36Z
zp3jnD8WdZKGuAtihVOlpf__J6KJk&<$Rl56Z0h`L$0EoZ6uIoajtrWC=ISD-B8Q8K
z3{IODZT+1~rc*>;Sc`dq{JC~K7ZG*7(K#s!$UA&n0-<Ikg_rb)!N)3l1n|uo@b-AP
z!uy>%sB0h+<$c*7KM!XCiC2CJrgdXr&2BpE!{`nXF5mcSN&JcOaA?u<?oQ+uS2eY$
zLlLLn=o(_@NoPE2q!t<8GLRT5I->bBSroOOBx)0%98T1rGB89++pEa^47ajl6j0bP
z9oMDVBD4Mz*9#<1FJXMwbF~Q-sJ(Jq@Eb^7*S=tMy6Q~`uhg*9#3VduJy#sUBo(*=
zo@Xr$q<S`=pBwLW&RzDBk=pRx*N?4vA$cqIT3fMFCmgIzAS^GjLA(lw1x{1zQ_L^J
z*dSKkYYyq#5J{fsY5yl5jfIv9pOnDSJM5g-8IC$!1^N3aO(%EP{iTuuA78*mw*d`B
z@@ibh0H<9XK7;mjm!a9)S~9b55DXzeVMw$Gmf<XDQ-$LSf6)eb=<^4Zg?h319cA)>
zDMZdn<pG&+$$<JS-`UN;5|$Zztyyjf)HVTynkJC`S{5JEzc1KhFQk}|TKtlnuqUm{
zf@+y*fVrU=<t$N%HYbrW?{){6DMp>llwS6@->m^D@NNw@ZLdOQTS8kc(7~X1>}X{e
z4@Khfl{$(|uA5MU9eUY>+W0ue6H?w-Iq=)j(wyy^g%d?SD9y}qWxnCQiJ@rHfFo~1
z8<EIHc>ZG=9O>6LSGFOpZqGoj2gh5?d=wG*2Jeb3Xg#WY;!cG~KkilJ>27Opfw$(4
z&DRR-<X$a=lbciN{A)b(Z?kN+2!Q6J2F$Yge#}|Kt5{Yb)b<66$be(JUci+L0m;dU
zIv*muQWJkh`mrl6cn?PE&=4y(Q#*Z33LCX4v6DA+KWlekBI?-GDYX;AEqc6{sp$n5
zL7SQO$l4UH%>^aGuIW1*Qp>PXS$gorg3L--Lu{R5zVzoN;7%HVhs6iJK?DIOoKP~j
zB5XV9OR;BfXCETuT@Tu%(B`+8Q{YN_&sm>c6_1MEUqZ(zM=e)}CA`&z2ek^j4b=@E
zi{95!{Se(nVw1?(f|ZPK{*IC?vKT8F<zyIa<|<f?5X|`Vj=wG-PJNVNw=w^z&v9N1
zN{kihitw+HOyf`Lv3Kg|O+@$KDM1=I2c~<WHKX(0sR9wv7AX$b_eeXr!9G_Au)JWZ
zMGOO}foLi-{~v;Y;J^(odae;;Kk5hm6&7{3@iyvAlPkF-oxT7Ppc-u7b7Y#fGbIcG
z7Ckyv2zQs?!((OKc}3lCW96#9MFJQU-ggKU48CJNtQF{)PsE*&C(_Q#N;t)89Jb;_
zK=fFGTxmF}-(IMOCDS6k5z{{N=<wrtQd_by=lSn2>2LvXssO*)=6Gza>oPcAyhk;s
z9O6M;BMJRp{?R1}G8@6g5-vR24Y5p5LT#Tee0~aga@?$m+WOOIYa}VV0hb%S2Cz=*
z`||@_aTJU@H`chC=y0(2Pums|!UxawK(2@Fr!g3<EjQQK=K<UC^fQG>s}$gXjpl0R
zX-}?0y}F%K+5&JudyCwu-!OE<G0Ef-kDzlrGCO}k1&R{bJrzOo-fw%=t5*HFG6Dl>
z7B#94{9X7@YRi$4weG%o0p;ENmU*(|5hY$9y0Aj@=Cz)Jn2dBTCEmz+%mO6lA=yA&
zDNH|kiu*>cB*TA^$`VO_kjj+8C(_5>OO;?9XXzA_ac0XV86~*Dq*Fx@5XgUiDw#`p
zg#o4+tlz>J&<<Cr*KT{sT-92Uq==HP`hrON@5?C%8`-Pb#F<0jQf+Gf9NO(V{1dc=
zOpj1lMziE}qW2}Gh^HAPMSpP0swl7iH>a%XBf=>QzvuQHNO|wUeU}N%+qQSTiuy%}
z&w5ZrALnyv5!_aAamzkL2i(*vDT7T@e9UxEV2!O<VKtj@B1=-LfIV23%%c_5O{{Dy
zJULu1g}}>I=~GTc;WW9XIRl!ZD~vEyLCQOeU>n<k>cqTZI~JPw<DMi=Omf|&LEvyn
z`!XcQ<JE`l^rNpUytrs(x!&3kK=PqjN$e@szUpKGwaa``vVrjT=eOcKlL#W(n^Y9d
z9Q=2seTxLX*_++<pjGrr$@s{ga(eb-=mm4C5ZTr`FIbjWH37cW{=}InEut70Xl_)0
z{9@tusYR-3j{9;GWw!==T)LiqJKPU)aUdHo%>mOlI)8%K=OYt4*i1I3d$Qt+K2%LV
zfJGW?OFCc8uo8c)+M_o%;{(m)rIy@O;7bY}+q@QQlxN}Omq}qLp{vK0eU5aWtI0{Z
zv%!6ft)_7I6mhL<t$U3S-7G_fjQYMqy|aex@in%sVZ3n(v@^+_nPWR!?VBg+>aSPx
z*^+m}{wllw<poH90kx{RX&Ch<@n{IIfm2pZQ)n>$0RJP)nsMV(V3!wb-#Ju~vv@gS
zwa}H8tob??rB<cX$7;v*ibM4FWlU6Ari?thcnK%PrezFxtl1T{iWztC?-m%y{_m#X
z$jOWg2o*&mEWi4XPX+^{QdVu>moM_J7Y6uX>^cpeOtg4r(VwcC54O9~AfKTxkE)t?
z82o(db_tDu)r(BBx?tIxh@2)0z^=Yluk9L0{Ozt}pPQb{or!8o!RX+37B~2lWbqgD
z^S2%)LlOBDCe8Une}0PN*m;L^#U}L*K?AK8;2PV<VSa-k0vo*5#A7o&+;ns`<c4#&
zMGn6dm!@0|X*F5-n_HSQNhyr1W)Z91j2SSbFevfsxTfw$5od-*i3YL!?S23I!+2>R
zG&@eOd%VBy({t=R6&)B~8oiG^vj&{<Jxy9hvIY;R2%aGphN2BwuUX(Dp})TA4@t4;
z|MJy%`g&p?o^?(pe-DTMpNxfnI}*`P;YhR(Oz?g3lW!n?$ijn|S@0R(B=`A?Z|09a
z3xEHcXihF?szrZ)ym7_#L<fBM``Tk1A;10m-pJpVcxm)I*!v$;hrf6-;>fyWM^Qfc
z<MaLISO4{YiEE>Px5kxGAosfk_}{nSZ+C|^9KHY!rs0GCm;XM45MlJ;*hl~Nkoa|{
z|M>z3(GH30$^F(#>!1JZ*H7#vpo%X~>^T1y*D5g@&nf=w4!_CA8uub)^E>40Fz!>Y
zQq?d`dz($iuJRn)L@jF2YF6jV{rs%^rZQ#2gkOGd-=_Eu=``Q%#lHCiXZ3o021BV<
zLG18pZ#9I)8#hMTN)0VR@OyRpOChz(j$b*zrZ(8ZrE~;_++`h5=6%~q2_SnEWeGjx
zgqqwB(dRpGAtPTGOvpTmS(w#je>aA+p(q4w3AMe9B5j?%+4ah<eap<+C0JG}JWI)z
zPgA?M9{0P&2q4dYu`@96dyX7&3V$S?sLr5zS_SC`ji8}e1M<42Lr1|VWIeJV$zdt}
zm1~~rnA8(WxiR4mLQTP^nCktAo`Jwh96_K5*yMqv*%V4+w*BcD$=RDv1uTXdUgd^R
z0gN31_RG7{NuUzgjFfnEbl1i=?gE5HE;7kdYFrbQ(;S40o%ZFOsM}tPHxA*!ZZ7!q
zJb>B7vz(gq?=LMmQ+;@ywF@*>r8T7Z8@w=<V&8khOq}fC$wq840V)sw*PZ^?7sT-V
z1bNo|GOzQh!>a^N{CUHsW0zo%Oz6tf*aIWGNN|vrQO#D@FRO+U2%&lJ0@XPc=pN^y
zNfXE#h*5nsjw3c*P=f9FT>1*Mz#Cj;4cHzjGGQI(cCgrc+*YR=5i_3W7p>y$=07sX
z$lvhzQvN09iD~z%T^vX*xPutGoPt3womvaNPXfxPj_}l#!!zO)Y}(^id9_EQox${F
zX&^^is0?g$W!3ba&%|w2hTQAzeP|D;8R2IQpFh95@AoH77o~lf^fv!m)Bm=)8Ni~X
zz<T`Y*M-?7@vJ!>KR^$WlGZTpe}R!N03uj(9IrMQ#GW;^po%S|S?Ph?OuTjXm@xWv
z7;QvDQd21hwuiof$Pu4Ey#O@#maIJ3!Q6{eV+EJE3)GC~fPP~sXr|L|2upS2GM%?r
zyPKM}y^70oRlS?n=PWLiQGZ4rC7c)GDrKbOq*1Njo4-2*j+|^^hEsrrNaq*jnN%VX
zT!zoPZY{cS?>xGg3RbGmq9$@be$BpDtDD3TQ|*qr#<r71_jJwd435XU2eMaKpV6&$
zy?Pz(@HG0?!UOKci_#1C+v0wI&F&*^r$yIejsAFFaRThTMyld7jW%gfs*&J**8&gq
z-F`tU2z+g13c2C80}t?E*Hx9t|LrWcMD;(N#mLx79m>|UN3{-<TXBt35$}ErsuAe#
zWln&#Sazxjt~%+-KvTagg}k_jyS_LcT-=33N2nj<<3&H;a!#t_=zk_y9&Sa4onXIo
zyujLd(p~<@!N@w-n*y^9dC?=&XH1?sjrAUJZVu^vxYlv@cueBMZQD*|4I!<n(R_g=
zHWMsME^C_!!sw1MYn|Jy7b{xTR-SF`V5VoRMg*lzf2FJ_{5nMQ$X+_GU|Z|(c6%ia
zS98?tVngp)G^KgajpeDpe0H^~7DZs>wShHIzKj)@BJVE~yJBlW;ND-FKT%q?`_ZPx
z&5_9Y(#@PVHZsC<=&5)|`&J{Y+P#Ug388_}dSd4ZT6eC}%g?4_>!N;-RfmiJQy~R|
z0XB%u$d_K0zc{LV*&fXdd~|*yS1{9fd5lg}3zXunW#CJG_h4tWnEwm74>5tE>vMC^
zLELGmKHSnof+J_Z(>WXzb1LLqTCC99yu+yr0Xf)Uu{V&*-f+s|(lr%lc<i?6nol0W
z-0hiBTl9M+85%OVsf6w2aa~|Nv40<WnMgzDwh^KI6|lYs$n*@LT+SzF6`~P@wOD7=
zBqii7Jju@xBieA5ap^q+9|HA<x*)}~kh%1K>EmVCFZX=_a#>EDzk6-WqUKR!;WTVE
z1@7B~eNTX4Fpee*4!R6tPQXgq%qL(giw63~%~uKumdSnqplATUj^Y%ERKehoBS=~~
z-n5a7t=(H4Q!^-8SKISY3tYJPw<N6NYR#qr{$w7YJypC5XhY1X<NU{3L25XjF0Jpw
zl?a(cSY(F?|Nk!*nV;Ja7FkplUjt{}n^?>8_pQwWVeG2ec}pq0#rn5lyc0`Pn?ZWS
z-U*-`Bg9}9wqs|wHnVWaqIy*|KMZ;Co`^65%8+3Yx1WQQN{+x18akkGc*q7{dvO*s
ztknt!3btp)F!3xSwA?;GuHvS%sF&D4R`p_<aC#qx_9Vf%u=_q=Ic>f&hI!*B+3ay5
z0gW{biC=IE#m)o$C=R$zU%SNIp+9-hNT#;4j%R)V9I9~pzB|ndB%<f;jom!2n`Ww3
zTil~6<h`Z1==cHfxfxJ-tY(S9STQ!8_hL10wFH`Ya3_UKfN}Q}@oEw9kT{^A__m(#
zz2yp#2=l}~jEY)-#B03l@Q?&Dl548pO;un2;o9XG)kZar<3v>d!?mjngqDkLc|ge`
z_}nRWFiBJ$61^5%*X#EiGXW^qya>)Fe(#R`lN);tCT#7H#(X;>tOUk#PUz7lO`wU1
zkkS5^ahJBwmU8$^CSJ4IP+=x4mM_l0l{HemVgG6P_K`>Fbu?BW5MHr<6Vy3|aKBa&
z<1}Hi?Ykn|5S!w{roIM*3S)akZhnW*FDuJ{a(24tP*hbx;i94OG;A-y`SWvCE3$GY
z5toe>Q1)RfP_&4U$=s<VxLb+nx7l+u{K4P&sO6}zc4AohOR{uY`%0Kx^xr0qJp>Zc
zS0otZHgQTmh8fu{7sNoVuK*GRo)ukb<WhYVh+RPGiN9w7KE!uA=K>MJ*$rR3Q?pqa
zEd!s~nn!QjJSv*}Bty3eVrv6|SBzo`B^Fzump)S#oEzu^IT+RZy4l--DFjSwmbk5I
z#8K&tv<Wzith%u=aEOa_D345LQueCX96buu%4sPa>G(itVX{FNw#$L$YeZrPdW#=%
z^S;3C?GXZcT@{GPx^DAk4Q_(GX2$JcwK<}b=@yQ6Ji#^2H-X*4WRxy!a%;+k28KlT
z;uXrl6tOa0RN$Xb>@0<o8&5!A1^Vluq5MwKeJ21P%Kf&IZc@tFBvB6$tCS2CX2EhE
zMi_5z-thpltC2_;Xp!j5o$keK;H1S9d`Nz6W;1(V&4j8+MxNo#s;|v<9U(X6l}Z^N
zmZK-AO{Phf`6~BUD!|20WV0(;=?%;TI?V>(rfuSUHV(c*=46yw>+6!Usmy%ONG~pv
zwV)|qwY~D?-D=skoVt+QHx8+*+Up=+c48whbS6Jcjd!s4>ZgrF*nR;bbvmY*!uOq;
zCdz7E<#t*jNBErm@dSq!W6}wRUmsi4=J3o-ig<uD+;t~9fB;e~qrZQxJ@8)*vErHW
zG7Bbqh&3tVZASKOZ}2Isb#9-tzbcl0eL*IV;j!8A7oPOfzKe)aXcIlBuYY>-xp92H
zV8fM+&r;XW3!PHOfnBSsdSe`<0}55;&R~fl-zTyeXvm1`ulN85ua@4QRX|+$F^&s|
ziy`j;I4`!e5E!s-x*JZtMns%eYFCp#N+(I;<ke$}N;#u+u(1TcZRUDxY82fEW^zXM
zURfbxWoB@x;({+Np@od{H($ILz;VG;^l$4SRJ|Ilmo(-lM^9iQ{S<NTnv!N%@@Ffi
zjLs7pVDTwJSMf4w1`BOEm>6=&;t{M;NxjZ9+r*1Dx`@g}%)D86eDTiQY$r&<LwgqL
zOVD65qeds)&*6BOl)7XGl74?Jqez0Easqwk!Xu@XWUV3MY?2PM;T-zJqwIM6$Tugk
zJNgBClGprbF$G!cV9pxZL7BsuQ@Sl!9R0u^BVo)+*Le2EBTmG%SnO)O?WvRcI?emH
zM6mZd^+F#dp68gUtH5>J0xWRfboyrCbF(^UvKKP9DhQs1eUbbUej)5awb_W!F~eh$
z9p`Spx~oE1jG@FSHGr@R`rJ0&?HdLVGhMy=;R8D4(`W60M=lX>R%5Z3yPzF-zJl@)
z_T2VzQ7eA1TJ>m|cC$TJCG6J2+jq(8{!`!e-|eYaAW|DjJ5aI9|1rOb_w%*kV60zP
z<*9i>CuLSXn@)FJ|0MalN3?ejVv|VxX;y3{O7i)0y}n*X$3#yogHk4>b!>|MftN82
zNA#Dz<otz^5J?{bl!n9xmW{Z&`TSZvaorI-n;%|X1~%2<4pP1NHpqQd$6V@Jikg1A
zVHYb!@~);p*}_*6p3!sz^~*^M$kw01Q3a>@z(C+hvzlqrCK}<rSpw9Iq}>I}k<ymC
z@4g={e`q7ER5{xG>V`0wJCdYGjwox^ejv#aEhj53QYpy{ej|CQ7%S?HMa6@<Y~UIj
zwRcNcudp0t&{SqMu3)>mg8NCCcbd@LF4mF6a>1cRDTCvL?Pcr*l;c9*z^Mg0D)oY4
za9r43McGciBqx7XKWq-z$)R1@kK4EVZ?_NAy}i>$qQs-vnIBz&r4#TnhWo3tQ*aMw
z%7@_xT>JIwJF4~6MBMGDjn7oFtMU<6gL40ehIkCYEtbFOFa5zRppV8NZ_3Bf7qmn7
zoVyWzRO#jq7T9~d<coM%`9-wY_7GH7An24_8~vnA638It70+&%`>g?sFZojCd2o6n
zX4BIYTfq+$om%solX7}T=&EsL?-oY~MOSF)WeP@<2L=JG#8QGD{xo@<2ET5K<O=Q)
z3$2=+TJmPFBh*(t8BBvGOn=_HIYUbfFLFwK%&uErwFyA_xhfDlv7D@TYTQbptjKF*
z<yN^vCQIq{K#@gw4`PR8iqqC%LTVEK`s3AL1}BeT!r87+>%;rQIdMFM!^zFQh>kr$
zW|aHKEmB+D@pilsERxsCUu;d}EjciU_-Y+XMOo;JHW!a@X@PT8H2E@fB`z^fhyK}t
z&mswm?OG%qB_D+&hg86$mqNoY)fqP3=`sC*TP3eN*I@Gd{V3MAQVj1A>xVqYJ65ZF
ztPtng$5uvKS5|QMkyLQLDYJ4`XmjL^)<l0NrGYvcW#V}sJ%x+)=dX$T5(|O>?Oj0F
z&b#ex1iyg5>^pX`N@WnybLdq0wO2DF!%yrJ=T*L`2wb3IIVO)`t~O(Qa~rKzRrWbs
zKF|3#%GV|r3tIl{L-yP+T-xR8mq$+&{-Au3=?A!+Y<mvvcWzPfYqbkX3SyVKU)Nme
ziG7kWp#;3s7kqClzp$}Aegb{>bvWS8ccHgh{+07JAg-Ljh?d~u3ZTm!7DBk}90=}!
zoS|W1mz9!2f<cg`?8L-o9wZxiUkllBODUt-@FH*s$^9+YsiV>wGkLCE4W&NkvDck&
zz6(w=6=f77`V<F~C|c5&K3P+ZaFM_Ds2{k<X1pW#nhGdJa=h?|K4n%`&>|$x_l+fw
zai_e!dFH0CQnKdJAOcURl0EX#&|Av$2F@v>AQLD<G;GQ%XVc>z$eKU59}Ue2>@C(t
ze005?HE@~`f|HcWb;Q53;2fFVJ!Ihw={-7b{<YiUW&xUpEaL7%fDH!C-??soO_r?B
zSUeOYI}9MGm6Fywf3KZVjP@*&T~;+64_mG4_XYCxWF0<foSfu%p;@qQBE48L4(=T1
z^=dqW=Tf+kjeD2$UpDT;AGNy5`>;nZ<AL;>jCt=4oX;Gz#a&=c`H8ayXSwu69Yl+9
zP(DE%iOg?p_ML+wHxB6#@=cR11uJf>$1=XAZu-@y{{>$D4Utr)1)U#TZl}88A2=cp
zn2ZHq2`&Uj^Vk?7Zw7w_`Z!PV6O3G&T!xGR5TO;VwSp^Qz@<cX`Z%uvIJ3yF;4(<&
zDEccNAzTHmo)12)?QDQ6`nck^<?r7*8x-IVSzTTGxgP|#VsjA?U$~U-^9S#X{@LQ~
zAG|L#@hv1DeG&UyGig_7Mskq;bE$A_7DP8AF8)-()AK9R-l?eGIdlF}(<CymDR)bq
zFHT`$@%gA)b$A^JVseFqtNKGnymj%?aw^wky(UykV)?xiT$;`wVMOEPFaJUCvd_wL
zXt2q&sE}oAXEUyN4g8g#EU!nni{}#(X*CM4GeLtA4jPkA7RzSp6ATxpde{@Itm0kJ
z=Opf+-}m_}AsKri(qw%7K75=P!crZ`xaj(Q`xs1VO&|7`I#PzBuFMgjcz`{l-gD)(
z_)|X&FGm<KN|k6lgTc$3%gEQvi{d2rJL4|#32@?g<eNA>`Yk+|sE%1~{U3Ez9+`F|
zr(3=XO>uU;!{h&Af;H5Whm?7{^cK7!2P^b=t@n#M44T9SLZaqv={m>E+9`(dnclb<
zw#jM+E^QN-{$a=$on|Q{MdpGq!VKG_@^+92$kI}h0nHwP7Mq2XDMC}3p7qAOhUjXx
z$04m9VYfS7A~g42J2ehN|9rBDY9#NBsg~kuoFi1&nqnD#wWW6hrqbn}8%*Dn_RV(-
zv2fpfCj%@mPCthRi3~VHHFG}?Nm38D8dMBrD0Vogv;&nZl$)&d`G--6!cmG@3$5QL
zD&B`pX>hA1rn{#;AzepTge=*0pje|yb!cm308{YaHfQF;9EshpG_S^_Y%SjhZ7b#K
zR8`)5dD%#r%B>RpI?Dy(Gw)x0*n_jC*y$PX;uT!J)SQ_yP_H`T`xuHGQVrZ!zi~5z
zl803RbLDCM->ck6u3!(?<yr2voH<!%umx@_@y~!hw!!ft*+rbM*4Hv>et|st&aOB|
zs^)VSM2--kNB`jUIZ{St#bx&<$r9m?b0~38u322n1&}17v7&MJKWJi-*zz|yyc|BC
zESc{sh<B-5LxML;4Gx`VQZ!{;fxDB8t53OC5Tkevv=iKqjefGi7~)-)vhqLZojl`@
zcwLEXX3Q-(bSufPP1^;jlf>^3pNogs)P_g&7oX*k7>sIx@0p1Vb>vWSR%%`LgW{|q
z(gXd<3}dm_C$<3w+V5mSxJTf#cbdl;d{bL?+qzQe!d0~GZCQVR%4u`v@qDEuudTe`
zG>|Ost4*%iXq(VPeva6vI67P5NGaut^1Zq#G=)4v`|+MJeAGP^Zvu)ynU}S~`Tevl
zqCRVPuoKb+UNCd!1osE4z>GcK(BD55YY4XQ?ACPpj~R?-fZs@+le#dD6+Z+3wWBB*
zLP(}dH_Clqaw>;`bD5b0HBiVtavuNCeO!vn-?sB{#YF0pn`?+2v@1*8(SB6xLUBqy
z=6Sqt6CVh~oNB$C@MfnkQBTS%l-%ApEnq&cZC#5qIR(vCCk?}xcq~|~YrgyXRYwr(
zX+ix7A$3iL7kEE;WRo&Em&N_@mKBxclvvh$t%*6Nyt7h-$h%I=EW{JiGMHfY>rr>q
zAF&}4-AS~0WX$d_@)wV6p}4wsIAZPxlg#)d!X#TI=hmTq&P91DhdpQA!msUYoD44q
z!wz^~-^gA{IR=J8+Ec=m^j*WC7OVsjQh%k7?-GIO>M^Hk=qG7?(kCYx*a0S&0nz9j
zLp5R#W$2qRDNI;J3s_T#)qrJQh@ea?{paqbHy<o>%F(ZbFXUuDF!7uAb;0#0GF4-x
z7@OuHw`==!P4ybW3>n&~a~6DBpiRcOb7eR(nA_Ux?5%bcNFFW(aVw!fxj`erGH^+@
zt3eRSbF*R$M)l@#DZ0p^DkcwjdKG8!Kpl22U|%_ObfOYv02ITBm{2rqU2;u)4>nXw
zlAIvo1G3>0!p~p{**n<EgTPZ{14)%PTh^{@Td5b`0!ns__~~^PVZp~(V~?ErzeHC@
z)7ldr#XGvH4Fdtq*l~u_$NlJW@xDEVpzq%1$%pk&xyciEJn8?}_Vw2fdy&x;dS0g3
z`^N%Iq$RGATtC=dm)FBuzQ4joL3&bR#Sx&iLB?WDxh7)CqhM9GVS7ZUR-?pn2_`PR
zm5%xh)R<)mDE0+%H~LF$3`s7Fx8pL5AZ|J56cIIQ02k=!?^;3RH;ajFT`|-#cgm>D
z^Sr>hz)8)ni5M#F+xQYR4Yvo5ZX!ly^m&3vT3=D0M%z@r9xWNYllNJ=o==A>97Dmy
zrGnJsfKhXJF|9SM1M_PIsAUWK!#YlUwice^$2Yw&Fg63$C}+?(Uj4A4F4bmHAzo!N
zjUW~s9R}OTO7KuBn>p2dVdxL9vdeq(DFsuH#)2lCiH^}H9w~tZ-X2rGPuy?}b#AHz
za$ranJR6~-f3Ebsl+Z~^D*me<=TkF+#^U<f*r#U(JaK)@QNEr`Qdep*s3e4a!obZi
zl$x&lXGWY$>-X<T865hQLJ@E7DoDrV8=vBDrGG4YmxlR<B;Gq_v&Rx|dk;8Mcun$s
z!tdTVb34ZTy=sf4Pv`f;%Y1Vkb7T8u+pZ<c+ph1%<L6dO-9!`CFD2M^>^nzy&Q@11
zZ$CY!1jD0bsOh**gZab#n<hH`yWlmD0^_V5`&ayt&zT8GN&wL0EQjBzqEzjC?92~X
z=ko5YX<iTpw>Psta07p|y|UcAC|G$nKHxZ8RF2kVBFPW#A+Hm_-ePTp*&sOme%_VC
z%ueStH}!o56tgyfd8Ye?dlap#aT!g@wmeVLT2;POvMH~qWp19)aTzx~E-TDBnZ1>n
zCrbM25U{TGI_@Zlh|u)mW;Z>4|C_9EhU5KfYp*D`-e@tzw2hkes9v)7xpX_gnQZYy
zGf0clhLqkf^Ly391d#mwdR}$&BKX4TY~<J%K?=mqZgSb{-AMLx%H^Mf<6pwC{JzeD
z2z{8Vi_Z9|nKA*s_q4q<V~s=a_?zqO1wZeJp3KdSTP9s~5%j~rrvI1ZTjZx+P0|>(
zy3?l>vb}!%XPT#HW$!$$IsSB=%-DU&Avm_&s_O>dPlrWfADMg(0PR^a+KomE-3^`Y
zjg(?3x3#;)(8BG{sobRf#yo+)lntESu~;<|cIq>Ni+yimxz9!f)-lR9q_4TCPKuMA
zP%=nLCB^nAKihd3Mp+@uTIP_Wu^n#<g}GBi@ZvSuwZ~+8>fK7z%#pIPt;+-XDY<3<
z5eGZK2Dh9fZ;vhmnA0UfwX1-^3jDd^(_+vv-yl-TwKm3FiDqqsiTL_0w*C|NA7?`H
zS&mM+Z+{)VJxa-IAay@<zFI2nWVq?##-|uIpWMYr8j!3%Oy1#B(O*o3A=kG{0j3+3
zmA$h!W(SDToX^r4EWB<D%3@O~ti7Q8x)5XLc5UPI=O}DO4dWM@8n!MsAD&^u|H41k
z)RTGp@N#&@^^r4UPy2a&wGay@5e4VBkrO-#D1IH|{5cKBwY<!&qUs`nGv3E+C3MPW
z&aihd@+mv<4=9bV%LX%l3%2bNC868)C8TgsNLfAqoTGk6W%|d`dcS$bCSi-MYQ5o2
zl(Y6Q8~|0;TKz9dToSb1cV|tfc50Ml0zRpz#_LR9PwvZeG&7I9RpuIk=132f?aV3-
z0eO;aQA*^vnfHVOqlhzI)OIuUhdYib)#ELODb@Y+M3if1@GWM`RMur(Uw&2rbI=1?
z>d%7^>vu9su~hGaP*mHt(({Gx^s{}$=i7E~5~C;Ze&)n`{@c0=39C7O{(U9+uP{*)
zacrbpGM*wEHzZ{Ef=FA8#bo!*)(>Xdvmy%Up1;JOy`?%x)ZI}xc!}i9C6X(tYbt$}
zv}It(@!<}uwz1ClIQ+;F#5bCwC(RWU^l<_29y*h-mm#wHP4fdO-E(r;bM0B7eB2t<
zVl3Al><xs?bVz<_3`&2QC9-ZX4S6c@g%9^@M+b?fOr5V&u<Td^2Q33QXbTP+@#taL
zT)Wq%C|GptSSYP?E7t;z4Gf6J9Z8zYpsf(qxiD9Tk00l3Xv-pKO)lB>yjild7TdPN
zN25bZETB`h?nFq6^Ksd1F}`l8ht*cC&ql+9*ZlcD#c;p6WHf~*w#A}j4dcr~zRo-D
zaqVqM2~5v7XUmH}`gI87mcz;5ybsYd%NMNH({3%{*+;jnfwtS01><<{Zi;iz0~Wg*
zv(GQG8|ypWUgsL$L+QSeAaQg$!QyVm5@-9m8#~vsTmGVaZ>%#TIm;~@MEA_s8POpu
zbBToR(mLM!zTp_}OXraGhuKs8FX*P5!?xGk8nWF_m{`PS^z!j=yPqRxv9@D*EBoaN
zti^<N+5K+d4aGU*-LUy2`sR?enpUo5POc$}?McJ=k?dmUv`d*)U6Mk6Tr9NgRn<K$
zAZnQHPjUWW2gL7CzRoA#++Oh)Ls=rUG+P$w%xm8%vK=>0RK(r%q%Uqv8Zi_Q7P8m*
zs*@7fnwcW-S%v53ezEgc`=Kv#>eO#-Y1{HsZjGB25+T6??*MK-ey(y@u8{gI!DX~i
zIILASJxs-S{r5Wsa?6~@XI(DN3#nD7E;F%QwR54Hxsm3c-lBGP`+x0quxJzYNyJzB
z^P0oPeuRBvV4ueEg3|b^KJ`^`3`~@|EZ(&h<1w94u(w;{j1GTanZkLd@8Su*GmQMX
z@nbQ_RP7Gsh&Om`h|@ot$UVJxmF_q1hJhKy=sA&li#Zb)6ZPX^@ruN!-(f#RLnI0B
z27wslTRpD$$Hx>o?TQM*U#oPBUB1QV1$Li8g!x|xTS=yYocmRTfMRMA+9|Vl*xv7k
zjO7S_{;7X`1QowyqYiq%Q+~w4L%xq^&8-{alix{06(`Bbx)4%syl<VRfLWh;=B)Q5
z@^?-%9`W5*bs7mL?RW^*vXha9SoAzVMWE{w9qWZyy4m9a)ZCi7a|DA~sB!`!cQ9eu
zK~-v+=ld>)N*Ys8wU$&bCTjnDvwzzSP@dup@R7(Q1J3{Xv6bUI>y84YqmL8lzTdBL
zN9f`mvaXJUWrh}@k?dn!WXQ!qF4xCWyld(%dkg6mbRGwvs9r6T?CB(R0zqsVOhEce
zk2RhC%^Rz~hD)va+K+CU;pf@6c|#A<`7FUT_RdRnFyGJ`uXZa~d;kBi_mxpmuJ5~B
zMI}^pP*D^R85#scq=t}??rs5*5|L&|K@b7yZs`V5QGo#jDQS^VB&54j^4zc6{fl$<
z_MHFs^TD-TEQXnR-{*av`@Zfg{=JFC;z1=o2gb(ZYYlj@IpDZi2sE77uR6ch%HQ8<
zGUh{g9w!~9rnvv{kp0_R1s{r87A_5oY+7K!U)Sd6P5b-n_}@S6xC7tL;)RHK!>@1m
z|Na;5n8Bj7G9$CR@AKb(^B+ot|M>Kt$E2ZJcudaAd+I-a`v3M{X%e_j(w|L@nEuDt
z>4wH({C%>yj!)LVx!?c3kN<foG~)1K6~x`Xr2W@*(wM;)RUBK>yccV{Ym#&1Q8n`G
z{#QCQHd^Wp#m|U{EB@|H-YbdxQ9Y!@(RG3F&#!ITj|ayv=_bt3jW3phx0T7bs_pHJ
z`4~ROLddCq0}9qei*Ii%<vw$K(}l#|)A$qVbtmyvC_QPUqnxa*roO&xzBUUmEZMb7
z^X#^ZgioT%Hc87&o;WaKyZB@_YL!J^QxAHy{q^zx{%dAq;NnJ_NCp1+-90mc9$^S1
z(>q`<cYfst%&`aWA01kNn4k?e{J}tQNo4N@A1~h7XuL}!SAv~0(xtv3(M4RQ9akJb
z(nP<55uBh&G_7316eN00ELt!(Ji9o0v;X0Tgz?#t+ng<VQeVDm%w^*R*vy(TGh2KJ
z8mw?B2V)hwlFipvLxV`AJbtxkdveyIJ^v&dqA-rfU|$w#*@`AB4H@_>?3EaZ8|A#}
z-^3aXKO+Cn8}`3_KLs~<L{x`jI;j7=D?*Q{+1$465S5Q(6GK*)_7K>8EFwP7Fm)UH
zA<z=f)sA?2xdF%FfzLU$>~MF<1S$hvz=?Fg8d%EwRi}dv+%UZ*F2o|pOR!jJNO0@>
z5iZNV8(cs@>xJ;nu>09vBHe67j0Vz1#c3$!rh>HM7x^doZhpCQB={{vC+vSIv$sI9
zOQQmCGs#6o{I71l2k5&Ae5eQ2br<;ZJ1VMf&Y7ODR~vB~yuf|u$ZuA>3LV_JZk}sv
zgunK@=lBxeH^6b-*;BW~ZWnBJmJZjJo!$Ld9wmAc@FV+ya>pWkf1CM`tV&=4zIPF1
z7X9f*R$i!EuI2-0(xfWG<35c!`nnA%!Ot3yq(zR}H5)30$T`Nqid)_4y)O4GtIe1l
zy3Qx}6~4VWI9W1aGV{9@A;b3%uc{1_{Nl5JenbTeydd5gdM*gY$<SM(3zmUn(M*+K
z-=BQV9JPy-tx;J`T<7_uiYq$XZ=8=RF*xp3b<pz0#>XZy=~YHgFMPP}{cZO`c<*&f
zlKJ0kuR^0#aDXm9qlx-;oCymVOd#>BS)<5f6Qz0YBJ$+@ut<a(*fp6aL!Y$^{Z?DF
zY|R?bOW!RiR!BMDUgt9FL@$>*2|8~s%-6XSD#u22(RD}13Rgu^_7U!NC|XYOxOEXn
zGTew{jKX*D*$^2LJ_ze0HHq|qW#o3DA7JkGJs>ExGcxkkrpBoFlcxwSI}naXGi(vr
ze-xmxZfv+Zc#mPyvnnW~Ew9CavwdVV#~T0U6h_^@T|V;T-Izsp%P*mM$v$Iwf`+FB
zhGSp6xvNbj+3xEmX3yxuYp0@@Vt=kPsj2H&^BvzHs-t`>>w-fuc}cFD@N9KBk`tZV
zt=l=%b8CW^G+^nP-`h|-J?3Voom|$3z3d!5?moL6Lk|<}@|*cN9!O7VPua$3*S&c5
zO<CQ+dq3AQu(WVpu~SPSX|AjP)J-2({l<L)TcO_N<rDV4lXmh6gjnKZZ=BU8rJQd*
z+_=cI?^pEUz2MovZ)X;!FDOmfTH@M8E?xfZ{xgmOlZ0!xl76kCE_P4E3J6@}yAG~Y
zK|EnO%iI3Wlb3F`xP18rM$H2tnACwjqaRqTqc9Gd7I}pv8ZH3W$qouido?R~(WOmL
zZ<K;`tOl78F2K`m4=6|lY499yt_05Bho^Zrcx%~r$a$8<5vUJ4lp|0jSR;ZyN8o|g
z2BfzEi$>!uL`Y;7e(;;y0?)p^afBvvif@jM3mgOyFa{xPaF-N=8m=7i`6h=H136>D
zGlbbSO0_7|d4$g6TblqwGC_R<pL6{{WE(}~@<_a0FHn-0&%-of$!!lfI0N8aH2|>K
zA{b$W-?=w~n75a~M1b_ZNR9&n%DML|pZxam(DBRPch0p-^AdYic>q<({Zv1o!3!Yu
zwpY%SS@sHs&)bH?UM#>o$qv!^8HTEuuwx@*)KmBaC&3o2A1>wyt!&)b(O8JOu#Rhw
z`}WcRq}ozvv!}mOi9C4+Y!+sfN$0hx;YrWyb_d{E=72;K&g=uTbOGU@dFV>=DV>aP
zx?0=<64PO1o6nR>oB^wG`%S>N%Mfy<CAd1v%_<-621648D<V{rX>f;Y?{0_IkM1J(
z@yTt(;OZ@?(n<lKt|7;cg<@wJHtuC;nx8V2v(N)HrU5DM5Cw0NkMKN&UB@lISzO<w
zXo^?W7N01Eq&)95I;&%-&k><MRb{<3`&?fXJhZpqh~3On(lD3RgAeL1_(><^(JaKj
zuMT%`Ox7d%!+Y&XiA_Cx&oh6hHVx{}Bxy=UF;E}4D5*P4C9cC)-BtrDB_~HDA-zKQ
zBH@_=H#wusj-@{F^-~cScEHqO#Y5t`O;P1`F6j8b+0A?wBh?+7`~*DX;aPVGpWH2}
zd_0F+ZCd~>;KojHy#Rw*LFCFcHC-^miySHkB7w(FE{$B*CS^&+p4L#nYf^;2jo9%%
z0C(XBFlm&wPrErZnRA=_C!+FC#ZiRxd-p%kxBjYhAK?Q<o$LA`{nsP@Ot&XWSb-KF
znTo(IybyMq3dVZdtKe0~b=^ScZfanoH%1YFE9#wFd|Egs>!^&44=#6wlN)OVKLI80
z+;ktF&?k^74X?^}ki-hn&z6Ea2ix^xg)TU^hK{d>jJDe+{wRfc$;%|J6p6crhxOIw
z^A2dQEbYO|OVWV;AVw?jR3`g;?1hF&M0NTJm{%dH>Pq!se`ivP=MNOlZ|UDOu@l{N
zWuP#oCMo_QKiQUuAaJ6zhHK2VeclnD0C|jBpA*=b8=YsX4z&6#>ZN(2R&?~<;g|W7
ztHx(7d}nyULA0}vtNO7?d{N@VMbNFR+{6GjskpN-zjv7<RR1oO1T~5M51x$D5`ZYt
zFY(tSoS7Z<d>mxiQ2YnGtDzG`6VZM$sqTB*KYkQo3prETp=GCt4Pn7V@w!7vpcs8k
zcNg~=qh(Q0k_ryd=?fNm@Eh)-ei9hgf)CKCvf5WEIixc(kB$N9dg)D^0^M~_^}Aj<
zQ$R>u+CYL2LMM<w($i)q$E0bQgI>4Qwc_F2lRL^d4`n3NjihrbO{+*oAyR_m=9ii~
z&-}RgiAXGN(_Y&cwM=b=WEX*^>uG(HwcE2rl!?wWIhuQIEo`AiGvtPb6OG<I?z=Y}
zob?v}2+WyyL(oWFkddy!*zB=4tM7i$q0`B2YIUR9K#3&O@64WCD@(=dnm{%~_jr;f
zLzLvr?nrW&g*?0ASnC8KC+oIINK^0}G5+plHUqmZAhoyaFB-dXb$mHVDS+vo7pK6Y
zUkVB>SxBwNL~9E^=yTz-n>rmU#CDy%V2Eq&TDRkIk^$Pb0-M3QnUIO-?L2J8;;ETn
zzh&y34|sw_-E?2y0!`mVTQ(tiV;~#WIK!QU-!?2w=;u|h>Hf9G{~$m_Z~Nkod-PTm
zmbjw`DAN8pco~&@H025Q^|GLIFgf4N=1$-th)r~}T({!V+-#W*CY6b8$jf(GK0#33
zVk?iy%eM7vSWCmzQlgAwqhGw9q^bNFv|4w3!jj%tT`T2TfMFfVkS--j6gv?}UvSfU
zw^@H56?^<uBc{f9o%L)~hmZr>`joLh|9J~>r9F%M;@h9R1PO$YYc39KQAe#vLj#Hs
zoRG@l;B;*GhR{XQ@fIdOm0k#SWmULA9Hm(i!()5zE^zIfKfY$y)3_=?aG4ks+FiV#
z-K^PI9o+>aR95<J+v_N$`B)o*Ui&w_+iDQp52f$tv`W%@mxZ+$+_8D{oHJ+LZ183p
zTvFD&1W_!i69T?i2Q_$@Y0zP&8V(7-;?$hgFdPA@W<8NRjcE(bwZa}AknpoEbtUCm
zM{9KBa6b1vy7*}FgVCY3G#XM5V;K}8w$y}cmKKyuT9o5SW5FGG+RUxRUa{9Db6fFL
zGXUByu@hb!Wo)E3j-kYv^gU$m3{twO)AaPZuuUgY;*C1r_9`bJp}iZMwWoI2Jn#T8
z(ggb|)Z&Oj4pwq)uL`!O-TidODDUr7IaR1o+C6qrq(CVDpA`rshY0r3jbWRGPGNL{
zem!G>jA%cuwIev6?<X~Rg118juf2J}_GZxHPTTRh+dNk<cVEKufa!`|+dVW}=$6o+
zCo{&Z&1U#Ednz%33)(+ih)VhXg_+NaBS-yw#;|M*-A{y`U*HSdPRA+XdA$5w!!((D
z{iQ;n-Xkm+_Kh(sF9h05MEdNnrJ{(A$G^HnX5puX#)KZv*7;Zy>`e6xBPe+xwUiqg
zjjC&CosVaq-s@2gB9FCK9pTMie$mEampMeq>}uXNW6}~cTo;Ner@@MLJzn`vBabbK
zZ8BhBz1w$ue1J_qOh_x$%+K2<!#sf`1Al+4ZqdqPM>{q4)xFjg9CN>A&z{tah)qp`
zo>9rcJDm0Ad}xAp&Ijl9Fn7oKd)AY3+VE~UGcem;RE~DQt4_x7)`$C4Y|L?pebN#^
z@tR~_*$Q_O5O@+h&V1v=kTS>6k4R@y-wnWd7heG#U2t)SA43vUvMfMsZ|nWLRzCvi
z>Pzvk6u;_wc|7!%^3=zc*HB8|Bj`lWeWcm?%#WS>L_Ci=#(4Jyo;?I`uVGibGhSr>
zP(p4GQ*|=#@i@u1oCMq($*0s{AC_TQiy~y97XUVSUXIoxkK*7Wdb>%cgz7YXeNfB3
z_bl9)Sm9?1>bvhiC$Qb|Y#Aq&6wO<NZVJ!Jg(}9$%3$R})GW1UQ2joHwpt7p!U5c^
zcYpia(0-5alZY?I1NOavXzx6RNON)7_4Bs+tEx$zW1o<ep|#-{WMDQ0e4u4RarThO
zdL7AjhKMYTo7o;xQ7sbZD&MC$tP+cs*DRf3rm4Cd)z{qR;m2p&>KnpJm0c4wA>DnG
zPU_6A<mDu$>4E3=c2F7IJC}ne?ejG15udki@2=ugv(`E&@3Q(L6m)m?VNMg@)a<B)
zJZD#{^>dhQzbgH_X*GlO;liutnCl`sKpEjK`o0KL_8|3er@}=xl=kuAVnWp!-OFSr
zyafG1OVrPWn~%Eb$?Qb@7MdBOe(~%<-$Cuax7PRgmmR(sJOE3o2Tc(SX8cxwI=oVN
zwg3?-QscAsR!Ip7aD@51mT6WeXSASjO~4X$Q65QO(EcJ|aus7D*jSDfjgiS%n=VCh
z9-`2hmNfZ^0<XevZ><)aH6=cgHOFst#J;PE<$8%Tp7(XP%JY*<z+P(7?!npf5S^Qo
zS(*ZkbJE*<A3i(UC$`WTTEk>plF4E?bO1iX40C$XEI<T(6-uoB#nkGTCI~v$r$0Bu
zHoVjkQ~Gg!@GAoN27+BNQc33Qrjw9#l?GfZYm9AdjtybVk<i3KUj<IZ<w%8C2v!Mg
zq6*lZJBL|K^l}miRQfwoL{_6t8l?zBL~)P{B*^Ii<gzM|77r6Oqq?K#+1Qifxvf(<
z4ux`9nYD?!x8bSmPlJk;o9Oc_Qu@j)P42H{36AHh+E$7OQ^>9A?093j)?C+dJue8H
z`oU>xW-~4!G6I`+If|Uu?jfzI!}*$5^?Djl@0nbs%i5zJN>Y%uHGF*X=F(0ZjA+!;
z+!&uPJA>pQ0h`^TIq3am0U8aviLdMhGBg@90~wByJ`e2?UVnm5q66Zc#&PEPtB3t<
z!s1|zJJH(?L08j5j%JCC)@SO|Ln;Kt&lSe-E89nWxUNn)(|M}Ea>wY9iI*U9AZL}`
z^)j`J@A2|mRr(P}GLuZ;p^W7vSR|g$IZke;&{xqy)T&JMc~0ih+@|tCQCB(lYvVzN
zfsb^_bDOwx>N@OqN$v6jH?g-)D&zmT0sK7>sPG3M6-R|q@&80Ko|6%%2rtnsH$&Xa
zbe-#Ytxx12P|<r&?U^8|GcjBrkOEHf;aa4-k|8njPO0qpSE*sozu6e=e?~fHlah5X
z?0a$MU5ckTZn49;P)@QII9cN3!RjdBC=M{JWxDmkVAwSg`zoM2%+1`uts<lYBU}a@
z0H$0hmM6o%e1GU*$3rhaTR`%PmIN=bC|CRk(9Yd4(kF6C84a@VJW%^bmvssqq)MpO
z-Rj2jhEN=OkXWANgXe;plEfKN6eaY$@h$_<$|TTBuq2OjYj?H19OOvdu??J?K#<y=
zr0LUyDm=kQ*Z#b7`EA;|FQ-|xPf52P_iJoPGsQI}Op>V+Jhvh6$+eX0%J^P{VrH4c
zkv^5;#z447)jgW+av)vmR=MvbfN9ubWjRNCSR3Y=GnIW?wf{i&EWffW_0~@NAKsh3
zK`7lxY1EZus16rEsI+ldKC%G0PXXUqLw-MO3wpF?D(>}bbANsa#pZcSlgdpt9|g&%
z>`G2rt52iC-}6i+DAQ57#dBJy8F4GD3XgptVr;o1n-lZ!DIrTvNa#e}i%fGjZso|<
z6V}<sEvWJCuse*+xH7;Wx)GjNrczmR)4vV^uM(S@?Nfh5ojt~*jvbFCtK{^~!oi(e
zRn=eO_sQD>pF4`!EH26kRVq17{nq$<=`Y^BWYK5r4swMHXa!n!_g1^v9+pXNhKSrf
zak;^eO2g;!N}ca=$$chowBHXUFbVfpMHgL;q74hWIp{Q+I%%NPkY_Cyf$Hq_9+aHO
zSUGFM#gJ~5p%It$zESkGKz`Anm~)YtAp5eZUB7k1kB(1$MqnwEG@~p_XJ28GD@Zu4
z&DgB7(Un1N(4Ma%p-=K~Vt};71j5ZYt`~WkO{l}(%h>HA;}S0%O&X8vz4s~UiT|l*
zLyFx-F(Ex(iN6rb8$yT23HJq{6KTXmik;78&syJB+=tGA9_C1TA8dM*Kc8HEH|`M;
z{oV*jzM6G3+1IV^B27su0FAdNBAN4K?%ifLBv|d5(9~!g#a%tR-inbEJ73TFipPT-
z?=S?gdW`HTT@Qd=a}P?0yC&B>qx*^<#)kUb2m+L#T&Fd2MK#zbC`iv~*hgaIWgMQ(
z@GZY`saW;6w-aC6gCl+BPIixDa$O74E4AOvHVVVL8Z*<}3!mz$8NNL+&Fl;i*J__i
zE_vy;ij4cg{$2$)doulJ9JYski=)UQ6l>S=?X}Or+PcQOdE9tYspcu^Rhy}!xhT%=
z8^`ShdvV!Nltx3$OYse2t5EEfz?%Nzhg0s(8r;@ZPatdQkYHBJEA#5lnT4A&_9v=`
zJf1WN|JYs*yQ6%yCY>`K2aLU@s%92%r@nV;$lWOL{$3QexZ1%Q^w&lHyB@>S1{tJ9
zUupjZ9?p0^s-?fR6wRO$+Hjc&BH~RIW4~83MZ9Hm*4sriW_S1Ki>{-jB)wB87if}{
z^qg*f9t)S_WzaH?RNdu>HMV0(=|`GeWg*(?WN+7Pr2HQIoZQ!I%i9#6<ShV}^L~(V
z5!N!XXFsLSErlb+?0DlIw&V-V&jCO&*aOc8e_pb)++Ij@j;mY9X~n_pu$Wv~0w1)_
z*#I4eHTKH^#Ul1kCP%~x_G>wlvQ}Htaj@}s*oI$FskvC}M;Cd9h~O5d9;e^?wK-(|
zcr=nM_#{_98%<n5Z@ise8!oIv<fabXW!UHE%`nXQ*_(r#z{I#~e{0h?dG0=pv(^B;
z$&5V1MC37FQLkzGO-q~7Ebu(m8V<zmk{m24rEOxfz}LGeIeRhsKrA*h9+5Q*)G}2S
z^cY?DzMeaa2<%o5U_&SPdQLEkgAULero|L%sI=CTE_P40oSk@EuRN00qS9)c3jU6B
z5fZYiyk;2Vw=3hr@-7eV#Y(+UxrF*`uCE}7aFlw}RvH2?O-_d4CUS^T7bc}<w0D)G
zp+ZW3(`*piTxjSMV0~;D*HNLr$RCK1UZQ%)3F0HIsLDUDfV2K(1U=GRy^6p8n$^6@
z3B0a$lTtRH3iSD+M_wyTeL_6cdJS+RQWXJ@_dHYZ+NoGCkIvI3s*kuOsjy;yyL_Zh
z17b<w4?_H(bA@b?!z)1?APiyc40YthMg|b6qkp=LbqWX09DlH9${HHw?M}sjo(3S{
z97J1x@>T2#UM6(g1osayadP{JP4mo-4AiRR>F1Pts_OU1_?=&qswX1Q32gndM=fuC
zK|uX&2d*!79-G3XH8yzL!q4yQRV6Q8KfbdLtAG;(wFw^VP6150X&@Xa2psa-Nt$D!
zTvY4qu50=Cqgbfq#j!@xE4=BGG~9^)yxSK&0z9vigo?TQ5_4JmXnsmwB#P}vN9M}a
zOm`pNF~wyNKt2QKs^o0YtED*g6{8%Udi3ihHq&BTnpzG3wN##OUQN1y7c7BzD^U_*
z?`VNCmQtDE95Pm^V==Vu(KN|OYInRE{FE$UN%2d)uoBOz)T+v?kRiXHHTt{K!3QUL
zSHwJYLX3ftx>yS%`3E=5kB3YMv|%0bEGDw)O#q=GU~F2(!OY#Fkft@zX71@*^`;;2
z)7i;+1X*c~=VIjyT6yPG?t>Ww!tw5Q*hz$gWh>;>)`GqjRhw_f3nk@Q>E#ID_b_Z=
z(6X*OR}o?Y^t#w!^|>;gC<Ho9y4VZgu_>4q<Zv?by*_qVJ6XUDr!E?3hEd%~pjTPS
zvM(yRmXal^-badd4Wyg*XzJ|?A+1cWfcig$WnMi_J<g-gNTl;}#H8T(BvGVxGAvas
zhN$X=E%v9~ca5lWx+?Tjhpu!Xg1)qOTRiytKy3N=n&)bDa^)p}Jn3Z5Qp|)tR2bA8
zzi1D(w(DN2pXK;i`i!G>cN02wx|0$jn%qcd-Odf#54*gF`9wk!(I#4}fFW{KpeX0T
z!fys4f8aF*KNp?*%L@RRid=Ta5dB)c#}BW>q)~pNc|=5sZ6PoN754g%l@>0Sjr++g
z%gRo-lk97V2LPxwL=y|4>?OMB6x#u7pG#Kf!ES}))4!%3R?dv(;Ji|8@b<xgD-S_6
zvW_IK3Pkq4LLHkc>DAMFcy{}u;IZoDl`+l&$eaoF4uG2$JQg_n=*X!hFK={lN>V8Z
za+TdqxV74!^#zaLsp9<fUGT-+E<@{ub%+XX<z3_Ng)|-JPwMw*K%l+BTxSQ?XCRnf
zm=oiCDP}KGdo@%8;cofajRmqt&&U3NQJ1{E$jBQuaTU4LPG}0Zia&3e$(*ZEcgX1n
z$v`swzKrSxzG+}#ddN|1uI=hN4xB5G-+A4Sr8yIQze6iTwm_@I-vJ~Ptn}R_a~+r}
z7|?zeeKqrmz2S0rog!71NSKh6(M1zRIYQOg*rpQdSCxi^CHKb2mEV|Kn(IoAovT+&
z5=NTI8+>6kuArXXcej+dLLYywNy8u+;Ir?Or7J3LYi{=5%f*xS3;Dns-Kf5k)Kj)0
z^fkpgLV{r~xjM-_nZ$>1A2?p^+KJNN$>mq@$ymcJ7Fi>WaV!K(-91~&qoZF#aXgIr
z!o@l{0-G}HIbqBkmdv-8)Y}GG&p5m0mrw{RaesSQ1fFbEd}TnpRsOMJdP(W@t|Wf;
zBH_x5z6@#71HDd1A|J6{M}L`M&*2{rnJ0B|K*7!2KEIu0hXF@R5nG<%aa_BtTGPqj
zO}K@Q;xni5w%8sf_|*(}9us|!o9Eu~9NPSXqvqN1{1%)2Q9r(y#$>QLVekFC41M%V
zb13fGYQ}j;9&nTJXCU;g_@DHx*sN;^eQWRg;s%(rnHUOff~tj^#N)>bZICrsM~csU
z-SVqciqU9~HK8;~dgAIKb9QVENvnCa<i0#&Hfi4OZQmyBnK>RGO=lvU@fw#-MB-2i
zP>_m8V~8Cei~M@569WsYyMfnLlhmV|u&SMCX}eb4xU3Vtfxc{1%r+@LraCqYZIXK2
zWH2lN%e;13hHX*?J&tU2TlZ<^I=JIVKh$n>aLCq9?qs@t-2=fb_o~8VfSFvZ(1+j(
zzc2kTVVrLfNus2K!e>(Ao?B@l3;GpDyC=dW8QcdDfUor%vM(}r9YBf<4^jI}i%Yx<
z(){`L>|(fxRUW|q9mqyo(o3$ykAAFuz5g}DBq=ok`{hLRAZ;s4K}CxS0pImm<_Rq$
z3ok}NE}5q@4f3;{cscF^nBAohIqOD-+$O?}uV-4RpTA;D9DD=D)LW&N6i4{^TCGU-
zOCepJo8(+3zxpN8$U8Yd3i++UX3t$O?m~C$olj<RF@JouESt{M)GgZJF{t6ma%&Ri
zFj?nMGDu=bB*-1n9DG3HmATTAm!QvKAm93SGrHz_cSxeQV^oivYZaGOs=KQUdZLBN
zd^fZU>$VU(BbuR^dEynxeCz~guNm{}4RfiCg*|8eSjSG4^*C{@BT~yQ%hnxyx91sR
zdj$@3Z%y`zmpMR8=kba~VZy~o!wp}hp*rWqxXt$+c5P{uv*?1=mg7=x$tFD?IDa!L
z7kVCskO9JI?f-(56;KF!D~xfJ3R%(8REG#kT0ZBNgYckY9mO6;*N_n+`hipPD%Wb$
z^vM@&U)yk(&NFZ29m}&N*w0O|T08W+ezzkAs3R6KX?`q!RuZ1aqyQ#6Le5=w@(@16
zZV+kC^V=PhPd}geMN^_XcbbwsC-(i)@P`2qFpUy5xje@`ZHl*|qS!Yttk{sgvX$pn
z|N8Gc>t8=T4gGb?FXGl8cr$^IYS<a?rq|>$z8-|xqMQENg%Y3(uJ^E#J%NU=qXVD5
z3yDcm8c@p~52Y5*Eab^?Ka9*Hsm4e7$-!3F@XNAO?6!b<;~;QNwq6V)XFT!63W#{x
zdcaVNWlwRP{dT_ZH>3YU_&4wn>P-#(C-GnW3u^T94j=I6uqQgtyd-?cZ&!Fr((fQs
zCUyiWzOVQ!4d`FLI7)TJU>fSt1thG`Dq?WlIcpaB@7{MXtKp<DD#rWWwO2Tb-0A8o
z9Y_ECYwtUy{zcxJC@QWeD}+Nx^T*x4S_1{$e{r|aW9>2nW%jm+M{mmn+3()v56|mx
zi<yp41pmc2`|bVtZ)(bnKd>UouFCTLuFm}XPV)c1?O*rp|8~9q&x`z=;@|YGEQ&Ir
z@Sm5z<By*)l9j!{lfQptZttM1Ut?}7SuEkNo#Joa=zXUgP`Nq<j=Z4!bCGmN;7=Xj
zgG4Qc<~U^P%~TY}!D!h4F_Z%khOP@DKLk~@jZ6Bv0CC>FLLH{YHUxiM>cbK4O?+{}
zXV!dK8-N`>&k06b)30u1rE`9ROz++fF$<q(W*L*UV6OdrQUmztL}rzS#%QHzmSvT;
z{z6v1k-f>wzgwmXlJE>NN`{yHA|OgL9NOb%4Ha`El)VS@2QVi5j!<F*39Em4hfM$y
zt7k$@{V9dJF-?ylr_<pC5HU`Y-DuRc`AD+>C|CQ}3_lXOGJ_9Ke(aDn!&&kU=`&4!
zWQ^a=XfQGxev()s|J>tXe*xN1S<ZC(Z?8<`9ifNJT#2d=2U2J$Q{lU-G)v)5nl9gf
zOwloAt>P>PrZz2A%Hg$ZB?Pt*`na$)8Z7U()tvZk-B{f6dTw)B-Y}n@ZlARFKR+*I
zhhYaiX%I0*{R?7jB9KWI7(hffNGG-4^BIhw%V7-O3(i@NmBgqMNL15*P`#$btqEqG
z*JsATB|=pxKJv{dv^bkyRZu#Po!j+`+}^J~PWcY(X-dH?;2rp-McfrJa?b7?{S@=%
zj`SYt9q>Q~fHS9*ctOw|2hPZ!abH*>OV1`D>@ISsR}h+tEh53e1AmVxUdR%jGK23f
z!OVKbE%R+f-?1Oye8Lm>6~KV`ct*zuz0u?)tvQ7=MNbx;?2Vb`|9Rb;(!-9&84;SA
z{Ohz~q`l%9kCBzUp~f=(FBVt@46F;GmwPbxrMOUu>U497NYeJTci#^$?!7pDYXW8;
zb8yo7e`jqFz2=CqQIWoh*F?KCtAEf3aQkrpUygbJ6{J+#vZT9dqA!Gr-cb&G)2dnY
za9<FO^tyja@fXFS)6J$hvZ~QYgHylR7){S2AK2fZ7XR0VzlY+U$xiEsJ}BB5#zBM+
zpyW`c*<y0IcV(+t{zSFDyFh<|;Nfxq)S05mNJ}24Es>>39iv#1z`CHOwzaan0(j5v
z#P!#Bf4^(OPoMJw>XX-IoudBhA|>FRy{SgEVTVK)4qR_$9))=)okzhcfNR^Xkp$xr
zRsA;2KVC|nLwen7t|~=gNs4;_<w&|-8w`elgiCR9O(hM@y5zcXW{1_d?;Bx3JC#e$
za+K>L#)PsNgs;4GQ}#OEc2uQr)M<`GB9A5IvCYbz1=}hhg1qu?I#o<h8)+y$C67xK
z{$WhGFRc4gUh?uv@zb}OIjb!K-aJ%MWUOv?^#OcqnAF<ECX$9rT_o<-Z-PCeY0ky<
z)U@7(c(y8=ujw|17h`hVBTri3XoFG&$1ivLE-AW5&4|P(mu^#7y&{`a@bDt%t+zZQ
z?{?#+rzP%SSf|~)lst*m#4VE1U+I8jGNzuYw~~CDvYtw(S#4dPq5m^`3;RHOWDVxj
z<u#e9#}~#&V>C_Jy{g{p+k4xeTj?TQPl+jwy(?Zh`Qh@@Z;8!49YuGnvbl8V$O95d
zV)qmi4K9bx1W(o_pL8Q~>%k}x+E2T)=M>K2@2L*mptLIp4yzZ#zT|jjiy0z)=`mu#
z@GxX=IBl}-7G<|4J`w+=A})-6yu}THqzdWXAkvtJznMXLHUTeR9h=$pOIebwdKk=1
z{4Vj@Wgnbz29GL7P-SOaf-H{B7jH*{_{zZ4uoQH^rJ%-d2)xYG>3a{cQXc^3(ljEU
z&96j4FBTxe(gu;Oy8u_SAHiWN5B6Q&Rv)|c4m!ndUoXVd`oNlZ|Ap`_;#|Jpun6Ps
zeiof~9pu)vXO^$#tO6+2V+l;5oY{w0`2=>CL_A8Nr7uPFa)?g=!uYI!Y!ZQI-=VRY
zLoCT0Ti6@8BMDNUDTo&uG1M^cWea%hy8_zQTZ*_6uAWoVsI~-i;|H*XIf4Yr9-*rm
zvLl#DFKg9qCq(G>%|oC00J0%J&h){fKefHNs1MHC&(HFK-}!C9h2>M1AHF|SslSX%
zU%j%mGbwm5J^3DiQ<5Ndj?1{|EGzey2j%nq@NYs65_Wr`N@PASX&)6wS6vOhXv1Kb
z;<2^d5o5aF9FRJ(2bXE42yyBWyeT=41bsRnuIS}3m+glU>h#$WO`|k+e&Fi3^82(g
z!0L61;4SJJgjvb>lM@$5?)rv-D5@Fd*Z4TLXA4}82axR#HpeOOaGHaTh+qHaesJwo
zt)aL)GVRt?{k<9Wq&kE^M!Bt$4`P|dEu(Y%toB8Ij;Y8xM;~?u*P_G?fmUgs8~T9H
z_9I*-q=BT=Qs4Oo)@Tvj6yq&^+`(Je;#7Km?Y8ZXvMv{D^t?--Nffi@2m(Ag!u{WI
zGY#bIdFXr`9KAf2R=Fa26)kjj5HI_gO*bF018`__HMOgF`&xexSc&`eJA{&T2-t`N
zB%@Nx2h+RV=^~4DZZK=_ad`-#iVwiPv8{B-3`5GIH3nSn^9Zk$b^rYF3chvRMswiJ
zKKf>Nep?8A|52fu*ij~c93XqKZ~P~reo+4EOxg2V4<0R4!R#KLi*V;Yx9ipt*^u_~
zB5hw9=jJ&jy~R~TRT^v`f722ZfQ&jhdr13wiwoyEcC=@;rr}84Tvgi1Q-UI50Fd9g
zh96!;*Oo@}7muJmBm!A%KbTRauye>_#v)K{gfGeIUmC3cE?s}t5#Y!AC7AhFwGKj%
z)1DCLd7IYzshW;;T_bKojH%$Y((5+fSy5;kB)VQ2OMAG3km~jZwcB06Lh%W+Yvypw
z-@W?W6ihfZqYxzQ6Kt}S1-O-_Uo5oRi=hPSu9bdXsR%g!Ckw3))WWGf1=>3TF%1_z
z3TIo|Rn>JP*KvMSS9cB#M62%1RPC8A>{!<LXM+GEbniPw>?5=mjQ-s!39hG7x^GeC
z6{bH!{2Od2&&R%Jns!i2B5+EDKu5vrXW|ba27+e>EQ;@5z1=RHFDr4e0%mZU?F%3+
z9y~=NxqVX>9aASHbK?P$(Acy3<GO!|h&|B73fxGFo!<QDBiNL=JZb3!&NFFXb-BZH
z?`l<}@U|H3iB(%ETMJ)u2Z_69W*9b2y}}V<mp?BcKs#3)z-HB7=8v(-Clh&oEW0>p
zk1P#sX`=#UvZp752pm(PKNsEGUSD19uhk5#pFRNhjH|76!pg;?ifuThWW?A?Y`(8#
zbu`ERIgjPxoJQ|^p?-e5WRZci*H;r{T4LU{$)3Gz7HH1Owp^e;s%4uiVynGw?%sQk
z5uTdO9I`pKPX>2|rP~L+B|-@Xc5yduy`W@T?n+|Lq>Tx597>MnqkGA?YxRU+mQZif
zIlX_~8c^Y^jf?Z_M~<CXvX!@x=>@4#<<KZ+gPo;tS$-i{@n7jo=uWyCMT-{z9&+BM
z0AXw_o+EjW4Wcb;a=xkl#P9eX2kp~VRdl2WkZQKv&GXmxKkdj-l5}ESPVr7%sS<#o
z#DYfex}3=oy1u+5pr9Tk3>wdrUofqlP0Z!Y`bpI@Td0LVs|}?BS@uqs8c`eAbZtpG
ze(cQT4IVgi^+fn5lF-`8%yZ31f3YRR2Dbsqx;{{g*Bwf*#m_g-MBBE3xj}j(bHkS)
z^xe6*nWj@D4w(}lkb(V9(O&m{OwHyu@A#zPtyw<*N;<PSnbXs<IcuA7+d66E?hQ<K
z_Zg1=W3}<e;fymK`hDSl>3&Qf;ho}NQ#cTL<^HrMxJK$v{u*%yD%Xdyi$*R~;to`0
zno(_lW?wIQ5Q@cbE;C7bfO~o?)z?*0m~E9JrbuGY1%=7>Kw%{44U+0Ro20jl2UB*5
z7bXnoeo@zaG!5wv#dsT|_-n(t_aR!@PYtTIM%)dcu0=#TQIk$8WPUL!c0hOQKhuEG
z2s%6bk4f0b<nol5Tci_6`(k}R)*`GmKT=e$#1w7HAoLv}s?Ax+?lY6#6J_27vD0wK
z$@uBEUQ&)y=)!1<K|<WJZ4*Z8+>NZR+pr`-j(neE(aYUk7hrb1`?UKgGN1Hfz=@T=
z<gKTd7$;pP5tBwaNMx&g16PU7I3Sg+>L0$hj4(H;SMQT8?W8TlPx4h+!E%+)GuyOR
z{K#{r2p40oH@%)xkqSW(44w-hF25KnE<VH`7{6h^iXmGp6pWToNGL3wh?5X*bn3+P
zkah3%p>}il3HrL9c$>nMQXhTB5F?3SxZJ}>&oKQ1bYPzsMJ8Cd@*1J$q+a!(8ETE)
zF{5V=Ef=+3Mq59xm$YR>uWn4#J%%su+}Py8d`}Losxs%E#ak<hS#Q9fU}r7Ra#JT@
z4y#=?^D!c|H}k~`t?xzSbFS$dU0KeY>!ehRX%HUbBiLO_yfVEbUpi!wYaH0rVN*3)
z_asdD&%*a_nN`^cm~-y1B<HU}wT$*kb+YJru)MjY^?7BC?o*tPn9TUBxWk@8;+O%0
z>QoiowgKNYwvb*%Y+h4z(;hqbF59(3vV~~&r|u7<H}km$ov77|RovNyk77?oJQtz}
zyk7Sthj9KbAg_8`RPB!?bJ&4j@vwNQN%Gz{hUnGw(d#QAs?+H|=&W2(DVi7h61pwq
zB-UsR?%WHbmp)A4_rr#Uz?`{1b2G|#n8@~=baPQK_8jkNJP)!<*vMus^$#4pZ87X0
z0s>)(gkvT)N=d-cQ9y9=@JqR%x&_YeI}Jd9-_>JWeEK8I$A5hl%!n7HnbdMBXbU<I
z^v+WI=dNvcluL|g`=y{mm~N1}E>#bfCu17xhWLiDL<X}yJ*d`masJCoS-g+Gw+kC)
zRd9{Yn+b+xvodI!4oFb^u{uwsm51_?{6a^Z7%Q3NL^Bvn7Z^HN5NAsz>0RWA4V!FK
zpmm3vTANHXSvTi69DAkh0O(<mSQun*l7>L&GhzL;na&`f$&uDLy?of3<d$i=iqLwK
zPLCRc3Kf>FX4=b){za$U$N$`PfQpj_#w{-~IVb<@(iKdAANLK0IxqXa0TAFu-@kQ|
zfWd}f-%ei794#8kra<+WCOdqRar~olxbWo(>GAyw`IPlE3%Qc-(FRXMX;$8-D>L5C
zMsi@B|C1nhy+U!ix*omNOjmg#1RbTUF)5WL7H}SGkR@!`69K}+NW!cOtW4I~co@m6
zsZKntj;#zn{w(P3252V;fkC!9Hl5e82Hd}$j&PKb+5FR@o;NcgqznX<s<fXu?xd`f
zY8CcZWI~Z8VTRq`hEd~c9Lu3^5H&sZL8*aA&AiC3_hMVX88KC7H;JXJjbtdVI}*Ku
zP?LUHc=;=Fv?<Op@y@Jf$$L<W5alwRWmH|FlaDU#451Qn735g7Xw5z-D=LXtq&}1|
z_yHl*a^x-V1lP_{*%eNd)dc0?CUhiF^A=UkQ2dIh4H~iAUd>50+$dkd;dyMQJ=T%j
zc@(1yCRe5YcgK4($ZsuWy_jA`FBjH}Q236T0l}W<sK>Z#QBD*mqjvLN5b0zc&u;)@
zl_vnvH9br}|EohyB+2&vE-#2{PZksmn`Fhl{T%!mMeZG3tFxU_aSa$ld*{N=s;LiC
zSxD94pF?9pT2Y!^5o4v&g4^wI?&O&op>~0!M%!H3a>7SFY770*NcG-V!$KM|)Tj<c
zs(G%o@J$~RHTogpu!(&GUvf2xPpQJ&YUX-AK}W0g@gvDrh<K!G5j*oPGJWVbsIjzs
z8()kn@Ry)Gkv>~|UdfTln6ad{>n%BW@yMr&e-yruEt4@RiW?9XQ!FVKKyQU}<08<k
zDTD{P{2Hk$t;#;3(Y_@)%3025blKOV7-8vbRf3Z_>ZphFZ5LIb<WpCKuey!Z$?J(9
zKXp0ASnjkJWB;qLnwUI2_RwBEKbuH2P79ioJuKD1SzbDuZq*i-8(mb@`k8NQ9<p9G
z?(GDCblpX^q(Bku$R}+6xrbBWM<CmFjSiAuS6dN}k!ma4RCd^zW-a{At<)#5jeI65
z0KLye$){AyV!HVI<|uspAW;!iRDxZnB<D+bCazdHpj6L>C1EEnvew3MTN|n^BM$^=
z8`lNs@7}tc_wm*^)qd6{B8$aO7rJ~kiRJw35MchsL(leI$ZHsQ-A?4SXG}L$1V&10
zUPrMFo=2|8thk|6!AZYYfQ^t@A|@Yu7J<!akxgB<bG!7Eha4JhzrtTvzn&`9?&l;<
zc+RoAew$u=iS80yI0JK<!+aAlM!P$vKqFbB>MD0!)Qb>rMl^ZaV9eliKKjv{koKo%
zs7F3bw2yz4zxpmk#CLm^I*7X_AWG!ISBM$wERywGs!QuonM!C>)n{M)q<|>mLPEY2
zbuDsszqx&8*XL{NnSaCph1L<a0>9BM`Cp}o*cp2DB|3v(;Uq@PJd6)*->`hrhwBMR
zU1~-WihfVfg$kw>aY}Gl_}E5f@nU)M7#PsGP4}Lxf3^*jT%xnql%0m%XM?#CoCf3k
zor|TEE{vrgKUbzXiUHwJ`ZvyrkC&n_PmoAz&M4Jnv@`#(H;FlO-r;rjBH;!p0Mo>m
zPM@D46#pt{P~&Y1jZsjKj&17{XejXCGCJh{ysY?7Ykz73Kh`u#&2vgVR?dExce6*F
z!Ar1}$me})BoG8iSoc4>GqJw*$%O0avs8o<jE+Z&YO!~ygXxNVN8E<myfI>ZM_k8p
z9?}n3-{lBTiWU+Ed|ebK$KT%<5%#K5s_Io~COc^au-EU{a?hOs?9WRoeopu6Oism*
zVA|RNJk-sp+Etj)&$Eqt^qCi)2h>}tihIGNJ;b9ZC&{iCV9&V4;&IL3sc^sjbZ4Ef
zYQ*XEOqFK}!{A@O#9Ualzl{^7Shs{NZBvgpybpw3NH;R=Ac?I5&7Y^CC?(QPQxbiR
zl}2Yeprrr?kbMn?VB_XDBa58RGi#jw2_f8nS<j)5GQn;liXZb69C<w3fal#^Uw>P*
zXZaeC$;p#k`R+21SRe4y`ps2WYJlk;hLN*Kcr=4>ABIeQQ{dK(-;Jg?C!lGu1QbR&
zoW9(>jE%VMHzFs-JY`e!AY!HGsi@)2)>fz7wS=d=x-Gm#Yuof<ONj34Huyz91<Fm{
zyNjBWNIR0+pMQzx`i#<2)}FYR(|q)g+JKZi#$+Z$#WMKVQX>FVPwQgPz)K9>#$U}_
zs`AM+oDfw?M9yL?p2q=|B67rPq#4Kn3^>WBm+kut*!`t?`fpHPB?#HAi_>ZU+O5xo
zC@|pfU$0X}9QnG6^s8<R$2Cc`q$zl%;DqU=-u<L?`Q@tI9yv-tJf<t*b!^<ngyW>@
z(*dv&tP>2^yxog>NaOcXn19;<2N!)mO(6}upEvBMdDcY+*VZ1}Ld;<*L`tABQ4S~e
zh{XLi@2*0EPxy?>yXcMQy4mDDh7koT*P<EhKHJXtVwlEfAA3J71yWZx)39MCKK)>F
zN5ezr`cvH`Ndc5J+oyY-lG-@JSA@9-pg8Dpc|{WMq;#=3<g|*BT6l;|yB<PdSIAlZ
z^6<k+qLr%%qV60O_c+!~Au)v#xMwejv#_v;nD`>^`qs}Nwook`^N~xpS94YK)INdg
zakj!FgCZz#s&p$WW;weHM>v0cIS}XWgM0YQo@Z8N)*^uAjl|gVnz(2}^4wDG>?QWR
zP|U^1K8u-er()?9i*Q9HAkGfZ3&+zL$*K4bXLv9|^tf?7GKs`s*PhufF;65?gL3%|
z&DJHhJ3>7;l~^5yrvZ!b*7^ME?Ngcr0bC$%Ot?CmUIqqUIe`!;xZ~)v5thEsD4+7A
zWQb-U`n3AaxDTn`)MkD9^*9nB!RW4SRvqwfgETL!jNFX)Geu5mH-HKHdPfm%SZ`*{
z2gbRg7&J{u{11kEa@lPylkjZUtfF>5K0%*Kc{bNkj#0lD8IT2JG4SQVv<Se7xqQv3
zy@o6qt{EGP2UG>0fLfJgERB@s27$*!kjh|Y_;)5Xk<?Jp*)Zq);y5VK;i1`L`5Yg-
z28b8CXpVX=A#*wm&jaC4uYsmB%l%PL@X&UO)>RfiE8(!obTH~-iUUG9F*!XuFfLgu
z)|BjkIWGpB9vrbG9ec7nUN7;0q+kcrS}q<jVn(_KJ(}|^?bB~6-~p=~fTYPQWIQ%^
zO?l_DDeH+rk;<AKep)E$E97TSO~IZNd`fWlA*H#>4^#x2hTzIsEk(Bzl>AKb#?p<S
zkvQWv6nCy1KWk_mw7O&r_lrbTtB?8csCuBUD0_s73Z&@ATfT`U4&H{F>|$^`TSTvI
z0aT5q0DJg>kr}yv7CDoF?^eTncj^7G=<JqVfF7@R?*rX*@gg3p%}OEN1me+(yC8Q}
z%xwa=U?1ZA>%gQm(pg|g65x$J4O7gTqQ>XAu<gu_96~>CpgdDP(3}s*;{Ny{u3mUg
zL1Ga6dsy<CjDmRT1MD_Hzbi5P{KG10u#spKq;QK3I!VXFm1yeI`mrUDI@<NlhX;|-
zX06o3Oss~{E(g}S_L0`IJUlJ!XJJ(~O}nV5c$xH?PK<Zz98PKu+7DkL3A~IFD)u3m
zw}+1~6r#2z-_u4hQm4q=V|0|uu2ky9;#kM?iRV47Du3KFrziw;RuL#+O1%bp6Ma_Q
z!~}D%DW^33)^3n;(tSwcK^(%mRV`d0PItL*lAjd_V~(-gq~5+rq)MNQa-jEhLgGSq
z(>b(L5MAV_zOX9_hFsPJ7Bb>K{#BqQ#Z9RhqYGuF$dn6|*+1>jqD$y*WvtL_H%B%l
zl8i4n_PweTACg6Pc3tk!yCm;i#HfFH>=PUCb+$NoS%K>lM83j%apLiWqD}%9o|_-+
zPLPlLc`gj(JPZ{^BO_}zRlSW1+j?m;kCjS?miLv34t_JX7J@~ym4}v{`9%%~rXzt2
zx;8up?Rs{GH5~|hJ(JDai5OA_$&k}Yp{Hq{KGt(x7CR6rb9ar-miHpV7AaLQ(k+?;
z5f3QztT(rBVK^ioeksuA(w9rcCootudfSwDZSO)Q+-5Nlqc*#C>HAXF+fx<Y{10QV
zN(V6Z+M*mb7FP<}%y^%gnZ_+45f%|KRBR-!A1&M@r9*Wnd8W&j-(_Qs$}BwTB=%2g
zn2K-|V(PX07)-rT51_K(MCgk;eCW&&uKKtO-a804tb}zj*x*lFudk+m*?KAQU|E|v
zDRQI^Tsq4gtqu}IQ=Ij@vV}XL{3Jmn3YGlJ;ifW1sEO>AAnp--);`HxdEtxLP*BL1
z7qrCdgD@TUhtka%XXvaYgST-u4b&<+ZMZ&|I}(w#LxID6Big<nC}DdzgskYQi`Whm
z-|9;dIW8z(Cp^m%4f}2l>N;a_ugYWc>$<CYDZwRCbq|31@XTQlJ1OD%VLK$-5qOF`
zl93#AO`4O-ry^eofO}@n>pkJiVFG<j5E(*Yb<<t|TrR2_q?xm9P_!0ujF`fSY>mPf
zHH3N>qU7%;<}OvyA-pnwt52z7T2&>BzMGl3;eV*@i1#B7OsuaYS(UonFE*K5eML;e
zoOIip1S*u_IbXTq>3&ZPRi6jpETf+vNOI@85)i|qI3ZfPGXQ+dkb23zZpu@&;h0uN
zL022Mtfzu$tA=<U<6o$5xlhFhZOfev6CD}W15XWM2Rnf3vsT90dD3BdW+%b&td0VB
z#K)QYcZ+izIA*ym+JF2f#|+c*C&vtPn1|#j4++`UEO^^<uhNgr`I?ysmqw`4-nW@>
ztN%(o*XE6%;~L2sEYgNa=UII*CZjSwC;kA@!tm{c4j|FCZMFK9IP81d|Du>_UcmfG
zF?&6{#vU5^DkSw^46_=)%<*r*18v?Y#2;+pOcOs~S=%B{(!ZHubKAW|%&^NuI|kok
zeJOuY%ozWhV%7`nGZGUD4=~QMnJDhCWvfo$U=QF3l^l5$`9fsWR4)%T+um~HO!8v<
zit#3dR*cAKj3NxP7j3{WbNQ2DrX2|kv;4_l3^TjJ|D9nLzHRKqE=Uo#N&DJ{_Js{?
z+tOZ64n?a$t2BvbDOhZClNj-k7;&#U$JjEfn9$TyuN8EGVb@)X55k;0!CG%8&af|6
z-oH5<F+>86Fr$+hepZPSZzeMR@^!#rIl<4KB)LfCUMVztU(Pt0CR!A+dB(VakRD?+
zZ=^M<B{&k#8p%tiVv;k9$q~%RCoTv?+!?Br!)8uk`>e|CrblGvE&=xP(g<GU@Y9N{
zDBHa-zOgNKsWmp5+ayVU&so+b`w0<=7BA8Gd`n&~$0z%?PG^0jsEs8pQ@L2dweJ3O
zQGOz$CzhK{wO(|%t=1)HcFS+V^IZ&`tcjPg$yJ2|xdHRMg54z5&k6X8R_j#z3moma
zZQkKYksC$nDhmnpj#c2Ou(?ZdX)%7Thx^Ih!A(r^QeoR>;#Evq+J@`um1!9T(snrv
zYs0AmvsJ=`T=i9jMZ%0N{?@N-r<{lUD>|>0&y7@ZP2>6(IdiGg0V_@+9(QdZvj6=E
z^9(d;JW*YLAxT1hcs6@%dWQCxV?QBgX$H?LTd$Uy+Hl95my@tWG9=3^8Kz{<zM`Tf
zPhax(go~zE?Nw#Y9tJ((b=2CCf9?lI(NMlv2QjJrqPLKpJ>-7DO5uZ;31ZWs!r}?S
zkX9r*t*xe!k4SuhUt8;)E}fNa`jcBi6z5o6i%Iu#8dhqZAcP5FfpDHUk{|6_WX2{U
zCjQSpA2%crD%jYIPiDUg^wBHNJ6u-2r!D0Zit{?JnOeovFi2(J`gP!(6%jJ9W-Xga
zeJy?sw<P=)K2pZGwPFQeXGn6(089@fXU0JCKLci7(Pf1_B$d4?vyC0BfByBqLsbI<
zphKwVOwu^=CuVQThIh8G2&OGRg^!@G$$1fQU1&tm?<T@UyEB)P_GhODPFodtTY9h7
zg7~BUnPZ)80pD*Xj(h$u&QQozHtP{lE{mT0E_-eyXbEu)`#|%H2?rlgVIBUe^16uV
z@!woELgcE|u^Hn12B`XbTF4Ct<fS|KMwI{am;V2E{qN@T|Mdz>b9&PICkQk%zj}7S
z+|9-YlITXs?=Y2J-i*?9_?lo$Ruc#|VIp^ADgXJNBX5>XiL54HzJkAwK?Q@uC@f;t
ztyyvn&R|V!cY>lB-$YyDELfUAw{0=m_r@~*^ID5TwDtoYJTD$I>ol#Oq%tBc5f8K)
zWJQ_1yWiCy^Ue4^^sT#DXFnk7s`i(VRFHwuTjA(?`&fciMX9Y)SKq<A^7o2w_=F|0
zz-F~1$#nnod3&!UBt$WW{u1&fy|VM!MUMVQhmQ2a9zIuo8uJZ|HjN6wmchP<#ZSHV
z7R9=hWaJwo$gF(O@%+{TBw|~GwU~R`HAEPL*!&FjY6O9p*AhH%c$#gG9PSoEQUn(u
z+)c9JEzDKR^9svy$;#Ab-KK`Bs$AB))s^Q>2R7!ueolmBupFGB9FjS|*^VBzV_!e3
zy^$Jl%?7CpgH;WOfH3D@Qm)k;y>lv0t$d2~chAOicG#19V{VgZ{{qg|MehJJ`T>NW
zEkr2G<z!KCb0Opmm}{58_@Zso07lvft9t>glkJ)EUkE3zNgJN~pK{8M5Z`)Uh<_jQ
zB`b4Ik@IdDL6QPOP0>_u?X}>8Nlse)4Kt-&VzA?M%t~~KD+f~rC2U=S#08e>*O=oy
z#$an4p*$4HSZoCZGW(DDZ~Ag;`SRav->&#;L;L&nyc-M~<U#TG6TglQPZWW~cUpO6
zRG_MK8Tb+5{wTh7)MMoaZS|)w#YVarHHJ|ZzL<5y`FwE4fXb@{vR1x;6#HDYeu>Q8
z?a=5)cph37y}FLNgErD1TJ<4usZc&yfQD!v9jAkHDn_b<nUm_JJBhq{Cmp&jy}y2`
zFwjvWR(It8;?#!%C{Fs3@cF-PUfr4h;*KdJbajL~_5htIvZT(;-qefOq#dN`7k*##
z;OrjKq`~t5%cmzpHkP;#WCG`y|Ayf3Z;(la5?<WRMZ5M(-y9PCNO*x^E;R_)qWRij
z1lSLw+_572LEvj97hO+DSV0~^{hK)}U`|x$6Fj5Z$&p(H={|n#bvvJ+sbJkuHXTY&
z7s#n>IlEo3+o3k)8)PohKAq;CEt8U?GA!|e=`|$?FQU6Dx$|~mEzN)UJ_yz#3jl?<
zsVRP?I$xKr1UxFgx61TYGS&Lo(i;qZA_C`^Z!*~@ef4Fdt5q7!2%b|WG`26?DeDoJ
zG_K)@bE(b@StRPN5u8cxBTViPjp<EUq!R39U(;M?sN-q8-JOQ(lk>FxPP!<r8!Qey
zuMVqaf?2k)oS~Q0E%U<Mw1L3%645qr75@5=JZL7MslK5!q?ko*Df(+rU-W>d?38>b
z`vYlu$`vVti+%LQRZMk3TO4G^zmgA%dwQ-ID_<ZQZx~98J1EOacQJT&*mrANLDy6H
z1!u2fyHBNenMkr-OXSF&;>C6Gc}5=>-Y1na?h+~OCCjIXEPq!xA-p~tWOfil!};sz
zo$-DTsQCyuyU+L>L)t53&NlGE-~-OL0?L|C2$|J>rt9MX^kfe}Oeb(-cH^7XB9aX9
z7HEtn>&x&WYY0V%K)e9@nTl{DNNoBhOvd;K%n&^gCE*(Y)%4J(Jv0m`i>mgWS*5U$
zZ81>eLo{$k{PG#rFkV5}^aem;8((&c?&T<AD-jKh&~U!F7a#h^K86oK^E5!jiu*1i
z4XmvC2YCf=Y#J`xK2wer76Y>nar6@;c$k=&7n}t>Dr*Rt-hri7{X7gw5$<{+VhPWD
z``Lxzoo(P6mqCtEIefYy%a<Q%t6RFPA=lk*Z)cPB%Q1cpe{y8YpxjjX;DF@UY`NF1
zhP6HZz6f4SS;<GB>nU;pHpc)MsN(RAfvaW<|Kj<UYe{s$CjF%bx?9scNt8$5b=hYJ
zY-omWP;CF7yf1U{UHApDp)rtEErN@c8W)s49t9mmKo=|42B{3OEe6O=f=Ly!FPuRW
zh-D9E)u{Hk`vEZ!+d<O1Hv!Zshv%pZApku9f}5?@C6q66)bVJmlE-Pk6tfU9;HuhL
z;G-1Kve33gk|meK{VL6wR7ai8Qg8jFn@O`)j4jjSO7P_SK15{g$;>Ld48Hd~G61&r
ziM9_ck~=P!fhS4AM2M&o;(gXf2L=uGw=Y?a%?1b80mL+a<G8Rncwg|rw4%m%Ci{mg
zoGn>+IdPXvlCYWD5VE6W946;~ZZ~I$(1#h}VDTaxtj%nx;uG^vj2L!w85~qb_;cAd
zATCpTs;qn}HDGr~!9vj`WFoTX*<d+<2(S3S*I=)V_$sdJadEiE!OqYERIxTAI{>)z
z`lR=ki!MF~ielTMOX;8|d_oFh$wjL9ow#_!3%MNJ@b+8`fR+YE5tRE&zFn6KS{*f}
zi2q$x{%zsHC=RFfxP}tRpF8GrIeZCB7tpd<k8QyntGFXMk^<#P9db}VfH9RFicRu0
zm4ONn$@-CQ6G_2fy@_a@JGU8z1H3&^FTo(-Q*`u{&{bA~00`lx@wn!xijj=Ca>gHX
z1esHN)nX+jg?%~NxT)u(Fc&1ha{)lMSq=PYFq+(+EGnHKqr2rL8h{#*ctp3Hj3<w3
zka(9}35=a{K@w{#kl(=VL#GxNVh*76LqjU6bWs|3bRLuHN<O-P<OUTf(~A-?b?N)z
zb#JzyH&FP%3ao0vMXr`<@Tn+*YeJ_tCF-=_0XS5IzDj&ZjlWOfwx-BRk;rKg70xD^
zdd^Q>53J)h!5pTUNCCi2IoA3U?P@F8Z_rJja2=P=*!t<!${8(#j~j2RhhIvMzK0u~
zXX^jr?!AMm%C>h=MUbGtCJRVLKu|IWl9LF6faD+<0Ra(FB<CCi$w@K_N(Kp%BqKqB
zWRxhN<ebqrR=4&&zjOQaxwq>5@#=L^UArlo&0cG+Imh_M7y7{qeCdz&cK{&?Eo2Y*
z`{r+oUnZDBWIGdlMoWm%I8v4kM~+yCYF<Zd><Xnu8SE10cYx0{H=KbUx)|pxOtGLR
zNgHJBv6VC_5H<@P>HJbeH%{gWELEqC)cHc+Fod#_t+>9UzJ#prsE9TAQ{SXg5;{Hu
zHu*S{QS{Iobd51KzBa~@$Wxp=K|Z^gz}4vuzl*q^jYTY6>%N5jCi=%_T>XA6<EqRG
z;vhn(fpE1$_%2O-68=c+B7bw}jV8ydsx2fc(EVnf`lk_6w*5~!L)DUKwwG3l0q#vO
z;2CS$%$6@wcwHZTQa~)kx<NfieYtM~p>K-**f(MBDVWLFOF&;8yT)fw=3(g(iKT`n
zu_7y%atPBwtNKYfy;?92eTP&sd*u^XOH|h4kch^Rs3>WVg=G6@)jxk6iD;~l0L$hb
z89%(`sC$7`-FK?Oy@W$Ym<(R8{TXwAi)x;7isnD}c1U5v3Z8hrw7=Ug&~zYvVb6ok
z(r+!2B}~27U5D!VNO&%(qPAUHTyzrME%-D=7!_EC-60~9X4(u^u#R}rBhB$V$RNW9
z?>ftpzVK|g$FYg2cafup$^ui?)(MZDf~hgEsorIa`~;@H3>buQOQ@MrkKJ*|CIOo+
zh>l%e;-xcgeev~HG0*c`rjqsT!aUg|gtA2I(oE&S3T4j+>y<}*9|{TvaJNv0yk*FB
zCsMw_^icj?OpLkYZCjrpFswt#fU!e<s8+ni<$kv<&IAG5erlGDMs&94*!Sp(?#eJV
zcfD?re(8SOq97@UZ>28Css@)SIH+qjwf>d+8pv%$>mlp-Ikp9GCoPWSV|hG2Q7Bb%
z_UJx$+U<O2BKf9W`1n9H0gHLbz5N2Gp(lrl@kl#*Io&voqA)2kqMW?eU@%rIudlY`
z7{AWN-KXC0^8Up#vilrdr|+6(yel(T<Rr1A>35Hc-abyWD)_#T>~y1>)tJ7kUD)u@
zO~8I7r)Za;TYC!?@2p|}srk+Gd1GPALRB%vZL*$^PWoNfm0o9!)V|b<jrk^3^#nFd
zid}{7H^AzYC1O%&UyPmi#&%;MioL;v*Y&P&<C;<`;#QamcD1yJ;6d5n5#K&qXy1*w
z)uA;jr0i00VkB;w(a_$U^E@blfr#wr`i}dc%aCxx%eUzvV~zo4e3EB>7m9yh|C$j+
z44)?a{WG``p1wGTDwD!HSv&3^YuL)*J9m)K5eIlZ`Z<FHFfpOu2oj~!zA)Y{e8#V{
z*(Cc<b1~zSj_6)F@2f#%T|%mVxQoq@1hFy3KG$OS(_KswFUYIwmLB3@PiY)vynOQs
z<80OiI%QO0`|Wdu$ts_IvR}2%CSyp@%>8Mm@l9c?5WIw*JBfyDVo&ojGhUD8<V^9N
z>S@$%D`hy>usAI764hBlb7THAPrkLyUEdv&p?D|XZ8vWG2ZBk6VwWL^?2<liPs_QU
z6p`@|l}r3@!qmEFRbEf^2)~fo5aIr)Xs?t}mt|_yQv0w|@)YxzU8gZAM%tLl&YJn`
z7@jpZ_Zpb7#dXBR-*T~d89&-sd;`D|#h5qWv*MqmsA(3vD58#II!XxLSSp^$*T^=d
zTyCgzl2%-d?BQ7}aHnFfU4XbC*7{WWt~j;?&Cl(UH)Q8qu611dxm{C7c~gtku2eeD
z?<L$Kk}>|d-!#d;FU47sp*3mLRw4Ywp;AX@R4eNB)OIWBt=9JD#di3EKb4Zr=p4n_
z+jz%W0x8AZQiRWZ=rpKP!){1zzI)K@aNhupDCol1OK)KQ?%2{uW*+qL(I7`p(B^}<
z@0N;lR#I0~Lcz#HFBHWRs2M4JIbr9PYYB`v&F;myU0pLLbkJe>MzK^<l=s}P@HQJ6
zx(8wws#E9pd3*Ls`11{3$Mq>4l4HvPEsJkAsZXv5+nO3VX2MWx(&cj`t%8-RK9PMk
zm<$Y_^SiR0x2M38o;dVoabQgSL=Mbe1~skS`NCy$z3UOXO=f?@#cU!0<Ex@d<E<}o
zvgNL+^(QN~?s`{!AL{TC+F1S!3{l~zUgkE?a7o*}dugRprvgGC<y=H3Ww-!hO)=g}
z1oDla*`1>fwsL&A9b#44OB0z!J!QRBrC4`A{`LZp`y$t=>O$y!-*r~b*a4fFXKU8%
zalu~cgjbV4Zf8x5!_|h=c^|*{+nt5qGx#Tr_}+eS`{#W)m>J+j$9uxR%`tG$Vo<C%
z^=R=;fCQYf_kg(}$X8>Ufh$h>%oDnORW0B<H7>GVG*ZmH(ASn7r@v?uOZ(>KW9S>5
z>^>$7+Lr_m`D<ld5GuL8JsE23D=tw3S&H#`gV~#N&lP5(6gckXqFqY4nMF^KEUlWI
z1t!vR5zH^4MzAeIGTj=qv<x=u2CMANZLLHbQZKYe&B(Lme_hB;iCVbkHx~QQ8v1#C
zR}Gvuk}0|K8!YPUWP3$%AH>mwv5&Z;Qu=BJ;w^u6@2NOy_!&?MJT$z=H=+6+JkY1N
z-eTH9omW>|885heX;WU?g7w}D=0WXceIj&^>>`Ex22L%zbZ{dUbDCXvmlLiJMFY!1
zM=XHx7+0?Gt`nvjLiG~vqXzFarx5vd=3@8a%U#uw4@!E(GWTEbAt?kOLh&p9I__BI
z(g-Cnf`*6-G$qA?v=|>(!3?Li_dawt`lgIVD!}f62%I*Gjhh!4=$b$Sp=L4)%#zJ}
zuaw*UvY@(cJ;x75zVYfc{LO^GI~h#>d5@R@9cL!Sb=p-*2rN&l%~fs&&aErZ$vT~n
zD!LDFk{@votLjy6F{cR__j{7LTb?Jz(J^{^$uv6cj*+kn=;&<nfh@!@*RlphQ5d7L
z2*QI(RCiSJGgzIdt=`*@!^lJ<9~7p3a(O?h0?}(c*RYtB6qTEGhR8gZjhD@7n<Fo;
zU)sG7l$UM9xzEMv)3LHUmr-Jf_JZ7}jFH;2AWCEL7ukYy;dWzU?5iCPqXp>wB0Ds1
znuooesPfAKlV~2e!KxD(SU)yrrkQ?k^M7{~q%|P<9#a9RGk@$xhL)F<3#OE21pQEI
zjvv53ZlRf_#~9WuV7l5_5eWy8=)r_HSa_{q<!%HwGRXUGfLY3*T0TB=UP*$H(T*W7
z{=NWRgRevx>sIkTv^Z8*-u&T!^}T_X5lmhl_peZFy_Q+OX4Hr-W>EcPDK_ePjr{%F
z%-Bt)IhB_mTNf}+M>Km-U^*@4{jqt+dDn9mGCsb_Y*{}T4_L~V0L-D0?i3NB$TkHh
z47JJyv6Jsg&xHk9(I2K`!?n3dQo0v+M6DhovCeG!OsB9aa<?85kZvHH7#)JJ;+m&O
zf~e(z9NB9r`V*qt0+4CLir-*>gYk>z^!N03uN!jEP-I@z2>k=D_saGXuafiwyzg_!
zt<4}`Ms}b=@!$|1gHTZ8l8L5(0k$FZYn3<k3%R4DiE&&;z;S}+kb9Nta;bGC`;Ber
zHkBMnH|wI6iqh+lCRa51z6G=U`9HZ(DR+mQz*6jf&Yy?nM_1!a;Oq^#FscglCq^!6
z7a2-yLv<@nSUG2s_pVtxWeHF4YXpUicVea(Xs&b&d^ne&%~4uP^vy^#PcZIo@WGUY
zz;<VyQF*=XwMnta2)<O>JM_b@-u2Le@-9L<y$#~kxs^Mh+O0vc*rpTOYHT@zOPi9t
z;*5EIm@9pkr}kV_KPaPW#8r>F5o2#LR%WFH!ecPY{Tx!f2X%Z_oBcdV>BlJ6%MOLu
zlFE{YqF;xLq+P~bB3th|A?SjKhg?@e!q-td2qpSRNqzTd9>k-^pP|KY;eK<onq6~U
z^TSguowmxqKNEj{H;o?Pj}wpGk7t&HXAXa2=4Q~}FbTp5%%OEaC*CpV&=Obi!Ie<C
zLPv2$C5Yr4JzAf%NpnHar3{tJ4t=MLWl(4Cqe!?|cO?nOk9+nQOMlk7JY93?oH&|r
znOqa@F?(tb|A^Y(bD-HhN{Pa`p&|Jz!5fIp_&@d6$w-tMTzUas;g(isnU#^L6K?cr
zEFC~PuS?3}<XAc=XJY!ZfqD@Og<-p-bPfn1qMAwPY#ZvljEwg5b+Ae3%kVn^`rP@g
z?SP}%>(+F=*Fg(<<?(k=DP64N9obvi(haoiT>zxzx<$1Zncj9i)o09U!9LAEtqjR)
zdI1jNfj;ZGayFhQuAs{g{G93Z&_lAmj(@B0+dlK;aH&xj%j-Q~a=?j&WD5Qy&NKnK
zxFOMPmNTR%Z{sddV#Th1A31^NJ1GKtwhspJ0E(=&#GW%x0bXe<A^!!NGaUyGw4qFr
zc`$%~(J-eUxDEQ9vrZ!$-t}<2i!n@|`!N%5N_=Jgl-v87_wfNst)kYajq9M7X<Ocg
zMzMos)^-jf9zg|S?`E7|^xl625RC$H#^F2pQ4MJD==aaCIrjTdH9lxGnHi<Ur;9Qg
zw_BJ#Ahe$ajaJ~@HD&x(hl9<zRYX0Cv}J|YO{zaIB~^kpNJ|$6jwMriQ09FB5aEHO
ziti6ac?;d`xHaLuCpQW?O!FJwW|wTQVA-A{p_&;pKJr?UcY8MT!~GeOqhq8;uuEl6
zy+rCh0=mH?#fT6m9Q<c}zE8+{VgZO?x2hG_hS<c5piap=8trkPr=<vc$qC7csN2HH
z3wxQve&L@S{r#3^|2e)If`!GMkW^0Ug1net;hh~fPSN}TR*r=vouj6YW>t|mu6O+z
zJc0Mp82j(TeXmNhqbw!9EfCds=L2Qx0h~GWe&-v}vA%*^9R`v%G7N#oL@DBB=maFb
zcj|5%V3uRm;t+HpHW~n%+IaI1U>J=<39(#<=##+1(|EtckjA^->kMXMS^;!P^WVS0
zopeCe{9%9!9E->80iJa166@$a@)nstr**c;-8P&?>HrqhMw1-~S|X4v1%3)?Ihp>b
zD-Er(lnZUUxXX;Nz<~q83`Qh=4(sdz4hPq#R5n1n_yj2Mua-Vnm2#vMD`wJ6(X4Xp
z_Pqj3g`4eX2uB^(Dt?cP;VB8C%7$v{_$Eq{=1UiD9U~${L@H!r2c~6bhZn}aLlWL#
zkn@HV)|@}#*br=<rDDDv(v@bPX@narS3g(rlU^Y5twhkOeEGm{HK!Ip_t}opb846b
zCs0f8E0bp-w=@W!E_1fouTC$09K@{S0(8gouz#WXuRA!F`unYbYtATtd?=`4_oWob
zZ=+sLA0x0tFO4HQ`WahIB-qTU4n066^M)UX;*)xvHU@(RnxU&*Znl)*(%Y%UaJn;2
z!a{w$*){MPW{07hA1!KYD(u<>)-vHPwovW&mq`lye%v7Cb6Xa;2tf(_1zOjN{f~Ho
z0Bl(Bbu(FWS$J>nvRVE{vb@>%al#2w)_t~e8cR`G5lpPm>F=aH-9S<8lpZR|rg9i*
zSjH&Fagme9>c&lKc&in1?2Ng=BD^~7ESx!YnJ=-IW?*I5J&&x2%VW7huBZN?Tefgh
z<Icm>e@u<_M8BEliq5jnMx#5V@N17VqdT=Gfg%<Uo%2|y{`v_l?ko@1H8AlNZlWyq
zt!STZPD>!^2$+S+D01-u%y88=c_qPj!yl~|oTZDb$8SuLR3=eGUEE!poa~Tb@!^a?
zZKMkq_{HN)U#4>ozNNn&V88yD7s7Ker;Pz_>q_c_C)`SsYMX(*fg#S;L`wIe=S^02
zx_U5g(jb%U08wr^-V7$JM1<jYPLA?veovC9+q+_P)j)O_PMJ&$_mT~q?zx59Y0kcL
zbox1c4Htj&`&1fWWcG%_p?AHAAUPn17>zrW%KgG80OJz#(`cBJA4=x!^W<=r@tC%w
zs_|x+9(q;0bM2pL2|<L><|}AoSZZ8M6E~hC;TZ7|ox_+C*=scAMV6_wi5m-uP`C68
zL?IA=ZJIUH!QRE`iJtOvI-TSnz`-d_T@%tU9%oBb-iI0o>*^AxQ_sl@yc~)2r%)QC
zDYz<R!Xb!1gP>oI<#sN`AInOhuyQ7tJddmGgaTU>V!e1cQmb0e%+Z!Ne|}F9QbIMj
zGzRsAg{0X<ljI_I0P^O@jji>Th?;v?&h=zWqjTjLu=S=?s&xqO?ZlJ}Wqs(cXoJlU
zLSw^uQ#19~F`rYXgs=YyT;I|PDb{E(VMxhT>B*?;aX)8wxu&hLfV0p@v({y=TySj#
z=?>}~TS=$Ajh?c^Ed_7`1?o}j+diK=DMo$o9DlEqzXD?a{e?0$TDNk(Gp-T+>0dGG
zIdo`?=?ZGwMpuVWB_%*?2nt4Db_7r{L;t0;*kK6HoVR|iviN{YLJpRaS~B6+%sl6G
z!kq}FMG*xxwM(p#k$}@mshwwm<M3V1erLy<r*sHM%g3ro$T3^FI3*{JL1j&Kh?;~E
z+nrShh5O_JdrEQmQEt&~0OATlT}i|PafrD$XYkZKVJF(ef+WM$$2X`Km%y+2hW@^I
zOGpM9o)dUhhXw`P=B=<kivrE_bv<4?13#*ai4kcG+6+1oGjJ>B#AA&oBDo)g6K-4x
zyW0iXof%?X4_;Ln*CGC|nUe@9Ip-A%+u5W$#lc~<!E#@LCQiB|U^bhjkJo)8W9Jf;
z6^pz|j~BP?gz(Z&&_JL&B-u(<t0mu`JzcHYmt_rp85uL8R!WuBMVf`rBx0RoiAkcz
z?Dms`A4-c{)e&{{qKtH&<*u&qiT)vR^hct22EDRLfMB|@a;f&t4B1El&TW|%WAdjr
z#(O8sQ{6NM?6Y#1IaIjf(`;F3iFclnZ(o{i<fc;l<mh>N{-wY4$5GupTxjbnb1KTx
zE)Lzy!!XiS3(v~%E>BbV?6@BVWYCwwP_AZ7Fv+w;WKdJuoTzcw-b>XIii~vbjs}MO
z*m0BadFD{AAyJZQ|5BdakQWPek#~j?9~Et{G+s|nF_sb83&PQMbC@ZQcF)?m*mArw
zT$q}_r#0)gX_Xu7=BA;qD{Q}UQT_D0oKpky{6m~O{fS$nr*)foL^T_?lv)BaW|s6?
zvPR6cKFT{q{<j^CmJf~iINLvx<T;$ce|!<&qLo|Mw05h34DB>al;@OEZAb@<qw<?j
z+__K7UiKYa|9+{tbrJWIvpx6@1o<~~>#l3!VgaLhu$RBw>%YBYnkY6L^Qp5P`oG>%
zT#+Dcs_><MNZ2K>*K}!M62_LjSC>ndDkJq;L-A-N!XG=7N=SpZxq!qKAcC&|Ve;UY
zA2{w?Rqk%a!T)+3>V*00;en9#zyI=5fX`{RIYsd6Q$~nmVAEb?Xq2KCwgSW4uQ1`O
zsAg!s2g7u>5yFXJ`0S+Ev^BhqvxmOxqeRTNw*x*8o$`Vma;z~YAn7_J{CprG6x1s(
z-X$m@qkBwHa1<ehSB$C=eJs)_Lq^e2QGD2N6#4t2{&{Kt_kRyipT_xd-?nSS`1jL)
z4h?G-;lXOZdz>4@M7z0_E<pDqiOa;l48q>n4?AxE`AY`KPGjBj*m+~{0h(_ss-~@T
zJzU333h}f6%e*PXE4}vb8_i#-^S@uHOL=<Nzb41w_x*Lp5~;mF$M6B7+e)t2wNhBM
zgYS9iW`e}8qi#@q3V``I7|E&7>W9AsW4c=q$DKJU%=DkG`Cs4nQiIAa?q6Di|N6)M
z+YiiL5^peYIw*1Y4|n&kclr-6Zu*_V@W?eZz4ULt_y6=imGxou#Kdt}G5?1*`1?io
z*AM@H@d7BnKu@;!Hh?2{S@T7{b*1`Ff4IL(2Vo4Oc=c;Wn~66lG$eM01c|1?smfdI
z7`i3)8t)3F^sn{s_t&mB3!d)btbo)Pzb=*6A>ApJ255s)2h%n3b<B_e!?6>s%7^I?
ze`WCQ=^8uusoh$3-mbf<Tb{@2WmtD|bD35tLj3&cG4=D&ge6e$2oYU4Vx`qU`arR~
zaht~{+f-NQ4(xxP>``_??9TooYe|be@d@@Ton_g|lh6O<CV4qCz-p=uV>y!g^-;pa
zPwAJ?qAY-)!|?f}zQPq8UB1FNC=MJU8n|l~3J^<3XfITY9VD)t##_(PRFJQx$&?L`
z1!uet=t8eRPyWC-3M7L}Cas~xTNgS3&X5ABS%g+yKlA2_Z1#A|(I7v3qIa_p4T#c)
zT&1EPJKy@GS^+{|ReW7kUN&s=xERcrc7;QTO5%tbK<2^%RudJvSsNE^^4z1NR4y&C
z@4U!}w3(~D38m<N{$iDd2fL$q{>87KNqc}@1J0z}Rzp|FMO_^;ckF=7_y}AT+BfGq
z^KoH}9|`QM51>tA002M#W;N|I&mtsmsvoe%jMW4OF_^PIB03JE;A4F?g`d~E3k|Rm
z@XD@&OJ4+(h@^u7kMyE##X9V4hqUX_Ei}JBTLCc3mIUEkmA?>r8Vei=?k`DdMmCm(
z<KO6g7g0q`R>qe;LHW-D=DNkYFaP+AYLVJit1+!YGF4-(&T{|x1H$K`h=T_@tTXlL
z?^Oo&WnM@tr(mdgmFnGmqLqTDmRtjku&%@uH&6@Y*v^~eo_M~h1KhKxhGw3j-St~~
zK$bzhyE^ke&}jyZ1FU7A5T)JC77|4r8?96WG~AzG*?OPt$(I{X=@O4YG&2YJ0M=y_
z!?(mJa1Dd5v9}4y1S!tU{rz+B*DfmwH-nWNi{IzhExVua9s=r2p(|5_Leusp_#@=9
zHW;KksONjvOCB|B!taY=76))H;T2u<o8xb0lg0Xn&%cz8wO@vJwAoJ1XZ&kl`MVbu
zfpBSX31*yrpD5gDSR4MF9wP|3zdui#=KwTED;JKyXspOJdSfnP5(HChl*2C|;7;Wj
ze}{gWVqg*ufsv)gl?&Xafk7vzv1{z<ia8vKuGP=i*5<~&uC975txE|hGfk@HDL3qO
zPEJ;IS8%NUm<rvrt=)<BYp8|i;VTU3yI6zWJ!R-}<K6dneb=Hm>T{-ghW&ofc}cO8
zPP@duTU(<lc$Mz%TX@>);%tDg)$;%@|CIyLgX49##v_&mk4jQ(lkb9t_32;7rwj4z
zrN7jLRarMzj1H?<P*U7Cky@}b=sB&M+Kb28%#lhwj6Z4nP+UZSU9(Z7aiz`ep;D;a
z9UOxw|GU{Oge#3b<I*PM>UG{?Wy2NYJq$e~;b9d)db5Pog#y)^PufPK>Qf7w;&$5H
zDIc(HDr-vd>vX-NstvE#6zvIpXyEFU8sOY_Z?ay1w63tvCi{?JRcI`T-&ojfypQ>7
zrqVwA(Jr{lXBUL8iQPz%8YxhYTOeMs-@Y(>P`>}JPgE~8#H>=JlD(#N-+m+mFS)6Q
zd2;?yjq1M3!(0EjqbiR>&-rF$!({2tM<S8d3o2*j8{yP8H08kEF&TaQNztq;o@?I7
z7QC5bU`vu;o2Z>Ds9uc(BR(CAAt=vv5cbk2M8}r~x}EF{Z$kRv2O=e5I?$FyA}I@-
zz#h0o`*{SMimnn{w>_gO6}4vQnp$(;6FOS|0cybyC1#yp0~wyRzosEa6tmU<i)k<P
zh9l#vO`u;3kqbOb13$gHVAe^Ga8dlYKO6esngUfP3T&=4V6B92gYJEO!wq&H(r1rg
zZz!KEx6vIn4!LR?ch})L=$hrB`YP4Z1100o5Ex2Jf+3#6_wkBCYoK~+pRDwz`-EQL
zqF)nD^EN)T11x2%GJvuOsFga^jZSt)`s;hZfcs)S1!A~ZWEjQ`oDw1tYXF%`z*g%6
zj5Qd+ThN#^r1E-5H`_~^DM(i68NaF|FpXXX{6$mP;T=MiGm$`9SD<T=&LixXij9jI
z-h)8j_#|E2S57J<*0@8Pw+#xXd-j{)G8K?l)-T`xN{MC^puahA_vNiaoC#JU5d)F2
zemD`*a7HeTxxz(1H+G(&9e8@o@p18)7ec^b%na*>cZlkk*ZcXMVS{V2a1=*^-boQe
zdJNjYrZI)w#S%fh=T?wLNw^kxol};RDSj49v&!}0(K(+mJ{<d5033=}lt4_^>?|+R
zH{uX;_7-Q(jp`1~bu#y74Qz{$!*T_y@O-r+2x(TT`rle<maZV_O$&+I%g2?l8Df+#
ztAgRv93br50S&uAqJ0iXCnmkh$UeDwG94t|+*W3Kr`q){%qdK{{ad(_;{-yAF#oAS
zy3`2-HN!Z%$3o2}j-m|sJ@<uOiZqe|)hq#9q=niCny+b$Tv&E%@4=Yv5d_IQJ$^Lp
zN0vxsq}af7`U>aXI&~Q=y-p7IjdQy7v_EdigYS~ZS0DlL*4fr<_;(8mk8j+52zO%q
zNV8uEd+2eE=7H}Xns7j|Lr=1x>JP|08?$!?wo}p0R<9I3?$G+9I@SH%|Moo1qHl}U
zCj51P-Vb~a!{8UeCdtA6$<7saa8{M5PJ0P~PFfG}ee0je_8;vI3C3~_8*{PRavj_A
zt5@HxLMhy{nCrgnz-suA^|KuyX4Or_INP5gHo*8m?BIh^a<EHDfy<N^W-^YO9#|#h
z<1S{m?%G{E38TEmzT|xJZAHz~%*@WC%s@tS=m801CF1l<Vl8+gzKZavIyLEhj$k>z
zP}kIJUGw<qQfX|@8XP3_BIXtKlh3ta$EMbNPmIL03Q<7j2KH%rzQwux1=^FuCoAAk
zW_hAq9P&j~20{_vqtEWbGBCZH%9h#dp2ki>Z~EDwg{GB%*zH)5<eC-@wbNIE;Y?Zl
zQZAAMLEQ0WB!0K_^2iWOEr>cBhof4{<K1)8T3{D(9tr?{weFF6-gmY0OKKz2Y8MyO
zMohLa`OgXkth~-)D*w>(5#L}G813H;Z@G(l&jJl*<vlTfwxq^WFfP+}QUAQC_I?SW
zOL|nPZiFM$LaKSX`5+wi&oCjdPBE@CelUx;s|oE$lIX?legv_N`JPI{8&$7Z)$|O+
zH1jh;q0)dM0VbttadeZgy!@>=_9v{dT%@eu>{-8-SoJdJnOvah(k8KBQhzS*4tqYu
zl{bS0oO`zg=c3QHSwfCggS|qF>g|>Zeww1z8Rsq`lCng_oWL*h5_;L$UEy$VCyl~9
zyN%o?z-J>Fr>5jVW{TB|;AU&QjJ)6lin2cAAsdpc(Xp?osgA32N9Tp$G4Xza-=mU8
zwr`Yz#Nw@jb3JDr#6rfZo;J~-iQ;A*4CkG&0*4!;SH=<ndzcSZm8ubl(`9%0yV}SK
z0+DTp<A-8Q7^ronSm?%^I64@vw8n#_Peb{x@t2mwfCb^92Q{6|*$7X%!T!P$L+j}^
zL#=wd&?@hzv?GGg)YHh-R6jlqo?`k;_jh%{UoA?GCE7mQQ>Ndoloti#Ixv$y`^uxZ
zr<-{D!H?X=3B@UqbCwG|()q9ye+mQvC9~^dP$)iH9PEbOS##QYT(eMsiNaHhR?{1G
z-aF8ksM%y+JUo12|H2O#a+;~_^mzREl)Fbs<dG9lo@PMqM~~@;L-HPvW}U9XXPrLz
z{RY0rz7Jv*n(Tb1^L{GFn~8Bk!6R<J2j^B~^-dNgats@2(9=+6xny~vtPPU2_zB{~
zGr<DYA1Y~Ed|Z#xC$McU(hewxdkmqKQA*;Dkoy)d%5-hg4ZV+N8nwN=b>@kb8WusW
z^GFxrSBWglB=0YockPS?18oMEQn9ZN5}A{BgC*Vdol1)>(d!!8ez&@m!_|vt40f>Y
z(;IC|)cP__V0e>7CvX}`Y8cS@dO4yNI<M}B@E36{zb1|ywAMVp`Eh)~@NQeY)cW^@
z*J1$+i4`jX>3khA*TsqoW}R}Wc~5Ls4mLqeeGj{8ec6j;2d*q`2h&Pc@dGj)OfAn#
zV@Q~;0G4$VwzoqbpJ9qEYZ<QH4(T0PiZwp{GCWe05}XlyZ?v`8=-hn;)<GQFem#6+
z&*O?6%$dwi(!HEb%64$_2H8w}))*aQ5&I-2Vc%X|5$WqY)K+^fM|cx116vjvLLD2o
zpFbhG_rJpr|N5N)Mg)p@G%{lGYh^sdqiNkBAfL!+#7;S;)%m2HAi?iTw7*ZaOG}g-
zx-5RdFI9=6b*Sd-;cTUij<=Yj=q-7althtV_#A}}Z?0B^L16nS{WEljUIMhN{O4H*
zsIV|;<2@G|D%GBaZ~1LA6ezZLYWDh;=wI=2zwc;f+QNhzWzk6J82ehSqk;yi;~vB5
zEIK<IYH2^abu~;|or5gNc}#SE6wzIPbc58bSZ=fV9FQ~bf8rcsQs^rJk>UG$bXvZS
zGE(zQvA4E-7p8`BaQ$qo`4r4%Y!KJDtyql0q#y4dPJ`8N(kVjvR{58$<!>alHA?T2
zFzowSEOK}p+**$4ft>ec*4yX(&dI-^p*TCVm@4e_=%Gho)w)7RaObp5IhY4ZROty>
z=8gLeO=Ea~LtlsEkAQJ(jwdM8LK2RRlwPgh!1Gp@Fb{u;_xMM0@)XoxhB7Pn@EkAg
z8rNklO{=IkF(y@v=q>RSv{h52V4#vre#|E!_i3$G8-TZeXnxY(A^6#s*`}f2_!<3a
z*9)>(jTVwOdvj=Uy(1YLXg)TW5KG&l&eE4rqdx(Jyj0jRL&D;{-euedn#R>R!tv=J
zDBh_|!5VJ9XdD7^+B}+lUl|&?ohEzwn#q=+(vA(gx(Ij1w}J)|L<M`sK>`HQW$uAg
zDAM9PmnYpF8zzxFojzV)$7X>h6H@`YGV{o(!sr99qAZlSk^7uH>RT)JkEtCzeH_^^
zin%E})?s|_%L5y)#7-^`=gVKmOp<L^b=-D=4*qHuPa}SlDv7UurC5y^Q#R#;%7(*F
zkbu2agr&kgL^XhYBS-AxxW*kf%eia*cVik$5?5YdXrNgZyKMBs2u}}Or?qAJ<7rSf
zMk4qjK8RUDshM!p`bB9;Awy=YtrR7BlP~OLcpUr4VN945_!eDwOVO{Ku3$oP@#bu8
zt5_52PDQ%!kpxTF1}^il-D%&bzT=1F%d>9SYmCs%2c>if_)p>Q@<2#T=&?T@+@!u<
z5vP=YzDfMZER%A!S>qQ3C{F6dD;hd$!Y^!pGJma}Ys~>A@ZKlor$m=u_&^Nmm@dy&
z6q`<&;738a8ay_}gxlnjFZ~IHOC(0h3>&=jGUoOL3~q{*P;A$Sx?(7@thNw>0#o+R
z6A4jaLy2YO%Y}~)b{VchIOLWIaH5NfB2E|izJL<H&zhB$-tQDi`4AMTTi=2NQ@5q|
z<s7rFg9#MT>Kz?^bf=;)WORnOcSuXQzwjJpEi0yLpwT%M9)UF(Y-L9!{F2U809_Nx
zQ*IR#%d1z97v;y*bFaJ^EeIw!rn}-+fUX-vbezp6M!Px&gNA(U_$}DjV@989>8%bs
zIAV7*XCqr)0NL>zVS!ZFh#I<wRw0EvQli^@@%9{L#)_DxLZ!QI$*rG<bi4P-GIaY*
zG)lyYHDAC@Ik)$g_J6t9AF+9*es0fRB>D^74I}8q?<#rx{Z0ax$&DMG;Ht!q$Sd@~
zj7mb3dXM!iA?|lJaQwmF925#$1|0^tA*2Qdvzx+5fr!L9(0!S7Cx!8_SbICO%5{G_
zK!d50vxEmm%6**48Pi+z=9@yxeVGAo*8rwzcRGDo3~rPkfT+)k<oJFp5eTolzpuQc
zRM6;yi609>W=US*M+L?Xw`)&#{5+Zb`Jv46lPS#UKd{tW^wEFs&kp(4=0pxNxYjCT
ze4OvH{UU1Cw2KMy1u*BzqL=s~g)hPg7N1183v8efu#cdr$TV`x*Wbaq3~;QS-gC9B
z&6R*)E*X)#!Or_pF#ln&@NsS+BIycp-Y42wB>rjuBCLB#NQ8Yvz>7Nz0l}l_AeN}+
z$ej3gdYO1O&s!`B^T#ogzb&}g&JdO#{J_oOW4?1*v;i>01v`nI@5m@;Oq3vgfZh%k
zTFRqfv>KWoY+~*9{&PJ{zhy*@`5Zm+n1LTtW^qBGZ#}n_d?~>BC<?0D5Pv;V5*o6x
zu8dC2{(^o(?QVy+U<@@r_7rlZ=lH1O!@)NvR?kYz`n0{8-*@sSFZsmOMlP`)6Yg-b
zX*EF&&f4DO?6uIXsw8r$GS0c=?LsmRJI@%w(}>o0cP*$`ACn6IR9`pm*{pN`nSxpC
z9hp~luSz#z8`i)`RJloISu?zEexV$ElRWgIyoE{BJ1)OUreD-n;~glTYob<fW(Ise
z8X+^VB$Z`)tGM*Ui?}%4A$UW%Mh+%KidTr9W?Wn&w%aqJ{=LBd7S|WYCG?yM5Bj|^
zr0rt8|D*{yhtnUh=$23)A3}I)(5b2CZwRS>fJ@2>DuYYYTT{gC%OkArF=f_fA;QLu
zhNLJQ8p_v0jo6viyjQNUUkSLDB6;|HVBb|Fe%g6$T1kdNZV=-DD*=lX5^L2X;>ymr
zgg6)7Voi!pwH51MUyo}&7T-4~5uZE%JtQInj4ybrpKWfORDv=jDngCX1ICpwd78cS
zjIYMT_yLpOHinWf<866Xw8w4KE_tIw!p}6g)qmY3bvb+w%Wiz(pN7>i)hzMrX-(t@
zK_e#q%k{*dqCwUSzVPVy!h<Z{p%uxcw1E1b&q6)?8^$jWy~y(2*#-Z<r>uBAEZ$uH
zWc_gK_!z`iPd&*i{dgTuU&sEE^M0pHz`{Qrbe`zJbnheTT9jw`eqn47+yUlbZinAr
zG2LswRM612j3i)jG^9EovnJv+9!Ng4gKn`cICH&^+GrQ?bUr7GWv48aiz@bY-s2%%
zJk9k={8YLC^de;Eq$r+xaO_tDmN|QyEJpLDSmc9<!Zhq*mtyq8gj=68QixtSH<ql&
z$?feN<jB^lyaP<@V<Xc7q0zq8O!>iX(>bq!CW9M%sNlwr8OGg6(||<Ek(+&j!)Q^0
zllQ^rO0C$bjEUtuMGH-}^#GGV8!pnS=0|nYeM@fcx#I*3Hw3UA*<%aQ4av~J^o+)B
z`?+e5=PKCzbd6rzQND+knCy2)R9miE-p7n@T>TOTayNFCEhSa+eqzOi(nLD5zdd-P
zKdNvsytn?~<)61<YOOwVHq{8VGOdlOV5eUL`|XlF>_EXxx!Un}h%u0PrL%AwHS7eX
zFQAcS(~)Vl15z>X<YjJ()-%7Gg1>Dbyb_TyOgA1C<?TNZtKMkrDXCJjW0@A!bqyZt
zk)rd{6djIS&o~O{mY1d8`a5+W(mvxxU9h|<we|Q+AHnAFTa{g)UzX@XzT8Tp1^GoO
zmRHl)O=OgE&Ux2EVD88o_-Aq|c~puOG?AiR#eIe3{;smu?azE$SsAJ5v89_3PX$M`
zrL>hFdSz&z&4KMuS|`a%K=N+-dfA!Aa6O=S5lp1d#LuMUkTuE_-IT8jbJx8oru?d&
zy~ZyMTd&D!B%JB1$ol&;t+<kWShwFZB`cr3LW|xKZ>cr;o96ZK>>o6*SLsnEoqjx-
zHZ68_C+m%IFa+S$%XadYrkntOmQYuY{U#Atc@uYtaxYHtZ+D<D9&e=<-teIk_ryx4
z6Uex0*GTNB)2yWjdvEd9xfQ>=R}JbuTYfWqWhCcDc4r)+kMY;{&2us=&rYDWEFG<u
zeEc7Ev&>RX%LpJ`u%Tl~@&OD2B%mRlC1>YJOi7DdM!ozP($9_-;1C=eUE~$6{uHrV
z%s;!50d%GE)l64jIEVR1O^K{9=hOXbaUYynOF}e`*!lp<eWn$ANAxyVa@b}-Q=Y1l
zR3E#~9oLMqlgJFhCkb_Ji&B+X)gEifi}k*|GimkEqkhQnOG~BbRu_-9)f&J?wvEPl
z*Vd;lC_3WJXcyACt&Fj7bV=02JqESc$+EeaJu2)5QNe~7){om_C)-ma3$jaLL(c)w
zFS|`0^|xE*?{5xZLi4d`a&h$BA1ly{N%|B`B0~K_OUf@F>u7BE$GDF=rUGO7U4n5B
zg;r9@Q-Ni)rkmIkWpMIdy#_MKtqBYXd~nOES@)*owTQ3+_kC<C+IgQq9LdWDuL{qM
z1Lf<Bv>Y7zPZ3K;?6I068vv#p41D=t{7>c=-VC1OkRW)pP@$88DR;as7z9ymZZrt9
zOR%6TiT)e$tGJO>Z?fR7&~Tlb%enD$w)g7rQ90=%s8eYoc8ymi3&6O&Js%1zyotG9
zb^}b|)-TRW`48Z3Jch~0ZVOMj<@TB5rjr)Y!^dBr%if#w0I06vgdmdSJZ;+P3Mo9n
z=J=)~g68w6%QF?G9|(!Qd$u6<#(Jq%ub>UlU|wv&r;DbPZ`nOF!^5!;NGT;Z`ooE)
zKd4=3f`3rE8t~{co6y>L=P?fvficmjH9R<^6vSH(K_Fft>>GU%0wwK&O4+)ub(sgE
z(LHkp$|?6eS)=kKV9QL_5(&VMn|R9hb_G<BHs#?(F@Oc&SfAFXwAJA72TRNn!Qf&l
z!ADu61k&l#=mdlx3&9QxZx$`wGoI4j9VDA<^9@7}eD?=(*0Xb0ua)1;ZXgtYJy@_w
z0<)F|=QHUn(wbVYXmxOz^BUR%?xq6Yr?@AzY`zVW^h|cK?XC*HJp?8gdS}4OZ3sp%
zH@E(zRdFl#%=vY_ttD#VV992G*Hl+>-`tjqXO6yI+%an!m&EXB*=6Y7qv$#9ZR<$g
z>e8j}vSOC;blU@NDQA**zOjg+JQf$;c+*=){;7k29UKRoaF<HzHTa#Bir?@TcYBQU
zT>fN;tuag=$~(Qn8g6p?=MD0Z9yUm~FlLuOCyTx4SnqXlMGd7+EqJVJ%>GFa!!pMV
z7rVJXvQ`L=XHUF~5)m&OC&eaNo4~^<OUoE_jp8L0JnA2`QyY6;*SN7>XJC6bK^dhh
z7>JD`B%l|KZhMPTrMpPqENS=Q+>eKNDNhh7Gp(NWr<Y+)J)?jyD|VqsIenMBhUq6;
z{mZ9L*{SGBx`UxoUO|k8&IgG!zwnhDqG5?-3?sb)Ubn-?CfG0c+b5wvASz4a$T^eZ
zPD>wjD|`108o8s1b)c6R5*M0Aj@7bggT6q=>bH(#0$yCibgq3I(TEn78~u`m4406U
z|M}u2+LsMG>zlD`G5z>htZqMx{1lJSpZkklb6ht}U10~q7FwDKayNM^cX2KIgSoh*
z3b7G%^gOV_i}0&&`CJ7ng$TpL;+!<og>zk&W%ZfPF=H^3P`nUg5{fy`wn{+y^Uyu|
zT}zEmfZ}J&ElPVL`m46*);ln~o|}i*@T%>DYqYhfR862Wt)$-`ucP>9$JwxacE_*Y
zMQpeCweHhAbmJR={Zl05e%TvW3=dSe{|d$tDj|TVCepu?QN#o3Br|9A*S9Hdquw6A
zbr~%&n^tXb&t2kYWzELxszoBreMnLaejJ1!_BqtMJHG4_41Uz7mNQj2;+}P&KMTL}
z`?MRzgbEZEwl4fLCugZCH}r{iwkcMG=E;O{d?<&m=T)O@O;*sPE-C@O<D5VxmObrV
z{LyJhPB+>KE{e(0dO|i#hOSp$`rC8(_cwFI(7Jms?P$IKYb38Mg+3sakmJ~*lSgN2
z24@J-35!h|&%G|x&f%p)FKsmzvTb=Wg%D>BA`->GU_WoyX!SzFO)xkNcK$-RbD!xK
zdyID=y9SQgj2RBL!U=0r(AwY@;I2)@NVYg<=s%IJRnm4*YiHR1F8ipj*Liw|{HQV&
zXXz_UoI2Wv@mDiwnt2fV*gD<c(Z?`UGrUj#qK}FEPx=^ohJKec24)7GhKvX)1tzpF
zILXp2<i*1;`|v|EopFQn#S(MrMXe||`_{lrLQBosti@23^;uD|=7zU0`2J*2KeJ9A
z=u98ISyq>2g6qbzD5yN&on(GO!)5II{(*hA@zLqs#X+td9g$nXIPxn3uD4}qGEzvd
z($qW@W8C>+>R^j`sHC&iG>lpItvZ;-EjCA4NzK#P1oaUg=h$`uCcx21HwOw{Hl%3*
zd~VcLokODp<{0O!-zH=4azQ{!DF+FABCg?sH#vjzDb5?`=F@P@Wn==$yHA@u^Z>+L
zM#M1K`8^>CJ$CC&4BrfWQ6H`km7Fa;_lw`$(Z2CqE#$7zm12$i77G?cJkKVb7N}Ib
zp4^~bzThww<s8D}vbW@1uz+81*~~qMOf^B!{$+lE7FiQ^d9ZmOkTJ<V;_d#qIl3Z6
zo=T6sMAG_kn>~BY$JZ5iRqPCvnI1el%lxy#=ZUV!9!{$yMpfS@?Y8V~hWtH>JSpfh
zOBXm&szeXWlY(OrvRTcU>DsfquhEosGuIVV3abQD=d|8ze2ih6c9U)iB=7O5+-#{+
zZ60h9{x5qYEwVR;GPwLE6!9q{M6*vXjA!L&@zGFA7L^u#MPq1v1Vq?b5^2TD;5ojt
z%L+Co6%WMUzn<vp|3<h3&i8@!gGAY1R)DHVgzJ-3F;yt3*yWVO*kJH6#PPq;-R_6N
zY@CnZ(3<^MBR9Z+)E#zeqp$q2#MRCenihFDe)LuL3qqH_)En=RZFIZ)R?eRPYok6O
zqs5M0(l?MmgO-p4x7g~bFLNtQ>$-(;7G|dly}K1^NH9EtEF8Um_kVwj;iWx8fFW6*
z4+X!lSMhuT@qIDJ3`v20_Pg)e0WN-USNuXF$X$nZo{s1f-emObeI}a4h(pJ?xBdOo
zTo+GIG<4Z&2>$}oiY;e$geB1I&|h~KpjL&Ja|x%rm~SUv7=Apje)&(f+E_r+HK)z5
zB?0I0Ckx}<@(Sbev4q^)=>oK{CKUNcn_o)?&PqxhkbXL=|C<l<pC}D6$nlnab4&EE
zA0S6Z@fSfEf<P@tH}RjR;-6q`OoGEt*4=6c^5Jz)^!qV?@E!WI<RrF?eij{o(7PR(
zo@NYBF+ROXtAtkeMp&X)>kE-*(Y9v&VT>yX+lMr3lnldKYyoZHhPL9J`Fp&7X43Ie
z0Y8iHMe#kQU$l|5MJ$CRenoip;|2@(hP0&Mh)j&>IZp`LPgo(?9NMC9EkE6W2^SBN
z@0$k|ifEE9Axq$?mDjX-SM=J@MSmXKSAo|bhVi1sBG?E$UK!&{>^AwwC4QPPM;rc*
z!R6lK>A$Yf|J4h{AXuM@mSv-w|K*MT8s_|u7ysU%{U)LOpZ*QK(Xcp&qi-a+{qw=|
zKi>5BX$0T=|KG>Ac=Q)ZZusk|qNYEuB;Tzx{S+hM%O&zL@+be(^S+2|I8XS~mhpaV
zIK1kAu<Bd^`PTkoJNQ~%L!Zw)>r1%bIkJx9gRN}v+v?BPt!OTgT^k2p4h}kzrTTm`
z)O6;?il1%#NHij#Pga|6tvQDsogiL2?(kRy<93%p*Q~RWottcGh#Tp`(RnsZ%^KA-
zv1zrC6Re8Vqkq2rI}*T9I=mpe`+MV}h&d%ZJeBDPs(Q+NDn!_k3=_>w@Tk$Ethx5$
zU$}6zr~WUwaF<`OJW&WSd(w!_+zu_pgN%Bxt+@fTRb^{+;2Rrb_m-O!M<)vcyS55!
ztI)h?&<C^(xWbo2_=b4$F6u#ZCz$|p2KBp(LNawdGmo}2sTcG7_qv>?+C~3$nDuY3
z2k?<?-u%ia(XZ#qf)@>ev>@T}mOnntgoTY(iD?iF`UOjXTVd4%!WUB@18%~3e-r$O
z+4|XCBmDnwIdRVKSi-oj#<o@3&6>`4CGb~FrEh?#J4-pMzATzHy2ZHm%{!e)DCAt_
z?6WAKO~^txmdu+yu+ERwuXWCL2*017Q=DQnH|1rQ%xxC@|Hg+#Rdy%vw4pnM4F$G`
zAucOKv>JETD;zhBzk>=1oHFGg@DpXJViy4Af65_*vj}gw$A{By0k$0p*}5w0d-c<-
zZSkb{u~E2m=&TqXevu{zdhlVYzC}BHJoiKNWLs4mrn>TP@>&ZEwt4dODi^b9=4Mn^
zo!@MTlzw)*ck)lmF(L$>m#S~;7k>k-;`-=kaVZ$>7e3l-!ie$IVt@86b?hIT6z`Sk
zAQ6%Cam7WGM5h4%aF%_mc!FERw>``63Y5AQrcJH=jAB)1+52gA|8uSLciGuHS0Hk9
zKSkTs*74Hs-~JnZ_}AqUeH)&Ke)5vDzm`f4%V}SuV^9#Ik3+hi1bD|WL1cR^l}SCM
zr^JSH*InGyEw{Ebo259K4HpaknXb>M=X02J&6e8bZY32oMO?)+k_9c7%CL82qHM2i
zp47Su6}5;wq;j{w?D(1#d?K)RG`F@kK0i+O)Z^ie1E5)SHcm}|`P3sASluI$5qqBz
zp{5|_>e9IPA<jT;ePK!RXT*CtvXFI=kWr;^rrrlXJBQ2r9!#$Bwhat_*LEKFS4^|n
zH%iX9d}Tl6O;gXow<mojpN9?BdJd{co*Ec1)$Bak-<?SYOe9OxcZtP}X6I;Q`(l$#
z&b)h;&3H@6I=`|~IWFZXgAw!#RT)XYy#Oj5lh&<BtV`3QU)qq-?@PIGF^wGV9&&B+
zKiiG7N(oq7t1_Z);o4f+yyE6^V)wZ`vZn}$bK|Z_Ax7k@O4*0g;~_34U3Q@~gsm-N
z95uB8Tlo<`qIPwBoJGdhFP<^!8J!e*K=~lN!pwmEGrq^i@sxVza;ad~p6$Kq*niw+
z!(iF$@MkRJ|8Zx<zYy&M{A{-Ut#9`<G=SG)6|a3!0||`@^=>%;Yvce`Y_ERN1-%P<
z0vN6P-9s>$Imma9h9_P)qz7bZR}l#=m_&fi<tr3ru?VZH1bU)QL|K$>DCDplzq9cZ
zX$X)gy<LL+;x_6iz~DZB!Pj$jF>d%Vfl)9PEB%;GvsVAJZVoX6;07aeWdBk~;Fdvh
z1RF&7M-0y(!<4VUu_K9YXj^r?!$q^OLFa;SbYVX70mA)k>VcPA0>RqtpyZ^N3c!nj
z>d?Hn61?8?^AYE*&rtXG!^m_L#N?%kqaaZw=Q5U_a1+O9*`5Wd@mo9mYlh(Z{no*6
zYi750Y7GcQ)uZnL%(?1uG(KqxSB8mx1mt|CfCoiJ-M%h|#PnSuwR=Ju8TV3=*lOIu
zHqiv9xDg~WOvr`kE-=)MZ(Bbbbl?KBl#vY>N}2GJ&dp(XAflen4&yf5VA|ybGJVD7
zaS$_9yF>IqkEJFScBmcFSa$QrvfpI|w$eLmkVSec2Vq`(D4u5%Fpg$)b_lD;%gwOh
z_$#v*m#)|!Lf^}WP5_PhbKJ6_xd_oVPZIcT5|OPN5FNe)5NMhrpozQT3$I?pcTI3@
zRO{sCr}A4yDz@cePmrgQe;jhc+Y}vfUE1u=^>F5?17;OW^2u3BZk!2lRKL{kT12-+
z4@@Ok9E33ut1_2Xhn2Cx!(vdpz7p+eSna>X@pgwwn0D&=&VcGw;o61S3XkJi8nYkA
z2S0jKKHzrm$%{7io031?^bm?!#GIfjs$+QZun`;4@s1=0%mPcj9RzTuyY4{Uz+R?W
zV?;H*aGFopRS>V@9i|x2T&^Z4Kpg7Lz`1)9=5Uw+{P-K9I6_kGRyD7T6@NmDo&_4e
zM{r3yq4lvc`{dYD7QFxGdjGp^47RiZI^@)Qs62Y<kFCt>km*!ICv0Tw2)hocvv-Th
zyyHMFDS?<@X<KH;EYGfib)%v(dip6}Nz|S=Z7J}e@13F$b**%J*dlA<YAxgB8t*57
zq%xKORx>W@D>uD8<%8b=WI>}wslW$y2lJFf3x1vZCkDqeWRJ!yUmZw|OB$Z`ngm`Q
zJvb+)Y%dM>6r^$*dFMgWI+Ko`%gS3tHS`Ud=Eu-kXJC`{4f)?mhetr-2zbh5zF|_{
zHy=S-DHkb9uz?aGrJ2x7(utpch0Lra>m^Om2#BS<EnIlLL_~X~7-N7$LzDA#q{xVA
zYJ1N!;`}ja1#wbiDtEl=WvQK-n=_%TiCX<qouoxf&w!<jgq6l)ZGsk;kL#YX8Ai&@
zai-w6HoQXlTo^oo%7&9gPe$<2vx-iS3?N$Btqb&TVcgl{`&h*qb&TNo&0g2OJdO&(
zkgi=n^xZgwYoV|ma_~)r!Eed)u4LXJ<fm%8L)gKX&S@^<=JI&mBJF{va2;C5@(a)H
z+5;Rgp6FB62X1Qz2*(`e$wluT(hQuDFcBn_4n|3PT}J;<ISk%lW=j;S*j@`jNwuZd
z4wQ?$W3ml=eRe+?kx_o@)BH6ME-|{l_pPsK9#eRXU9VCZXqT4v)&s<j8-{in@gpq9
zrRjkCOl0gN{E3b@AP`9!E?o#qT>$UXa;M;frynlGWnJ$f7(0MJxe7w>+N8C#C~13x
zi8iOTpY37i-E|xu0g|g?d_3=Pguw@MEuAZ~hpM^mNZG`+-|V6S&SEQHaMqq`O-{*>
z)4vDXkQ)<2Vjp2=6Y2VBBvfQ1e5PODO5W)jn(T<?&mSi%_;TyidoXw(Il3qvhO+z#
zG_+_@x5M~|9YhIuY=k~hwO*&T=yLC~0C)I5?*IOt?u`E;p_D*7^qR=6Y|GUYdqu65
zhj~~$uo+Q^hW3i){#65!9NN}&#3NNhX~!f_wJ$<-ja7SL+b2)eC{lIJKzpH)eDO<X
z?%w-{8WKmWS*5$V7n*Xlx^v@-LYH=YiiEa%rCtO(0#y|IFh+&t(K<xhPAsa?8Gltc
zrW?ss?#b=#$@RRPi4u5i*bKvq`?p18%-;5pY0&A>n;X3BwMjn-d}DwSeY^H&^NWs-
zPx#A2m?Y1ylLS=3F3OX6UMnYnGhl1ch<oS(H#^HFEpF1yrt(`<swBGCNffS=a0!q^
z-|JH@&Ku>Q&kMFG{m5VftErsJ5{wC(_I;zAZ<g?hW8B8u3uYq0xK1K^z3VE8kLl<t
zWJ=q<qSt*jLGi+Q`mw;XTk%mjQW#)G;e>EhgGpcDM9;7*T~(@OyiUSyKKM#s?nSU{
zXjHIF$;YG6``=_K%ENv{cVrCNP@XqDXH`)3vYTp0EhI0L_vxzMD{8rwstb2JT^k;a
z_&S^M=r$jBU7IEkSSSl#e&OuYS$V$4aWWh%;rM1R;`PTbu-<vPXv3TNfFxY-q{%zf
zyx@m~rxl>rM}vj#F)IAu)B1Zj%QAsw*NeZCcE3E0I_!+@`%+DjDjm@j4A#1em#=%r
zCT`74H1um(^*dhMR8Z%&T+y`45aC$I+ko=<@YAlVs%`UazpoN0lb3fyJ`Fcr{il*|
z;M{4)cEkUrz?DEl9ZN)H>l<?G6sGqol)a>G@OBNvpk?eu)kx^~<B77)E<`z>wFlbg
zGqKPwv4?a+9?`(s&NNw)WkSc@AHg+>au!s$vY<pv&+XZcolyyK6Xw_%u5KguS=5O{
zasysGbSN@-?a7qGQl@%SOft@FnH2bEmA9JM%m;`>E5U|R!P5zxF#SqZ$2kHNiA~0{
zLocsBU2nilDx!<S!Y#+0KI7|UhwUvE_}$RMPOdh7^mvGKLi^PqW!wu@{BUFuqcEf%
zW6|BvyR+u&@Z(zjL6{E~3%jL-;nNYLfR-1mw^C@7>h*J9NiMzdr&6pj;M>kjF29Qv
zqElgWzQo_@>Ec|U<+q+@u7EUrV-~EjfhH}Dsz)Qz#qgG6KfyN8^eqATqRrs4u2i}C
z3nh~jD*<aF`~WtZYY&o}wy5@Yu|2Rn_povIR!bi}U^_W4WpYe|Mcl-l%M-H<OdyEW
z=FSwPi2O1l+ww+haxldHhW5qicSBh_557{)V|X~$e*?Mk$Gxi_VUw|kDc_g=O96-o
z{C2n_mofh=0HYXAHKh7GbuL!;PdrhPppd4O^A=0?3Ao@o;>W2hHItC`0FeDy%tFKU
z%{aE0LpoBLY&r{XP6kcyYs`U>Q#OU^YVB=KxCYj)yw=82gNX~$j^KdJmMBlqE4{SY
z=C|EsK=cLfS?8s??VUAf;2r>3e={KW!Kb1iUhC>;E+oWIG3~VxC#7j7o%jp?>1S6l
z8zuoFr*Oebpjhl%CQBj9#NHyCNTYr1{i;AuEHERom56$qH?(lj?;xPgsN1uGV*s0g
zwxikUUT}GCv`LrHXQI9Sh-SQrj|hR<#8cGxMab;bnl|nU>=1paceDrI*)$DJomOn>
znJ=>%VfHmB3PTKkCWyUl1g$f)Kn~T#bScmLTcQ7UH)~|8Fb&VW(0t7aHug#jE%jM`
z<L1Drl>MgD!U23)))8@5uaL(O!@J&Cn`tLI@?_n+9ylUbG6nZO+lPl?)=&q00%OQj
zFhrR9!kYh-VN!!*-$&r$k|SO2z+RW+YD>vKb$ZI@&=b?2A8z>kFTPAG*7=J9%1C7h
z6N`3>DramW+!JQRAUYzvt=1Wj@I;w|VuFW(=1E$fb11<MX`t9ytdCavIw}MyIlbZ#
z_!b2N?mt*sy+IlJo*Az;udf12D@K*W0ykE3BmG;Os|ts{VCQ0h+0vU}GHI!u`716^
z;0OETkdg=o`?GdwYe-!U4xTmzj(7&07Ei~0ENz6HW%14Qf3UL#h_Vomij0Zz$YAsX
ziV1B<8DIuJUcf~tR3+R$Kf|&43YJ5}Bol$b9W%kiFEn2(zJK<qXys1pV=#xx$k0<M
zetz*$JDp7F2-MWlGF6)(z-S&yZxrf_r~%KK1IXr0P<aVywi)#LeVjq$2X>f?mzC9S
z(#P}IZJpvTq&|nC@Lp5o%?D^)m?`EpIrUj3V}YDMu0y2hk?X|D8Vy?BARJri+Rz{z
zDj7dnjjM;@4xUaI1J-LPt7)>w_zxq`!n3{tF(~a&XmNl!9GB4faJ$C_G9DSzMRUXT
zKi-B?i(Gde8`|Qv(Z#6ApnEAJbr|gL-Sf^s@9caJm_Nu!b+r9Iq`h@iR%_e#iwF`5
z+>{7Nr+`utO1FpzC@Bqsgp^2^bVw=)Dvb(AOQ+IG2q@j%CEf0GuC-jxdcSYIK6~sv
z#`?p-f;;BCu4|rgoWCQXM3~g(3nrNk?V>qv+AWIBi)%#APMxUS4X>ZOUa53xGs(XA
zMBDG#DfqRn-Y?5|z+bz00?+)SZN4-@9IP16fE03|q0I-VexT7z+lmm7#`O9C>kLD&
zt`Nhc{(25IyRUmc#$dYSGA)77+gv{LYRY}tOsAVFr>b&Bz()L8$5%^jh!OMz4C#JX
zH3hB$)Mh&k3Nrs%mMG>(&~-{CgIQjCgFAp0a+YD1nmPW7I6nQAZ8Q`5U?iDY_7TG-
zvd5w7bmHXk87H7AwSOxq`ZcSRocON9dcX3W2m$7RC+paf*|uyl;>5WuRiM(W>+|0Q
z3o~;;ePk-AzUTgmrKUbHkwuV6S7edaoWhqU$|)qTZ?u*ukl)1)J=k62EPT-MVzk<g
zEG&(rmZk|)l4_eRw(poO+4nP}e5m)eJ$R_Kz;vUui2D~%{25B>Z<nwjgDPfmN?n~p
z-YCwz*Rv|vsm4LY$P75@STY{dn?{;*@zTwFSqWQrAgS}S?$xhQWYLs4WtJqINA5gg
zbIyD-AgrEwLtgHpo%9RVUVX^+RPTX^Ojy6hy&T7pZdw#%qG6Y%y%4dc2@+LrAl|gn
zA|4^Gr}+#kAH-T}FFd#~C<*h5a&Nkenc*C;p&>bPAsYT&#a<j2W2^qmf8?5YM4beK
zO#WLk&Yz(2b<HnbO-Vi&1MK0mL;SX^U_jV+LcC`w%)d}dvohcg-NY`c1x1J3^VCqP
z=boTpbXY)lz5iRk&$lY<=#Fo^Pm7DJVf%LD{guxer(W2B*>|2#)f=jQZ5?o-+`x9b
zm#z559C@BfDlaWn+|7%WA8BK!q4Q9Sd#!msMZhkpOu2;ZoFnDKd<lu=DQmyT>b*v$
z&?SJxC4{WSyQ15(#fP!0-=SWVqfrxIc(#0*49jb?X%%ZwCB$m#phy!=;luoN7^mgU
zCrM~AByJ4YQk(Gj>X}hm-5*(F-fq85GT&XFC8eofML8~VdQ-8nY84(9rOKnuCrth{
z5ti<6lMKyHOCFqW<;f-ZoONHU$|{}kTWS9s>p^^PCE=oEof~GaonH#)Rqi)7DGi#x
zQO72a;(V&a{O5%RfMI&dnFLp5dkPA!f?d=`JP_Voa8A01)Qv>&5}u%)q4B0;?d*K3
z)krsfhf@!~E*&b)g4?q(YRs=xkY{3y1;^Ftg8J%bM1$|B!*xonM~t32pK6DhBuclM
zxHgb0s(Nwzh;8x8kNY2rqf+U@M!;n}=J60M&9<4dE<8Ott_6K~#JxinV?=B|=y=H(
z*G)yq<#zo(KWkaN&iBr$f;I&(0F)CZHi51f$|+un$mD%2XF4_2@p^+b-|(AuW4ux-
zx0>iSkkG}UU6o%Btmt9S91**{H@`O0e=II^v%pFx30dS;{qJ3O3i-68cF+~Ss>%=L
z7L9Y}xOhU*LU0><=}4z{eXB{ezLIhY#Gzxeajq5ZU#`1<fal)uYw$V<7tgd*ytgJh
zN)58Zn2p4Rp-=xUj)a8UgmHl2&xD&<2CeQQ=ibv0Xa14Lche_!c^X7mr$nCYZ@^rS
zH;y+V@8yUhXb~6pjyPFgKCOa@NaXter$lZqn$c8JcaIg1j>hvryj=i}ldYFd;N_}?
z+DZk;Wk(PQT6Hzm;m-c$2Dj1Yp&gq;W|(CiP=k5k;jdAW(TUfWLbYxYorQ;@uRBz^
zcz2k1Te1xMy!@BaE}OFpSc_YI{-vlRZ%aXf!!ExnENbUZWfJFbQ_l4J5w|?S^h>^w
z7a(m=<_LaX%oz>NkLNq-I0<au_DjDi(Orx~L~N$SaSAFg45nN}AvL;P|EK~I8R7wn
zIu}sLd0wSclCm3&(*?<&mi-oJ8y{0z6-E6)3WF$Sgkwn7qXv&6u}@;0FqVs&qUHOt
zKrU7O3L-c<aD``oqngScYC)s0+9RAXn!~bPXg^F}-Lj+qzxJFjCWf?T)|IkxpZ+<5
zmQSKwaxe!fx(m}tbnFZ<^~|`WN44Q}cfO;t$#<x%SzA8U(<FLNTzkR;J2G2LGrcPu
zN<QQQ-D1#ucs;cKDv)1%Jt4ZYWIgX0gh4pFSAdoJ5tsBz0Qugrds5@5&~jRY=&P=a
z&H_z)X4u=IkiGw@0M&9-S-khW7iOhDKER$Rf%_4P-*_?lAd&noQ{nFfugUL_$&T;!
z%}sxb<*v%0uiq)Mosc9h{ZvU&bycEs9MGXqDti*_woXTiqwUTZaO12IYZ6ctR)5Lh
z&JJO5<4NWqrmu*9N-h+{X4TG0Qo28Bk~2kgaN2$x{_{qA>l3R`*5sFXO;&=m1Xe4d
zELREh!U3FTcYdf)(BaT!_{O2p-G~r9F#E&8q#j?Wt<I1BCvXu6a&gwg-JhVG{|_k~
z!a(z1qO&y{tEu*1X~Z%g!}4R5aiiUla(gB$lR_Pog&HKG?m=X5RbT%pgOlybj+Pbs
zBsQ$SoTjwiNz(ok96f#bKW$Q9a3&C5zlk;b@o0wgawvmz$lhhG4-E|@2EX8E-g45R
z{y*t%8AxTSB9GLruXmbALcuT0XmcK(nkD&N<Jgs-oI4VP`X1s6_h~(#+Jy(g88h13
zb#W{C_4v8q9aBO>%}dhYj8GDq&L@uPfqOQG;RNDS6T+T5-wOe#6BLopI0o|@40i<}
z6PJTQE)=n}Xg+bImS<ZIMUz1=2x8i$w%2qA`40y^l(F4xsRVt!)d^xbV+|I1j$TEl
z5*ciIBXbeCoC7*P|D)*}g(lc{(=rPVy?3H5hRh4ZwJk<UehB+0O1htmB9qP<<u8zl
z+kqzkmxlMOr&TPjBC8lwpSO)y{0$tSvRLGT?`<SF;o?_}=JABArtE<ogkK#5ZT4*p
z!kg}OIzDq^({s2nLvgTx`7Xbx8Jox<%SwsZe`A!?h~g9j2y_#LBgy$p*FQjT<U!nB
z=YiW-;RMky`I1+rCEntd?UDIE>Zkt42fH7@u7UgW<{A?!XZGgXng_G*U;kFX4Fnxn
z&rj2u`2SkZ=I9;5o1r*AXqz^~{0277TQ8RuxjWCk1}h5_DUwq^s9U5Wr^K?M_r8fw
z{eYJKNd`e(AlU6Pv0~-jM|JESe)2shDl12f5eV)TrH@cotO^>>Gh0Zw=sYFpLPl_L
z98&ux`@N!@M1|Y%sIGTQl=Z(;q};q!0z=}1SnGlXbh+iz&lG(44@#J=YA3)or`ALZ
zXCLObK2)GWNbAmE8!40J3LUN2E7Ha|7_u4zGOdZ|mWH+!UmqfmvqR)@sW^W6p@=*#
z73FMvYtikh*QY0^BzRG0dWNuDBvh|kbhA<lh5l0@XVjP^;h&NK$mV_rl>g)b#G02M
z^Lv*5xnL_`!S-<icpvtoXA_<y8aMsJ)(*aDD!T?kW?31Rq)0NVLWJF)#6NKf!?laY
zPLGVPYP$r_kbU<0;1idKCGI9DqH5%uVhHSy-3A-66<6#fW$ICE!ku!+AWY{z*T>8B
z&pu*o^KjnxTQLl)tp+mP4t7NPy1@lgLbBYdg2Pk%ZwDRkadf_L&9@4=U7$YkGVlk~
zg42c&^_^h2asRnQHyG=62@s?{mScm81pS_G5Z}Yn{=jk3`0*4z9p(isokihhk>gi9
zCN0sn#21CLQ}EIgT`goi_ma6GppkpUQGZ<LNe7nNiH&<{Pu4Y!b9NSBm>2IW9<^})
zgUY8@Dp)@U1{x;a)U|XMd0RyZwx`w-Lv_a?t2B>0B~Ba*hVG2<8nc@)|F)LH<!x$K
zdGc2DA3+bH^4gVDn&_Oy#W1-o-}f2)wc$Oj5g8f`T@jh7lI6)7chhf^;&m{yk)TMW
zF!ouq^dh81o(8)v5|J_ldm6m>3#hipeM|KS05+_Qu-7eZt(>v)NZ5Lb@g!Il2bu=c
zQfX_Is-Fi3oW>*T=x;t&)nuW-y~?0pjT;=o_Xkexf9m1xddnd+H8HlQu6<FldahsM
zGA^_Z=1vL@KuT9HL%iE__$*^ws(UCr!--lY!oNuZ(>|5{l8TMxFFhQ0=>MUI>vl^a
z^;OKIZErdO+5a=JOt{x={2~l$QCBI0dt)DG)4h}t!D<1xNA}Of;M)eJ<##gqy(_+d
zu<BM@j+?xf^hv`%$W$ZlBil(dEJi)maYG%CYai8=98*(uz08K7+ojVjbw83UG3hgg
z2I(r!GXE>CCQYnvhU74RTrL@Zdjy9bj4LHALMY3L6W;SYu?P^Ny)i(;seeehQtx3f
z#&+}^23a*#L&<dRhEs8=(mfj}!?kF~FR01L@16Y^AdBC7ic~NJx0>gb%AbXe%sX1n
zT-ydi^Q|CczeMruUiay@8K2d!J1pMs;O!Tf)f$ff<S>(uv;F20LZ--gpy8b`dvmR7
zGdv@iEe(4x+ozXSsk%jL?1=YDUtw9aOE~3$6Sgo*eTV)wvxTMHV?11nOrfy-b5cUz
zOvzU%6Ac#|HM!1fo&{00=u>W*9$q`;YFsWW4_Vng9p{VdZOzA#-3-~)m4;$D#b+Ey
z53T)nX}7C_*&Z=hxy`zsnV7sY%4<Iw&+<EjP4XG!Z1nYQV*X1QFJ4V}<A7Dy=J>wX
z28;OX&q<^Tz2{F2I$DKkaOXht0%3k!eN%yB*F$OID>-Mh+GyD#hej{3onbD}nLm>{
zinn?GC?-#Bzpt&B*-M>6{&!H3Vp-^M-=pL%iT$gpz8HxncC&9H`30~lq=V61TzZI-
ztu$6T-}L6g^G7}$K=xhAJE;>N+IimvQ1RQm${g_{jY;?f-J%&#qZb^0avQhIt2P~t
z+5EAtvC}kTcRH8PMcjz+w;8z#l9A7UDj4`HBUdKCK%EfK4Wg_GQ-%@v$I|(_XOfJy
z-CNk5%&$h{1Y)Sam%MJd@l$|f6&fXRT@z=ol!-(Bk{&8dttx=%-R@Dt#C)hMsKl>v
z3qO{vl@%Vp_dRIc#qzfw1O<&UHjG!Le-F-oW_+UT0(s~2w+gW>puQpV6PCUd^b^4H
zbl|4fS`{Scp!kGerLTVIjgZ`s^|$(tug}FBv2AlQFIbEQ{jOC3>KY&yg(M05A{V_=
zGyw1jFb=CYur0F-0?tugzsjCOzO4kW|I>I_)$PWR^?Qa6&*$PMIPm%$Hlty`eSKf#
z^(7K^e@_VlwbsA?=Z%gmOj@q<|AnIUzy5s_qVqIXo)h`q^8Z(D<exuG7Kc?qiXGbc
z|Ml;=5fR&kZWpEh<s1HQpZwR)StH>JVxk+{{eS#>vvcqT_|7<tKL6c)`b{E$%LrG{
z`;+EBlw}sdPwa2616cm$u|?*JdFggDjaLzyhNhu^QI@HPVjSoeaR@xgEA==X)i{q?
zr<4mV{>!IljYGaA!xZM<McSSlXsJFRRS?ZG&$(CJ(vmeBT69<d>Bmcm&mOOh>Q(@N
zzJGh(1LQ6P^XRpRr(r|LgsgzgsLnb&^9T$@sgHrNV@P>?UpEf0&i2pnw49$Dgz`c3
zT7cRed*HUUv(4;(*IXh^Oq*8{>@gCnTbcWBT<C5}q>F7OJwWr<H!{OOU#G?L7B7&@
zzWVA2L>r%>TX+Rf?!IbsB14F`GJ@T?Y8|CPg;IEOT6mH4`M1^P`K~k*use7J(>1Mt
zP=DS327w*MaT_lSH*J7eBNBwEj4$#Uf}y7H8A@$F+9|rAF)xdU1Xn_jP6$MK{D(or
z0q5@!wDuw-xBVs?FmwOl{A1GKcnPY^p_Br-?gH7*UVYaq?aFDomxmbBB4Upy#i?t9
zn#cM0-L+zGYwmWRLU;T<B$GA(ZVk6`oXoEyT={}G#9msa?XQ?%e2H_pjzob|>w`2P
z+%9TCovx<>hKOGZ*dzkqN#GkmyGhofbx0csQN0jYX<i4ViVj!v%c4U&WZ+`l2Jr^R
z^fK@-`3NHlJZ?J=fspTw=v(C!TD=xQE)oNaOSr5HEZ?Sq6BPr@B>K<n_p$1fuq7w_
zLo*kMl67NjmJw8QRu#mTB`Y8u({Y)JnW;34>5kv22(#wvb#;J@3Agv&o2!XwVLb#(
z;$c6Z!ys4~gTdzS_camh-dMDNUylTF^nKzdpYRg<X(_T^JO%=H99qFM5Q3HXd@=z|
z@iX=Qe!v(jc_7{)E@EwRk^4Xbk_UsY_mb;m#cz{Nw$dN9aW|7(-t)y?fDAnrsmtBt
zi+Ifix6sv>02tLZ;1|kPebX}k-5p@?l~eR)fm{slRU}_I4E+|j_^kXMaQPY8t;__}
zf6*Awa7ZK!Ao1pLS{-JFhKRr$kaerZjOTSuWIq7T8;53%ISPt!eOFm$`V??aPFgSF
zrRaDzv}(WCJ5bTPIg?clZ#!do+4A>qn}Ogu*yLaD{?)ehCc&m{psaglR0(c?H(*wO
z8cJHF#>=H<U3D8OO|ep*fPu()s6z}&7`Y!}-B0~2mA#;TWSn=l)EE{r^Ny6+SeYMx
zN{!jZifqltSR1U8GOXNsf)(+X?sp3$F~>N)yvF#)YRV3Md0UW~@AGGNLBkXUnF&`G
z>%pn>{%?#x;YD1dRCAQ-Bc@Mw5Qu?auIab-a>p~JjzjE4|3UQH1;4!Q<FF6w4qzLC
zb9A&_lU-^Vswj%H$@DO`Y_#;015104`{LL1b8+J*Cv$n3#h!bPhgM80Y!6R+US4+@
zKl@^IG=EI_JwJlDirqOvPcPe#dir*5mHhC-Nxx0Kl}uvkuFe$jst|HpY$@?rVwqJL
zXH8WHdsFU_2vB%z$GGtgmc|J3BSh(FJnA}jFCFPk&H2pD_DOY|nmfMMJ#%fphS|z4
z!RG15(b7w?cz#2^OLZLvUijMG-u8R@(?dGpPPkqK*1uoI7O!Cs7mIj46#26uNzAwo
z%9|EAg;vsbn<#BKAgs!S3PY3dvK0*D@jx{IQKAn6KY_<?Qu=Ir1zh1)NKh|9(Cc!(
z0ASAd5?vQBxq#*5G`QeAgIrPjWIN9b$Bz=C>ok9#70A{+qVe@Z^D{mkj;t-$AA;_J
z8E*P>hHyc89mamK)v~GkkaC|3KgK`!0X?6g2`Cb>oW{bBUihBgifnf{7z;bIpz`hg
zuw=mW(Q-jEZp-vxRLzAHbkQ0(avJJXzJgoS4x)+GepukY{Gs&<575cYAyhj56oKOo
z9zZ=3nPRvC9Q+tOf#FUk-hRjVP6=?MjB=u}HRL`n13H-1urKA(SA=E0f;-o^;#NC0
z_VdK4&>NC??US9-k6+-@+}g_TQP=)*TWE!mCE@Y2Y$%p$aZE1WV$V}HJl{ZseL2K#
z8)t{tak>9YGp%sPcroI}civAfzzPP2pPKH%1OgvEGW7rT0AYw4I|GAlsEh-4BLdhO
zFiQ)=n8+s(9Dkx&xnUJhhFOA?RQitzXYIgA;k+n?Jte;bT(17}p)FCJ0dU;uB7D#P
z6VjGV6E}q_FR|!h`?1s08Bx{*)AK-K=XU|rSvEO#0bUeAugbKcVk3pO!ynRKz!wY6
z9liep(iS^=RCja&#ed9A#DYX%g0FwP9P@={79=;(bQyJM!_Yr6Z5?GsIH`y{O6*(-
zlGudOh>2>-8B8JdU}U0R;^q8p?j^@dOKPtiNmmsBPFNCXG-e5^V+K$s-R(RAQd?)Q
zd3G2wYuL`6E#XWs2%5nevX6WIIr^xdOmBZ4&53P*cnNo}1{BGN@VUQ`Y7GOP1Xa~u
zOx4eiV5t@Fvgr$qo?X8ffDF$5$zR(v0R9>(J1LOb)yyDm#yx#~<@cTX8_7*L^IopJ
ze;=3KC~pwIUnku!LxvMGohVB|SSCbDU51Rl7F+gx-RXemTslok0`4;ad+PuOe;5>S
zOGuR{J8@CYCLR4udL30Ml-q<ZE;T<W1>XzJ6D7|L;`GVH>?LZju}=44>H~u(OJCQ_
z0*{MwUWg$<g`SveCCKtLwT5X{#9+iZXN2vs9n_VafOhZmonZQQqsogsbn;!$N(yD=
zmW9Fs-L+ks-8$_typS!toiVjqy{nf_&wseR@V%J|y1}<)LYvUD)ov{40U|5g68aoU
zOMxM}%ZY46ZANI!7fO{x_ZgQCbjba8PCI0x=%!BC*mz3HSEo8M0eQ0y>_Eh>BQ7lS
zbBB4qsZ+O#+)KsU64-~(ma0j7^iT~pkeIC(sme9d9~c}4a{Ge~KA2~HT;5IMryHpn
zpQ03;iN5<VF(dhNre;It=ibbD){<6Gh2<2~U-*<uAi8JZ1esDfQpe4V>sfY%Iapid
zwj1Zky~D*wz`c=^9r8UM_gn(*^Y{?;hpQ=g^RXd6h?ZEl8xRWHQV_0Q0xs=HlpMAB
zaX<WI1hKg66f45nWWA&&B{n&zE?)^9yTW0pUp$!d-^O2J_Z)WtSc4F$*Q*4?qT{x|
zsa<<<l|?c87T#uBN_6j0T@`|W3!_V9h$9Aab>V*S9C@P0-Uvi8x2uaxt}a1U-Dzd0
zw6x&UL8waxdR{1xpmIm?Z5bm+f;L`K6Mn(I5Uo-Lt-Nm^vdre#{|Q;se!zWu(Y&wM
zvtI_WK8d80ipR}PIQik-&Rz8}BTmF_<plJW3Ao2rBaLy!D*;xux%smb*SvsJckOS^
zS|QdF!>qt)tuj_Px5hfY-yC+h+8*Wqd(K+c|IJyGxSy_fd{Oikg9r>Pz6f;3?y16-
z7dd{JVF>YzmWVFc;oi-&UWLQ7<$AiF-K}Xf!kf7mbQdrP8<~Te*S)D}s()}F6?TZW
zX5bpJZhIO}zDFu&^BS@#pCwWxG4t!O+=Dh~tFT#YkLqbo#-aDdm!6pVVQ{YXWQykJ
zjp^l9l8VbkG<Mr$^k$%`#}=o%a+gz+T+PSdm4?TgwSm;m@wuHu2o?~!R!bW5{YGl}
z>9W<<!pM7!Ml(yLUfbJADhG|x-tpJXlFR3)^mcZDIW1-`e{3fCHry2K!6zS7B&aYL
zQ6m%X5NFPpxn_ru{(LW|OpKF;x5#wWS7k5x-T|RC@eaF{21?nN)bADam?Lnm*IW6l
zU*JDI^i<(bx|;jPn=kb<PIDkg9b)>d^n3bN`7knX%_Wy+Nseo%+q`*41mjGfL6~nc
z58HHIA(rtDBm!qlo;!wUS*ALW-jmC8B&`+;KWinG>G0-<9#Apjj8TvNWUA$323gS6
zoxq9r;j8h{F^xjJ)2NG<Wg#Hd9LRQ(_3l%;v(y%Dfr?yUg6_l_t>g4|?!&*3YP>f8
zKapyQrGFvSB9j2ACjTF#+VYeCfmHhl9j5EKuC2bGF%uou3v{HOGQZel!6jr9H^b~A
zXgH-(kdjBB4OF!T@BcwnW0&rnZX}vd574;%{+oH6H1Wqzg{p7w&>aU^-sR?~?%YB!
zXF4&~Fyg{rI@v(6F6i%q72#P<gi(&7(zR<OEol6Mmd1l=6Eji1aJC~!Y^)&zC7{N(
z?@(7NCuOLu4@fm5;`aqKW)I)4w45z;-Cm3-xE)wq`}%_4L%YmB5na-E8{0VZySUbv
zhrii=w#(oku)L13fO0Z7TMm5$ShZ!xjp;8Eq76FE?_?Th*7-pxUYe|vn_$%Y%FM&R
zSLMB~%v!{h<dCw(UkD$JBh94e&jP$k9QanNPwNF&$8j#WI)`>gcFkH{i#?ujt;#YC
z3nVkecHLB*aZdldJIW#VD~I~mSvchpgsr>!!*J=ZbeuK{O*E7N!|*EN5f9FhL5~2;
zB8a6Vi>j8m5Y*|4>IO=A+s1hC89EmGw^uQ|uq$H~9tIvtbv}uhU2o}DW|Pi%lJ7ZA
zeBOhxhoB>P-IHv)C}87Hc-ON*ouK3|PnjmAQC$agj^dgPH{VJg5ca(@(v)1sI%RdI
z?nokNO5sX5V;1%-J_a?4yw+GXmj**#p2MbW;!LF+l=er!7RZdAyG_H7rg7N+a+6DD
zi-mG9c^SbfPJ|~iXW;)B(b~7)B|QkbhOrhQp9y=-O8<|WggP&B2_W7rO~R}0z?0zy
zW!D^8mA28Uw_0wwzdeeiPW0zix|j*AzuDQPYjuwfAA>gPp4J~QwL0Oz1ll$?(agAF
z=6g0N4$&4GU58U;o&3@F5M1F2c|c?Z9@8*^q;jZ@;6%h(+RtILfvpBKYYNUR<h-W0
zwp5$AF}(7VzyZT_THvvAxKl{L>X*&Gix2<$k79{r@>iKOVsU>LODsY^fn&D`XtDtA
z8yflons>Ta+cJJ}y%uw1yX%w8g1SQ8yS+b9Oy2I+raiepE0{N7DI?>YN>e=pB5s_G
zG$G$JctqP*m=_gUT-oxb5SOfLcvyo5E~GReA^FlG!806gkZWq{_prapNKcA4Q5&aQ
zY?oJEXZ;yb%}daedhl~^P9gG@iF}k+f9fnzPW0vTmyNVu;!e^87u_pSekVh6`2y*A
z`Fkg%ul&gk31uP&jb0g+M8&9-C^KK~lsqfrwbgMXx>mn-ZnG`HWz{3mWlMB-w9h*b
z=DBA%^7i9-?O*O5yMr`Hzcf40fsd?y0K{EZKzZgf5fT0X>d9}yk3T3TiB`|*6k>&i
zre))iQO2EBYZLh#Ln^;1QdOv!aJ>geq~@>rEY)A>7fnf0Rsz7N&0k=)t|XQIa=YdI
zi!yiu_1`e*N{pNFH@P}w)B=Rb&I~|Eu?%Bh*_U{&hXw5=g32G{zdA}iP3999n9Dz4
z8HG0rf#bWA%JHu-DQ2+xg}49JS<IExr&o9gR;*T!KV&Svm@emP5x$>aj`mr&yVOPS
zrhSS+hW0T6RI!3|WY$Nl5<*Ocq2*OO5{HLQLp^gzUe@>0bqyYZ68*~Jv&%YVT>a5!
zuM5T7JgB7Lf2gT>DBNZq=t*Xx`cfZCU^pdWo=491DTXTB=7~>F+WZY>>A4WN-#7P_
zTk(S2!c+VY`%#Hr)JCJjmhZ|x^$gpm>q3L<eBTbt=L~>zpRT<}mjth+Vojp#Qig@b
z{*J@28O{6Nu8`eVd*{9%C-RuwREhm2Jg@+rk~YjXe$YHba%z(cJLkJ#Kl;v_TX@x|
z0sDKrPPMCpxyo*@IpC{b0!}@n=vG&-nNyB?I;%yX@NZmD=Uu>(Z@bD(HsUWEPj7K{
z(NKj@#O#W8_PN20Nn#up#>nSEt<|j<Q4L7Y-iql(#Mn!$X5LA}fKSfDh<(B+oe@hx
z$Tlx^1d1?1jEL>U?rXbe%s_Q-iV{okI@#xIbZTtgmu*t{{<2Mi!RH<8i*?%?M-Ua8
zSS*w+8G86ec8W+2W~iiHKW}Dtn_-I8qJsQ|8iu^l2m86(&!1$l%cEC7)-!Sj|DvX*
z&ZaH6Y|c(FwAY89;ZE@@SzYOXbm+WvB@s=@b@_Gnl<sFt3b%Z;2?QuHJXgutu~9vo
z!5$j9Is=*GN*+tOcF}eNBaUinaXnuhi-j<~pc8xx#q>0A`2+hGks?Rb8$=DV13@r`
zU1hH=PI=6n-l_UcQwY`Hl?*?(LY2_CEf*y4_q5%6o>$a${r#X?jR%#$VBmX)qRDDz
z(fXp83l)0Lp11&5f0p`=VyZ9`PE{ZQjWxB!2-pn2ca06SU&Z6XE`~9vk#8TcPi}VE
zqW|d@jP`FPmamw`X}-$;ask~EN5_@XV0x1$*>mZBLLGTU`kOoGFwZq)jk7S6r~gr!
zj$(;v_dS{9Wl;$Gw?q8x!~Ez~1x5X5?17Kdn`K+awlvy*J$i(6VwZwkE)f6NOO}C7
z!B|k5%r#mqc9#8qs^$>@mdy1A!k2Sc&pp!GPd$P=_t=A%gY;WpW4=t!TbEU>Jgzr+
zPc@!FKA-xntWl9pzV+GDxo3#E!Z6k@m!^95os+?0%layN6~l#U^*H^XZf~x>u@MSQ
zJ6E?_l6AqgSP$FIKSr%=Ftz)uW6!M{W2SSlXNP^-ML;<#6IT(RK9}^+2^j2j!uh>d
zch_Uf`uZCD=Y#9iEBjhdiQ05eDxba5vfS9_zbB_w*;1M}C3~n`9R7Zd&q7SJe)zT7
zNMw~Ob$k3Z-fJ+nWl@piRs|JC79;~z^z8CpwCw=liZ{pggxT^8K^4|j+3&f|>1IU)
z(Sd0HtL>`y4kNA7HO60?$(7doJxvo-ns;eUGgyeeH7_G@D+M1hf`x&rGP_UCV~|m^
zasNQQpjwR5{4Dj78Q8_R^cRBTuPcOMlMYwv8%)qeEa75wmqGw7L@wm`&~Wp|6e8Wx
zcdmKa2A5pj>yNF{DYz7@Exh$`GBN)d2q0>*2d?UGJI~KggUn5V5e|um%zQlIsRyqO
zwPe=ksDJpNI;wXZuA6dSB9cb(ieR^H1>xbxU6J2}2?#0uSj=V5I?Z+@CEbYBR_ec0
zU@cOskKC6pa9`fnWjmH+ja=IM?rzuNX{eaX5)!ssZRWW!44kU-vzfhKyQ5^DElwS8
zC(6eSHGKs~ManN#DrCMaX4bb$b5JeLazDIoT|}U^lyRV!wll?hD^(;|^66^j2EX<2
zg;5s9**2D4Ys!W}khN%>sJdqJxdG*))=@a#1|AG|p%mHT6q0%ph4pXy;4j~8kq9)O
zkBLql^W1;lDcYdG$-o~~&NoLbpvZ8Sf$Ppo)~4VY5kK8M<zuM8I_%Nr@(?LI->4s7
zkWIy%X|&ud9S2Vkr5zYLv{kGL4QGnp%ujoZ&VzF@-F{Qdr8dg>#>&!ls~N?P`b*5P
z)#V<+PMvv8&|KhU$NnRr=?G52Hy#ml8>ATt2u(0lEj75RQ?W0ThRCyf$=5Gv17EE{
z)#|+-4q0O=N26$y6~mNr!jFn&bsuKC_E~eWr089k;={Dwc*s-`*nj_dOOU@`a)my1
zuYg;HtrrZmNeHGL+JP$wGqe(<49g#y*5qAZT@*AD*}wkE@KWs<%x-I199l?GpWKff
z>|hSw3K!n^JPjEHzFysi<J%7vvcex>BvZQXyL-rF3`if|`|36=G1XjzS<FlnpZ<fi
zW<<*#H}+v<`9g6Xw|v{&M2_!>FgzFeOpLTwxWYdx9DhA=OxMlSb&2?{vmTk|pRGCz
zg*t*wg~NSh%La`r8E5f)Y>-L3QLKJhGoQ9Cjb86V7wiWCq~7!l40{+NcrQkPZoAEX
zBo^u_?@SPf8APluv5v^;-tO;H6~8{_tLV+BS>7JBd4nI52Twrmg);J5{lckE2emQF
z?zytwWhhexuP%WdIcZ|0>x+eShh-@2okKUzdA%_nr{s#isI^AtA?31t`8DH(ZnrN0
zN<ZK06fC=Wsa3+~Blriutrz*)_c||)bd6;H`n7=H?C?<uv|9bOvWU<>&*KBsDG##~
zME~_kcC=?&?Fu{NvvC_>XHGCU9>rzn_?;s%m!BtlUgd)a&q1wHYu)wHw*(VJpFLnY
z24>iFo-IQeQ(~T$JIT)$Fm!iQ#7%&+Ao17^%6p8ksB}?9GKNE-BP1E5g6IM!{x|v`
zx2yYz`DaEa-=4hWFHybqO@SDP0|5^rE+*&p4-E988J(Z8&}?&`JkDv#E@%+80#fV{
zKDt~_<`NbjMKmHE53bv3d98KP7sFyP0t0PPDXKZSpA)AFQmJHVy^-!b%|KzcT%B?r
zWPsxNM15fMB;SyFsMJy6R8LLQRfufx_qEn!HmX?85OoF(K?M}u-bL(LoCLkhaSaTj
z(taa+3iqS<-A@y&ji8L{Q%ZEvLJIBS487VhZ~KD+{T>qe2c=0%4KPxw)c8ff$o%Da
z8kj@B4S`xDGHmcAgp~cx>qy<8eZ&JBCVWW^6V1b?pGzA~@`E@1G*lo7j5KvAR~F&K
z!%yUN5Q7re=+VsOYo1V~@m*nmI0N02=Yu{~{GtmJs-Q?%=aTE`fbuhg?eie+*Mo)^
z7nPg_V|op9#8e`Y+_^9N`=UNanr&)D`iEL!g&0c>FxDB*Q`=Oi+N%0~G1^(VPAliG
zohR$9LqHW1`suS3J{;hgSc<%gW%7}TVDTlz^Quct4p|%+_C3Y2R%)3HgSXh1*}`E;
zob-rKC|td!PZ~cDT+``0>RltQL21-yUUzd5ktpuaYsXYaGx^_)ZJ!^63R1e`+qoyg
zs|&8xyr1o@M9qcY3yD50_H1pMv>Sd=9dUmtYvXoQPT=p%>*sZmur2rSi`gHiVln|L
zvXsK((+jREC(}91?+}3OG<$r4@CmCt3YABG{jnJZ^%gwU3G}{isc5H)y6_Caut(iC
z8w-IK(Rzd94sZm<+u2J<ONlPte$9V5`uPnRLqev>yDqv}w#m4GtOd#{AwFUW&mSZV
z6h2@fx&k(#V$L1qAno+MAv?u-b0dKCuzx`PBI7gA8;Ze0TuK+J`L}LgUwAW~qUf+#
z0om4>DD|#l?lmarp5tg#rP*(eprOGqrg*v-nNAaYBjiqx-Hj*AHlYdJL38O!Nr{fe
zkXvPj%~wgkC5l`aT}XF(kE)@FZL))O=oAJUJ{lSpv%Iu~DoKdgcU)c}L=skZ+lWG!
z)5c-@PK;fHYu1P8vdfP`w1W)_<@JcCu35Ft21@63<uanWgr8fPe_go#BPN6KYu`gk
zXd~lnil;K$#X8^BYbJT*0*y)UbYU%dTTP{)S2=}U9_>Od{i=tr#$LT9l91Jo)VSjE
zrRBrtKwO(ZyT_%v!BGuKTDR{#_^^8nVJcptY@x{p$JDW<+#hQ!W>aBOG)K*KHh9Qz
z)nNs|qS$qIRIS$A!ftR*ST_8KN-H5s%gPq6HhnIgW4F><ed(rJA4QMW+i>+EHOJ5}
z+>=*Sa_nQ4spCrYA}%yByUp^?nf)K6joM5XsI2iFU<J6!OKQ7V$LN~pJl91mN)7SC
z9ZNt&T*pRoMbk0;(Y7aBAvk)~T{kj~aSGAI(*702{Pj@%6<vy>p`+$KFiz6x%KOlX
zeuY9JL7xKfDBBv@Q<|wIdtt1@;6&y6#mYGr`#jlMnwRbVj}|hH>Wf?ih3r+Hm0zH5
z*U;K^y`SWHkm<szYoF&e>%$0+#;GnD3WpGnv^VOfR6UJZhGAX#y!O@kn7T&SSv`lg
zYBT?nGTJX&n-SBqDcc_!Eq8hn-3khNvNZap53!s!2`2emWU#L6<-_!*+Rx9*k1Lp!
zwsnDM%I7k0nvC<+*)z)khB<djXPNL|by5()0e+BC*wwvZuo~8`V{1Sy`EA(PaD3e7
zntjTP^7-`jK{)YtyQ<HS_9k^aCe)vQy}`Gtq36=5bM<`4mv=fB30JG1k^#z4aW=8A
z&f^hhV6E)Nb_|u<yp(;1H}tkkxK0pw|8j|*h2zD7ukSga&uV@JEv&~s$L%Q(Sp~$&
z?JdIRY%H9%X9M>v@GIbOvi*@(Zu?a={j8o9NC3>M_o)hOqe7~|{$@6jp<P7o<<p2Z
z5%hqQGN)Vic~SeV_dGccH49A@^Et^1&ro&_JE~Y3jO!iUIe*3WDw$T#Rti;9SAyf)
zZF7jDW~uA$T)CHCOl>mfdv?D~M#uam{fi&>D+gj8-m!ce=oqfcF<j#6nHdEgyz6ap
zyRXav9l^CLOrH4Ql`-HJUYTF!R9J27M$+nhY75;q=JsnMPnnex<J)iVX65p|&$vld
zS(iIGUSwT4r}zEdT-{`xZbc4HxO#P;8P%)n4i?TQi!Ukfx|fujT{Cqs&vK%z?wfZR
zNAJw<wP<(DD|bkmapj*+$kO|E81$4-u>+bQ15P2X4v+AElm58!pTh7GIXvld`*Y2T
z=g}%N?=6w~8l#X^dw_ik=VW*&>#|;zV5p`f^KN*roa~rTz$L;=x9xMjF3>n^s~>mN
zTJOWhXu<GW=@dBl*DUMTMld6Vx67fI7x}d%t|~%I@O{1v1)ouT2(_M{UaMfX6|k6+
z)h_kEMDgeRjX4a^mE5N`5+I3YX#yV&4rh|pkid^$EyuQ*P)94SO+~+Z#}OrX{oS$C
zu7CZ~d?5zOiP8i%skI87qDIa=Qccz>6HpNxaFCtBhqySIg-E%~#=kjCexfawQ=gh(
zU<VByd}r_sR|<7KKv%H@-mouLay|T))v$H_sgeLYk6kO|x;5WP<^R?CYP5}0n6Kks
z+5WFDiR^p0n15XaUAiaZdF;`wr~eAW#anS$AC7~-ZUvff=V%s~HlC)?Fr5rlxfJ42
zT+NtSxkLBsPph9k@p}us{6dHlIM3Z$BX-T}v;lw!|AVjbqkIOkzh3h1_xi`%dc1<j
zUex1|T=>sdOLoJ_)-L@DJFPc`7@0L_PZX(?5+(opPiU3KArvnnWmk_oc}bfDMF3MM
z#eI2Tf9^N=18Tkid3N?|%QgS@BK-X>|F|^CoP^@*eX*<j#{U2I1^(?%e*LvKwF2Dd
zh-!z&7yo&wfA=SUe~AMkSd$(L*1YP!Ikx|E`~LZw@MTX3(9nMV<ZI%6^NXIdoN#tn
z?e0Od#_riW^<_)%<|h+F?lVg_s-{KKkJ!ZBxFRnZr%FU_dgDgBerQv&t32Ve;fgOL
z%rZ0Ts}0yoKaUb$I+!Ne?})AqJXqVTne#Zz>6k0p)pN@!?%ULj^$`5H>%yifAMp1)
z?Qb-8|N6g@lW|nRO*gqYFoEHh7Vh6Qg8%FP1wJ?p3;KWL+0n-6wzHv1;QX_7H(c?p
zIa1|Cy%cvD;B9?M6@?Xz_fPQBzx-SL8X~T_c2V`!FT|-3gSWote&Z2@c;@e5qw-zO
zi{zawF$Mtw$fNk)fs^Gs5O#ULOBXG@8N)_`edgOF9uccBI-I)7c6m!$jUl9pGc7SA
zCgmdUR&~F2>g=98+`Y9MFBst*SW9ig_SDh&u*3epi05d%_dq2{CU&>v#$WvfA*xf(
zL_RlC?!3ME*7UF?hQ0lB6hk{~lxz>%(XjIzng;Iy-(qy^V{rQ7xC&=1NhuYOEd{_s
zlwV-j))y&}8^nf$gtYnL5O0Y#J%TCWch(~n-ZKi5vUF`(g3BD7jt#Z+7L+soCZfwQ
z8=p6jZzyH7W)7aHFTtHgbsAIuW6@5V`;&5H4h4pDi_O~z-5+@bnpE!2)rW6<Gwt4W
zahVVo`n#6?um8Y}0Nske0Q2%MqvYso=&iAw0&v#xfia#+=C%j&Y=+h^56tKHiDv76
z9L?6N9(0GM!#hNv_Kh=$mo|IpdwN7BeBs+2p+n(NqS3<qqv_7Ncba|615PSan7%g!
zCG9(isjdy6G~P<aY1sAO(OVAkmFebhIos;qZwD1!w$T@twHFHv>jP$AXj=z|RE;mi
z_9nh0-~wk<0eG43@6u}OKa3o18p^qt@rJ_}92ckj*YEZB4{@I(U+~(MWzxTvY4jy@
zR9TKDD4ZPqzab=}yO=>tm-c*}E(bk?+J-Tx9?zc*$#DF1cMLrJ(nbji3GXAL=I`K=
zVAAceH@L*KCchvPdQ>2?di{^P^JJXh@GD4Qs0DU-di3PiUhezc)m#2dmgHXgI%3zQ
z=CEz&_u?T(YEY1_2pcOZ{PlaZHW;qEt5)EjTm1d($#_(zN}4Pt)81L3d4ERa6Og`O
zX^H*kM*#epSmEXO%2!sey|#3lX$;p{;aqXucTS2eQFQ;k;!=E{4))wWxxzikKh}2s
zbCh^od6L)3z_@0Ck<49aF>ivC)dJsFc8%QZeb%WKx{t8a?)4IU2c(|SFxCov3jSc8
zYUVQ=5gW0Cos}#jF~7_$g?ZvZ1;yO4&rThWjz1B0h>N&)5=4LDpJ97p>sUTlSyO(p
zFG3TqasnymcdjxZ2M6sRuX!b@lQ=$+iHm+pl`A$ncfXZWZGCRg$K%F$WU&A4%kp@k
z6T#W*d3TbYy3A53h58nrnv?h<B&Na~WBBfHKc68ZX<#}{%wSwJH(Xmg|MF|U18aSW
z*Ee$UE=I+I_oOg^$wuwB0q-=0ui>|jJFd=}v3x&v@1HmbH!krJvk@D<enWT8U}BVM
z^ub<2pvi)V=%D?u-7Upj{VQYJeY%HK3XeLr1}6_@50+=uFTBumyV-)@r!P!t9B%0z
znSVYKONhHPLGg<I*)L+&_~U~@iMdu91iPOuz1~eWep(?YC#l=4#fyIOBUh?<jC6Z&
zVs&Ev*^<5JDCpUQe>Xo3z$b7WAUTp@`h4(L=S;Z{16Cz5FSJIEC6K;0KOBd=#LTyp
z3EY}5+J}+@;vHmK#V<wNN)UclfT;)+Y1=NB4jbVME$0^@@n`8QR8Dw-?aqS`q6W(B
z(h#p$HDS<{YU5#Q=jn+Y6OL)`LkW)$fgPNnegcg)4!X9lU~1dtDdSfx{EKGcf|EfF
z#X{axV4j?!M-8RCcXAQbrchurg?{raNZPoaWugDp21r2*XmHsmGgGFZQMAc$oHDFM
zgu9Hh;5E8c1&7Hj(td|ig7@^9b6ilvQ84BPtk}vD<RT^r-3S>0K^oN$OeZy=4#W!p
zi)^dUJ>f8(qtGn!!{FLA@Cz)`9~J&+&$4$2Sffn)aZus$gSTj^762b{MWKXu!yagq
zad5=Eq~No7Q?PEydrR-)W{U{HRJqj<KVbNF<DicAp4J{*QO|6WgDR{$Pk&4!IiLc1
z6TD#J^V0GwqK=+B#bf&uN}K|)Ra_=duFb_8BV!ZQ;orj>PCpX%)|`bSfnP>O#uPH_
zS#VS2L2|VLVBy|^7sBgh!FUg|n(ccS*e3aM1~tZ>202HI8Xv!vBgn9q@KxjBW7oyI
zt(KDy;kfZyzlb{>_dSjbT;cg3ljer&9QZ*zV+2Suhq;@QNyG!VCsk`GXU6JrpR?yA
zy^O~Y<5veIn{RJ|b0ZLD9r<pSj;4{~C?NhTk_@hhe{ukAw58xycex!L=Fm6nW&H31
zpViuP=ELq>#e=3VJciR=O0dGK3PItd=CKHpnl{+2S#Fp-D0+&+i%sWJCJG`Zsj#_s
zW@7boy>2{t#IByr1usT(n@<<4m##-3YAmZms4afCbjP{|g2XnM<H$H9Xd2}-%$eD^
zBC8F2x8l_i&{};gHV-2Hag5bb=NYyp(|`6Okj1{gS7QgkpC%IMvQ$KM?7B?S3I9E`
z@$bQkcLv5ZmiT|{4tb;vAXrexaCd~ZUBP5=2Rtux9T7Snw9ek;sAauK)Q4eZbw4|G
zX-aiJ;S$g`sC#ynK-gjO>g`5-C22|o>5mAcXL31L2JJq<wP{p4nM4GsxAMJe2-Sx}
zG(STFQu&aU`q9?KzRaXA&soxY$WnqxxxHUBZgzPQqRc+RSTfBJxUsTNo@STiV>i0k
z{&7C_=`xhmB70xxH-T)Kr3~W(e5i!6l_KFEi^gCMqry(le+<S^O*=qP3kg!udQyyi
zX$2mxFV%OW@7`7slD19jAc-=dIL%dC7W{-G$MaoizGHHZgzp(95q%7T7XI({!_^4B
z+{zhLSyr&mABP8j?zoA4hduR5!reG+sEHPFp6$Dx+nfO#Km&sTxsN4rALI1tc?!VE
zs&_xkC7jQ@18Q0}GM{JPGlfG8P)c?Eai>e5OA+*XhQNH7dDU3Y0SrMxf}~vxsSqt(
z@S)v6lo?mu7r-(1s1q(~5;|)U-WpKgtX5qaf_B{H+2+m^sms0M?-=Who;u)L(~tLf
zpf9OoL|<3>gk#CLO2X2Lm{ZU3JZutWpucTGXQaqJX+&j1Xf+7q$*k?6u-uVUg$nG=
z?|r<>D?J_O92gWg5Yl<CnZP*kz!}Tn7I&$PxOKscqdU1AKS9g5aN=F=XZLB8)ko^1
zoF*Z|qDR~1s?Bv#)+;g7iR5GoVy|zugu_!-tBr~aEY3^5`a@wyYyUKQ!fE=@lJ?gE
z8Yj-2Q<bNhtU<LEY%K;@K=l(ei9V7+LeA3b*8Og<C(UJO&17z69iAC|ySGZ4sy$Xt
zI-?tkZ|IrwF-~J|^~p)lg|QYI=J_hwBm;ebr@^(EWAPAZnL89d70U9jiirRG5EX@X
z@-6Z2U_bdBks!#Zw&k3TSUIV!x_zLjuD7{vEbM1-JA;neYD;bnlm&0JDm;%}+4V-U
zoh#)lS@gVAxtOT)zOB0O(qa&Ih>T_{&!HkF<XnQpXj_Ol4du7#YOaBzPt~1vG=RvF
ziS%YUHv0=SA?$717GfJ|`5f;l#4?PXx-Hw;PhlyE{anha<@KrM7C(`I;`+Q}aGc>S
z8Ye*SL_AA(;Jm+SLB!E;`Wdb?-TL}#0eMn-9qE@EsLWM35ZIJS2ws?1YZhC?kSh6C
z=-si}Iv)^3fZhzThDgCzTbhdvtqEVTf^O=$E-={?zu9DUWaJ~6m_$67^K3P$T;AIf
zY^qo7Vv3%!BwD&#Wi<%a)M7aorCm_prlr>BJTysOpok={o)ZFy=7%DObKE~b@vORj
zR>C;4tW0@M8Xow~#BOX&R7Pq)HfBF;2RVgk(|FR)32I8(zGPGO+dD5Qxw|U)pyBT8
zkXxKEM*K$fjBfN+Vz8&Qp%%8I0cX9=C%w5@^d_`Pv+1b_oQA_RM^Yn?Dy-wt{o7pO
z*HOdmNn&E{!po&7W#c30)J(xdM%7c@v`HqFLov{Nk0h;s^+63YZIRL6`6w3fx%qHd
zBv$UfWFkQ~VUjNcLo4a8S)-f5Xc%6o`?YSd0=3U+q%GHFXNVdo_fF|au2*D#V(^^a
z4Y|x1n&XOhE}K>$HORIzm-q+~3wTF(f5MnN?Tqzo{f+m`a2QuA0k#`$h_5)E+Qa$5
zdcsM}ApdcfPR<!AT2vGMM^QRXGVoxGUT=|5yob}1f!>C6q4RHbQ>x6y_?@V#1pE1d
z8F~{%Dk-+o_$fA4`8^fVx=)+Lespn9A)n}TI5lzez4+DQ1Xjm4C8W*8+EdmE3iG!{
zXc*ls3^pOMRJCJlu=b=TR3W6U8o$JubO~3%nXX<GdoAE45ym%^q7)-ZG_Kw4X_!MM
z`-sqKYD{zWs_(hTmkq&n&G*~Y!tSlc88zT;axPIoB^nPiR+Y83ICCtB5GUCAZ0Ts8
zA;0^~Q4p9|*Il3cRjco0rkZ8bVJF;uPJARGyhc^u{fLM{bGN(O$x~;|XjhnJul_I_
z=7h<|)^fvzW~V$>2WO-H%YauZQSvw^n?&!gQNKI%XtgwwO7D<F=`GGT+GQM~qX}x^
zo$)|N?9+ri7knxpma?E0Hl*XQct;}R2Nfg8L-KDK2}S$%@nhQomaL3KmczT#XPg(x
zo|CWC5QQ05mxe`4pg3(PRpzl>Upe;49sditm)8oQR!Z@z>owV&OYSCas<88W3P7{D
zp7m|cPbR>)e&O&Vj1Q1kdIB5heg9VJ$K#xU0Ka|{XloUs>PTZB6{nRs)(<zk`=?)Z
z-a3RxjPK$m7LB_B{DL;Akk2V$fiq0qQmjhO2XzN95Ab@Gz>N4I9lA8Rn39%+)$B7G
zf2JLtEt<)%$@}9QbQm)IUAHnjXQi(IKd#=FXcSLV4l%F47bU1kFMSnb_{#G2Spvyh
z##h3@;{C@h?4E)(o8dC<sMv)!Z+35O0ftOR>&(ENKW<sws4KDLrjhU+DV+!~BTpM&
z15U#_|CC$!8W&1iXo4HJar=rR!DD!<r~(atdB}SLT|`^|TSU!_PHKDigykQT<KzsT
zh%vTF>-oa_U!J{5gi*KQ8?3UMu^*3^q9XdLogeYg$8!;V<GDm*m7MLo&OLb<<#jHy
zjMMNiG-y5ah}H(-CrNuQ%CC)9hxy&$o9trQR#)!>u~6qDqB}+X8}y=A>2tXz!#Iq!
z_EYq&NV$UvZKTLsLrSkOr1*qkQ!8#S_VB)lu?>`Mu>F-V%bPF>e=zUPh>D7=c?aeT
z^1F|!v5n<Ac^in|EcE3i5k{Cu$Pn!ljSf6s_|^!Z+*TxN40iuWOxXQ$DYx=#(C5~m
z$d<etZ>n_WdJv7%n&++%r1g{x^ckTZDDPOpJXKi?Hd8>pkhfM8h^N3qx{u>t(dVn+
zvt>**+}0CrMEUi@lb|2BR8*n}Pel>hn9E9g%3`SAtCS6+kgeN#1#`0$ghwyMjwT4a
zbBw-%PH)5@(KrdU7SKYrT@WDi$=ISVH^)uUfPv6P+7PM%3JgA`40OUzSN#3V_~{?<
z^o#MV;P_u<!53Z^A8~?V;2onCsZ*hS&=U!D&N#=xDf!TTrMQpc$8wx1GyV%_rwLsq
zI-PbK&rM1O!}NB9M=OnC#(1$bs<TU0xNzaX1$vm<9<klQPtFQ34=f*_QjFtrV;&5u
z<H;akGRNvIIVyNUSes-HBFj_-`XKe8{PWWBv7ejc#eYigz5ZMu?B->l1fy>D->SRM
zw%x_~{!sc!_Ti<~W>{(+L~GHr0DS8fv4Nn{Za|A_^M`21VlBt7vekc{+y7ICs)d3R
z;_MBMKT~aSbpIQ)__Z_*oXEgS$Ub9H>NOCFt9?4{eNkFxazeIIImvxPeiS^07)^;1
z4bH`X#)dVH$McrU2<OL1+T>+y>Zb~m-ZvUAp_Cy8<IZ#z#XvV)zOfiOM<43kh{Mkq
zpGRopn5h0ui77h|7pIZo%{8Ead_ODQ_C;-Y8zHu_sY@r(qS~2Z-La?j<E3hZUTcVq
z%R)JVM0FyGxC2v!MMKDE@a*#lPpLLRaJw-%RZETGidqY5X7??|OId}!bqMLvM7gFf
zLfF?TF!Cpiu$)bB8tpnjtt>;VEnAmMfm5x^@tlj_xmDIMcpr`Mr8CJkKe=<boj)4e
z4OUe%`&33uFT4Ec{7p8LzpXPg1}jYI4q#4dsBOH~7f;#~NeU;mz6CYQklw+<e=#Qu
zRTk%~m|j20(RNQh5vDHP!a<*g7W?Z6u@g<qL{`ea%6^epA2B(f-#<{jAIXV(78(DI
z0%i5ZkTyVFdXpEV_ApC5hokO$h+L64Iy9_Ig8jB~pdeE)A>Md4Du7Hnu6FWHoNW;H
z<>1Ju$;SLLQ@;31E$;#Zr5oM-(9RjlDbLyCEU>4(Or9m$#~#8yKjyxvNY6QsvDwaf
z?vb*~^c+P^kAHd<WH!8fN!ucRS;qjDwQl_WI>1FH=x&?Fk9X#T=gCS;Zk7*5Wwt+D
zgJ?p(?^a)~?&)Tn6xA8AM?3*rn5W&3h?+<{B%E*mXfu@&T+-lpQN_w3!)OJTcqmuu
z1BWWqob(ks51ILu2YU1bbWX>T2ThH4gfZ_Cc3A+B?CH8*y(j&oiV#@c;!%Z(m?F`w
zD5=w(Z*I&7zL7QMAI^P9XI!y$Gu43?`Sj9GzabG}7qjbOK5vUy(p{!#bUWupDH$86
zh#a~rr(1vT`xhyWI_13ZAM`3*?^N6q#C(N4u_fEj;#?86Sc|^v!doqzlLkvo1EIzd
zRD||HnlJ`66m)N}<4eRiGZaDwk{`paQ8F3Bk#oyJ_CkT5v&G%>L>q!>H%@sFN8DAf
zD|;t;J1;?eyi1N|yE;k1^a3p54w-c<AzutV=WxH)BBF2o)CjvVLUB5c3Wv%{-(`mM
zteut}`ZW#0$y}XT(%Bs6o$m$DBCB)DKQXjmF>UGA>1yLxSv0n4O17Qp%Ji!*o?K|F
zfh36Bs--79S=9dvVmcK<%p1(p`1%20Q-4&LHl`Uj$tmn9S;^H|J>VdlF(r91IirmE
z9>qs0fMq9Urbn_E&njQ5ufx$;|C*HGO>a9DPGIhLsk-1L3QuUNJ-!|MpeH+B5mx*B
z1!T2T<IH4DVb(HB`*~VoyFjfXzNVGQ&-q(f6P!w6O>}!tH$7Gl8X5jmlK+<~I~tBO
zV(h8vldbeuHT$hNnh^=ho&ss^C*yEGsar%v$)TI$Gc5`iL`Xj(8ru~)muPQgT>5iI
zwgf*&HgrZ*pL3Rtqa(~C%Y<&8vH;~(7@o!^iIiB!qN5Q-1hTFN^_waLa?am<JJX>`
zz5}k7H$1BJDo@H88r;4XMzq>Dcb{30hoMDvD-Q@tz8wU+&SO>GGFd$B7X)E!Gf9V^
zGB%icV&BY-SiR7MU$n635&__{ECGVKOdJ>3>1B_6FObxc=l@`iK6Msbf?KaK_>=N3
zu)WnJn3~d{oQ~5KU#k0YlXVkUlAzxf9v=r^6@&?<=nQ3g+{S=z6xEA49O#i>2+j^W
zkasAuomMA0xmVpJZ#Gnw6Y(Z6py~qu!?;|<Hlek+sf#<c%s7fU#z^6eB?e&ama39j
zi$~ifQ!U9k@W7Ru`w8%VUB~5SR{KfdnW+<m!EYwHno1|#eg1mvf>Kb#ox5}$!CXFK
z{Wwt+v&Cc1b=K4M^)x%f-8^>r4ZO7P8f%_-c0=8(k6~l(YeJSij^1V`7IvsF(ZTF-
z^mJmd=){1E=9vLf^|E=B@vns=hd)N^1vx+$mMMhdXV8~)WSM~gsdp8;3CuU^a8B|$
zjXpYqQ0#HXCpQe5Whs>1P1eV5of23J*SGjD!WlWRc0qD|d1JWiQ0L>O;OLy^%J0G3
zP*$Hrcn=pz2G0Ddr_-KtZU$LaLdfHBT|B*+syb5alj}~a@gx=`VzyIqZeZNGWsgHA
z4_n1XTUGUqA;}M8VaYMzxm%V%oUqt)E=7TE)|VDd>>w?t`>OryEm95L$6YdxP$pwD
zv#%?G*5kZAV8O(NtOkp^-@D>*6G}=;Dr4DU)mbM(F5?I?`5h2CbaAD8rU9!58;>A~
zaYby8$28sk?fH!}6WBHnJ@o>WX?&i8T-LH*4;v#M4_hVVM0`2$WUt4G%5u1@Ab{VW
zjC-rK_*6s)&}j41;~#}s_2;LsCqDjvxO?+>DBJdbJWC9U8M0>ILzZmG*q6wzlq5pP
zmMx5IVPr_86k247QnEy{?@`tkS<4#AI<l|*j_baw=XpM#`>Ege`{(z+x#x9fX0GeJ
z&ht3l$9svXrzNiRUGeTAk2{&YhQu(BtPv|HDcbqoffY)&9SNO2)Wh$kB8)wurV=?%
zas^!PuqG<+sj$4ZVjn4O_7>`zz=pD~i=^QqdQ`P0V!M1^b#)w`9V4xM!ej-v<<)i^
zM1W3d{C>!@zkX3@@s>YOUiV|~)rz;ex5`$RVqVhDEqEz}Ks33}05}(P=2}-POT`jP
zaMgG$yspZX=#jj*c<vq<`a+f<>KV7g&i8;W3YdopVad0_T($t3;49Sr#pgfWwoodb
zdL))8b}~JM6?*0nMX-WxV2j`|C(D0oT^&5hh_RbAh)@=K24S6_?4D|Tp?b@>yc1<z
zNwvZCtw3%lqQiYIuI`-@%?2a<Dz0VoL`llwHM+HA^|4#yDN$98FW>dfC><(bb8QtI
zD_hZ6oPsdsfq=ZF===AEV3>K|;5Hw6bGW<~ZpIDG;am<B<?ubm6cY&GIL3Gv(imPY
zgF?1jA)+)|saHZPh<P+<Y`|wWpa!0)lvmy>+!p)A!6l2iTW;=8hs_=YGf!(O`-z!T
zlMhKzs-#0cr!p~V^c_TZq^%KKUpTa}wOw9){UH?0;xtx3iRU~_f=Z_)taDEycx{n#
zb8S7Brs}K$S~>HgLOmJx!Rfx|Pi}>yr=d-Hfu2vpPO7tbK&1SWp;t@q1E3H2T<g*Y
z))->=E5NLIL2UE^C1y7GMY$3VAze^<TyCD)VGr|CFM-FI0-s+ve>l<~hNDtv*c^&0
zHqHZ+1*i19PHiR$s~oY~B2cgu8{Cl-c^ta`92qYYp<&bXm?HPPQ29pg{-X=!;L`Dh
zqQUKaY%05ECOk9m3+h*G^f7hJzG)u;vJHHBKZ-(02;|uxE)q*3*FhMXI&q%p)akzy
z!|YmPaS3TT3ZZ#4Q`?toLpi<%)Umt#x3%H`tH0lf;dyvEOqJU1t39<L$}IVIdS85>
zRgy~%E3OIu)clV53ABHyC;s6Z_^aHnX{#j%!S3$ebCOVQS8ZWCd=R*9XSREPPIJS%
zwL3<K3uwMNe^2A{l^{A7jsWiInuF>OJ}(<N8l`naxk8Z}a&cmRI}dSL5J$dri>hzK
z8zwMJ_?x&c=vgXynFiS7>C()NMcR}Yf}(HOKfY>19%8Z_On6gicXerqzOJ_{0G6_6
z?=nI>{#>8&RL>AjBkn5EntJC*T8_m^{2UYNO6=U8Y)-P%L4Gmc#Ff83xIrJ}9kimD
z;@QYL>lxZ4JhxDz?H~S!M&Fj(n8b}RLH_S3vuec*%UY;)fY?BM%w~<7X6J+M5uewD
zl>>Y#9(ys(0y($ARwdrUUmL$yThk~iFoAK-VH~ct%t&?%p<9JdW%Uk7e)eL>ZUeQn
zpRlUea^+y-U9c03DjQ%3%tRNYU^<kwvAdl@H20h(BBPynV(`F-?gTCcH5G#=HE3p*
zFw({!(e)2z=AAwOnSnCf*J|f{O*#7*=1omgR@BM0v^8nQugVV?3-Wfpuex%uo!zI?
zWFS3sg4aQEWZ!GTms-LLem+i`o8hSO7{UX65RU~;HQG=%npCi?7vI0Vk5_qS?>Czj
z;A7i2Xxw9}ze^u_Iku9}(+Y%o_XdrW*c&*7PKE1zR&So;&&rJ`yn#`%#SBsx)mF3#
zY9*h@-2!u^ZXMilzJdN$q_rb-T}Zh?DFRcrCe|2F=Q@i^Z3Fza7Eh<XH5^UA(A}9e
zb*uYW2a`n4p7=h8B$t?{sK*!16VPT&bW^6}*=6Tb0>8a(SjR}+#qgXpT^PZ<(Md9V
z?gr!ZLoZ(1r9P|QWE#Hzb)d2wV`5`k8EDy1OTWMoF!F7z#m6d|<pbf&+cSoQ9lbJO
z;`ek4Ha*RQj`>Eq=f;bw0;C(HGcjs6<@IKmc+T+}RWyIS?HS;lRjI;!D~j@URDh$+
z3~nW0UhwfKAFmXjl}V);{X}M|V>*Lpp+1YCP>=LWm)u(OGc$m#z(HL_KldPNTq-V~
z;#$Q@#por!T`7_LRvU}ur0oT9(6?)?P*%)Ydd=OG+|bCKv#s!Z8IT%qVF)wmv)SeE
z9AP;)kv&7a{fBiqg3u@K%Jee*tp8LE$-~G)6l5EZAX{KQ1N|jqJ?zulKr1$T8c8t0
zD$L^fn{!{IOQTL0%Bt3#-8+9i>G@C|I)|2ilTm%Jhw=|_q_7bPDn*SlbWi;lxTnSd
z29_5?%U*6k%5N-5&VTm##V;39bOiJZ;;%j4`Ci5M0ge<z8V8WvZYKPSDh?ANKj@0x
zWE{M>tp`>xA&r6mMkx8m>h%&#^zTysYOyo5cMP*&4$#|}VJzHXg5SrC4?sG#?4>Jo
zh^I>EGwmH|#V_3YZ+($d-){uOThF!|bwZMG;_luzHv>Uv-a+-T0WAs33hK+VJp~%|
zm+wPV58HwDYZ|m~&+|0+CmB~|y3ySKSuOwi<p1%)J`}!~y-h;T&z@5_k{n%@5A+lV
z(7}o9x%8?d;y!*AVQOW;_ZxfT^^4{*7P$kTnP}-1wFac@55g5K^M*~1z=3xl4%Ln|
zng1Bg|0kC4@1MR?6<}df7V=+CL%zpJk<lY;?M~x!Fy(as&q}AY)?XttVKGt|aAhKX
zMWNSJWPKq0NXF?MzwDjvfB&=p_!<L_J<KGWPyPU}{g;rUYpS{!SRoH|7J%#Xqy6|l
zdx1tSGiInj<mV#)Z=dvE-cQ&APC7@TW$yne{`9|nt^XG<K&EPqg$ps;I4vyrNE%Hb
z|6kWum6e|ewd}b(80(r-1bnX+2hz#5Bc|gQyXGHX`1&GrmTBor%Rjvi6?iUjEr!4l
zQpV-+@u#<Mop{CS*<Wz}(;;wIVS8WU@BJR;oT~4S-anOjgpry&;<yMiWqwYZXtz6O
z3tcFOqVJ=mZWoixBR6*`D|&xBRS3<$oc-ydTWv{=N<u(6kh+Hu!dk}X3vPz$a1`T?
zCX}AG`2R!fA|UJ_&v$1}{kYfpH|T5P7SJaXvfi8p)|dm(hsR;E)@9<6GV_bn#Ygs*
z=U`2SmH58%ij8qx?%HvjZs$R;4ShpWjW{DCV=@(k;FqA7F$gEuhwnw!GAAZK!Wnlj
zV((0=N+@Dt4lNBoC*G$YpcH9Hywx0N$AN*ArL^Zr1HOiC#D3ZhX0sPcn=mVPAHfBy
zI=a=cxP~u<VV?BaSJ&_amC29d`xU>ve#GA|j-SH`I7!_9<sSA=Pu2h4nmqmo&FjK{
z(7aS7^DizrF97o+83GV*=~NRl4&r@~%m~;~TD~TG_IrZ&*pnfE>^)f#bFnJ$Gwc-u
zt$fbu@-Ad25!%~)G_#~B16z=`KT~fAf=~|7FpKM7uti8Ap!PXlaEjMk2#ah$tOq?(
zzk-_)u$(T_v&cFpI05&g^dbEAOXi%s>&M&!)+p8h$zhG@1wBeeT=bzAdM&Ym-Xht{
zG{cHky1mU4|9DRPcvJNxLU20Cx^MZXK_7L}4)XIvnVoOV(;oaHfc59wggZgU2O5gl
zC{c3TV~RuBbGKb`P39tdQ3ntl_I;)vGxav9pjYQS>fEhfU^dp|g>1RJ&$CtmUoX+s
z4bT;W4BLm-dw<Jr&3g=B9Uy~jvi`t`&|}o_FUUB?qm9Gy1UFqdL;Kt9Q?(RnvP)UC
z*?xv+XX>Rwuj~L9MzL~;&T8jgS?6QMVb%H4;bGOS4Z^D}t&x6uajM5mj(uIi4hSjI
zh3wg6J!&<s>0iJ%kZk4klyStb;(B3<(J!d;*HQ4V7ol}X{a_PI6Yz5}g+ksl6VI|o
zj(I^kvN^1Q<HjP}1U!FTfeD^t*R0`r-9Ob${O?S#poRa=1Z(yBzh;7Y-g~%Q+eVe&
zfEy;;2%}%0z2`Epevs~l7b{)S{gw)^ts^t#eze0Wd-uv$hq<aN<X>ogqug93YsG%J
z`@Znu#o8bumVdFfEjzU;;699VQ&rk&*(-}L2`RJeZR+X|@ryll{(SkmN!N+obm2u6
zpI96Auw}RI5P8~Jw&L{fI<uUy=VOP1jlC9nJXJZmDJ7d5q@K3{w9~}pR^l`E?G5uH
zPwU6ULcXIx@{|?6l)aPdW79q}WiHirJqq!CTt#0SYx9EKQ(ojw-I<aeC%NuIvr_-y
z`m_2a=Ain(8d+r>7Us{jhVtzLf&2Nt<GQbp?%4Q5xl~KSRg}D&rgN!yYI4H+>cyr)
z_t?RCb_UMh6AVX$;J$9^_wxQYGC$v}0ZEuGY+iXP>X1!Pn{~3#!2xVtCT-@D>|&)5
znkr&|2r(scurIuSlXIBEC!rK#s#-z#cdP3KC{??#AZv2~t8=8IklZVOUIvLtRUp=o
z82}t2yT=t=wilb6-0ktJ<sg+>Y?=eUO{43lvJZ=}g`DE|REBU&?*`0a{@hPg(@1{j
z(l>~XlU2I7D0zihLJsJj(nuKbWsK{a)KCS88Zdwix!uhNO)k6eeM!0^Nu%=Ey|3ss
z5{8M;OD|f1g2NjCk~_fMNCu~;B<By0GYgUq`8-};f}Z|>v}3yrJe>z9IW8DmbWe6Y
zbwNDJp6@=#&8&zdgL3$}Q-b-J%uJ$u$J|G%?d`RWkV$qN{KYoJgyR+fCSl=xaoBfv
zA?VQovQe^N_T?nD+1Dyd7Z#3?wICt#`<wY1Q0~-U;T_*14%aWO19zsoU_PK7$zxl9
z(4QojslKTIDGCKoij*A7dtmE0DfYJSgH?<Bd?=4t0lJYs@Rn?^q2)(u=HKF>52cjO
znNLsWg$;Kw3F^k=0=uT$@Mo=CE`AR+C!gBu&1iSUfBT8`){&Els}0-wo*Y1iE#04+
zcX)qtskkRhuK0$Fa0wxCg;&xo(DSrG6Xxd)k?={-ak`v4J?+kH=@B}I(@bWt;v87W
zPZ#dI`b;@re15=#YJ+n*84lni#IgD#%mvHnW}uv;n7afvWO@^MqH%m(2bbHx9_qwO
z5MB$0Y2}x-n*wMy#=s{q^>9z30X5%heH9UGCY=gV{;^9DD{SA0vGpleSBAifpcUq&
z3?+j$SSZgOW=khH3&DSsF(jO9;NGLSt!E&}a9Yl2hcwp*0iuUOeS(0o_GBx&IBWt=
zCHh_*7t;%?`gSTp*Zg+4lDJLCT}b&WnA<<+$1wdiX#=${Ze<NXqg=aB4|Eyq=gSAT
za(k9PE_jOVQzULTmT9vcI7?{Po6ggLLs}X%WtFiD2<3M)W{wP9e>!+=Z2d5!F!RdF
z?ROR#o`Fqdbiexje`LgmVlmgaPX5chq!}Fwa<G#a^`6tdy)UD?mCIom+lS0nQ=q>n
zhFf_t7$RR!uEh^D$4wo_w)IYacybCu(hIxkfwp&uha3DqA7;T8eHpCoic6S0FEb8D
zk6#gfRRwN=PD`92o1Xn8o_26_Y0QM<?!NW5d>D*KiH=)0ppOTuc0a=@uS=-hcB$4f
zw<-6XaUP@e6D6fDhq$G4Rj331Nk5}2<RyM9zjsTBlyd_zRxL0EUmpVMR1A>dam|=!
z(i|-vCzxTi6_p{_Xde%+ZcrA1{5{Jhs>T@**XupGzmG46<D4GeRkwo~bLCmvBD<ea
zgQs&n0PAk2iq55Z>uUxdS6@Ggx%~BR-N!>9gAO`b;+xK?<zcI`{)G>TKc6A%LEP1%
zcj!z|j6Wglzc3)@*4Me>$$P7ix3ja9ua2X?GQdv}M?7<Hg17w{%R?2&^JMat>fhsw
zO>Y%l{0hXQM7M#8(paG`lJO$cgB(;_4k7Bo)jDlV@;)Jq^m?v0X6LCKlj)tjRUc$y
zWh3PzU%V9`_ftNa8*uf7QN)=@0_{`O{+pGk_#D)y3s%wG=1tVt0-^A1ltbmG?uAF*
zyiaStI}q~4r+i{XOuCyGPgutHIOaFSuq&8OaCWieg2CX9Poq*z9{bhHigMk1pS`qg
zw>2Ws!-PRjZ)6|rgocBnyqAu~@|GQWvNwTeMC~bGSWhZ%)Fv=$OgEwm>NY6n6azJN
z9xh*){dlsw2^{m?w9M~5noz2>S^{OCi&90)8$ZdKz$QgT7ux$skd5suj-g9_*)+#M
zvMxYf(`$e?${X|MYtSO&Tu8;<D|$8)coYZ2#_siz<-4u}89*U52$aXDJ!d4TepX+_
zJK{PPT6#(9riy<4&@DHSeTter$w_oz#n-iM7jjc2>rL<#kE3gsRH`zD0_YnJ$>W`m
zd>nK+;(b+nD>iYa=h?6D?tg5!RL_%xbEp2VX3G9r4Pd}eRb)mZ&ut3FEJD%b!Wn$&
zvT@s1Dy=8ACaL#$t8U>1klG!uC_4$(4!%a%Pmc&B>5=+W>XP%EN!dD$0cdP$2pX#@
z%#+f1y;);bU?6Rg(r-ThM!smaaWgD-rXIB&6gI4ybOYqEA=?Z0-C*O-!8siP&Kk|U
zHrnxo;Ivc@NAOG}Mh_@K<UZ&jEF3Vp24NAnsH183a7l~zfUf3MajAqtT^_ZBu$qgG
zBU$J77?P&Y#aUUvPn6>EWpvkfQa71DYp;0WORDz~tcAVoLt;$+D5j~xt2oL}?VdG*
zkm0cc)>VyjpXnQEc~g(sNOUoD-NH35GVWLFAxSf>+8JW9a$7y0Y=Mi^frtf<!>2eo
zpVcM8#>krCiox{G5hvNN%Hkt;83r2ba~4XvDVL*DlmZ0F%a9oxV_a;XQq10EpLy)$
zOm2z>Kg#fEmVN<R{UXD-RG0AVgdQ<+LhVhgqbi}R!0s>&CS6_)6+7Rk_(~#e8`6|B
zH?HAY3`C!So296g$<n90W%tYK(4`uCCJfM&ws$2TTGG3_i*QyzhKA+wM!O`1;cnxe
zEn2rr47iavOoN@3Y2j+_i1}*N#0}~I@gwhO=Py0pibQ=_;<DQ@FF0Mtb)3+oKEpvC
zylObKvL!SLzw~5=q|Brrw!`>a!f(MWZb4`wUy@&M`&q^7GJwv4_X<!kg_PJ*Hn}Zs
z<P)L-H($kaJ>J5K2~cmXkgK=`?4%%C>~uFvHtpmJ|3o*6enT!mP$PybRybCIeqZe+
z{hAG)n#hJyBf+v|M(<|W-TZoUZNY4nq$4~y8%wp#?GaCG4t8)(!S@~u-Gx2-@}A`P
zp{sPfGE^FSd$8l=n(k<dEJ4CE?xD~m31Ky3zG1l!r*7dTc8T0dux49m_Nk%TmXsrk
zBs%4m9Xbf=f76%RKwtN{9VvB!c7Oq~w)8v6+ms(*z|&IzE!HQ!VGgOI*NqDfeMjce
z4{SYWWOc7@ukgkQi_9*C`e7Pgpf%LE6WtDt#z=RTb75ck4=UzX;q%SXWHr5evm`mU
z-()dt>gr5CSjk%P?}_7{T+pMp=&mC>lchxN!k8|J+89ccmaS!wXxJFQOVyP#dbbM@
zi1Xqr3+!K;ad9@|_!lXPbcKVZhYat22U=L0y#U82wr>oV1Q{1C6CQ8?hAx30;2*wy
zGwAWT1+dgGGa?K;I~ccKnPuE1>>9PsLFXp<%iZ%I8?gpGK&Np2&t!j~iNaD4xXKZ-
zIq%egSe#$3TN=1@jaP>l89+28<DtffqF8sSabQpI(52i)k4PSkUbmABW{E6$a*%`p
z&x*cT({JUuFUEKn94d|}BfbZ@CztM`=KQh;9vK(FO*{}K!NE@$Lw706LLP-2qc1PN
zgQpW@yF__J4?IIhs@IGP@qW8iAJ}wP{ZWsrsZ*<Hw}cRcS<<CcJu^Vc*vlz+78)1I
zz=Gnp6csvZtAqpDDI|q%D*ax6Kdn5O!)sF@gl;`2NA3!_8)cU>mziQ*^5%=bw$~HV
zN36JD!nKAB^QH`sA0{l+FpwU{8H5{uYBQ)v%<n+b$xl8XqaG7k4S(2@Dg)MAGmeu%
zIjoCZD+RX>1aV)g4_HNdOK0h_@2ah{kpGa&H+`_#uDz7}t@84dt9#|ka$=e!z3cMQ
zBscCo0bA?sFWka5TssQE97^hujzzF;-!Qn<-I+B&&)H&!JbUgm#&YZxUcK_|@7s2D
zoui}W%EYf-`5#}*BWxwkSl5UjPwJ7os_bFl0#W7kUXZ54ns`2v>mkn}4=zF}Q6oSX
zwavZWBFK{?UZ=s7SchWd#$IYepWKr;R7#_=o_D%j?&&#y`b|)&h_+RA?!uqAyxP!V
zXXe~1*^S;E>Uk~l(MVs|wcW$Xm}dDq_|ie>L0loI^XuO33NH@tFcoHz4`YtrctR!;
zbSWELLqche6Wj~XLG%uljcI9XjEkI`<59uf%B-Rp#2GbE>4-!94*QW5hJgtO(dg!;
zb=W^fh?evfc-~FX<ue3RDI<(FCubJ~kUr)ulbjhBS#9B8t@@gECva!TqwvbdQ2j8?
zeb(o~&4+it1#Wm|@+dwPwGxAu96!gbKlGeA-I5=bx7_cpqSP^d-*rOM_>|2+M(%E4
zWL7ek1<az#eJN_Ka|d0h-L9GvCmh7%{_!gRc+>bG0;S(AIAin!RO8+gdJwq1PkAgQ
z@WE&?)lNyf#$AjS@U7a^2_*AcaKWP)P&;TfHPE*#L81Th88zRrOY({Yk!QLHx=t+x
z`oee7i9)s3SsqpGM5q+3Po8D4?fvk29A$J0T@G-K*qsq4d#Es!p2jKMH2@!p8?N97
zRk)As2$31N?pJtMHTIgTppV+Rsownn_{XX0;Gomr%}0snA`YzSz-E={Mu|;CxMz-<
zo*h)~ho3R}joAfhVm50jaPt~T7K9p@Bj(LzKnbj+@d(_e_9q-}xxQz!_7vB$2%ahO
zlH8pRh+3?y&zc}kQ(``@&W|pYfs#<f!q?_3sBKQ3zJnePKU}=8Ujfr;m9+B61rU$#
z*Dm!)bkE5~6l?EW^)9^g6L9|*l<{u`x-dJOB#iCm`9CT4=xEhrwOn{5lz%j_2--}V
zKsn720iIPLd|4FmXz{Rxm5`<_MeD&yftsK;@*#R1w^Y|TmSC`ZD)Qo(E|C&=LSKP_
zYWx?c%TvoNX9SG&@_5zd?e2mC`>S7h4;EJ+)|3xzV@c1mFASP_8j-a7i^ZrD8sVC@
zm=P&C+clD0S~zUJ!{IXnJ&@f>pNmAf|AK>`XtAATfPqs#2{&3=7N$~;HfJpZt_}{J
z6`h48{7Q=$=^u(NQTqzUh9>h|%cpQ@ZE`ms40S|=M64<}JHxy++Jk4iK7e%&H~qAy
zg<?0`gS3)bTf}QtMFhEqXOgzspR3glhUlmg6Q}+<k<aVHRpiJ0E<q%;4gt}IV|E7o
zq%NkRG#NL8w$~=utR!z-BehKheTR%_?I6(p*&cr%s9UvRCzT+ufnE1-W#$C>GIzsi
z7{bcr@Zz~&W=g6@0Ow}k6~s#>;~HfUl%)4ADlG<7F>E4%4#6+Hn`waIN>)cuPQi^(
zanSUz^3&IQb02X6Ig`@2SA5}|cU~P<1kpvoUl@TwBhKN-(5^Wvz!)3~Y#gKxBUFi^
zD#Tm-ImqW<v7BewGB5E+i)Mo1`n_zur~sQrzFhw?&XgxZbI@QF0>9*vOXCWG_}V-i
z+T#~<Tf)hIFK?C_vb~xoU~z~rQ8l<G-5J=ka+g@l;c&01QrQc$83eF)alp##GIW~D
zx8r>Q-}Gi2%ld?R1<Wq4WiHf6fwcp@L|JZnpfQ_rV9CXN9~V#X)3#Xr=-u(bu|%>O
z_5F?;wY1P?-!3?X*Y|90MQV1!+RoShlmwv$V~Qy*k-uHwpiV|cGF*DaKk26e+B?(*
z5I&a?sJdEdIk5+IOD-QkFK{QMipHU%(YXkx%ivhV0kniFZpWNDT86vhgMsWG9$y3*
zczVNvq8sh87WK?han+{6Ypi))?a^CkN=^IKl@`ec7K!2BfW=RP+xyUC0glt(KVn9d
zdNx7sCb&L3D`1vDgQZMku@S9k$)x^dOt*E(F;hS25?YF|FD1&C$5Nk>#D8OS{Y$LP
zw5nom$_nZ&Wy!?_kCfZjChESzpuwcFH}tA}+AecdRJ4?+q0Z&97XufcP>V=$LO0#k
zeC1=lO?`sE_!(!eH(66n@El|h%Wz%%!gE%Le!9kdH$+M&KkTC!-aT#x2-Gy7Lk|ZE
z>2E3tJ(}3Uv&NN---J~*STmK_;(gdC@x5U|6)h;btZ9W~{G;BaC6)J5bG_%h&cpZD
zL2@Yyy;Qeh0n({si<3e_q(;*_H7@Te0InO3KDlH9zc+-v@bttNWA>@~4Oi5*b}W0c
zoP>#>u!84&ZZgNypI!ivfR$7{kxo0QZXO8itDVs~TYa>Ei%7$qI%AG{S|E5!i&K5K
zUG+ja9gp~O3q&g&U(v2Uki>upu0p<V49);sZMfU*iqftC{&Co+o)=-MLHc*1Td=O4
zcUmH8RRAl$#Nc3<+Sa*lv#fZ63rJ=|O|>^eIdF-TH`Nr}sYNYgY$F#W%g@x3x-)kc
zN;XbF{i5H@rRbaU2}pSFppvW1LU%oqM#}SU+TWg4s-*KU#FJ3I<@!^QbOgVHTo5@r
z04q8+=ZsEO)gPiK4*~z?V!#*GMkvp1u3qJO%=0d6wNU_P*av`-8*s@l)8binla_W{
z1+*>Ce(HxGW~T8@tu>OXc3$JfAl)LWlDf8Di~00f`STwig~<uIo&tWc@(P%A6#*{k
z9cOq5Z0W$x^-(IV`VO3fBJ@u)NX84OF*DM%<LT6*tiyJYrZK)fyrOoxXV)o9&e%Dy
zalc)IaZ6X$WP%>I7}ZZ>a<h=c#_%l{>eU-AmH4=+6TqU8#ZF3hU^D7=byHN_ZA+4>
zPXWiDVvvpI;@(NoD%K3TNVMB9fNj!v$kK-sQ4@$qVRMate7K*I<ahAlPPIKq*fZZ-
z<aFb*?9QKzFkSf4K4OxbtK2zGo>V~^Uyk85*<s*u`BdXVEyLX*o85My!QA|Fh{&(&
zEWn*+54<O{A6kGiWQnIDl(~_N_MmDLdkW>2Df=kK#|l_G`=!pGH>`#*&nT>XR&vDy
zwLmf{uH~H`-t0B>BYjAW=A;RR9YT92BYC~7*pa@21dUd+aZ_zw0R}riE=&}vi~LZ5
zW>w)Ex6|9HiV~*|tt<;Puh(#{f-x-rxDE8quHcPA-+kVSV&=scN^;*GI%8&DrB<z0
z-4~8Xk5gG@do>O)%l@U93DBVu#WlX)C3=9{v+kS$w&A84&SR0xj<U88p6V_p@=$bn
z=#^7;2YHWe0vPr%{Zv9|Ntc9;et}^v`P|+_A{HeQPxsc^IQo!K`mA41VJ+pL0r$HG
z_8O}x16&?27yHm#$uEi#FKFBAQQPS~BzZT^E1%)?f0>Mk$YW~(jJk7{nLGq%G|Vx|
z`0((S6<N7nz}0j*9h2(Ot%r=;dcPezqlI8RJo4am@lQqrIZE{mDGz4@|7^H%h&Ync
zc$?1MhiHi>R||VW#uLpzSqV0P>l%lG10x^Ng|XrC)XRk_=bWK{HN;X{xe~*1&S69O
zaKu^ArlU|cL_7KBjQyJ|!9W=6Uz+NEc>IIQO8r6cp%2KAo~iC7pny+Sp1(I#Gzh^*
zas*QiB(h%_i5=R(6J@O)4CheMetbei2`Q;^QPf4HfaT6>MZrF&2fza2ra+SVD+oi{
z%cxSrR75h|2#Pn$`3H8u7R$OM-#EfjvH?h^9+qo>>l1rz5a9^Vv)Z2Fa49X6bx(vB
z(TYiiGfMk=z`Gaj)=w!-geNu7Hx2q=EQUDOFZmBqzGZ1_Xfyc?BS1YZ#o!q=VKNC*
za~j-vrTOvhG9MR4)jDlU#5+9i@wgTTUAISqzt1~eA$!>rB)1~KvabO2REP4Wy7Thf
zIAG|&7FGlXy@w9hl1}Z{SxxFXL(A$pEgNx$QopI<@E5xr`}}^`2&W;D(y_MtDIMV=
z7?yGBh&10NJZrDyh<>IFaWth5wn*7Gm3u^QNm$PYxsC*a*t)DqLoCt@Eof)RixnRi
zHiu32XoO9&W>*x&_kin9hv?E*LHf<Vw0EN_z&%|7b*3N$=T;8)@F-U{uhw*8uGsP|
z@WXkG<M6TJ8pwv3qxW@HmFGODQIlXe-F07Zvui2iBZK$FDqwu7_k!;mz0e|a$iX24
zqRf@R6{>YZ@OfO2467FQRlPR1ewnX|p_Dwnq`Vs<MHAfs-(S~M#tN42Sqj+Qw{J#v
z66v8kp4b1Lw5=LQE|R)$o~h>N8P9Kg8AXU><x?TL#J%LZhB3K@Nei_?LR$!jdH7WH
z|Kgm{+uXp+Noxn6k>}Oa`J9o1V?Qv$(?*CC;7b?bQ}OzXakdu!sRFPO?FMTLE9_7*
z81qS)bO(gx?cwc`rl;s2Jmi^)BgpMi=6!a6-n8=+_A7e_gU!8ok96I&Z+0&*X_}@f
zgAq9Ve=^R7jS<FKaF5)`a8m`?KR0V?+V}ZknvyeO4>g~XAzD^2v10DA%&*!tXk5d@
zMKr6Cduum{X=@5r`Hbe1jS5>m3G6p&+k?5kyn8Kw1L7?Y@aDX=O*+b(^S*%DGtM6h
z&#Sm6?>y-{E*?5;+D}t<>rJUPmhv|6EuU=hX~#$=z33HO)FZb$#F@0hTXcXo{{V0M
zo9#W_wBwTFnS`8Ha2kmr6vPmuVsJ{k*lCJbn{XvnV~6r2BOC0JTm*>(A5l(B&|JL(
z#_(&3q)V6By+qzB*`)rYJ8&UBmHWx_b}c>SI^oa>4fV7KHov{#1VOd60$9DxEE_Cc
z2fOH29gdX3ewqyX>6DT96q$=1Ov}~{=Uv{9*;LQGNejmm<h2|*R2PdxbRc2G+7;=O
zmdx?@FvE8+5}sf$I(gCH@v+tj)c&jZZ*n7M9|&I#<;x5(!ysK|BtqPTV$Ps~t778@
zE?btDRzCq_Qi7Sc8tDrN^WPypqWhk3rk8*^Hkb`rEsthTzHFmf+1BQ&m{Xq5an?MT
zos%E3mI-+FWbg`@K1^{=&Fr0SJXpp*u#ylLaKaH5@#GLe$r`J}hUxS#4C=_}36<S%
z70NJU%Q<p7^mApT!c-&4+^PJ-hd|o86u{7)5uB&CSN!HUn@<qo8!YJBKQ3s1BjN;3
z{hCO>idTI@H>x1l<otI_thkRvX!QoMS-XPbG0v!Br(o2ouc?fEqbs3>o8VJd12Zrc
z7l_zmc@5R_0s3jm-w}cmjGzm=+2NUK_Osv4LJ=e956D_hiy{ARk60+&x{h7YE9kKJ
z^9{fHP?(ADmnF(I9g{=VWTfLIj=$x)xXUO864~UzW<z){g^t8pWmfAhM#Uq)J5cPe
zBdjp5o*t*4=oc3``XU`?AUcI{SO@yW?8`iNdGTyN5pwk99d#pC;GUpsmJ3doh+75!
z*nGMNl`7E>>`s#qU%9uh;gIX!1$DjW=|@7FR!1N4)d&9$gHv7Oh4yWY&B)>>>B$~G
zW=R&ND-d7(a=NF0595oWP~D__tBr&K#wbk-|Gb(t?Ru96U3>=$9*+GMzj%k>7fvIA
zoIn4eYM3gNcqhR%eWdRmV$zD7Z?o6`t(OtTk;0cJYBbRPj!gEy*$}_`9{-#EVmyFE
zTn-#gTK?T7`p4nLUvDPRAWP?p%>&H;+2{WJCS&(t1&jsZow$Bpng9M-|A3$VzxVTN
z3=Mv<&3v6%WKzg=xUbbfgPFb}ay*B{JWK4Ou!L{I&aUa-Zy|6QRlfp8nRlUNA^X?8
zkz%;?Y7{WpBQDQ@0pR2xaToJ~16mVlz?!9Cd>{`9+HM)8rBbuU$I?_KuQ9prq5g|w
zmil3d()Kxcd6l<#1?fNMXgvP-n664G%Qnyw_O*bVXv|b;;!w|vdpaGm^!9^5z_J|l
z{~c)ZHwYCP$N;_4Ainpf57RU``H&iI*lvQs5z(iXrSD*N_-+%%V0}9$de1`^dlop5
z2z2cyoTwq;T=LTO?3V(QA8Al`c!Q%cbT^IUzS)l+z0yyTRGYICb8Ne94`hbD<|*<z
zsc6*VOn)~p%6JL}EQ-$Az%bYnb-E>h3#bk@&|AmjXuI4X9;pp_XnKDeL$IL9+IV>1
z<B-0hO$)N=Aig~o`(uO3i>6U7erdPj)@enLmG?;eAM=0{P-50qA0d6M(-O2inVn>I
zFv>j(*8C!cK^wj^_3KMhZBXbUR0Kz7sL*(~!pwKpl_>|UC&AAjTWl?vDOnr8V@8C|
zuOuc+C0~6I)Yz_YYCCl0o_4t{<8=MT?i@{00j(kE20ITwckMNe+r3v#=EX-ZJzb@$
zWB0wm{yG_~d<54e-3PGVKpnUf#ULd3t?sTW%kEkkEak1Uw}E}8HW)wBgAmHDpS_jm
z_;Q}M<-}l*_@AECn(qL)tf(+ye_7|fivI^vS@=UKR4n2uHbQ-c^#d(#vV8ouOXr4-
zTv#g*aNYs$J0h%FJ^9u2VQm<R*gX9Na0P42%r5zmttsm}Kz|q$TMa)4Q{QSCh+jRQ
zJse#&m^K=5=5RJi%<52JMLn-?)Q8`5f1r&Cgzq{*7IE^|BQlIygl;>>;<@uf=FM6~
zvkdL4)PVS&=f9dZBq1hsOTqwlysfz4W3_|dEkr5{0f)})-gx%?GkjvRGNZ-Mo$c2j
z{2wnO4#U8wZZE^~&uHWW^s2AGEWV+@_GNUIdIa@2JfH7e?m}F|$pw?fn;OA4Ma+tF
ztVgRizg_X<D!PI{bgmjigH=lvTZs2v^cK3WTO!TqoP=y?FyEk3%M#gPEZGuKO~qKS
zKjDc-L+@D19Jq2#tOR;!uYuUesD4NpP{Q$*>hD1oCOsrfHm2gfCEK|BH}+Q_tG(|+
zv&<8!A;>h~*yxkeM=38<Lq8fpH?84i?-{#M(nv{VZ^_ZB<anc{Tit(^a;v`ZW0O){
zn@hJ}BG-irLD~U2R<!w;8E%Qe*HXTl6<fkz=TUsC-dhx0FKAD5uJ2a-__6(R#vtzO
z<GGZv#-^J--Lr(M)EvLNg{=MgVbz@zLu>i{1)f{;iWl8H$4*;`i+z&z>Yei{s6RzH
zx}*Hl#w#gRWmhk|(7?HXJquhb*E~9|#*Yr2<~rBn=~Ya@VbZbm^7+g<-p=^jGNt_+
zE-drwPxWt?7*|ev1+_ORwkn1H+Zp?hnD{?HhU!BArAdKF;fb``T1qZgVAwc-3vnvv
za;(BQ3$p#ZhaOKF@?)jJ6<~MUMI^8ejNf0f`}Wv41#Q9{qz&A&91)w=0o>^BLeEj{
z1t`(>qMZ7Q-JYbKnfmbLAukE=!GbD)1#v9l+y9lB*xNVP0ne30YTB=Z)zxaru38MG
zsKeZsdl0CF@^rCla!Pm3$@lERtHW6k35nDkS+LNZ#Ig+UJr3HMe3IANnk;Sb(C}WX
z8YRc)R(*Cj8-7s19wlAjfK4zlHejUI%L_DHPSq9u!4rnWm5z|@qZ2iNn7LX7&TD}?
z1U+LKI}B{c@eGENkYtBo^*VOL-440_=rNG6F^?lgHMO{*@HIM#=%f{>8oQ@JZD-v?
z(LwSZF<0hNIQt6i4QSMD;C5ANf9VGNc)+#4*r1+W!m`A;sW1<99F9aU#)WgWEZ;+S
z1LdxG+2)_U#3&llK<&3XP^7QsW7vN4y{Ph1*kpHeU44F4VF|X!FPQYaz)W@aIA}`$
z;-7)`GMaMuDG#oW^D=hY(foeT0+8A4!7|}~P3hGjQtvO&TT3NrpmiZ;wEoO}&7hpC
zuw%Py3>4rXGGW*FB|YRMg|UWDVie!_0@$0~MnAAUZ|U1>?*KMd1_|?uRHCaI)fj5l
z<r4gs56mB4&YE}@T<Gv=vRp2=oF<GaMN2LNGxq*e8De3%7({qZP(m$GyHL!1g7wjE
z3BuK~*?TAnlegaAAe6BmJKw-S5jf~S!4ZCK5tlf`Ap@fIPb*;qp*=$LoPJbuJkuLV
zcaWKDDOB=HmHCi=G+E^h0XPEV=adI+a#jB1vA2V&0ok}#&7TSUtkp53Xu758-lH#f
zfP+s4^(}5rq}^$Q)TCDM6|&_%6LyZQ?vbFu?IZ|*Xa>nPr-&Rv@w%~2HB5sKML&ZY
z<VcIgg&pzCcrEa$dy=AYX3PLBpo%B$P(@3sCGGG`u?~AP40CmZvYm9==vAa5a0N0o
zq6wNZN<Q#X_1ccNyqfapA+=4)T80KGw0L^n%gp>b3;}W5lL{v`=%7tG!0(?zj2n-C
z^cl!o$KNuN69z5-bjXHR$LEu7Md+Y|j111ug}%uPUV74bDzw{XMpXkWz6|MuRaMEi
zfu^!6Ft!{<#-JN3iarbjI5%$nuyZ>{ZX%IDn^V@`O-Y);>fu<-*U3izhC=S<+KL)>
z5^sW44Xxe$bYTZuYX~8Y7RXOzp}g<RPMo$!Ja4=o0~hm<M1dT)YXs#H!VYWP@XkUL
z{8RAfmxrhB{d01;qz=!AK$QM+2(N^GXk!J>!$27m&l$~X?$5OcDXBuaqqo51!%i~R
zXR`CH+!^V*b6^a2M`XKxGa?^|kg=Y3QfC+^ad^z8+N+iB>TR<L3i1@N3q8@#9aDEP
zqJ}y^V4Q{g6naONphr#c1sK*$^*lfKR=t}7TWuq*W*)H!K7l7!=<}GL1K8B1YQ6*>
zXphYqNR?Yhj}-{v&myGvBt<a!D_thDeLB)8mor4?!a7Hds))~TLV^;?9P%UWk9g)U
zx~L>>i8CvX;NooK@LyKpd6om3P(i5U36I*1U8H`~cakJ}s`FLU$f-lQm-$qTf`CtV
zb=D7kVh$vKS%8JD4>XqVlpEPP;Lys#D6|o^om=tk$O7C@k>k^hh&+h;PF~3f^}VSi
zcU;qG&}HQTNc5$c3waP&Rli8G8TngA4VVCPTakX_BWL(o+<?Br1Af3M%x?4fNk=V@
zR3?OgbUQL|zSvQV{J|RU7mZ!UiqN9U=xOPn(fVw!<$BH6NK@i?gG4dVE}T{>SdkoV
zMzth)(auuT*P;;K*aF_;g(DyKXK#am;q1%3@r-H-jC5{P<=3e*EjPa!g}jr~A|!b=
zeW#z+{A{cB#qp)|QBxj|JtT3A1veSrYFm7_w87Nv^zyZ@<E-x+?yf=!KSkZ$JfArh
zb?vAd8k5aP63>`-R@c5pT~m)~<z@YYxeDFK*G3~y$8%5)IfUFCl+=3*Oni1bRjUu?
zx!@;e&#Qgh=`5n0-xoR8g8*`M35EtY2X#)yT2tmwmdUvLyZRvUhJ{;<9l=h1n&<3a
z*56;&)ja65a=H;LeV@k=Y@VWAPZx7)VI2^b*NPS~@q$GA6?G0ejIsshw7r{1BS`?y
zGCk(aNz4a<4~(@5FO~6IjXwxh-iFwWEJh3_i;*s=jP7AB$DT&zFa=RgQU)I>d94a(
zDdtM50FL#2=_>BNL;guVv)wv9m4YHI2iMWhuO1x8IN#Aw;S8#TdnzHp7B3{3HNUuM
znLBH-C;4)XN~9>e0Tw$0oK-ty-1*=DuUD5(>>9Tf>y;o%+qqjRyYhM2ig>l3^Ewps
zig(G)2fqwBj;$3`B4b{|MP@4DS^r1f_P_r#o~mYQ$G<O_9ciS&oy<l`x|P6_*)l5<
z<WXT*LzAn=a1V<z_^Sz_?+=Y$#}_;wVSROaopY4_FSoI(89oMeAyhH96?#tpFT1f3
zMuI5!tx$8mzwO4<X$heXdUxd&JyYlkG+7ksIdm1M_{hCR7<={&-#luuKA4GrdQ9p%
z?&i9+gT7-@2mo*|M%G##=<93@RWG1yQsotD@8e%AJDj4beLTf&Mi3aFdy1>k7r6s@
z=T#rs7~-z=3C#~Nt-g|OKUl9`ftr8(WMV&<j<PXksJq^ZJN9Zc{>hLL_TI@k&aY#a
zODRq6TVKx~5&3u>T@(C5mM#EA{s8S4)m`zxYoyYsDF9-}q5=t}`H@-C$pY+8z?i@(
z-O^LYrmNKYl+6m8&}z#~d-fngpEmoP$qPqyNpVh-t+EIHBcIH=3kCzu!)T=JzIrQj
zLW5}T&1-LuSFHBwFVHQ`f7(%JO1Q&g+a@<Q%5iI^@MP>CAL#%n)$Kc?1%7k^si`Q{
zN8D*7@7MX${N9s0p{L2Iwms-lGPy5?OB&aiK2ALkJMV4E4pri}#C<QpaI8f1Mh|UA
zWDPkl1@5(ocmyU-XLUpdV`#iV1X7i6c1oAAGM8<Xq%ZlBLJay<)QL^aec3Y_f$vH@
zM?0rqbrXXrY#!pn#EbV+wKxj|`j{#Rtp<!IJ6``L@Hs118moQrJ-FI0h^*FI#=m*P
z#JR<f&ATaCu?#~|<G533yp3C(i>+}PRPAQM{ArFKq<Qep=6UBBT$pDz{CCqhOSlH}
z#2wBJs|K@8EV>>(HNnrB1a}lqq+c11-~WEXJ!Iof>=QiN(3v8A|1_7C)FeU}w=bZy
z9n`9;it#}b+a9bZi`V6NSNqn@${G?AY4oZm_#|ApHv6dF|9y*&e6g&a<sTdNM~B%4
zR#MseqJ)kvRN8yQu1F7r0^X-ndGo^VTCVYMe3D_a#4In@T>ePP>y7{Y#QcwUh7sfN
z6jqm?G)6TvLF3U5PZDg*X|T$i{6rhZfO`6o(pH>6aYUUMnP1h*kN==`ux`YZ<!uuv
zD;~*dTr;!02pJ}3cMIV(ElOWQ>){Ldo@#t=Lm^MdO^|NvludjUdH4Xq$h0y^fci6z
zG@e@118p#Nof_CGYMe=sB%h5Wm58u-Ptgp#fW?|o>_ER6N`l>oxS<|0GNDTuL_6=T
zD*zv_$E#w@$9p^XeV%^>=F3a@iOoAmv8WAGhRHKlwypW!9oD(AqIG$(Y_cNHS(lDo
zgoJu$<>etzMws58V3aPeB)8gEXn;Hkcm947A~FXT;+c!-f@knfe6cH!`lUU^#34-2
z?NB85rVmx!X&j}d@SrO__U-bQw|sUcKqTv;yYsQ)j<lm;&ttJa$!__L!w}?AttW8L
zkAtm2mzy??RMQ33O_Ro@DTXcsgAk2~<1l<Z8jrzi0o6xwJjpKlC7I_g{4S6_s1?}P
z6O@PAXw$A0{!N{d`-?g?p&^E8j4h(wlIQzt;U9=t@aZT8$EKTFay++A-*b&6vQXH3
z0Q=%+`%;c)8Iy+)E9AXC>o5k}>_Sts**xbti%+Gu!EP*m%_wF9f?+&yWCesSr%=Z>
zfE4AzG^~UPE(J%o$$5$0+PhdIu$+h!`3vP%J$$_Oxj|Ru;L$@!Dd4WYi#pEj9D{8h
zRGK}5z-pNs+r2=xW>Y5<d2%B7J@2%Y1~7Vr%aj&g3*BF=znu`dL}xRB#6G<`tm7CX
zJ5bvD;D+I$*Wmwg$j*d%`@qPs!Ji<V1>~wOXy}mrJqQIwu?cuIbnxK0>-5_1$V;I4
zuvc#4hw`AgV&ajO7C`OciBUb*1q*kcx{gmBKM&0WV|UJF2JPO?{QO7eDdZubPH`*{
zL)X!ZRY4JF+hSX_7NNHM<O^OC#5s%Qr(LbmP}>ljm$v|u`P|T$Av1dKJ|h+ijoRI5
zcH#7R4fJ;#_MvQPuIy_~8DJT9-b(EYn}ALA%OnZwD$9B@yJ<j@?1rTntVl4kYjh#2
z;SoZI@jKe(Oo%I7rZ&slkLFCUZ>hG2+AveH!ZzQhdJ7*SUIC_<7^w<B=<HtVt_3rL
z$;K9N_18Zcel{T{s@5|6=}ILcS1PgoNXCyVML}v3L5)<5(Rv4>@Z91hkS|t9zsYg&
zOPYe+;e*zwGeFBaJG>+!tEL&6ABh<1zDcy2)5*5~<O5HAy4ID1ofQ)H^scGyV0`x8
z<MFJ*K*rz9sDWvn37cuuMPnA8YYcfc?fSME01O!_UFTs3M!^gM=2v}d7G)xRp&#>>
z>0rJ0C-A5_nWKaDUIfyL&g2sY2c?BpY2DM|#(NgT%yJaKT^Z4X@ZY>T3U=BUCiO1;
zFdc%WG}SgjcSqzF8%y2Rj?<<<lruvm6d1?K)A(bLm=YsPDdwfSv;{F)^)cUqoj~&?
z@}?me5nxSGjt%W&2&)FK|7?~GP_8UL;Y?z7xVIro@0$l<;+GeL)Vn1$19}2B8#(?q
zv8v`C*Eb+e_oS+Y#|*zMa2JOTc%mKbl#brfDT@^WTPu!Rol;i+fR6tsl-{5Y{hPVu
zGtAHa?Ih{y@F3N;>C2-;P8kS2<2PXh1zEgk95IzBCJS;~nNZSZ)i+^pGSE`0pTn>x
z<Ud&8dp<!2_uSB#)lx$;L8SRxF-fg8sDYQfAuB%np$e93$pf!C1tPBZZ!(cIhZoKx
zAD22fc(Ov<27@=;(&uKa(%*#`5!vw*_^0TC2Y<7m(3HbR!`FB4*rJbOsb|znP)usC
zm)>&J_k_xwR+_`8b9*d}4`XF{IwIU7h$vF-U$#~b2sdq|cy0~4?8oYYF^6G@KLg&%
zK`s8$k-PV1iAhl&-MHkH%Z18_LO-EXS~N~CW8g*N0^=hYzz7|7oEzWn4Ryrnn$})p
z&L^27wVQR@JCQ_#UI5~z$;4M+ZlxS+RBEP@M6`fK{!s_c(Tuc&(;gBwCQSB4dx?SI
zl@FpEaPR7|4HfgTQFh^2ynf@y<#M`?MDxAC-g!C~9O)}r{2y$k+lGfYp(`y0qF-Mk
zBGicrfJUz<lEXKNS_`u-FP!F{ec!da)J6Jva8f0OPsXzgTSU(KXj&8b6N<(XlL31=
z{9KBljbs{)&Yv@1=y^e`T?3|4R+weZQJjaqL66`3dC(4@=?e~hhTb)zWDbV)<!}Q|
z610!rNF)RLv(Rjt_eAzmMYm@~eZ|cD-9HvnD)1z?(48p$InndyyTrBK{q)pBbspS_
z@DmDos=LFgfp_)f(GC?N-!LkkM$-blUP_*K1_(_r_KS#^N}NRts&WIJGp~1`GN2y-
z_Qyo>bOFL6?gTHlGC2>;FJ^=@;$>fW%K;x_EK1vt>IYmttwRZflQW=#ulPVe0m(UE
z-fHahfdPey(P7Y^S49ksfx%S)q#7q`DU<ICrV10n9arc2MCoHA?ds`&5b>dNB0XFb
zwT?<VS?=q$yPqb{wVeXh`GyO107y-AhPl>28ZQQS(zpmeh2Q&JT>v3o*4W>)ruqd=
z7S%VPmd>RdCu;BXq-vzU5A%~=ME!Tt5+rg_>&b?<<=h^-rJoK&tCyDGs}Lk?f=k_=
z_v3*Sx|E?J5SMV!#89bQB71%xzvczQO9<>GNR@l{FFPnRwOdngvxp6m*WLz?sGks@
zo(I$Z<A$he(7UgbYv2?stn2E!@~o{V<dR}Il`PgY^uHRQvxb9#m?SnGdRSG`W5F4o
z&&f4<c1XXRc*-{ApB8%<44$V?nQxpwub+ZdaEc=rJ3*0VR?9HO<AXJ(*bpTTHK&9C
zK|@<B+hE(20hmLGY^o;%qom1?9iXm<qtQVO0VR5%r_s&1sh_gob$Vh}!DN}OX`hH*
zH}jlS&XB|j#gV|cL$qj?`uF<>?jQ-E(K%9yND9C`iKHh<5^7XwrBaT@%mMNJbKIN}
zO-h|jiqt8lM;WHo`p?fb1&1n(=YEp<vNho*(X80oPuuZFB>x+NPi+?1xBRFwuQ8Jn
zImy*X8q8<O=G6X0PO8C^hhBa{=9xf42fUU%3tU9ls~zdexVwEn{H)5^@H(>!r~qm;
zvtQ&Sbyj>fo<KKd!fY#oe~AC`T0)_gYqs|N+h0yrC)~A>2s3J*Gm+L|hsC>_5L!}v
z!oO%qN>EpcHGnhl(#7TJY#yhr$+C)<SC~chBqP40!@8La+FhKw8)p^vs7XTWIT^1O
z&Q2^gw0sc6#+vc~(B4ksE1@&lM}C$?F;Ftg;I_=$Qc}gsyI4m=xAkSdD13!kKv#GP
zmF3V=@jFE}^%Mbs=eShY(2<(OMBT}F1rT#0b*W_l`q-fjkA}OxROzgKc}Dk@#rv%w
z3^yteR}mv>Q&lG|fU`M$#<0GgnT4Bf;N{Rv%Mw*nC%PXJUOxEjh-aNtt)P<dK#9_H
zTs}vW^71krSBeRNEz%(3ASbZfYDxhr)y*kr{KQG648@z7#Qo$(vN7)BhQ1okE6<i1
zrYztvG<y`|62q|E0aQK(-I%Lo;%b{Q&xm4^2I%mzO46-)J(I3d7l~#o1|hv}F27x(
zt1Ta54-RI4+9sZJAda|AATIP=DL7GU$2X9XouYNGZ5HLZucFD;%I374)${Y8EV?W1
zI4(?d_`SRli(-!PkMv4b7A}Nr(+6eQ%q9ai*>3ZHatqSoRsl}Y$$7TA|Kt=M;7g~Q
zDZpQoMO<ky@t6K46(#QDQ(6C;RFrlJGiR@@iPLV_hxk}Y9Mi@=V4df5&^%_Z$#38H
z!rfhS*Et9napN12*e@{<(F#$zNsp$KYEMvOrda2_&Ngf$EFUJ<rQpKsXQA@>vOncQ
z>_k+4duxi4;y!C~v^F{QUoKWPQvdE^H3c#OgjCdUMVp-chmBR$+yAh!ng(a(Ho}eu
z9UY_hscLXf!W8(__w&{94VHQKgnI6gFszz+=nzdX+PECpv(<Ujh;@EX=1a+knmOrq
z61MaUFofMZe=u{D--^~J_CYc8h5c#5S8=p5Pfp|=-#82-R7v+jgD;7Mqdr*4IV~Bd
zSjE|cd?U&;aW6RXne$?!#OB4EG>sp34+QYmNsTy}EY#9`yw<Qcu|1%Y@D&_TfIW6d
zoR`$h*PMZu;)8;y)h5073G#>94Ys;Q+JkQfAzrhWFrmYjQV~O_-Q~(al7^W}qB?b5
zWc+LTwMMdC`EmmMUfM1N@?w@HZwN1nG!>@fduSm48+)A%QzX#5dNs0qZMJ;wHk`Sv
z^Kwp_yPO*AGG_0Oj2TP_Eyt$vTS>P8P5;3Z-PD^ql8rqZn$weQhKD+Q?v-n&Ps^^I
z`MMqvq>g>bHviF4^VQ@tvN)4z!jw|<Qa?yPKFt+0m499~#cY)2<iz48MwCQIFYu8X
zpYHFeZ_smT^rIZ-9hgt19|+;qmf@{xKi2wrowlV%*!_3hj=##c25s2YdW@g({4BIX
zEvW#t@N$b=+P+^Y9a`<&{{6dRCjE7rhEyHl-WvijL9b2m1v2}1Th8jn$lN-st6sm$
z?kuY_0Ta#?Af?-tAgr%`;8m53baBV`t&1IV-rr#s(7rVg&{4V)FgLU0kFA<jxWAxq
ze{9V^M0uk1?uj{hs%tgh)OLiaEQyPBTY6vJ&7v;26k8m)tIANDs-WOvebL*GxY4WW
zHXbxgn`UHT2Wzcm7q15W^K|Xs)F+P+-qDAN3a?{7wru00P-8sF{%Y%}o9c`@*K>5?
z$k$4D=U0wzuRS(H{IP6EfjP>wyTD~93{p<}gKKGPxEl}WHd}7xrT{b=Dt5(xn!oa2
zGk`1V%K5M9^doK(yqzXa2k!_2oeacZz~V5W4Q6%6@`YTZ$#qhhsJJU+>PFY`#9lzv
zmmY3DB_V|Y`n?DfI$vaZb6tYpE8o?QgYT?_`WIfy6*!{L^WcKMT*O}ymrHQByBa?8
z$NSvL3DjC=_x2vUA2<3D4Jm50r_|dKN#}_Y<o>phc93WLj+EWyBA{D{H#V#NUGaA^
zlc4lEunYRg;&wfQ3m#UuTn$8sS2dwzQ!X{^yMH3u|Je=vt6fn2fY3yov%)xjq{oDg
z#nr%mVh=c;0-3dUo2)&|2^D@BWQ96&6G2`03gVJ@(Rq<a^Q<Z#0C{!THRTVMlL#e6
z1a7e2PCn5;1l9c8BKeP16#5OZP^$<$sraW~^<RGN1EdK$Ci#s1=i~I>KIy-_FDwDK
zsktTb`JaF5KYoM%7cYQn2E9i+)i%&hY(uX0C6iz6ZaGNLJ@nL`TeVz-?AY;S>QIPx
zdOdm4CW6s4JFh?_JJC~yNu~JrTys@Y6Id%(((AB4Z&m1Rz8xmDNG{O(IcV`FKAU-K
z_l=t8p9yWfBX3l8;VUmlDq=Y<mV5%|fr7{6VG*%>zLzRgNipQ-PGF4b&jx`GK~Qqp
zOf?5^4e1bz7uCaC@9AV-FH8W-Q~p+Ez{>Imv(L;OCNV-u&YtYG+Y3p5=!zz<0)1^s
zE`y!!$9j|@CBeN8Q%waS=G8ar-){$w-8=!VR6O4xM%?1^%Eu)bSRDb!JJBu)f&ju{
zI-=_oX9reV2V8q3AxK5<Gn}(=Gw+b+53JsOIlL8s;N%3>=B{f;tc5SI-<)Ocf8s`3
z^yI=sUdZ?DXXoCgw{CYsd{QYCv+vy=P`l^_3^t;{RKljCP?#q`0kexs*)q%M{&|B3
z^XzUpo(VTh1f~qv4NYEGJ=_2D(*EOW4{b)==)7`z>3{sV&i$cq3ZI0yv3(F)=3FEt
zZdG9pItc7ZU=}Rd4A(^<MqO*bcF2U0JL6a81`N6jSo}s%_E?+Vaf%xSWuRp3AOXYo
z4QfT|AVe5q5BK6g$adgY_{q9<!c7Jdu^H@ch3)GCs^v%g&clc`=t(}5lFR*R8kD7z
zkDt8z^&@$XMPA~tVh~L}UpmGF$&b%ulN-H@3K~sx9mbqRulhxMEnQvu{mv^R4$m`N
zqROeCxg=%WJ3{*bTroIvVn10KXh#e#nS)jhdrEES00h(Mite&|=Lz|TJDDxd4A9X)
z{vntnyaO|!#|)c{im}cG8p|1FRBv+M-Qz-S=dYZGw)E?9OS|U+pI%5=5(KMPM`--P
z(VNX}ap^#;Hd#{suvf)V{%LymA79`f?*s|`BYHS0&0cHf@BMjHwva=fpePtZoSfYm
zLcH7w1K-DBOzCpsgUG6=FFOJa8~gu7Rq|(3FKzCyI)4Q|@nif4Z9|2Dea?2Drv-ky
zhpKA8<L5#^=lwLQyNOXH$m7@Q=cq+<xS{=shfurPcF&LSD774BwWeJY%`Hj4d6lgl
z3RC`Vx+UIW&dx#T+FxyiZG8bcN)!;4<VDhq9!cB{td5`S|F%^l(N`0+`u}nE9ne_*
z{r@F<-eqQQLKz{Oj6{+>BC|rs%HDg06e3$>7Lrjm*|TiQ%m`)g^?zUA^*q1le7`;C
z`JdD2aG&D7?(1`XKJWMI{TecoDzADq@>J@hI?7iL<_^dQB!pfbj@i^&8{2|CBdzea
zCo=Y7Ce%i%N*wp7Yb5KPaA(QM4);FrNHE*BwlEIg|7<$pRFo3be6|X!z%Ejf`AfK1
zC$CjcRNFk(7!K{rq?5+#>+v^wPPY)SOb~f{KVe$j_Utrcs_ChTG&$&1p`-c`=_HPI
zxF6u<!*GnNpgOZUgNfDezhTev?(DMU=+(sdXB8w>Yy_2x;Vk=sk~YK<*4lx%$GwIV
zHnovrrjl32S=YO6*w$n$w-3?WyHz7@7|gxYb7WIg*eY{}V8r~wfl~#|gcd{3<4GM4
z*VJIQPiB%e>||aAQxl32OhQhZ#GW5L_G?U)T3g~*=QsRU1d6EiH@+NFyAocll)jUq
z?$SBs@MZpBdg<DUt+zy6{qbbYB<)P$6uILh<*~=Q=cN|WKpX$s{}`|Q$G4?O-7}gZ
zxeb(3n0FfqGp~YiTpJ8>e>85hCR#BjT9+j}rW9xwTZTs13E7ZFz|rb1SYPo05{lRL
zp-XVA73d2rdRE19gvB;Bd)FYNkoA_>T0j+`xa+VJv?5kczP6@FoSaG#P#JpmiXwRY
zy>}bwmKzTMpkjqSGS)Owm^It$F5(sSc0fySR=-h5Vt4XP#{|e(W?Pw}V!rofcYJGm
zHVeGy5*0AVuSmB7fW#i00r1*M_M>Y^($OSpxDBd?8&ta?(j`)_^Z^Udt+JMD4u0|(
z)t!PQSUJLQB;I9zA1Dj$Fc-^&;KjF}u2IcCE(hilAHbb{Ek3x4=UYIjX|zGR44Qdr
zqa$1p=eXBT06pQMpe+z^Isls^N`RUxl(uZ#&2q$X=nlx|m7`;|ZeIwN(@4&mi{&F+
z1)uf#*KIIEP<U9gcRvAKwUU|o7obMs0&Bd`K-j_8TC61w@{P6tsNz+%swm8Q6J;ch
z&FR@f<2?N05Lgx$z~SaC5}gR6pDS?rQER_NUX&M<4&EMojxam(eC}lQT>1LwSy~Uu
z*8?_AKgUpku=m48{seE)nGzD!MSSXhvm1+pc|i=|jyENacTII6zVF^l#Fa7u!e~o?
zsZI4{X>^$4i3{uk^7UcY4y<pc4J=)Q5T7P5#0;io@Y$xc>6j)ZSA81g4|pgISA{VP
za}fN7zKU;o8ZPSHydpvnD<cwFNUL(cpSv^z28QQ`-`=|O;ZZd3<VI5Odm+{Nd!Tdj
zF{%#l!mQ&WB&R9AQrb<6@uXU!On<^lo4Q&>CGPfUeC;LI_H%Lrm->e?3r-;|s~*>A
zQwChNjS4A59_18(5kAJsSS+P{FRh|yA(Gem#ck;|Z8Dr`UzP!(Nx?DlEe+mDOLiJn
z@Y+b67(T{WGt}P)N)#vJ<-O(!#$eW6QzMbiexTj?G+sMBu27UcQ{B#1MLsF~xcUGl
z0MU{sdp+Ap22ki)pNLI|F!8y&2v1z_@ZSA(fub^lzIn2F|1XI~4g(q<8J8_bq#XyP
z{o>c=*KgZ7lmoGT0>tbL7*5#?7@PHcmp9M$!^vi(&Yor{hxy_M@J3@a%iRAomg`fW
zYuf@HjRE02N3TQT!w<Q$C(CA%4jez*-mmy#(b{rs+9By45;@N)MVnoBR{*@~es3u0
zCQ53gtNSkMIDesi|Fs|bP`);Iw%^5Uy4J6hmIcOp9CvNu0=?}Q!`gSjgbiEUN?DFq
z^|`}(|I5fqz($k9hL>+)=t2NxqyvWuKbg^MX~iA**>hhiAxnN(NMqT=I`2A9@acG?
zahQAbB^_+sBM~L+&yo%(*w#EWw%RD2Da@^?Z(fB*w*mqJUqs}#_E-$x-vMUSs1<Hm
zDP+}Ld%L=oxu(}Hq|rMKAr#_NGKqrL1t%PXk2r!HIhNs5bKqDm-a13&pWNuAsX;t4
zt2Pu|xD4KO_C+kUwDoJiGj}w?O;lsZuLL#F>spS7jvSMYfR(1ye$Cx&5eM{^a0#^}
z8_4l+vd*_mx!9n%Y|VMhx?FfmI`%#H4}TySPK>J7=d1h1si$X%!h%HH@#eMktOc;B
z)Qe*pMZ#{}h~u*3h~p*WSa9I@_ST(;b=fqYu0WU6yje3^(d)OzB7fjb(YCVW(I)1+
z#qr@@!hC#Kc2ZciwYX7K(~*udhq*wa%Gt`-{yB=z3Zf93GUF(pE@N-VElp0wt+Ici
z+Ec15XHtf#s@3E`FbYRQ8SZB*oAsVsVmr8=LCJuzCU_F{s4kQ5Hj2=I2b(Bl5=Gbm
zXqJpi;m}O+Ts7A&a}ClA8;L|o@kFX5eG=V9OI@34MqcY($!>T!1csa0C%bhzlSgt*
z%f)4Z0zqb2+YdYUU?;!t&b@tS+0~`nbfvjcNn`(qjd!9ep{e&ru}K>IgDc&|TC0TU
zqAl-n`7aN|Y$P-GVH5p6wEL&OAG|nq94q=S1`NTu-GM@{d(te}39H7G-$9u6Rt!6y
zUc;XPO+ox<1Zx#I=A*8wlAzIgv^da(gs|3IiE0%Ju~Pcz&>Q-p@cly6h(DU{N<~GD
z?^3|7nrfyA0hLD`xjR(E-Z{Z#ofU6!1C_%_IUwJ1QYHp#N{$C|7a<|E9I9W@t<1PS
zNkgsb@+SN3dtpZxov2l}^0Cr0U^MF^9g}5H^fnJ2Z%g7XnraAV#xm<t5XmiEnqW7!
zcdkAvGUCU&M7=iH0!C)2<~(xVj$oZL4wPoK3WHdSK{+T!wXLEArdI#!M1mNV^0gTs
z?NMWSLz2X|^VqtnUqBURw`lii(9`9OJuX|PN|a{wwBZ)^14iS=SE~b>%h2la5P~49
zSplj+N-Sh)mh3Q4beO^^HsGGXm}a`4q$Jkx5br$7W50B<aO(Y>wWy+bguMb7S~-ew
z5BR^b8BF0uw$#Wccs;ZsYT)ToqBK(B&%gb{Wa$)Mkt`jORe|Me%>#Z`)tH@lH=DK@
z^2SXTug_FBhY!}HO}AZ8fA#GVo;%COC=?d3e}jdGcu$V$ZXn($!Y92_)x4u#7=w$k
z#r4kT#BM;4s5vpM^ulXe#$>msQm3^V{~g*bu@`+&P^*69Z@~X`+wcN$O1-H}B?>m@
zzo9YRQ*gjDHtK2c{FcK=$#MA^!4$fL7nXSI-An8>UuOUG0tg}mVq)qo%I3i;%>9j=
ziFYvi@F06mALM6*N7o_n&RoA($%~iqj0-`)0@k!o(J+sCI<1t@sd{uVqH9_99!cEy
zsZ~9t4buHsW426i9;-KUGI8=*#LE)L)s<*usin`X`boNA;;|BhS}xIZJ@(HQap~h=
zCM}VmFY=9FfbJyZ^=%ou+vf{<*y~e+iUp0sf@UATJ|l=%&-js(lJ82H{tI$3tXgez
zb}Q3-V-C|q#Cx+#kF4t9{()cMd{?3w+Vv0$&IHP_H(~d5yPPz`KvA2EL^4LJOF(Z_
zlF!ad^6>fx34Z*70ZB0IP&z<4OeK(!J#QSdf>@i&v+FdO#Wv)6!Q_Y5kn^V7b7CYp
zl|_g)Ev;Vc8Io5XFY?)?aIJ?2d)HXfPNDJTm>0>``02ur2A$;2l45xe;_q&dCE9p5
zAGv5ZmYgziD438|hjJg+I&t=&p5otM&r5_(c0gQmndXnK;N!WvV7mZ%e6|2|0=&=k
z?4iVmQ<P%Yu5Mx)(@QU1RVjb;U9H2PjWPhU^Z94gA;YqPL@~{1aN23QuBq7EbdATY
zr;Ly4*x2_&^665Dr;aWuCTT`SZ8SY*deak#`@*)C{Dy2OeZ!W_vQJS+tYTnfBi=FM
zC0Fo<;Um-Bd0HlaoOJOSbZ>Y78=gv^3fzk2zM4kGvg<!2>T-b^ViWvZay`6JO6ifc
z5Z<`=Lh%uSVAPZB)rB2k=<8fJ3!n?i=D;TSBvB>VoCTAHk>)<gFWe}?R^{8K>BqU2
zBiTNtrSEAmdnM=&i}4@#YMbtP(Qy&xrRvSkNsJZ+L?LBGsaQ?*h@adDUy6bYA!9mS
z5pjSkH*?k>cL|wI7T@`3fR+;Y%H#m0X~XLX`}0hyF?Q~linKdQ(bUTy=cM9PLz}Hv
z5q{)aO@Q-L@JkBhzLzYsigf^Hmu7F*sYEWch|0-WuZ#S;C9_75I>EWn>OYiKzSL+I
z9uzGtRxWe<Ust@$#(&V;`e2ElVDXYmsA>O3nF?eMkdoW23{r!~CPabfK6V257qwy<
z_kp{PLs;C*j0f*b>-XN1dn6p@(stZ&6p~&Z`i==6E_Hsx>1qbExHR&&{c1CpOgk=?
zzbkrj-iQ|qs55UzG<0NfPjgtk26vaaPsElIKZ&N;rlKD#^rrK^X6^{0Cn_rj^2G-b
zHpQE_t`M)@4wMjN-jyMmJbP|R&yM@2e+z#Ae(>Z25CE}0P#Wx;UU~EG4$Gg<jPdsk
zmSM%1N$%HMr|pJdSs34`docl}UssL)w0&)P;|+6$3lW}`!y;9a+^M!SCe5?p{&SR<
zz6n8b<J%Xp<30UOogY7aw!QobdQY7^t?S+K46q%eyG4>%_4Q_wQ1P&y?(cmNv_gOT
zC#@TYzkYXfhWcOUTg2I3qRzLVZ-9>AnpRIGAWjHv9l^d6Jj8x#;xkiz)&}}YOq=!!
z|6737+d<01!SUe<_`RxV6fx`<>-D=oU1+)gDQa#VdN19C&WLpX8qb5UpeYi9RPqmM
z!wAyz+8ZpYq{pdvB0*{24naH&2naI%wj#3l!LAC`t(pcffP998SM!V6*O)lntUle&
z?Y(ycX*`x|PmXnG$uH@%gV_klkq|e4THPPC78Q7BLM&yPFf&qezqNA~KIVMd<_%(n
zOmO_knDV#>{a(j6c^Y#a!+dRV>2~M_&uKCbR7tm_n<%N?dY(ahd>2Zs?#-*!lcBhC
zvf_uKZjSBW3_-ni{}y8Z%}Ru&Fccg%g6<7X6P&<P#KRZnbTPck2oJXq7jf{HGquuu
zv{bTjoQI`|MA5IuUD(bDOOvg>*9NjZ^rPOQw6`z=r;;IU89w$Y^@ez`aJOB{BF|+u
z2tSH3r%?K|n%jK-AlU8TnL)K%XplY`DINJ(*V7K|&u=~(!Sh=RV9nULGgQ!I{B$}|
z;7jw=!(Gzqw53MicU{L<r4ky$EF6dDws<&)8$}%#nOhsZeCN9Wpg_0IWz%!-&Yk4T
zr&DFKSRQK{OqVvR-!R1R-sXy09D)w=A(!SON=IR~!eNPT8foX7`lFlOBfNvAcGCO_
z!xq(8Z^^~l!?-ovjf|AQ(Q2<qo^-7Zh@G>bfN7115LKzqsWyY~Q8-@)B6c2sdMV!$
z7y3aZ(m5>0DPI0p3Tb8)e{}lYe}ypSh@-7;#xiGei=Ah<OVe;(8Rqw7r^M_v8yXGa
zim9df=WeX#Wf$rZrt4BEHo3@43rV;*s7;y54BQqnjv)vRY>Jp&ZC39Tniet4!Mglq
zMf@=LI@6mEN)pnneoe5@lH0;XeX0Y+hnr#}I=CNETe$4rF!iKAEWu4%589Nw{t2cJ
z33r0`cJ01EuWEOl<EbPhUoNGFt(nYIj`J1^&Peb3u}Pn6VIe*J0Z`BJWgJ?0&IAq%
zZHuCU`;~c_!L{!Ou~NK?z9JL71~b!JiC1h={IS$sx@3nRkbZh0uUIf7twwp^RP5N>
zIB{~nS+a5R2g!l5euJ0$<|lbk(qJ>vdg?~J?~rR69ZTovCjv4S8Z7SDd*{RtXJ0m2
z{s->yAKzx7jIK|p<*d>EQN-k+p($W^gPU4H==t?f%T?Zkf|72I>#NzcwTiF&Y->>Z
zNYGI9G`q(;`FO-Av!~&#G<ClU$2G3>@uSCqZc?3`C@J=C$OHJ=@Z?(Q3Whf{RW|*z
zPuIiF%j=){<E%4E+WUBMQTS6X8yBAU`Jp#G^5TmTpCJ|8flN?16}xeQcuQ|Vs0wvm
z{P3RUnRNDLF0R3<)doDIhl|z}jMzGzPKz#s*kv-`uvN8j8f4q#<l1c%{#X_t?qMQ4
zv9@|7A^Cj;B0*`;-|qz^qn1qY?fhyX32->qOloI9+i><rq#OI?(>KSZFBleN-pQ(e
z`{GC5Q$M&H!EW39=@+)W18y|aJ$1WOx|MG*qC_3lTO0W5w-PgLa>cs@jx^MdDP2@4
zHzQb!TWm|5<~0+_b=E7N3Zj}L(2klPhMWDc?6Pq~fGw%+Jbb1WHR^XA|66#w0OEi7
zB3Z=gcLI`;E2|Gtm%%DI*{yoLE*N2z1X!}qGzryI1r?PL4N%Sr%=-(qOBrP{5Onns
z$jY~F>;#RZ0)_Ws>3~*=d&50j!t|t|rN<=V>6fsN5uKj5S*mB5>%PoE1yL=^+(Jlh
z=$(Z+;ua8)Sne<(d4jiw*?b80;dl3c76x)kiGuCg%Lvkp(4t;~opd^s*S94kbCHm=
ziNrd?0Gqs?@zS2e{)9=hugH;Sy&?9CI)V7$8C&Ub*yD}9K(8~2Zsmt_uMSOGr86OD
zi@+dKtOkG(B_aPkb1-u2>urmk^-GwlnXc-!bD8RR86`nk@L5bIg9x#L(yAdpOeaYF
ztV4&2l`6<)c$*wFT=pL8)?qy%y;MLiiuzL_yD_fVMUzXwlFDe^3OsIj1K7|U|D&S5
zqD(LB+@s%|m})-F(=$=Ov<}(huPTkWD)at+>{{|FV-n&hXXcrXzvvh8=w*ap7-|3h
zho^@b$*YDeKR5{D2CTH3u#b6&C{w;#T;I8#xaL$m9mY9Y6SY1SGq^wesPE=$I-Ow3
zZNWbLDw~JJO=PSt;-HDSyf@`vbd$8w=4B1D<D`KLi|mO&j5YJEnD+n*TG!UwG34re
zGJa~8c1vNxvbSc!vS`J`00u(eld2JSjVs2y+f**L_-tZWYXw73YMgB1$p`7jZ@|o8
z#cM6xP$=YDBhJpQB=#|IU7iZ>>`l2fLW!k(_RX{ZV$r#E6_(b1ynLkJAAlz#EG>X&
zcznA6g0MEe>yIXYeM~5^YjhCuLy~odT>2E`G2UIr8Gi?=%q5`QJQ%N}U+|Us8Axtm
z3n6iih^{aF%tsHD>Q(mHrA^04XUO<qRixxMf8q{mAEn;8Kx-HUlsTOpeFVM{^<#W3
zVCD)^NY)(xiZdyWk>cArSiViF@-(x9w=FLQdwB3fDWY)epAqdZoY-<VKQ7)CJFI~{
zCc1&+%LGVDhrqB1Zw$}oz85|!CtVVCDos+gfoq}=$nV3i0URS7pY7!~#1d;p#L*97
z6GIT9)us}!J_@W|td?-o$GJf95#EESjeD;=1MXs+!8Wh^0pP)vxm7Q8-ZQdcyLweF
z#i5VtrU9C4`;xkn2I3_xw#6noY4$5IiZoas%?wtNJQFjV$*SO5M3{|kUqggCgndT(
z8Wu(*6v>ZGfw#yL-37;iS^CvtLQsincgS3Ci)NCie9m6k;MV9C#EzQI0kfdXmYxjw
zTfo(P&_EXR9_&O9i_{LPL6X?DeYVUV21w`ie1@9G`opI%sjQ4FNgPB0g)U(Vk@m-7
zWY&j#$pM-<4BFGhf`|JqFj?#LXIp*%%-niCb@7K*m5Tl>J{tLVNI_09?tg?Sts3&C
z=+@PDluE<?(L|L9cEyw`Y4JJ=kx(uZPv3TVuw?=C_c)@lWRT4Fbno9ZolsW8qL_R?
zw<GRC5*R|TC$YzetsX9bxjiMY*n#OCPI?$C63IVdDTakzGv!LL!VB-cFm#;v7($&#
z0P6}n;=jc3QOJsDXL`hEo^G3S_=4EaDX*%_bq>}-#$q{RlHtWG_05OxPa@pua0Z`$
zOJ>v|iuyITZ801fG76;rG0LyAmpV26VD&T=n$QWBjevwT?0j)iwoa?N{sB=($0?43
zx@^Q+8vveMDYV$lAyB0gyZHLUVxE&E64i(<Rzu^$<e$co1WAoR35)fQKgDpCVOqxZ
z3mz|N1A{adNo>_5(a@x>8BF~N8=7E$d9jlO=4y2x@vokiDWY|`xbA7SKu%46DNL~y
zD5XDMzlqg07`d7j81*BEc;{*ohOvGPzvuoVQ=C`3trNhKCVpqcm2)P=i=Hu5QK99u
zFMbuT;Kw?{6%Z=DqQY)kv_>Cd;0RpBbqJR1ydb@6qR;M(cw^xx@{!6>Z>~&C{Pd5<
z+-dC4tBJWwjndd(n<5wdK`cvod>gFSJjyC}r^7Grt#&V{-(EV?@B+qA4OK_=tn^0W
z(41d9yg1kmLkm(++#1sJ)$1GGe0lO?Aip1xs5a;%nMxEHe7dG`xYQE8Nxno#EBJ|U
z@X8>o?Co)~$yhuzrXI4|DU~UJ)ZcC4uQ4wAVv~;dO`f)Uc2ltU!(=pG!yEP~)6cw<
z#}A6~`E}gL2F#Jk5IrryC|)E39}%ky3qf3{`*CKm1pPI6z12-hB~<BtwL3_a>u)5?
zWF81|D5_x%j)U1nOcQPo_sOCB(L&{fq*&i7rQCa44y{$~MWN3gg=-LPzPVb1wvvOy
z=899Td7X)ha4E!bcO!_RS;C_YtS4}%*}fA}UhM44(4DQ6aIU7SJKLhdXxSmf@e6l5
z;-D1mxZBCRLQ+F`{OgfuKb<ct#Zxz~e<e`^vb~tDq)}xeK8SjY^fh5#F=ZR)FW(Pw
zJ34n%$HfuT@mNt(Dcp^lnDk~y7`WfE=Nr_2gPMbM>jTT;fRY9?3&H_5&$;0Wvk2KH
zLoI{qkydEyqTgR}@zKU5T?%y@IJ+fr??{y)YVwxk)9|1{JhC2<YwjEc&mb`B?(7nl
z^CzNdT~}2$j@>8F&Y_<?VOX@+qFr2qVp_{!RTe`%!|$}~_pIlSL&L-iRyZP`S(lBl
z>hN#t1fpjcbjo62WiV*TiIY)ck#>gSCdoFPkKTF3OW66+jUcQ*GAWl)Vu@;<7fEW?
z(XhJLWTeg!S@d3e%};ErUp)hSo>Tfe(nqdW)KoYSc(N}FtITyLTh7plxfIns+{jXE
zIla9&z^h*B7DqxF0}$tnFbP@li?=0Q(#4s#j_*3zu<fTpomEKW!RtgSxzDAVo|1E_
zX^uWV^}?z*mQD3m9HQXnGwY_-eP<qRRaB2zW-t&QyZo+C8BOn^nV%={bVrtN1cu)~
zqcg_wptLmG;E>N!C9oue@ngYHAKnTC63n3GkrOux#f?UH$V8J?1?r`6+XvpJXAg?=
zGr1bC=QX|Ic|Oc3@XnzkTz4Lt^EWwlBAi4{_@b){)lCW22cF^+8(2Hcx^#<88j!+q
zevGJ?8<W_Iv?l{<8=m-NVddYp$6YZ{d1s`|^LaLG(Dl5Zl*k`<l56r!;8V5acUU_i
zEV9X;H>lUop}*oGJTKNJ^x3VjUZNk;yxfnc6PBKNm*tv|n2~OVJJVMU)fs4>df-ly
za$!;6%1Q~fJBV@CS*oIPD0r7#QJOXV&>(&IZw(S1%&GIET)(mZ(Z{E~!ug1E{J>3R
zSqz5+!n<h*$S+cA>CuarF#7Ui^T*w}x|%(1FbnZj#l})~dE=ixQ=D7ML7fh%9U#ss
zb%F&N7F2&xXH`4>(wsFb2;z2F4gQOhjV~n=K%NmdDJuUQ8EsvM+QSVHuUrN+)Ic5Q
z`DMD4%M2REjJ}!ZlzngbJ9Axew#HQixIUO)AIOxMJ|bmCGKxQ;DhC<Gcx%9eU{19H
zMdVGoeBVkP%>FZ`Fn$fXiXbLdG&fm;|MVa(h3Ll<=+>vMLpnS=xOSE>(Bm6JkNCh^
zXqg-5<2fmFc{;UofvBhM(4E_N3(vPgrOXPcb5CK;%eC!gK<~?rHgI1XdxvIIBh+>d
zsBaOAA{%)zbua~mfKUIgi4Z9>A`^r)#yLxIz&UuPArdG7Q)GV5eK)U^=Aou5h9Mks
z(|6}zT(iIDL}h^oi}r=jR)1-<>BCPoU^+p>2Lub5(0ykB?i<}-+&7UW+LR|!ug@Xo
zCx};XpXacoRSgVv?O~c4zaZccjm4}&e|l8YH%+JDCytbk8(?hhh?kCO(*gRn&SAv)
z%hxaosdaYng`|G{p+fW|hIdDgzC-38O{9t=I+TSRP$PAHAA%Kh;S45zYnvG+PCv)#
zfTn=wT&cIqtS$iB_<p&lkq`!e`MXxfPo-3WfFw6%G9|+LFXo*3|G=D+>OyF5_p>8u
ze$5j3KU}$gW0L5kz$aWT)+N~Tr+DK(n&SWRtx5^<jE&@)w*0zM`FkYI2OWZQNDeRF
zD(Ncz_0jWB3#Fe6AxAnb4%_|me*N1O|DR97cQzEck9}4S1^?wj|A9REA5H3Scp)Y0
zWD@bAfW_yJ$CNexEPrl3BG~;MmkSJ<RcW;9|BEI|WmTT)ulEUbPZ%mbJA~R?*wg~Z
z(8>k^W|`uV7<go@5b>M|yot~69p;DAgU9aJHIZIGJ<ifhz(evFLIpmJB9t6R56D+)
zgg}CJrL=Kt(W+E5wn6SeyXx$mV$;qniEup6pjdy-@e4dDMtMzk=m$N2GEM(_5C7*2
z-+d%!&$E+B{?DAf2*pz=maD2&2)Au8O54Cy4mvVM#QS5b-o4Kiu$e6IX&_x9M@O4H
z73G$d`L2Z*6rbk>t>X3V+gR*jp4Z}JvDlx>2j1<Kk?*M`y{g_>4qYlcTz|yyJt?kz
z5p<YOAp^Wo)_d$><!hLjz3R_%p~I0qmu4+*qL3sE*@^E#!FL%=5<923)`Vur{Xv15
zj&KFx9~3uDHqUJKO<ny+yMCGEfz=dUB5GmGe_k@a`-pM-Q74Q1pPOdHwMjtx;vna+
z9prYDuZbXSEe_`Pw-5$~Ct{M`3C7aUq3IzxjB@*dM4$gxhFeZ{;^+82HCKITW)Pd7
z;ZeWfW&m?De!J9fs5I+&dWDNKae5D<iIs(}Zo_0x*CH#nU>szZ?Le(dy-hW6Th&y&
zvAzDxhLG#}-F%VnqH^gy)Ao;F(k>PR?R2_Lw(|bdlLW8#6T!o@Xm&~XPY$d(KN_l{
zD@A+^-6>))RHYF>z-k;6=Y?G$5k@~+=)KCSoWu?|GT~OvyS~gBpX?gEOBHohbYzUJ
z9)DoS;D|NxDArUtyc9ho2!$5wS8LK%>>+{n?uW;tr*}5|YJo|hqy_!k-PnQMud!Z1
zv3T0mk6-G0wBNB2s4<Ik?<`3D2OR#lhix6P)TW~`_{&j@Jwyuf)b4>-d<SGc()x%1
zLrK%G){m7(CB(1x6uvlDOVa)F-c<<TB)Du6I*rp?7H+1sKm%{oTTLO*{-pL*vv?lx
zP%kxo)+=n@>HI9;DEh0-jb0c&wWpWw68w1*sG?yv?gHe6>!qz@jDdLXBhRi=<P50t
zbS2t%?OP4uIel%VVFux@Y-`>*<TyKN$3U36Hcq}Q5B}9RmxG^Js^WM`dro~ji13(*
zTfy@Bg!!|PZr_(L@Rgcm_krEsHrSw?WQ<~&#UOzAMf{%t7%~M@+h12!eSC&1Od;!W
z;||-fa{6I|pQIsqb1*vLo(}pL*Q}6rqEzjXVR~7F*MVKOaCVJy&*w2gD>TnWy15HU
zetxw%B8M-CCdeh^E04Xry9!_DIC&;a78W+{kPms#>_$+@6h$((<V^Hr&r>YR2t+Yi
zk6nrlEcbPLYdw5ww}G{;X)ItO-zbqFi{HAEqMmL)T)4h)r7)GrW@f?r{h(`+XAi*=
zdsxECbF=RUOFb{WIQQ`JBAa%)cHWc3)vswhpsuMkr*epNIAkz3INWF$bLW}B{qFsF
z+)|~`j7s0sXj%B6?4_6g(OTJZ2S&?Oq3hU&^>n#+=t@Ej%`f{GscM9@I(|Q9=68yD
zjyoEi=VmnGBPemo=4~76+mZW;ZAiRcBVrK42L4pfXq<+xf}IyVR5+!*1!Gnv5&A%z
zdfP;%_iYts5;k`M?<Bjlnf2V@DX6sNkB?ToJNB0fxht37h6nmt!$RzCIRPX78^^1`
zlEd<opcz(z<P?q;Y6m6dqgJ7*0OGs(uglus2ibv!6f3S40EqTzdD;u$#q+%F8CwT<
zgRz;Lf`efU0FnBLNbc%+kL63uvFmoUbZ^PI4Y&{;DGU()H>%sClK(HN8}{E)-I~5-
zSrBK93hpP)B}o^o0=&>MadPZj!Z}Pf$Xj*Q<R}FoJ*Q(tVG|MP>RLYmhG1OIU+jXF
z68Pj&ypLUp*Px;s`8ferLhGZX(rBi%HXpvTmPD}Jh-(Fqn*q{!is?0`VR}n#F;-tA
z74IvKW@SXD!zp<%($hJJxc=!as68$LK&yUCO%$^QP6`>M8v9San>F>Y8T~tmx^o>D
zq_U*87<D>Fwsn7OxLk7Y8*#}Gu*&!nFjV9$R=$egXi!>@NhKR*QJtP?O$_L#U(cgu
zoZr}PB%&!uaAk(LDGZD5&Im1}`KuYu8##l{dE}Uo7G;}k*L<EVWqf0;79q~vgI{;t
zpVCI~dPh;B@$=n;^WRQrV4b@J*BwVb0yv`$ZmpF&Z8r=|FG65J@yG#Ci0O-*tDZ^W
zJ45w7eEIb3=P#BjhPf-(Dt3VQ^c2F7TOr@TFlkie;{p3JK11PJN|^W&kQ3uTm27Yc
zhJ?C+wY|2YTuIP6d-A7y?cbI-jz~WulB7ZSdniF#4Or&d!-?Giwdw_+mLz*F2TtAt
z3^NYAi8QnfT<;z1ib6v|Omda|R66RTII6XvZ1jPy52DVe(_i~UL8qrmOLw1v5Uez#
z3!u_<!Z%jq5eK%CJ=i8CuA(v$ks#EMuJwG!yH{%WB9Z`?I_~#kDDtUW0lX+0F?<6d
zq{U7Yl;eZ?4mXJ}f~W5&D~I1Mk;w;yM;H*kU@4P39JR5VhA{W>8k!ldG9#cNC0<C<
zB!MN6yGo;0Q1qC6K>M^zH)P=R#l~g4x+$JiM@zTB@RWm}6liLq5T#0NyDv!3eC7m5
zmY>a<j@{}fz>&jb`V$mXiq9|IQR2dlK8wzN0z}N4=&rEQCQ@83d&|fTO9B@BhvdN|
zzw@%}MK;4CR#<3i!>6nk!AatsBYyNKOr*`EKkx`*A0q|Abe*0}^}rljhJlM`gNG;3
zY*_c3+%{99{T%BqF=$moAu$VF&-Xt_A-Bvbyc8o*&J&pxWq#fNU?d5Kai!u|mM?+p
zz790YRv?cVIi}O}pSfx(d`>nDCF^?uz0Euyat+D~({Eyn0q<6pxJ_E=+;d7(cw&%6
zes`=0V|W#~g610&APLZ;TQx^AEfA7HC;stfw&ch-?fhvz>qPcztHEB<Tv28s0n?~8
zMBuw!wYKB|S|(u59V7GyjUi!a(UV2(+NUEWW{KE%?q|?%t4d3Vb*xGFR^zQuR{khU
z#37y;q+Exs{YLn`eisgbNQB{tp6bH=04l_4y{RL@1F9-^8T8%FzFQrl`%#y-;bKc1
zh6$}GdVJpPykK6rHi>-zPN&b0WX@H!AZan%5?HECvZE*aL$#Ukq_E!g1fw?e6CSh`
zeHTN;{gm?pQJsGk0sp67{k{12{lJh$r|_?w(Fad_ywaRGe%dV#^yFfZQu%sum5P1K
z_o@3He7i%Nh$F;q0jit$%W6jyOB(#&`YRQ?1&-*KG~&PYA82=%8U?xS*e{Cn#B25!
z<9;vj=4b{?&S8;-Ysu?~emPQT8atbjHhbKfZ>F7tL}U->mFA9lDBedb`y|U6IrAC|
z^BQ$An%LBuJkwD9IA*70*zeEd<)bdem2j&s+Z4})<QZbbC|94z-f3^^BU%>3B#jm!
z4So-_0?vGVot$8v;H5=lzRwTeK4f2GCP=*ASoW+tH|%0EE^BfaMG~%}gL#h5pcR#+
zR#TbqP;@cCu_H;K4;wZ1@>39J9-&P`V+$^8P>}Og4vZqSywOB2Ty~Ipa@dmFI+$C2
zE!#kdru*f&LFPr3dXu}Nh?j!;rwnypwf-Nuv$@Y@vm0iU_WL<=)z?Do)QYcnJEp1<
z%)oo#P08eX@U^V|R9O+sF-AsJZ<V&TEB?lUi|$VepQ`R8=L)`=qb*n>SI{F(oMPvq
zoZmX0`@FnhZ1wTZj0>cKl)RZ@j&OX#=$Jiztz=PaExylN=gC&*fH!IqTIlcHVjG<W
zZ|&HiR7|H+(9voX>i5qZ@pI$v9!5*4UThm!U*l3)t9H~39NacIw?!D&Z1AkPGTJOJ
zLt&{ljQEGEuX6~%8H0vp^NDUuP8`fV-}d}7p?aYGFVzFeCm8>)C|h4j=2H-Xsk?~5
zn!iJd+a+tcc$VuKQSqq&bV|P~&&Zv9qS|*i6vX}C!{*kKdhM9?2(=`sNCz~g=q)B<
zKTMD1pdy88>gKo&^02e#rM{u7>IW0j*rTh~NE|+3*CZCt4Nf+4niI<PVO3Y1OpCWe
zf7l_yB?mED&x5qP^VnS05oR`eJYcR5c|=QR?1yD9lqKI){xpv8IU9mJ82N_<1Bg(V
z*I2+A=#~GG9Ltu;{UyCdmwA0H8-1sn;$n09c=x49@INsGP^c<Y6OV399Q|~LxXql*
zbl<Nqi6V8(<#P3sIP%I$!L=@+bmA|oen25F3}qcz>BF)3y966Q`#WCt^`;g(5MHJG
zsDgTz8(f@ZKU);42E5AB6JU{cd}re8AIlpl53=fcpVBNcWIN1COQ_`qw^>wd;#1kJ
z-QZI$slB!<5}YeGS*gYgk591BvWgy4DY3LnVR%Cc*0C5Fboa`(@fydr^VW2dCK;Q;
z&Wm^B5gsiti&!Ct-0RkZn%R}LT7`+rp5lir|J3Qj>%MMCQ&-)|-12+hTJ+_t&*ZJf
zs@cbSGAkd7=m{8~K&<JTdKr9)Woe2WhBc)>sBGFiE_)r%pL&Hf$y;E*pzmybKG}tn
z!R#_-$xgaqXV6GaMMsc0tWr8Bd=_|CpK-I9(O%fzIj)Mg+jVcK&B8uj59RheLTbV>
zAOsy*<keaX%TgkD3#mJZa~@E1Vvd+&t7IEjFm+sAEr%*m)<EV}kQ@I&;Cz#9!+ziu
zcI@QIT+RtPSyj6Y93nT$=Ap%g{l?)Uv$`?z#Q<#7y$rF-fd?_)sFw(>GcFP+EKmMe
z22;q(m=sLg@7W1!y5P=HmPM>F-&0*GBhd55;wO+y9C~A%it?%#Jcd2hO}+Inu8~!d
z^sv6s3J4WJYY(nZAWXLn;G~Us%~ntQg$HR0JOua4?suWyLO%0FK`~w5W<_&0FXMaq
zP8Mp!*_q~k6RMg7!82{gVmI_7rW>aVTk0Q81(G*Ihg7APwe|4Z!6K}UU{gA7Ru<15
zu2&c1etE}k)Ayn$_eatFY4b0ICU9*M(SabPA)o+n-QE~9$jY3w!Pqmtt!2hQ94pi-
zhn!#~AvQ`xMN7On#0~GGBxbAr&}wN0qnK*CuLm3@35Tm!p){A1VGDCRjcdrUPzAn8
zTFTXF0y+jw!5MTLa@8)g7xU@jtP3~{Py_>s<;(Y;-(jWX(9u|Br_AHA(IqHl=LmYX
zw=o|cWYoox&u}*jI)se$XIaiG<GMqN;ur2y+gazZT^9!KyezPeN!z|zx|n?j<D|!Q
zk_LdPK5K*bmmYyA@HAYQI5@&N)a(t^5`Ad0jmH$`9s&0)BST-cI2tkJWhog3Ih>r_
z`*lb{mmN%R5dHL|Dnt4TLY8{l{%C3NrUp<@S$R1n7iGmU!f+MPEm<GX)Xfc;ru*RR
z=9vlT!Wcoe@ep_JhTLyKH)EC=i??26XGkNwu$<Lv{T=Z+y{Z~6cLz)k7qH^x?Xb(f
z4$oYFFh+5M64}k!Zl!QR{`(V9p?qED5o(z-B|QoaYK%iHwfOA6wXWV9>*;%EO}ZKu
z6i-I*os$w*BD*>VzaSL^^`SvAQ6H>aImVxcU;1#pzwf?Q0{@`!ILqqKGzR@y*u%2S
zTg?%!tX%bCSkO%}g7cRa6K<|nd4uVswRhK|Y^MUUFI+np$J_WsG4hOi%<&9?YJDoQ
zx)Ux<>G~BLr)ovTCU^{^?`KOdE71}@{V8BkO|_?->#V4Hy6VSQL_gDWY2nIu_xo2P
zicr7eQTC${G|~L+u;w4o5caw2z{O<f1z%qalNwMfC=-gUN)and(PN!`?&iZ6{ShgO
zo!}%0-3uV3s3bpsbu~C@Y)_4sZhm0wacC~Rixqor3l*`9lnXk877xqWUi2j#TJp=0
zlb-c6YUtjS-P+J=mKzBOw@?la1q4u<=<+jLcW*JDM{zJ0hrY8CZA+3i>IdOQf`DZl
z0v)r=TxZ;Etdh!k2^7o+bZ8<WO4A=Hbo>EFDF7=7z5d4MRnVHUZfe@NBQSGF&>jgv
zb@JJ1(th*Pf%GClI+ifmM*GD5Q5Us)bu%L+1Kphjy#>5f-TnO}`G<67q{bAiMK2<I
zJoL2Lp-+uxibAD7^R`ewV+z~afnG0`>@cgi`#IFLes{3n!?E?iiedi%dSC7{ScLr1
z6VT~hTKulwPnCl+CSK`LiyYP5fUnqG`~BXy9;8k7Je%bH)vd^sl9IIgS3+(-{pq8$
zWcEz%hBFU={Y*GzF`%x-<qPd`IU|E7z)_!!gjBZNOsAGngKyF`ly=1f_+&$*w@hGL
zCR3)k*T}GCYx)$GU`GI-eT9f3&i8!Ah>AtNJ&roQhlbuqj;!22Pi>QtVhX<zYe<QE
zKn%U%Q(-ym2cCDF7fJaZ#$(JfWftN(n07F3vW|x0-pg+=LikN&O%izw0n1H>8Vyy-
zU^|?<+nvkClie_qV-mWqBH5gSrbupPvnRCMOyy3!Ab)PG`N*O;$8l`i_?AuO4?&{~
z!m+UF5`52ZIH;OYv2l-P@a(%izj&cHH&`2N4PCC(xnm9{hPi$PGZ6Fis<p##JW1Mt
zJw?gJUaWx4i?R3mDV}Jz*}04<GiBQ3f<9UJwKG$HV89OU@(i|&p_R$?36oGr_x(rI
zPpX>O#~JV6>zXGTtJ&$8$!;nRl$1P9o2Yf9Kf}N|?jO^#5+}Iha|)Brmw~;vH%qvQ
zVN#9TfUqGOfGy#zxmb19B-PFqSAy6FT>RgPW%MAdO`S;o4a`KM<*Iz<ATc$f%0Tfx
z5Ki-JN$TmE&$F#ci=qO|ubY-{4eGTBB(}+mQyL5*)HlM60*}W>Hx;ZHZ;mPRWfzfe
z#mw_3+4%d376=e@QMHMMPTmrqPGBc)hz_claZ6%Xul5k$T3E<tTi0&9IVv(}I~G0k
z>@I{j4x^{pnMn4(Hc)td#6I=Xrn0!j*NeH>`b#QV`0aA*;XJqP5^-IbZM2)?8VX}?
z-T|}N5%7v^(Jkzy!`p+ao6Q=Hl~hYeV*Fsd+lQ_W5tri6Jv{Gjg3JGX`tJtM1}me>
zgEReVnW2Ux`rir={j)wiEqKQbj43c!>$KRd=CUJdcOf#WIfQAmxpW$AU3~K2{P|ES
zlP@L34x}`E#k_4Own09#h^Uh)+4M2zQ4&p|2`c0giZZRiwMf8CfT%g$&J_9h=Hyo-
z_4nt~8<%jd*v!{>ACpguuP8Ln?@icZn47^!>|JpKr=HG2&_OSei$MDj7WUbxBlfTh
zqq5bgOMY_t>z<pnO^?qwf`FkGb?9r!L)0=AP=!V7Qde;mA2%Iy7@fG?2O}UYzA#AE
zsA30(Uc6UX=3WA_gyHIQ!vuMRJ`0P3DzcNcD6WC4zIg)yg_DA29|1N&{`NA?5CF9B
zBgF`sEF7GGojX_Xsjz0h-5R2=ev;Sj1_Co+$KHx+Yfr-5#Q@7!UZcUooc&hGM2=To
z8G;Z|Ed}5ZXXIUv!HOI<E%NgFF+*Vu`PePWGITBWs}WbWg?giy%OVc?Opv07423c-
z1ae$BW2z+knS(mY_b}=H0zP4p1qB*}iR@adSfWAKZ^{(iA8{^v$$o~&+XWMI=u~@V
zX?O97QR+3x+vN|fzo?vKDm=5YZnUJ`KUMFnFFoPs9qSPUAq%279Tj$@ocj2Dt1W>N
z;nmK=a5V7ei>^L=y!Q#UsJjPJ#&kW?2m$+L#@|;BU9B1RR9VkPt?rpHAqD7QRar{@
zX-<Z`4&!T>hW7QfbpI~A=88g}%rb}Q6V`{m&H^qLerfe=zq8hecP3>x&UvjxQ!ri(
zN%vU%oE2_4R`xm(j{CqLz?exNGyqTWA<*6rLGJ2C9!5HWgwzaENnziRuVDaJ;DSkS
z+SQ*xT;jc&NM42v07mbC71UQCF8i0?i(FHsWWZs_LHI$}qky>P4D*{xB%-V#fYBQ)
z4HDe_uuydL=(PXAT@yx=?+-j+ur&t|An->MmW3o{0K1$w{kDwvw3u8vZAd<9l!SHw
z;jLA5|2uC@>QhoovrwI+kO8%+(rPbB_X&i1T=w*t7w1`}b(xGM?7SuVgCB1OJDTyB
zjN8D<1h+_NKFNV^XI|7tUbZvvL?*s2NKGyAdPmu(9%cvJW}v{obL4-tpn^w55LtxR
zxsxR9(D|y)A=z=z4wm2pOoERJ@g)dz>jxaUPSi=4Nj`cYRGDV7lnKabh@du}g7WhV
zuyei*-f_O;3>t{L%jZ1ns$LSogO#*(c8Q<k>M3+y&TX=o9aAZq)dFXr54^Y{^B4&u
zP}O8^uLORW^-;3NdfP)_w+%|{pX!M&x41q;AUIsNjWaw9Su>q+3`zvq&9qqi&-%;y
zOnhtlKUmMWq9+FIQZRCdXjNNJlN*mhZKp+?pMfu%JB6%bfNumMs>%HZA_PZU2ZhnY
z>I3G_ZlgUuZLph>(qvKKASMZ+Ymuv+8#tCYDJM@-xiub|N;LKkz0=vZ!}6}yAu#1)
zj@{=9a@hmCm4C44W!m)tyyFyjpN}bsq0lEvcc1S`O)WiP`)aA;{Mq^CEs4F4R^;1D
z6CH>Zs8WhIx39SXg2*HkKK$2M=I;=h+lU<OVmIZ1(r-lfB7I^P*m!S-VDZMm9TQIl
zSdp_M*sR8NV@_94_AF{8T?}3LxN&i#M{;G-4wDfrWS`zc2LSoYvy-v-)8{7$#abs#
z9|>Kx9fpF56Km=-%*re>$YNTmcTv><lqaaLd2n|h$y-o2HhuwrHtK+IFjwU7xr;?@
z>*IMU>KxT{3KGcNCqFS%VQ7mG32pV1rH^5_X5+{}!Wlq-`j~d<i!;EZiBcT%%~H*q
z-}@q1;@{NlFTO!Qsp&c$6)ag|>{^Q0r;%qeP%fsAsoaHXp({vuNROXaem2HG=Mm5n
z1)bynozNy1CV80@lMph$ZFa7bgdMDa;<j#Jp)pf1I?no~v%(z74)^VMtu!u^zRd(q
zr+68yItm58(KvMSvR;#OO?^V1QX)er{LxTw>H@mk6rZXAIYZS}8G$k9pvb<DW7Gj}
z3oJI9`#r7v>)u$<G8Y{%6v()`Fv$+jBb@zYCL&8R<HU~WqV1_!;<8|IOzW%Evp>$#
z-kPfCX=UWlcQuVVWA${@={8sIbGIVv5anX+sf6G?n~9q0G3bnxgd!b=@l)Kj_3h;1
zglK2vIf8GJFzxz>2$+`(1x-rCPo@j#u>RCddX)U~W^Fd&EI5pTMP#|q_&ZTw%%anb
zVn3R-W3s|l|M4VuTM5MpJH61*aK0yEaD90QoCW7gz5?`_)D|3UoC;NBO$tDSAc-}5
zWm9h#7=_q(PUwHL`MlHFs+gegE8T>EQX0<wn-jIvKSOx0rv+i>IFo-OY9BP+YG&HM
zs;RMTG`~12-=817(4wK5u76(x{0z?$^ssV}ph#uzuB&QQ(}uc{5pVUga9l%ZZU}`f
zheD;_G(Kmyqdi~c`ZkD?JFDx@RH^0Y+~P-1&M~7}iVE{X_g+j^;-@w*av9_uRX0vH
zXYVcW*y3R37;J417g4Pb7K8Ym-C%sx=OybZskmZq*ZtV`RdxF-BM}>gX4xj4L1)sI
zHo-9~VlGPeg7N3F7<3uqPO{Y-A#Pm!gdFk2qza*Kdf`=9BJ7N42f~f-9!)8cuIWxZ
ziL#S~A!Nol-fIdgMGOnU9k6wN2Gb11e$%+FXKup$R5(NA&TkkJ<ixSszr1&_K3oh{
zHY(qPO;hTt#fab{hoiDx;jPJhY1tCa0z*>oc4g5Y%p9vcZ)%cGG;cX&WZdz7Pz&}C
zfjPmgO2CO`B-YogU@IE4ZJ;)#Q;G5sB<H9jA?<F1q1h!4{IgrK6g4akZ&Efh!4Iaa
z5W3=VE2<+I(IOkt<2FFBSUH+IzNkl@;n84tI#PlApo_4$Hm7KCur}yg_wpGITM|;Y
zdyX$Do1=rQO;r?hU3j*P#wa=fWM`p#db|HjR!*yy;IPpd_mujY;_i$=^q-u7I<nmq
zhJl-`cqQ&Kzo8rGk>oHmb>(`^NZ;LG@6S()=#rdS7Y~0{;K08y!LXWyFX26qnDc{X
zN}Rn<LEQYeH7eLKIlPLm&fV<yzRqA+g=0(i@m5Ys=;eX@Q+?ZJchL(m=1=!{ybI2x
zHU9CHpvB^cKu&(YGgB2-?)*)S=qpoSFHc;?zQX3((>W%@{VU3g6ahl)!DNfL!}<q)
z?Z$-KQT;I9+_pxd>?5cu#2Hs`5yt4FFACM%mN2}LS*H1QN8;(cEA5zp-C6IN+}q_(
zF<#lqu^=!XwpU#V!oWLjAIfjAoHI~||NJZBn(u9dwf4G;;^0r#8cd1kvX*SshzUal
zqfzuWg}32X=Ow@B7jXj5ns(?;*eTl$`1Q+NoBN9z=Wbpuy{4l&f`b^3!t~F=5^Ma7
zq5BVkjW_xs!L@Il-}qow43V4)mHTDAa}B9?UJ|!F{L^Sg93{0Y8DK6`As1jzb>(3a
z0iLk}!&9T%$Jo8BakavR2G^9v3=Hn?dmGyYgZ-qyB!W3OeRUiv5KGax;tRGP%Cs<N
zEl^f|!GA$e?2%UeHL)p!zL3=s{(tU}PT^>T07-}xt?T|@@AfSYy@mm(a__Z0Wm$)~
zCNpW;43dxDm+yBz8GNm0Hu16K1FKYG&eHcgZ?NRY4Blj?`m_0lot-pL=k2%pGv4d3
zpZ4!xP{zXViU>=M_~V-E)^8di#2DwFU#PJy-@Jc?Pr*aamU@2c7Oa62hpMW`CuP-D
z-C`{_(PCNZs$adW8ln8vweVvC#6jsVGWM(c|0@uc5xFVMJY@fWd^39AW`sOss$!P-
zx6{b~{-O#qTtFCl^YnlJr+>@Y>ZL&3i|`k%p8bke^6$6gKfVq_s+Q<Fo&V~K{I}v!
zA9PB8<PKj=w*J-Q|F?9=-(Qr9h6~_4M-l$V3V~OpE}s3%XRWk<=iR}6-z`bqSk8hX
z=~2y+n%VeMKOP9D9P$xX8!YDm1Fbpt{=k3v_*D3jkNH|N)n8Yi7kx?=15Obc<)Ig#
z%X_BXbsh}Lp#9*3pg<^U!yo<pBJ8v<<pgCw(=~EzFPO}-YT-M5fF{(hR_!x9CN^7J
zKfhMSjJe)QMy1_}wP9?uVt9jC$?s7XJblr4kiDYybpDp<4{eu76R~=Y@EPpw)Ry1g
zD9uxa8-IEM{PXFqVuP>|%g@ic{PC{S&SKa3Q=_A0;TC^hSci6+Q6>mK8uEoc<!z+=
zp!M8TwHSGOkMbso2G+-O{~I;UgXI!L$?DwiNlCWAJrdbpc!~J`w0$FW=*^JBD~Vz^
zHs{?cva%*dF7y~mJ8oduwl8}_dmIZh{0|-kS_)kjD`ON|aJ>+8_8h!@s7ClxUPcHI
zQc2MqvH^FChJF(D*XRGYUoZlFll=6*g%mcOb=jMJeh&a-KJ(tR>4!?in~-rn3%2Xe
zfP$y{8XPuaV7T25cpoQ3`Y0A>?dH)(`V_q}to8Z~g5zC3*Ec{eM5Gev-jE30$_W<c
zn@4c-o<n3t8)(C`z|zsx?e_5v#Kaw8d?y4!U?vxmR<7?l8=FvNLP&oYKu==O8Xso}
zc694qM?i_OiFIBY?DB#zAbRaPj+(#S@V@bK-h$@QLd2x4-XJ{YH;XpoGdXJ~pyE=U
z!5;SfXNDd!&!GVBXp@hz;Gg@Q)D$|stv{P!NOK+w(yy{g6=Y5lrM#xMO%Srn$RrgM
z{4O+^yW)p;iDxA^5Sx$AZGizWc{5Gy<A$=$jnor;u%@ke3IhrTPY9!12Q64&{@k9u
zl8;mD_&1r=;OZ4{D8E_wnX=%*_nHU>?M5$6?wOpc#BVP>pGnTp(k+_!^;0zh%d=yy
z{_>xf<$DfE9@U1MRq_Szc#;5Uw2_)DVA3YCq`Rukg%@Wifz3>Mc#$p!mx=|$`(!32
zNaYRB^z23Wwd~wk>jMTL6kgwo6ZqFw@OK-B1`?(mo7q+N#}cB)K*L@OD$3lw1HOv@
zXRxjVBWGARG7&_Pd+akE4Bkzo+V9?~-r4=0o%-=4-GKRv|30|#P`Vlx(H$M^Tq&lL
zKDfrRiL({>Y~{+jr(O-+t{hbbk@o^tJIM=0P+9%|^3#+KDs+tFQg0mYYb;QsuWUge
z;;Q46n+(q3;ZaRNxn<`~2Am}p^XC;4JEt96UJpnwUdqIe2NKBqN^Ys$uGhhY;;3dX
z=ApslMjbg<b@fz5x`wrrX3iw{k1Y&k)g>mn8$DkyZ{It5e?qWL>N-BY(RJ&054N~+
z;)&FpiQjD4w8QS~2?A+ha+sIMXla7T_AxI5ooPcVo%(8Zb0J1g*Gh28nKuWPIUNq`
zrV$5Co;BRZUK80Kf~aMppLi1fM*@c#g!+xCw`zK3U9wlLSD*j#s7WhBj?}S@mhs<j
zBUdoG#Qp-)d13oZ&&~0>&u-o==@NVfW4MIiYGecvL_ozVx?JRsMA<e+(paYJmXF-0
zLJSb^hF>geW!MHjQ7fp|zd=>@<hg;H0$2>u>_MpKMbyLPw;h+raLYilxUDMUa=9Vn
zc#6ooh-d|1oOp<x)qvE>c%$gfH5|4idOhn>&ABghM${lc^-V@xpzOaz%eO<zdk>PB
z*1tIP=q?O^Nt%k_golGLYy=qhaF4DVhN+yUn*e5-)f79DvfBc?%u{fLRdbr~bUx2x
z<O5heV{j9q6!1I(80Q`cB|n4z#)$dLh(n&swUFKMtll4sTnzJzPP@><iYd&*9i>J$
zTd-l%sr7SJZ82Egf&C<r;0we}tVekt%G~daH_iwK>ELZB^|Xay8@`(rtJQ9NR@3aZ
zYZGn>eVmlJYG?kfqEt`>T@k{ncf>4=_>IXW#%HwwT_u0zN`LM%JA}x!KDYy09aSTk
zY_-iH)!o}>O3T(}DjP2BU{7G^$&bdaT>f~m=r!USGAbY6f^4e0<^|Tn{f5Wu<Odip
z?U^^HU>f-(S3SEOA~R>fw(A*at#!0u*Q3H36q0=dTG1%$8p`Ig1Md|_2WXtXFo3Ir
zpW_*ZGKTefi7>I{++n7o6+#*jjX?wVA5}~FzuT?7Si)O60QS?_g2La4YDWI^WI85=
zqrdsQX~z2iHKfC8WK0%#OW%<EknDsBpzmnE0lT$M2{bB12y_nsuC^E)x|p{%Y$JuA
z-{nv0z&cM8kJ&dsGD_bA41f76orFilkG^3l$YOETGadF{r3JB(Xq%1K2Gv+C{c-q>
z6BzZ(0y;#pinyrFf+1t6ndmr<GSH6PPr}^U_P}c6;fiL>L&QoAf6N0&M&BR`v_zZ;
zVRO|VyM8<?f7j|=5=hP@e+ig_yQLl4zPH=uzaq%{QNF-?7b4xOK$L~K&Q#?v;u0dI
zt$-cu3mCkQM)rW~Xf|DXGVar)Fk5{vRQ(EBO-nVP1Wbxvl()%7!_01=DF-k}IBBWH
ztQ<&Vw~u?gwl8N-rz9N!n0Pwdk7Wo9MVI{Da*Iuqf=>`|C+T48W%<(A%JNvbgM|~}
z`6SaLt@LojIA8R8=CdR(dS{Nvw;rUe6$n#I*ir7@;aZe;0QhdvExf!m|1DRZ#u>0|
z5z^uj!=0aBzk-|Lyv9d!uDaWSF+V{J*Y(~R6uQNBIm1n(8qr3CXL#zX5YChE&U^>E
z6LrpwKy2L~`8ln_bCxZJ>URDSQ@|0mgGxexN<BKYuc+%MWH^?NkXRwiw}`t_OBY%S
zahY&wu|~De#;Tva6@;;zqmmd8X?XXOb1NEiDF$@vFu|(qksccdyFCY)9miWo4v{eG
zbwmrZ%!5_5)JQ1W32R07XU@w=dye}8m55E*SHw@J^&mPI?P@lf8fX4Ae>R#-b^~*E
z2C1I=!kOkQld55nrv@%i_*v5-hH9rXm-@1!B1~ao=KwU<7md5UYy~eXGzexw7WG(u
zEW#Gz2<>A^7x}j6a^dco9tN2pAyV%b$)(m~gSow2d6jY6ubl#mjXn8eKHoOfaYlxm
zydNFVElm(BH;FZ|#u6fLuqa7X=eE}U`T<-K?!Y-^b*V_wlqRnZh*HY^4zJat<4b)V
zUuAS+4&8>${^=3N#Y$YF8nqcP!I}d{ZYK`1cb;S$nJG@zE5UbQp5C<u%F^7r(eIKH
zMFK%P<fJ7nKfeM+Jyt!v_D5yB!}8ani^HVbVq)`1)NHVGLeA#c#IfMUa{Y10R)jN8
zzpq09Y1Nr2VLu1b<NJ9R=0)XRou&Q95!`P_c0d2=!+hs|pTn)@8y>Y^y_;=tL@lAj
zBo1oBxL^<LfhufpOiRcyp8gmr%yb8I2kM0ne**Kf`~52eJEd=@&MMZRGf<$P&0+=5
zL|<CE>NQ?44wmf6!!=>w@-m&y3+Jv8y8Na<I{leRAg)m`(JpHCs>@lW$=z(pqqf(u
z*p<cQW;c7D9|0lpe8<T13)|(xcCrJ8kJH-lj_FSYRwz{)-P6P&Tl8#2&p>I*j@%q1
zxR63}IWp(|4P%_j|3}_?$7A`o|KkmnQQ;y&LiPw*k#PyxWoJf|tPrwS$w)G?FUgh_
zva*S+>|{mRdvlSk?{U7X`@TQ-{l5ErzrTNfe|S84RIYKp&hzy;j_2_lPb?+0vWBP%
zGV9CKto`0iC3z<hRbNfPFTWXlTgc()>KvH5DBi{=mXD1-BSE}=*!Y9j6<k@`vJV@!
z_y^n%Hnh&L9%uw7FrlPrx)n^xl1Jo<p40a;pVJ)3p-ni>j?0fYnvpy_zKb(@>HhI|
z<Pn))a`4aWZQC$v8>AoQ=mYgU$%U*m9RtfLM-TyE_$h~4lyP^j#6``~aoZ7o7sO@z
zAi*i($_*s2@7k~Lf4$1;dc7paXkOM_bM(87{(YLz-k%;WhJja`gp8ZRW36h)Xfu6q
zcVUldEe+b<{Gl}lHDuG5mpxR;i{)=mY+M)|5yp#4g@I2Y9^q>6i4_s$<>p?_A;*=W
z<Tz)$9Th0kVx^;h%Vqw)NK_6(nwO7echY{T4c-fZgOki%ry+bMr?n@h*Cjpf<<lwF
zHZCWez_iwP9<DnL46_gYF@55m=QmzPjw}P{7??S<Mr-@ngxZaxuQMrca?i)e8=pw$
z>iSU|ajp9^^B=gpj}nc&$KF<FYi!&$f-{JD;taj%+D}WgMvYlM#Uy`&5$#OXjA$FF
zjSSLAebb%d>ivtE@|VG-OjlRM=#*1zZP^o<YBxMDy{AEtwc6Zdgak`FW4wx+M+Vma
zGCMO;fBKRwVoz*zH#zYwkjGy0yd&rlJpJ%pm&2rwHk=sly<*fjC#;Q#@~@?uzsmwD
z1kv?~{HpiTA1YoxQ20P6lbA2EFiVso`872$ub|oNtX9DwRI0{nbW^pw7dbfnioZaj
zkV*5TFpF_6UZ3WDYPm(WuVU>1i;`Xca6-#NH!`03aevpm+x0{R%Jq`L*Zq~V8Fa~Q
zgSKM39}p+B^MqWjMC)WR&Cqa|4s!f2jyvCw6<XRhT>f21M;;P;PLE?K2#Vr+LSB4N
zZO3ptOO7i^6IEV4`?S-AsQcorvhKAJ|KoJb*i=5qRw|EDy6^X$&h)*Eos+RS%Ncrg
zaV_);-ZQbWIlpNsr)mOM9m(yVc6B!HoYI{W^(d7KPao_P3Da^IrK-Z{d`^v(1M(*y
z32++j)7Gtm^dLqJ$Ku2h!l%ziF0l>M;K<o<fC2A43&=w_zl2mbFV6O|*9IFc){-Sx
zs82p*F+2{&#lPLqrU78@5YaGvt|hs0a5gH__}u7$#z0O5H-X`&@h<S=zTcjlqgik(
z;=&fYy?aeLWAAybNnPEN<5%exuwcMwL8gjWq0w^`D<NKJ5%s)y9akCN*!APpR*;9v
zf3c|DAk&pmEypho#v1L06bA7PS?uwiq_d}zHzfoo6l?BL8glUvs&>N2noC1bD>py_
zmcEku#t!%3HW~>pxalFu3{|vg`gQ$3Lr&?Zj$H4&zQ0xT&s>%}I6)$7vPet<P+sV>
zroo4n_!KW2$OG;$(FD70+@zlyRpP5Ny(u;)u6@dedF~?7kF&IPw?6bed$uBlc4+~Z
zu9%O6G)R-yT^^l&`PE9QS9`g`qWL7C_)jJ^+*sIZQ5y`A#S0MkW_?E^*EA{4V=YqS
zv0wUSW|f-c!uwx%fGsx9_LD890QR`~&E`&bXCY3)bms0X_#E^WsVX6MKvZl3L}xOe
zx`YIP16SWlJ2;0cQGgB(%5CN_8v6tZDhjB}fUTw8Plh;fA<ymuRt^oK3xF%2PGQe!
zq)ujV%OW9J+u-;XrSTf*;z?daDjm1Aw?Q0_LwxGi#kYMdROi4xH{l0cNf=la%XNpP
zeTl(8HVJN3s8fp<=k|Fr=hWBK&oWZ-XsI7!pq$g?JC~Z?SwAQ3iB$O*A-mV<BT9h%
z7xoJPqeX=}tMFDO*lT&Y>l|ya9eI30+Fp(m4#fi2GY_Zqa_^PvKp$?o{FysVXg{b~
zHopp>g8T0f>j#xYL-h@e2r$X;YoxB-MtEe3Vy6)vxNC#x_C(nCz0XA0Bz@j%$8wV5
zgeRe(C95tT2CErch|awCjmBOTVt{hepx_wLC~=P^AM{K4<*ygL3?Zq4nF<(@{#kJG
zWo5e^iI8yD4A~Lf#T3`JhBqg%5FDTJdMAKDU=)y9d}FpRHy*(1EAJG$k6Al{Kao_?
zhkwKi{4E^%cmTo37uzUYe@-vOk4V!75{?tmFM>BU)6|pOxML)^%(Nj-o+?zfM9YqW
zpATF42%%S<TN2wzN~1b(aElosl<;=TI18TDb!breGdZ<tGUsGm7T=!a@-!PugH8mU
z<s>B54xFVo?Kn4{5%W3cdHZ4`AO`fJvcgiL?*>0zqKkL<b%8Go#p2CEV4;{ftE$CB
z%<%YQ`EnII5z(1Y+63SnH{OL&W)Me^{T9Uc`1=Nvo)R`UlEAg>8H`__>uHfHC!%@%
zigwFK*&N}9)nxm=<bNFR(!g{0!p^b0WzduiC5#ZdTfDKUv&jH6zBBhU3T|nrt~i1$
zB!-1T<aU$~!$xNey=>)(7r_K{nw9>8d47^jfg&6DrX4Q<0K^+mReHyCf7rI=BGvuO
z@Jc2<vpjo-qO1qIf?V$i<>nDTQuK!vf=dspa1LVucHo?&)1YmZ^m<<o>W8Gn-oFgf
z9LsO1SwCXxj^nqud0!j<!06M;k>B9Y-vJ*X;u$#WAB6tt({~Cb&Qiue=T4k?nr@z`
zU0IZ|9wi8LBX(3*F_0Bc(@|m>H!@8U$-rJd&aHky=TagraS);5c6Wf`lZ^|p+|8Mn
z-^J!n;#O<63ff*-_oS%z=GYh|(wHIEyv0OAM^ZC}lwuS0)bEFqYH1rIf}0WJ(aoq@
z6rH=qNh*z7OiH5j*L>GCR-X+%_;89Okm@T@C?98rQ#&;<R|E*>47qc(lJ>xqwZBYF
z#c)B7zBTt?`9*(7RS$QNAa2Avb{ZFc16&*Lh-)PS>FVGkDi5AzeNp&dTc@l*dBLK1
zY<hJ2#GpHXfEPu$zw>RZOn1*`snL~5_M30$n2nYaA1Mh@M;@Kx|MBQBxaVHoJ)n3!
zB!%^zsK?Gg8ifPczH8qszSNlYCjXi6_;cR;wd5PhelEu!&K={KvjuCr`qPOe0oP%7
zW_d^Wza-8ozrq<|_50_{^*A2GKKB6_fMcTR;$?AZ3OcULzkvZMWmm!7XLE<!aL6gy
zKAGXR#JbRubt%6_NPoP<C#7zjN`-=E?pueJY14Sck;ChjFvfF4(mn1EdB#@5VMd=D
z6{-y;c5_^Rot|e+#B2h5<Lu|~S}ZF!dTrY7t*He2203(yZpJ?0K8e+qXM+XWh^poJ
z$tWXIs{GC`c#qp3C54!?-w>S>I(3<vKSeRl_bLCV920&06F?_~ac!}9!gRfD%v)K$
zHg$CW`cU|7B*pFDS88W|wt+ogRP|sHdiFf~XZr?~n_p<wdh9Yp$@xxZ`H%O6QRNfJ
zr!Mf1J`2dE&$#6?*yx<+E~_KcwMkjHjCiVDC8gwFZi0p;rcz0~mEy+Sx8Aw|#IR{g
zb?Th!_rF}#l1JTmw^dI3G*y$@2zLJa;_~+>qzH+^8+gh4{dazeqR0gLxgP|5-sF8S
z@}A={akJB<Ag)}xg$2nCXt6QatLKbq6FNhDr)G~Q;ZJP4eo%V=!kaR1=WkoWi>eXw
zwgqLtP4VM3G!@#Tg!&I?&r2YNYAdhPMoDkMl=&NDlpe}zsNMa?9#<XJSuA)xh0T^%
zljR&*TmVQ(;O0}NT8YbF{XE3CUgQ!IU8ogSCsix-ccIh62`HJ^o$(ByI5D@rdCZf}
z^hn?B^Fbf*oDnsO%@Y&$r=stfqLq(x(Bjm)nQ0A=Og4?fgsj1vH#!(%n@gclK<Ddm
z^M2KgCBJgoK&d21RZC(85>fiJ24$#LR!)@H*>GP?DPC=$DmU00BcEdAxUb0);QAgw
zjp2txalT>XxnH4G7jl?g_S>@k%NH%lcb?<>9svD$8>a?O1rax%C8=|x92!`d0%W+2
zZG14(*TzYI#&IP#)WOGLRP?mPRmmXfd<hKDO1gW=gFa+{wI0Y~8zS$zKPC`)Ve}Ux
z?LPIjRj)u!-FL0v73jVM?Hb{DAr|tuyL*MECIKYt2%qiEemQ6qa8Yr^1=v<5YadRS
zUf$AN(0{VywK-lAeMMOe?x03*EELkIK}R<k;c<@N*Z-WU!S3gbKnaVb0VWg&Z-c)9
zm7FQ0p$d;MXw-X}Hg1EmtFJlPx(27_r?Z;%uCP?{DCxwZn&A*SN!uwwJ?j#v*;s4s
z*Zw^4N)SW5I%u#>DnA=t4jG)4L}!GZ_7-8MMG)Zo({3T2<%{(1endYvz@?z=^6Knh
zb~^T(n&!M*>-PY6;$Nw0Bv-^+PtbimLE}cms!RbBUw=-J8BQR8CP|^QAri-u{S8X5
z;XrEP-pj8x2x@3*)g_cqtdiCyfQeAv=_3IMFD)Cg5+Z{Mr&N#x(6y^H`!q}S(U-{L
z{f1!-A}llV^oorq<hvu-!MWP+wJ$L=M{&$&2yqV0D1|rD4s+pF01{_;BG+mrD);F5
zGaz4dC<Z8^1Tea-G3{!z=STORjMrak=$`(O=g738skHqA{j?e7o5!9Xv5<kw`<)|M
zn*V7cO8dk|N{KB-ZQhfi*;CdOfmZ_<y%}`bFtv9T-bpynyB#d|VT2sTHitSYhs)io
zhj;hCiv%|=Vz*;Egr@C=Ittu4H#gj?g9R_Jqn9dW<2LKWk5JkmdCVr9%Q7fl*_8AS
zn9x+p%5WeusltQlvJe@mA__75L_IaOo}f;vQjU6Q5r(^AEK=tfDW5U75l<>?IW@DU
zG2{fpnInZjlf6KPHid+t45374x+##*K8HA&`94gq?7dJH#jl7~e+DKa8%A0^r?BP?
z&nbqmsSS4!+FuX5tvkS6JEggABOp;iq3xKnWyQDJ3@a$MJOVT0AsT4yD}&MVH1oHT
z`xzb1sqR|4N<<8|fWc+9Tew5s%o3a4pl-(<mfEZ_U%{rnqM5P5sP8gtc1m7^W%SMW
zjJIN>ulL^#T48IsB1$JCb-lUR%}$vg$cv;qn#-=eKKngek1o39o5lJY+W}m=*C;Go
zKXp(&{(*c^a_F$d5kC;taD+}9x(9!O$E<SZ*1S3NEUVaSb{belP1W>=Z4^DPAU@q(
ze^!(>*fleE`Q)13OU&Kifz!)Ii^DN@Mz5T+bEoky)@*lt<|G^|dFTkyqeD2fhYsV4
z|JxtM$1gk@$+Zb&?)~du-m8j>51FqYedld6N^i|iX9-8On8E16#Hx&(*CS;UpeS7T
zIe`qFD*O2M(=V#Ti(rtHh^ngQk00W3L`H=y-sJgiPMwK66NEbF01#DFnCvw}V#YBw
z=N=6~Wgp3KEf&^eHStRy3vPh**#$gu80Jh>PM-tZT%UnnI5f1v&_}HIA?XSKl+e&M
zF#eet)B*E|B!HhkxtglI%tXQ=xVq#ssn(!GU(kM*;W4rNwX}fz`A4!MIwdsf1w*Rs
zV#uNELuW0Fh?aoSCfXASI^>k8r%MXzY=<!oH)sMgB(|D4zu3U0$@mPdcB<h9C|$<q
z<rI^5)^0{|Mck1wE!K-f6(74X%0hf!<fDaAN!p#94Z2Sj(VD2IrmxJb%-S;Iw&c}O
zy%u{EFlOAC7f)iciE>iYZzQ#KhHZIMWb&iD*9SL=Zl{;^aTeb`es~?1yOCYQN_;n(
zV6B;vOWTGFlUSQo6SNbjX~eK<5*zj%vsJ<c!H=$#fl(jtQZ(dJPJ^$0wR{ej|8Ak?
z>?>xg?AG}K2r8NL887Rt=-FL>{Z@&j@kG!S`R6(LYwMIss3PA=-8*67&EG7#?+IC|
zyXe<ja)1F_CUwM~J_(K|YWCx9dIj2HwyVCHYGYseJk>^}wCDQ}E;&>7j_erCfqA~%
zEB12}n<LyE;3g2#?_GP$VkkqeD!5rOUQ9GEFhm<}sLtau|8X1r^X4*j!-;$$H&yfq
z=hw|9e)Pf$G(}fPa<|Uw;%XOz5b(;b%7DSkX=W4sz}Wq!1x1#q{AQ162eS5d49iE%
z*$WZ5`BNKBAf>iwmzceR-I-PlnLrvYlSIyMps^oS+?#B9sr7LY1Y8+qt0eBJ+<9@l
zxmVgNOt%lBX+O^1L5B(xWRr8nIpj=Z$f+EmhDw<nBwT9uAW~{kDC4g2Lu`0A{7%Xh
ze_rbz0t?UhNb+j#)g<1UN3PI-h>(!DZdJV$Y)1_2SCfP~uFH<JQoZD@d0apg+Ac!o
za+)dHj;Y(0X-DpjE}k%7&=fyQ-hQhh$?A3$tAPU=$Z_pbcdDYN!u{#@`d>QLh=Orr
zyNJoQ4G))}N6_7OQS$>5G*__aW<ZcIxkv-Urq2~^QSwFfh1#4g2+)B!#K$)sJaw@d
z(|Iw%Wav$-&*=5Ga<_R&x1!_Mzm5t;Z+*f%XwSG#Qc(t{FJF(<l)qNAY&~LT+Abmk
zuRN*fHqlC-upI2%_;8A!r$DDZmml2_OtU;4<*&Veu5&;@CClWt{+=p_;>e8R-H2X4
z^S8-j(-Gg(0K5{y5=D+Bx$EqzxQRxGbYPycYcTm6muOpJ<?U2mQ+*sx-%=F0>Eb?1
z2c+IegWxIt&xkU-T|{YT6VuxdDM|~4QPc8M_1%a>!UNr1byi8o&V?M|g|04iJD0i!
zQnIB54n%9#2rdpzmG+tC@*3JSwL5b(20=*9C$Vmje*ZY=5x$(VV#WK(wux;feOY`*
zKY8;pp7_3gmgr!`)RcSF{I`T<z3tGd;OMiTX5B-$B6eO~FnIa($v<z#f8X>T`(8g(
zWANcT+J*4v-DygSV~ck{>0@*DF#N@BRyx|6cxP98Fexr7){<bt2E7Z{8WwB~f7)H)
z;&R6dxcnx+B0?YQc~zUwP`Us6Tl|kLR{hx_c!FP~Q2%~l<e6Z%z57tz8$P2lKMu!2
z#wLogB2j0?+=Nae>$T^@lzT^jW8p|<)JGo9+yFi{-&{#K>pmE?W%qgD;dmBxL^J>U
z7XB|^SA3sDd`0ue9dW{6d%eQN!#+RM1SNcQ)nz}!RMC|T-}(!TTvH<ne34VZ^eGPM
zlJ2ODq^ckm%uan3f=)ep4`7;`)~=IgV<1)3Wd_@d`|hvU@8(L(uq`-s#EH7&4qbfL
zU+XB7SP*b(QrEy?Mep~w``ZJC4u@#(jD$bW?-NKp_EGI|02O*6!v6gaS*{lbxce-t
z!{MxIdH@iy9q3Ua7Y8}25^RV=-mb$;&kk{Dx~w?Nyk030k5tYu_o%j^c+UU(HUH+4
zyj6!wQLIjgZuoWj%|*Sh^C$WCM!O*YN*?$bmqGlv@njgX?my2&9y|`n;|hfMu-RZd
z9<^etZ+cK?@f4K${>v5qk3aHh9l<@IT<H8{^ZRxF?kfDpuOaWL6@yE=r5htnrAzlW
zC-dL_%I~`^p#~+s(s$c0DCiGo``^F9`z|S{nY;J~`tkqq)&Fll@~KF;pN|Vae0c7^
z78HN~riXAMnczA_zW&s3^Z)renZAWvh`&EEdHfIS{r6Y!Egv!51$==5?<9T|K5x5;
zpC&cU7PRuTY;~xfIP?U$_M}%6@)j_G^+6YWgOdLACXZ8n;R>YQOTR?=^9pp9;DOqL
zYY70@W*%FgGdG)>v!<TQ4_<;H#ATRNLl}Z^AKdM+uT8`p0qc;VjSeHbOsnjU1IB=3
z^_7VxEu7_VEWoy+<bGjeYbE<xow*@{nK+6#wU5a|hWX5Klu6X92X?3bxK`qmc(hwt
zB5FFn9vu#QoC!JE3qDYK<_z7Bhx9ip!3PPu;CSOs<K>tP<z*|sMeiF(sI2i7Y)aN=
z&FFGCNqwdJ6eVBNEigJ;f=OJ*msAKlDl`1xyS+=<0&%=0h)*|%0r}-jLw$|f_68#v
zMy>Ec={ttnR1N_CHX=trMr9gMHkLB8a@fe7d-`&7MGFMp>p{qZ6)-WX%-V?E<=y?x
zlx_C;QU|8C*$z(!`W$umo2U2pYKm{b`f0k~kqG{Ei=GO45AXF2k~Y7AgNdHCdpz2+
zN*GQVq#>Q&<pzr1Ws6>pO@IQ+kPOuuJB&8I-uB@A<pBH+wpmQXyKz{rO|NPg;ZB^{
z^N-j$0ER#bK!2q|c2gyg_7Q$pGL=o^Lq(?dMYtp6-?9`3Hq_<|Ec62oCOU+)Qg@&S
z`v#zcHs@Mqu7QE;J<CTF{N~|QfP!0>eRb4IboI3C|8#cXP2FzAr3dei|KW~LzX6X$
zrk&o*nO}R}RpOBMi!f^6I8tsty9KP3kz-HoS+~L*FZ$mH63u*><FY<6>Er;Nk~_KH
zgU6d5T2E=Q$Z!~@?AiDbTiu?edb}vFU**|iM9ya|tzKVh)SlUJDI~SA1y)`y((Hqe
zTVuxI&`lOUU{x$IG+p$~WWevgJ+_b~n(Bw#@+NQce%<oQ7Jp4EE`wP{s{RUdup)3!
z%rx*0o>8xxqxLk$E7&5_nlF<H%jZ>g*;5j<47@#BGwgX!w^X83zT=DPpaq9RxVfi$
zUsm9zv-@Bj_79t1!YUDLb50E?g};^`&gTSfJM@peg3mT!Vr1XKQSf{hFxb7Kb};fB
z!)h>vs~y1VF1IilBB_abTj{qLdsKEo^rUB0>4ZkA{CW?3HB5Lnb5RsVqB?>m?#xbN
za$KUzPC5;nnkkzG2k`=a2cbK`zJg1YU>1_$x%goOahF*FwAlTo$`?Aa`V`d#+<kv^
znz9GRtGY9CK@H{S#xd5xF)6arm)Gqav8~ruu;pZa&g;~bj)*a=nlTT>RJj3_!^E5k
z-S%AYS#PCulHSPs><k$m{hHmcRZS&jCeI8&xTYFr;5NlV=cSa}GF1oDRCYyS%N~~f
z#5*~6pb>CF*IsGTn>HdIVq7(0ctzgAbKq7)UNRpI-$TM<#=7LjmEOrXiekp~i|azI
z*+yCoiy7>~5}q6P150x@RJwGO;}*KVo)++NGiW-%vlH)IT|H^(=TgRe!>8iJQQe-N
z($Kl%?PBze=vDi(TYgD4!nLB*lKAd9uuM*WEcqXam%;`ep%Ak3bL)HzCZX_fmVn*g
z4$TDMXm-W05g|K?GodRG7n7;LZGIk@Oy$rEe1-H*chg=3y1xaiOe>&Sw9hTVFOA2!
z%&~5QlRydhG}a>y>r0?*-5h&y%kf)}pO<A%X2cI-wEi~mscaFyzaMl5tBo{{NHQCU
zQk^Pf=k}uS1lfQMqvhO{*rp&zUG6kkf@)%^Q<Bg=j9JNHyfN4jVKjZ_6@WooF(8Gd
zF#H(EP^6PM%&+xKOQO~boD+6Kq5&PmP9xc{gZExqeg)O|5+d}80lc%`2t-m<xqx-1
zLKbA<Ab?iikm#9KMQUB}fH{8htn%AasRK5ksoSwp679cnT5#mhaITYnmow4cJ}^1g
z?M6UuY>)6oUh)_}BX_$GED<aCv15*X;AH`pTx0<$0s$Yh>MHO%K09%Oy6PLuCamDV
z&<`nvAfL4DH>DS;1O`l#r5?x3ou=Taa|idTM4>*Qs-X$P6L%4(u3gyA)iw!*7r2Yz
zxY?SQA9&#+?$3;B0R=0S6)-fcfV#f~_>?208f54iM`(P7Z#q>d#;A-?FW1DbN>D5k
z$1KOt=&)Ds^yKd8{Q#q&8N?816DGGM^g)6btcB}`-tZ~39d0vg3qZC~;6Ih{_iBA8
zP)`5$<n$F5n^&5<u`KE<<eq*_^s~VEn<3rN-Cb(os$SS!hkbWzW3_yEpL5wm179Kg
zBYe)*W#$+H)Pq-$jUjf<W$J4Q*adt8Bu(v$%ZYBjYWXj_3g39yKk^t)cq+{`pROzb
zUEj4I-H12-H?Xzx$W`{0kg|yX9L>sC@m;0zC}V>dwv~6E29^F525fUMh7fqQ17{QY
zw(~;}LvFn@aG7QF%wDF2yq^MuFSU1&oTntOW06`B(9;V&!Q|!=xS0)Cnfa4!?8|g?
zd1NnN8z87Y*s<u#yCZYiY2|h?j2P3(R9-G)yCnr5|MyMxFmAZQdyhj5SASI{osV$A
zmh&>tVS~ahdG>2lMO-@YFeC%{@bVlGf4hOH?dh>pIxWrgJV3FK7!$&54s7(hXj2)m
z_iWGa_HYVToFTz%F$20=dEwT}Lo5S;!Mau9xm+N@HSTc*Mj$;2!8Hi+{O-<8D0HCL
zhl$V1DPzm3!n<~Xgk&q<rE16e$dTb?7(g=5fi&dq*pnu2^B4X$ZU^uvpW6pxNNr0-
z<2{Jsj8sUNfqn6M*{cW>efK1IDA7E`FMlAc`nkbk>$?}EJ0#v*m!&}ZcWwL4FD5kB
zet1@Uyw@3`>%|rMte9w9GUK_DwS>ESp@Dz--Hk-V>D_i5$r56AKje$3k<CG{7k`Io
zEr0{Z>~0b#I*qT`q2JI?iY)P!YVc%x)`}j<bEyvW54V)<tZm~9Vd<Vf4m><8;w-Mu
z*n#<(4AWhL9@R#WO>-iW;&{gx8MrJCd@wlZc5lmqbP6sWV*lw>dnW*BQ<pTGgVdiG
z$0EjtLpMq0q?lHoDRQ5A1>*XlD)8Os@7fat(kOlQW4Djq^)Gr$OE8c6mPvGqnG)8q
zQn7{7Y_Hfpt^w6?-A-G;BK{I_nu(!NTT@-VN5%C*3U@#nwC}0!#v(DdNoqHQ7_){k
z6il{?m6D3B=P;cE!SM|@Ha1RoMw?r`d?eDPBQKwtzXvqA$xXsG#=s}vrAf{)iC*Bj
zlvJHtn(7!gbb>#%>?<#OJ<rG2W%=WfdEw9JSmy)@3RpWsm@x$}5aRBg<2HQQ4V7ny
z!LqUY<<9<7G-4t>i(op>kdGE$v5;Skj?zV24Y&f!od_e&&2uSx`0_eukK+lFZ%pmk
zwnO!EAyZL1S<b1;r{{wGSLt^@;GlDRn7<SQHsPv!JvRRPw{(HwY1Z4Zf2iFK;WRTO
zR2MS;;~qbRch0tn?m)(IpT+vTWyLW_*)Ht(9K4DGUAM1u;VRlCm|9q%r3+-V*QR<d
z@>pNi-fW!@CRfD3WY2Z=BwkIcosvjax}AUHdU8b_4ayi3^|o*aHM|Gh;FYnXSb^|-
z_H5;bYk5X^d11;*Ng`P3@ZK1~1DI0#w?5Indww*3ha6wQlvX%Y(2uA7d_WO>#y)?a
zUV&*%JWc{a8cc?1v~adu9yg(^J}uj0nXU@CEO<d7to(%@z~);BNZXzm-{gY)cnVCK
zM72JT<#Y1Lr%Fc}5r6rj_J;=NT3cSl6QUwSpd?*2H=Ca9f6=Q<LH=W_5smUGpKAL|
zu*qlL=4#N+Y<k_?J*K#A2K-@K*qBw~y!ndgwgS*^&PAe7a(o=Z?SkXa-`I0pT7?fd
zeykeBH-#eCod0VBbuw4#DN%T1u0JI44n6k@)hiR)REpwYENLZWk#}RKGjwvy;GN27
z3s3~Pd10a0R9X7GsM<b@pD}b%m5O#(U2jDbP#*P9j>ttbaW4l~WufNhMWff-3B7E3
zUW`~lUbgm>S#`uCo$()VrkZ@qtt%JgGuKDC|Hv&F!>>lqzG9?vlAX(yN|EzCsY11k
zUC|5rW7}j4{Rw*%?Z&TKZ$nDl3yqgQQOO=R@lh#>-#HmyZtT?lA!|A($(!xT9uvU<
z7df71Z&F4Vczt7$@lJMlGO`Q|&*ts3c!~^SA&#^JN-+g`B`VKnBx}$6k1;4Wky817
ze;Dp`)rijdyrO&&9T7`IP>L99E-6^>6s|Ov#ysaxPbR+|2dAS@OSo%ttg~?<x-z}4
zwq|O<w?QZA44%@&VK&YDm1KQf`VJ+OuPNt^c(HG}1bCV}$C5CM;df;2=|+^Cwh(z^
zNAPNP^8`VBOu;rJNI2P~)LF@$C)Q9oUopX&E1)LCG3leYn@~YDrGS!E;Jf%jIY&Lv
zE(tzJ#TLY9dMfbQ!QLGg<x}2PW`GyDCdOXlBJz>yO@mg)X}Mavz?VMcgxb!*_LU}?
zyjMf?@mus7x>KpILZuzFde<;lbXQTOus6iJ9m9%yzz`Q4Y<N$I;MBrqcdMgE>QMct
zGlT<denBwJ&HeLp-ghi3IwS|r{h`P{bYv311WnGXbH9-i|3^5U_@WrhyRWt_99Lt!
zMN4r!%b0^jcM@;aiu;_5OhB9f#FXORl9KToup!IC2o1_xHzVPQus#7in+eO@*Xqx;
zBodF+n*eK`pMPGIxq0`MlmR9~0Ov@3|BEm24{kjh;Zl1_Jn)PLOg1!WVlKsLe#ELy
zM@TuZsq^1$B3izvc>aT}J45HE{vQya+cvlXXZAqAb<f?w;hoM9RQVgQt9{}9fIn4r
zF+Jm((UL_~vs#SE4&qrUb`A@t5{)|m@%jra$Jf3_vuDIfI5&Te>K#tH7O;1aAW(nv
zf_Q}7T6jcOo9bGNVMe9mepCr>pBwHr%8PBmv`XFbuNA8sEe#zyr6U9C1hbnj<lm}4
zHN`8D?6+^D@rmx$JI8%nd=7)kv=~N01J#_jZ_K%Ls2ebBu}2tl>dtiq(Qc{r4?3xu
z>Icf)dpZyE_~|oy2wQpe9;#!5TH0uyR^(=)d}Y9gqfbVZjP5RK)TazM#gk**nt#I;
zzZ>EIQA(^4Ye`>T^c)NOjoX;s#yJhj@?vnxybMIp!nk6KzIJUEX(YxVb#0^UveJ!q
zZ<2Nio!66Dry}?X#o8;5%>^G;-Mn}|o|ibTfM((wuuLPHJ4A#4yME;(Yd`gg8GG|;
z-{N5L<|IdoY2f5_w;4sf(N-$$q5SWX)x|w^4DA)CCNEkHmBsqtJ1{vt1oQS5-QHU2
zD<mI&chV^lMJr8K@EPpMy87&460@kEK@Q75PcenpwXrzZhiS~$yQaELHuS}Qai9lM
z4H~Fz>WDM;Qb<U;OF<LcglTHQgs^4EYCEU4XRVGmLV>vCh#2*>ODQ|=-af)*qMs3q
zr~b7wQ4$+@>Q15rNN00gfq10{dDGb5Pi=5><c%I7h(6l^k(r+`kD4&Wd?pvrCP&x|
z%Hf=8uU#(#0mK!Oy;Ai4tdfv!PweuiWWUcMePV_~r$h~U`X~hs2ZO+*XHP$@;02F$
z_=kMp`LyG-wM#hl_7f0%T2P#rDD<2dB*`?pald5L*W<5N+q_We7@z)f1vZG6FhZ#1
zCDGt5r%|7F5#iSn88`}+<?A{5L^(@{scga9&r2gu2pyP?JR}Pm)SfXliTkp*;|#81
z`@5XkBto3E(WLpC)26j&3QRHsZMCN6BySBQ7Fl|7j3};5!&GEhJxY;i(ekqnw~aoV
zft|!Y@2EAK67%SOqUZ*&<>bqwoD-eR4P##6h3tD|?nbzf`<Y-I<&k`%ZBc6wpx}Qw
zMr|?fuHX5D_4mx?f81@qZ!(rgfCC(%Rl)g*d?^Rwu+eRWZ5hLKW!*@OAbSR}9OD{!
zA%5cmC-%X3y)qK3nN4uOAlu3;^64zjix)6~qMA^sf6Cap8%%C3x1y!SLnx&md*N^o
zWH3eDtoXS(^kVw2T^fs^?SK2As=eY%z0OCmqPJE0vZm+hzJV$~?p8u;tZMzOAIm@X
z(Ofsx;~9u9GEjs4oO8Vus?xlb(YISK;;kB-FF(ATIQ#5is-o#yKan+z;?=in<B#*f
zaYfC@RopqB8{7NNy9}(uOai^oqM8<nA+#fys$}v8TPy*yVts>gKpU#?S3bAw$y~~%
znd<wAxO7h|w4Ly8YE3q+_!v7)sfm|hi|3PRd80dZBp1bTaiN0zsl1>Ym?3ZRvws=}
zP0c(gDwi4720;Y$$~p4kl+7oC47<}e7o=}3V_f*Ws(+jmck-NJ`D9<kaCnV1Ep|-o
zVCxRoiuxC1Td*VEJBIqx-ktD?_$|$%?C-OePy7fIHw$}8uD?c75Vf#0zLLfHPqrvQ
zP}Iybq*y~AYjN^~x&Bi(r>i~`Z&=TeoK{I>Jj2sfakjVn4WUV9K8EhJMxKDgA+kF%
zIqgxXu0AQ=(inwF_WQ|NZjg1JQK82iV=ZSGVi%$5kuj<(>C_CJfT#iX*sxm3KsD%X
zUO2m84fW5AsbhO8%%t%H*az2OgKp}(m$(TN#gRTmfYoh-!Xz>VJSQf=g75ea17l!T
zT=Gin6y_q&=zZ0^XA==0x}>^3ftE>cT|{xtiND=nRbnTmQ}|<rdYr@(lh1lKsREYt
zEaqvi-$qZl>@xo%Gt>D%B$<xpijHgR%IF6*_srRpoK3|I89H|bV@yk2St0!bCIWxf
z;z)bNZqoPXkiU9uN`t#?53aIgvP-a;wc5;@ks6$p0gFc`|9DuI;x;Lq!25-NKRzy-
zpjR^t=1k`?Y57eO{8NSc=w>7u5RAxLEwtJl%*zbV7w-kHp`19h^BJ;Aa5Sa@IiM?1
z=mtS!Q$?t8#K%?^X%54<0TBWM=UB~@genYp+t!xc6Kgj(@%TM(E90H2!&{9OntD7~
zz0hAU3L~l0S|zPb5|lEaqPdmR_<F4;$R{WW_$WTIWBD;Swuwn~w^Cc#+jOxD(~Zp%
z-+iZS&ufTjst#ZLYJWHJIz5&r5>+7g(0ud(J@z3Pk_hT-39Jh}cQz(Zh)mOujaCXb
z8xpY~-NW=1HHloAaxe7xvA(EPvRE<MT&jIq0c$&REEC1w!*{|S-B*#`E2I(ybFMr-
zLhBdKw1Kr>^B&B2xM&I5RQu9^PzjyZT3}xD`86zv{YiiVeH$gVcUD7ht6iu~;ZLam
zg<ErV%SCUefs&hZx6V;!dynLXE@L~n{@+SUadD$?+N~DSoTfWLK9Y(gh3SebBWId*
z&x&$+ltKz*zDKww1zsvcZ71O-N{N-LyNz!f(52^y-pzoy;CT*X_%7h6w)jm9TVWwe
z@x1MVA6$1nV>OgyoIi(kngEX@fU|TWw6$?}zO(v4xD%JtcRc6O=r51r3P5(|tP6CM
zjW-^daXWNeYTG-sbV(H|jF)<N8A(GQsBqY(b~ZlXXgFO&F;^dq+Vi&%QJTicITMO)
zB}_A2O84e9B2i+aCqsFeSD2$7XCPv$#bXha!GO8Hma<!Q2y1V_Xrf%2e8E#2Hu^bd
zN<sqmxynt1ilKe0)BOvxF!bIu2!qPZxURX@z%|!~Y~WpQugrjCKn3F8zc<ib5fEJ3
zZXAC<zq8w;H5X)fg*>Y=W`R1U8fee)MQ<MSdGH}x#r^;nNxUy-(2=~%pt50KPP73+
zA~ueqXS$Vis<=UIEV)&^%`WLIRMz3L*Rfa3M2w5qCP$TAt?psu-*)pdyzZ$=@BsN_
zZLZ8@FxBPu4o2rTH5vx)_K5D{tu#(}F6BBn{fs%wpPk`spzy{@Y<g5~@|{L3x^x^F
z?%%BPS-8;x$z%@-LDYD+Er53Gb5;8O7;vOlU44h>cT8zQpMI7|lzDkS-`~J*)t7yP
z^Yr%<z9<pDiU-fpmuD+b%h(i$kAD~hAI4Spg40Z1Ipx$(a9*4m72+5Rf^(M_K^0ja
zv-I>zF4PQPJwY8#X0DEYFfZfkU&yddAV(($io8WrfXu|6CwV;AoGFLWko5{wO_iFw
zKAaKJN?wQdXL8^1e4hCvLK1rSw--QiWyu_{AcBV6H*ay764br}IH?Ei2!RRDrGw64
zw0jH`6jO*^pt|M0Ccb?sLv64A5a@F<5K9?NNRqkt_~;3y-l(%uL*CtuD=N3XaOg_J
z1+6tiPT0$CFbZlzed5-ykB8%R3STAjKz1SRfT4$w8e$;Y+)UM+iPfqlsMgY*fCMcK
zCxKFTnf<)7$f6dWa2?tzx7|2Q<eu@ACk&@DrdzOaXK{jMMP0wI-=QN^gn{hZ9k#?V
zgsW9@?JusD`o0_U+SBwrTpAxsFyP|N<x?s;&CnZmR^|xj1~4>hFAsxJsD5}74=SrZ
z4CfI@^H<y>-%bKg89RhGZVUwr5W?E;Mokq@>2!T%X$;!J$_pWc*!>DofLW5EDobrp
zIZtj~9MVW-S@^=hAm0#N-aJmrc49CLO`*{miYD!`7$erWH6kadEw8a>7lx9rap;wB
z7r4~tv>1aFigX!c1IJacbRcDfPIA7ah4?$9+;qLx9o?W-)Zs62%u{DmXCCDu|1j14
zYHgoW*Fl<F_9^F4g@N4`LyEZ6eR-@rR>}3gY|N9jy|g{IMB8~gsO$lC>8C~?&x^Sa
zU%hNoVV%m`1=<h`QgZZ(nqlD2VgzIuwmv&P0sJ$?zgAKhb-GzNOC%W}<0}0J9>7PH
zM%<O-)nc*+!opNbmyIfOYs)#^2oWy?HRsVdsVx3f+o5UxqBi$G42>R%3n75JWN_Ip
zz}<zGHdttf_|}n<l^2b%A@9C9=*Dw(xIvKX_b({4&y}?|Bpx#!uvO4I9|}#EOcL{V
zxv6b@#=u~5VpQou!pb?kc%I7G*~7mnTSWJPQti)xkhUVYZ|7EiU-L@t?AW@5dl1f?
zYX;fv0StQgp^0U%)<Yc0$Qc;?PiahdDR5oixU|mSr*aOY&WYyQa@*ivw{9s{!I0?h
zv7)JqaJO<fK55>g2l)LvNBu*v*ABkoxU_JMKmVGgY~6(yTB>Y)r~MyJ1YU|-#B77=
zM9E%q>>am^`i-P67}#y)ecm!tJX5Rbkbw|hqgW0U`OgZ2Aw_U5Gw|R%`))bbFHMPE
zbkvutW27o1yQ`LUGdl?x7!b(YUliu+03GBS-={Y&4*``ku~BcD4PglB(rUmw3zN{e
z_u6)`>3(jsMK*<VV|cdCOC3#=vyV-9S|>u#Z3{}_9)_;6+JFj$+kDT5=Vt#jG7EIh
z5h==k+g5~Ot5Ll$@wT0Dt*0ujC`VM;!So!XD*mp?+q8jcsYEZdKYAQ{a~kE;7$)M?
z&@wTwt6)sw8R;G2(z~@fKK|uJg%F#8M4n!T7;KLjp&Y!H9MUVA?L%C|E^cp%&P%8Y
z4nD{QHXEIA8AycHw%1IlREhTFaMPFZ*WM1KDv<xs?{XyYxz^|SOfmRzqK9B4mU9Sf
zXVu&Z+?B2%ZbG9rLcx<29d-iEzxKSpt@0m<AH(HH7HG9Q^3NJa>V3tc;O^t$&Hf5{
zq20%WnDmbZFZ&oxhFW6x-!CHcEpvd3+G{6;NUQ`&L{rx^dMjR*!baJKN#iup{}H=C
zviui+%Se0^$~r+3&+Rp)ARFMl%4AyYNg1#zzL{<<2<BCitU@#4+wMG^>;9<iRPD4{
zXeZj?ZQm~O`1Rz=C}tCh)WV7s0uqN(#LO5B93Ki_R9eJ1v0mizF!Z`$l5}jdnLXEu
ziK!dwe%U5=5SohdYj$xjqiZl}3?@k@Ool>ZojKOeDRNIMGE5@gFj~bZSf;iBL!(^`
zhG`7zu2!|+0p0<z&gs~CzBI*z(Z(UE`%5rlbeN*+2o7@=TDM~;sMPBt3;lwb7Dpz{
z)<3|XFLuh3E+f|Y%z!71uVQ^n?58T*{$fN)S$M>~Ji%5BqjBLtn|ob4A#JO7jdTny
zPhiqmOwJ2@NG+DLQ?TIT)u;_>6$mKp?T<-l?l?t~w6prTjPGiu!DSd*Juu&oKHH8V
z;>mW#;?8g=rWIR)`o+WQWuth<4eX$|XAUUn+NFHVHq6J9P#pZjMa&3v)uBL=e1(ML
zN*f*tZ8^dc%bM>IV^$QLE>zMsoqAK%$lJ=B%6nBqq-D@jq7(8mBd``s3mMzr4^*{Y
zpTEmI`YN@EWO%vlWjYNFiN6r`8D(zhm0Y-^UZwN3gc~{}iGHq_x)SKqtA9PCyV78p
z9eX3jTpufIA<pW-C3f1SamHDGs;I+Lv$OR3i=12KJNs9c%R_2iQ=|72yB+g}%BMuR
zMXpS%4BLnHI^XkA9iGN4J~;I!o<lmrVI1%Wk<<O%?}SUyMvdgAG^8Vq>upvB+Qdj_
zouBM2{R`6Ai4xLK(-J0yiF^7&m50fg1~HBwypDPH+_vY0LHT;1*dNhrN)mXtKr_C)
zW51{#;s9b=Bpj6~raSaFlkPTMR;6G2h<|r2sW3r?4*z~KJwM4=kiNNsjey%#XTorD
z6=&?l|7-5nAspc4&|rTJgbqD+p-*`59G@cNZ6<#9WyrA^Ciii*gczsFjerw>QMXKS
zNX=b8C0)y~V((sqDXLmutUCHJP+hk|)>h+!s4jC}T5zgRbdP7hyBx}>;MTtqC+fNr
z@(D4&TaJ{7B`0YvuCg56{(Uk3?qWRl6$iqc<XOfY+@A|$>WxEM_7)Octbj4E&q53A
zEi~&t*jto|lhMC01XB?l1`bhhe*X%l5;K8a<$B8x-NE94vaj0nN5dVC$iI8Le=L3`
zf;F(6#VPxRe`)F9K~Ux=z?j%u_5cw7yfNR$#U6k1_aT`X@E+WQzCkEftoI%!0~Cch
z*Ix=?zUcj#9;JL7me}}xGTYyBqyK(?{Pnj9IY8dZV($_;@v9a5$7lbIm;B~x{a(?3
z{)hj+ziqn15fxG)_ekw~6hAV)A9SJrTE_I#N8of0SZ)IA7hP&<wmfp6-=NykDF3iH
zd=d!zZZyFN_47a7He*PjS%<i~G>*>@YUlH&!n{4bkQx9(dnkrLY*S<h_ime+%amP|
zfrm4Wmo$5r!lC}oDUF%SjSZ5w!AM2Fm8#ejnmxN6jy6`aHUWso<bU`{hwhHftkCNF
zql8;Ry}%}tIB$^0gUV=$bYe5H>WlhcDO&%wbv;fEf8-Gx)s!UW!vto055SN9A;N2+
z@frED1XEEvBnq?O#fur`a}z4L;p%3>zNe|@5J?HNpeV%0czU4=p)-Q}?w)(#BzP4d
z79w_$20B*iE+Ua+R{h$$I6yxb(T9-6?k9y-t{ZccH=qwxAL?@AvHB|L2-Zm{I36f2
z3+qAfv7{6YlJD?`Y`+~x^U&xj0mJr}mpg!SZ^KILS!!2awUB;Y+Ww0JzguILf~zyx
ze_fLQ*f7#<k3dG8Bar0nnd9ATD>oh<>}@Y0h8jp_*YXpbqmb8Y3#mWZx)3%vpJK41
zqE}lsROwoNGg~KOy;XkD25Mv1zf?YJ_y=|J?mqx7{R$k`ui%A@Ff`@!6EA6Gt~>ED
z3G}O0<5*R!N?(RBLv?0^?tDXZL!HY(!I;yBIaxZ;7V_+#17fwW2eJ>*$9C;PjrI%u
zd7nO~ug8XH>0j7Uw012zwFWE@Tgrmm>VpAv=GW3&1`gc<X=UcslvU$B5!ruu;Ke2B
zfZm(<oOJ&e4^RElVZ5&sX7WjS=5mM&898ll>c}$KHQT@>-T_Rs>?80WYvoX{bvxD^
zBX{~~_T{l<2-+`A98V!8+tTqD$=dmJ$^wrKAW$j?jBIfa4%oL4KF7DYtXzG8KHsn6
zrFKbxNn5Rs)h_`VCR6hyJU1>rS`8nvR)zH`rj*b8*BwecnHD~<%1L_ZU%N4oWWljz
z7Ur9!-2?{{8V{hs%4VxE$m2a)W8h%~2{G7Ft9@b?O;Qa{(cFcwCw^!g&vyPup-);1
z=?jXv<-M8kp5L@P{xO>P&s|WR88+P>A<lLDpPwrv>``q4TU}fb!i3blhcW&NgW-u>
z`=Jy$lK5Pn#Mm|DQ8f_A7zLV5y-)CfhAMSYHH@YhdjmF;AxlqxA?>ZToSeI=D~5Gl
zcY9Pjk><)AMP^#?P9hRR>7!kfv$EG&+*zG69Z=8=l>R}nlS_cdPI~~E!{kf^wULS<
zTYXSx{FWX5WEATbf#pXrMw_O_3kR+8uS~wvwc_P<JI&JX+A?m4iuH|c*y}5@8H;p#
zuwvQP*$b9eg;HD%@7L}%8;3Rf8{dBYT|?l_YK)-oLAaedO|49ankr3gY;(`%7H?L~
z=?&)o#5-TfR=Z{Hkn}g~(zI(Zw=ODI$8Kn3vdHgu#NMIw)7y2J@jUDI;G%AiXX*o`
zM()OkXia>q-~;7XohL3eH(n}eGs5wFTs?q0&o+_4H79w=j`w4b+s21;PX}MM^mgXn
zwl=-2$4njYiUhN-^3w2J=s@sv{k(6LkC?@I7N2(ytE1I59lQ@Rtz!twdS&FslSIBy
zGzi5l6MR9h7Lf`z9jvIh+jPoUlGRGsk1zV}6qQOtgOXOa?DQD#PtuQc5u`jbkq`LQ
zmFh75lLnUf|AhuN_+M#Yj#O49iGx6oacAj=ePWv?Z9L-4*fRKSZK6GZEk2G+D3^e=
z5=<`liU$0rtpF;$e#?*K%PL@eJ1wsq20l`|%p@YTRI9zM)g>Qt{S`wT#BkwXcwj`v
z-U7DN3dHwjWj25bRu03Zb|@>_;$aH@%nDj$8|drH=RuNhf#hojAhlPO2W(eX>ocFM
ze(Y~L0zxHsvk!>_`$+{ePaqXWHYoJc>S;vsRAsPQk%Fb&+651r+3EMd1(=5kwKfd#
zNVx?2fuZyjex_jG#sLF-EC-pKWn>I<O$~6mMi0QbX=}AJ+Vr3$AZcnJ86D&fyDp2O
z1>`yq6Sc+j;2P)%9dRjK#m}sD1r5@7TtH#J<ADE4&Ad8%mBn6v)){dNqT;#xG>lpv
zluZ{E7%mXr(-|jWzVQ&?%%XX)f1na^daDc2ru&yIG%CQ<-6F^QK7WVjMwWpyndja(
zjr)vRp4iw^`mr817yUVXl4U6%863#std9k7U0HOGiNZIl0?W5DL^QAnUr?+t=X)`U
zlE@_+k!B2u&G}WV(KVG!-S(Z%vTj#;hDFx_6NDNK7^ILR3~+bdP22eU9=e?Pi{yd5
z9Nwp5p|?8)8{OQdyU(>B0Ij&0KCCP*G#COz;DBWpUk*Yr3%L*@>MCs?k-nWS*XzzM
zjb<Iat}DQnv6+8UaHD<y#&{{%yCHV?B_MclUM&`!`IfOWz;zjjbR4@H(@=w8U9+wv
z+9T<GF1x&zXF7m~m6pFD_QtG|OqiKLr*Bv3yLZaI&DP2&e^v3=@by9&izW7yiXp^o
zDW7H@7;R2u)w|!8&F^!G7&MmUOD<byTcjfFv<`*ETzLRD91(6F%`}ol99&RED0|Wo
zT8ABUrgEvgXKqF=zmEU1)7Vw7|6Yea*12#SogebM;xrY<nQDJUg(13tNaikJ2gtL*
z_2v`!icxJWeBm|!l;nd(12Khu_k~zKO&>j$j~r;=96ZFzg-)wu912QTW>ZXo!~PV&
zp$JoeeV#qE9DCm6%`P@9b@dY%2E=JB*PNhv4!+(6>Gt!zRfJPE4e|3=uy#Efn#Mhs
zFG<}+C+z;kB}2#Tdc1sl;V&u~x<rPF*1uQl&Z&57ujezs&TKMgXB|hJ*1q~(V_K0%
zm}E!dW`MI0tXMaNMpW`3<)p$q{Zv3?L82-RpG4>8H(+iw4jNvN`=LmJckdDrN#Gzd
zv(`WxNas1EUnI#-6&;r1h_kGk_I70o!U~9h)2Yht_r?8lye|yy((PX$@!b|mEiD4a
zFot~G#c)@?8aY6LXs%ITdAQ<ER1PJF4AY@CJ4tL$9ht`vaP|dOZI~hh6w_y?gYX=a
z&WpN8fv4mq_G_K?i-y@B2g^S$wAD+f<&CSfSE$AVdC2n`)R)-t4EqEj%Oe?0-uGVk
z`|nXk1TyhpV-;JYOWLoKo-+$H=i*+?HW=5jGncf}T{QRTyT@jq>%W;nxxoF&@UDbH
z$B?(|3%qZ86?|9U(y*^~$8o%7iVk!fj*k*%&Zpe7_QjjW*PqT*UD^?ZTUmAY`S)l*
zhj&WT_+QT>nGDOTF?Y%Y=x*KYBj?Z}GHW+iB32;rx(56-w?cp+b~m5ljflvQ+%oQI
zi!dC!YS4(2SNbev4H&5Rhm`^&hR`m<vv(w}Sa$&UJ<-<)ghFb}TR{yjkEF^un2>I6
ztaGXj*NbCsEvsh>xg&lDvUbdIg;hmS=9bT143RK}2Ed;C@=^ORkk}9U%JW{UFT`>u
z-Z8IMF{q7GYjS-EZdV!0OibxPitm@dm>FEMc{iO{nsBKIt@yMqyA#f)=(oK5G)QpF
zu;*g#_eRpHVWzL^ZgmF3D(j+hcTBuwt>sgz`?{woFK!C!C*DyI`rQWpeh~hp=wJza
zbkG&|d$28jl;J@Wo!=Hs%!Xv42kA{0=oH-JJRU_VU))RJw0S0V1d6h@{D;81ec@~$
z&?>3bD~f$Me(fYmbV32;h1q=Xc5Q7aY~<=b{7B$o`66%X*K%P`PkKJMOeoA+-i2?h
zN6Oxc62^Paz!5e>7gx8+<+-g;xHFd}GnLAk<(c@Vj;i#Z<TIl>6cBr^zQPd%in*gs
zn319WSVd>Mh$2(_%R+oceOW#F#FySKU?d=ZlO5sa#*@U%qu$+n4c5Sy7#;Yn6}2T)
zuy@bcgQT;A6>F-|8QaDCj5=7p6{4-3$=9*-&)9QTjCfb0I$0P3UrC8$YW1_3voB)9
zF)TWvZzUyC2gvp!%KOxxP{;g0aR*Ezaf%p|&yR*_?1x2$x82oTA3N(@FV78-Z})c^
zuYZG;&m~+uQ|YoHkN2<x7-}#b+*vaFTA2HC`jeC6eyNn#*hAQ_rBD(LVN-d%=}Uc)
zBT5}%3AZ{zh!8bp0E5JHX1p5yA(s=3j6DVx*+qL|cZX~Xp0&mb@$%%RAAE}~jrXlZ
zl`4>uE1^!9$MU5~@K{^xi#QFKr9nH_Xf!3IDp|keXf7FBr{?+m_IYeqOpjy4`wMa+
z@izq>RH7WBPpA)j`YTKR;6IMCZsy!K^a0SXGVHz5FkWcYkMP~f*E+rFCy`s{K8ujg
zA2wHfM0-LP!gus%=bmnH1hpr3cNKQP6mW~*W6&vh`!vICzbS_%Gfadd6@PzI3Vpxv
z*zq+TMr)02)l&3vKbi){gpUyn$3GD+t;KBs_Bp{8*~}%KPMiA}McS0C^17e9C*c(s
zq}C9iC!D!i`I&%;$VsLz^ieV1#89A8T>5-I+bg~w6M;$>_y{Cl7C5)z`42IiY*j=V
z%e|++WRSyfHPB~peJG#9%`nHvY<uHUuEYti5v4oyxwhL==IP1f5dTuqKLftXEo<Cl
z%^zLIpxYzkoS)EI&{sR<fFyD+r>+G>b7`k*p^?+Hs?aE6z@m$CfBD*_q;y#B$^eF{
zh_muXN+mwSS6`^y{rE(0uaaa=7gv|7Fw!)qQ~kLgx`+bDlME%iSV`d-_nqE<Cp6Tj
zH*n%G_wf0oA~uz1Z}&Br-c0ywj_o^jot^!bDMTJBwdE4U>>PB`cnn$83VlJYd({N?
zUV=ZYJ$H4R<b{3t!x;`;XFpq>t{@EGNFR6nV=ceycT7?nn1+9Im3SW2<6Jsyk!c**
z9BErUo^k3I2JoMF3xG)Kh;prYErtK*Rw4e1IAp{cMD-tChG6D7D>>R=Ei3y%VI|FV
zO_iSRnfGIS3XZF<L^N5I;$le0xDD7g8*lG(bJ}QLKf;8^%fR-cxH3ldl7tT&LA_Wo
ze9*hz=F;K6<`Fd;D1U|uGUMX6c*H}sZWz8|e?@jtXTs+dzQ48_CBL~kTdYyVo4Q2Q
z3WNXCu}QvR{xnjycOmLDNHYP1zV@)RT#3&xaafjAl7F6BL|>&dXfCOr!r-QkQS5l+
z^UlDu)OIrloC)YsC%5Mjt3Qi(IQJp#DuGaenoBFAl4<CkAeS_$#_R@NO?=!BXtjNo
zNFRm^Dv2=Oj_BjV>-q4t+f{CN0Zs%xW!*&O3@@~ES!;yvNs6&ujI)7-^nRavcB&f<
z-pKm-gL#7=F<+h!dotW3Otj<gJL8}?$v-U-YHG0C1GfrQ{@VEf42j&SGX=vwJE*<c
z;M2B@i${$dW>cf4*g_)=q8xB);%i?whSil*5;~s?N<%nsq<azAzaGN-$U_)QKK%y-
zQken(9<~oJZvNWZ?QbNYywLkAVx9+kvT@ZD-)^99C_E$*L)p#xeJ_K7$hxMis@5$Z
zKVY}etB0(J3n<)GQ45{s+@C?i*ziqKpxnx3m`-u*^6--CYmbF71Z!Z714<)JWuMzQ
zKh4azi1&pwPCL<!KrUt%+yYJMt_9%F>Sj2fxk85HNvCe!#5KV}tN`pXH4C>Vrv#o(
zU=~1TvU=e?9H1O3bvm+xKH(YN`f9@t3lMVVf1Z)RFo7o`j%0P3jO|%wAhITPqp1we
zcp6)pi#4cJ^EkHW6(Vo0zq-;BF!Ri;SiQzTtY#C!Y9d*-)TkFVkYZ^4M*HRqa<ntG
zD1=4jx8g}C`?%gzJ!@CQ&a^hbu96e#t%<mkqPZsGC8niv-aT^;_7z~*YFWQ|cF8As
zL5C8&Ry=Yu-thzHvh`0)-AqL4HAg|<a^&Zh$MQ^^5lCCvta`cL9w-jk?gubKc1i33
zw3#075<^uwL5;!nBZq4s(4#gKYWh6soAtP7CfaT!J(K9-?GPFFISZ{sE89>sj^}pm
z_VEvhBS}Y62vJMo4LTX;fzW<qXq<dVJJ_%gU9`Is^T8(tB;&lGMp1(c4{9=VeV?ER
z`Ykmo9W+YK+)_5S2W)Y`fmvW}-oQfde75>)0oBy+F+1-h<67V)szZr2TyE3}tEQJ^
z3G~Vc0bhoe3lv}M#Qjj}Ee^{d!(zfjsP#6JM05a*Q5xDQ)MPCz!721{lYV}dm`1>p
zv%>>ba|+au1~kAb(=R+%+av})ZyuKN-dyK>UL#ld`(IZW6g+<3bsqn`v{rv$to+TT
zEd`F6E9sj0s99q$O+5AsYQM*4Bo~y3{Jy0BhrRa<iZb2Wg=ZWbZBWrBsUV0nK~Rt=
zp#cSyoO70-L_tJCgMcE4C`y(Lk~4^a2sSv9C5t4<Ip-$FZ#|Cg+57AnzxO@A&s14u
zSuh=*JFIoBD}<U$4HJ8JWFo6riC+_!5ge8u86=jFVSP(WwB1S$=85%|GG;LjFU^)f
zx6`TC_Dnsf{4BG&+T^Y0gvrl|!8V=Y%(hy~Cmo@I>Cn<}3z}Tr81c`|%OY;8!4sch
zQ(t7J{V1+w&PH={ABm?!lpi{9A144O<!o<8SDuQiyju_bSv2?UlBpC%5(c)yPRvn-
z_~D#xEjq{w=HVbea9MK;NDcT6#qDf-3uWZA$|6<y2bmo`o9UMLM^r;HJE`k$`j@(O
z3&YO9uup`%NDG2w@q>~Sg3Nb$wHn;1y00v*)1lEjp<Bp-V17~+fDNLbU<~czFpgOD
zJ1G2JTSI>BJN;_Df@GXbMe4{sDCnp#8$Fhp*-xiE@UPuZhW#O%@EBrvD{VLJQR3vo
zIkI8wB3TJuXL%8aE=BB;ml?tpn4nlrvqjqZb(ip!owFP7_xHAgZNfq{{mvBAD)~*0
z#QV@Lo6rl{`&O>d_^QCNCA$qWtzz1!I!z%G$_q~rJ0Lnf?U4u$-+a6)^jrDWm+~VM
zkst~ULxi<_9jq(t`tMXs*8uYiZ@vT_?x)V=wo>EjL=^%n7O(aZ;R{kDrpxgWm5Jrd
zR!95q6B>Qj>px|EJ5ZiH#J-lk#|yg4(}7YADaEH0k#ptkGwq2`1LTQ_fXhT0i~>Ic
zw}j#{!weoIp#~vQ4h)jl^@;Zi_N9*J?sCOBbKX-Qvrh&kN|o#DqgHl+f0j+2kfa%k
zJ5j#Fx?>xk9xf2rc8hacKh1w|K@?NB3w44Ok6${G$JqTElDRuS<oSPd&pxa}N+yAq
z1>fuKH0}dv)`K4nQ=BcK)Aj=@WE_*8>M36wxy+>y&BNDFnmK@^Tu=_Jlfs`JdWa{H
zZh3`Qy8K1Sy9+Vfaf@l-DtY;UO*(|4`!nf?d?e2jo%d;dymOwd>}ne+?#ma_CPdqL
z?-OH~m!(C`H(Hh=p9jDO>195y78G-0&B`p-zp^gejS#!st|~gJecq4`BhgT$)~xm(
z&3^0YgW+A2+mRaMStX%(gB_O7KJTgt_hKVLsLH3<fvM3S71LE2T=&G=>(%M}NH5b2
zS70o-UQMn}1UvPExIiJ`RbgctKQ@S@HMaOeW83BP1rJKa&={FF;wBCaoHfql5o<Ee
ziMaL*^-ObwM%ACM5^BPnLA6iLO&c1iIkb88K0~e{JqL}6ev@GP79!h13#7dsPZsmB
zoTp^H-{2w@>nAhEyL#J4E+kE}zq7l@cHvsC^a1vzHiJ9Bg183Fxcun|K<_W@2~w6c
zwa3e3`ETSiZh++bMh9LkLht;OXA6J<WC?3JsA-lDBPNWgwSwt$fFin7!#dr4FXk$(
z;t`svRMEW<G%x!kb9cF%gk#0kMCDYYFQTH_0;6<VglZ<;_oBeZ!oRH-!4f!fJx8ja
z6RcE+j40ML<${7@qc+dqQb=oT*s7X!iZ3Jbcth_i&cYzH!rrWYeQY>gYsgr@k%%-O
zA$s9S`tx5`PeA=5_ZQXctI9l-Hd=`vTcDuxcI5vF3;xO&GJm6y)%gyOw4r`Q$sy<m
zC`sM>-$xR$pvk;2-zZ$<rLdRN>0Ygl6uGCpl#(5e3K$izfE>>kObXGH;)i7Xm>%|m
z?-}_S+-TruVz#V7@mA@ERJxC=;!drkE5xO&A1e?M9HK!rXrc7FjgW*r<{wkroP16r
zz=ai@OQWkslHEAcf{$N`{rW<1I&BgVdjDe}xTLdG+eC+87*S`*b(GzX(6xs)zF{1l
z+JMW-3`NtPGA(PG*KL`3jkt$RS^f1@BZYQ4RBc-vCQxy*6VYH+BmyCL#31}ccR*M^
z2ot%xq2x>eg`Ur_MU6}s{hpnx1AEjcBP;hlC@JNH_kp9be#DbX0M|)IKsfCNN0<W>
zL<kej>qhcb!wuZca_!Adt#5DZic+2?btCU6NJ%O?e=kGqTr7-3M0(%Ea>Nc}NUctj
ztcug)APcK7i^*SP%bcSX@uFhzespY=5g$lZjX(|l#aL3NWEnrnVgjs(Yw30*_K56S
zVa%Jeae`5PnA&%<fa8ho;Mtex&VF!2?N3bLX9u1Z>e^RaXqQ#S$72rz4$D}H8<&~I
z;+U#b^%z%zAIM)&zb+Qhpf(?^<eHpAa5k*drtAJ~08YX-tR4)r5+VF0C8q-4P&kl5
zhwicvRnqre0U(RHB38hp`p81Ny48I!`C7`E5axh{V5f*JDlZ=-%B9;?l@xW8oK|nQ
zrAvB|Mv*nKkn!rPN}FMHNB9e33J*Ytai6p6r1YXfcv5syjT6@8rCef(LhHgNIW$?F
z6hht)STs}8xm^yFQnkP0dK)`cnsMJQK(az_;QX+|Jp;eNTpFQK2d;)gjfwC031!>!
z1QW#hDCH~T$GCn~mzv6|sWJE5z&<ctYCn_=D||PVV((`Ss0^aZk&Y}l_`L!B?EpGl
z>IpNa^ae|tp%f^avzx<;1aa&tL1*dP0*^C3dGt*ID_-J$KP*kr^E573<vlC&f+hHw
z$p}jtSg$;Cz7;h18m8lA(Gn-n<&!lwp_;l|oc9r^Z@e)o3}}eZvO3V`^8JhB6|r9C
z2$2>WbVD~KRf5)&|42wl<d>2P3JDni&FjERu{s$&`d{O^)k=X=y%|s-?I7Mi`YBjg
zmNPUU8Ro7AJ7;y<9(z%DLZSjotm6fscby-7p2zh2kUJy{s5XbK*IbgkXPBVkYt7@h
zW&S6*D;gg*cG_PpabNgPZkKY}wCr(pDHo_d6F_O#BC5E9IsGNvOCSRe2T1QT$Dr<D
zrRxSs1(Ykj8sm+*Fl{vW<vg{6uu-{JQie!w@6zF;VcfZE&<lB16JUOL%)ef6{sV|c
z4Q#>ltt=bp{=GhD=b&ndOEivOeJQ&U-hKmh>d!?Qk%P-^snMx7F6z*+x(0jD9#dxI
zCe<bc4$#p_0Ai125uANsnJU7tvD=;gm^9rS1q?m?Mfn^=9H!Mi)W}{)y^ChDQeegA
zjHhWKcb>t?mx`t_mWv;ly@~fQPi+Bk()m!h;Ip0*8T^aJwq);H!0I>%fd)xwyIdjd
z{cm9Q;IdRQT_;ki`i;c)(EXM;x;hhP7b9P6r1>-Mqez2Q7{?`wxBEb{7gsI4+fH|^
zG8o&)DtiB}Tk4a}&p1xc4|it+&Wf(cOS&-=OP^c%PeNHZjS95WZ3<w(bVIgyU`b5q
zr^<u7L@$+9$1nc7&({swTR;d~aJ8m=813sBjXJus3bsDcIP$w|6$RY!dUdR1d;$cK
zk=a9zW@27yWp$A?^k%fJ(KxNUOc{6A7>as;92&D*I}gdVN6m&g4EwQz#lmY{4C<HN
z6LQI!I_xQbZ;nW%AD|<<&F>~>iYSxt&fUeaX61rrLlR$>=W1-s#C#Ew7~$|3lA5(k
zFWx#=fGspDk1xP)c<;WRYpC%5Zt)em`R^8AjaGr&pMAOgMVlbIm7p~#5PqX6BD^Z1
z7L#YgINQP>*_lzf#?38XIkBB`|LblogFio6$A)pFDpMB0#CDi=edUXy5J6+C<`Z^i
zQy}Q8L5>wk%uIv*EAY2_S%lE)6$o!zlH4P1H2C|Bm!h(xQCZQx-1j{!EhX)>2(1zt
z2&1N{Qmd)*WD}ltoQ9-O+jr@(8=HfH#j%K**DZ2Qi5=NE2{}ARlsx~~x)(UIY~U=0
zbyJJ8i!IKvA$or^<G#wR(7Fpwx#^Tb2S~H+=O~zt+an%bW?61>R(dm%_X)`=6JFgH
z9AcPPQ$tqc3&G7<al=VQC5m2g`!)o!MEkmlYji6+T0^Zej|gJMDcYMI$i51Yl?f2g
z6Agq+!(59{%XgRJlk=7OoA4QeWcm{qve&Dirj5qM%Z7e-RRt5$JyQeA7~kJiZ_1I+
z%i+UHUc{NxK6!>~y~m^lN~hI@w(Pgvbqm_OjlPR}8g#=8B804IIsPeai+vfHu_oWU
z>+ca4w1vE-vQIY8O{B)%nhB_d>H7oc#m<~5oQOIkZRT0x>^O`Qa5R^X(i~KKAL-+w
z$x^jo(e&7^Wm{u5)bZ<0s{K;w2{>?-sp;D%P9Jq+nGiBCf5h;Q)CB;EULYV*UJ~OE
zJXr=4Fypg?j!OKKgK{C6$aAo{oK1*uda}xO{Ng3~%A$}2i%Yj;^_Z?&AOAfEtN<Xq
zGNVeWtW&LCs)h8&|L4DPX^2hq{7@~BqxJo4H&-IsFv4FwN)>}#2ug)Sh&zE3jOzqR
z7#1$j6t7!b3^~T&8v{@O{wq$cV6n<zC6%_GYy=tmLC#Xgr6aEmmP=DU9-%n%&o@fT
zS%kqBP;ou=yTax9qX%LOpC0*_=d;pGLDN`jM4ZYJ|ND35r_zyMy{k$gh`cpJ540tS
zD}`+RNP-JWYS9VM*kUMySld4#fxh6!CF-D`&3r#My(2K*@SpLV(^N1ber|=$`~5qo
z?#5vjJwrfK`aBPk^a4vw#fa+WgGvwdZIBNneC!gJ{Gr9sazQPY-kLh&df}hFYLq&G
z7TwNk7klVC*XqjCAD&+-okOO`91-p{lCZzdwJYspgRb)Jq$ke&b63;C0+pr}dBu|x
z|NOcB`vHIZ*_JNk3!i21bNk=_>i>Jye}?A2KVJBO|Ne$Av=b$;+}*kW4qtouY88Jt
zAHH&y`5{6b8PJ7O!TzPi;mfC=ojLlD*X^I5IYK*p<&Jb1bo1CB{F#TuWC67!h|`y)
z!~fLjE5rK_Sk?GfyepE%g-r61UiH`A8VdBQ*^7Gb?x{W<LM)=qz5*Pjk`bmSx$Ss-
z%KLiV*P@yF7cEXdL6g3%H^^gyH+Z|e;JcZj7V+WQN6|#o=%_Z$$ba|zqVhC*|1Zz4
z62_&c|4YxW=A(zzetRh5w)GCwlFWtuNDG=%GM!;Lsgu2K1Xmj(SCSW&oO!F6YcTIc
zrXV#&Vc%L{q58pt36=WHBDL!cmHxfJC@k(xu7<B0u*q0D<}EK8&I@Jf%q9HfxB&O5
zYGB10nw*xm{*L&xkv)W(@I8ojOF(Z?9P2XJ3+j|UkeAFr#Vs`81NxC-NPsK`sIem(
zDlLar4qtUl77~hO1zoswb5_9_AQTl-?HoPDg~=UM%xYT)$m<V;*wGllT~i1E9ELiL
z=wZpZ*F%q~w<H}VTRzf3%{+%h$dtI;FBO@9LC(EkCV81vBZNBwQ3Vc2q&RDCI2kID
zG5qDkA0=`K+id#13gg<`)|*83i1v<r)}>>a*=3*Ee{ul<3u~X@!Qg_&1^yr6rB@OM
z(C<Ed36U^q4?;rbIId*LevHFDm%hW}IIE?Z{yO~cy}+(5YF;FZa35s^N6U_GV9JCQ
z^|sz2Ao*J}pcMNA03s8F-_7=eQ;ti8`n-GU3(ZGoR)2rpaHz9dD64^8`OH7k#*mHq
zkO+Wz^s9jH_t%3fcK_iDwxIUI6^vSX0t_H!{gzjy!FO<fG7RWzhO(<p8mNzkyQ^hG
zqQ!juMn~g_DSu7~_D7IuYBHDa$WpEB75a5@N97to$~!K?UPm1GmOBM)Ta$d7H6xT1
zHlZ7aQmJhNs^#Hw$vM{a7u|*s;EA58+WwrXNXN0JYCe#ZB$8g*K)#i|vzXOFdS>-V
z`xnVwr3f+A^zy(Pn$GT%x|4>jgevVZ;3wXu`+Xf&m_4F$&3{^VERTgL)L3#-aYw*=
zxt@H-`>knQ5dWyj36ERJP2AhZ$Gx;H@hgFDGVxhW4u+juEa`G-iN#db#G7R>Yfjvo
z(jFqCJu&Vu=~~<DaKuf+RmSev8|68c+_$PXXf~?0l{-h2A!2?-_kOwH5!`gq>|y)&
z$6rW~1m?OsYn+Tz;?wg<+)8+}WXEo5{fuJgrif&7Sl>hWNyS!98(9{*Wf_n$Dc8Ac
zK0UJ}c>2)SGyReeA1>ciJxMZFGmsd$Uy=%>HNOKs8(T`c4^>}-W@lB*P^>3jQE*&l
zX)n!hD{+F=+g0#!*|fV<d~@N_F_PxN?-cNV8?OKRU;i7Bd8jC|S&0DYBu9}<mjA5@
z*vLP#ztF9-0BDJjTZ6m@FLl;H+lUU>nl2nNIEX^%EB7Yp1C|#pN06ZCc(_R~7;#6H
z9Akbn0=?o&7&xK?3xU&6A9SgG&^`5mqH<7N7Fb!;Hjw>x7YWBwjS#y=mV8(e^BA(|
z%1B8CZ+G)}l!1t}1Ok+*=U^EiF|;{wQfOa*slm`T0(+Ezzv_FC>kiz0^NHzWLt7qT
zgf?(E9_)GbG`Z`4ZUGkmw-#WsjlnDv5Q%mNv66|d7N=|Uz$v%kMIl9UXBFC}iJI%p
zC-#9*^jUNn;tK3RmX;X>X|f+QCOhHRx-Noqn2$g#k3|q&vj~%xFgRhAhSyMK4T`=&
zVjvgY!UY{NTmcC0gN-zesKX$vGfE))sINFeq>ffpr;gNpbInTJ=<)|RXX!+AtaR<C
zA%}KR(znc#O?||foM8iD$RSAD=n@OQkVvt+A|!QJZbsF$ACd@1I<(DVqV09Jy~bN0
zW^<Uw2NR?NiAP^V(zkS|5v8B4s0bxsOhsR44yOZjJrCmw(y;WzaRS^fG3X>qn`pc0
zQ#ObNd8oADd5y(XcQ*1jvnc898IXKi^(L}babYjh$bMar$xB2pD0}C9fD}@(&c16@
z`lACrI?uWv@#GfVTIugYA_AqEV@5mOjt?cXRx9j5PUL$S_SO0>KwRb&d62;Zy7Unu
zDF3i9=5oj~uFVyTDfKB>u(A;W`qVtQ5UeeDBQk^qA&-p;Iu-i@z5%0n(ZA}-zb#$T
zXrTBFHIbJ6vlx6RNPl)oDc7JBe{v9UkMd;n^+DQ}b%;kRhRmZ+(tlt+KlV}v(TANs
zcKL~}$#Y(OmpM?a<Y!X7IMBb#)OG*_V`z}@n=c-70Wl2QEhf@f!YF%6uq+?J{2J_q
zLQgO<qRv2q;!1(;@(s+x0bm0lDFuTij&sv9hTw;w;7R05$ygHji{h{-Di-r1@Q$)u
z_IV06n!w4ihlgnI5G!V3=-VlEt%oYb#H4&>z=!J)M@i5xoUv<tPKK#UPtY1^AJC|b
zg$U_c#^c}(=N^h<n{WV1R{Y-Xb}C8~a~8c1-g)OH?8~2@-9U`DOP%k5c2J=l<Ew!&
zzyVzn$!p}`xKa#|N9jrKw(L>m<{%bHf|$KT1Eb3cq32a_(@*j2Gd@NlM5BqG%7AZR
z4mI~Jzc^JStuZc4hZ_C1?zd!#2N1;N=i2F&4I82;O2{e4P}VqcQ*WvOj8Q%uyXjwP
z5!ne&U~dhdp)X|^qL<C1?NJXgPj?$_-A8y{AgN2X;(Tm=Yw$DURUVNa(0ucNCWrEU
zEYi$Ij$=i!GQ=gy-9~Cqz_N|SHp)v`Cl{?{k=sTSd_(rY9}Spe-TsIThNvUroMI&&
zJHQi$$x_*>H{2Uu4|+#c2!~CJv{s?>k}*d|$OA<Jzv}FHdhLb&(wOlka1of_rq9!m
zH_^=+RJ77wAV%Rs3u@fK&(FQloU}fS^&?mMuFBebmD#jPS2fopi5i5yxefay2K)Z$
zvwk!5JF9`V`=9rcccsrMp@<6U7;)3vK#JXUXSQ2~i+)1z-sIpwk4sVQ^-{wo@K57V
zP(bg-0JCl$$am{ng__Zh-4+I2_hGMY7UyiHIN|%vrbiGPi<Ldij7Fz$k6-qGdQYEK
z(mt@pK=hY(P--wh{FtvC&+!P&54A*x`aR)z>7F=0fS<d@bYOXtX;Sq_i#g1ov|mPs
zpXCbFG&O#d<;vvZE4PC_bFby9_WJNCgk{P==2eydRPZu=%tD;6$vLu2#33wu4-RYc
zh|3AJM9P6UbJ?$<8Z0Bj2(Qrgi`SD=PC*=1O%m*ZsNHjfh7b)Y8u>rSBDz?)K3~_^
z7c(L&9~-5qBX_;f@giwW{>rI{Kh95SXvCrfN{=#~doFi{JVNtL0hh5+LbB9)V}<E3
z;siGAjnX(r)*!)D@>K3gMEg@;x!vd%z01m0-EL%^&eU?`WDzx@9*VpDuz>EngL?WL
zS<A+Kvd<fw7tQukgt0^ASleskUpZ!8pt8>q7ALk%&qZf#)0;p@Xqlg{5d&G*V`~qd
zC9O8lb9xLAd04Eo;*_&TjLK_jB{UC4W!SYCq?J|@-3#=cZk>+2!P+X4w99lpeGlfg
zV-wN1>y50&=Nf9UH@4jP?dQ^7ubpVrCL9mX$Sk%PZkMv>j4kL)k7L?;W;TK4xD~_9
zs{#}t_IjC>_+CENCEBB|DdXzkY@3K8V={G+wg2c<7KoB%AUkxcuX7_RwK)x{>dyzM
z^o}EDfEG99MrS*V#~<NtW-XmbEA834be8M$4O8Q`+RBDN-)C-7E+3|atzvI{iH(2O
z?d)??*y~?nt-o)U>zK~w{)YL1n4CWQD~e(zu?#mYJ%hxuFI=;tT;>kt0Kon*Wf@m!
z*o?$5nVZ;~OK$^lzY~Y~*2tjtkhx3cTIl)sU?N>KN4Qf;^jP_G<4QKPrql(7yz9WV
zDo9QAqCx)_5%$j6aT?E8i3T&GX{l;yPO(>K4FY3;ca@mDudmZ3!*QJ}S7pCGt0iLG
z(4SWL9N|9Y7+JZ6f)u0@-oQeBg?G^lGf7I%_S{Eu^e=Iueht6f()rq5a(IR<i<5A8
zYTJ~y_pZn?Ak8l@1|iP8Z78Ao<3GFro>74i?=4lLjI{=JdSp$a8hFkNJ{xv5jgA_R
zm2i{(RQ}X;=%x2B3MF%d0w1X>taf3zMp-tq{_P!Suv1_U<=_lkYmwv3_cnRn{q}Yg
ztL7JJ&KYcaw()u-*sO)#WZY7+p$>KSe=Mx(X-9Yx*~S=W)~~-SsNm9!j8)x5L&G^V
zxpLnHdi%X-cxD@7yO&;8MlGeqYL_pO<(&eLi(<>V?uT#R-&#<eRpN)s%sW2;^JhEr
zD&w$J+Ku>M^S4ano$`q+(6ggJ2a3hYO0O^(z4A_8@htHsq1wXYeh&)!a6cV5JihOj
zedkv0gP)smKv))#sfH;?45Jy#Ked-(nd{!`r=;?R<{r(H-~``X`)?0O8l7zVmSgD?
zjKy9?=&u#x<D|PR;$K-=*O$M!PFW_Ht#^C}u@JD>q`dpZKOcL&`pe+QM^@^Ksl~?e
z@eH_?{6&1A_t5ixkA1fVJ6X?5+44B~hESW*RSgdEH!b=7-0@zPI=PmrR+@9nxTE#$
z26>z9SlVRbJw!eoeG*|v;mjxlE?+}&I2h7?U3S}BEBdf$S9rYKqEV_Kd6$9xdMN~#
z-`Add*zFWrz+u(5HrI1G+r8|WaR*LCsA`_~>Y^rbebNM5ZkGtAP*v{FhxA2kg2{pT
zBG4p{tBp`AqVr|>BO>TH675UlR*%u}S!<1F7g(Y8c<3%v-qUfHr-Eucb1I_upK-8n
z=!G-%XHU(p|M0g^Kc42LbanUu?j#H_2zVzUlme!Jn9jEta~fDYaphZ-LF5{mLx-(o
zs-B3jm#(PT<`Zf2CL>wh&wz0d5v*SD=YDCA-@ir9Ktwzo3JCP5C9Fn&Q*@lE3#Tjv
z%aX)73g1CYBrBtO@Svh6mJ@O1PDCQONUF=Xg4VMk3e{DYfZD6P5=#!YzBfe`)fHn0
z$Srlt^Pr=5S<*hAMVzkMIHH^}$T4rk5&V$I8NGIssr@s)E!<+iab4M2!`l#HL2*Sb
z#mYdXL>iYX><l3_gyv_@A}S2d<%W2tAYXRWKYiJ%HIc2zeKrx`^?_dc9oxbK;xS#R
zAm(myQehrjWz-&~gM-f-ciC~a%Q#bo_UApYAO_@9$BYOp-&uu>C^fOPv!5f&S7j!d
zPdxKt2fQSeN(ZLIt<U^07cxjb4nW1M#X#$8f-&U>?A_{uE?5}$&yo`t2duzs*7;*K
zsQU7sc0SfhI4Sf-aAm;H*z3wwd=N*pt#cEMw&%6U(Z&L8=SNxmjnEo^S!4$cYv0Vc
zw#eM<3L?Ula*pQ@#jdhtz4h`#tFii-L1j^z&5PPIO~P;VK~)q;lz}ujhWW$`>5KKC
zL0G5Ve{)0aY$BXP><2xTJ5q7!ms*MEbZeO{^bm(tj{cC7YP3CVDU}9Li2Aw%k;R3>
zHj=K6gGtUFP8|y!^7ow}U?Ht~Z6>m_SRt+TWw~IFadrnLpDLTgQ(1W(LPY@uP_bQ)
z5p<~2CI4t>fbIP$^2UQK-P1eAjsL8oFJu$dQS3-#Q;&NRa|pQaY($?N)6QZi@_<kO
zoSci>FzAYM`C;!{P_;4`C_78|UJ?P9lgP-`Aun}4?h9G0d`uq)z(`$*fyDbY`7>w~
zORC>|bFm6${hj{vcdnG?Mb8g#@oj8&gETMNZ&Y2yUQqfZlX#VUyoo&{$_KOeg}+wS
z>~@kRt$zu$`#8pUD*@^4?CSgADlUS3U2U^A`!x`ItB&nex$yp!qQ}$GPs;QUNaR7V
zC!P^r`k{9{4ZKx6mdDdpBiStWA(LIR0Bo^SC_bO8cb!O2jg5R6cQ2j~(~8KKf^AZl
zkMATa-T6+$`}aosuinBu1PMl{GphP?{b`*(9Kxoy0nCHacbGUSsbNnb%9ft}H~>_7
zbuLG5Lg+8!&|5qoU1jB#mWY4t12MBp@f;)lfn4>EXN)E2zLBBm7i2eysHb~?3Orwt
z{&=F&|LUR>XqjsA>jgQgU+9stUO!A;$@Ydjt_!gNUZwp+{sK4zj+CG_D1~72Kq8yX
zlbILoUQU$2HP^v{Zll?SZoGBtw3jwW!JTzFG34#odUC5{v{-%}=xJK(G=`0#PY(3T
z+9V1@7mt(eAT~zIaPKA{BL>^B?n0{x?<+WxQ0M$VmuaLGTGZ<HD4Egk4Xkk?2*37V
zWVbD38&mHq3*hQP25mu=W8hZ`78H!?%@LXnf{K<&U8PKiucD^50}LzBa~2EIr=7--
zu=2fKMozO^L1s131@qfF-<yj9WN+CX7VP{d>vMA@dLIoL2aGJvj2!sevtV#P--&im
zgM4D2j^fumBc7LO51tSkOU~}i9wcG0d`-{tlu_alBWs@VRW(ecxnCD=o&*nT75iz4
zti#4vjwYPDUS%L<+_S$OW7_1=Ea2KCe6zW9T~CC7_34X&|27+h^0!|PY?u8QAS8(a
z9?E(${ksca%R8c#A<@HocdJyFX?dI%2N5zAk+L#N?&7QE3l0bm7e%gi-ILOrHuTBW
zb5vvAR`#x$$KNC9*axN}zZZ-r>nLtk%>)eQs%L8KIU#B%MURpXIbSpmWFxb6A9f@w
zbvW$Sn)r6PhVkjuO(Zz9KsdTyZ{@?8KFJ}0c1>TcyWTb>3UdL+C5Kc=3a^S)ipe>!
zwX+Q+XpzgrKR41<5LGjK$6gkCTuN^Q*WD~%M>3?#hz<Hh%6eG(CiIm`&%eS1o<=n0
zod?L0o#}(C6G74X8P2OPE%sOh)Ny_J=3_5knx45&xjAd)G$@k;i?N3TX@RjL7whOb
zxB$RQa~-MY2Mm+;!FsF`9S;Z3_)<wBhi*W)E#gq<=mNn`w?a`x#0MI5ZA*EZ+0`Ke
z_<k{%R8x^BJTA%j%GbHXBDu({v7E##6w+5&gL$nSE5=m+7Ob}j$>Y_4Fh{-XPp5r8
z2l!GVytT_)!0;7o*X~*E(B+s0*GSuX@IsFk=me#?IT$;wh$(yTT?}b^+iRFkz_vxW
zx(Qf4&&hsy(_HetZqB4qr$#razUG|0+B&Iy0pJ~h)#^qAyc2~sQ>nYaG-@Wf%d0r{
zk@K=eUwi%-#GE$ia;Rx|UKKtBp=c*}XU#;PVNN&7+Z{vioR4<Ka?DN`P>CvxoTAM2
z|EYJ-FGQ9IiSqV46}IoC+bvi^9&maYuwWJ}mszQ((xexQfcE90{7__(&4MlPnS>5{
z%Nvt>0|HiMc<hEH7QE%kiBL!aqBml_tqfg5>7qmeqi-4}gxKM+V1fFiy6+9Bm(Ym0
zoIy)U3C0?f+L0d?BCT>xfzKzCyY?+ZPY8h9!d>kPl4YL~K7%-a1`=@qIB}<7xsFY;
zF6;sQdFe$=vy7$5#+1&gwi^o9T?XGZ?to-_`V56wJ|vBZSm)y3c#BBAN?DT%h+o6t
z?hw!E04p{D@$C+B5-4PAqxXLbvliI`pvznXGWD^l{I?b>;yWX|?PXi{z*Kx}%mavR
z5!;CMl~fDHW;us>eEB+~oo}_u2bd*XJ9Oq=fSDc4J8$N2y#11`d)Y8`x?Pbzgj&W@
z7Y63|-ScWG<rio6hR-Vi=VRR@o>TcrJ>u@AdH#g{@j*S1PHKB*0`dW_r1-OJ1y_M?
zKdf3iiLo<c(VSm{K)2;+a(8<WMV)%Q?C2cq?L7z42t~I!Pw`oYPjGPF#7c*!5ZJO_
zizV&VGQ2uT^d4Es_z77%FdK2+X;oW~nsE+c-S|?HTuf(e0)dFyR!KL!xrnn_V<Sn6
zN8-733>go%VVxm^fXngxc1R4cE1nx{By)i_RDYPC2DXs9UZIWUQBn(#twt29G}s#?
z6R^Fhk%dfmGGMvN!g3wVZFNuDq&Z51x%4tsSNJj_SK@f{@*)xt*l%E(`6f(EI==WO
zcZr6Tvep8IHnc`qKi4b-n>^06D(T=;XgK3-eSya3cKn1PzRWV#XvwAP|EGR+L}s}!
zF7keN@URrS(*Fp<5^$1jXH=La2@dP7iF(}_?4|S4uTJCy^0Rs!asyig2Ap?L?Ue)(
zT(k~gX}hZk{Rw@}dz)o8ok+b{+Oq<tSQ~fIbu4np68Bkitr$~OoMn7tgTPB^DjItR
z0&rM@?&N_-M1Rz0FnSyF7Mup!lVnay_2ak0X9Lw#RQ3=oY(_;2IkB)o8&@5sOOO$`
zPm+8I-zZR2Kjo9F`m!T;(Sn=2oAQ}uJVsejCF3m}`g|R=iTIK7O(3w{inK!{lhzUw
zjI5y#<V;fIstc?-bNV&f{7w9Y`IN!K<kDLRy1n8NQz<`u&ekD03B2j9Be0fj*oO40
z3{DhZ1|CY*Hj-j7H?-R0Jqxy|W6}u`cM!yxUPusv;J7!Jog*}YO6$0joh=M%Q^tm5
zkAyHU7%#6;bSK@mc`^q<nj5&Yfe|@@&Z-83<N$sgRS&~Hi*X+Iz2==}v63$ELK4Iw
zb_SN|$#1l*Uik2;+=p9!u=GA9HEC#Ymx{f(RaTuNPp%bE=VGW>XS7C=Rtd11KcP)#
zx&tuo8rnF@=ghuWD|BFMCR>i<(sE%<$-j&jHu`HtK_|GfazA8U(mSqGC#z5fLJ4Ua
zS1+92qIruBt8>kol_BUj7DA=8B|tZ84L(!x&K(*<o;RIXHD^<~V7ha+qp<i;JGSEP
z%<##-y`p~*l^=E=*OF_G>^%N2p^$^bAJ`B9GPey%B$*K4M<FuQ#>l5&W+4LC*lgkl
z;4GzRGLhdIFKvi8+#JEr7!1jT-gOZ^JZAkB(Y_hczDd1EBBR`qNdu2vhi{Z6JV`OA
z%g@{2zAyQWvNS$c8?k<VfR+p+uq#(Ec%jV*@7z%~50oV&mHgIN)Gy@`SLoR|rv4z}
zpNB%JgN_Rm(^RMG1|qw<x@UCZ&R|<yQ3`VASYKb;`#thB6MTL5zBqsO+-91+Gyc3W
z;+INoDMf^vhWn9Ah?T+fJ$FZ(tyGP7ixFq5+p1+-yAXULKp-D`_9;a!Yq+Mr5;G5F
z=iPC;x^K=_d_lT~@g^kKrV!E+jE@)g!BshC_HA2|V+4P@zt3#HL@*N1;>a{lnDgyq
zSQ~HW489r57=x18wcCmIy%UYoS=k(Bw2>KqUf)wHDm6t{=mGOlcRe|x>8}<F;V0XA
zvT#fFT{$R)7~iTl^I31xOOfCc?&iQIEKY#zk-%yrkyjJm<knV@gVFP2kzbsjJRDE)
zmSFGJ%{~gG;c#wSi-@~A426`Hw?AWp1j!0+)?uDY@<T3&Z3#f=y%9yyX2w!xwA8y>
z76$!zH_JJekOr$1#F1%d=L^=4`FUQ42-*AU{QNbxn2Gox%Z^B$*GJ}B1&v~jV{r5K
zOr}S_9xQ5KVaT3H8@GktY1bu}S0w56%gC<FrlAGx+?vhGWUqwC9&8n7^ETiMDBT=u
zDOdeCUastLb0%;vIpi7Vw3=ysmJvYxMN#-0{N#!y9B#>vgP#6?!qP&n(0?>cUbQ<z
zboRCR@qoc*{Ie23&C8wsOd&MCSLVH<Vr)WqJmqQ+%w<86N4jqYa|O<_U^c9d-A+9E
zqdfe3aJE1WIx3y<cdcyS8&!RTr>|LY&4j+i`mk1)>Ov@&^FTpBbCXyGU)0*?fX663
zI6j-iX16>M6Kh}rcj6EAs9<+^^<3v<fps8kGXFQ9>#x@2!4*EhhrVL@f2h}Aoj|eB
zB$kV_Vy)oT>;>(__(IO{D8$jK^5q?PU$1CUEck-*k;h4kUrnn0P(Ic(5B`8_e5@j^
zCQ1+-Ycww)S%j){e`XQtt;2L;Qhf1o%PKLIz75Mj%JiZ2f2KZtgGBZaLdOTr#H#ND
z7|N?xjA-6F%#4pW$1PkVLvMO3MWgrSW3gXhKy~+GLhijkYknQdUs1A%<H0vlfs<S_
zhj7gblJ^7N2zTnIFO~i@>E+6+uSklO^<z&k)B1KmQ3@YG92UU%r*>vThAn#(jD14F
z6o3C-j|m({dbmC$`wL^xEbq3pVh816W8?VGkGEeB{~x`<&mYjgJuCRDk}j+e%lfeH
zAAGC7?>PVVU(?~^g%Mw8{rR`>;}rgnjnRh4GP{?P9e?1yKQ_eQzR-Ux@++*!9`X#7
z`JeyFzdmvyBj7e~@2CHq;_x5u^vA{uAcl3Fej^F}fANlQBYd5UhL(R-KJqeWpPow{
z0MW;|hkSQ9t4+Vkmh!;^?AgEmI-BvkQaTsK0VMk}*9=Nj#pJS-jHKk1#Pkh2!=JtS
z6*Rm>X|>vaQAp+wOQ|hFgEUbHi@gJfYZPFZb8m<h?rDq`_yWHCeekc@H#+SX-j4>!
z^07U<k-p2)byeZY*@wS4hW?WLGz*%6CK5k!3Up?nWjia)+pgR5Tc-qd8VM5!%l!hl
zEKp}yg=EJjPAg*DO+<(1%;<i06SuS?j5-Sj8ISMJPX5XjgEyGtTE@PNk%RO5rDJTl
z_+bTsG4ge2K{xyci)X-<$PTKq+Xar&7wC^Ir)f&dyr%<i**g`FkJ=+actT)?K+I=>
zWLb^Y2JYdjmqB6S1l|rTT4i~PA>IDfHdO1Gwg<Y@OGYk3VRhMQzPDiuVa;9!8uvZh
z)+1K3u6-I*Jp~pcac0tDYdYun;ppoHUhB@qR_7Od>Y|pP(l}>w)?yMkd+JJmvg%$f
zAs-Lh(ffT0)?$pp?=08b)P<b40cC=GM-9SJ0(s<@i;(wRWCW$Z4aAer!|d4(93-ql
z9HnXKjrzb>Qm^L48MYF$?%j9rVDR5ijcW%zjFaCYez{0ymkkV3bD(gfSdFKo3hnmS
z>5UT9W&d@SwJ%Q}hP!Uwh?8#*psq;6kL`=%spxJsC~CikYG=F=)>&CChqzK9xrHBY
zo^Jk}GA(JsmDyg}>YSRU-jhc4lV=!-Y}}Nw=DzRe#YN8tL^eoH+JX1ut?@`y2gz}2
zRX`;*(C_>p^FEp4abCB3t?Iy6v$(lZx|io_0a8Dt4_9nd%wTS>I}epm(wW96Sa0(i
zQj4jD3*h7_S2Qe`nceN6+ZSrSOlTH$vMt-Z6}(yoTj0|#8n_W1ysFw)4EtZZ=|9c~
zWV1iKg`g$erw&Jck9kcqI{+RYPSB_sjsn{G7zS-&w-`6rI&&g3k+-1)fUo&|I12ab
zu}9<eM*I!-&gjaBF1u>b2^7M&;vf4(<|kXv{0V$3j}O~@hmBhr4zpB7mG4-*2g}y?
zRg}67ZUCh1X>CIk<?<#gS>1ULP_C`q3@#yz_=}b-s+P)>TwmqLt3>;DR`qbgD*et<
z;!7{<CGXDOxs%&+?6w1`erjBIJXdmxj5l$-;m}_E?RYoobt<z^#C`Ps$zD(~_8=-K
z+_iSnJQim4R{xgR4L6xFU|x|pq<hE=EBRg9iEW?vajluLNTXY_s?@33IzOc|X5n|!
zHC~Zk)IDyatj;?wN4-_+_ha3*EZ&Y@Gb1$WPUa?@)TrCX971{?Kkn{o!S!9f)53=-
z6hE@c#>lwJu%b(&=KsPf^q#--wgX+Cs)-6qRjzbJPOEXc;}n&3Li$_=jJmq1yU&a(
z-=L{x>XOK_^YzyZ()TV2o;}~o(RQNl@yP4vMe4YWOnT~U)4rxs2>(0Z?B{GKwi%G>
z(mvku0Ye#|TijDUEs0;SOzlpC2t&=dkX(@|s^yUd-Z#`SB)#P)4<567ucH4@)h5o~
z-$CD8HcfZ#%W9QYbx-2Vu$3HtYW8%Cz^)ka?VKv1Ywr%}{^WV4*}_tocHJ=-{&qEA
z@FW6=Tl}(d<>7rJS0JJ|02Wq8EpRt!{dEx(q3<E$uoy~*R)k?<10Jm+JToa#Hq-3T
zq!a;{aQp-^jPFC(4746QcI#l6Tnx{tXBSlVchNQo?*}+du;&RP)@%)f3-0Se`$M}R
zTy_9~=^aFkZm?U&556{iKtPOkQrLBd<>a;p@a`BANC@=OQN;5Ou}-3^yOUkmahKZq
z^+TcQw8ZGTkgdQnbiAiXSu+^>`6NaI+h@SGGbg48#MHeol$b=!1)&%Vt{H~(*5%u~
zVBhquN^=3@zz%4I@q3OygakTQ!Kg@@r_7ce(5W}<{4amAhn{;n!8mv5<G{fFo|Elw
zkmwTzbAr8bCdJs<Yov%xdk>iLBEpn~mpYc|SW0OJjYv&t(aL?07_1jz*SbvlK*hX(
z=<)n1ELWiXaRP^vJW~4|0IWBcXv{s^!DGMinX!G3&eYD6&cxTP2*D{X7-2*8!uL3b
zG~7kt?eh-)G!T$)beaQ6?{L#aMx^Otj_G`c@V*pozdSq&8^k^`EHd%*Gldwzo2ctt
zRH+3F4;$fO&TA_~a8;E8C^(Op0Dg0%12}1iCaks)F3E!nX@WeGJWP=z#G1u-hQ#Q8
z*R={6uLdMwwg}PMk#wle`KG+h<k5P1)xIE8<)FvXL^wVn_D(rUVJP$_6-U`df=A9h
z$hzrSgCT|Ly~eD2?ue7M!dn^@kV7HbU#er=Juo(^(FZ%kQ-NvLI|#iLXf8S_a`Os|
zGjGQW#pp9d>8Ef`5P?cD_o2?x4L9UXwR4lWk9{dTwz=+0zD_f3HicQV-D6#t_R1O*
zQD@TGxu&g@-h$^l4@>9BWG^hxi8(2&r?DvG_~O*{SKz1By*o_xVR0s`(!;yeV~;?$
z4rZ6mWw!XG$8q*9Z)ai#5ubwvL+^g+f|Qw3R!k|?q(D~THn{TJAq<UOdlv|kS)Q3d
zO!KuXtTq)+bXQRR<O-RNez2Q*`Cm*C!E={f*Jf)gIRkwkpCj0iA*9Yg6e*(jx}*cd
z&7D1XrBh!eCu@nw0)6kCBO|>A*Nn>wte}b2HW3#|rW;S>-jU6Lb64OC`Bzl%iCjeX
znMF6HO^0LgGl<%&)=NZrqAA9Wx<U5&K&RcjEr4aX&K}8yS|Y8otohuvsL8sb9wL--
zoQT^wGV5z=53xG%6x_;CrxK9hN1OvBAnvVy2xEI&E5Rw9N?2}4RJaG|vILVcL6pq&
z8YJwQ{*`|Sra!pOd0-!w(J?tLt@8xlp#6>W^E)@l_Q7>h)w2}QWz>5e)?UB7<YT5g
zD>4-AOMRb?7CZ+K5qE#J+$fcH*Z3rmE!O!Md%D5QAFb`>(%PlbN=jBBVU?BlctIR0
zFSl*^tWyyu`M8WEO^4srdI`Jl)mEEMywN9X$M6-mZk}M>hDcn7uPWjG!_Ji_BqfYl
zqqyj;?m<Ply%zb7m+M#C6`1uru4f?Imq1*gl5lA_W3f>4#les6v6f|)HZw?!)U|PM
zR(t(!p@q`**bn|F0|qklYfNf>3Sf<34z2l<m%y->;=5&>E=5RFiWK#5H#I)hTu$T`
zh$`OUoeu8ceC|J<h2kMm50r6N?NL8{Au3bfVjjVoj=uPkrJc{!%t@wrjcIL-aMda+
zrxVY8HM0msmbZ-yP_y7Ae9Jj3xLym9q_%)K2BPrGHYTxF3}n8W0%>oUVy#-PvuPK}
z2aI>7Ebg?~=eC)3foXTF_WBA<yGwlB_!Q`5&~vcwV;5L6v(OO_ICK5IxlU)&S>QXd
zKn4x;J;+ch`Xmb1y*iSCSjEN$qTVO|`rGdkYc+>rz55O28Vp+A^3<%%wsdE6iVi;b
zZKpfaLDMQWmZ0-~l|4Ab9Znpx*a{sw-eWF@zHDWiG*XoQ_NmFi`aq)`Z{hva+Fq|S
zKf7qG4s5YG{e??O4T2$HF%_$HZ4&7h8PFAmGzUMB7UT7$4{7)kxBrU>aKAlh_QBCa
z0P(4Un2t>6w*!}R`3-o^=3U9d(1#P92xWqt#D@lSrK_AfU`0}&F&($RV`kcKi{Gfg
z7TnwlRcD!ar5EqnHT7GW1P*Ob8O*;2L5fc_fbLm~O&=@d2;p70eNe!MYs{v8PEuue
z{A!vt5|jwJi&R-ztX1WBQryF|N(yPmV27(u4<)Zv;K{b^0+cGSdz+MLeQ3+tCX631
zhp~vE@zLiN^xN6D+|z7)i5buFl1AY?vA!#&s&8Bi*`zNp=Vjw|31YH@$_Pn&7^eo}
z^i~KWPFc&unh{2{ndqty<0smld9@!Rpmi46(fG6m7}^rHSn2&L9|;;-eKfTBEvh^x
z!{}=?QehaL{4IOtv^`3sLhtCriXO?^7)*givE_tMhl4Oj<D*;evOUV~LrN%4j9u6j
zBLk)hirhCz=92Pk!gPJKo+Nj_Lx$DpuIgZ-IUTD?e4CmLb<p(1)E7McR-Wei7cjBW
zmyE;=@(VO%c3A8NVPFy9WH3Uvr!@@opr&o{-w4%{ooso0xqe5>f+w|Iz;DcXDBi1o
zoSutoA#bnDW#F@`KaRH1O7dd=`DGii?`{9z_s5LT$~g#F^PK+kAU%F<>r#$??J^~W
z;_@i-=Sc6m22)x7c>e)AQm%+#0U`ADDXJjvyj6qWD(qVME`m@XjzR`4F8w;$e~eP6
zKq*hoUEthrxT7JHlEV-5@{Cf%dIIlU@kD*gOx)l%R1gj7PQt4klxL-`L{F!BYiej1
zzq#aVpmnZU<$0*goaFJHPuyBqZ7Drb(!1Vv%WBNFI>||Q$9;(S&}2w*{K3q(O3`J0
z<7Po1l+$uGdg?uPLCeQl!}FT2<^@xmk9RF%57U(E7{$L9)+JHZUgSd}71m!e=0**b
zI1fc|o4Nd&B;ofwBC<QD?k4g)(I4X^R6RE94IyM3nxy`mph~5ZSKXUNMmOAc!G55=
z*U*?<_{-g@v1Ps<7~RVi8w*)v^^A3#Ao2jG^J^Uh(S8M7dhmmH|Nb1yx#bSHe^rb(
z9s0cM60*`ypIdpgL37dtr&f6BQ$PIPs&?m$zI^im-$q^!S$Qb^SyuZ)el*m7^Qaz<
z%R{F@przCKu03Osx<Xe?;d$xud5n3^S$WSOF;CH$;2IXhy-Uqp7&VO`>4Z_=>+!!n
zOXh*R#rK+W)~VaFEwB8?m+RV;SS3wgy>=faqjDq9FpMP|+hv!IR)6MC7+XnHijt+^
znc6!0QVM-_w1Xq>QTaJO4c0=stBdO}CcMQfPfyF+w0!cEBJrHI)S_nNj*^xN>J`t2
zr;Mn3!N5RpVdn6z@FDddwae&I^gQ3blq$NL-j$sp$Q-3vtBPB3MXE|EGjF<=sjR4@
z4jwp<3n3tFbQy>Pi8qe)8Jf|mPdUaoX)73in<9fFdne^}rll62+4~HMMdode5T1RW
zL+tp)a{M`q22$U*1+g_%9})de(M{pt6&f`aAyZ~zf`N>TNgsO=L?j|kD<pi%R=&<J
zBkOL9tCWtl1y;MtJ3l_3WFs!U9;~DTq^PZt11M&D!pO~n+;zs%U_bC(SXb-F1B;0`
z#5;iPL;NFX2YK!(HvPo${fZkPRho1o5^UeQRvc052JEM=%MGhh13T5OMZteurCljk
zG-h{8vN?jVqFx4(A5b4!I4v{w<OqgIIgZiPlI0Ual$SiXR@brVe0v}axmWk=S=3vS
zhOb_*gM)=CgFTpoR0}P{^HQ@|WjpZYU~czj-skJoo38^--8kq7)i%GaYVkQ0l+sNr
zaPN4&KAA3HIGMT_ye#EVeaejdovg!G^ynu(N3h6FgDIy04|xX)m#sogjowK=xO_N$
zF}p;g^OH%W*kjo{yogt<WmZn|m|<U+qx!}Aacj-c!+Yc}%d~>BF{r)PNIJ9=`0y)~
zp&|)LDPFYj?j1I3ZW#?x!`~kP8<SpGxSjHe9g9*+mR%sVt<Jn8YM_^XaN!$0N+!r`
z5Jt5W7g;~X?Kvm(lF!-zV|UCz2KSRI&3uT6QgwfJgyydbkKqPgcpswH0_jD*4wx`G
z6~TD;J&<Mjf^%z1eIG`9uQ-T5KJsZa%IP#Rj(LwLNFpY``NlR8Q8$S=3K3#QXHD|+
z{m17KHOEaZs`1&Cg@FO>lL}@-gu6{4j$Huh43tqBILA*OoUVb}e)yQ+<XI;$ZlSuE
zNtn*`CSBb<ui=cS)}SA8qNFX>;`C_3|5~!Y{bAo8!DU_Ni_J7PkqC2k{Sy9d{sk~2
zjdI9oeROwAd6Jfi@6LEcoxaOdviDQdWxD%|S3BQ3*_sXA?*qG!fKQ6ud9TaL41fdL
z0l70t=hDYRyGhAR1PS-18Y<!t%VO@^mpGi8i0qrNI|791kWYvsYpJjskQ^egb814O
zat-4vi2ON!x=}bF#xh!dabn^s|2`B1U(W#hJ~rFeM%?jpHtyBE4?ANV&t9HiPx5^g
zjjO+#DCzHL>EnIS0qOe8%!f~Y5HX)0fV6mcC0=KbY-Xg6$iEccRG#Y6q$AqU6E_3N
zIr#zgO_eoceEGx6&n+}9URyUg7LG?+ZGD4aE?8iB=%8etd78cZ=3t{TS&CVEM90o0
zqcHBFFmBR(1uN&y9Q~N(0)t5$0T^F`ef8L{TNBYJ_WbU^djZuY`Zi~n?m4>bMGZEK
z)(3YxK-KB`l1WfrFeXgXhI*Oxwcp)osgY79Y(+sI7@Gt|hJ4~yP8I@cTnvuVfpQv#
zytJUXhm(Kgg~-;V+*dInrjx4+r@-OiqFQBV_PQU^quq!|B7-wps)Q8xUl`C|D6WzZ
z5oUwo2d!1fgJ^UrGcurT*)%D@f!K^6m}Q>7j0R?MVs}6T$fWY5+F06Pb=HAR#k}LO
zia>uL`3uGVXt|9pr9zSD*LCBMz1zJD7pqkjM2UIJ)JSiYh&XR>2VRe#=Gr8VV4(zR
zzB*HG%Zu5Iosab6-*_*}&Pkr8-D?e}Bi^?@kd{F2Mxjywg=T+_{ic_DFW94a2k^-3
zffd*NZ8<;9J6U6ARPFPXqzsqTq=v;lNj&%r-kf7|(zKG)+`K9J$sgYc^r)M-Y{*)<
zPi`{zxPgwsZNoa-g4?XgZ}g*@0XN0lZi#|^hU9Up^XH#CyzdoUTwEUrL^WT+3ALA~
zvRIS~j|PaJBl9ss7-%F@AczSGjl1_s5a5JDDFdBnjMyED#EPGJtEPGVLhr^FU(Ne{
zo^tETE1i!6h<>Q*NL3~8Ryfk?gkd(g{S7HpmH7=(>@Nhy!zJuOM7r4;`e|Kfhbm6U
z&>(l+1*865ED7O(mfG1@O3^~|?-EJHBk%3Q<p8x<kb-82CVb0%uhgyv=a3qsR`bK(
zRbiR0kE7~+K3upr5RmKAKz(xs+&wm5f2xu=ITV?6JAL=C*jd72nEG?Cgg5zmdE+2K
zB180`+Bd(1)1jxOFYy5GEAevJxPSJ3R9lYT$G8`aZ`-xV$?ClyPZ-cGXZAt76Hn0D
zqVU(M#`>%A0fP$w2|R7!zOe#gFUs35a4duakqO=^B45KDLw`q(-RI|#rx9+Z$idUT
zm3!yY*7=AD6;Y2ywkWC^wiVm-KOY0w#yQ=C0wh6Lm?Ws<+L$ny%k8;|88`Ks_TNr~
zl<`dScP_%9*PZ3G6#^u{%!Wd{rk!}N6}kw^lZSq`z>5&h#9YhgJKxS_tq|0txc%{+
zv8Sn{*|(sb(=O)?MHxpc^^0|FlD1Qh^DaDkS0{DDl>TUGDz)#65UQNUn;zvTW{JPp
zVIlc<Vtf;r+ZjI^g4S+l0l<jW*u8rm`l;DOI+O~WIsn-xfPwYd^|d|z0uF1&ex*2Y
zL70*2sVMFxz88o1JgloH`0B++<){i_c33)oN1Yfu!|IzppYFn_NB&Igd;zC*{glu2
zk<O|jG><O?BW@Q<<G=v(!vs@&BoXt!f)N`oWZCZA8{VVP)En~VKJ5>WM}ksNAWsj=
zU6d`z4)xCl#PIo7X0F82w?9TQyV!sjch5fnwj4TCD$9)i4FgHuT#W&CHTaCw1D5!B
z{tSKTa&nwbn^rWO`Mw0%xlnz8<S)Hx`55TUcvN5Nwkw><+nd>ZHK!~L*I$M=m+Nke
zXt8u=amAmw>SYgb>v$tasIopfFb6QBg9i)ab1dJMD9-0{&aobS+4wc54z5(13s_fE
z`aZJTOkGt7)v41hv{~Mw(I2Qx%|+sbmh%2vHVg|<h@(9fJ{D~FB6fOX^rBdu`ZV0y
z+v2UpLY3x<l-~Mc)yEN6u9@d+veIKufdcN+tv2WSe4ei3boTqxov`f$ZRbj1RD*x5
zHS-+HKh7k4Xf>-Splq*O&mi)V3&&Jfp2<pn-0p=SbDqmnk&D5D3n8ispJ;q`dR^>i
zq()-S6)C*<$^7rtO=OPARUK&XZ5Z~)If*NDk#K^|9pppm3Q`8@AIz~W|E?jEH*e7g
zq>>ovRi&0w3hP)iCE#vd>3@%z47`Ha?LPrGIJ!cczHX8;nDO8zGGggUW@+L*K+|e`
z7W5Z&4D7a$vBJ8#`4`}uS8~+P$}V>18j6C~);sZ7hZp?APoxe!dqgV2j|kux#%3Pj
zRDzVGV(+(AFMcVU&fh{W7r4QLLVK&1xFZ>!h<LqXtu9J?@)41xr9aUxmWn}_*b1y5
zm)jSzK-;DI$J%m~e6ZDhaC3eJ8xBf``Cb9?=I)fwq)Yzmn%mP~z(MzA*glZ`Oy_e0
zvW(CE`uyD)=fQ`u#dtdDhO>bQ?vO4*ICmzGAIdJSPO6@%j_jyh)z#_L<8?=qWwz;;
zw2$Ba#tS>}Zg6P9mpOLvKyl0OsJ+pT3iOF12Woi6eYWy|@!}HL2+W3g@kf@lh|ji&
zB0Zxds7;G8Uqi>E#`7f}q9se0s;RXJvPNlu(_QDVR*nD#Ci^|M!Z|2D#%>!Lu}uJV
zKkp#!HvSVHtRUMyH7h@#gU}fK8ENv<ifH<&9q>ClZlO1NLdl+?!5=vJRj!L6`bX(R
zu<jNa+w!-eA|MYKx6}j`M8TSQX`qr?0$a8!I=ec!<HlHh4p4?GD<>3n2qUodR@Tnj
zWn&8;^FUlx97oR?u*I!y>Q2eC(Ny(0Iay5EA&esreIma}59~D6OaZ2~VtbyD=eG3}
zJL=v3JOfq-4x!4%a9RtT0MxvvD$U$SEUt>qf6eIlA|_2=ebco>qxpB?{Q`qe3S)MP
zoij04W3B9iw-QS#^0j7S|Kyjg2Rgg8<5eqXjD9wPgo!?p$3-jb)X_hI{R2d!+1Ke_
zGd>^@#bu%_fGW4PNo>aKFDkRuMVEu*YjPR!#(C`F7I@|93K^$^Q<D+lxnb_z<wfRS
z72o;e1ErNhqcHrlP*HI(b3PP1CsUcaT%uvj>Mb6k#d01bLrtIZn_gXgbPh!YbIh62
ziCK^#7&IKGx>Pw6*qQJr!OTwi2f<7S3QPu6!AhRK!?d7pKE{vjC4%p3KTQ`|w#{$k
zGiXUDOFM?!3Veb&uiit7G6hpD`rn7{50traH!87o++YyoP)_&KJT0`#lR`X{>7L#T
zxk6ElET7auUVmqn@st=C^gm6u+@imFb29Gj8%RTJJv;$P702Hs!n{p1m}9|@KABe(
zlLfiAviKs=uU<r1NW)I*ky&gH1qT)RPWnZ3g!u)Y)b_m><LZ`(5~Ez}j1`v0(82%R
z#^sg7?(Fdxto5`=<UnK;+Tx&YfF?W2?||-YdvR|?fZChQe5us^CDN8-nMj|&+Re$8
zr7gzH%AiFwdJgFi(9FU)`taJHSUTFjdhXgbI-({RA5>jW?y{G}DT=w?Svoq!8;h|#
zYPAj?GQ+7{+wD0$8EKJ%JnP7;;ev*XyZ6M0Y0aLBtrOwnP0FG5@?*q%taFl)3n0f~
z4vg*B96jRz5awd@_0oY_lf-_jyTYa8Yx)FPTJI`(tC$Xb-=acC-}o2Ia^V545u)y`
z<p-@|NU}(7tB)gYcUcUk8U%$C(>F-|DmDL$dU4VJ598?|+=a9MDnBKO8pI>Gm!8_@
zc?URvCI@<&);(v)BWc3B3vB-}3$cn*p>E-h2=~mqJVJ{F;hs@6lmgV(gtrkOSdK%8
zp<|zerdd3DA2Pa-+Zn5mx=dXJgOZy$&le^`%PU=)@YDPRk+B{VG1}BPH;fad4u=8O
z%5a_ADCRx%SNY3xkKX5#4r}A24>4`rfvmyTdja%Q_cUUs5q*M&!OkB<s!c?{%_oN$
zEX&Ew4)xd!^A}B}L{&ed?7d9?kelOH6Vah2u`IzKj#+<vir{Bh>gL&sx8upZ>bp8|
zEAmNmk^JWLI*=7d<*dJv&uAc4v6AFZ^4U=Bzmw10CjSrenN~&tow2RK8E*9`GcbKp
z@t_ekOCU9irF1L%|JZx;c&yg;|9kJY2B{RH2+5csnKKKS=SavUQ<qs~CPXC3JYF(H
z=6RO65Hgc_44G##PrqZ?wfDZCd)N2*{+@rI|9b7)&2_DHu5+EoaeO}SkL|RR)WqH=
zP?}{qLy;RJu&gpHyL%xg0(&h2TVeM?^(x!Z{le8B+Avv&L~M%W?Y2TTLDf~A@d5~L
zrPWB+4{&@%E;!U5icM5@+Q95}QfCYQ>S!qD8BK&0jc}up&jqg?XrInD{*l0jd$-w5
zLQYrjh%g!@z@`f-wU@`6K4zroW4z*udV*_|jN{#{n@x4Pmf?nu$=r_&rsWMGb<-Of
zEiLn~D(hJl?^b3LZ1hS#RHWG_S+PXvrP>oBu_()PjZVa~)*MzZT0j9h(YSgqLDqhM
zmYYZm-PE7K_UL|sO#7v#h}~hIF4NV4S$0C2x+eZen|rc??e>{%JRMMj9_)6qam6i}
zu(ds|qc=>oxm?y=hm!qmrkjrccB8Rb8@i~=sLbNLv^_Zsi%!?-<>V*viv*HZ{y%6*
zf5~J1al~YL1{3WN7N2k5-4B3RI*GkMQ+%79><IN)zq_@Sw|H?<=c#zE+`N4Hh^)RU
z-Y%w46$uyHG?RF5G}P8Ug=fW%_q%6B6MlR}KWurkjHk>#xmJi7>9q9|;O(9-dlC4L
zzxQANbm0LYN@nx1R}<f1pz2qUt^1Mk^lcF=suX`!*4_#}v{aI?t57~xczXt8WAX+b
zacPYDT}w2DO6ojbH`f&xs1Y00UT8Xt+vz+(q-^3Nh?Gr-4QJ@&2d$gz5x-Cyxey`%
z8XRJo;_p_RB(5VT4`}6;Pl_X{t2Q8g-MU71N*m<uGwHl}L`RNd)8*-&xSA|{@)N&d
zJJlg63j(b=5W;sl0tm~T#KH<V!Om5p^ltH8qLe!uIoksnrO%&?6R-fPLr>l%FjV?z
ztfrGs&vw@-c+S|d^NB*wGJ-wy_@W6gw<jtmAKe4Nn^fQX?}gXD?>jCvSK!KJOIVQq
z=+fkzpfkB&Xf}utbN0mTJ$Hqt#61s?AuTZQi_(m5JQXU#n;NdDAmOz(N!ZG9?tS8M
zhI9ZXX*0JnF@bYJ0Hf(q@q9zqHHs~9cRN*Uuig>*UFDhf(XGU(!_br;1jD5U-yb}`
zLWOvpfg9-6eE0M1KR=gHF6DLb6mWs)uhcp{uU*Ov;-2lc^n~7^u;~~RO7+}33QRQ$
zKO7@@hs@TIxgf=BCan4^Pc?(xy(*c)-Xq3&gpEVXVr4c1FW^}a(U@`8{?}Ui$8~Zk
zMkHWn9zM^1(J%bB2jcA0_pz`dhGW_LzWo40{p0uk5C5D&WHZI2Jj7@IhadcJd65^Q
z>hfwg_^kTlcK-Lz{Nx)aA>zIn>nVd9e<}?BrGNdikHaVhB_)&8TUo7N6bAqP``(x#
za<~4Nw<CXjtT)<7@Z%mR=vfYoH&04^7uAIycL`ZJWqftdfb7mqhF^WH59FLbUadl2
zmzNOL%wolXKqfPE{+k<&n=cw}=z^2Q3=FFo4jT&@fNeTj66G*V9eNMX<^D}|Z;Mel
zTdwNJ4w^zI$_8{eMiC?yL?=4Jgrp1S2iBBta5!`LYS>H7lx*Ld&VAQ`zRUD(CQ^9M
zGtqudymW8w7pq060SVL7e??jLeLG>l`nz35oA@$y$lcgaeaoO|dWTTA5&L}k=?|})
zDSi1yL7dQUFdrZ!gycL11X&9rI2i!?O<a5T<%kVn7|xp<<~#y%f6vZ)9C3gG!$Bcy
zH`X$?k0m{*6ApE=i1wU$8;N@#fM<{){pIXsKZ)P^<*JpeCf+?78-TZW__<9<q4YJC
z)pzlU3HHxBU1t#m0CJ*CMHFov=toT3ZFn;#XBII`E|iWx-4NB0hnES>IF0@1QhM|0
z#D8krp8rqUw&`biZdQLBJM|E<!PBOq3^&H0C|-807^(6gI=|}~+_w)A@kI#V3!!*H
zZt$HBj}S{mVqNvCzKU<filE(J)52tD@B~WpKfkK@Z~e_80qQk~U>2H$28GZwDvl3P
zukUt%Zu&b29=nq7z~22beCFaE95UkC<+0kGlvGC;>8RcFa>{e#d0pl3K>CXW`Oj9K
zO9<kaYPcx<BNxGd6`))QFbrodJnbAe7cze>`z@R0s56o@3NE9=Y)9zaHyJAGdvvag
zL~S?2WPNcji{R$Ov-@Qi34gEva0RC9Hx|4B`BkOj#lqB$M@6i3LK#r&0dJqM9O<!U
z;pvRTajigJAn93^Cp%T{OsaCNv!&g6r!FOE^f9q(jk9*6QGElm{qyIO0C|I6AZ`BZ
z=!xe7XqFPds4=gA8R@r~_IT-BVDPg)%@#hfuXEj`kWe?=(c`%<m`fX9;Em(IFK9sD
znK=)tIKgBi_s0hn2?>M9)2jY5c({za9DQeqB;)6fUIl(wGM#L3$G&g(Z}Ai^)IhYt
zhyPI)N4)cMSzKLyT-?uOae~brPVPG%1@7zAgIW<P`|tu=(S8V>5i?L*?TnvS5v-1l
zP?i)Fb{O5b8myJ^g}Ebs|Ei}53*oZEot)?b*`&b{)fCgz!I)t~AHyWJtWyO{uNC^g
zeZDcC@a_w9vd{HNi_8Z8F2}01k$Yhpk{hS>NBb4T4KkbEdGeS#JZwoj=Ti5tN4e;h
zo^C24H|)2LXLLQCN7Wy4`bDOR@I6;qw|uFBUInSo1r6tRlgo(lO+_54E*vFM-7)59
z8QZ`eVD}9ObGJSk5>qyE!>~WBB?}~Q{lc#qv@b5o>#kAescv+I&E6HH_LjCqVLA9t
z?g+2T_ZfT7oHiM=ka_jT<io`{(^CEqAtFV#f|W)09Y&)1&ZI5zvkiDK{qz;!5+Vkx
zT$)62`_I55rl|Kow4XvYB*EV@0R&(MbSN82ai4Ew>j=+_%cSS}A=3-ax}U+*SL5*j
zd?~VnP!lb14m_;l6F?#a9iRp5_^ve+W&a8B0xm#~X{ZKBQwOd4O(#V@fX$v|aqIxW
zE;((K3~%|q2iSO$M%H~7gweG+qgsRHc5s4at<++s7HZGM)xtg_2BZT7A+J<iPP?`+
zK55||)csGn+mx=&)#L!*)onk1^C3PPaLg6Mma@GE5x_;>oPBlkz{o<*8AwdD9V2q?
z8twvhRweSWGJs-e2FJu^;|Iv`2V5}H3(x42;>dwv?M=R6`v=Iwn1?fLQSDC9%LR7S
z2IoU0XR%^$-W&;(M2Tj7qIvxe_Ua*0=iOzq<oJ+!U4LaXx1C|1XnaBXdIEO>9>WvB
z`YdER@1+(p^oC4im)<nmj{-H!{wBo3E^~%7Kp4<rf+O^<iouxv4h}g(Ubcu^A#)gd
z4=?+Ig^w6tg8RZ_P6Bjy8*q!nYJgTPLYNxlJmsDRq~>037u6}N#nxkrRsy4UtNXXX
zUbG+*niB0nYJGP%r(^vik{x1$%)D+V$UFiHc=!%XkOx3QHM<jIB_`YtoOz-$^zAiD
z+5U{@x1@7T)B6!!Ug12nryFrH^%w2UrY|Psr@%PIQ>X7fongNyw`;{Fo2D_ekt4nk
zywq&@aL6RnEdk)vPB!R^;sjbgF2s6WTbV7fm3VK}E=jw!^vy-S2g$RvgyF?0kA2KQ
z>l5=v^xbeqw#eOKIo&mp#k1<e!3dDc2G-yXhkPyUtyXYF$_qE3HtYX1v?h0jWxy~o
zav~a=K6FC;?KZt5S9?BalHy4ql6?o5KyO6^an}Usd2iPlmn9m>9zcBV0Al59=AOJ%
z2>bU=j$tW!`{m}(^Sp&{5ZuGjGxVB=rWThFSs~X@#QTRd;T8}tJh=n4Z7-~#LFCmQ
zg0eR}Sl&RABwepyAEhD)cqe`Y15bwlPOBe<$nJ4fNA|ATJHg!^YM##^xnLH_!>(Y~
zR^sDddjAJPm0YV7@iZ&nnjm3r_6o(`so4C;Jb(ldoY}i`1P6N;WjB5v=7ftyFiG8+
z1<ly}9d$!$X7ziZ`YTJ47!~xKMn=ud_mnkAq{f{hx{uEzzT&GU5doqBu&koc6#jvw
z+JLlsus>FU^<p^>{^y<6OgxMP7PbGIzNpvLDDo-l0D6w#e%wx8+ok)=UrufK`N`}u
zgB{IJ<sk`{Gtx33@@)=dsX}}NUbm=O1AZGPKiKpjx;X^>Q`9T9tQ=PPI8*%KzJl<`
zuCSLFQMw%S-eME+BTBtbO=ghh=jz=n`5aw$t8k_tXM^(?bd}6!fW}9%cUPi=Ayb|>
z6Tx$OIdPCbL&w>Y=16<09PCaCmPGG=OLR*2yC9{b(Ztl7HiL|c@a8!?sCT%>Ua{?l
zzb$IZMayr1KClF)!5_Ngcaq4kiH-fKl!7ix$xusll_8Y7ij3(fJX<n|h(%E`ut>&q
z5bR0x7o=*p1(;|QTqc3bTei2kG1XSpf@_x-3EUo}bT7Slm_};sK!NLbxa0-%Z|m6c
zf2Jk~eSzGOwU*b{TWY;YCMT08hN0ITHp8cCFbVRjNi&dOqTJ&1ZyrgfqKJ+fhsvoh
z-o_gSyB~1yahf>BxhB{cnL;G@4AYz<>?Sa%rR0rRpvdglFq91&ItCKyd;SJFfQB0d
zdyi@m>*4&p0!orDw#)PR3E?)BkAfW6ra#vnP<RfFfie&OK!8^9t!N)|y#^t*bId&r
znE`($7n_QdUX<=EN7Hyy{$gaRe(Luf3E{siTxYO*j&Ijh^ZVPac{M(Jx#;4^xMo++
z_4`ddawErbI|ap8RS3$<jz8Zw`%uN15n$6SP$xzs^RjVAb*RYpx!uOgMu+=+scDY*
z?OW@Z3zV$c>UyZkoQy);$~F*=5@24?vMzFBob=Dvb8o`dx;V>Dq0@}{Dw53>gVGZQ
zgqVz)dvzNYC6d!|7L%A%y6#>{<Hk_O9dy-3!%Q$>hU<5Iy&>lSx$*Jd<JQB+@%_bU
z&Uh`rcAo|?FS5!jkzBZeo`_=2{kUO|C~<-2d{}r7E9(akZa%NWinn8zVVpJ9u;{8T
zu$v5q)3w7UiyFvJuXxMzVLYfRj)lFy+FtfJ(DMi-zJOS`2F<Zgxr2x})Zf#Da<%IM
zhNxMhR#MVJ$tvtV%d!BBp`7(%Kh|MMNrondpiy{~_}3;o(I`nud|z8R8p2P!<gDQ$
zg6&zRYl-TFYi9jAmo@GpB)<U``TY8NDI`Xqkuur&+C+wa%zY-4%F-M=jgDPLj_Heu
zlX;$Vl&dQtR+y;e7;->m7<F-z#<8@>#*twP+tP|<`3QT-3HA6b=5de2LY*O20eHTg
zdO{w9IdoW`;^~HoR@~J$Go~N-h|wlDTTrWCXI9r3uP>LGr(mb=IY_qi-X%7cC0-PB
zN%o`xnQ6pSG?BfBExfm?3=f{LnOFN`e`2%KPVC!gRKWd(jGoE;DWBe7A+<>&NQ|PN
zKI(psx_L}%sX-H;tNsc#p`YJ-;iU&FC+DZ$<~UJc&{+ogop*3NJo}9VtDpO}i(h&q
zs<T%g^yVgc3AG6;v&Yc}cYH*Y^EiMm15Vv`u1N~f1%y(aUQvxza(ukGTIo+|ZY!E3
zMmSYpYv@Y{Y0{%mch%+;TBnDQ59ILs1PDaQTI5VC2ssr!O;y<viidwce(8(*>5rlY
zIi}d;RHtO5;DgH0?np~bd~9qn#~F*X=_`G-Wic_rvUrbAB=)}N8R}{Ysfoze=mQ`K
z;!9!$1u?Qev;<=dZDfih>OxBv3X>>2tct^{zU0&tL!O^E|M+f4iK($4Q=@|(ii{Fx
zPwinN^;F;mDOb+b>mjNc@L7_-$j7cD$__hI<>z)9YwM;I{j1H1EG<lz3W#uP8OR|+
zFmAGBfiz(Mtk5Dp;R3(j`9aQgHW_bh=gSJ%4Z@LzjZRn}hGEE)6Ru0q%fEAPIi<^H
zbu>cd&hgF+4#P4M_vbK<7-9xxzsfg9hI{x0un|1B@A-M!j=@#v0^|qz$B`eF@>z9|
zp4d`ZI=<E~yeL-W%I+xyO+iaJGM#VQs+B@Go|8ur$HJ1`%1Ih#=Z{MH<T`7L>aGx+
z_>$VrB7jbEXdMJqPa+DJK1TVv%bFyqFLEmm%~Q44%XXeHcszTucKjRVy}%N<X{s_$
zF3t?Y6z(ve>{3FpY!yv=*1S>8Hz*C(kEq~0>%B~Z^E<KHg@x+Jc)N(s$#XKuFovvb
zuYOg>tbI3_iw-;TYjSFjd~A~N*i~G6!kX^x_^rfp!zRB%=FY3<-qcgsJ*YRm%~2a}
zQ=GV<PvJ>v$Lp{$Xv6=<n&Rv|JdcMh6Rp1NCK`LmA-Br<3ru9mS{Zju6iY%Jr8cSi
zH$O-36Ub2Hrn}23({BcP^a@MPVz!Mb99Lh$qbTlkX}UZ03yiE8(^4imEvH%=`PV%D
zus@?A(JO<>j;wKd4g-{3mHHe7E#ln26nv9#f?FS>$gEwqgyRVT!LG`xS#Vd0qJc3a
z9A3(>Dvf~2?*s0QSjL~9=It+Mg%KN)3rhUxzI0r40TEXb5@ZezcXa+hsT()W-Xt=)
zmMJIOnBcDK^{_{e@aL|mctA481@zF7wc6h>tAyoUi21$sJ)-77X46nS^5R5JHR32N
zBnS97e483+z!2pfYk0lc+FSbsjVgT^l8@U^o=JmxNSIP+(Cc-rxK1tUKH60r9dptv
zBwxX+x%wgETx%B2dJvp{HV27hYZj(0LBI%+so&m2I1v<k*tuGQoIZ|i(G<~HF_q>b
z1n&|kAqJ80E9k6PV!Wa9SdG9okMOJyc&Qwrnl4-C^BGwjmSDAO7A<P77Fz%lLt5BJ
zu*6Ku1kHl$X1SV2#&(jCDw8ed_0Bl6Ge#u=M%c}wrOlO855{Rp!xDwi!gz_?f$X??
zs}tF+o}QcGXpX`fV&fY)tF!DV8sR_Dp0Y}NofDX(R@sOfZJ4DROehOlSR7qRsg=H+
zY5l?u&ttRfbuoEoU%__3F)#bVd1yM6APLr{FuNN%Go~9+Vje4q9?2Sn62KT7MK1<E
zJMqhZE|7gc%UGS4>sQ_^{9$sC`Krh&dSRfbS-74ya@~$*=rJ5}GT7u_tLe9gP1=XH
zmOOKRXmzmRd@#wVm)Gxnuor39X+_A{t5)|A9n~8VwE+WKH!ZDPV#P0SvZJ+)#5fpn
zEUP#k2<D0PlAGVa-z|bcF~?^re(4Ub^v@I7tbxq06?SA^M1J}#mv1bqz9<%>I1$;7
z`E4k35Wv2cb=M3vVChM6#hm^5(&IrI&pyYz629*!qgd^*ZciZb(+(`d$VhvA$ibXB
z!cR!<%C*(s#&6+4J3zbXs52fFAwEyi;!QoO577c<MAEQFaGA5AWT-#rihSZ5vSo2|
z6UWo}c$?!$xGn$k-sA0a71zqOJYg@|DuyEz1xj|P+<k3LSQsaJj*-fAXv>I|N;#>9
zracE1h4PWN0RMQ0be5oxoTGE86G9BH;mw4p76mLY{~jqy<6eCs{24+dKv*?wxLOMc
z7+F|rwO|rCnxnM^BHf3YPb^Cn8)Cat@r2wp240F>tQJgXmjb6g^4;Q2SH!!72<}dw
z+_x-AnDJQ9OgByZ7&~N&6d<`o)D1QI`0;Tt31gQhAjy#aOhc@(6k5Zl_3mK|(VSGB
zBebh*3-I<31*X-r<GAZrDgboUhDnr14+)iJ@n<m<m?E!kz#UakD$T10!j_PQi$o^=
z;_Z9&Y!)ubUtG-@zo(9^8T4_Pm4BFJe<bv;{2ZPb=Zb|8g2k|>`x9heAFFZ+?<OH{
zmeZbxP1<evHl2t&9Xs>$(<-x7*r_IF68xHXWnVN3b6ra$cb`e3ip%t5KdBP>iR<@}
z$#oz#>%%L&ZV0j{iRd%45`+6h!wWQ<!sxN`%QD0#&gUp_N!P1NXyH!~GN!h^E|yoK
z)4)F_4!M(e)aC-s;{3|Csj0nZK?QdOfB4n}HY4kZiBfjTigLxa!IAnTzq0dVXU&E>
z_7r$ivQ0Ik36V$Z?D{uw+2Mt}yv<}1+wNCRQ_9?jyIK%>p`9z;&MfK(<?0gg)wIl^
z)8C?EwjSsVO(906n9{Xf1_CF(<Yn)@Q@2S7+6$FJBpnq{y6coZ*2ExdbM#4Rfp=x{
zHR>`|15(us+mG=n92=;`eyVwW6OZp_gYyp$2Z-UK-HBf{K2z6=`5k8+B^(9Lzam)L
zcK!5s<wa<zp0<hK$3DquX^47+{jSkVm1^WRh^(%xP6VscaJ`N2T_vRN%+2zv^<Pk-
zOE#Stp__qvT*v?N8(>j6;I*g*sDUc36V|`{9wXIKyCc<!Y`>Z|n%5>uQw5kOow_4d
zZ{Y{81<<8l7<;J5!ReoXoO5!BR)dJtZVVqgY7=fu7|!Ky$Kfit`eO1H2G57RT5}S|
zleMWqRga9cYKh8>X&^4Tme7oM`3VOGkX&m=T7`84J55oNdwYRyz9U1jgSh!`iTk_O
z46~DL`a{yg(Sj8;8g;0ctKr0-3rJvjcku~=nv-(%1c`f*$;dSSZJfmgwPD}~Ir(4`
z`(x9_Xi0W1COA-T?kt{AF6oL}Gb5ObQU9RDv51K|ZSwQ3W*z|yHPq`R+W$=1xf^#v
zA`BnPx#&t1^fUx>mK@YZ$%3)T-V1Lb<s*&@FfH5p{C(>sgkpU6SLP@c-B&8}i?ZZz
zjJO$qlwqU$Ye-wRo4k9qOo}vlJoVSlhg!Ev(m}0mLk4sEm4%#YJZVv|n%`WQ-+N@B
z)gx6(#P`XqG^;6HXrmYMMxSVn2Q#ToCCql5vE!gW)Vb}^{Y2-s`=2^DmLxagzjSV;
z#CR^#eF*#mhn|BBaj<{-<KCY)k2e7*C#8khP4V|abntBjHorXccOQQR)y)}wbU^d*
z57Fxufh&vziR8K9pl3Drae{Bw#xt9Sf5;0Pa@-i?>%Zd|bDA)UE8solRYN|;pv1!A
zIP1Csq>(h-4KGx-<d-Qc)r!WiCggYb`?|{n!Eah8OiuLm|M}dRX8S#rFz!USXq&r=
z7kfP>>R^sd0xvl<6fXH$o{x_@=jeKNFk%^pIM$M+u&yj6qVGRBlO&_QdT-XZ%P?OT
z^jNC{$_Fr`*c}Mz=F7k7x+pAZ>UBW8dNF}>dZ;stu_N=byVs**+7&RNSUR%<A|pKx
z@gZ*x0YYJZhJy%;X3snK9)091NwQ9toE@b5CU`yQA@5gr(>oq`XHpznWRe0He?Bg=
zLOdIqxa8i47IK$c{!dE@ngEO7rSVl$78Ooqa6;0x0<KxNm%}#-6-f<8DoBsiN%AYU
zStLu(si$uPR;Oc+fTtmW(fmIZZ4R9QtzwWj>CfG9ai;YSg*0{X1y((YZ$Kn)xT$fr
z0GS*N+kk)e73iArD;RS#ecMk;(%p=##yWY^9`WcwAMXxQN9<gWsYek1tliZ%R00Yo
z`0`nHEz`Ht%H8-(hXLr#=kj)@eGTtMp84>Q$9(7U>DJ@!bSe&1>$e?F+Pzfb$AfI+
zWWJPS&6qQGktHB0i-iU!)676o?#qdI_HBy&<$WV9#{iFnfnDf+Uio~?ZV3O`7Z-8V
zboFj%#cn+A7~OX{o6gC<<ZS-pqTaVQugcsPNmhR?(NNIY3@_JLC6?2plWiB#5hGP%
zUT^1w$M_w$EzNI_79r`DNO5}1yIPVs7r-x^eejChb9&OKtT46=-b9CjmECnPMo(hn
z(8@<$V#@DOYJ=_^+2F&f(-%vWN!S;Rs(}A6-JfI^4w3KyVJgO6^&SGCe(EZLR%{uU
z$eK1}r%9?P2irg&c5BN9iA4oy?ec~y$ye?b8l36A<lD3ra&<MxWX)8qvSXkFULg)C
z`SW>qlbE3DwkP{ULoE<eCmq{)|8*a|%nrQyr#`08Qz{ynrMSxmQEq;Y=;~wmFw<w8
zIX_v1^>R6G(!cA-+D+7EK-5RWMhkvH_7)D`Z;d#T7Lrj1L0XK;X14{ZI=b@iK6k1u
zyqT6D5E4M|NGM&pi@0jLa^Aj^;5c#E@O^dRwyFOY`4|UuQ+ROxzG_KV(t^>jH>7^*
znwR7ZFQL}KiTz-u(!o<S6-uc^%*UCDOFut**&{tlGI4Lt!%5})E3#jp&RrMg^USW4
z*IpOy1NNI^Io~Qm{y<9;P%KB)OZ(vqqzM2p%%87WLiDm?v>PtyB}@_`N&_VM@E0g?
zS+_4l{v~Z2X%{0g{d$WD*lBAN_Mw{A%fgWZ##WaEq>Q!5Ccs2aAo;Eb=Jksh6X++e
zkiQSCeqM|SR~z$K^732nTtd2*njxV<6~IH-EGBrVQh5wpM>#Zu7u4vbLiCH3cWZ#T
z!6(h4sqKC~7MhbaUN<_3Iw{)RDw%2Gjdmy$UzD;C&N!Gk{V8?3&~cxVo>p-hG9dPP
z40r!2cB9*GRKC!E@y5e96>_l(ng|}DrkQ5(jxoMrZ-KgMQ`pS*531TE4X@t?#e(w_
z;VWm5Bv&z&Ex({p-UlFW(7G}2o`^R!9wNC?OgCbhWs0rA(cO8h$w4!d1=&&AjKwBs
z4*I7%Q$s6Il{~?rT*B^-3%$9Cg%;z6&GaS~OaT++rBCYD^J=D0c;ZJLf@*KuU4NA;
zaZ(iyGG|{Knes9b8c9oMnR?3Q-<ifFE-!*mHxC|EyODQ3`fT<Kh_Oy>)OJlP8Ttc0
zU*xZJy&E*ra^0~F@aVF|2)B$HO%^R&><WS!qxSXLPgUW4523BDKVFL)zH%|aI_?we
z5c7N}<!0)HTAJX=7ryyjGky!OZ#Nc!!_{pL$*OJn+EDdS0@?#HQAR`F)<braGX0}_
z^@GDuQdG_36(X&|J5n@1Uw1q-Sz&!<&dKtBbqjq?gxL_G$%7odnn-+eaZ{SB&+PY<
z?%r560tZjZAzBTeM&==5R1-$wJ?6p{MBlbnhe5NTqlzOu{UaRjTf@*Pm5t>Mo3D0r
zwN`m4w)=n_?|e`ZXZc&zllZgTI5#I7bG^-dA#2iUBC=8KMyu^|n&V)Ln^m*03y{@R
zDbPCJX{B+0;6Hf>qpZ8mb$*sS4DxH3_mm9GPWQ=!Bkj5$K`%Q1t)s&@Y>i)AB+E6(
zj>`K|Zm0Da^l%fVwHO;RG#L*6ar{PZd`a%rf)HC%;yp-%B&w?3wm{M#`xBn5?Lap8
z<B76XqCi>9<tE=f1n>}LKXVmLmb-fc6Dnd(S+sVZr@*hVneTY<MYac<`>$zvXiJzv
zQu47oBSTD6Go)&jyPs2)(};evj^8pKl`jNwI4{1fZAQ$!o^V=TbnM{6sX{~{&W7uc
zDc#&vPIwpou;r7PMPSX{*{60`-*mN)$e=5^5tYmh{O31TMZ+8<Cwlm%i(ufjWi<|O
zo(&j0O$+3h<L?*!apneGaApfxChO`}H;JJdS%ukog;_6b<GtO&);f&Udn})3H<VhF
z!I?dYs)EjBBQ>n+Odkq5Lm`sqvzmkvnv5>e(?l~-E^>EOW<A|5gq_w>6PlBJhQ;8K
z=)<5@X?dAKU=fj`T~pGo80BKCv)N6kG;c8RjlJaeO2teTYukFXTCnc#a+t?D&DUu@
zWA)C9OnCA<nba`%)rM@#^;MlYc0zR=5~*MTFAnl1wnqu}`1iNh)GjVtEI(>x)7rSr
z>YKGcD7#~r&XJPZZbMmi95X4aqR2iynA^i9L8Q7GrS@$i;um9S55z1n@}^Di`&?R@
zQeo<?rSvTA-qpz8I2Cyqq`HWHHz_HvNe$>a_F)*Ag)+t!6OD#*)gN<j4j%9_HWC-Q
z!nK+-ZGK<7A<~YEQEf@J<W4Y(%~*Cijqc|JxHm%xhph12+p<4fGcJS@Pkq~&9x9~H
z;c+Tb3tclWz{;9`z{{S@F^43L)n2lAm8-DcT&eHS5=Lm<<M3H#o<>>O;t1@%7zaQl
z?>T<uUjk{}>QJ!s@{;OmP4L4%y29};D}ILr|B%f5L(e1>iDU~W{vn?G{e;cr^0e~c
zs_5%VCi)=Iv<*&_#An6fN%{e+tfxAbYr;#)H+%k;e%t{G;YR6dbHmIMY#_4*Fi%h8
zI}^_3lA2LmkIb#qyf1<5`@>SJ+R@ibPxBM7M=XqgeggAG#C4P|6P!PqRC@RdNu_3k
z#fNPwnAl&#Ur~ss+DD?}K%zFD!#tXF3;#Wh`nRQP?U7XlMAIXQt7i?^kEC&UKMyDQ
zp>X)`FKnO$)j~LdAlG+V5Mv(}0vmbI80LRm8@LsZAtSlIf%Ywk9ca||FEFR@R~Ha<
z`X>&bHMocT@`l$Xzxd@Rkzel0v+?uQ_TQiAyTm34Qqi<9@9F;cg0Qm&|No!;$BO>d
zKmYq8&v*(E+>%4iAdLqz$@7MVYe)o>%C+*%QA1T(N3ZL@&l#N~+a<Oxi|<j9o)$mt
z5uWn%S`>6_`uOlXak5_fp`3cdjpvv21q2H%#_B*<TU`ma5bN6ug8|}eZi|H3p|TSA
zuNogRCWmZ*0cG;o<qMR=lOJD#iv##iaurhTaI~15MY8&+LoH-+D4uF|nq9W&mAcoT
zqLSJ-n1$(xC~<oJKxK8;Fl9}uMRAQbY9^a}c|G@g8sv`;@}J^714E>fpLZ@^@cZ6F
zbNMpvW4sf}^wm;tKV_D#yyS&)V*tjwUs7?R=7Fz$t{X;2>}HBeq<GRVp7G771hg;I
zzBorY00V{rh=cjk4E3d}Y&c|%AtWTh%<NKV(rSWo;$(>VD|~1gSb>43VL)o02Tphc
zNI3N;26Q6SKtlonZZYE>BE8!*$d}tacIMI)%)wu)9oR!pS<IVQrG$htF#IzB%hD6n
zGn;uA`4G*AJg4`h!R@swn>R;yZ_o4I!VwZ8`k8VIWQ_o^bFILl)BExGDv-HA{9~^y
zm=7QTGOiToHshg01jf3D5hd#2m>Gk8VFVexgSoYYU{U!Uy(cAA8p-whPiNt1HsX4#
zId0Hb`dU#n&i->;a+ESq!;N-eQaqU;7di8wDSRB_wbbL{^98M-$a@DfL`I9ZNCP-f
zj^`8N9LAJBwx-$*x~z<<lkfsuTZ*-A1>(7_2LjG?A)+6Hk7ze<)+1vi2T;1gaHFzf
z;73JtsmHTNy@kBmVHchQT5p?|JmNyDzgU)Uego;%(f59*zsoy?rm?6V2GVkR)`UwJ
z|D{=5F`;&8fbXTLuC1Fo@4>f*1VRn%XkT@01h_@(+QMLo8W849rx8v7FL>-22zMpj
z5b#b3oCnwnjeLl-K^JWQj2WUP%aO)Inbd4nf>F1rRS|cZDpNeUm4R^iCxoA-N|;CT
z3nWkuC`KD5KAih;%shV<Q6s;Gw(S+>%|Na=@ME(xe9$VY>J2#RJf8DrH})ptIE}rT
zoQJxyxx<{u?)EtTT&8JqNV{*JQi|q0@;5b&lP~__ZxSCSF~>FiAo^WBtMG!$ox0N+
z5r}#*6@=|TG}d1Bx<<ay_^Jv`(CvN1#MeN>F#<b=<w&A^u%-8m7T6&1GQA3sBq!8d
zVr#jiq^Xi)(bYVStr1iVY2OT4v7BJOh`y9;@X7PS6_tA;{lYuYb#A6V=+ZFXy8aFx
z{z6cY?5(^^A=goqONk?o;a}bfs%KVRBAQV;2&#^`;I+yBd5Bpx?0k(<zn>K8v(5aN
ziXui~O`c%|yo!g8bR=`?y0@j;pLq7iXW*JD47yJkU9}2T5|gn!Si<S<&=3?ZsVV?t
zqS+tXOegF{8A3h`<qazsnOUnq*vkACwVdSY+Y1fv`LCbRuX@x|)1Oj$fAM2YVt^Iz
zhQJ8wgPgEKK*4gk+X_V{UdhP~gEXh=os36k`<I;FIIc-<%>+8d$S-e%Une?Jm_>H%
z3=es8a>o0s-=O5{b}w+76?y%#SE!V~_}c!dGn08Lq&$gQ5;0A1e@<-t(m1n*^k^J?
z$<x30X#D;brP=@Ah|+X^h|&gOfVhCNpQhe`!*Pi-@O$Jcy&*8L*~4|dC;ROI9cjf~
zZAE89l!m+bPf^;~E!G_*L!}=|X|`Hz@k0vR-pIdF*zB$TQ<nDsnZnlE)K~yps@YU)
z4E4rMqC4%r#u-I5z77*L1-X!Rv1R>nHDLYOm&HZrgOWlQB)tbjY0egbwDRk*9<rRh
z)uv!1M3NP4q?z)S$-LrtKG!p-hEZp`340mc6A9QuxWA`;Za#_nWIT2#;UZHj*n$p(
zeaT{9e(Gcvj7JF47ZDNlDWqj6c<n#5XO-H>|Kw?XzK`*dm&q(Nbv8Cs!)lS}r=_;=
z{+whG`t_QBOd^ytwFM>DAT-bhgRb5lKmzx`-cMNw@k!fG20qeX`--h*T?@^j05Q6!
z+t-PB5Fl>g=<q4`o{&26*`$@c1S9X(;zNZmP=fQPc8z=vS8~%<{MH`4Wq6x1{}icx
z6l-D1?ojW9!4Koj;A3Av6SY)u$s`u^;fO2*#{)BuxOFnX8{Wusg+vfCK?b3n8*(+U
z<o2gx%_jTc1kH~-?RWQH^oztVa=^>Xf@{cDH(?9#-BDVJ`J71a#L36-bh|?<Jk0Y>
z%U)WWYto|?!qb8WTfPooT#iR<mLn1|`0~pw)@AsYKy9?Ym)gqv3Dl6P2dMKiT@Ef$
zRXI(Fws{Y9aNk;<Apwwvdwf%Gp0RSOH`u&vB`f%GM*e}K{l}S7VG?UjuuSsDPjsAf
z>CHup)2w)c2yMEqv|qG=F;_;f^4S!0pI(94gS)~R-gjc_=Xh9eDnSTc60lVddu|9r
zAJOj&`D0(Wg1;bIGM6JAr@OOx32wlD;nc9C{~f26WIkMe$Wx1kG{N#g8cR=rI^mN#
z%w_vKUiOu_G`=c(8GmnQ_@FtJ;p*5%#X;FXF~HUYrY%uXb{(EbYobonWWTV9vtRW8
z5~-%>|NkDTW|YmLE&6G%bx&6GJm`<AgsqiRWE;ad7}#252F_tFbmsb<jsR(<Q&iX+
zI@W9ny(|57qT-&K`>=8F^Lo3?Z9o0v%nO8sHiZPf$u!Q)+d~QEy}#AsOoPgl!9O8J
zb9qf)Hi8iI{yM9+l3wh2!Hte%YYo-5M`x4PtC534_c{DhF`B2ZfFW{y7|e)0l<LsC
zr(BaaZRev80)<dOUsss56%>2o2>FeiIo(EGYYo2z3<cDz%ZY`#DtFYVS!2|cEF4i5
z`3!5at8Zh1x^~-39iSc0FTyX*PJ0ZOY;H0!3T?kW{U^{Zo+(ugqO{~X${ig93Du)H
zaAm&)tB4(+DB@)G53BH)z<9EfL>LFB^>?PJ8uqG~O#(>pZ(!OUPwFF9`4Xr3P|a_1
zy==rc9^ThGGF!L5`M!mvtRgI|>`UJss2#;<zHQ8A)Z8;JG?BKB=a6sNf;ie6?Ikiw
z+iRK5k#_3_jvc#QykRq~*%L2^T)tgu%f@zQ>+D#GykdcG{fmR$n+)u!%%Q*6JcSD)
zu=5wSp6Fz5_b`l4jIytvys9Ofe_5DS-Wk<#9kj-!QOy`OTGmXnHc03QwrMUMwe^1{
zO7}6q?LMBUayS*)=i~N{-@0m-V4Gu1rT2_!-0>z-jdb4M`csldr<!!xdiyD}Vw4-v
zGL;V!{u+D!&Rq&tc}2T<9Q^}$b_pO=GN^>9kX&*tl+xg&t01-z^))u;yJ{*g-R;O%
zjwpIE1kBOQx)R{lPm8ykGH72d2{e_uYdl1fo6=dzX71tCOpSe$+9^;%#oCnY!^)i4
zXeT5KJGdl^(z#cs;k`^cP5+u9De$RiEN^6H{1<srE?*VTQLWVzGq);@p4nkH=@1AS
zT-$q5D7oWps*0!J73>xcc!NYD%bI?wk!%yOg3zGwT+LwmO`N@y2TsXnvhRW5tn6$r
zU*~fpwtlb$peGC#)Dwi&yVpwDaa}ORc{P#>7>2Lyx#GI(&`o|&PuQ2Nm;Na|b$}-&
zKXqVd_T2!dvW7VEBFD(Z1a2CNNf_U*gi~wyCEd9)2h+F9i={P!yV=#i{*5f;e9a|R
zu~1Le=w<S3k;*=V>~S3h$~H-mtabH$DZ|drjHCD>{{)6BlN1gIbtTvre{5?1w(W%?
zJ)H5ZHiPdy9D?bpcLG-$R`2kA3XB#-A|nHtv;zd3YWeH*<R<1M=H6&Y5Y}`#oIX`6
z*`J5yjHt?#)&%6Q5ZQIL)jq~yQqQ_~L(xa!rZ9u26V;iz^CZu=^MW4=319k8LfT_2
z*_(_pPU^Pf!YX2fG;Lm5U)55wmAK(;b(@;;S?f1u=PhZa;5LZc-5rp^Gp-v80vq)9
zd1<__QkIsG%0xqmh&9SWuiB-)YH3oi1;Y~R75$^zy75_ZZtHjHOh7euTP+K_DVyBL
zM7Lk%>;qF6*Q?E1UWzF7l&lLu?x4z*%|L^cfGuiH!ihngLdHvv@#v!yIW%1@Jxdtb
zt?zK-{4qOkJ@D=k=80#~H<^a>p$+{eVE<zse>tz5)AhAW)r_>WPi0}x>Yf=%N;;+Y
zM)=aHz^7WX7QjV|-YX9O2N&&Co?qKg|6`V&&ujLopNW~78}Yt{gc~i(DbAE#{gr_f
zn9i;h6Z`^sXZ}2cEBHMXc*TIFZKe*vII9p8ISwXJh+esP7!)ZyE2&TjVZG6`^6V$M
z)fnR`HgETo>FDqdYPlYBWDiwM01}VzQTH|c=yy|Z74}vl=oW2?qPmpJ3CSlLrp048
zk!McFlj@wP*&ZG*9ixtJ{Nru^js&T@Qef0KjfvZOd$tzjz}l#DO{Ca-RICBDNjieq
z0VlIdfm*W151P31Wk732Q6+4pEWHgc$@xgdtYUXMg;ivujeooi$(h=qhy*Z~=3=sx
zW5=a{jauR4qHmOij_|PE42#d2a&JdQ5n;5E=FQeBo)Da~MZFjj?x3S$Vm2Lt<(|`m
zkYd#&sQcROGqMIKG%ee1m8V5X+Ml5}jB1c0m?#TU{-hqLVP0gVY^++e(fz2%G^qfA
z4+1lW!iq$yINI$E@vv<>Igy0l>(`JI=~%`*9jOP@^P<EYAL`A(Jw6EcR@*vF+A$g#
z8{_c!S0RQo&)(}8B{=T>?Mp{vAuM4e5dn;$5uGbg{qf?k&HN=3ORvFr2%;&(4rOBM
zFL?h`Ce|x%z5x<-y%h_a?Ty97#uCWv;3-O!;#q)zEN1kG5DBCCs0jKBBRZ=3Wk}if
zLADLel7{RSxgde5XjRX7Y-)!bw~M+A8r~PGqp#%1QAy|2$`D*(JND@gTZs0n_#8d?
z9F-h`=d8imZYC_1d(MLB2|kPH$)c83hl1h7ja{d*mf1CnMSgIFG;cVD>@*I@4=<7*
zOqp_I8Z`6mePuE}f{PxCh`=V(L?|60`B@o=8VyW?8M?JThWbo=IkT-X&vIKoW!NkZ
z8E#0G(Lv~)X;{yDjKW9E@uGXTl(q<mde`Q9KYN3197(N8-i^b@?|X;H+z=&L7E`js
zt<{&=l*uDXcf&7NBc^k;u*o=j$0+P)1289^YKBjj#|oJaRx~FbO<ZJT5}dAdyWA|S
z%G=Z($9+Su?rb6evzqe_+A1-$@`BwmbW`h~BSy!RMcP%!q!eGVqj~k1khKT%8|dl)
zJw_Lat@&hdrUg+}Asq;rrAZ{Gal$0Gd|qPau4YRr(ziE^=5)zMYRrJ*uTCIX-r?4F
zZBE<i<jd^Ra(1lPPz(E<=0vL>lv_?C6ZU1&33iGI?9+iM^5dAP7R*^lz-dIVAi<e|
z@bN!ajDcGP`(Qu!>}qrA10MW$JgWoKayLe>h}PFh%uCxqxb(=dg;p~tr(3V<wh8HY
zb86$dm8J=KefnFx3gb#%nK8%rzaOWvx*D8lXAkjO-1yf^poiT*X`GQ~__;>y#N$M8
zw;R6o=6tLkh+vSKV<*x-ug&*&L!2fn8WTsNJ*2Y2bxVo&>yOsXKMuq{{wY+2nBPXc
zFY-VCvq5NkBzW`sHI{ZzST1~gf_F;t$(0c|tjBndFg6#N12)0_Vx+l5ds9b1G5A%;
z1q5>a4PN2rR^oNsPHvw)fXd0?gKjyGRSOW~w01{{P`e$#UCSHIUw%}6fw!Fpzl2V#
zm7p4ru(^A0CA8vIXMm%Twqu!JE1z&N)L&F3LhcsG+L%_l8$v087YG9JVM!rniGZgI
zAU3xEN2(NKhSerONqKR|ry1vxi14KhjEDB`Awls5s-sm>wAfD6LBt$JL4jYoDx2(G
zqU~x{j)<;F>kK2f6tNG{+jq3Azm6((ucqA7zQfUy3UQTtpmbc0)mv^YWu(G*bTQtX
zf$q&>22ACd&AQ31Q53E;GQ4pf<^vR!oGFG@gV_X;u^F4B1eMX`w(HE&*`V|bGG}#m
z6wFA2H#PBxw0WdXMKVji=L1MRnoKED)J_V|wh&YZYDrc$bJ?{M4g<q%=1);c{E+Yx
zp?_|7T(s(Y6z#t@YUzTXI^LlDaU)xQ!=@w9<OW1qiX6o(L31k~^*Gf-ILUTPN(@7P
z?bgY*FlLM@Ke^E|+{b3s#$g5;f|$zXtBlchENR&&cl<kQt2H4i1)P<O=XJ&Jc<jGc
zl&Nb!d#25etSCi#oIx-4?h_yaCYRo$Jxl_%2$rvhaLc$!Y3*bXkUf`8xfYC$TEi3B
z4!5Q5106y`j>Z5%0J|#d@;%E&mfFon*KwZ}y$lW(1m`-SH<ZIH(#}tG?{7#n`;~Xz
z7r%DGia%3$s`Cj&_4zw@MvMZL;65k`Z-*3@%Y2!b8~_1gfLrnxzOn;$n?(8Og{nV{
z<=JqZ{p}pRWnPe7HOaS2|29%|{>VE&<$6@0nt8TW5N{@Hf45b(BVD1-W9ROqy`?j3
z=CO7nD>IHA@EZ(3Rw_MPP-6?TjF)ndYnitjhtsJY^*iP25sW2oFgkWrBIcc!Z2t1=
z*d75S=HTD^!|-2I0-y4Y|0UNqc#2iOcYPm@Zgwmp0{(si55RsheH442F%JESpy`ir
zqI-9fs_TUBKIo*(_H$?T264>7t?PK)ok1_p(TCXxwp0p}O`boIgm9ZRI-m%zo~Wwb
zf&(e-veQ)z1aPPB>ma8o3f@&u4aRRKdQn-iP;Bz3<|c!LE#~c8cw18hT$dSBlcI+^
zv@NZz=JTb6Z`6(=%5FZj?E#A<o~RRef&J||ych>810<9AaL&!T(<Z+<8FdPYbGytH
z4FtErC$1GnS(HQp#Zju`Cv#KmP@Qmgp8d$Tv&GLHO~?7lp?$l(?3NedmQ~8$aOuXD
z;P~rFr2|yf^)%Ea^QeI!O$B=W%ui|WdU)z1RWb9k#XSN+8sbG2#vM+!9+G*qiU#eF
zCJgVsZxysu6OA$xgI!+v6>yYj)JQg-??pN=7PXaI2`hne5NCV1a~=E@(ej^7`omk}
z@?0D`UsWV*_NV^r-Y$*CENS!I=A$;}0Q<BA{)J?jvtw>6gq6+CNn|WKqFu)y1^Yj(
z6ehNp=fYE=W9EC!jYS)2sgwuotxYs*!<4!(Hx6(3Z6TTu(WPqB@~DSYWJ){sa^iNM
zb05A`gw)emXlYaB+BuVM6Zzno*G)iFE>GviL+0Eg`MG00y!IN@;dKPX)7)z!mA+3L
z{mv>5#RP<<$Q}lLewB*x9l868G=+_iNKmSkDr1>15=}D>)|kDh*L)h1a}=D*ajZHN
z6iGD=mD@Guv{71MQ$yW&yKI2Oy4?TZTK#yrv9WxnSb<lVA7d3%spMJ#{obz;J4Pa8
z`|QkCWb$~^rr--h^lW2gZu-p(!JTZQTTq0H<3&A4xQr@-xkK)RzAb9B0W6+$iDx--
zn{_fl04-Wk&S5jgIdhLwvJpL!zE@fV&Gz<sw}O-pKJTUXd=5ItOO;Ai$P)AKQ;6fM
zB{za$ex^ILne8?_wK~9<O^;ZkzP2KoQZ{o)V59R4otO%smX1XnpKV}6jPm@fWnMg;
z;1p?m4|%mF)&t|8qYq+zO24eZv{ZT=9hEe=sg>>EJ>w;qxvDdxu=RLSB|*0M+N~wk
z6jJ+)fnkmriymY30+Wk|_oD44E*=bZmdH5w)x!QJ!0gvmeihdTBe{69Y?di5lR2~e
zV<6VEZO!%!P2P$MY4&i;=q0L6B#o-;0=nz9nmD!7ZN08f96$Q#|MIZ2pMjYraqSJA
zqlZ7lD1!AZ#tv}esQZ+EhJfNpF4@2fdh-zF=7YxyD7P3biItNDaP7p9h}@x)Zxr{A
z6oIq%Dh~_=-h#=Y?U3bmb^&l~5(G)FNVuB4u0f1Xkik%?9jX;jZV-93G!HD{?Y1x$
zUE$A$;@JBH2}-U!o{0p<){@@(yrjyyF_%>pi`W4t4^eJkBmw1C3(7(LToY_No^r6)
zsUZQNlYn+0O)s8F@vlT+Zdo^vaIA?R`E8B%ldi&4C>9D)6jNoI&8Q}ckex^N9s`gg
zy#)9h`O8F725pbc+WF2R6Q0o*lsPM7LtbC)<WEt(IadguyoW7xgVOUNl~Ttra8k-7
z#2K66K%KWgS?nh++IexGK)5JDIfXQ4=&{;MZ(E5v8pRhkq}U1M!zt@Smma|Yz;jj?
z{0E9&giNu3d*vbqn2M2{0qVsT<P{K;I(}<&iZt}rqw%b)PM6(C$NGx}@S3v?^e10G
zX4!K#rXncpmQZ3}h!oKmK>5!it=twz_w*;=Ou21w6GdX4{V)Xqy=uLknFNb6KbPl)
zv1H;o1?L$Hlp%T7K0cNtlCa(<kiHMNHQE%G&P$DY$&WjEGEc(xW8~-=I%K2%3=dc8
zMyexy8m_xx1tgf1ElKN}0ZDG8K}!8ucm*ErOeAqY>DA|s45}w49&ZrmtH_aRH!JI)
zj+Rtq-~FJKY^}UvBl~3uj!j}Kd}+fh(90?$Dl^uchgI)XZjfPA{#EK^a0d5VG`K7x
z!&U8Z>&Bv@y3LQ-Wi-j;Kr*OJ_PTjf_Cr#iA}8PKS%yijTcu>pzZz*hg6O*LcQyFN
zhpnOe7cXBn=h%MAq<wEQurS=O0L<ahnGda3Hs4a+6taB)Hf}lh<Bl-PDeGo;><CGc
zk7~RFm*t?e{!>*n7GF#xl1}fideYSWEtUsFF0PDIDJk9+_1siOwa{Nms-%!&3<`4n
zBERufEfJz_{oTBpUUO{>S*#D*XfX%)lS)bsWeJlT^Spc}sw6471?TIs%v6X{ZbSU-
z{CUZ8{N0K2<jn0-_MBglxap|)uo+5cI9r$AmnrxGa0@o}z%p2=xr_hoVFgqj$!4#}
z_<1<QEwy5mu<O{Zw?Bu`yJgM=kix|2N$e(@iuPQbR)y0(x2NH^($dax>G7U(GfbiJ
z)ut}DZ%AR0$ghN7SeOnoc9i%=JQ0j_MI9Zkp>AO>_k@}rdzNFU2HwJZJbp=%1Jpx7
z-7N0+hV%ku`Q+_#0=`_2KFfhZXHTlztYjsx=a=VmQoEjl)mHSCnxXj$o3brOkK*vE
z)GMm}R~GF8_~aW1pxJ@gJM@#@#hOgwNtGDa3aJcc&z(Hg=}20$`NrA4YbA4|?qkJW
zrJXy_+`lG`Uwhn#N9t8Z&dN6c5avHG)6fkJsvk54f4r5lJ?f(=;~iv64BpiQH8<Y$
zr=6ryL2K|9SCFjjh(S1LiP*tY3ax<FJ5d@OnR(rYsQIj_Gfuc?R!t!L%TtB!Ue-ip
z{rJvp3^cN6yUwvTY{{OR`SL3uk+<*r$E=@93I9Ix2w@?`4SMwaur3b8EMT~8(B$!Z
z`F2i^CpznT#UG%Bi-J_vwfod=fHaE=t@A>;ygs~}as$+%+&J;GjkD2gHv{@DTKFT~
zi!A}!6vDO~%51Jfy#dH1`ZbeXQNgP+X_JkqK5WeD<$^R4kGFGuzcvfce)LL=OyytI
z8C!tIC-2G5mxZ}!w(6*McWi+Hu$0Fl0ybVI`L%g4y&Kvs<KQ#%;GcSjKpWXuvON|{
zJjVvW<%`c8zB=HB)m)m*jeqF|4#jk%Zp`F;S8lX06FN4!8V$Q&O<U<P>aJ*v;T3^l
zuJMKi5I6N;3>t|(N3cF}BX_`Oeg`<wKFlPepxnfGYm0y5_G3+rDG_wkAA^w9QMw0e
znaAOUTT02+ZafY8TQDJoP|W8uyKSYnxbchQR~laj=Uy*OwhJ|Pe39Ap#*>%tIKChc
zelvHcp{r_ljEIMF8HQFz`^xzZ&>r%xUV_ZgAhu<HHs7b^f)k{lG2L~Iwe(KuTP)g@
zSb`PMQJ;L;Rp6=4{|AuSbhM9%qpzgq2`E1!UE|Ko6GQ?VF&<e2MDw5fd`D+LxIJkW
zy_d`?$yn87^jbAMu#bT$&tghvG`oFP+8Zt`uM>ZGw+sZp-9tyw?(N7Jn<PeBKkjR=
zpoSMS5}s5+Zlv{!Hl4f73l0|*w$?MH17%|8gKvwQHm2T>SV#D4dEJ<-8+wy`A-dx6
zPTdCNGobf$KA4FlQL-(>7;x*Ix9l*#yqD3o^B(JsQ)XwBDavBCykX6()jZ>gWQDlQ
ztEq1Sb)iywvi_NJ3qQ)9A4TgwDkPyQ8mM2jeaXwd3zyhW9x1|d_O86~3OOZay*kVa
zEh|RkR~n2Ifp+2=Vrk}=9!QBAq4>*Ky0x6ka<9KTinE{MB6JN*72muC{-uxl_aCOe
z1&7Xf2_@VgM+a&|*TTv}hDe(fIItM!zT?%Vj1at9g0p4ZgCoEMu_V;#CJLghfyr2v
z1HrZGDy^pcZYOLk=X{W?{i=z-k3gahSl)c+8oTI+AUd>5F`z?>n+9V5AC7yWNmp7g
zYs1kh<h8&qdgSCr`qDd!s}WmphNz^jJtlM?d{_c$j-{!%Al)7Et6R=lv&!%*I~kLj
zH3kZXEZh$zz%1NeptH2jj(IZMaHj8<L7@alXdFcl-JckgvrmOSVgok~#Et`y{QfsL
zEkOn|s$OWOxon+$_Io+yQgchFvK>jPLW16guJ1*)E}q=;Cg!L=ec}3>M;BlH3fPt<
z1DCo`23N4=J7j7m^gSq{2jGFr;mHQIMhO@%(c^*NqwL`Jvtl!%N_)H^UT>NlUoQdD
z8H#*Jc^^Em_@Vds$3pu1p9WdbH0S7vc{}zm$HHGOj>}n$@bWEW^yV!1xmWnGc=>Of
zn9wr>xc1=;De3ox{OiB;_lkrRKVTlXR_Gh%W&iJiYs8G8fH`eFHT8diK9h#)#3cRJ
zJN<usom%h1sy5CN5wHElN#<W_KDcc6)xbPyu10KX@?(wO$55WV<^LIJ$V?uDaQ}7z
zsR!SjIO46}V!~)mz+y@8<0t?1H!`vzUxQ7=JLb=?;Zh^}1vdA`f{<hv;tw_>ojWKF
zv~r&h5_kan#US90djXPpHE|j<p8?FN6Id4QqsOpPeBT2ns_Y%tHxn=w*?@lQbg1`w
zSdq<95$VAloIy4b-eA5NQW}TRWQx9Xi-MECzG})*a8DtcT<p)u&3jgHp(r;hC4=@j
z$+BJniwPMn;Eim6Mq^t5Yf%i*Ul>CX5xFMqdis%{c$tz?{QLzVyO8i$CLyV%NT6A0
zt~Asok3l+BRFRo|2~sVMp|osXv3m5DctVAG4z8B6>#ZUycEXT+ZSe}Hy@7!ui`x$J
zNFH&pDoX9zN{VxN`FdmDZx(I0h>DQ;p=vsJY=zRWw)?^!{9kmFa|br;bVr_&{^GkS
zIKuT-84hcu{63R<MtL2aFkDtM$~N^^6FW49C!x1~NS)F9gg5K~%AfA0aE=B?pb}*0
zH-uQVGkda8rGfA-6M{W8B~9=2`5^pCkeZz)2~h>9*~|`rZU>-AIxyFps|j&|(m63o
zxGGhpDQ-;OAAO{<phZ+unZpSYQG09rLm*+>)`g(3K`77^|Nf3_5Fy0{uU9RrqQxxX
zjs3o?w=5lMp78y?JpEOiv_d5E{U+bF{>YMkcIC25KMcA`-)E@v?AHqJx8a<fRa&B5
zRLAQbfO-Jwqo2R2m_fSIVFT&{heZ$f6Rd~18IZG;0>5jllwMzf#EUe(2Q`f*#%buw
z5^)ur=LOo5Y<xCX()NsfOB&M+_-1{hxBJ(+YIaHsc4YQ`aSO8M090N$_lhIr`!;3d
z{T?w8p&Zvb^-)W-zB1sdOQQdvmHiAQ3vHnY)sHL6TQxf3R|{0PF^88<XlWz5_fCdF
zM!Q8p$2Kv1xIV4nmQMKF<)*QYf8N=@x72^^c?NuNc{rj#!+H2acg0T7H4gX+?oV0X
zgMmK66rY5Ix8u&~tKhZWeiDp59#m0zl3_DPG7S1^v_$*LnE_FXd$Qr|l25J;`GRBL
zE<3tXnTC2za!o&>)qLEeKv>XHnZ#bsdB&rjAYAIaG**f5PK<esv*f&2ZatC=D{yLI
z&}Ml4Fq7<jFsY8B4E;x+)@WN^_mu{MUH4+W=ztqvb<dgS2Ccr=s#Tg)c~BCLUS96F
zfkk2&#qe=5v0jh0(WZiPuCB91c%_q^<%IJHCaXYCgF>>!!||3Cy*rPkGAhREe0x-V
z$FYX{9zGODc~{m1hr?2%o@^Lb`Lx=Q;+)_8xO1-S<^&Is*U~l5LVrsXV}tEZ3iHEm
z7B+V$_ubavhmn?P4g73o(o-5!7GWA6vtISzR<(B9Pcii_%2-Q`%YVbyn_NU{a&oKO
zZ;!U|cg@;bwhq#$&-{<*zw%$`nNf>mnC^Ywi~hPP|Ah3|evH9@&qMiIKjCQOF-Nk#
zUmp3-(uKnI^B8-D@6U728z1Z^-`d2@HkU^BpY4K!b`;LhY&NjbIp8{gmLn3`XZis=
zFaUI}X(WwAB-adDt_v?2G9Dr^Q+LdvsbRbU5h%q_qz#m<)H9gDesxHN8uiz-?)~WD
zY6@|i2QSE-5yICa5No9$oIQYe%Oa$PB85e<d?k5R*O$KwXybR1XCGeS1%Kxs&PC99
zD+bGW2U0znRk~p#@cf5Hx>cvSEJ09I#5)nD8&OYig+ReQ{slD^W0;_GIj$$SX&<}u
zP>#OzA)yVtL``dulEY!##pv}B7?F-z(3!;Hf>iZFP&&Lr+GSxF9uWF=ohD{ax=$jN
zKUO@Ou{suI)FX#fy2F`3JF`2xTLwJ>r!xc4)>wd!uoY0lmR)_JTiZ7eKfX|nFK@}D
zo>!*>urn6B;4sw~e(%ZHTV4Z*n-0Ou9>oCZ!7983lwjJ=PIE$!Xu224Oa;6lyYeKG
z(<6u*6te@wu<RWOQ&ZZ&1vv~71hg|N)#q>YrW)T<f5?88=@9(}8|ntwBi_No)stSj
zQf~&-cI*Gc-g`zxnQd#^25d!Cw2ELxa#Tc+SOfvdIfDd2$sz(0il_);iR2sv0m(^{
zqyhv%k(?1!a*`-H7w>#*yZh|-?Cp8a@9+E3W9%_{R|EB|XRS5oyzlGQiASX&N-pfJ
zKqUzmuZi*rpPq(R_cc!8-o1rALMtS@d_Xi#^UrAg>u)LM@vt_l*ooNbEE^%%(BT^I
z`|#y2vyoY>gQiDHcd}cE5!xkngVak>&fuxy^Ct9EZi8oWTd)Ggh3;;%5G#pY&s5Q}
zkSEMLRxNMiCOnCB;poU9^@-!-L8!Frpn^-yUp<j#?jDxE9#Lkh5p3L0zZcgPcEQ;c
znKf5-!!b92u=Ct=m7a=<TgG=rGIQbAlq(L>e-MIM?yfMMQYrX`O5k~GrwSnxWA}^S
z!2Ve|k8t%2W_aqs7c~Y7q`OsnX0su|Jvz(_M%}BRzS*h(nH8ZpS7$_gm!N#BKX<iU
z%4HbDFXLd{aB-|YL|KCLsJi|7UE@>F4*zGv2y*?uvVVJ@)ZhND{ih;1O%?7Qs7!-)
zM-X<-7wBWjX3%i)KLHCj4dPM>JBUG7?lfhT8SOP{#V$N{xn9Uj(wCCGCBI7uFUhci
zliNYh!Q3k-M}8lf6?9OA28jfNh`sXTkW?^QG|A1mFU9L_S4sH@y1h{zNJFb<nSl%^
zCpXlX;kG(RL{vG5bo#n7HFN804+jx<)1XH;b`Vw67IaSYr)3&IqC>b8A<_;b(2$1$
ztgZMhahMG&-lgH*>ji4;6tu4V<!Yz2pA``^C&62Jd8QADD2-nXJ<koM;D`!NKf`@!
zCc3NLI@!q7bE4ZGyrs!S)}Du3xeWw{2LQ`FWaTs#B$)`k>J2^<nyPxyKnaHFB|8~A
zOV~K%@1B}wUe%h%ADxT7P3H_NkG9#MnH?P^4n6!I%eR(wEk!+wzuL;)6@B|dYr5p_
zr{FXYCb|!pU(ALsbvy8sRoAcQf*}4>)U(><4t@;$@~Z2s#*!H|*A$QpIjidl@DOtA
ziB#J`n=?Om%Q#G(ra1tuz)Rh6wmw2m>%93(-xy-2U7-Yf%L6jPinY>~uRGF%*<j_m
z2uBzftervII8!k6ypYqK{rE*ZroD==n#;zS;iixUn!omUj}I1Xy^J!X`Pd~;TiX-_
zXs8YI*%kc2{<EshR_H)U9a-sgwQwYecZZNr`r4*#z|8agYX%xL?bdemE$7MaZc3^d
zje-R#Y$5kDSOt)15A*qFi;yxrJz0a~YxD_U`eFCZs~r&&3VKvTtbP%>!(%&EuT~6L
zkl}j5eoq6{jF0%QTQ;2oUy3Yy>xn#raOS&}36AOR0+CbPHuuApuW0yiRdpFOv+g=5
zdC0w>>Jygxv8>}3?1m_xio0t!&8z$@mfhkGeE0>f{kL~zkymmDwj$R4re&PH^cdm$
zc|7cu#4ekSUUmZgom-sH2iY_Vud@^FA3Ro&eslpFST!7Jm!~PBies`|CRRU^q%@7H
z_f$AN^}0P#${45KRa$#zi-~>FM_hY?o^1|`vvyyO8FX5Fj6EmQ7pB29ny8iM^;tTN
zQ8^AvE0ZaIes3jJ_~t}4+qHlP+Pfs)^jv`#q+`XRi~{QVgVYdoA(#{u4*+*+XA+sT
z2Wcd0MINSOKN<Fp)_mZ#pwfq%e9JO4-9M>%WX(^EHQ156Pd$cwYc<&X4*rd}D~fj>
zV5sT#Dcd8r_2+k71JlMpcIhh|?x62~!JItIW)XCmg32Y91DJ{jipxFt_3?W*ELCPY
z@m3OmTz1#s!!6OhPCs64fX}nll98foe>2Mq+`assyxQ6%3BM&z1LCuGINwZr1u{^Y
zVTw{MzS3)4XSb&QLba+uJVPrzrb`@-(lt3sMqe2gcokpXb!NG1;${l;Vsvz<?DV=;
z<I>$j-{+s5ZPQOFeN}O#>5$*!+V+NCVTP*<0hO7knHj;zl@?M@8$sG?L`YoF)alN;
zzv#mr1j99z8Q>gk(z@_;TZDkq95oAy&Zg2wE(UM0o-*$cewyaDjx~#*{j10E@2--s
zbYVWyKwrQ7=f8aHOmhjcPqmi@ekE_>mA&Z}p9;OZbz~qgY&k7p$0pDFXYi(-&~2WE
zQ^RQkmo|{(ZCh^I$GF<VODFCQOWTE$@H{`|CiR(CIhN>Xl+k;ocz+bT6dPUZ6X|qr
zq607t?P44c16kKg3VlA!upn=Xl><%C>k}}Ivb7H7%w=0ir3};J{T%zPO-r_rW*CPN
z8JD4O?;<fA+mx<3*<1%b&ux3^VVM*Wu7Wt*+4`(ZRIj0*kNy1w&RxRoVO2W~VJNNG
z{X;)r{zRL2G4wDywjr}w?Sw3rx>Ja0;pt{DQ3pz1mM)Z4eKeNo<_|`8cSISFZjQZf
z89r$|e7|5`(=!?0D)Z)|;?f_P92#G!rJZtY3Vw`Gvo7T2tjmF-r9;**sxh*xkoE{+
zteqo{ZMMiXp80vt+2A%lPT-E!E-!L4v6&`3-i?diY{~5^=Pjfp@^FqvdYah5h2f|N
zS@3kyV*d9}M^w=y1ti04vVjbSidU-R=Re&E5Gs^CofcS)>O$aELW*&3$SAHGut<6(
zIfGEj@r|pIKv+mGH?hr_Rc?yf3oX%vLv@qj=FLd|i6mQ~+iH+2qCN={peBOXj0%xQ
zl;{g3&7Be`$Tr$uEaGM2BKULZ8bVSt;2MLqHe5O#KBIWxg5NIkX5#g#rZY0}pI3g8
z6pl0JH6kPq=_8t|EZzRm)0ZbICo(a~pOOZlM>X`GV{bWr??oltBH10t4r1>&g{{Pj
zVjDUmdIsSNwL^HtU#CGM5+=PA8kX6jveampULm&|QNr2ldSnyj$uetJsb*!DNCsc<
zaB1+saUDsx8klPY$-Gf}W=32h_SwFf$~&@0_dJ)GQieQnchLd_!Lx@&EWC!NwtCaE
zN=17jKl#es5l><73O`2giEgL6PfIF?JV}<05)g}*({<eB)^XZA`;!+lm68YQ5~?q4
zgq+rPQRhhK9+<2vRMh@WRrE+i=s@H5l->A^fA-eXkC9s~<ZZQq7mGf760to|Ja5;7
zqBY&MV!k;fgmc)dATZo~GegX{`AGn$_2aFN`_i{5-nIgdtnV;(9nQOyB~C?u%#xHT
zwaQEE$|t;fn?02Cu6!iSI(M*WDt@#k@3KUAahBWE`rGKp+Nk0Lym9lbhytbR7NUS0
zV3lJVbxOocjd*!n*6#J@-T7FUPjRu&9+K|0r3@h0aiP#9S)c5VWDa)|`n3R{P+nr$
z8rA8-tcPFfwZ;j4+gKd^c1rSmP=g{r$BmceQ;EK@ID9N|dhqkmFwJRv^i@MoMChr9
zh^cU|me#6?^jq*MeCV_i;GIuiKKDV?yGP%}Y8`Y%<q$?c?k2$_Ta@(SISJcsqn~a!
z(*rc710Auromk?uoldx2KcGnIMY1HKhFTF$RI_&G#14=%$GBRH6eAtQS~XiOkH<Xv
zh>S+vb3bWINAWZm_7z!bntGa;Qe!M<y}ivG7;J6(g_q`tySW&DYQGkAfyjKlKlQ-b
zTmT~QFDpeA;ED6TAzHPjf;&1<GH<%dWfxdThNls8122id+zHGN5mOow)8Uer-u?~~
zQO&vWjB!piRSO;bL8CM;q^Vsz{8@B4TvDc)e2jC1x{@#Gd|3-P6iZC_3LW;S>0si-
z3M(BF1{jT7Dx`TZ?8uLQr^5>ure1vTZom2ZHFhq=%W?&r`|}4C?W<H4M%u!$sSg?B
zUGoQ*f!R-<Pr{G%0bV_2Ci?HK#((N^rbkfa9r)V^t>j0<LcI~PVGPIhx{(k`_7wE@
zh1rLKesahjjMgbpxveN?lMv{nniw(dr-_1sU<7q;Ot$VBebaUTrM2`AE!AtJ;`*^P
z`_%(33U)m2eJb+mFVz<wMRX2SUr{{eT9^A$DSCQIh?{9&e}UOrK1pblYDUQnh=P47
zSu+$-&nSXVK>rPI#DmknK7kUm$dUQXTpjP_aRU>P`BG)BUfqCWbGo?p1#*zG(R~gP
zQ5?Cj>@;5ZmcFY3JW3XjyQG(04a$RoW<c*oeY}pCn6@YB`?bxJ!(zSl>?xefB^c^H
zpJcyvLbmxZuD&VB4Brqtc~+D7l`XM%hLu8B@=z{{?eZ+1@|g@T1|)}7CiGk7&}IJi
zX#c@jad()Q$8fd=w_vU~Q`@&xKmV3CBd975l&02wUh6*oD7{Hp)FsmMylrk9!#?^$
z>G@&Kls*opb`GJ=|EA~ps{PcP0Ed6}fJ^()r$fm*dTUwQ;SZo<#$;o~8>NS!_R9$g
ztc12$Cdc_SEcMdxA##&9DQa@O;@;tE^-{}5TB;DjnciUXl(*|i(WwJ(P7B!DSB8Rg
zEBceV9*cqDxw;XUw#$r<^S8Uo9(@v<iC>aEK6ti96`>MuX0UI>)pQfU8Rgs58AH09
zaWXYJcS4?1MN79YoIT7!vXY;qf(MaynSdZT%r2HprkfzCFc4U2^s4IhYa4Zeqpjv~
zeKhD62J%cPeLgmMx(ik%!%uIvzJKkn2`4B0>s;cBkS5u2=CS)KQiC#iT#Ao_aAs~=
z1{J<5`kn4ZBkfb*Jz7pkD*A|I8F!~fD?pn|%>y`OYT>1`>9PVfaoUI!ZK$c7-J*30
zIIws14|O&n=cQW4dgL0n_Z4<-q#yM^sJSi$`aZ%x{AaWYaxU^6gwYg)W$|TGyQ46(
zs!Ra2A-mKMQ~me0J*u6@mrKTl8)S#_9)9o08bPT=@#G^Vi82_UXqRQe)fqUGPUm@H
zEtc8X#%%P1Jhx0PrqYm%M)$>5ojCCLUxZz7VCHj1ypx^NDru@0Ju7+3y23ukM0GRc
zqo|}LJN}0#osG8qfzltZYW0$vSetJx&rW`95M6sZZF^B&NK<9-@>l!23Gh>|g`*0j
zA6g7Qpk01Q0gx#dtQ8!gx^65C9>8$dkBw9eH6OrOlI=wG7S&_;C#?EwJhq(KzB&xp
zj(=tINyhT698kwaTSPx8i?`HQn~6=6g(;U-v`=w8CQ!m~UW@mkupnD(iNSl=8RFwr
zC$x)(%vF0`Quv<^EeOBAli<4RB>jf@lwrqNl>vnAF*r^kim}pVUA=N5OhfdC=BIV+
z6s6>E55Q9B3x>%CPX2m3Rz*aNLc#4ymQ#Pn-&;%$U6>~KV~D)r)@ea@_cMr`OA;Fv
zq$ERg%&jECNd`OICkOGm-(Q=BscS5@-VJXr^0Gx#=nDIzV2@h*zVL#x<wqQ$t1_-r
zd0>TZ#T**x55ZY}mwO>sY_CRnq++O#am@RU$sr5Vj~5(Q-Xe}P?cFZBX&<~Cq$YYj
zTapN?RkM;|q8m=s`d2mA55|8{bJe~=wLFu@q`%OTI+C%NAx)HGnhOAZ_OvdiLaQ&_
z^wGY(F%RLG3vLtR)^!^8#x1MN0ck8NZG8ZtahuxTST~NJorP(_3i$L=!%nf1c+6FN
z&^G&`bBoH{j{<8mf@9n&O2<-C*}ZHO>bskez8h%(CzQGIh`hO<j+wf!$LOlpwTu~P
z|Ew(Ufie78AGvTiCu8N#lI%i)4MaIE#+~V&Y$BT|{-8RN)TGvImw385el}D}ewp7K
zV_dv6KCI|+-vkaewR5OW3j-x#nB>j%CM{7pPf-g;j9piB3w*~5D{#cazSJ~B(W3l3
zyp=2(C8teKwvc?=ULEM<#lUUA{9Di)?M;e&fy;ZIBv+mx(YaYzNS^PfEHXe;`0lN3
zF5({jK6E36N>h*kZZ+0xh#!%u@O`rTIDx2@n*IFf9K%b^5w|!s_TO;0VFo6HM3)mk
z@A@_U1B&vJKhW~evv&kJUdM*{l4#Ox2n;<s*dgjSeXv-lU-qY5>!@_uUve!&vQ-ZK
zgaz;eCQl(*jrz@GWokx3#0OT|4eX6mAaM$7)6JY3ivoR1D@niS8!KpYR(?ZZdS@Y^
zY9fiCV?^WqEcBJa$O-Rn7$|KLp)tqK4{|NnNTqSoix}QTEbhjbyp-E94`&VT#_+(@
zBzzt{_Cm)pDmDwQBSQCbi4Ii%0e~_O0F-$I$B|w@JhX0Rs@PMiRiYP`(wv}Ua?@q5
zNE0k++e(-@En<OM;rqEQ#vvFJ;|UK?@uaVfgHp>=@--sV3h>5UMf!~a>0ZylhPpmN
zk;e;Px@zmBz@LmV)z2S9RiF}HR+|wq62#fi4XNzCcVcpdQOn@$O=cF_JROYmK3<5c
z95f}qg^d0eUS8Fo0FOt&*mpSxb9c)e`er}d*=NO?4Vy@!=HwiBVoz28CqLaL^kdyE
zwz0)O&fOd8jktFj8pQDjQ4dWRi1X=8$UC9McD+soGfw>j+J6CXXBqd##Jes@PmGe6
zUx^q8(<|Qe!{72SzlIap`}-WA1?`gi>};n~#u}e;8pu=iE%ihdaXE$TQX6L#`}aa~
zi>TRMrCmJu*u<ch$Cb;qSNQSiYd)q{kA_xCmSQ2e$``4yJ^8kH=Vaz5_}kN6s>G{q
zr2X{!*(^d9upqD)&F#cJg?G7uMd1+5gz2IkeuXb+;G70-{oEli2s&FJpw?}^<kAd5
zKIX9PD?QsR_?eo8M=)693!|KuF4R$FnRL@v{r2^@qufv=G>31+iQ<>>hm>fHLz1c;
zGB!PyBGDa)UwU~wY5^A5dLK&I+jLBu(RIolo-r+UUvQdOq7cscAElPKNWul;_-hrh
z>HCRKxU5c(H%cFQ3p_gs)M7e5F6`=UB*4)#sjYQ`;|L*?G1q)|dp&?o_a5}%>8y3(
zdF4{l(czA-BGS(2={r-FIZ3=tv3#ll#`NQ0iCC+=f{UQzYCr7JojeI<)e!ovhODvl
z)VL+Kx<VLrDkFleGzk!F;k4*bvNGfS)It7uj){Ch&OZfPQeP=1o;66T@k1zzu)|ql
zzjV=!hAUwLgpE;jU3~h{JB+(qtIMsQK>P+j?aLiUyZCT{T)L6q<2oey>0kwEM|4t$
zlFPy`L{9?OM?2a~Ly#X@`}w>qV-KZJQs!)<a&^)PES##P4tIMrn~Hi~;O{hOWXor>
z;_x-R4XRUMHg4q-2(EEmfJ)lkO6n3*40(f=VLXz&F<H8#R$Rl~Ymz}Eiwk_uA5pKl
z9oJK~CeoK+6Dq|#8($q~5s&shN&sP3>#divLl+d6j+|u=v0PveGsa&gX6^0)Pv1yi
z5T1|4Kr~M`9a(A6L*%q#llY<8GQfVy*)^-^qNob*mj}7v{-qWugT}V%=91Yt6$Jg8
zHh6UDO@D=p<DPzpHHr7lUhC&_0gs0_AFxg8;o<3|Zc%K-J;VS%NX(8eU>x6Sr_t~`
zyTBunLr2(B`J3zx5dS5;_(S|1evWth-Ga-#0f2`MIWSl(^-kD-idw&BN>%cbcw9vT
zVKr-)U`%U{n<l0rk}XHhpOUTRu5plT6_66{w+gO5WccGB$=1l}|4Xt}5WuM@29m9#
zwp++JDAs%3mr_wc(}f|+j^ZfS<v~1#56zN1POlgFrcn@7v0FaJq>DS3)Z%dw9ZMsW
z3{18RIu8)cim8#P;GiQQ$AEChpwBf~iaQgFm9b@TelT(QQ`AS%)@mZV7IBpvP^9Bp
zHEK@yC1E+_vV=K1#a%0{8zQ~|q2&!HseNVSL?|)W&gYnC&4Rltoh_KLO&fsYyZhhm
ziJ0jZy$0H<DFjF|N5y3p*onVP7si<@x(nEB|AHS`E%@S6q~1ekZ({FXOhe#mV_YAi
zq%Ci7e++>RYXNmD1{#TP&RnUG2L+)^R5}XQo_15whaMtz_xq72y945Tl8IGz%do(z
zRGs-)`=G@u(7ZcFQpD7dzrt$tW5BxI7t9~d4$*@QJ9dXJJPhPiHte@6vQZ`uk6!3U
zBi4t>WSsI#Vv3g&Uc`Y2H737If6lJP%T8fC)Z3(G+6D6-qsos~t1G~NV<3_rkJ99;
zGO3A+wq9WOHqM|HAYNe0e!1apJs`fj5h>LarP;MXo4&mDm9rZsQy_Q7gt4Eq*)pe7
z%WSSYR0a<2%4-X+)2FWo6#fx9vOY5U%JD_N>a|=qLa}E{Q2_XuI~Pcu2ffJ#SnFm<
zlPi5!2d*zXPWT=xx!WVHGZ{r1x+T<_<?|b&7089?5lIuB{Pf?G5;{b(MX&xr4Cy6?
z-#CD>h)Pv=>{i#r&d?iiDf%~F!acRj7r#Fd)_Lx>VLpNq)B2%#Tw@Z*@M42g2Y=hA
z<UsVWkrL=zzf@MoL!svVga5%p0Ij5*JwTD)?=p2PPU^B;UO`}ndZ#eI3g_b>TFEkm
zH9#4<E`mK3#xp32+3D7GnTvos#`X*I{l9UCT)G5((a2wFt^G&3&h7WU0~MLKy1H2f
z+#xnAnZ8;Um^4pk8*u&9sU)0IN}Gd4oLrWz)Cdue(t36(yF?NYRwLH%(PgVmjFkTA
zg@B0k1v~Fqy|Rbtr03ib?rC6q%>A<Tg>cbWN_0bUrB%Z3X6}zf#E|Gc>K7=tUj!~L
z(E~^IJ_GdgXPOLIU_HB6eCUQX$g*=_F6Gx|EBEW+K#VCWJPt4g^MGnvm-w!i<5)%j
z^w2n@Ah$WasQYt@|6k_&|Mh>0kRz9}Lr?zZ4>htpmJVSkh2T2lsH*?2*b=cnrG<CO
zgLKWTkU(M1)#>g=d&s}yjQQ<)+DFuY{B_5iFh_sa>-ev~*gqYRG*)<A)6co){@>+R
zfB%sG|HuCQRs2T52!o{m{v|(lzVwe~YdV31>rWcs@xXU5>K=2c*jZ}ic{i|^+X$+b
zJ?+{vzj?#iW}tAqJj1zf`&Z@Y=k-WvY7asrwR^H)(yfi{j#V}n00a3bk^|XWYG>LD
zF00rS)3@ayEV*=&OD9VAz`Lj1xuR8k+!qhP1b)?MpyVbfOl-Dpg{yW%gF`Rfaw@S%
z^h05X-ALv3{IXpZXm-504Dzm_(i_8_SWNHtMqF4V*!>n#mqrItyRg^NdcP(n#-Wdd
zT4e7k$BTx5nW3!u$>DOy4r>n({c&#{hOT83%`v@w^->R+PAsKq%48H?c<)2%h~UB^
zbZv1+vV*MyQpC=~w0(X3dN>>=jzB4C)^^t}RsO!HrBPyNI)^?D$65@>9f=I;>@Nu2
zAFYVt2rFib!$5l6{n}@zm><X2Huisc3knE{fTRw)M0hs|FQe&Fa^)idFPMW|&{Sy%
zwv!HSeabN}rt{}5HvBXr);cKu<0E<eR0J-zVyEf1eu>J{Y!5(t#t|~m>{DZHWhz!5
z5_UG{Vf1Yy8_u@q7I^zMWL+Te7n{qs-e!fke@`m$vPS~U=K70R)0<_!*z_4kV1FjI
zK(D=4#?1iY?~n+B7@)-=z%glVt`r{@!Ha&l<8QmzA45!f3uvDCx^YM@oT{Q=x9>NV
zsK%jB@PGY=<bfjiG@U}TVd7(FuKfi5_{B*iPkPlZq1sKIzhqhk?rrr}#eSHD4%w1?
zm_YGNqksQ|n&ztJdux1Ie%NoI8(au<{cM9&yW*d{4iR#6;5+%a6uPsS;VXYjzYuA4
zFF-{c7b@1vWV}V#dQV`d1%Q%MvHo6^C<=%WuV(rR7(J`bqf5Uo)+=A`JINID4k8ll
z9H9b&9%Jqha)-5pg~=RaDVFo<D_=Wwu#24*jn<*q?_zxlTCECEwxmUi9Y4*de81Tk
zUn#>D-OKd!6z#89cvW@?s^5nIq@KP8OfgMRr1GnU4ds50dOXA~%t7R_x%r@&a3qxV
zc}7*7=G*b1Vbbb&J3nIuE(!q_fzFwnx5&?Q?CJa)^#9+chmWTb1~dOhZ~b4wZd;Vn
zXD#o~oi>PYvhi|p>^21P?G&{jnbChD#URXTsF6W^r&Ce5^amitR$fe7!`dIQTXbKU
z>23AhC0o19oUWL`4$>)B;X?AQgrxp9yDe2=wUjosjH|4l$@@v&NOol{khn`|kS+fi
zcey@eu*aX7Hha?7NQD`cRkdH3Lk&_O?vj2xw3eF6*C{59zjanD5^a>e*bzV57}mAE
zq5IRIroud$uq7L%nP3p*7SKxljpPn`Li0+DngZKFP6&}Q^y{tVSV@gd+B!+-iVNO4
zIiZ$Q`SDBE9|B$RAP2L#RjcZ8Hs&4oz}|x#)0(d*S|bW7Wyo))??u_kjRsl{p4cTF
zJ(**ixs^*c|NZnt<ou+J;l;xzoEcVbyjZen$|^WrZ2(_kP1#ZGwC|Qu(^t8C+QDVR
z6SZ?K#+Fl>Z+N=S<%IaxwcwfvX)RYzNUwfe9%+4A=`Cb=kqImpRgr2xg=E&zq-8zv
z!5?m7^EqGGVFoGl_iB4phAwTtkB_4~yqn%4sx{T;R585SubZ^EIIB5gjeqC$yTRlC
zrCJrqMj?*zuL$r#a_T2u3&Qk%?pPPs-hR0+zyn#m#qUoDl)!%WS4O5ujMZ0=Vh+Fk
zY~-6&=>{0Ad2o$B1Q}C_3*=OO1(nQK7|qWhlbK><(5UnVHi~7c)KOB_+Cd~iUD#cC
z?em>^Xhhn8al;8OHN68mGJ9w~sZaCA*83xY2L=e+Y96jv_DJGe49uZihP^3WPgg@q
zZkrXb4d)>R!-fSdekQ?R_oIrJ_EwsZ2f-Dg(wCTbrfe>UU%rLpq;$Zz;eL+C9w7&I
zPHi^i;<mk2M`-}k{12r+18oZ1CA9TGX`6ixMvK+lX`U*xCG%x7J0%|l-oO^S<qk-)
z7C@3g&t=7K2r}zze@3iaGk#8a{$9mu&wXIF+r#iE(Xu`9CIt}5WSc{VhL&4}9g$eI
zT_-a+uD$V_%FFIxts`t9{Ea&yT`g66yVmPKvnm0H#y2EcRDn9~I``R5P|LOSs!qkK
z>?XgEOkb9nbAy-e{^vJ_3Jy%tmlBKEf}UX_&HWUkxjum+Li0WVz8xXd;qfiUZ}x6|
z8y<8N?OQ#%-JdPu_GV3mm)Qu2YRpH7e1_JamxiEqvsLz+<;cr4Sm;!^NBXXdH&ZJV
zmsO~nH6cr|7+eNr2pqwur<>paaxr>;Nj^K87Fu`xEGC)O@B*R2E2{%Xq_^YNqLO7J
z${?Jh1%ceVQ|Z9(@^WB*)n>Ex&d%|1`>wOSmDo;so=rCc!}bU^?Gdr0nTq7yCE$y#
zO7>|=u}J(#bsL6W#!@%SSZPnhfMr9cfP%~0fhS^5kJ%uKsFq!Y(NeQ^8b?sYl|kN<
z%VII_)VK7>B7^&Hs{hbln~56Sop*G%R%a~Cffb@=?U9l_viYY}O(@H}f<l0iPN#QU
zan0-Xt_`q7u2&Y1RLK`gTF$e+#t%%gSMB6e&4Am$nF21(9)7FdYxY^J0xN``G7Ww>
zQ+M%Qgp96oM{7rD=xo~{h_qeane|7t5KDwj*^?w^(fmGmeYHq}b@7}c=I@NEe{)M!
zk|mm@CHXI9{%vrn1`$&yYGmu&oG4-k(OFm`ycGickcN>2Ox5jWyB0pi{D74pH{F1y
zPTilfGjWmjkviW3HbEd9AWl^d9bg_MA)Kv_A{-|TI%mTbedwk&SRZ@rkj%b9=#ESN
zsXYte{IA-xr_W{&ma%89b$e`^&2L-PJWqqP*dVtD0dgX!Y;YfffbC2>Xu2cRHB~-P
z5q=ylCH0jX3PY4<M3{D19O$TL&3jOUT;_-VF)f<InZn3Qv|$SVf<|Wt1exy;Db-PI
zPE*bHhf$)<HjHKp;#wn=vo2&bdfhXhi+NU8(A=v~VmiWW{T@7FDEsdnb*&ue&6^Y*
z?AHt<<5i<?0SCg0##c$PR1zI4j-2(0hz4BfnH~`HY0>*OOA|INj7Ws71?;CoKgZ(6
zY0ws1rFUvx0&MQYb0~a<pc_#@<JSog!EGtKyf(F2ClH@>rF!g)QXwL(T_=kRW(qMP
zEC3w4YeZ8eVVs0(!Z4ya8FK8<ha`29`bp$mXL?I|VyD}_YE(oI7DdUxTjQ4Thk)F}
z(@||51w7O$cSU)G+2v2pew4u!G5V$xn~Z*2pB!(&WgXp%F{~-`>@B^>?C2J2-=m%G
zCgIWqLb%<poo4|*G?P8K?$peo7ZZ3TbB2WS$_p;jqSR-}#MOz2<_$<9FLzw(h#tfx
z;d1ihn!^l7ZtT4vHnJm^9*>9psxtm8YX#`GI^&7I;{^|&fAGkF3J{7WcWXwv>=w~;
z>aD_icn^hzYpZ(4xCU1hx}!S>VH*U=T5U*A@tR7uJV%F1jkqm{b?wk?o7>Zw1AERq
zJ)zHEq8T2_&GAgjezrS`4QPoI6j=RIOVk}iydfws$I8&MhT)>*N?9XC)7Y?bb%$MG
zn3Zex0disYy1{nl?wzB9*691rLL<3@Oj|Es=W*V6&6C_wWjSr<7ruQiw=FOEKW9Gd
zKN2b-veyy!*C<@%Al-v{YM%|#=B<<20W~sEaul|99yTNHt~qRvQ>8<d{Htiq=M9L~
zzJKsNP3K%6K}W>7@Obn2SOsJClP&ni3pHWRmW}r^n7Gj3@qmX^fVsTmsIe{?YZFR<
z_#GWv$P{&4?V_T`mTDIW+UxOL#;fe1VNSG^lS(O{@8mhqbsN0UqB%mKOAFaI3Ft+s
zOck~!*;vZ{D4$QY*pM|593`q(YKkENiBgDAZv^%9?4;L3nOAnyci34JaeB??RKu0J
zFxof2l{>E6HDGkexl+3%@<x?$Sn_+E3!d)oexUI3ZLT#CO$y|4K<zg^>QWR}__j&a
zB+*Q|<^*w=#jPKj^ZKgC9n1)_@7UrbFW9Tyjgvp<?$PD8>^iwFKX+)t{Q6ww6&60m
zkub`F7S;-(Ld})$os9Htf*@(5S)Q6IPqpc+ydNhy`fYU*Mf|)oO*x5%n0i!tS<6yw
z-BR{P>KdJy<~da2>Qp;z@DYwUI-*{unG2zzD?4W`vxe8Hm&0S!#~>ZcuZcCogU56}
zHQ|l;JgTlPr1tSy#^&_ErY|v+q(~nE(NRV61*s|LZNHp8$+*}hEO9jeYu-IwEOR@E
zE>4m4e0iFzK-0mDg(~s1guSx|aXiKm1Ve8u+e$^qJH;yEVE4Vy?eY|zp0s0^*Ln2s
zG4cJz9`yJqbdNNt1J!<c-O?<PaL&7VfMY038ULi1c%8&IvY}Vf!7WW;=}joh${+;Z
z_>&wPE0P(=SW31P{=<iK6xB8+gL*Rcq#SPw$N8<&=gCXHr&#JOUlAQzG7DwqmQy!l
zgpBEx8d3zJ<|qAw@v*PBroy$(N$p5^_P`*YEE&dTVeDwgk>;AVi7b6P(L^5NgPbAa
z+Nd8=O%bfpN_U32D~j1j3?ehNbllyVGTR<s+UWdTpQ&3}dN!`z27#yDdkEM_6YOb{
z@iD<ga0HV?*xRx><fi{2fV4l?Rj`sx5o~H%@gA@@JL=@g%w0DRBdjN$R#r>y^%=N{
z8k9Lk|5!KSJ6tJN26o48Tp<h)x86E(i@h-1ET>KFg^zIe86pn~l)zlY8EeJuAaOV8
zPogstOZy_NN!xY8nB;1Nuv1NPOSP<v?ilkIEBCu&Nu+69D;h&El25NK-r=yLA?Wh5
zn^cjs64?-d``H^9=5WlCNqS$j##IO92nV9-yOeA4`}T?KlNP_B7LPyqA;jBqZI<(9
z47jz**1b-%XD-`%^r|(U4ZA!}_@H{>ZxzMAX-=dewzsfxX{%qgI6v*B>Om--V!H?6
z`Z*ys{J~mEDhQUX%aRH$^>eSsN6)a`c!|+uR*Z`#8@X!0vVP}{&<okG>LHJ$YmVV$
z{*=cl4TI=^)fHmN6QVO#<Iv-~S31vyUA!mfS>-y?xKG-M&)YudAyl40lnyL(Ci3PG
z2Auxj`C>lGN9J7!9nJHIPS@U8m9)>ID#%031pyhaVqE1>c}Nk+CKY7n2=DRVu&~NJ
zfei}&p=^9j@F0bbN5VLm!!xQ+c!!2hB1zDfa{EwkAY0vCh|L~<J9#bI>E$r~&&NpQ
zVX+4yR+4YbQ%p8-PBQOUa?Md0KBejF!E1d;!~9j5XoAWikEMrEtjEuBH?4CnSN6Qv
z%H7tH-6YUG>U7yaYEYBw?t4Lzf?h4%x|*MD!g}F3Cc4dO0xgV08yis9gnRw`dEY|K
zGGObdE{w_aAFiG4kY#UHr(>=9Q$8mjz=(6aesPKSydn}8rT|m0D$5LGzlI)LkI_1;
z&F{J>VboAMUOS(D=-1F}5>YI5`m950r{vNO{er_5@8$Lz8zGUhRQ)reKN#B$sHF|<
z$uH)EdQs^{&A~plpquj2apf!R9$;wDzIiTW$a-gEQG3nUgbANt%P4&@^I9W?-R1Vc
z)00al)m8OS`OKIC(M^hAu61zFwuwHcAcrWtWtN9*M+l;6U4aJm$~57iOiS+BfRmT<
z6K5aNVYvAS387AWRuou9!mOHJky1JL85q7OK^ck~5n~oX#*sM+mggCMtJH`dF_{ia
ztiGW+SB*-xh+}zy$7CS_Hs7h|>BVZ*213>A3-0UX;??nS?fTsy$(LFfbM;snVa!c}
z%j~CXj7M&e&udE?HmT=E)?4O^m&%~@vkO}I3HZQ0GJZzeZuKS?t}g0ogOuW$2khsK
zd-E@Cu&z4F7iB9Gc^FRE{^w8zB|*f>)~eNp{?*M;{;3Z$lOJ}bUkSN=Bk?KRq?$>x
zVGxd%en!1`Q|}1D0(InwzqlAGJoaIS^!0oeiL^`B4k&f9;5dM`u-8U%nX70lexTcE
zq-^0i?GDAh6NR^z)`?a!+tAd@I6XKcHKZ1^u0dxc&O_g>BoN75mlT+NBSwAG7ux+6
z`riIwv>U1ybc9i=7FJTR-C&Y41Fy`F83Rk*i`i-^+$yl<x891yHA0(5`C6h(E6z#s
zD7j<zhb<<_AB^Y5>J$wv=no*apq*{Rnjz%nyf9GVyJV&v;VF!r>C{hd8$q>EEb+kS
zH{NX8Vb?;A{4_poK~Hf#ar~sj8j{LxsY>Iz$izjCe8ZZ^H~bi+Fm!Zl|9rI$cK@rd
z7Ia6_pOXSpk17fn)_BEQ$*yQpdRD>i+<jU_Es+|>N~VKoaFuQh>yD=@=(kwN9K~hi
zM;Y#}=-z=tNmoM^RHy|QUufNAq$KKR1(W^3_x*<V{T|+cb39kRJNRLjp3(;xA_Htc
z<8DTrJ`(MBwlPBl*=zFPwI4Z!@$xRKQC?Gw<>k<UA_>9F(CWOoeWOG@@6Mm!stBFQ
zG|*H&)zrfYA^k)(tt6AXN0Ql7xS7QJU2_=1sAeaC&YZo_I|aCE-am3t!Vxim`=7p9
zg4+9Dk>!YA)0i^`g-tAJ-DL`J#<P{Ai%yE?9QEy65WR4OKoAoJt?kc!{P>;cL+pXS
z@u|IXMuKAE@?J6i+Qt*1FSUU<qg<V;T}6-L9wWJ(o<ttdTf{2w22l3+*bw9CEVRlV
zn(u@QM_d)Bo=0$8?^$GysHnnH{ej@?D8S@mrB5BO1%-V%HrGdb?r`-W6eb}L%U|P-
z!1E`<kr;6JqSLzo@7X4Ll&SzK`5{ax`r-AryB^b^DTv&;5f(QUxAtfhJFWola>4aS
z_4TJw-RE}^)6LdaqXFR>LQPwKdQL+aF@D|jMF@VCSCZ$+IWLAqYbGqh^8ks2{X@v)
zdTN0RI2xaI-3cshbOY>+1u8D`ZeOVXSbE3M-VHrU9a<EwkBetCy~B=1-;rYm-!YR!
zRy_ze#q$>)w%uq4UBVe2+8I2gfa{0G*wac%=p7{#?n#}@-Mtp?#PLvl+%Ea?!Nbm^
zZw)Vl$f2Tg3KYA;Uyc~MlM{;D)S6<YG72JhuV4a@1>Uf$r+VCy>qR<|$N0_%lGO+Q
zd5!OT+<@Gk_`<t>-MrH{51?2M!A9g{0p37gsH6fJbXCvbuw*h<Lp6`wB-WI8aXN@T
zL*)#Pk`5E-NKJe;lrpZFOO4tAX&>y#nViv+eHTX8aCCL*y%Xr|cH>3;u7JhAH+iy)
z6x-Q(c#=UrS{(Z0(oSr1rx8lm!l%C}U2`|el?0B~G2I&1yP`Ukf_y@wq4zNHDLW4J
zm<-LDI=NFyc;R1^E=GjXwL<+%*X8}7tdYG6ziz_#H>4|zfjTO#0@1I(QW<RrXP_s(
z{v|-L<B}(Cn^38lPJILg&eVJJ0h^B(8WiaJ67u+=CndNn=R%Kf$6J7Nxe!$2P7B)U
zK6e<MuMjkwoYY`*TYIu0p<gAc&-<Y));?pk2s$PVwn~Q0vfd}Uzg||`i0qb%djYTB
zs+EsiXqx6gZyCf2-prLgaAmPe{{dj$Bq0XK_-qvHdRGbj6lE>l2{Ks$PG3+v$3cK(
zsh=U2LsnY{w*!}+xJ=WPEU&tAR)xw6LG~)&>W&b)KFs7+%@30I%LZ>{V!Y4xxJNuR
z5Y?TF0LEOx;JJ_nGX*y7e7dbKg9Hl1B`E#PztG1#pzEu}x3VJkTu(S50}W!1l4icC
zG71jux<6sSVJB*DB4p6yP8k<5q{vNYbHw<r&%k(TLz*szoK=_YqyLcMTDv5bpJTNP
zMb8GjCx^{9ky}9_#3CB^smw1#TWBR!x53T8+nM(_R0WZ7WG^@`Yw>Gv*+smcSPm|}
zXs9^eyJ&{&yJ_SZ_@@Lc(SxuYpHAAgwUJHQjWU;ZMF(RTf4a#4C*e9Jf@`M>bO8o&
zBD4>#>DbhfyGilmHdjvRQk8`WwXv>b5RoEt3@+;d47(Xg&08d~%mK&%yJk(190}hG
zlZ0M?OfErRiIMsdrr?5omvTF}?<paG*-)q}ZHb;qS_UP;RG7GRny`{}FZ?pK;zuYr
z*=2{)_*e1b6o4wRc>SgVrGDrf*@W6|&?C8BY1F2dJHmDo!c2n$cEbA%<W=&41o=xM
zc=1>*pl`tx=aA)fVT7t!b249AKk?c!g4R>pMsAL#C-L>4%HtyUKn}lZ#ET<a<$+En
zRgc<Fa97vzrUJkH^w>wy(F9@d>)Fid)<<7v1U`ei#o};yccs(w5KNXiTNkPKm>ZZE
z^D{7Ta^j^-;e_@lJHd}(ANIpv%>gHzd($}fDzmx~2Kax#_LA|t71;@GFk2S(7v@6j
z=BPobrc}2ZUk@D43oSw7EY~lR*DXSiR$5fqL7#pt+TxAaB|6WJ{=mqEDAPFUMKVp|
zUAo(7Sx|X0aE1vmsu`ruU<M*=dwka3!2%YzPWq+L9F6e8(r2!++szeE`w%(wldfcm
zJ4gFQhZwfTLXgn<bjEd|d!OGp^)AOzvTSsg%k7fXyk~_Bmbr>_3DGm&hc$(vCdPb(
zwQVJRk|{5y#Or!Wl4#u^h~V@eW?n<}GSK^__oPQoxuJvmLtd$DqISn+pbulaE-=?O
zk~GMGGPgKkmg~n@p&7=0V#zFWhUezanU|Eb8>}BsU|$|vzOl@7r3Df3?4HIT>M{rE
zAF1o10~|Is&Ok@J>*$uNym}-3>!8zKy7%7iR`~y=n*gaJu^e4(NF?13bC-b>bhpD=
zBZ~R4S5$&8QDbPSWlkPO+_MC+GNDS@F2);vc+GVEun@;0zze1H2T`7LJJ~9SI===}
z_J|NBQPI<Gvl`g$<bYwwA&^{wgUUo`I`_Sgjb44G{2pPKmY6bz$9ldZcMcLP2(<WF
zDVe28GOezMWQU2RgM*ilJ;fkGv_VFfdVl<Mmz~?z$|ALSNZfHO(qK1bPC?~Aoat%y
zrMl}o9r)l*SDJEyQ-7YW09{HT&xh#KUfUHe>k38e3jv1yQ<dV$s6MmVl~fM(BL;oR
zwoCfqmKYVNM}J|xL|5@d%HFS$a7yS;$6G7z(4Fm!^y16FgSYLguS4*3h?0c<>q3gI
z`vS`o$bi9V|Lc2QtKfnQxwwiQ_m@m7iNzMdGoj>)gQNVm*_D|R=(=xUD<B|nql~-n
zE}F`-5cm@A{?EO8Ze1C^qv1_|SIr2im+zKMQ~=TJvhO!&U7VUL>{JNac)CihjHFW9
z1+$ncgch`nUSKhGLuj$)+upo%yXLu(LT=9ZzOGFKEq0MTFSbECGK(+$M@e*&Q6IR_
z*^f0c<}0Xg42Fn8%6nZgG^Vo$``6#QsCG;5%iN{C{{`x{n7-GP%obJ0jEj&M5IFa8
zOpW6Q(toPqM1%`yh)31k!T)YvOm|+r(#uCW9c7pJb$8witsph}nlMduqBaE6%p$xq
zz9oOkLt+y4`-VI*&z&}&3#&0qbhz606`(48Iy(hvCJN(?cWkA`2|3Dws}&MYkb?Su
zJnFR+)T4^`{XO<O{_GJp(q`@P;QD9G0Y3C0KJ?j*8D`ZK-esn-+4qHzZ`r_t=H^=#
z8l62@&PgrGVqn}n8pug%-hyG{L#y*`*Uzni580He?;FVC$K`g$8)FV*{#9<qb}a5s
zxmo{%EzY2@g22SZ5%Wn4B;c|Jmoon4<9$<6Tp|87$?}nLB&usy?CI+-=Gcr@FEIoU
zw*6+V>3sk)RKlRqVVjjPye(F6F<DIHYi+g+;%vup1_-xci1?PZruZUnw?swOMb*<S
z_Q`xsu;I#8$E}*0L_cd?m*31s=9LZ%S&MaCsd8GA9|vj$=fQjBFp=F55Q6PlwxcVk
z)a|9<?~R%i;Q_>e_1>!+Phuym5~O#pRYy{hHoKs8_|RH>XwwZsms-HYn~7vx76Max
ziOY+{al~Q;oA@X)3Q(h`wlt>k9p|qF7$!u=N@m|$eirFuz#a3TVSyOYk}Y@C7Ax4@
zvg;Jrn1@!iEklpIYo$um`26*5Bgvrz`<#jzBBObrD=fY!3NZJx+<j(UsWu&}^fEbT
z`86RlU#O7#qQ{QSo`Q4+=a_lb9E7lO=2_Rv2sHR#e*3vb5QY>j_{6Q!`!oWkH9V&J
zBe7Wb)tT#a31_d!nSEyqE#}P4Xc6z_lbvznz%Jc2)t^!^UK!}xrmhr@iqJ6~d8c*;
z?Ja;_NX{8$j6OVK6vy?FIQ%6}WLLl*6vyi>8uE5D_^NK7jq+AZCO=xiS);!ySf%vm
zDn{m7F8A=(`L{+-2V)19U2Dqu2d$s~c;Vl_-2CJ<=M6oKz#{&<Kw(NV>w^saX0ZxZ
z5U_EXzUQ5CV_lPh`C`G|x7SFAD9t|*y2ek1LYL36cU7PIjh2fAA<T|zsi6PTI?$u)
zR}JyC(@w{Jekxq~z>OcK!w{2td}Ty%nIe*5<d#G`*XaWKD<PM-j~s---8aue9^FQC
zNPmNF>hNFxo&P=(|A53Cj-J=d`b&4kbinnrwbJ{WriiLV?eb$NVfqkecuP&a0Ey(h
z;4PgJJw|K0*H3SVTqM8jnn$^t{&A_67jV-Em4oNl!K{d#9P_JUHzohG-H1yHQ7eqS
zX*d5>-;Z8;)#kD^l=LI(s3bqYFRUYLm}Sy#{Q8~9#6%2j*vqC83nj0BH>DVK7AfO4
z$`wR#t%7avE7TcL(__>==SPs6eOXQq+Pzn8roN5+Kl?uXd_j!&;d-3bul}mpRU)#S
zTu_SqL`{6}NKSv_drf3S7{?9%gKxKI6O#V=>e$MOHuMT+{RP8+(ZK#)m(LRZzUgS*
zxxWB%fizbkAP-)_tS)<AZI~(Q!NK;P{vTtC(klE{RaeEt5C7&{Lt-V*)?GCs`j3zE
zKk1^l9FUK0AGt&IzyIj}w|}Y>!#{elVA&v&{YOCeg#dv9Zo_i7Zes!j(Q)UKQ@?mM
z|DIm3??90Zyk^-W$+mwj$?4~h#4jSKFAcl@SJan8(Mf0^q*E3?B}irU0A}hzUs^!a
z+s4mOdu~*z9IJw%KBjS?!X;wK=~JuO;0c-vC278G5Kx`kYWdxEgRI~#ZfHyw$wzMe
zT7QlE_G1G>Zi^X(Y{`I~<>8aVqz=M_jiiPDsuI&Krqe%l7~924uiozuzS(lP;guor
zi?G#0nZN~W9@!MOx6d{rDT`37PO!BeKcaT<G2oLBg>y>Jt-l~&$MrH-g@6lQ55`H_
zZvP&ucp<m1(C><DV4Zjk5(oD^Ig|pWs$!lo>;yFHU8;(kA6fFOLH}0Qua`aE=;ouI
zskcJ%n_cx%TO^cfmU2nB%fH$-&NTl*08=H6jR8oH0Aw$H1&XvQ6^y?0kaKe#BoZZ{
zyXcXwqSFX^lLJC}7~DG07ipUdfIR^8O6R!mY3WzTYo#xcP_R8fqR5~<=x{;6f$h^D
z7d=JvP9exIH&yu1>c_RYFXiye6LX?-dmTV_!Ow$T8dIY}!W{q*uf}6k`%F!rgApYm
zd-ZbbWBaHN{ugGt9Ft{czpNTd>Rrpse#^3WA~U;Enf>=K^ByEDEO@Z-KKZ3OeI<SX
zRr0AbRQyhNFaf9_&ijp^A2Vhy=a~e}mtJ)iHj`3^deuR&f-O`~QP9*QMKNNgKDDFX
zD4s%vC6@Y;TI2vil(EajY+rXEMQh$48gIdHbCgHAAr)XxS$&%Mbywj+?jHiv$8Sd+
z@BJW<)pTX?e}McEL%Ww=g!-?C_ju&UdF)uwJdEY^CSWR{&<*P<1GPiW6Wl_*!5V&g
zd<Mye+-rvr#`UV`>4hwmXVun>+k5L$?g7K&@kc5Cja=a$aA8IOe0cgC+WA*^s9sit
zdY#Nc`6FO?24JLYi%?g8^zvJS{IXRuN=adpHS3KhdyLv9q7ki$pH}bS%%}oAF|M_4
zyFJ%Hn0-yvZ1DAKq4dfovaRf$HjODCe+#kJ?~1!hQKG8pH9@yE&4gDjudt`P*=TG+
z$=&KxjaZew<pRSYD=}Y;?YVUF=Tlo~>r^H8D9<XTh*?#Xk1N}VmBuutfYeaoS@H*2
zk9<Yu(C*Ut866K97Ia+uR}yY-SA#SPLJ}c8`v+z8k!AiB!LdpHr$U)R$-IaA(wsvz
zzc61tu`{xpHnYei@q=>fvXJ40r<m1S4ql~tM?St@tY3FEV|}5>bS~$;vYT?>1~28c
zw{=Ge*S;uk=L@_c-j(M6U`0GGrq9eKb}o6=5q?DviP1~t@3)z*DplS)dM9CL!$ob`
zU_JC9`wQivlLk~7tPcP~&Ocq+7NO)g72;p2f){2<BBbdh1OKZ{^BXDi3i5;!j{E<0
zn*RF-rREP$s3Gq)!>{YX%aebpxhmRjhsbl=#8|zJDhI7@g%9GoZaIE|zYC!a&%-3`
zA!ykb6SRx_OwKvhk!zN~1(GH+;9tWfvE$@C&_r2>X%#=QIRq<GQXPE@@1{Q8Ne?YU
zDiC|<u*+Z}E#oUV+eZ*bxKmKqCNf_EAxS12;XxOGllQm^kV6-~0hMAJ36)2fed~as
z-)IX)0^z?v9siBv<M~mw6s2}?=9*JDY9zbo{S$>4?&kR7dwm^NrQ_lEi@!P8fhEcj
zAq^UFAcnVI_R5VKKR~o2;EnqbvvSlV#UBK=_Q3XK`TU2V*7FN+zk5lwuJ+y`|9{YN
zCHyBH7t`clI<A)=#5|(+mJ-vj!f=Q4*ac+pcAm1p$R1Q<TMQt6b3#mNrLf#LExp~5
z!u}gzFI*!>LA$iQIBV+J&Do0S^u@YXA0<d~<6+KRNs1_LL9+978`|368d!5@xgnP^
zwTRkg3b1UE1?t@irS!K7xyc;tJ_XYYxfb@ko#+Bs*>nKPty?4G1F~8jp}sChAX=l!
zPL{-Rg+9!R+fhBKYj|q0rzJ%_I;@jldxR4B9)gOBznL2MI{b>>bfMxjHrwww_vd%L
zR8+nK^Cj=iPk42K?%J`2n9-}TOL=({;6^gUD%W{hct?*Fuk~A|HqgB!W8F}#B0s8)
zD5~rcMb*4=n9-U3;~!Pl%2j65YF*R}XRSe)*Jkes_{q)ngr3B;OU<m=XDS>W`^_00
z0bsIiPbKb@e)taK9$~%e8^WHdMHZLCQR!Sq+`tWF>M*!&TXC>Raa!<V@PJtV`gNXo
za5XwYXMJP62Dm}<>n{g)CV0~MbYd3Xv{C)9UywkjG{tx!{bN6%UU8`ot`$QH4;N7F
z>SbzV#}dh*tfWqN=iQN9hgniNf>!tS)6g~!D|DCHpfP;qbr0dAv$MB97S~5nWEXbG
z5Ofb@It4S`pdrKl5e&0tBHVItu-6F~c{}%C#$FA6XAM2h`_-O@a)2LC*i=rE;3G!S
z>4Z*69KhiA=$wLiC2DT27Nw|VJrGZffX%Lz8Iu?6rG~gi&?Lmv^elY<Vy~EzIZ*4=
z5690-QEC~j1cD4*TBRvNUWCY$sqnV;M0RX%U4H^1eF|jQ34A2ABME)_f|m~fVS|*_
z>=$1G{wCzAbRs!UQ{R2lcs;;l=RJuMz^k~~KOAe-SW0_K*U*k{JAcX!r(dgZ2ijnY
zyv-iC1kmdVGCof1;G2vc{SZL4FxmAJdzIYt)GK>Qg!$AEEb7sTs5kbA@yu?bY0*;p
z9#=f;ONuMHQ@shZLA>(I&PW82rSbuZRaw!Ces))=1I6>IHtN`8KjeWivGPP;*42L4
zNunwD%n34|KSLOiOw7<sW)HHu8L0;WL}Mz<L-*r6m`;l2YnZY%iL(kdE&Qe5(C@!(
zf;v1WiFBB#$`O275p}C|v@A-483M{YmRzL2`t~CT5V|ZKaIH{U%DhX(#xk#k24=y<
zM8R$n_KoG;K&{M_yvrJJVrMye6Aj4p-+_^oYL!W$&$w*Z?I<-5=50M@2elEw3_+E|
zQRtU;AixNPIK8@?3!KDXi>xgw&GtG7bf?mYJo^KJF{_9w)4eLB#2H-NHpoS0!K&ic
z-J}6_ByN?Gkr78|5!$}z2w#{}RzE9P*0|X33Bh1+N4M9tQ$T-#b9E8cxr6f_Yxjge
z6VE%3e^WU97q1K`CAN?JRRM~;+K=LelyVyMadAkIyjK0VcnBIDYuJW8kF}PA^Ztph
zU!D}U-6)f>CiY^Gf3A`&_S#X@`j0S%XGc)ZewUaoFdvmdQxjJ_l?tK|i9V?_)p3*X
zY<~S~>*V>h*>aZw)iwM$zIVrca&~^UX}0NlK|w+Jy3n1TH57V(X%+Dn*I_eTXhJ(M
zr{joGXe>Pg_f&qXI6Dq!5De<#*Nv5hA3;ME8F`6quDh*__&dgf!CzV~D?C!YIAa4g
z4-=!FXDGBdy-7m0N=}AIlpoB=8#}qEjZ72?DW9Y2={XV%jKZTb4&aiAolxjo^bGw|
z!BwX}dp+=uy%Q-B;TKNJ?uJ%kV0h5*rMs@xD^~vxbPxl>C7C7`ZS~W6SKCz*cTEhh
z6@OE^X=eMLh!sa{f<o)jGt|xoKgY>c$fsq;q!n3I;VkjT7@kWvjl?ggRwCoZ^>~{P
zXVB=jk}8}TUg4H`mHOe<zN?M5a3Dm@|LCQG)81IdL>)<OBuv?~-8hUrjegI-@WyYT
zl50^bABc~e?&ux$Iaa%*+#VldDl}TJTTiX~$(HCUJ;N(mIW<l4ULsZ87WFC*R$*~6
z^3DlsJCid-vM14o3?}<Kw})p-lip2Bc9}Gb-los#Pr+%Ti|Pth4N_gBr`YH8N(Sn&
zaS>8UZt|IIjrC%g=Akwh{M=IdyTQ^Q31Hh4*|%L3a#JypZ@4&YZb$59coeKeA<;sc
zb?ojzt}rK3H<TNjoh<EtqHR38b~df@i_kGCIj+da)7LpT?2JZ?6jggQ^-P;l_qsU>
z+x;!IBAyKWL-GE1>r<qW=&t>~zj75tgy}9Nc<N8bB*$fZ^#F3*!}e(TP3^^qOG-vW
zF_rlvnuRorIQRBGuD3YtKc2^wHs%v~z{YHMe`_hUP1^G^j<Shg)v%-_>D1!$T5?pQ
zR8|oAxeqIap@CvXmkLl3YdKlPVr=bFd@N@^WpU9EJs%+t>rByXdA-t|B7f_<5nTbx
zYM80CrG*E=IyDM!4b~CmEh5I9pg$&8&a_eRJ-c%NCtp21c|+rwaStiF;E^`@TfK9O
zJI^O0wY^p_QHP3a#O{MM9#`v&%KhN<{T<RX_f@9nw)B~4Ty4)9YGr##|6S->v%HmJ
zVkO43^LhD$#*}?6U*_DW#po`Dl{98esC#qHxakk_e0!)`XJB+?H9?!BK>lMicwAni
za>~ObT9)yJXXh*vhjO*UjmHwYxf)9RKM1LM+7vvCj-X%XoXK&&IH@;@>=+A*^b@SM
zCgePM%t`2ik(e$EkLS60kG+ynJhHVqxt1TVXARsz|4OX<=SKf8PwMd&V)KkAr~XyR
zx;=m7b(H)1?FdX?SGgVpTh5XZ6ABf>lj)p8uC}ElKfLh`1`9Lp`N!y3J&JVCcnagn
zEpT%WU17*!6d0B@uS+{Z9#dK3dnA>~O<N(<zC0`Wy0^8DR$`DX*?8$GY9+GVF`P>N
zcUjl`dK%xm6}JY$S2>9VtXQV*cjlXGqhl(|SaivD!0BYAPT(F-=`CQBNcjfb<35e*
z`2Oee$6wYb#>=<HeCEP`f~op6MKRrx&Qj(~+|7Ad=noPC<ppQLiRslkH#4~G=UH|!
znT9vNU!NX~m`T#E9t7Kf(lbN2wzLSG6X_wN$9K7F-_2Y_h(Q(lR!4_3?$Y(E>;+LI
zU4$myMGd!CTY7w}jr`UitMAR&DB?cL*Sg$cUYLsm$%od&1p{}#32OG25*<vV{Aai2
zV)%L31=QX7<GfyF3>Jn&>eYP8^5B2bWAD~;{l{xa%$m>W4<AyTt4@i`Sz5TCbfNdn
zy#FWnzq%a%@?idMPMu{0EtQQ)?XTT_h4_H+5pudKf!=uto$dZN1y3|__K2d<72F@r
zqCZm2-DYf}H8eX<IZU2C<#ryXMEWMg`DJ!+J52Yr19?x>l^b8E!_8iw*SoeN93bH(
zqKvnNwgV--=3>OnYdFPxjV6};(YirWH)(irDbY4=pme+u7YU_z*{3GX)>+RyC-+ZV
zJuI17BgWbE8|Z`D|A)P|4yY<^`-MjbR6s>H0-}T>-6$Z9BHc)Ca8uGL2uPzKA|Xh3
z2+}3pASFsjw}601OLv`X`Hb_v=REVweDC@9eE-cb%x3Mi?sebS^}Bv?KNcXK>NtjY
zrdpFs3!G-8M3^izRPQ55V;vAL^r$WJs%H8fXN799mA6T;QaLW;IA^nGBRkq{6WWkg
znyGLp#3g%MjiRdUX6w#sx8C=`h>r{g)wx0I{H*O_`_XheW-koJR;LGy$=A9~s9xR7
zP0sJ=E#O1u3@szCpVxLtmN70ojczF=r<E=mF)x&DPXw|}_ISGA(hrC;vEX2s_|H%w
z(pJsh1<cRs0+;t6C~Bx=;@LMIB4*LKYQjR`mS9^B_UsRg?O>a%{3ubM6c1{~+Jm*x
z$sleq;TIs>$Qyb0>ZPE5yPVSXfvF$W+gilt{$zY5(ku2asjtwBN1Q$ENdCF{d!s+!
zCZB<R!JcSSCZBYww=lGc4JV7-;p;Oek0&@iYT!0h^#qQO@)c>j#EeAFoRa7MC}q<t
z9e^1*qbmsxGF}%<BD!m8RQeBKzqoxib&{8@tpZcy%BETtHmQ|a*o;CvgwP90mAxF@
z(Zi+j+FI4*knzbf7J6BkXe2bXQ41g{IaQ#pl&+<2T|X8}HL?80-(g1dYKs%tJwi=`
zMq^-V;;5c=V0$Dzh{Thr9dM=;&T9;Oary$Rqp)l!dpSKRcd~rE<NmHcedZ<)U|wyW
zFz&y0px?vpWWlB}!{L>f8bK=HWP4o<O|*bhHx3w8oN-H=hTKU}oEsdRW>0cqn@<%F
zJu^jdyd(bP)1GZTZcH4J>P0??^aAInC#}Lt>b_0Tb9P&(7nfF<yk4giyWV9GfeA0I
z)U2(c+_`$o(7$v5EDZ^72CGOG&++R<;xbr9^Ok&+<fd9pn7<p&J2J6n<GS(oD{BcO
z4o1^@-b!T#G3l4b&#rKu&qza|+d@_D$Zl7GjW2D%w{-fa3wFru$q?H+Qq4=|)w^`?
zDLVP{0A<<{63!(}r0uerc`frHdiek{v>!|-l)mAh>K->ra9y)kI{=oR!n>1yzSw{9
zy3Zm|KW~8s{iUB^)mf~QIk<#l?#Hgg5xpoJ$p??VK0Dn{y#qGh#yx|aad}v5U{nPu
zGHPhri(aV$4<Mw+Wxln@?ydP?+33qzP!aujL{w0tEdB0HRZjkgD}r&R{Blkt*zU;t
zHe8vqDb}KBoA`4NNV4fiT|ETzMq5Aq=zt==G+O{lL-|3C;)7*be3_BiI(!;YQWa<h
zB@WMmXPZ{o09{)J01BA}G;h*}(W=>iLW$Ersb;1!c$4cq_t&Q)!u=W^5<U;iaE``+
zTcPjATp}R|vYKeLhMo`ggEPvv6hip+e?9<t3)xAdxU0n?krxZ=XUY-S2RnM+DqO>q
z{+K?N$v|oai5)1r47NH0pZ&CE2YwYhmppE!EGHb2h8xSf%4=1(F2hE(wh+-6i>7T2
zz+-xMr+u!g&uYwVe`(A<7yPx&Q7R)jxJNBjFH{Wt|JI!2H!LZH#(K%0>W^nOi$3YJ
zN`dW;UJcDY4cDQilEmQUlazE>tH-mYpcmwCUqllV*-%Bs^NK!~sceF@{bQJOQU~Kq
zQlt3gxcp0~FSqy>bel-5+-|z0w6pxx6W?CiowsJvg#KASJgQlx2=d2m>~ra1n~}9;
zp^;(+h@}~$W8gkR&{1j9l(CZ>m?DQ~cgn7Y;^SCzxSA@GQ;9I?W!|I5;1D+@eQe_Y
zU~pqvA(|b+?5@8qOFb-#w_Ee;DV1`dI`Mk&ZCw3JFF2VB5y{ltX!!!d$i|COrT`$)
zY7M9!GcTl_flyHM`Mw2P%xN#b-A<0^XuVB9A<{(xh2O$_w|FuwjBdvnV1&$r<gcuO
zk}B^{-BvyoN1VBCJFKCm`8v`Hgn--jP9I2&l)ZeseNVPYvsM>LS1Mmlp<bO9#$#5d
z6M1D)T(Fq6+4b_Pe73%%{F1^F4Td)8)d7b=!La?98@i^q+`3m$e0uANl(^kKaJ8uc
z!NIK6ou+7fn-vagNWI{TO{=3;03hrwHArqf+U65AhAuP1-7j`6{IJhCzL}$N_3E72
zrBan}c8WQJ<fMB|7A2&<1M_yu&4DwsKn93O*Wz5N{Oq`?xo2_+-~Gwm+|NqrM{T^D
zl@%p`Wszg{@Vvrw6b9;GTV>t-aJwc*f+O0_<UH!Ygm<+_`$5MCX{!k9bJA(eynhcY
zFCmLa;DO{HsV|*eu!xR<{8*tN(ch=9!osWo<Yok>Q+%ei(rQ8<<R_wm+7qSC(g6XO
z(M`y=C>5sB^}w<L>i44pNadkclI0A;lc&-MekmoK5osuMPg=H`;bPm%i#LK{xHp7t
zAtha3(oF8$7RcAZzfJ}3svEC{0mh>iOj<r*+kvfkiYD(cypRMqT8N}vKY-`E#5dLx
z_u@moN`{ApHt#4s2tynAxM|dEbW3Bf0;H5yIlaCkhkwsj3xg(;CzfeTAQnlBd^~nq
zWS%Gynkov!M<{fJQjOk~f^-KkY@-6d_U#aC;s<P}@-)H~&#%fys4Xz`oabBYY&uR3
z6wD;m4&QFP<3VE`Vkf{J!O7qwtsJojG%2%Bh@}$Kd7f{tr4BS05q=X#Sgu>Na5-OE
z6?4@}{n!zQN@}WR|7w#`ERt`Qhu7zmTIK?{l7OVoY|?KtPzI)>9*EYB!(3&~BRldm
z&XK;4ypq_A58lHvu?RJcUD)I9jTit9^}OC>lWCdv)PEK_J{T)$`UY^%lOvQ;YTR>^
zVWC#0LMnzprjx#R&D9WDt7gq5Z?GImO%AUZQ(H)e0mdG4+<v}AEa+WST;J+6u|;dB
z@Y@OMFWl4Zt(CfvzWx=419?T##qz~HNUXM(Htu8C2dDcRV2_&L%yd(7b+83*%*<HE
zVx}2RShiC|-2)<WIR*ZXlB`z=4>&agZssKid18Giu+3Wd4&ej~lxD>#h}>$#Z#Z*U
zI;aOoeYbqGYUs9}paNSa<0lucP)lc^)7I#g2Dh_nA&+Fl8f|+hoi@<I$rN{f;qOlB
z6K9GLtbZ$?+Taf)S@X<Cr`K56sE17n*{4NIVKG$JX7_zM84`<ig*~8A(b}ogo7{;m
zNQKVlEY*BSQwE7)WZe{FY<8C9%S(aDCi2wjExb7Az0Y0!K0y`Q#Sw;Dr3c?QnUpzt
zYNwXRilF{1utlijv?!MIFwDv!LF^}4dKG!rVwr53`0YzCo%VUV7JH!fz*<yZTB(Hv
zN-m#9d{~cA_wXeIbFo@(RKousml7#un-2-GWt*H~q9z~Lf<*)zbNg&0-qM-*kFvkq
z2iyF0DJ~=dQ|fA$n~xOrHKozHG+r`(hj*ofVGn^N_nFXL6(y5{P<F`WCuyvMiYuJH
zAWS4kmH4<$?S@Lvl+(K!HV&20vT$x>JXTn~i$<obIBl2Am)kBoE3<K=z;PpVM^`bm
zW>mM_OmUd2QiuunVr74wknQ(rSgkE2e2BiJ8G+B9i5U9cSWV4-Z=!)lb17>SFl{AM
z9HysUDE7a(*|e7q@i?D*WNyh4n#E1!2VP1*!gX?7`ZMvpLI&^8yx3FsitEVEMRb}h
z%Kattl7}-e`ctR=p!#K&%1u=rO(e_bbN8Y4n}cuVs-M>*wHRGxqnIALO%^EDX%|k@
zt@B=iA=KR59^FD<w40V|{vHF`yw{Gy49yYF!$Qpxwy9ps7&l<5#+XwSoUooB_lo2e
z4ZOyt6*|RyURj$;k>V7tMGThq$zf*dT+AhZUs)UDKl>D#jXqcRYrNkXcqY=TKqm>m
zBVYFV)qW@@btq4ali#*{z{aRlr?q#QphZQ>K=_$Xf2)VjiieL9!x65TYPttn0?$x@
zDKecQQi35H%8a_krxS=yhURE(#SlZ64M788Wpjk*Ae4}}twoV-K`EYsfC~Firs#rs
zW`pfYrX3kjjwS$M&tAKILWJ#^aG3Vk3!F#wP}L;ilR$;GOJ5Cmcw!4{weDn-&dGok
z@L1p`AQ*QTlO-h%hxilQ&U!lnd#u6_L^8jj%%wWsy+4V^eVU*Ro4^#CfPDG`b24eB
zc;RGdj9WiAT*cqs&H4f>?=hRSBDsu3OCy8UHVq99lH0)Mr=%rsINdoS!Bw4P5#xI^
za{4_Rj6_(EPn;&b&}+G1*An5QR4v&1LzaUC{Ntu|CIvg>dCBr`X9W<l7BZ2@D9n${
zxzg0*7@Bk)Hc5Z^)^;VL)Onn4PJWj{z^Q1!&*x5D+-DpDPp248Mz^pMGmjn}z5L7@
zj8-kn+DB^<I^j1w(QCLgr|=kJaL#z5o3lex81!Pe_U^mBeD+y3232lo(s%m{ludQn
zDJu0Dfy&yiE&U9GI3g2sP&9dhlb?}5Yh<KTmo0mW;NGdi!P&z!wB!0#Mosd^sJ^yG
z5FNL_o3_rjq})8eFS8iEm8)U3#UJPcTU?Q+L0iNgPT{4iZ8q9$*UdIq9+cb<D9BCb
zbw_ZK<>kI+apL{^1p8un72y@%=Xp&}Zs?p;5Gw8Sjtc9K9ynhtikULw!5qu@UzTyf
zFsd7GT~WQIX#Y2~Oql8_T!Eo|${VWdzxwTHY6YqtGcJKiSed_jKE-l3#Bd<3td<T=
zR2%Q`dpZ~I(EEB13t!Bo0P_9VL)KHX^lKjLH9FM6rdL(a>0h7o7lk}D>%7R+^>enG
z6!yO={#iAc+iqsfQx;yBy@h2ytW+vRl4CU-+EJIPymV{##*I+|?&(k^583>M6WSo_
zUV+-L5_`mn>A)_*Y5yL~;)Nat{09{0kDlTVCw#*`{=q+b*gJ`*ag2GzWbrfR(D1Y-
z9)Us2ZWm|McWAXnJ$WV+z~#%GDse7_Q^3+TT5umiECXYh_D?)RW>jg4@<yAfF_yT+
zsUmXTOF4G3Ls|5;oS5x!P0Kl@l@A#UoEY0=rQI!69&VPuuYeAbW93NA=-cIg@eMjJ
zVsX5hdVT#~JutZ+a9j`4j`dZVtOY9uFA}5J{2MJ&*MHv9#KJQY>U$vx2UIQ~zt>{J
z)Y_8uSOFzf7bXSs*(a5*{QVQyI}5kNHI?DxnSWiM&ew~VA;mb7O;<R=qXVLmbipTg
zc{1=1!dD{eg&imW8emv(nM$jp0vbmzu=_Ep9(wU#PE`L5bN{b@OhF8^aK%v1{^9rj
z;`A*-5}dZysO|p#tNhgxyF(8PD$FBD<gfnYKU~QY5@T5HB1!gl43NKGYznx@YotSU
z@&DqAosmF6R_*(L>$v&XQ|?eC&XWGJ<lkod{Ph)gYC%^$+#1pQ=RtKR4<!*yHG6wV
zKV&yg?N(2H9bXD5^R;WHSI>0j4*wQL_2+N?r~5dJP?R)Rga71e2#9_~%sh6$=c$s^
zGBYD}esUl}poOIw=1W&$;+^rX5heW`=CPNE1f0KsLv4XP4%Py*xbBIe6VFIb7G;sr
zIs7^y|N3jdPd7~9o9x+czUKMY&A)Se?HVH*<u5C^`>i^B0<=W@q@X35029r?nd=%q
zDqz6e04!D~CT$d%uP4(y=M-EKV*ZMZ=^wSRH$MkMkR&!~NP#*>8)qB5tcHQf;IuB3
zsmy1`d-&#=#pP%GrZ%nM0*^56-^l-Ay_^++3p}Nlqh$EU1%ARJhq8#x?O=xum^#5q
zxkBYgI?(~RlHh@nL6o=?yxI-m%2|&Xf3g8Vu|Gip?>hC((07QMPZuW1W(+9>t%#X&
z#aAn2;Ee6Amr?~eWeXF}PLrR34oBuPM8+!_7>mDkYV8Vm7-BesO2pV%j^4ld4@<N&
z1KuDz)Asu;e_Sm->h*gh$m#Jj3_f|u_czFlj`-Tv+pkuT%B%}9Q)iJe&>)Z%%)aN1
z#pQ(G3*SSs1j;rWoUo!GHTK~k<X`Q_e$DH$>1~a8^n1*2OQt#lLMi{W;17e4EmGa=
z8_6GAWb>It=u#)3uDiSfNFE{LK)oLdpPSmPhVA|;+;inGHj{sUb9j-@mCIIX`4@YK
zO!iLChoc#xV;l{)B{OY2Xgkr!igndoP?BNy+GmtGOM47!v%BGY-7ffMT%DAkYoQP5
z2Fu@1kEc@B^TEicY*1cl|2xmL<eUMft@L0jUr(^8m;I8w(DXgsdAoLv4}R0W&4G@7
zyRCdR+>8g%|9|WGdIWITS)rXe{0b};p2nn$12Q~%Zr#lp5!$LN#69<Szcs!g^4nn7
z!~{0WrS)KVx~&^EM)p3@=N>3Gn>y#qbP2h}ZdBAaB}}$tu-rJE^JGR(Q15_4JcVps
zJ?m8?e>2CG_Ipnp7cFZ8uJRa@#I0R_vDY`X)^f3x`Mi|)$dV1O81FG}KtKjBz0DlA
zK~YwLI)nCJ6IU0-h|0ys2mkoGuffLEI^*#Nhb*xJ<pA^M26&OaQU;0=V>p;<b|xnP
z;kUta1__!A_rXd-gA`-VB?NJL1eo!DZ#LXIb8ZN~YY3KR8A^G8D38m4n_CZb;fpYy
z4_ble;Elx8=mJQO1H83pd2BvtTkI}=g{~KGVxab@7*7_4^(89+(!2n|M2+;`8aO5i
z)Prz))F7b+%wDX@j-S-5$Gw99ndgAy@F5B82!?mJSzYHDc0Fi98<4E3OQ~9Q4{izb
zfcFmoYL5V|d5%aW2Oza*_+9UPX)i=lx(FV6x&VUysGo$^f1}C`_z@G(9^qXze|J~i
z_>EUZC=eHC&Vl>LolygYlkK5Xw`+G_4LI}uj!=4L7?&cJs}n|kc>$D?W4pWWkBnEH
zdtle6G65fsH}&Dw<Df4$TbT%I_D7drfOW=gnOp4~81-){L)qH5rY{^rd-)L`K?A+<
zSjw&epx$4~*}eu!G%}Dt3M72rfp%<r?@JR_mN%)*OFz9rM~kfkiL_*s6xQmE`$9h%
zCf=aVW63UPVbFz{JVf!m^ISjdNjUnZ2nwn59j&lEGDxOl7&>4wucZ)A&2$KnQDnAS
z^Q0=6Hoyh!@P07h+<r%ScL<WuXPBUYqlkwhaRt~@&!OMF&IEHTw*cvHynV__gsGVv
zwgKsKwgG0ii=b3#ORH0*NO?T<q;@}!6;R-<dRvXru92n8v*CK3I7^v*6;y(MUprm6
z{1W!4bB!8lZ9iZ5&O96m+d=?sya3h_MCu|I_25}Um<F5!aaWd~50wrtRFN)DLr}VS
z109{G4fMt=-BEx+X0{0FHJ%|pCH1jn(wDKqtKxXS_W1F|U@RUmY^%<bR*$qtG<OYk
zrzsZe@8P&p2?lYqb{wvF3!O#OH+?MV-Qf4jV0Mq6S9|1y>~sQ0;2=_OUfBLUsR`Pz
zR?#P@V@bLqB#iNqqT|wMPKMBs^m%~S@T23^<xakBBGP064DvM!?FuWwK0g@irAaBc
zY@L#wVZ0cQ<%z;!SVq%sq0C+3J?Qws@Wrx-RLU5tJ)09oG8<uPTGmH^_uFhRJP@lx
z{A?yCyTCl~Id!;teNz&O;czq0{TwPs63Y<~sItq2Z23e_IFtes4i}9k;O*Mx;sjrH
z=gQvl>l5)z3Q|}O(D%QHkR7Z5jpuWCJpCQi`HIe;D-1gVBR|6*LLug5#MezLW&mZZ
z6AhkZ=!zYVIiT+C+r4oy<+B=&<#gkNC3j$*t=(h-Dx?;1!C*$!I3(qe|0u0J%t=$p
z3lj+JeLoI$E`LHx;2MqMDi|ux8NB9JbwS!d;&dW(OGw$a^Z34&f^)1vIGX>JmFt2g
zi>e#oS(t;gb%m3%>}C-QOZF<KTIEdCmNFl9|BVsK&pv=m_Py5d#IK@o)*Gt?s0WRy
z)iQ0F+H3IioN>TognT=s7Cj;@!#{OoI48gUng%)8B7&n8nAwBW^iBq|?6=+5(3>oR
z+!wC70@kH`m?h8{3b6f2*ie;{@ee^buPUpsNww?4h9xB<5gO@LB2}Jw(trzYDU*E9
z^OsQqyD@r3!qtt>-9CZr>j9wzsjNa+qy#;?v?*qd`1hC#xvw9e?U^m;viX)`b+V<A
zX>fXJe$q%-j@8pgf}pS<`~he?RdgHcnXb?bTM^Mb9+6dfj=DY1?2V?~xEaa$v61P>
z@qFkq;PE4eMzPi(8lU_9u%eXB>q*)gY1w^K1n^UM<V#j|yci#!2U*4s;H4Q$?fqfr
zHr$)7vR|y5^u96sh??e$Y#Yib;!vJ{RYZ@7_^_&@UnSx_vr{CYaECk>N#U93NJ4qI
zq=$u-j3bB#v^Tq=RYU<q`+0vt%$)XRSJZrZ2pFroay4{^OqMcf=OZ8f`+@xHvHWZv
zp2<;O%ev0}t4i6segfx+>f_A?2bJ>U@3(8`k~m`+3MX0bCSr+lVOurj8H?7_9VQi;
zn1=xF!DsoMr4@CsRJH!Oe6CeVj}E7#tpKg3dLFN#iR{&xX~e8dBq;$P*mnwTX|~P&
zBq0*92C9>jx|!i|R69yM7EaS(eskM=UWHf(F+(v~F3G*k&~`J+QDrxu-ujULBkok&
zp1vLtTUI@h1&56r60gM4$Op7AzhG(eOl^E6s+~fUY=XkK5rZ53UM3RP80(VIVKQwf
zL*25QPejh==drbhe+GS62Qz;gIJSgKJg2{C=^t_a06Ng}-|ka>h_|gC&AnnU^a<|Q
zLfpfDD1e_GAyv0~4$s=ZqTzp!ho+rN#f11g*QO=$-DoUoQHWd)X`=HPqdHq;B8*Yk
zhNH3~LH6;y>QMzL-L&G1uU1f+XVRDSOz1Dk^y3*=<nax>{0hX-ACCnQ?BU6$ifIva
zX=~w!eo2p7jmg$CCJ?16O@!7DDjV7|NUqNYyMY`AdBjAlI^KZ&<`U#~eH5<KW}NYs
zHHaGwEp~HD35>u*sC?qDOUNFRyoO{zHl{A;?jj@ufmaAFKJ-m`L^Y<AK9>N!Q8?#F
z5cA@;Bb-eHF`oy#UNCrO3lWvWareHp5M+*JUZOt1eam#Rn|Gy-f%eJZJH~Z9W~`R;
zZn4O|uf&&H_t*0I-<H-Nhyb)Noo}#cl*$*UNq-SRp1FzX03{d)QpR!N1kP&NMM*PV
zH23Xz1aaQV2X^J9bB{dHJ9sz5{L|Oet^oGd;ep2MZT>CnkjvGAZ9%tAM1BI9$u<d9
zBJ&9B!WKfH<G&tN8Q4OCp(o6FJ1rYVE;ZDpZl5!h%r+c^7HFi^w7=0qwomicB49?u
z=9yo0Jet{zI}@XIEqn)c^vYJC#=wd(EM~}-J-$K&XoZdvtv4tMs!X*Duf7URp1o{{
zb8`_O@CisKX2HF`S~?fk0Qkfe1QP}t0iW{0x~j}eBDdXE?NbD)*fBJ>StT;uHpB1P
zJ**$l@u+<VJCgF6PEwhU@DTi~PX7wS5#phBVFt>Jyva1eIJ}%#3Mf}?*|LHr;LOUC
zx?P!%P=<MoDcOYNr;fUk#l0Z|m>a1j>YjUc@jWr|c(2pL?wxS7kv0JW&X4DMeA8sb
zNzF4yb%cj?lw3~}>|zs)I|XwYSVOXg@ns=57q&X>K{z1r@a+b5y?M;w7n*MQy><H>
z=_SmXC96o@=4+&TV81v!XWz%rDrc&;lrG=>H|7mLpL3iXycTH#Y|}r*3(o?m6<&Pp
z$lmn==5K403u@RL%q>}%f{}c%LiHKSBoF~tE_n#yppazJ=uPSq;j&iq@b5*Y<zmlm
zz`z(E*e~9A<Yp-uu7>B>T$Y{3AFuYnEzu^d3V}B+&PpBljalH||IWk}BVnw><c$=;
zm6aPy@kEBjPHw<vqMAiGGkyBkFXj%Lsjh5$3mw#nM(8oVF%v-CD)Z>evS?qKU&e8x
zrXuAu4V`%igS|)35=25!KQi234{k%1JO_NtBElmGG{NDLwRQllzX!K8n|c3H{X5-;
z2QKDBLWM)dBlhwoSvS?~r7BG|y@J{Pp_-btJ-IbW`rjC7o%5kJ(V4cueV5@7$U&nR
zsUFR~@!t9vS_6*6c1iRyW;}&ozW0Qin)U;=NeSW|x+Miw-l5qSS<ne`>OHtdX<jOg
za?#xbSjM9TY!@RR?<48i=d!j^O=L8!1$Z`1NoKr#mV<$KllDEd#&C%lt?NZtg)!Pa
zfZ(pj$9}9eWrXI;&167Um}zKqkW|E&MQc}DF~oMO@pCd;2i>};Q-D~UIPZakMC(#E
z^O3+%Zi~(k{Zb}#m}JsFw1GOmUFcLc_-{wdGgWMNB>7ZNbu(CF@(Rvt$iRDF0S`6B
zsOM7c;gVIQx-qa)+X%%ane@XOg8@Ncjnea_NBMYGJbrcNsI4NQ@Ev_10a@0^LkMM-
z>sNpHs*%Zj`V>&2U<gR_#Qj!0ysmVIPS~=(Rz3AjnDkgwG9gb415ZBvs}Lx0Qt2d`
zwC4tU%RYhkMx?}LRX=O$`=6z%r(X*5#B`T35aNhZOSSrP2ExV)ISocRxSAh!ufK`^
zoDZ7~b3{tnpz?4$?KC6fPEQGXYn@9?nu!?xCdHv928rGC!73Nk6G~ytw`8sx2l`-0
zq`~{f)b`vcvY250DBw$&y`XnC`n6~nD#<1B{|VGU?4nOe$tubH+-hg(PNHN)>$Fw*
zi-F%y3k{T7Sh|#M(q-Bg?%)ISCA$A76TtW_8|pF!-jh!N<Sr-O*lj99N5ab2z6qlp
z$5`{r7SeP^J$B!H<3q!LW?wylVye<XqFIffnYwcrcF7R<9DW`oJ;i%_31I{{$p2&n
z2#j}Ww`TWPL!-CsfQCkLj%8Pwbb2aONiHJ_oKg%9Spzcy=$}PA_tR7S*1g;7gpdM;
zc80?fB_9;-7dUnvL4#ITE}2e~|K?9x0BcPD?P|U!5WgY+WFlt%A+&a~V8y=q<&jB;
zxmGd8ibvU-w;xWRXt|s4tO4vuZV|#gvh|;B;TEELAM&U>^`jh|Af$QDtG)6|cXoq1
z6-dLz`Rn4Dm@`=dc<0l<tV{2zRPMp-8x_OLy$kdQy^B(s2;;!bjj_W-Ok*IQ`a7^N
ztIt9iC#Dh^=|e|VjSL-QwQ-Y+s3<YebK+tbv*)OW^U7@vgdlBHTz9TUhWO;)BmciW
zdnWT+-T0kc*Oq^|mK_E^Z6rthd-2TmF|w-DR24pQ9<{(trb5nP#`>!`52)yZC8UK@
zXZ2;x;7iFMva<e4s@b4BKR`t;2P5A!eW)aCTD4H?k|NY^Lc3ckpu=wym4stc0%6w^
z7m4bQAT}V)eZ8s344x~~N_wRfI`gtbOW?A|5J5|>4^5|B0QcleeI#WY9(91o5fqmS
zmM%D6`hBvF{z6SEQ7YlHY142pg)(N}KLCWBJqk7x*I_O6EkA?TcVu^izAEf~A~0<7
zXNFev6I5woA1h+l2zy>Pm4UpFY8O-%2~1^QJ=h)k3jA$ex>r<bs6~Ozi8sp;k`y_;
zQ&2^F3Ki)yf<XIi;2V4V4Kd&54O3W!qFNfpjEqLsV8y9CIiL|FDw0Ctj!feR%z3B;
z{U!^bQc0BhjTqY<z<@NoZ+hfZJ0aM7vLtat+~L-4>7>XI(QhwY+ih2b-;NwC(#Gv-
zfi=X!11DyztX5_BZt6NZ>5jEKz$96pS!M6Jf^LKwPJ>8q9<a~X%a_vB0>D6P92}$M
z*MK*tI_#0nrDd*q7P+ZH;wrTuKr1V~;UnDh0|1bXo1Xzt@R=7SI$>woxIP69&l9|O
z-e$1?(x?E9c_*t<I4y7`Xp6G$Gln<*hnndZ0Q>$o%BNqgepBHR{n>Oq`+V&Yw2T!O
z<WH9ER00CY=%ookWTV>QyNIS~JHl9EQow7#Hp*I~9YHQ~QYB@m6vv?`ifChNI2-n;
zV1@RCJbyrTkfcJG9g6fkxc4<kzAFU8rMyqjOYMPA1`?x6jAt-q$_M68EO?LM<#48)
zm;{g_*ve(<>Zcc1EoUin`SagQBGk?>{ts0|%uRPEllW<m!4J5rC)%=N4uw-h^;Jp3
zm)1J`xpncE(I?|eOijv@sVb`IrkwB$CaH%&pOc4pkgyn$tm)Shic0T^!UPs-f$5^Q
zX8k?NNOkySy`(b3MA*EXP+vysib}b10IUc;wF9&yTQNXDl@9#qOeGhzW7f0=y*F0a
zoDEFE(=x)RYJj6CzWfc`#B+CAc^}#u#8IGX5yK_RC3y?sll@4r@x{k+e2Bf-o3HYY
zL3DC<ZysyxCIIU{7!lfd<qxa6&Q|racTbM12dGByF6-+mHFxdiJXQ+d(bIfgtcc@(
zLzB5s$t#dTXspA?hd3-iBps0)-a^}Ui3%{Ig&~~5O<74{L<+qpT*=^Rpx-$@qQZ1q
z`XB1#-=2LqBa`~o@YV1CRUT045Vd&tsPQ**d6}z&)tZOTGZ*JCW!81sZrB8)SjeE7
z)37b_*QDC(=cKx>9_(^R+U8Z+vJC4)?(2dA1c$57zyZ2$51(58D)f6iI4BLNEuBQ~
z;Swxh$B1!8Zqj!I;y8M^8BQ=oVi)`$v+BFjY}N`tXVpo6Iji=_&>@<v2OFeG8JYAM
z$J^Eq)Ip`!a~E%*9i_*{XjsSe0Xs&IGqOo|M#rOEN2%%sfI&33ElH*o1fmHWzHAi6
z(@d^cebSx=v^~ZX9ks4wszh_NQBjNFlZe@KTYbVn2%2YuD_AzqcqCQ-yde*pOe9-f
ztNZ&Cr?HAhG;)V*URI;+vZ*Y>sXcL|T=}Z40D?LDui<oah$g(i)ua0_4#Vu+8aEo%
z3~sMJ+`y&j#iPmI<dxW0J9%I=4P&=6P<D#LVxg@E;za|DBbR3u5Xcfn_N2Y4BXhli
zVqo09)cV|S?3Ar#OnP63K;fhs&uLlit1{2k!^qf)wrdZjS?ic|P8F(NMyz-fd2gQ3
zWn)YT<IV6@pA^$JQOhZ4qnztlV$dY|2i^j7=%xfP*Nee-{sj-uo;WG$QA^Iv;r$dQ
zNUtUOQvJ?z<=DT=2yy0QC9Qf4!z11_7J8h2C_+D??yN*m*K_^)H36i254<t+)b;68
z+2t$x#qvwq;kOK~<DY(}iqn~^K`r6(4{Nux6!7VVk<EKQ{@Bm}=vNwSL4~E>z`)?3
z0Z6gmr~5+FkeXIJBCh@%)5&N#KDmhOU^6M?{N*FH`FMQ-#Blidkcg}f2+3)pOHv~@
zPYkQHo{{`dGW}nr=LR1#mBNGt{n3ivF}i>Z6&n7S@Op?aqIPB3n=WZ_`y79Gb(Po<
zK$_2VOevRFzzNDE?GaEZRk_m-v)VhLT$=ze+pf#ZfR^zBcr7nw$|+#|l4Sp<>mU<>
z`*&J}M)Z&Q>&ESq$UKXHMrJm6X~-~!89}l!2dgmcrbedf-iWL&`0%dhDKC&pzX(&+
zpqBx;7Z8B1b)JE*e52?LI2ZgacEjrlbCu8O;!=MvfcdNGAsO<tz17&7{^@J|y1jp1
z2pm5!wY@bTbpOvE{ljJcdA!~6hNV0(K`68O=a2sC+WzfN!sjR60v~_{f2X&X{_4N}
zr)To>cRMGL`I-Z^693=d&A+&&*|V^wc$3NLv46cK|DXT5EjJ9P$+YS&{Zn24?|1S4
zt;f0nz!$WiXC#AdPqF`BtYmmW-vd5Olp1ZC$$M|$eE$uPT5tsrtO7Yi@#A`}|M5eA
zddK@>Bv|D)Oa%&QNo6>Rtds)5>N!xT3PaC^{K4Y*+ZFe#aQkC%C)Qqw1h@PKF!fcH
zvUe6)0jOu$3uH4wDf*Ejf3fWTi_`Jf%k9wX{XorD@wF~(l3Z~tcLV)Uj{l2K`8^7D
zk7p0|L;vUWi@kHOxDQh^Khph=-;1;QNE>Ujq-6g;elWAQ;YHqz$)x<#*yrEho&Rtj
z*j~fYboFDg{Lrsc^q=nWpLafS97#R0%JlXBhxYE@cft2}XrOSvHXvy5Pm}jQ|Jtv0
z^~?=Pe67u3xcNU1s{8*R?Emos{<^9E`xE+q^1<#|iG@(%AdHJ+=HGaJEz86kXAWTA
zylLYCe2j&cYa83b3KWOcr#`t~d?sfy5_dl!$*G>=?*OV!ZKU?g;~e<zBK<avm>*~~
z?jBmc@=$P{BU7{nbi8#1s2qG%(68!3bGJ1P0A+lP>RAf>n=mP+MESF+|9Za^aG!y-
zte~-fAv6LJ!SNp?_e2M-PDRwb_EtX<i!c}v&zZW_pDmiUP;>Mpw<0`s#?`67qO{N!
zuG9XeDU}7r-PSTgDA<pp$s}X{TiK{Y^b)Rg!VpWr+y7iqC6b=3rrGIfu*9kNjTJF;
zn#Huz*L2g{pfZZr-;|Skq-n1))D|;+(5Lv3MlnCY;HO2t{am{CsRK9y)&m|tTMLaX
zH7WECc1Cw8YoUg`g}9nefM1N+4m1(f@!tVLeg|Faa{wH<+qkFvRpwp+m7t+j=>bUj
z*Vq7NQ1#ON{pVfI#7VD}t*ly5*xE~4&`YNLvB9qHeaD@K4X7LYgE{?|i-Oh=RBr-c
ztS#$+4~n<HJh~SNu;vxXr0I9A*YVqllA`u57`-B_td#%w-2`mwCtltNSSj;$2Q1So
zC0v8ys{hjQ#o3OfOjc#RTNai6F1MZCE|FE1wkU=s94+R#&46qGpIF<)+5sLH!i$ox
zr9FcDn;+ZN4KL2&fi6L2PSm_aw$mtFi;(J%_`bi`RTGtw8|#b^tN;bk-mSpkCru~a
zu>v3VGa%-%Yp)Nu!pQ)A+#8gbN#o9~Hnw}*5cb;u^L*7I?X4E1^F#0&6%lVlW~L5J
z93K8-fN8uT(qw$QgQG_#z7}m1*H^mUl0*i+Czu@=z+;poW^qDSwh?{=V$#8~Mxa(L
zaR|UoJ-DS%fR~7WML;62!Vbt=nzbL8gI}($4|VvXuJPHf<OcBKBSe_i*X$7jJhd#@
zG7|7*_7wV(r?prlK=d*YV*I&hE69F@GT}VnEXF&)jhW`7;vR<#?0V=tv<X)LfjAFo
z9kxQ48(=j32@%fye8j(dy$3Ai!Ab;SE0^ZcX$lv%Seq952`F361Kup*003@1jOd5V
zZDgAk@Ena9rie14_8frv;`y8opgrT+cNN>Ozb?bNJA=%PjuGYxSLla3xL*=RRJ7C6
z42SJWb}hFK=y>g|!jkEP4u_7v$5`1f6zrW*=wJ3eOETHN0(QSkEAZ17kld8~!}*68
zH^&Ig?5(SBWYquj(JUaJIDq92rb@Q2@BBbaEA9)yB=)X$X$R1CYjTJmp(^@pP`n_G
zVkQ@1Ro{H)0t1sR3mgiT@^2_&u3o9+Fjv$&GHCQz<a$<NEyg9D-c_`Xl+&#DM>I~x
z8HWXQHt_$vfTqaloh^D}^B4g#QseA3W^M$10UOmIxm@N2i2JhK0Xo-WIJVo0C&hEw
z+oK%WAP&|A$S!?<x&cJlSEzb>lpPCVlOjk=XeSb%)4-f96py@TzlbHvgEu#u9LQAN
zx{&Pq^t~Du(x$%yWWRi68vY!PA5+0gO@&dY*lKX~%@Et`9q?gC2?B{}CCAFOb$=6&
zwdx(i*{c^W(_yKq>w2vK9{U1uzV4+olR=aiS{DzRJT|=e8>@ON%~J&%5i6psz1g&}
zDf(DJCNxm1FCO&a<Or2y@%r`m>$Ddw-Mra5zZ=wk&noOjOyE%sbI>m7<F<=6tFNrE
zUe&1r{~!~8Pv4Hq;;^=Dtz;OURBQIR?q4M9e?Q_B$e(Eu92odryZ8kV{Kb<Mi#1Lq
z26YmvoSbs<yf{6GN;EOzIuhkk`?M{_SA0^`XRZbry%0tsYl<yKN}Tn?C2<d7G;gRw
z{-~VrJsjbrzPn`dCuMxL=Zja;xDj@f`lCAR;^+l^SE+u)u)c<6|Kfq&T@QrK{45%a
zgEKa}o)l1COL|Ve;wml5JRs<>ybb46sW$%rbM|S}EM8JNZ0t4}-_vN`6bfFF(%k)w
zAGPe~K%F2Rw0y@zO>}zM>L@3k;xMP`kh*sD2o8`u{JnR%w3sXl55YoOGrzwX*rz8z
zjeDFQJji_1-Ge91t+-&ePq$M!o8;%hpowni;ixrXbdN3(SK2~)?+F+s)wuU*ORsD<
zP#m!xOyIkpLFUu2uFNjOWGt@(wvEjJbII#5@bGwx23v-EkxBXr?C(!-q6rUtFxkY|
zs;-T^b0`vg<R$t&^gT_C`6Q8HGawGC5U+-dy;+V5q<{(01SrH-%%lvAKKt%z;$Za*
z+{JXHM(X)yE@=2?KAKN^xMN<ncoeDXKOSIEWoS&Yo-*~pOnh%IR-k4woxf{tFVDfo
zIw4%@-u?!o&E%jAxo>w?vk>M<epSo810U}-@^y;V@CbWeW-ql-G2+(Lr84<e%Csq?
zo-9IEt^J<4rj;&vlhg&NBis<OLW2>Vvk@%vGKKv&Q<JB}0w#&wH(w^{#TWatd@onk
z^wk9~51FhWgAzf#k(Shp>uD(um}0`a3>q3AjrNK2j)9}zq@IwhQq)Yrp#qnTMpXfr
zZaMO`^qW1|{ggV?R?_lz>Vvk-4V$bq*+%XhrW+$np7G`*E$)kIp%gsUo$_{XxFQ8g
z1r)xs7Wf<Z%+ck#@rr5j;wJo%za%yt;kRFut#Ui)VVgU@{SyoNREmb?gQj)4{VBm|
zdRT<{ngT-ICcDp;So`mN$$gShb-R0;aa7$hr@25(Xj<0M$Mto;b3Lch`Dw=XHPCFW
zJL||E)_{~=)u0kxqvE5_)!l4aEE_sx^gV7y0FywXZ2q$rx#NA|<-=6&7YncTHl^4(
z0$xt##$U18l)B49hqXt4;&WFXLqN?jBT+W2<fR*Tw0i_g-a=4!D3tCA)(3#`c&4^j
z17*pBydEilCLpOOLbJ5T@4e@F-j8Wh?UkQi!MwYVs;%deK*6K!b}P5S`}g*Og>X+G
z_Uq-dNfImD?#J8i<WIn97!L*wHF(@yaL@-9!YEdydod&ICboM8N!O)NL}EN>nw9h5
zaul=pr30Uk09PR6l~dpGQQz>m(@(%Tx(?{mg6hfjd8!zi5#;FNjeCz6_$v*!I*mg`
zGOJXd)rL}Itz0Y^vzou6(mJ=d4iw#?w3vu%nnaI@HH@z(c(@_fN@gXi)6O#D6F+Im
zA|ID&g?4?KefAAF?SU6P=H4ogKYlg;^3|S<Cqpfp51MxnmgB=`4pks(x;A?yJ7ecZ
z?eWngz@aNr8UYt;&taI?ps^Mko;#&5umdV!Rz&Y3NTYD@^E{AwvoHX!*EqEn?<>Or
zB!*-K<|_^x%Puo~x(I;;$)LInVxv+#*@szXD}dB`ez=4ilA|OF8t=gBv0j>Q${&zm
z--Ttw7kCv(_Vr8+4i}vSth&JT%T&<O_JB-f1&|S4GUsi13(jUw%}(L)_*CnU-dqAy
zfgz#a39dHR!AvHVi`c5I_@03vl&mmx^NG~x+7YGDMS}Gvpwc)CXZ>0D26+i;d#@it
ze%Crar>X2*7T2}b52nOd_!tDx%Kge;d}+IW0QG1|FTD2+V;0z}1gm>84(+Lf8ogd{
zf3b%eD29&Ihz5wT-=N=nIJdEN^4f&|hY&FQa*g2sc#lXPd^G01xUTt662@Jfzt!h$
zGW@uhFfOLxB?I0FZ^sDdXJsC!Hp>&&jgiU^Y?sbMbtxCA`+}rOHvMeDNtzc+w!lBW
zIlKiTx^LUO^I5iT08toOGCvq##p&N<4-1FG-TC@?pL!r=UALmhd?>rm55<rdLY_o)
zZ!8XK6y;Dt*v;j&%<ji5nT5)`aubcRy6;L`%`g<U+-QB2N$b<dL0Bw!R*6rtwvA=|
zz_^5z{n&a)W<0M9C7;!11C`9mLf~iD-Hh0IXFKLFl_@wKvMgrOjou$cGkZ^_HA^4?
zHG%n{57u1P)Z5cAbxGWcBa?sIz)_)P7(<ubvDlukn7nufAAtykz!OPk>J-l_!nGfp
z!zwjJceI;3xp_fl|ABB?yJJknzWu(Z*{k=zv=E8=f+TFSm(sT0Ji+5-9kxrLRb71$
z356IRHimw>(?Q{&N6s#&CQovC;q7KsL6%GV*z)#wo-0CMB?sgSrzO(57M)03YxiC_
zj4@E;yLa9n$SWH)C-)h6jHMjGJFc7{AaH)Z{C>at+o~F)#Er^1c5ABC_3|Nh6B1Y1
zHt`$M#U^3YOB(dzykz~?e|vWRh7Orn!lM_Q`Sgx4pZ%&yC86jUPMsX_3u!!+>Ql`x
z31hh>k&&$==6qt2AIDJT3gi=8mz%_}$;#hH5oA(toxGj@nTrSFx*2sX9tCAXF0)Uy
zf(&r*9*-NbIxy}RG9s1>dY&J@0Rpv&;d&`2vlAZqkvSjFmdjxT5+8D`sD<{r0;2^3
z)~r~isus6w#-gHA>KmVjm77ElHad<$UH1u@Bo4M)Ls(^FDy5V?$nW%5zSjAilS0K@
z;c*1Y=;NOE1V~wRhe~HMt<lo(kT=;Za8Dc4V+f@3#^;NiL6Pqn(GOp7R=F2P%=!5A
zhJfU{bt*=uJ}(c{BQAQkFRRQ&Dp2Q}ED;ctJCBp)Vc0vN%#I)YeX=ptQQnqW<#<$X
z0Ctz%J#bf>cT{n+6S!UHz6M1YFP_Chv@z=ilrzurzU*@a6;e!^EcZ8qzOwp85^}ZO
z`xwNq);*K4Pd-Y{F1d<+nRg2&&QX!ZpQ2*uC`rH+wgAc>S{fZ0j`NC`$jJJ<jv<#w
z1&)00{rK>D)7Xa|v5n7Fwd_;!w)7~k)XK%M_3{|>n3Q0m-gN}#us;p3ploWydhiKe
zRt(Q`9Q6<&CG{?d==OdN!zWNssW)yCp$sc{Da#ayqmfllXpt@_`&ubOiGZ61wqEH1
zAXV?KHm*YFAh+m=-?0EKQXh2&;xF0EZh9{hFHRd_LF?7yv&M8Iy-)~(OG-a*3=Q)B
z@_b=K+4W@}x8Iw!%zRA0Nvu&4l~KyOdFhtSK;T+jXSaS!F)orIsH&>9<_0W#4K6d4
z5nk~yA%gt99T(8wudFv<-45f*dHe;-L1ne&3tLdhHa5+g48nxd_a?Jn;nJP#5^>EP
z_Vr;P*qmOnfjvZZq@=}ga-gML-=jt!JQ=nMvW}H83l*t*EhQ-&#GCMXg>Jh_H8c+U
z%s&RH+xEJ)s}IgI0kZu_I<me2&87V?i9*+QAlWUtM(v#nTL(YD<WjTELS&q}eVYb%
zOHg6D(@|$~B6Gj~W=Ei5YtV~f^YJ_mTZ$aB9QD~NsrjC6yyCXtT~MU`<U6!YjUpfB
z8p3PFq<BK3xW(=Z5iY3he%{igF_)E*(KbD5H7Cf~imK+}VjyJjLAai?Q&@`)cC{3}
zw~Phu^{bbr#s0_vKA%nP3>v1bE=-0L3A+w$g1IIOkzC=54{2Y$vCbjrETnkV1h*`*
zpJ3r0{p^w4Qnz4>v03h-XL+OGO)1;F=li}-$Ko=q?o*^B#M2U1Q|vLQ8fDK#1@C=O
z-&%HEcK$4UXn!mBc1Kro0B>fI4TsW}d93Qbo515PK8yfCXLaiQC&3z-#T#=X%P}fX
z%4%fqY{xj!<j#c7k8jE-bYw&2zETsnO8#;5&W|VUbfPJF_s6pqI;`B*V^T9phX**E
zb*huQ_^89)_}g&A1lC-TpDF+RgcR%AYxz6cwVOC;S3ixR4iI-xmcd)noEY^;18nzC
zi6^w<rSXo6v{`$Nq6~=K*>f$bITVKi1h&*I`Y#=16n0y0)boy7lYx=XkzQf{xr|e3
zuMWEBoIl_mX%HS9Lg8XJ5wB(F>#0>V4&;1nap_>XosgaTtc$iGvGy(7kk>LdDN{Y$
zkyU8x0Q=sFGiRwzR=kkle;d&T`qltE>R$W-8>cfF-Xy5x!%5=WS<2sTnF;xb(4jDq
zDMgu}5xQM-_A~C$q#JV}7>1VM9ch%ur>$q&VLYkkgTD$>`XK-dJDvx58xwKGYgXsv
z^uXV#$b-2L5mT0TbG49<U2nxkz1X2O-aVm#iJ;ROcu>9a`jOJEEFVs58}FNej7HPq
z4_Hoj37U;7%KM7dHp$|0M^G1~E=SoL6fxpVwiRdx(<b>fqr_K>SeKI2+EdVT>ubGW
zO1eLwE^(60)&I49MVE4Ae>p)j&rp@AIgM`0ys!nGCh^{7eUOf7$@yF<j<d`8emzF(
z*2uhg<50GmYEsr{dVJIMrQXzD?cN(YCPjj_q^`eF>gpD<Y^k+O0n2naWl8FFmMiLG
zNl%nY`4@cOjZSR}^0_Lr8NAo3BPt{onl99n_tAFbbnQh-wMh$sTf{g=z2`UYsLc#Q
zc@i7B*=&m%tMw@sc_=l5DA0!!MvYUKJ5t{)iH+D-oxV>_eI_KGXQ6jol8t6ny(G^^
z7b<H#Uh5%a9ysLfm<dnEc|*}y<0t&kt*H03O{Nv5d3d^oPMp^YWuGZ?<@838S+Q)<
zu+dgnzM}l&83&}$bc%kOi$Q%J@n=7(ix(uH!%54}${lbz|7IUzk*U9RHIH0TgZACl
z2mQv%S#{(3tF@<QaZk}rTh6s|kP8`qcU!$8tr-1W!&~_Vc6Di|b;bUL95EDG`2x~G
zt-jmC+f>>cI!jH4?l*taHeBld{!(i-kXd;AoD!rBdKTIrD{rc+@)S&)8dmnzXm^lo
zd{OM++vHdCh(@?5;~gz!GK?=2qUq(e{gaPw?PsU5PAgG(Qy2&<3*ZfK-?0#OWwVQ*
z`n`O-Z{L4V%_QUT1Bx)N1&t-MXo1`holgBEt=*<P(e2_^%}PRVx6h4|aLz5c*tne8
zpn)j~&qkl)EOA`+y^jnc@vXQ==1y$xJZ~aYrSA2g*?4rL#-+P~*k^1lR_lv|a^eZ8
z;h|9e`0o&%xnA_d0Mi`Q7YFQz+#lLOgi%sS$C@T85L4-sZ9Xen&LUP2ozD8um+5L#
z1jfndM!4wb)N&gd6Ej~?-BkqGWm7ow(KIxdy&#*~EOYNByQ@I!RHFB5rwkT$?I5KD
z3#HlJO__nW*?z2^&YCL`?GY{UEbciAj$FA~a<eC%aZ5c=2rQ(x!T1L?(|w5eD$RPA
zYVdVmm3<lAPR}{rpq_>+uGdqoy=9p-E;0Fd2w@`Z1Uhj0eJR7ca?P*YwTdZVi<nV9
z#U3%3Yy2SS)xb*!Q44YH-{!At2c;OLSoI7#X$D2!V669{#7A#rP3y<WEHUq|N|QLI
z+kJ7`Hd#zvOA#7=ts^5GoNcUpQUSe{-OPEPg1x$+k<sa56z#iSYxB(S6|EQ6`&w>p
zspQu>!2mzDRQAT0RDqXuD^Sx*mLhosin{Nis~2BR2ksXY%{3l8%%ptN3p&r(u<Yo5
z8!naXZQ6oFwN}oa?U|GT<B@>vJM&h`sgy=aO?-JtN7a4nH=MtSwJM5@gW*y^x@ixe
zQQ^0dg3wIjn{f7p-iRx8sJT=;Wjk0fplLy1?LT0iFr@ZD%-}J#0Mh-tMm=SEyB+#<
z890%v`<Hzh)|m9AUk(yjg45mhetkT*Y8u595SgUEC%7L`@g?7NXDR)G+0+s)#3>$`
zyQkQ`@zzMN9k*gQu6ayC2v>f7m+jr{wSkak^9^#wn9>@XfD(-W^YZ8+%arATP02#E
z@%JmE>-wS=>LF4G=&bn1^VN^T<!^2F1#sVVxW?WOTEI|=vsp7Z(LQJD#??H8_C~3q
zz>Y@)bx_<r5IXcjlUY2<aqd$QM?jCW4}f1KwD_(0*E;jgY%tbok4Yy=RL0Z05(=@N
z+co(uEy69jVg?DZjTlGGeUb-p0wHAw93i;L5d=<UC@UH`?8l{gH9Tr}>eZVE{1!Ik
zh4FB>*R+gVhkdo=Omvi%<$I3$rr^{_3A6?ZVnqs->uSba(b_y^%_4)7Pnev%3`Y*(
zzZ3&Cl#hVRv-A5CSub^I2cm}(8z~E_S_z~L#u;WISdt(czCS!NFxC2XjOd`)MxcX8
zYCvUFWE4p5LzwcB0uk!2mXAR<M(3l*NeK^gYxYcQC3g>(-4kL|v*KZw(Y1dv$uEGV
z7}sBL5}e6^ecloGp_ncYw)+(3y104D<-6sA44v)qHHmoxoNIUPhlho^b)6<&jl2?r
zSeZ8%V!L0tC&j^FZKUY@@(AT&*vx70?fUn^u#^_Mh1NNGcKuNWm)`zKZ)V=pc4LTy
zz>X|8XzwLXIJFk&Xq@$eG2z7@jxZ;0&%N<ZiMA0@RorO$4p?N=Y^!_qwfoK{MHaV%
zU)>y5$+F=4uEy%t`$S6N03hc+Y^{o8)YZ`hA27uAwJz@Mt@a&{o4&kx9f2(;dWYD-
zyYT2TYo&RSMYHoYS?b=n5s(TUV(X3BRU!*~@z(;64lnsQr;+7k6~UNHI#*@q@ii|>
zu#olDtM1J@#9r!|9Bn5U`nfLNke=o~v!QbG%B_V_|9Dpkx;_HAWd~cKh&aL|+JI5D
zK!`WZiZ<CGU&xAdHOr}uxOb~^jz8zghu1ZtwbkWax9+rO+k>Wa@0OrzWF+mpl@-gp
zG^WG9yD*4OHEOwNp;dJJtZ41NO0})8LatKFs!DrrX==u+KJ6PjS%JiFPSHGbyRp;b
zrRn*uiuh_~g%g%Lbl!_=^%8TRd|zmNH{GMEE($G<+3E39vr7WHIEPB8C8a^T9Fa9N
zAW>2a45$No3a2|s4?j6!w$dVW<~uE+TbI%R-<C2e!R+mUxHH#MF<W0HJX(OX)bIHH
z9BQgH7c{l}hHoCx``w)U!h$F1b=vSJ$|)$P1D`4hWVZ;*$Mg`QI?NenB_KEO3Er$!
z>CPH38`yM;XATviKo#{a=<mu5n8+SpV;9RwNL2_|X#&wm_Sm<pxU_W%ZkuIj!>FdT
zLuTBo*zVT6Q%R3Zjthpvx_i~$R`f6U!Kv<=#msPgMYDI|isLo<)_}|&4I=L;YhcT|
z<Hr!v%;DYfu`(<6E|b?z%*06iMuZ`8d}&f#5&D}U;(V>&sNG{L`6zkMC#lQfcQ@cx
z_t}++&EKFI%uLLPKa!f-Ov-m}6K9mh8RGa3GZ4E&KO1ZRZF1DhObW#jGm*Ao3AdJ?
zJFBf*lje-i<@AGs3mNqD)NEnz)RG1%Y0dnlnes)te!Mumy0q_izVqjfs$oClwwa~v
zxc`xWr^LxkKX6onr)7R0oZBoEIbv9G0!FBJ7_&&{_@BUn>lY^zbv4w!ay4ba^QRld
zVceaAM<NuGvuXLkTA$7QS=0i*%1<d`*7dGvL}<$-6Wt(VbSt9r<AtGAmbth)^MYhn
zzqRR-(r}B_p2M(ZuTM(%#{vbCV)@nr41a1sj3Ifs;lkVg)gBXi0Sj^BZF+&QJXQ2$
zeh+_-NfSp;t8J`01^aS7%eOp^!lOCc?u9x2X715x0Wu*g428<4ou}D9=yUjq*Ym2x
z<nmqSTx^@EQgZ^yH?n05FVeNE#;#qD-z2XLdsuwS@<x3~!9sGEeVAa)JgaNGAxV>c
z(xbVpkaH1Zmfy%3KeUpIH)p2FzHLq7_@<`;BY;Q><`{qj+-@z*t%-ckug##XwmksB
z&%Wy>LaekS?&&EA;#lCZJR2hDHne#$u&d?1h1B8OF{`ZO?px6iz9G<cmg(?Q@t&Nf
z-L1~O(Lq*Vw8ecW$=ph;5pLCR)h%(QGxyh&KX9mWAn)f@<;}yT=twaq?;d%DJ6*7A
z@>W}-`IV!eqh;~9MG>jVsrKyV1*7bU4LjN&ew$ZYHEd&TcR0K?Rw-|jP<B1GX&*+d
z2Ww?ke=E1e&{dW6uvFI4ku~}31~ywf7v1B|ugW(!A)4dp+0`dIxp^T{=GmgioShLX
zo~P_XB6IT1FI9tyl#!yPD=?OO0cc;I^ko8fN0-8}P6O=?Bz?sbv*<ahBa){>Ci;#Z
zzd;BD{5q_;9%#1}HhU$a#qUgmeyuvJTJ_LtFX=K|JTBGKtI;t`Xn?ucWhB2~yO@BJ
z2|}pqEa^O5=Vn(Q@;jF=H<#v}Pl{lra6eep5iZjq+RHrXxhgfFg$v@K34BC<7-LKo
z$@yMKNg&*wwD9}|L|meydQ`<f_q@p(28`#nYy<{%ZdPZtR`^hQqF>{15x!aff%@$a
z#Q*7N=;;_I=17T5nrA))w5JGGv18;o5H~;02GcYgH;>%H3YxUI5&-$8byz(;(fZQM
znc2wYOr$3V;Br06-3cr2p5Nq;A&y~SJ4+$RKX^$LIm9ZsmO*0XCUEnOf@Ng(42aGt
zWaz>-EgeTvGjM1~@epgi_96}}Xn}bez`u@AW-#Jve~kqEd;;GS=cpdqv52cQ3ux8K
z`xW-d1gG%HNf)mxvEdol3BFxvF!4{ZKLVT6vLC$nMZz98zPvaVz}cTQ%>p-2r{~VT
z`lB~qM45!W>pJ1(9ymv_CTFu(zd|}DO6Dm^wJ8fHc?}Ti8M0wg(EdBuyPoHDb<f4>
z1h1^Fmd%U08RmDlx`2b??&Ur#@w(KUv(n7P%B2~%p`yzNl<QMZsfm2D+Pzk%J;ql9
zeXR3GiRMPPckfylK(}<))k{|PN}I6$JNg6<fwSw*=fm2wi%$_86sH8zC1Z+*O9~&@
z<)t`d70O&~ld2d`&LA7|goP5+0A%!wiw)W4973hNfiP?@X+?j1E&t04pvI8@3T#sW
zT|c*}Ip>z7qNOIOcXK9HwTf;%xo|edh4){j7e~9+y1plpX(W+h`o8^rh~K0#?xAl%
zOVH<S5PzdKn3Be7*vU%!HdRMN7=P|v#VtWQNawSMdty}H>`QWfX1t_3psRJM^4t55
zMhAw(Mec61z>rt#U7$^czUD(kfshh>+j?+Q_%X$jl3}#u^usc|(?fPfqkMTj_^G3@
ze<piO><{eFR9FA$kvSoOhwlPMAbdQ>tSa58kJsLIi~Z?%gb6yZ=dnG3lY`^`V(hJ>
zs*JY%Z$%Igl~O{wQ5q>pX{1r<Zs|@5l@94n3F%JR#HM4@-3`(qY@|2u+UMSL?z!W=
z_m1CR4hIbOex7GNYt8wcbAG;D2zI}F!oq|ii{mprranhG*EgJtU<!nQi#&2DoT!&l
ze8kE2wd>0HpE(|^V?>(d`MR0S;?W=Dv|l$su77s;Ydzu@@qob>Xqa;PoHg1i7@j?I
zHDU|CO^%$CY058?sCvG5UM%!8eaV8cnc95nL=ZX$1+v-obp#*%bR%kWv5NM%P|jS+
zlE+L?A>vT_{9H3%a2aUzOFZ$3z?dWeRDQXWI3D`)0Dm)&7Z7vX&f5(Jkw0~g;6-`(
z`XY)%^_-gLd9H}>h!YZQ%$CIQ&EVS)1GU~}ALqN`Q}H9;T?=DN-WOP#&3C<n5)t`_
z&0Qd@*8lsNTF7p^Is>td^qsTw<EhQf)UJgon|X7Mmrd9nJjnSWET%Ycw*PKesakPs
z22=)A|ET#c=w^=g3aQEeN_P6_&~HL~ikCPTN7HTNI%~OXS50;yR8~gjSyph~0xB;y
z$`r}25c)Z5l!jYVex01XgOW(8)0n5_RuOZJhqrJ+`DtpQQx#|fsBZl9yMnammqT??
zp1E)6`ab$rx3LHlIQTku1Qh(A@#h$xwq=kjAmtwTyf~fR_Y+q9cs+93pKVn+C;57P
z!1cWGuS2;NX0B*q4+zm~P6h31NYCPG)sm{ZlLvLe;>+EK4W%-@1aloMmIPkZDxJ;F
z{?UXCt9!n{&s*Q~swb8xDm$0dN>bCPsa7b^;ltp&-^h24X`cM9q?%<QY>^%EO?_N<
zvjG6Uc8bM67pjHW5>jM#D}RXF35%dshj&5R`?E7nY}w3<vilvo0k$InN%Bxd$oAW+
z72nhtl6}{%i!y!Im^FAiJNV6)%#2q`cdfoZXR@5?<m6}hLZ^HJaTTjpeFaJ2ei`e0
z$_eZtfvp~-Ur`X*s`uCcg~wiD|EpC!>b}`hxzOL0qWOxeb`wx<Q3NkcCad9IWzwXg
zm7uB4aCLy?1ndJTHga@+HrLZjd+K)I&@+3_hQ%Cv^wzVaJSQ}*tOsJoYqVNk;`;p5
z1e-_@+oE6n#e;3R51)_8{3dIyH?TJNOzlf8=+;HK14{W!S#J6sWieZGT?eRvNXf(6
zg&69CB#e0PT7GZaZ*!#%>#ve$?8mHfX2i0wd~k-0mhz^H(5gMJinWaIk|NC#k={jh
z)N+g|)O5pru$;LMjm$UkVEo=@{#E^ztv9je#{SpfWu?N>=L)4ZY%I6^PCr!xt!j?g
z_|0-`1ePAmc8s&?r{zUvwU{8`&HH}UAoB;AXHLw>of=?vYkK?IX(Iz9i-ZT0?fU;7
z8pVHUW|ote*sAW^URuy=zpnHCm_E~^DNJ9LSQUc#*YRkcTW?H5BcX)5<GXjw*3^=M
zazbxPeaVlV{Lgb*>mJwaL*{*>kMBLQW1D_8e-Z`-4G^U^1qbUq^(Ul}Kg|zbh|Hsh
zx=@?~rp(C}?3JI{BJMoc^Ef%Of|jjb2{p?x&Rgghh=q3QC4J6t5h-&lPz22oXqly=
z;qeTMr<Q-I@Y^>da9#9CiD&vRH-OLR$<{qEs+d@hFEq{jCq41qwToT)7IHRUmS0xM
zaF9`4ZzMFcyHr*`4b%$EhKuL2nZNEKl*hMWc=dC6hu&QHG2Lf>RZQu=|H}T1axK#q
z$;9gE>jWBW=;=HETwByu0xl`H2e1e=j)Npvo>t|S3+|=^L}@kdpiFMlT#%J;>X(^f
z#DRFK#8&WJ0E~`l<80w`uUJAv@pZgediSj?4Z@jXT3?rj00Crkv-$YB(2U~3=3SkU
zH!55t@gd^g&|R4W>G3qTtrrbFp%&au5wC1DnT>_V`psa`5z7Of#oo{%szteCMPmwU
z3LLdF-?@Cm53(bJc}<GJ<B$4zl>sL|_cE`vzO%uZWv6CC(@Lm|5G0ot)+&3KN~4u2
zf9y8X|GqtNJ0s=N8+W;#)=&|#z~ibLVJ3yJ?s`>X&pQ?WjCU|=yk3-IS^w9r(C~cn
zi#hT7fLVDLEYHkEC9`C9z&Sns%*4jrcenZ==a!W&%-e?)%gnrtU0g@FKYZ-NXLSmV
z4K2gbAs&@D{Tp=h``%d-3~7o}83Z>LBY7mX*@QAEIf<_zZ4-SNTWjL`B`a)fB{wIi
zzAprK#0p8~wT4RcF1|~-YBg#(Xe~}}ugGBgLmGFGo^s7jL-?BAK$v=~ytf)dzA}^O
zBAl*MZ2Aw<%vVJk<?<D!O&+Tv#J$=n`1bDfL6|l4;ZBz6)GXnbrE)?A?;?cG^qiL8
zZ|jur)`<>6!{p^Eqe0zRFPOHkREHmgo1q!Zdi}Q9Ys>9c^`tC7=he)=nwaS*5q3|P
z`S@WY1^4D<Dae%vVkK0JtO%05!?Rv`amSaVN`Hq)k;}+#6a*1_{!%G!nLuy}MJ1V|
z`I>bN4wF>RO&?syoW%$!5_^1#Q&Dlho*V1WPL%M(%+5^H)cX6UMsK0y*IFo|s)F%*
zi7(9=?z-%V3BjK=vyOhCJIoRWvG%gyT^GLnv!3LNiTFpZWD|c15Ws&DJr))EO9=x!
zx6`%;@Pg{)s?z`=9;W17ohMtciHt+WEdL`xw<)>qx7Po$-g|`9t9`WB<^-ofV~|4i
zsAJsd`FY)n(qo{wrkL7cAv}wRKc*fq8Rs~AS0HCfe~92IXL@$xO8>f$Xs_A7ZPW~Q
ztxS@52MOxCR#uwUsD8RX{DC{Km;ZbIz_+Bx(gBF0tJ`6nX0_w0;o%TD+}e4-h(IXq
zXIZodTiMxYBC7iV&W$9q!%hCRO!KFSAW_YH&k=zKm>;UnQ)Vk(^vD)aj{cZyZ9}Kf
zxwKc>jK>%$@i)!cw4J^dd(%QVT1{Rpv1PH-qK5K5UXH>qIY97yx*9V?Wzf-Zl4G)N
zFePd_=%l16FTMK+a-L+CVxfaNyXY=V#;bTgADh~$**w^RcIZP3H+|%0$hGq@-f$$f
z-!qPq1hc-;@v&z4YQ=WiO$LO?aW)hph3Z=wUOFL7d?34(j+$wa@0xopTYI_<_flv#
zB!(CCaL?&){?t5r^J2ZOoii_5H$f!=EG^smn4qO5X)9@OE|{WsmBa)m$`6F9lmq;b
ze;$47XK{;`P0MJeX6~R|nKM*s*&_cnd(`#f1+!ENW@=XJcf6moy}i*ABbHE)vT*Q5
zUM)&p4_IHO>{!h?WYi<IxfipxW@(z;wHh9Alpl?=keXUFvwHi~DeDhhW4gf#WwW_a
z`d0xarmgtfs%@DMyZ_|Js_IX59qe;zb;v9$|Lh=jnNTS^RJb;st|mY@=ajIy&~kn3
ziFFMvZfjIfSmE{XR3;mj3ko9?UCiA&h%{UmF8p#fnN_!0Ga-%|znphfc+unTP-qX!
zNsd>7v(u$66O~Px4w&m$+UF9=ItZ|aMHr7q(!G~7xAS_Y=P~u5@S;m+>Lh9`f|;ve
zM3Rf?qcYJ?7Ti2?`Wa_*NzFE78*u9vYMg6x7CO)`yHKN#`xxNLHOJ*12<5&{tNE~8
zihj@2fl?(_LPV3HEwU^hFE@YPO4qwi!q~I5$VZ@ABvr9`9vtW{)|)H{F8gUhYX^a5
zW>`L~a4)f&)?-?<YrI1=)%9fqU^4A`2ETVM(L&(egw*<YV&v?Vp5#~+B@`)zh<!Zj
zA$5|seP0P&>(thAw+Ry)aPPJ>r5a`P+4k<gVA^3C>uty=pEuT9=hVWATX}iaWqRfA
zpvy)+Q8#G)`7lXhE1a3Xcqv|_Qq^;G=jRt@&h3S0w5F!S$B)WAIbNT>Tk0rRJH<JD
z-~C>V*c$c;kTD81-@Rvxhp$O(HRwqYH1zVbl9XcqEN0s_PqD~baFX9CO5dYX#<i*?
znT!|CRk<jM&FtQjG`_RGkg7M`5xX4zThCv1!i<|nUh*=M=lfe*auOjky^DVoZv<L4
zY6`_4?a5JGZl~~=^hPxqU5HZ%C8;%joN4+@>kwa|nP8Wr%$=Sz)1&{|8#)&HXT55)
zj!bjgN;Td5YWJg!#DS(rDX>jEWA%E5ea7HX7Jr%(?pQN-9+t@*RCK%N$8Teveq(K!
zRrmd+bCX&>?kVZKcv|V(^S)6wR^u607HtcBm~pYUS(6#XWWRBa>mNE@q>e2l`Icg3
z=jzjkYy$H4=iWR&<AeI~Z+1MYI*qGPG&bPb`s<#}_Jg61PS@b-ZJA|#km2#55RdaG
zulDS#s~E7H4jgy>WTK`@D!P<^z6QHt17q;D6=^Z&JCppxlhR+F`i#;@OwG@91>%DP
zb|%uv)0_PPX%*)j@7gp4<Acp&Q+8?+5v^IKs)+O_<%E@x`swQ*g880f9vYGM?}wE?
zQb44q+*^qABU5Pw<G*~ET8Fnopb((R4o-nBVHaKG;;VLv{OiQc_d-f?nbkbEh4ome
zEpNzb?1hBQtvabKcq%=IM~TV@>d3~Kb4WQ#%@RvDsOq9Ug~_dXd?o+PsZdDvz)q)~
zEWB%qBMV(33Kf*n_&C@;Wo33_e7}KQk5nJORoMR|dCUv;hXR+8o&`%94@Cm^vp+H!
zOvY)mWj^(z!mqVk%&t;g)3!<zXtc|u>Qaf#yM5{=IrQ|O-EhDDZ7ZGnDfcPn{4A=c
z!f}48MRMhYB(0tdeC?&Jc_JR}cHk?Pi*YOjkz(E?H($uShg|c@JEfiCqDgf%&zY>C
zV+ihe|FdS)gg!ORg>i*=+X1+dqD2ef@x>Ej1A>e#p52g3PX4XuWj6Y{MLwYtp7^Ib
z(6V{BQ@GEckT6PGX8Wur-3`B>{K<3uZ;u_KeiAj~FpuwS7T8X>q+Q^XxQIDu_jCE}
z|2Z>w$D8HJm|YW>2P^!kr_)vbxY}%l+a<YuFW?J|0e?(Z+DJS^?wLLm@Vo|3X`;p4
zK46)gb|Ys^zbexGKyWPj_7BwN$%lB?Uk&Zoz8n<$Mkn-E52-lv=fWr4_I@j3hq~mH
z1a;3^>>9s-<XMfBm)E*9f6zJGyNc1C9%B>+xW>3&h1SR>5RcJXt#8G>^4LOL3RVjd
z{h&GOrMhx9SEt6bbNq*C#h-+lmM_QG2z8yK!`?iT!fu;i1$2Nj0imYR+K(S#yPIG9
zH9QToNSrR^Mv}Ii7r$?UrjbO|MDS8He%2VbY?R$PIGff#{}cNQ)?S{MxjVG8YH`)1
z++ys(H59fvQdOX*DivbIMKCc@3rJl(+@aOrreC8oywDFF*EM@7KK?jgQW`*xPRmVA
zg%QAS)LA8FEO;w~oVBQQhlt;;x-4AKWZc?SKvB1Zqu#bIC$<YMULOhuL|=NUv-B5W
z(=p0n@wXh3bn<&g^o4#Srw-21!}ys#q_nBYmL8`RGN1=?q9Vw9Ez8$0?23gacp%le
z(ZW(fd!Cp-97^=scQ5$!d_X6wENk6M#feA&6(m1fE7A2fBL?)pIu_AHSRq1<?F&4H
zKZC?5o3OG<;*}nabOduJlSBFy*W%uekGR9S$E&Q&T`dBX=Td9=HB322Y&oEw07Let
zuxRX(f1EZ6sXOax&SM{U5uY~J)ehCE-U=E6l?2)R*%tk+f~(11F}Crl^C^z-qypWV
zy_-)zNkf?!+9aBWIun;ts@TBx>^41im#KAnH&_=^btV6EoUmJJ5TBMFxolcQ->UVv
zC}r^Ks`3Y^ti(OPu}J$_>EHbkLQ#GeJM+%C2HA<blj}5nUA&?+%mj}PNJ4;Sgi^uJ
zQ$+8r;o;(jou#_05gzbL6*pU;y}07$l@&`W4x7NKFTh9u+oSjOzRhfL+yo!AM?Z(w
zj3-GWt|(dm0oG#uCi2})6&o^LHpyD6Wc1NnS-pH5{Ua<074?iy$Sx}D8#M!hKN9}u
zoe--cmAci<k}kieYt|M<Y3}GEho_Nm^*5;5UQW0x-8wBKXu>-t?FzM-=3{EiCSu!J
zq=@j&Q&=U^N{TN=p?=AUtDC9OhG~%#hI=}_UllJXK?MvF5h*tFgH^IUA5!C8goU){
zKjnPCS*n$}$%r7Fo+Nzbl>;S$F7~UT`oiXsi9@*9ZL07WX2%!OyZlKmrr%%BPag5i
znVa=b%j=Iv_|#82n)FZhx&ncLv#bkd47Q&_))!<>^*IV$?}Pjgtp70z)8Bd#5c6>8
z3<2-*b*`FGw`(F|6}U}pX&ntI42yfXU-guV3HMPsp}1#nR-z3@ON-dG?|wS^FIvg7
zzt}%~3FuzD71eM|_<@Qk7eh%*AYu3wtuNT*;@&%y7)=9J$A_ZPjLg3<m^Htl-lxN4
zqLfMaf%5Q506O-<(i38y*Jw4?oS(6^bXa&;#($<ZSqPtcsZ?W(+kJh_6y|49-Lu~U
zlVu8{(nKI%tar}p7=(<VY7ecV?3IvU?x{2s-Zz`Z@pE2a9%IM+>hg#kZ5~X$z!YSU
zQ=Lr?zR?`wV96hjd)!k(XtWxC&184S#v&V}e>3JdYbV}{!;Turi4wIYPL?vYT4Oax
zkDkoeHAd2t@WWiXh+AZC2JlZtYq4&I`1R;JpHcH(rTm(~3`9sjrMDgZvmbTpSQ#c<
zS3H(MgC$XFM&vo`u>~B5Vb3#NlmB8dUq0Li-Tac$kagbWaVX#48Dl3F%#DXjzaJ=F
zth+qivF0(57KnYLk)my$?n0VHT~4;St&_QAHVstmUxgAJFjb=axsq&30s{HT{^v*a
zkQlb8uqsj$app4Ihdnx9*uoy)<asFpFhcZ*7im++hVnLLwhG$~FPwLWV)VQ-5+V-e
z+}{dZ?unv$T)r<v%fDf2$mSe1$*8~58<@mQ$Rv`a>^~AtisQ3Q6T<*k5=tPYLQ7_K
zW|pF&IJyw=2|7#eYJm&j2pjY7k~QXspkEfIcCKFkg6B}Z@gQz+6*G^?BibyIo$Yoc
z+I+}~<394c;!SYQD3X*CglYUJLgmrBnw^__<6vZ@nn&*&JLb!z%9x&b<1r0Jxke(0
zh}7&~tQKp8QAr3lS8s$yO0KS{LoCNjlJb;>sxv=!<6J*`*sc#u=;}4|wFW2V8TD0H
zeL138aWb-M=iUQSR0}ra>OEk%cDszt4~JSO2N%A+&a#`Ja%a)_$Fqe5-jg04leXNP
zY3eevNjdHj6hwgHUQ=rQ&*9{A^f!<2?x{^_H;B^RM6=HEm&jZrL{RSo8ZkvTS`<_2
znW(nm_-@0|orBG$;+xH9iN1qZ-v%10D_(SPWR4F>T@(3>ba%mzPGH8{*k2>6F#i;h
zN#7+nW-h^F&snf4d&dVx%V`!|NeyvD^7)^*sV5w(^N1M56@!_v@lA^S2P+l76n)-5
zWDJ8>JA@x4TUkxVQ$^%<*w#l}9T_Fsw11C&%BDYS^_<vK;etuKuHPsg@id#}w4>&0
zjqkvDmiD~GIiGhx1A^Mu`ipt^;uH1w<LM{FjkwY(&3l->9wZIXZ;ochHEstKcNVkx
zU#E+hIx8^1Z6}>`$5l7nl6}m=>xhW2qmR||h8IIPRcNsf7PG#$KBH9<C=vqAH*#?U
zYK(_KW>Ix^nDove2$LY(dE{N5ZZ{f`3$b&)#|GKcoACU>h_5>#8?`M~f^0+0e)Mrd
zE1L2Cpttx!@0Sl@6xmY|3(a@=SgKVgFsAFw+Ty2=b!ND+n^okVWQUnNA!6is>paTo
zrCL+@T-DOW$I@HLnL2x@92N=qvT_se(P>BW+Wd(&nhUUkvrXOwr7KDWyH-gss36_F
zKFN?6Zr^)X6U#D}<t+MY_N}u+|7sbQ1B#Djskrl3-3k8a{>-LdgUjGGfBbbd@E!!8
zER=Q7QI;N$j@eYG=rpJK!=6}nCxZ$~B%`QUoD7Yn!dqwDmczvoikcA1cT#bmXk=?}
z==d6VVn5$Fk1lx!LSvSyz7=e!*9kGa*9|8SQdFIQ*?u5)rk05Fr{eEhfC~xH_XXtR
zGje*Im;|BUV(Av((q#w7=%o1QZ&G;^&B@gsDiek)mSCLJJIeCm(goLf4xT0c2C;I&
zkt&P2FbXVJ>xHw33wjY_bdLzEDJGx4a;Zwcqh}>MH*CQEIM*&Wme3JkfgX7j5o|~h
zE-X;oa?ouD+aB95Glj@GkCkMFD~rVQ?Ol~wPR<IcS0ysxPgJXEbVpI~ezg%EL6aC=
zw8~bakqmKl(H||(50B>6WOemnOrG3fT9jmiyrq@Jaq^W_BdDDkr9`>-(6giDnK*RL
ztQ(5Z0<p&}!)V#IHC<S!Z31qs{nk}oZ&_GHSySOzqt@&}+z>4X*tO!T1~26LtjiF}
zfg?h)yN)jMAyL_>wL01pPmjajIJZsIa&H^ZVe#wA=$cwb|8z6nMi7~%-YgmTJwd%K
zPEE7yJ2R$%%Qi*`vS^852uSgZ5r^&iY)sMY?RIxVl72f^YYh~j4PI#WeXU(iFQJfl
z&qb!k{`)ymN1fxg&lP&Qvnfo1!P$)MS}MJB8h}q8-xvQug0UU)%C;r>kL&B8A#-6>
zmV>aEDa&%CIB%<?l&Q)4Y~m($kv*uGp`7b;*LNG@d414ytFXKW7Z>7?nn3wgK*u_b
zZp|xvY(y%3_ss##Tcz6)&O~gyE%kbIPy+9-<lK|S$<Hl!J*q4{yxzKmDU!TxziKfS
z7h*115daYKycV8$*%X}5x*KY%ha)6hzhe&9B_1D(G$-p4m_m&!V6a?Og1dh@Vu2J&
z_A+(HFR9R<nb7c_Z``Kn_F=7!f2SJMa~ht2G-`*9JTZ0zb^aPJisWSI=o7DoBhElt
zGRO|?IoaJ87e;mi@~@Rk_O6gi**JNo5)lrBO^<3{B++K5i`>+e^M`WA`*S<WuxLDc
zQ{a9chm7n7F*)v9lm3EUL2CxhvZryQ>V_)^Qz~}~&zZf6cC0j;Q=jI;j$U!H>-Xxs
zaq`-^EPcgUduqdQl9TRUy>uUaQD=Lqdvpt8xLo@=yz6vMxK^IYl8&9}SDw$nR}S8B
zi=MOtFmNeZ+T_Y#)bbQsu%wCyO-(gG2-w-4gC1yC8w1Z<S~ZdC51#b}L-OX<#Jb}l
zo_gG5JI>E9b@MWbb6BLLrk*DY<fQ+!C3JyzXhe>GFhsUkCTDKNACx5a8;J6yGvaHW
zBrmf|zD2M54IVo3c>m9T0x?iD1mv{2$f@5|n-HM{exYKg#XTm_t2j0v`mf)5{FMtu
z1JNbxo)PbQ*Ag?8hE`M)AYvDC00vXe%cUBg76pjV)lHW;{ro!egz(V+0bLOt<i}Qt
zQT}qspZW8<irg7y{b+L_O1mYrrFIpEwLY*|jcmYLRGvyuEi~I1_6+29)ID*^e`qSC
zblp|cbOO>c$gg<zSjG@`GSzwLOOm*+qHdBb3eiiiE^&)>48Q8}nl%`ES2mS}o%I1L
z_N|ilhoClm$Y+0|5mIMku0kxQveO!-<x%v_l;8FayewWCrsb+8M27?ju>lMG!ko1C
z9WOD{9NR63G&cIJGDjRF5y)^kK*bf_QWOAPEpjW?xComa-uGW(DSADQRuf*f<G)mz
z65$L9y<r_64xc1Q4S-U=F+F;!j#{y6@b9<lzPLd!rGY<`DkeJNF}U4CJQx>f`ZPCB
zQkm=hUm)+Y0R|CD0FLsWz#&oJaly6oJ!a)%$vXLV3V*8F{jCTz+bx-9S4=5C*^Bky
z1tfI+iQ%GM?fb0?;GOk)@5cD{_*NvM+B8cL%jpI<L~dXZS=v-uOqdcFouXK~iO#Ye
z7k4#7ejvNJgVMF?i%&)1BjWf%lLbRZVD4OzQXW@|yt30*InF<efMOB1)iRz<#Y#dr
zX~Als%R?x=)$m?|rbjB1#PL16hz#ebZHpI&1%qlAsF{BkZR#<7D`D|kQ}PAcdK3jN
zR6(T~{%g)fBEJ)j8RuP0gK2gu=1X%BjY+2SLBup<!X;JHuQ-WT%W++LOOWORc(v_2
z5m$?#l!jN^I+uX3eq8D%tcu~=p?Rc6bu&8DwQKGwc0@3lyfdaPHSH{Rq^2OvuPf=3
zxMSXJ878!P1gsDlsT~h=Ymc|COzOdC%phA+)>84N+4S<$+tWj^<Tb#O_ux#qI~H#R
zf0u2TrRzWMIdXxc;*R9VF47(Y-fpxtWGC+s0p_BVj(N?&GXXyz@x@>-^v^IGd$b37
zPEKSL+)fA5{_(*6^KVcdgK`ZWfVF#P_o%#)`(Uw#efzNYOH>b~5-`qCe6sOUC}JQw
z6V5lLQ!M14(^E+2DPea4I@P%Gz1IHv;vjhDKIP2W9c3o4z5eki!%%QxZ}eLkTL4-X
zyt?Gm3~hdoJv~6e8a_~L!hQ-u;DnA7bCGLJHJtcc!PTP=nOAYKl99wtCvWA~v1cjq
zccT0Hz^aHv+7mo+2(&Q$;{Xu%7BV77Tuz2Whzzt%Fxp#BRpQK-L23`WHztKn{6&Ob
zGz=}{BgB#`aJT_DoX+#6RL)q9BI%r}Xg9jQ57fBc6PguM7)XmWPIkXoGoY$Vrc+oN
zHD-+k_k^>0!M{oeaNbz;ew)D{z%i(SKkUY{#I2`9AtAoJ*G&@BSM`)etzpbKOIP?k
z>qy(SbQ-zG4!q;N9LpvQbz@9WTB{wJef8>QYPnupIy%J8D8VLkh!d6Q&Sth)2?{0X
zigCZKN5U=pwYyiAhV}bl(?f2Zqain+7!F;RzkUiw>&K;$`sL^gz~5V1T{gS}iY;5G
zgZflAqA};t+gdW>z~Ut|nycs<UeI)r-og5v@CYd+M?CBDo&yU^ZL(1q+F^y8{tnV^
z*T7a}b&-dE)->-A#R=lP9ni4(gEY0C)8XMUwaNZ*Bm;e8_hi2?Kf`eu={@ganhGC<
z=I$#BolP{D3h2_q3`Pq%qsxN+e!idvul;K3d4l7A?XpDF7!{OAx&b|yANx*{#*R5C
z<coCwiUq7j^WfZx7(?XBYaJ2&5@15$@rEZzgnnXl{=x;8GB6@K{NCt4)};Z)V>VQz
zsdF$L;7OV|0%xyANfv^7%9ONn=`*7@HsIuQ?imql_X~ecp4B6UBn2N`kur;kWf&O_
zuthb42BSV;H@Veo4wu``&m}W7F!igr1SafVF)BGylRlsC`aJzDroQR$H*rGWu_bJJ
z>F8h!Ul`QO+_pZzp12f`8;E&u^0ZUt_+s(MN4?`dGt$r1AFWNv_i8-C`W$>-B5w62
zZ^kjYrY)dHx=h9OiUhj?K%#V2`y-}bP2YaK)%!`d)4sg^MKaH(ezETF{PmXMO2<0U
zr@JyJ5=nqbFmPk%`(*1m81$;`%)ig6E`xLC4+ST2tyC1XForL4fmjO`%OPMI>y}{)
z&RbX0f!5J7rtu9lU+V~}NEwo4FqOHl2d^swQ$jzONPr)O%;9xz7-ttp0uJiCat7DS
z$O^9j5tj#h;!(tf*qTr8!0cjAJVyoN9swBfWDr1ojEDeJ6;+~VH;vT(P>oC6932&r
zk>@K=z~G&>_or<ACW44Uoaj9ER;`Pm{=m#!JvMFAuK$>M^n+gJXm_;lgv!xYdDKV#
zRsLj(^e95k4UmZ)%yrasO(}M&)5~9R><G6pR}ZZ9stJoU@QlwIHg%{RjhT(AAKP4k
zKHd$C3qz8v<t$%LU}f=UV4NVwBi~Xawy6M|NZayj<waKfn!}CnF2W{<!$%2X*{q*l
zwiC9jFpI@X``nyNY1`3(F?_?S?iW_Hz$$~coQDtNYt1au#T4ao_?)e>y9p2;8(Pvc
z6D2cq`(0I;`8DZZ;8~G(5ZnhR)O<<3KgmB&sK8cbu@%6&lmXK(Gl3%XYJ<f8@gq0|
z3Nv3CU<5w#B+*@hJ$iTdvzwOlpFzcAP~W^q97;1jvx4@<610UXP?|fKyhNpukv$z2
zfqotm$dQgL!w*3#1GY!)nkXBWpffoZ^eS;64NMn!{rx20Tv2N#Oi^BoLe}QZe65Wb
z9Xn0aqgZ?lbn`S9R$q@jFr+%5WCwSV({?rTyv;gZRqb7Zo~MOd9F-4}Oduf3eQQ&)
z>|>ZNOtI0>F>$g63=TI>+`xA4L~_mtdt*a4+%v2K&1I9fn-V$_1g9gS8p1@&qn}aL
zqHrCs84@sg{DxNcs6y``6G<ug+?Ye%ETX0*b`H$R8U<J1BurpnQ%^pAJ+RWAy&xKF
z?ONhOlmQLOZ8u8+6F##nYoG6qA)pXEy>WI@d5@GoI0-?lj#0Ovoj{{R6`~SYg^{nh
zY;tJ>CYx<>JpqZ<S-s@p(;_CZi?C!04%l<r4<}yj{+KSuN1#=Dhyq1EqqwNNXhOv#
zG{ltu`y%~~Y8T!m)zhFezM~ADZ2Yp=%TY`Z?<Abn>&eSO91oA3+ti(Xb{nv?j6{>Z
z=aVy)S$ks4Yx9U0Nb-9H(I}6XS&EDYsjD5SO(u8ht4=#tWw^y26Xs;@oW7RQ9&l)a
zV^aOD>@x*DMQ*8FL{1~q0V6%Q?H-2!3GQG_bTt!xzcuDWJ`-Qa5g1raz}z6#>Rs8f
zrqpcBzeR3vlD^UIDL!=iuVWV(@%(T98_CBI1U?DbJ?}!be&W=^7%VXr<4WjdqwsW5
z@r7QfIe+*XtH&}K95egk{Rtq{2a|t)W%)*dUMYjw@{bs3yvJ3MbIR^{7{;2x?X@n1
z^nD-Asl6E<D8`GE=(EbkvQHi^cHcFsl27BqrFqEWN999j%hi<M^E@=BRf6zJg7BUs
zrq&744%nILQYG$!A=SDnq~LjKpq;VDcC<$o(gENXp^kl<Rs~1{&r8=HqMs<UY)-Ze
zLhe5}`6k%s7%B8OslYU|MEZ%3<)}(4dcK5&N?2&Eo3n!+ykH~H+;i<qLD%ywCBe?1
zoP}MZ;>Pp4IFnb9XqC7hj7YEZSj~4%Wjc6u8I~*hEY6Fd>sq3eCp2aV*%M=Rk--16
zbA;b__nX#r@ns=5aZo**8{{o>aT&z&paHttfY&Bb>Jo+rH|b}ohetlrk1t{L97yZ<
z#@_-85FvP&SV8016pWghb;~}VRpbP-vCb4ELz5{aKGy0Kx=`m^jnZW)A_+CRnos)O
z`3*VXs<Zr|%$1do$E9Z%z)tNC`6_%+6ZZ8d-t-1&yR>`x3KGb}QMYH})0yC++%BH4
zEZuB76f2-)p1({H;Yqk}+j40(szAfdw@s0C+Rh{WPCI*}bxuGw1#gk|{WaOr#~pJ#
z#pli1lID1Bs=DKSo@Kc_Dr;5Sd6*Ih1z=}`i>cX!VFtA@c)z(i2zS|z<s8bQP!M?+
zQY;J^*aHC#10eHM?w!8=oAr1+aXrJ~ON`E<+fJ44^DxDPkOBXctqKWb=*KzFGWx%*
zy8pf+N1mcEU#Pm^9BIf5)7+>&S#GhDcpE9~nk$M?fkSZ88~I(I;M@bucP_M7YOOpI
zk{XM)4-4Xiw;cP6^tM~ZPejX5G6tE%WF_uqB~oLMcv|Pn5z%VZ+RkvdFOP_izc;~f
zCb&P|nBG-Sk|o|PapWn+<Q-;<i-1H6hjoBlYmEwo4(=LSGXHPim%S4w3E+A+fI?Cj
zR|MwAJ}L?vaDi_wR(a9mDW6ll`^|+-s2k#b&&DUKr>_qj)mzLXIQb2C)b^n9y$^Um
zQvx2tv}B~YYp%MG-U2Silbklu6=0xXuVxZW6eby^b%rrmrg1?+8Yll={D%k*d-FWS
zrLyEB=2v1Rf$*RXymgSKv4T`tC*I&d8=2>P!ysC<!!RqN>-)Fl@xLXU|2_R&qpTQ{
zWtuF}j(_3+0KL%wZ>$YWEGVPy_+(wbq7m4)8D5r8s<fD{lN~NQ?5~4Mqy@B~3Z&5&
znV;W!UgZ)WFtv-&!rRwC;eW>7^O_uZDY#@&0OAU#<wOC$$o1$lhH{xNu%np5?87}B
zjQCpqdDaJPMQ^8(*TY>PK!EEH1&pN+NCJ))DIh8&?}P5o+GST`<c@U@T+uM>>%YhU
z|JU;&lUM<}h-+9ql5VKWjL`TH!PCW`#<J#$<ApW%|Gz0;sRF;?A_6r6wH*KZo%!GX
z2Uv<o)3Z&Gf|QbhA)oryKoS_30|X*_^Z)!Mqr$)^IA&F%|6kwx|K(@E<`Aq<L?2dc
z=f7RZSE{g9Btv102QVT2GJn6N6d%T;kVvBqIqs=cxk-NgUs1s!wSkR;+n>6b|Lx-b
zAxwPD`o}HNiqP`>=kI+L7GFWKLud%82u#YAyw>o&>9az{pP(Y=nAD#F9I-AxKS9sk
zY35xp!=Xpji|N9=lp;D0mcYk91CDaAfpOkDkDEU|rqPlF(E$Wn1(CF`RsQpjU){ey
zH|u+Mz09p5eA%Z|XKif7=C(q=FL$+Kaehhi(cI?GkKbv1Hvh-*Wq<)kN0GYNi}YfD
z`LC}`719b;DkbQSj0O|x<Iy!pa0%$>^bR!9Vo@2yO#)_=WSVTWI9zwxh?g6F`I$ai
zzyRIL1Avz|04c5!3FJ9R9`=$#8l+`evs8BCol*SsQH`g1uA}bzO<N$h+tr^R{1Dkz
z7z5I>_fG+y$yK=04Sa_mSQwopTe>*)N7)#ds~G6Sg3W#R*4!avYh+54M~xC^GCu$w
zqo)?TrupXfqDLeFU@c0d$Ab3(YW@U7IMZv=d0--42RYP}vixIFLtF6E8?%#tX&3+F
zdbte%EA5I&ME=D;lOF>asxOiW;!odn;?~kDrZFM|c_i0E{MGk_q`x>s97g^$H!Ae7
zU;4MZ>fc=6VvxNa;>)R)m)1aXF91Z^HyZSW>md`ZJW=)dEQ=K~FAPyM$0fipF$J`h
zcw>H0J|bH^0&F+-c70PWmV4rqPf;rOJV*;4GR1+KhK?;Do295CEvn~tAW`shU?$yr
z@n_vC!KhNF9&`o3P_UTUH9`(5kbY?;Ruo9nfCj*<Sy%?9YW$>DKF0tQkg@>GwQ32G
zTN6lUf=I!v0yk8&v9iv6#BpGQ`(I1^f7<G<us~dWrtXu(_0OBu8YTH9;2dcSgW)UA
z6O_te4R~J(zz#+mIi2;J%}g+0h8O5O{eUy_3Yp&^<J?2{Pj{A$YL!L3E9<*mE+-p(
zVz&*Tc0jgMS@amd)k0|2jBK2M8dCfs&D##FR$QE5-C>$wE_+cVp2e{plL5non-sMc
z%+n5O`kjR|{GGGf1e#gw^6jD92@<nY1+Y+$OsbF*Ur?jzdtZUBrVh{$4H@_dl<0*6
zK*RBvb)C0CPdPTXao|@!sQ;fgoPV9G&qToyEICr;_T*pp<p&pdQJ9`AR@@x}+faL2
zD9E$5ka~z=5c)<u$WD1*tlG`o|Afhd`taWU;w=Zx9U#smA}g1gz5x@wAmigk4ey!(
zS0w!x41Kl1QX0Ha!;cdqzYB5v>oI(_Misu=ra3nNVfBDdn^zDS%B?beMZwVWvk&sC
zh(KL>jc)>hU)a>i!dg*yPy;zoZby9J4{oLc&%{|f<J(}aI)Xu;KS^DX9MZLpA#Wyl
zbnGTLD+M3;LtCo187lc;L)edy#N7q>30cCJJ0YRZus2g8BBM|DuIKy!2Z7O%-_g@k
z-@7R9dS?sOX^XBgF?@F=BNtq##CviLcRbTLUJ!XOe(WxGPPv|mJ?^^)W&1|Ad<A7^
z5e_#?ICH1Cymb>sxSDJ&9$zER=_=3+1_Oirj7a6cL&s?)j?j!K@b1U#H6p#FxaVuJ
zQtU9(@uvXEi)JwIvUDAwo_PmoMQKROiFSv0P5&O(v5k&L9ny!{VeI7NIOzJrm~8!*
zcE3(ok32ju(6`1%A_TqG`_=f_FwmibnpGQn@tHk8F%tCDBEQ~|{r>6OU8_obx=h#*
zc;G**w07c~8|Lo4FnGfCIPAOwGVr>+^U_@Bl=LXCqyKuyc=-9H`!KkuwoJ$0{Vx!t
zcd+NuUV<H~(&YAC!^W$bxN1o}=c9Ls%X8xir*Ck_XdZ-pme<7-v0M9Zi8tq|%QsTM
zaNSgTw}aC%X$c4=(ty|Ia0$;BVL{sCn1XgQix+ayj*@Rz_WB<~``q-+sxnTQ%dIas
z``mWUL#MFY8P~<ymw{HO!L*!<pPUkZ;OwRMyF%xrxm;l9JoVOgx?MjwE3$uDsZ5wG
z7wwIEe(ccnIeP@>A1xtl+?vl|58LKJX|a*-3ENx{b=8PJw|Tb1IM)5d5fUww?t1en
z^MYx8rT{$wB(toi9y@lyu{Hi>vi$A#uqB!gGl=o6*2|WlJ6)a|(aSP*Q|itwfT4>6
zco!XH!lDUWRNAFo?J)RI3(CVO#*BY4yL)NN-1>khd@+Q!#?NbhheI+!u=zxMD}i_z
zF1uvgm1?UVB$wGf9mX}JdKKJkDROfonUwHq2czSq6GicpQ*S&tS%gpU9n<^~Q*4Hw
zmv}~uPg;3w^M24-k@Y#s=dkKi5Y3dI5@|pF=T`Jq8uy1E0%((}WevYy7#b!()%Cq6
zHBq$zuig~QFS|j(WVXjF&}unNFTuI6nFzi4iIo0M0Ntw}bey<>WY@!_vWp;=%ymW-
zgTYQT<sNbGlKc91M=-z8kFO+1bqCIu4J-ZY6amE}@Lcn%E5zLgBj9x5px*m|FD8;5
z6r1QSffX?mC=g@<cTqK@;bSs59~UWUU*meOywiJ2&j-*LCLkYf$Ng?_f7StvLdOAS
zjvVW+6%;qq%DG11KB55-4}y0efTyxOujo)Cm<fd%tVUE&H9FKR0qalQWMInS$PjI0
z(04ER7uCE<UX(*1u~`%vfyP@keQy`HxFwDrQW_-dE)R{amkDW(@x4NPjZvJmex`)k
z1BWq41Ox~R=WI>K-<UF&LA7AhDf2qbw@X(#f^_G2;6<ETe4F$~G^~>Yq*Qpe4hU4y
z$fleVAOJPo<X8hP=HH0UL~aXU#O&^d%4xa_=Kd0pj){|UY>Chz@bORm%zinV(4D<<
z*7!v|;7W$iD{m509*21-a#pDHMyL|@s+jLKMpb%LI_|lB=X)yCz?{d0G&r`c7-H*y
zK}U_8sTx0NTOnOzm_g$E>^Z%Dzs_{{W!<oufOI^1<IQI_kmMRW8G8;h+6j<cM%tnv
z^82MvVj;fZdG{mfBqym?V#MOu=9AG{_t0#6`xEtjX1%{=+)dTD(18OcwSV~fuyHtk
z7U?xg7Jrp~<keG)rfGT##@*wAq9gXn*S^;cgfaZhe}Np;@YmmU69o271j5)t@DX|4
zj(W-YiVArU?A(-n2U6Sm0!O(<kihm2zXjZKk?YxE%YzsEq?dv_d<jbtn!`b|x0Fi7
z{3Ykn`bfUy50a$LoF#ydS$4X84BLOiR5HSqlD!$)+Flp73}PDd^rbwgR0Bh~y*y|3
zhjCW3TbUr;4PHow54!sxFkvmz4OD)2%@39ntf;u`6^NxXtx`|vW!r8+dx`xoq>JXE
zw)67?URp6{5yR)BDhGH7qn%>vB<|0=QrgyMO~mbg2XKf|Fh<n>9!Nz$t5V6rKa7?p
zj|TX#{5pd`ubV|MX2=O_wiP92<Js$H%}98C&JSkQp;xMJc5-7Ocy8r1h8f>lO5JJ}
zfirA_RZg>hG?TGdw~r~zCfa@M$cW=J2sw(nv_X~cI1EAWa}|h0-;tzp4D$eAD6Dn$
z?qj62FEfY;CH=^H&p)fui&bnBat;XqZtNy7bb3kwl7ST*_uI4SMfYgVb}!uwaVQ5g
zsMh`WTcSpe{PXy7FG-(^4hk3&)B4FemQV?L{W{P1&<0Ql5`a8dy4-b8<keBDG#Vmo
zR6d72WtN2Ab|$XRD@D@%;yG<K80&<%@K*fPZ`Z*-J)79flGQNFxCE4kXb@c}E_t0Q
z@D96xF{0_g>4t&BHK<<v0W1Az_AWItOs1Y8i2rBgfJ1WsAt??a&y2wXL*-f!fAF6?
z!4GEc241P9Cx8Mnt{~{OBd70(nq2cZ>g5FjOawTR-TX|{o5MXwVh#xJxR=q#PT*WE
zLn6OcrO{9O8>mD3y<|9N2(an!XU`ErFC?X1wE${sgJk(45E*+~;OdY@;UdwUhPa86
z?gZbc;@Le=DnEfpH=e2Nmsrbj02`OrG8=hsBgjoLH)MmQdmkT?GF<?Z=Qk%5u`N*=
zt?dF2wQyc1(|Q8<aT;CygQt9N5g_3ckdxSdOJCk+EThOAE{=oB0Hs{*poSJ1^=&-A
z#H_@Lp+F_V(CqD56N#O|2?6Igv=>~X(=0&!dckX%OoK3=MyPj+Y8k)nqUEKnRC_9R
zU*ac_;9FN8vVssWtQz*rC(H{cvsK`r2=>-pwerg^*pK9!RIv1PlPSq??pjame6Le!
zCdF*!v5SzXvn=wM{0yB$RYco8JhU_6;RBG`x}b8vwgLZ)1&JD4VsJskY15-2?WMM`
zFELLT{E0B!w@K)EsT#9`#CljeS?Xrm<4J;{`+3maqT+??30(FTz#!Fq@$wvL+$i4g
zca@6mh@Fw%2R@0MLw0_0hFyYzZ+$(f+BTNIb9W29+k#U;`{ZOKk?zm}*>I4%K8GIy
zf2j47heGYS$f3I5B5dB0%i_o4gt)9W=o>A7`!KtMT9F_AUT<`9;xm0XrCzjoR*rRb
zZ$QNH)mJ77<6YsafDp&IbnuajETJk>j7M2a<AI{>rnvaJQY0bPQ3a!g%xcEBPj1mf
zldr-M>qHuJJiDx%on!MHdzZ=u6oyM`4f#4*6pYd}Z8cq>In>eLMaFj`#H>udfm(3c
zRV88c4r#zRf3vbKMgSv+;4exxhAzoAYMjxK*^Vw}97~gt^V+6qzYV(AEA{Y-dqROe
zXf_GQ*5vrl$loliiZQv*=pHCz%(Tzu9%8`wt^xrnDF!h|b<;JWE-a*4t~zhDRu5|V
zg!`Yx4UTbs9Nqu(I2}g8D)dl3|Iylg{}V-=<Lxan_m!`<;ZjXCnm{J}G{4?dQ*wKf
zTKOUZ@gSlu;;$VaAI45NL$k5*6#3`<(Z3p_zFXyS0^RPK-lTVm3}x_ZBBrs(4pjwd
zRbpZhgH0S};QJ!Q`}86H@8hU=4yCE?$f7@g*Uy_@;ZqSrRa4$nUUp;nQgOOsqtukc
zCzF*7Uwbb7X8-Lp!#wQu@-%9IWJGU-U~PDT?vzH*Zc+`K|7831@5AJN6F&~8SHBTq
zt@Z~WoU<cmivK>)eJg@8jqSZyWJ9#aL}J%exp%}d{HQSb43o#4(+!RI7#ubZr+s%U
zUE*6JkkB_Todx#rtvgK>6I+N+wFaj?@P~+%!(180O$^U*?jQ6f{hO9>U-E4Tc}Pqf
z-Xd8VBXK^}bIW(kD*cKfik!5v8lT+DB!VO&F%)Y2k{LB=M~8eD1!ULWO}x_2_B)qg
zpvF(1DfnE5pBSPTM+bSEx@}^bjkVx1?6M{@ig_x*a1T#+K_Tu!<vfkTKTF~nVK+dB
zrhdwAJR~z2>x+?yT$F<owLB!=(>wfwUF+QXC_FaNIY5!{FIj@1db|Ebfor)M!Qy#S
z4!%_16W#^o&%f=!(Oe~gZW0n>iI*;OvUK?3f!v7(VVsVSHNb9!9NL$L`E6P1GLN=J
z#j3IxObQKO{{e&OGtK4x&?a@qTbJww?_!CN1-Cg5<&S|H8ebAwC%A(P{GX}Y_1e*K
zatof5%VxOjOy|68R^Kef<H4u<Q%4pN!;eR2NiPeRi<?n7C*ymmNi*6vB+*y!#AOB=
z&PrVhh8=CP39y>$8(;0v5d-(naaI1Avd_PLe~^rD%QDK3cw4vurVu^kS2E)c^KC)7
zPfshSd=#EO9v#jb-LMSF$XAu!%_wI47U|p6dKd5FKGV5FZfKQ)X-XSiaQkaTYABH9
z{ugtR${2J_kczn8wG-viLps^CqULY5AF%>4J-M7!=h^9)8aeZGfBSO=)A)XReK{WT
zBW1y|ji;JiZE_yGC*Wl*q#%zZQitL2i?X0LR^{NQVERHZ7Pu~oz_A}f;+^jgUHCO|
zC7z96+Aw_US9R9mtG~HBE-Iw<%}$(bV(@-hWWkg(XHM3^_#kB&c9DO7pxdYE=EHp7
z@iL~lK=3*qz-ER%OO6|!HIaQhAQLy(ruvbZgtyomy-klgUB^yfqLX%>A2x8<4A-dq
z(EE(c*hMy?Xm$7Z7YW`~)hA{$UivlOvJ5FL+?%!2WtRL!y-L3C31jeuSUS$I#!tl|
z7dj;7A4ANQTMmLa+elGQpCciY57BsHtHUtZ?*?a7cp16Bt76jFmc8Yzo&Jq;gKjDM
zN=42s#Fo*Tzh`Ti;*t9A$Yg<4%R^(-5~V--%tmnB>#W=0=!i%;>4wH4n<U#$t}d!9
zN%SRzd1SW<S-TUxAtWS~E!*j~&JOO|zsZj#&#Xk^4$2tjsP~Cb_I21nw}W8NyPyAL
z0sONz^C0*hZ7Q%}>{O89qQT;*#PC2dGMM9_JU3~9zrNFE$))w&(8?x=#9Q21Z(fWa
z;=r1N<Wvrc+}sy^AVz&;ii;3TmMi@pkjcFuUT1)JUrYwdzxZ;K;E1X=TEyZU5NQS8
ze<XJp#FwU26HOR!T6y~Jo|xXYk-@vBjY*`D@WVy>ashSS*#-a%Cdn}M_()88?Yy~~
z@LWGZUxk?eE81rJ9aYYukF)z8Ziyzf0JKQrN6D8kW}ml^O>db3i}&h8JgBO@`W~Iw
z7{1%7`ZsHYtLC+zBi+c~D3rq9kR1c)k%%ap=^G)0z-UfpHThY9J~MMi=i3W$-d&Ez
zAzGqOC|ux6_@Bl+8N=iLZtdycL1?^+Rhy4h7MmHkRD9q`J9(STHx-s74_C2oCbSHR
ztT}-!@y4+jyvLW~7!tgBguz9-HWZ?hp2E*|gL<fctCku<uCmlN(|q2CRj~M#Yn27J
z;%^$^Aws7j6EJPo(J~{D=n}UMv@l)E2JY8fLVxFN-pia*bR=4+_fqh1=JH1P9|fFC
z&at<I6_$&=DOw^_2%0=C3Z6d?Y&D=+adE7!3wXBc@V9K($RBESX%luzv94O<eRa`~
zfb5y_Pgph-2WgTxU1QP3b_4M?DQ7&^YT*}0^BeB(LvFyqRD$JhYe=)=PxPfzZA=Qn
zSq5Q-&{j&aKSfy3ponv}+0etTp537pZ3}v(9=qPwW<@>aQ31RyKK}3|_RKJ!zY&*~
zJnw8kv1+NmEU@)04N!`|4hiEg{fAO)z(@0gzMd>ZEE4k*re6w`(oA2hD-K5mL6oE-
z*7BtB^g?G3(vaGd0qCU;wD>|rwL>De518yd{6$JsL_ZovCoW11A~bNi`IVsjLPHa{
ze<(`~Lhe^#;FEhg>$g9(eTY9upeN{H0=gpXk|ly)F-HV>5I*Qfa#eqdMoQbfdJxNR
z`Lx-#Vq)L#mEXD^U^g0ywtRS4v5j7)QSo^-fH3Vj4@}6#81RA{j#cSUOwn{q{`ip%
z^cM6SnxUy~1xa!Vb$7&{LC~)rv{MkEoRXRnQx}u(q+UL{*x_hV)HhJUA=};~^n8Fv
zd6Zk)YWi@BW>`Zba7|o;M&#_%z)0izInJk<`^({|p*R{B-=btPHs2^+_*J~qwU}15
z=@`lJ{HP@Etg$A8ul_P~!cVYWU$gmurd(?{NbL4lqqAx3Y4s{qId;r*^+`v`q=P8V
zP^}9lEs?4*wTa>U2mOg25iRIyzT;1I!xg8E$5Z|uj4b1{_2nR4v3CqOeF-f(0Mn@j
z_N*Y+eAHP(hz+yG+YI9|E^3pat>lHVF#$if-9q1tf8|SW)~m4=5#6aZjadhB$B~D*
zn?IC>tDBw(8@hJOOGkHkn4F#xz;Q3JOsNzkj~_#A7)07HZuuBHyOuvUG?g5{K7!4#
zN8GD(G6HKH??FHNQ1%W6l*Nf+?MKf&h-dF_ok%FO&hLAsIL031GQKpE^o57RpYW4D
z{l?YWP_{jfV95+-&E>AYSpA8K53IEtlzrM;FDu#wEf`AF9BMrkN%dUI4AT(F(<FLR
zCF~y4TfRvTZ&KomXEZ2{3(nugCeG)hi)_H0Ovj=L2@nIr4kv>^N!>NE`pbH{R@%+1
z!47OtH&fBZ$IV-q{a#=0oz84wO{;BE%im+T8#&_BFNLZ*w1iXZS95C8Mc8`J+h(>K
zQ%pU`6fqP+OLf&9y;@xt4^)nD_R1=0npPx!IW8^Vv%W6d4`!<5u~siCU<o!9(#g&j
zcEvg!@Qa(5?+MzE-aIkyc*0v^rR0^gGJ#3hP4QYJ6b1`<Hz0>B6}j0k5^*Y~06sO|
z?@h?2mk0b8;mi5On9((29n`$H&NkuyY`p+lWa}3ydm!xv<2{!kd?#62`8+{SAU<2#
zK9pRX2|_e~at%~uX}G(d%8A~AF5OCqVm2)IU96>#Ca)}p^HE`dh{~AsB9niizu0KH
zlnX*rByNNnLLwb!P0vppdnP-4`;>3xp2|t+w!wQik&%j=N_NjIh6eB)_RRhbk&Ij%
zh&u&63HU;sN}y8iiE8*;VCjP!Lm!YkUdmG9FDmRb85;m?zzV`X`64Uw73{e@eA_J1
z&9bdRISFve;*b`q*#rorYQzM$xo}h=HZsWa(;lTslVm`baJvj6uYJYe>ujVV40qSc
zVr5!YYB_H{3Rs&u1dVN)>VeK*ULhMuF4Y@UVv@F5c~ruZS^P%Z%D1CSvI&N?rUVwB
zuvq?4$Z%i795wFTYgcoOm(LK?gv+e(-yH|x^=<&w8^3mq&~y3_Q=KLCQ4)LjD9xY*
z!x(9VCp;7<h&-Lv@?FJW1(C(hc24<N^2C?OK4m(Wpd~UWx?LKpQEj+f-`i||B^9}Z
z<f9^0X_XU5F$IJ~_%yx7Je+4=d-*D%2{<Na_Xtd&m7{WuzVm#u9{$=A5yuOQ$3vvC
zo?;`<2gPtt;q9{yW8CYaRB{s_L?nvb?kI6OZuI80^)RYSAWd#zK%QZch1r|FWE#tf
z6rS4=NM{`O)H!APAYw_O-ekc*2LC+qsCr}?E1lK3(I=AC^^Fjju-vkrfHwu*5DX|v
zu|5mOjWQ(NJsfAMe(cW$L-YI!N;1^%EeLa?J|z7rm_FbxUfMd<O%yKVj34ojHVl%z
z-S>kU{*{edO)lPib5eBpVOqa%YIl!*Km5dW>;Cg_Jq=8QYM}ZJz^RBI+ya>ir`D4d
zPMGX81HSH}V1Vr_+(m;V;$qgM5L-kvfThb#Ts3d2go}L+-Om+g;SM2R+_CR_Oxc0|
zn8=D_$aXXCqbA8XdpDZzRdBe|L{aHBCc5I&Ihn=W#4z(&XJi%1_VLbU?h#B!ZcgPP
zaIa+O^fNB`DRDb@BYNe1*ZAog6|wSror;w1w($?<v>7Wwf~LbW1!*-R$(szHD+^vi
z)TgqPj@~4-!biW`+wB?iAr0*HUYv;%Pz-B~*h{wEhfY)i`xSoMziET)i1MK314B3e
zA7gJFRfW2??<y)G4N}rwlUBMUq(MYdIwh5Ekd}~6rCYkYLFw+6?w&|b&O6s$Ywvx|
z`Ofz{WB3p1kcn?Sao^WNX7HQgyPI|jard@uR>^Q+xC>Yi37-2t;APs~EV{iYFH!B2
zfDj=EdA;z=8{xRf|L!1O8h8V*pq|UbJ^Q_B=(OmlUJvFh_``~A4jt{W2b|L_6Spi?
zhlr^de%k`U9&gHl$SbF@&qGGxu^imy#y0UBm}@qtLJq19-trKJ{#eV+ILFTK?0?LP
z%0RP6z_XN?CIGscxHSWciMyNWeY*o#p+)YSYNG}XS1+7dnf4kXB7I>t4tP*ZYip5)
z88-qUNzZhau^0in`6g2h_T`Qyv@_;`tnrA4ov$1Z{xBL&l~$Z*#dwBrdUi1#NbOe~
z^y~fW>gumC5LUdNh056$k-vSNh@>Dehfxhn_)~J*sB)$Jj(stV3DEBR_fQ$l=f6g1
zY>lAr+8in>6@5D3zy8kgrU9Mhx!%%^YM+yC;^s%>39OO8iT~2kqER3KZA$U{t32HM
zoa|}E+?&)ayLsEf;u@_EoAeQbYc2Y9@zMcz(l>R~=oR76pN8aWnY@3BQ$%z`o#RJ2
zsdsUeoKVUwk>yNCF4~3^!yQH*if4u**9*LE4iHKC{UHgbNWNS^h91e{=vNyKs+DX>
zVF1xyZl6t^uI6Q7$P<BIv2XUm%YQaJ!0+tgfe`H&+jf*k$xhgvm!+}X+8Q%UCD*K#
z@n$~yDnWq#T(`R;4M&&wRveSIb5rZ`>?lL2%v36=qarri7*}V+-t957N-ONV^u~$j
z69giJukDC4v4VB&xJDMLS$vAi*TZH!AsMY*?r=H~N38Ru$CLh!;x#e;_+PL5xawp_
z8m3v~UXYUbK65u`MncD$!0GB2<r3sZQf7HcmRr0Z<gPo+yUU9Fyxe^K9pv$BZJuBR
z)W<?^De^MS_k?}A6NXCio_KxA=Pe)jJv`Kgg>al+J`N;6aOKxT5pXmTdow!tKI!)-
z{s%@MA$CTi@yiW8TSXt%B80p7?XlHOlY;_iepP6^TXo?u9#IGj#>rS<M64p;sYhbN
z<;Q|V;w2@K&wvn3M@t&l;4~<Q5)xXqH&^{j+@;F1DI!@`+O~oMlb(VdHa_UiO4`83
zs63{Vi>x~KDs_*Jp5Pvod=5Z0rkU;lj8n+)SHX!y*Npp;$fqdhtzPB%HZ!X-KMZnf
zyFhzl%OuJS0=FjKkX&lEWT!hvYCYeaPwc;6AhM^UxoF?Uxgydon;;hYHyZ=dpRHPS
zFWch&W*?6UvD}LWS56>0s-GGrRMB7LA3##X(+`us%N)Xj?G~0E_T}TC2pCef$|m(Z
z&dtulq1ih&Qw-;y?_=~CN>OYROvPcrrIqGw8j%BId>Y4je*Yluhz65N^0Uq`HKH#}
zkgi5KG{#;Zoe=AE2~wXPf9K1qD$KkwGI%v>YV!>Bw=4#HNPD6G6*jE_uT_#jj+M$s
z@Cvf3CzLg9>ENa8boMmhu%793$=&@}O?+g$FLU$DL$9?#aa-r~8CvKcO2OCA_lcM^
zXfTIIHQp~Gr=4z|mG7#@U61w(OC|{$#{J2{HTjO9K}X8*)Ml1@pnLF}-$#*c;w<mG
zqvJV7&&eOR$2-h+8#JufQh@4JeUo)yK#Cz32hDJN^tR*D|7u6Ep2?(9g;TRm9O<FJ
znDh2}PR`#_4*f*RHi~br*ke#kNMe@Zm4S+GGWV+6WEs?Is)TG5cL0J{%T!a*QndN8
z@qpg)+MUXDHLRCp^~=h3Hdoa8{e*>lnEv~~_TLSf+E{0aZHB8-#R(>XTPWTZDCvy3
z`qe;v+=x=fo=&JC*7r8lUttY%&FZ41?{}TLe`|rp!lzF3k_979J?9ZKv7aX%6Kke4
zK^hY2Z$49nl6$=>nH*j8A1*JBJ%Jc1sCJO^Gk+Wn=YO~aDcV*%UWeCE6Bh#AcMGSn
zwa-rTRysq-%FEDxZqu$r1V^h$F>v|F5Wxq~h?nmj_G7^es|+|$yiLw)kTNZVA~Z#s
z<R@Y@tSd1q38sPK7ubPbY=WXW6w|ST95;93?kw&T_B@_>=NIA*J%}eB7mMWhXs{a@
zmk@qDC}RUr*K*AQC|6gn)lKxJj8|K%{3!03%>2|r7f{qvszvSG#ol8fsXc0&s&#0e
z{jAy{RVL8Lh&nr4u!H=J0(wD7Aph;fk5})ipQe3ErZU2*)*3$+9L`$&>3ZLs7U6T;
z8OT@t{D8cE4+vye#(D`E`c)Nz$eooDo6hWUAi9By#fiLzZ^?Qg9tk&@2P)S$nI@S`
z@jI7yM+G~_E3x(QNuBz_=2>64?ffU<N(|Ck(qDtWo^AeSu($YHw2}Ok+~|j@w|d@;
zzYp0-_!WKivuQiQt>scX7A?AhuF3uROcUR0Iq&%Tap(f`8M5xQAdp-PW`6w>RxV=i
z<JJ&FEmi(-?Bi3hS}}xop>NPnevaI!mWk4Hu;u1y(^%M680ZF__@;`kIncS7eSKB9
zJAn8Gwv|CYR3~U!B&yd~Bz`Y;9Y7-ss}gr%B~}kRGThTSO=vRmvnT4#b#1xuj=UeU
z)=c{3ey%VoEE#t#`Mhu{Lh+o@a2u9s(8W6c`~5q_mK}dk$O+&qYqtXmJ&fiCyXD|p
z24-Gs)U$V0vm{A_zY+mvb^!Z!l+UW5N2?J+tXFt+ougeHJ)Vk*2$~YaVIG`}|Hu@4
z5FZm>az?c&UKaI69zRMQk5W+5-7ZH7x9SdG7VORrAd6X}8#tc{XqcYql?~z@_I<#1
zCWIS?Hdo0%NT23Wf!at~sO>+=RO|+^u=f2d12dH<Xy|%(djf9+E>6o|zqH?{^A$|m
z5L}7ql|ZD>H2?WAn$@a|IkYN}6*1l|HnkrUUsj8cE}K#PppMBekuzpPlM^p4g9+bF
zK7yLim?hu!ib77A=UbeL(qqF>hInR9ne^_+I>{$aAu32`B>X6r4=tXGoZ-=xX=(b4
zr|A=M^{0TrF1=3(v?I^+H@nxVTxztS9-u*_^X=A{DxRQ$BZ&~TU(5GAH^E=z*r?ka
z-zgRkc?KezQ7UGcopHNx^OK=fBg!lFqS1mFKjv6n?c+_AYVj7w9unuMp8m=;sFBK1
z(D4-B2&3P*v4~w#qpSAg8R&zN7Pj-&wI`MBK5XmWyRVtmYrLwS)~0NwYWto~33iuQ
zLy&Sv?*Pq!;G6ZA<7Ja`*V{^HH3(R1!9m?Atb(X|5hzIQyXi_g-AsXMLPNyo>0USE
zV(KTa&@$5kyeua#24mv?2wAPw0yN=5xrHtvh)scU=L_-2F^t?TdJhr}n$fH`cf$?L
zsPWLn^Dh-33qa~3fXtT@<<DwMT`G+JbZ!0oqdSuh-l^_Lx46(S^Gi<LD9Bv{^HiK%
z0Z(Rn+BXR67YZ8IZHGs|L>$lSLQj%~XVz)HN{iN`oXU9p4!1DpNr%f%ecJs42Kolw
zF;}On*+c1%5>c(xkS>=lD9f*NyW8j{LUrrG@l>(HXTcpXh@^L=p%ZGEnCyf^>(Gz2
z&7`?13Y7@`;1qfH`a-=OK{M(G^qd9*BqO{Wjz)4#zM>ZF4m5p9+M-h|vX6J{*^@cb
z&_ZLujYqE>U;a7Os4?`o8|c6)Vz&o!#%s9fk$1=uRwx=XJwuqzQIiY)KTpk@*KRVG
z%=3B}<8S_ukq$P!wA8?=EKlzLMF~Y$8y@8WO+nobtyerct+9nRxwEUgB=9yKNezC*
zKW)4b+s7bAC*Yy%SXUui@s(6_I;s55)7+V|cqR|C^~dN6`}CG@We3aXM<_}toP3Gm
z9C4P6VnX+g_jS->=+046m_Hgawr?+K_zQ*B&h*r*4=8CeQAIM|EtI72VT%X`bi*f0
zK&thdjwtcNZ{%2;g950r5Z#Z{vCj~aq-dVm!%lrE>>Yid6c5UeS8MPG-zk*cKPFWo
z#D_K;7g0dNMq;Tbn0pM#HRyKTB?cm;>+c=;vj-PhC`07Jb*Q!bq|Bp#HS*$Yy6NM|
z#nGjuGEsrytGo4?6o*AIGIoEOkeA3~Q&vI38Y<otLz^2&w(C)Bp4XksF^~5VG4JYz
z?P7)K5c9j`le7DImX$@oq^`$#N)>dME<q?orueU1gW9>5&3X|G{M@2`jrWIX|K;pr
zQc(PJbh!r_np)$y`txC*c3OjNzxvSHeU7*{8cQ?sygfR8F232s-h<gAd#fBzQTi4`
znp%Ah1E0j)sG?6I5ksW+As=^a+Kp+6Eb;4^)Z>KkpkA*cYJb>*iaH~6hPP_-!MJ*L
zYIhNN&kLJ{y624XHC1tw@@e5-_0dwW2&bebcOdb%^cq|*dRyXArzCV9J1%JkdPF*|
zujqD>P|Gx;(qaA^z23u<{dzTC{swXN^j>u0?T0`IJ{%k447#v0TN~(WR4I^XvfRd8
z8J?qT(gxzpE?|(4tH(O%L@xzp%hk|~(lI-NJJP<RqyB7aH35xxUhZ)Jxm(0%`rA<@
zP4k?%g=Nsp{|$9$^cR6*dcORH9!6SjCGq_y0)ZEFr$TyR@a@t352x-$D;m)Rj)>2a
zM4N<Pz6%RmKRG}+GS-ax6;X2rWPqQJPQ8}WmXW)#&kpJ`L(jZkC3%UbiYu-wG!pJS
zp#ODX)d9uWYp<G*q`l3gbyak)1k~Q^X)RsX))XV9GXy5^>ggxZKZ2=_k3#m<hWoH!
zyfq0=Y)^}lvX(dlf4X^afuP1HPE_u0H-}k^j-S?Y6x8$)rP#OlPDA)pYt1>18Y^I7
zcQTzo523^Gg|MOZDLx5fSa{Bz`DWX5()iSikPNP~RTI{5Udqt~+n&C};H&{6p7C4^
z-_T-bphC0?TEl;DrObbY1x+e6Lih7(`}+_xzrNdkmqrzM$oBm8euV-bgh@QF&t{Yx
z5}P&c-^zKn@dj3LQRQ7!dH`<Yp>32s42{^0SjaErWK)T&qkz2xyUEb1_sJ*NKr87X
z&m!@UOf%gQnVoR+QWgIzpJJq9K^nxYgOu5NKP>2zk7|995go5-JW9dKX>G2RZOMC~
z(9-=5<H$Z^K{c-ml655he))m2tWj^J+pWEFG<i2HZh`qEQG?%cHLDK5dsm-DY6?Wv
z>mUPgl;EPHxiX4W;!jMcOM%Z<izghyM4-ZbdCT#z?=abhkB;U`!mCcct>iU)29M`$
zk1L+z6ZW9e!77vK!xw<80!501LmATv()0<+iORo@=ZbK}(f_jz$B@MJX;Q?is9}|o
z$hQ<qZ&F>%z_+%>@@wiAqjWvE2g+Fepy5ZDf^W{&h+N}S2dG7z(r!Ld+WN3a`MQaC
zn`AKc|BUqcRH)w`^=zFdo`p~B$?avQ1-_%v5Z{)=xo=hZn@6S)4HG`;a*g~p96IwX
zD{TC3P8}eV>V=PF9RLb)%``fa9|tF=(*TabLBwS$@8_NR1kG3+^OzLX9;>nm4a=uy
z$;~oB4XVH|f)aB?$s_hoGO!8{t)nyTbdX{J*S_QNEBtI-`vF3JALtGLug3&k&mp95
zxbG0uFFjo%ntl(F8-M87i7)h}<@9AcU*4kSN4negzOmF@42e@jejy(z3Y8%E^g^KS
z?KC6Ow#O|cwJ9IIb90fd$Ss@%baiFs>bMZ3nT!5N9KVkt2F|KSsOL%kZ9th~hxlWT
zt<E;5X&*>+dI8mXHA~5G+nQV>&W)_nivqLDK~Ji#M(YjKd^=$;l&wXBSxzIG0-eNX
zyI-~Ni9hEClf1x2@Do0$V5smm;c9l+$LdKn#p#SAFk|+vd{>G=CCa@|;N15smA+J?
zk^ug$vjclYrMW+hXeQ9n?-DtvL4seiOVrtQaWKw{+4JQW@+NWjG$)R^vVv<S7L}xB
zjxp@U)C8FzPq-|<S-cIHR|^ZFl|&!QMwa?AV$8|Qi@NK4%IQ2`-b;!22Xrao!Xjh-
z__YKk_?)nyMpZF}z5GDl!QGvQZs}Ad2dACP*R^kcZ4%ieqe$UW8VhG`;8ki2f@B0V
zI|yPOe^?%4RpY1lQrxgx!0K|E;Kg#nVTvF{a=>S2T3vn^5az9lAB$45jbMrg;p!@N
z1r2tVbQ~Vwe1+o8UiK=t9PsY48<m?5;ljqwJ`T3*OjQ2$@IPb0h3XY1zr_+D4P#3r
zhpZazz?d}HYiBSr!t1%9s#^EmV@aBIucl~*is}c8F2qIFjl;w1IeWYYIM71C{oqMv
zfYqkGx_OW^Z#Av@LyzpHz%xq``tNXf-9VXWZtx=W2+z8+<Jeu{GAntrz6oZxbR&6N
z9ys8EXFmbDRSIxULY*Vt6lLVjh&ZNW{FvrZX7he7{E8TS@m@OxZk$f(oD$>yM^xM3
z+qT{t69bHw(+cqsgd9{WJsNb`GY5m~LSCS7b{n-pL1^HKvZo`D5%JE_3(Wa0qW*I%
z)cUb*C9p`z(kl|GLHOHK+8yRjcg!b@Xos2oyfu;C4Z1y73qw3R**D!JrIXW`pdr<e
z?u;1XTdy_kMzRk>1a|`qFnc&Y#voV9E7{+EY>8hZ!_$=ymjIEPNX@m!qi7@E?({OV
z#n6X~Hg#PS<J(f8__AE}mP=rDQi#8_S1mXG>=%bjhufa|3Oj}yi$EG_$H+IyW7tkU
z@v6zjkrvVuX^!OcW3pCo>%%t)fjDAEvG0>*!<DcX6<A2Y$Q|FiBY!d?OB(QNe(tv;
zL6y<@{T^9YejN+nB>cFSIov@}0I{saCh%l^(QHzIobGCUkD-fymmby6KL8B~yj(X~
zwdt(KPhPp8kj6}{2w(;Lc$7h18k}a=(>TE#9+I8vgW{#1ET#CzAc`%LeiRpiAVzag
z!yHO$r}*tf?+WQ^9Ke3T9g3uK-{?oM9tynVL%_IYjM1=?yvQ1F6X-7nWd}#s)HuXj
zEmA(^Ty&!h5|0kv^3+|@1Ek>(1fJi;`EPk5WreUg;+N<SE6$TY8ckw$bZ}q?wq%5V
z3OIhr5Y8Px_|O2BdsO~=W+0AHedicHu6o)z<L<RH2VW=&)OH5aUVfrmQgs-iDl*1d
zeM3wY$abXrt4h&FI?^l1iSa<A%{1%Ld2rdby4^EU8Urs}M8~IwrqWHc$r{!gs-AI*
zDx=Hg@G}fLm(6fVb+aTz=&{tj{|T)`i&{7b-8Fu5zlQJ6m)sZzD3~1-QdQy|d!WOr
z8AZyh3UT7Vp8*^cZ4vUT$cH$R4dlVJ7}v*Pk2LD6wn}y6mKJGlf65|#H~pYj{_V-=
zV=Ia<TlrZ)?btug67dxw&Yg1v$2B+$99Me$t*9&ue&nnoX4y-lmT(V?pr%lHC|;7Y
z<!0FN^cO`C&j<LA6S_O5aRU1<{PEt-R$Fr$X~&^Brw}rGE@x)C@mb_9V;j37>^(&|
z`*8p5ho#*8BHo*JwwfgqjHD(bm7MxRpc<Tv0OvB~4^l;+;wU=+7a21#iL3D&Ucyf%
zZxHheCR(B&CeO9emxqs-2BgYc+GqT1MdLGKu{vii?%tLd=Jj{%p(ChZ6?1p}7q&yc
z4SNJdQW@1fGeu=IKyoygLOD7<e}hf=tz!Sgk0;NS`|UDQ^hd)>xBUaQ5q{+3u_pnt
ztd<urSPq9ChGAFhQa4cRO*~IwAd-r$qnE|UqSg4KWo^H*E#|vT<0+oVZ$Kr>`i%BR
z4`?_r$-cm%jr^k3XkYQctZC)@h`4iD2#(T%vxTiv>?a)I!%LwVW=16kj@J%8Vh$Zr
z6Jv=CCx8{C?SNF%`BDkXcq#y66oqdTrF_nke|F<A?uFW!rK1X-T!*@021Al$>@igX
zwX6eGe;n<pDt)44uuziWHiazkiuC-79@xsr!zhp-C=+LMF2$dhh`_QsR&sws7rXxL
zPdp)~zfb&6U-ta(xODPZDP07#x%K=RF=v^m))>^Mi$*!wSfc@T?7MS9ocZU{>caVQ
zvhkll@+X$MKB>}2Yg@bKuvLx9bEMOUXR-<az77q6oaymN62ZQiYQkJ1;ymf<Vp1u&
ziY|F-sFD$TC18L$p|T|&th_NZfx(ZTf<ULvjyL}&qdG>rSg?0RFWq<PLN{NCu9`zO
z>x&Sr_WaO7Vv#=mVC`7zvax{XT*J4uvF!U$gK+{mVTs!Dn(dQbM@j}|*)NT8;=>u#
z@A-C}nHQA^CFj|z|7_05w$R>>nm6e8fgaP364I=qfCJ<M+&X4T)!4H2rLR9~QOc}@
zz){rBgr<SgPlNRKoFTLyxaKEg+*UUj-Eypyf1M^r^Go)2x<**SB>PnlI`0=59%{0-
zTpGSo$0p||Tl^*DRaIICPPC2)j9zh`0E+AOy7+jqMW%Kds_<diWm;hDNsVT>*JA^w
z6;a@|-Dgkt8YfE+J-dGPuV<NANb=4wo7zc;?Hl61g89%k5IyWcY1eiEQ2-Aouso0z
zT3-{@@_wcj<<VP9mrQ@^Y0Rg0-xtUAX5h9`n4ssmTFLB;8j6<<$&Yxa0Qb!&4&nwZ
z(sNRqNE2EO+nfI1yu*lp?T48|UAk+0RqUFGgE)~giP^bY`L4IKBzaC=Lg$Tg{8|`o
z{c!ED2o46p>|jQ?>`0=xKSXN{pVlcvEAS(q;%Trxs|?TY-ji<fqvoeNq4B{?8r--=
z7X04=7I%rxC}llt@5p}56qB5b@q7)j$&6M}<jdLFn?Tn|#HVElwrT6}ROjgm!O53k
zxeaFhu+=jC6-lyRBF`A)3l1^4d$qeVc=c4HQn`mah^Y(ChC(BQb(vBx8r>EZ6FD5c
z@gRxnEx--F*!1kIQ!|{uE%8B7y_xV#qXz$Y5P|O9E!LSu*Lb+N@Z8%ure7ix&Lxth
z_VfcsF5NC)_`cZG&A6#6IYq)r8#PJX4Mk`6S<Q)A%*^&v&XPCJRCm>ju7Ofl5^YiM
zMc52rMzMg4Ch5zHdKEhAnnY{G(kCzsI{<Vqk0iX_;eEUQ#gA@@8W#MQNf+1BANI|r
zqV&1P{PKldy~TTd_wrWk5yb!Q)es^)iukM;&0EMfcP0z8^5;JCN*@?4C~)<~h!Gxr
zexO!8f~&!*@^<He^a4mv@GGxXIu4~x+!_UV-$MdL@gB?TslFO+0X``ouU8cP9eRxw
zjI;gcp5yo?okwwPk?@$4wDvg9ICmXb8IiMW>biF+?i?YA0T@K(u)OcsI7MJEM$>Z1
zUT#3E#ttwS@GP7(!!=Ym;REZKdBphIUd*$@nnat4EmS2TPl#3-Ey}ZTr(7cczUeJi
zB3#P)zC1Aw-yO2oXr2f#PdJk0?Q<Wz+7qiNlyb^G>So7d!Bs7q^CYos5}<W|#y`fB
zdsyllfYX+&$PZ-O&mnGp;WiYqK*`6ct~+{XzDNpx2>hcbQ><hc9H!n@E}`6&<%i_h
z7yY*HP>c`IS+~yndcAbfqAtfJRbFcu#T}cs?`hTC2&B#RRSf1WvC`Yzib!ue@&r(x
zBqPn@9jLtfp$8&m`s-(sH;*)3b>T}%oD|X*85d6I1tjzALn~G*Y+Vm^I)8}CN?<mp
z1ZYBF<t;NT(`*|B?{}(Q_(LD4y%_i#(eN+N(lPQvu2d=~M^|>J77<Agl+f0gqH@rm
zjQz0&g@k&(<jBLArq-^P2!q~dolqz5I5{8>>Je*$s^T+k!g)A;{trS`E*9(~_BtyJ
z0+wN=DGp7b4;fC8jX(eV%QOE+91o*OyF115L0IO!SsFRG{a&2u)C=~FmY83Y#lP{%
z%OyF1FXW)sp~jklf`d!!Jdqh6Bfy6C<&nit8s+}&9R?sx(M+aStRCANE1pW#X&Co8
z9a5Ge-PokBFb<2uD0?SYqljP4pn)r1oJQ|<Ey{U5p+ZkGp0E|ka@Xz4*$v6Nn@7)*
zPNE1e(Aex%H8Bum8&HFhRRY23=ti&>rX{hIP%)LV<#{itSfpu+P&nq}9U24kU@s7b
zd=nk&vC~CtjVz-7T*V@;=3;W&ga*^fmX8f)+ao4-c42v2bA^F}aUOaD2@-ki1E=3y
zq!b;6Tn3zqU#iGipkF1?>p>QkE>+9Y_c<{;PijH!B#?O-tKrg5+ZI$-1P~C<>NP=4
zH6(|tvTS;sUc1SHk!8<zFv4LZd<=2t2X`-kqQ@r@kjRv(eXv|MS^Ug+{T+JB1K{2?
zk(iAuUj%ZOo)8GdwO7x(X(|1n+1}v|PNi5WODFjxc=zw>9tbvOKo7Se?UV?HheGW%
z`^RLYHgB;YtBct1_fcq-!45mcEeGad=hIT*l=9LpY?bM*NKDV4E>bL-0uVvPs^Y4x
zlz}xY`@_)K148p!_CRiUS$yuw^2fhbw}I#gaD@V{lX1<rcRNy|8WS!+ZWRMW^BoE{
z?2H2+VJ@a{4Av@Tx>PldTQ<;*JncetAPmTs^59>r>cWwu0Z3nS1@#q?<m8=$mDu_C
zsaKM1(8+%cwv-5wClr(}ds4uiq3C;#bAPe<{bOM1){;^1hj>j>849Xb4}6Uu%3xaZ
zzqhWABbF1#_zom9+B0l`b7f6AfYe7t==p)1N4>{55)YXk>J7%IJnsqcf8>fN_ZrM?
z^~&sSqTjYKUO57;QYT}q>s>eC8-GY~ph2<r9{C|85?F4>tJ#rJ1>sgAsqEjRJ$?su
zP6S)QX;oJW<Qhe8Kqt0ac6tOKW)<G{#MInUseG*6m-bRz^a==eHRN`I$7}!AHMje2
z>UNyRZU3WAOH`0sRZ*n}&L;hZX8cw`EJ}omDrDUJw6#$IY>{1#tOXw-2|9p`0cO+^
zg7TFDDgTCwX1ILD?Pw3_vD#)RLmnJ9Ww-2k@zNs{iAIz>qV&7fmvb-JoP6NGXH*w8
zM%*Wv=2sSR6$+YpOS|O*D8}rcs4>NY$bOOwRRXULy(T}GN9%x4*JFp|;aU(5_C`Ge
zoSnh(+AdAii|!8JTMbaYCDnz`r1EtXDC*nAJ|FrB+*nGt=4jah2ZlQX4t2M}$8C>v
zJgp@;8+UTQFSw(!R7F~zF;HuIM><gs^VqKw@%xI*Nz4`7)G2N+0`KT_&ZB6)l?2;u
z*$5<c=mN$2E6Z9X>_V@f1>~wAgmCbVW9L_{@zgwgvQjEb9JuiNRqIGCu7=O+<?Ce_
zzw7Dx0IWgDeb$oxEUQQ!)(Fo$qhJ&aTrii-B(rFs9p{5r4R4u8OM%9~!2r@m84UN*
z8`o%M{@aDU67&E!0|bcD{Cu{vjPH`rO!Js=18%>9{1qM*C+l4kW-0dSuRKTSeDaX)
zrAFinb&`j4+Ozdvy~NbZZ=&2`i3@{Tk0E6J`W^;u?(C<%*(5E<i7n4tTt71MJPv%0
zfOR)B^SYIJ@8eHAbt~pxxa&+{HCC7;TlkAJplzj~;@Xko!!Y=_)3r!QoCD4?M-0kY
z<pLoh#-`No=Zb+q@i1`~RJ_ImUV;c+pkGNN+iyTj13+v(v1SAxYa|P-exY74Bt43F
z7i#vq%{Kp*&SuO3Y`~gkpAyg{?D2$pcD@jQ)8!lNjD=TDhSL(`m~0p3kPJ*AtYm;U
zcL-8tyR)bIQIg5)I>>U7Z^ih&FGi3F@rQLRsx#tplv`oR3pIef$S{f~av#PYxkV9Z
z$P#`ku9IUW9_{PjdvV(WAOokuN?@c~b&c^1cz*3W^RLrolk%U}-nUrH0-Qt&@h5*m
zQvM?SYr&0gIb59={R>K<tN8<VaiTIz4qE;L3<bNeM2$Bsqx~8;;>+S?=@7eE7UEq+
z!4nCo>prU6s5(M^E>s$b87F<HU?k`10tn<Z6Sm0ii<lqZm>vSKS|RoV@PMAw%oms%
z)SN+TmI5XoY2uEp!b<_+$Vo?{@|h&!y4_=pQ!vEY>iN7cnyrQt5lo&TpaLsPSl;iX
z0Rul~;ani4fMdv4LCtGMs1f=uIO%9H#2YO!SIVkwpm{fDot}Q*0yH}fw%GU%oh=@d
zykpDijPpBX`?Zvj0d3D*n2eVGOUDA<S<wirH*$z};Iau({Pa?~4E%3}Xwg_6^I-l}
zV0H8PtQCO0_+%EVnl--{7^tUzi@Xz(|A;x6T<w5P3xXr`oQ0|?LWE9&hq;IJz$V`i
z`5na<XH%jW-9%W<z2ID6Q~~uAw1QAPiK=gDWN%}DT3BbhAGqRGGDGzbxvER3;V2j#
zte(%10|@<gM>9kDk3)Aa25f~`4a$ZY2Dm>w|6I%!STZFENF-6*GrL4)jbjuf2yTg6
z6x{PAF@Qhty#&jX5G@95TI_}wEMVqc#e!XbJ^T3b`dup6OOhzoc3@=bX7lQ;%nd*T
zKr9Xl|B#te7R$QiUkTkFcV>pBR4?~KD#I91%Kr#7{fNe>-lhWs&uTjri);PcL&%sJ
zWy`Rujo)sO!_p!7IX_02FH(<tvqn@IaZ}njV-*NKO3JtZ8DroQfGR8QM9s5wPNTmo
zN#=HsYRI4Jl8~t^8G_eYGrJA?6FH0|Q1a^2-lk1ZX&>6(e{P5%N%;iiSn{p~u?N5|
zw`4%x=clV|>HU(8HjwXb?c5-KsHj={6cI9N+Ypswj=Vh&4ES=fkhm*dyv1ckNbJOx
zgDFF0^h3$uEgPE`P$l|;kRkb?4eui`pg6M|JydISFlG7;b9m-=sXKaw(kj%wqDsX>
z0k^8$ZH^_@Vd#g;Tp*BpyZXwi`3Cqp8{A$kzNwIZZ|$5qV{tHfj_SbWn;)PpQ)}+J
ze4gdq=oz9oQOb0lB@3_Tm*RkWK50C&5T_>CA3U#)#)A$53&#RouO_Z@o33{YVP>0n
zI&wF!1aQV3IYkV6etb88hss#Hp*GK>(Ya?t9qc8;(z!bG-_~$FdZ8pv0Y@>s{XC=e
zPl^A(n4kY#j{i3018@r&avATiX(K$rcwG6Q0Dme_YBfljNmJVtmsWiN_|v9TJC?D0
zMggi0`uO4Q0q}LQSk3r2Ojsi+<G-bU4A4@8$U&+meqYcjHI?#h{oQ+J{dVCxQmF!A
zaIcx{&y@Rp3){APpGL)>t%Df}^VceKI1xh(^q*7c(3{4!!Uk-K|Cet3fB)3%kNc?7
zrE;F)=RrTP5p)5#AZ-w-lg8}t{YEm1&yte(>vz2<8ior1?6LuTZ+kv3@N+=NvMTgJ
z_VXf$fh~T2g+|j}z&z+rtb-MEisvuCoL~&19f6NUkLoLLlWag5HE_i-tR@?VwBFwx
zeWf8c&OB#CU`K@h0i2vjU|#K%iJ*PmS4UPo{}R>v(bMn2a8QVl7^t2KAuMJ7?_`L7
z7Q$zV>iFQJcjHjOGpT9#S!>4YGi39B@E4b6TyyZ0+@@yMv;sgWuG4_13(bd&`hg6Z
zU9~gT2VkA!o&{3Cb)e40LH-0spLs7#GM^lQ`_k^X9jPDuTubG3aJQqGrvqR^V?ox?
z3#i%JE#Z1oFfP#RP0T2EeX-IgwnpRgZ2-myFv=Bo1|9GIFWA&#Y>hnd`foZL7U&r1
z8T+C9MJoIkzVsgp`TzD++laV6Q1nHEenLe&JH$?r<L5I@5ZLxe66nwXR+aX_?3?Dl
zeC2>M5d~@ny`ag%^Df4G=f7ZbtZ*<A>qG0&zfzii(<T0oU$HM>{nn5#GIkBBE&|{0
zUA&HyW|GsKd2_AKKT3~(9f3Isuw2#%CZ+$KI{p9uKL7PMo8kXbv&8c1zsu(Ttnq&x
zwtvHS|F=K;A3xIw>R((7f&<{RoekY)Q52`Z^`VjRZ)MC)?6@;&F;0bZD)zC<xUFUg
z9zOa%ogVngK7tbg9;ItV{%1L{2z<Qn_VmKFCGxSBg-?Q&y1u4=9dV=W1h3A)m@p{W
zb7xnnqxl?C90QhY-&%wSYbd1R3#4j9Y%zGk(45#ucc-xCNfkj5I>J;S`u8HTDU>^)
z(W1U}z35f)8tXYGLu>rsNiu<=Qt%xorC%-jpQWO}M<Dk&_6Z~6dJT6Nx4#_$XhMdY
z{Ys^x-}6uN5;*0-gJ^?&1*a0cub7iNNB%tjPDSD6KEYD?FR8nSp@fHtM&VIF(I*`a
zxUPI)jHA(Ga{~#n^n`1sw>XeQGAgY1W*A%CAhb9zC@;%@PjMWocPr*~8k1Is{(Sug
zIQ5yi24ARv259*|%7_12{Qvw9!hdrE-eh&%-<SK(4ZMwjpi#iT9x1-Qlb03`Jk)*g
zYZQQg+z1R|ydk=M?1o*)QkUY0_CK-oM@hxz{-)gg6ectL2nYDu0o8vGIP&<kCP1B6
zM14zY+lBr5GSv4&lc?K|($$;}vE3&KDI-ap_2{5g<gQEy8Tx}&2NH{Q_;m$O-cEr%
z9Mr{|sDPhX&YgVz{r~+_{|{fqih?~&D_W`apY58?gA9L&E2gw89aD}8a%<)}A<fXS
zQ;=ZygP8`nAqE%L6zqWXPT5-0bSg`9C`Bdi<^fQx8>3R>){p9^O115JX~VySRu5{>
zbWD}c5ldok*R(J(r8O_>Av=iuvk0YQhGWK*p8!j=UIFp-WxrVW?SDSR|G2v;=s?yQ
z8k`$o0+!(KABI#ioxFz`aH*tTk%{?;fDOY5V64wql+GOBcM}|cGYUXF`H5&d*&wnX
zn<W;q&E;AXvvA{KP4&b2h7CicOpt#K)^S_vaT%xJ&u1#otPgbKpQ%KXPB;4craqfy
z{&np?3iykJwGLNWA<5d{!sSYTeO^KbfT-cw=KFB^H(C?z{bLv$Rz)uFy^5RmiPuF$
zq`a5Hz2HZ#h2`lj*ub@@5g%>kL-4Q2G(B!vKi>zuy1lJlLw>~9gw}O*ADKsbDjVkT
zE-oXbYO`5<u6@*~%{_Z@%XZ26GU)Q=qM3XJZ*bhyw)!mW7V0f5RBSArm*#N%)p=)2
zbLR4e2q@@O{Psod1NT2Gv2|4xW~S5Nb<H1k#~9Y{9(%1@&l}#n?qeJFOgdfu9g%!{
zA`DBrU8mn0dmgxByWV%h+d`S##81{&KFB@KTR|=*Z(9FiNb|@=yVI=eE83toLr}tW
zkfhruO4h-SRn0HNttjI`axQ}1SL}-s7`OfXxTR-3XH8A7NFwAC^9;1+u#-M|&aQgi
z3#pGSqo>S<Cl3O0`k1KJryV=e+)_=~GXpGLHQphv^Wg;XCZp{EPEMDo4u{<S63hhW
zJFXIQsku(OowTzOwuB|NCN~npZF#*CPhqTx1irA!$!5$v>AU0QPv`vYc`OjU^_Rl}
zD`G^3+d9S)e;lrk6HVj%6thu@+M=N+L1!Bw%wDk^N4GaS7Zs;|O~>cM-f52K26pE|
z{aC*a(oA#;16R%xKGo5CtzUS<H{{~KHe`>~=RaR-R=@=jW1g3yCJYW*BX@?%fkO+;
z-HHF7C-P#`y<T#JN~K5lVASp8HKA_2>7Dw$;?nv5lL>RG8+xd8|M_o{%>EMAHZ6kY
zHHs_j=pI&@V5k}{Q0zWO`u$ya9VUWw9%)p;b|FA0&F!4nq?H-E)nY!43I6*^8c_k8
zZd&D%<=-gcbU_3N6FiG&eGooj{05dqe%Etld%H(sKR^kJIGH4$+8&+;uz9qz$eRBA
zl|2Y5Xq0f35*&Wj;Va??i=vilIC2I;uOxtz-NUyVUpa;W{@_jrAi>hFgFGyH*y~Rv
zCtzsQXAS-Xaj`d{5bUS}1EDqm84te$NbF#gdN*^^a7g{>;72bN5PorkY~Ix_0{S&B
zoQX5rmhp`E76_3ho?iv9C$5XE_~UA^e(7eyHCb2mKC5lLQvk?wn%V`sQ1~d`(>#$o
znI-Rg&m<5z>C~?>X!H?=Ze+M$@8oUJi(JXD+yN>`UcdAG5Wpy@^MP#I0+mw!myh~3
z9HtyD>ILL&@bC`~VrgVWyJvMYE^Yo}5xfVGG7PPNLH5KW>Ie>GL;d&<<_2HaWu<{#
z*YjMJ5AfF(RwqGg?I@KT&;o`!mf&G15~)KZ1bDtrE7R_Z)RsW{CBt{pg}c$O2df|`
zbz%vnb$gJ~wmx0zeP`(kf-MsendJu%FIfRXOvUADXpk=Y1spq8_<h9>&L>_WzRo!k
z7c8r*1AL9AQV3wU%`2~)`DOkq3e!wW@QO6^L&gn!gsvVqdEc&)FJ}G*U=KEU)CFG4
zUppB9tNY878;%qM7)W~R*J08~Dpx$HG4N_d=Q&_OICj$xvv;CZv>~AO0O=enkWv;i
zKgz~ijC0y91gOo?IGvU)7@R%+CRa_<1P@LGLKHL09an|&@WKVb;L6LZt2hRDpq~b1
zfcD%2mdU!7xAq=(gX)MCL_RF0E_nf@4XVZzEbv0txTSDP)z<bS=sfuF8_>tr<Dg<h
z4^JOSK(zpUBg73hp`ujp1pNo1V{_sGhxb}@IFhW|Px;Err)cJ&pIvFycy&#@)Q!)a
zo^h-FaHH~W$!RIbMN~kM@2Zm=_8U5BCVWOTjbZaQ{|3dHi+@HSJW>R_0JlnfAZ<-4
zX$4nd@?rM9Y*Nl3pq*{3H!k0myiq$tCM&;AQoSzdk-Pr`Hf=dv<lTWi1Et>1D=Gi=
zyu=&K$VixrYN^()CI0J?|3-Zb!W<PiSfyc56S}vP<#Ar9=V(L=WxC5)*(_;=x$(g+
z*DMbxi8~0Tyf&Eih)&<Iv*z9ZCcSG1DfPlvY&Z&L&Gy%#6q#!_iN|rfq;N5m6~EGX
z&OC3<LvqsL9ZqgH`FKz^S>Be18)T|YLTQ;x;em~m3-mSp0w_*X4Xkkx_=*d|0P83>
ze4yTe3SHB{ChNP=&h(9)tboIq482J+|N1S^Sa04jk^|xl-Y114udvje#fWkSd-c~x
znF4dOZ5pr1Zs0y)G#@8;7}F>{>;6w<Aeu(`AnFL5xgWPz!`@EW@c=SCErVQ-;!EC@
z4afL(WJ0>0;nR+X`wYN2tzg5Jh%y-(`u$X=`!a(Ke#^O0NfAgoooPNyWebFsMXG45
z-vSHdQ4b+jJ)FKrz%7B_qoA&B8xX_H@~D+4QtS^ONFkp@V@QY~`^n!{GTi=T^BRpJ
zOuW4R#nVFBgWEv3lg}j1pcLnh{6?PlRe0wsEq748B<)2Ld&5PLCI14owTDYOKawZu
z*e4w!19cT4^guq>pxwv4$1dXr+(Fb>D$+ctdvM&xd-;o_PY@Q|d;fC~&In|pnXo&6
z<CDw2;@3+5R+)Yb!xx`MeOEfwPrnjifqQhE7iodRtq>jnl~~~jyGoKkC`2J=k36J#
zdZbB>DN)fVI7QzXaz5+FTj!urME;}?Dr*DcW3Ng^;(p>puL7yw`^s+*HEjTa>#1MK
zqIk^=I0}}nl{QWbwo+2Rp~I;F!F7Qi`gV|{;c+IVi|E|Xjt-!%+`umxL6=9asFT!H
z4vgU6&7U*2BxH4cbtjePL+sk#)d%|byPwuB<!?1-C{y6<`A)oObNQgAZ6<c=QVQGX
zpvW4Eqo|=_&qI5J_0klwe^>ynaMWMD705XqM=Q5Ghs@T0zWU?OAe|KW?vX)=Bp7YX
zCxx%$Fxv3(E&`~t0*9a#vCY1ZHa=H8WW*gCPL^fOI}pVbz0J_}0C_|L9AHyzFu~jx
z<g7Z5EpDb3=Tnug1jk|KT{I7E$!bcGzm2D;$uzQQy>kkBVnRECw=LHva{6ZT$9p*S
zrjR&RZq5a{53682yC9LjQkEDIUv*asqT@$mq9fn@KKG81*}3v)<YWSGj?)3JA>mH~
zjOaYhDf09+-<i#7S8xR=feSQWg0PpD6Uku2EI>QFnzWAzH7FAK#|_qFbNV~O@0g*-
zg@3pT1AU61A|p~0IkfQ_9p`!@^Ajp_oxn9A(E`n3<I|J@G05MwKT&Rku&9xP1RbZO
z5ll^Bk6EBY^5UYa6Gd5FzMiVB5FzVm8Q2-QFdFF8K$N=ab#coV;7;J+z6c*QfHNei
zMcrDdri~(mroN)iLS_^?$nTs^!@NiUJ|ZN$K5Z%I=<%!_sG8VhnThmrGgPqiJU&5#
zTy^H$^zKxNCYy>|(Q^xBe#aC%7Omy`?n2bbNzf^%Qdr+tm3#-CA12`V)ov?`@ck&z
z4`c$2&I1V?e|B@Zns4-~6pqx`90xwK=Y`4by+0m26%QI&`WL1(Jr6lQn)B~D81OLS
zQOIY-2ON7b+pC_LF&)9Im(w{CGETwKP@L*W^{pfC<7S+F9NIz}i?4(3(<J~DlD`ox
zf%GC+v1R)9?kmPRE-iIEZ>&aaegMIJk4{{6RI<luZ?K{oeM8&#V}S!?ae93nLjL*`
zhTe=rycAJchgDi-9D_`WE@SX0m$(l_{B^MX-b4vbM2GborASyZcNLgq7E8~$$k5+M
zgS&-wH25&E*3ni_Ia(((o}4(<6_fV1&VhMl*!`vUB$Y(w<3vR7qYQs#f?uGZyaQVQ
zj+9Kwp&G%8TC@$3%Wu5m663a4V++0ce|%$5&SCYSb*pd4@-LI<)jv#%c1$LBlnqv_
z{uoP=#B8Yw3DziPr<L6-jw#*6q^+ivb+8g3gkoLG_NE4F=@#8!cr=wLhkkS%zRbW4
z493Y!W*zOFmO%OFc*O9KO_^CmF#-$8EeE)4r(~Q7hQYI)l0Vti0BAM%h&orVF{&jr
z0CFbwVG5Q~>~;c!LGB^nGIPdkjwA{@I)7Zk%`3Ah6rEEf-VttwC>G^RH5~@or?L}n
z#^_SJdhvw?Hz<__Lt&KSiQKg7^=_Fog%Si-IMq1c(lMJ=#F9E+qo^+xI_2zBD$Oza
z2xfTZ$TKOk-_`wTEEp<!@(j2cIt!*SGfOfp5k{Gp)kYgt=hON+?%5t(J3^`BJdzoH
z6jRQCPUTdb3hnJn!A`$Z+4?mlg5hX5ifl#bIzcvEE;ccZ0nBQ4E%Gv*Y+*WRj=s{?
zd)FK0$T=oFN}f@gp8Y;y%7K4ZxQzVo*Q7UqKzI9>gzR>~*Pzc&?s^j?dE{stcC2S8
zbBMjP1v2qdgLLM`QCRt><h_SoVoqYUnv_I#cY(8MyPyo4I_pT{zI0z(UP2j>n_*Ep
zj$yWTsx`ZHpvht>En2Z*4wHp{Xm2_!J))F2?pn2}q)$~<kK>}(B_9rHabxkGy;L}g
zmpoiMa^T7w&x*}>Go4BQVccQOFy%gsYiQfco5XkVubR*Z@#7g%!~YubgZS*B(mQ<|
z>4>&L`JZ&(tyZNyH8CfnQ{Nj;Qg2s~35ejkCpnl=ax45&DKxfe%ZhKeYu$QiOoyz9
zML4GDY}?2_v8s8ke8AI%Sp8;ush{c=2uRV0EIMAdF!BehRm)4@x8jpX(7~EiLNS81
zY@i~<C3m8ielt&!@HzET{akXT%;+tu*<_}vk-aU>IenU*@e8d^Mm2Zvqpp1I^>f0f
z2{DveQ2j!mvA^OuD)T~`_xBYjW>BH!I$_ompgAPi4h}Di>b?%j&p|wz{4}|yxhUX2
z%Gi;XU<)d6)?Y-RqWEKOpnn~-6#n9=4VK>6<DFwkX^u!MC~yAA{9^Zrx~ik8-<u~w
z44qZX69G+66r?TWbEDE+frD-*zktiA2de`J<=0!a_N}!PJC4RKbd_A13-O6a;z6u$
zdObOJq>BridG@B-D>9L(As6wr|92+0T8kUpR~`iZTEFex&Tp3Qy&jFub!WuC#}Trb
zzb|TZ>ma8wgPNs^m%jQPBu&~tcsEdB@Fs{NJ0=h!R-E9xP-k1s-Gl5+hWD=m)5w<x
zZgG<DxXA+7DDUa4k?y>Nq9@(azZYV|gok!BjRqrH_e&cyqQf)uoeZ^=-vwrzdI{JY
zW#du3G}^K`7Y`j${bT7xxN@|XOvH3;k)OS`mTZ1B1hsMJed*qh3uD*EYG{HH5nt_$
zfBUOVo<0q)r%m(epZt4>1|HV2#1lx-I-gJn&G^&jH2nDLuHm#vrruZ%iM(Uf{(G72
z1<Qb@dGzJ;V9OTKdN^=OmV)LDJaP{hwLlVyuZW~nxD1U7FW`JZ!;~pCDpSQ|LX|la
zz7RA?j7$<aTlPCgYLP$m!=<uQ*=fSoT++lo7|$itFno~z$3IC>Q-(P7Eu7N6$czPp
zx^(c1cx`k1LYH7&P4_UvhPkLcky|Yi6I{80s5{`G<6x*AoRaKF(C<VQqnB=<)DJb_
z@V>D3#|%%Qx&=5WUB%Zzn;A)jDENHm`A~@n{2%U05#>&&JId8?ltdzLM}rJMi|V;5
zT1pb_N9?~QAWtU2vCcbYRo3ff_$zE``~}+s8l3(VE?_yCgjipAOq`}xbnrydGG=M?
zdw?E0j*#y}^|(GCV7o7T;^%m!osAx<EF!IqrxSy!7>uL}hBf}l^RG)vdlL^O)SFi7
zyxs=M>dKsBACaek^ih+zPLEycF+BM&6Dm2&8Z@?-nc4icZKj4U9%1h6hCF7(g@msm
zmYwQonNgi22!M|8Sl3JPGYqBSZA9R}vnbdAn5nepV6llxl)A0|3uVX4{#T*rL0oNm
zGVKfVKlbcZDP9B8#Dgy1nG^jFEYx_HF!zxoKHiZVU@66)owyQx`-#`LdxNQI5a=b@
z{mtWhcu?IOBd&p#hnuGcCH{dwUarp#=Z<ucGL**^dsFUGXt+lm=8~s*#h^v^rF-oc
z3Cejg`X}A^;}8<pp`i8~VUZ5!&aI}H1U6kS@{gCP-kW=}{Hb8r0XxgVLM<21YhEjU
zt0_#{LE-yOyRUkcc9HZ(QMS5-f!(Q`gnh5bjv*-oE446TQ~nSvSVsktX-o2D)h8BT
z0E6*A*_y;;KefB6ST2#5JXhr>Tcs{>(oT#g)Cjc2SCG22PIA}F6<WIeGbGYK{t_nS
z4XmM;Lq?ivN4aNF%NcSaALF|P&v>(!&&>NRy*pf_afk9)Ch#a*GLktC+aq0fz~Yp%
z*4f)Z*eeZA3ia+R78_wwnh-IPwhCD<{|XnOjee@roX$T-^OImD5HWoiu`~sI|6BC>
z0DB}wDic3aqL(^zJ$m37i6?Z_ZpI)e*lB$)6`l@**Hmb9?U^J!?`~?2_x1yEhfy_$
zb~T4Q&x3NZYG!^y93i*z%>-wAkbs0n>q!K&XLhGf>Z(wsZ|iG=dcwQ2{sv{Sbay2b
z@66DlYIVcxtOB%`2K=A0)-gx3wn=O(baGA5K!>rvZQjG@Fo%qdMRpTgs6c6=gY>(+
zh2GvMf9BMn1X8?Mi)p?4(cS&OMgh;(gw}mF#(EY>LMt0fLnOgTrK~R_H$?SCT7!mF
zQ`8RU@k5l=XoFV?G%~)?ns#c@REFF4&CB<N<#~DTLjw-OKpP<FsI%pu<ml))sgqGy
znC7+DUIxs}NbkpZ5=~N?C^GVGy4z}=*_3V2__9#ras&S8_TpGP|3|s5Kwx=cPh_YW
zw%^ZrmoAjZUXAYrN)4Yd0MMlPs$Bd#t<rHYhOj9%YvS+Wet`qMcv@RH#Hl@g8^hFq
z@wJYd<rFG#&hBFk*qO=MCM8mb5%aLlaMlE@Sfi*MmC9>?xT+-)%@mJK8`vZrVYB3L
z@4S6A3dL5EknW!gV0a~!m&OAOj-BvK&L|7s-TW-yVEw)8_NE>yq5}OGsTENAL8;ky
zW09VP6@&;^#wTJpGf2&4Yht3tT1UBq)$&96?27_>K4+kM#V`)Chd>9|V2262k0pDQ
zE_;8*j<~{wHPwrmY~wNlw)S|1i!+5U`&%<(z3u+eD`2oSCfy|}NT{4`v`wxD7LW%k
zQ`|N14*VQqcPv9rM}xid?riWb(iF$uJ79Q5YKCG=U>>U0Bqv4?|4`wsKWMUM{JtPO
z-4ccgk8>g@d4LQ(Q?9O6a~vIxj=d;1eK>PhKxh{K)sMqp(-9q%R+ThN;#9wFw#Pp@
zyQxEgC990L@iQP$Q6N}j>08a0tj$t|KiO_|{bMnm^mD~YX@^Vj>?x?iD(icFuKyy~
zt+4@$Q5b1}TEc?qx^lc$A8mUE_v4(LCQB|MWj2B_Z=FE2Kr(<(BNEEp?-%TDF|0a3
z@J3Bg?zG4xe#@^o5^Dg*lNa<EJOoMi#B`Gf@MePuR+`R+&de}VMa&Wn0`<wQ0)Yfd
z_!A~Mx0QyD=8V(>pK@ZOHzWhr!KU7({kokH`Dr1Mq=0>Wat;%}$vc}1&6pS>DJ~ls
zfz5&e3#Yk4gETXY&H4{!1LSZq)C#CKVTPSsYpRmp9}E7fq_h7(hxp-3R!>%p$FrZ4
zAtJ*`19~%|o|uZwyg*Ia<dLsla~&wcHBw`h6Zr{r29PQnmarV@9et%;rqW@vyA(^T
z2E*DiQ_3O-Mh)8W92i3x7@RG#0=U?2o?pFUdjZK8*L$jl2V{}(nwUka4C=X0)9C#h
zwjXQ*Qe%hvS(G28l1B#lmF3=q3P{=;RSEU+)L5rJIml{11{E&E8}>$4r&N|C`+#nE
zo_v?%S%qswq<=)8c=63BoX^5`GDZ71D`l&R@QL~6bu7p5ifMrGOi-Jy)V>kp(Z%+7
z=rz)z5mX4OYDy_MgOjV9nw<g*=kD_`61ZR`WanwGl-4UL@Twj6U~lc?@^lzr;gm{s
zoOoC%tlN0RHEc@$D&R6K4xQY)c3!WY+-B$G$=w>5IAAyL)Pyl#ppm;4c5|>vojSc!
zBd>>BI%}Tt`*z06Y3^ta$wv4iZhw^7sY9#6_w|<OmgTMLgo=wa-5qChfw004JfnDD
zhXGX?<H)`~C|idJUo{SChf$?;>}s?H`Agda3VqC5JFpx7_3KfrXqEiLfY!Sxycb31
zvu|jVykS?k@ibN&!(>!@(gx+jWS_P86D}3nHsS-vqGr}Z-jmDSY#hvshp>JBR(-54
zDcqM!$=0lTRedvQmBg>C`ptT5B4IDbW=1aApTFmnN+ZPBbqj#&dEv-&x4Diq1#Q<G
ziss@vIPy*t)Z2D_)eKfjpT^S^gysBbkk9d|x|xKi)q-(eJMYlGHpa+v%9I3%xVJ}J
z0hRq%dSj2W$I`7+#caam6qlfFE?_@7<$Zci-#=5cD!;$$+UupMRnuCKIc!CHH`pN&
zT1+QqjXzKvteonK4=d=CN)oOtc#%yhf$FWXVL7S1d_en^jas46nDIkVz}VJYf1m$`
z!GTHfG*IgbdVW2zUMUUv^UA9a`qG@t<U-)Uxv=g&R>A#`$1nXqhA#}{eSZq#p3wf`
z9PIwAu~!{@xKP@e70|jL%{zN}rI1oh5Fe}noyJh-7=jQ7DIB`bXLv$4<R2Uvn5>kt
zz2#>`pc=t==-&QKU(;W>E~OP9d7GSwa5btNS*JPLQ{J2ZkbFBD-c0kSh0_v6fK;eI
zn+vgv$wAKsbgoXz{I`a?usl=7D!W4=J?pfIW)={Ys4dXo>`5!Gr#C`Rw#^8J;-nVK
z=cgrJQnNk1=yVJ7%#KVH{l#+dtw)6K_vc5^VALb$Ut`zC(AD-TH-37;+3HYto~^lV
zP2t0V#z>UyTVC^OOf4K-osB1jXO{4r0Nw*=#Q-y09olo*y196UU9+u0aa?Y{Q^>6Q
ztT}iPCzV6t&AQQnIHu4G=K&RmuZpM2&(xV=v7y1A1@GzRpw|XeSmcFKNShzjbQ~v*
zfg24f_<ynY)=^QeZTqk&5~5O4(kLM{fHVjwAR;9y9YZ)E(w)Nyf|P>NAt513NDtj5
zAxKDfcQ*rk*X{Os-lyLE?%%t<KfZ6R-&(Ww!eKTubKlo>Ugvq7$9Wt^O1I6wl=g_;
zJpzqIu4=)~tngxFp|%D~p2dV-VJmJYmrp@RB*~msey~U_C1L+J9mbdTxjeM>sua&D
zW*i&*iS>n^OS0^2<<ANmcP*4@u}%qSkCJO0+|Dx@Er~f;sc#s0{-r+r4hNor{D*Ot
zN45%M2R+Ukx+B5U5?Cm8q31vI*m0(j$9N-AwKuQq8NQ9eIw9Lc%>OhR5f7>HE(|r9
zi@K$-buYm`Oma<cL`cVRRr4S-d7LYZuysVD&`!D6^2j912{8&_ow>}FnoFNpaxHT;
zBk%Z)ukLBE@Z{_YrLyjcWcPyMgSpv{P-yJQK*dzOHV^+o`T>p}M~pN)<ZoAd437>{
z6gz&@KfiIyc<CDX#+L5ju?lRrd=OJ~TVi^Ik28ZkB!8E^O{!W}3fnF2W#rE4>Xzz~
zj1)HXNV?_4ckkkHY&ZQ3=xO%y61U#8p|X983|OA)bR{yITM0s{SgZfBbw6sAwb7Y7
zlVmH#v&E@*q|>Y_*d{(GgiR5oJhSr)@MhBB{}OW*e$(Wwp-YWPeqGkKOipC6A`vit
z4-e3~wXjXIRE@kfoIB#6RSQ?+Y>`T@3p0tacIL}{gXcGE>1t*GpQd05;JdA$tUrf1
zO06Rpz}hlUryp`_2@8fP+fY=FJn%ytDzelU!1`PFG}Lxt79E`OJGaJWvU8Y`#Q0sW
zx3N)bvnh&+T58w3FN+alMnOBOqa|zhrb+e<T0P4|gofc@n|C$)5H1yWPk--AwcH-F
zYMsWDbnhAWYpK4~;&PG-qsCNYfbe;X*SzXVZSq8C7J<O3Pyj(K4(xjO*Fam7LE&hd
z^wr!37kEe3n42d4=@C=fG>DL^M3g|{S<+mFS2}YT5=M8TY8c8+h`MUpm-NE+Y0)c#
zKy!Y3$YOLqo%3usYI^LhB<&fX7ksNyUL2fWt$GwZty%DOo}M`dQB_u<2W&05EsQL}
z=S$9F$*2jXX?Zy$9BVcARzb7FaWX<{c5hZ_`TPK36PEu}cW~Z?<h&gk!sKhMqd+C&
z75Q56aaIh4YPoGqfLnw3ORcwALETxqH%;zzCmL#hiQ1fT>LKdPG@?%AIAFl?x<-G@
z*GWP<&EF7ZRtqO{^gd{7Z*17v9EoE+2x=ZANAg5pt*_F;Y~`_RU;H4A458OlT_+BX
z@%^@9cB-rD82`NF0X?0q)e3O=`pOkyroLA3H8!G7qT}?;@B3rtx&p_sf@Cef{ZbK`
zI>3^iXP<nN+nwbzANY>S{E12sHfqON+<t?10dV!MhqgBR<h_k$u6R8yOSg)5>Yrj!
zts_Wpo8T`>M2qA?S~*p<t4G;<%-7m~b2rYaZBJvhFCd>Yi0_FYz3e7>YFr-v1g+fQ
z5V7iP3=bp2G#(7@2@ulPJK3ED%sWl%(@d{946hw<BIA{{!nr~+mefRHLljuV3wq~C
zFuPWRY>3|{-Vau<1DlZmUrClNBHc|#80iJ1NS!HdShc@kAZK}M{cWuG6u6|PiX~*o
zfQK|6sIxykyNFVoNwK)zeVFA|-Gd*dyd*<rX37*maTR^j+|VJMO8r7C{;9<vNn%YQ
z^N!v0B=h-xwxj%aVN=g=VWOh8Xoo}s?L(%jZ>mC;w}`EAHMtbZ^Sy!QQ7DTKt74Y=
z;!wN{$Bnwu0iTv$4$g?#0F@2n$3j&LZOi5v-^#gmXAq-N2+^;y)vXHoh0|%b4r|=b
zIi`Q(RaF>&!`z+)bUW0#Oig-v*Q~dK1@L7)C_XjS+K9`ki_$d0hxLeqU3o}3zTd@J
zE=)xUmqLWhpP*-RNpwuO_Ae@5?hFzilH>5MRz7Zb$=*T#ks6yvEt+?m!OG#E#<|?j
zZpH1noPw@)B*fX+G+iTC2<^?Tpa@poXb!T+&UZ!L_B38equsaHsmoituJ-13Thm>g
zXCDF^Q!93wvF$q(5RqCRe6dmQ1JU3fs1X||w%nE%Lk*f@oI02}&!=0J(4Q>GS$+Jh
zHKJKEH?{>hMS=u#&6J5ZNIvE?-P6(j(L3LAhdgPNqScnB4c8gcfV8s89q@a=(m4uR
zavd?I+a^~t_uG$oSi>AarD^+$%sF9@5;v`lP?Z>Q1j<czYT31Qr$m8mCNrU{P4C1r
zdC@KQGk)ST`1M<L)BRT01E&@nBqcN$^7CgcsIlW+ZPis-{e?Oj9kw<RiW)z?C6{u!
z9ODYNvwGf&5^wh}2?k17v<&&|HPKrHHjQUDv~4>bQF(xlt(~DO$9xmds`$L^RrNZy
z3v-)Wa}Ia~m0HdK=S?vFNy!;G6x!wLH%h2FnH<Eubnn@kadwIFMO6G=Ibz#uSHD%_
z&U}!s%aJa&=c1kLk2s<9rwwTqsGnBxZ0>~1>JoudmW}ka@N%TmpeS*V8hZRKQYpf_
z+G$Oam%j4HdQ)M8W%+c03*E5tTH^-|w0K5_MyjkbI1~2qA8hm=BT}ygzQpqf4=HXe
zx+&t8&7{t#*hRl1m-C*S*0l<~71QDlJRe;fh~-LznPHbnc@^olBanMR*81mjpI(Ov
z!$AZak*A)TLDT~;*TfkuJOd+dd7(&SsfqVW7y4NfD>wl<ly<4WuyrmakY8pFQ9Q$S
z;>Y99H|J|VB-!?Ic}fGV9Ihk@E4{_4kMj)vCQM@KJn2F>t9NyD`if%@c}=(LTiZu-
zc)Ed0f}44q2CCY^%A<ANQtIK2yfkpi+ak4mH+&S*Uro;tE}PpAbPE@Htx%v<H=^zk
za$NJco3Lk^Xm|e&=}D%MytdYTGhh_dVmYr%=tV^n$fA!Qc4Ex;4mMqIsta4jt55w7
zy7g*paD7VsURr=CTro6wdoGySEBBI|>pPRvfoed#MII6@U7#s&gt=!cqi%aO@d?ko
zI%QuM<w;%Z(#X5{bm+Z4i{9rmMJp56`T#-TI>?sep8Ktg{U|dZX(Ff;kEuG@8?FTN
zyqYu&;~R#@0H`?JdT3i3D_%?A&lyMP6194}?Gs~u&?=j9s4<f9sL}Fr+~^u;3yxJ4
zoLrk%ZPvURlJ?n*nlr@@mB2A<N@A*LK7zI<%uGhwjPfCs=%SQQT6fA3Ym)Jwgx19)
zmvjugF50bo`J{USa)zsC&iBAE5xR@0ozaH+wpSKsNI*C^Cb%zC^M<(xI{V0M4Nm+w
z8-P^(#K}|9x4j};g%H&=I@YOa;2cVXm<WsbbKtoX%W~ABr~iplC##2$E(jUVD}}iz
z%_GSp$)4dm&C%fGoA3MDT~pvMxF=>e^MKxd-gZ(exRla|_f2iw31hP18tZcIkM`TT
zx^={v-Qv!6k~1mc3zVdGFvRR@bzQAV{Ql?A=HwL6nRzZJZcWa`80V~CZuTuG<%wj%
z6E?q9C7DLeG%^j@_z6*jFKdHp(L_i-#BK90L-fBD?b>QQgBe^y@`UVUTO-@fG+QI`
zzaLMqz&w|@JGI*nc?3NH`<--d1fq)Ur;NlQ-^EdwdRV=+l!Pe`g;aUXV;#|C_41<b
zyM2z^xieyq7lsz5>TC`{%c1(R5~nmVl#9Sj%0FFb)QSCIIdwdniQdLYIgs#fO8wIH
zU93$pL|{t$mu*c3>z@&;IYZf`h;>8RL-x3kWEY)Hxs85h_!HT9|3&uU^`@#%4{PCK
z`-#yZ`U}p++EeJ!O)c>dOV)j(4vr^uYJD8tOst?VIL5wtuzbSz_!3C+e&uyu=LpyQ
zC8mO&Rwd0b-#YDT<l&SLe2(A7uILHuyfx8Y0i<-FqZvX-xh=Guwm^K@lBcTMcr3Ks
z6imCUP|*&GW{H-LXN07J7+$Sc%$zKgM5cz{Z*?--8HGl-s8wNt%^1E0eb4U)w#M(W
zz-Y=TLKKJ~-+G7keELV72VXsU4{r^?G*&};YOGc+0mLT*k5A3=R5{o49g|1-=Y8Qo
z6Xm2$hx0ZA(C&StKVt~wzjo!>mM*}~e<PTp>rQl1G~!j|10!BXGS9M9Af>M6yT~^v
zS4Xj?Mg{yxs=7;vthinn=cX<oShN7&<_L&v$m>!cNpYa-ceD>IUV4{Um^7;{o7%m@
zZ}X#b`|QL$TZhelo|-e9cQMq{ut|n%m3!2rMUvD(Q{#aSPKGx#zOg)0L#02tOAE;`
zrA`&mTv?$?me7nyhkU=}9<j0ECfjo2NoeeNII%d91CfNem$#d@WR?~QdyeuuoJeQJ
z_HG5(YGl#_p*HkHv)wz*gO|o-mNT>IZ|Gr}3pqsFyu^_v0R~H8R=|Gqk|7c187=4M
z+Z)EoVrb$iL-}@~G1PHvQDp&ZmHP#A-Hyvq_B3`SP~ZgJmiX~k2SbUoab^V-HbAxy
zwLbNYsTLmNVPlmuBIi0tJgmC1y||nVAK9{F=`_6Mdn`_FL#*;R>cCh_Xzk-pil);?
z%MGI>9i&N&yJova%$}+moV4e$lMZ@(b>{Sm;P=68Vxeuj1b6c3(t&YX6%!IFEJ4d?
z#OicgngflkqIV4iM?r<plxM4gY*TtgR_fWDzV^-T&qg1#1g{R!?ZcY5H9L~N)}Eu=
zxdEo=stCICl4ZoGjmZg0P04mA&F({)7LivK9z*wp$cIh17z(%{YjQ0CThPL6`_il1
z=LQd7Y5OocO_99X`6zpJIe#ZzcK>zqHclC;92QP>qwbdF;;pu%!7X?3C!35U^clOm
z%^rlEVeOl#w9}5=kPvd2&^=GjZO#UfO%clFCmuKZGcxbxc3IK12h^s<+<;$D+K5UJ
zGmjJsRs$z@de(-Ia0WSs_tc`_F-5sB?zMOquCZ#uS>T*?cmG7`{%$|qxy`K|DI4^;
z7jYVWJ0;seXG1R5rLTTUSMfsUIS-;1IduD59^7~#&rcyKa4kb$vZIyE=-Z!iApUMQ
z<)wWwGK^#Qx_U&2K%eFfEz6>x;ws^=?O^Z)MsVbGEnc~9AJ8$(gM->cFc3Q5bKCXB
zPRcDLx^0WfJH?&EkV<Nf-VRNC>%I9e^x<LOQ3brRq88Umi|j#0yMM2mx`(8AHMn!i
z)j|xR`bj(^OJlL(IUF41&d@<#jz|d)_<fKwV)@(yemSKo9S!o=^-GMtu!-td@VQcl
z6nxh)iJ*w|V5X|?Fzjp9WH$m7lP!BN7%sa8=HTIRS#SIOb+Dy5O=|>*@;AJr1MXq#
z9c+4sQ&cSOG}=!`!nR@zn8Zoz@!W|TY><(RQVjfnh;wXPX93tnn@`8p--qOzeO7pc
z3)(&`*4da^UuNFq%zWhrVX;xBp=aK{_^eCkcD-PaS|}U?U{=%lsRxV;4d>PR^nm`l
zzO!)Dk?aSb^x#dBOST%%A~cEvLYTeeFPPnZ+JManTLmWEd^tLbO$?&!b`>*qt=Csy
z2|uULv$C`FH1=XzCGu*-Z;zJH_mC+Dj5+tRfZJ$%mc=O%p1`_<49%@%1dzPkTIS<@
zqj{xGDVOzap(@K2VgdxMcj17bv1hDhfh8&NV?pt{&O+mRoV1@lD?g&!k=67~5hv!g
z1VEEM{kLg(*7zCHTGz0K#pQ^>^sIx>gfi(^`3vL^iPco)*yLEc7gYjjEQR)Bc1Cxi
z6bnh|=Wpn@S1GnNsr>?=+CO~>8y}G*I_#>pho@9SlFctps8>@TY&7qRewT|tnxvcg
zS3vXhWk-ot1dGTew}cesEs}^b>EruP6(uQ_Z@_F?tVO?C?ZhC~3>Q5dha@sq%)1vC
z9jX<p+XTc)&m3a=lD9Qa-&O^lzCDr_OFPpui9u8lHiaBTl2@ir2YO3lZ;#A}@C0>N
zRLmySRnnT@eS=HnDrEmFPxoS=o)mTclru$l{n)aJ8HlPo*&;1WeI})Ge;xa53I?E~
zdV0BWTR{w^dbwto5R5URJKpxU6#Re-N_xC`Y4eIOHE;yTB(Bx@uE^b!ox}}-wshO2
z(~=NgPZ^t@R{Boy!Ki&eWvdJ^wWpbO<2+{K`}hGI+g4{KA?Aog(zpm6GVixe(+jV^
zC5`g)8+WheDa+uf_2PGHsk_{xA;H{b9B(D<M0v0Uyvu<#3x5z{84}$##B(}0h;S9b
zeg*kK2W=clZ^&uOi$#(X`|%PMf(g8%0-VJgS3I&5q6_P_=}+)GnX-4(d<;8s$@&vP
z^|e(i=iMH*arfN*;{hI1hz-q>nUx>7Z7e~Ck{Egc9pN_*6g(1gX+Z(y=*`DHydFs$
zjab%X-A|;H6IyHJT~mhkXLtW%xNOYU=Y0LN?PMf~f@91)Sn>tb)gOLk*<g#-oMWn4
zJ%+{%1+nxMij?wk+<B>2f~sN@D@QKxg;wtk>kr|?EqpmqUatVyel5$+D<vFk_b)^%
zvsOIIMzx&z*l`0_1Q!#pj^$dz(68_ly|qa2&^yYxF!ooK*n8WnY0slUo=4r$;s%Y1
z)mGiRWHQJ$5EWm`S&i+UN27R-M6rYjbYU#C3?;u;UAiJT81|Ma{JR{P+4BcagXmbU
z;{#VeYWin8uCJ9|ySg^N9Q{lRz8*R}QEAH$*dVe;lQHBdgiptsf2Fb@xizrGSj+NB
z{Dr$J-HPwiHH4xJQTxNkJdCmZ@=H91UlI>u3_3c^)r*xL5Uos#56GnV(~}msjpnuR
z<iPhHOY9GssdSr@2YTbHdTV{Z#IkUy*Sx@nVymN=g1Xknj_#smH};)cmKjX3*7dC;
zr8ogz^Hsh=`nX&I7AFFhvg|{R#!ykCwqO^D_N{m6eu52&0hF(}avqnPIQiM}bKKWj
z9~oz1xy%zBEO@cFFa^$g_wn~O{cEmgq|2TKqQokEs)_r~I|5xXlswTcH-FHlolI}C
z7x!|I7P!!+j^|P6)at);iBGQ3S-dSW&iU0vj8Pn60IT!X60^MZMQuj|aoiqBGGxR+
zDtz$>z3`ck#UI>7(P)9_h4^wyA^OpU+rvuI?RdyAt=s@fg5r^|xKRabVDn^<e8W1E
z#edhG)pA&W<p4Pg+-@>P;bmQ>k)?i!+e*|{#Yl(DoR&}7@#Bc2QI~k9b;->NcJ#>l
z!f*+EP~k>(h~0Wxb#kK6B%CMHsn+LHX;ezbpn|w$h`W`;&Ui`kwo%Cr`WKkRrqc8U
z(C(5#qQ3Ke-9OA6(sTqf(^pwuTo&YAZ(*?>nJ!1d9^UST8Uho{HFqLyABzQ>UQ76x
z64c4E=?vF7f;Abci5)YvCC)2akO%_Bu-6NV=Bo(n63|n>FukfyGruA9<5>4Ni8VSc
z)1YtRf^lip2d%cPd5bkA$b`~r17M(n(Gz2dw+pb^lU;HN+r0y@ALa#3L(?aG(P>tR
z=?t*;JA;0TwGA(8B_c1eCb$^!Fv<v0l~&_hbI2ZulwxUbSO{G`e$X^ur<e<;6zgQ)
zC=4;ZRG&%+PV}8p#6HkUT*(Cn$vC8D=n=qxB7H;QYQ?MY<kVwWUVlZJAg8$f33;h{
zTfIa#0lM%RrNE1)s!vN#!SXJpcrn#d5+<TGCvEBYiiMGyunQJ@@V$9TL(frd6!S#1
zP^%v34sN?<6ru`{q<PpncbRyi>bLK~zYZRZV|z+ayddwmaNN!@>^D)w=BjVvIfy-N
zc(5@kaK`RL*?Ak+9?;K=uj$2j7H4?yI6p`vWIaM1-JDx_ZrvWoGcKnmWSBm#qK%wm
zABGp{u(sw}sB0p&yExaXf{TXxpQ^Dw&=~YAj;Q)H7|rzkQ`oVJcs0e*Xmn(WrLzZl
zr}6NuAih~uMCHSI*<%2z+)=E-LH<`y=Qq^FlZ3xPCam6(dzDQ{7%%!|NA*MN-DOPe
zd-K_QP>;RPx4JfrV^-#m+Tt>nIup;Ni#OR|78&al5;QkQkA&OOil-FY<D;cUv%ecL
ziorV6rxtyAR<+;VGM^+=*yqHGF;(SgyYN+3y*H2F<UBz!rAbb0Xvv2pB+Wc|`4nF@
zz(g}R%*_Iq3#sLVvMv?Hy;eUgB>763c~9+Uz;m-D&>yyB{!(g~n`J?-Hba*pVWdYA
zG?&epr+8w~yrf?7Txdz=P5Dnt4)uh<DZb5In6=Kc(fIA>{kSwXFN=RByKG^Gi{`*k
zUpg~q|5&@B<2wl-GP3Sqexs9+qnBhmRl2J12xoni_CY2)Jv2SGKMFaFdhA|R->4b%
zGtwVp*|X<Z&6?^820$8gC&uanx72@wj#ls|YViz~h+{p{XxHNc!rOKQtu8@TKAy#}
z{+1Kgra^)eD@BD7w3tNDeBSE6o&z=g*8hG_4DfkkY9F(3h72#Y)ndgBGQP9XpWt~f
z2`0bb$&qQ<v!Kw!p~J#I#nF$s)d=0mAN1Jmfnh;p7KoRkUUaZ9t;ipoyUZqfm1(IT
z0PCKP?)^6~Sdoy6eJtg;TJgxF!lG~OWVS-9H+!0LTZOXWQN;&S$z_Sd8B%zHC#0fb
zf$+yZ)uL#gHF+y~OM7aN^Vn6iJsDNUw{LsPq(FRj7V<7R&acKR+@k6KcFH&)l@zht
zDMnvyA?#U9iaSY6^$vG$o92`g?g1=5DPFfE2ub*`riN9sRq-w-T=B?Xd{YhfBE<5K
zs?>69CkuSs-)?8NbvU2ctRQ5eMR%Q7U3)^dO4#`Wmt)6PVc~J|mOsq^Q=JGHO`#UH
z$@*rSMeUfp;O2UoTALN`CIQm_v9g_c-_D~-Afe<tD>afU`>c5@7jJ1}8rIopY#TIi
z`IK{;_GC(M-9v}*MjqLfiV^%M$8Zh>uN(IS&3y*zxnn2C4Ek2YM`dk)oas7`ItFB4
z?IDs&kQ4Kk{Hlc>m&H7hl`xP;$x=oCEYj|4Kkp8ip_kbJDoKzoXC^>KCBj^K=JUsu
z;~6}(B4u5d*FA;4e+;YolmRp>#$|aP9*cpBcIgOfOAdZl)b9$ly6~q5{Q=5rlAD;#
z(e);`2yTr?$@J{OOX5Xt^z~$x(}(f-Y&vvmKQogtEl^%xj0f_1o!O7Q5IXk{4}QiP
z8PYznsYI@i(YnappDkcq=e?r9&!_ls{I)AJ=RpTD|5+P$C6`>-3MfCh_X?QZag<HM
z9|PcqwT3`!=(f6EG-WYj)O_`W@$UsG9d{0PNL4(%wIQ)4y;F~TMUjD)G0mR}<Un5r
zU0U)j_bdC<r^>>UK9QZPMK7~2ol;6d0{ZgT`Vs+ymwTXIvW(IM2$Jmb7IDREvVI7h
z1Q0{qz*c7fS7OXSsH2s@6-IJYcnk1<em~G%b{aY!S@kCUr={iL0}A{a6rh-7(oUCN
z*}QP-e;P<LzkNCCG!E=yQcmnoCj)1|CC)zO7AdZ69d3{vm|6*e-wq_7ZY26{i);wf
z<S(54w2V#6tLrp8UPsD_k~Q&18k~K*5z>p9;`Kqwehn18I|Atca3clr@{95f391u*
zKSgw2@Ix%^(9-#YayF^}uDqq)C17anY{+neeI5;I*i1L(OCGjy5K7?5q+Z`PS9s=r
zC%Ae#(n)3@YuP0Jn||hk&}wreLNPfh@|5vY99LbK`bZrC>;uFUYXNQ4&_^M+aAn-k
z4vzVt+;>Y1@;3$G<L1HklHjZywLDz14>l!#Ou;Mr;dS)$S62i#Jbky`Aiskud4}nR
z#6)3Q4Jl`>2`%#ad^Zj;wrUmcRrE#)ScSSzA5b>D$}UmAh{{vp8U{5d;Spj<a_TN;
zxvXD=GUia)L2ssYC#FH`lk(HXd^YQo!S%>h^Ka51Z!H19pm;g};Q?oVRQnkXt_>mJ
z;p#P3SMbV0EybO3o;SgSj7H;v0JF1x0@yb2?;)P)-y#rqD-dldt!`@JDu+QY@pi6)
zkmQ!m8BPMhf&caJlEBNgfWgwH=+((xc!j}K%~Ah4^268qJNM7EaaU}-oo#8uX$|Ls
zvAK4!0ZfZ07jMJcGVs=?&y_?5i}#0wJqMY~aMU}(wSWCluM9judt|wrTi-Sf!i8jv
z*{2wf%OyZ@s+jW`BbNuwJsN+y0ev7=qsE*!K5eG9(X8-%E%QyxH2pm)pYSfm=-L+=
z;vMMcTAiSiJ`#nxd71o(XpZ5B^f@2;>8)QAUZ*bcC9;<v40NT)$-&0g0C{DtvqYkk
zQK~MVavr@^48&~&Wu6EIk2(!!PY(H0cV9JswYa0|e7KPX>rbpB?al{Dc7NLK__>`^
z!2IpzUj(&Eq;KlgE^gux!}o-DnWyh1j1xyPD6`v%;~J3#ZxiQn=6_>K5bDqa#30?B
z)`3tDMBA%VUz<ho`Bl<6@CnQx+!I=ssc*a_PIM&DR{xMA`>BmdrBL727eAX+@#o{>
zsQL@Gkc<p4(7v9@sbEz9_&Hs*xMX*GvxHVN-x{R08WbQL;+e?omiIvJq-GtmErfL@
z+z~px|I%lJWzW)nKrYcnO1$GxH%^=uh1o16S1=Y|w9EvJ)8kS0Eib7AanyF)xT0xu
zR7r>>r;4!rL_U(m7Br^S@^5l!^(UTK4OX#yssx)$7jZla1awE%4|j+QH8dZ0Qc{1g
zZK7%c_Av9+n{C0jTE|dsAIIO$dQHSu`FN0@tbzaz@H;eT5T#+v77*%MQd10px<Qo+
z2IoUfUi<+v|JSMleLL6ld%lQVScLgvVN1I~*qI?1;35#*)Y<Ous9`XG%lH8eV&MUb
z!y@eARQFs8V#C^t6j_I6a@aBk6GS{5w6$5-7mwL18M~nQ)*+`0WUf$5t@&}&ATr4t
zzrkDU8UYI*KFbt7Bn!o}sof?u21ery+9EF=``-k0!AkYnX>HbnA`eg4#hI~EaPHt1
zyd+mV_avHMe|`iA-qdFO_*pkVuxQ{NTUHVGoFh6nnY>AV4A<RD?>K&JG64x95H6{t
zhyv*&9MbUv`^bQ=-bmdDP{gTF@g=ZcVJZ7U3poqiQPwfyixw!3MY%O+x#KR~<FEou
z3IdijGdyjUVBjSs5H0nETR8!y^g(_5wz+z3B!833>*JUM#w;NFS~M*y#`1!|B!FX}
z8!da6eOD;Rrj|2tmnvNgxm=*w)`w4G*f-^kuXSNDpIN<fMR8%)`0h|9zDC@qpE9UP
zkT_fgrtFV+u<r9PzDT0W0dsAJ$pI`|r>`7L+UjMl>g%5!^;gF>sp#o}Nzn7VcUA$b
zqPyP=t)d5F`0#nx7JD0hf^=8E>zL<pdv}$S;(3(QF|K3!_TL3w;k$Wr(NZOVew`$&
zjYI`Z^ctO5mq@i<__T9D3zGK2A$^W>#7>&$67K>=>CSJ96q=Eq`mTHKm+OBWv@fUz
zQ1V3>yg=$YaQvcZXNs8*a4H9<ko$*0`#{e)O_At#!qKuV$qZ}bZ#|0@CEC3;dCZMF
zf<1XxA^wv!3@Sv1%B`vu6Xj(POEN;5pKkwN$MIHsZM^}MAjyt8-GfZRlaSr$1^d$R
zypTks?nG4U-xu-a!MzCjru#eeu3vNPr$w?yg2cE&T&|$idNUn7ezlnIKrLqd7&h78
zdH3I_Rsa4^X&<R;m`X5>R2y07xS60en?bsHh4|V9!8klEq9j?vaIT1v9Per@0%d#(
zuT7bAwO8qHLc`<<=Nl4t%Iz-ry*wKAdZPaD%O4#)XDDN!{TitDD}(=|NATbL_vj`x
zMenQUG}xFOHON<Hh2IvL<G|sLOvpNS=kB?nK(vSL_g3CW%KbspM4LK%)*YFna6ABK
zxSdCH+8F%wf;{^?&vAsx(b&o$8BUhW&M!H;m$`%s{}H!685nBel2U!lzbPpH&BuU#
z%Mfy|ZHmry=DnjIQ9>3_6Lj=E>MF-3N?0i#_nZVk<`-r=uy+W-l<Mb`zF^R)Tqha;
zzH%4Zrtddz)d6iEPt2Fqa|}tI&gZU$hW=t!GW}2kH^GrChY3QW{}&Ih7vNfIT#;*J
z`pt{`_nY_U&*(9O3!t&dZuh@>d(dHf{Sq7X)8;AXpTEt2``iEbF91&h9tNBum*hha
zezJbBspA0y%pvLF4rlpiVDR%is!NHl<))!GK8wF3?x>(1n%ACGiC=X>1vH!g<nI5&
zuU<2OoZ_|77mm)~-|ELJY@*eZ{pvGIo%SL?>U<4K`wm%@0(a>%VIXzo&g1;^<{|*0
z@S&mfc`|*WF&E=#FzA{QP=RYah&nr|2U;Zh3RJtsfE{-U5Ru)#tB9S*CG`jnp>Fj?
z{H2d&utEB6!g?br&+d<f#c=+DcMX6t$k)JukP95^iXz6q8WkA<%2u9hApN6R174Wu
z4*CaPkjxp+i{8&00O9ZfSsg!M+=>|j+=cFe)^m<*(?q0G33$eQ{tXacTLB+$vVl|u
zoLT8-cn(ANk72%Lq5qWl|6>vGk3V1b1Z(}uDWCe)-+e&?DWU;zc#XymgHJZ7>Q;TS
zYK<HDlD6Hu8gCAi1Oyw`S3q}tyyRpns+z`Wzz`Szl<A%4Sfpjj$p>A%zjniLn4zK+
zk;nez_ub;zz_S2nz+?bx0v-r}dQK#mmOA)&QR)#9V)*&)t;cWv+r|2i5t3&I>^z~J
zR$USDU;CC48$0wS{;wsbGXgFU4NJ6vw{}N<v&jAEg;OvRbaaTB8;~JYn$ek>p+s#C
z0A&+R=dbji7vQ4t^?roTdGu|s=SnT;3Jkl+Pq(&e4`<R~6oWE2J|S-=dg;0C7_Y3@
zBfII&w23iAaUm{BXfwHlo{>|~ck5r$yWG6VB<K@z$l%egkY&#std5S?KA1jQbO^&a
z4P1KqjG~9SCsrfpn|qn1t#T}LbE$fTX9f3ZLo=LnTAn4Qp`JW%!3Uvu+qTK!eznV}
zre`YGa<}WP;rj))#YbZKH#`(M5ZBI(Bp#s9RVF#_xpjim>YI%%jCE;Xc7W9I*8YGa
zL&J5mA)kR;SM=&uBOGq#Z`*L7Bkz8>O)y8l(St~+f8P0h(1?*6K1ELrI!F9kBM0g)
z#2haVuxMuO-{<^ZVmtg^Ad*Ua0O|q)hEVE50Tyb4cB~{AiRS^e0EOK|z=Jt3mXre}
zi_P_rfYLE?=Dl1Q+jOV`H$Lj1p1!W0P;#Z30!#u!^9X*RNqZZrOC9TShFX}NfJTDv
z2GBof&v699M?OGmKoWAhFXdeW`@(Ruk)rd?Ytw5Oh{A()cLa>TuejcoDf)4bz!&H^
zux5cwM4;=5z6>Wm2HbRs4*=o*H7a0Tb!Y+@>{NAt1vsy;V3k{oV>pm&sMH8^-9%6O
z3BO)W0QR?&fU|aWy6G(e<T9e^IvY_B`p9H`z<Ou{(qLE&t%{$S2Bc&>AV#H_-O<1g
zBp-&nJmj1Knf*05hr~{IR)wk$zvlvpliowd^GwQ`kjmjCP)yllj+zHLLb=@!IedY`
z%bj*0-l){5=tw*iqKu7#%zW#uJ-4ZINDJd2&9E_b;N_B+3o>*cp!TqQ#WC69*ag&C
z^-#8KR0XP>@f=%l;Fer$0|FYNkov)|To(c&g+aHi@+)$hdK`~6-r;f5cZ42QWZ}i3
z+g9M?+Km;8V4kCTxmu^+Nngc>AGT^-Uvi4D<d4fyN7ZiuL9nff-Xbpa028oYxnsDq
z&@~B+OSe4Qfsu>hds%XPpk8Lopm<HlB*zvw6Zdi*SAtOXdfj7QOkZdngeMyyXz1qJ
zE<F9lOJd%A{8-<Lb~RN$t>>OtdLqjm2wpx@wVngsP&u$24UFgRPmOwIvDI*iBJ{Yt
zSWdi!Sn4E*$5$zIZM+Ina}GhSSqDA|?dOo=^=i~aLQNz(YMzKd0?2qzf@w=U6i~#e
z2iZ8?Cm@JqZ45+k_UnNf*!*Gp9-*ski^^X(R+0q3A}?x^|M)x9gL!6gNbcnR-7GKR
z73b>Xc!D}3Cavqv(D~s17<%gsAiz?R#7tKWvRU18erICq6{KIQj$Tk;Du84OcZLpF
z#d?>wfc`_5EDc@(5#HC5slY4VNtC9nIzF0^(6}xoB%o1BNieCLAno8%gM~XoYbS|~
z@>e6Wd9UvC#6KhVkt68xyGju2*;b01tNEJ}^)Y;s9=WGaoX*I9@2Sr_--EA}d)gHD
z$yibwu&{;0Y3mxG7;YHA9Jzsc6pY79d<OJ)B=HNr3&y2K<i~*pzWjB-Oad_E({h1u
z<B<T7K9dEsm-VjsQC^lC0k#vRI!38qH>mIP$a#>_9=;p|+id<&^Tn663UMMhz|9M=
z49V8OoWH9<%zz1byU_=WVKNL7pw@=<B!1)H{-y<RMd&h^6}kecMQMW8qHc}6KE3M<
zvC5gDA1yhgYAWibIk>k8ZmjFEw823m1zcDGyP)^Wuyyqvk3|LeCCxT!ffwNf738`$
zwJ#r$*)Bqc<UHT`dwTQrBAF4+KsFV-HOK>s7IZ@Qgs?qR-<#j)P7Y=~r&!KD{9JTQ
zeV;g-aU%uBL>SJBZ1|F%aGhwji=_p3HB3}4u>=T}l)8k9esF}^fW?LwGZp_dvnP_6
zNy7N=8r4TXa{(q+=Kpp%9-vFy;fl>YW|e`qk3RVyej_-%_u2nk1fLXT(2pp$+h8ii
zO-JrQlw&-4+d+KC5*p(FaG_3Iv%B~QE~K<MmdGSn2`;woqj(9SwL~C!<GKEF7qTR?
zeB*xA`|%WOd_&M@)xUC@CNZ7gtCRoxD-B+f@J!M#=diFGaI|}ASt_&-sjb<PZXX;r
z7^=sU6a@AYt%00EQSoVH^k)`cFRl8`4Dt5sPmP!=gHvv|IVVJct0Kw>H{0*8J{90?
zdl7)&K|*?%N_bcmaGtclGctOEllQKHQDpH_od<A@4ZwdB3YNV%FUKqV>hej&m9Xdu
z7G7dKqvV1zg=ia=kDQ?Op5Ku5(pA4;R#jF4n%-*QyvJhlGjr`|kzTi1#xp<*OToQ=
zPpL4noxevh0Dmx)*SM({GaZZ{NK&8~9lgQ#(X(htaUvr6d3M?aOY}oQkgLrjyuZHY
zdnMrbx*~SBt-(j>$|f*MX9?aBf4r}#ySzCL#<bSt#=gfQR?AElJKJ`^*KyPn%`lLs
zDYV^sQ8&;aQ|H*vL!8-<EHKfSPdrgYzcraQ-`@uZ(&AxKd@v<%^2hkt*%lx1cbzH0
zq8Ux-1yL3TB}K~kXrkU@oS`aw%R9ozbN#|#`7bNyB<pwK#(Wm#wME#0LcG7RlaN%>
zUE=eN!jD;*$?>G~Ubu7%;)}!nw^tPl=T~338GBdnd>oQ^!$v^LIk%SyrJqL0jSbkS
zBKyJ{f1z%8lyI)kedh0JC=Q!f1~%%6BCjmOc_n|G%dDlxUjkq!ve&oVld_`mHp!<#
z#k-iyZfDwYTje31Qip=36DC@V$U-=BPb%COuQ^LQt&(F9NT%Wi=SwB2%>Y*ZHI)nZ
zK4eYgWOxjaqz7olm=is1<gcZ?8l~@Mwv#bD6q><JQZ#F@jOmdiB4FH*dGJDgu-os_
z{xC5n9zeSSyl_<?lRMmbvlgTUB~deJ0+&ujE$Q)2i6UasO3H3V&&p6J#LBkDN_tDR
z_C^kaDZ(cg+t?0nw>iom#V!&c_sho}-<u2p_V6D!$N3sYVA+r8YmBcV#YRk8<EN-3
zi_gveKE!v>Q>}d*Ln2v=nHIN3>J&l=b*>KidJRgc6>TdC4mKP+>dc91&;%UxbDlBd
zIc$H>Vs-G8JlzZ-9eSKF3^Eqd0-9Lo2cCZ=%cLcFH<2T4BdYZK5OMR8QuCJ7>B#R~
zVbH(#A;J_tyE9A&?~%Hq&je&##Zf^LSx)pc9-vxR40Bg}7jP%3$SYz`7$1YkdU$<`
zZP`#E)9r|;liwcCJoj@1E7V?s)ME%-Em{U&r#?E>3@Aubq0X8krff6N45tyXd-f!l
zrwXWgi_@cLEb1M%#D|D6j7Rq-`ZSP5%&+g{phxWMnNucmdiL(d<sm$%ry514zLu|P
z&#Q<QKy2~*we;lufi$yf&+z#Zo}k}0xrmN39Z0i971hMN4Z$$Y6fmb0_$_(fI7m@u
z#IgDrk*xvsyMl-<1GXe!W^FK5SIgWCCr^Nrxx@@_bp&wPt{OeciW(xB!Y7v{rZtgm
zT$DKjBIXY)sNm~jhDu{ohlSQh>~)Vqpq|wK_{g}ygQYsyCi{yx0+rxiSPv$EugYbB
z6nf+`BVTFxwSBJzSSNL`@RYy7`QX>nLZc2(27vMJEBv@R9BPXnHsY&>Z3(XdT^3fc
z+ox08puZ>`v0@F)qjJh$EZ`qd^<cTTEj|X6z)R=jh?oaLD{8S(hM+z4mB3iR4Tse8
zVu1yS3;XpAcWWcBUNIr#7}zRvy(_v^kfvJ?WX8xrMw5`+%AU#p))>&p@lEi#qbXSd
zxKNa<Hdi0Hd<A{Yj;GDJRz|(Z98x#B$|8K3g@^zWmHjTc29<4D)u%^*3#N{T7PsB$
z7>#B7ww!>!R0ndoC~B5fv-8=xfES5fkQ=>cHNdK0ph#}f@>s=G`sEuiAKgxBVA@B&
z92{;*4>Cg8+*$j2An>?6RWz$xSqTC30;tzB=xf<gnOv~dQ?<O*_L4bk)bVgGaE$o*
zHD=7LPo=>IlvZ!jdsYE$c{V*3Q6Z5M4TAG4u661Tkbc+YqE{Sv2pno<9DNPDaWg(^
zfUFzEL;Rd4FSB^x0h7I^G1x&UGV{&`WWpCo4_(im_p!d>qM}nASXmv2Ou34_=4RLp
z<<AD>1oaBA$_Bq@WAt5Xi#OO{q?ND15J2Culi^L#f2`mByPtoMgL-kOF0JNo#k~%8
z1__4I;$G^{_IU6PSRbjFZu{`5mFF8J^MSo6dixi7h10(%oU4G2y#11XZkmD4rSXSJ
zLEL3*D?rb75+oOw@qWaLU_bPq0TUGL+qQ#<DBsC?KR>X>#G&FMfd=~yST1qLm7ryS
zg{9vIRlJuWzcU6d|F7D%4T)ribMgHd6V`y&WC&KnRW!E!07&e&lQXt&XIgSkq~1x>
zyM_TotJX`W8Ac+}Adh3B_dJ+%;1yVro22rVY_O31st3ktb}!lP-lw*__SqcLu~qGk
zJp$LZ^aseBsn;s^^(pqNKuyH$dEevhgV;iSP=Pomk6TPlSoWDfC<WaDMZ5MFf?{Gn
z6);FO2{IM6q_<txt-m>a|Iq&X$IAbYpPx$A=>Pck_g7wibY6{X(Of+tmK)jH9Sk(u
z)#@Inj@luj2oUn)V92tP$CZ=IOgepp8L1Q<0j2*{J_ti>Jqd%zYH{TENWHi6JRHt+
zg5@J75r+n59a<(a%)A6}0q*tcGfzICc6k#h*(oSKtnX_;B%lHAUNWhv6`~%sgSe$i
zdYu`-nbwI>sa6HzdH7V+bxq7jF-v+x)J}FFLbto06C*Ig>jqkuycdk*#|`UW@0QAB
zX~l1gx`2A28Bm<Gx}#=obsPbls!3jb&<NA6w2&@g9}-b`P_Sb4{x-3}FWuwrZ?bJ~
z>R*9buJSUVkAb9eAysFwuGqY<pNk{e-B|i>jrRYvEE};k-Tpc2;!x$b$U_*|+rOl|
zkk<szh7a!Imtu+J7`a<XwI_s3+9^ZovvOyfp7^qYC5n?j{kawIRQb;CQwntS&#F&#
zawPp+cU#9#_lg@LI~3bR^{qr?c+l-%2{wsfCSY!8zCO}5I}`3-*IuW+1xCCYXZS6?
zz)S{VT8=-};w}YF1-LqL)3sUYkYQ{SxPn%gIzl8*g{hTRg%Ds-I=c-H+f@mil{`Ts
z!^tpejeLRa8AO-}vy}Jd={f~9i@U_NU$|7qh_1|<DfF=HT-<T>?~5&Ux`=8}L+ih{
zD!3dwur;bQEYhEhAs#c4byO?d-XC*^<)yg*py<#?flO5A($K{`S1eGs6#~h7?iR79
z+XZPsmF*U0v$625tA~ZmxnY1{&2~3Q$@MQSPJZa5eIMZQ`@AryA#wrROgGB`_-OV9
z+#q^rP}ve$pqT6Ld%&zBPJB^G(-In2_@%<i9ep>vj(#cz6oF$G>Cx>LTYw(o`1U)U
z`^tMoAiSHfW}$tk<hIoSyXij-4SWdEoq?)S*ToK$Eke0cJHM1C-vx`CTomNJ8a-qe
z(E`-jd^KjFa(Qd30<Z$FV=K)BZPhH1CyQaD03XvyPd}rh8PpCFky{p8Dijm#fa?^%
z6+dEH^~E`L6Zpw~<5FRI$&0WXb?j#bfxuAu!;H|e-vnWhh1S{Yu<sWX0wCDaNa09S
zs6$??;ps*km>#bfX)*7MFr(LnT{_M0Z+bij#I%=MgL%5nBSZmxpkWsPEAAs`e%nz2
zV^93;p2X9_+(iG$*1|4Ixfm{Q$s@kg&$#r+o%j8)Bb*_Y*#MhpJ4sNj+QVe^r|G?`
zfviLBp4d?VmmSzQ0E8GF7BLd;PH@ys#5~eRdsrJOEcIRdk{sV$W~tyzvFD6BJ02%J
zI`Jjf{=9ayi91`1eC2s7gvt^Fpcsy8y-hAtdq7h9%dRB^m$a#4O&<>Z9uxYLNL2HP
zk)d|{(JiytC%@7IgGVG%kG!5Ez@1hkI%BM;;p+y9grN{fC}4Z}JQe2wBa?hN+ffNP
z;77~GQp1#Ep&0}2Plh7Bz}voST57@fM|NB=GdFPnqnrC|gof^joF~!Y9iW-mf`v=-
z&STXauFxTQhADf-Kwj@nWu^zF$2}y?f35;Pw+dJ#ycBm<0ZiM$&QVAu=48QlI1|PS
zwE4{KfTRqIzg~&wvm>y_Z*ryuy$p45sArtHYfvu`43zVHJY;Yn51IVaqvhk!GR?Vr
zG=fc>usU{2=A0#u6_7RPb87~EF!06?1qzXAomn?d2Z0@7#hS;u`}lU5v`AA05DG=f
zbC)eJQ4pvK5nn{m9wzJdpn8<WvUQ{skMk0_)+s5b9!#Yx7qpuzUbt5eq_hM->O(50
zR;ru{VPj515^Ny=BkpT_5Baig0TGD40I=I@ac#}6J<Ll}{4#IW_7<UO$GyukHw_Ag
zujMSRZw;TX<IQdz1$;~G5xm|BE0;BBvO}lbe`{wR%~@XwYKdQp`1zo0W#Oa^icpmC
za1FNmNFTIo3K-nmg0!e+uJ&$Ac5015H(4Y!EmRp_pv%9?*r#+vdvtwE+7%4;Cdhuz
zU!~z?+`WPWVL3sp#wa=MQ=sEo<{!)dhDL-hs0IZq>U})M`Y1}?zWu@`Lz9euuz>Vd
zjok!N#K$-F)@eLwDIw#Np^4i6;lXng(A$M>2I<B8hJzKcxxtO02!@cw764|)8C<;5
zMR4yEP!zf$0@{2SAy0pFH}@qkBI&$_8{~tz;mx604aI;r!2$x;^DQuLm8P=XDlsL?
z+45dlh6Tu<XnNO}zNRv<k(9A9&&s>M1x7o5>Y+fh@at@d{oye1H!(LPT4VST)I23<
zee0o*FVD@(7@|T*;nB*KNV5B-%jF=mQ9E$_ye4QYp*t@_X@KPUWCjp6{5lTLW*$OG
zjE$aMMmz4k$6McvJ`>_KXxsXG5?#vE<)7~`jNDPzHe-gHM2-6^5TA$8RYxP?sZ$UU
zDzB1!ji^=<8s!&~$J5Q*xqqGL^gWKuM?75PSm4z*H)6lJQiwFT6^2Rc2~KNYAcuR{
zO5QIs((^VK$L02yOj!<R_P)3DNR$>YcS<0Do>84Mg_3(%rNII87p4yp#i|=};+jLl
zAde$y2n$mZjL|W}bU0T{Bh;VD5r4bi#GB))Q^!EWZHHJU+YAOuFIs@ODt;$5G0lF_
ze8E|s)N{wupD;)_#>o>HKOGAYEx(5kHSo1@PAk|NtG(Ovp6v(NZTcfj2|yCvPyrJk
z`0eQwTuBM<-NUfaM=n@4UZIy5$qCwsTQ){M`aAydC)GvL`x<PQ7_y81P*o__V3XC#
z7}Pe{)HYDm%9tC1dTkqtQJk1AtsifYDf$o_kq%j~;E-Z`FH-2{&RH{wj0;Wbeb13}
zqKu4w#~R&OYd}8cjsM;|-~p!7N4qvM#NxqyQPgG<e$Ow25`{yWeX*cK{n#>c$s2#)
zJK)6&Pupf0CFotTB_{!Na+yT}pG6xV$lgETsa54^)8&2MkG6r-D5&WG)!iVnz#E_Z
z6}}n9`WNlC<1&zGd<jqF<NFwvemg!1WUk<J$0-meZ$C=1e*jO<Iib*ZVH_a&mOw}<
z;u_Cy8`fL^(#QEoEZbPM0t2J;s^tA?7EsZwX7g~)s>JhUA|?T1zEgHr%L;&kwb-`4
z^WeAvJt=3N$>e4wqz)j8RZ2`7_6>S#ITEn=Z{-K!Cym9-wiqz=8%%v$&>Kd#g}TWM
z#>m!d)9Cj2XnXO?n%3nROw}Cp9EUhas)`wi#TRz%93u9VUQ#P!u73qS$YswJUs5O*
zs?SI5*ZYL&Qy7l46G!1EQV%`d`=Gq6CYk`Gd<6jVhT1q%M)C$_dpYF+AqQ9e9>eXu
zwnF<MN+dI~h1Pk5xc$hG2kY~`P@!YNC(Zs4lN=QtKGmdtW733Uq|a%zXT;v~x@v>I
z2hy4Q#5$oO10;Q8>)qoq0)1s}v6WSxfRFU=S@AzTQ}YP4W5*wHxz6hRu3X<py-$+2
z0a;0}gj@F2Yk<te75^~gfBHMW-pC;gv}Q+BGfVz9hSv;D;7ngkoq4Nq{ssLs8LGaK
z`M_L?iSV!VVHh7sc~>)PF0|(Ve)HUCur}>rP+LNRt5v;by4+FBk~(_csyAf~ct{={
zFJwPP+|rLEG5Mks*=urWN7F_^NG+IEOp^JpOu>J;1b@FqQc=*yMbrf9|GpG%?`uHU
zfgY&aR&Rik%4LfMFs~^(8InW*%2^LG6*u0&hhca-1Xx8Op0y+0vQTf#59AJP6(IHI
ze*7sUxw?+=lp36{(QZp$bOIH;{zr5GdVv8C5ooRQgePzP?G&&iIao*usRJUAltewx
zMxk;Ff{C+=J%eiAF(&PG>Bu5Mb>#zWnerJ`LBuy60I};jW)?JL{`NQi<cC)=;Kscw
zb%+r93peIh^YNdk173Hf!L7RcB)rxCFRhb*z5F!>EMO5DaO=(f4R-oVm+9}((BKDj
zy?$Iv<NM!{5_qDaqcmDU!<F>kzuaGVhLRu5pHlP7etGk+-|X*e>ra2UYy$PE=epI#
z{?ESm-!1^o6{zE^q{@EdPww%b{ObSzw|{}6b{tCG?_N-(BK`|^?4OqJAFuav!czn^
z-c@Ya<zbk2H9+ZoIwj94c!>WJi1VK|*{k<8xrX(1KtoGkzVAqOLkPetzMy4V9*WKJ
zFYM62{E;cyu&Y1<?FSOj^w62Tpb2X*N-+uA%UL&T1OFIM_}i-Z>uU+;k{6(lvgiV~
zXvEE8I5sLG_38Wn-BR|iA%gZY5w{0I?{^ga>z@O9B_Jh&y*v>f@tFNDg|2^H);vDY
zUiKk?YyIyCPL#sXUKU2Ah5yfW-GAEDmkpu4JXxpF_xlk1KfE0{5Puj)S?gBXzkaj-
zw8Z}XHPX=_luMUvL^%HMLb(?d%KwWx_n*GvpRJ7ld;9<I?f?J4?yru>VwER>2vO?Z
z<gdRl1=oFZ{O65`!-Lh-YFnY}G;-%9FWcg)@EUM_z&i&4NlN&E{b!56qsISvwM(4=
z_FK+b%sz(9_0~2pJrVmw0fd0oXwX1Eu?F1WOJJtMwr!C$7^1pN0%kV>uu4Z9bVpUq
z%+FV@1oXdouL78XX}+&{xh?-IS{US#7-TiV9j;5ul3Xwa4cN}t!1+p)6U@9Cf{Bys
z!ua$Jw`5<+mj@aZu{AtvyRC}3<4^oFvgj8Ym<<^RBW$nk`MX?AzdGgOWHnFakz1g8
z`sZQzRMKmc<-qtn^<B;P#zt{bx4zTNHwSD~*eJen-bP%8z3hf*%3CKybA94E?q)m5
z|E#6&Vmk~Ejv|0cDR6F*)&U5M1UMMln__<EcnVymmk`HH;7Z~>QMu!o*m4L^nmXQQ
z8i+Eov;qF{8jOtda{@4A@`p6x6)<n2E<HE~$g7@8Os|&#d5F#g^qnMfQqDmkjwc-d
zhiSUr7?RB$HCgqUqS@@l5QSG2KJbN}Mptz%R1@{<+A9eIpGl93rjmmuYU|5@N1~@e
zB87R=W5B3E>jiXLI-0)$uAzhQN^ZS&u;y$RUk3X8<K{%5;%EanR(?Qhn*{^%_x<@q
zU^+IIs3JJ@JS7)cF&C1}OI^}+L<w08@Ssds#y-x8TBek*m6=0$Z3P?-#GdLQT7iC-
zV{FJ~>~lTcKRftvM4$`p4p`iEpv4FLENX7@abMc$1Ndt$C>n$sI~(=^=Bz~^O|yyl
z@+p>`2J}=6o2X2)Xmi}=0e|KS<dLJ~Py+F!02r9J!3Pn2A$b?f0bm$RFG5Do7-xNb
z;g@O^i6<+jO1?gHcQ2kz6az%t4>Zr3;U?LjIcH^t-8H_@l~NB>$LfTtHs4BBf$Cwa
zj)Q3r!aeGM2>?}}zWZu!hv>3T_33WT=`kb*VFGpJMi0Ypz(n%cfT<ASQCkrN9&o?$
z{9~tJLLhTklB*I26B{8YO#L0=59{EY8iH?p3TGG=1G=_#CAQNeA?`CkcjY>8zI<86
za|G&?M~W{uiH*Y{wbgd8QW~+Ed)`Bc`kU(l?2wiEB(Oxg0_J9vqd`-wJTw*<yK#;H
zE~_EHtd|~Qqy@S`yJ+kp>V{mfg~CSq)`IfeQqUZ|(6?})NuV&VsV8G_2FPx|NNXzq
z&Yc86(t1Z@vwha+!aA5w;WLdnd0;4RS_J1FH>17-AW`Jl%Tw~UL@)%WQJq@6z!X21
z!X^O?`TXaY@ay?-R-Q3z6zZra@%;HxFm~H+57@ve*jX$PCc8ZjISmnv#v+J6)>{^j
z&fRq>ZBd8@wdDxvhDQmQCe$1F!5PfgOjJwj@Z&DG^0XUpR56*jsCtd|)Aar5S_LCC
z4csEHu?`WXM^3KwjF>R8gFu}*0aTYfp#LOH@8bmd+lrQbU8&w1GRJyP3oGdm-&RD`
zMpO~O&_n?=zz2_8U6Ed4Jq(tGwFXdZtG5hWNglat%>-trZ0lK&Wpy277g!4?fAb4K
zka}GMcxd$KQ70XWaSw?11Q0fRjx30FPEZP3uD-Z*E)0-qogt{p?HTt#m8wZ`UlIhI
zKzxSicwj1357{@--yrj&sg|QeF+N=RG!b$@4*<O~LaQe0uei-W^DIun5-~SvTqe9M
z18%9&{fv@KE`)Rl+W~0i<vt7WOr8BnwV!?K8J%}aDa*UYxd78@5(v^>f>mtXD{Z(^
zQ_zO)g1W%mg0$n_)%k~%FJ;qj{@Klv>LjJK_O9lnk}(7$)yV`4`OVsgZ!g~HFl|XY
zb>Q6I0S;t$HkhV?Xv1DqlB<HaVVX6V<h#Pz@@ileP$JnBSvEK90_=&d87&fR8lg&n
zN$4Q_1%)0#o2usiF^DCJ)YzFsnph1-7Z2OD&&&L1=F`-476-a0*QdbgffXPQ1?8k$
zSdlahLqte+^>6tKZ8wxPfkuk%gSP@|GMAxlT{dYA`!3Xes7+*)e%3B0YDO@e9rpBF
z2L5O943_H{*WmeF5c(x701xjcJgM23hV4Wz#5jvN$AWs5tQJ1#(fjf_829;g{*}qR
z44JnbJbcHkXZ1*~ZA1c9kr_b0Nw{>G!=8z!2JiFR)0~>#{j(6?J@7x2$w?uV_6OKa
z07(Jj{qLh_7IUZRUzyMl7Aam+<nNO4B`i_ha0U|-hjt~9KOnMYXy*cN+ugity`WFB
zNWPX!-le!v4EX7<;ip?s5Ege?Y$aE^1nNzqfKBpgb#HLu72VuxyU?KkF>Qs2RU}Ue
zb<{9jA)YhIRTd1Cpbo=uV>bjE_8~5!a)V~?X|~>ol;6`mfsk*nj}JkELC}Vv0HWVP
zE#@m1P6VemPxi+p0Yq}qQ{@Y7`{2iFS%&B;LOaPgKu5D*f8~>s7(7`pncr-zD-!j!
zon!3wK$Am}og^#P2Wijb=~H?*WrMLrK!`DhOhniz^mX!q=%z}#NW%Dwp&Mcy3icDu
zWx>xvn$4%^a~cKRC7TJ+y)jj~Z^A4cc-Kr2MD_WzBL@`Qk-@LTi77n08u_<*vB}@%
zfU{2*i)emFPQu8YiGIa?RG8ann2P|lS!UrEchR)LlXE?KJ|8KiPl@>^T}Jml>f3uu
zAzjW+N}=|xB^A+XALcuQW$%^T%P=DMPp3*{74sJv#rqlczUzu2)gt5on|k`2{Y+2x
z)`NLo^monTW3ek{pCGo4<a)079tgLw$|l*tpYHU`QpS+cHcRHb>FE!#ZEz`Y{E~A-
zoD=UXfsjYexM9+kQJ3MesHhGVMU{&GhqJd1i?VIkM-@>(TDnU@0qI7N1}PCK=@>#v
zK)PW7X^@ac5ou|qySt>jOLAbS0roxb`+nbAzqP-0{PwZ`@*o7}nftk(^SZ7xToCiU
z{GD;dBx~Z_z2OWQZT!LE?z{1j1?QBD?_;liMdkq0o{hkhn6q<|F_q+3XrrgJ)TX)+
zvC*S<K(32C;hw0aZ<KB$SH+U3zquR~+=_3wWrj^|iV@s~?el5mghumK^A}EZComiW
zwDVx&49s|qeKXab{aO%8y=-w(9f9d+Jc+b5{-nT^|Fl;EAJ<gvJeLY0VaPY7Z?Mn<
z0~@j#wTUhZibyB4ZK!KX^aslHeXWtRF@bg$Md51K4)(K>^V#R8CAk={5-SCIE1J3_
z8qk?a4#=Kf+j%Ha_T@|?<gnB$e`J+B39CyMcTTC#)=&1zUj54mdWoL4B{N|Yxx8*J
zuVG7Q>!V^Dy9NZH!}TC7<c7i^eBBN9z*X%q<Q29twTYNPJkmhyXG-e;?ZdlB@K?8$
z1#TV?c{VmQ55mM-!Bp`8B??I(_8E*4+1C!QXmYJCEA?8_a~dFOJZm_p$>ZX{2CQoP
zxd2dc=1DZQjf3iGpGCCO6DhtjPX-jHH0^~8K!!-rC6I1!y?1|tKxf|_kfmKmjNphl
z%-+8ijR0!^>RsFC$}W#Z;4i$U*ct~xozd3+zPLs`-tTbx6<3ZInnw{{gH#T4NHhUR
zQO!5|2lHNQI+E?~fXu+#+-%=1jX3O%Ww4I(5Vffg^{UcKz7n#k1&aeOHoUPau)uh*
z-y*8*0DRi!oaSGzAOm4EiE3!f4Gj)C2cF-6(uIi}S43k7W;&pnKHN71AjG20{U}Al
zwuWiKO##i6NQ@rIRQKApNZgQmENnAX8hpjMo!bv8M$_+4t)%AT8d|4|lwAeKAenG-
z<_KxUTm<OFH>v9e7wGi00D6uZ)+22<!A$);P<O~g$v?&%WU}CTt#OXOQ72M4c1ekL
z^ZuoD^4=CIiezN#nr4`%ShWIHD&7RZjwG$r;65=1c3gQBO-MFdBkDR3#!~tmg&U&o
z+_R<M1$fkjWZZmKPFQc+t2u%j(3E)Ez($!$+a+#y^w%Vy%_E`Gc^7@W?YQD)ITmxP
z@#r0?&Ok8hcNX^i>0K20tnqy1=c2nv`U#)=QRS{HSlnJd+By<7Fan%vyCVzRD=;ex
z22H~Ei4rdN>p&U=EixDHloQ~x&?@|6|8KZ|dBWq2JtJ*e1YlWgu;Vq-Q^-K81o8OF
z71;K9&PFs%k1oC%N~w9;_l*QtD_*<sX#Sv9T<il_j<1Piv}9y5m+D#&<?=%l%@sx~
zD9B#M0BS2_u1GFdLQVk2Tb#8P{)?Wo6}VMc%JkB(oqGs49^oJj=&GJecg=|lHt85j
ziL5X5T}RwRiLWpl6phA-`KS*e)A!zTn}`)`9k>y17^q%{Me#<H^A%Xlm}>t@ySm&h
z@f2P2zea{IcRikqmiJtEUi30?`=qkM5JC|d&;%~kc$Y=QdUfUzN;2-lCquw8>{|@$
z=`z?W!^F0c83jmW)e7jZn`invtls>IY%qgSjpS&T^Cv}r!Lk<?sK^*~q#WVgiSkTM
zQwwR9AQE0QPyx8yJXKL%z-8}(2lLEQ?5%}Q02jLHzFoLSHIr$6(U-h|B9>}=BR!xj
z2)TUt2(a0vz8JSU{Q;<Z@4|%gH@ho1hZlPN03L6c{<71z_|gHQC2uCQHp9H{u1<r_
zu|=zm+|u1jpaB^!*&v=1-O44qc>op&xL}E>Q#TSh$=^%42ncAaR?kwSfAM5ENlrIr
zUNOS%))$)X$Je~yHyl$HZy2g^;26*yU}t`EOs<NW7gEXI`^g<=fdkf6w+%=W=Az#V
z1fGe+U+Hh`yUF2G{Uzf%cK^9GIrz)QaBT4BsIg9vDezkU5ozDI!9aEPV%8#J$g7B#
z?%t2J_yr#ZBZ2me_le8xiEDd;?EF(oe;8lLZO-;>v@_KZcrK?+LET&JQxPhEWJ<4Q
z?{Jj$9L-fLCw1n=g<{h`q9A0NSbhwMAn)0Wu=ziM1bD_Wg{XEWLqt-;`x2_5eLY}&
z8tSaw1JtP^;%4dECaqQN23t%M$rgdGrvj6|J|}0`RjjRu+BByov6LUy!X3NZr#>|X
z8UJA{2r2uFjcu%Cv(jbriA4u+@E>`eE84tDpW>G60dlwNb^{kicq>cU_gI1oys7t%
ze*q|?yho)yq3ukA2(h+?5V17R8Nh){O^RuFLQGKl7>+W3rA2_x$aMHpHiC3{GnKRV
zpyU^!r%BfLTDL#z4kjo8?b7=Tu3es5SY~l=LmOgLHb9gn)bq);tAIWzb=#MqsBBp=
zHp~s}-B&mUU5F5-&k&ORqJt%`bh6x~jH>aOYJUv5x!g=}2*Q{_=#9$gOi5?ls%gfn
zR?UG_(B(7bQA9$CS4SVr?;ei4QuC*z_ltcks2&On4KoH4kC~`poDufsAns`sZ|q&1
zVJHS(eX+w_5eYMr$Gf`RE{zeYb9}n%I9IeHr0&KBuy4=davK4`cSfd%Fwc7BN@k|u
z9;(&&9yukvG-bc_)fx`pi@wm%{70T!s!D91l9u|$j3l3xG{IvfL<g#vvSF1xI1C&S
zJoyyU;soPNOff|w0@vSc#PYQG71Ro~nCP9N_|;;Q?u_*HN+*SSe^}Y(Z85w&{(c5V
z9Gy@{`o(yG)!%G^=U0GWVdv4Ha+Q)3`^#+@>bd|j3bQ}+Hwy~vlEPAG$K44M#YRxQ
zei|`l!)v7|BepS2>GRv$V-^HnO>V0yEA5<yb7#@?2ot=XbjWU?25rx@5q4AI(`bY3
zl3w^Mwx96f$20EUc(FL{?+4+h!gf}>Oglw&DA#0N$bQufu;WTZI&nQ)^ut5kkkx@7
zx=i+|x(UP(G9oDq<sHPjZu4B>4WSsdi0Sw%oHIAii<DWzM^7Lxn^X+@cnBRniCv0%
zsx1%I@+)lX?hjaLY52;eA-M@ZBAc0h5$x0a5bWdx?tby&sd=jXJKiReroPmS_N6_Z
z#6@8c)@`<j>KYV`KKot$_oh%wK0g_1W<s2l-9LMFX}a-><x13o26|o{=qS)4V<H&?
zJ-Tzm-S}<sYgDV~Zb82szujLj76@B2?hH}(%gv&#-ONeeF3=2$C7o?bWt_S%Kqv<U
zlWO!bnn)LsbYFedhiA2RLjoV36FhgkOMti_T?Na*{aC^T)d!}G3#Xhe+}}{v8EhT}
zddsg)ZtqT4D-Z@A$NUgp8Pe%z9#K8Apkc%9tvBa^&?H<4if*J@h-F~wdW-dT3W&|J
ztZQqCm0#?vxq`w}L!BX%+=cX(Ks<jol-a+^iPjA}f8~0T={^+R(Apt(sR%gA=)Q*i
zf_}mhh9%{SpT5o=r1*!uHI&Jep|FPx{>(G@yk{M@ve%c%b>cE>tJSAC?RDbZVDMZ;
zyuB{Ealrxp?lR#S{uVBZm+~k<l*In|BSiAcnfspR4#oaK-Vz5Ki2g?^ZCs+G0pdO|
zZU~Z%L*k3r2$H*8$BE;R2@n@%FK+YubiMV*T9_t#YLaQB(75kQVS3XC1a|CoW*0zR
zP9*b3j((ysc2`fouvA~c9))3Km?PDwu=8VsQRx<%6-B1tesV?qH`%FZ7Euf=-K-V0
z1Se8R@EVMFcb`slj^02JxYp=VzA#+qKQd`9s&iXJHmvP33)_e0`CW!s47@LXr@P+1
zSnKqrXFXT!GBq4`_C{&!zZOl`gx8?a<%P9kR2(KGF+RmLpNK<1>U6dWL?WZ_CDBTz
ziJA8&)tVxZ+0clN=LERZSGrP5IL}Ej{{SD3n@5?vaTArRtASvWHqKm$H-98DM7Q$o
z8PKhHV9@{AV)5-1Gygu^b>IAgle%EG^GfIxr6oj%u|Q-Hmz1_Z!YGa?Un3TGwD++{
zzys>2z3`h_-uH7wM&nR@IHS+}(fK`pl8bZD>ryqmUC@lhouM^RM-}(Ni*@<<-by#3
zVa1KTkDIFTj~I>N;iD1)Eg8eH9_O*mVVcpK{)ZL=sU9z8$^*=q9&MeKqm;~q{%S*g
z<Q!;0%A`-4I!YreC&F3iT@%nm%49c?q6c)hkX`^wSdR@LO;kFG)AVn1l(tJ|9nkZR
zQGK@s@e0;4ZcIok$ag1YwXHQ}zE#<8hv56D`NA^3qMI33)D7dr7U(3Eh-A`u*>v!I
z47_@i!8*3M-Vu_i!*uHP^HOpbJgiSYv!WBbC#s0du<j*o+U`qDd+K57CPE@||NK&-
zSfBaLy2C&-Ka!bE<voUUtCf7w+t}@p*>~lJ@Be_KF|=tK&AT0#?#o!kHT?}LT+t6>
zwR|yd`U6U$+3`Zwy{85Q>%=L~q%%U+dI^judj(mq{7o;Cz(EjIht&r=_}htWL~<gU
zMdb=t7wvNMNmimk+{4ghL(Jffyj0zg+%)g$OA7YXXq!gi3;mpf&6{?)W3TQ(Q5=}7
zRTm}xP^Q%M5k8+&ME%gf@ix)3`r8)~9diC3joslLoc=&PNr1>-RQ{y0p*QASuVfe(
zmHWqM+AF@{=3F?_fznAUy*`!9&}zb2=1UXUhkcex^Vjp>3?U+I{4(UE(~_B73I7~e
zuf&Ygn6+u$dH$&;c`3SufFt0Xo^H=Zh$@=LR)X0(ZP!Qx>aSJk>#pfN>QEkSgsLc5
zZ?y51G(1E*$;v!oOHTYU7B{}a;kz$v9V;CM#+J!&_%#dO(z+;Hcp5wQmeS}1VR&or
zUtM7E!@F!Xw1XQEagGptn(PA$cS<~I87{*G2gKHyt<c-1Ob6B8S_6CpWK3pZ?84qf
zd?+J)30Do3vW)k4*wuzW(>BU4QQ0G}b`12KBt*wk?eS+}8l_8P{l=$J2)?T+`Y_%y
z1T=`7$+gs{GYS2xf#%29MuYlHn2gQBwOH^^NVLB8M_P`(f^M`)NPXF~dFuCXc2iZ5
zq6TcENY0@I{O0%pv{RB;TuK-!1y_T%YK~}+eF-49Z&{o8h|F_G1*u)$BWA+JtDrqp
zJR?CrE!P|_pNtF;@B9&w9NsQB4LkO6<+f6ng^pofYfy)9U%ME1?Ok^8i!`G8&PA8h
zY^2PYb&0XF@-uVJl@9jt=<*y9UkkRrVN2Vn;l9>4@H+nqqesUVw!Qn+tpX1HnC23)
zUMCu|RHB~fe&oEPl@sZxjOrV2m0_8&`f-1IEDUuQs!%u8Xkuug$ar-DW-NBX+HK$%
z@Jh4%&9Lz=ezc|9LSt$!_Bf8f-+-~X8V$2Yl*2;2U|Ta~bw}2H6>g_rF}a$sbl}VQ
zdk{8qJGj%(;Gcnu)o{p`a8{;wFtWc$ctm|iiWSIeKZ$4ZFefwb94x+BH;fdDI+1ID
zu6(n?xB>p3lS#;<xOvulW7H#8CtAm)HEl;uXn^XG-t6}X(Gj{Wn)ickn}WHzVsmwf
zgreT4*`pN<A5%p^&-iS#4Hr|0O$LlU0>oQOf0M_t9_?~bKmBISbRN!OT_YA)a6%)A
ziAs46p}I#7#qjDG-l0)^^lKGZzuQ$gLhq9$*0Vkh*F^QLFE!#!XTI|h#s5$EZGB}c
zlyPJ2@445F+!DCcPeYa1Q27WlI(W3-C0=Q&HBq-Y*qa2LfThnzgP*n~=r_G{L)OU`
zO;r1ZqS8b<rG3G2cp)<Nu?%vxKRFu-*)OH~Tf7f}yLq>)`lWy=v2@r+jPPQbYkX?B
zGY!P1wF@2Tjq0oC_a(k$T&K7uIsD8iQDMK&7K}`v`6*4dJUwpVPK;MDRFpp&f1ZdH
zE*8uJd!@rqqwShgOz3@rKg}(sXBI0XYV%(QuN8xJL_0eQ$Cxz&f4m*{Ltc2LZz8_!
zM%I$Or!GIAz59~+rg^k8x7Aoadal+bE9skQdy8b3V^1&4o6>h$Yeq7ei5848n-Z7r
z66N>%$`Aq*ZpV7bH_tf45Hs%h?T-#ByDC|#SPT-+zdzs9A%Af=`J^vv*Lej@!6PRn
z536S06?SA|>#}t>V`84m;E0{5YM-q7d<>FLBwE$$eCVC7S`UG##q<9&8UG+l3*YP4
zfpx*CR_D$|$HSJQ-QVxya^b*rf10=dT;X9PcGliwx>2z=^$_lZz$#WqxxJr{E*1we
zI+*Fi$VHX?nCY-~oacpg^XW1T6lX?-pPy6di^_VfmFO6({2UV<bLbEpn>`UK6E&GJ
z)mc(a)4Ba1x_$j2WmU-g`*Uxf1+i7?1q;q-2|f2`IhmERDqOzXA5J<RcMiaaQOJr^
z9$Apd2`^3W{Or>!Ox^s&a9&s|I`czodsp3O+a(1;xQnqoddf`k?7A*}^P8IiTv)F#
zK{$30(_zU(Fa7sk67{e2H3n}V=BsUl^hQ~e6=UPc1!$1r7Ib1s;SL?PA`ZVC=Si8o
zuM7GZ48HXz3d@N1pg~Ry9mLLo&Wnw}wkcCr^H6RY>!*(Z9JBK7j#W|NZ?wyYzXmQ_
zZRiO2GiG})jy$&cKN@Bz<|Z7Lv9AY3`s@wXn2)tNzuhAl-R!LF;QhSkIy)fO@0&a8
zb{nz&sqvuQw9L!YAiOvMCv#)?Pkndf*{6Pe<085o#s>DMsULLk#B_7ZyTVuKRJpE?
zYWGUz4VgCtHrNN~sVwm2F2Y~n`(p?iVMFRcavW*}H-Qb+>0G6DAUgl~Lvvw1JpSGO
zry-g49JUo9@0x7A4>-Yo1g+wwI*j28X*9q&*3^=ih-prRfp4}~om`4*u21{^x(QHK
zfGPt_N`_E*t|);6Nlqysj@MOTtOT&%BklK)Ed9*14AyNpzid9yh=9R(!1E<k&nE=m
z4%sIwKR?PKzH-dkCl&uV#<m)S6iO$3PY+&&2b9wd2ETZMTQNw`Z|2;ad!T<+z))1)
z`?Mty>)}Mp0|F`DG&9MN7dl>GBGS?xg57n*$@P*{%*(Cx^HdA;fwS>aH5gw7BgIfj
z&h53P(=>X0(2N%9tiVu;@{rTNsF1;IcL_eC5lljnI-V!@D84+<CFVfxC;Y{Ja#$!1
zqUo;iPX+qwzw$i9{kL+Dp-_a_m(uf3*)kABp!2_vGDxuNTNtVcE&J}?2F3t1N`$e`
z<F96!!n`O;TOYEbrfPR%6FgtG3#c|SsW$4vMu@`)?=5`F>b<?6*IfMAmB>G+1wZDC
zAAPyhAd3(?y0G7;*dmr=8q&LrFW=xx2Rzo{eoBiT(ay*~%PNs35swvK-KlZq+xXM7
z*!jw3aduwgdcpf3{nBe;cXph}SHt<~^-la-X7|@(CnH%_#zt<E<{u;-og&iB?$-Zq
zh)GG)Bly$A<5LZI@M)G}tj<GddqX|BdVKSfKlF(9a6pDSvoxyQc@p6TQjh4J6wzi9
zp(nIs_O&l`#*G<`>D_&zRNNIp>>!m&>MJhortbBVf;j8XX?Qoaowe79-`D)kJ8zg%
zhSDf!^49qhit>r9LRXP9r|8{v&Ttx-@0JwJv8i+N8<CObrL{|*T*wVFq+$<*d@#}p
z9T=c%s(j>8;q~}jTX^j6@DmT~+AAl_bB>oi+BjA#f89l6?y@G%x?&zc6E(vju#ma%
zQK@TO4!Oa;&^Cv@*}0OSJ0upyKdwj>a>^mt%MV&CW5bl<t*8T$RRb7b=lUW>?>z7t
zu3eN~NVxyQ8{Iq7#w$?>2A6C;by5(5PF5MVmRVXlILh~P(M;C-dO#icJveq{x-#;e
zi^a32|MCKu3*1Ry!@4}hrbE;K4Toxjb4Iz})+C-8&tT{SOpd}2>_N`>bYUVpD#^as
zEc`U&QO~jON;QK-*A;xD`j*R5OPe!q+TD_@n(3jopu_z&a*?ijefib>)J253DzwPH
zf@wfdtMZ44P|<!DXB?BVqhskz&sXQJx7C)Tdq;)DClrp=!%f_9+J3N3H4t~raCELG
z*z-+4ix-y{fXT0DGN3Ua5a|{~QsO-Z_by`ZU90de3JV<`91)b)8L}Fy8Vz55NT`++
z$a0=TF7T*WCw`HU^x5(^Hkq?^*HI3`cG>y`)?xw)_ByD<FOi1V+vL7s6An$pFP<AW
za=i8OejbZ$BIo`}F041v9nZ92xHRX58zi`lbQLqgQXhUr^;;PWW)4Tovksy0sV=ep
zLr_Z7ybv8DUtf1fW0lxO(`fgJ7;e4R{B=>p$Tryde(P|V(pJ@?{RDX)tD|LSLS&nv
z$2MaVe_L8i%`F)1t)4L@2|Ne0zD7|}`%$r}+x*vEc>-Wr9=bSYY%|mT>bBWE@sAZo
z*+dmA%VYTQpIA6UNR$^<d?BOv><)=<@3H0UlDw*Z#Ta~2PH_<Umu5+Kpulg-rspI{
zjr)nrc|(reIF3JTF<$%K+mq?XG&@z>Jeu$d8im?z5h&K5Nps|=M#Rj_a*JRc&*5u5
z*8KSOyt(^tWyP%wv5?EbOR=lTZWs8k*i@wQaO?MNr}xw?T#%)VWykkrnyd~-?N(MJ
zjU;@Mk3eI;cTXVwCky-z3+qG@SsdqvX-TL9{i&24`_sMfd=ll(WZS;^TI*u2@|*z|
z0|6T6UEc-o0tfTB#-z%vG%1|I%Okwi*Be6aQrFVvn_U)P`@3*WTZhH>(w~o%HgUvJ
zaA{ZSFdu7AU7^ap5Z?UC`5M)`BtgnE<^y-K())*FtZ}Bkjs=ISw>plJ8M_Ec1_F6s
zKB~9R&EmwQ`p{-#ha20RT`kvEg-e9VZ=~Jav3;|(n7$c*5EHFjU5hl}noyOZS;mOv
zJaCp?tTF#wF4<=oi|-hEi?dqWk6b|;D8{qBJQY-ne9_L+ZK-Ez%4>>FiUP9k4_#5s
zmfm^{%rELg3<%h{({GBymcxux4mSrlEziUEhG6cWFybc<Q>!>U<#cJZ9X05+Aet9(
z3W_@KBlkw8x{`8UBC7GTawd<}iRqZbhb!C1KXP03n#GF51)XIOsJY+HWMUq1+~#lU
z6XPMOUoeTM+dnz>cVVVe{LpH1$soI`VIwx&F4I7d9f&o_v+6?1hv<<ERn;vjB7;gK
z9E{~C_YM2?y6LN;_cFzo%sQ5KOxg;$t&E9mz17lTf$tB;6IrLEjoK7ATWharysM$o
zeQ-`=i|y-J{llJL(AY40`2dLZPMx{nYu*v{tcq?_1Gtv%Bw3O=uk_)L-OZai{E2nI
z;it_|oy`*^ywq|=mha^R2dQTQStXqAEWDH8WYO_U+u33_NwIiN6D0i}Sr^}Kxx2xR
zcmxkWjj=X-7Mam~SL%xpN`zrh<@#jT`j<CMsfRb-Ae4M??J#t*GceJY?hnNA<9BBI
zY3M*wT}9_WT5Qb{5QZ)Dp?-^`V7Ek+Ej*<#e10<?a3l4R;|BMvpWyVl*XU#=#O$6h
zP<E{gAC*t2WPXIh3xa1cp`shG{-5netRX+R={($i$U?O$Z`Ak@+^!J<Is!4+PMPMF
z34Rn1dbCb%x{eF66kaa4sY)oZSucnEa9&ALh5d`gXcwX01lLMRhdx~QmGrU~v9VGr
z>F?2Tt0enb9cRkolznN$1Vdln&D7kXH|3nbtv>q5on@DG|JeJ$GqbBsevEhK^_ri5
zG>INyA*B8R%#=as$JI0KJX6Q&?{8Rrnbm4&j5?=2@xoKpc{q%({VIFEW-6f<iHXOH
zPfBLVP+;|t%&dg>(v$q0jFG92P)sF<`=d+Uu6>am`yxAUNQLd2-}}v~A*;B%<ppq?
z!mPyT9Fmdk{(2(<k-lC|fl1^HBIMiNjmjRSK2!u*;5MJ4hkfG~%gQtPANT5dCFf%i
zuXE}jM<2^lqYM31cyx7XRUvvQt9OjA(>WFJVdR$1n#_>1Nm@iVS%Q^*-EcFy?7pqz
zW2K1TrqHLZ%MyMR<q<cTW%Rz1MsQP1H=GCsOO7nOyi(G9RpqY#Q4fOdYqW892v%4b
zDsjtPBzzu{Tsd_n*y|DA$Ik4wQJWf%<4L#>CN`tu;+-?Pe;ZPochKF;O%N!P>YT;a
zU@m@yw*(!1MJoJ=|7jl<=0)}L!xVb;pu?`mkr>~|g=veq)}GS75&TRhEJE%X`=oZ1
z#8perK?c69>ML9aXhwDQ3HsZt{U%Kp@?D*>9wRjAtSCPsTg<iSuj)%JUkJU3VB_X7
z?>}6emS^EMQX%7f&&K@{&*+|Wj6`!eCqoJvn$XV!Pz{XGl~{ln{zkh!<eZ*a18t?P
zldHVWa@D93!}rN0O!SU3hOaQS(OpwsY|JseWoh>g@|`XXE&X+Y7gJ__X00u##CtWw
zXyB<1-?TwFagr0(Q8I2}=Y4YA(kiU=>Lgq|qT-K$DNm<Z)=@#fLcX_+SW?m>0kaH=
z;9Uy`kqBthn{so!2O5n$8)&(Ji_~q{GNAohWz8bXztw)}Rh5i-ViS4YqAasRq(<u^
zXLlk0noGhl&6;`9qZ?{pgo9y;%)Ra}Z18JphJ#*|;tKq1iUS!;=Zixb>)1?8V*M^V
zXcJo<K*ZkZMu81BjdRhQq${@2COqsbTVy?K-lA=;4-I6EQKijsVE43?zE{N@F7eVt
z>X+rhpqm8m^i|+99<ytHd%-{S<X7IePpg*J#7P>RK4PsbTc!%HVvjy0e=eGswfF1&
zYRUx(YCS)?Z&<=$a&l_vi2P!@K>WSbT|D&%12FA2BRg!X7~#X3A@5?R>S%-{3L(4k
z1<S+5LOYh-`T*?LR3Y4hAunDwuE}O7@vdD6(k8r+B^n6QcX(YTlN!T;JF4_ns!hWp
zH_h=S(@ZFjd`!I-pGR}JbA*JGNUIr^0&qheLs;r%`PYt#^}=(1v#V<CH(x5Kw+Ulx
zkBGyYIt%ip&c{SM<K@CUWa<JB$;fx`*Mi=sW0(-GS-{T<l4_3&So#ZW-&}VZ<r}YM
zJ~@uXt=RWeB2O8--pwiN_NG`@BUBX^G2?c!f}_Ti{E4k_P5;~?V@z9k8uKDOckbgW
zLOsVi($+=k#JbvI)SC2rC+`U*?``@!u$)$PXOmj|Vszc@?f1G#=}zIFakmjWz9@h#
zn589WaWF-eStsP50Y#;R*i0S))VdSzjz5O-d7z-&CgaYWqV+wLPyz@zi;%m-_%}&}
zobH>Cb;X^Ki$)oe8lURSiq6HPQfOR4mLGt$;yr8(^1b7&KBdeB8klV+cs-AGe=m~U
z&)haaV@A1BZWk&GP5g?a_Hj~H@FFX@5k14JghMo0>rh!%bdwA0Ks@ow2Q~X$XA+L-
z_UApfvfuD9V<0m+v70A9uT^?ZP~W!E%lKwQ7e8B$K0-0Qe_8UBnb=`5*qDwIWh~m~
z<XfwcU__iFyCNE-a|0!nlCAr8N2}@BU_`805|12Xct(@oZullG$dUhU*h-6Cunn&o
z2d#DxlZ@C}tyy#jExT>IVse5l<yQ8mpvYP#ej3Xt(~4jz8X3}e9+vkoIqFO8Ljw5&
zeyheYXXEmBq$ClR=P1;>i#sYqOGqrMUU=T~CFPan;drr%iWV@HO>P$Ke>j;fny2`U
zy<{e4Hn6v__bey;hAo)>HXO_GQw?T$vAjs;rSWoZzNRqo?g&duKEkIwp!z(jo;Ag~
zqg!M^?ncMLI3D}X4BcKukuROsG3G*AJ3BX*HgVA8?`N{+k1QgB#cKxJhwA#eeV`>F
z7h$}p$NI`PV!V`|<Vf6Mgw8AHcOmFxSJ{+6wpOcWwLVXTe?z-jb`wF|%-tp{4io3s
zyZV8jcU$)IT<RMd%zWZWSf})wy~T*x$KaY@5rfG>bGXJ8{jcB=3?tCFq0(T94BBbF
z<ho#Et@&4)tu${BBpI}1uRM~NqsqHJAuHpe*RI7dvFgp`lZfssi|a6#IB}yS*W7~$
zI<oM@C){7s&`TaVeBvVAhR{IEtO?pu9$0f1VQ@}Zw0RVz<UpHp)?>2_Y=PbLa~Kcr
zXKsF+*&;pPon4vfNbL*mLLTwntMM~%!_4Ov&61mkd0A>JXL4B$3%`a8lv-W(B7Uk-
zLmY^vonYkFhL~)nexKccZzuCu5DT&YG>v$*i932CpZxdj^77nNxpK=ztQQv|MoImI
z`w4)bJ@d8PJef^Q<vEcq)Tz;4jXU)MWEKT!kyNYOZy^K_thF-iR>L?hR}fI3LY6Xn
z6xp1IEaa2oOp0k+lV=EZ)w~eqJV}AET-CFg1IaqXL4B||zqE5!(jsXKM{1MO;Z%{i
zx>6fvA=D*6X*9gCQ1?1mQ<&*jy{X#{krL8Vs>|Sdo4_?2UzzVi+QmshT@tevS;bPU
z_;=}~_O3fB<FgeX-<S{EpcgB7@n%*-?gp`rCFRWntGwq6Qf%*~m{T(GR)bGQBub|o
zfZ5j>)Rg$|XO1*#O!Ot%a`f2j<zPksFj{`%9rP{ts&3!znK)_nJ5c&=Zufg?4{C7f
zo}UcaoAGCp5(~x7E}wY`Dle*^Y_==UoJ1>NO)hVFp0g=3Uu55!D82suc_q3rMu8^<
zR*;s$7GBt&ai6gANKMo7DeFu<El2hNpb+<lsc5Z)+$0>owfde=4w}~*R^c72)JJ$S
z7=ee!Omp>1JJfG-)SI>@FILZhU&-E2+1G{&)|Ri=0()qg-}KxwOU)`Ha9@9FTrB)%
zMJox%2i9Oy-|gB;D=q_VA-s4od-uo?_)Q0`(yy5Rk(oCYRrX|mN-#=pU0rMrCX}p2
zi|q>2iTI(fyHlo43VnATpQ6-932l)6gwJqJElC@EmH6HtqjmG4C1il|=a@yZ;h%Bk
zbF-0P_0g|XO;^cD$E3FI2`}x=+Hwa4I`n_`@-V*^@rGmC16#|J$xK?^@g>{~PN$82
z8zi6SbPsGT;mZXS-YbbgG&_t<;aG4RctTLiPpj^S8zmP)gv-mBCl2Gn*?WiwY<4DV
zR5x7h1UC{3-+o9K;p|<c3uLF(JcHS9OfQj;BVy3_nMv#XWEPf~i2J;flow5BST0_*
z8sckCab?_@1@;01rW84Gmwxp*(dgGO3cc-Fx`M<D+_tbDInt<Weq^0(#<PsNgXTx;
zh&>umh?*<99PO;P;VMVv9Dks=t=e}(syHmJ=XKO^k3dut!`HF~qH-t465BjI&>icV
zsd`vzC1>&S?8W;Bg2O^Qp7#U$!M<}-A;~y1<E)MRfm0aZjI-Y6!#&5bA&dxzB~D&S
z!fYHa0qg;MIm!{Kb5k5?q0)m_n#|)FHRQWg8rY_fNALQ72{Xp(S&&P<>pP>?BqjRQ
zut74)IW8yI`u>p6CS?<RlsCfoa5DaIvQqT*Z&vuzWZ!L#<!;o@$=^?KCi9FN+eB?9
zigezMpXKkOliL|~G13(NQajD4wIi?tCE+qf%yjgNbo>ZCg73$ya@oUcHH(q{F5wlx
zhhCY?g*hDzd*4(IJ~>j~<e*W}J>Y_hkMgRm7dujEH?i|XFl?j2n8pPw*lTpP%r{_@
z(i4?*uBNcqA+7W?rqzBzZ!RlJ*;X2enrr{OY<ix0@P>;Q1Lr%a_yd$VU1jDhDh!I0
z%M4#hdLvG%M>kU}uU#`Q`YjN5WER6SY}SIcE?-yXX>5LANENY};3~7=2~eSm^i0b#
z8H;WaHAd2W((gM}PH(TJ#p!!-)UHHfn|I@REdtjoZ@0DdYBxL33HTh}k%&6W!cIXt
zMqF*qJa}(JpT6i9{i7fa1@fo@*bIBxiq;Od#6@~EJX!)2$sG!!$*i&#9DKjyR&fG)
zl@<TZcZEMDRx=OS9_8f_R;tGk^WJP_VJ;(A7e4nwurs7^gtD0_EB<P_jOgk8yrhHQ
zEh2pt%=MOY4!Q?@Fxz5+l8S7>%(0P%Ui6ZM|C12epmg&6DFokX7@xOrpii;I*QPcO
z$MizoEkz97TY2X2pZUs4zAA6G3U#0w{etV=MERG+LxvwU>*!ZVjs3X1h20@{RNcZD
zRrxJ|p_N`!nIDBpLzOYGj8!QKitd0A2kS^@D^==g_tV=)*ypi+^@o1Uq3I;2d}XvE
zsS_N`g^{xRKG3&Q$zAD#7itR)w%N(;Zw@98J(3pfJ6L(TknlyWi{X><j~#9ST(qXi
zrJ}UIkaXIN4<9Meu_3|A_bc@{Z(fUHV6sCflcoR#7s00_b7Km-YlZjv@p{M43LEIF
z#CBD0X=`p&rOxX39;}g+Zb_#`rLa-W$tdecZ9KDn6-v4cJ)L%ctQ}O-MtTBP>#$ZE
z6Oe)yM?#GK+p^(77Y&e*yqiBY(<?Z>^P1K!*7;qoYVVJJzx3L%O$2P{fKO<rPVC76
zH-S2J8*Pcnx(nGe=e=0Zu&Q?q#&NIrPNqwgrj^X%LX?$e?)&y@OtZ+j(5mLM>R}KK
zGApn`loru1!hWfEU5x$iS1*&ZBM$ny{9X`Cr^R{aL}*4fR=3-y@;7JnTaffpyD?5@
z&UGB+)_R0)HQ^#@_*MIuZCCYZC@GM?<w5=6NDt<W`G}nqb+@9y;P+PUc7_etG5!MQ
zPmQ0X!j}?0-;da<PD%6V2xeF<c~5m0n_&rcc@-E5Z~s;-UKawg21HV1+{L*Zt7_Ac
z&X`xbk%iAcBf(mHBNdGtWx)?U+C<YC%m}Vu9T`&SF<qC}5>X~;{ca{PWQ9=iW0}Q}
zUq<6!I_KgMrSL&gXxST|A};i3;YH`cey13uV%__KD~gS$sE2b=#kjzZC6Mm)Y4Bl4
zur-e7NnzY&Bp=R>xld#k+Yla1-mt00XX~?bS`K+t+xrs%V|?+K)J(NjFVC&tQa{-l
z2i643ebKsAoRn|Bi|UAHJu4>;lMo>fu`5hrYLE4r(wNP=RMfA+5Reh$^+uu&QrlL;
zvD$gU!g}!uG^E4jBHHL}SyYJ4kI{qYKt<xSmvp*X;9C?@RYnstL~^jk>`UbuTX$rq
zrL|wa?-#+Ui+jJM#A}_^NB?QGJ{WZAI#w0;_ivVj7FkK{cWy9bmIxn!so0ZClTks?
zwlSIR^Y2KfbMscMQuwPeAV%(PK~p3g$HlwN-f1I>cU0$Xo37zI^WthXkuhd4)Ymj`
ziD(O#Ga~U+_{#4T-%J;&Wl>Q~zAI7X*FGlqd`0p$UW@;E_#a>fwN}hqz8uIoE_i}b
z7J$9@CRmw-Y1oE55UrXvW;%N%o%fBJj;hC$%>!1%SBZcD)X|1ca%;j2S%?v-2~rAj
zuAaNpLql<@x2)(^Ek&Rze*E&O5*y`T)89@Ml$aGMvVv=<*!E<vyvABIhQM1#Z8Moc
zl}ENdCuKIj2p$|*&g@yx1~Qa{VvZl2(49lZsL7&VdMV>hZe;UwT?<)#UEB%znnNXE
zQ~H_g5~5P)|J|et1hIHN5<NJ#@N$@Q=yE1T>uU`w3-OW>!efRsp9hPjm*Kd?6Rjq_
z*OT#zP(G)=K@Qy0vk4;=&@mB`-%?2^&5Y)Fy)QqM<fBLS{Ki+=;Az~cGL~>thJCd%
zE<(EGId`f>$mJ*{uY}?V=Ol@`mov*2WM(l>Y6MdG&<2*PJ>Uu<D2sn$`ncN!0$PR`
zJ=%L}&-C&UUva&=t@)SIkCCRz+~{O)?W+Ct;$}*}U?(TJ(SG6-V_=WCDE;>AJ<+qZ
zPxNt;6NEv!lTkrRqkX{!mb9&w$&RJD-9_NpN|i=*khQ6)Ypby~ANSc3O~xne&o3ZP
zX?<TB!5%%5s8!}zN0^jCL+eLuoLO}>)g`7$&W*WRdEX2bD#X8j6z!!AWN4x_kwU~S
z%U+-EOvem$-PaN0WqTZCcWj(Do1CQ_LvY!ys2}_$^v~o4yNe}**@Biiz$K{##==+E
zOp%ta*|?{yX*y3KYj^c9aD9<Nhh<s&mu%sq8k5CV6yGC`r-D+frl9_M>8=FtnsRSV
zVfghilX?)jrtjV~FdrP&VB`46p_Gkp&kQye{u$#@-dz)O^p26Z8FSsh@THVABKXYr
zwzFB|CaDXT8>-H~Pqikne#S&m){bjH>DVT|`U7x!=Z1*3#mOq9%-WPGWH<XQ4XMrE
z1d@hK^eV+gPL-E}+dr2U!PR;1;HmQ?f&NXkiB}y+`gG$0Psk6G*Y@q@aH+aFz$hq?
z7B8Ov&6sgVR>?g(Pm9vrq^RG6DVR&xN%;9=7vKcsaq&X$dSh?F>98cf{$aqOb+L22
zak~*&CrHJLjpnRpYBC+=9KrAn6A&y-l+ZT>o{)%3aci0DOR*hlYfR_cFuUZD57}Pi
z%Ll$AMXau-=)U=XwHF1eDdiqh{~!vr#HOe4f9`K{Xwy?hGdm_>l#084N}{qI-PzMU
zoen>$Cv_~Sm?=^hqSM~FmTw#MHKqM>$jUG7i%y$z;(q0t6*(BXqr*(Wpu*od52(lo
z(tg51><3W4jcYsstLBdr514175cb#1Yv|0W!k07dy$ZB6E?ts7-J#yh6wS{`(Ycy?
zm{hiYD_HDG=Q&S@noBYh%s;Cm<S5Vo1Krw!u~QtWJdjsW5fx7UeUe>h=v5xOW_E8T
z)STh~S5F!<GA<~W+P*w=$?W~M8LZVeZr}|wg*vq>9lqcp!4Di2VRYshO0SC(RFi+&
zkXNWN>D{l)tYYyWMG=(08p(Yoaw#&>y<zrk9^<G31HZQ?d$H$KLIqt~-!9x>iy-L3
z8uZU9G=AT#V&N<q(W%l6S)hQIAL<ZF$Lo*Xq{&0L+Im{S%0|$8j23f5rFhE8v7o-B
z4ya3KLYSGE=KGU*o>NC~Kv7rB{AYTy<PC41h`pwP=52Sq%#xYP)!|2Yle1`zppFw@
zuhL%54R;4bt2WK7cRkJ;@5R9H>{*Esb||-6HmhjISRTSKZZa*ou8+YLb_w+1?U55x
z>f|Z^y0S|PVF<7EQKEaxO=%kSbX|h9TzdufI7B4PlG~-+cmpOU7&s$uBSu=@=TF9*
zdOym8G1Vu^?ZQ-vrTryyh?3Vti<Zg`PEoFVT0wmrPR&9~-@}nNo6xx(hdY>b4@GLo
zt&KR#%FT^X?@@l-AYi8OY{UwxoLW!02*o2vkIAY1a~dSfj{44<-FtK)e1?=d$%#{?
z*hHSj`A2^?OY}&*7MEikR$MJy#<7G2s~1d3cqA{K90hxaaPy@kU&F4y#28W8esafm
zq)k(OlxV4ZdTmGg%sVKiLRhnWYx{~Yy>_#0OAygS6M+GJwY{5S+3i%$xp_rBBD(`I
zT=2eW$gls#L4?gj`LrKx8m8j=a#78^`C$$F%snr$n6Yk>X(cDeYqhUc{H1IWdBKo4
ze)6bf-lufER}RUIA=ZRbBP5uyO}XOVp1nGY5t7tCgISvW#GCXM8ZoqXKc&dz>j}vj
z&-km4B}F+x=ghRgVh!=A@8`gbZG7*L?2EZP6q;BY#nV~Ze_y7ra25y&h2dr5NweHl
zO*;jJCV^qc90Hw!Ba-{1JY16$W|sqsL2FB0JdUJ;N2v{HBO81y-7^4Fu%!^r1C^%w
z+v04A?hGHCZoT^c?qz=FDtI-JJ(u+>Ks-C699cU5GBmrkY%sy;5DO!R6Ro}Xk7M;t
zXL<C>ni=gb6+}i~vrSlhRTKXF!H<f(%mUSkU!(%%OcHO=$VU4wvG<DlIfEH#9A9Pi
zhgMl&4JM^?k*{*jKnU7MBbt6s@hT`r(A0RhhQgJYs~(?oStXR3^fG8Pj2;~U6MKEb
z&&f?OBx@Ozl8%xOPMp|h!eW98&HAeRg~2d!_`imU&g*syzF}ucyqxym?IGUmP0!-#
zZLeD`zlmOU2y9V!XHil+WnncRpwSL(f7bpSbI&yBnt;1FN+OtO4|5XZTY>I`BG_}J
zk`{PuI|_jQDCYb+g~WOYVx9FH^RgLn6B>o~I)Z05rp_5+t`!vX`G1T}a(2RX7)+Jn
zz`Oj!woawiq0iu<6LHOTPw49c$peUUUAF@P8+vY}uv)xWMd}xH`AtaY-%khR&iJp+
zRX*779cv7RZ#gGB?O#~puh0JTMNU*EdK8Z<B?BqXkN0QXpdVKF!q&lF%e-+?n>i%t
zseXRQ)l>7Lcjv2NbFX;p)C_7)<$i_Ncb;b*l(`67Y`EOvT_-Y*n#UT+i4I->8|@PN
z_;L(UrjipG^~RW$r&j9OQ6$T69$=*eK+RV8&>sqY?bqoN`_7*Mq@m4|G?hoc6`Yv4
zj-_y#%`CruZu@Ca$9F+fm{Xkd<d{sII)8^~GGO%A8*23FW)gqO2spP!$;CQ(Z$VCI
zPM+74^H0$ZqvUAaMfEkQws)5i3W1an$F}i>wsJF-YSd4TX&*QK0ilK`0;ylzLGR}}
zutqDPDdUs#QCx$Ri~NSEd@ge2P)<C=e2s9mpMT9xKRLN1T-i*SVo#Kx=E7^--JW|;
zX`G}mxCw+CW~qD<u^XxEPs*TB4#V2Zpx_|2*WywxrJVzSGT|g8HPrtOGqnG&Wu?*I
zJ4@t0J$S<dm<$NUz~3m`f4?FBzVulB7-4>v=zY?<Nlk-<kJK~H+hmE%UUZwr<ibH3
z0JT6?dV5@^G*3A5^?KW#x2>_WZrmeNpF=fal9flfzIZp)E(VKj7GF1|6OBa?8@;;x
z$JIN0w?9~37IP2|wCGI#3|eabnQr6`6xPJ|@ojUR%!|<(*|VJ6bJ0xh`(|Tv<4?AR
zZEXJe+=PIrrWqSddm;W<h|k9P&YkcWlU}A`Ly83_T)j8i)Exch2uDu0f!G$dPWSd(
z-JfXSK&@oQcf7n40(3$=tGxm>2|b(me_+AqGJ!5kDH9d7T9N@&&ZfgNs1Bw0tW{#p
z-#L^p?W*CgD~%Fr)y|MUh|#WK60T7=GXw~t(^Wd7O;aY}5Jy(^J<i|`Dg9PM8~5x2
zt*q&x&kUz1WD#ENA0M!B&`1~7l@N5Vb4wYtHq{<KJt=GgbBj@EX1ud5Y#Ms!n@_43
z-|b~lIO9>lB86Rb%jX}KD0@@eRXN**E75{8EYVLLw4CpaAOl3swBWA~ka~cNBMFUL
zhCP)&<xIRA+MeAyvBGTYtX_Ri6G{bN2J(;3UEJ&6!+<^IbX?YXx&Xi-WsX{XB3=`*
z$!n}9eFwGW55#RNVX;hzAVFR-3vkyWDF}<3j-~lqc=t`^ItLsT%^Kfcy4KLOU5NwM
zQ{4{#g{haqd;-Pj%Uzd772cMtAHr%$#uaHX=%mZ`lEih8nSb5D;Jo;spmVl-vS#N|
z(JPa0ZY4WF5V6nYs`>|6{?|eKr?Lk(pGz$P0sC3|qna$<IM!F!ImD^Dc*-C>T~syu
z)9yS^jd^d}A+4!|pL(0@7_ju-5U7}fLPa||lk8%EnW*(N&{XaASXfs39VQAu{S&cU
zH&lxy`q$nS4c5&Jr3qEMmwHBH5a)UvE-M-tzFBEChCrwno9vZ`u;^5Bp}V7T)3|+^
zS8%u!4o0yIrc$qpsZY%6v<%$8D{!M-s<qOX3DxkASbzP7pq-(5-+Xr(Oi?!a9<HE3
z5(L&TJv)s3?-O^gkuQMX_SyK$@<0k-Vb>7o=`+8!D76i5B9mcG6B~fHU-HIXZY_B{
zuwVm>_Zd3*W&Bneih`H}$w)OIE41JTV;BoZXGf;V@0Y^LI)`0vQ6SO_hpbwBE~Bs3
zyO5D}h5|Sst>^XTTDUOdq^E}?<AsLFXAemj5xf$+5?8`J`#B-mALKyUu&X&MAGkV0
zyk>NCmk&U?sy+GdysJW~b43l%_Y&>h)46*C9Dvo)<j_%U&P{Q*OZu~8w3KIAtX>8|
z)NW)`4nLzPuEqEATnhF^^B1^WLbFGM!J2nQp2ND<G4kgD?s^!UKkFk-+>MAhlQx*A
zJfD8q7iS@!OvPkE->ZMg*7<NIH}k`&LWUm=++ip92pM9#%q52t{6L^=RjD;Rp^l3?
z?%WijoVw@r7z2rwLC@P8)JBI418k7B?3Lf-C~3O!fTtgLG0%drOnFZn^J-;RDYHyw
z!}&_MMa=C7uTcJteD&PRGOAUPn;7PcTB>m5;+MI?oXZ%Wvvm>5?$K8bRwDl{a#r|8
zEm+w1g%tqLrv}mb+y&%{Y0h-B@&2bL4IWC&*J^HTcGH+|nmUjKEJtFk0}0N6?8@J-
z6)z5|k@+fh!^VNTdEcUrQ|?@dS&e4f01?m70xaXqO&l61TLa8GiVvHK&BW!k&iBr`
zL`w&WKI?~49tOB1`#6Be6<lagy?^szY_wOqJz|-<zdG0sNGai3WF}Q0#N@K@4a`Tq
zVYgFwO`zM!(qPW=Tcp>;AkUEKG@96<e?(y#?CmkHPw^$Audz?K&*}yU`+04zuPSfz
zHZu&dvDbRp-udpF>d|*{oI5*J!0J`!$F)Z6yXeHkVNsCz)TneGCHtf`P@Zqm3wBYO
z=ZZ&vrEg6YoZCc~UrK!!KM{L{zKv)Kc#N0dr6m8DbQ#H<^6*I&y;QQ~qi;Aw_$uKD
z!s$IwVts6btV4B7C9IdU`JB;xpY$=vGz6HvOg0356Z(e-fE~R+jiQtsnxMsH8_)Z_
zojbD&RGtnKg-be|kiF_Z0^ZUTIRnR*3=QN=(&xr}9;8`CY~ziY{Wvgw{Pkz#P`Y_m
zO;Mo}Ja`qQ&x`O6TfwZaU`#G)`OAC>b!Gqk5P^Np+YE4widP{@-EJSHo&gVu0`csY
zw7{}wP_bgXXFMRfDmJr?v(p^sPU3Hi&b01Wko5LTy$R+lLLAP&E3M-*;Il%<NuiS%
zhh_eMA`dKo7F09*?jEN2xH$Fa%fP&X*bY3Y_R9n?|HJ9Q0zLzx!1G3>IPODMLTjj|
z7<0-}TFiJFUUr&IfKUzjL?|%>O_F7SgQWQ`#yOylx;XL`5DZw^ZS*Ir@ERbAi-Vy#
zLIRLpLgrM9=RUtzfSAw0@n>Qqu~^WTHE8{Y*}dXJ&xr3fOm!6Y_4ter?gi5M(Ak|-
z$L&5$k7-V`IT4z^^FeahaYz&vF;o>pla6;g2h@J6(&5U{i*JY>AsMqEQOkR7SyTtz
zL@Z3gA@ClACFS^%2?L7-vp2h26h39D!h1#Ny;DXI8HMA;DJlISORa2GmVYx{FJTU*
z{4rdGx7f{dg-#6XRFhx#+WxvrxD{|951?PRsJ&0rGR^4*E<6;bfV(kN_{Zs-=Rrfd
zrR@hvVJD>POtj%lJf2kDA1h2ps`p(g-2v40KU~Wt{(uoo`k?D|iz{7#kkH{4^vkB$
zMUU!E?u;rOt8k6^ps56f_|})eG&~IJ5FRXQZRQzZK$2#;;!v)(D4_79FYu$uZh|HA
zaXGQhquxr;`VUTymqu;!BuX0a*?!Ar<V!TZ$x8(u_9fqJX;o+WpOl&n90S=hMbaBH
zY;TiyW9IJJD!bYkxXF?f8q<1-``pRtuHxVzPR)bMrGgXrVadn~+K+ifNoOlWhvqVe
zZq?gE7dz;A&S0LVAjst?0M@jwMMYK>>6rNg7!K1tUok$-6e#cw{+8~9cR<^w;>svj
zT{eUx6p}YNqs-GDw`)WeAG(83i(0U99*(|9zIqix)_0WZ?iZui(Bw5A)sZy-p0;ZL
z%hUF6!sY)stbc#9fVMf<u*l!7OV)#InS<X8wixcG=lw?^5YLE?Ek2gU(|e-I9n`Gr
z&N9oT4~~)9%*zwF5Cqsa(U=*JyxdWEj2AD(-E-b|dftZ*#jg1^W`XtShot}|SL@=r
z)R?0gdS$p@&%DbKRp>CCDZOjLaPO2!Ho7lEtax}E00g6SM$o1Y+HV1<|L<k^fBT4Y
z5KkKnZ|1=>Gc`}o7TAdI0}(!Ft07_7>GM9mNP|MmnT9p;-OIagPUt3kYQVyx-TDii
z3eRnLGZF>301GD|CfgON$(W8aSod&HZRd^X#6dy2qD;~&hK*>W=4%uWQ{rR$nxZi(
z=F@Rfq9@H!MLB*{xKB<*rzv&W_9fH-7#+I;$-I|R>Bb~@4V`o-vgZ^bLGN__Ux4^O
zUU`3B^dcaUq1w|_mYM_3((`~S{|Dr{2%}v@7n&`eHJ)884I>1$o-XDbseJI&yQ&6f
zwAN}Yr-n#V`zxPdwH^6hBp79Q6sDl+_PtQN>-2o}bK}YP;^AD?%>4;%)5vEco@92Y
zd6SD*3um>DdWc}x#r_A5HXPuxa+)At5LcdVIC|?MGR<fKd1fgAZcU#Bl}`8oj(l6N
z?&5K_W{%(cPj%}*y{5O)OTf<po?3-WZ1ZOG`0+$5HH&JH-$563G2ow#<A1dYfaj5h
zeBT{3za&!qfBO}r&rlcKR<T`yk^F~Z3rzJYpmnWAhT~P1FaGDl^5O}2S6G8O#s2+l
z|Igp|e?NC^fqwWm!MSW~jQ@Xs!oU9C|Nb|<&_zc8t7w=6FjD$sD`UL;4_5<=429_`
zr04~k(7%4h|MK^8G9mFy5*xo_<-cwuaCQ+xw}|Mz@_uG>q+yZ0_pWXTgK=UQSNIF<
zT-tz!b#{BQ3imSTncNcaV}Ku<f-&d*!EmFBJn9X|cox3-J=`=rl*9!9ds*a*htoq-
z1Q%b{BoeSik#UD#KM`dg0r9nB?yyp#^OsVZ|I43<=MSzUeKSwVzrL@TleRgBIW^Sg
zeN?3osMvGJ@JIy_MfRWZBH+m=CyYV7C@ASSa0MV-EilYknA`@U;34HqF>lKC;qpiH
zoIt7S(JCtsMQ@7UT>SQ=s_S7fl1y;bDbTolMFJG}GuHFK_<~dO(;`cHWsorgD_?In
zDO_5=kTs4rnM|m`<7-#1PiH;w+>Zw_f3UeBA2<()tGXT>03YsC9k_+)g6H@g$ZK5I
z)N|<A!6;ZRSD--VHP0jagzDf*5a4L&ij+k}Y7gZ9>(c)>MES25`}h0S0J@<P5`qeF
zN&mi&Itfug+IEa7j{~qjp*s4I2^=~0K@PSHXy-C@!8&dZY45XQRf^t)i-N}<C3-A>
z2G|6(NCOG)lW#Q6AjE6?yywe$Hl{x)Dt!bu!#jc_{g%CZWdWzaiEmd$^hjF3dV=i?
z@I-5YK!1e2dMUt^@#5!LC_)6}N6dpp>JO80%A46Hclp3OP#8!n>PO3YyUJ_Uo7Zgz
z43ADBx%+>nX+WsoZ5z$a@K>@#DHPa{c#|$@A1JT8K_<`|i0Zpvh0NUp?1LY^^yFT)
zGqRcP8)T~I-|Xgrx5|xuB;SN(3<}KTzo7R2d13zbQS&15950rY8vgq;Nz!*5@4SW6
zDnuqu+tuyOf#GfBp}^bwoH(ocfIcz@LQ1SaG12~&@*P;YO*ep$#(e;nGz&3cln-j2
zYc@CjKa{<7T-0m3_AQDC(t>miC<6l0UD6>Tpwi%wgOs!=Axd|5NC+a`DJY%N4N^mQ
z=fHc7`(F22_ul(?-uL}yeLf2ZotfWN=W!k9_n7;LErMlJPjs~^1yxPmEUe#GR5hbu
zS>paZ7|>P?guzkdf%tWqF6c)SngCFTvFRd@BbKUZ`)((TgZH4dMpWVeDt@NWUVvN;
zdlJ0qH-1v+^1Yj_ez$2m+2NNR?OBK1G*|$C2NVZjG=#38i5pF=qSzPUmdMDv5lvvq
z{QqT{16j|6SKsDM3jf-rz0zequslQMDl%N+T4QN9CDMTvlksapTFJ*B939~4jz~i;
z({EdWKuGX~F=;A2ymSX#U>6h>5y~6h-A{D_q<<HEHe`URTN5fU1LSE^f7L;HF<Qub
zfDrkLDZo|H4Pl{uYemW1WJsg-p8vlFC0P;X8y``|2TNgN^*RmTARIVOH?4cWl?WGA
z)ag9#iGe%|{xF)iP-JEa9wsLyj?^__)YWs30(_vK?s^+Ly?@(fB$CyF>A9Fr%!dE3
z$4G>ShH<86+U5|a@4ROsUg*!tW$W6l`>ybFcb)NvTu<)+7+Ub{Nx|iK<ImUhxSnwB
zfnpw!!NI<8jQuaS436X|PV<cXNv|pR&*s9avFa2(zO`s9%i8tiq(6!^oqt(^U!sve
zf+vCzlUw`nv9c|h?J4#9VjR-F=AA%(=XDa{!E{sN$@g+U8^d?~XmNE9(oMS@!Us-w
zgUG$5$PBc?cvX;v^RMF(MT+l+JyXnqH<N4*6AwZ%$yl5>`D;B(f9768TamM?(ummg
z@P2!P!VF%w+J-ZY#$c0SgMp<^6Azs1JjDDo1(>5?S1t;4RRIj62u?Z?j{eKNx(;tw
zYjqk}d?2>Pd{vt&#i?b|ZqAnE5OHjh$6~HirXi(T!V@u$MAA!~rhQP!`{;rk&lxZ>
zU&@;_H5wIIEajeBB)bYmUmK9%!F;wJ?5G_03M1Pzl*z(IX$zoperld>v?j#wn?~yA
z4NF=I4YhIZHf~-3A!_Y6h)os7^weW1$`0Q3t5m`CL;*a6j#QN!bs}H(R)II*diNls
zdS0_+_|042Sy%k#Wl`Iv(?eag`g67;?tDs$%F)~IJ102EYtRX84}u^A8ka9_(Vjku
zs}h1ayHf3Dyydsfq;QizMTHCqFUeoY3!CZ>w~ct7Qf+bJ4jDCq8~#@dkPkRUdg1)G
z@295rUmCAdvz2z3@H!z%&&l)p+<tf7SqysPrVh0wIQVIgO5Rvxn`E<#jO%YfFyQ2@
z*$!qIV>Ny58-rR5x!4MZUWdF)!GyR2lEAd8=eb5v$@>J;SS;}gTd5xmPu_p=?C#4A
znj2^cKSrX0NO2K|pH4?y4I>)&%=~&}VwJN*G?|;-8imx_h$kQKleGpD%H>^Sj{V>f
z2WAkCL@|vcm=L4HZy1^1s^O5G;gCp3xp-IfrE$4(=(2!S`TDW^qRn0i4(jAbp3Fw_
z*Yu`izNzPRR#D{M%brnsb_QhbYG5HK;lSe=%m;07KC{J_JJ(mX{KjoKE`Z>6iaV@b
zW>$mBCPl5ZPFC_B$7N|pV<ShDH&(Dbl_*dM>^jtzMzjOYX)6u8KJ3*13YlX}di9FS
z4Y<v!QJXp*t}u#QIU|l75<j!lpLjNFRn|KVRwnIQI>qEabgimJq%|nF$Ie)MQTXr+
zmV2n6+S*TeLLDgp40?42LO-~Jg(h~0wb~wF7~xFXa-b?8L&x;gS8^%ZU=wsppaHm`
z?U@}97yMY%@pI+fi^8E9z@@B<yhZ54rT*zSGlXwa-f#wB5my0!QtY|`99+*!Q5}ev
z!a+zm8rQ7%pq7{c9H(s%USejLT8Fu}29k>HP#;2gepd?16YNkWW66$S4Xy<{@Q^4d
zF;lp{*c1T<))z_yD-Qv!Q{d@%eom@#B2VXHG0K7uppx~D42$3{*b(Jmy8zKih@rNY
zlRQJ@w>096r8TfJJ0o_$H&8Gp;MhVa=ctc15tHs8M|>2hfWjNDGA&+%lhPq}<j&i6
z;b~N^@kytu$Dn2YdC~PbyI<wEN}SU)0K3)qHXxBw71?VU<Qgt-8QlU0=21~#&JwHk
zM2>knfxgY}7kgozo2IR(op%HI{`Ucm>fza8j`-G_D5yb8WmvSmu3dm@+i>3ae!n9_
zRm2er^ECtHzV#(O``2KTd_!qf>=nljy15kcG|vryhq2*$h{t`?nAQOl@g;Hh!9$D_
zM)6<MTJ9HhBzs;|d2U3IbC$|(+((jfMW0&4x!!twXngkW41|l-*t#uJrZ_vzdEIe7
z%h6|O4@V;-set>r60z)nbHhV9VdFg<Wgl?Z3IpV}AUDcSIC9!jaOv$q7sLT!wt4x|
z2#7Y>yy$<!Srxq7P8MwGo=ce?kgKT!#Vk@Vq{?P+wz}^-a_ADg3H?<ZdJQOMj;Nwz
z8iXU;d2ZRqGoCw7%CF13W(&)<em?ZoTl}1+@Im&>RlDb<A8ir`2`J^AF5hBm583vJ
z;+TGOEzJQ(s9>C%AXGJMEPzwI@!Mimnr``IsqAM;v#hX2u)B5kg19<lJ>C)j@)y0?
zLaz9;RTN=&*g_>n;JgpjV@E_~JOzkWKf|VkVtGA^ZKRNJbz4j6p+#cr`g|#PeaYEs
zz~}7Sd)%Ra>_~14s2J!ZAhA*(ylFcI2R<LezOUq)@s46TPW0$OHC-UscsW$;9L{e&
zS-u+uER-TQ^9yzW;JzW01D}PvqY6a)O@VoC7kczyd9a6C{4gEA24&*00rpR2F(~uX
zZulf<#lf`$1XueFQ0u`NO;;CEzZwVfwfSd`c%ovF_cz*}7sALkvR9kOp`T2g7en~Z
zE3}GD^_tGnH{O;n?J52?rCk9J^~mqNAt0QggdZ2ck990a3oWN3FQ^qAJS#ty#Y7`~
z{tsd5RWO7Ok)uRHi&96;QPfaw?dao1*S_mcdZ*cG_hs(1P7@eC5xoJ*L1J3p-R5)*
zxO{z6-9{#ila}Yc5On`daO`{2-b;IZ+OEX+SnyGFXST;N`}IEi1jw9>!P3rKWW3x1
zxzU1AERVS-vU<J!W3531%><}xlRJHF7#~p4wgF8%_WN!Vc6wFetRzLJ_k;PFyXovi
zk`jzWF-jP_Gz<yFr-zsldEJIxabO9`lEaLlv3x^%PhWOo4~or^VAW+Jnb>2p2-O8r
zGB<?jQ2JquqM;`WV5LN2`L5wp&M5l=fx3AdCFGv-0*ZC)t_Sk8DvK!R*rmg)fOow6
zYLE7sGpg0vu>EVD<mvidKT>U`ZUB3oQu1d`Q2M|%aidQ<Pqh-zc1-|f)PoUyP>C;^
z3sP!Q`gwNIkYtQ|LjL3|$3p{u_bItr-(GhbYXj*y^hX-t9*TWOXm{i^Ty^+*GNOkm
z`QL7E4;LT2RXE%T#P&=mTL}s9lI^wA?ku6ZlH4q|U7UL=CNI2H19C@O7%PfAA*s&0
zIGiydVGUQOWExebgj_^TxkYuuxA&DVp723zzXp|ZMq_n$$8jG^#`lIBrw3TmGA}vD
z7iKpaufI@og}<}NEG-$g$h22BXAH_nuzKulQWNgY4beF}M**q7jb-%#JkSAkjV!Hq
zeik(Y``euBjH~{aZ%OYtX<n9;%KC?hc|ayE+^(9%FQuTOk8EA<+KEQV<Q7)DMuhlU
zfmF?26oq%^8uWJW*fU@kCcuk#VS}{K8iSPb>xnWNR2lZ&0f(^9`}u|`w!KWpqQUV;
zh`X0nx#`?XGhYdVr6EQ?ERlv%`S|PSr!*QxMy4T*Os(UJ&4I3M9UcoKLMp;f%!Uj(
zL$Rj(DkHuvX+XtGte-C$DM?N)%{^hnUZ1G@ee`(t8nZKlJM?o%2aNn^@PI0oYfB+s
z8<hA@@-Y2nzin!yok!C`2jq6!<8x<e7s&)$#uRJ4-L_WZP_3wKv`Df2g0A?681#pC
zqd93k6WQCmiCQtSzDL>a_az6MscSS}ez)qG;OTjEh4`RZ3CbC#{d-B^itX5;d6LMI
zCF_%iXF<himp3M2MJV`;oI#bC8!QqnnjQB%l%+8}(<}b04n3;c*ybVXc$9G&^d{ns
zq4Jbwcvsg~v)2cfBt3R(QsM6T<cCQ+u^3`)VjTS1pv)7&lYB>IoNLQpL*%41&_-rg
zevUu-!BI&7$9QI3EZEg++GyUC`28(=<Bx`Z1TuHYFW~u0oS_t97}~A1cxSOvR}Fd*
z^ZDJx1!bYc7@04s<Q{pF^mWkR;GPKnyDANap-T(L(Pr`YeoX%O3$b3SR33eV4w;-3
zA-|QOB~AqT7KT5>SAk68^Ii{s2&;cId4oium;4^+mU>KNEv>HtnMt<qV)jTU&QMSF
z&J5KKH_t??Iy<Ga^=A~i{;u3)c$bw|TsbYA`zM<EZ6ZPc@bsF6Ko*hFXzc{FJo+ZX
zRhmex)ZKQM0h*OX^N6SAFI?G&vE-ezo~neJW^8p%NwwT7;PW{mZKB0q9LDb8%Mk6l
z81M|&(Bd%XRN*;2V$Jrt6||nA_9%(d*!SIr2&j-El*;{GuTe~1;bT+radZe0n*8RY
zDk=I)zMT>OHi#;j<)>Xz-6FWg9Mjlyl6#_%3@@iCZ}C>ew2YBnV;v_44w;#31GUh(
zpR#a+-*k0w%5vHcGnt@r&Mm0}8kQ>F<al$d?}2Ri1LsSmEs*MtwZX!CI!c5|y<u>K
zMs`EbB>kliVOhuR`8TNB86>HDBwX@P<txi@0xFX3Z~<$N*#ws>q!vWJYC8KE9?5h{
zyMnKRRw>V_p{+xB68~(D<Kr<}A_gDY^X*vb8)PqCU&QU^xGqm*8O*cGj%HElMuZM$
z7jEJ&UZW6w8>H^1@=4FFqpwU4dii$*>3*Q0H%L5T=W5o=mRaK!2B%SZzZ(uvJlWNF
z%t7vhg;H@*X^wlwP<wKQv7vNg+IvX{L;)X00IFoFmEcMMN70?ynkXqO{;;g32O^KR
z%%Rd(q{cbRoM_qBBG}{Mm#)IW4?{A0ez1(UD=QD7x|y$9{Frf4|6*90g!eBNfb65H
z)Ke$v4hGjqO206zr#>wZeyc-V^LjVw&Kn!0pQ3St<1SzJ`h_rGk>(N`x4>4#)e)Jm
zXP6yGr~J~byejXmeP0cSEXMA=rn4(k=}b+hsKM%5{OQ#*z}5V$N{(fR#P%I}@NOK9
z2KK-okrI4PGBtE{A~`Kb98wx^GSc)kDl*KD-fgF6q`>SK>*%Dbk*rGNzgFhA9atUf
z>p05wpu?O0*70WjF8wV;ItOw+){(9F$IS_On|u5})d6ZH_W>23kMxW8Fb?=L@6YOH
zM)DYnr;Co3aJnVVny#n~CYSP+r4SHkFU|5ub=pBF<hElua;ZYRD<<FIR=8efB7W@w
zUWqr6H5^CM!B9LCoy_0Je#s19EkA8`cy@i1g~yFEcLiY=!6!XfRR+H+k-03NSc}cb
zyj}_ZOgx$#L5BW9z5v*7O~T6JkJG)r#7<Gz^px3r4`jw)_;}hHpxDIBZC~-NsIDV4
zV!YfpUPjj|9!Ib?P~oAadPIQwJ<!%C{IJKs<7~-iR}=_54eT96(2>33t~{+K!<l!t
zX*f$6N5&PqA1;-l2Gpoq2trF`QFTOi;I5wYTOUq*3c;Y@Wl`sPJ>tp<a?qMN{^axB
znzUrV@SWXr-;xhb@n@GUT=dimw*Gbf{ef&0)5*#6mTQV$w>NKN1%vfcNYe<he}d<u
zBlF=1%jN;>SFCt>Y#7<rvwYYxe%LqRQATZi6n0?vwVt^4C%Nu4Jq-h)cJi?s(I&sz
zh_A7G-jXwf`nkQy!}arF%$PrBYBQPtO=x*d6)721OD^?F1f_)RJqwJz^Iuh$7gop<
zRy%JWToux?vyQbEAm%qk>4ALZ`?h|*425W$=HqjI@(v&;!Bw3a{gI03OHOb6nww4c
za{_>~{MSb%EtBzJzui0J*ld{LGaaCOIE1E&W9(bozRUVi%4F-b8SVXPw34ftw;EY4
z(i)Bf##cpEdJVXNeaEQjCcfa$pa_~|Y9+9>#ic})CPCt7MksgrW`y|uM3LAYuKL|8
zcEH08z0ROjJpDu_mqMXHq+3KMYcIpUwTa?f90T#yc@l%{$;}pJ%1Nw?`R6V_AIQwe
zs@T1*v$A_R^i+@a_@n0UmU4VtmX|h`0-Sy0@9t~RphQbsm+)s7<m-%y%odr}-L`{h
z^I;)B!R1+=VX4rpdRO9elZ<E}8|{bg-5A4<BVVW9uaTEx_3(XR(0cD`eVC>7@t@Vj
zLPuR}qw9*p=Hd4c)23N)Uq)h<yW!HvdBPEAp~<+}ih7(<pC8Tde8hgdD(q}z&~uq4
zh<mW~=9iMV{uocpM%Jeqc`@qX>(iIc&zO}AyWEB+lA+c@fQCb@X=(6wbYSi!KP~K1
zC=J_gU^elwvgHWR$iuFsMTpqKH>0NdtRGy#Wj11xjmjQ=0(xL3H&ACRJ>45Kzi$P<
zrIS`3vXCZ#LEgk36lCk&Z8)-v(W-eU+LFP_CUd1e<#Y6O2(9vyQoKjg54cwPbpE~3
z&(#?2P&{SfJ`NRRwv+Ub4l$!m#bvX|tFa=3FKr)re4Ua+ENXeh$0G{4h91zG!eD*8
zS$anuN1<5-X@Wp3yHiMOUmdYg=Qje^$1v~G?AGe`S^4~qa2b>U76!jlpcLK4!J}$}
zV?`=$Eu88ed7;e27sb@@Gw&^8t8a<5TT+V*xG<spJs*o!{TQJ^li>I;xy7!WoK?{t
z?0L|kxKY_Q0MHuz%jC;)-8m}ym`I1bKA<xXjHJQi=Z1Z;7FD%roQArTALZ{|Uo>9V
z?MVF$=QSgie?!zIPhJof_&%jpRNbQdZQJ9UWf6`OPdNVtmOyFG#>8GgGZ&gTb<E(+
zamFLV=!hxW25ch74A)*H=zv||cmk!W*>1w(Z0r0mj55hkUtER(HuEoikV37(V<ea$
zEm%X}+_@$X^)9F%N_D9b8o=u8?^MW4wROdYzyt(C77^&_QLXWp<}LstECew-<LLpC
zr%i+miJg5+Fj2<mDbRnI^0n*OVskJ42mSrn5(UDh6ckZm8sB60R#his5-|p<fT>;^
z$`MXTCiRHR`+%%Mh3W1J$4Vyd!!Thv-MuJ99;$PaOt6;>FIjUVGD^uzedZyFxiubz
z*QF#6rxy0;;H6N({Hnrp=rplnXkw4>?pQ|Xz@e|lk>=WZCrfPDRw^Xg`v$eY_t?kw
zYEkFl?Q!F<t#gtFf4)a234LyDokyuEDjU9OI>?l725OHpCI+y)2CM99?)Dfnm^=#O
zHD%Dmd-1v~mrJQe!=~JP;E<1FoY*PpolitaoYx-Kzyy|SCyFGlZ(}^^he6Cx#Ty(C
zBt$StOUD;bc@if8qCO4|jympT^qP5RVlpJgR7aJ{42i0L^moL!AjN^KXx?(=bC%)=
z%N~|iTA^AXo=itGYXG89vK5DSno#4NP7fzChMn+-+~&U=Pi8$EF~$y9H^hhMIlA&t
zaVIFYwo0LeeFmFWl+K!Vm4U@)u+}e)D-K1iYil0+?R6c^;^~o{fb`t``J7SDFlkCt
zDi(?Y?DlDJa47g>ONhEFllz=s6xYiG)p1R+GI4d`pMLXc%C4Py2G}$<Rumv^J*H+o
z@$>O3oHVJtma#;o--&QZZGf!ff{I|7$x?_Cg86uKd+<kWWF?$}fCI<&_<eBq>Ja=L
zr3?rOk}Y27j6%fr0#u1*yU)=N<Ob+>5Q>kXnw;lnEEZ7Oe3{}{w8t-S$v7~G*ywJ3
z$ILC|9)*cXiR626T*0m$C5E~Qg>Uq@S6J<%@%P@b<VWT5Kn0esfpD+6qVM_aj<j*5
zVc*YC&Uk{36u7ZE)JIhieNlQbAZqIV-NqHRpS05*VHARM^X<bT7YWa`NLtnSX=Px;
zB~en!9@e+0yq_S($fbiCh!$X)9K%CEBKunIQJURGaeJJA@xXKv!#E9<uU%9XJ>NwU
zNac2DX(X*UW0K-sA=p+D%pFeHh}^PGnZ7%ZxRGthzoGc*V_dvY1l_p`L62p;uD5T6
zoFEgG&e5d@Fo#^8hlk(1q_PHFay$EuVb1HQTWuc7rr=MGOMNTaO7~Z>s=Zo);IU|0
z`R?Hs!s|$6R9?G?E^L?^G_k@EriQ_rssXp165hFk-vL~s<*`w%Dr4mvuIzW(sn1$D
z{37#O>qb42x|CWAH(h}a^p2d-Z&1%8q^w6ZLDWeW?}s2lwUKv(>zV<Urn60-1IIs<
z*;I}`az*a9QN-xgtu<UDcitzo2w>o}f6EohmY--ytPJC8CsuCr=S0e_*A^TVDToc*
z_LN!BX4`T78g;*aZ2FY}rG<HHi>wJtUl^{4j)YZqW!ju0iYs;PTJIyZ<mIP4JRA`b
zazcU~+wq2-Y}gHkavltg`|j|W)2`|E;jL1R-roePbad3Y<bP?a;IP+>*#ZxYlz|kz
zqYcqa7DGJ8*1VB^6wKn-2!zqoyw-2<(<n5%N+39zfm_7qj3#wTBusTGsj(8lZDI>6
zIxUYpCE*<=CEr*DJSF{zgyHxL-)`@>I^E^c>N7#}oU`a1{BXxj-e>6p9K+*ffv)5!
zi_=rV?vEyGCu9YZrLaX+=j{I7DzIM0WpiHAVJ}bM32*k7Nb08HIJ@iTtLMC7h?+xf
z^jtM3u;aY<GFL`h>Z2?P<L`S-la-`3`oYa!YNu?xt!KGi7Y;vz>3V9ol3LZkF0^IV
zOB?-;wqB13tg+o%>7xIK5u(zU@IXne$f!%IL!&`7eZ(ewp2B7FK;Ggs)A~6tvuQ8K
z>r+emo&H_^uOW8&&I!Katsdiu7x})8LPv_6{t>*n?JpCviTi3x#e*G0;HMioK2dfq
z6Y=lkyLng{{(v{OAkVVr@k7>1Vbk~bfQ~*?=;9u6OUYwX9GHSkfC_KWR%f;EYCYG}
z86gq8w;s=EPmQaPPy1OdV53$36BR~h{1~Uot4%+~Ncv_nG*)>Q3p4juAY5SA6{70h
zNt@`~&#FH1e56h4ZHMJ_lnNy4v^<;!l8+K|fYpdQ4PP)n5^JrH1B>(QbI($7Y?$Xh
zq7&tyd|CZ#4+LhJPDehOlxNX=e0L&`Lp0q~zxS=AV~_W0L<GrI^>*Z`_cRfP?y2zY
zSRb??(;2sDAB_<M`S=7Xk3+II8@I`x-61A*&%o8~aPa|`3K$!7`D4!v?BRW5w(ER}
zaKvXlM!&a;6=|F~{iXf(b9}hV=jH5^IP=}}N``Vi^5cQo*YQGXm`YMlA@a@un@wl!
z91d$x5-f>jlI~W*yhp>1h@_G~*vAm5iCHLa#<O&ymch$$-hXl)!3j^4+ujRhH7bZY
z+<lB#Ox<klxO%SjtV%oWiw%H)MoEQFs0eRGV_`KLcb~I4B{>6EV10YlL!O=7fQSSN
z*@37xOU2b;RbA)OYGkkdMwtrHvpt2^vZ99Ht)^Oc=yyez<_Y^gdqY+Qzg894WDhpg
za8io+=H6QV_w1wud(`*?TIO}zAu*-J*jv3RSD#8a_chz;=gZh!M_#(ZuhvCsaD(>-
zz-7VO_3Hcay5}MBo93M2u<IN)B8{pk#V8+P4}uRBum54gZ{xVB9KfK2u_SHVW+Xlq
z5>1Y=!b)($ZMExQ%7nen+QBhkP7i@{--*1^5-{fU^NGmzt^bTs*Hetvv_Kg2_)}px
zD|Tj@yCUmFtHzQp4whVV7=%3Q!DWen*jIyMc)Vfx2v2cmR97&@Jo}698Tzr-&R#pf
zH<nfcfk3}<59}x0E%llo*5^=AtAbo|*3deFY-t--F8ljVe@@AHxzjJNIs2X;J<1jt
z$GJ7@p&4dLBU9(^f&$${m$qSgUh6J2KPqaYm}dtP)YPkWbNvb0Ht`57Pp}cf$WfO_
z1aN$`hnF+J!<n*CE!~|$3k~qiQ6vni98@Y_G0B+8Ye#~-l9d%?_{e&Et-U_a$=O*F
zoPzy9GpCq$^mEx$W1OyfA6`FlZnMJ&F_yD6DiJk}@Pf5712Z)z2Lt4W7FkmmJgpb8
zjE+R40jQ8qAJp=p3}_gfy<FyZr^&tb+EL%Tx4Yk;gh*BC5krEH%^CI6=saDhQAtmZ
z;jHmN`aF*-A9-1Eb_K8ko>!Cu_r$be6?Ge1B8+P}V_0zDt3bF(wj~+W=L}^#0aC#X
z-w>MtV*WQa`P;^+P8-rnddSPs0-|!m<pPv0EE>UrlT^VJ!wY7w)@g3VER`anI1lb|
zj9|jMzXoA5VR`Um0CR-M^X>%gr`ucsc@G4s9#cTykLGCA5wWcsn0<5^5PFs0vx&BJ
z&5CFvH*Cu%iYZ8nIm6wGXv}wba_@`VH-c2DYX+on#q<|9W?^DX9!{$1_QorTtL_t1
z<tA!2ai?}-&aEJkw#u-|HezNI&lwK&G3(ueB#g^)*!nRHM)AzK81M^yUNY?0h3vx<
z-<*?-+O3t<a3y`>v_FW3o-#NHJcm$}C1Ds&%@z;mO=9umO9i0pJEdBzYJWKjo;ab?
zg%)cWj1`Y}OTxr&Qr(&dw?NhR=Gli=Z&~<9^|Hoy9C^EwDys-C5jQ^5m#HJeW|F}c
zXK%bL6+i3ws4L#AKvUC@ofaiKt8O&GgyscJk#B{c8aHPs)zO&#0F`A^AP)~S8l#V|
zILSLiU^bo0jJN`JV$*m}4QNmmrp;vtf8Hr2PXF~C(>INC<+NI@!l2yOv&-Bsz6*i=
z<ua7L!d6{7`qs|3Bp>%%d8Z}YEqpRFzPoZ_f)BOJ-^!;%vP56X-gOcwV)MywEywrC
z{>PAq%duvEzWv4Uf}sF~s9`UIJ-GZ5%?scr4%Gn>mP{fl_R99@E`02(JGo<@l!EJx
z(}k$;iCbkRNRMc%sb9MCQg}C#X-64Zy?E}-ytxhv!(WU^E9uwqvp|v!5K`v7#C1q$
zn|qCF-2N<d^N}2)OGowYM>6y?s2NV*c$_#d%D=gFX0Km!g<{9|&634MN3itE(`%-`
zF0D<q(z54JDrDLy#W$pD_&gCRr3~lKGg`O#wV@R+6%EoacCa&_2vZ_j=z%B&Zgj@u
z+y!?02--w$eKgc646jAhmS&g`Pqtby5xtrbA`CKxF}85=(V}bswp?Zs&7qNJT??Ok
z>2A^8KeD`aI>sq_=fM)ic(604PPR=7oF&C+1KokLD=WC=!C2h6vzBzJB?CUemDY74
zxCAQ8K9C(8kv?vJC848m0#3e~cZ2kht26yM_X`MHx_V0em)^k)ee+@byjhwG2XGx6
zULro+GHcJ=rN@cf%yl5CtyXPNyD!Jqyx_)1F7&1ntMyQ3qe)VM*8k(M{>qT4aB3-J
zfOpV;$j_B1dPClWpll$y`zHi}8~7}oACv+%sGcdFkc@~%SgvNsJ=;`YJTzbp(T~vj
z_;1h1JIpsr4>aFDvARp@gyrKx-M>sa1W_%XYqFTVG`f3*879M~QYx9+#g1ulI!<c6
z>(E5~k@}N)xe1}T&4I6i=vcJG`YW{uX(}OphazAcPJsBKNc^U}IWy41kim5wK3rXx
z^-I!4Rr(<bb1^GcZ7lDpJ?>4d&o$|owwrW1PtKXnRlJ3_t1}x(9z5utC$TdBmAqCo
z-|v&`MkZ9!0223V2UVzl^nvj`YuY!0rpKE(zl*>(BEilZ4jD6eUyu_TNy5kTHA!i(
zsr(*V(A4Ql7RfBnC5L^}D~IK8d4-SkFGmkoAK+M%z0T?5=!tIB7-1=vz6F)8<f3Xg
zDzJ5R{b6LTE4w!c)$Dx+rSr^95j(K>Atp0}fxRx~nd~b!M0<2gu=4<1QST<1SOrG2
zJ#Q<=@72_(DY!CDWnO;~xIYf+3HKB!BCAw4h0|e8p0dHvTxc}zaV{=wotBbGFCuGa
zzKr$v^HL0Z3+*(e=?)eSnAf)s5r7H)eGNDV6Fx+)pKYR4<WG?%jmuTJ^`KGNE9Rw6
zE(<Q~on3SFhvU|rdU6t1nOb7}RKV(}Eb=>)d!45G{<P5MEnlvt_k9LS_27c0%a%*L
zk~tUZiK4yNWO`EmA|F^YoX?+ES&IVTi^cQa>?%-K!e-iWvHMJow_%TlDmg0OpeCgF
zO&~Kx7o$Xc@j`o<$1g?8KjjY`p&{-jhLA}W|7<1uw8v@xHFUsE>V>te=V>$T<urQJ
zT?*6wXL8fQ4rFB<%RiMPWNkLdx~ZIo)qb;WcV8Vo83Nbrdd$L3{3}ksr&Nu3*B*R#
zjiRTWYhipi7Vzz?OS55N<CY25OnFw%1f8s<^9!P~bR{aV&EUi$_lc9jJYe|equxL{
z`)=j0VTQ!nigmrE*>}XPcsiwxSzcZ=zyn^ZYf+#Jjh}i`VVjVi4_`F(OX#}ftuQx`
zph1OL=tyLRBRRF+eBt2Y6W{4GznQ-_v~M}gY+@$%*j-MS(M~eM)<~XZ(QTBs#y3gK
zi8X%MUUQ|p2bM&Ed5+?v8upn{al|UmaLOoMC=Q5CQfPp`(5YEozbe<yX{i6@Ou>|&
z&l(Eh5VdC?f0f<1b}isobA5GjLD{AiCY{O3&%`13DkSV&oWe|NeFTn}Sx^cCyUU>Q
zYskM0n`Ac`0)L<k5LkraP_y2|5p7oUfn*{b_^PhsfDF>iNeo7?FCfv5@kpAH1PwBa
z%JrNW&ecHX3Sr(PxSPNsF?yGj783d@j3kjhM9j^YF!45rO4ny($nCTG8US%Yj^G8c
zmimv4x0#_?%!et6jqMjdQP$~r`umBEv6nBJu+H1~&2OG>UKx{Z-3foI<!Ia7Vnm+A
z*me6R)mZ6=+d0jUXTD2cY(O08jW8K)*W$U>n_MO$yNh9^X7t{%=p?27pE}jGp#9}v
z_cZY4vFriC=k|-QXkG$LnuX6y_riEIY{xr3^4wH#0M@fzY1KiAnuW9tn*P`y^mlb!
z_I>N^sqZ_5`4DwvtfxVVgw`KO3rs(G+fB)rCXek2prQw^9-s^ray}2~Ev-}ve0MuE
z8v3?BmiSI_R>2bjNvRu?>`ul^P9E@~Vg6Ru7GhVTNsX}P+Xe+yI#ovNd?xgaM<}~C
zl-e%k6da%t-K1>v!_17=;SF6%L&t=^qzr)Mk$4rFtC`#N$yw?m8=2|j-Kq}|Pf<d}
zU<EqgAnp(TSf2pVtQ=_famuu~8li<jD9f(~{QAdM>%<P6)R6~D5KA5r%wHW^JXx>B
zF=^mJm%twrWnKIeV*=3`HR3uJb^8(D2NFX!yT;BNv4?3Ltmh7h<g5FgS6Z%eXM~Rz
z(Yo$$N@}Ybk(<9#71~GonVFFxVzOl}7zTUBz20>!CyruwWp@D3fHx(JrBVA!8P_~Z
zQicoO$8Gll1>Jd<e#A=xsEe6lzJ<l0Yy@GjM)k|cat8q)%MnR!S^W&;CeV5e7$c!=
z`mWqKrirRi-zkAq-vRmT-S|v{3CZrD$2WP095LX~C%VFW#g+PjxUE|AI^53cL8bC3
zLrRAT-hu^1{7?yD0F^homg5|@C%jB5wy<{n+XR3)jh*1{W89I$6Fq!ZkH-V%e8foB
z!{Oq%F!Y-=i>dVrG5DB9z=INspSrIP3VL_99^?<gOp4b*%XPmcem2JAntsfy^CJIb
zDzPv864))!hv($<n`n()3L_UE%|7zw>#46yQ)`-)|E>NkHnkR@IFq$5YMnk-WMx$R
z;-K#6k9cq?(Nk!w-g!mF8VlIgirzwn>e7Y2izfStie@D9vJ-?*Kdudm!Om&wo5CjI
z9o5FIS+R9GvKrY+<SYc`iN9)%1LKdsjJJn|Op^UQ&6!ZseCgZ2raA3Vc~5jmT=(=?
zi5w-@#;^R79nh{s?)OVO$H+aBr-Z%rCo|sLLBr52^4NF0k*<~1lTtr0J-1+%5b$mP
zm!p^A<Bg=tuc>Ac0j(Y+>U0%Px|%{I*rNFONDq8j?!>beW8&t9g*aorJN<OoocmTK
z_}~{UOmbbl4BY)YHgMW?Bmt{R`hk8kQ=KH4J$+7sUAqyM1q>GA2$K`zn|>ozw8TP-
zpToS8X!%$bZZz}?0Nnas8n~Kze%S!_i=~711!nY5mrZJ7`Pn-qT6`gzNC~BQK(eme
z^=<Ka$-^67(?9;z^k>;u5xRh{j68fE)m#)+uj)p1EM2+q-WrjPVs$XV#-r0I3U~W$
zlQ}eR4ajYwN`Rs@wb`0&z%WZljh34Z*T`jvX1D3TtMHwfn>A5tuwO(KIIs*oZ*^i{
zdn<^p=tN`;=>cnt4b(a@JHtHb(VoUIOW_V0zyFi0G`^os?d61+D&rEx^}w>#{W-YO
zoUj73j$XLyc4T29Xm!Q%^zL1gEM#NY>A+tn8q=3x-Q@4IofP8{7coXSlc@~NavnW0
zb$x>rM(12d+Yj28Sl(3;e*+wCrr`_~sdE?~;g4<wgr8^6$BEh<0O24;7ECOjSA~mT
z`Z86stxI2FyM~@CP|zLP6__W~pv+E&3gRkBIi8xVK%_jxA<uK|=?1g6eL2{-K-J$j
z9p<3exQBCr@5GO@A6WiI!0QU4WP5c2+{`(h+wZc!X)>EA_l$pI(ZDq{GHH~>!r~8e
z9}srRu^!3_k$m0Rolvjj6&>LsHO0=Ash=%l!G*e(B_l@<#Pkf-PFR#gP?`w5l9~0b
zhXwy!rnqxA{y>WCvWc`|PT?(s#3@5T7B#(Sl728D->jR1xFN(n3+<kahte-J9)I#4
zb1l!W`|V_319R&HY0J`+AJI}N8W(o|6!d3&bqj`rd)hPgGi(oXOm+F~wGm%*K;|}v
z*+2H-mvOu8XbfGscU|ME$?M8!#`;jM^psjBdSCA$%J=oS2S003L|AbQY!meuCvG0J
zLA?Eq%3KpOHLTBSp#mEQknTjh7Q#_k$eexn8`gwjd(tKxfzk$J#sj}0^0XJcoLej}
zF5t#WYAzzryfy!naf8qKab?crVVYI=*o(=)3dZp&%*#oacUsZc9&0paT-X?dYtDfH
zZ4D^HJ{6d6^a&O6I&!c0jKV|5|I@>;`pc(&0YUKsN)evSdM>1me&Sh*xhfr`ZLy6w
ztdAP{t%6aZQVvKFmk95#OQVLK?oc8u=oq1Zw#f07&xd&gx+|1*LhFs>$K9P$yPcPY
z?nl4)J^WwNy8uU};{+Dg&rVsXxRi#Dj{$#}weewM=U=0U&Qv-CR4~m|{OMFM7T3#H
z6wbwWz3>GlFzreO=S)gkv3~Xj$UcE4Wvi%&3u}9??Zpk!V|0%@KD&c*J3=X9gd{p!
zT01eSqlm7kg(sMtCCAt5<6`^>aHq-5ljptBm%UHtyxrE`uprc+E$UEdDmhmlC&_Z8
zAY{?83FCbiE|EuOayhFBY-&t&x+CVfO1aXRP<tKR?QpLvz|k-a@KQQueTa;JDxXNy
zYo)?yCa`6R;E9XM`~2MyLLdZbID4n^J#r{VrGv-k$`m#{me~~M_0T%oa0Vxt@0?t>
z$b^2wgm6%e)}*;Tf*o?>!VUXuz?fNQ*sX?7@zrfrt`cWd542whMr~|BWl_~Shv|%L
zC02S*=SV9>y{CG6K-D*kL9u;Ud&icWmh$hU(Y)v-iCQdGl^3;ONoucBL--P?Bh$j&
zQo?(_>$S6$%E>1**v^TLK4Xu-EmfH0XIl4#2T+sQYG@P5hUOFF<~F1RnsLifei4sT
zz84k{P9wD<+88k*8F^?;rR_JJJ-f`mWgtKITh-vWy{is|zU!pd(R>$Pa@hD9L0r(D
z#MSPN^s6PuaQ$LI?u~_{JmR%gAI&2!Wi+#j>C>G#A)6?&<K&C6d~;NBs35N7IbE4M
z1yW}P8FVg%6uM6wN^{q|$>!Q$to!rTfgFof%P|L@?s68$F{+h;A@GLr!{x^r^L+&+
zZ4?y?**h^IV&V<2sCwRh>+t6jfqFBv3zU2#RF8<7!k|KmID|@IB>2Jf;`k+>9XR-+
z1lQ2nd|nRdbE(;$^B=A&DJ#wsrlb`r`&p{9DCr44UO=2mQ0v~<EZ_`+HJJ|`{&it?
z63nZy^r?bq0RuWHp_d<|2fQUJc%x?(kO8Z0@A0)Uvp^1Z#p>|DWP<G(ENJqxL)Xvx
zpjQ55+St;voM|}zse$w4{z6dNY)AUE#4qYCNjyFw!SH1j(WBvpZ41g$&MoiH2t@yr
zT-{Lz?<Y%Is=1{$>J-mD;)7%QM<(n~Uv;6<QD2R5ne|q>?(|fLRzmN5+fR#Z4DzLi
z-hoqZ0Yd`wg>i9sbG?Uh*pay4^vih1$s6ueS_er-uXuA!vVBR#5}LbSG<h^Ai@d;D
z%VhsjWYnYbenoq%7_Ze$0oO7re=5XVfz4$p7RP136ME5*AXb|@Q1S5%b)AJ)=33Eq
zJbkwNm)Eq`!OZF%SGBOL!l~iRS_^pI_Q+OqlOvtsL|7dFy-041dRD2?lmujdFDYuJ
z!E&HK9vU|`s%O0O*7#G=rl;2WizW1==6Mhg@QnK}iRd3)R>AMl`|0>s*E8+#KbrRc
z<vaP)&GxDX@+@n<=rtpPF17JDU(r1OAv?M+ipXw*R5XNen}Vh#4>G4d+j6|F(7;9`
zmP>hxfh(w=R6XDCUEKlOX-!^+Tp)3ZA&T|dj=mdSOpaYD<4+%dh>%KZsP`d8RERk&
zN>LMKO8CBAf6ZzkMTx)W-+Kqf!|TL*Nr<mQUK8>mueKXUhtkw|GOmswv@6;5{e=AS
zR!89>QdkK*so>Fq#`eVRxNy<e7N<{94}{=md>ru0I;ziaQP{^nyx4YW+}L0#6=$#0
z=<LaeoO>--=t;VeJr=8jU@Vu)d^A}z(8cZ-a+h~A(txK8p9>xgZM=kZ=e^FNCbo&$
z$z9scPO@1)fMlIL8aSzzsO&^@_RPFhrTF1Mi|>z|_xDEdeECnUrJymxL%hV`qdz0Y
z8tg!Cd5UfKWc1AxW|g8325uSrlIWf@d}e(I$Q%V`gEwh8(Ki)m2|rEJ|BvoUuK{Ur
z8p{l5X#K+w|MxTNM@Qq!K}BXYCOp<rMZPE)-_vZZ2S|WaB|3_!cmBlQXq;$~)ML$(
zUpGiAJ%*qr+#Pqr1BX@8WGswD9`h5N;_^-};kMw{czurUM-RZV_k|DduRX@TweSJX
z*(q&oFBaHzB9wH{!mJbJ-hk4pNMmo=R8P{V%d1gnEZt(2393;If)j5dlzA4+GqlB%
z!1XM5;Tj8f7Ta~ZFV~pw@>ZU&T8;{V9BZuUTHpj3JGMjkcO&O!T>ew8c?v;*oYf|?
zx0GYlAu9t0)q-O8C+lzLeMn0cq8Lr>HzY6wGdtiSwGo#fl(*8JBCyHQD957TJVj{W
zK?A)wh0AHaM%CUOEI%MM(0>v|_d;n>!^#njLe8*r$YlUyMe085Q^Q9Ss;8V<&s-p{
z9gtfnLgNO%_v0%%O+By%rGD_LXQjXN(tMPl?fo78<zz1~yF8<hT6kYVT#@zxf3%4`
zr36g-&BO2T^sQ5tbF}OZ+lQBImE%U_h~@~6mK62vk3n$V6J+J|=nxaOy_70rmEBvB
zuXprR#s5`--^_Ww#c&j#5wMx#;p^$t1j_IVyw{#W>ekV3rEm^iQp8wkF}xdr235+T
z*K59RuDe7SrGmJBKs{ixRq`x(Rd=m#iq)7+&V4j@iqn;^k?0ms#FT^WEA{?ejQ`br
zM1oeM$QhN#Sq5n1oKv8}LE8Q<;s1wY;?E0lSZ;>z(Luf*XQBjg^2w)%s)e%!(T+TX
zsHQn`=?9ok!ddCoI*$TSCHr@0h7K6;-=*Q}7IN%Zd0U`cee4|3OQ}}4Wx|Dt?VV|{
zz^uuC9FC23XCEgq7+iOiZAQq^>@C_2Hl^t;Zg<kzBjJo=p{4{k_I;JPE+sKveYym2
z<aO_#<1`{D)~RaN4_w60N7?JkB70F)J0Sl-Y!;p`NlAG8dbk(u4%Hl%-mBQLWA7y5
zvscacE+L;UK?M8Q0~5fe53E{A(mR<d>AQ~SwS1{rqvmX1kw4~kFBaMu#M9S0jMd5v
z=)tLn?pau#;ZN^#uTzgpnE-|{fFma*wxs6h5w33>`%>DhPKi1tR-$<wp`zfv<^Tc0
z))_RSAD_0~y!W{+)A|6_Lkws%9u`bE)xM@mDPIYDpX6{lCDp3b#aJeAcf6XDCHKY<
z3>39D1AWtdfhK0F4V||BSRDdvRkumBenb%;>Zh6R#|jvmdOl)h3Ij3At&2y%<tQLk
zN3CPJ6!^}z3bZ7e09$sxEk8ooR=u;wt-9V5#FZcJ;n63cDA&JdxVl*vFGSYm2WYeO
zI}VUIroO>*VPX*;mJgID?N5M7eDCaoiOQ)mPO19dTtG5qx6akKP-gXtA}|Ealp-XL
z+r0;5XGUv9fbtzL7VP-x&Z6WNwEt#E{fLYEFBBuDzR=b79Hhc0W-B1AOFJ1;5AR4M
zaE=bCil=<v(BJ7oWve?yUXp86MM2wcQpKULH+m#_rd|KBW1uKjPd>;<s!;X;vy2BM
zf$06^a@g(8j1U&7U;piuWEkjl=XN7>#^V`{fPwv8oofd))p)itBZ%eQ33iqf%xP#D
z5kfqdpn!y>2vl$1-I7U{FxFS22;7wCp6uOUHm1FZZ!~k>Yb8B_u~iJK8ypZ(tkIPb
zjDjjxR}oGx!nYDCw9eG1?RTM&v4=xdjThsxz^Doi!FuyY5obJm22Kt%_Ou879DM}D
z1F=~c8T6A<lKe7b%sCzH6Xt1KixLh`;U1s^)*Ljd#I<Mm3k2w5J`GoIay-8H4tjO{
z)J|tPo#-FNieeE30j7sU^r~N)M<#AGxqmu5C=(dabEm?y=wJ(JC%|&w(DS|$=zK2~
zn^9~_3fWD+cp*Z{s`IddDU`PZKa?O->Jy0)n?|Y(7q{aq;v|@y`@G9Dgf`&>6nyt(
z30W>Qj3kh<k2AOMUGaZ@tnGMF33oyHoz@i$WxU0?;v~&Tyu6o{%JCy<<jGB!fug?o
z$f31y73*h1n&4tEw-9<2s>j~maD=_*(X4v6zPxBSFNDniFb7^lK3l~wx_3mjFO;Y<
zaLga8Ji;l{bsUr7-ty=C5f84awu-Mpc;?wF@fo(tneq0CJ9GL&OBO(52%#B&=MG;t
z1D_Y@9`dF=;r6(myCc53A_MQoZQn1kLm+|W!A<Mf29p@DMPi%c|Hsk<oH;+LrJwL5
z2ng1UMe$Fd+y6mid`EfT|Gn{{tY6I^Zah6fbsa>7OgAAgVZQ<@)0=EvpALAeNFT(5
zz_#{e_>e=!Str(BUqR#5d8pkUs!7m<sOLS=mXHl_F|FPG$P*s>XbKCIM&9GYOKGH#
z3_-|Cfe9-|6di?V%V!&G7IvlctKP*c-lr>|wVV=4c_r#`K#~#BJOjE)Rd{X-4Ln=$
zmR>zLygF%|RZie)^-DO>IBo!Sw-BWKPn$5n+(CB$YpaaN+K}1>;ZuKiKHd0aw;5k}
zVso?J6Tv`hCn>pE?<a1w522Wu)_0j%Dk0ipgB7b3#yCnA+hufSb&I*BKlabg?f-Vj
zKd&JR8k2~uUy)}k^Vz0^?<VX0B-g1rPU3s0X@ierh?jZxZn6(ztyUf&2<biN<96LQ
zfsVoIRNWScR4{YC7+9!g5EdTRd3@fA3guH$N|8@Ujju(5-m!O^&4+ZF(o=@*ivBQz
ztFI!gkqOl8xSk&#a;e~P9^<09)(yx~3PPUVuMLVoTEie&V(i?_JlN#@SbNi0XGJ{h
zCLFS6BsU+#0=`LA)P`>ge1J`#;T~J4O8G?W?EzEO#$wR3>@wlQZeNidWm)eow)Fb6
z!A8onq*0@L2HgGps=}W14L5ysG-K*YU|r;)N9RYF4DWC{6D#PM<(Sl@)+irnFq)L>
ziKqA348vk@n!($M(M#Aoo^dhsn>NQ0zQ>r2KXe#ytb78pLDwwjiO7|%CB!2K%e85k
z2buj~(`PGbI~PQPJP4H90QF;0N7pBlRhoH>)6a_QMFqMXyU~pP-M9ZA<Ly63ANW5G
zPJ&ks9w3nX4M;67?c<>QoB^{oGVw@8ME08cONwv;`MSgZ-c=h`a4fFWtLFrtXSP9%
zeQ0xHc&O~;lW3n|EH9cSXL)I*6cYb&g2<)U)~eNh$)YcN0lsKmVPqx6C3Ol^O6{-t
zB6FtNh&v(FdpF2c{V|@>vYWh;|H`0nO1Xyvd{*GCWdDI8JQjOe|920>f1fb_`~W;M
zPYM;C;K`AB-`xmkuXg=r4%dc~FwC!H(a?Avl|uR$MKP3vFLoLmg@H#@4#G?F1^mU&
zWp1eB>U8!C1es~`{0{L2%`*Das5<~&r;NQ^woi=xAl-T@s%zkfOhbEy#KD7>*ML8Q
zYRQ(3#J@YxMvO*ge7F(-O<lEVL^`Lv<Ka|qU-~vfIJC<z3oA203A;=vUWVGu-x<7r
zfdl@!^!wj%95`MBGIPLj_%`k(b77$<ilxA>M)IBwo$itv<T@Hy#=e_Ib<%?vtn{K|
zki6ohCVL~24Dp(m0Pb<QNLz>ErhEr!J$N^#!B&x%CU>ZC3F`NpBtGjzT>!S#3AzAX
z_j{9apyy4_&#b=WHPh2w6lP}{+|dcWEE`a)wN@Iqo6kV83Oez(tIdRXQ0=DEdJet!
zb^f?3IM97>qV6wcIJM~iuOH=qem*i!Or;c|5q(hcrUa))IH1rLV7SHCfi{1XVNz(e
zZ6iz^xu&`ut7~ItnP1|YWS;8UX359)Z>9hk8j`3D+WSK#{Qu{_9uLQS!^WsnI{cY!
z1G(ebsB0SYVj3L7ohvF)y^DX3t|u~}S3#4I$-(R&Km5N>oqzuNKVBKkh-x!dK4i1`
z+X}~ljzPBzT3UPS;@yuY?o^leeInpsy#q0s#RU)lReiZycl+unL(t5BJ5>CW!t!sV
zCfN?^%czelUh4bn<zz<aS1X=Z?WWqx8kp%@Y@k`6AjjD;(iu2C-v^_51qT#<2n3E4
zQ~hqb+%UXeKh-!Gt!fzQ1;^ZTq3zW4&7q!NfK3f;46stKP+Py3CA?+$_31hPJ+hpx
zcKDdjUS!xn+rYxVm;?Xo+xqtp)fvF(ANJRI`}bUZhJ}Vnu92hS)F*yj4>ijNR3Dr3
z!?l=8l?QU5*vf%QA_3koL4LyKVUe#XOEDwX;eI!}%j{_zzla2O<S3Yt8zfq<ydJ=a
zFkyTaQUv%@9CxT4!{49I`w2@(sZ4>M&uNgbK2wt9FRorbFbkUK@iHbu_<ygW&O5?Q
z$-pWw(}c~w8LMT4RR!*@YbtrO`Fig-5lb!<hwCRT8`F=@WQZ`hz^51_xYz<`H~&8j
zv^I9slWsJ%6y*MEp8Hau@im40uh@@gIj%hS_{a1dT$^Yqe)hCLUTL+>k7e(GZDuUx
z?LvSJ$ivtMO_+|L{U8I?HyI^!21tK(Q7S^YQcMr#B(lC$ec!T1t-cP!NyEzkrSt}D
z$PUnPpyU5Sp$xY{Ui9j?atm#V^<kLjW#$wx=806ogZ06I$X)au7z?6j_*`?~5WGbK
zvIAbuTSwxbw}5JMa;yH}m(5_7LP2)X>pA6pHQvpDxoxn3+k@NnswwavsH8Jn*x%3d
z|1k7{#O4n!SY-*Ug8p%k4kXk9;13bbW=zlTr}_2!UEAQf>^_M3U1X+l2d%jjPbU*(
z=VCwJ_0Adg7MJ%%^*6sq9*xZa@3({M&G&%tqMxWR4Nl+=odJHPgbfENJC14R!#Rv1
z;Pe*aag=B1glg-YQIn%y0_alJJB8_90yD9rZ5*kLXYn9VH3?wf>-M`?W{UOZe?Opw
z;1MDDL9y>@U;OhQQuW`9!9Qm6pC5+b1B>4m>7k~-MGY@mFbRtYt(H3p$XqbXWll9d
zZ`Z8Q{Au&O*s7}I9o3R5c%N#W!$6x(c(w1(^M2MON*Q3s(=HLv-kW4HhA56(qy5`>
zpow6%SNp$PAI{wd1t&mpRWm7fd;iDk^xxmmpQED>$|;p6!pi&Cb9rL=2Lx1$ay0)I
z)g(0<4$VWWDYOSj6c?aG%FU0&tdn@Iht9rk+8HOr+vf1@b4eX195cMyC%yh<JNQY|
zW|;aI{&FFKA2P4n%xIg4k6YGZE5<sQ?s4Wz+iNI(eF`af-E_ImpX&XS|7wyW(KGe6
zzecy#7KSKquXiTQV$#(pP6(d@YJt6YQ?p?&XO=0-!}V@+lWNpHU6{_bTg!gY-PlBf
z%W&D>XU*Ywo0z&AI`AFp`<FwPOc(-c!Rm2R5{F{nTM9havlS`!-_B`n{qW-WR?c~8
zkq{o*&wh1gnFQC2*>wT(N^OesWbX6kw1ux{bhb`!ez?OE?_K$)mEL%I#i5feQbPH>
zv7E;5DlgVN-jA^{Kx3%rO2I{Z;`8=lu2Y1Id@|go9$-9=96%P1_l$1<090T9ljX5o
z35YVOCZnxipdap%aGrJCF??RcgO5mfmryIxbZANa(#?W}O!?Mq9@%9CX;a&rVKbAW
zXS}bz^F(VDvU;+r14NkyEpVeQlKxA7)FS+R&AP7X$NN3Da9-|%(3hjG_0iq9xGyN3
zvQ6x!&`P1u7|n%XCDp%&Vl#@u<eN#wdjH=K*&3a$4dp}lJqxlBHu(D*k1L8?|G?5w
zdd?qjKZpeU3Vnhg6-3>Y{o*Zo`nV=*3j9jl;ElDpP~q*>acdBHZ36%h`#bF44<m&E
zw?R~G@w&?8asTmiC+b;*y0pdvWg?B#^wTF;%2+XP%|3=}eN1F}lywi2jNbR74kLuF
zP}=(~`R77Y`A#VU^n3EpN?LA~oqzgT-(#-5Y_HtdSLhxu?DDI|{8!JJ1N%wYSgp(Q
z=0utKPM=3lPwLLq7I?Sxq5uV`0<;3FN|S>gb1`nPzS%wnJtSpBPY+W8!eI?KuvGv&
z$;qguHAi&IB2%DSh42!U{EOACgrYd46atJA&=mNW9^wg}I2FCuf7iyH%J~wr=|__b
zSSTFVb91r*?@0ub`s%C=056yv-ZoSI4hLBuizN$j*pI8}P*>s)6@VtPQ8kEdc-lnO
z4tAR)F#!2kaT8fr3+u8to0&z<l-PyPC&2t1Q4mxHJ1RB0y8_$Qi*3u_UCCRpSYe9f
z*F)k+4THo(QjdAkK>)Um_1HP_U`Lq{)=-R!cdaZ210oe;wy}>zK<z4P2LP&^uHnf1
zvR{gzYeT9q38z-$PI0L-uv=vNiEhHQS9yiju3dbY!nC#NkLkY)#NI{*HJ%1Jc0P<l
z%(7PeVStQPfK$a*O<@73FR@_lfYtH4R*&f@yi$55i_L9AcUCdX4Z(20HOhK+nnC)V
ziBPUAWEmT87k?QgU)$3FxIyh<*!5}H8bE;j+O)VRXG538f7ky>r)adQh&dt>Zy$BA
zSXnDeLs}VOX`C}l>(p(QKHWA1AgcNr)-~Nh4wYTGq+IN9rX2usURncAho|_!rqsRW
zdO>@=N4o~f_8#Qzt<6zgTjgs&(1@V{WP+83<z8D4hi#YI=oQwFjSXA%wrMptx`M^c
ztTT_gM|MmxksV`TL)XWTJjq=_-Mo;V;m;31X&)j{N7H-m_$M^y>)DwFRDg1#F?%O*
zMfIRgRZQ`+++qL!Y46Pgp?ur^;gX7ov1Ex1N>XFZnlZ8^*+Lky3rP~PjJ0Gd%VcSW
zvS%s8i0m2r$odW0HQBf9J7c`(^!?uV@4lbkeZTMhJpa7^JpI)&xvn{{^Ej^KJdV$z
zj7I@=mMKp%rKWx&D#$p(G9ZmL6LD>?{HXZ$A?TT*IF~`yA(%4}VM(&7f%4@}qMkau
z$qLR=1pBd@Na`To<6BcaTvyDYD2=q+6$2D&7Dd@habI%c=Wc|9xU3(LV$c9@B;6&=
zp*6{_YlEH&2N!sc(%!sI)fRQ6t2lfeC9}JmdbqpsM;Jnqm};Ddd^&?n=snkUHt*TV
z<#xE@_jAh*n%Wk*VTH>?a0;IT0w@oBS`AWSDS2lvy<#cu<eI*h|C%0};7>b(r5@T7
z?7sSld#*;y3F{=Y&$Bl?EpLI{G14Ef|DoZVN}dPU&3j7(P8svJodI6X-CL&2ubxt0
zz;V-F+PNE@W%K5*+cSV%Zwb5p+p?SOY;wTf+%Bw<3Mo520fbKu=7RsATV=L+R;iK}
zOfv^kMay_z<v`S9KL{QWDA$C-kC1oJhq4FXpa**7DDPQ2QD2jqX38qvD`Kwy2slk5
zsSVc^R?q8Cypx9xXwdNsHr}A<GeqZ*R94CWpvQi!>+X*=3`|v7Y^R){AVRkdkzN}+
z$Hy1gp=4>X1T=#88r`D9UBemmIN6SK&Z^<U_Q4XyBW(&%2tqUv{@wr*SwW+)HMz-G
zD$GG5-N%Lrh^?Zz{(!~)Pp=rJmw{pqi*j4~xeV$7KHhr1q@C@+!)=(PcZ&*-#P=!k
zuP!o-PJt;#0dShMx_tVIySKqi09L-gRfbPKnKl-&<9O2x&~xm^W-0slA?N9Dmo`}~
zXq-77=Hx>>_*GS(zPSF%r-xPjv>s3KC-_>+c@}=A?t^f3IDfHVRIe~hijV4^1~S+J
zM2LHVZHY|z;TV)WMwLmQ_66i>ZJI$y8HADlPV2L4zMxcaaD{oGJ>?n0TB1E^d{rIe
z4{UkX6GqrR9Gj$DZ9NILdmmcEfB)=S<7bm0x140p)_#@hh}#LC+K0io=&h|bJgF_Q
zALs!m_Q*=2U4!tTpu0wQ8+tISjuru(v}EUNLki0FL+t)k2VA>pifWB5gj-u*i6xKN
z>MT<|5vr+j&Oa`Oy!{;2H<5`CmWlRkQtzHVh_<DPWR(9U=6MqA!6Hy%Uc{_t6{abs
zrMfK8mAmlU&!iwk`$0fz$Jecc@@piU3EM;bqrmYne23|9wTMP<k<MZz$Zd%C?$S1m
z#~hkTd(Cwl2x;r$_ww*>dY+~I1oe1((p<lHf#VO~Tgt%y6Pg7SgwE?C1|Bz@UiXi8
z;AlOi6W<TYy4R`rd1x?%#tnLD{J3fm(6rKc;aTUI;^9^oLgM=!b<D_6Y7X_CIK|z#
zhod`|<SENf_k*{+f#V2$?^b6GeX;ocSZ|#%PCN~Xqsz)hpi2VCg5HB%gs>6T`E7L~
zH9N)=(r%gifNC>2b<qBWEyms3=w5Xijm-YEEk+WCb+@`M&$C_q#jLx+tfx-i_k+55
z_f5PwBSDrYjtkXQ7lR;k$ONoBhr-dC1+Qf}b*;E_dJUFTfc7ZKv|`KOKPG+Oyt(B)
zw77Q|0sG~(oaKP`6^$_2vY+#slb^R$E;j3Tr^M*SMaa1X9jf2a%lquiEnaFNt!D)`
z*UMir&KT18vQ1u4RW+o?8Cpj@dKP$~j$F+11m;PV{TT-wV!7iDJN$<feNn^efjidU
z3s1!uBLYC>@1RJUXz)FAENM`YT0-t);N}9EJ^IKIc#wvw(jDW#&zDj-Y71<P?&~<U
zQ|H&;>IxEDJ(@NI;4|*k)&kC~4z$!^koh}&)gS8Dj8=J^Rqo#oIaB0ER#)8n7IRph
z3?beTFE<OhUBJJ*x>|Ee?+ybGW;OR1gKXM0fV3qfBi;uhRoX_BCB+P@Gtm-(^9HG=
zA>j|c$yqCi7c($5XV|Ksf`9VmBvx`Ncg_P16bn|VDM&`~<zQjM@G7=*g}(HLJm$zy
zUhx?|i=jMR<?A>VbA~$8Eyg?-!*Il>KpMqdh^P^t#7DuZm(L0=H(FP~->bxHj~;;F
z_woR`V@4Eu)R~}`m&A29Yw@CtWJz-x%{-jM>vh(?`8Q$e`28&@zOx0&8wMc*MW?K1
zFIjm0WQXqOa$G1-PDtqabSD`JQ3`_wZhHk{Cm89T8~WXA547f#+z&a^^}*(4T_gBP
zdw0$RrKydnR3k%p^G@YOMMP__(X^!-BEyqd@6{F6Ec(qN+?B?CNVDBHhDFwsX5cwK
zU@|h|dH>^HM=ax#`$nSkcFX5CJ3~$OU)BAnTX)l3-ZAD^K~jw2?ytHar0<NUE4!{T
z_S38`zCRRU|7ns1XAC}4<SjbTEff5`XGAarg3=>u1>Qh-Z=H&$j!4nCb<bMQ>zjW%
zxBHd9!lZ-R=95;Ff9!T&0uI@9uW}yfW~!g*Bdj0D7@b1C_O4fU3kbD-80KI~wuv3c
zwPK3vx`Db#jAZmcDLzm>RcFa_NLy#Q{)bs0VG{d*qJue=<${=gGaz@Pb3h|M@^;gO
zuW}Ys<Z~X#_xV7727j5_(J-Gr=5Qc&gMbvUg|9!Roa=Lu0{8BeHO!-y3bW7UC8@s&
z4UcoZOum)cw*TvatIBp2-%cKfiurJ7!1iJI@bXq6%FmdqpEyqZ1%^p4(hfu)(o-VD
zhuMlP^dEvHE&OqP>zy-3?xyamvDoSh5o4x|p$?XHP@8;Qa4GK|uHCpvqMEL|QQZ?!
zOVIbCt6w`-2PN2CJRD%0p*Ia_%D<e?ZD{@au@ZMU9D;7r_n{*#C_-9zTa>NXjMnGF
z*cRgisUyNGA=*E%1$jUD?MJ6e!larn+i!ion9SY{+B-R$1CSK86?RN+E=5(!_8q;G
zAxbR#=&Ht}_iMBDpJ>SSmurJE`DC*Sq(k0W%jKV1(50&nHEiR?*v4_&Am{HQ-4j?J
z;%w=uUMupS&a;tAj-z(*-#c3nQX9H=9X>Gv5F)SSYQ260;gL&lIv>a;>(TB^5?b>r
z*#kpjh=?{sQrmew=r>&HFH2?|b<j!vCP3v0BnR|_cAVZxR1slQbz#gWc6V1)hVNfP
zn1xK?Pw|nfxq}oNy^9*%7X->hzjZ~jEQP6s#UvC;1+7eyZz(dfMG+9bmv*?+bpt6?
z4l{C*x)aFXl12`*WxhDpty3)@sV*kWA2mc)Sb8M9{J9K-S(kF#e6Kc`6VY#Gt~#PM
zP|g#&9k_MUs0?&nYZyLZA4CtNs~jbI;Op|}@<gjvq)Q^}*91WpHNrvD1BS%i>Sp~n
zsH7?LB2I$aPSn6Mz-m}9XjF9d{85Il=Qy&tS>x%y?rND6*tame+<9y1$9ikok@?$S
z3bRswrm*GJ!dB3R&(@ZGZ6E7`d|K<Ta=nh|-!PhY3*0~0JSX^WtI8ssxn_QB7I<kV
zS@#>SuxrMPR7%!YBb)C%Q-qRHau2v3tE~_SD42t`8^s0Gjk{=M{t^e<a|rvZE}J5!
z5e)h;lMkcJlT0jqUK@$6LY^vLTru5TxE3BNCV^S)uxUcpm)HAnVkU4IN7z+Q1wR|E
zHe(_+xdBH;sw*+!V#C3FQ<Lef@95grVs(@As*u^T{ps1WtP!l*bx@-NNEE-=Ib*>H
zMQADRzPM;umhk)$yRKixu)E?aTy3v2(VnZ%82zk%P3><;6;<0uLz(6;sE;;E2skG8
z^LiAms6)@RMNcd4i519w{E4o7X{bDOMtK@}qvWMsXC6373GVAHA$J$XD;aKXInJ!l
zUagw+&>s+SE)nKA9_W0>e*0|}eCKwD0;iLLU#5+Oa>dDSk7gsLzt?DjqnP-dRnoiD
zam~_gO-auMj1dLem7+}1rD&u)W@uuDKj)_l6SKl!h3780Rs9ZZqfFa$(mZzm3fqv;
zTZ0uZmlt0Cij~{#Q=h#8zp5MOwte$K^}}J<6+8L~m%}X=!Hcwfn{a~u8}`({_QA)W
z%RltX@f*h5Sih0G5<QE9%Q^%{{r79aPYuoIAC0VYUb(xfl@|w!-^Z3Lw^zI@u}}N4
z;7^oyD=&EB>_7l_0ExUD$2dGO;>JYMzDftFW7eeZ#Ugo^=^q>b9YLQE?}y#czG?^?
zg{N{Z?EL<DF_n^H?h)AY`_*Yc6#;|d<i=dE2i{)vnq~Y{ji$=(ADZ~qO{oJPV%=aw
z)m=Viu9?uwXc1VOW&pGfT71_;-%URmM!>z)7*%bNDp9$oV)6^nB+e8w&N04cC-L3P
zT)cgvhlXrF=Y)&ZAQ=N~hgcJNAhhf@fT9f=I?=`i|CGdmVZ*x7JjI2Fd&f-Wnq+*y
z-HW&%SJ#B1Bj23KU-aYgpVlAAR``(>Roe97XJJ(9uYh6<0T}tl?ALS?9xK`;I>E9n
zfMSQq5e45|41%Q&da!&!eQ1a}5bj_V6`$po<GOJSv?RTpZXXZ)h}?*$tlKj~ou@C<
ztGT#=eACAZ-nyeeuyRc`t?!6%iWc`}7CKXgLXJXLX#B9m^^~?r@gI6S&yRLVY<^|a
z)V8|h+K{{>TSj7H?{{>c@)0;~<~#Cv-cf*5F1yg5sD}Gi&CvhGe&k&WHnx=6IJrsE
z_4|4)8Mb@B`P-(9ot`LmZ#OIYc4;uDv3*j{XPw77W6g~mLvkz^HBIO5c8hnumZh<5
zeaBN8kR&$Inw5z+E8{8PzT%vPN=Eg)@Uh4qbm3~aFTZ>qsQxw<v96i5;w=hS-`-&y
zDhMIwFTV`&pUa!Kt1YrODKQUh@q5BV`o6%^eoH|JWj!3?(C^n<s94}%o0pl_d*Im9
z`x*bhB5c9+qobLEs0uFMve*eRKhycia{kF!$MSe<k&l8XVW`x6f!dW7Q@mBx;Wp_t
zDuvd)Y#V)sq2sZD%fyo6=E5sMv#)o*bmgvp%yAF*=a8@u04I7_`0tI&Mv>y9tH(f#
zKfQN<YL~k9J|}C#AWI~+#(qSg`Y!EFt0^XA*abpE+5dUx-Q=Z)3bnt;fjrKnppT?Z
zE~C19yT7SJlgE{6Hg_mhxZf`T2)VBi?GSG2#?9wu0=bBA%c014&H?8Xlb^b8Y-qcl
zv<l@!l2DU0N(QIs0_PaXKN)Y3e>Jx0of^}t1kF?Sow9>)I}?4H_--)I!!j28gN{ZF
zXgNi&^1m%LVB2k?&u6XAP$M^;&p(}M{AJ#7u6!c;ipBQe)^jMyYWhsu`D_TERx*P(
zq1I{-Xip|R0-ys{qdt9@ZGtiqs~n6$kXO+;IGgx-pZZ=`9WF8Z`rikFqEW!z6-;4!
z9xR0Ey2-b4>#>Ww%4>CY45j^j>dIpw>Oe;@XXFvMlCPQ#najehVCyBuT2iHNkgWAw
zC)Jay!MRi)iR0eY9!iImTfms)gpfJ5t7|Fs-8R>N`KkMUxIm2o@>`oD8hh{rSanVL
zFcsGs(2$!;BM|-_0PY)OlQ_&;tZm>|{Wc+Bh<h#z(HJYm($ML)0?s~uLja;=&>r38
z&+7Vz5d%1vJ^^!y9FAYf9M1IQOn@j554kTLl6GmkDoJISz6~XZ<|}A7qrLi(PS0t`
z1%IFqTYxVL$&J7$0c@Y6s!KBh0$5FDu-UD34jh?W<g5Egpdv(;YxDW(7X30D0%D1M
zrlx}C_c4C)JF|BSYWC(gHVXq4nz|)kl(Mw#WOJBlf!yx!((c=Ou^O%7BnIIFtc>O;
z4X%Ov1cm;`7U4VkG*1n}CkZK*43D8xWG!O%6H2|}o>H+r|5wWPr)eZ~bmputng7`I
zwx9fxmKUKRZmw%00q<&s(*;W%kSvIvRLk>VVrRei4ILoas&ZGB1QaQkvye4X_;qH{
z3CN_sKqVMCC@2Ot#@0AKKP{b*5CYkbO!45bOwZ?V8ruz<DHMc{(ar(Yx<@a+M(>Um
z_J)++<h-&sGK0WU5U!2BVzzMWFNmgl*>DkN78!l5YO~vH)!WUJ-K@|F_job1z?*gV
zJL_kyH5B~8G~gVHa3L8UdUfsoVB*xl>|~s7=Uk!sHbCo(g-p$XgOY#B-G+>N8Of_~
zGSHqEZ^~t!0}k7_tc$MUMG<ovy0=_FlW=Y4M)*;-Ch<;Nj9+Ag)7eAZ^;Db9h{h^J
zw9hK|ilSw-^Jlus`R=tiUxAtCAV2UnO*W|Mb?FO+2a6w{x75mQTV&pHwms6@rNXEn
z`MC(@3gmEKt!7lxF<BQBHlf9H5?agxqDz4nnNlv=J(g|$80y_&x`jSbg*$%OT<dIW
zb+}cEhfRE~kk?SNnC$CP^<7M<c7;UWR}R*RJP-CO@{_AZ-V`;`)&n~&Y3(BV$4i1X
z`W{2TS?GrA_pWngifvW=j?;Vb;<j4ii4tf7J$0K;cb2widpIYY-A4$szO#80c^OiQ
z1}fHbPk!i<ZL|53x!BKx{meL>2UU|)`1HL|ll40JCt7J-SucvV`+;)tJEC3Kwa4|e
zf{{!6)%(qvTW}-&`V!}=E#vB$hL2=FrMLEGg&&>!3*}hbe3I<&PKZ`i@gq>3OuZ@Y
zqwN&+pUyuJA=2IN%HrmPe%u}|(O9~&x$l41ch2|FMC3rkIc(<gkhbeYN*6~%&KD^{
z^d62Q7k<+>)yHsc$p{pleAMG1mV$pCXI|Bw5_lZ=X@z|}rRb`=<|@w5o_^#GU%QaS
zV5=eA({NSEYO>!;@We@@{1CkNQK<ZRjLBmlTkL9~j1`El+3j%IQv2?M=$pL60uyR(
zvvcBrYjm|)Y<+q;Whhtys<Jlvd-*vrVF)%dZm$HWAi+!u1*g6&i?_b@Ya|q2MkIW`
z5oV}V-^i=C;0gy{61l-o*3d*_>B44(!_42Zky}4whxD-~)9coZzQ{Asz~&Q#%E4-3
z(x&4~)+5G2hv<-RtJks8Lu{Yf$V2bjm-;`$@1JkK&NE|kq|B$o$s!IrclW`IWmi|N
z2_m5$>P**`zgp+>RnwJ`?wR8msEKI<gYy_;jhVrpE0Py+Wcwm4;<4q@8aoUAMw+tK
zw>h(dhT`$e{k=<T(ekS?vWm+D>f9oSW<nqLG-&(saAkA<^pEw5S|d~uZh1_=T1uwE
zycl2xF)q#O-iJR$gC$UE)cIFzza_Y0b~r}a)VAJ5+QusLIJ}^BHq?UT0aX>;1E7iX
zb@>zD(0kATuF5sBfSu!D8D4WgXDf+RV2CkW@KC;vMaT6ej2MN?=1Q53y0g(~187EI
z4TIxYp`7O(G`<F!QTPCy(mNd~gUe5w_5%w0gIxv64~-}=2t%5epVR_iM!IJw-V@sX
zED)w-rkMHDk`5ux(|ga<e4C|ORA@MIi91om_cr>r1yH{E_cr50L7MHRg5mdol>OI!
z1}4KAX-u&us<&*#e(HrZdz{O<-0#jn+#=iYvB9cFwd4oILme98F^^fJD{F7@O>RCt
z#?a#Ne2Yc7usvjgFX5cI69~#DF+yI=COdRR_D%bSqanCbvU`c{x$Iw@ZQ-<<6*|RK
zEi9=!jlXn>bqTuc{hQ`7<$$F%Fya6Q-RsUatJqNDZZh$!-SxNc;ErC-(Vi#ix^G$7
ze6keLT)UAH9;diuJ|dIj(R*xYSMpJKK$$M$lDpC3Qx@l^HJo|o+Fn3z*htrRs7K;T
zeQ2#`Ow*@nInvUQWk{`GFY#4M_e^7OdFd?s&gnWgrQpv{^L0IWkL*!||7rGL27<D9
z7kryL%fwnfT*co}1WKm#M^R_^S`)Si?n-YLKosg6Jh~cap3)^-#ux!&z~##WiIp<Y
zORlCZ&s<&;f&R7Em=kWMp=okdx!K5TPc3)NiED1P5!XcBouA)_ska_c=~>O#0b*$T
zbk#2+*a>^QrnI%!#R)c{_N%5PQ=|mF*C)d$W>C@5>#J|sbxsK9?DC3t5Z2PE?cb?x
z4a{8&Z)UCllGA;SS3+M`w#(Z~>e9<B)ba1ANIolnA0-z6T>b0{?!qRneLYR-4zcOo
zf!0|fzqf6^mVJ+p<0cHtYF~4BV=L~25Phzo^GltZ-lvB=EwB?YIJs*}1x>v;B@Sb_
z18f?tb=urmNij7zvl#69IYJ6)^;sn1L&U;0-oc#YxaRczb`exEYpEuebHogpcgL3Y
z*VK3=<|K4#_hzJQVC=;C3(6I!3Pg}%cL{JIonajD<32YtTa2=&jX3+A(w*+J0;m3r
z2Ma|7DSjoVoaDb#^ii+@mh&myQUMK`l;ddang9+^ko|}-n^4Cx5K;}746FwoE`%m*
z`PGL=z1QPbS>#RNU(yiCmzwdS!W^2nif`3cY3<4b7Y58VyFs4iJ*A4sldIA7<qTb?
zJP7Ur$`@0jhPaoHTQ^+{OOFmdlX|=kxfSJQ8f;9NsI2Sb>8KhRRfE8bqf0olqc_zU
zl68B@TJAlk+QM?dvbuB3lc-}lz2}?7>mN)8T|(8IUWnNpba~@rqO<fS)a?gk7&j2V
z&SErKeEqT!Z%fbzW|;};aM6VzJl{O^>Uqznc1TfC>_YYFJpYeDcpug9#-ZlR3)TY;
zTC}b!#VMQlaA_rfha@WXFY44BZ2arXdw5ju3m;O?FZMd3gW~?+NWuNP)>9(u$(~xG
z`CpBTsMHrO`~e;pADWSud1<_|=1P4cIehz_G*DnxWm=13A$=o?yUOJ|8yPnvZF)ka
zcJ;VUYkGGy*amtrk(e}y0c9=-1d#il3JAY;o7-P_lRy6)TVb_75X(#?;5gdeG;b~z
zR`$K9ff1%YxFDLF<f9H@{@TQAdfng4h{FEb1}nzn4qO<2!YT6nn0>`z{5n<zf45t|
z)I7DT74^`;?pql=G}m~ux?J=qEzFYY6dB_(m>}`HVe+b0m8$xWtG;nck|!R`KmRbL
zO5z|X<N7_59v4ljq;OiWw=rRqhNM<q`7m7_{obGx{e_pIp0&>M>Z22*oML(V#k!_U
zXU+tclK6IPUtPm2v=sYs-piNoX4X7DsdVXt7dTdNi|Qsp+j6IE1<3Z?ilCmpg5Z%n
z`H9xJyB@>-#Qw1Fpfy0D6ZH$b5HC!nQ7+D5LTW8t+{4|&h#Z>_cDjaa2CogZQyuP=
zWRW06d-piD6LdEjpX-u}(DzyDPF`)ruSU7RDpkLpA2w-w$8&Pu9Eo|B(<V~&P-)0!
zH^GdjNOFxT(0)kb?dt_rg8I>UNq*@yn*Hqa@J%_bt*)u(rZ7i+3X(Nq{#d_&0?m^h
zS=Dn5LpKx-1U~L;E;{YJ3KqMFsYFWBXpnkoVS(SGHR#wb40Y>wOSWNlUVg4aK9D4!
zi6#9?!8dt!th#@_3xDpWs?~OqpW0($T!g$Sd@R=Q$@~2x#m-IV8I_I1iSE88d?FF_
zGt6>$P6pwL<VB)IqbqY8vu>)h!s%aKv5fgwBl9dXqqc6zk2Ne*n-TruYZNgvoWn%%
z7f%BMf$>9<rgY1(V~v0k&@oV|&4FN$PIbB!v@_61xCGOwO<M6#-_Jo0UCPh=1K+P$
z+S_2=5lY6^q;sCcS3vbsoVP=L;J<9=-fy6h_Y<%oKc+nng#VlrK0mTe7d&fDezV+e
z-YM`)pNeeqm2$`)RQ{H^ZtK`e(ph<3=x3*o^Mhe*I{2BE7Ln+QV{^AW%oQ(Ty^Qi-
zUG4ieoe=r#W_E}FUQcRgQX)?BYm)@--Au2&Y|0l(ze;2U0icUlE+Tr{81<NzQ50Qx
z_TY2dwlsA&`enUCCs7ezmD1_~Eh=xt@`d36<~?E}DJG<dmZvjA$j(v^?tIQ=h)0m@
z{wdvX+xvCUWz(Ww=kInw-{JGrsD<&Dv*$7S9m|y^as#*7bIy>Y3JFy=4-O`QsfP0m
zt%-S)Ig<7rJ#GWBbTDIRJ<7C@;T-Md55qLaesc~hhT48Xma2re`#(G@llRl5j}o7_
zADV-ffIjo?m$-#aYD8_JMFEXZRkhdlpOLwZV|v=GL-do(SlEbLT`2e+oy-H2=<IkE
zgcsRPjYUuWv5YB+ckZ7lcH<c~S!zn^{~eW?YM?}d1yX`CkTX&Fub||os;*sEwnP+M
zO?#R4o@`@TkAs~PMIvEN40|l%{7{P<Ot4kHXNtkX6kQjd3qlun6F*0_yzt%~@CSS2
zUiu$bQ&uD43C2*I(hA|%A3kqn(Mq%@;$EZ@xo3{2uJ|RXQ{w<n!(ohN?x%|Uv3~kf
zS8sM)rx(R(tDBiITAvpU4bN%aeQ0ud5<t3I#;?}%dgsZnHOOU#;7>~rq}e!rTdR6#
z{eyL{cVeyz4GD6)+;8jKtEX=F2=@B+HWbSytDmxB73ws0MAQCJD$x7hRiUw|q5wc$
zV()f|ZZa2f7kedURe>3vy%NhNQ{JoiV^g%-*<2a$J)A1F@mj5>zd0}HT(uVz1l~zM
z$Oo8!Gn$ZIhwl_ai$Q)W&GT02u}}2aNJa4?cB<wRW|%LWL*G2_AHsYa@*m9~3q%IW
z0$1TY0PB)16CS$QC;jGt+Sw_*CX_L>#AWR(?}g`uV!XnQc;HcDOryHXt7cjd$vI51
z_On`yszHCU&X!p<DA?SRd?c4{kFrXPRI*1X+epR0d7HERKe4iTB*XdF_{iN3eG`0^
zn@+`DpJ2%A5BJbqnM=c+`&)^dr1nZ;P87xo%ur}UxHKPIqu!B3hNYbP=#8+gpW8gc
z{o>0ZH0!QZd1+vV86|f95Fo{Qgum<8Ggh4AhiW0x{9PHdqFlR#4$kfVJkJQtnT@0L
zgij}uxY-v&5cEt}4^B^6F6YpyRyC#i-ae?vPUN6&2_E~QMpWASO1HWi^J<?RN?%ki
z(`n5aif*P!aoiSQDD`_kh6)czMbB!=A$n^|Nk!b4Kl?UDj7NWbt|G4{-QnIh?<&L-
zvEHErI_Gz)0Pk&Gp-_b5?GxveCRG0e-?F49KiuT|HlIs$Ny53rOstb85+fYmNE3q3
zr#N_rj{7@yL;2697`}~mU~a^~HOIXPQ|94MtetCQFE)zl)Ls21`RI43b7<Op7<7&H
zq;7MBwZ4|E%uY{|X)Fz=hv=<x%SUl#=sZrYsb?AIFkMkMF#`+^RfW%Q)sqwU#t)&m
za0G{pKQIeSJ>)QW7o~k==iL(K>y5|`v6Ouf=)!2)wMWVoUglY8dF2r`&uHs5<d7e1
zyYHYBGnU6vR1zwizSf^o9O5|`%nW#GYp2H2T$1Bjd<wKWy~0{fCL!M>u2zlXDsT8(
z!E+<m^{0DL&0IMi{`+|a9vRCD56x=!rftotdv?z6EOhtN>j~bf+}o<kD&*8(>7ODS
ztQB$ZjmBa=E$;alb=u0eeMwl~izmF=d|5IJT9;G0py$6PdfwpDU-K{3BkA02xnej|
zazaP+nSyceuY~S9a-EpOcx}SdQ+*j6aO)%N&)BRvFAX3tpH(8CLry9-28O45?@o9t
zEbU2D!M5wg3oN@EtyJVr7IE@lhE2rFLaOp<^-dhS9;|YwAyOk)ESL`lZ6U~=5aT2X
zp?+L~EsGtaFoM6%un6`8Vi>XNzO>MIMr>(pN~y5*qw3XWSB=@lu3Fb&k73eHy73p;
zw$suh5I0wzwo=@PxF(jjz83wgUV}fxKFK_gy@;EicE0VKH7cgg^aO-6VchoUa|7)w
zsHtlz@#;_v0O+&@xm`Ci<W@%O^U#~IhJ}CJu{~O_^u8>fmcTi30g=zU!bh!0Z4-X$
z*4+=O!(OSL1TD*XI)1KJ9zOcTbBg&-x>Mszia&0%sBB3gGnK;|q-OE6UbF5yr?C+t
zm6t@5?{q73Aby-4gY3l+nZ4XKEybf6B#uA#T6sshq&Y9S`VB8u4j2imSVVDS)IhlB
z?P$@@qo8Zm`hl)grDPr9KONNAzVl9(OBj3adfZkqt2%~pA2fW7b-13$t0Ha%QN&Hm
z3g269^dkqdzQhL1i#{j7@}wDMR1vr|4O6D1IEy9Xh;R8xS5baTxGrzW+c0k)YxiOi
z-o}se2>b61lEe3D+X(!yBWqepm8fhh)Q2^#dY49;IBuvd#o*pgVb7BK!-lU}oaAjU
zuOY`ASaS+1Bf$bNkY=9n@?KCt>h)^;F3jh7pWb{<>x67wUNa7KIq1gL^iq!0!E#Gs
zDpj8vd6m(wO8~~zIv(;`<R=#v?@9)*XZ9l1(QbpQOB!v<t2(X449mfq2jb77W+2Vz
z`DllS!55LY3F0D+E-4<eIW+1hK|3u6Q<-3^cTU{l^l$o3y`2-Vgv;nIRAyTbdFRj?
ztDdl%8*=RA{D&|l_+B@bL0|hOb=o<DQMIz#QtPT~xOh64xGJF69|C}1M6()4TzX39
z-4*qk#H1$GYzgV-0}NO0AdS?iTU6Uiw^jFK_s;x4W9Rd5Pj6ao0z|mS<=(&Ivi0Qh
zD{u%=>4Z+Tsrvn3Vlr%CG9I#WulGgBPFoVm4~*)FFc{$;pYL`DA+sENtvNv(TVwzJ
zno{Mh#IDMCpSzYl?aP@!S9!o>NpXtw&+lZD4Iwn&f}w)w5;*&`VQa5l5az)=`4jSR
zls2E2YNQ6}_pPyF>~DOCenw#Bf5j@ga~=;ln+5q584xymDddvfM%AyzH0PIg8v5b2
z`t<WT9Fv;CQA)wA&u<C^%L<)%U=Z%6Pk(u`r@r3g6tLr&c>|=*Y+u%mljQT?WO2OL
z$P~grE@3rRUExh9BUqtT1~d=#+b?U2=Sf1*bwB+^IU=>h?<XnmysF<bd8stU8uYAW
zxn@5bA{9M;wv44sKJ!XJIe#h_RXuBc8LLr+J`HCN=zY9Zj>OIOLTV=6Zw_Bzf37cd
z;+z3JkM=3)b@6Uj04tK<v;8W2lWuDCiFJi=NE&~!BCy5O4`0$pdbaH{tlHXh8uhwW
zNOzr^_d=p>0S1_QjVG)Q{!;qH(i+0!LG16hHa(xssr121FRY&lh1@)USOBFPvw3;6
z=NGho;5QiZAi+7==BeWognS8fKG4rw)A|E{e>#LX<)blTJ?N&ilOEXE4X-#C*9l?H
zg5+jF3NtyxIyR#uu51?k(4`;khgj0XGb!5J6;buljn{WYRcnN(GlWh^xov$A+<>Z-
z$P{P}XIPX}K+b1zct4UaX+hTrpU`$k5%Rx|;A`DMJHB&DC-nv~G;cdmfPbELN-T=*
z+3k!`>Csn<ei%e?Vv0$?w&T0a1wE{d+z6#7I}h=fIhg+A(R;u6C*KH-RyIqDDjcb8
zse2lhx9-yNJ*2%h>-paAHhzxFkzoBntSM}dl?B4=tk4+Y2q>;Rn{S`n^YR0a;bDE6
zF?Q9qFR2S`?0pVv2Q?dB&dw#lcACBnPM_s0hR*pT{5`ew6LMkYI>^8l*I>1`q$E|;
znmUv_j(Z-<RRz_g$>F%vxy?AQglsoisg~JNWuy$eU)8=8tUY+{;F&^KDf`18<v^fV
zvT9gce{7qRgH0T%o?)C*ZpYS--wv?g@q$Lu+6o14zu4cZPVo<I@44+7@*3<EENfr#
zy8N*IYN+=`e8ft?XV5MCd?K*R_eb2%LFvA_b(${SDf2R^pf`dO2srUO!Hy&DB`uTC
zMayi+z@$Sa1+n@CP}p@(azDOa@El%syAOUo6q3$1u`>$Mp7ddkn`=I4I)1ZXR{3T;
zhe|H`xTI*^J~iAn)@=PYIMqAekGgk5w|R#FzEa!BaraC6MKu}KWZr=0G>(b1YFp;v
zOwDKowz%W&U+nx`lHQiWZJJ)OvPk#163)qp<Qg4<19~5w_R6F`mJGjWk%O~ORauL6
zZQ%>c1cIjU5G&{UL80x>8xxVXVttyHb#jqh5-Cjl>8x>{>+(|8*&fwzg9X+{bXodK
zqD$5bTD+wI`;ndV{DeY+_!KgWAFb9;DD48kxskHliYf`R-?J>pYuFs1>V5w!t~y6w
z-FEY>Km&7{E?)i);s<%FYzBdup&B}Cdwi#fy0@|1R4YTr>`HJSY9z`v%h8NT&UH$j
zZE({M02PG(h~~M4p?-Qp!JEdQ6Y#OnaoVf<s+z(t*mkNL*=^QK)3mj8VckRyy_gUC
zT|wlgux|_SKC{!}<-z?uIZ?I(N96U<dZJa#*_hu`<=?yG&q3+UBTnA1hJh)@wx=Hb
zQ)cy^MW>ALo%;JoceJj+poHgLWd4~Z>dy}qsusQ9LT5@$LLSgnB^^|nz5+x)(Tg^d
zXPEXrHDtJJ?X<P+*&J}wnIWYKK0nF(%c<BU`c<7XHhCXb8LiVlIB~DzF84~EHaPZ%
z|0m2SBs7=BUsyXvL6~%4$){^O<!I)S=5p^d?U2Bg%aQKt+^&ry!X)d(VkGHd_HzWy
z!W@1>6MrYfD%Zv%ys|-pcRT887rb~$p@j@w(`4=-E@^sOeZ{_hG;3)ho{9X!E-M-$
zKmA&=S={A$e;FRwXudYKo3$Bp0SO<qwdj>I=Tx$?JzAQ5^_vUESDa~!lHb2Kwxe!?
zhP^bkS-009YY1)3pHq0SUfYb|7`6nRx-atAsOfLHpI(C(c4%bnfj{N6dB!F_n=Y>u
zHMkEX+`pjSZ_3z&;_B}<Jvff47%Y{&t7{gT;&gY#3&6YV({CadorEK<WoTry0)1H7
z2=4mIJ8bFkQ`8OoZfs1i?E*~5UZ(xV-t``p&FMc2bHReCd9ML1sm&0-C94RGWN3ZP
zEk=>2mw-F+1}mOy|6&Ta^W8DA!|fZ`^vv5O!;Y%ZiSbiT=_u4N?xvi`O*Ty^;BvdO
zlS00Cy_^~ntmqE&1Nw{>5PBtEO$GyKw#kE<h;WqiDRO8lfr3UK@d}&zaXEi`O7Urk
z$DY}b@H{4;Pa7`m{hzxYHh=tCSoM>#oF*HKUJV&$HDQ*(-Jv6+1CRQVA!@xln>CWp
zFEeh=@Vhsg+BHAZ)taG;qBkj=N4Z*pRj#3CB?%KhiQD~cDQ{Y-G5wrU#Do=a^xBm=
z!agPhPW0Ou&An_bdGY%1PvXx{Kl?yiJuOxE)g}bl?E37FNUvrjMvRV41k{or#^uL?
zS=_vTD5kCi3&K9pOBr5hyKo!J*h-Uvn^mn*-29*`{1f>}g`@GdUy=iU4mY`T1;2LX
zI}jv5S~KM?CMT~$1{y!THH3W}isU~e16$HJJw<?fb}vG|sNH1?ofYaXN07+)4opgG
zVShY`c<Lm12uZOL_%KKGL<Dp^CM_|a9An~qb#iN%n>=|Q^A?OOP!^U)wXfDV2E^%*
zU+$c6ifAf)NNBoxC%GJ)bKM3BvF#uGWIi}i5Z#yavC-~D7o0*k@<dkn50a`toAom0
zTG!?yGgN%_C$*LBX&F3<>11@2J2yD;aCm)@4(z6gl}-6s0Ebk%ZX#RFoELbdMGK!C
z4;AK@&v)g2-+;zT8%Mtuwz;|o&97s_e{z;r#4r@vI4+wlfj(L*5M<Tum^&CCAo}2D
zd~bFuQuiw^V|qL1c-T)KbO8phWs_T3#Q6R7S^k}gH~#yh>1#(d_pY?^Voc82osP&W
z^7CQqgaSp+^M}(qLV0_onyDIoo9=_l`hqk>tld@JsqLX9uLW)Z7omjyzB%<fWCVOF
z-hgf6*d)3p=m`bdZ<{JK?_q=4*Ofc0S@P0U$i*%=<|Jc=O<}%>+--r8a%Z_fBR2mw
z+xI?FB7u99S}fve5Pr&#C;P^4gcKz>^O02aoeGnDj9*0${qEGdH2|q2?7^F>PmCX4
z)Gq($1bCFTUF#{D1QV|Sqp~|PqfZRj4#YGf8FML|f?SH_#e6x%shPyY_U%|8V1E4S
zoe?Q}+H;D$6HcPQeGH5814*j{hdLf|(`zvopPsk5C*ePjo@uN49%3Lc0=yOCh=b7W
za)Ken)Rb+CuW=kLu^Se+6ZQaqu+;|LrS!M3J!}tM0Js3g{PYlzWaO_kZ$cD*qGw%_
zCj2mh1&0dbOz8JG2y)x@5a6qooFx6nd3AQI^d155MO!~CqsdKN)!i)Bl>v4(V1Jls
z!5e;Q?9K3JVR+q|v32Oc$t_FLIJ<hl6i0oH>!a{%N+aPKCK+ewVv!S9s1)-%lK}|u
zJBS>83d?39RxcS)Mk*}6Oq^S8@uwcYPW_P5NQ&N$LHph7w7iS58l{PO@OAkF*TYu}
zuDj-?CFkp<N8t;q>!uR{Jpn+BY&w_b!5s8f<sd&iu~}hCNcRH(h!gHmTt_V)ymh;t
zxg8?vm+mul#WHc<OcnkcZjb-?$RMC3W^YsTfIp@3P&PZf2g34YR!y|^;MA1K-Wv)^
z-sIkIa)Rhgw7@|2JKYJhA>JU~*0RPrHXes0*BfkSm_t{;I*ybC;k$D{iVtf&c_6XK
z#dHBP`m*7cAeqEh$uqOJk=QtxqeI{1X7;1c9zR@G$j7MiopyY_XPLipDurV2?cfwM
zCHK^q#{c+YF6py^!N!+ESXcmiXp=Zo#)dO2S^F);`CUC=Lw>@{lC5zV^y>1BC)PgT
z!NAFn29OsJcGkE4{+6hHd9jv$2p|mnzXMv$54Ax0JAXqf{5|hor=m=TLaf5j1mF*<
zwh0o>wY$2&-!(<%^Q+@8fn9i>J9juw_yPqL^fE*q;NsAKXyrEyneD1vdj9H@EcI37
z;$^~M^xeMoHcPUb1vy5H!UZV)eu>(L=IM$GeF;c1BH?(moQF)un706U;IajZ;$ttq
ze;)qOon{UPYp&=a0bg8;h>STd0-)UPsrBC&x$TAI%FjVp`U1cB{^uo2h1z4Zv;&~F
zV$Up2ncgD<KFWCC23r@HpR_6Khuz20Ea6(h(qdnv)A%xBcy-RsK$2|EM6WXy!9mjP
zKuwG+mJtyV7Kx8<)x4QfI#e$ANg3XvbqC`?kz4ifg~p7CsMYJUGD3`9{!8wJq>4ZG
zK~iq%6B)aS;;mMBd>Ty|P7T%*hktC9GUZ*=v}w@4&xMy(>~4Pm)h0+R!LCLmgs;zT
zn`E#@Jn3a5r2zKg%+%o984Qu;734>JjWZyT<^Fu?owQydjk3{VE^3#2WLigjb+824
zz0Rwfk_u4Hu$GIOBUtIai%AzA6A^K;d%Z9COky$Fci$d7BGzPQ`hyTM0~wIQ$sw-K
zoxfj*%S>5)2-N!g()IwWs)xjdpT4R_-0z|kX(-<EY*D_{2{OaA*)(RqZ>6FU%+{%E
zywU0Nh8k!JL~55A8I~C$Ve&VKz<+tw{RJF&N<FQ^3scI1glFL{VU+L1DmTV?0ys14
z_~#FW{y<eF6IlvtTB{XmJU<5nOSG-k9?Y*}4Qg>Xti^U^e-`d3M)}B9X-B`#3P2+?
zK&EqrykP7WqL!jt<H`8QTK`fnVaGdl{Mgp4m%m`6_#6PW>+l7%!c?v&D*MGL-+Qh6
z<TcQ5Soc{`+mE+K4zxPbeCwFq0S`1~B2wU@0n(jox8JM2)iy&!Yoc`MK_2UPiki}X
zqwBD9Sr+7yaCn?wn5^?8dtbgj%rSB`@)!GfSP9VHiHt@M)9T$lrD-OFDLu!2@AhHB
z{lxmU!C2SO?_~mQB`(RlOxD68A+Mxj3KDqVw21^;0LY7v{-Z)Zv*d(6$UzoR7?Cp%
zJlGVX<`4>g|H7vl7;9$X%o6t*y9Z%th}`7b-kfgWHSvJ?c^C4u%f)%$(T-xB-omSn
zsg<7Kh<kCZd-^8_o6azU@Y*})(55O?d0Ho*7i)}TmO}tKW<cYk4?CAg0a5x{#0jKi
zJ}Zge!_QJlF(!1gOPSMr)2ESS@iz=dx+U-Q1Ck%>c!`;)@($KwEfzcJ=A1H2I-o$E
zur5emZKu9`4L<&MRK_S!NGO%-oR*si2t&$+ypj!3`|7In=@7m*irA&0vk}UVvh&=C
z@L)HRa`zns+6VJL=(hGZ5><P1blRWWDY!co6Bmd3dwV(pQm^Ta&krwm69I>WY?tBR
zT{`PtK`cUBYKgR9A0Yn9Uu9zVs_q1MOJ^M3ch^k5+nG3R0e3>hJK@e{SbU#wrxauX
zl!6Q|A+GihiRoYbHlLYFZN&?q&a#l*(h=MBm;0%hh?kX$Y>HpJI)2oY=|+?OS0}@j
zGMl=@$Fw8r)R14lfJ4(9AbYn6T_<Va^X|$~|MD`{F-g?_J<>>EJbFBEee6O0zqr=l
zPbwe?tTd=7X?_S0LyI>*A#cdst$#UnrTFz10bS|ZPrOgnnvgZJ5#*Z?3<o1(=jRh3
zMk5ljkw^pnC*Lz9n=CMpQq-i7GZ9bEd<CQ<A8Xz~VJE7ss<ujRt)$;|LH==2<@ZtD
z6a9;$Lm{O7Ns{|7_cla>gDn98g5sZ$`nVZ|WfUEWROCm?3#X90x+%F=kKb~y8=DO%
zB~}~5=O|GQ82|#3%xx2FgxE1ai!$Av^E?j05zUW=++!A?T*ayYJb;cfg)zV9>;LZ3
z|N4eP`O<#6a6~ur|2<p$>(Bj*zfHIT*5AF!Y4X2($A9tB|GFN}tik(^EA(B_KRzc)
z>)}7R`RQAr(QUSRw&LF`@&C)^tvJAA3LU!-`;RX4Pe15c1@r9{8RP03BLDq`;s^z`
zrnT(Zi2uPI|L>#yr^oPb7au%YC$$x0!|!4rq}cwPc>V`_>F+<g_SAV?=d}g;*zwQ&
zJ5^wc*+W5&C^0|0@ITlpua$pNm<mIH?RrvU$&`uYTs{^`A(?VqsJrn$2sES+b-*fn
zpg08{{QQD(9g&wKO7%YrmcM^Nuo$?6+LtUOib~uZAT^sMvH1Vvzy8Z~^?!e0+m;IL
z)BCQQQjgta=zkf4rx@&g$j$v9uay-5K4~y|*B(3Zd#z5#>Yx8V?UVm__492&XdAhX
zz4iESKYXMdNS_~*K1pQ#$5;DX+QPl0kdB)|T>5?|;{Py%{a#h8Dah@WGPl;&_`gUx
z|Myz|yW6R~d=Vt=Gwd}cf3ZFOc02#y&+yr63N@%z=!(JRe;Z2w?px#~fHY=kwqtMp
z-`;yh4B%C9B`06&!+$#k{_lhRkMh|6KG=U3E&sfH|G!1-3jU74nSl;d>$oG}mxh{-
KYJsv<(EkG>Upu4#

literal 0
HcmV?d00001

diff --git a/_images/benchmark-1.png b/_images/benchmark-1.png
new file mode 100644
index 0000000000000000000000000000000000000000..0c4ab31f1b11535e865fedfd474399d59a16d1c0
GIT binary patch
literal 62190
zcmeFZWl$Ya^Cyf2cXz$G>&4wJZoyqca0%{C@QZtZV8PwpAvnR^34U>Z^X$9-wU7IG
zr>d)ay4p_l%$%t+zcbNlDspJZ#K=%kP-q}|X$>eSm~|*9Xi`M@f12BJ3-Nyp+D$`F
z5~_NN^yJ@(mz6HaT3H#2>7R@U1qV$G_31w*|40;?<bTMr&<s$p|4k1A1r=!v1^3@=
zRQ~b*9EtzvKR*9Ihy4Nl->v`Ye!%>fHO%@C*#DBDN&n-P@2W5Gk0CkB>$yQe;nV(y
zpg|gdD<~*&D3G*-mKXGyKY}UgpFejSg`1V;ZELPhtw4@R!g-;mFH|Yw3`hWR7^>99
zd~s}W;wB{$pmD}`1m1pc(>L9IBCXQcv*}#ysLHxYze4e}f77nU!^7HG=*IeR{pF%#
z8b2uNJM|Al2AJ<b|C8PF!|+9Qjkx_m#e<d+$5MmC{U3}nEC~Ao6SuwK|8x<CQE2rg
z*PK-U-^|6a5UGBI3BiRs{_hzWU_=uB1<D)Ia?|=>qW>qCBqGBOx<moV|60&LXYpy8
zA7LFwoPS9FkD7up|7Dg^JC69D_5ZhAGdKSNU3e?;%Kg6~{NHq85O?SP<!x?ehLSvi
z9xlkSF;1Kk6ElC>M${pHWwIzBKMG@scKgy=R$U35W$*|spw+)IjF9Y9myFfE(9kic
z^@c|7`We`qG|j!4pVaR!yr}2Wu`$~-Eo4y6kgs~%`5wgcc$H%ZJ;6Xwvl57oY5L*f
z`=)*=)>&iFzJ+AA&)MD3W$aXOL{U*I*Sx1&WutR?aQ3x-kCu#`QYdfY0Guo2IX<r{
z#ySvzmb~WF^Oj);bE;tVgEhR@x*^l&X3eoBW?%=ilv^-8iT!+TPD8KDr-{*Sv(x(!
zzGSXH-ObQ22S7_3?!2#TY+}M*X`rB>@FlQ<gfbpl46(98wzDHWrbXCfS1y-2UUW}i
z*_8*wOTi<?^VF}AozlhPFq%f_a#S0^Qf&Jns>L+U%aaZK=;AiS>5Tn%o4@`pYIg1w
z`t~y6orO=e3Xtnlw5;)hBsichQB+bQ+w8DL=l^`mNpVWlYqRnG_EKC}C@C9DI9vCE
z$!gkO6L41n8DbAm_saK_e(c<RGd3__?(%(bOdIq$gb=0X<k&CZV92wI<>FDF1=%w6
zt+?G&X^{ce*sbmYp1eHcDp@NNCx%NWeoN@(R^{bUgdzr6kjPX`?WKFRXpi@cSV4-*
z+qmr1e6(h^J5n#r16*BQ0o>f&x101WD;=ISPr2FIb{D|zhMSuktKnE8F4JD90h~gO
zI`Z)aALEqtbdsVkHO5^4Uf$2UdwXd4#|7r`K76xgU*)>=l<~Gvc5*#8lSasWl{N!j
z96(A+JnX3!NzM%oUAJ?Jkaf3dqPa<uF>J7Iw-%@t<lazUZ#nZrg3B9XDe&=nLn|U8
z;ts2C+kP`i;&;B<1bNx|*wpO`fF$Q<oFOA4?_UOfbeyd=DR#e|J58i8YW@z`<Uqk^
zN!)t7@Z_?c13aCm27Z<IG5B_sc1iH?+%nj3Pj*eV&gK$29Mf=FW|8K+Wbb7{Ec%u{
z*!GKdCCObH$+K6J{z91|+INh5mwBBUI>CU{U+%j_uXJK}v)ZDCNVvE_b<cT&jr4kx
zX%G@v`115I{<8c`D0O-;8prD4VuNlHM^%_=$t19A<&gVgH-?>WZErX(nN~4PUSD6o
z!&|d0;2>3Je-KC2{=qO$yx#w6TjJ>u*kjv$G@VbOprfNhU(@~k>-D~+C-vWhS+%-#
z7rET-YQiKPcOvF=e8{#1lv7kbKPP!i6DlEh$O*PXe(;i*A-fOyO^VIl!|&9(vb7=h
zyQQ{WBH-gh!rS@})rxPKe#Cuzx*qJ0R;HZ`wocBZ0Fz=lV?z_Vz*ozfGp1kAHCLlK
zohfBl7sZO{#56n4cc)g{y}^){o;T$w!BdujPu+J_PGddqckD&4o8Cvct{-m?6QfBq
z;nSb;9^NhkO;iL=08+*YLorgYo7qm?j2OaKLEa^LG@e^eQ=a>FCNEj)94u0Mb-Eu@
z3*|>j@VSh4h<OtPu%N%|ZEnhudwf-FBZZ05fd|YcoIhXsQP^=eEf;NW4kw4dyRX`o
zX8!7Vg-k$BO+NIiJ1zt>MSPpR^}v9KFpm6(O<&cE);;thP_n6;R!~~Q*6Sf~c=v5a
z3L|`!k2B!yZs|Zy>^aFB@wMmU;X}WdaKNn5dY-mO-H$x|QPXWu)VVW?j9sYHJ^{`b
zbw-0!8>2$>+<)ZmFZ0yz_uXR=T~gbX^W4p@y<VPeJIsO#v_T^s+Gc6Y(f|wEL@+Ji
zZn4su>3mUORL4>r3uq!s(fop=j`o3TiK?1nH2?fqSdJgYI?16s%CU7PLQRHscj7IZ
z&w=Z^)3K=#i65WGEs=({HXVVXbMV+KQ?l421mJSiu%{aMoJ_U0uBy@1)+(!gCIE0;
zf1t-uOyicL8$0WS(_J#E|6-6NkalLD=)ElFDGgSl$e=9dWqY!>sq?FEN?f_kfy#eB
zxprN3)@_t{e>y(-PNFjXS>6p<E{IL^)m<i|nwf?%m$9O{kGaSd+iLZPF&K=>%!2J%
z)lA0x?R2rKU!8Tl+y5mNI!TtD_@LQo%dy-0EFti<sM4@)PB-%7DpG9p_GJFxUqiSa
z0CI)#ygMRtMdbNrjjK+}wGw5un(7}1?179A>uw3S1OybG)tps3uWc>P^G!QoMYbeZ
z1oTl7r%tMgbap4S7z=ALRt;{ODIRV6?q!CvSiimvzne)7Jw3+2w^Ne?->sKCZ_t(X
zKHtH|`{l=PPo0lMVRrq-(atgf!?%sg0N_5Kq#cnH;@1UlZ|?$Xes`IP)yM0-1Cp)B
zh&BOgUJbFzkc$n)!T494;XJE3x?3W~SFM$t-Oa0lL7AIqykAxFZ*;m$*I;Gh5ue0X
z4<(IKL+WxGYY9%MS%t9^-9om!gQ|g<)x!q>IljG3j|(_sRtJv46Gk4$vaa!YVJL@J
z&j8Wu7kZTM!424ToYg!=)zRatLcTiKb!6T&)yu^S^{dC3{QEZI5@Tr&b)$#N=-o_#
zRK&6COZ@~Jw>B&K7(N~z(cH%TH=6Q*`xTq;BBKiJDz0hDi6{vTajVx6vG+Qnsl%H=
zAN(2G{qKw`kcB3-T9V>ToJ32TdX`N+D#mNRHdDi_br*=~X|11Uwd#ehe~rDsvRhd8
z{ugomob_*CO)`5r6!P?qF_~0>@n~>)COUM^{3+~(J^X^Ux{uT28X(o?i355|4?zbx
z12H-v=&^ic^kO&^ul_|%nci+?794Wq0`NVaDTs*G{`@)9H{LRp=AJZlway|^OGk%9
zhifyv6N8(JYrhYHf@c~*KV~fa3*M32u5Oa5*qau)_dZt9@y8y;2jVqyd}83gVL}!(
zRf&wR?`i@!7_deZx;pd|iYy9GzB+Uy#pllZTzS%D)1;+z&%QnP$?+($prASd+LLC!
zIw3U~wolHqv<pc>eydE8+&8Z|;yj0`A9G16Cl8F$($XHn(_fcfJ>-e_DiH_VL?qlw
zIW2b>6E_E}h8$!^IOt}O%E><5^pg!XxMjk9_sD4Xd{!lnazf^ezNp$;eGGn78yOX~
zD`G_VDwl>04q9OG(>Tj}UgTB{k-236J5-qasn4pujM3-wbjbHJmgf2Vx-LKbf?H)!
zB$vSV;tRp3+WvPJakJU8M0W4{T_hUbwAEQn;8QVfmyzZZr2H6((hTqjz{xxe(C>be
zm)kKBF3AR2gQdFNtSNGE>`fX%H(vW7@xL`ov+?5U_O0%qGkdxu`Ea$rwI3AAzw9(~
zt@zZo&)ar@dxM{ZXZ>;D>|~ZEF~nYH)>T_JeJWfUwxe*v_fr-xKB3@`E|hD=P4gWo
zt3AGNf6?^tXckCN5WJkXkQlWg4|p^yP)}lWxmCs>n4&KzNK$~!r}@{4fz#!cbDCKl
zL4(q=9Mvi|_TNlv##EiHMs2-D$^?28&vpht<$ccg!?2LP(CnVhWtS7rjl9~Tn@_e~
zfT|nYw&&xbg#E3L_m&yClvU@}EBGT0-pIg5>XypCivrWaH?gz*w$Xv7RV~Dx7G5<l
z+ER}c6WwU6=mSDvE&y9<a;m-$W@F<t6DOo2*Nqc2-*iECo%pyW=(rNkQv9}6DJS&J
zmNz-5^?P%S&07ccl}kd(<E^0!phWt@@pWT`t3z`-?bpowQ<IkLWA{`Ffj3)VB+|vh
zstEH?&of%Bcq^ol5zBT>xP=>m<|giSLzsPY#C_{WpmnSHunne(+C&;_^39Y`jdvmW
zvrG&h>+vo9K|{!UVI}Msf9egz`&FbQg=_qROH_vF8qRa^XzlO&w=tTEj7(g1V4g5c
zNA_Ny$>f7{iTw-k&*GwSkkMENv+ZiN*{Ph6fi}NNuJVL_SZT}GKXam|x{6z50u(p8
z2W@{44w~+Uc2RF`oEUpX{gqw@>m|KC6_H-wg{)JKk@0s9Agps5AfvIRpM4YoEr8CN
zr}$Fp38Y!TY}!>`5w_#xc|EQ(dFi*;Snesz_^^4MEckj030)JW3!W(#R~s)V_{9WS
z8U2G3<bgH^zj)G1Y)#|<U%0H?G=fwd))SJUbmT~TdA}8+O0)UVm?;8^X`2yKpKdSt
zt4ua3Dyo$D6V@sou7uI8^FOKHGU#Y^dvpDR9y8B)7YQI3Rj|s=6;6{`3BAC@R!`4L
zkISAcw^_12E$>%m2sDOkHZg-(4&&r=T^CCX@LdhnvrYvcKL<+T`ZHdP-B8S~-rKLi
zfd>Jok4bT<tKm8UoU7r1AS&K(Hm#pf#^XECyBlkH;5(gDm3yy5<x99-8Dig?5ocv%
zeokAx=E8c=Lqb(^xjDT=-|W?OBHSvrxJ-!C-c-+=N4776I^ajt0!Seao)7~Lj6O@5
zYarYPT8=W7{_-MJbXGYE>rXG#wY{MQgakDM(!cY@L|04qDVI1)uRY%bF9nl!G=o3;
z*XG$1tU8Gf2WHc1sIjMk41<Fw(Mk6-UC3)?#DtZiKKZ9`B59&_dnwkDxs3G*dMvUN
zDWK9jBfmNiMLMNLd!S4J-;cfs)sX%D-TVVwz_dbI#>^=}yr(&RTb55dTuf6hS|Xe>
z#OFIJ7gdGxH1IZHcoW5!?G4L9a6}yZf_;kL>*UL2tgF_!OhEi&f1wA0Q#WrE_WtG^
zjQoQ|Q*QophV7ona8FzcZSG!jV|G}6N8nJumh}tsUnf}d!)KgidD}tupi%lx4EyNJ
zzJM@az_wJEx)@1~vgFlf6IfHA#Qe9Ju71iuc#)Qlv3Ied?|D-s<K)iKw{O|gwVW2i
z88$@uoPKTdRpF94X*&avBrzm`H{Rk9Cu*rldul}CMfEWikI|}m?#Yuga<(v23wJ~U
zzi+d?L(z)K(gbz<zAZGFGY>88ee}G}-^?~`M}>L2@BD1rY>d7pmJlaUd-Q0z*YW%`
z^XK<j!s+h_!I8HUiAvcE$AV3zp-}EE;ue^qigsR#-p|O1`rF^e%b(ez?olZ9mD9Va
z{~UN^O-mhuI@e4$nDsuxf-{DS4$7Dv-{M_{jtc6K7i8*#(1d?;6o2{__iebOoV(^T
z7pIIdzHndtD+Ch3^fcdZE}(USjZy8EI)P9;8u98-Wo*EF?q<hyaG8&%sD~E|a`<^r
zrXa#mJ|JLEP=oo42J}UWt6EQP%*WFAi9}DKDcir#=^3NFx{toIYUgA*P?umnkbem2
zWam7Lr>{Nifr6<O)mWTOlIts$qbLf0LWsiUq_e<57ZD@rlh1k6((fojG+V!fNamBI
zqZ{Cpw6E~bV)ykC8%f`%`ERL)F>}~EC5jcr{49p8GmTYJv(rlPBZv7J)azA7Fdj_U
zR?nXE!TH+8U^x%|Zo}%fLt<{J4AJo&K+4l~_yCepj_%{@%`^!Css~TIm$Q7j=pX-=
zd*0lH3!*dK^KXNTe~G&D;lwmFi#;Y*#&W)RPq4NSDlaQ=dz|TVmnWV7G~shHZcdUC
z$k|as<>&cLl77BVuc0FL=5F!la1d2iW(Q7XmlN0Wr6k*ts~TXpSc#Pfw3#B>WoJ%Y
zRefhfv)jTKl8^1Ntl{y4uwOg)nVe<94hrn}Nv{%;utccZAR>WVoEuv4MX1n7j!39a
zM#&k*+`6e<t_tX&>jp8zY%mPj4JS-eprX2hQl3vM1~bzbqm$WF9vRd7pZqBwIaA7b
zYZPH=l!9g;!k6uyL*Fo^KS^^qLWXwjZ<$*l=pTG_>8H#;&0Zv3SbeeB5dMJO7mapF
zMS$+a(=JuMiiY_jtSXq^^Q7C%wd-rX80mvVI#CUQYdh#9_f_eu3GdYJA{9gM^?xT!
z*e@{Np<L60uL5xf1+HN<w@M_jHiD5S=RQCKS-{{vsyccAsQ=84*6i@Jx#|R3?pMCV
zQiAh`N*|)-0Kj{@f>ifYp-P5pH{f@>_SuT}<iwXesT-5llDu8AUA-r;3qc`R8tf2B
zeMHJ>;1_xSW6}a}-IzU;H>>)W;7xIVrrI<G!_bCaQhZpsoe&3$yV&fK7|GsG=QS1K
zYB6Ltq+oRrB`_+0i(iZGCE{|%kwJO7urr*pTa8QH@hL<p!@mjeyH(%b-i-!ilkah^
z&z6_)5q}yWex~N*cbAHt4pjf^S(El#D38F`hr-3lh_cD)L)N|*BYL(JKdP#_-(G0s
zH`Wo-QMO=POF&3cr02Flk9fV~Q}S{|PVD@^GtJ-KmD}Lfbg=}VLxfYbgW<dkKs3EE
z&Gmc@E{Uug%QX$D^^Z;{WX4LMVqWrsW=-0l_rrIR@$yp|2BSY#vPAO4zTW{RFE7?`
z`8b7-Ip$^aot?tWiZ_$#%#oQruLtbFwJlAtva<P&*atyY>qb|DH1{d7orc6&>|;%z
zlj3yZ`VAbb<eak_`^1lGrqohx{IPde2AR=}L03Y$TK~OUXxD?mO>o-G(-QFlUm|%D
zkZ)-;VRMD?6vg??#RPRqCDCb`1OQYzWiH2h@=?23G*5qZ7^CubnUpAZ+arP+mf`Zg
zYcy;-YJ}zzCW#=9`RsuOhdr*4ZgO~8nI^`*N<VSKtjjCK)X;a)BV3VhC?=v4VNLCd
zeetRNI8oP$!F4Q9CuMZv^t;vljOA0Gw>P8$1Nq=&=mkKtLA8}7Bp$%;<Q3CIt-fh;
zkc8o>HMT`?i7@7vh%iNDF3S+`=2FnvUX3VQ0#B5P$G#=xTf6RK9w?n*OTAI)W}8dl
zm6m;*y}jEfnCy3CMLS;A@G_B_>}n|vkE+pIio7h=)+rtE;L?4@_6bSKc)E%$l{cKc
z)56|$uhHHuKryu{%<+Ygdi&MeEO<X%f=F(aKAi$Q4GEo&SX2GWN|3OJ_*D&B*HDc)
zft?kc(2}mugkq-cLM8y?Mn2Vclk(2xV45{3MRc6<15Hb1bIq7l%-<$~Wdkh~r&S{G
zcEV}u1jFREdyby9ThGL_XG8sO>PR}?n4%qqX^3;EdScmMBcY;%b^pKOMA_N))DN|6
zVvl_ov62&Gqo{W_V;z^>cB4F@_`fr`M+lLsEEa*V%pC6Y(ah0oAOSSHDkMBBS<fPw
zt_Htw*>Sfifsx0=mNc=42DU|Ml@i->Qdhe6GKisT-ky=*ho#`A!MfxdNb=*M=OYs-
z5Reo@z|J>Nkb^l`tAkoKP)62B=kjIvFX35=dmUw4@#pA)GE|EJg2}xEF-PlFF0$#;
zgUDnOW54U6fe01-GL7sGc?gkbavdKl6?7hlSqKu|sB<t&{wiB!J_BHhBJi~?lYcxr
zs{eT&Bb%7(p{`4dYkACI4qUG8w#Xrt#AUb0!tKK4*Hcrd2A0cbLf5_js?!Za1Blv7
zEC=&{sfWWh3s*)Lg_E*XqgcA19yI8b+l3_uk#ZXl-#cJYjDmd+CMZqb203p5Zi4h?
zi);Z3>y%fZ<mu#}r-AJET8X7iHHsUNfyt|?`HA8%vV<z|C%ws{bz5KRl-uOjDUYEs
zeSGt==;E8=@CVNd3YZ!jcUC&I*JJXeE3Z<AA&!@!NN#aIrv!Vf?@o7(o*RFvw5jb{
z8``oAU*yIVzrzkXR5K#o5uLuKb%PM+ZHyWJ)N1&8u;R(V^StiRQ*j}WD@ySY=zq5b
z@r5Ax*>-zLyq`I*fPQDQ<WUbIv2WOMxZIP_-q+n^VM#u`D{I^=veQxt7|m6`hcCax
z;}aO_7^0A;KNqtrLKP6<O!t_az5TFnlERh`5a|$2ZubmezW>Psdg)JacCbOL@qR1)
z07u+iDnf<L&Ds!tozP)rak24JCKjKu%kMCG9qIto5u~uis$G4N+;mT(()<aziyUmZ
zaHx$Y1~~`!Mn;o7E_FV0gNS)lxE$Zom*-v-HtGom->&yGTDG^G^;Q8$>F?6^Ii}wY
zH>`Y%?~5rERx^@mciP4JV^hs><5z*?*d%}9@22)0Bk%Z6*PX@+JOi20O>Iv_aId$I
z`pwZ#eeX?=F>5nK{4sDF&pB$q{6B^pQzB7V{#ez#8J?YPrZgtkB(-K!3<$U!bLWHp
zxNRk`dcH%^9>Id%AYx`!!{5SEy6?H8N8}?2tOS?yVfYw*iL~z>@)hQg+?+*jhBRA|
zXt-krpJp!s@0CuG3j2dr0s7dYB!#Sg5*q3WxJ=t_M^oq5V_5*_j)`^GmzxeNBeU|?
z?K26;ohDr<BK&?`apb;ljZ?Bv{uW)nJ$qV@u?tM)gz|SaOEQD!jqQf9S0hJn&l`MQ
zun0IA%Sn<5KLt4IX|hI}jiEK_#70IyZ4R)F;n;!X=`>Gl|Mv1eZFb?Jrk63+HDJT&
z&SwvNa_^~JGoeZ_5s{UBxacHN_99a}^@F6#$UxRqAqObQK2+1TTnseon=AUXTY+wo
zuc&aBdk&HNE$MNI<*?kvWy2^i`hY-LGKJIiVGToFJo-SRuwLlX7>+q2`leb}f)px!
zocGLkQe(>Iqsz*WU3JQA3~kWoT)ANkrPPOJt+C<$(B0t|n~P)8Jqsf>ZRQ03W)le4
z-u`QyzhoP(w)a6F#yxa`WZToAC}jNKoJ=w+oChBraMiWpdO5LXHFkmhc#AQzKftp<
zMVXB7b2N?{VKl}r#%gjCxs{T3>_b_y(^fcU_|QD8_A}Dxtwg!&KrA$*Kg)cfLZbv#
zJ}qY0Vh-T=M_nz8q16&{iB}AnLGZP-jF-y$fDqygJ4w|U1Y8|DdRtr7v8?M+Cm3=`
zvgOaTXr^6A$If&&7N~fTGbM#Uh)>YCzd3{&p^Hm&K7^lZXdQrkRVb+6=CW7o{7mB3
z(MacT>>;;z7S88$``)@I6n;+WuWqvvD^c$^$5us_i>A`DFz`7=f53+Bvoigv2K}DE
zqk!2Uc^&i!bcYz{`Hh9|Uf6jxL;n2_GLFeHGn26gO|(4tiN05z@=_(K@!F1ynKXx_
z%3_<v$zRv6H`fPFRp&qxY)x|Q+8nrO{uqooLE`n3TaD)2-L&-R`u!^66bkw&Cz@76
z_a&*3Qli?3V$ZtuZr>v98}dj4s3fet6GWfO#*(G8s+nW;d&jWSYSAlk!huM5uuOP>
z;YPvhk{>cT4kbh1@ttHT3)BRnZZUuef=?r}DB1~iHVUR!me+m$y?JwrIW(%cYhi9f
zq7Yonk`e?Sq`Zf1bSzzfW~nit94JOLrX$7_eY75a-L$zolGsBLHj6#o=THk(!^_?q
zw!RZ#f0N;`tmIchbXuCfc=oIzr9~E68Wk&`{CX7}K}2?4G4kC4B4eKKUa;&M%CF}w
z7cdfvTGH2l%)iHXO<Qar%<4{@w`acA)Ay*o2CvLznT>!GF^xa6pa?KUG(Eo62IjnR
zD2ZJtT?b!%GWGVfq=MVYm>&|o7Oum{g|4DH8$t^?G<uB%pcAs&0^gOv)qZgKk!DB~
zPPpYzuSwzBAF|i#62TP;!bU%*+DI(3X;7Rqvw}KHjZtIWhMH`;yb1tO6>|!Ue_MnC
zza`q!0~E$-#W3YIqoJU`MYtxqPl;fQBX&VmGkWk%(4g!3V%ZiKTPQh^IKu?dP|=n<
zDEtk#Y9}6-)cDu|`)#(&dOq(_K8xbwt^ss^1S@ac7w;nU75kzMwO9r7*PiEmqpb$b
zyp2R-B|P3-e0rH4B<h3NEm&9{p2kr{e&~~dwpw11qt9daSQtr4WE%sDu({~G!Tc0i
z$-#LlwI1l8Typ}=McHC|E&`Kyhr_8<7^23OUXdLXBJctmc>Eqo=0ChDA5wb~W1YHK
z9oxlre3sa~)v<T}f}b`XXs^Nfpw%i>YSewgV~?m^*dgk)Vwmbrs+BBBLme>(zZmr>
z*ZPtYQuwpidAQ}s+bwa|3G5;GEDM49IT+82E@nX12(i(AvV+>)`*eS?T-NMvpyEX?
z5LMd&!4M7x^4_fP9r#k9RwATe2s!-}4o><4+ebl_{8FsI^w;$h;%>{4q=f9&s#otP
zg2=Z59;20@K10z@bsuk~zE;CO08f2-p;jEDe7+anMe_#)1aXN40=E36+5y+I#^oWY
zT@<667}LZIH<vj80@dCd^m^(Z?4nl0I<IL<@8ex{*OPrw*Wpq3eH@SEOzzOYDWY5|
z#fvOT3=>oo{!Pwo^o8+KAqDZg2Frt%lF@8n0yvVLnCXoRj9FZyopz`bt}!r!vV<8g
zNeYEDo!Fsr%HGC!9r0=m<TdE_)y}&;#nVzDH}L6gElYII6XcO$BhP1r_1Ou463R1|
zcdzzCT6{t3ZQ(r$x8tz!`ehi^=x@epvq~g2C*JFJ(0<}?sYD^FDxG#wmHJW15N7us
zLHkIeAY+AopKf%Tos{V)ej+JELrfM6*=l9;K0jF7KgzUtg012qZTybA!E)`vX8fKa
z5AcxBs|8=k%9rS(zmSe^DxH)?GcA|@eA4j=On8)T0<)J%xu$Z~W9=HT9)=b3F5lwr
zDT4opF$pY~@!=~Tfl6fVS7^HvPfC{XJI|51G3Tc93d3k*^(FDN(Xuv%Wq|r}aIl{i
zLaKiNYGq7vo1&f;ArBZ4-YF4-h<!Y%&X=q3?}QB)G9IH#dLnrs%Klr*=Muol*S7YE
zayD@=_QKZ3_CRJm4i69CsJCyJcS}{$pFhb5+3_}Ms~B76C`Tau6auyC=4Nc2t-G0d
z`ygY<zeCb{2v2c&b?tU({JRC*r3$uE{)`RWoI?1V7??Rn9Vf6cZE8fVKlMD0rH851
z1@|Ul`kvZtdl+ssQm8B2Q*!d@>V&_~#$l&}(FhJn4cX6vd|@N+Sg6poMM4Y~*-vAe
zdvcvITzUmb88@F@TOSsssFqo}L;hd~y&!=^hLEfVpzVBq$_R5M9QIWC6b4l7gul5r
zfWl6rSUJCdqNJjR#EDaII;BZ|>CsORPpL&wQ0FLh11KvCSVciFP5)-Xb=O##3;==u
zPiQIx|KZ$1@3{WJC$o)e6Zn>B*H}}f4!E?(fSpJQwfVv947Py8(X?W;^~rD{;m_y4
zmIl=|>5%rOIF@)j-@zhq;`0RSY}J0PB{QX@yX6rzqPU-@zY`3aeB)TAs1;+hDSP~p
z0rvxr52r%^)e-uk)&5o{1&i@^u`=-T!TIA$q8CF<5}4nz-D_K4D)1^x^&MqBJBnob
z&bgoC`RiO65!iLIh>~CVZAHw3Z7Qh7_S*=OxaHHxi<3UxgLdmd3uVd*ql@dx;16@l
zrkMA+qXdO?thN&kjlHD6fvz8w-Ftv=qiS|TmhM0gv9`CL(AXI8e@Ldr9&%RY$a~F&
z+~@hB@*{zKX<5{ukd_c<_SMf!4Up{sD2g{JhIEQyn<-@%d1Bl(<bUv;>KRrv)?3<&
zi*x@=mTwauiz(kx9w7VB(OhuSyoSRMAn*j;ws+*nx&r1n!OmZ^XdFi@TclR@nQolO
zk_n(V*T_N1cy5E+!PBYiz+}Obzg;oPhsto23>SnDj`{t~1|4jmLW;->WT2_vi^FS&
z-cM49rmeXDGj>Y3?qL91*~~Zx<SjeXA5k4sOmCy{EihQy53M?glE)rQ9r5>*1zHbX
zE^gv$8}DqL(HD4<qlc9q+@shjI8XlCUof+!+);iYydhi?f)49Rw=AX|4VlENch>o|
zE|j%NCS1ykq`gEVCg_PNsf?ad2`DuUk|xgpig$;+r{of|Tm1=mvRC~iOA#`?ud_%2
zW|e7jJuFiKsk%oztAeWJ;3aL3!-XG(jhnW@EkrC<ib;^8c?4AVNoy8Y@vGiQi(9Jk
z@PbF5zG%1c@aWHx<-0N`6ul`I|82!HMpePK2Yq?6I{KQxKc_G$WBCQEB=Fl$=hje1
z0I8&voE3``SPNShJ#vQY_xUIlmVRy|I@#CyJ@0edl%xaMck4*Y4DqSwQaj{R#ZSQ%
zg7ua6VY$y|4byyN+0t*233jKnF#ehEs`U=a)%G5eje%F?L(yO8^g)_t0f|x+qm=Yh
zVTy(yb>>2MvPfD3PTn)pt@@NKM3t?^sC~XzXhRXF?6+f8Y!Z29X}4EOf^FwN6;Tb%
z2Io7c1(G@3quV<;_dkn$xLl{6Cx_XZaJ<i0*Wv2khim^P$Ae0PKE+FkcfRhITc$u@
zR&{F~^fg6!UAxBCdoJ_>`lE(^#aU?_cvywjNrjTdJ|hfmYLdb<bRJ6ri`0%j4yM6c
z!6ft0&_dBNwDPO&Csn`QUl^dQpBAG(4i|T>vpAKaERksg;=`8J=?6HjfTDJol_Sn*
zk7u-o$)RAc;i^_Tse-a%(B+NU!QmGfdH^&H-yJLF#aHPr7q#HNo6&eI{B8ysTC@JA
z48#udpF({MP}go3h(bdgpK3d|Z>Z0pWG!dtZe+!Tr>|~Ar`LF$1HEkQnihH^-6#={
zZaZ@Zva*Abf*VU5nSZ6YVs0!g5{#$Y9Fln$+X?MnwT{Yjh)1fNgbPoT=CKX<mQrsq
z4Z=ifpxrBM?H_z%!~B~6>TAmwg?oq}78N|H6rvdYKvV4{9MvMlw0)YBz0mX$=vj_p
zb}?Rn2^fET_llR6|G4EBaFIve6hmM6v~{HFkusKUgUVwwG=Q0couDI}#Y>1k&j>5H
zv?V1&<zMrYe+-%kji8yG>~ct655bO~byI4`T<VP4ia+Z4=ld80^>3#~U{mF)4lO}E
zivBLm^m+M~9g0!?JJg|4uNC<FN|5nnem08@zB4IO{%P?at&;JBv)Rlto0*JRV-ajp
z(ZOUfzs_xJ1}O@A)?`)AB)2uH^pr)PVs?9JyL(`e{LB7jjKaZPToaWJZJK5%C_P7d
zg`hIGqEg?Cd#ysscSNAj0n4C0an5|5*prfoj-GN1jUWaPc^R`#_h7D<l*8~UJkdVs
zzq7V;<h=G>V9G%y!*X+^fpjS>`vT9fxsg^wx$w7FszaM>?aj<qR-L7QT}!c_mkY>I
z02&<sau)Pjy9W-NH_3utUE%yP)-1@u%ddiHYGeM~+1crMhtuxOQHZEhc=o-vVUFS#
zZvn*IHg`wg%I-+cCh&%Ue1hLkwbl~7VWf{D2MY*OYs^>aJsI}eumcp6{X%UUDGaZA
zc5DL98;DL=;WiDCzd%+(-GJ4I0fixk5{jW$WSj4nz@n5a%ZK>tE4|DN*D~ofUyl~P
z*2<#DB7xaj4R(j~!;!o=`oa{W0R2p4C;6lS<Sm>oSc)dpwWx(7jpE`os+Si|_s}Y>
zO#`mEuGjv%ggl?Lx1;2AQ0`jLgg`J8Cc{eQJ}yoM)Q*pm^hz1@(G5L_6akO71L4JP
zbC7{e=pOg{KhsKiW&}U$f(F&;6*df(6Fx}MZ1t8kQsv>+Yk3^Sq|?JGtK(;y90j>~
zIaRFplN#|d<T1YGO&J>qGwBD3p_tLBkb+VBjnL}P#Vv8gUh_qqMamLIDLXtTi72=D
z9XP53HI4dfME#?Z^)L98k@Qyz9coA|-_&HPZvUpQ+xHuJfkjV1Tmj-~m6IcIU)k3t
z$G5v+>$ik*r*o@gkHXVuTFDk=FQOlb6r2&Y*T?_*RE}nEe$BnvhUX?7X&Io{LertF
zm?}{6@hYmVN4F%n{B%oppPxooAAgYzO`R74Hp>W7?LokQycqx49j!ad^JU$cMwR{0
zR_#7b2ZZ)|O3`ll(@>;|ecdqY+0Ltc`-(0^!SfU=vzwNSEC79sUv8eLMaXcKdSuZX
zu=eo}D5Ts-ZD+C%jBbz_JrNKuvQ+H1NV~khR%u@8j%v|XR2KDkm~B@kz&Py>UrOm7
z(=l|B=?GhE&j4%za#)d~d07Vm>wqHs9Xh0vOP+yq(q4tM*OG&byMlvk7iU}6L(Adj
zs6a)l=EfSDRV&4l)`an};*Dnac$roMK@X*iy}N$PL3rsn=Ng~Sq_wg1`98V)m}Euo
z7D}sYJqJ<RLU;j#Q$KNC*k2PKcT6s7YS6GRCn*%?7?+SXED7WGj*&`fcbcr9`Hi9O
z4+lDP%LzincE81^^RH93&(jNOmfX1lcVHbTu?UBqRKczhQEwmQxERzuUYysdSoZ5t
zetW}VQ~UEZ17WM#&dT+ZimhFptAdU#qkO7`B*_wjuT9axh|&h-+?l3ghOSzI@lZ7)
zQsa8NrI+Gx(N<7Noh)Y<??9^4%A1atxF7*4swE6#J!x2InnNMj7-CtOGAKgo1Q<}J
zL4!M3^C9%whxO~QP;%Glm#*cNmq`O$@w)D+KLQBwhdC)h_HzFJF3UL{^{Kqb<Z+Y^
zw4q-*KX(@ih98YVeaW4aYl}~PS{ok|0XSzuaa)gl@5w>vY50@FK<!+|nl$w~k=@@(
zj_uX@H#36sKEz~eS&6m>yza+pIEAfMo``W7`*iN-4$((?-el}sa;M#ASbp->ZTmKg
zxsjp^?yXL$mFV(u@fEyHbZ59MBlY$-!uqoU&`CRRZO0S;aF5K~JBiwt1jb$x_$<1n
zuO*lNR(kPIm-Y2#m#3dwtlAFIWC=)Be%m8We_1FO^7h$L(73s_*p<0W{%K19nj>nl
z*tK+$`u01!Rw&bxmQ`MBcXS>QaHSO%@iHp?`I`Dycu$GpUjkuxlN<Nwe0QuU$(hyJ
zRO<Frc>naJKQW1+Qw8+jMeIh+RyZ$s(piDml>>$<=t6ZakwjRW`$_DKr=RCmSIl12
z6Q;hOBv*>D{E2^0#z=4uMOA*=h@yr`$9i_-?lgsYJ;JGe;+6dHXF*fU<D{Y|5-CRS
zUdWL6?i^x*G^}-LAKpJzdN7apmw;h~wOPXJqH#fX-rjWR3pW#)Y*1}q+$v(#Mh7O5
zg}LLPnwTlAMDfh8SaDJjCcZ@Kb~HeyH&Rq(D3`*HQ-`Hzc=T%041XPSr&>g`MeGI7
zpjJG<bLXh3d6KKrL#Ly&zNhdY35##7abp3?%}mgZy;<9f;6d}D_BMe_8pv4fxZ-O2
z_cxcC4dJc}0<ZBlTiTj2*P-JE7-5G?BfkBhL|hXoSuM8~8iTkXTU}M*s~=P~4NGwy
zTiJ|xYjqAc6H1akx@T8G{pa)htmW-obj@YSAtLQ*E++6uqZKC|d;8yC75byaM;@Gh
zC+5;&MNPTGnvmox!P#06JG3i7`v5qf^wQN3$MLKBKdfO*haZ}kOL^RE$RtX)<rvu7
zkxF~`?B_zGSDygqEcJ=-6sC`1ND6nLKuW(n-P!U#;1}I4l2FxwwfVu)>}4}gPyFHR
z?9K7LBr6!p;Mm%&G6sb+i3C%W&9?Uoh=0dksyRVB*VZ3`=P+OhL@l63mW82C=8u@h
zCP2)yInUk=rD#LgLC=My&&*htE6RhGzDA@Jwit+~#Rgum3=Ul+v>3W+&#8H<y`CDU
z!VLmi9>{0E^7=aqn5AHe`7Od)`ATCl9&liT;7vJyjOMgL#k$BKb^B#eShpYgKGxq3
z$mNOd7}oKie~+QT3S&0cpc{aK*M7rbqXEXS(kcpnvcq(NtEOpXLn=}Q=Kodw2)~M(
zCfW*o=r08XtnzNqt&{JesDuwYvK)h8OM%Tb<z%7_qnH7($zDo7v`+Y8avC#+@Qf+{
zL}^2I4(kUHfny*t9dijAP#<8NGec<^M{>f4^*xa%;VY^#LBh_*Z_7dV$bj~9Ogc?R
zrZmOj0T`P)3tXks<JEpa?@u_aJp!c|__?+!7lIl<NEa^4@UC>!0JYMp;b%28Ac9zw
zsT<uWD93QSgkaqyqF?O!3-aY3T#nB|e{)SV*h#j5lhKJu$OwihI|?DTC`2w%LmRIG
z3%Sn?uW10*O*8yvvk%>d7JZM@1v1%IJ&G_SAUN`Ix2r8g9T<*s%3&-tL6(+Fn1dQE
z%nil{kT2bZI;6Tj4HA<oIv$Qpn%^+{LGnMp?9bJ$T+v2i6=X|F+PNSzn}dK2XWFi^
zPDQgRVcF^nvT?+JSR(j~aueM(`FrB=pZAQiw%SH~?N-aggtCe^^m5}%lK}=<IlPSj
zzUSpBe%P}~D7gCXgcC%dbBhgE2W$r;DE;t^nBAHCYbcqzX6RK|VVPac26#%VCEvMg
zzJFK%`Y#h*hxMy$7D*gROvTo)vzA$pl7&+ihT?Tk>e3BpQ}U$ppJ9;e$0@}c7-kH{
zWB9jC!i^2zz~;OnQ0R$zDI@Pjw6?RPe?!!})Wb*{`~2zTX2bF6H>QAoKBD&VT9fW;
z_?73LOB7B4l;(7|A}Jy}LGPf5>=efxhuqpb&acvP3l6b{dg{tCGLOA4m64w{=B|k7
zkYjsFuu=M6rTjY52eRVc|9C0GM6nz7JTjU>=P2olG9T=0p}G-%L2WoPe){!@ZvQD>
zIoa-YbYjZu53A|dY809Y%JEoR&1Hexf3FYcG%0MAXp9!BI0$o3Op3)jxh64}l;b?=
zx<MGu*EIH~=aGTorKsy3{K43}0<}aji2f01hFh{Ap}UFQ9A9QXVU1c2I%@c^RIHr+
zXyS+_TBw+|Z|HfH+I-4yguZhnhLq6FNXX)__}x4E8}j8FT`(QBz3})}uCaMI-mS1b
zBrg&OQ?H%5;NU2<bkz&`XiDYO+S~4?VpKSsX)wo*BzwjEqJ@BS(vHBjt1G!-xFD5;
z`-)I@m~b@8Ig8|ISD$X>LH~6h*8C<%em)<U=B@;jC+V+r6F)V|iCRKsvcJ7N0Nqaq
zQ73V<!=7YJuq*Mvmo~w@?(gt^>%*=qS{kPY1KYkXoBSy#TU$m^QtwFAMbZjb=EhD^
zEGJ1_$@{lE%8quGC<HoGYZky#vmXX?q<Pv6SzH`aD&+B_LeN5mjSJ=)mP2Mohv`0m
zo<``H-#^E3!v-6|#_{AB7;>V<F?2rmzsB}8!zDD`rk}8Eotdr%>BW|i`&+GUqh<<Y
z!edE0>V-cpk)Zc~raIfS30R9E`B^(hGdA-_g>(*ORuVm5cSWOZedNm_Ye9#zPTX1F
zx6|Z{l;lV&v+L|_!2G=IEtu+9(~1|``!Q4`Ls#!BouY<|<<eo;x>$e)C7R=4`Jro`
z2xY9KtV-W(Aohl{ohydr)(OjTb${Z7JKl~DZ4xxcfAIa9CQJzFMjm?UZCa(w_(O1<
zQmrQ`(?hmp2jK?JM0I-+B#s=_l(7KGeCOfSW<)k2XsHj*N(!1Vza114L?iU!BESE@
zUKI=HBVR%>L^+NiKoBrp&1N4%WxP>a3z^o@BGk%Fl7SvoR}x!kcT2s1OB_g~m|<hU
zMmyN|eQh#PLZr<rf`(IOSA*N^GfOHI-TIT=L>NlQBP2cJ@9cTMfj5J^dH*OnMphJ2
zM9PlsUxf035dK4NrGEJObJi6aau#xIajy<2zxLBQs5rX`digISVG5Ph+WJAC^wTmv
zvlCejt`KHY2gy@Ivvlunw8;7atfIDO^SmUe{SE0zf2#fNXVgYks$~?QL|c<+pMa>3
z$Z`b<@*m6nma)??D~O9?FfWdOjSm%ThYjNhlCg(5c&Xbp<)n`3jN9_9byh66mN`^h
zs2u&m$Pt&hJkv9|_J`%0s?*dD{>Xv2pAQ^TvPLg03U{X%3Ih<EIw-W=Od1&Z$+d(i
z|0C*X62UHSKX>NP+RKH4`QQB#!{+fr)`k?X=u55w)lQt+-&s=re)_;f&`xQT>zt`%
z*{%D5L6CZVSc6T0e~A>t@^X^UbK-w(o+?uvgfxkGE~_i%CttQVoI}MYjnsDqUbvL-
zH+jwTTIz`(OX<M;?vlFD3BidLt0iB2@$MDi7gm(1yl6NYH0f?R$@+mk)t8LP3Rhuf
zT|jPD4TY5X_WKe_lBS;hiGygV<F%((z>C70a4E+@asx)Ua@d|*%m&O~2QK3%R9T@?
z+V+Qb*&dJwrYMTN4rj8RbarI)hxGfoeO5P~g3`W;!8JGz=q(bQ`57da=h^vAms2g<
zVD8)@kYwE<|1-j7Xza5l0-<>Wbiuu`Zo(&BVh|9dKoxZzxu!756k#IfPq{5!sDfCb
zW*C#CgeipV{S`XuI=i;S*ZFOlD4Fpneu{Ub_r=Qo(koL4S(@<S5%TKuv>V7bVKGsw
zi$tyNSKlEU@VY5ld8JEjk~NEATMMSv{{D)AZ<P6Z%FFczWUloSz%To*Myzy*6)3w8
z=!3~(W<(@w!DE&a?f8YunbT&9)($I58<kN{{IeO^P4<v_aD5$18O!iELL=;$5I3T=
zqW^LVxoRFz0_Y&mx*Z7XzU@0^A@tq_3=AKUgKpa_>H*N=tdliKOTa#>AUX%cQ@>jH
zZ<{9J4@rt_g!we{2z)XoC#j?eiXxS_tvVj5i(}SdQwCAOn=-MtU=K+VYwAT$L4S!v
z4__BnZRo1hPaPEpp}7Z9vzwGhu~YCTQ<%T=Q(9b%bO?66r%=$S%1uy<ZB_L|p7q$M
z)7;q#QRsEKU_h*^V*yI#l8s5}O9WDVu47g74<oz6$}ZE>(z=7!&cq3!bIsQI2Hp3&
z=uaE{Z-@G9<29)rCL5s{5E)k{-1NRKAM3`53X|lQ#AA=a4Zh{?$L|NfmU6a+@$AL~
zSc(|hrWDz!9-+KQrca_&z1;0qx!Z&SP){-|<0lQ}xxBBN?N`jxNa>OeWDe$vZVgI4
zWi1-%*PP%>f+ySrW|MN01C}vFZWA+ICS+WWJ<O2C`27=)ze|EAGusRqj-~_O8}y2!
zC)jVKl;f;83H=NPs4EF>EJyb0CzWiF#<-32N|duBj*!yoReHg#SW+>IZMOr~IT!SK
zUeuPfrLq$T8#Iv)l|1f@X1)T~$!vV~fpQ^B1}AH;!@q)a<c|mX`9x7)Ug{*N_85dP
z^E0$1rl)9k*P0~l^8^#;?OUN^(X0{dng|LO{B*q&cHT&%ILQ!TCJySln(%a^orJV8
z8Tb}ez_JhUiNuM_a-=K@lycCRK=xg(HLP)V$7Oexq&%X^k?y!%1gwWXn7^=kEoPG0
z(SEb~G~zyfweIb}K)8u31)TGNS%hovy6t$qI81h<%XpdHKFD8PIr53{BjO36J*^Jy
z9(X&ya==u(Y>uv30}XZ=Yj_jsHzBPcpAgG|uS{PWV8d2Wmhi-?=D%dK3j7)DZ64SN
zSu_i@!ZOs|3~{p!K4CVLABo7=kNr@BVMkOwMhFISR&ol8KVUCuR4zmb!q-Y#?%FOX
zu16)$mfVwHA%<p(h`(y(VfrMU-&PLPp9$fc2d5ESV@Klv@o)?uTDxf^c5QmAipJz-
zr`yemd;^Ch7=$K<5Q@i|AXbSVWYo?r9G3VQH!Wg;#r)d@6bSWX@?#vx2ECs@NgBqV
zUL&o0<lz9FF@A`Ejy6I+#nC~%{0Lv`aLV=K`#pl3MN&62E$^l*A9l)7_gVZl`Dh(5
z1&tH8JAArZ6Xe?98R|8G-b>&c$z#e@?KT)4#T6=8@O#xEgC`C{PUfsuD_fr4p5S6c
znt@4_Z7Qto(*+uZ*=%9J@ak0Y(T1JzCl>^^A{FpY?}r~hDTz=`!>H84Y3q0Prc>!)
ztC6LfGb0B8OI~(GO2FL&6iGo1>^fZ>f-^JHyF7eR-+-I!tqjb6Es#6EhRk%D4_48Y
zzrE4Fn1MXZ`YEy;ZYFGLw$b4N(%_abWBA<KZb7=y&Uf#Ed+1k;@(Y-{fo-wAqS~Ra
zFZI|ghM_VpCDvn0))*Jg-}EAZz*MS~9|Yb@3i#%9-0)ndcZ1)lpQa(Ou0tuLPw$<5
zeVD#+0ydOP3V|%o{TcJAy5u<(65S2uTJMlCQ5NO^wei}~-%Yx;2h0udS-4M>MhWc9
zDXC9{(ZR(plz!QA_wiDM?IkaT#R~HYt?Y${gChb~CwWY*wF}U|^w@}LCq~?`bYRNF
zp5p!ZNffL9M`kE3H&T&pZ!`Y)fxmCOkkw(SdW6B$?zzm`6!b9#u89n}b}#8P<&PWc
z{SP)-oWlW)EvVvJ3PZ6H>PE_qu`e#h$OP#?zSL2P%BsBV!xZ(}T5dl|BOiDx+9(u&
zJMTJ>j})=JA9YG8LbXNpy{VE=x>3G5BgY{v?1;A!03Cx+6-qL0gUH+pJ3dkG&!$&j
zsv&DRLZ#x7e77km_9}z~y`a<Y)8%KvJL)0QT{nYaZjul6P?#CY6{J^~snw2F%@%`c
zlO5&-g^yL}SDVaiy-Rd_JKl1L8gU?E9xv`CVYU{F(eu@$id4KK3YcCn9X2Qa=`=lx
z<l{L)%5^r(7}AuCVA@w$@{-1;pyx$@$><4EZ(2*dOK8>}6~jN5D;tB=-R#!B5U}`R
z$iau;B{^_fa<!piLH#uOZOMrKHvi)9u`jS<j>X&9I1h<K0s}^WdrUySKJfh}pTzOo
z^GPX5zjDRvHJezl-vtsdQmwUrZ#v+fTdI*&IXWy&bS0Tym2W@Q(tWwsY?w*8Um5bc
zXy8Otao6d6vtKIqc48K@%OS!#VEB|YKn=Laksq6qfKdc1LdG-Fb96yqrJmx&u<w1R
zH9c(q_Mp@fnVW68d+$U(&B3TDC*pq@-nK65NejCx$VRDJ)%M97Nd=rR^^Y<onYZcB
z5UVO@Jh^=yC&zl60jD7Ph1m|dS>FG{-G12YI(Fc{lQ6?j$8%P08xFBULGa69u0itP
z2&0L?NKg_C*U)Dsf89p_6!;8^(GwLG%e<^BOKNyB<l@scjdjS(dNh${mjuj1=!6ih
z8RQAfar$*{&tY9(ErMcSKn+6&8q@7La3n#fVr5q#$>F~5>$!34G0j?fZy7ol0_1dr
znroLy&hV(udWT3%9t)J;0tX-hBw5Zj9e%3|)+(la=Z<D!g^*!mdg%Mxf*xpw-Cq+y
zyE*Q`UgGziwcGRc4v|fpI?yN_xI{LNd95fwYQc#;^F6Me+TEAy{#&zrSQoDoovESV
z)^T%EB-T-%pmDTyN&DW7Xgt>BsaBa?RsO9<A<v)jIme57pa1J$06)eL_PSb2_}c9$
z*_m3R>3^q68;0tGrFD<@IBYq=&O%sDtC8yHW@ld;!v}?^Y&f(NI|;DxuUz{^L%B)(
z;h62|ez&3R<EKfh)$lWrP6ZUrI<oYr%*EJ&s4st$M>ReK&E%2iT&*laQB<CxW7?|<
zu>8rE=&Zny7m;s7`=*%b21{3GjBMd{6>bOK)<>S3=^E8BU>r*>NKx$=Z-wLS{3!F%
z?y((Zt9wYp4?%PaI(XB{FrCGqy52j2(P=GLBU-P+eS4<Ck)aQ2y#h7Vr0w5aj_9>w
z5D#^o)3|iqk$Ck2vi|Lh#_S|;!y(OF8P;jp>Dp5GVJh-0Iw~T1;Se*Voyk5m>sH0$
zbM_*S<5*4lw9$fc0MZSQ!IES{L0Prz{XQ{dzPYB)h+z6j#NSJQR3@nG4#w;!UQ4Ga
z6!;vzuPUgx18yHp`~PF_9-k}wqP9`TcG9tJcWm29$95;_c*nMFJ007$ZQC~c?BA(6
zRp)vCh4=ZsKJD6T&bh`M_ceCbdD@7IUNcLP&oC&v7WR)WS=#LRQP#q_mWlE%sZz1h
z`P%Sbx->73c%4F}w7rzV8(2dv0ucI)nH%m|h0>VY;Bm+g^kt#7h!&`{t_RN}$nBPw
zN*vrJ!|^5-R%GE|%vrtW6;E^Fz%~vQjE)H2vf;oy!TQ0TCFrlSDG)VxOC)l1yhk--
zR(kag{+4q)?%)L0c6!`;gZzPL-BO#F*<tDxo8W1<g^OOOcv&*pONMGVm1UwvupNtW
zWTLMYzh4ePYaH_r_f&}Mg<69=qNHHa$Gj|y_oe41E++n`rc%~Q?@z&lr*V8|DQJ?W
zc$rHoO>M!sDT!%Z{))AdJT6=Ne#20ctfO#F)-rRwV^i+53=L&ad7U0+9|rme$w|qy
zen^Y*P(gWO++n+KRcw~^NE@VTie8hOLzbEy7kFj3Uar}dz|vY?SqT<XV@scCQAdn~
zwUf@rEbZgdXHQ?e%~C)0jx+%kO3XddiOCrANFwzB|6BGD_VIT$iu*}9cr7ik%h&HW
z7R^9I_V83OK30?ydV)~@f;}=qpUUHophc}NXH|`<f9FfkEXc@{gk|H$G@u*Zk?1PC
zvm<jP-6F@T6?@CR&GAK5Gb@%546yB9O0t~`T`wOc)oM#@HXEu=4{_Qq%o!I5Ees|U
zcV;4XznL(*_OI<Ehb<^)nIyqgyG%N(Tsh%#;fL8!#E<*)ikghs0P@9=zq5a`7+G7k
zTr+in?Y-tdWoQxjsoc&N_=_CYE(pg~pzB9qtsZHbp~GXxRzkhKwz##+1$9-UDCQ!V
z=c31Np#s1mzV~fG;ahrDtTz>eCds=D!s1-o!$7v@kE6GQ3%1k2PT!5&=4O}3yN?96
zeNrr6^0S3~Z!7zhB)Ir&5(_MD(0L^&6R2{)S^ufa&D-GOz#YL3MHFq&yABS8qYPwV
z8he84x_Xc>X&&n&i}D%@hi*Vo{Us@B7WWvKK?V`T(+`sPD-@Ssy9d^18ul>oA>&4B
z_c2!*<+%6^FTkp`@Fxj$Xg26yAtaxt+d{sHbEF1EwV32U|F53+C6mS!`KA5&)yzF7
zS^jRoU(t}K^KX8zSdch7rqNjFLqc?>y7#;z#Z9&QJ)K7cnzF|}f<Cj90HSQX*kq_X
z`Y@F0;nvc5x(~xCUn{-3tc+Cp6~%#vd~^}ksWInb;%TdJSItp$2(Q;P%u_OKmfMU`
zV(zpsE_!s#kAXd_WhepFHaCg`CBXBzwpHJfj(7V(g7BOF1fqt6TtBl2p>7ImU8ffy
zXvG{lDZw$yb#^~DR3Lpd9w|tqC4a|3TpCZpZP>W{O)kU^7>d8`gPyWpT+#Qe(n`;W
zJSF%o@-18-e?SzdI!m|vKyI1(#PQDev%=GMPrPIUQqUGwkFRwXF;%kXtRyoE9%{zI
z#?GfdSi|;s0)}{`epW`%^VtU7B^c9PdttzPxx)a!XU46&ohWi)To_!Z)f`i@*J1uq
zMZ_=l9vFKz?8+kGysT7_bnPwq@XT%m8tPo*L-})k%hPn_UYZCb7$TP$R=dG!&Q5YY
ztiZU~=Gjd)^dF}nL45P5o8-Le-8pBssKvRoo+vT+;pw2T5uEEvy;TY=<fNT3(R|0W
zGM~0as9C@xr*>MV8SXO~q6;Bn#ql#K6D30XVMzJZ4!TikbGx)7KP8^8p%%YBIbx@k
z*Uj|axP`Sb*4rP+`FksVz6mP<{$R42DF|vEy?EtQtWPmfE>?vL7U}1oE4n7#g~|$m
z2l9W|g5M8co2EUgXJBAi<Zr-m_&}?>5?kR0OwT~D=w+kZLUJG-$5b{i=jSa;n|w<K
zFz-G03Yc4$$3p)ho_7H=yFC~iL5k7=V{b=Tm_)y=8mZ*}!*P5T!>ir|3&r*gRoL&L
zjjJ^3^Kjeq#&#vnO^o)(K(Lr(aRjK4bcu9rIlp5co&?7pDVMH5b_OoL3zu+WBJCOP
z&=|-fV=Q#Oc#gnLEZ=<?A25kZ+06Ad#vpiekdLT6QWyqO@MD!_4DQ=H?HhT8kj{ES
zxiSH>3I<;uf-C&4@M`J)GlNv_TP~?4m*>_}LvkVp6ykLLeHv(eeBAUf0d~AwvOKTb
zawCUKVoqdBgE}IA#m>ptwH;xy4WtO{I5X^#@@0)m&C5H=2-=~kx0~5FjUAX3lS>Hh
zi9x+rvyH=T48tO1Vf-zCX$%HAF#2GLpARpi@)5&@HFPZRF-4}tD=`cp`y3?0tgxbE
z;&(knRVwD;Nn<P0*NBk!aeF4liMAx9jSyl-C4PCm9Y8)hE6q*ey>1m{Y~2meI4KZs
zAj|+bK{(HwI`Q#*MPk}X2&W7BcEch{O>i=7A}sq1<5;BK@-CE;eQr6>X7P>Vr5<LY
z)2L(kH=~s4d?mat`}&Q=Qlzn;hE50=$D8aMZWc9=+wm3T5%))q*M)*mMa}?27n=P*
zp(4VhCP14E=<A#^gFUY4H<(9jF^8a|sK;M3J1g^|uJo1!@mk+n5C2cgr!EHMtyv)>
z>PbKE>ev6!f5K>imQNvEjd}I|;Dt~wfRs<n6XE|p{=dJ>|NaU7_x1SSmEiyOHF6MX
z56Mjo7tZ9%<M+q^1mR+s!Rka?y?>Gx-}pHbarLA8cKAQGpD3YA6g%HQsVSLwR*J2<
z&Aq0p|Nj&CRuIOa_ULOMQh+B~iY?nm)|wSMXk@}9AizcFszHpxWs}=f2#g$Z=im4l
ze5yka0qi3o&L>1WVw<A<IyAh7Hm0a%l{^>!VwEwV8|*>BE~exmRT%^pYV^%G^lkUc
z122#?1}}Kdt17FjF>_}Cn*44v%Zr8(HHr5?R%8^&f$^rx4Bou7WR$4R_-i<YZ-Wxq
z<p#*aR<8ac8~S;D`lE>3*A8_x^1GoZ5BCe&i!`F&v#Yzid+1~8$WdBURP}9WXec2!
zm&xwsZl&~%k9#>6kKZLdn?HH(E<Uy@+_|A&>m*|Eh1XjBM83gf1Zl0^rDD~Fqi!Sf
zLQsRsnI}6TD~sOB$|@l<lMdKkp<M2yuAWS<-6#d*fEd}#C*y<zq0${AA|in885Hvs
zznHkc9v4K-8ou#3?UMcjet;^MX|Eq>z6paZHkeKZ({}*@G0TO0qC2_K_|QH+dsXp5
z?7P6PK3pr!hYT(+JLxp{ra?sByV0ew8+3y=0cwg3za^-<AGXOwi9OWQdsKx;F=W4D
zu^GqIHHZ=6ze$=ytfl$hJi_q=bOwE#3Df^mMQpY^wNzA84*ynE!|{VZUu{tRdz<kM
zdf9JvI5-@qt6`r>cxJ@QWf1(ni#-CK=#nOwdOEunrX0}xfa#)9>cbU1fF3pCLqR~d
z{hmlVpL>0I=n7Gsyg1{U9ws@+UONw=%M`gY!iW24j9{CUYx+;>t~K|qH|2}FiT8dU
zEkK@vMu5$)#g4=0L`QOfm>r;gsgWf>NIuZOHKpD3>Kj+9vsBwP_1A>A6VzB|{F<_`
zAH;jt%XZ>c2SE1U_tz1SM*F#Q2<s*JpOCYE=XMB0a+1qcOJ&5xgFBy=b>=$&f4Dy`
z8>Rqiw9A3%Dn^<eu2uAU?TdVzqugj{XdbmeCQKED56@idT$(9%%VZb$ao69H#S&ir
zcM@yOwwWZ=o$+p>%0RTO*r3NJAvqZc=1>d=Bxh$c1VSTo6r~QPZtznyTK52<wtC|s
zxO4&U_9$$Ye$y$eGoZC+tN)y;?_pKOWxf4hj8&S=W|jFN0g=GQY(Phk{#s9Vu;mE-
z?pm7$NUg~jPG)IL3%qyW<zctBFkqH;@JKbs67M}1$Z7xDs?h&o07_q3jVql70jbQe
zDOqXKDg!^*;{+>VWdHS065(_jZv49%ov5yT(_w@_MV}BKu^{2&Fi|n1e#h%U{%!lE
z+oHKVO7)EYC`om_3QPL_uX;rg({l_=uKJ9?M`h=2KfL3-G*xiIERVR#?I>9+c3@wb
za^WOP3nwvCB@-|ajv-rf$c2-UO|HPV-bwu;6s!~R7jF>Me2a-p_`PH(Q-6xRLw0Nb
zo_%zZL?PW}YEX%LYi&18NxVyOdtI0Dk6m?=x~B{|6P12-M$&q=%}-kd5+${Rcr(bR
z`<5YSQ#0XWa-J>6hp1DU-*msKlR^nSq&RHX$#HOS{sXQ!x7`X^JidSeQu0ZT=QN7l
zy`LnJx0*yOA;W(*;LSlm09f1md4<_%97!n7T}-g7#6j=*Qia*w{*XlZ|9kcaejO#F
zD{u+JzSP<KdlVssrmHjKi@u<ZE0b<DYTO-3O<nu``^bG$`dzG~%70PWFd53Unrh&q
zwe9=Q+v)OY#UM4)u3g%$=gZYeH9dRl)#-GW<N?g0n{(4X@NIBt2q)p+Fy0^Rie>MY
zEh-?(2A?cK;qNj%Ihmwm*Ct+u5F`tNsqO26I-KjhKZ>9|ZhJ>lYp#eX2ov?^#v`0<
zg}lr#sPyX0mWh6d{O|On|Le7zd$=fNq|>>$@r&o+ny;?nuL><6&|FEc+OqXTwxOJa
z;Pv&vbY9)f0mTHNH_*DiF35Gw2KM*D517|3J2nrL6#zXH(td_n+M-3#Yl%L+iUH)s
zGkx4uAzQtWqYXb67o0@oGZ5o5N_uE-=TWtwJtv=u75sJ!h2cbJscGI6rDt}V4nf2_
z{7;9w<+k6jZP5yyMB$hyddimZGlRG-oOWKjHFiF|hvjw@=}`+0tXDyn1y-!aP48mR
zq!pY;tp*Yf@58mg0$UdY?U^QOl7B7#n0PK2SHj`1i^u(wx)4$->y@Rd@gpgt6vRQD
zn*(JPR(vA#f`yf;4qfroaH}P`HR_&&o-TjEig*1T7(UD5;`*U+O~3YQFX1mw)TUV7
z$L!C+c24PLt4*Gp&IfXS{w+TE%I=9Tpe&ULq*zA(`x=M207^vAGz<*MH9#xv*rzed
zjllMI;|y8uo+UP-`^8EPr=4Sb;m2I{z}$EbH2vnO%Mp<u0)BR?@`#@(LDy1fF0S3#
zbWkzx-l=9gcNDARp;TeMV_2F9Rzf&B)%XyWflWx*{r&yqM2tY{a%6JQ7f@pK8Kw1c
z1K-#gz~Tt!`+3ambey(lonQ+$QLvPq^2|2Fx3GD-%+l?5=J6jEh*y<~?01W|@k^cj
z`ie+a*BTg^;$3hGCQ@AUx#NkQoJGAj+J2@PMT5DFqMQ92^Rf^moJdXrpq}?Tr4(Di
zT~_K;?c~Lh0o%lsYCT9`6}zqLt7pJ-@GTdE1YOBVMjJJtozB{G3;Id5347)(Hr)+B
zgVd>;$Xcl4BguN?{nQJN43mrKe6!yr6UnAG#n^sI>O?fZ6&2J%*S&eQ<S($_ShUcu
zP%jU!Zfsm2t)Zjyd;ANCFxuvzFFX0^4u<Uc)nu>#y3So-NBeaHJNhi@<L+vfKgest
zT4xZD#Ysu077Sm<U-m12ug%87(g46&R$}hAIfIu@5>C8LJAw(O*1SD<3NFWfAFcYn
z2RlW#%wn{jx{_*yF~n$)zErAe#j7E5OCQt!D36}5o<je}&RNDxY351}Gn)sn-e~0r
zWU?{<w*ipwlVO}U4BHR?m9Om8(yOz8<Y{vok$bGlDPgn44B^|x&bN7#(PZFohl%tn
zUm=Ubf0X0YAD@NRG6LDBNOL_8Kl?%t@qa@LO>Yyl8Ohh$+pXa`xX1z$Qk;5HKml}8
zcF8)oc<=s5La-{G%>q}JGj@Sio`^}i{p?^#eJ}0@gBtz8`$Y|>&%TDfSNeQ*@cR7#
zvY-tH*fuqOb4d?p)#i-gzs|YZH^00a>_3z}Ey9}}zaju1#^JfH;mLZZ8-!R)F+qoA
z+Ax*<e8lyL8<@hWa7ZpXAH1L0vraxmRmo?+%g|p!UIsPvsGCt6GP94P6ftwrMXybn
zY561kTDny>bNR_&xxIB}IFY`RatrAj=~XS)N-nSrIx*UPzJS|3$HwEsv89$Gyul-W
zNy*X`u!@<SUMc=ke|IpFDXphA$xSc!!qwhnKNy)r!@xpFUiLPM#jrCulTa()K!=(#
zI6-ATUcfI;zRA<lu|vOvD&JrVpu;Xa<HWB_ZQ)EFhy4na(X}wt1n>C-aHfu*k@bbd
z`pxP(k*je`=e7e^a=A#m?l*H2KUo<W)}C_H2T?kfXnijtoe?*%6Sv7+!5uz2O1U({
z?TM5C8&<Ay-%&H_#&M^`sO!o)Qhw1^Y13FMcmQu|Vo^uQT-PDBRy%i*$oy8uZOGmP
zy)Jf>;mLIN6`I|<&H`4ZDHjuyB0b$tX<jOA`+2Z24u<o+{s&9%QUgxG$GU>%$xzUj
z#3-HaB7@I;EVJh^Dp_sE1a?Ca(}8u)L4BQd;~x9jCh~Bx=Ch#<HO3d0#r{4^*eb#A
z&f5A8jV{vf&n06#m{fm`X<OQ!rmE(1v!;Q+fWw>Rdk8Qk1zjBh%$s9Kv{cAQ>mZd=
zsWXb_S8`2X?S$s@#re0%U@MQ~=0EK|8Xa2Z78ir;1j*Bjg7>)WHK{~03S0MB<h7oV
zy5rM@zT$YvmO0JQ=)CUK_5YgnkZzuUw~mUAFIxakx=@G|u`G@1TX)PUh4W!#0Zy(Y
zdC063n*?uxSgF>wshQt3UFHJpxc{}1#YI)(bNkALm{Pgqh?!AxAN?+6*BTnA*ya$~
zvmorA(#PFa9hfof{%l-=@1Hs0h<c(Fd@qsI;n}#p%^7M;npR7+dW9V!<5N6d$#E{4
z_N3#xuQ3z9ADebeXRQ^)D2S2uX_F+>aa2mVEKg#SECo>8N}exjAZlNTGVCpB2=GZ{
zvj0#xO=EzQtdq(rzw|kiHLx}>r<5+Pi!4f(X&IOc!iVLOtazI~<|f|~4Mno8iz;ke
zV5KIU;=w<w_dK8SEV|fwIAT#~Ia=^ek^<;N@5?508n)h2lHDCCv4wg`_r5t6-3|d*
zVVHe;?#Jo+rLoCpy>RB`gZodZnYv#K?s{_AtyHM1xWfdUDe_~vsW{_wOG`>JsA2g8
z{)3#{69r7wFrtuKDzrC0QX+r9$?lIRb@4MudXaaym96e37~Sy8+0fLw#cyhU*OnQ8
z9F9bW`*-gX+`c{c$F`+%K`UZ<Dh!H(eee3bns@?TB%vpg#eij2%_<evp*29#7*RL2
zh_B^MzTUp}L%W+5cZg0%k6Wr6XIE$h*I*8>2v9q+k+|re4gL|l-!~|pS&I3TPWoKg
zX@0VBvL5)nrk8wtM6M7A-|D}bFUPg}@3-QVm8;Jovf%n|Ku4q}P8d-^3q+P{YqDX2
zJ~#cA)bAU4PUy{o=#|AHXsGS}seS%haR_ueTZiy{zl%t4+xn*65>b(v+sJN_AVV?!
z@N`5CCktRm?}iiUEKnnHw7uy$Q*0@=-dlWwVmG7%{#;_m%Ts}n5=$e3fki$GoGR_&
zYD{wTqa7|Q;3C@2YNFSkrH1DGtb&i&UzGQ5xzBrLRbV9UjMCd}4K!BPt95(3z*e*I
z?=rrzfg@4|G19x9q!{8IOAx7)Gl&tD?WqOmD-ka+A_9Fxq0l+I7`amAv(>S$^<KHU
z%eXI?#I#-Y;O_$VK1AGT)_RsJpzA9rA>~zO(GTl1lp=uQf>=s?T7608iRu7-<j@Tk
z`|!>SokRRAP&klj+zg(>qZ?VZzZT1l@D*Z0dOUW!CHcmivL}$Yfv216f&Am-lzVEP
zQKxybHNl=V$?U3G;T34LT2YelzwQPa%WF5rJPmT@vf+2_XCa0V7XFNO;-t=+72i>A
zZ(}|O2@Q|F1Gy%33UQr6q{ErsE<J*j{84F1HSwZhQc%3HY2hr6lSatI7E&|@BR72c
zca|W(cAO60RYMB&(~+htk-+hO76B)5<GxISK=V1myQv=Ssoe2EE7&I+7LYs!u44M0
z$1Yd$dt0zwwIy&#{|&*^Q4%&Mwv29KT-tS9Es3pL@!rTxy$v-SQdyIOurhhuN+F%M
zK9!!Stglo?Wmm>&KMo_EyX;qm`h9@%Nh&9CglzI6`>UU5`=hM95KJfTnth(crrW<-
zuWO<es-s%3uoJ=}gA9XkFZ-SQwJ4q!`N%|^z=ddE*G{r9<a$YrOdfY*_d33ocFl8L
z{WS>L7pLD7QG(B}0@AD;El0mEB-Rn3t%VmOkBXZpRAMh`ky7mFu8Mr5sVit~t+Ew0
z_A+c4|HfNmMA1rX2j~rh4V-~56k%a>PRsV{#unh_dev8;C3;^>1sOi)uDM&W{sq(z
z{%roq_lI3o<3{oy=&+fK{zC4;zAdYZ;omdC@;mv#x20bO$-p~K$>Ztl-?}_+-@s^y
zyeJN$6}$6YItopc{ymW1Tq)|=kD!ZEj?vBkNzy}2PtX2RJ4B+*%jqbXi=0)vmeR+}
zDqs11Eu;AHX}mN!@6GY)x`E^bAotJvb4>VZuGO!gAR@POQ(w=D?yy^4n!~C=oil)9
zWbk$V%M9pa4!@hdxjJgT0?B6ffq|^fUTJvzkdh`^6{&R`^QdKd?NRp_Gsu${WS?7b
z^g6$hll~FjrSR7L4^9$tkWmc6W>6G8qS5@V3s9{y#QJO;@=Xn4V(snmz<vK8>n&yi
zFfVp@D@hRqMGOw_#R>NFse>t)jpxupe*B&*5uQVlXdv#6rijDtX+i;?ctV={@6mm4
z1{lz!;n)e`Fl7FuXgqQv;@m!BS+`f3U;+@x!U>Zcz#Igzbf(1$>s{~9VgJy)ugZ>0
zR6E5nJ5?;2jmWXIK*N$SET+-a8gc{twLnyf>kwDimRPR*ZxyU4fxQ~WAA*#L>$N+}
z&@#eMx&&oeO1CMuoUMaRXJmIaZbY^l{<Z$l_Qb>vWC}?|?tlQl7?x6P-_NwmlYoDK
zJVT06a^3DM>3v9law8!AObN8>Fiuw4hC;$K2vB|H!&{fTz$vUD-54pUNwRat5kxR$
zyw3kw-0*@N&{j9iSTA|r)SFI-Uqb-``#6U3gV6A`Vf@uph>Ln&5;{{;Q=e77AF-sQ
zL}W!Q^8@<~@xY;8^vx#f$#uE_Rj&VTGMDtKH0a!$Qs7_lYh7Qq+ni2(*Z{|wY$sWu
zXO$AG{cUdFUZOVAqtY5y+G6_z9A@c3lRZBY!(mh<8SyW+J8)f8X#=EH9;bRxc+a;J
zwws~kGW1cnyEo!I)PmAM+Jv4(A1n$!VQ91(Ayzw}e`Opu_I#uTsa-NBHpPW;IrzZ<
zuzDD_ih0S6`^8IikrXgvZhl}D7}?)m1uOcaKiOoF>S0fzI~8GZEztjcS+$e1f9@l<
zqL;9Ggzn5!<Z&ZR<k8V-T207{M}!?{!0#isi42O(CNo+{m3@7rt2&(TK5>EVJ1jjJ
z#XRF_6gFn;miNeVJ=^+hY-L_^UJTp?pQ)V`k+a+5%Vb+{VvuhITJljdP-Ix{kw?Ts
zu+QZv&p+`Wdw_ApOlExqWNU7vhb;?1p-h$cKxV$O8gQ3dGwaa6qlb>n!?WzA`Q@v)
zy0LiserM7~gHZfx+eEq&^0BIu;_+hi(A?Gcjvn+MTI;7tE!$jB{Q%v+=Ns}{l*1uo
zO$&)f)9yHq4tqjQY1U$Y-M}Dj7=nTe!7bU<HashMg&j9cJuv0&k4whn8(7IXJ`Y2d
z?3kAbMtBbA80-nFtdbYgl)__S5{-k3H(WX)Un1i1EQ&{67F2{W2N1cp5k%S?|B%T!
zWSHr=N9rnJ!y|Uee!nz1u7O=SxM5tL84>Yg?3FC+%eLkmX00EuO5ssrtfh&rAEoQL
zP)l~G{R9M*tslm`o!z?x4Nk~jQZ+>}KTOASxZCJ7srB}9lTpeQ_<`5a*q+9%wXxjS
zJ)qXxsYDD)1oaDR*vHH)&~EX&o!sjG7<ju|`Z}fRo$tE{CEp}Coo@lp?Ml;hR)r1z
zjjRS3{r!<Cw9E|(cZYHf{!{OM#<r<GzjAE#&Z%E%Z9eeC2Go_rabVK4F&}|O>RGVw
ze#Mo!F+a+crurs1`$lcdV5cgMwNVnu1K8G*Di@W?q?ZQ1pG_Ta^aarliu`pL<DnG^
zOSqwkFr2)VW@-ce#yB38pqTgNW$KYpROnk|FCI>r<+V}?%NUzz*?<g83iQ}et}aLX
zoYCrEvMl+pR#vIgG97hy2rmlzYlP5k7Zkn3ttcq7+cYZ4G#HieoY2GTaXCehOH*aR
zc2f3M8%LfKYK@IN<%F>-sZAQvL_@njHB^7X3-v7)Rra)M(rpc6UDYu}?n5T01*YXl
zSNMc3<aloqIMRW73fp4l(e?R9w<&(m+YA~Ndse(EQ*a4~2+zv5R=*(<s&*>~$rENl
zai1>b@&-8+d22CY(C;deJ=6oN2;23ec}SH_@XU)G@85aB3<<MVcK!44#tQdGt9SKc
zQ#vH^saxP^P*rbxFGXKZ{}QXeF7lM|sa(Y{N!k!*GC{F`G%i<X@?toFdvLZd!@Ie&
zA1w10wnCnZiz`41nSBb$GXIL^v`lEdtN8Z~u7jWE5Phe3e~BJw{SVrhb+Y;9<5=vX
z*$^-kPmy-9f%t|Fc<q_38fi@x4w%ECVShzZ<GicjHN>V-3`>O$2SvOrCDDBOX2n(^
zj~F86pr#X$EQXPp5poDzO+yj?YXhFVt()!lD6s-0(v#}mBB}*BrS+9>X%DG@%)b=4
zNEy#}#MVBI2b6`~f{XZ0^$T4OKVWm+NPy)6Q9M3OR0~7st{22J-3N=pbmU-(wS?Pn
ztVnR}$C$0(Le#Pp<TXeY$Vf4<XmBcwX=%=jN$Du@<cH+FDfR(NNTlbhO)3Q}EHILd
zC_1L3VGo#AI!~Tw0uZVwi;B*U9PCeN^kDz&2I+1ja1K&8NsJo)u+LzGcx{x8F&=dc
z<fDbEj!3`G&q(WBDZ^IK#K%qEl?@>0^ZN@iiy+h<o92}^;Ap?9Bw<lJv!ox0)qOqf
z$Xbe)59;5!)F}=@5LijI>S*NigLzhq1sNCMfKBl$;8xo7mNq!!h575@|M*}5rA-YR
zl!NWKMLM&sN_}o2Ov(9+cBb!{E4KX(t*ZaMm-jnfcc!nKY0*B{+Q>kxfz4pg;hA9P
zWBbKRF?;b(!Zf0p?Y&?PVLVKSsE?0rK0QHh6l}yO-q6%NIKD4-2~6m39n=P1{OMG7
zWq92TUcLMKZOlmvTSs&$PfnPF;d+fV-F50%+wyoQv3%@IGzOD*DT)C&fnerPU#K?Y
z_t>d1*1=PLG4;2lbvSpA4VH4Nv!q7s4qY}E>|8&Ph|$%f9!)0%qto-lZ3->Njkh%s
zw#+T!mmc>9{24tO=!!br0Kg&ogCnF9idFD>Djmp=XoC68If{?JB8S0xKUrb-kP8P?
zQ1lAH75*B!i%MZFZdA=;5J8tS221eM(SWV22p1AS$-Cpj`24M;v14}^XlalzH<L6{
zNjMp^<)suBbnlXjX=xl^Nj>)Ca@;rKba{PPGNZhglP1uBoBPsd`w8wsOI&n%psNJ~
zPIJH9$C#ERHc@(9njbn0CqX(KNfNZeNQecE*no)6QrZ~`Y$BtCI+Fgm5+l4u<5JcT
zijs2wCo=}P(xV;G$=(+e)2PtqCzsHSd;dwDQ2dn)Dp=}eXH*?}isYa(V^ldt`d6Wh
zucJbkRj9TmE6eXhCLWN#w&PN%{@v81fv<UG+>s@rC=26BsIdI~Y1gLZ{<>M9J0_{s
ztmZt5Y``F{Q<-cMeO2<M`;&s80?YL-#^n#|*@viEYVqs6!1Y(QFqY-FbQMP%xV%aT
z+2$eWA#5L)rsL4WJ<T0spbt@0@0k1$^vO#fm$B4~#!Caa5pgs~Jol7{Eh;RBG~1N$
zDYba!?`d<Z7V)Wvh#z!(#ru(FYi>><<3}VCjKo?cG;m$%R%}B!Y|zryL#K~YB<{`C
zPo~&|sVHfxvqk;tPGVV$<bl0G#N(k;22%X>jFt}~`PorfFUoZ%)64*83R79+Fr`0N
zMum6HxMdCHGFh{Pp~ij)k%7&fv6|~nhLRCSA2|sO#L}}E3zMX{Cy6ZBz4(1N!7(^n
z*@)}=C{D5Y&;%Fp<@rZ~(Du$sAM28fl@DFaqIs}kuuGEBHo01XB>idjTt>#bp3yE<
zFBUK`wY$0KC7cRzopGmbaXsO)cAi#$8TzB)MR_*Jji(XOc)mqiXkgi=W8NJr(bOWr
z@eYHUId<E&f;yln+4a)$#6)y(xnH0}o851iGZN(Vq1Y0Q&-|BrPGB0BdS@{?Kgo;G
zi`qRGddLe5L&T<gq(#uSZp5V23dAN`W!7722FJlCX}tzx+8z5Zhxm<19L$Dv0&c?7
zCTB4$dL7irMy)xG<3{9kz(zJM^3z?Bb@$#*WO>JLz?83{aFCeRh>eJ6`EIz{At^~`
zwF3_w14Q!ORSrAxis8{A(X1i}DDg0;D1Tyjf~~`|_?06DQy>^TI-C)*skU1kM7ti+
z*<So(<VX`1pYD;kRAn3xHv~n=pcOapiDfrlAh#M+dHcH~b4m4WnZx4xEGd`4vx|**
z)m2d~SHSd*aE>uzn=jJx;L6uxdm<x8yaKPpZ4tcQyHm{@gJM&4Pn2qgQ!me4hNG*E
zO_e<{JZc-;I95_fAoYv##oHNf6037#`F1wmlmcv*$smggD@InVRV1|&?CP9p#n1%R
zl9)3UFe>};_oy23?f9zH^{wxPwwRi%-;&9sUrWbsUUMUZlIgNKp6qGseldpUjTr{8
zOM#;~T{5k&74F#C*&@T3i2+}qZ8gfFDlZ?KG{ZQV^0lN_{|qA179cl`lV~WlrJk5*
ziZ+RDS#Mv~F3(d-%FOapM4*9K@>v~lkl`A4n6I;lk!CM9HX1+q%8YVBX(b_XH*3BV
zP0fZCCC?g6FBc9H|0noN68--27NZlB5^v_RGB~6IDTsT?(u!dr%-x;kz8rGR5oY~@
zK@#tfXm(w3Dk4$DGAZ2v?|3y!PKh@w@{}D&ftXUZ=4?XyhJ!t#-`X&wu~R(i>G&<h
zLbh@Y!oRH`bBrZ!ROE*v_mR^~#n~RG;QWS;Nq59fj0@$&PYq>u$c$mt;{<N;?62M1
zl9t*@=tQKsI$~aE%>GDyVztUX-%uQWJex4Ko(({*LwUIGkOnuO`bEuUlnk#TZ!Bi;
z=UPtv<g}6IB)jE@PBHdswZJ08L#Eh_*vv-`+1E5mHvR^Va?}C8^pn#1?~Fj;{UkqG
zS8xdt2^eNB&YszckYmDiGe6kVE)LnIF4F$}9MFZIx<>(qRgCh(HR?^r(9mOMIw)U7
zlwYN(sMhu7D}#(Wv59jzQc@^0aqaCi`>^6B`U`2jkC@#(d|$d@V?qZ@WnLG}2<xX7
z7~>SB$Nkp^^fhsWc(<q;p^?Es6JRV!7Dlu}DInGlQR3(qW<*r!iHpJd<?C+ksVF(@
zyf%9jKRGcdnV&7?NG+`oDHt;dh?%HhRA-)3qrxNjp<O44ZT4ynT>HYKjuAj3IVHAt
zy4d_U=)8s<VFS%KP}H&-2Btf(aQ{$Uc6#1jDx{gUS`iuMF>uN0lCvNprC0QwRPsRm
z@Z*knwC{Bmp)(*gs_kk)7H##++q&Lvz`YUe3c_OWFOR}AsBmLvI*X2S$u!l3mVD%H
z)t9=py{e6toOFd|d4^lF5KWK1;;cO{-()hfq3Yd${P2}Ic^@GM!=0#=j*K1_M9{tn
zdob)l)N%T=N9!l~xl0J@c}*DST%$(mjy0(!*9Cr4d$wrP5#E{n3wyt7V|WVP|4sha
z9c6gGSnsrYjsUA+n1Ebq|L}A|SR1EyYPzC@fET1tD}i31^tWFS(Js-V-Jig*-iO&)
z^S<ww@$ysj&LED>6AMYaj@>2VoEjp_fN<f&tX#H6@A{RaW4b9=0ah|lhu*Cf=IMKN
z`}<3W!*PYx)#ZLK(<gt4x6J@A(P4nbymOJ3iZR+@T@~W2y>?Z<`6SM31Ni!KLc5(+
zpFZ53QWOMPsL}cz`y^8T6{Dq>iLeB;wZ}-Q*tmKkx*716++^f)a+)?4Gr5Qfs)Q;j
zo!eoy-$Dc&&*)nZA=;i84N6=Z;=QKQO)N4!{g`B<lLOE7A(%L>V*)QfxjBc3V~pKm
zZ3P4!G01%~CfM9_w_(Pgdo1{1-wa*@dAV=feX5U#ChI0=Go$RMH>637E>XjKZTEnv
z&#d_-+WyTG`nqjtX9z2S-`ZI61gj|+CE0dDOmdhw<X!!{VIclr>-eUjC*sSyTF#bt
z>6^vwojL_;r_rY{BHuN!7t<h`>a-;4rP>a`^;xsghvRzlEX5ESwX#mvJRFD~_~s{N
z2{u9k3137&Y;W_DLZK^PuTQ11eLd{7_q<3nN7D>Dx0#gU@cl2|i^PzpB37k-ehe=T
zXJ=(}q1J~`83&nIC-d=JZiIiL(d<809|cNO69h9EcKVe}<Tq^eCypq1#694z2C!!`
zWrYX0Rox8O&Ue~^h|WX<getR)?DZe!)o#1<iT5}x!SskjNtj_-n(HM2LKlb~f8v}(
zJ%pwqP9$Zq3CM%QmL?Yc-8XP3k&yfpvM$yeppg8QfIXn(7#*k~X7ZRi-!@59vqogS
z^N#G{0nZsF-1_^2aQ=}o$pUUq{XSdb{W7**l4S5#ZM1y2%Esn;{;=WbB?R`4fr&fz
z^P<r5mNlZn&QX+`v2Wq<Z+Iu8d#t#9+!-dThRXx@ve1ov#9%36!+ow^+fU{qKQdS_
ztrcI0ao{cqdP_fUH5{6Mw#oBY$|9oSg(idc6&aGF!CiGRr5``(QQ%QP;GyC~qh<?z
zS!$4#klKAIa9x67K{wvIUaG9+R>$HF79-HcM4Dal@}C}ZJNh+mTp)3Fz=}}2mdg>y
zVIJ(3xvdqcX?68rnu;4H^0M~JTXak1&c!0dHV7SB^vB!jIS@?J;nxPbX6VrZ3}ydV
z>z4~x+gt?Wi1<>7_fx(-L^e#~aUli0nx5Abl*4O(NEBL4OXII{*1w}0=JNgE*et)B
zQ(2Ia6r#p?Cx@fN51%_FQewA`W^ls3@G{K3wxWJ0sA_5n8nEN8|3eD;$|q8)ux_j<
zEcfIr81xe|CP<algo1=pg%)?6nTGo*QQG>^3J`OoHwH2$AK{Rh?gkglbaFH-3QoWv
zX}xahD<6vg`y3(8z7P+|A!7;i=JWN7HST((MFO{9K6De4t6s(OkN!eLm>SZe4sQ~b
zE3NG&8kSjVrnv?7LW+pxwW?-lAu;~TnXfR(SxLVwb-%o-YIVuNqH~dmiZ5_>1#a=K
zR1N>iJhE5ebW<tOxw~|;NePMWYUnifo;jI07g2^E1lgkY0_AV&e0NxRc`4j!OShbF
zhl-vu!Yk!}BxppQ#?6oY2nBdzA!0hU;6;d;iK~Xqy~@J~KfbG<#?4I)SSsCO>AkTY
zlwJoTQCNz>rgMHrb4MP$`oJ;_bKHg%60o?R?ZVKJb=)zRe}Qsvag&8G_e;JWM&<=~
zfdzTq`T_lh^)-K7?VU}VC>Zel-}nmNiXJ}D8jQf2ayK&-)|5$}a7^!@-<<4e+^NC)
z1W||-Ys?uD>+@~QIT>>UH+g>OHu^|w8=Als+UlL+i)`N43GbsaZ<grR9c0|FWQVQZ
zke-a^dl;8|+~@eI@G<Ad%Y`tHghsPH<gl&oZjFn?0FEtY7)x-CSLnb@r1f;nr6N1~
zR`A$fJ<^M7+hr{dps%~**SQWaIpLrscjJUXuwK(2PLr+d+`bT4a>(C-;6^&oQw0U`
z7zn<_iov4$Vq&o6wsxt6Tt9i{9+ZVq{b5DDcIX4-f0M883%~S%x0eI{;rLD^k6>37
zcc9ZnH%oms#GUrU>^K_rBqCqEBJHkT9sWg~#Jk-oPk?gMXm;-7!ap*K|J*7^>CiT(
zY#p-&)F8TCF)b-kh-nM@nUe9yk*>oI`G00H0&)<qvS&BAYPA!oTL!*!Io7{};w?XO
z?@~QpP1H)1_aySe|Ncp0yQvrh#a_we9ga~u73*?7Q=!-d1t|^Mz#<oIW?^>fcwsBr
zYL0Mb6-z52UA|l%pZQp*KiV&SQt`+Y+~{UoG~g`cwedjP!!w$Q8O29cMKJ>3$OTk-
z3+Z-fWG|CS9`>3)_VX6;HVJv4I`sSB`twuhpjO|2^iL8bj1B=v(WK^pC$inl{DD7z
z{lyWWZp-@C>-)~Wi1HIrLJ_I({k_6N{?SRCQQMu{lwo)xNpOhqamCL%4?=Bv8aXa{
z-`YClb=3;s|6o=WagHmK6^QMuonNqU@-6rb;kh9caY!-P4JD3GyLCruuYo(7PQpJ^
z4-o0C!V>!`>}&Fy6@QAyrenKLPQvCq*xQ+t#I@aqVkl7$Z3UH=I8CxF@hS8beXv}1
zdhr8knw#v6Rl2+UL2N1n$4Lr~ls!*GU(-fqf8IY>1pe<+TCNoy*U8sHVT$DsJ`YQH
zzN%#-@>sN}`!4!<Dp#yTxrit;s<E7Im=)uzDlCHsRT!Z2*JHw5>q_{2?eItbe0<h|
zbRsCO1npu5^nnfc{D&9He6AbD2Qi;n-c0WnD}Tlni@(0hatU5c?Lt%#i!+WvrmVb0
zhqGwl=JWHwY2CBzV}B-v@i)UT<%+XKX)<Jcys<Nq^=9F&?h3r;mYtY13(2*0EVie`
zun&^Io^u|_iv*spNdV}T1TrOfequ_**39x@SFJCA<S>VyhA7@RA(S5MjKkr2$ng9e
zs?W63iU}S0C3t`tCDAqQK>%PblQE|BXJnNA`3}qIA=jTHt#gp8Cgo4?u}4T<E`XJu
zq_DNuYY@rP7g|mb`_G1Y9bP^tC+RpWep>Y#(c;(`juh#8Jq^oG=PsB7?2NcrRF^#c
zasJ7BVcpu;6F9lunLn(8q5t9zw82%OCuN&|^<uj6v0P$x->0ZUx5?t-?@5Si{`nl?
zOEgTxj{$_4QefD&LLx;s{$dno5C0tOyQ*uRBKff9IqJ-k3imJElnSmIw;WOS3uz`?
z4QY9s3OG7wSV*7;HGBNo^WZs0!y~n5l_ScH<dmkuMf#4ujP<WiWR%M!N9Ky-W`;n#
zoS}qS!j+<+CQ6pD%$Ycb9zB@jD@t7;BO&1GxbWSzDc_r)U8mOIPuxWb>0>StI#vdq
zt-vyUaF~97)uEz>C8e<gNVO}nnL_`E(wX@ch<p9Z(nY9*p|j5F><jnD40gTz6i3Pq
znWErYiPea`tUw_m@fe5b_g!$2$#Hq2VYq~W;_5GEXb!<MNPoD%DVA(1Xs%u@ZR*@m
zX7;&6`P!q?iNy;-U8F099dq_Ij#UJw!qicCyVNo>KzbOF0#_-vJV3TdF+AYaDuVyM
zKD2)?wt0;|jO1$dqus&nvCETcUkYGyKwxrk<vD)=V6s`3-qj;sk0Ha`nltR2eyBj)
z`{pZ5UExmI9O7Cd-;&rkC`U=OE~A1sbI!b#3IZJrk@Bqb@W<Zho{y1MKsL7NAP~B9
zz%XLOiw%h8BFDf36*pGFM67a!xx7FP(`<j~GMTjTdcZmRBjMsu$8Q*SE+Xg526{JE
z)Nz*9HxOch6et~eZ%gDC@-}w#tY!ah;%N;E@9wU6mAA-mU&*%|6i#$otoOiYQ^|_k
zZ%IM1+8LagZ*M{%gReG@K5sGy9De9j`gzm^XJt&!%fw=YfV%OPAN_-ir#9M?0`Hk_
z2NKx2Asa1`6sNxyjlDb)Z8g<9vEHGl5<Tm>sbtP$Gd<I2Dyp7^lo;>9h2SWfp&g|5
zm@FZNQ66~Z2uhkLCnf5;D;aVQ|4B~ZHF{z^yfj?|hd0vF>wX7v)#hDP9bG|HrdDZc
zkDiqkl=!QU2pTHde)aIvQ(J>x1hGB*VRaE(T4c+P=69&7%}==!sExWB{!LP*>*|z@
z=DycA4F0@a1W390DOfVj5}c=97|>)9bayglo*=v!iYQnQjel&Iw%f7SPuZrtuERKF
z8}zW8MfuHZ35GTmij6j=M^lPWLnkS<7-)7LEU_s;hVJ7Ek{ltb7x2GNtjk5P8_}HL
zhg)4)GhjSLb7SVonGaK_W=n|8Uq#dq55z{IHBRx@sG{oU0#N{V(XevVh93KaDc1<7
z*diygHc}=*K~r;9uA<IAQ~Z11RtUA;?i0=Fqws7=jqSk+Vu^K$@i_?0bmVjR@WCuN
zGg_Hrn$t!pshen8PYOtSP%2P(e$MYDqW&VD*0$MP(4gva(^xmoReC2a)jT6b@!W_}
z3S$)OI9>nDa_)WsTxPg|;2dz*Zse~mo^1IVrcNgrHEm2XvHwQH=Rm*X(&X5F;>iLv
zAQYPCHEK7K6iY&}?*0mkb`v5P(G#gUF{#)8X55P6*sq3P%sDxiTJIw}ET|}y=h=DV
zb!GAfbLbW>Nr%cO;#Rh=^u^$f!o<1aU^49j+=#5nL0|$jP!En}S4k^F(op;=<Gy8J
zk68Zpq5NA%v%nb-^eBBS=t_IEiHP!B0N6gRt{D$V$YzN-li{2Fxe}h?_F|dW>de70
zY{-TTLl(pt>BRHc0n|wA?wARjy~p~*b-TZ)ZkK7!U<a@I$_`UW8+l}fH*}V?Lh@oG
zslhc4bRJ|Un5WUK)>Q$(T(g`LP3U5U1lDr8f-_q#C0?6>WV)%Nc@%Jt(Lk1hANNG(
zr{)+`dF5~kjjwkL?p^JKoF!lLD`i}B@~ZPuu1*Cx_c-L0E4EFAUOoIAmCx6)!`oE5
zV?#sUOIOId0HNsz?lu+<(O?M!(^VB-9L5@{nsF>Xlz&zV0olYJmo@B-Re|0vh>p82
z`476Ga6yJCwu{ksJH4e?%$>z=g9g+L`!NVEkO6&nC-;g@bucPhrhJZX=6*0jQ36I0
zn6BVelwmZ5YSPv$zsuG(Wyx6Z!A`8qacA5Od%El%MLSw_VK!9V4LHv$U?)?wdWdK%
z2OWtVCTW%tDf?K|XB}vV8Al)cn9ixac0}4qVlCTIe?X8hs!`{{yUz*Ro2U1Vkd!!&
z4P&oeB0^~z${C?AqcD3S7}uh!Vf`aYKlnH9-C_%gIRNXt5>ijVWa8iTluARQ#WJ;J
zi<Z6w=fU;DW>SRf^7c*k%|llB&h`<`^yiaY(v{j4dZbDrarsK&Q#2BQD0u#LrT{C@
z<G?!W;)Z&zkow^NvjF5oB5kPVgdL-RJu-L7*6BDVAHd1CIlK<8<F(%YaMElhe{@7_
z+nUCrT_!&g+4P}i{sCvn8W}q~cC5NT9J^D&zVia$5kw)NEco~Nh#>BI!Ezg_hHZN6
zM$Bmto{mlNjaOUSI!BTISngBfIR<)wv0zE!iLCo$`{TFO7IOk1FGW__6@VnfifT8W
zSp8~a&$hWua`%-uN){|y%&mJqb0Q4nnhECye86VIllz%?j_aTm5}o$=1|l6&QXN*u
z&-GKkQ}nOFwQU<Y%+@uzQ!Yy-)V4re!;q2#*UNmIU>MId4x8`gG3MbtD5IOIAxdWG
zZL!b00pcA*>H}6AB(=yBu?_K?>%oUMB_AC(!M0i>QlrL+&E~FHYStZQOCthwOk>#x
z+QGBxgzcA{Uf0W^RYo6t=3sul<|5?ip%BEd-?k>7k4EHZc#B+=WR>LEVs#y6RIX0W
z#+-7zt^<eDVOvL@o4C%ozb3|6`KpBQXPKR3AF8Wn(k&S`4~Mj1H_jz2A-J_doo{^9
zFT=RdIo?=dnbH`LR?m2Mi8lgy0G&@L1#_3eSq`0rX_YPn+thnj?$rNm6qC@Ng|m4z
zq^0EAgQhxhT>_+u27_z;D>K9Hoa8?polty9Yqj?wkYet;G$A$3IC$Y4h~X5DG1%9b
zFAXEYah5Vp*|J_^M(>a3QOarxQoJPEb3RC7Qe{W*eB;U{dLkGO{TYg|5`|WdE$XCE
zjgy!@_+tDp+`_Y6cSb9)j3~S*Ao8)MY*%xe_q=1qp5^bmzo#fZ(Zrha;)qj-gEDID
zk=UVv&h#(MU$;_`gex{F1ogbdy1mc11_ff6SGs}`E%dilqk&ORM=TChST%c3R%M%t
ztw<$XhE>Y!5~z5}8GuaYQS3!53AsLuv809Jhx3c4L}GYBqg0!V0+Q66GEE@V=fA4*
zxT{qf2Ne&5#v<mQeI?wDXs^Q*Ci2Rb`uD5LxJ7h<W0EzMY00Gjn)rxF_MtL<zNYZP
zJJbrYl0j}1auUx_+N`StrY+3D7=$6F$GHL2Q$OMHX2`|3>MG1|jsJ{z)bCpNvaF^x
zd@+*_lU(|6QB?KR+>+xT!`s$&2<f)5C}qc%G@-lBHkOXap+jKpba;{@k$}Lf_Sc!1
z(7KM;B=FDL3kF=R<B%MEX1STX*MD1fbOuE0!6jTdf*UJ%N-9c+BmU&Is+YPkPL;D;
znO@yYzmwwa@@3tt8MT_drE-v!8RZ$Q@u9984g#l_LrlCVDKZSqHVuj#lwjYeIE20Q
z8_SjvWaZzwymjj=#@c(cMh!37T72s4-1`u^_%^4&jc)DW6w@G?ep$SPpb^0%{@7#8
zjk+L+cC}xDT!UFzHQ;nY#dUC3ZYOyi!R4(xQz%}ua-yYl-TeD)8zd_yyr3E7p>Stn
zXS^jitC-Tz*tmG9(PlYE+R&z#w~|xsyqOf64@VP~@$7R;!K-C+xf7VBOX=28n12<X
z<Lvqmeg<OC2Rt_7UfVHbj6<5$#8&Wb5d9vQ$3VJ8Ip&6{C_PSZP#a#vr}ZMTsQ*uK
zHCy_tPM9FyVPido6PsDr;T=37+X>B|m>zSOsjMs^ep{b;HFs7CC9)58N6CUXmSx)9
zbzP#zfDdo0hm9|VJs_-6$9L0fPRY-!k47aYNpD+&xIT_@sd*Nm0QFYMfj9lQjL(t}
zqE|(HN8U`lkCdQ>VuGd`;(rjr#dj^zd7ytVsu@skcst-NHQy?`ue2Xz-gYwUaR%iM
zFvhcQ-8G534kmEmYV4CHc3UMSz?B>Ei8Hx$9o0iCCNN6fEgR7z8mK)mgW%wXm7XR#
z#{`dA5E~k+TCW>0WMW1cbZ<WCw*eijwBQUQwuvh@axpp)#nV8zTuq+^y<&#7W-_*c
zmuidFwqG^fKsArYHCyKkHB`mH%UM~|JR8nGuU-4w-C~S2O46Kb;GDTyCtC=Jx&VcI
zct_2Ms*Rd0;H`aAV!FQ=s*XXjxjhikB7-{_0Mzy+ks*-tA>_Z>=<76YFsJW6pORN~
zMb#kfPt7RV2rWcL$F$0KA<{=0VPDnuHH&)Mgi7!S<rDjNpksG1znyXWpk%zsx4IHd
zx>3p=isuG>x%<ztH<ZQHUs>|5+7TQ}Zqg(?AM>wfm3R)IqossVsPoiQzLWwS=b`$i
zq=OpHLH2!C$O<OZS4#*wUiuJ9O%~YpMcTAp7~{O~?xwQ*K7ysOIW+1)e=CG}Aa#e{
zFNuQ3hJ#*jHC%C54mX1B#2Ve(ms=&4c%>OSJI#k~sVmc^IWLR$*@aRH_S<M}cR*J_
zr^UhC7Oyg|p+i%K;h<05%zX}krt_-(*SHvjOfc2?btJBHIRaq!BVRG08j-QGyJ69Z
zUGYz2<q>nqRHZDD%mcW-%<M6n6Cx)}ygZ>?7OWVl#?uZC3;5mIZq6R!#1)%rQf_)X
z#OSiSTK)sysaIlK+q6YT+kaXu=&G#4%S>p@l$XsJI%YtUtGyP^S&^3&J>{6EZV@As
z(-#QmZfrG+){G9)c4T7TyD&`bfGo1}(&plr8%u`&Xl$Xs=m8!VmdRG@Fwo&8h-k*6
zpu%scj5M~lbF*xU0)ou@I`a$4M^>Y5guX`%|JHUls^_fhn~=78c`3ASMlp;P<{cef
zXY%IB|ATi`^!j0hhG&0cZW&3Ji8=&}Q$i*v;ew;^7xnyB+-CVcc1qn5#C5RB{6tz#
z#etwH_JjG%9pglUs4P(f<---_fzpf85GzAh?0s~&rzp+d=3tv4(@49a<uBq9z%cXG
zJWXj4B#`h8fg`ic(ebYi-PW%(sYNz`&!+Ai<A=kPkLA;ujT_6pt6~VwjD0_%J*}Ze
zBji&pve66Dva1oaY$`onRO;RT!`?Y|R~B?@yJNdMww>(Qw$UAQY}?#1I=1bOZQD-A
z-m&exdA^<J{DgO$udCLmy5}m4T2*silX+*`$qalrnemMD_p0GZYeyL#dB>!+Qp2fx
z#7f9}bgL#rQyI}k_JZYJZ4vwdbp~M6F_h?R!1X*8-$0cpJ=Vw3my$zJCU^)R>MQ}4
z&rBD+?=DY31~6J-dyyQNbbiCJdhZ2m!HrzZq1OwUP^`RpKGo)1m_xqyAMTX$uFrLZ
zUgGoZA&ON$|J<7*<@;A0B`6q6Hx`D-TZzMV-JA;nj`I<E=QTSm?b(05^`@JopkPM;
z=T7loqt9=)>2vzJ$r&J;5nt%^0G20g&X<QJi`f+Z^f@=Niu;9FUxPNs%TjX}NRwk1
zz$2}pfbnsY%H#4^8~A>I^R=&1cD;ly!=r!`jyx&7txnP9`q8HU^?z{G#hl8;JXC}9
zd)@vFaTNF+U8ET^u3H=ZkJZ_FTns<nDPmz-&d#zwVE^k7i_`5-hI=1ke>4HTD8fO2
zw_3|6h1|CmBMP`ilRbiHW*<Z|$vMUP0A}lF+iBOW47WR`9zqB^$S~gKuoC;{aW6O?
z>$zb2kU-OPQy4ZygO3?8NlXUJO;&Q7EArvb&YrEjxMqfyUzxBJhXK;k)cMxSc^(&)
z+=v<X;BZ-r3?+Cd^&9Q8F4w5LyHhRbdP=$roK%blydNqpKAs~1T0&I_LaT1~8Mc05
zp$Z+}_R#?S?*j|2di+;*Qrc}|6YX%kuBI*0)S?^WRK3}R?pTZiweO^13%iPa!q_@r
zDSoD`<tAG3%*EKU9Rh+BG(+M}Q72s6n9KSsAt|LF@E4G5x=02FyPlA1$<U+9;h&uX
z@p@3f893$L&IXq0c46J{kvgB|cYDakq3bd7M&ZLe$@dCG*X*W+bv%n=uz3~FW`6QU
zB@<lAe#q4xPO`M1Ps46gaoGjg%xWW4ypr#BoHB#;ABYYlKUz|Y=8*y5GO1n9vT=EA
z{gWC#G15;pmXq*$`Q3@idCG!gc84S|u-d2b+TGV5AxOP{rbl=?3iD9ny64#Ubg1*~
zNX8eOIbMpduz>CCH>oxBWUr9GoO<d@pp&K;wqHjeU>Mu#cI4273M7VZAOh`8z+hGK
ztwpnc5qrT{w##R;(P^*<^jiA`m~se}g0jPG=01HiqbGW7{duFCMIqLaZm>y%xw^ia
zXs$PhBQqHvb)Dws6|F!N7%vdpsG8}!;J_U6+vdlI@U@(h!TAElQekZ_3L~Y0l4Rqa
zmkS3;ouT>PM*W5!n^r7(I5zb0Lg4yQXOP`FKJ*DQb?F0W20Hkfb=;||4#82T0Xg8-
zX$o@_ZBB*<eW*cdAJSnFUEt}@_}0<4uF?o+r^3|2emG1%j9)0EpRpu|WRYa&JDtHT
zzOd7{mUt0&m0Z(E7KXxniWO;xjp~Z>;6uzgJx$lwE&q0F9a!3#w8<~)u{|in_R|jY
z;ozuTD`);vl5sqlVHS^5V8qw7qQ6e|7?X0$Jn!z}WAuBeo0OAGhn{#vI_*R4q08}d
z5#eDzXKc&KAPr7sY0iX6w#jSP^dkA74v`0s{}2ynfpO@4_KggQ9Z@lmROo%@GBg52
zXN!yIiimG>c0bF{!(4HSl4x}t($k)lVh)DD?7!aird-g*k!=k+?J`iTME7oII3A2X
zq(Np!y?q5na%#05fIy?~k6TPe!_fpF(pnF&+gWsUq6=_K6v=4atb|&FArjB>rzB;@
z&_hCM6!!H7JX7*0DO5j_@$FXgFs7w)u0QUYZo3DZ$$KQOrOtb1S@=6<S`2K}r4SMv
z$E<6mo^M*{b#|3vMu&r?MfQY)0~WE)?Cs-1YsUih*jSV_O}cWw<*@myU+y}w%aJ3Y
zY$*+iKR*DJdB@{{kNGsn?VNH?ZzavPyDnKS=7l*|%FI&<S9r1J%k|lyJ~u&j_FukN
zfkY!%U01>N9S!CjrKGlEqTvub3emX5m5@(t%q!4nuDBv{`TY#t5NKURNJIPJ4!goF
zth@TMa*GWy%zyRx@sA$*iXT5WtG<y$eccC{J{RJCP$lk-^h04F!-Sc@HZ;v}p#Vp5
z1ZAUmb21+G0j1?XW7mW%PmMSGgsM!^(hcA-po^y_-Fv1LWi&Y1HE?nTh3Q#jg){ca
ziBZdfA!rCV%Fw3$C<VrXO5pa#dgi>p!e-*X;;0n5-p}p6vXHMEs<lD1tEgttLgs4M
zAZ6NS|K-hMf8Iqh$h6rb_OMN`QG-jd!-hZlPe}KFQbXT7?ZtSRW)`TMfISHLZ^kxN
zK8*C1DM~p?w&G`_=khVT;Sg~DG7&r0&o}spk`T^YM3Nz8#@P{Ma^PhR@XL|fUiUD(
z#$9?}w>moQtit1&y)r`uR{<@M^n}?whZpc_$L$#p)?G-deR|6#kl8#9OMlLQ(|r&6
z;Y`9Jj>p#5IisJgVRf2;2dDkJ0qHZB+%pF@sR0M@)DI+!jVmHdbs+xd7|(%=!G1bb
zf%iymvp>QAg9PIF^5=-xWha$g*K_eor0MewNTx{<YDD;eyNzEXHkivM#bZLMs;4K%
zZPnI`V~d)CLR_svgG4fhFe!;n9d|zYE1vz$nz1hVoL;P_t}~eASyO|DA)eIU6D@b+
zA@9Z3!&g1)tol5b_rSEL_3G^Rl3-`p(S`yHI0yu~Y2SLE5+iOut!gKyqB?Y5SGz<b
z1}`!F^J9Bb(W(yBoXlLJk84e)2}Vc|)0F&i)2o_Zqo#-J_sny=?9>;Kh!}R}VOe1w
z)f*SjM$A%K8FI=5a#dg5;hR_nB?RZROhEm%d1XM>c<D(!O6a$O^o=i;($MJrMf)X3
z`9A|#<ub<yV*3m8mh&0L!k%8a(TCh6$zh72{bB~H^R5>1!6V9oAC&epNcTN3(Ehdq
zvj><bBuUG2gd9tq!^l#PUNOC84uH(>t-^6@V5bnbJw&-MIE>auFh_vtk%FN9O?_H?
ze5~kjz*LM`A*$!R6STihyX$B=yH%tU`6vfjG*WZ@J~zaK=0olw@=h`LC{DD5WAfsA
z?(Y7<!Pt5879-+X0qdUd)e)E=Ec1aoEWUWb!B(U>^hGHK$+FLU5bl7O0_aT4FTq-@
z41MV67mg2PzQKrg7;^yapPpkOzdXE%FJ8jIL|S}uvMY~m*&{5tzY~>=H0t&d#sB0_
zP=o*J871zarZ|j7dK@J#`bP`?-^IeDxk0>A+jlHq;3cL1!A~Mag-o=E35YLN{vQwj
z^Gbl25S&|V&#}pW*Ylq)quxP=q>wN8FWgctEM&MJ`USE7{P=gU|7ycu2vw<%09Gbl
z;=dXeLW4yIqgWRPFG->KUt0Z%zqFzzvPF^R{#P5Q<X@zxKu859)&HdxT=+{XF|`Hm
ze>DsyPAv5w<KQxNi~p;hFRcMzT7{J6*#5V1vHvj+ovTyse`|&O((0dEXz{;|BS!l&
zj@;fe?SE^9{n82{EIt3#Gk*WG?EgCAe}0eu>xln3p#E=0{NK{_|H+I9f{s6eh#OiV
znQKxfzj&)ZaL}d>AAjR)6R#s_N3A6}%)4!#gLFEqo8s{(H>hTFZXOFtm20JT`70)+
z!K``At_xe!m07&hf$ZsHl2x}UA?i-FIj!zKA<*|4X{y@=n%yeF^oPhwsw;LZabCkO
z-t4FN>6niJnrpbdo~(><Xz>K4rAO6~`uLYHTT^wK3wNE*ZulApF^nXfOZ)6}hI#>)
ztA0)CK%RL1-rop;a4am47!oTQc^iO>C(L#q1LIb1zhnG1e^Y6sCzy$tLRGyl1@K1h
z2l=|@oQh1}8b7biEv!Wgj8Srebg8c&QXdCzvw4Hsn|~F4h&PC_^)KH}{lRo8?Nsso
zeN%B(#RRB^J-*hzeq2RNnxc?Ko2?^t^IwniX|=Xvj3dGlWH7N|htL(o3MrCRDH;Ei
z>$15zSFkshbSis0n_;c1a&5LJR)C91QpBWBXh#r;R1m?EQ~0)3Q=(U~MjAt?!vLym
zG>5A&a@)9leR^VhK502MR@2_8@qqNa=`|_`9pAVW<{?}LtE&%s`Izkfa5P!f4u=az
zrf%0G_@aPwcZsj!XF=SMzCZU*v>QvjE&JWfij|=F7wEX$^uNc9x8K7sJJ<@#hZ~Cz
z$^J5{7hCP8VBNLQjB8f1ozjf-bQy7T<)^1hZiVCAdi{6eHlp~zPlIg6RCRV)KzeR)
zlLkl`nQX<%=?Xn=SLOZ=wzcxfNn<O(kt=tmPBjb~gQ)LvUL7yw<l{v(cWxMGLgA;1
zN$RG^nj#E`tsO79VTw~c)bkyaeq352yRim&)#bT%vpqIEg3WzSwV=4i9CN@yDUL?S
zCzUxE85`V)r4b0~?w-BBUvD-!J{ujE*H!M++;I_%^o>nZh=-@$RQCN~0q{JYeFJ`%
z?cF@M$O6W5{%w-QI*Ge6oP>V9o$w-Or|*3e$oS0U4O$B(DJmB!L@vy5#EiXgHETYz
zrAfHIf0d7m?=v#?JrNs)mmfMcN`OZsuKsBU*pdf-tKAc)4^v1gDq|fCcWanmB=cIM
zHI{uBPGBUXRlqD#U>-#%U<F|qC8uHCuGHP9TD9oO#Q*IW`A#?PC%8g!CgPD(<u1wN
z**@tq%!{}GY=mc`yepQL!NtAZ!M39hc`z82(s505b4XS_DmeB}TF0$kdf&0NaAS#<
z0l8u>GG05IMOEd@;U>rHlaV(g+T9V<X?w0q{Dq|lPdsJCltP*Nta^efGq1jF8{__K
z@-U{+oec?oiWdQ@*8<WNowH+CwC`{IQ5(5gxTWAD|9<HbV{O9_w?*Ws>DzH>;nCt8
zb(ddoU8Ua6Aa<a?`&cPWw;#>W9hV+0amQ~PUMo4Wcx)z-7l%nml{3*UcERK&Psm&6
z`b`8GOFDe4^-qs0^`9jnY3I`O796UT>B_%rl(O^1S&n`D0<@yG<6W?q%k<AKqY*dg
zL$b||eL8sfF^7~`GRksQ4Q*|ptkq7H`4#dw`JFmi=TJUuT!urc=p?4-c5tIJ^6;AA
z(bSQj&+Dn6oAK<!ed);IDE^<BEx%kl9~3Mu&mGM^z+PWvCNVvBo+Qdn%)eP)NXy(`
zOL&mp%eP{<i-iQ`EGl|mpxUY;5u<I)_%3wrS0&c_IoA`TJt%3WvM6`yx@XD#nM|<9
z!;-PCYqD+bP-!dz-u!9Ee4diSCflNGgZdgXgFi}sx^V6kd1-yv8yCH{zDRVp%us8_
z^C8KaH{MvU@N$_UMO+w@y^kMkgh-3<hUH9}a;ESKH{iCd6&9D96s}l!5<$-O@6A3B
zkN?0q9RCU{E5r<1`wKXoKa9@>9!L~Qd3px$dEX(eL`WxUsjj3D*l`5lMMue`XgQ_`
z8Fl%710b~7d+yMP=0el?xY>Rbo!OP4dP#-7`5u&j3RWU0xap_X6E2?`o&b&phH-Cs
zZ|u)|{#e&lp;Ff7<S)4CSuL-{Yz^~+(g`Ngt8yOs&g`q>X?h#G`t8IP@e91XAR@`7
zT)3EAv>cWJ_Pu&Af73|Ka~~Pfaz2Jmf;s=|v!AjZ{4%=*1;!%I$=^!|QY!BURn|Q_
zKAW(InZ`32iN<wll}VIU;1g1|{8GEsT7Uih0>>9g)AF1c1FeGJ?8k<Ual*VR_*E!}
z4@M{~=1pwQ6;O6K(jI7*UhJ-VpVj7m)131Ye7XKporc8c7FK$S$>g&Jk8n%0fUzcv
z7QLR$InN`%v}34`KuOQ%n!w*S1ph>|{2+%<FFhMEo-WfiFINTVKkN?rG`zV?JNV`;
z(SorWe;&|6#Ajc@gf8sYm`g{?Bzn^GYeD=rHO`ld92eNRPbX%E+!eAVmE$FT=XNRd
z$7si1kHD3MIa(N%f?v&XTjNjbxLZ{X?WsoY{#K`BN;$E&r@D{fk%Bw}U)xyQor<GJ
zyhZqk_EzWb+^r8kFxu&re$Gf_)1i53mJlR@7K?k(y~UR?R%XK82^*A0@Va_6ypm5b
z_Jxf`h!1qg_`|%F%@5#PjXVz5aG3~v;6dIyyMFNWsJ6S$NC+m7mjoRSPD?3>{T-HL
z0?;31p16_qn9X<~{jF!l$Mbeps$=!to+zi7f`kllR-H$>jbExX%ZJSNDi$9aWnXbJ
zQJ!1y)5Y&9eLG|zmwQ&WaFk!-<7P4muy37+pS2B&i2t*1wO5_eal{4}7-DVl^0U?K
zEwqT3cD}91w<`t}7C^8#+Kle%MwK)E3fuB%pv+zLa^8E$tl!%117j^1`a8hHX7224
zYe<(a__`J1<@=?i;F@KY&6PD>0;lfF`vh?;K^7^^mGTi0pH4g>8dbP~a!I35p8KQ1
zKI$l2{q7B@uc0g+l88pOCI8*^9BgID6E@xBWxS_#4&}LF4u~o^Otmkgv31h+cq=bI
z)qm36%+vFFw(`ZGp10Rp_DZlN<P|-jI7WkWgp;xwkX|K>bB>8pND&`Rm?_ER@ft)-
zd~`BmOe#kk%*9hc$TT<<qbR7&X_p_hRCy_b&76c~Yilcnl3n275?6_CY#s|q_F-YF
z8nLTon0t?sCW`ZFUccmwBK!%ti2^J4XB6N%Z96n+ktSPA6&Wn}b4XLuo7emqGLai>
zkFLO9a}Me@T;2u7#9|I{G=isd)uSBI?53VtKu@F$^nfWRW;&$UbL|vhFPicDMNo<E
z$!{BJWaVmuYx3!6dy4s3p>Mo_-XtWGNYLRY>R6;yB1*}0LS6J28c5Q|om1+|2}p{(
z?^V1<!HIh3EOUH>OjE`bJBM+fC|bO8q^ko*Wg`5uh2DpjO$8BU$^USNKmSTD4+69Q
zH7A<pX`LQwK1=F#21sFK--CiRE$$QvyFVRkNtX2C*hY@~XeKF3urVsdYncEt<gXKu
zypf}F+M4%axIV0}i#ED$AoEdhd$mtH)p#^3aL%`J*~QTmxO|K^A+ks-_sW3gGc>Cf
zy;y$2^KhS1H%5l02O5Sh9i=a<+#vD7)PE$g0xd#$mO^eGbNPn~2FNPO4n{u;AUu=|
z{lK;E(!O`qXSruH^8I*aOQ!ikP4s&jJy*6F>9Q^2SO|`!>*Ivcy)Nj#WZSN%xrmm(
zQ780|*n{rRc#k`Ve%HP{2-RRSA`L??aRLlJ*%P@Cv+1E;pRT|zLoeCU%!gGmZYfwP
z#c9`0^3J1s|61c(7r!T{adz>$^OzBJ*|H3KsjTOzDput@Fce%~a#8O<TR1t4hg`Ud
z9dUUSrQb*5zdn;&4_{mlH<OoKaE|wIjBtFn*`(wwuJ~nmrj<Q`nm8A#AsN!wFi9A*
z#8UU*b&_p)lqord`UYCm1drZcN`K<@Q6u6FEab%i2vy4opzQd*b>EQ<Msc*N7Tg+p
z`yCLRm2O(`P@@pj4ygrUp=XS9Niqm)pY!gV3nD8JCZE|dBw$%`q_J%V#WIb#FXu8z
zE<|v3ASOFQ{JW2O1u2erxv7@DqIc+E?6H9UOsuGcm}3%zMfuw8EXEV!Y78W*iT%|0
z*P__Ofra9ZrWr!N9H*Da^@HGNDxbrfjYq2^D?T?4sNbaFO5JQSS!<gn8_7&~>>*jj
zqx1|ZNPO)wx+S3Q$58S3xe=qp$bcqnIOm<pUUYK70paCBt5ID2{5R+L7-ofW)+yG6
z?98s$)Qsn6r<Q`>AHSZfI(|b9r!sP)$@JVDM73f`K~pM_{Xj;XR=11q?)5*H;>DM`
zT05v4h1`C32r}1-XCY_`d^aePNdGi1+}iXM+q7J#tE620({AK)IF9>0s7{+UZAbcM
zrnXUKMMv-`IWymf=TL>~HF%5+WF@cBPZ{)@UubxI@3fapmbrr;j*OH3jo{K&XrnU4
zY;0#G#_?*3UYfq4YkXGXVnmB}Wrv1qNw14mGU5%%;5$sn7E3cb(33gsleW*$)=uEh
z;hIrq+UxZ1E$u$HT|DO5uz{W%gdi1iox05;+x_d&%m{L-^-2e?PJ`yRw;t82A-nHX
zxLBu-*4pXb9KhC!s#O}KCg|tAM6b#Th!?qf6y%J?ZqV0_?|k+CP-gH=65Ky&L#a=6
z4ps_y8iU#};~VAqV(>OJbnb1rcW3rKm!t}|^9-)~oOYT^@MrIjAMU`=C5hQM8}@9V
z&IX}JyJt#6pKnqEZ88Nsxv}jSS1W^C*#we`a<94I=?t=4KL78hJ%pf2C2}l|2BLB8
zNV|6%zu}c7-Kq`R&kA?h=mxd4rhd?;NCmwwDKHJLXw-RgE|6y{yz90rUfmdjRJv6o
zMM_9(v%Jo%Uvq_~9aVUbSDYwry+)3E*;hOJuP2u2-(vM^WAQi2A>(n1(Ru%IU5`TR
zN_@XPI@Sn@HJzX%HMiX9Y}|y))}bw|_!S!l$(h4$T0AuLIaH~2S~3>ls?kg`x;o6f
z&*Mjwo2fvbiKYXx8$OYB8w<XmAMu`iD9TPaj(ijXk^<S?KN&*`1WNNCi|U&9QY*x>
zI7lIj9@__f|DaVdDrAOuUL<0nC7oWH*o`UQ8HA$&ZdL-u_lGJFVu6CcXwEB=sA#w1
zl9IIK*iZ~nkwqqZAabU<<%+*<dhG2^VG4qcL#YDo|4E|-U_|=YH&;9F@;AIF^DyE1
zL%ZnSxL1fxCoZzL8>^{guV=c%k@sxA{QLva(sumb(=N6uDt8s+FA4Fw7%g!4W4o!?
zx5V0x8lYUrb!?PnuTn?ulfIq?q~Q`<@&i0H4rPUvs?jdgpkOA?2TRdo$=d2|J*(fP
zEN(<eSvDF)&t5y~oREeOb1i@;qzWcg#hb-+(oOpKxWmTl#<>Xe5i7)R5%_9${c>*9
zMiqoG@&s-a{55eHLiD&f(xOHP<2f<4I6zDtLfSY)VZ2P^o%<v=t3ovUm@*4<c6z{t
zhWRu5oktov|GZQZJFlD2N7NULD$>ziNXy`l2(#ar1!ICE>h33foAS`^V1aS3H93~+
z8j#WBmn(a+p^*N3XfhVW=?~3%{lI9hJfEP9tQT4A)UNJFz)ENb%WRrnND8DP-KD{v
zzuo0aMtGiw#Q~f1lRxk2>?RulvN8;snHvRPi<hR|5w%7LH*EROV9?DuQqc{rl(ms!
z2GtGpq>ZaT;$rXYZIy#MDfkH3^F|)lb!Jg<f35ZAV{`Jj0AlnOo#O~R>vp>2-J8MT
z<D=2(KFSf|L+^D33<*w3udFw8GUb*HSJ?9Y>%I>CBfn;?ZWny}!??;6-i)ROPN!P4
zsISuz)QX=YbEYaMTa?irYhhOjPTq2Myh!Z+Se57ePAwU4T&(_Q+dh;>o+AeIU)ola
z3Gq5=le#yILdReC$SZy<8@$Js&SA3CrJ|ac)&UsjwW;4Lg4I2e$~Xl!lF|;x?>!L6
zi|ux~`APIPsW(ZitXw_&6wx8DPQ1LuIPe4ieQOU)Rbuzxn{Ul*=?10VO&!Hq&0z(k
zB3<}#2AF&M4Cm|%odTTBNe7xcdCaP;ZTM)X)jJ~E1Q~%$KB`@;_hgQuFNPQlVUI-^
z21ah7Z;kuxsmk|O$;-BL1%{y4d4Z5*0$QkAoEv(n15t`<6ylAB&dkJEnyoZPjJekE
zgSyDHoBoX$f4*~$VogCI!cl{Z@Fi83jQvxs597W7Q?bE7DF)U@V6i0xDgL@yJ~+d~
znAxlZxwO)sf&QIV9OYp8InlT9wM&$0vDX#wz6LP?ReSCRop%{OW_>U9P0$PwX=tJO
z+e9_oEKyg(Ud)Z*qG{QQ<yA*$11CQ`x)Y?$-C=?t$g2#eA;x)e40z@##*v=`8yYhb
z-+&VzX<vqii#+=Gugy@e8_H(qeT_+{)mafNg!^Eww}LY<cx|5(+V6V?)R%7m7FWaV
z;=*3-izNWo*w5vP*Z6Z$>f0D(K9cH1YsnYJM618rU|SiI(G_4|8}weY#ja7@w2#s4
zL!ul}A`oN#b{C#AcWHpU)CH5N_<<P>KE>L%$L!%rpNQRhp}_TX+8-t(fd@$#NjmoN
z^)H(7ztQ8!!iKI+A*N-KNU>dV@hc>EW#l+eF%~u$TF2$Ox_~DJ=~rBdXpbc6izJC|
z@Y%PjQr#FncY@=a@#j&Jh<P6e1E(e;##4Ud;{fG%)nhAD5u)hDXiay?k^sL8(fvqH
ze<O_5g|1{ru4jT+MZ=rx;97NqfW>M*X%t0NJf0a~Qm56X&uJ_6{6=_}BQ5Mzj*QwU
z5df|9%$n^IGKUThO@LDJaKr;{Pu@efa)q)xWbabqt{IW=>bh2T=ZSy;g-DYDWp|4s
zi0d7iGFn}K_q?1SD;QwuZOtKaMA%~%qCDp=NDlXvyGDKNJyb8QKP&{mP9lj}A{SSj
zt-!aZOYP7E50H!LvJ^{Dy!49P=S_kL?0emTqLZ0O5Mx@q5V~EDggc0N<R{7S)1iDx
zy2XIc%M62j-b5-|R>nBrBkrTD>U(<bR<%n(0JJ*sUW&fjM8vW$wwtiN=v)fK2Jui)
zN(%^Hz=w<Ls&%sv`J9rOl!Hd=n&r=~rD6YGU5ayVXGt6>H{4gBdyE80R-9FQaR^k=
z%)A56hC7J@`i?Fow+3i^vewxRXdKOASd>hMqmmA-Piy<c9C}6c>EA{}vU^ZEUM4<N
z`fPT9AyUmQuk+8NDZBg~&PA|TsXvm4>oJCK`bMye)Ka^X*tJaHc{1%Hzt+j*u-sRo
zp*hE2R{U(%ynIZU(lCL~st_il$fz0Jh7WC>*UV?_7$7Gjrof>$7F0^%i7TU<wSK0t
zd6QGrO&E?M4Z>R$T@B5r_x-uUyHq~Zzt43AJEhjWtQrK5TI^o4amm{r=pa+`+28m&
zrmV>_h)<C;ItVt8^TS&S@665I-zs(acA<c6+bZ+I-BDACtd9zBYirJO*}PrCrUO5O
z3KBVU@nZ<QWPM$?_3E(pJ7FG9n?X#;)tI^np^Nm~iOLjP;&P`+*xsjkgBHCN{N9eK
z+UYD-LWEv<y}P+0H$6S?jEU!aGk<}_`rI67GO?k|aQV>Z*t;1pB2BVFO*pcz&wrz^
zu~$u?VE0qttAs{M0m~ia;$8Kx${1#^!(KVb_9;5HlbYaBpF#`=e&E@CL}kRd>Lrsy
zxw@88*%9eG4n)a#j*M*NWVkLOX<vBe3&sxULdzWd^Rs!DG*-svm#J<UV~tepB!JeR
zHQGtY=%8=|q!sB;_>8Yi>Rbf41OEH=ylNB|o32EwasWyh<h1NmvhK_Wiwa9)8Of&h
z3q1*M^{PMKKCk`u$6?7Rd7Y~50bGoZkEwL~#9=g?2#!{58*O-0ZSRBEs!=GSJjv|c
z$-t-%)q0{&SFe*87d|=)Eyl=<fXsD>{>W18R;%4mP}#q&cqX|PIort&Y%~ngnvPYn
zN9&J3{8PUWqVwx%EML`w0fixcFHk-9g6+l?g1I^!T5k{UW-HiX5U@qei*sZW1+eo2
z(R)*bR9I^k9?>sGU*zukEdu0Zo}f&;a>f)7Va=K~qp%ed4!6}RTVE_&NGC!#)H5y(
z^O+<{_ww_(-BI+xTQik;vK^NAQ2s_7^`3LK+g`uR>nwC1!JKG;lxVb~cPvCl%`mXY
zGze-9-tj54zk1aA43i7v`c#s%VqRK##nI<Lq@wQ!9>a?VGv`yxHK&j)kIBqyEz$m6
zh8JOot5Z8kKuV!vik{p%IVJdYJU%+ioCl(h2zS_Od8tl@_oSE`njF1lO`oLvjj(jz
z$>RAD(*=PS|6Gt1+{-3RbX7bIvj>x(sGgBoxe`t0w*ZeHhWWS6$7a`O$sNj_^RcTQ
zWqVP9Z&SrA&$7p-5^*y@zRvs?{iUL+LLp`^*0F;x@29i>KqZ5pEISfr<oo`qx(J3m
z>JL<elAp_@ihf6L7?9y(_m>K7XbSbVNcK=+Tb~3T;unGO`CykYkFHMuSR4oI0b5|V
zFi^ovTVjq2cJfm%rYc0$r)@pGbAXm7%&B2L(Zm0H0dPa98Rfm{m4Cav_sUyQ*lH?V
z$iu7hFkRRD6YpBv=4hN|`loFyk<p@=*StXk6j)ie!Uc|3l*i*LbMhOe45Gq-@60RL
zZ<|tWXhAZ&(cU5I*2s(CaQs(EJ*3CKtLN??E_FmMn%{G1^oW32+UcIz7UOsh(?sV$
zaJu@%x%EX#0{VHHeRu6i9~7=d{Tl<gvcrKE#54j<wBNfQqmeGYEz=B@MK>G#KnROX
ze}^hjU$$qp0pc8s%}TBI<dKr5H_Sbk@U#gh{AcNF)tlF29BhaUjB3qVs2h1m*{A)3
zE#f;mHMI_73CAk!SY{lXW+0S{p>gp#>w%@m!%@3R%;U>=y+8z;YGlHLPgugV#{Te(
zo?a}6uJ2E0-U%<3NJmg(o6b#T7El;?5xW{~dMpoFZGZZ}=cb+Pk^ScWnING#r2FJZ
zZCw%WPc^IgQuJvfAdBGsI|3x%Hf?_wull~eKuAA$GpjfK%3@mYHo_lF#IZ_t+O;?X
zeQ8NMtjHNxd`__!eqboOzKr4Sdg%xK_RD-+YO_@s*{p0`B&8S~i&)_TO9-EFER&`Z
zn@fw2*G|)o@FM{NVDxL)NdG!n&71_@P<4MTRUGK{mDDsae{H`F|14568HL1x3-NrW
z9J}U_c<QRU#7E}C=M5h5M4e<n1uVXGKJGidxge#<OoPWb2Ym^@TGy2MU|AAeZz&rm
zSfG!09tl}y!-RY`@wEo@IW#xvGqBY92PL*-XML~Zak_6ElJ?M0(U=Q4QP>9y1&4YU
z`wp3-jca$4ThE}cFQ0a;gqp~ykt;JqTnDXD6&#MZ&9%uDihd#7gF2g*N6Q-$H)J2M
zILIb!{25QE&`7j@-)vrxBnPT2Ule0+?Ds<_TNJHG<VvZ)Vb{lv;R@>lV|Utz#k0Wf
zjG+DI2JuPR!E<WJbZcwZck<l92JTayr8D5gYrAznWUNbQLYShUIOTvTj*XF;7&@gz
zgo9wbEEk5MYH;<1yAZ9{3#YpN`tc`i8x91thH?(GjgER~qO%&YY{*1C-26sBWk{>a
zKAN|ly34omOYJxZ1Zj1ekx>1D2w;vo25g!a`rS+%XU$K&QxD3EKf~B7-fqQ4Mv+ul
z-m-?IhEW86lm1BwbY}vK<c275)w~|NlDPmEOb#z_9G$Jk8*cl@f15iKGOu{IayU|<
z-^^cza|b^Py}foN4T@dD4}H%NaIdHA0)|$#%oVN_<;>O9ZCQPwsS059Lt_7!nYb^A
za5Y~OuqK6eliy|Pch1*L^Je&U)o(t%frvZ3gx0?nu;i`Dw+ed29HSSSCPCSg8t{9$
zdkK$9Pdd3go+wOjj~Dt5bqi+Rh}co{c7z#}nV<velXH&>mUAmXL&ihnC%x3%`+UJ%
z3u!C?R)eaz=(fzLU$xUB?cVa6qy+{gA*p?C#Avu{^o6zAi<#R*cidQo3h{ah?}_WN
zwXhhcbRgDc8fAMs`0*=bx&03?Cpx{AnV<BJL_R;>zsV&kAzLO^%M9DrxnC9sdNcNg
z(01q{WM8x2QB^NkKNw%To)gx$?YVB&OA~BXQ<{~RVWl(GZT*2iy^M(vZfj;w5<*zt
z+N?a3d03n<#P@_V8qAT=1uGtP1aRbs!$)n?S&k)OqnWrWGG3I%QyIWFL1B>7N)`kD
zz-;d9&v3CaPD&Dd&!^8sv<oR4kg<+cS>$(D;eoILD!+)~1pbJ&Ct}s06>iGcsf()P
zcO3p@d9Whr4U0Fq!T(`NBSX!+>TUIF0Ah~X+n~L0hU4=wb=w|l;ZFUczj*rXtp&wk
z0w=<Ccd`2}i|As{SMF>x0KG@YR!+aY98lI+&@I!mPp@|)zW{J+KjeLLJr}0AthzV9
zv1nq$G9pk{R3$?_ACz@nAQH}SL_WGbV*L^7#tE5_aq4t1`UWY7{sQ|<cC`05S1F%u
z`&|YuhSetA=UHA80koNaiiQb*#*nViC)LwS3m9c_YbcuWZAUE!T!z9@&|i*NMyY$x
zfbfYo&Xi24qWdH7f8&HI{CNl>H$Bs*-|<^IGEe$#_KBErD77F>xlWLu)b^R6W6f%M
zn03@5?RdO>&CEsxq}7A_2g`*hT;D6tZrc;_BXqf-Ly_6gMWG{kIsq$zzWI<L4y9z0
zGfN7YXB2$-NL(N1Up!%|t$okS#4^Y<Jf7*rqh^vW(hTvsi%MiD)5EER$!-T)O#8#k
z!=qcoT*zER-^CkWiPGTpNX}BpKKnSEaZ8`AMo!VA#Gt!FvmGzp{Be7%f1Y{SK~kx_
zKbP{_(VP%Im*j*lu61cEhhdUprtLgPq7%oWw#ZJUDKc<~1K;&5kvg{jdr%jAKY6fA
z?T-Tvtqc@@RF|jmi$tPwaj%z6f1nIf*cnRkIxVNrUQ$T-@;Ap-b$A5zOOb1wPpg|K
z)7(k$sfMgczjFbCS9$}2l*gZV3{5U)&nJ_nSB5_cntLu{R@JK?y0~20X8+#s)zaEl
zhCE#Tsy~@^i~DXsv3If#1(Cv}EKUBWVX!fXj$*T~X?1ZP2_uKSIJ?3rh6pH`d^qX9
zVbslVtXwhTK0rw8X5%TOYq7k~%?5XQ3RGx?%1~@=^v%Wq^W6#BwsY!6#z^6do%&$$
z(wlM}hQpoY>T25)etw%0beALSn16nq;%epW<2zDYb}|jrHWjCoHbBg&?q+ui-ckD3
z6Hb&;_9*ORXY+48*t5`M34vJid(#fR+(Tz@LY!gjO&t8_r5ACoAcODkDMi}Y9K}x6
zslep*c76>HM@|J>WD~RswhqMz;TeW-k<WOHq@bJ&w-3~81B#KaEYrbFTPoBILL^t~
zwemOf{fhodE@_WGluhR!?o)&DkZWqq9+^pWVUV!mnONKuo|!>&L}XYFsmCQf0$u>Q
z;TwZ^!+XGF%d{tIBT|iWb0H!xX1!dfKFi=z78+7~QfACYuYCLpRsB|Y5PG0ZBVswK
zooM^-OLLY+SNwNbiWjjitaV5Wf-pQP^yC3DQ)loo%Orn{$n<Q9vPXxx#yKw_%XH_=
zf}&9S$Ko9?=tEO4_ENs{7iQ6qQ4>wp{iY6zWq&kEQ+D2cmkQ<<xKdfup6E8}Jm4kF
z1<P1gI*s;u{`)6mh2`^D5i{5uR3d|{P;O`N9;OeEIo1RVft_7)4uQ9XZiolg#%|eB
ztAG(<L*IpO|5nf<Z+Dze^Qb~<V`rrA36n|n_%O%2h;Oi$`a_(`ZhFM^d8n!&C1*h=
zIb_SKaZ(<_KcYU%ToaBl4*So%l!>Ygx&Bcg?Td2yZ_Pe|iu`ii%YwD_Kf@~u0#JA!
z51pGTYb%U)9A|gShFXW4xfWEE@i`;i19B`L9T+Bg+~w>SSe*YxQr!572eXE+rdpW6
zB4e2j7*a-;*QXiY-zB!dDs-0j!Ko2esAhb5q5t$eMo{lOsH;LqJz@fBHXnye+36x3
znXs-6&;fXp5dIB3@`QUm*b55UsU%u`D~<p;F5iYr>E}zI&sdgJSbL=|HkT0Z?+S*s
zGfJ5a+pY_^v!nC8Rm|sY^rWX<p|V2*g<~41-&S2zmq!t`qw8>xLfDm9<N*^(9PET_
zSuj8d@`AED*F)CftI!AgO^>+mmDO7a^I5bB#;mZ8aG4aZvtZI87NXD-`W+O6uJ0E(
z-awhl+a<Tv%^!HI#D)Eztu^m=I+pl>JW;~~A#)pRR`|{4uikA%=<RtQui28us`Fr6
zVLl6~P!#%_iGf<%yKD*u&YATCK>@h2!AP=eXG^cOuU><z{=Bt2>yW9H?#)Q6$tdI>
z93I~l8rkBmlJT4vnPrc&jm#xc2qk!e*6V{Pg*Ve_obUq|$utaV3c-#oScJ^P?T(om
zGoFZsb6?Y<fU&@}S1iRkrk-q1Zv4@JB%B#r0oKbc&{6<}QuKoriO=m`yKLpTp#Iit
z*?IS1kOwhl*YR^kLQZ&1Ya_n8reH$+4Gg)-M#HPz%QAALZzamN;++!>I{b2)2`UEM
zfv?G2oJTY$6*S<h)35vduqmkzB%%`RKOrPQ>Rby*(nDCCJxK0nrubWfjFS-MU9IF;
z>#i>C&afi}BB<C_>iA<hVR1zLv9^l^vTFYzg)#UoDgV+ja6Ponm^baaH5Q7YkUn=8
zF>W2n!fBCzhzclpXoPjw;@snpF_>#Ashmi-Q1U9O@b4#WjQyUTXwnZ@(2B1m661K0
zFqB~#2=6CMow{B;$(d@a&OPU9d`}+GA7u46{&(u1Hk|9$)7EpE@VFNQs*BW|c~Knj
zN~6gc?Gz)-a;bi^7}~fYSvGXa?<zZvU71?j&}*sLO@OR451^uwAd>d4Mnw>26|Jgs
zT)$4Dm%B0>j!r@JMy8_w#DqkDdWr7tTLAqq3c-l-Jv=x=02y*EN*bEzHw1xvstge>
z?rXg9DE*iSwyZ~7X-|x~-wIlO-#um8jZrx=1rPDVF&GJjy7d{_gC;lPU+BWRM3XJ>
zS*4&{Gk>}x3<K>{m9xhscspsW7%4g5_c)w<uziZwr0WDjf8Fz%&?X9r&l=5`wT*fM
z;nt5^a691Nw}(031?9Ln(%?AN-Oc}U1Df>xSc(MA0L4Y#W+fz2kW}E8SU6!<uSE+b
z1)1a<nR$;77G#RU{EB=0-g>{0&EpoJZ=<d}b5=;A#g^|1@-!Rwh~#CcwNLi-Z-Jre
z=|%fq@YV648c<Ql&?5yW7UXqkV8UhYu`gnEsBo8DKSL#dz|_>MM_M{Ax)t=4GldPi
zhHr>`B^DPsX4h{G=wdwVx(Vn29V7hA7Ql=yS1~>W@;gPq0k~_tv_Sj)ea_)VXtK?W
zkXr9sFD&DUBUj<yX63O=7t_&9j%0Fras{<?PxF70WQB>m{XHM?-@MJXwzlIw2zh3m
z({Naq<Paq!nl>Kuk?y3b$OzU#{1c+fHv7POc!?CGI~)#McawixI5$cCl~gPI_6=|<
zEhhYvR2#Okfk=dnunO0?p<TYnUEW8X*ZSz+ID_VRVPVPU`em=UUb15KXI7}Qy^>G)
zK@EQ313wkLC}FTuEJXGTXQAHL;W!u6<C^dd<mGsio8Bm`IYR`<O!Wd1kVf2{esc`=
zFKQnKEi)(#OEBXN4nsM#{hv*UTH>lKP{gUojE+M_?WzB!J0Y~2>AD7AG!1K<$SQlZ
zoTyh_G2^VAgeom3Bz!RXpkE>o7s=-I2eE;AzjJ_(5>)<DheoY|KE=Hb-VuyC*Cyo3
z$52(MKAkoLQDF6OVk#hT(}=EHVwvA=h-3W>g?^{q37?Lw(v@@8ZfL-tV<h>`Z-GN(
z{A|F`8*#H6wjy6c!SqZun(Snvz<y!_Eb2h#KgVv4;o2zxA`JM?WvwrKh>yA$kwa~?
z(?1aj^ST$x-RUEeNFkE#okaLM?)Iw;o7l#lLyK?WmG*PpP&@yWE?qijcswiTW(eEi
zaP`K%7Nd1d`!qn=DHPl*>Ivwj3r-Izv;7!WZ_U3p-se~f3X0Z`2CSLCnX)51HS9Kr
zVr}PE8Bx;1GU*x78Zr|xiT4Cx;gI3~D}U?sCf!?EbS)s<KR6-@>-7v(aL_qM3-$(f
z&LT?dE<^u&&Sn4X6+?xsFUEoIEkO9h74u04srCXerzr4G{y9QFrCTXJ*DwE_dl)S`
zs$tZOfVD5?QD#%>LO9-y36{RO*~~djsv>hEa2qW62!cBE=Y=(&A!W-!4>NEC&vs0Q
ze)?y{@QF(V6{zWjr6QYfz|!N7gGP0r&F31s*M9#BYgB6{6!$-+(!m3n)<u|_Dvo~0
z+uOB^P!ACP^JbwI=;ZEJp8rY9Wmn#evXy{>%CN{|dtgDeu>K|j{UYygFyRuMEGm=r
zgA$E^{V;=yLmkQxHHPe**~n?lWIeP)m$vhe27;H6>)fCb9Vj8S*2^kL@u920jb8#R
zqL6f6hCXJ=z>g=+ex!j1lg@@6KRO*?u_yEo*@h-XYa=j(7b*u>ioK&o_#q%ow1#hw
zIp8<yxd&{(`=oEk!z{+nI6@o@W4NGDTAt<+Zs~-k1bW0A02s}-fOqVv&^=`s2jaXt
zST$|FFb&sbVIQ-CrO^lcowqzE_LAGzSzL7j3e*ZIiLOK`0Osc!=yurH($hRpF+`ep
zp?TWlaxEtst#z7%yL`l-_#b5mpk^ATam;0r-;b__!8>tt>^F9McJYD`2{8u@kkIex
zq{)X0^eZ1&Uw^0a63fL$)K6;XzO5G<pGia$9m)c69?x(Qg4*B*(g`tEtcSP^r(>%&
zLuqb!u*LUNo_dPF2+5c(2E7ngtELpVNVoL7F0batyt&vr`QXt3vP0nD9mZQ$P*NIn
z><Uru-JTxycKNmtrZy3Xr5L<iIK8X(Rd{b(iuA#C18im4Tj*dI>zP3f1(szJvBLyw
zf~ZR85o3<SDK$<?gVVdW((~YT5ghN(JC#o-iJz$0#Li=ElC2?O^INE$ggy7SoIhT}
zzSwBl(5r)<TTU&P<U(CR1rU%S#b5iHAFE1)FwwKharX+ti$oV5zkQvqOjK<N3tPe7
zU7zsZ=8@}rlDih6LVVt@_t*BVh;8TvClz_(4yR?a6G~_|0XuLOl*1j8jCV$(u?}H#
zY>5s2Z}pdVu&YA_l1shUwsaTAU38N{N7&PZ%XcnE-C9}abN)$%Ld|x1Apk3e%m>$=
zkM*&XJ<MAl!rzBvO4;5<efN}iM@N>wbXXWEk-EO^p$&##7e0UQ%{%TLIUH9BddWh6
zWjD&0a*ObCtvP>JyiTw$T&F2Y^>a1kgA*$jGaK(nBz;m^@SONrTIPGRyjF*r=i#(d
z^WhDWh~$l`<8ztv^UYRt8Be8fG}dc`W7OY>k>Qq*^@2rSgQ+3C?N$|FoHv9I88DDA
z%uDlA2-FL$&Yv8+wSN2AcV4X~)r7&++g+rgh*#=xlcF7iB&Khw?Z1w!)$fCyNsG$5
z*nUeay@Fyh$b`&?&_2TW%+fJImRH1;5nQ<RZzvuj>(L46m#?mDlr-<zw|A@Lz*7w^
zKrFLsJq11$1{o<T(=(YPVFUBhwp~`UK&y5NQJm>J-(U_&80`%)n~4o^tZfuCyP4qg
z=eZ7?`>(}D-8dYPgKu!6eW)Ve(u9j?sgX;Ge|bZdS1ujzDA`on+#59dC@ffN(W21(
zDl@ktD5W0J0E2-vhKEBvAU7r^Cjsk21igQDaX)5zIocJ6#1>fH2J>dsKMH(iFMVaO
z1NQcJppb?j4ext*QOP>$!Jt=p2q<R_7^xq=3w}Gtu~?(`&j1JQzA-qWejwMoG9_l@
zm05-h@YDDPS<`L?cd2GWV79H;7k0w20n29EKHL5U!T$;i;NLBOzeu^PerF)9c~zWl
zz?klJ_cWfz=UzN)SSJAL1i#X~l3q?2*pt}Bj@A}*q|EY_AjpxM%X+g%kEH~Z8Q`YU
zKNv=?LfHxQ3NIV^BD(r}VOXhEHP_H?Duqe=X?`zNE<{~FWV+PRrzW=`WutG)`10@w
z<m>oIvW(*W%oj1{l_S5{e5#uf8|LYhfg#vryBl1A0FoLsVS_Jqe%~FzG*BcEih0o(
zb1Q;5Xz8IDV2%p%pXO^YDP4!Mwaykh_dibY>B@abR9LSLun6wEP#sMWE?2v43~{Ys
zz(?N4e@%XfbsA7EIKE$O65ktQ$?GRx-iyA!b-B{n+2Os^Ae8kx7b(&Ch>O007(hv_
zw}VpfbCfgQQ`V{Wuetw$E#>%@J)5m(a%n+1+XOce98hRvPcLArj;=32K$t)-7_h{T
zTnz9(;Cc9`PztQatzb|eeqhDj)`s;RbXV~5a=@quD}iKtesY_xXX!exfiFQ4{;OA6
zU3Lx1++RWeb()N;GvJ`mO<a7=Z03nePaZOiKgN_ze?AB|NBV@Fm|gL^cHqNabe*CO
zJ(5bG-QJ%@?l0dA#n!TWcZ~eO_6T{V;cz;}cQ2F6fj3E*De+vSMsJRatyUZz<r?t|
z0dCPv>?@<?OE;Dljc%=Khvn3(j$)ex>dVovdo5Al`<z)kNBProTZv^tGIVbEKjrC)
ze)qp7L$r}?8tfwRETadC&)<6RYuB*s7~NZkWg?qm`nJgN%iHL(SceZHgG?2_Pl89d
zB|nVQwEbELTe~eE_4;$ZuYM8K10lNo4bvRW&{tb5^FWN9rlk1a2i`1Vsd8S^Mm+DW
z@Z%jKb(85s_&q|^&-o53Ab-Iy`@NQuj|9Y&HeHh@kRDv8hprup?RT=isM+M}oGz4!
zUf22}2#^5RxENE?Y0(h*77*a^FpKU%eLg++cU0WbWS`i~;DB{_3PzMD#f!6>RNQ{E
zk?GA_FH$u^lCzV6On5M9?yqbQAGnX^H`j2g`Gb-Damim#Y4m_RhK%vY9d!V0TeIBs
zhu8<x9vSRApY<d8Lz%aE<&;@I*}jPRTF%Bt)!C4P0B4(<YnL41iDN<qZ!&1MXUQ~X
z>?3q%tjip3ef^v2_&d6sW%kL(`=6LWuaP)2S^Vo+r9SOBp0Y8(66|Blp9>1smrXjw
ze$6LYrIgji_cPuF+UZFhn|wlnSguVqmHSO}_zx)AicTlS-24KlsYU2uSK+O@)OYSn
zGr{^t-h*4sQ8j9r4~tBen0|aqK5%h<u#ha9ZR#ADC`$Oh9V}BKV}f22x8CpcBPPCj
zoZzlmJA5)SBl>q`i@*M8;LMM{QRU`E`tkG7K(nIJ^+bblb`<iz?NvR)D*XtYlu$fX
zQ}1A*JgTS|WWoCxA1RIsZg@u(Zx>>?&60+!0udc`Qe2KI#Q?{5;gO5hXx%%W`_n_v
z?#FlflA7Mhq!O?96a!+11L>maNP)-g{SzM#)>BnM;5S`aacF`!VB<N1N8@x%oaEp<
z(L-`88BsqCf#zRYJg})brnAEnGH-)ZO~E3s;PY8p>(9_IC$MfX-3`_kg}Bi$fgyqp
zeh0LXo`4r|Y3OJwA7rivE8MZyWOvMVR!a3BJPBM_EQ~rI<E#Ze?EOQ51~C~=ylY8<
z+B9q)!&cXC`_`_6R)pL>xVYRri1GQ+tMPw;_cYcRQ6;^l*CJcIz2m3ze=`Q#ziu61
zyMq|Yrjf6_8#tf$vV3Ad?$qPnm4o%*TU2`TecN&tSn*E?UM!9O8seJKITLT4md0y1
zq){PvC;lmisqujM|2zZb1Zj@Wy~*1r<Adpi_H5o45_VLb6Ki@LJDDn*^$>aOf;+*c
zmvTC$taVr2@CTP?g`x;$G%_7E;x|W{Kb2Q;j_>){RxXsWm{T^s`8Uk~*PE3t)}DZb
z?Ow23duSfEmUE9$3EDl-e@!(vem5p@2NEWGwdtOw$OFzA4tv*+>*pqaIPQj|g9)C6
zi+GZ7G~rA74`FrJ#{g8cqC;lh9do=)WajRFObBv?#{Sq>6=SC7Rd)iCSne=D|0vJa
zUcgJ5ZjS6ZLIii&t(czbs57fEwN8(DWB5eCm@q%~|Cy?h2rvJgP_YfidkOBNf?ti|
z<f@VTZ0X(RU#~BeJhe>&H)a68s10PT)XeP0hC}28q}G^;KDk4*PkMQU|H|S)T_^Ko
zc{ZE`fr#>Hg^a%H`>l=L*zGo83wBj8{o6Yty-D-7am2*$ff~wi*X5sxv32t@XH1s?
z!`?g`<jPz&=YdFBpKTU8a84XGSp17V@g3`*0+4w(O8;Jb2j38j*1^pu>J@)LZhS|!
z+vuNVI^$-Wcw`rCH4qTpSQuj=3&%$?;Jp&j;DqwIbz1pG6$zEAx@*XaJzJ3-%D+_K
z3Dl2HKNF$Na)$5KRYrU%3&}JjZU7Jm^Ung;$_U`eYXHm=<*^4L{{DGRwF1j3ux#(a
z`hpIGd(S$??VFFuhTpJMKgB5^5!`64PvL#Acs8N_7VO3Iytym1vBxBgo=XtJiVN<&
z8JQ^bn*1ILy>U5qO&xN_pQ+e7xoKpGerV--+H*z=TPeLPqs<a-u_3xIct2Ug7{>m8
z?VV*)9Nx0-vEXily9M_dbZ|n@K!Sw;gA?3?2Zvz6Av1Wez~Jug?k+(?7$CTFa_-*e
z-u<t-AK|Qb>qU2UcXidO?p5`B_HggL*gIdJuX)Pj-iKa3_TE4OENe!EoPV^`2x&~A
zeCNZr94ulj>Bk4Z@T)lq>OLuWddBNo2qHrJ@m4{g%q954UAAeUv4PnM^Ydv)&i-qZ
zp2&fx4L6K>@02ye-y1Pj)DIIg_3hU7bxoUzC)a?@Ij?HuxkAD-%y`Tb#)iJY0fq3o
zvz*inS-ULdP`&AQ%Oo$3=keJUm1J4ifxx<V1yZKKJKwJTLLQZ8F%jFm(F3p|G2_Gi
zSjeaOkfQj{z&q1-2D&$ydmlAKzfvd=9nozqT_niC3o^VO_DMg2^k`QN8Nv5}`5S#?
z&KZCw(<VR-ZD@Jsh3EqC^ZO<}tr{v{-sjtU^2J5beW7G_>@I54!X!_QnY$kv#sJca
zG(2TbsP{+rw|p6M&jGzLjri?o-z6_t#WVLZxx=L{1UAy}9%%F3+FMS9a+`1m<_jDW
zcMW^(i}8L(T^0l7f5{l?`c8;D|15U#p`%PO)3<oo|1gnRvBR5crtmHfd1gofggZek
zO2#dMIvUlO0yI&XO7~#ukbB;9vRJ!S-P3m7VJ6HNt4-_>*?9{@0kZLEekRqB!@Hc2
zwxCg$lR#j8OE6mB40G^2{zmhd&JN$1qJ1?xq&zLq7V=ln+Ysp^M50^F?sbvfPS`p)
zdEL3VmT!~0GOJW?+nNUReJ$PQ;?=T3y14mJ?S$~n@nF-ooUgpeT?4KxKg<!cvU`@j
z8t3M%`C}FkXPKwkXi<j&(IkN8k_==J$71yHhZWCOKM6V2@lO}I33WaIdJO*Dww-jn
z>{E8L;=A^1wKgf3zW@vY8I#?W!M-hK3iiGM`$pT`Zqmb6S+LFi;Qk-1!M2M@A2zqH
z759f@;B6jw_Xl$bj9}WFf{kFaN4f=&no<zFE08-Ytp5O8GG7yv{`pnF75^wVP*fOq
z;5LZ%@>9&?t2=HTT&M)ya*?l0lBm}r3Gs66i{q<Ua_i@sL*3ioVrT-lMxvbtmDRn~
z`J1yo-f!t!&y!Wu{ZArqh8;^UogxS`F5fLStYi=jPEXQa#Hy9W>#W!W6ki20?kVH}
zf{mr-&GYC-&}iG+!{<-peWkBPc(Cn-(|T_jNPu6UvT4^YTOT{@ZiDtLxVnN}P42j~
z!_#wdb9%j<5nB#QHM9}xqdbH+-mvk>g{yku46%N%FqG4lb7F0Fa`|mX*7~<<+s?nH
zR~IyBotxcVEc8xVRT*uBLg&0+*uAlSo*~(*!3*jz9?fFM!C8=_be#=-(h%Cy$x=tB
zRQUN#i4mabP^MRk@A3EzcjoV)f=GZ@oTLRU1gK0Ysqn0P$OYB+)JoB(4g^$A43P0&
z-kaf#Mz4?D)6CXkRH}T$U0$s}gYKxDjT{3rWW)jxeac><;U<nE_cFUlCQl+QZPTSf
zam+^dLF!h_cJMagzMPrnKHQM5ceZpv0Wrb0gP%Yz++g}qJnkscfbA8$OZ4;VfEwfd
z{Qw?%^0MymU696Ss2oWaLk&^yQN(F%3BM*r@w|M_+{4rzIAd^JN3@+YY5tKMm%ClS
zG-Za+#NRrvuAG+emmqxJYhfj}D7q{D0)uqlMwZ#$1PkKp9s;P3BBQO0li(OXMCs9S
z#%3f5qNO|D(aa388k<t3!Tc+tNtMP&U^VCmQoO)M#Bj^!h}gt27eh|(z-&5Q-C8IT
zZ#8mqCux$fOBjnVXI(7RH2KTEhcG#JbVaT%zi<Xs6*1X#*eH+TfV@4CU#!dcWTk;O
zkQBu7b@cIV(AC)pmGl@p(elApvKRbZaFl&ukDM<+P%XS~^SVw*?{qnLbB7gAC%{_^
zJ3tR@>#>Q}O#mxA!?{J54jRA5!2M`6S$3JCkAcT{BC$~-1hWoEm3%|D_VBGLws!da
z3-b7evuD<8FDmvab-#h6`I9>DI;it$0J_~LcHe@rE({b5jxuqeT}0Si&#%lED|RFG
z3&~hKleCx++L+WY3#*ZRR+Ni4pXfsd)ea(=bwlIp<Kx+EwpD2Pk}@i>M6T8w9bNAe
zCigxKi&!o_ZFjenfK;Wq@3h%^WB5C?<iYrCY%@L^jMs2zoHsTq@wrTqm4m&P?laF)
zaN+myWJ~5}0M+W`YbS2x(!c6l2(0m&QuFY)K5gg9ai|me*SXy%Fe}2Nl~Vlg_ji(l
zoal(f5#PP80AD7wQu~Z5bLsF{KKfKkEW#_-$BN45pDyUFxb+R!UexRzAwAzkq+jzp
zf`e^`UZEK^u??Tc<aTBfQ4VQldesg>r;e*p&!)#;m@eBgqjvrpRgg7j7Htk(lp=UT
zxGlr7G;Omt%?Nvz+Q*@s7V1mxt8b-|O{Lh_ln$)|Bjv)bMU@$#HkiX*I*|~MMJx!Y
zz%ux{8WqI2YKTt4hy%EYF5g_I4~vY|eI42<MG-e^i}4cXGnw0lSX~u~R3T)4#<Ps<
zYfg-RFdlgVXsln2j7jkcywUWMsTn&+UY#VVKy0!Wr3Qj((-jX?-FPBYO7S_u9E=Pk
z^gT+)?cR^#NHMxc$XA|^R`?l#I8WGD-ca8vHt=(Jn5pabx_TaCqfw})w1~$B6)A1H
zkQW=>K^|&M@rs`lw~iLj(?ex!>{+i_<6hM0Mz!L{{xDz;av>F6?NaMT*+o2#hbLT1
zBq!<UawvCllSIu#HEaAT$Y4`wRrDHs0EgL5{&g@3+bNZoup|mKE$U%enRKMvEd#8i
z1x2l+YJwkMm$`e;pfo^gh({T-;?9Mxc+u7@0&AYmVX0*=Ckj^4BgXx1f7n15cqQrT
z&pb^^X3g(+dLHIXmtP9{$5@VIP%pEGnGD4`@Coffb)~3&mdYppuq1<`7hRNC#$CEW
z^%e`QTD%{V4aT%VE-(+!(W&nA{;a~Lm2{9WM8i7AyfTWbcgKlGhWj@r+9`^SX^Y57
zZ#>(SOX!8{)lf2OhP#*Um)_n*T*I*ip=VYEq3LEsWt4fI!$;!c4r1r;hA7sR^*r5G
zAdk=?MF*G;LZtH#$tW>Q)PvL+rrfcrPoc5r%DUqs*^q?)3-|b3Dt5o5KSVk<fvskf
zAvBxy{8g1j2zk`94X>_y`l>g}gqnTXl9)|5TW6B5&D&~r)CVvNt{RR&TUazd0N+>-
zjpt;;G#W}nMX+3`kbzZ{+F`{TG7lxy%j*4om6Ae*L8c^SXVi{Lg)Y695GmY1g3rlT
z>w9~NyXbr!p2qkcPk^!NIw%8zEDFYgGwle1T0&RQj<;=8_Y*EC7b&T?DkVD(TPGEI
zC|QE84e-@rjTMUS4HKWYp-UreQ=-2Z&hKe)zGHNs$Jqf#L-qFkj~d8yB4JpJ(?+9G
z2&cvnT7Wg(I$u%GL>_^U>nr4e6L?g<#a?;Xt{f0ll{?nVMx}<j8+@{#Q(DXPMmv&v
zt}r4d?oh8wW8Jt&#f5pdDAvkb67lj?$b#4;lN&*|K$6m=Bw0UJ&yw#)r-`xib}uzO
zA=4B6`|4VHI#k)AnZWduIu}B;SSjF(<C^}sRx|v$dn@KoITg=`wi$ER0gLutWzHmw
zk0=Rm_IJyI-BAO)AKGTmJ<~^SaQLk^KUr$Fo|GXZ>NR<3#*T*g-D!p!0xeISu_sn3
zB$EU*t(YnaU%W+6soM4=mAezH>=|`zQ>%x_i}Bog&TQDY(Qp_%7UH2b4!JOQCV$2E
zT}A9^UwOImK9lcy{HA7K`;2w34@5J&l^vKmGwN^I6XLUa)XRTfZF;hF4kPI7XD3(Q
z=%X+<QB_8(YHm2!cXQxC=~fWN!Qk2as1Ol@4quNg<ev8QRBB(#spCN!@F9g=+%Z)f
zE3nRf_B$5A@EWeDfA~;%*-Sg$k1!2)CH)eMaOMp|rFkN{*T6(_BNLsoaH`a15^Y)N
zy-EusR67%}HZW(WepsCdHva~FA|iU3!zYDr*a{zPAiaoRcwMxIV7Yu-Na>_&>?aj2
z<}P}$B0augI?wDFa{#j5Row!^S>9)@NCS#|7r%GGM#W1i_s!e(X%HDjBN%0uA&AS@
z{0C)`8fmwiXALx`l(83Mqhwa=w0O!UoT9q&pNx<p_i>?bBC(eiQsw8Iwq5Q%k$Ou<
z#8TAV<mY)eGxxK=0s1Lslc}U%NZt?Rne>J@PCn0;9TF3c5w`<&GV8C+*9Sy0?R?cq
z&a1hw)A@k-SGUmZVFAT&9{XLh3p;7C8=~Bzr0Iy;N;>-evLra`tts`d@FyX{^tY=p
zodiLuM;^dqwjH4=6^8XP$9q$neIXA#C%zL)L~4ZgGk3*>FH#p&(m%yLDds`#*tZ<V
zpGdt{o`jy2>wGD8j1DMu#s2eKJJzFv5wvE}IFmjJxGyw&C8TN8kr{={LNsd+l!QR>
z5?zG#tSou!Nr~%4v1KuLj53m1X0JY+HY`={e?YGYJmRni1o0pf(d*GJnYD8%&6dPQ
z2tL|JiFxmt2`v5{-+fJxkILueqvcmPow|3vxqtwqvhUkWrhc~y)H=7w^j&sHtfvil
zjS5)d8dUm18v@}FuN)3h!5<N7G4#i672-~xj2dY@Kg2xQj2X^)9?8(^0-Vm5=XPb7
z6Y!Tg!EpX^^Tq=_jO$O)cPOLCCgzOWOppQ{>j_!AWUhM$(Uhw6(Lh}BS^<5@;<4|O
zj|-!_)y2-JS)mE_sec#i)w_Ek-J(59>q*Xda%uMm<u=A{PTlWZ-%$tZgle7vN7U_1
zzDkeKCzj-7enjTcFT=9hDj{MYaa$C5Wb%)Duvp=C0@>|dznVsCVsreO&6bMNHO_p$
zJK>wgkhi{Jad#Xa)bpkZ<6GG}oTSY#+WkV{%wh%Amk{G@WzQ`CT~aMWAWz(6<~0NE
zMXq~=IZH{2W3bli7O=0l`yt5Y$9mek0z#h=#GgH~va4;|OCtg}`4Kl<h8arOoUsZI
zphg$oPJFZ_#MbbM&Zz{~{Z<HdQv#Wq_FMazpJ4CQb#V*+CPcoN$T`nP8mMoh!0G1l
zr{`YKy}t_s_M8E0c`_G&X){|ybgS6l>MK8}t4RWkao-=2F3WA3yngHZYdtSXi0fp3
zPuO2Jp+J9e03M(bQ1knQUlYU_0r7a6QLo{Tp{7>uuR`1pk-Z6NyPn&@1DNs-mwxda
zOfh*<%369Xr>P9vLVw>nI%>?FB&$AHMTe@BDuzm*x|JXee|c}gYoC>dFgXExP6nv*
zZ}9tx%?K=JO$@yJ?C<v{*^yq(+B(@X!|sMYK5=Nx3DEpfm>*X-gOE2K@$$svOm#+j
zUj4xzP&>wVeD?}n8Im9DPWVLVcJ9}@Fki=d@XSR}lo4;Aub=?UoP?xb)_CtW8W7A7
zSF`VAbBb;VSlqyKK$*#)<qAhq%)uck^J^wIGobkd3YM%tU6|Hddx`SOu<NUpflkH1
z&hXz|@QUHWYN^djFgHEzjp01NX}>d-6m#8ci?nb!G@#{&d&Fw>lVV2mwu3kBl%Pql
zLmjJD7?bl-9igY=@6RG*srvC;jDFJ7{wR6l^=H^zq?Knxm3!Hf?%;$!Kv;}FLkVV#
zpr|KOZ5XpUel75XcAfI_0)TFd(pzUKqmU<=B8c@RjMQnx8L^0MIMn2A7dz&KQah>U
zt3f_Izuol7nqG&Ie}5_HRbqMwmr9$Q#2~3XN_g+x@YZh3rzT=#-9kpvyIBV(X3j%w
zO3Ecv*iBl{Omp66eHEf<Qe;T6qyfQsMg*k(^nTPkF+zj^p^vWto*$IXQp*d}bbO|z
zq?&50E<<_h*T=Wm973-vGLdU3?_I9LVoMkKV-1NeljSdC(xO<_@lnazBDPl0D$M1N
z&O@>XXp040_uP7b|6(yTTIJ}1Q7t7vg3B|B!R`jQI;8z-n=0-(3W$r|tDaVS76-#x
zlr_7P*FEs_*{^P$W|P?K)qfPd8#Zyou<yop3sx*-$EnKRF}&$4^!|7aeS`@b1o*<{
zL7Mz70*$M=Sju3hgOVwoS9ooS9@=1~814^oW&DdT?Uh9(AuSYfLKXWDs>jPW><X#x
z?T?NmznL?;<{oD)Xdkhpg$$h4aVuqyd@WyeX+Wig3BgfN2NBT1kOfzv&>uqU_ZgbV
z2gUU=Gd|eghFG0zeThisZJnHvHN}(}KKn%}<|1{yrxIy?70?I~kNY#f-8{3aczf9M
zS=74-+fv`&E3`l=W_zxUqIjxb$zJ+K>N&`K4k)8sOSMd@<O7TBJKkS&47OUxS8<(}
zn%tK+J^uumP(EV`)t)YJsjB9Ogf?7)#JM`<hWakOB*01*Jry()W9El~2H5*$&X`18
z1da&<4Wr0FJ~YMm#-l=ygVIaLLA6}uuutg@{C(YOA**>jE+@t|&0g|u4QM3OqixY@
zrfEhyGk60{V9h7qn^zSf<Xq*!Nd1vM{@o<IZwC#QVz+CyJJbI@Acr0lBxeXy_ontW
z2Ah>LhE8(xZDJ<>JQ9ZQ*TCi4-;j7%sMa_x&R8#|g?wHMG%Pbt%kqF)%_vkePm1!+
z$LImcyL_LZvg_9`DOX?DeXs?m9NcNwrE2QiUF9kUa?piFW;atT+t_$NM9}Nu`k=mJ
znBH^zoaQ_uY$<I)90YU#C-9(}8oC|j)tosWa%)K=O0=vPg~79(v?0vXqerY~0DJPY
zHw4(7y3{p`A|P_U9(~?0(c^^L1@%e1eG!AklS~*H|E-KEB@LF)o6=9LqT)oDX{~q;
zs_<3UbpF<0b;SEcZh=Vok$!d4kNmpEhaQ0CHNfq-1>sJ|>0WM5!^5migOfe5TF&=<
zgMi?AbueBeF9=<^1eG6WDGdKNCyWy!sS&ZgiAoq(-{ZHbV^0TiG+qi}5Z5=Fkb|DL
zjWh3Dwu3LT&~J{*A3m%v^$6l@|GY*$NlxQ!SN>`G+;QI<aLZv0A009d>I6pbc$hxF
zAAON+3e#C^6WP-C_UJ?<tk;?JlJt)or8uG2=K(Tf65JR2A366l5p4Lk2GXJ}cey@(
zv$n+BBb;!luQ2%vF~jPI%=80>7+|ws$8reJTHG<WmMYDyKgL}(wm&k+e-LHe5>Gd;
zOGq6vG}BoAiN`R<EUA(8>NYDTqy$G3#s+`R<PQFPl9BJX>oWd8tUSlgCk~N|4E!>-
zmOpvw@^E_xK;40%`)X$i&_~<OV|P0TS1BTJ@T_Op#?&@)-MC3!1e{?4kv)Fi+ytGe
z4fB8!4z{FrIX*Mg9!}!hH@}ykSu-U6Oa6u4r?iHXl#zu&$E;3<J^go8Dn%}Eet(gQ
zs?a<N{<Vy8YeMP3F_Kp)T&Qm1jca8f8Oum7j;Qp;9x_>Jh)Ik{EKkaI&os4f^p2OR
zdS;hjSrx0L-OGH;VcyxJFm!?Cxjt<fxwt4`cvlZxJ2QE|u(y+9Z*PqcKx41OQXN3Q
z=AKU|jHMcqxygUfjDJp4ZuVcmTv4YMNl?e6aKp5t_$|K|g~7<Mfm{>EPe{xymCgoU
zxkIP$$stqrmL4UF#9%LVb6e=fC{#md*pP9d<gj3sIaTe7vU!PeHo4GMQ#dtBTvWbD
zp#;QH%rJ)p3$H7@DdR|AA}YBFFgGZ)fRIelf~Lf-k{&!ZsZDJAy{JevbJd|k$-L&V
zOa)Sh#P+xnZpK=TZvtSVI7K2S!ynTPscJ|~cY2VJhbH4j7(R~=8!<^hL=$kcbBAGJ
zP3`M#w&AT29(8Ejv(o4)1-F#*OSRj+ra#g-qWI$>goT;<PL1;}-1UyjG>pp_gVBIz
z%)+dmsc>8i9o*(lBN&oSuIG5|AE7oUMSdu$P@s7B$Q=zq!Sq=oobA)N0_-p`UlR9f
zt!uhE(H+K_9g5z9vlmWVP3ZrV&p>sT{?uzzl4gTbJ4-RA01FY$JZ**3k&(#0M+q8R
zMz{Pn7w;3{_0@!ZBiv=VdD(%u%yIwx-h>&V<H!<wZlKXKNPQugC`Q&TO1@cKnzI&8
zo)eUj;QTY^Dvs#3EdOS>bNgbIhU5cFoEb<!oP?bbofFgZ?HnnNB8PpcEt@L^6;9w-
zNDB2)nTtA?S|_P?@NQpoZEz|lRkSKO`mymM$m{RuD*Lf!Gqf7N8wD}Dyglfq?@sUw
z<$lRI)DPbs%_-%`kCiRv+pE|4rXB^cEMAkdy3_g6iyo5g)n_=Kc;~rz*_q3c4sX%7
zN-*6$OEBfzh?ZUJ^>EReR2gorZ03;b)8VoG<JHmb(irEa>1*6Je!8+Xvh9%d=gfnw
zN=kK^P@qB{sxL++f*4XXVr_P>eQN1TB6M#t(U*Qp2B~@<{X!a!Rz%28_ABPVL7q8@
z6`6$hK9bhX@wk|_nqKN|fn?l&(X?_6VR>t^ecLPDkA&%o_vNIS5Hnm|tz+}eDKfW(
zyChC(BV%qVG?*amyG+y*(%4t)m!q!1*UH>?`lN`7(bh%JR;QIAn%ipAQg2O+64hlk
z)$G*F+&M>wJIi+y_H)tWC1A~Lo(9(7N0h79E_*u4!<(UthMhDi_H=XY-vRqvJ*Z>L
z4Kj|3e?NI@ZFwpDq<ac^Ng#Vk-7AB0mAi3@lMynjwd1X@?2HA8#mSA0Up^N_g=~T2
zz}ZG#UVZlaE+P$f&Y#jqP|1(A%r3R#0;VS;4{I^V>7UT(B<KvAuLx()Ovg#fVk6(7
z<*67FDSKx^c)G(6g`X5V@x_xN@Tqm05Sogkn%eigiX1P)JL6Dr)P?)80{4}2kPIcT
zzXlPPoKdhzpiIb1Xn!ldNepx{@7Qvy@nf4<^^-fxC1SzR`S^!!w%?pj3fJlbt*PB1
zYy$ghAg3syN;<|%1;OrPAO4Y}upu2`w9~E0dCv$p<}pAqRzs0uWkMkv)ITN4a*YD~
z_DTGJad9LSVxGujN93LCW>j$+Cx}b6=c&%5<{xtTL7K6GuUj$;8P1bJ-^U(DG1K1l
zMr9i3;S9!pbWS_6SEG~fkMy_$dMgZfcBV&<We?+=&Jq-8jI&L18yKaC#I(iA#^p!L
zMd;IKfl^G@uhJcEMalQ9&NsxcLa^{h8m>h)PO`l0&A(AuW=Wck26=9?#ODrE9<~JP
z130pnVw36@C?_*yS+jw3<7kQqmGtReQ)AL4WCzJ5EsA<+$|szV+_~dWPpj<B5?K$D
zs;y=+j>uFY0w#I2rE8dfcK{F2$0(i)sw#G(U8BB>Td@r@!Q>5~q7Mreiyo$mdr{7H
zM9r#oM<e6RomE^XL#^DslX}i*>rndILe`tSip+|@8;RZNQOoMoH<i4gYpJ8IRyJ@K
zq$bs@Cu%17N7&(&ec}abrs;dk`CdjXlR~D=8u|A`eO?jn*s%#*`S(yHi8|Z!A6ccm
zfdk><J1eZt)?ggS{YT6dJ(<PwrpI?5f}hpLqe+CHi*L6+Z~0CKxF|1EEIVi%h-6e!
zG3ml`8b3Mz%%2?I|AX1X^7DsrUZql3A___vGWLzpK@^}QF2**FsJt|d{&(+PrSYn=
zTX$j3v1eK7^@T{w@J~Lr`J?>VLf3?1$CV}3ZS=3fVLFWsCO!3A9*{)~bdq8C@0T$B
zVBCAm3=WVdC@YvpZoVz0O_0yAyD6%g9lKhxzYfjZOeU6fz}DgP?<@_yE+f$4huv^J
zzF2hZDi@%HMLg>Rg^nSK@&f7RG27B#zrIpu>gmy`g6zr8W^8A&X_4X|%n+ANppi6g
zP(DPyF>?;cdE^}sQ)f&Ej1+4y1(c!Qquhhze>Mfd4_Tt+z$cb3gNA=PWpM>J(DIff
z?LZ|U5^+sQa*V+7pu=}-C}T=^1j>@>1pAodMHlipy4ld#yJM$0xUWY=sgJ!k_n{Q#
z+X`^ik+RTwB3#*Xh~9V2{Ah)%lt93;qhY=v>0qR`$7%xt-a*g(;b}qlunjxNDEeWt
zFns7K9hl#W=zvRS7koHq{l!43GFFehvfuI&^tG(3xB4c_(6I7%pgC~_Fl<SN9cRw}
z6*@J4a0)7bAV9qo8hb51ABDsBEc`AG$)O3FY{$pdQGIB_Q<E}WZn2^TeVb{4wbGjE
zA6omxL%=gVO=r=H(#VKMofm9!@HRzDp8J*?z*OaEw0W!dE>G*P)`9L>oY(JY!}hb=
zm&wA4kypzs>n6}AO~%M;bkzs=8CUR{!TCv>eyZJ~U6y>G1V`&qQWs|M-yMXMZ#&@3
zf6hEQC={jP?2Xp?6Z5s(P!QidD?bqp<<iV3yJWl}7D);9&--^D(aohA+{Wy~+gHoN
z{cph^eBsTr*LDacd;f)M-r;;y#T~LAOXWD+A_V`UNe7)rBMCQrBMk}<?tHdr<v9sK
zImJ|4qv&;*I%SV_!nb9d=<GN=OqcDK{E_=tylsAah1uC{VZPJ8p4u#8QvBa^Ya@L$
zn`7B0S8#0A@CKm%O?zL#Ov8>HzuzwP`i-0Ou{3)|(HYCu1|lZp{+;sy0<CB-+)>#-
z7dq*+?vZPkx6Ac^LDERQl&=;lE3OI%K}yU%tm!$zFHtUIk)+VZlUEFUhw%l<w6B9y
zTW-{%)M8pl@lzp}A9xm0iclJUe^cWs`xgrPR9#_{;>ZF(?v(f|2V9^ioXLcjKoNQ}
z7E%4@k-ZvE7|L!6EL9;Y=0p3`=&HIuLPXEBy28FOY!sG`XGd6a!HG8UMGxUALbbip
zqpgj{e?j~|@a;`G#>zTwW#fs*SuN2-rSYsmE_Js2Q1U84^sOzG?85)!G%iTnl@Zhj
z063uW97Cee`p$R|%7dGg$Kj4M6AvZroV$6hgW7p04bfC;r9b{1^O)jicmEx>a_%>$
zY^taR!vCFj6cTe8)ei+NqeSc)OR*FO|M@PsWl1MuK6S?D|CQ{&5!{U^$k?kNdk|j=
zr(gdQkKRazOxC!z>}u%rKRNshlCH;)gVMnDO@*uEAN=r5U9`8Go^~U)ZDjvIof`#x
z#$GMAVS|7FPvG~L&+mz}f>z)>MVGq&;FdeTK;d+5$;+rs{7)h_2a2Pj%nQecdf5D5
zyZ<ogB*QH3M0~OO-{AkZ^8dfK`DYXT`$k516Bz8E&a&(DGX&}HQBl-TD3>!0{x6x3
BPwoH!

literal 0
HcmV?d00001

diff --git a/_images/benchmark-2.png b/_images/benchmark-2.png
new file mode 100644
index 0000000000000000000000000000000000000000..30f537cd0fcdc49b0bd832e0fcd0ba169d37ac58
GIT binary patch
literal 142903
zcmb4rWmH_vwk;kcK^hCe-93;%aHny1hY;Ke?(VL^-CctvxCM82ch|hlx#!$3IrrCN
zbdQ#?>0MpBYOT5EoU4OnrA6N$zC(n7fOsP=CL|950UZPZ0o4T$2fPE;BCr4+AnoKu
z1t7|Q;q3u0oQzb(KTAnL&;ZZjAz&ciLBPHo0^E5ZasT(Z2qYB*^uOMRf`AA#fq?n<
zG19=}%U=|5fBB#Pd4$e{{P)qodzn!GIT|`B6Z${Tp}JoF&5^;T5O_ea7E`l>fWRVu
zxkHM}lbu39@Ii<R@hdt(9;Ut0#pqt>S-BMXg1~keamk>BA;c$#3K^a6;}t}8|J3aR
zi;0OT91;?go$2M1DNjcwocqCKTw^!Q@A7(>^<ytZQ<Q{lKlz&T)oknBT=Lusd6WB%
zg=PEGwMLArm>0}{-7HO^@Hy?R?Ih%F6%<yxV<8Zt{=NmJg1*`kL4y9z4K>rJF9Z7j
zzAxp4NaBh5U#9_HzK5g<dgT8<9|IbNDI_|7<^A6S6XQb%PCy9r|8s(omY%hA*W+lV
z5@m$J58HJz5eMY|J96(uDIzW|uCI(yC8XYXL{U2!U8}3QI3Fy(y*Hbd1fiqbL7)}<
zIlR6R7(^^OsyFZ6_l+d>nUO*UB5ZQNj?K)d6f2cbyuoF+za^QA{%hQpLLC#VBTHB-
zyS7eYFf3AD%t6T_z3AfqHSI*6Xqi4q?L0EtZ~l9_4O1X6yCicqg30)x{<=lA!;j8!
zdb+2#y!&%XC_!RkVo?eO(&DPBZ)<ixMbie_(nB&bGEyrQ$HBnC<*UPEb}5R@{{3k)
z@OV6Ks@AMpO}0j7r>8_%puP(jSq3H=8gK+5F9R_#u@7CFa+c4RydUm=o(*&x31OLQ
zU|^6Z;E(u|uHdn%PX^|6bmM$$(DrJ7N)eI70QF*ma@dUcf2SQ1WK9D}6?}Z$V&q7p
zh@3zRznT@;+I;?KGJ;8ng3bT;x8#E=r{sejyUKu6`Fq41A`mw5D7=rAf)}m7pC78-
z9~=Yk^AF#f&oTuE2aC^r+qwzLI9O?K_X`Xx_%uD1Bf6EBs10t&pyhPC+e)l9nV?XS
zbnoOSB3ed8L5cZujq^Q|SRXHP-0hcC;58Sncsxi62nyown;l&)eO>AA3CHd3?d_dE
zjC*Qq{FuyPmoHr#v+RY<s2%5zgq=S#Ox~ON-h>4IlVk0{_FfF^(+6ryA(io+#lNne
zZXlnbVXlyn(C+xVhwCGW&-9<{ccZ@XxN}aaeKbB<ZneEWoKx&Yl?b*e*KEqJsm>qj
z3$e4|+G-4Hx3{ya2@kX-OoC3}h=UN9(Dt}9N@g|)IAYosZdu5aNW8q;?&y!Dym~;w
zW=NuF?XK&*hc`k!5D<uu)7Qf85&enjs;@7Gg7-1%@Lo5#i%a<jn_jpH0*mMR{ZZMg
z9lTGDlGi6IjALVC{Q(_ZjS=XS(zEksGv)N;;JDGF`X0?nL0D#NH+W5`zgDX%0@v@y
zi}4}_N(*ma5^s4JS?OZMB9f^Rl^M1}kK46Z3)E)QrO{qq5Jf5#>B^jNC<Nx8;6jPx
za+@6vz-vGKw{|i->+w5(t$6+}8|k*0?7{tLty{gZ{pJ^~Vqw(#FQa@f8?9m4rGz8S
z9EtqAFPc=oM44X0H57~9@cwd7zQxI=NG>P9vX?EMRz<pk_TxuY4cK9X&Onqz1ayks
z=f-Ce;gG%v{7*w`h6}vUw}Lt<>I?;avj_|h`;%ic#uzh8-@hJQl=^Z%T~3@I){Lm9
zUh-g=pO&dsWot7|*BJJLG<d11<7t$}dzm9MGBTua9-nSj^7y==?pvyj2Eo9kwAg9A
zo;P#csd2FWfs84H&9@Wx*Oj*>h3aEMkRaA1lQZ3jSJ56roXPNfE-EvXNMx9@agvqP
zO*WgW&f{D2-a?%2Fqz2jFW35hiQ#^K8I7&&#82p(U8CQPxG-4>v+THFF&~bxqhCH(
zW4usj8cx8i{(OJH$jwvgI+#FT6K=5N?P{{o6J92zu6KXC5n0#Gz{8_W*y&Q`5e7T>
zZXGck79H!?&9bYEhF}$i=13H=U`#lI|IyWqPU3Q_D^i?1RRcAwa!;8?!vMC9Tb7nQ
z^K(ul4~^^f;TR(Bv_6JRDKz+SG)qtcIj5|!@^#9U&jPSTENTwdU8<17UPV+g{p1WX
z1CH?B4t7+f+8IhVz-`y-4Cu#@T7bkpu0+m8TC+3#E1*=0LWF2O-CxyhA@pW>@e$J~
z<cA$2AK+miQ?bd;XW~tjswoiqULj)jMe^K>@jM)|dcz`B`H5tjzoW&P*L$PUWSgh#
zlbN*S@o)rnK-N=<IDN4hD>Lx>`G(hIraVpAtF(8R2V)^KljuU(Dr?jIb}exI=*Yq>
zQ0Wa;t?w%o({6+TJdBDMa*2H&SyR0y-;Uzorf}Uf**;yi+8uebdp)EQ;~bH2d-S+^
zU^xx|p{|Y`9dSEXJoY_hV}y^^M7~s6!NPSF4_V8Pci}Ne64X5CWar-FmW$fVlrR}@
zwmXr0HYyX|F*Bz$Ms#%WO&pHX(+4qhkM#eQ7CLw!sc1!3Nle3Ey|;1rh<uP-$2gzK
z^s{a<{r2HuAqDOGs}d{>Au6d##Gh}@6ErNVu9!cGnP05>q72Md8PtJ{9SqpqEB&*f
zyQgSmokB5bKbr}(5Md}>y6zVhY<6SoOdi)xH@nFOh4WQa6LenvemH5DZ*y~==sw1}
zc;`Q1-_b~+jaC&?DJ=l^=3Op*cz%WhQmF3UcwU{)C)-W2GA_X;RxH6wDV&*L_ERYi
z-udvPkJ35tsg5d^Cj9o~?!Tn|isdG#UbY)S9VtP#LM24nmB4auMnEK{)54@t96f2U
zSonN^IEaOjD#LTvRqkVOaJ%xWP%a0C5k8Jxl}Z78u|0f5qdtA$jfs5z>(j|DJqCLY
z-5%UkWL>)nt45=X<#LM}5JvK~46|0}(M!=1_(^41WYReE3^=4TN|eiB$T{lG=k#kz
zQw%;MVXIQLRkYm~Tl#dPn0a;3An<47*Hqw6@&;kMACxs**3BspF(YFb>FDU#g$y>-
z5;Z@Pu0s6<VTd>&sYI1w`<Gn}<OD&9^%e`J!Uzb)tM3pcGK1$HfU5vgWhu4|<&TV*
z&T-TBa6A#Rk@j`DD3Y`Xt^6CsLcYTr_lN7LvtTM6$WTJi@+^8OERJBAKK`qto?(z>
zWr<1!))0#S;3ccZ!Q0Qn`ES(3Y|_Nspc_<OeQcOiZjP56hWWd&^$DmFM3&uew91G&
zEv8ND-uG5@*9jy;^LAO2@JFhllqSpet@VV=2i(4nI)pom05*GtGQ@YsBN|{%V(u<>
zkm#aB+U~Y-osXM#h88rdu4FyhE4D)7KQ;ylhhptPXc;18|Ahd(Sm8Tj{F55upgu@7
ziu?q{$3o3WG|G@1qd=*MG)9wTy`F~Zeied@K}%NaeGXT$*s$VJLhl_E$ZBI9q|?O{
zhyBf$=Ph)<0u!>P@W?lgoY;a=jRt9r8778Bf-FB`XsWkFcf&HgPh_FkjMK-Nz=f~r
z7RxvEoe!1;MYm%3u%J}ZrF{I<MM#1SGE@8RSk^IxUY;a-nHjcrA5D#UTrQ$()YV-1
zxfJ$5RMKE1o%>4jV{g}evR!7^4}n0N=u5Z@K{+^9(j<eSBxZGQ)&thwAuJpfxdv1O
zH{B+`ajn`fa@-6NdSxFYVBT9#32Ty6j{?&iY8{yN>1%mJ!{Jm(qGGwG8cI0NsMVnw
z_VD6do&Qgy1l4x5ZoRNnQ~M!Di(j^9Sn8gU1l;zzrJ;#G_x{}3=LnEgvO(B2h?cR@
z{-A!HQfAL-p2Y?kh^sexb3#x!B`lxY8xGX76n}l=P@pWn1CpYxJ#o;UW}|f$ooXeK
z=nh-#8mkYK1&A<9i~FPX_j1nE^u@mr41i|%u;7WZkV1OC|76+Tt=9AR!$bv>3}|oG
zcEkjM$eN@IiMH!y{0{!aWklX?d;pqjQj2PNrEH*JN>VVBk_wtotw<9$Eq%|@gfc<r
zR)7}d^8D!Bn!$pkKoRJ;`#$Mdu5lvBCa@zGb%Q_bYK<D%|1ShAw)zfIg*vb)*|VN1
z2mncwKT1yL*oDd7z+m+I;mq8F=Xx)PX*DpYh3hF`(n)qP`74>!j2qr=q5K$tw`TM4
z(#ri5(9Xr8z>i84Mv{r~SF!-mxmc*z1g21K*c(K|7}w(_QCe#F+od|QM2qLBC{(Q-
zXbPnZf%W7ThCUd7>@M@@vLc;!56*>{A~au&#y$-Hn?fnyAyqN-*V-aQn(ZE~%D1br
zuLpJ1&j%UXcbjyn-+NE!hTFH#dfe|7aBeYdcchHIvb#B1;VmPclm08bky)doTNnE@
z0zfn^_*FNN-jdG&sV4o3+l3Uaext6(&C=IugI?sPpBz^0PxVJm9&`Nax~d@0$1~9v
z%u+7sJNZ)#pUYzYeKL=GGo4zsP{GC1K}GxKjE-k~0=>rmpr}@SWhGM?)6et3_dl*T
z%fBCwe~Xr%ftP>KZnhWy-YlKWB3y7Utga7l+R`zMk(?sS7$ETX82O--p`?{ibzJty
zE%I2{3^S96+)Jp15b|Ex{(ibSsBI8)*C>$2*HY<jE_|#+>2~_*hrnX9E~F(0H{+*R
z-824*b`4#ID=XASfFFib6^~e!<`ps?RpAws_h30UvjI*h`F$qVYt!Bic?6?T7fS%h
z!YcQkwEx!DU{Jtfs{8P2;m;^)O9|;`lI+U8+2v&ex6nd%Jso;lHkin`lQ-C>V&+in
z6GRdj$$g#F;RC;Q3Z%H`5KP+r_4v$s+b!^$lV#VFsSwFv5c%fKis#4^Uxf&GvBe1V
zaHKEZ>?9n6ciS!4aUh7oO$KD^7?JRIdr)cAj2OxyR9`6)P<!RVFl>CO-reu!MCNmm
zFWscFTf`WrJBwgL`3tzL{e-=HycoH17ReeKuKP0H%IElJB{&LmeY{k1p}YD9LpUy>
zqgBTewbc0W_N>{-tux?_0zE<x+FSNh0sJ-~V$9z^o_3;)*O{p-xE^oPK^9O4i9BSm
z8b{Qpyuo1+LiL<2Rbzc3Nc5lm2tVHcvpIlJpG*;l^>diz*G<Bbx}oT2wIlR)k3gYz
zE;*30Rb1|y1@i4WYEb)Sg?*s#aAJ;V<o<zZ7||k+0*x>AD{m;{QWxN~2|ykd`r6;e
z_UhHE!9YTf!gz^`(5W_eSMFaQR@`#T%J_l<6^rDBt?p3%0P%d+sA639nxGqk{)8WT
zhLxXxH5?lc`r*8_2Vi7#JrUPz>D=`&>L?_;`E2IJ=a%yU+WC@Er-`-ZFiX9AyHq=v
zZ}~(dpQvF74};Mu?H7qbr5#{+e^{u&H2^>o14%961^!K!Y2?{#^sqSWE4nW(^8#OC
zIs`M0tYru#qP6Xpr9_BKl%Ftdl!Hi2+_~dGrwWLhn`~|5iloY>elPR2K$dSE2AJ{s
zhr#(~|9siy=~|~cvmkHd6p16Mk7N!|c}+f4vH%%b`L`{fw=YZJVXkh%GFyY?l6vna
z{Y*qmn!Yd`Rv4MMpPij-R<*<kzwAc13$#%7XDTY<Rk)dd{}~jmIlcYLT*>a|zLH=2
z;($JxBnZ86)mry>TG+eDWm^T#vN>m*F?ADj%J5RbmC?rF{4DgXB=z%1b|VJLow1uz
zcwd$*y^jb4o<JQkV?f-LO?sTsxFWp#X7ONN;hMnn(dGxY2~aINfP{w(@Ud^6R7vT=
z;SEAVzk8g7x5XWjuI`m)No9$umSn=?00lw#^IE4XnVvVu5ECzuBtBu${bp$%6ujY-
zQfal)R@=<ku0`!Do>&otNHS^A8<P4BRfig#kT`gUG1zs_ELeM08d0GUjmjML(RM9f
zp1|#te+dR#enSLNMUj+ZqsSR0_7iC4_U<m3VpzXB1oQq00ESj(y}!I5p)UvLZzp7U
z)mCgX{)qBKsqkb3+|)KYjFTd8XKjvzd->a7)oO4|xk%908|1n77gu_69!C+Mj-v`P
zJk)VtNA;nCb@=$aq%k-EAelDjvDj!c-)OTxNe#RXx;}S3s+*}bqk4@T^>x`<l-2~O
z<`Q1ueh3zQF@0i$Q<IutQF+DjJi#>d78(c3Dr_Ja9yKiVL1**Oi`b+q2u;3Nkvh%q
zY(Dw%!zYA~Qv6*c2h(M=A|}kCJt$ZnM8vncu2=hcgN#WVeflBSwYkn0+dhfW-{^wZ
z@}WG1sXg~8XCxUB_8UIE?-wn`pf@xdJjPdjr&6Y#=c!ih$`B#*M_7$wgR;?~Ys|;X
z$Ls0T0(~etx8leoQZIpNx!*6T*8Ks+;qlm5KdYxE`!f?yAWE%Wz0Q==m*+L8PiO7q
zq@r~U3L4rW&1y;R>Eg#L!d=%pdX4&BF5G->Rt>|)KvCj2F+XJ-U-apK7Q7xLG;v0o
z*Gi~St1&UHFyD%avfW_8oc1TnTB!wc<B*7%!Fj&HI5Lok7*v@{l+lxLHo;acXCl|y
zMxRAh5@c`GmB?%s+OvnUXGG+9JH+!s`>QHd5#$nt8LvXqmwa?UFsJf=1!xXP%z!w1
zW%7_S1ra#c>=;4{5usq$4u=sY*QM(DfCsCtYvLW4(tt|jbRKtg+MvpLHB5TD43jHm
zLCe=vCbFn&#fYJXcjsI2M_r;(M(<lusYIvSoi9Ewy-B`dW)Bkp*8POC@Z0n-U1?Q%
zsUGEtUl!M>KpurB)Hhu5%V0_uB6>!pFfq7F<#w%Cw*NewlD7nou7a0gK*dd@g$<b_
zT_UrLs31H1L}G~m)}uos(?)TePL>p86cHCg<2^eY@*rncIbX~E^J~SRpUOAPoKi|j
zOg`m3cv=>iIP{x-0_PF01WvW0hh$&1=GYM>!j%RDLB3Bu)zh-ij|2FG-VY4gEvmFZ
zOc9xeflAg=BGGmAb}M%;mqFjTPti|B9SYq>8@=Myr{F-B@pT7cftbZOmfMWMqWigc
z^u+iW)jk!smhxV$?Ye^vmp>ulHzB?yqUNb`2Ju%Bv-qR*g_Q$VYnK|QhIYy7^U=}K
zoPn?``-EIhObTn@!a8{GZd6P}CM>$pJ%l*9mQn=MC?m`!Fsds&`9z#<r|ae*Q@?E{
z(NBCVFbQ}^???<ntw=uaGJng4618tRLz<F~sp0in{?MW2F^x(1J%0lt0cu(ch`t-j
zJ=Y7xMt#oukIii<0=4XZl8}_h1I~k|5EcoYiAt9igz`sHJ0RXBz+8B#lhqf6Lomkr
zbx?JJB=0HEsp>_7yPA~Akxz)fbgK4|`;^|lt<nAA!->Qjf!mGta1C>uBbD;L3g3qX
zUPXn<w{S|U=Cf3pwSb=A0hPe7umv+o$uMO&jXKl34H7xxH2?rjEeF>_HpN=NEKa`4
z(A?;+FZ#yHw5a5Ixl4RBUuXBr3poNxte1f*<Lte47RSd?4a9rxLB*}r%0rPqcPbKm
z2WNo#lJXCW^W&xObu0#@TGJaJ=4F~e3^o^u;_DuM)J1-QMiL=2+!o!f`hDTjcKgh#
ztC_C@{0!Zt4LSTnsp(vPfl5&l`a_BpHPwA)#p9xLTus#dSXPjQ@ll{quZ@M&td_o<
zv%Ifv^#8thcdGj1#1qYvCRnYjf8w7@P9%eFO^ZRb9O-Dc$A-b^QAU=^iltYv4^isf
zLt~N#jX)i(O2KVo=nfReVpeYnzOn#hxikzS`I)U@X%{(}xQ&ri?q6jYp9lG9yXS93
z@?+~V9gn*_2I~x+icjr2*qB}Bw|=Dh>zlQ<;#6u(Pn48%Ox%HFfAFzH-sY5*)@wh_
zGknoExge*F`zW^Ms?KaF=qOS}b}Ij3uK*Hu)eTA!HEm>)T|ZYI6sZ#y#}yCTSMPHC
z*Ki;Z7R>=3|MLYUL1I#FO-6q7<NH+OozwsGVIP4n7Ev=`FB8`T--=NbJ7egAnlARo
zMf(@+J16OdsLu|~-bqO%u^v1a4GQIa`*$hON8-J^`voAvA_-vq4*)tL!4mR;PI~Wj
zudpI9Dk^H<A5vDw7RUm~k=+p-62QkHp9iGw1xv`)l5G>&NelmT%DeauZ|GOKt~bZ`
z=L|Y+bO7NW1+W!cA%oq`T-@BTtw6vgXJ=Q&Hny6r%u>d%>(*}yGK5jxCc%mL=4_|g
zl$t=bOZ&V`z`;sv{3n*^vw;%VQz6sqd?z=x+1;uwXv3&Q(L1wRjKKT*(Jq<GxjKCr
zP!n9P52?2X;?b?FtSpr5r<y-n-NSt0m&}l>zx?Px4*r?FQDbC|4E|@pWUJ~>HlZF-
zNh;4W>xp!6w=38Zho36M4lk)W>I;F~CJt~?d2V;-F5m6<<QEocRm$ULv|XjrowsqQ
zHR?<4j~CVUDs9fsdI=?f6X$t6-B)}5_R#nZU^J5hGK>~>i}}9mqXmoJ@p%NG&H^f{
zCq#JoC}7kIkyc0T=5s72lZDYQaFRm$+ZXOGcWh9m&o0UBv8AOg50?moa8L6W*l!7c
zLWWQ9WREq%AY}&hvvbgyEP;X3-7R>%{S&$4JX?vn<B%CXI&-k+(^VoKw~Hc$WKw24
z--d?^fDoJ?jvBTA1}JWHRQBxVCK14rgL@MNTdO~OS=?@|%g9pBP-Pm82C+CC50k>)
zv4QueN+{9L(6+6c7hcdr<j&4cVkzVHT#Iv!`D`V>NCaNdj@Y!fm8O6fVzxx3PO9Qw
zkSzy`^e?~eMr+J(DSwQ*BN=AB?e6EZ_5L>FS%>yR@3+N!fawW|jEt1lgodpIiVXE1
zo$tj!&=PtD(rK5aCv6Yqr8@1}8kl3F1u}%eVPUo1tZk=1k>=?!!1~tKl>}dVBk=Dv
zMW64sQ>`|~L<yy)ro!X^1|k{Y8eZeF(<u?M1ADTEXq)RokyWItXK;@xItJEV9^U}^
zPu#vt9F@xw3XZETzoTeKm@6k@R|j=g1i9)GYbZD#zJANZH(jbmy436t3ozhc{856<
ziGbX!ZrTNnB~5;OJYyj`8@NDyziiWsZ@V>sxi?#tQ_l_n`IzC4B$P7g7-SR9=lzrl
z4mUpjSG*?UxxsJ!$7*d3-T|RNtCM@KbRdc<8Z6Fb!yyjFzQ$}R_%j%6azk0t5KF|Q
zy3XI&?I*c<cbK<U))<cr0D(P?PJ(Jtl9e<_ShGss--3=R%%C^I!Ko0~fCB)rd~_X{
z9V}T81<>+lCHYGc=0kE&Glb}yyS3lw$RVlxtIiN?4+Y{1V}k#T2HsrYj+`relk&wf
z#h@U0P>n(qHHPi>5RS#3pwBz1@MJWDEkQa`x)5Li*JM6tlY$5c7{r7OMp`j;0PW#v
z-hqvRLd2NE{iO8<4=2IBI+^eK+5Ypz`o+d+uz0ZTp}rrPti(_xA!9}Ge^$pT1<|U5
z4vNi%;&JT2{O5SGJYDBE?<_3_#(7Gu`mhkaCBB@kcVE>5arOlhQma+xW_H2Qd1*D<
z%W>G1H`!;uu`Cd=X86Utp07_ntcppiJc%O$=pICUA|bgfePO0OEIYFc$BRu0y>vg_
zCu!ai9*D%G_`o9Vvyq4oYkgfZI_h!9MzdD<yuQiX#GoFOGx1YsDMu9Lh4s!U_?SmX
zg#R72sMM9icTg)CLjv<0-AZQvRQe8r1OA_twCKf`CE#!WEv*j8GIs4=wdV7%;hFO^
zkIB)-q93p4RI)@YND;paQK|bP0(l)vm8nc&RvL-}3~`p#C##V^banp??EKZ>wSI)h
z>j)V*Xdqb2f#j3G*&CCyvHzJXFBscbkgEM$3l11RXfp<~1OqpLO<Gird<81Q>Q}^{
zjVeKq<sX_N9AP7`G9=bHyJ$|(qDf>-93=fCZ8eI+52dgTT5iiVhSq->Tc^m1`ppoF
zs84V!DWemdkO{>jKblCWLU7yIDD)3&)JP5gQnh~Qc33vY77r&4<h5zUHJ@wL7Kp@;
zc5UV%;vITYwltgJ4@soblOQy=65pxl*|qQYPF4iTecuM6jL%&-QHK-2E84DiA>hxL
z%s2g`DgSGDjT{I}&|qZGV@sN{v&CecemfbSa$kWATxLbblf`_@4~9_!S@P!Lx;hV)
zCD%yNHXyWAU1Xy#p%|Uv@7Q7j-r&w_U+jz|QY++({eFI8eUJ6&zhwAL8K@by=tpPw
zrqkjW#XZ~xLog8rpN<ai-R@G=_+`QX%lVnN?Rl3?lkHtK*K|X{P&$X;vE_m$Me9wy
z8;?Q0Vu`jOt>gB3Ge883@pHLce$mS4snuxM^(E4uOn%8kJbvbteU<)qfp2grt`24r
zj}{s<WHxyISt&0&yhBO|h?%~G?0BrLT;<=5Ih|5Em-0P1#d+{PC=(EzH<oHQX~Nx{
z<i$6vBr?QM6NaHPDxF^XeqE0GAj(^C?QXyH_BC?J>Oeg0R#8EEOqD_JJ}28Xg})J~
z&MX3tBYWiOsA0u?E-yPfd-JD1lI`9&xygKOQR~FqpSTWy2-fI;{a|-B@Nx1RJ&Vhl
zbg)XDK(!#J{XR*~%3I1E>5PP+Q|qR8SQN(uGm8?GZukCS_`Gs54}<$s=i|61yH9p{
z(dT;w3PzKKa@95)A{OynWxv{^!w7k`!SOWlr4hO9?d?NppH!lP5wS$N$p5O6&KbO|
z9e9HuzAw_8ot<%_h?tw37n7%RI)I<=b_kE=u&}Wc4?iBOP|pFlBOBVS{qZ4)ZqWt&
zylAI<@7Bf-lFc0rhQb~R-dKrl=xrg(8;px@e)En(LNzwi!y5!D?<VM>GSd+dXi0wp
zE4J9T{`NSsM8uXSB+M`r|8huwRx`pA+e_d|mmFHcuvy#_85TBK;N<K)_bJ}uPr&8u
z*OAO>yctZH-WOledQ`svbCXIn+U9sPZ)k=GFivp{TIwkutjf6;tUX;HZ*4R(03<MX
z-4x3WKsnBiI<4XLnA~UIZwR3Y46(|C)Jk2;OEGAZ-g@JQP_SnMTM`4XwO-FgbJZrk
z%Bzd^%j555CI)oBx$Hybf!Z|zIoaf^vS?Hx+)JKrW&7RHUF&c#&7#`@aTE~ZB!c+I
z+Txpr?>f;Y9?RgNk&7jWX>9-e*`o4wx>g<-lf(1b4Q>Ryj4$GMi29Zeb=l8Vx25z&
ztle+HVbUb>>;VW0`ZP|5YR5GXcEB^T3$bb|{A9F}nwDo8%xBap5mVR*v&ZaEMmJ{z
zc4JlTAZY(Wzp%*g1c$5)&P=^@H;l5qqa)4H_R!*7acwtWX%^OYdBQhYs!x0N$2pyq
zrIzCq^6?23|IIH*mdOW6jW4fEaClRxOtO-ET7`i?L%}Rgz;d2!oTpl)|Dh$DSTKO>
zV4})El%H9FD#&7?J|2+hDQXy&t(ge^gyCXRM9lozyo5AWIy}U|q7CBXMckl<UGCs_
zUmM7TwZ)fs(dmO}4iWSOWot|2s}04QN<sJpfNjHcvQY0+t=*1vk#bo|{k&44EdTNM
z6CU?NK#QGfqLUPj6ZW-5=!A}I03_;xY~j#cm#ckwzNJ6`wa&$;R@WQEf&45_K!48z
z6sgZixR%VyW7)z-*FW(3QX$VSFEL8l@>2JJ?U$!qrY>Kp%b#FcBf)GIZVA@wPZxz&
z(@;q<M#1M){7qDAHp4K?GiL)4#Zav7H{x*<NvBYrpAK#cRSsDvrKjzSTA-m#l<hTU
z%cEUo57}hCCGYHA&GJAveGPkbd`vJOKZ)17`w^x0bkH;+XGw#+(wp$Zv@CI-41x0|
z8pqdTJY<s}x7PglTuzlH>wn_O4uMSBECHY9G36gmb3kpE2M`4~E?xu;`T<=}?ERc^
z7As|jY-hO43)eifpGZZ*P&Z?ocGAr5sJDkxDGA*#qaOj)`vvuyjAnY*|MV|f#A<Bo
z4#k$(`IUAR2RKYB+R6b6Qw}f><lD)uLfFbpmT9QeoL*c^+@x8|*N%y!+2HlS4?%yB
zqbiVnZPSj3^ZrZJz|e`=U{VFH)hWucg-WR^@!J)e#>gkYSyF38t&bdS*6Gx!=cbux
zBUJ`iBHBj$&1lFCV5HNzimp%_@88L}1f&)?cUs)Tl=qPJ9cY<JY*<ikJ*`}u9;`Ls
z3K(syg`TVZrr5`(dSV{!<Z}4Y7EC|+0PbhiUw<8ouQEUk<5@an)6xKUL%@km9Lt+V
z0m<|D<;S0@0EZNjvKrD-i~Esus=aL_vZn26{n+u_d^9i`os9^;+vS+(7C^6m2(-{J
zYzeFz-Y-VTn~}S_7~!?h%5XauzXPVN&gU%#My{PIpr;AqQf5+lTnv9erVXm+S2*l=
z>oY63$SR1iRBw?g#V;h#4U2>|`zwh_&qvv%h-f~_&^eGHA4ez%hD~cfK7ksAZPIu>
z(e?Kn{NB}^dJ`y~6G;IsaK!d2RgrP0!KSU@d6<kEK@xNn9|NfuWfI4cxGrjkek7ye
z0XD{d(E4DhW|QrFnK2+5-WQp<RQ>oYY56}@H{dAC-G#T2Xug(RImRTXsi>$tNY{3J
z_&5unhm!CED4B(k@jepuX(G-H#`!p^zRPmTqrAR<i$wfO*cNBwJotp)jS%3{4^B3L
zr6mqHcU%;5UGX;f(b7KJWY@cp$OKI7p;^|7ti~V;gI0zyc7TMGgb)!t;G<Gv&!5az
zm<)q`I!5|U0XzqtQ5u^@SsoN<%j;0585R~Vp$O7?Vz&06MD0jSm+Lx(4QzxaQ+BzU
zQgIN`mh@IdYRr7RoRFE1hqG95ze&TBbKoVmu2XAsYg!J}7<$1^fKZeGIM8<4CtmSi
zIGvk5fvO#DtqP1s((eb+90bF!2pq(sBXF;T&kQdBN8FfB*GE`O7N_cF%cbTx!87iE
z#a=!hveoU9^&PH?CBpeuE)+%^MyHK$XpSQ_6Y$+fO*`omE37Y&d9k8;K~N7&9i-&<
zP&ySleCdt52W*vLjU|-wjEKI;A_b%Ep=6(v9w1(b>*{8b7LXsrE^AiG=PN#!uB`M~
zBa-8sg#a@|)Gui()Tbw=$9w1T#cuC+Hyno}vHXk}9*bQ~*pae^WknWC#%Cf1=GZ~8
zi}!_v0Ed=TNSDqJ+cI(33r!ZafKaMl2VVAguz1>n6|T@|uv90;0}On?2t;3aZG5~)
z$^_ZN%uIgo7uD7c@Op4Q%)MAaj<S86!U(usb|plHU@#;MZ-kk+)(KCAKqf!VE**fH
z%yCRIgGY-(Dkb+J1zGI!()0I?Cnwy-;atu9RJU95kL)k&>Y24Ecf(EpmR>h`p|sOP
z!HX4zdTBquWVRJ74$j!^!;%J|hL8|guR|6pg$XzTwX50I5g_T#T-E47`UwP&8&}?k
z<9||)AmFBv#UcdMK>H2)Vc$s&%O>HdO}KvKv!|H`s})=xLpC!(O-h1$eZ0m<{j}3A
zZ()p;lb~2X{UEx0AcSL;uBfs{yKZC!5{xrP%Q92}a^6a6R+9r1WG?~AN4@MvuCAF#
z_IPOrFt}yFdjDvT!N6mg-il$Fp2w6|L-<Lad@cVLF4j-Dw_jW-##aE!IUZtMOR(N@
zi3T7KVnO8K*>sF`lQn?vt37W3+;jLt%Roh&x;4rb--TT;#l0OX!!xGi=QQ5r@#J#E
zzzL-3yXyt3x=^b0T=fmlvO{a+Y2A20eA%V50{n!{I3Xo>@&T|IAyh2A{K1kkmnU{^
zVG@K)c?}{2l>dXHnIS9KpSoM)0P=ZGvLf^zNhh=L7Ww(nB{ue2+w;kR-<^*u>b=hi
zEtQH0V9~3Ke}hO&P@f{11vpqomHGTEi27f*A)PTzxLK|9mnxm2*2(3FGz0dYIH52O
zqXCSxRX9Wq7oXBXv3%b)F?MJbc``|FJ~=!#)3|yufi*Xv{IkbWVx=g}J)P&57omy(
ziq`4$Pum|G)F?aX@{w7(q;O09tQqSlWnW`ka+k+-dUmfY2&$pS=4vMku-vt)3cKzY
z*3(n7i>z-Kt8?}u>)`(6<>Q)PrY3lkO5z*7r034Xoolc6@MGLjMwb*}69gqYnHfd=
z43G*uqn5cqaX9v4U$$$$-hx(^<m^2LBIzQ}nLV1!yEA{^w*mQ5sfa9%#H_gUzi&5C
zOaRwdJ=9OOSBL~{xl!r331VEr!Xh6ow!Wai_5ipqqF(VCFoAKx<p9PGP7NlEIX1It
z_~nh@;^0J~o{x<`-*&TYH6l<x$2StLMCXR~PDu$xIn^CgriVA@9`@Y)50aeY>u_>a
z4Gi$7WmS_Lev9xhBk{9sisY<nUArpUX|h>xk!!cZ_}2PB9coBFW%$ux1AS}Rg;>7)
zi`V1I&2O<7m-Y+8q{R<gR>|`X(el3@#}ye^>)(F});Z7VqjrEU^6{(d>IVO;CSTVs
z%Ws6uz5XPck#q~gVM7fDQ<INrEzQb{5mlgi#awrx%fP2ctea5936kEfv$G|xN{4J%
zhkK_|!PubHnaxb^Rms`)rxR7ZW+~@y62c>m>kmU*$B`4n2sVid!SykOKvjLe<`eST
zTGdQ_KySS}7IdNaRjHvzCWA6Mk=*f`>gto05SM9f^K_5rzq1GiWMuI$Ohu~HYOrt2
zJL(OnvoyddcZ190)jFrT)pIcXnXqrssei74LiLSVAVy3^U_*>RoA&$x7afhX@yDgF
z6m)smF8Nq7enTpOHw8Q@cb1g+N?6^vFqpj*WZ>QIUEJo}e1ar;vQ-x-F|9Y<#p1CP
z_1%0`f)s3%rau$k=rcvp!7<PR;daKxXQ9DrrY>4CK+djOZd_IZA)y!WxgIbwcM*|%
zMzx>_82uJ&8_LLGpy%gGqfcI$0N}g5%?~}@7&}>SnGxSCNN`*Bgs(K;HrsxA7hnCh
zVO`=PLgU}HavwU(>JvH+PV&A#9wdbavy5h+RI}K8W6hfHjr?)`1%VaA$nXk-Muhe_
z*7x4FWjfyl#kl8TTJSgdNfz4qYvYwr#TaX*aD&fnUAaO>G#d3m8O>|?<YU%r>^JCq
z=QugDdUb#isha*>vN+lyvj4E^aR3S-1N1B8$vn!@F*joVJ3!P8A|Ze_Soi<PU;Bd_
za{zyxM)ytaFEyYJ&?uvht6)MT=?RpenTF`bWE?VH)LyF(>G+K+KECJfh(f8wVbPEl
z{_g@PN&|tYR=+QbxXHm;Q-u>KX!4<{Y64ri7L9CA;8@NgpIFS6nA2K5k7vBvDhx4j
z+e%l5GSki%x-0vxR7@iN;8-$2kC>j?++QXuyaZdo%dk!rsTmeG?(fz;Gf1j1xGGK-
zH34(jRHQlsaWfvLFGF1uVXzb^Zwvz_)>O*URfolrAf2`wgWCel&4jJC?iqj}lMJrw
z2^>yh{IoptIXC5Mq>9<u|6$Jf9MLjEO&p_@HhPLm^GWBGVzD`AMr(*l_Qul+@=h+V
zK$wY{mQ<|FP4dSr$4VymzKU4FbrtgTG!5^qY^+w~7N0|%Z4DJA!;`xDRIxA}{IDMu
z-C$(lpv#M=Ria6RtKayEhEK0!X=)p_D742qjpnedG~Ho?DwLn9Eu@hp|6!P^5k`HI
zVuE5<I@2!GoAdJI?nuU6=Ut90U2fg{?2~KN+i=G<L*NXvMY5_bop@OkmjsJb#i)Y!
zKaUS<sPR^2KzfJn?Nz=-KopA+@R*?|#d+(a1#GErLLI|{qHdyYCHlc%5!ca{XxWtK
zj^{&CBWW5X|Dn#k@kscP^pU+quqHCnzfkAJX;6~!lh^KXIFywxx+-(Apd%H=34Mrn
z7?WyhG1+2@oq?2C?+#LJ%G7@01!q9k7m8{Dv>&ieUTI^F1bkKCus`bkG52phW;LIT
z50*KQ$MNkAA`R`*w;#iKYI#J#Dha|=!Ho2a)GOt@H0jB5ptQzo+GrcwW14s&N|G(v
ze=6&!68IOt&gY?|=<O8Q%xA!mU_u~s0#O}Tu3AMhg~aAf+_%mB^?30zV9f>_C`pq#
z2_ZmTEh{<YmtQoMDe;6^QX^IH;+T0)1_}cU6;vXe`bx%ZY|p{93cXvp|D4EL0o^N}
zuRDGRZm!v(1i-p1!+DvXPk?%IoGX3Vd!j%_qC&gXyaE6=eE@O7uuyL?QE)t4B{*HK
zmAcHtb*%rw*PcRlm29%f)K$=IG|$Mv9-=#FuAh7#0k4eA&`M+@VHqZ~h>(vUv=*99
zs?Ik`{Id$kqHLgcJaK4yDHbVuc}yuOcV5SK&Gv<%tgqte>FG7M#Oy`rU*&hjn$%d1
zq<u1WFS(IhE&>c6Me~5sFSdACTM+OQz`75c0jh7a2C(_%9e4qp0Id2=WGDt@QQKEA
zDy2{;vFggm32M-(!HIz^D@DI$G$Udn2+I--ieZHF*FkCa#Ah^j!!~trjU9i{Ff1ZO
z%nV?G0{fU#`W10~Nk12ViNm!zuw09lzI2XavK<VfDqv%&lm9xS^|ilLt;TRTh0SMN
zhTHWz4q%`pVkx9zq#SogiHp?hieD3OsV06&VKE9hauO2@da)?Bw*z!OOoi!Gv3R@3
zll@AD$9@0wOmrH6zXyPND5KcY8X(vmsHx-pQh;<zDp)0h4ARA#!1GJ9Eq2QDQlh~k
zlE^S6g*G;i6QucWA|%<ub-MTNuF{y>M5EP@CO(TLzL~F*ovKu<d}IGs3|o_@;2(m(
zTN1so@z7lyo|o40I6=cgsYV^NzP0#sCvRvpj9T4<9tLq5sAabR?E69vEWHul?~Np^
zz~p<+=k2zgY`pmbDS815gHa9L&+)*Ebf<L$_ty?y;ktwjMWPXSV^Xn{G6{e;Na48k
zMtfo_K%)zvgun(Mt^4okRV(@Q_4O6w*{zza*L+}Jy^;V7sNL#jALkI<VmDyQe`9Rk
zPJSCj^{TV}6{V4<6tBC4v3f+Evwad3?0M)=R6<D?R7;j36a1g-0=TKH!}w<JD<%%#
zeqF2N+~#5|12AuQJ|0Ek{jBv89+R$@#WSD+m<kWSw?CL3J_G|suUzd`m+CKLBpRs>
zJJF&9RjsWL$KSVh$8zd#_KrGORP!Zq&Mqz_rt7|*EUR+}8OX{;#nEdh16J?`L-*UB
zIw0ydC$w8!?=}iKYjl(G<mjX8u?d~0G~vqhom$lwz7^JJz+3r|Q6al!`7Uo-^q&X>
z32J19#5dmEj&`JUu)`#@VLsZx`RMbu6QD&3I6Heqr+S}d0B}oIRW-vp_3-eJD(HfX
zib}M$w$?vCgsXZNrdk%7e<ZR1w5-T#yB@I}rI;QCUa7(+2paO{Z4v<c(r&~!t0u|O
zLRF+^Nk@gUod5Hf7G-DdZ){xqG65BOOz?UUtsKpuNlQsWhx5Nf7tlvUB!g;wbB%rD
zMPgDuHQcMi1?>X_gE8o7q4(At{<iQlvZPe>azMhoU{}IY=%Ra(c<MKP&CUTPNk=LF
zE0(;D?n#=UU~_=?s0)5zpa9x!x(WFP_SUB7s!c10@MaYeZ>icH^2i9jqf(=I)GJ5m
zgec1B$)`uNp0skc+NX6{W0H#hIaOk*d{qoqRE+YUmQ{)u{XO%PO)`>#NrP#cGSO7;
z0$$^g0Khg{DwTaNmrck0Dn1PNgY;K#mf*vD0HDNsKHrI2em`R01KI{0*@Ry;7QXw@
z=3xE}m+)lPeHG~WxWfBPSHrf!eKo1@()MFAmhJzdU{X#nyEi&;Lw>l@>*R|whM9|!
zRp{1kUOa4p=kv)9;0)j(uPTNS$=EyC)O4b|6e`1uI8ql?r-o&rWDOJ143CPiNa+hZ
z!2)m5OyyQ)Bw`?(CIA}aQ)qaYGLG{(E8xsLBDDJWvo2F6gXcaH!zURKg)NY6{s@p9
zKwn{`O1Y*uJGNykum<S?gdG>wtg3x$=i^uQB(TWUQXXE^b6z$p3Ff&KujZ&ehs&b@
zeCZ^8x~7<uxZgM(iZK|qH8V?zOQ>NIxEo`!sO5zKO_l1U>E)NK=`VkW3V<`^w0XTx
zPU0<Dre1fs78C>zG&Jeh3*b!fmL0V?UtG=x@z)NdbGyQ9c#$8!$|4d_%WDkyS4WT_
z#KBH(>8Pb5D*hvvW)@LOUTf>7&fQ){?E1l0)&{!kdm7rDlXB``mD_js_aOxZf%j{&
z&!;ps%HPiA7*Pu~N#iUy7^KPMpFla~(oX|<TRXfnVxiN+@r2)lTm@=xDSG-etlH@T
z7i27;BPMU<e;58S`Sfse;6c?BLa7M!$HW2+69=n10PAsVc=$(87c#1Ky-yJJ24hCl
z?8Pf`Y98elpMlV?RmI4qep+Hg#_+j$MC0A>jc<)j5<YyNPAFJVL083yZrT1moOM1s
zBsHIsj2Wo_L(u-XC}yIVlf5{XohA9(DZ5m?O|m$SiiBU=ay(y;o&(KAQjM`<_<XJL
zgZ%^k@czKqMuFkUSq;nAW6dU)t6~pMe@>2<o+X*07hnDNnJ`Klz>=D{(Hl7k=o9uS
z8vq@)qCAV)wtt$@J3$*J@z2utZUJ}HSO?vzn2<uU&F|oDp)wTv!o3-Hcx|)tS-Y@A
zte+4FLEGLV%p~mmE+1$qSqV(CF@>!1$SgS=)i|YYB^ZQhZYG9IgZG6Rar_9lNr;UG
z#na%zjrbxTa_y6=&xd{TY_Eq*_LldM^MbT5EZR$-i6o!(6Yyrgi!QrgU>Xfej$^fd
zQ5uAEFpntY<NzQ6cCLm8(A01k<9}|PE1w5$W*-3nid3wlv)AAcK<=cU{#jO!y!g+z
z075ekq^f#!nms`1^=rR0N(@eS1*g4;7k18{7rq43I7>}|ysJ-xvE)Z*Bo;39l{ms4
zc-2ThMNr-(j_i`Sxjz}M*H(~m(g<YS_lI$eHy<61pN#z&ka;Zb66)Q(PF#LVnfBd2
zew^6IDC4~M1FY1GWaqS%qYg;lFNLiA(LCFdeRtuQuR15im-$lX!`Z6)qPT*Bg3PEA
zZt-dugB@M5!q&XX@P9-_OByHw0)pl?;`Qm$3c&J_8Y1Y?rb-j!d^mA+a~;<tXc;fE
z75W8Rfl92ieTinpii;vWnJ#HJ|JQ{qqFU8DIa@fdhZ-5bzq?^CuZ9CE^D`<Rd)b4A
zn^M(8iApDVp(?Y+XWA}f5n9@UD&&+h9_nz<(@6xMLp9~OCcA=PQT}u)LE@4WrX(1N
z+-G;twzuILo2Ls`TZ<^9O4y}d|K-JwQUZd04R{oGhg5CYPx~s*eHb*ejv7C+-L>^<
zYh!B$)c&!V?4xT>Z?AZ%Et-FXlFC%(iwe=g>=K6|3tkH`Y6k-vME63k2Fl&F8Kd|f
z6oKv9BZu>s4n8BbFy%VjSVgQIp|VL~Y6$~X#cyCv^o&xXlGZCXARo$CXmc+R$4cHj
zKb*`Yh?AA`pW~L0k&}z<7TkazyK~mJw+b-iQGyp61f+*skF!mWM*i;nB=QbJ)5TJF
z<@`e=e!_?WPi-DWBVsUc@Oq&%a4H^mzZNnapT-wbVr5T(lI#(jq*$aaC=*yOg*!b^
z?pfcDkIyMScaB*Lx;^=IB%cLVQad}4HyO{)t7q>iw$B$uK2Eg7r0w4d#kQYc2N%B<
z8JZd-k<*iJ6?yah%62s+<Z_UopT9l^OBO!br3r{N)lO&nFSZ2?I#nhdpFv#X{=AiC
zg}rPgcd?Sf6agdAatxGqfobLAf3#GuPMB)a(4x-R8d8KtFFtiO$s!`oTfI$X*fGJT
zz+)a~VS;Dd90p%+&T#%>k+mVwR4}Ac1yk!=?hRy9I;ReA_Lf?Q({{)q*U7T6IwFPI
z0=(Xl1qlWMULss}^U9|5cazX?ubIB%%xmAfrf8~h2v5y6Zl9ZeQlkzs5_x1u2keU}
zje`FGY#d^s5FxFd{lJSVCc!VM7t)WC=whcO=mXYPs8J+VyH_;B;j*hC(i_U_`?$wt
zQU{a}>9wVx(m{h<;qKzMDLl&Eqw@joU2y8%->mAlRTa&@!6X!uq42tenhcz?S6Zqx
zT|RtYKQ^nQqX>4@Wu^&K%O8aNUuz^__JhBBX*j8N+Y872knlnf4HPzk*A>yM{Fr|z
zym4rG@zGA3-vQKE`mdx&7W}wc1SU9U3wGyicsdvVr&gn}So%DVxu5f`lUiyJ8H)X8
zZ3}N{X21G((Kd_;?vRj)Oj+3nIAOpi=yOwsI8hB2R==Yl?+F#<tBk|1pT+64d;O1C
z|6;zLEr~%z-@dd-V_O9*NNFq>1K)5WSqfe!1@kj#HIr4eKR4gr-|tuEy&$VVR^m7W
z(7Sgu1u_}Gy81rMP}dS|=Zt(}VV9%eos{L#ao}sy#-{ODrb!MDk<t9Rn;qtwDcyXW
zKj#hAy!YZ&z*{ASeX73R2;k-PZm<XaMjR&Rd91A_!vdkK@L|U8Hd89i{RlJo-*%f#
zcc@?l=4vx#lwc&@90n0s+9qd-aDt3b&oeV;!f!VIV>Eh1rgAx5T>y2Qr&jaXhjZfe
zP#x&y0pleR0lpTQyhe9`td-9eLIAwoQq@M8FgQK)-@jIv&s8^H{8s+YZvhCu*ur>t
zc<OoN<m6sJFgF7TNxjU!Fg%FGNCcRgf&+B`-75j^Vxts?H^+e2e*doh;rJ8K@zORa
z>r6}Z(qL+^=)99w|2#Vf$T8X|rE!&r`Q!kXln69F&LaaI!LlJmLk~$P3|%6C^R4u*
z)@nsNai8gEQJWMfz<D0Tz1YsG7K-~)waiQ;NHA*a0)GwWu(B8by#7_SX@r4un21fH
zzxg;d8<uX`&>rZicSbEI*n$`j`kDdd_)z&rbG)W75yN_m`9p4amo`s1cY5uPf<WBi
zbA6)bY)zNRAm9eisRa!CnIDgo*^0Af`Mr_g;YY!kn)J}n&{sSZ;Xy&Daf}+RE_FZ^
zEYIn%A1B%GOY#-Sa|4Mw9@K%z_#j@T<=1y)M3rr9=CiVZnSg{!LK%>2xf6K`pZT+Y
zyfy4mIR|)FL{p8xz`%o)nZm>`Fay-n3%TOxTIk=s2eN@?=8=Pothk7`%mT-450(>J
zQQ!6FYfW5AMG2gRfJ~+LDPj2+P;)t8t6~s0I#Oj>Gy5aa#Lb*-^d1;I?r?e^m_A>z
zA8*+}1xc7|l_j~2H)MsSG-#11h6^;hIQ}*)lXNchjzd9gO!z_(Si#E0Dc#LPAztOf
zv;MAN2Cq$5TJ%4$!CC?I@p2GJomnyeyWz~q_Axc3vP@LwyR_Ik<LfsuuU>Ru-7?nc
zPpShRzn{Loq)>u-v1k(YJzzb=FS?$zvH(tdXJDaSRROKgeSi{dJ^_>|X0*8+$if$E
zfrQfo#&gw1TL2)ayYlzB0VK)+?wiG&g{ZvczHc0ML*?nX;H9Bt);yqr$|bFG6_C9I
zZhkPUvShXBKPPItP$zd^lvK1;hXb%&MjWW~>&_7+inzV6FqB7uPCiqA0JoO0(^9pJ
z>FPZK9R*iKFK#p-aQ7QTawpKZwzXDSC9tcbUhGkpd3?xK-V<Q)R7KjFcRs?#LYx<z
znC_22Pt?lK6b3wSXA*LB78x?0$4+YhCP%Dm(8a*uINCtv{9s~=!(1|okG&;`u^q+6
zae|F;KUv_6%y&n#?qQAK=V*w<L?+6$KD9=AMs<+$@GM^wFyx|eoq8R)16pt0*690|
zt7#3+ek6=Cpz|8>ejyex;~Y3wGF)l_&4!07o~_y+O4CJ(RVvEMqGgef5Pq2fmX!Ku
zJs|Of8=V5~1I#WQfiZtvY8ZFxR_k{6jqne-fbdZ7;3EUrc`PP6o-n}dr8#j2fUC3R
z7m3dCos&x+N@PRSZ)%D4d{XT0iB7EmfT4M^q-PXdP;S!;m^g$2RcLC-DO5|M@$x4l
z|6!Hb#D$W5j*GD+6%PwlR{CC@E(kfzvcY4LW(9z^`k@Qiq|Cy+At9?HtLw23MxR~m
zUYBQ7DiZ)L{<LPYToVIu>faqLk^!3I5zq#j8RNtc9+P@$s8=*1fgjC~(qLH2M<lNX
zObhi@y7<DlzXG!p<wYV1%Cxj@4(CX%jr?;R+^I)elUE8ToLMLO3TBYW&8uzc@n?&(
zg{Z$O^kI7vChY83%^>E`GT>Il^fHKmH$4w*y5D`I{KxL`G1V)zpyC#*ENQzk{aLt+
zRjt4L{dZ3Lj^a~3IiFB!{nrxlG<iDY5(#m@FLadg@r(OB6h#9KM)ReSUwo@Mtd_a#
zRIwPfOSPEBDlL~d@qY&<iBK8wKA!q&ZDF_r+!N`GB9y@2#Z1yKsSG1-j(P!4ehQ?H
z>Nfz=XS6``JD{01td37XuC)Fc6XeH*+C}uU*>n>K+3ItCsnlDXkm{%vHGMO#RmGEC
znbYyPno;d7O}jd0@(Eq>55FWbMy3ZgdBW`j<Qns~Yri#v$oXO2a!)-iD>=A%cm8m#
z8KW1v-8kKSy%zLI5-cIAF*#FrFO^c853~)M_Y}$GBYp-rrek!&Q=xAJ7@w#j#E>rl
zYUv=CP4?TWb`y{w3>oreD@U&ZHQ3Dm5a9T*$bNcrT@0~~H(2Rdcnd<|c9G_9>AUpe
zf890wP{|~n#;KHHpsB_zYHn_etr5G}qZ#Ocgst9~L~{0~$A=|R9_{^;G0;?D@D}g}
zXICJdL<hYBv%p-C1fW#~iuGNrhccSpC{ncot95Y_B>`#D=K!g1m*BAo@ZLucd2;3$
z80Pa5{h$XWRJ1T)lj`T)7$1;bx0mJ@#(w;2&5D5BlCai~IkR)HMV*JJC6jX)!e%Y~
zpA7Gvng~H>^-?gyK{WLDA*%y@0N)wH`Z)`n&>hWN@ovw=^YDm6xsC~!ausWS$g8)0
zBbRL^AtgPcQ1#Y_mmsK)d54sejIAsb=T`_AD5SULf%?H6KKRS_NJa%FO&6^|6U$ln
zt{0~-t#aw8xY6ozuv0HU*&NtK=fWZ8?4gtpZ?_+mf47H|ukB0$ek3C#MQ)(ut+*Le
zI$GQ@JR%~Cm4iHvDij)T@K7ODs~&xo-a(;bWdHz-cAZxv5C^l-Dt$n2VA1=Xqwsn;
zEB3zDnEyYn-ZCny_51!7L`tQlk?xX`lJ4&A2I+2)4(V>BySqi{ZUjMEq`T|6?sNX%
zbI$L1$ruc{WAA;fHP@V<g<6I%uV<O0aq+G@5}4!opspKuJ$Om<$khDiuVz!+a4ttN
zIZG3C3kVStda$_XW7<9@qxHbI3VukWwF(WF>d~b7SMiOBhsmfQGCT;gS3twFm)!{o
zTi=13pW17<b~s(c&wPnULKi7dL-2+QTPlfe5=XJZJ~%3C+^3Q>Cdq#QTNsWHy_n7f
zOitrEH=RjI5|*D4%0a9MOA2`@mC^9SfBpn<o9qrf-^JBAGeWJB)DP5E?p!~aUne8D
zH~;aU-pYX-%9WAtgkD?L>$CE}#SrncZ-U7RpK*tjq|;a>k{I=}X!)Kq&}_iOaf_PW
z->2`sBJ-ry`Ju*8?0;lXYAmdaH6!I!r$u1WaF(397ZqNW9OBjRyLg|QfWhdfinjXC
zRz1H0k29yHB<A@WKHKPwtgN%xf1oU1Wu)}`B5-gzU7EuI;i3wMH=1%&7rK(^&6e(4
zqzby?PfZCbk}6|ma!BBjnHkd#yC75aL0L}&E#7G%?6nxf-~-_kJrhIyW+u|-q?a_R
zasU5b%n|hErg05LA?XP!I06RF0mG)c$y~%Dn|H)bl2BV7hzQY9$(2fT17q=OPQPi@
zv{w=bI5^$O7vCs3{&PA1J+-EWcvrvmALhiBB+zWqirjb9zLPxUcvG2KfOt3PZ$O{#
zG05+CXW^vn|AE)?Gu=<`Me@VK0TuMTx$3J;as&iwQlK@Aqm%<N%wo$0coA^881pg}
z0^eBVbGsgvzZ@W8!BTve$4~cvlh&tbkc`N6efYJfRM3}a1;Sh28Cj*g>)Q`_Qt74L
zTsX+<D0+JJTD&^$6BY~&5;Q+&XOG`y8gkIR!jyE?={W}mJ~UMOS<@O}o$PAbWOBuZ
zppt~XNg8@)80y4W5*+H-*vcR9m$5!Kz^&W*3kk8u2lyKsXFAOL?O~LJVk}-75nD6&
z`5T7J4E!<rh)8_Ftf;%u+@Z+_38X{-mjgul(`vMuyzq<R#c?#?!6sz<&rt>@Uz;zW
z-SzSY$rB8~z;H%R1HI$SD@n;1oxszV;HC>{RvF(gmx0@l?0xk3BAt(}##1b(`+DF^
zg|cSr(xl+78dB`b?QzKs+^qtE*;>UWL+Ts5CG`{K1}Cl+m$I2EMN!=Oaw=5GwN`vo
z%Hesd8E1R>HaLyO>Y&qL5B3*`7$`F*2uMgtpoG+zzi?j<quc~?dad1>piU87|B1&|
zq8f~dZcQ{7H}*?5{V&S7uSGO+@N<dicJEd>7IR=UA*t3=Pyq}AbVWiUi!hdn2$SXj
zo^3m^cxvQ@bOy(y2kzB>v|BBNRX;af>9=*jy(jQmBzzCv_`zny*>XIQ1xxj?keFuL
zLbqzG$cc1kt~x^6r`XShz4v0D>kY+~?4uTgSfA<6ZQSGMMLOmDVmjR#)n6VLBGw&u
z&KIH<D<?KLw`X#j$-R6--4C87OFeQXbN3?am2~Ov3`B+dKavA`V84aQYv21XJs>hG
zO&fHb=kH%!+#dj4TFokG5Wsc>fJtc0>4KAxGP^>araS@?XUO1omF;}IwRDHf6o4Jt
zBMV#&!ywmL`t<eNx6m^n-<m<FND%wX?o3JE=K(_`s_3R=m-NPn^^iJnggM-vTj&H{
z0>kf6B43p@#{UWd{zozt(7J&aGvbUDvgp}t`^_6yx3jYo4G8Iw=Ed>>)a#qz&>M5!
z>rBNIe!+ytYfYHgT-YwDAD|iFuCb1W1{qD(x(U-#<E~aPZdtiSIAt|!x|D3PY0w9p
zj1*5!DZepXZtzYg>*<#o_v9M5wARWO9-OLOJDT<A<bD1&5jHYCJUe@ZEnycKk+rwb
zdQl1}c=;ea1JRlV4CX;-L5J2{9<d@w(*WW8<Ycn{sCfekLG7?q)vp2Ng~3TbWMmIr
zfU>>nyg7i1!{t=`;qr&kEyg`@z=i$sYB#g_1kphhA~}kx^gU|uF!e3}Ve1|PwL)&-
z`Qfrwml)wcx5U>`A)@suVua<#3jF2U0&UI?on<Yu-#>ciixs}R@EK)dTuC(d4@RtN
zd-}VeXBizeH&i(7OnLZxaO>X#Q;}N%ACN+EVrSeE3C4mk;7`Z^EG;IyR4nB4SZ0xj
zI{V;#LR?=yXzz~AB1%ok*NN@0>l-@^0iBCvXYUEer=ahjLnkIONR+B}iG29Lvvp+c
zX*|+=;&7KqP~g%5>#<+J$C!{8f?UGUV`(2SPaz!_ncg^*)Ddz)m<ZY;ORRBVl^X<*
zkp^Z2lX3V2sb<e#4#V^SZKZL*ygNWBO64Jc4YEB%>Fl<1l9YxoK}5TB1C^-{v@1o+
z3I+{<+$C%j4v+UY9tePFHmlWK`tQKc1_}C!4wqqX+bQ|EW#4elMZC%IH?eb1Gaa5o
z3A?TUPeM{R{=<^P5){!oVCy9CF**#hi&kv-sA63s)ENl%9QMz&uGfjk&U{rijv%0I
z<q^81FTLa-T;nQsg|}RRfQ*#h<tz~k!yPo+wgLyJ^#(`l*#K$ukGF;Cj$f?is`T_$
zrq>L(PX**1q5!4M%JiThu3Z@(DdwBcFMsJAca?Qz?e!U-$gLdHEf}WoUy>5{b_+i4
z7yX$yjbylcx`KxO4p5Ov3RUgIC%ec32Z-b*=1;#Ai|vS8syyXNlnc@z-hC&lWno&g
znxrCCq1&qjQ8t<#z6hlv9|};?#sFu&CDx1k$$+EQR5-b8{n4Kto=peH)Q{$Ad?|mh
zs_0VWudAbrG&ta_AsK2&`nsQDy-q3>)%i!Y!6MrmUu@9x(2nz5W$+S)Sv)>m>G8nT
zmM{R!NR`jIHv2!fCM-h64>mc5;7Wxns6pEXQbxj9Yd+-RRrAX>(ksDiUT>#?&pS0B
z2v(xg);Lxf_5O?>pO_QA#;@tkhqN!spoRXHVY+8#DPHs3mWnD<d^4mS8T5nGJk`Mv
z4i~K*7K8SL0dG~urDk}Y?fvCA-_y)5%|fihaUTH4xB((%!~8tNxK{$&l2%LBZO3$H
zXq#O)N;wO9yQ{s)yby@@cuy_+J1NoHFy~e-PMJ1))z2meH&l%bB%6Y2e4l2k>?onY
zsh`E#28lC;m6G}!YZHI$sy>NKjPvpLL#-qG6(tveB#1y_NHJ0GOgssdg|6<ep$Dfj
z<#I(8+L)}}hb$fL%}(y0RC{YoiGjb{mCLnNo=z&T_p9EpeP9bHK(fh~{|)->f;ZpO
z*ensfPGVzV4-4JGgDDOEz;(Ia<E~2JT%x0))%%C?d7w}wRr9Xnp9^ItK$nqdQyeF}
z^ylmO;|gL*55lh{Z2m0*TaurB@Ke+*J?@CUd$uID3RRz1dmkT6LqZYR#xv`47b6pe
z<ow--QhA*(VeVVUOPCz?feO3rOxKSuNy_b-DFwXrfGgnqA{B1aR9_t>cGv)9!CDb8
zdjKzK7lhFSebs44AI3rRUb^iGth>&>Js1Ewr?{eIzylnKHJy|!8!P6Y$+()UnoPa=
zJ9r%g?!gz*s8j>{p}n66zt<h+c%hpEnzkCXm#f$4wPWB<E8=JKBg`WA`CqEc8Y8mN
z1HO6us1OJRaz~C^Ne`2_ZXMfs_y-Q>%s`P|uKerueugjnUk;GzAFum*`y>dqR5it^
z73GnL?d@w6D@^Gts0333`M4fA?#dnpis7vddI(=#cu6B=wbA(a&DTceR)Bc%tp!F_
z67e)OjD5J>i;Lsqi6P>}FJy+&^#a5_E%MBf4xzl@FLKP^+t0jC@<l<mNDyL8_c70N
zFo){Z`Jd@&8)n^zbSB_n&!kyV*T&E*qDMGOY9XWKs=2?rV+7zwoPNPjc?=ZADC*yQ
z@0e`3D@+iZ_O(5<oIuxAqQFEAC$4QK&AS5%&_%#W#HRKq%RMnd?*uETW=3IHx3X2P
zwcnD1IM7jih=ZvYOBi*)6CCS66Ba23n<czO@$MbxJQ6h-@K#mo$ka4e|KZqq+|gWc
zrT8F)BYQF3$ojU~AWLg^CUNvGN}^<HKHKRFLHKS<S&hT{eUIaF;0HV>L|1=J{h$5;
z9p*#FnuXJII39N7lM}|@2?eWboM=1Uw^u)HyClFou7Ofzs5{55n=AY{HNpRffF;rk
zb$o(`BoeIep#8mgc~}N?8b@^m5fXmm7%}V^l`lQto)6Ou5c|PBBxFDYYcr0bv|gcy
z<76dg^v}Q7TEl{rPt(XrHKo2gh``{;ZJ@_(#^T_a%(ar@jIE{>ucMAzr_;mvcxhff
zz0(#{^Cm*DAJ9(J)(X5pJ$zxo1NAvHGFeHj(xhFGFJ6Il7VazT|N2EJv;hj}#I)CP
zj+@XTv_<Z8EO7Frd+D~Fu}nsN5U9sKl0C#?+a!-p2yaWDu;>5v^#v&AChf!INl??_
z$7#f1B=OL01y(T;T@}Pk0AZ;W16$BZaR`vyAe4euoQ6WsNbOWlr8&@bjzhx9IA*_R
zSP)BE!|b?4tX5uUv2i-fn!OIb@F;ecJD8V$Yo#b_s1iZGBun|=&abOky=*U)t9Z+>
zpF(NYIF^($+)wuYJ|U1mgI?^(jmP@w0Z&^PRQ`dw;ZuW7HWQH{|A_D6L|X95Rv(0+
z!&MV|0xb<#Qpl*8bU$=FFZjh1pUFk}HVp_G-~Fu)?XFn;uLU*@KYg8YJ6RXWOe9!i
zoGn#BcheMr<hj7Kb-p2UEsF~XemalC>Zn@v<~et%z9?tvC5q$^m>TN_-gK(vDN**q
z1Rq(7cCzO%G392mWdR^m`7gxaFONNXo<lv~hXSdz0uy|3cit7V%xCd4#+f<OqRn5l
z$<|6!v0fk^mU^3QZ4tWeSyc>qYx}=+8!i_Z?P1T{WJ4HIds?x*3r+VtEm8?PsZm_3
zoF{*`sX6$0#Ex32M~}K<up%ZECuI8{uiK_A^q91wk$C109=#}=tV)btY-)uIo07#-
z9LG>CG&@ee`nZ|RWh-h1vEBO@Fw{Y%)wNVv&G_U9G&s__Xk=9u8U-m{cj_!V?A`$`
zoE<2rv^qui^wV&xu_z|wGE#=5<3F8og1%upZlx%qs8Cc(FRDfBDath{^9(`r-Ydv`
z=KJ|iSza5$KS00y0+>8w5;3;k#axfmR7*l&i%4zNn&VqOV-Q>3SdEK0MYB~B?JlFp
z^LCapbWT<|7bky7t=uoAa2%VeCEBkFNQLDWjlL%18Q)~pEfUQBYVUj@q#V9HlQiAa
zOjUsYUbV<rMuQg}lYTqVIY|(D2bEukvbmZFT}yT-5CPpu(mwx<7@atXG8qFGPOB5*
zN9+HUKSYeEpZr6a-xSB1dlOeq%#@5t%{$gru(fFV*6DaNbp8_b(M2nc`wEn(0;4;J
z6GbKZicraRmWMTp)0vEl+KV41vBI%xK1<|kx$F5_-%%<!H02x|uM!*%M5X?m+CLTK
zE1aZAAfPE1rC(#H;*=>b_msmkXMA(Ul5mp5P(-k)WyT&K#e#{+@&uY@>ml;4DTx+t
zI4sDxDExBYCeX4#n{gD>w9cOw9eJ_NE}P9;A(j2*_sm!(cL6X^ak1d<nv*c0MS)6I
zD4DA}nP}I{fVU@zXy8{C(XVeaE+XZ<AZd^{ADkv<Z*T9g`V~s29}$tv=-oc5G2Y){
z1sjr2tsan0Sw?DIlbIi&b#Xpzl~$rMojAO|_qc&os~}}L^J%~TH1Z3(H2!$?9y2ju
ztK+Tn0Zj)2)0%_BgR(?u3rztoAKkB@;gWQNZQPc%Mft|istp7uX2;17l=JBPm+z?@
zfxS1$w(CJP?Y(trT@`?{NaMb@qLP25z)}j5bB&hnpb35vz69WzPHD%0xcY}nW}rS;
z+79AusbKtQN+NR&+9VS&IB0YsgE0p|`GbxQf|(Kz&_KbY{+1V-)|5l^dy*&me~;}k
z$`}Hs&@nx7F*2xaRB>9^$rjg@F)c>1S`>Ks6UVIp=z0-Un{;;x(fTTyBOZCRFPnmo
zi@vfRGCE~+fhCMR-znvX8gH_dn;j#rF-rb8ZK0ff&qeoOnyT~<-$+Ogcs7*%3QT(D
zt^I>+fFF+8xzvLrr`#XqgLw1V7<Yp+iZukt<ULjQt9w19etF=J%k{AVsJoIz(%I$4
z1MDHlGOMYCkhdP<<FB_S?xVggV{Ck@1?Gv5)-uwL&))uv%JQQFAb=E+Pn^!Jh%z%8
z0#a`-RpvOZlx@;D8=@k{o;QspsrZ{bZqPef3paZP>qf-2QQVWmjq5aMU{*!H?44|{
zY~pf6zVVHLr((WLV_qGp;Z*%ad<??!yS=<uOO9NjkA84(t9wPhehbx<uci>x5XQAt
z$I2<jk<hTht}LyVBnh4v0PemQ8)>Y+M0Z&FE40+f6s+8)@+Ia>dEWMQfS%TgI~?(6
zd>IzfK(}+e@l`gX)St@*uFTdBAFF?*TR%L&Xm%szYB^5HromBYEY2QtNmDc2iHU6}
zmux@Vo32PCpV2GvmEaoQ0n>`%A@6PN--ltFY-&cuWWT==p5bdK0zX=1RUizL9a94$
zT|_bZ1&HnJUJLYK2to-jQ>kOjSLs#JN6k5obSgl{{xR+Le*rS7*$|xP=o)KNC`({|
zIs3(TWIl0o^WsW`pp8&>Y3kVIsDa=X<bfJ%md%k)atHlU&srB3wTp7{h63KL7O(%H
zid2E&pn~;ug10s!Ny(-ttMr|>ZY+n1%)N)c(vEytey1()NyH>$?k6Do{46VHtyFLw
z4<h|;@hjqJgf`pLMu##S)^%@Ah7p;`7bS10o2%L2J1l|NpS?X=0Lz$t5d{3`jG>H>
z@d+>_`AlwqwZ^e21;KiPJTT0hYIk>pbi)v@7T^<sTRkc1>moeRTN<6aQ$mG={WAbS
z83LM(-EZEx9-k^woI!Pt|5J^Q*1E?%9rQozV6~L~_3@g^PnJavaKaJfWlx^f=JBD8
z5U*)Trt<lA`dp{i^5h-GmaRTL9z9v*P~-6MHC>s-r7b-oa8`GgafQj04~?a=m%pIe
zVzNE6FTg%jLaa>x5^Ok)D6$!#d!b`^{3*6qj=HP>@rH{(UFQ@JsyR{+A&`F|z!T81
zdXe>@2Y5d&P9ST@85jry$yq2T&;enXv{8P)8u>~&Bl=Y#_REL<NW7veCM1}LHrFEt
zFifA*W3{7X^S3{KYYECx%!9!r^s=xbSN&2UlfBDC8hX*fk7j=(9r!;&vs=dD_ROcl
z<(HIj{89u^?#z~-w>-bE7shtnLHGRd5OFjE(6JBtE)!n_8UyX+gJV<g0Ze#sDiC;n
z;~@z(H@D`;&hY_#@SP-sW_JK^GFTim8_#8EC0-9}8^1ivAixk=J>f#fBt@+`Y3Yo3
zt}RPwt}N=G@a8<Sw~eiwz7$`SaLa+K#PvC~<}>yM&VKm^orFKlFa&rw-(D4Q!to}L
z7^~cWnb27IKBNw)0m`(%ZV3FB(^}^&{*0P?*r440v$3{D3=n?mrHqEnZMi9SL;l}>
zOJQzk^s_pqR><*7NQ8cVXIo`G%NU#NJGp7?_4mXxCJ!cxP@RSn>4_l>p+>RY>Ls<>
zh3?5I!)iSn(LNM1`r5Js5LXUNprhXm09V#IrY;pV#0YxJ{c0#qCB~1Nh={*`jr4sd
zFg(qD_aNLB%#pr%a0Bv-vA4oYRQogenewyM9GbUj>)aTYeXc~e{9me9J?jC(WUx;#
zmAZX`y&|N514@@e)dY*Y0_ClL9Nqw0)RBAp*Vm34+_$1DRi>!$c-@>^!^!h9>ZF}@
zC-1Q5BtPiCU%C2?PZce%UNIltkW-F1WW4719PZ3L^~~Y>`=e{h->{l3Ua!d!Z^c6;
z4Fd=lml)*Hek#*aIKr-clu{1pVY>3I6_wQucNjd~m}R`0e5tUQ3n=1tQ(kh<^d#RT
zd<*mVJcrd;UrX}3O;*mCE2}1<Y@Sht8}IKrvNjtSbu;K53OlTHPvSCo8x_a0)M>fr
zBa_KleoAUvHBVz!n~vfNTt6?@kpB&?lY%#7`+2PBx^r<dDcI!k6H%H$ET$vkjhOXa
z08^rgP^M8=C_^krNfpuDANVZ@i>^|hDsnYN;&*UJ4o8&Luj6kxOoo9w6f9<A31D0*
z2@%Ucyjauz4$OF8B=LHQC4p2q1W~T_1Tci^+poSlW6*1!2hZ_&jcfLPFlReZC&f^U
z2xt)C3{Xr<TJ=C-_IF-Ov**R&gHPxHN|2f(%rLL_?T=sqQ>RUaulR1oO9fI`%Y@|h
z@>kfB%oz3&nRw)!y-Ur4mDEBoULSazNK(E%>biehW99fNh8byI5dVX$RF73WdwhO4
z_n@d}nEWJBqV<)ivXx+Pa~%spa~wba<j_?8{D<!M60=D33Jvk~9@oehB7J-(mG{~l
zKn3|aDy4tbH;9JkEips5j)8(^a<Se`y~zB889u()VClThs-E}O`&^*{g%@P3xLBdq
z5Wu)QvD5o*R|3Tmxd7ff=e}IMMlggq<qdzyPk^4`jdBD<g-Z(<KAjuj2VA^NV$8S$
zQ@uUu<=;zBW=GP`tV$m_9`Wj;YpOItlof|Ml}Qri$+S>OXfP5n6(#Sda}`O@f{J6v
zb7MlSG~&o>s}@{YjOMQRvRTIV##mTP`$pXI&NfDM#z1iM&KuA&JGYxSD=Z*s$;{I*
zk9?mrpwj+CCbUcggrGqH5^>^=p%$n0&Od<gzP|_rNY|YLU%;+q4t?k>sBX|um0WvY
z(xR|?Jt5qxHrePLIGWcxnLohGhFr#6g1~3qAmIT8Im!V9Q{I+6!7k3+g9Jr5rEgEy
zy*W@8QHV-Be%j)=ea=bMUk+Ncl(^b(pK89Afz8%f+Ux$b6kUO1mc5w|;iq7(5kvA>
zOc@2Up!`fT$1-NoHmjKW=)PIxjsmtsyux;Tp?q4u-D~AF+8ey=tp~nhk_=3F5NW9H
z5UIuY^8Ch1`V4iX#35wP+&W+?ALoPTCy}ot`ltgiFpZUXq@6x~67~C}MbWmhhQ7D1
zIm-@(>R_nURo_C|J&FRq4z&AouEjPWL1|YBF#tn^H)JWquiwb|JK!{)V?eb6)?OV@
z1P+u^u)$c9t}vE-A(Gm>myvvTfe&tmbgT81E`W-c;V1#Z3y(~ay4cHk0AlHgw$61n
zs<nVQTYx|?A>F!<w2}mm$SvjlHBp=SrA^0$Y7^foIN-~#gcijKRl0n<mwiWsC(m(M
zRK&^dTn?8?!s>|K%z3)u&L)B7XV3Zec{hu{1<m&6k~s=$o=y)l{K#OP^b=8H*Zt||
zxfH>eNVnFclm|5r3$ZS>TfBjuGjz9t`f#V)@tw29{xL@+T6DS0--Tgzi%lnyijoD#
z*qwGoFxOi#B{VUkm3eEYy2D!ME@^lj5zoKwy16Cvl)z!28da#n&_QEDeg6An6@LxQ
zG`XF?NsL>P51-2{6A>MMhAKu*l`sD>O{xZIky)CV#{@`DHgPW|i76SOPndDvHD<=v
z$+S2U2EtE;#<(D?2lXL44v(g@Q%;@lNMb$e^A6RR(-9k8L+ARjte@ZI`SxMVBycCk
z=!x4vOd7}}iyZZ?@TMgx)>E))oS_7PkwotbGpt#?nq596PnV+;3Jww5YZER?<^$na
zGf$#TcC?tG{Wyd<itXF`1JW2>)=*)W-kt>KX)~pj^yDNWkF%zD0dLyET*o+!tajgD
zbmoX!YP94k-%q8IYs!Mmi+m6H$VYi~Xr^1|lD>Q@e`s+S&W=*1z^<`Zq>~eThi2P1
zemM#*vTFFhJ{&*V=w|}v4%ZYiNIL{3I08tciMWd2-Uq%AJOkDm-Azw%PUFb-Ho)U%
zOxG+cQe(JZQ?J(rtn9X5LiMuf`o94zd@>M)MKzGQ=)Lq2<fY%EPd*aIkW!|$%G0a7
ztWN@yH2e9W{nW;HAu^F)$c2gQfs~pDVLjPq1NMj|$_W<Pw-S~uI88-(HW=258fc<F
zD2DJ<g7UZl)ZVT>X086o4>d?Mx|c!5s;&c^>XPJTUCch-C0#VXbt>sGXyV@x_e7(0
z6IL%24I4jnDy6aRW@=nh(opDLW=0HP@@8tlf96U2>(=`FT@Mn;Z;d^ck_n_IG>1ze
zVw5DfslnO^d;+@}@&@6_UKlc)j$mijMNfYr|88{-j)RHL&zrk!s4dj}o2c}-y#D5!
zTuy(!1kA|?XA>1>r8vjQp`6WZN|U_=$qL4<``S6?DcA?`D*jp`Cv!SHR`b>O8`7m6
z0RLrKu?}q6Rd?!gskIK?6mhhh5noufNh!)Ug(yBnn87G6_HD;`<d1w+i){}d=e;X`
zUnR4<H0EtCcQP5c<j;;POlodUTz;)63ubitm!&3irN-^MQ%|0m0;FmrdexGz2<god
zOh<CwT+zIn7{hw?I-z9QCVrdpf!gwlhSIh`XP?U;^4vmOw_t#ZdQxLQ8M{nc|LSdY
zjG=qLo59EVjIU*%>N0EozBvB6b7)bhyuDdCZ^vmC+N=vVj}^XKvpI!Ee~!Hm$Nx3w
zW*mHhdZuW^`w=%tS_P2R(1pC##Yg#-6=0NP2F3sztD$6Pz4tS^$Sd+MBcSqe&R1}F
z1U5Itc;ToZI)f}pf431fi)|o|<qXh^)`>Z()Tf}7g0ak=J`;tAG7!7is9E50OuWE0
zfRRF7LG4J$0fjE?e}eM*4U5`6MtmG(y#dJ^H#j~iacs>tFvn>d6+Yo4`!ot16wX=z
zrdEU^Ps!OWjPZIGGXD!0`b3M(fBGj+DIol)@I--(=(SIxIoXZRVb_xnJ1V6f84SVw
zvA2^bpG+m8M#qX+2}gX|uYp~k{97+C9FS$0A~sa=j;U}%g<eyrq4_l0)wqhKavVqC
zH-vR^k(zG&Sol3{qOYe{s!p<Z>&2V)-;xuohnV8;dE-?7^EhhuK;btE)W}i7ra(Ur
zN<cQ>o}w|6JBTB|>q=_=K}D@mS%bl#KwBYWY~g)>Rw$iH0u=j->=Z-6q*|Pc0Y$5h
z6Jv9wn?vasg4<$oUL{=B$ItMaug>GAhW4GF2G6)qJ~tVD(p$HraT`6D$Ba=cvYqL%
zx(p7emA#>m{iDq?+!i^P=oKmA2!|l>sr>&wDv|uO-G{{sxOHVJf9zduuqUg=g%A>U
zMuJ)YMeAM-fg~Yx5>X*?Q`veke)vIx^5VN(=d7@Z3*C67pvz@;%*p)y9ot8F76OI^
zUlK*l2z45vpUSalL;Eacv#W3_#y_BQplDu;`e$#doCH*38T{{Y1oio=FRmXLtu{X&
z&c}pjtx756^2Sl}EU(zkgwNMeJItIn@Jy72nFw7wQlnD)QXAWaCYoNd=axYdv{f}o
zKb@$z&X$JKx+&}y@!|Fc@y2LnkPcb*xpIFZ<968bpBbG`6(;}wGjK2rqYi+P6G>O=
zj1yLukQ9<qfbwOD5rX06O(k<fQHy`VpoJd_!s*pvwy;e>&D~{KnqsFpLj~Jr&iQ`R
zo^c=HU7hxGL<zFjeK{;RN5f$yu&B*bB*Jf=BcsimCh5BfJ_4*au7#O%qlDMqNNPfo
z;|BgC;NBGO#t4NVk*CM7tGxap`OhKs_X8?f_-ot3SMoMnT4Pq|`C>0$;oWmng(2aa
za&NXB`MQMFywKlg4=q|-4Lf*bW|-TOFhv7@)VGq|c!h|+XJxl@Tc8TEUhT~@_8Qk&
zSTL5f0+P&xLfZh``L~yh5Lg#E7Z=xZK1RQ9ZhnC%gi!#BFHc3K*}4DT`QVfgr-XdZ
z=*?v}??^&&WyoS=J{WL!D7P-F$ijYbi<fne6)HGZufIsCC+N_&&08Lp42z14e@;mn
z-(v}w>=27H$WJ8#VZ)Lhn>cFn61C7BMDn;&X%<;8XTfCBHPc)l*!E__Y{IfcH6UZ)
z1;`?XpT>m5nb8E?6y`q$B!tz?;uhl4Lr;N!_K&x1$3+s7Gi3k8?n9sr1tW?x@9iJb
z<wGgPj{K$tR{$Q-fgxV2h49`%QW9-M9x(AKU6t8S;qCDcgzPf?3uEQS6VS)fGQ{EM
zIXo||Itb4&zVU7lA)~T76MF75X#l=Hl12xZ(-8|dy!n5{!~c%U5rMC0JtN(0Kbyi;
zE2~eXgwx?~PDKSge|!;CNIh4UK~O*y@m9%hB1A~O<JBzBtrInEku#~6>=ro2zbDUx
zUPC}eP97K#udpuzH2tvxp{XCh(>(MX`2!Ka)P%jCh0q*W!&v%-2E;cJ?MYMjfU05w
zWb;Kj^lvRd$It-g&Myl>@%@}&Dv)<}5kmz=aj9IB^zKo1L@U6LP4N+o{yQjwAbKFM
zts)qs<~$(tUh~H@#cjr6xBwtrT8|)q$<v}`mBp#(3b;7QnV6Vv)y3Z-d?a?<WbwZ~
zUCZF;KqlnXMpv(oB0Ji06AaGg^VQ9U^B2#`0ZWz!K;Dt{hOA|Ob1*)r`X?aU#slDg
zW}A5S#N&UipH&i=6joUoGR3}E5ppPs1a{h{2u8%?gBv4U$<|k=sMRnhr%dgoTJV=w
ziBmU1WULc>!=yolu<EQC<fmH+gdR)8A`Wn^$QC>fEp!V$GaZ>KwDzTZce29pQgS!o
z7TSc6^^uLlm4J))f(r&ESByj}bQLH%sek0_Bcu5qo4<q!Zv<;l45|^!*8A}zMSjV#
zT<PqXBM~G))rA%ZGM8+U)Hl_@{7SM2wkg23Uxtg9$AW&DN=D3Jjq8`NPATyK)<AWn
z8@E{Hk&{;;^&gk$cC=E2DU-AufXxf#GK!P@eSLkqn>A#I^qTd>ol%yREu!fCfQ|RD
zc1Q||1P2?txGKBgzgH21=rh_}u!f8r7fV~4)?mEa;QM(#qgMM7=b;14OG324bmR@F
zK3aJ6@2<<v^INR8(W8PebOIxS)pchjh+BvgJ7iwo?Prt$oa^&)GjX0OT(4N&$ZCC^
z7e#M2ldadc(;IN*i`RZ_prWA?B>$nMe8p-U9>Sg<w$KWuPw<`2oQ}Vp?oJFEcQER0
z;d^3?h{FJfe{OLg-}H_s0}=y+IR&?%<NL%-t{=oOEjT<L^$_zq=mQF`CbvPqh_E;Y
z!d8-CH=m39aW@<rLOJj>&Y`3&h>ggi-hO@`V|4wcqsI)@uNboJMvpH|2<96cLO!83
z-u7?JQRhFByU`^^jR#ZzRG|w{e({uzLz8)U?UiNiYB$CS`hI~T7}*em`bogFPLO?c
z;=bP}y~{s^2L<dz&vs~3h}30^&ykA8DO28f2z5yldL>ft@5#`L6BkI_f4q5{d22Yd
z`#{rg6mf3hWRz6&jU(TI9CL`ETwJGoC-?Ft*+)u>%#j&^W4AAcykV(SWC*pPUvz|s
z5`%jZsT2g~!8Kj>3amXyH&!$Q!$-P$3=nW+oSaF04^oi1G6Juu9Y5la-o9pE`VtN9
z(7g9bCPtv(2Sr8n?!!G~NliZZpyK#^eF%iNXVvtaL6uAEwHYhsq$6!&+qy$>WuFr7
z&|yxBIy7Nwcp1Kc0PSv!*Cm|~o|d<*R7K?w{-*U$u<PM-Rqnr6j~ffL=OtZCfz&Ov
zS~&VG&mk$VVnv^ngJZ8O&f)>kc=iy?=tAvYoM0@80c$B^1TrLz0W$AdDrnX$3hi1g
zq52m_fuURNpFvfs@1j#*4tx-vE1+i0rTk={$0TSK2*6hZ&GsAlfC|U;V|nW%z06OS
z)EfBf;?#hMbgqFPo#YF+&o=-vX%JXXGL2i7Q`z8qsC?UeWl<BzEyt3R>i7#WZFuRP
zn?4no_Mt`5ReDi(?W^nDzZo3;7#0?evT|kyqohCz$#NrGjxfFZE==?hVw-_^)x{@G
ze^UWL0@XV2sUF0Oc_7$mRHDk@QRK%+-#>FtLH`4*C!h<Ru+b#0Wxg}9yOu4kp9Zpn
zd&qm&_iI@4jaC}V$hQ-@A}<M~;c5|?91*>@q2%oF;RzXENlUiyLXV*shd7dx=O2C<
zY1e<(Akvn?ZmYWt%ZNNC2BS7&)AAO2`qPU*G;0b?Q6tUt8P0v)f1dyOa8Tu@ZhTxp
zt_-U5FrU9K6E`^AUsI&=>?vGy%6dtAc|{=d&I1$DQQ~5uVtI%KmGMio0nd~x+9MN+
zK=&&a*{>vUO$lILsS?lJ_p-U+UFz^%I=}PvBMZXl#kweDMH~U`*JkJA!(cL1@I&uo
zieMjmN(!3qv+mJgrq}k^M4{lUnA>TOs|{qf{$cW)^y^i$Y4XvpP+X&1b$|1T>S3TL
zTygm@NLu_coPGXoG{kHd>U^6p(IfZ5Y`9m>0&~Yw_RWCl{sx=VF@X?1O9OgT-wra_
z><atI=Whz+gDUXE6saFZl+bixVyHrO->08R?9^JN{?j_uzVeG83tf@1po&Emi`#GO
z@|AVZKZqk6o@$Lp&>6rkDMC>cTbmwl{;GcTZW0$0vv3$kO&Ngxv>3_7P)fuzpq7-Y
zuMP0OF{z)YG(T^x!FYTZJ1TrM%Dz%=DS!g3-Ema%uhBQbAyYp&sPvVo?sFVzs0Chy
zsSBg^-2x&Mhe7uQe+z|N%si?vynao)0P|G~H+ea~>_Kv<oc?FI<iUHH5OVPsD}$0b
ze8Q?c6$4J1ff_vZHpysjf6N%cPK5vSUg)6Jko<o9ICHpF7QFwI@8LISp{#5uW1@JW
zSZ52sf|^q1yIeBw0U)Bt)rj60CX$2+j)7dMP};s|4zi`L04%X6p$O?w`${tyBjm_H
z636|*nzFE1wH?_{`bvPT7}iHbm`+-9!sS@v^CEAF%YG+q2!Hj|2<p!H1kM&MCVCOw
zeuof|Nq$$#y)wP~PTK|ZQ`L7aH<=Y2JmdKqyB*ad&9G5JyoHT51+=`wFlf%_N+P9^
zdaDH}TaVrN)BoqKU_!0n3t!1vunNmJetoq=#smdWUAvbnVvc4oO|N0BNgl)CaYb1R
zlF|AwK;#P4d0;199jgW<YD<FGXqF&fhG18i+sEEJlmX?}B&C)<8Q0iMpL%@<;Pb21
zs&qa0ctyzi-m?%CM(LOm!nhITNYG&q2~D<<X<HzzcogThyRE#wwa;YvDo0ViM4H)G
zIpp9!V05>CqrOm)o3Y++(fiNS9@DSvB@`jsZ+eAX*;=eOxSG<D(E)Mva#B#;8`szZ
zKh5%iOJ>oRP)VV0xrd8r<=x}}Be^<-LKO%JEG8c-UdV6?oE-a+?Y4lvZhTWv*j@N=
z6JSM0>e7}kp<$RKi6DMouDB`CQ<hjKN7n5Eee>J|Y(vr$GYBM$8%h6yMd6WywQATF
zBMT6}H&WCEG9$Y)Xvw<_#^vsb>&)wurL7;4M9iBAI4P0Kyu!>X#f4Rq@j2d*r#Dq3
zh_Jkk)hpzoyK5X*74ThWaHeit`uEL2F^YW|%M4J)IP-PHNIw)6%ZyEH&|^za?jUm`
z?Ln>}`dx61M}i{g-&O&D+hR|B?R8)+qerIs7Cdi|XqlL|TsI#IVFM_!JTS@IWR7v5
zB^(%vP^6>zseqe*M#_J6lEJ%v-5cImGX}TbKd_C>{Rh~6T1iSPfN$?aj*B+fD%0#i
z!UnwVx)9RK`bco3;jTAd?xu$mGl{6IsBn?FaqLn}Y6Ic$XC6t<t8bfpTPcXu0Wh?_
zJ+7L{36kVz)c!6f&zb9PYhCxG7iwMdnLSlBUApSxNEx(;mT#Mt%M(}4voy9F4mAJO
z*X)>}ad7g3=f<-ZD7&6`)sY_Sq(#CbV*OvxN^X_%sK~eAVrAbmU&RDqPl<fOWp&G%
z><R0UT*ZN2#36oz*)y@SqQ^-5d2B=6S%ge+e4z68b$*gLY$Rt{9ey&)n3%G%qN)Mz
zo&-te#ev5hQH*v>n)HbP*+*1jC5gB}X;?lwp9IJr8?`pz?(67*nS~!KUrI?R5WTRD
zs*^erA<_%&3cTK)=nT`4Xjc@3c?=0YOQ*CP+8F9_Iyw<QV81z^_Z{FKdUX-C_p&s^
zSd!jYI#muy@5b_ia-A?nY62%Kbda&-{@!G}FAPn!GiY-NFNuk=_P-jVoe0W6Vq5-_
z_xJ=IXA?_eQv7CWd|t*0#zx+Nk9N?Ay*U}!ElCPldjwP<zDa}pK8IV*G{L^(=6G@u
zq>_O3P-H`TShiQX50pyM((E=%#f-Oul-V;15h>pAFJDf#$WakR<F04)uJ#9~;e^a4
zkd+XcNM2YkB#IEpTWm{W!B0|CR9eZQWiQL|^z=EGQajsH6_aMfA#OAru@Ck0*(QFa
z8%T7!y>_`Ex^Q)sB+rO-w9j$)Q29w%B+KroiT3|AqP1vHtls7~zk&5q?!$?2u2SBG
z(B1a^>YiK4iD$0pe(*eL3^mN;z4!!KTz-1~yio8f$PVK{pu03wunLdkv?N&b{-ELx
zI&QkhmdKJhwUt(`Fy@0n+(wGar3sB>54&v{Br0IrUp|KAEQ0PQQc@+1q3o+l`o#L1
zR(j8QYYdCxMB<vFr_EttZGb<l7RqW0v82NG-v*u^mgp1aF))C~0k;PE$JdTC`Me<R
zE8666LG<@aGab*O-_#vUX9p<Ld0BM9ONBAyS1ih;RzgUrL3W6f@yBvx5|oqC=KYeF
z!q4=&rf4XMhs)n>NWYe3vS+YRZ3&FnQJ0w@xxFpzsf3dqe#oYaB{j<n)PJ7}im%JU
z#zo5R`e<K4KN1x(v#eT&RpI<7mybjH|5mr)BmJ5v{8jTrN8-Hhaw%01Mdt9m@{Fi-
z0tu@hXkDh;Bn~ZLJk9pgK={YpWrsYfE=IIIbW@S_fUMM*z4kS9cz}hBVeC@L@bi>6
zSfYW`T9TWIgxM8&r|7W7U$}T3Bwzfn;M!+j@H&<*J$+MHl+CW}BWTmyPzM9^TcI;F
zNAY*xoWrnn#~b&k`E9b$v2GR`GkMHw;liS)OYDCNoWJ$pGH`kkheB4z5?cXB)|zO9
zHl_KiLDx`6njAM^fIfUrZ9P|>I!CkkjLl}C{o{+*wxW__{Cw0E{=ND`J5>$`FnZ#v
z9wm<X1&0NTr*z|)w91`F8frAjBYB)SX=9b>a|83VSK*8!m;Nv>UZksZ7^9KO-;CpK
z@4-+WWJqw?b0a&IBw!ZUm{T~Xc~QH*Ct%#7X-vlNe7Yr9ylNj>Yu*+Pyo`5xYdKH<
zuekqrf1c{cUx*^iC3%Y|SQKX-ug?lGAA_NsICwwj&MBdV37|;Vf`qQ}vKC{SQcu{5
zx<&y!a{waB4k8u406Ix?G1DdB6h>nmbj=_|(PlBK!(JSa&%aT~-$x@$Sdr{X&Wg8u
zx!r2KES34QE<;?G{1iq!dpkQ8mB3Oe9haQ^J+B{zZ>wF%tNS%Z8N;Ug28E$FgY;zG
zB8?9n8n{aeIyN*a>Wgf?{JF}Zf4UMmohCDQ9rS$#$K=Kx3!Ky8tM6n*|CpCq@XXCM
z{YQuW>xjvJ;U`TR3ZbmL?Ez~W{~A~>)LX>2KV1Gi?2N5OFPek2X58P_^?raQYOXCh
zgZMDu+$0BPm=Js}C-EdY4HD4MOs`A>SFJJFdv}|^Sq@na>?}+Z%IoJJq5oc*CJv+x
zfu8_uuNaVFU(Rp3$-<(ZUIUh*=HYlPVsq4O6&qhP0(APvkK`mI+NxFB4106s4wlK2
z0~Jq3k&8{5mN`sfrYm}Q^TUtco=#`nO}6N~<|XX+HzD{u4(fB2nkDc{m@+mczJfaQ
z;VYPX4usi(*^f78v9%w26?LL5Kdbmuqs$iVCB|o64DaAa`IPE8vOkh;*;bN8?#)ek
z3uj|vW0O62b`gajKaw)GPs#b{bt<Ed6sbr+*L-lej``K(dz2*%l^XY*wBW&RJD_NM
zxTV;4v{9+lcKAOUY&CrT8>qg}!9j4`8UtBu@J*2=-wob$oY|y1S}jEekLO$qt<+pt
z<TtllhoBQH#RCvA7A-<(U%F)6b7R9EQ#}PK;uc7~VKr-JySra+F|E&do5$+rQIb2-
ziecailp85mmDIb|gqHj=P%$310!WK%m69kPWBajx<#_AVvo_^G%?`a(?Ol`kN}c$Q
zb1;zBn^)+OVFbQ>ohJF5Zw!}{O@_Tvtx=+F+821NlFIXP4Zc+0wh%`XhiA)G@a8Lx
z)+%XcEn7?HinkW4?_v(y6lHf^{`~CO;cLb$S1<Z`d~q>7EGm=U(n!E{Fci<8&70hv
z>)aMq`P|G<n>#vdIy;-4&?V8!J*mv;h)|n63wML&*QUGo{ow{G_#Go1f>tNz*92^H
z4Og4)YA0yqI#)}sU6zMaY^_M!@8_HJdkGKU%lzEYp;^h6O+5S`3!ul?zTw4xPLzCP
zD3eK-NxzTl3k9K}Qq*h=G{cA75Fcl<A_iIOt3HN737`(@&>V<3^zJNH`75+yrx%J{
zcS6=$p5uFH^cQEuRf5<`!H?jfKdT>?TN^$z)zs!oE32Uz&V#_s;aggomz>Fp)y;b&
z8JyFz#R{ZXE~b@KQ`s}N{LC@|#XkC~?I=(qT&S#&g(lN`o6S3O73q+Pg5@f38IXgG
z&gd7b6I)MrD^J3)VjVsJ!Dcb43#Jezz~$sY=sPhXA|f&fG$!h<$J;U_iog{MMgO@t
z^G4gwo5Bn=ic~-*tE=p*5sQ8so1NI_aG!9rmHqtAc%bcQfbq;HA&c=Mk^U}v_=5d0
z{ZFE0iyfzjgdGlP)j1)Vua#8FmdD;jC$q&m$6z9NwDiLcmc?vi$32H*@D%gg*y$qK
zA+hWd-YVIFWs4~GVau0W38~{t#P0qN<YK=_&<GO5upT=1ntMdW20vW|i=T!3luUN1
zsF$z^p==G`^S9V7Mu=ZN@ZoWZc;EE?KzHm^+BgFLLxKOidJTL&8Wf2z-|^SyTqI5o
zA;+)YA;ivpmXPNCxw*@I#q#`v&%N7~8KB!NhJSju1Ho$4mFoQcczU9S$u;_BvYZrF
z>oJYx@Oz0**ZBaDm86s&1?l>DuU7+Jb-=PP$jId-J6d^+nMFyAy1O5pGCY9H?h>Z`
zZ@2m^9*^^hIWKdRBI(pz`V+fRA+)}xrzne2dySTQ7$(}u%Q(k1Td{nNxQsY|V{n&%
z%@|dGeo<HCynlc!#$N6Sy2o+<lM`|dwAI>=qTl!D?y$*Rc4s8gUv0SV(z&Vu=Zc&u
zSS1|zjFIi*%tUV=-G$0*a?a7Km|NEs(jvKx8I#C#UaFiSwTn5LFvnAD#bS!<rJ}D+
zH@~%LCAtNJwr)?^ht`DPwnAHX#>7O2$~rs-7-q*8U1t&jDrU!sMzxf#;M-7hv}>hZ
zSj>0uwJ<acW5)dX!J)Hs_S5-v_LK+B{&&KqKnA;Kk{7Q*_>!X@>DAi&id`NCt%9O0
zAxE`dtht`+!d7L5+QSc5FhHzo(I$E#?LVtEP=uX6-3)Xvq5krDHliG#*XxH4TZ`v{
z@+t>gT{+*-kPshS3e`9E1&dsp-Jb=2b-34zMAiG_a>hMPY_rBEKWopUiJ3h<@}$;X
z_rB_I>XVgjM#g0ohw}qle>(Snb!cTjH+eR0T0}LPJSG#x{AnL*Xi*9RlQ>{7-i_6a
zN+Fh1qFC{{t5~R*r<6`v{RI}h)GfJfMBcWeyv2BtjP{EjERM~Fte?~MzL@>2ZD(iQ
zx)H^lFP+M~ctNKaxg2p1fGw^2z`!v7;M|!`)CSn+EGie+^+56RNcW+(&H^9+p=eH=
z=isR7KWr^O?p$v-8BUtrc24V9M-y4(Ng$0e-frelAdPz6*9}HQ3DxfsuNGtHqG<pb
z1gmc$|6w?uG8)2=@c0k`-75~(7Cq{Amk0#Fa#en&9uD8?B7G|7NaqPrBl=WDz>dI<
z#)5VK4)rY!Brb>zZNcR!Y_a$uD1C}EK=3veAp9f%cX10yq0{-(?i*auj3D~<i_9Fg
zvj}K74ZnW>7J}rmaH_nNgoOMHB=YXgJG-Cu7_}Z3$ykbH=YF2hu!+Rx&YQ3Fcze1G
zA7*l&`#erO5t{ih?#)%kRcOr!RcgvmD;;tzuPErr#3fn(lpuCiswKQJxOJ`lv+YYr
z<5AR4c)cAHLOb}p31gnz&xg}^Gmb|w@8mm|{OC+J=11m%n;CWO;RM^xIFrv<I~TYO
zHt|1xqy+lEy2UGdNA%4?dB5vMn5Ed_=ZMOuXN>t^4e0JlN9xH|LEnf8$V*)K=zhIc
zBue1a^N_wiHdHF)Yh}2)qHkKtrHuKoM$C5r@xg>#OchAI6h>Y10=50@PB>@>_S}n?
z8~>_J)X1T^uNdPRBrX5!;e8>=IXg*y_9VYhb7Xe2HowihpB9D(C15NK{_%U6K(|KH
z51$bj%vqJ%v9f$_Eo8>v1;zCTAVeHFBAO;**KELi=Y&i+C<0^30WeMncFpOQirFSW
z`t)u0ZqGt+8&f|b`ZWMUC-z0+?VS)leD@e__cVVg(gRpw7aVaXhL8oNFC)|j<Bq|q
zjUnJBbk07WWf44<!gwo+9l$&%!f^!7a8_dSK3Kwm65_V@qw3nVb`V&eNyQ|pflvGT
z)22<PYno_7eZ?BA-AXo@Aq*wnUdyyJ1}Ra^u<snHoiS>>QX)_(6+O2^y1(WNMtB>>
zb<EAFXxNi)nOgDiIUVwU6^%$O-zB)n9Vz))oS)`J^G=zB!?!By<95h$Uk9aaET}#g
zJn$_twO+F3R8zJiS$^w08(FlcSP7ea`0&xL3|WD_+LN#|7(d|Y7wyB-_OEP4UnsS$
z(%A-NL!>Rz!S@%}<xBP7FRu88#%rBYJoS#|;dbBevYQuQZGq@FqgK5|kr#?l`~?<1
z%L%tmmGRL+C3Dr&`sv9_r>4P{(%JHZG}mr8CcV!4R_WBn*2hIx)`j=ed!PS&IBdR<
za(ZmvZxvhx?8?b5zKh&>-oAU@S-N7W4y%htaM1M?nEL2k^A=;pOLY|MfVyzzhl1<Q
zQ~`sD!Kz|i@}c*~U=9&0z)L9ksml0KC)y0g`ebQU<9n;eV9}3Y`ue}G97Ml}a{>HE
z8Mb*!bXv+AoJX09S8c*<)Kx}|DX=H0%EsS%S>AT{AAzkg(|eLe#o6CIgdP;q=#ojK
zOqCejWzS^aDFP_Sd`}qqg{dNMsO0x;#lX=5f2>xF8<r6A7VUPo8YzoDZApbS1~V;G
z&W?(u6z%I30Gkw9J?BtD6?KhuP;dwe3~csBNo<sj7Ls}4C7O6DXd4_7(iZ9rr7Ft%
zaIQ2D3@#<?iKY|Fa7tqwIC=sibUKphA6@nertWp7;3o0EKnu1e^kyxzn-6kmpc*GL
zvp6SHAK2s?UCXK<xhD_NtX9S=&<@@Z-IfsgCq7|AZ|;3B=YLTY3*kGIi9OhOp{1%Y
z<-&St4j0>C959BuAQ3~CqTa(x2>6Bh<Ld9$lcz-0cKJR%YL@BmmJ-&hCgz#w!-hd<
zVrV3sU`9#6Y#gfFAMwT=EWv0XaLunhR5qK66<t`lSimYcFw$|tJO7-@l>4QxEHyXQ
zs4K~VJwdJX&TD;dyG-VFQstf24Z>k!Oa^D7yEPP#7K_cPG-t6Ff#ddQk*1S`ot^B>
z%APmS`Gg{R0OIYz6K_aoSLrGX_VmhcBUe2C!_6!%^^Lawj!0+BGxx4pdc{I{hQ#3s
zxjQM?lBKBr1Zt5SlAzgrmERw4WNEv4unQ#U;wT54elk3LhqudSqSXTCGjF~%r4eZ+
z3YL_i^Zf-S3_pxq2_r?Lk<QGR6KVbTiFe5HgfP=-n}LqZ(@XUhO0Ha_yK#3(IC_9-
z_EL7{nLtX5C<g`2ZvK7aH5+GDl7OFkf$72g3=6S<QWfn2-(N*2Q9ALbm9?AVX-%UP
z<$3Dp&<_;E-3S{Q`MdcpgyRb3tPx)%Mv@t`qZG$G!I>ai2~wVrb#hza<dREJ_=>;L
z!fM2GX|(%?^^DcI21x<kHy1wty@(0lMrT5xJ>`Z(UykSmBD0TBgy|93H^Sb#ND%Jw
zbi2oBU0{lcu{OQIS9j3jq}ienjEPz&N7~Gu>WZ7D#?0`}RD>hRhnb<TeAb-FIp;P$
zmilHQE*Hx5ejj6-lc!6`tYbv835KaI=3>f)>4iy5FwM^xp;koH&0ynq12Q4U6Ki5&
z*(B>;`uzJ5DYsVKwzy+K2W>4UGTx`UB-i}7e)_mfJrh=Stj?*oTsY(ow^q$P)Jh8(
zyRh$bSwl*@I1_d(m@<pxEl{0fTcVnq)vc3;WGnd$aOhvW#G2f@4|DC6x|5Y7+pY5#
zc+osqaWY}*2IYsQ`cB^TgLk@9cC)H7h6?AKtvIBO$Cz$<o7h%%s;P>1fyJ2eJ~5SU
zAIMeIRW2A=wzBj3jfa}fHcybOXQe1DatDNxFvz{&QeM`|ASp^Cjb6p2de>W<(L?t|
z$IP(8GkOF%vBH~*Ep3c>YGD=grQ`~2QhjefZ{4dzXVbC_XVetrWjbQ=vypxg&kJW5
z&T(`U8cND+GPS%2Gdz>cbO2?s;L*9ew}qS`o$g}smuysv=J)I=W7mmyFpD`Px(WvF
zfg4t`(U1nwEm+7dBbU{5Q98~$;==84o#CC$5=RBT-Q3o7LLif};y2+9mhlgHgRi$z
zylE=56;+n+h;+_JsZ(6jc=a8Idh+c#9p*N0nFQE!mEudT5>WeMrLAzgd4}qTdHZZr
zN^raB<{|=!XU+*kLW=k+PljE?r|V~1_~`e}3H5sV34OHzaJS(z5}Q4zG?Wc~5aJu=
z^wiFJ-q2K}lSFPmNp51!#KsZMv*~IBA+SJnT;w@Vf$4l5?GGcC?>+Woc(fgA-6QZW
z(__ei=;`kFc@yvT*`roH|70*p>}FKa%7~{5wvfL|)ngF0IZ)lgQ@^_u4nZ%5B^2#F
zMLW1!vMbRj&+QGj^c932Ii{|xr)x<>uPF+IxgMO?`6#iac*A<}@_3Y((Yr~v@4it=
z%_P{a|5s40)ABn`myhW^eFY!tOc&u?w<!xAXp`v^Yl8%5^?79nnttS2nl04tec@7@
znhz6VMdm+ESpO19k$Utj-IG=g8T#Ib$hloQb%Y6*`w|(L0>y}a_Xes)W05fuz(Qp`
zhpN9zI5X_yBPBPm`+7n_!{oeR%D3$e=FS(w$aSNlkLuiXsGO{K$q#cY<%@q8zXM6|
zOAg+9#&?-~;|;qlDI@$83wd#j{4D*(jC1PJGiyzSib*ZVp9UJtCk5}w(j<HliTSRl
z1r`|!9vs<S41%tnlf5K=)_=w;VSyI=9o{Q|kF+!5|KsW{!>ao8?{7t<L>g&0)S(+h
zy5Z0*Atl{NH%NDPcb9Z`cQ?`@CEayzX1+7?``^#KE_I*1Kk;5`z1AhBr)Zx~3n>cJ
zC<-0XD1m-NlN7y8I52oLp(8<0#UJ*dSR{-5&`%obdS+lpP2_~-0W2Oar5_eakYKjt
z^nvs3J%QiXnbi4P_0LwT8e)6wd%c_#Y>l9OmbDKS=k(j^hyo(wolc=QBuhdEXI`wh
z-%G=53Fi^U%oX`c$hJJv)MoWu)0Id&B7%u>_cBiuKK@Wl!oW%>PD!`9@BWEVMxJj0
z4yHneHi9n5UFwr}%#>)BXSMnjOO>cdQG)zcJ#2j)<>nn&=67r-b;n0!>JjWjw&crM
zRp%OKJ$Vicg3)^j($&vCXMC2aOGphf!ISFyihZpUb<7P`EhweaWQO>^(MIjRL2C?7
z8e(+Im5$}X>et?nRk}LGHGvm<V^Oewn>1HF52c;kT(Q*7meCd)-AgW=Hkk<I3Pf`5
znN4r|?_DUQZxtLL(w^O~*w)lu2s>-+*9PG@aH2BF+2%!t-qy7C*+nLJO+pL3NhDh*
z*WEeb{~2fl9op4$3FOlRdpJpcF)51szWR!X0Ru{}@hwDAKvVecCj{t(y#n&D=2BLd
zV`R@dNgW(gKcI^#+p(U|RCtG&O=l<<TU{NqNl8f=kbd@-K5TTCpwCw0{d;d{CYGyT
zqB-CRqcYfpUrt<rm?&5o`)eZUQO7CD!ajD}gixkJCt9NkpmU4fp@BIW1T;X((BT_t
zQL#EAU0)EsOBsv^#aT6eZqH4Hf9<zuE~}0<4RF5<5*Jb^m{eJ%Y2Zk-rrwS$ETY7k
z_?%t3p4NNE0eiv<Zm9I%c=EPS+Q-vVSrv+uKG>|<#^~$8A6`JQ`EVvg?9C{TBrZHH
zfwSk=K&-(q*#L}*3vVvya|mp;G^v94hU#UTt$z^0QCV!wc`@z^eh5;4IVOL~I{`<3
zu@Hip%(p2Mjv<GOoYrLsQZ)QSPjr<>_Pk?ge1Wruz2PJt5gCj7Wovv_lP+5JJz6hq
z*DD6*aK2*Rhu#GqyrMJ$Eg$igxSB2zT@dYVqv$8<-#?1NG=7M7e<nKJ(NTM2SNv~5
zK;78aTkC3MX)@$Dmb@^r#&?Nn-(lXT`~}#T#ugZSo4}EM-{#<lrDYYM)JeSzd`uqs
z#DoHjXA8q~T>!=Z@KDePLXT~zJYg&V!!vtt%%{YNZ-HxMIbrWvKL6eOq4lbq)x9vz
z&T?5Ho>o1R2<_(gnv#~slXQ`GPBgE3<Kgh(iZ5!Ek^XZ=xBhZgZ}37|f?(q?<SC_D
zG6<@^(8Z*k{L#%_|7=YxZ@lJX@Q-^+Mdc_rul_voq7$+Sb^%*#JS=j@uP1Z{A@gi<
zPPFdZGlO~xJNHuxS(|id*7O<<CriXutuK`tDsA319_w0ZO*wK=*;US4E_Y5d2~`q#
zSc#`Ql>29!G&V+0&3VdOuKK-w#|o;m8)HaFmvLq`6p1P4t{u53He*8f04;4ES>mEY
z_d4PPi_57E7EoSxP%4D1^}MB^6|}Hkg?^T77=r6F?|x3=vMTfalq$W4n;;}xDwY#w
z13bwzE?Dv<EB1d4Gx-fk`G?*bxx5?eXF+;lpW@I6#&Ei3pyw5x13H7aZs?Ruin8;Z
zqD&Qz^T((q9@@+4())Am@uqxV@XMi%%c2iRml*qOjf)gWHL52MG>dtc7+Dsje$No5
zXlRe#f$U^HW-fXb*)r0zk<HB*j0`!3aL|0RpaMd&7x{Ln)t*#k1)qqfMMYDmYvN|2
zt%>)3#^TAZxx*aPOP4q#nZPl=d%NEy8@UK$$b?<WWY+yNT@439#13_*aT|rV;T4cW
z*Esg;%rGAg3eo{*C=D?jMTF0XpYhW<_ljhGkX6+e!j<CChdLjmMLMnuCnA;UD}0iF
z5!w!u=G?X9)^~QQM2<$_=&2vPrMJPa7J2Pvd8cr+BY$X1d%d0N!%KYW<L+59gVrc$
zJ9IlPMqO|MCylg*$Em=3YgePNm!U*Psg!<j*Rrir&A#>$+QcjI)iN!2U;}M7IAZGl
z{qTOE5VL_Ff{g8RgC$p)C*yw0Q<L}k6*`;!s&FZ}j2KpAg`=H3Z_Hf{EJ^YZf$J{C
za!m!Z{L^lUb(33Pi?+d71x-!gv>56b^wx*C0;4mDJNNF4rZJ;#UNlU5<qEzBg*9Ph
zV+Ww`VA8s3&#9q@vTsNTJ&evbeGLjaHHq1`ldmx<alb1oEx->m<PFl(O2p%56c|KA
z;Re3ZJx__83_=@-Yh1~tT#T%MDSy<dt8L@v%*hueKMYSF-G8fkpl_u7rXjeP8PQ_6
zo-`4>jM4|nCn;ImC_g~>;Bs#Y4t;#c5Q5vilVQPE1xX~6adD)@J}EQ<zr6tK07dSH
z16n;E4jYJ?#FK5bkdbD=xUR&W%zjIHzrYusikRbEsP;)Lvb;8m8OY=OMQV^Fi`6hp
z(IB@uOP&EMpYUCeFk+|q1f26ncnPl}(7VrYUei{EKo#06;kk@}Oq&=P>8MlO&eY?l
zjKbhLiv4GKI3CXl{D4+=FX;)w%vxdTv9rX`^Iw7?#^FBOIBR4hXrMyj&$;g9Ua+wC
zOtyb!ielQxVA73IxEziO>nkz)wN~Ng`-LKWi^$OY7<b3w#Lf!?a~W=qic6G>)iuq2
z1Kc3x2ad~ku-SE{{rdK!L7vYZrY}*7T)9VV?=RoYr&82PUWLY^-3jWO)-O%j^ayfb
z?w7CzwJ}>b!Zz05B2p#U29W5#j7^X&(uYwJzd_btpjz*Kz8HeV3oCCAjyDa|kN{U1
zv)lk4f;Q>~q(-+6!WU@PhTTrfvEGnAz`z0OXf>DW7;OvHfPvjh_*Aw#ctW$LymV}!
zP<m@&(|`O9&5)h>T)2+zUn<>?I1+wnMHa6U=b%u`0r6p9x35VQ*?GRPE(mQ-4WE*%
zvy7Y#E%nBNy}MMU3}za!Bc{tQlCi~@%0-i6(#X%?IJmS2J#$aMP@uX+IB-^CINA?i
z-y7q1Nt67H9QFwL*>Jl#aO$FQSn?0=v&e<3x5?mc(ZBbHe?_G;u<-~Q(`yiCBog5Y
za%a<z(9gI;t(HH?e(S4RCB0w(N8-M*063aWbfG7ERgjup#w2l|FM*l-oIAT`DvL<u
z%7C8h&X|j%_|2Z{Fu-uj^UTm%@{G7Q-tdmaDV1%h=%KV7=STL2q$N6y2oxoGWN(+;
zGu4P#?z7uh9)%VXxA9stTfGR`*m_+}-DQns$GV6k_|_lmMGUXMv=ZN}YDvemFMOE%
z%JJrCiPel&1k>6Gd=ylWfu8XpIRP2wV<b!%5u{49x>_XVdg|uR=+e<U{#SZ3Eu+jn
z_W&d5wrfZ1f)sY59$-w2RD*+t_8lUySKzBoQ+xXLZ=-|PQQ#Y1d%#U6?vUolkzc7s
z1^rpi{6IQd?Mq_jmsPtmoLMYM6c&U*T4UI0dGt#7E|Fv?sV_;yg5=@O6yXE6j{=3<
z-5Y4k2~TkLr{8+)Fr+hy5x$J!5@){0rN%d4Viy5f8mIjE;JsMF5#LrB9*=NLlq_Y7
ze>EISl@rnFg<GxQ_7Qb}eo0vHiR6}!#F%kEW1r;#6xJ8pl^XF>BdcAmnB23-Vl=Qq
zWF+|v<wqtrii=693Q0*I<Z=-vY$*sIQU9ehgjGS&98Tu@d(VkXE2$oHL|GlRa+tk>
zR;?x((d|!7=(je_U2__8FKVo<rkwf#+nCS8>+$tul6)6fl$2ZG@kr<5m%Q$wf%W?P
zk`wk0&31t;aFz8FQSh5~TIu9^g@uM(gA|7<t8w}`Po{vjaQp)<@O(~z-QW9~R32OP
z-LdpM){(GSZDYY21{;M<#`5USgQPxGQc)*D)B69PsQXL`w*X>W{1XBuf2$6rI1|zw
z_mJWa%k=9e#$i#+_TG#fJ`!JRb7y;-$@^Ao0&*r`=F0#QFgg_{Ek;NE7}k0XOfPF4
zf{X(a8wC3%R<cHW8spYkjc>j%=*K-JVz0BqspTRICo}hqU!r?Tz)|dpN(E{!b+Mxb
znBD_od190x^jW8`n~mkY?^iTtGn*w84bs&eQu`CDVky~fHm}td6Q3>yts#{LaT!0N
zXz^jKSV-JJ{G)H1M2eH-{zIOn9|f~72&|D4VXC4(J0>*cdcQ_}CENmdkf?u*_>QrW
zYs-i{;s;9*T@gIBnFn#NZ1b8-XT_YCjL_xqKB<CV7`3-ff5gC4*-Rs)x%iy;$Ku1S
z(VZVDrfol*fCpRF<{&R^43sNq*NQOVLZYE_oo0`-`hptfsTiX>6*=V!mj5$fz1TxD
zb3#Tv31nTLRwMXr1WaiKE}u{6g<RkU1)KjQ^(<9ekMN#d_|!a|b|a_vE?VM2Ijk$?
zGE=px(gY@%Y~M}W@Ltz!f20D#2U)P}p^;|m<?RgYpgDK8v1-{LT_DYF-;=-#=%yO=
zvN}BOkz!QxkdNOuD;Li;NY`2XCJ|bF&S<%qWE_lp*h~$DUUkj}7~LzE#3sv?Aw6H9
zFp}y%MoBF=3_7Ox50|q@<|<T5O#PixA?rEOEz9AUqyO*g@uQYh2Yz$*`Q!58g>VO&
zu&nU2-ZHe8K(qg%p@cU)#XbuhrH{j3)2OYDNooa?(5Yf)EqOs-xthQvx;?$k+zJVw
zN{16FN4};xI+}<%TDnT%=td9<hjw02o>65blR67RCq+Qc=L-8`f>8leoJ{YDp59(l
zx@x@}k&-B;z3tXKS#jYJwK&)*l`C>VVDg7p$ft&|L+c^$-rW6PL)goEz`Xs);5K}X
zn+8WFhVoJ?U^`cLs6qgh7^@YGZ!;;}NOJ#YRv4+V9BL@F`bRncR~~57TLz8`%8`jt
zkf%1zQhpU{p`7rJki?h0Uw&Nrw_V4awBZ~$`BlbmCJ~Y;^AZD1%XFBxBlU_R;B6$l
z7x%k<uZioMH-H0X>>CBQ?esH8ey2wcf%3L2{%7R`?2lMbYi<|P{u(HgN(^jBgOTW(
z*{I;@pdzoXt(hXlAk!^xk?qMD1d|Ram)aRZTeg{M=rhyXPsePz2woE)KB_oAVbq+E
znZP1WUQ+5TDz{@Vg!4<ydD9mmzPPQ^W#xz5es(jK>^sT2&NwUXDh)G=w79LUC;jfu
z-wEg~vx}+!2#Eg-=Ro|pm7$Ro=}v<E6$62FQ_C=CjBqOuP(P4dKE%cmzGpuNa1954
z?hPLlhS^u0gknY|XC;ds0ef{EzzWY*>OCAX$55<zkT4HmMt`d<G9ZBo7X?WFlCjKp
zEgIG${gNxuw$uutfhPowJ^e3J+vqRdpV=fNf!W4(B~!bk+|P>AHBV7Oe4veER-LBT
z7Bjkkn1|$<`iG3a89U$$q?LWof#bZLEy;kZu6~^w7^~6Q7R^Vto^A7ge(+CTHN^Zq
zmwBEMF$;+o3QR{HMe5gCHEfY#rs(DqII5=-c4_?KK<CxG0G9P4aUv<zbYQG3NR*7u
zU87Ayx>!43*K5YAsB*Iz&57JmrbIvrDXBz<&oBt?x|hh~Hj7KdzmZkyEugEut4`V_
ze+~w&3h{u5rg^S<fCi#qvu9XL6nj6y`1;!t_|4`(th|M4m>0=zh=?dPD!B!sWurhK
z5|TssE5@YrE2qUegX(nWE)Cnu0wII>@MCedqMP1wc@v&{g6PCXBl6w;YSX989t(wi
zKKfBqWnbllScc?p9)Fwzm_2(hXT`>Z`-Uq%&<tIU;^iG%{nod-_0k_tTlIDdN)Vm5
zmjjnq1|<iEb1~RU@D$!-G8DvC>~65gtLZzX6yiDj?=w+(57i<|LF=I|%J2uY{H#U-
zm=mJ+X|!V^jM$aYXEtyg!d~5dMZ*BkK8v)^6EFmD+HiEj1o6lNDCMjXsz~y4`Ixa#
zjaiu@AdNuC60c~|9mwwq#DN+^g2-e+LQ`?_qMR%w!+69V@3(k-TV}~6VkfA^PLJIH
z1f3LH5|f_)n9pw@1VW|RR1+@M4rFbd&w8=^E(Es++AZ`ue!gvzQB_S0k3FGfu^)+M
zB5=P@vBp!q8F>f4$mM!n3LqIo#O?rBz@fdcyKT!}w33xh=;Qe?1E<lKhlbDDe}(wN
zUYR0BxD6Zrc&x-L00MDLc5ET$(rR5uqUi_}4>Am}q9Snb(MnE=0kZ7~d~zg-(d`pd
zuS{qv`qYQ1s_?KIAa-)pX05;(UxUcddKq1+9{1%&N1{Ks4$*r75DE09-#B5vI<8*>
zs<wQ*r~HC~cP+%v3hE~4_2w&}*`b7ZN7xqno7p0IZTGWA;S$&tE~_?h3VZo{r?%6?
zmtM@gM`)w;$z-83<<Q88dm<L)ywN~u(C*m1c$pB|2Va<>OouyiUj26FlaZ|C)3^aT
z>Ft#)8)hB%`$1QQJ(dTX_q#Fd2hx$hS}=2ejG>u2=RBU2i%0V{6qaYx3k5bcod^qO
zuXF4=HUC=(`uFE=0R7eSL9~uY9sJxTAOXO{P7b#A@IVyZH-ifzvbk<Qa3u!$q)n>|
z(p@A9cn()vo0S0YdrSt9Blv@%kHU3L`f5se?KcE+QKZ}b&)YGwXuWBmrv-r2Lj2l5
zdc2_sSv4mG&_@^~t1-+UnZ@W0NN=BTc7y?R)n(5UfQ3*}x}J6+@dT97ChEOspjY9w
zMWD)PoKU)n5#DdYW)<kt6W{CqeEivBNok^sDLoHg@vAMNV!y5U!)rM^xYGUp(%KAJ
zydMT8SCQ0x*>;?#0KQ`H%rJ!$yvDxu=av8V&8ju8!U6#Ma7M7&pE=@lIh5k_xF|h8
zKUk#&ht*g`_PzEtXukjo=liM@f*!P<qf<J~KY9uDns)R;h8~HX8=^_<R_S9V4SM}Z
znunGBgwx1g0KQMjDrS0On+X5QkIXTg+gFPOvmq~lu*saaUNHFNr<MhT<5W&0|01&!
zi2oJ0M22T8z5!(JeOnG9vXOJSdcd>d0uH!ue|c_h`<~QET%ZfM6aOWEFAcn=_)`M8
z1HQnJj)*d$i)uS3Vt3}Jq*wu&X_tik04nz3!5E4Uw?Fh)rn^g{P^D!0J+CV<fV1bE
z<!_3xWY!pDm{zJzV&dX<j{|`k<pbq*wI%Kaxv`SACUTi5{=BmbjCmY`Itevg{kc_<
z=#%_AG*wms^usMYx+quEp{&tbWrT3wqy|>jgyLe3cjp6&v#Zg2W}eycK4f1&^$)JB
zA5kE%(U42!|LbARCG=V}gCYNRmw{e{935>=8tx4=L%-5LE(zg&0~-q)ye?E4@)jRU
zJ}#i5fxqnjaNEKT!SO=msMV0_U{J4tVCBI3_Ar%qRYX`o3>9nuV!iP>J1|RpWdvnD
zrSQ1cHB;}rD&xzM-@u#loS;Ns&N$px10_O46sgz{kj5+PFV3e6%x;u%|8?KZ8;DEP
zH-y3>MA-?OF$^@fX%cu3UkP!iStQ-(OUECP{vl1~hb9t(o>uqfnnLzru^;8Y?&oBh
zgLS?ml0<2J{9xmNjLA;z>S0F~#GsfT`fSZ?);Su%7}67pR80o~oHytSq=oi|tgrr!
zvfaJ@&_)XVj~lx}qsW7L`MrhH_E}rV)k3noce&ug5s)<tBi;oN2Hb~z11@<rHObh>
za#V9i%2>*wFCr}eFszuMu8P3J3Zqz(Ayj(h8y`<OW=m&1b;IC`eOJ1S>;amEhXD=;
z>!iK|n7q@?8&<YRA+ILsJs4AFgBSxlvnJ@yhwO30gu~%7uqhupY-8dEX0SCZWx%SP
zJ~rTQNl8!+@3nW|#A3MG4LoGoLRrG(Y;((vtf2=NzR@Ad1u84dZm<#kecs_MU(Cpc
z@mL(@u%}E~LXHtjK%t??%Q44)Eq#%3aKc4)3f5;#gXz&#)#5!+kn5iq2-xVlfB#U8
zb6<oq?vEJ(Sf;5e14+&I33yVBVNQg@4I_=nYaDu_oJM2#Jo^zak=CSJRC)@gODvM5
zwG*k|H`M{Okwy4EnY%x?d<K&`5FF&V%qBOipv1-Q7oY~5EQby$*~opdpAx8{{RCcb
zBdGZm+gGdY+G?g$Zz}+JicG}?tZu-$_hAP?NZl=CxwJ&6t{-zwq7JQ8it4UF$D(Ze
zm-_1T)d8^QW0g3c1v@6x<N6S#0t6a-RFl-nzq8t8Z^onN6~^K{htk#C{WiK65y5-#
zjp;K$s@^9*wpIiIjI`+khncNF`GUib{X|RrAZy@&C}^W-dl5zQ?jr#^UVd&(5G??P
z<I`x!LLD3KFHIqT5RBbLDl1$R!$XlCAqQlM(qZDD6{p;-b#*6DrnQvJABx>yk!FS&
z6XoX<%=6)k@6{pG6iY$vk^Y-$5&n_|1x*sZRh2~SF1e}l8_`e7c0lCCJ@_`{5K`Kd
z`N8OGa5w&r5>7w$sJ>~Q{ldHRr#5FaLOkNP&Nm?^A8mJL>NqzK5xxrOk>rjRVx9{e
zN&u>&FwcU*vT5J!RtYB_Ce5S-i;Hx62MzA~&s*g~b^8Bte|wDxfF}5q9&HNG_?b6T
zy0zc?(xgh6^d0r#BHHXy-4gy?t{-nr334jnpOfQW0l<hZ?6`t`E^xb1s#f18{WI-Z
z^r%|9&tN|i`A+0Y3sJ4(8*vnyPVKH-Ylxk_sbulSDMgJXq6*@m&=c@|o=T58J^fKu
z+Hr#FNXxq)2cUCJFI2<DXb?C6=<><*@(Lyoy+v_2&62XK4GJR*Ex6(X3r{3Ljg9dF
zq01j)_jk-GUSF&zQ|-gBEeMxaY6=&ebx@9aSETaGZB3ST{F&MDc8pPp15p%M-GHQ9
z?%(3jBnp-XVa<Di_FEmI6AVJB!2OIx(dM8#XfMkj9r;B0?t4CCi}7k=IYXhyJrg5v
z-!Eh`8m@8KOpWo}Aq*bT!1}EyV2(z3B!)z!u!ly_<A445h<^QWdk2{q_9GW#V(UlC
z#8%2AStfrZKiwBJAqmKj$e3|bl<>uf(>Tae$mFlFbDt19hoONgrp8B(i0^MxrSg2T
z{p!xX8(6AovjiC27TswDQF}wA&8q9zug(9po4-<QPA#Di)}|EgPqh{FTqNzz(cV)h
zk0NR__X%46G(!>&w*K^0Lp71ucu<Zkle+`=W0g@~d@Rdf!4G|vlz4Iwl<v2k*@u8C
zI`beer!#2KXVT=og}g;SB4FZ^8Ayqgs;98W8Q^Tml{{6_Kr(66*ObcDc>m!O?F8A!
zFybr5Z+;j7et!E7*jwdVt?FhEr#U@P5&WNxDPqvZ^u~5HZc0C>^-&|Ym?T(p3?cfG
z&zZX+uSflNw<YB>tug*`4)}*uPDFA7IjHpE`2?jnQH5ZV_62AXqECGaEG$9zR-+CI
z{p%wkC|UdYDV9^hADc}{^aa$7#!`#KKY0Y_)3d`MV+zq%#)--G>lyvUU`>sk#)3wF
zdyqx7z2RRMfrls<Qq9tX%5@zJtuPQK;Te=D0;g}-ok%Bm5!Cvn+C0PkI>}+qaymb(
zM>uU>|LCcHHFgQJB0+7s03x)L@y9JNSb-4cnwGY+6`j{>oYE!{$&p9v(;d{s5Ct3T
zP(r;YNY4_F)?%0~mI?rY;3?-SdM}<nar0P*Nj<w~K;LRME`PNbLP`ydjX^!w{J&|I
zNHQpU8oMaL#-U;+<xTNf5F6^semO%lH<*WuSXXq}D4fCxVT%H)m)`>$56uWlAnGR@
z_8Af5JHJ!FKQ5ba23tl%_KIej>&Z77PyY*-i{GCZ8sx~T|7s-Uf`l}CeV~aX2Vb2n
z`mhT0U~`cq{>rvs(ZxVCOvyVd(od_yzWFZQm&?l=EwxKQei^~fd<(0)6fB9^GZpH$
z|87n{l&VM)2X}hg7gE+omXL_scM^FYT1&hcsoxiCY!PJJ=@fI=XY;X>F8J`RB*fMM
zg-8U3K7I9cGu`ocNm)#>E8%ZD`0>we6P$t%Ff-DH`|88kd;#1iP09!zejBg<0D-?#
z^7^5CmXaa1UBNEhf(>*AvvcIX8L?zFoq+^?=@sgAM*mWo&Sn-bgVibh?rihzJNO>Y
zNM=g}^ImH$|DQ~@!mcBQKS-^&bf!3f3Tby*77pVZI4{Ibvg|p4+(Zi;J*V6|=u;O<
z0gk950M=&cVng){w!km!ByWU;aoB(!$&wxM_d->dfjT&y$c<R3(cnA&N%YhppowL%
z2RBBi*>p8Q#h9w9f8Es1Y(P-!RN&TR9N-IcPerHIvf{4y9cSg>Yfz{REZ%4)(#6cr
z!cT()g(Nka99_u*33@623+xk;GRhxg^l3_U2-QTh6xhw(XskK>C^Ek@io#Xtfh>}+
z<p~Bg+p{W;yR24?lJU?>%0MM7wJ{~7jT(!XG8`9|!6Nf}ZC!FHujNFL2QL1h32=H3
z4XqJN4W$b}48+o4`)}d2{u-cm_~YOj+K2v__5>*2hAiC)C@gT`YL0zW@89dd8~yX*
zR~N_fEnj|trtJt{0g1Bk_$@8_1>)sJT*ha@Sau_esbpdA+)s`Kaavcy5PP<iU8=-U
z%cau-W@S1`z=Zn;&$9@jO0h+s>eXFot;xuD%}vj<%&8nOv;`SLq50%hD*WaH)JRWE
zHAv?ZehgVD#;Km+rzG*R8iZZ`L~7K?w;LRgHu`iHa5Ta9wMxO$k@~nf+r@ieZR>~=
z`6<2~3Jz4c4xaLG054T80r;7(#mHdn5B=Z0dbeBSj(qrSpeF(8gLf^f?JU)I)Q*5|
z6)4*%yJ$^xPiPh%-b>=XvRQ7KIRn6;5i+zcMJ|JJEUMir#MQZ~t2!J6IHVM@vO4PT
zvxC1F%`NoTsqM<;BdmE`<y|`0H6QWd<UNw2lVplet+n4vn5h=&Ynj~inl7?j5iGXv
zV>vthVxFe+%%Rn)7AKk2!R}ny$e;2fm+&hx6fL~LdzWVxft4#%l0Q{+$ilq-8<|{G
zgwgg|h$wN4k7I@{H8e1OdQUIcw=1BA7@*atdA&hbO$6E^bpA5zIiK+mcd{7XsZu1S
zs)*0p6B^Amdj2BoQ1`P<G`3U4UpMn+wOmpCrX9(OV+K8v{BNd=*HP#jU*a<FZ;>o>
z!s-$ASFOlz&G2^J*Tk!Mx|A^bb!9u}k!{5bc1NmZ92UR~A9*8JX`PkU6BX`+l<cZz
zesdy<MiB6jStPT?iaO3k4x;xTb{@5GROi0XFbv(S0nXX5lQ|PsnG8-UJ}JpSVJ@3L
z51gkFoq}hO!2}lmKEAKwBp4jDQ-+Zko~EHkqw-Q<I*r3{TF%Gr8&)-qiyOlI*@BFg
z2SdJN_WiAx$cpo%vE#gN{dF$uW=ms%<nr1##epULV~cj!I9>%7Te{GRGaciNKn!6?
z*Lx7<VuVyghG$2}$MbU=dwQ2U#NDv>Oemj#_4pM?F81x38|Ikw6JDg8_YY!WRe${K
zDE!ani74j;qhwF(Lcd@KvRPuAjhY>n|2r~V4d*`)ZCv%htn>WjMJxP<PHph9Qvr0T
zeX+bP<EbhFS0bO4iqB=E6pTR!c27T~5AnQHdtPl3?{nuB<aKiWd9+li$ysAg;#saq
zXN-zU#0K;@d!)08oFKjE)!{#RMEd$G$KE6mCX9Rh&T@Z?Ou$)~n)fbOe;@5{YS^Jt
z^FX3(z(I=u-_vQUQY-Jiv1Zez!6Sz)@7L=R&}XL4Q~L&)#>mnLZy|*H=BH+0gfjqX
zZpQ{?_kAg-FEUe8*gpM>^`?lr2psq;f-inQZH5x>tzODc)%3j{t7N@{`HB<zt}3Vu
z{jqA`0^7Rg>)GFrbK!Ka@=HhHhRNZ7!|KJ4Ub$jfi^tUFo^((uZ@g9nJc23xDRtv(
z)X2$BBG_o`$H7t$0{PFek!xpV^V-%A4&}>j-}1sCr;U&M0ZCIeB4(QHX5&~j{@oW<
zGBg(NZT7DGcAC)qc5YK6dhE>;+-h|rE>kK6)VZf*L~6p~y{WHY_9ngW<;BrwuQBYy
z<+muPY#XP86D_A7Sk93l{75%c(b(lS6zY|5`WR5TSpqEe4#(=9dl7FCB1Ip|nPB}I
z1fC&35Z_8+SgYDYRRwsWk|CoR*(jpw(@<qu^w6pWWTFWq+oF9L-iFFS90glEzmABo
z6OAmH4%Vxi_qVlhe*OKn-)dS~%Gsp6$xVW<F#f-!cR3l<9PF7yw$*srG3u$}QV$?b
z))1`@24Y()h|KBfp^=2?j^(c;#TF&E_Y^}6O&5yT-1g%7=y-WIKKEoqc+sO}gu;~f
z_;K^(P<Rxuf}#}a<O+N6f0K6JNe*<iM+{{=#EP%{?XGg_<c3E;il?{uCb4tCW|u-g
z`!<hN{@s)>FBMzfY6gyH!!;0tB;&s`|D@6u1S}N^AwQXxS;Z5=oQnnng5jxneNZ^#
z#o=$TqGI8#o&i+LeuviJ&Z)$K({)ndBmhO4%Gx4M4}4f{{ejFClvkel{OEAYuz&S{
zePg?s;ByH0GLp|-@&cqW|2HS<z!U>rhz6=WPMO03i&Fr98;ZePQD8O;<ioqZZ`3N2
zX@92b{A#RLg<;d3i%#&=hsAV~U{6{V59Uit4{OYqSC5x(XA{R)8fnv1CTpC#_|H3C
z4<Q%+S9X)WZ-4J(h+7SGGPn$EyldWx|3xB^a$bVyv_oF>H7DNn*JF2R(Nvw9-v00E
z6hdKb*WXuT`78!$HTbUKVi4h)S+F5O$4i!|v?*!)N4(Fm0FA?}EOT-*O0O7*F>p4r
zc>K)m7nc|i6S*gH`!I+e|7j!Dm?9{6Bzi{ZmXstH!P^b36oyG&_R9Yr9i50~oR(-G
z9ANmeoffiFq3AlAule7!^?LBD^nv+;U=CaU37V0O{HJs|$WwfRD$qBs^($`%qZ+~|
zdJ@2SYZ8HV59tOYn@oMeM<Y=#nU77)YRrA%B7$5c*Nbl%j^@e77@p7fU7;n@K+JH8
zjChCSWWZ7>bZ@DjAY+koI9zwGjd>dI&YPXh+e--b!B!hK(J!l(%v*cpXy1O7hCbSO
zha2B_2+?gfCOjPI;IWym5x18ur)KTbE8HV7m2laCC1hwmmSBePyP)#xHQ4M(6v<_I
zXNtH1TYs$@#o@Lrx!2i7cfd{<`|5<VmhJ9H(G%(!6IlUGE@~Med<2s$If-4-2i&{Z
zdpjVDB@W!R?J=RIBDrzBgBz*Z0&AuL*oX6R?Ytd{A=+1T)_z5Y46f(_joDb(+xy{N
z#fSeqH!oU|pFcHZUj4jz00nwXbFe&yA!?7bvmFNcYt}3S+~A)<YYXxR5}gZ>;o8aA
z^-2gTDClXUdN0nOG%LpA`I&Qya`#7K?ni>hIyc@&y+xPPTxMxolhJG@WtK^tloi_m
zT(|(TrXfon*TFPL_6byUrY+W8d&G2~NB5mxanFH@<?x`2;3jNzbQ%KFglLL`*x632
z0Z#}I%wS@R7Cb!sWn5SAEvK|~g)rX{$4fID${>dHD^|w7js6i80DL|P<36CUQ3)jz
zWhWZO`u3+Ms8?&ng4TL0M5$sfnCx97uDs_BW{S+YfRhS8C+a{bci`Dj87cRDX!5GP
z_%H3k;BS$)IFV!Jh6CS2xdNw-SchK#R_LYb+mg^&U$}d8u++r3yjvUJ-8MSGVv=5Y
z<1TIzue(dK4OPA!{kM2$xCQROq<}JuMUo$=P@Qkv6q@)0@#XFi^<w#yN=f`>s9c{u
z+Q@c4TnR14l%8eovg`z8qy_x3htq1I-rB+#Q6tUXG*}m69|n%QQ8ym<-;zr12YTXM
z4QBtwFz;5Ha`O9Mkgs#|mdpWvP+warcdACcy{EKmC7pr^s@iaL%EK3&#A28ghRY_Q
zA#TAW9}^Wl%>cPqGA9@MJPOGb_r+GXNsNLl*U>_)`gCO3x^9>C670B1HGl)#c-y8b
z&+!zRw|mD^hxhFP@^qap2r^Go(USdCx?F2k5VZO_tfQj@nL|I27^nk@26FWsVX!0X
zI48Jg!4ilYBJIY-BG*=Xs+ze0DWxS1qQj3pcldrz%$=g^^iyxI{!9uDC<AX1NJm<#
zdfd^#HF*@}RNkcu?arI5ekjeDHE0)&(0o<G-}6j4IGLMpDLq4yr-nloAO#v2r${8D
zWEVX=OyD@*Rs%$9X7!I!(#Yj2%ra4nn*Ag*p$Bo3x>}7i^+wk)&~1<Bk~qNVxS3`}
zG5rQ-If*zz<C_?a@*xQGJU+UrF?i?$J9|u2JzpA5L6ybp(xHOwCA8-tdEzWR+x)i%
z0jU6Lib`5n%B=?R<>~#M^}{XVA;G@Appa1QAh+Sgyq0SjQilO;C|qeT@4;`6hhwx)
zpTJVF#F2u~85WA3zz~h|eIz|(7um2#*jKAk*AF=R-Dcze=2l_i(X6*j)5+Ja)D`h6
zR3GGpQ$?RXy3kxqH`Ie3Kz8E(5HryI4w<IC-4cDi@Zwn#k<Yy*ctwLdqKB$_j)v|Q
zygQCVc_;W-VW)C*_g}#`v(^Zx%a-wlUit-$DJA#5^#)VaPFV?<ar|&_jh?lI*^ZLM
zWss_(E#n$|lAu?8tcU`7tIb}FZCW=;!}zW{yx!#jD3hjC{Hj+*=XKof5@S(BXy}Z=
z<7*-NXD;%3GeE}Zctad*3sel(j<au{0Pllx<)_$?kYt_ip4X__zfi;AR|u4kOOo--
zV6EAW5f9X0QeysdLi&|^oly;(1{+)?OQmpIDNBj(Gv*ip<#h8}zO?4Y$h7{l_(nYB
z<WZU?va<f7K;twPR6eQyjjJ1QQ3AsvqK+x%$X%%lNFoOA76SE$rxi{Cv!EHXVK|jq
zRj%5sq=p>ah=K1<=H)u_Z2^B!z8EMSfZas`+)J}+CGT1Obm9!r)2q)4ZEnueg3_Fz
z>NWaSthd1A_@=p%r+EnF^ky_`5O=r~&1t@&-wbj=-D?l^z~7F4-oE%vzoS$dEpx6_
zKv#LJzm<XnQ%&lqVE?0gwfjE+JYZdMK><-p+eFJnGw~AT{^sy_Y<N%-H{oF6eC7wj
zS66qSU^sOl)+h>@)QROs0iS&;@vG75@ZZ-+&@Fnte+X0;v{a#)3EtKi8l%}#g^_^c
z)yP}_whCxjs%i~L7B+$StawR&gG(oKnM&>>r<|Rl(%~ZFeG0}TOJ~9P&~bFt|AgWh
z#5x!P+BH-_#QoL^nxj}6s@{*xeZ_FLlYo-bBzplw4V&lHyka><MINpW3!b2)^Kt`I
zsU$6(U)g~qBj~8CQ4cpKvKWCd01x8U9bgfL&6uAKAW2H1RCm|wKRC@2%Q3RbA^00=
z79L{f<&ip#4mc=7(`8o#|7h}fF)`uLO}Y;AWz{mtLf+Hciv7$aNTQSs+%Nw_93Ci#
zyDU5th}P_XtdVa-5|7U_G6Gbbnroo6+)ePg(+3uP1tRr}ZiEw<N94Vm(b2k}6eHS?
z#Yhd&VFqrMm?H}CI(pu78pT9F>ElGBF|qr$zldw#2?>W0Dujx|s6@F0{0by>#-xx=
zwktBUXv8Ac@tJX&O!(e>z9s~cK+X1youj3Be?Du|qYxAtG+!W6!cpvwNu@$IFEaVH
zx=K5J?_5Q=PIxsm>HC%WvgNn`5?#=}LdSfZxHls&AM)aa`_iWif<q$W_Dny<QTP5m
z=e{zq5`rdngZ3saE;?gC1llNpRs;mYEDG6li~)kHP`U9LzXhKAKRMA{j+U?N^`hua
zw2Q9;nMNt*(-b2zJmX2aK6Q+es1SNcn0EYiSNaHW0fPAQXJ%A<=HU53%{)Ay7C1Ed
zWow@+_h#m*%4J+UjzNCg8nY$eRTC5U;hLFRJI2xYL^qb>=<n7B^^nQWZTTX>|A9<F
zT*Pvq+(Nhe4m2;yCNi1x_n7jZxN*2H!BG1#vb<EtJeQH-BtV6|6%K%s7h%kjr&ay6
zF7iNZ;|2->D(+*FSd75Vk0P*I3vcf(FoZr4r;4XdSGWDX=7nayRylB=H*(cSJ#~SG
zQ&(j648&JR?lCq%IM~~x01gXic<5VWF_1*lM`-4u$<#M&q|3^2j(e6-R-S7GqcQ=s
z3t-n+XQQ<TtJsBx(u)YG{SRHz7Y*mqp-SrGf{jqMCguRw<$rdKmXB`(bq(R+_nDB%
zsd#)5Z6HX3q{qwk2>?kjX;eSK_`!G!;IYj{i4z9{1c+kGY3B)x`rq-U$JrNOn$ufR
zeu48B4<D*pM|GiravKV>`SLgbr#S!&FSY@c;v9<|*aqwOtER;baG^LSX^{^k2LWv7
zE9L6miw9`1Acjf;%oI8LGXopv-Ipi1lRl$6m1DBSNv-%cdO=Xkrd|3)-(IxpsRB86
zZQ>)RGnAH4)N;-L!vYAY-!-4Zm4Z_{a}xk|`UAA1lN3S3x?v8p`|vy~5v7ChR4)fD
z8h5y*oPX3D5IHpSQV(mM?ILr?Pj*mbYEaQfg{_Zk;m1DnB>d%>0Kmdj!2zh5@M6U6
zx0>GPI%!dS#oC8lr`YVB1LWbZBRr&ykWX~cXyg(NwZ1bcw|5j%?c*O5s+1BDj~=4A
z*UCbmp`lIttsfD*eT3{bzZs<@RLh(uH(Sw?Xge0=0b}OFY<H3rqz7rFj6wCa0ZC0*
zcP^XH64x2rC(1on3Qu)A|NYos7sx7f0GDc5U^RTrCWPVyWK{BA^Uw5zwumG@QPVkq
zL?q`v4Mk@@p{qx>E%6r}ppa0#;<a07`4Cta)L)hB;dr`fT!E!7)_3UgM<+pGhoc21
z_s}~QgCUILny^0W&7L5j;qzK}7f%?g`FhFBU}?J|Z=U8!f3AiKjgYV$ndthNOM$f@
zB35Oo$M0e<_#79-a{VI((g5E^0Y<YTb)DT1o3jt#XkE^&VYbhU!#2>p|MMrV>c{Z+
z3YNprAjK$|IIGP@%9GPLDT?iD#{Y*8qZRa%JW%Vdc0tAf3}w72S0pr-%xaQ-d^L$A
zDk8FfIu7ha6Y}-<Hzy3QWx9V6gu@H9AOKaN7X1d0316GcynViqAt}6qS6>^6*i2TM
z>`UB$HnekK*8c~2`4vuPvCi-V=z1Jbl;w#F1g+XlOIHDdf;-KmERQQ^LFmeABeXyh
zbB)r+Uq6z90feUsj_Y8uUh|HZbB08Q@^s!s^x2f_4kCO+*wk^s74mKhv=7dFqrVbe
zoXVyWz2G@&8}9|fo#hoT@l+8zd=VOQMo?^yL=I%_|Bk<35$XDfufHc^cwL;iKR*yJ
zvCmeGNiYKDI)XN8aBC5K&y=^suV#x!<zLvoASO{zH~^Aq2Rn?V3Vd<e%DDw}p$~=y
zaPpnSN=4ax!0tGGT$S-_;?=S9>LdtAGbCbm{HsBw)zUB{-!DeF8H7QfsLkNoHlsql
zT72(rCB$KUU4!_j7aq)?ECl8LM7F@dUzf%~n^XUoUlY7fVH;V7OjJ%wSB#P1RJ&1d
zPVVtI?BHxW2d<bV6I34;={<lHBn|<?ivI`i+~5Xy(puaQEj}!wcmXu31Fct<3<L1$
z(Y$&}=Ad++)qoB*c8**@K<cE|BLD<ta4gHRF*$LBFbc~tx;@5)GB41eTcDF39GMV^
z44@@|5fJn!c|S0p|6KhVxlfQU@@}*PYKsUNoMWl}_}iP2FR>GMmeK!syt94*!z*N!
zGKhg{TX9!FC#>xv=!}1M$5W8aVidJSuv_J?GeVUnW%ODq&FW<T`-KgBg-dJ+Tr8fh
ztHuci-a{OO0t?z^(9aXlb~)f(;53#30{hC0iEoZy0^ko`%UK!TYm4}zOp+zqBj9#p
z#|3v8qlRG8fQgBTmwGPezFEyGvjd7Jo44_M6Q7CcTTVgd4Ch@OKv4~wqZUK}XvD?h
z8vviR2EcS(?pL|p+1ya6&}maUV+t&v?h&02u<)!rq|w}x7b`W6Y3{4{_tcEpzc)Bn
ztR){HO3RHne+Y?u@TZOE=^r;4)osEZ>+z8h6CUoIrv4em#xW`!Ycc)Lf9^*us3Z3h
zxCf&Hi6hdz?fU9pn#~TJ@BMApcFI)%qYGFzJ+<60QK!F_qJ*55R;oX$Va8b&0Q<bA
zuebR6Vn{{F8j#5C@l8aqxjxZrs0PM)Jw&6HPWMF+S$(e7zpJ1olaw)%)kQr;f-e4A
zR-6+U-34V6m|0t*TA7W+t;gyf$@z#LWazRL#%@YSv?Tmmn#?KkUcC#*_8h=cw*>-M
z?`dIV76nMIA}BIaaYPl~#s`PS`#RDNE0)DpjGlxLMRbB+kQs|!4czzvE9OtC*ObLz
zk1(>e)1*5&^}t<2=w$AOk7~Y-OwxmfrU|=Y-VyI7`Y0yrDVuGCyW6NCn*x%^nX^yw
zC#)P5_w|yfi)hr!Vm~D|zW;9|sDkFj0(dWfY|FDbA4WgEb<UNW=ndMbP<pw)I;3b=
z0XVH>%9=Xo+&Q%&4X4<S=}Pup@<9tIG9~n4^7-Zd{{C83<0vgaPx9dFy-!c;)q3-_
zR5eP!9eayu_uDTWuW(xdYSzAhW{2acD;&IhtQ<1g#Nads=s1K(^Ed!qNN8DES$S{B
zvK>idzFb$9SqqqvT(ds_BK|uJ-N5(>2mp_~;w7}4*RTysDgBNci~`TrxLpHWS#4s3
z4sef24+CTHt3GlF|1GvniDgg}M1b9mb}VL}_wbfu#bO|Q(Q)4Gb5R<W>%=-sp)eiP
z-1%_mAf!y##j7|!{BZF+HgTb^q{oZ!K}tqe#1K(_tk7f8;h#Noz1^=P)ZO-FeAT%_
zT%J|F<!8utYK}0xL~LoH&KnUWISc_5%#lRfF^)`1Q52=e5e`rw1Wh=aY>=bi9sN7)
z7d+RrZ`a`=uFYbz;|7LCE-o&s=`0sb#ae?!d{<GDU~QTwur|eS&wU12TI}=^5a{O(
z;1gaBGH6i(E?Zb6S-+?h^MUeV?&w{&*KQB)s~JUR>rLU#?)NzR%k>qZK*R2Q_#a9M
zo2<USJy#1sGEnF%W9e?^;t%vf$w+KjEP7I@+D;pO4$*EvU{VYPIIK7>i94*iLY~gh
z6jp(?_Ye#$q}5xeE}X9xN6B9+*k3GxdZ}v~q@T}E55s`nI}dQFTsr{;ML5i>Ym|Hr
zfM=({@+ca@<og~7wlTCnU4P3TO3t$OI$cMX<^H~ge1-@VE3-(q|NW4-naK~0>eI00
zd#Wq@bPP-6HW7YOw`^$}u)C-LubVDVLJ&X=MHvHwVB`e20}VF1{JxfgmHNjTf<;8I
zk=5jjw2QMtISewnm(Deu$0B=gnh;|H@pp;x9e)RQaNB61Hx`+>AF`2QorQ4V=`S>J
zfNxnZLz#}3tOR8oU$C-KU2<knsi1;?l4QbRC_z83FliH}Ke#0uk;OC;RoqJurett1
z$KjaQS^hSaG=KU&GmhHkv(0D{#%4Ux5sjc?^I?kB=~U-i2u|61a7Pk;`)kWtY4zB7
zTKF#r9BBsNT!ZLtZZr)5O-NE(?@7<$NiKmAkov4+ptL9qhrVFyuPK1L{6UND5(f^m
z_cWKOk=J~IQSus_p|+4IyW$HG6$c!GOgi@>nnv^VBa8ik65jDno9z)$lVRMBw(n=-
z4xDPXYv;A9H?j%`n@^Rp8#)j-MKG{Ybx<rqJra+MbYB26XXVpK2H)D>;Xg!hG{P}!
zK9Nc<L7pd-D5@0!K*OJ__XH;?c@X6DkDQhz!1eprVQ295@plxb`7=Xd6j8KHuW1Yv
zIEoInOBcFDgj3S84S$MJm=wuVTQpUuP%=1_9vMDMa?RD4sA}R~sty_z5EcRQ2RzKi
zGqj5)1FDp0y*)ENRhC9AWrv9g3&v~zdf=46bI0=HdUgW>besyT^i@A{9~-`B*PDdT
zIU5nDqRL_t@(o}{Nec^a)ONA>?IGu+vyeA|e$F`h`|@*wyqw;qZ7llmM?duOPH8a)
zhLGC2weKdI(GHT0#E4%5q0nj$O`o34+8!1-=hmJ0Z*H7O_`=XwY}v#{!17R4d>Tm{
zNNRc^Y%}SvcNWT7$ro;*@qA*nT3Ll02HWBy9PxUgItK&;+rrC(S;<%Mz2q+3F&R60
z1tFp2m~0^>Su)cktnigF%)W2WMq`iJhqv1+UWhg_L+C_^UD7Vo01DM@l=$4q<*pSQ
zog)Zkt%aF9(Alr_x&&^U2`WkiLA=bS5<HXj>m$?X*7V|E<mA+Ddhm?h$a-CVOW%e!
zZi$rTW-n>M@NPp-Bw5h5c=;Np)djkbS5&$XsvI=NTaonzQP3HGw7JwmDm|R5!u8PK
z?QXX4X^YrF0;h&lOK&b{U5K;+;yBLtT<lyp2AqvxoTQQmDLmh+7U5ok0Cu4FD>)t_
z_^Z6-e2BG+x;)J>sDA?fs_{7Y^h+yD7dtE>_Bv^{h?EirQI`xBttv>f$<8`Jt?PZ~
zXsN)pMiWy1`XVVemXF)@;rzIxme}XPw`5WYby^qsc+T~%^Jimd%m@NxWS`Dkj3-I5
z%In%IiM(Z18HD3)Oi-Mxg4nP;4#NF%NqiGk@}Ze3B4<4?y)r>^5YyNo%yTFo6J?3s
zfrNQS0F#<6ee1XIYg%&cUI_eDJ?q=(pUp5Ch=Aiaef$vP4Ih&(`OgEv76#=*YJv{L
zX%rG*U@lf@QPsFd^N&2d<km6mV7~X{KzNK0D@eAD0Z-^TP|%<%n`kyXwRX7syPxyv
zeHD<$oph-=l~7pv;Dwp+fL+M5e0tmXLwVj1SO?pfVi~koUARnywBcSRU*IWHby~9k
zr!??HPLk$*{IOyV?RIAlemg=k68>;n^IvZzk-1^4e5rVdv<+N^HW+Dfs5#8F0FF9t
zS#5t+2k19)y%3r3`lQw<7q$d0UGN9RFL57;qnr&SP%aSEj#W_MNT4tg<!Qdr4n^1M
zq{jD)Na>^AT_z*hI7|<H{f)tt(#-n}#PIkSff2&XTnfhC;#}g`L{q56=$qu9div)x
z;}FdhyDHzmmd{b0{kc)FPxvF20Isy*R5%biov8wZ;2f~s5oNTeOzZdPbg>d{lg!#o
z%&avfcuQ)Uf`hXv2*DjX#U|vu|3P-OXq15|VRA|^)d#q4IO22BJ_Q>C3^;5pK86ia
zphkz!C>HZ#_o|ZD?bZ`YU|VwYu4kaO<^G_gtbQtO-gysYhFE1Xsz<Hb5I*<X7{&{1
zK8va==s{c)e)$oHxku~W6TkU<**7C058_Ecfk?YT@JXl!4wv(LCa)zFM<<CmviV%R
zCBy4Y1XY}z78P)mP)VYhO6Iv|tC;Z?u0ARsB2PyH!knZn6p^-SwG+Ng-L~`14@#gr
z+?$`En->Lj*hVC^m1<Z7TE!?}n>qeXD9#Hrw+nRPvc1Z#Wf9?jeT}>_Nlzz&*oPa@
zVF!5?ce8(Fbw6VZo8F$3=g;pgIh~%Ox`=BULGZqZ0yaQNA?U$*(w5j3HnBeJH{T<+
z+coWHmE#iye!+!Z;$}02-)_VSZ(#T+MSTc7pvXn=lOBtT+91wdB_b3E??p+o&dQGD
z6juF0(S$OSVb~H$(WM3ytXm5dVU80L3Y~*B8)$pZ0qAO#7F(C<AZE+mZi;)iKu~NY
z`dE+(+W3JGGLQZZIt(Xp$8*&h69UTG0AF^`=GlT?#NuN&u!E~d&jAY&>s>@!jMG<;
z_qHn9-w&3LGW16YhOgcr^5bonO=U8v#78Lz85E~)9cQ=V0QMKY_V>qfS)-CP`Y`ws
zCstyub6G+G>X;;GMzgQ$4g{LgrhP2d8E*Gw32u&dxk<%8Jn6V3EHHA6LTFhsC~<%k
z7Kh{;b&A}g>K44R;_G7d`fQz|0=S*RjHq%ySff<@396SCJ@m<6r4DgmVB@&=%ayV-
zaW3e#Rj4)IJwP&?saS@M;-^F3h`vPBodR%=IlFNAj&2QB0beebHyZ(*zC(kyOnErb
zI3-6Wio212><d@kM%^C;;4kVTd@CV0L;X!Dn>3hF9tVT8rAEtd^Sp@IH!p+`>?<xR
zP;-MfZLr$(+YpA4vU}iM{Buk~0|Uo!|Nh(mY)|}#C{7hy7tD91<2&=!yir6=;e*OH
zf5uLPwN@PKd(8Z?LaNZ7Yo9by^bCTG=o8WRHo{B%6a60;3b$0Cp+IklOVm00;xiU7
zGt9fbou_9Bh)QCHMO&-Xvo{7_>{Qe?lPyN};PEDt)~8PLyVMuc7r}aqsJ;6cJ(l0#
zi=ha!m8nE!qV(JQry^(f#OOc&toImRe@Nt=9SnS>fL7(Sw+|HDF;Bkn0L`oM?eg|@
zA8<*5=ElmVcGAUB{1t{W2aMG<;jBlcn>T}LN}Gyrl0jLqo7k+za+c}Tf80$1O(uxv
z(W#+2MRYu;-Y%M+qxqOnCavD|t}&bJe;#fG`9Ceq#b#+9z}RA|abmIGn))k=C3DNL
z{qwEpA$zsJG4Wi>&CnoGK(qF+rh3p3l#oP*z-3aZmuOUGa`uCKk*~~298-VGMJpgJ
zO<@t?oh1vLj3O*DV~^ZNH8=8coFI(vZjKjg{&N<4JsOZiQo#LDa(+F9P(x+;{S=ps
zP5#Wl)bKKV!<|*!!c|xcPsgKyY`SypK6No$1ioaWD)eQmZin$=__oE$z*OyI!u9We
z;_I77=>!LTV^O)jQJ|6w=lc<AoA|W=KqSYzU`ZS&bgLvb_QdjtmeOCh`I=(*u0lFy
z(agcmJSnK52ScC(<}6K)S^5`pp0-ebHn8IS>xaOQf{q1!8d+1WkU)iCc}bA#<vQ7&
zO2~FpN+jckX=^OYCu3I)+;sKj1v!ltRI_7moJ_Fv#p)c8moK88&@4wK<h(YnKLY)Z
z`{~|G3UD7IDa2cg0JVhqcb%65=uA4T1VG?St8fL8xwMu1-zT#!!VjN~^-)s=Ivf?6
zIyNxcrZ+n@fYu31%Ih9Ap*H&Mu^D_VZcZrnNXXVplxv$R%2XHQp+nzrTh9#ldjzO}
zu5T{?A5&i$Rpr)&tB9~c>6UH~Bn70qJCsK0+#uZ`xasZ`1nKS$3F(lK?nW8}r0?42
zJLlf}*WnnzTWiiI@d*g%iMeemni&hyRCOPOj``G<cR(6y{YLDIDtJFEC<?l|{v#~a
z&h)>(7e<e;a0ZU9)WTSF0nK;i22j@cSi+^X{f@S4c#Q?%0sOsn1mE*P7(^A22^UKs
zxK4=2zu$5hW_F-|BXN+^r~WW8lb9%S;Bi`Yz~MH@d?TpAzTCtrTeoS{+6`c~+3X)~
zi2P1F(OXr(Rc;{|un$niGq0P!hv7!UAdfvB6qU^@y5Yv0NJR`SW8>gFWjq8mZ-d|}
z8M{)8E0Nc*>eiJAF2hGy-vS@a|9x{HoQMS>GllB(e#fN4Qc+~D;7xDPI?#EsyU=(x
zhTi}20W7Cf(7BozxrCP}F{}nRz|E2as(P?VNuim-;B`_s8+sw&R-pjqaL?efZE6H?
zJY#tY)>1_;L%LJFByg2JtJ06@GjSm+3%6CZ@K;3OMle%JDm3hlzw*r8K4z>kbQM$S
zUF&hGp`EnJ?3dTG(@;9`*_yO%KH$~{$!YLJ;c<{OG_Nlq0b2mh5f13w>np&iYyy2`
z?I3WXQLgMBwq6>*ttd1!G=1<*io6A|7J}JO8=DdVAZ|MaZIyCB>5*g7A3Oe4^vNnM
zg7OA1s?cZ}xaXF}odd9p+`h&3pRrK{t4|1l>C!flb$Te;p2;`qieJ;b{hI<=|HgZY
z^^7As0aq$&jQn&rddj<b(+f{fHNOeIFO`VDmWlhSL;Dr7KM=};?GV2EwUtpuN`{h5
z$v|GedoQk!f{Nhx5bEz>B^UyhDi{I+R^ZoPFEuAIYgZQ0A%2!Bg)lF7ae78$ffOyH
z-=d8c8%|f&(4D|!5!>n1=m|F*mNXpQHm^pwxGkC88~BWDGzBI>iJ@`fyVBHG9O2R!
z4m`j*uu6HzMV0|V&@0m@VgXql3rvb#^EAWPAYFI^gq<eB`*&80;A^orrnpkSIJ|`u
z`Y><Tp37_ND;FzpI*n*~cU9~#b$Qf3(C|g{@=ZQbw2vzMp`i%JY;c@n*cCisboW|%
z*tlwMk7mkip-NKlV3{ueBr66X<NXb2j5pc70-@2Y8MH*3&CO4OFvwX1JujTzADjXx
zGKv4NfqIzxfmNb79e{Ugz&Q7y4<|e=e-$Oz-GKWFognoh9va9H^xQY0H~vN9&dSm8
z9%G5v@VH!Yv;cf&v_-y3&1!{&JU@IzIUYTsT)s;|x|eXdNsCU*qq+1K8XzHw=*q19
zY2anLZHWB90Hx*J>9HONbX>gXKqjA|%}I$?br7R)!sIp6iQ+$eBdds({=|JPgnj_D
ziQ2qJg(#w#fETS|RR|tft>@5BTh|X{vNrE=K{2OsJI?iIF6i7O%c3yB?X90zwMPgp
zqlT6XEp6YAb}yRhx>DC>=feMe4WrQ!-j{!2#Ni&UzEqxuS;YHZ*hCYXg*&A=w9qie
zeH;NTp7W!hF@$#Z;`$}Vv5qHK$EY200gfnC$?;yiJpov^G~2GWh~Kw0hiUjd+Di1k
zm^J54{c+w4+X0xM4cW%wVskEdKOI0bVbfyMa}I>88zg)aia$=F-zy^@VU)ehHOiGx
zfNcl_bDUXV8(k|L;h1~~nu|4Q3C@}GW-~5bu<o`dRA+dx9;&RA;7qsidN!Q5_yWA@
zItYE?N1$Bj1IG4a0pK+f9gEjC5D3|@23y}Cw&VrBysT_xF)Z+KxqUhCH)3jKt|~|a
zv{8bvLjDZ%raTS&6#K0I(&z44+reXiF~32*&(%7yuWX}+vK>Z+k#<p99#Qu1nSm&^
zGS;|0#JS&>Pk1?&GJlWarNUhY+IJ7&RQFuScYvo1g4ch0FTm>zbQ<-8bWp1OVh|hq
zoaD{VJlwx8o@y0>5}_K~MIAkj{{7i%s0V^2YNKIz(rqsQ5^kEP%}+sKTbxd<l`QZD
z!Cs^G*}apP_4=|a7m20fpKJ`NeIohw1EorvTTT2u&pxM?CQ^UX?~V11N&`ixl95D=
z82r@VeHbAR3?_vf2cu2(aR@0Fq43h$#j>0I0Q>`0aIY5tW2zjtm2MsdLBv)lpbU|x
zA0+f6qqE?}AW1``T}$o=`jBXxOnAE+$a|(5HU|`+7}+{VVKiN6u%F`ZAKC|J4XYc0
z``>IgX@=tIiyRsxug2N9^c`s4-oJy64`iR`a`dincbN$XBd718`%U||s!L~}6cGcZ
zZFcBXstj5w9HtAlyt_GZ-I7s%xskLeH#^*)_cM|3a=5l+FyzLH`B^dYhK(G%)LD#m
zde_1dp@%QL_?>7OAUTRQC<Bb=G~kC}{)WVJ(XttZe;VFe45T69eVj_)C4IYDBZZf%
zT^l=YzVTch_2j2Mn^BRM+gn}rL~6c*jNuZ!jl>3tJ@)`T^C_9WAY7yPtm8SbARXLU
zncHF|b+kZJEH{DrY4Jb^lI>cTQ1R<@b8udT)Pd~kjWPFZN_*_f`9Vp7+N;2O1gQk0
z2hXp~>;Jt_$K(h<sEld8A>+poZyKX7PdxCy^t9fH9R)B>9`)bL?~X{+Ex13dY(O+H
z_x_|${|dR7I;D9I#aI6?#Himur4PJ#drkwD?qw@dWx#8`bK4#Tmh5IEJ6-)zPsGX3
z$(MwCM7Vwx5|0ALew!K%&wTF2e2^?O>L<LqxK>h|fNZPyf3BEVDpyWIk*|X}85pG(
zCIs`aCv!V)DHNOh9+@rE=ZD96A363AnM2gpW-D!f?a9Yg;cb4;dJwY!H*7$jbfjO&
z)gv670}KegxCokD0-P)7rxJo?ZkwNOs;a?L^gbS-h&Ob|tyeps7!rT7l-zCQ@jO@K
zQF;4)6B?)BT%(M0!$}V@oEcpcwoF;7lxTow8`q7nj4{FTuu(pP0Ax)uQUfu0OKAgv
zn=h(GA+f|c!XFB}<34L;OQ1|kB#BChpcr9xOsO*>tOh<3W9N!+Lv#1<;OIihZRr0@
z;`)|<cl@h=X_So8hJGe%G!IreBX(;2D@-a4m-9$Bd8^NU1m}4RB_C}wF-BKC^Zz7@
zqUp-^%GKsrpL?9Pmn98^Y>tmyZ(X?83jZ^2q9PznlP&985tK9ZBR}8}8F#+orqVI6
z^)xNhr@{iE2LlAZ3e}};K{1&h&00*CItnAr`6u(I?pyJi4r{C%O^(lzM6KuRpsDth
zKH|yS&sbZc%{)EcGZ_2v@hr82>oljLCdX(9CtKmUY(TKW%8;Dht!SH54O-S`Cy{YA
zL%jUHdN|_h^Xnra`-#yZGA4WA^0pO$6;2KShWpGCk^B?r@RM!=f_W%AqR)ore?1ca
z_O$HkgBz}L!8<vF-}6)jIb6llL;AI1{T7T^rK-2uKa}SG83v=@LntD}3ZIsrq&{!t
z1uXEZ-P6Otq3az}9FzTlWre<7%3Y1dkOkj0$-E^y3ni!P_ZaS{KmwqWDuegHeJ$96
zHPc}K?F-^@UU_mzfY|i~_ELLsqyZM>1p*hCni5-Ou`~ZH+gKQ6l0+?ha;eR6F6H1b
z8M2W_eOMx-Cuk8;6yk6Xk^ClhhA*|W<?{Kho>!<wwPBNcMEnID5`e$cp%6WT2d7mW
zrNB=BZWxBKjlr|*iTDc93@Axl)(QnyVK_7#ryup#UnB1%RY*|0UM}mzmLqi%m3)*S
zEB3lUJC^N$3pGB@;^2$>$>HdzvmB)gLPtmpyujfdU+Or@Y%Rd0Op{Pf78-^7hdyCz
zq+Ch`i^mn^ZhU5%ABb4r>a!JZ0%%OmDDlK*|FhKo5UzpbklCNlVKUrg)4%4|bM;Ob
zn2h2sEq00iU5|17(kVYN%z@@yD|sc%z~5XBkh=h5if~xNOVK}6cuRXPj9x6YXGRJ^
zAT5B8cDOsAeiopMB7E|q+H9X#$g9ECF{6~TSlXs-x{vrc$uG~0tbc_j2uJ9XeOLQ&
z=d;b$%cV22jMDQ~sc6%iwyx9J+20gt$hj|aCn%3bV>jn7bp1n5cX76P65@z8YF6`S
z_qj(gKDXm@uI6o`sXjLoGS7IQobf-D1OX;u087TFaQmZ%N`bH&kQ^wD>Gb$k^iOOj
z@?*2^0!d^PV<{*oz_3KZ5bZjiiU7HF@FWG@^${+r4&oFG>@Iy<aWAiyQvP0{>u&jw
zf8vL6<nVIL3*561r6~1;+>lj^{>;I-UtC4$#5sA3*&b&KYQ;Vor50P-@L&ZWn35D>
zWx6X;1`QaEcwj2bvrP}h0UVvd(-o{5Xy(>l4n-x4M64l<+Qlh^MbkO!YX2b-xPJa2
zp8}Kr+2CF#VSOTbSESKD&2&11o8}y79(4!Eb)PG8VDHAvALX{fRlOw<Zw$+>gLxW!
z?0Cwp@VSjLD_S_13dv6#I<!pjt?Uh7@A|LXtu(vw*LxtZ{U38Y-(Jb<`Ky1rJpiz1
zSg&64#olCE1*cTIuHe<dCwkW8i$4MMeFVPH=-xJ66G2DvAuZ}>Jo_=ONHwv4<5y63
zP0pKm4rOuWf;Cu7f14Ek|63ra?JhuO@AknYRkENGVPO1WdU9)+WZ+R~_jQ9d1I(O>
zBx>!*Rgl$hQ<!9ueUz|Qyu7m>2P1-)dLa?HZD0GgGksZr2(@BK-0GeJEkaa;vkGik
zkl#i!ayPn_Mn9Mtoih}V?m_$DEc<g}ZUw+}Cg&l=)a#o~qa?v-dR}zmD1<^Nf%T#(
ze1CZWJ;8qHeY<WGMkW>i1ugO`VUFrI?(?hL3Ty`5TzMo&^8X?R{{w==P#)6;XZ^9>
z(GtUd+=pN5PvHU?Z|E$y{SK&49PGvu1njAPY(wIlO3EbgTh-*8G^|HflS$MWP9ln3
zJ+q!Hh(RNwoxQPG&78#QY;rlUN=muw{jvu}|GFTN*OgyFP&jPnSRHN~V#YVtc{aOD
zpV9W?N{*JSqA_8S>Ya1oD(Ox!QwuLlY7gL357kH01_>Hm>ElBea84V2;`e7vT0)w`
zcOp~&Ilca!NV5DtVRx%s(K?Y(Mu-?0Wy2FE>W-Os{qI*ZA)Pie6&=`}U~JP?{s4f%
zSC3))!zrB7GfJyks57yi!MW0iq<#t*xDxhFu173x&)CHjGu_ruJR|=^8a6ZW$?NXZ
zs9Ok9a!DfHUPmY_w;0GtNy?w}F}PcHFuU6t^cX12`)y~U(q`vmtMa1d(Z@LcAUxVk
zns$DDjwL@g)-8n>1DrF77!O*;48|73>+__UHmxDZpGoqF=lXVb{{6(9KJkOECW{kb
z>a2f<Ou2}Zve0pZ-SjMloxWxp8aDOh=&hA|5coq|y54!RP(UzWgl0X`skhbF{f%Hq
zWDbjtB<2~}B95I7%E$Wgv-zk^m!)~1qr)1)*c}#3j~zpl6bBJ?G8tQBzB(+I`#thw
z+?^Hd+&0TwV3qCzA0VcO+p{RUc|ha_L~%}0)<-L!;+F&QbD^!Afeq(|tK+TD^biXr
z%W|oHhMBHe+t*3$*4L}m9l2)F#j8nm@$$+=lnh_+*{A=`l{0Nr+7Se_&|~SH(3P<x
zaK|63{9mlm-|u^hd4}bLV#~G%SSMQFPn3L%VgsU{*Yj1Urrp6<7o@OD#-#|Brk!jj
z40<tC$~H>zz?uG!u&Z=YOn8*^<=Z6pPoF|3Hy#(vnRbY+6cv}nf<a#vOr85^h3;ot
zZB&&T#M$VmC-W`t)p&WUzgrQwL4rsHr+z^?sKn^^pD5O3ovnkWfwG1XAM-`VNGc6+
zhCR3$5ev)mfgo2A=s-$ii;C@Jkk9v~>l>pm)y=~QSY3);`ol?YcnJ{ks=}c1GV<sL
zQfH%Wz9E%28MCGuh-+MxRrbpMIb@JLVO0~0BoDPFi}>*5nojJlld$>$rS&Y&5>3vI
zYQo<=T5)O^GEdh-(B6INSLv1L&4-_lB5$aV2Lpd_Sgs4lda=D#D$<hnm}DFM7Q7(e
z{~?>mPhg248F6dztJ}#MJ_lXK>>+tVcQ?<`E6A^`FW+6qL4hub1r?^CWU0A1&<>^R
zp1<OO&Y93R-R;p{zqoDo|G4Ztt#6LE>~7bYUa{P+==Cg^h7UzT+OVKH?nRet+0z4L
zA%z0RVTJF-n=A;H%R-Lzb_a@B444oV%qhHpfOPj#o7t+Ra?9W>S7|Wlw7iE3W!Gn9
zxsp<a+Ef2w<r_qcN&7~fSf*~fe{XCLo2a#iVVk}!r?UP$Zd%eiT<bX4lshn&#_DbU
z3p1inb*rM2u@TIC-PYGNWt78339(tm4;)${x3w<ydaoUt!ke4EycIy3(yDSKyDdBE
zhs@B(oO7^aTgDr~;cobO*NNe$A#UIiK~9;rcKMg?)@QLs!$6D8fX=CXHd?}A+tC)D
z<*@+Xs>I_vFrp?-u0pBN0vt3!;!TF~qo``tW+JtjI;gHLJ<+CPunuTI+60qC{S)9I
zk7wf@aiQz$4JR_L|JIG3Vj?WH%ltfc;06Ut36JSr&HU{{Bk;XLH?Ze<g?q)+p3H(U
z-ia3Ftbq}B2~B&FBhj1#k=NSW`XMqe=^qgmC@&^*fjmHJYd1@4NeT-HriUM2(dByP
zO)~LkMq4s9?@0co4XS&3xoUo1RWi%P*=@$*L-(STj}vN@F`~orCMCU^uPjfo`Dbn=
zypnHX)i!UjCQR%JgQrP)uA@tq6DZ~{X-V&k9$x;74dNl#v1BmoVy}5^{6rJFT6x@f
zeOY;E9SPi~Esn6j?3zgFa<J{cBgg`u$GUqCe|N$S+dn&F4!}&JV`l)0o1StYop^v+
z6K~@VPnq{M($K|(ZvS9+S<)r`>~Fzx9$@b`0m6eep<7Lgs2ciM6jh<VMWBPaO`AQ6
zR?xf%Lz56O)T8goxdUjP5q0*P2LKCF*IG~4-TcX15Ctp1;YvX(1A<}LqiVZ)E<2{9
zj5+Z#>v6e2E^BJ_5n_trfu|WyhkgfSv*~0tV3ng7)i7?qG!|1-eFIf54qy49!23}J
zDJtWYvEp!dwjri1>WF<vVciG~=a|*Mr^D*!@5J(S$Rx}$RADZ5&l^~qi9=AWrpvyZ
z=e;e{(|+;Vf&%zTT4@b8jzv+~olq$>Yks6(qK$8(v9<zMr&Kc%3Jp&t&FB|FB5DN{
z@g_Szv$&s{1GPjQ5ta&Y+gEgt7e_oAQ{ajlgN%y&X1KxY`>B!4VH*0&;V%F(HqsqO
zDt{jXZ%Z9-cY<R!p>jNqB|k|0tlj1DyPkk*OM|O0#Xdoe>xU2`vWNQCM5N2iLH3SL
zjq>87kBF_lMMyHFP;Cwtu<<S^WvBOKY?{@Y)_b1t<V73?5vjJ8!IuWT!#g9<pRK7W
zgr``F-=8@&Pr=f~@c(;9#K`^b7(-$`81DG$3xm*8Zyz2cHXIxzVXoKPD_B;IAlzsE
z6;}$S@eH1(9;IFTU7h~uzA$EmNTdG_Nwko-$KX=}y$`af#dwO|=CD5$F6lVp&UZBv
zN@rgZqo;rZ7g@tG=veIUF<%>)e!2hPWeLI_u{&V2PVeI`ewoQUqG^hS0$ARRm`_jz
zyt7sQA|;H{s@2ThoXM|GKDzBH>J#^!5qz;Qlb6IC6K%;G-J779b;IlGAiPA!v2F5@
zZh32o7R0_shzEb^-@L<Oz_&czL(OFk$A9~#Q=vPt56|rwVI-AnaGC&ywH6>(zzI7$
zY}#Y#=-Q^WoE;<l=pCybY%vUaEG)%!TCaQy;<1%bJGyX!Rj&t$xkLNNjg?1VY7|hw
z1B?_n1`1&n<Lx+|9o1I=y}d^N5!YY)=7e||UB90jfXvlsU&#Ie0wupzV3ilXWtj>t
z&v{Vkv&u}dR3-|k>rF@)`>a%j7*lPxoXOmZCaYJ^B-9tr#Yv665QzN1r;ct6X8c@D
zg!=K)f1lB-i0|I7JaYJ9W~N1X&8+<Vm|5Z)NLh)Xh-KIV<K#ClIL~i3fJO!1ire$o
z$QakBofz{S+LXhF@I;UXRW!`6S7j1H5if4=4W|^vX(5cYjp4o5Nx(T0w$)s|GWnjV
z6HSbl>;->US1|o4xD1uKoq8e9FXCKfp;mL{I<;~hmxr(5Q>rrB(s|b4KJ`~b_M4gA
z81VZ=qaiXZt1L34tGRR@fB8zZU1TVbOis)-!M7PsKVdUt-N}vI!l~}_zD|+qxjPy`
zn7l$vZ6E_(UJiLDF3WWGx(0l<|4KpiQS?jVc&?ndtUxpsvVsgFPJ%45TCQ!#>8VHl
zpu<6I${T@`Zk(rFYTXAEA#O)4TMl>Wa)t9je3n7zq&D7Wnr=QT-~DXs^|Spwfb(en
z94Oee)3wNGnfjhc%>DuHxe7#_*u;!gEE?u>aO3LaiU56qX9w-#9C;&Nq*;E*pRPs<
z&-^K!hj)3ML_TBPf!CM`-~EYuPRRB>uz@f4?R(^=SZYMn6yXwzw(L!dl8X=bgelG+
z@Ff)Crw#FA4L(ZC4(@@EdS@N)PcN?px%Y7&r|C9vML*Mhi}BKOc-PtShJ}cj)-+6#
zEGs@~qV3%6pMxrr3y}e1)8TTV)-hA^DQ;~1m*}B@X6}KYL314kw|5bA<F#{!Rh!qN
zyRn!fU=$PwW%KwIOcI6^fnY9aDSbs)5o0T0h9$xloUX*m%VV%8rh|Nvd>8lgnI_^X
zx!<YOBChgUVnfc?KzB9PlV9>Wea-6c`uRac)7nRB15W5BI?+;n0OLDymd22P=N(Gy
zrUGg?4ux7F?9*9!|8D1ekk{<1IZ^ey3Fc4yzl$L}A)=+g^fKhNTWS$ebV{Sb#C+_X
z27!dxY9K~J5oJv1acSr;)}w^;_`mHQ)N5(LG<eRjIs4w$534eUaTeLwd%DP*d|GQ#
z9UF_q%%mtt{*m2j#kbI4*m}$EU6gW;#*ELl5#MyGmx)8Ox~wvs!sIHq%wtZt`STcc
zkp|u5TVYy*rE)fk@o9GPfB555qQ`RYZb2~FV^*v&e1|_2(e&vb#lCYlsbWE3U_KTp
z*x{C3A(Z~)^;GO#mC=w<Y?_F_Kwvl5c@l@Q+p)@<h)_27kjLm0CHd1t`Eq9-zlQRg
z1iUUw>k{LvUy$NZUoLVhq)rx5pcN;^46+FHzRYO7zMXK89p$90vj~^966W~e|8rdF
zoT)EgoL{Ctxyvwh<HFh3G_+)<P8lMJM)3JVTwH@$ug`2It^p#C9@~`tKL;|V6-1?l
zEGAVX@X_-|-{7VKMg(p63Ha{kcwMMSGSwsZG+xKNmdhniu`g7(y;@aY5E>?ls0tT<
z@fl6U7&Q?(%+h>~+g=HsbRhX*S6E&H3Z;nKl?)ev?P;DZ5zy}1ya!;UAG%l_5hp{e
z4NouCZ=M1K(L$rc_o0vPIBT)*!e-7s-K@pgb9fKXq`Y=)({n$FIK2p7t#TeTn~oT|
zb4(2PIO8H>a|`z3G%Qm1yX#4GGVUmL^4?}|tMqiWpx&gJsJI==R;N{=#LLrIy3lF_
zL)3JB^JpZ#aIl{ISM~1ldj_xNCLYx(;;8Jal*#`P6!?d`4;%jVL`>p-huDepc(QY@
z4r={u3MfxiRjB4G3_rtr%f|B*iQPe;T`pky>VB49qEdPTXe|+_<z~scDluCCoi6=?
z+s4AQbT-fpcO!1Yr*9i{OTC1({=<~;TT$6(fPBCOIUDsTlum8XA>RmMX>2O>>09+1
zO~N%8untaW@MC!l+zqTI%c!Xhd<>?_v}w=Y*~Y8o4*m^na$7}c7W&xdJl^Jr%m`xc
zZ@E|7qtncsU-mE7FMw>xGU{hr_nZ79&0@0Mm6~-4oWz<Ou*+!1AF5i1*EW7FZ|=!M
zIQCY}+@w$F4kMLO>a+IT{yDb(q4+4su^{dB8D5C;%B2t(HG;jp{b~EdowCQpzNKKG
zQs9eGJ%EZa%bNItWyS<_F1E>=2%7*G1$hm}M3TNs$~j-QOF(Z-XR{j;WeSLIjMlYn
zlpIK6wHrNy%cg31I&~O6%9mzamo@g(1z~dJb$3g+ulONv0JNq8G3JP7-1x^g>Tn+R
zi)GAF@v1@GN&R|{gc7dIWxDXBE$3oc2enwclKk|g;WCW24TQ!TP3752?=+f}`}7;T
zHS#l)skL8X_KQ)yksKL6x|UQuF<nKeWQSw1NifpqCCYZyG%2*UioN(eTNh5-8_SqH
zWg2cjKe0FWs<L_A7*ZJcr|AC&OFq5^V*wc&<zhJm#)|-p6}x~#2IX+3uulH;nocR5
z0sP(#=a0lLZqYiCLb`BvLU+WfC?ccf!|(odoOtOa7wuUC%G2c>K4lG>XL`Q_!t8E%
zdf*61lc6^GVM%yrXJDjY0>k^lXeD7qr*NgX<l&nxIJz+FJJ%wHXaTP+%=CH)!(~++
zhXs*EC(mf{mKk{K?nSq~;Ig!mwZY=Nc$LN@yXq_GUwDH)FhZ1<w<0ZDbuf)98^Ui9
zV%Y9dyRrLBxuO>9gl0smR^6jrNdJN+2S%z2yP=CoYw&LGPC5BcDS&H(=yJG_Quxfo
zDA5%$ynlI!EU9qh?gcG9{rq;@vf14WAY~AJK1MDTlndFwN1!Pe$E0)srNj84^G4B)
zpo&lT_&E!Z5^c9SZqp*4gwSVHyzl4C34V3=HSk@T9&ZYtbJ5~7$adoevf~3G?qkej
zorAgBgK5q8-oIFQ>(uj=Tu9Jfyv97m<g6N?{kGwSll{W3@Anib3Ww*E!hB08W5G@L
zXV*b#<E6NB>*+qU&#XzsWyEyKt!#SOIM_L+^A~Nl!dENa+D&dTImwk@erDI^0HpE>
zN|4Df&p<*Yr3BddR>vErw~6hEH2ux!b0po7=L3WJ(@qsMSh14BJsIUn3AJ=KoFt_c
zSPQe2!n5I-3lkq(5J;>4hw6ME!~77NxNQ?lKS0-NY~a3$m&Va~51gm7>*LmGDF-{)
zuPw(VFh?xsY6^!^ID1`ZfS3lV{??1wi8crDX@uj#QRwK1a3tl^4}IUebImK@?{YxI
zUMhui_7mQi=&~rmEgS+KkXDjmpqU`82suNrovr#2_;H+}hrvteY9mfe$Pfe=?gt&X
zVUF_cF%towF|8}6>?t<801PsW$sH>C9Lx>sd!%|zPK7Oh#^n<=s=R>|y6zF{K_sAc
zuP)~8OyQdBRS7rCl9MfapO1Nh)d{z@1vykspQU4ih>%p#Oh$#z;wM4C*6(=Mm#o4!
zu;(0&f%LEKn5l!>8NMx(P@+ZJG0@i@G4L30iOjt&W1`QTdV7u0B!?ZRkX%^VY}aNm
zvNtvRvQgh>Q$<!V`lTm+&Ok?3vv)o-iDmFgyawZ+qc4?CHaT97;`EkUjSY<C_I<Fg
zp;+Z9w3(x`9~^MAB38|jhn!2eX<U;~1piZn^?4$4|LE;6YE3!&y?JMGdiWgPFWJ3Q
z(D?VKI0gz%pTPE)MpxWR{iLig5y~^;fM0qXMuZ^vGX{kG2s)|$irIubzNblXr}}dY
zvN@}BvcbM1J$VH;8J8YCDfDwQ>BUs<s7{=HKm)7TE07M6(0-4~>=gTj-7RXB@3==4
z_X$zXU-*u3kWG5<1&c+j^c(nJvx=CtfQg-6>xy{c;0cm#XQNg6?48#GK1Y)Qy1WjJ
zG8gdCRqmfLWhNC+p8og@5fz68LWj_?LlLs90mrX=5+7HlT`g6h4PpP`_TfXi9l!+Z
z1amNrdtzsmdPJ82PK4=uI`{^Ffwp+E52>7*v8a-==X~cSzjw*RJ*cUuaxK_<aA|!<
z(rq`|24%kWD3+|f<Sgh9q)B8^e|K-dHX1#UIOW-Q)VkFrA-mwVall$+Ls`f<v%x@i
zePkEPK@#Ott<siJ-ygrmVAG|-%3GMX;O61I2dyE%F?;y99TmcCgy@%DLZO_D_CBci
zA7H4A`ukawdCat%0kcj`pAfw}nduKYJ3ZQ{;j6dYhf&p|Gg@c+#8V~T^mNMI-!<`C
z=}z`xg|1%>Z7{9bw_v>CI>iwB`@Kz5qew%xto`1jP^Y%&;0)k!m7}I~-tI|j+C%5D
zYW`KcM}{M2L&=8AXnIYd^BH5Y<C^iq*|&IcLkjqgWk9^4^;m!|Yl3MkMzy_(^%8TX
zHT>Af$J2aEzvgF5K*xfC{!4_$;73_cVW27puFtpQYiU_0mDq=+{ra;@*E_h7tSoRM
z!M`qHfsnx*?(h`XKOh=DdoQUlw2F{NzYIppV7mV~%6}x_*O%)`ogdA&#5}-LaTxxy
z9|T-OXbJj~s1*>_^R57vGylRjnoPjxcC`CS1!+_27Cd_>JWKembhM7NfDP)Pb_HrI
z_9Q?l7;~B*D0ld!lDdNGF=CQIqTT!%uaRlH+x=>u;u{M9Z9`{}F#O>mp6JvoYc>D^
zNhA^5f%kK|A2x0?i%KvD?jbOhz;!ige)VoF7}7&Ad{act>V(cWv*@aaKeL_`TRt}(
z7YUSmNnXA><|MWJVbAw2d^V!lG^`8b7g{q8&94>J)8&(KX*s#7^qK60gOP;wmDN7v
z{V2PI$-v|}B&5J>t}vOkz{xHIE|>LH-|PYWx^2$I8)n|~uM3_^#)WU&lEZJIxwIOY
zw~Ypg&{?L@b4wO1!yUy9WlF~LPbKpy6bW)aWGWd-<2G+PoF&1&g6l^@bv$}_;IJ2d
zRQgw&ozri3f3~q*Ue6)>IdowJIxrUN3eo5UK1EaFA6++5(?2mq^t~nvRNo1<0I@>i
zZKcAf+g7%w2yQeJ!Bh4P(p`(5vm}hTzLq05>Q*-lzptnFJ#N5IVn_cqYAXK_LkQuw
z=yRd1-VLJ0*%fC8*mJQR0p9{Ejx>#YB~tl;<6!nDNV1~0h`;0olrY~sH$tS$A-%G8
z`*Iv9%0Nl)R*@`A47eTW(F4W?Bs)+vodP48HJHE<%a)ke5tD5|;LT?zwYJCpC|Os(
zi_1$~pCLSDzuWm1#oTyoz?6f`O0p!-$$L`9?jt81y^cKQ01G@*`2b`;?XLNMUJ+6A
z==9O&pp*1nIyU^o9Hk>`1-c{sWA1D@xdMj2_5$5gL_Bu!@?F2`x>Gi;H}(l&%8R}-
z@7hurf&}zQWCUn>(L!R9QT@zA(DGwVbuS0LUz%yhLV}o$J5_|q#+N+syYS?DA_y_g
zDgFn&0Hk82F+-QE`^?AdJvaUSGl=O)Fhk16!-U6|rq0xHKNZ-PfmZZFg>Hb-*9|kp
zq#Ojpcq>yqMf|+Oo)Ymv&x1^2E=wuWJT~^7!U&FKcXQ|i4GXkYiilYtfN(YM?&h+7
z&keJ@VJ-Bv7?;8ebygy2Kb?Du=nQ1`V_lW|aMe+r?^r8=h_ZgTSUP=*gq7H~BA7A+
zBaa&WKBzVy^$^?D6nqq4)B1hB%H@<HxviZ&3i5jA*pMI@JX?$8@A_6H3G_{F9J8=*
zF%WpW`FY(>P2}g^@VrbTG^TcCX(gs!|7e^gxSAYJFiOEQ36yWz4l`}uo_p>lQ~Pj-
z%lwS6<#^7wqM$M148^`x;HQJO(<FOFU_OR0@c4W@_dssM;-veHVH8Q@86`Y-t#tMs
z1rG=rXcNQPWk1i>1RDk9Tzt1+Tr3Kfc*{UT`+^{*DY`3kcNw1&hW0WceiUfRQ?j=f
z%(_^<J@*(nX5{c~MRDb#VbY7JA}fv<XOt`b4+or;rs&^jWDUDvZR%~$CzU<HPKxCp
zRaaAN$@;n^3l;dJ1&z{-Y3fdqja$8YDi*|UsFZBsaU@bV*8XrT)YFWF^B0^v^>|&5
zH+R(G$(ZEAJ}vX5gP39>;tJwBy$n0z_U_+c?;OrPxP+y!3c%ckj%`I3(vzNzW^lTe
zJX76pE00N&`}>kIYBwNpe2G-rf}y7bZ#66OBhc(m1R8PMF3dzeIileVWx?%QbR#7)
z*}bkIy}!W-*kQl!z-KkENy1buORCVg7F&7xS)iFsshRSN@9}qsh-hKkPLNdsO-&r4
z5c^7mQ;W>nu!2CmCQ8kd2-YZV*23nk*Q7aTn?YQ~n{6qaH2wE<K^r59rMbSJrFLu3
zm$n>!d>g6_?WRpR#hHBr)kx`7+kK65-Ia;6fR1cECtOPHAz6}_O_~2*e_6JSlwWtN
zY?+#I>pPS5gQYct!OzY~i`OPhBoXq`yRXOMe=xikxL)RqUv;D9Gy4boIK}XT;6USJ
zP7X1(8C*NB`&Z8r^FJQZr)Q!#hF>CNr?8D%AET2EyB$$IWV$k0kt2Id_ZMI+O#=gr
z9KZW0wlRr;@q&b1VE`Ed(ZkcAB{ydY!|?B*@AzQ&u`)TCm{~SSM+c|NR*Ja}#}>=j
zudUT)>Igqum7!>?cYH?O5F#F1z0_#==_Fl;^$ohKlV0+9Hsv0FEPp@tv;g`TV&Z{R
zSlrq!%mz!o9pp24#yRnNxDlVQTrdA&Aonvr29lVXyEWeKV-hOdF0`2JQ#TcMF?Ib}
z2>cev{0$@#vr1v7+5bF_7p)}2u$t!uG2G0HY^=h8{h)dzz978OTxTxD-c{YM{$Ccr
zmx<Y7>&u>gea<}D9QPGGo_1aPIQf&#1>V!y&`lyz{2K4gh(5VO5;tsb7+YBS;U`9c
zhpkUj4hCmk?j%35sW*1Xllt!7-$7-qe?IEl#1XAQS=6_7z@ZHN?}67A{~8?AX{%W&
z)r9;u@)b*^)Q49*;psvI26Dcj!8os6cZt(&dhPD1aV$~Q_HXQ)7eJ_WSpf)vanO0K
z5cg(D(5?OBk&PRrehz-Vh{&j-(wVK8+wOO#aX*CTvOG_*9-mVnVeK2=X+UBdjuAU$
zPKFFyJ<el+Sb97#>Xa0N!hl7F1cv(A#&BUlZ-io=A}E^6;l&9RUlf^@?k>?lE==wt
zNrr@kBLMI=fu7DsNUi=u2SRZk>TWkqkbQMPg9F*xDVE>r=SKD_B_Uc~f5OHfOcmRB
zHyO%aLPJWo7kj^o&3g~)JVPp>e~*Oqnx8%MQ{bl_jaE^gRHR&8*@D=NW(dOdu=t#8
zMtX$nD|wkj&5k`zAy=BtleJOY)v*t{l?p5+x}(s6Y1)i#i*J36Tr8A|1$G7)FZ+Y1
z7ktc~SZd~%ESz6?rS<X{p1jkx>b72LU!!4Cik>#^Yre<m&l3!h+I=ln6wgsq_TRNn
zky<&48;K_q4(Es~wiFDw3I9503?1on5|*}h>q6t1L2+0zJ|;X}aP5J;IYN`L#i~W*
z6~|k0|3hI3{9!*41>&Y9c(MdiiPjgNK%Xp`yVV;pvyjL1TaSOhd;}B87Vl{O>Zv)P
z@GwW`1}TtHJgyMR8%4|lD#K<Q0Y1dO04oQ_Cg6KK2*=OhvzvLcn(Vb5*xOsED%Ras
zy`I;j*t4jo@;so1*FQidLVej`Wt}MJg4}>-Bx2<mkAkZ~kH~^RH!jB@T3jU#s@$@z
z*-(JhPyDqXTA2QZ;WgV^VbVCx?Bz~u#?->GAC-@0j_(_1yz1gIWdIwsLq0}w?b`FV
zfs)g@H(Bou$_9S8o*JJ->Rq%#&&J5+b-#uC62|fXKBuj3vF*rBf68Jt-rVl8rmc5S
z1(X9dUpfSPxOC#`n$ZZAW<<YU*Y?3ALlsGIKa|-{X77xED@n&R49Q!d&2MvD4%XNf
zN^A5SW%_HJp1(#jPzw8DN>K1fS~avJ&|#chU8$S=-r|b2TGC8S;4-C4b&ifqu-d%-
z1qv6JFB(TBDsZwKkDf(yE<J-tJLCL#{z-%UlI%yUM?a>bxdhco+ggi{t>@QV&-ZJj
zK^HCW8+VxJ&PIOR5EH5!gfYBIYQrfIUc@Is9LFhC#W4DKJY7wQ*x=5FvpHPb#Tmpw
zPl+9w!&^aNaF6uZVTpY^!U;<8F_}6%T0kaYdLd#IOXI?QbTDEtA@f6Kb7NC1!Tp{6
zlkVGkM-om80|y!Y9(;x+GH14O`}DJlrh~(|#pDH-M0A|`VNbNByFW2_4t<rHulWqf
z?i~{)h(^n*anC3$^|Op*V#Is!p;ow4FEur$j+EjjDYhPC+s>8fFkYEhcs3eMW>!J%
zVr`a_n3LN1)eZKh#ZuqBdbPn4Se%=>_pJu9+QN&4ggU)9ZT~-I87X2UleUt<o|AzK
z`tT|HWR%O+lJkP=lIl;sW1b^V+%et5H*|3zfX4(+hB0N6OUC5uCFZ5Av~0CV&qjS3
z;U((MQ>)^608`i4ja~_pk%~T+d^D>2l^8XHBiLA~lVdhuBb?KL6-LcuEOVq&J&w#3
zA@;*#vWR*jyw`a=@VsrgA^~f*aeX=Ua~6jZ5BmLxRk8yb=8NA1NE33&dVNtrHF1>2
zW@#@%Zk`)ZO?$IR2RfIs64aI0nniad(Kx)~f4;jI_n{Le*ZcP#-#?`MKW8Gv5!Uy4
z=|W~TyL6ea{MKJjwG@PKIOGH^YW@8HnCySTA9(vVNBWU_3WfxU_`N>|;RlcDR8Zp3
z`i!eCf8I4-QVPl}#3f0kASCIC^TKFW84}f^v#T`=%tb=8>TiY*bMd%+Cyor3q@<J#
zSnM9!bc_@qUz_{tymXv8Ox(UsXSj()pFSJ1eYo-_$(ZHhUkhkahE)D5a1rSc`-F=Z
zKmLR$?%L`px!;3P59)pGZ#)1fi=bm#*qmbPXqM0nN2n+l#EHA6n1)KpHX7yO2F_#V
zY(74F)Vf-ObZSvf7wfR{)9@sfRyTLFZ}%M$uASZ+VmI6QG@DL~3Ax0eGp^m4|A$eo
zumPC%oK+O;52&vZa*e1<zspqs6?}Z~n{TqFS8H<@Q(?i;XaPITt#7os_vIjt0TW%0
zh@Lp=5j<mksX_KTHWd#nf4u(Xwn9Y;sH4ph?K@#8b%`WJF^8XzkzU3SzKr#9z0u0H
zkWl3Pp90AN$8T*+GAFfP{&^i%h<ZnXuaBcQ0GLlN*l<Nzz4tIuDoo`2LT*iv$s1PZ
zzMRnUu+#Df?Xmy&V}Q8O2(qD#PW9t^%miFjR|-Lyj)0Y^vt`flpp?SNtyiSqeKuS|
zJm38zMxm(33Vx|J6Wblp<4Ld|JwZhm<v4@~-P|Q77>x3@!)gh^@NQK}%&5^U6RtJT
z3Y%_Nw(RwxHq7u6d6~e0z-{}D<^NqYihayJPJbE05fBxDD#`t1Hl9)4MxPfq_J0UC
z1PLL!`*$7-C1pkj3;$zAua$H<H9l@Rp?k+xH1<1-EuC4*;9p<NxvFy52?;lT@lpi9
zeqlT4G@6ujTMT!%`0L!fh0|!?gC3nH08s}GAnebfXJsnAjV=dsPe&f(0OJsdYZ!3Y
zWntxR$`C?u$2US171)edw;4MHO05~OwIxFl%x0%uqi3$bkTc|g(R$;o5AHEXkcb)X
zj{-}WO|7qPTlMHo)>jtj6|MwzceT{)j935ToS`4JJ;r}$+!bw>$U^hi37|29OIPc`
z%XE5`^d&Cq8Hlr%xF-LR&8PT}m+C>-Skj@aPMr;-#Cd>!D>#SVH7+sEG;e~sktgWD
zJ+f4-=jXz+?B$rsol~U`I=?OJc9RywDMlewIZ<n=uy77X(Z|DSwoUy!r-kbI13P2i
zpnZ&zT#SDd8+OM^z9vY#nLGdOBe`M$i3h1cK7bE&N^3nGV5;NBstxn_vlElI2v~jS
zv-L!tfgG=wnq|xY8?JLYTGAg8I2(EmeKt-ItgW~%8_rgncG{(6sii8^)b!=Vh)n1v
zZ!yFHY!X%O%0_LN(+fS1?AFNC4NMkC5Jkt{!dXCEKVv+<&q3$|)7FXo^i24JO9H8G
z!v!t)T&Uukiz%14;O5L?SekVOVbmWzfQ$e9?>0L~K5qmhF0<?gUrP^{sng<8T~mfL
z;wg+GR=!Vu4<z%I5~4K*6X<>PJI>~_TN3CHR~^HhJN#9gbVuDG=>;DeK59HfPtwXK
zY8mwpb%1Cq@X1vkxqkZ_3(!ZD$DqKZ9~^3|a5V!d?*j&VRuveg+*bn#HUMl~J%9D!
z#})=w{U7(fo4zTx?LFPkc{agoA?x#Z6&a3@dJSU+d4Nz-co1ncd^L7S)d?DsSX1q8
z{tG7HnFl+pq|-~GYd(|-yfd4a7p6uu+*L9^9kEgDN^K<^ecBqUCrecpJaE<(@0S9~
zmE$t;ZMj!dEXiSpsFnSt0u0uQ!xEE9|L(?9n1qdPku_H~@d9rgs|481Q)_OF;um<%
zj?egcDuW{<>qh_5DW=Y#EJYCW<jvm!MwbO{2LmRnlVF`7csd!J*9AHq8W1LQA^C8k
z^s%2d2{PsjXT)oOv7jyPIMN=-+2sMmw>iRpq;Q#Ti>ySA`0`4lROc<g3+>&3v}N&-
zI+3;QZ?+T2!w<!O<<Tv~{c-V6$Cc;BJ|n-&K4eCT$BfiZpkFjRB&5>5V`-I5{unF}
zgMUxf56Ed*TMRrwhLb4>>No|G?|EUd`Y3YpawN{DSXfxQ_c09@O&W-uq>oP_$%e(i
zDkX3+V|+p9wP4>ni$YL_W+#w?d9++TDrC)?6p|1`=FvCO?OyH+Wacgx2QkuQSaiB<
zb~_9>ybFKW$p`_CU%l-PVC|=@!i~=qRs?R3yPjV3=*|Wg*@VdBLUh>BdUc#6u!zhB
zLhA4B$`tgP6tk&N2cLj7VbD;|d%!+09@$aC-L(vwXRx-^ja#S%(gGW{7E@}>|5?<;
zp7<4}O%3vgZKZlH`L-n&bV_dIJ2TP#a>L8C^0X>@9)D>ZtPixq2b^h_2utJ$ioZET
z1wb~&d}Kj32rbZ#bq0QL9Yfg~aiQB|@f!m+i=Px}(=yJFCFpu1i0L5;V%V~zKt*3l
zn{3=0S?m6(rTLg_tl9NQL7%|8<0*(3iYvUhk7zgqDwL@VM-U@(`uz#U%m8G=q-c}G
z@8RCcaVe=h2<#^V0M$9VIehn*{p#~Z6ck`qQ}U}9C%r+%$tvpQa4A2$`==4i!zIX-
z0GI%=DP<5`2E>cQMJ6WAvKoU{@Q<^{%E9kLuxa<5#ZL7Kli?*baKBSC>CxAR2mbv6
zwcJnvcz$1WGzDlWsJi<C^t%^?N5XEXjoRxWPVd@%y!}34Zv$l3R0Oo0oJWzy-d=g|
z){MsUeYB5zOdPb;BNx~Z|MEow>*x8T_?J-^R&=iEvp+M)k5-YTTDFSDs!^Ydz8=w$
z1rvvu(s~3#RAN_OlJIN6LO_Ke;`ue3DLyBhlMU^PSKM=OW>&Zr0uqTj<?MrC^y}q^
zd)IE>;|nZz4bUH~tAlPN>G^z2rX|RBJh_jGF8JZ^I>vipcm|oGHr$J|e}j0m6)l#H
zz-tSxb6cIXE~Q7vWo$YjI^%Yu7vLD7I>eE5wdvH+^oTLHU4TW^laX@f2J18blaBBw
zR`Np%EBG!w&k<R_!Sg24oC+Z%ju6baD0X&#AQc-NsbF{SM3e`)p$8XgxpKKDJC&@0
z#&sC}vVD+^iRI*UN?^wqw@V>v3DfEd3<?RO&Rbf?_>nJnThUH)HH;T~-6bX4FjnH*
z&@)eCGc|}G5f5;Q(|~1c2M>cYt5dn<Eqqqnp$D5;)}$7$Z{L3rEPRQ!384zj|KHYy
zD}lJCu5My3rq+~Hw+yp6lo=pZCdd$jq;X!BeM%Z#zBpQDUGfrMqOAb!*pLy@PU$a_
zQc`mv=5`H5Jd<&k0jPK{8@9yW=Omj>tw2~%Usm_J&D$@6hW2A9m2?cdg`hJz4X;i=
zj(>3TUtkbUDlTuEYtuXJg;ci#7k3T_&2p*B0-hLdn+WgVNfAOI!qn5{VVBm-L1scQ
z41Yv}&hd9rUAZ4_BydE>N|Xgy@G6x;5oQ^hwlUpLq9tM5W4{u|?GbJm?BM~jE;YOl
zT(cJe&e7tJ*v|<nvKISds1Hmq?vPSjWE*__7(aOY)bQm&{8jKdlo1+zIzh)<VfFrN
z7VNezmOVzT(8Q_D?T{jbUW4`3(R)KPvoIpNJioY~(_u@=EMeR|0p3K<FW1ustDMPs
z4wa&zo^dw!x%2t*|9lF7Uq+IeYB<OCninCuVLXPt(V_59EeMTz?JbHpn81`|mUh!A
zWG?W)-iLRz&Vm;JRUkV%T%`+}-LmuR)ViBTypD9maR`}Ups#b<Ra;&S53pdQ!u#bA
zDUyLH5hz(O2h2oPaq>XhV|A7bEc2wx!Tl+qQ7DlXHu4HMDBj;(?1fT&ETIx{<2Gpb
zliCQOvF&7Q*abSh>>Ww}H}c^vqGoG)#LXwmz7+1Tj8jSt;#Fcq+rIZ356a=lh@1g5
z?)>Y1B2_nVaWjbe!KHeXemP#t8~*`fc^i{cqoRN%d*_{IjqftJHiq?1gzibcO@=;B
zISyBN)9;7Cx+JSy<5D-JQ@_u^<dS(hKVJH@=#ik|JJ@s?mtlU$Z^fbNcE$rOFc6#~
z`(@N(NG~Jud49pp9m;us*y<mL)X7Q8P9!HKLWl)D;YsIuW09<qcL4Czjbe(An+cja
z!%tpg&PX>f#Er$53@U!^!@~b9rrYXG-E#@B?Q3qJs~fdqVIyma)1%)b>`pFVSW=z2
z09qi(l2WJ1Ki4sMvcYiG9<t&5@dof~ue{*1=87<)3f5*Gnmu(}M%vdKh4eRLBU_S^
zcS)?$4tzY)k!K^%$EkVbPG7VxNeaX-eC=ob4k4l({d6#rT(X*8V@$Gi*2o>e0JFUy
zH#AdU9e;>83cfR}cwCvBY8s|}ih$I?-8V7Q0p09BBIv*LwTtPa3fR0a$CtY_OW3@3
zd_gQ`+EVwlk2~d*4055t?k~DE{=Wks;)!@@IcDI)?*CdqzO{~(Ry)Jmx5-r<<O;QY
z8}+;99renX1@_&F+~S?&e&b8@PtngLP{*WuoIzf>?g`P{uQL7lcR1hU`f}Kh$T3-t
znjafTnzz_B_{;{@$Bd3W$4UX6J2wubtp{Kcr2!Wt0gMbwwo<acHq|&pN4V6c!Z`F=
zE}izKhLsY<BO7f8jNH^sTfFx?rZ5>y41D|_F5+t0^DHBH#sezyX0zS{L1y^(bY902
zX)OO_gPA^Hc7@6W6Ci1PIy*H>IlYsTy_%$%7svmC$4Ddj-qNyDNpOCljZK$~&7VIy
z33vU+AVwHBq0i~Ihx=`_q}rq#w>kz%EPk!rV6nHZ%zZe6o$4pTB=*%>TH{lvLk>Gu
zF;C!*_p6ZnM8f^x<!KOy<p{TNO+@Do>6!TVOM1EXd!<wu*3u9={Him;BU;#zDAd`n
zIMy?}?vo?#OX5*xYc;0SjOLg5Kw)e&8?yd%j@)XVqS*ba4999ZEp6Qp`_Y>6E|$63
z>30bkCQ&E1e=<3$LO)Te^k<gFJ@Gw^7nhX^X%MTCzz|YGeH%G!rA)c#(^ZB@Ims!^
z+_~X0q7BfP8<UK;2(*VaPGb;vM-k#en@P=neP)@SU$|kR(Wg>u1rYatQFlZXzXiHr
z(HAZ<Ujo^F0=qFeJDr@9yKEm8nv-}NMKqb{q~)B~74M&%P?5)?-&?**pBFL)jqq?P
z2QuS_zEl%oVLmX`;FLVzIgsVZ`Vxe5J#jcZZ&!#kZT3s9zkb>^_Stf>S=h*ttV{wh
zMaH=z*Cl%3ote?`*_FS%Tgrl+^cPl{nHOoiG!$6raV+{z=SBumF+<1_^yLRZubch!
zz|~f=EROrw`!sL|<6D;+Vke59*i(x1KNfD9jCil+FY#jEs2>Xo-|EC5tt$t)L^lg%
zvQ|&;S1r{RFJ-<GR}xaXt}p*$XILTK^{V-%!>w4AH$qOP(qC})KsxGQ=L3WcA$a9j
z^_d-Fr?4zT%gOW@z*a;EDMe<CQ?gax>Aw~ZjBpz{me-LvddKAj`}nte^R)XitTbBD
zC!Dc-G$q|b<sCGtK4YRqDCs{t9o)hxWuqeckG2R1tOO7M0wA|L(>FvOJ@`sJTxEE8
z@%JFq{B=!+g3P+>8Baf;tmI-D31Y~XhV|OIA1#u`LT_SC)(<60(+D4tF6R7C2>2DU
zK?gtpM;yfic!E<T3L=Ba`w_2Gd2%GK34>Qf{WqHFCMa=O5AmD2m$IXp`2&Z4ySTDZ
z=P(|hD9Q<bkdNBuI>-5MB?Mn33Oh{TbLMHVDhKPj%6BXV%a-5-IjQa9P>To$$<4hz
zemTi+W}(;ga>bBY1?{?45vEVJVe+wPZa8OQSp>)J&{lzejwP^r5d(`-@5NEMCfrlV
zEAl8{xG49uJO}x<pr$|M*2%(?x6nm9ElK8}hd&~ExFN-wwsSIvIXJNQ^{gAt^nNf;
zw&-s~=HT{9gE%l`ZD1e(QHWs^{XO_vY=F<Jf~&8Ki^_%%D3P4y)c_gugP&-*s(xwX
zql3cTOcM~~DlO=L9S-auiOzcptvd=c!0rhh9tO!Axt4+;qgK1w1)2(Vzmr5UGN3NG
zgWqL8DT=B1ivk_K2N|jrG1T`h4UKw}MNMd)5iixW4DOs81#Ztfy>KQ`dZN(p77ZP6
zrjL}q4ut*(0^wusOyox~>o>g~#sWWI{%r9EzHo2=Tl`TTNxAUfmCqcYHp1rI@oF<^
zu*LnX%GpK<y$PhNnatH#61mtfI#0`jw1`}=n5qFu9H8VvXz1zlJ%`{+B;HGnXaFL}
z!kbP3S9)2SYXhNSPX^In$rgK!;O2c5p~8g1jI81#hBY}7jXLG+eou*HG*zUWl##}F
z_6&3TjJ?}^xFA<UOwxT^hwIpC?Xq7_D}#D$`a2h>y+jD9D*A9E+b^Rm9rh*}Iqbc+
z1Y|p!hCoP&ypfSnp7JFSa}EN3K<w^Cilr6<@@RfKNm@!}?lc|ew%~TUqY1$9FP^Y(
zoPkh9#C_}|ikUL4%FeoFVD1qB-$xFRu{57?X1bOBn16Ij3e;{PDq;Nd(tQsTkuVg;
z9$b+f66TQVv&{F{G_AoZ4+&*kXL~3GehI<A2iqU|p|~UMmF|@T#pee%v8IlCMr~UG
zTfb^8+rhJnX#gR4&9kAKf6sE~b0zD;O&Qp3H%C#gltx8xcl()!^i8xi0T?bw8h;lR
zZ)AtD#I8s3D}k(AD!A|DAk#ZP-9awZaW$gTn#W~di#P)=D3Ak+wWk<~oMJp6>X$__
z48In-=z_ZFDXRobH8#yvrh{@I(glXu|BWaoAMo&M;ECD=b-%dWqe|t1U<0^#Vc`f8
zBl1uv@BXewl_r%cec|)JJo|(!Enu7JbsAD$ERN*c;z~%9Tq)M{#tvi-m;mXtOXkfs
zaKhJam#TgXd~LU+2lQgQF8v_jkpgHe>)xCnF6yF_3GB=#Ee(MG_X;2l@xj56n?{fn
zHgO)`v{#^Aok#2MhWRZ!n}+MW9T&9%I$M?dGjJ1VmB;y=%WRZyiZ|qv=HMA4z*|m-
z5JOi}?Q)fU^%67A%IMYJTV|H!rH@s#B`rDoT<#v;U<_4H`ae8<WmJ@1+qS}pgVX@h
zATyNGjdXW|NQ08nNP~oQcZZ;)Qqo=0ohm6H-5?<y-yZMh{eHTZi#1$xvG;i%Y0p^x
z7w;k>;FgYtEdM(NdVH$m$;0!C3#}rpQt(sSMSoYl95mZrSHd&M3yoTa7X|#S@jM&g
zZo5ghmI#fBNkMAj&@hQ#%MAmG8|~ANUX{CZbt>CGi=P5`q&ACQ8M@?41M&%^kcK#<
z@FkK`WMTh3_#V7WF2M_L8*QMk|A97;MdmF43Oag85T93<_qYtz`+k8b^uYCjJ_nsF
znMXAEu96=<aTf)k#42@IpxUyWUX4}KutMMo1yXv%3RMHGr+a_P@8T6$DyBA-S9_LP
zC16A>`W9{Co5{?2=AB%8tXVBU#?lA$`4(q)Kp}Pn)`+>|cve~HR^jI?t1~f{_HB?N
zS<Wj4>t`G)5&~#}<9DU!4+JA&KaNKPRN!$R9(KG3Jf}Q>2)y^~n-8TCXSntzb$0W+
zK8quQ+G_r(4?kI@fH(K;F4aA`LLI{R)V8`QS!&he!`Rb_uXZ|DCXcxr;_JLQUKwW8
zmwdb94N)vVC=f!gbg^HT`20xl**ibvd#?e>`jFY0I+$Z1;N-&O-UYSn0%wVdkO4?-
zE=+u;CeG4&Zrpwx4Bh}Kcw+Rm=m)sp?I|$Y&MLXL@w@HY^e4V2YfaF4*dFj~Ia}f_
z9X0h7tc(UiNWJ(KB!bhqEd;g^2SC_0H-vV^b3^j-@}>p*fhAlHd<WivxT9}?)pge)
z?IOenZaD%EL^?BW9?f_G+NSg~4j$GFS_<d~M>Y#Hf6jDGZxJSyP(S1_?*_5k_CoL7
z6;VkHHJZ!a?HX>8is*OdEw+c3@wuVwryvx`biib1kD74@SI4r%_J0sAN65XU0<uW3
z*!x|Wtx^fkT}fWpobr44TyuyuuXm>HtR6%;9L?r4V|b;7-pB#MkI%3ueUYltH)Lvk
zYB4oS3wcjUzW#`y{&ki6h=|Ye^Gv*rICUK@jcm~L2aFG5Ujb6JDuB~q6`IRSI|PC-
z6%-XCe}RaV<nup5jA(L!`WIw~RAg-uvd0=MLgt-9_(R-grk+t}kJUGZKX*!OLaF@u
z-2X~P5-;tb**^_*04k?scq2tzTwL;4!@;e>0pIjPlHf>wlU<lOCY5X`eU|nc_XBu@
z{?jb&GO=b(>%flXzX4<UZ&v+}n(9?wB*Uabw}OZaEpU!0QK$U1ko(=1OGfS<xcea=
zpOhe{0nyixAx_;Q664D8SjkXb9}@g3qMEPtYvH|por~nyb!lxrG5WU)iXOK)`1z6a
z32L!Xu(`hiE5m<f6#9w1R|!@Zb<31+p#+d{vLn1SvB&6qI}Kvk4ThnxhG0X`Ig>Jq
zN<cxu4D_8zM27BSNYO9SyHZBqd%1D%()EeigAO_6oIwN0pWks?POyVcMDY;T`^~VY
zb;_v>3j)nqw`l0WU3>x6@4^_f<St!nK}hwhHXBCc4o`&ts--b9M&~7wS!Q#qby>^g
zI``4MwY&6<!h2J?sd!;DoF_ICmWMv(rhzSZ`#X^p;^Mf)4$>aD;s%+AWLF8&b*#^m
zFUU}n$h1CrgQo%hGIF3WJKf>LZ=!2lB2mp<FGw|F*#lJKjp=?Qnzp>0I~xSqpn516
zY0R6o^NF3)#t@s^C{u>bl;SC*$@3!e+e@W%H1hlHiBW^(*f%)Sa5Az(-Y9ZDb&zwv
zH26ZIx@U&M?H8hM;axFSf+TM@$%XGK({g5jVE&uv^9tu3_gi=dj3D}HB%X?ywCjZ2
z)0z2??{6C@7!gTdJR5}<-|aAyb)aW=pvoIAAte$%Plv>#pG<|l#ayZ!sW)s+(|@>v
zfDjyf1}@(k@(7s&0F74jz2Dv7=k)<zf!5>JtNo@^)uS7rq%-XlK?EaDnPBaPW1*8V
ziCP-y&09f_2*V|^4vhvvOh3x6NnWbW>9Mv)!Uot>=gnYV5~q;wxu-@W!_Gq_IA+&;
zJBYccdyj-0an|X;wy$Y#i+zXvx9LBruc$8MOM+}m*LAH`e#F3=Dc9Je?pzzW=tR^I
z(9y+d))P)~!BB#assH=iq4c+4N8$^TQ?f|wpH9dZ^OpMU%N08HndWbvi`gVE6Azz6
z3$9h`lfiEi279mg9u<HQu8IbgHIcKAOxiGKO7$l2ZEV29_O3t69#{t^Kq$IP_d_{z
zD*Qt|8W!oas%!@wT)S2hN-e#YWh=vh=k3g!_~NeJ1X}0fBK?+<ispT>h+0DU6u5uv
zc4(QkzJ8!pdj9Kez~5*pcT^8XgT`thERQI1nYKWe<(EMk{0j2&@kksUUI<*oM15b5
z`y7U2>p?X9#mEknb3qA83n5dXuJ_*<7Z*TJF}s!s!LG^SbjNSGMmHi&kwDjhU|2oP
zNq$5JdavqKf|7nIG`B&~<3j4+-}wM*qW!P@wf@T3i%$LzB&44_Z?#_+@5l6Ti(L{<
z9;}+opT1_gpH6uTZWx@NtrK$a>Vj#P6h%OhB)Gnd+KJC`52CZV8FoN4Sa%%Nk#%TC
zm!Y55YHKwx|K@qTx7=O7)tl_5a%)`<x{1I{!rec!<yUIX-@-+OsePw09;FzraFZPQ
z_6`M!zdNI}y*X1NGwZl?sjZfhk@(%HS&U$RGCDaPEi&tqC~^o}c99=#1)0+d!hcH4
zdg{&Z$XkcViHdCha2?Hw3xg&OilWY;6o+hig+A6_BWV|V7&jZ9E_y-=*hiK2br@)S
zsWo#2{{d4+cz*tWzWK*aSK<c;uDNM!tbkaI*|qR~>x5*z!*mDEDcyrv&K8yO_}#4$
z^7XYC@~&3IcI*;x=syV3`j2ZoLh&1US@h^*Gl${+cl;0|&#A8lrB%K3Y3AnSe29=(
z_5ss#vI~EAhT~56_uo1nX*O7Xsl}_@ws$vZ=@7fU>Rj>+Mil6iFG%>hNDIM!Wj!v`
z028O5f6%=KNMNk$^M30eKq}PnQye6)6uabAjq<KwX7DO6^WiWpAEG1B-vq=aax4^I
z^E?)2wboKtXlag;Bn}#Vwjy`!G|a-clfP)DA}-&GKmGbQs?H*yd8ZEKtGk4VH+~@B
zd|=AUfWmaQSk+f9pUOmyse~GX5qW-H&_Eo5o^o;ruZ(}ybA_SbgZ^W6k{3$-!1JKy
zXr7TJL2u)cs|d+En9*ey#|pcoTB!S?!%EMLq_S2$$298XF>+d8Kz!RJwbH$<iW_QX
zXY7NhX)-k#io{nJes`YKc{4LB*){v3Ia#H~>SrZKV608VQ+EE!a+@dxj;`l-!B8<P
z?S?=vA0?Tsvk}ahj8v=R;}J@BCBufi2ChO~hC4K(JWka2;kfSFRn!aac+`37{hRPi
z9hKaFS53Qwud1?rOS`{Y(47IUrTv4S94Bzc@$%ouuvxn1qj=dUn0;of|D#C56sosO
z)S^dxPnyCB7zcpmcw*Hcc5}YyC1cY^9%Q&)kQG>HYTcRUQfs`xz+^)fKW!P~Pu&0W
zm#U=%mFv3;n9b8{vY;}++AJ1%I^Z3Nq-Ii&XOhYS$BFy&ts=0=0DD{{!J@y51FbRj
z>Iz~n-)wloHJM=9eLlo#wW@8B$o-I%wFikY%!W>jCjS3@7s(G|qEl7fxnsI#lX03y
z(gwG9o(Xp}y)E(oJXMMeHBZG9Hz(zoZg6^|(D$o`--T^86%h||6{E4__QtXUO2xPU
zVaNyIiPAu>ELptUmF4A<L>@#*V3QZXMm72NfN{}bzCb63qet@b#3&$BlA87#fd%V;
z%HfVez_pq?HMS31i~6iIT#c98i*J#cr!utpM_tf|iE@9eI#XHJyGq4{Yg?(-80J(|
zwiuzA+UIgSAk$<Z0Ql1WofTk$Cq*H_lcafhvu5YvkT*ulQG-O<RoQUeSL0+#<PX7k
zY>(e$pr$RZ3dY=LfEB0+JNR+Yo)A*R-tn^9ezGKuHUwS;GRd-q@g@IR%?HwHmvY_j
zH++O}q?|=U&Vc;&sfy6wMBAQ=!W&eQ7O+K?{NTK~nrmHs++a4q#J98aarJ@^18b>&
zzRUUUcx}})vj3*ou)spOv9HDihgkgN?P!5xr=<(+&%|%fj2-cNZ!0vv-n_1reBA%X
zA<o(a8WauRm7{-XT8b_(Pe9p<56kw=k4CB+khsJ(tnW{l2uRPm>ey&u>gZD4&;k7i
zZdG5AYx{lid&a?Pf`lgOP{4+R)CJ_KdF9<7Pb)6YkH||Eh&FYeC1g%#Ka^A9@B+~l
z$)f7tw*Y!u>y7U!@44Z!N*-7VEcf&7%uOW5=T`;8m4mQuhe54@nMJfyjoE04-;|QZ
zR!NKuK@#mr7JoM93X&tq#|9>`AX!0VD|u7JFLuvhB*v;f++SVXyuwT7NLc9$vjzHL
zkJ(O;5o!O{EKKB=WY`SdIC}nwtC7U#&ifL{>vs^{zisPLI75aG3bq6P#@ruAS$L1I
zEd1==5fryL$Je^e=f-Aw6ZoV|>Q>}-0n_}Kcw#ba4(=>9%@Dj9wKHY+=UU>uJK7<~
zvO^cDoSM!hRbHY~8!)?^lB=m%TF)2=cm)4`;cObQb{Z(@aanee#TmH3uu`JKGwHs^
z(aikN?029h4>t<82-c3xpeW3>!Q_4AhnpkQG}}aFH9?_dOu@;0`Ep`%g{ji1B5(v`
z$K3011m#Uuji@bP3VA?#q{_B{_(C7{=NHJu@d?CkM}{OmV3#l|>AfU=lt$FyoWPWt
zZ4d1H@Rb3@)+?j9^o_OXnXA`i7qP3NnLd|W?ZP!#D!9-7K>ysGgFUrG{h6V-q?Of=
zFGl^Ov@D<p#H!mIiX|ml0A}9_pk1b92~4~Nh{ObF)wG-joQv{%cRV#76ABy}%1NLk
zv<)i1Zt(@)p!n2O;zhnbxh~!U$uK-u^evhXKdm1Q?UnS_t(a*|#2NY3gcBv)Wx5pA
zd4zr?<EZ3R!%khKdK@X(&nP_{f3b5m9sB)!71dGYOxn>WV!PMW61uXRRFoefZ;D%$
zgTyKyJi4FtVs+RvYRXg=3qR*}l)=8~TK6N!761e?1&AZ9_Lca3aRrPIIzoQFSpM~g
zAWyU|p%psGJ|0m&kIWeb7U-H>(dOuQT&AD?0r$8MJkE*@TN)1n*X!sLjnX238DbD@
zoX?Pj5+2gmDjg9BL;q`dNNXHmU%<eCjv{!iPl8UwGY}U3nBI7BzCHoZ$_FgZjb*L0
z*<bdh4|q5lR)?CfXS&?~#P@DSUSZP2dMbi|y?OH{=iaHZS0cl%XZfKXzq`^a_8rM?
z4aCVak{MU(xHZ}c<4JoG>Rn~TWmlEOp)x1dwDvcx;ysG5tRn{6czx1_v>RR^ai%aj
z^q1tv=K4Kp7Ta_^AjTR>|4LuMcol14wwOxwEsVIyIVe)MQr{grQ?oSu*Y``vl8y@#
z>9S+!&qHOu?066dfh(i<s5l54MJ5U#<jO?s#)Rb{O0r_hPYg%IH!7X~<qfhxfr*e$
z6TI&v!aszWKvh+3$lZ4T9&PvNl_o!f`o2xt<Xq|=K4eMMkOKM~i@=C@z);`Y@n3|l
zI}c7;+N(HA#@fN8+`DQQ{D6CMeVJ4vi5>9g>}pQ<68m-I_vulxeMRHt5;l)l%8ojk
zgDR(*1$xOI4r&!o4OqUqeP(^OKUjlU4qK7P8wRkn&D3`cJ~I@;4V*YJcS(8)snwXO
z`>1HseOaKwLc4><r~}dz@dHNno~7kPJwwy~VJuWaTx#b>p~;mI%G6xS&i5u$yxBO3
z+NORZ8SEJmfEOa6zZ<p@P*mdCH1-#cc&p*x#|v&y^Oi?}8SU3<jF@@k#ChyHmf&y{
zMJ6bNTXY~~UO6s0iBUG8GJox4`{n8iXOrwn8)x#?6JlclxDoe2SGsCXCuiw}kYK1$
ziZsF6PX#?~Rw}-P!MgnfVSB{{tfm!jzK<6|Bd;?l@l@=Vr4uqjW263qHw_`uOxje|
zXL~8KDu#Snostsd_L=;Hb6mYUfhjo@Y`}7m(dh={FMS|2J$lm(%;1{N(cTuAfXGo|
z0Ydk=XwptetfT0d7ojR(j<ekt!5X^<&x!p)$v`rc@mvLI<H=Z*HANI64ge^JyK%DU
zn~Q$f03u%HLFHH=x%?ZL!N7KQ^2q1+cenR*lk`0e)+tbAXhUO!FTcQ1+HLMb;&9$W
zKqk;&lbYq|zjN1==>GBg!WTt)kL(wE7b2b;_bH`*)*Pa4We1K4Jo7)H`V{)f_kjRI
zj{gG!++Bl;vjx?-0&QVjDb;dgDYZfdSymMF{JMKv-`ws}&R7y?$>Pi9T>JS%csK=|
z<^-#Ter;r(DyLu{DK<^2BO;Zwtmt~=okL?KhvYJ*n4X$keted1a_aePSGGSftMl`q
zzl$7-C4(JG+*1OySxK%^hJPQ@_giR?0t!lnpyiuyP8WP?07m@iLMfjaj75e-PP|oL
z*th&mZU(?o>k3j4u9(i%?-P<b@4gb{Ix>2n3n*T~{9|(QxgwR{(l>zfcwP&{d`A4G
zKcRq@2cY39t%3x6tD3O9K}Ubq4=JX?M}5?_@3}rQ1VxBOL*gij;!qr>rRF2<5d9Tu
zCdVr;{^j_Jg4l0}6><C~RHyySb@Rz`F16XOsJ(oTQi!#qd}|k(^eydyiVlLko2jEv
z0>gG$emG<E_gE>FY6<RirdagiHwE)`2cul;ho~TOd4R0!)3n%1N~XJ#AjP3^M9`<_
zyqN1*slJWRR;$-a2V3vlpnwQ+6(AW#j1R~q1<BD{mc;IU4O)TpXwd6J9BD$oq@c&>
z)!OKIem_TivRi0y`V3prK-#B{VthT>Hrj%kr@FXISNq~Bfy5<%K?P8%Uui1S`yM*s
zD=NvVQ*KH*sZziD^bPh!gHC#&TP8$NmH3)EB>gwH)<=!0RT)zb8jjtmHOIR*PEk5G
z(PhafozEVVE|?D~P3_5Yt1UTq3q0Nr0VBz?M-X`_DS48rSVGxlD(XCA{kQ&=uamb#
zo&s%r3u}D68zv-VWqv>KSq{|pgW~{wt9}o8Kpzr<+ny{PoI=69<9Zpn0<MF<f7jIa
zvI$-ye?hN^mDp`Btcy6_C7CLZWAn3@b>QuY#etl^U*-PWCxh=%B>qYp<+Zr~N}U4l
z0V%>o@o81j6c|mLS^)BE%a2YNru|r*{v8BwCD+EVBZsp87B?^L9`lXTpY#nDD?2Y(
z_qxor$D#BVw^z|FhnW6QQ|mN(%7^p{8BNlJ@1|8Rk*e(Fux}EjdZyN{%3R;^i!>gQ
zK4aqHeARpK7GMrYk&m{c%spPnjz}F`2!wNFn4?&WD$2;o1T7*QfrM?A=>f9S@h@wa
zOl-9)vu<_y#>L94(c&D-;`kH(k#A?a$Xp(O9Y*i>K*3h)?LQy8L7aU9M1PZLZJ@uo
z5i7Jar&fN6I+j{&HKaXC5-O%JQU#bBz20#k*(AKRsF9I^2jYIiL_N07@fQN->B)-p
zynRUmGT{8JwwX}}r;j;h4o!emFeZtk>JL_%Yb0;9BN&a^;KbE?kOJ-C4O<D_Uj?Xc
z-FQzrhy*V+A4~EC(>k(jJ3cYdG&j*mc%R_Okk?yb8TWoRpMMQOd@ulQs$Ch}J*Gg(
z)B2+J^^emSTI2GGkY|1sx*=!5d(Ra{oe8vJMB(*?{VfZbTfu`OO{Qpjw)8IUxk#Sz
z>o)a&PdDfrpNn<>7I!h_$z7XK1H_}I^Y&M3yC?)}5riBfhA5B_xste6sqUKfC2)<y
z2H=uU4M3zfGUfbeLV*?iA{t2EEJx!|@vIT+<RH5|hBh9Fpl_7JY3R^>LX<Z#M5La+
z$#h9xd$jNAItM`P>2+Zruv_h-@EykkndW8q0<cK&SYX|M`$yz@w`#P6zNpN1|1l{F
z?NA`w?~I5SigXv_<)b(czH@P5LW<}gZ7sI70k@^ZnCllX`~}|YUt*tv%TFW2qKGuy
zW$|0nn*9Wu$CnR%=^)$~Y}Z*HqK&Gf`YOO?QX(Wg-tMlgj_n_2DSNQc67=>vXy~Xg
zXrbPKZ@?1;|FEN*y<`<w%09rGOnCDLfLw_{cf+rs21R5S^^1gd@|&;60-wV(>&r|w
zBY!2SxcnR+dS?j7*=_>&+>6SRO~OmKAVm(}oMm<J^Qd%7@58Ytcohdg;~+QH^U|hE
zMdi@A27_>IIVURYtwu$oU5mC9{1@zXtHLj=GPb*JAykgQH5Eg%s4!E0&tpye7Uwdq
zVpZjBGXK6~btzov0|+1=M2hQ?tcpEv&yPCE*Qk~+F!f&FR>61OHpklccPPOf%uDRY
z=d|7jV!EG|OyWAe11E5+cr0&qNT6G=yI3PEzDGUTD>qhBk#V{b>ImvtbU37&cpZ=5
z$6k8D$eR#vOZvv2+HzS>4oR~J05;8r<bZNW9-VAh0o(JI%&xxOb9O(WI$2OUvXc?X
z9l>a^{CR<bRuH%A5QEcnZ=^+C_-o*YIrsoXkDV{0IcmqJ6&4!#%;@{;>o>Ag_tnF7
ziN!2)=dVAF9v=PKI5q>%qbE#iHQT|-Y<P0s(V+KTH_scY!E!yf&twVT6$>@^3$(8+
z2>%{vTxmfbk@=<PE1-W|rb-2QA;(07<8pf0cu%0Kvarq-Jm<u4Q46?lGAv2}rQd(-
zB0}IGx{4!==gHiAMP6{^S#yg!I9w?~5fRlZ)&4MfLR+>&^N&s}jVZ_YuIO(JDD5!D
zyHb85Gb+t#$$Q`UURn_9kOXnzVGTFMOY-6UY+U_oLFq(IBHj7ZQ$M_9v1Rn~v)Y?K
zlq5%c1c&Mjm#=LczZWF;hu<5QuuhJ74#A9f-`sT2u_hXNUcb#r#FHR~pVBA8IVo$S
z5g2*y9mJ&D_9Ba?xBsIydC;Kp;51`#q&?OBkb#~eo(kSu?2~i?KWd8mC7<-|qz-E0
zSyX7f0547~d?qhLx(pR<W~1oNCaqk@BEq^a+4NA(>GW0~yWk5lE&q4@9%KT|;Y5CC
zvUINk)Fxv<u-fhqy%2EOHsua08DchUp2lktU7tc$?%CqKl9r{@RUFbQj4+Q2;Wy_x
zCppKMMBQKV0Zj14$C1-3);n`^M95<*mxYKI6RQ=<?O_Yaw!PiEaOYz4i~3lqFp{3c
zqDA^%*`{2^fhfw6>UQe_=0%425XB_+n^|;R)rexz*t%W$6bgD0^(y^lomKj;vYReo
zYyK7&de4r*O9Nyj(S|O0-#MQk0`EmnroJ{tc@)?AY6@3~yf1}G$2g(0v8ysQ^A?<!
ztsML6yfJ*wPzpOKVZCoZ)P`V@(b3b>`@c5^97!`6JTwzzB`c9~Ut~AK)R5R563S2i
z+KA6>^`ks5<{M{qjBA3oBR*eyNusa7Ii<KMyAUn0vfb*O8C|#;&w~_uiSG~YkL=s<
z5S)x5aPRf&2Ge|1k*}hED*j+bvj*SHxeY&m@!|V)hrJ5)wWbz69mCeVAg-^;Z&m*U
zOBinAYTTooAkcGFFs5J0ApShtMZGz>o&O&na0*=|&ZB!=rX27iaM@s--R~Zwd8)Da
zANWg1zL*5qBKQ!!@b=6{5`9LSKV>JLO8x~q>`?rV+E2Ylr$f#p+|N7T7;ax$t2g8|
zyP9uGE;+C8tJh|@77UJ5N5<wXdPYhgqTx52k2hxe>oAe_d}gs=_{`0)&~AM`Y{cYC
z!eahet*7kYz!Hw_=MPnX;jEP_#E23i4SywOg*{)AjD$Ku?%^uPL=qbvpT#~2_6k_V
zF;(ka{5<-TK9srN?&5LHi4`I33XxeTy3H1T0*^`~Me!^iCHz78i$cZf!>0st9CVS{
zRJc8PCz%VjYpp69eXC}AHnGVvZK<pZ&-db6DPn4A=rwwDr&r2b!_qtz`6EpW^?7>E
zL0p)M(IV=e%KrctMwBA4q3{qNBwPxoD(5yU6Xc^;qAOdVmU)_-7@`s#N0&tHKOMkB
z0KM<^i|0Tdqwo`3Z1(8gtx8r#dEa?%ay4CnQqj}CCat=p$4U(L_GXFbnlD@Q6=mn^
zMBCVfzCy_@p<mDr!JGR|C`23{i&k+i{fPIm@jFBC5wOYpjvQ#$UYaQL=h|CdoG-i6
zhJ-{}?32l|`PjZ3=m}2zNV%RkA5%<U_G)q>T?oJ2?DoCDs(*WenWtl8Je89t=jm=-
z65Rwn<kf!;dw<E-VC$&t6?AGAUsw_4c>7z_V=F%eC39rz5DEUvM#WL`f!ox!vF!SX
z&<Fl5XE;k|_9{^6*HqBIJ?t+94VuG^=zz@}(OUC+_#_q(to>&#Y_6ck25n$A<R8c*
z<tu85&r_=_ZgJtc{C>LD&g0^`AVG2Y+#+J&lA_RYf2VD*!$odVrULruQ7H>n+=B5D
z-B(UNma}syF6>gjZ3`sW^&Fa)wqY2~PK(2h(5ARZY~e8J!918^y2|TZ##c-O+Dn_+
zYWi17;!JXkn$sxKWY5#}+xYr^H8|x1Z7UDU1CLTCMhF?r8u3M0=GbKRrIT4e4c5K^
z_T-mpivP)TKEYv>6bNp7i3l3fTWa)Ravz5mo_qyIx6SElJ*(BDSMU3PB*))zerW-^
zWThVUeScuMzmjoyrH^c%r~ni+h2x@l4x^~FeS9;}|M7+)v-gH>!-r32a!(nZABDvj
zNSD7mN!99mOqj<7%y^$+N7g@zeO~tK-A<_^D7kdU=?71i8Mp(gJd&zD^eqj7dYYiV
z13$=~j;q=N{b;@H-!;}5gMhQtx!ydy50&_#+dzn#zE=Z!t4`q@nFeJ0#D&9SV?HP2
zGCYkf0$JZrLC#!7V2am|nY9UJ6_vXBb8BxWE6%o!MEE)pU+I6zt*LOBnSWOxsT_nq
zVM|Q<#Ev?pw$x0K@oeTdH|t%wWY)E#PsK>m=8eu$U4!MD^y@d(VMU53RJm8+hNCX;
zluQt|y*}Wu{*L`hV@s)!WMXdzL>$guTMkt=8HZ{BCM!sP2pJ1~rKRr>#4I0e5*kb`
zL%#c?FhR*u%GqUz1Q}YqHQ!W+Nh$nmN-%66S>s#W?`T=aI|DG7_T~xn4P?wW09@T@
z7#Idl4@Mr$s<)}-N*@8!2nJskA|e`7%$L^<cqpc8mseJ7=e{!cAg8+lqE}uBKwzZ+
z(4grBkE0-c&xhC-q(>2|*uzh46L!*mn1Ey!g1(#tpp1ah6(j4*y1KhB{@iN<a7r)Q
z-4)1Veo$ImmsR|6K6%G+DtX@H&IB12115R~J%4%KDi(vrKHvtT21ND|P@Y_9;cUBM
zr4VZH*0BG3t;F&DZfMvICv|iDuvwi~*IEANt8Wo|crS}g;vDu$k(ph|aIS4(WXwE<
zWphsCkjbTw)5SlJb0$v3E=Dqi7Sw;bC^2YfQj?ag_X35Woe@>Z1O40#4V@u1LFch=
zJbI%kl24G{jATrsN!bPC>cAHTiWPhb-zO#lHMC=~j>r;F)2j8s-qFW_rbw<La^viY
zDSm=<d}tWJ@L6Z{Ti(@_EdW5A)oTRA1ue4QQS`_yt~y0WMI|92n^J1EvgV_33Jkt{
z+#a-kYLR15DpKmo%E|fNkdnR|)Fycwu<^)1=f#essH<^iy1Wwlj-WKDzebTO`p<A=
zx{a&N2T!0uv!&F>F@;TRR=fV*S;TR{*6Q=v`<OE{M|T^EjBpl$^dq}qlRcN1$s)bI
z2O+3Zw*Pu5{n$i(1wCB`dIuTsArHQPR*h3!`RL@3kv{RsilpH<q3wZIW?FIWpjqeV
z{)gVmv6U6(>UazR%*zfoit-7W<7ssF;-vYRm-Vve8BzGC*o@Ys>B-=>!|4X}@eTH6
zlF86$1Z6rqm$5SGb2}R)T%eAj1y=CN=JVx<WH=CN#b|zq@=3xyE%vUv<0PUfJ$EZm
z`1)B=INP->-UD<cIWWJV;PQFWngC6VY(=aD1SMpBByI#m8w}$;ul6EeL2c-y#Hmv6
zS@VJeAeI>rxC0YD%C9&$I3QSg_(Nt0-@Z>`0VRY4whYUC>p(_T(z9PV$WqmNA#oR5
z_4?5i8V}6q=|2CwTNx$w?2E4y9eqAxp}^%S`f~1HN+nzp_ceQya*#V2z0<+MShm5R
zpX=OsrZ#t{j$7sQ@yQLN(nq_(rs0^a%}HEe)Oqq`ZPn%t_g@vsp`-Md>GNwYn{e4G
zs?B2{OCkeBi^4FiCRb1%eRlnP5Eq#G8`ibCd2_!_KO`38^Q`F9|78IvGR1)^K;7@E
zT=qzp;_BnTo}Xun{3shj(d$@IM7LhmnTd%s0L-nf>l3?*?6<_2ms5O^Dayvq8uYDE
z7zvkWD5U!|n;B$^>S--{E)-t`K=>n{MWEp%Xgvk1Qar!Q_O$i!!PFh8e&)5^cwSDS
zWUj(o>aFp9c>fk~F<52%r9?fOw&XsZvseD>e>^x;_)zHI!3OjWj?f<*e%4S-e0dZD
z<~I{gUM@TR6GfVQD=GC;K3{##a4y_?<il{vNh#&0RJBd~RLfN4**|{J+QqOIoQ(n3
z7r7{WGF3{GwdR$A>+elJA+88?_J^V4ew-I`8A6_}$)e9JTANKb{4rR{XP+x70#E<O
z4g@Q!tzf|I2qtjek$vc2|GVlruo)dbExJ__t%t{39K)d4Vs_+9jllD_HvD`s5jYF=
z9Hl*kVNJERa|1Xc*T1dt)WZvl@UIbJr(2L0nzEnpPqzlqeG*vmdy$eC?x3$Cpl@ul
zW(6z~8A>mgwkxFOQuNRR(bsDs6=JfKNAqrr=)%$G<~dkBLw}R?><(Q}Ta~}P-5AN7
zT>L?Sae8{svE5E*?*H_?I*>mRg=#FylXeOfG)x+3kK`hmbi{*S{glqSeCHQ=;S?oF
zMs+n1bJy2Km$Py6=DE7k@|_)FLeb;gpmVqHc7?hun#(yF8N>{2CKMr5e-y6@ChV7)
zt^|*B;_L`pSp*x?&&G@nB+UQZG$9*x+x^d-cc;gvSWHo?UJNm0zIyT$fOzJ9f=7P7
z$g;~;e!GVp8(c4AIb`-NB0kZlM&MZlC*tpaDBfwKt$YK9mNgJhT<aY}$}NvfN435K
znM!qZrI8OOq*qmkOvb+~sq}Nh*b3f^X7P++-taNQZ-YSY^<yeu@xZWe+Y~su9l(UO
zSc~4!AxLN(1vuEA=oNq_CLoc4WIs`H$hbM)Jt)4O*IkzPsg44<LEG)rW-C7KUl3DJ
zrDi7A-Oi0z>5Zw=Q*;Y^8-BrHUTc_b8_gHRG}X>6)(2=o4QtXJtiF{}vXUD-G`59a
zBWm81)6MZtl*c>)lYalDYIZ<j^>p(PF?8-QHp#=OZnCu{)^#!`>T9zuh-FxE+*D0M
zD(t&YFp4Xzs|ek%hA^8=nj(M-HURxiDN{~V#-?Nl7{4grG%ee?jl0{Yqjxo3OjdjL
z>>W(oAKq)cY$*F%vA}|DO=;{3jH|jIJ3$(~_FX+#Mpi!Y=%^0Z4aD#~veg9UAtsWJ
z#P61q7N$UIdrH0<@fg3_FBq%+xdSLyO4b*b1?8-k{fR<pos=kR$lN+SUD?-wN6OQ0
zaBL+-A}s~Q7I+97Loug*-q0`M(y*v6$8D<%*?fUxV&e1e@d*i~2e}IK6(FHRZ}HA2
z{#}5t@QDJ9tW1s}UFd6yPZ%~v#b31k!{|6~FM$T>_h4p=5(->In4B5bG46bgxF-RX
zB)CSTBh^B{2|FJ>o+R>txu6s_0n}!f>FjMuc=9msx0})bM$T>`KP<X)ImdUKYjd1V
zq{VByH~W}i4Y^~&pgO6Qi&2}m5Pexm<!eriSGr!W<1lU?E@}sZ@`&Y)8jz$_GBeM!
zyiH;+$PZ)Jxq0Bk`Srs9uRpW`o#wn!EV`L4-EZJYSTBWGAhhY8kj%2n`bIhmt|T}6
zFiyIz402WQqsF|SV{A8TNymvuPNXPBNWMM!ySKWpvfk&Iu#M=3ST<~Dk{tClMCChm
z4pjCsk5U?)(TJo-IM4nSn8;JyeNt$l9wJ))5)8hdHet^w2<W{Y$&Y1Sd@g+Jy_6eb
zjuS<>9vOrHZHuws8o&GWm2Jn_g9L|P=i9@v$63x3a;=+%?c3PPLLmG$Ae~wX+V0Y!
zqY*$n;-=F3^`BZH7FXMzD~ulTMQ)X2i61~j^iEUz<qh2&SnSPpI~9&TgWIky!hNcN
z-cGcb#y=zNbJACWZ%Th)Jfnd;5)aUpm6)=7?B|7m$p#SXB6g2lEOq7nw<T#qHMrBe
z)Dh{S6sX_DTA~x4lv+PpjeX5ic3m#{gj7UDPL)UGUV@b3E$3}M8P|j*&*moLe-pSU
zGC(pg^2xuZ#ZReJstt;fG(9pQHjbA9$|9OsO<&SR$z3Owl_s3dxpG<O4sx{>?e)G5
zD)YoC_h+QW<G%9?MZwlp`M2GI26htQ1c<<Vx5rBk>tk8g;*D&2*83%ufcCO|r%B%{
zG4VMK`{SVN(9c|7Xg;3W+xu5YYCC-=(2HNI)_R{nkl{x8SX20MS$27E;Kz0TT%})?
zj7pZyMge6v|8xd0bL7XZA}i=*c893Hco75}417k%#D)F)c4R&fvi=J>ogdtv2SQ06
z8&^9h0ivWoEg_@Qg95gcPwBB5M_wEk&_{MU+r~Ds5g6zL%tFj8fvI$ltL5FP@5w1L
z5v0@L;(wndSYB(IQ%TlcTfy9#d6V;DKS?22DPb|$4mj?c$8l~q(X|s2)9D!%7bG1k
z3jg=JKSP;8fFx`IjT3E<o&%Jr6WJ0Efx$7N`~Hzc7Q9Rp8`vw&_|{WRnnRAORfIEu
z*?YAJ8xo~2_5h(@)}(oNDi`gE#N-_L!K!ulz4ZXeDGdj40qV{Z^YezN?IpvOvNl|m
zg+;z7F$!yD^#s-(2Zpb&QN^>O-1){-tG6l}B#T|$@_48>9k(g`rYj@WqO>RV|Nch9
zMcJi6f-rmQMfcVOvZo*+4lsj1tm={S(HaM!(TQjLTHhfK$YM9BoE#G0ZA;*KY#fa2
zE3P#JB|l!EsH=A0_dcpAP|H-lT@o+UZ>SV{RjSLKT_%cB7#R_PFV1p@EL6cZO^__N
z;qw9T`rI~hmMz-_e>OHTW~Qdoj?tmqHq)xWLoV#M8fQ}jthJ|tCLyXEK5oNcN{h#(
zmYxp$0@R(^%)RwEv=+o5s(`ed7?A~bHE*QH$=j?R8w!3>-VJgHZ*^+K_kq6d(cxro
z73G?47eyWI3>md%dhS;ECSd9l{gutSZiDC0BF~W}+a)jF96TSrHLP$u!<}Zqu!T+v
zt^#Fy^F^ajJ!|{1`BE7E(??EW3al*}eQ#1Tk$WOfW;a+eZ1ERf1P8fKc)h03)DU;{
zoUnlptspZ)nIsS*w8xdIc%TW#`MERGgW!922)Es71d!Y`Ap4zo?EfXn-*p1xd`Rsg
zvdI@Te9_*Wy}x_YHVV3Bz%;)nj6)du8lYDoZGiJMmD(C69!iRX7!V_`N&wEjK`>h|
z1II`bSoY-r1^*^Oj29UvI>!pZ;~3~f88*n^v9;0Rat!5U@4hAwYm&>Njy{;JQ2?Vg
z6H*;Z8v@4SQe?wua@Q?bI~;)>;1P}rmnXPSQ$TKl92z#c#T@#{Qz|NPHv2swH~vKB
z*#3a14?GA{^v{ft87T=c6>9K-sMFF~ZcfgV`0}R1?b9)6EJL>I#d$2P{`s`QL_w{f
zkB`Yt0I?C(huaF{41GT1U>g-9qqVUq?O*cu(CbU*LBWJuhU7vw*1Z-(sZX)6^U?F$
zuM24mp0c~7IsV;aVE>kIbRP21mVKjq2uST|DTf2mw1*Y4tv)yJ{zVf5);$gwt9uET
z8dL2XE)wbtjrKT&36F5st%f3+)q%NL!K<Y={GCYUmxrH#pC%|s)ERWP8rXb%w@ttE
z<{XTFeIOuWZLe;zHk})qm;hF*CW8iGe_=u|0WFOO!~(HhcJqzsbjcRfxYe;U{W($*
zyGBrYGgjaTKqk9-fv^4a{H2z`T5%65jD*iI&)`D5xS!Vv?}>kGD`1Hfz<TmPJ@oEh
z;UyRvH3h;BXl|3B!m%YqV(B=vKL(Y-@=F8$SJfVGg1Ri@+Nj{MqCzlWewZypQETE1
znnL(%UA4lSPtVl*xn9gR=}v-<b7NX@l5g3s1$K|)aB1_EgVZZ0EHcB7%*ZyyW$_mC
zjR-6?b3f=C=cK2L4Q$)?Pk8?^CHYyG$(@9yQO8TkYI%FG%SOXb&)sPiE=QSzEI|&5
z`PMjZIk7W$Q7K>&W1z3FSfzrNH6f~8sI&iv?5z6R)q60F904)S>>11KGbWE%Ac&uC
zfUvMP$LO59pdd`MQ|s?D$zEZ&q*eX+YM`g>9338hL|`;dzLPzYDKOXVx=s!}E#X3=
zS4O}P;1BLl>vguTEEuaGogg8N^UXg^{*d>xgakHwlW`J1=;a{j3f*lU7|y&NJ*J3M
zg_Os}_}|0${9zT)@wM{z-+F<$%I3oZ!gDnqWGTJtP>Q}|n63W>R<Z%?#nWFH>+iIU
zJRa;J&U}xC5CHI|8G0VWzx~X<f@)C<%Q$Wkg+kN_@$J!L9O7$qWZ})7lU9lLIG)Im
z0<NaRuP!3nZ2Tw6Gw64P2vfmbchT29B(liFY?Z@LV&v>J=a?MUlS~3UiqgoOU(X@X
zLYi-<IMUqFG}|kISsDE2we{F1zp<w{->a*+uGn&Ht23cyU>SLRc6_;kOx3>sZtStR
zF!)`3n#6mzDK^~Cy+N!*t&XA+`*6Mu+iFV?<64J(L4Tf$p6O9zsiWV_orn`tAdz{1
z!wlj(S{<&Eao|NuP>}{t19@9RegHziVbX2lS%2P!-uKm}iyhk8SpB43+~0R&RuZ?*
zi|vh1?`^3`!EvA^9HgE5uSS!847Ir4PY6oYHVJb$v%l-}#5mQ1tz|d(N+AS?%v$em
zJ*~%&!RHIti|s)P<r^S*^8o?uYqd@&T-tpAPAiq8?}{}+VjB|=lO>o;SByJ4SP5U+
zjoi2F@(ywx|8eoKz{KzmSm5U#IFGE1_!D9osgb%76BeflaDyafA7IL0^4|D62;v=?
z$IVzCE*U$_Gs4)Rjv!g}fi)fBD8<rA^6@@0l&6+?-6-~TwU2hFSmjp0Cok<xMC>TI
z+o$yA4evl^P9^yY#u<juVHQq@V3;R?d_1Q&c7EKpHUNv^zA<o=B{D9H*m)V$bD+?g
zBltQ!t&Mc_2#tiNdHV7uDBp56y<~fINJ)V|2#~FrmsGN_JdF4)l~A}%sw|8>Q3Z_Y
zyAniuG>q<oY`B-*T!|r^r>9=ecAn6bR3E|}aPa#m#ow27@hQ<16v4Qwb74~3CQUJf
zpxU(&S#mr3?%Kij9Q%ClJOiW<u0&M=Yx~Yh9|K9pV9;$N2-hhaO#AIP@_sGO<RZ@2
za_eYz*9>nZZ8ewA%VAPvBG*TWo+3Y{`ucbJ6#ui-P;&WM2%SiK_N@nn)bAr$Chy8~
zzQ)C{qg5>o)9aENJTd}A%=-$_WV>lvx6JWx(E@KVx~~-D50w=Ae^Dn5c0DTSMk+mG
z-5_Y!tz2_K3S1|Y%!aI5&lf}}`#lGBtbaG6s38joh3;>Ohl>3Hx$3${z02HrXv|1f
z*oeplWO-Q6>!o(W?XcU0(<Ss@eOSX;4VALdx3a5WLL+N3S{q1mZS4?v53=Z8anl=_
zR3)UxVN~^XpX5xlGb}cS@R5$28ut&qHE3Ay_WQ?-Amt?fU-~?f`qs0qn5lX*x8r8Z
zkp&!CdgSoDp{QSi14)N=Bu{D1!8KeHG9&&dv*rhAII-p~7;>q^C)~><obt@_>~>fW
zQ2v&6D=z2ELm=-a(X0LmD*6Z^&dcw{fg^t#aP((oDX1JJ=o3L8Y&igG)a5Xj#`Rjg
z|Gnupe~6vt1Y9wSl)EyW<H>-<`3JfQ9;@5(j^0j!Dh#Nefy9J9jJ(>T8tcNeCZ;h}
zUN!lpfFh3m46;h$C!OXlo69F`MyHb3hp)}^aqwRN`qJT=+%h1EKafD0%d6x_?DiSP
z(lMS`&$lAV)f2GuR!vzmW{8w%yc=}926Mt;i=sArp3?5EhdY01e(}e}y@9;>87p7o
zp#Yc5;8I7c;eiWV-JLbu`?5xV9UUrtWQydzwLXoh!T9Xnh$xEzS^C@Q&cqUQ%qgi~
z;yErye^#2Vuy|C+9NWBC2;!_9+Yj>0uvA<ESVP>S{9M|0$;y8si(k~SMni<_sUJ$r
zhQwdGR`#blS3&WK(rzwxw;}vJ%_wdp?~T63#>T!LrkUsKqhLoZzf2Qw`$}uG|A7-{
zP2l+cI$tbFpa%}2xHc+#uV}N_KKzXxW~}PEvc;2m!O(_B$JkH7<<UA<GpRq4GvLM+
zBaj-AL(z2yLogGOO8j~c8)-_#^9uPSuINM+I+O_yRZgvAr*uTwD!}~M94jyo5&*{i
zkfuBnK^us>tQZ=?r^J=7{dx3H&T7YKh$HsI0}pX+2S=2Cr)8o2oEHw*o=(vCyCS^x
z>>qB$o?lC>N3Y(l8%07GcBq0ILy|b>zbVo*z2trW{o(9sjG{c8{2;YGGPL-leY$08
z$9rsoa9L>GkUVu4?FB;@Y1OM-iX$zZxb07B#;2_x+jn2h@e}(Mp;ypdyledXO5ogP
zoKlW6p?q+5GP%}eaM0!M>@#~5Gs0=oCdDp*Zt%_X+7#Yg9gY4c_n$~W%4V`YSPr>F
zeJpENMsh)pA>Zx{SNt%JSup%XHLhPcj59?n-<m=csARM77mn>(mzXh@vJGS|^*v-w
z933lfw)-ewy`6LDxJGyEp{jO6q&0@_j3(jqB=ddH+x;xG5xv4xmU_o$`)mWzU`WoJ
zoax@X-C}rGx!}0}p-RwcvD3Ec&Ms)~i%WC3=o4H~)E+Am9G0^3fxsC^dgE(!9I+vG
zwxtuFvEKxVqRVXPV#TSi4k?c<KWEZ`5oXxl>4AEgW|}+S!50ReB)^?+&5Nr!?_TD=
zE97;PtqY30r(5tS-$b0}5#4e5yCj9d-l>nz)|SsiP}QR*kE!sR(JH@mJ>8%jTJZL?
z<&R`;s(PM3&`_K85)w{{qGryXFV&6ew||ZMg`9|@y1Nj#OjX&6PlyQPb+`;GjDrpy
zr%R;BM|3+3y~unWy&c#9$xG<VU@=I-*!FjUP0HpiJNm!MqV|Pxdu@-AzT0_+o?e_X
zS1Q66(EKhR$f1z+o#)s_hT--1*iZHp4f$?CGQAwA(VfN$=e7A|EOhf=BHaI@8@uo`
z)MviZfK0phW!7{P;VZ#m9(`LZcrDXS&Z)XOKGqb!E7R49j<3b#Bhx52T^pyrm>2zn
zRZDl9HZgv1TPr1Gh&?Gs>XVI4o(2~UyGBq@;DDJiz=ZMhEu$bA3x3z{Z)9$1UtQmP
zl=QgPmH0~<{!U+xev6J)Ivf_J_V2}z{+*U^x?%R0HOW|Uu?#)_4q>~p64MiX-HUW8
z)c6=Lk+o9<)}u<!(ZHV^!wHAt0LTROabCsjstngKP!65)&A^Bop;z%HY%qB2TvGPO
zW0lxE6=LsTo_unK^pLt^pU=2}bTf<5Zbs4K4cZ6`7K7h}Q*C$24P+RTOu-U@Aj`_4
z|0Aax?J0_;a+mn;cf?iHfnQm3n(^oSI95HKdC$+*T36y%qegvsepsr|OEfJUr|#z@
zt)?UTpTR$%{nR7cqw~hM_rLN}z$4~$+`44lwngt8=-ifznP+=GP4+tXi`b9b|M@9T
zsmldyV{Ev_uf%miO)tkbj!l6sWKx#|e{H|is&DV}bLg5;$k22uj|Q8m0D`AjO1ckI
z^<fTl75~21DI9LNbQT?i9Z{~VurBoRyKtPM{fBz(_zAfIQ$@<*s*Qz7m(@uAtK9Sr
zjpt<>Y?fVau@0B4!T4MSAn>?#8nRI95a#D~p{L!z0m=3Iy}HUF{+kR+68N35I;xOS
z@>msR<aWi?#gTtwpqzG*D^3$JG=qNM9PVP6lD0drc1C>m8VI8KjFT2?F^Zn-OT`)x
z5`Rd&QWZzfk&r;j=a9;KF){dXNi6>FJjt+6Tw3rbwrWZu+t`qv;VN9^wC11eTLQwH
zPX0<!YVgb_Mp4G;dYupkb^Lki-5$b~L3AE5$beArIc#qRS8_<D(*Q1pw13ez^jV)z
zo5X}gSds2eY5wY=422hkSvp+NZvOp;cdCAa#*$_i$OYgIbeNV}uM{1fzO*eNjH90$
zU<M09pD9@y_x0zu4`Obg<a!vG9nsmTD6^iAR6mr0E8@ZlUeM!X7FS=8qp!x!iGCBR
zev6FZD#(}g?lnDny9$8Y^82XlI)8oiX?}KE3aA=*5w&<%#JmSFL_Q>6<SY2Tl6Dz8
z75^3g_ccG$hrnUhk#r7aI$}64{h)OD=6!$f&Qh-HsHJ4hhut6kwW4LVrqzAbtxDnf
zB{HIk%r@tVCNmjx-1Z;|L!DG_91ao)ON}wIAP|8#S{rX`DYdPm-2Vw*N1<Uxt~20C
z!vrC0LM3KgVYDUDEbe<~RkVFdBr6Kf!JITvZbf2^h9=kS@q>v(#JUs_*}l8iXM+*S
zp%@$V(wzNuNdG~>zaJyYh6`QIARZE~E-it63A5A`JjwVUf&#<!wfV^twmlPE`H7p{
z$Be9DIdC@K?mCG0AYzd@Iiz<qq5fX~SBZ;v;l?(<mYC`6smwnYZY{-OHa>ehA^h(_
zKwUtOP4Fef#NmIK);jc4P_c=uv5l!@-?}`+xH{~)i?H+)zOuOS(7l>=EjpG38m*{Q
zBRP{$r(dz>Nuqe{%(wP3MPcR{1$2!F2Q6pVzjtgE3utIOIu#TcQDh(})5o4>K>kzd
zx~KbU<jFyU#2&J_4WI`pChV8FH=_MU$n9tD?FtB9Ov8GaX_EVnV;A;ClSNa$TQcyZ
zKJ42&^Y2ziOD=mh&M$^cQ*5T`-?W@&lyLIhlQ6(1W%w;#;P%;U_M1r45YG<&Qw{0D
zmH!JC!3LjWCOfAWz_R4`b7JCDiAZ>CQln%wL7D_v&`@0UFY&j)6EMJB_bcAy5CG5U
zdkGQ!ulDO#urnl6$d!=Eb)Um%PP_0`&WU~=$lODDInX69u5Yn1YH|aK+DNH$jt$Sx
zW!_HxbV|(svVdjL3-T+dqX`naOqVOC-(Xgkcc7N8?cQgt^527@1Rv<4SJ|2Y1rhWO
z>>X=>=Su)k&j2cpAROe$B`<Y`48|fPR+F?0<G`q^23mLM6_T0a^73d9On<@N_6(%w
zryzTB;SLrkuRu$TP1z3kX9D2ae$;|3Q{!K^r|095FbxpRu0304|Is!Llx+2bsKy|g
zrbVygD3HM+A$)i3q{q`CDU~Mz%^Qdd61`9#S8s>0ib%Qk8JOmcKZ$O=mE8JOUdf$A
zBfjBnve7s8Dp@{JdS~eS&9Y(HOfu6><d$UtyV$?^ZA|>0^(R$qG-T9;EHI<m=!S#&
znG&)9B-^Xt;U$8Ivgrp~fZiHQMu?bbXe0vTJh_Gp4n!Kx@rParX`@^6{|2%N*9tQ1
z;z3DtKQdP!1gqj5SklY}66sIROLVFr@y~%5z<RsC)4ARM!O?=}K|i3xHaWLS0BTm7
zGXF)5nh)SsLGOS!s49_3M;-7tGd6?oaBhKCpq8>{owdKe|7cRrz8@6IJ_Wpg<>x`L
zQP;tGk#-XBFlm7l^AIr%V!$R3vZF+G@cF9q!t5Fyv~;1pyaYQT=caW9>HwGa;`L4v
zC6yW0X4XsKH2tQ7^>dXAM}+el{{^RF^@{|1j&UYDnl32@ACGS;5y4MU<SaeSmL@f3
zOf==9<wKLL^2O);sUEBK2(2NFFHSJRXr0~Dfu1O`e5;>DuI&w4WrmaaW?;!_rGhQ0
z(zUhv1&>8r!6Y`#H&q||T~qn;;23iQ_tymwt1@}ymiG0@X1FJ>z`m{xrXVq&e9(6b
z($L>DfC$rkw}XYrGvpdf)Mo_zmQ@sJa&HkIQuA_x0_)eLb8>TOd3bo{a*$aXtFfob
z0()G10L;t>Cv5F5(&q*+uFqb8z^??rmrk{XTCMFl<*zq%pFHb!6TRgcbvI*_zId}z
zkkS9dUlzC3ei#+v%l+J$U;2qiTNrKCIqhxv>i{gn+}ZMg-~6lhhm`A{ejW@T?vKwI
zSH=lTA%wny_~=$BC{tAF+kLE8-oNx{`EWKP@3B|*JEPDZv>UY!&Q)7dBXPss>55MO
z+TOg69svFgS^>h2x&w9AbnwO1{MkR4t2+>5N8#Y)T&tNftl9u5ua`ojzClSaa7mjI
zIRLcA5RtL~R54r13c9-KF8D@qDSQuBoQ{46;LJK80bJL^^oIUy2o`)&4HXSDw!S`K
zkpw9PK>BMO=WOeAQb^YMEXZ2?=?=kzuXUna7Z1TB90e*2>(U}*dlFJXWkbMJRvz96
zs7{DiW?n(TBW+be!HQ)^4(Cl(>wF;4R+-@=)+<{haDZFHG~x)I#O8he+*Bl0`_f<A
zbj-RpT=Wm@HFv3TcX)04`T02$R#Ql!K+NO_W2zN+%aMwC>sR;qN-Gb?q~PASH|}#o
zgTkarp$OKpAPl0*qGT3>+5M+<bhFH?#pK=1<)?Pj2uqyz43;Y1h{BkdgzBlb1aGU6
zWNI5JZw{WKydR2!=RvQ2tQ;MDOt~nGVi@e(QD#f<zoCn_Rvb&mKvPOCrK>^BlcBYT
z`p#VDY3MWwy$qOM#o}kCK!Xm=rZV8D%)rOZ#rm(SKBk7^9gXKu9R2taCDw%E-+btE
z061zSN{JNsUU1oH-$a7C?<Z}ympi3xB5SyWuUH#5sYrm;^cPTIPk;y@CKD@Q^#Le}
zNG*$Q6W{|L(O@8p^_U&ff!bN$=VHxqh`qNDoOY(8Vs=F@i_JICN1`6=J(-m01IHjE
zvLcMo4@gfH{Ma~>JL8c3L5;b;Cy-f~S_UpJlh?;Yb9AoT?yf`>OaWz|67<KM{<@7d
zk^;na<7>XC-z|@gS|ea#2M}nD<oHiB?4Y?JaBjgtW<?b-FG@fA*E-<)u4V4Qkdu6S
z(aw}yi)e=p=SK5y!#nmb=}SuSx;n&luhbxqV17uMvejSr)W2KbpGMGkUtb)rfh?xl
zVh}T#ga;+UDFvn9=&P`aUN=qRLz&g~q|84s6PJh(^7MXSh7f8pZs6}B&ZK8zC6Z+#
zRn-(`50GMGG~<r87I*)%LF4)SAn}KcNP>62W!KD8r@GynHvy(Tp5<lY%6pp)9WwXc
zx00?ci)WdVL_g!kOuIQfjphj*Hmo`?7Z7^E;_g&=UHvvU>YeWSX3+bXtu#*?SVm@H
zh&JTWox)GXXVOr?KXZ*X!y8yKiDAXA53s^OSNTjmYOGKwRTnl#-Mhb5WN&_h`KqE9
z@w8cmf*ny$ehAqHCNA^^5NNYThLp7%^}}_=7ltiR!fe?bS7r5r`5q7oYu~T27#Gia
zf9vs4I<k?8DtFJM)nS1Cvn<F|Tm{U0or_LjFwF$r3XKm6Mt2Cc#j~`GPf;{kC~Dsu
z49i51zJ*lyl2nj5;`N|gb{3&e{6C_;GAydDZCem!kQzd|JEU8>yOBn^q@=rt?ru<0
zq#LB98$<*p1ZhyZQ@%Am_j`Q)MrPQvXRqrz6D|!~khXtOGn&S1jj{`5e^t$xh>>%t
zcENk<Pb-**1$s$h+EhUD3=5sE8I~`C>7+YQ$t|8X#*|{m?Y%GTTO56IC5vK)5K^HG
zhv^?Nzm=k;RbZgQt+W?Vrobd2(@3lZRKy~s=|WF55hf$q*F?#2H57Q~VYYpQveD?V
z2|V1!0=&eWJuI@Ql5C6SOxdVxV#9P-j;SKOv1jaQptLq0<JkF;gz23S?AUu;{(^O$
zg0_&l0gnUoDt5}+O1N}I0y}s|PR`74__kOad)=JMQmGqfbK_~6-!(R}-%Hu>%wIfL
z3?Da05#G`+>;g_wtKnzIdKr%dYOq!c>vp$MFuqXo*GmPB1%%W7K0@KYw?LP(E3RA#
zVyC)9voQH7mmd{FgM(^!{vcIl<3OHefBq!H{4{vuvA_}trJC}4j=ckWVr<Phvc{#Y
zy!;7>m720Sb(Zz0k4g8bP`^Y_>vD^K1_A=K_RKr5$Nso}e2NQ`b{N;nGvQQJ>&S%(
zJo~;66VbKs4Eclh<HYn};M^yy?FT$D6LT7G;|q#=Ucv_Ove|8m9gr-7*<e9B@UX6T
z`7%fOxyR)~0T^wO9wU(5Abm7_lKA`u+0ucy%nf8XL#K9T8%IvpyKlTF>XhEZ661Kz
zlQ~f6uEVOV;{}-C$l5;+Nv6M&rvGsvIKoOa5Bi{2Rkj1{8lm%}590_8!ez*k&w?2m
z8edoV<yJ0albm3wcoMGwGbmrmbYNvVP{U{6#V4ICr#|O&n8zdZleQyXqAUK49@>-2
zzVveTPIT+94u^Q~XP{Pw$Re3hO(X~)uiAHT)L@1m#SfKDu1Gb4a3Jl9%}8Q@Q>tf_
zt`ZiiEpPvv`>qqXLgY%jTU`>#=vyONrER%wY^FL`sjEp0M!kpnyDI*$WMue9vRYNC
z82Ykaa+;sVXGB)nh@Mg{b7)T8=7rhUOi^y@*$;jD7S+-LdKLDIyPKi=POpcy9k>W?
zd#>G|n$s+VgA2DojL<7!&MWchncsP-YJns9O=f_X4#ag4!2-wl>17XV7x*{@w5^}d
zq*=$5@90_QrJ3iM<iF4>%9Z1#;aGz`t|;u;Q(a^tK$4R(Zujd*)S&uCJM%)saOqPi
z+V9y>Z2iG5uRhi``yX2Bt|NCV!gN!_vAoVz$H!>g<)eXr^floIIi+!>U474Pa(|D`
z*%vs^x$Pc@e5w8L`@XwZ=n)I4Mf^8)l_OP}52U>khoGJNm&!Z^-@%wXeV)brE-*dQ
zdk#CIl=dQWpD#(U(xC0<HJnfep^9(T-W^Au`k#@F?RQ-+Py`|^oQ?w4?0{>2gMUFq
z$((eg@NXm*wco2u=FLXA1Ln*zQf?r3Eo78FYU!uPX~~mB$(F(UVwC%YJ10^34jQ8&
z@qliax?{5f638inwlKIfwf88jppvlMaDMIWSGvMy0XF0~ak((g<(GTpA5~QLH_r`6
z%s(n@v{gyPs%Fwg+h%DLn%PFR-Sv0ymTbllAVg{_ucr{;5UTf63%Ffo%J5btQ5+gC
z&TV;cOJ+>a*-QX^;%G;cpv1Dp#TOl2Zd!_{-9e>8W1L-XM%vo)*VC~ozORrl;y#D&
zGv)Y|)Aoz1=!Xli|AOt!$di7W4OKK*Mw#zfNeFVfIyXIugi?}lBF&36cmiHq5zx|m
z0(THfDv%`FLKg>p-lzvzkPNAVt~_!u0m*v~0^CjzR{Bis!p?qKOzhuE{I&_+kp0;z
zgYTc-sl{vILe?p!>Tn?RM6J0<8jyL{(`RxFG}Bq7Yk`pl&m3^v5~N+LNrPJpdBR|;
zlf692TtvqGxv&ePTA@)>yl1;^y?5cmhp};X5lT2%cp1cIbnFi<;ukwJ#1lVfOIVit
zj?Rl;xVPaQNj)}9)#$B#bLxXi9vpWGBvJUksQhHAsX<FXaIct&u}tY+R+piIHZwhK
z0xuoSIQOewJOnC3?0y;&ni(Y7Pg|XY^$Qz9T&_=Xm~9I>-a5Mq)V+REVTu-E2btxN
zAx!CfW4Zp#jZe$z3l=9U-3hBStevKGB$3tyL69X&Fe7JQS+qJWO`b5FAst9y*zydG
z-lPft!pz9wm8e3?<mXZu47bAmdI?Htafj9JZ<!NSqHT+=awIPFbl;jCI7H>XLaZ)!
z{NMJBgQOrOtzZ%H)Ra@7bn89zgF34=FHK2~&3lkcd(IK#|FD3wF+H_CQcRn}lGXOm
z(B^XN%(F3BjZQluHTP<S9FZ{Vw$jPW45(3JM4=?qfmOCF((jI-UZ;&j^~grC*Z9sb
zq&lkFQ4nKq-vmQ4%M$XgM`zbj``|NTBYysVAM{j#BIf@O3Pe(>mDo3Cw$j#5WICD!
z5RLX;<#)G+0TEx5(D|$cBN)=jv&486c@L7`(|IXpbL41^h$6ywn2e~ux+V6p94ne^
zS$EXsJg7LqkbHyp_dBCFB$;_2T|6fp7wn?*GGv}x<=$_GH8Zua;5pHF6EtOcBOl6a
zzIYrDojv_hWZIZkyu+WX^;(?R1dF$ru2dS*<pEE`0}F7XFH*7wr^RXYk(>gi_qFQ)
zUYp%GNltUHysUT8wNHg7ak=3jOjfq&U9uBWP>sZ{jeI1klV+#i<9E+dk%B|mIam^d
z-xNiQB|;$<W*^6##1VCK{?DANi!@5asrD181_gP9{cj}88`{y97I)lnDvqQDN60~2
z%>!%99K9a-z}wb%HB#zlSyeJc9WILXTgqs&ImPQ5;0~;XU(O%PFG>F=QyIVv)=d$t
zv@*+e-bTUnWx2x^-CudFKdbcliTqk2bulpBjC`=IkXSb4B?&OHG1DQwHRm4tpNAPF
z0K!YM-@g5(4U}AB0A_!Nf-uf!N0?0T`)?#hd@>#Tz~}s`jjy~s{tMMTDd{7wICx|n
z;>=4;c(c5WJd$(vigo22(QnWeYW^WUvd@5?DIry`{QG<O#T4U^2)Z+C8p|=vF~xok
zEojSW|CW&>_`Tti>frsv(z(m&2fgHv)5@5)88IS+4a>8Op!(0{)9f6XN72InaKHgv
z07eX)4gRqN_9p;X9YB74*QxO9kdh4_s)+4~1Wo{!zM41Y+)=(&<ZHP+izT;-jLFH>
zBfmnAeWfZoeoDPGvh>60F@R&my_g6gK1n$Vn70Bb+YPsB^eimOAUrT#(_I+<QSgyh
z#Z!U~GHrNv3H(~T$^6V?Y1T)dZ5hgBxe=2i_=AJ%az3w1b}|r*rJX(oKWwBV?{y-a
z)|l2om3r9?UHDDs*W>4;h|r2^qjtQXkLq{5fum5PNCrJBGP0KZ!g@S2NqOWx0xmQG
z({+8p)s<HcUZ2rSZy0I`KqwLZ{Eqqzmu$tx<^OlbKyZ+ll20g<us=w?#-7ma?rI7p
z3Doi)VF%rQz)bm<+IsW@VFgS~h^a^1#9BWb#Vn=Js-q`Z(_jP{F^2CjDPhZM7`b7_
z@+MpO+UEVQk0xQ_g|}+S<Ks$%fbCUf{?X!OEU@1=gNTtN6b=M?wFRlExD`Ad@MdrC
zYBnY&KRnFGuPv~0zqF`f1yzF%(tjWk5_#|mU7l21>&`DeLopUJ#OrXB*RFbt`n+sr
zhFY+W;~5QauRFhq6at#@W99zb%Wl$gIu@Di8@i_`Iu3N6kPabQ<?4H?S2E)Y8JE`9
z!~UYBqZ2E|1Scu&H*GBEa0*)MCD1yh0U47K5w{J+?(S{VHi$XS0L!k%mqzhS3d2lC
zI$BzJc&=v~kY>pND8k>p<6eRSu5Kj+%<Syffx*F1P<OD!TyH%^1{cJ@*5N(~uuR`f
z)}5|KkSl->$YBNS0!<d+pTOZmiD_TB%d|B9HHIkAPSt~aofXdq(gjw{(%5_UYOu{K
zf;>pS0fqDzjMTF6jpkx~SHN<1HMy%|0&vQwgECY9>Dyq0h*2el#R~IB@=^t6h~Jic
z^!>ee*S6AeWwL(>in1<IWS|$|oyl8d{v3UWfnV54O{=Nxs4^#4CN8Z_j0*egsmNs+
za@{XY#t__o);r<5nSTf}u7$=s!Oo&ip>8uJ=kphk&ZF>*8J@S?aU)ykd`lFlcrR)#
zzqA4kl)nb>-NP%ZrDSEtfG(904mkb1B%cH~r^Q--6yo(#mw&q-XxMN8d%?}{2HauM
zU_F%%B90&nqT}Ll*UKW>MlDEgfa9Wy3e#Ia^)}iaL(&a)*D(zcaYHiJn$q&qBL^Ot
za!~QJ+OSQ7o#$t>L(X`Cs4;v9udp1~kn1LaiJ_PWl5DughTcGaBLvt8`<&HQ5Q;nK
z691?0<~M`NaarZqwb188zF$pZOZvs6$aj!R9GhkPDQa*eM_}4U%9xL6@a1!bav6@{
z?r{w|a!cBVL?~HzJcV%wFFrIF218PG>v)n$Ckc;(q_HV-VJ^F0IS7(Q>AJg5JKnlm
zeLzfr;~zk4umOig?%%%;Jp;A~1t2AR$3raa)0peI?0s<u+~^x0isI|Hr)k!@k=vwN
zwHsq!gTuvX1Fo&!m?14g>reW$H-mv-3LVykuWWi!r(OVj8UtFty*41@{?P?)5+;!F
zaG|Rb2<jzY*x+uDXHkRoe60M#yAi3-T^3YlX~EyK`YtTs$5XfSeOZStqQ~U%p0@@9
zo#UBNeOno=%Y)+OtC|1IU%+D4h!x49c-}OY@hrzM>WY@ksJt<@v##?Sw*Z;4p~!rG
zGG*s-Hl1d=!`v?(@xN_Y1UoAe?GF7j><09hoY)J7;uMLwb*fiL$yzv#Ubg`%cpv1d
z6WENq4C93_bd9|OvBwUHB{qsdD6R<Sc6=L)1!ju!xHRRM4Vu89XW)CI;U|dRpb)<4
z-Vi9@U3yXIf2`ZhRn{R*1aJg0X?-y9B+YUgYFI4X4<#ww6v7oK{tHJyLKm+xssNaf
zzQxgG1dzFO^P5fhSA{G#itN9K96utI;yXnfcGfRgOlnkAYQjx($`;SqJf*BmYHIt?
zcsgcj6<SC&rJ))_uJcI+h4tFA6m0XBQbI<OFER3L$La#2u3sp!e~pP_De=;)=I_~!
z75E+BEdnv=p%7939;ARd$O^hoO;@xzx*v$q=K6OL2qqRKfrO@ED<|<2s64qD4Ap+d
zl8-ak<MqXUizYJ+uJugr1KG`+THibmK>tq~5L#Y!wdaAxoKHsW(tRk$S|fzua1vNA
zvi^^l>m@*-qh6K4<^tgj;)9KiPD4es6ITo}gvtE&UtQY#&sIXLv}ms&aL-R#I0GH=
zL}4@)6b-5R4Hm!2+;r5P=wz1EmIoT>+)0NQCO6v7Sb+EF0)PPP@=>nDU+gM!2+m_H
zEv+@_Xg4VgD<g(`CiT1L%>9_o@c2Y4K+$g0flhF`Sdv!Y>@5M_F@fw1&UA1`kHW1e
zJn;aguOkv2r$_>rt}zfPT3r4t0L$)TcIL~862VElQw={HETu5uGjkB6^A&qQlKQ;<
zlDIZD|39_@9HMv#cefN#cFiT(8@spkgzp152+GbKjdk8+Ig7G8=JRmmwY|yV5XfV7
zI(tJ~5?88D+X^w+Y7dZ&mgmBz+*Fiaba?CgWiA>A@q2T0a=9I+<v_E)Rg(2XT22!=
zNA<voRz@z5jdu^q%CM@)b-xVefKX}>r*by;vF$E}_SIsZ*IHD_MdBka;1p}zM|A!6
zUNYj}@Y#)X^z=cx#sp8nsOBh7C=6n81T>MU=J5i9Z;%NeOF?%5l!lHWsr4$7=x-!o
zB{GEZcQEybK}N-BCWraPdV*;4K%~-3{~NbIs!`evAoyADqa>8F=r56KVa{vkf#>OO
zvFn~2j~%ZMZGRA-`1+RPZ=v{`=;Mj+ivJUYw^G*n7(|Z<5RlO%`S$&O$gXPN9K^I<
zV94F7kRrHk^sI5L=&^ftG-TIW*gqsCH}I9?_k&&#e}w)_51Ph$_5OYLjOEF1VcsLH
z`xDyD$&0(vrTM;<HPlGGaxP-pl7T$WWn1d3uRb7DIZiFxKyyMeF7mGzYdHxa1nHSZ
zkz%6k;QegH|1nEX5Sx@95nz|45VMiMMs(x)y$C&L<ay5rO7m=GP(P_w`&owz1*&t|
z>eLjn<@)m*s<y@^T{}f{@ELXT=z;8*5|{Z*3PPlCRK%B)CX`>=EgR1q2`lF?9FePZ
zfq1C|a3FI|iN(hN%@=GnNsLDdC0SdK<elkeX-vLixgp2=`lTsLhyuvGF->gM|I^SP
z$-XP)STwX7SLRaEJ$s8aZXw#$l_w-<nxvH3K6yjC%;gbp__Z{BY35v)vY=3*ha%A)
z@kixSG9wc7qX~C3Q?wWZr5I5R7!Ixxr^RZ&5WVcV0|(T?ft{Q^WdyKu;+#<Vgf$Yv
zV>YJzW_?vIQSreSH7us(;=hstwSI^R$ty07xea+GtP|Ssm^xgV;}J*Y5|wK&!jmzO
zg5dFnd_vc<W?c^eSd9Eg>;mxQ_&x4|6HrATH@mCZ^S+8@`+4xm<``<hUVjcEb^EEb
zbF5Hoc0?^h;*@`Zcw@+<f0@1pm?Qs<lMHXbJjs7r*W>n-35Ke`lPwYQ9gQ>i&5<;W
zM#ZR0dXk)5FIj^t)>fMmtuWIAp)s6(G~LLSm*?rF#Q(g|)L7tekX<vHwF6rKd_o<5
zjyQ`hAL~i^)b&--4(e*?kB=UOvF&x8%DxEB^R~mLe3IA?^R*Q_$hv4UK~6iLNXzR-
zuNYiJwf_Bg{$vo3;6?LOxfAcd4@QVujh0edf$*XJZO&FkR9t<JUG}UTRD+ws|KU*e
z`mBbB$eU}bJLK+F5Qzjf-=V^k@nd~>hXK466zma@Wn9wa|Cry^f7XZ*76MbrX?(}2
zi0=lu(*YP=Yy1LL>d$eq+(qR{u4Y6zExvN{@CggqoN5uHA4IdXD{sskMaJip#mw}A
zF;i_1yhYtT$F|qC|NUAIh!!gjvt=*)mE2ao|H+QG@<yr3`HP-7?yjh!(wbxg*$Iv0
z%%5?r@5~vTztfix%*>?L&dA_HX%_4WC~Ltq2WLMdeEPGp=+sgPtu-`~6Pfx@+Q{))
z^f$VG^FrX$H)7-h?0>KKIVo6-wn$znHUXnOrl-I0JBqu-2Vj5g#=Q2O@Yn@zMGKgE
z@m%d=ksO3Ft34iD4v(?;)L8q;-;~l6+*m)rAf(KJm0g5G*=-KoNU#r;a3QZk8k4-w
zO~A|nm^NhLad>K^KDzOWC}z6Lyrhl){iWiRYOEp(*lUHolS{bCKU#f4{Ja&B8I9kF
zjtuZs|FP>Hv6sXvqB=`6R_i8w@R+7^oX=zZqfC+0Nz^|`o^N*UKNx~Wc<%qv@5te!
z(jBwWmn-~KvBLTkaDI*EKk_!_b6U2qNGui+oNu&ENr=2IC^M>oc8oB`4*8C^tG0=)
zpEs{w<Z#<k)pTBD-+|)--gs9b?dFZEn0Yke2YujZyAg2k++Uz2B-F!qqZpqb0T0qj
zJg#<E63M>Z$cn%$7K3JzZwY}}YS-0HF=}l%FEReR(-OrWg*;IofB6$7YxM~Uga(Uc
zM<<+ym%{?frMGx*;qq^)bHm)5qwJvJCl5$7v$!eG%m>mf=JV1C?=Tt%5@k#hoMlP*
zU+N7iD`Y%*I7k{`z8qE!v>4_(ER%D2^S0mxJob`S4!nM9Yb$@2XkgcX6%bislasS0
zR_yH9!0Bgy7z-_}bT+R8(-4VkEDasqC}4IQeq94fPEJB1q-yK}WHq18hG~HfQ5(Lo
z!Iu%>Fp&d%&K9+;W|>OUeKX()%*@R#kAC$uJIEQry8r<Fr_t%-4o^3{CyR`2Lr~6S
z7l&^P8AcviZ<V2nY>XCx;Mv)%TwU(A`xUWnLf+$W*<sP^V&1w0C}Hsm;Kv!rR<d$c
zNY<$bjnQ3n?6$(dN>BpMqYHi2m%B9N^S59ceE|<X#d8rF?C&2&W$XA=(GIxJ@>dn<
z+%{i6=eL8_EWd3DcR(3@SSBRR`0ut#0>Sl_SE9;>Nf<Qbd1ZtRUt6l*P-iq%S=y}|
zT(5b?SbQ#XJ(<qd79c?yUwXxRpp?R6Fw1xR_AOe=C!7bRjETj|EWZ+DnMo%~Ve^8H
z(^-Uvw#U||Etdf698FObvE_jQ9YxB<hK1siP1Wm8pY$VsmS2D>V@u@Q$oD;{v#<tx
z8lw0uxbOF!|BB^PaKTUF*<o?I2sA<}xHa4oh<&LNH6p1twn1h?EmBkpAhXioc~S*p
zlhYTZAix{mW#kT;vgiT4lRJy#QVodUlbb&^i>|?c$Mc03v6+V*0rByIaUUG{HT>C9
zgK&}d!f_Yin&IQr?WA?fI(K}cb+tLrBDug5@ZfRxpxCI%f`ICeAL%Vn-pVtX%dAJ1
z+W}Lm!^|0G{1Wnleuq~b8Ug!fSqcz+Z?EV$O#zCFbiksBx%wJb?-kJ8L%Ic2-<v=-
zHgy7Dt6zMmE`yI+`k+H<h&-@E2=s{@)cy~fgFiSE#okqN@KkBN<lW<r3d%6mA^=7l
zwuFL}ei?oiTdT5ewx3k~%Q>b~Pjn)%uv?gmk&{g6!kOcElOoHQWJM4~|6=i<j|Ap~
z;_u@XUV`pKgA>G$Dh(IlH4XMjU7e6{z6L)=)^{1a1Q`IDBG(%@9#v>W+{63&*il<f
zpjr;ztT+b94<<l7rtL?Fae#`nG{B-=k4g?xh(I=w9kMFME^$=&dt{-ElHT-uKiyJ3
z171sR0q~v{leq%v9fjgjb|~eN1UN)0+@8Zg0dIQ!)Z$#t1#Fg!RA8HFG&ub61zRPD
zcY;i|0b&)eD?vc{#n#rg>!!)>d-H7~Fdvc7fln(Lgdt4luTR0%(<8-)Vvqw|0&$Bp
zQ?RRg(1xZ#E}aP;oz3bPJ2=sai2w<Oo>p6R?*G4(xzM9ayODDHfXTG@xI`eYbfJl=
z8?3C5ih5k5MQ2VD^LGUnc6fYTlG(Xw4QZ<|r=vL=K{lm&jI!ruPfO@*ge=c6K_H#(
zq>`TRq@2>UsNq7Q0%Ihx75S!)f`NFa&36D!8TsEi1SC?D4QY*T!{Wc{oE@!Q-lp47
zh}%5750fiM6sj@#8<{?hvX!sWVL6?++7L;#4R-M~P^7UqxeYV|>zZd?1?`zxFW>wB
z)`<8-EfZV2z^kN(m9GdCB+dg*_KA_$=*!jKhp>sEZ-Q=BdwKfOJcxPE*$m27<JtOf
zj^IJ}0Lgb_c1w3`Lc(2Q;;t~LF}>mA@4J42yIg|e#Jg;n+l6W9mQgQ0^i1Rm%*YUC
z3wgUV8^K+v3uJ=s=Rmc?Xr1p}YE<%hQ0L!l@o$<yM^r*$w>1m-{Lw{RE`(#62G&<<
zbgL6%M6N`qOMBq4FmiJ8YgMB`o3uSybaiaX@+#+XNlzX6W>T4PBB5s4@MF)PR4b}~
zT}9R1(2<;pt?7J)Cq*Wd(l17L*t#<aXG#@-4X$`n{d}X|qiM63`DVCaQ#FmWJga^H
zsiw`P-;FT9(#}?`DYiVtonmnY%m%;r>s@p{=IRTo|EBL(Z2YdxYUg~Pj*O&)hfQpY
zx6=P+bB%`e_h%S4=7h~OREC0X;YGF7)7N@Y+?Xkx$31P2#)a*oW7GaG3qbu7C}t5-
z422uhtr8O1A~8a)X{HpA$i=pP0~25FeY{fv&bBJQ-__d!?a4e>gGfvV3LAEB4C>4v
zv+!%6<e5M0x$4X0SatLNLf8M}e9H9-P6bEPPpad6jdZfCMZ%8fZ2}LGX8CbF4sZAV
zFN&F+{%d2A(AL4UJdj$)vwEJ6+H{ZT3w&T;5~%2Wz4yMe7Zcmmy$8jkqtfCI2w)Bp
z|Nkv*NGWE~ekWBk_G<lsS!*FtitbE>-TKo*d`|Hu2hT==Z3^wY?iht}+=JH5WtWnE
zqmYWo-(SLQ>*0U)j?w@8ZLoT6b7ctf?s^X~Q{E9YoE$bLrn=uIS&>I^R?Mmhj%-Ap
z(4wFI=yIsA=fY{>v@9w<-27>|LtL`}%4koD-2^X52z@1eZFNgxq2o_?FMEtRx2AZo
zNxe;8)-G)vw;v3JovbFGb6$7vpIkKtZk7{1=r?E?$Uh)GLD-u4gFa{1646XY$$)Qw
zM%;F^vB#qzrryB$u!4v)-0?Zg=@(eJSdnyt(NsV>=2MrRC9G3<zq{JCzFfB-wWhpk
z==<iZQ7E*fIU7)z&C4E|M(|(?@n+h}1>hI$s*<=~q$IHbP+P&p7nZ<#C2}J3>{-64
zu<r#fRe%H@M~z!)c6CCBy|ZkQQ8n_(&^ae-&Gn0geoz)X`DW|rni{Iaj<N<{sum4_
zuJ5P~9$)J&aFRE3z64^?wZ~#G*8to|1%iH`BOb!)NxZ0$yi-x}SsPbvu9L*>B7-(^
z)$)|t<KL@z^{8V+Xg~jNz*&oUl+91a)xig`2y%KBQiD8n1@#JwLiiWAXui>Rl6?#n
zNT|<zWJ<Nk62pj#LyZ3`=tE$2rNQx*VArud#}hO6CHYE9joAlTY(0)@^@eyuFL_nZ
z1;+?9Klzt#zo~wXAGe5*P$vjq))$wQ>1h^o3FcGGD#{KsI%Ja-H@U`DzHXBEtsR*`
zww>=dmcLM4V(1f~)4tN?;L2!JntFEN+?D%nO_!@ngLAXRxY)z2(te@#I8Zq?@$^$J
ztASE2O#Lt}@!HjTEHRjEbi+8XuAFtb?NgQEe06+WC5p1&W(4iYm)16Ha))n3TiM~0
zM$DNRpLUrSujy)5n2MC$hL&{QeXk(JIYAOOI4QAKI-A4d?;O?{Yy*^vi?iNwKNuu%
zFtSR0)MjA(o}V#$cKWtDdz+ky@$M8J@<IsEIRAjJbV>4cUqZ*hz1J;ejo|}KmXIl-
z;b*6=h3VM311Jt_o<GV|$=f!P6)U^Ri?JxPcL0=YSyX~Of_pN|4~5{+R0Aid3z+#A
zc=?xs^p4jBl;~PfI@)Y4o&Las93el^0E&)EL87T!P|5TovI;kXzwl?@o1}Y?f~E~Z
zeL^3x?2C@~5(mV*rW~!%`5b@w#AQ8>ZxpnJZdSJ+U=ObW=-`D%ZqYcfN{&~n+Qftj
z!{o|Akut0kKhdp{9!x`aQ(?agu(zIUTkE=9NrE2b=J86I+IocR^{s{zu><TvtmC{_
zPa>&9(Ak<h;bpkY*p9fB`ut6NAN91EB01dvV)=>c)4!H-5L$=`{WX9>*dLe|=~Jm<
zJ}5bSvgp_jg8JeGaQhSS1(`>shRmsY1;Z{853%TkX_V;-N%q;DKP!Y~aHkXj`;bf~
ziRn#We>lZ6ID(R=bgTd)Ff%&`fm-vtdgq%)8_(&^9|-qVN0PDXEmjc)?u+M6_MNyk
zGM}QCtUJe9{#@XhOXU{uZV@^eN7ZPbRCoEgDZDXxOP+KyxOWy_;VVB?QJrLsjFH~C
zb-$hVidjdFUXp_CQ)7n1VHiDmQp04s7!i*HYmLw?%<p*J>159wb3{}TZ={gtk7aL$
z@o3u5sf@OV>WuuKlZ|vWqAGc`%73(zG~4q*B{YXpHu)>%>Vrk&1A?R4eI{NeiLU(9
z0-@_!3D?doQv0u_B(LgH(e~yw4)Qy+JW8xdafpVveru)v$<dyg#-H&e{Zzn*lo^*6
z9hC}HvC12YC2QW2Z2TozDTU`J&!&*Js-YxXAmD2AVLvG=hHFAdr&zoSV(V@n?CrVf
zZgb6=;6yGGpD58xpYK(eS0G99suvZZ(n&YhUEGJD(;GRRA<>wza!J79Wv1@HXYDI^
z<<!H-73MaeUBVBcV;~Uv6O)G`|1vzl$y!vIh8hA;K)B`|#QTLkN~iC5tV<jI1<Uym
zL~7RKqd3!8;~`<qkkC<e*g@V7h7=oI(=Cumus`9;TECanlcN(yp-6)uL4!>`)k8Iq
zut!FDyV#705VzPeE#{PBGiD7^(A!>$m1fRKvL}=WR^C}cmYmV!)5^$*D5yp6N2}?K
zXAWh`^{b+VTn<S&{Z9^?D+3n~d<%#jc}fc{pM-dw8ePf@tgAj2A7}4+@p|=smu;GZ
z)?pFLil<<wWQ!%IMDF9rl&AhW)7Q}}{MUak+*WUyTv_nDIP~IeUkfPeb*Z(BxBape
z#`rCUnR!c_oSScZAXy45nJ*|~m{t=$PtQ`rrjlro3MqGDbN;Mrgf}0(g?)qCpm+!P
zJW+&ldtWC3kbD?ShpRw6iq7q8qw>>rh4lMA0@AaM-OvU6Dspk_+;{iurKiarZ)^h4
zDj!jI{L&5k6}I<WVpf7A@7)?)01TgU_)nk|8l=;wUaG%}{_YedETs8X<DN1?s8)L^
z1C@xp_)PGmMVKT$n*W8BV;+((Gj9Cs6-qsUufVf3E@Y?l#b8Bi7O|YBTGQ#{py9)b
zkSp3O*L<J)qK@a{fzgy3ZT91L_V0C?X5;Vqarz$xibL>H7#riJr#b3v*PjZ!_m_Pg
zkiVek+1WouP=2j)!dhj|{a!P$fQCbzO4R?d>d<g5&p(_b9nFiyO}To=8JME}#IaFX
zw7}X*El+vnCE~tfr+j|Cm*Em#*;tYPi!e;EXfZ@As3S^!gm5r-{wh8DwyF>U!$uUM
zKem&y4V5qFBYG1sG;&IAEEyQGgV?@NW@Pxh12w1`{pRR!C~syz0O=Ww;xEWcowFj0
zP~Urxm8@LI0c?M#J!6-EUuvGL%e8)<2@!>p6vmIEb^Z|4CjK8B3=u>+a;KcH^K~}Y
zh#tW&Evi&$fBt4%;@H&s$)fauWQ?2`HzG&vm*Hm|hp6M~#@?IHDT&DE7$6(iC(;lP
zC%wX%wk{(tH;E)k{l`(Oni?hJe@$=??$3UI@!YU5w<5ajj=CRn>Y|kwyh!<aVC`O~
zEG;E9%2m^`Y7l(ce7IoRKII1;S0goDGDlQ53~(pE_S>^zG{5<v=G=XVX@P_GAwr`;
zVqlc>_dQm~9H+Nuz5kHUt@CYWPM4>$f%nayCntR&adREcx7kJ-lUFLUpzHu0d%alp
zeiYk$-1gNoOoS8NWIinH0+SQxkN<LX#XlJ?_n5?@{=o`l?(K8H4~e|4P*}!dfRamv
z_e8U1OO3mFnXVFybt(D=?=q}#T$)Pb-#RKq9_fwCGt3mpVxEhSqtN^>0!oa4`tusq
zUugWDGouZZ<G}(8>jYUZWPG*e2fS@%Co+6843|Ee)Ip5q*mKA%EQkw9>mAt~gVeDk
zg8|@`Z5JBL@8hp=X|PvhUl>qC*>}#pz%51-Nr2XaDOb@dHVT7jP<A91y<w&x;A--0
zWvMuef3&*!P-%g3R!LMd;9ff3&bN@TVVJWB!M#6$_i}&Hjwg|Et5a_o+6|E$<#jZ)
z9wGbQ>3sjvwWDxW!g2AF^j!68+NuOmb?qu$*0fk{=UZN!XUmPD0!;hZ?@GIlW0oxa
z7}gaAwlmEh$1xz-iR3iRHJg4M%ToS0>~GD}s9*W`rIO4?jZDZ<PE|UTnPad~ujCh$
z0Ycs%LR{5c2_X?d4EBSf#!_%vP~Dl-_o1eCMEt!*kFpvaW!-=lK*!}VX%(QDVTt+w
zP<Jt`BdSNi<|R4MBc(=;D5@jSqTV7A>G9A%Sm}{+I>v0*Qljv3n!$xK)^0F$L=QqW
z1Wq5i&A7H#>LX{H1{?9Je|aq|x(92Mk0Wl23#xnZXYY?LVh-)jVf?3p7i6>dCh(*Z
zUAZ@ct07`Z(EgA|096rx_tD`W8UEIS`q43!2|@suZq+-|7Yup@saBzJ8&((vc{SA8
zqBYue?*<7!7*W|&bKA1`be~2{WJJ={X(_?rNoP@RJ`@pV?{0iLOt5}=zIw?!DkI49
zBR|^s4+w@?5oSBkAps_(DhcNY%boLo5(qUkXi#L5BvzcUc(UZXN_}3L(ad+O&<aVE
z{@Dl3)Q!aB<deems4qA7q@*PQn_ugP4CU(OeVylzc2CxO&qMGOOLm~Yted!a`flhC
zE)-=o+={;D0|x7toP=1xiR{;(-|YTRe1L+)KBg4OmdO7$%*8(uG$4EwZ9PER)#@&s
zwZR_WN}Adp%fZRO8zpE=$;G1J+YATN>rR$|<eOZ@Kl5whk5E;0l{T~im6mGm2fFG3
zsq`*#1lgwtx1bTf(X+MJOb=-N9a4Au94g-Z#=#GZoR7fv`!OYWD5TfsviUxG>$i}j
zkM2M#=Kh*!LeM>JhGBAr2jX*$5CRAx2qXQEX(xG#_*`K??g>sXCw_E)-%g21MJWOL
ztR_BBN_#t*oxGful>GHu5C-G))y#(TGGsu6ugcsp+qN(-xi9$}0noxon!t4$EsjQi
zeD~uw8)+?{h<787?9Dg$XCWh|t@8Ku!kzvA<u7w-6F?Tf1D35230~J3<ABjSfMc~0
zk;C$Y5(Je$*P09qUoKHg-E6r6^}ixqn&2qA5MInx-IsJ^v@DccG56`8egeF8F-i)x
zxng>Xcf*5uyL8w#ly16@j4}nX-R$)GWTrcO$4PvJQNPp6Pp!^uOuQbFZ@c<XOV{({
zkzNva8#BvHHX{9C3Z7({r?d;s-OJ5~$wSfNkwN#&<iiwLf3l02wRuQBbSVOXh}xx%
zE_-KywW{yt{7==>B$B$!A2{sEa>H=7$F)V&ZzbN~y$t|=H3fI7@UO!5RBqz|ayl*(
zNf&4eP;0-TPC;TmN8GQ<P?T&A?8IbeF`7UigCy3=j+_aGeKsrR=+(hUR)}|x2XRZ0
z_`BzFX`X-$C;x`FJzOg05eBCj1s>qzrL)`bZ;WHddtE??Dh(?6OdmTw9leEIQxgz5
zq8%)}`Ch$m7w??XiBLtJ)HL1b6PBuA;Z2dK4-mM{GK>Xso;O<Q=8g9tGRInA5WhN!
z!U14vcjDVFFW;#q=6;!W{l9Db*_29c5DAxZ-kaaf>qybwsH`LQ_o9m9%@N!DCqc<M
zN7W=$WYzTdI?d@av!frg)=mGaV6?o{w4xg~@F<hlc=Jh%waAtO1AnHkydb|>RfOcz
zB9tJx{?N&GJ6`x?+MvDU_LZ~7dzZH-(JGnjJzKuOsMYPh+~LjZ`KR-8*{(YU6;D%>
zien1sAgWu}o)LdQ)rgX(Dvf^KI|Fi#6hCX%yms(!hU+LoO@dOB@*Z(MKF$_!E$sld
z1D6S&?<893f8ki~H8_IpMFAe%tn4q#dIq?Id0v|cq1}0u{)kv)aDycSMLFs@r4BF^
z)ULz))(kK>LjBAR=9?W17I8N#vd=_|MX8j(0-%B|u?L`Wa&WCB5zRsYh@>5$6ih}<
z3e!z~hEj+g54Wh0Mm%`Q=OIyA`lS|d+a`Cw@lud$Jzsd{4opsqjml7IB^4lUc!NTV
zl-a0MY!JebOqS@Ih~w>Xk4gIS+t%6UYyS6ao@~=}a2u2?dR@hKvjUV9`~S8VpfQ64
zYtY#!E8{F#RvqJWDTLKNcRBX=SeJLXUYyw!m4vpX+pXM6=I_mEz2vJTY5wj`gHFPl
z+-6-sd*rO1#A}nB#rzSM6EP)FfeUv3NWkr<tW5S@m2sJ3g5BEE`@V%`ztag`BNFk}
z4n)G2z$W*?ThcrNUg?nT*6IQe8j1ylkl?$pjRE(Ym4I?gZ7yU}B9K)hw$*_s9%5fu
zBK2kYeaTCZ=&n>f>*b_=ORDjtNAf7n4pj0QjsFtd#{?K~cJj$j<1eM+h?D}uII2RX
z6HJt8k>4#+a+`l~018>J_3eADn%6(ESq#3MB2&SGsdW(tOd&hUd_OIQ))=ikdoUz+
z0b+EKR@EJZ@S|2Ov^GrZwfAP)g)i`)&5FU}wk`zQ0^ot&o<H3)NSOtE^WNqq2w!Bv
zdj$^Enhl0;iV{TLjQOC}J_^E{DKSGt9UK%MjbjiWM_167`9JQFUkT}Gr6P&ITF&ev
zrS_^4jx<mp@jMRHr80leH!sQ-x!Cl}GCQFcnXif}w%QX`DsJ>)hX&!4T3L*aR%H!t
zaO-2oG1e-qQr225lSM^Oc6IL?T;}AaqlNtR29kpps}hQ{Kt@)<Sg6A$e;mIC?*jm<
z<+nFW?47VL-Y8TX=)phUt@bW`E1-~|0*@UAxTA<R`Rqwj4(N9l&lZ~$fAmn$$%Rnc
zcC&Ux*pT4ALTEYF#l}Jvjfa({rT;4RLb>FXv_c6@5_XyeTU2_WLOLvwM{vF&fdlp7
zw(V05-&9T<$5@W*?}=NMOV7Z|fWRr=?!&{}Q{hR9@@1Oja}ocWA3x#9O@DkEJQL+2
z%=!qliFrhzws}IStYP}VY<N2NXKQO*WN^p}4%Y(|DHZgVFO!An#CwPK09fCMFMksr
zebV;!^_FU<LBBUhj&(K=&3H`FKawSgd3(IZf!WakBB)x1yuZVc?Sxt<U;FO!cmr40
zi|Y*%I27Ja(cwkQ(UE1c725_KjM0s6S$*%ek+^SyM))dZCoY-w-x&|z2In6k$eCsJ
z(iK>BQ${_ASWZ6eQ@DgukgGP<uuc!0jJ|WyF-Yb&zs8t1hiSb2qc1BlK2f57Ar_P{
z-$m{0!p|!ny@i_cp*9x9VVtaCEWb!D(z8h^#rbEpoy!Yt8qV^D{YJVO@Uc-MLk|fB
zyf!+p7Ivm6u)g`+f$L)bV-vJJn{_5EJ5c&e4wsy6Gg9MYo?w%b?q8FI`auVGrh0q-
znvZ&I95o7rO0^wIbig9N)=qk(2wKkHdPOEQjHJ+yX{rj-Cia8KD1+o37xrCzPJYKF
zeq#M-BBs}}y(%1o2a!zMCU*o9oEt(e0g(hqm4JMZ_`vd7|6(<+&lNf9f@g(O!3({U
z!)@6x{xo0E!WR$@b1^84mV{Gu%~lgc2@z>_`^wN`PHg=#j^(X5j>RlHA1#N6TU@Wp
zk7NIL8i2xPZ776`rmo|d#*t^WG?i(%+~Y;h1|<)J^&8r)w1_ybxXg1}gNrd9$?ee(
z2E+!O$)RN#nrtyE&N;r8uHNxu7@Bz0HI$Lkr(wj{PP`xUo+C*}gwilMB_4mhJ`@>0
zDjNO}+e7XW`U9l~AW*Z(Y$x*+K8XZQ5)%CM-Vqy{re0tCBsD=}*WrBlcFd8iabHn{
zjvWx~$a`?4Hzq<CuDLp#mdW@BF?*<Ljy_gY*GIq`gSbR`ZG6vvZ(?or+-^Zw+d?-G
z);Kn$1887B;s}#BJ4)R42D!mON&|3Em*R7iX4{!TN7+nz+8oehHX-QdFjsAqLl;aS
zArFW|hrSSjbf_jDzJ!)h=Epx7o*lW6nYS3mi`{|U=ejRllrb2{rW>=d`x~iB_W!NR
zaE1y^A_yDnd9>9036ek-4d;cvCF;n=GD}GVRwZS}**nIyW%nP3gpU1oOc~?Wjvbs3
z?)|M;tf1un%&)W@r7vU<a6I(6Yr^yLIX;ZIeYqKZn<cROO^D~OaB9pXKbzxpp)^Yr
zwbNTb@}xs09vxGte?puhG-HRct%*yu;c`6haP%7LKu?$hei_q6iT<M<f#Bzd9>6m9
z?pzm(*BcOkx?iK{aSGQUQJ%M4df_|h;24^Wcsr5f)TKw^Kwxg(;e~m*H*b!TPW>u%
zFGvVWBxVP--|6n^c<xU`L=%9N(_zQ)-;)hxwx!R=*eo(n-yueOCPKfbz6kt#H<yCS
z_=FhZ1drPD!n!A&5*VI>oh3I!9bqnyrlL8~=ZoXM@sBg_*#wW``tL;3#0zD!A!z24
z>c(&lX3Kq$AYH*+)EmM7A;Un%*G~DUhiZv?`CI+iE}peeWP++L6X?CSwxi)}S;p5%
z&&ie@tGnvY73mBj7Uc`{!ivePC%C5iX8idH-CgpLywf_bH(*kD{k|H@z4tU&*?g|J
z=8%P&V;70fw+hSE5!GU~xiyX~Ipo+iDm#1J<^d_!sy3Y8`qaiQE2XNo1B1MF8KF(L
z5Wk~vVK6M*kpD|Gnw2Pjf{ubSu!z(XR4e9H3xs3ywOSo61p8Rxt4mO|NW4ORulo-@
zvJYnn)0DY5wCb_W5`%{POiwmY?$0JC8qgo^`!dyD446lKK|iG!|4O(?PvN&gQuucC
zB`sAZ*ASI<pZn=KfkeQ(>OAH*N%i3KGP6<d&jjkkk$k_a7}7W~AUxY!UJ^pcpftKq
z->GdAkN9NFL#W-$x#tPh??}3Ps<Y6D)N4~1N+|wm5vg|sV;a%w%fgV~i{W;X@>n-v
zxJ4Q<$mtO%UEwWRb>0;v+1^hW#m}S_zW*oA!wpuF3L0t*B<E*>R^~clz4zWo+pX9b
zla}cignb<(GW*e_?(Mzz^$lG=hpceI<ZcGl#QOFrzG<&Y%qDt6Td}PO`y*m2ZgHeS
zb6M5lQg@Wn<$SrUXFbh$W{)JwG&vz+P?TwR(@Ft^)D6g5`FxsyDep^}e;>OG2jT`P
zRHaX|`z+J*Gi1njEVQhQJkY`0EX(i!<#ssYZPZO{=;`T(F8X`Kr1~(bCkQ#Yj<8IT
zH^aZUG77X-Vj$@0<%)mP9m&TVkTRjNO?zRF3PgoE43y`SJ0sS)Y4}6gRJ|+`wW^*o
z@%#>F5*RnJZ!qxmQp!70_8>O!8A>?~jQc8R?ojhjmyaPElt`BP!|No2j2a2j<`ip)
zaU#b#s@4d4PXDOv#@C%_b-i8f<sA=8%miY@1-1N*)QcS?;~A87_4PD^%3Q4KD0cTh
zE}vEz?Tsr^W7|5*^PV9giRUw~TuXb9i}27hmre1h4PtuJ2fEEHQ@ac^<bMS54PrM?
zij`wPK+3O(&o?$<_uKoJy6J+)bmA@a>ZlJB-qBEdO)zZGuCEAr2DjU;MgLg~c4fo9
zlJAe$qrBcnJ2?mT{T~}{%}MKZpl~Z^rPlNCwE+%_t-U?S5N%eO6+G-PA0b`@OQ%K4
zvAOK9l1umrkv9Q+*@=GBwVw~7`xVhfRw)BR^)(uYZbvxeQn?OktfIl1bDPsIa@<Mc
zyCoY%-0-c4J)_v@QFSUS0qZrjglh*ra)@!YT(MGXS(uog9zE>hX7$wKBHM^pmg5nd
zv6xbnUDIMJ^SqT%G{fw_KXw7AS~)tsKid8gI5l7tANLzSkNtNf<%T?pcEJ^r(I!a)
zIZ>2ZUUvt;_bncbhDiz=yML6l6#o|A0rXOP?t%(}v%{4Yy5^QQvkHle2g{x9$|ACp
zB(!`qpX5hnF-WAjdL@zs$bmVT#V57}^D7}Ksly-H_jnO490Kk8&5JKMXi$~~VY5x-
zKo76nPoG5Q?fzR|9D`8MHS~=0Fd-J?6XO(molR%C+)W8n$>MDE_wdW#x#!UM)%Cg)
z9vu!SRecIk7g$M=b2Pw8`ObS0Z>7<adu>QKMO9Nqch8uj6mLd$`c90+m+Es#@E50_
z_~yff*L*hBIq(X?4Kkqg-O9JC;)0u}>|@1%X1tPy=cA<;W}X%wC3E5+nXr$pG=GPG
zb@g)?nv3+bc_c;n=pu+haLPd0mSQEgJ%y_Re2$KYurZ#ENX3(c$cQqOO_Qn8&?Jjc
zI(%3%#r=J5eb(ni;rAEq75~CJA#@HJ*~gevG`RE5lk!)L0|9mG?eUo~hV`bSEetDx
z0IQSNO1>sR_l((mOUV^ttt_$=ZgMS8zqX0%8C-JE@XIY<IaYPB@2j5ESe;I>(EL5B
z^D>Cv{G~oYIJuRf7*PhXAOJQOE;NOGxqsrf8QDje23eez_Q-PTk_!~tUPwu0;rvua
z7;P?3d#I&SUaEOL5y&@hwx2f}q1y*iB75P`1A<Ex(GCdhKq7@~Y0lptDg|DwNl!&_
zLM_&QN<oq5U4=3d<+rvsW&~Pf+3DL7d4&kt#$(jkIkgm?M?xZPC9%mJb(CKpXc@$l
zDAwKuZbj$0%nV;EU{7t3>(Y_6U&`guM$0x?$kP>8muVo?KLz4Jz0mgjSO4BI4Mc66
zfa)BmR1A`zPWQ#g*)a0Ne`tfX#<ZwIvdliiwMtbvFdOt|WoOd-^lko7sYvpt$3+(@
znm7Y7-|Ng}+YjM^=9C4E1HO?-#&UVgvF0;x7}7Ovqx5xg9ZPGEfp)pF1<Dyok5#kY
zFeSnauUY^@MsUtMhKKy&cN*9I{WPWS5QG`l(NuB<1BPDc?_5#ZUR<`1`i3HZ=9PMt
z1<*P)i7)$yOG2%QO;ZUBQrNbqIBTsx^DH_3_BasoQW{G~`cXo+?GIu$1YET5s*?-~
z(q!@mH)2<Weq*2H<Y+3Pn5ndByi@_tV2BM8SPPZgeubdgFOwI?lj8=|9s*OY;%^Wy
zYZ-G4zFCVt*vzj!(ttrJ4L-u*Nx&m$OMvRDnV+2ol>2R+fb5|8!YLs5Z}<W4ajZ8V
z!q5XPZ*PV2N5O9yf%_-^sZVi1c7PqRl|t|eJl+%#dX;=GtG!{2YpQ{hk4F<ID9dBd
zLFZU){fg(3bHwusLO#dbb21rOS;hAfVHTf*5Sl+9gU!_8L42TyX5;QeruWP^f%3Ja
z=3mN6**x4S@?k;7=jzLluHuMdYaJmqe&H4`k?V`Us1MI;1Sq|VKXqJIjO(??c1F#g
z_j<?0SW9y_qiFA+1)&I{!y@os&x|$22x*<MwC3l1{^c2VhSG)X4a%N|sbj-ib9`&F
zUYXN%FB5hbF?b2Cw`BKT=F1VbP!o3}XR#NMoEu_`KRvM>wpRb=+@Mf{2F0#AOY%Q;
zB2%SCA&!fUZM2xqA*mhLX;~b^q$d-&bU(O|<kTA^-i>vr^lh;u_rGdBKgci?KcW07
zAQ0XKZDG`@UShoWPDrryeC^q>csRJzJj3RJh&jNPL(34@iu9e0R$Qi)wuRU<JLiJz
zNnTJ;h+%&hPR7WTT^vOc$JX}}ae-YRF9{lFr(o6mnDT4IQU3Ih`>R*G>nF;u?!P_<
z)6D#8fA)L}ZVS0)RsPD5mX=q6v)PLEi%E}K>Lhc@--BDLnpwMY=bClO(R>e-gp+>#
zR<5}(XiF&7apXG(Qc7;uC~dR}A26;eO(26!jz1SZCbS<Pea>i1+_RD*I~E*2Y~|w0
zc=vEXw509-A1*!Yh6Gn!Rmes2B_N)_yIW`@&|u_Z0e(v{4le{<3wgS3=MR)IY!Bx;
z7wdl-8yp6xb4ou?(;zw+z-&4ty1CdE7D?Y7^UNVmxb$c?6@Su8+bH=QrR`lva$ic~
zkHqn3gDC_INzEvwxg#+;-;;d}3Yd?hwtB*@bNy_;DU%aDL`<mPc`n?fPhkBj^}6lR
zHNXq~^WlOVvBGL_WIQ8bxX`(--oh-`Z&6eR&mZg7I227kTiCbxTunM>7)|I^qHwI8
z<dp~(nN$pg*Omi=Ns-ylT2o2l<JC^^X}w6I0Ybc^y;9?0f^9oXDjoh65K^=@kGa|k
zaIO6SA%A9_>a)GRk<b?G9`3;bCo)WG&3fRglC%PG{Y8dyoxh}6OfG!~>t3WRe9n|>
zkeiTeqUa6;OP&Tu$tGs$3)L&Mhnv5^x3DRRr@8>3H+nJ&Wc%6<G?QSAupLRWF4CiT
z6OP;&`V_tXdkCl2c*1odO}^%f?LRi#xW(?d{|?;Zphv0_C|sLqr05|!@5F-l_!|v*
zN=^DOm~ls#4K9ZuNX%u#m`gPdTH?>7!$&GsDmlF!sK8K8-{OClgZ5ir=b5Z!H~oPC
z>1hK;0GW<Dp81%lY)Rq>^;X*ASmvI73I#EKe-5&Fc!!sOjwhkh8hMIYiDOwLS{v>M
z6>2(}avC|v`?z@iq~=FSH{dxD#*NVN_FF-i6eUG?@)<nqd*(TX&vJMOPZG<wFSM3H
zu26Z6HWy3po%a#;qGvH*2w~3#-o(CDNG%8INoD|g)(P{C&U<|XExQii<%@#s)7Xrc
zvB$x0S}N+_7k=8JWMtft^TA#fERm(M@MP^dMOk7Y*J5+O>&=Yxo%c7&0m7I(DO_^I
zZO0}6nRo$@>8a!=#uV0**^9dipu(>Mez6&m4)^<2+we@mn)i##C5rf_BCuSE!X8A<
zQIao^Va9a&GBmBB^8Z8P84$e^q(|}Nlhsmk2qED|@;OAbmS_9Q_83$a=JyBGE<bNi
zEfN@iOn%az6bqW7^NxzdB#l#?XHMm^a=R(IGf%$WySYBX2K{{F41Yl3(z*634|UBQ
zoQ7NtfyzQiN@6LjyxJ*ry?WVC<|tX9U7(wU{+Lzb*@#gfP<g;=4TdqEeTRy~B4<G|
z#Y=}hhYZ5MLnZB5KrK^qnFuS*AK;p9BokZwd?zke)r8UmgK5m!f?d!tW91ScKEnVP
zWSKAgHUkv3CYlq5{OlH(W^({bRn*!ofZ)d~PDDn+)p<fd^Vuw%uK;0LmJ_N-jaEny
zhj?e|f7D_mP9JY)i&9bz)SKmH4=}2KCM^6vb1Zz!m4>7;lQ2SlAVsG!5-`pj2pdP|
z`0+MIL3BDCuY^*HTA59T-@Y@f__l?>u>dxYV}`B4+8J0Y4z?$aH)dcjhvvb#+#qZn
z8xMu!ga<;lF}X;Q@U3HDc$GT#E1}dNp=r7o;<3of-qTU$8py`Dj&hDU#Lk_@1_bMe
zB=c;1A&n_>0E(1qU*fsn+63mn`DSo_s}e;&%yu3EVEhW58b?VLy$A{ki9B_QsBceX
zDSmpac)HyAPk2krPkOs*=zcgfoU|~KRGLB(7S{ehh;di~NdZ4L(P&HNPVSMHU2C4q
zj>VcV%f>*B_He4`-*h?2Ao6>ctfe1&?}R+`+zli-P*>+P%W#5`)?Y2uAAZ<C#Y{4j
zv0-X~<%Pp2b(0Zw!^I+`tzr|x@7_we4da~g4inN$2@d7>e43t~ZYH#S488^pH1X)m
zH0a>;KPmgP8~F~^eDlXZ48HHACNMtc7(^n^6nG~Ltpo%vX6Br#eY#KecA?E7Ky`H;
z^caky=PE+$fMz=n3aD7K9$fAH&F3hX+!IkGg$I26BbK1=AgK&cN_ErV(~b-cg2?Yn
zPT!PJL<baa=0hf2zVCz$5?g-reSG%v>F>tY*Rt_henL1NJ`B7fYGvM|9ZcLxe^)G)
z3uWA1gdTiGnWovsz$B%Kt{nxg(I2(bk&%!bYq<mB*FC@^r<D(!mV(;pC}Xgl{=va#
zC6T#UyrUejfO^bTZWr`&9th)_ABkcCJpLEpm|nx!C7clC5_@b|G`{<@8SI1I$-`Tf
zOs4gB2W<E1<OHkP!0ffaE8Q4gt1}qNm!ziaf`?Ky!B7Fj;Yxt5@WG{={B{(^eZ$55
zgkk~&GmO@e4Wr$7-Gs`v{0rIpcLl_(DAc}w=-NtvOvGwhjtZE?%ASRnKDYQ*73c7p
z>N-89WiNv{Vmw_k-2TI+5E5EVqHx5zMPEL(_7}SC2?~Ge>Uj~NDE>IXm(y=IZj88O
z-i3>xRMFHxd+;HsrthKLXd1F{pTB%jQ%oQJDftSQx;s(7DN#;>1s;Oco+k>6zz!L$
z+ml3k+N6n3_Yy?&D9TWpd`Q+P9P$1GnTCCUXL25$98;D3KdR0;EULBb;)<knNC?uY
zgfv5UH!9sFjUZhjL)Q?}AxMgXbc2#Z2og&703zKD-!tbu=e+Ot*ZJdI=Q_ilJ$pa*
zy4U)xVM19`ZM3DM6M0|srRp~gczK0Ky6u4hPf%qzjB&Juy!Xr~d9d|mzxJ3fj;SR+
zU#Mi9vU@SP+X5Yb5XQP<%+yOT`lqF|trzU>+yyWazZm}^;<cHGcbVC0D)%6$-h4Om
z>fh_KA-gE)_;yQ-N~nV~?%C}Ah)dN%M6THqosZQbzej;{s5D#nej%}lQ%)l$G#nM1
zZq#UH%YwofBf+RE_Amzc<h>!zIS$F62-$xcP>v>Zt+2$}#y-`G$Z%;DbK9q7gGrfM
z()3Ew?@L}J@~66|B}N^CV6RypMHhVLcjgW4^CAD?yM4>6ekK$1r*o@<JSMi~8uJtt
z3~|muZo4}|^%Aun)2{-*n6=Drx2CpUhL9f|)GyQy^q(UB&_0$L>@Uk-|2O2$LmnJ*
zo7jc3BcRe89jlSt+ivt&rEaI*#}S$%?q<ioVbuPKEjB^BN;d`1P$TC)iV;LPokYNz
zyDG$*Pihd)U6|7HYi4-(7o;!d6Z&}=G$FCxyGdT0ky^tvWNSHC^ic>q52xutpK$+Y
z$oDcYVe$e`RX0@b^E(^kzT?+_y`KO2xxuyxSP8yqTw<*#U1@02&EWE6-=oLh&)&gB
z^kJ#2rna9<e;JcnC{M|jf4@G2J)~soGA-In`-a9TNHxo+o6;qJ-xUq^*v0ePnPW@{
z$|*66Ithsi5Kpc9ex>JxCsM|X9>CcOVznP`i9Gs%=mB;;|EP76Q6=gR^<)RHFYpnT
z<00g5UETG+Lt-!o!iZQ2&Y6>5)GRkd4bu!_)#Zf`Ty~Y{9ATMMMsADhIPNBZLg(|g
zz$l@FotR!5-(u+I>i6%lcu6;E*$1Uc5-Gla2Z_L=f15A@^E<=peFZe59FyX(RcqOJ
zWO<c?Cv{@ke72>S))ZSV$4ro9)zTd-viUR>S;s26D`7^tuOV-UWbsLw{(dR}`Vc7!
zW>c1qXa&@}uQJIi#7s_4ufet|LB60vP@}h+my&}Zkvtd1x*ENjN^xpR%cK^GALBY*
z7B-mBoq38KRcsHTysP%)`_zPnLhd2q50{awAY2dhzrTqo@?fO%QP^`Ujfp|_CfmTg
zS>V$V;^dP_qqOsuTj5dLwR|M^@hHq$qS#wpzs6N6C5HPmi<omV-nxqfy>YQqilMDz
z!i<emYJHPn%ELlzJuW*}SJ)ACvm!I+bcQuulwEdo-G7D$5eDF^5HWr9=pSHcoWh$7
zv+_Va8+98Hq!|TAO)ytSZOv61J|>AvW(db0ct9p7xQt_da2UU!a{X}?S4tK30Z{x-
zerxq*a#hHtHQDyfL<Vi<hf>;IwJeLqkE?w!$>8mx`5jH>(f5?4I(}SyyVU_ECoex$
zceFmdM^!%Ur)px_pbOW*=I1(WpFQmq@Oj_q<EC>M^|s0`Z9lC|+&xy5-zVLd==yHy
z%wM{=;yoY$jNM5T_}eI<iHdxJQ-(#NP@`p5Vi7Y>X6Qkkv`Nm9TlE=`Yq4-hkctCQ
zFOB$9ps<x!57Y$heZxqqu?9tcKMR0G)$<aZ3U}@NkV{slamq5i%KXoVa@$pa2R`q7
zdKx}$L1z3qVSUyI=Yzr7m>`lA7k{+PvKx37=`ZM2X?y#X>{6)@PFw1yPNCoSI<lRU
z>9LEroqm0jk*|~`n<1^NeF_(+hQ!62<e2-31|x-5n`>$*p!6rGay$BFXXjPO_GXS@
zv&y0H^THEIVd3jGGoPKva1M&>%R>#|pn5ga=9^A`(7xNICQS{Qlw`1cu~WZ0e=me_
zXd&>KZ@sW?PFqVDhmX|El*5VQ@Cde=(4SYkYfYcPyVL~pz4XNZSG%`UUs^Zdmll?4
zUZ>|cfjx8;2=+W{&XY%gQ=Ep~DOi!h0j<Ha-*b*OA5}{N?VqQE<YUZ5G^(d&Z<stf
z!|_uAwsW9BDX9R!qO^<E9~olDSo{DgQRQ~2My~jai-5Jk^c0xX&5KWZbjUuJ7*N6n
z_Notj0EWmKw3$oK6scwYRiyz&`6LGBEq9Q|;sKi}WI2i0E-)%MooY=)`<frF52u1O
ze4n99>$@$uOUu5z4<IFoqQb)51Jf;Myc$Z<UvEyD<m_HX?C+t?IK}KoSF|Tt_FAkK
z+EPn|7K~sXc<e3Sg?tIhZIQTehj^Q{4_>dboSSWVPqpvToUazsSap!)_I3EU<KBJ5
z5HE3=sG2LRMLLvgE1Jw*Weu_$=NqVvB_4(p{Qf1!2A;eAjU-(9+HL8rlKn4AXzM+i
zdlFSeSSutjR<4p-gH-1fIyAUGKUWT082?d{xyCI!zu0>H-bn07cgkGZ9AUm_k358;
z2J2VbU?rg2S^$t}_}<fj8Dx2>Axfv8p*$z(*y#<d;CGof0^t+(N5Zca>n?%wZ%CeN
zZ!tF9FS*TCz@iJ2(tPRH@wEgH_in;NIsbh(1`?);2Q<yiw+96wD`F5o@?^Bz{C%38
zfSW%BNBroA3^UmdbR8X92P%2%P>D2_Z*NW|W^MrO^c`3?j?OPOI1xac-B{;UXTXdz
zQr8L=-K!wgRPU6)prBX;+T65j?VoQxiiwS_-A=re!OO32{4xZn9DPqh&<Ad90-iK`
z?bN7I`&Q-hIZqo6l^JVzIxQ@v(vHH!Dbqc-g^k$IupgwnaGF*cz&227z<TsOiAB#Q
zNMbBTFutb!z`n71LTPxtCek$RMjDaT|GdzA#b&f}z~dy#o2oKA*I(WK*GG<k+@S7q
zmvgUXW9}f@3a7_w9%V}NnaNQ+QAI3{)b!^ONq+UoS4%9N2A0c21)E<Qt6p!mG4}qB
zqKD~YQq2Sy)u{?GXZCjYFC8o0irFB@&vuqV8@(plXV#mucZpB47iWHbzHDo5oAng+
z7M<-dx&>)NF{a(=N9)rKD@ijuCzox9ur1&`s>dJn%-H4pHFsL=Vu-niEFkKweYGMl
zGNi)U*fkc@DDsL^5N~+Ms0}=5Q-kijzWDnmRm!jkUtI2vcCx%L*9<E}f&-n9*xgn_
zI#rTRzX&J5G);V*^V(AMdj5-f$kFD62KbHI*6)+tf|#TLd-Un!=N~7xkw_6&OButu
zO8ly>fm8VbYFIxFCNS#~G~X{&$#T4|@N&9Hj!bNqOm>{9(!c+J&0^T2N^1I+r_H_W
z1+S%uGn(s*imNedqcZ2#Buf_NXAO~JJ`9;$rgxizdTNxkt6ZUyjFr-%*fU<I;a43`
z9zTH56MiFd0bGNo@Z%^2#j(+%<Zy?V^PRB@ueEncM{BeRDczKZmRoZ<69^?L1TT;3
zda)lHi)Lb`;jcECD_H+@P_SNji#pZLs`aat?8y_1u88yX1oNI#NzmHp#C?W>c7L+u
zEwBjaVwQCYd+o5x3c~IgebVEmJYU>JVBk@G!Owv+F^8vNqU~a6M(Fb+)0W`^{<)R0
zyHE3}?`H;8lhpP$dF^ug(+D$I<VT#Vo8#0lY*913&fz}PjJfuv)~N=278L+xScB*)
zZ$}<9Y*h`2s_oJ5EVP^c`fRv9B`+%v2DZ;$36MtuKGXIOEk5QV-OGPv$wBpbL*<+I
z08iPG$m{&FL8%d>SrAB?><qklNuJfT&0pZ?HAtj-UV_KmmKuged1G)BsE0v&tK>~A
zs0vM!=JCn+{vC^2ID$%$<Fi3+B1pvgtw6tE3p#SkYldY?JKu?aXr7)N;|UX3SCUnG
zN6FVStTv0xjjcZu^3f5z8Ph1x(}OwA*GT!!U%tVOJ>I6|^x<2Jpz3tDdtou?Hd|>K
z_CEUy6c$i_yyesyS(~rYzxp{t$&NRkbI!R+Frkt=Mi4&cm0%&a8r++csI(%HggLg*
zz)dq?d@9CTIFNiSR!f+TmAJIs-+vD`s-OS)eZ_^M?$yX5rcW<pT~-KIC@A~`0va3j
z(IR<$B4RE-m<smu>ZBOHqFgv>D#x_qL-7lYO=X}Qw2ikVT08800+)5~y>*xV)tY<I
zK64a<M)+8IDAInhn+N~>?7aeHs~2sMzbQfG9rLckH3z6^9BAAO?1dGpMqGgID(g%s
zR8Z^VJQ1HBwCj)GdN?GKqKNF~IT69nuMN2LoXLUtD>BHN2oMs*L~}dx;kO`_nNx1k
zo}N>HNxfzS+Ct5Oi>ED5Pzsl+^XU%-u=r*Y7B(Pd*ITUTU+@6#lVP??BPhB%P!{5?
zm!y*_9M=@%1o6+#h6cAut;8kg@#m|=sF6G60PRum)jCSNc=Y~1!>Xhiog|z{77wL&
zm?ftLs5x-Z!o{90Q_@4RjK5PVEJ(wooOLw0q>sRA9neLDVOVSAm?RA=0vPS(kZ0#P
zt$R5KBzhJU??U`$^<EL9)2)^^)N^yT<jnpi$cS2h<1WO28l(AK6lF5;okM|Hq#F<Z
zH*(@Ng9gP{Q<VXkUALQ|uDy9Y4{69;x~=b%u(LJ~#+p4dAyGUW7<x1lEBOAs{_vB$
zwVutHiXw6pDmo|5O*41g(@e_RGt>~7KJBo{9{BuRbD`t;eSt2AvVDUa@u2c?hjG2f
zFNAgDE}=b5hlfM00x2n(WG&&mx?9ONrWG|lQB*qfC5?FPr3td3XvL-YemvJnrI=u0
zXXmcOdt9()h4$SxBS?3?r``P^++n<A2}|K<K6myYQ2}3t`cRVrQg-<97Pf%^GT+=P
zz25@a;VunP?>$yz7WmqF{f<-!GysHHN81AYGBLi3!>M~t2LL$c^B<mGBo;7mc>hti
z7cqWuOS^(l*e+`XHDa8?vJt#3w?cT@nrSnk1D6jMtTDeYYB7!{RT361i6*pd>r3pm
z93rtUIfwgicEG8ROtG|LzZLGahYWppYJ=WI!b}WxKiLjv?hTZ|$XVAI$l?e>NF2NI
z+gZS(*V~1AwrhvPuXz*~oJPF<c`8T}Dw5tUeyV)W83<!{ea;{H@Z1cBzH15q8S%P&
z!P&w6*VL~b+Rm?AQfdrlzq;2Mt*PsLFzY=z%!4|z%HCu-+kA-AaO}(fu|L5QYjo=h
zdz;xgO3+D<O93z4^+|=_Y<_o?N;W#K7-~6`jE?j9MFC499&O6!ncDgQE%AE{jt@TU
zMnov8;OVcg6{t?RX&1Tcz7PwTc_0xqDvdK|j>Sw>>v2q^-RyX`@#Naa^ReOSqjy8Q
z*N-ioUFf?HLZ2wTHrwzEqk}2hyT0|Ih|iM{F+b_15#r248x}>RQ-Znb<*=Pm%^k|K
zdXFK<x?Tl2D(*q27mDo21Z#JQG!JArqJ7V-pMvrJAyc?>kn*m-1RzZgyCi{}^0fOr
z095^w>4UUJ7PW_v%K-elXft)dk(SR^Wv}SwJT~aAs<KMmodjX&eq{dGy$dOYWd^^)
zUbnrM*FfIIUq2J1D^bsoclfusuZdc9=dN@atv58_IgtIji`Q?$dU4}U)#BO(?50vC
zFq@P4x`=)buk;cZDjlW-;;ZyEub;kg{8A|!jG6A{q5@cp-Bp3ZgpsitHOu_NMp|PZ
zbBV$}7GYYEpnMbB)L9IKX4s%YtXcrEq3(Rhd+X^gm9Vt<ejVYj4Qkp(ubIhB{6GDe
zPOU&|_uO<|S@P-$VWr+8sFW;eu4kIab&%nB(Gt_*Hv@nLj<4?#k`K`~mr_WZ%)2uR
zGcYkc3pzmb)nP3a`A;m1b<-Be1_{kn+r}F+b}-Z%-ltiZ#fu#>{a4?kLdE0)X9DY{
zK6Cl_xGw|RX!K6?-ZWROJQrE%Be{+B?@$Qycut0F+yYlESe|QqkIYSLcqadF|KS1U
zPm<sm+4tjx=6#-I`kX7y?O6E8SU!`Ks3Vf2z1|YhXIA2`Eo}>`)$3H_9PVtagzi7$
zd$la9!c6CaFce|n!nZg9_JA3fW`{IgGWpUbfxytVvSNeyRQv?q&owYJ{PI17pRSmb
zFP#ynyOHdhh+iL#2^{5EYV$}d6XxcP6Rb1<f)|I9XRoG1ini<!ysUr(0|GOXnFr4~
zo6)=E_iCN7n!R#lcRd@OAS>c|jku8Pe|1}eH<+@as|h@z>hpT&(?@$Gcw5@KJ1y@M
zaXDh|V<!-{JHkvX$!4g!B{M6AST+e*eU#Fmqz68GW(|6<SL)B(L{@GB2*vJf_ySkB
z6K4+bD>|!3);~bIc>jI-k+?l=*($2LZ)YLj$Ri${2f(C~B?#T<M1M{pd+b?B8`EHN
zU3O196xfbK=-HV3pGuox*>-C_3w=*0e;-SZJeokaLt>wu$W4Rt#MpPEP{3uBI&F}q
z%MFk#Y#g|V@|6`o)+^$gf0#!-Mw*o9su6PZDKcvMwg^8N6S}a$2U6Uami8uEUjDT0
zWr}LFrpDQS59Gi(mw^duuV2kgYE&qmlB*LNlpXD_`iz0t;g0@vN<gDLQj$d76Y;}}
zsJCj{5yizUm{z;<`8SA{m2v)7HYHatvASZOFP(OYXNgq-3Z@l2^9K&_XZij}Vb^yv
zput(VQ7=of!t4Lt##_}Mv-Ko>!*<~NW9=jku4p{vfA78=%=dQ;+7y$oT5n4<!=J)p
zC{Kk~@AnmZ@g<1oawXhTSXs?Y^WHd^H=_7a7)MykRzlq^*|ez7M@B7+D5G*TwRONO
ziCtET=DFVqbtEEHpIqC>5S$X76HgT3{2TxP6cYlBznW<<smv?9xv~puB__lRPr=72
zJVYc1fRCe7o*bw1@5ix1rHuqaPT#Kp?z;*^@tKrE_vH@zp?(JNZJKLl$`k4YhF4f_
z!)XhvKOO8djk=eWiFnTsi;7hwP!JT_r{MEzYK&{N;x4P4l^Lud@ks8jzpDREGq9pS
zi0*b0s%?K|owDecVa&*^k*qf3{TA=+Kp()ukpv0aCj6C~QDfA^&&Jhi@SIfI=_e2+
zXb%Y_Un_7p5Nal5eCjqG{x=<A<%f#-?ujYmC}I<B$Eov;!ykA^>a&Ej23bEb@$_c>
zo$J>LvUchgb+`V1?f|b8ttBDmtjk4Ax02uAY(CJw#g7ukkl{L}7|GY48~j^`VfbLW
z=K*4l^3TjBm4&h?`+SutC(S|hmIx7xk69mb$^_mI6r}g#ZS*IdXa5P`X`=Gn57q`d
z%!<-f@(6i~n4S?iE8kq~0?h)gOupM~h`%bQ-=(d7?aKSaF3FbP!YttJCWHo@3&oQ8
zl>ZnK!6#%wRVIyq`4a|jD0PWEw8Pvh0DeT9^ea;I4%J<%R?VL4x+zV0{=eRhD3sv8
z4-N?M@Qm`@LTCAh1@IrbXw+AIsrTOEIuSGDc(S;BnociI!`z%cxV<O|BN+1vyi&Q)
zFtrPHMDZ9Bu9%Jd3Yct#K(XjkzxiB0p9lt2?x|~M3~NDfl4m~hDw4O0_S<PxP_Hjq
zNzaks=mn-j9RUB!vL6Fz!i|v+;f`}n{6FjLoFpMhm6pBzfC=G%f0qkqYqx2Wv)R8X
zD48~g)6F;Qp9KVLfRL5s0W(4O*TM+;)4Fo{Rxa@g6FbG_K?$#Z8B_Lz$+IL2fyOI_
z_hZ9Fm{;H^p<QVq7o1kmZDC=70D_BZ9h;TaT$4xNj+cng0+5n8!KXxDLIQdn7uv2>
zn}r>EHL`@XSOs!3m;e9=Xee}Gz)}2cmvK^WHzvQ+a;Q$hZ2?6Zv9t>m2_fEl@5NGl
z(_%JNRR@f+uzY<#{qJWrb32xnK7bG1tGpYDy39X8xt}3yjjdEs84XD8p_He69=pTu
zUWsL@$i7*#L`QTFZYiaId|Fve%(zx&=|<0~NI!(1Qlh>@M3PZ?4*~?puHIgi98n)e
zKEClb+LxeVnNZaC+qZ8-goNhe%~H#ADNa-?4g@&U@^W(LJIYmoM<r|lmJ#^tIQaM)
zRl}dVy1UbN!wE#hdlW^cWf&=<DKi~*FL&-w^5Vyt4<4<@h(j<)Nl9%kJuk^Ad(E@L
z-J5S(PDfZSlX!;)ASK;!i1#_~G-RbECaEpE(Y-qU@G>XW{VC0$SMFwKEAVHXI*+u~
zJI%bH_TQ5_2k)otCXv+Ul<|SrGl^7bgZ-tsQC1KC0Q~Vk4-QS#D>}(IlXU3>qOT)G
zZYEPGJe9iGSoq&9*L`ROq=fOZExR2`vP+6Ey%}8Tcc5E-eDZPrj_@V%>UB6)WOf4W
z?U68jq#sXt&TnK5MM5*+6#yZOS-Zj2TqNeNA;uNt{@#JJk_suPM6vD5m1sBe-|t4E
zKmna7DqOO$vK~N-fU3m=%4w4UQmFBJaf{c+Xt(d@-`EErxNxL1{beb`p*#WHo`xV>
z!UCELa+1MzIZP7Nb{GH;g;j5}Mfd&o-lkX>b)3U&LnR5TCT(Od-snV&k2?}T^ysk4
zdSKSPPRwK+C=%mCm&gW&OIiyvBnr1ww<4$8;`Zh`XQhRV)-9$L7;uq7uOD7Z>5U^0
zirm2dXSx#UQBCAB72mw;;Y9Zm*(U_*dGvOj`x2SNs7+c523Nr%HyM0BmHp-JZGQLO
zqLaA#T97NsvB4wj(E^}J%9G8h^e{B02)jE<ng2dqqa^z=p_E(}>C(4i92$)tLo?Xt
z92tVDHnTx_e4q3bmt&XNJ{H%<U8sqhe$me#Pf9!F{G(6xYVgb((zf9LvSqi%C(f*4
z>6;+W(2TkXCBzlr6PwuqMHA^bq&CZf0fke4SyAsdX!PY+dc!Uq2a2x+r$EIKlVY2B
zAu_%(r_t=vcHYzGZ{Fc2M?;`F4{jo6WH5zF_%*z~^q43Rh{IKWr_LgE$YM{oI-ux)
zKwM!9nBosWYB|R(YvxUTLY8n`@imf$BslN=VTOaCxlkpwA6aGu0!T^7yhFgzEieS=
zC8BY6@nX<*lXoKh_?dz(DS*4v4;~j>psJ~1Zv7Gi|H-;>t8Z;U@FB9Nd1kQOJq`&i
zqHQ7YD#u`?OpN4C1q9DN0FK>ipgn{s$5_V|2f>%Y3dOOqQY7j85RAA=Af(0N6N=e>
zG^&!5f)M5@23XDMO+L`gF2bx4@{IK6aHMEd)9c@W15%14LLux)(gSaH$2l@$*t~42
z!}-nkSVSRM8$;!0pb-2U=g*(!JoIw5G-aNv>s-7+<Ogt@rcbA9-G~iD2IYQX<P@U|
z>XTC7oJO`KAzR*ZSVW!3>52&CI;f2mndj?~waG{*L?6|;E1=t|OE(C={%sa->*#$8
znlYG=X^olLT+rO3=G=PxLjQ|dFp~t-WSPUV6?~E9=>fJ2PbylLr#!c7Ck*fnK*DQV
z0CNX;%k5I><e}Xb1k9oqF$;#Ss{ntVEFF4(O5iFO5dh+Ezslf^S%O4Vv;)9FTSwi6
zjQjfv(^bRzUam+%Ma>ChMe|_IzzowFHaAK}{i4H>ZVZg$<)Br2zD`5w4aQ&=*+hbQ
zgaF-pm~%ufuV`+$hP3j&Dl%>+6Y*$?xpyQx$pMq1D-F=AW@&FP7jD@hAf|YU<b-+B
zSRH{BN8P~HUuYC?ux49J_kEZNehu?4e^>qU)P9edMTkPjJX>Xv&{B1toA&0pkeFkK
z3XQ#?T1Kwnvz)Xqs(U#@29-LbPV+3eb5p9)A+2-rwy4F_S@#HqUL8SE3y%n&n#i_9
ze+#`Ph99X?EWfSaL)OQjhCGEMtA4hZPi9=|sd+IUm?wFk&_>L!Wt-~e%VM>gNQs(x
zm(Vi9G(Wj4v}u=WGwqg(4$L>kfFR1@fnIowInv}3LoGJV3Hsqj!YoJI8IrbWkA70D
zhyNfFR>5SY=YS4bAV{P?0?UT2Sh%+ckWe%<$`PQ!CMl-VXF-T-J;aOw)|oW~5b)pb
zg)87K1Ctx8h)vLWCe+=Ra;ZM<cV+fUwp7QXK+vT0<t;Dt50OQb1^c`a4qXP|dkxk(
zO_XQjPFah{CXg+)0ndH)<5G34cbXP1AvMcdYGrIvs!z11)&Ki=ABDbY+~v;cQk<&W
zE#V+cC<<}YW9awe6Dcl<X)3Io<+sVhf{4-i;q@v_n7dmm)?fTe%IGXA==@HL`j$Qf
zs_S;pCbbmyl|(&KswjXKh9gbwBQ*<m=qy`{Z-EVjhYuyB0r)6OpW(_U?|nFa7eLgB
z1V6iY=;K5nt6<i03S7REe*DG*Nc_m28~UrRORP#H32w_KB59!Z!x4HHo_37Z&bQ1r
zd0Xemb`-K4kq*d$_X_R|X$9j_&OEw41K-b?ub)7(&@b#h&z49wig|q(8MTxj^W!)g
zEM1Q(m;|i*N#fVZ5(l0zF~s~NbecX-{6-UaCI#Lp@3hI3PQX<`2`~Sj2d+G=q*6px
zD1-Iojb5aR=ms4;BCuMAMo1$+diCA#t|KRYj3IsINe?{u!~Uq{ILbzl2M45G`IQvf
zbB<Fw*y*pQ5T4NzhYMk>?-HRug!`V@6p^usz9b-5B&N&pzOzrdRNIMt0_bms!K9dk
z^DN2?nxw%kqHqG}=uC+SJn-NG-t2xJIFE|z!o)~i`2bFzBcg05`85k;)TalN&|@GA
zgJ?y~YH}@EME`{UL@Ec9f~h9ibs9X8rvw<X4)emPtCj@`1Y2A=9my)_QR_iELZIdl
zWWz`?nV?BVavc&r3OMLgdjb~&GO~Q=iGz*5GY@?-+t2Bq$vD%_8%sL_PV$CU{BaM8
zs&OEFUOmDBk4orA3bFYXbeq{0dd4Ft`Vp8rs$?lNs_Lb(OaqReQwl^2@`)94z0Kua
zo749+4NL#<R+haIX{!GF(ps6KHss~w^Z>nvd6w@o?od^M)DG2NSdjI};RGdjtPBg&
zX*)k%vD{&p^6@CzP|c;0m;Vi>jxf_?s(y9=c|$rfVW<HiL}4SqQV>_STeOu?3^h}l
zbcFxFJN$Nfv4x>K$bgm`apc~$;q;bI%hUxRmVWt#y~Men;!S|msZ>z=8k>Qo_S08v
z^K|H|D)pg<CUsy>U`V7eg|!w>hOa^YmJQKjju?J*tA@#Py)x2aDF}=l-&DfF%cVll
z>uCCgVD1X6Yym_!8!agWj}_Sp`Y0H&9=r%b?_*LRbPEZ`O$?WQBH#81HF8!sg7Mrh
z*@KoV$%n2;hR&x$$?jfXCqbFi6v-}qSETDWIG_8i*;8I$h6knz_J&~5H4<3FtshGM
ze$1FSjM2ddvl+r~MbRyf=KW%p%MfU008ECL^utLveN1RHL*aroU@x|_$GB@I04fTD
z&!qdY)I4&YDYE8$F6q_2cpC^9-j*~E88^B<5!@*Id1X)Cwp7wqk8E0Q`stk*wj+tl
zU_2mQ+!7Ril<?m#>P*J$GT5@H;VQ_Z<Ej=5>eOjlks2R3_wtaa@tggN(ufvoY4!M?
z>}NbUP)N=;1Sw#41@SPW@`_o8I-Zg8tP4;q5i8T5Vb7?3zFpyFVcZd1k>ef4?Rbcu
zi^K8Ve<ySLsYehx?VDoak#x{*YbEKDK8A)O<o&u&$-^eXC5o-4JiG0GU%amJNST!L
zyyKok_3Heuch+C|Yngfoq`kjZY@YNVgLAq!u}xXaVvM*n<iIRhm4WmAJ9u3lkx>`S
z<eC5h-%Lx9T7Q~r7^TyH)yvC<r{MK>F95hVCEwF)ih|4{v;R|A>~zHEkMCeur+oF$
zG^0UWX2Hr%Ly3V4F$KoSl9s<x5{q<_ffniq)J3n+l4*;eaXbI929VK84xvAO#YW*n
zK0bWFf1q44hZrV+NZM!m#~wXgGE(0WZSBob^0SvqILn<)T13*D{7H1no_t=X7Yv=H
zK9#<XZy`T;O~40Ndm(C?_FV-B0bx+pDR)g)-E|L6W+O4rmdCQ0iDg(H$g1@lk0b$r
z8DRZDy@L`=vq^Y|OsV*<U)t9xSu-Hj3-pOYw7VL$B0yN~maS7#Sp=f*PP8|rU=ir+
z1W;~D_6mikXJbzB<Y?XRl6X;x;R<e{#Yqleq1QqpOG?212F5@b|NhP&pLzZc|9I^2
z<9we^$IoBq!Y$D74c{B&_nA_<E9g)>wR|_I*Cx}U8qmeU%$cMv|6?t|^QYsa-TNWq
z^+NE|hmhSprd^wLC}#Lehoqguq}9MvTJiv{&Y6`2++J36oV-rI-axBvHlr;ZCD%F~
zQ3@temk`N(<=z`G7d%NA;t7+cy6?sM?G;|~6ml<K#)B~;?K3nnDGtbYGr2n9G8}lT
zVi$6+a7TC)IX-$iu7pNW(-R}Ho-cXvz+HeLbtpix$jUhw*fUnD!!qn#7kqC<E5$uP
z({PYtbk%YF(Xr^>drj_{k1sy^D(`O~E9R7Y7!=gA%Sbs4^cp>T@RwvG37DF0t}mN)
zP_O)gBM8`@(~cT-e$9D|dY1lVC@F_5ivcUL=uOkPEqCA&sn6Dc#CQ_<WU7g6F(yI}
zDJ)E?6yN*LB?!Qj^rfXmC~4lqu6k>A3X<+FB1llDKLCB!d6MREV5$JjgcwV(72P}B
zCHLCO$;d1LxtNJh`V0VET}@pHIxYR;LinU;>7)r{I@E!+&|z5R_u~L=fr3dN^Aa@a
zoDyrayC1^HwCA9nHy|}3{hgI@*e--hNH#9R7$yvVDEDFCn@AYDk};Wh_q_;4&R5WJ
zIU`;o5;_jCcOLpp2J!3>2WJnzSrCPAxGcuy*+e3RK8v8!F;S7a-Ivb__gF)X%OZAp
z+>^$36eF16ZdP*+Fb5pxp<j82XDxg>L!2;e-VY0Y!yKO-7c~IEkAXGXu!kD-V9uu4
z&ycCyPvJusMi-U<(<U4OD8_@331x+k$alq83OdQ{LL>tIFR&(N;?B6s+eLGNdPC>N
zZlX*;6wt3)iNL)qB@A1~b|EhqpkRy=m2GhwFEju%2a^X(+NJXlv(n+`$9H06Owk#F
zF3ypY$W8O+jP?p?K02g%B~Ao^Tt*|}{N+!iNFh1XhH+g|gP@Rya`T)K@y_B0Oe?|=
zY#Ob=^%gLChca^XE!uMbjGIjNY_i^)*D`3oo2(E7FX~Rnpbt@^<~(Pdq=W5a4i`hO
zK}Uz<0)fucV4M(!^$;Q6Vt|yL%97b#9zT+)+(4evP88r=J>>OHsxXa(_=Na*ar)@>
zY{qm?+dJ2hn7DQ4zPAUfcQW)xOQ{5jTi)Bhq6pG;Cw}4QwlE~qTd>mXZZ!i=&0<Vq
z$`gM{OmKg2ciD)q`Ti%mSA}}lZIjlBL?TDPDPxh_e<_qqw6>L&pM-#!1+Pk0sH7f(
zQti4Vy=i<juMol5+qm3XG*piGo;lGRX`6nycOW2ZNUPNX+F>o-ulC$5NCLxN#)92q
zSX^RmL>f+*V3MX@i$9WQ+Zcd6g%;F<0!@Ki!iqV-Ms$uwU*Yd6gH!E=(NiA+nEfQE
zvXDv+tCU6o-WR8J1(99fm)-_+fh4{~k4&^8*yTP#Xb>rTGhpj8eGRSir9A8;XL$is
zq02UJ6Y#(JHjXFi+hK&Cun@$*X^PIAfXOd6GC+oP<*LK!NXRgH_>Qom)QCP;@*!Q1
zU|brP&h4a)JIQwUghZ4iGwoL|e65Fewcsr|waY_&^1Q-COzstL+Zgy(9ZcCpR`M9|
z8H{TpX3BS`u<t%7?K7E>Q9!ol0+~(b&N6r4mAB97oY!`|C<{Fwv2@7Ka^o4_1=vOp
zafIN|OasKV3vV}#w}T&9RqFn*j|_^r^f^Sx+`EI|W8NcDN=k=N13ep&9R8lgo^PI_
z$uU2(UAiI&Pvk=a2-v|{K*kg%v<78K@L2mMY|;kQ0wmZUk#3t)IY1S^q}Gib@8Vg%
z1H7^$Ye~IVvZ#V+x*ot6i00_5Aw=U~QbYIhtZ^3Q+blp<XM2F!@#YQ9npz8lY{mT$
zYPEbE6E8wHK-!~{W9AoeozyEjI#<J~bu*RCDi`IqeO93qXF|Wl)ATagaf-|#BFVh#
zu027A;tKzx07K#p!Bvg)HpgaYC?etR!l#t&OUuI3?ssV2I<;RV7iG6zLE7fMZ@=?@
z*Wpxsd*3=YsF_vgFhI`P%f6?RGmy($Q*no1>mlbw*Qs8hjQ3uT7yUUxHV1|cl!f}4
zB7fV1M;}uH2r+rcSQU)v0zYA?*pC&doya{`_{+*DDIotQ^DQu$>ANAql=N8g(ysOn
zpHb}6SUfIqTF&ddR4ifC$fU~{8tB$B6ZM#7mHbkqo^5|wV$@j0WOx*L8b2%LeBV1C
zCm2c&3V784y3{(=ILKKPvAU$xHHLAX&mk{?5O(YQ*0_OBuj~yLAmZzDE23k?8ujgD
zN2?AX4lBjjl5{BLOqUiDyJrY>rOh2|1wvKIGwGLogs8-r`JLgJv~JjRURR$+@mEd+
z_ROzFZll(}#Bbiw_c)JAp8SgUNQjeIT-;tP+_SSiD2^qHukGoxKLX-PZLOzhqX#KL
zaSszo8W=~6pD6CL918SbN_?)*Y+xwUZF`T?RF?y75PCKtMAn+hNzO)?mrjG|Y4^*&
z(-Ipr45RC6p7n|b%<&ZBn}~M}6J8MnPo>6XuduKhi{3*$II~90H|Bfqu_^!F3-Cq8
z%||$>g~2rhkLQ_iYlL8RTnX7ipMN~z!(p$7kZJGZy`#qP=WQWo5{+XuYTti{?8=PZ
zeVWftfMV(IO>pjPP1#|4CI3|ZS@usJn!ACQ=N#_sl`&t9+G&|0(t|jT{*w^}00^qP
za4JD=n01FU?+g)<`s8dU^sK<Pr`r4ak(Txs#kB=dmPdV4rZW}?2LxVNuR{bs&A(D~
zzV*&y3Kgy8qI&Rn#aDngr+aI~sef>j027sazd{dkaxkQ0&P%hg7=568_PJYq&g-G!
ztjtHG5#X_$pxvk}(((e-siFsi^n7hVn-h2&497lbQ{tjkMy-5`$-Qh{fB{S09tK5g
zb~LDo8InR`_#<#7FKVd;OWFt|JsQX=Q^r~Gzwg5ZP!p&e>E&N`dU4R5A+30?!6;w<
z#8_eglTm@--YwZ0`@OQczYnIU)F=p@{Jp3N+9&LcjEV+zc=dTjJ6GQLPbM(A5d>bc
zi-wII-U%qVv`hS-Og_GOP!UWXrN2N!%%zK$&M;6%4gZ<bm374y@8?&4yOb%i+jY*;
zxqU6DxbOm#^8SGAFT(FqP&5<ieDbF)G6<I>7Ilz>myvCweI?7aXRILP?}mVh8~J_>
zIGwX?jYwvuKgtQbrm&8Exj9Ivlv*cgaM!#9!9(P>P+=;e^3vEZVAvP=rnu(+LHN%^
z#*0eOPRGRp$;%!j-+}xLDeSBz)NV5RlKhyG{WvpRG4lz-V5vKXlvPS0w)~{OSc`qI
zW5my79{23o=E8Ay9Z4d!HiNAFoO$z>sIinA4ru<zRVq|<Dqcu`GjSr#Uc!nSJ+o!Y
z3e<j?`?Wk=jP7>)ZTH^v`A6m?jWlWiKOvBzf2`R~4f^9bxDgdJf7ZcyYn(_EZVQk-
zj=U+{qqrqxn=u4|3kJ2$xox(%C>OCw{0*D>|7PS#2p6$}&VLiwEig$7zmA24adhZ7
z(v*nMsh@Yag%{;4h4L(~$-DV%YoEBurwb=ksG2ejtEnvD2D}d>b?a|=I|`$xBFW*#
z_l0LOtNutI2xhkZK|z-{^fFqITWl>Nn<0uy<)+@Ht;j34L^;V&t&u@A)tmKPy%WAF
z3f2q75V}w5LW9*>WL3!KH~stVIY$50SF~l`%Lcn9rKW~8?~Dug-@V+Y5+BJc67{pj
zjZ_ih{3id&=2vbrB1Ma3AP@0|?Kc4~;eBUB`yl=z>bs}-qN;YOF<+1v!^lPQ4lZbh
zZ~{AO4KwHqf;?!9<eb1CwlL^dBTx#KM1a<Nv1?enqOoC2OGl2FpKo$SD;C=&SlHTA
zp7isOf(=Bckw$ssX9_=%I%;|zFpQ2obD@(-h5tLBv!bvuXu=We4~8h!qfEY1n!Fd_
zSMKu@6!^mQ)V0NN)=#$lk$`U9c61+;R1f5mCwxKw2!ZM^@rC({?`SQlFqFq43_+65
z9+evJ-v%UJ8KhNcBg0=}yjV-G$6Zm#HlA8+D{x-RRbDOw<}2>iM?ud6{k&X>2tC7X
z3?Zlc092X0WAduFR3aA!^P(2eq;`!&-G3Q)NOu*nOsjrRsoPJ;by2e3aH0Ax%14U~
zT+X+6Qvbc&qhWVYI4><!jT?<D!Yx)lpStcY2@&Vp@;f4~GaoABGC0cYSkM~KmZZ5E
ztCRCDTi)~RkIa^jOH+C9PNBn(ELHRYPcLN~5VAoM!w0+T<v7V*Q`AHq?(dJJuG?Pz
zZu;+xj8H_~V+_mKYsn6eTfMJga1)_<zFfXG>M*!l<-2aKLbQ8>@bcM9d<<gR5FIX_
zr%D!aSFRU0!tfG81ySn*rEF?7((di0B^@2qQQ(;GI(lE|^v_)lR)zzFa7t==>zE(i
z3zzW*9m%Wrga_`?;=k(Rh+S>Hy=I~rntJ)p7$dORAjNTt>t56NJBD6u%|X!t%dwrY
zH<F};GWLyeG!o)!^D=)36EhXTMtv!u0HN{Xv9MpfLMkr?xLMOfs&oG3se@%FL2$$m
zQSD#j%CpXj7S4NlS{L~mnIE<D6q8#VlhxP)xE)qGr$qg8a2oU4fPpEKstN~HuepC8
z#74rw@y>XZ5C54)5C|dj)&7(Zj+Z1A8^iQB^9cLf!U_K2>Pd)+-^Z|uqcr%SvIxC@
z@=Vsww7Djxy}3~qif$M>=FyC=a`~;=`Z9RZXh}`1IhS?Y7<|8yl34kHc>l9e?WHHM
z5Vd=7f1Ua+UJH~7TNa)SR;tN9y>C@9?FB(1W8u$SW9XCh$>B0QBSJV9@@B`<Rq8GI
zM~q@c>7xg$t?qoSo)0O0&JG<hr?0mjsTF$E5=o!2{c`DdqY(FNg_4VcTL8J$HAZzC
z7TnmNFX;W(JRBuir$8HDdRHq?vEcjnfs>H$@9#>-<nDUsytdOD0wpUU+=qfnw;NBy
z3b;em3+Iqh1E)0To#AYn6n_5CmW}oYAuS8{Aj$F&hkSapxYI}Yp>DRViisqVE&BqU
zn0QvMA?*(z8zK~vdWF@avsb$Howt!wySs9m#LV)(zyd05YCnan3z|Be{tf@{C0gVo
zU^Ntyrtfe|)%4qZ$h2B5lIbfNZg|?=|3D@%!5G65Z4@l-&gh^1Z6x1a)A_{~tkvx4
zWXQ?D$M-CXgpD51{)Gy&XIy_i*Two+0v2#J6^TqL%CBG7%&DpM0Jp~g@bwwF2MmWH
zFK`%{TzJ8MFR*Qx<Oa%8;_$f2AV(n$T!!66*=*4FH+|_d_6o*~NFAE}KY8sLZsa1L
z|7@hweN@N=cEI{>#jC+>qTBBai46yf8?RseiVH6IY}_pH^Jf2K4tSytkVO&ri@<uu
z4BBXlKN&TS?pA(Kh))5hv#B<~D{|>u;-4^DvBT~gOo~1T6(b&&!d515|Is~~_dk-)
zG$Ct_gwcKZ(4y-%RP#hUsvGfKJNBGp&pB*bBR@&-+oiW$b{cSmB=k)PI$x1OroBqL
zf*PMMv$!N#hP+VPqAXL*vhKk{))FFDm28^t-yDJMIXJ-EXnb*55Vr$9p}&Hw1xx3d
zV0}GT58zJ}!mmqWD0t;Ln|{-80Gd8i(2e|y5=}Z}yG!BAk$g`t5N-P;_zHOKZZZt<
zEx`g5N>)ow2KxArEf&R2nwqdHu%19}CHqpjavz1;7IRu*qq^x{$<@6v!uYGh^gl*{
z^iau>7IPx|E~69b$bnBfHs?Ngl*3B{*?-{4Oopq+`VLH^<EmFyH#X_@OEeULIt)xv
z>8_zfWLxi`yPk@gP+f6HMHrQ-hoZ%T0S6)evjLbEyHSJY-q7}JqYgkqGXqLM;*;_Q
zX>fS5K18Fl1rs3*S=9q15)QjeFniX3MYD7Bwm|cCtl&I>v;>{AT_qUvkmZHXfR6cd
z4g8ta^Bw58551@5<>mW<x6?rkmQ@dSQUWad$S#|jk<d?al4241(d{|Nr@-Gg_Ye;X
z8of-RciWy3vGe_`-dK@J%T~98H4hJtrrEvEb3S@a@o6j=aXm9m7wcPeu3MeQHzI;m
zTW5blDAr`R#){+p!Mai#EOk`6nAI{E7XpBRU91ZeAGrnF=E?brG&}cznQkH^HZ2MJ
zm~o4jBh0G`bcY>)N(ChVR?-$L0K>mDytKZ%ySopWycK*k58l)2$n^X!b5m9~^5?LO
z{l55L?;S~Sb-ZL7kd4gm57@fCUYMdeqsY3Os8``{e*XIxv-YI7b~h|(8^dC?>2$sS
z<D>8x?^t+nJF#_dE>@L$tyZxrT<v$(-7lq7zVrfG>Qrh{g@<Wccg50R%3O9cYSd=P
zJ`%njnR<`S7tOdoe}INozY}7&D8c>LI4-8GzQ+i#Z%75#SmY-suvsyScy4_$QQPfd
z%_Ab(j1jjhP6+_rRcQg`p>eYRbE)Z3&O1U1Rlh}4dxvdJd4R{I09R!J9elD9x-xp0
z6w2eV#;GsNt$vg7h|3`;BGojkM|rR<S%utcn8Kz)iwWz=u^^u`1lv0_=fWYW#|bzO
z6TsHj*FeW)`S-{Nf#h)~#+q>oCs2*E{$lJ355ib*l&Hdq+~td?^(HLnfw)X&`0s!_
zIVvvF4fa1DZ{9t2{xF~)4c?gl&o#t~k{%%m?dyG_EA*InI899bf#Ymw^0$}9t<WwO
z>QplsZFCNOx-lXF#k4HOCqAmlOMiBkUHLv{)1l<0rYP5izpqL_Ux|@A2Rq_qB#8^u
zByFGTwSi?pf%pEhrx+untcmXiQ!|babQU~ozL)D(m>36M?JG2AjnYU<4@1_e8o?%c
z6<|hELB3-qmAI3RrV{M;tH2=3IEO<eq#1@!lVe5s1aqR{i|(&n!1ZYa*($wf1I);~
zg_%q`{C|GH6yaaN8kZi7q^C3HFI{0wk+vViY3CEvC<Wn@@$&Gf)d48wQS9bK-<SFY
zSVKe7pTs~6k1$FGdU`9(+jZ-aj`J6ehz2&l1Qyo#r#b7F;*U3P3@kV4PIjLeQUer=
z#WK++z3(m1Y@V&(Jbi~;=8!MP4XD;J?wdHa$<_Gy+Xhl^@27v6m#z3>O#0oHE8puK
zPNosyccLchRokeQg22~vdb#8tilr=qZ@^wYoY-*p9y%P5W}UqloX3#P3d8zxrXc8t
zi)!HP=naF?M~neRG+0u{1|W&3b&zENZ>IB45JRxSMtVL*YwUFD`^HEsEKYakK$hTP
zBx{!V#qSSF(;Lda=n1UEac(4V^+n39hSdjowOjm!kT$Y^y()X*cLYCs>xkc+Sq3$W
z&3IC&!_J_xDj%~FEnk*2Rtio<?XRs}K=+%Vw2S7>slSuRwiBBJGEO%g8}lqN*i5RD
zx8@sV#4~wh@Mwftz2?0?voF->+<(AH-y7?SDY$dP4u=UVuI{fy5;@KjtBDMy_s;sW
zq#Z7!S@^fGFf&hPx;YOrs6pcM^P~<Ii8ThsxLR1{I^QB%<0X=w2z#c4{rs@Hz4W0#
z>F{uyP3G+(ykg+AU&oZ|0k2GEYWjpXp`XIfgsVvN(LZ^LN`ummGDp!E)aH5yPt7$a
zuK`^}i4%JQ$x*9g3?Y84G)nvKUBiCn>RG=Gi-bfW1Jfsk8q{>n^}(m20TB~aqTWA9
z%ZFp!|3KVGQO4^K&0Qi}ll}{7E_DBKnf-m`s@PDFoGHO|Z-^0YJ&39qn&U+z9fm#i
zXFMismz2HdE_j5riyW6fD|?}~Hk|8nnvfc*yL#~AyVci24VB@{iqw)8AL}U^l>(*j
z%5H(z8CBljL}yOgM0^!*12y`M#hUuM)i(Rx#+cz6c`ooS;?vWVGGVWGp=APa;^8(c
zOq+ZPg)dA23}I~?$B$Fr-Q_ysean3#gOkc}Zro%t@?xHsx>r7IGWLA$9>v*mwYEr8
zfxnn_K2!9>>z5odCP6=c<x`#vWscr-*bq@zgxg;RluU?uTA#s^R<WZ<IBej|1*Zi}
zX#`^z@9y`FX~i<~uy$SGSu$#&cd;k_VD80X^WZqXRwcyb&FP8=TvzI<u4QalZ%miR
zYwg9otN8tF>%gVsbah-)N5te5gA`?Flktm9VfDX3N|Ioz_&beiBxCZ0mHy`kFK`Jv
zCV~GRSdXJ|KX1;|SJXI8jjm}Y_{lYY`Em;<cpM-5kUtTWwdk^LshQ5=5PM<-Z$<s?
z$qXrw?tZCnCoNrJKc?;6^0|0xvQ166-i0IcHu7CzG--eGaS+AsEbqa<ZCh&DbPiV#
zjdJn_8~=HErQi^ZlYy+M+ob6SDoR>O7{89gG7+a}ho5_0dvtXQOZS&sMZDi93&@`I
z81B|+?6zJ`+IF0;T^}Uv^Drw0>YQDbDqJYmW~?jCx11NAI)TEy2I4CCk!}+^U4T@T
z$BM6vJ=t}p+{;kqLZ|R2h!w)l0;Ondc5dV2bsk8=EzRzj;!9Z=-v%|?b<sp>9;NQU
z8Z1Nr^|;S81#T?04v+o*0gb*i#hs8<yCC8?5Zp)-Kb2@5Wy<Mu&0Q=n!Ssq|{j7Et
zIS0+UfA%Q)XJU|ylfF|L2z@&5yU9^%++3%r6We50LajY}3Gn}^$lOQ-A;Xr4c_y;m
z&STohxE6SL4Co_Ry5_Fps34<Jj{e}C{(|b>u$7+Tmrn*4&Bs3;@i20YK(j4~dxGpc
zL=F|RFK`P@Z^;-<Z~PCIb>dYV2S;?iK)tC23qD$vE+4$XI^Ob7@uD~ztsY(as9v`a
z$s3qlwHl>dTGGPfIgZ)5anh`+C3?%a_2E_0>P$s(f5_=9r+W%}>S;JmLNWD@WhMs2
zH;R_xw8KPSe9%yp1hK)*<U8E*+HReZBOPS;-n{;f6bsi1lz{=b-wb0HY=}<c%DUg2
z>YW1DXHG2UR4oo$ddc71_<ExjT3T8>FMqtR;rl7jcn2K%8yD#}N&XodgR>&u42&_X
z<1@)gt7u7fWxc)y;xW5j0j@mdwD1RfI5^yIJBnCc06Ml~ed8WvzVBxg4tHZ<PuXPE
zbIO^z?$FN#&e!M#G!OAjOdvK;dMvV9bcw?ybt+0yYG$VyQ?^GUrcdIJcGxy2%DHpe
z2sUd)tqx}59&p+B6LcxwP1;@aYoX-bt<2gojgE{w^nMuWXe~}KPLTAlvD;fNfP-Ra
z1DLgPpI#k7Hx8Fn%9PUSwRnzt+i~=V7<5t`Q-6x!_raWT83zfha!W6^RRqD2UD6C`
zIDT6GVMzL)g9)Y(YW{~0PA_Lbip&9P8kCx%$#sKXIMa*MT`r01qoe>N@W@wL^8kuT
z`sBr62wJ|UH=Zd#+cx9wGFTBwD^6}8o19>(zc#|_7{c4Em6n=n%U0t|0tR2%eNgar
zd?;YYx!b}mnZlvZsDPv2TIYu^!>-{=Y51$mxl^42zH)8CQ@~{>l>5l+anK-Jfp6@g
zRgZ3xTDg^)ezWf?n}QJD<qqY&LgULP>QlQe_i{SQaR-!EqCycx3D_|#-0?vA)GPP3
zjf-h7YeYo<Fq98!*btDTB_PBM9MV(Im_$0jlKM+~^Z!!B|8uAhj=r;0pljk&$P`+#
zMW}25JV<3WaSHezJwWl9cGZE?z2!%!8V4NzOku49{JjiNh6%dSYXY8?XiA0QJ+fcD
z!P1bcL6@TJa)MX-0yD*S<z=iwrE25B!Qenh!Ltn>Pt%YdwL<0Nhs?!XlTcVn${|-<
zS40<-GTtU_J@oWT=kx4B*OB}x_g>G&&$ccbqucVEQtY-D%{xVgj)91|Q!#e3S~~~X
zSM*}N(}DD=I|U>7>~<uYlmbr7gv=zhF@4V+trTq|_N7yvGTqJKb79)9XqxubE_;|i
zc{newU1;E8_xh$gWwx&6OD%=FeyQGukzaWVnTcZ@ZpzwPtJyc{(m=f^1a4~FaeqI4
znP*>uWn|*Zh?TQY!vg!hD<WgQ3M=_8J0QnU{>k`b#@p}>*&{7+uWc?LUD;SF$h@h|
z$-|fsu7tVg8uQ}}g?V8rz0~8pC8g2z<?rVI(Rc2nVoHS)Xotk+FClnOHYbKj@1CNY
z1*x`NxibzWN_VL4pGgt1N(5u<ZuSwkW@dWk9rQ1HsixVT^?lKGrzT;9gzf31(6xF$
zB_>*4jl|fg-7P1sTL@T?(XN-0Q8`hkCUZ==ec~ZCg&m_!t_K()1-nU1T&g}e#JtLO
zcy&u71sXjncDT-{WA@*ehxfm|Qn8x9XeF|5iR>kc8(`S}*k2p_B+FwzL-MhE!i$#L
zY<k!OOh(-`X)KEuMz~_d|NUF6*ia%Ca3}>-0U!=zGofX_hw?gZ-4#GV#Qp7q{opff
zg`77`34tO9)gx=X7tfmC%oc{j&o=}*1*C8=z8i5Qk-AhDK8>S%J5^`H=5ujSgO!`L
zBl(y6dqZRgST<`+&epzX7Vk>t{_;0DKV5X}iJ^GGBs2|;$i-!-NDCK*h7=e5h$7-~
zqG0&bBTFrm37B1CNVVQsr`%5?jy9vl>i-=~+{jQO<_K(rfptp>AQUkhy@1A(B|Y7+
z)b%74@op$fS~O89PNO(=*U_}ndnMa{G4b(%a`7}#c=0Ig&F+gWK1kGIcSQEbRaaEN
z?qXBlK7CybB%fz9GdTe#`sct*+A3&0c>scQPD;jY{^yBMxeJlI_`$SLuraFJu%kKz
z*iP>ZJQ{H$W_8tvXp2+n17u)}e}Ml+{oj<4CaShP?PGSmJ$nlEw9Obuu4!Q!(l-aC
z^w6P_GrWO3*2OvRZt2C$E3xlmfIamp>xwmE?-kc$++Qf?^;NK4mqCyOB~fBE+O(}n
z=eE`#E;0z5NT3<-2Cx%XZ8>?b0#x$Slq~r%@CdGkQ9AEAx??NEQN4;*n!k?aP`X#~
z4w%@msQZDz%g5=sT~SETzV|QNxi#z;_|xXbCjD;uV!w|GrE-7~!<x*fxa`0B%!UF;
znRgu&;Xtw2)o#B*t}vnY1f}V|E-Es~v4TIZynaCEJp+Qu2L&Nfbjm~;tsUsqlXu-2
z1IK9+8(xM}nO5l#F;}EavIN426_OG^>ENqhl42SINeV528TvXb%{QHZrp1<A7I&Cp
zInEIo7`^t4TO`h@1H-A|gT*R&Dy&!3wgjJ$$?R(oXiLnHmm;g0%$LJ9G(dl_2vmmY
z*$FURYqp&&V|$e#A@TyJ0>Z-6m?Cn0;NNbw0T$7CIj?}5&DipH1Qn7v@hE&9wU7OW
ztyF*Cl>^Arc}1T*5-q7!&c)Pwz`v{w017Ij%=P8qYJAo;S^xY$S3{b1yz6-i7g0Ln
z!eiq{h`k<j`EMRqbKeWKy6HWc6Y{<tERqGmFRLe3rKhtFy1NX$Ee@p;G%F09WC}E>
z;*uKbNeld4g-_zB#SB2w){DW@%FP-#F9hy01I{-b!Im`V-e53H1tN*BVC4WoJ%BC_
z_|BFvT>w=<|GAlGVkd^V?CQG!?^RMxqnAoF<{zQ2cBX4pkUs6p4tESlC#&4y*L(Pr
z+C0uUNWk1BD|ctwr22UVm{+GxT)b=+X@-I(jJ=eAaD)Ib=i{S6ArKsV4379PpuF_I
zmkKEWyHJ7pkula(&X-!5M!riP|JUAI|3$q<UBieYf+#TvC`iiCAxKEc5Go*D(nv{z
zfV7~Lba#hzcZ1Sh(k<QH{a)ia>V2R4`4e7$@-vPzSIza^d+oK?8vh+@0d}(!1T~Xh
z<C$pL7I!cWjN}JjL$QO<$2l~+`KqeDQ<v_&;p0zZ1q1z-9PQrUFgsL6M)22Yvb-q7
zly%#ZXp~o5{mr=E(iy#TV>$zQU^rlFk6MODsN1(gIm>UPP9vn~cSJpj*UEv+-u)W8
z=@x6@aWHlol``YP>3X0y*v+nlpmkFVv_I6t61|K0PW0vK10Gk~XL06?{$JHsfr*`o
zFl|!iL1k}fWc_?p(92B1nm9yLKP(4e97R)tSrg@E+hoOH72+lVD<iyX4%Xehbg1~4
zqcPF)FB$_7b`#VG=8`nPW5D!Yt{ct=GX?20n=IWMf;I1E(=IJKGj3UY4@GI;v~F;l
z@h1DxPH>J9o8=p0e16~hr(@&OCEo%Oy+Cr$pFovITacej5R1(-IUgXE!(SBV<g0<f
z-$rMtIj7eGVcH03OdUD<LH#qdQ@4_&WYK%Fp0i_Zw<x;G0Eon164p1;9m6qU02l3y
zVA_rfeq5XJb;(*5Q_J)kvU2e|wgUs6LRFMZZWj^CfN|J&L+`d%Rw$!|#6(Kk`(LT~
zH^Dlprd}DV&ry}L1<ANeIf--R+%Sg+O4qSTVg@ImYo_0744Fyr=Utc=kLOWpTW$pP
z&{fWHcSooI8ky_KB&Z_BUrwXjo+1Q?&$X&;9?;r(*<JwNXCOs-fc3OnsFlVq>ZHv3
zuP^<AH<FuB=GaLb4}$a@0FhpHvzJX`t0qwE&CS8mL#OmB8g)?%j9nc@#T|``ZJ;V@
zGB7~aFujcGws`+}t@9J!IM=_=Cu(Fp%o}b-Qpa7n3T0BEVm0aR_j?K7HB!>}nKyc#
zve-;btLDBZu(-5qccR#nqYl*Mj2Ybf2s&Yn^;dw+!Q!N4?_n%{zyl$}@e%{;34}kU
zNPN2U2(o4*^+zDsI|$`4%O{y*1J=pK>d%|ZNTdA&$t48U=Rltj)$4Hj-L>5Xayv)2
zS|^Y#ST_P(TNjAEC}7=gI0a$i_@?biO{0|L>dMSL-+;g|gj@~-@NYis3b=>$uE(q2
z6exDG>sWV3vssMkMkq1qf)=JtrvgCpWTLbN_Kl)IWK}jdSD)}UVE&!^;r+x{NKgNJ
z^F+DAL5Gb1jqjx4Eak}7Ye6VernHPd>=8$06xD*}WA*)`V<VV`?bcldbQ5`L7o!NQ
zn{A&+&|Sr)ylS9F@PX^HxLGLY7p$iO;$#{UDPOn}Y_CZ9P<vECZ>0bR8QUGQkBUHU
zG6ld^)|H^;U=Q5Hufa{23W)4Q08?89kmyuv0+6CZ=d?!{z5`shf%a2TEg<G2GcPXI
zIIC5Cz@`!cUe^z~^8lMh1Q2K#&=>&d{iF`)%(j2NadU&S8Z?zL)$F+3r?HSh@~XRh
zI9q7B)P=oAPGY}87R7CocL8)!;?rUYgFTq>d9oI8$z@!t)<EpeYl4<a#9<oC5K}*j
zus3Ji{(-okAP*zV{qG=sXY-FW19-B<Dr~^uFC`RiE%bJzUYm08hYyxcD}Q`5Gk>th
zWr_M1Liquykc@|VsM4JsFF-%znsCAHaLVDAkmwNX>9qts^R8(%S?G=Y@N)AQF{Lk{
zJ;=66tmZhqEVklf1O?-hR2(s7m3YG!u!V*@nMA>Pmka&;cK|%87tpAE1k9Y#Zf$`6
z;6bdP#f^Gxj!yb!JlXt}`-KvQ`m0WV(I_d%4==bE@@&z%RNy1daBg>HEivLb28QHR
z^JW)|ndMKyk!ZD-CiBlvKRGPlpp*|PoqcMD^2Pav5CxdyTRb5#(zF-NIf>+Xp60k>
zbk|khXpd=^N?f!cYF5n!70IZSLekil$YF^(u)CSp=V8-J){-17PX=~*F$HHnW>2qG
zT7Kh7mpFX%y#1xafOe>z_CC8Dg#xx?2XXvJQqXF8ni!hSLi4~N4z#%AZkF5ARbx;q
z&O*1mH_UG}@T_H}@~&%ZfvO_}l{xo$xGEYl)@LLLDUvfiRbPfgQrO5|%e-z3^z7>t
zuKIYL2rtt@iO&$@8SjR5^BS9VlbV^mtfr9ta?wS$48zdKGeLK0L+?9YLhjLN=lAc?
zersWWc}-n^H~ayzEf!8=R5cvwpU)43LDq(T2+h9RcT!t*H_{bJy*U)hp7%yZUt40P
z5ryw*cUq6Ce%bIwS8Lk%cVd%`mADvP-K7zvWJ9&V!TYO>cU<8Y%x?82^i<(U#7#|m
zOQ!#f`v$7c4d<dcWFxtB`!p8eH>$Ad$fgO0Z5@YlgB9qAFx?E#)@z5ibLFebu#S7q
zIi!JDn8`Bzh}IPV+7idl2q}fSSe9lS3!bytprPDA{ddGa;^3JaOA`CA9`zb>I$kfs
zZ)`?_)<d&ICI_J;stU5Oe7lv+ho;7JmsHzp(6JN;ZdoT(9Rg>9H`?IZDFn5W3DEnr
zj?#OEgrKJz&D7O)zRdDX0T&3m&#UVP&&iPf{Vq5cP9T1+ft^<b&6C%+VORaOj$EY$
z4ib~i;SOHLItjM(*bY;>X*7l6>|bWDoZT@|Qz)DX@I5C?%Hs^6lHDp$R&j~3>^Dpt
zy6$PY&-U6LV|(($=>>nUJT}C;#yFmVz&0dYPsoHl51P&ET;o`5fAr&fZI0&Zc2fDi
zQ+9!?nPcPg*vd9to~a5tOFRgv)D*-}j`aEApzJW{RobD5t#5cuIt}XfX?FL()6E1t
z(lEC$nCNmhK~w)^0J~ux#@&C<(hbxUoEz>6Wd}l9rg+tzp|f%i)iq8g;)k7ls!Ub2
z^6*`_-wvK19o*)#H~x|N;_25MaO?PFcuuzhMACsHVh#u%n*bX$?DOm@8EI)v&<R+W
z@af8~HXU41Er<|`?_x)=39CerN(BD`kYJC1#OP>~7mgar$WL=%mE_-@z{gkNwMYxi
zV0H!?CVb<Ct;9y)d2T7TOVv9EG@eX69$9#*S$4AxgQmsE*34ZWAD`Tkl0k28@7RnS
zEh5fGf966F_nE6Tj*8jmMKXcrSQeZ0wn+7;NSsR5D%{Xcv!W4H_|*0eSR2G~P8j^+
zN!7}q?|i>G_0bh^V;*izP1JreTJ!ZQ1raOvt3IF)P&WN~(xQ_=0jNS6^(VU%0p+W8
z-lMb_Q;SxEigAR(3&KAObb&E^>=`Z8);9Fsxj}zH<NT;s@XcfXUC`H?j<EAEjZIq8
zrD0+Uc;Na#t*D_v;C%bj^?D0A%CP~m)FIbOQkFJj#ZP4J#ZcMF(dM7#e+~qa^9NE=
z$#F^aH@338IyQ2l&=cR7=dF3RSF#aZuT;yXlKpwDUcSkmhn<L&4CShf;4%5fW=rR{
zCpevDngAt#!Z2a;p?R~3=<Q`1NRI_OU2(%amx{!c@{oGz@w7v)xK)3eIN`E%$Y^jo
zUAoPCM?`Z+8JJMzW>&<Kgz<1r)FmicQQk<;WLe-f24r6Dm)ArfgH5y91&uD8H4OIr
zU7^M@mMmv!Mr%`Qd{tYZ9avMvq%`bX^^avlyD1;vOz7xd{cVyGu&KpAn6JKYS}rL9
z*M&>@bPiB=Tv$syFWu|71|`+%b%b!rESHDmTQae?!#PTfT(UDaXL!DFvMuq;H3GYX
z={yejJ=g&L0T4oe-`Rg{Qryjc1k%V?X4cb!hm^pneUcp<oskgfRZ{lf7aEc?D_P=`
zfu+OOVw&k=#vSj+zg{k}Uz`w$w|{I}^x^!bR34*V9)(UWNHX9*;4NQv;nHeysG^~y
ze?*Snt)^wfciOLTm)OKe&u(8pA%td9M%or(<GWu&6Dv#1=EW7|mTcjuZ(0w)ubI8Q
zJw)D}MiFc;`Nkh>ll$lAo0CP?JzMya0%FIC2_koaxz0r0*&3bAC{T#g1%E!-tM5eL
zBV_*|u-ChlYUwg3wf+O`j`X3UdNlT|#`Rf5cETEc|1*apV<K-8QfC4Omdv0*!o+E`
zLxQ$JheA;H;ZDtp)cHm>qz?$3g0p6CznVV)&xv1`)~aM;RKyhP#EazPDt+!!>T>3E
z8*_5$+_m75y9trQ_i8M<s^4WR#H34*uLmXQ(UC-ARhr-oBY-OIjRIGP5rkgRX2w^4
zi#H+E!R^`U2?*Vc1zki}>1N4*#8bQ1>5>Us$@`#4^cYcOSFxtL+)WUWVneS7&6W6*
zss>8a>9>h(zaNe1ku3XdfRei-_-hZd-0Dw)(Zn-PS3ddp=eXqAEEzp}!iHrK<3_Z>
zNcdg6`t9Z&oURNH{`otH?!kIeyd-+@c4_TCp8Wq_mq^aWWTZigGJz2cLw+i~KaxK>
zY;EKXCy7KcC|3P=c{IVWRk&K9-JQZTkN1?d6<s@wB@ZR!NyxLkP0z>vJNQ%S)3t1@
zo#hC-E|^V!)wTN8w$GLEU`7zwj4tbm&EAQ*L0QRCU@I}hK4JW{dR^H1#|tDR`V$c$
ze!0AA=#d1|EW*|RbS`w}n+G6d`M<3-=q%=VOSIuU|LiUbfkJ;yfr8pd@hsgrNct-2
zU#xWpQc0Un)sK#drs~}MXO~ECp;GTPZ%0Ww&q=x6CF`p&E$mj4(R}?TuV*~-2}Q($
zc<uGVlyS>NYS~2iw}7PAt&|w@wq#N31D}8_%YyiBkZihbNOn4wG|XHzfySftOLV+B
zRK}iz<k|bTZaZhiMk~A;+lK_ua8z#KO&-^{3ffL{K5*L!e-+PhnWy-ATwLc57vS+E
z_%L2(GGS$N7*2tTd!93m(xeF?Q22F{Nb4b?%8p%;%;wu?K=YeR^8o{fReX{0Cm@f=
z+$*njJML>lT2`b?Nj8<b7ar6^l)@bJzP8kE?#wo-1=QNFW*Z{<hC2g+Ysjl=X8sQU
z413VGitit>SZ1dbG<4K5n=AQ7=Vmgl`j-Ksn22;Mc}wp*U7{n8ue{N-UOc)JEFcOb
zkb0iv)5xZ|MU;~Er}fFsfMRf}rw((!_Ekw9vzG^vbtFQ{<jUaFS!U)h)0PND8C_tI
zxRel&vyK$TamJZF&-#dAWtV3)B}7c*iF|_o02{IWO480dajzsJS|yA!O$c<3f3hI?
z{r!J8hZ7P?ez_1Li4ouz!1#j4IUOIb8wXF4{**~WG*Jq7z>3YEZ`vx`A~%aTJ0j24
z5@%gmR5*#=rBHot*oH}$_6@QhsL753v%gt&cxwR3_F88AWyR8JpKfn%NuPIM*dulQ
zO4#&hmfHujRt9Om&C!~iQNM5JRxgH{ZTxV((ar8VH+-{4d%NY3yH}}>3EwIcr!|{t
zW8j#4@4BudJr@P21YALQHK9B%2tCW!5|2H6x%3b^MK+#J9*b&&+bKl}h98q-IXXy^
zyHZ3yK1Ot9AP{O(&f!n?D74uAEv#9M%b-)B1T=>Ct5FT_7M6*O8XEGzaxS*2w<q@%
z1^X^~Xc>$YZmL};TwkPDd$?t2yyX1K>fo~W*g2zz+Kwgcdm1RlAA$bimMVscIjpw6
zSH`j9MGt>HoQvYW11`yWlb^bO=$_g%mzQ7^G}94uALXTwZJR3uqiEbUxKd}J^HG0q
z`Xtr%8!(Vm%d*)V=~R$9SV?W3$&=JQ97Ycd$K9il)G1koz1l}xf1pb~c;c9v_~wg%
ziR>_39F+k(ivo8%ErIuFv{B7)p)E*3MrN+efIjGp6-3-?{$MY~QtrwXgNwRGWg>K3
zBgv=(H_?hH8~;_5I&oO4)C9NH+CrI{^_cD!Om5<UWtl3&lDqc${$6UK$XsSdGAL?Q
z&6jFur+kRdqv>Cn=;$3DzZi%Z-z&^EZoXESa4C)$D>7fl4!QC&xEL!aX<G|3wiSL$
z@P_TH4+!_8a1X|tRV!8r%(!rvw6pD!((S5CijoJa<W;6{>l2qfimbA3z8t`7yC{%F
z6SC@AsU(w3X7ODcdpximo95hS4;qZ7*t?mZ0U^!f4Nl-tDaP0FzUuivu}J&d3!{}v
zSWpIMf0*S0;6CJs+OeM5L-V7rrFyY>t0{C`dnxRw95EBuGBYq7&vfO8LX+gJ!Np#^
z5CIK|=H)*GP9_cq3`!c9H+VePof{c{#nx@~$a^2fLHJ!OW>BZ@9%gvO1hZe6`S#5r
zs?V+4o~#OB*l{~D$fz*Gz|8boz~UYTM%57y1hhZs>^(6gW~}{rmzZbtFn(Q;?`tA=
z+qoo@K}SZqx_zyjfhe`^#r{*IHC+kbS<$M*2#&*#3Fa*i85drQkPcerMNrlXt-{2j
z&hjdk2+l!VD9&U}u+In7ou~-piP2uw8jGNI`Xm~yRzY*t)V6=5`s_bzLr;3II5opz
z#|*OojQh#=1dg?~B4Qnnv-<MP=UbwB!bEB?9<nsmdU|al6hv)eNeC!4!MUuic6dFA
zkdBy5!4{QCKVRp$$7@dfW;@Qly-zn>wxUOSg<<RmHeGJT>$=6^?#bs{d*ip%GYs1B
z7&O94wV_iHwB6U8Gx@bRZboT#MDXkpMbns;@QEDzRt4G)_&(tlcu;N~iwI<Dd~?EH
zI5=k{!|fd0{Yo-(z^y1BQvPK8z+pp0YUOkynQv${tzg(T?eu};z=`l>(^EcegBq?@
z?tsm8R8p*w!ngk<fh@6~)deWpLT4pr#Q;$%8@)4_dS5(Qq;_}hF#N3yW~~`ii3N@M
z(Eprw7^=@AzL!(@Y%jc-!1Zg&ohzXs``O)EN2vBVm1c_NB6eSqecRz)D#>BrI;=pi
zH$KlsU01>JLk(nj=K9LUM#2MwDArRpBcq{=f_>2>)N|-{Oh?W>#R|n7!c^nv+{zpm
z?IN8eqwdaPG?!IYr#^K+n#_2%FVwTD4BBVV6<kEoK_LXw85+o+Ux*rVg#D0wHVcX1
z($G?-%-ic(BZki3F=vRJw^Q4vKzl4smWE4`R~qcd@`g04<<RJQ%Sye%7)H*#M;EAI
z#Cd3_bfvy&nA3GO#RZ-Rlh5=odqjTHxJ__aiQY;uyE@8pT2sm}|Kx3q8XeC7f^VDm
zvXi<mc=WHD^$|Vo#+?l1zSCXyP_&$LpU5VG_eTGuk#KU<Hiryi8^3K?)KnfoGHa(>
zim^|CkyD9BFEo6;*gPAoz)KB!^B|&u`p|+>RyY%(K(J+xs02=Iv!h)39QB=6Q()n)
zt;H0fiWV3cr_}h}3Hh<iceRTtTanR1D56T+B|j$@BGws0wVl)y(Lt5(3@NAy-D4CL
zUTc=tm2wO-ewmbf)<62#-L09$j_a~85G0Oj6+K+T`2ihtlG#m~K!-T)m4K}3s?*t^
zIl=3c_H10`EGc!nySPiVvZI-6%MtuZg+}Xk&slqu9x`AH#_KWXJ99J4U(@Bwzr9t-
zdE^tZ3w4FlBj14y@q9TW3WNJ@t@PX0tzmgq9c`A+sAr;%*CnOt-{xIU&IQ0ze~h9e
z%MFNDFyZHkUs)=A4z+DP6`#`2YQPq+_8ARD=kbxjx1~k-$V?s&dvj?N9++kzjSwO%
zW^#9+)$|b?3BE0fEHZC%08Qyivo}YN=3^=-yA`qMjBZa0tDtxebgt)|K3ss`>pUb8
z$}Aka_Q(q$H8c~MP`MQXsLPqOXATcK!8~=qBnZ^Dc-*VP#gdAuUV+Basm@gUBY8^4
zweh;sS>?S)u~x%Mra7r=4E@XKuWC~aRa1F9tMbKg%zW}SnB3<@GYn(4vdt3=Z;ydw
zZK^Xj{89&6xyY$dV^b$Dwg)YP#NwUiTI3c89@W%^;w@KqrA6})yTW;pcZJ>-?9)wG
z5>}~wn~%}5XVhHfCvA23o=^Bb5Jh*GMDLP$286{w5=-u5|9alZczwQF$ZS0aAgFn=
z<fqpcCVR#rku-y=K&<*hgGf9*8l@M#7x^bq6fe3cw?qNX!?~~V4rt=;rnJ6t$Nh@j
zUeD}K7RUyzK8Q2WvA#xo6ngSv;$S|egJL^cfHPe@W+}MpM1*SLQjU|&RG3&uP372r
z-j5^pWUo=dia|{fjeue=nHPOLf1O%RK}7q5S}oGqVc4?NZ01}tgk!I)gMu9P%*wdH
zX*~UjVrtt6Thq(M0Cre=UnR8cIUiBF&ci)_lwLHu@ttyM+_E#-UfK1_PNOIzE0p_6
z{wz`Nq8OCw#+R77F@h>yOhM@wQ4j7=WaJ(?<jjmDuG&cVqny>f^mv03p#2beRT$n+
zX^(L`{n5cy`~oS~Kp`8=(|<y%#4l*ny0tA&MGt603|JUagw@?+-UZA*q1@7?8yTI1
z(P?dUrub!2Vh!0Po$_BwM2l&Os};q3CN~`9%-o%=6Crtb`gQnKJ=H%ShlI?JEXwtD
zM_HF!yln9zzV`W8v+&ua5$@6$o|TWu@EmSf%R9>2_Yakkw$;YmMb=*^6MxFp#%hBg
z6Obo*1}YMLMr{()*IfG-?#XzEOfZmA=&xvqP0bi6Okc?96Opb^&r-o^@2oIdqMn9p
z;muEqgQ`F3+>+Ny@XZ_E<%XAr@dxa>=L|Lbs2*Y;Ph~E><Pno5`js{*I6;m^zC(zU
zR7w^C<wLGWGi<Q>3onIZQFrtfzgvWv(gxCJvJAN)L7+lN&XJ^jlPG^c4mm1kml{<g
z=}&4$6Kgf>BQ8jgq6j2pW`~LR=s!r(5@at<IA>A39=;dKY9{K9=^RS-87>x;mPq=0
zeDK6I$R3aR=p;JF3vz`xB0Z2VhoO;T(LL2f|Gl7C!gR=$(+B7b1cRd4EV7EIa8fq#
z7TIDoYTQ5X$$;MQYErmosN%$!ZzGS~T95S^4!IFeiv3K{8G%>+lZU>B?^kdfLr3(D
z$q~@qTW!>xgV6_uu{Zz4){r1Fk)K;AmS#o7X7Q5sv4qu-QJV%$2u1$_eeBVng_6@6
zz9T+T&K*h8FRO134Cyy|MhUrZI^Koj)qKkiPc(!@Rt${x6oTd1Zgx}j-z^3BsiE(8
za2j$I_?`UJ641edMZEwkth6a&^6yB&=wZly>v)`!TB$_F2*ec&yyL+wH~GI`<po&~
zp=anQ$J$Mc8m$>bLmqmG<g92yU5WL3zF5L0$T34SMMY2{EyHqXsw><AEL=!uLiO)G
zLUMM4>MJVi92(+M9K=roL{69lO!>foPV~;7HOX>^Uyi#Ee?KQ1FHwGPfemCn!FBwS
zjVkQ#(h|N*%t`r95RNyli~=)#@WkH*9*#x2Y#z?aia0_4POBf%;f-q;tql;wovT*F
zi;=VZ3o3sXrYI7wWM0R{NY|r<EJuWlg$jj}q8Jq1{WF%LKzMEg&j9NKUy-hcmh=uR
zPx8-jU*rbb-(P{%_7WK^zAd!4a5nQPBpF03Uq2W8olH^K_wD$mVB_59hhLB&Jjkd*
zbtr#MBNDPE_|KKYHBPxZT<BmEtg+#w@7_4bs{CGdsL)#=Npu9HtSDDZgUS)sQyU1+
zV<2T#Nd!R0%6kZT6b2PMySK5W7n6^S6x6jiQ`r3kJ-e~mRKe1AnWz~4K22~a+Fg?8
zlj^p!?zCZ~a`!BuYw-wGr&q*wO~Chx$pIiyCI}#BF9-CFyLQS3Mi}K143;%CAk$G(
zmw<}DlKNHE!yf>JtTidktvCq?l*`{}hQY$OiN5wG&uK}%y~=xHX=$0;#I4bae_!9w
z@8k=mU+))QEMW-PLUL+L%-<9H5&j+M*lRJUl$8J>?+H*}RpUj_Jh`_jzT&_EOR+px
zv9Q5}gwLsFC|_d+gKLGfoQa94*W}xJUNujWygm@zF92;@^(N;)MF|n%lz=pzYg-)%
z7))^jOOuk9egsMY8!`b>ER_AWZDby!00A(r-YD|1+8iGOo$L-aek<1>HbCxR28z^W
zOm!y>2pvB<CMJ$BLB2g&Fs>1xpBuX(+qjOP`2m@-2?Wa+ASES1Hg}6VZ)?2GB^WMi
zeR<VMaRgZUx{erpZ3H7dg_qVd{dipodZp{3<69FZCYaL>9#+$Nr9@WmZDRP$;<E?b
zcJoKs#R)6s^S`ZGlrwy6<iktxi=bCC@^Mlz>ZEH+C+PODLZ$Eqcbs%eCFbwbX7DO8
zIk|xgByB>mkV%^K*g3ZmkgS@53U)3pFby~Yl`a*~5M>NXf)Wo*%f{Z7b+b>50-+f-
zM}UtSaG!4!Z~Aj@6@XW`cT<TFaWpr;vYvyAb$L{S0K$$j<mbjne!?M;ZN}i)fm31e
z0g$U2Fd*UrKLUZfSR~(DvdHG3@tfUxL$SB`jWNOk6Q`Uu#);xNKs#)y9s&D@7R87H
zP}Q4QXb+hR5ZX<>KH-bGs@E}^PENf?{Gn`!k)hBuuB^PAs#f#Ov~70x+3Ib{2W4d+
zdzqhEw%AxV!|;`_i*Jutn@N<&`twaXT^_#SK8;i_Gx6|e-wA%b>9095R$D%!#TnnH
zsz>Iqu(-&+=Z<T=9@4paCf&I)pU*KlQMtuM4`8zxQPPu`%FL0ukH3!|X`nQMnq$xz
z;o!1(wjSq+umBGW59cDIGE<$YbKC@wJ%`;xK<ATPRw4w3u~YWzxr~D{v!VIluBuc(
zS+@ko3=|#85K1bsoR-EbeM#&(ZCF6jCJx61(Mq_VR|`68j=*8z3g1)iFiLlb`|v}g
ztd-cIBCuaE5rzeWOq3XC=hy6c<2V4(GixpB+4wzSK*y+608#-Cpi4wsjl5<P;pYHo
zD1EuFL)Aa_Z+D{O)0b&1^%SBr0ZL4)Ro>L;Pwlet>g9ug=T@!pY%5yadVO}oRtctM
zZ8+Hhya!C>msiqXO@6+(){EhoZfFT`Xa~RDNo)5||4(LN-75F@gi0qC)Ot6(uH{Fy
zdSehUqZh1zF_5S5c=1mK3r@O-CLGy*Ux3zZ?l#HCw+P5-V#&k%3UKKqHcn<!(DCTs
z3rK5;Q);H|$iQ-ysy<|b^VgYcloQ{SEH(jb34exLT|Cr_)r3tys{`e}4)@SC<vr96
z3J~u*nhO5pGa0c1l_$%2%=N*{`KFYPG$U%(vYebQ(3GnKAX&kmT^&Avy5T#z5^#^6
zhLLs1xJQ!DUcK>A%;{A57I4lA4YMQ5$yWG6KSJk|Q&UrS9?Pv49%jENZENVYEOIaR
z^0DtAxAEC=b}d(BW5czboyEran31su`^nP=DaFrlWYitP2&A@2<bOjB=L3p9&`H4Y
zZX3X+@4Ylo?mqb0GzAcSV^C5r1|ja)OWy~x0%C08QgcQCRN#oCl8grgTKS_sU<x$c
z!?Q0|4SYRfM+Z&>{pdn%LtaUj0SS(ZzCj<xsy6qK4%hT`(aBSnfotCvaxO=oZ(_XH
zmCeX%5I;ZO-mXdHa>~vvv(Sc?=IojIvSq)~isQ1`Z6TO8m@pe^xY~BH%(2T>7}lI%
zm|DCFY?crbg&eKgWZ~rUR&Q#?aT{;8&=FfrqQ(Eba-2;2ph@lj_5C5iW-}<G3PHcS
zMhcHlqW_*}2{~!+46vY8RZMG&Zx9JPSOhu{=CX=40FGk3yC13TZ?2g_5MT@BC{%#l
zgqmW8+g!ZEmcbn&`3zvWNW2D#A*J^6ZIEPRlorti;rmH5qC=i2`}}afg}mc_QI-%V
zi@UF(ss@kjO2*bcFet2YW81T(jkOxgkd`2=1L^?7wGK8BJCwt0az`E{PNQCGux#l{
zU1;nb$*|hpGoz#LjR9kXIO^q_Wi$2uJY@mh8buvnFsi$)w+5SO6%<BSm+zKn@Xl6u
z><iVM^}J$V+VwC=TCV3gXwx=cp5n^KshK`3n_`(s`=nymrP<kX!t&_-l`9+>ZNauy
z>eI<Fj_rteMVJ8N0oF9gXxL|IWB)$CgI~d?U$}pwRh2`Qkd@IFn@^2Uuc~*q2SBeF
z+ItsDOoX=MMw8)tha6p4Kf*OPHez6gNt%`^ou$OolR*)?XeRv4GqX(p2oAZCq4iYs
z-ov?<BlcEdm&63wcQfc3=lV8xu2h!P^fqGXOj<UE6zIN<-r3*Z=*x2-PW>8jO39zL
zktf>c%1;`DW~;fziP~g*g(7A1`!)-H2`3kwt_0kWXF;|LL9Zxt>0>PBm3-)OwB&WS
zHHPvZS_l+A%WqOP3pyd?fM^Y=_&#L`_<W!__vxG^;4vErrT>kV<SBpFdQ=)bWwm<A
zr4my0f_=iWCU;aPC{nxB+tS#l$}KEy|Jb=Go$rRz&4F+CGH?9;k66OA&I~+3iprte
z?*5de4<<o%G6Mw!ABLaXV1FabGMv#IiJ&2dT_LS9*TgF<VN389Rx`La=AUp&#vSe|
z<JHsz3l|Bqm&HKUfr7AsvXs*5Z{#OJ*_UXYBto4Bn`UiisNW<jFbDB2VXI&9pKn5f
z%wbW}^^|w7-*+Hq?zv$MA}^V6@F>-d`Qq}ol5zl(TJvg3v+@}zM-F|2tRxCz;?#SR
zrGIl1aU3AsRGhX5V{;_T1EZ9C42Hwjupsk4BDf&Poqda3rCMm(ch9B<g8$bSb&*O1
zCKz=S4i?46#&%_3m41e!;DA8y{k#m@?`<Rvq=nqEu*^xNg6`KyA_C+8n+vdbe@@Z#
zIrD7tHI$<bK}`GMNQxZlpDZbb-Fa!TG^Y)n8}U;geEi+)-C8Lc?C%ag6o(69$Jlxq
zVR=2A>XKs=2jTfZ$t*Sd-<`BUhI%#ILI-Le(tP@Yh(;A%Wr+R`QUX%8Bcv5K=kkG;
zMb{nv5;2Vr`!bw=-*E;3QMvL|umF0hlQaz(ge4yC)G`uyN8on_3zXk^oX<+p>F*1W
zr+pNZ{_acl8Qyh5ndb6mRY`eKS$BB6KtV5171=G4GJ?;><Igeu^{zk`FngMT`}ZM#
zX8$WE`o9^+KW9t$AvkNH26yft|2b>FjtgleB3WAOo%*xs|9nr%U9i&~tha9cJ$V0o
zFCwO1*jwKEGuD5mB?ANin02FFTz+R}|K=FNjNo%DPA&iZT>m|!G=ku;v552Y|Bpk9
z_$vh>tF64K<o>fG|IlClZzlWC_XfWPpQCy;338WTZ~S+5@D40EX_}a<6xN@O`DZ}w
zFpv>5e%iwL&obzOyhS#Rh8p!h-@5=3qW^#S{~tiY@8u`ut1Qnyn-WY41gT6H(0w*k
zZuy^kT0jF4ef=6Cd>3PZ*5eLu!D`zTQc%v!sdGFZ2Nc5GoE&-@zXH&t9rMgQc@y->
zid23ooBjagbJ=aSRM-2GZW7%@JA*DPEQ}%&CcxN~9N$)P$%>XKZ}03Z1|p{_pleaQ
zmyq5Z0-B0-K!lGuiQ_Q|tlfT5eLz-3xsx~MNXI(?LKI_@@sdJ>kvoSkeDiSQ5@>OY
zZxcR%6h(bL_H7g4c8<JrPodmYO{K=xoc2D<sN;!d1B8&xFq9XtMwDci<uX<P=gt`L
z4u`=6<ftWvLz6%*$_`{Zb281c<Vw`4t&c#Sl_F|hl>7)(Bw}kpt%3c+N?%R}_-BM-
zb=ia|cyv#ix60ATbc6~z;dtraMIIA{p9L~|${Sze(kaO8-o?nrK*q(zEdxVctGGXs
zfqcWR;H2lx4MjGh=y%RULvrg28eevbn#ynV9yL83Av#oB9wUz7oqNKLU-k|k{N}7n
zIZECC_nm<ifkc*O-rqExq|V`oxdxoPHg1k8KZdYUzCGF)8wNu21eMS6{lwwgSSls@
z^$(gxZ4HgrK{)pg#HQJW>z1>lkZB+QE4xAz{tTJGHG_=s&UQ2O6<o7ctLH&e6Z5hJ
zmM}i@9p)FLX|Ii>CI0i%V-p5b#2O%gbfy0HvcUlZ<|@V0{m+1<T*2kU{F63&9H@kA
zZJas|oTXKqARqvsBJJnE$W4w0DQ3F@EE49YVR_1>9o=IG2x3V5BSl0?a|zg%&J6UX
z)Xu=9hiz#xIQkvGrEYg5^W&vdg*?^Oqt?mxmP4qN_2?n5_J^XSF9SAw<A_d;7%nTr
zgLlImpT7e_s`$)AqJe6q;`;ITYp>@oTLB-a^$0M0Zb|Ce-9+;=+fgEM90!Su9>_sW
z4&IgL=>?A8%jUY_nBbbmzVmIqzISE&9CQ(gn;;&&W8l7epnu-USlzX^zn@!CF?uwr
z9q^slay8Lr%IXaVheQtll>Lrr`IL1TNTDVW5jSY~_dd12H<wm;Bq=Vo>CUYJUfJk>
zMdFueF|4Ggq^X&7xH{0<JyHU4-AG7{8IW(O<vvackthPl$T;A@kDQawB}8l{oYk=d
zwoD-%9bK`8(<uzuYpG=%D1&Xf;Yd7TelzF)^F9(L-$em9cw>Nr^2WvwG()k?eN@oG
zk=lF6%XbiqJ!5+W+A|`TR1n5ik1;qwm#_&)olCU?c!v!~^19u6us_V@s{({&%RyOL
z*$~vL1)H3sAE@L2Ch3~fFG>aI*KcUe5G(-oIqN3uSEE2(*!cFN_i$Sh^j)--hsEIE
zxwhsAvjufPJ=gRTsDI=FkrQ^%eYbN9vB>3;y$?z!)NH13r&|R+B7>ReHWl;QrnqSU
zu5H7NmE-c^1?F?Tyl+?b<=;T{&7W`I1qH3>t}{J3P#Y-ZL$y_^6;SVWS^=>*2Eant
zTzp1CNcMPKO=FG^Jx9}VCEI}N_JipyaeCsawXE!hb^0c%ouwGFs<>&A>_954#~2qs
zT`|drhZv_jg=}Mv=kV+SC3>>z7#S)vaRCaKvzPQuls~&^9`31A2q4<xw3qCFRqshF
z@CTpxk_VVTJIr%6Knt*ZOV{bHT^#MKBlwD+cGqB7jw_(U5_9z-b}#56gRTJw%Qgpd
zbdo*5igT~&3Envawo!y!3xR_F3=15C%D6ag8w@+9@7TBPk(uSbIbC|1)Q16zV|Ce1
zIl{30pb1CvB>Q6v*-tBK_Zs>()BJdRbvNOrAkgG)(@RW*-wDXL_v@8L6oD9ij^G#`
zX~Eli<YU#58GdG!gwxipSk3{Wr58RGmP<Wx>%)iToE#iBjuCH3Kz!m1snPS8<-hOF
zC#KG~t@&W!W-9Tg!^u__I|Yy-Y62?>C);D-?zjnZr91h>SP7qNO8K|tcG}{s$7w1I
zUlCc!J#!m5|N2fe$f=9DwL)7|R8)+&|A6%ssZV8aM1&r=u)QAsAY69?diBvgL3VR3
zgN$`ljzAUD;{3jgM{dGrD#ln2is@rDJqPj@00z~GYFdHThfHultY!%jTt0x*i~@Xd
z7wu$Pd3)Qo%k^L~Ko65==X3kO;zWUF<3_R%<KCN7UqQGunRk|#UMrDJ?EcbPg?`A0
zf*{HVU$f|=MX>!xh_@ux)J)c3P|7|yPZ$H-uWf&MR<8|^6It|<NXAxM*>xRGE{|=!
z-ywk%dBZQhc`LbV4$x-zJ#5~gV{PZCUC1@-iv9E`LH_X*fS6PR)Nx9L05fGL@+bP2
zrnHfDrBWY@4`RsCI1*&lps#`jEe2uFNR(bk1C8JZ4?|b+XjU3Po%AC`sK-ov-0$w*
zCsyE^*e9*(v-;%0WO)Cbp9#+mFRSk2ED0>$*9S5=oNsimUY8x6rft`?IBO9~e)oC%
zgvq|1r#H`TT7_dIOQ-{+zmqT6UL`OoUsM|dJI*6;0?)HQD%`x`u5MKE*$mWN{m~B(
zM_?GZ9X-cWR1H*+A`c|wLW1$}W0VtOJT-(w$IyFSMr?{ckjir4Noe!v68Nlqqd0H%
zst%rd8zTXq(X;DedoDdE<VC47m%Z!3KTx3t6@hrkIV!1#+{EPRd+~V1*emSqax3UB
z5>wwx%=82Hcf^xT7Px4_JMG2PK^c-SyxRoxmhE%QEuCV77v2)Sf*T)hDCouGLk+x9
zqz^n=hC2@E`=7$}34DmNd0koyYGxrVfP7Oo57ZAePAGQ6NDstktV(5bbd;``KZ1xn
zN(4ao_coQKq<#`7*Yl~X8-GfTSTU@SA6C}_8NZE)@qOrgjDP_J)Kyb^ZB?#_z<2=7
zcAmGJ@YQ-#Qd1?#R%*N>j2kt&p<cN}AN$v6Jxd>~b0aidW$~11;pb?d@3T|pzwTR@
zz4MY0a$H!W2?<4l>0e(g^QonK#xsb<;+jC$Sghi~e7<jg<l66g+K6AV6m7Htj+f24
zeYlWb>QM_%^#aW-j^Zc<GJ$szMrI<748;mVXkf%MxfBclZ|fi)`D`WS_J_E|+xMz4
zHM(ENYJtGR%r?&5kBT%Psp%)slO;6Ac6VuanDcNpNN>;$-Myc!j@%)z;(!{~y$hb3
z0Ts?^HvK@v7fs(+9+X<em2QT4NzItpnK??^f~Q_g#3R_pn?zsnH{YIU+ftzuio@eo
zt<Fy41*|-4PUd`O*G0aQSdVJXTH$4Z4Oh{96@J`T{+4nP@N1d}G5EJD(yDiYSx!Cn
zE4?7`pyNaIp(9%?e8N$7IXkVf4+|M&eQk`$ADmIf&p~sN`MyskzHDuTB=2FU!;>}4
zDEl7=T8U5aqllB&lm>iU8VFi_A8LyZ_vfi`m#ft#i5di4aykPXATolj8x6~kSGxBC
zXu@tOCIG#n9@k`i?NSdhCbRMUh`UVlIfQad0X{9?4SMRnT`V$va+38g093;k*eUiM
z$8s2Q7(_Tthk9>(t@;wQ9*lu9xhbHJe4Bc1GV+WQrppCryJ7>4ZH+6}rWLLyOq4fX
zJVdY~)9koweSQ5ix%enB*tVWLJCd5w^V$NDVwI7a-C`JHy<eG0>5u|~GE)t>xZ2Ym
zenF5>9YO5|+?QGAX7=G51&tV{^tr)atJO@Hx#L1J2D};<SgpXYrQNSl-l5GW6{6EW
zo!eGjxP%i}Fgzm>70+S+4m5fsZ8_v*s)7>5V%nD<gXotv2Dm^JN9S6O3dJfXn3=}T
zd%Pt;3zeiv0EWEQhGXM6qH8~`A45`?Z4y1K1$t%jSi@BtRFkE=A{W%;SSZD<7N~Oj
zD-kIu5yYa#gMXvMbWHl%Q~5b?zGr+LedU>Xtf()+dnkAOdC(T8mM5E|dnSEa+^+PX
zh<@O8Sp;-yGvc$HC-KoT1laQ6NA=3*gvCi~GC)1mvkp9;$hf;*F7sWZX()bb3O$r&
z>INl?4@@Mpar8|vQs(6Cf?l<mb3n9cCdzkX-E+yt*U&BrO)UrAKeI+OG$B0^iEpU9
z)Q<}Is1p6L?2wx$ml^Ybj7I*4BB*f2fJp6GUFN<hQ}><ST7;>A*{AcEZ$7<;Sb>2E
zwsT|Gq3^?I?-Pma7J{~JDYr)sR;C3006s{Q65N|5p4lQoYn-Wi!CW`ls2Izfkk8I(
z%#*|hu58qZD9$D_PLFk<>(LXkE-JHW_4Kvl`PNd)iODYxILH**^8Fi!v9Zi}wOOtE
zmh-l&gZ)p_x@G`#SgWbHc_v-^aBEI{eNm{I>ozl}R92(f_WD0U1nZptCs+^pd_bm+
zrJbz#_&+J>#$x~%)Ul_x{I84`38Dp3)t07)m;VX7$6kR9+0jJ8?>|9z69p*bdAa-j
z=TnLTKmvX~9!mg9mP#+-mi;uBwPFL9T>#Gx?ICvklTq}8(7azr5dljCYC?eEAXa>J
zKB&!z2VVj}P)Cs=6XW-|hxoxIOLV^y1-#&Vp9<u9(eKe*J7@jq|AGbJTZ8fufSNNa
zRU+08p!h!)N+WCh{qqe(x-@1|J{7)HntbwmuoSLvCl=EV-d}S{JVVu2oki$TfS-F=
zdkB?^on0kxOYZ={`&+8m_X3Ago^-!qufSc%4(Yp(II{d;{RvUqe4YW90i&s^Qbd&v
z;SU1NVb_-|*;l`pF)Il1q|>yZLA=mTPU;MdhZ=V_7kO2I{FrUrt>5n>Eg%C+LMY&+
zjYxk}$eaPz#Y)fi0$^{y@?p{_1k#eyTl=4VNWlVE_Z>Er7YL-Kpb!$8@sh8uSs8O{
zKjQRnC6^zWbNY2t6^Hjv;_zP|K>8JR#awdK;3XT)@3OEQ-HmNtvfG$irR0lRh?SCw
zK$?4EHSs0*-^t6kgT0kkcCq-o?!QJa{J-cvf3_G)7#H~((|~(E=uJ#E7ZUg*@<Ku=
JPvEV~{|A#%FK++<

literal 0
HcmV?d00001

diff --git a/_images/nll.png b/_images/nll.png
new file mode 100644
index 0000000000000000000000000000000000000000..f5f7d0b7d234666c443eb19fce5b4ac5a9bbc2b3
GIT binary patch
literal 55999
zcmcG!Ra9JEvjvJf!J%;i0h-|M?ruqN4;I`TcXx;28o?z<aA;hEySq2;+|K#$c{tzy
z-N!q|?lE?6=(ShbtXWknQcXn`6O9B71_lOGUQS8_1_nVF1_l-e1qu2j7!S7u1_ndK
zT2fL?UQ&`u&Dp`i+SVKfMlLc{3t3xpkT6#-DM8GtKoPYKm6|&s1zUHYyh}z7E(F(%
zDq?=PFvy08pt_)pI*J@y(9}}>*jHRiT89Xo=fi4oK;`4m?e635nExyPWgh3=*s?b=
z%r8x{L?F>eWGYo{syKAi{CH*glH^=ivVA!GeRzTJELPJ#uoR4T*kVWeZ6ge|U6r8R
zua|{aqiCe~ym3UB$UwfF^zqT)J8T#bfz~?~m;m6;=r_WMV!)?-+Avs&R=Ft7ohF5-
zsU1+>2;JK_L??6@h4*A}e0VT3WeMMioH+}^zEFZyByz=20!YYB*IHW#g(9Y?8UlH!
zBT61(t|(nOUQ07Y$_F<-zcIVl|18a*mcTl+qx2{`1?f;Uto%F>PEh_{XLEw@Alqd1
zbX)07x+y_~cj$roA8Z?mC-u(c1<4qV+IIz!&fe1rS>r+c@B-L^5Wc_-Y*Gm_rjghv
zwethwAY&e|V`#o~5S%+4*bAF=a}T+f(UvEgzMsLT;F{Q!P_xTuJ;L0tq@6<E88;K{
znODcBL4s0~c}zhrECa4ax`2uS&Gcgg`K8Nd4*`!^RGqfEB#)_lHkTiI#MWJR0!*{W
zr;N96;xy;ecXzNxiI%cyjb$0m1B1##)I9yF30Y(bMYGY}Q&~ULy*W6)?3ck94eiIm
zNE#>i9wVabV{{dXLWA4fHgk#s^2(URXF*&fi(84omG_~Ls2Y$4tbT&CuaM*i>}0BV
zJl3(@klul<uJmA(PPo8;H}9%Tg9oWdQ5L;Oy*6@M=>d)bfX{Fuj%sX@G)d;d5};V*
z-gVW8J1T#Wen5FT7TViaw~SOHa<Q=TMl~L4WT8+#tdL-R5Nl#A?i+7KI8`c|yKs*;
z<@x{zmh#CL|8cCEHDbEY)c7%<2WSlJ29f|$81KSxgCiT7U{9mHw{Xr6vC25dI)%e`
zAuGA(DHAFnr6&1(s^y$_8T2m<Lvj<a*TDqMa<A+}O$5`fJ@Af{-`}J<kuSsWtyfN9
z;oS!3u22nYfSE_pp6O3IM%Tl?6;#s*CaTx6Eacf<UdnwG{McwhU|s1y5-&7+wc%42
z=IwVg{_yruXg=)KHj=sfjp<&^)V%aDU>0V7UcMa1rZ-Lza(?JVt!E&)y7?KhEk@oY
z`h+R0V@-FN^CibCyM5ph?M`~qgUeW%P#F)MqYiCf=}vlmYC)oMOYV1FaveI%<XQC7
z^)=6@n-wuv-42(_SD4Pd?e+EbKWF;$@-R2vohSa=f|fl2d2h^NSjaXMz@g4>(hX70
znG29SzKXCR?RFxK^%FP2JFH^`i2)*zP^l1=LjG8Gxv?N=c44p)r37$Pla<2~7~{>O
zxd#ca!&t%ecUn8X(MQtiQg)=Q%(Xwqy$ZZFh7rfcm5{cI5HStXroM^b>=yG$q<RMz
zDXEdb^$rh*8uv322PK2#Cu)KL!eXR_crtMn8lpt^ZI;Q%D=LSCabVcbeBJLjUqa&~
zh=F*RvUn4yR&)Z0%rZpzihta!xPOyY$f_2l|H-mKe!xBHhbfIYFfPr2@9Wgo=1Pk)
zH#W5;{1laHHls~Gi;#EeRt|sEE3zS#9X92J+!Q1Uk;;x%?XuiZJV*1PH;ivwpM$8h
zV!g`b<6@xRAh5z^v85%*XD9&Qrr<T;y}=vnZktvHslLN?i<p(dY9i8%SO6b@X~CG_
zHn1kx-IS;%Pm1~=`B+}0*a|oNgV_7G74#=`IqwW;+>%A!GcaZX`56%zOWvcV?lPn@
zo@fXZxGEoHPQvKq*vj8f%O-nnTW_0h+ilZb5IFOC1{8|dl-{Y~0vA%6Q`}MnhIEGD
zhxE+N%{k3gceKp|FUWS>h8|N`fgbPuHU504(KPuHqm}rb>El9yv%*g0kly>$FVffF
zN;S~GEB@e~=A4$Dj{nK^QTAiQdsVlFFC|~v4!LKvzI^6Z<St#5`epjd`B(54t!b~B
z!NX67Om1-xOxFt6G>45dkK9%yMcjql7hGvp$CFLpA^OS1)+NNngH!tX%^xpgEQm0R
zGN-h;SnNdXNKcq5^rF9;SM<&f&Ay*nESf2u`Bw4c<Y!K?L9xLPU;P$I<f)oTlcViJ
zrYZh>7sb80%0w2;^r{x^MwQ#Ir^szMUN2rgZ5nNDUcA>LM^o3qS1wmt`@H+2T%DN5
z7^aw|B(_|9HWPvpHX*ktrxBAl$TG{EQkW+gTNvL+?YV8ak9j$G(AaT_D|iLWrftg2
zjvcb?S{Ls^oZ?YO!E;i^wTs)ugXq6`a^%NLs)`yX9+fbaC6#GQ^^}&BkVnc#aYuJE
zPq@9URM;&uEHkQj9C*&^*lHr`GfdQNoW_%E2dhVF^=&*?FM?BleXO04C_^gSoAsKt
zJu*GgJwoRBKq{KWz%#?cYa7{su|&3n*`RfzCdk{u)S}SBc=NrjrLEI1*!NMcTW&nD
zJ8_1_LGU{t8#x;J7jgk|H!>QsJ$@d(b`o>;_iocs>U*-jF3v7PZi~CF{`I^2)qgH!
zcw_yxZ1+aoDqY(*B-~KlGfyV>FBWH3Q_p_?xsD?pWuCMv(5qY0^1$+_@y_v%4-b|7
zDx2DG7e+`ERukgXmAp>8&cAe-u}9D9Q3Y28Z|LtS2-*XkUHFOgd13gEG_V_pScs72
zF^iD5yRLDqjkc5fmHLr&eCg<a*?kduJ$V_=Xc)2Gwcjm+`3B?Exe<WH#5$`q`$LS0
z?OiojHM1Niko`OLzIwHeqlKfLqfvz3z*T}~LCD1U1VKULf{ZOoOV|q0E)EZ8Ijuge
zkVq@_lT(O62wx!sZY)N$KEXZJ5QhR_1{<?*;ZIp1WuJDxQY1f#VoYpgMT{>AuYkSG
z@;ylRsrb$Sa)t+on?}2=S}&jSj)(pSZLI1~`G*9uj8=|jv*YhIoc)F>>~c88T%}1G
z$r_E{ZsO5N8ZGp1L5&dFEeGZ5Ts5wZO{w6>X1sV+jVy4cD_d+rT6P2D2|E2(&Cvr@
z%WqmGw@y0Czn0U5{8zsiVy%UDB)u~A(3_d}T`J{`e9RQq@zdGSX@g74mpB&?<nq?M
z3D4foCg%@s9&Gl$Ou3l6h(qBH<?l`#>`RWOYm7dYrY($PFw~q?I=~B$Zrr1}$Gad!
zuV0V%DFe$~L>JdhG*;K1Z%r5fK#t~$3X!guIxc0-y(#Bwx@$bRb3MJ<B`)RM9ox?y
z%x*R(<Y4subZL9JfNtS2j+$BL!eVYew!F5-KdQ%nw(_)nuZ$i;IHEIgtvJB?i}ok&
zX3`RR8<D+k2rDkr7ro<J(DSB4N$aGdxr+IYxlx*9y8eUPRn?Inx&GIhdb`ZW+SuX9
z;m~1+igUdhmy*7r%~XRjYOR^7(kd2%XZz7&ZlYm>Vcv?8^3?JZZK9e=r?%dQouiHq
zpFb{_V|-uMGuQsrbl8Km8fHWK%sXqxvG}3k;L@UNS5Qx)VOmS?W4Z&aJ@fK|tKMY8
zWe2-o*@As_o9*Kqc{e$^;Ff#CG-!JIm}B`@IblWW^vy9v$Kzpcr*N#>8#le1!(-+p
z<Qw^$w)wn<*V({x_)%nH<fAt_6h8%&1od^}r&Y_&YzDsUTPa!FisHH#HxITPoU8X2
z3C3kolnGV(`YfOhMIzE8sS@TLi_!|rcne&*YF!rZK3ewWHvhq0nsn!RIu3Q|+8fKZ
z%oX+Q+Uu@~HYAGIH`V{rbnMsMK2vdherDf6Y474t*0JeS0{OW&Q+mUGLSK1rAm}T6
z{?JT%Oq%W>vtrwk3OX7)fiw!9o}UVJWO?6Sw{?5N-|h)f2?2e-zf?aZZWJAg+P=;&
zZLB)Aa#q(}J3VbZ8Z1|)t6t>!92lObJk&pB4%U6$8yk<mM!41YWj~d^Rs1s?xSAx&
z{Zj5_et9@y^y2^PCidjWkJ(T7&SFsSxlwSnx!vY8{=olA<m|aKd@}xl(!<ZxN8pwJ
z{<1FchmoAotnaVq#Vh8grQojCE^!>$k;G>-<LW|UVwlziSQu_Z7!sj8^}fMz|JtB|
z@p2A?7reAQ9P?iC0D-$#iY4?{dwck52T~zFao=AM7_b_w+&RMc=wU~4fwIA|)Jchu
zy@1ggqp8cut`3F^&OI!r*JZdLKN;_upNZ;@eN4vHVKJZu1)sULyoItd%zNlP3Jh$x
zH4Fmu4i@?$fqr0M;PT=A_X&b-KK%dQ|NRFTnE5R)DHs?r7<no24<Oj%9OTvyz4sl+
z;Gpl?H3g=R3k2ddj5mbdkp^Z}><t+JMXoQVk=$tj^qW;~cf=Z^<0x%d0!Fm=gdvxC
zmQNj7&3n0Z9v)|>Z7svM&pxLs?qk`RmWPjHf}94ptSZ!S)Kvd=P1wU3ERhGnVZ-A5
z>xF{w*-BmfU;l(1_#P*rphSU+AO<BO;9stRi9(KleJK^(M`|ib#Z<BPfPa5$LV>io
z7=qNlT|cG7!qVOceqj4|d8j0#hyHCZ&<%cw48UOi2=~MIU)O~$OPKoqp@i?au*AAQ
z5j48~Rm=$JvhU&leLHHja2)j?P>X|b{&nd<dx`&1&<x%ADDPSmyIEh;J9#wman5m}
zO3U%wO2u@R(6c*Va0HJ1WU+kc*>AU_U9a2IlhQoDhi(rqntxg}RsyOVnL&nLO$3q<
zn7-sWBmb2gk5$jEd5QWK_mjgneCsfcy@YyArz<T)8XTS%pFJSGL@h*yptnA}GaodY
z&C?BL$^Gs&Dpm7DLXWOw{%He}P^hr6Nt#jLP`I!IJ@<0vM!DzoPum|)UDkYW4I=o%
zl2bqF4l*~b2Qh2Zu6i9c+f=p9`rhxP+Xo^NIzFCtbYz*u@$2drVf|BK6F`J|S;eC5
zluZUWOo@OJae(Hmwizbxl5HJ<=`$ElwFyh~sk+yhT1cB`#Z{B{_32VD4a2T!Bg{UW
z+}%;w>k7F1jrYG+m4iC4b@8%>zI47gO0|#G(;XzPP4`?S({nd-1-hTJRnQhDuSs_Z
z;Sb~ciSOz~K?UYTu{>WOSt9?mwph_$fB)mG1R)4Sq6%Q&*@_o^Ba`}3%V37N{ciK^
ztg7&J%Hw7<t8Vf7l&UD^+Qv$4HyUI8oigP=T`vC*%D}{=6@Tenlo&j=aR))`hjR;J
zKF54L&_<|ac{m2~rc6A=is}7CwMkF60~7C8@6%>dDz~V@f5zE1UP1wFUaTL~!M-PS
z?&bNPirlTk_m?&)GXMH;XV#IXDsm5QyBK0Tg$m0W@^oA`9(G$+(eSmWO!~h-Ay(<H
z|BMOu8Xz2|z(6tD`%IX3T<9`HRE4y5>nk!A#c^L8xn2anA<Mw~vT}_(6^H+lOdS;}
z*?0bLplGoi+zdV^)A>BaGlDcYr;$tNVdlNvEQfaNEm=xmuBS1&rmenM7l*d%Pj`O3
z1X|@&_zb%LG@zOQH1-rMOBkia5EQ67<t53smVepI8Cz|D3v?`VW;SNJbD53&pMOCW
zw6_d(k@YYvN(!El9P>0Cw8y5Da}<OB64Ol)D(3G|5@{s_l4O|SAZ?xyUUbss+B1c8
z7K7!(vMR&#FP{4a;meI{en*C02eiD_lf~7xR$r}4tw#_TXaBiJY^Yz*1{E|ypP$fo
zMoQAi#5lHj-@XYWM6YpJYquZc-@~A!^xVej4{K&^J(pmsFK^!c>Yix`mFfTJFTdcg
z@+KC(4MdCi;>aa|V~IKY+8}+TCJuy4>7m+VY&{tEIc~G4eQA{bPk)3jM>ykRogh{^
zZZjW`7~e+4{?j`?$U!AV>pI}U!48~7;A`-l{;u8ynk>_*s`df4e_8JgM660HtNcCW
zd-X^5&6ZM!-}C11mEr&0l^6m^8Z_)Nv)rPZP^KwidAiLPu(MWKTF<*#+RmbKM8NIL
zP%kTfJouKkjK)~L?sc`V?jNZh^M7Q6ikc7i7Y$*Q8=;Z)8PAwTKMiw~XUWziKJDtG
zrmFC-CF5Ys?z0=xtZjw=`l-&}fMMeQ;Tj=Ek9T*!m$xE#){4i|lw<$<ASiukdoWqA
z_shwzIrqA85$_G}IN!AY3g0?xX!z<lYaCz)s=NzG+*z>9b4xdH&onDq`Wqy|5HQ{;
zG^Fhb%4pN|IkIJ@EA)8he%WX>K3xnq{+2QNuLX;6_h-N3XthGkco^R4vF1ypf=$9}
zufKs%u+bCVqGjYe(@T$qV|HV?;M7mP0*=J%DQ&(S7xjOBC&m8Hj1cIKyT7UWB)3Q$
zW>565*#Z{l9HHJ-uJ<2K^S@vJf6k0nj&RwlmqUzIP0=jB2QW!_MwD}fn!7ZM_oM%j
zaK*N8Cm>_7SsqdM_$mSLJ$*66ThQR7@aQM^Xi7KoTgY*pl(%0nPj@qD8442!CBmXL
z`oA8RpZwR_^6S^yI96DJXC1Gr|6`noqbd){Fbt<6nOL-3CQzO0!DNZB_>tj%k%&O|
zzosTXcP&G4<qDKfrfY^8XAYBuYhcCutnIWYUc_1S`AXBI89SGWqu$lEa^LfxF?}uy
zr)zXBO6Wxo=ca0WJYRGAv#ipYcg?5nXsx58WiyI#IoJF2pfuO(5Zi-T_fQ0#K>Yvi
zBan`sDn5;F+?cki8D)fTE5?3Gnb&QYsV+v(Wt7{X%CK#@pVI&MH<Yu@<7@pFNKsET
zLsQp*{)^XdloV;>-v^~GjFnA=M{`wqTf#v^(L5c@=tV4q|L~N+uYYrtg%v?H%5)ga
zaER>>HM`t=wSNMQ7{Ytnd|}=9wBH{88EOA<Cn{+uaNuM6;{Rj)|9lNpfZ~(}1OE#B
zKN-k>)W`t0SVW5`MG1Ya0!_lZ$Sf{v4QnDT$B%#U^@FDJ$F7{66$ybmNzhKS=Hza7
zGQFz%UxV;9{}gim)wa0VTROfSDTQ#D2(vT2h;6527|%bQ?ewsdZbW7t|GfC#n)0CK
z{%E#h34dmd+tOMM>iY<XsZfx*{>mAZjxx<3=+N;zEBts|#~r;>UN?65aC>Hye6AkT
z4vppW{+fBw*QdHiNIxZc@LaC2_wSN4-63VpQ7&#z_cjCtua5g&(%V0>l+92BEuXaA
zuG*JoS{z>O>$LJU`0P&<*vAXsI6_I9%eQ_-D6%raQTm(*l5&6fU?sSlVfLnRJ4unp
zcPX(~e9et1&@$JPE~N#!_0>_4$K$w{+U&^4zr?KV`DV#}Bb3Pfa!jx)%dTOee$cPq
zq9B(Jj+zY-`}>aq#FCw8)^<m;{y4i(=ZAxD9xh}2-?q3_p$RUR-%RABe(JYjo0s`}
zL@4#jA?37CQp>LG?)6gb@3#4I0Y%7`s`iKXBNWAFCUhoY#1@W!k>`QlmoTuD?tCMh
z@;4^We7=RXzGaTvaxj*N_UwAi^uT(T6Ld?zp|VXWXRsAWudJFzGcQB&5O*;{qsONB
zD5JMHmP`m#-Euvvx8nc$xHe(P(snh;Jv!b1z^NJ+c`TQ1W&6!kJ3xKP+i|a~*LdWC
zvyTs_W#E=72&HT*P{~g|={re#?B#g;hSn@Cnwy2;D*Om6Eyo-lJ(#>r#Fp8ApvthY
zZ(MPAsowzTD?l6zvU8y<w%|+;$`zI?TTV%aShp@TPV=1Q7)$<ooEm@NWCbRHklAvR
z)2Zm|eSFKqVVNF~1L4=aX>@bP(+;csubEH77{uuVP^GhuH1rXze{BKdDmL|Ci7sQ3
z@v*%ywE7>Eq%RNB=N&^?Pg}P$OiRRA@SdT^ikr#oV!Ol9bswpNtLvAM^>2ucH_VO8
zlR3^q3<r5vItX9C!qhdQGgSwW@;iTycHJBr!O&wcL;)rHyguKGLmY7AVnx^g)E@pU
z`@Q0^w(=K(-V`Mc{$)eo=SG3m6g;-tZ0}Z}DdJ5JmuD!XnR801)zEV%RibrYFvG;a
zeZf4MWGjn3KBcseR;+~@CoqnxdOZj|tmYi5kR6teCufh#QQnrd*Oy!SZ#nMHP_F@H
z4F~RKTIM>wmQ}SM+BfeIdfcvfv<TlWb3R?4EH%>Mk&$;vkN+*~0==l!G$ql|h@t>>
zf=~+k0sRc$;zMSMzDXzn8O3{R6L*)?kDpNMDyrIUbnl28<pk3Go*#C+67rWfem`Ce
zvm!f?klq=8)`9ML<@Zr#)&Xj=pWhR)Lux3YE)X1=>4>-4c2Ju28(Jw$wNFAJnGg05
z%|GIVXaa{*W5maV>A8pU1-n&(X=A0u!-Lem$pmmta7k8VZ`ul0b9p>zY@yEuQ8%sW
z@%{3ljXSF9XxFo;_!U%OIgu|O+Rx57pD7bAA@%cVH=NQBMTQpRk3%a&hRqlQz(!SX
z(($-*w(N5`s<lI(Mo4DJc)RBB&-Tg8wre=iFR&riT=MEI?*HI+;42Q_&Agc;r90FQ
z1k!NY?sv0J>!(y)SljOkACHV)NzzsbkNj^JZCGU=<7H>f;)Qp$6++VYpka0P0*Uu;
z;-1x(RE&HH*E{=*srejmh6_J_fO=Uy*8`E0poA~_M=qhipK1Vj{N0|&8e@FqZ5YGy
zRysOuJG|VJA6oP6IsC0%%Sl7znpuJ48B^UzlF;q4i}V@eO_7mE3Mg3pt<afHGRPqt
z%o_XNJ<AU-@GI_lSO=JKERs?!wDusq``xNwe(87&{O##QL1nMo5~);jEYBhkqTu3V
z{pYqG9$JjK7x>}i8VRJ_?SxC8*__4K)2<GvfhyOE-g!a;A9u0B0(a6A3MjOG!v1n)
zyj-+?`8jvuZllCRTXL-wUi$@LtUGkb1<QG)7TKchUUcYEd50%fG5D+udkA<vd{#46
zE^n9NO78yzLES=MWOATV61m?_4vX||lbIfBx>ZvVdA^$RinWl26u$AU>Q)Q)yf(}s
z2sPtnM`}8%Y{3;}CmzbrTgUJ{NLCfGPRI8x!Mq7e1FQPbEi3ZhP_Rex`tcfwYU79a
z{mo5*Bv7Q3jdrN+(N*Qq{(Q6Kk>T?ZyH&q6=%Z^NZlgP%K6&HfmolD;V<uKp9r#(F
zwaAbq9Sbm%bxAi;W{6Et<D0rpr>A4ywkR=X)OFZSq*A6T5HwVJWJ-_2LqH|mwzjwa
zuebhPoG_~8y7q?(D<Yc_G-Rj!P?C@0)Y;NU%H8wewDg(vNo||b5S>2TQ6eLjRMgKW
zk+JL%&s(nR4}t?>6&LpmiCb%SD`Wu5-gU-W;URnvQ+Y4=X&7!Hx5<!Lqe!CM@V#u8
zII}_ep#HZI(@4wZ)xKNh!Lx<Nn4P9+Z=nS3_nj6QxHwjXa98G6i1ni}EjJ66lzW)B
zqe-n3{TuAv^C>TvP=>^6Gv3LC^5Ah}1uYs{Kg_S+Kxz4#6X7lUv`|94zS@}h=DaqP
zh4ozbK2mIhv(>hqzVqJGIfwsb?t6q6Oa6h5(v^<tmt0J99Q_NYC0Vr*M-Om`uwtvp
zU@V2Oz>}0^(Byy-!^O+v8Bq+Aeg`xaG%9*-hT!7E<E60|;LAzn7oj?mpbs$&yT=Kh
z^4;zi#S<2;xAk{ZrLF;hI&F;lYCF8Zp5r1XXlOLdD+Xg$FTig@bCUaUk4BU+iaGpp
z9R4B^(iPqDc=zt?=)3w*;2xks-E@7#nKec)$NY03t76~$<ClSiw*@iAY@Ln)rc^YF
zN4RpUHB14|_!OSoiJh#c$+~L=4pR9L6l0)qvepovg8%H~MW5o#tEtj+iP1Y>rHl_Y
z)FCAhXf^7Z@npu)=l!GUFURVwU488>ycT?bKy%5sNnw}~yC1aUa**kb*U+xF_MbZ7
zGtT#BS#$-$TqYnVwE2LN<xs;4pThtzI~#0YJ4lyl$PCcUXqHs-Q^Ss`;QxhSI=_(_
z;*ivMQ2D$%%cvk;&USM!^=;~g%Zjn`p{*I!`1@x{>JGB;iuQ*?X}l$z;18QUVH3wS
z-BgnZlJD!>z;2o5X*E?YMv+H1OAeG6T+Kuu?~&syVuermW;5M?>L8p^BV8Q_O0*34
zX(ChlI9e%Y4jwbTH3Vow$#-t`Th>)#z>yl^&k~U9r^PSV`a=w1gcdnCDrVg-n$&Qk
zQUN_^N*NhNuU3QwZ>A`gpzhS7+h<Sr3eCk?Yz-Dwgmi*@>Tzw@)>?{*vAnDE<nHq(
z8`Y?NEz9wN6j*YR|5JO4-6MRlx7zrCt^!kNW|A<@?%YIcY9&qqg#+W6y++sY9$RvF
zMFdIoy3rDpA<bjj9;)^>QsG+9&$d=3MQj2*kx<4Cn>JC9G8?Vt5Q)dKdO#~B8U6;7
zbmc<u+W<b^q2}w$b59Hz#phpy7|np-r(CoRbpVB$VU{(G{$!s_q^(HYdT2jVXO5lH
zCcttt)PqnW`-wt$<NM@!=*urEvDd+is;(C~d%T!uyBUU-PgT?tC<uU-%!1IDfF7)`
zy%}=;E>NP5$ZEGrLw2X8yzzZ+<yqcK3B-14VJ-6Q(CHY&u=SCSgWnTHRkE*DDasJa
z<_p$<=1Fy5t<|;vhS3DZUnQ8l!gKzdFmWz^A1p)SYr%bhVm0AJx>p~mO{bKh##Vn$
z`d0@V?A|uroJ&jcYy+{XWYkJfHFj1ZuxYSmm?iW_B3q<<{)AaYD5R?`YgkTUT5Y0A
zq?czjgp+!*p}GW~d~3s~UEBTjMgyrBBjs79bL0ExG*MGNDzY1VV*^3&w)H53AL8)#
z*_}Oqa)6Zep~vL#lt_E$pG$Fwyl=Mnw-Y2e(wEj>VjqQrU?bh5HP;t$FXA<Uj-uaU
z!%&azbQXJ!R}tnsgffjg^T~pFAU2AD##Mk09W#_LP(}Tu0%E!wdP{}YFnq4?m+9}-
zDPcZh5j6gQyZJLjnd9&Tp51SGfq}U%X!aq4lY$Qn7>?N!T{YiANW&$CGJg1-cOk=b
z(h&_%3URZCWppBAsnB7uv9WuK<@GT+ud45F!<z@bgzp#uRq2$p9gA&RN20iJ4cS$B
z(g<3iq`BaD^G9SO!=rRHQ@RgTY|0j2tWYw?0EhGtUiRP-l-*J}*4RX{KC{YjwW>#w
zDo*o>M-8U%Ri?$2Yh{?)nqyH;qOvw6iF7ddskM8Ca_0K>hIq;6;->8|B=e=9P7Wiy
zxt^X>V=5ne7<KR>8^XK}dn!CBB%3s3uMFxfjAmIZY&GvqeWH5>(P9x#smTtIYG0)H
z!@g`Y-u5I43pYT?XTob*6;EkVcJ=4PbM`bgcIkN?JI33LbWTN6YX|I2abKxF@uBVs
z5b6_K_5u}+HJ~V45rz8rc}!_lp}j?KI9qe>j&KUG&_I(rw9AQdBKD98CTBU0mJm^2
z{qeYs3<uxcDHjDB?Dbi~uP-^sN0l^wqgmU?`u3ZViHvV2d?;Q>1lGN@p|;QDZLz54
zZR3)Ci(0%jQB$6P!J(9;fROLs3Nnx(Q*0EInd=H;9)~B++)?tIB)z<)D*bRdZ*Wh2
zL1-7w?G$k#85<Q)I^(0In(^eSxuh$PpT`}ccl(68B9!jD$Oecr!8;&Xyx@m)V}WAX
z6eBl2N7AMNee$lucwo#h`?T@TTgCln+jjiIFZzs^5qjiuyKUyqhdRINbdx%?vI7x;
zO!d6=c;L*y_h{I5Bz3{S;?fB4#^*N6iP3y=cr=9|29yv?QhenUYNQv(UO;f|z$%?}
z7q%3jG<z#3{!c*$t_lEH3h38f8*dAdd}0?#5J7@~`}0h<E8A`s^J9z?#0scRz-I$A
z@fgFMlTMqQ3e*sZ1=)Q(AW?|H&-{sG_c)$jM{!Ycp^6py*rzUZY5bnKzbW9H|A=_D
zFPUP|ODGTpj}W*6ic5&Ee>{%Hzm-n8?Rb4YGbu9BG*ov7>K!MCbW2g9sA6cP2m-K0
zhfrlwI-XA)Djq~HUY;(;vC5l}dk~*XP@mYHdnJs**_jyQO~Op~R4>rS-ziO^aCXzi
zUJU~ak7vY)>mHEflT`#5gs`^W-KuFt;N9kvZE{QJtbcdM7kuzMm1Va;?e2KFt|Bz)
zgT$>BVJGw4fC^(e12JnEjEd3W3Bqs;7bL02C%UgwIEOgQ)44ixn#=}Fgd}n`{Mp{j
zZS~#MBRd8#ZqW|mo8y}4j08DL>GfEA*O6F6#w5WPieO*XH7Lj_iAzWaNIhA{7OW;&
z)|}fOlHA`F@P??_Tg9_|H%{o>hj%!{3d32%ynwp+K(+j1K{&<Y7bC$7*qdqd!#7oQ
zDe`^JPZa1o$q?h*0z`Onr}a#5Rs*U<b#d&5$RZgAz2Hf`W?4ffM|c-<4}<w)%rc;X
z-L|jKB{vij!Vw0Ew9Kf%?LmTcKx_Blu51C=mZmpTbYMk3;-krS)9K0&--$phrdENh
zTUIebB_z5a7R7}1vFG!@C91jA4=zd9JSeE*E9dJmYmURD#tk4%V|)fAu<x6PC8%JV
zoXdyHAX~u;w(o7@hQK1uUNuRgiGF@6GP5M?Mq1XNG%P|P2OdxfqXZ{u@a|ypx=x59
z2?MQvVIvU%<1p0TQ71?T{8nu1QwuU3?8GftW!o>ID8w&8hB#@5bmh0vAcW!|f)UIq
zONuh`_p^O(7EIP7ySqDcT-o6VVTekyyyUXF&VW+3Y7w=6(k3j}^PxqWi)7ZI<)ZLS
zMb{G2PDEhy=zC7%ewd&j{Kl*svPFYd8R+;6To<pE8;z=fh=iJgJc*5d!ajr)1-5Hk
z*=2!_U)ZK1gUlto)HCAvcLCv(s@{2oDQsymi@r=mHb<Xh_%JY#z`;|IK+9sR9UEvn
z1BLUkeg+r<U{jm)q%a>^JmX_pE%GTEFm(MR-Q_4ovK~hrmWU7<ErJ*dUlMA00tMi=
zXAE=xPg|71ur#0Y*-R5ur(wIp86LOKS3|DM#KdsS;)|~QFv0NBqZV!In@aOUs3jv8
zv+ex@soaY3g@evPPz+~HVZO-p%gD5O&>CWDWs@`R?Ab?A9pbV!{y~jL4Y3u9;tIpS
zFsN?3+l&t8P$iGlalu-H_+SJHv8sXy(WrOO9T*OY6o<~i5(h2NDTm|})05B^ic?RM
zjiFd|gU{B&HljX58Azx)Jaosya9QBNdo!#y`Yk+5ai+7xb5VO;M<Vg!9W&!vWL=I}
zaBVCEJ+A%_z_2-9SQ3P^#r^{Hs0k-C&*h1K3TC{esaLH>t@o_=&;D|A=KtElR*6n@
z?_XcD+@FD3McC?~^c~t<>Xdkn9Okqxe(GIwqBKivrFBV>y`eAg%^y+sV=FZ_rX#Pf
zs*kQ8EhZn&nAiO3r6>RtgdBK)^V|a-R4Awb%gu2&Um<0vcCo*1&r4%Ds!002(M_9Z
z_Z@5*#>jbkTHg|Smy8y&()dGeXzWw)RE|f=SP#zGhSZF#@Z{)Edi)+w(id{Soy}PO
zr?8C(aM-t@^K(qU>mq7%-V{s#k0a2vyC^;xb3!;-H@y8Wha{Tfj?5tNd5*DXACjT-
zfcGy9w{4WbckRKzkQ2c=ut!q!T>3?DUV);|o)*efOK6(2UbxYgWj*d%;_UXg$8uzz
z`SCf|L13(GCu4CWFAM%5x?8rWC`j??hg>A3?W}ZRYXMb(UCrJ7!aNwu%Vld9lRR<_
z$OZ|Ub&S^n;QH6QZ4xwqvA*Mp4rhT%L>ByCpYa^3QRlEltZ&ey?VFy&mi8=7@Ejab
zP6~M6(;K3ywBiUr(6lyFzlQ92ODb++ehlS%6o;-dHB#LMp@OWEZRmD%*=`D>;Z`|X
z7!Ic!{*iyGDJpoTid|)-g{)#l-X;Kr#u?WJp&h5h35to?JDO9xssLI;c9qu2BoJbO
zBlDd3Aas11+4<(QIQb~5mytY7iwt*Dyh9w|cf$jT`nV1+%jen43<=x77MTN?$r)$7
z8D`W(-RUF^k<d1*QG|AG(!q@1B6%m@ty<v$K++WnELDynA|s?V5}9-q&{l@u=B*LR
zNV$+FyvAl~1DJFOw~CC;?_O$uuhp4wHQBvomeD5ApG-6Zgvbn`;EB{S#yj&>Z_2&$
zHv4!NyLK|p7*){?LHY{zrolhG%|(g4)t76%NLAC4r!ns()zS@7ZQjCaPvLYZR%}%V
zrie#3@@9wqK23iN{f$6>){8fU;OnFRsV(Shr=aS7%d^aA=)(@dC75wElJ_a(xAgNx
zUv-G(h0mFA>Oy1kt7D@`)^9oeR&lCrhI<8p*1*~X!(8-l3L#gz0>s{+9tqm6`lfK7
z0?|hI5TT+NP_Ux0ILQx}m@?yy^k1PD6V%wRU8G$S!smSzj^yS8-VN_o2lP?;8K)wR
z776^mF?l;gfl1XQ)K=kXATkZuSfYbTpkNbV&nF`g&vm*_VZ~dKRhX}A78i!<Z{y3$
z@04{U+hVrlvLs(L{C>DVZPb=X3z2!m_GetWK?^Rw33@7_n4HB4;}r-KIKXs&Td_Kk
zbPR3hZt(iuhDBhjMq}OkUkuWRyC!`4t00qFY04TXK?)uAZ0B8kw_3*>BI|vSjuCoW
zk2TE@Qd7z69&+Wm<x*gK>Ej^9IIX8hdO~=j=N)<>nYqjVRPk$8H<DK(m5gcYN>hcu
z#8LD)iI*$Yr-S7`UG%$DF=b)h42W?&EP|o{C&`G0@h{6ZK~98`_e<qy-hAtEQFr@@
z9<r;P@O{t@Vz(_BAqX$m%w*AHr=6n=Fur}SQnpMbu<d1`nlV}t%GQHB-a5TWj8?~w
z!nPD~#BK;f#%mX330rpyYIALn1+2o>3mclC{^_|4+M)6^bV>1bIL6GS7;)tw9Ta#f
ziW7K0GBH{CmMN#}IalYih4qhQHsf*J#Wtpb3bWmW9r_Xd%S>Trl&oitO}-6W6>kg3
zO_X|ur&9RvF<V`=5T(gI#;R1W%GFy;Ch>CxDQ$sn3_3f$Rr(PtY7lEe6%%=9(RO3U
zv)$Wgp=esk-Ctcw{CQ>t74)aGDt}nng7ugTA<wOQ3nNT;gyyj~00(jlXPmqB%z9`S
zDD0)5F*=hXainys#A!2<9?5~=fjF(bCMKMgmZvPH5&k#>N46DLQ~L`%<qbm@I=s6(
z*4&12FcD1a`H(UH8_s?G;KqbeXBXzWv2AZAHWRvBR}S=X7n~sq3SJOK;%je+H>7-{
z$wTNJN^%M6YChpj5}U<&(VL06x7%2CXO6{BiGFFJ*!~vJ8!CP-nC8L1N1s6s>h6Q^
zhj6QJK7Ep>fehk4*a~rdH=AZ1VjMQUfwp$B<7QYVm6|||>kF{a+Q1g@-n6DFrgdjg
z;KKke%ckXha5#oJ);JOZkEc^~&)N4|5b}!e?Mi4(jFwl=B$bem(z$NuWhdjoTjo`4
z3(-(5p&)JDo!DQ9XY3*l?f1LPeNv#z^^I<4Ie;1qB&0|wAwGp&Xv`{pK<^^Cgqsc&
z?-&HgWn*&d*$Qrm6OiJor$q5o@{J6wwv$JvkfGb2`f5{o@Nj)(X++>#!c0FNiEPL1
z8GVa)Qt7D%88T#N3Y6G~w^CG+6Fi}#;LL87T#{+N>^n)Y@S|6?;ssc0%tB`0Qd)6-
zX9P{3D_t&b3%+2d5%p3qZ^K{zAe3m!&*Lso`9qOP18U7oXq@)`e|rL+f6L*TvE;aM
zBsf>YCfZ5QDRF9yT0j%${;mg7Ay4&ZNA1HT0)1M$eAOmbFFQ<`=E{tGgx;_;jrbu(
zLJMZUC1`WJ=LhpIlq^_^V82L|dg*`%3q{s>sd!k?Zn4xhJ#Ff)PLc8RiBESkcR3BA
zfizAMj&-H6LwJ*&bZr>gI7oBH`~l86gh)+16T7_lL`Zv{L6Z@Rf^!A7W*=^HsZFr$
zqpDYI(7ySGxg9^)uW_=;L{?wbEY~ovhQw+_IfZkn?WzAXHAm>-A0moE@r>xUbL!3r
zFts^h9rDnPATrTyRr{)pBX5*-ON@wGLjIO4qlCW;(fsR99H467!8DlQHYuH{q{c6d
zV|bE~r>>sMi9AZLoazq|4&gHq66L&UvE*RmMVD_1>*ioC1MmSDOQ?ERIFWgw$>h+Q
zuA&_$eusj_<iu2B6wEuE#`cb?eq$BAjO_rY{pwYYMxhkdT9TbM4p_R)a9aqYnO31>
zJhH`GX^BdJ^l1oYvY+wA={Jbd<(DNy!GGMDduvY^Z<)gH0N2*6YdyGD)zD<0E7u(p
zi8?JH^KE=2VVZCklONB)FOGq>V@h)w0D4)Sq0X0XPz0A+!@=ASo0`PXrY<CNa!Eua
zV4q6OM5*z1J_Z%Embm66?9)n-O>sFSyzWZ#*6d)%#zi`APG{HC|D+v;s2a1+B+a=+
zbgO|hG9lr~Q6&51;~kGP*G<a`Xi0s4oD-xT526C75`XI*4=$-$C%H56;e5L~?T^7!
zOt6|l9J5sKy{vc-UPr@Q<V~|PT`ypAW_Aj*{B57a5_2|f9^xgPr^#4g*tr%0($rEU
zn8dxcs~4W#O(O}b#QUGwrDJ7j7$K6e_8rcVE2cn>QfnqLxe%uVkj)iTPGCs|4oJ%+
zIBGq4Y1OjWw#6Rc#;BgkF<foaS>xOn8;l|-(XiC1?JnE#fKFcAqPxs~<)wc5@RQt+
zxx0w0Mq~krZ^jP^-au1q2=#G<a~y*LV`q6PagIZ>zq3`tq<U~Y`C;6)1NAyPkl`Aa
z_Pem#RCoC{0WIHZaEMS$ER9n}y<~CMB0;5O;P_>&8Mg9+@;n>Yr|A=K!+VN&WD6tZ
zLF4MUKhzac&cbLeWQ2Vo^_yTfK-g$<z&L=jwsNacTa87{RGFb<6%22i+xAC>H1wIt
zYU)IwyHkyTu?&!4Oti;7q#(2|dmpC7A4EMaf)%2gyWSO$jzsaMCSPB3JObW2eoH|B
z=~0;R{I@t;${np_mm0ZIxO1N$>vjRE_fSH|Hr8{7K|DvdQX}epq<WaLLNmQZ&xl`k
zJg}HBZpEQ$+l-85u9irVAt))DI%b_#f<#?x%Q8`h&lsP(Yu;i+O=hD<7);EH%~)1V
zi*?}5Hkp|i>If8c>WI`J;jFgYI0*1yq%cNNLS?n%Tjw3hFj`dykb>P1UO)~BRVaPH
zo5{D#2Uia}Hq1A;k-~-Dqd8act;RH3DW0n|{&zi;K}4zo3czmA#{?qB2pwYp)!Gjl
zlr4G;El)c#U%~5yn@k~E)dA7=Iv=*MX~as-4dGP>K*%OLoF}Ds_4#rA#z^-%fn-Q3
z;}EK^6obS{e%x1focOQv>y{=Hr|azI5;`0XB<(c5V63gGlGHQnc-H3aBx9>U%-l~P
z9QC{mvMa?xwBxNSm#Ro3LZZd@!<s@in<CwnWwVTsh|8}}Tw7*8#v^Vg3s&MD12uXk
z-2OHV*-G%fLz|E)gwkP16a|0cF&Qxiklv61^cxRCuvJKmw`7qJ0S4Ft82GgsA+QW4
zy>)~@Oo&)VisPnDhIs?mxp0nubCfulA@kav+r=**i?;C5atiwkp3WccKI#-1qoW#B
zQNqz(*5!_bEe)K0(MM_}<Cgh->siFQSu^jLkZeJ?>wXkiyYGEfXm8mW9}A2$Op|UT
zYR){((^8bVl_NOA(pudS5qm8@HHWv3G5V*#rYK#)nONiyPYZ=2Xt`<1ppf1^Fm+Af
zrdMfTGKUZ_5(7A_;FW<#Kz2as=FwdakEja$*;Gt~me6_?cp``$#f*1|U_9S;LpczR
zg7w?Rf@dG?MaT}o;<(c>@DEpI|1|9KvwE#kB+c^`!|ZGy2Fg_0YG<&erpn~T1t|r5
z4K4|3<DD?!1&*pR9)=u1WDz$-LPf5Yi$gJ<{hbO;AL`Qld>Yb3DNf$JmFdB|_)#j;
zaakPcKXFxiyoc$Y-_JAI%*<fc-U_ATMV9Mt74*(`=H4HD?0DI0X_Y4$dLPEE$m)ss
z%Gb7iK;{4NW|ipE8}$bEKJZ-c`0>Ck$jum7djo@o6XSN^C6#TQXkqWvYZX}2dq^`}
zOYCnR+9SMO23hAc@8>JM47|za8AFgHQ%ovXnZMokJ)V)dk8XQ+EMOQ(KkQeIkw65Q
zOubvY6TZD~OY8<`m!#4L*CYlWF_?iiYDQ2CRtkhP(kGKtSY`M)qs}qtuz<mS9cjj{
z3?<76L{7e{4n8??lHQ_t^pjTI^Ye>rkChd3O<09?(e!kSzqLC4b4ce&A>RTDe_XBu
zD4cm1xs$<67F5LRUkeaDbcE!ZRt2O494z1!arD7<vzbSbxPl&KkL;2V^EtL*B9DX-
zDn*ibG4CNUW{qe#uu&cQb?AaM($@y1Ag-cTt)2``R8gc$VBqX2o0Uap&bO4h@c@Px
z3}bceC}Mx>U^Qc9tMLnbvJ@+e4P~}^{H#Igjkq2aZzpX&Yf1FrQz68eQ~3^4rtYk9
ze6~aERi<P4q>p)W=xAT%h4gIK68CCgQHta#JX+|Vqq^NYVUZDNgk>C4C}WtjNqVdq
zm|D64X#}#}n_e+`pq955P2#BU>_N5@ug0HE7Jv%C`+gU86yt~iG}4`^2j=cx063D~
z+C1YNW$xj<5bKCp$m3+UA#gVRLSG0&fnBYZ7kgZM!23ZxH+WL|-jXBO4D3l;B~OF#
zv|Gw7<hFU|n3IDvaO>OZftgIBL&0;ElQsv=dInAo@$IDSXkD$KK+GEQ1T^D}@u|o-
z;Y_r*JwfE1@#hfRcd8I78m#Zb)!+mRL((0gI>I?;E_d7c&%}z&c!JwFeK0m_5%jB}
z)?2~7c><*IPItMqygJuOzk6_VKo^m1C6mLiD0soYcaRkXcd~z{(Wb}_(QeQHvZl6r
z%O8!UR~ntec=5MHY_<+O&JDX2BOc`WL+HckW9mu(^82L5mM2u#pYwSUHm`Mtp24K>
zwj<><X?bR3L4qe3J2^tT+y=;XO&8<`;G=NYp5(i!Tkvqr#p%rLf98W7xKNOy&0Q&~
zkqMMTC#SAcfh_7;xTt*|DGD7cOv%LT?psM4TMQW+v7kO~aW*1O$xP~P1M=tuA3OI{
z^j^Cb))Z=vS|(Ts?0YGD8|p3q>(ZP`G13L*hoa3m{l|@Cb#!6~%8PhSj8}D*)mqOn
z*b98cBggRKM7YJ(jd^P4jR{J+QmjrTJx3H6hu#h)Fwj&>Yz%)jrzNYj>`};4-hB|F
z;W}|3L(Uz)-a&ys(mpVQH^Z=mC|QzG)b;V8kKF>^O-i(G@_gk>C1st*x{7+^9dBG-
zudFkud85)f-?du>5V*J)OsmHg#m5#(Q|l|Q_%<r`NAic5j%6MJ54?Fd6(g&1O<jE=
z9sY26d$f+<rS|$gh<2wJuLpl;a5GdK9{$rOG89I_RuZLkl?S!$lLaRdcH4<h6ws5^
zncna^d|-MWx<6W$$|Cv4b*<egnynPTxr0_ZLnqL*B#$Mt7SKaz9Kq0bBnqb&Uw;za
z62l1abG|C1gx8aThVYYYn}*R3hcr<plY*!1)Yw*#y5Z&xCJp2BhsAZv+ajTU#i;?B
zA>UaSNR%;u(u_-=76`K>RDdAddWXk1$G2}^K9jv`{!>)K?0i7{Taqdsz@qlRWKZJo
zw7maC!cq@w(GbVG9pPk-xSBYxF<<%|R}GeNR{$VhE^o$UF;!s*B(U5dl6*R^9jcz(
zPE+-)5^j0j+fMJlt{as&clxL7?<7#^0O5;;Re^XTv7|s#`GPW1H2iZwCb2=5uXtiU
zUZG22fo1iWSR=kkf+;7dejax)VYsBM(q2~`fUOw0%2f&A0Fu6It1#F+h#+oqWQI?@
z>g4eA`?Xd_mlY_>9MVNX@4P)4HM5yHgOc3E<)1W0(+MNW1JX!4m+ORCA`Ys>p2JJS
zB(O0vHG{7!@bs?J)@@qFS8DH(7FOvnkQKxUkL88stc9%4SMCs0A;l1vis9Rkj>;L|
zE-BUcvg{Bb&hT*VdBtq%ke=yk16WPX8HPFO$yr+(m$fhqy+P)YexW6S(A4HpoSZDK
zO%i5_)E8XeJ;1&Dw%g@Cr&dlc7Pp)6e#9A92zI}S4F=z4Sl*+2-z=arAc^~NxQ_iz
z<~-8Z8-71&{^((pjT?yJyArw97}V3^M8Cn9Y+%gQui`}EK(;V^MVjUPZ+4f)hun$-
zwGZyCL7VM>UVg=z-idChXw>Tjwqb%psn*u02aDX#S31SRLg`GZ(VoJok56CpvFe&u
z#F|yTqf#!&nzCxbju`rTQ*2@^m4KPX$D?I?poomFfFm853<B9)%J9fU=Hc@%<DUmV
zEW4*1Ld3lR>6{nY0<zB~<;iFDWw!YvA}TR!7mHfaiVvyoU*(zbhIDP0YtAkWf3Sk5
zx~c3ubA>%kK0FAUS>a6@*e`B9Ze-5n>fCt|{15fEQqHFfBNU((Aet?hQZus}Bqig>
zFz!pg&coPd`@PQH?JY2<VaWqvVnd0r#8;1)3_b?HI)U3rworl%EPPE(cK=xP+~o!W
z;a7y-l5KYG#$YgI+_RK+?u2R{+-5HZWZ5z!NC%5!ur%W008?SElen*%PH31+`1bRI
zgKR*Sn2V0qjU{McRCo#YB2^(y#iywg1d|Nol-P8oiLec~q|{Fqp~}`fekld5v=|tZ
z>=Js9eb)e-3+Sp3UnB%Akj$W2Rm)<2bSX@nCkj{Q2h&?-5Od-h%7I0ux~^Ko77xx=
zgy(*7mE94i-J_TTLf2HEkcRf5l)l=<JZ}&Wsl$-ANxnY^4Z>P-9@#f^|5lVcnml>Q
zA3;f0!>&!?HypzgQXNQ|n$v1NKuS+EYy~@8{P(*M{QlHEa1;yYy)6RCx5QQ;ee9Kk
z6UksdNHzQJ9uU4M{8sx*`i{D|XIE!H0udO?eFyEjhv$-(4FAhb0JRVXZ0fdkvy;`?
zbv+^k6zrVUTVAne435TRPH~7?@{}Nti8TPwT`ckqnUe9yFfNAGPpFs(i381IvTD=P
zA%fS5J5~TL?8PAsArau#v<GKBHr9=GcY?`v#$5SbD01p~ymN1o`wQ?DrAmp~;+KPt
zxCW7_#dPQmnBj<6XHC7aJm)PI5dI3i1nuOzr;PeHE7qvgB8{Gz|1e)#obQnZp<{zc
z;TYNn#&s6LFtJHJ7=+|R<nWyO7$^rm#6E;e=x@h>7Hnp;2)&VeyKYt(G3qaB4*)jj
zCkrpfKwiaTKUL&hUMe~k495sT&L1H4^C#^e5nFoEx2}a4Og3+~jcV%WHZ#%cYEaqK
zisIDiGf*w7`*AVi-ZNXwe>eYWtpyPJDvKWcV0wZV?AekADp)4cM6kXc>CHY;zpA_W
zw#j=3=9~x9I#5Mh4!X@5%oWC(CHD!LsKo>>At1&z{Q4lC*-2%tz;=R$6x$joU`-cS
z|BNzc;rTK_35G}vGS0!IsKDQQjIj431P@t%%OS+KzsM98Wb-j2i^4nwxbL%KFV>5!
zN+L*ldm}V7c%UkvT8D`1$}yUQ9|&T!dWGcGCPvC493kCeIQEKiMV^t9$fj0<x|>s<
z_#kH~sCtvbx4@bMk>s-<oW*}sS_6H(Tnt)RZsPKvwtOW7h6#pn6nMKYjg5Y4Ni~oE
zoVhOU!blZ@m$Y8JE72wcAouFCk9F}V4rpQwHl&OVDST?z_y&x&;9A$cNOcz`>@JSw
z#-bTQ-4Uv?>ruNXxCQTsgzRW|&h?EKRCHw^^M(k>W<?;Ya)(@Xhqmc|Z>%?%c3HH;
z&`i117<!Kv=j3OMQ2x$nSRRkb44!9;x391SJeP|{`pkC6{`Tvb&EJz4&@zY$ZUNzo
zsa02mNgXl9L><yHAvuXXj}{mm?9Sbob9i4dU6j&H|4BDPS!!Vb4=E7jKS9i{J>aYr
zhYSvpq|cobpJd>yUf+I)5Ko975QlG}4N#ktaCEEqIR7i6s{2?U4nsn{ofUdy)zOx|
z#)LHldi0eKpe4xYj*-GnHNT$i^|dh^&8gV%(o*7i+~v^%M#Oc0f>0K=4##m77qL4w
z79%9bd_>CW0Xp|Ju%cvU>0Ojb2?AvrHB&G<cFUOrO{UM~Ccfbz{*`HdYNEMV-jD?A
zZEN+`utvgoeAug(Pkgl?q(m59p0)fya!lZ;T*4@zyI;xE)+#bp=OLr@sh)2l4~Ip_
zjyUSMkI_7*1Qr;LqQDFp=I}Cmil9ee^TWPndaIAm)h*NX#++fLs^Bh3#E)0{=Q3Ae
zy{S(Svu93;niHm>BFeaU;Mj%Z=7N%jwrBpbv64bM@??J=s<~(m&*P}!hB&bTYcHn*
zO63;6CzAO8!`@kVMZI+opOPFp99kL#1Vp+)KmiF+N$FI&JEU7`q(h{X?if0y1(a?O
z>5wkpGy2rM_xC5fYq@ZiG2b}*>^ggYHcp5rQ_JFGadLsnCSE;bGSN&Yvo+KfdrnKa
zY~_5uIWx6;quW@h+1E?sY{Z8&sLx^M)N3$%GHcX!zN&Ofmn-F))Fh2$Z{-$Hi6x;_
zagt77Ts2iieIh!o!#iZhMRhp(d5=2>x6Ki=L!|*?AB8n>Kf@R$+QZ)-E!qm+??}Gz
zl&lcd8tb&jTGvD6PKNCCBMlvEHJ#y#g|HlQ$Ma|p>e@z^m*o>~qEAv`&zZyLN*Dgr
zu3rV}ao*t`sYvD$juy8rV(Yn1(lQce6s9INYno}8D9JUBd6}5})gUX5sWXP}nhHOh
zTPBHy+X44pX713##rI*aCA!JcBcM#sDyI*2McJV4+#hNFWtv6%#yw2u?N(HLNCs;8
z=*$=UT=5wtk>TzUBRz9gYUphXM=QzU*7Pa+)bw0$>7@Y$RNjsl(ZfgC)Y5GRwtny_
zt@db)O;%QZ9V=aER^?GRQ?wb?vOoeh`nJ0?B?<clq+zAaXWsinjKj(uci{j&X=glf
zpdb?VWqMjc)J^c>f#PT|*>2caqBH(NwrE8WNe6ZC0isjpYc8vF)#Yj-8`kRc*?c=<
zcg2=YN3}<)aMhy<5-!2T#4h_*B`E0i5)V2wB9xq}47=4FpgvzD+}A{Yrw+zM^xazW
zm;BVjFAv+8Z4RnGaJ_^TnAgraD5`Q#$wY_~2Sq(_;0)$i*5HUr5V_a_MRD{A#~@WL
zz&g^&MHZ=Bvx@fus#<X&I7&bn-^;~|PCyHLz(XkQVIZ5F)gqhByil=u`#i>#)SLep
zJLXH$t)+cPpWx{jt$C1)M;&@la&8TWI<v}315tiG!$Q1yd3Km+4&=ivvIJ88jXd1P
zd+BoIp<^;P-RM|2F&=FWpG0=+7`$XPw+Rz#8PSI$HQmi;Vb9IoXsq909h5(&*zGBk
z_BAj~wAV>f7VMPvQ9BJ|_iUfkj5?ILe5|9Sr+&L&>_dMf4;kLVfi4{7K(2i4)4`tI
zn)v2?<6iTY;FB=iTpsQ+Ml4&JK3r+7T&ha1_)FHG2`@#fii>n6p$aQC?*w(OLT_)d
z=y?sZsek3JY2_^m7*8x}cxA8DLd2!x(rG*X#V8|f<MLSRcQebP1`4OQ_$SFA^pxl>
z5@O+w8VU>aZirGnKiP@-=ai+W(j~EU=x|OBU)%MA{t;=6PGLu7MgxL=UbUhhN5+Pd
zkA<y}T3dsZ6WShWtCCIn9ZbJ2#g*VeGHs0$piFGm&y>wd#^fH6!)opKZ=>l$MKw7&
z4a3Drk2PB+W5=m`Ml6B;15~eT*b&#}NOJm=uVbF}R93s-S`UNjJx#kdd1P=p`LbQf
z0;U#xX)&{06N@Ymtev2Y<^ts&0?k$X$!!(5BbNH%xV`So)-bk%D(6lf#U*V2{F^q0
z`Ivx9RkP7Ihu&?sZU-jYO#4K>7(nR>*Ym<Zcn{H>QKC>&ElQ6njavBW1;3?Mr3P`g
zfwkVtpq9^8WDlrXs=+{8dS6xHlz5i9FQ&(_%Em%D=do2x$7R$J@-pH*>7KpRMQ<jD
zlzVV7*13vMD^h?_&u6)s2~cu7V%+Yj87sbwjS5?{r}gce7?oTI@JFTdaXTr4wEL`k
zw#%P)s>);-LUU;&SUBp2I$?4L$rbXM)@xr%mJ%I;rMdzi#9@p#N(G?bD&;?cOR{Dj
zwP3E{#9)om_QX5!^1QL%2_#m=^gS9p;0v^+tOG#>l@V9k8AWpluL5WOQ?=S<^(1LV
z;RD(X+XJGlwi`6L!;mXpsW$U5Q+_}3qY;^o&pQ0yDqatS8fNmScCGa`vt&;7gjV?P
zdrfZ?#aEnBuRPYh;enY}TPLV2TT5%T5y-vl@eGb{bK~paSg8etY(0mA-7i)zWpBS+
z4Tg{#wFOs%tS6D1u{uH_D-}ImRa0KWaW>H=%n;rrO@qkOf~^E!pY?~l(-fP`dP$M8
zE4dc?CYJOEfefAx)k2ozn>nodyl0u!&^eo`#Hh%JN9DUel-qagmaZncygg|RZ0-=8
za<y(U(&Wv~==M(ZF4g!`JFtxQarcXKNc&<CGIF~lN53GUe8lS{S;Ku&d`yLx>O+3v
zZNY3PPSH*fNcX^Go9zG@?^Yy=g}lOjoi@KOB@B9yUg+v1g==$%8xf_$4}*<XayJ69
zzGU2+jjKv!qdQb#n9mfDx<aMww=MnJ`yiQvQ0+-pM@a9A6w7g>smFMXHkSPd@%#H#
zT2Igm+;nnRG(QQF;JmA0eSuf70u8dDErB2QalcmfBw)W`rDdHScb!Js6y8BcRY4!9
z;=PJq#|r0Gpk>kB1_dj!7`3a^lzO8tzUUG0chjJEvg+>3kUx}~>c}h$@TW6N$z9c+
zK93SyjtY2(x8~<*SvnBVQ}U!Gjv4*xKr*#v(9Dj!sskf{Sm6`flYu0M6cS34?^p-Y
zgdYZsp1n^SWsY}g9wuHkQGnPplo}WX_$~)j1Y(f(;j}xbX2kPX5nx19)$aF7B$R<F
z9sWRhEy>4HLJXpy&PMQ;zu>=F0L$k{V^wzEMRl>R?(=dRq8*Yl$sZugYSPr;^k~nx
zEA_BvKzt$#ZX}ppn(MU<S`><51qbQY(zv`QrBj9>)Y23M`F6|atdV8L`q7~BjlDKQ
zJM-~{yHAF+GR2D1(_KNrn8UydC0bkehl1mz>=Oyk`OD`(k3@huk^+Zwf}qh^&hGtt
zS&;7bQyuH-=Z2Y7kvO1>$xKr@7};O+VuZbJRcPL@DhK-hWWo2>i+B$EI_mQbES5eF
zcN7MQPFbrYF3aW}(^%22ACIQ8te<$^Jvs<xhlB;nG?{aB#K{>eysUs{UETTqq#U&4
z)IxEzF39&r_C}hMTe)l$!C}YB70W!7lUtCR{FW!oF!jLSyvO+Aa`-yM#(VsnIR=Rd
zaQgDX6wesnI*lBIXO+$^^Gk_QLH3-Hr7+bZrV5O_yL}J(>_E<NG^z#{mB1P@@2=Eq
z*qg&)$dP+*^u<`SYDp;9#v@P0Lr8m0@j6W)Lr*=PF{a`{C~wQJ?*c8K`BXzF`_JSX
ztdnVL2}6z?`i<IR7iu6GTZ_`#On=-|Q=)y`S6@5vGvaW#;M@mW`pFa!U!OZ^R~st6
z3SJ?}XKgQcE8$zF%M)$P%>nmjw=l5lCw>Odeqd=<SHAk55|+usHoGqJtFv*N#Ye7w
ze;tD$%vC`7>dpr{YWpM9#7_K0>|r_@A2Wsn7IY>(I!@iK33?VEJ3Y){d?R9sG8~YD
zK0#0IbOqfX&q&(GDG3YEktMjFz@^_=)vv+AXpe#<S_bYM@ET#YbS-04hGhyVT}f8i
z?SR6i%NG+}r_WZJG{WP+0+gz{!k6FbdLMrcqcwq%z%`PHOVJp8%y<q^S322andUS$
zW%UjoT@AR=`RHM(zzO*d6cm>+`abXZoHll{dvZ?Sl~`1UXR}t_s#4LLQii{QolPyS
zeu!7bKAvN5S#q65*}F-13o2p!_Tj57fv$x)ygl|z%6>Tt4{Ik$qd5+=((oVj6Um5c
zH*R9Ab1o+(`jn;x9!-jgve;}@>@JCvA~EW1aCLs(%^z?_^v}SVear+86v?o6?rofg
zD=d><zzuK~JWY8L2yrH81E}0c@AbOLEJ*9w<6CpM%h9-@K3+aG!E#riRPpWLo40ri
zDvWBc2xx|@1zS_Nub;+sdU7+(LrHypVypJj+NP<%9TEsD84vm)+JUIIODE0swxUSG
z78aZA#D=p~=obpveS{RVXWfDXSKn#aVNVe2_hFd6y$QQqS|sgwvDQXurJcZ8a@zOe
z5yK9;Y2-d!I^~S~veUDNah~8Ex$Uy_4JFsv)8>j!_NFj_AAw%W9?&@S!<dVuCKsaR
zVAX{t=_I!RV>+wSab}<Q^atTy%cL`{!FF(I>fK|x>=%|yKC1!xLr*b7*4tpBrdCU&
zr~8ykykz?sqF6gpo?V{zNj8ypIIV4SmvNucl_c03?=<B|i`bmYQg&W~jyKT*^p6Ab
zUsTRaIfGn=DEW!@_+s6L^E*9580QP1?X@S{p$yTF;B9w!IhsvGW$J4L+eRrL@s{Qf
z=5dz2Zt5|x=;`%59OIg?{L=mOcA=wvie*Ictmn!jsAnwmp%B^hcSQM%D$G6aG!F-I
zjXn|NK$<_M($^&wcoOXrC2V2;J<;bIWq=HMHRTW5ZHi6e4Z4?+_=c%`yE)2Z(KFug
zr?A3C*t^JBa?cks7M}1W{)6t=eI>WI;<f`^LT=L*wgZq;bzj$6`<}SV{W#k9IEV&U
z1ME9~TgG1l+k{YC>twPSveok#3{?|ZoU(lILjt7;PRHs!K6%ZrEqRi>2Q`+JBXpg^
zoNHp#A^{<Fjz@ZYSkzUxn~GR0&B4kOrR&ywcUV&hu31r-2wGJ)%ohTllD`u|m)elJ
zkE*(dLY*LA(q$JKr=gflb8dR?7E_|sk1&=D&A3DBgLbz*zpulOf*UrPu($I+!bgYB
zrVg??$z{+#uo|)6fr^jrkeqT0o|aq~`VDwP`Yb;UKou!#cN?~X;f8h@{9bFa^TWC<
zK4uSAW4b~#5{M5W8LL9UqzNXC-NU0bQ{CKUC#gTsP5mz6`AXdM>*UBB(NK~{pTr`n
zvbSUP(C4b`S=X<kEBugC&J0Z+vrChiU13v}2z15n!$TBG?B96uEM`-A*yfA~>{}fo
zDYK%%56NY4(Cr`$J||)V;;THtqI=FHB)p2Bnnr(RWnB`>zoGN<>LZRcaG_Pe4K5eW
zj?S`3A?on%oEtV{nP>k<iaNR|#;dKQXNQ08*9O{X7`1{iJjI=9IwT@FqfZp-d$|4L
zaB-$1MYEMOGcnhzC-g)#>~qqM+L?-V4~syU(r92(Zf>}}uV+N31kodt@XVc!KTRM{
z_yWH#=PuvY*&FcQ=5&UAU}J_ya;zlb)m63I_of$PBPckKF;omZh`W`2t;X~-2+}+|
zeHD(o{d>4kQE<?jm=z~J*6^NlyxVZJ!E8<x8*^lJuVVzpUQ@opcg`v<pVE(H@!Fy@
zo>OS3oYq#N8{M9f@1YdrT+_U)=Fp>P@rxgwTsyDrRjwBC;cfUiwWnp+-Ik3DdXHJ;
zH20k2EC=rFL^AAnY>Mk9{ZOo3Nl2Y!eMEH~cgX$yXMgUp1?jwT)qa1>{eDJtiBY|9
z{{{jBN?WMFuj5i*&JP$ln?Un@>wd}U8=QDxprkUx_y-EWnko5wf?!c$JCSrjyp{|$
z>eeB*MSJs0pg9272JXX6rmJwB)tu;QcVVWxZI8@(jQQi}x--OE!|2pXyi$$g-cDlZ
z-iMB6$On0Pl&_4E?#JOKhI_YgC|x~I_zWKVhiS~W#|i2Y?j3(MimHF9QLG_x2e)KX
zuMN9j8BPz?ZPNPaH{($FMue~Ax~}jN*)t+nB{x`YTX>TxHOfOJVKHSdb)Xfu)<Sp5
zQhn6<-7v?2p*}`*{t4=3865T0=P>a4kzN0H3f*7!bX?4OYLNb?n&E9kWVTsY!(Cr#
zC%h!w@Ka-#Z-$7F4oBGAZ`FEFvu3=DHor*LE5uHXuIPa%3TX0oR@?*47b?90ZW4I9
z-<v9kVV;ZcV{{rc(%Oj_m8lI2N$r;#p9XFTb_KYkiV<pFw9CbJ4ztH4OS$DJ_41aI
z1YOfwl@i-hVR}Zjrg@$l`;sinkPi?;J!M`j_KuRm1l$-NX}-V^*d+HRGjx|x|I!>`
zDt%ku%ifLT`UN~+<>I_Ep6OA70L<S`>=*{lhqjz(HxgIN(28U9@!AhHVYFq8Zv?+A
zv}YW?W8vUD2!C1-!7=4mEr1sce^Z$Ig>pb!Oaw39)2j5&WO)LavX(T3W}b9C(-7v8
z`d%o$Wpt2uybl#K2WfU>?Cn4mVJ!-2Z6UTi%?yX8(`C$5I?%@2Z5S3Vpc{CB#(S6m
zr&n1@F{AGJJU@o+&ftO>0ngEO4~V+fJ=V}1Vcw^LeTy~cU}BhC<y|25olY@2#_i_z
zV^e%uATZ7~ZRx|nQKy|MdqYM%jdK!cSotE*?q?shp&G`deDF!IIe3OA<d15_Rs%9Q
zFrB}^h&@&m>!NEyYTi_&u#886nn{tWh51nR$?de0SkV>NcDf<0jX(^=SC*G-G8{`r
zDr9{MDLe<}k15s?oMO2G2OVnQH4*!80n0}i{fidA!YG~J7_V8b|J;nrP0r2z5LF>5
zT3GAu>|f5lBi7Crs9*G?!hDk5oBs0aopI+3KD))%eXU=A@;iNus?FANuHaC~i4587
zuD$F@KnFL&21a9>ER>4?=GFP;buDdDNOGa*?Vo2C3Jbg)1C>|eia}c3aO9+~N}eGk
zp6E+%L5<u$dmv2x*sth(-crFHmj(Ktu3=8B&<qn{mF|ZyMXr5_Pd0;bT==_dN*iaO
z4fGPz!fZS3-*8S`=80XMd#lD7(U93|?-R7E_pmoD>4{43JM~4Bt}#_Mh~k%G_&#&o
zr#Hx>7)BqB%D=T1wZ%?2IzV*K(bl|%QY8oO)-}AC{9)nHFWGsXxJ99}JwR+}@wuv{
zX0CXqSDQ$HIqZ<5j$3{3cvY%?n=@*a(vq?0X@<@#kBSn8=y80rtpggi+@iP2aGzfF
z-Q2X!d3>$m&ll0=_O@yHxe3KinyZA0>6#uLM4a%D(Wg&hV1ic0{yjm}ya#)^%`rdc
z`*aI;<O_#;8>0Cm59K_i^r{knT`4zZU9U7NEI&&A?sLPsp6kXr<rv1Y7UWWw1Vp}P
zQ>}0<zD})%(`Id40*3zdhBd@(V9lMF&{gz(QImOC+`{dAO$Y63xfbgk3B^~}-L1tX
zUvB?#k8=rG#iP{nzR0};l?UT6Kiab3sEed0CUO~L)RRZi$&zNA$+-^<(FfQ?47;Pq
zyR@-r@T?q!COu!?8Mw2%BtU3V6HQ+Jv<qKUCk>^2o1t>Kn!xOC^4F4wbpp2hwZ;2a
zcSPJmxX1RQAZ;RTW>1p8n$+Cqv}3Y;GI>;4Os?7pGe2v3)DB^x;dK))Cug9+T2<GE
z6A|{^u6&OfnIvV74!df-<T^o8TF~D#KTKRPJm}53svy5h$+|Mv26@FpcW*%Yrv`4J
zi(77InAcM2vuv8`p7{5?qoKF4Do@rhH#%|rTTU=HawDs2uUjk8OWFHa=5-RkT`GXr
zPxiHlD>pC51vloy(;!jyTD=b)c~N(3Uviz+>*_y#$!L=Zdwt=+@6avN%O&t?%@P$u
zVikW!L|)C4;rS1SeY1mp_-K-Y#EN~uDgBPb#`el3CV#-YKaHo)<nH)#s+cZq+(WcK
z6Sb1k<Y1%vkXXPP@1zyAKE^8NW;JdNYV;`#clYvT+vAD0(}w1^8jKs>`CNfKyU?!|
zXivHxNDbj<u6|H8$t!%3IA&@kbW4sj+K*pG@r5q*>_SWudw0LCYUQI|G;#!!@G3?)
z{dt!qdD@LXQ8JlC#IBy#L09|4s`)&Ya8SfuuK-GMMM6z<S-9ZZ8@KbRZETrwYkE~N
z6xI^fR49ekbSF`z0_oajmjd0BFf>iQAhi@z1CE&f`$r=gmrVp6f__^YEYYHez3=-=
zj9`(Znl$}tr<ij0VZP-@kN<E{RLPOutDbc~aSfBqCusFV%faS-vc2VJ6iFCF&IKn;
zYO1K7VI6R5%^HE_B)v=FCLuKQAfH_-G+b8sg^BtgA)y8h6$kYp1REMY$=-n`qbO8)
z<DxIt;_$E-3!SV*EyvsHxUb;~um|Y2*Wb{g9HjEY_!I}kwl7=NT2@JeA(77O8J;%W
zSq3B>KU)J21J@(4t|Gd=lzduEEE7_Od9}3DS}Va3jW=cQP$~=7pqqEyn|BHQjnFJw
z_R9__&(H&K`mEV58%S`}Tf!un(DwCAV<<)OrdF7()AD@{7I@&agpm_&H3lYJxdWrS
zV=M;a`6}OO#4CL@*-k@3HbDy$!0bq}1A<bO^n0#{S>jmSP$S2OEkojN6=n6ba@hRg
zCXV8wnwkoor#{8*lUDJ_&{Qm=s)Cxuwq+F#tl_FMHP|&4#t?0nB`E*x((G9LQo4?h
z3O_Sip0(=ZFPUYAwq3E28P?1;S87CZ1fb(O+#H3J)pIY<f==I|y#V!CS0crpq~gG1
z0IQ>FKVmVg17L>sb7P~9r}QL>hvVp5qVdSy-?rIWp36*1j%IEXAKm~U1kOB1Cx(E#
z-W1Yv)azGoUqBNHS}{LIiti-!88V?ctMFltpc^OUMJcNoK&XYs&&RIc_^0`hsE|Z*
zKp&5yCl}RmyBGxMx5&eOUP=ThBF;=kxW4ug^Rv@jC?j&MDY1SqJU-)+yHB1xMq)}b
zWr6r4l^c~pV9o}Qr1l@W?KPdYdy4Ar`z&=5mJ|oYEei1B$|df%{`!S4n-Q<3x$?4)
zwi~^~oQ#umu+Ay;_?X!ze#F28AbWV*9n$fyEalvo=A6c!7(V{O(?hj#^*yys@J&d_
z0)(f|KObY7Os`yH*U)9&Hf0-81zr~8zhJ<-J+UmH=>%H(m3uMzaL)~;sf7GPy^4bf
zxJb0jjPXVHTe)@mdAUGOTiNQMZ6_4g@^yI%TOF+e(_ln=EkV5<OQI^1quw}-R-x!2
zns7MMd^*4<>1I<cYd!M(A+Ut2mSww$#_N0J&K*6r;2=B(fG@P7w~SS~iUce7Zg;!x
zp7sb}FbH^3TmyE61E6T=xAF3C>3RP__NY=IuxcwGK0VAD!^%RMx#1{kk$Y^q-{pz^
zh-0daphNpLER1{KXiBvhF0Bk2^vawHUl?fan)GMAI*yFSyN!<rF<K?>#zF6gxLEIz
zM$!Oqh0Y{N(we556ZgL99tD3`ac8fQ<*HsrjQ=)F#|F!0jWrR<uX)quIxO)E{*p$I
z2X6OJpr^FEzd>7fd<efY;)yy>4FSgNDi6F|1IKE+;ERj6Ty2EpV>E7sN?II6-UFJ>
z*KhfsRbDbJjmbWzZxy%NDz4SrSIwn5?76SHHGFW^_&wz{6}sbe|4Y|7KIpbq^b{;M
zh}B3p-}ZcjRzI3B_FA9&qj9KhjZdwD*8ATi$Sf<Aulg#gE+kF?V1Kcp?ccNxzC)fQ
zQsf}CkzzBD(fwwd(^`qqYnEzFCd#1MJyxpU5{LdWS5(3~O5$Xqyfc?fy5W&}C6?T+
zZ-b7bL)WJZdsk{HyI(c*oo_6@aI1etuZJ3v=~(gf+|L2<2NpNR=kci63`@%nLquG@
zR0WMU7h52IEz>WBK|Q`{@Al;RN>%m?WO@C???ZAiG2#~%p_G@O%hE!}!WijC)X<8a
zn4;=#-r6a)7S15W*!&4_HPnz<1oK5TxUyu-m0lGO8FJUk?bl)eez3#^I{eoFW;^pK
zbdxO~y1qh;q3DjM7=5k-=q|Inje9f~Gf!a@i9|Iee_{{Nq;k9EM34{yjLQYHB;VQt
zB++L{mZtIJ;yyCEXet4SWQ`ST!+twXBS?IWqysKOG_j2xmRX)cWjS)<%^HD$TqRBa
zqo!_ljGSg0A{36X57egkcOx^bMh7x(6A}J+gtDD1y#O+^=;r)p-CKlhlmHy0&_s3c
zW=Rhc@h=+7G{IYWMeDk|&j<_dB9jc&&U>&!B9jcGVsuS%)J1hCxntI3+|bIl-%PRb
zcYo!0{=D)%8|o){2Ar%dKD{Sem<a_a$oz!n$;?KchNM0qJNDxa3I|U6458r$D+^w7
zm21GIcG^P=`?25EN{F6&lfD;^2@qh%>>%TG@}pBg$?38m98g%g>5n5u<09jKVLGKc
zoFT@ZaqtR>;}41PSvry;y;k^VD+xvZSI(X8Ul~e}Qp1rbzeNL1#)q(6sWjGq#dH=_
zKyyD()?>?4=SD`q+0OKzz~1^e_9ixnI8+rf*-G-x_~ffWyzakRCM?l3^FE`0vEqm}
zOwCQv)>ESpdHMV$3WIz!B=zew&x%zr`=9mtpYLS_-%GX8OT%du*u&Fgxin4&;T~vC
zDN8PbaLUR2K3yN-M^;Ad@Fb8bx9)yYKa-cmE}V)K&z^?aX(_`&xjENZ3-L)$XY*)A
zSet~$fB|;jf5U<BhXo4N*Z>yaN>2=aj2W@W&1gm6-A{qY<O`Lxw`lIUNK0L4|868l
zCbW2hB!eSo{Xss4CWtAyxO7~AUR4BTyYd8(l>#x!CMZ11AOCUdDIyRo(M0|59g;_j
z01Ju7V(``pT_0edbOgWgL>YZ%ddFeq6EV@B<~THE#P#mz_<{pbA3#46hZGD0fW45W
z;(d!HCjmIl{@ro9C{C>n^&L!*#uhNdqEs=6xpQKE4hSXG#9Rq#|9Y#}#3O|=$C>is
zOC-ujD}CU=>GqAJA!^A3n3h>`7lOT0qEV`6b925v0nwy*D~>3dm83-D&WnWd&~uC}
z1_~r-D?4}sWjgml$9j!S*RG9UdSsY>#0NZ~+(OJqtz+8HCjxV;pwKTwT4LcQpB3%T
zIf^zBz_P)dNNX^k`M1*kI_rs(nF6;;?fWlr@53xRA`0kF;xV2Dr6J#g@u)W^F<t`@
z=|>L9IVKcSA_G)2kvGXv`v)cnoX{-A>!KwDU*NvX*xDb5J6%drF+@0^A%pGrO9Jmn
z)fG+EM$Keu0h<e^$kp_~>Y>Mo6h=MZzZUH-f5PNrLFR7(LT)?x_8g#B@Dzi2DyZyQ
zkqx!%9P_jbi4a_X@AM;G!_$m|iHx~`1)D{N29%B1DF7!1MY&->po@YI6*(sDfmlm!
zTVRLwBiJR$^180ibdRVI|C_1x?GK<2@hRW>9)H!a$qpiOxV}8amm{P8ISW76vt>&K
zd{99nul>cg=$jaaI?%w2k(24(Fs_qEO8e)<!4Uk`&oY@nW*ji7oX6F60o4lYb=O|w
zo_NpAt!gr5jrP`#=_7EAD^jE)u@p!=3y_lPTk`B);11Z5kize!E;<~l0`gp*0g-}-
zInf!wzICi5PpS%S52TG$A~J_rUI18^>t)s;gO(?VVl{9&X&h5;*by8QA2r#L^{t(a
z7OGLwa&xgi3rv@&)!`@N+pnJJEdlXR5<4D`mH^NaW@K2|zXdxUA=t^b;nkQ<tjG|~
zey$cTP0D5j|4W-fJL$0xVD8R!-PYd*2YbuhWs}(QW=9V`O=N48hJ^hUrTE_+m_&p%
zAhe#gd$tne3Y9WTv+r)V#H==Gi-qN9czpC+i4vFvR(kb?mMN(#f|v!aDN0rU(VOU%
z)Jwzgzkcg8#K!wdUugF*D-5elJWZWopQErk+ZZtJ9MbmvXT<d$%r!VW0^h`ec<lM;
zpWXVOpD758@?b)$LImcs<knlabr(y(s<&Ip^xtbNY6MA4s_wVipO3~=_Pzl=YuQ6g
zhJL`@DqWMVep77sHxA)PWuRd`GCcRA9}z|Srv19OOdq@FZVau&d~wb6D^5{upPO@Y
z1QkvYc3CBlCvmmkzO&u-{s}9N;NRKAS#6jA8EHVC_aj_Vo=DSUBX^ED)oK|XpN0y;
zi4ZzPcMd!%F?Rq>`VHo6{`iBBzv>x@pAACL|M!=_FdcwXr<(-6r$aebFc?VU*+qcG
zz8DWDOwI(JQd(EV>VWu%J-Ru%v<RAR^EX6dfecXe<A!;x3iPU6Xp>hifZn`2W82~^
z=6T|TbFa=If;a@he7w0iZJAJXBE|pf)sYE*IH9~05kp5Tn&cdj0mC#x)!|wqwi^Ph
zT5JEj^FNF6uMb%?n7ui<G!9bs1pdNI`@pO>*spX4O&)Je4ClT(UpzPj;wE_V$}(mE
zfDd=-U!ML=B*2vr#7vtCX+lE(3kE-d{>gG<nNGoJA#02BLAwQ@s$67mKvjU%#C5};
z`i4K8$lpIV5me~oIO6aS%yBwopmqU<irTZBqv6MU8w4|#upb~egNQ@8d!OBX+4$;~
zaqcz($?(5YDS3_HR^A*w<RS8JxWf&|)B+Bxy=8>E06+NMEW`afK`GdJvxC6@Mo_VB
zrY-ZGF=*Tj+O~4zLVNO|rgOuGj1XZ=`(>4XkTj1KfRuR*l6NVh2`9G(ViCU)=m#v;
z-|VaMAq2U$@p89ujxL1i=^292S!}7)0{-9)UC3#gUtqax1~*?<N$cN4_bww4lAuMW
zRLln=$fA$)QeJ&D7=em;RQ*PSF17`9bk}<BRBJqmAGVYS0Z0a5g25^It&%8(oh;mF
zD1XiaBxzft8EETttQ@$n$bGh`#qh_Z|5qCQJ??<3h^@X^zJucY$N&LxMDVFRuMT?V
z05fF?fl}K)FUW~@h~TJrG-woAi=9tR?YUjP)R}`Xe%6Yh`YySH7(;QtfKD92ncSMG
zEfql&C|st<K0^M}6!PDNkR%7%mJJnlkJO(n49T<13Rpx0`qII;JJQ~9Sge+k+rUo;
z&;(yKScR-F7gn51zpf2S6)aaPefYcR<u5`397)kTh3|S<X@G?~mR|&LYM1S?H&QZ}
z@5l9TSmplMt^fQyMj%Ho`tZB{K^;d(d6qwxll5X9`{sZC{a-$)YJfX0{IsoV^7p;*
z`<5t@5h^*&GraMyV)sV`C0`<hucSgI_U{Yur*7E<;kr5OvhDumlD~@;!JjxI?z^E6
zPxk-5?^p!Vxwk&Q?(Z-A_qE4KK^(SuSA>N8?`q&Tno*7!tUtqg#`3*?8})x*6>%&7
z|Dpf6*c8c<wQ|vL#0yjXg(iM;W_${4QKO2)brr!B6HK$cxMA~SiRbBV&>r)XlvTj+
z6*>jTcCzMFmubs61d|gT0#*ILz~&hU=IwBOPF<KJF~J#vbq(U5MZmtd5wMx9d;1;$
z$GZ`%PEH8t)jo*+>Oqmh*Cw#(N_1*%+IlO1x5kHji%IIQ8^u6@E-P0?T|t>Woqpso
z8en$P)B@mlBcRd>!*$MnZ*@e8+P)Fb#&q?k-(efbsZf?Adu$fYA#6TO-^ac2;lGM!
z|0|)Yr-&VK$?+6THme*<Nnp+i;oEY;H2{5Vk;eN-?G&)C$%rO$6>kdwj(CVfa~eRX
z2l$(Hlbdts=KO=Cueyi$zEi5FR}ZtWFe3q&JPO!pgoQycyy;|t-=5mBNCX^@D+IfK
zhr8rxZ$Sdy-Llp6_dPc*f)IMK5E2PO3ltQ9cWy>d^Jf*<i&ueL%{P%FkADgRf(8Uc
z`*S?Y?Wbpev5BzSd{9lz`+@Hw{-JCr5Kc){f=>7+%m=Lg2v>CJYsH8(PW==9sm{Us
z{TznAFhECb!Q&z$h7kduP2C2eOcn3dfAB8;Q|k0TE`S9TvEh|ou)U`rNkS8R9Ga7U
z@B+a4Vsw13k8v)`bir>n%bj6#hHf=p0AgksMo7aYY+b<fQ>XkdKxTiobZ}ql9$2au
zvRU0|a#0j(00CUNIC#H@$}vy_{v7C53Sj6S30O^*5uhsHS-uA4Jd}|f(`dEb<T;Cf
zvpKI2;~i2Ao@aqomRke|`4AdT0OPhWWS9jRIZ>+90X{?xXB~-5KLSdMm=a+(P&xOn
zTCV?nKt{v^KDA*hp&v0oi(|C63cLjIS)DAI7UTAyG!dfjH1eL6izmph1RG~)Ne5=<
zf-PeM2!xhp33nU*D`kJw(LOs^&}T$oH@=cvL^w<R0nH$*QD$NR<QZ=`1;Bgf{E$)G
z53^$=ZYITm)v^IZ?In`;01K;3x%7=go}}!rLLym=*ln-dvGvjN3{c!E?4}7)XN+_W
zwcdUPsT4xE##y9BKfsxOZvbO0CVNbQhfv;sbocLC5?3}$N2cpVENDQnU`3(zK8o!y
zHU(HR%wRGYc5+d9H$_{)eVbW9*NK;|Li{bJ46|V4ia~X^g`582Au$kwj0D3PZ22w<
zuAE<#ose_rwU)>6=2!X?w~`{j{;>I2nCFlqAL9b*66Cq7jsU8=%lYDe754AXQHTj9
za($d#&(SDCTY+L)<*u@;`iX+Wp(<$4s@x8C(+UDBzW;T9siPx<v!An8*DGxP@8&~N
z24S?aD->f%{k!S>{7qMTV|U-_``9B+(^CQPSzFczaC_|$g9sKXi!6XQj+^?ura|?e
z9Qy!-?<aL%;R}hs9DS>$3<3fySd#39aKr%xy3Kypkptnqrdm}zJ_~1h=vHY5ul(m8
zMDVwxQXIo;F~`~b!M?>{=m+4*ExAZ2q0#3XcHod0aMyguX_|AH2`U9M5J2JHM&`e&
zW&g)<pa_?|EzvRS5MH1CzO)xC!#E`fqM(}l7Gb%FX$pn8Fl@^C?uC^N2LvzLX08Fa
zmSox%*=OcWe+TFL2O|c}$e6yPXaBos{5GJ&0H|M(;V5ta2M+LG&+CujXGLt2XD)0N
ze;4@Q3wex~mf<l(-Sy<JD*ey1WMv})64j|(#3TII#rpT>abAQD6mlgIAT01}WiN#<
zyUbV%xB}KmL%DI!3xYQ`|MW!)62Kcyj2rW&S2afiBy&aA&BFW!pneBl0#|bvMB1*z
zEXqZdns<f&e#0ZPXbpU>hY7-j7B(o4<LZ8okGS}e(-yD>631MOZop5s2RMG1Hq!fl
zsQwXTp!zqEdOZR$$HQAeEVDBtxB~f2@CI+c9`Hd2qbLQ2t^o@g;ha}{6t(~9`1;@X
z*-MwTu}j5$p^dhg4rXr9055fMuwo@zq=8N*xDI%;UpRpuOt%SwCr`Y)K7V_F!AFub
zoP$Z96vjxT48T*n0OMM%1In8bb&KnWcnUbuls%_^(kcI=>InRY5aHaq-GsVq)?Kvk
zAR*tA`48Cz$0?W=8>-%`0Y-HHm!OQGfj40JM(MYX`=`^f82}dMI2J!V=>J&hP~a|r
z{e^X3?eYJ<kN<owz(<*e)QTTUl@s^}Gb2KuQ?Qba`ab8m%o|CsJ{Ha*g6Rh5PuB6W
z!jA&3nI12jB^jdk{LDwAyUz4Fd<>}h8T?wzw7}U}dHV~{NHt_kZrNu5=6*Iu8;rOu
zFWp6ZqWaT!K?-_e$~$rPsj}`lv-T+I%9*eI2GmJD$w?k(y<%m@3l&AAM&^J<B55}m
zkn?3B^w%afnC+{31%l?%M=DSOn>0iegJ9LaDN!3d`}X2>p_vjX?02BLsJQ`b|7;YL
z=RLFi_E^4(V4nv_bd)|mD}^71H~zZ0oCCELYO0Y>EQNOKR%Ssxu1@^=H^nSsW+giW
zH{KJFm=HrfRsf)89Z&3NL^+1$zysvsWZ-~!zYBZmqUE|8EAQd~50~{Dy2uWpaR8%f
zl-fq5wUz-F{+lDERhfUPWnOalt?Y$|Sx-<9su)quq55ioTZudsG4Q8m#%2qQ0$4DE
zHh`%n>8-$4{kT`ZN^}bBW$lkf^JG)9T2K+$*zA4D??FV#e1n*31eF)N+V7wnI&uY4
zTApg(O-{*-8*Z^ecn0F!IZ(^ArRRI?3@mSd{nr;doROr-cC5181c>15_t1zR6o{Zr
z$29SIdsxZK1}8SA(DxI-x{<^n5Z0_gNZ1E&>11X>Nt${em}JLm(27(GGGUF{<l3PA
zh(W+@-{2JFeMCly`hk(A^0jI>1!Vt^m&3j{#1IMq1~OBfg>pOB0C>dDQAI9X^elmS
zw~p)m4t#`_b+)yj%K+&1ltutFpX-ff&`Ed)w)_HOauR3r7zQEhJ0v~!@gR8`_fA7U
z6aFC+Fs;dTtEe*CdmMOEoX8{}5OV~^1HdqxMp?1ekvd@61+1sHcnpucmAJVxB=^m<
zfY1TbV$Z;<_kv&0MS**@3%JjAjsrX)I!0kk?SbF$M79@yr%S~`c}7QL?wh5WrC7F$
z0o2?_0NmQ4Dg_8-i`$&!+S5TQ0>IQs`e}^=ba}7X*<yg98CxD=kdh$!FKf=0wz(r<
z;c&wFaLwhh11gA>1>0T%^879WeGQp}X}>C5RG5sl-zYHW!5QLc1<?5wkWHn4aJXr2
zgGaCdvyV0lEUG;4Dh~x_?E(;!0s@vI+9;{#z!a;Q!h=_&U(&b=aGdttq+SB{;co6b
zJ#3A37oc8VyXL3=)bth$A~g68hA${Pf(X9$a$Vw=)+4cQIAWH;z81huYhiKEfYCTO
zjbH}U+#G=M?an$5b227gd|~ZZ!ilx15P$}r$A<#k_|fLqe#CSML#rTjVd6Z*ASAag
z2I{>8qYDfjB1AH67p!q*;~Gqdir5q3Cwn4NhakPP>oV`Y?y`Xll;!<XkW_Z!Mvf$9
z{n!iA8X`sUGK6M%J`tQXm@k$Drz(epU)wFcv?}jfF4bvbncx2TMxYg>5{V)Aitl~=
zavRcn)`3gj)nyi%990;+Wf1j#uEYRQdc-Z+6*oMv2o@50h*kR;Y@hs>hPCTv!oui7
zPA&;pvn{@tGYi7-6beJE+m&u^dOW=CPdRw2xGSZvswEUX6+(7{kgT7A6O#@r9_#N}
zQUUD);w^?2v2U=j9#;})(HKnTWtyGe?{*h$dV7mZ^G@B3fCy@?3R<2MijvV%Q+HiN
z?i2@ddq+^%0thNb(?P(qbO|pP4JKoLo}87%Sr1~XIRx)A7?+Z)hB)hq5D=D*8?MAM
zSavSas|I<*5GzGesNEfqf6<#vDq?uA07mPY4U3&RA(ESu>m3=FU~Pm>B&F2unc9W}
z_qP*d@11PG+y;D_dt}P;@QQE5sEJsxeKV310wLhKQ}2F*YT*g361aj4KSoT15$u<C
zr5i}I1jI%$DS^KN6V_Zi3FYg=eOxMv-FO^>XYQLuJy^=^^my6#)EslS;NboPGre3S
zS2^Wpr^(-)dmLo=`)CC147hEcowDVI$O#!2(n@RXfuk5>U+gf<@wnFfOdkqnnaK}n
z9<2_1eR}3t@k$#Wemqa<QKFQPn11<V(AY&ji3KN~)rBx49>Sm@zx6{vC*FLa?|odB
z>{;*L-AQK5!Hji~Wj@>Mf(}*GL?7;?UuhF9k8+yO^V~6TYEujr8=1t}X~*G$3>=)~
z6|-xAJkYzOlH2~m^nUaq@+jE6rQ5O69exPG-Yx6Eb9_McK->zu7Q|RzIZo0X%10I0
zgQ6FKn&J}0=)mVKu`&tw)YIX;0B3Zx{oizKsbP8~D4Tu)tY_s8k@f{sfzdnQ7$n5l
zmMoHd^)r~YGPf~L4}G+6l5K3FMU1_zCQnWJtsfSLk?DKmu8AEa4><%Zxj|0QxS2bv
z9sgSM=A;dSmZZH~^Wi$Dr4>pYvAVAvrx_VUMGXE(SjOU$#okT=KFQqE$hdSu4YA64
z%(BNq$A0{X7zWEZyAs0b=4^P@`6E=7Hu=o5UvNQJ=>7Nu24qp*A<>@$cANC6jTMo(
zT7i!T5zTKL6NU~D8FZSjxi&_rUH+crr|E9XVTfTH*m|2k<dtSCiV8Sa*saksIDwF;
zfa3$xZK}97z);V?&FB}qu6q6A;c+^!lg0$CeNm}YWIC{!EKxA;O6m;kf65aYq5W3W
zfamaUh3!$Uz~OfSQ_|cc<PFgpKtf$QXJR=d4d2S5`ZIH?Q-J_VA9nw9XcZkaU-<Q6
zQb6vUz9T)}rLZWNU)AOqas=jwyu*%VT?S{(=JjnpGDdw|idM=(YY7!3n%`E9KD4d)
zNx+J#^E+&~w!y?fG*s3l?wR&Q{S5S--0AvLM<=#QTWC3(F=Sf!=R@30-2A}gG7MF7
za&Hbi4{INkanxz;$mns@i)#flk597Uxy^eZ1wtnn$uWK_4D*37awKZn_s3$!x}AY$
zm6*=3&riA)KExcb8t0y1)0{x9lB|rWLJgp1ky;-SCmL<+!B|FtVQ|EdmLO8p+~5x9
z`&9`tiWA|I>4XMsC@>v3GPhYI1z(?i2K<_5rgMYSjb|I!alRNaTq+Tty^8*Ooak3>
ztOn#}gCEH6hs&aWy3boly`hHk>DR=3Q@6J3b7#~AR<Dr)9%*=LEj^*^TclXOpeiWo
zTca7j(8ph4Pm8ibr3s|UIJ+ozt7h49i3o~Fkh@^2xlC`~)1y$V;NGQ>xEuUTmUu$M
z=jYDkp6Hg=<w=_Nt})~K)Uc3easBs;VV)lEAvU|j)Y>WQD?v6sx0$0F{O->Qy>j=o
z8yl=5YtGYijs?q-lWX(75sqz{F`CL%8PeQ&?QF{Q;<<|TPwI=B{W^sJ4pS8jx4F}e
znUB9{Y0x-&NK8Jj1t}!%*~fTNX<|Ib!RBV>>5S2_6I>`3$HzyZl8}O>VJO@n`Y>oU
zF*r*NelaFvV5E!oHjkUr*)?E}2z=L?9=O{iP@E??XTI0tP2}VKUeftT4a}qCbD@+M
zIC}p&GokA^IPFiwUw(sG^42!g+cM{+wY4;<#S~g`Lk)qT38vkw$#bE^Pd+Z552qSR
zm0Pt+tx?A^Oxn`0@9f97Mc!N<{Nk+4(tLY5`<f*&n^`8!gnV2|XmcvW+UF_W-82WG
z2r(tJ_?lR$$tM!W{S>Ru>R@&0z?@LB>0<B8&|ZvqO}#T%kP>N){IKdMCDZNu<%_H&
z!AOcNJjFzp*BItIGpvGVhJIU2gRJBPd3u;o(0%wm{&wz~#Q`Iw_or7vtQAV}FJn8&
z?%xi__Rk6T%YE>%bxPJ!zF>6nwTN5GeuEfZMwX)ECsh7ag%j)P*&4b!pvKl~K6Fli
zR>-;&_RWY~Y4?W@bdEu0?X}{}gbm}pA~9`#sZS-GXxRSM1aIevp1gSM2sY-x9^ru7
z3gyCNpNEENnqVKlMYTr6NdoE~lNSFthBlRAQ}jqa+Vgg9@mu$wMXJ*1->m7+e+&OO
zVE2GW{?h5LycN-jXUnN)F~uYPxk9BfV$*yyR-$1~Fb!m$q{8_p7{>ICP1U}4mVGgZ
zV^F(V@2m62Sa`gB4%AcwZ8YO&+IrPs*I{~stv-nOtTXRuK~pJakZ-hxJ-&@*VW0{L
z$ov$|*C?*+lxq=8;n#r`mbKZ&ojQS`(OMyVttd(nnEdmJ_iM^a48n?bYJD$jYx`)o
zat?0YYL0~7e7nV>HPm*d(r$iX^>dSFid$QDxB>~9F^1l=P@Z>Dgr@OpC&kvf!o9<8
z1Iw($vx9PeM0w&Vp<xV-aiK>Mt)hJTjQ0W&^3EhLMji{B0h)QzM-nIOw%2+;Q{1`n
zOMt)C5iJ?gLT>I!ll)8l!xI*|f?Wo#UTV@%s^_o!Z%L#h)hOs`8ksXZ-Rg;5PjU3_
zR^LC!OSMTESQet&<%w}0cy@c;*I08p0-bi2qW@w)x`}&${VHhKN>8xlsv~tiN;C@<
z?L70ODu3`E7|AK5(vED)qg%c6vdG_6DW8FFBDU#g^f%uV5@cgp{8$Z>9OJFarUAh~
zg`b}9bEw`mJbLkpcyE;LNM{JuzdsG*3`aj-cXMX=S1N|v{dNEKj}Nxi{j=UOLETdf
z=0CkuOsBp6A-|r<EHQ}MPvt;UK;KdNy=?{9wR2IZNL$7_Ne7sLHTXnXQ_pne9FeIj
zAzqR!1;dkZo+VE+w0Gz8;k)2rQOmWAd4AN4>bl*$AM`?Mka?Np(?bJ!j!N<kXBP~g
z?nV@fH=)yaLTFCwdaC2C^4x}#%5fu;FFCzq?^FoT&9JX!n@<Kl=c_eee7rE@ITT!M
zJX+!&lsU+D7|$7Nl8SWVw7`Na`}}tQU8x`3!}%SPidMw@pZfG2m7b2LZOjasx6g=u
zhu44P{Mpqwzb<|%v__+7HHmds!YXOUTmO1fLm>65=zW&R%woJBn0t{wI>AxzHc9Gu
z<q$Yc+#c%A?7l@?#@iHuYcn;lQr}Quco>mBn8!zUVYS_KIZeW)<pd$b45n$Po36|x
zc+^8s?U8Fu?}yQmO-bLO5Vw1V#bSfmVVxQ?^IYeGBD#a1z4my_94e|f6B*e-(Dy2I
z5FQ-bU3)1Vi&xIoOAsx431KL_7DJH@a6L(L{!E2qZiOk!7dR#sNJJ%_jAvL>P3ud!
z@{zV&^hbjHkJsB>_X9gbeOtK2LsUicbyIzNPjPOE1eSaDRZ7?6XH$i6GY3xwx?{4G
z9+&SCQ{~*Zz%=Jfojced&$bKUCi{Lp3gdsLX2Kk7=Yes?+<on6n7FW(-yAq5FvU^S
zVL?fMHh!^`p;~556_sm0c(Rhzaxx!}J9jQW8Qiw8`{f-75YA<|csmLxmmhxm+?>iq
zD41+Tqen<ZZsfQ@N8<fjf1$n5R>E6?!GQN+*e`gxoeyQH@?a0<D6Sy~|7^!w%Jz}W
zZ5?R24D{(w{+LVOeQPV``7-Y4&#2=0Zjgn*75-KarS2gwEE(I-T$5r>(d5JSS(lcY
z4mD_dtma(?hF-xyL*YGjG^AAck4(zfqam6K4}`cf&pKI7-tC^XoZpr@pi{o4*pGD`
zJkQD$BO`>T_4DnWO%40Fgi3yf>xXHfQHB8TYbW4!?;wzmder))h11OPn&y|aIqJv#
z{&bSjBQj@|w4TN_7BnOIRwF#*zc9lgOJ02Q<8q|}mtdkt)1LPpWqUoP0PnB4wD%El
zzCS1rYx5_2D?YE-FATO|njIl-#hsw7i|!4L4BGSQORf}*a{PHOnvtJ`N#sUT?Qs36
zmV!5w?dLXE)_C%&E}@<6Y!e$gXtOcnqL@7%p{UUBdqizt6J5sfTHjp(klYkCec4y1
z^l~N7(CFPBm~I$jr^#!=Bue_QQ73P-htu9ddD92ARG;j6LncK`01^t5eEw&~DWu%B
z=c)m&w=h7uLK}TekXY}SdcXAGag5KlG40Hk5`Fqb{)c`RJML!li>{8Z!Y3yQ4X9{t
z_CsogzHOz>`yBaRm?`bXEDjO)hpW%?p*)azmOEhkB?J?O7kg*UzW7G=PSG(>w7#uz
zZ5WlIz#feLTyUZ|R##M3yyz9}f-I8Fjb1s+_m4U*UR){9oCXM5_X*dmqgRY+FDDIW
zBuH4-KNS#Y1|fCz6YM44mX)!vXBNN`E%Nwg(;Ne7CrCtcl<F&3g>cY2M#j<Eu~{l)
z#6|aB?$^{d6|*^XKYKXl%BmgiME!n%A(iJLEc^H~ui!UuwsK6MwU{>|c-L4r=k5|o
zN9#{`rh95%Mm(+BdC<!PABbjKvJ;xnkF*l204V~kIRC@d0s9l<`U!&MGynVjThM~o
zwq&Z;Z5-GlrC)u6ckUXs%EJnxw4On5V{fI`wpoNmT3ik#eOEDIV#f8A^K)Iq+zh!C
zDo$`TF->_ItpGE#FqbF=XLCW_Eq>T5jHN-T<y2$l9duC&$>+``uEottY7~3l<9XT~
z%hp&Q=DiBFQs>=12(R&?B}FS{YfCQ3vOFbc8Q$o@l$(Dpa?$BWnm}yuv&ieVRI?zN
z*GJEuL)-7T@FilcA!=j+=zaS5mP6jVEy_z#3Ad?gs3<i@2V0JHTE4#Va${8JSRdw&
zK|{a&xwqgRlzvN?HWiFsG!Io;i?-jNoqXmIa`4(JC)_eL@+$iE&)dqy2KmiD*vVy1
zbnUxHWu9d~V$*4tj~cr5Ki~ddWR7)54!z>D=n#YZ$WN7Lf>o~tbI)Q#+l<{f1C*u%
zaWgoDsW2D$;1NByBrg0){0_%OTZhx8GxjSMO`kbEn7{wMJ7qqy&yOE}fcCyvrc|w+
z_Q3>+rHUP|yPA%--;~&HRMWN#+Jk`r+B1JzO(^D~iQFy4EyW>Jk2Nr@)k+zDy5e_&
zC*C+Ca`Ayys7451l<N}u70H^Wb)1U(_5iB#tAz}|HnkKCPf`mWGPR{ICftw>!Ouu1
z1eA;)!*t%W959JYaTyzFd>K^x{H2@T`WABWB-e+it@w<i4`%(AIQZCPOth6EXc4#i
zj{UuOo4couyH($EDPzCl4}Be3C?u_6{_DqL!7Qa~=|TO65cl9(2RxG*wVgVdg)&Q{
z%qI2ew!L}%R^fQfjPn9{v5!nTKRnt#Ck(pbF*lVQ45>wZ-A10U#K_D`>mH7~?~Xr}
z*@k)TlKvWvGWLVsvw$igmVIMA7Mc>i+AFbR3m@#WJs6r_WcB)a<n>*FS5Mz^eXLtw
z8Te){P@+g`p>?~E!HGJLAYF-h{)A<TZtVFp%%XtS5b>jI|2L12)31855~hS9A7Z?N
zr40=3Qs$Q?Km+JuIs@xa`AlLk_=k{$kmK>nV;51<vG=Fne_(zyNrZIWQEyM)*gN+1
z59^F~h*(34M`!C&30tck_Bp9!!HwM}W>RGH;OY)2N^Ko2z3M9+Sh%HE&QU-v7#U_R
z{xf-y$Ej2y?}4~zyxN0qjNA59|7HPnSH5Sg7V@6rn&O_ckuEPX4$am2=^DXwZ}3bz
zA6K(qId?R~nIqp~DsR3_BYSbr@~rcI46HoO107ZVmSQt;NTSpzaSj7acrZX@I&m%K
zqntx#vll-KZ`#cRDAbnAbE-d>)jVCmI`Sg&DM$P(^yqDxoOBs>!CoP?owts0v8EhR
zY#o`XcUTMhx_-s`e)jPe3}fUy^W7cTZX489)v6AEs>hub&jKx`YVp~wVrY#`R49ou
zMUVf=QxbK!Rd4u70&;5ayI-6oM{%BpvQppsh=eiHl*)0PsFdgXEzOHffoP+e!pt-i
z<;a_5(`W~(O7Vx5H_=F?erqan)j0L^S^0NyV`5cMG{W%nTbpD<@R3QGwPJ(z&<@Eq
z`kGI_`!CY;@bx6y*thgWXP^l_b6#^EcI)P)@I9Vb{iWX1@WOZh(v7CmAt;YQ@KH&J
z@9N}xpD~<y_4l2}t9l#()<bR*_>3YIPbSJ62RXgD@q2xI;|cRqU#^7YnBN;=m{(6s
z?ZX=K;cD@2DQMtX_{HtLE1MN0Ce>U<XCTjv#U1b4B*uue>?um%d6|w9!NU|UcvQkG
zj)MPD=Yr?>7mMf;mSeiS*+&;yl!fj4#x_@INXF+Qx-pgR`<3>tPtuyFOjvG+2@Ynj
zVz)OhC3edmK1IJbOMFRrFAa6Yb0bl_RXKm`(khp6c*rmQ<lv=#ai84VPC^bsuehXm
z0~H;%%fN*>77Hv-e|5*M+3fzjxI+|kWAC%r7%#@gnq!rU5bwlEMG}q<pO2_-Uf2!r
z4{SIWo>*XwM^gr=%|`R#=Gdv|aiY!n+aT4ka6OOOB}d`sucr0m8~SF;tW`_pA5tU!
zAVKfuTvELFXZ3=7ocDA3_|%r&Gf2?fw~yMc;q92O0WiCRD0n6Mbum6JYWB_n0rE(V
z;iu~IrK8=td%jq0bKhRe*I4<aJTk!n2<7VIs_!mOS*vU0!;dRj6C_1OOInyqj|Wsk
z&tL0seOid<b<?b*e9pZ?LhQK~h+^%~<vP!lR#y`|xkspJu{G3j{_4rJ`4SVC`{%3P
zLDiy4Oq8@YF~<$lkKgFCo?JN>tMs4*t8usRN5&^dC1|LzF1T132@Y64c`&;R*ax*u
zHrxM?rL$m*>igO-ATc1)Dc#*lNJ~k#h_o;u0)mpm07G{-f`D{)3k;z{HwZ&_cMLte
z^ZQ@#Cpg#MXPv#*S?hW3Wi**|-P4Om2K3b&frtyC^*VFK6}kv5VzSJYuhlVvTHDYt
zk%fPpl6<^@+D?m7Cymv<M?RkSw|_m69JGN=?9yds^JlXO#$Qz07A|_8lCr)|j};2O
zTzCm->-upa_>x9oja+tZyeTQ6;VHGb9f(b8O&QQ~Yn%8_C9~XXN?Tm$wuST}V&y0(
z8N5HI{kK!nN~8KuSI56I6-1*M4ZOPvwq*VSmS5bzJ~A5q!efBt!@}?NG{|k5Hr#<T
z`cD&bsS^VLdAV^I2d?CR^hqvHAT)%puf%A(U@Lcm?AM>iWM17Y`qbnkTs`ToHE(<e
zQzS5O>7P=A?6ui_n+1}jNrrE;WqZ@VKG4XgG!g2Gzp12yhCQ3Qd`7QG+yXxvzS~+Q
z?gz!wd|0>eQHrf#{w_ef1hc668>AaD<2C8CA7Lh!ssvQa#3A-~yqAf-72OGtCMC-F
zH>NH{7jR-MzUJhN^?g(-Vk^g~I+=@ZmmWM25;TW66(sEW_Vps3Zz`!n&RbF*(m<_c
zx6sFwNO<vdV3IiXG)L^FU=Y(J&A6!I9C9HkqJ7yrrkaCeqsN;4V>Fn-%wAjQzT~n=
zixqbssSWGRekt8O;m7M%(eZ{Ti=}o7mdckx)}7dtMwa0ubyoR!O*>44e%@pOfUEJZ
z>)!t?SiFjGKBZpvD9ac6rw#rUAEkO%@Q8%p{?r!5Tvn>GQaSH<Pi*Y?R4vxAr+MUu
zNG4S|%R{o;C;65=-K}2(zHH5oEJx)U+&`k4Ely1YNRHworl7;d9%S7PO}3~k3#Uj_
z<ch~G#yxu=-_bmh0TP$H_@Ip_H?|=i6PW)CS*>PBsQ6MjE~LxVl)&)(w_nvlb4*SR
z0bPY>y%yoYj~L>9)r`<3*o%r5`TJzJE(-xhSW;~brJ~NI@)rA8O$PAmcGH0G{Wm_V
zb{eL4;(z_noJ7PkzX=H@X1R`~7he)Y0+wXPdscsGc!#~dS7utG9}UtRm|Sz?<3L2a
z#KRQ3cCb?hJcA#ZrEnpmDKQ^(ZI*3yr9C2b6UMs7RFkEEQc8jE_Q-Ct1Ej9}K-GS(
zMVLwLB0#<m%$$iQ=r|C&3PEP!{=65$fm56<p^GAaSMJyZ6Ks%|gS=0jhx)55|3RjP
z|9-?vMyDAP-OkMD`2%yX>np0(|6{oyBzHS})(9Ji#R6h=L6d$D$7W8JH1MMBVLaY7
zx=dw$Jc90=2_TEIZ>Tn}gnf5MZ9B2cr_W4dO}eGp=Rah+4x*W`$q$plB(#=e5^SWX
zlT{P;2A`U~ZIR`?68>sm0P>V-sv-!kocb4EYk4YU+9yx3lBtlT1?&9w$9j7yy<cb4
zl0-(0deF-b6V<u9;v@Q!CBYQ<>&AAlb(&!L6YoyZq*qr_JLud*MjhI(P5_uR63&vU
z?e;Jj9l2seehTMEptcLnB9kHs-@E^moeuRP)6(@#X`b0(lK1Zv3RMj}^Cgmal0mZe
zVseQ{SoxgU7#L}0zkijDbNS9$%;w&!2^G(;Izon@jWt-6$_cd&rxJhYD~`%~uT>W#
zWsUpJA*?-WKV*BgS$6R@4kX84Ec1#`Jen)Vi7Ka?yscSVIDyS3yc2+TdBI5Mnw~+}
z@o<y_AwfD1=l`f0U^^M2?-B9BA-uvA!^PMTB*0Hv5mlfNly_tkeGW$rwMx)c<qCV$
z8Z}3$)ie>RQcT@M!W6Y}5ApSGKCy)-cb0i}ZIFaYh3$=lhLbt$%NS1+B^&6Aao6nm
z_8i)iAl`X@)p^01S8*K{B)3Bpr%3zjRpZ&hEZCFiT}pw_ij>Og_!WkgU(vf<_Mq3D
z#iU@LqGR~cO0-g<XI|RrRGSSwLR2Jcwh*tOB5InrZM49uK@mPrynnUqX_(45D8AF<
zu%I2CxSN5$l)P$w#f##6msPbL%5^pmiT2uvDXJ<`e-c4ik9yOU^Y*j`60oT3$~`XH
zHIIS*x@ys)1<ZjubUfczbRPm4?SplbxwZ1}r)QHC#jl+tD*ewh^Eg!;)r4cz9#}J*
z`Z<TAo0@FMDN!~eISYxmrB`a!Wde7LH2yAw`0C~-9n`$~0r{^KP1SL=C89UvqWt)T
zWzI7mIatzIGSwCyz4$~l9vdH@sB2*X=-Gedj2i0z4$7lmv=_?{m)X7{FTw%>o~bCu
zF#ebd#Z-(=Vf<KO^;OxZNeF2}Bt&ElQo4_<-6+8K`LXG1+4XB|HR8nP{e%}UwaYS;
z!^gww5Vk_Y+IjK<&pIhq19I~1NmhofJdtYsKs<dji??%Fig{!BQR!mk^`**~-8NnJ
z@sln8Ro@$t$TKW+;s(1LSYnH9y2u?+ZwmJ^T4;^op%g=zh@e_BCh|psauc|pEyc91
zW55nWDm&(~gAh6JfS9rE%;%6u^Zo6z?o(tg=sYK=6!si_1Aaie(K#yq&GA%YS=}h(
z*^qFLwPfTl$-{pYvG5A-#FG(CSD2Iz^<WiqbH7J%N6AOig8S2)*8NwtSvVMhT08sh
zzi!hm7>}@VDaVwzwXuyK2%2Sd9Oz4w$)fibu6!>3`Y3Yv(#N+3V%vLCJ?~E7qT_Q}
zGKzkMcJiPiS(cZ0>d&eut9<)&EL%2Ob)K@bSKU5;l^79rrm45}a>(uDmcdxLFIJzh
ziKe96S8V5pQJ5)kszvKG^Ji`z7D%=j61$Uud^7XTozVQEGf`thVisDCm~#s(znS$Z
z2x3jaB#F_)Ha7hoJ-5(d)}3nniehLUzWG}wJUM{E)x+%51|p!U;;Y#X5*4;VdS<S?
zL^Xc`Cl;Rnb?Qv7Ycq(?Y$`YKsAg%)L~vCzNSfbXYsz-8SciR5`&;^=uV2`o0t;8v
zceH3$#NWFRhIaVVzf(F^1GNo5Pdy$DJWvIifo^o|mpt_kk{ctsCHCUNbp|2PBE%r+
zVve6wPT!k~s{ST<$voq;$3&>HESMs={KruQWnXSa&gJFYe$jJ~UvYn5qyh9!E(+<j
zv)n(>MB!^2&s!gPH))qZv7|Eb=A|AkkMJ@r3>Y~(_gldJ3vCRO!cX*2s^IKjZgeYx
z{-;5$s{qT4W~KIRKjrM<)WJ-jJ~y3HG8Kg)16Uf9y}d5>p7>z&0M~cz^Vk#Muy~hR
zNS`&=P1^UEH8vif$^M6g4)q@E0|UhBv+gL?*xcN!yjDAnGvnCC`=Xm@S3A?hQ?}N%
z8_ifY31$-x(vuIsA-1CVf%~(B;aUwDW_7CTO<X_WG0bgVt~MEN;3V~>Oe(onW2JW4
zh<DKR#EhFi0vuQ^N9d<Q`^!Bw|I%st>6#w>yTWUI5@gr2asZ*kwy3I0sCBa={c3A%
zJKYewPt<JEFbz5dx)ZrStvo>@YaGyZ!ymYumJ=9OYq7wKP_u_HQ;s~lklsP3`%Dk8
zU3aQW;I-`K{ej~qck-d^js%U4gL6)4J3zFf9OR;dTk>Z!S7tzbK-bVC+?7zQ&XJVF
zCs#p4&?y@8C*S=?#SY*7AEn=}JlnENHDn5h40&uravT!9L42P-IZ_hS@ACLxdD4Nx
z8R$5&<>}B<HKuVx<Ld)y%f9dtYnsU4V+bF_?~^|8kAFfkjS<-x7vDu{O@|$_K>O19
z>x2I7&6YWMOQf{f&qzOBU)l>)xA(Rg1f;qrlpVxMVI;HLWi$PzY$(C|^waLIt@DAY
zM?rN$tJSHXo;aql_i-UC(n;1LIcc8fb?j#H_*j?R8<xBN+MOHzykVU}W{;)0fqpO-
zJv^X2m8t9OM8q6asSOV|B`AQBT^kgIoki@BI)~DPfgBqUPTG$-wR1d_vF(wmN}-k(
zLTe!FeBt|)!DY$aC-%o4u~sS79AMJj@ZV|hO{U#OH>)+XPTQ`TX<rw`J5^?F*dLuS
zphD-AEVjhXtG>Y%tk<KlY-&-kf3N$~?Z>iI*Zl!=Hv}ufgEmIg=U!nK8)gOrF+yXa
zryt)N-_YQEC~|o~-lT0L1W>@f^6L&X#oiaM;<O91@Rj#j(clmiyI9^71;kaK4SR8E
zK^Qh(?<gRbT<;{Tc&UbcLt+@?x5NW~bv)=erOvR!m#2&U?PM8-jeLMVp)w%IIlJh)
zuXUI?f>qQz*+NR&f&Ln!)GY(6BGY}CEUv->i=1yv0juRQF6^9I|0+ZXcVD0r1*b`T
z+3>BoYyKNoT77<EqpfX2dOJTP)UAA<feYA@Vywobo&7f6n5MoXH6%6eH~Q?vnf{Ne
zm}a)CpG0Xb=!=Qy#I?Ya_HibTMD<!xp2Oe-<*PwI*JHMiTHFZ|mmlX39GMXJ+&?8?
zJUjw=oi6EYWT{wK-jaRdC^f*C6{#vM``YU|2X6cMg_s*#<6Nw<ge}!ONnq0Uwagxt
zPwG(q!w4;L1<B?wkvV`P?%mt2Am6Ugdb^1_<Y%4;nY|-CL@JvLtaw%*^msExYpu?f
z0~V8ycR!|UZX7u2csO-@XvA%W5;hb_*J#n83Mn{+EFLkqef@j&V+%f5gJJPMNZ1#A
zMw=EAm1J1dB=uLn{`SS*G=CoDbPa+oYyd=>_@Gw!Oc0Ykz<hb<_YZXf*|nT4^7cXv
z-hb851zc@7$!D+0`LY_gNYX9;unGN1O#zy`j90{s8&bHCe8^TCS>cRQqW;lt_fYzj
zjG_}`D%#vpT*W0U8lefs!pPC?rmxl6hKe4AzjpLS#GxYZM*7@yY{b{)A?B>{GlXU(
zsWhuw=0l)!-{!qrg|8jH#J}Xak(;_f90s9wBkGts?~aLL%@iSFy<;(cm!xm6tCLrS
zA5z-l&O8LpkKGhv*+Bfy#QvhNWV9+iy;LTBwRaHhH}p?~^SEyc@-aEk|G9fL62?<I
z+`s0-Yj_*=i#I;i#<UN+>2{cCjO{tTT(@~$cz!DQ{c3HtMg2&r_Q~83z3>n$;nMiM
z9S|;V*Aw@Ji+Wp0>QEZBH}(S=9ppt-VwB*%kH7!Bf~DTZqQb~wuhZQzyK^5?HWO!v
z*6f0h&XMy&QS%r28Wq$bHq@@(Q8OTADOTKmVJfV*o%5jmdqQgbDGSGIhEU>-QlPT0
zV83tHP$?*PP<V&N6l++U8|WgolEY6vw~&0QD-d>rR=(+J@mCe9wdeiF$N_%j6%=hO
zNX{OW6YU7NrsV5>jl8k2#9XuUHUDFAly`#zS$7w;OXT+J{#N(eX&fH`?(9k#tSim6
zK&3cjANHKkp!E{}Di#0rHApu&$y^3xo?SFuff-n>hKiA<&5vjIg?!DUkN+P}FAC+1
zIk)DY=bcvcC8bEKDc10kd{x<#jzvupS@~ygm#S~Gf^ADSAWzgAd6xcGf3<7uL&-i3
zPGj|?rHha#xKP9o=R$C0#*++@Cx;ePLAdU+OF^AK-%#t4DACHP)0KO`V@5UT>(8?A
z0Wn~ePrDWpO}Qmi<=-!Akiq&MA9FTt#U#hu+I7lLi1v<~CgW{7A5GjL1KYte>$PAu
zWkT&<ePWgV?X)Cyve*V^%$t*<eU(D=-Vv>b)-gPlF4Nt5Uj4D{#mi7yT43N=Nb55Q
zus>Q7{cQk+5B%C;v8s03_MB}arF*MWNN6(Ah>AOKh9jmw)UOq@qBu}bzeO-TDVj9M
z4L;{WFxBP3pi-X0f`k#B&>3ymf#T)L_qpybSNUdM!Hi*bIZyMuxGRHq!v4hB%<7m*
zr|AsmAuLhQQ)>M8)W~fOi=~i0su^qHI3QSL{F4@MB5(tIaQ&!&rVTvp(LY;ii*?R<
zWV6F`OhX0E3vG5yEaBZKnG#J($${r9^O=8Pzq1_2L@B@HDm6(UUlE=<+P0l_G@a&2
zBwTa}uldny!0~(+Xjf#sSo(#x=6>Kb!_GVKchjA>hgI2Hu4c+?j`KFVg`;=x8k9^`
zJZ_B@Z@MOh{K%MhiV{CuGG}rwu&<bjk-v(c>$jvaj;=1ctlTNJvC3p+XrUP6%M>`a
z=CQGO7+ogudZlv4PtV?G^)7Tn-q&oVswr`@`LNt+xKcBa;%;u}nMa!9thxTYWOi>V
zyg|qRC`v35hHyms?$=}R>6g%+L9MWrlVU8c1F8JdFH5jKA~2%9xcXvpvV%3}6c|M+
zDb3t;B@;eBzHb7vg(%w>prybqHcJj5)%~0}841A1$hG8dTxp92*=JbWZObjM`Z!Ox
z0<E~euIMjEL%BWy^rn;fLi&IGO76ILV+?XCqg`uY(?t$NAEKOcYywW$SUA3ig8N3D
z`A8cm6>0#q)$;hR?CE5&!Fe=?@5h~=R8tue(}c$P!23hNUFtQ}8hXhEe3n>u>?0~6
z{Fl@Wv4MC6_znH9gjgkTh5-B+QmXh89AvoI!ap7aG1Ja0mosx{Z<qbaeJHvzz65n=
z1W>em{Ny$n9t1Z6Q)#ELALuZA7J47GUD9-#-}T=44_-q{HOIvtch#HdjmR`p=W6-@
z=~-O2-KV@sPp4HA^yOEF1F!qqOtv$mY>%r%IO?@+qORXjZ~4s-*3ghr;}>s=1pHAF
z_VRIl?1TWuHGNlklSmFOb8~Pm$Ln&NKyosOA&{YV%rh_s-;GoResNmA@I{vlA4cMK
z!0o~AnNq}aColf6mK+?d;QCizn=cr*DnT`z@SY}r(#J5&P_2C_e{|^|ZOOh0{J0d3
zYKK8GrGeV10}9rRx#<BE=n18AQMZi!@dQ$x7b4`-S{>CV%T^imGq%|Grch4Y_21cZ
zc)o?<a*=55tA5zVmriP9*GIhfQ2$>*w+xN%eBeMN<BHK9KzO+^LuYt%j#<*v*h2KN
zYPQ_5lgq53LK9&zsyWM7_Wj3~{Qr-jjCaA$SRexZHcQE>*Le&x+Y@iQK2981S4R$K
z4I+l*=n917SFf>uI!@(jJY!l&1L~O%F`to*SLVw_SyvUmTOb3z1XT%*aRv+N{t@uq
zQ{U-k8jM+Mt9TRmZS5XkP0*6FgDc<iK?<s`*HJrMYGTtMEXBFOI`~Q(Isn}&7R5Aa
zmmKt~)SCnG@g|Bx5Bk}#4P;}ZEsV33&|hlhcj-D;NF&eUWdH77EVm}q$_Q;w{_5Jm
zczF}jV>|vlw?CDi=Gmy?lc|@zD{&k<#gNPy*y2|$r~Yoo(O4pZ3wCU3qc1HqiCnw=
zw6L|oarv6+(z;LA6u9bY<I3+v5act=fDzrwiaYj_QcK{|p)lGM`qt<5E8y7cPOhZJ
z?%gdxn@nUASH|-j$)Mrv>XF*|3Nu&spX5;2@F)TX3}2-vkfp}#PaJ8;8hU$2I3}C8
zTWL9}PEfEzC!AGPLrc0P_3l;xJ#eM&e8?v^j&hIJX}qH&NZr}qTI{0B^mqqdTEP=>
zvJUU37Y%%?8s%o|2g#BhU~Ll^_;4dZH>s7=l_#0Dr=2-agb&-X>&xcRPkH5-*C`z2
zXYpq?^^Uz(40D1a!QaF==d+Q^cXH5}OuFUQNi2O|i<y7Z*NznANvU!Z#M_pmx)&N-
zo7}%&RN2T6f+K`dn4%o3OZ~ErSMFLzs+P&pzU2Z96Dv-3O5giDIXynTWA?XCX@su5
zaGpy-IzaT<my%))@>Q_W@dk9;@LAN7(X9u@%Oc}D{Nw{t%}WO`c=f*AbQ+Cg{1~cx
zHn`T|soZ^JpZCd@n9qUdB*FbZIdZ0JA~mIQCdmZw+7%{#Z+|K*72P`*F|>1;ldUjN
zRKF)V_)XV;n4Z~~n~&6MRAPtMw5~W(F_(=egz5s{9M;dBS8?CrPBmfOo5RE?SCGia
z^K5lGZli36N+FZM)?=gt<-|pD!Rn9&&nh(2U40Vl=6h9VG`SsT@kC?q7r(Y#Fode-
z{1u^)IDT}$g*=>%dUn<WJ&Lt(NTjS>YzonX6ml-zH3BhVf29E@$12<Qi*Z2cfyv7W
zL73RWVSX#mcp|x%7yU^b9h=m4EB?{)+@FPmcp+f)PD@MQxhc%lIoR7`Y|wMt=dyi-
z>u_k@u#-wfHBITfd&TvuS5@a``n_)??~}n1>#Fz`S_Ti-vqb@bw{rgj*QbKwwIi_B
zl@-f()_Z_}{DEJV=ahw)gDjo;hLhACXiR?KMEZI4+7+$7lv*|mB)|7g$jRbxJ?Bj7
z6uI3kPR#f&kfDE_K1XvqU_|OZw^N!`+~&5*oMi{yH(*%(g7GR;VjSmPzq1x+aieqb
zlFyE81Q<Eea$~H#U9SI)Ri|h(7BqgK`S@`&cLL{;%GFexd##*Sor>zjP#AX<<N2fS
z@l5J%yzt_Q_3hCc8+r-4D0-!;$B3Mp2)g#g&TvToOtv<>PCr0O3RPBUxAU{XU|-mm
zWqX@g{pkB8haZ>StH5`yjSFcO0!Nwp7MHKWAU9#JYvbp5oxuwW^Lr858#aZg*&~k^
z{+%5mDDy&Mim=M6nTaY!{s!hRnexy2N)55TKb$J>XmLLFS8Kk@6?Fn#^Zn@4cY2=-
z#7kj8iY38=jjGHI0JN_#g3GP#oZI$_Y<2X4Wv^XjXPH;Y;PNDEnOaePg29(@-+jQe
z14DC$GN^e}lvNlQ)DzB>ZcjQrv6aUmzc+j+siDz&FtopkYT0iC&E5!;jiV;xH1`sn
zzEkckquY?U@O6=2@V(NQz0yY>&|!U?)cJ;PT9nLa{g2?%KKHv+N_A{Zj!(FpdS?!l
zI?Cs%-FBkb?<#;|Y<Q<lMY0;abTCh%rUHK1cc=nAj0l>9nmkE4J;0nFl%Eq7O?+Pq
zp%e}J**%dYnw1#8x#V4(#?o~&HYH}@^*F8lI4w24&dT)R5?eIu>UE6!4FDgJzI7q>
zL|*Nsl<7cAn9;fnccB}`j1{)-o9mj=x8@?v9L)raJl}Cn)e{QM+OFkX-Olb{XB(D%
zoRVFl-=L$vDp489z8$vvdQ`g&c;-M&ok`iTt2Cl|H)HWQvwd19(pRhABOA3Dl8c!y
z_Uso7tH~NC<<~6Eh0n^#hQEcBK6CKl05>L$aORU=bHN-J*B(_O!=oPOZ2C!5_>}iw
za-&x=|18iiYH4>Y*>fDi<fYBOq#e=pE_V+<8lUdI5L+C-nNl9ZQAPYF6&DNLY_M#;
zZGS=1l|bf7D->9`+Ebr>Re|_@i-3uB*78X^Lc$@uc4tZ7S3zUo<@O&0gaFJWz6Iv*
zE;CYHy#hXK&ohVpDLDjp<l|=QK$a14p!f`sk0b8F2UvQVapgMQUUO$44R#{?c0yCX
z=7quAo}0C^|4cAeyB}@nFV82XGtfR$MkARHP<s0FrYzL);W>IeqAp!*;ZgEv&avq(
zvo>qgP*SC|dG~wysw#8`u50#<BwX5N_^kQ*+M0GP>5bGkLD{*qvd@NSiQOGCEZ3IV
zZ;~s7cJz=yNg24&rtZGCYF=qq^$FYFo!oP2>Ba{t!c9Hu6QJo?mgwa`9kVx&6N%`H
zWfMi43l!vH@fa$@{YSrQ{XhmimgAoF(+d6YT(bb=b|~|(Q00!a>T-mZC2BR~sh{&r
zh^f`HTEf#NA!1s#ix^6pVG4AafnR>S{cc{YSBQ^D&AeUIylnfv97%aclj$+wN#lZ?
zdhG0*#|3+?DpwUVeE=CoNY@VlQ33Nxu-(fq!k7rOUAm{&nG1J7vX^Yxip2Z%B&y0}
z#eC$i6e`9Tb>2jGe6#c?X{KgSc>@JcCS^}+g$(RO%Tj;FkCI{OiBFPH#^A4pJRW3z
zDkxb^i<K!%7#(O?dhn$kx18qtes`YzK$iT=06dwqEOGu5VSXA&IsC#84<RuI8r_6J
z;1pOZG!xNTep2Uc?lROOr5aX$!d*TEJ4V&+r@RSbeBo-%s*nuRk%_8ZE71m*+fXRW
zMI|H=XqgYW5O-NRGM=zH{Jl{n8Q)cpN~s6)?_^v!P%r6cpC{AZ7pJpyB(2jcYdXh+
zMa@CyNjZ!7WH1JWuV4JSflDv=sGH`sKBe!MqeZP|Xc~<i1hM+`qa#=x=Bh(h-b1}R
zg*scgULghPz3Afryf!lmGQTYr;;ILNh1M#YJFhC8Oi7)kU|g)O-h;KH&;ji4oJsw}
z0jV;KhZ))O|Gq6_y?ta&V8GzpV{Oq&&5qtxrHMW&mVMlFGVkiHi>-CV6{zd<3^tpU
zto8lf+$W<hE&E(KiEq1_We{>S>sL9Fgv@?KLM4)XiW2R3c|4{VT-Dg#H95DMLotbS
zu2z0)LG+rvn!4<kQM*sOqPaq`lzHL5Q2+^a7JY2;Oa}@6d*SQ;vn&Alr`ZR)IDYwx
zUV^Y+!4(jP_=9_R&6THem&=~raQ6KKzdPiicjZ3;f^8bguy#M-A+lP+7aVpKqP0J;
z^0OaEU$Xp)uqaFPLeBjyzw>11!FQ1UO3~94!!l0<e%Q&7IigOK(EMSY@?<WErCPEo
zXHyrO`1IK=!BBFa%pMq=T03)^Z1Xq$dUrirY**#NUc)9^?)lF^shq4VynfcF;&#gZ
zZU<{nC7o3KL$m;#U^^O^cNa?t_t-=`H@}Vb;|QzQ-o&`S{r4e}zkdSC<GZp@TvN9R
z_fenZ&ph9OAa@pA8-E=w^D{*qIkUDL&DjlPCU=+d9+iJKJq$~p;dFZ@om3|`uG<ym
zN2~wz6((P^ydq^ZbDv^l4&NM@rixkkc<s57B0NKtzM;R-Rw3=R?e&`<<O-L*UDDc1
zfmDvY%_S0_1`ufoXA1jq<w`8mx@6)Oksh-FWP74&RP%gnBE7q_R-7*M8GtkZPv53e
zy0CaFS<&(4+eP<<Pd7Rx)R>8-8ar7Sxdx*x^_ylyq53RSyfln8UBqF{{T&)nYN<w@
z=P8xLiRbb=KRuT<V6}vt8duuz@-3y-h}XUG)_(+p7!;xt-tAqfg|C~xkK8xzJfWk*
z#s#+Y?;hOaWcsNFop8FqUmFs5X$nTxve0C}@_1$GgBF;3ARR)(F63mt{=7d6W5mYc
zkFYNO60VEIgN@gF*VEsbPU*`YP5j)1fUnc$QoQi3rDi7VpE!VBP<faMo$#WcB-hUm
z)aiLFa+A6)czN%+D%Ue4xlTH!1K)a-P!l%9&shFU=~ZtR0vffnp6)fSQF4j-#(dgM
zZRvLRGhovU*3J0%NA@u^75|SC_4~2ozF5*=LTW1*pjr!fFJ%WUsU~8<S!>h$Bz~=2
zV#oQ5!t2uU172$Pk9_;=E9<YkHpW=~RPDmP7@RP;`%Ut&?qe#tvXttv1|OGojVm%y
zI5!P?&zj#~QBhSVQDsL1hOzhl7*?Akl#;7jqZ#Y6Q0tBnFo&7YntTs4B~>H`2vvL2
zPq5Pm6LJuP2rt?h$`OBmK6y1E7Q@850|f*{PIjYZ^uaq$_jmIDHe2?ST3d`1=|M2z
z^Nxd}{uN4JG55ze5D^()60|r|2DsaK7wrJ3>ug<o9r{Q8emkIydsI+{b<A<no`Gcc
zo1eb(wp9Nx=#8tdELMg%Otq@kEuSO6&71X^Fn8$gGjzHOG)SVA>K&8Zt+gz|x{_p4
zaxw`vPixDOjuWVMBZ(^Ve>-m7Irg$}`W`sq{GEPFZ&K;PpMG^1h^EQd_b50e=2|=Z
zqH@CeflxYDR?p(w=AQbwKeF&-Ps93c(NAJ24k9U!+yP2;>p5OnmMK&3nTTZaA~7Rb
z|KJx2>yfX82gLxws~KMF`!Z;8+J;x;$Y)0^gT?^uk+=q~=vr{y$JusElNzB@zU<sE
z%=>m%66%ng$M{D<tD!cIw2JzOxtDGfyr;hzA<nYvMxr5@DGnpP7s{4@vtBt5_>Aq{
z55UMj2sQ6nrL-ESr1-uPt<e<r2-GYa>L-ys)GQZb-D^^^Q5q!ou-sp9Ul60%e23jM
z7btDZvw$05YJj@2RPY0c8WaGHOk@v47#>Jc6i1BXbKgSnI(=TUXLz{hZ~ov7B7!X<
z>_4#wO^@oB+@U(03II`xH6nsTBI;A7;F<Y2JRAt1?|Zo36%z)b!%FjU$S<ET7eAAy
zOI-_vT70WH{9n`B-JE2=0~O9b4{e)yY%i1^Cpbt%QP3IyoIIX27uQq}=LkCcsN{mH
zVh1R8T4p#p(B@oY3VfzO<*q>GrNB9gz=sgBd@`I;pxRb2faCv4H(6K~z>i>JzT`QH
zSsfr6ge-HM6lz;h5&Y+wAfoO=EF-riPZ$bw6gbaQoDPgQ@0g-DLC`!*uSzhBi+J$2
zhI|Vvv<9KeQh|g7>K?(2KeF7Sp)UM8V>8YhRE9hXRCWqfm<mm#uqC3*F1|lJ_z7z&
zkV};v-+%!!Xy}@7FtHGH^d=;K(l5?`(Q>q_fa5|U@a+XSnoi~~$M!4wvnoHoHA+7#
z0Ly-|XkCZkaj47a&a4Xg2obiiU&n?Bei9RYq=Y~(2I0PQ7W3=iGqH3+9M#id&so`r
z7AL%ohQ`dXGu)tLQZM$@v==wuG+@Gk6ftRwb``^H=g*pmcO8Y{o@=~KXC~|dJ)g7V
z!gl(WiHf6N8<S=N35gkhSm#5%VyXmJdeWCC46lQ~$pjNiLvgaJ`s`p!k9X9t3!=hJ
z1x#lxP?PdNF~Dnd0oM))Q1I@%xU*o6YR#`6u=bJ+9879KT%YtkZ3O>q<z?hZ?HDV1
z>=vtiaEhgj7C_;RK_Y7ag4)Y(eb?4YiZZQZaY9j+GY?3o$3CBvoDJifLBb5OfGwu1
z3RO@bl=-ue1YwVr{Y6)c!<1AwYZp;>v}&u9mJ)`Y3}Jsy|9MVNR5rEo6%*~ggs22z
z^e+j?CM7-S6naFlgyd_&o-<lzDrmec+^AzQw(Eu{#rb&{OKRB2+1X(h&Z*acy18(C
zU)ayxvXazvYs!>31@Wcg8K(_O)d(hxsSAW2J2m0cn9wI2fhiT@7P`DQWoB_k0~F_J
zyitrU^=oy~8<9rvktoi<*9w^BJ7aM+qo?p1k)nR^AN!;W2kDi$b)KeT&sTj(%r&Ba
zgtKq2mWksw%$|y6nY}tUvfC-+YM9^!5ig|Id>;!&&zIsJ`s45u6yGR?sR-LyMr}UI
zu;7Z<40(5~J=OLtnhJOcTnyS_8senK;T93&FLdu*MqaBw=2_?_h)G=c9aY|gS^2k*
z>JN4&H2d3uC+zIo8|<v-X8-$UiO`kh9T#7FX5*rV>e!fSfoQOmP`c@w2(@5#!9Ic|
zF0rF^!feWkOFE3#gMwk<@2i)%e+^SH2R|zQ@i_v|b_muv@KGt^781q0^L14gC($R%
z2no0#PmxftDW3ODj$&|f(Lni!9|-4HFzpDy$zMcOWbZeU-frx=Msf)b_N;I@bcecg
z)^RA<gm0e=h=&gkPnyKAsBo_E)qNiU<a#~Ed`nyr@?(B31Bkr}?VZH=leWz9@9o$<
z+YQrdP2}oe3$ezpRd73%T|(oRhb907Vu!ymWstYbwxR;YV8^HIjsnVKi6dYjn-caH
z8cr_CmjW+1KJdG8UQM-Ig^Li}UtdB3l1gY_%8<#EX)WYkzs||a`-Jqf?-FREM=nc)
z7VD$G+2mr&b7ODHCwXc~QB`B8EU>o^pR!Y)8%IduXm^3Bp-GhZ^D;!U7a9KVY$kBz
zEke)h&cZ7m+6lLrRlY|u>(2$!`PQ&;=ZE&=2^{h)M<1Sbak4qzdtN~Z!gs#;cT6#_
zf4DmFevb*YYzQBA50I?m&9Prj$gx)O(xm}EdUhYWVGfIJwDJjWlQ7&K%iPOFePi0V
zJLwBZT}}{Dxn>^!#2c-elRnPz9dk@Eq+4$%`GJ^f>)qDx%ewc>p@)>|rlN%XbOBY4
zy}nUymb_<Q=6>*id=JBfTx=;vjJ;gvM~+HCL_@hh+Ka%cBhSWSNAh;Y9+4+c$j}Mq
zP9f(x=e^Tj?0+P5VL*D`eK68DGl0UWM~oXcbk3p@3->$zMUvyM9W2kZ0Zl1VLtM?-
zVuB-J4Pw^@{mX_wPO2_%m~(gcp$ad<PkcdJ8XK^)bsaI5-<%q%|6upP2S`5J;mFh{
z`sAZ6(^rg1DLsJ`c<z^TlS5Yns4vmV=IAsP(4i+^lnms(Z7(6Smrw6}oO^QUx{rbd
zwLhhtV!;L^BL+*39XA?wmwk@ea}IB&gKWGEtiC$_22d5n>?Qdv1s|`kqm`Brabm_Y
zVu+U3EeneY;BQT&HqIRik|AQn>g@9-r5yZ@aB9<RK#DfAJ40MOr(f0SlN+nm)c`4)
zAFGJ940wbZY#|F|*hV~S1_UqFT2hijCm?EEr#G#?@IM;YY|*McR|s!z%u(}Cxx8QU
z;?37duOx$?2ME5jMSMSV1pYMkdx9J<F0jyE{L4Q7FvoosjsN%X--qf2=iIhK(CPD?
z)|5bQ;QLQ+_*;)pj6|Ix{I}HJMU<b_7)5wC83lNbs{mW>c00t_oIXCM9fYRjf25yq
zK5wn)5Y+f6(SoE>->SIFQv{PE=L<1<anqr3qHi;rw#WvfNUwpur)E^H7m2=ObYsx=
zm5WOW{5H|h)M)BH!xo_I(a%p`KR?ScCMkj*oMwt%iVR2(Wk>bJ)D~K`5(bU|uUp>Z
z3ekI%C0?6&_#)V3rS*-)JN56XO?@UNi5>R&8s*nc8^OcBoJbdk#ayKWjbAvc51;<{
zW*#=Sn(b834tQ1*lk;sFR^}l}Ei0aA?}f#{12-?OKGyf3*V507yJ0|X{(a+vcYkIo
zh#Xpe!Ji8F@cz?q3b1;ghyKDkTP4@FkK|A3Sr{G2b}thMKnMxpoW)Gu0ni$NxNe%-
z%OT5yg%~cuFL;}j1h11Y%gz0wp<#Bi#oq0WLP(i~r6Ugf5r+%LKqe>p;NAvNXHaLS
zEe%Y;VubBGFYUN=CeC)?PP|3Am%6MgeX#Z1jqr~AelR^`R$Et=A2uyrz8z3lminba
zl92bJW)rY7Ip($^PrPKqV$((o9~w$bbWi0J*cZtbp%4_p84$7dQs&JRqBxlQ0&i;C
z_4Y|qk7Y)ibdxr&sC0)JO2R}@+k5OPjf_GwH|ZJP7Ivms+PKSNL11-@I96;f`5fns
z9WqqxzC4$5df8@6{nQ(lBM0lI94*a(ja|^*#Y>G_Y&e#_7x)dkzM=ED+oraGUVjqX
zSW`;BJX#eQ3TWuoYKR_q$YKt|m=mYPuBz1--tIjhF|WNGCXunZB?J4i+%m|KlcOpP
z#EOi;h9+txu2A4qO3gj=@B^TR&Y2ZVs#X9jPFXr;i)a}d-;ZL*(3zpiOgt=8BXHsj
ziAI|!(xlx#0P-itgj!863y~@FD4GAWx32ogOlJI>gEqI&_|G?!+T$iZ$-!FPJU(zs
zcywXiV!Qd4BIk@!S`K2R_X92ShgXo;=!$6di3tU{CbfSY(Y^1v<LqXcW(R^>Wkqy|
zU)j$n>R%7-4}4x95gdK_6g(m7ZvS|c&XlXClR7as6s%kWTB8X~`gb!Z%rirOBOJX6
zvtDZ-PV4+Uq(?aOQQBUTHULo}IBn8s{p9-yB@*Yv+<j+Q;5GYQk=y}C^>EaQkM2$X
zb%`CrS4;y?=|20=tz_9&QU8ObQ61qh@l5aGv3rCj3HZfm8bjT0KAL=e(O)C_aT5!9
z|NU)xVcI+y)`j;E*=)k}%g}B1z3Z6Xqaa|K7R`FOP{`8L&^y@oBDm!8S*%A?bTxkc
zml(}=Zl;YPgH&W}?UGt9@Pf)-?@Iiz{+8iL&dzUpadVmb&4($&&Yb>;#xF%m2@j?B
zz^{hTRcbN^&4p5(GQA=|9Ed*AFae%Qz9<T}90?5)=6J`Jq!GNIn3o^quW1C#v|%)q
z9?+{E)M<ThJGk8&yXxVF>USqvWevuz8GUE3*Qg8Ho@{x!w##TFy@ZKwi0$>qYeQrM
zimHEb*!(4lZYs?$qo8>^|1dc*Q1`c=xRxPOlUu(taiZ*2n1;LFerh%E?W!PS*JyA4
zc0kfk10WeWMjti0Fzw2{fz!r#02fXA(XOK0jJYGetK!zH%ujTA&m38uH*8#oec$8N
zOu>6H$Gx=|3;v>f;GCEUEc8&@3evBI-mPxIU9~)&ld=J6WO7LMaS{iAF&@yuswP0R
z*zm)TM6!cKlT$e~TRLW<6t#KkY#j|fdU|;O!&N`z<@3=({TD=Gm#EpqU)6~zI<i)i
z5*PpI>Ga^@W)k_KFmE_4!i`SwIJ(mmnCk~`thW;`pU;C?<VAqL0?ilL)rRd^?yyLQ
zK4sw*Sk!9y7S?wjAq~LAVg6_UYU_*$px0g(|CbRTBLZX(Wv2_`*c19Yh5oky;>ezD
z`KESfz)=Mx-T9Y8ZA)UP7A6C&8ncc?^R+a7X|5}(<;Tx*{<n@Swxsa$v1X+2tB-<k
zz1IP6&$TT7CNgsrbLnziBxg$cxf!ksE}*Iymes_oQg2d5_rx8$l+<~z+Zbm2m})El
zHPpQ2^*-kQzWbXBLphl5vd;T*RwbqG;bZ0t<_LU_R};9YZ`Bbs4z$LaagJ&zr#)G1
zTPKk?E8qioK(&@XadnMKQh7?Z9fO?qx<baqfM&-}jniqvT`KcpqeS6RO)sFk)*`K9
zmaH<>=`iBgP>DFd4KJqFYUsnLD%)x?J}4B*Qi!F(N&Eqz@nN=&YmHoG;>%QmJ8URh
znj7P=xVu}fld?wr+>_waJb^Km<+0+iuNosFxpY7E1BKYhthqFFen}eR#yTk)Di`5?
z5E~FTg1+#>@1P^*x5jjyTvYbDbpS1x!Z03qZS9%-F?Z*5=uL3gY2>~vm|#wzqIfYs
z0G)4~8LUc};MqyqPijN#HlWspMBYJ|Tyy~AvuWnud$2e==NsyUm$%oz@17fBGcQ8A
zD2Rz(5p$jdZNa6$AIY24j8ekO9I&>mUC)~Otj(KuNuQD)5h`wr&6tf6Bnu1;`_2f2
z!MdF;`1TsFftBRS7(mN}DZW|u0ubCMwJpRq_bI6pzoCS;R<!`19>;t&gec*0LDb&J
zBH3&RdBrMnWY_Grh`|@!e3+y`7MA#}+Qw`;>w+^L!^rS020^WZuRd)cbD19f{aoMZ
zI$&lhbViRqvVXfJ(CQo6+2KT-pPQ@lDd{t<E`?KGRmdp`P_!|NrFvnc_<Cpjj|^Rm
z-B~3Pzzwp??58UpELyU3Pd^4OmQ*E4_b}WTxeT*=4)wSDZL%;(NV2H;8yb{KS<=E=
zoT<5zAWH}4vzs*Ld~c9|c-c*cpJ9CuC3X21@3G`qY57s5cGdU>^`h<V&`tICfr+)L
zuCRo|5z8FMT;vD9Aq&P$N;Io8%5+Omm6)0J?m9z-R(XhjjtOHKX}jT?nfb%$kSGUS
zcyLWz=RGYX=g(KMZj3>+QcsR4^N>y&xLKxcQ2S18-A+a~s{!sfn0ykITSJEXbn(-%
zPT)4^x)Q$yg);wGv(>Qk)~)7$c^Vals9?;}(n3*5!lRQ)bx56KUuJ01l&IhLXh%S?
z2dy2ih&>Hj=v3#G*bDU6f3kU#RG30_9DLzN#nuFL37%P;WW7Y-0{jFgC#sMD*<g8D
z#dxc5-bbR?O`>~4{MZnDl{_<2f;|@$4UEu=*9i-chGn7U!*3WT*}!Rhm{b8B4ZVEo
z8VZB4;`7kMMNhl5Q?k&pnvV=YnX_Yg2{vk-yeO2dB>g6Yf=~{!0KoJMVnKTq%?%bG
z*9E8MpYK(icBxW4V@{Y&BtaSo``4yoboQ9L&Ei6NPF3{AgD{5w%7|5QA3a#xG`o9U
zoHNB$dMv~J*Aif5Zy-LzHD`9b^^U?Q2h2vCb!qQpTCC-b%TeR3=%ZGein~$0z#LP;
zIMw>Al-q>A1n(^^l;!N~GdBbrL>7gZukM?Ox=dmrV=?D+%)|?14ckt!^h03_33HBf
zfaK<pU-@sn_jIS=&F>hSB;;}mD(TSsJt}T+WKjcV4}!`*<Cde|f%q3dKck6gLA$Y{
z5jOQeCw!x4B88ik7JNIP?N_Jdf3dsn1TA+AAHlw|u720Vp4M<UBZByC2CgW1qgTg|
z=RA-Mw1BB!n$JYxr(An1ma=_{WsH#<8dM8#U-`@4kZQi@wGktt5>hUc_&S_%>9BRo
z_p8a^CR{b|p&B&eFpr{P+<=wJXw2{DXPaK=S+P8~qfXLHrJFHwYayzU&tjt%^>~k!
z0w3?48QLGwRhARgRVun=;23WlAvzjib^x^%2q$akf2sO-a`@rveHV^QO_`Av<#?O(
z7_lIsBdzxG$M-y>&hVo;P%cK7%zG5HM6*24aG+-tz)=UM5RGgyT_DjgL`QJdF%X0G
z3@_V;*e+wtx@=wIzGPGoEVx?kqxZc;e;6@;FI~N_*qFlhD_oY0gZMSg$Hww6AH93d
z^Dbjj0~8Qb&r-MR>sij`5HfYQd{A36q{U@4UIXvjBBVKh_7f(6B9;ToK|+Z{$zHQn
zE^3Qi6v-kt89sc_bX)q6BWKdcwDvW-^fL2Q@~a}q6a^4Fn&u=Zlk5AB0`Q(bL-~34
znC05={KL>nvnqx(rMZ`t$(zlB;yB2|#u4VTF>*)hBwumwY#Io(a)l2ke4}6i3Ah7B
zB%x_ctSNMnFa<2WLQv0xslz3G00wc0m*_BlY5y_a6Wupj9D*&{X_(De8I^xIru@!1
zAm1taHnYFFx?3>S>;T?%)jUI)eGBA5LGg8-WzVQ}$g&~^Q(zYue5jPz@`twQ9a(<^
zdWnB@K^)si!j;9sL_z(LU4iU&d22EUuGUpK@;#s;%D+$+gLBUgfL27e@2nL}%Gyj=
zh{@b(A|}@p209I5SWGPeD;u~sqdljqe=)Dcv#_h$sEaB)I<Qzi{xYjoa^`AeoiK91
zlPJUx!r<49T*2T2HxlzJY1iCx&C{ae1y5}5Fh-my8^fQ6Sh452QkAK2PNUq{#YZ!F
z7D{2H=qR(RHee-QwH#q5k>#x0@n#s1{*)!_!@m1=1%7zy*1zWzO9nXsuPlh(`u*3Z
zLOlP-@yz?pSUDIkz=<jM?O$3(S66x$3QA&{k!;5@Yl$30y<9DA{s>d%hI(qk(mbix
z8NO51LB*Fb)6K&>E^f@G82|ms1r}<qxqkj(Zf_qBt}r99wb@Qc;xwW6c#syHu?@QI
z_VmP747Oe}LM;|2$REg-x<5x0qi4{asHlsWVS`>sLZ`H@a$nIUnV=lUZ~gOnRKj8i
z9H0a=S~wF71cDhhdqf!plgDm!t3kUE>5I>dmO)>AZ1uG>R}pQ_Nn$m>)^TJ`1Ddh|
z&yEW!=E}>cPcj?yXDKjn0_0=~ZLUOypcMk!QTqZ9{ODne92uWnect5!<cuix)vB<(
z-Fs6tbf>pMj*-cnTc_68TzJG?49m=Y-gy{mv{W9`k=tSKF5vyqi2@@;PQjnV?L6>N
zQjZ!j4k!_n!;Se*ZXl8&33c^D0d`RN^qeS$R#^=!$Fa`wGhHua!M6kvD~nNggK&8J
zZzqQ!N7X=_ZcIiXl>mOdpQz4ne~(88!8~{o_Y5PDN%(#@x`e)=!OM)q#^7vNrhyio
z?QHy__r2SBq}=v@m<a1;f#o);8F!S`7c9RHB9BRBldN?W3rKnN>BVa}<e0yzcx(9_
z5WfFW)TSC=&4V_i5mS*mtE4(6--e6+ZanN?k&Q`L=MPR4yGNf`$Bbh<(B?;fZr)qV
z1%Qj7({uzYq3D(fIOT7ItrK;~a7q3~&fRqJ=<J>IZ|ICUQw=5H!?`ttoE2_M{rAlu
z3Ilh}uFurE>|5vN!i%2oh-~4V0w8T)3XJ;zxhzbv`^#dKHlR>=$eddxgnRtEfpXcM
zF#r=|a7X;L=t(_%D)|@J>8D#AD0QLSkZuI>pByFXUvA>9|5{E8wia!jPe}>U96N>^
z1BY&6z_ai%1NSQR<Fv{+YX>_B!VV%Ti2{l(Y=>!CaxJy+N4hV`v&8#2-fCyJ*BwiR
z=l|LnEnAHqalTjl$yg*2C|=ihRL~bh|Jxh6$HF*%9--}%_gW6Qquf%wvW{Fb4oH3M
zZOb+Q9O9wly>II7<l#0%u{dLGy|^!Eljs^!l8Q(g2lT0vtdsYz|9j8a7=h57`^B9f
zR$n95+d`j2v(22c@G*IByX{g-*MtI6HQj#PFJBgkHL^IJctY<-{LUyLwLd?lz9EVH
zZ&Phi+7*cS3h$*2qlq&pnP_(>;0uNN?!TWn7<^jfiU^B6?YX+K)TGe<Zz+@Wjk*wB
zCKoEAnr2ih{OB?C#M5<t(GRNq^G(A_`0NE<z?5QF;xDoGwHNix?OS|K67h%2SF;u?
zqrS2y?H3*mY7z6R(Ux^h6W$B%DQB^v`)Q~(aAKx1=F%>GyU^;22kHSiI2{?d4BhMU
z{2&J=EAv&gn-fB9QQ;z8C2v!`KSN81;&p1=0Hx_Yc6C|*4u@Pmayv+jS2i!tBhtMt
z^BJUvzL5LaSVW@G(@lJ>zVlPr|6(r=#U4`!AR`XuQ(h-1p@li{B5I3(4FmC^)`(ab
z$Dk-32dUhI_x<V0;>kS|2l*95drz29{vSKY1S2+o&`R?=NrGaw1EytzI)#ZA>u%zG
z%jOj-n?0obVr96EhW27wRZ-pm;}FM3Te+U&81)EIi1O-Bi0j;SPKDf%-#0bsVxl=V
z42#<o3yI%1>ro5aKzMR=@3ivgK<@u~(etW!!sG{4Ya<>9^{5TO3veg0OkYq;5E+7}
zeDg+SP14{a2aAPRQ=cBv?D6O)Bk0|XlZd3&IV#uB#oOSz#;svhCEBT>(K}c*m^~yw
z51^1I-vrpgn(<Qp<HCsaO(Dk#A9Y*>t6R5!P{b4a3}6QBkNww#9iiSbMJ}y8E*%<H
zdmB2aOU@(b`egoP>EPnMsYfL)(STJ$d8bV(+xEM=S*RZ(*0bH(1Eqr*_B+LaxFC!m
zS4J^Vvp$3f;i&}tO9Y36@5=qMr^?^RNCDaYI=5NqI(c>al1uLkJ~VsTG#NoSe+2|D
zz*i0}?l~)7*(nIq1o<%b<bznJgtlNk&$+6w)4Y?vA8%}=*9}N*v@DCl3kXtKt@7+}
z;7NNg+(Gc|^Ox#i_xLgg)IqoEMojc!ZCcphVB8EfP==56?!SdN*<%+|;r0tYU;(|)
zueDKk%!ukX<05K4;Kg8w>`jntx-{xn?<-gfw@GS~iO?QCBYXqK)P_9)_eHEML2(=D
zh9K0~fz~5TJgIJpOEGlnAqQd>I)?v>xPJFbD=6CPKH}-2&$sRgA2fQfbnMamP$u&{
zKp_Nm&ZAhVP3)3q4;v^L6opPL8G&q4{i77oA9g#eKUjMGJK5sZe(_V>hMj=+cgLlS
zoEZ!GhEFTnn5PZ<ADV*8VE@ULG_^L2P<kYl)||(W*T(-tVzgELFD(A|xp**kp?MBw
zsQDB6od%7F+dVuCk3DEewlcqf#PZ_!Pb?=TUL92Zt^(Rfv<=SG=gA>bgNADnwABAN
zWX>W+`u7(~o){pQR5OI&x!#xBaH6!{+j`mi`g#-qs|0I>(Z4Uzvkq_YgyFVd#TL;d
zN2@8-9NxfFs*4h&DPv+bqF7pB-jv?zvNS{7ZN1!55aW1aaZ}0XZ&F{iw`W=C&%&EF
z`zKk35)C5FT)MF?(Ci@p6^!NtFb#Ggy<$EJ<?+7%5!xK1aimt!V@}|7VP@tlZ1_1=
z(5|wb|0FwSa&puwU=-_T`(f^bj`5x!agcoI`IZ0#WTMPgZ*Od<%cKLK#mC5k|0j`h
z${=J}rq_+mS)?4C-xnE^s;{hVk_`nas%j9SQ(SnHm%b7|RGumGEIH9;K=&!)#;b-~
zV^rgIvE}>+rmauwI0ROfh3M~ve_UA;Ku>iCK(5+<k4n31GS%b&Hl=EuLqLpklVd^u
z4N$@=G%OJR$r;x(0%E+;Ouajdz<-a$qLXhnW5(tT4&Hyl2|<&TzbcQU@Zli}7wePq
z(RnmMEaEK)C=_?{puTKXOq;5l^W@?NJ()h>u}W@8uKUBPy2afjTyo`hlwu|oZrmrI
z=@W5hjas847CJ08x+F&^qZvVHc_|W`{q(SN(h|&KT5NLYU*Ko#k0w*^!<j~Ol?I2r
zJ+){Ye2x+w!8?xKt3ha?`o=>6-VOUtD%1hMi>}W1DC)5FbLIbQ@7kZ5y0);5vCU8`
zVHBc84PdX-@lvKz1*H<IqHU?52ooNSK`Ek03<W}gn4H81R&2u*YpQ}kLBYpG-Vxy@
z!QP9Jh7c5K3<!}T1w#_Rkau!U_o1{6oqPX-=9e>PCTC~w_4v+Od#`+JKie|eNO$~1
zzam!O%=L=hzO%NXI>c~;CSNRg(i*NLH^-?67%4u{H+D<bGD{7{Pr6euH}#TdiG;jM
z-wKyB;I2M62pj20_CU{`{Cs`>@+H{xS*=8XBz4$T{;ln1M*ZhUw4RJ0$&0s51-xg`
z{o}n+KH&wLRngJ7lDDv33%O~&QT+)W??1`rJk02(0{ErBa;a$4_iUS*5=r2?G?l!Q
z0b=9y6K?5gRg8N+UHm0l_jM+?7nJ)cU#FUX5Ja9#lLoR!0tJJ5mT%)eE?>A!-0z3Y
zF62~xfSRwsC8(j|mkHgwH6GbMp`fWJ*EOw}3!4j(oCIT?zQ9_84}x(QUo%R5Qnbnq
zF$Y@SSXN@ZP<;Mar7*P;AJmO|yxLEan>8`$!ri(fG^TTZ@b2txG3ct3u=eS@^&mSK
z)+{8NLxb;683*We>6d7^r{EmCT9mqxf>xh}0wtmreKw3EOD-b_%4AjA;VAUTQ0jKQ
zfIn(}d<T3d-50ZIzpX1Jq<7QDm4|5$)@C+*=lQVz066^uz4eZGw#IEwyIYhT9`^fq
zzYWVgx1f0P=B?YQyGRKYGi_p9TjOh6$3I}&dQye>pjUCx<)h0N9~5r$t|Q`N5nYQI
zke1#}-l#D<B#>Q_nS6AeTntZ$yU_@vSI2uqmb4g(lh_Ug<mie-mFctq*rz<tQn3E^
zMWIDTKNemj(hi7dQ1G6}^;q89827`QD;AOm-@zI^xv1YZqNF(i#R*%K_uzlR{27ZJ
zVjGLamSR715i~HfkJ07tc;vsT^GZh`O98gl&{y6pwgGxb4kxMaAq(UOg1eTz<C%tf
ze9&Pmh}(MB?2-Fpc{oh&4|{Y65sJz{2?%EviOe1T-KZrBXz|5hY(3afWF}m0dck6_
z<}YE$Yl;9(6KD|s49Pys)N@js@j>~sQd*KKjPkA{<Lo0Oc%;kvX%?Vn)#iM)N(Xvc
z*_Dn-2w4TGo)cm=G=Fk<>3`Mzb90@&7(KX!?3qfxo|ipmLbB+@=o$h>f&<pOvhv=)
z$m_7iH@T6JwKsrSUq0y=njD;NsWlhWUSz3EEUHa89p;E~k~6S{&)izHf`Rz*(N|n^
z+P`6)QRjHqsc^?T*`&HsG<VrW;F2uoNk=?~^G^1E4v+Ja4F7f(AOoMSeonIght!mG
zMXZ|;9W(kPFa7mb2ez0L8&9`?1X%XtBh(KROajK!C$Y1K=y*t!!8S3{MmsT)RbC3~
zNl2y<Wir{l!UR^0@MvgHlrJ=nhE4~AHeaSjeEJjano_a#mN+jPG$ZK-MWkQI0BiGD
zw`ep$l+1L!!FS;%eHpqao;O_n(|UIdlgn9rpM>a1a@YhRcNT45+3|E~*m2M)qHD+`
zG8v$uafGVhNDZ*@Ce1vTrnFP>luw-`)IQ|p)t&(=kB=I`^QK0!P#x2_>;0eI0;+d-
zpr<eyz*rhqmRZp)aovtk<@HpL_nKGqQzr8iN(Ec1=+>del`Hv$xal#Yb?;y7{N9>G
zSJACnQJ(RV{P>}&%NC|+jCn}YjGj`&^1w3qBl}b!32VQCdkUz-^^Q1QW|^&NLsT_>
zuYi}!p9UouuO9-vsC7q@*2XrfAAgTnYmv*Q5m2`W#gigzTm}m4PM#HoMG1|SCyDlF
zN;H<Zrs(?)`1us#c0E*zNGf1d>EU>DHDs;M<%|IHwrQ3EZa@ua$y2&jy|?f(h0zzY
z+wx=9SEgfekDlMBZl3;-1Z!1r&ER^oo`_5nFT^RAceZ}-Y$XP_>t0f8Y~O{`3xK?$
zj-yjL>OCb>*=71q?k_E!I;9jNHt`)BN+H>-BMqL@I9}d}WZkTJ12#)YAt+h>n4CQp
ztlHe@DXOf}o`QhS$aH`Sw?tQ|7CC#!oNs4;-A*}3cyT~c6TJ`xJ&l|B=~zf$8=!hk
zAp&~7f&7+2mL@5MygzqT6Abe46uw>K&H~J<r4;m?1^%sdP3xxb@MqU=C6$d`^&wHM
z*Qi^VK%2oJl3Bf}O-j)fQ57lk<og>`+6sW`jPAO;SnqTYG@ciNz7~fdeIOKItD;-0
zlw?*PStim*MQQSrZ}Kh({LdwCzx7~eD|5w6oWX9Lhfs4%b7@-U`llM_fc}q5-9NCR
z&%(9Q0rMf-`K}}fg|(mK$`?4n-+Z|NZ{zZ5-f`Oh_e1Sxca>xOd`lO(aOOYx=iL(z
w0MAN~qP~d#A(z?t{5=t8heTBfBm!vaM^0`2(D2Rj1>o~hSj3Lo+Y(Rw3dzJUQ2+n{

literal 0
HcmV?d00001

diff --git a/_images/ppl.png b/_images/ppl.png
new file mode 100644
index 0000000000000000000000000000000000000000..a40a8289b62a8d3179964306a598c405424f1ba3
GIT binary patch
literal 25483
zcmaI7Wmp_-vn`AT3&Gvp3GN!)g1bAxgAWeD-Q7ZPcXx*&xVyW%!x^5l_qCsQpX>X6
z4ATtLcUM=fTC1x1CR9OA;tM=3JQx_*7b!_mB``2ZTF|jCEHvonPC@1r7#M<pg@}lP
zl!yqif`grzg|#Udm}F?8I*f+$FlLr^T#V3<92vNFI8t`s1Z1rPf^Kn1hyYY$;*f=r
zKfkQ7FsgD&Ny7+`d5z2!PrSZ~ifLkf=1^J7^R0LW-tRr{jrqKDUuU!KkF9vXfYmGG
z$5Lad!Vt@A5J!E6`yKsVDnC989RC0U{Q#0DIGxdGpidM`BWS5J`Mwd1)TWYGvi@!H
zT|XQ;I(z&RSg0RYX7c!`{{u1@07ISpE0{0!?r0%qNFKJ@Z;Bvr;Wo)I*4-xQu<2bu
z_Fqbps89Bv!K7*Nqqxw(W=mrVu^d=)f-H&p<b<<?V1023jMm%QhWSFKi5vVlNJH|U
zB5sJBSl$a#1<HmuO%RPAbxg!)6k-^GyRy%M)0dqxdKD%^!5C$<np+cGhZ%-zXFIZp
z$wn9<9)X97lgL)WFN$69iy{$9HRL&=4jwZx>Epk+A$gE_FS-0Sk@1A_>HkKCDO?=l
z{4(I^`x*FK>=%R!M4vk{<JLY*Kb<v4IMpDHXU;8-5vFpt{zizYcYX(<lmlui{41xX
zXM-@2GQ*g(WKar3wO9@@4ZM+R2*I`ERxbvJaagT}q6mkPR0f;(=aB7(o*1xZ0ksqo
z#3<#3<h@;_QLN<*3Ihq63%_4w0Sa#3RhVBTFa<L{yCgE2P$C{)Tn~yPj0O%OA;gUn
zc#L6D4$wIXguz2>?HJpKQFDqLMyFpoisZH7^i@0tUWQczl)!aj9K85MRFLC|qtO`0
z_5%8cw!4%4VY?vwd=bg33jBwO@nDzS@!U5v+o-UA`eK_v2>euF5+RE-<rfA-!t`&*
zhddDb0CarIl9Av^-knnt^$CQ6${H0oNMZN_xsU?<bpVX9k*J6sG7$2_WDmivh-Et1
zc3;b;-h9T9s@HKSO-RurOoqsGZHD5oMG?q@Q2j$2n!wM(J+@IUj*v=O$GZ4~_bykn
zE)pi>0J2R|2gECx<SA5dG(gD-@LPY3X32MEtR{>Zr(VdPL`8^+_5>?nTpJY=NND%|
zS*ye&N_`BY@UK*-oueDUg*jDZys?TkUl+5judgLlc~u(?F)S*4{zmi7-E4Z+26=ej
zjX#k*^DP9uTZc0AAR6sgPcMj_V9$XaEJ&51SoKHATwWZxlWOaVtZkWG?g$Zd3%-2e
z*R-I#&a}*Q&*&I>hJO&7a%D64j`<z!GfOS}f$W3W#`L0a#kOQiZG7!#u&ML#m)lzo
zwYyazC#_D$>vyoO{hf`CjmdMJ1u3vQkFHan9bWTZ-)uxvu&?-=(tUvrh{<}e7YsSj
zu4w{H(0g6bV}m$NkainLzl5+upy7x=$p%cCcRPQDR_;b%!b<REslqP<$1p%!fOq-D
zzXA3GY_QAXC!!9tdiVFAL={=K7pOOW_Xc3!kWq!jeuM}Z{n8-43t{aM@{A=WhX@r>
zieV#1Lm@>qfo3715m6(>7{bhhUW~^7CP#)9>$3B8D)fffE@qrMsN}a+5sGDClrRo8
z+7}753Ai7WJf9fEv3|=;I{#pA!7G=L&rP08{{i!adO8SJ5OHWwkODc-rJ=!=6lQ8*
zWR0m7mS{YyK`;lEeeGNZdE76sDVh;9Z4cA*OXN~CBV4}Qd{gEE-jhl%x^ZLvQmzf@
zUHmsH0^A)GBSbn=QjAoJG&M;AS_2v)+E`Ef%y)o1IjVEWoG4NgmU75q-(ep`-<Q7j
zKIJ|aBdqFdQPRWs6DfhbAE?19LbN31RHu}g<ho?e@dC6obg9(bbf4(*Y2gz0XcFm8
zm3VTTz8|JefoW!1OWl!5#JlZS?3nJ@>`-1|IB>f8{`pp2@SuQ7y_nFP;GDn%)C58T
zwM|V;Sxx14HB9}k@OPbo&k2mwuCzW%lPcB9hQ$%;u|@Q%i#ZO`yA?of+C)pS+rk2+
z&qXrD>@%!05;M^yY^oBf4Ycyk4VL+q?MLjh>Xs(#GVBFQqV-1g4)y+)>ND=M!$)dI
z^v+RF^taNtWJis&&+I>NbJ_o}U$G_qIGJiHy3~o!v&hHE8=ltr-K=^YVTSc3H+5Qr
z?W>J|4c;kzxpsJwX?gz~aE^9*DR;JDwy?bTv?Mc6H&3_NOQ%%?X1aRH@ObBlewzEY
zqs)G7MeJAQ<jPi!M!Ea6GnjT1_cwRXcBS@qcg|aZ<LO)f8^;^<1I`0MwyrNH2u5ED
zaIM+6tR{Gctpe_0&qAh9V8mBgMZcUPY$Fun*|J-+pK!8pz%!%blymYJ&sdcipV(#C
zv@JaZ*hj;S_RWhL)GY1f4S#Oo$dnq-ugq<nc$WS0UF17Ofwt_jEX?1sQPk1B)Khkk
zA9Bp*Dds7a9CjS%wM^9^bt#65R`%m@*27hQYjmvK)~@^$>s4!Jg-fAJ_vhT_tdEV3
zwT@vpRPY4TX*gy%IITk)5SH<mzci?yD)4f)(zi;t(%lucx3+hA`+Gf0_DGJ$_QcMT
z+3^-}F%iHMSQ79MIOCJy?{jl-b>NyZ(>l+DDeg;nIXXB3otO5UeCqZNswS_+IU~Kd
zt@r;rS2%TS3OmEOq@GS4TrJJ6C7!oT-bUe#GECXzXxA>QyCS((dt`b<2M0=|NhA*1
z1Ywc|RR`F2$8X?la4%n{>{BtiRzg%l>iM|w0`>vtSKa~x?g&1A8<=$k%mnbW83YJA
zoYvXaM>`0-3%v0=EjtI__TKp3Pv6E<8va`E+3uBs6@uA!ZTce7GtSA*6${Zbkyo)*
zF-WpfGZ&E_C{}6yH2Z1uQ$Iv|=q5%vCt%`Y0wbq!QQZ1#YtSmz9tsC*8HEl7pFkVw
zi+zA@0M{QH)JTMI9gIg}Ad58iEb^B>i<6~)hz2wUWkb1fWg;R&%OkvSIeBcwR~`Xc
zFL@8Tm*;2@s7Vwn@^w-P<Q!DR6p`{JQcp4XDQzsx#wSJ9tb=-T%#tX1Yz1*j@k)(_
zchR458_jg?0gaaw+jifpvJ}`hw?zFzo6(}>mD2lCotPqHk}?|TPCrwnDUTk?n-{9*
z-`i`h)UPD-`K(#$A*}~@#=X<`QW={LT+3$vRZZpB^w!+fY==ntEquYl%jThc7o2gB
zLBQ?bJlyPYop3dE6$Q&4$la4PJP;pF*%*EzM)4<#Mo)Q8_7E*Nym6oG5$y`^bKOR?
zXKA1KRd`<QL}OLW#r8~IF-$mHSb$jh^hp6j)?FD_(?jFogVWi~9!>%4-q=CLa7ME!
zCJUXH+O_rd;%770ak$i4$FHWgV=L?X+@spu=c_L}kKaE>VE)ydxRn`VtfwfU*os^J
z+>T|d6~Kr}Z>fD!19;uC%Ws>KF_kmjHPui0nXL2Vd{cSsO`wxjU1yW}ToXAGIubZy
zSAL;g?U+9R+)C6fB~_oTEU5gd`)WIy$Bs3kJHlC>UzS*Qs)1EqVc*{Ww0qpCVxqcI
zhETMkZK_e<bkqyI7G#C@$~kAlvZT^*cx~3b$Ez*eFr%)mnrug5%dqm~q&?Me-O21-
zx@cR~Zv8w@&_h7LyY1321DKgPVOgm!!z@pnxjP~3d_KzR;*WGjbk@E*I$>Cbxs$qU
zU&wBFpYywb9EHJwIY!hZEa8#m)zONckuN>B8nQh2A!}hRi0YEpJluMCp*Wb!8<k2}
z%2(m#xd;ag{X_*VkC}ZUNWnAf!E^1Tex0}XY(9|HJc+tI<-+lD66n~yKbB#hCFs_@
z-%}l~hZU`3q+{81;@#XaTYh_SZre#@>u6Wnxn-Y!S+YM{aL0T~Rq?3H>&1WZ)Qoq6
zmux4#YTcO#I37E_Y~($=IOFR~_qe}p@9}`V-{&LdqxLF#t9psu%smpceqUJLT(fUu
zt*X7Xf7yQ4U8zWxzsmMJ)VoM{s(VQtu1(t?8;`z)y4UeyJ`=l_nH=$3ixXsjD|0u!
zKAO;f^Qpg!JU#Yi@aBIo8`gep<Xvm-usVxA^tll@f9(pMihd$;^)~Y4dFOt-uJtR{
zm(-v0s()R&VR%{g?{4e<h9dDd_7&cs>JJVMSlc2v82cwMT)u3@f#GqVnqNcXWh_u{
zXi3>9ru_uIJP+@L%b(wEZ6T}d@c6vHdDUNn^(lZ$UO*LvkNhOyDIFe5oDvS$_Z_X)
zpT3^z?xeY5-AA&2Ux6qtp?he4#i~8=G#pn1M*uYxT&5aQX5YVqeY6;`VBo<PU{IhV
zaL@}E^a2Be_zm&jpFq<7hWzhiFyD_qQuVHyfPo2tNr`?_0f3*b!lgKi*$ck;<I(=w
ziqihMu7HBNg3-?)E+!WF<4<TPZiMy<*(1$a5HU*bpC3^76k#$GA>SsOkX^TbDdsl$
z(VbBb?7kdNk9XR+pLv{R@?MW+HnS?<@?EbzuQ|$PpV_&ps;jE2pHIO!ON9!N#=wH=
zn18*|BSY+K{%$c*MfDRUMv;X2_Zu~AUQQe&$^ZIP5(*R1k4*)$LZ(Le-z)v!b%Ug%
z{@;Z(xS*yKG@HH-!&Cgn#XnBLrR9eH_eK9P!C(_vQp9w#g`6(of7we8VNtY${hxMz
z{1zxe{3S8)IIc|L|GC8sMv{+jjD`!X->+hIKJ=eCW<TrPx30fdu<h{ebIZF<r?1s5
zTUEBV=(;SL&T!pl0k%R{?R=h>C+)D>55!s_5xIUZJC5^9bnYhq+e?_>ACAbuszD6x
zqD3+A^gPyHKa%6x3^*Ine{-M-w0C&e40a91caU`5{Hza9xw7-VPCxS}dNH^kbM2^o
z*;~py_uRK#_qr!ojo1IT7c|hJSZ=Ho-ARvQVap)E#fJr+Po{Ox-RF-x-e1pM^`EAa
zf$$@(Yqu3=C$-~xN?UH@9&2Y4B1os+ugChX`maaot^9Wjx!A1DX*Ey<FY^E66?#~Z
z3+W{#OxRKVW{JBLnomp_mrvX8*H$Z9&v<CJ5|rf*J6=!iT-9`F;ow@|ZY$SyFMeU2
zxGdW=p2@kkC~JOi?)vmq!ESi`KYRiG;fn&g8q(0NBjhU1U66lm?|X^fs#*uwD<*ZX
z;*w7~?%U6uvtM?oTJ_(ql4aoQ)2?nSecA&8R!_`0wH+AV<Np=yE>e=1ARy_j_#6d_
z#wQ5Dx69~}qv1!8<Ie6HJH1?Y>GYcuj8u221RsavP&=;@E1f{GWdxRl{jARW56dt<
zEUP!K(F^U0fX#58)6DFo)7gYyEw}SJMup#-jF%T^e*Opu`OEP((K}%9cXeeyVdDSa
zB5r%BtDK#Wqcw3xmB2_l&u!GV+w1K2Hr~^^=@t-DTvg)Um*Yt(?9*ESP9ok?|H><a
zZy(tH#zR^P3N5A|=v+LIJ-Ua|{X)%;qjq{@3W{q|RGBlGj$_=pf7YMZ?p^i;-yd6F
z@7H~9wEuG{6qF%H70hb&8o7^KTc978j16nTrA=;`$>;Hrd`OPGO>XL=kbW}4JM`-8
z{$k6tjd#`XZ{Ki0`{^hyVy|46hjyhA6E7!MRfg;AkeXujlSE;)OauZ8+!s4dpgcg?
zqB|z{*heLxxJHwCx%<_%ncd9?Gw$E<YWuh^y&<#a&@LWggW5KX#8SQ6l2wdKK>sA3
zon!iP0uz&98wgqF-nXUt&P1;V61?WcD_fY2*sIC^iGGz2zL?1xcBg<~B{m>ySKhPv
zW2Y_^;G(<7l1vjcTwIu?HKzG^5iGdEo%trt*)oYhvCr^tHnOG%q1>AO?eR|3U>5de
zkF%4el;$o|<2%Zfs+3t}>*}fPYL>^Drnmb(@46<7ucqQ~<PApti|qeIN=zI?DPm`t
zW!j`npZq?|&BVYE51mc0Oqh{C<T>~~X*WpZ<4L7^dDeQqrVC2B%Ele@|4B4*h|l35
zdi%W0u30T=zynd^2KT%KVj1ZHMnhc&-RGtGui~MQgo3Y!^2z>0nbkGN6|D`8yZ8T&
z4<_mdc5Ae2ltR0lpq@GyOKA$Tz1^k6JP&2`-GgDUsa@NTbz4_K$>}tTL-!vHZw(Fd
z;vzzOg0n0XE)+J+*!RaF2>~TFl^1<AMh~@t!-mFoo5n#xj|lIv$Qv^1E1CbXou2E1
zBmgj`x2S&hP~I=wM6Ms))ymm%8M*73Q0aAQMe*q}x$|ZBQoFTOXP0)h<N0h^Q@1+V
zG~4TLaWm`XKV(Bplvut%zfW(NhXO(wx$2iuqNp09&ks*BFE5>>N(sI#lvVHNQ*|XY
z-b4YKf9uVtZ`aqHo165UKbnL8_RBw9xgY>*axi~)cx-L)KbHR^Lsb?ZSe$lJRb%`2
zYM>QB=j7$ocs=oGr~M1NAA)F&`{6D&-9(J<|AT>luGxU}fu-z@u_Vg>C12?uq{s1H
z-BfHI!jE;Fuhs*ooHfONlLO-6n_QxDu~<!U(9&h7ejrY0%m8;f$O(3#%Jy`)UtWK6
zNu<~{{x8gGqxkVjn9|)@Hdw$wQH$ESjcB**3*0xaVs%^uo*k)oy5D!cKX>x3BnyIW
zu<HdS2IKu^TK+sr;OXEjFJA5l-^jg?ceQxL`tf>DVErGJ?=lkx0eEyqTXSBO<YHLf
zyT)U#HfuK10Z6kt?tNE#=GytJyHDKrA$qF2yCNV{s@$Q*KuU8qTj}k5{|l?_HwdI1
zbFUxS6PS75w$I?%{1kUR{9lnM4+6nkyIi7{pA70BY9AzU)p0$k9&|C!bXIen4A-mU
zU4I-?S5v~K{Xk}wH>^fCABZ8P4|bo|-`{NSS6$D)dTxGp{m20siYfalZ?9N8=@0iC
z)Dp)mEtwq_m+gn40M0$um)nn2E06A$?K;myh@KSgFQa|hJk=F&Y3YJki6o(1;Qc-X
zdlil<Lm>lNm}#7AJL1ef+C#_H@3YSDg!Ix3-cNP{86Pq4N)jbjS7|V5G|MIa`MkW|
zvFy6;)hYwh)YX1Q@5?h7!G~|IZ`av^-5<35MQGzHxzp3N4@JTWM8o5)$9nfI`OvMW
z4ce|bpLMD4!)=6ocIePPs-6Xe4{+hQ(;PR^LSa+A#eLk!HVU%Kqethx2cCXx*|+}-
z)UJYW#3G&l0Oy|*|5eYN`h%RhTv%C~|8F|`FdKBvm&EVWs+yDk<^TVFGk7KjDdqo1
z9t@4%q!*Od{8`2U58is+!s<-HgR03H?ntQXfaMB(1*NblcdG7I@Mpb?U&q>hp36PN
z7a(vpTnEn|=hTg*E+PgX4RM+uZTRcDt~!&XmFpdmE3bYC03SJk#GQf;{2yvuZQjq)
zu1Vvecg;7^*>2{)UvX4@xjWW>Z<+b#?q)tSdW~PkrlPjg>gGdQH3pkPpUHh$liY4?
zYu5qNr3nJ}gB(W;4#(0LO8OMiCs0{b_`E|UxahTomVNM#T->+Jx3}P}T?D09v1{&T
ze%oNMbCbCje7_dV4DEcwAyRMK!f5S9<h3>r(|a(lpMK;4fH5*nU4XmQ#r7Gz0M$E`
zj`H*n27(1{a>Qs>hTL~s(fq=?K(~75z_9`X{4I;!OE=L=cNo4d(L)yukNN0_*x>MY
zqWW*AufLex7jU*0J61Kg9fR~narlTUnv8hQvgiUGRVe%IrjV||KfCN3Av&S1xt`8G
z&-~#Xi0ivOmWedui9o#D@{w_;<Gh}(G1&4HhPj>uR;Vd~_^3s1YwU7BXO6CJ7$QDz
zn)htjG8_m54C&KY3BqPG(<rQ)L7$hwRWDG@%BC{lm)iTDIq{Zt-&XjX?DLq+GWl}v
z^X`grG2i)iJFSokxfZUMSA9|W{#2P6N>oXv2W(y0{i?mBz+9LGN_(-z9I6ff4^`eh
zDUD=P0i>-z%{3+oC^E{CN^g9WmmsmpOl?qny@=utDC=T`Ap{t>f-1{e*66=AEtiwg
z-gUfh58L7K5o2EA+ADwf#y3(-1NXGopVS6598wzUuCvH$&wd8;IwiO5P?(qWmaVe8
zitKuwd+6W)!D=r8gi)%7^&nXf1emUu>YiGzf`T|Q^<$(w{vg9~>@eZ|Bh@T5WVPQY
zzS@V9B7A6$t83dfK`Wj9BQQRUv6EFLihD1`L1W_~^60q>Mu6!{qP@Y@yHyxij$=NZ
zZZRyXUQCh;sG=s#OXAJ`1wrn)`ImK^e$2li;-v$`KNUV7nXU8-kS&XP4<fDbUk%>u
z!<-E85Fmnm<-MMa&$`L`gI=24ybJQ|Jyw30l^#;?`P;PMtMaD5yw{Pk>02kL>{fs1
z+83Yv5Wq7~>8*EP;o4sM4B~06jKap|Q5)&@f<sa~)pgJFJ27A5*bJ`AKD1UO?+gF_
z!@%UX9r?FB-klhcB>t5pzXieP#c>|G7PT+$$DMDb2hgN+2lY-@7EXz!I{9rQj~X{S
zAFA75JPzjGM@Bf=EO)Q>IjuBjYi5&o^-zxDEv?!p@RJ2LiRY-5UTxpj8z^rMHO#Z$
zU-m~(n}Xgz_>lmA|HKIR7V3j%S$3uQ(>&YD1^f_HpPwBxfI>I7ic#@1@XBYm)$Fz(
zl~)18%=fO@&nx>aov+KC1go6xO^*(x+N&k>5`w1|^8a)o(G>@x<18+5954`ki%B=J
z$l;^H2pDt`(e5+3Ez|OX?Gg7{uFyko&*$r}yZUdYC|bAom-6?Q23v_A;S`fXwoRz>
zhK=#rfn~(3ET7egA0>L*ByxE^ZPX9N(Cb{HA4re*<{5gl7NObZ7OwbQU)&E4gRj7M
z2TIh^WNg}JP|dZ}X3#C&4I}UKG~N1<{$h%#s|I3NM6e_%teq%*p&|os&z<iclQI<^
ztH&-M4PVRh+Fe8D`Lb!}gt$TWxehmklht-yl;x+3Q9EzPd%O2LyVZ|WV(I+p%Va`1
z`Q6TadaI~0&PXgd=i7P1uZNF*=Ax&LS-#T@Jysd82Wnru?(-SFil6s=-u6+!9X@1Y
zzGbd{Za@R>li3;ioF0*{*7R2&kMnj{ACNTOZB`%j{ygXvIy8PPdjH7nw+|X53e#+6
z6gta({A>PIyHkVbTx1~{7|m28z*C+;U|*2`^8PaZPQN=ib#>68p?jEpl~8z-)Aij;
zqHN{HP$e>JI_e{wUVx%Pfn+SK&)~+yNs=;x;;;rdL?~qK8i^T7=}ZaF5`s^YG6RMH
z`X^)+FW3|s`L}h7(=UZWL_wo*4#J8Dy&=z?c-PBmq+rYDAXL5*dYmZ!jMmr?fO7Yl
z(=@+Lw?L?%pxzv%+g`+`RF393DiT0ws`u<Kpe@bp6?S9vY$?WN5;48(X3e^cKp{%;
z@BMjhoDB@R>W=n#*%m-={o^qqNXpkH__!1QtO1^xqq=;aq~UfX!hG7haI${a&}btF
zhFR3VxpCNe5y*H`on1aO`Sa#xzT>X`v;HVcBOoj5#ntED)eD<4AK__o{rz=){BL}<
z@X{bqjwndVO#uJ#^`oU05T2WdqG7Ifq=GjBNJ~S?O2QC0DV#Ujvq-)63ciRCd8h*0
zp+J49eWdA^Pz{1ZJBRr62w*gJ`&+pF%NF3$6!oLP=4p|CJnGR%6OY9)gb|@g8Y$q9
zo5utLk2mV$zekZ+90ckRmFmBWeRYVPdQah2+kW~OFL(jao;2=sd~xL~A^cEf>bwF%
zBL<kfNp*~hoz428zdl4WUA1+xsT3G@)~b{7)kk9_{9-40dw*MhUrvgOx|q2w%yPGU
zPn&NmKst9O+62#K#wO`}MazEZ6X5gk0vtSDw4KwRr0==OF%Hn_nKXTL&&S5IuG2F<
zb75&d++_9fWLL!?t_r#am)PAVyl<LZ@fBRZ2;#JsdNDsO+&v4&sQ!cC$n(CK{krJT
zxPR*`M69Otw7h!e64*#;S$xokLMw2cPIgojT~(Q78z4xw(iVLXo|;WnsSoPoZ%|$p
zv!bi^wnF~cozRzu=b&C@dC~7D0U^7(ppTF*%j)i*2uGwu8l$|!zlf2^$UX9r@g<6&
zk7)Hj*k=nuCxklUx)o$9@MDp%O%o1Vef9+#s3t8umyM$Qy>u?{N~GNURu#-|4Q|=y
ziGe*@={{R3F8$pm+T~(Gf_n5Su#bR`*yj?<hq;IBOJb^l_Uw0}=Y_$ig<uvfnvOyK
zvq^1&<tTK+b}#EE`gr30wu-EId)c)?@6xPEqg}h!7-l1*VE+&>@fEdP*!Hh<7*83@
zn2YV0zmW7Ef$Er4!UW+cHB<4mQoXRE)Lx3ej-?MagNXz@jo(@-U^V!;K?SUM3#|lL
zufO9DwLeyagw>?MH+I%B=VYbxWzqL)lpogIRl7DM*jI<^mX*nl>2w9)?YBr!1lWcO
zHVBgAZgd9rnaPB5h}Jj+&bRKEyYQD)qq0&d&yzoa>TNZpyxC)p?Zr+OQ6OS}UEfr*
zEvq6n`-eM~y(4q)wp}8rh`?*4Is!<hULUjH-4+MybQ1G-ml4QW^{Cw`T$Va2rE7kL
z+OLC%o`?B-1VsS$TqcxC_iU}L!x1*XP}4+XXrqQm?v6uKv4KnqAGtI|X6KL`|EqJc
z%=ebMU2i0j{FCqrCOAx!)Fn$3`LG8!Pvn6g_Zj<q=Dm+y-l?!6j5KkxBLb-s0#~ta
z1v3fV=027D{W;jlt)ywDz*RSkc4SzQ)|1Pc_mMSP4<(z5#>q9kGfhcSI;pP**iW3K
z@h$OkJjDl!wS4l?PwAbH=`}R^TSV@iJLnu|=Ln;#3Wl%?U2?F$FRYWl@WmrK8Koc-
zCi)5<>fLiy?cuFI44{e(L1#$?!C%;zl$&f*2D;!b?yS#AQI)7q8Id6q`qjX#ir<ur
z){HL%FdPQ-A;bDep;1G^QW(UM{oXS8tBh+C!g@$P3Ikp<EG`lcUGo@|Yto9^Odl#e
zr36Ka?#Lz~&0(zks74A6{9sl^RJ5=e2OQ4%+}8wKKCMJs^ZWoT0KZ;4%{qx=-gO(-
z0ImHBB?B~slWF>PK70~iSp2HEn^ZQJP!gHDO+cngHN~I2iNemoFk@!Vr0IC8#Ty25
z5~<Vg^Db(i36=)lXQx)aU|=K?wZ|b4p%+VuTExBPqnbKuX$c{0cNNk8fE64-hBOH&
zix|37l~VKHZA%9;M^=qvAfZ87poY2#kkzko=MpJ0J7;Mi6R#_K5UU9;{uL^;#UD!}
znr6pWymCtl^gA&IHB!&gxD&9I^zgrG;b$^`XOd&X$!p4idYbvfI~j{Nb)Q@HRY<}M
zUjmlvp)<^pO$sp{?!Mn16aIuVVnzBEw2ytp=*)-x&*Wa*(IrmZHUsesWg2{yeCsu^
zSAY>s#%mzC=Jux}3$bP%%d4`qTU^8Reg3@)B%g=gT<!>U2Q8>?%PiZ#?tn4Yjiwkq
z^#Q^GCW!#HccsLNq(;H()bWIye5GV*L(8&Kyhk!+^?;*9r{yuH2T*Nv{Iy34CKG^9
zU1O#u(4}9?^?X6at@wymDdTnYIEOQQZ54>)BMlxUSkKFGBex^(9NMmwOCJW%r`?)G
zUcpKKc~vNXL$~_tZOi&IP0=RDh%^bmBy=IH3}ciYJe6A?YlW^;jQykJ+@vaXQi6zk
z0>&T7u9`k-@=l{l1x3+KajM!Dxgf+t;iYb&NpWn}v!$}sSn@m*hw>)Hkj!Iy(5$ZU
zjqRbB5ctnf$w~A|GZP6xf=O)N4S!oa_xNH~9Ff>DyXAVT5fOQNF91!BdVrZi6FpkC
zYG921VIJNwqZi$WTRwVvm9O=F%>#>&)lR94=6-}5hHBL9X(q#U8*d#DEKq){F--Lr
zn-T#7k>xaavFvb+Tl%l^nCq0c@!}Eewkc-M{6Ho`^3yjWZsDEgaj!G_Xry(tzdY(S
zkLsQ-ZS60h;g>5NUItbB%)nmD<|1o1m-4ol$mF5Fg~V>~A69!uPhUL_l=+qigyY)i
zgv$65Wez8Cm}ua=EdF2i%eOYrL<0+g#W}#C!^9dln<VC$X~4WhTvnK(HhBD&*ki&-
ziN=#aT0T@2shh)6gjyx=oy^4UCu=D{^Eq@V<S9embogf8R;WRLkkSS3bK-G_f|K+d
zi48_H9vwb~n}l}Kvq%$`&QQe1XxH>zh@rQ=V;P@OS@n4gQ)CfBdy&L(!+mpQj;xcB
z17g5%HQ7QZFpfz%a8#rkVqzqNeN4tVV(HvS4nd+z)^slY<g!j)JgG&?C6hDpl#}_Q
zqO?|C^E3#SNb;w0b~4TnOo13ppttg{v)0?q8Pjj^7Yq5&SN`o69B5%8!z3^2B$3&p
z-r-tuyh9Pfs~8M(0dgXf&YtD+jA9$r^RVROxVMn)Y*7w28EO>ibJ+VO7gNjJ9hw2j
z<*(1;q<sY^{Z8^;o#!>hK|LCJIHg+c5e`gcyH5f7*w9{FelA66{IPMksu5O;x6`~!
zg2tL0vr;DKKd@2In3XZh3~*_3CHjk&?FR|wkJayIo}O(PEk`FX#kCec=SR3l%fBb*
z{V`&u1@t5{DLhGj*B#IgPj;aj_jn~7Fei%rf+l?^Bi^dH$LG;VHA;c@nW)J?b}HQJ
zJqD@PWZWj41@ZX$VcdHz5%~>~O0Qdj`>t%}O;d{dR*x{M8KuiQuky>CR23q#zw|;E
z<!s5AYvuSZ2gVXC74J7Mi3}SS`|IfYr7rdJS@01!DhS1r#wP@DmhhNNydpICBr&Qs
zSucS0+{?NA`;Gkb#lv@i<6B!SY{0zDu!;6yai~!&szA)ALh-4vp=Lcm=<iB2Ex@ca
zsc%wrukmbfp$%LeC3bvM=u(~t9INIGM<bdIU`=}IuY0+i2Bl7z7+_F(j<Yk}x0km$
zj~<P*8)%>u<k4pLl9@rSREiX9?YL==HybUSm;f6=Mm(bFJccumK4xL^mJ>1~o9v8`
zPB+>Z1sT&XtvOI3-$v3p>yT%S!<sK$(Pnz$h&a?Y;5&$CGaz*STg7q^xxd7Sh@^l_
zASMfdTPy<befCEZXrc}fc{v0?zG$NTlHD4=P{QI1TBCBrpr`=C9)_s1G_s=C1+vy}
z(TmZ$bv@)Uy-yOoPQpsO5}O_)$40ZT6<5~W6h{u-;OycP(PRwwX{iqzT51_M=xJj|
zIZ?8d94Vab<L}MH4M*e_tg(I_8KQOX<C0PnHEbMiDNwJxHDK%O7;+eO_Q5yZuF+1d
zicHMe(ot47?IQ01ld;sksxd2fi1fY-pyJC>3AV=~e59}`+g@%@Ffvx=P;;IVXYp!_
z{JGthCccNbSZbLh?n$g$Z^9md1(1YNV*Xu=Xroc-s5c5Il^=D(*RLIOn?B+`P0JPJ
z+O#6LR4O}=^)MM`cjk0CzJAhxo=rq<vlKjr!BJC}ab#D)hOTBdpINZb!xFv;Wb`yN
z9D_#`&umyS6q35uJm*Aax``3G!;7}^U@Rc@EL03gMn@9$lEVy4gsrtOt=wfNOj6^T
zeX8iED)9VxHkULzS~CX=Zt)$5BEVsJ{z=xdd+xYFyi&)WIY2KUC_)Vj)|*fbjfb#e
z1_+M`PLr?A@@)tjL*{-LpOVlN?T2bnU8mWzafs*-$<&KPu^;h*XPl|r5(A><{UUF!
zn#m<`zwyHe?J7K|St&WWH)-roO@PYe{h(6Ts6H$vLR~7fiwj~vD|3QoB_2V4e3K&8
zD{r>)#Ws+%2g^0(9(&-HFes3fF^LS`<(NKiA=Kf=J|D~e<G9aTk@b1V&rI*BhXEqz
z{-Jh&Bd%f*;mZaL*4lPC$NhyyV=P*<+h1dNpB|aT1kx7^3`&l(sG}gk;X#tEnNr|a
zH|{)ZM1FOjX(XExz$yMcZoJ(rw{$j6Nw|PBq8dXuUoJKeQAOhV=_LG;tbJuE8iKBa
z9Q4-#F#k{iC_8K(H2<ydkFKN5cr2CDllp_On*AH8;$xR3IGHo?TWe8~7UNkcrS3=}
zuwY($nFxF=YnBQWSM_re>r6sr5%I5U$2iUkm}CVFlaD_l#P@eLDWK_%=RJu>j2M&!
z1rMd<ODdN)m3?qzA~?p2ATuk+=m-^v27ff!u0<j@5&$O)pBROaN4%WJwnafMIjwZ=
z&E4y&CpK=LjpSxG<F^`pzNHBkEyp8s%a?H2*yd5&7&>g!N;@XRs6LmQsvwtB(DBR%
z%k6s(Eq12etIfJCn=Ala7Rq@?&qlaEBW4@v_4}y9i_={u8MY^8LnJ75Xh>FX0Y#Xb
zWHO8pHXKUKAMJYt?`i|O*;{Z41eC2$5B!{;P$B>HB}4m{J>n(mGV5OV)OKYGgJp-z
zV+P~?%Iiy@ok$UUI>>F+_;w%+uscY>*rvUfSbeAzFv{!y8w5sFudm}sQS;5(N`~Vw
z7AC7!kesr7EDQ75sRSQcnMk<054fz3i6(~yErPX{gW5|JuMuh*6<XN$cQ8-aC?ivv
zAb#w?S?&NM{@krbgoB-2d@V@(vh_%kgPIPXgf|1COtw>9No{J23^`2nP<@**p(N_X
zYH&cSp$ogFF=yZ<zIIb6{Q*9Gsy9b{AUl1AAz>O-P~dAWnbz;VP>)~@`}D<At8~!;
z7Cn?@TSB<qM&7*?W9hb5?%KHfi1Qqjdm_3&w@DP)ciMGr_|oHX$J_7S#&K{9@22Yk
z?eKuIQ%L6PJA?&&6uQqc57T-$9%i&YY65-Fi5C>@BKl;flk;>W&lo4s<tAv3)IGPI
zyYZgkPUWVZaIblZ_O}B}^wI?si>b(-!hQ@MT0M^%cG__Uezv(^m_ldE;#7y<usoP<
z@DS?NQHkACWEGw!yc4+=Ds`p@oru%RQRq*^QbmC*<0@X%5euY5%hk+>n$96VEC1}{
z9PgDgod5wU>(b@kMUiV#S<kCn6LU-oz$PT)auaj1{nx-tg@N+d;G8_y%@o@HB^lFb
z^r=0q_p9Co!5WiAdB5=(e!UvD1~|7HIKpzCd4;4=p8ZAEz`t%e56SJyry1twN)+4m
zyYWxYmI9IC8k)HLCHlP~gs(nHj11W&>Y2voZgc!=`h%4Xs<P%@9qo1%LGVARQ4)L7
z7=!#Hi=A1jzR5g|U64jnI{TBBE%a>MF6PJ!leU%Q%4EeOet(3{>&7Geb5>vJWrKol
z%O;5}TaB&;%%;qkQGC_j)15)ULRnu3Yy{M4+Cu?xJ^dzQe0BZ*nA1W`H+nHmiyQ;c
zjJp=rsQ<Jnuh1M0cu)ZAs)<NQ$Z~Wm%jeAjP$Omwv{i8TgP=#{rjSN-RkMZPTMVyo
z$5D-X|9r2{DwR(CUdw@`Mk0M!dpQX$iB>FCZhKUBSff-{<OC4s25#RFGtw%~$=*?x
zTjn(dVNpv|^lP&1x){fkF3<{7YQigJ0%Y+|Z=bTHvX6`$s0R1Gq0`=`iVgo*N80Du
zmiMrU?O=tE3{OODv*DXPe+a#@DZ}<C|I;?*K(%VOqNrje7+(u83AsScDhMF+!XiAX
zkpld%mb$H^m-{ST-aO|~ay8N6PilDG(1$kLG-sRcKoiE$+_;M|-<vMjB!xp%7X+mD
z6bkj)fF#mLSde<5u?HhWIKM}HDIV8)>yu#m!5sFp7{$&KPqSJ{@3-i2yflYQZ$;+k
z1gR+Itf4ylSHGzW!9&3z{y4WLyS8kt!b(>KhvBbbH8VGB`H?13cf~}c>YGC$BuI3K
zXY*&$PtdmULtKHz6nIaChw`@q#=b9&kL|dGDBwy6lXH^xf(UfQNRJhf55KaC%1R_G
zjR}GW8lG_)mNX*KP3y&jo>;JhCD-lmW=&*vVEhZc)*O#|4UBGLbGPBJBjDg5mbe+!
zzK?L-TFb=YsNkkuAJr#%Z%GB|P+X(6nNb+~lU2nr&*JXh-dki3RPsy=4<l?3?sGv;
z1B>R9q<=ph<9gyM&^Ij|hVA6A#h6qoJo1FLigPJ~!K#O0&;gpE6`_{=MvD{#d||14
zj{Fd8yXHfKv9$OLP#B{^o85I0$RV9jLMIkVOnN-^O@g&lwicHqMWPgP6jIvUY_xLa
z3cg6s`Nc|^dTwbW@qMr0Yo3jX2@L)gxlX5;1)y?6LFpJQ-3|(~Do5RTf0*2XhisGU
zCP%v{u_x=DM^2r2QSCY25u^B()x*@MPRWaA$gUzQtJ+*T9&cl!lSr^L=5SRea2!9F
z(ydChb+@GD89k9!!@?f^XR<@yQi*U;jA}8X#IGf;CG9WH{V9&vQ=IEh!kYDr*)e|<
z7xwBFlryD-B};0p{+Q6h+l(l|mAlGnNka>a6n8;?DQF=%l7K?4&x06J^9j?m3<<hp
zaXFW|kH2w^5G31L@V{J|)6HM(gC@Ecq{{;%4v`RxSLWf>&$?D_No&k1@0T^p@5%;-
znwj}1i0T=?cWmeihj`2f5Ksn{PUvtmcg<$I^sUNzzTV+GRnv5!gM%Irv|k>rc0`wz
z+f<&D50()&zDnt_kkS?J*Qx4DJ0B;S6W6oUC9%FqA+ijVY}q;oyv=sV=*}!3iBUi&
zs?z3-&URh-Y_p@5KJbvHl9rQ3;gQz+tP~-n8U_!A3G(OS8Y}%6$>0>QD;Zg<K5_<)
znh`LR@pt#yfQ8e0eCA2SULo1Rdu`px^w(q^b_DDX-k!4b%6+JVkb2A#6$v-Y&4GU_
zk0$4==C1b4Ko72}ej_M}0t~FPu>7YUzj(7wXg;@1^_1krkO7j>%;;Gms|W&=Sd0vI
z=UvtO_R~!HAUjQC9JT{|n9K<3Yz<c%Aq;zrAFpxAguocgRVps&)Kzqj+chWstpK&5
zsRs3I<HuSUCcnM`jf#=9w++7N^H)=^v7#l+!sQ`-??J}{^7r?e_>W`L@GU%K#R@mc
zD|a2q(7;m)$lL{3IO|Am(#$w|#ubl8`dkgw$5%0hg*dKtr=M^yG)SX8vVX;*ZYy`l
zW1v$E2=J=|UIDyZ@7^Q&?a$BXO?$K?$No&y8=Dt>*(<$-TNGW6ip<$pd)6jsymfQj
zKu(>$klLlVPZ`0MkKdEtc2e67h#R=p`x3U`tBc-}!yN8Pmt73=Y2<}pbnX02y7uqs
z?b=;DS?(#Mu`8;NWv|0au^z8gQI}rpW<ex=b=xC8{;Y4t4Cm6`>(n(otJn_NIbU%3
z6}1bx!|R!XwDBj!n}D{doFK`N@2YXS^MY;fq*6;y6FHWtXU6#7SqDZL%GCu>bLm_E
zY!Up7HEIdsWEsF|sIc#Y1&z{;h@`o5#dL>kf+onlH}Y&F=^cYq>D={}#Z6Tqhh4Fe
za3BZ5dD8r29uqQh<tqY?GR#eh+_HNS4?CXf#@rB8#HD%n;pFEetiC<WdCCt`!UVM9
zT<;8V3B<1r7Uc_;y0WX>EXK&DsEE@AU*_94D+GlhTB$1|kdMq5k(M&CH5(&hzyPJe
zCN?9`!?2BLuCWcA433cfMIWIvI}_uOYvvTXxp)LDyiihhBDL1F=--xQ@%dCKIplsx
zKS0Tsak^DX^@s*wJ5gdYPpXozg$_SPXTynAZ&4bqO=<4>xoBHfhQY|^zS=(cV3H2e
z;FXh-GYT>ubH&n`XghZ`o$;3!PT7HY2F`8{uIIY#SA2rtylc`;zx0vvx_SQcx*CNS
zw}5&|s~Ld|U3b4m4ZPN1d9!KU<&VGho{^j@GR2vq<uyly!1V{N+tx>HqJ{w&F^@P%
zaor<t*708Rxa2g48QhaI-NPolE%^bk!hL{dqM)`O;nujlNFXAMUQ)w%vS-8#2u(Vj
z7*a$D5Ap*ZwVRVdEA{%NE9^dm@sgKgyi#~{Gxy+Qgiz&edW?7VC|9N5ryi-~MCBcI
zh&u39{DM1HGC#}3<7$TR!`_e&MYbz}{8xY5N4S4QV5Efb6hTp)x{PAyo3}Hnv0!WL
z(Eu5A={N?@t9X<~e%kR4!X58Q+N$fWl=5AFp6bNc-hM9C*)IB3?*?lh<p3;GS@-A7
z(mTx9r^m=~i+`$9u?&32X1vjF`(0TdS%%|U@|{*qcSv}E1s`V&VT1&Sr8r`4LOJNo
zh4@+8!d9#C+XAATo(c?M`D|z_Uv;@bJ^J8h9;t~+bnM?(#B~tzk0Z=`((AQq_h>(D
z4uD#TrT8{`?V2+Um8RO9j+#*6Qc^}?Y9ZU9iSR4+U49Cd?Rsx^SDQJ-WAJF<7;<4z
zp?<mDhKk#0V|7okrIu8NV=;5_tf!juL#Sx7sznjbXuC1L4CC^Jen8fw31dgkCkt_M
zt7y@@q4!@d9B9pHc`hNOy|j_Vp<5U$4E?eHrrFrSH0;rC3|IT!GrY&|OrP3IdCT2j
z7^DDz%tsxURmMG;E-`sgJP97C<F?g->kns;pLO^f4rr-3!4?5#_SrrBzG`7nW8I^c
zUeo;20?Zb<1;rkC=<w2<=C0~?oQX(W2QQ7Y)7~JA;4`Kij1|>Mgc*^q7R<b64XdsP
z60gsXKg2Q&iMIljcvZ4k_^mQx>S4u6rL+R8bwYds&Q;A>c4uh^$+u_Pog%atuk4o=
zO3w`p>c!*N$@o1Ntt2JbvLZ%Hh~jpEdp}t#f+PvBX*2YOQW~yUoH}{tRZ%tIcX2Iq
zV+JeP$M{>;kj+GHEqty!v=}>&Vc=-z+{diexes}J%XVFzFE~qc{;VVoUN948<gZ(0
zF1g9;tBX|WW#2m52Epg2Qs@p}ZIf5Zw3}08LkrP0X%4SdtBC@PgJ|*Lf^q=;cU}QC
z?#=T7^DYh^>wFy&@Pk{sEZB2tfIN+7C{55jT}X0HRh->4??o(t31#Z|JLMKfQsxEc
z;uccR6jgLI2C|1IwFI3iHZ~SMJY~Cdn=u^;;n30S4ZeuQN~zum**rj=PLV#i#}4*d
zO|tQ}bX@M)L{fI-jqP-?t-&2CLW<A|NYdY*VbpKSE<NH<kNh?n^%;Jbys`r>t4A-?
z+^d->Hcbr3IzYyFiL6*r0<++p!<_J()1u&H{Zb*vEKQV^mG=^H?e{oSGAIq}(_qs`
zB>x%&n^o?Q?sK&f)fl*9k;vs>l7D;UecDhW{9d)F9V8^4=%!)xt0cEi(}Y(g`L9Ky
z7Rv-}3A4ugMyX)LxA~;_=%>l&(rK~v3GJY#LfmxPJ${4$DbZM}ak^CXAYj@2aba_j
z4H<!g$I$$eIEnN?Ek{d(P6*3t#HF$`YuizCj;g$u!m#sma8MDurw*rC5JFmx`SN<i
zwA;Ab>2rY>C$#pXFp+L3p)b)vU1CJL=2P3+d99aVuQz)Nt@+m=_>qx~vZOkArIlo~
zGFmn|n$FRm0VM6l<srVsBN@Ov(bT$uPkLTN;8B;RB=(i2>uy1|Gxu}y@p0JWa~`Ud
zrzcc7Mi%T}5;MgdI>xVW2DycE6E+^chk%hp!OX|QauW(mBa@PEqd%O1_JVn!xIkrv
z8aIc!YGuM1;VG6A*AL+6{j{0mV71l~8v>qB<Q0BJAK-d%Qen`$$PtA%i4AQ+rJtd(
zxI`(EekJesF4r<3^D2c!ps+1EXE}?=kyc|W7<EnH(rlwue50iK11Q8LJfq@Vs@5Lq
znoeK}g{NiSC^)rZGH(DX-rmgbn?KeG=I_b__sC5#<=Fj`YTOQJy0+QIvVr=MIe&B<
z#6yx@V?UWx@x$}I@91Y3Dn)(!*&gSr_%duQDxl?7JsxvHLiAcc7WH%jn)rrwN+X2j
zAgj#>CaALP6^gY^)Gu99p&2b#)@NMG^Qg>1>edR1X)r?8G{kpvNmb%9j=x8e1c_0k
zYzODCSS}=cL|VME$vuT$8lMoK{#ukK7=4=IDluNUewv2&IUUc;ydF3fID9|qkt`}s
zsh0w5(|f6n<W2W{!U@iRjNj{j`ukv>`TlBKkzIGi^S1KvzHd~d;k~n4*5}b3lR~a@
z<W=j9+5_!lwHZE}xnr7Izj7xKvFNV(<-#J<+E;6eypQG<dfz=RUm1Fe4*}z~QO&C{
zE5*9dRS}SPwmVsR`PaTnX|YYfW@YSiF6~YUskmeuc$>bJ7(OJ+se*|tJz(O2Gc>hK
zyWe1n7YE5GuT+TL!hs3F&`h{CIllru{ejv-gn8UVk$19q+!7;4%6WX9lM|xHme%LT
zFk5fP)MSPGj8|qna#e6&lQ=q5n~((oDZG|8lHNGUr^N@Xt(i0VVo!dRigj$3Q((XJ
zBbd)tbIFV9o3&eMN9&dO7q{MCfLn-QFhXLcGkMIny8;dNDz{Z2T^3ce!5ol`2bH1&
zH-w3bh+DuVC?Kw0n(r|$dO%9sv_Kev0h;2~5H+@1b$tqEY2ofTZ$F}Z$@R<-O1oU6
zxMw+Oys32Y@TFpHJGJAbA`-NtvZk^>WO83*%-B&?*Uy1*k>8}Kl_PS_cr5+y?Jmu&
z3Rlm&u-D=}hIw9K-n&pjrAH=WB$4lY(S?1K3lDz0^gQK`8%sniSk-WI7kbi?V?wrs
z%Wst>cOYDwK<O)GV)e>-0AGLM)Pz}`d!(;P$|6+oj5z~F%~O^|c2d&5b5xjAX@PKt
zT5wwfGc58@&EQbJe!B%no_05OTHai4Qoxk7x{T}(httyMyopjR*Xq};E*!cxj||9~
z^jTE}!cjHmB2b~*vc98-!H!n!iUeo!OFzUEV}sJlkh(2)TNE=DTXru2HJ*<<HUU=1
zm((z-czX<=`@0l{zxFPi&zDF=(|DLEC-tpSj4?AlL=@8&_CL$#r>}s~Z{-Y|7@ka_
zv_wUnm-?A-@n=NuUKxUJCc|u}L{u7Kv_Bq|v8gC*GB#GXh83BMS5Ir7A~*Yn(9@(U
zEsllv2ml*?{RZ|fK^z_htKomzNFJM3p1I}n>Gz=s;-a5ktxzg7*2V6ayn}YoKGhs4
zJpIrb!l1|+Qxz$vE?Zk~&|Zx+N8HA=yZ>u6p%QTz^%P?a_>nJ%$6ahCnQ0$r7j$7T
z(FIxT|AyNmj6+wP`xcPd&@di(f9UX9tT#a5Z=b|H{#qMnudXh_mn$C>(QL6>|HP;_
zc}|yaSrnrhaY^!GPUHABC=7h{YiErD^It6#u;nU7NN09pxsvGopZ3+xi+{bKA7iiv
z*%1tlCTNzzy4CQ)3)^NV$n6#_fk1N(EQo}=j~zt0s<Ou2Zo=pxE-aekYAA02Ki@tH
zbVMLeemlllz1+j-BR*fWaCO#QUE!>Zm>&~2@(VgRTJ>~uV1q(W0BXuYUx>O$Ak}D)
zEV)T{`x!GyV_umWG1{KUE!jh0f@ZZa)Yrhib#kK{1&F#uqUIXKpzS=Ky>mQ&-)f;t
zwhENMM%poSgwpGX4NIsEveLao>ga(PWP7clU%?kXU&IJo=Izu*E)^-RHB_aK?H=N{
zbTN=Rx6z!n<iJbhMEUZ@5JK;~ZV5!P6Q($JexUmF*k=d^;;d4lk|D$<61HhdN#aoe
zxRvf3+fQx8)_8KImIwCTrExTsO)K9U&#$6riTJJPspkQ{<1sHeKY2YB7HCAIDF51v
zB%b)IvwP1*8gi=W>Dnk`ZCCJ+-tyv<ea#S56PT$?(md8%-g_U?i}`qDx4Tl@izMhd
zqqm)8YxY9CYk52o7UuTxFDiPP^|e>_J#bVfRp%PI*l^Ai($7PSB&%h!D4V(>%NJ2z
zvQ<VyC7(rL5SPZOmF2f*9T}MO{?_k-)7)gvs$Q0RX@SPbI<{leg)GKVAQk0Rzh33C
zR5zHlpm;v{ThFkL(^>wKnlGp3cU%%x4RjL}Ff*qa<lj%d9nd`QJxPK)?z+MM*0e;C
zeR)s2E@Z>`SF-`Xrf4ub^Y$IO>6^|cte`;z5ke!#*yLh41nl5o*LngLD}w6|Q80)E
zYY#SnDXhUJ4B|DKI172({Ac#<b*Ml)WbkuYjI64;X_FqscfoMwEds8=1EPIaV5~nK
z#V6#W+Y@ua)_`WkgBP<J=4v<nOd%XvK>Tm?imRDIf%=_tyG*2`d;yyuzgU9Ng*0P9
z8xW~5HctsSir}wP%m6W;R3C3D(!2)l^G+kd0>epFifL^Vz5~N=me4ybhnSw0F?x6;
zX7`T^-HvJG-GD<61qOV3%+WYFe77BU%8TVSHoKyaN5Sf15`3VI=?w4~yY*2P<3E9s
zbdENQ9no}gHY{P69)s_KW7!utTk^Ngwx~$#Z*|#92gl@Z<aozng{~}6)P*A`yOC#`
zUWlgUrKsEE&3e1|ZXD<iqH6O3vjz8JVR{Y)0)?_ypnZ<=^bTXv{uqv@cvFl+)eQn`
zQk=MdTCUF0uBdAgoIH549{Q*--AV;Q2vwX(-pyy+lB*BL1-^}u-ph<GuUT3DpK9*<
zA*!cs7`Pw`0+OO24K9t8(k#N#%@R@)5=)oTETF^!(hEz2q)LZNcPyopG}0;E9q+F9
z_r34?dH#dvr~PTqnKLtIX3orYeUL){Lhf*{X&5S*ud8;|p7eql;9-5kKZmT#Q$2$}
zvObjS1XD-8J|qb|K}Ui6+T?hVBYMB;)vlof;W^~IptG#AZ;SejY*ywfcZ#+9vUr%r
z(`@pakTlKJ?=mPZvP1@;VxT<%1V{0R#=cRZIh>hubU{R8Q!n7QpG%kP<rBL?_eHI>
zvY=_A7bNVOc2YROzvtp(lzufRA7|)Gj;2a?+G&OX3Vv;ZQ=}l-ty|*A4J`_WyMEm5
zzinv`-dW<U^L)%b=GZIKobPJ%I-wwqN+_r{LKMidEX|&^JGt5SzH0vk;&pg2*<l;#
zyaI3=r59{pWQRxfWbiKzbS(Sggy<@>(3mB#2$19K(%saf&+xS(?vWEpb^F)`^3w^%
zhLf+g@D4E>w-;E99<V<uPNY;9aOK*h#%0ka=UY*qoCQ6oSk#huk1X0(_qtY%($Re|
zJt(9Y7T}^UQ9d}|X_c0w%;p+Deb*e>?+HTH6q%j{@Y`}t!L8B^-1u18+l47k6r?^2
zpEn=%g2Elxg>SMHz)XAevsq;Vz5CCBi`x`a5l+QSBM*B~CtMG7yNynDsXD2*=g0L!
zjT{qf=l0DX^0j3yi|0f!-!YE(vT{aF2c80#{CQ-7zm<4+(H2-Fy|ebl!%+zuRbmdW
z(^`g}(xo=`d~{jd;$<Cu)EaMK?AarJBvi}DpO@c1`sI=3E;SRrQ*r9%$*Rghc6Fg?
zn#h#dRZOXFaxlJbCwNW=WKbJ#8WI*rNaeYq9@REe;92k8q43gC#FZ*}75kJDw1Mlf
zG~>HHkR}&ps85yBexxJzx{x{0DRinzpslDzE2GEb7-iyi8z)1)%8A0Ou(fJ<%Dkaq
zGk&U@P5Ja<<cc4dIU-V8qL+NQM?U-Eyy)GmhL7$+YTaw#F%2tSj-xLB*z5tKI>J%t
z{&z1xkb0q%Qifu2qb)vGwbi5aU$7@R#Ee|G^XoiMCt+n<AhY$mEjo=nNlnCN`n}Oa
z^a<M%e4j77lZJ513ZI+gNi!VD)40gcE3ZKD8ivF^Dqf;<^y;s6QY3Gq`*#D`<xgcH
zN9C_`iRFxSs1LJ{pcXyfZV=rt%p-@mRPCIg6iPoENo!0N?nf_O^_x7zirz=;f}~08
zceR3K5W^pHvQ2aFra3>CIWJ2hFvySIxFosYDRvF9*%%D=H9D+-SW5iYyUN)h2uS0U
zPcho_S$rjsB?icn_|aoU;5OP`1gg?6_J|%XS88IAoLx-nI0$2)nc2OykZL5JyO8vH
zG&$fN^Kijvy<MvzXk23RP1dU;QqeJ3`4nutGM*}93CL0c#Bru4pIdK)Hw9r2!lcK1
z)M^)IR3~{K6gixT(M7_)Uxo@&M!xnnBYFTp?&a)Hf;}rP(czlyL;O$G&A3ioq@fFZ
zhh#VWWAUU5u~0Uc11WhkjA;@U_p*SXi4X+Urt=fHmDBATlE|t#a#5J=;NQHrjkofy
z19y3!>s{`eG`k5xvR#4Atqo%K=;x2jETdD>VF_FHEpk0w_mGO|Uyy=`Aq21R1?If}
z$MqCd@8Q?}*p)Kky9Lta%W&Uk{_PAj314pGMV`|iLu4F}veuuo=iGGpk}Rw#Z#*s>
zl%SH7>1CuxwqHp_56z**7-2Nv3uLdZCAISE@RVbw^VcQrey=~~NQ*G>lHG@(>L1^Z
z^>>*Qk$k$y3*H#{mAGF=yt_2ndO8N#CMYE@^?st*+T2mu#x5V^s3%doz4jdM;ALXt
zmL-((I;XBCy5yT=0paUY=ldtI8kR@)mpjOm!zupP7ZMST*>>ZizqHz&cSap_`K|2c
zo<4E8a}b^ZgHZZ219{!s6h$^SJ!+y^t$wa2nMYAQ7!qZJ>dMui6Hez~@)kO~1e{fP
zJsi2y2T<a^Dq?kee-ss@sKA$4c3qKFRLAS7f0yems|pjB<>>AJ?l=)vgtVfd%cx&Y
zK*$q&Tt>>-bAHCUXz<gF^x4BtDkv{cA$e(_cq){Ox%BJo!iP=f_XT)|d(V%{>&jPE
z^J91!@I0_fPh_tNdDQ7=r-9-ffxQG-m*@-@&u4q$@>zJL`$P_IPM({H-*(yIcdIY<
zY_|ui_I4<6m5K>ue|qwL(4w4v5Kl&F+yrcxY&`v;#3BiIfy#UOL!d8qrx)>@U(6dG
z=IpmxXFZd%?*06+r)EJ#c&BFR0*~kUfM->DDVsq#$0Q85MhMccbQaIdcj6q0DiD=D
zdrN=Gc%kRUg%XiBqU}q229NL9Kb6b=l)X0!VYW59LeSo*6Lq2Xhb}4Q;Ae1}tcc<u
zNp?rVEL*N&Zl?G4Fz@e@ykl>V^P@Xnb}oRYLEmx$W2pVi8#)#u^BQcTn2Ct{0d<<2
z7T;Dj2Jo~vLfW6@fUb{Ba;4}Z0r>7kHQT)7qgA{(OdWPX#iz2?>gxNjg&HuylWlkP
z;#75%+?FDhZ+?@qK21<s%MvPzzuS7L*H=O{8Q@`Zl>H=fsyo0mO_ksFijFf}d83c&
zbJOa;z94^uIkGgOd@9<((EQ#vbhs~2*De3)cenme6t_A*y(p?j{R{|sOYCOMW$m)d
z#sAQzB_4Z4-H}^Y3MHM0qA-U*te+6e2ag4Dg~>wHx%&9^d6$Uz*2(FupMSL?3?Mrg
zx@uUq!%HD6Aq}+e*{!_kI{176Gfh5@1)O#UX`U=iz~SHb@g}r6f~1VN!WA30TtX8P
zXe-o1Tix$RJ)A(a9<OA_@p=0$Z}_*7XJ{;jbN7YUQV3F@T7HaH8SNU8_bsRa(fM#W
zec;m{_kMUX`!P`mK_3{F(-4~QEt-iVM3>%N3#gqLt9-9HNzy}~h}Y~S<=$Tjwe-%y
znx!B&)So(y;sv5UJ!i~7!3OUx<CrB6w<&PDv@g!|a4ul+B_q6}c#d~9)+dPDgo21i
zobP1`WLZbeB;3wK;ZY7qCkOQhow~<+Js?(3Y2ORnXFt-U3@ej8xN{rI_{7%yhHo|z
zZq{b9Rr(^|0snCdleqQU+s!+79@k(jfUey(qj=U1{8DxkJY}{N+10PiBDMn)By&-x
zU+ShzjRnAy(az`uf~dKw<y+}2`yY=F!ZOf#M;;Z1arA9z9R4BXSzEH2UlI<!3~Gs+
zY+6dDdZ;i}L>LX>I>W*w4^S^XW7YSDiGhhZ9Q47PUttHw>4N}9v8lW?n`;}R80bdO
zZ#(0p5LgUPMW)Kw>C1@Tder521hK~w=zOQJ!c!!@>+nb@%b|DA=4u)w1H61;Wlmwr
zWP~I%Z?1~<%?ROB`91_B^|Z}#*?zO+pq*cS541EV!23~^Z4*GFE*7ltTpp?3wI*))
z=T~nI$c>W25V}Pnkg2fsjGZu<Po{00D56a8Sjc_vL{pyjrDlFaA{OMbXyNgFPsI&J
zUvGoCh(yMSQDg8Kqxt=i!u+q=V5Twv>AdEm%~IGB1s)4KU7u+R?NlEKmDB0d+}%O&
zi2RIfK%0g`N8Et(TeanNs0Y25d2qB|e}+)-9iCaXOt`q&lAE&$)DZ!(YRrU8aiqE|
z41{y|ZXPz3=4LD%pZHtq^M!|%H?+HoJ86NL?tBm(i>OQhL)~94jo99+H3PKJ@A^N4
zXgjV1R;i&kPD=M8mJ;<9lXdS)yF9lHa-D=NzC3m*jOF&18;;a$oJXd(M>ogNMBrfM
z4AOB@W*J4z3GUxI(2rcGG<by&==5(F{Z3xcSwT#VS4K5Y%^D2o&;6PmM=V@6d~;YT
z+AVY|f{Jdw137%t+g8aOwk*mvSm`P>TYuPCP*fBg%WL3vpt2?Qw83LITK8L90Ee&b
zJ^Z=ipdjs#cTUeZi6o8&lh-a3clC{`T@*TUWlWKH9RWPX_SB>f7fZpf1dsOz>y6oQ
z6!uFF=?ERXi%z-Q)~L#3;cLx~yvT*o)pr@i2_(BSb|#lonz}>|nXAbLij%)~R>BJX
zE4|^rtK%S6wVCpd5aK#)kK!xX;jHz*-TKphgMln}pOX*7zNY(*b#&w=9_T<>kMHH4
z>&eAO)WCL=hIsK#?pjn|@8<o)Z3VX?DhOxZ#g<anw{EOdUjXTX6yinD??O6*6<wh9
z=12+70b$=AK~DBxwoK}RU>Kf9N;mFCCxPMlyItXr0jv#U1jd{Wry6g(!putt+3eVo
zKWAhm5zREUJ7Fq%%lQ|+-&a>h$Vz`vX%SRn^Q>^Y_&DuYVjJo})ijVgCD?z_WY9JB
zr4lGKf(Re&Kl&zfS9f|Zn4%+#g3Da@;GQP7Md9d+!lZ>Dnu4);Co*v8^ZkYkrL(d)
z9A1lBu0YQxt*Vb}(_7U=I1{5Xl$3BD(_kr=NZ;j6T{)Xr)<YPv%YoVOWec<`OH8-)
zD<($P#xAu&KkR)^`3bt1_L)7*kR2-Pkm#eAdi&>1_kGzsGEg1|>2Y=Om5qH%R-AT*
zw9%?mMR7OwGZifTxWrnBkxOP*y-<Zp@7MTWEJ|6lZ3LDd&1$-Dn6*ir@UbF(76S~)
z?(*9-;ZdG$fA}FyzA5q)<dpy2d9NEUI7q7Zxn9@dpa7^@v#gsIy>Ems5vf`i=-cwj
zrgHx$T+*D`w}(!zJ=bCl{W<;u9c^xn69G!L2sTi96ZM)(W2wW79U$~Jps$Jrr%;pu
zwM&27yr(iS?xmQN2FiuFnYs$ndM(ESt_1>neMCTCdJy}fDi^oQ0WFH^%e!GMEU#K*
zSasoX*2oY1&-|us!`~yyZIB&Kw)pMAQcIsZ^pjm185zV!rlm|0C|O)LmA^<3kUoo`
zH3o$kDQ_aPhnwCpH*0;gHLCLth0`5?i*;!8z=#_GX@XC&DpYS9_GZCXx@5QYx^5{D
z5tf39S8w{O27yYSlOXKUovNv?_S0Qpr);f+PMn55uv2jxKRoB*P`A<!>m-?m?+rQN
zawceAbpG~Ik~U(F`FGKg80?d`f({p5Q;Dc9&{PF^W%JT9<LXZuVAg^iM%9N|vF{rR
zR8Uz=rN`sh<6uX`b=w-=gv<^wK69vqn!Kko*V(vIaMVew1Py#|9^ziVcM_-wQlpc3
z`!(*y7XbwjvJ+`E{3s(WZv8_QMp&m1GSp;yJeLDT#~>m=uZ;3X(lMz_^r?KxDqjfs
zTaN&0Q%<d>c{BV`L2cR<AZ{j-4V&B0Yr)n(@6O?JTIi%x=|xA-!c(*J09<iXUbka|
zMbciM$x?7-K=8qASA};NBSJ}lum|Zo6M~?SQ|$TdF<}Nf9)|Z`SZzIR2ak5)LwEos
zFBZ!=^h$b!gL&YqpI!;OD5WfMC~v95T@<T6Yj`W^*7Gtc6*yliEE0y2Rnjd`176<t
zTj6%##^UAX`su)&i4u;Q_&?|XV_G#*<_zTZ%b6HEjXvx<>W{a@Jl{RGx$5n;RB!3u
z(n_-HtlUFW4l@2nW<{qK-|XM<n=cM2nVBA$CB<_ExDh*2UI2ccafv^@@5Qrnj>*Ob
zN_X@Ei+NoyoYx;u8qEp(fg3Yj*G8BwI8$4;!Xs5zTNN*@SrUY{J;u<3cv&6^Z`K6K
z>Ti5}>mr8u6WP^In!Vz4s;Kz(y?Mm%4HuFT=|vJwR^{RKu^u08$#g1dX9wJDw#k@o
z;bKZrd3k&t^Uh;0(uExr>_5td(jCq&KvV-oxfjv~TpI^WvR4i9sO@`$KgI$MccJ<I
zny%g7#nY(vF(bAtdz!ZKO_LxV<(%MJoX^X9)j`Z)ALZs=g&6BL5gruF-A=uPCNc4R
zJMaUy(7l4V3=v%=rCVh;yY|M&o7fM|5B%lUU(@Tit@dY@Cl>gR<k0C5Zg9}&TI*)+
ziIyr(SH{SgKE|PBIeOw9;T67Rp{<QusC`OP?&mSHPuag6cjEgvZZY$&A{pa=0B;&O
z%7%|1cJt`#hQ_=n9A#o_E_Z7H4)K#E1~_oz1j&Sr$D=fa!z@)Ss7%(wt{<p*RwZkR
zs`cebH;pkNT<H`-OgG-j=H=c`mp>P*dr(XgQc2s5a`Z{~UK_{z=#P5mpD7jL(rXUH
z3ir?_`X~V`r49RqQ0dk*TP|&v!WgI--q=Yl!$RWQVkCVlZ8MYEea@Lv4Z0uxx<Rkl
z1T6!LUKC1~ZBdZguDozIQ(u1QMTt^W#LgOUq+tCz{ob*(+aN}iZ9q6CUpPLW46!NW
zoMnlU>C3#~2~%Ku6)c&Tl8$ThjwW7j?*U9&gqzks94X}gW3%zlS{YGWOG<*xX`y$q
z?z|O$Zi46Y5F`$aVk1$r&!m@N)h(zOUiP-4@*Y-XO@MwlA)ZR5mQ!0OM(pM@D&&hb
zW)J~<;Wy9XCo`(`PdimOn;Q`!N}uHX=E(4pest*cG4t&DM>?QX1Vh6TJgV@`QO3b#
zSA9T)KGpk1skS5aFi&+Kl~pO6IRkcv!>@A&i9HNC3=fZm%*|#M@69`KP34doSH{su
z8I^7+8%A!qkVw)ln5<l3QEzl|k8|;QmRY%He_ou-J!dwJ=xY52;M7QaZeo$`$4g|6
z%-Op?*`r`4PMwMz7%Ol;W6*UC|6t+YjS*AQl=)-$ObY|QLFc^pA)e@`G^*J8qRF!?
zo`+rwG?JLU5MbD2tjGfD(#n^z%J1W~&$rwq=DOsPe*!skMcYPo!YmPXYJjNhEY%Bm
zcpOc;h`q~q(0P06hB&bj$dc}>RWfz>olZ7sPyD5QC2f9@LqdLVQGqbYFbjVOzMRQs
zSWS}$M3Q|>FqNhFA$HI=mw6}VFf0z}Gr%CF(a3(a0Hvn4!=Ma{4xE|T7~tEhCv|N4
z)0<mz?b9952eM@z>BH5_3idqZ@6Ih)xbcsA5{UPt-?*h7tQ+yfl0OnQN3=GQ0US#x
z`r4`=V#m!6YLc|Kgew8$KIXnEOcz=gN1fPau64v92M!ClbE+(B2M)PBK*s7gg4Ea%
zBy>jPWtQlV{d|EtF?*%I)WnK?*Af(za0Icjq%($voy?j^9mtimwHElb(=N5b4p_Go
z+*!%@W*4;eMv8op{pTW*IFA#EM>r$n&yRf&;#62>u$QVr=My2%sNrG)kgct>%8aX&
zkltm!xYvS=*pG6asxdkKGEpwKspHBDABZ9ZrI+CvH^@q^mpg#`(4`{4K8qD-$?ev#
zIh~*2w2+mpiM^}8ztOMKka`bZ^e<|rkFh0|IcSECqF+btbz$^$EE+M1YHSLgwl{yW
z0(-k`A0hZdrZRY1_}lCx(IbFz;SM={CPvk=gWbUmD7%;Xyf_|j+xGN!g1pr{)TpxH
zBddft$OHTrs+gw0g2440qFyvx5`WVu+}`L`1X<R*wmx>KF5{sVpkfltc*s5ZNz74b
zd+LQvFaTq>j1ghcNnL7LT*v+Fv+6?><&K=$$%T(5CN>@A0WMRZR{uh3>|f)o^B)gt
zDF%eZHQPV2cNMsH|MQ1?R&CZW1DBPc3+oNw-9x#=?3$NKEevD7dWwh1w4xl*9)Od}
z`wFkHmJ4Q69ULAeYWUKtE<geaO5-%QBl*r7nj1YrYU=uG<$^6hz_R~eC+6&yG^X+U
zOeWKN_j+*2SNrjdxwqR0h@@KG&LT<qVjj|-<FxUR-P7GL8%Hz5mQu+v*J-E+y*Qa-
z2-sadmj?$;(~rfyPpZh6WUwtw(*B7_CGS0dx}86Cl6;l%*A#Jo^-50^!U>gqAF5-+
z$<;P3VpUDPP4;ls82w0%^^sE>5u~Z>mY23Yt8XsVQRZU7X&Mo^>Iq?AGmyi;)HyT*
zVtS0H4eC_-qNDnNbBnfcSu~oFxwAK)(%wghreJ*b32TB!sqYG>^7=&Mj$7%4r{7<V
zrp%Jm6U`~jNGy^@vqg>NL<8$&J_!4KYF}$bdjs<#AVZ1>!t1F5iTa;YE4k*3Z*Hs@
z^<NN)@3rVV{`A2BB@}qPf@#rwVMd8Pa{L5Aur6ABKgpGNtvZb1FhicPhjXj?FO#K@
z*))?|LUy$@CrY%STRswaHXjH9Lj4K@j{J9ui-)uA<^S!Qys{2#W@MISE~*eeh>hl&
zybx5)smT6onrz`<kJt9Q^=5TgEWK{N7_Int@B;#@tZ%2?Dcc@?1X&+fBh2-EL7Mag
zn;xAjp?vc`HW`i2NQ~?7KKSHwy=h`Xv>~=t)_xF??B(W)5F^J%U!+tiH2>$R?i-A`
zf|gh|@D${j1hU_0Q*!!jBVYGe)zJ#X=Vw)n%KKC~X)!U7Pkuq8GZi*!C($XiHW`uS
z>H!~)&<U<zFsOva2_cQoRyih3uHLIjuul!=KB;Gy*sLZhTQv_%-Msp;l&XQp|F0Eg
zTF7L|lR3Jo$k}g*oN=h<KPr@EZf3<M2nYz=dv;HFE$v<J{kmgba7Qcfq&#|i$M3X0
zQ37i%!^e>{Vmp8JUI0sBStLo#67BB940`u?%RYauAA)C}^L!v5-M}rB*sJr~_PGJf
z<3nBdWp9Kq3-DZBzm~gngNn7jXqLgYWYmT1EQ~cnPkemQG~I*oVfG=Us@^$`b5H-Q
z0#CR|M^JJhTBv^VC!$f(=^trk!W$#i9ACUJ{N6)aGc{T=hFPU=4&(2aVh|t22bV<<
ziNZ9yg+nxn<Dy8Kit4R<4{EZwgLPj?0%m!8eM(=$?Ekk{{H<*C#YTOEr>kv9%ED&w
z<&T-B#kruH;yoe)7$Qqwu@=pcyq?`{M6Y7y&@b))uvppnATGU~5&pNi6mY8OF~(9e
zaNscy8#WL0k8(Z6)&NjsMbKkBjXF&G=S+#F4Wky*v(OR^vVNlTxgbZ|O`@IA($k#V
zwL}?HO_H<UYrxkL?mjxMmQI@905qP5fpadH&^*q?V!D$u4hxerIt8{qZ3K!wgnWPU
z?f3kBA<7v}!quY6{tYx*0EzH3{Q1NZxMfCc_%r8Rg{iCr{Zg*52-PZfIVaAWRnNE^
zj*Bk${Znee#~A<V@vW~96R-3<$Lm@p4|MlzPD^gn$oh;tf@fakiv7ujPi5*}rMD%b
zKM3yQG@*?*cknyWEd4tBrMYiE{y*6@jf=iimAVN;SqKXolI9|}79fT@Eh3f;U^k#Y
zt?5Ea7N1?dEcN<HAqgjbhE4sU1;dv~H_c)MmvOXWQ#}`IyDat(CnQLMDHzF>uN%`K
zlPPkkGV|L0flIm9#OjIC+aoHDXf#i?seYb}XK#bE<GUu7cx@ovw5272MoFsP`&7F8
zV-Ja8YAz~inJ26zgki677g`E`K9PT(I(w9FsbdM^h~rVkL3;>={%9lSP)W5PjjQmy
zimbr@=kJ@5Xv=6b(zA(&G%273qu5o1&8|?-tjnd}-8G1-JHtp$JwYQkO>w)J=>Y>R
zngXJi8_~C8jJe+%B9+Ob_E&>q#CTqeb&Q;`c<%vQeJtMJ=lm`c?mppsUg4LH!T8V_
z`b}CP;qaYb8sEOI{+GhW#gCE9Tz4&Iswl8AcA7`O$l7~diIVquoZNiccg%uU&|Au9
zYG#Xj4gW*@6TFM*ks-#tqK;lTTgvAsp(ahi;oaujoAf}+JS)ovG`t7|Tn?@9_@D1|
zGe`es=du8&$ZaeTpKHVZ=asNAje&ja3t{372@qfkg7OhsoRV$0{ow0*@8sD|EsFcf
zc++5f<FAD!)UfcdE5m5|_GTVea@~GI9>cYq!z1l|n}19-y;N5IFP+R(9%Gh~!mrRo
zq6bb|&Gs(Fvpl^yZEmg~Z`>J$qS?uH3{}^!)Y}j$b@bO&n&<+KLDBBjj+(^r$d1IW
zm%3N_&r5k)H|@WY3H*bi@Kcb@_#2P(*KQTR(Szx~bt*aZpuzu<sTtk<U@d4-rgaBx
zOUR!l5BTO^q9e!*BLZ<H)}?05V6<b~Xb<jPVbW+Erl-wdg6v%fpDit8xGAs9(R2)?
zGxNIXAfY0K+46wE<g9H6j8+SOZ@}-d{+F?)K{^2tFMwXFW9GUR6eGh?U(XQ+3+pqB
z#>j?sXyekngre<6a8Do#!F*{IlxImf;i|B5{GVd$5nwPuf=87RGNzFvrow0B*ogmD
zMO)l4&foU$%gK}f{ORu_n@W)?RBwNDSJCDCPaXVgdg;dupXol5(|>3G|MpbbFx}0%
z%L4h&5b^Jy?FbC)ReM^w{l8(akC?aY|J-d_ZgInicc+;g_1!S8t)i@&Oo_DV$NvXA
C;<HEq

literal 0
HcmV?d00001

diff --git a/_images/raft_idea.PNG b/_images/raft_idea.PNG
new file mode 100644
index 0000000000000000000000000000000000000000..b3fe93c864ef87e848557fbba7c0f42943a39b6d
GIT binary patch
literal 121926
zcmeFZcTm&a_wNgWfJ(C>9R$HbQF;%c0!meybd&%hO=+Pep{Up>QVm_Ibfwn>EOZh`
zfJhBW2@oJc2oMs;jgQaodw%Dff6u)$XXY})3_E<n-dVe?^?t3zD+@D2uHyp7nV6Wk
zZX4aa$Hc@g&cwtb%fZHYMUBk$=NHU@_YC!!s*yq~j0UTxo~a%a)91t!v<F8S?PE`j
z>;jpX&RzWT&3yZw<ZmV>dhzX>dRC#%YX>NAp`OD%`Yc`YjZf{#$+tftFX6y<`M1`d
zKj-_bcH#|7<FRj2yx=ox1~1Ob%G7_-I}v@3E&6Ed58ekIuhiJya5#N<ezNtvc&~%~
zd9MdaUDr;0jXGE3Fotn*!eFeOv}Ys;JtXEKb^a!g)(V0xwxVL6&KSvmeOXAvB>mUV
zOiYh&@t*omOSg(c|Hyw@AJ6I;{ingyaDn64e_HuAtm4f7X^p-Xll)I3%Kqdt*8j9D
zuKu6stW@Ng7UZcRe1Bh%vWk;mR6AVL-~WAYW0hnzwKla^H?X=;=z~fe_|X)*?hUvm
zNz&XzFso(AnK~srDVb@L7!du1`gcf<_N?M<^+0OhOC>n`f?J^GDkA(QZ|(YmQM!Lm
zm!WstcJzdfI*pXe3D&-Zqqpk2F3V`xzkAiKirx!xv_XX>y%3XZ%{JiOA|(tuTl(Ia
zFBv3mPwTtL6#Z)+<i}Hby~41D7$f(_HhG<H+*5A0>O@=M7a?lBPmgGZ0k1;99?e5j
zmVKia-R42_;y+69b2T#Xk01Dw#p0Fw4SK5bas@&X7i?|=_%6z~e`$c;n9$YsJ#2YB
z2#I*^<IY9`0pTM9xS#cFNayR=@oGL)VVzyZu7%3oJWCFJ6_**~nxAeZqu&1+(v_Gb
zCp0Vxl(&BUbZxRWj9#6?=7viAh3Y?!0>yjt0ILR3?JlX;3B<37o0V+h7+6vvJ#1*L
zzU~t2SQ%+mo#)i)XE!q}bB?J?oroCRf0HEj*Pd#HcVv4e_eWDbu#IwxoB10v&pWcA
zwpq$jAztafB{=wpu5^eNcjXW6XsJtZ>4)HJBqqS5dmjG9bpvpL+?1J4{&#UFBvx=M
z>zd@IO$5a*Isq`Q*poW*YY>X=r>Hh`gq@)uzL!~Fq_s`j7kh>tIRGuiCVg-}O;(Q=
z1AZu3L)fR5<rxBhp-9?O_q~OE`?gY^wp2{2#4P<gME}R3(nV(_kKcDr5?i=<Zy+kj
z=<sphVV#;wneSFjl)H;L+AedK9x7Qlf;$u=5r08<YmW$9XTTK|B@qkx+7`1g9{pOG
z#TCuddc6VQ<T_-kqQ}PHeKIF>fn!3RRowhKM_KmA1GY)(b5>D%!Nj7IV4|s=h7n70
zJfQJN`rJj1W4zC9^2H=INc_8vqDXJWfC%1aS~Frr!jYE-9b_kHZlBz6GcBT%^aHZQ
zgv?w0Q_omA^b1bt^-liRp6U$SS&U7R7IhDFMi4KB)!@$Zw<yb<5M+*xla^JVeT1N{
zP=V(h=gX%L)?t;xtTKHopAOUx<#1v<w=KDGS}A=Fw(gfSHvEe=*I$}wIlluRU_R&>
zk+s`ua+e}e#uIV^-wr-WSD-}uzLB^6r|@(uo852#)PdycbPghzGU7j<?-UTp!zb+J
zbClW9p)CK_qeG44Md){&%bCc%`U|HwL1yi*Q9lm~-ns(<HQMxglRAetr!o|3s#lC8
zbmD5*7pdNE3hD^Co)0yB5vtOFZm7d~pz4bDB}GAX!0+3O^UFD&w}tZ)@wtQWAMf0-
z)N6HJ+)jyuvx;~nB`Q5_%w>{JuMm6_@Z<1m=~6|+I+$HQW;iE^-!43~rtmE%SJS}n
zjndH4XR1aiOJqsYMAbD7gRyvkIVf#?Y4-J?eX#5NRMx7!-}BO(<b#AFk(up<zG1N<
zNU4+O{<mNnV-JoW!ZDfsOQkG!n00i+htTf&NwycjZM<Qv)syLoox+ygdnJ(92+z6P
z%AHf6*NfkN%VgiwJX#fad-Vy9tZXINK$pTGTUXy^w~xW%f+vQd_`ba7tpl&p*$9qV
zg$e6rqVLq%at!1}`tE1<2qG$j(vQu!{CI?I^HP80^0f>f=JgJO2qg64C|>AWc@^dK
zU^{aHeJ@k%@^WBmb0ay!`|sM$wsleB2>UUeJmJ043C62jDl`w7Hg<!|raX7(i}IOO
zxwOm2eI$K+&F1_wE$0k={L_l6r{*?I$;_Il{J9Zt*cKRtwRK-2(v)VPD#_dlct2al
zPAUu~ibU!p6}v>PRHz2Wmy9d2B)DAS@M1H1gF6P}wwYosm9T3)DeCfIcr}|%+&lp|
z7GXLd6x%O~or2odKdVwNAua(`71h9dSQdWe==f~k`%Z=JgAaLLiVF8_t(&Od!xp>s
z&FpUC7Vp&H6YC;2J;m*755Bo@xAjAix{!Wu$vrr=+~lyy=^VH9qy`ikYTpPjYztmh
zdy&3|oY~HfZ4=pCYIwQ7LHWGR{3^TZX~f+fQo?6X>j4kroje>#q;6IKTS6igKWq|}
zwT2sN95zO(6I<RIS&koCTEy|Zy%{!COhy%$-0v#ikPXxkUaVb_6^-{cCy_iq(4rjd
zYuXyK>xnI*?=Dv))`T49I7wv^6c$atw$~dWa4o?L43n)`jQYP)v7XJ-+&2}*?iBhQ
zzEk(~_MWym&^+?)?*v)E9eKB0FbX8%j_by1m!a1Rb45zJ5|0hgLydp1^$IEMdUT4m
zoe>-6g=jB&s&EH)wQ8QlE+Uz;oGzy50RL^I<IlTQr-578WAp1@%VJSCW8-lMZ~U4)
zJ;sv266^6>8D#Mr;|+VuE+q)iPHT|lN*$*P+wE)f&tX^~>nPum>)da9ztLLPphl?A
zWQevYBKoP_0P@@Mky_N#n<cr#-UpaNuHZ=eZT1YqZPy5zzi>Nqh6OW)@a?!h7a+}v
zc+osIeqs&=<FdW*Xv`u1Yrt#0-kBCP^OeZ9cj^2&{)s%>J`IMo_P@?^=i%NWP^uBv
zOo^g+jT3<+1kW#5RQ_A-yA(Cpz2epAuT5_hU<-P|a!xFhHnW-FN~GNw%K%ZK%afq}
zq3{IzW}Nz<%#Q-XY#9X`>d02zzwk!;`mLK-H{vcZcqd`dvo5a5q$mvh%a%M7A5oh$
z>btLr>oCFGEN1ux>rB{Q;`rMElwqxQ2Ah<jf|tl=%*LzP-PhAOGj|3;X$~tar)=Fr
zN~uTqj*5fuDH{64iPbkuv?+IYa9U}f_a<p-XSYPgEv)-Jj6?Dy$s74^4?XWT;#4sK
z1(=<dv%Ep9VW!m0C-xo)OjyL#$QMryiWQQJZn?>1UY`g9^DV;F6F1^m>c^)_$6LzV
zKYB(<;ZH7Z3fn;{C;c@=v)YxV6fLhnKkm3KS?_~d!Ojgawk?!83wyII7C&a+7dDJz
z;jlaPPONBMaJj5m$_EEbZ6)Mu(T6MSZ6}VpTi=eYfG^%ZfnNA}C%nTN^1g{GyKgU2
zInaKro4fyjPdz80+Iwn)UW5%XPd7y@3i3xd{}jLUcOg2Scah^LYl6xz0c8Ez+FDuo
zGLMxsd+c9otJ~CJHoopbiAnMjlLn}r_wwr-1Z{#oRz+BRy6lZ;WC!FZvYF}hKdv=L
zn$*nz_@)NnOX@}$=VUX_zAv2@($<3oolBRv?{if1@iBe5wc8|1#hF=R-{c#wmOL5%
zQqTNz@e%v!Cw<}ReS1wb#b1@WhF$3ZBR};9v7$RYp}ei3Efc!81}Yy{Gv#rAG3=Q?
z|9d>ac2LI&=f5?cucYPv6r=*dmgYX^ho%QNUqjF0GtOEo+$h@$bD%@|CQ@b<-Yt<Y
zb!`1Av6&cmfE<h@55+x&^=kT8K;n<Qt%g0>Vc}DcHt^>S@R!SFUZy~-&rMNNPaN)B
zU&>l~Z=LJ7WhH4SZ>lRT{{%Rh7_G-&cFMKN-oVPfQriV27csf73A;?Y0Xp#2etGY4
z<730w*7@e@apkR4<EHxT1#Ha<g1Rp`wtsTPUx}@HUo>R@w|)9-b3$Q$<(0A6Brm9h
zhOOsC&!OePh^WhfzJUbC16_+^?-T#bOVpdC6KvwpBnoIV&S|^TQ7b7;c+Sta&qQf?
zKQ`itgCdKE1SOUtf)p#Ftp5r$^&DH|Jf7KUqBpm)9;9XDxT76Beh+4T2g!EjX+l5&
zL>gxOUX{=*_~cmHlKgJSt!@!`5WBmquwGgF+>q#|YZWL2r_QC{-7{+Js^OvtCjR<L
zo^~DBaLk^CG3CK-!T36BeE<nHhC9?M=BNI=!>J9~x^F^irAeDCEQ>L9ULw5_7GL@t
zq$)dK1io_6B(}tYG&)CT@n*Z4998vK?eC~Mp&O~8Ghwo_hJf<$u7iV>mQc^~bhP(G
zeLqAmca)ni*{%~)n^R_JLfs^5+pW7xg;xgfUq;(!lr5_v-cg|;<Kz|1d&bBT9+`}g
zLm3w$<%plXZeCFU;LGETJKF8TLk>1+{dz_f+}pVoJ9dl5Y^kE5zGTj)ILhC(J{Ld<
z(CdvZoh@D9$>^S>y0g1B{mM5*&iR*aH+wYxN<5+-o1|$z&rEr0cU2&`o^7o*HU=3u
zpn$u+lp72i5O|3?!M!iJA{dbH8g^?b=jj+qY$0hn2iMo0f$lo~sWnE#Yxuij8_Em^
z7}xal^k|%3r1+Gcdscts#aA96Ip3=J%VpD=Y}&&kkrVY=4DYBRvG^nM8lF^>k)!yo
za%)i@w-Scu&X!hm>(IkTO66-mv0TEjy%2AAzc56A>wYfFtB&vp6P4-zNB|99Ul?RV
z)*i8*2ntzhIBktQg1}`DSoPhREbFQDN>IOM<d1IBZ1zQ7&hj)a`|$QmK*LP<k|UKg
z9cLW5*1kx^6khSY5k9TWE`D`aq7wIeD3j~%Fxz{=lp8O4TQ$Z*%AP0cV82zdSs>gO
z9L)3AhH^Rbe9vS!I$L&(*%!VvCC%5i#3U^>-C!eCiro&e7><O#8d83bdo0qo(VbqY
z1HD2r7h$-JC;3g&rN<ol^nig0aHhB+&ip0Lq;M9w4?+}UggN1RY$)ZM|1`0<zDQ9?
zYXD%cI6tlEc!f2r&P<l0?EZk!YV-op-+W-JTkk#-_WSX*kReHvBau#ZW*Bb)xcgQ{
zpTl?NaK(xWf?6eignPP`>5qZ5a^L0dW|{U$5`vC$keZHW%_;*#27e@i?xY!m4mJk8
zm#D@*N7wb5MFvXLmxf`a4R336*M=7g552%@N$c^!cDc6e-$t9Kt=GR9(aaup1UpY{
zHyNWcxrZEzw_I)Rsz1d~+$2M8?hAJDsOQu#?714M;M!B4OPjkdHC^ZLfG9)orL$X@
z3)<teF-cMYHe{--9QL5S@vkYHA93g}UE<V#!TCNbS`C%&`bw56Jd|?_*6eXq(8-}j
zuXi{<O8%wIfNK!|+AvDY+HH1EKeGlLV&$B5>;ap_CfNs?r;E<kv%$wyu3b1ya?oa%
z7<${8kY9Cnj$mJ{)KT&?>naC-lAn}CCJg4OQ}f&GOG;v`dW0cdwacl)>0Mi*?G~;O
zuk8mC>oVX)mp<lIs26I*d>@0PTHfHl%Z)P`aJu?qc~S!<E+e{h+_VS1;(Jv?XSd<D
z711!&s4zvsE<$VE4!`6_@p>oX{ap;`3JE%#Epv4E`w1PD@TwekQ~AM7U&MgDwIC0%
z>iv9I8-Y-76g)Ewej!URBH&{N|Jx&?9qfIB>bYcCk^$p}b@UaDB_6mhb=+TU{EQF7
ztRbKBDvW3(uk&F7aGA<edtPQCq$8fU2Dc){HGj7=!%wm7$z3GboVwJIk)cF<*mm+g
z7HpPTOJ9PS%r>JB%{sUMNew<bn<Ak5t&wG73!}Ik(#Pzus`LO{g`(2w+Iuq=9dCuo
z2`MtN2M6nhwZ9)aso+F<Kl@1+7{d-<T<x?8xlEe#cDbB*rZ}F^es7F=G8&8jM;ZuT
z>r9%4?wO?HeEN2}dV+4x#i(=&d5?)nD*mg4{63o9CRTJzXqma?%79ULGuu1Nq87`n
zXcM3Ot+K|o{QR)Qhvr01<BSu+Su=s0S9U}sjWMcg!eL8b%P_qGdo~G{Ei7IZ_6k$<
z^ZX-auR}76-N7)$uX$tLC_Tk4#mPAIF`pQRmwIHHoOhOJkZgdYvFao=&ru>JF(9K}
zJE@qkdDsCrukx9_(_a|2>!zXD=*>UzI#V-H;T-C%5AE|9Othm%_}q_=@*MoUPZXJ#
zHo!SyLnM*DL&9et`<`sB#w7)!(wbRDcudmS0V7g_i9wI*rIZ@S_1X$QQ|J_V!?AQY
zz^;)Na_NfncV7d>mbh=H41U3ID1*<A>#&QPPVw!zj_g=bBU)q9O9OoWN<>da7={6A
zbSg6<Jm%Q%+E2JIWHD2^zC=H+yIDNh%61uN)}v?@exD7UjgXFVMPNUz>F$-Dg}qE{
zyQUTx;2L8-W9)aOfz#l{qP+HxDKm^H(Bo>LZ{_y#@#E@6%y?1Yp%d-H)6X$Unf#&f
zj(e-bQ%fChMk>EZcH-p<rxNEU^g59GGC57jx?Yhcw{=wY`odvzv`!3iw<kPNQ|8lM
z7v5U6<#tUzlmhZtcN@vxUPphx0UR!yH6Ad81a*~g^ZAvmoPwr*l;*Je{PunL_Y_1`
ztBylc=FZh!an3}BO<HM_ulsN7W<7u)u#lU5wX3k_k5<hL?X9@<GeM-KMtA-^tos#6
zQ~?2;|KAOR5G&HYucakB1t0mfd$smA?w4DC0MCU|MR)-&M8?TWKBK&;T2D{&QV6wJ
zD4Z>fTv>jRdabg{5<BPF%9%HLhv!+-D5ZyGm<0r!I4!A+DBN)v67;Ebzou{vy5deu
zp8NXSa-zW?(m^nBu=W|8d7j%teWYEt|0uUUo7@EUJFC)Y*xL8HPm#==iq~{5?Sd6M
z7&}U{>BrgSm5ctcE=X_!6*Z+-B<dL$gI9in<lPYg^>rF>>GjqG@q;eXskiD}oIjy6
zu;|WyYuL<|6}o)78i|lgvpoox&uX3J?oTn4UxZkxE+FC>tHprD3is~UEQjlGU#qC9
z@M}n~5eakr1^)bDu{Lfi>&g}BRor~*ly*wpV?aSk?Y<ot^`t#fLN<YJ5|adHmx}Ch
zOFKqvu@Ad|d0}9-IQg##Q=lHFXY`%(NJZ6kDUv_;g4!{T-1;I4#4uenvCm@ex;usK
zor9)1hH3sci(PocD5NH3hho&22S~ab?Bm8G3S*|V>)p|`m}nk9%-xA-Jrm5`!%nFE
zO)xs6MW9hx81b3x7*QTUeFQ-Sw@X)8suo{=V%hy`aXVaqqlw*s-2k{sNf646wL}!6
zK`VfDw>NCPE^2JO-u!Q6-XV6DOfnPR!(a@BM<VyGRe+7TM%Gw3_Q8S>n>e7d_uIOc
z-K>TC_ttdu{v~RS$oCZ+t*=;FQ+%JxRGm?vXOBP#FVMn(UUzAVI1eiF${Tyy5hisS
zyA4_zWLmxGf6>^Ad-@%9giO_2ib&Ag=%vL!wF@oAeScgGyV4ftk2g6`O6Q6RO`PI0
zG8*H1CiL#J;1#en_<F^vMT)G|BoY1y>o!8<#AgWl3;QBY5D8ZUKp{>$fcwElbq8kX
zw)kv>|9an-s;ld*czN^RvbJfpcB#^seJA34cvMlN7%)wtfD|5(&4@pTc5NmRxa8+~
zW4p5!>o>P&D8Us6*9L~G)a^8lrZPhW8;U0<R*>u#`UCFri&(}vVroZiI3T)!ca6pq
zCh@b=4jE3Gh1RQv&(g-0prY?X0@nJ{h3s+A<`305k3D-vgI^|erJWl`Z!pBn29_@j
zDa>?iu`Fi(vWD##?mgCBZa<-7RkKn<_U_jP<zRzOAGd$+C{Da0M_K)BG0`c|k44q}
zfxPFnD%}yWt)DI{n%i5A?*QY)UbFm5B1JO$xsL=PV&1Um_BE4%1Yg&i2eW!)v5C#r
z!JDyD;~-B}Vad?NW2dYH9+{=~xr7hv921)Lk+=1hRlgrTo~67u)uOZ>#;IRIxegUf
z{j84FH27MiXhuHwY4msByUK%HR=eq`UZT3<VtBuUjjp|ahul%Y+oWP|OwY{OqAkYd
z<yr_AE81^@j{1l9hs{q~;s?vVv@$g5LAsSi+;XI9ULUB$<YQHS&a%yCpWMjQp|S<$
z6W9`#n5#(sHS$sMwBznYZU&Jd-qR^#g;g=F4Hu~>;w`=AYJ{qT6!dxeKyJT(L34JH
zmJG0Q1Gm~CdeX*5Gj?-kYFPe?sPEYZW>I%#qNc`{t!^f*?$T+EWq+t)i1*vRFA0U6
zw(v%;C(RqXzyIBPYZ^^@y+#Bhl{`T5EAJaCu7u8{x3dBn6h6&>weym4r<7jO9vG*K
z%>L01yYvp)01WqTN{wq!*Re}%u~gd*tRM%9OR?j4?CofcBO9e}uSsInKjK{Wd`%3J
z^=^?>LC^@PtdV>dQ@z@h5+S(VAGsHeql8<)5h3O!xite1M${e_M=#xMp}8Y4cknrM
z99<e|2nRjVup1zTuvQ`;@NV5ig$_Co_yZq>-jouN>z~x)iv6<FW@J1V<OO$l7p5s6
zCuCa_HXy^e%ETN|z;J%yrZSe%@x73nX<tjdw38OKH$HV5ROuNV_-p)D@yF`f0&k9e
zI04@`JXT&ve%lfs0;kThMg$2fNTrxN_lL|jP#-FaTZVhww?a6OYvGOJ?#1I`z{*mE
z0;OFW;P@TtvXDqkFJxOfI+zeAY=^AB>!s6Vb7YvYF=`f7PuZAdYQ9mRSfW)e!uIhk
zw5V|NgO{J&UVaP-qL0;+57t$JcC583t`~6)cP;nq4wC&NBP^vfMVO9jiCB*}xoD;J
zIfR7yWbA&?u^jV`C`h=zKeX2WFWhPFearwoo$Bvc#n~f?wrz7)jz%%%v$MZZC?^P~
zKm3ERn3$Ta7|%3I+SgJ(X?uU(?6JY0%Fj5)KtiQ5F!iB(m4CNY#tFtN8DFWl_}|3%
z1A1v~3>x!NOft%{Sd?ibnA`aC-!2_h7{JK!i0nk}zIf)Vz?BRZNseQ4R^N%SG_vx4
z9#Z)8kWXGBkkn`Dbu+ptt2T%i3bFey@tcV$`v?Ok$rUXK16M+<rRW;3>Eu)YC8#sd
z+|fVpH|nX`_+BSk)7y)})VB8MN1gl+f{L=#{l9>A{eOzq{{PPYj}XfL)r`odCd}3t
z7WZ*IK?eE~pjp}|N+4NG&FAc$j(jTW+&^<78G(Eu42wsT1Yux7=(EQRg>8q{{1&YF
zqZ+_~4e*^S&j^2bGvg~?D<(+|#~?v@9QxyLU2Uq&5`dHtp~(8ieN6g)e@ISh?|1xL
zIiYW&`}IK511*k7@LleC^BJA~EyVl*{or20y!qWzdPds}kd=9IupIM~l(nwmHkTmN
z;oY6L!-D1d+o5h4t9Z^Gwqp(ig|j?pwxcqcHSP+t0GZxbj;!J%=B(o02G=TdA{6LO
znzB8)qTln#C$L=LAawFPh)}CVH<GGO?0r;Q|D-46mcce+Mj!)5$*2t*M&)b%{rju>
zyy*ES4<tos0vQbjo7GJi^e47KDf*SY*ec1;o!CL*_q-2`54|iMlccPP>Z<&2lrM49
z(`9(`^>iXSz{2CI6@OKstR3Xf6@mw99##tK8GX|c?p-ArDm?Xdw5b?>^@o)~91Ykf
z!z$LsEqXrH-?;V{c+bg5aIDn*0Z_3{me8hOJfwBx!LY~~!Iyb+&}KO}!f1!S+NM4c
z54geeukF{g*z3@42<k6Rh6(z!dgM3i2{|>T$Mg&CoBH|@@@~N%EfZTVfJ6zfc+p}?
zn0b126A32l=nsC}|Fi9^U8X?B=VkvjUyBXI@~(;NUt?^)7Ar#aB@A|(FRQjke~1m8
z^Re(a3>_6YFZe!vJ!LPiO_It7mOp;-*-gO7XO;X68p!+XTZYArtf_mr>!-jy)@gZb
zJq_r{CE8(1$r=D5eZvE>7ZF0^dHpy3nj269>GeVnK`w_xBF{oUodPZBl0HD}Bgpsz
z1|7_uSkC(Vh%L6qeSN*)tpOcZH&{M_2z_z!I0IS!PhB<)9kk0?_I-6iz|#}HTUEi|
z0RXR+7;%+HyF22qpvmO<rRBJOH4Wvm5@m@jp?COIA?9K6E{VNpJ=%j8{NE0Cjn&it
z5NNPg;Ry<KnC7i@Xx7svWo{qPJ_6bOvD?>{EMU91;g15Y=kGpfM9y33`1H|Gw)6Z)
zZGk$1)csFb$S#Hq%v(^)?O)BMs<p;=nUb{=iG9mWPjx{Ik`B#HAA=q_@`*9^)bI5s
zj)DrrX(0iTls<~A`naO$4fdM=;oeDF^K=8e6fBeoTp5a1%BiynL^C*0yuW`*Ps&SN
za4=}^PGltGmd^`>1zCo8&bvpji}ZQi4d^gV=Yd}~CeEYum96)05-GK4Lnm)=PC8CK
zp{1JkVe97?e>a6w@^z|8k>(yQ>%q<`*;+~^{a;H+nRr%H1%%ty2mJk57VW&z&0EAV
z<Lce~NT_K}&zl@%<j5NljJlB(Jhy*Ame3a5IxUd#9FeGzk*eZ8Br5bfqc5}Ccx`Xp
z7}i8rxdU<nR1SQuXiu$l!HDH}6SO&gXx&dl&MbI0D8q%{y^4F2(0W)XL)WBIVX|eV
z?P$W>w1&uwWw_r^$5~hd`kKZ@&(&Jdq>JQ;OBo{ei+0#V4b<7u@6PJ($`Nn-f<v%^
z+Vv@uVLmRcaEa==+F<{p>c~>$t~nTP3tanYJV6(NzW))HS>p0@w}Elo*!vZPmpyj+
zSJ<A+$r3BdU3t%~($X~Pc~iw@07#pm)VXBV?s!viDfxvk?UnO4r*dB>suFPkA|t#2
zT3quNOr@@Z!eIa}o)8g{j#O%2r%(HYqT^0pd{Jz4%hXxeo+4e_WXH7>LbTigp=I07
zIFxSZ+@&}$yyOXp<(X$bi!w9g9m>_F10`<X>J(>evU^7NR$yzZ7)rZD-B$|rCKk1M
zSl%vH#K?E}yhrp<q1nzXw7J;a8RGEj0ADFYdKWb&xZZT|rF<n#Ngxn@8Hw5LnSRw;
zqx?$hG=d(!5|-d=VUYLH?G?TDhA!_tXp`!}3sH5F&ATfrj=d=vg|ZF+vN(rbkW=^4
z&Q{F;!{tBxRUs{mt%%!ikk{))vDYqWLpsVL)f;k;apc5{E&M!To-^4jU4fLDNS>iw
z_Xkfmcl;2}K0b2SW{fyEvz;moYPpOt%bnU;G$MOv<31$XHdF~?^BF;c`F>Qo2Aj1s
z-{hTDfex|-_unl)48luslw~-YvS&?a&DJ=Z+8taxJZxWt+cQq|gOFC*V~q#}|B^z_
zl7+rp8OJv)w;8(i%KE$dU(Czno6(ipx7n<rNwBKv4`DSyp91YY=vA>=^6tBqv_$@q
zsKw5gCWAEthjH6Y_C&0^6o0#r1wMnc>+y-S>jk@u*Vv*K_#hGy^F<fq1kK%L_<sZs
z{jOt&&euPZqJj@EwaHIL=XL8UoW9`CRC}?_yt<pNbE%L%(lesub%o*mQCrUH3gjX=
z+Z$m)RBW>eH=(oh^_n!k3hst+%<u>LJa9_)uW%D+zFa4`Nw&mKYu!I)-jvV<_5Qv4
zGaRC8;LQLiK_cI~`nmI*2Y``>cOjoj5*u`c`!*`nTL4(?^Nz-cd$)@5ooC6ynjYi!
zgi!NzsCmuAU<G+6s;!^J!=MCt!+-24jG!Xj=A`I~&hHydDOEV-1Pz7=_q1LlNr@?6
z+NBSM_2!FlOwfq2YE0KJxtUwpOG=*-o`0w@T*JcY)N{Y`jz&aTtB1}ZwcFg>UGZ|>
zoeBnOB%5@_T?KRL+qB5F1wIw|JOCSi+`k!}>%MDbQ-^SBftV(CpaUvpuf8m*ULPe1
zt1G2t>ChQS*IXb{WTb0<52x`AiIz=bvjbz5wRii<yeN0sH?3N<O;(MaGLSRZd!`!(
zJ$yhq>NOP#WhhMwHN8L3A06_ur%V>)gnwAvvvwLRW4o}iW|P_2Dgt5TG9+uwg(kX&
zYKECjWyF~m(y+}XgwaIdo;LL?wwHoUii?1svbeD|o!iy%{hrXHT6f&4!=MT5fA6m`
z#zqHjarYLO5^bA;l2oK!cS2gVbsr%V(V5y4WMmJ^uBOCLXRF(%ds8{_yDgtD#cg_i
z7!2d&yi+K4sg{Z_h_KTR=3bu&qp&>p_+;r@-L){DF47wUj>~n*WIZwM^}+L4IJ?;H
z(l0xk*Z^psqQl_v(1gKYM6Y7UA<jrbmG`~VIplRsd8HYRn4~Mp&NFWiAT21x`8z8-
zSZCW2Te*V{U8Gl6)2^pC0XO@iWoC?aDUWrXrm_c0Q7zhPdAo;WN{<+`+p{L7T_`6o
zilu@}x9oRnDMz%Wq6n13`LlKAWphvd%qPs9huHV{5mVig7?6m)1vYeamUe!g2O{q)
zYwujq>s5*>NB7kEY<(yPB}X}JP@f|#comML{T7NNjMLUbDa-Ax{i8j({J)9G;PCtR
zS|!95DO?d!50Gmeb0J&pjq;DZr$h2=Gl`I1dJg=k`1(Zp#v)Pk2R3r9jM_2t*W`N-
zk*2%~{8}d_9)prtx!H>YFRN(hPm2Wk2C*ifCsF_&Y{D33<O*_^4+ir|h(<i)4>}=K
zR96(6^G}HOHCLz?P2Qm_%EHpQL-vmdt-KwsCSEg6I#=}k;VkTjve;zu5a3tzRFwNK
zhW6RGJLJrF(Nq=H8G7oijl(<-RwlPIXn<85)Bf|giXFvQMc)3W2XctJ!1i2dZtkw*
zBn!z6CCozeIJc#$Ubk48fBSCdto@dX0~{Qk`xXB*_j0d`2h0&?GM>AKnHXR_H66V!
z7tAgm0he0=v9h~(m$_hY>*^d{X^UTopJFz1){O^KR0oU(>|J@$P1AK=SKBgVO;_hd
zofI`Cyu|v+X718g#TKh$r|bp$SU6|A_{Pn0=XZ*)bw(_pK(kv;(bjqFO@=m*0jIKm
zLtqynMe*1o<~n5l!~L+3?n3_i?S4G<rNqRw<3^H1m45!BwN!34l+Sn3qpRnAb`A|r
zdUxw1#ftgYf@JaMM(Q-T5QT0b&NNtxx9&F3*rNc&!gAK*T3v)q&TUoP=6#kqr#cs|
zU3G5;xqTQ@6~-K8epTV*K?qM>-0aYI#d~DxZioUgk6`V3_?t^0i?Dj4u{3sLhB-%p
zS%MHztC*P`ohxB?X!~O~i^jl)#h&t13d2+j{2kK^uZ8w6s2mG4T|K71197UnSQ{tW
zyQ<gOg}sMJR}Wz)akqV%BmVqi6hUuq#h*AT9+)9(cAxBL$cv85J-v*&M*eEm$A&C2
z-1Ps!T(cI;{a6|mNIi8%DW2ryoHJG<w&09DS0?*=vJ~5&A}6{740Nk0Y5X@daNjiR
z)s^A^vY#F5e29#yrKo!6ZvcB_B;E@**Rx~vn`&8pE~(yOy<tBjF){Nac?@sve%L1g
zYu33?h_g5Tn#Nf&;YdAtwCq$h^IgS9f*Nj0!ax+aYgQ16Yt}yY8=U!x%Va2Ki1kF9
zXGSJ}7@Yr(S@^N<p<zZH%6|O|wAX7Q&u<p*zX!yywPM5^L3U4EP=2|txr)On5a^x_
zZ#!xEV$PeQ0l9KtSECI8<Z>2!u8>tAZ!J3bGmf81X82kW_UWqJ+0ie;SFzhaA{YYw
z{Bo$;_pcdC$aXukOqB<EM)csoHP5lcE(hIDl{d<^pSTR<hoEWWN4;DW+QTk<^piue
z>WnRM*H5m+om>(ccqPsAdQo=buZ5y8PyAw%=rkgnKYz<N903o_l8H;_Zl|#GVC<5m
z<{ss46GhsM*$=)n<;Es4$F}`T0@WQIo_CpHA?|F(IV=Um?8_`{#fW9Gr>!@g@tYU(
zc6X3n-!{9mr<-E!9}BiB%T~}Apq@CA;jr%RcOHG4+j!ogjtqt1`8z!z<!VMMsg2pp
zuM;mP+OgFogaYMVAmyhi|7(Q&B%1;7A*AB!c{}ri)oKa-w6a(Uc)g5M&!RErg;<eI
zHJrzk+6t+0SvQvU&*do38mR3LpOg8Su8|M|U1f(&1PsNPpocoPRsSAE`-Ev$@rE=y
z<*G<gKW-bb8{ZJ=p!xB=T-9ng$a=@=$<l3lc2_q#7NgB^z@^=+TCi(5H5~sG76v~h
z>lw0Fc6kUD&I5fS7W4_zkVnx1-d4dzedLd9S5vgktl0<_HB^`%Cqcd|Wd0H4vxI-N
zly61gPZgIY|FZ&cWmh>StkZM4UMS#jS>-E^282fp)gSkcST*t7EimG4hx>c)fPFn9
zB9+;}u=e0#tHjIMKU}d;>2XCVB$z!}A2#1yuP9z2Twz;=GfKp?q=s97p<SN(ON*HN
z))q+uEyhN7uM!>D+(Cb-A!>`k?o;pXf&LyN`SM6>-g_8DVRCXspXLQJeU-j3#c3o#
z(lI=p8YEA0XX`B3O`@=m4#C64n$3xF#|75vJ)R^CdBpI(0PLOA_^~qS50dj-afn-T
znO_Iju@u-0t3C2+lGx4Pg7HiGF>+>>RnZ`T)Br2}W3_cvrltbjQfPFA<4NwlLk@vu
zGiVP3)3+bO+kbKTgw9IXy;)tEPGRuc18uupRZ_y;KvSAx8oE`6I5`&NB*g8aeJy(G
zGkztwtn`ncffWOW3q9cB5|WE=*hN*G$O)?xvP6&-IlP<(YKaQm9%u=%Ymd^6iJ##B
z)VlEwc+{y9WW2OK6(?2{)@k7TocO~kcywz$WJ+nPIbyudclUM#fOv!R@7h?d2;_T4
zOCHpn=wvwOg7PSVqvz*^YfXli*<sKuPJ4}*bdRLZleL*Ti)@<OEG^PrQAK#MlwXtb
z!XFWK*WUYdoG$m)9o=t=0$mi5`$n{;`I&{xqfU7;d|UQ7`Xiol1Lkvq!U|mlt?#2r
zdN5P&mHI8ic)i|kQHNPQkZ`Y`e8cnsd}uqeE%EDM5d5TCMb%pYbc?67gTm>99n}E3
z`BkT(0CYKb1Rr$#;!q*w&u01cS$pj8u*R>ZXmPB_Gi|>iocEqiRqsuj@adY@F|zC)
z$CgC#0Rmj+voiBmF-bOdvaK81iYp`hQOG(?KK99AFwwphYm3Xn?-V~8r%8|bzw*3^
zUmBT=O|r2+Cp;Aw411MciiDL38;hFsPpoPDb=jFpW5j^`<N3senoS>VVr5Lw9Qjaw
z_jpYDxbF0)$sW~m3gp(gLk9?N`?M=QKkyUD0&K3Smfz?cYw|h}F`hp7EjB-XN`7(p
z<ifgILN@aVBQA;!z^OW{FBeZILd@4a*_|GGN^_L8zhTkC^0|ZN?^0<jY+Vt_suB2q
z>}J$aw6^A8c-gnS{T%yNK&*rfjHr>YV1H(@h9N8<%Lr%bo=Jb~XTw|TmY(4xCrr-m
zc;<81P_fb)`d$p!bA9|SqTDSJh1i8z%cjxi8{Kq^6Y)Qp1HogEiZJ)i3{_*bL&anI
z`$u%GW0O{ozn}11ZS@b6i^`4Hq_;+X>Mf5gtPeihxBJEe(eCPgpLa{BN&a!5+=KGf
z@}V+GsK(Y;B&K<GQ%mgbi&hc0&8Nza<R6SI-7H~m`O~xKQ=aEUog;?c+Azfb472!m
z5rPeQYg2*8y7j!i#g5!Ej=M;V_Tl_i7aOT(ta!hyvHSfXLOp4Ip~FL3)(eg$`g-$x
zBP*Pi_IX0qJ@*SF3s$D-rsKe=Hd;|wjoFm3vaD4{zyAB}XWUVGQs0~MM9!NZk%`aC
z?VpSW3-v`U<eO+m4j(2=M~7<9<;lpZGvXS%)b(9clJ?eL;y5Dt=@&<6Zkbq-T?r{+
zIx(Z@sv{if7O_)ef~@JS&XK!gyx8!3k6H+WEaA*9Lr2d)6Keb5Dejckht~{vpIT~T
z$*sDy;3uJz{wI+$%NNhJ&6uFXf3hHMDuN>G2DhxtJ@e)PJ&T9)8MS5o)|m37pb7f}
z2_0$SXP;&(K3*K4uabl#YTb)VHL$jrav84jNEmCw7P|5UVe7#yDL)3Gc~LH^$?Yuq
zwNQ<teD%ee0xXSS>wSBJDAG51mrBiwAk;cYJ@n{6ZB#^`grV{NIdO6k#&rHy1VCD@
zybyDBW=zrwr)}^oYG-o5fq|%$)8J*RdPWiaLp6-L3y>$)ogN!JJG9fLUFhw?%QWbY
zcDuPKOV%up94z<tNw!Jrnc?#S)5o6d${5Kiag-(OMK;QpPp{<TdWrj?^WR==(bETU
zt>{_9oWwe(>sPfRB>*~+vOW7(|0IO$no+z<-h<qy6e|I$#gw1AjV`kNqbS>{E<<Y@
zsimd`HEAx=`?F^qiLpNGeK9&NPHc)q?Z0xXE|^0<9XM-y-uDnxc&n1xx#R57xAu&D
zW*L6TVBdepp>R9bS&imN^{mTvTyJ~VrdUqXR@8fo_IAC?>ftct_Ua_t;^7y)pvrZ6
z2<NxFKY$LXi)&b6UO0<?J5YD_*%|dn5&8Afrh-Mn@%OkOUq<vTtEdxkg;Llpwy-wN
zPz;hetMNYZ5gYtox1w_y=eZ)$BFJ9?k&!oHt(*fk$qn%>Pk=!)Qa%SU$sR9XqQy@d
z9Ny1ZN@tw&Pn;@i(-AEhk1OpFjUl;9jwvEBQ9XUN$rHp{6=9*U#bXI)7KdC#-N#tj
ztEwc-$9SKO>JW>GnZbUm`$KKBv8xYCdjN-#lg&G$>Vhg-jLL?A9d~1HdOMbTZIO{6
z{^jgm!pBC+ZL=J{Waq}E1et>nOrX%Z@1wmxDf?dRn#M-6>?@I;HC}F`HakE5o^QO2
zatFH*i|db)0(xYA@@d_l)<_=0Cjx&>T4*$@Uz03s>UMYL={|~N?Y*Mhy#K)XPr1Q6
zpuxSrwG2K6J$Y=<aBhXm;#6e@K-0mijdTAxxsdOgwx1gt%*g#WH;BXJ8_rLigoQT0
zhh%)UJE1>GOrt(kAs#LKX-{kdQu>;wWwWsRIe=A8{Z+yT&(-$K#{N-;M<^>Yy^~hj
zscTMhlFBRheWWG>4YM}#l-JI>hrN!Sw=)=F2-(%RlrlUaU|AC=9161y>rTCxmJaw+
zvr`x{*Bh}eg#CV4T|g*+KmnQJ%4mmh+e(+Le-PeQ)0H7eE=iuaT7W$GZMvg1iFXO*
zs-KKWdd*z|fO{qsZmyfc);OZhvGT=79hm(QLUo8l)2WZB2otHi_SAJ})Ws|jCyJHR
z)!=x(?KLBmvfP+iH`0w-Fg0#R5V+YxRl9azJv#%|k#oa!M>|`~n+~q0-an@W#+7xf
zxSj8)-s&1vw1d=Puit3S296!?(QH=s__x$6`*=)JXP8?|lB#<j$*g5SR_W1^IsK1{
zmm-*t0u};JzHhhiE&h9<Gg%q(o>KGV=KHmHpWG%ek4j=xkNx3<;Z7wChd^{;{j6f!
z_TE_YG}I4)KlArZrO{%NPjc%Ufl|&o{=cl{y>*?>v5H3<KNl;)x*yk<BE||iy^KxB
zvu>lzSns0$<%-LKj8UlvG@n5L))*qZY$#}E(cph{ws1%rTXrPM3BUSNr!Zv;3<@bd
zK-&N9F`H5D)EPF+$ZeJv=`~X7)wyH|Rm@&xjQ#Ie%R%}OoutJ$)kSLeSqfw?VGyn%
z#`L#*>-TfsXROHDCd_)YTOC*bk7*H>^5+Z?nsHpusK)s0zol)RFg=DL211sAm$-Ex
zbWF31Pvl!VGvwdT#rI!@bCJjXS1|OiA(@!|->8%Ozp1lM3i*M#it0{L*eSYdz7kIe
zkn~4?r0Xd5)-cKa4`z!jW59w9M%^Dl6bO+&<aoqziED!4=QA2SXcI79$k8V)j2ZiD
z)+ajP(957gSB%7jR-onV@9*t7=`WCcSiZ$;hs!2nh$GEF4*i{p%{Q9mckglyAYA&N
z()gd0!I0y^c89!BwmFjmE*1q7!JYU)cs!tDC}D7Cy$zKFFA<M=Lu4rFZBB8K<WT9u
zO2a%xn(wAIzt=Y$_I!r2hL&EVC^m=-8N*mNfj<wo9lZ|@K7xgzU|G+{Uj(N!*=svK
zU!>;!qHn~0eg5u}3r!8wSR%Mj^0wTuZuJpON?~>sVil*pWu7Lj{6<A6Y%}6iI^j6B
z%8#+40%KJ;d&M~Hgc++UZS>4DTIK5W&%KDj?t|Nx7+}ylB#+Yo`!wjM*deRuBqK@X
zqeke$rj01>)!tnj|0Y6t83HvP^tFH%{&C~0po>cy3<2oQW9p<ZGC?m3b6jbg0l!l@
z#W|O2sHu>r#WfIbbii2ecsAejNqW4s3`LFUN|^-51Va#-Wa)I_0wgy`jDNZMMobbJ
z!-!cn8wMu)NV*hQnigw-+g)xdZ`5=SRUBacv1)T+V9d^)MELo|l<aS74TY&3@1!t9
zh*x3@?&*Wigcyae)ebZNTVLWF6OpnhceRxm>4KaaJ28xoQs>XM)YSa!toPaVwQ<V(
zlY%?bDuZGqoDd`w_I+vjH~(kSD3?JPtvgL_nMo~%nA^{UO+rK$?AkK8hH9kQYlChO
zOIJ$@_Y|4bN*O0ty^eD6L_^hp$ux6w+ouc?@p0XT+xgD#6WG)P+cyei(n}MXoo`v(
z_(-}8fIp+{YhbM~bo{xPB>esUR(P1pRuk!&Fpuw##WPkJE{w9+iZ(GlqyA&Ri${5%
znXr|N9rRui&VKtz=Yn3ZZXP#~OkrptDsmNkE3><AbFeLpk@Xo)WQ%5uL^qqHO$)3F
z$gF4Q)s5FN2Om)9edxH6P#qjQY}3ySGKz(Pa*n9s8SY6sH}8a^h|`~8+RB+z-_s7s
zN_s}wdgMj&7JR`Ue@s(k^3%ZOk8+4g4t*PtUQANz={e&vaSpE-vs9el1Mtn5q-eQS
znE}NR=RPWVLoTOg<zW@{oco3Fv%_awsP|Dmv`a&%h}s}0Xi1YEEdI^=T*;?<ez(#~
zy4?rcbuP^i58nhn*^sqHM3yG*=2k*p0|;5$;3P$dO+!ycBF<#$n9Ft;Eq{M#aGL!I
z<m5A%s%c#XmS#0-|Du;DW7STr4iFZ>eK_Rl5Uu(L8K=9>^28*KF)}?Yj_LIt9w$hg
zdd7K^ck7JMH@!AdAUyB%teHLK>W5nnlsOCI;-5yTFD5{T#qPr}xU}g6wAw{Hb&U^-
zLQZhsQ+Ap07A1TQUZK8g9CyERC*!G|rllpv03JKuUAVJ$;9~Qh>llNC{clXP+n8Q+
z=zl<2xulvj$Rvk^2ZT+=F;-vW|2)q(pL~xM0|qnBS}T3tTEv(Am#pF}C!ehWFg2l)
zS;2;le@OnDQUwaeqWxK&x8T)XBpk)SF7k-UhQX3C^DE_rnGUzb6lam!$vKAWyMpEO
z+56NpF~wIB4X|tb1w!4TImbE{Y)@Ns{6iU;=Cajo^?LtEEWVC`DhI>^TR*WDZzfe|
z;=#u6w7q6Vrc2dRyB{3qQ^)1Y86}{3KdHcuy>9Ck>Xf1#sXeKc!1kd<x+0NSEZ-_}
zK6xas)Sa9spm`+E26#aOfJ3wYNx|zd{~AV0*1EW0L=?dky#&q81PP$IcQ?BgVeP?V
zTCY@;w4_o<4(Ol`paxzuGvca3R-i~0(hR)p>JG^@Wk)$WEOXu3KR^z_;TO?FwYlQ0
z{Tvfxo+?^)YMCng#x^!t@HhcfSesAZAxi9~gfXTKJVpV<`!h;+H;pH_@6<#(E!J05
zpHC&$%%n`%M*O7QVF>ct{I4>+`R|Med|px5-f>UeO+F|9la)j7Ju!1al;Xq{Wlv*&
zWhtfQx~YdXQ^TpKiZ+oUn%TDSh8D)snWzi}T3z-WVmOF<zJvW!Pn{#0e*_-b?1Z*r
z4!5U1DFkH(kF&2Uy;(&|P|K|=0H-Gcrlun<TlR^4@9fytw(Lk9GB2-6m{rXvV{u+<
zCUl=7dxOHR+GX&qn$08K?uHreYVT;&fE#6fA|n)=wJo?}t~8bVHT$JasU;$*=%JF;
zjIXLWUgOpCwm=ur^4s$<?HEgklbQ0`R<o1wxW%;bm8Sx!UVv0IIP0d9*L}&yC$`Yd
zdD!xefO`cZpSP!X86`LN+uZWr*R03B?(4WxoEk-teTkxt1tEP4RSO~)EE95c%5YfU
zmj~1lkXCYWj9bk*q{K?#et21mTWg-Of2qTEQ>#|hz|;69mxkd*P$9V8>+o?NVKmU(
z$<b4|KoXo!YajRKt?k)zDE93uc~f^lKjQ0QFroAC35>H$H2krm@ymU3ZbV#Els!od
z!t;qNJaWAp61IELnr7(aFspW>#<%%gjvLReTrnEnLt0fNK{}?cE;S{(#19<c**Br&
zk`|K5#wuZWc=0hKv2@h-fhfIfVCpP?#NNFfK#!=@pr-c?blt@)9t<`2(xZtR6u?;`
zpSz4kHwFfK-_BFJDOu*}P^+mbr3lUYk{Gd_?t=M@`HaVV_dQ@T%Kl55RVp{C=!zFG
zg3oI8dcQ8E4?5Rra_lf_V#}Wfg?Q`yLCgy!s(%pk3HBDKKTd2}8BdgsML0Qx@V~r*
z<}N2)93H|dz-T+5z*<4*bk|x6>6`g;rORmL-+p!eQi-GS3!=7FxV344rm<iYfi+}0
zX%#s<R9`sJ;PmblfTSUy9($u@i{vvQXxfylx-GndO<k9kKnD6dMqHViFl+gM_Js@z
za4ydZH1V}*k|#&*&^C_tEUm7IcBEhS<R7219xLPG=w4WEerJnNRQG}U<Cjb_66&>+
z7pGuuL)?`@H3M&VBezSo@>Y5A>4k&Ds>5HG692@c=Fl6)Ii8d)^j#6_mOOf2;2@(u
zGV`)rF)6%;B4KF0xJHVWX+fiX%7@<D^9c0TrD=<3t|P1j&q<06X-us>CVLhXg27o~
zrzXq>lHEzaYuFW5eJaLOXhtd|g$F&vRNuTi<BQMNlx*B_c379RK6)c&$6`^{(?LyZ
z-HXQx^pofX&0H}Z<1vOa-MX9ek&sa(8Lu5yif_E2<f1_%RbJzI0kONhk+D75G_rc<
z>jjrJ`!Of8JEIcsW~n+=P38xbLPE19sjxajBXMZ<_#_iqerBLM*gxJILkZ5h*K=S(
zQF=sZG1J-HjxUgn03$vvwVF`&0gq1*Vr@R&2ukwn2m24h<tH^7#DI&97n=_NGb)Pa
zjAh)@F+k18h1U_BRmOX$S7pI?=fiW0E2?lY;Gc5fVR_N>eWRM%tc3FVpQR!G5h}4s
zj9bxQtmEbaDSa>!)PP1;#z9wAD!_;_V++J<hTkKljbX~{u}OxKL%Ju=8f*S8PuR5b
zkSZa@D%Hs)T7k8$$$5jJe8ny`6!H_V${dKz?^8_3l)%%^Mvbc*1+D2NyM&r;whvbS
zlm{1T`0Ou*ljjQSgA5I7E<-<EP?Y*H_bAkC2PnQG#0^L)ZZcd)*xL87*X~Fjb3I^`
z2G6MKF0uUt1#{xS5)ugc{vB#4Aei$X-oEmid$j4<_dm&9#GoLB!PB}UDi@5^c@E!$
zYjJn4o4)?RLUiJ4Z-EJ=z%Do7I!$3?=KehZ3`5^q#9;)9)6!{`g6Ik9eYUMXrL)Oj
z7mR~uLwC`wmwO#>h`=YkRhLld0~no^p$gB5r0Q+KcQD(Kec|46<tNvFaXcg<rB(5(
z9yCTqyp-x>6+@Yl8o~g~e1o}Pm$fU3X0-gJ;V14?lpWfjM#JA@K2x{trRK9V4?y0x
z)otvFAYh+kpmy)@VLPX#nW#>hU2x=|e^Nnl+81P~XeyqV=pF-SJO*i-Xgyh8`!lkf
z{`HPV6!v^dSXx|SiqhNFl3<vso<z@^jVf<-3Sq1n@&+pJ&@_ENb?DJ2phb7C4@{b_
zS+kpOayZ5~48yY1?X*J+6G8L63;f_Akjs6RXJ5_To8}9ok8ClvPoagYI!2ygKFMmW
zQW7+)A}xS#pck=NN+g0~;W`g`0?$o;($5=fxgWgv403NG(;mG)n%i}a+OLD!rd5pQ
zd9_&M=uV!q8x{pVDs-xt9Dbi#893s;y8duKDu$a=ntY?60<!)Ad@p8oOW$kKl%^pU
z-_58BuLe(d!vgPu6UPu!T^a9iF8zKAYWb@j>n%Y^V`C!Adhx)#4-SFC<iIx5_uhcM
z>RtXPqazxkuz9c2tg_P%;T{$=VV2mXrT~&z@ju(}rlLBV`?jriTXWsP{i`26z56tS
ziyxdM67OT%YOv+ie=eQ7;TzQDz<9ds6*rhh!0uS~blw1WPzmMfYM-UGj`i$*;=Y5o
zlo8>;#?$2|s(q)3k-~wdsqTw{MbT@$c~hu$6)iyMJqo0xROzFQG!gk7UqvcgZ7L$)
zvrJ&rirZN4risY-^&P%3R5-UdqenR`HLT4RWU{1ogiDdktu&s8*#8Q5T^_HOsn~BV
z%xkc);KQO1fACa%U&&ceo$-55E9nmjs4_k(F6*(B)kYgfH?lU-y}}5fEIS$Pb1tIS
z9i-wQVdHjExN-&e{-Xbfz4Pp9YW>>1ZWL^QTeqShU_(TjQl*QCNSEG0rHPbCCxHYN
z6_qAkdhaFl9)cp$TOdRT5Rno{fB+$cBqW@L`~RHh8E2evKEZkGD_Lu<J9FOix_+1A
zFXS-QLUH*?^p*9n)DWfuBJ_Fj&bK`Si<C$<5P_vJ9J1c#&?xZX0sth@<3D};i}S^m
zdd_`gYC{hG-2LGW^F_C-DQ>fmnY78EBm$nz3Kmpr$IX4J_lsh(hh3{0PIaPHmvD~~
z_OTJ(v+Mz<J|gU)n^)vW?W}gm%g2pZ#2bf`oCztZm0CRLXtma}gH2u`N)r7Xh=SD}
z(~!v0m^1mt?a%wBNNawP<}PF0@DftWa+6?rF14WvzxU$JVChU>!r?=AxexaKU=F9z
zgVUqR0DP4{`!N>DL(a0TRpbQMvJY~W<3IkGnGDL}n>noJ8<ve!Q@Z)LdkOnmcTiS#
zw84ik&5Kd(((r5f$x$QwayR0Ujmg?Qg1<ZiFV(CV%j$p1SLb#(Z48Pwp!FKj6Q0b9
z4M&qU2JY2!OtnM{4Mv!;kM=LQWn~|3)}t@N-9ReNDao*4-_5bz%(~#@)%X-m_au3s
z*ZgYub!Aa<3Oq{N-s#n{(%+$7L#ESyE{;>*2GPgK+%tGDlsWg%-7S~vDp|#+6e8MW
zFC@qZJ#o-g`IT&Po)UJX_sNG@pD-8UQZheM@G42=JU^wlXaFny2r@)eS&TXTbWx1I
z$`mt9ZV#6c+@>dher~3}lN%W3oi=w{%2=GZqOyOBQY6yI7AE!kxx=*_=dn|9N`LK7
zZGM_`b?G$YSgz>a(LP?<f3m=h2i*x4Jyim`A8b{)ljK0P>epN3hx1zoty(P&zc6f_
zMdbM<s9-2=1mQ$1E~jz5&7{8K*kCf3l4#9>^-g7WX2mtqc>MdWH!ysVT_GJ(LAWI>
zM2I~B6_jS&bC9>VxbiIdsbNMwSyQT&sFZkqQbW4i?sm5O(UV(mgw05<5S=g`{~8_S
zZpaOLg+6EQ+#QF;K$2=7>%Ow+RZ_ZXOLBW7<BzGFe&ndXH~(5phDAE9LL|*Z?K?8B
z0e19?`wN>l*!{5=(qZ}Czlqme`km}~SK6DE1a}y$TfWHo<uHQKRi!O2gVR21&B1cE
z5I9)_mZdCXiE-y+6|8$|43~W<<(cs7kOtq}8Pl4KB&qQ=3O@K6sL~lq4J#$vNA^<*
zWI<Qy<NO@??I7D&QaGt+wd_P$*)qp{v;CaSmk27U_<N>j>V7tir<KJYH09V#t+Kd1
zW@B$>@d&-Y|6(8dPl%|oR{P!AI9e9G?{go|HTfUTPy6k%63t4)*JY7}l=RU+Gcn&o
z8}ERI&dIcd!;5)tLc}ni?VnN;x&*6A_snM#8@ZA3skoG7_M5`Lo=Ee@tVioW!hR9I
ztoMc}Tb`|3(D=Nq)as>ZV5&$Ux3u46l(O;pr?)Cf_u^bpmG5Olo>kfaml|bxj@rqi
zO=C<|L88(puVO$o5~bfz8@tqXD{R@BTxU4)%JvDpK-#$$U(weuxW<<&h^XyAakW_y
z?n}$-3C(IgnRR$PGN-%c!!6lm*zY!k>f}f9v;-n_Wp8_qkwtE5q*73Om3vQ_ndF%x
z+Ey3_*u`3y<k8dWtwT057%9+5mG|lQdSp)$3O#as*VCl&Xo9Bu?Qcv2yR0go(DPHb
zCCRL(-K!Ese||uou9cm!lp8$pS2ImMQ8}Az7Vzr&Mqt}o{29|^9-XnFnW0HG@%5Hr
zFYcOdca)gqZfq>qKQ4r=GtO`t!9GMx0bJyBFmnXYm${1KGg?qpJt@uR{)zD;kbG-g
z_a*m|><6iWCk0R6SyA|+P0S5w9-F&wo5^9C>cYNNIy<}_$Brf}$T0ir4F!x&-?EeK
zNLWb^NoU#67?*h*>}(NCkyr~oEWAM#C`C7$x(+{u3OQ*#h*&#Bu3WE_MW6&BSL+E>
zrrAeE#?0jPD!Z3rU~7WC2ptRBat%vso9GO*N%w4`u}s<<D;C>+haaz1e^xgA!1*yV
ztx{>KJGbH0PQJxjyGFgc*~i4JH&{{f26D>w@Hg(@nr5R7dm|ndtelq(WeMJI`;Cy;
zP@62N7*f*kQ*j?QEqv%65};nyNktX;ciL<`h_k{eVPD=bS@-q&p;XX)PF()U9r<Ik
z^_>;!3-+KTpD$d{fNYudk%zS|dRn4m&H9QalfLjP`k%fKlQ6M&qV}@~=y831?7oWC
zJAwcBUoMY+jtA%C@%z!a$~M4d%n{C>cZFF*ZFszRiZ-kL@|xakR?V3q-lhvQs3sSa
zn*X;eZ9obyw**EhWGPGlTjP}h7SS1>osGUjSp@s4s>{on&wNId+(F`lw&b62ehJFF
zsn4O;LN>6DIiE-yp!%)xeZ~a$2tNG_h83$mBlqFSmU@Z$;seWOwUeG-VlEDx>fy^k
z;mV9dQ+?9JBdpY&#~}BcvL)m&T~-{X&$AM;TiPC{2DBdzihtiu?YtW!uV33{)nr$`
zQ#b!*K<v${8dtl1kJ*x0^S4X=vq5lobeo9@?|DV(g9hP@WsK<>qfIySyI?|fb4G{y
zqfjPBgj2?E;N|UZ{tF+ZX=SYrPlw7dj1-jH<9Y+HK7!<C^(nQ_Xn|z7ZkM`Zyz5vv
zDi?OIn`Q}xXVQ;xwP$Of6(8UmcMpgCZhMyD-~fpJ>%~*exM>XeqEVOkCqx-fosHL$
z#lC)NHfOdY3166(LYGWMvdur+*B>D&dk!VHG*-;F<vI%L;U<%BgYSxm?VKqX3U?5V
zz<$njp`*X2D)6*;xTVUMQw0O61lGBWQ9k^no~=C-tDwK9!h5P*OwR8xhiu!cpuG)u
zE8Mr>(udML+km10eq+D+9Xp{IRLNZB(p*kbk?;+B{4)N0i-Sn1FE8hLXZ8=+uf6$a
z^OM;|kuwTaanpC5ve+xG>3<4>r>#Wt_%%Tp?_gTr6Y;2FCV%t46_StnA${3GmN|eI
zY8dd>ORh1+9+=DbjOF43FD~77&iiiuvw7&I2`hcY1N9>`-U}xBGUcKx_*hHQYc|cy
zua`mKfvwro;fs_S+BLmRP#SrgQ*2V%&0I`O{QlT1AfsvfZeA!L1#aGcq7Gr15e&%;
z&9Ect&eF_^jX^VxZA;`26mdd#R_AR&wYG+XfUf|E&i7P)-VN&HVCdU^h3&|aWy^4X
zy6(pcveJFB^I;~dE6}Ef_QGf>OF8j?wzfU>*c!fQhxr69s+1A)38*x)TfAN+P(36R
zM7(LT)4DT0(IgU27Oe?h5S4e{%FC;e@8(xz>tLqc$bhtGiKMZwRKEqKv^_z_7|h`R
zqsfBsb)a9*hQ{=aht5w79Wj#@g+1RvfnP%rTmalm^=Is$yri=_`ie9%s&|=EwuD${
zro7GbTK+vX93mj~XqI%@NySt5d%A_uA@TCIl)V{~w2QZZelV;2MT$&T^3PAi+yoly
zvuZAtbp=g-c^S(`G`j5qdttk7FBluUfXIH)uMh7a$w8iXB6|>?ljOcpda(6FkP6ZI
zo0O+R#bypbRcOsYEoLYgnq#;oj*9b439k$^kI=hezl8XXM5HL@EO`7@OmVy^(d{bf
z@YA4LeEp2MJX?pA|G%0r*z~Vmd4xk~V%e^~2JI-v%d*+arq9H?l4l?PFnnfZctQxD
zqUxEd+MB)>f{wkR*@h9{{xQ_(P?Dx$k$JM!N_?o6;_rfTpTGFIEETpyK4UZF@3D?Z
zDB;`G9h$H37T>o!%xu@5=u1#@Rpco2oUE~5qYeQbooKw;w}18nO3354X7<7hG?(uH
zU=OmvE>I=}#Ei}`QipkQkD=}?1XM^xqhj`R<aP@f7)GADd7ws;y8%cUg#Rp!VpLM&
z8ePDUA8bzesff&5r8&t2I;~*-cF#uDFWUa!s-a&yZ$3()dEZmf{T@i8vCvtT3T<)T
zax^+!bJ^kI3dn^xJwk7Ij@*!W6eASU-+OWpu{>kwcF53^?V8jb5_tH@D?eJ#y0gNV
z*_FRsvu|uGH3`J{3%IW*`KoBZ>)nyY;Kl7XpP%|C;Dg=_i>}J+v(`%nPc2G>H<UV~
zLhPTNY!&d#>E;nxHW6%3NoqB(8m(U}w0Ith(BZ30rz1QI7^kxq4*MfA0!GS{A!?I-
zgX4p)Xrb4l^Z6t@{bY_oDbH`o>2q#`0-wdZ6e03E8;7Z9MO7^q4-J(bO!MX}yrjuT
zyeA#UfZKKnDq8-Yf{#Cm&_MIOeSvzC3zoWR|1f7>#icj+c-+N%ODaAy8E{2Z{s*f3
zeb8zMBe6w6k8JCDPmVc_ixuswUEHdR0R*J4Y|-1Ii2+^KT1qv0yU`&al@e9CM$Bfu
zb!gwiA}Qokl!B6@#BI+3klk|1XRy5RA{p|Cb<AWBJ47V>)~<c_)a~!ool2&TU8}&C
zwo6U6Mz;<cV%L|=oVP@Cx)=VoRy)7yT4`el7X3Rw^#&&t{2ko30veY(Tl3t?Zt#*W
z6`WYoO*ZRc%R=M0x{+|#Abs&Z!jGiYc90I)M9H%aQ~zwXM3gfkJ;UA1LgBb_(g^Wv
zF}O<NBy#d~b(s{+B4E2t52n{|-mQ1YciQ#o4X2{tGR2Pt$aT?9j4nT|Lv@>D>8n$u
z))USMK3Nhpqt&-J`}s%pO+f6>&M%mjsZH5P5kpSKt%b#ua=Gc~a_B+T!$*_}2Oqo7
zz-y?kAV}~QVuv#LR_s>mQ`+DHDRXMz@k;eB{exM*(=v|Q7ZUnNa5TPg%<4Tvj(p83
zs9GWo#Jymd7~ztjast(F;dRCg`?9}e2KzqULUQ2l=PAhX)C`%$jS8LC6X5;zZ}z67
zMC^USr@snU*)v-cK@Z34{rJ^uoZK*ah%u)O8$-_*iy=9#dh0z;`Kvn~@Oyff>`e`2
z(UqY7!0(3{{?@UQm6T{mFM5GLmL5t8rG~OQ<=S?E6eCo)e$jED^MwS?`({fScAL%h
z$zOWf$8La=^`<}SPkh^VE+`8#(n|ej)4p+pPYDM|f&OB59+;v|LdV*cp@1aQP#cz6
z0hd1O=W8m<Oegh}FE2bY4>e3$A%w&Y(zt%`rfgYUxS+?zuw$?nRw!h3JV^#8>lz|0
z98XLA3S#2TufT)1w7&et84w$)ij)Q!w)ex5_`jiki`6|<bte0iX40jbW=`D_jSehL
zG^&S19Yrs`L&>b;UcE?lQD+mKV63dZb{;c4E)qP^3aP=HAu5jP)+)PX&%akHo<mfg
zBO3s>+27mdMe_|-i1npq=|?86F$oOdykHew-$+0NxWU9ueYfn!RKtnBJY?04b(Z&Y
zD4(sQaJt)Ljf~ZoyM)(~5Ac)oO68_mA}Qi%J8D6<-4eg)ic(UyGurGT;|-S?R?vO2
z<#$_L8&1zN2cytGMk+7!{P?!8zoafD2j#d)`MWHsSCQ~ka7==I`|mNifka6T=dDCh
z@e?S~%a#_+oWo<JVC*Y0;UhXAnLD-KiK-nk6tzK@qyVeaGGvAG%D@(&LhW+nz3n>$
zdFiy^O&y@i7~(ySlDyYWmZ?2%An}wWNN)Q_VQJ}`U72QH5&cV;y2=t)7okJxl?U?2
zt#68p$1xoqNxo{nXYT+zO@#kDdcc@KpWiH{{KrDq;7pjQHr__QSiAeKe3bj6_^Y$T
z^AJZoc6iTb^H=A(?ev}MB)F$8s3SDpAE{hWzuDs#4csv|E3X~Ad<7;d+k6L_lV}xo
zjJjWk`sQ~#__*r3JBx7l)3@j2Zy9c$4Ll2%$`~H0lO;*)sxRTXvYJE8<xg#r{QU#M
zVDKN8os8&N#ZrH(ai%)lxV7YO!8&$k13Rj_i)HxPVaw0@JDx=zQM3Ee5?NJu7w9i7
zQrf0-{;dwNG=Q$cw~$`s7WQ`H@KMH}8d07YK5SNd%GiQ$QlFy>i;=GFdjtOITJ>^6
z$u2+y-*&6<+*7v+@d{TcoAgRG-WF5(Zx7+}r-!H$SaHO4A7QaOH_V?#?OAT8L8@vq
zGO11!lf9?kAr@vG9k{t~gFEdCYymkG&zA@f-+c8&#mP?PU+B&!FH@;Lvsuskp1Yt^
z{<<-v97b9>1U67Hs^_X^_u$h!e2B<zTTod$I7S+O0n8~QehLM^oMI_h?US-fW}4A2
zr_*7efp4WnHEu3jAMvaK0uo=COyj~(igP#vSMUhWIq=aDQWlzoteeHV{*@1=V2k*z
zAX38aLyiV-6c6B;vjPHc82Dtvs*EMA>bdf-&Xxh&u{D#96Dr-JK5BG>M9?RJSwdMA
zM94U}{X^Jiwriy@N};(38v8qrhNcgpn68KJvH?q8^W05Sz0Pssp~{(y7tQ%Oa5D_X
zIjda|^^RG)da>A=rJ=Z31K&_vT7SN@GQ~KFa`yR<)f+Uhy`#3goBc`KsZ~7fz*!s`
zNKu&WUMXl;f8wkwxBr1d%TG&|RH7a1##6tDx2iY1<aBG_)T+qu{hOkBTMc?a*ku5c
z*dKp*`)oED-pLVYYbZthWqzsZ`7}%<Jd-QL*VP~ia`S%SUddo{dY`t5T+wg7k+!t>
zk-GZ6eB`P6aqlkmjk%&4>=?-kxmad15amIQ@}K%jE!v+WJE>tf{kz)MgHedbJs{$?
zK_y(rWH|ek6|c@(BLOiX@zl*57=MuqUz1*uNRJJf7pJ}6C+SO9DITzCK&Ez80Go#S
z-Gh@-b{-ghjaPV}T3H;XgUCg-<6b{?%~g6Qs#8K1zP!uEwZ;$Fqiir>GcVC^sPeRP
z**=S1dC5L%5@hJ{`xA-p-wpL0oJ(&p*PukkT~|SX*Fo3Ebl92HM`>Fx9T6R)nw=YW
zzY$Z)^by!Tu3z*EGP$G^R_b39?wk1WkQ=hNwXN`VvJW)zTT4@({x-S`F>$3ukcT4&
zN)sQGOKusGGek-o>tb_(^T@3}MSZ;b$oAi&$-OX~gO#ojgd`ZJ>R;!K8dx}MVQo#-
zdC`x+%z}ACk?tOO@8zZovZxL3@Mhmrj}{Vs?sR^dfaa+V4|r>Gm;Q5fx%87@2h+QB
zi39fOI$Riq`JTuPxSppx_GV{)d=Xpy2e<o0E{DLlthV~^kW)Q2&xjmzCpy^u>R*Uh
znoaUZvaPHa^sG)^DI4Ep86CO{J#s}?7j)VH&<Fp_FhEa!3H80)-J9L7t(!`swCt}9
zJxWwJ#>0T?IRD#hnMeG!hd|p9#__`GMYcF0-f`7YllArjlUs$_f0HQ1b2pTjjelQk
z&B^L(5ve6y0(5wfRYMbIc(|Qp6uO^kP1zX!B@ct#N+Zonj<~(7I-_z0(*zY}iCb7(
zO43;CvE}^r$h_u8iPm<#v~FNEuwIJ@;Gitzm3vwS-tf=Qs^9wn^I2VAZ2_pJN`VzF
z41`i39|0$g+1MTSZr<N5@E`<ETQa{Mco|uL2L}1RIkWT1Wq#^_{*%B$z^o0CHp`gU
z;_?JR)A~MKRN5$KPY)@g#?xV#Y9kd&)Cn01qdDJ{@vP64${!MDs&qejJVLUP=WJK(
zk-p&6H}9V)NBH7z?_h$Fgs8=$=vBh~`));uS<rU{>8iVMYO30&qe;@iF9CQ=VJ0yx
zXZxOYsh5ou9;jl6Z3xGi0A<%4PSFJhtbj6iPtFW_JI(H4*?YQl9h6ftZRg1ZKpxe?
zux++B$)M#{kpS&ybZl55?6Lni-1?E7{qi=u>DuA%f3E(+>ikj%>#=Nzm}@W5Q{Ytg
ziuYLmkks{3eekSTe7AtA6!<Est+XCEbdy{|ReR+voVVV!xb%Wb$vCKN1z|`XjeC@N
zm1OvDHy7nzFm8OXlsiXP#n#F@`@n=C$S>NWnvEhSdjkoRunFWO&0CM6pW9Y|&ClPf
zOZ91bnpkllMb1z#HiZ{p^hoo(r@8foF)&1d`a4Bh=O?}C%p~VT!DuDLe+VFZl}}{*
z&B$5hm0vC{8^~+-lBGq5^JqAZLeQNZCsD*usn{j2Vqa4nc8CL#l@7^`-YIex484I|
znGb5cX`)yGQjRML?P8ou<l_OpFC#7NnRJPwpDyQ560ezy$6F&w@a@j4y|Sw<p^u@A
zfO}ero#EV5-Cy*j5xFMD@{>@_f`awI^DuLYMhK~p+Pl_mt;@N;yHS6M2*2yr$HFNs
z@tV7JB7Y)0JHWI8uzgmbO#JF)&hJpl{?NdO7&?Q-T{5nZspd?Sgdm4+!5s@7YFhg7
zg)(U9z>cUHwWRI*Q9(3w?)zTSrNBBfuU^JwQE^ePqZO1T8_gfx$|3VmL*fcP!cUf?
zUV$MsLmxdK>}j=TzC#ZUn(-1uBR0M%m0dmbO8V_(CpWje3sRQW{f8=15P2TX_C`Mb
zZn>AhWkm~Lda3-eUa1m9fZSCVni!&)_ZjiM*1rAMGht>2(*<PQJKl32t9Hv!GmM9h
zY;F8eU&KmzVHo%7!H`_J*WKvG6J(F>SDBNLX7Bn5Y@@uAAvzW|<%+&&FgrItJ<gd%
zXr5B&u(iPyZVj~M%?IpS(76Qha(nkCbc)jKSPct+PHinMa(eBPUX#~{`(ttwbXTJr
zabM433YBUqi^Tr47Wubg7@(}eue8bo%XCntxy7d~NaCK+^`>-wVck_HYP}qPc8J;9
zkGl&&pQZ{QjKY}Z@cPef({}r2a5W7q#@Ly>@MvN{Ddb;bQU;;aLE~UY7Fx2;Rw_Oj
zx?U!a#g0_C)(|fqyIDG~b)bPZ@)27n6aMt0wv&_U{ArbZfljz_kV4qH?Bbux&t#;z
zHUDD%1C?CLPwE}y+|;&0fUJl)43H$F9pX6ReHwnKkXE7m=`-nD<Bg+*VIcrmNg-!t
zVG7<T^B7VQuW#k;Nl-i|o`=uncssvUNv6*gx%`uzR!4kQL(*A4<`W3gKo+|@vfuOV
z3)T1ka0#)l9J6}9evn|PK~GXWtlncil>~GjXt`R;M}pNaQuIvA!XxUO0@n+R)-N6i
z6pZdwGI29*6AC6$wIwYeO7XHiHgnS1b4Rv(<b7!KLDRaWV=mWpOFA?(A<6!kX@UBD
zQT1--T<L&U@tboh$2Mi?1I^Zt1?%9!jGV}kR}O}$N;TWq!QT*<|H!n1!-V>6@nSsI
z@6WxAZV+%aOFehJZEXxF&FhcTJiZ<tgw=HkgzGx4kp`@OAo2v3>+O!bQoVM}${G^(
zGWCB#pWL*9{tYl_TmQVhSg3ORvZbLD<vZj~Ek5J+%E>y8PWD2G-a%(;)U6+|gWLW!
z0MHls4cLP!dp;<JA5~G3`gJK<J`)hM?5^Ae=mluQ;Iin)+ut26Gd<UfxyB5&6-#mZ
z;L7i)_vckC$-9}6-w(fDwp@Lp&A*a7V(8|zEX=005PbI6fqE#acOgExCW9K-9b8jN
z9-zn|a>)o7K#D4p5mm$+5=dD;F*=@6Wm0rtC_K?-53X5+H#c6bodEyACUi&p>bxzi
z&|gHX@fe!1s>iA__D}Q!$3HO9NA}H!iyugd7Vc!Q4!BhA98WJ!q+!aIV*dKNxcXq`
zdz%~H(>K#-wi18$jOL%s@RJ*oHkct6tb4{S@!)`8EAWuRRrIuF45Q^_$_cN-Bi7sO
z-I>;plX>WWv6VtfZrZ6~!D6L?nSEQ>j>AXIkYXxr!g;tKY-bd?naUX~ghJArL)M0#
znwF^w&Ga*TYmV}ZbuqyE00ypw2>UY2#e*%b_Vx&MR63xg@8MHsT5k;GeSeu-85gKg
zdo=s`6J~{l)rZQI#0jJQ9~SVg`(wTJ4j{MBx9o(=wweS7?W?hyuOP~%2iVcV?mY7B
z_`FQ^$QUEn4zp{fJcc=2TvE?9>TjhLdL<3>7b8hkUA@OLu|Y`2uKhupRd#WVWxZa&
z@)*<XKF-2F0k=%IqH?vjRV?6=D*)mZp4I;BL1M)WeqJi6&VoP{UJ6j`egN_8L@ppa
zd%(0ypx%(UN;=4OK3qF*IG${@xL+j0?^DtcMBd~_H=OGITzPW;7KW=;CNQ(!RZ0Gb
zDYF(`O~0pt3f#k{4LRgt%+gdT0I74;zx<QQr;?KbX13oNgu->_NBv*?S-YyhQmlL$
zo&-Fxqx~k^|MQEVx%25oFsa$~n1EPe?G(kl&~Bwbd|d=tcvh~(VJ!dl*)tsq06GfR
zd#2=E(lgEIR4*=X@3*r(4i9r)-#&EcRP6~4XvM7QsZaCc1tiW2`o_&e={#!}Q`RP5
zYPXurpZOvT(VliSzc*W-fuL@uEZ(n3@$H8HSY(lX`hQ@Aq3}RYlC~eJ%*@tJvLr1M
zv2tf{*P%5XU>L64OWJ0WIM?mC=jk0olf9VCq!)_O%rk-Chsev<_huV#99C{3=6=_X
zeLVH%Gawf}e8{;V?B3W6X-NIho6oL>x&e4n50x!!iwU1x{E7NHvpkcuS#;=Sj^_NW
zO+Dm--5?Y4O7(=PEI)QK4Vok-t=Mv!gI^7F009p^k32mlAKX-FU8jng|FUUZ>{2f~
z{K+HsEShxjm{oNBeVb`Vn5XA5HGw^y?NIvDBUk3oFm*Y@v5Dx**QJi#Z#h7BCHOoO
z_nw$<jktGuYv7LV<jPufxpl@>m7b~00$($iz@gbEcMolAxZ?{3)IgGlTa_3ZrfN$P
z7U1J+!;QE}aG-WUw<}t*1*-Yb)sTcpZMH52ZDx3tt@jd&c57AdhDZyO_tCU$AT(te
z$Cr>2^+r0hDD2H)X-?6)F2o{c;S%gI^UhS@Av~bV4_X#FW$K8xX*h(q^52ha&bme^
zP+H)~vyQ^RBUgqmk!M>#PRoI!O|Wg_buCgUpNFIh$Dw3>cfjZksX~6O@LdGtRBCM^
zIQhyRYG<I+4*F{{=e-|s_hw4U8<LKkI_b(s2$bN`xwi1NUUBNT6ixO-H0*4E$Dx<+
zfk)DQ<jrS5(Y*o~8V-HF%ks5dSF=q(H*)ywm52Zsloekfs-vQ34*)L8X?U&$u4l^w
z|EYARk*$3diT?eNgr%YbyH`o%$}r&OC7e495ZL0amg4UXP6qm1?{B0SSIaw>48x1`
zT=-ab@!UZI?M>G>RpfY3krN~5Iq(=K=a|<@HGNH2Uikv=X+sF${IIJ(k_V*Tuim<y
zZNp|q8dT`U2dbL`Z%mI~B}7(pNX*%tI>i~&yRg@CNs7Sr3#BvL$s^ee_POkCc-}b;
zvbjnxr(ftKq^?kJ?3BS>T=;(ImcyoE3xwgVxUrv8v9!5^OTmvE;9K=wbATplu5gNU
zs|<3e=`^4Kclb;Fk<;n7KouW9blY01hVjD4<sfZj&TnGg!oeXnmgp!&TPW}tYb-qs
zsNg#kK3n<J)^9%I?bPfC{f<zwPfmG?c+q9<Aix{3^?Ha0S^l7wlRFEi%%_{mzJ%!n
z_%b6Hp|HGvAOC;M0`05yxD+0h0W1)o@zf<cgKv;+a`tB&Q?;$z1Q~3s_Q;>wr4>sB
z=%v?QHltxD`{4SM7ArN-4nO-tJRJXJ@i>InGm#U-C6M?m78r9%Qb@`j)}>@?;=QEM
z)>|6493rQKuJ`Y!Cw9ot&UGw4`2YO5|1(F*+s#9b^{}08?HLpJ+wf&IG+{9m<2z`x
zU*5AiY~lPUG<oVh3@YBeiJUv*F3__B+(Cza+E@Bs#5OtoGq%%Z57!~&yZ76m)t*Jw
zC5-Ckv^Q9)0zicMU@6oGB>awqyHjrJYLCZs_aD&O7-g<tI;!&PAIHagFHjM7asO1&
znC5f*^D?_)QTM<tqXvAp^D6dhKqlXXIPiP-7hiWRpa$2(hx;j49dhIfNxEqeD*(Y&
zpza8s9GBjhgeZw$^X`1!&>x_|vYL76U?TrMDveNBbuKihmml6gupuwC6{MRg41}B<
zklPw``scK2dW*RvuCU8VrU6dL7Jg%YFD)Crs=|k-%0s=6T<HXIfzH*TCB_r$b=D((
z%jnCY8U!LkMYKevMt2|b;-dRsypx*`Ida9OhUxTx7x+S|oB1gYO!fBro)X%pYbd}u
zBIWp==Yskpp0eBjO@y<(=-`mhxoaZt{`mb6vn0E^&Ev&|T@TORfqM+40|Vi_E^nxN
zdke-utIyCz>H<syO+onGFyL}pc~57|xlOVzg|ge8Iptw52MFNJ1Q{_wpM7?Laeec)
z9`l0b{_kV=-(NWfzV+u(68#QhuCh3%V+#Vs&YH;oKt2d?h1ZSU^9*zdiBuCO0Wdir
z7wf7j2Hu{Tw6;6n5(|Hvb`r??nhm;f79tg{i0nm{%dO{&FhJ*>3HZO=-({8gP#axq
zpkI~j{+m$8lbM~Cy97j_P`OX~<lf@#?=PfxxRT+UJ2-CTxWRtqB{!G*(Y@bhodx%p
zH_JxPwgRW44jNi@)A1Ha)(g2R{`q*i$-&<wd*q6K7WlD?+_8H*9Bf2~OX@whfo!Lv
zp7kgFouNBUtkwYQXce_jY8n|A_Ko9GK4SoFHt-^VKcwP6=~`!8aYs4t%Tew7OfQA#
zHSEypo%Xw~UDW+628;zr%!rQ0)+1L`xOg7^jsz5<n%piWe7Wxvy%lq(!3}YI#eY0b
zrw}y1%YP!6WcSaUlu1g~r@S7X5%%u%4qp5=`>HXz;g6S*Ka+?Z^Wpi(rAKUGbx#t`
zQP)?wZ-sK6Z{&FgcwkGJ*U{gz5NdND$%Jy-{i^vVPslZZ1*DlQv+jtCHIvfMx?RQU
z`upuE6WL=IgK(BzlU~3#P^QrBzbLW*umUX>U<pZKQvDlx4!2q7ew~UE%1kEu+mp}D
z+q;fvp1Lof9B+iUz6GM@YI7J_jY20vfimGykp^;m+J8#J{azq&vG112PPkV=C<VZ5
z<O(u?MM{4i$FUx7ucr&>u^#fWd5`>_YoYeWJMF{U)8c1#rMGWxxFi_-bT)zhB1<h?
zl<pbn7E+RkoQNKHXt)aauMIa{%|bpU>(LAfpan%4sKc?h`UifsDT-p$R_|W;3Donk
z`>8O2HPij{`PY;Pz*GOFRK2G(RPdKaqi-D=oj%{P_0JYeWE>3b-nDooEt3>vUx@(^
zT0{nCsPp*UaQd$NJg+fLY9~JWoL%B%Mf#z6yno6k2e!h-2<1Y+y@Y?@FEt-IvSD5Y
zSc&XN@;9YJm0!4~A2zF2NdBq2BT2VEqM*UgkuG%PipI#rldiK-RvVwSo3>S4&%#TU
z*Zz^qej5S2k5G2G*zn5>w3pEw6Ot#`7dlNj^V9ENet&sEGa#DSXVSwcc#Nm;&pR9*
z8nOHhZnnqEI^0?JXSZ7bjN`q+GY?HudIh!W){`kOQb@mJB(o284fs&%;y#OYuW~PI
zU%2Jo!On>|%y`Q1R<0eDcO3mE-P<a-^=6*-5ATWz*K>8b7n6>?e6PrLR~Ptfv!%}y
z9wm9zku@?X0YuX7*F+#<>q4mctkv=RbNzQkF<Gzm7l3bD_21`un0DKYO{D1b2jHJ#
zGyb0s_3p6}hxzd#&}N~98spNwW0K0@^}moO^pyT}wUm*sx&W>ge`&Vw|DgmvbQM;C
ze%)EV&l1c}8QrYxjZeA+s>7i-pi)AO-e&8p9lZ%INQQVk*IzUa5jOUS|2S^I#L`om
z_?(8<S1uxeP_$4%prOx@ai*H#asN(slFqGu-D5E0j%sJG)E3_UrOa~G_Jh&twF{IB
z;1Uz;@9|<F5+$p59!l0R-HMq<02j!5;FZJy(Y3&l5-5v;;@`O4f4WuaS)uL1;4U*8
zWd<(`o#_)Cy`w@Y^jz_^_=73Ieu=y;xwF>vUWF?pHM%%hnlpd0G5gy5+H?SJK<)tl
z+fumJR)3XI`LMrmYl)Snm$`d@GX?~iWTMZdb0gdG88%v}M`vFW_jY%j{Z^M!i#?xP
zGI!$N)WrH;gS2gy=S6Dbq>;`OhH7Rco_X0Bw%NaJA080<als}@v)x%ob(Bz*gLd!Z
zYa9t`^_sNRxo7VJ3e#8_1>xo@eEZRrd2%tADl{`_kCMYJ#-zN(OLX5m>FozBDYcNN
zR8pXcmFO(A?A>H$#<J2Bjq)?aL4%_P#a!KK#S<NT8D=|oTgKM&Jcfu=-|M9NO;@kA
zs8#=c`th1>Xi4MFBcyA-G`t(Xxs)v`m|}%)S$nz2Z6bfFh1{*a^~+|cUX|qOy%KP1
zLeUs%YIc5&91ZW<i{Li|o3;?R6Xjh=DP&&4Vu6`ipG^#A<h(scAVzDb{So>`F8H$h
zC?gp~H_l2PlI|y6a<DgnC2`yEgJl~lhnv-U+~qy&kYi@>71$S~-Eo1>`}xt5kRqd{
z)`+dZ4=t!k{}_s4hlcGaw6I8JP2{sNtNl)Hw6i=c^x2f)M`FAb^lfC4VtDLsAYUU|
z#cGy1KHYBjy+P5qP<3StYSUo9;=#uB3k{rQVJm|hE)}3^yW0ed_m~etql^3$OB-k}
z^O*80)r~|{2|lx((cCiJQ<6qDy_p4uM&`@~2)fE|qlJW5c2{N!@^(ADmi*0!_738T
z)<`Rso)V3o;#u5b*yawFRi+~B+G;9sF99PT#R$4FxJ5E@*N-^CT&UW^%roDb6mpKZ
z(SFB9c6E>J3JJe5t%=O|@x>D`=cpB$6<zmlY(--Go6A~u^j(=2H{eG{k)yg$0b&V1
z*Aqf{5MS_mF>ZVlq}I;6HU^md`HTG&6yng!8F9e=$}o^0V*anA=DEW?j5#Cnk@T^A
z-md9ib|w=r&;8&teloSf{G1DbaJdb<&jK5s_Y%rrv;k1lGMtFRdDQy`<I`KuHE-z6
z4NTf*yvf5z`HOpAYH2K!A_sQpM;cYnZ?FpR$|p{M*6?vPF{AD3o}d)=!sNEx_S%`A
zCN_}(=M~76N-zgO(1ss}-1oSSI6N>j?Pf!mrbKNdkQ%p1#qZI{$#w%#MgM+Tm1&bn
ztCG{+S(qw1zhT}u-F4JKb&!<27U`^b2^@`a{$V4rU0B)gtHpH2UA|O8WN5Ws8N#Ns
z$wB@0TtlRT5Pgwd%33vh@E!2XeyuISDBzyRmIC-nz$~j5@DG^~*I%PfA4bMw&!bag
z;<&7KNa#Peu-o!=dPR~T-^KdquZPWehPB-H0{5YHvg^Gsyy|%#S6*F#C5C;5K1cQm
zgl7bV4~nPw3&iw=8`|n7-GmY)6Fp+aC0B=H3fOiOZ02{K;We(`$ZD||tjfyV^Ogt|
zt~GSIL3#H)LbuW6mI^B~X>?&}w!UPmY2n0MCqfKh2?F$}yNM?jEBa(36Dq@$h<Tkn
z)M@aPRna+j1eAb$sNiPn01MdP=^??f)KfqXLpyo5@*%TMd55%2R8{cUx5F=8dQwGp
zXjqTtXHhz`V4X2e($t6LmYyr+^jt)Kj`?^@6ly&RZn%ch_%)y9`i{~&A}QJheY-2&
z+2=UK@(`SAjQDAuTxs|M335;F9ST6rvODY5<C^g<?3`ql0VsOUPEjSehj9gveHGJ!
zAllu4YX?k9W+ScuUanZTw5@GhRWw*-v+*J;IL*LiZySS&Q==!a-<p}X>{Shus(s4u
z?S-Wu?LMPI*p2$;Ztn-LAdi{HxzI4x4Ry=SS++O!6OmaPwb)nbTNxEdh=^k;=Zh>G
zwI3d7H9J{vT7sWkS8VhW8b-WtGcVSG&bljIx*l~_6Z?vgg}<ahp3ya7e+Ub`D7SLG
zf7@!Cr0fs>1%7z&1=#zbua~=yG0<ftQZ9Q;aMUwhXZmvE*z0`19Vh#OSh;LFX(2<?
zqly2Z<~VA$cOF@L7v=kh028s-kHk@KsoF#fk==Yw&tZTXpI{tfGPAo<;?ZsxlY`#7
z$)K-9hbtJWg&51vdv|5;d#(L)gxCYlmlbOM7Am882VW2b`p3AidJk&tGA?_8zL?@6
zjDLxW>umI0(I>xZ2>=^{C^XxFJF4kPlce#e{&?~3v0#;@!~+r#{tGhSiQbOd@m@eM
zBxZye{G*#?58a=^0P|h`{saWho9%S(M{L7ae5De9;ny*S6|$hxk<3gQ(Bfof%fu0_
zN*!Yf%TE?dXf6RhFRWpv85Mu<Gcxx8KIfS2u!rzfwD&dD5V#8C#^fEHWFX^CS-dO-
zXUHKGf*%2Q&MbE7SO@SrO2J`_y9zKSmV>w{JIW}Io`^{Q_SEb24<{EL{`eMfTK!NC
zkf_mL7y{0`zE!RCIx0<ya&>(J`WU=~-(3-79wByO?Bw4r&Bc~f?8ehg_t&@SK}2fO
z)Nn#40G^=@IGwRGFyr6Js?#9Z*S2@KQj`ii!S!~6mhYM=B`PLvJ7!Ck?n7GfNPDvk
zgG7A1@F{;4<fiG5I9yZp&9qzl^0|=f6(B$UmKSe+g*Edt{9fM62<I|go}am^({D@6
zTWzmmj|evluM)j;vM<rd56T7)MPh$<%-?pSN}|t~u^uvbySmTV;S=i+k7>g4k1!aP
zamvo4nO}p#&NoZ~%ap4C{}V{9bSnd<+0y3AD!sTeHa(ie*wAP(+)<LSwz-6wAaH2L
zwYV=&*lMx793m8&)28#e?rt^7K9*%zE}A4Zrc0AQnTmnTDxZS<Us}Vk&{$CdNs)0C
z>+8lVpU(D9KvDm9TQ;aJP`bO0p+o7!w~!x+(`d=w6zrO>4iVDo04Pu36IjoD=!}2s
zQ=H(b;a;yexR_frm=~%wpUUDA<tDUGd(f-yuN&(&vlWy8*&E+O1(O>W9^^XO>ak_*
z2OMcQF&;OxsWJg9N;9aI0oXE*XDGPg<l$dU_VS(te4W1Msq0`RwTu%-XXP8Hd<mlt
z6uoyt?#<_?wR%qVOdjsZGPT#~ORgAR!@hGX3obEy(GE(3Y?(gv6QIS=nbA`5b)gV7
zw8J2{=VhS8kOm#5MG@3zy^vkpy$3JfC?;lB_(0w-F|Yd182&R3z5>zd-VF@*((kXa
z8(wa2rQF_lG|Tn`Pf+bFO(WAC&TYlRj#V;2r6k5PNfik@`8HfUdQvGj75k(jN^6s<
zFvU^3r4`)I-z~*(YR|iwbu=l$LkTlx`RuhKtTTAfe}5!lX2&w38PmY%E6vdFdo6#A
z_U?yS@L*`6Es&96Z_c-2NlC8tkEy^&O<?X|gW4R>B9Nn;CV|AdU{{p~Nm!h?_xfyG
zp{vCE#rpf_2;|JEm?wn2_&QT;Ozi|IRPnKAXWg`+3y`GL>Uyf^nDiCKwSUb$Wj5AU
zi!z@|M?Xn5Fozzw?>Jc$Jmu|PFT&|x*%%0z0ywH7OjZ>y5Xsh6N_CQLGC~_3symq+
z+?*0qle#*8tSTsR?QXF?H~Z4y)~F$~v8W!eVuafFz0+NMYd$vmaK8Zpf392V+{f#o
zELL_#xNowJzM@a5Myz%&knlLu?^i<dl5|bkuQOO~CKhl^8Kx_~IX!it^(bE<f$_W@
z*pFj+vt@Xf3Fz6`I8R^InH4YGser*Ik<$|wbz`bqeLp8f+*E6+M5=~6btuB6Uu-4L
zwkW0jjqXUL_fj)wZ8yDHjFi=OL&EdKa8@4ZSKQ&Z0j`U|6i^1mNzrDsQBM!f1%&tE
z6Zp$dB-sl(Ff#+{OzDs$m`bGA7YbE@J+aNYhH;(AkB`igfc9ub3lD9F-PaB`1vQ6V
z6E2+>;P|*Da8IW{YD$?>r4nDxnA+`2tXQVMx7{~+9>GQiR0vF)dP8vcGrR6Kt5?i5
zNeW5;>_TnR-C42qp}l;X&*TFys)9S#97@Hby)OGsago<Qq_Y-|x<{PES_j{zSG)Fl
z5#QgH|KPHf!IiE!m2(;0Z$rXVVHEF*BaIxKu93R^nZ%Wde{PGOSl+?<NBNA$WufJg
zcs3e|5QIWy@RZlWuzf>lH+V3y4$91F9lA}EENV+rUKas%i@Ib*TU~xl(7_LJNElaC
zR0WGUAS6gU`q;;QIQ+s|yiI_)Q*V8O?Q_AemAJLT3JzE_erGMW4y}R~uE#afrqtZ!
z5HN02t(?cWzZ<R2!foGG+AMCL8e|-O(VDe^2CGAKwP=4_MO*CxqS}=<Lx{pe{4fjp
z-b^09jf4yvmhStsG7<c2M6)j+`DyUWkLUyOo8J@Kyv5>OCbz}<TwgC<#SSG3>WZ0n
zgO1qwItWrN?IK9ONsrUykCilx30kbF>=!JSW^-=nU|;>{Fe9<aAd;(L(UY)*dOUGR
zNQcRs5g3|D#-!R9omQGM^h6xJ5?nDH?4fqQM3u(eZOg3-DN@76(XbP!&EAP7tERdg
zO&>r?&CgTvq|DB^$Cp=Ax~n<Q=aH+Si_${j?|$=tmfXVA(~Q$f^rlJ00~Q*mQ8FV!
z>;eP9<0ll839Z<7qnq3DF$$F<(P5GwQdRGu{8>dWCDgku@4&w-Vto`Rv{?X+;qnm2
zOjHbw;b7jPH~@2-lFRgyFMNv8lD!`KofClBE{%QCdw?ryX#1h`3VxR74j2whcY}Fu
zj_dj+wd<{=;Zos_vyu~?CV@{gs<$t+0$Oa_e-^aXY6;^%qM#NkPBY^)#e^ICC3OSA
z@+l^M>E0Y22EO$|^5KdXo+}>Z$0*O>971;rA>ns_?UoA@P7Hr-xihr3OaJ1Nyzc}z
z5!{Lmo`ELXZ{n0(oo?QElu%o)2^svv9t`zQDzf#z+0uR(voACCyTd20(9T$ZoG<TS
zw=zLx8Pj%_yPLRrlXuD#UBprqRf-JHIdC9CYaW<cfOS~yPWSN24PBa^YX{!pP!jL*
zTGFg7KH*(QK=ET-e;S^5$C|@t5rgHqLw&HwaZ+z`nD6JB)f3pxIK?ev!A*YPkQ)oN
z;?^tzLHa75Y$=1v(%?$CFUt0hX9wZ}nhm=wjI<E3mF(jbej)myB|B}+#r`g=FgMS*
z)}zW~6{Sh(3t-+gtS0dDP-=K$>P%;vR(VU!&6h8>`~yyjFr$A#TB5hhuaPidwYd`V
zx9!i-hyB#9Xctum>-!Gi=MGN(+7$;8wF0I!hWJ<Cn7Rli@CLF7wRKrFv(hJv1+}=B
z^XH(i3Xt#=4PGHnZ1Qzs+L~4iNtKc(?(YPpC(?Tncx0U|MR6(D%JUT*X<I$Sdgj<L
z%klzTK$)0Q9}J%9rWRR|Z*OGd$I0z9XT67ut|+3aXKD<*Jry@A<8SV_f}gw?Q}~8x
zd0*7?QmYqNyk08xAC2(3s(}RSX32_j<~j8i2MfoioY$lF+OLLZw#fXg+}Ul~?zotx
zEQ$5*vH7d_S<=IS^nJJV2!)3@-4A6;=Y08rtcu=x5ch{0{PRQ3v#7MeZ+tNlnQa9}
zGFnxOn$$(T#P*Op-zGomerdB99xSJ3yfP2Zu6=DlOhg^NCR8q+B;eTK`=n2E<nk+7
z%efGpYNej3ucuy}_p_GH<#-t9SeK=TVmoG}pp#V^!4BGY=g*LBMF?i($%H2Ge=x|$
zbX!U2Yu#MEa>m#3lAoPPVW*9ynt!R9pnu>G6Z4s07mM|kDrTHJ6VhIhS9dd`p)-r6
zCZiPbc2*JUeE5QfkbdjBI!_w;3uD&dgh!0QIH&nGlV<3N9iZx4arpdR*JxlnzNYN4
zWq1RA!?`X55z3l+7HX3=WXjDOSv7^Kbd;wdLo%x=$hPLOI*j5^3CY>^54}{Cx^NOI
zyyVAHz_hfv?(u<P1!?_L@yHPM&wjVX;V@~WiQnM4v38gKFTLE8tf+e4up%$C9+M!B
zL>$v|C8;E#16<ql9L;S2mUE*ae>MTv9l~@agA~<%O8@>eI|)AVU;Rp(=f|!zt#2s!
zV5z*KxpB2UReJB9Irm?R`~LZ^{9xn|<G5=EPieT{sX1NY|6PxifUZ3!0rrbcB4iMo
z)cr$eBYWE4T-GKR^BeY=79!-JN;p~Hh1w3EzbA%Dn3o(sULovq;$`K_rj91q#k<g;
zh|+GW`mAYp7P%ffS1#Alc;6hY$cTuVq)#<<_`>hN&6dP)&=+dY3fD|*)|n;t4MFoM
zde=przT2!5d7ll2@lTHH(tqqj?GM<^g}JqSn$j0CO`5jY`znfZYuE(c-d$)<EPIR^
z`n*rtlTTTx;e7#YJ{U0$eJ(XzFs`wCi9dBLRGpQ$j`IJ7ffad6EXl9F?u#UVyJC6b
z-b-98)^@S-^W%zBUJXUHqLt%jqQtGBEK|TM^as$CS`MygOm{!+v+j0nk+>_o2Ud9S
zH1mI)nZuHmOw@2N_>*wC^pfhs_=m3zXVP%}0g4e`r_bat{Ga2GN?U@wvX%Pp5wp%&
z1#Ow0w|BPZvaj(Sp9-F^pSMHDo-M^#evx?X|HTFZwWYLbQ>O#_7mp2MFWykwWe@R9
z_fkDk&Zq45YmccsM6_vN#XI1B6M@{^@8{z%O6LT_&!+}d9{923aGMKvR?dwJB>t4b
z#|>B$j$Oa;wb!Xb0(>lY{)Qa*`M3SIdzx+$Oo~w|X$zMYx46vaw$uhW{;h~026Bey
z(AMJoUjp)Rn&LNDOz~9iVz{{z_2xp0*}!mn8_zg<@K)5BZKdiuuKopAS4LtXc-Opg
zt$`w*aSj73*zz2qY~5r~=`)(71}8E)S0|=Oic+f+<G$l#ovHc_S?H_^hNFQHL(yUj
zgFI;KT~B{*%Ew0W>LBu?6vdG-)+FhknJ10fSmX?z@3^Q}A*524#NTyEi=TQ{9lG<E
z&{%b+wr(^fdp#|Al!+Svf=VFFQg9$>-TNp%*{XT?C8jx%9FJ(Ln|kjRS;oMt;2C3K
zcTsZ34Q|X|>ecb%?~XsBvgIPoV+RzhQGLcqqi)bY_T<muP`%VejIUYD7+|>}-i;O@
zJEW3=N!KPlUy;71^f?cpN#(u+JDsmRK;`{gZ3L2jRKFb>$EQ7;h4WXo?gb2%@T0^6
zRX7S3WEDR=PK68;ferihHO)GO^@_p#4^<qmhH3!6&A}aa#mV{EZ5T9%w`gbZVJcXE
zu5ay9zF|{Z$15vHS?eKFc>bqdP^L*&M&Z(|Zc0mRn+F0&1S$CxTE++#mA`$(pk)3f
zFaC^Ar2oRX+VXh++AL)iP*T-xb>FgF-8aN&zL=$<T)pf4oV12?^2rueb^iSazc^oM
z&D{Ze2`C1UpHF>ZkR#lYBgo<8C1IRGQ|Ptj7OQk555!Y<@pVY=U_|IdSb-_NXWegE
zJUj2SWq5M|rrQk3z?NN75iptD`+RpM#Lt|aqc54cN3;oH>eG{c_F;s(x^vzYX1F7_
zr@}$)^Sx&i1S=p{f&d$y`!##s@pkP9#-|mZY;9P*m(IXh9IduU(%NQKG-7V?=A1$%
z?-Wl(^wtDXR?H^3un(iPNc_++AJx1QrnpI6_5@%G<%+WolhnOS<Ff^{8*o*T`1M6B
zTxPJdv90DehU#Z-%Sh~NRM|0%96^HCdr<Ie2UaOC{TRTHAlZ@n>M|ZO0%E6LRG314
zd+LhKDjOGai9O#>4X)YSD3pCv+5m81zmQ<v*5WHx^{Y2?Xx}!NE<h?$ieF^<rQWIe
z>LI3YqkW7*6>h#arT_>j_)*zXpS0(<WhCbc2vl57JAi;piurSsBscA^e$89kUkMBC
zQ`S2s86Z=*^}Hpm_UkiEj98VesVDya_$J>wjPiBE6q$ZFzEq$HFg#B9Us72>%HWbo
z*9yK(y;AH0OpYGh|CMaqf?bHi+iTd1nii;H`Wgh)<J@x_UnLHWQ(j0KPS5&S<ZBZU
zBS4K+{g<c9_vsd}m4J@bhnOgiq+@@IOj3Qmh%_s_t2R|%`?$ZPaPFs@@gIBE&&pkj
z&&^<ddCMG5Q=y`9@4HdH$7xa4Tc1)&(o8?~CP0-q(XN|H>~WJvdqH@8kw%Be$}FI8
z{|!byK%f_2`!#H%zT(bX^4aRd9C)%|<BQ&`QdF812DXB47ml+<amA4?O?N*>E$a=x
zt+S$CU04yTaKhh&IvDNoY4U&RP#h3Ps8-Tv@)>S}Z?^|?`@CCdH><oh@?Vpw<S(S&
zfcyRbDYh^IrtedX#z}o8;KV_-YpJi8HA@)l;iRCKb7w48LtaQVAD3n7Cl=MP-ToNy
z^E`u?Cxb)FpRM^OYgU@ncndSK7tpKNluW;wud?kXP~QIyv=g;&<{yJj$uSU8OWQj)
z;kAGUIafB9bk?j@OYftTmn4efhCTsaJ(}Dy1pIanq0)bUxGK#$0Uaei35BEPO@Qgt
zNbso*b3fp+51<s?ofQmG1p2@iWu8|kACdo_Z(CoI_%K06`TqG^y?$D^>o?<DLO=A`
z+x%N_OLqxPS;ZxTvVWnBxGmEbV=1Nf+*!R_Y2uTryTyG0Cf4-K!DIvV?p3%Xr{PpW
z%j{Bc3rDM+zgNW~`C!GI*_*GTedWngpn$Y}>NNKarW51tAY&<xtdGLp2`gSs1?~q*
z*K9v1!typ!!^JROC^2Ma86-U?R51SmOWN)6ID62$`or}J)vA$`g`#|fl+YNbxvmzE
zc9qYLGq?!WG)C(tcwKfu7G6qe7SCrMbFNIAdQ*v{uuSV0k6wvyxwDME<~j|;XT5+F
z=Gptmg0=El4on<0?Xa6=t&LyabN1+IU85ArWS&snUdCgb_4@x}@4NrmZvTI|wR9QX
zwW+GwwQ5tewW+E-TdM?7dnKXT>azEywPWu+Zf$L`ViRfvK`KE=B+gs!`}_HxbN+$z
zIFEDk8+kvHE7$vVy<XS*bv?(v1`L(;T2YkA*WevY%FN}i#Okc~9T%AN$JKZ_2B1=N
zcbn>4YdUWJq{#?iSKvJ=J$Uw7llf@?#fpS7+!#PrP7giw?n7}iHUSbQN1y0e$Ls`}
z)<oWQ+v74}brykxINYoC!4gn=%|Ia7gZ*zgyB#~p)!V-v$6HvrSY0o)UZ$T%{OD85
zp%x`^$7c`46l48t6yku^Qx)qd=F)(=d)v@Ui&VO1@Kb!UpHzJCOtC9e11QJEuRXXH
z{pZn%*3#H#ea4jj_O>6Vk1XZU%fqVRK1<Smd;7I14ztO;B)W$)qiqiIgTIIhBuey(
z<6XVa_7Ca<E=)8eyH%D8CA!m;`my-32O$r5jZ0gMn%}ohbhi1n8LmTl|FF&MnL6FS
z>G3Pm&8z3O$C<RJp!l~;a<Lod!R@08I}kd3gdYEC`E`8vDI^%Q5HtglrEgeR^LKBc
z%MK=^T}cj%L<i2P+J6$@rsWXv(fKWK_(J{lI+(F?PD81yeDHpU0@M8i1z~J$uG3R1
zbP|U;J>jZUZDWAtMsk~UfTgh&^K6pdX>ZC3eE<)B;0MS_XJ!u{e)B)@OR`^WXy8m|
zb8@musQI=lV?4^|qj+fHJ@{AuW2CRhBlh+sqBMSIH5%k@{JS5EF;X0Y&L`biRkmqn
z7b)Si)MwTvI!Q%Lr;-4r#^6;zX#?`M!b_jzOuhX`#Ts8^nI+#a6ZLfCC_EBj?)Jnz
zm_LAC+u!DwkUSuPDB9a16tGz23Ljvb*EzJ71YKzC(*Vcm74p1tW>G&OK?0r58o<m9
zPF8;}q+YNbF7&RT^^v<x(3HDgqHfje?8BzFWcentGQTBu&K8sHU3JIOgxA?`Gxmw7
zC?d@cCj^kZ(OTFfFX)R&KK2jJ*XZ|Jv{`Guf_9$y>*S9NTpYMNS`dfBzcRJknMDk=
zTukvvx;(**XZxn<WW(}Z39fL>?^5#{5fy14t#znKKHJSRP`N`nJEReoiH<ERpfSLD
z@%Hkggld1;^ih3_gkHuv!$EZ^28L_3bWIDNoeXpPL3c;xDoLnkj=g%dt2U?*42;F?
zwXLj}^W<`eqm^5>zG+bW+P6$v^OtK+9wS_Ih_+VS7O^IK9nPZ+Yb(1KXKs-Cz8dG#
zk4tZES4B&lv#Nz?KIaD;mA@$WsVL9DGBw#~syjDk>goaZ1J%7YFSegplO3Uatb0Ag
zTq5!{35mQBgv&Cs;=@8rj55qymlj0XOAH-;`do_1FQHt4sh6+T{0=OboRRjAFAH4S
z;7LB{hr*V!=lssb4c4yP{HePIEyPFgqqfq*?aA$+X5os&*h{t?HO;ND`OfSc%+DvU
z7viCR8={fiqU^ftD;v$pV(l%<7v$EeJ+qdI(nzLCx{Cpe&QqV%*kkLSB#gA?3i;OS
zYNBLWx;M{ttlVpI3|;-|B7gI1B+7%5mXaZ5daI|x#HSB~F;db?SR*%y?yvs)SbX*%
z3?yJz8#$fYva<m<_X+mn<o-Pe8UQKXh3%VX1+CO|sg#7%L&%#9#+<WFNm7j8S6w6c
zC{=!mfDUMvOUGSUgv49F-@Mx)b0@3T&0l?Jg-g?lU(`mG>l02r4lr%b9%3M^X7}G)
z8v}uYub#^_WD3>0o3qy2^yN5tXec*Ujcc1}SwFIR+9N2P2=s4WbekGlYqFeR5cd2m
zb3NMacNn93ai3)}@VFLaPZLYj9=~nvYP!w(z9xRxXx}1oG%3K$XT-1R8b_QN!=u&n
z>Q=U*uM<#*XCJJ!kxRnqo^T{tv@1XahNL~#<ey-k^rvx_QYXJa@()mqN9a{=dn<?6
zh%s(od$l2ZV0Yuqh7jRcGP_CXwUReu^2S~i0`f2~qo51iS3Co*^=2u+;_sDY$)!6g
zRYChrq?03?<T4MxJqo<K3}$4=BGFzh2^CTF$dI1BZJ4ymc#GEZWbD=FFkFsPhj>@3
z%?U|6XWJmq?%T{aVMOTqb)GdFbXIw9b(Fnx4rC}m+EtV3<=PMJ#&QB-NT9*=G$%e8
z2)J1jI<fm@MfkI6B}A+^zP|6YF=?8odz_(2cwBs6g$C}XRC?@+Gcj#mqE{_;dqg&N
z{|FFZ)}D_tp!XVsj}T{)6sd)D$v=1IUN(lJ#<ZM%G-l0b9n=;byxud{yNG2S18uX7
z8z1dMgIfU89O0`Uf@A#HNvJlf+`1Y5a-|WVzmKnj>F!%-%0I=a<Q*fGTr?0JVprQ)
zuzZ^?UhKHhk#Uu>;Q0}7uWVViosuAdxvi-Ywpu-j^uD4A^V^$#*<6O(_d~Ygad8Kf
zAzSqkDf!aM?^Hg=?@1_evRb#$$&lRr_HvUW8Ir%vW<Tu3vh`zCQm0gcn2+MuW3%pL
zw->O<odQ6id?I-=SckG%t3>Dlhtrsnzbw+xmC|WA)p<gNi@seT+M&;?^gN5@Yp2V*
z>{RkKMY8uf{^vu5;r0IS2lTpjI%_Dlp4ZQ8_030v%Tb+E(2Zhj?_)kp5pAp90j@b}
z+b1?5k+}2DrmPTw?I9m-9Jt!OK1KsV=zyxnK1=_EW=$j@86DM7)yJ$+E12}N_VmE;
z9y!Rzx|_((u>1L>-&o<Y#PS~N(e4sRb^=v8NqZ7h@GaWdT_^#XrM)gT5o<D_{YBmW
zxs4Zj{PD4V9T4Ky;1e*E4*T#%>;<K?`bdJk&t|>=pawY*Q(v>_1)h%RHA)F?;|ZT=
z<ZQEwF9H>}%=_OlJh^bJ7anb=m9@YeU&&{&QCR~>O~IX+jT}xW+qqqFlZWpftWn5Y
zp(y~ttSZ0H*^_=Iz)O4e<Hk<fL(uuR&48yvSUq5_>DIL~e~rz~0Q!9OCB?W(uYLKw
z;@*>*QF^$K;d&2NB%kLQQPJW@F|1x{jNGhV<vXs2T3ZN2dsg_Wy8^^G@99bbuXzR`
za)A}Bk*?>jLtNM>Aj^>^n(y$S`BG2O>NgfTx-xj6F1y!qK>@o%8K4-RsA~@8;VEOz
zn-CHsvjYS|uTMjMj~$;|O+4LKC#75N2Tbi_gDlwm*^qP(5|ju@Z)U$mhS!(%%uoAF
zpG2?q8&{K|AJ<n;hVZ&LySW?ki66By_w;6d9d|B&o(3}g%fDx7HxCX6ElG0g5KHOr
z2{TE0tx5HplNI$q8fyE1IM~unNw@j90Cy^jxH2DhXH^4Lkp`2``)4i1m=q+V8gnAm
z$=E-H!bvo?zT^OO|AaD<b|Ob(3fLT?z<J9~8QodEZ$-Q^J!7$ZoAfAXpvQgwE-Sa<
zrb){*q!0058xiY%=MKm+B{+7X*gYtR_)~M*$fvh#b2h4|jXV|MFWb1tl{-4(C_^-Z
zxT8e0EUem;oZ<L3H=P7R*+DrnzKb5-%KX$V(D%kl`1O<qMPUHqgqhk5ua8<XnDbp3
z;3`VZ47I!aZnG(?aAxfwI&$zt>3%ozH9&=CcQ2HCo?$|2l8eNo&pY{J)-v&9kprk(
zN$*UEt)0eA7|CFHr#7zvd~{fnR?>RQbC?Mtzt1Ltk2y(~IFy<F3|Oo`D!<<b7AqEI
zF0ZePdR|Xx0;eAEj4Srz>53ViPwp&zlNrR30)5z=u;i5sN20Ih;2IyH9_((m8FqDY
z<=x$5VmsM2&y{_aoREZ%q{_go9lHz4geF|7ejc%9(zhX`^$qgvrm{Q6L{c$;zqza;
z##!IUV#B}Zeq-H{PflW;>^s40>C^zZbu;}62A1{kd-3|7@wFP)R_)13U-!)?9xuu*
zUudvV0z`{faP0)Z%ff4MOEP6B8f4flbs}=oz;L29Y2@o(>}ih4qbYVD@XN~8cTE=d
zl6bCeSG13}RE;u#XPRsk{mRrT<zZ~bx6H?^&y3UeZCUs&jP|8BvlY=wHXilPzBf|e
zGo4=#pQG|W!ye}@^Hn9w9{_{lx|<~HgJ~>tA_7LBr-U|s)JP(M6d19azKoSY|FUx>
zSKqtZOJ5H%%qV}yu1*VVxl28}T;}cLA;rB4vXje(8SVmEKWX*?L~|IBBO?+PdMxba
z>Y=4WWbrhPr~XJgD)Ae!oH)7Jg19IbHnykR1C-eU1HKtHo9<I>T*Ix*jJeotz6~zb
z$s0%qi~Qq8_ou4mFxc#`iseEFTf+?}Skdz!;Nu}D@m_|~>i8W(NQ&RLbzRYZ7T7{(
zS^HOq;$H{}k(U;1SKf|06gAL~Q};RTuyzV;Ye!o5TOdJ&3FTp!&Cd-fEvqx$=Ju(X
zhP{Zpuy?hr0s9_Ac?(KVX?c0iqjDM?pL&(!LML|N3Dj?(=|HA5u&X_*DJ!jc!<})T
zYVaRJm7Usj+Bs=o1;wm+@8&ih(pp=?Bc^AVbC0YOdfS`uYVF-LBB~bi+K?+nz(T-i
z06fpRU6`L+M9`7MSzR+fw_-&0PCx!xu6f}5h?Fx$%h0bbX|vjLzVYolZtF?cW%sO3
zfLbI^V1Df51wP^zrYX~{S8JPgvz};v4k6D+zBV&KnF+Qy7RV>A8v@-gh`IqvvX~}n
zv?^h85-unlh~YEog>eIwsN?}xHm7APeTm;do!8@!1Nk#0NFG#!TBBDvPkDe*5kj!0
z%Y9<YWS^i$kD2<#RZ|4_;d&78)QoU8!_9%H)WbX(d{*MgQ}+#m=(|(*lqH6ZiWI(^
zdpD*!UD$iq{+L|fE9C<0!uKW`2HK}$maSR$OufL59bp_X*tlgE%Y={JqRR!GiWzsw
zk|j!I7?aX8jG(wDIa5c`CP~eypOchJp#Z)p>~(UO&3bhh^p3V~=+--H==u@hMb3KD
z8=~hW&z=zjfF(7uhIsdGMSKe2h0uM*&fnVbJpV3$(v>K5{NOH6Z%2|=Tywi(+~7SF
zNq!{{L0++2o1<GLfe7{!>3*l(v)^JHb}9AvFuBC6pAJid2F*)2hUB*o)<Q{;shCxr
z1=nEtuVr0^W_4K)t2RPQy9on6AV5=ljER;~rxgxx@f>i8M(yizx9=#wU-<BTC&`J=
z%-!MX-4l!FFC3D2!89C7{9WEG#qC#f8msTcDf6zX<dG0wN7}GLBJTI#SDhl?j(B9l
z@fqZ5jt;b9=2A1%bAOYV8MVpI_F?mp`h2}FQ(|&AD?)CRUmy&>Y5qumFK-ob{q6Xh
za+vb$%e-}&O_r%IdqwTf+Z0`jKvN=y&Y4IDmwzl;zBcu>$afLWe3J_LK~VDCepCD7
z1we`$rJV`5EeD&YZR*$0)gt{G1qHx8=p&x_cnm^f*ogd?GSJm$)})NAus7x#*u${s
zZB*|92tY5kP2wibh8~q4xj$ap-g`PbTtTFSf9`bCD>E#VRy`PVo_r16eVy!evB~=E
zUpuVdno7TcQ9I#6t~KTD&fofIxoac2&_*do@&)dj1pf|MFg&PCzoraivDBuENDh<|
zzIx#2QxhsXj>We39dv2OFAHoi#}d?xTK`mn)W}{biH(~!)h4?G_fY%FP6TEZAFqw$
zTqow%FPj@9cQ}iL*}?bLLD#F^h)j@1nx?8}6bR}nlS0{+<I_WpK!Z*Nm4h1_?F!o1
zADbkhH}+<LW*ruTOudmu>b;J)S!yte;UIQipY3SscEytl0VisQ!@c8TT5g`}c=H@<
z48C~VgTsiAS7oyKVz~qNBKPWuO&qi?EZ|<31Z~lfm>j{zI1tpY&wNS<TFsWrNW<%Y
zTb1vlGas$p?icA_w3AE70sUOoImZrg(=3B?XgsN?t}Q9i*u)3ORCUvx3DJH<=*^aF
zYvrjfXP9A>^f{}6yg@g0)7rHc+Eeo5PRF_#aFAiJp&<05_>{-S^$?G2^+o4wT`bNp
zuR2mEoiD<VIzMC$qoE2p0}SG!RRf|U$^NIpp3wX#qdQHZ_7z)K1<f7NJ^00+_G?)C
zt@|kfFr~zoNPzJ~#q_n7r+P&Erj!&85aL($``vEHeoVY;^6T1t87eA`Q%{recce?f
zVtjOt`+V5u2h@9JL?w;tD{0X1wT^}oUb~oH&r-BKzi|ZD9V)8OR9J+%_U&UCj6c8s
z`Q~ckJ7Spl(am`h{(UDk)kDBVgDP8O&S|lKxVpssr?xtUs~Q}j#d7Eg_;CjXA!|~E
z_p`b&;u67+?^>H-H|GLa5+?DInHm6JDdZ`TmxC?*Xx;k&4|g+Bh8qq#KDSuvO&S^%
z!f&xrQQbNHJ)3_+Xo*XI<QF^)j8Pmty9N0*v3bVI(yPDcKb=~24f(mS8jzV)qdRW;
zw*12>`+1$YQB35bg)tQ_{gfB<kU7iG-q+M;7%bHHLKLg1ktoXyP(u*APBJF{>o;|-
z2x9+0Ge=jfqN#`>r*q)&xSxBy;7om=&!~Q+{eF@l?)0cc{*!W7W7i@T6&K*yL-q9L
zj>PcIQl4o@?Y%!bavpM+T{uvGdiKRqf^pC4l)sbyddPV?{n<DFC4{Z~80H4#HP(Jc
zvf)@X{$Y<%-DLX@0}S|2?f=ed`q$JCcx(LA>Dc~#{lDw_Gq2PV-)sZpv4KF;6P3Sb
zS$UMgjO`E5aPeyy1ZnX-3`jQegq~?SW?BY(>b~=rNaJ?*w)ifx-DbkthNHb<tRDOG
z$0#B=`71qDTcNf;_>TqW&9f>R-p%&X#M1gP@hglXslv<y$AHuOB<(8vr)~<$`7vo(
z{3R7j&+T*5%BA|c&KciNs72pI-;Xr5|GFC?*~Xq^^p}wvU~eok+Is%z)U&FC7eI~t
znA?K#e7i`u8?Gcg9ej)Nt<@jmRDD#{G1mX2`aw@%%A4p?;S}aqHsLYK6RR>;om>Il
zRNIrlpq8t!PyON4ued&Q*^^IM?QfyKA7_wOYz7YvSN}t10=$DZ78(RjN#wIhVp<@8
zjrx6OE5X<L>FoSX&Sibk`ipUQZ>}%A4ZZ-3X|am?+{SZpjVbe_qQm#%>$HtWfGi5R
z>rHyW`jxZXX|Ra;^63tq>4CKMSAWP(91`53qO{qeLT{q?olg}4We$^Fx0~{LNunIR
zrdRw2X5VM^?eMy1o>To>!!g+^hzm4uAJlx$+4HiKYOMZg_*!21m#L)E_S&ra?wu8Y
z=N91~q6KD`xP^1o?IuyU@IkvykjdEP(j@T{fb=jB=(un7m1|`u&VQJ2(3f{>eZpMb
zqUzY*a)$xY>w7$%cd+r$I1BLY0kt_)-U?|;KjkSYl3ai3I||xgep}MpDZxx8e)>Z8
z)&Kb`@U9E&L^e9lX75W1z#V<5pSp`HDJ95PF+no7SclOK9Rg^znGgxC3YfXkaIKQ4
z9{=Fbr~VqPi>Hgt5@2pY4%Sw>{fC`5!6%Yo^NpNe$pGW%DQ2nZkaBza1P*xmNmEKr
zN~SZ~@*azhM6@xtv^>%-P3<I?7o1mh0msq(6?~LIz248?x+9b_5y53YcuI;&N3{`&
z`CrJnq^p2Xk|1BlVSrWsue|S~L6nkoMrWF#kPeEok;){oMJ7*DUhp(I0zuXI7~l@s
zt2-{Mw7n<Lo;*w}!;onm)}qw6sY64Zrz4$&y-(Rg3jk$$07xdZ*-HRrDjfJq-%Q_<
z!tDCX$Oo&+!}hi4z{l_WcDs|9RC+##aM=g^>-mN5YP_gQ!MBreFc^%>gn~G^I`5A;
zL7tQ<W5c(*cL$!JzC8A_QxE}xZ#qa4Pcj{h2Pr=!gfNep!~rj-TOsdRRWvw(U=hdO
z)2qWiXD7>2x`jyGMcOs#{v2<jZvvVXxPN{M)lj(C^|zKsS?#|%!;W=Ng67R(-4d+r
z95H!ehh1NWI!6J=t}=+iw~I%xXa-5S1*w41*yV_Wpq@7UL<^8*!s$M-HjHC|WKU}~
zXVIN)J`_Ru$7sgr{V#I(`)+$rsCHyVwW1-B{jE0Ox4vHg`XS+T-wLw4vA~w>XX)T`
zmM$%*ia14#1js(*0CVG$4ek#X0*c2pQMDXsO+#NB#cd7r&w3v5T|f-Tgu_D<(bYH|
zW_mZ`!e0ps)U7);pq<j5l7#YEbUi@B3H6=jOQ;)Z7K-!Bq3<J1?(}Yb&d3g)>Sby1
za$~2N;T_`5{V)+IBKzEW1{PPO_7yd}>T4BQx^;_XYnQgX+UMt~iBl^dPs9P3(F_ak
zOH4UEdL}v-D$=+vW|lZIe(fe!Xie)NkhKYHXSTIC$z{FHDGiC~gy=6t`00UQC%;xW
zt_Qw1AENZHTTQXwvI5aVtcCrkf~Vg%mDh*xM4pS4<#;2HzEJqwv|Q}=O@Y$#f(^j@
zDOe={BS)K2j^6fU&FVH*q;{#+32KP|n2IO7YO7h>-6r24ER&f{Gzn6#IKE?<^l-(L
zK^aa5Jg9iBsXae<YU6etwN1q!`P7C1b!dQWEaO0&s;v3ZE5zzTE;*129(&RlIh5Dt
z6+q3_<kpE4fYbOH4_;0hP@^ffo{m5%rq`t8NA|(7p!G4d9LC3CD8L@?H!AHhshe0#
zFz003HrHrfX}!CjU^maWuJB|&>J=I_ufMBzx>h|d0J`?i5Z*fO*})du61UUJ9mFX+
z$0_$uor%a;&Eb)OOz%rBuKGW^(+Josg8h=Q=U8SveII0BK2H^FL@N}T3J_s)K_46J
z@r)sL_I`lP%D^9+6)*7DZ7Nkv*5k&kOHAHwCdkSpt%%r-5ye}!@NJnW>wZAfF*-Y_
z8PHKNVP4k*qJ+1TWiZ#kN~Pxyr)2G0`FlkPXna>YM^B-I+dqCN7t#!Csn<`*7Dr95
z^^MspX-X+x8@c$Ux6C~QvvTD|uPXTH2d)G3R)Fs73+R6Yz`qR?9!tE$CJm}z-|P}_
zkWe(4OiV&4if>dCKQE^PTdTh_l@b%>D8AABV9=@1S{d~cg05<X2nAdS*T<Xql8gE<
zUiuu9uoHXriQur3n1$Jp;oyz2<DeOviM{+jOzblEWzsE%s7VJ$8GOB?;nDAR1;ccj
zq$?wdXPm`TvDT?|FINajFFJEG3zv`dRg_q6y{-ZIAjv+fT?#iV=?|;i%Ag%(<REE0
z=M1Uh%4ApAh5*-}TOTH`NXs3xljLI^Ue+{6DTQTTpD<8sFY%rrIE)Rj{sztsd>pz1
z3hNwp^5tQ0ki=0E+D~8}gnTL*P_D_oYPu?qXiZztN)SA~+lA|MXSe~w>;P$lX<e*I
z`JXe8lk##T%%KJ_>>_S^A0YvP#e$AHLZ|ntk74>He?UGOa6lRTM91-u1xn)*Ma8iW
z2aX_@gx->0jCXL4?6!B@)RP>R!EZiH-PPZ;Gb8h`KW9$qxAF7!Q);Ac6Np3jt=w-k
zaEHW`rS%<zIDQ|#c7bD_pw2Ksvl5vTtp7DpDF?`g?zu2d`iX;BwomMZJ(IG(la2ac
zZ!({GJzTguW@sjVm~2e|%Ftd=?1Ws%2&Lr$_&&dZ7nA2HQDR^%4541-jH{6&C^z9l
zCyudU{uNGu<w!|CR6Ihz06Y6QMTfR|Z!t-t@Yj-J_@e{MvYzk3=tfsW9?j#3mL}B=
zLSxN>k}lnqaBQtpD(}Z)=(i%=7;f3wtaeC3a%y={dLm?I%dR%7xIGcIinw#%VQr~q
z09r8i0&eG+Z&0aJQox{D9+bCgS}Ny*n}m-x<d><4kKGOulD^?!nCHXO$KyQXJ<$w2
z9+>D6>&AEQ252B>4Vzk0E-_8=ooLeCwZgJ0e1R%2@M+aMrXFw}*vk?lRh;j1Z}rF<
z&720E=dn{DgIfKX+$Qn>i|C!dW2zi&Q=e*gnAE3MyuDSJ<r5G{5ErhN`;BWO`?VTO
zifHQBl4U@(@GOSsnaPQyUe`pxc)7dUjmWnnT8J8zK<Oxb8k)VZ$q4WgOstIb=WR@s
zguGwOganSZ66}ue!FOry&1;{jx5vLY1~X22L8XDB|DAQt;7wWWvm%>B>u`SL0!7Bd
z&{2By0Iue(5GGE9XF{Bi2Ak%k*O7!g(%um|*nLy0IcxV3C;ob)d~-rk+Gx1sSGq60
zo}T_c+TPV$i@Uk1oaQKG|7+sI5aR63pcE*OyoVMp8$nJtL>)!4(keYG?9q?Y7Ruc>
z7K4T!?$2%<L}dO0(Vvv(WSql(nalW0$EKiPh#mS#0vu<$4y!#J=!EusyTwdxAtnnX
zal+a!X1z@25Ns@@iSCXLm5N&c&A2&88<k#9=;h4jH@0AU=GMwykV%fcG$TvPuCLVR
zQWa%4TyY7Avlq0Ro99)VXwbsbnOccl1LtSbco!7(5uD}85XT3B8X4Ity#tM|E349q
z%S#tCKVJ{=92Dul3Y4ieRzs)t(E4MQzPJ1mr?wdPB+QY-icGmXtvmZ?xRGngOZlAE
znzYyV6Ir<wfPEM1u9B<@L!2X##h*4{Ol2HCj}@s)H1a;nU)zP&gg&*fUMQNP1tj7v
z+0hlNnY%~R0ur(tDQian>u3U@zEMja*?p+TWORV#YtJLmVfUN>X+3!>TBK`4T4#R_
zVdq<kqKzNrNvaKg0cPbiZzIO}zTuu*huJGHyraB8;6%{#Nx98-z%IFDMpV08&2F)H
zS?)Ukdv_KU73Dv>4-r=B!9O?golEndxsEZ#Z@$%ne9jIm;w2$2Z1RGYO3fqpyQsF?
zvHr<tPxDS&Hh}Gsl=h;w*w$I}s1RgIXz3~%5u=J8${bjDmBzYH?SluQa_a%7+NMf!
zC&aEah1G;0FEHOn#|@FL*!Y6j()9)uG+!|e9WKz@bL#sVzhZgVzy}{9B<5CYvFZo1
z))b4Ln_hn1W)okzj)l;??3<OD%D8z>w3k&!D>+mSp_J<P5W%8x<iP7hC0r+clkUy)
zqrN+cCHx`T;A~jG<N1|2<{M|<(DKqCevDtAKgvU)M}vLJ$$m7;RmQ)plyf>cz-{MC
zBx<t74b>o~`s}gRgr+yqJA~sR)F(P=NOdzKpez8%^i?{3F)FlZTH3iiRM+2EfGWal
z^@s2|&-kD-Hr*b+Y8$v3%%H`hgAqA#&Ws9s(^%s0yx9?!P?0dM=JJ_iOE&r)Xcs15
z_cuWWIu$n<QQ`L?(a|{c@X5|e${<*3f}DVRk$rKLHa+11!Gm8!x>v{y67e<rsgrHh
z81n^H<DI0l*5On`M5Skn=K{($wczZ+fn4RaOEMFmAfIRcy*XpS1{<BXPVY?NWeM)E
zT<N$B=~L)*=yfkO=sfh(L=!cFVeSr1^z@>A(%zc4qafv;=`{7{9qM9+wW75F-Mx=Z
zz4{a-J`X2}&^JXxabj6PbOL)g>s_cKgLw+-H3gEou)t*>)g90vIYDW8^?TehuCh<N
zDlkm*yu?%Vv^t()J<oYkQNyj0wpxDdP$TgKNol!P_;P5&MzkNnBqB?-t#~tzpii<K
zZQ^BpF7Ogs&czsrkC2iveJ+|JzAWF)ve^~58NElsoL5~NWYIR=ujV_9RyvMpWVrhw
zL86&IoOy01xz~L#``oP-N$2!Id8r!eoTDlW^NBaKwxvWqdDaOwHFqj5htaCN%KQPp
zLSaI~{<CCOg#GQ5rK{lie(Srjl#ikaqa{IBBMc=K_oDFP2<<DQvDSP$1q0O8Ne4e0
zwv=pO=QC4Ref4e0f@pQ$&u6s^Y4!8M^96`ZyewRBFyLCf;6bJD9=*W6uuDl&%TbMs
z?1Y5uvRN!}R!DJs7=C&WsEA+uRB6Sfy?vjKDC}lRtb-_nO`SE4O)}s`ZP!)0A8-a!
z`-TGvKu2F(dSh*^)%L!Cw{p_2c6~Y;9IfI%AUsD-b%VCvejbWSsp+7K#z{yFfM&%1
zE?xA;$FqefrJ-Y?FiO=|_q)p1E?od7k{|y=#WM~qp~7duzW)MQHJz9bIwDS7d0s#}
zfrzYf9xYk!ycsoz<xh&znQ_4D(VxcdR$^vY&_sf;E@4Yb#Oi?-C+P3MhefYK$ftnO
zqdPNfDo?!VixHV%h)nc$gpXM!OaO1V>`5Sg_;LsP!O+5QTBBSANvwLNbjgQ%hHCo~
z`?pOR_3HC<Wj(-3x-m4YEU%&wm!;)e=6wljS^=D*{cqj}sjb(%tmRC$#|pXwL^k`f
zeI;Wu(;y{l1Z~$1Q~}DY(HfVjO_Xm<u=`;hYwV^5_}nZZttq7X(;~8X4WO2;D4#Tg
z-SdNiIwHDow_hYr`U@p(!#gpz(M(Dk?Tzv_ZWd!O^69wU^~cO1QBw_O%XPgeOM4eZ
zqNDk*x$fSzVYJ^Pe(*{o!sVlK0|q5JHqLo_?=q|#v1@25Gotz!W(01&q>|e^Y<kB4
z%(w2x&dz9YLv@I0t!rK0hIw0C?U_7%Tk_Vuw4BM2o<YJYE=IF?3p_DQ8?VIq&DG4K
zAZVk?fM|MkGdV8LU+v9%%@Do?8~~+fu3ngSf`)_hSIe%udb`a#@a(ZP9KiWG-#0P2
zh4|J*z9#fWPnY{Yv9%?e<QA6EH%#1BPMEUXSZ`<0@={NPvRA%|C=(XVfFEpVvpKC?
zoJ`;)G9NTf@5ddHuUHTXtE2rEq5%lrJ--VF@<;d5s{K}i#_HC$Kg~f}t{e|}oJMS6
ze<SdiE_i}Z^5H?YZosf<Wey)kQ!Y;XnekR(2X=+ko<o=<9wDheUjm?xv=?wkG&DuS
zWXn${3$QqweHl0hOmtM`7J{mPXL#g?e71=X7w;ZcP;QlaoszCiBzH=PzRaUz#N!@)
z6z7aZV+tAJiGy{LRtR0Ds{E8_?xUVus{=XYFzsF@-Kdw0v;=o4;oN&jcc_q66v4#8
zn#q%(*?h9JnRQL}IeMJo%cVX-J!kLVpM9o-gSfi0Qb)hUB26BQ!r<mM#(m*AB9B<J
zGji;Xfrqf|#cn_>X8%Ta2eDkkVE(E{soE?<eTR3A6I{x@Z)N;ziMwm>PLR=0rP{0g
z<HC)Q%3=s{Pev4{qd}T0Mj|)-jLVW*&{vXSsswpJ)HKaWJYZRi<iB^S-zm^G^QSx6
zCu5?bp^NYa3<>XIKa8KiftZxwiWnfV={_ALxA96`<qjlh1mZKL?@jGyieiJbiVNjz
zW0qzR`yz=~ROv#C8E#xZX+62yzWF+}!&?+9z7+#|D(FbNd5lfc#=9xgJ{t`^D8?G$
z<PxaY0T2U`>=4FqSJ}ITwZeip)RZ1<_|oHk?!dSKr?GIF+s3{EMx{q`@b5&FBvuh~
z&OBe@@H0LJdDoH74D^a7kW)y+>_&;-p#$A7wWyq#jykR0_Mfw`^12EfX!Sar@g+=u
zat$Kw@<#;&CVeli-12C_<U#R7lGJL~)0<!Ehll#aHvQ1GVT*PbF?Y)&6{cUeVIDM4
z1fM;hq_xCslB+5(74kRR*OuP91hGqu9g}N;S@kkDNyXoa$@|(R#pvt3V^IpuDFd;Y
zii&Wa-H(;dG8XQ3AIy+t{*1`k?M^IEIPLX3Y#p*<!fCsl_5*X+`~_yg%d2MJsO(zQ
z5}A8QfCT%EGs1)P8`%S{dziZ4@o%ChuzP`V7Wz~<roj3J;{-XKVx!vH{<!}V2b2r1
zOK8)<j{ys;675p~Zzs}$GuMxSyKR?cn&CGK#Hr>1UMIE`+4=oS7|40?S7wPEYTD<H
zw)6}-Aq}jGxx`yIF#eR&SR^fGFEDeTD>v9Y=HsC$pY?U2jsQ>FGIW%eZ&+^MUf1C*
zZ>9ZJ9t@$0&J12pZA1(2k3Xw;Z@w(7nX^GAA=F2HP>pty$2nSrz!qh)(ngWCIr-kb
zBBuXvwu!P&$BpR=4;9g1^)QYc8TiT`iO)AvIxHU8X%K(@ZxU#naLIBZS4Cr@Gs{J(
zUTs#j0lIc(R?b7qtjs|>jDPJ6Sq&<e1G_jD%<2BkIubHKqp@i{L0DuyUGL)`x$K!I
zbeC7kcO!yY#$DRHLl!Bbgc)f->#!4eiX2@azyTB)_|g&ilhAQ}67_4cci=H=+w#Ez
zn}q@?C8U(Ydj^+2&>K@S)V|ZCQqC9hwB!&=mOe^|!rx)nai4>3u#XSx-xuBYO=XWf
z`9c>OOfvv9c{MW=eOD4DW}TTowyA?5EE)kxXVmCc+pxX(xcsdl;e!X+Rd`7bQ~X(~
zq0Yt@ebiOyQ^BvZG@st@``JZy4tqO4X<fba{`~~AmT5|n_36F1?4!8s$@D;Nkoz9w
zFS__auD}Din(@Z(+oK^wcEn<1W2Q?xa$vXSL2RgFg1(ImuBf;^X*or2BvGQryq3NY
zR54?5qhyr{@93P-*U@A&NlORSvBHH7xcW>EIo9cyW5a|ljjy-dV8p>oDO~oDqx(IN
zf#&4jVjQjUBm?NqC7IOhXfLw~pjjF7wdui%xyeq_JInBzHgLmcP+VeQzq21^eJrb>
zX;f~a+?y5MmszYIw>8OLQZi9rSZa-18)NCKRHq1>JH0i-HUPvzGf{*k{)xi<VW@Jh
zMOBHR9akIoKK->R8iok5ls|gnS9JM=o!d|`tKD|uxs=vPGg-$YKA+|0-y%u}jF_q;
zebUL3PqeSEw2|f9|1SR*vy&MB54bZcn!oC<Mh+UH4^*Un`|<}2+umEgEVFHE{g0~m
zwRSIUgGutCt<?=E`$t-(zDtraDQ1xZoQks`M!AgO_}Yk}%Z+>Ik5!n1s(8aStzCgW
z03du*jN2mvVyDkgw9mFWcT||kh_9~gU@bxBd!&-c)j^4rhJ3vF0+$t)gx%4<*evIA
zR1TgHnIMYjZk+1;`xGSQ`&*1mxhLvYBxf&o)xU~%wGHSuh(6EMz@eOs^w8KMIdpN;
zL_s|pDEB4ARDF8f*)H)p>5tC&)%;VPbB8BS!$u2Sn4VRl#2(zgZ3y%#hAnjxXpB;Y
z0d!PS1D(NG-nCf=8|1t}&QtFuIPn!_;`%;0%6^k@SdXJM6Re|7poL!tn+^L+r8ORa
zF_1l5?;+0*1!5JM3@$L9tHoPx?f!tfrgdK#9Q2~BQeGn-uvJF$O8I2mN@_iIWW~2;
zgxA?GIT=OGH#l=xud6dv)dk=B^>+=m&EYMdlqgIEXMwmxw#BCOaz`e}Y$E%gxLG%c
zca@HdUTkI{;Wg!)W~_{qw~Lh~k~i%l=6}~TYq3L!_-7{hXtA%Ok5}76)5PJ#v(B)`
zd9$DCiovODa64b8ytc=>`>}-7=fd6c;xIeQ8`j&sFOJ<uKFGBJY8PvB@|<#@-W>_Z
zP}4yB0Rndhh_?_9rnf@;txJ<>Cy;>4f{Wkkzh*mNtcYz4=zRB<n+GMV{o+a`ruxPO
z7Txr&!y%2&YEWpPEzREuAV4<nLZRiLhsBvxs+GUm35>MkKI!2w#R|vka&>@^$67N-
z=a$MfP(Mc+$5tdQiv;P%Y!4@9bPhL|TqGBLMP4?x78TbiM|gk>BDfac2Ph1&crrVU
zIvQ<xkcG*9K_S!ud^ew(+gE$J>?R-FBiv-LS7ugFDCNq1mZww#cdhfyT#Fz4h2LU|
zSCTrzUAyd)vKv49YK*$iB$LX>U|!KfJ&`if`RTCTil*=SObDQ0R#BBrH=iX-10{2s
z)^k$biqCnUV7$B7kdl(ic~0X^o!qOX7HDtRvmJs__<A}yeVPTZ%y`v5?n6GWDgbo&
zRBz@19lmb=29Ihom<CXJ;d`1Og5EGql*)yf=T?-_H%_><68irt4<iq20Av29?|Zym
zW~`k>Q5aA71mQR-tijiKMS%2MtJwsB&Yr<(XAkL_4&9mYp5DGBIRjh96r5&)hhV_E
z!h!C(n)_G%@raDZ5&D)&(0{-x>}S{j2YQQT@spS8k#MCPBGk$e*2cmp1MH2IBl?S5
zy={P-b)_l``ZCzgARh|ASajtIQ7wQn77)61%h=5eT-W$1K_|8kOS_Mp^^LQ>^v^E|
zZbSAb8U(&6E64H%-EU$DjNaV^jN@taBf#J_MwhJM_`v{D^32gG>_q<6qE#It%eSb4
zoQgHJr__si4Fs~XaKwBJ|6Jw!bWidjb!*KW{xvh{0f*_c4JEO9J7EX^$f3t1pbekW
zGJfd$2oNB9o`@<2B_$=L3Y-)vh^6LsPMX-Spq<SYoat^2EH3c_C<eueaG%7%QzjWD
z{<uvqN%lnC-ArY}(2(hw;nC4=-R1>dd~%r|ws#!wBYR2KS5bZ)X_(TIP;*@_syHx6
zi5wK?+g4Z6{UCePwpN(`qA$f@wb|-Mlhg1FQK_PgcvUlcdFOAC*|~F5krblN1Bk=G
ziWI_gHZ(Ozu(&+hTG}dHN5MLVyiP3eu|mCxHn%#e=y~^wrfpFywMAh)U`h`m6g4(k
zD+Dq)snK}aUbAj)5xAN3&YCU{9+fw96A5_QpQ&m7I78fXYA748LSHV(XpGdHd3TTm
zp92|{CbdlFy2!fsF8%B;+{!1*HZvys38$Pt55z!r`=67dbn#YKm*rV`anfcs`Kh{O
zh2p&IJq|jB8vw$BQm~F_86a_i%acQ~%O=a7CjsyLQvagTdeytu)xf|ASv)=Vq!PIg
zh|*F~-F%2s>KR#q*ksk)uQkIH5*^P{#d}oo;dkYq20^7oPcOCP`g0T}5<)&YAnkk&
z#RBbfZ_=kPLpwE2=6_KUMIa~6wSM?NhjEWjp#W5iw-1B4`WeA}ga|<#xuSMlkShBG
zfUK=}v|Ixv5XS#pdRF!HQs7m<p3*KRf)|}|AV3ULvzDU(EbuXaRz-03tUqK6VQb*&
zFZupw(3k&=nTkqgyT`R=z@>AdflvG7J{|k%rvPWFe>UI0Uvaqh;*NPz8kBCM_^LUO
z6F^`+ImKUr8FB-zU@x6s{NWUa@L#W|g8zQ(pMSVK{&!fXe?2Vy_hjzRTZPyEKQiI}
z8qCG<$uub|Nv?{rM4%Xo_vG^ZvZou$fvk7ULoVvasytx!USWA8qM_PgxAXp>Qr6`D
zaHq=mP!0tynRADa+s|D#|CO_P;gR_xwlkOC`0H`9pJ9Jn9@@ci=7G$i0tgvp1b3N?
zGHP~V)m;$1*Hm5F3U?_rf)`H|Oql<P6)LL#vw;NNKd?X8^z-C3<!Cv+70lr;?@aMs
z-GrEtnSvQ)WFP^`0?H?rF=LBfmbRS6E7a)Vw{lGXjy&)U*NO?2^_(hHDf(TwE_0KV
z(A9FiH;k`2&Z%0fp5b0Do(LUR{`cbOVgQYyP>5{Y>yMI76W;^8<yz+A$#K5FTftLi
z7J8`CUfkGa<|bJ#dcjr}6uhQMUcm4(s0tNl+cC-U*0uitOP&y$i+<N0rksN=rsMv+
z8HcsEt>9;bX2c-tKURI-dCs5;jYq<M;_o&3@257=3bo6|PMH0!{apNvU4u;X??B7~
znci+$(z+|hA;XM+lfnHP6X^=CatpM(alxX8#u;s)Tq|Sihh>&z%{9h1z;_MtoO2eX
z@w4fnpcmomL$YVMxAR`(Ke}43FuXAei~jFlxM}$&nldV?HYVSSx$ZTQxa^l!C8sFB
zNQ<_YK0N#^OS(6Eb<niD`)>eCnhwg$Y)Sl>)+9YD6>uNMwb!bs>_f-xQ{J6pF*2%i
zeo-1Y9^ZY8wepKWvc264R5TBGu-7^-ao?1d&w=ojHVXc~%&jMr=YiYr%55+0NMIt5
zTMvi71v)FCFuFW$CENkHep+sXo;`U)3IyBp)O`<waAUA~x%>PYq1AOnScWk)+s!|b
z%YNW-pJ8sX{J9k4teU6^`(Qu`ynXn7?c>4%kQ5!a{N}aovCM%JHpV9mqSnJZ(Kj0z
z{>_dd+9lE|8rmSR-h0y5{rz;a>LA5)T$b4(NXi^N$Yxyp@X-W{4yFKTyN}4DFTdS8
zCab<L*=HGRJvRm80VcQzvJn5en!R?;0`oOEo#sD@cKE6yqpHhxecWhGo7d_{>bR2H
z%H#8<tv0bQNT%(F_uu~cz#DIMN+#2CKDS>+Uw^AJ#6E3wBt!Rao(GB{-w^LP{?XWd
z^$|aD*1%c5DiZFUj=~qj*#h)O)}XfA&Fm;mNlzzXe(KK?Bl$x5U;4t*CZa1LG~(WV
z&jGxHkD}iyWlMje4%~CB@;h8$XvUN1`*;n;9ywT_kA4~S8}p^44KRm(B}L0^p<X3c
zj&}O;K6P*??V2@ZsONA#|M*p4_W=<0;SRxpj}MeL68g~2LQKX?C3`sX#)nI0-5z<b
z)!uU`#jVZ->u;GfkS9V*lug=iu6-6z3!IwYE)M-NidN;~QP!`wF7$5ardjiQ0w~_A
zyq3lGcqP~<!@cbLMb*(a-nxSO-qtScH>O|3bY+&z1YAk1JkpibYax5I3`@2LtcJ@@
zGb%-Cx(59!ugRR-l+j6X@534VeqA_pJdPh6T3{T?0f`?t{YqB?_3ozmL#5l=zeU2-
z^?aaN7ECyeMDSyb)zPZYT0h@BvRf)6?wyaVg)iT)ed5+^;iZ*H<^4%kqWOJzUrdpj
z5n1XWSeK#o`}P{~Pb~$OU+Zf<6^)n!8^nO$UaV^mxHaG<%RqIK|1wybT5$e;n}@`O
z6$jynAVE*d2TTEubhpQ!sYf9QAn~P&_R|Fj-Te9M2<gUzJ49Y(u8JzPYvf*rIU<gX
zkI5B62?)aiE_*->f9EcGW6P37y88Xl_NBWuC)TmuX5HhHI`Kw+^BEUI?V*{$@!P1Y
zT{IT`Ap+mbR$Is-vhkk5Blu>HOki&~$vJXdMc%+<)Y`+KGRoV;ot|TP!THF<k{8y7
zZ^Bcx41?TfpOh$_pK6ag=Qxp;#a9(5YR&n5XahmV?j`o~#=Hax>vci3Sh6zcT|v1?
zo>JLHx=<lalo4}%|H5+8ZeW=!Qxzs~dF#;wXuncY`Q8SuQtiJs6CLkfyQ}p0W7DuJ
zou0;OJ5tU{!FTav5)0z{+7dUY8b+lBSOF4$DR1*<N+7;{;p0SOp!&}+Y*2suw%cP8
zcAp{j++@&Q#u5&#K<BmnIxVj&%0G9N1>bWavAnEsA4E}(1?4575SKWfV*{t-j*IxM
zyl@fk^z0{iHner`q1G{p@-^u7-KZ{6?Nxlu{c}fVd7ZY0ez3}VFhOMhM!A^sVOh+%
zyT=ds4VBkVOjh@I3U%%T76tY%rf((D+Y6pfu_=_HfBfx<1<}7dQz@>}!T>7++t3#+
z;CDdiNwnWIZkMDh<#^e8PIc3|!$P9c_7#thduQF2Qaz32r2XM!FV}Wvubn*j<xvn9
z-dc&MZEn7Wm~LGh_XKH1hPA)bJNeMJxIgXRYG4ChNIGg9iS+7CA9Ko})+(2N@ikqV
z7Tn`gtw)s2DwjDo9aw*R`9M}$hKE#CSu3#rS{H4v$DYilWl6UX7j7VMZrZ0cMhF#X
z&>qt>D<}ZoCi4O?*gvmDZMk-?mFY+wGt_4SpqA6fgR}Kj*E3`(wg7KgKSlK}D?bYy
z|IQL?_?N-^nUgfYamPw)adk`MhP(P~<EV6;l7e00l+s?mt5!I_lM|gdSo#-XK4~@V
zyHZx|iU&8xBF42a3T?8%WnUsaMxR0y+=bLKR{q?b{YW*~X&J0EtdelR-n~SZni#jS
znp|+C5|Ze8{8AbRxAPw}gX$SER}-p7#JufqCub*hI#s!;cMU1^{WO%fMOrZzdW(D?
z9YBS5UsSK`&?I4RR@xpVL}pw6GWEfB)_+t0vG?G<JwlAKyovUWwh|qwXzfIC3irO~
zV}n;_#u6=$<YW9lQdun%*yHXtBT8|*-V&GXKfS-!N>NL$CpakjX1h&8S%PiQ0blh>
zM3rIBNZAMEwOrS9Z;I)$!g|tDW9#Al{H8WjHqA3RK%Ci8Lu&8XL=KZ!<@H|{kDn$F
zGoQuqT2n41xL<G)p!L}_rOtjS;;-(`fM|^>d-nXB|0x1(x%i8!ciWOKscN}LReqm)
z*^70Fq!!d4s{7UTGto60;WDtJ8r09AXMYK~t~p>bLC~qQgBzb$t#cp|?mhHU%)1$l
z9!vd_f~p9w(HS_KI%6miZkM@6OT93wGCq`Z&8-{QUChO-PF0lj+P)zM%?xK$)@WdP
zjSKr4nY631xg(dK<A0z<baTyIKa76Kj5W7N5Qg7;!LHJ?_XDfaz&4;Zz40iq)Yix`
zwLPu+FWHfREAo@Ec-2&nIo*`3h|Twh9;3eCzLleb@#L(9ql`d?AFu+64;-V4n_DgA
z9B*}OjvLj`fVt{+(`mNQfz!mbYI$vzd8+idAB-?~!g;GroU0g4-8udV&3+&>?J>1q
zEKtqXE`?){cwCWO_L-(s`SvsrS5EP@B(ECz%alm-PjBQaAr#FuJ53AjbIh35)MG0P
z1*(^Cj0-@xso)cN=AeYj1AV`e3j$O^mRFb7!H`XG0CmhkzD_bqlVLFB++F*%k7^w)
zv2VP%6CE>JajF5G8RE`2Sc_|%RC=Jy#7WGSthrqmS5)5<W3kVx<wluyS>1@)iL7rF
zw497H@<oKR;QTgTRN^oxPX`L+Q@hBVpe00Kg`fAbrL)tXc+HW-wc<xnsCBTG(}=U+
zob=Z79S{AOqi-1Q6d8X()weq8qrn+Z8&?fWr}KY+oWKMe#tw;|V+QFAWr|kL%f*mO
z+07>Lf9ZYcqfz{B<2h7P(-CRc(KoCX!m-_ZkapcHITL6u1m5dh2@c?8UB5~lR+O=b
zAN@r?`Hp(_5s#08o8TGlt9bt=5l7aKs7S^a8y#Z$_%~F|h%N;3*)1-)7$ocwrh82B
zvW|ve7{}!PCm$b?mDK`ar&eGpv`N4k%2zT#qhG%i=?PT_ZG`qrRW^tk%;$w~MJPoL
zW!-x?^?P*s!ZkA>ARanwHD?r~o%)`B&6x_64K)?^wN5HoR@fDHAM#6Z>TTp=m3Ep7
zRQ5AEor<5M?xy*fvuc*;eN{rjpV_E9?h9gmgu(r%s498@1!MV#{F=OT>lxrUrsw<i
zcQ!9GZF4pE5^h<(a{&hizD`)b0jZ?6c-L(Gz2jyj3nqnu_>rD0e$iongE!F(p-XqV
zvGR1>5;RmfEg>rOooCb&hTgRt*Q}P9+#L+O@%GYs<L|A2o%dsbfBlU70jaFNE+6S}
zZdRfetfF|Cp`B&__{*D3gYwnAJ*AUlA(_d*>u)b*lzc}&8@P34%CP7(roZG_wM%Y6
zj=G=uGR}1z(Q<#r!J9D>|HL!Z-rq;SNyL`y*tn<s=HnIF<j-eWSWwaqTN_^n^hJA#
z6Opd%@-Y#pQfz_Lik#Hn7!G}3#5d4r3-FLICQ<d<)3C1oV8sJmq)G#MsnI%cH2dj0
zYpz7N?Iqqo)t_iMT=3j<e36NOs5kVQ5Q(qzz^*+*+{F5FcgH6$CUMbQzdjP)eklmj
zrH;C7g!u1Y&i;1%09}$JHnukYnZ5sm(49WJ3C~0R!#2mHeGh953D+WI^;0*&Enak%
zlkvRcexziZ2J}WcsFgugpvmvmcv^+B0pGjgiPLn~E4lEI&LGVNg7(TrtMn|*<-CZS
z{Of;8a$e_D@2*WY&1BuGs^G;$eMPZmDrz|YE_K%H@td%G1^Kqo6=s}r_1Q<l;jtb|
zdzmIYn@_jk`>xWn_RGh-My^?LtvY`K`Evq?^@9fl%T5I!XG`4k-iv>+iAsKJ-&lNR
zZGC6$?H4*dmka(Y#Y>hw{rVO|fuu2G+k2OK^0c<Jquxa0%vL|o%L_MVn92&C8?~>p
z;rfl>HuT-%b}khMRt#wH$J9|(hwK`9;mh^Ce$FZhu+mfMUGdBtZ{sBrGV8zjlpSXe
z^{;fAltqE$_zXv77HH~i-Wyr_tpV^@m7Yh8q`eS#F0n1y|MW;~o`J)norI^en(d?_
zVSEumppOsVk%O42-`eZ<)x4iGRr4gKzpDKC>*L-Ba}4sszPy#>a6NL@)a&a;-jT&m
zOJCVvq!?*YGhi$FH}`Gi=w-=brnfqhU(gKj<hv>06HERi28Mv*Ra)w;huuhMz_GZz
zs!xmSiJoVW)L8{jq=(4LS^=+<Vo;<S`*7+^Z;y{46<WCUf~vr-f?i9HEqz*X(m1yT
z`y;hdC|}gcl*G|3Wv7xO9<XpPa1&}eP_k=IzutvnilN$rwbgO{M_c)k+VQ*NT)AIq
z0!A7OHaox1)gdv<sYAWvlJ5N!@8i{!lBREH6^-WlQfqO{$oJZZ!vA?(3D#5uDwhRc
z)yd|&FY`6wGbBA0X(#)7JO9x`tfHx&<X$EuJ_5D<)3!EOYUL+O>{$58XQ>;QOxceR
zXLg|<@0T!Tp$p*P`f%I5#GmZUSvGNGg0OQqeUFoyGsLBLw^`RLgCk+d4AbC!f6U!~
zr&+Y0?&G~VGi`>OJKgUUzYxTcS2ipJ&TY`;ZlO3c7h=Y0Xg%0Hphu5xTrs-}&OXtj
z<q+;|Q@oU)Kf2@%dXgA7y%}9#Uc}XbWEJ1fmw4q#Tr(3w#8Su2{Ke7eB6ITsZmh?m
zyp_lBARyZKK*_c5`R%ENelB};MAP@D-Mict;GV<&usx2R?W@z78S(oy-UiMu`-r07
zRGyZ5QD%U;WGnBIfqElXrk~Ipxt(DL)v~CA)njI%0WHxQVKv)6M%_R9{f(JMXA24r
zlzxPdmRg@?PshAUhN*>Wf->15dV!AnF#~$}XyO<mVa@1gdGYw&f3bo>M2kF`Mtw#)
z3%bUSE-$Oy*!+p-#45%eZ7`=~-AY(`Pb|T<DTuiAL=2|oB>_1H8PDB1a2C@Y_@n4C
z9WCG9v8#mmYj6js!)>A2##v_KDah%``<L|V4BEdMM&G2+gmsD)<rR(nBwFC>C;3&f
zT~vC8-uu_?(f14kg)^Ve;n>{}dMUE@+y7wiJ)@dh*LGo-4VC7SrAUXR^rj$ELqr8c
zI!JFKO=^(P5}E>ni1gl-UP5mYNFtzgh;(T|qy!Q=ga9G%P2790{eEMdzvtI^=MM&h
zggNIk=l$IMy6!>wHHuse#G6yK)5ikXxQu57__t(nySex!+D7nhfd2_A7!ym4Hhu&s
zDs!n!+6twzU{bU}ddE3d!|a(|>~)lQEw;RF!|tly%Cd>HYyT+OgsLSWF!`AWNR!a@
ztVpsndp)|?g0ah2&s8mG%%@j%4hvdKQK<^NC6$A{3$73LSkK9$ZX0LS4#EsJGfIBh
zDac<cxjisvti9j!+k3WT<?l`EskDt5+qv}6K563)q+PD=z<jiGOzwZmFyq7fXpH!d
znQYXE>C^cplf(2}zZ}5DS5$ZH(i{M#H3SaGIkEWG@pV@7@HIPrZ1n#TchEfkVyerG
zbvO&yrsRj5w}Ydy)G>6LF7|yO7Hs8^(v!8FI1#XiY_zjPPO_8(f6Z2<y{dTD!qCL?
z_4pzN;PZugAR1XJ6rE_@P7l7UT$+9QjM_*e-i6R@ksm3>lVo5TN(h_{^8<Es!-T&m
zLt-(u)96rT2+#GblRYCxS9IJ#pU+6+vZ4LJu#e>d%S%kZ-+|I7lbkynEuFXisi3)C
z1$3}pz27~f@>3lbIV6)79<pVbfd}B%T4S~%c+;N+DPkF84*%nM0M&UEj^&obUCiH8
zs$6-WR3kHi@p(IO#XbNb4&BM&!-kNHbH4zM5uWcSC1_)<En2^v{lKQWsb$jRXzA^T
z6ZUe+J2HRQRtA5f6_)<)<HWJ%+g`4C2UH#WO{yI5``3HZjAI^E8jbc_v*cLD%_Mv{
z^R6h2+xF+~+WWo}Ira_N;+Q!gRugsWvGzzE4L!|tfH~qmwF-PJ*}=C~O4c0SzZE+d
zTLgsYn$)qa4Bj-~n;Xideksu)i}<L#1)YWl))3%{X42MQ;0ch!IvBv;E)J+u@Xj7L
zZ;t{7Jt>curD1F%l}r3q42ts&!2Q+M3zF@ZwfMDTz2~s1En%DnvMRwVY!C0X&SuMM
zqq2JwYB(86{05Ys9JB~Bd0!R3zw4BA__+`|#1oZ(8e${odic00Jt_CP%?7}MUl6`K
z4?QDoJlxOKe&MF`KP;!%>F+<=WKC<rDuJw>zTEa{<p4Ooz<2<c`HK-ZPPq1CxY)n`
zcxD%v$^}$5nz8Z{q(?VjKkLnt;<@-A&szNyNciP#13xXMr7N?s!K;o>U%I>rUb)vh
zVu&+u{r5S(`<=er&)@fK@6+Ac19oyb@Yqum0p-eah+_a^)}4dwtHJ+Hk8Kv5!dHJj
ztPPHB>IKq$82<9wpU(e2)8>WW9|k*P0DANT2MO;oLGYf*8`VmOg?AbvZas5F&+Y#G
z)z4@Cx6EMp&!z$%%Kt(;|8JBJ0wk5XX7ID!No7tjG_$`RW;lgRbv=SAYlUFZ_wjMj
z`xjfXc6$d$DK&>Y`*h4xc&3RNxmY08z5$t-P@Tx)N}AiCP(7xlxp!v$llFBDPr6Be
zeAf9qb{4<%3w&#>8)3xd&tD4(i8*t|>T>qC*Ux~V7Sw}x_t;tw25%-sA=y^Jh^}+%
z)lMYQuSRHPX7D9g4WCVE%}al<5b-*z4nQX~|3dM3qn!^@Fsf|a*4Is&O3REJAGvb~
z)3%4~m1U#>gCNPIWHb;Mc@%7<xdajUe@sLmaJ0cOV{QF>jOZ_Kft*{)^cL4G3HN%d
zz$3f0Hn7?0PE0g1GP{9&@qlK=Q2J&jv;9gd^iQR(NsFcZ9y@rTEM#5j2{E75ROX6e
z-*EC;QL9?uX;w<R)`VHJEOrfRxdYbk4N2h&(e4E%HOJ5%ybeI}*(k9zd3fo0hs222
zVXV3~vE;?p@<O_jZ$ZL%G$Ke?YQ$*d&Lmvw8hfNEdXnw<T-wQ;ykkD+tg6Y$8u|lh
zw|Dz{{~x5Q1!Gz3N~W2Ju6f0zfyPm(MOsqVZg90+C1t&LyV-D4(0B8K7<X{&=aD^I
zMSo}NwM1(`HOht3|0Fb7lWy-LR0^ppPWs7N>C5-+{w}7;oZJQYjD=iZb?KV@UbH#j
zSTS01a{2SI@q_#`vbk;7#325Q9w1C+<24Y^a7btxf{nc>6oN|LdQSa8uls{(dx>eH
zI2M3G5?|frs>*BZ<ZO{C-rV~kyB*dr9k;gV-~kdKyR{xxel?n4@-V)^Af~e-FCRw=
zlCt}$OAncC$<SZi<xdM@K9VmK?SvSZ;7k)Hrj$$!d-rpsXqNlTTu`ugKJzp-q%C^!
z!0N2eA5Ry_4lHt^O*$g{m~{BB7Z*!-sTM|RI?eQXfrXOkt~H#%Z%i{6fBV#5O2WUa
zLfgbLSrnIvMmpT(?)ZB1(9Wr>bItQSk26_nH?)NgD~>N8aYuC0YGSUMb0kIO?kzSZ
zSVC?F1%zK*$*TIaz@G-bqzK(wc{0YF)BU6WXwDRq#ciw3U?tl9wH-jC>YhwFF`xWF
ziI@G}$^q6K_Dc&W5L0h2wxN1hXVw>xjvtJX{NK(^NlT5WHe)G^J&e$roGD5bFB|6k
zREwNB^kbDMWf&Ne!Rbwin7wtHppi!hrWIV#C_er<r8p5cx}poaBjY0rugj4f0$8#_
zTk<D8khOpBU>MXBBKuQ(-egOa0Vrlj_qS5j80Mq6XX#KoR-`$5_jdWm^@GKFIBPE+
zG<<>dT0H}|{<x|lkbzFIUD*p396Q?~sR@~~n|lu7d#9G!{tgPFe;AlyxANj5k8EX`
zkTie(^*d?%!Y7)+)TW#k5^F$f)WC56bR)ctq(1vo=6;pGyi(WlAGOSn?5~C*QUoQP
zN>KP8%HAfdD&m^w+&po9k)=NU>f8pzyuD}*QUL#WBxgEy1a41}Bq<i78FAdl)_UH4
zN7u~6pis4tz#;-8S~tK~Eo@ZLA)oQ-txGVq!Hl_wO%Jf?qrl+L#)aU~JXfz{SYF9U
z5a|BeOQXo8bw*WBU2Fs#PD{!<x{rE!KD7=1!2C=K9yd4TTe#xdEW(~jX@m+m{AX!j
zjpAj1wK0b(=wIfXu38{ZUYYG8q?F8t@NgZDMC43F^1(yOJRw!IyU%^^YQ^KS{{m#(
zwm;*roc@)^zgMQtK-xq@VitB2;Hh<P!KGuWYvm3@?t(dE>O#umS&2+hi{L#1Ee!w}
zTH?Mm;S4OM`QG@QufWUG5&K?iFEuFnx6%`Ts+c8+bwD^Mpke8Cc4^*GiZ*2jy9W4E
zLqrlNq2wd|-;Hac=9p!lS4vkyTZ}Im`oCH_j^bXhdwg}9_)P3lNh%NfHe4i~$^F@V
zT$r@DY&*h~-$j?Pmw|b?HJ1B!k1p)`d6M&ntoNWra@b&RO^zDi#D(w}LYT2etn}j+
z(3u^<Oax|m4F`R2Rf0yT%kox$&ZHeGEC6uwEd`jDO={DZS>j%wMF%ohank2F9y$dU
zR?A&kAQ|5y$6;EMXEvKei!{ZN8!czB&)tEs#V>e_x4xxOfSJZAaV7jfjo9ER4NqP-
zqSmkDE3Nvy5A8Y~=9Yt@<AU=;MfVR)nSUU7b#>%9>C9+qj{P5pp&5fqfdbsFyWSWh
za>H@>Rea{kRxrQa>e6c#)>qHnjXUr@3-^57q!vRa!Jw0_ayE&p{^Y0gVG8?Rd|q66
zw6L*w>d$G#34r+nbECYnp<vjCWZ`cHSg-8bm>P<QO2>w2XJrs&y;$VPYiO`nh}b&o
zzg{zKrd~O%SYq>`KpS)93w>;FwSFCMlg%Wo^UMKT9dF0Kkl7d{zgm(KBx{Fx(f-oI
z%8W)>VhdVTj*i<;-5@9^;I{GS?{;2WnEU!-vdOq1@RiGmtaI<m)b+H0ABpxdR(Yl^
zo@s!CF30wcc;(?m#@_ZwWhRBm9i`;V@y6FpiniIeB6i0~>%9&V><eD=x;?P+ofo}a
zz0}T4<ue=XFdjD5ZfO1ysYBn1?N5A7R?GcTsGFbn*TnKSbq5fkF}w0SO_T3XV(6l@
z;j9nkC)=XcQas5oC)(D=k|V|dyHD8#Mzgc4l|7{*k}BkX?H;W6E87nEP%SJN-Fix(
zZ>@L&Ciw$FE>^S^mI;N(=mF*hSUFkJZWn2uk-~~j0FtM*47vfyGmp4gTl41s$e;2U
z?yn~An`Vr>%K1pv%?atQ3-bEuee9*n@0HlkC@*nxM`V87R#CLq^B^FhIEG_D7346u
ztxr@GkO{isK}Z>Z@~5e^LRRw?KBph`SPrdAnyR)=KQ)wn=|7uTU8G*K7QZ1xuO9>L
zomEZPfRULjEO*{}<`;=BvTS)0l~*Tt6u5e<7}1-xg+jHI%3oL|%WTC;4<N?FOv)MW
ztgE*x>@-&Mg?C%j5wueL+DNLc8&tH-`iv8%!ZSf-W~4USshrCho?J#g5EwCu{)Nvf
z1?%;|988+$Gsg=Ezr=`x*|Uv)xs=EuaU8xz#R!>1`t|Zz5sWNqe?9|oA4HrGbFS#l
z9?^ewIB_eRe0<*%?UG|_+0u+YDc?crEy;e&JMPlytu;0}82h&4P1EU+zVq3LF?1zO
z#Om<4WXg{jTHb4ZR8tRTqsv>})@y6t^he$5-3Y<z5ncp<Q!Ic1@l7$#B@P~t_0?r~
z1*WVq%SBsEc|(6?PqNd)Fd1!+ju^ji3}c#>Ohew9XU)q}WeXY~_m+0_+Wzv<Drhz6
zis4h;UP*&6x4@|-k1W9q<_d~s0jT$&sl9-D%$~&X@F`GFpovJZOjBhJy|erhu5EAo
zI&D*KZ9`D{Y!VmUY-LxfLBzKdpg^oT*lvI{R-*WG?|@a`ezsxp|C@~^L)L%r!<FZR
zp{AG_10BNL%iQN?5mycshIF+?qEweQN_h)1ZD&5GVQEv?B$~%z9-Jo!kHgv`-6tiA
z-B>-GlU#H+L6nYSgj<hRmsb1UtopXh&SwKtRDVU+uFuYP<twO)_v{FBs7qORe<6ud
zJ?rAy^VrfvFR~Gw6G8&n5mOS`zT^*UrXkbxxqKsgGQQ=(g9pbnb(~OyVv^HiY8d?W
zK+~tqIs>`zsa_6vGh+FZ9L0J@>!d-5P5Ov>DP>`+3<;Za`@x;qu-j|nwf4Tj@EEQi
zc)HD8t=P1+Y)O{6{=9^?r(Py=^$*8|<=q(397veX%ig(P??$U)(f8c6ye+^!Mn2~U
zfOZRA%j1+1GSV;Bi7J(rGYfb8(B;vg=~zNxA;}XhEZ60CTg+AQP<zN9L`CV^@Yg1t
zZge`Oi5c;oft=%$`9Dx&5l=%}U;1_P?#P^*8B_C~P*5~k`D~<S86G(#2y&gQ0fRHm
zI`6HY*6p^w?P*(4B!{0^&{*<mWe!E*328(tm{*kPVY)2YMcK;{NTh6Zs{mBw&KAXO
zQk72-VpkL;U*JC=wW!&jkImHQvwLFn$b43tCi+=sOiknu(zb+3w`{gw<g!k2=r%gf
zGA67;oSmb6p;uetL~LK9goGS&Wlcgj^cI%|BGybQ_V9T+ztDbi&Oty%4?|UuF@)v7
zG(reXJUga?$`;d_SGN_`r_Oy7H}d9pGJ4WnD7b{?y(DLf;@6+zK{OrfpWtfYLYoIq
zlA#*~!{jmo6(TyGB$j>d1xYm>nf(np(D)T*wcC=_LrN>`Boopf9c-_^;k%y|eDnV4
z{s}$`!rX4hVJLP^!kGJa)>dN15X@(<jeB}oaa0Dae#7MSPCtDsS4H*aJIU-7id06r
zfuW)fynn4lq8%nK6QdKIKX{0_9wk+*DQ{PhkK?_kEi5|EHDS+WuLn)(|B}<cFK)hv
za`H1=J<fYbp~8f@LZ6LV?t9x1%ybLy$`R<9NZA@IE5;-GL+C%BG=GwiPG<>XOr<+?
z)his?yU2-(ogpi!WY#b4ubC$T$yjhC>x};63$A+hq(_=I#!HP_`PY|xD0!=xI#!E|
z4z}O|5nsyWc;aan$I;dD?|~30d)sxJXDdq=8(v7OQY93suN8z30g(E3X>r?C<(A-e
zm|S+r_EmP9c|6O6)_OvqJl+wXVfa?s%~7$e?n;AHrrw_EN#M0-SA+rz4J<<|5wOmt
zzd+`t?+eV^!?4pD#ZbGw*~9Vz`K9u_H&ZN-ZsuY=?|9hMk*i>KqnV_H%l7^GDA8sm
z_ss|qOnuJ@zW7M`u+0VwhSa781e6Zcb9o-G=96tN5n|pnC)7A?L@F{f=hFPLAEU3(
zr(VPSvASt3`jp9nzeP5Fc=KWLkO2lJ32{5rW94nOCTP~nfd0ytja!@|P^vR#==>xH
z>D7LHLXzAxN5fk)KvD7%TMlXS`i{hcnJV~IIz__Of|IX1YE3vFReJ-}lvaSiGwC05
z^tL*R?WD;fpcyBCUnV-o)``F1!F7w6y7yB8W(_2CpQ6z?XMH(Bn!aq*ZZE`slUn6A
zxJagI=>$?nVH`(VDU57`{T8@bL4rM0ZZhXn&<ect`XU(!bS>&CYhh7lvO*&q6lm))
zZ|{p{-XhgQO_W0oP8o$yR&+vvbE#~f<@woExyl2+kMGICt}1@qEvd=$a2xdk>rcui
z65X{55xBF1pfbxLze}lJ@fh7t(}}WsJIFCUWzz5mpOcZ#CFDUNE5gxKb`<ADHkMl&
z&q@+r*zM#TTJlo*r;OXXe!e?VaY@y(q}KD75s+~TTx5KE_$r$O`ZIs0HOH?k@F40t
zDWW*VsB3i=Jc3G`1P6#a4=ADfeS{6&|1PlZv$^oagE;&g<P}K3<#o=r!ltS@RcP?G
z;JD~?j+>hs2f4_)@*(>JD>)7X60{bWOYo|HkT@$ibsyhXrJ-3&^2~Bxd_~!PU>Fq1
z`ZUsS(@qBK(63!-r%;?@W2d^c{gquCK9ePmv|`QCA*&->opD05hljGvNf9dBvQ`Bl
zCyht2HcgI<ONL|K7A5g2c7%y|LMaz&w&YM`d>Efs%pOIR5Q+*bHEZ8dH0dnCe<NXz
zzL+B_v=ZwP_$nc<YA6V9NnZQpp8c(*BnoLw;$%Cit<}h&+^DrEBiov0SXyd#XOHQ>
zcgwp=e=L0jZGAej$B!wRUoG=BQ{y0Gyv&V|Forve0`-WoT@gdU*{2E+`8(qw9~{!N
zab=C8)v0o&f=o!ayGhSuZ{Vmg7%lw*JDgya=+oL=Z<Dg<9zVfPKv}~Q%@>knn5}^=
zh8u#J%i#CDXD;=IolrZ2@>f-p<aAO>KK8-9{C_FT1Qo3NEg@er7nf1{rZJ0Pd3?>}
zhxWrStC4a=F^$1+Wf1hO9bdU-RZl(Yv&aP@i>aLzW{abZGU8Tm$Z{h}Zu!BO!$wiG
ziaub;CnaC&t#E$oC4O+WT-fjft(hz#bpcyWyZrYhN2_~!47dFsR6`I5X#+WUZR+Jp
z$*|2@&Kr7MM?H!+OO=`1>ZDC~rvQ+#;bFg7oPfm$upJ#SyeWq6<P&28`$hizq23@H
zK4&ft?JLj(!*$X6?4rEO_A<NZnuXcV1rcsbdb2MinMX1sCs)!<ik8<3v<5e?9$zz1
z6ov8#amALNs*Mcq$7@+FEu*N9?QE&5+m*A2+gL}`LQbPeAG$O$K+?tv#AVsLoq^ac
z=RyZl`{7qTORaLD6+2<K;`q{vLP+j|nwyH<BN<NJpI<Ux!3{uV!`BYDH93MxA#j?O
z@U3}9znZHT|L*Mza*DpS(Q)_biAqi>)`L~H|B<!U!E_G&n3~@(W*MIUY$nWaf`P0I
zb}yxNmkO1R+0j5-NkVuEP<hP06qYy`2?enm?LHt>lO1G>0_%eB%}l@avGOV2zUkt#
zYOvL$AEtl!gE0gP4SQe1SuDqzo`hU?cq$*`;kqtAJTPCS>t@^$v_q_~o4;c9H^KK&
zbB|Of=?8YpG{bh5e}N<t8EgjjLc>g|wIuBH!K-CNvoXaVQ5sE<R=zYjK79G0&zxMA
z($^nN3H3o^?nf(-gJv2Y=^N6L;f6!>?$(N)3O3&e?-3Up8raeAxU{zebGN<fU#ns3
z{8*p^%e?cMvhoo_5=6_&cf+XX_yErXhy3N~fpzO0-LZBr#3!wp$7FR&n_q^LQcq`N
z!p5OqV5drt8z`Af(1!W0e?rT}F{njJs>%S#1y(;2^Hdb;0SZv8>5VHrH+*(zERetp
z0$pj)uNLzzSH-TOLALOlIBmL7Q-i<<x7N_S1kl7pp+okX+;-tODRES$Y+$gDGg3`z
z=EM+FQ9R4Im)8e<EvV{fSD)-qU3zdMi>!>D63ZpNk=qEyzJU8ya$blSg%qJLEkzOt
zY!+HlOa&%mlJQ_dq9y+;RQ*>=?stnB7R>l0`m*s8Mj;fFiBSQ=h-=RZ#ZdW|EdS0u
zkC$LcGK6KZPYKatJ0n^548=LkwZbcU3C0N+^eIKj669PqhDxmE)e1vqmkee9Aw=~~
z8E1(xv;4f~SHFi?fT~XFE;P;tu~=N-$%}j#)Ooo9jJG;st){ga@soi4c_%GFDm14F
zu<K_83%KZZg3W-sIo@7M-Se*Km8T_FxY>vJ%5|3B?^WN;YKURre@nV*((tMQzw&Zh
zlJYQ<kK7^p7MAeHK+X}Q*EeN!;lKtinarWL=&8x+*IAM>@SqANP)v$?dsH)@+^9RT
zvQICW|M>7xegB6yR_-|aS5xCJ2F*NK7=yJ)>Yf8$)9#5+m`>{^t;&4lrM9=!GX}`$
ztpl$n^@B)YJ`s#aU*iABV)-c2$=5;v1)fvC+%#M5_GJg|n%;PccDpU#qQ@*rc|4D`
zVa~jvTw$P@1!0(y^x4n^^4tetIKRU6MBP$8hB?@#((B!GEAN><lRA-*-S@+tJsXMP
zefa_p22HZD_&~mIO^L*1xv@CaxlP-~k&xWd6V&d?%ls=CE-t6<%Ii{{TR*TnrWt>Y
zYCF=S8$R-*Or?(%bg0KdLKz1u5LWM}s2@y8c%l=gR$tm_RWth6uZAGxn{Ow9hyDu{
zN?m8ibXK|>XS0Y`h-sg0VF{lrZroW(GqhyKviPO8Rj^XAM3dpnsS9{#V`<r$yb1P=
zls;%6KEU;vn*?iBUyDJOqfW0^ffH-fSgl2IULY_kv(~)&=+wN}(WD~eMEF59jru4t
z?^0Fl4thog==Is3LO>Y;`p+}Sulu#b6$ZT>)@KqZ*9t6DS=PKjc6%k5kpNE%GG??=
zcm_n1>M^2gs6Oq02SpkTNOuy$^OoVUB#q%bKQHKjd|hK^$zcN#X&N~s0yP^u&QD=3
z<3Qypn8NgS<5Rj%9&_kaGu%5cmMczbUv=JsYFSg#gt&0WUTmQvx|X@GJ>XX*td8qv
zN<?J4Fb(*Kipvw?pLj?2HAZ#G21jdRm6D2$E$$QKHaUxKXRJr_$veW(ZjIM5lJ=n~
z%{B4RL3}r73?I{rnYr@tO$2vz$)<!@YT=7+%*S-Po|}iaZI_Bxe?7E`9{kiQS$gLE
zc$6^mledf@LmSQ4;EdS6K-#LrrILfGbarNQpNA9DZi0Pm7pN>deQz0Qz9qGE?*t(d
z;mz_rC%d1X=XqpF^Z1w?*ma6R3<_8nX5#mweq`vhI;62<r$p?n)C>K>bE)o=*yT*(
zON0sM9%jSS#=gLHG9%5-m{CWJxsgFBFK>9NY|lTo(XwkDmq|ySiJiz=zSI}%8)a?c
zi}%%Ywm!rch8y4)QEGKeM~2o^cB*A(laVK=)S7L;ZmMT~RW8@bqHs_0^OhJ|Nf>$m
zx+Eo*^znJW_Pl5NOIAH1WVyS74Z%aC$>}=Ojo7yQcqagH<PCpNH85if&&E<U-a`e)
z4c*0H8ztKY()PY94nzEdSDBK9!|PUJeKcRmd3$N#eakf^ND0G_^uOO^Ri_YO%<_y9
z!(cdCVAS1~kUEeelAvL#@l{e-&S%??nz5^<*-IdmV3CR)LsRg)iBA<>ePp?cvI`pU
z!h!@@hi8F{ya`P=JYk$N*{Xb5NPiUx@&{#nitKvJC&KXKgIKxSW?jfJ?@Su+?|<`l
zagLQcl4Ay&2B}^JfSntwohh~iBi;Rzg{jS#o9q;YRAJ;7_Er*M`n$GS8B!zPv`EV^
zTqE0HU=pNQ?D?ecH8N^x$M(74WQy@Q*8u^0R7FuPb9E#!;z_{&Po@nf%H~WEaI`Z1
z!TL@^-Vc=driq^R`e-{8SKa=|JLWmFVfy0qPWrqC?RvVi+hi}h8)hRi7}#88rZnly
zF^6`(ZsN@}9+&GIa?+elyk&dsR92t@;$u|h10-r0ZopRweda56Z?SG8x)0+e10(ns
zJ{>c=9@N}S)aViwe@{li{gP!H#HKbdiDDtKhCFo;5zVs&nfZ>)PR<Wk`{?fQLpvh$
z_FAjB=Pbt66DS`u7pi90#v)cBYh8!GfX*>6(0mLOWEJ0~JFMGmsafyUMn<l{@@-tU
z57B?Xj{KDyVz#w)@M|XRf7&vFuNQox=&iTiAknD@2`j*xE1GFFrR|EGaOqf!voEiU
zy-uGn>^$cXR@;#I-^Rict8lyJF?vaH_4-3zA@T0Wj}ur#ipaI6qdYJ~#>a}l-sVpt
z=wiRs18%l)+mE}k7bi5}3(4IAR4Che`>as4XQtnU@jL+dYTh8pb<OoUQKFVUg**F^
zjrJP>ssus57F>aHtiftJ%#3jC`3Ddb4!3am2saLcz&pYkTtwett17kk+ae|m{S8jY
zoLAnQi>CUN)lsvEaK;ndeUS}oZedI2Fr_ckDff_q(psH?=<&(hwXWe%1yqY6R(_|s
zi5@-ZoyocRkdN`HNCfmNX}a60(m!DG*!_018GvA^T&M7<EWSZ*IEO*4N_YdKQ`ET<
zHcUEAUkLhXXJlycVXpK51C>qKvA4!|ZaIT?U=an=^`brGqm9M#lwsj^f^OW6kDefD
z1r|q3*(i46#4lv&j@)YciK*wh=do7tP_4OkPRw*q+(b&t(SV$^b8(~vVNi}xHRR|f
zbtzdsh{G5(zV<Qqa(@1_bUykj6Yf?)-HL%C4Q=YIPgCdJTnQ7>abq&^DKqwurOyd=
zqb^00wZ<g>$f3{C%rfSLFP76~8EZ=@c5BJ;+)QfUUj-)XDD8nuXA8q2P8<4-$R8<|
z_e;m$?D-csIZnrI9Y{@pSJ0weiKB9pSfRxn(v;ZR(wLX9I2b<K!8PnU<-qCZrcqBf
z&6)Jph-L!HJ?_qd?VEa!nVIq8TS#X9MzKVJgm;i}Kan`tirgiZnasx*WE#&tA?j@m
z-NM@Hb;to47#ufnD`Yzc%;SJD^HZ0w=Tqx9K_WgrU2|xIUEdvReXlXiz|Xdq(xGWE
zxR%eAK-|l{iv>~wJEb>H<#eg-2C6*8OKB=6dK;s0&NQh;UkakW*fbBz_y#jOZH@8&
zf&6JNYCuRKFo(s;N;kLYO03dCjqpuEkB}~M=QaD<%S=Yv$NK+FGoJ3wEb2*8i<#c-
z)e168Y*<%U<uG*zmD&JGwQzZIxx_d{*7epi^q@jX)lteMSNT3XwZv&=P%d_~$miEA
znJ*8UR#f=l74;`DE~{pz&!y}aRl^alTOrk4J$;XGo?K!$!(8}i*FIpxH+Ay6LBX_4
z^2Scf7+=zaHyl=;?c-)wyV`X14ov5T+G;qlDFKu17XUx9*W)y9?LhEa(j@;Qa%Rl<
z!UUq!r``Hnty>BnAaKnXK#64hGzo(D$68gQuG872zPx!q$AOL(4|nsG7-|8OWXfBR
z$#?pYwBE>Qy!80dcrLVP2Oh|Y^%z}$^&h$GiD4zYNP*Wy%ONz-6!^{1XMH+Le4t#<
zC*?t!e&&C<-xF6aTCiAzRy{b_9qF;$f%d3aj3>oo|48dxGk0*D=rjP$_7uKCx5#~8
zZbiXMh_Qw%igHhDkNk38zFXgi&*4oTrKWTz?AGY2aujE|^#d4(*uvz~ED7#g<cb~f
z^>mHxws=wIqJV)Vf@o+OxQmRM)vu@Kb^%58Y?L@SJw*h7Z*eME5i|WTVMJg1L!+a$
z^$JOz#ufpb&$vPtE}Qu3nLO^$iz{^XW(POxVKG!A(8?n)Wkk9%{=LLXCMTUmVrhO~
zU}rc(761G=gjqwhqI+-b4SSH2TCaR{q+4;M_LbTk3%XGc+yzuX4SyMS?Y)){3B2qp
z++pXZJBbtYC4TI?fTA6{%D>VcYO?c921QybGxaWADnFQB)YR$RxgR+U#CBC_p=RXb
zfWwD|v#VX}gQ|`t_Z*3A<G&q%)$JF1I})T&I=B6qrJibh2vBY=;#%D7Hy`1iyctKD
zmhPtweH6M-m@{=?sy=zcJcjTij0juyJ_p1fl2ZbNyC1JxWV}z??S<>yp$yMw$9j!$
zeA<gxMQ45hrAh@FG7hjXo!|ve)gwkjyKCW&m4!@<$65rdKp)8Y&I{F5B>RB+HS}Fq
zmCYwCTX{}RujKZ2P-A?u<J#{9A5}N@m*KO_CO$Oi6QTk-6}sl(>kE~5!yJ`c;mg@K
zu(orHg)SISyneteJ}XjJ?kjf7W5s40R9#s>h!9(JYnQ^5E|ty07T|%l<wi?KglOV{
zQC0Ses~uS(f-o}oZI>-fZ;^K%sV5udB-2@&I3b;IBLXLb-xe`C;Os1CPYdYo-9Q`_
z^X9xKp6bDEJBi6kx5^M$?xznSl)r}8V_$8!L;)j_<W1Y}7uNCqTNSRGlq%b8Kimzi
zlU1a#Ei9YRs&}X+94x`9?luFc>7P1OGQq?P!l%#d&GhyCkVu)uecL6bI%2N-?N&4b
zAOeS7OQLTsM@P$jACsF*d~3D#lS0pOQfqD#D_F){!k@3++$3uxzca5ex$o?X5%3s`
zKTSJMuiv}3|M>nYqGC=iJFW5G_Is^G4Ce!X=*@FCC-3r8dm7C?%{cOsEnNA>r{dSL
z*-!9sN`)U~%||rO+;4CjG4u3oko>+>W~rnsU9`wXhcQ+%N+KLWtc^|v?hev6l3}Wa
z!8A*q;*nyZEH;JX-&oNSB$Ww`4#bU4+)J;-tWuaWF5Oe$WY2azNsL{V!Wu;arGvIS
zbg6UCFt%1Vhp;;?_SxK>85J5pMPMT1DJIapMZi=%?`uPzTY9_1>b*+a<U3So439k3
z)ntb@N$+3k&#P#--{>1s#6c@nOPAx>=M9!Vse44FRw0O16dk`EWZS=fghMde3Urr|
z+a#(EcI7lanx>S-O7Sa6fXZgEo+}DHUSAfVI+*>=IQB5PX~Qb8b<CSY>o}nlwjvww
ztg~x;NzRo@UEOwF{}}=-<v@p7f7|?$iI+xWdsNe!)gWA%eE7a3hhleRF!soOO!Cd4
z!b?xF0U9Dqyt=a9+=zCeFkxPHzQyuJ=OLO9P*<W}Hh@p*UJLBDLsfA~kbUiOr2}&d
z4}ol#P?7FvO>7>V>Z=IozKwtDKI9!?u+wjw3q?VswSxbQ8=-~~je4F;H)q)N-3;g!
zi?Mv(xuAB4f>shs5Kyoc;3%S6Di%OAw#j+v+22a>8AK9rvz)ey!}|oZQOp3W(OX9n
zY;A;<(M`V&gwy`fld|nA*cvOkm_SUGi><;V@~UD*J^v0k{Euo6*MU>zw<ie|?G<-0
z$;ERNDtte4W?D$?vC=bvaGN`*o&<c?LEV-SropGEUYftXRZ<x-Q<;K&_K&c6Pmq5B
zA&?{119c)0lrv;@Jvg@X%Ug)cZhu#W(09j2wO><vNI;tj8(*N3c4aQ@ku2yY@|{?r
zBNL6;E?a)9Z%=jP(^k!#PF5mj>WUCt=SZu=l2^R{0R2*_(d{PJ$)DhcqqP;B>9KNE
zoV=s+*WJnfn|j69^zDadUrO3w$_h5ac%MC*C`oS|)YQ<`)^~2(6BB<eMsXVJcBH%G
zWn6H`D~FeJ>*>FFQkt!v#>Kx>Tgntd>sum=7)ss#nA{S6!k}c6=iJnImar9$7SPOH
zqevS?tsHG%XR`DhmE!YbKhE531O}tA4t-S`Bo&9be$EbmOSx8Tn4-)`ipo`y(jVj4
z13lV0S7P9kKM?h&Ps`Pjf;4*Z6{iw;iBuZiPUpD(#Wyp3<7=B|BP(jDJJNW~Ikyc5
zq?g#D-1Buc(pLB*PA^5<u&6bpB?Fh~P(^=IZ`6Es4!-_=H!x-_qz&NcZY7Dpa!jN)
zB)ACIdNOI_ivRpL#r)EoEFOgV)v>dDe4Ae!L81Dl)dzFSacvKE9!5n6l+UFb-ImS*
zLUuf<CUd;Da5CD8pLwL&0@`BbVIuBtEm2I{h7{dsDOp77Nkx?=zhzG%s@6hK0m+g?
z&w!YD!{|QmZ9-e$QL!bn{Aiv%1@AuV7P?{-#ImbA9!TxGZV4S$wc1YWaP~q?({J3Z
zS<mvb&`yGRH;j0l5Y1IRO;T(DVNE#|R-xa*s_6+5xWq8(Uujv&v8z!Ka7V-G-iR{`
zAgzrnC6t<3Ds+@qCj+S{j>(-BFXbo6uH%D#DSu)me9_?Z)!Tj>DF+P8q?-9CJsD;*
z;CQ6_rE5gVDI)6@UzFcehZS`a2Vz3R)K8GM(@K9qd|+8yO<Cg;6iX|=WFl4dfRczS
z{VTLi96gUY&u_8*?XHkW3|G|tb!*G-dcw#r87iBZ38u0!bnnFq3AUdb;}3L#T2i;J
z+FP8j5nCVpu5Lmp)_VW(?}0!j5oDLeivEh}FO`MJ8@ri}W7`c5LmJ?}kPL4egbD*;
zndIbs6ZM&0ktC;|(PH+3c9ZBcC0B$St-A~XS<P&;pFR%WhHR*&Fh58MkRsZGjym{^
zD)caqnR5r~8aZFn7+9phlB)Sg16h?Lae|Ca-G2MpSKRt6{2bF7-^EL(0DVd_(}Ag~
zI7muMs}%l-x+;zI7@oDa*o5vpgkkt-Pqdm!1dyDgRrKRKRlZhduM=yQ3*NB_1xRj-
z-PMK7p&ewXw_5JoOaEM%)5wfg1??;B5pLC6{whfN8_=yRrMuqVYFw0Qh$1#EXh<$>
zY1gt7{S2M6GA5hkZ1oJREh2I-rP-TOHD~`q*$_2XG+apoh%K=?XZ}=d^GwqT3fiJ!
zLj8^W!4AQ79sBQLDO8=5=TSm&eqy&DGcK17Om&xhC0*j!sg&Pq^*Y%euMLyKmI`aM
zKQ+XelCL|dDGFp9P&^fFqprSST+2^66=<X>w{(^{mSy~30!^Ze^>K4P?B7`&N--U$
z@TQP*TXOb^+%Jp!A$|WRskR-}eM^>cTP${$H`NM!Rvwu}7~E-rFv_Byn`l0RMECf*
zqZjUfG&{JnSIqr#Vh1ji&Q*Cw+hyry>CB1T_@5O-UO=K3lJZP{OY{_Mqh}?pLw^H>
z41)YM)+)5&k!_KC|0|^iH%fX~U}D{j;B5h;3)TbL{pzN<l@s}!?D7h-+1gq*F8$cu
zR1_<sF+#IWVS(vC2_#(=vl`HJspSj&yAomezr-1Z;A}_v7nbf#Sy(zKxZ%BU=<(6)
zMfi#Cp<+R2=>?wt3|Z>XC5hZ(8tor1XO?L>og^Vw&M7ndI=>4hY8%WAfeYO>8BZo7
z?e<`{v4Srf*L^MG*}&MYjzQI+BfA3X+yldt(R<f5YUiWUZ<$;Y4r}GFIUG1#UwMPj
zp4MF{43zf;godl!Iw8;(=uQkIG%(z^o7KA7e2w7T^YvFYJB6vsce0#mB8c>`!}4q6
zAkkBSYV(W}Y!&AU@~$_yb6O5O_H&O@yUb_rhNYuTgIa6ZN#k~VzhjbZ$pLqg8}>}F
z0YyWd8Rnl?l|@@FYbYY7nDx9qbWaKyk!ti+aRj!WDu18EprmlNT6pe4V4CmY*ORE<
zT7L47obcQVNO{0jquV7;$sa_Me1uS8kZ@@d5P~gVyhG|}GglZHx|w_+t?J>l7Rf0e
zbnGUR4WIBxZ-rm&4cOP-Bv!3{s{4gN@*SM9E-?GCxdM8SM2O2(roS}7h`q1^5$=g3
z!x>D2BsNB*Gt4<<gx%Vk);$CF%Z*d&*Vx9`Bq}Pe?=|gutMTpnK)?x?A-;n3f35l5
zaK4#9BhGMiF%;;0s<xZ`d7T3ru_~q*_lcYFJ!jqUvr>6QME4Jrnu!6$pSAg2iDw{5
z#P@+<V1`7J{^3_H+l%W1-w4(a@G8tqlC!w<)TwFd@#0n(0I=jJM*_%2#^t?fr{5a^
zo8{!5adx~781vh!8F{trSGEL^n+d5|9fin`kr3edw2!8^MbHTEVm7T)kampsvj7Da
zi23^qxQgzOiPI`~R^{QbF?M4i>yaSkCo)Kxm}c!`fD>(@9olLybU9f&sr1RBC(0UJ
zY-sn_nH1u>!C88bVOy<o7>0^axxCejkki`WQyqk{;4+H+)+1(*W}nQAmyVB~ZCY7?
z4<yPo>SfK>7S6<uem;1dJe&O(`^YkF25@$dkd6M2s)*G1DY*4UNe7A&cxX?`ypr02
z@U<vOR59p%7)>C}A1L%11Sxt5bJ<$~2C3z!*qGV*EveWWs6+lSOx#QvkNEk}Y|m<c
zeEm}pN-SW5)n+N4$ha+6&}pUNkYR}naDos8dBwF8_rh;Y7{b6AH{Z8JQ)YXBHiw*Z
z9=*Dn*cYSQ{iMUWuB|i`+DZ<u=`z3flS-qq7?%oVS#hW99xf?TZTH(K+cL$J?WwUm
zg)QRh<B6Az&$8T-4ipU-^Xs>s6`%8Q*y!3nN?)BKZ@t%Qe_M4d#`@}(=AhzRa`E8?
zY|Wvxk3zHs)8260=uh2IvC1I9Y@$b<d~OCFiWD=&<DZb6o7GK93)Un7!jIY0GzXkh
zUAP=|=rFJ_>MkKL-s@0WO1=cB0&^_!2;TyqtTo>N_uz_lO}}yH2SAHba@p9wA%8CO
zVqO?rft(+qpYlw7GZx>qy3@to{E-8lb@y!{RQH0W0i7e{<%Bd1M<nve02=fjN7wzM
z(^%OtIdNBhCQ-05j=!m*{&oYfm+v{I))!xaA7>eCUC?V}ZnlOR$6pUE{c*ofA#p-|
zr*WY1{`*va&S%$CO5JzA1<3wLmNtP6!V+_xCQA4rqX#lYV)=94@;AJ06^UuVt*AjM
zs56I|ds5!k1oMlW<=tCaue5`I9Tw~twou9OAKps8Hi<;;i>`XbZE=!*RyE$5GCQtO
zbX|NS=TK-_O0)hGLTP<lqra&V+g^2sv&_K*A5w>wp*YU^mU8CLik3wB8-{qItF*c}
zRf?w{=~682xdklc4!O*6^JcMgX_UE~Zry+<V?ZI2AE=pEijgeSb?|1b^(Cn@0&b$+
zv-*rBQ%rO1)?>Ml<q_Fzk1Ww=1?zVX;_HiPLYfc<-0x)yV-88}-~QaNuC}poX8BC{
zd&gA2ld54fP;2gNZ+jMHJJUWXZoWX*{fv~Zzw`sU^1BK{nA?kKk?_{+*k*MVTCg~N
zWD(!n6dQj(hcUOwMS|H&4C|BKTR}1Uj0>+{p8H|CN;F^XUs)Ie;_?IzXn+R&nHwAq
zvjHZ@W?%uDlYt-OPyg9la*aco8?{SYIfE_MX}v$0CIkXRD#+^5L=#jEN$=i!m8|ik
zis51plP*J5s;?N<!=Tb2dNHpUxB1ti>RaTGd7T(1yvOW%L|EH8`xS|&<Ag;?I{WkL
zolw)F8do@~tjJ#sJCHSAPatEGDutJJ5WG~`k5%lWyCP!&=p;X!*wUd!-UzIc)5DQw
z6d?1PxyT(buvYVjpK0O9;Nuqvg$da>lB%_in4K;;6oyi4E<dN))Nkr(LXf!Z><E^`
zD$UUF)aULL9#9JFv~r=Cc~<?c$%^aA1wG??3VS5MfI+NF$(q=yCG?H?KI7HFBc9k6
zQ27beOFGV*4tTm@!0@mm_&rQI?USUzYt0LIoqY?p^SiN{k$DnF0Xk>cq-|Z?#Jy|1
zI9Wz?zbmeGwN?5obZ+{N27>lUZ(Z20PmjW8*^=#jAD*9`+KZkgWCPu*<tSd36BxI^
z$awyNyBI2?`f2mLw7SGr70X(Y<;mD@l#ul=*EFC$AIN6AmG)r<X+85mg#yit^K_hw
z41>9wOY}g9GwFR(2cMGPv=)8<SqKgEvX^HuO9aZdnCNu+0z5Yj2x(`e?x)p}w8rv~
z$LYMi$3RcN6t(cMOpT?PC7YJIJrAS`1FpBc!q;%UB}L9_HX*+Q8JD=N<_M_fBIO7G
zdzHs;c;jw}l+eeS-to)=I@(eEJBLd%&99y|A<8n&&M>vh$B;CSdh<pigFbz2d6b3p
z6>!SwYN&9~X!?!;b3ams%^f~3^=hKW*I3wFolPQ01myiCpZ(KZqVt!xt5ALj6IQiP
z5kMB+nrC5reJI+sSaeI`-c69F6uQz<+7DSS|4fbpy<J)2WhJS=JEW?(+D+IkR;cRz
zXvB=|B20aNbH@gP6ZDJ57h8JkVH7J-S@Gwo*|eqX3*$T2D`aQb{bDNX)w&PWAVFiu
zpxprsUbtqey+zo+Se+@GHUilOW|mlQ_N;tezCf)WzyhGJ0Y#N@D%5m?TKIH!Gtk(T
zg4Z?CzV~+lzBQd4-py>mC0R+eOdBQh4F-_Ee3a`rSk9hW@WY|K{VYUdGqfu38q9dZ
z@HRgsGO=+Oo-?weliNpdWdQl}^+KYN{#)&&z-}s9DE(ZC?_W`qK*=#*m&1<dkt=Cl
z7O+J~L)m1;B6vG57>S6a4~LHo6Di*P=hue^QoI8|JlExmQ%z2$o@Hmu5A;f!X%hq^
zfuLye?W#gIyb9rdVs?NWT7<sVmn|ilAj<rSzhKSpb|f|X$pjt)#;~=Ix|Roy>Er9h
z$<StmeZPr`!g%1{+BF2Pala}t!UNE@N@tT{J1WPTaHjcKd@#_y!frG;meb31Z<3>p
z@*p6Skv~uF+sI`W{FArNe#KQH2FM2dAWG1v#o<2m$SESb;0j^XQc>RkrMkQ=r=Q4N
zfekVEZA+fc6lI1G0G!^J0z7?^#eT<3%GG`Y9ho-@uT4MWWag(l20352bLM8tFUZHY
zaoge*^s~@=#4Ds*??a)+O_}0n&`NxlRwTv)8PB$#`ECQC#(rkt8K<DuT<;l|j(8X^
z_6`VM{>XNoCoW875AxmQ9%43R(WyOtPIw2V&)3p805lPn?XSmi-$pwYiUGa2S;M$L
zUbBmzZ?7i6{tCg?09YaS1Arg!HDpE}glL%m&TAFA9x=yA{dq6{>g%UMQa5tY$q_*L
zNvK-nyS<;=9)5U-8+u@LPyANY>TEW^f)xBIEHM5cH?DP4^Dfkg`g1QB04DnPMyF7>
zg93srW6$Ur?}#)<9Y4N$?aSgRLlF8E&fPRR>WP)cen$YHg2ko6K(A|nE15xGd=^ts
z$l)5+Sw|Y%y#`j)*X_*0wJJjM+8-BJ@-ZAi)BzY4w)*8v;yMpm?A`Ue!8AF4yyo@S
zZAEE@ao>L38xG*1{d)&S=n?QcXhol`W|_KKEqHs)eSnDUZhOsH&`!PdjXe2#H9?wt
zw#R$LeHehg4FHzErJp|qmdnJiedCw-t0k_|>e$5Yw&D574IkS*hak@@kS>`{%H81D
zeZH)$E&|wV;xkVzWaZpt?oU+RJuAhnqU@$gai*ZAF(%+8nJb>YDYh}((4+DujX;K<
zW&l&9<4fC<^v6*~P3ks}lTQlg*ttR;ysxKIjElO~>UfI%Jde}rs4Ov}*!ix1{OaES
z&v`ONlh<Y`9Q<IK&2sKb5%++bw=h+euaP^^=m4MtomJy`rl&L$%X5+@fzaEcDjcuz
zc?j?oIQYI>*N+`3Sx>gNquc->^ucq8*2NY|Mc}~}u)TjA-aKkOC3#F?1b*y?)iZe8
zl0hh766e1T(*KjgC<&Rtv97W+CUFN=;RUkI%>H}4^$VM$@fr#)JwLd(La_gNJOFNU
z4)`vY9VzEuAj3d_u+vYXjI1omXnE~_Fm4haUyc6{N+0;=Qvy!l)&A!+uT8x;LKOno
z`nG;t3XU|QEOutJODJLli`JeOG$ypp)}%}{wjw9Pufu)@1X>w_i-U3k64xgFtc`lJ
z;2+P$U>oq;RPvmgg8*+k#LB^z`tdsAFWCUttc!HO<%SH;vE*UVAck%(O0XAdttB=p
zZ(z;yd*>(e&eH39!LIalgu2PNVM&rtg4WpHY=5`cMr~l)<Q1lS?GtbMPJW)OZ~ju-
z<oVC-p4qfZWPF&N)peNXtn^*Pwfd~1L2xWTz#di!Mx#d~w>OoC&+iWX?ypzi9yhqU
z4UE&6?w7{_EC=qlO|091j(7?Cv-Mf(^GhFA(Q20;_C$!xGK?Jr@b8d8LRr*0z(dn+
zb@=CcUzqjeE;WID+8I+xGjS8;Ope5Cf78=$Rr#8u-86*HE}tiR^yrm`1Bw`Nql-CR
zgy!Asux2bt@t&PapgsfmF%NQ&qM61@*DwmOj+H3McFq}N8Eoh5uU;~s)EwQH!|koa
zxoFH@4h{l$k7zwj-LzHa3<vRtY&5owD&};p|6*d%3xe#vZ5{KydhbfQV!~RVfGvP9
zBx9^@BK}LPI-u0ey^!5^QK`P#h@m}})h;BKI<!2D;t9d(ubx!j4mW9buLJtJH~jhM
zg9^(RZiL?ZW4u*ug39inS_Ni9K>Yv!Ce~kcVxO1pxqT3ZN{;-QN{U~OIB?YL2*wm}
zg%9BpFGI3&Y>zEAe-qjNeny)Yk{BP}xbRIc2>`qCk2^31Gd91v-P?FH>!?xWisiX9
zfr}Fq|9P0uhwfjodb!sm*sI#=FiE*{)pRq!E@+P*VkH+I5(SsF1h94a1-<|^$qAWj
z$rikt^qYnD_eK1bctc`EO@i`6?3@6Y1W~mjBq*8jq1M{@9It|xehf-oS<kaO&}tL?
z^y|tPyN#U{u2!8=IhmXO$=!WfX><>THkKBU<i@%K&b*_WmVGPt|9kGtydmhrjwRWa
zNpUbPBXP9+mgC|@0x)_y92)+jz8;_u!PMU(MoLP6si*4BZkgg9{SC2p9l&87+?-*i
zG`7eJlkhB{n>O3Qu>e2WNyv#A$D^O=xMasap8g|7KYInZLPtz#Y7Td!>3sjFdcNY^
z^W#WM<xY*uRSDjkDc>i};;S<ZtL{1<3@LTB%+oJk;7I~tx5u+5O@9})I+*Um02Z-V
zYYG3o@m*6Nr7j)HV{a?19-A_}^*mm7;)MNZrdayYM4-H2|8u`{jjDD|uQVGP85Z{S
z)lG2?WsIFp!Lbova)g;eGu>|*HE>@7Kz6o>`Hi`j6sknM@}+-~1fSkEvuZte{vyB;
z$aHfxGpA~R+xDKq%I2z)3gZ`oQWs!&k2b7C)0re@jy<oYQaj(X)8yDWQe|0Pgo&Tc
zYPKRV7(UnbroTy<2^Ond|2&89!T@Bhe&#Y*$9oZHnoE8jrSzm$OFua#7#oj({nbqW
zP-SYRqR9HMuPq6voa|)4jO<C_KTnxq^FNI@p%yMkksX4z(Zej8zlUM{J}(-9JI|Bj
z#s?P&OeFc41l0nJS9`syQ$K4%=K;+1>nzKf;MjoU+3*`Re}@{ny8~w1thLh-SJ~Z$
zVQTFIu#^|3UQV0E&40kXuMX&!cVjYk?b&YLY*e+S>Tmk|bMHTc1(-vRs`zZke0E;M
zCt7BP0@YMq1}hM4n9W*B1V*m?0|gq01#!GI^JRX#3sAJ|%^sh?0X|cY+q5BYj#FDZ
z>cdsQaejZYh^9vIl`3n1kL`~Pab3|8rLLF0!@^*Ngeaz<ATyeb8i>!F0^nsRac?~Z
zAPuv{^9=ZA1!r|4H5;MGmHY>ZT*sMI6!iFyjt;D3%`4jtVw_7KVWV^xe)x4(W&M?^
z<Ni;}&1OuCp=$i60B%~J%I+j|p~jIKMD3sTWah*xb=`TATEE-7Y2Kj_#Wj}uI=lWW
zTvF}=kMbmhzyBzKL)}WBx}-a|S-FNvOPpPvcB9N>(iN@xxU4kaqNQ<Eok^xP$s1~S
zE)0psFh{q{Z=GZXO>YOPg_UnCF)%p?>WiM3tI@^$o@M*{!n;~c+zV;zTc2+;F=h<8
zgtk`w=AX6EPO+UAfb{7tQVaVOTjogYBT98#L7Yh>a_j89PVRhsd$h=r9Q&H;DL4@b
z^899@86h?Gjko8)jOUfwoQM2nQWRI6D32R{1B=})elx?&=%xiV@fWc?Di1TI0q9O@
zD)~_6A+dW`tHFKUfjIA#-$dWPZ|t>r{+KM!VAiQi$-R)jdNHRUGB~z9^8Af|Y^D2M
z0C9R|&mrVM;-vcT&Q^z#=(aiIjjo!Rkkg6F|2jTD0AQMqIns82r$V<6r<FFX(;Hk)
za2qtjT;*!5UmE}Oy)$2!9|7Z4<W32>-1gsm%RS)MZaa_&ia+rApXTSn<Eu5{0QFec
z{_cC17WufbBJa(TC(R^XhQ?&uU-sW?x12BvfoXR;=S|E^0Va6=?;D~MzDU7A@qcdn
z3^Sur7w2TaIt{}PU>F;G6R#OhCro@rM&;mbPoSIsHFiIPf%($v0LYQDNSo&SBk|Kl
zyJd{i3R~<avs)6?p9MS966Dvr{cl}T3=C2|#g6WZ>@)qdAon+?$p8Nw|4a0J=6@;7
z|L<^%M5w>iwyzB2HyV+k+}|&D11PoMG)DdA(O7M<2n~>7#>eLTM-G3Q{~7v~_Y~r-
z#{<>|jNMk&Jv7(<N85QuHMM>1-gBfKsaB+eD2NDX5Rej(CMZak-jps?LXj2}6hx`g
zrAzO<mw+g}_hJBr5FqprS_ru-o^$^1d+!+ce!B1ef(($Iz1LoAuQ}&4e@`r~R3hx(
zU#T;ulKXYx8r6~es<h03O{v%X=-=>(f2UAu{6m1b0AAIlDzs28XXz+f?|&(ZXI!2!
zymcG;4_Sb>vCZc26IcRF2DoGpT8xEGFM3x8NbsQB{eZsuFRHJ#_5){DAV7jNp8<rB
zXfiG#vVcf4K<h*{=(~?984z;QBJ@%({yU#uJG03u0~oUqpd|kOt^bOR>hc7EP16?f
zK;zUO0gbXJVu+pg>l45;dm`=N9I)38;#`^ptnAwVqTUD=r~v0{JmQpw`Nt?D;ow-|
zO;q2F!i;9WYmGvnJA8tc=?=y$dikH#)27XvPgUb*8oz6S%zgXrpZqu@G0Lp-Cljp;
zWub#xmc)T>`^36wAw6bK%Tq_m!?NTF%{!CH)yR&6CrmyK*Js_yz5%w<{;$gQMQ_Q~
zT=^Gg2JYIrgSeoUAA&FA>kmS-Ob_gRSYKvSdVHqLy=<8<5#iV7EJBJW%r4~y_p~nx
z_?I+gY&~sxg_s4~G}fq)@5#PJSKt2dV_&z?Q2V=kJaW#DyJGWlnX;$XTq0w)+b(?Z
z=rb#PpE#G#DFeveIMW0tu`vnV#DOQ768rtT&y?cQ!u7pY(nhI`IGSaW_UHzYf>q(7
z7iWi1Pq!97kh*<u8HwE(RZjD%<X%p`syr~CciKXH@rdU+`iX{rvqYqBMbl@Ph)=hK
zIxaab-|l!h4<`A@6nUR}(KFL~V#!8@fMl($CaBv(uK0>Jy}IC=kPV=O0rLburG~;s
z|2$p`v+f}NZph7?S585UO>$RQ%KN9|CbMYX$rco4-@VrBpDSk984#I9Z+VKod3Crv
zZ{9%NPAS4$DL>mSA`9SbF^0a9-Pxr~^klJPr#_E5iqg$p?S3RjA;m5SMglrl_Sy0_
z@6}!rCL^iG^n&016mukT0{?`i^a5p)Fn9X!OiW0!#rQ|hCOg?O=D^6`sPGylV*f2K
zC$ypwK-$^ljxJapzg%9nF=FcsTM%>VS`QBwG)NANL~bVKv^3s--mDR6d~5|8o<(nw
zbB(`L6NB|6q=??Y@JNQKt-p5aY4AtfEJBcq1m$d;>$j-#2l9`8Cf|FC{@$FuzTh%^
zRE+6JiQ&we+1%f1J8%-}ez|RTE80n>`=s$dv*vZB%Eh;1@+BS2iNaT!O_9dIQc?u&
zbFMT$Qli5$BSz?0Fta2DT(T6^SuH;aPKJ~I3*=n9R=Skb%<T|_`?cabr<rgYA}ac(
z#vOn->7E)2F5_J)yWPaka+@YdM`u38ub#K_XVBoxh<EMvh<*>WB_c0Z>vQyFj96;d
zt`TO66p?-LcL-~_)2eaF_?&cwpZ>?1jM1(BsN|qF8b4O0B#6HlO#`lXF`kd<8M%{n
zu6v7VG-ah^$e#Yr=(N#RP~T>DmixBkSx;W8acSL@2wjaLaZiSmwofeP{MG@H>Q4=h
zGt7c=0hqyxu@L1I!?<`c+SP>8l5yPm?P~F^n~%h~g`9(%y*-1k&Ok(ll9C>}uP1et
zOYlCCCe&PS+%j5jasm`Al>h3&GsLaRQwby)myv%s#jFPYaF_EIuLS)ZN7wPXU!uBK
zT5G;jMei=Gx<z73cm75|J(p1Wl#o24>HVbHtzuC)J@$(kN^#}@8Ek<}qd!{`BDJ&D
z&BR1EvU{xXIMrA8NaCtPF3_FNpleMA3nZ@h)#}%4b*}i&#h@}5Dlc_@^gMCAM9%;3
zGMX>=gaML#w!(H8Gv1${K^(PbqGujR8uY9PzEH(H!~zHymfG+lMhmsA+l@A}f$^s<
zk+{Q~EFY5+<=>J(H1`&W*RgyrD}P6+R5K{H*?FOW^bk`JtzDe$dS9Jatnf!ki{|tl
zc@qlNg@-~RTn`*Q9$XhV-1}L58`0g9R^uQ;yN~+~a2HtvQ3Ak$3nW<RbYpJ{?ak+U
zjMRJ?@&1)-&K5ht=JhExO><j_8;lMGSL0mmlvi_nRFI>%@3WNJ>_CurA*2$Km`apd
zfvc=C+s&t|F;Ow8%_0WUUH9z<NQUVyp?vP~Fos=o!MsM-fkM!j=4pB4w{2kVFDT^<
z3rD$W^T4V06$Q$@f2}`Q4n&x!Q=j^bmCx-?0-mrz$iFWqSL4r|mQxAq6=VI`8S{Ie
zk7SFZ**10*WK#xOA5|5DTuBd}%U-{&B*=eV%LgP~33xSjq9)|<2<*a}w@zc&e5uw4
zgWhDEENB5UtZ$-Al@f8Btz+lAsskbwv2E4071+x6g491>@}O;tK&O|x!bG&}1jH5l
z<J=z9P~|UGayk$^*P0@(=W?p{4lc@k{izQ7GxD=NLq)G2OuDqVRTub4-+2|!c1cA~
zkP6&@M7O4)z1m5!QKoyI=j0yZCUAY}C07U&LjmI^<)GEm0PP?L;(UbAi$@a|HCuAX
zh}E9Y67ybd{&UT(Vxg&#bRSrb0SuhubU!74N>_tdyeMN@H*H2&*SY7xrPf=fwbGjA
zxeoNIP~+lSgaIv4RCv%BgcduyWq47y2TS=xNj7L>(VPX_n}O^6GPxcBgT^Q}8xz?s
z{~{e$HHgq)CIUr~U3_m%b?_DZVY4qDwIkP&oWj+~Y3Yf##%Z|SFMZ<exuf;^uZ9C;
z)KTpwbPGIYD)EDFPd3B>X4k{A2M^!zvKoPXomp*8<XlLb$P{gp*bH~{GmO5w<-Cf8
z1|4bc{rL|>?+hmkOitpa(@I1D8%WdMGf)b{b*G0<$Ek2~i)rYY;M=3tog91CPC&y2
zuud>~ViyE6%vtw2e^E!7t><n=FK5<t$LN#X1zBtF_;Y6-;-eFJY#ha*GtnDj{1tbg
z+}a4b-l1X^S!YD(#@OYzRe|@(d(d7Urr%J-j08tt55WZbGP58w-3RQ@U8-qhivd1X
zg7Ud>ET^S|zS+Mrd?whdhdtz6!Pwua57!eXr5ZqHYPwYlf;EU6jFak&H7TcKM^2H#
z$Pew`X=t)EcPtbp8daQj5sj3#G)8A3&<Foe0MEF<z<bS~hzN!+!rHz}78)gTTEfe?
zW03m(AXnMx%!7J!`Bjf$1D%{T-MDQ&^INy+pihal*?$5%dbh0Z=@17gj+%fl1C<_+
z#~bpw5W8W&)W$eM6YdK_<T!vU2zR%Utj~2cZ>^;sKCj0|m7hXNP2zVg?)=C@7)MhM
zPA=WCI2{A?GuU-(CK_e-`FDcltE_VTy_%ZXIDhM_D1}lO_Glyf1N8lW-o_r_Q_sv0
zaW+jaCCXP22Xcv=OYy6J9jVrsh+-D>9p~vE0Q;4LR6u5YG7a~+o||SkvRiYW&3+R6
z;Z(2qDqhRm#NYc31|1AbWT(1W*IE+WH1znUKy0cb+hLdFfEIqJTZGC$V6YN<P0$&N
zadsXf21Ea3;jS>f^uF&&W&!?+-zFr~4#jD?QgkEO*|h6$!~S$@)oMG$AuCI<7(j2u
ztQW&W)o+ui&g%$0)Izjb3AR>`hicj+QFPXm>G_lcKR2;oDs_LTY?z_H{bLF8;lJy!
zo#0ajtx+f&F0$|^Yt%8~c)oe%52jmAD>9!@M(}pr&{xfE13F<LLz|%y2aN7Zq2$GP
zIoJj@!&0jsv!!q*|Jcjsp??a43?qJ2_Q3$x>!D-Vm|3l*OTNSBb+tpp==rL9e}v}T
zX0%!z{zQ{qWFMa4tTR#hMaxWr%0T?@Yi|BFzizJoaJ#>YbhSCs^xik5^##u<S6ePJ
z`XhTtJ}5%0siJ=HuM*%iu(qa{TJSC|OpOQ0zFXMbTh^uq&4e#rF7(Sw-etu_C-f21
zS;dL`t_l+}UWNazoegfQV80xoB2PR8?s=u?Tl}`H&cw?!2cGUaq8GIL6=xerco_Q8
z6;mnM^c)MdN~ik0qI+R`sKChMy8dz$#r~J0na7mzbMM|z1|R*9cJ5JNSXfQK4z4La
z9pirCzZq0(3fVI8wyH4=JH!A2za>hllzC-T>(ry-T15M^N0j4U@gg@Y)B__mOYb||
z4m4qB9%q6pngQz+vC&VM4a+&f_%`&<_g+F;3Em~MoLSXsh8oq8e;pn$?n^}Mi|`WX
zg}Td3a|eZpKaZxyTmBux=TveM@4e@j$X*nsKgAw}S5^FZMOUa4==zl8Jum4Tsxcqv
zLe6q@ehhExfw%@g7@w)YOHtM3XFD)x%}7En@2~YpipEmFBSr1ab^}y&Ij;-7vf}7n
z2wmqD{@OFh@ArbE;#VYS1NH7UxB9uA>=A*eT9lduafFwt$WX4;i2&1kRZPm{C})<$
zu?xJlkg$7b^^mJ97&)#|8W73X?9)5`#0%ARj?OxkQBnpH5Q#YmKuD6$Id`qAE&sKa
zjX6HSC>VlG)rL8Xft8$Hp(}E)=BLo~5l8d(ot;V9?O8gZJO2*kx$|$`p3!0s7QVmX
zFqgvLWHH)dm-pCqSlCQ<VD;NW6GmtL-uKxa`VYQ%0NzJNo=wlL0%hDcc<w)<xLLb_
z?7e`KZ9^WX{odFPu<7ePN1in#4-1l&t43}d413OkV%yK3wDqn=8+v6M0F<jp`Q)qL
z*8Q4M?&$=!xId7+tE(4f*c$#4oNl`iOM!I`U%ID?sjLApgB4Jr&Ck)|SorKL{}g+|
z)KWZB0X9Yj^E+r|9clj<LcF?V7g@K2PixI7;Tu^QX@5606TDc=2oXI@Io^<LgJ-d#
zfmVPY?S*?_=EN{W$CI5R45gDhep7v?H4pnzn@&COUPZ6gyl~=SKQU}qwhNkwfZn1m
zrCYk6;4_@r+dV6;9Tz<6eY0WgNjPf7MVdx*K5{31t@f*lru*b+;iIP<`HKiP8qefm
z><G<jK~43xAk`iCRLll~K*W6usF_@9D|$RiuUl`Q(zV(92vMT?sCR!Z^eGP*q}k}a
zR6<cp<4Rh5yLuXVXG$9JXIXcc&jA7F9<jXO%r-+=v4HZ_<~hC1{DtwKVQX<D<4T!&
zLuJ28wONI7d8t7033CIq>IhlNWS))uOYRh=NH;&AAX^fiZ5EU>>SH|9%m+KG!S}0~
z;UXwfPN;jRw&o;pW~pU@B;)k#5Om;5c+)~OaceU`Wdjj!d7I|YT0la%F+S?N$HKC6
zHI!}a30y`WfWAk3jCp*zz`2y{U7Mm|$-pg1rvT-}_NL`zzU*vNSlUJG<$Wotd|&Cv
zqqIuw&g0)woP(>hr$|Kx$Pf6$$jOHL!@+044c*nO-@TW47)|EayEF2QMgk*upY^W^
zFbS}&8ph<zjR>`4Df6MXB8a{hFslj1pMmQt=DO5oz)7vQE#?OBdDa%n&t7IP{pXf!
zS94LVt^w1U)Cgr^DhCmg2BQ?hTqyIm#G+1|f!vc>Z0T>3u#Afh03K6p(o0qVRpWFC
zQBnvbNo}sSJNd{_kn))zc$T4{#fEJqL4Ml=J$d*?^kzV$baU8I(_+EZw|mKcG01pf
z2wY6!3KCo3<*Bo5w%Kj>LW4Jw=11QAOe%>n+TI#Ml%#I(w4B>YCY2hqLK7>7eh1mp
zda|X}O^Z^woLLiosglczo8I3M&nO?W9K<c%p;1UxV$j0wYQ|@?EZ+!x(*{%=^D&F*
zT^^v&0kzO;!anPY5t5-dUJL!&kgqC!uIDJQtt{6?W1#lL@4cx@7nR(LEwQHt=SCmJ
z9;7K*K^?nRKu;(TZM}WEd4iSC#(<2#m5sszmiV-A1-l6`PI;h%*n@PM#1QvSqPQzg
z`3~=ZY|K9kVk?tu7j+Id)lt|BBJMWTxg|2m*VLH;v+9J`ded{AD^Y+0xVsklYr+2N
zTZfaD=o#$4r49{t@IQk%vj+E6kAIgj{$QN=U3#SC;<!C3zj@8JnRy4pK)5Dv>Wmsp
zg8amNTaV8dkOLL---ias+Az|{Dt{*<Z;8ghDBFTtzi0bw9=Cr#y^g&!Ewr@`ET8c1
zU9o1SDgz7Su1&8x&9*=YGDxla@80k+7nN6=FIH9=ia<Jxo!nb-6;xO-PHBqUY2#eL
zg&ml;Tm_ga&rY&;HuKASPA5&=+OqT$<$7Kj|5lhD7=YwZLMfE0UeZUYBUjII5gAe?
zt=RU)1Cz-L?&-IFIPnnBmlxw{>_dbc)t8+kdQQOh#ee?O1bSf+tYOlu-TR=_c+Pvn
zbhS~a@X~au<Zt1x$tPT`lIwB)jg7K{U+6eigi$mgj0Z?8NdjxM-sHt{0$j=3pKdK_
zR>f6S2q2+<A0YMFMZH`|FDFg^rzpTRgUmw@4f`3ijk>jVukCXH@!Fk_nZlZ{b^QQy
zpGP!%{Jj3j%|fDTktFDXqGJEYXYRoU04LrtBe)d(Ka$d+AKp`ExmR%)Vh1hPbpZt4
z2U?K5(|oHPz_WLgGGojcy>tAo;Y;Hz*CwAaF^76X*WwvJ*9JAGPBbRA-}jHfDakgE
z)^S;Y{bJ-X#Xna7eEF{b^G{&ZYxqBx+U9hvxJKXN+%r_1bB_g-@{LqnSx-n?=$jw^
z{RE-NkgcVj@vW@8&5XXJhbBhpU#qJ(5f$)3dbx1o^f_tLP3!rOy^R?N220jBm&%G#
z$<j!(#Dt{5_<hW`B^yu5cC|18BQw&UYxr#>O8QaiBI=p*%?W;Wv1~S9(yed{{P#n|
zfJD=;LCWb9M>Xp$WaRR~YZ_mFGwb;~qWs|ZdPt1$-$zs>G;sqP7zv8u-5r#~#l?4$
zYn5=&0f+J@{ro=o6u!}8f!Xjp9a52wTL)}RUTOFPMsh%bAz0x~(0|yyvH~>WWNyb{
z?*A~2y?y|u@#unc>*R9#2dM@2^s0~kDa0$A8-i1JTL4J;8vrRYli6**s}8sHv{#~&
zAtwfne3FmlX;ixLXRwkAB&*kBKCfu@K>k@G5&gj>iBzi4BjbAGd_lIW;r!d8$(>%j
z?DnvsveDjGo2MLfJUxa&EHSwK-@662#CeOj8@_*#aW%>@>|O&h7*<kVtpweVmK>3w
zmJaMhl(ukb`}@&IcY7Xp`XL;6mjfNZN^}z&UBUAEPF@0cC~cfmq{`)zZ$d6<Wx|#I
zF#NQ*FzP@OakL#biF^G4z^^Q_UCbnFzEWQ>UEiH${xl@NX1MQz@ZVhf*`emmamfL!
zN41_IYab^NX<ge$VD~t_7sr8PaP$+d;5Z>b#0iTbQl}`){kW3zZ`VmF;MbhjA#a5)
zA$13{e~9Jm25U8E45a+>NkRTal=9hzuvh&^!*>f|)dq>+p1h9Wy#D8nz_%Mu61UkK
zaKE+}klq4|ZO2ZsYI|4yzEu%+3q!Pz+H4-EC)jHvR2@^aOBF`{#u7hL?;1TjJvS2-
zHl1$s*7QO%x8TQpJakKcA`R0qwV)685||%b2mJAW@8dmECIz0c5`G(-khnowD+_mL
za(gh3whamZ!C5vQV|@)ee&BqBHVZ_P%x7aBFkI-^krRFu$qoR+6ZdgLj`Xqm_r(cn
zhmohzeH>(IKdG-#8PxW&0Kd!sq~1T4Z8dOptM_)?qh)2B$>laM^X0J+Q?epdh_-C@
zlTB9Ocf=zi0pmE&Y%u@EwB)af5Vl@dC;RnDG(T_>QTMQP_ts&j8F<QaWvE<`n`$*L
zcQVMCk81i;yB<kJ%4gwkym>5|Ly3}~L?0W%{o;cA<VdQ6$Au3!d&PcwiSjjilxhiU
z{nchlx(J{|i#kipP>@koKZ%`3aINQ$txyM*^RJwL<}FINbKpqp<?#$Wm9r>XIm703
zp>Hipc-V9xt66ht2IYHkie*vPa=cXCvOl@drFZiEtEOd)zX%_(d$D9(>SjSu7=J1v
z+?`2QSqa)K>Z<~{!up5;d`81)QSnGOrdsS8CM{qvG7}`D-Mk4|0~tlR2Nw+p!KS>X
zwVQ*;?Omq?O;UM81I!JT1>&jyX&h1iL_P6T=2U$P#Ea0EIqbjoekiZ~yypKt*5zg#
zJn<NsklZ4)lVeFEUu!JISacrE|A3)F5WJScQbk)*FT!ZZxIQXQ?(*xUb?;lTejKL?
z{)}f1&!z?zr>qzdAA}F^-BDuL{Zt^{-zU`@@AiIMD5DwI$3gyU*wETp8389e%JDeS
zf%;)pKd(*u+j<Ar&wyo3gBiSO))A)|9rn`*A|J80mB>THN9UR{Xc6^@M*by4Q?b!r
zuF}pgc1_H&VYModC4#?NDOF*4VP5c!0qVb>Xz)cPGdHXh{f+gkY?mB~8?C|*j5PUE
zC6BL3p4&1R`vsxjp9EGA$rdIL8AgW1$(O;aNvk?%Iz*;3;~1KHSH%|iM#fYAo#Y!a
zP}yfxvZ{Oe<a3}H5F7&lZ%h7kH0kqlgY7~L-JR1~A{<FAl{1`%K}@o}cOYKgb68J!
zFa4a^uPLZ8PpnO&W9as9lL`~;NoqC9tv3z3)S1q^9NR(bwT+>kzF%-y1_P#4W+ZP+
z7RbUzpy4rv<;rgm@2As_qT>Ph9ZHqT^S<Z;?=#x-iT>`9NBcHXCNU&eOpd6GEkdi)
zB3=?my|{Dt55{Y1&^X_RlV3J2NiDhG3vfSq_&2y0wm)8fUP<P4%ZFUEMLf_vn?)fD
zXemu{Dho$z+oXb*Yuy`Y3Z<rEM)f-yMQu@H6ywoCEx&Cs!g|6@DJTeT^Mgz;9uG!b
zCZ?vPQaguB)Y>t>sfaIOIh1DW4wBwaIy?NM{J6>IP&ICJUonI^x-t$D_LR@-_?gm`
z_1%RH_0#%^muiLc3I118sR94Cn61I@%UUv!Q(KLTKdXn=Gs+Mz4iwU@PK8Sh(wpV`
z)$H1vYozgqIC>SOyH8v=Eh*0Kc`HU4^OQ#iT8cHtnY{BhHIo^umu4wc3LvBn6y$!{
zjjE5Kh2;6VASNs)#dFwS8dUuGCQYY)XSrKt3`&kFjZB;s$9tRUVuYS<rnbZ_Lh$ap
zh9+XxmW^`dp!2>-S`S-SVPZNfG1waeGm6h;3pRfXV080M%Jm+l-5F}N4FQqf92e2<
zMVl_PqA|BB-BS=KOhq`x_e;1(eaTg5J)!_T!}x-Y9Oh(rgxOdey)nHx)hSp?`=$!d
z-Yh`4UFP?EGCiep3Xph9{`e`ZP*kLc8s89~REme3%En&wOPx9(pMwpkzL0ivml?Ba
zUfc;UU#k41TR@}7&ym>2`i^NZp=;<<nAwRNhiTJw*tImEq2u-otK+fo_q70M)H7#%
z$-xt?(U5DQdj^t4ESW5PN@Cg42d+9JAJPjv<IY-zChlz8E}J~Y<I!gg>R3c2vIT(4
zjUzAikNbn&pF<78M%YwlPD!@3Umw_ZK(=iKs@uk#9!L#Ldn}?ED@y5Cu&I4?Ii-ti
z2fcfgMf%)RAw8vi^BS`a;@pjCfkGXYVgW1-83)#5SI6pqMCKzJe1#k3Bp4y|`KyT8
z=~s)9S5j@8flyhq<wS}XpMNwGhg`M3l=aA9<e*#tk8Rfkr00>k+UvplqCOYE^8fRf
z@+N#G`z1`P<|40|UYJwjFE8&C^hLW4c(3W+kd{xryXr}@X;fzjBuSNnO@r^&q#iSO
z#fhhPn1`wR`{oqwr<8-lWLRn;!6bs9dKC0+e>U;NjgMuUI>o6+$B7MC89{0int~Ki
z+%UdM_3ToSY4>m+hFwtQagXCRnShOYZyR*hqkH-`ih6X98o1<LOvNV;+K$6~)>v1U
zj~DEE_ebwMc_DSBxWtFbq^9wHEmRTxPCC~Um^07%3!&nCwoVtHynLPGn^=rFDZcvF
z7x4*pLLRmRZA9Gi++azSRBqn)IyffwG)8CvwT@%^sYkqQ&Y)67vd>^#6Z$dhVsElf
zcsNE++&yNeo$=_v&CV)&kDw9jv*wf<TQy=;2eKujG67W$)z^Wf#Rb+RwVsQ#Vw!G+
zJFO;lvzL>$b&TLWK&?D@loxF5o-tYI8Q$em;az>Q+W<?Y682TOck%+j-8KZql2bp~
zP*yNVx6(;Rp=x<Y=-4A?|K^9ja*0^ZbjgAHNZYpBpGa}fZe?!O8gZaB4w`&F?+&4$
z3!sHisDW`=sZwV52$;Ga0J4TZ2T>T^kMa-gIZQte3lR~g@hQ&Va!y(tzIg21vj%lv
z2LdV&{`;Gx9L&U*pib%7RKDgpRqm-#CZRLKpf^o-{`-NukLWpN5iJhEN|rWgnrZK=
zp_>__gKDI0iNUFW>PLLV%#*X~@7x3{WEg$t)r^XfpGLWn+S`dm2M{U)E;F|+Rz&Cz
zkV>H(Irp(xkY$VB(r+LBVylsGR2HS;=loJVyxrWwC>AFs3}L;eW|B7b$j+sTIjDh_
z!V83LY$)X4y^;JLCygI<EwBrg4_`~or0p?s87p{++=VU<KEsw?o+g@)p9Kl1xu!G2
zLJ`E3AG((=CBzt1tM|`W_<UIlD%(C`uajgHnlzQ}>kcf-C>0_!iI+-T_b@&d1I5i!
zczcXkH3BcR(e!yIF!hlj^nshng_H1jxUSgJelBR}6yc3PH{A3DDzEgiIU;DEWRBw%
zViR3FO1fNiw{r;P5^iE*Z;VMM<6n_V9=TYK+v1fafc(a2lIzu+>c<g_wJO6O&r5V&
zj(cL`ad!9pZRgcgrDt35K+O$MN;YJZA6J6lxvkFy{!o6BZK3b+2`jtC?lrhY2`wn;
zc9}N)T>V2{wI}8=W~y2*e$|tIqE~=Peq~ySXQiP{T4+pU`xSp)@bU2z6RiT8p!Il2
z0A5VnU*FwN-m!Cjeh(1I5=HDh=Gx0ob9R-HvvgA486~iN>B8gNs_~)6wY3L&IAeU@
zQOTMvk_C6a&$yKJ7+rWkXcM1Smy}^09px|OiFB9mST2U7uh_rce0Lw;1JVLlJ9&9f
z4RpWqS-4^InZ5JIZcwKzXm4GYz~Zn-MP8_nRZi(DhtX}0B1YPZo)7Lt^#}6E2}z_h
z|5JhX5>)#eui>Z2d5PgnNMCru+?TU~+-s+u>a+Xv{E(NkGGcpO7cZ7Jc(vv)=%gea
z1hCiAx=Z(2_ciBD80n%9z7`6ajnF0z)9MfsYf?T2rIJz#@udCCip&+_XcB(-ovC`M
zS#F*t@9_QHd`hLQpIxcKBY+Zlmm@gQGaojfeDQV<&BHu*>p*v>c;9w_C9!=C1k9o7
zvIU?NIl%m7ZM~~c^3#>{u#qRwCTXOq*PXFup|2h1ntwzd8ZG@1YkfK-H=ix4^ZDVO
z>Lrm5_CmtW=t6GHh5{{13T<RVZ6P8p5#hg2qF=(*sU0roMGhQ|y(ABNXi;j6NQ}Qy
zbJCH%cJyUrFO^85LIzPvB~W`ySjO;fnAE8mw!1$8@=gxI?kn-!m#r_R>~rV>^__vk
zTw2kYvFQ+wu2BpjeJw*_HxXtiHiag4T7NK$4SmSne~uwbBs5iz?YT2=>8WGkvGlrP
z^k+so+K`<1fEH`ZY!y3AHq?gya{iny+RBBt`^Az-XF8nsFfXR0M5M$$TtFgj?D%_3
zTk<Zv!FmFeNb~qB5*Q=DqCQI{jEAiWh>OUEBunHyc*a>#b0+KgASF|(jraYp)6b5E
z?ceJWWjH9V;jL;^tMx~r3fAfR+456?cO9=zuiQuu?eH(IUL+s->IihtJOhZL<HapT
zYcp-kd*&X87;BSM1XpMJd%u)PAzaotu3@f;y=c^cxPN?FkZPaSLg#$j$(nAy$V{4M
z%R&ns==oS3^ZmwC)5?cbE0!FK>IKGGDb5WoG(Jv+4Ki5fsPDz0n2ZOo-nt56r`CR1
z8TsT3ygqCzxS`myp#eUw5yUoF`@Dda$gP~#gdfnj^Z3!>7kx~-nu~hQK&=0&Y5N+2
zn$p$V`+6aQCU5ZR5uDLiEKFj6Ai$PIe`=!W;y-obryVZi*b@^kZ8r#!AP^AWQ@=X`
zSwG(Kc#QJyvbUB9Nxb{|IoQW@Li=6}YZPTbFcQMFZYk|C@7l{np0|C-JFu+Ky#3;v
z9{tXKyM~WaPE$~J)tSWaeMT`*C97N5*RxL+<!ct-S-VK3w6`jRJrj>U0p$XYLo9PV
zD3t4#4Y=80Y{sbKDnUoJ-K+QL__$Fjr)539S2smEKV2v)xha%_RqTwNOONK^6?neO
z5?Fv%x#SB=%PbO@Mc`?nljF`-Hu8J3DX{wS^KV1<+pX>wgRH}8L$`OEHjJTSg`7fr
zP7>3&l04x2Uc!_noVWU}nBa@14JicJIEK_{Ukbg|r57BeP+45mQQ?#z#a{=v1q#lu
zjT4v1wVuwAbHqMxF<UlVG(;<?3az%_LsEZ8Yg14U$O!v-zlf9!wkz5{;9|OSN4`4F
z?^)L;ruRj{XYvzT>kOMP3PX)wj*NydJLUsqQEN`g?v24+h0aA3E0)gQFtew|o}h9>
zHoFD_)?X*HX|!i$Q&#nr62nG{da+_l8u50+?$ScG7snB7lVhp#X5+98e#_>V$eHMo
z8&H|zZv8U91H4GTM2C|7PWtD5@971V*q!G`q^G12!O@fN*-#H3*ZlpI+!)C%o*0*H
z`PS^J6MQIirK*?{R20`3DFbo4JsHsp<dxf~i}>s`OSb1kfei_#unTvuU}TfOl+?c=
zC67qrvveWk#SwNn5q6poY4!bA_vl}ii;IIhwiKv!pG@@y!`CZNb2ahss>QouEZ$VD
zBI28`9%Ix@1LKd3Of4F5t&_v{V?*?rrxJFuT%%OWm)8GcW<SkfcK~e%nhiW0QvIx4
ziU{X$tGY%m1DM-<nJq1^)S<#egfT=~QsCDvi{x4t<7mRTr_%Dg#67#@4KimHMzL$3
z37D%D1t5VtP2agtYR9UIQPgy1s5lh+^00L&X|LQh!9ZQn`IjVlL@`B{rC|zKRcf^L
zi3MhmcE{Wz%t+*+AkE6kQRn`L`94qcwVS!EIk8vlpoRT+V~Xwn{QI5wdKLO3Eefsc
z92v0dwA-@yIdX8!fC<l#!arU@$@Zq}^V!JqU6*P;YUhI{aZL+oLO(;)<kFjh^yG>$
z%_6nGq^T9rR;RsUJVKu)nJ?b-pt`AJ>l==9qLc28F;E)Qw#JSCxlH=|`tU1#8P-AX
zRmTv`2vSI(S7a_px0-KOv_0{%j_$kStQyB-QFCOxZz5##%gSk?D%`qPO1+^DKCTtS
zpPl%fIVAUHL^Gq%acZ~67d?A7N9rx5r=KZhn@-}-1#((;{>U@g{<BfDwXykfOHtg&
zMXhX;!`d<Y1}D{OzuxEZz{&SLdiw-Y;!}ej#HmB^DHlg<0=PUhhRNu(nCtrHizs_4
zS7Ze%%bUMx?f)UgysstTcXs#i=z{~t<?252y&sRZJS_3!pZe4XyMUt;4U%SgbvrJZ
z;#(6XtTRYvbSRebtsB)<E3qL(`qXa19fa9)zS&j_x2S=4s)5a>C-ptLwYxz#delDF
z*)~^I=&Q6$VZHLL7WVUQH%Tzqf5^IDC?zh$hN}EM;FIjVUUsHnwCfzKIS}2sKVCnn
z5~zKcyuK(|D)G%uGNY+OE;mTQURkI4-sbOK{xrAw1*`D>axt^bozE64UrPRpK}^Kl
zOq(bR>1Jn*S~>U;59zdT)`%Gv?%jK<qp~9&KS*}XpM8J1vRLR$ets2XGT0t6G}ONJ
zG+B@TN$Dn6OcAaRE%@eQr}y<RaiVdM%3u-q8eipd0)H}sJ-4BcAdyy`dernyOVI!E
zm_YdCDG6GDoQWAf`t6KLS^L>q8rqy?n#(p)sU!a3CA0%#cV17ODZ;*1&C=gqX@c>o
zgqlqB-PmyXc}nQ&-lxx$!*}l^>D@KAVwy?g&Nf;U34>02+6c*!dKKlX*(X7{ZeGVe
zpo1ER$NKSS5P*~r>35#Sz`0p9(X~0Or22*u)#IlXzNC458CW5ulQJg;_xlP?8%;}j
z$IgBsYA{1Nm&GC#o3B4<9j39HIErKvVXYTJN^w@x>017(N&O~$oYK(`n)td8iziZV
z6Gy7prz+zFSs3i8IF|?le2s4e8(-$cQ3PGPJTm<V^FTLkF6s|_?eS-bXqsrS_7O!f
z7+dw0n>HL)KBCDHw~Ya~0(f91_!dszQXf!JI&!mr-)FVoUM<TfBv2XXwrPyj>6M1(
z!WzbEfd4>Y2Hi`^k_V<1&q9K}jTVc0If&I2SS@U36-f@By}}18v8zWXi3bT{K0m0p
z5A#V3SyoGpBJ^c_fDiidy=>Br__Fu#$X0rGZ>}Geaw>jHW^EpC#v|!nqVt?~_^M#o
z)_C>L)lUwR<0j<Ze)JgtK)2T9;ncLHvv$(^dvxp39!%lxiI{SmVjF3p@|fN<?-IY6
zouo%z=Hx7uabuV&dGEO5PVSOUeI9{dYkZ3>6LX(Jy`Y7G^L>U|wU^oCHwhZwebyk8
zrG`Cai$}V7a?s^+Z>>1?FHL$ybNMOqn?~|b$uKg_Z{@yrHUi@C*~Qi(?^E#~GrjN~
zQt9gVK5>nd_d1>RN2D^n9|MUb49@$DMOSMcuL3&RwPmRZnkXN{m(AS!0tl3&$I%|<
zO7ye|SFQUu-{tsVRBv}Tz%TCmsTii6h+vtUlEd?Z&d4QpoUc?bf96`f-jTS;62$sc
z6|FVH3%s5nS59nCir9Zxgw0IbosTD$nX0$wkDVp{Ok#vMiP6vkRkV+O;Yb_eo;wAX
zkf(Z=C34wK&1m+x!AEXFxKtZst9=DtZN^ZSQF0d*^yj@!4_R-yndB0Ya8zaSe(`Ft
zXYNANE$z8R^Zuv=Z?zei&iEaAn2?VgaCK5U!Dl8H?B4emE3zo<p5m|y&aO)CgueX*
zX4Vg@Jmk5sgXI&<)fq`eYc@^w#D0}QEs#^3#Q08lm&|XOCc5>v>?<}e?%7`W4|i}>
zz2LTwqWFy3i<E`P)a)M3P|RHqvkov+9&DQt4tJ~TcxH_#U5iGC8|eiah_{q&iaUH>
z#AEaXtfqE+`E@ebv3Uflv$gs1N$@|cON?Xp*<c>CvZEvDC(OaSC_CU#y)CDc8NOii
z)rQU2_|soP`ex;-VBi+MUZe7Qvw9|Ta}|Ws<1m@E*38q=)7sM}n?ID*pGzx^b|oNt
z+xpm+Y3-tsbs4Au;RzLhB0V*_f6v^vgkJbt|AuActxpG>Y2$aUYRjK#J1H*vW9~D)
zzyb-ztmx<*MoT26(fwpC#Eyp9w2I|(W>uHt+cM^vQT7Q;cf4D1vv}!UQ-9ul>#Az?
z>}Az3z87RL^NnC_&duvWwux+6Fh-Q!&(o###!+#{=EPK7z{Ak-!lJ^Bai+I@)B%*L
zevvg!u|ePBWEj&9{GjU5^5uV_Hnq<Q%NWfZS%+t|G;~D6in|N_!v+hIUu9D7fvTh8
zoAO(*#>*f7^jsBUn!@UtcvZ!XXpfjUqu*8m=10$Z;$y^Y8~R4?mkr~;msD0CCi&$K
z{WqwJumk*Pn-|6<f?-3uB0Mj*Z#mnB4@^G$Gd%XqreIU#x*_}~oFq$frwy+eb!4b;
zfU>{LlYy3C+#%?S_7<!79ox4FKc|c8tqLi*d|cI2V1$Bf8oheYG6SnR2X=d&jUF1u
z<D&<0{>|L-^JdX++RhuinnaOv$3A7(TH9HfuE%<16PzA(_&fDPgn?e8!)ukdzbT^l
zWvEyvxcFQ0qA&dzTpOke8f&o$O9h3c*QFqftny&+=#b+hah044=kIfu(vsQvf<NKG
zkQjI4+M8i}%PMrAGK7{9D4b`v+wO4N+A)u4&+V;?@KbFcMr>zD%h)nQU{)Tpuk~#h
zD)P$rqI)7ZT}1Y89l0cq-oLo)uGH?uR*(z3%yadnbAL}}UdGC?!~XpM4|yLah6=3~
zOG#69I=Fv_H?{cV!GUK<#+O9`>dThPUQ)?{27G7lTsq#%8|rMbU0>XWY2U+J!n=-h
z7usLjYqZAb`ALz#xqy7X<Xl?WIpsCZr5R}1CJxUT2ulxET&vft<{Bo;Bn3#rK8zF^
zV>pz_wn5zBn&G8n=ln20*IFyKm&%`HF3x*tOz*r#>CIf;a4U%^WWi&ag7CbYVm*8p
z30len9Sctq{ALNbChjv#1p*BDBhZ`F-3tNYW<?e2Y@S6e{V=ryhR4g^)_K#XU(e`N
zaiCbXGJ{;6`Y(z8n1v;|Pn>s;nbx1Ck;Z(mFp1MEodGj`l@P1jUOkwlIM5s_Vrf|`
z+ogqjrK(9_ay2Q{7O?tZTnPKq_?VgHhks6d{>s&ntzhq0>M8Dof=n98ROjy>Mmk6K
z(MzF07yCJ8s&F2UmTkl?x<|K&nqkdo&S7igo@vSt?_|;XIj!}DSJd!Cu&fDJ8$C;T
z<u~xm*sT-a4STkq{UQ-e^HxPgh6~e0lmR<t0hCieQjLb&w-=STPvu`&4~=-caOhx`
z6a{|wAX2jn_9}0L0LBz|;Y&VSD%^hkEQLZRWlK{~2^e-sy8=Y+a;Jex4R3&y3hf8f
zy#f8OYl>kl4Z(A+a3f<moLkAYW~G?%_a&pZHa4F5)bUp@QE-iq__bI94_bxAdM|^8
zHQGdEw|LZE(gwCvyC+8^Ey-fA@1FP6U4(4889N&1aQCQi;v>1WoF-E~C~jq|=$dC;
z-WOGIE_0_nH-&b`dd)Ot{!z#(k}KKUMS39G{Z7T<O;=0eho|bqF5Y_>x1wkc&Q>uW
z(W88d%Pbz@TIr7x$^BwSQ(RlJW4Z6$pYMNJ_|kJ-znG4DD~!qh1#~lMMz_XonOW*N
z5Btgj&QJ0-_??$?<mLwt12=<RKjpLr+U#5EC`9TZk^MUX6%K0%)~J}24Dohb!(Pi=
zpORNt%XV9Y8O352?z>_l%2<6=9=E&FT8XY@8*O^ew3Q3?7V0@BVsLw>j|+d!=w0Qk
zSbJIxHd-ko!kVU7V2HJg8GMiN=3BeDxNL|_t0znyX!piQ1ewc>1cpMJatA<&tr=7}
zCwX;(@Z6MbM0fS-Tzl}r+(ffZ=TMI9)?pPa^Zbv%B=#}`p*cV4C}c6)G+b|4dP2fq
za(DFHXOAX@QI41h+e_TJ@vQyOsWQgalP{`pfV((z<`5_f6vDUP<eD+Jopl&aX8L&t
z8rH&@7tJ_U#u3WW4k0cptF=F@1ewh_Og&I?*SGgjaH|oo#&!<sS>jhL?)Zn5`gvc+
zASlClLn!9jkIFPrL7@n#_brpBH}R!BR{CtC+Z*Orzc~HglqDdLE!-d-&_7SYw3Stw
zHf-p_jBz$g2JppDr1JDX9Iw@knhf-?;bHcw$h8>9ANijL>i?mdA+@ad;&y<_ZoG0K
zLCZTNp+m#H36SS%Qc{)dA@15KXAd@$ntEyFJW_;wApSl&S4>xr0*VOQ^H8{?%|Pr=
zOL>s;DWYi3+0%GLZ!Lz}mfs_I^rkUlW#tLbOoxZ^TFCf!o^5bo@VKFsQnC@$4QD!z
zmp+ZFYNxn`<g7eY2nty2a%w6NqY^LLZRfZD=;O+oy0sC<>fC>1_hk~p5*wx<q}(rZ
zGkrQN<g6#!Qhpq|Jmr4mW8o;o{lEp}^QQ*{p6Wh@uRxr)obl~%icWiF5IFDtBC)mN
zqmU9AKY`qasnvQ_;*2mJLDJEnC!3D)RlRRIK1B$<Md=>AeJnsN{-K1Uy#Cacd66IV
z1=LS0O(2w1ObJnqgZUo@WK1Z`x)Q|DTHO{RN69O9t;V5YJ}RyWS~(_{C^+S1QyEL*
zxtK-oGF_?(N~0faK9Jz(!3{~o`%v9*Zr-0$y=2b1SBkWA-`*azQNt-;VLPG51cx8X
z#}dK}{I+sRs@2mD@t<bOjO7!n(focHBgXy*>zNLDKCiV{*B}C=rq!t8$zFV+VTqrF
zKtO|thfG%0xL@B_`B<Aazi?AtoWw>#YCW!Q1YqfB*G{cOMQ=Nt670XZH@l7Z9!A}M
z`$ouJu%x9D;m?V==s8yT>2c?svKYghqK4IAn~Ime(25vy?ft;FMOs^Ko&+ubQo|sV
z-r%!X*JpFAA<Gh_akBp3&Iw-&HiWB&?kyk(amaX_LJFvQEiMy_Gu?lFtYXSM($7?0
zxq)Rb2)t~oDkU>}kg#JK&RcN1U^n0F61Ca6b7zYcJ&N#YWpe~A0Js>&$|goxf(6y(
zF2h@elAHH4;s<`7mDHz|^X5BeSTnB(Ls|*0arymhMcLhY&b(T21($^F_L92LjQpkY
z+pRC-2nhlc42_)heAJ@M!zRvRgnBpXbEwt84c#C2VJ9~Cf}bx#fb6KJrQGmt3d~?N
zsb#OX%=^{W?I(`YDd7QvY`y8;=`zT>c5y6k*CoupNOc>2Ml`EA`x1Q_`b*>M^DNQ^
z6XKQWQ8BLUaZYuT_?xpA#aCAZ8p^!{c;9O{Gq`%76(%^|wdKWtey&0$`49CJ<i3vL
z#(lk>{(w0qw=Z1q9a|v3zQGEhZf|}LD5gb-LnX)W;<02mT5L8D(-B3AlUyb(Z%PC8
zES_0-PXGeqyXIt!#yDqIejy~T%KnKi!Tw;v#o!Za38)MEq)ewjmMDGbg&9wsDF_V1
za~PU(V;wZm_xyV$)2PX0-@j1`M`&^}*fAGsZ(8*Y#rO?cE)w3{p2x&L>~VQjKd{Iy
z$7fi&GJ^yOLc?2Zg=AdKF#k<8HIw<@jo#+(h+CmSs2(d(=7#$<m&pE-ygX@TXxjXO
z->}r?Ku^R3{~Wsa>ErR-@uO2hq`mhGLEI}}7ABkxC18E(2rYV)l~}CW7Z4;M2a*O>
zH};>uF(`6;n&;Wj2tO)_eRi;``E4&~APF(*ldn*WC8p>c>^h43Wtu(KIr!!oc*3TW
zhA{_qyy?`5w98FomrhM57*2B6QVoev<K$pmO{wC6)e%yO1k|B*PWv*Cc@+h5X<DJw
z#pcA7C7#!<2=ddzVfV3k$i4fVXXodmLS%fVOh#@{CF0~rt=xj-0g+ucl$TA|=?2Hz
z%j|{*e3FJobm_%E0*Tr3gT2UcAV>66vN}bjCVPz5<<CPV5MG#|%a`hS`u}N_HW2tI
z@g5>`i=LuAhy}-1%|UGhKdZdgO2V|9&i_Zfxmv<|$l=p+t+C#2YwAAsQTAe03Og`R
zh|CBcv^?t9RCD?A;GNq*HvE7DfyvvmH4HO68+hwUs$h4ACOmp~b%L4+!4jJo<h{b4
zric3OLoEeh`J7oPLi^5_Q7T5^wYrw4K=yVl{X%Da-e(&VTGaByex-7ycChxug1$Gv
z>Gko>a#^T4V=N`;y{|^B`N%}{ct1RAbskCWc(nSS#sJkJ2@jUagHq-9*23FQ$FrVx
zuDCHpO$Qu*Urkm!F5@jcPA<m?yARYY{RCSBw%MaHZt8*`iLj2@c}6;~?7S+Z8Efp%
zvc7_?@cxa&g|JQW(J1z(7|$dLw5QoX9(F?T0Pgo|4-c9ZktX&dn0Z^<N1Zn^ldI;6
zy>QB6?&dW$o^((;tO5FG`y&H-A2rW>T)c!(nst+L6l$ru01fK!V>z=GGKFT5+bk%_
zB$x4dv{JymQK830Zj+UjlLrhRjG*O<lWid<5cfz6sFs2m`ni>Wf#)T1uCuW{MRb<Z
z^jN_<C&b#9@}_jb(GLM1xpHyduKuwvH-u+h8u#s158~`yS8gbShq;pr3S4xJaCv?y
z?^DWFaW=e@xZS6Mi%gS~j?TyxIB#iD)qWh1v28OKti=Swc7ED7f^@x~?zyLY?N&WX
zsdqT_x4$^p+YJ0TOGgvlbrPN?I*0ywS_lzy*yS<0-)X(y^wE2b9RE(L6aBMwRkx}X
z4j=3qYu~Pggpblh<D>f|zjJ1q*0g`L!)0^a+|P}t8oUu4AJvGu@Ky)?Ii80OHVDf3
z3%`Ki%otJ<X5Ly`nA*`<8f#y6_Dn%niu$E9w>~ak>0jg0=NC4!y`#6E3pGfEr_4}S
zB42n3&BTCrc-ivLzH_bBwW&vcNI;vKk3BIPT6f$Wg7$M%U%A6NlIKk|?p&P^@#UJ{
zJgOJ7Aus6_<b!EQka4y1nO*&G8Jyh}v?v=xfWGb;85b68d0g1Pvdin8mVbikTsxyP
z<Sp`RQaHrxl202MKEOPk{ZA3+6{jtDuho5BWIJ(Rq<*B18|$-^d1toi^IghpiM5S+
z0(&<6wHl@M=yGbr^$70#=Gtr6kMO~<GIsT?y9Uuf7BFR>kf9H<T7rOa)OrI*tZvFN
zKuB~c=mEfy1!PE71`CkA{;=oUW2)XZN13;m9M+(Q4dQ|$u0^mMNc!NI0Ayw_<sN*V
zk1G73tU~j>oMN;jk{XhLt&9@!%$y0AsVd-Tl{9S;I$MwB!Q-=oo%&PoTEb8npX3@v
z*{(7)k9;{$R2mzC3s(U|FpamImh1qo<^6bY^l+9Ik(w`YiR;Xzp*C4$<<_x7gxM=w
zj_$QmG0SwI6sk?@UYi$}Z3+~(Gj0BPM=4W8%+^f%O5d>~u8(mnGcdX$Nm%_Q-d3@L
zvNWI2z}2o!y?hUjrZ4?G87OS&ohVUc^hlGz)w!d0zG*H6EMs+3py@jvXp_^aWV@sQ
z^+~7vf0GA?ZEs8{i31U0NYb;?RJhkkMX>zgAFn+;t`#1*HN%uhSeBNo!8K{OZOM9R
z%`xhc++g={_vP)ea8}@T0bnGzDTP4ytsfWO{rlWN_=)&E{AREB&t9&dzDaF&%GMfL
zMpY^b4^Y&7B+$m|-Z0MVF5Tc^_`vYC4-C(P%gfT&J?nWaJ}RlvZlBRGU^r}a_2IL*
zIf8w8{6c{Oo*GSPVh^dQjAL!_<!PU%6mr?4?HU4$fFJ&@eg=CdwZxQH=Bul^8-!0Z
z7{4P~w&2I=&g|W?M56s~h3rBh_)=#H?t&tf+%lLC>7Ue<DhRvpNH{HxDccf8qzl|$
zf#-@CDnS!_Cb<?EA2Xk*eIE48!o52i#n`{Y!3}Z21%(n<rz8oj95v%Ua}FC_>**HT
z6)Z|BvJ4~a;a9+oHoHgysDWd_)t;?;b0VA-e(sbu`$q7sCms<!O>85dHUKT7vqR+Q
z<#tQo*4QlnDE!%w|C{Vp0hxXBG?SDVr1=x@-;Lq_4bAmG15fxzr3+4$4pf+=6!722
z*oS|GzTo@+Nxd`abpO9jmM-c8@xe*fe~AN>GNt)+BpS|a>_w6iJI|4jWhpENRyg@>
zrUqEX$S!2eD?ZzF`S=ZUbIM-N1$V*U7k)Sbo{s8_^9I~hA#`M1u{G;zV%aZopJ@g+
z^8MXII4vKEtS{Q&Sj6zC$BW^8Tka_^@;Z3>!+-8wD>=|kOOKWjQfLHN$vKvRo=4lC
zz+W4;bpP<)EC4y!l%2}xE!{iZ!^m-0c<cm5DphP_*}Qyf`Q&TIAg}so=R5s<X->mu
z&cD4RZW3Ld4zF`{*YXIAP~oOB(8BovweMfMPXwTalmF*g-+Yq&3f5yX+8nho<KjsO
zd(qaFrKV6@g$Y48Lxn)6v!|txy&|c}823Uu4l(iE>pa6-oiC}8^roys7k7V*l||%`
zn%}`U7vj%=Rg!eL$TGZr`?6_>hPn8mRpw>i1k^^>y#NaB=1`)+_Pdla*>&DT87gV~
znsML5rTy;kBDIK~z_$8LA2T~3(f<vDV~xcnKkV-nX2rbe3kE9lN7GPh>4yZ+ux9&T
zdfa3fx@lx&)tC4%AQJtnf`ZG)^$>s+*hQ^E=av{L|J$cgSTa{Dp$tpE=#M(sp8`T<
zqtfBYEipGJvTtMhVu?TMcRM%!nDl`FP=ioa3L(o{p33tAZJay_nQ@@)Lh}b3sB^fP
z;~ox5dJgalqzrflMB?g;0Cfq$(R7{43otxDOb!l4vj3alwT8FS4{q_CTkN$Pz?<!P
z6CgOeFd~D$me-&TKD=e>!B}}7^tU$bo5@Z;?VXfV*alC=$RDy@b9J}dntLnK9u31U
zzHI_@px<2|uc-lLh<x9S=zfcy*h!2bS}piT6A9hgLR9(0LSnJZq<`t`CN|y1!Zbp0
zluG&*i*vkX7oM2>Ot=5|JM^}Y7-&+p`b%z8ozjZ%0<jbZ4`BwE2qd>+RC4=3r%fI5
z&wQ@qB+2?DS`X(>@$N#~2@gq<SWiyK8rwt+Ds<o(PabaYvUv6Kc<<ViiDTHk!~W2E
zp1~Q{2Gk%Q&SK~y_VGW+CDx~`*RkuFf75EkEL&$O|6wc{9Ad8fP)%3Hts_48m+ts!
zcchkfIgL{smB44ewD=$LM49US7g%skRuTWH{yzC%#jVN{zhVQGl^81AIWdDumMeJ9
zmE2L{&#WSj`%8}FRXuLk3W-$v*frtl5`NJ+39y%EZ+>Gtpz|~iw^P#6dw-7pmfxPI
zlY!M3KT){eZY*km!^2W`kaucZhT|%0)U+PuNQKYM+@<ej5`#6pF{|G^l6OZ%T=pMr
zq>1^jiV$m?U5mpP!5;of-?Z1UXoC9PAEn_ipr9C6-bN;;B1x0#1itJAztoa^US@9W
zO!Hq?EQQK9va(RC6CbC<A$Rg3d~)yfOT<FwC3}@H=1;7P8@gr8CYOe?Z<R(8r`M*G
zCL0Ku4!i?IebXh1ptUC|P{!?DZHzI(`^o3yhJ)psro7k1O@%Y#jVj_kwUt`dPuPh|
zUKqErbN&#+Cj{VZx2J0JQx+=e0GRK)8L!V+kga8KczR)jDVy^VP9tXXRQk#7<>Uw?
zYh%26v`L(Z8Jj!L???Fo2s9XDHdhlazouCQm7y0_%v0LTQ&O7TCRg3=HS=C<%NyEM
z7;25(KNzH}a1nEOplWoCDxXo)+jz!n53y#R)QSrw#70j-UJ8?b=F~N2W@ooJq=Cda
z<7a?0kt5p9C3OD=Ogw^f2_KtGXa-W!WU~3kns<v`4Ny{*tTAJ7Np5+bo5>YfSqDEq
zT3;$sHK`cwQbRX6iuwC&aLjjqkQ(@bU@7R!NwM02`27}Uw%7;}Aoq&*ZyZ{O$^#9z
zhtn}FU>;V50#X)E$;(}NUsN)I(6>3w)~vIWVw|R8*8LB=uxchSl2>3lWk66qEJ?Tm
zFMgOB{X({DF`ip<;a2^<hn7&Z+m-~8db~X1yStAjavk-Z<$%-jz-~H@aP)J1t04QO
z^CP;^9>oxQYtXPuf=UW$9`LRW?xJqonl(T>t~W%AFuyfGhu=ofOOF~RL85#d2gEs4
z*nLnHx;80)zgab2gL$2W_63-dh9S(jq>`n|1P3{-CP1+fACn~FlggL>7kTg9)#Mhn
z{n{vs6m=`oK}0~BQl$$jy(kDsS839bDlJh!ut4Y_NK;DaNbdwJ^dh|zkP;w3fB+$c
zB=17q`#jGWXPgi49{+F*k(B$sSD9<hYyK{Y0o6v{K|h(Loj5t!sMj~zf%Oxh=&Q*t
z^Z!(kkBL~3RhmE06|K}4J(~ydhx|mk92wC=I06yV2Yg|5sKvp`-q}Go7@gu9D;Q_i
z!0^2l5Q~w$zacG4N`I2&^nTtqU#T9J@y@V!`;O^9&pra^V5?Gw>AIbMpSL#O)3zN<
zW@+U$aJ<-2_WrhGMDRWI7CsZYn<t9xAnR-WsJNa5<2gHJmJSh`#Ab@J?wQ$7ZTZD6
z>|N3rAh59NW2-NYSJy>XD5kcP1iZREQkB$1ijIoA2>vjRLQDfPE-&NIHal7!rJCt5
zIXdLJmNX-<K6uht!hT3ZzeLi;kenvdrxn?4ZE?1@1JZU=)yD6spGvaK&wUZEfYQL}
zXkY9VC4?M_o9BiZLx>ZRpw1w^k~n7{e`)U33qsA+`6nia9`|4u2sMg~taeDYr0I97
z3|^nRR-!j(_X_7z+P$xIRTG`FaceF>VEfo|$6bKkdgmVcDwJD-+Wyh;<{<=?mI+<g
zP>4*m0Y1s^;xI+sh8bDD+uwk_{DknmlwMqJK)9|}rx%;J(c<Q<vlRg#zgtUKGIPu_
z*N?l7w=9GO3g%9(AA`CHiFuWii1!p9k1W#!YuekJ=$dTwg(<B78~JB}2%zhAwY9!A
z)`z|zyTwRZ*Jx`(Ft@*DSOEJFSln4@Zp7nhY5`-!?5D%is#k9n`39A%1m8INqGRME
zS33t3LIX#9&4<+MYdhbTW;rrcWpOF7(hq%9O9YWxz>zTLs5!%RLzU4_z7dl8Im4jy
zM*Ne_)(xJ=yX?y!u>EH>$HvRs#Y1RyZ1+aSkL<j6_^wxgwEaOuyPrvvK>Wtyiqw?E
zhK;y4pu6t6r6Mmx)YQ@r17oNnh^!N5@9K}3x%U9;^O>b?Fv!<=*K{=4tCg%JS!LK0
z_Oy*gyF7Bnb?Pj$2N(}349y)=-tT~-SbPNadfumksQA#rwn75!VRvf`(JTHAj&(C>
zUR!HFHPn!;yJ%|SMkujssf`&Sh+H4Ec)4Yx@*UT`+K3IlZr5?PgGM>|m`g7vy*SA$
zr)rJEh`=b`9@*Hb(eAD_{$_WdCPd`P;r!ch|2v{NtbN8cx~VJ1t_*r2xcSD?v*_yW
zC0vibDD_ICU|D<um9y8BA_Y3DAy<!|Yz~4Sn3w-dWg+04K5FvHv8N^Y-eBB$*t<Gj
z*g!L?zO#T{P0qVVFxO=$vt(xSiid0Xc3i$T&e$Y(LvS8}_qsi?6~e9PxswAr$`g>*
zYzWf&7l@WLhk4|5X<-K;JEbgVyG!%q!Rp`V1dd<-TR{f`$+Y!8JZ;7nKIZdGY)7c{
zg=6TC;eDkefS6jswFX~n)$0-Owds41)G3|;m*BH|!9f`~z|1G~gtLlU?c?Bm^R4P0
zGcRL#Dvy9#PHPJ*Z}AdC)#hw>g@5pTpxfTLARAAxeLC~~M(zgt<_-HP=Vo4mIF*Q%
z-i$bPRMXmg<Tgo`$)fw&xMbr;gyzbRXA~;$EKc3PCx{w7)nORe(b1qeFz)`+rz^!H
zGxdx-aH7&&`UB001%5Bp`%9kmZEjGRDi(+$52v{g!}fub>d@F5Kd*}^s>Uh3J_P}C
zCi+G}XP1%(jHL{_pp3(~{vsEVngo7A`IBQeg|rJc?2aI|gUfZ-703P0$Cp3SHIGH|
z6ep!UL!eJ9FD0TK7I~PHKFv4KTC&$vUJaY8%+9Q~-zWYC&_;6;0?c$bYUWEv^+!G*
z<`?z;iQn3l7`9Iw49Ih>ht|EgfAc_Ye7^sb?9)4OE7g0%U0G(BPY`9FwOiSpp~sSy
z7XlW34JL<EARS3n*V^#wUPs-J(DXCRQkj(B=I%zmer74lJRv~zj)QH@mV6Ys6w;6=
z>1p}B`(pB~oK^h&<}-D^uPR~NjEnQ}r`wcq!e`BPyEdmRYCoBYEmCD`$=IFkTE6L<
z!1z5fC^c0ngiq+*$+IIEL0-}C3kGFtDS+cW1pz(N(8BMlm&F2hW-k~ky5H)5zx;$>
zY*kwE%N)*4T<P`Q=}UWATCzrHT0k{*>UHoNYuyp|c_>PqBy-kv*$G(s?f&wz8b0)S
z;Kfn7Rwq9~BeO54?~-gdrHl3L)I4@Nhjn7tlyOwcGD+r282_m6=)vmZk5i}^+*PI7
z{(98S(Kp{|*q<D$*=*vuXE7to=5KCOMD=eFLHC~ujCpk}<MYa^=mGt&hdmCPqTkMY
zH$CT|Xq`$dqfKdZV1ejLAH{?7fb}03M_fYKe-JAD&f;Irl3zgd-l?D7G24F%(6+bV
z#Z+4C2Jk)htF!4Q*wKoZU8a`Ux$(ANE`Zb!<5&j$HPSrk0``>=SfNO7`{9llQ~hw<
zCpVZzV;tQxij-}^GZ3#dt8tq;eH{iNF={4a6CsALZI9LdD8}@UEvxf-$j`vK_xT;g
z`ERq<myc0*sS9$R+bh$?-20er2cRW*)F-T;aY`r55#Mf75_%}|@aV=NlH_{)nK<J<
z=3K-Lp0Sn|d3#W_CNapkt}73|b)v>+=US1s+n=|aWn1rZ_9IZI!|u~r9=@gW{Kols
zgJJgnjm6W#n&7A2;SaYeDJ|U1_;-YY{?Q=6eMOSp+>L?*j9+6KHOp!trQn5b4chLz
zjbU0D-KYFzSXz3ysq=n|t@bL%Fybz-&aXNsR|9&8wD`lsN1u^y&M7*o45Ga|72>qL
zr)^!tD{I8}OmwQ37E0OV9(FZPs~eML8$0R%A=OH19t{`fw_l0t)$#T22#zxOt;Z^5
zJ{TStjVs0PNUOiEs1EWKU1wKS3cN&{c$pnIij?`?r`X1Zb7oDtD+NdS`d;eP9(+fz
z->&~2<1QrP(X(@pps&j?P}FdATb|NVZ~)*S4W&}xISZ-Ge$KDb-1<vRBfUD-^LV!g
zIt_C@Sl^aeQAr`c#OP6V9Q-jpI+l=kE#ceXV6xkNJBwsSHZK{yFcK9Hxqp-Sym%$<
zLrh9AKWoB`qEBLq8ODgc4gkMJh$x!4lIi={0H`+3VOG;X|BhGKvX3UMI!jH3QhPC$
zD^lXFTWgeG&Qv~+bX#7*Q8U`1gVg5Cebj@7_nM~Jc4GU4=Jf|jCT6kgMJJ_6Lop+E
zjVh4)o&2dO+l3@fYgM-B@pg+P99xC1HEz*bkYQ@yS7)zrN3dlG+JB=8Qu_uMeg*{>
zYbzL-q_MC)Rf5H?P0}fk#E|pyH!>55Qn7I1J5s0|`ya%@WaMdKF!d@FQ9wMSn<@9;
zMt@3miyB>NTHoB^`?+Tfw$@>B(VXB$?ONPNEwg9+4_c5xdiAc?H4zLOwz#RxHm~5{
z5p)}@af8QIrUxI;*XE-jL$6ecmPA*gZ{LZiwC@41oCgY-xDQX4)UZLa(#KE#C@eT>
zg~E>zLG_h<iSDDXY66lfUEtdOL%-hA``#z*E0M6JiLNupP_<4<?W_1GUMDp}O?RCe
zeMe+;QgPQo{p`F;C9o?%eR)2y!N$1D%ATg}I5sM5$RJvYi|p}x{>f^(TV*fB_o{#l
zWcT|i-!R55lJ`S0-v0cWov0fd-5s_Dtx`o4y_Njl^+vK2BrWhgI2SKMvsM)x#-Wqx
zbiJ-&2S>fP2p5N=bqjrnUb2bTlKy?;8tIis^ewa$PNAG9Xej%uf{3S8y<b#}Yz4i-
zH#R*m?yD0j;G$+45!tG(9$ZYU^c?<o%k>0X?s0k4JXDYbXABu4um50;vNtJFIaDbG
zt?z*!6(aS*lQc2F&(f5kLOjA{#RI-OG!E19og@K-+7pT7&^F++ym|$GuJ)JfM`@X&
zxnE}HRh-tl0qs^2AK(&C!x}Aj^o&HNV{;4pLc<(7p+kekcWzOAvJl}B>XW5TlBG}n
z%+g2<U^!yShkZ&>@g595@P7Wy&$#{qsmZdo+E7MBo4aXa_;SK~ShLp$)jFStSDr-&
z0VfW>#%`$PZeTaTbVoTNzr-DR&c5S-n=CajSg{IGS%fc}2??Q-SKjWyaZ2sUo$qUn
z`ccmNztm9-olk!1MpB+@VvY3<Uuw_^Ed61(0UrR<ZwxKqIG#WBcly~=D_{9Id_$6?
z$#AO6-nyJci9?Z@pW4JE!aCR?bq98*3Flb0i8&MTQQa`up|CCC;*R|hBs=A?rK77(
zg6ywT(h=<j(8)6}iS)#e<tsapH>i^obW_E|K=3fDmB$)Vm_qMcm?q92(Q$38fW-y9
zO8qn_%CIiULtvb6&HZp25;fzXuHJ>EQ(on=u|wMalaU|;|0<_R^KwC=3_*i}ttYnm
zHh%0iDO0Ter>B3pSv(gtV#?I5pcuPIL1xuJ(=t=u%dseKw`zpwh-81}TsUZ)#rP7?
zm%q4Kh`b-fTS&O&?A#y+N~~GUAiG(E@w*0V0XjXgs{|&Bx}n;-SdRz8jVvh3da9(d
z;w1J;#g;9fL4Bl~P)TK6Vs7upM~}9vKYI7R7WuAK*q-m>l9Nue)X?kqe6d_3v*mGW
z#k`07-bANKmBzoD<rz>lJV)ttkbO=(0FS+6OXE)?eh@Ts#+TNZS77@VD|7ix!Qa%N
zl(6MOp(320PW}^Di^ei}!GDDvrhOgO>$eFzY9mnk(1gKnR@au~%TOzhQvp}HTy;NV
z3Ib(@J&44C9p-%viH8Ei5ZRe8x7>VmB!^0UJ+$p-a6S+Hv^)LET5htc=l~-K*8St%
zM}Bn%Wi4Yn?0B8uA_qTYpNaeas72vir8|ROUb3O@*BPNvyLofZOy?QUb=etTJw1NZ
z8hnj_3C1H%I2(jAey#oL*~f;h2Bv~6*;MJk=73LR@(;y4^^TAAE$R;zA)E`F&hq3r
zS|-b>*!j%{`|5=kjymk(L4hkknqe)XCR82WSOnu?-6*vTB~@On_m+W>_>SOaW7?^(
zw=)eKBi3u6l`Ow3fLXSG*vCSiX1g5mf^YoclR9ym_O7tF+S#}-J_A365F1@I-~U-m
zVsd$qJ@~c$u@#4;&E&=*#6n38j+ICBzF#OZ;v%OGl}6$bpDcH%(My1mURq~t=^<&m
z>!lU{4X)eZHGs_uC03xWHTY?>ce`}Xh2^eQo01NpGx%ic4_^K~4=H$<>xjQ)571h#
zFZ_hw2Rc7$8^Buh_1uZCSjpX)YplNvi{@$TQ?`PX&`8~&Lrmh_N@_0T;b{K6P^CzI
zF%I1qJ}>aR*ihJUdMIff^vwEeFGBUNo%dilVs#ul2AB|U%^qadSPZbtSrxuaw0IYk
zu<qj_%1U{qJfTUp%<Zwn8|gQpoIQ_*Tze6k<equOOy6kHMBwV!e~;T8hbXE9T&3?c
zojH&`(*fq2Ui1ZDo=ljW2?|L)>O7PrnP`5#Sr6mPG88Iu|6su{Wxd)1xG;e4^z@66
zx$xe*GrqhYg2wM>`tD8^u{~RVbda2XmMv-au?cgk&tKJoISoX_1^`<eMmB5jMXUiO
z>XW6^E{kLYkDqH2vd0A2H3~cS_B^$P=rt}sF%IstDtv4{lTvH)Bn_}pExNxXJPQ29
z{Dtq_m>SBxX_VpCU|kQC*K5tH-!-=)x!x-1PO2tC&f?Z~uP+&C@2jV80#SIg^a<FQ
z`+ho;A87Hg*0u^~tG8MfvskR<Ta?)lg01z_9BGWW_}HM!%OEl99R*Z2@ek=K7MI;S
z{LZH6qs-u8g2^g#&L20K05P!GrL#Tsv=5M(6uVO{Ed_`U*8WG%C9H3lNB%U<11p}`
z!7pKVnChgGNIj-fg&ETl_A|@?HLNWy{7x)T51Gy$PZezVGyD0SP<DmbMJZLLZ=Yl&
zulVdvCeZtrCGSH(hrHarmx>`UM3h%{s_WmwzYz~+cZ|RQe0zbgpZGG@#wAsoeL_Zy
z!8&ei%0lp#z7fNKzC=-3%Epn2nb!E_%d81ocs4|}5hJkvIh5l!4OLtwUY5jvj$Cgk
z5wV_>b^Sa$`<hyLRkX<eYh!Xn?=a}UTzb=eDHRTjSS#{z6m^A%nNU&!B9y1PBur0d
zLTuXK?kKT7jnEIhn0whje%I76<BDsq*$@7{8Q0FWZ(nBgRtQ*tMrko0eR(ghStax<
zwlSyKIKz`9n`gzrh9h&9l3VajdqlaQsN@Zyrp0td#bk7;f{}r|RO6L;`^8VcB+7rI
z_``OC4rad99zGd~t>Y^h?$i_)$Hl8Uu83UbvvqkC52?k+7)UjpQQ>IsBDlaBC(A~(
zQ`L3<R37dh2TA|il%vhmFR#q~Y@DVQ`0E<*+s)`06G)tUf!sx?6uC`fbt>rh{5jSP
zkPKIkah?Pv&Pzwk{maY}9xh6M{%=+Qlo*diy#iI(q_kLEH48+f{XCZRx8#)wh}z)e
zf}(Cm|6c4GudrZrvUc-`WLFHt;@8;0Ju;b}$vWokt(9i{dBj>*hXP&9*#YaQb}=>V
z?JIL@3r=GYTHI+$Z?_EV3`P-ZdFmHGgBiF!Ckn|jj9>Y6`!4mbM#3uRGN&9H^$zFH
z*;%Z4L7l_!>gqvhUBN`r7BT24refF}F=A3Vaxi4F2)#a#3yuEPL&1a?#87b(7RM(q
zkpFVSnsETia*Pw)A<nLiyJC@9BBM5*B(<*$Jbx~AlX!&vQ3GU@_MKg|s+p$fEnXX@
zx^v-v4>gf5LHEb?c`S>&8E*8iYlbk$4gY!N(cfo)QuV)o-Tw5qLh<`2eE<Bd#Qgr(
z>;DHk4$6o3|7ZE|pZ`@pRQ_MMB)<Q<KCFA&S3R`EB38hNqifraT#jj>kT0G--Pq@l
zG=^2gD+H9ZUGy@B{cDWyvJ1v(eF;>#jz~O3gn6g-jfch>MkSgy-nzv9j4@K|PH~kp
zgvc3zQ<0N7hn=&5O5;v`qWbu*;=e?yW2g&C?ScS33tZV5O>u2Zun<;=N3+_$%&lX+
z(m_$+RFx(G;@+w*_mtNJzpIYe66>AbH-)KThki4ARDU0{+Xg7WgM+9U0FRXdNYFs9
zHMSu&@&D^r&i^+MnkhWy0N0g-yn`pFA)?UaJ~G-X2u)svqWk<oRjU@f@l$T<ai0Gk
z`%9W7WeP4M@~A7M&FAnl+2YpdWApjD-bkI!<>^V$$9pSvaU3wU-SxB~9pbL2xXzxn
zufyF?s*;cMb$WF|cXY4W;H{=DyWC9@=X-eaNP5wqScoK_gT<v(I0#cx8>ZvtWk~C<
z&+eMXzVa_GBlM8J+}X`sUwS^p6cqzy&1SJ-3asTcl@7X8tW<c=AKG1+66;1ltKLN2
z1VEB7j!V{u>c%el*mG%<@4^kFvPL#X%L5gjH3{_TP^})&l0}6asX8WmE~dP<=o%C$
zlfXXg%wBS(>%I#q&kZarKqkFlPMdE!E|OYIWb%2MTn5ALuUo}hbULsR5(k=0^^sW8
z9*!q(g?>jmcLJZA#E7awyg<VTMgwh3TG>NUZJM$d3fyybBc+j}*0ykbJZw19dfFfU
z^n>CA7SCu*swY=x@!Ybo++iK22il@0L_`z3%(kS^GeK7t$uVD`At1D9QXZcCT#yA#
zRbauym}MV3(y-&Oo)|B=X9w9C4dgNHP%m4VFPz%mL%$M)Y%BhWK#m<-aDM{eBw1GF
zSeZcnfV)|D7|KX(S&S9$CppUF(mJ1NxV6$GhbMk}d3WkOZPSpHwcK)??jsGKocWV)
z-?j}iKU<b-*Gt@mJ~}>BlU-5>krKGhAR_h%x2|>x+BrrVhA=TBqx>{BGq+Gd;W=rO
zdpmCX*X9q*8-+75I){~}Hta^s1e3Is_U+6Yo6N+X&wCHj)9v2cle~ZJYRmSv_GZNq
zk$k#N)#m9Hm<dqNnFwuQ-x+lV#j5ei{dw#%X69{*qW=T@3ujqNhO$zzR11p_^IeP~
z6@7LdrGrjV_=<W?>&^Rj)(-rb0(eYIqcwTT&4yOIWF>+7(WR_kMS~(G01@;N(6Exp
z75Tkq|4xz&YQ}5Sd2lo5^#GXVj^FD}YUPoc*WA^V?7f`_{fzdfT4Gxnr4Mfie6!tw
z(VfRb%Tv?ZlTWAYYT5*-dn>+9U5cn9rAmn&Tst4Ig>#;r%M!$-sLAiYo7f<wNPKqM
zYbFQtEY>(62;5v8P0>V{Du8{y)BC})=*hY^WPfLsqo{d!ws1{0{N+0h+7XF8>ub~s
zyYV&wGZRe%`xU(J$<!yA`=h;IC2;o7>L(H0{z)?pWL@=1;Js!e<UZ37s6WY8<kCK8
z=U)#P^R9PC5C_mNS?ora&?dcFa)yM%E0Cc=y><nK^te&QTr4htOAEf*!0M`_$#cu~
zbY@=+teoc<{;-j?Ei0nJ!x~I^eL~KF<qFC317l+a$SYmDk#u8X{bMSGy*<2cqmq+2
z>04WZ%=!cij<Q5{eSpF``x32@%sP|%epd(r<=ak@*Jy9)KplGfyVR5LO7T4UlUv;0
z*d^e%n^#blC7t{b^IpTFrFVI!tenB(0-K!6q|P0-K4+*b2^TD0fWtn!lTYten^&y*
z1tDbr?M!mh)SDi3T5DCqUQ;^Te_m)c^*5B-XW62R-&>U2w^ve=&0r97@A?wNlAVsM
zj;J_&GO2oTmxFHb9;E^z@dv#+TFM%otyqT&Gakn&N{9v(OcSo4u`F-(1tm!@7CF$7
zdJ%6UY}IM*uJ;`rr(s*uPSU%XvW@j7C_H-AJ&;Kuy~ashPu3|)7Y|?MyBzPO-3oEE
z<X^85O>)c&zPUQ4PkC^$EYhJ?qKS>8c{*(eUb!}VOE*F5XTc7i@a5q+Z<rP+S(F|Z
zy5|h{O&cP$omw&-&>fI2bZ+0LCwUJI*fZ#%yfF1Ly4T_vpUZ$W@UEux9J9n?H(Z!t
z)!7!;53e*R*D*b|`FMn03m!mz$kwI_oiJ#;JD?J{m`AQYXRND+J9yOSCDZt)Gz`i{
z55d^Bmq}tvP|iMuhM_pbtMT8m)@;vaoR;0?MtJyf8Yt$TlYM%VYOuDriYbuM3=<T^
z#LObP;fza`^B+}e9!|XcMl>_2xMOoN94Rj!^TWT#OH*CxeMtYyY5mCtw5vzH{F{VJ
zkGrgk7bR*XVtHw$H|B@oWeG1_zV57frtlWnPE1xD%S!NuedSz;)FUoK$8v<xiUh>g
zvr4atpL7Gdm&ZU~Jky%pRPb?oaM?K)+*HQ+p(ju+{AICw1wC4?Cqzw3xl9oU>2?8@
zP5I&r>+enUI$zgh)aYDrRFoj6iD3Rt<zt^OpHAYr`N#*DX*jxwxjZyrDbPn`iyb~5
z;gB8)e*D-ZW47dsa%WoiK8j21(Pi$N43%uGj@xf|C%4RMYayC9_|EFx3y#Gb>)&ZT
z^fq}Z6nilwpm7~Ka<MHioO07-`CJVh;uG06lEM43S2B2_ZY8cV5;b<k{^DOMrPdX%
z&PG8BvC~_gg&e_$Ga~WhjYGJYs%3F%d^L~so+4UDMWJUZ{YyZ1Cu;6t9l)Bb!zt;V
zS5~~CtsMMJxF*#L%~~J+1MB{jnE8D<jNSv0<mj7c2QGNWO{Q*}M~0WBN@jUi@S1WY
zSi3lwSdhU^Szk@`DgCit?@p=I-iiXP$5T;@s}&F1RIT@2j5}KhH;47C83t0?>S<hB
z7%$vSiZds36OwJI6k;S+f%#pnkRx`r25!-jE3@&5g_JAYXib5Vu_?CU@w#qkDuTP|
zERU^5mFqPkrz}>QGzfHqGdaZKS`t&{Qp^aNk7Mby@;hSMoys`&_D5a@d%B{XQ98j<
z^Beuk#oP8vq!8bZz|!r{(rv<nl5i@nZ_TcR<@;hLV#5DX<l7g&p5h`OBEoule}&&G
zG2#p3%rfP?(v&$R@!~vs=k5sab%1rZm!1Uh|1fo?s-Vm2nz}(z#i*N-Zw6K*oDGns
zQT?)poFew>G`&A49+Y{y3H3eY7#tCUqR}{2_I%GT%a^3J3DaMLqJ&p?jTp-Es-y+_
z)-p;=DA{#7C(=*1w1aO<x58(nAYaazrHPL_7ZP_tjOdOoJ(b?p5czNH95wm+w6a41
z8C`9w>hbj`$_eAN=Hlj(O;y<1xtqX&R|;=B?z^`gAqX8yn9dZ@Bby|KrY+qRy8kni
zbD>lw&DJf?(nZ0K$HP_1+7YGZ;E=vLV}GX46kH8OM97+_$zbR~V;On+l|T2kAH<+a
zl78V4N3~R^i2VHl_08Of`AWHhKrK^|!^zMC^5eU+UQB_BY;UywTp9k$OHS(yNClS%
zLhcLu)fyoy#E+Es-5Hj#Xo=2LT`?T*A;bvPo??~Q-)O9)=9LzVdoeLee5Mq-V;5Lr
z{r>XRzjNuI4G_yo$PYZwtz~ZDK7#DkZ=Q>K_5E01wBU@hcL0448$Lbz)^x(?45oro
zhy~i<;0?1}FXV0@O{!u2fk#imKVIrI|Lg$f1ysap#&xf2OhfNwDZo(a^v!QQ%?8u$
zXW2JL2UfM+uqE7eF{<+jJVJ)r*PHO6$^wmC%TvD)<eE@{>|qNDKtYRx2aGZH;+W?>
z93kSX{<eM-V;PU4mD;cT9bH~oj_)V(<*?Q1#M`$&R9%RiWcAqJXqz>Uu6_vlcX{{+
z^Ob_D&rIVb@ACJ(8Scb$H9wWgmfADV8CP2>?0JD$ysxBV`5nQuL+i`jqKfiu^{@Sk
z&tC;uoZ`}ce1(VM8JjEkNLcUuN_0}%)`*kivCpxN4j(*>6MkfkE!z8KeCK}=_kGbM
zZB8MlXmn4?|1M^DQnWlZV0{201%No=@9HPgrOU>%&Q<<%DP%||$Z<$StpAQ6;7{FE
z@ES`8p-J702W#rom9tM1^g3@v`l2#T&PUdR#ZM9k7ll5}-o3&1ud<U5sJ{g5kF#T|
z9re=268!Xju1=<PpFKcpuiujy`qsMz;O;CiwqMuAqW<wLtx9UbtSCjmtSPUpMY~b}
z;52eC_zn}QU+Z4pi0S(c6r3V&pOEOM?Yv~R$4i<0bwy<O`~|+}XV`h3*CyE2MB1Z&
zkc61uOueO-20gMHoLRdw%>qEHjMmnj>`fam6lpxI+<NLAE{EtCd;YFrO7Vkp_#qpp
zmfIK0wCspbxuP!8TF@GSe3rFOaTCWStzD7e;h-+N@ff<1@9ca)|N5}v`bOk`k4i>~
zK~z)aWgFepS-F(>-As`l+)alF5%>KkhM}502wqC5HP}v588D{mc#i$yl%_(wyJs}h
zzWnh^a3hI81Fa|OB~2vE&-&=HxHsgD7$A&l89jfst@(F8A0Be0-(8va<tChN`S8fE
z^Ukh(W6uE5e*D0i?Hk-V$=}_yI5+T2P{0pLuJ}UizP;`x@&ifKItW9X)m9}8P}SZQ
zLR}YP4Ub{fmV(ylnxYmX=MEw!UKK_tM(L%?xInFqrkVhVB&G(O;@rRzq<Ltnd4peJ
zIc~2W9~`wRx27F-T*V7^$w%ianzVk`8}$+Yd3!q^+9XS_ywNKcRl1tL2Hi$Zt`6FJ
z2vdJwjSe0`%Gt{FJG!G6{qt4^Jxmh6DejrTzO67{jMh!vYooK>>-wp9$$C^_!?iBp
zq0KN`;M88dabU!V_{|(_xq0upV0#T|qtQr|Bn538s$dF-1PJyWg2c9|J9Qb_o#F2c
zndPER;t^{ejQWO$i_R9sHoRaPx4+}!z_LAo4v!=^RgbZ)31zlTB)&>FTznyG^IiYf
zb7Ks}Lru6+LPEc1fA-G95uGakZ5_35i*l>S7BwdPSIvs;(Z26tQI5t#d$!&%(k~z7
zv}rIu5)W8}loqJ;h-1uP@p3t*q_)0a&Deob6I!&C%5qT=^gb7I7RF`I6u^ZpRWIiv
z=+7<REx!x<w#&TO2X`T4cIEA&ZldocY-4d-qw5Qe;YqixyG8dksvPnKwbZMt+&r^Q
ze$6!6F9=ZmtA7TEZ)wUG3)nP4-H7(h)K_0T5T2%?(4>S#wzkKU_U14}!S?;wvZ)=_
z>*lq7=AAz2o%@HOtdQMNn&oI!fml(8s&Nr1j9h$DeDIR^XZu0n99E;5bGLVc4U$S>
zpFA@KGr<9Gnc@k~k0oTN5c`Nu<<kL5CB4mR6wes!DDz^P9I;d?UXmxFA-w5px&+A!
zncOKRvy5-|mRuZqGww4w`#^A1JU_MN=mz0gcr~{L)d+!+Hd}o+G~)KAKOY&!!!B7W
z#D8Le+^BcG9h#;UiXThC^aktuWj!&K{sbk@y?Kb-P!~$hak_4f>5Ek3858kZSFyvH
z<Xl+4%C^`Grz2d*g__0Zr)~@ZWn&f2U1=mzRCHm5aXcA1Y|)#9jZbqieT1oNkQh8r
z7~@m<*8}hNk==O41yvf+1&T7(pr%_VRm+xjxl{0ugHJ0fMBQt{VSZ~~!C0z0O3BIG
zxAdK0^@1a<K1jNsv-`$8z3(?k7Hru{QDiHCU~<U++6!j@E$fMw1%uqi4WnmYnB(z*
zYeSj0Os{=;0ZfVph1IM~EeQI3uwu8xzr!?R^yme#p=*5#n@G+5%<i=iOu=`{%&q5!
z*CCI$M!u)jDR;nt<=rdN4FN*<<G<;!muW}oGmZ7>H*)^9oQ|@pdIXb16^7gc!^HW>
z0=>W&jZEM5G^M#`U!MrDi0@Y|v3_o&OQ$%R&7QJ37&@fL);^(ikMa=x+y;2dBJj{i
zhk$5ewMf=l(WJ;c^uKR6D|jR`&fXj0;7_arxWoO>Atv=v|MW!bTYJnE!r8L1Wd>M;
z)nrn5;UowVPX@L9y(~ACqstn(6-|t%@Fi;A3#I(_AkwJVB5`~AUAVST$A%r$d}qe~
zjsu60&GUe}p_myuRr!3I$6o}$A+)--dl{jNm;K9sCrpUvk|wo`|8;1%f9t;-4fv@4
z3-W-%+44!Xe%i&kHnweK>e@9O@jg`iTY2(@1p_36+X_dOwU9jwsg>@ky%!lE8&mlf
zt|yGg*N9(1OORdv-t9VN&@&Q8t$Znh)4CLfRM-oS9Qj&%Pb5HpaqjWeBJ)7Ou5==T
zQQH-xgD~Tlk6N|(M*h1_?vp@|>L7}TX;0QNYzcNqKPV2)mPliNA0-%3H#RVO5kWwc
z1Jy4--ZldAesK`k9aL~$6banxDz816mwu3b^0hv$%#!fY*Jm@nFnvdX?-7_QQ!^CT
zMOG1rsLpS2QP1i%At}O86HdFUiR|;|J9E-0sqM?0JufK<vNLzRm~QHU$cpqYvitM7
z5W!y3T4axhZ{Ol^Do!d7*VV+)k9OkM`@=&D+oxJ6m`)qZMiuoq8>jnDb$|5pt>p@z
z|0Afn4IC$Lz)+nK)UP*%qeS}HHLWVG_*;9Xt&CN?L@#LE>MJr6Ng1%6y`Kk1(dIde
zsP87ei?i$f)&z~NRPFIFPQH@iUmPeOQ;9bc^|QM>;wRx{U$U-Ez?HR?wK5R3c9AN{
zizz0dNXVAr-}(3p^u{kgHgv(KzMqF?MY>X}T#r9Ev|kfdx%(knTd`X-aIzLC-JdZ^
zQQJoZDKHY)H|UjM9xl8uYO=yGx@!_1XF>b{4Y3AgcKLLwcYkiu?2$5e6G84v-%iJp
zdGyKiP(NUkx$$;TKB*t$Tj!^(k=ODDtThFr=)_S(lY_U~o_43pFaOD)-Es({9f;=p
z!Kir6x$qjO%wkSs7LAIVVuY5S{`uBNwFK0dBH1{WmrUZ=7CA}QCK28z{tphdxL7Bb
z^Y>pz!$XEk)jH5lDYSvC%XIO{NFV!?x8=&<ioksY+0(C_)V{SthKor)#Y5ZABnZ>}
zf3Bcr2V{5kft0_5?A=8p*LuW`W(4f$!?{SO29tyz+~jX>CMMf@yhu$~C8VQP>3eiK
zKGp7vgkYjHUW_h)1>z%(C@D)&iWvRDNTb?jKT)a-cxJnIo`qXl<ABCmzCq~Zj6!ZU
z*gx1|?Nc1QTl($!n6(md9NcUh;jvp6H|&!yK&7X9d28IyBJSW-p|%{@aeA6vR$5Oo
z6}V{jkeVh&484wWw_%gMqWmCj+BSU&_%cU8YJE|8-C1h|z!I;bxc43!FEGv5wfsG1
zZHZRK(>Kk0*!vL2nv5JlWq&Si2^}&KZg9LUX(l3k%dp$aKA*$6{a4qG^<}WA`Oe@F
z2LeByr8ucJjmo8aaKI+2?Sn)*G)IrGmZ+8$M-qE)?{7Rx#iTXF{HO_hphZ$QlR54A
z!)&l`Qq&Gk<q93ZnwX{;A|3>uUwDnyi+DdSsq-*Nj6c(-NR(G?)D>rfg?=}V=W?EW
zfIoD*w_fhmNJ;5s@z$x!ilQhsVb;glEB;yoPSwO-IUak!EC}s>(2FL7b)v<D7T+Mq
z`s#}hKPa3})H%7mDpP}WjOk+4ht&wGziKYIKCSg5PM&E+B-XduW1BeReYlb5BTNY~
ziJzC7{4<Q2U*BWcP8r~4i#-u_oaobN@o5!f(~YqX=Xm`vaW~ymmaAvEqXuSY<I_+J
zSwkY&+NZu{WQx12B-%O1nn$F|8OJ{;(64iTJDIMCY`@Pyt^FuQ=HWVwI3MSnxPgcc
z47`opv$nkfVZ+oQ)5M{-KJ$!ueJ(co<(lg}os=-`N@6s<>J!AWTHS^TQTnpCSk*wv
z?22}f?)AtSykJb=7t+2LY)psxdZ9QGNSf+|adtL}dv)U0KKZPa_D>V`1D)C-l}{-e
zClZfxV-?lov7FX-=ujVDh1DQw5(ZM$$KP2uSx|CCLjAcG6Od(z5T@jb_d-@C*qs=;
z_gUl1_&oB>q@>~!kbM4?zQr>Eo`jH+ngOK|2EO1?iEtEB9Q6I2=Y?3v(C$-!7*wV*
ze4xvVDK&ED&exsj8qYFaw%@lM$GPB~H14RYDpLecL$???`}W2AR+np-j@twW!iEpb
zEuQ|FSY}QjAm<|Xs>(0s(==L?V%`*XP3S_}=K3(hh$}~FFR$yNgC(Jvyz?qDIpJ|~
z@=>)JCw(_x3vQ&SeWRC=a8NB%*lnh7J>)B?o|j#}kcTWjNX2Fo2FrH&4<VVPdvv?)
zU5Pg{)f$?$5NC;ns!t#{3EW5>YvSl+PoM#mgHRN>09zR&PreDK%|~9gjQ{p{DEWYt
zAf4u@moyo*sJc2vU~livflk3`^Eiq+(VV$I;k|a<xl)7bME41r+i@EW;J8slsFS!r
znWa&hX%i#cB7c;IMdlUvr3VT#r|(&B`ML?Scew|oaGS%hDK^j%a#Ns7rHZ+Os$bTu
zfHB&JLA1@QKQr{B{X$iEqkNQKl#|!Qz9*-nSfrIIa@{DnIs0JdFpZU}r8QIo&5@J+
zi;Bl3!nN(x<hx{8<xwDB$#T#Zv<AtqW+iVlT;ya*(V~uGyvgSHqNo*~lfAay7qbSf
z>U^a;sV8ydqF3&-K6xhCD^@gfFYi#!_o~DCv{fvEioLfg#}Omb-RZ`VyZXLQ%FZCP
zf;Ev8p?5ETRZ7cGpHNR8X)+w_$*_w%Z4E#5#>@B@8RkPOe9|6K0#a6@0)4jiYCh>k
z-;PMf^s0I^1F=5rk|rgt%C{xuLWSX0>#G*?CQEl8DWNpF)xIBKL=LyJEwAgdxL;0j
z^iqKNT>47I?HMJj_OW+u7OQVOWfspQcD)89XOiP8)!b5}{}xZdGEGNB9YTFE5YJDe
zgJ@tneBTmOTKZwA)DBzT8e*@S;|{3Vb411<ne2n^y=|ec><vfsHd26`pGTS0y3dGW
zd}T<6QemTku~u^H=8oDWj6@_zezBy1!@FPYxeIId?IW80L%m=e+goa;m%l^km@)LS
z^XvP?O(g{s?!kTOAmQUWCcHK5>l0Ib-;s*MD~>HPh*pX!(S=)mD?`umuio}w9CfXu
zJoR-~FF49!Wuq?F;~*DLG!;`9c@**d20C(EQ%!fG?Qtqq*?tGPn~#s$elOE9O(^GX
zPSLs7i{mx(f|Arp%r9l+?J2i+{1kKD`P>h*(l*eEdM2~2bQRi8yt>tn6ww;cCrp6Y
zTXg80yRC3piP>L>H7H8v7P=@C!=#SNwtB(@n7vO8U2_AN<l-1OqCf&??e~Ki`)c)8
z$Wn`0OC19;*}dqVkg_XpeUI@!D(>3s2B)lI&Gq@9bhv&e=ilk7aWq}!dmKg{7QWIM
z@v=P;wys}HF<2UJ5_EBHxx+L%LwW42B5>S*+k-QLe`><><Dy+h`|chOo-MvqtIhqZ
zKVQA4oY7vSoy^&<2VY^tn%j{g)W3ja^o7sW_dZ0goQYuNCtn0?WYaEO=yq(hFwx-M
zhjYG+zDdlMPWbTHOUE81ZT`Y+Ba_ek<cT@I6{U8U63xp2@0T*IX~z#LnZj$cAp&^O
z5X3Y|xcW*^)EuNjSAKls#fFl{^t0Mo{#$@GaV%IJv%YBl0S>|_Rr0sXZCKLQ_fw*o
zJMCMxP4<13VBCjHcAQbMF*DNd%{j#t>^x3iykJvo+)BJdsN$N3yu2k3x&Rpl0uZtO
z5spb0L^HDp?5Zm6Qc2<;qRHJmE2T;s|6CiJm<n5t`)Zp%qQi0d&q=J=CxUZ9ddDa-
zqym25nKm>eT~$>s8MajKO=Z1LYOkW}b4JWdN>V?ek#)3mC2-X4*Imyf;X3<_7HcwR
zMy#Kp0{z1(_3i3U=eLG4OYA&@FUdSt;<~Wrer9jKr2!#yPbM^0wk48Lzy%Opvj?W*
z3i$_mJ!BkZ=2{rbycgW(e&4k`rFPa)8)z%09pleA%og^hbp;f2_V@{G&bH*%Os2e-
zf3U8uozyFzgQLa~=P1`gxw&td4m^1(?RAvQ2d(-qoFe$%rxlc6ztin4+J&_JIb<re
z!FuZy*Dl|b>5L4>_e^WD^GqDqSCOWWEb=%fRVD+7p#?`bY+fugjBugY_XB15Kz;3=
zGcLq4r(0tz>_*8yP{luEt@#Jok68~Kk*CM4|AJ7TsswyM7^n8b{cRn(zxVFZCFq<y
zKUb0Hu)-2$cR{nvYk`5YpV>>AR3l_6K%<wMobX$XA;v(mFkuiCqPC%67?4TVZsrrB
zUSYU#rq9C@3hq@8#^=i|udKF}UvhS3IfNm$r2})Du8*qxE%c)=Z~Hwih(bTr9?eSx
z*<aCWcg_*V9TVe~4qtoCbc$14J&+pMBOO^3$V`<F+HP-sM8`3IIb9vk`cDN92NLX{
zI+LffD45NIqMmBM%}*Qv!CG-A>8*Z+aDudCTYo@a+eaztJWoi;3TL2+K4s!?Ww_$Z
zCV-*pz()Nt_?|4(Go8*$Sd;rBAb8mY%KXy13rI4zN3y|DkfOVGN}u#PwQGfKbdFE`
zf}0$*N^^zJ{g24i$1B%3A1FSG+43JXg}^=S?k-hbfD|<GzE?~v;(RnPF!Rda@OX%F
zkU4+q>=I7)k<5l?@AI_OA$n{4;z_12(1H@d%;6f`Ztc<f<#&`EMYNA}Th-d1$n&ce
z{syutmjiu}wKU2(6q!!;sBKnCF8yGyR42`@+D0125T|_S-p+Cs(!&gr=q9jXxiaWi
zw|uq4J`Z06#@5JqW2<~m2u3F@f%skjA}Nh9(6VL1ojQ*kWNGS?gm{Ym83ac_3DnS0
zY!4+?o4$ReMy&nKN}b3#dsbDs0-;~be^AUX7HF!a>Z6)EOYtmatz$ccZLDAaBzpzy
zda3HCm42CdnzE`|Awv499(lq{tVoPdc!JuTqTVE@V;Jn!vg@;$N4?mo=+?JuO{q?&
zA#odhouKg4C&9^3Z1jksY`yN04%&TxE+TyVRRowsF1lW0&o@PMK;rqm{CQE6D(OXg
zM^hwDt^LC}H?~t;CwDVxlL$W`w`Z&B+B_wpaOo-!(endKo+Zz<(rGJQZCo3vnDF_r
zJ2Mp6+Ygk}DpqOkA;SKqUKoK3m4_D9<y~MPw3rXUWzxy5{t<Xwdbx9V2K7vaDOOeb
zgHM<*hDmc{Y!uj1j%dPp%l70xFmL{r7Z`PWy5l}8<wvb_E6^*ochkK<n?G|#U-L2H
zMW~it_U-eih8bDxvmK0PiMmqBfAxqyP`Fc&DK9~tgs%Hufa%frLGKU;nWjTnt?R2|
zw45K@W*C^Dhu@x3H~CpM#%aAo%37;8{URIyw;njjEL8Zw#@xB2ozvQcVal#ExW=oV
zTA6uQK5qucLAU%q{g;^fs=|*kwoyWELd~I}8^eHF?dUBET11}buO8u_B7HNq6!x)5
zN0X3=`V2W3v^}E~f1P!hH0aQ+rv7)=0lWl}W5nJzatc}Gs7-tFz0d8|yb?KQ@9TTo
zT^le(`2xnrJ4SEv5s5cTxVKgFHmpwxX{-7n8fMlqaJAHi4720)obA!<;<O)pY|_Aa
z(y)za<RyRDJiM$?VR|e3f?UmrGaKq&J2VY@qvefi!HC2e`1vM@Sh>iI>Hak4kZv#Y
z7&Z~ZLu7hFAX$tMP<4t|R_`_g5J&FCl3D1K-ptoh^>{%)RKj@fJ28l&8-}1PTP0|H
zW5OzQ*@H&eWW|6(xAo4&gX7^#dlRbRslE)N0+H$Y4RetuupSm)pzk}ag6q=aRmWJ}
z5Y?TwR5(cI!RKt;gFxN~wr3bAsR`{_WZD3r<xWn|G3UJOLJeEnD>bM~AJ*@Fu=uq>
z23w!RWqVSlOqy6z`X7bKtQjCHSm%@zO2dh3K4oy&)R|zaK8RoPxr&OQOF(O0k^2K~
zkVI&le0(>2ELzR*1}{H@N7>cB?tlE(*+%Tx0fpI^D8J{IGGgO6g|IDEDvi@|V|%d~
z?4GTv>7H`_VL$iICE0hTi;8`|-y`3Y$@K5mzGF$X*X1fm7~;$_sY3{NXfcS29(2hG
zy6skVJHf55%8LS1foskt7!MEDBqmi|w(+xPce~0f`X4QUMg)@+Vy~VCN6q-Ep9vod
z?zi1)(HI3trIM}-&oOMf#w)pU!)YqL3e-tyZIP(p-Sa{5{)3R)!(PlYZ-&m5+J6;~
zfsmi&VxzWneu$k>u7de#(<`$Vpa&bE*Wy&4Ydx0aCg+JhgL-pa+fOgFSvo-$n@8t|
zAGEnUHcV1+{TXeWjaR7O{?qc0D^X8_Mg8gNoDP9qyi#gKMkt%pb%U&fIW-5RFf>q!
zr0>*MUZ1J|CPS+-G!xNY{+4H4=HQB#BNcNcw=rQD69@C$WH$`X!@6{hO$KlJM~Hw3
zng1e?LHm=$U!s5l4D;&Y5VWh0(Ha+vD4+42Oj!LkD-YT%Mg<!$$OSMa)g|~?wEyZ=
zf1xlW>2T2KqL|nS+pd^@6N{6BA2eLgjJ#&!tDl{nKtsMqXJmnI8$v4FgRe9jwfRF$
zs@AZ+71v8JxpJG!rM2E*!!IuBN_&WjXrJnQ#Y(`m?&M<FG2O`xdT(Kz^77F((4nFD
zf>E0nws+_h!;r2+A2B~zh4T=+S=wUvHip|oas9}WK9;iK{tQQf*-V)0GRMI_dVu<b
z8FC2HjfmnR---#g1lyp%s^1HHXNWEg)Z25rww9c*Da*SDpb1>`jGhOb^7U(OSqZPg
z%FBv7**(@|h=E_K;3w^Fwq%RZE6*6%8<^9lF0BLKbcLuzs5rLTqD1cF4dt<L2;y3`
zMN?w3j`x9i4s<Pvh)nh6v>G<Rq?RlN@X;DH^--)CM$Wv~rr|H!;K&a@gEiIQlmYzK
zkN(;Yc>jvoc#b>*tAGG`MSNdH#lU!Et27zS$|U3zFQ2VS<9=Cv@LKB4dq;+ffFPO;
zbeVZyXG1I|;Pg%&43r)_F&8VG0?s165^HwA2{c$o#;T|nuuh4#yH*QQ+?mrRzYo)~
zcd+#+PHO2=p*BpKpVGvV!o5VaX!5>r=?-U>_o3-N-H-1!o39IfWN58=u_>K0h%HUT
zLtFZ%C+2C#Q_lnoJ9o_rjOROl^hjV1Sv#l;8NF)HuL$MfD8zd+G+Bx~f6!(TA?v+}
z&6hY3H<O{g(FIBb%(~K%zyRfH0lyL)%pI{wuo>txgugS^(6y0%qM-+ua~5B4Q&_K)
zN_-XOYOVO`kz9f3_nF?g-ip&_t3_lKbHdjiwlyeTJk?bAV**;vq5fX@PF+C6lleXh
zCfXDA>Y901JpB1_pBj|Z^bL6gpW~9VwyUEJC0hmiJ>C7TpqJ$1I^GMbmJNCoaJCl%
z=3P2^CYVnQlB5q|PfY)Y)9?{p7ymQ2nnX@<MfcKOuk-w#zR?lB^3yv&_tE^_swK;g
z<<wU<Ui2TC&8lV{c<Va6Wt-)9peoCOVmF)x_}u=h{P6C+UFFm`N^}Khygq<UMAr7$
z-nhEvcC^<mAy=siM6pMCHmpV6gM@xCCAH;or&DJiSPuK9l=X#~faKe9hk)^+Tip4S
z_9$SntB3HfDdU`94Tvk6{ob4*!z;pQP2_77YpBAj0EM7^j^~}oUwSJ)SmrLIm4n9^
zXugK*tP`e{o-vDA6}bqszLfv$;(k}GQo~=Suz#u>m)qkMTIL++k-f<Du~UP-6He<0
z9mZ`$wwokkP7P=NJgOFP2~59e!?@9&p;unrDN;ohX_?rBD~RaYT+0l2-Z}Un9cOKf
zVyc(Z_ibHdPBUN{s7eJ%H26Sjhu93M>3NOeyR;b($ldEo$p%Qa)M+tpBy)PzWS{EH
z_H@qn;l`wWyz9CrQcyPqqKNS<G@o|-)rABb?39Pu`2ls}e7+IPQ=jsh;|t$qW4F{@
zi2#r>Y-wNN(`S4J9lieNVF?2xf@X@?OJ>J1q$$Tm@`{5BZu7z}S*yWs=KM{))(nCA
z$dLyxC-CtTKgHGWsKjh_-SGG8|3If&K;@;3Aw&z5pBjgwiV|hWH!uF3Wj05MsK?+9
zfOIwo^kd9yBdPFM|Nq{5;eWsP--=-hAOtp3fVXQ0h%ws#w6S(TZSM9zKJ3pWG#meC
zy$XKt|6AeoQAI(xZsL{GdaBwtGOu5rFuIS*dfCM|l(M7?Ecy$(*X-Dj?%2yeit%QO
z;r)QCrdSc{TuibVo)9y1*!P@LK}c;eUj_rj*UY?sep~Xl5d9@h&fZ?BS?vU<!^UO1
zqf}kRfwOb@@cz63wgG7#WkvoI%rgKHdkS|C-94Fe$B(^~U8RRcOuvO-8WOMV9qI}P
zivagG`2&3+)LsqFS_PACV(Wbvxyd$LcVb#5*6-_45@CDU%9ykU^=3DZBYuDVMYg{|
z<N+1cOVDok$|FSQ?y+M(`N6xs29<=#wHM5Uq7A4w6GO;%%h%|4UAbKK`ZoMP{0bys
z#YjQ`(BASEh?>-0%Vyy&8!0%(rKI-H8=&>C0VDCvm*<JKZm~kFsc(+`^sn?0Ss3rr
zwOMu<H+5O{S=s$v|C#;#iL}X>Thu95q627%7s%zBU<i_~n;DrYuUd$$!U+OKyf&lc
z8hu6{>RLsQX~hvo(Zjv>WcwCs<<Galoofc8_bx-op+Nsz>SK+gKv-(}ak5Nui?abv
zJhTuVd~_|qGFLH09;?vKVcT=^>~1g*6T&j#Q_G}Y(%8OE*FQl~`vq?1&zW9xbA=&U
z&C;TyK)4{o&Gr2FQMps<6xX?Pk{h8Yx-3#<TEkF%(FyL;sc(>xtsPari|w;cqq>DB
zkntdOFG<!=!R)yCWz!8qS?m*P57Me%<b{gP$=)H(t*z^hV}hO9=ccTf{F<(}QPGH+
z4J>xzCTf$UsK=uw_F^7F^VEEEjFhy33>YI!8<WsyDm(=AMYI^U1<i)QdDZFY&K#=&
zVHLFi^?1RGNYoph@0hUK1nHcf<4ppxv5>S|iogMWYl*ngISg4bw!Gy=39ej%5E#W<
zrbxIQdaWuO!c56qa<(b@4s;!(Im5YN9G-Rckyn4s@W<NX+p^W7@tj$UpDDTeo|v9F
zy0=F**t1Y_y5gcDimaJKY9sHE3dqy4K+&e*c?F^f)|PXq;<*6MVUA>6y?~Kr$E9Gx
z^H&d`x2$$?qD%2?C-rKL5;_Yd?#)zR0MfnKd5ky2E+~qR9dx{yp5l^KVK`55)Ro}^
zDnnM}Lp$xXLjtrNarRIOj}fu(Q&Rig8eB{A-=bC~#V_AIJi<?3={5a^2g<HP0-KGn
zu*@pGT~`DBfN8NN{`a1=5D{`n0wp-`k`vQcsK;-D&tm{foZFqF%Ea;3qcgC@>`fSG
z*lygvZGBUz{YX+zJW}P8;vPEiAJuibD2x>~{Ke%4YJk9JZqi6^3ku&EbU;b%fr7at
z<8QtF9{ys;fHM<mQM<K?4^*I_+4$))9E79MEOGnWxx8Z5TV9e9I(vscDP}kXa*Fgb
zR^6C1R!6LAeKkQc945JV01CYHiSIY#Hx63zyT7NT7sYiJ<vHOaQV&{U!jT+NF{DCF
zUf~b(+~0K}6bzA8z420paXi(hf4u|cED}L6YV&wAulsI^I`Q_%wY1Qc`_2V1er7jQ
zS)$j)aXxw@biW34_KDxK?buwPuSyC;S2;$}&g61QFJp3Ttq!))`1x1+^`~8_a=v4)
zh3yP;Y}V$SnKmQafZd$o;(c;MhZs|;g{6U?V+We(j)iF?y5h1D;;O8+t(>u70-rr=
zXk9pwvLx~z0w4ILDbPb@qoP4_5b$}JT%vHg3TmDo-DRi<6hGFB^BDoIeWr$OY6CGt
zkM&27iY|g3E*j$toe++NqVbdD!G^k;utjyGmqCsnCbYD3D1@L%+AcB&niC#B4pHAt
zQ!GY(N1-CeH`OkA#hP4z&<l6jey2K}=!2}v(Zn>YU`Wr;!$M`PKz4Nm)Qe~EdmD}s
z+{XTpNrI2BgNP;beVf)TTkN#OlL@C8%~n)(-<rkwh4uZiPhFg!SLC$839vME+T+B{
z#nkm9#Ulj@S7pLGm+$3aiQuc#c5Iq}k?CXx6$X&Q_D**kdXq8+{XD2}bcvIXM2cO-
zMOLe(x(B~Bo|nS;!wz&*am_&!FGGtFxt<vaj+~Q-!lceI+=;Hj;(@0v&hHO8^`Vmh
ziCKZLJrxKIsLe{ETKJ0nTv-`h3AjFp{B*9BI&XKy_t~poN_THT$@(Yn+*^ZZqh+%7
z<rP#-uPecw6?Lf=ZA!4g@GeMQF^mY*Ka+tPsaX=oUV;Me|E$54er+{3QfhZqE!)?>
ze^VXwJU$EH{`B;*lrYO`=*-&CO-;Deo-36o?uIiIv?WaapYGnfFX{e``>#C9rnbz?
zSyrZ|mN_A{GBtPZ6<Y2+ae(VAE6crSh~^-ixs{xm6DO$TBzKC63rG;XK6ZY8zrU_q
zaNPtx9FO-nj_2$7)(Z+BLVDQVaZI{-mqn?h$4kFm!(_D+`|6#f`$y>X&*aZ@^qStu
zQ`|`<k)ZG?KWMcfhI=|NDuK3%iE09GHvjB#KPzo0_+@gd(NU%Rdib`kcD|yM@KkzI
z$=frNn8xKFE(Rg{P3fvSDf4Et6wMV+=e0?>{WX>Oy473j*P!F+-b3%AKmQPjaeDb>
z;V;?l{%RB^5Q*FyI0oDL8VQ6bbb1|}g%y&k*8~P`LhVIBC3z{FVP&)U8U(+tVfv;^
zUT<XzPS8mFX0q(c*;Nx!G^}{X@w!YlUVr_^eC9?PsbW7iYOAq~r`rM|^GSCa8!+yx
z2m3?emj|g2?~cQ{_KY>=YHfI@+r0D0Ro;UZ<2@M-3z_-90+}<DbMuVBUwNw?CvJ@F
zA~c#iJFE!n+Yz6Nd%lbejn_cU)r6m0piaYoKTSCpZCIQO{Zs!m8PQiU$=^=#)xCn6
z#+K96)F|h;bTzGj1t&dbfL&!uQ<7AQv8)SL1$y!8@~=m?N|PZnMf6{OK`l8X23;ji
zs02u8m^<(C<IO&O0ZMOPkt&n#mP7K`)V``aQ0k#+Srs<5f3fcYgy4^<W4$tbV#js_
z*{H(o2ztWX_(8I0qUvX4%{%!l{``#Lo8AF``igx#au)XWpy#)==v+E0Xs6-jkSi7`
zwiHn*TIiJY$+5K%IHt@1LzoNXeyiodkoJ9HNtIKYUq+Euf1Y{fa^y%DuruWmcfX(x
zUse^R^2WcKE1d%-rosM$%|~fK+QYJ$U8uN8MsclsR_Xg`OEaV5anaft>fytc#zdC%
zv4`=h)}4V71|uyeMF?RMwq8Pe2-US`qo_CU39(yna6!hJ(W+7VtON7ue5xmZ6QT>_
zbccfd>2V*yDHi=X1RW*Oy)Hz0G2GlW6Aq+R3aYO=i>QQVCJ_oN&|%&#tu?yb`by@>
zgjRXuXk4BKM5G9(@E{)?QU(lX`~pgL{1nB_&dmZiD<LFA%jWk&UhiIZjF&Tb#>v1)
z9(j6z004-?^<Qc~4NSp`tTP8^Tx)3f4^b4fs+;eCYndK6&Mx%wo9PZxa<sj<Az|vF
z%9j{YfP+l2BU^t@hy|tsM;xw7YGgtZf=8aSODVc(j!3;Q5~M(a6z>x%J}jG3@GJcG
zc2WUbU!fMNBhuShpgq5vSP6b9(N1{Y5;mzNfE{2<9oO+w&*oYPAUlZJmDoKc(_YXJ
zu`4sYGN=-J*Jw0IF2OvGdJPoY-}7}nk3^DPFt!Mf(LZ3mBgzf!zEoxld%K)pf1K+~
zO%6AB$OTxyRV-tRq~^$Kal96Isx-Q{)BhweBCVj*L@ig;F317XQ53^sZo+NB{pVE!
zm{~G-VM$%EYfse^J*z+&H@ClQJ1^D_O}XhS@~5dO^?tKY6*Lr+zeiaYTnzdh6Sla7
zFd$4zfdt86!@6{xhg4)#bLrMFgKf)@mG<@FvISBw^q`Ss$I<k6==r$^Gxki)L2tpL
zl}0c3^3D3ZY&^wT)4N#&-EaxSjPPE%-Pf#e57wa>iyW5z1mj^y`4C_#kg~$iNf5+c
zg}e-Z`c5PiydLBz+Lk)YlI`4B4a5YN=G5N(M2JBDtwVbh1<fjrlsXuEaqlD!kS2i1
z23*i3lul?UzdlEQVZ|Fy$v#~tJEdh3YOv5iN6DW4wM4tfKdteL$bLxDSBS7{Wf(j8
z5Kj-M&Vfj$>r;QW${&PWNn*EbI^DzSt9!O*!0}Q0$B~>2I`bjz)7Pkves{)?c9j!n
zJ#N&mlhUvCUD%oqF12!>4!wnaJnoz`Y5~i|WfjVDBG5j0)l9R4Jr%M?G-fzv$1&Ta
zVv*<V+4_;pTw08MU`vz#YX5tNl}1;XLSvI3Saal-`S=Yz`4uo^ie)VrJ9Hvxpy<`U
z!EU890ZH-TajP1=HQg~^d93Gfw;33}B2*DjR$r#<b3d9l8QB(29KJZ=ewy%cU)A+I
z&QsK2payxpuGDTsytb^U5NmpLX@_AJ6gXqb8!nuiRZA6p>-(If$YCNOVy%>1N=kx^
z%yAHE3{TK%tZEnRJhqjyLNGcp3eqLP#&$`X<}6390$Xs-VV{{i6)V02hcjp2Mt|GM
zZ=Ag(^UpBxb8h<GnZAdm-rVlZ!~WmCwY(Wv)t$X_`f#|cv$HJX=3#Utu^<nuAjwfw
zr&BQgZR0AKtK^bQ%gC00wQj);hx^Hk^QaDeGa~I;YPoKlsNcW03!d0rQx0b8$DDG#
z%38Va$USlr5=CC!G}ZmwZnV=!9oGL?)H_aiDgol{aMfXL$RONa?|uko3Q1*12VN`V
zH>5|%2h{doW_7k8VJ`4VZ8e9)i#4;9?EHRbvPsZ2T@N*mM$`&=8w7{A6huOYN~)an
zwdQhe*bHE&6+0Csfos&m)S;bEV0H&b;mU*rdpG5d(812Z7vpc@3NEB4SUMNkBM6!C
zKX3{Bm#E69+H9Ze?o#M(8{yquo3A<yM}tIT->=fuZEKOm10FYZG?TJwFQeX9W;NBO
z^CcjI4i3lS_gjU9s0;xscd1Zi=K8vpiaY%1ZOXxj!1*x=uS!(APxIjciq1C|T#8qo
zr9_An1c}luk&{D&IN(~4lq%NB7puV%n?WRhX}~6$yb-`3CYmJsT-5e_rtY?}+Ya$V
zIQ_L+caUpKpO*tlaeIs@S?}MUZi|;Ko@En>4UvndhyQw&p622by(mljA|Un?VGm>w
zKP%xMKPyvA7=$4gw*$@g;tU;zX?Ovvj!D+jjj4Ib?7m}?yZT}ZY)||`2TozHGMac(
zG=)3UXiZtI6Rbud;vt5V8UbQQ^1=#yax3;a^>yb%jE>7n1H3a>Th%0DN~jjt?WW7Q
zYS}@9pY$!Mf9UUFdj49dVF{lRe@p}G72Julw27kTo3~j+e-Tlo%2lp?5PRR$D}iG*
za#=1lzSb-kTH8cJ4cuhO9GCC=&ftHgjbp2x08MhmEET0NktJ)=$I<<Q=>hBhc-{ts
z&TGgw+}?jW6i2HBY<<9xHHR3Goj)ZY|2Hu7n2=C|jbZoVijbvqHoBTFD{@N@aMzBs
z9O4wqPR5l5*5{xVrQF}VQfr=yoQ`&btr7Dcs69952&4M{Fq=4VsxHC(>Rla(?M{Bu
zXZUDN%`yKzM^V%As0Nx(Y=%`VAuB`kq70j%9Vz$V6*sB7++74@oDVKyX)fo2Zd93^
zZOeXX9$q|7uwuS1Y-{2h*v(y`8@AokzkW6>D_&3Y$~UiMDeJpDt)61`NwLZM71y#J
zhvyc0c1FjXq1`HZa4S+NLTdbZ4Yv;HS@!-F#S0}VRX)LV6{;-d!?6Qm{k!GHlEk}?
z4<{jU$|H_j)UuNM=Yy0I76TFC;5(JyY35a|*ZZCgM(yzaW3JsI`VxYziplfr$>@xy
zmAr?o8f&n0yFXJy-u4ys=gC_XYtNr?-7-?013l9n(u<1GnzItK#3;DMSQ7B^fzf(J
z+p<1Z=5oL(tH8sjZx*C-k>hz2g3>tsBHU3xPHSKtVJp8Kq29dtxWzqy_OSVEEirn+
zGRriylEum%IV|6#my+|%xtu6E>^D8LThJqt923rD&0W2~7?Pacx41`fW`^-`ctTW7
z%(q)MBlxC9C0Jzny8cY?;NJ?Jqyblp=cgM|ZAS^kw8u1n6KO41<a~jj^KVUZ7Dk{o
z)*0{&K}gMecFVMyr@g_q_bi?;RR<EvE5#W7G$H?PC$WcRDy*tgD}6CEHqztV?<6-f
zo^9(Ux{fUEQTupp>BPVG%@0;2GOR{d)~ZLiyi?5}dF__9Vp@n>1zEqW)F5RiamEuN
zti;$;S{&SOA7+NUuOd08-gjFj?x||kU!5nu4gdFCk<x9doHyB_>2Ddf4>KftqR*DP
zHNRM}W8YRz`4IxjpR*ar96fRbh{^f<ZisrzlLd~W6X3Jf1qy>TVA7`t9APyPsnyIX
zJs-0-%#Ww+t54U`0B*fY&)R0puntrPz>87ndt(4zE`;)tX5*O-v-jP2IuPr0cGn=b
z9b^=(aKO+x_Z07xE%Gj&`tO+K_>O&nb=IikS+LIpc&KD2b)?n(Ku615TZ<pD0ix5}
z{B@2O_a|p>zY6Us+bg<t|47lbCoZigQ4b&Qe_(PQeB5?aizc>gP>1sJpAFuTOL4SZ
z^W~Gt&EV;|Ta+`N;h@tS*jj;$RA(RBvtH_~YBKfd6I_Vswi(+Uo6p%Cy)=?w40fHb
z?&(ruJLj-e(f6rVqvuZ8O=H$b5z$3IY974Vs^`+ER4V-Pzp@a5FPHH<J@bjmWM4Mi
z`YRZo*7-{je;8$aOA`kW737su3l{9_6m<jhz~aE?Qo)(tC}6VW&*OL7!?OIC#KKaM
z_iEOi*<D1=qIp`(R2tj;Z<WPu;hBT|BvVO*hVh2jVvY#Dd@*e?P^9E+U!cSg495;c
zR|s1tE}S|2q#(s<h6oK|Cqkx?o29kK)=3}ts_fT2W}v=nh~iQCh_$>LOW)@aj6j`_
zP9@?lO1li{KTk$uwywOrUJI|BYix54y=>`$wN55DQE&|uL=5?L21Lhpk~MK0C@~wn
zU6K(oqK5eY7g+Jjr&VjC4XIKSwYR@=ILRLOEuga27Q@wD)5l(&-zuha;Yy2OvEJj%
zyL4HXf4(Y1x;|Cb&2@c4!v5Rg!)ubx)Ye$_Z&ifvUe_FF{#g5YM(ck^e&_|_DJS>E
z^6mQ0dmLXJ+q1D6p-)ZgJ*_Ur*6IPNgNcgj&%{$wVspAT1kc^*d|9(LEazg^l&SaH
z3nF1`-L9zP4gXN`TrJ(LUQzv;yM%d;u1C=BCy~Wh$|cMUA&+JVJWN6>Y%ST!BwFdV
z-S3&q8q~peV$C&06#Q+j28tiQRl@&vRkVmKFEt85Z?7+eKg(?zdwlVqB+8C<e6aDG
z3r<5q6>T<9D~w3xi!HQz=MeNl0VnNN0^f^m8ytMJNnRq}yX{jX8)*C*GS=4LM>&N8
zUg`KRR=OV;y>k)ZzqU}_OT2f-VA2{8+2%?JD;KMV83KnmiGMn-a5t1*Lv$%f`5%m;
zKCU+}LF^6T3@FnjrY2EVmuH71Jw2W-8na!T)-JvrXR_+w_M0ZFm}#0&k}ED~pmt_!
zbzSs<3S#+rOuf~6!N=oZ@BSZ4G|3R%W+Dk=2nFCjxhYOF8qrtY(?h@JqJx&>h+F_Z
zNH-VMoV)V$a%dGCG*of6c@vzxsu%1gm{p9LfZ&5J=y2<U#}iA~H6>9ih_X<0*WlE|
z{vE}}A4xgyrZmU-kGU+3&LE1n$^Ie0W*k$VZg$Z40#hkV%1PJ=T#s~7AV0fuV41Q>
zWb)X8awZ(^d_^5~*G)ACp4tVR2cVFPLc`kTm6pitf#p86#EJ&k{Pf2xfd+{J#6SZr
z#?sOX*{!+EuD(!8hQis0Utmw^29&Pu*-^nVkeUYs>2do$3wtNtg7!V%GS}8HQL)#F
z0#`3p$-mi~!nvP*!?$U01XUY0)ieCV<-!ccviL@+%PjG`)bf^RrM!bzXS}~sAMl@u
zF^04iEWK7pUG}A3&Jz-$u(6`Om(ZWOC|~$$J^sJ#SVv)(cF^_7-Uu)=niT&zp_URO
zTJ(|Qizj0_8Vj`#1(}Y=)E=@ttP<S17UrU|Fp>3AVqfLCC5>7kd4)XL_d&dF0Ewf3
zih%um?xH(~$ZOs?VU>1bLbQ4JpVo(sP+1PUnd0T>@BNxZuV-ZMyz#lHsCQNII`5h(
z`v8)GtCYk#J6@SiU&v5P7_FrC!9Mc%2rbS2Y0Eon!P~5^xmqnJ@%Y)cVQ}4V^j$S8
z@vamD`Q(n0v%(_51R00Squefyufr27s17e$<C=p=cZiPlWC|Av5fWdL{%Z=+>J51>
zWV~>XUWz)l6=AGoy2au>1J4Hogw>kmepNZj+c{H&z*o;8M0K1Ys-Zbk1(L1p-#LoA
z3(BoAGA-eu4KvV*`@9fiGq2l%&vkWegFjmz|2H8eO|t51pn3;Q0l|8~qCT`4Sd>it
zlIU+F^?5Dq8^rej&<LU+5wwxbg!Ii^{x{~qN2rLdIevw;VpYlZ<>;CLZKzN>_fhzF
zhV}SvZy;GNj6_OHJ&h(qz1fFBKuEm4Vms9!us0z9vYle~nhj^I=UXOaf6an%UF{&`
zYQOjSq}QWNDW{IBR&<!Zr~!oJ$xEQPs=OflXs%y8xb4$F8zV>#$ru2eT2pf?;SKuG
z^OBTnOA-=+$^4i#6dt5Nopd^P<Rze_LGAAjWbBcer!>w!T^XPdDe(3P4^P|svfB^1
zdE>=R7R)M<;v>2R(Gp$bi#4mEID;V>mnZhz@izC6{-EI4Sq!m0W}q6kcB)K`A$6^9
zz<@i){x7IhBlI&lZznzzrvA+wS5Tl}@Jgs&o-<_04VxI-3JrTK|0*WGF<0K?n*X4l
zV)eC*lI72dI|(h)O1Dc0dLc1O@X|KDmvpCX6G@t%a*bb1UIFSiH6nZ?=*P4Y=+u*f
zh3!wPw8t?A`zp?A-1@;+|4H_a5%JQxq=F)2K=13VR(0woB&ilVav=4x&4tV{^7g%p
zI6ojm$?jp>%f<!-IDHwHfrZC4IuUpw=_@!CqH5NDhS<m<i>Tqq))FmK<PX@rPj<|c
zqXv56i7amAYF7*GpzUk!94G9DtrdRYxJRLB_sN+xYS{(RuBD`STXf}lIXQ*+g_jtp
zLJxI;+id@<*f%w+?<N9y$%m`Ui-I~o(uU3=l^I0iDO&xLy-yroHlk}QC=x1!g-A?{
z7y0)&mV+rR`K;?6=9SVqKyJfUQwYA?)Klwq^y}Hh{RDFlzeYuM5~7vj+Z*|A)c?i$
zYjZVP>&A;Qw?~!4rB^re`mT#zfzTrC-Uz*#sFg3VC28GjWh`p(YA(o^HE23c<fAxi
zB}XWKd=RMrZ}E_{N%`O60cj?8mGG6-rSx%4mU<msu#?zQjtK#k;B5vF|2sbY-@vg*
z_D9Y$=W2m4z@Hzb6J2+=OJCQ-^d1udYt$(f&JL-tra0{pHQk>H8M%WLIAsVL%_Mla
zR{q%adyc9{tI9~j0X5S?vr_Lq!N^ZiCwvOi!p~J#=@j)c`{Zh>dOXBOe<#%aTC!?~
zj?$eZOJC1ut1LD;^fxQEmj7r#N_z11Up(=dCxi%E+U(55>DQ9K_D!6u!M&<AFa;+I
z**n*Ta9OBhWNjyvc(xoi_G8`oW<70UMJlOth7fGh%eUa#K+^%7f8%Zl_Rn0GRTF~Q
zIPKf`JXj(PzG+U98aO_mZ2BMnre75NG%<3cG3-9gE_EmXY0{Sj&HcFhAoe{^=%4{a
ztcxxbeunH4G#=NEb^-{1S-wvr29eFdf+H&ltF+G<E)Nz8&BsB*Q>xiA)B>jeIm=Y%
z7}GcP;ne6ZbDS3{IR{3*HM)ZwqezS1n>ISqp40Y_rXZy6_Y=UJTe))oWndz+!;Hk%
zeCs{=%~fMB-}E=v^laGrK5)i~d|P+X*rWe{>j`I|5mCK?)!Cki@nFed`UPEe0zxGc
zZg#HIlA<d5=BvIFpM`mbh<8)>q&S1w?L0%682>?9+)fJ+H`88lQF3!wvU0J1_xH@x
zSZGi%I+rZovnc0IsQa*0@n<g%h1Br@Q@>^<P7xF0|ErBNkD9kPfms;#%#)RaD-5=-
zFJFhQOnm(G13s&;?c`rnS=*@GSM8PFYIya*Vtg!z@jPCHcLq<7@6i@ox-sj@%{qqo
zhfx{Q>D=r14=OvmFsXz&U&eXKF#?q3qIjH>u<SkhxU1%U;M?^dPJROIv8UCG_CR>@
z$`D!ptEy>QN9bfyfECTw%h`A|7x9b8X1kfnDhZv#oIR_UTpx>AJjGhj_S}E;M6TvI
zx9;P%$GQ?%cr;zj5=Gg;_Yk7Ybly^G*C)q%+c&KD1-U-WyZvLaEHB!3+e<`OYg*%!
zqM4o5s<dx6(ivgjv?89CD7M=2!7n>qvSVC{Jh`LoKX__2fSqJVmtptW!D?xGnLXO|
zTcp7W6!Y|74%lie(&Qq-9qSURglr^buLD*HcANF!nZ}#+=i%+_Pg2&ku<o3KapGbI
zx7)Q~P6MDgeHq}wa@43(zJCbq)`LtMQknTMoF*wnYrMJYde<)6vtrsV&5~8iI)68h
zywtSxBZ-&Fm^)!KbvNfEHrUGl{30n0^Qm5S#r)~O!16~|?T5f<_b6B0=A(`|)gId}
z#HfDSzxdpdm<b_w%Hkd^Cd<l-(D1X3uCu4`yc}mtaC?+^U^EWsTT<pH-bXQDr6o~7
z>mWZ;J94oWK3ubr{-~L+oul^{K85&3AlzN{_=3mQ{es#s-X6ntbW>SfDQD3KQpO)y
z(u@Ei6u(|S%Dn#frp8W6`bY_A&lg*0p<LtCv%HeY|9yMfNC`3Pw|czb?~LZ+g;%*H
zxdM3@J+jUpS%#v~t7lGH>C4H(a2Gd@GlUMiE24jraYMg$t3-U7Cpw*tL=r`|#3a7Q
z8oL|jFA#p6s@d_m;9C>T5_*s@o?V!5wV0Fe-1qcpk#k07?lQ)V1+#Gj=;EhBiZ#d!
zQL|)tpS{e(9F?4P!p>eOM_rGYLfFT9sh^AW%tK6k$niTjhCHoG_$S(~C?1W6<ULtE
z?lqB>F)ZQNEIu$@=)?bl!S~S2gFNx)6Vh)MRQE<Fb@P@P)lv?*BjuO%aP@n}d)<SO
zC!LKt&8_%hxRYqAmV=fmUGY$m@Pq*ym^x|f&;z|vDD@5D{dwm@$qJ)JvJL&PyVruw
zcRr8smeDy~MVv|5`lpHF-hj)^A67+%jECND{Nv;hEh4HtW;ME-w+k~Zkj1}rK%6P1
zedR0gz{4&m45%b|0E!QCsZN~Ku?HTO>bA$PSE|LUm@U99_^Q<Ad;=vuI(KdM*SNaz
zJrgT2Son<Q{bGYN&?vAgIB8JYo+nXMEEj3Us>LgW=BZKJYEN!EKB@U;c~Bo4p2B>t
zsmgmV%V_%c@$cS5pNl;)jslYF(@0v=ddsB0O^gIj35S?&2IYGNXz83!P<JjgabCwo
z_@Y{)yl0WD&<lc$Fo{B^V;0-_M*S~3T))RCNC8>`qY9$c`_>CXtxjXHmUy5`E4ArV
zPH;Pe99uG}kRxajynMq1TDIvWBB?uc;attiH*x89h2AyHv+|`bJ;d+JB%}Qiw+T6F
z5+42$9CM6X>EPS*1Ku!;OY!~{6did-qEG(r4W+jWy8cY)dZ(r7e?3`%WEcmXc_QMM
zTU~Z{ZadI>GRP?swRtFlxX|zjv~0gVUI%??@<@XN9Rkb~2FE<d<1pjMg^v-kw2&QB
z_M*p~9d2BKfxnsqRh@v0c(OWrhoZS)dGAI^qjw@qd_`!*|Ik%2+JhgHVe47BYM75E
zzwZ6wW8+88%sLGi?$UKKsCsc3F6Z@m-F}`lA_V{DKU2Zz4qb|(9{M8M)^17v7yz9#
zt4D$7hA9Ms_%qpP439^Dk7qYAwFxjm!DUmV-z{x$C%@mh4;+Ij+Jr00brNu;2BC%6
zJrpZtOCtW!=hSJVric@sBl5z#$pPCpttY=2HfHTx^nNihd5qz~O!RWQ=_<Ii?npcH
zmW9dYw<3htznw%48DM{%F`JbCsGXXeTDJG_l8LxoK3QTq5Uxu8eL_RaOz)9kUG-5Y
z<h_H;{DzMUESH!#mp_qqSj9tvkg$hvfEO(m=Hf=y*sXsAv5E}*v;2SxH^noj;OJ{}
zbfKNVsQsTVW2L-*#HO4qISmw-hg#{4!b-B_TZm~PExzZPE-Z5PBD`cLNVpKw0_w?*
zjc@XII5ZE91Ri`;cZp~k=s)h1v(`QJrl<BnBSEO7R%qnEJVoa|!|LHm124d$ZkvBR
z5$iplJpSzhOs4#%1-;Q}^fvcjnpT2kM_HhC7ixd}n&84Ktn&HejVwQ-+=n#3GnwCa
zuchdM`r4-hXH8#sq-<s1Li~s_B)taWWm;KiEw}JpeblnHFS{PBG~I&&x=(w3WBt}S
zQ4cjNtAvMGS6x%;RN$?-&}{;?n5E+Z9*t(EF@9Se!mb=z)C5LW;){-|`J$hXZ2K9Q
zuzQ17m!G+xI-+_kYPx!x1^Sq`G7@A_T1AB!sq$<0R1{)I%}+n^)iJi*U6Kxb|It*3
zve-alE9oQ<q$NbAfca(vYeq1ZaH6-kxvIehJd^Xj!OpnWsl3xIvLL$zJr0`wQ2HnQ
z_#1BlN7;OZjf4@l{<{Xrm1_>aw=mA=<{XyNR;F3%t3tjd={uLfYiRn<KR%5y=Cd)m
zADF#*sMS)wGP#=kiaM*g8$DR-ZRP0@5Y0wHphTL@w%<anIu(;vf8<Ehe{|3|rZ9jY
zGhc0TYSZI~S7g(vvYS(HA>66%*ZA9L);GlqkIML;H@>PMhBz{$V$@PVoxk^`<)%^H
zi++{7>t2h|VM-%j1*^HT)fQT}^nBQ#hyXB&3!o{~F^9Ixz(AFiY)|@KkzM1uk6A$W
z!U}y?7o1IAl(g5kiC;zgTUZ`!s$46)zH6S}jblb23yg9@3?MvHGhgM`f=7cuO(UD(
zX6XIWZHtW^kG;!JAJ0)nwPL1?i<D(|mxDxq=GLa}#5}Mfz<v&Gddvjc%%*IW0Rz)h
zTUt-*^RDHP77?D^n!B)GgRd%5Z>l|5XTVP{a*q64$XAtrfw1)>$~Dr$FvJu*)?XL;
zGdHm0mi;#~Lc7IIU?Z;I?CiE}kQQCaAg-`uo{nNMGu*8m_WFgY3^c7>o=6G@?!VxH
z(|3vetCq+r%ReyO=WUBsW*)Y|dP^PmxzwO~0$s{(&)g~PVXzAUsgt0S-}_nGqdF>8
zH3Kwz4Xrrixww8w^cy^0Vdy;&OX3sj+F!yDmX+`ciV@ddXXNBRs0`jL2%OSE&t3MF
zkcUawe56=apDo<hGhnO`awkbr7e)iM*4q1*cGu;X?X)yEmiAp4f%#jL1juha&wbb^
zC2FqBSE6gedp(!^)~cAgZ^*zmq2mU9cG_qzQC;%~%R)|icBn#vZ>rskl|C4mw+U!}
z5E_1yN}KQt3WL+NJZ_Il4#klt*u`#%<wx|c-P(HwoKDpa52sJgSRr##@-TTE&wJp%
zZdLNGJ}=0EDJjbmjIi|RV)zT9joF?7>%d-UC$HKC2HMx`Z&}vY=HsBy=reaD&h)X%
zGnPRhYJ*eXj&}Rrxg+A&-^6`5$t?|0V)_?e!~G`)?J4lW{gu0H!bS#loCfP;#leGe
z@g;wS1_G<Szqx_eng#WPzokCpO=tULq-Gm%@AsERaXSj4zHbqB3VhQbE1|l-T^oPz
zd9zO~Oq@+HS8G3^<1x-_MaU+xKRRLTYD5n5szuPQ;rreiLqVFVnJUoBfOlMW)|n8M
zJc-65Ln=J&a2nt~=)=SmcUKHfadUV+3>VERnetPZKet9~L|`q+9s9f2Z#yFZNu*}s
zDAeVUH1SKv$uB<yJ1fcwA!6ter-{PJRJK7+Df1q1zw6ZED-&dG=JM8iV%K0Z(^fyl
zf6vExOAB6acu*%YVkKf+@2`F0!~>kaUi8bgs_WH#j<8(3t4c>fimmKSAR<ZVuiYku
zh;2S}+wf+!F_jn4%-sv<f0oQqq%nJxb7mQ9g^XRQb{{4DFpM9yC7k74H}E_Bdd9h7
zYUwQs%{k7BqI^Nq`=`v4+nSuTuj05`zEf4j1KMh^&%d|U?gH&((sH;&O2eN0R*5+j
z&a-a^>iwpU=@ykv=A-yxxRV;;r^rjootY&eUvJK_uX!_*d((7~fBMrpI*)UGD~rj=
zqt10_QXgqV(fJc+?r>H^Aq24Q4nwW)3wLAQF!nQ)*Rez7z&Bkge@kV<;uONJlmGU@
z7-C{&lKt6&-IJDY2I?^Hc7Bc&<4)ln%g_Dln(3FtxWB^<34oXg{~m*^8nqj}!L9@i
z9q1OfSrtIBJ?TuQL8b_h5g@~6_h$kVBd2)bMBYVyoU<#|g&g#wIdM|G)=4`*x-HqU
z-s+cpcd_hJqkq*(H^sLvvL=QEhD@{MGDhvV2cHv(!C)~P_O(E#ZCOdu5)511@b{cU
zYVDo(ceG%)r=l6gJu-;#RxZ00E8IZbO*e&9;zrLs1ob;;tR=g!&8UrTYW?z=sI?Lc
zWLWrHQ72=#99)}!5$Y4|lbm{dDf~wTHs^2<H;_1b6?mmr41oSpqh0bD`rd1kbpdZs
zi#FiuH^TZpTJ3bmgL6V#XT>YXwLuI;GjeH!XeQ-Lc!Efx&Ud!t;`jyCTyYBhJ&g}b
zzHxHdB?tij8WcH!DB4__zd*Zx3d{h;zz3Y*)$KQEIVSwwKkAm!W&2TK#|ZluH>!?$
zIhh(vEx}9y-SK5W>g_o;ctm>>;fKpaE%JAL4!J=kW!=?jWJm3eJ2kOO5)6xp<zEf%
z&7rjp14gj*AaGp08jJo4cid3&vwQchHkPK;33K`j(V4>LZ9<)iNb%U2k>gE?%cf%V
zhB~cI_}+vlI;pg~rO|B6gX_vvT!|HP=s2Ah>84zenR_IY%moGn8uZB}gvycBa6YdC
zQ+Fx_+lS0umP&hXB}Rw5+((tX@!dmHbYpFUov-_jYqeyb`+$9-3tO8)mUVC35F9*m
zM2`LNbDldlc$w%R6l}9?<WtBQ8ZkV#%PL03xHO?&>kOO1Gn%sv;`CVS1_6ds7IT-{
zbEUEBpZ@Ru9F|<4V*cr_x@3k-Pw^$^B#bN$=ae6<w4H9_;0;-t+y?VLASOES^$uy?
zJnu3+Fb1U9+M_O(kh{Il@0?QAx_u(*-;<=7V-R+$%Fl6&jX&rJV&EyR2a?rICe7To
zSI)?NKKDWM^JX}as-4yku+*Ov0I1zgnMJCdHm1(s5H^(QN5&_le=l2(ELPY8SMBMZ
zckJ3<Wx;<yjO}K@NnX1^d(fj7YzwG}%86}iJiXsm=&KbV%ky2UjyJ!tjh?dvJS;mU
zoNqaQECY|A6w~I@;e*n`Q?N@XJe$xDKlexOH$1#Rn>MwOv#!X!@TG3D_uG}S?iL?X
zYr!QW+3jO<%&k>7jRK_&79t$pO0Ur<t3_wF@VY&EaY6{tVl=1vQ!F(B;QA84USr>%
zaEu^VB{q58#5PZ+yvF$o69aEyLvgS2M-GEgsX+eH=Vo=8uTxsgZwwD9l?(*-8Oztd
zDPNN#AAb7QJCw}kyWKNOm|-Onnj@24@ldJ&s1EN(r@I#?3Ng7Wz+M)mdxJr!8UHr+
z**fiSSG1gVAKA<3=9Tsw%ps09<F_zpO=x*p(iL~FzJ>#qs*jJq8{Ioivr;vLtVU`C
zTFq~?RT%<CMLRC5ZjEN&v5Wu#S)<_f#a}jP_fQ}s(ar@F(HRfjLYgX&KT)PCq2UJ4
zW98ife7}_Ja!OwAh5p(Q0Q60A07Crt7oP?JHjzW=i>mVqPC|K(dyE&?Qm?k_$2TJ5
zU<bVy@N4E!uH7{mE{Kxs)T&<7Wi=<7#jTLJ6x^$gxWoW|Q73u1JOXfLO&%K4yjRVQ
z34g=4^~?UoDeg&jP-}HRUXn`d$zfdC#59XYc(XQkmTl;n+@A)q5usYU7VI{`_T|?b
zHY4ZcZ|`Ok99d?+pCMJCHJ8zMw+6+I90APd&wf?p9L;#6L)Xmr0$jjnw&<x5XVV4#
zwFH=BcVI`E1n0fpj@?>gTtu)L#sX%6a^eTu9m{(OxX{!JS<IRY8F9?BY3Rfce?Va=
zh8+P^Lba}ZH*t|6X^w|M8fZ}58K8c}{_3n+-T^*pWLy9ITzjYFhGTc|O#$zd1+Bm%
z+jxv@qnLr;-(ceTL#00O;Bl)|Op2}R2Ig#%yXL||0#Ar8IpE>aN@k*deNI0!dSc7f
oK2z^MVb;z6XJZ`4FYX>3O)<w~Ic+o0hY<{Q6`iN0PhP(HKME!Ql>h($

literal 0
HcmV?d00001

diff --git a/_images/raft_reward.PNG b/_images/raft_reward.PNG
new file mode 100644
index 0000000000000000000000000000000000000000..f69f56e81df52f35a72644eb8f2ed194297747b8
GIT binary patch
literal 221269
zcmeEtXH-*Nw=M$G6af_w5F%Ycq=O2CE~r##(naYV=_N!23@AmUCP;_Si_%-Dg7gm3
zdmyv`0YV6Y8^8CQGw%I)&v*Xa`y*qo?469U*4%SG&#aL;T57a3tTZGfB(%@fRbG*h
zkikeuNGqr)h<9++ByPmZ70*{{N+e~2cTvPI<aUahiX<dIVy~TBT_t{}c2_s{Bq3qq
z{rkD{{1wkW3CU&Na}`B>KZ~7a$`9;ni39?H`R;A@wd+!ETFK*?k}4`HT1j7ix*G3k
z_WJe8Tj!TKanE|*e~tg{IruH7<nH&J_f<TmH9Ug()ad4z4VnCa7mn2#&*H6R#j(Gq
z=FR8L%|VS@-k0#vOUn2jYcZ$>-@h(Tbtz!~tp0UlOP|Fa_8*UkmqI<8Oo#vatp;2x
zF$Dg<|JmUG|M~w5?Ee`h@!`yKXh>!3I4oka*sY6salrfhr}$nI)Px^bw5i<U;+WPC
z!CUw*A_var0@euz5_;P_md3{Dlp>sn8*->w5S{O63!H0Q%Nabe5{;3A8i{K3t$1Vb
znl8y9w#?tez<0X`s~+K$mTQ}X+I};~ICgvMf0qBumCOg0%Xp@@oomMha72MNL5_}T
zCu<#pgJ?Np+|ON5i8s!88{RCKA&dc5{agrO>}yz*0@PR>8qkwB;)cunc4|G*hrlCX
zA71ArW_f`Z5N`D+o346+3Fis=u8Nu3A8km_JHff+J-7_aK0dE0bP*fgQR-#BY3yMN
z2O!SQ0-P6`SZHZ!?<c;vpWOGTq_D8ZrLU^|$B&U+j8&=VuPlr;P{ZTY?{=BfgJAa5
z7PztP0(b%P_j8zrtFRFVAb#>)_DmfX_T!sh7496>%RjTd2(DiTZDR6*?10YAyFHKv
zuPAH=uA*Tbl#_S<`Y0LaGK3@~VXYSWA%W6pX(1b&>>phKa&^)NsMVs4jGae-WTaeX
z)|N&?%QIgsj#-!(r4SA>LBze29)cqaZ8=0hiaJ*UHnSk^BdW~HMa5<9)l(MRJD&y1
zS|z8<Dn0SNg5`7dmdmBp#pQ?@=-@=nV9UkXw#U5pRK0gN{#_H}>Lv<FSOMK%^anW3
z2G#juf<*xcF(hagROoGaVq+rAp(m7{?z|xEFNib^htgr4{aa3Ls(-ox=PK6;NDwau
zZ;;fC6R9*$8ZtW61-9Vc^+SYGu~}GMyE7e+({nQ+SFlS<@U%^G{V}s;&`6sd;?qQG
zuA4L2z)6on<+Xy8<>snV4K%ip-g&l$mOmp%>|*f}5B}rULa4aBz@u#`@4Ax&9B-{B
z{o1d2eKD7Pj2feC4#ajjS>k7scZ->fw5zk-4kv5fEV18Dj;;@g37&POo*Zva(NS{v
zE%gPxj_N2O%A5~SBwOs<cZmONC{&j2=-=`iB9`bv03xxBr6y)_6NkZZm$xTdz+fys
zWzneWt*k8+C9T`Hf;aYmg+H`f`0R1yg#q@rm{yf8K0_<Jq+XslFKGJ%1T=u-!2o2b
z-$i<f-+B5VcDdYtz{!2*NB7&&+~<FOyz3rxa};XExjDPRYyO_tUYo^iHjs2+HAd#f
z#&^#fuFc1WQ2_ac={k>9_l-2)JqZ`V^B9@EKSE8sV?V^wws*@K_hwjzTx#b(rMe+l
zm@WJ`DQ?`|bkt%*4iz)mX;)^`tv5OXWzkI-5wA@hZ1TpeTg!Y<iZ~iVG=efvFZOWR
z*e)$Rj$5t48{P5N%ywd6LwddY<)iw}#Em}TmO0!jtEYTuU*n6WAfhDJDA#3h+aIqz
zYW>QoJZCxLR=>`~T=s5F8Tv2feQeu*C+dzUsq~zChZdmM4%so8dJbpO4yB-dFO;@s
zgcd}X2kzB(<Hmwc-^a_Zbzp|%7n?H-*bI-d4wss-3youti4{~2{CzR@_q1Kc?!u7t
zg5RnuQH(e(oDRbI3mxcgu&Kv-E$F-<eRMwuK=>Vk45b5Z!bXd1GB4pLnS~;Y-wl`l
z5$A8av$qsOkUnylquZn5psB`%@cfM6nGid~WvsMIprC85^+gaMzaK05st&d%K>LMB
z=)F9Bv0>@Y;P+Q&Q|Z=)<!YaJBA1A&*4^N}?X&&vYix18iKP5*NN~oPyUTn7W31fC
z9k^yH@*D+vEjv`wfIEku*Lk5{8KoPQnW1`X(W_h;fi28KeUM}Qc)*$<&V8ZDzf88=
zxU!3|E{pCscMFjp3Q93vfM^48JaDbF&(UnsFuGZX!`YQzU(Ef3FK`>af%1JBCXOYP
zmq0PVp6ekbr~r!zqrSC2`ck;LAM#X7Q>2;h3sDE-V~;re&kwri@&0{DLNb{2qy<PX
z#Kfl)a#6ol%h-|*Ei0Bcqm}>!pITv?WbFffOAPPSU6*J~IecR*!AQ=SuBY|nD%yW2
z@a_)oBR5qdPKgI|^*3|+By1{uaH@s`v)73B`I_(-<1B=5IsmNWAg8(*05QiSw4UmC
z5l43{D&jO*GH-Fu)<vA)x*4huHn(UA22C_=VJeJ*!P?R|RiN2VYid62^c~=<V)j)`
zJ`TN(1Qi^Xz#!-L8Gj)nP#X7q-570`xN(lo#%2i*EgWuOiT0)IKZ!SPO~A%YegXDp
zxYhJQE(j&^xBcaDAp8vEvtH|yfFpZK80)OcB6YCk#ToHOPL8R6PZ#v8+*aVWz+(=H
zs{+4b2A)2%5|BWMtY6b{>hJ5;m9?BD)s!qWZS>h4++N2r4+*;Z`eKTZ*Pu1W=X_e<
zH!rr}u_foC6N7&r#8=IYfdn;#e?rCW8d~1M=zNV|VlA%g;S4n4mm&jb;q;Bf#YOXh
zMa)9`5u2dMDBvPUKJz@7Xh7*J^2Q^U!7>0o3@``@wVQ*W(iO%srs9MC7{<rS+80S_
zwS1v*8$jnwKWgzC9>*Kw^zcOP>c+WLniHW|Q4ei*DKrBGI-|_OHxw~(Y8)P|#h@3V
z+OZn5V&JlWQP<l;mS$Mtn}Vh90b7MPB$ywW)m1L7#g0JUA?SM!yQ?OP((*-q?K(-&
zN|@F-Vjo`i1VzI`km)Yk*d0&`I1c(B2f}2coFvw#&P8$WzF5sqe@4vcrJVDMyz*4F
zBzrXgnHb#NGi<VJF+U2$F7HfN`c8gGp_{c67#fnpes}5;8a_gAWXm@T4`ei^o^xhU
zOH4$WqBn?a07gtd0&EdV_=?Y5kI@^0M^-O41Z3%8FVr|B2>vDlnqqo*>&)q23BvbP
z^%VCt?G_1pzKuXd$M`Z3qSarJ30ZR=kRiVad0@f#Dg^$tV_P*zIP;fhbI@HmRaJSv
zv6^%KTlQK|v0ulMbE$L~kE7d+ge4(#N4@7p1D~h>B@K3iacBU7-`y*5#BLGdu#r6S
zMHFiY{GMlK;qD3=QZiFNZ;T7X_TpP1P7g9r9s29Wn#;ww;S9m;2Zz{4nAru&U+$My
zTd8&afU{L=xe<s<A`pHR<|vd?CPaGL1i&D$wllmqWuU%)y6C%4WCP8(;i|?pK%!N1
zhGfjg4?`URcDM>Z*Wg)G4hf8=U2*fGHT*B-eJM;iz`)zH6m^2ej0yB-EyD%F_TYo=
zVYgb<i#f-uy70%u*Je0E=dYIn%)b855(AVNE-Cx3vmh14FWJ7;I&S#bt)u4!(PV(K
zyr2}y0Y~7`MsL99LoB)$yVk*FPXrnMb=44CMEFOn%?Nfuqmc)EU$`jz<Cz(riALz3
z9piL_ZtJIi5xV`f1M<tW`e9e`XXu!=Ji#Wjo?pCLn*wJ3`WI;?X#%Atfj(e30jGpq
z?>NVmCdRsHZ=fhU=s4bH8k=7P7tdF=pF;{Qtcl!qjtic5!;)8f4bfmS3)_@W$YD4P
zp<!`vu*j#}q}))704cn1U!%kp!5uw(><k($OO5)BRELlJMTNmJ^5y&menY{{FgiK=
zBH8WG;h;Z21HmExmUR$%pz7scY(y8!5ta)!J3xG@rEl-O^w!Mde(T<O@1rT^l4a}R
z^Arw_zHo8Ew{8=Ge&}fNaKW@mhL11Ps(U`atTc6<VjtIFdtK?c%F%NuFesqrg8rPU
zSFzO-`YO+y>E@{tM_<K<?yb|BqD`>SvDx8JmjWg+j_AS~{vatJVN^>S2(`WV1A0K%
ztRaV$>pyKMT4TGkYTzH1xU7DN?nqMj%X|hH&1yQmv2=r@tR#eNEjbKsUPPqI4{!(c
z;kvQV?V`kuY_w+M%ix1`V~~WLJ@WVkPG{uWjkLIXR~@OXV*H+ch`QUP`jFqFl9#79
z!MgD0t!-dfZj|zels3Mxe$Q83amL$Y0L)p|ik<cI659&a=Z0`tZ_0uA!_xh_uTx)i
zXt-s??H+Lir+hGSQ2JdFkpOcC-`oeW@4x<(2!&VNI)5F+XTpTH-=mh$WRL&X@$TKR
z_o3^NzEHKzoCf*>SnX>xCxTYxHHZqCPxK{@)~6LJy^D~_WCMe7!b23W-T3pBK1}fy
z34_^W<DX+nq`6BEc~qZCv|08_<ZI+V<ZS)*@hRcmU4a*}(OvJZmWH)Ic3`uIhxJJy
zv3x^}B8GRg!<URXt$SrfXe8ou!$cxPz=Ps`JIfCW5~M{-uaKtYcs1g)M6Ed69^3k<
z84x$jYL5^jdcf@hOypz(dz`qgmfzPUX8F;?J=GM#+=v<&V!N&XFQK#9*Jw=y*=J5a
zIsChE&^OB<{|iVcpO6z%F<)U&cb))q(XZPjAz$IJ7fH7aj3{RAIo}jB$LPD!<kY{@
z&#hW|BD(6Jcjh0{C>3d-V~ku+i1|J7HtCL~lo6cF3eI``W`7#tHb=T-Bzj|4!{WC)
zuW}|E=V!7c|8zELi;1Y(cLKm$LbMW=EG+guR*cs^()<Xqok|W`wY2^#J8V{n?X3QQ
z^hL)wQLF-IIxj8L|4Q1CXzV+8afi;tm(U}OkiA@WlJKWZBAWnqJ%=g~W2s4fgB-Eq
z6P@aK)}VYAj?|HHDA_>k#4{Qte-X{h<kA{mToD6fF%6SJJj*b@gXgZzU8ew1rSGGi
z5@B=sGlsMjM@uEBheOd}uRhwoutpwNRvFQU+mPn(5%Ec|kLaWpzosXWDuk%lT;jvd
z8BKJw60>ke!~p<K0V~xTftZ~A-Q80I4DkpXzmx5V-u!c-CgT9KuxuI0NJbUI6mm%!
zgT<FuIKJ_{TD-wSUj2c@Xy~0}#f>MA>|`_O^O0qlHrtL`eXF-YK8c3U2{P}FFTFqt
zjIoX)Y@5Y2A4Rs~3omjUEmG<4DU)u9doo*1hQ@N2P$~6kfIsXu)xhRy)ETv*4n}%Q
zF-4~uRT9HNHbhPdH+V=z0h?Doj0VTC*w<(;Kqz1?dXkH$bb4e1lZBhgLLh&sVdu0W
z+^|BW=UQK7h%KI{<@E5ph(we0?Fz-xvh<t&UH!$Nu>RaQz&o>8GExf`-3CV8>=AP)
zWy(WiioB-~{$ii_H~BLvl5P|1)x5_7v7+T@1#ud)_^A&bpRUKQlzb9%aaA&nR8A1k
zOu_f-`R%kBQ@se&u&C|aZV3Lqg=Q&~O4tx}DFW{nd>&GHNj6~g>|ZeZ@lEEm9M9q@
zGnBZkH4ssCr3^69mm-qH9e)OnkNy?(WFd_BHe}QgvPAvXKKvp)ba@J|bi7fl=tmye
zDLO&gplc@>rpj7tAbzKj!Ft=+c04L+(f;Fr?Jeg9nC4W1A$PlNC~h-m=378wx3m82
zeKO7Cdh59>wcJX4-)@%nI~AjCsM{tDUZ3onr^w(a^rarE)mNpF6iE~q>c4@pN!I=Y
z9sjN+s)fQyM0tiCc#I@Py!Bx8H?R7x7kqsGC+vkn0r#ncgfSDVory0Qg~wPlL#KJC
z47`?*Sr|BQ2C>zHS<g}nSg~9gi|gdgRpcJ+(o~lgQfxQ{mNfg`F<dy4PS`Mm?Em7A
za|TCtZ~lq+TBI0mukn3!Mwc-oGKm6K9!1pdiqBWu6e<JN5eLFzf1^=6Q6pT6kVGhV
zB&s$es37kz#Fgqv5^>NcCg`u)WUMh;re7Xp79ygi6Yh12OHfM)YDUzSsy6vZWv}02
zkd?mH_9>6NE<%4=#mQaMdV*O`5~e!bk8JE`Ew14`b__1Itz|TF>cybqd9L3*N~$YY
zSt>0FU|l{MQ7^l=x#D?ol{RNFl09ef^J&wquND^%a<K$jR4-+m6vHz~D*CaaCr{XU
ze06@=W#@$;1-Z!v9+kXR6dB&(_~v&0rJo3LheYsjPA$T9fm4X^t))uLsQgv(ZT$hF
zC8TlxMVMKB+^^&fu&n)CJIr{dm7IVuFl4FFiC(izRcvoc-4%CbWfmN{X+b(yQb}Ku
z67y#Ei)>*Wf3Z9thRlW8{#lunF8z$*sD6vqKyqQ>%ejKuA_Lk=Ak1o&PYJ1fe;&;G
z@r!(n^`9(tien`mPx@Zv8>)|U<vVUDzhA&Z7Z$n9Wv`%WCN#L(v~}jFgCG8%7#U)V
z{rZ24kiJjtw@z(mSs=;^!Hc_yS<%e1M>KmY@Y2>^7Il7F72K$Hj@Z^~mC!)WLR4&(
z56Nf)cU;U<3Qa8GYV=EfuIY=nlTnE_>eb;gcn6v-{m=qL(sf_S$VT(=)SH!!^(F>g
z%=MnsDdYnbH=PUoCS^mddleD}gP8@YL2;Akx4yOEIg557nI|F%PO&RzZB-Wq*X39b
z%)yF3&)bk*g+A4*HqHM&NHu9IjL}jT17pa6$55514~^DadOhZQBIcq#?&^<C2;?Ft
zH+3~Y_+f!`cHy4AJ_SH!c;}1X?~sVC08Vx<s`ioY%%DDL9SbQ&4}wW9XA7NRQKU8Q
zRe-ssb<gpbe!Q7W2U~e{U&7?PrlyTDUzzBOyJohY{-3B`Yjxmix$2kXBDP}`G=?=O
z01<E8<v;R{ZPd{5(F3}6m4_eaO0Re>e6dE};mJ~n-*?KK&a?UN<MG^dX5s2pKT%8#
z{8+~&{h3>I#|NwFDsI+0{Hf1!>!HPGia;-pk1yq8{Dz&(#K$XY<g*pYYeES0Q+L%<
zWJnw|<w?ygRqR^WN7yf;t&hz*$vZy0G0irZ%an;S8vA_Tb;6bJN-b{hI@?$Mt+<35
zE(fz%ziyMxSqp`q>rXmUleib^uJ?RWjj(`yflp|2nH=94HJ}@_<$Nu2W4UgTtCX#^
ziq>7E5H?wWe3Q(P3=e^DvZ#pg1yK}Dm&n{cIJM-SZn}G=ycn3lZg2p#cgV!`{{HVX
zYyFsF{AGlnYv)9S2VaqaH9}jEJhN`ak!n7Hx9EU}9?poEs&P|7)nw#IoO?%T%Y3Nd
z<2Jjo6YjpG7&ZhXmr>eSx=^QNkMi1IR@VYw%T{}&eA<4ipy~9!fFR&m#d{_$$rf|b
zYpBCQ_4Ks0hZa&-4!#A_3$w=hUARPSn8j`u3Npxfh_0x0a|Lgj%}htwML(0xxzU!u
z+IdU+3g0;+su`@}w~A#*!(aJm?M5C-rzml%qa@7fc`wA%9pH*p6mcgRo0caVXd|N^
zKT-->or)`g#k?A03L{dV$-hB=cnS$ztPTsqqxk!rI${#zRAsytwLQCf(fQlT#x2x6
zhsIRpXAiR*I;h;$xJq}?nLm{ZO0LUMjeHHQmm^uu<fi(r_GDOJ(+<@{m;D~cPek3|
z+>K^7l(C!Y)3a#xp|$sGy6T?}n;N(o=wt-MOQ#OZEcnaVKIF!<rqptR{C9!skV97P
z7?LfjYVcylGWjg0Y<ed2d+$3cgahprOqDWKVp#tSf#Jpiz+;!eOrFLRdh&jkGs5kZ
zLJ@0`@REE$>-U%c0Z0d3fl(_VcUvb+p0R2`bRmwDtWF7kT6aD5B0H&92821Gfzm62
z7ttb9Y>%U*>X@_J2s(vN684|JzHMJrxaH7H6<9RCqrJ(qMk{}Pq@W{i^dp0;&*PA<
z3dVfu6w1^A)r7CtWrw406vMi5Pu{fihrIv8K;AQT&?HTg6Fby;v{TE;j(L>3fdKUU
ziRG-{T#U0gf1b+>y57c{q4XFj7d9U{B63QfK>Z!nN=WGn)p<MfLiy~#q-n5qA%e0H
zwL`+uw{4E~&K=THnaf@5v!(P__@9u=sG2yhjw37%iLB%d5wn&Mn}UdqC@4;EmW^~V
ze+7En`u62AoCqOV`wF>bP-j5cR{VqC`WeJ3sKjP<1jX)<x1^EW?Sm)$T=hq*>gDI8
zAGZ-vOK0))bNiqK*T>X}sfgH3B!IISCDS^vc-57_IXVmebQEKGkNuJ@*Z1-Y(Kr;J
zE?e`^egAAds=%YS^;JuT{h+_{YyVS9ROkjbXpNMdF73-w5`e)v_F1SxO^y6mj~BhQ
zpfXi@HAKXmWgH$hqmN>L@n`M$o403Nf+X6m8}s(x#}$9e54KPp#>Xk+65Ly<J$V!)
z+4V5ZoB43w$cZMOj#oT%79ZE<d>J_C6dv!;D<QMg=g0Xj!%{vN7s^qsXA$CVpu#f^
zH4X5j%4cN0X;DWg5A_TX73zy8eTi`dy!uO=vssjV-xn%hXVshZhwaaDbt9+NYIDV-
zKNTN2sJsqFdq6dnxgIAuG`Ao2yqh>m-ssl0jCAjp`I;UqJ8GrIrHiqU%q!*en!NjY
zX}lw4Pnik|qRzr+Nu0Ud*9QFZ6Ri4h;C-I9ngQu|WBd1l!v%gxnC=I4?g97!_Ylp_
zxLMMay2Olg_)_h<ahRu@5^hhPyJ>0h>vj3>nLZ>|mBP!fhT(<E7Tb=(FT)z(S(FSC
zlG--HD&B+-m9dn_xe$F#w_2BHCPRVtGcg4kb<Q&pxn{8`>q2bPNjlSJU$&ywU8e}5
z1GxjvEu(kqaT0<>9tI7PpcnmB<<KwIhw2}Z)asNRDk-w$S))%Y=xXs?R<BJw>6Mf`
zl;@WVqAs~tY{uqD)E)Zqyv{!QjiXK5(!c&ISMlEq<`E<VDqo*JzV)VvPnf6ast1dQ
zgUsmhwGh?S4~5kcWtV&(MbW%w9FeC{s|@KEPsz#q>X1727B2?03xTnWpAww{?Lx^N
zVtPfpxtlQ%^~rvZP+DHjkffK#SE9`pb(#e$O*;F<Dy+A}#sXa9DK}T+w`G|bFZ80F
zf1gsNgxOMN(Wb_g_>rSZTSx|R^xv$yIay)RpHdxZl0pdPmqVN1eX~$q08TS^;VXWn
z#RJjYiN{}=w^IHG;wb!Ek$Z2Usa-tLFf*4oy!J}(Is(X@(v{Bljn&HuTv<Enxym=L
zZpA1*JyhwO0+S^c%g2tQ$R9@oWvod?VI6Vl&zA1`k*u_ptBU0c-)y&kzdH5xVKc1<
zS<3)y?9Q+Nc{@$pBObFN+fm7R?@6=NNLwcgZ95#;@lk3LgQ#reOAVQ^A+e(OtPn<(
z=&*tHv9DSss_&=4_t-rjw(FKF`~P9nWjU}_dAA7hIv0+61Pl54rYO#Kf?txVK1wE?
z*%uB)*8XqlhX0N$jlMQ~+p+6;*oP@~0b2ERF!@N8t~%AXgC}mvk5}%G)A8n}VHuVN
ziB7z`^ymZ~xBm&SQ(Q~UYnq>d4lMVr1*;aGS*Yl{!T{mAgY!}GJ*VzD$A3+b5*^sO
zyq|fU3Kp4KEXHT=I8Ro92D6Ap6<YOD#h4A9S=kmIlN2d_*5{MN<|jQsD^^3VOkFLH
zJWEKbz4zx#wpSSVX$>m4-@Y4KaeCI0a}ilgFDHEsg<Mx>lWG8FU1_V1Yh`sn4A=c1
zuJ*@679l}}MzKb^K0th}<EuJ(JGb8M#M-9M%b)QRB(O6ECi|@pFSktNIVl_9h+5*M
z$UK|<*>2El?d0B%L4Qn!OPD5a={KQQYKRHYL_3#MpTf;b$JvXHDieRH-9u>d;)#_B
zQc-VxIw<9Ni(?s6B;a_dTW<7gmp7^6f?t^OM;gq^=Z&La1VETUV!~Q%w!@Q(6;YqF
z&HcO1S1wU2DSg)!HFw<at!ZvrDpo1N&VT<KimeED*LSE%oP!@wr)%tgJrf8j)qe8s
z{l+!|i&PFVy3_T#!CDlq4I-JXDe)fAoEI%)d2qS<s@ujvxAY!pYitm-<pw?W)(`Lo
z8}C|CxQ8|DI_VlqF^(v2aMrJGs0*K0q|S=568OlOL8R**V3nhNRbyH!0&DISt6C4T
z>|`WoFw%r4o`So6SDMQ8>`AuFCEEcRKR3)D_|1lnYrZB1J94g#vTFEQCEyaU2bHgP
zf4LCT$G&iKIyA55V_`cpc%Elsl2!_4Ah3SYjF2ktY^Oy3TD|3_+VN>Uf_ef<w|<$@
z9bQm}GOHS(k|{{@FMjgqKayoN_dqEDBI7SLk>cr-#|KLtnC`uvDtSQnLVq_jR=WKB
z$G4c`^72J$*fr)yB{VR9r(zRN1d!@80R9VeUjJ)1J@Z`7sL?I_s^W)VUzpJ+Rt@=>
zqHu5Db#(m<sQBT1^mfNl7KCF4_s@f4az?aF%dY17)aA1M?o4$c`M`?5oa(Q@LASmH
zW(@f8%ukz_1^}j^_~YA*XlXPV6q`k<3A$G#%WNR4JvpC8=Pwzk?~*{F{rp>rwy6|p
z!Py@Ns3tTk_FN3J@SYB~J;fEJ#LRKcODWi%((&XDSMJb6?t4p|Na#cO^h8k45ql!&
zWnrGRRGQk!PGy`H*{3nTlD)pPjw<oi=-*o~a6|B?E#BXO_r|LXi++r$6YYX=uMNWT
zmEoJ88NJF4iG0n}52dM=3b!X43Dma+mFHf0IF66=tT(E4c@;h{!CP2cXr12mYPqrV
z)<XDCOhK^R3hhh!fqKSeVtBN_bBkfZVeiGv6SJ0!St@u&<WV?3Rnj@Tho-YJ)8;kY
z%g}uu&Me>AFX~3}(VNq#i1dvgsfPRYMGv<7`ugRS0^ny|P`SVw!68jeO@qaug{R8D
zPX99{_c5VuF4hb)YRVyUe5IH<b1`}7yJ{aA+l?0y0)3x|!OZ&OL;z*0*Ci|oy^EgJ
zc*B$s+rEER#v;RWji0e;izmd4^v}=O?~6$fgo$~Lk`-d}!M(p#k)V1rO=GTz-;QvB
zvt1_&VM@z^K4)@0+LN2BSCclL-|Bj|dbU9Y<0Xa3Ye{PT91kIGbypBjbfPsCb=Ufk
zO#T7G9{~+*3Qf)@x+KtvhQd#RiqP(47<yng=Kx-}=6kcltOdC1R2Jw|Y8K61E=)g$
zn+PhXOQWOCR@8~by-Z%YK*hepoO~)VOu;80e)B$Gbm9DXUjaI7;i}aL3+;N4)%&Xh
ztBaB8iRNo`;2q@`Rt-1?bID<Z)V0B6qswYsa@*v&Mj*xm?&N#H%MLuzsb7<752ShA
zfd|!}2JOs@o@PFH4wK??@r&%!oR;8isx^A){}p^)A~1$n9?}8ktX~nqGxTT(%UdcO
zF5^6Ho_3LKrJvwDF|#unu|t5>+UDmp%=Q-qG+25#h*1zcG_hxDMFA7~XJyl-;E5Jk
z@C}|#?JXx(2@RSy&TBOMM-(b;{!`aOY#FJgs3aYwOyDnJDb?f-Wl68zpdJ=K(X*`o
zs`Aru{e?!|hGB^*R@N$FX)T8N59<5Z3bhEUx@QXWMnHW~&I?6D48Km_GgI$p`?qd4
zI>_le<%_^IMe?c~3r)0*rLIB{qF%8BU+3<Ei+T4wff{LaFW*XT+zPJr2S|07!-j4W
zD&2dyG}524%+|PYHrXZ87Fl_Tl-p*=T67o1zHRjrJao#Qc||JO_H^7QlBCX+;TwOM
zlI#VT!4l`slM|~Zc78|2E-hjoj^>7WP^|pA^|i!O$$oAnD0{}qJk;{@AVh%rMQSl>
z;uY9SCqq5ja62fOz_2p$>h78#*+4om)5erAWG_H2a97O52`YStSba`uLUpv8dbG43
zdN`O18?YN`bf>Q*-(I+B@00X-N~o78R!0<q(KfT*x7uq_;2H%i(C7d`EG_B>&(gr;
zp8vxC3ts-NHHmAy5)$->X?W+sXz8fjji_Fp@>e|WW4@$TLP&Hv9C9y$TH@~Meruv&
zddK{#+Xfz|8b;1A<rafyx2MrjqHWK{jMus^o{9x(eG~_peoS%V+g5l!wdVUJR7~+^
z$mcg3A4c=EMZVAH>^&H*AzLx8ctCICRhs0gyE@<pNRr%zV6xuGV5|(^^<!zujrd+a
zvXBLg!RNm0&PqKUj+zKFx_EWVcp=nLw9;gFXF=)A05PE;t-Jb?N@^ePj8f?wQ{zPJ
zj76v+IF>7>xSwngBFfJM+7G}={(I~N&L)F_YuJ0MNj(4t!ab2>o2g*{+i9v2_~4%J
zzz3{;7IlNUkTB@P@d2~IkpJJhxRpQd{hYDGqptk_$aMZAe=#~u4hlkp2EI6Lb&4Pc
zE|Hf8YXK2*dE5ZL_}i|T(~{$(Ai@5BWN*N9T>@LS0@|o;CKdxq{w{2KzYsOQ*l*F_
z^7r4ehUKFxiT9o-(Qam=Z(g|>B3eZM`RzxVj$=HtRQ-rXvV1M;>K-iOczaUROz%~K
z?vQ8d-Tec$&^N1RTzkrvwwiCLH8r$(Z#Nyx)Nka&e~{`(ntE$G8#p~!@*E4^zc3Rb
z=8j3GLv5MLSKJUff>n)lEZu!UztLi6yo@^({ef*#4~=DG*Zo;*T#MCy6F9YR*c*FV
z2Ddz^+o&1+T@iLr7n0N~Lth?n=b*EN`!4=Uf>R(T#|(4NZ*GPRf6{M_+RuJkv#wN(
zOnf)|CbsJQ-o0=mYvUWev$+X*;*4Pcb)EW)s#bQ><z~O({!C+%%R@0t?AKmmN_dD^
z(!BPnOE~y@0Lg7~Gw%t5(_g~DA)LyemO>~w+^=@O?B1f(M-BxZ{9>MW|6`Zt&2(|T
zA-}pIfB7W%C<m|{3qI$5ggAXGt%QtWmoYmROH0-cjrG-ruX-UU2SAcrI}0nLa>Ob{
zNfHrnCV(24jWuG&(&R1T8!DbFkUgUtlam}cDEi02I<Kj91V1i<pFjC8%fPHpUoICE
z5bykBA4$SPXC1`Cqw4w)>cM;N?>F7Y%|<6kZ+1!aw0!8)Y)3n`>vJ>R3{y(@C}vm|
z#6&meDJ@kN82O=7gfwCsqMp%!{3$V|Ulk2vpXI`SKQ+7BhVU;ehsL9z7<o$9GE{y$
zxzR^S#nfUK16HW~ocNS<ym}n&W4A{R?nU6#BS5dV@xbaMKv9ht#<JgUz2ly1>uBCJ
zQP9@)oj-N#9=xIc^P!G6uLd~Wuh4gSX!Z0V=lmJh=vC~!`b!C(skE@8b1XN9_wA1j
z7ZK&B5~Gp)+-ZSwqs$pf4LLYE645K-d|~y7w9gs)cdj)+4V#BxAYKusH@2G5>w*)v
z?dbO|PVZ3}xoi|)mnbH1{wZ#@c>=Z=@iG@8c);?OkH)m6hv&M*hMg6A6Mn_&xoj&h
zCz-moJ?Bpgn`++kKHagC|8nh7T`MCK`1{}lB{T+mAd6dzzae%te-St6yFcI-d~5Nv
zOL{-CJm{+YX&ar6o?fV3x__5Vu;slV+2BUrG08(o7_od}LBul}aUwOFG`J-)6MG+s
z_JnhB`^T_wF3`hlB7+ktWr6wU-@3`1Imfr3h$x$n(f6L~C~>dZtLROTG(5R!FlNv~
z^a$jlAbU;ZsYZjU!&QtnXRjpPijKV$RA1{3z{y%bYcEy%tYz-^o_#uG=l3fa=XUw?
zL7zL{)iks!m^({T(jx3hdM-@)N|n>2ZAT8MEj-@U9r?^~9q1ck=$K4*b^Xv_36|UN
z<lfbq1xS-cxQ7wz<jVB4`2K!#ZKl-(4ZJp;L}T&U^gSg~efrap&i&{<2BD0w{U2ZG
z=MVLG>=3t56*+N5Cx@fj_Yk|?5r9Veo-eV73$=CIRDSca$>amCr+kJC!4ga2y!HYv
zGWd!H{1hf0yvYU&Od8yL^U1EdS<JPCM*ktPnlx`pl#x1%2Fq9~%}259B0RAp%(`jU
z#jPWfKRu!xOxS*Wd5)Ry69xw@+)n0Ols_KVGw&kytu%_dEyzB!M9fV@(U{<|X!HjI
zfOmso_2bXT4&Kt@2Or7~7O)84*i2|`O_9l6)iFo`Y-S#ONpMxNCmWbgY}t^50yUB!
znSXW|gBy=c3lRM)Nj>A?FB+xfN2EXCl@kU{Xrc*DCT(Q@^>LrH2<cfkYx<Yj65U|@
z0A7LEzvT<>#iU=j>kl+E`Xx7MgSa<S6>ao_8LTPp8=W@ttADZQY|-wO`C8&nVakEH
zi7Cf%*T{JYWlOU5c$pfKsmWh^C81R0I5_U*ZHd!4XkeATl9Os^LpeHX$dhWsWsJKw
zT3^cr^_BU?C{CTaJrAN<N~m!rHwwnnQ!Teed2DIF`S|Wx(Y3nms%M8*lT8xuhik?l
z{ho(e7)zDQ@aPCODNa8h3X2r?8@86dG^`s7p(+@KB$$pvQRdAeK7H)?2d|D7x!Pu-
zYiGhaR!3~QWxKIn>helRj|jJOsqF4Dg`eJJXib83WzZf4?+eLSrEX&VRa8EbyVt>i
z=Eh5eU({o3aIGtC;!b=ot5n1J<bSsqyTFrJ!djk+h`&tj^^_bgu}#Z*8)RZKHvPAz
zOMKVz88`zG!*YvVSaa{z*492s8$2_iW#P9R-e@zrL;22EVXn?22VZ>Bn<zL>IeW4@
zYyQkAex~nH<8?RCw&?-*5(`cdeb+VUv@uj}8{Cq6oTN~h!U480O^D6ElPpa(kdzlV
zpGu@M)1<v>UVy+~?)#TJXX@1FdEkF^dmW@RN95ho2ScitJprFTW4CLY{Q-Nl5`o!t
zdXgNPWR!rSa4$(CN1))a_3KE(>ALG9-&5n}4xBoZ!B;fj0VASwv{h_&<hkp&D%HCQ
zH&EI5pC+9=3U?(M_=*-{@k(WUgHyMqoK<YrxvAi3A)BAjob^i`>fOQQBgDk7-+FS~
z!V!BeT}G(&aePFdDFeNT_^|$No<h8X?f2<cX=5P{;ia-M)*guK8+M$|*9SVpB#E&H
zvV$to`x=9sKkqkF@Tbks!Z{xOY85VPjr8WH<ppNoo2>kFigkxpDs1VhKewoJD+d*W
z#(d+xUjlhLViX>qz9A+`V%HFt&z_=LRfgn0zZv@yLAua#_JL++Vyf~9?ezUx<KoA)
z^50s1HkHl{(u@_vxm<;@ivEfI-4_@~xAw(+J;Zk^NW+M9An3{7_4U=0*?&sMraFe1
z;!AqAW8IUQ((_?x5IV4ox&+9opIYcb2m2Ot9Nx*C@YXDN9^grAm8)YNUA(-&rabmo
z8FYun45hYvmC+QHJxEC?tSEx6FJ7ETPkH&YeY`4jtwIA3A=WZ)fjQWD@CdYY4L&jM
zIslV&ZJhDA00w054nP9ZgGDk2Ql*WKvxWe__S4@ITQsc%*n%srNXF^<Kj4@e1Eti;
z;S~KxlI>_Is`3x#MlqWY3<n^GYCmL+`o17r<dW5JD(lnI;fs3ydy&BFt5m$Cx|zsa
zM*%jQMKZ=Rh}A8D)THdAP2T&K+OK5@$LfxF?z`j6d7NbQ*HN*4n_22Ze%Dq~!JpTl
zsb5`#V??HwU`q`S5@zBPTT^=}s^o$e@3ybGp2dvH@pIShuSP)FBfXZK;%@APR-C-i
zccfF6U9sm$1~XAxJpzS%n5^iDzQGokuFnP=9?z}OD4@|7d`|E3vVquztn&s1C!-d;
zKL(+p6P(pG9&#&$zN0rI802z8lNvhWt^=uhGA-1J9?_j(p56^G{Q<Eu=l|y3Xye6)
z(-mzZ*FMZ<=)U(f`GY^9mR=hW;*R4{nKk6jj#ZurK7=tMjc7-2qo51|+={%QoH(#@
z_!BCd3&V7}N@O!DuBNkMhr_FUj#+f~j7+i8CTICj7ADhz3%(qI9~R0MmA$RB9eCer
z^l%8DeHVXv?iST0!6NCH5P}3I{dKJ<^lbI73|s1;8_~e2<X89d6+z=W2G!8N-Hzt*
zMj6iueWR)Ehb(h;7T^*ZD2XAIkf5j4#vLm=`u(E@Zq??qT6I<t^~CR=Vv2beGk9X5
ztqz#!I9d$MN|kbXAO0NfX&v0iofInw$e%oB4xE@P@~;TWb!m}z+tM}>YRZ585;#}m
zl3E-+c26&0a3bjbC6ulhufvy7D65eHWM`9db_WiilZox1)9L4#@)m{OVZD{G<G$Em
z=Ud6Y?1s5DuuUMgn%&IHEjFJ|Q|Pb4r-8o-hs$P4qq&i{e%Ky6Iw#Vl4$ULd5UXcC
zT|fFD3-6~;Nqmbo*sXI#z@+Dc#V40NfVqXPJQ6}AKSy^IYRqgq(`CVJp=y4oWuEZ5
z$=17X<p7l_ikL5bV&NMlou)VBE;r;ny8&UBfq~uT{LQu&=77^>j|t6oQv)@=@IFB0
z?32b>+nJeN+CK{w?0M{_!xi4*lP-nbNcQOaE&(@sR++PcaLhVI>>sL-m)xz*Nfi$w
zWbUrda5G#7-WJU&d*w-$y^VmlpRoUFdt5UmXm0Az5UbR1Vt-MW^iald-NsY-7K}+t
zL1GVE5cT-RNag~I^$pHeT9oEJAj!JkiJHXPo0;W^F8V*VQA))2_BUVl`~$gJYYx&K
zdg)TP*;s@zgD08;s7LVf2A8)clPKIUB>HIC2O(oIGy0+Qx?p))>{?dku*Avg(Nr4=
zD58yWI2k{35s8xhFstFKnbA*y&6CIHNSul!Y{jS&u4c4PHd1C=ji`*bRNb7FI#`ev
z1iL9Khn_b=rxrhI9-EH5Zi!y*<v+InGicX}{{!D>b?e+c*4VBYC|GQuj;1{XfQ@Qw
z4(<<V%GP*nQ`pg(N@oYp^3|4GY#ea-CPvPq2>`1XkcFXG`HQ;Z3SIw4a2LzmsqT$M
zZ&uGYo5NWYF#PMkY({^-;1&;%=#)HC4n~S+cz}eCm+wW4egKjV5GNfPG>TT`bOJEY
ziSERq)|v8_&2u8#wG(@oBe%;M2SK+t66F_WbDziWC%2scl8iIhU2j|@xHfH@qu={<
zGN>(`uEzR$JSY|TE!Or!1F!&m*s{Qc@xXZtF(yw=92rfqOS^t2;{^|_RL=zN?!8q7
z1OOaB$bM*<EuJ_e@qWmp`!?wH++HM-urY<X_)3>}Nv@8#dz)YboU;i$j6Q#A)lU9*
z)*`$(NKoUY3*Ig>&l1h@>~g<05IIgM;xgJ?2ky#Ak;O1*q(14}1zXRk^naIP!D^<V
zYWTJ50N6!UfL6-K=VNwnR!X2Lq7F20tt@mM7^z>TJ#a4P0_zRZ-e141ks%iH5}D@E
zZ*+uuc(%iNrX>}PNa@^<9J6~%6AN`mKh~(j`?T}b<^z9&s-PxX8&az|m$sveWA#Aq
zqTZ(9TX*)NyEWehjloeYyRHo<2`CLUj0dZ(7ljL-=D`CYJ^Oel@_7*O=YuiZ!zRdn
z>o{K2q03q$t>-jfqbsThdkEa?E}`^aNO&I9bv@&PiX-iugyVa^U9ro|b!<ug$`uRn
zHw@)a8mf=UMXV!5%~mV@GbL2px-umRTE$X@YFb3;`JLxRrbxbH;;^2v(?kHq{L^}l
zM!4zbKA-)_c=q@(I>L#Li>i52(I(s5L`j5(S~al^a*yK=+epFckM+-PqS33HtJEo?
znnTZmzpsoM6IhU%#9l0ODf6Ms=|c~VtXTd-L`LH>UBP$4jZS5%gs>`mkuhY6{E8xo
zDx;Zx;>hl?Rpp*~=SF(OuUfiwOs3^J84v;;;~!ZJayv|4zg2xVrO0VL19VL))Zc{(
z?LFEL!QBPCKNy7d24*~E@a#(avOh=1dFOyarJ4|;SnfhGKyU~f;H2CmAIRl<=}T;p
zDPcJde%P?a@3+&%km55%n*uy=1quQKyM$7;0Ci10UPb4Eiw=Pehv}8J;CcyxOY82Z
zmJP@KX0n4mIR9iDkF$4{F7UlOjsqL%$%5Bj2&0>E=2hhj{J>xX4#Ht_i^P9MWN(TV
zrYtS=n=fV;FD7Z%Djg>TD5DY2^fc`V1(p7-{mSsgNq?XQSXWmw_e`JWH5l8tSPv)f
z-64GiMi>DHahQ$OPi(I4-2GcYP)Fs<v#FL9%gbX+k9KPQ$LhLzIYFiQNkuODwBdBg
z4vmt?)i?*!fwlyumQuouVQC3)-ykO!^i<2R$u_t-KV?rEtqcuH!ba9isA$M{_THiq
zJQD#}1~>Aj>=|8*jp94;!)QI)7T_Q9og(&=?lX)H?(}e0uICV&Zi&TdB7%-a)aEHe
zdOd;Vju|!0meso%UuqgFpI8c+eBix?eA{zAol^S&HyHjyR@f{m@7l(jp7#3d$#l#s
zZ4ldK1RniMTzjCSDO}c86tgPZRsDvV3DcB^LGJ%iO=<kva_w41_CR^SysJ%9H}5sO
zcMKp=w}V&x7Ik5WcMI=MN5e6j?~uZ-eU_M%AzOMgK6!Wfz7w;J?W@-<D<akxf2o2G
z{h*V~?=yymt})vn%NzajMWgVQt59E@Y}&ybZm%9yYZF<yW5&^#f(sPwYt4Q#u<__u
zlPboc5<2<hwMd2oOg?twaK7&y)5DY@yWT+4s=Iv~^3gWejsk%I3`>UgRFi50#aQzj
z4o#HNi~-!<%y(+yhiq4BqvhU|+{3JLrv1~a7#85h^N0R#1(>v2ScVp)=<}?~iR>|4
z66x$>IW?yz={i!8^YuEa<9LCJN}akxiFD({%B`*Pop$@TYb}spRlLlX5hu4E*01Hi
z4(buEIQ?m9b+F#jwIUcC`pf8oj>+;kaOU#u!9<nQMbxjU3&5`1>!A6b*ItChAE#Yo
z6{m2;s<zi&82X0|Hust4pI*P_R@g!M$&D4Gt7I*o8+xLg2M2td38i);wS>F85r!An
zzI4+V)<1jUt@dUjZ?Wy%RPD_{o#^ZL8|n|)rD^)QGNNsKzlZa#4z_1E4z3`?EBwx(
zr9bmlpS?q!ii=?Pu8Fvv*T44LV2MI|Jw&+NH#9#X>pY5u?Nd~1Bs-czEfhU;lox52
z&vZx+zxIEdSt$692Rb*o-7%C8W;W~x|9pZ@@z;Zn%NV!jV=Ei#PO+aJUfyun#&>qd
z)vkzMfR2f2jz$_bvH=h=E!(Vau+#h1^0b|KPJ#zEsY0K3UEr%;okq^^);o=XKv|uD
z`$r!(Yi%4K71LQAAD@pVBf>H+j`GV2i(z>vvqidI0DAS{fJSBFyAgefMA6V^H=4|k
zIjmZFy7=*<spIKPhxv?hr_+i_lZ(4~s_^zgkzKvIJ(ua9Hhbl&TVr{khV&M=G$6wg
z{Qypkl68XP=rD+3%s8DhBGz^U@-Lfd^J*bZR}w2m8taNS3Zb;WGic&3&v8G4f^fU(
z*0{i-a{M`;d~r`nZ{@{rBz|o3@oqeB@pd?VV{~4K3RV0nl&xQxkkvkVBWyi@JW-4_
z^!9pYCrC2_6$^hKey^J97F?_S(|{=|=cdwb`99X{7TQ25SGFbcK*}-TvpZX9)9i6W
zj%rzpF^g_C@cn}=)E5eq2RcwgV84cc(T^Y0L$Y$r;&yUTr7|@lmy!)}>hI3vZOmYu
z7N8k{BmdTl`)h#-!L+0n#nWDLQ{AQYmKlRncOgIf7lYtcVwyurhHqtI6_y|pbIUrO
zR6Sii*Ul?rgYPpDZi;yviEyEU{j}N6#q=)rREO=kg1l&1-@t+^(J7Wn$hZP(-Yz{P
z$jPEBV#o{?nc^#;*<^`i&QryRI)2EYiFNu+CUFSUR3z)o+ofVXwfOUcW4VTz-qqq$
zm*lmPiY<qP<RKMVtYN3#CoaT^Ai0JjL)-h`B%S1gUc*;z?2PLh;J>vsiw$!}e3B&S
zJ}DI*g(1RZ)1?dKt@M|C?|;jJa2Cqk?AgzWiwa`@rYn1Sw8~Hn-eBRs2aK0O&a00D
z4(7m$K^G8u^L+*t=HTlis+TdF#K5BANSsO-f49BwGQMw!3vp^0S8%yrNXfPA-6u(z
z)4t8?t$}u+6rrW0pF0~pd-CU+(+$Pd;f@i%1vQ<9tyFcl^DE!BTNLqpH%l$?E<Do1
zuDC~|Y={ZRFK(wd+SyuS#e|juyyl-rxUL-*33KyF!1J4#7(cq!`;dV0X5O5=816et
zf-YpnIOGU-m@H5VHLQr+n_P%wpA8J(0}_W>wWXS6j<krwbV6t2xf3nE`jdo4GqjHo
z0a0UeVRjkZTfWWKc4og9$YQ^zN}4CLS4RdoZKWK}*9P1L9gX;{vMokPH!agX-YUo$
zZuE6y2EJ(}x$8UUQHL3n@of(UA`ac}V_1y77LJrH946I17SWdsMksp$<2MGFWdUHI
zYj|NRO~6$c>Dsy4o0kVZ<3X(F)klh3lyoC<>(~uVheNE6&_cTP-$7i~yjPj1jgfxz
zP%3ehINNefoG>vy)My6`s%JzKC*95ihL3UvkyxWIwfrf3ky8XG`Ww;v)9Uw!TOY!X
z5Tei7Q>YXq9BuM$zCo@0N&))ctZ@kn5gA{CE4b&lj%MW2<vu6-m;g{ad@wYoe`}&_
zHccb`!0B7y6@w(o#M-*?h9xbs9Q$Er+9Rd=VB0~Ze!o*{=XJ`gRJ<;51t+H2w~Uk<
zr^+2Hm)PibZt0g3HtzM41YIqib9QxYPwT?+W~e6Id@Y$RSSvXJ87;bENB--D-tEID
zlp5m8hEp$LQa0Q<kJ2Ldr;Y6cP%J9ZZ2k}B{aD;orw$z4QCXswbq;D=q{8LK48u27
z^fc90D5{sG$~Te%k|*6%eg-qF*u;+<jYr?tWQ(>wccY)zc-`<_K^j@A{tS@V_J%#0
zo+W*T(#Aw}nD5@(*Q>1;5!$mEc<MxNh+-%@*O9dM-6-qC47<v2*#!3hkrk8nMJ*W~
zQyQwYgU6B`iL}dzEHW|fKQua5D<0HV+Eh+Zo3Z%Lw~^Ogp`zf3SsK3**3Zo#BFkFh
zwcEXY!rLTT7D8+k)EH|8Ugr<mI=AW;NuT26@R7(~E>~7}*ZOo23;~4ub?U0C-F-gx
zlTj)PE1#gCDQ_>5?e~YFsY@{o(?Kt$d83{3gl}_Hx`Qr1c02(~Z`231SK!5|SGV(D
z%Adt-n4kDd3^cQi)KotbX2yWDr4yB#)nx6rQhl_${flc_VCzE;Jx77uf2#c@u+79W
z4_8Pk?3083Gz`XiokR8xaU8k5HH6GuZ^pN2mECVcW-)RIwC8zaf1JG4&N|~cX1n)~
z<KR>4-}FI5U2k;@#;{Gvz9NWy*}%~$xCIkOIKfSmSBo8=6$hu1YBGLb!ClB22d<4R
zt^T&Ol;;qmg9U%|1#7&VjL*drvv0&nwMmoivv(#x(|K@3FgiwJzEpm*$~L{M>6pi$
z@iMz7mes>Fvp$$`mWCJr#Dke;_NYhAd%cEW3JcL^4<1vo)tABUfBf7|&9$ysTycIt
z?Poo=USNrP+}=eS#;m0BGki%Gj*>Q!bT)J3ASI<{N0p7GsSa{|XjA-N-24vlUNWZF
zQbsWf_aY~rg0~-9teZ>iemqO%PcG9IAkdoM12LPDU}Ufozg;yNJD3>7FHWG?zjLt)
zOQli5+~}2%Q}HQ5!`n~oS<Y?21qQnb&&X+U-+xMor@xmQrNz+8gRW-dFQQK20Pcpp
zRI2hzigr^(?CC`dS1^*5hm8sXgT$d;-khkD5}oH<U9RQ|wihqwDT;KN3-R0bRG$<b
z+yFdl+H~d7Hjh4PBlbA3lFuE><kp^(pd>AMUk5WShXYi_kSh9vk2%_?+ISTs_Tyuo
zGND-9ya!~js;p;0xLLhOLhcv0>EAM#xKs3?05%CH?@GD1{$ckdgA?}j`rqCbs5iwo
zmYlf3JK<h4>bZIl?liI6sI|67a;VO+uRwR2U3Ak_iw#z=5vv`6#*wWwf@4q1E<HrP
zN8uU2$$%r*h&`K#bx1_nne57l+p;%Z7Q>+<U4Ee(GMMVqk>X>0dSCii6D2pe#&4{g
zwStvok>(L)r(EBP5uZ+xXs+yx4mLz+Skl6&5RY_W4aT&}t+w}On&&DJ$sbc69P@KF
zr*ACo(~?h}GZXVwQNi1*nc`(~w;z`>yU4@W6*`NkIW#OAy?V|>cgffL5E1*+4<dY;
zI)ieIg*5#OgQy=J3+-GPnvyv4jYz`z`Z%q}E~Y*R1_{p|@!>6q1?D(0grl;)IKN-O
zayavx&ZVGXiF;^t!SpsU;<g!|oP1iLXV3&IE<zQ+fQ|ul8OGJxk`Fr|biM=b6T*6-
z&i4AcHtvgS{mc@U`*YW{glCcNP~kAdJ;&Mn<Ek+&;`nS03Ww4>uP2T;Tbh>Mfvk;X
zdTn>;gBm_(8Yj@9BWt7qMZ%>3zh2@DBnCm8Y#yz_X7#3Cye)d4<(_7w`KvB4n{mBk
zU2VdCi%zX*<f!H4#a$uHP<zJz;_N-#nq0HCZ#qHg*r=g{q9P@Lw9tt_1O!Etru5JS
zrAQ}2KsrbX2uO)^REjj|MZi!30YQ2-Aian7-tIm7duHaH_nDdRAINbex$kwYYpwJA
zog)T15Q9P|EmQYyD(TJowC`W+S+b*g7_YyL?zT*x&HHf{Dv*4}K{wc|By&dJ2I#@O
zI#v8#o!l)7?P>1$`wDD{+G|1_@TDVfzb&w<q^}wEV+dSB91t22iiU60rSf$)dAlic
zOLydtrzI0VyUe&<A1wT$la`j*eMY9r1$OPy)u2uw+JEIXpy-P6N2fX9u*&&RkX3Bq
zirfc$eBWSNh(Sv*x$=S=daG7&(8zt%?u1M&<}KWg$bf7GX@%g6KS>8SQ}Y%R!tCY;
zP-UmQ8~rs+wTDWBRJjXSR8=F&qURa~d#3uQMb@^W6mZgAWtpYB$QYJKw&`~)`06K!
zs<}qT)VS@*D61i{(lYDfN9u5LaUvb8&!0Vhd-8Sn_W*U*QQJ7PBXi>fy%rNK5`#fQ
zQoCEncDe6Q5**@}(cScNk|B9P#+g_T6yh?2vg>46+;!F%i<RyC#24I7-$6rV)Gvz3
ztI8;=8bFTTTm;O;3R{{{?by=CI*(BYc{LTIs1clDc^%QMO;o<+<x9|}dpToSdbHP}
z<d`AZjsl9S@xtcCOgBi*ax%@Nkyd6qIfI(@Pab+n{ha0(UX39%K08ioD0-Jeu@=pn
zDqHjSQJ2S~U1#2+7AGUECY_q<)8cv?NeT6jtLsbdU+S~vIKQUWv^`fN&GXisZ?5)Y
zv&2Y_?|vo?S|@>6QE0QbUX}+oSFJWIad1By2=a^Tmu!l+xLv!KhLZZGIn?G`Sf9|d
zZDY(jC|x$$w;Uljy*<@<%9(8`yQR3fj!{UR9eG}&;Xc))<Sx3EYD(F3nQUtD2;fe%
z1AEbVDso(Yd3S}Z86>N4^B7;||H`4Xro-Cs)p>cm(DrUwfI`D=iA&8@I7Mo(n9Fib
z0oi-^EwG1Tv~TlZ)wePc(_j<2bMwlBfTHl&oAaZud_02@8?ZyjyBC<z7?$nprCW(m
z9IEma>>tW60A5e2b!cibLvr6xixN2*(0WZoY|Rikao~|5A=9kXGrFzLL)+=rmq>y?
z;XE~rLcqwN!<r+R(P)^*+l@Zk{(X+`-jBmLpOhnEItQO8ogeu0xK1ts+96g#@Z&gV
z|DZsvOu^%?;Z;15?ES(Ug)gwqHpWgoQ)v`X>fzq^JF4St&;7k{4A!C2xIQ^8Yrr5Z
z#WhbolN8DbXWACrw_Q3Xe#m}DuNYa7W*O(Acnf*tYX<N(ENONlr%Xkz-DaskWFMPu
zF5J-0L^mcSW-M9{)>_;7x?b0^ZDiFi!F-lswC9|0LMfGtvJFMvn?VYuA97g?-)?ee
z;4y7Pe%QS=D^#W_|B<ST?(EexE+*Ul8;#u+?+e|nW-+c`S2!x{VlxtzQfSA)ZwBer
zSX$8<jhr*U3hCLlXYAbr^IV1~{)bVjwaad~KOWpVTXn^{%`Jhg>cJXn{W{bvi&Uop
zAW%{ek#)ErG}ivPyRvu`#e0u7F_;Tc3b(3FvkuRfQF>2)XcDoQ@#fn}oPfk`0^ITt
zVvAKYl$icjaTn0p70{v9&K@TzMonJRE4rPyf9E0ChhpT2ROpOFIl<l*J;y`d@6r2`
zsrL()y|qXWoUcdw!OIfCpT-g%3I>bd&?6}0FRX`ucv7WSKUh{P|7Xl)Vk^@-IfXed
z>e*i1=e}Gfg<Xz@o*=;aFyNDOuSnnS44Ru0t>iIMHgH9f_vz_9F44uK3}+vspFb*O
zM4fFLG>_}-bulXs|J*48H9l25SFzNWw7*s->u}X2;N_A$x?xYxot#GeY^CwEhcrJ%
z72<ntWuz>#=R5f1r=}@((WsGl-_{%$8<}GBHD<R)gfG=1eoXE>DGAQOl%x05AjOAO
z>k+!02VT~X`Z`D02EF|Wb3?Vs!}pIIAZR@N)V?e1oKXt8_fZruvp>>n7)xN-y=?FW
za9{5`s~@`hksPqWA#ZmTlPYdD_Iy6|KCu2QJ@I}`?p#*a>hmcCbE;%L(h9FCJmo2_
zWeJVW+Jm*IfU|4@_$xjMek+q>+x+R7_3SW!+xJ+!jXk+8W*!AzL};-O4DnYiKd(zD
zsM1qp@!beXH$SY1D4L+l{vLE!2^M~hVq^FG=C8;!v84u)k8|yZAU5d@c@^F{Jr*N~
za>*_Bux}ZoZ96=z<zAKMg0Vwjs!z%<#Ct4~J98=EvsW#7GMG^&EJn(4voqz&-}Ery
z4!$bfiacmfiEI(*9A-k8<yueSVIhGIYrs0Pr<(8oa{K{HkL!0a)XhVRS%(JXL9HlX
zzjv}f?Iz{Ay3gWAkS0fX)|xAqX4AcGyu?(v$1lw|v+d{8dsEzgS(n7#N%<^mAkVL>
zS3H#^3%06Qyqe3N%>BLUl0M%|5pUDgfdFg6s$xKL{tD$i5j*{uC;tsi;X4F99-I)H
zI<zA|3+}D}2TvE4iJunH6g>9oTh6@JxbZpBAN!bUT~ykL>cudHG`pu2W_Mi5Z=G1#
z+sL^IpAvcqR`KF%!UMLebADdv>Cc&I;{1kSSd^OFqMD?F>$PCKZp~7*%+T3=_|S8|
zlN&*)VGH*HVEvV58661BQkmDdOR`QAAw#ZJ>N*7mPrB6b<IS8NJes<wV4Xy9N|lN3
zR|u-_p+RO_2sL*N;U#EkD#GW@QY_)?mpZI?LgwydKR6gaD?C*xaKt$=5wY~SBNs6<
zd7%vdhhqa!tXRD4-gtCb?X7$7Z5e=MWP(i)FE|%Nq8wOIHHBbX>BCc(A!GxN=fHK3
z`m;}_siCfX76h5InI!3eiCN9J(yxx#;6WIK-W$Py;=Lv4xs%)e0S;L3dfz_yK=Jz7
zlSdTra?`$VWtaW(olb(3An3#9z%78M<50q|9YT8bIylUD<y*4t?!eB%?e&*VJ)aLh
z93M^$n>aCSUuYJdp>}U`!k4P8EY_SlESy(FibLgdlo5AFGB*Y;cX8uAc9%4-LJcP@
zc>S1w-cQZl<sH}{_)XxQN|obj{c+&N8Ko1lWrcn4CQN)VP>YT6J+tc_uXE<U93RoD
zTb2QXyNxYq5&z&VWDqK|o|dq66($YUy@l}DG&JNR9KUDqy|GJ8vir(<eCtM8f#SPT
zb76R8N8zhWL$(%Ob2RCQM|CVphjPrx?4OkVs=fr;4T|qaCIAihv-LAx*5?<3esWVF
zqnU495&rONg{XlEsm`?@8VSt0P}t!NGnXE7xU~gB+JbfaeHm~)ijni(x_k|%4k(u+
zCg$DNbT77e>OQrtZ@Z{uxN^OX;#0btIg0O)-1z$1<6#qr6gp(v(Yy7hp_ItobAz14
zk3(H&xsjGoWcOET7MmCwA9@Y(4JA8|Q7cgM_Ri;Q-+{&)@$Kts>`CJE_%(5VG{TMC
zpP>L+<ip;bB4G+OIx&DZNYg#LK(k}7Lv`}`{NksJlvWtU4_X1mm{wiW*Be?YYpE_z
ze2Ap)Ibkd0{be*gu@3RRYcH9mPTdn9=&08Chp?yI(Myp$d!m9JVY&B#o5?r2&eYdM
zFaS|ud)O<W>vpw|@lsk9@1rBy%-wY+(+yGKkuQ_YK3Rge>zzTw<id{ULus)lr`G(s
zCTkj`&eJvTPD%27zdEF>il0p~+AY5{s!TiFhYb|A59k}z4{QEG2sz(v`|Kjg*}LUd
z>xMu(Q*c#7)`8{8rlLGJc`E6S|3uo}%k9pd<K2iQVH5JsW1w`q=kzY~JqCY2mFG-=
zc7`4KIeql?1)2!i`wt<V`3T9^aNE}-+dmFJWIR31_~M&#oY%NfKya^Q!)7-gzfdfG
zpSqu^<Il)(b=`p;_cY1|5^(?g3eAJnhSxv1B=2V!0lE|a?Rrsbvr`lB<iw4YKB2kw
z+t9C&Y`Z&+0X<wwK8x&eh#@%MnU3%(f6di@92|5|p-PiU5#3DuIs$rgv`xM%tTuPX
zF5n;g;9#Iw-vBa;(A~&WDhn4RF*G;ATMQ4fd9MQ|q1jiSd>s2PDQv$`-=S-U`>{aq
z9aCrESzszq3``_U!#sMlf~~{dwp7K&XAs>#mDn~>eH|mY1uuGcshz0a(!wB5i;2Th
zWN!o?(cIDFHQAC5AxU$bqAh+L?XjQ=TN)4DnRy+1w^Y@q)su<IUxn_0`xt2;d33A(
zv%y!4w<z5x1dWc`>tksJB;RhaNqgz=X4UQ(pa462cKCblR6ntI-tx20W3uW)xY>VY
zL5GZtnlvqRx`l-h-QAJ|E_{Wj?Y0?!v@-FuPR^ZA`&4MZ`Qni^mLpf9SH-7UGi7QL
zM29fAeV=@lBYZy6lkr(2a(+_`Hr)5Tf!Vi+0l|{_N@Mi-iT>Ml@j)=?g_PI}i;UE1
z!%l*C?HnlgaOo{+K?XHxbXADk%0VumzgK6-dNvrtm~r?bw>2h<vq#)R^2A5R^-^oc
z*`>|+BR!h56sY8u=-jg$NRc*+U;Daf6vcI!Z)|s0mUmPgBJK4#=+<Hs3M&^BHrR~M
zOp_c+%Dg!SDxL_x-Pd;nKfjLF_`s3rFZ?3ipX^%BWW>(anNsbvM%!bWJvTlr-@D4)
zzPUDIb%#wl>kLyK!{iy065^|(4IyI=H`;St|3p#%@}+OKIA0sAt=SC)ibZI{R_8hZ
zn3^ve!PT6~K3eS4Pq9#g+uqN)+J7PcfDMaS`nu)zx&64t?9er@H6Z1Za=D!%C+8h)
zBcKi5Y3OGt6}(tVg;w8QIymr&KYY1&y8Wbkee2yi`KQlp*o@`Y;qlMZSl{!njyKI-
zzJ6b4EJYWps6pUckPd3@Xh(tNTj?BUi7Rw~f9M)ji{`C-(QdZ5XE3N;=@;Zt4jbNs
zPRSK(#H6q<6WkF68nPOIv`@JEuijjwI*PL=;U?wUj!|5jE9tW(uN*&C;73O0FLE1F
zu|!M03xG?7H&U6=Xf{&9lLH0-bNJ2|bO*v&tGPBrr$e22^f?@YzwRU;-+=NsA<OM!
z;i3^mG^8r}-@bK1YjN+!Y~7g>Xz1rhbIoV?rJx>Y_)dHbFXxdm|55-+*X?yxpN_{S
z5+e~9%9(8>hKALb5sf(7WEJs4oObPqU!NEAnY`Fy=h~jksGyer(iy`pFEn;=9q=WX
zG?Lm6O0^+~-QtTu5A~MPPHW|P`1EQ73$MA56&Y5d3#0DC->)UFGp?0YUwwLIxPtPx
zCF+4f=ud*fhut-U&0p<PoqOS}O!-nL?Vb9gM2>9n7=uA4PbbC89ipz=m37A0vMuWu
zbS|YXpCnApvC-w2&g-1xnG17o`)Us#n&vQXgYdFpa*V(%vciXcmXcyEtXEma;X)6N
z&@^WmWX(b|bDj~x2;(D_R(@rX@?hCJlj1^fHhWjHTWx`z35y-3r1Fdcty30=gEZQp
z@g6dDOU03OXt-jLS39kmu1Rpz=QC1~n@6NoSb1=Ba9wP%&0)#Xe)K7<=_T%KXDOzE
z`b4m5IMK!LH6Lk4pzb>B*=awn%(LxLgyzhjimd(Yr#6E)TeC7>o4sopz?8wV;(CW^
zB2+M-6sjpIbTR#SN03+)7rmKfEdO3%YsrDIF!zS==uqFba`g?gbnJz0+}dXW@Uk&c
z-}Yoj(CqX|*|E36o!Y5-hHd}YuCRs}3<Bln4|pl=-zWy?R8#Fvcl>v<j5%n91JaKl
ze;X``%joVE^p?H9w9-1@cMLIC-CR~p1IM#YL%_rJ=5D{lLpQ+(aZ3nauU=>D(fFbK
zGwH$S*0NQHY!O9=vZIQpG4@LRB$ABh3}08hkZM_w&`<4)nR|azRhDoe*C^m>+Hzp*
z491$<Lj&RmLVuyT<7bM!6zY2C099jta*!>yt|EK7f83<VFp&P3Q4}3Gf2bLlyhvJ4
zkCx)2yKgB3w%RYtYXr)~gzk3s=#@2R{<^boP@=-S&O^5wP=Pm8YClGJCMwcLUea^#
z<lla>SZAYI?Fp;jHogtsGTC>0VU{`aeU}0apOLlk9ORk^DZ&m3ms4OR^_*^J9=VqX
z%J~xzpR(AR2I;wTIdxVN@Ki;{*G2%=jQ7C`qv@Gb%i?{|=Z3&;Dzc?zp}3ow(H;c;
zT%Zw_-KXc*HILaDcpcX(&ysTQ{P@0+#kfoC<Q5CRv@ZkUjk$);KxbUWnLnV##bevA
z)IrbfJny^C<wg-NF^RGQoWkjv{eUo@;(a5jwc<$bivZ7b!EMnx52jFu8V*Sk<<)oe
zQuljUX=Sg>I6*ac#%Wug^YN&aWJ$J%Pp0h{sLkc@@YM=cltXjJ-$*#EEPY#bKOHK!
z_FqcqIu^AU0V`Fnq2q0umX5zTL59Dj28TH-Ft8^nYRT}{yEURGYsZwx9z3utc)h}_
z6D}=q_p;|A7th?n%}mM9=7)u&CMDTAR}l?DE<S5IIcO<cY+6mJyx{i-+g5Paa(mJb
zZx77}sjm9f8hI9@A4`}e(d>uOc=mmhHt9wYZsy(&FS#`-+V@R9zruRNZNKQ`W{JE?
zqaV?dvS)rfCc4woGqSzv{&q@ziCmv&0Ap6)?#n#)d6mJncqnBO;G=livEI|DU0<nr
zxcwS)@OR$ObJ1l3*Mn^~zxx$b1!4#b@h)RzXiRd?(TB<``)+_rg;y$<$YZz*P@J~q
zm8roO;0Yr}pKP8>`dxSE=fpp#t~ozJeXX^oR9TsKH*QEcMmgcJ5xlLJNxdcmzhf2B
zntwI@2!;=<Ir^|z`&xTs*n70I5pgl+9Ml!kt~zJaa#>ji;>6}_O*jE``MBM7T9=dC
z*OGnNx`d<B?=*S91*y4jzp-e^t1=BJWm9>rabTD*Hg-B{ah?3gwGoZndk(Q6prm;f
zjL4gOFB41yC@@hZUYz-&v)b-)8$@)PXyFv{A$endqgfskn!YV+e)Kq~r~yIquKh-u
zL!-%{7LmJ1fKRQ1PTrFbw_3#b<$(X%_L|Dh&%FAwqM-JD(n8X~9a^@2HO4mUT>k)U
zpH?Mg^lLUTYmr*2RRX$AB<hX0yJH-Brf5gX`@KzUyE*7333C~4xuls!gD)t6jU`PK
z;pb<Z5IR_uyk$o<zlQd{YB8xL)j;P~dC};vsOq(cY9l8kGZNCges;7tn)kdlyGF25
z1)47FOjX}gO~$wWWV7!yH{})hGTtwiprOxx#=%z1VytAGhF0P-ZVqxJ`d+ea7^PEU
zi8{H}(!<&ZpVAOtONOl<@!`VH!lpl+DHVir%X>kkuS90JOX`Iv7IYO?K_<GMjC)=Q
z+P<|P_G+i}Ou62lo=j1KZY-fAs#o`PZjAKt1IfK`U+)K*D~JW}5%=<`PDH%<@saoP
zP=PasWc$evfS7ej(?7%E%ztm$5qxQG5ckinb1F3F|Hn5wVw030xmt%NQ5GsdVQgHV
zP)SPSbGyh?UMpQK8s%@wk9oT&U?R;<*b0X|(0NW?iA1V4KYj3{bvtctfP}Y@(4~ah
z6zg>);cC{KFHP&%MQN`vInJ*yMG!!?=3!Q2w=BlXXadbue7YKWxQ<AXOAY5|eoFJO
zJ-<F1r^(0RbA>0L)fW-N{}92i<vdbgK65I}aQn(G`jbjR9fNFORyjCEK#h(w@S*z^
z`ePwAWWQJ20L5EN&?6VpUN3gn;qa!Kv&fxnUG6VL5tz#fD~~NBPE%B^){X;raS(L#
zHH{s5Ss8wl(yz9Y_NPbLRvLrBs{X5<?r1#f(iD7yN5+(>uVy(JN-Upz)t%x=cG;Jt
zdQk)>$-5S6EguFB-HD@0p=lK??Hqkx{SD(aiVy39lo-N49k1*B*!rAdymx&^2Z`3<
zIiM(K!*0D$?18|N;A$S7WuT-ljEodLJLcYYP_oD7(zWj$LJJ>h;2no?%S4_+zUio;
z^!Rmtj)w3CIeMBKS5BxXat<6r>69cZL{G+F_b7riX_Z{}(wIzkl#xaOTr*f9o`!8B
zimwTAcMDXr;~v>gMZ1pDBKx2rc358im1N$5pG|2QIVCF9-U2NJs!UP-<u60oerl2{
z&;$<humvSr_;x$QrFSg(F8jmouD&|>#u6v*K|i}Cn3%Yp9`H%gm&~=fj%re@-*;Eb
zU#$OE$~<5n`R%UzAC48Rq@3ViIKMrJU9I2?H@?Gr@Si9!mQgtC0urRL!&EBxW&!ns
ztbL1~Y=Mm5-I)SItKoB`8#lhC%K!P4A{-HElOANzrIz@)l!wDUka@Rmwub%%#p8=D
za6T<McKzQH_d-pO(;O4jPar$2PeZ<vYCxgUJ*3i3HezQsfD-hSfZ_v5h26eF+b6>{
zKnYfPwqP4KzGJH9c}CYo{+_pXw^KMu=F$_+nK(VZ*ADdD+?+b!<#Ek-AE_Q(Q7f7|
z)$in~KQKV8Ln+^E1}HVXgq0m#3)*9$zr{0MEEjJf8`$JS5wG9LcYEF31a9in`k5x9
zehFAdq5Oe9_xwOYGH<#k3(j_(^Z^v<vCZ4I{NsWBtipL(Rka@=Eq7;)T)~{e$|62~
zrZ!0-i`>&t3adyTg?N!umd%P8Wbx=R=ts0TsrYPpgu!l54w&*y9F^tFc6K`xbV%!Q
z>~NX-r(H^w<$G-oOS}We(a1fIt`peORFKEsuJfmp`GME5-#A_cj(~*lj60{u`9(0F
z;+Z)U-MuqBfVNu2E|5}42iwwphIpTSQ03?s3fU8?-shcX;-yKu;!UgcJ@O_|_?l4w
z>CEnwlJB-B)6}MfG{R1N^TK4`EhNL+SJRv4nR*+jHiLgKiheo&@f7xPS**=r(TeY#
z^HQC5z=!_&h_#qY##5@&Q#K~b+|Zk~m%E~-`0NU+;u|c-OwLTOpC7t&55Bi)_(HNf
z$MJ%8_OXA2D}jO{Qgu6|@P!#x9WwjDSJ-uA=lZwVBc0#X==f>Ys92XM6<R^_?_om#
z@lxk6NMOA<04qgHdW9k&<qS?Y%ExZ|N3yt9m?vy%S_k)PZ!`9?h0JgD@pAfKm^Ik*
z1RtVpNGeY_)9*W<P_o%h1#czY&mSt&R6Dps^N`z1Kvdz?&qHc`Gq1}n05$3Dh7noo
z9n{TZ;g$gxb@^YyxvC5TAbKU)t>>^_pY!qVch#Pq9)aGTg&r)6G6@GDJ-YP*KP29)
zX5XxRgG`8`2=B71o=*T7{<x!ej*3N`ni)xjz#X2#mO&|yGIiNl?t3E)Gn8PBgQi=#
zKR)MtdchNHZ9wx>nkN-7;2_WV)Xq2v_mLAHe{Wg8p!$4;W+f&==XX9c`o{1zalFlY
zZU@apwjR{)^VTm*z}62Gab5CfAH`rx6A@O2*V{zDk()iD`1U2v;*>>=+<UU%s7A0L
z2yrsxX-Ixi4BN$sCGGOaJoC?Gh*9z0H4t_Qh>NWaW~@mhsiWu$ZHLjsZ;pARjQTXT
zj<1gt%cSahy<GHBi!npl5rro^3t+&lmw*L!J|E6y_nmROF8&C5LgA{Ctugn!(qA>2
zxB5#)gKT<#k$((QhbUsjHen%4tD8D(<qGA0zpLUAL3N+^+cs<7i2nfOHa#XrW01v&
zH^Yw{*S>ERa_VS8HsqpquM3kMSCC}nZ5rU)4=V9}Y!fi<3TY1RIk#=wWvr!b$E4M7
zmyMo>j#YDX%^C5|NPN?{qY&H|)L$Uj0N|tJxHL!<x4m+WeCan0jQGA+jN^yu9@(4>
z@DDc$n;5a!E(9crjzqo$cl`y#T>le{3DmjU|0}bU&_E71d)=pjeWYz7>NeOc2sRq_
zj0YbhaA-!-YChf}q(q;Z*J2$=HH=vW<o>*q`^r$}bnY16*oW4Gpa6c-Y4cclUn$)C
zo9e``-W08D(^Z+=HM8ZYK5y`oRV)S45@4!EzLrAWl8J3vTx{o)u|p0xGzLvKBrwT8
zV~STtUZRsM_r8te6y<Rm14%k+03VE9cz-CFfCaE$s!d;Zru5xD3^QEBQVRDUzZC5_
zW6KRO>v7msw*Dyw>O7+y5GQ*ipqAV0+M-eFqFz%$xq!i<*oJvVTUJW^&JiN+z?RIy
zJQhK@z$l6ECR5diOWBnlpc%58mcz-|XXF%4=wKS0VOIR_oG4a{S!jyD)TP90S?KR^
zowixtk6bX}fzq+uo@2^8Im866;8@0F#%UZ=<jOb$Z}pO4;`_`H*KQd9J}_^R^N-%B
zb>Xeug6h}I8wi_kD7WRrsPZ={&S6KlcBE_BeppdF@sr66JJ!k93F9KO)60G57sOM>
zqqHYr#}ys>;gW-5%R6M30nJ<`a;iD37X59jKZmKX4}<eqK3y{ws3eo)DzevgKDhM6
zM?4ejjylvaG4I>5X={Fv8-$U!x9XS;%iUsWzE`<O<O_AU2RES0A->)MspJv~%r|oc
zGgSKxPu(|CcSZ2|xirwkoWQ0UzZ>!1ET94CSE8rv6|TTfN#sWxlQC4KM{EEtK(l_8
zl0v?$<y*<4-D2pCx$`?=iNLld&H2^Lk=9=r&;aP4IM6?Ry#T8C|AqhAHwD8_UknV@
zSAJU+Pd;8;q?U%(dpqq`M&cdy;SM6!Z87E=)rTA^WU%alo04!?3HP>=imgl5T@rU!
zs`Td+XzjBb4wF8k@5hHOpp$qx?{4U?8cgUn0jjy0pO&DqlS|n@&cYZWpspQ>@wOH8
zQ$Ay<F1jQZCzH_JjD7P0PjZ&4(3+s<Pcix-n@9nQ!sjKD+>={m+0+1LD-$?&nOH(4
zJ)hjmX-F|{{03NNxw+0>s-0g~sd&@biPUKc!`!jwwWB~_mJwc%`|5$*c1qtX!Y#_r
zZPt31?lEewQ7ulX#5SBG6n}&X;YsJtkUJRgc1B|a46X-^cicFe9}gBs%B24&a|i0p
zScnPT2X@EKQDqcWfmF1;0Dn6$RQ9@94OUB$quTTatNP9aa4VkW!1}zQv*o&K9*Z={
zE@^Iloq|lOd*#seHygUuQGH%4vtsLN;z=I@2ttgeC*Eq(j5;nSJlSrIYQ7A9Cr#$D
zcgvLJSFTgUm_>9CjxgcMRK!nheKjcji8tKr39A^HaH?Gmok1`eZ)hE+(6>W>Vuf+y
zB-v3>qs5%Lz(x-eip9SCGLTu?`17to)YFKRPR(~1wTP4{@0;7EDhA~i&bLCAkBd$0
z`1K(yFGBPPw>N8&flboEFMj@*D0SbnDaOeYTebaT$^qoyl^0?-YX<Jo3R(d!k*^*n
zVU{Yzr(P{irXuXQT9rp+KYDLidIQX2ovqhFHl6&9SM2sU{)RCAgDGSQ9HoEZq085m
zC62s|2GwWNicFvNMSzd5LNz3tq7bpXQMejE^7y*5pNMeYD0!_E9el;J{?qX_b=bYZ
zAytr&Uv-RLHmNa!$jfn8Pj;>Z*6Xh+<=5U0Xe<lM_!>5;Ve5x1@u{}l>F(^=7a4fW
z8fgcj=}N1vJ;1&pdf+H-6rOaWw{7f)5*eNOO<I(he<{6E%4ai%8^slyf*2zPA2)^p
zw!!S6z^aAf{T#`pk%l#Pn9!L5`&U~z9tp(Ldh5%&uNZ&)R4Qlz8$``^$B3hkvlo*{
z2l@2|vJ}x3Q&AWZ*Q<~kWyHof3vdud&BygH!MTE_OQAB=+NDQiPmWsi(>=+8p9D)1
zQA!!4@2SdWYKejLJdOyf;UMImvY!dOfW67CjU7aA(exRvi;ecqbhyc8-u<%X7-~W%
z2HLJ(c83Alpso@q7Gbh31iM?HzSVE+T-sx|hD9ZfhuJN7v>frlhOz3OoGt>STYARw
z=FFcx0-t%lY_Y&Jm?L&|Qj@HvgOEEXDN=MKZIZ)#?&Rc@zLU+bz?wT>)sfQh7~TNT
zh5HUL_r7Z{&BUo+^rrODq&i!fk=ZbR{i#c~RLb<S-16s(jzLHHk#^m)^Ryxr9Un-Q
z+m7#z1yARVqRhQ65MJwYkAAptr$(%fEAw$q#=9(h)Ns$y5LlPv*oNGcFj^>@PrSM=
z@~<D7bA^9!um9n(=~GVVQ-^4l-pL6)9q6no8l91SrwmbW=FZgO3$EN|n$oc=#E&OZ
zGYQfthu|gevn+svwx&r-Yo5ApIgDqC<<(m11t7qZZ%@pS(!SK;i#5IxBvsD@wH^<1
zJC29uehLZH&-B~rWIL&vClO!K6zacFoyeqdF_E=Yuz_AF)ogdrU}^x_&nU2~W836D
zEtnk>5-9S`6aGW5am(%cTc4|3b>56seFFW?4}X@fVD3FJ8Rb?NyT((dJw4zXf3-}g
z-vb*qA;HAoC4^>H9jvlyT-}ONk|)1lmXNjFD<rADQ`B_dURUad#-|M_L%|cyJol;B
z5sUl9d?tmNKV!}EhnR;nX)O+#g&n#t5-*3)zX%ymq*%TD1x#?7QkErH<w}I{`OQA%
z-*(a5`9;ew+DkQjgIuaa<56RHf6q&i!E3uZ)TA==)ae-?$t4ea+1DS#HB?pDZlvkP
z+mHM#jWK_1_f75-X%L`@V0TN(GOh#PldfI&A0IjYIINMY`P&oNh+>CVmBihXAnqa_
zk4ofD8v8&>gS+e99*psm>a8~~d$-<2<I7{MXVzlQ>98<hg9I}vdjb*sA-R$9J><2a
zm?<EN#VP>wb?uE?Xr$`&m@lVvW%Q3ezer6;{#dV)gxpQ@Q(3_C{x90^tY84MyelyE
z_YL?jhyM9wXh6LR8;{mT^8wyUD&%?##~|}2Fu5>)<uy<ODm%Tm&3_QaC`+s(HJTj@
zVYFwwlv;KF2m%{?lHWMqjB@6+V!kz)oIzG4|2qE(dH0iceA2lKC;Yp?ba=g5Hp5`m
zZz~vz%kW0#nhMa}6b-UhEzfmQmy#+nmDrRAKjmYzAx0E5q;x!$<JF|vHc%An{y37;
z!HD}Bi<bAQ(uaX@Lj$a(h)J0w`PaLQb0<9%>drW?AUe2GZow^74)=>}nz2R2W(?X|
zY9xi=eOm@GrgeoczptS+s+;Znbq5C5OEvaTpMf%J?kidwhTBHFFAWE+4&zVZYe#(X
z?Q<sO6CSPdL^m;<Q++bA5+yl>S==5nNud``)vPZVzipX;3z)24T`N1xps*+ve8@K^
z#i(cQb7?cglF{^tk|HPCm9Ixb$6f6!Z5T(}hdI5PUxWENT*!@AnqEB=CPBIEAB57_
zG&dHN1{EYQBKWXe>8Nz}TKm1j#FS7IpZS$veO#lmgJ}xXp@pGoBl-!((p$x|!{9qt
zF!I^{9o^;hDIz~oN>oQArL6tkSq|!Cthz@em9K59B<(+(9IrI#`Yr_Yx+&xCIlrOt
zqKEBD<x779-x})CN!B&Y1LlqBzi8WOest)<g4dL!^y0e%C68|UqvSRW2aSUtw)*h*
z>1ysYnKPPg_#;DriB^7&c=KXxNE}XyMA-?d6f$e)$^nlpYlz!FW!)?cJ9wvSA#ml~
zg-;AG7g>TrxZ}WQ4G*;8*5%D%l9jJ|^$qTms<-SQpzngScw>hW!Q##7h)S#A_+lw|
z=Q0=B^~d#B28HfE^`6vG>S|K)leG+jeRUeq0aV>cIPulbRSTgsi9|)$MIM+^e30QZ
zj*;m_<z3fjBbTovoBA@9El?$gAKbM29_h7uFL%VBva3I299H^qF?Yx}qtcI&e(2{H
z<Ag_~bs9NuGV$*re#=^Ia>KHjHHzhHYyxIwU8G4xCpp$YYg~U7YQ5s0`9suvnN`lb
zd*A2M4(kbH&k)$N^=Vt}TD3V}!yfy0x8_F6o*J;a7dDwO^5{lzSSK&Qf6nfJ*X!m5
zxz&wXV=tl=LcZD7He^T)zJ>H-Y$@}6oXIrvUJ-VZ>SEp!7L2##4Z03gD14nq!4w|;
zmtqiiA)-Y%<}kIh?mfr1ecbwk>x(<p&8VGXuc^|-#F_4}HC7QtuJSSfv;W6M`p=iz
z{U50NK)di4w;ujjNh0912Yxi6T=$Bmd&mWg7j`rK5U+e5+QHXy&r)G_DN{v$?z&{u
zp{*L*cg!Nmn1da)N=~f?@(y(pIjjHJDXYb>oh&fMZyTWr(KD8q!b{Pp7>1-a=P7RN
zZe|9hTf|>>k|GTajBd35CL3%qSzqLQ4{I=dIU!9x5xDT^_oLUN0%s8R&#_v_i?5TW
zo3FBbeA07XSQE)e%`WE^iQx8b;KfUdQpf<(5s_CA<6K&Oc%Js<R{mzt6G}E#$hA#h
zuFL15X|9iB872GC9NaXIX;kISw<82SPbT4WILU|GK3wYN1MlB+s^v^k?eXOAKf^3m
zF2uRGo;(`7XaH(Qu%Qptx=Ligo9F7oiVyQM(tNudZe0{7MVlK{e|9v7Im}DwM`4aJ
z4H0w|8XaYax&%2h;fVp12VTC|cN%fMdd^X;XHB!#nsPQP+@e$DR;Jypyid3}Dstxf
zW!A5n5z_8y2*KU_dGK17NdaBW8`81FnlHWUV{yh*Xkx&n7<cG)2o>c$^Ip^sdfr_f
zvc=I4A8V>^v;U0fa`n~-;2_!N%BEV9&eOykqjJB&yh2OA#Qgq?YqrLZ{r`mC|8FC@
zF><sRwBA%gS2@g~fmQb5ptKYraB5W`l<8dc>NJitLdjqnXHtj%z(jCo;Ptm)jCVLa
z|9d4mkC7l~R{x4oGl+Ff&FEM)0HuaV>YtKG<JG;EPvuK5$IpcEu=$a-2$d@3*%z@#
z3Os|LjSWVNZ}Sgczrw^FLB#?3>LJR?ZAb9MIi6#2JZR<gvwCmk=GKnqefEAh=T-}@
zUHg_(x0T~48sA`D>3h4yEsm+dPEo6)<qTqO^`2m{>PSEGJl!pg{`uy!&l&L~8ZyZ<
z`c=BO-rlTc>4HO5OkU=)?`SI0g>r;4&&U_AW@6Q^ff;@QLfn{lGhIbYPA0R&$Hb#y
z@$Gj_Z@D{`YFIv*vK1a1R7>ycYEPEE);wBx%#hO!9Z!Nn*c@8{gWIvPyzH)-VO5Fx
zJs#gJvrkLO?eFF)*Qk!|qhrnn1LWWt)4URmxL*$%38(916?_sMxZ^VKWQq2YqpmIF
z)QbQMJ+xPGhH91Qb}DXR)*}FEAzL}s-gS2V*4xb#KYpOL-uVv(tMi&~Z2eM7%^Odx
z!S}rn-93YBf@AeA)($8gK5mO~39`NN#<UT*#6&8ofESK8UXZsi-V*79A~!Z)$3~w)
zaNAInRU5lK0D1j2i0m!n2Cmw$&_cmdb1JP6qP5?-uyaxnIt@`Z$QPPa7bTCuea4cM
zMRXCOR4vFx%Jua$_ffT{D=s}L8i<wzA~`a|ts?*&pugdlY@C9NH+`4E2zSF?8ceg1
z_0A2_1Pfzc$57QzadtU8nmjeYR8?@~oqg-QwkY;g)&ri7!Cg&qWtR2$Gm86U@WXz}
z7Z35iKcPDJGj^2^7<QjrPWsIwIuZb4KK_hU`R5tnKjm#4e}B+FZmbO(6fSv(V4Cv@
zEm}8;6rY1ZWD7OMt&Z7*1?<No?OMx)MN=cTOdHjKiMDBDv(>>zSg)7ptVm<{e7H#L
zO}3bMz~zvB;+*U36Kw_xV@V9%*DM}@QG-z)-$~T?%wjfXb$R`W*V-OGsKumyYEv#F
zs$6?pc>kWnqg1r}3whlOe?(A}QS}=ZS`AcP)=!1fcR4{RdrxeMzn-J|WSKUBE4S?2
zs&LGSOtz+|M(q<Zy~$zPRAPwIu$B>A-8=dCGL!s$OZXX~VdB0l=FeQ2Klq8}zjH&V
zfDD@7Q_+8{;QyGL@-@bZ5-!~ojsS_U{nQ|zx04)>Sjl|Ao0r&ANNPZrtyGSzus`3u
zy~bm4da>&}1MlsXC=ku-H%~NqYKaaot}mdju^IsWsqwIphG!%Jux6O;(d6>ctBP9%
zNYW9)x?>e#l<#vU)dGpPao(SK1@!Fi>d;lsYEU~lcc<a)eEP5Rtw?|0l-J*Rx2@^!
z?Oz#)AqFJT5HAuHMk_eHUhv|CZmJNc2_|O2Rx=YQGcUI0JqRd->NH=|r+MS{atsw-
zd2qKpQTPA<FHk}W7o}5}TTG*+d4?Vo<G*4Tcq!OA|MbO0ybbf3l&ErrmHcc$E7SOf
z0&=PT@|E$+GDn=S(i2g&1Vr4oI!Z!^Iz0@+TaZSdUC&9--mYF3R2$UHdN#7IUvu&v
zwRXUYdwH77=p%^BV~KNo42<3JUyDo4t+F58j$$kfviHbT(<a5BSH%OWPu)2U|485Y
zomZ0=Ho^d5gyY(l?rhw25^NLItaax4;LREzy6bxveJW5?-~qR`57q|WxDC4Xfn5?v
z|1Yh3W<Wkj<L?}kKZ8L2r#t&sfj@EJACGbfn6<1Lf}kXGVF(;A$Sf~H*R+Rk($?kn
zd3HVcabt|#yc-Izd|CZoN^W!Ue?DO&NJ#TdRH2t3jP&<cQko-)_m|n6WtmKi?B%o^
zl6lR<SiGDs0aWQ&$;~-YUSTF)ZLQZz!ZsYvJ}kgNY)~#!{UcQ=B8@X?Kj@pw(bh47
z6zf=F80=pkGJH%;g1NdKMm5}pG4H}G#!g98580K|3r?Jf1$K5F%clT>5dZ)8Gkz@}
z82+38b2<?4lvBZ#GZr)XIc(g?r9$2%u_u%1nv<Gr8v_v%bTb=#j0}XZ`N+?_Uv}Ta
zE>FCU?NrOVQ3}l?>A>F4S4M)~qM224A_}V{nW}5Og=s!2(||yE%3q3Zlx(DHG&7f#
zd05#q$#0sVU$^Fk6EIg_&O6sl{F(S<ej}^EruE{a*R*eQ#a8OMCe_1`A8fgI8t7Uz
zD4wwRB`bT`kYQ644cs5WyP9pMGvZq8u_FZ3p9w~}^*~6F|L>Hj|785f2d$!%Tv&>Y
zWoDu^NK`ga9tY$?W4B>kjU1Z_0csu#OZvCxZh$oFxw>lBRb_!N7_J(Ut_W7t$_Njy
z9-={8IXmr}nOos-lv^Q9X^1w4FKxrz;1vZDb1v#u_bqrjSk`rR(_~pj8Kd!*4#HbX
zjzx8@$aL_zK9%P$U6iwLjQE{mv}APNt}xZL?WIwa^@F;r<~9M0{o5D67lpe?L83DF
z_Frn3`z^0%jg2rg@b&xAt^5wISx^TudVZ&+{h8$TC;aqxhL18Opj24V{=ZXvYTtWH
zZZCbUm{N{h>@Ln<pL*J+y+#R$my>;&Hwi8(GRIdEdk-(G0p!&bSGJ8`hWv8F%pNKu
z^HN2A{C4L&KnjTAg7CwMKt-fZ5GE}o)&ge8ADg5fou<d@_~6PK74eCQjuMrsye?m_
z)PSMG(qjDlX#`@G0zSQ-D}j|C6<hF|`RwQI4+#`xQOjhkpiISbXY95n23Q-@eNZxi
zXIj(W^hECs$Xg+_SV#r?CX!D~{xR9<KdI0M41<7xW97_fGSan`w{0+#_si^H(}&g|
zv>6Pqv-iuByI~~;_COw#$QJml3m@B#Oc9?M2%>Kmqggo|zg9Z@k(ue~fm`;NG*t6F
zH!2?)k5KfEMkrH8JF~H5w1%YJAjjU5fE}XyX|*WP_5QkK@aCNDkfbdmq`2Rp(qM(C
zb1eT_(b7FkLnT6M1tDv?TDB*_rM!5z-_XEnH*Z2AphEMXW7z(6Tls$_7*$b8E?9DA
z%0uY}LUQyQW;9WKN4I9<T$Ch_ue2qf%He2)_}+TTWkbrDAkO^r`QZ){^4*3%Fl^ev
z?^mQ@uKIL&hzVS^n4H-Nu;aM*h+~WRu9q*e*5>1=zx2b{$AvK%^sG2N9q2+4gS|@s
zyg%vXvWr^{hD9U#3nzd6ce?|L)&$2=R*Z`zC48~ymvJ@bYsJw1kD=}7GI!@z#lEJ7
zJRp_pyNr5aI2j<`==2Y1bN~D`^^}1UO#J((rtt)ibtcqh)(WuzYlsOMmGhvHxH1aE
zXM8xmU41lp1VmRSr<+r{`QPHiUyuzLE-x<gKi7O*$2f7Msg~@^p*-;srtvn=?pGsE
z^<gIntO(vbAH#RU$u^_6Nz^0CoHD+2Htx)F>?tV^4=yi0O%g}H|3;h}hiUPMXQzL&
zGHJqr9dL!@wvRXKvd=JiT9IfgE_@L=Hz~9ZW!52#(-)>RR_rVs;z`jS2IW^@rHU!+
zh=AstYcEr0c9REfUp9K{ohdt4>#U0LjlXR^IqcK<x1_26;)Qi*z<qfS2jRros=26;
zqHxt&w1%KDSeyDK6WM8K@xvPCIU2OXNW)hKq!yL!j!RR_EAc+<Yj9ZdMA$*bRF21@
z93%a0<~yq-xqLCIK6oQ0to-FSkZpY$HX0eSM;)Bf;^sv)j=)vA{Wx!-h(@w5_|NM3
zAKYYXs0ouS5Y-!Xi#ZpG8sw`ofjWGN@mHeAu=&B=L~F0j@Vd?EsP~q0$6cV$^1873
z^>a5yKQRvkztpQ)T1SbV(c%VN)J%R{?jm&Z)GTkqFfnQsPpPs?M7eLS15rwmd~_Q*
z4(;DYd_VTo4wMToiUJIDAZzcB48VUqP5p-v{Db5*Q}WMhNcmKyK?t99HBl{T*<9uD
zm1dm$bAgsSsvk}|GA;2nXW<=>{evvJZy9FPQ+54v2(oXD9Y63xT23wzpI~eoD7)TJ
z^d6lBCmtnKx--35VJ#!VLtzl19CB<i!KT=zc$xfrRED}(fW7@6p>%)b{wbAP#+(0^
zYyAC?=g&6cuMtnJ$2yy-WaL3J*Du#W-)c7aO>JeC7F!|NhmnA%5jRstj$lxh{i!rY
zD$Jn3ylY#}K5!e*1a3Xdt)%8*_2!rbycP#lnSAN)64^KUn0O)iI>S>2aO(*}KIG$F
zN{>Fx5l`*B_qU%+D77CitN}*>{BQY`e_LGqxx@I)y8XQq%g+L`-2A$v|B@#?zkSxm
zGmE-)r5z-hAA@d{DplQTQH8A&1#l8mC?MIQ#o9az!+>i(VPv_M!$U21zEW}L7Si?$
zIsEf6Z61J_w%&C#WRuPjE6i$hu*yo^-BoLmzwWU21>J|Zs|_&T%YWpjG^UpU|7q@j
z^Pf)a#zm6lK1n?kQ7%^Y2WH6f1Ves6H>3|YPjmn!haB<F_{gJHOa#YD7xt#jV4<AR
z2RL2clmL-ncpq=EWbLEwb3DWF*Y6zqCEvgp&)cLq$fQwIJhRs;A9LQh?jbsPT=hr#
z<R7VU|MJrKSF!p_kF$*XQUtXrUM!xW_xB4$^&r~-f5}2ifU3-nFhvM4X8>=u$7i>v
z+P3PYmub`gOLry8<AA9NZ%!eFRb_@$%PcJM&z_gNrp$7piTrkswEQX=nbF6t`gCM3
zMH;&GqnyI|%+JRP|Mmg<=P4rah!x%eqVF=yGm&kp4tg2V^l}<s&PS;-Xj*@tL4|U{
zzw+x1bD*rgz5}g9s~F0W&;&_vd}nvVDmPJr$u3PsTXiC+qwbh!6pv{w8VA|1h0))H
ztY;G~1;?qBFX&S;#Td}pdHzv6^8U?l_Gc2`-~D2BOg5B06b&+VD)o-nTIx;#65)lG
z<}XZigx6L--<6I!-Wxnmm^LeAbt+!0fSlC$-)d!0@#%%~r3SFi-Via7FVKh=aJ4Z=
z$p_o$Xpc`r!`c0+yt%N8HkG4^*Cv*Xf11@FI{lG-3XEIj0BOs=7`6YID*4x-9iR7d
zBt2<n4-b0Kr(AJB+P0q&IbXleCt#6n1LY54P%;$BlK;AkjjZ8h$j0w1`UC_oH8*AX
z!+O2J!IX1Mj%MK~Hw0I2qSDp;NIO#a2i6+FPqbm&$&E&m)NjB2m>?Km*iHsg7J=N&
zKcW-=st|t^i{H%p|A}7wtpNK?;V$YSolLR@JRX7b!{)rKr&NHbLqm2$iOk9+Yfq`I
z^2-1K!&9T^&fnvfq`OTa@aW->n9hxAKx=Hbr=la{^vIO~UDc?Te?c;|f>yE|txb2%
zaHh)tHRvq`Hn8-vqfOS;TdRtF*}!4;XFTHX^v{1={SOMs;myhDH(O&q*{qIkd7~-U
zjOIR<hZktuzKGT?qqmWfpg1#t^O%4;k*?Gu;Cr1mZtgwp9h_@mHwOI^Dt-|gQV9+6
z&!-?l`&p>f(Brsau4s>;kElLB>TcV&fm#zxAO|Ajt)qA1DT`F(*>sAaxE9?XF?v2E
zPq{9e>{6;ZWGEGUq7z77qm}f;jHKc+CqnMcBP8A)T2}uO#3^`h14PpQ_-8iv%PL@N
zh|qq`ZP1kZOo4}6Fpm31h>mnVeGqQ<`Fs^o9Py*sc6&2U!(*X}XskhjMgF+S)luNi
zOPZ<S?=*(;+P+62V$1FGi<y7UpE~kH_}21sSQt2bzSt9xMmjVp1Sp9gN*X#J@^mlN
zQ#Z$+0*E*@;eHMeMzTlp2aF?*QpJR*t?7pR$P*)ELKt^x0O_0R;7_<=TC~D_;AsAL
zZ`VImq*fy!Nf@DsmXsvV$+}nR%N84U$#tR`Ln6U@z0llpXuMOOuPyW@I97lrF4&OF
zw1r`<dY-5hdQ(valf(wE`b77AlQ;f;47dL;6<}abNBC+PH@u_#$dey8NP|xHze4zO
z7UDrL%>6OUH}R1PFsw699>XbeAT5rfz~RJlok6vAx|MN=D=2pj)W#RxmZV>%*{Tqr
zeB2`8ti*#aKIaRWEe={N{rBJN@2bY{Zw27R1?3A1iY)F43TuPHf+guv4BowAeei^c
zwj;w9OG%K!vFfZF4-6d$i|_aYfn*HYd}Wq+xOU(xPZE~X=y^c?qPT7jO4j6Ndv+ze
zd@B5)!-da(!=$Z8kY}v=%EBtU{!Pf%em0Zx$s^>pi@M}mV_c-a9F7M)s}@4WiJBie
zp+S=~n(ru5u-w(Ca1@WnIAQI}`n8N}Pe~j6DuEO2zj}=j<XFd>cYw6qX_=r9HR<JR
zJolai_xCH?sC<vKqtI41;pem4%E*2QNw0@?2;>_W;AL*|A*MCww9A>W4Xgw6?@-P}
zp6qRCuoqy;%NN`Z-sfp}^JGwy_mP*-?5;Lb@(USQ@vT$wxl+)hIg-t!rMT;FBRz=5
zu(jhGY^4Y19%P>ACzhRF!_}QpcjYW``<y;$L*k0B-Wu$){$-J{QYvKAXZRoVHve$+
zY~ZQT!L5M=7;{7a(<FF{C7Z^$b2da=XF7Gaix&OF5Q`8Ax<0L7AFTx`5^%)v?lxa{
zQmcK=8gIKUq(-mlu_?q9>BVxbQZq+>xdhcwPffs^b4w2MdT;WyEF{^$oJybE2kT<)
zY@3{uO@LwD-e3|nnoOaW9k}XPCCb1-*z<ziqqXsbXvX|ZVN#Ym;nIQztA%LdmajC2
z{(tm+{;pTsz=O~iF?{p~DlC#)HHBw+R09RB3gJ(w(Z*09I`}}HPChB?j_?HwKo9kS
zbhASplv%^I>C~w7Y^F0d<IpA-h<6qL<;ri{NZz#yRc$+&(cWfSd#L(#UQ_=em1siV
zD@EbWF(y)Jo*ls*<2H{(P-@ip`xE5?(xRm6ZRB%nO=x^yHuJDtzYZGIaxqx2Pj!gm
zmRUQ~HSEut!}9;~iCbRDVP&~QlY*equ6~_xuDKMc2MH2uj*exm{?J56n`=`=8dGHV
zffFweL=$~(4D>j($_1o|^BGW{hGh|0FTWu@ApRk^DETbd#*UvrQQ+H`H_H<kT}<1P
zeWquP6raVya6_QVV3TI6S8gIVoL%2lw0r=;>tCccv5IpRWjuaGvp@MddAMPI7B`;a
z#>fluZou7{)=|mm^Qa>Q&Op!quF-E8Q(_6OsSf{enPrHLDOrlU#GFRbnY$SBPVK-!
z3k^y?jumXP56om<&-plx+^T}09zesb_MQW{?n_;VU=7({K&9rm+wUtk`x+YbGYQP?
zea!1=@^DPULcnlK5yI?r#a{B!WFta{^*wGH-N7u37m`$YC6{K$)-U8bQp$K2FTL~X
ze&5XL(OP>4ekFEUTK$$z)^+y)mN~j>^<X>-6Kos22yOXX#(eR=JN<!U>FU>UjfZ?G
zPG%uq_*U3uGwNb5lFiU|P;hKkfxaze*m4gm_c=B_K()O<4WvJyU3lfUO;H+|fw^mT
z!HZ&mZiu1fNLRVSgDk$I5+w42>3~fS5^g=&MSP4PvP?G43Irk;d8yIkmmHr($*Eql
z2o@$4x37{Ji(E3g(Q<HvdD6OkAJ!wsVMoPQjq0?)`lWe2`g$u^NHqI2<3Bo@{-K0(
z+O0i&ZFMM_Xq*0OBYKLvX0I{UY}Xq(;>+6X1a;?Y3DsUxuYN+BT_VCHdL_oI!>;Y~
z88aX<*O1?$Lxr8cIrZv*=)#vFeqrEPj<Lj__Ra}{%;M`2My#=|9tyn?A%T}FB;0&=
z4{n3^cz|`k|D(VXUyh>_zu@+EG^(gKI}5pFRN6GLf48Hy(!mt0+n&b;moAB5ml4h?
zIJ_}qup;|0?Dxg~f9r+L-S6>wT;<r@^6I>pJ&s?V_%nyTcxTRF^?QARD-DOoS|qE)
zEr6%HZ@D|4p!*bGc)zsi(4lbJzD8&vCXUcq3du6YJ}N>ib#EQbQ<GyAMAk#<5B+>F
zr|MOr$_DyBKbt~KOOsxcvZf2`7QE2*TdiJWRCA!+>NxXu$6Fbz-BhwSkbfK3q0n{2
z?PB(5U(u1lZgGDFv>(8^($$lyk{0tWb6&(H`1;{J07tJ@dj5b-ng@uM-kb<fAM`$p
zbp6Q#qh5=t*yA=t5VK4?W#ZkZqIRM)c~iw`2l$q_YY(Y`aS};9u&O9w58O%j<o&X!
z4M3szvTsZCjD1LHdarM{8*Dgr$1Jl^tgyuK?*Ad}E5oX4w{{hflvsiwA&XFu?uNB!
zP+E`<k?!u2M(Kt{gHjSo_mb`|r8^}Twa7Dh-*=zu`p%DY_TJ~uoO1C4=X~Zf#(3^A
z?)w%lt<N9xH%0qEEgphUPBuQKYqt<q>;Eng(d$b52rE<P2xY(d+_$O9RsuaKLgu`i
z2_QM1H;GK-Uk<8Ri>L}AqSB|E_WCOUBe~5Lj>(dO;w@B`@a%1$twbojUH<$_C^q*2
zJZnAmVAFDu<+icy#Icv{7;qYE&KeB0iT5T%YgHVvwzQnt<8v^$U<a2k^8W35kIj(B
z(0(P)rkZ~}o^>14jt`~Cx?ZDqXnjP&--GBb8o4;V_7HOYWB<O`Y4Iq^En6uI!W!>~
z>^JConXDT?qVI0F$#c4zh!K(N91*P}cXLsV&W+eUn$mtcjMzA5u>CqK>NMO3tb=dX
z^*U7koX0%GIOf#{%JtKUKPo=Q;^_GfwD|8A#6+1H5O-+UnR#kkeS*_mB<>K}9E)O{
z$<HQviI(-{Yd$f2Y^7e{8Vv6F1TZCW{od^X6=z`ez{Oi4Uj#o7?K^l^iJuW6wIXe~
zBb!2XAjk)Xt{|J0@%bbvi+~v`g$H9yKkfGjq;TT7_z1;EV{<>-^HmO?7{E}N^9u>h
zKE4b^@EX9p&EQ2l!ttgqwqUkUM#X;l=~9iiU4IBhs5I_rYMXVthdpD`2yPgs*$CZK
zhscaHC1=-1oqr)4*oa2CoyD5E)N_Edq@JqnojryD2<}i|{v9BKt5%s=oQq#of(~nu
zmqR)oo6hK6{v<Jn9R5VrY;-N1P0Bi6e1vSWn9(~9Z~3N{S)h~^1xWQbHd#Qdo7cz_
z4V$yT9&4cU`QpVMpKS?nL9~+Gzj)TEx@=q~+&3}q_ij5JrUTEsFWSLfIt$0d7@trS
z?leg6PzKVlW3k7_m7tLazUPe>fB<k5;2ltVZy3U1K0<X3Xkz@7fgFQ$J=PO7>mApJ
z5T8l)9hh)$$vS-+x{3-icSKd8L4xYtPsbCXN`Ecw(bH8DcCV2WD3ugb*R*N^+)QgB
zRQvL~Nq^EFI;<1XR2*C_EvWx925`5_5}I^&(78FEd(ReYWm+_&z&98Y<uaV=p<rQ}
z0{ejem~EV~>hXv>+h?KVvMEg+HR0*2@~w6c#Ks=l;MC<vU>x&qdk<%~y#mt;OCyxB
z-KSV&;_dh3$f;s6M{Y~&cvN>jvfZgoko*N#E@Cm-{g65gEA42>R~G$T`mJ0_$zG_r
zy<bP5CoP%e_Y#4yLyyNR2uRS@nE^F$AnE8Y0Ok`3EZRL6!eMyF`iUVMo9#qAiwe7m
zaKO8fFVVm!1)H1C`q$F6_DSuz$HKi22H+;w%Xj%45IBPJULJnPWYJbqIr2q7Vm!c?
zWHG`Nsra+`-oQ+;ZxXmU72kjrNo$b#tvM?k!oA0ed})x0;DIfeWh_f%)Y7Xq{bLOg
z!<p;%sYto!B+v2Ybb6J|4;daP)w*2Ua>^3%T*!6PKnwAxE3FL;FInHS=v(i-=G6M7
zY>9p|nUe4k+7SQhV~kKZa?nn<Zs;hrsjiG95FRyk<7=}^+cj;@s`>rr&HA8$LBF>y
z1j3^8?OMQBZ(}^q1=)N45a3kRHw|C=Z2D|X--HkF4UV=nl^g3%wrSHp2al!5jC~H2
z!llaFh~BX)KvFUZcJM?-`fDn1qUHyCC|jT*z;{L4#`UBX+`~q>8}Wkb1W9tU3neVa
zzQ>!Mck4cd(JD8c6Xy>tO7Tf`xPCHFCh;sp??(67q>}T8&dwBEB|s^>or>z&`-j|4
zEGBwxKqFPpw~M<8bderFw_B-B7!6B|vy56(@Yf$XRa{eWfGJpVj%0%3^ZSmpn7Lc1
z>_(JClS`yk#f~KhPs6u-&{1}gb79itBwwup@`}~gOBMMJ=TV~B<j0G}^j!GO5g=sU
zvseqicdNU7m9%LIVb07mQHq7zvBCESuSSy1?hGv<){P10$Jq?<K8*zf-+8+<e7EH5
z<@*q9ka@l)0q|rE)+83i#SS)o{?Bzx(SP6appd;#VCP;t2?JI;?@Hcs@|CTgtdGWg
z1LqUp>8FccCEox)WU`;b_hNRxKU~AYf%jG)7twbY`02iv+l?6-doLMP#U^ny<gye`
zpNjx?OGo=|aB?M)(*!12V1sQ_+{G0EYL-ujk1Ph#EZ;qvY%sv(&2~m25Z&))_p#{w
zs@5d-%FD~${9dK-+RXHej)OJkhC-`{oSQ7sF{6VL-}X53NXQd)+7{yu=A|fvirGgn
z22t>cJ$3#XB-fT6$eOu&xz3RpwrTl2MKmUnhkk~H4!p-;(7U#{H(T52_?1MKk{kGA
zLI%}|WK`X!dNv_elZU2(JD9Ab(QO|t1J&aI84AmvHxD1GvMU<!Z%zG3o2K#!`R=|C
zrzUlLq3lIqKVud9SePAEVLG7(UwM4?bUGrGQtVLp*>;#wEX4Qan8GkZ(3@CZgCTcO
zBFQITVOVrL`HSHvi%}fMWT>_k@~W(cB|hsBjB!|dhmYP*_tyRHDrUGG5DJ16*05wT
z>orXHr0lHfBRTxiq3Y__k$YI`C;s=&t9|q>!2vL)o#?{jcSo99N+*U^3Vt6|zcD15
zEjp{;9=DXZM*U`uk4`uzX4C(3YrmE@1jE;{JC1KjHXF{|yS4vvpmrdEEU)z5T4jVS
z7VPeL0jX}WPqgw8GFmH3;E`=027cGD?-`e_+q}zuIR|})^OKhI<^!P3XVSNVAC>f3
zX4Cs9+@Rwee&;g{r7<W}FVXY|+<OKlV{QQfxp@{q3a{?Oaish6N6G4x0HLR_)7|Mt
z#3R_{Zu#BMDg3+v5>u9^p-G&6<z8IRcUQ+;Ic&cw`HtLdvE^G$mHhT7fkFYt16L(>
zq{xAU)@bN$G?}Gfqs2I#TNJcriw&NEZ<W?*yK?h3*a4KZ!Y*1buF6mSX6e}&EG1-5
zFn=I=&5<zoO{xW*wYs}}a7^b(w=mlZlC)&M(F}H;bDZ6F3{gw*el;=?TGSVMZph=e
z-j3GqQ0UGmT#&98?%T^OF0dpQ*TNt~5IveLJZ!Y`IiJ;n1}y{+^1P03RD+FKQV>D&
zB5xt&qzX{k=bn7uuu&OGLpcHN%{CpOm`I!T(c3c>ywvOfcz@=5*7-S<FL_i=XX0YA
zNs@Ih=R^=5=GsJlY&JixeKcm1w#izAzdZ`6NtC(a8hJY_*!?v$2?}r288w@%IZZ1(
zW*!0Tl2#S7)^8HebIz_1UIE+s2*A<ECnf2&U2R?KKsZ>#>uwG46}o7h+#5X%xnkZ2
z{Iljoxw7sC?7VohNKQlF<D#cx3^w6^%dkpH_5N~R*-~USl#b*anP@mRsdk$E@VNkn
z^tpUaD&Y9#A2XNZ)OVx?K6sA;ELra}bYnESkGId&?4oP-_LVuK>W;g?dt>bJnxU(U
zZs<Nb%G>-Qa31)&alfH|()AWd)vn;K+Zm*Lm0Dey-)4qegE*)6g?+>`j<(-1)DnjW
zcee-bv$G2!p9tw2Mr}@$op}(4`8GE^io3-g6E#Vt)z`>Q7r2_!lEZ3O{nxWAr|QKt
z=tB8s`Dlj1Ipa9j$?kEi!TGH+r3zHTNB14V@b=>D!hZX0=P7d%tJcuEBWqS?#n%dx
z!D*c~{J58zO~wz2S#>@?>d)-V{W`#fvtcxOO@A~t7kW7MM@y4u6H8dgjxve$gLil(
zH_K}1z}w8#@n2-$lDb4vq4Wn5Kqb${T~T>k2n=|7lk5)<UX|1}J8shm*E&<*d`~%V
zd>o!sx^@E%ec7-69_yyk87q3zA>6|+{Mlme%g4zFNz84Pa+`;BotpAc7K4+=Q&0iQ
zQ~U^VB+QPQ5+GN?^f4c$N?cNt7Pr7CxvywSw|6c1+hM+8b(B$_&!^^H7?YqLY{Zbp
zXHM$E!|o(Toh!fHZF{Zt<xJ8bW-wy5b*?o;?^Pj>deixJ15_F^^}v*IxG(&+?)pej
z(St54`UW{a--Q1eas(N93n8Myno1;Tjy~?*)jW{!%Y;k2)iw<S`jNmd4>+q2N4!|F
zhJj!Hx>u-yU7!3iimpGvjdW!*z!FCI4qQ!n*EM=Ti|7>Ru}-=NynVlIw11!wnO`GB
z${&~|npLaC3}3fcsz(?pB|!^jDr1nVClJ-n2>*?iM8GHL$;5U!WuNzD&*dukE({PN
zPAmKP3c&(+y`2sNqtLiJwEQQcufk3?#8kSyw<N%hfu~y-`i8!6^xw+<gyw*pJ=8@(
z@s#AbpFCuoNq$fts16sGOYMA^7Opq=Rsm0y?zDvyx?1727>T)Km{plWw*~go$%x()
z5`H}y@S?WHI6l0bCm#;zFM>mXc(y!K4%WA^THLE$+=rA@?GI140Lz}=nCAU5-ryLL
zU{IZaUv1D2Xok15xI{pq2bH<&<&}JC+!g?C$)0Kf7^D!=8|@f>wIZWa$XD=uz(uzh
zy&Zqh?t?!U?M1I-cJV^k{jM~{vHz?4#Y}99Tc^Q~wXwzJ-u~ODB8}*-n_B%~tW~>K
z)bqtajQm6WV%^G0X5gcPg3D~yy^yNeWzZh;*{kncmAQlI&S1(f3aP>%!u+A1KC0lZ
zo2p*jEMAieM1R$%b3@J>&e-WJ$2)HMO#Z-4&a6crRJwUzmiz^84$9^hyy2)DvZ-iK
zA=e4Ypu&Bspmy|Z;S+WgQ}7oF%;d*P=;OM63Y1)H$_ZFm58L}`qdRlSm0BaPYL_xC
zD(wr#T6Yzi{V3Y^JJj^;`l_~5IDP1DL~!vBk?TGaWZzux7Opvte_R(H|2gl?x+OK&
zX*nIvErm1a%{M6(r!JN+zW4dX#Kt(VISOcmHX!Yjk$|_-#nEa;FZ4LEZ4K%tuXJW;
zHF)*bM7tVxn^DS+Wb;d}y+)d*+Nt?A<KJ9KK~8Qt@2ah(z?atrKvY=lthr?OQ#_ix
zu1UI7xJ3^8*=E;An`Hk6ESS~gilPt!aD(d=64t=t?f3_3eoq0@$Ck=Uz=P!BXHvD8
zEq|44hM0->B7h0dg<{Wy10Z`{g7146*yx{shBRuj(6JaXFjg~qWj^_~8t*dv0Pz1b
zN_SuRv|pZN`Q}^xbI#M#@sZ04Z%rMA8_~&hRXK3h)YGgslgr&oU*VJFw>F!@r|LG1
zfcwb3)DCi|+~(5!t&Qic*3+ByT7+Tg4;G!;_p3>R-0QK)-;H*qLu}U1c&+EFC}%BZ
zd2KpE2Ia}i;d{tOzs@cqe>rx#eK$Js8sWd@pUCn>shMmB>4ZFwwjuF;JWD_Ep?_st
z6K;dSYhxlDY~`B2?d!7u2v)uuQlpz-E31ag4R|edIPqD}e8$yreZQ){(7(a*tHncP
zhp)UnN}+0y{Ci}}tT<rp;>SChoMxaMvZpj3^1L=H;1Z`LECggHR==;T7IH?XdnPSx
z{NQaQvyi>ti-KnK?d+Sn`B{I{9v=`3=^Q?p2QR7MgCjuL`X(;)yGuKr^+J#HG2e&A
z#!wqgC7nhGoGH7xpgkR4_UGh3JgjE>oo)etV5HHSoI&PkHxKfi;X;<jB2$yQuN&V1
zm)|1N1<sOHv>^e0%Q$?jx}dQTdTas7IwShFQMpp%UFZhOPZ+rx-4yfFRXPU>p+z#w
zv!ZBkgUqL_P|wA6AG?}H;Ji~dcBQpw-6DMDlXQ1xdmP>Oy+y}t5Vk(mJS;dK`q@g!
zLu+|w=ADj?)juDP@JNxT!Jx|GdCPgdis8G-TcFgLFSSDZ?bd?GxR~37gu-w3?Dmal
z@ME+u>FQwg2*Rw3j-mX@na$|Q(C?5nv|O@SbvF47+NWI2L!hl<@W&<fd=#2j>YjH%
z<{OR_{I-EPQ?Tko24H3u20AQ$wh&gkH_7Pa$~LAB-~DruPQ`n>LrL%hV4on=2J7{W
zFIy)OtX<24S*kV${lwG}5-{{1mX4?k=8VyzS8Y$8@Lebci6X3RH9xQQej}Ja)C0Vq
zWz7ZAR0eO+3U4uYFqK?xVx~C$+_%g*AGcn1hP|1^ncme!obUC=E+!$9FfX4e$9H;z
zV=Om@$31_V2n_5Ct<`KqPSjK1fOyip<I$LPPRT)Ko)n2@Wnv~{3}JvEdK>~8;=G3S
zbE)0`p3Q%g3|c(*`r>v=TnCsTTH-k?lUgig!60GiXW~nu2<0z9%B|*tXSA~PxYA8W
zk2;>_|9C=On(&dGN$xAZlxj+yZ{BOQp|N%8pFYUuy}{#i{XsQf#CpB<UET`UtFF-u
zmV;1{z2xEJbuGwX7X9wlfulxE^L*`DZ>VoVpLf>ob-l>%&{7_x!o2TEa}K`u0n-5=
zLa(3h?iJfkv|aI2#G<8d7SjFz(5m0QMT2>29N>U!h@#tgx;_Ekfd}`2<78|M|HmzM
zDXib`cGIuxYVppumdLqy>ToF#-}S6vx_JoqXP)VU+d$qdmBb>urXy+ZJu8FuXx|fB
z4X$GQ=LNf45{jx>=C(u7JLK@bN%o;P#e%n3(b?N~tw-Z&g>1?#OXl#XNe{^WwWn3d
zOg%&qoPT9!0qEu<`GHk0OM;uhGtYA}<1^ux#9QrZejc-YkTrq*CQNhQ!N8&-;ju4k
z8>7y&X_hAw$XaSZdsNmE-zD(Zi)yqltr*$DDI?t1Q~=4&_QmA;LuyRcTOvTZMYLhP
zzU@?S_*%pZ;W2wK?Fy8RenBdlkf<i0Mk|9JY_(JtNpO!b_j8>yFcYLCy8(?B>}{A?
z3@ZTFt`uRGU2rqfM)z9M%Cunh3V_AEdr4yEt~6da*@-szvc_#vigEh0zqxp<2z_Wc
z9~<Q2zwr>$3ZKrLGN7sk8pm-U<v45*dwfylz$Et8MCZKYYW0**JApmJ`)=FA;)Sk4
zb?)fL@bgKs{zWeyrX(|)t5u;@zOUcX;EN*jm4G-HJ`C8utAn9laD%;ZyWD$UWR=m-
z3U6VgOQjwiblXc}-d~Mt^JeX4Ga9iBm}(5L9VcGued~;s<~#T8yBY~a+EnUo_`DoZ
zlZv`qI9*;agACZ!PHGEk0D9CvwiV6*O9h=6&J>9HB(r#)Q4v<A{38j!;Ckdn+Pw3N
z3h4DRJ{~0kI@dW+d;{pe+;H<WzxK_t6u;Z_yT!=5oG6%k0QF*`n$*X}9yjn5G2aPQ
z-W~?W^MM!C5u57aa=VugMz}FA!?zbiD+{#cAQMgr!)UO-f0mH`80*lUntyhfUF5-$
z#_ebPE9`Iq4ve~^z2IrC0I6U}AC0stle!a#72&fNlCal-#he}lfV!!lUk#&u)oL{O
z7SD1Zgk**fUnzt08NYd<4mQ1Hz6y}%Gr_sZhAl`gSL6?(4baC936gKSEJ}NJ3B^3A
z2yA!=z=~;+%wDU}1R|~iWPHpLMZPR;eZ-vJsL3ybzM0XcI}ig(YqTZcpej>8pyMS4
zOfG#X9u>X=Y)>P^FXmQd+vy-_Px>sv(DfTX5!V+PU%YkQZ4~9IJoDdJ7zoV24co91
zJx21}QC!z7nk6CI1M?RiwxGW_0LJYQdM@z>CmcIEuP(?40FfmHkI`D?p1WmPc=X0_
zbJ{g^*7nt#b%DFkye9l+z^fNH)+zy8|6o&NCiL_8Ujj@U`d{SBHwsz~LJu_ZTv`jz
zG2)6Y=lPh%i5`%hh>fSdLR<#q-4Ya6djm(8eb{_)zS?Ht_efH%uGZyypV>ydzFRio
zNzUdH?d5S7ZInUFDat(tN_d&`nwaXWey5XX@v(2um4uv##?6-Jh?-x3C8UJT5o1}h
zn;;2b`^Y{JUwuQ&s+y99!)ecxl!G%~U;G@u?s;xU53ElIRO*FtbZQ{+){Vcl5L^mf
zO7bBf<XjcInKLgU<KZdVR`|FR?%c5~o7q6zTT}hOAFD}!;6egcmsHOu$6Q{3!=%t4
z5%SS3dC!Tj&6!ipa0bkqX8F7l$$AqWCd0EBxRI_FZ!md0m?=cU1-Q6;{gL^-fJ}!V
z+anXTChU7YLKEU4(wWRina^X`Pz6=!6CL;b!Zs)^xGDWY6`i(z9YYq7t%)0tMh&5S
zoQ@=Zq2UKQ@}Z~oIo6i!F97sDP}XDKzgAXw+tT#pGGQAVRzs^>OVG>3N0q_RH)$_@
zev>7{mkNoSuERQl351dHa@lz5%hMh}&0EZn3Yjs`H6|OAmN<vzz+(@lEO-J5BD<(5
z1<)u|gfs!I01TG(c42_@9V((HWnf+ueI;z}3HDqc?hQ?IO$MsS60wikSE296S$o09
zId(bO=ZFPw^BFx)vnyh3vgTSIKg|8Y&kn%&6tWNCTRvh4WI$ZOFDz!+o|d(K<X$P+
z=$sIR`m$3sKjsnN!}q(8z(~3Tjd=u!;?B{szayNvN3QC&a>Lo3x0LxlMx&!NtbAsg
zdmxC+N4tedF$|=wmHz+?#;4=#fQ`z|Gh*GiM!rt!pAV6reGDUQG@<Hx;Q!@Pzac-l
zFFfEi4@M-D?(pT=Bg@R!J&sqjwoc6ha%!$6(oGk{FOqN5c7?)Vfv{3L_3-WdHGWK$
zz1FgP)f;@HQ!xPOJsi4|zFvFx^SJltj2Qu_g;8jVx_AQ7a4BZynLvk<@-c<Mp6aU~
z$MSIM*q|2yCU%o}$RqB_hW)wCpE0|f3w22ip(*D)UJJqlDW?Q|%p|X06-KG7aH#aT
zw3%+LZ2X8&QZUlf&gG07<CWAgkgU37c!98$68wq|c#(*CpM44oq5Nb%;Se_Lhu;o~
zKFNer`K(~SU=t3h8~gb2{6)%Sb=;S>&>Wb`X((X-*2;TOUxE6o2D=wfxXJWNuoocd
zo!TUA$*@pX)mvDl?7L}67}i1VGO}IfS5k5-c&Y!z;V8jZ_0RYH2K>ka-VyneHnbBs
z;{6MP1Tc;g^~~@VJ^*FzOg4+O{&Q!akc|9pazEye!+pA^%RHR1*ABQ$ll@CQ`{MH9
zQL4i1rN2swJWm2lMm0MX?4nM)iY3Y;8e_@mlF4_PqZUGJY-t&Hb2f`F^W25#FDTPP
zP@{CE+ic(6Zj!T6DS5GA6&MKs7`k^fRwU-fBVlr4DX0qCsY{xM=@#`Mi(j%aMcKJC
zLOdChA32Mu?LAm<oa8#IgONwQOrYqS!n6aBC7oP$2SFJ=`GW~u(OlE29i2h}Dy)n|
zn1nK={3^#XY4^kAJ97405<uk&b$O)`Gd;XVKg|Q5I?^L;`c0~NeoWV~6rd)XX#U2H
z>cWZc)j9`9zt&vG$-4?KBKFY7@t_r5KdDx(1FN$pWD2s!=+`I#4K~uD&*{0lkY=Kt
zd-c3%{5EC{?e8jr#DX%3?T{?0XFfD<7u-Cx?K)7esGoU04q$4DI9$ewDcNjJ;8{|i
z3RT@s3TCWCo8^9CE=_9Q;G1|-^g!+0Uv@Yb!ZnXNo(;wOQxdEm$u)lQrf;*<`}P=|
zj}9wU{c9BgN}1q0#9#BU|C0Ur9~=c8Wk}w3sN_b@utY|_QcMx?TPQ?OdeDl;x{|Z(
z{X}<;lbULgX|2=dkjo&=B((w9p3#~NOOX_y!NW64UV-<|k=(!>uuU)Dt<;qI$Qs^l
za%()QfLQYJO9cn%plZ(ci32vpM|@R51o=KY;f0V-DzswXDMxqc382obV{(wts(t<r
zo=}<+A?#&~<~iA-<)uIsQb)9anTME<V8xvj@^vKwm>lPw)jptacYtsnmB^-^Y?tq|
zBa<wknoA~F1ZFzV?*Q(J;Fab3PUH9E*x81tV&(oL{oF!_W6cEYsrvJ(9aZXaAFnFC
z#vKCsxL0rjw$*KN9&lojfl3U**2l}c=dM^=qg=vOEPsWG=504Pb{;?FuuYQD;L=`*
za9<Uc#Icp;JTiVUeM8a~e)~Le91x!S_2XwD-vbSA^lcwM>FbkfOpPJ&DhJY)uED~Q
zf`{PHZzc+)S-TRwCr=>HKNin%DWIrK3n%aTKA;o(F{W#n^0T4Cy~1?Bd!ZN?7`u1;
zn{ETxq!;`>h>QG(4fsE)Gk}CQ_fwRc6!X+Y)_o@J+v}t@7hg!!^=0PTYRqfFseia8
z6};2t4J}>+LIh#K-p{xnK3xre7Mfj9GSeanEiCpgC$P>f5wMUQnX7ry8_q>A*oZgU
z0jF)+fr1`C{}>dL`k3-#;XXS?>b<$Jxn;s{0NVoET-vkbz<_TVIKsJORJQz@0`Q0n
zUh89{Z9ZHDW*qq9V!LTb-(WuHeQ-o-C+U3zZ@A4m(u0d}o1m@8u!?eSdVY~RkZRbW
zez8kJ_4_u!s*lWy3?D@^IEhEjBsHh#z-a4zYIeO5E)9=ih7)RTuf+H58L-jb`l7Uj
zB`%|qZjk-&8(BUU88r*ZxrvX??M||penh=Z<go*VQuc_~m|aac%OkX)Xm?a#7)z4<
z7Cr9jHel_OTnyxnY7AhVQKG0l^bXD@t=q-<#bZyfJ=yx>(yt-b5-Z$aj;nRdzp@YR
zD^!&^S+bMD8J&pq&3j=B^EcJF-$-QUH41;TMDEoC?o|Q*=CK08S=M5m1NmUWo1cZE
z=6a95x<^2V(xOfcwa@d}S$w>a2p@Ndyt<l0))3b}<cgBg?CndzCyKg*so%EyVFr$a
ztr3>I<JtAS`Ic-R@~MWiOfMyy6dMk0f$y5Q6-iG;>G2|<A@GNfBs3^5N=on#=KCe9
zk24Tkg7h!@1TRqWMLD;sWv$L61l|2Qc!EX@WoSsRhGjPv{VzJ}0b`p$<DFDlJAfA3
z4-GsisRc<!)7U@J-^JTJk^uzpMO78-^beFJ&6-JjUxFnZuap2Q@=x}+8JPMhf#&^W
zUlfF5l6GPfc0m{lWo)K-$0A~7^3N0i?U%J718Pbi;qfkRJizHZ##fg!G7k^6?3Q_~
zn^n>SwZE&OQ#PxGbh{KTS&U=>ICy<AfusCuuec*UF~;Pek!pc=%vnL2*-Qi^b>Dt}
zqtY0D{h)~KR}ADGToswVtrA&u>DTvRQvZoGDONjHhp-Y)Dj8MJq;|vCzQ!e0_#;Zm
z%EgZwZy|Q;<=^g!W}Z*~qNLu(zQ0F%y*>V>xcpBx;iGjfn6Ioc<6};EnGa+DP_;)I
zT{~R*_l3TzwA?gre&rXjSYY(G76g!KxXh&o+@5X-oi;o3n3J94M_{XPQPp93R(McX
z(xtuj5Q<ACrcUv}VO;g~?*n3z`&1(VzDP(mQo1AA7kzon5VHW7rkkIh*G(<lQ=)AG
zF$SvMP{g3dI2r&`r;;Q-D5DZaF$5cE<|;Ry;wy|D<VjR|H$-D*0C~s>bqoV5>-?JZ
z$OS6uy-W`%T8f~w>H~whp&&^>)oIL0X{`mHw(g{|Q1?nNsP+zT8#OQn;_`H5U6r6L
zBk+v{2IK5Bm8LgITv|)uiM)74zBYB8bl??dhWsuA2Gw&<OnX3LU;pA^(6rB^(MbbQ
z&)X|dv?pQC`Li$Wi(zX4eiGgX2B5+EdwM+WOcjOUIPs+cM3r>Y$$lgya%%U|n3I{v
z{t3Z+)ho-ST~ThU{asjZ1EOLv@G-Ym7m|W=gt^|N&n&Uk1aa^NXV}A_ceC``<tDVq
z-*i{)Bq!;w@TT~GBVaYlrpiAOg}&f~@0Hf=MaQqVBaSZiEFC-$$D3>IO`ic&9Cfga
z4VgcmUF7Tk5P>tS!J;)4#N|9F)E}GA0t!k3@s-GY-yofqe@dt%<v6H;m0~B!zN5(s
z%-i&pQ&HQ$RMJ=APY-t=v<_0z4_<=Zi33J&+57A-0x@AcJMl%gDWNnbfWvJ82qeGi
zg24Kb=HY|;D<0spe6HBYCv&Os#T|>pnb%D_x^4tPU0;q_xP%>Ck+;qIMsCf-$LL<Z
z4%ldnO~3V&1T+=BDPKQ(=^f?6Hueo3bTO4h({ObjRzMg~@L1Y?R|&Jg4jqWJoc7QJ
z3~*5P8I`2G5vOc*8uxT^7267+I1q<x1Ovu`4(?toEx=Kub<s<_3AXX_*<RH{O7YmY
zZ2_ZIO9l^o6)tmARP3XLxwrP){UXVDWT%gb$`ES(Meb_5B^AT=BqY>HNA>x&Wpuzk
z^7iItF?fw^N*p%0pV$7shr7A(gx(X{I3MACyfEzZ*-rZ#0*mcw#OB88c+1Ld?%wN8
zBcjm1^psW&i~)A`!wAdi3AYi2=WUC&rZ{|<@9HPQvYVbCC7FaJFNE9g5WGnC!+z$$
zt?Wc)fB8d}^%@h<$ouR(^iGzK%|&?!{V~mhN3+v8hDXLhcTlhChvrfa)#DM(otbO3
zZc?2KAGl$;Kgf-yJDw5pd>VZg`KG#iq4BL$XF#DELvFywM~@`clpCBEyqX+qeeaMP
zEvWqsOU=nR9tZOX5xhzvA)&^rz4UqH{=iU^JI6{WUT-bMLUkx+>vLldl5+*qizz+K
zd_Ss98HcWlZjAG{VjKDNSWb>nl;RirUT=?x=(j#V)<_144TbPVFMPumTENTS{A4E9
zYaR8ks9MatI^1s2M)h<6DQ-DX%(bhtUdP^)D+;Ieum@!N(ag2-%Cu<NUOv)JR)CQe
zUjK!vMSo1cHsntu>$Q&W<Aa<)g0ma^TqS*}Ci`McXOI`o)cjVZBmS|TAp>zQlT5;T
zZGy_GTP_j8I`IosHKt;c>&}O++<-XUVdj~u&PLOhva5=<b@_dgU^7MHCee%7)#G=&
z@-;f)rU*%xM99uC9XE~yaVQ;kKkBI&A?&MvQz0~Mgc*0OGstxPXV6CN$KVLOpBPvQ
zIFze#{k<-ctL<S|FOusOR^zTLU~3IC$JyPvAbBhCN@U>NR(OeKPAVaFt!9U!VxM@=
zer0Z;<oVCY?+x}1n8nGUmSd8j=+DYcRze+o)(@Qt8jX~es&_%ryOHJ#^kKXCt@y-m
ze8W{p*oYh^m(V^a8{qog8l%OBQ2D~+!&-$v(SVClK{LotgxIip&dwKg2kl);Co5Eg
zE|^Wmc)hSTc8#G;?!~(ytU>fon^m-qQP)Ep8j*#*lr*mOrLd^EYc>M!n;&7T5KL(Z
zO8KJ8VWt4dLZD3cZPIgsJ|6q1;kv<@u-S5a=QX!f8sXWJ;P3Mdwu-~LX{XC>a|6{)
z(swx{bC|91aH(rC`VJ?_UsFRaPVYB#*n;^oE3fwB@(>#Z%nqed`(*rw_&)I+$5(HT
ztmvhGze_S)2sIJ2bg59#FZ_Xd_i819`K?%zu*c&C6?*PYVJ$b7j3>2{=6xLMwOQuB
z-%uu-j0Tx5%R2Qa-~0|THSAd#2{QG_zO@#w<c=ZvXPr_;N+?oBbRQlk_L>A6G(6a;
zS*s{REK965u<m+timYBNy|FbatPwV+<PNY<Fz5A)A@_-W%ET?u>!;YE@Ux!X1kC3?
zR1yvHzNQH9w;^y90WFIlM-5ifYNana%0Xu>D5E{}1U@`Z946&KQZ(1E7^Nf~`UsWK
zp2)mgcl%5fVcRk3DNqn>K%Fjvw;b>q(`qv;<!j0Z(xeKuq=UCJTVR?wV;O-1?|!-h
zR6`u<;b{e>wri;m^-~I<LJF*8gyP$aA66DGHi=k>Ir!b41TY-5-Z?+G<``}R-%{l!
z-=*J0sY<?JF^9)u4xJJh+@S_p*IUD>y|E;^YIVRII^451Xz<wskT*p1_(ppu8YQMQ
z?igjQzSrY2a$bvp6}1~nhxuhB;Yj54deGPO`e)^ls1NHs#4IALh-L=TkoIcY^bM)e
zR!;}Xlan{{YA$aiqf>>;)a4b-i9@(UA5Ui8|AQG%!8{4$ext~tdZ_sE=4g@igkuf%
z`C|<}{rCuKkUGHXzU@NOtmK$M4OWIhNDGc$kNYHt8f*Xc+I_Z88`YV#=!L4h<un&i
zve+uHp0V3M*kDV~I?!fJ5Xf@}J(T2FLC;!Fpkr0SCf7o|EGZQiwtJGxh7P#al3%mJ
zrS{JN$HLOk!>9}p)Eh0g%#8Ng5oUmvktEs=FYA`HSB+62{%b|_bh+;mVkVUN-@VC5
z{}@I`k8M&tI>#5fPtlQ*K-aildz5bCPuDLe-;pxg6J+h<;_(?XQNeoMh?1yZ@FFTb
zp+9t^A+xWZrtlHnS-{;2SY_IQX`*_OqTZF9#<npZbMPSir>;_&htYI1am8cTYMU#G
zpuH)Hkv-=NgU7Z+u<N9R?z{&{s|bpY<L;Q`XoKCP`O{-(4_jdpWS;1T$P&~oa_Cs+
z%bPf*(G))9U~RzLpT!T>U0ydGHcZG~?}R2Pz{HvI?;{crnzT81a<raR@N?&AQT`K?
zr0#cv1r-||HxmXOx5md3l!By%GTE-up3e$t!Omoaut8K?f@dI-Kdis0c27UhB*z#r
z(dab3dM0JANR;#2P8sUmY+ge?5Sk!>vnv^{7S3N34ECZ4J6^<<egc+?lPM!U`r<}#
zY88GMOAA`8VflcU@6;Yr*tmDM1ivPcG|G}t6CqWxja}<*N@k(;h>v&7$H?#P|2R9B
z#mxFT_i98puIZ(#foo_-h_udSa)f?EMt(SQR8J}fN#kwS6n=Qhv{d{1nQ!+X$T@rD
zntFxT>ZctXaKRw&y*!sRwneL&o$%=<HDH@<%=qlSgW2JN2y$DFMZQsSNv>NXLSPKs
zJ&1wu{pbMSu~e@SBUeq$<cQ}S$DY+j+~9v7KytSQnK}S*1Be$6e@Bh?;!f+L`ts&w
z6#x2aO(UvkQ%0;Q?GenMzhmVwg7~XHoF2A!wyEM7Qqial;Ordsb{supU&3Y9<R<Z5
zdfnh{MW*xfS6jEbf|6vBtlV!nUTYHrW;gt^{n9CB!wm!q+hKu9dG{0l%nPO6@gcp&
zZ=mK|x48(<57S(isP<5w`an&yaULB@ukD1_w0zzQ_h1U5VQlBMqG`It2t#I=UV*8t
zuV_KZ<TTdwyEaIc%}Y$%Uypqc;Iz8J7&(g%_yS52RI^I!t$j<BjgbXdgDWu`F**0Q
zPT+bBB&t6DCLkIL#E=le5>bH<;hBXDd(Lm+CC~7RT`O(_VoQQgg(m@)%f0dIm|}_@
zMmN^pmq+%GeNJDh4X1~FZ+X)}hY{K9H!I80aqytIfzcfSkzn7(1*TkOF8(|U+n;+_
zAC*RPaZnQmhG33X!X^0sxCW)yNKo##V7|j<EYMEd^&!aQXJm#augQL|AHHlmxV8eN
z=}?(h)`ShZfEMc*R$R%cTl70j8VBd*V92tnv{tW^SMCQ8Mxr&D4inNylx1l&`U09F
zwg2b|f3m`=V=>8nzin!>88H)$ro_8^&L|v`;omfp_b3pOhIla*cb!P`AK>`rZk?@j
zlTzsMCJ6^@8!h3s-~ic=y2*kGq_#_C3CG*9o1@U5jwgune%5-;HX8FS@jBSe7pHeh
z)HkdrnhJQrFac%!g4ey#n7eH00rmr2g>X1e6P^y${)Hb6x4?YwOj!9%%V9+(bNPX!
z>13%;E4d?ITPpZdPf2iER8VOiWheATDy-~3Tf?8Nwmb7gr0)(Te@%oheItn_9Lo6U
zF0awSrvLb0J6_po(NNR9d*Ki)Lg5fS4#Kw%w|6u^bApH-uD?e^wUz8)tw_H*9?~Ol
znL}x9E{*j%QaY4z3pih&JtS6?USlDSCHqO@pGt_3vOXqP2|9I@fEAc8*-N&h`z7ea
z-0@C+a}5Lsx9<lpZX3JTT8CfarM0`EAVR~+4G1&C4$y0i@fx1v8bk%a-OJYA(x1}~
z)IO0DFbOYb=BE659J8K9#~pqj(e7`5ms*4{CXC|`|Ljb4Jb80oqc6`}gG><*yh=6-
zaI}{4@p>I@yF65vBi$){-9QPo*4yYcSY-xal}49&wdbReX`jR9aKsLvv=q)PZ41d%
zO;!eyeAfPlaGiF)YfMRzAu;wXQ}?VhDK!y7|17(sRlXd1)u4;A(*%|@O89H%Qn$fM
z;OO!QOY+qwao?NL60aWQTV(nvb-;O+bo)1r?6Uq;&5NZ`%~t^;-k1Dq`nODuw3L#<
ztF-PkoTx+b3|Ghc_JV|b8E<|jZS<A?H2n`Yr99R25J>eV`%Rrzg!`yQ7<4~)r&~!<
z;fe)RtB1pZQD#v~-Bv?Rv-8t~iceG`4OeB+L3Pdc7n(o$YZufGJ{v%uKTjE2C(;I_
zK=;YI)_}J6KY|<&f-jpUUq1vHKJ7!*@55?L<2@9HZTp0w;>TicduzPM$EEo~(i@C^
zBp&3z(iGBp*y2uNaN<X1<OL0zBav919&lJRmX1k=Y5u`EKb?#VZPHCYj9xnO8tgJs
zqp7$a|F=Uc9f4Sf04YNh%CN+BXIhxxbJV*U8v<Q8<%-pGz@i~48jv-e=kY;$aNb;A
z@=rD^(h8+km6SWTjkLM@o79v?{JCtQw8kXA=V%zv(pSDvQtGfVq_6!R9Ojt!PEgqj
z(t0-08y<&$w&4ZlpEjdXr=KR2e7jAgGWalx1~jfRM`gkqnr%wg1s8foi_!A=GI)I1
z<Fsr&uF_?ru^;=KhEoKT1vv-#lN~N_Q|3L1{SQ7-u7@(bPh)>moii1thF;v*hS|=d
zs9c?b3~I|T`*CA->nN|!(hpRo%5`QHMnP4x6b@-T%hh@q3#b@0p1Wclm2^}?pX!Wv
zzpjyY(&!s01<^^->0Iy^Gs2-?9Gu_%_&8o@&z@oG4gNL%u0*@66<_kjrGR^GA?<Ra
zYNmbm%i;3&P@{$J@3kSh`7P=}wTDAuDPM#u*2mo1(o{RR!$#J^6$g&q){8UtLxJ)w
z6sY6wZ+};DK+%8)oNWG17Wd_|b(gGA0M_7!|1TGtr_>!~nNIXHl&+0ecwDx$xfO%x
zVKX)nAyHQCyHstJH*|{Do_oAfWr3K3HH#7xhR-FHC?YT7Az|Ol!tIX<Jb94bQp%qD
z<no{->mT8cWnLvp%cS<1vI6;X-AEb_D49}meU1u+I$}l`-*-r&xm!T?t9AWuKYqo@
zZCx~nafgoP1eL{$Ox0X_XRV~yjs(5^tYO4BB&SFl?@=`0RWU|<ty>H6O8zeb#^%7y
z*a5f-v**-1-j_swJ=y)4&6G=e1_Xo@e0w~3JmFOwjN5@9n3ry%-t(|9{w*a8P@2uO
z=!h#SkZs$o;5p3Y80C9iJ=Hg{n1ejZOej@n4j|a7{*xSmajf1LRp+o+t1K2TMPo-}
zogprUHq{8!t-1qHAS0;Y&xgLNQTMllxYo_M0%7=YuEgCqE1c)}M8A+zIyF1K*UO>L
zIQ}TC?J#in3XEU2N$8u$kx%PQS&1qeY)ZO2-Pl%B^c*V@!aAu3sU_}j;lOnrD6A7v
zJvjb(touqYjbkk8y?ABBl6eV@s5wRIWCm9rIm_MM7&o`#$#4z(P}^2!Dvj|g^9zxj
ze#No=)puhmRF6E4_)K2pYTQL$%<xD`e*X36J6-@D4X2dmB@K5V9tDB90^cj;U$&}z
zN_34w8<av%j?8XA(*uquaUMsv2+df`;$Io6w@rG}TA74O;%H<v9x09;TA%t2<hald
zLUq|lCxyn{Wzs*I*1k7jL<RSU|H&|Hmo)zzlW^k5N5N-nz~NSIPL$HJDQQ}{{^jw%
z{iLEYK;cWrpF_vlQBAb*h+Ha68=^=V!5Y~k2eKvd#&y)<j`~B^+yClJ>L(fzPeN*r
zrect(h<7>mLK_1nWGkw$yB@P#htMGne4o^>qtXO&NULTaIE0?2KZu;Qc~$d?Jd)S4
zkutZyFVKBTCTw@3-@%8^1gmkdOHB*i4X#Ogv6UCC+sTcj89$ccVh(4vmH&5l5+{wE
zIX1kBm@F9~u38qjjWX1`)t)@R^P*Ti$SkTfgy#Bq-uU+nX&47S{^;B=2(RUl@8)W3
zqY?c??II>MeB8PgZ)JCy*rC3K*$EbZ(0`n7TO6OA(NX*R!R;VaZ2v2?zdve`#PL=J
z&z`VO#ngit)ebn0y^ZON{P(KD>^#^AGk(uYxSjKdF_+qba)h=yQeA<=HnHl5FgNNa
zGr9cXvLA+}UkHyFXcAOY(TI}k86^>PG@j+;Z6vH>Cx`xr2fv6HZ0<1b_C;bEx9^s!
z&CWGJ@q1IDi>@LU6jX)t(qS`>4TM8^^-r7j&8RO9NzpiTWRRt~=>_CVA$m{+G=e5R
zpc;@NzQ6t(rGbBxC9u<}lDXY7hc5KMI-KM>TrY7Wal<IQyYN4(tqBUOqr?52S<($N
ztO<`cCuy_~pRT+PB!D}S<POnOHF;Q}@!6pt+%_+}i9B-1`WCzE$xF5`^mthT>KgdK
zgN9cGl%g{0LQuUZi?)qd=(EDVCN1o5KO`AMz|n69S+9#Fp~73MyG9>h+D`*sbL@Zi
zvXgwn88){YMk`!YZHh*Se70MnOF-Z7ooM(46wrh7F$YHCU9d=o#jZyYVU2f`M}D>9
z&WIVzyGD+kgLzb3pV{^-ZtNfQ9=@j!1AO^t+^#Y)?S4H|39X*zOB+44`q}8fH33C3
zNrIa`&i(0iG}tA`!J`bg?kRFSn6IM9!BPk~9}F@*>RIXfcT+es*xl7L6R9q*j|vX5
zULU~UkI%|~0X<Y={owZd+L<OtCwhqm_Q~8E+EK0-c-U(Mn)`v%cD75~;>Xs}%qq>l
z@yMO4tF1gEl)qR;`J5-994dP8DQt|-uDxxOwL78vE+WL2MjE<>eekUo#BzykXiF^)
zCw3^)?ywG|P#R(z2v<-%Fzi_r0>UC~IQhRFBfXb?Z+paPz^r<|@L(%p_g(X1iwp-_
z4@OvSQt`Q<HN{5L?t5_L3Z04DBWf`B7YJ1fRzY}wJ}!;>h81tn9*Tg9|2WPr4fss|
z{{CR}57x!;JmlH!8#6TpN?gAqT|Nxsjswo+*Mc5Y8^m(2cK>9Oug|$!4G#qVXK#2!
z<sMYorwCRO(b#*l+c5rCS(+kJ-B_<<zxc!P$|n&I*>_4Z!X6LW^hv0j%Ae+v^9C=P
z5~8{NiM$97C-CEVmKoONx!mD~dLfA!lqqORozp)SA8P+>b@jDD_VkEYO07Msuxq=C
zTyfZIrM7g!<`(AA3T70ZGDn2d@%3Zmm6>>@yRn1Ozne{y@I5$s4>%B;4G|;Gr7!^O
zAg-i5w3K3)qQ3lq8Neq<b>0(;c`%2yJeUR?Z*hW#v%<}t>8Ib2P7gh$uGg$$fG-cY
zxvaZo{t4*cEZLI&O`J!)X>=WQw8dA5Z_1`kC9W!I{HpszAb_QSBi6}&JZ+-y*2NLx
ztG{(Dq!8F>_MgNH;3$Dy3i?7A?^sqr+2=<T_?UrPPcEO0rrZ_Mq(w;3GW}GgTn+?r
z#=Jg*2Dojq9%e#_#TKmy6t*JWTVK>e1$qMmR3xbftCincZ%E*AckI`;VD_O8w-;nZ
zbjLirVfRx>#8hydOhkaFd$;9}S#omLy7z02W0I~YdLcx4A>IO9Ia<OMbsRZb6k`;e
z0Ek2JH<a_QL;7C_ph`y|t&`I3VJ6aXr%G|c&3hY!zrUKBGk~noMkFoDMD*M0(bi>s
z8J(2QNww=tu!_4AR!Mz4&s(nEv6|fDBk;t9mNN7l5Kgk2M<0#Up=t5);udzobIfa`
zD$9)g1%SXbD686QlJXUUD#L?3hM()4iLdsr`gNJ<0*<*nVsUrx(23bfKq?H~Zqses
zswm6caW10`AFQ9_?ns_s5BIn1Gm<ow|0nN0XDle6?Kk?v#u)U&7f_{4RmgO^r$CG@
zpA@O%G2A-&gnQI%$T>q#c|L)PD9-VX07oW_FsTAK@+~*~A?kCW`795|JG$;Y3|aV;
z-Tyjl|JOqxb1%==;5u4&Qn=!`G30@!m}9CO@rYo5E*V#9#g26txykX9*l#&9*u=-*
z$zwe307&r}VG}+Vd#?_7!2K1;f0Ck|a!SaL(^E#6DYfQ+E1m@&DUb;9v?%V=;lJ>q
zzX~rc0RTCPh;>C1w0w!BptuV<p6z(Gsf{fltJrh!%p?PK)t+0!x3S0hp+b$+pN5%d
z-6}>5+^99W=$V+Wf6M4oUE9VX(P;1lOt1>rmL2DmBQJK3!n#gG4`<-YcfZP)?%XgW
z`R<-Atje^-^WU|b0jNAy)8(UzEQ(N%-XHFE{4cQ)mo*HsJ&`Rv=y(G-gUooi{h^M3
z*SvX8RS99i_Z8zk5FY$b&3IXfH8>`!KLJ3tN$9}Q`hR~}&J&+^1ZRJJAnnGb;r?Bm
zN|N+P`_1U@rH1-G2Y2rp?awh^Fu&Hves~$hk_c(cMSpUzN$`Wb_>aUgDaPpsT+Aj<
zw8=^WrRV+^rHvpL2&gquC+Bv<W_f-d+pvI8`R)w$eza1p_JO~yPBTTGe}Kl2vazZD
zueWHG0;f7<89-BH001qO0TdZHh1gbo05Ul{P{0@7qxtNAm+;E}pvwO|guMB`Ab}jm
z;QZ+HF#c%qp=wRX%V>THb7@-<Fg-qx(%I<?A1XGyPkExo$y#OIispS1p;v0X{qELm
zXv0~77g2izwFY2^ZZLI=n?cacUF2auAHh2|q8vv2e$AlbHC1rmQKzvT2Poa}xp?_d
z(1vBefBS;pJb~ptdzr<4r;vxqfBxJkA?voCu(ij7USRqq{G0QldIiq64gtclUx~c*
z7UXqo7%N!2QqfcEGVWjB4F*APc_e+x#jZ~r0FdqZk%A6&gC7))%^>LFhMqv)kHmML
z0#x8=J&mHx)ZZe6vA9kfz?0xbMZNkvBdOZ}bpMUW18ctiH{>b(zwzV$S8MC9W71aP
z)5%<h;g^uKD*w1eQAg$U>nbqK@heI_9HR$XjvFh#k2U#7^e4F|H2fkj8yGztCO*lQ
zNUsfsYQPC@|0pd*dJr$*04d7J3}|TuY@KBg2>S?M^Ucs+G=Wzdv>Z#&K~=&@DVT#j
z<xT;dNlr^@w1%s-S?O^>=iSl2Gh91WWA6JhsX)uE?{7Ra^55@2+#|UoDoj@F*<IVx
zWQcwxDVqMWQ%-vCiFSM^^O)7?H{Ovc2mNPKl^{QX#|tvUKfzawZ49fRNsWHUyD+D5
z&@fr3?&9vq;ZOfCyBmg;nq8W>5x<QW6AF|cb=$A69n5Ly@~G&$i$dwU*JZ0)G&_>i
z=(!QQ5%?Fl7IRcD?B{;32O@X=f4=aQAkzUcS4lNTMnaHXET(`jI2BCLLkJO`_DLk3
z7?NT?WeJj)rS&{>3j;$@MptE8{r<dK%e;Q+$aaGz2LA>+X~uTRP!ioAng`8Qy<C~~
z{k&Dr1#0#Jw{6odO%5ep3%glU#*1^|{qjt;WBK}P;8UFmljZI8EB4=U46_1`rC*W>
zt(pR?|NoXjEq!Foyo|_;1B^%HN5dE#C>xZiXt+{2S?Ni9s0Cg+l_L<IAiIrB1MpzH
zTs)}wh%c)9-c*QNe>6y0ccB`COwbzbV9JI7;)%AR!6wGNuN8!#Zy<iH7`&1%2|DUg
zm!qdZ^$-oY-XBkvs<GW@9REBq;Ff4YqW?P@VIM+4wWYHovnFj|4D_i2|GS+GBw?c(
z#?K!E!{o_=^XVxhX--f;K7T59f;QH2Xr8f2%=s#S63KIO_UH=ju4sd7_(bPieLf{C
z2bp;TXqEw-4nVl6w!BfUW25U1*jFjw%hsYp>nt?9L}CmcP7wK;TuU5FiplWF`crSF
zeELmQ7K<19XUpa_f5fkBCY>Gg*f@a}w3xtO)k59q=OV1ZwtJlaJ}$QXjr5ED(*pXZ
z9|x2d|0JuR&xEj?HCnp9A<|&54tZXo7+Pku=O={ZvrD`Az_~z-yt12+!x_7P?3*qc
z&S~P<`jB6OVRZ|(v}i;r&-tY7$M2}!cOgOVxRy3XzkFZ&2!Y7NOpaO!JW$RXv9^AP
zE2+6*-0a`IUN76Du2@rZoGvavGQD1Nbe)XxU;oGJQOv3HP|Xd;`Ss}I9A3-q%h+e&
zzlfB}pFBN*Y;(W<QtPk?MA!FL`BxjU{l`ibU+VG>^Y{lmPHl~~;L|S;0l*%2Z(1hm
zL&dyj*lg*`GanEPN;F!ZNto@60ZJu;@mU8e2brpI$w_|2d<TlY9^!Hpk3Wy5iJv#~
zfd7L1{^gRaXR@v)%pt(!2HA(c2KNfAOKsL_ME%!sbp!h+N>|%(D5eC9vu?q*VD0wV
zn&XtOGpk;~%W*5RGA3@^c5Y|ze3$dnm)ruFF_>bb4%Q<?M;HJyG^onJ7GQpaa-k|i
z^~W_1vky#V4JU4@*$pcV2YT^gGj|VSJFce7U(%e&(D3rb)!`C+dKP)98|mKV6&6T#
zL1R9jp8ljlr{wBv{J9)IzLR4~!!=d{)_=V(oF0vcOJghr2ICW!Ii+^FmvQOWT6p44
zI_u2TDIit_5r*kBsl1UVy!#qjEHIBU{!qHXc;;Qn!=p6Wum=$&!!p7-_$!J;FyLgU
z7_<&ls9SYU9j8D4vaW;#xRTR$?^oGONzK_dH;pv;`nWfjhd`x*O!w2o<MaAA*4%-t
zE-3j2He#*)ulk`U57-K^|HhuEN8tz2;uC~;f45H!R-CqIP?Te%AGd^H&nvF-4hq$0
z4l&_IPTR=qycUml(yF*B-fbgu$838bK+o+&BkXhG@;2w{!>in~0&3&r-@DTRZtY2j
zg0|Fcjx3UUQ_{bD_1r610IbHFqdWssImD+t6lz5Ap?nBS&b8MCVNjuC`_{aRA|mW)
zeKqm_D1rWiM>t$=#lV&zd+r|`MZIdFjae<~n3gY>udtuRI}Zp5+d5`($9WL^^$hib
zIlW2JaP<yT#iW;ZHKN!g_32?9<@8j4JSYS3$lLDm_1VftP7-`5|BJNu3~Orb_I_<h
zlO_lXktV$ggc77k7ilU=FVY00hu&4HNRRX;B1OR<MIcn^M0$}90TOyoXrXy#*7NN5
zJ$vu>oV{Mxb*&Fx%MZ@XdEfWA#~8o=e>hE_N3ipa0mDjeU%5R7JMtX7pX-ZO-HDiu
zdVWsV#{PUJAeA&139ra5Sqdx245Q5;{D_@%fQz~P?;uU#3!G6>Kw|ThaCQ6t?NVQ$
zzbkkD=#fQHSS)2rjWyedjWoGnnw$eut4BS1L!3}yf~MMCCzbDZQDN*@MQz>%VfqH{
zv7ZTcDvqkxuPQ|QA#{S8_^D^=N!e-A-I!=noq|$XoT@gDNhY0H72WU9q-U^xy0Ce4
zr_FmS^r)ebdQ4y4j4tg7<r77iuI<AEpnS^n{$JHqCM<A;cKth-&mi&_fcpgBiTAV{
zF}!Ah>=bZ{Xm4jQk1mGSk{zOg<-@61Vj=EJ0=hh_cPlZXPCwuCu1r-6J2&j>o#JwS
zd-#f2&z_wedb20=6`o)%I*3!3GQvC7fqkM5gks#s8HQa3H|FlU)USEn9&3-fmkTb#
z<yXALf%yHV=Bz{8VCdt4{PkP}TZ`9quW`-MF@}q^%&t?hKZ<`}D5!@qz@gcWoSuiT
z^E#qbfX?F%b<B)6XClL>j~+jjcTWdH8jucrAfz^whSPTKBLGD6_h=e}9sK>Ewa<mF
ztNAGlfNxlbOYn(trfUmnpSJv&2e%Y474rs}z;Gcs5BliQLhH(6%D*hChPChYZU&$3
zcB0#8b=uUqvqFQ<*3IG9MZLGCIq|OY+^&qche~Kqro&>6+_$g3_R?SmUs3ytutW~*
zL2PXVM!!)vSBj9B*HG^Y%Bu#TU+mm-Br82DAORv)(QUP~nVW7ATGZt#IJh23PQ!)W
zwnm`q&;?mSP^mnwJ(;_D*;1nl6tint{P9bx)qlQ_{om|23(>%1kG<7l;<Mv@ccO%a
zO<cdcT7S0eZm0Y?4U<!40<9OF9&+>nBa56#;rI!l-8a?XA>Z2{GvDR|1w3j1<ab8|
zhc6{6S-QeZD4E>RT!;;s)Zv)S@PmRosql|R;6+563#WsMT@;76sZk*0Y0wAex5jsg
zc3&O*IQy!i36DD({YpJ^Y;{G_F&rSTXgsG$o-<6Yk1B2*%sfUdOM^9UzvU_NG_}Mn
zBO_S)lMY#3GUaTLo;6M{7cb&|{ev&RzP4s^XgFE#i+l9!*<#ytPny{7PyW*9owvO`
zE<wk?S@gM$7)arFESLp)8vGAC5X;%`a?Unv`^r*TaF*FR@@L>axf91q=b1?-wA|_&
zZH+q(F(=Ow2du}QY1=;$W*4UWe5*pE7!<Xn-}^@LfU_;`Hs$nzZX4BBgwH@sq+Ag5
z;CMC(s*%_Y<?#A^le+2bLJMgrd!{(O!sewbx!Xqkgx_pFTho`UIhnX_H_fHbrN?~N
z{7>@-my6`LZr_igqMJO$<4iS{`|fRT)V7w#7lajTDeZPXob+AJiXfP^Z8x64Y11@2
z-)QFo6Mg`ptcYbBNlMdJDN$Cks$T|k@srRO9EMk*y_w6Rk&x2QJE~jtSFhwysqXxg
zr$&`S7Vdc|p0?yr>-?s6n^9Dz_(ig$qYAcp-xi}#`J9nG)`$r^rHu$a<<iA^38p4W
z5}zQwQ<Bhc5l%6|&>yO>MbX#)C5ZgDJi3Tb7T_oT&k*BZz)*JtX)ocIvan|_=N1%n
zP@F35f&mSg$I3u6X>W#eEvasywr8x8?Y!JI&$KY_jq$RYQ*8I(muLKvkpuqQ4a>EQ
zqUXRu@IVEdx%Y}|gRnkG88C}1i+G2aU`%(w(KCJhx&R(<XB<ltCM>hknz^`TPem1o
zN9OSLs`Bgzp2qQ%`UXFYPu%#7c#)ck_~qSrV;WQg_X6cQn;%v(QdNdmF)@V5t?fAr
z#mAd7tn!IebrZGls$6c9(N>r{x{}wCZE=sQY=ic3&qKecM~rp-wOikhXI|X$<&+fA
zpykqZRdjEDLzD*MIutP;eXsO2wcNG^d-~JyvyGb)Gq+CmMkrTKAhO)<Kud!_6{acb
z21`BMwyhc|m%`jLI7_>si~a`I4K5hy!7hUjUvx79s9V9nCwZob_FKNW^1(;*5!!ZK
zsD31P5py?-MM<J(+H1-Q=sJpM+69ptbgf*V(*)72blr(-vxvAIes+jL2?B3&@cX0k
zSE8on)a#;X!w>jJ$AC0{^FSVOq;uP8CKg66Pdb!(qJl)T=(j34`y;3%m|+W&_NQTq
zm&qU74ZX#`qZl6gY#(qRYO2&*d=o8KD=(NO$$!y2U^mP^p*K~e_4}8XMg{7Vq9@1f
zdjabFOZTi8CpsyP^Olhwkd!_gM+`SrxQXng`8RkBz3}~3{}~z&xf{C_#_Tot-*Z%W
z;Zt82r_;IAR!z^?V%A-VT(1=)bMbEGO(94M=ZOeKHFr;zxkZ%>rDqnE+4-X%QM`5f
z3_TWB=?}{=;@l8xnQplz-xuxg8xy9Js~#@i4ct^81_$)nwR=of+B?f4gj+H0pWfR(
zbeC9k@db%;*z$q#R{6(b)wd$L8<mDSHm~8$9_J#N6*P0EQ@WAMFZ10DiW1aD#mRi-
zxevv<nS2Ki#nwGg2SM6f4#|?o#f>?<V8iWaECosLS~2IVr#G<>Xt!WBP*rWen~N8G
z&W0v`m)DGCI3rjdI8(F6(w`KYpOFQh<8o#WQ&!DCKeHQJx$l_dw}n`C1?(5+YS6K*
zw`?<C6BN@zgfPF^h{J^BB*k+i`f}o!C7nUpfd__TUmiN38f7QHJ1k$}uUp<K?-qhP
z2otke^7HaTz+Q(RM6}W3@JDDl;8)OcYnvnzE6D{QoG@Jpu_=RjYIQZ3#_o!@v~n~|
zac^@(1Lb~`(qc^7Z;JL7_N8voMxShmMgR7=B*$`cQ4VGmAQE(~AZ9CM&Gww+ci&7x
zodti7&s=rtUI#4o2&sk-Wa>u;+xs=3I@nkO1ZV?pI!t2D4GZjzYFGEUd>>n7Lc5x)
zq&(w~6`-g^L+PeuN5iqw0{bq`y1n5XRPgUJ&Qq>oyIV_Dd*^~Pdwvy~)i__R*KM|<
zb?*ET$bcabSM_%f-NtgwYM9ai2jku6uh80pX;cu7;G`~#Tvg{Iv{W}cYIx&W{#0Up
zG3TLpH&gd*R6mF#v0*rI5pewQy2Fi$-nKoMw9M=JGIzJZ1wcblwi}P_6y1Nd{EMk>
ziET1?tXcj}z}zSEOQ`A7i5tlau}zg&{p&MoF$=J!(H)C6Unwf-71n9s5WXpD9qIoA
zHwL2j(9TdCM83(PFF9V<eBy}ZbephAu^2P;HnrU7(3rA@X?+6(lz9$gaOYWVy-Iv{
z24nEELU&x4mWK0*1iG=7B}F`2x85~tx!uz_7`Wt2`(bZt<{iX%+Flm%8DL5@M~iSe
z&;7W#6<{wUcu2c`QXsQJOw~LmCv5TUlkaj;-_73`)ML-#J`(3?(D%cQcV)t;*!i>U
z&H>5f#kpX6^{pggMhHe?RxFvSvCW{brq4&KI<$1ePZXY?0Zqk(oethC1(l!ful?%A
zzbfcm;OB>57YLkL^<>0jsF(%b)yAAcned|?(fxd5f-~=F8<t8uljYAJAunP^`d3j#
zFT{lf$G|bgS2VCmCSy|JnD`v4fv2vbeNqCml`@GM$IwIv#uQ|poLFz8{hs+^d_V!@
z>3O2t3yqliZyTR=hRwTsEu@<;(n{w(e)C}r>Hll%<o(cXF4Q`w%X0i`)y<_A9k@}}
zX~Om^FtMZZbS^co4?6b&u<Po1tJ51nl=;!~onVz~l4X!B63r22Gg-q@SMm7A5YMCU
zOqYxwdZPlpud&EYe~&NszaVd5Tg(^JJ)M*5%->NHx20p1jr8>5%#B?7{*7FaSbNAx
zvoP7La1}2IDGkGM##Zg!BxgYGwl4`7MI){Y8Qo>VE%!Emtj<xKnIUGc2?}gG+#+h~
zZz{8!<ThA|HP7lVeQQ7;f!wTjsV@wm#&yPEycYQyvZH}_1ns)pO9Qdi0Gab`Jb_Oo
z@y~esx+OhdUDqxhUTHf0o_5|tf9F^do%9>Ltr1YID9~#-Dh)xG9-PjHT+UC>#lBy9
zbolux<?uRVXFRB!GxmmZ&`-s__uN9<!=fc%N^<c)d0fODuUs2L4qKzE-%Y;cAC`Iy
zmv{Lw1iwO`_^N;k4kf!!`29)Cw={xB58V+E>!8XK&+rZDzS&nm{161tf>$27y7THZ
z`+$1~b77Vj<5mrta?C|u1B7L3OLz|4t2_?RII<q%KKR%yZiX{C|1fg+x>(yfL|REC
zmL`NK`Iv$-9#~-o%Uvj~UE@29;W9S6)k!~Ojq-pOi!@=5tJAWTL}Ptl&Q5pW(!utL
zCrME^F2T8nzvPdFba-~SUzxaz5ykZy2g$w>UV6KDnt8LSw<JW;{wAZ}*<}hz=a`)(
z!B=L%ic)xrkze{6j1$5JevQ8!jF}@|PIu6Ic)2Z<`U=QnB(o$O%wrr@fp7t{$pn={
z`#(n<<;6l%+x7KKI1P?#hu6W!!oedS<#FvL@^5VN^YgbgG^)W=hU)_FHohe7*`i%P
z=R+RgTJEm)pXTaSCv;<QkG~E{35;~zSr>9NU4DN5xaX;^5HiQ@NDFw!$2W{cc?}Xl
zYvsGu0lJjDqZ5J2p5xD&bC;^?oj`T-qsA`cxritWMV#>uGCs9O`LC>UKPl~ScKY6(
z5RXIIXZtRt!Bt?2uw6Psx;V{F#h>PO&VWS<_k(`<-vML2$yD40g+<oYIfU>5@8++|
zz)BhY_=QZg6g*>pP+;T0qp~^czr7hsmX)@<R++AS_?Vv#c?QF;Al%YA7fKJLA9A8^
zq-Q6a+4fv_ck9VfGFw(+T>eHELuwwan;2?#<CM|Y5a>RK6H}^~c7COn@#Sa{(P=%T
z&z+5-=kHvGbTw*StgpJ^)+IG-6{PmV{KP=7T<Ylx;V=mqO9#pzjZfbz1yQ9i%b6lR
z@|G{~T_mh*+O_`E{cQO}x2!2x|NBq&Ci$$so889T`g^uj@M3v~dTbJ<ebSq~{k7T-
zkx%ltna}4XLCzq0Vr+h1OGV>Hn*_FMPCS^hi4?04yBBe%swP3$C{YRRf@jsoM9PD9
zr#!zFCTFzX*zA`YZFcg5J)@#cM_q~%j}_>CNKZ^RFMP70I$B|?a~Oju$Z%e}inU;n
zZ6x8Po1i!t(h&AsHF&@)m6olWx&cJrnss=cNwuDrvBmrgqgci6ZApZ-tp<Y!<L&r@
zDTF3*zS0-imjO(5o`<D$;#R4Bjk##7MeYM&z&2;hZpoFMNxiUvYY@)2%0zVO2{Lr_
zI%;713-u9kuTVi&%HCVof{y^*yX-|z$BQuml3MEMw$#M2-^>_$70}SpW#V9fvQMLc
zR<cvB((c{!l9$Lu1biyJ*PtD*aW)@0)3#@mxny!UeDZ{MV13k~&xkvMF1;ibSlY`H
zXn;U7_zdFnFh$?|GGP6VE9dQGdA3A>je$Lgo(x+gOLx?ofK`?oPVyITeG*RvEs~8L
zH~3(qvu@Raic)|Z%z8ZaY1AFDMdl_WlNMc7PQ-vV)Y|L6bQL*TCxirg#yjqm9`-e5
zIn$l8PkLQ|0qU%&_hTHD*|!Vi*)=XAfKlnwLkeyn8BYgBv`%%LjhCyV=j2TaFDzDP
z$olQMnhx*fCJD`%*am^MS|ie>Imad#zX+yJdg@hk#<(%|e|u!TnG=iNUke@JPxp>9
z;EG*;iS&umTw;0rX#G2RX>R>ZZYqUov`1*QlQ6rxcbly6D|si5`s1|@h%nCH#`eWe
zrStxTC5<uhST>dxBz^pihWhdaBk9C|LEz1-&<0qPS^i`oG3Qz1R@b#|e|BkTEXzbc
zPu}5Of!Ft8C;M6PSd0AorN*TYg!Ua4jOJOItV!Z|X1jSlv}nLn9NTQm?$0xjgICIg
z76tIyBBg|wl8L%MYU{tJOTRYU|DaV;!OVbu;v<sf4KEl5;_A+IOQnFe9f~n29VvYy
z(_%cU1p#S6zbJD^KMieqF*9)JUZ7W|oOVYA9wHJfAmXG=NSW<=+z!DuD~7qd1!-RN
zOGIk?GWRPReM3Egqq)a-OKr1#I|7xEGCW|&!i_@i`3U@}ms)VIy4<(K=b6qr(wETq
zOBVc;YRxmYIf2u=`1=^IULC<KwcLTd)l<9Lm~OgJy17%;ibuJ;zXe5{O)X7q&)LCf
zDeCJ1tUYX!E}~GES>TxctrURMdF5Ql$~;V=uqhK7vyAC(Y$WNu+h^{ed!avMnQ#?n
zsx!wi412Uc)k?TAHq`9!EIaXZFmcL2WKkFpk5nWyZeeN}bNEl^BhNDfMh6q?`Il?I
zKKU?KJeFxvn7{x9QpL@eDbv}Q(nQG9m0Cu`e4S=pA6E7=hzp;WAU~rjCP2P#O9@Wd
zAj`pMA^tiCa^<h!D!9+sq)ZPIL5#^nzJn}pChCMIZZw*JpgA{g;ednI571fw_2PoC
zEXp4d;M*LAyS6fdX}~(esWT0Wu*}Hgu_K#dOYOLi!vhwjP3L}NA3rOKrBNw$>>5m~
z9SYMSW$e_M_oV9St6qI$ax=0k$lY~jGQO>%DQf-m!MSdCAx-QZnm&cwmZewEySNk9
zJs8*S0l^6_OGDK_<K|4W@CZx7@iiy-Xjzek!qTwwxB08(+`|me%ZXgxV6dL1;jIVc
zy<o7MgRurt=`)Bm#Ed!OwEaEm33u2FdJoj1lgnp=K~D!6ybU=?@U>xj>U~eGRxGl-
zr;^c}_Yq||Jk1RgQoo~1rv3JVAgem1xYp<X#iiIArlr_cY&(1>b23zhH=@=hvGs<m
zF#l~#<hE_2@wk!#N<n?L%O=tIU#EZ~Y}*zNyC;}p4|Mh4)PI}i^ylG#Je}1`R@y2|
zzKSkc-}v_-5FM5omcz!vYNFjZS5rbouT?r_hYU#EWh}&h<{VxvhdY=7(Z+tY2&mN}
z0S}q10xf2kTNpcH8;e~O-&N^=L`5;AH<G6%UK*ytp{$mMAIO>Aev|H7lp#iRjtdU;
z##I$HxZMj4gx}v=d=3fH_DZ?U81~{O68|9EV%j70MS3d-u?e?1xUs9;2t52?%bXXe
zr8agQX{C_%I+m=|if8vHVI)m4;=T7$V22$1P03`7tn2AXOwgi`h?SxRLl9qZ(oYNe
z_e^@4o;%cfBuPp$`oquoBEsI(@1AJ}YX?m#HYP5L*oWt{6864f=c9r0l%?1g2K1OH
z&=>haV{mu-eAIz$CTZ_4K-R*4fjpHyApE7#&doK6umW{O++g6Zn(Q8bmDQUs&5%bJ
z2_g5ptdy#4t6It2;2W!F<jvTF`C8{;hBV5npM=XRSA#T0Q%HJGL%7KD6W0sG2Z`fI
z&ZjMp#5!{iz7o?+S5uubRktm4#)kYxYLL+PhYI^DYYKNpD+AI9ElQI2{nb0`CP_yU
zl)ep^XhoRmCh)Z+b3t6WAPz_+yn<i#xE8kll1}}&_dA)^V$Kw7&QC8uBOLqbue}Gv
zfb$YXcVCc@a3iM@;t8<%@E-<7;1zt}ByOYYg!GElMS88|;L!s`Jb0^I!K}*y8*vBy
zO^OGIJEv??5QChLz-fkiknUUouPFXE8+>1FQ8L|eP?N0JT3V})uQ|dQ%P@>l@=;l4
z`$Z~Zu6gAm@fl6-O``AYpl1T_e2QoX#r2dIXX@BCYS>PR4>}ie4}Tc%Bt`a6GdFea
z$xd+Jr*1D7YSB+dt0oMBe&8@l2FBnfrxC+e;HQ2WIIVhYOW|wn29PM{uJ50Thzq>?
zOD*`!u7LkpxJ^PkTf1~vL6se2M83iq1akAyb^9&CN-vUkcNGRmz7XI~g=%5??=xCG
zf&e-l&ipSB{nG945OS@bJ@?NKRqT!vh~@2E*+s|#X)*zZ>lpKz#Juj><A|6!RF&*$
z%I`PN4-AeFjOtUY>=mj*_#H)>G-p3pd!l|H1zDvuya6=l`=86e|E*$1fXDy8$rC)e
z7C81g6DxyX5qgi)f<tjw3G<9C^*3jW!fk5V8sP=6b+!#ZzLvwf+Sjjn_$~+K+!~Px
zO5qj09Q%anS?fFZF7?^oMK=ZKJ(U-LY?UX1K)TXP{!7*OcN!-I=zCxz>;-6%c#Bkq
zCjBCP_r5EN=Y7P$f;|8eDW0~KMc#7^JvwlrX$q8Fn4q=s4+YgBI}(#a&LaojKY*yR
zu6ypxWek+u{lkBJn(&V+vk1NFbgtMq_?fIZ=2^wK+d~tXbAEfl*jqKA6CP=PdKu8I
zvf35Vb~opZzn5{R%V_L+NK*{J^jCwbd9TCbzAa&ESS63y0UnO=Kl(6VQ-RLJHGshS
zLJ(VD`xk@tZ+7Y9zh%DvTgPU%rAX@>+rvnc-sq5ZZDK!3@V$AQ!s$}5iuKm3mC;RV
zPg#oHHn|;#MtTLtx8E?;sS4=l_cs(Z+#crgxs@|ZXr~OBFf;sd6Z9`auh-OWNiBUp
zfz~b6ea24Zu0)TDOfBs~XB^PIG!3P?9qrl`OtUDp634Ch1pC5_{lOiYFBhBEJx-Pi
zvn}{}>jovb^VR>^VmJ*eh*S~@azQ$Od#(TO`5Luy|L@1jOG}Y1;}_=F^;>eV*(6Ay
z+nr?~D|W+5atfGhXplUB5i3cP5&q+L_FtG#dfmV^&Y!TOABWTuBkY;-q-Ht%G}BGO
z3&vFn<BN2mR_2N|K8;KkqguA?^cu&hD0kJybqS{M1lp2ZO#4>lhAO#Hhe!tdKYWU&
zf%g${CQ6vMu9=dz20Qy({CA#TsXnC}Y>i;`S*8hfm-8M@302U>?q{_p%DpEJ<~ga|
z7O{aOS{<`XFmdQa={%5z*r5~BCH~=qd>qJxI28Ev2m@J-dU~b<HglQ<eM&7te1xx;
zUtn%h8gU~B?#ppIcyj9Atd-i1C-*yYWs2;uj!}!Zw%;bdH#I#k6J~$PY>8l1%)<2Q
zX+6%F9`^~?V!|l>6N1HK1h@d&_M~6!w`I^rk@lpT8d{k%k2oguR6i`QE6E7uz-HdR
z)&w{*Fl|z!8pYqlX0VN+1HjSS5k*x$Dd_1`P7f0Ck17+1&FWH(lN~&!QoM3vQtuDd
zhW1{lzxt<r76PXQR>_69k9r(k1jrgHTbcytS{iMI7KEs*PYk^>UDiIyI{lB}L$@QH
zb4CR~P9@{RTJO#Ls@gR<qeEFVo;)c++L&jCL|%}pUhN;9B~gmF^-|kzzaTkHRB;5i
zCG1WMG}KV1L?P66!~M(kw=P9q?k|>VGhc_t%D?tT$I<K~$WpQNmXycCH7v#|Ry65l
zq*|@e|NGu5L29L_OP4P8q9=S=lB_rKYf5IaGBE8Ls|b4|@&)Fe;(8nrX<kP**l@yD
z>*NoJtHO?O=l3^3DcPnL6HdurKbi@w_W}lm?9Kk4{s}2mzymG>9)O2Ff%oGO%>9}&
zxWnSrs}|jb%Vn<fPnEajNT-7@xDM+@GRu=S-#cVA0u#Bh_6buRJi5YHngG-RvwM?`
zW)@|K{;=0bcoRUZU!}MTW@?mq59#DIX1nYS;Jk7R51554vSeHt=i;_QcJS|BQ8HbW
zbh-HZd+-P6W_^+~X__=<OW_Ef_`pR&4uAI*R)#9&9p=@4!r$?d^%_<bbSq83+-co-
zIw^TXe~p_;=*RfXTVE+P4cme1`Q}b6Dubg}U-QLkY9y>ix4d7@VMtFnaY&=OFni}8
z?Frf338o6$fVW-6mkY}=6{#rw<Ak`Y5GwPw7nQDL3kh*-etQqDNdBgql|5-@!AsR6
zXb0wm`>3u1?F7tIF6g=Q+VvnDFV~X_&Nqh-{>jOq_1La2P43oXxCClbJUVNjM}_&N
ziuF=GV<=!hvf)AY(YRW{b=ld@2cOdwD9e1YuE*~g`$NbMmGb+jl24DVf1ZO2&5T)n
z;{MTfha+sMWR*mgDw~Ck?6PFM4(lb6sKB<0PF4_<c5TX;vJ|aHk_{nl9yz4>)uXJL
z%RAQk`H^y7t;@Z<)DW^owXB$sOp|Y<T1pasc+>DxgrV$$D1;;Dx)DU|boiJTH~k_V
znp9;@nM31lzdiPnfAjn3F$v_S<BVb(qJ67Qb1;=83n2arMEDte%hGc)Ds1$?fBF`R
zP3&IGRDSx((~GUjM-Q(Xj$Th_Z2ZLrMpR>lj<~6=dYOB6*i7)PJqmlVJzM(xxCGMW
zMyW}Yrp@{UJSyypvr?4}iknN$`8eNnkA34eOPemLFDGxK-$evJK99a<&0j#94ne3c
zwVwK%CX+%x{Lv|5u&D?O`y9WB@f!lmHJBb6W{YTPC}EFD&6dB=r~B@9LIilP$Rgun
zdFkI$V@^|FJ0?YJz2`O0)!$|i@jP(EOjk*_1qIz(3cOk$aI2-xD?(UN?6A#+CP?DK
z&N`O(v+JL3>GMKGSl{P$q>s>W0>TtM0+vI>tcu5WzpRZRN}n=f+&i|vQtG)+s^$Br
zE1e~x&m_eFBDbwWef8+&+SB^ULgB`Aai;y2V7lX-iE_kqvK=%1xyC?hR3&GNTI~I;
z5L=FOa->v>{DZKN|FYYZm;t;!_J8zDX?}U83X_Zudpl2HD@Js6<`dMIoAcu2YTIhG
z{ed2HR(`c@yG_1r(JqzjUc5g~CHT|r*VAw;ykcXB@1scK-FP$}d*gcYwF?)1@2M**
zJau&Cn(zZIH`c}>Ijt$?$af_=ESfp{;!rbIb}?n4ojDyJiouFywym053w4Ledy|Vh
z%Ubj?hMZ|9F>0S3|MX4kIlOyuy)~SowS&nedOL92{|GRh-&UOTZaOgySiXlce351W
zJpqOjTrMf=VANOuo5kSW<i(x0fez|B{=4?h`Vj-lD{nuRBvn7t7Etc7*Jx|uQ5B+m
z%x_sXTHJ|I&&loQ7q;5!Nh3MEK*27`^yNkKda&%9j4H)(XLsfPtanE+1{&W?ILGrP
zywy{wKizS^-E-6ELMuN#e`-AI1-B*b^)PbY9?I8v=P#0vSls^#%|$8%A3fqsVHcX2
zPSv(k#eY^)w!3OUsoR(ys&DQ~<X)hqmWrbu=~LPn?Ip=MB5E<N3iCdkB~g=mUGsvO
z9Vr~EWykl&k4Kw}yd(|`aWsJ<ISGU0+-7~3=cnEKO*?LSAFmdoi}GSyRGFMGMj~Uz
zF0vayTlxr?DiI6Qk-G)dAtm}`aGIH3llDM=AN}q3`H2<KD)$gfcli1}nbsKKlcMfj
z1q@>U{agKVk?5b@52I&nApM-iVtGGck$#dNbT)ojt?8S8JS;A?8jUt2`Le{CDRcAj
zvK4_*3iaM33%+lBMNE$FG2>f<ea_M!fjIjvl>Q$@#pahR*CJI5OTK)8CjFq2JG!vl
zaF|^Iw9l<ZiXPttX8lww=L9qBoB489O*!9ltmu`HI^AjA1;$*o28K*jyO<N^^?Z*$
z*hZo98&(T4Tmg>F5*X&wL2v9he}$)M(7G0wc89R4=HP&E>O~Ah^YTq->Bat{4%?Yt
zslYBH-!<iOx$JwUvJ02bmL9eP12pHgZpG}xNRa$K2?h*VNSbxk?xaJm204q!Ai1%!
zBN}N%=16x3-nl=#3zbOZA{CW~J*D|n9-EqG@ms4~ot79@xSniQ*U@Xzzag{$ZpsDJ
zFe%#sbW0H&uBxRKqtPAj=DyU^JYHrsh;X+h;8vu4?tPF>QlSC|dm>fD$U1vdv5H;z
zYlAFM=~a<XeGEo{=?D28TZnwiC*i^mz^8oP@hNV;!IP>50Fzp^j9dJ%Qde486dy^B
zZMKd+0wxTQ%oFRm{c$+|4Wrk36BgBnyxRaY^OMpm`lQK?T^hvCPnYHEBG}ulMH{yI
z<74}6wz`RbP)JU{YNfn)jTq#5A=_txGys_2<m|7Zm@a9C0s>0z#x@KWd8Z#c&q`mh
zUL2|U_V5xF;>$$p`EJI|QFPN$ElO~wHalu1ym#)*Thiz3-j9s4&yj<mOOk$H<L+ZU
zy4htY@F!pXDApb21mq!Y=)Yj9Fw*>^V`lz>2ebYzV|;j^jb(AXx4I)G{n=69&hJKg
zt#(=arJm6q!d%azv+dwGNmM{Z`?L5xKl>{XlYQWhpBQq!Go?wY!V~`Z)dvfjdLq~$
zFx1({%J2F7`sPmi<-!kZuUqG48ZWmnRVyEjmjQ0Bb64zzKGp1DGTqsqPD~0O_Q%LL
z{OMXd?FS~<TX9v3Xe-H`9tCnXRZIh2dJ(sUHk%=S57R>pI8MR!*E4;K5DE!bl=qVo
z(MC=---as?gMCw56;Sy1m^(dW{p^%^ADDMNXi{Cy7@oy3U*3+*0&HbS<~<!)*S^ue
zIamk8pT^38a3ul75zmIjfXp(K9l)pZJQ^wA()s-7&u$F%l(i2yx*fPTsLcW8F!b{O
zRY?XHqY|G<q3n;ikW$)s{8I9XVBurmg9+QF$MG5W_R@i1@A9kKE_{bqNCku)WW>Cr
z=|3e|@uv7lY}{Jssc}W}l;rW$KJ{-~tv07Ig%M|^(A!nY0jSv>VpQ}oB&U@eN}^KP
zkt^l3!WMr*51;=0{pKGBOvHcyef{SO+kUU%^b%n1*f^n26j=%6+fxQN_m|w$DB%U`
zbo!^+IDoii8a1@N&28Ynw=%e3rup-7%p4&Uy}ik7%*x5)l|@0{d;h3UDUum~Ye7O@
z8Nj(CrAzn~-7mL|eHC`QUNv*KY1YfhM~CHI?O@iNcvmYV(VhHL3-w4V?4*NN7f}EH
z*>13d0yn=HMO-MYwkysSh$*pf@kEpzl!=^MX*Pgjbx(lx+UgFSaV8faR)!;%(#HdE
zy@SASrq_QyLb_Mm)d}q9$ceBt)K9X*%pvPFJD<2yvSuygJC4=aBoArQ0kVEtTlvLF
z4CnM+)E8cM@Y=UqEg8X4A8@;oUz(seTWk=XE*<1Kx<*0yU+VyWlkH9i8nwoEgqu#G
zlGN1gOGxgVXtM@Cgkl0G_}0+it@00=sdt;dzu2l>hmQj^(at;b)0T!gW24ve`$aVM
z=fC*C#V=TyPh>v&A1bhZ9CEqyrW_roZ08*MO)zXp8c^1_(?1UN8o%!t7)rHCy!}pL
z$6*<9LN8Hecf8d9-yZsJ7D7X~xWS~e0bZW6=SS$WQ6mRLf|YR1+$E3Iq4pc$0IV?m
z-EXMb9=FunI`Ua}cRqqeT454ydT#?jbjMv=DQ09cC1-u^bi}cs7RWRwX$Zkj#SVHQ
zCpF}-CoOViCSneS5QA<?Qs*Mzk=<G}l{43ZrYLBt%v--h!|Ad&CFn$IV?!-Z_1t?Z
z><pxCAnv-(*k1N(4TFIHv>0KRs0h=Rahe2sY>d+wJ0tSf&wp-<drvt@ns`mv_+S#$
zrj-(jLON0;!|bQ3oMu=R9f{87se)Hp(6RR@Sps%0e);l+(cJ~|;9kobbvnB+%?^{6
zx#N=56_=Tq$tiE;K%iGpQMcjA^uwWo6S^t!8g+iMb1(9O)9*VpxlGNp3AZX4F8ZNf
zN6z{)?z?m6O6<@;pR>j)^Wzz@htK{fnI7N><01!<&+g2(x&hN=t=7L*ds(_~ejn@5
zGj>{RoSF_ioXJZ4`2FRtpPsNqFc-+eD@J_O)jpUo5e~nc<4tu*?xqprQ0)C#y|{sB
znlO=igzlr?Te>?gC#3ZU>q}uTfRW$MZ%mA7vTu*kB_J*;mKC7^KNm^GAFp3;ImFm)
zkdx4BscAD0e1A<W{~i~8+L8BE;!m%iT0qZ>UP^ikprd!U>b4gM=n-#W=0Y21AwRyE
zD(Dqq&~f)TFn3B8fMtl5Y~@eV9Pj}ERA!Q5wDjvbd9)eKjKSCVmm(}q*Bl#piKTfV
zY=?{TzQ;B*$-R<@Z@rV+cHvGE;ldUJL0VVX<?_5Xp73`&jnWJvM1UTgGI-Ned7s=g
zWuo@(D`9rASfkmvi}|nqOYZe|1<?bTP%d(&tYo%coD7hD5d$HE?;MEKb@n=hpf34{
z+E~K=-ex6iUC$qYbJ6v<zShfBGOq~C+YNR>o^*2!#Yq~Uz+xr4+EuN3zqZ#AK3j#6
zbvaRgnHUFYaf{|vF(G`xAywsluU{!Y?-(m(n`ex4K{4;XV_^G|{aUIA2{~aGUCg~x
zXoIs``s1BbE|^Z2N~eC!kQE=ZNXR6A>6;^5gjVND>#<Q4=VvE%iY*FnQp58`p{P&K
zpE;-l^|n%%@1$DhAkjS<!R*b(>T5Z2l(Q7Z&)qt|&Ove!LWFkK);H#l2bpT_cYJP3
zstlktm4UK8Jg97%#Vf@|PGpS}eYIFp>Q4h8;oNP%-t@H|VCe^dYpZ6q%)||n8?m0A
zp4ysNHi*NhKu^{^hF`ya!RMfL^ppk5m)UNOC4LzDG?qX8b<zcV=Zf2_yTBBQMa)H!
zL-H$mfbYRZn%&X17m2iEs6i-~vdJ4$TbJ=`%~^G9K3gCk7LTxMS3a2>-z$l^!fWDP
zancSU7a@QA_@4%>^$oRc%TYlm2nbj$j(IcG`glo9@OGw!U)mHxTVqT2TKR$c(s$8h
zRXxV<7;zgXs|Pzq$YJp@6LjbjFcUS|EwdybDlB7Gxy_<WPs>-Ri?4_(ga}3*FnxQT
z+?L%3%QG@lAx}R$$w+a?LfJIDOX7ESW3Yx{PWb3Lq=11OA>SWisoUnz*igB62N^u+
zanBeFVR3w7_adM3FXCn_Q{v5HCg^4Rdtp)ZPR0XsZ>XstFBsp-Ja)ry^%oz}s)IIs
z*nr;f^OIa8i@G8GjnzH%V!BVIlE{l%kf*^*alJn+q@5<Ob+fPb^MNaT@@_SfVihUz
zL8C0O(X9*j%~tv|6J1rdZNEQBNMg{cP8%?hC7k0_O@K;bnqsR0__(z~Wdfo032plV
zXr=opuvPnZRtFY1&LnXeCX?(u95`ICI`qI<gAwf`>!)Qnb?dr}5Q`rlcp_biSi9iY
zUag71!UHT>8wX~KF0GOWOMUW=dQ7W*fwxrz{}t|2=dY%D_CfxoZ9r6!9~XNm7ivat
zsCXbcHOC;Iv=`Gb;2B<<MI{3u3hw~*tjn9RT6RiAPkDb7uQ}`gW@BVet#fs9<Wfbf
z!fAJ+FtP(SrloWoh=--aT@hqj$70KYOSSAsLui^F?*sEvcrkB8Kw%kCsC-em$x9>j
zH`jPcGObwuTt%_oG`H%2cmS!9zfwE^(4fS=>SS7@-g-n4s{tb(fK|}15H5Vo(#Ax^
zZnydQbtcDeuN{7U8W_1$6eeOy4UWx?5-nI1<ySRXI<Wk)CzEpV^3|FaAN*npx<B|_
zOsIZGk)S4PGxeTX%nY?pTrR7}<yn<|Pd3#{MdhcMMFH|8RG9ayZ(;YlJDXS&SaJ`{
zLfS0GO<Y`DqqdLB7%Z-Z&9qZddTIl@N^%*>$7U=67!@Sq<|R`sZl9)?ik!h{d~l0}
zcm0P9uDn}*8(WfOIsrwx$NzlRf-O`?FhnSdIT?w3r|)&#F&YLHq<mGuP5lKJrR!<J
z1k-B;dl8Id{6v0de!WP5VE!QQKu&UopOPBcf2eTm=V_`1(X6)Ps^(7NPiR$CV8%n9
zY|l2B&TyWYJhg7Wg{_ZQ3QBLQW$UC;PnKNdZ*Vps!?quabsFc&<7i8tH51SeP|h6y
z*^*8FG=LFq1VFghI{e8-MVvpv?&6%1h$$<g&qsWw&($W3D;OCJ6h(Hco6aZ-bAqI`
z0XA+dE5)nhmhGJR@8gtWnVqRFjN9-sa#0<|h2YiaG4Kt{EMEjvT;6!I!2Mr7K;}8~
zT(wrdKg;4evL05jx+(?ad|zqP>G<lsnB{%tkY{_RWU2x;J{2j1as|J>ZiaWV;URFE
zP|fC5@#n{wevy@5DI&og`mfXyc$aB*i7z13WF-18Yi#GPD|QzQlHEUMUI}~v#|3F+
zN)1wY<eP3LyZ*9^Dyd1PG;>H4ZaB9FLC_WGUP%-<7XboLL<GEJnT)>x<P&pan4>eD
z3-DJ?y0m!zU$~qh@c;DFg>LKrYn>+0Lc5|?zge_S`wPA+Es<(1XN^0K;R=>o@Jp=Q
zD|f3A=pu#;v&0?|HZ<VWyKWM0qhPid@%-S?fQZ3+B?{Q~#*+^x1^;_zHyu?uO^v?Q
z)HWR|HJ9g>gK1|7T!}8sE557YtWl7A<MNi;{1{MrvGpK{l4Mkw=S(qq13ddQX`OK6
zk!4y0Maq5v@R@5p*Tb!1Ay?HvZdaTUi}d>65U5U;;Eg5nMd5n^cB^FwoykEDh?8cg
zdDzM~85Zq}O1Uk>ADMG<g^E`p^w+%HV%9i6z3L@blmFoJe*JxbT{gqRb`M6g!0#+e
zJ?U@93%f977-qLVO(#+<GA){7x{Rb(d$J!ALab#{e197pXxRnPcJp`IQ{vf~q&<pA
zdly*)XON!aa@{;&|NXT^KfrtFCqxUPZ)K8yA2L8VC%b5KS^?+=pg`r43K!jq{(9}X
z>o0$@0V_>v0h-Pqp44NJ`<y68{sT^ps!y!X_Kxq+F|Du0sCjg&je=YkTj2n@F3i%D
z=qFYREI^KJIbH*bHX^`e=`om2u*{R$F7!`m*BKzR_>Y5BGDDcQ@2xsp=hhdc@M~^i
z+Ae<O)ZbXY*qdNg^42RYf1LRUm6!|~%-fDVWn>XLRf)&>Vo#5LSWIFNE4!<YHadRl
zbrUpsK0q35njg}qV$l^ve1aDXmd;={?oaF3w+_I*?vM%(z+<O*z`R5FcT90#eeEU%
z_-{IT(hYXS3&u~UpbW#WKaqiA#;?+3!wNK&w&l)t_Df1>jva}d>UGL8$%*L_J<1P<
z+~YU5jC;v`Bsyr{!Vmgupx+5VzL-iku=B(7@A&qr15zs%x1&oUxoI(bp4?-<y6*57
z1kCP2il*T!zKVpvLts920wi$6Kbu2{>*%%6u7>}PMi|NTJ=?~yE>v~Hm>sZ6w~ktA
zg4=+@gwNJQPmD5!2+ta^XFIB4=yD=K;OH{HHEIHiEo)hWoQO(rgozMQy3XXkPhdbY
zG&ioYUe?L%j(W-LGxz$4Aj4vgc3iYr=OVviE{bH(Du1=6@qj<Y==nRM!$B+Ro_Fgq
z#CrAP3qO;4Y3``jM8q*i6m;gtV>|>;nQ?@&t9!p}M7FH|M?gHcGTTshVw3POo~jt*
zR+CwOHD+15H16YjnE@Y1#{PFEW$OaLf<t5yxB|{BL)@Sho#pGu2Pp0)nYa%3gaBE2
zV&V9dV9<xz?s;4(SDrU&OvI`~03uvv9Yx}4>bU3=(=n=bn8hs^?^HRmYxFU}ebIC~
zWR^*McG!9h<gBywcr_WL1NkJ+6O4HW;;LJ^8Iga>_ZQr2W4?J=GtqYPr;M&qZ2}ck
zHF$|i?RY&fQ7OD*9{X3sr~1-iVXq1l65@i~zn&MHyp%STU``RoDo{=s#0SYcKt?=}
zXFbUPBYI-kH27(*d%@`$w5uKuLhO$k1L&3|7?O)5W;SO9QN%RSe+TZmQ;roBUgH+T
z1ktdwvleB%B6kjeCX0FfPL&IkF3`$Yitdgc-ZZgQ5UTodR~8VxQDxH--BB~O&JOso
zLr~R8P6K6wtr~&)<HS6-;BZHxn)z_1X~{IYKDvTVX8#h}kdy5u`Jmq?DD?Tswzsc?
zB)#}j3ZR7vqWzh2#rLLsQ2mK#jNtR_rr&ZxwJYo!z<hc5Jt3AilP`Y?AsufI>lvPx
z_j!g-n?(wmJPq+BcPJ~WrTitH=#tbXm>Lm=TYhMU3tfIG5bt^#>gKb*uP<Vc)pyQW
zOK&eqc8439G_-yB;=2)M3Dy1u{FlguP8IC!g2Z&!Q(g1T@nP{Xw=ao?DodQ#v|x)@
z?KM;CZCCdn-W9AymY0C|3Qq@Z0A1J?M3DtdSt01VX(+RUP)KNNRvQ16vo4T_Ny|(;
z!~_ob!+P>KGoS?kI5_GFRZdV2UaETw;)HM%cFdP&=8qxgiyr4?%2x#akixcau6i?}
z`C?ivqxXrt!fd;_#ti9^DbRv?-jo+8a8<`HjBj+RMjKuU;)_6H0TpDPJg!r+FRaug
zL1{6rpdVRs*3=7*TXaIZf3oU~CFmLn7k2)M;8Qe9*>bjj?WJ!(NyL(n$`V>WPSrD`
zZ;M918lzubrgT+|xidVJxK4cABi?~@Q97n+_VI@smkadgAGqNGMP$u3f0}SyFp$wp
zCm??c5CN{5`qz!h<I7w8fqEn2vr*9TaLxh{wZHODXWjGe0Pd0-S1&JoxKEcy_pV3n
z(8dgD?jW9xIyP}@zp*7VGH#kYpKR>B8tOX+H#K$Ot#fd@K*U&g1kF-RqwMqa(iIvB
zujf6XV&a>pzQ&9V$B0w6&VoWMl!(#2%^_`9t#+6~2AWV8>ajnIp{ELo+6PS1*FqKz
zfD^Yb;}eUtM`SU0Cm{S9^P60ysCy|(FEN>8y)2qOR5d+5=jEMF^|sdxd2MX&Z(Bfr
zV*-cwtxYr}f!M7CuLyuC#n?(~BNbk^c`ibL2wd~<+n#6Pm^~)t#+MZHAYXOgXF*6w
z0C-MQ)Yvj6O<odk$f_?>NxjUMzz>J<V+%p>9(M%7II$E1NHaJj1LE9`C_56k8Sxs8
zQw9>Z)*1=Y)cJd1{VIXgc4?Z>j-!VH?@%u=Asx~;yYrH(ffgj9&uLYLZ|1$ys9P6X
zV8$tL1k%IAHURotQ8P}VJIX4%gk@(~5+LE}wY2T7Z%F)=?AD>qgn(($+jGS7M;99p
z8@3TPj47}V5Bn3KvDc)TE@WCOm8*9xH5-uinwuE#@NNkK@tx3Hf<H@qlWO^@sl`r%
zOOTXIOWUGl8&Zu5@+WL%_z6{$zoq_M-F72W)5*+srqq&#7%yKM!`Lk@Ac|_C#mh1T
zn$9rOX_fq?^+Rse-9mLW45G&EM}b9E0Z@oNqxrW$_7kn6n^v8{p=Icf2h1!ME!<YS
z$eOa7@?HX5(Z8>t<(MOAwxgA|^^c#A6diJ8fl_wSortaa%F{BOZ!9d(9&Nvh{{CE-
zd1U-$n@^B_my-Z_07)FP?Bv(NL;c~iimLA(fKY5791)PVW1}j6m$G7W3U1oup8sG$
zcmR{&7T`7D5&ZpQ4sY*t;#gz-W`J46N2N04mZg!Lp;?7g&(A{6scKWn>I5adNyBFD
zU~orqzXLNr*1aPCTocH&kiZpnQR_?}KXTTCRn_jg6+Hbxo*UKQVPR1K8p_!}D>1Qg
zOP#L1IV@8#dJ8*SrtJbKB|oZH**zI~+*S50wk7(K6G4T)4u%ic^tf=q01n(!X!yt>
zpo}Pdm3QmuQ5hl&#W#47X8~TfB|TRAv)TLTx^;IRy{7n^vDU5fjTagsNX0(6b28~$
z?F;D+@YO!pOaQFqNkUJ`a+}`MP5K{OWnizu{Bi4Q6R%%GaV5Ua9_-Ber4gM2&dt}z
zMTNO7WT9qR8~VcC(_^pzoUi-y4<>!B7<Kcv4=k$_PbE#|;@CPCPiuo<T><Vpr&n#$
zIvp(v0@h{cq_;WI*i1~}XA8)DX}*AbY=ttjI{N&j;gs>mU8C<HF-`ZbcpLMkixy){
zK-Q|)<%vV&qC9qKFTx!loP{Q~(#ea7*}%JRj8W>SaKtt=n{Q3T+<T`a#GOOh-CfE#
zS#4Sh?egY4OWT&ks`k}nov!-jLk)^{XIc<QApJEA#})*91{As#*w0S^iAnsMNO`{%
zf;R8PqApUk@hGTZz-pZx)sJ~MH3S+;(OA5jrUNxi!Gq?nQ4=J3gw%sPOppe2e>u%I
zTLD-((+|Fa?niXkbhzD%S=a1T@T!5^+ePO+W~DcNo|-KzDCj7V2}HBgahJ>_gT6vS
zQvyAx#$x*9!M;jiC%T%`X|l2zDAR5=4q8i^%&2~|fw6Gt+Cg8j%GFd}>J6atV@6up
zF`5>jfxl7IC31qY%m_DR_V|Y_sYWF2P3MH1)LeYsmYTxksTW#6(up(=daZBnPbBt2
zD}<#BqTCWOrGtC0#JNoEZ;7H?6CB(NT9q9*`QcRCQ{G_?7*c=#E|V2!B$`vQDL7az
zHL`A71j2UG>uIi5bj(YBjbC4K59`+<(>jlD-^a)Sx(q+_A$;Xi4YLz$Rk6IT-B3<U
ztBqFaJqM<ksjgfgWv+l0O%3Y-!l#$&!7sMUFzyN6FV%%$C;GJsae%@sqep-;rfD$)
z`T+P_xzVg7C1-wy2>!_&H@dnF3xY2ABa-rJ+g1VcVvybGiuA3{eQOZ0Vb{Y>8#1LO
z(`$)D@2PC1Rm*p9`sZSOQmqu&)|Q7fKP#t`J?N^W<bbL;p~am!jtQDB2up49sCF3Z
z0}=t7^V&T8K;l4Pd9*!Z8&L7FaD&rap4h>=$gaX97`=@<FgoMcEmqOprCeyBh}PU=
zC`KK6o~Yan(mfY^8?nRYuYV~K#GzHaD_36c_9Oh$z*CN4+S=@Bht&sW?#Gje5>R>J
zYdy_hDaqA3((vo?jaz%t2_>n}ols20y9ZmaN+1nRuNeK6Q^8k|&eaW+VAi$Vf5dN_
zpUv|aiag^uR8O}sERwI7Dm<=u6R-<W7o)6DkQ>RtpM5y4P=tWblc}H|fkPy1KS+o;
z0#JJtO!0sTin};2Uhrz0hl!va@=~Xbc#JMu4jFiBt=vfDxnGpBm)N{8KH$S6%_lu>
zSyZ>G02@>22UclvmTAyH#MifurkLGZ{pY6)CaxF%YG`gIe@yk&Rj%I~(%1^I_M&~E
z5<2$bftfkus*{X4*_sOwEg0)xR1IOJk+fqsig7aO`~-xSj3zzYMG+$K)Q%C`wGb`?
z7-v}E|1z2}40}<w{qZ3%@4-IT9+Z4_e_0?`{S!hG$bvqeXL5o`(XQ>R!rip#QBe2P
zzBs`852%z(cqS9YeogmyVX8Ch(NPf(jCAtqs2G3-OE)=>f!klL;+yZ*HDGgl<nSQM
zOjE9L>`_Y7DN%+~)vfb`>L!G}iq8gvnNPE)sSRit2ePAkB{}0kOMnFq&v?~B7rmr!
zEsx1;d}}ock}<=|?AMm}$u`7{0YZSkhIjqHeJcN^X^b6EiUkYdxWqo)S(i>hm{xxV
z48J&sxxX6h-0NvN-!GLqoOYi9b8K*R`wH8iO2CGbUBO0P`-^Tl0a*2v1A<u&;IL7R
znO=FCIhHt_dzVVvt!1Z2tlvZMBgi09_)$>*=aUu4I+U^2v|1z!u<ECRww7BfR(Sje
z)Q|alqq4w;xyVN@qAB(QsT;LFg+6CG!14}vt-Kgh@B<?@Kmp}!hzLLR!^-^)w9gZ1
zuS}>taSPo#UpN2V(r`HYj*tWa$2k7meEn0v6)7Tv8)oXfe~O4PwV8+lEO3SDHbln~
zk%M;Ih-GHUBbuz3)jdDhSpQv$8TeFcA28d47%}*tfYEq0z`pd(|L9{%DBCC0Hwy#^
z#7onOu*8?hPy)ho2!wwjz*o1(?TK3-vQ>&PBD+8&HssO`V}*q>1eKBzwmz|7DkUX+
zw<jVwdRdyhDqA0#!OpLpFx!C!)F$VDs7MIm5(e`MO+55L$4D-eRP?n72knv{%A%Wz
zS)wmfN?jlZvP;npDaPeuYeq&i{*fRd#bIF$C>kL3Fx47E#3XAL7M!fhc?iwuSBs4R
zo?b*a4T6km!0CU@ko%)$7;;g>y82mP9KKNgjPzuyPFP~n?oI%ZB6WHL*E-IFneCvl
zG@;;af0VJMrUqJ+7{yC3LnU|L&QF^$#YBJD$8>ZvfX80WWOOrKRcrVqW6ZFJJ;xzu
z>gYfCa@GKH)0`(|y}k%^jv|4CkqU5|;U(PmEwemF-CD4~W%D)D{yw)np0sKjs^M6U
z<T408+b=tqbh!5#8CW%LNv<Dh1^^))cjzdyiM0}*U9gw26^G+;Tj7nh+E#MWI&iTV
zIQOA#*Cu1(uKh}5ZOke#`T>~rHX96BDJuQjN|7xN7^<cAKhFF60Nl8+FX9-*$x_Vx
zKGUWF%48vJIlQ&Z@k$>25wrhB*%6@u*~Uq{W?=%<F=tq1K*%IZ4`4BL5UhfLmj38s
zdvi#XN^QJb8WkO5mf{#fbC~zp%~H4Yh=Rq-5656r!>fp%)5O)`0iT(Uz<)jKf1;)S
zuC<!h6T8uok-68z%s+ryu{$?74zhHUTNDz(8Ae-3We6}{nIEiGGQ&cK-#jvC01oyV
z;H?hUdODmQk2Q4w3EWZ}<stuKdmC;($Y}29K5$GciOU5fP7QVVIGfK_eG<{e4TJ6N
z>YdR+fQOp=f00at%K#D{(Eljm2?4^?#Iety3WLRzQ+2z<lq0&}a;}}F!Y5O}eg*Eo
z0nl4kK<f=%!=gl1{Faux_6#Uz+e-3Kn{Q>VeR@xJ9Z;Wzu_Zh^N_3M^#Y!Qa)bGm6
zx~Pbp(5^c5^#30g=ihZZ1FPq%KlW)M<$g{8UMv%I)4nCa`g&T9BIHk-NqfImO|CI1
z!tAA<2(nz@&hkj)zyJHFQ+@}iIHc@ZK4ho@;(WC6U=>IC0(p$?r|R9~|HHTZyI!^m
zFuL@=(2!Kuft<jDiB1Ukb_P+x6fbzp#9ixjMNJ;+ZY^pSGiCS+f1v@Bs4;y(F#-gQ
z)EFBgyH=_O;2|0R<V*wL0y3?j9Q9lTfe{?~q~1NfCx5;hr)>v)YN!{qvM2Bf3HCU3
zfFn8r(Z~c_|AUQ*RR3-TtRE@C!2G`j&%Z0R1Mbz<V-x1**c9hVLo%+uLZspY+a;~0
zqlI@_B#W=^_sbq;((13iY2Kqt*Hl9XWC~Is*ExN9{}0RY@A@8pP$%mJAi()(u|8MS
z^QxD>$Ot6C+S*#CScR(@aQ2Kl*KJPAqyq`;J-Oef?)o=tvQ*u~0arLlR_a29`RQ~(
zRBzYQu*kYe!UZ<I`d@a%-+zptJmlwxQm^`fxcLTX@7y4m1unpd5-9r+VsTOnBmn>t
z&<<o9Zvp0UHs)!!qGE2lJut|1^$mavPkxUg&^Lre5mVd%G~Vk97I5<aC*R={x;#Ho
z7v|WoKk9=;a?gZ3c9AANn5W|gD&XgdD97^nohj`oO|pIhsRvxR(j0Cqu2d*RP&&yd
zcE83d+|jhFy7q?_!i3c?;8F5C$8*%sd1RbixSpC&{O;@71hg`)U(=^MoXh0TR&}y?
zWY>!FOf+oQN57Wsb};=vq`h@ilx@HEZ-Rmnk^+K&64C<FpdccWic%8NAUQ*~NVkI0
z3?QY%&^>fB^ngS6Ff<H94f$T<z4!0kzjr@-J<oIR|5)I1DRW-u^^N2B943Hvq6zr8
zf`CoJ)Xl-Tpg*~OcmMXQon1(0j3X*XH8l*FGTy@LXD6!cqXCf|QcN&%QX!Hq5SUKf
z0$#2~^g0I;v$!b1O7NQxrEu;leRtG%S`oXH<o9LS^$EjKP3wXg<EMYu`w5I?hK)F*
zXjPM++-?mf?ZhM47(=*2oE(C<@24Cr0nD)Wj9|tvCg4JX((}~y*a1>lJKb-9h~_#l
zymQ8Z8sjvdy^^>Bgb4ljf5pFAF9M!&CtW&w4?uBBp+>399RlzpS`?oN0>@wE16eNr
z>P)?a2e3JLk(aHGI#<p=sd(^{V#OIl_n%p-e}n1$ES4hanmsqMe(Bt*xQab22P;Vm
zC<0VRc24?g9MQFem#Z+B)8Ra~*RJ&t76FCt-!3Xm1KIAy=ZGW)@>|gr<;+zs`RL`4
zq5PQ>p%0%Hy)Ftz+Wl{yr-DJ#=0%m<tGV{#sVQ+08vIhk{EOASPTWnkYD%)l$L^_r
zuCUwiJOI;~@gJDZf3+S4G)8sJfDSy>5|MM8(QD@!450OGNhrI7XZs_5S#^JIlFiGo
zTLeFb{08I`s;q9_|I7Pwyv$;8r<K%|wF|#H5k1>Oq`ECQvHwA2ex`ca9|;75=xA%d
zEF%ihReFAW+yx*^fHvjt(X1^?WwGi0BrSMRkqG-Tsqc|o0iYz*wn%j@ZYH4r(0F#x
zim$aM?dtlZxW)Qz5_1w?lM|OxS=9abY{(k1zPJzYg8zPf&o1Py!5Js+ycW=SakecA
zU{+F#rvSuU>f9+X35^EC%7au9Ktn{YaL<_%UmTtS0`%u~E}J_*FZZG8)8zzyqtBj~
z_+_N8S@LYQe{;EJkoLd%b>LNUz1sr8q)c1O9j4EK66D+un1EbuJVo(=nl2N50Dw^~
zc!9gF*!WrvP~w@y>$`v^FZOa}GQ3X?+yEr&ABbny7v=;pM{~Tf;hrB7NPqd?ye*XJ
zh?!E8G`{sFi;%Luh-lJKlM3X5R$u94aE#!(50uT&bzP<AUmbQ;c&owWdGhROhZLEv
zuAk!W^DkULIw{lNG-Lm#hD#(K?qqqJGf8_DBJEwfP6u4d+qGeChg;JiY@bM7TyNUN
zDSFIfy%6KiblMcEQt1R%&j8jK{ivZB{M!;}p#R(d|6gwTWd*HZLlL>Fj~7=r@IriG
z53{p}P+Q!E^SRF2ywQGR=>u3`n(<zks;O)xnkwL#pCHG_&;Bo-3Vn;!E++%tvP-4v
zrTpM1l`tirRx#(bx3?H2X`>}~LnspQ<AxFdepkm^X49<)lror1rn4?@IPMfXAi|h|
zdRp^?e5v~%+?oHhVpi;`*!7@5LacTbX7^#)vSm<cKd5VSH%%}nbfE*ceYYy+CBz?^
zj8_Ad0uo`Rc%4hQrwmda<rEPN`UgOt40f~yfQt}Sy>s9IyJQaSda8uM0Wl=<HseU8
z?R%%qiP65A@!|B-!5d%zm+;>{n}1qSKH#{4{V^Cn82wz7K)R)irg=rli37uY#epE?
zsaL}JDfx_#LV&p~71&6%@+2lQ;mBi!aOZ*d13frNWGBbHRnL||fW3sTL>P`Nrm8_J
zvyF}GyFH)rig=*sJGvSru_X#vE$C!$-D<@%9_wO$im%&$_X1mB8oVRLuJOzT`MuJ9
zw!!V<>`))hPW)C$4j)7K?^i`NlzR%g+;KEXrQDK%KuWME8SgH}j}~#Q<_1&r-Ei{l
zTMEzyYJ(LCW+L*RH)M69)h83w-V!h(4g>xgqp=HxCMD5H0+3Nr0BhvX#T@(rp(4^w
zKyULYQe=iU*}g#xKNCa7xZ`Dxz^tha&o92lnJZ@$uPa_pO+xL}ke+h2y!c8iaC-s3
z^?(1^#(zGi4HF~hL~e$wM?B{z&~wGSIcR(Nj-l-ar;yAkZR;!}Nf9%8*iiy8WGa|P
z-fE)t-UY6&n6-TqKy~L60?q9%eea)2R~rmb_OVw4bywrBJk@~V=J9)RSd;!Vnb}Yc
z>N`YS4PVL~ET3gil$X=Q8YCNV75%NQP^nq7(=5>Z1|Q9Rw8D<OcBxiD8Nn~s-+@27
zM;E6_Uzl_i8~x0;)&@d!(iLN;)%mUnIz4toxhv070Kv9eq(Co|BIWsn1Rz$57;gaj
z0&kJ(KgGR<s@Qd4+W!ooNDKI!x*aZL__GsVF_c~v{qc~Rww|q?mm-Ph>c4v*@?Pir
zZY)Usofh-H_vLeQ+Vti`LcRhD0Z!^AdVg?6^kI8i@NEh|tynpZo;NPfrYyhy?0@PO
z>8-=rD#7lpMhj3M@yb2N8JPPKFdIVKowFs|ED#)xn<v{kw2>>!0hflt$-ySxq*(Jx
z6g`}^vgDdTOwFymdu8}ajGmp<;QFxjV#VKi*E(|UBH_oCe8vbzeE6QYV?W)OaGKU|
z>$c}MJqItZt&9mU866eR@8q?oNs)igP_?J+axr`qYYYL7ViYh?;Ecojg!J6z1Gnf^
z!upubcY=>*4ja(RZCN`@H_9SpcDq>dT+n%7<>JhjzKY1Lqq##&NX*T2<JNm)BSn1V
zY24+1w?K*itnn3S&pG16E$*MMh9}%IwgA24<&e$h?Wb0o=lfaIK6W@uks)@gCaL`n
z`-0HHAQ>`~7wY_3Xcnn?3cDql^Ir~kV9Pci&{3csO{5Gj0SrNB)&X!&3?3T}6gGt?
zE{;G+-wND0A^53c<54@G1K#e3w}t=>&l9KZnd!bExsq0Q58a7y&cD}+>Jj-QcaxiY
zlZR#8=+h3!S$Scxzfk08jW*KUyro<OiQ**AozKS^{<sJ^ce8TfQ`SAAYRIF0e@ewB
zqTJl;@FcI5u8D%_Frf7@3_SdQFCTXe+!aN-T#1Yi^KT#4d+fJJ9Zvb<oosgEv9rLO
zUKAL>PjQsLx2^b|Ea!-)h-#$c#Ue)l{&H(r!>OX2Mgy<2{ySBx^8E<;4+K#8$02Pp
zSDC&twPNXaR;Sd4u(32dBh{UluFmMW`Dn)WRwn7FN^kq7(fY|bzi3Dk>z%`u&gg}&
znCQjC(1FFXqS@f>P|N+XNSFWegaZ*baDUZg=S=6sme}i`Sv~@!?3JNRLcr-l#{dim
zq1mz1){Vdty&77{pd8{)<P5yl@CC<mXMj2U(*HC5{-<x<o7Z*+qSa~8bIj=4A%fQt
zf%B4*A|(<&h((dHd@SQw#k%TP;mOumCPLb{up~L4&S`?U5MmctpJo$S&p1o0AnilB
z<%4#4+gX<4aC_X_<>A)`_CL5Kp>ArGima6pTfN*^!K`=h-WdUJJg6rR8ez({d@yOh
z4fJ$_SzUkt5kQLRf1mRPVrInu-ARXX`fE$xC{97FpzL#lTYLqb`Du+|6cCx)CsL00
zUDj*aT~=Wd2OH(NH~OSX-qVz9>CId(giKJKIN3!mrb*pZ^V$&wgCebmmP8<2;d@f}
zE#ZE8w@d2im%w-SPvj9(HIAvD7U0X@wG2T11k-?K2pYceD;KiXFp96qBLLww;-*T_
z#+#ECV!m(LO)E7Iu*1$$+gT5*a#g-Ym@szJSc2aJ8?LE;>xChBvqsaPNSlM&VUCW|
zQ2ZSsVX0AmQZnN8Z*OuL3bv)d<<IiO&z0RvRfdO5eMNG!ve1{r6s56o19Bs=2|oqt
z-X$e4=tMPTzN9;IT&4JZ)R_%KT}(Zdh-9m1HmiKBMK_?t5}yDaj&+JYy$FTr`M{Hw
zQ`J&u)7ra8S%hv_Tqe98w&amj(Ej@Bi#|HP3)nb;!hz_EYp9vAY=qqyc2%m~rIS(L
zrI~TsrQI&o)p?!JMdH#0gWL6EwYBI^tjOcbpB+%%QtqTq(|>B74RJUjv*9C#reHo4
zkZDmpWXhRBaEta6UbI1@@AQ=(<bwAN8^>f0mr}*D4%?Lvz1Wgap=KoFB^u0-3y#Cj
zL!ShB@grg;6RJ8K6XT}X@8)JC*=6$Z*#}uYK#(c_)o=Vc^8Y*vu7+~ULucR)R=yX-
zQ5ul@<;vp&)4CA&i*Wg;?m&p0Ok&70|Jf#3o;wSaZceebOJ%dUHsuILtP`*F`>>jO
zA$(P@p&%`knf>HdA~o>?h^%dBu_>dXb%bjRGI8W?I}5ql>j&<yUSg7DqC$j97*zTt
zoRD4CeaZBknx;Rpdc$8YFCHv9FY(Ciu9wCQBwM+A{lbRgVwiZ+L<Tha6`&6zJ_;1N
z-0|?nJyvx;a2-Q;dl%=2z^BtDh&`n%=yjbZZWfAfN9aq0?`%xbDM&kt)_UzdXwc(V
zJc@qJj%>pk>TuT+Dy4D$DI5Tu6;l9Z9D_g{)(o#g7TqjII1D}`Gy)pAW<W!O*5%~n
z5ZmQ$?9LjTvdT@Ja!tZdqNg}2yuSDV%z&efvH^8{ef>9n^KuEZ0c>2qyyh_s6=ZCY
zt5&tp69?a<4r>O)+7sap0bhVxoC~bL_WrtC@kwj{Uvk|4IMC<4UNkyd)y5$tRU+Dj
zL~$%sOIZrBKd2|8;1nWk@LS7!M6MCJU#)&X0cvZuPI>JFO1qW$(iHwn0ullvxJhq%
zaeuDlp-P-0<k1~Dy2aC_CkFmgB02LOiw7L+Xiau!O}@rdG`xmeis=0y*qE~C&GwQo
zzgjd-(EtpE(_%FpJHJ>8EGN=*QMSE_Mm>EBhSdgnc<Jm@9V>786C0jUkgU6q9juhb
zhV)6v+f>TtvmZXsht`?ILko+y1%ns%;?+9q5UAA6QRXkFVodDsziUw|ZaM#T_<@0w
z{QPFsY*g8RG(En5rGM=Lw2-bm@eEVJfXNYX0sV2A0|F@$U@OE401SeI89i?B)$he{
z@dL>NO+a486x_;;=oS;B7eCf-k@0y<Fq9qZOnO&bi$xzSmdIw?z1i)wiV|{?0aBm-
z!LR;*o*FC7*VdLo9=;G*72JF75xRn<Z-4vT#2*q;slN7(qBFWR{qk9fc7XX52fv)*
znpbHrEM@0Truz3t_t<^yqa1mmc-5$fKl<VqdQAfDe8Z+XJaP;85RTv54V0aJQN-9O
zFdFW5SMe`j7r;7oNIul(_z~^?u}YJunxB$fA@(tOXmJ!LI9{)4gB8cKYB%Yo2hr#s
zFnWC^+gkt$;#*d0svc)D9*J-jFzq3O9k*M%Z6P2lLQrBxMl9_XgEQX*X(Ras?XlYF
zG0xDW(XOz>vr4~MscO2U6X9vH&3@2sptjArUbq9FVPEKX-ENH!Z}^7&Vxqx8C|kQz
zQu;F9`mf>||BU$HHy+m6_|7)9YYOP?OMpRxG0<(fUOmTK2BE|}OmURoDYd;m0g3^>
zc&q-`tDhn@0quwicKEbRIr@7=^j7D-Q!Q2ERx3&<X7xsn@SpjpwhIuTNb(;6ivROt
z5V#`1KB%XZ62fH|5XmlE_N<Mj-FzvWEYtmsul0>N^crTBooO-h@|il%`NlQ6z{D&N
z%j?~n&wFvkr~QMRBJc8BF77z8Jm$ND<p}_Ljn77VlLrWi<xu<rf8ILXMR%2MM;J$R
zn6VSl2bLxvE%kC3v+<YCtW;d*vX!EgM%$x6;q~we1D-taXabU-sqCYLQOCmF1YIer
zSQ^z13u5WkPHczKNWqYs$=ap8(+wR-u><C*lNsCU`OSlB<H1IUt##ifOucH$SE=YW
zzX66%(-1q9O&athtZRz&#K}HV3St$gU%MToskR;XTWuq_9x}yL=&=z_9DGWZY2g(S
zxIYTk2N!}3fb7+<y8rw6ln<K1AEH|&*9*#~KFI*?+S25HQ+x!3`hCqN>yIDfU{2-g
zKPlMaar``$pR-Q*nOxFTDx*F^yetJroTZ>SLuM+u>xw2`zn^Rsb?ghpSl!I|YlQx%
z^Z)1gL_H#dygGAoAxa}kH;A^sw)e<=%|L<@J-u+fZxkiV^RWZc&n`rcP5h1=SFU~C
zP_w3tC<#%u$hgsSdU-cYGv0YRRN>b}=nXSn<S=LDp7cvGd!fqbwU@z53@M?seTtHG
zv$ek!FgKC<B33cM*GicJ%?;d2zYk9vny)W-aqs5yXWP+9r(BTFq+qBG5=ZFOBD@(~
z8+~aeNbJLv2=s$y52Yv?-BF?pn1$D(S?GDuBb;0tNjvjEsIvqmp_3SGVeMA1RP9Ef
zrrKs`ecBj>r*sVs$v4~11}{lf-!mtMlLgFJ<GRsUHk{FNf60zT2xq(OBf6YzxX`$L
zMsu684xB0mUJ1RF^JTdP`XFvMaUlG!aR9zw!uv0RnS5xaJP+EwFw3PhrHDD^aM60r
z@630VyLUdC3iI1)UEFc=dMY>~8q^I)T0Qg;LvH6u@i`f*fUqbQ1H8uX{D(y0Uw(~K
zKDH5CniB>kP-LLR{9;9j)JTRApVZs#1se1hwX>)BMELm=e3<!heJ4<Z<aDvS(xW@m
z&wrO+wL1G|w^HG2p<$;FX{Sx%HaE?}e~#F<@1iq1G++KA6_0g_Wc$^g<<m1{I4xz8
z8e=6z9=*@<jVAn4N_1ic0@HrVx6_PVkU`8gH+B2(4CmGly4%o5?##<od+%+o?AQ_N
zogIuGb$)?6V4A{rf~C=m8#9fazlu@<W?_3=ryID|q2omF5>KS&SOo$ymJe?b&FY*#
zLWOrOWNE3j;?DXzv;EIM`@$TrPefSyODIeSvCFv8rUEqmvb6oSe8YwyAB~grb$0#f
z8|5c1Qa3!MDhLm@U=l7HmB|c;_Fx0=0P3*S8wMPo#edSMqIBuE5k2?yp+F?@UyeFH
zvdXvM5glpDeRDwwseFa0*Kiz-7I!r%@xnJwjUAw46l<Pr^!3%k`2ue%44^*|#lW%3
zkZ=g#^C<G06MwIOyZ4CcpumlvPg6Dq{R_A1K|O3$-T2ou2Ok#n&x0nH->7Gy2+O{L
zslvu~vS->ZgS>tXxG}Gtd5j!~mxyZluN~8I7bN<V0#@BLzapcx)4SRu>Va!|RLcb6
z{OoxxKYnTku&w`!<N-soONUY%Oz(Zok<}Ou;W>Mv5s<?z@w!_y2pna%bD!M!_A?S^
zgL;WAq_d%kN1u~djpbvuUC-_slb!R8uWlB<lWB!AWjuUwcObZOE(k4_!;Z<FegX#E
z9kfn?H@amOvf|Y`2GS&m2%RLz3hOrfy=yljuhwq)``2v-vwARmHc68r=xp4fPZ?Qv
zOF3Ta7Coz&T}(j^v@)IO1vK7<F(4b<rR?tE45wt4FFT!-3mTlPJ-n47OqPj`Xvqqs
zC@FT~r4Fw6iEff(ne(1MSWik^Igyt+%|YKBtAhKC<xUz{?oZ9QZDa-u*v>y7Pog|o
z>XTD-bCNB)B`@%wA1yE*?I66*6y{MTY=E><mV-~cQ0kx|-Z})r<>r7-(P=gizu7$+
ziuMUm5t*i}+pJ8czj+AEP8~^YrC#j1)%Ic$i-K7orpmG16OH9R3VLg^V^i^IKH^oO
zDdb=CU3_%tKhJ-2<a8^9%EuJODQ2u*cHZ4ecq4o-Hcz@U0GDRGflOOK6tRC8{CRoh
z+`}~jaxTEui=(74S8M7Vr4SfOjw+IO=(znxrf=vFi^%2{-dwBwhW2uOFW^$|e5DGc
zvoJ=bNqA&rz4{}aK+xLUWEWPna~^TH(-Y741J16t2pMvnugyTt7oG97uNI|*8cHA1
zGfbh>H<0c!>@0`tm81uou;4K}maT)PnQ2|4XsG80v}&CZn%&h=UypPROy)C<rA;|l
zQ#`^p$)ryYT7wQlBnHit+1W`tdrxm@&?vY6XeU;We;<)z$bKcpxcJ3$)54He0`e9y
zks^-w4>;%CXT%KKK>8#`{v&~b@!gJ%EoN7Xn&t=2@0_?uTw`wpUkOrS;{K83!!t!}
zK{>^67a95aJ%N^COGUze4!bq%i+Kf2GtzfFfkI?@gdJdFIkLjp5Y~)d36CK;@b=F*
z<y7&}&~DGI8bZ8LlOmWB0>ss%(o-GMG#*{Y53)p%xr_<}@{u5$_9YkT5l#Xo&kX@~
zgTqOC8Iim#DlNSdQ^O#yUpd%UK%Uy_U+dL4s((e^bW?p0$4s}75<(xxqiRKy$*Uc4
zXc7Hll6li%mEN~Kkfq`Ls3Xn6s8MFIfM%ezTdKcF$D6lB-kY~0yT~vi{6zmV*K9Sx
zw9M8cmap5J;sP##X5m%_{KI-Q+AeO9*{U210_f4u^uiYvH>w9{e1Z;_VbN7NjgiQP
z6@SJw$?#55kM$LaAs;shJ>d;QX!x4^RlYXd*+A>Ip^Qer#4pjUYzhbEf!Vp2<YB>k
zqO#Xeeq$%0^(EUHCv$^P!nqaT??gNwJj)muuU?9jSxl;ecJyW1HG7*dwjdU3=W0Uv
z7Am^E1uQ4XV8U42h;QrXmi?j;X%}~;4fh`AjP|y`jst(9%31wJ)Uv0R6ehn966+D$
z`RGGcxMl9gd=n8cJja?!C^SW0`7pEKz9Nu8$m_9he{_bWnJYPDYNsJ)BjQ`q-~I8@
zx#Kj8A+C88flB)8wL1Ny+l(l?JC@yKXRdi=<QD}#IRts!`Pzc7X6cR>uy^Av{7gE8
z!=memjsbJSLK{cje4Yarb-|J&uj)iwD8;mY;(}FM2s!lAIk2s32ci_W3_gj^nwNgc
z`!xupn!7aD@AhzE00rhm=sQ)!ZjtX9JbUrnYX@LD>wTgB!w&i9so4}{M51R3_vD!*
z$$f1o$ujQ#{;*Rn;;>Mb`{BZl@Iw_)L&w|I86j2cOsaAZ+jvtCTU=5|ce|&Djzgu2
zf#{35pU}oOl~!?<5Q|J{GE5h_-uiMaZ>9#dci_OXcUW`-qyK9`g%4xp8MOQrOdj5t
zI}nQbQ8PXEDeU}owb5IJQ*@!V!)g=U!6%K_VT~(yK3>W<RIZL`L6`l#^Y}ea=gF%l
zhhsD{z6TuS35_kFdetVYqSPqIlhk116T^T;l4{Beuf?=#mo8t2#hs|?#k0wswlM|F
zX)YjX4>xdgoJc-$=%@86v3%SBP4&;yyS*l-7EpOZmb~3>%Ez$Rkxf=$*|M6lW5bNY
zS>v}_AZuvQ3OIn_IkFH=VobSq`a1>{Uf90dn9TRhpM2D_B0==RL{f4?U}_l67?CQ_
zuM*z^6MJ||VG4Da2bI@4Kw~tiSHXS%xiJ!Ic+Syp-kad3z~oKYi0iX22K;Kqc(8|~
z6#&SXbvD9#o}c|b)6Q6`iUoWeDeE&m?oY-ab#HSh{kp$elNIS+a!2WzE09N)4%pJU
zD3gS?TB&u2mhda+m9Y|5yXtuY1|VJYk8kk+rF7q2N)dt8fq;C&@Y-x0O^(uo?1xpD
zQ(~$lz-sn%Hwn;*mord40;oxMk;7kM%X<}$I(VJ7`CkDX2JGd=Nt8+ONp^e=jIc#v
zz6{>K9xs3FfneZQCRRbIf1P5t3Np*Ui?zFH(qhZa`<T_l%#@0P;hYrDTECjmgg4E<
zEIX}y+^?uR|8XSZwdi6tvBL^h!6@3!)5IdISGCSK8n~|yfqOfc(ep#Rs)R{&u1g1G
zV8cJtP?Aw33Pwd-Z<SymOt^^liW0H7)v+FH@T{sN6*SOQx0A+<T=>yIvp6!-*!r4}
z(cjyXAw<7+E7VqEGh|v~JE&fKJ4#bc&?fNr^g;VM(8`?EGKF|cY)7AoGkub)WePHc
zIS8qG?!P6^^ljhYJ|%SZc9RhIJ$_(7H6*TgZ6g=kVPeoKNe&kSg4UM_Zru@&+hw`H
z)W3|i&UxCKY_qr?le4Oq6(Zwsa9!8H;qlF?kl&O7k8qB@<cZKm>vL8uokP>cOA8gM
z<;x0Vg)N>Y3dVyD7_>Nf+z%Vh$LkOERlz9Sn_dDN{Wz2FpH1(oCfNxB85x6Ffy=<c
zco3lTh`6#fhU)=|ig>+i9Zs0@u~ot81EvVCki3M<!KDU1*v{wVJK5i^w0b4xE)RSm
zHLom{#x2E;0Yf4E^#=;=`jEkq>Ob_Ko;TNI_6Fs)0u?jdfz&Qu#E--&5v%8#FuFsY
z=ay{I8Sp@s`bJ<P_mznv2~LQ6PaZs#gD(QdyB5@&&4Au_ISjD%?}=wnA!Gw{X5wEx
z+ntE7f}ZRB94vTCLRAX|-3q5l`rcubp$@ogflb%J<85kiPTW%IxL>Zat?VMIJ2MUY
zxU@aeBTPKR<7Ig^F)R7=*CTTc8GGZKf?w8MXO&(gl<aUx$}|wG`l8-$jAv81Y}HD*
zY?jsm6X0UBMq#l6T1wt+%WoPoLtH3+Z$+0@?9IBoy<P~hVf{6W!(&!eFBGLn)e?x0
zDcZ2{h@JsAH6*%}*a~`D<v*Bob-~Fhvc4`X^rvJFn$Pqi7GyVD=CC@~xG*t)0Ydml
z?qXP>b*sP`>zqW)wERrW#0(*sj9WIMcg8=59{o7x^3svGSGI0${Cb*U<Tqac8oiW+
z@_9q{Q7LQiD^0Wzjt>!IMTe-lzBN#q!N$NdF3U&{hu?`Ed$H)VP=XVpO!5)#@Y=gP
z<K=m6?c<t!bm{p9oF~@vzSTR+WkPx5t#($OAmF)F1_BQN@o6^=(e)l+1&(eo!jClY
zahoZij=QQOo50*v)#)5?;yI8iNmJr>MhG7;K~AG~2jxir2xtRnxKgqyY5S7D`vc-M
z8dAxfPbUUgyebUXgb^Z*kAanL^;WIdhjp>R@-N_vyu3$;g=j!#NrOLZTl5-4#{Hzn
zle6c)oe6=&kG>vuD@kfZ*%cLZdU_Tn`Q5U7Mz>wNZnL=ugz!)i!oFx3!PO?@4fb75
z?Da<rSPeg57ZQ8KciruAbe5G*_t|?OcQ8jxIcHu9MGAy4alc;$QXo$)@&2yWiWnJv
zHE^3VxQlnYAYnJqn861@RAUFkfN5tAnv%w5$vDcOgmoZV-wc%m>t|5<f{Qri6hJ~l
zk>vG)J{S&oyZ8WK-M)nnOU%LHD|E5CQe#>-;Exot@AH$ZM=M3}#4xCL#DkfU^w(0L
z0VscGsv3i+jUIaf`IkWrBUNsZulv1~y1Zp)<)W@mzVveTrefn|2yWazlOS2~B6C?S
zC3aaY=hBs`Vb_(Z5ph`udbBNA{+QiZJyx29@vyad8k_lRYxI}!!D=}Z$(qY0pMwpf
zBNTCG$cp4;3)q3D-XPjbZ}_698azm4iCx~W#-t|i%zmv}{7TCB3|XR&X=$SwJcKvU
zKJliVxj4S+C2r4>qt(~cQXAOi4u&!gXsnTgFC@ig3B-^D<PLs%3J;KQHyN6)vY>`$
zK+dF;&^)_V`LrKt{j-B`VW4%!hsHTP(0YcQyrW@?HA-k*uuxB<;l$=#%CI0C?&y(t
zTd-=zBmZXKy9m7ay3ctq{h25*m`O>%uSGW6C~|4%<-Qhv2jEl?>v2<5aNrB3xT&<j
z;J{^I+1rla0|ziPhI3aoSu&fPZ}@l~PP>`}-*a)8HQP<$_{=HJx#H%PGfH^x(CB6H
zPN@W6U1#Y&7_c;}R==?<{gURk5Nby+pRv`c9=|Uc1E+-A2rq@lvv>eIrhbkf;8K$n
zX(Ee9_ouEm5StxeC#J6cEdY&WjYQW_ngB6CXz@TYxxVQ2#m;Eg!yrDPX^5F5lkV?I
zAjOj5v8tG<*|)|SQd<6|iVo~dHda0VElKxxaM!vTxNDR9WP?foNVYv(j0Gjw&bn+a
zeV6e7qN93QqI(1Zz#*bK<-3}a@SiAB>D2+RcvDapYDq97#zy{h=o@gzDhmWv;S4y0
zGx#OEr_<)^;#cSEJ}z13$}``5_YF}`YPK&X-0XC6&`IU(72tdCu>AIZk|#8M&))T2
z;iMscc;+?omZ!<MyU4|e>Lbm_2TY6ZcrJA~HSdeQ7Wtn#LZpr#8#)S-Y7#W12-ALh
z1?Cf!w-1Ojg-(lahNPyA-Sm_yAw2N)u-_PQrr8*D_S_hDCM;i@Av$rLCJlCxB9mEn
zxlgj@b3e*`izDS^*y_k}8|z?p0e5Ikw%wVlb6p%Kt^3f#USzVD<%${KQxkCPcjDF-
zWg<qyD|*D<X^sAP(I2W=VOf&${&rqlUcM#Fa4BFvM7KBIQ`GRF5E{3hA9whn!hz*v
zAk8qc2D@~(;i+5RS5HnoGfZu*ZS#-71VP^?LE$o_K3H2Z83VQRBx;Ac6q$hvU1(9|
ztF2HyM@{Kjb8dt#6tR8_C#Mulau2xfjPchUA>sSua`*;7ty$A`Ioe4dJH|`d06~hI
zB2tIwF%Q{Pe@hDu`Fm2%m#!05?0ajY`72jiVMqCdzL&F(P;!Z#p656;2-rl;r8xFC
z;~h7EFc$fa@$JU2Q{o5tNS+c&rYPNf){!>>81agAZp6BC77vd&AYxc=z-FbtoivG-
z(yImp4r0&nsw|Ohc3*ikJn=<ypR=YLEx4X?@f^m^kQIsp4?gTTTt^PHgHGUljT`-f
z?hJMsaNQzGASJDSz7*VLhWf$`=V|2zQX#vmU-28*$H?_7D3Rb;so+dvs44^lWyE>}
z@@&}4to_~RY<!wrW?HPmv0x2b^it|+jAKxQiFneM7Uy5~@8ZMTa28}YEBpaYgG88j
zb^W`sVnYl*Z*r=bD(Xc;7iZ0nfmpemMR9n}aqYNBq+Sdwk&3rZ3}4XH3%kUqb-MWe
zMPDOHQej9@QpZEYZ8~nW)RJ%s(yZRsgsy>pM)pD^D6CuJv(?Nu@m}=?``{0Q^xH3C
zdKaIbxp{s%TA6$1R$y<TcX1q9ZP{cdx)5kmY*D2-R){4TtC>_a-jJ;maoH{<1a`I(
zB$Oa`gLuK60SjJ*6N|~k1GIh1biG}tfBjZxET2@UDxXXww8-gMG}1HOA`Ojs(?~N^
z=gS9^o)g~lTy(z0FrwD~9+Czov2PH#CP&!4A96d!(R=YW=emE}QpzlCQk9R~apu6>
z<rN8a$Pg4p+E$zaOFI)#-<oWC2CkQzX{MK(iQ6mXmVPYSjJ_Bq2;HZDOk01_bAlU1
z1{^Z4@S!C>9flklf0?4MS#G|HswL>De@OF7vZ1eBFAE+-2Pav_;&+;VZbx<@&YZD3
zAmCWlY0CLrRWuUuKHhs+=*0kX$&7pgp0>d4s#4$BQX}uYB8LHt?=^hmwiRQ*p>&h9
z{b&)+>U(>s=)R5o#_vM<C_u80Z0U(lbJ0M7-qOBt*!2v-18)l7SGhRam+qIUBtJO<
zUb#W6(UQZVvP8cRHYe*^`b{jQJxGA!>QufIK#EnLAFrnh$V>oE)XmykfOf0DKhzDf
zQ7UheA3*n>Au$+gon&9O)NspLze*6&@!Fj;X3PV=zjWShl7r-Gb}O_xuj#k1uGSYb
zy&iV>1%WdKB^ao6R5&eG<cMW`l?&=&HsqjOvr!oaCn=}(^4-(A8HX?WB4^XEHP3w+
z0m0s)@G4f%{CMwda1);mybpL+yEBdjJu^6h*RV+q!(z%s)(mn73~zjwOLw!Ri*lyc
zYL~6zM8GXGs;5K#JZ(#1n4aVutB(w8H9s+@G3e^vO0^|jX`HqVxAa{;(Bi{rHMRwW
znsmIkG(~a<h9Yaq>&ay%r6f?P_r{SLrXxWFGc`G|a4vl~XE>lAC|H{sYUm_IK<Elg
z$R#$Tf9va-G+G>V7)-mg+OhDvZkCfCxaU!ktoq#Oci)$Ui5}m*b#BQY1i<=VFDfX_
z;12uU#GY(YOau(<F8|hXk@;@x71bj=%+^(RuE9B(7UhhVv4efv$1sJ0^r3-zA(8Rx
zbX*5t;f`0>dZg1-j$T59P59#)1TruQcfb+|@CCP>Rux9FBZ8i*<?#02Hj>b!EzSL2
zd5H<?7}&?P-AoI)DPrP&5hk;3>fRN$=xxN;jYk^<EI)bys0(c(0TG*Ts54R@Suu#Y
z{8L|p^O$4f$=HXhBK%~&R>+nnAkVDdUJ!8p00>w>=TpCOCDWZP&;Xpu69a@0=A5Wv
z0a=j4tRILTF;w0A86Kxjq-+a%J82fLUyNH$u5&bn^8~IS)MPdTel2HqE|x+lkXANU
z`ok)w7-Ls2pUECEF`Nn)myq}bZ{y8{8V4ruOV(UH^b=khW98&t{}+CJqIO7U60f#b
zRKEA;K(QzJ50TEE4{2*&%@TrNFbYK-#-JdL6WCxvRD9fW!5cB<-FDQ2U}d?q^p|K6
z{f$=^4Ce25C(7K?FLrele9Z@LL%BKuhDjvi|A#b*Q_-|r8QAEsW#F~e!ShU*XRRnV
zHjZCJedDu3?7u=4tF!X$SrGLti6A4i`oD>wN`H-4rN%2@wYNj(P!DVXTqjh)a81?N
zFM+8`n2>Q9(6Im$Xj%yi27}`~8oRTdCwhYyq@w8vzp-SE5Xm~4D4*pUDZjBoDTsBw
z6wL8m(Sno9V`Um#8wYzcmS-c<qWdTf5OSriINb&6z~RheI5Y2e<kYV)($qmp(vsET
zQIIfzc}b_zj~EzQN|7ab^)@=_B{vy1+r~3;8z25<W%b&BiP>GG8prA1vHkd=w|8L{
z>VwX%LGLIWv?SR^N9X+@ylB)`GP-txF2ddZ!Hm1(7g$>UYd*|=%X_Ky0wB?Yw@)<2
z8nBq@;JlI_!ok0v96(IBpZ~#il&*6dw%KrIU6NQy_um<}NUI^tm~nEsYX(GVrjELW
zJr;)M2C^-NU&am+7B44C5?p^EzYG97KTp~>PS<bj$>*L7(+=1RjlUfT3v&6sAO@If
ze8Gi@lm!u)U8dQh?)4wL9p6~Ym+Xn^EfXgo>jCs-b=d9b!pS|l9^0N^#$dFtSq~-B
zHcp0)gUK8&g8HJz8O(indOhv%+m!jHMde7>S0Li9IWBs>HJA_MUvSo|X|tqpmN>Ay
z<fs6EJ$=mGS!I?Yu>55w)MsLSM!+|e{^Ah*?8*4vyb!T}mq=wFjRJL*j@!U39#Ajy
z+<6_(v=fX0tCZZR)DF<n_7AKPqm@C)NeUm6pO$R9i>~2Vi)RH-OroJDACT;>odbrF
zw`gulF_A_$lG=NpQO_hx5!6eshf^4cQq_eHPW@W98>{K`sWxwOrj=6a$qrF6yq3my
zWhk1O^%iw_H5G>CisP?t_i9J-ev^4E7Hr2rVOy7&LcJ}+pye;&Eu@>i5YaM07y$Zn
zC3SDiYIuli%V1~gi~<`o488kp=E-K5Kj{_;o>`RkB4wSSd-S7Bw}-hXZQ!}u7ENI3
zNFArp!@@?Qp8L23(&MUW7XT6n6zV{Jsb&pc?;84?h>a@AT|pP%wh<D&B2p2S79H<V
zc-Et-=8|eAqOtWJfUxj@Pf^j0S*4}zgo3`;O=%uS*}&rz`!l<@?;TyTkgPj)0hDK(
zaLFUtrQptuJIW@+RE?9S*2MDjMLzpnt|}1-8jeFHt>-s_^wb(JRrecOj)k?7gAVa*
z^Xw2%_nuCUJIxg-9}i#sWsv1kSjvvUuK1p0Y!2=rlIybK)`;8Lt?x2)Tsfz_p=QDB
ziqIIK)TwaF!es(a6`PnT5>)w#Tg~#(e*_Xk^M`zKS7~KP=MGUlg<w^s8N$dSN)E!m
zn1OSGsrpq%a;&o|0N6;Kgt8X;Lb?kOeYPCx0`ZfsA0q&009e6y`7p5nUOXcU`KF{>
zYxCs#O3lLouDiKiop48`hzv_Ha!Vm{9-n*5BWj&HiLHa1v8vTZrk<z!k>Q|^MX@{B
ztze_s+{2dFe5Y5?q%g9o2-q}0Y_;h~KC&6Q#%-Cv*HTWD#(Q)n{;=^*u$1Esm+b~Y
z67O>wH?Lle1T8@tF(REXDUm!LdDiyS5NRUVi`Xt0sE1Tc%mT({XOS4^NFB+EhUOr<
zJz199IQ!zAHk#!?aB{PUkxPmhJtt@-T}U*$)m7&EQI25c{{7)=O~lwMUPnXMZrwd+
z%zHkEOR?8zGh+y8GPP!tuS#)Mg@$URe!IfMZiTQdu%Fyx%V$)nT<3m&lap+`{@a2|
z-6ya_qpfN9GM#lniKz!g5;akd*~K~_NJmzfKPL4tY{L^1bY7!eqissO0Qk}l(A|mE
z0l5fMdu%{EmoV7@{Za4*Q>Q|5+)D0<dt=HNfmVt^4`4MecT>NPa%*=N#!ctg<*xc9
z#eo9s5KbSJTWM|-rOL)4lGT6iRM)1S3{@B8xKk1JaalzmV52}buG@`93zRUw%7jqk
zlE=eOr<5(s+f0oL3qL<4vEgw@aX?0j9(tc0mpMc@uF(`={^YMT;0v$^g|r+_?tP4*
zN-T_~$ML^H_%@7Mdg@6&R<V!>H@rz_-~gQTVx|N44{iVqKDj>(zWc5WC6I0Ev+_X6
zhwe(d$hV@s;(WaMoV%-8u;O;L08RfJB{k029o*@FfY-5v+k6>%=seZs=q%sj!-WM&
z8&v9`?b*P+FbzY=>%Lo#%ytYQR>;N1b+0EE#BWDO?|*?Mkjfj-UT1kiGe0yGz0j+r
z+L<AB&}n#lcHN)-_Oq4t9eGND+ktHFnWk%=9R#)sC#8I+P5TOq>aYfV)*i;DnAHo&
z&zMS4`q7517|$s6Ia1eH>3F$9RKF20uffR5s6CH9On7;fq>~VXOI4nyoFco1FR1;d
zI^{>j_Ph@@oaluJVrfw7>%t*}1vE2DNlUs6(O_FfG25vkZA&r5<iCcC59<0$&7zHM
zfwf^PAmoe&0CnQcKSWY92f=nKsS2^AOtv+PE|qkp@)&?&r@OkUa#X^fOFaLU?GzHz
z^4*^j<N3<P71af$jPS6j1VfyL6`}HRSLZFgCfxd`qBHy7Sc2zRI#yi{KfT&i;#Y0Y
z&B*|z#sMbL>!~mYLlvN?v<m!e727#_77G$0QmMR_Jb23HR;UQwJB|gx4DsRFNq>E(
z0skF*hU%X&<R4b8N+6pj2XJo;4`aT6en7Yi07ca^x*3U1c{%#N=O}f)BN|5LuYX*6
zj{kEjXl|>9(uG^_&ys9sV(T?8tv@ob4e-DzCGDllI3{2}FAKs}S1$ISIx)Bls(PYq
z7{>N3#>x=qW1GKzkCkD4L^lF6L6|xCpu3)ka$!=?vADa|h>~}xqYaS=<OZxATF2u%
zbwf{U)YT$FwlhOyu@^ni%*CRY``hc`B&s?lNZ<Ud_;;uR`;FA5-lNz(cM8Ca>E!M}
z`fH#@MvTx7@wjl>f<6-K`zGQ#I3@k{&tdb?BU(`q@A<ep2czeSQbZ~r?pQ^^lGknr
zGqWUCJ>UNN$P-;PbXXP$?iU98um}Z;s6m_05|&2c6*mbZPpHBRz_JM2|F9bXG*n<R
zlXjvStKnRX`Z6oBc2xi%s7%+;7Lm$_DL3>+%bdH}RE%nlfjaL^TS6K3i-1w`D`^FD
zlX!NbTmY0})cuWup(gTk*cpt9GMc-%VCoG&c4|tl2y<%b(Xg@nO5g}MM0If*9iU`8
zHC;`;VIf0JW4zp%T?i*d%r!O@r<xCMXQ|<fch--$qO9`zicz?<E`Fl2quUrMr<tYw
zL=*;dPCWxo0N277xR;f><#402r8(;#_U19)`==`d0AQ2q(sUe*u(n_w0GrDG1vXuZ
zNPD~^)p6KT1^PLOr{;`03!}blD;^8Zrgk5OfwV=tOmr>T#56d3RldqWG>%j5?&=<G
zrM#D-u~(*#Q0R~h%MaG`7d(OZXL_CyoyfTIA8ge!MXlrBpL#dVVs=hQGX}DxY0vgg
z65a}5-#54iYm(#b4Si`t#vH#LvlcpXvx!M$%uJ#wr4~oKGf(ffF{-dLo^taiW^9)e
z-R9?U6q#|t`=+j!mnXbiCR9cHoR5!(Q+YgR+!8v;g(b6wEjxr_W<tB1;4%3rT!9mW
zhJCHJ-qI}s7n{LK9udvqCTM2%L;WY>&~s$MdyVL=ZRIwx=!Q_kSwY$=((_ws{SdSp
zR{kAP<eeU8!lkdsnQD`-+AW?h2qN%%gyM++Hv2efN++?iFY1eZ%kTi~WU)LLI1!yd
zbAfK?oBJ6|W9U_~t_g4cWZ<#-P^9`w-KQO-s*67AXSt2HX+9c{jX*wX+p3l{C>eBA
zfv_`~cfVQ>Yw||O#=>>bF;j7RIjiueSapQxXDc91S-3C<i8B7$uYkP!2$<^7!hTbM
zy6~HWW@~-&V7(_9xXr(i(@K_)Vh*3iJJ-MYvh=7PNOiZ$P*kN-?2m!taY%nrfr(XA
ztQn|Dc5`O+$_%oc!yj01wN8s*)lIxJhmH#G%SS7<eqxSJi+*1k{{EY1o^NStp^uPj
z{|4_Pv>_U4yu6Tjw6jhs;IRAD30F-Fbvq$SIql&TEr-vK6~X;P*V;`SJQ@W-xIsAR
z!t?OQEZUpjeFpv0_NTLKBu@;bu+2TnUz{;&0c`D2DS0;Ja9_bL$9+^nn)1;#U>n)~
zIq$kEWl0JDK^sGK41kiVVI=nfmqun76pbL4cC|lS;eIPsnuHZQ1MxHw+uSuWu$dR|
zDvvi|J$DsX6lcQEXy5HiQ;%|Ye4g--lvz_cJ<kv+v6yTSJ>C+*q!}y<EWocHSx(4C
zBsY*znDe%v4Yf>$Q|Oc-b=qz}+vBloJENyCl|;jEE8>p=(>nLexwA%1mVXfR=6!5J
z8kgle_AigP^)K`E8=Wr@0H8{v)RgeCmI;Elqrxw>`K%s9%R49Vy2E75n4xO|M8VU?
z1P8U4+bIYuQ34-vm0j%!v2gDZp(O<pms5>`*y20sX8g9bk&(}VyG2T~W^jbQJvdQw
z@95LqV<xxm>Lvdus!c%k+EwuqtbV_r7|%1Sd*+R|zaoF&>bJcIy!pDHIgi!H@uUK5
z&c<ZJSe9@4_Tzn;3ZU$Y2OL^LpHDH@doKrZ2#voBmt;d&O}x9g+Y3OsO(~NUXSC4n
zLRw$wj|sSPY!tOJe(|3--0NWms!P8x+1T@p$%s~)KWWn>FkolZy8Ctc3ow!Z5WJ+B
zy41IU4b~UgSQ&<)G#zFk5+#59PkMIILWk|2PIel&1;}N;S-Qv8LWnqCn?7or`XsgG
zM)|gY^p$SMukKnt%yFc#YQ)uP^$pOYukE)-kBpnqD^fKgP|p+bl)?4q8>8-g8$;eA
z<!cS>z~q%bYTNw|lzxJMPsQu5pmk17oYJQGTDQPw7T)jZyz^eBD5tka55CDH=>>$m
ztKqeXM6zKTwB<5UMFX+x_73{aIQ@k?AkNkJO#`k@S9Gsqv^p2g=N)QsaW$NO4a-S;
zKFY0!u;zAY@VW0&+aTjob1ux_GI{B=5G{I&ZjT<EN0DM?k(~?YBGT=da?-67tqn~S
z_P(LHL2B}7*W8S7?ZdAcJ5xV7S2zz^R(`py)?2yZ{O`J%^#VB6A#RxTf%P7}2;o`h
zKzq|oZ0hYqQF?oCY2s}!X&TnXZ;s@iuERF4i%}%4fIGoV0zVa>ujp+)A5mh`MlmXx
zMzi}aFuO~Ywt{4yE*t)Jgpr*Ru@;M|d^E<pz1Owh>Z75thf`;lN@BE^Awqv7GTJ<V
zfBEI2fAD(u7mSl`!*d~nD$vt3{N%RpX16<A6@dK_>EWtFlumL7vY~U9_U<#J#<jd5
zylEW1j8{YuktTP#Gb~TWr0cCZe--@Ky~Ef%vZRTYF3+;(w8A|~U%%T>^d{dRjOGb`
z=D&Xas}oMlSvesbrfO#b@f%@A!CF7*Yk800-X)Y<vUqsU5LZQ2Q~7RQXr=7#hKjD$
zqyg$>!VhUs!i6?S?!v5G(TiM+no(Op6|g+KdE|jn9=Rs`Gn(#sL5~hbPfBcSn1o=K
z2$Fm3Xe1kJbe_Z!x0o=!ESB@q=Tlel>m6R4G$K2@DQ5pmqyU43|GsF@lwrS4)Bw4y
zmDv2^@(k0$5@FE%7~zYBfRouaKatx_zqJ~}?m>O7OPr#GHzq2XNGOe}F8cf*dQm7$
z3EWbck)~9bexhxYK1MHk+~^P5yC=7kFA=<ebR}uF-(rX`oP7v)ZRCt2^>tLs>^?3z
zyruJdD^^tOpyaS2kYbD8FLsMQG<J(Al!D2Prm_CQR#8y<fIh;oyFsSTyZFX@eDS7m
zJ&M7wbLP(Zf@b&PjOz}uqUW0bRZLg0TJg((oLP>3Eo%5wY=<-6zL~FLx!WA8moube
z6=Cwz%1J_=qWuo>6@ZFG1K>>M{BWt;UY&vE`1lZBQzbUyULXcB6flstyc-_>{*8}=
zF*BQr)*zZLBPDUrX6F9|di>+0AA<_nSZ676WpMNqM;^?V^0qNlSk>;H0^{w|j;&vJ
z6yU^tWRP{(TIhYVL&aY-$K=$b;njMgH@8*qMa4_IQ@cjOuG5envHI)}^MU5d2y4hq
zeNT4nF}Z#;Kc?$82UjzpwBly8Y+5R^a@t(zR@hPkWwkMbVw~&~N^-afB@qC&(*uIn
z^J8ShAg8E6^O^gy8YKNhv9#;;cR~bu2QIoDV{5q-;;1t2UUM#ulnvP*zZ6c5-oX{f
z(+b&%tJb*QuRL+(JDnXpE14ekE{+W8y{N0Dscc-kZsYon&j!1VEZ(muQk=3L!_=OO
z&czvtZZx(!wE4<Ag^isij?Gt*v=|u&ivV#7aC*^|xMwWNg!7-0v6UAp_8%JQ35^@+
z2|mn5ouzCdGK$LTP2C4b&g4f(njo7qWb_Z<{w21_%wsR2YR-h~i?O48uQ<>LX$@n#
zzc03vkV|?eA98&gKX1B6A@^F<p!5$~_EaQs{b+4&hU=%Qv66)kVh*1p0U$JE!GXQd
zyS*W8TxKhqJqKdNh1XG{u3=XKo)4`h0{Ce6@o;D$BuxgC*_c<7z5xU+**zFzhq$$P
z!@E!2CBsyRpm7JruXbPT{3E2&|A5CTih-7amGstA61LIwcj9_4I66b08-b;jh`5lE
z-&iyv4j*6GvcAW93AF)nl630#cwG>oUiwn>nGHUq`fN^}@9gy=Dbl1E38hBqPMl__
z2={CvGd(V@<dTXKibOY1o#n#9?m8Q$OWTn7aUr>o&3<5(sF5{6S=F~fkV=-){JfD@
zSQ_<9UE#-$&tzkHz5wa5-nr|qfWwHLWh_0<_a{INgt7=yf4xiW8-bR#87~+ZH3G@;
zP>=jJn)u-XJ8PGfe6pmZw|8`7W446P1&CrQyHUCM+{W%C4cTr${S|kmvS!!pbgyUT
zsHANZ(GYF@>r?|7?chFusXj*Ctgb}5>9M+uhMQmpAWRZ<r?}{yLv+U+pHv$JM@Hj%
zLFT)4z|6aL%zQqAx(c9iJ&#OshmPcOUyEL>el^Z?*otKZEibZydRkdQ$E>ZOq%ja_
zR1^7Oblbp|JH5iRz$Jz05&Dzz^<Z1*y#U*yt>82R39|QLwxE43RktnPNs>ma)Rc&u
zc!z4y9p2}9C&G7@=@NPFb$k9;$G0q+$%9!Y;4ie|59}G3xVz(C1PL8BS;1=b%oz|R
zrG4ZEoDc$1U<LtXJVr|&|19fIm8c^(5}RdH@J4l&J$fwWb{(ObB-D)pQeh;AZu~>6
z#y<wVq2DnJjqgx`CvFpQl4~mkyzLnr+eBRW9?V>Y<*_OuAt#a-K={mI|82p606X?u
zZ0LwcP`?<OVRF-@YI5mT2f<ZUT6Sf17JqP52zSh-jFhKh7OC8rv-Ny$B9dIQs(>b`
zJD&5X)Vz;cg?SUX2Jd1doht9vyL{|tdlqqN#m2wSPR>sM$rWs+$pm!OjqOxts$@|$
zVp52l^{mrqA?N*$_D-2LY}@3T#pZJtGrCvw;n&;phCbRq4oZD>N0f4ucl&{QDUF6-
zj@*P&RGGt#!Es&*%P^@2Bpw_kNsiq{b*g@9xhlgwn~GVfT%ZcL;KANwNs=_-Q~LCA
z8~fJz1@qJLM%qSV(J;aW)t0u0Ale28k>ft4w!!O|>eX1$y)@SJ?wQ`z$!jYtpwR;D
z(-(8tJe>?}O0O#?Wn3c*GV&iY(%8QJ9?g6{LEG+p`$yytmED4y79W!xYAfZVehBl3
zZvGk#qSh42zaF@oulwcwbO86zElB)3=Em|XVpJ7fKfif??|aE-{Zvo$?#!&)C8Wx1
zXg;N7N63qXkE7Vyt7?T&3T|swOG?!|E}W!!dJbI<hrW)ISUJPwbliGZQ7tVQ7yEY^
zGFWFOT5Nj7)2Y*X(@+NOZRacdiCOXv_qeByhDPSoxPkGjd0Ocxq5?b=<M=**%FKg_
zC*u0$UvDN!8CufEnNSGv8I<2v8Qe&6GNqpwy4KepUUl1jWPcp^N>Sm7&@=3C2y`Ic
zY^7i+ly2?S2g(E%bt_cyYh2@-+!Tpx+_#SeA!Ucq?%yF(6AW!M9mZ$f&f3Mkmkpxa
zsj6%ni5Jh<7dk%CW8At-juxS$7+lG_G045J)@%j!8FiM8M_<x|noCnTY@gYw{h%KE
z>U!fR?R&psE{&n$YBTy+Ta`*43Bx!`8vQg0!X$8^2fr1C72?iMd9iQtU;3!5_3tRO
zR3b)MidV__sYR&|->E!d;3;U2l9NAr3?=iON{UiwkO(PZQZU(psYmEA1S)84cAg|7
zO72V2gtrI^sLsW<TAuXXXc;Xt-bqd@IogbPImORPHrr|T<K?9x-fLSiLXJ?hjxga!
z!bP`x$=iLkvy7)3KkhhWK-Ry1cf37VUUaz^JZ=8{^KTYA?neyJHS5>|_ZRZS=3A11
z>*minnoc(hIP|1gU!{w_U7-1_<vS6%D3AJCGBGM?aD~lQNpu_PwjA2<<QW$E)$q1Z
zg~x&Mo6+D<szL953nw}42kKyErYl)4okY-SjYwI`qeV#kHMIjbUIUnR-{{g>`NdK6
z#n1HEEzej>`W2@k5;n^NuMFEg(!`rdH0%>WY?dx6;-!&x2}$YWVwg(i=vwJy;H%Xx
z;3o!jL;JYWTd7n}haAbMEK9Xse|m8_b%};K;3G}N&6lFr@{%_Dep>!|Zs*PDLbctZ
zW+k(57T^jxsg!s-fO>gxv)Z<^-i_tR>XiBT@JRELV2vnwd!Kx9zBlhp`xhsJ+8OUA
zWQpv!!IX2%+PiE<G3oBVp>HoU1^-%`FHfbMkVm5X4=m|Eb-R%{q0Yb$PG7+_ziDO`
z7F>Y?;etnE`m<r>!yW;Qe9fIkcim1W{b@@$6!$erAT*)$GLaouXE$25ue`a&sAi<*
zB0Su~Gxl!qQ(dESTl)AsZ9PA=iHu6eU^|AEU*rm_`E_Gk*(YJv+{KDcsbN*w|A(=+
zjEb{svWB69yVJP42X|@Q-Q8Ui+}#Q8?(P<xMiLwX!5u<^OK<{TXYQHjm2cje`Najj
z7O+;G>(t)6c2%*|pV3_%U|6)*uuLB&_Por>hP_rcN{!z8EIqQ^Kgz&z7iwxkS8a5z
zJAG+rGIdr(gPV#O--$>Dz8A#;9Z8uz^y;GI5=C_Kl<iphkk~tVG7!D&2d5FtXr)_t
zjdE^1t-(y2$OL|z;}U_x4=s68?P}z<bSIN;5Kz#CEb3E7l{y<+7i~l=OHNL<$0T=l
z8`iQ%%-Z-79tZ2@U4K7bR2kO+oh~7mRkV5&(J?P-2?<x8EfFM&+`qo?d^WQWwLThI
zj}$t7a)dmUQ9(ZOv6Earnp`OT`a01H2qLCb=u*Ey71k7sln3rM>3LF?Y3GlueK6_G
zeqcRE?!(9KOaw;<QviEC>79S~;m*Pj#9GN@P+chsL9Sa>t6J&ST0~E1&O22sppgwp
z6OxerDmMMR^#e0mIDIj(+Q-xcYPlke45IJ^pOW4SWih&N&-vSKkD{+$dXlJvT!T7d
z@t&b0Mz>_{2yTLP62b5uj7<WxsEw~p$h@p^<3&<2U4dp;Pg99wSxFX`ztR0Jwai&l
zZcbDeO~pMFs2smrFLvm<R&^n&G^{2Wh$@W4i4u=gChI-?sQ`Cf)WXD@;KWToE0^|X
zpQLvw6)(*Cr|?DR=PBRWzMm05{V;OZQ6()Oz_Qk0ey%M?<Y%`;hB>l@2JM~xR)f;D
zgEd!Y?*v|U%g$=y!$Gf)9(Aatyc#=p-MEz!a@NBOt2Rj5{}6^;&lC5M`I7d~-j1m|
z>b!|FD>T1MYx^p2T^mN5P6l<0=~{RLA5Jg>8zl04%GnDUH}rRD(iLR<QSFkoNx-ev
zRG}ytq{~Un%IXmIQd3>ML?jC`*>Vgr*&-aW;$^?^?XeKo^lzJU*UgL|2V{Uz`?9sm
zv)KYqk$RR0HkV)j9M4nK6wGuSwi5`OtB00kRhm~3a$DzluYZy;D*XHO!4mQbTf)yL
ze}1vlFbi%0hG424ewxAq5+?xi_YS<YcD!9nDaPSea5MFxhkq8=JGSt?_}q8SL%;#V
zE}s)8Z=Pcx<`qHse!B0J)#Pw?+~aCy)i}W%+9gg7t)<g;sci^c&G-p@)&Ljj#_AFA
z!PFx|LmAQnjXQ7mnF{T@G=kdCp8X1`f8(!FRJ{`q4qt~EuDVv{$KpBzXr!aouM`@D
zq5935&dg|dI|Lw+z6e2=9>H16)nO<a5=K>iuQ2soVW`{Aj2el@vdma6{k#O7TlV#5
z_Q#sy>O&=M$+Aoe`gWL@>~wpV>CML_1b)5vLD1tD7Kh*j4>B4wq~TFdINJs7aDjK^
zhr>|jF_n3FQHtez#@7#QX_w5qZdWutFuNFyhW9Z`X&1Jcy{&S3eEOP6EDW&GLi*84
z^M1S7Il-SSB31&+5Wezr+jd~9cdykQQ0<o$y+}M8$N#Sz+KGeQ8ZQX=Tdh#zu{ka=
zaZK={xAboGfHeCvow|#~LZu2BY(Xxsi#D0uov(Xh#lV25vCkoH`}OJD1KHa<$l~lb
z7elYjk6n;~jkrzj%|(!XIo0Vlez~^KY7A}F|77JqAnVW{qcI)Dh&e{7iH@a26_9IO
zx!37SX~XQ_CxI5oCz;Y80{V%I?54&oE;1f6O%d6YE8pFI*T>=dv1|G$;yAq;oz76W
zyIJMM?)gnA3&R!LdD|PFbmN-b7|jp0^`13YB~mm|yP{}JosxOPNf<_m+R#EpN)0U=
zSJq-Be!yO(1zsyn?8N>EAPA${!MVwJ6rZ_e9Rc!AFe|`{6f?1j_#1G}A_g|?EB7zY
zEZ1AoHo1Ud=8#e0RooT-9WLt&MKEalRAzub87RbyWPvxjb9Z><{Td~lh3$>f4}a~}
zR)*JJkB69A^u^d>xCPoh_Lm)5bDYsmuAc*;2l1i)xdBa}jgh*|fyjOhp1t~~uM`c^
zfiZUJ!tv>Dh>8B_n-gKknhR?ogu*M4_4)2j^u;-zI{{zOeVpCei&rUpEISEreYOMJ
z<O7^>RXg|y$i_xhUz>w{z8HT$pMdq*`&Zh1q%b}FJtMYtlqc_OaLx91&31J6r3{OU
z+Yw9X?c(jv0FvZttDLuCw|hB&QKU3?;O%)}9I{yKZ|5Ne1A|<E<fHN*iesleI~k;J
zwxQOu_-oQ{pR$X5+)A7v-!7L?t<NPwLTCJ+1(Df<3l^;_A$m<-GL*)Vt1HK?X%C}c
z^y%FeN)S<}<7ZURpD2sqZB*bHCsZfgMzjvtWb7Eb)~$>O_T!8^Y%I@}Ln@yujdo@D
zc)IQSQU+N#dbI{KcF7ok$C+I{n0h&-!V?i3EE}Q%O#(Zhd5^~ghx1Xm{zL7FN2OS*
zSS22#`_QhVj8!SN8XXfm?@#LUKLP~2%+X^@fGnNq*ujq;!yL|b%*J=3u@~wSZoLzc
z0=aEwMrcb^RXcMe4b2GXW~BCAO<~K5J(EZ}MEKt*UfN^pS9+KNB~e8cJE1@I{@{$F
z_oL{R#V{+_wdG|*%~!pJhhhv#>A>gyaYATz*~jx$G-?fBBY1U-^7fHB<&H$WzHI*7
zHbclcqpOEP<7GwDxMzP(Y2Ndj*_CbRk4CBi-y~Y5Dw}d@_CUydR;2fcv4*`ii&bh6
zw=LZzWb$0C@l><L_}=#G*|w$%UDRklq^H21mA3f$?WORo)R*(`Saw`xRaFdi4UHwn
zHfAn=9Tyg60V4q?2!KNS9|EY?yl4^_q%x@)XO=u%at)9ie^Ih4g=^^#nT43FZ|C-)
ztzVo739*uLQUt{IOKJdU{0R=<$@neiE$noU!)!1u4Vmtn7F+7ZuF*%9njd>#{hU~l
zYaDT98Me{iGy=2w9MFst0<qEIL}S0^I*&fwV!Ix72wwXBW_mauWQ4IfAL(xi_+s}3
zx2h)jK^C%}G_$k%!To_I)jGHY6CMw_zmQR9q7RzOJ!*E8MLZ5{G5d!~hyerFi&=1j
zC4usn#mk|(B3-5XunV!@+X(uzz+rXxj`{?9y)sivi;B6II@j?@&dy$_1WIaOHC6@a
zrH9xvMC5L(ovL%5)m1NEYu@gm)C7t2nH3g0^8y?!h4_=n)n_smrgzs1h6!ti{K1kR
zt*}}~JRdbdBIq=LdrzKwp&`xAtMHrhwdPE=XNM2s_J>KKG9#fE!%LNy`9^v$yD!(F
zh}IjY&>t=C>xhYa0=F*q85r5G@FJ=Hu897vA{SddWYgNYP!(yQ=EA^%P*YMZ<13oY
z)SD~L;XMd{fuOEi{v1LsJ?|<(uX9cIuLV;OsYQOe)v(zd7BZ2p0q=hI2C|J00I3Id
zLCE6PWXL=Cnu6(|ho~S`j3>zCL~}^j;+Lr@O_2R<53v6U%C~obH^%F8mhnIN81fpc
z{EIRD&B<Fl2avzp@DE$E8QE+)Vqfwset-AEFqh`1%-`Y0R|19Z_qft?^yy@H_*?X+
zJ1sjag42b;BgSI;a#d3P1I>wAj{A8Uvo4rCn4*n7J9}KcNT!Q&fprpqaeb9bwT!kB
z99Hb!Zsu-O1pJTvh-iFvh8^knE3zbmj?F|maN*c<GnuK+gw0BFJ4v=xc9mUyNWGC5
zXm88~`CIX>6ly6BK>gCJtV3hi9CSw|G+!BR3I*JsW|u-O+6`!bsUe1?(pp2IKg2LZ
z^MVX!i0$#@&iqz<uB9Nz`Fe@?f&+}}-Ggh><vU2-^&qW5{)i_S%S4o{t{O)|H_`0A
zY}ZTV3cRn22VeRW1BFE4urtrbz}qAv7cc1hH$OEHgpI@FYqYu61w~ETDr0Q|wtBoS
zU<YRPKm4gQM!G^vlELN*B6-vjONEWD;Y>JsDN;pDKOH*2SpD0i4_Wcdf3)>Y#nhC%
z0J0_G<(ufA)X;^u*X=jKz?<Cdo!Z8J31#I(cQBbpO{HkPK<^fH!LL&GfPa|K2k_$m
zj-Dbtxx`y<_MAYZ9|L&ATRj$%|8ig5@Q@Kab22rwLN4EshjIVAFK#Q;M>qV3Il|H<
zp78QF$e8%fDES{{SIfxCu`<*v!+)hXP_%0!cE;QsNsh+&k{yW(tex&Hhtleh$m`GU
zsH!3L3;l*QMhL*fURD}I*yaFsxN(g+=0^8)Y$u>&y+I8qnv^Y7md<RE2d_u|=+Nf1
z#QGozP)ZC<^}9ANrT(C-Yy4q0t5RNF^NF~`ur%5^O;DobcskvyI&iY}Q%0)3c=`>>
zC~%=^*S;QYg<3?9CP5c3C31XVBjRUiI(>cW2ZVS75?0N&HP@0h(xIr>tF7<*S-q5q
z0hN&@vs1E$qP$2TJemE5*?o44Ni!HyW-&P}&N`m+8q;K6o|qQ&b}!_cVVG$qnZR~Z
zB|J(MEixSbnJf+vG?jQN(+n1&9@MSJS>dxqgMK^G+ey$LHrO0PBhQ&UetT{`XD12X
z7miBNvr4KnrM-OCwMm%i0=BjU#GK84;n*?AD-#2e8X)m-`1dH_z$0R~7i{v+9r#OA
zFanNz=F&Qn#Zoqp#X^FuKWMhSh#=dp<pVPIjE7Bi_3cdVE-`sX+z&jdr_jteR)~2t
zxpYi*uAJ+X|KJ5-MEg^os%>;zAkjJz#&Mj;?DspN?|2~w4f0m@gEZxEl)o2<`FJqs
zJ5Jlr6W$REP&>m}5<2lp&_?w!*5GKlb;-^NiT=)Xk_*x89w$}awJEoMR_v}^po6UQ
zBCLC&bFSVS=$odsTswtR3bF7b>ZsJ*xh+K7t)NlU#R|mg@nB3g+6rfZk7H;1iDPW+
zg@Nu*kc3${Qd&58Sd>lSy;Do=-zJQEa)(RKUx}n=bz9}=h-rYemOyG7lVE$nwp{R6
zgV|ymIG)qPqgFYLI{b6Lt#iY?3hMrNq75rg2y7R8It<|X@Oo?VriEJAcbH+YNGY*l
zq0uvMrY^i)n?CE8>RpjA{5$*KI3$O7tKYxC2!vL0|5XDZcNs(;fS^ed<mtiPa>L?p
zy5$;c0)XtXUPZ6m=)MSl)(@j+EZ0cNRPWyt;##G<PH;!9*NnO$ScNl%77j8*&D7L0
z90b<3N`>t0UQ;2=ppNp64}hb3M2GjyRQATEQ#qJTsn1sf>5P;75nJyR!RXgQHRMPH
zAN8s{7OP2B5ml?3*BHP@&nq(?o$Es6sS`rqovFDQ`|`0eh!9QUUD3E^@7J)%xn<w`
z0~+C!#uOH2?qi9Z#rNFTG<#|)zDCN=haOu`A<hifGkKbQ=~Wf#7a-N>BpH60OZB#!
zHrr@nCZOteKdlqnc=Vw(-f~slxdDDZn+B<S(jeLE&rvo_6j)XQoWA1Y@=~!Kwk^lm
z{Jk0OCmTg%G;9cyUakTltwEY>G&*Ssv9SX8iUFn%AqV3!4L%!vew~aMhFm`(%@>j%
zKeiuFFFLmgIs+8no=CN#mUBlq18^!Jnc;tM@^y){*eV9optXeX4|$>id90b(JR!Z>
znNX?VNZLx=)d(L1G;}FBIS$8br#8XU1?{FG)>+~+ZNwz7-r<^+cxxYKdp+)#Dv^-6
z4Y8?pQS_t;(U|$%af^pBGvI&^z<Jw^mlP!c2{{#?ikcH+RmslwgSLVYG2G>6t=k;N
zrh}R?o_CZqWYvW4eHP`DIb~0j{tSE@b}tqGwh5B^WYNg%XQ#DF(c6NbjKMnbSZ71f
z$OE4T7>0xDDDD`IrGjmbp01KQsFsHz6Hb8dUk?jJrgS@-!a&LC#}2SQ9dxmQ-F?zS
z>~i)+k)Dnjj+?Y-qI`kbYnf1;Ux>xKqvb@y;=L<bJIpopa+{%}&{e1Zby&G%bl}(S
z{oMwVb_ZULg>wYHc^@QQK(71PpqKF=O=Re1wfe$l4$x9y)${$uIAj)e(cR?1o)~kL
z|9SNqy4G{)@yS2AAmN`}pg4D0@GlMJ@6rE@#{c115Rg7TTy9avY=M;36;#q3MJb4$
z&nj3`X9byrCT;B?;I7~LX(=nlD7&YL=YWowQ7Am`v!Uhk5317l=r)9tsu#8?O~k`?
zXqw#V6n7s`v`L>JY<VDI-E$)Zaed76_gFFTLKF$zd4TMj=X7w8H58f-iL}`_f<^rk
zvyXa98N_27GNt2*VBc<|7=1+T>lxLKY3f35-H9M#kxAHiRv4w%?XL5*FMf+a6CeCP
zm>T?!85ct9V_F|c7F3fT_{)i7+*OphgVGJ$zXY^nq&zrbCd}3kf|*KOzg=%;b(y1A
zFvpDt`|NgEdm=RCV23(bOYa_MCj)YtyL`Wk8*&Sh&Nhk$zg$bFE$PqAQL;09`dU2A
z6BCxK<5(;IobG4feuC{Yu(qEve=hOfujb$5V(2-yI6SpLhR>r}B?Rf)TwVf*Zj?~-
z@Z4>9sB|)a5+dbug@P<|&?)+Qrqbs9yxZe_7SPxk(*Psjq*(b6-GyMN|D{wy!h!8V
zbo~!$`m1*}BTNqc`gE6`#=%XDi#G2GEeJ24GiwTNM%B%`X0R@s8qEvv;PBff_Q$x9
z>1}(YQB-<B@VK7=Zv-H-+sp#|Q(B0(3-!ZxQpg<zensK>iGUrhF7{prW!U+5tmAKp
zqTm4+qZe3P!`I-+Chh>_JKQswzyRo+Jf$YYD*l-kUez@xfP->OszN{h*qrfKkXk!s
z!|^Cp`1_>nfyfjZ#4P)UiqFN)`QK)cjflu=gAFv5X%q_(Eaio&!{uT9k)DXnWz=F?
zm4vtj){?bXqg}kL%wAqhe`jMekQ0RTbG~ysix)HGWI|b<O&t03OT5d>w`qgQk#^_R
z<$mXRa}}v!iQPzUV67tzY$-)@M*Mz~zwuNm-2+gu5(`!58%!{6%=#b6Lcq3V;YucR
z!q27mDiSdLR~_bq&+#8C^_(qk>a7${6)qBm&-rXR*VljK1A(>5sml1@?Ea0gMjkFh
zkP`lX=uH1Fq;%r?ow36mBdbdbQAfoJy#&^7Y5a2Sj+cGt4G-m#D5TgS%Is(4F0jQ6
z2v<)MZbk%W(RHDUa!(dw_6p|k{pP(v_yON|?HP%5?UkJNl?yTKMaE*`2ZOiw5tTRZ
zwNsr?7a$onnck6A^I5KjMN1M_7m3+zi^5N<8>x1a=IH!Pi!XjT3!d@(pc_mw>j4Uo
z`WWc2TVCm@4NTM**Pe*()efPJs;D8=22`l?enL*edO0;8n1EwKsxffwtwW6QxqT4f
z$G><Y8R<$Ybp`jz8uBvXEYEJlIM~3&NEdI{@x>Q>Mh&ZU9xviI>$;7j_Q&<YqKlhA
zUBR<GK<XoA^zSetuTYv`aT+OZ#gggiEwo1o!1LMwr3p9p848`WBIfj&1ai-5;`YCL
z*q_s5g(>sn@|pcsZ)5#NY-=wnpjZ=_)c<`gcSO)gvGJdGV3VC&{AN#beeU0gL`N|R
z1eNywUmBkv-Y(se$JlN|N2{kU%uaj&<=K~d{+Mm}frzg}F2kk+A;2R}1Pgx|vyn(O
z;8SVdGs>v#KrxVZ_URc_%w$0PC)*sq`weyc@7!>?{?46bCv=O~epm!X6mjuM$VGHK
zXYUZXG>_!aqZRzM`Ui1`Ak6^r!Vjz(4L|i7V};Vpbdcoi=Q0+`PorCXJTbMC@3@w>
z@<{u#^u512R`?1(NvZADkV>G@;XOy!{NY&j_(^@aiVHQ7y{1|wSD@7{kesCz!Dili
zwH68L@_A%MG02M!yxE(xaZa<PZpg-t*`7%p`pCKO*8VdH!PZ!q8*d<=Fj267C8DcM
zAniN%;I1V}C*&Vt<tn@5)Vi?gMbTo~Bh0OmeOc-Gx)gAzr-QWy8g2bs;CMdN$u9jG
ziI@%>Q`7i(+P-WIe!GAB6ZBB{b`w+s8wlAKLGo9vJ^opJ(f`a!8t_}KUE)mi1uR<C
zowMaElzuKei&))PKloF==$q6!<Qjsg@c$ixa-cdACG@ivz*%%fpK4YA!KfeGOyPi9
z5buoy@{ct84S6a0&0r#(fib%xRP4iEz{p3mA=7Pf;7?8%(D$SfnC;vv%}dU#&o+$x
zf>)^KHzlrh*9fsN?^&F^)lKPlTMa$vnU6RH_&eumn$<Mv1`jB8F-iF~Gr&@a8j&?l
zl4-Gm63qq)&wDZpWty+!$z}2hC36gFR#CxB7v>DRLfVxh5nyk`9z&(h{-wyN7T=Kq
z`jLA?f~<~x7-PJ_-un|_U?(-+ADamC-zD@yU5o@lTY)gG;DxJ=C!F8exRD~97OrPg
z2!`BLI6Bku;P%$xQ6nt01MZTB?7j>%A9N?s&JQx^EFQ*?OdqKt%Zr*~r07jgj;I;J
zk`vKe^=L0|p`3p!Um`)0xe(;&Z-6vTw*PEyGaTa4Zm^{uZOtN9Xr5{J6#FsG{xaz~
z5)KLq@{0FOvh`DE_(z!eFLJo|qj>6tc~*%(8=Ii&t?X!mSXnx+plj_bCZrVLi(^RO
zby(O{EQl_u-|kmGk*q$Z!n7MMY>ibAj2l(H!0H6>*Up8dNsx-JK>-`MxmX-LxdF0E
z*}XT}QSerX&qd+F{zRGwZ=22+1J4AZlU)1=IZJKKrjyAqOt2CF3W=Dsesa~Jj*`59
zR7yjvJYrae-4FLu90mvEMNC)P!d`H;ZtYqCj-;{g6l2^`9WC&h?~0o95&yU?>4pN=
zc_JsAIuQ@}|Ktt0DbKD?p}M&z@wD*eRbe7bHQJ9Pb8;gnv5YNq{Cb#kVLLa#jDz-|
zVz)J>9t$Fj^6Qr!_SmMgSo;wMD6agJrxAO=iK>g28L`{yW&Y*GHe?P#QI|NjIBuW)
zgE(KoD-ULr3NON7OH`&QlbxlYjwUp%R<^<dTg}23`fu8Rp}iF+RsM^^;TI0^n~7u-
zSgyWwSouPapLK9jsF7RCPaDtrJxj-kIZss=xkWz3>ywb+c&gYE{TG?sIB?Fg8?URF
z5%5Wa(PY2aUs87GQpF;YEJ`dEBw~Z1Sun}6<wBR2Ursm(0#Y^1XSz#!JqfQ3e5UJP
z)i!S|3Qphs#D))4Ltcw6NZ<AX&=xBX(fXknHFd13E^3Ac*xU{BItg!~1rYJp4-kM}
zq|Gm%%z(qjN)YbdWyk*HOc+fBh}^$RxCb9aqZ1N3z)m>YreD?*Z!ug5Ws*IS?Vl7s
z^*2K$N-F)%BP$QpHkKFUn&gDv{7jANS&g4(!C0U*YvGqF)Kt|(Y>r)7g{0G+d1RQK
z5p2EjDJsP!#S^tieiR~vHrOh);y&NhTv~k~EO-mVA7f3_6i$BQXM!8ZCrJnd<fLXu
z?vWG*T(0Ja^-f50b+!#NbTq{}-8k7?fkOZr9kz7mhL}nC{R(5x4qm&>T5??hb}(Bn
z42ZDl(@iTzfH|?$2RDKGF;$~A+0~fi*UG)0{f#OctiC<$eULIC|26MBB_3_>7l@Q~
z2^ogF2U(k?HgVHXbN9UAF`NIbvM^A;y1y_lZ{vhewt%O6uHe5##}5v7wv1iseb(+t
z!b<XpoG=7;#7L%CAG!t;M#9edPm?{c^5qi>q=$$Rr-j28H^DZ)ZLvfjlZfs-`YG>U
zGcFO8>x{4UD^{WO<phf(WDvhm*rz_U#XjpRAYY%%?1=pmfjNTNAIBColbcX~@wxxd
zi~T0p3w)#`MGXY<x_m~Y@{h78dLu`<A&JEl0dx3ox-OOS&$+OVITeJLGfXC{=K$jB
zrEa6K9)LlwCw255)m@Xxujz!pE~hl8oB*RvJR$OE-M9t<-Fps1r{+|wX<+`mwsNBv
zY#}2<M_ES~r_M>W;9%bjzd(TzTGo{j*0By&bB#<ARGw;&^!jM!#*vAiioF9@d<74$
z7|zfBqS7q!LvyReKtWL)^oFgtB+s(FN?|<&eyz|yHulK|o1xOVy@faO-CA)_7PV;u
zy~ng8mAF*=r7*i+WtwT|(dNhFFtUw9gdeqRn`QlBm`PG9@Z{d@*FvqNA%{P@g#JjU
z8bGpbfl~We@yp?OzElPi2j_!wSXh`~eZec>-Md*x_djH6guGATJw$XV-CE9-Tu+rZ
zdxBh$j}q$;f-gJ|ac}(9R-7O|%K3{W{wW|{wuI4|jFKOIxz8^Yq`xc~od20kt>~VM
z#Og~`R%NOgQt{$ZTH3`SmY~K3)M_xw<NSH<rm?$#8WrbWt{cLFrj&jZj=gVmz9kpZ
ztnctHHh1EaT54FB)i{P1ErH+-)sBZfa=*}LLXW#?ye5Bqm2lC7^^WZrHIaSxF)w0g
ztifK3ai=b}-^Lr*zP!#jMa0<bNK^qZJVAReQ<y*^cC;jT;n<N97|k~A!I?gi(eL55
zPD#pfE8?{+x$1V<vcMPnJR%9k35xnN!!((qsY5uocEVbJlT$#tp<?u}P7gJ*YS9fq
zIMg_cW}Z2lg*e6j1}l#v?sEpOUb3-%+(Df>3^C(yOO|0TQ?Sp;wGr?=8``@cBW9;`
zJy(Ov&sX0gahQ<=BHR70n+1DNXg;Farrfv5Z**<b_|YuIHe<;k*djUAxD;+OJpQni
za(+Uj9vdmCDJDozN7JW9H4Ki9Bjr$nF{udo#dRogRR^ir-331823J(J>=`?PrJ{;O
zY_TE(qx$c9E8G7~S=Rn<mF3>kqOUtJK<@*G*<0FJch+K~^J>)l!)~qMFDtBJXQ*mE
z4SO}3#FyD!{FCXrqotFh#+}xPvo(=!NEC>@m)(4s9(UdXuo=u}^|rQDmy4zIYHY!J
z37kqcu5~ZKzqMh3Lj%Qx{MgE{g5d)tUh}RuRg`Oz@M1tb6>rkmiW4@w*<=E6AQ0<(
z{Q=4t*$<WTZf3C3PNA9aBcd^?&pQnsl`+>s3Lrxf|F8=eMDRiXL9Ga~1H}M9!|9WD
zYE9Lr6l!N}2*e7Z@ul8na@@Lvay+L#fo{ZT`P@(%^<xtb&|})QQq|Mn0hN7Pz@#Ep
zHw*OlA<8dvg;IcaS)tq!C#isUSPKfxl#K6Y8~KAVkxO0wOt~8J3KIv|NSi;r05{qx
zF)tp7hd=j?^677cT1S4p$S%wFP|FT%f2tiZa`K~ed#??4!cM<`&$3qmDUilbX-bZb
zmVWNmyB$2f9`w&dK8%7=GUokXQCo_fX~FO*Xl))QGa+TWg9i%={XrwW?bgP#w|Bvs
z{jHW|{ols!MN8~ky+2FwxGsSfdDJJ0`&GeJT4Jf6Zf)zZqzR<$o*_)3(bu-#+OHoQ
z^B^hQ`9S{`$9^vDvMVIW!hln`6G2s*M9SZc_<P~?XiFMyRx%&5OXNU=jYPo^C<J|{
zA>V7q8u}|Dv<3Tj7wR$i_%Ww-=b=<HmEBFF3-h7mhzE*&8JI&@3zsY7GM#d&P7(aa
zX;|5Mk(his&}Es}F-BoA@E1f5mNZEUAi{h!vu$cJBy{WOFxbJ_=}9ZHqdyz&AV}xg
zCsvHgDj})Ajt5yeu<#Hw)sCZo{e>W;9nAT$zDa=OdJ|I=M(KR=muqqNXyQgJ9!z<L
zg`6ahv7s06KCllxHIt@h$^3a1Whh)~*V|VMfOjespYZI_ELaK9vr?k3?L^FTk7hY?
z5y&cb=Mtb>G?y%re_pf$<$`PWCcHV<yNqJM^Re-#jw940Vp2}nr|LWjTlPN~kP$-4
zr8Ycl+gjooNcZ&qx+4C#uZn9RBV(p#SIlaa<1O9y<P@(Z*<|V83jaaFvwvN=r@c}X
zjrhea-g<=WWD#n8|1a(Zf%y)93j~ng0`bIwO|ZERBd*CPG@yos>(Ygd<#E&aw9u)e
z=G1Rv8$Tb#U1=3D9OX5BEpCIEvG(Jf$esM8l^c=P5*96t9e36(*hCw^l0l{y)9rV(
zuM!h9c9Yz=^Ew#CVYOcmY&9budrTO$gr0hqI3M*;o87sOxiFvES_PCUTm>9^t>m=B
zs#Q|C6`6~+5VfjV)rO8*`QNO7$M`c@zT@RpGY+>sRA&L%5kUdcJ7Y?`7GH<T=^X{A
znk;qtVwJ16>tSqR726_gZKTTcd5Q=hZJ-3xojAK^cgq1@XRJN`3Ry{c-+W2AknQz%
zysBVkU)vxZk;1Xz8dkPVYo;Fjs+$|Ji^E8gh0H@w!mwFnIWd_5>Tc^t&+N9&TyV1q
zgGz;C`!N^y`{+bTraqR09i&s>;$HQ%#l_MkR-WYF7`{g4M3svQ?VS~ZDhYkO*+*D1
zKZN(`JT;Gq7nsj&@hichr&G+-BNtK$;U^zt1>>xO+3SW|<R6q$F4UzzTMjaT{)10z
z%~`%16?}IaK)NI|f8t}K^=za2BrMdzuJKsc8DsOrpa_xj{(2h!YY>{|l`Q71IXGg>
zEKvcK)!7jzFRTi-NSDud1tA?7tr_xg3F@oa)XKD%!&$LF#J?%<?gR*d31P^~u9o9u
z5TzIA8S8w@hj;8DY1FlF9(c|RSIVO_IKq<3VRgdN^>O{I%WFIU;%O!5Eqw6vcYnZ!
zv4l1CvTF&~3D!p}46gb-y|PXx-vhh>Rb%n`cz&mFYe&A0ZV-N=DnRhZ0xu_w+)&J#
z8T{bkHcabtgHi99L`2?IDN>C$wWW(t7LLNIeZe+}4UvR39Oc1S&-<u5|Dy>~zQ-3v
z&IO5S?Ad<&?8@GV7PRG#M1@bYTX5LZ^!_KGGA=K9X$jhVwoMcD3w?o6Cz(Yi%`Puz
z$()NZK$=x~FioZO>LIiF!S+V2PEa*;F@;vM)BTlpA5=uWvWENMrdwxpYJu%X(2>%U
z(r807uukn_RN@54%wwv3U72GQ8DTpzVHCrgo03zJQU4%B=wQE-BE+u!mfj|>M=rK*
zG>wHenWVSe_b+<>PH9$=QzZDPSp8k`QC~w1+BZSNn*QZ=$x4mNgkKQg^549clV~|e
z(J+x|^10$e%a5M0At4~!rHMv>8S%A!ojc<>LRATCj76=OY#^wAaEBGf@@a39YLdfg
zIm(-2Pfv*SH~aaj51sR>2m!BiVQ_P<j9J+>JIXdYaVn4UnD7w#&SQ8c2d@t+#D~G+
zdEhkudK~b@aY|B6gJU9>Np)F0>c@BM+x3}r-SuvIOuD2s8d5-de63N@7uF_Ad|CJC
z7T$DZRLm;0jB{{HK;d4968g?M$b_O^Nb7Omf58(!s%mWq=ThU=aCPscbEgrpb;K`y
z)#Fq3A|G!DNpeT_NMG@4NXoG`P8t_^Z7(d*UBQr1;h5RaQ)-mL-01a8n!WT}x~r8=
zcy=E>Vp<lZ!A}v*KY8hp{zw%*oI~Nk=8Kui^C;tK?J61}PZ$u}wkwb#K46wCbugvq
zXnxgGEA7;087WHx=O!E>iP?d!EO6T|^@r6A8f0Ji)7B`jMAP%*!XJ#P>RrOMpz=MW
zXVc%gO$G=q<*9eOOho+qC3{^vJ^y=(IdJd+JV|xY8hn)_qjI0G7Fqn~>L)?u5%*>_
zx(ypvzA|Hk-9H!Yzge07HzQ*k<ekmOhv>ake^s-5h}p@zTRio0Hx!fzLxGjLvFI<U
z1tkTv!)|ISUC@u{jT9oT<%{eZ=5w6gh%8jD8#<B^8Sf47xOxu1zm#~m@n*in+-Nz}
zM%lM-477luhm2!dKOeopWZOYUys&_*34NfnqWMl{5_hJzj^QJK#4yV6rDjG_P608;
zPzk_sYD6`1b-z?El<B7&Qwv_L=z!aOkwDu$+yIsDeez&kVpogdZSU<S=Fn5iBIAem
zL&jG7re3}9^$fJ$1#s+H(Viwt&#VeBe8YR;rqJL}7Q#9L;=?v#;iCzruY7fwk<h0>
zIv`Go-m$3UV7o-Hb`%b+UQgP|=Jw-&Q2B;J+$fP&i#e4U8UK{%kQ=Ny8+{ZwrKAzC
z3EALcePxpXD?EivtSi?5R(X2@;1R~*9Z=7DrKbk(BKc|g$GMxf!8g@ajcUi?ErYf3
z=$@AJmTWV&SG5pu4B{<}dIgyEYG!Im%G?p_*q#W+HbKjs|3k+#L%HlkjXsOc22&T$
zA>dzdHH;QH^Lv9e<EC%aE(u=xvOgIFtJ)AU{D*4)-zt{iNz@#BK}=qMcE=2l*FYYF
zT(x!mxP>3;I0(Ad-wTH4W5Ln%C{eT3s1K)e>b-2LkKR8mvRXdAmwrkbl-<SzQmOL$
z>`!-joNRX49EEw4|8^6?zIwkE_l%8qp@&q69T?3fD<-WaV4uT8qcm+_uCvrbKT_H$
zt`@9dibbT#|NADd1zvg2A>Dgjvil)EhiRgKD{}t_@!fmIEJ`lKv>&*K<_5#kI9m7!
z)N=d!8_6+-02yK)U+tpwB0J`lZ;uycz0#xeKj*1`#|n3EMg>=ZhTCL6+dC<zaTz3=
zEXT)B(Drs-kw49#jkLE)%-MWLGvpLR7|6y>sI``gjohtSrRefi1~%l9#EE-gxjm<N
z*TqkPlZT~s4eDL_%a}0$hnt23S$3MBtWlm9aIhIVv_j0d9yeXuF~(t&ZgX5$vD3;)
zwJ#Yidn`$MtjT9=ts>@oglX20uX(>6sio*jnk%%=-DKO#v>bnY5TLFfn==@$d*8r-
zAN}t&%xxGh;fpKytO<!9XXlMEo+<-dHHB)AVd95A5cJBWp5^=xdi`rC{@)yjAVG-Y
z>GOZs#WRVxa2FW{Iemxp*Y(xx8Ag+jOA#0lh9K46p~Zu0Z;ltKjm|TXjb1xQ!96eB
z4N`kN$WQ~e*a6|5QAgPeceAkAVO3y76GZ8P!VkxF%x%umh}lG9hsT68btlwZ8bPh4
zh%AYCy0B?ogYl%F?dXSMiX`T!_+qYQy*<!hHGO9i@)YtCd`e-2`0?k`Ob_!Wj(XYH
zEnPrz&K%}p!)p3kHAT@vBbZ(GCzywc<g{nBNkTv44WV*ig!)xRdq}inRHWjNWsn^(
zJo(VKWk|p>uMAy2tHy@Hf|$!QF<5gbU~W#zt3uAbc^he<bJk~U-WMPEh61@clu;I@
zz01)C!#w)hs7*8Ro>H})m118RmTTubd3RRpo>skE$(H-oCQLtef}~daT0+LC#Rcgz
zKq3<3FYOm#5yPFk>=Vgp83yl%$?G9@Y9%y3ruW}CKLHEF$9zL8Ui$M6kw#7laEacF
z0!fb?V}mqUac2KIyDu^a5N&;tknwLDyPs*UexFo7Wupyr#(3#xEB~jwBS_koYAMV5
zT{@~y_m7}9GNSS}Y@#{<+XfmH6BBy#Rgap{c3C9KLQk92k`T8@W`_uI$?03sZFN3A
zTt&3RFn8cZ+5~onZ~YWM<Ls><%Ax3$SBQg}m$zb$_E{7w^&ts8d~qz<y(n%&Vv_pR
z1zPxSGo3ukSV<b{M<T}=?Ka(9)h+#yy^Qxbyy>M?{V;OysTdi+$d}JT2L=jiwoEc|
z4g`eyM056?)dhW?bUi-~Acg*8EA@b&_^1)3b2r<vM7kpmtOrJ#7makfWN%th@od2>
zjblP)!6uw<DV4kV;a->izi+(xPR%y@|9C3mbP=C%5?Taww|{Z+7TROzEx2U0KFf_^
zOY(;+{C$fsu^Quzcw#zAQMOc&@_=8gw_xGY$`HNSVb++dA1{qE>~E%s$Z%@Z2gSJg
zD8`LqdyqgWr!|8DVidzFZShj(L*#;;1a1P6arNe{y^8r75$4S>>8l&U6Z$?Z{M5No
zIdsaK_0bknsz*5qM7<5YzYa0TuJo_d%v;HKs&aCJWN$=2_nPK?Tg3kD^r$2-=jHDA
zeQzreQo&0g`uW>?#MI0Yi_1JdLtak2fFtSB?Nfx`*#wETl~KhPQy9PVN#h+V;dYPp
z)VyCSA);Yw1U7C$RlA!pjUPrgdn)v=)326<@$S^Kha0-lGE54=S;O@;%#&}r%B-?g
zW#d%?DXcYa3e_(riB9(Et5Z{o)YQ}^#l^}~OysrlY3q0>VQy*>xvz>>I2i=kT#dw!
zOL*_zA1=gsNwJH!dLjJZ+~~$8oX5OW`_#<QVNjaYNydAJo)8)dt<!@*L#n049h0b&
zDE&48KukPJ^?IP!!}Ve$0ehNWOvn^WIOL^064jM?Ia8e+g#r>s(V7^5V=1L@><yc6
z;F(xb1$lWu_Hm|ow;bCFUqm^An`jK(6#P>b3*OOe&hbLUTzEtsiC~r4(q(dup&B6a
zMGK#v;o<NcP|?z%<+gMT(|t49nI`MEd)3A40Vqg_j(y%gX<}VRKk)$UWF96pO}XYV
ztH{)HJ*=MslnkRIprmwDN+%D4N9P%#G{Z_F%^~Id9bGHN6a`PJHgW)TCC}~+*o3~x
zUwH4MP}3i1#p}ZG-F`3!qbUxpbAGg-=T?Y}1nH2+?(F_;74KzbIC|5=&iKRYBBA3f
z>ey>aAAQxVL}{Nfxk|8d@IuzI%(L8Q;j^0Cya3zQ8{Zk{^C0P0@7k`^mhnF{)o1dk
zxrE4ga(Y;EX_Pgm@_u|Ud->|$XK?bwH7$V?FkB>U*iji+k~Aud!{Re|47s8CQ!dr#
z%*)5&A(&~0B*aMTjvmD&UlU)B@ygN{_f5rF_TDo#pSg1M$4(B#lh9pVuYtQxZj8pZ
zGFm0(gPf@&1M-3qC%yIyW>c>oj{CfSjE8wYRf2%eZd^|@2t)0hpm`T<goZ-Qc!!Fp
zof<t3U0_P&a4exNO~Om*)FZ{`Hr&bTMX52hQ(yiFE4?=nI%?`FOr)*tk(*nUQ|k#6
zgfsKMDkenzGQDa*_)p6>ukmxzZRzK7>P2#@u88dJqbgBcr`>&_sCYU|?^~29DOi!(
z6P5ruc+etoZL8&GR9vRP!CZL+8476{J)h@{@yL#@&m^q)+^K(J5SO`)`viw$!D2cu
z8K_r9_8f^Ce;XLWkMw=u+!80p`l8mwU>1iO@AX-$yBG_GPcj({JOh-qg)AyGc-Abi
zQbd#F0fa(Uce`kI%tCsJID|}r(r_qg`v@pkWRVmi?tdVYpG^dDu>GS<e^(@DoeIL5
zoEyMqeuHA`nM8|g)e4r|v-B8VScp#fEJ4h;7UopRWB#i^<x+Yjx69!q4EQSu8}1<q
z{3-YfNHHEF0(J|VHAVT<B;!n+(DUwMh={|*^y@3lN=h^AbL>6UKtF~OR=?7zJ^tZD
zIdRQbDRYz+su#I_frm|wBK>~Zpy~-){F=4cu=CM|wVE+k{i|JVv{_Ad2AN#q`eVa7
zHq$d#y~FHz5W~EhV^s%6vbz=bJ=(b5J1(kFH4P(1AIUFiUY&A!n(MwzFYjNcMDu6W
zk7%|>vW~xOfqi8J-`kTQC^U{OyRAO3?~s!krpM10j+pIixEhM476EAofiZ#Zr=r3@
zkPYnq_wp-m>uEJb<cW~&b7JBz3^KkxRA_jeJK0NDKRH=2bY_*X1}M7Bq{*LVQMA{y
zFu3Ask8y0}Q)V)CVs;=lzb~;?Y?V#Skjkxav613juloqhG%52I?hwQQNAdsj4JJwf
z<q0>g`!rD9)}~Mi06_b&K%X5fEdx4)gF{P8Go23rLL#Ci#*z)sVVYS@w$cYW7~r~<
zEN4y);dDN>HU7Rul?>^7l=2jtYvuD>@4#YgHRwo>nw_zIea+i1E#q-e!@JSM4ONMe
zE%@>02SL|`y&4PVC`JbXjI#s{W@P06W&5mRs04)#zHZ#8JnS@_G=7rkiNxVjgMwl2
zkO8w-Uel(xkH7;?gfOY7N0gx|)KU)h8JQ=vLHak$uAq}m-TZ6*n%vlAoy@R5RtziI
z_^~d^0ZBCQndZI=D80Ckc*ZDxXh~5*p?+Rl5e&J`BQVgDZm$*`cKeb-z!wbTBtPt+
zwLdlxeSx!1>0aDh947@<IVqhCr2P>umMB(rQ$K=Mru6uc^9LDWFz7*poUZ1Fj{CM$
z6tNLA{AvA<VPT|ETp{&UCI-0o*D}MqCP0M!qdV6{)+N8T>wawsmYe7fi^o(y_3sZ+
z$QD1@_scsilHb$$Xsd0<ehTaMzUXn4EK4tL8)Ajkn}~d{aG~E{p}w+GsniJWQ!^DF
z=z#URVQN17V&|Wq71mu$;O%mQDAaj}LHcCPc5av)ZICf)_CRToz5)%%+0Pi+>2LJf
zM;$2D#mES`@I15_-{q8@pdY0`WU4R2tE|mc{k9QWgGaP*lJ7lv1O=(6jVYNBkDTNx
z@1}ozcAH&rjep{UHW{-nttr+U>XlSksF;N4;44y@oZsO)jnQ@m|CWvCSpQbX%tNiw
z@h)ObA>ayQ4Prt)|5u+4L<JmoN8>uA$4d=b_=#_zrngu&?Q%(pkL}j=DYqc(;3qZL
zk_XlOP1R0HFrT4z|00q^mplEC@-5{@^(-SeOZ9{TQqU^BS8Z>?%{#=-yv9oHP3jdj
zL^k;Z2kzfG)o@BPJ|15@uEzKI=cr{{irIc;W@;p`6Lb_G6}&|dn*PEfZ;cTQr7dn#
zf{You5}#7eiCgE{ipA^?<KQxdf_p;2og=0#;s<~O#!aj8&^Os7gFX1pw*z@AX#YT;
zEcAmjW%@@pgpsd2PcN>-MyRq{{&>bCXN6+|D)(XOUDn-ta$FwG7d^b$UY(}o3u4RR
z@%F~|x6S+@45lCk;&(?(YfBs2YpcGV-Likh9Fon24n=Yu;W5-o-^MGAYBn0G*vZ)_
zHF2916^b^ATFJ0SVd2q~P<`{ZM&aeTmKCSHk=UrZpNl-w#AQ?q0ioU?anTU-%fk`D
zY9z+9ZU0K^)$#e|zT+wmix*X4zL(<4oHPw?=1j1E=qO)`fOvG=YY3CoHN!?uvBLBC
zPzym)x|oG=)gg3OFA&d&-;zlDGsG}-BxNx;#U7tWmYAD4hU9uabmMTh@3bBEaPz%X
z7t0Mv^psU>F2<dh&nT%^ANjaGTxU(vk*da#+^$LBtg5Md$FB&AOvcjGM}i{sy?yha
zF9dao@}uL$zoHe`WKu^fjwc??TGe;#q;M0fKD#WwyZ!kdVH;wSx)1mKTM);DAnzt`
z9T>=$s5zehs~tcPFDJStB(Z2rnv3_0w<Ci+E$$CrIdp0OBX&;-_PpwHm1EV}l+vcr
zo-~-k?~zzJE^WCX_{XEaMuIa*#>lL$RVvgXS~aXZO(X!6(&=7?m4ZU^dm`X)*>=vu
z0i}=dq}%IDcQaCDC&|w!D5r3yYR!A0eb4^oG*2uY&!P8~3M}&b7+)OS+Q|p_$wDH`
zyuD8>fG3=2rM$|6<lD`_HnZNrAKt(32@*n1#vwKGc|b16?dE$fhuJeV$_f`u7;pUG
z+(#;Q$3F=1mt+a^ulwv~`<5I1XU0NAkMBj+II&IgU+{(^DERaPU=HhPnmXGF`{tEn
zIByior=frrd`!@}6aJ-yzig?Z*H}_6+|SSdv?H9h3+PK3m|_CG@@PJ+jmm0nA~XM@
zZIkDd79kKW4%f`>&E?62zqVk^mP{AItaMKIa(J$W^#h9TC*i)7Mpy<1SB#jvPXR@p
z?InW!@|mN)dp=xDj%hexxozf|!bTL!#HlXth<P@{%F#{k=H`WrQbb*DtZimBF_Q)y
zN$#%csdYJfqCCwcf!<)RfGy;N7C&TmzSAUozphIvZU^F+Trf(S>tiA*)Tk~yMkMeg
zSFt;tMtqR%g-uFtVO~4XU)6K;)Go@KFgD-3abS+N6srb1M}T)R1c~khObQNQDIwOD
zRgeGS2_aF~pbP<7E{$C*uxtBaBC)r}Z6ZL!jDNqwD?7}SbTntS&$Tb#Wl69JLC5!l
z8_i$4|HMvb&%6Sb_B@)bBowF;lO0ZxetPt`EH!vnG?J=X4XyeTqX61;PMdf)%bz-R
zaWi-5JQYV(lUJZ45k?O!u3qmupZ6l8>w3~nss>G+DVFnX3&~GGSb6zy57Hi*m`N-y
zVTRK2!*O&1+hqSWAKg4}RyST>P`=2W8V*qi?KeTlYp%3u@HRclB}wG#>${~4O0YjH
z>mQlD4i?C${kfwnv%M%?pGyTtzF%=Wy{e}S+i}?3bjb#pNqb{hnw^@p>84J~J@WhA
z1-fgW$mxFZru{Piyt+PBTn-pbL8Cj$mfvz8B15~U;WDUmqObcvX{y8e;(16>`x%u5
zu<)DhO`~S3G_q?V8EojItmWf85^|B5iYBgcSgL*xv8!)!vf8h-kLWog*&^*A+G^{<
zP&(+lo+`dWv*F%JB0@JiO<{lWeDaLf#fp0#SPRQsF59_sLECGXz2B*r-Q#ex<BIgp
z3qLucDY<APkJrNp@zbH*x&8Re!F$QquQDO7YR}<$q`m9yChX3XOO0-zwqz;Q+N&Z?
zI~m4E#LVW*okBT}-IVcrl{%5weCB;=@D{`1g>8mAos6-<ljiK_IIGBIlpBe1;01(P
zRsNsoxUT9yHDUb4$ERkPhs6962wL3no5(M~xR)+d+$N#){Fs$+$wgl7VB{^nrAv*_
z$Q6x^A=XcW#`JbH9$M4LxkdxbaJKRm8>l%)`oE_bJ7ckZGPy>ouh9eqE2o_6k4Xy^
z)~KB|OYx^wmyqdX(v=a@f}K)-x33t?+1g3^XfPn1+<el=jcA+L=QH#~@Qd@J>Zd2g
zj7Wx{m1rfs++B|ed<W%*q2qlM^Iy9vO{Q;vsdY~LAms-Pm+}6D%F=?-G!_l^H5;7B
zM(j@;>`376=wMYyiTJY{0vtTb=!z2dmFiBZH}9QSR-YfO2>%1ecj7lv!Iln+T}6O$
zTA`RY2pVDlO@YTNJ#A4?f|7Z!a7Yq`<#A3au5`<_5ShYLKy+zWbzF^0v%)`u3cbd#
z+br``7ff5+Tep??Y0<v7^&DeMKM1&<v|J>yqi}0XSfSOS!DuxGrMTZm4CjCn!Hu`x
ztl*8;gPl<CxyWUp_I)n)8@JMz21hY{I~Mg>&3#ofn!6(%a{gN4W`aWm6(*UvW_L#s
zUBKJBQQe+Sj7jVkb~+(cNYC%cq7y;aN2&Ku(=j95t>DXC2#-+yu(QE#C@P4gGm9V-
z#2>UX$tQIG4YQZ%Q{tmw;%;2khlkUzpoc3;;xDxl%2K7lpK<XrTr%Lb&iN&??zmp7
ziu);;4O6tskycQ(GWWhS+)7Sc;B_i@e<j^(WsOVIYR+EC`Zz@KYKxvB_{V91XK}fJ
z(Kv15mlb9|7=mmN{|^H8|2Sfv%r#LeVnZwBCJj=EZZ*(n?h~EM6sY)ubtU!M@C(X^
z<g~u{_6{eeUMo0xt7YO&F0J6hCQuc?)%AhKZ+gY_-2V}0xdaPeDlI;G!;+`luY)f>
zWo<@FS87$GlJM1e2_%K6iPXDBeZKJXD1)5k&YDXvchNY#ou$3Xxn5FAfnQ!<-u3e%
zmwT>zvUX-9I6bPV(HotzC9Iid8vP`xerOrKe{gRz?ow^9@XGpB&|6>+c7y1J-Fd^4
zNZL6s8v2#Wb@N9cu-7fzFbQ0xgv~w_^nMIpaWE@^%Izd0*KNN-4!FyUMHL`p0i%<5
zOAy^G@{ZkipIwOX8rZ*agPJl+z5fAJ5bT^DtdN=&+?N(^HrzRw8~h1i6*Ae1);Yny
z8x1n3w=vMlS@Kwu3`yvr%Lr&a<w#?n0VnQhu8ekQ8wfh4EKdU>#^njwseV#B>lcI*
zWnK)o6>*i}>6FYYFnW5>*V`pq#F1!Cduk$uXYlH)uFZv}e3r&p;@XL~)&9t-j5I~U
z&@fuiD?ehHM$UK31yFatvT&#EkCy{4G-)y;iG&iYqwNY0fV_Rc6hg226dUb~u-KyO
zVe^I5W?xExhJs8G@tGicdkSf^{l>y=Y04YoDu)ihAK4YPiK?+#p52OZIFd5IERo%~
z*J9R|A_e^{HYaAXmFN>|7^QG!@ibF8el(U^3zcW5V{aW-JFT0bL@vzhDXbQ@g8h;{
zpC-?d{<1q~vX;ZN;jXoDbHj$%2&mtS<ms3U=r<U+$rMX%mp(bf_)iYsf6D$nnH?`{
zVBHaN9r@x~%r^GyrK;;uy6Ydm*iVRS2e7&yD^qcNMCXZe5r=?L%VkUQN)Bv(mQ}t@
z^<aI42yeVO9D{HDhC&r5MRr`*H0Hg#+B|oNifCrawPISeTx$k6reBdYsZ7K}eVsuL
zepL3S>GbxDlG>vTWd;1<2QSnxM-iqPPu6w(PgL6HCW((dKlW7If(CQd8uwP4=q_cy
z`~iIG+;1pi=n}rv-uS$)t+?uj?7YT9klL{ojhIVkmM3<Q72FIw6vYGokv5B-g(w9?
zA!Y{S7Af0KU0(YjZ`$Adz1TNs73Z&1KPnYRV?l9O;u}3(4mVD~mOzY#`yS|kcTW~0
zk*{5TM@|n9bK~?*dz2iI#)CQ!tEIbEJh@aT6me;m8cA+Ewmo98yF9m_6#zo3cf4(L
z!s4ln_oyq${un*R?0Ffp`O=lPaZtef<f$s#v6xs|M}AMj&c^7G!N=q-vp^Kupp5AB
z`ClvEMNkf2M-~Y`z2qJr7)8)0&D|lxaF8{;3WhIuNXD5&qOML=;4eqKo?S0$HoNvG
zjk<1cBsY46FnI$q{y)0@Dk`q9S=WXGA-KD{y9Egrg1ft0@J55Xdjr9OyK8WF4;CP}
zHUxL~Z{}TV?XlNC_CH1+^;!3v^;A7obzgUYOg<!)#`w4d0XhKzzmb}hN?2xKgoXpa
zt-uHGh3DpUG=txnZ%9O5zG|<sb73#|*AnphG+^=mg74Ls6?2@JbG-gR8tCWpWhOO1
z20n{4s=`ugs}&_2!*PdO)}~@+(N45N<DJ^O83uA`4ZZV-HNkEy8^5x}=rM_yfg;u4
zJ0Efxny{uo$@wyOW48@V-nabEsaISd-ih-YC3hA`q6xlf_nZ&B0<W9L|7s)z;>`a)
zg;;SC>D@u((ALDahw((eo4&GHNZNnu62AY4OR6351MaKo(E5+G3_&#in0u`C#aqJS
zLH~#!-o<8rUYF9SK*9p&{6WCv$F<3jd_=ofoi|^ADf>YZeN$PiI!)KX;j^b4|0NM(
zg68+Qew^Xj?j7>&C<uljepnX2KrM&RyCcQQ!DcBf$Q2dSh0It<TDQM&G2K*84(40P
z#$b22mYLBoQUcrdXnYBfDXCg4S2|lPuBy`ZMKB6{CC&J>f^IBpcIf5T1qs~k+lkO%
zghS~1v-_e|*tV(`n=iscM2x+Db2Bg2zZ*wo9mJFB>N$i|)UnzN`6z(~{-_CHWD^Cp
z%+4$tqz44!JUMo3!++s=&*88@9J9tr`tlyuEO!GvMq7Z;U;ZAe53%3Gy1ASan2v#6
zXy)}L)`H8(bvF&F=T}wuwOp)eHq_S{+)a5^CZii$Gp(BH7G}bJdm&v`Fptm!@v8!~
zgXX9hDiCL^i{NEg`u-GHo-3H*{)HhO4qs8+3_hVd?+bg6&mn^D`!#$!T_*vboemMr
zuaRiVr(5ZrGaKof8yl%%C){1la&Y_ZYn2f-WN23g8hnA@#fKY=Cu!Cr7&y3u(E|Dk
z158l^iT;Wkk37@Ir0ii0u3-l*2#6>Ry|1|4PB=r|JopHAG1eCB;oDEqapI1B<{&c|
zv-w>c-I+KCw!~CopMe>9e2A$_YT9y0Q=eCCo7TjLWpl|=68%vDZK$+W#lznPwuzKU
zO4OdKk~L$_<vn`ryjv{iX!fwnUPBGh5XuVXh_zB1@3R%J<kF5xHWLkZ)XN$uR>Wgk
z<SLH@)suI;b~#!-AOwj_#(ZGv!TGeT4Ryc8Kyl0ae|0J73Jak3e5bCK)%*{ZrViL5
zO9Wn2=KVa@S65?xa9J9@OV2)S(2D}L2=5kl_MSb~_oNoFxHYJHAg!7X{#sGb`sp`9
zf0JJ+sI30X&wsNLV<A~PI{B^QC3KQ&Un)9l%JJ9Ivn|i&V1GTCeEs0I<IfmuAf1(0
zs)W)*>r<*aeiOHDEac&?%-zeN-C^`?lK)J(@jPjQr2OL8N86tO*4t5ga%aR1C}ulx
z+M_VfYv081&v<dgo?%T!!l0lf#n>~ikbJ{UBiKx=*<l}0us#!k1$@qi`wZlieVY(@
zZoA3~5AJ@}P3Bd)K}jK{6DyQ9&lfD~#zy@?OyW-6lsCgbJVc6DOO38fXU?n?4yA%l
z@|L}&cyuLp+kr6?(|y;)e@i7SaF2s~Vi<+y<{zO2F&gH=nA+;iLfgDmXg0~;ppH2w
z2{-8~;P~Z`6f~5eH*EZ?38BUMNTJ1Z1$(va#cj3m%1{vf8l4o?4;c(A6ifF$r@x+}
zW_x#}`2)Q_*cySWmd}$zlUP5JOw}Jdek~$C6MX>4sKzr;=IVucrCsRrDYD!Gg#Wg1
z7+_ry)g9+G7j&YbL3AgFHc}R51bxnZj(Q{S0sS|!B>ZBWp*K^PNE1|RY^W8^T{RDd
zZ}|I@-^vZ}%uSyRY8NxD8rqas4#hR1-|xoqB+WH%x*A|z+7iaAX@Dy)y$Vx#$g5gV
zrJ~B~+H0YmmMj1gPS8YnRIQs5h9(W}_H3!(FO<<CI6b9Jg*WVuT0b<>H$PrlAIwnO
z;tyNt2PQUtkZ<*0IE>wLKbz85CmvOlNaapl17dJ0=$EgdX8S=_KqCznA{;wdXyMf2
z=?Soqegv$D;CVb#%4~8W-1%lvqqh=)ed_U)eA<=l2`-DLD4r=c>A%{^4fge5-7^c4
zRBZpefK&_KpR)DcQfa`;Rek9zlO-OhO1xI5Xs#rr(fC$s@3yIKPt?V)%(&V1VPP_*
z1rg>=^WD{<bzY0;^&}WF;LpNHNTOP7P%FD+c2;8^R|@6HoK{|<)m8PJLb#y@ozgN}
zzAF8auwu^UW-6=UR3V7P8}?~{Eqr3E)h1)arM!hpa9rRaCi)$}4jq1zmzUz?0tQ4F
z+*aknA9}XM%g)7xb#ZY4v;#w?S)UDL6ck>fMPO`Ov3ahM`_*o^{^1CKH9)MspdqyA
zz2Sj&bT30S9c#fH>$mJFj~&3r<|cJumgL2=D|}BuK_LrPVSU!=u<|*R%<#xk9!vB)
ziW(O8o9b&iiZvUf==Xp0uGb<fBCn{Xz!HEr@N99R*u2-MiIR?ifwA}bdsAW7!>XFc
zn#xjFVV@+OH{WWk5><9gbQ0YBk%`Y)427R_^a&o1$J*f%Jg()COLJ}O@o8ofEN_mS
z=eDB^e>r=MR?Es7W*6Vj$fhx_u+bM9wXs{sXs#uP(-jOdUM_K`3B4qOd7L&-sQghF
zFa{(M`y$AIoW=N8x=M@4CdjAh4eO(l^1MqjGnE2Nsirp<quyA&K}KX!6!rk#psD36
zYUPwRm@UpPlxExgUyB;+i4|y3BY=;6kljIDNNPL^y|Eb%1BERA7KJ_J7n$3viD`jp
z$H)ZMhvlIP4P7ZDWbh>kxg$UR%LmSIL~^BNK9zUZJc1Y(^oG&CII~!<F`?~yrWRwC
z+EhJlG`>jr&INUE*P@w>Q++?$RMkH&<W)DK-e{_LlcXQVLMKkt@-f$bs;os&{scpq
z2_3>(eSMhdB16{Ps?2oLqyy8C0mEH|=*CA~e^QBZ+eDjLDKlu6S7ZEZXcYf|3j>RH
zHMq8ePrDFV+p5GZ8h@j_udKFdTi%|_IV!`kwwB;x=^H%1VPL>Z$%TO_t8nzMfsRe?
zXQv~MjxLH1O1?n4^}(pQ4I*Yh=QDMRIW!<{qK6fe2xpGz&3)j&r(gCr<OZeT1$SNL
zPy08$+p&XE2y7Bo>vuevGX_?c;e9PbbI*RU2TnOYw|~~pAxPNg<$@0W8wN299A0=q
zGt-=QVg}3h-;MI03us4>NJXK0=EJntkHzaM2R{EKOq%1)*Y9a5UJ85E3@;@?0ogY!
z&X_yp>KL~zovJRIy1?tNn7BCmRmri?it9BmWhno)>RlGIxS!Hdfd)4Ina~BT{-WxC
z{JxOTVlBg+ABUIMRwuF4C1H1lj?Ud=ZqjQh@%!7!mt?p-jWxCKE3ebC1(99NBhThN
zHdfbyQ_qnPQ=eM8^!v_rln~PE>gxW*k`)mqxBJ<;eb84nJRpQb)y)R0+DwWhHfHa>
zL1fz(KFrOMBqz(YUjWp~+k%&a0^t&mwhQ&JUfI4VF3U;c7-D`^U-;I$Bujk3-z-Gn
zPplmqIx{9#Z`l`r&bEYtzB)RxkByBb37ib+UBAR--)NfXT2q|~=Gr`|{P5@C;P8h^
zU(Y{tzpiXXoLwd$a<-bRHfU#kHxVcP^wPT5so}@4_9$$xudEypA0O``W@&AvlyOCR
z)yu`knwt4u2JEBRf1%C>zvoOa1<$xj^c!E6VS`cS<JsiNg7%`sbKTw!^9w;m&ri4+
z?ZpsKnx<!XrppI;R1)=mxTWeOcD|b5gD-D|qLb|W;WfFeXcb`pShzkij5OjfQIWM=
zFLpSW#<_37*wZM$q-2jgu-L?d^BUoo03VqiA~3#QBNNGdMlj0K4=s`s9?XA_#Nj?g
zEH9THg|__`;QoQu2<(T#CfOBH(w{v<)eGJV%tG5fJ#jZ11Ddx^5bBr<5+-E<Racmy
zu?Gb}51O@dirEa_UOL}9m%XmIH&}hTw_4!-wITpFu7+zmpHOuFde2xnh1`>pNdiBC
zB5G#{gNX!5&~Jcy8dr5}xx`nTE2OAkMGB4mKyS+`QB@>KUP8e0rC?EhDQ-8IsjhsT
zO)Qp0kB?VXy?S0yM&&fs%0xJFjZZ$&&78Dhc?FR67vqldtIbCh{^DxcD=EUieVN7o
z`4<ZL?e_gMKRcq)@E+v*A~AEM{4jLj@Q)RxtrvT`i}hmljy-Seni_$aR6p6?`i(PS
zLUG*t-so&73#j`3Cd;Iq>rhJiVuC54Q(o}Iw;o$nS`<H5Ze&`0NTGt^r1Qay)=M3R
zbg^MwguFY57C9Lv>SDMz=rNJs@PWSyRx&m*hb=z&H9(R>2l)i`YhyC1_lEfP6Ci1v
zf*<<GWp;jk_+C3XJ~a2ch4f@iUtPXUVd@{sl5&iPk5%-gjk$&El(irUwF->@mE6qF
zr`1Z5t-OcJt(Q#Yr*-sgXGeK`LtV+gS>uoD?I%iYY=+zdfm<ULyG-t7+*JNW&mIOq
zVS$3Ah;Q}cs4S+cMM^^HvgQwvDumCKxZ!z|)>Jxd8(}R6Z-@)IYQ70y2@E`aG7`KR
zNkR&B%d5Jc)_D~+LM8g0qv^0E)CDNE$Jv175>y-DLeKb?o|BlE@0EJ5^)fOd{ju<3
zvp1NYg+=3v@&+x)axnKYQv`p0{<1$YyPB}`nF+r?A}&sIfuEoMu*Njo&lO*g5QU+j
z_Q-p=O0z4+AIEg?hk$!Ew1OY}1L#DGiHS4Wui<Sn%I%CG%-zT1#U=NkEPLzVh)Fa}
zb6MK{GYgOTtJiY;UdQnB5j=MzP*)l{;1z4U(a-R-#aYes?4ap5JO5lNc;MR30!`Ug
zHPivcW{L)5%@4RwY}YU2G%&)fvZllc>Cdp%(>qVQ0(58YGT883$2dD**kKgS$fQ)@
zwHYn$@`w?D3ZmIrn(VBuLwXH+KDVsNbW>m{w}hxf!MI*ywJD0Im5r^+x(wLC@aVvN
z!wq`?_rHUhfFfo*S|Jl#xa)=Q_QU@x&=W8b8+@TWLgn%CuR;g$9bq=IH|!mpNQ9pI
zdYF^mkePW8eatzI)ipun40J3OH0^e=!gr_floiL}+jd8z>6h&{6G4nym~13}Shf%m
zuok7!PzL7unVCNrj49H$y7Qpz=E$!}_JVq(^jKOY!H50vz?Bj;nFLr`Ebno`9m|NG
z-m%jpdah_Fr|31M36#tzj8)!`r`qFlt}Y03XVkLT4HC0)qMp>j>9+Z(w;b)6``jq1
z?>JqUmnfxwKS<l8PINKeu0QEE?>#I;##@YS*54)goa6zU1M}DP__xA-w;d};`@!<s
z;E@ERiyO5fyh3yLo!!QS5_Td^EQde^+6_nrwY<^+PKzDzG1w8r^?Za_B9Us*hw?gg
zNX<xg>sL3VtKD8Aq_11#q<Sv!Q;11jKeyo5w*{r3=F3*%%>1Sin4_ip4_lD!7PErw
z8A5IRzRK%z+*;s5tV}+{CCyQMYA2|9AeP9{*}8Amxv(j<^AG%WWa(I!W!2IB0P}O~
zaK@NUzFt6otei2r0V!Nx>=?w%UU)mIzWdm>5o)A*+w`4pa2vB2DqX)U`FvyjPY|sf
z7wTUP<3E3}nHN2=k($W{h`T+0Aqj^q1xnR%@4#O&W)$YJoMt61Tr)k4bQ){i7iR(y
z)xGxYt0*t8-aYl`u1|Il=2>zz^)ITntOVltmb?AC56Nem3hW!#+RVfrA80IEzJdYL
z+<&5r{O|7W)-aj>KtGyD%u#ha>f5!_TZsSbp1<Tqm+!@X0hp4V9bG~~qC8MyKKV-p
z>#t{Wr-bYbP#Z$`%ZbCqKce9Mgo9tON`d57G+UQ)zy5*YN7Gd@Mu&N)!%`E4U7?_h
z<E46ymh9Y7%j+Wok*TZ6sAeOQV)quy$xIb?`=Gx@h_i>*T)wBC6VKaZ<brRe<tK$2
za#DAI?6Ydw@Uer6C|l4|^(6q1{?&PutX%0b5xsjhBRX;c*rP^`?S5{pAt=V-++he}
zFtUx|Mf7`KqtR2HIV#Fm-NgAeMslx3tE|&z7#$6Wm+<P0(@D-0%7sl(D$}y=py?!c
zdJB(3Ul=9m<PZwS(GOOewQ`HZx1*hCQ~7yCQ2Z9d=O)kGtJsbLlGQ|k2}j{b%&0?-
zffHMPF983CJj{+XgotPSCVd9K(@QYyjT^b27r8g$X8Uxw7r(;<KkOdC{4uUKB$jvU
z*4`bG5KnimdR^hccJTg}AUoc;BJ*&0r9&ORJ5r(xR)Q+?rglc0sx1QH3ZZebhci4i
zMOuYBp<Y(RW#&NwRbXPdH(4ts6I6Vst~{2>?sHU5*hKPL8Dd8nNO_O3X4?=Nv$@j2
zR7nZPTh3bLElEL#E~}wnwORIkS1SpftZpqmGs^p~AS}m~DB;;)$Co+jXE(*d&wB8W
z@<_cJT&B}nldXw1-3J`|cP4_5^05FrI%v+6H~6cB*=irm>V*a+>8l@awx0jH?v?Z?
zo!*jKK}I55Or%<Se~x@LCP(rLR60am$F`q1jS5vZ+T{`f@fB4+@%}7=`fF43Fx{8M
z$08<7M8C-~{VrcV?=3dlr*nKD!)*5B>SK3&RJ_h;lE`}VyyDnFvTD*U$|ASc6ziTR
z<1QqY<EJsU5xNs5O(u!G%F%%B;n-k)ZD78Vzl`$SkN0TePcQV*{qS`~nZeqoDQEDY
z4^v!XsGy}>b8J+Hd(?YQI3Ja0`9$+#*p}+;9#w4;>0}F7i?5uzx?8%^E1gnkRkp$l
zqJ2Y`pEjq9eD%F;&~vww7U?cT2C|Tml>GGb2NNhNanRDw4;y|@EM1R>Oy8(ISoK@b
z*}9)OLG33z-zH$wUKeZMdGpv15fNPgrmSTYRUs2NOq#|whx6qD-T(Nx=e=<MIJon+
zJ?kou3)=)K!g~n|f_re_vt1!ZQMH;+<K4J*&_^CtyxC5(Cf_s#UyjZj4%r`|MyIBD
z_pDxHo<9;@j`%>x$QGkdU&M<4D@3LxyUoRxqa8hu7wcTaTm;8EHGy_WJLZ7ETK0l2
zLb})FHyN48@{PwI&1+@Llc{oZvdPdPFa%%X#K<GP&+EctykGbiv8X_E?8|O}%&jHl
za)SW2^sFdvs8)v`@1qyP!)g|3AfhMz*v}zg0apFgb97w*-h9N<!sUHjDv6Xfh`dYY
z4N4A`wSVru%tPqfbG2(17nraIi$blwIVJ7GBj4Z{h(HJXz|g;!+{z|F+nkT!g^?BM
z%Ur9kJMSU(ak66YiF=MvuqLXb@`n3z)ep#RQQPGboobKY#c2}vMvh@g*NlBuRA9Y4
zLR4PWrtDFR{bWm$PE0&zgY(&(tzc1lQ>-i#i`FCO_WJ<+zWG`0dHCTs*VYzxu)^p5
zyw4Jql&@A)e`_>>4;7geSLbD^f$?`)BAEk@cj|VQi|I~QQqci4dt}*yYN4BRT-sSl
z(fpz%=P6Svf{(Kl^4fK93llNDH}!6ac43?8O9}VjAEGVkx%^z13ku*N4KAwl6&;g(
zE&<}HS6BQT@vo*Xvci;c$&`=g45pmXbpl$SIlE!Ql@_)<WZwTCPmkB5P}Z4``mM`U
zPH0zZ>y}$5(S@fO{+zvB)KULSI(;(Px&U8VMa@lB%zrmb?DsLYfDX*$*da+;_X+$E
z8~orvGTgb>XV~>yk+^m(Nt2yzm|w_aU&>mnzx`IkOk>P}ab8v)BTC){pO*y~WJm7<
z{7Zi)quv_EV$3F}2u8MhOYCg9>?Ri+oz$H&A3THKU+Edq1wun}D=V?xWZ0TnO$<z1
zjlm6sl#^bChd(Y9)sXXq7Eia&xTKm6_|uu$vW3!h6&~dBs_q5)HY}_2KKDR`mM80_
z>IHbNuQw7dN&c~<fh9>Jj9Xd~eoB#JX!~Q6ZDpOlS->#%0`U9UKCGx2B73`+`dIdZ
zi~v{Gz{F!N^hYPuKeDBedZxEjz%B<pJv`u4TBFLd;|dOZ)xOA#{mOV|JrBALSLlw8
zhO(YEcZh>kc+r2F#fexDbOQp$JFG1Eu2b~;Y2*Igp}Uun5yBUNmqe7Z=S|O`uTng7
z`87gkH*R^cpMFxD9Z-!69~IqQ9*`_m6nJX|0h#3~z0Uy*tKj$ciI7Hg{?)ZhpSAU!
z<Bq{Gg=zo;3P=Eq$X$dn1>61^3o=WK0xIB5^rNY^3$RW0woADpo6_(qobjz*P^!*2
zQJ}YF5YzO<Rb@n_>axOvcImJCD;}U5v)6gwd|^CL(Bk_<lC~oPRp1k5I#w(Lz^buh
zmN^OfpCyfJ&)AVMr-_?PxQCt;`pg2AwvYNN>iE5xD5u0JTFrTHwT^p7%gyMkS;aNf
zXb7urtiCnC!#e`AWAW`_lVa#UrqEClVxBA<JV_7aIY>pSJF%qFWmwL^@$!7CJ-<AY
zSq-nco+r_BS6XsWo00N}F6hoa$&nZRru}8<<Y4D5%D&*p`oQYhANLWpv+;@qQ}i}W
z*A8`#M^{;SDWssp7CluZj74g4v%OHB8D1!k_Zhss?(m)Can|bNb;DE#eat-{hCAy@
zczE}6C;}j(EMJ<rpNKKm7vMr(z~U|e)FT9W_>J6(xvV(C;fRh6g@}zT6RN_yFVJ?l
ztc=tb>g19c^vU;cp+Du>H_A@AM|}(Qqfu7N!?4=t(`;okF5PwB{Tz&|O%h~A;SWnj
z_y8}c6CM9u<DQ3uB)Xrg4x-IZSP8uq!tV)I&MYN8WEm@1M1kIq=VTiCmild~h3Jr_
zWB8tvg)v9tQ=z=vvJS8QCF^URxR_o0Y!ugr)(ChsQ47yV@nG_?Y|Y1yzstsi#tBXq
zdIf6C{{_J%jKT}bj;|k^XZ6*1!-i2*lZI68!BX)_HH!bJ$k_$jG|(~d1J<eVLO#TV
z8sK^Og2^y>nieB=(dQQoy%hor_lzDjAVXE_TOOL^!j=wyeec%7Eyh#I3>DVR?w8rX
z$Qe?PyIe0ESL(u+XAD|dH<<;aPi>c8L!sr68Jv?55l8`K;LZ_%mME;h@prE;Xhpdt
zfgFgwT3Q;tOJDcv1m1E3gu^TlsgVD^HP>sMMDSm><hQuf|9p97%XlI)z4Pq)N)h9>
z<FbBMs_*sKN$mA1bP2M~R6~FpLRa)W1vyq!Vd(Qj(%~Ibl&b7zzr#r;FZ8+?V8KJ>
z#6{b-$#T=S*KAm3w))9aFu5G@tBrU7mu@Tky78Xf6P6Uo4;JgW#CRku0XjhpSU08*
zgN;#UYF}@&M?6JoBXSt*h-)s`=o1ql>z|J8cj|mYY3cQ6mwb%vgRGGD?|EyM`?6L+
zi~n)UNPpXp=f+4PaeFv%lqH032kky9ef%}QhEx`g$~2;8ZLE8rLFI<dn|bAt^7#-W
zES-yCbIY!|&#g(rxKd+X?rG!<#$L=@wktc?YXIo|RBcaDO0}<2xpE@&JFj4*e(z?w
z6#l{r`68WgesWo<@yw%}Rchr--rg6qbL>^Qp%!4h61Sb;fXVR)fd|l{>hCzAd6{FR
zx<63hvx}hj<ArS9COdi9PWV->Izx*L?Lc;OR6j6698W5(u5m_=1%!|n{94{GD6Ivi
z{vj2&u@*S2VsUKv7)hcx2z#vZZLBp3AqC5Vsz839zsQT;foN-4s5z<0OsVLwDye~w
znk7@f=GXWQz00;@AVoJ<i2Y2x9lZ;DV-K+wsishbV!<lGo+Q{*ARNEmU8{O(%=d$z
zYt8e8d*aqn!D1JdY^}MD<|~0nB-2&-1jPgMSc9W<wR3;o94Znb5)wKNN$BCBY$Vy%
z_h|lu>sd|YVUj-<)l)(WI;i(2W>Qe1-|bmn6bijwVXaT-GnsU(v6iJB9JZIbiVPWv
zUK9P-&FH2rb9@1)y9MsQzL++;5Roc2sScF0i%Z?t3Bv1xS-9C^1vUS*ZQqkWdJpp9
zyE^Bd=HUG`_l?+^?$<|RE4RXBp(z~<9frkVnca9D)b?Fd5;EUwj4%|U;-T&%ieOiz
z{lZIqEpuT=9aDM#tjWIP4HwU$H5n^ue)zI=_j1ndk*{x&6>z%m`3(8U$;qijL(R|4
zW!AIJuNXcqU#%zSX7dyR_jjq+2uE&cy!WPS(?F6c=syI;L%G0ssH^x812Hcz&xRVd
zTcANGXQp`FZGMV1Rrf;=bH3uoGgZR-+pm0jM>Ahj9Y`SLv342s&F-5+XFP#?0-}nm
z&f4bfpVit8y1JoV0GM~0p{FLWqkMhHO^rQZ)*pGJ-QmEo9_^1dB`F*Z*H79<66kI^
zmZHZ#<BmC0O%z@CYUcPh>3yzwO_Y|}0WU(V7d_p<UZ_qoFmtF*jYDgE5~$I_wr_~M
z%q{-eg*>c^B*^-LjIxy+Bjqn`3$v+6dy-KgX@<hP(bE8pcOh4A$G4hdyO|esEy_#%
z)_WZSlPJklQzy=o1;NHF2iAB`^x8Yb8b$KWJ75LEZ>K&skPeLsiGbgXcdg46iwdi)
znVL=;<We9u$_*+g`RI4i3X><>!!*mZo_Ss!Fx)4&bc9DNrOY+Mft5CP%l?GpVrmys
zIYJ&=Pv(LxHfY6_@<StFh+}Y%g%RkEgAEo-WG$AiTVb%Fs!W?CCorC>XfeYm_<H*j
zHun7HXVAyjjuT+8Ip|B@`c1l&!_`d^`=3UG$@UIn&hSP~zN`HE&u>bndTHIAwuzw@
zjMqsfZ!(=QzA3GVn`Gzw9>b*FP-`{DTPJ#eLZ#{nYWdDN$mx)J7m?MXfF$}|d6-mg
zewM+6uW-FVdR^ya2=Qk_YLE|urCXqC*D%FxfkNw<WkrYrk=}qvj^`gy9Lv^I>4QJC
z=+@sa9h!AElP+f`MO3BNn<JM)IjdNEC`}f#SZZ~2mD5tg_amg$v2pro9+Vb$#(HrX
zL~t3Tp2b4y6~+>GL^ievz)S|#XiZM{Qol+|>F56Z$$mPSk!n-Bk(*oTE*m<7UEvEy
zCHbPNn)q1NMa*N)rz=!*!~Z5->w2IH*k079<vt+g=fz2(f{jIY`h;@)U-`fOO3V9e
zC4BXZ%LbJ$eDOC6QqgS*5<R%BZ~xpit<nk^0{H8x_>)usJ}gt8F8jT8TZ2w+PgX82
zhRKX_ul`Q-Co!n88~41MAD7iOnWFU`_BfV)sAQ5AdA7~IJtBY&%$RDSWDEb`62Woi
z`8dBkbCaC=sT06}PuuzCjAggf*0SGdi5Pl6-2ubdg_WDV-<9^I;}lxozM2Kxd|s)g
z>i&22m=joAUr=98MB;J8u?}iy&B->|r#ka~>Yhta$N*nw`gm&LI@fc>e`oN_a3dyg
z&Pq@Vi0R<s`#}Tx>5`C~5&Ub_)qHn^S9&`sBOy5RTHn0b5+eR6b1b(Jo~Aaln9zjs
zO2=b7pTJh<xFPt=ae?N+3OH@Gz9`0^;0GrNQ*r#UL<Zev#lXFs_H#igZ8yfUZFN>T
znE0I|`(VA$@X*>z4g}I43)G<<cv|x%q{bzw!C^GHGi9&cLQ6g<DkDA18@glRN=}HG
z+Awo+DN1J+o?2Q&=eX8=E6!t_EQ$X?Gj<*M)uMz-K<ij>B0+Fu?6aiiC(io8zxh9O
z1fyaLQOzyuN!gVqnPzhocz=nkk55X0Kv0*zx#Imek)G79>RkMx^1(Y>j=Hp$oM1{W
zIwb2>PrhWT3#^IwyRQ)wvz)U|^eiMYWzgXaOXmucL*<Kh%qUZRIXpQ%S0bc6YfRhh
zUq$g@ClEy6_)YMe$T6`*;sy+IBsULsF+8p+=;fa{NOSCk-{qVOMMw)rBoz=Ox+hxt
zXgxAe6AOi+`MQ}!eBPaY=uK;(hT-+c3`~f|<S<rb&o;z~T?{sDR@0#$iGNmPLeFR$
zx0K{@=Gx7O{}HT@@I9R>{p1gGRYS5WaGPh<So%%h&oBE!3iK5DRrb-v%D--EG*f>x
z+@~(uN-?{1K!geJAfJwgWNDjK>bdliX`#Uv_GHRuESYbd!yvn4T)@iArub<syZk7*
z`D?_Jnx|@UQH;4|X5!t9Lw3Sf8IjpERwp^jRZb-eg~O}$By5=QP&r@4WN&^<_Zj`J
zjEnfTL?#nCVs+o!sPM8O-UJ~h;$5ffeU)!jSHV*c0y`(eb64eKRf{q4qicAJ49@&^
z`vlo{AHl?J<8gp0kgeuFVTMHT_Nbz-U-{P@bMx`;41846O$r%`?H@~K(xBNgGJy}q
z4$AAgWv=X)>drqGeoGhg*QZG7{6Xe(-rL$8GsCjxQO;|3`5w5goR!`0^W7XRC{=|$
z{X@B{Nr{Mve8{k~!ZPJY{yydcs_R?lEbrPNC^9+$?`67-Jqcc>Uu^UD(^*}<WS`|`
z<fxq#H&>LL`~WEt9acHCq?3sB*sXHCc^Ez2HD#38@OBzBq6nRQc~bGr;n@M>+Ye`d
zTUKp983ARN_qSs<6kOZRgRGSP*rrIFgx~JY)E{3i*X!H9rRri)8$pWv28P7PzY*t~
zK`@9hkh**5?GTl|IQBt>3><$^&m8zDOqfuY2M<P;RNhy#Ls>=LAc49E4HGi2)M1j@
z=Y*v;b;Nk{t_RC>LD*$KJGM-r$1L$dLz-iTckJ~~ZrlVrZd5utUM!xi_u%fST{*g<
zA7C@KeK82@v79jdRQ0(Nw+CUFQ|nZXu&6MysCgD?ZsidB;a0wG-lCwHuC_&bcpi=4
z3;NJ}fxO4apxzi4uX7Jt<QD!iR_eoKrma{IPydRA`Qxv49i9DHfytW(|Il#M3C#(D
zQ=~HsPb7d*9hHtyN^zB2!gI&i4ku(FWZ4>^R^IIz;F5hFq&-zJu}6;Ur_A9K>=Hyg
z^TmsGXp*Gjpe0tRi-eEE%?JxO^LC`?IYsVUt`ay;eslZvVokLE_*H;ZpETlvEIjY+
z8DC#W1mWfHpz#RwWPOa59$4NmJH#7s=R(?`GfCfI^=T9$$oPVZK}u^K5z}8SKE(fJ
ziGx*hEPjzgcbk-E*UC~J8rLpj7&2~pzTb@CaWzcuQ%j4<R>dFi)D<_7ArfLLr45Z)
zU|PpUOWF>h-oU~_&!*paH~OR<>8(G<Iv^E2#=$CCz2+93>OL4vl4K7lGj+GnW1OO4
zE3KQ6ju%vzBa}*ZXjtT=q`(M7Xve_qMw>wxjYw`8j(DPvbXR`{4IHlruWf@;SIyRk
zWIbbXyoZPC)}=dW*f4k35RjF)MaNVGHC8=Sz5jYk-9(+m$S%^n>cG-eBV1gbQhSCc
zk$F0FEOKa$N!fMOHR`4bZZJ44;bJmgtkYO820Pa2RnC=pcn*0C%6T}cW}BK}YW%qJ
zD=E7;<PF`Bc;!gjd+eH|TTWkA$^K!p7}-6QX*xK=`Mh@S<HlfD+JCU0F4csx?g2sh
zKWPw&019k8Ql+T9-&%yA6=wndyI5z26X2YH+?NZ$_)~KaJr&%KO-#<{p%VO6ObWF7
z%jn|0%29)-mV(FbYo+GY_~qkWB!PTe&WO6F?i5Exkuv7O_@cH{QGWJK^FS)RXsYa!
ze5WP72zd-uxSvh;a;&CO_wqz?F3??DH=E+|=G0l|(^qLP@>&KTp4`c*7VD&%C<!QA
zi=Fz98!DhOEbz=t3Ejvx=D9BmX|$WX>>6e0<nU8x->l5Cfr*0(&kmkj^p~~uB8!(f
zXKd!?c*$!zNJkHfwCgx<ZTrwDX}VDd=M^wh3Dq@mwnu(Y!7G7#(BS+BuG4;aP@GGw
zc0PdQk<%hLpv*YN{JABcBC5%PorEDyM}Jq&?e1OD>bz;Y6kbTTS`+xxy^xpYALz>&
z5^CjCD9_kxYQ*h>7t0Ng>CVcll3f>~NOrVooGNN?^&(CfZK5!;asmis>M!=2!s{r$
zy^q5olh82ukWNUbqPG!YbD^d+PjB)uH?NvhLBn)hmtoe~knOJYn=`01ffohT{Ouzq
z+zf2_{Z^+}q!tTg_s^r3=a5f%Q54FN&0zPAy;%KXEv`x#|2<?#2)~OOSLroACI>UF
zps5YUVs~D8qDov@7OD$XX)uR}8nU&Ek=DWo-48?i#1?^58zZr?M7&HG^a_xq_b#nE
zy8PJwh8OnV86#l+v~?;k#CWinr%&+GR#6E%BrF5p%qJQ~wl7vz1Fx@d@?x&+_K>Sg
zhYdAag{_K;F1Ta%hR;wlYNRt9=JbxdpcbA^PKm1Ir}!}o=m+X(@MF5Oy4*trD86A{
zE}L8Sk;bklQSnAtrN7v;#D(YEO0r2UG81pu$<ZM-Slsq79l6ROu{Lc1a|5<3xTyxW
zJ?f7N4T>sTn#f0m?YOP^KloGBzdp$rRp~uUej1JbbR&N$Io8+O!qDvFmv?BXOOR?)
zhqtjo{w?@`HNmmG5Sd$+z#cnePOfVqGK<hAgSt9-WP&g5zLtE)VmG%f>cAfRZTGT6
z-Oea7jP?(ijAg9KX~%{q4b#=<H(@Nz7SchsMkm3SbQ^kq-bJ3>Gp?0N{qf+t>nk}6
zF^9M81wm&UJok8Z(W}cSg!FFmws`A%NQ)1yGLNoO)~8T~44$nNJ&=k}v|UJD{O_vU
z3aA9YZDu&Xer5<^gqZW`Z-xV7cedEqMzX29t9=6294z?<#i|%GrX%2v$EN))1%%M9
zPJ&0;IVDiwTJy5+a#qV<678b!gY~pv)Wha&xytR#)30dXsGcJEVbr;gjPozU+j(w|
zXZuZ87v>yb_~*3}$TJs>QoC+VWaFL<>a3}O*|kYpJ?x>`fXBmozm>^R7Z~hveQb&}
zd4cK96OQr_>It-J#B{6)GBfw~w#oYKYNMBziEHIZ9lL(Woi+>b`TtIP?|$mrP!%%8
zZ=>F|Ql`FIR?xA+Ak{T{GUW(S2TksRLwY@fZBX;ZS<hT4>wgm1-ePYKI*47w8IgDK
zqldE)n4Bx<f!R>m`Wu91&)W+D7|%6Q3){J{{JJaF5Q2W(1zG3{FjXz-%jxs3J2RW<
zD!RX2B($4cjEZ=cCZ-n-_;o0+&YfH^IGJoTl}Z+#M6(<nJyANSokp9QB(tCsS+E#n
z5$i%pqu>^4E8{{C1P(X07(mRV=MVfOmxTzf@OCyKn|!l%i7-mB1uh^wb62ognRk2H
zbXJv=HUoi9TVJ>vBF6+g@%FwpIeDYvb+?*|d^0+Y^aknmPTUjJXhth+#_0Jp<hINN
zH9omGo-J_fH<J_cvdRq**`jTR??bdn-JoZi{;fz$0o<i0#}&&9d|@_Z-d~9Q@gJO8
zoOTTbyZ*wjuXtcb2ZOhUTd)@W*zqo)UsqKNYlj%EalAQvG4VcHs$SN_UeT;rixdym
z=C!4?xIhsuCq$FZ_$7{(GdJ<+I*A`%o$hp6*@HqR7(9QoqV)5C_$kZndOa&m=@_qe
zD)m-ScE>u0+V2%rJQqAIT4&vETp^QGUBfS|)q>;Kr9xUX&-F3Qi_gqg>SY}M8un~`
zz((|tui;S#_odnbLSsIAzYt#PWEn=>V$?+f-!v*u>+qm&ODFheYq@|cm37^UvU+^*
z;<<7XUTTjC9&e&zKM5X7V#n6DDsrml?7t45#oBE56!EC2@x`<6ACO>;LZ`AdS(C!U
z(=UcE+1*pwz*lUcobn{BVEv7dXh*=D(sK}IqMMT0+-}4RPmmFB#N+j0?_y3(sf?nb
zk~teBgv58|B8oZbxtTAbB>X^~?e}PXlecZ;CfI-Dm(z8NtHg`_g4)@MAQBQHtmM9Y
zvWs1uTA}VNJ!-VFK9JOny&Og+I!QrpbqOXSS(;AD0XX&`Iq~ToFyT?SpIgz2vMQSO
za=C;_OFRP6s9l6hfm?e+A%c)4$+n{WS3LR=k?+B^5#;JgS|eN!q-8@aqF^SUPsZP2
zIkhUysamsg8Bx@Eh8nY4bxZ2(Vcr60LE`hc7?&{>UCxbK;T;P!pYt@^lRO;1Dbs0%
z&&Ph7cXZ3s>1qDih*1<<G*UAqr7VfKzTEa6BJzgFTWuIX*Q_u&zEuMW?D6Jso5e>4
z+~qkkhByD(^Fpdg+#xePQs-ax>pgf6QJ4kU(L}QKgU>tY3+H$8hnCGhhd?7l2HPY_
zckFf!;bMmoI+^Ic{zN|DKC%S-Kz~u_9be2_T<nGTT_ffWmA8aV%()S7FVvPeO*}PV
z@b1yIlxip=e1uN%VZ&2UeaEzgVGeIBg=mZ6Rf)f%s%U0wL*UnaLl654rT4}KSYC5u
z4x5S`oIs-DFgN6@#wqRal1w8+^uD?jUwx4fBz>X9Z~#g7KF7;fZWx6O{6gH(k`7*B
z?GD8Bb1$MI`bjZ%KwCLwhyMDJDE)U`tvog^Zn;U$b3yB#1JdBiZLvOmLv;0&U3Z;u
z2G4Q&G;5pik;Yc%(N$4X9)Jt~IROGWI6wUN%QJ-5rm}f2x+N7Rg_0g3d8>RlwtbKd
zx&c#CxXhPp*eKF?bMgxtywONKnY5^%Y^Ze(J`wZ-3(5dYuE2i;01NQ_2a0Q!FjsMM
zji6!Ej(F=@W{7U~O$D^c8bepxhe`uAk!S}qY(IGLbGG&}<BNa(7@3ySt|sBLTbYJR
z#xA|cVhHE$Z3Lm7%!o@MB}m)BZq$UhwJIW3%J2OW(~g}+CsVnW^kuEVdWoiT@|}u{
zEhBnzRuI`OobhY*pa;b<j{G9~2P(hJpafgc&>{-2i4U-5f{Gud;@8hUb0#U+_+ssY
z&4si<tQ#KRVMSaaFClSdJ-1Xovv2n$`a6>yVW%i!)2@(a<adc}ISX*U?x<SPth4}R
zO@-i+)d+Y4R&l1eZ?(+X@Vzs|X~gDcMmMmaWW$(jAmugB!B8i(o;@vP^TsE~H${5y
zxero!=&WA-ASTE<_|Z^E?k4;BFxNRm(YD1>Or0P}=IR`R@UHs&i?e#jxdRs!hvTwa
zN82lVm-L8)hc3r&*&}5@GY4CNrGOgf&j5#8m#~)Ay_Otj2e|YVVWda^wW;Ah_htp)
zum3-SjBZMOn23sY=Nt7ttl@1JeO@gMgHEr;WvQ#11tTpa3&#lhJsHq?2f@*wDm6bH
zPrmN&6?!Z>KgCZ{W%vxb{W^6YqR@LE6ps5WEg*0)WGsh9Ovmd}_pJYgMEbuBQw4Bk
zIuB>mz2Pv$^@)e})rF!pRY;4bWh#>NL>AS-U!~2_Wyi*iu}aC~%SmNX!>M7MwX=gX
zN<tVm>gsu#3B~wgN%`JSr=f(sJj8N^kvXRs3fvLfHN!hLMSYrK&Q)Jk5AC;*!bri3
zYb^P*RWByT@zIIaFxm$X94Zto(6k;L`HBTdAj!ij$>Mr#>CC3l&>&$oIXyQ<9bv*u
zZy~R8dz;?}?D@2v@pUwQ(A%(zWp%3jwh&n`J5cQ>B!KH<KZ|6c_8nReg0TwZo^cT)
zrNKy~w2*SR_^(H;GmWie>L0$Cs+WL24LW{5+Je7_tabBMRv3dc`DE$)FT3w8nCU-Q
z*Z2OAvbe!1E^t#_s3T-Q`H8?zL}Hf4sOyDa|G=P)Cjk6lq#(;n2`z44`Ko{5hE-Vp
z#B(9TJrNAs{JzVlvL5qO5=pmE6qt6y-nZ{1UDs*N?{0s3kH^!-Ci&lfq}sT^QwC%=
zu7H2i)%O2ULIeHo9=@<=CI3e(;g@T*4=FI>r#Y=wGn9?{lLqc<MY!)Nxi8JI(9r|j
z*FB{sx(kiMMB@u{0Ap3Gx7V|`I6OlS(UUa@1oGa<$f)tt@nn|3JT7zyq9$M1Al1zt
zVtZFj&P)};|7L$O%eMzHeo%vE4Z#i+INi0PyB`($7@|=+c<M^t%}$14l1E=tiuw7w
zWDv9@G&Mz70K@IyX)Py*p5h+3V`%FT>4y}AOah@qXZ1aGz?SEVMb93k)Q&+EWQH53
zN!Gvq^96ni-22!tp`H1Y*nzGS9NtGZSU3UU7;?v!w8M|3fKGe=N4_`U?6ha|>{N2&
zj8tkv!y)f%?8pAq1+;$WrtP&}2S&3mX;kfZOPCHAS_1_~M<s<wyj~S3P6`=O5(xE;
z{5Fj%-IUdPF`Q`x?N2b*bLbk8%`<vWskIP;E{|hIE5<h%{z$w%J<KtUn!?f`Lr>6W
z06~~tCW1dAWVRt?*sgwNnl%{MmySy;pcY%&!1>(51OQ$IW7<>cjPF<Bud#AF+-L&<
zotDxHLp(uXg(Z^lB;ruxY>?B!o_8%fACVNIjmpenfI_1A^;fPlZ4eCiE^pYAP5I@v
zxzcrL>}hnue@FU1=xRgsKYzDn;H+l)uWgWCOC{P=5mF&9?$=$T*AM=0cm81imj(a9
zcnYz(Rry;t=XnkTa{uQGsgFswaRR5xOg0y!TGl>yYwq(YRBtEUFTBrdfIWoK{<j5#
zlCW9lkGhXr?Tv@Eoe~=hQ~Q7Zx1vh`H`%WQDiQJBhZbVzhRE1C%Fo1#<6HH_Db^y<
zK|x#*NN5s8-175%9}#)PX^5dw!+$IzWL(Dd>m!H<kskiw5dRav_g)cR8HUe^X^U4c
zBf%&qE}{?LCmV2Lm3fB0bMQxp_zsy&w*e#@2>ta6H{yF9)M5B;QT#?fYROkVf__br
zs63-r3|{6J>^F`xf#;QH^5?H0?8jfI!vRYfPW{IuEf$`{f2gy*fUL6y<hnvnzClMr
zXIScx#$3B35Z|hnx0Yb~M^^d_q`9>g$Eth5jC}ZrP+Fealx2eP?hVteV;m2R_aQyM
zp(e;Vh<0zLJk)R1nC&n(im2+jzjuk}3qJ3fJ7pDca;nN+EDx!rwJ;O-T1w<kU9(?j
zbiD^^)h{`%9QXdSSdQ}z^vLOiAxFiq5MHh?fL<B+Rf_CgkxEjc7Fd;!V5CIq6uicO
z*dYB#xDBRqi2O;lNBAJX>r9pWADI$h`2g&B*AV~r6qhN=Nz4!=(uc-%x3)#<{M>_1
z%^#EEIE;ppTv6qG&y2+K!in5R%9ONDIH<<IlVakGM$r{c#$wQ-gV8BsoSna>^8aJ<
z@xayB5G^!!Smy=>TYb)8@rxq<w{eANYa@(A3Um@TPHt&6dN?O)kH`QV-KDCEs#?VF
zN9ZJ;zj+-u$TlY$HlpN>pDI-Y(V^JWGe|=Eq@A(R6VG%cY;hFTOa94-{$yPa{!U81
zcU*Aoz<y&S1$#i<!3YI1=$#GnbzcBtCc^K{@&o;O;O&V;_d4S{%1WWz_kXe%gO)B<
zIGhJzul2A4j`r*@N>~3088<~NLy_oD#JdUfX51ywVW*;p){Kt(yotb@758@H=1G(V
zKR<k9fsuydt46Na9&h~t+kxQ1NTXfm-SC~>+d`v;4jiK1=K^Oj((ZGyXfcg#5v#@c
zn#_qROJnziG%rhNjM$I)(8@(RyTMcm^x+S!rP=K*mB&GeZ_CxVdCV&{!pokd*+>xn
z>gznzv)?!ebL3<Xol|RzwRPem6XFXD&yzfOK%>6zJcB>4#SL(-_(B_HqdNerDxCH;
zh}ql4;5WJL_YK@YhsYme?$w;~`_H>c-~PvL`v3LR#ql`(#2XH6A#iUUZm!@%9cj-K
zf)u|=7r&N{vm<Z8J7lTof5&sqe;v>Nm$5ZTbYV6Zcagr6hU&xLS~}IuytX2Ifh(~K
zu_0ZbAGaU7KE2X($rywxeHR%dYkEV-p{waZ3)kFgr!q2~k7TRk@ny+Zl**sfK?*-~
zvJF=VC>UK2D*wxf4WCq;QcQW^${FPRrH<B~MOgS5VO{AJD?8GM)E$IZikueG3)K$@
zW3X4-uS^7qZV}fpYco@xF;Yk|TLd43g6f0g^&Q{AO`7+;X~PP`3r6)KnaxOTUOp0R
z;gfrvAO{xy$kf7*@S6?huu_D9odRQQ#`Kj+aJ1YV8F;wQs`%jkhz<~<i4ZXigoHLi
zt4>%oEY7W};f4Q3RL3};D{q0;I-a05XLwB7Q^%!oqO};YS4R8qbe>GA(CC=hQlNJW
zU*Uh)b@7YUbHNU;rLwHQuFp9nlfDnogQRs);eh>{(Wn5`<K`-Sj5c<3d)IO--U4aQ
zb=Gj5%^iWbr6xsD3x>?YjmKq%0X~-qB9`z{C|s97$SCw*G2u9D_T*>TF;NMF3L^Hk
zco}cUQVt3&6#BnsDdYE@pO`FUOq}lusQ~jePwS8ea{vsAbdvv%&x`k;BVH@+-!NcX
zXIh)#-pRy5S)}yR<N|QqbM4<#&-NC%>KY!zF3o-wTsv}cn7O;*YWr{W@Bgm9kx1e3
zoj=o4gegO8#B~Qn1+gF70yZXEhTQS1FNk9Np!(U_jj3TDzyqZyR2-(A`cKFM-!j5Y
ze*IZF2d##W*o}$@SE0sEu+ZC^@)1OggNMT)!f9FDL{nkH(L~LwA$hN29Mk^fVX(-(
ziwXrC493tPgtQoe=JhyUO2fetZ(K9EN6lj<k(j<>E@;(n&@~ga{Kxd5pA_5`#B2}6
z#+%QM*BJyMe)Wsq1VLT1W={RgK^oLRP!Hq+p2XHmtgH1lytEVGVS1=CimNZc;cguG
z+c*a2owu<C=U>>IzLfq!Xq(2;(z@K%LDsYFuB2yFmx!9W=G?kVCW`1zujjDnm&N);
z(f_I1i}LDW1R{g+KMS8;f+h!(?)0cZC~kPN=_h<DGf}xo#J?63gCcVf$R`m7d-$Q(
zk+x^t4WWnpn~0{W1VU;sf7a5~K;tN|ByPa+HS<kU){jNq@f;7DRbZaq4i(&AIP!lD
zx7-{hyW5Pzxma{MM?TD`lsu#HRzyg~6p2jG(}~-0k9tep3v(lTC-7em^Z$P<Izc2t
zCMXfsO5*3z^YlB<>Y5uNU@>`$i@42FTm8&i(588P0Z`%`St1YrCrP7Rt^bd=+`p1M
zS5yF|z!*cok;@%5{8yE5C~txAz|MrbS=yO&%r#LIU>avyYvms}sX`QHa@@9sPxEQj
zff}^#8vSnD7M>BlCq>QnV(?Yz;;H5C!gK(@KJS>U61nn%)_Dlbc;f~EdazCC;#2cl
zI4_k^|F-YdIV92nJ)I~5GFWCKvB-7vq128AC%!yb<J~Nclk+G25qjIf<(EN^ofQN8
zIzGhyzr)*IMNjNby&!q*7PvbfCw%{PW~V#;AU#c5lkF1k3nyis+pO}T(@I94<|A&u
zZkG3VFRqwWcoN2&G4}ZUY%maExjqG>(?Ow))8Td~sSAy3261A6STzgWIqeq28lQfN
zEa-Gp(AqX>aXDW9K`F9lDjVQQeEP}iQwXvhBKK0euI;x?Z6~4UyFD{SW5(>)i+6zS
z46h`cOpkHhXaN6Ai^{vFB9!5CPajfBrR2zyLuGwE@UkY8F8@Z7wGCwGSwCbnN{!|l
zpQ?KJcbr)IG6zXR3o6xU!rGEtQW})1+A@2iojb5wymJA?_jZf1nhe!G+MJa^5~z;$
z{}}RiaQFQLu5tg*=r>rr=g9l{#p<T(4@jVG&ok%szV5Q4rvLZZZO5r}gArpW2=Gq;
zjA-`BIDtE$;(rXYwbR^CeH(JVAy~;~;W%&p!Iqc^y}!M0_F&wxj#+0%gzz8_o}&bF
zELyU|KB6XEZQ(P-^P8@*L(OdAvu%3h1VGNP)(0yv*tUB~?o6LkX7I7dm0KT<u{z$p
zd%Knq7ge_=K1J+!uOuNH>b||vTe(($F__1;>D>p1!M`QVAQVv9hODkTHlHtG6G${l
z#o9DGsbNdwccV$Cs<;3dzt<PM=@&-MBnga^+rC(C>(@@zkC~roz31V<^LvV}l?&k9
z7ANpqH3-m%^ju6epxO|F(<R|CPi`h>Er5EJvaO@!!QyHv+MLL_-JGnsH9b$H))!(U
zG{cjE1lG81easO{BXC9Bp#EyIR|kDO#Uy2@L1n|j!?(OhR7!Dm8AR(%QU$$~tiVGo
zU=Th%r|OfD6s{9!?V)MsboE;sIU6gEB6{2LXw1B$3ncROLkWZ^@Gu?=2YOtg)Rrcm
zqf6f+RVe?mo$^dm%!sBt3p!y{WRCSik79dsBVxdqP~M2<Y2NZFy2vV##$`Ao`{+!6
zqDFqAB*OUq7L76H2!Vq;Wg7+1K7*N%qhgK*7W6v)^<w~wFQ+f<X}O-e5&wGP-Rhfj
zXca#Cf@juSLc#^V4FHa^h<kVa_DRke=l{j+uayd>Ehtfc-P3G*7kemndGG#u!GF`i
z{`;^?UG!47&#Im=qINWxXHRDlMz}mngHN}c$Y1gK{SYv+Q8|J&DFrq>Z^;5F0IxiC
z*YhwF{1(HWSqFANE8zMVH>lmsfZyx{u%@$OR)Xmud7E(OHggqeo$fq-vQG#X5>6x4
z<D$W_Pg*~v;y0MyYx32Ar-xn`=Wi)2vuv0Eu)Y8)F0ZMmEAHttC#i1<oawJV@n=v!
zhn<Le4DM|%zwV|hccg^frn2z|wQ4sg?&!kdQ&M7P<!$WG4RUQXSygd7P{Uuv$yoY+
zpA*kN@$N<&wAjl~OR~iMf0VstK%7ywC5pQw!JUNQ?rtGS&|rnTyA}k85Fj`S?yg1Q
z?(Xicfx_MORc?3Bym#kKPtTuI|Ib%vpS9NBXRYpQ^mmS{rBLUDH$RZuB6@1T^l(){
z3#mHcEn)a-Gdh!$Fmfoi+bs%98h()8Hfxc5U7VH~;m}yswaj?>ticCejlzR!p$io%
z(vSw|vrgqD!`{Ivz-QALV<%1gj3+mWKBk|Y{)>}@T)f(&g008Z--I5g#hK8ALyJs2
z_^5oDE{Y9>imbKi%datScrJ84NEeeEi_$BQk{_+zUL7q05bqhZ7^3;|yZ4m2XQ<W$
zn7|kEL9DMpjR+2#qBbabpMrkqe=xcK=^^y;{YfR@a4!RY++ywEpC5ev|7FkFy#||N
zdoB0U?De4gU8zs>6RuI;vt~ztosWR1Utp`okI+jDN_qlO1SBj(V9_)FEYHWSDD(%e
z2J7?dO;)G*T|u^qb_ytW{{Hq*+Sq^G5(BS-lwm#)kqh}AyAP!?W6KXmAI%Gu;y5L$
zJjyq+sPrpoF<(Gh!H2l|xa>LYAagHf+cHm7gAIUKm~hbuRzacOmiQSrKnZW)R|7ZH
z?>`w3;0_mM_!RaHL-$Q^sz_v4ONZ;xSebZW(8)(V2C^LDw$=L*LnXD<SPScNrT1mk
zCH)Fn+FM^)*)--XB>-leHxTs?j;E+2t4_GszKP!_!*d+FkJzr#aYC~xPP^{%n;pj?
z&1XZm&F5NY^a$ki{evnbrEXskAoGF<);q1%GQ}<PV{69Q+!m>FYgRb42&1+id;4<A
zO5X8|4O{ri;D7mz2@HJ(<EOEdf8LuU9l;U_wU|5VW*#<S_lDZ?juWI41irnbI=^y%
zJhS@7vsIlkTd{=(J3<5M5nuBV$&~as0i2Qu6&<Tu7NXC5GcR%)-wy8IzxN_2X(>$2
zxm0fL_JKYn%7<DQ{&&89)VZbi7K-u7{uLf}G?+DbXJyC{NA)oxJCzChNP*LH^xK~a
zDXeH*nC&w~Hxnl%8CXW)mPvoPjOJ6+9Y-DPvFGX<EvJG_uAB6=$k9Tct@NUGqpm)Z
zNxx|rjBBzVn<T-ghEY9^nMQ<dz9@b7QJo<6^3B+1xedP+WMz-H3p!QU`?G~#XEnp{
zrK^b_N^bT)YNBZg1Qn3JbFZDJfeoKD@m^lcohzu5eDAN$t5r!4wNtYb3jPAk#^_kF
z6l!508vZa<!CQ_r%j{MkB)~eP9}y01cx3dkeZBcpm)m*Jv0tK1I>IQuQOT(p8y9q+
zO?w{vs=U(t03+o?8Yu4aiQQYD?&LuTbsQ}G1G_H0=b}j7cFDkBFBwm!+)~mGRB4$;
zY|pT7W#iJZV=1ZS!{~OR)IT*^IY38;_tjoqxc<kmEX#rFl>eSd{vD49{InZsH|M*T
zNXA>^y&OV}c=xpyE@IZw)oA|r`**(X<Xwt!8+XY9GdzNLJshIY%=UR@X!cMG7u2Zq
zzYHg<z5Yk>71nHF{0>c*mH1c6jW8M4EIz?OQIhf$<t=J7R6bD-_y|vk$T7ZzxKyeF
zKv+X4Swh%D$vwVUNm3fX6G!}%&-7MG=a{+3wpdJ24+{%G+EL(YFtW~#Gk)|9ju&y|
z`+Hv#Hp^`pjPsrEVsJ?$!QoiFgqgn@{Fa5!@NTu#wJNmi&`K-~#uz^e)n)F;=I1CP
zk1a)Rmb!(C#$O_EKTUzd8L}2FfghT4h~5`dLydubBXmi_jnIvqULlWX>S{bwVTXoJ
zvcx+E7D4>-X3y+mu}EB5sps6gKV);INwp;TX(u;pvIdRC&CTF*IEK#6KH>YepHq@Q
z^PtvylSeyc_L@v=(Lu+I5hQj)wzJin&nD}(?`VqODAP8Y$&xuU5`SHs_>PZSIBVV6
z_MLbHRU{Z$L(=3h8;+C}EbePZrI=nUxmRq=q%_b&a)@QTGWAWDR}|l8m)2yYaIoPr
z96G1j?~_!2Nmo~ZyJ4zDl|oIX%YReaqke2bFU_yKf6Tn4C+)%Y|6eHo%kBFAYqoat
zM6LiP-KD$5r5)sk{q0CwvY{-IC$wFXnRaL?eE6mo{c~l1%sMP)3f2NC^FYZ!*W492
zDMbhm8(}q&UWdkTs*fES7KH!}i|YUOI!Fi+Br<{eIGkOZuy;7{A)#0*E^-i2d~6B7
z3dX4r(3yQw6kH$vhGyyfjR|sH>w;CJY8dNZ=m|)sfvXnF3=608;_GFFgqo|>Z5;Lp
zMZ9uhJwt+xvcz^<4JgDHCK=*pG&fzH!he~=O2dK_(Oi;-9wb^}xO&}~s8|}6Ci0x`
zKoVQqxIZl<elC~P$Sl&^*!E8TMzR&wk6O$wL=bQ{x_c6X8sEKv{~Utmij_(bfJ)K7
zon4N-5v{{fdEY^l_J)te%4sv?nflvPAv|FVA8EOrY;WOtkS2L-NMU<K7dSI)T{8Rb
zxJBj(lBQY9J-C`lyA?}Dk0R`eWVOSiRwFQB>4Iw6Je_Szdt=EdJ{?yU`oG>;|BLbX
zjqL=N=M#omIW_E;EoRERkoI>+3(dVoNmy7)W~?;XN*PZ1UrSUo>T7CqdX&SS<Gm`G
zIbjTr9&ssppJB&?MI-40>W!0L>ExGOsmQTCFjQk-eyU5i_b3(mc3o^UI{BRi5*~5l
z!vD3s8MlNcSTgXKIedz*RauiT#Nor)`KpCQw8N%l5~TVm%}bCDYa8!dY${sBQo|jF
zk=n{k9baX8*4yw89OoSe!h52kXxpNp-4t+9sHO?1rgdF3>+D?|A?vbetN0IxT}=2c
z170YdQmiHHr!f<Q`U<eTc=K-DKNgCQ<Oyj9rrT;P1Z~IMxHNU?>?_{aHo%4w<u7=$
z5y<6ooHdX3pHD!;gS?F}9>%fyW9};(_AhaT`j}O{v5j`4aNr7%*-o}CF>vX~UKjOO
z9(G_?&hN=1QYj)k;fhT}uzF+HX^cG`iG3Gbh%0I397`(Tc4`$YD@nRjlgs5p6a$Iv
znT90eKK7m9-eR4X<wDI-J*=Rn-v5Os{=1PbK<yumn}1m8dWl#%^BsA7qasy~<WH7j
z{b1D1;&1BymTM72zwsUdt}~D0P1sZeBE1WX-Hfe&>!Px|88iTK)ag)2+T%pywaf4q
zlsN5;{l&Vi)NH<)I&=Eg05`Cui7BoIck|mNdrSSHUc?&TyTE5;!<@U+_FEIvSxCnL
z8}TFB3`$B`^Z=eb;?TFz_r(Z&SnE3zT8aKNkSJ#dnSwzU?WKu`yHd0f^uazb<2yE6
zafHAAIM|45A5`Ux>}qpHsg1&PSaf^Vq)?ltKe3tU;%21}yBEu;AzC2SRQeY4$HNU<
z2Yleh)wCV4IB5Lwe&_?0_+YKvk&QIKgYyHY_4#qe2l-`R3|8=JIMlPM2lLX{adiX*
z>ZOL!4f(MhOtiTNwOofUMBZ9TMuk2XB7Ac5{AiyP|I-()5ZfCsJhGX?v8du5o&tVo
zMvG3b^wxVP?hef@hB`Ijo_DByf#KMdeu@2SAJ>3!&@@oi$iFe6f7^2Z#vI#d97&;`
zV=0dj78}tD&V>h-HMIj-BV4^(rBNX$_G{PuFRYq{gI@;eK{2WUNr`+E(YAhY-8>PJ
zqM7d-*4!zP9<X~y+hWhG7v%1QfOu{i<kF$F;+O&IUunKB&|tzoF4B1W+@TA-`9=_I
z+lr5U+<@&%>m7L?g$2t7tBsgQ?BB2xqP~@HINcy!uZubR!D(j#`hXzK&NBW9MNu%Y
z9_|L+TQYlm=D1-emThFo4VS1F)3HR-p41o~Bvvr(FFz(nZ3H#w#zziG_2e?ft)Ztj
zZ5l^pAY$?2ycWsuEMkv0l-;0OnBL_<9sgjZvY*g2=*_=8_q}CE+nd4<O`_1Q=bH1o
z)9e+S?zJ*oStj{+l7f&hsg0%m%DZtQ<-1#`x3KlD{Z%Gjl<3_t_eEvo&P>6Y3B>EO
z6(>}Gt}5?|jC`K~x+5Nfw0>zhq>F|UE8@Rl;(zb;e`9#xoq?4y;)Z93nDT?TD{}I&
z&m&3krmhmAFr_`K9U!5OT*+<1m~d<mcDuC({kunRj-}Tds&w#081Ujp+xp<1_P-Jt
z<oQI$d}qsxqxh^S!DG#IU9psI-9~K`b@`qQPB^L>|7dVIQt}+#3DUQGwzxL#Si~nO
zrtBMoSeAAS(Cm#==!iHg5}WQN*9_3-{qa_+k+nbq9CrA9bS?P1EIQB8V0X9(J>j_(
zMp?$&pAg>#=9q-Y*uulf1#69N$>ZPOxVhBElKo1g%NeuirPL%8g)f$1<V`f0?22bF
z?K4168HeJgNN>3Kxsn++_Omz?TkzZ6uu#oe3ugmz5o<Ny{f^QBe1yJ@JHM;vZ-yQP
z%B%FZKSbbCe9R@cQuv`+KZ*@-laP7xuHEild&{_*67Tp6#;oDS(mrEF6@6;o2=r~x
z7`6QmHz4!*vBLl5t$o$&k790}k>8Up?3s38c@(|~jd?(XdS3G>`(T_~Y;DG#ga;wx
zZ_l3bI?g9}u>m%mr|M9@vHc-$A-{gNOFh=HL?!xxMii@%*yZbXDM3sulAzc2ALu)f
zq^EPpR}DlpyOo~xfZt8;eiQ_oEX@(g49QEGN#u9LR}RF$^DH3nFtd=P!NnX$@dU7?
z{^-mu>=RP1Jq~N>rarT-l*>pf?qAa5|8~=C8u7%3-;Li7Rrg{m>>bFb<cQVzNS$q@
zn=9Xy(sgfjiIa{k#e3(<m1uwc()D$-;Zk2uPP>+vlJ+FOBwaO^Gxja?ZJ5gtxzaA`
zIJY;kaNUuaDbg(OvQtAAIYo%YW(L$JSrkG8HRcOwFgMSF0rEYJph#vec0x@c-<6k%
z!n6k#{?ePxJF+kJXS`qSdq0ik0rrBWT~h88%$|mk`)=19=R_VUb3;Dqk$0K$?`ls+
znDuw_ae{wZ^BXB;<_;R=y1_Q#2Sm~YgyIK8;}_y7)S2VXS-ih-WfM_Mj_!wg9!8QB
z8eRpT6K--6=RWmq?vB=P_D&E~Zgp7vWd8X{QV!L&3?DUiV+(h{6=edYh;K)jhmjPn
zZ0-hNM$1j{>+IucjV!72*q`?wx!}~PXW?RhWLH-VaNd+u@y?pRJ*z>bM6=cUsEd`V
z9=<owJTu40LR*SYd%-|<$3cVRUiMS0T}zPNR{x{^-6l+zVurY~3Fx5H)}_gi;FSP_
z#OyS?C%M(<9phg{B9C+UW_G}9>9)mMxmzPWhcxG~7~$)&erM5w%#>#*sccyoV<c^U
z8Q_oljESVHz{40?mpWW>doq1L(fbBP%>31W5RKkV86zrU?NsU!^GZg7=gpkY7o%?6
zL|=#X0WSgKwnxq=9fF!s*A;7*5?=Q&|Hp#=-#7hqQs|I!pcaH@ma2Cx0}>7<Thyui
zw%-4ZahFVp^oh=X&5aQnG~165FucsW8eG4<)<sZkK)EH4pa>|;F51L6U;<XS$bd+~
zQ0$mgkAKiXr7cp_3<QD3evhoOIsmMoaw`1&J>mj+E&PtUB;02##8hjU5|WXnV>lVb
zn-=G>0wLO;p{>{C8JMGrq=R-nYOiW5EzZ0TOkw>MP`p;Tjj*Mpo?VcQ%8V7D{H0p`
z$*C6lFkf}n1adM9MA_71vYbjN(TL?ym%ee%&|roZ(8d(fzM<@9m4qHlY1ZN51EG&G
z&I2Wt7Idrioh;VB8kT!s7wc$n0rQ*T3mCAYDBQ@`#N^Pf-htM6gT3cRa@|IB<f=oc
zJG?rZVuQI-*+K0N1&pOmXc{?(oTEl}UOZV%hLWnoJM|mkssCM{v>N$;<SJbnyyd!=
zdbYpzwb~;-cuN2@*kW+Z%6N|aao(&s@kbAn60$NR3Wuf6)&X{=w&H?5D}(!!srrJN
z^V49-AGR_6bgT7JBvSf-%=-jOD}cxVtM<FCl^jtN^n$h3$9PgI>+VP1-M((naFX88
zBdoW2M`PcDdVBosO?iJ?FybS`Xf?F<Cqo1-rGf;wRjMN>tM9Hcm$F>tbjyccwPgNx
z$wt^4s%YvI>joRaPPLhnp8cm!2$G07r6mY<ipU!=Fx0x6<dvKxn4~9PUcD)vn>uCu
zi%aY9l>3L+ye_}ULlZdy3Hr8n%YU;o9wEG|8d4rs2Ou)+>w|L`-csA58eN)vICbid
zUv{pU=|%Cen!N|9ho2)2it=BKG;HZP^+UE2>i!@O$5xVsHT+IZYyL8zCfs|>la}BL
zRvhab;D1ke?exK3pE#sBQPOavFckW}2T94d8aYIXH@7VaUC3#yFiuOE>ev_+^;H+a
z3dfzaC?Xo5@AWZTW1!JfI<UF?|46e_U&FGBi5o8T6F9GPvkYYo#fbX^6s=+LZuW{s
zuUVhYpSdNM&y2-lg+KS{$4P0cV^fOvgO4_iPl+K@#f>g*aClyqq!nm}j>}j#1N%ma
z#?bhRO~P)nMfd`*+&e6I1b)^@tY#1%W*w3mm&S)?8sztqotnOXI9rHbXzVYWF-@NP
zO=c1*3EMBFR`9;DLPv|p<ZhWIzKT;i6T7U?@BLsqW;i9J9Tfdu7?x{9uqr3DvogdP
zg!mH!Te6W$UQc&*PEzzYl5)sh?HA9*!=mXy$uMnFJ35Do_g6BgYwI!JvpFT$4fQcj
zUKmi(KSjT9Z9*J%1NC&;_Y{Vp-bc5LGVu(xdfP0Qs^EX5pcbZ%s$OtwL16`lbsdUG
zpZz)|XVN5ys(-%OY^?QSvxBgk&I(Kz$3x*9<c(GWbJw930f~@b^TTC5zTQpI!_r@A
z`7h4@J2~(V3F>b@aWSyB`9nnxG(q7qsP4WTEyc5lyi+cqK#sB#lQDwID?dUQ9n<n&
zC$w%;^4d~XV4>SBp<+*wWV-Yw7#A2uOlgkcQiJnuKSU=&bBDbCIZMZ>J3IGXlCuql
zN8Dr0uY)fizy1{1Nw%zB{m!Y-3m0jm>1c_X5$1`7ebhiwZ`+2t%@ft_*gl!mvV$c=
za?X%IsvA|&B9o_KmQ_-T773255d6q0z$ovPDlsFINo%TN#mI0fnC|umf_@c@kW-b1
zz!IVwP^bNNmLAi$xhO9)>x(}CR44mTHJn;0Pzi(WTa6tEkv5)KF%+s5Ah)^M{ZPFY
zW=;#8#)AgXd3z?Y4MEhVvTh!6Q>#YU*6%}pJ^)D2N)D7Yrz7C|6)A;;NDF04V)UoO
z(ge^;nHY6D=c#v8?~PdZwYYU3AVZLN$_z7Q1E^I0e`Hy&)Z(gZ#kWV>qE^bo3W;20
zdT2Gcrc;rmjXUtsevFemEp!v6(eJzpmk@gp3Q%Skihm<&@8F|1rIAcX@@>3Pfir1d
zI{2PNa~k8$K#AvtIz|6xTS!oZwqsVuIz}r!bnl7p>6<b9?ukx>hVfdDZ~(U9Q^A*{
z4N=rv+ji`7$@XB$M*_pu!OS=D6rnR=YEb<anmNNpY#m3xJ*%a>Pis1rOZ?4h;rp!>
zZ3%29Tt7?53<PeLjhUGlC#$v-5g7qzXSGjPHlAF0jMG)-hl(@fnVFgZ)E<~15Wm;x
zsOnqGffr0F_6`-Og&gdbfDIQ<x+wU->2YqLO=zJGSg2;cAM&JxJO#pqDB=h#>*)r}
zhw*sfR86bCrN9Yz=o6TQw~Y<PuMdQU<j88-1g;W#GW+&MFx)Y!yFSnCeE(12%WnGz
zONKMbkIFmGU8u6CgjfNf%dADD5LL=Q4hwH)cv^~}?Or-W6*SD=^t;^q;*vm>C)pHK
zP3v(PSa@;7^Y<0hQ4#ma#*8=K?+Q?CP9XS+h`>4PG%LEU^RU>T3_m*$W=_xapB@`5
z`r!#r1o2ItXf~ih92><OwC7S+&2UMXhH`WGc^VYbf%z&heV$nFtZc>(O_T_|<OCTR
zp2<D-i)(PAw`Ajlfmmt$sy_yb5OCu)x(Ynk5i$^}l6VjS#@MO3WbvbJX-vhPTQRW>
zx0j!<GRY!PhjGrv64Y#k0TCsD%EUO#1;mYB5#7jyd()P3i_j{9%K$&yUoJ0igFQMn
z%MWyig2Vk#VuT#3XQ?=!5puQ{*ucs#o^$l;=9TcTY}xqaaIhIWUu1tJNqlcGmQ0k?
zWBK;)l(m02$-me@5rqTPz@^bglH}+eb@S(L8n7{mh7=w*Pi&!}^*Dk;Xd~AxX8a7{
ziKTCuS~r&BjV`ndlc0sb5M+nPbga=Krg%rw8O2=^9K&H9fj&ef)MkhqJ23&spU}#G
zv<B+%tOl~D_8EQdaD;jY2cMJlZQZAjZ^jCTLv6EvY&CgvY1>R3S&PXvBi2=~{Qhj*
ztFMYRtKRL}_yIJDGpBW40~_yj(?(nN8FONUm98&p!&FTTFyz$`QsB=QGFl})jhTv>
z;C3`t(j|C^!)AX^ewF_CT7664=&3&cBKq?1++KcqE9>Ck<>hq>-XJ$_5R_bk!#kKk
z=hj|Qp@jS5=q@|?4etHPiTI}<Z=gRM4+A6KFNdv}T1GA>v1~Zm7%p?BJ89KE<;)g7
z8MU_}xwBOYthE$;syK<ZQ)RZD7Yd7*WCZ>NJLr6jf95Ob67a-u%<0)gTYgYa^$~$g
zlCKjvTvA(Y!NH0d`PVvHyJ_I(><$=DU?c<F1d}E=`~=sz=9QFOrR_66VhVDwN>-!a
z58Z#Qi{GF-jG6QsJ|G|kZZueFV0RmZON*wcM7B8xg@DOQOEvHQ#z0fRCX5rGV>&DH
zwKAn@*fEnS1wR5W-x1bd%cqFoc2|S1CSSmN>;rrjR>(Htvd^XqIYQMHv-g)?7Uq>I
zvF-${`E_$45KpYbJgd~RiovOS?bwM&nc}`%)0o#W+_UK&u*qeJ?0~(wx{sM8f04w(
z3OH!3ZaN&e6XVJ+&a;wQTA2XnyV^9A(e^;MHX~Fq=>ILqvyC<S*37hHP;9a)pC<!3
zYVHA;s!*o6^7YEjo;@p>`!#ZT_-bforzv}xtG2}=PP>Mp*d`+_M_P(yDs{Ndvb&5!
zotSM0hhbANhKX?$&$q&|C@YKu5@8i)c#Q0-%9`YplHCfE!41umo6xOL>+X}EO0d&W
z2!1k*t#s>saPSHBY4RUstJ{;nRHZ=<TRZhbeDA0bup5w#Yk5NPKJi~G54miBIeft%
zF~h6^vFc5ns(U^-1u@gk?`9e=bcOMt6dZFcG{PG$V{ESc<g*~l*Eh~YC+HIxLCE3=
z=*+_suz=cY9vIZ!ky+=t1xLw_je(r1l21AeJT{9_`6-P?O;SumB^HIExuOwmUo@`T
zq%7P-4GMvRa*0=Mn(O3G%&TlSNVR^4ir3%DLkhsDCC59d@!W6VVz6P>l7yIDh6jcm
zJRCKZ=|TcXu&B`1OELVbzdkj+;l{S*5Myy<!gkM%R7L@?=#(kLBU^1BTn|}P(L{gW
zvD1*0LZ3gL0y(hED;Z?u?7Cx4D<4Z8dX3A^a8eOit_7#bGJwgE6gd~sK&tt@@E<k3
zV1xqIqn&5WSNu^t?l~6Ocr?*;%t&GQ=JZd{MTJbOJX(1j+N7chjl%Y&Z00#)nGL5O
z5`a10_#Qf*b9Gm_w>m{mzoV5-W5ws_Cxj^D;6|UR{`~&e+9yFjVcQW$ih?nNYsusN
ziHAZuYtW|SOdd9nC}iimraeb~N%4Zw5YXdmsd5V=(-E>kD+O~S#_c{^7Q-c{N65G<
z9mxv^w-!9I2Rsha0LQMuusRJ!g2)AbIEO^qA!EJe)Y71*`JC#*&H%+>j>b5uoN@H&
zjM`{ZdiHd48i>)$R~_89UeFf)XxUkY$WY9JJ*R9s<Sh4xm?^<1l-`R79e<<)Mjj+x
zL8EhDg>-k}hKW3Igu_<JRoQoJp0SRcSwkIiQ>s^~yN%pdV8@tPQd0Fv>k>4gK@`I@
zzfM~|X>|)7cd$b$9wdVYPD)VCIy^Yrt!b5|sh4=VFER`J$HR_ReCB#O&;4=F{rRg|
zRb?1r!8!!f5-GeoV638MDT7}o)Z!^&B(v~KuC(RjSmcmqDg<WbNzBBVD;we7@h2d|
z!X`Y%mn#1c;SRJaz(e<RMf4NPzgAlz=%ypd4+eCIDFm2z)0$p7A*B4VOp=HP#G!4=
zGD`H}8kedt1!1iJM6lG>S{%EJR)g0IeuU$KMZGV9^XL3Vl$2l9W07nPlMPt+Ru7VO
zS+2Ex{1B<nBjuNQU-iPV>88hfLD1wqp#k(Pp9ZSb#{yp%3HFc!{fd?i96%?RO?%hs
zv0Uz#x?pC#!r$^==q&C?!|<>>QV?CWiMk3+HnbP<+d8$-O))yujRt#u<BOHR@;!n?
z-y?&fIrkA+G&x~YV1}rN0oFewA+tOg!{8UeYIRSOeHI{Ll^B7Lh;qoPh>FFksJ7pp
zh&mOdeD=U@IeyS6xz-g%nO{!O=DBAIP3|+ebfJx&^Ht~P(ghb~&TG<&kw1-GO64aD
zyO;@sDw_U?Ms@8~Mq6@UhZpZia!#pKH+$7a*psvEPcgOmCKwZ}18Ym{d8;sH&i-`#
zQjG;-mmOcxxNf6k<o>miVCRcf!>T${vVB#}T0BEt^d=3U0pKS*s1tWA0<^$8%ZyjM
zU9;K^zRhlg51$9k2Q4)9hed?YU!y-HZnhmy{+S>y{c#kIAVUH(pd)1S-Btudws@@F
zViK^X$t(Gce?7^`Nijca@|t&g^;$M|;M+EK<Wa_Q<WkwjeWmlTy0CuuQjEP9f2%)v
zXhUmlH;5(|onJ^pyD+u>%#Uj#68oEzXqV}g0bL`88fLl?h6{vQ0sD<v3??MVCyheQ
z?M5o-!KAZ>@9im4%D{Nic)TrR@kHEW_?2O_xTndY=jf_tPUYp?!I-a$>!Jyct7~(%
zz~v-rQ{!nf@x=*oyt(GPVf56CDcT4yb4fQWAL0g)h@CO%kS(yPqn6r+57*Sv89T)E
zy`ZT%uxkl@fXN#NC(Fu&CUsH>K8313Vm2Xt*$!hTo?Gk9+KZO%65JtegwQ}%jFQPm
zip&bp!hf^nd19Bb;sd)>uqk=J43z})GF{<?cI6cK$836UN`u>)mBIRiC5k2x%adTV
zcp^h%R10ooq#ZY{keahlQMof8mIP$dAchs(u8)mx>Tbp}ZIUZvh9MZ1IZ28t9*FCy
z2s0ZFhO}(^xECn(wNCKMf~*Z%=_MQ<W<!kX0OgX}e4DGA1S?OYli`Q%y;R^mNaOGi
zIc>37eStu$4SM?gF^1mqE%B<r3F`q*ef8R^(6rb+0A4e}+z>}@p!))QSOzw|&y9j9
zpZ{{(kPtl_7%S`#Bf=D}*#^DOdHay;V^|+yLv8HwO?i8P_XcV@^WU#W#NVA4b8&2u
zcj%a8Rm#fE+gI{rWS5Ih$&D*$<v?heCQl90il&#Zl&5)~fs?u0Gu8v73i_=IW=L5j
zn3nhn<T~ukoG!mQ-9U23(b#P-O9z@oa@l&s#6b@pqLr^`^3A(;Pxx2}RbeTqwGM4l
zeV0O!M4YnJ&2?Ir$$oPmB{C`7U~YGiCWcFoIr(rKeoZ}IFrft4pmAYi<cqnyqBQ3H
zbFDsyo)r={YUEF7T%trjY(+S-w1RS?>vSiQD$AU{Tggz!sm9X}8J!Yk_0VjtxWT7r
zwN`p3p6s=MCRpc|X+PvselBFykYp+6kSiR)wNX#h^gi@53i#-l+mVn}nU9Vvmq5P_
z%LLniwh^7ePhf97{7F-elPy6c7WFGncfW+DeoF0%d?R*YHPWFW;B?GaI<`=Lrpz|<
z!n=L)kG2V#Pu~=LPtb?iFiMB}`9EKV{msNlkpRQBP7t)#TH{jvwEG(S<TF|y_knug
zm-TSq93y0J=A^J#GJ5x6cLM}`j01QA?Ri^k4=|Zmo76~Tc`ZYBbnLS}M9X(OYp~%T
zTa0I8r3q;-EGP0#yk*h_5SqeHl%ma5Y1i@4!HmKQ@?lVN3Q<+Sc~H<VDUG1I`+uw|
zL46cL_rQ;Fg#Bjh^ha&;Ehk`KNl#y+diim0sIdQlQAvJg@K_qO5B9kZwCLM!JV(D;
z3)Vv_VpO-X>!H<?2`>dBY>sD?4m1RfpaajmAV^XaOfQcY1<z=dZ4-?!7o@>8hMnH<
zDY**ltK<*%*`tQx;|4HKh>Oc60VMRj7grEmOL9WlfZdn3&R_nynr}l96OW-R76Otr
z^rMQrPmPAN{4J$9XNiG=vi?%--IXEdAMMn4!YKb`P5Ia4?r#ooed0mM?+5_}y+Tk%
zMhT8RMJYN3Iam;BWGaQ3TDuygYa=&Q3u@ue#8pm14|5MV79CzVsM0AOOO1tPN%C5O
zW2&Ugjxc;DhUWEHDvxB^<@?#cI3Ck5p&C-9n_QP*Hy<1px%9PBIhBFO%4bkOA%jPm
zst<Fr#BR8z*fh5;&TeWqjC)DHQV*~5!n~@c#RV`g3q>Uz>vbb;%&V0vFxxr@ulf|i
z+{f(C9$N|zeK8LENBR(u*}N)Hygd9WJ$rppA@(H;xj;1U<kDa=k2|*3x*sjxY^YQO
z(Gc(EK#RY+u{7YWOMe803)Uq;V*#6(DJ5v9Ts^JNTK>6e4!=2Z0ST!Fw?lhF`;DsR
zNf;30K4s$j&r#8tQ3Ok;w8}L8EYt**Eo~w?+FQ-7p(C^A{2JXucXdlW4}EOx^jt;?
zVsNfKNonpf_B8vkRLMOV)%B`H-`@G8i-O{Ew@6P~{vfYJ_j_KqbxkYxN5aSIXb8JW
zA3`rzv0DC>I0H@hfAlF4ESD(o#R<dl#XyGpj7NCGgfNWUXH}yQvx1;{%c~aPfv8um
zzLx{&Wx_opWrbZ=x`w1IS1<w^l;@!rSOA{+={|2c1C2<mGEA&XfT6@3nbH0dn9t?V
zXgmn}6Om17E$FE`fH%PcCbJgEbZ250SJg+fJ>(dTI055S7BjXqg_+WDgBhtSEJZaJ
zfg)r(>DVf?YxZn6FikAbUM|H#k*y4pviUP)<mt6@sNHJT7+?gm%5pq#JdqM+r(I<e
zxwdi^te?EDZr4{g$Lr;%QwS85?r(PvOKEjW7&~xDE<3pueg^8=wDR3bDlBq+S!Nto
zGZJI$7UzZG6-$K$chM@-74E#n>KoJ+E-5|ow}(GpY^h5$t2Ndsu1mCFil84?Pg`rA
zHdQBQs({)5?p<LI$86JST!HZ@&NZj6DFq(abbFhM0l$EtY25fjtIb+ot5nIXTAcwj
z0LtUAF)N-}vJ<?%xrij@cc?`g(*k#*^X`YRDga{Z0x>MH=3i@wiY1n#W{BLQ0U%$8
z;zN1(KZnDmCrCwLZ<x`PgQ>4&eX=4VJ`v7PNcUWngrS<{VF;y$kJ1Cw|5_tMRI(QR
z>G|KNWZe%^nqDfyj>5TimLw>f)u|*Ku68X2#e-+WG?U$SOJ+}M>6`w%1PB~wD+_E-
zI$1l-JCb?X_>oO6b(n*U1lGzd)`g`Y!fXJDD3>@y^r?x4b@81)-Y{J{zgcXcid%If
z-pzY$ehd`E;evQr7R@ynv$;ZR58k~V4#FEw;s$m$9@}07AHRum3<O^}gMF7`C}e44
zIrPaNJM|46+w;vBBX$g6Ie5g=n|kDE*3UV`d`QqwyDY5(*iFubaYIQ{Nz=%2L8|m>
zkqxpQ`0^pGaJM|8V7oazFJX0jW~a3r06GU>4Xu!;v>Fx69YK+%*)m2}JP~D$C((fE
zJxeB7ETK~+xz;mDFhOdGHBE-tp9X)%8oZi3IdIE!I9@(Bx8crfd5sm@KA2Tfu(EK%
zi~-zsOg41l;%~*~!u%pIQHmH1J2N+gjFn0AKC1FIdDYe`K@DJad?VXhZyvIy-J;(O
zDV^lpCOD{)mOdOo?k9|%yT84jKPL-5n3(H!?HPr4l}Vj^__XZwuyvd}yEO*IPA}OH
z>h-)e`FR{L3WSC>$+!YqRC$_XFKKY256bLB{|;mR6Y%81q$K<Vrg&?1g-SAI@?s71
zqIdj^9<QdT?(v@*-i90sj>}INA<<#C+hs-Ba?42966Z)bkl@}`=pgT(>4fYJ*JDl_
z)niVgf*(NYr$h%A4=UH0jj}idc2g}tYHf=RYmHr>LyBP|bU7*5FHGx*n%Zk|!%u9Z
zcsDd%cSJgibOv~|<>s)ZTA11bpA9(KzM0pn&X!I)31m+zTTkAbrY~IMlFdJV<lbNW
zq|j6|12v#5@__6%?ZsCEx9`>ucY=hk24+Hl!wadv)hvp=-x-j@k+Q;rBXY=IYGC<n
zChb~jzlBJ48~TIg!NH(*Eijvbp<a6pkA?!iFYdjG*+``;9(rDYJiMO~QAs>I>4u4O
zfuhkX-Fi%EamAfSYol@M2~p)&h$x%#F>-5RXm~vmOggrl1>+|j!$}O<nz=fU&Ud;o
z5(C{y(N9$4dhz^V-i~@_x#ZpQ)w2Cz9-FPMYhrfy^oRCpfN(+C6gp1g-k>u2)kX||
zlT8OxZ5_|UG~@L3a+$+E0onQlZ$@&S$uh?5Tf5->qB#lI5d(wS_h1amI^t}TS8K7$
zxy}b?=&7n1btIc?$}ICXjw2@$kzSUiXBGX(_5#-7-r79gp6;ic$F-b`4jW|qPjnzE
z_!OilCeCgbe}z&j7(zKwi`ceoqz;MXATOIip4wwh1L-<)5S7D8ITcI-5>K0L)@2L;
z9>X7NViSzKFX#UTK1IR)IKTRoxS(8wM_kl*Q(<+<D=LGqTCKGJJiTzA$~7uB@l>l*
z>piUqUIo0n@k>xMS2OYt6%wuR!#OmE$*A?m@ki#{L7ONonfSqNa-BLkf7p^y2=XX2
z!T3r%aWwNu0n?Ia-#1L#P529~n}DQ(V0vk+VEv4T@Yn?2bNLK+cNwRnXZF~hcQDJI
zZ&ce71oSZ3L2*!?`<g?r=Ar0Nnp}FF4`OEZ&u_)J-08~)(0K9qlZbiJfUMzDHc3(e
zZi?@A(uqXYmC-S*+J|(d@{>8>f8;YTju$V_#6OIMJM1mYQMn$k7w+v0%BiVj%-@7@
z*s-$KKFS_54O@9x0d(Bdt)Fkuz)QLX`6P0+jy|EgrK|ZwjP84*`O91$lYq#yX606k
zQoua@@YP7st!p{#ppF;phr)en*4BJbhOZt(ln+1)BrMz3#++fVznPw_$DCld3pu!R
zqk^7zWr6DqXo+2`6o9;(3p|&x3*0353S7-?3Y5;ukhv@yl0D4(3aqT8tcl1Z3X8~M
zW2J(Vhk1&Hyr+IpV3(ezkXGdI04~n=P2!*!`%t>qH0j?k{usDXB_t>~!>}~=gd^Os
zQL~ztb=eNQ^rs=ZGexDduTHR=)q94J&=HmA!lv+&z*upAhB9f{5R$L_UH-e1;Kzd>
zH{C$*kV%2ArzrLb58c$a0-I6n)>FQvinh=}`4htI>1FI$JnYB`R{f-+W~7lh^w>mX
z(DH9?8_ciq!Ou}`GYRlPRa$Yf$SH_yiIosE_>@LlU;_)iPMJQV=V1w~2GN8q^h{`S
zAAV~|Y4grWX<1O~7j<t*f#GV-C@4rz1!*Ok3sg2YwomCQ2xOKHd+KGUO05+^EP#{C
zBn8%TJJ#{0^^CFQbec5qKTTir{eB~S9eZbz&}|-%o}d*=(oyw}=e(+B-rrHZU`Rht
zeW{5NgouP&SjIf-l?s|Ec@^S5WK-*}AupH~#F(-ZUZ2{!&iC}Xc2m%+)jKU+FoN+5
z=^I_n6X%>O8t%#WH16>_m~g0jxf)`6zS^;B+AW?3Za$j;tsmc7d-Q*;Z7@knSrwV_
z-}uUHCHR@Prk1C{Cn>7cq#>%!ry+_%r0N)Y?E5-QY+Ra6YEsq(wIIU$5*n>%Wxh+-
zdaDmZ^0oV%8tn&MstU^(lgdb3?Y3f<?b^|kZ|k+1*Dg}V8bRd`@&+Jv9l7qo7H=e^
zVTR%Cdnd}Yv?ZCr{=#VGGWv;93cOgO|A2y_Wf4uW!vrBto~6Jl1U>tcYOAWM8*h}z
zleiyE(ZL6RAI9dWqZKBMz)1TsfNo#|>|m}oW(`DH;Hd7Gz!LyIwH|Bgv}t_CB{tz;
zhd%sjVxZA__LS_oX?eNn!2t(22cHZ4)NCYBb^VphW~1<4j|zCwz3qCh8he!vS)aI0
zUz<}v$~qxrQ;Lq#En;yfov7eH<o4Nd1FtA&r^h2PDPU;|a30auIs+`^0f4*T^2uw<
zZDQ@@mtN`5Lmt+1waF>&f}LU(oZQC?D*ZrSvBlc_fDwG_x#Cbo-VTphf5z*O!S=~*
z&J6W+NqBDHYN~dFPjQ{PJ*S67>5HZkreTUMObnBOnUa;!X&x-&O2jXNJ3i54^V?0=
zMO~@IAr-so^9ExLDOFyzy)|*pzcu~<qF1m#pI)%v71Do*0`F_kVpkN^25x6*a!>M|
zKUaYFn!e5*tG?P!#E@^}QlD=<YifJE9YkWFJ-F!a+lRYRTH}>sKIQPJqbSbB<RV`S
z+bNR(>Nam$dz=!cKMs7>J0Brwx(O<NC?34N6)rfen>Rnm5O=x0D++pe{9$%*GiQCx
zQ^DuBD}b3<{bd{9e)dqV-f9<$CWZZCPsr+I`n4LPD&0hKt_^)UFARx}H^ISLI#Kgk
zS|l}YI{6$%_f_;*>H~oks2G+Eh$$j}rL1aSHThtuLd*8QQ)uZYo=o3V-52?rZ+gWB
z;Yf7fVuQ3kXqR};vM{eVDLh+%=OTqI_97np35$qIrcb!pri~&}>L|)MPssD+C{m5H
zEPkh~u}k&WlV!I@8kFtm4Ar#Od`9Q7pN`8X;c}dl73OscI3ecH;k`~N&IBeGJGApD
zXb3$AkNKH?A8T-rOK~SO^0$hJvg4FxUW%UcxQ(8!&+t)gKfI4RAr&z2%WP09SI_|~
zpMvM(s8hLpuJM@{DkKhhG%OFtob6av^AVR5_IE?=fXb)C@DsLk&ziM+D>EJDa%Ah%
z{^n4bruxz9!Mq)G4Pp12nuf}3XYtRqrjKaHBXQcRi#mgCa`55&v7t^@JcV|6M0QM|
znWc0*?Mm>caNfkjR?PgAYTrV0@M2jqWBvZ9nGUzpIew&Gjb!dFc>c^?!)hXe`hXu;
zoZ;n?zZ9)gR((ohn+48{;+b%}GI&4&oo3Tivl7P${=tWLsmGXNug5}pq`<ZkmX%)N
zk)>T>muEg9EmJ)qGTX2(9pg0gJ<r_+-s5avUl-Eu>x?T*V(WX#iS7$2%-<$ui?;B{
z)nld=dP%X&sp4o9;TSsi5l5y~100xi3mc)L_<e!il4VEh2a_x|3&SP-y9yxEc8gOZ
zKu1roy;hdCZl9vGtZ<D}9t!(AmdqQj11kkyD|xxCZv^a?0QO=y8+w`-r2aIbYW`y+
z^!ImH>_gm$Dc8Bi9P<CBn<5jiOB)z7Dvij+2{T&3gv-Jk(>QyvR`L#>lC`Z>U!#>v
z^>x-Q39GrboyFHM?#0q@SGF_;8*^4kpSZXks%MGbCg%=krscR$ke8R?Td01C58xIf
z33=4?D`IAq*Vh>;;qp>QMjBr#q77!90k0MwHY+L3eL*rztw_79%NDqpgI1zZ8L~A`
zO&BAC2h~jIR9T}<UfactPE%z+sW7{9C=}tz#)8WgbUCG4@gpZF`fBH>TgxUw$)Z(b
z6+?<WASSa_+o9GsTj$nu`<gyB9qOL@t1Q}$F11zaTJex!usl))f7*Z;bz1qzayV<@
zC=wgN(Nb|ekKNpW*&(;(i8g$*&5XwdgD{VI<q4sEhern&v|6h9+@7ksE^fyHZzLS{
zek}JNp0!8K81Z7y|Ekv%wkOMxSQuV%s|XnA6nLr=7kFZQ#<0{D#*`s<mv#WrZa7C^
zHLv<Z!1&`^H8p1tAudG?k;{U%b*BQ~*CnA^lB%ytkvgR(4lJ{0i;UcM_nyqlReCzZ
zL+XPflh(1-m8Y1{wCKq?Cc$09d`DTk%^Gc+kCztw6+!tv@>tybsjO9kwcYwp61!#d
zjklQ7l1i=F6m4u;_3d0@^THz9i9$jE4Pn7A)dXdLg3_L;p_H2;BeArC;VEc)6W6lq
z|Ax%KrKmHxV+>1dwhUEijq2Cy-C`yemWvAN(XMfe_pi(uwC{0ArWY{HrOhsLS#6qc
zW2UrO2PW7v&ZgDsaGRYAVCq!p%hjJ?WPTY`ua5pDVbszBLnbDpuLW8jyGWZm9kCpw
z32!$Kfh_>1G&trWjc67p(y&ZDckdn)%$x6fv9=bYQxJj7kAtCDG0@d^Ia}a-tWw~)
zGrQ@onB_sLu=zUI*Kg`J<mIyg#RI0^#SorA^<lF0()~9w2XEbZbAjx`FV<_7L5hdY
zQEjsjsa0<C-x;YDGGw$Y^OrdIP~vdu`kLioy=+nTvS88VyL@T)Tym+VH-FCT(|_^0
zd+D%8g7mT6!#t<V)3~Lq=0f*sgD?HL10mh*6(QaC)-rv{=m)enT0m2eJvnI?vRA<q
zh{1<7^A-s|wT7f%`h++?oFa{|W6vX69>wcpbi?eELY+ZNw?D0%3Z@~2?<3oD<0&aH
zvq=SKUB|?EHLrNM&Z=)Hod)ubdF5-#PU>)txlJ<9G2MLBlx)0N6^$zV5RBxFbV|f?
z+!~C~Vj#)FQtpj3qDu{G?o@{njb`;d7uq?fYrj#;th61xrd9kFoyTe`2T06)tp_|5
z*4!>&oxDRN;da|c0<&&DCL?jcrp)Hg)p`5PUQnIFPl%E*+~2tpK|8VkPRh7{uU0zE
zPPx%~&)=xSu@d4$+J74R=jBAY;UAsDM>H-cq!qlRwr!(*3B)W7G8jX<_7X%2CY)96
z-@Jy^3fR%QdEw;b<uu3rgfYbBhk|Wo^99V4RSL_@it!wcB{2BtatdE)3w70{1dn@v
z897kF6!ew(%MIrmax-eq-06mzS>Ep2=jQ<Uo6*;oYbu}<AGOXlp=(;L>L$AmGoWce
zkb~eFyJM5@B(S0EPRBj{X<}O7WmC{~#AN=kp<OSZhBZB#CLQyun^jrCxIxaY-(ypX
zuvb3pXe`iszQB6w`ZG80gvZ6@{K0n!`CPq$r{4PgqnZ!I!C`M1O5*3qxi<#rLGID>
zsJDb)ZAAzCHw?+<JS(DXIo0~t*>#yQWHNkjQPREEQS|OQ==2`W5cF;qNFE+kRj;2d
z3ik@fl_psYQgRGPDBCL`YO6C|O|X4LP2g9-1GaBbP!4J^9;Kf{E5$lz@X!Lnfhu}N
zNL<*hwV7=T1?{k-;sO^u@Kbh32fe%i%lkSG97VI3c+~MfQrp4iPwOqq(G`et`KWCP
z;OFCaEgRYvE4{ZOHh1<Vwb`}k$34+lSCJJIvt9ApDvP#302aNx_kqEd-A$35&f41X
zs*Tj2my-{mI+@k-QWcl+f5LED-;hO&Wt!69@$D;qyS>CB4vCz0c@-A!-{&meO#B$V
z?fc^&4bSRRv1*uiaMSJ;hDNO%VgbGMndMx!T~@_RKvN>KIhv=@=Bkt`(?ol(W4&~b
zc@#2_$Db9fXS~7%R#Rc%CGEOl7&M>3scz?laqZuv#HMDjw%WcP7hHL;mh{k|m0d6f
z|1b#O7S7vd5mtf4pR5aa3dPo}P<JO8&5m)spjQwdwB99dE%-w2175xbN<pe;T5ROF
z{dVAL_dHS*kKLli_U^dJ9{;ccSA6V%TRz9GE5ncytHN#QW%J8TSgl|9$^xXJM7HEv
z6lE<auW&+`lW>d#%7-0HaF3Y$a@=2XzZy7ub&_puo^=uTD0R>h(ti*}OM9^Fmsy(G
zHu8IQ-+KC1Z&lRxR|aXdJgbs^ap!b`)yh%4r%k8!s9vKMBX6Rm9BM*ei?avC;|MCh
z-HKL0KsDg7)Jtm5k+J`6J+}I#TPrXvemkiyaHdLu0_f$XzCbkyQ4=|s)@gO<eICY}
zelLMxiLWfL`anr-HZa@o0wt(Y|0Jj)8LC;04nK6gg$l)7*8g^e{;MB!`^nXt=1rMA
z(+a`ikA{BV6`YKK9&Ri-ZAa3&1(%VN#E<UoqFw(O!Z;_;dJdp@j_lTi@$mohfSo|&
zUvMqW6ast2cUL?Y#@sw(vxN4%x5L}*jvLMU#%->3`)cD%GNsA8uuk1vUD=G1TeE`$
z2@T|y*){~7ib08JMD!=}FKGQd!u+BptFY#6o#-B{6c6KkFH#WsR?!_Z6OH}Wag?N^
zlxg;96n)0w+;Elu`n?|S0g~R@D?|F@frmiz^1eX6s*B#tHr%sYPo$^mgQ}<L0q9}7
zqfKBZJJI#?@|Zx&>N@Zb=epW-;<&OLe13o19Uht6E4)C9f*-e)uzaGZh@|Ktk3v1B
zr!;m{Qh9r%R!LMWplJTejF`tFoyHmN@V8w`opFdccrDxG(9OCtSWmmkDlx%fv3Pb_
zw#LmXbncjLwn-0f=_{akbcg7;Xr!1cRfA@{lMj*Y_&BK4U457N*skfUGgRO<AoSrS
ze0s)c!P}>Pn+d))*7bZfkjt@;0lcnmbs<qkam?q@UU~|VGmsB3U!|v)DEs^Aw0Qs7
zJh(WA3X?umf3~)v(@Bp~f7Rx{I&&jI&NgSj#AL&Ru<i(U{X+C0iel4ab^mIt=xs%a
z!tds0igTzM4Q+_ug_o{*aY3h?#C~2+b&#%mF)=0Aw*3(sgTWZNWmwm%)?<~|F3oiF
zR%M1>WU;42WQtxs&R52xY%%I8u!dl?kQM|=6wO-SK4-Z@t$_O)<bAw^{)_FO^qlJc
z3SDq?U{T~OD_;P1S5Tk{hUG<7r*Q1JrHy&`u6;`GFFC2A?b5_qqXWH%#fC%RU<V^m
z<(AVGa9e#nE0TDqLFHG96}!BbPc&}Tv-Qi<qFbo7+BZ^)NfX>GtvIidNn9x5aG2js
z-X<w6{EA8TIE5#0b4~*M90GA&3(TEqy0vW;@<<LI(aRkip0}DSIo7Ty_&m&$K;+0Y
zc{3H!dRHn>1PU~AWo(-rZv3kBhp@C3fGt)p{<!LlQpNVZWt=C0nMK!%!=SNIja6RL
z-q|4Aw|egISY2?|eNM_tzl@-tW#Wl@KsdK_@Q2wL1+~X|*kdAN0hIES{HJG5^vBlN
zP1}~`P@V!8MqHjCX}=NNhtiNsFTcL^?Vj^h)*GeYh0xk5_g}*z+pJZdI$VfJTrn!6
z0YXf%qevj)Ic^xLgdSN5MIB7Bw)*VP*irgf<)WPQq*X2#tJ|)t!_!w!d1gxh^YiXa
z4pJ-=|JkhtJELNzTNovHouE@`?RKpFR0`5AZQTT4=={fv=n4~yd{RU&b8*ScZWx(%
zy-O>;ZASUIP5<FwedY{n$crrGhEHkKK={k3SN+VskK#ODYkQ6X_H#aYl5@1a&Vqzo
z9rITjowo@{)$Lpsx>g=z>valt52KU6c(G@*-P?*-MPH=J9%jaYM|w*8Nj$<=+XS?S
z>9c}0xj8}u#U*`17EawBJTOMFN_V~muqFohqz!;wHh*^HRKWxttoF)|K9m77^J9NR
zUQMT8*qg8$Fo*J%>!pn$&%Y8B)kSQs2Z-O);(vzDL#8t`<~_ev(N&@yf)3*}sj39)
z`ap|bwf#T&^`C#9oZi3n@iy}B?8x7#kxL1>6I%=f8r(9+BZ3eoQ5veQe&2@s*20|^
zc7jVK6;kpKPy-uitChs>*DQgh_9+@A7EC;Mj20|mRfXm{8trd@m2`pjHi6ybXnAYy
zG(|>Hu02IrtRTiTg;z^%cJ-X)<(OWAXyH$MF=Lot{$;CT(}WX4?^a8eT2r+A%{tSm
z>+oi&5o1U)m@#jRjGuFrNPc#IZ>+OT?<*q*NkchPJI5Ojk5fkA3y5H^hTLK=uZ(v8
zDFd?KALy6oRk#A~3`e#e+vQ)<tWy?bs0T3_6v%>1HYeW4jYV)Y$}(%VE4W{!gH0TW
z3UclD%X2Yw$vvdU9fuCH!NEZeiep;!s?7|TN6Bdnk+3NM%$+KvrxLU|!&eNMYviQL
z;ct1E*2@9sR@Z1DX9%$GEeP*Bcql0S(8mk)Ezy~>GH621)exbbdt%51zq)n`Q))h$
zYPflKZXQPC!#71}E&FGeY~Y{!9v%&CY@d+)o%;Jb@3*V}a}d5Dm%Xxs{cM~3b%EI}
z`nf^Ci|^?RCl*2f<zdFabin)nc@y~r&~#vl5nwG;dMRRIrj@IB(0~-0hh{AvWLK{a
z<kO!w__(cjRWwA{zmOq~O(xy){O}2Z$!H0XNf>Mbvgk6mo{9T1`6Xl+G7>zpT=wy!
zZ_CMQ-4JyVcXj0?Ry*ooP>w-`u5JLvyqt@{gq}m~yhERExy^$?u8UV1C?Q>YuWBuE
zA426@hE?v4GHfBuL|mX{vp)HdeyEnhj4c+~{aq$HzdkNjBFfqs5PZ=xX9t=H?8YyV
ze4G@3lsPqRy}WO_$bK(ynO*O?m7WXqnYYqz<j%d29U83C@ajAtRr-G@d&{sW->q+W
z1{9=K6eI^l$)S-R21VhIlz?=1cgGBalt@cAqI3`4NW;+G-ObSbj{DyGJ??!R&xhyv
zJPh;UI<IrB^{bVy)Zdf%b5{j7zZ*B+19+u%=0k)?`c>Nf649e!0yg|pv13ELh!jF-
zMp?a-y@n#xS#hc5jviViYL4IR#xP3fRvJSozwc~(_<yV=2HZYdW`y_x`(A{EHwi{R
z_V;n@*C<@i&soiX1k|~Eq~c6Ko-C#O=W+Z5?Gg-v0j&i-3_Eq>p!Xmiy7D>1m?sr#
z^gQM5)g{VybF@x!v)ZNAdCZXX^swYIvD6~z6OY;>pe+t1+il6SadaWqp%T#{A^b#j
zCl1)45#XXP|7yHSrbG)1SwLO$b<DxDow#tmQNxk1A^Qod<j$14_k=CuorLO1olLfi
z@YKaK?ThJ;wH}N7Cs9A4D%TS;`5QN?yp|(%(>jf^HCADSSW2{1KKf9jdBjVU;uL?+
zkBP*P%Q*FWj$(xaiXCV1^Mu;Hm??_|>w;-#nxC3i4cSy4^@WTQ0~7sEv2$ZWqyX;f
z5#+ef?7I{?dIo@!;9#Qc#g?v?i(6mVi~J6Tnda)3-={g3qA*0_#$z0uk{={iYgI1H
zUN#*aBI&dHmS5ji(CvWtOt)ast1#Am-II0<y_E8Udv*KojpY1q@a83G03#gaf6xA(
zjtx_`|B8-1e;Sz@a_fJUs%5N6S0mep&@3+#o0fT*WB}M#O8WQr;hq@+#2+sHXBfMC
zmCrxOdv3QVmo?j*Dw_z1+s0>`)%a;TKbsy!8*c@0{EdMIvsh_Czfv)v?+@8b9DJaa
zabB;fvQ2tX$?V7^;PDYb8XdFRVhi#!_UDX$w7^<RQLp`~kO{(Kj+*TrvU8m31N|Nn
zO_Zn7+VcmBDR<ksLN$`o6`Nj2xgD>@qODul9N(79sTMjhHnwo|s@G(%D=u$XOjQ}+
zs4~!uVBbaHnCx+w>V^2aVMWXk2@xf)<tZDadq|s0%v4%WjGFVjg)bXf&(z8#NxND|
zy4U2Pq+1milH1obBa-{6L~j+1)61rfI$l(v4@?kXLoQAz-Ai0C{Uyq6n~X`Ke<FRf
z7fd_22nXBp%r8r|_)HNKMnolrJqb6VFeuaATlMLKU&>i44#G7Ce+z}e|7D?Y<oJMV
z_==Xxz>gMoqHw#$;XOpeKbwwO8>HiXyQ6!GiGSZJtePLrl1*x%JCf^$^ct=1YvG+^
zwkQOI!A4~Hw<enID(+6Qp}Ce5MTtD7{r8_*56(aAALxIJ`j@7>L)Yk3N63;I@QO?p
z{Qb=v56dHqIfz+ok(UFK*qlcrdjTrfv@a^&U{b^#QKMuWGX<E$1mD^_fIS~RqKIhu
zn($CYeSyV~>oHaHb7{viGIqoLwvEmxZnl}44~8mX6UPTmL)K9<+Z(^&MYcrs5*w~i
zO;Z=qkK|5JRVXc5+&O+&M|<6AY;uj1*Rd5`I9Z{QdHKnh5Nx2(j_&*;2#)MS-i&WQ
zJNri?6<~lSAd`x{3_wy2^3rwE4i?Or_7(Cf<t3CYRFZcI7!w&OR*5W}^U`U?v~CAK
zB>mj!)|&Ci?d`*HBh#Qgle1JT+SZ>C2HfUl8+I_F;-1-%je)PyyDq5GzzS!_c)r>G
z;*RNj-DHcLSc3)OieNxQadHpZMp`Icnsr^P%r<FpVW!MErvLj+*_Tbb_6tY-;a8jb
zBMI1s@WRmB39VoGMYW$)bzv2a_qPW-l`}R3QL;hICl8I0%`ze~5?cB~yi3#N=D%#=
z$4OV)s!e&U#r2AD+Aaz3lV6~%ZpO6H=?cqM1ikL?MTy-09c_RBMp=E~XLa*Y#Wo%9
z0%%5|r{2);<<~~x_};C_W(Kb0_C=+ya|dk2NvYU%%ywMJnq`0m`|yv)qsh`c(q^@y
zmLi<hNglJsu<}FPNA^T?g9sC2=bB1YtMTu9sFVIH3d$w}ztpr_{#)bWzFLpwDrbjt
zo>Alub8XJV!z-62+s%pAqKT9!p`1EkZ`DLYW~lKFfr=F1MRRo};lmwWpqWYEEi4`H
zMcLt{19mV%6Ok1Y<A9U?&6(2{^GC@<&CVUW<{q%`z&-~h8FjSf+WH|~<5+j2;`S48
zEQ5{_t{XARzK~jq;a0E0S&w3J_A(9W@6fc+WS)T~JXp@D!VZpkH;<7S#f=7?uz}sb
zOcA`Gqi%_6mi)*5GUfm6v0c6_x0$QueA!C)&7XkW;_l{hdn7v?^DGuN9n-G004_4K
zGNKRt(YHWxM&sOF%<?X06kC-a@hwXCPO-S*OrylcWqVW#vvezrz5eh$5hX8^%uSo=
zGRyc>lnoatN%OV7SlNq-nD6cWhR_}z)${}}8^UK32^vbL=M-Ybh)x7%6ZkQKm}3%9
z7F)`{`6gXVbD~2w%QXE>TK;Iil@5k~8RMvx8Cy!Oo_6u;hq)zB*Yhctja#X9Tz%5_
zeo0o-V*07%g$iZ!(qQ^IbWdNuq|#nbPI+7`+lL2R4AJ~!m%VRZL_THXZ1g~Kq2Sn4
zE00WDW7DswGw0)gp_9eLdM&yF<+QR%HRh}@-%w}R*weBE33;-^=t2}oJodtlmzc3i
z;|f^fK6e3nW=1Vm4lyu$kgTW*F#q#PjfsQqg{7XezmJ(?+FQ#?26;E_wdV+V^vOl+
z?=@8=rL08i)3OT&D4@`vw;yM{NUzMP?v0_a$rT@C|5a|;Sx9dL&Bs8tHXGv8u7}cj
z?IkzhzHM$+B^?US${~w=|5>##>11Ci`vvAR!spnDp5I9w4SyP?1Vb2o*8fE4mtGLX
z+(PiLc@7kC6PdEw=OdZD1%54ZXjXzZdF#`*l1jZ5&zQX<A_=%ZkUTI$i`%iZUU4}b
z9fjy<>*(AybsYP|>L-T{#9>%4GfXuEy(a`TGvJSp7iwZh%v4ya+KC?t-CeA-KAprP
zW$^P^Ntp`rTt3+zV^w1ZMq}2w5gb)n&yWGkFl(T`#*hMpJkNGX2p{dk03IHm1})<c
z%@{^DG*mxzxVk{&(OF`S1J7d36J=Vm*>|Vx<eJrh+3{9DujiK`>+HPgK?b$Xad~A7
zpf<B!SF%!WtS<n1eL+UO!C|&CH-bu=iIHCmBhDN$BU|S(L#2fhJHp=H((<9|Ip^ax
zBmMYteieMLX+3FxN*G~!hl;;EEFPrtp(n_hgf6yF@nhwe9afd?(3q7m<8=ZH{+|_a
z!R_p;tMp=w7}7M~cq6tpedDs$<*@2>W5QN3yk?Hq{PIKJG^cd_ClyQ8qESE$kZpUD
z8bQvn_4%^-CseHg-|BvZ-;Tfv`L)n$$ukbH8WDRv(@;}(!)wZMuU<AkwkpMx1;%CA
zd+$|bBgh8Wet%;ZFTYF@hedwJG=^LAc^_-0tTgVe9koz3y{qDhX%v^unbq6x4&XMy
z_%B^e&F!f_udrm(K$HaS&I==sFrW3tuE7;|Ei|2Ov_TVER4ie+=Q{6YNgRb;_w^J1
z$7{5%+kV2T#h?MdjT$u%+^-EXv!0^Wg2y4=+%mAqoz<=8)XI$m{U7${9bJJj8di8{
z<u9pZcikSmICA3@>TNb%j4rt8y-ED?E1lsJv22=Oso-71D;Kb2`JRUo?3?3&KK^+(
z`prUns1t<{xio!un{UFoUK;L7&0rd=5krzmVoVx0k2A`K-!+UlWR7sx90o7ichY1p
z-0r*nj2fO=>8A*6IPMod-Lv%;-aUHYRRf(Z)@=fC%ZnU+kWJ)eUfu9*8TmHiZ*TaS
zi_`0=kr-BhZNup}fOqP=d-0T4M0+1{Fj>gHyR>p~v>3olz`>2FT!|Z#P!o9)_VxQ9
zqisvEoZ+Gv95~Y5awEh(vm1fOqnrDq$DuC)s%t0_K4!856||f|{KIB-F>Vmk&=pd)
zt5CjS#aF!d=DCOEB?Ckd?$dbM=9V&!5rZ6XeuTo|0Xm7*GL9<NMAq#{#pOdmlHp~Z
zmw+*rx`Dv%>FIWWDMQ@QS&y|wqhUf;k-H{#Ea7OG<9*gRspeK7Hawl6PCI?1E;a91
zM_5&rYO-Gcgv0rO$~702EJs;k;3Qxjd$)1hdi}HP{O4q0(e7>g3gh+5fskO#iUUS9
zF}I(uug!bsaUpXZEzY_8E93-^#7NtDX$RVZXUOkLPDIUjPv?|U>`iStu&@S=!bH`$
zPs+Q(t)BsqeM$U*$3Q-t*%xr8Mf{cV$CD_cU-5D&LdyB7#q)au!T~}perva%CIv#V
z``V2X>Rs;zbhHKJI@4}HCF-bdH*jo4;lLt6^>JWzWL<J9yT03eTsahP*CA|f*>r!)
za8q>?c9XNu-8R=+Lr9fRecr*&IzLdbyVCIVT%1IS{IXsD8|tRkaic%o^Fjobyy;jl
zP{;{5|APy{9F})k&&=SDhPvI3Mx?f)2?O3*j{ChK-J7l$Dea`q_uQ|M%vUMIPni_X
zVkKy!Ck8}8Qf@r34xRkJme@AkR@yXPn%o{k@5?ZIjme@P;7BE>W^bJ{J*hXlh)G_n
zujd4~)_*H4;j<hE*GH5lK8-~lOABE~n8Ib>U%E71*-@vNMpG2jr_v78cYT&o_}tB(
z5WsV^XaG<9!p_00bXnxEH)(##!TdHMCLe3w#@Y}$Q|ee(X*HKMb<jID`Q%rH*PP<s
zWQEl8H+5gsHFA-|G}FqpQZv&0Ic>AjE_<WvG|xReml)>8Y{U9%Ez1)tC<~OP2}jP;
z;xp!zR4t|gtXVk9ZVK64_GYtEV#mXf_^k4opIp`JhH#8$z<n687E(6J#cCvR&+@Hd
zE8(fysJM6PjjFcjqD9K!Dev1oc$S8fpmes2kR-Wljn)K7?YnS!m<*{(=+Jkz;BwTm
zidBAz<xkia=?T#RsJ~Q`OAz*gG!Fic7y1akzBP27Et`}TM9LUI{ul-g134P*2KIwg
zTvvNUTHQ>-mZoM%5;_L1M^cGmwBepVUhek>sjWzaQa+ytiIOlHzZl5BZu$Mrqu#J9
z@XUZ+f;m;#)kbZt#Zla#C*<^z1ed@2#t`G&+=6fWGDwEOcJ<R;G-()W>HIXZE*sl(
zj*Ly?+3+ilqF{24`C`6Ijtr398&PwiPG@Ga^QbN<3VKaf-<m+YoxQ%8#(&go3FJ{L
zm|VhDHC+k#NVfa}TU|8>trZ8~)K89tus#H)_15(U-N~UZBQIlF7f<%<4&?rG!hmO+
zUN_6=VQG%7uLSIa>ZU1~N^*i9$+ev#_nYqT(tELpgPlJ&zZ8q|^((^kG`F`0-Xtkd
zoJ2yh*nztd=&jOZU68?YQwu<)CeScl&|$U94!vidopJ^g#mptA?Hq6gNf4~O*7bn)
z?BWRZCRPzF%ru9=1)1tRVF|n<qS68OhAEK!)j+C=yW8tPhabAs6E%~LhKt>#S!Uml
zUgsZQ;2Fy5{OYp~ITDB(6A0x?71(kPrQ%Ydzakmh45H^X88!?3v#=~n%dc)5nv}jp
zX4`YYMak)rkxG3S3%{xg652^U5<(UN`zjkwe|74P&=(U9HY*tN+SFH!v>j1G?L#_b
zD$kO5-;s8NBhhLzYJv<QX*pag;6g6!QFA9#2qzji$Ga<kwVw{!&S;F=8pV`F_3)^%
za+FkU`t`>voAZP1ELEFf@dnW64Iqc*Vg#J>t-<e(5X$^gLk6?nHmTU`HB%XHL{0WU
zb7>uvJZfw{UewqB+r{ii(^lZ`AYuIg5)kc~=y6D0r}SBkfla7S)Tx|kYge4HONeO)
z#Ci?=abBx|Mn8N)O@O(~V4%w%^aPifx>{4E>*-|;VdPF@KPr{M5a1s)GN=FtzGl+C
zd5nL!3K@T#fNWLccJH8E-tV@%0T6WV<0h^k58>G&`($xG*8)~B0Xg4`k_ll72A6!2
z&b76v;(gk{+uuB=?LysZ?})n+Jv&J5*ez`dOj~MZ4d$afd+w7W=~d&%C){Se;;C1-
z-cw~L4a9TNgHg^Y&CX$N{YG*fk1&!VU+KT~4r^8)zCmsPAHvKy-XI}{^hhN`@6&wU
zhaY2g|DhJ?Kty=rU;-wQPGlvCgeVT(n>?&RuQiN;o`Ve_jhkSvCXd--vY;Htu8lR>
zM%fe78gQZ2Lt&cyB=$s+9nQ5p@#=o<xdd=2PG)j$yS0tc8zY0xKG-Lm24ds*89%9R
zw@+CyJ-2-(vH{6I_DqDhR?ep!hiRKy?ANO&3VRosBv707rese!u5X3nMZNV{)N2ri
z1<WV#XpV%MKJh;D>hpH>A*&Zwz3phobb(0h<l*AxwC}Z!qWkS@(OVN^HqC^ma9Gwn
z!`@}=K)qn1plzi|zqjN;qa;Ovd~aeVMOl``$*96MUsfRP=6%H7&9v^xifZvlOtDT8
zPhCo3R9$Kjk7sgmJ}z9kiRM;wPI!&|7Irn_v6rICYjXCFY|?RT=37mILK%Vu?Hu>A
z+?o1_h2)xpDpu=V7;)Il?A%RFkqqTYVKvE7FoWtrAk#jgFM-DtP8Tj9SGv*BW<M9)
z4OB-XO+@ORKQQTSrKPQME8r({NvNS-_VhN%#QRQk@c0?sLE4FGnlRo5^0eC7I2;vo
zqW%UG>uwx(9;&H%8+RMfVZ*g9siDGh7#6bA2C*rb7`i1P^GsO2`Qg5ANg+ZzZoR)!
z>n?h&A>T5>s1>%~kHN9st91?awx#cvw;d$pMBxVRZfejfLt7Z#+Hy#8L?LQ+u3o+a
z>2$dt1OeNdllDK;6K#9cmaJxF(@x|qs)^CL{yjvq1n!L3i@=K)k8j@h5+vUySkF`v
z1p+H9##oq$8*yfVKRZ{*<7{C&nt$@Tvlaj89O(dk`~|x{1Dc)f&j}M!i)Ce#Ms@m>
zW?U^A(4iIIm+CE+IFUtkD}r0FVkthfy|mUHNAbXCC;aiGsrMm7=x=jsXTiynzq&bO
zSE)K(!_wxOxa^GCiyi7>jZw<x$%T2JE2pQkvp%hjy=6hE=Jy&2dH2^Pa$4nl8bA1K
z>@Jtdko`UHQ*hKs<&Wg`V%7<zsMxJ(a)zxgO<m;?;$NpJ8(G`kTm%(Pjvoael`OhH
zWs2UvxeRexEn_LY$cz!LfNd}o&R1*jjLRq;xHmXYta_f#-9bkm7VG3F=|I#On_PbM
zn;aP5+iqtmcn_D2qz(CH{;JTQVTm!SWL8oxs*e%0+(O3jSW;iboK_r8MQG`Rm|z3f
zQnP2(?ogK|H3?A1ysda@$Lt8nTqtJ<UbGFo!AQm(<c~+bff+<)s3#iYsb4fTZY1hn
zP%N+;-rP*hZ3}1fI=Ljcr>uB-OGxw{7o6xxkvjw%Zw_43i?<vTFTz^UTwv=Eym|Wm
zJDn>5<N$7e`;AvF7doT=LV+=kO4VBuoh-8<R)(6N3A?Emu`3k-b?BT`vrD-^2b$D`
zJPnqIK||u3Jk^m|8x$K+A=gqH)%{%^X*C_zSoGp29JAUw?9gR95H$Nnw61`9u#-D@
zZ7Vq>&4@(OYh4a%7FDSE`{0;)B7TTq7?Oya_Lrta6YoYnRAnGf)B9{ts_2jimNsaJ
zdOA?t)|%mddpfa$3%&56bzhCtqX`w}vK$p`-cU-r`T{7qTRse8V`zzDfbSoytt9Av
z12i(GI4OR%VzSi2<w$Zw*RX4Wi6)PKQiFAyfA}GNHK`j*9p~doO^=g#zwhQ*eSM2a
z$c4!ACBL%Nvp)J-Ln-X_&BbaimjyeUdYrT|(mZ`XB_`=KPPEK<D=NzACZ*N+%gnng
z<+zE{>CxvJRc{UTMu+?2GnDji+m=_mT4(cm^2+sgL{4=^r3)Ene;yFuM6$b|a;J>l
zg5MrYYcV=QKa!u<$TU$NQ9*={)n#?w%2BznO;jQ)SaN5JdRUaqqy~E8vy+qWDCNS~
zIMYO5Hho>V=J;x`i~dJ+!avym{B92xK~B)5hxF>`e8(-<1M2UJimMDaO>>Rd@JUn>
z?G6!%pn!fxdJ#Ol*XxQ9IB`FtDK&-SJng6LKcJ%I2~&`RfjgbQw=RWy-4C6<_O@Ly
zxO3JY{<gE^9Nan@7fF9Tz6EN(p`xPKJ>ck{w}FG|c;JyQ#r!dqsPsgMy}v8meOs(2
z)MXKu4{m2jyBf$WW;`3v|L-B(^QF*>6t2_{HwmmwXkFJi=c|3!ru;w<`~o|@Uzlm2
z{$k3v+d;FjAqcsC_q%nTF?Ny9eE2ofE82x)e|Ndqp#1Ds9u6X%7VGJ91yn3sYw>iE
zncIcIuK9$&$i0zQ(>XdL9_d6*u#>9uJyqvs_QfMWz*_M%B2&NYP@S~Zgn@shzMIR`
zI-%P<wPN90siD=Y`CP3=7Pa?PdwHYZrYxO3n-Xdl|L9}%@;U5oPW3i>5NYuR?ZUMr
z7e8BlEFHm2K23A{kKJSqWlP#bTD+gb<vnGp<Z6{CLN9{Y_gwwu_U#k<J=UzW=MGqI
zZF&z*7H3FuJ8x9q+FU+ho2?g{K~*w&*$&4aKoeOktvHOZyYC<a@%g<ew(p<D+Ddfn
zUF&OszF(*DnA@F-J}Mj5vt5{TWLY17!K8k0F_mzHQhW_g>@{ZBj1#1;@~z`1bD(g;
zl*kwE=^;qlZw-Za1b05sFPEf9{J)<Q%y4}!>Ef50Ikqs{U6b_rAb4WBVO0J5`p>eF
z=q%$5AH{m>WNYhh(mEcpb4Kg<3#QQ<Chgt8ZyjBukl3C1o7+@Ze8b+E2Mpj<`(<85
zc$#)VBQG^Yt<?^Fl|WeM%BTB|X6(Te5VlBI70D&W>iz3)E{Ff79)Q310PJcYc9_u`
zYAUv*+H^IhH{^BHPA<;4_U3@m<dPt=`&&bt8n2+(H{!i)J?<C#wL3+JG<~4^i%xwG
z^&4&<_p^Rl*SmPU0cl?}koU44Z|x9BmfhLFGjK5xBT^V|e_Wn|$rv!(-dK}IW|Kqt
zah8R{)tD?Xy7QI9daebxHX$=)k0~Jfc_OD}V=W;~{@6~Hg{66BtPR`PH$h7$!|*Mk
zo%u+ipR&5zCtYS0TlYotT1aJK>9PPe9+tj?(RH+}MkOrhZ=A<F0-VjIT#_ve{+*@d
znUX~|CpT^m85n+pT{Ora=%v&IIv0j;-##JI<<pX`WLHqn>g~~A<60yDRAS&J7f_>f
z0vC>(MxAB?upDINdG)0q%akqoqul4&d8s0jNYz<fcZN`s*i4V}d5k!YEZeAI#niNe
z!;#LDH{smi(j&f_O{+iDUkgWP<AOp$uD5l!bJdP_o*u*??pn<HB-O>?VQ@{Fm%&4n
zV1li<9!D8#7W=PicC8gxQ|?gTKl3_wyT*v~3c@yDh_!qm>t0Zj68C*ACuVIV-pPw1
z9hHG>uH#fAgeQ`1umZ5NzfawHbR7-t>I!J=^`sEYsOxY5lV!i~yf6~Es=$_AC%kgw
zroF^a;QxN!->hD&r_&PKHw%4j@z0E~)xA3k_a5DV-n-wPRW%U>a&X#RjHNXsfnz>)
zXNs@Li4boKu11JEt%UO9WQK-z&DA;+Fiz>XE&1zvXK4a%<2sB?JBjXBS37v9v1Oat
zw;oTlc5bz6slKvGF76DU(?c*Yf-y!rZzb7^F><=N3L%Vw<ou6;nhB!lm6xvLd;pDy
z$Lh@0UKxznecVD;6ZuhlK_?59M)NDnm1LZ2FCb?PMtZWl`o_a{g*XQ1;p=>>A}y*<
z;}(;7f3^=LaJ6R+W-#J%W*qjUoP;B(Ik~YZ$iUcnpjP&EHY968Rc5${237H<>0_pH
z=D4bLMqi?Xg)x$y{=1q@Z$(c}pR34HVlp#-O$e|rBxdePt#GWY<JH(|NtT8{3nP=w
z!Rv{@N1$_ld3u*$13Wl_r#uf6+nN9Ug5R{``$K@T0$cNHOtJ6K6cV)c*nmW@c|XvL
zC?wiWs;c%!r)zpQt^2bAqcldsJ&KK?$CGPV0^ix}yE%{QxXy&*sW;ji9H~uo1?4zQ
z7`HX7bY`gN>7Us<`D4zOsnHS)@67XD(k6rhfxpk*q!y4Xa^Bm;fQgUbi8U-0I9)Oj
z8yN{psJ~p*!#GmdF`)n)A0&-H42*qutBg>*g+Dj*=Z)mxs98gw)pFG{WrLqa=}!Bz
z>d@OQdFL@@)rWLaluiyh5qDBQY`s>fYOl8QTY>MiO23I-ZT`~NrKd>~=<a5P%U%5;
z#F@RUh=U{oWW|sq$wq2T#T@PSHRq;HfeKh^7J*vBnPvA$_$ay1{M}4l?M!qYzV(3h
zPzN$bqruXj%vw8t-*#Z^R2!q7yRO~gC=MMx@C+Ru;~-Rs^3)y5JwF`^*G<f!wY47g
zw3IG)=_G5YG7iH5Zxu6ULo+lx=6-{SuY=h4uivH&A6umi7iMDoYphp&UAkLK{Gc(*
zW1{+^gx~I(CYG98CdwwP(fQoz8ipKW7M?#f5?VwRVp{^9SF6a?;e|N!lvKC_;%*}J
zKY>sdp#=5>N7SYr*7f{FBv0VBw-KH?d;nQF84T-W|7lzp?qYJo&CiwU?*5CeT6Y!)
zelUP}waNbVYSVUwfSA;UP`TfZl3$>}je$Gjex{A4{U0%FDc+0DA3#E~@9i0THrFzV
zOIXi|c_GQDP);3a6D{GV*uH~H`?zDWT9-Y_vp{<9*&XrUoo@7@(EH1Qm$K7ui&T40
z3+e6SsV@JtKKCjPKW_Xz6OpTS+rV)N>4o3V*wnEVd~Q8l%*WBvDZvOcrMG<jWFjaW
zpRG)FN`ZMuTE|vi3rbZ#354$*)774*u4}GGVch9n7$rvFW13T2?`zEqzperAyJG-8
z@CW~#QxvV|Zs}nIH*SK{Kh3Xk?@Fj+bz`;$_ay~{j#FIcDb9?rzLW;I>`rnqwH-+Z
z9g5P45EacIrOXk%BIF>!O|0MiT(VL^JAbev-R*RfkWcyvKo%i^&C6WN729KxFS)qr
zw{opz2q2ZvSxTqIb`e}rHntqIZKrciS&-uBgTEW--~CO6D_SU>>M7O<9hKV<oseTo
zGyg>oYEYJhoT#Q8pJ-o6-`gbCI+QXLyJ=#RVMwgY9602PzqbB7Q>ey~CuMvOtUY@E
zNn7|P@T&EK>a?v~lGjY%yH6b|FN`Y90X!|VeZl#RwN#Ha>l8UB1IhU(6q61#OOA2w
zjCz#c9zm^c`AYop?#Wx+4uoD<%V>D+1kN50>uGVVE2gU_|2PdxOQC3DYV>jp!TkJ!
zoITY=1C*cayK#(HKx81ctY{VX;OZ1}thyJ;L;{%USRQCrG=j_wRyM&`8Ao>??LXR@
z${|E8P>edu&lv27>BhHKS2<9YHT(Qtd>?;abEwsYx*KC}+K4or$NRl@M4&&lLZjm4
zR}+2xby!EBZyI4zXh6w&&R(|4Lx&hRYOW8KVjRuplX)qR@*-KiYj3I1WYXO8q*_n8
z-qF3+7=WB>T|`c!w(K7r3`As;@@(1PVdE0sEt5ArN;5dT46fF?6{T>Mk0$he)!&^=
zJh@W-V}O<sE{9a&YJ!w1aq%<GejnpmH>!?bw!3cq9omZ_Rwx>I0h6ne^e2TEk|bR-
zjG2%q)bsQ}!btJ~uA14oxs|r$*R$VN#G>q1#YXVF<@qsu4$*&dd)!`pDb%bY7K_T%
zAgmTh4ur(F{q`~F;WohW6xo7DuU+IYjdqsj<argoHyW^eXO6^+mg*}S8)JHXX}2gZ
zaDvO4?w?kE#g;O4y<>s71i}s+LK0md@A`x*I7L6o)+HA)&<qS7Y4;Ux*jo&**`*Bq
z@^9on)PP)ctX$n8KwAa-8#(G#>tv0YiLO>ZmFkl7Y|rnr_Pxp)aM(cUF>9<}QaLhx
zN54vd!zp|ND6oAGn&_ZXACe9of=I9pys8%%3rT31SQw~QyYL@Yl9b>=lrzrcq$nId
z2G|T4ZDpUgV}D$Iml!kk$$GX#f;F~Rb*rDI=f$A@lrW+sFR!J;JE8C6!wA#T?Y7=0
zf&{wSO6JEBD#;Pr*2j{SLfPEwkd*@aj;n``dKU^kEsn!)k5jrx9$O_7tr^LIt2k>u
zv#UrwX0$z43{ZX|xGpKTURjRboy#DO{weV$^MEyLl-%XbnDHVUfdbtf2KuCVp}Xe}
z-rty!6wwyHsDn<_20U~+^yQ|DyQv4$67!?OR@^TkpcjZw5J9hC&lWe0BiIkf-0PK%
z%VwrgVqRtvRr!h5HzT2GD<v2S!)%}6S@7-L+{lsXWRM*@M>tjBvD31h)V+YrkEkVs
zDuaDMpJA9FgIC$CPu50~;vAm)P^=p7ka~kQ$%9DZTX%V}ks!N&4;cRn+WTw220;|a
zpMkf#i!Xd+yYUB@bn1Tt60-!>^gRQBiI-ZL5*!IRSsYqL3@LIdWpa$0t8q!2e*Jw;
zT5$|G;Be)Uew<grRbL1Ys{}ajb#019YSVJ1ue5V{bOa@{orJ|rPwT}dh;9@1ldjUQ
z*lE-8b{Nd(E6>Y+TKV>rWsR;`Wf()5#7J3v)~qeG@p+Z50MDa{9R8!v?K;%Ml10#@
z(e#)&PFLGB#DdjwXx<1bCvYGHTit%naVxv;;ire*>~!1!(eX`cL6Kz-^7C1&iBY+u
z>(QAr$CM(l8l4mY$zc9{MX5KaLpEy1<d|*-(AZ~U&vo4DmT+4fjRfBUExY+<<5VRi
zn@$J5CQwri9<ef5%2Jn2xW*Ue-q}tjYns}cOoc8aW;}3sB_bB3EOk`JNx^S&^ZARR
zo(8ynV@?0&lL&60h@*d$y@(g}M1tLKpXD8McWROybHS0$M|J#Db^|d^E5>Q5Aage@
z*G{&6R2sy3U}9yTnc5~*Zg;(x=dOfw|2JviX?6F-zNA-=5}5d0N6yvJbwGG)zxO!5
zxRJ7vxEZ~PnHz*(au}sx^>w`acf#CXcjL__o?b$Q5ZM`<#vGxRLp+AD#%VU+XVN+I
z;IT9P{%y`!wmXk2lOrvEaECQ(Cuz=hJYlY1l>j@Av0=7FEF0KY#JE{))0rbxMM{X^
zQ$odQ<gB4^_;iInvemt)-u3o@b1qBPx-aXV43^`bZnvnuh26?nGtU)qOY^H*N<CWq
zfsE%SPb~!RcJ?MgQ6r*3AMTgeFmZU8tcNMO&VLKCz7vx04X$2S#Kfpu1I*CwqzDE;
za|hJ8*&@C|Fnbj63hE{pal>g6mfn}JVI@@NIczYXCpW`FXw<)uJEN0Qv{26}S1hZW
zsXQYvTA<gks`=+F(9FRu-j`5%WZ9s!_p=SF(uEwSl`~T{_IEZB{0y$eXvng$@xj3s
zzA^q*1IZ)1m4rg7<T$&lt!!Kq6pCj+gfzB1>;=~=1o(>kx(cro{iX6x{n@mo$C5vJ
zi_{o~CnL6Qf9FkoH<2kFSJId1dA_d?zF;(3<}N9x=H$lA-Gv$z7FEK>pR?ss-fP%m
zI4J?O;PQW>*cgf)d(xDASFFVBwBFSz>Zg_vXmO0EbN&rgNR89+V2(b*l=mq%&_L2j
z$-bg3+Us&HPMktAoR%im!EvpbcY7O6pW5lHe%v4)i*t^l+^wEuKs_OL;;GS*2Y&P@
zQP|aPKS>R;DH1Y2>R*mGpd<Or>*N9Y4QTrx59;VHL2@02X~L9_8Pv$T39<Rnpc_4K
zxYu(4)QPU;T`CUa^0Bs_oT-U4U{8#CI>aSiWc10DB7EAXvX#4kcxp7G#(UNJzYZEk
zjeWD(JxYij&ZMlup%joOJGHVp_Vu)6(WW@#&=R7h5T$Hf(Qw1Pt`jq~+-G?b((tT!
z%ds%g)iNV1MNSzN#x_<i?MQR|Noj@D$qVBGImsk-8fMMjyh+s^8oO=Gree64D`wyN
z`Eh1JxzZzlO{s`QE`+Vh!H5}kjgOuFbn&vWeXPgY@qjUhLR9R8LkANE_}3Huzb}nN
z_7t|>`(1slCU~I=So8O=w*QFwvD!oH`BJGr>kp`K-8<gD6D%Ut^i<Up(hslfWA+;1
zPs;y-p>TWlp4VEg;=|UCn1zm0Bd<m*56rgFg?Z(inH|r`7y;d0?1&DmNu&AvO2NT+
z&<s)~l?9uKk6cBw7Q9E2T!z+AkKtb%NX>Upo~|e}s=A3_?uEd9u&%A)yDc_1-d}{@
z4?^#cp*surw+pAb_oup7Y4?|Db|s+f3~Vm};N9O3eUaUbb&mallDc>3xYq8y-Mt6j
z?B-ik8__X(-9p<rzeWqxd9>ZvVm$6HH=qDFChW!aNyBiW1ymLxyEvf76nZ^nH{Oc*
zHaQHVo0}Mt1kuFH{|1_T{w{@WNsTxn@xJgShTa^e19-fFqJjx`-&u<W6HHSn5KlHF
z4*RU0tsM_QbxA}H#Q^D|3z1}KEHC?uMg2e;vTZT7vmXX~?Y99`HjlXNFW*qOBHo^g
zRV|9Onx?_{qG=YLw=&}bZVwR#0B+~i*Ra@xznG)YPC;4AABir%H+9hjoZrEM4IsUP
z%agqs)lM4M<i<mPjvVbemzQ98S0D)A&h(yz&gXEGB01p?3I1rPNX-TzA%GPXzQySG
z@E@#;0nuwKZTRKdPok#VJ=>i&Qh7q|OzLD5@KM8hqUHM}rMoD3Q9&<RgbQT(8~uA>
zjOJtkQwt|Bv4sK{fy@7tK7qfbZ))erS1d~2dsKtRm)mH69WC0G&&3ZP;?E>nvmWUW
z>0qDD1)bcNg?V3op&2Pn&~OpbWUEciBCSo$$<j{Rx?u0bCD@RH^bZ^I(2j(h+M0J)
zUScTQchA?1^H@3^e3YNR2{pEk=^6V%SyK2YIk(&IR$dU5B!b<ADRFLcTd=>)3<m!H
zmzF+qNfL`!Rzaib4`O}zGH=oxA0jp)W)d0-<04;WDgOn|N6LtLN#wF8zb)!`nwUS{
z@uMKtdNx;(mEU|mku9;r>{sJ)%NLpn{Q_cZVMTILKApLFwVX>!evAI;LJef@2Pzlk
z91%mua35LGKW#2PJ?N)YQE&2kqp9@TJra06Cn^N_^gm_QG&inb;RYM8de+}v{yt|*
zJbAZ`dQP`IfZ@olm~T&)gE@$MyN}dv?2<%b9-r-ij(}+I%Q){L_l*>CIX_%KJ=h8z
zXkOp%4-%&Y2>63wBR<BCKO_=G5FV(b+qoYNu0}wczsGF^F{TdQUwhxn%O>#=r4#2f
zlR=4b{@%RU5lfDG(ZS^g>08Hf*nKGWI&k%06!fKb!B;?cF>KFr>De^~4Rt&Bj2n2O
zzqL~C)lB#4?YQvvVOiR(3`X-n;x0VslyEh6kc$f-Ob@iNZ^2`HK-c-hs4DCSIZ}Ka
zApj7H6*IyCm{9N-1HNGmuJ<L8&jL^YYxKLJjXMl79xZyDFyJTOaw#<d@R_k~(#}qC
z(TKn{{-venTB@SNfqqZhg{T*h!OMT`@nj2|xAfZ&ob+fAeX7TH?q^(ZyILvm@a6?j
zZB6gv^2&YP!Yz9>>?l~5P)rEpE|L>v$B2*r|I--6`}Z^gY1Q}J(G_-w;Xpyp5)aiR
z2W3Fw&38~ivbv#CL0)ePL~^Gp&SJOzt;LLYEc^Bu70tw!QOe+&8g^o~Un?Tf$)+eu
zsLZv@U+w^&3KvBWP?mXMR4>M{Va#f~z&Cw8@8P?ic}4w{i@g5*(dEgGC1xt>>Yr%=
ze;B+kNJojehMF*=mNyGS(%*C*jC|VT(1*cL2n3kBBHAgpdJ+h<=6M{aJl-J^+pE>6
z=s0mO>ibJ|Gg3kNWODPP^Kz$T--Oby_07`KVCRKz{uUDl+H=zz+H>=!0XF+bS#k}g
zAsCUZqOKL>z3eo}h`df>0Us<oKgPe#a8YCqV)c`@@`je8Ml3vBCx?-ZX4uPXRRm%(
z5!k^#Li%fdgp__`b7|;c>!$+h{lx~x*xa*c%;yc{R=<l^a{u^_>tpJPxan$8at%5p
z125u_25$V)u|D8E0@%Z*U(yWZ{R_L`Hvif908j!j#HzKZZI)>KAlv`_$MZQS)+WG9
z@A9=Q+62$^so64d3APB(4CH-tblClLn4TR$2%QBAMbG&#kafK<!*XM<S^WV}!wXlA
z5QwJQ0f`+8FI>vGG3bJkfKRi)VKD5Wm4YjPCRU=^gpgOC!_{j5t%+b5hk~1uYY$Xp
zwp;3G^#Wo@({a<=ND_eirBCU3#9p3uqCX{+=3!OCQeW=JHJ)VZ$^M4@69^&)L;}3T
zc%<PCTIEDgLI6uwut+4U@#Q*y69s|;1;>-70j_1IdRQ<KxR;7dpt^4)5+Vpxq#Heh
zz4Wheq4;`lZ~=e4iFoc@IRE<+{GVf|5Z9tu-XdLdXw{Qhfs5+|_)uy${$<)oLw~@f
zu%OFQ$y}m!L9;*;mvHRJRWxTMwLwC*mRG5aRx<BnHtA@&ArE#U<JM|_cFvl#>1KZM
zjhOZZ(MSf(POA^aZ*@QfjtLQO03<FpZ7asd|4HlQN!dADe4F?YZwHbxb`*vc#JBjL
zHnAG(?_Ky;E*`n`=V2mWYCd;fmx}q^Rj7u+_nUt0-LVcuvlWP{1zhkH<{Cv@XZV%a
z^6mS;IlgimFO(RK%dw?UM&IX^C86E%KSJeuOf(}F>2lSXal{pzf#sB0t(rubtQn%g
zpP0~>5=?s47b-`~&;>hK&6wHkfNtP+aFxal-!JE!BmF|j?Z?>JX;s(scS2&Fv$GSk
z;JtCfLa$cmX_)WH)pu(a5@|-1by)rvb@#1($Tzi`>e_-x^IibGm}+S&;u~(f;S__k
zg*wi5npn^X9d+6-V0wvDv@1(<z#Aku+xUPnJ&HTBNRu~8+B6M*wN>&VF0-~*vN}@K
z*dUEwMHw;Fo4|wFp=d|U3)=#VJ+hp`XjPsyGa#OPOyJHa;Snt*PIc&E{+axHpz(cd
zC0dGazSFOr#y?T$NY+m5xmc8IW!%IRUd?Ck)ejyB)n@Qk1nsFLJ6ODZnBeh)8{n`$
zfL#x?!;RFZPN&j4S`7rm9w|z$dSJ@2pK(<=O{_S8%Q>6qQdCe{Ei$T##JMx=%rGPp
z1V}b61@lO#Ae>s1B+tsQcrWdNbnm^kKWXRCbGP*nG@sheD|)Ysf+ue{fdGvBrjZo=
z9@8~iQDc-6|4M59fN=zbr@n^)+|-`$(h@KL%Ivl?n4euxJuc`Lo{_fnOsK7IP=UHt
zWG6z-J!$TjYb=ghC@hZ9QbLV0ctWKkTsXW$RxUP0hQet%E5fP24$khQOl-9ZwPV(@
zzN~*SjAG8_H_0oTrU1%bs{#@MD2);dv2~ktbB_MR5aipe*T2PhGQWNA#u3DrZ}+34
zU#c3p9r~E4WF}w@wq?{1q?<1yso${c(jrXmUB|a_G;wyAu!AM%O!P{Ky^Qkz2-WxL
zVF!z2flL;Uf~`9@rL8+&Fj4>Hpc)ie#spYQi>)tJ5f74Ugz`2;3N`itXfES)a9&;S
ze4Ak_@Q1eDG<xlD!j@w7g;pz<fM)Q1#F_$TlIB5N8Dl3(qOHtR;jX$3r3U#utw9xl
zzK}6<(-OHM+>}TTRU=B2C?2nZ1w>?T34`K9Al}=O-#iLLey5Vw0*ZifAGYZa8ImkD
z-Oe;U91Qdivugt~0owr4)Q#dAzq`@Be<8=(gT`yD%?!e7Y!<xEc?IHl{RO6jUAL_Q
zY+2%|pWs~*<9oe=bR`uu&g;?x$Npr&m;KfDEAMXs5&&pR6eIpw;#<qFV_~-$YWSs`
zwS>3ZZDrW91-Wg%l79z+5QG$1jgbW7|HMwTbP?n;JL`MD8CQ<Q!<NfsRCy10eQLPH
z%~-2A?|IO`KO04U^#Bn4rWhP)ESllPIj&F;U4HzKT<fveD|f8nsiffN32Rl+&+X*A
zrk@<{(fK(skpL3VC4tTjTkQL|jwG9zg_E|J!vteo{pg7?UEBXVgcpkxj?a4!{Gi<u
zJcY=-owAhl5(NY^H2<K#y3Eq@huHFS3c$uFAyD%S_}~t6$|ZVWZ1s}BX*rn9$@swX
z1Qp4@Ik0k<6O0k7tfwSy`}Lo-z%M_*Y`f%7TXs{YEjU^~$we2kRFw;YeRWUszhI2k
zYv%Z-<A~RuT)@=i<t~T$5*e}*g24wN%eI7%?p{B=4UWTjA^%32`BGyX$F84xB3QXa
z;>Dsav!~a;Dk?>fTdQTSNt5R}b-EbKEG^@5G$x^*UbR!XY_}gtpWOImde|6fyO;K<
z5}cd|;qWqmheuY(clw{8ocb0P#`UWx4<AHS(q<TjkXRGBQRg~>5`GKg^bL{zS|?j>
zX#6xFqhWp&r!gOl)x-(+IePD?SwzkV9w#7O-{}sjo;t36vwhZqRXC@f71C$uRTfo4
z*dFnwuqzzCiO(`%=S4EurqQ7zai&&#+Bf4{<MSpWklNuNz>D8Bow&(%@r2`|!@!f0
zn<&^<1Zch#As?&j&`e|x@Z7yU&Mb=#ub-3-T6>mHdtl5V+<Vt?e|K<Cyy%}<Ma&D#
zNASqyX6~x(O=5TJ$izP2$yU&LblCNdZ&#`<1Ap*CP__<0R+mp&y({X(jo%UQv^iX;
zW3g|oE#^=wk};if@Z0GaDjY-?8KpHvj`v7XZf~Ya6(dA<(N+K9fe};%LVk21xJeJ<
znr@5%IcmC*tmCq=o5=I0J8#5Z)Rf3vhVb-18vg{hN+)i^jlkkAMLtbCd+9C4pZ+44
z4S`*AMJroSwCXFO6>!2jjg^k4cUr?9i*2typ0q!N38L5O!tha{zL-|-Zy4#{sM1Rk
zx@`j_hgaGUSOi}XWf=*Q!SuXlTTgj1qHvUIM|JxTp18B@hOqN8fohI>XOxppmD0(c
zYEv$;?R>T>`RUMRW~=kuV%lc3P4><Q?4)d0Q6-Yb+jO#hwa3KgUF)pT-EmYAM&S(j
zS<DZU=gsJTG-P-QJ?y8TT3wBQ#cBYwBSV<@?CCOHF-C6c)?>D7s6+pI#&}m@#(lqv
zX359+>O*(Roi3lo?9TqI?-P7`v-nJf%g!vUkASS`mFcI{J}PGe><0;8Ts&iC^Wy<I
zD(Qnl4uu(vjANAz+xfQ-ie8h1$jfYINT~kw5$vG7>ZId4&GZ)y8%JH0Og{eDNs({L
zF|CL^2OO`mwI1W1FArZ-I>c(y3hyQ6mUB~<HAS=G_2vq$v+lTt7)Fb)N<p0-K6h_9
zf;4}-^`h3m@a3*A;cd8X!2QqMr8hrMgq{$W;5@<-DV-(#_-u#5&<6$s(20R?=k$i?
ziS@#jq<~zFl2P@O@_2(Mw+YqmGyr~Z{^^LS!jE8}YzJ*~oQQhHwiqb`u{|?~SU6Hi
z7eKl8=lvrwdim{ekXWJ!05uw(S)Q6&hWgG*CL0S#a~hTccGcV*VP7baoS7nbpAUXZ
z?!rbf14v9bUZd0+cH3~vk!>;9KT5_O9+Rv|$N`k(vK}rXd&$K<Q6j$F&*BHt>sP!M
zCDYek?tLjYO7yvV67yr<#-brvqmUxjTGFF#T<^20D^FjCVW?VIk_PX2TK(StZqe|#
z9wy!cGT0eBqhEo)6Z5t@sADH;xl%q(u5tFYf$bS<4<6yrjGc<49QGVp3>5~!c^5T^
zJd49sY<2ZNr7%BR(>%?tIv9KiV;@Q2NXyN?hRegqPxwC;Yk5VL3E3isqbn6r;qnhw
zWqk^A*=5GPR@hew;>(fo9B0aPfqRW)MmXu_eP}6*fcXFOdHSO*WO@9xkL$TH4)28R
zV8&X0#joFBgkWsn^1zZ|fm2Q{-<t&-ZxQm{P=f2eT;D!Rm2j;rt^9GD!4#Ou_O}^{
z&BMwS=dMC==PuprsardmX$wF2;qr`!ZL3^Nv<(lS9(5(NMEV@~So2=rANLxg>^rl#
z!FME2Gr+s<y`;uZ6Iv?aab#p47!7|D#`L2oxn(x+#4@c0*9;l>)MP7f0LZxq{0E3Q
z5+q%J6fMIz3~(y>LVAePOfSFuB&<^8eAW>==e>e+I>o_|3@y$xY*`N6we@BSOvA1W
zP$A?61~D}=t%gA4H0MYMjpSb%uQ^hWhMr8JzA2OBvj=W{hNGPZKMGNQFJnxn-~kW|
zBAcG%|C|6aK+deNc7f(<E->BKmq0>&5>PgQTPAh%lPUd}_V*#lr4O*N)Ub#iwOa&%
zjt1G(zDmPKleiz_s_WevCvX5p*%B`lBxCzAC@C@Cca#ql7H|U`QS)sej;6Uh{)L6h
zpXvE`MFaWmtN(oqJ^mY*+@;gmf#79j6ndH7Ta<PDfKL6w2*Jz`hQ9sC>^0u@tRZh&
zcTr&%;B{$L!515Ac9=17x)FK4vs|>GG{R1vCw0!_+?^k{ea1g+)`fnj(PUV`f|7mG
zgTAfN-fqYUzvH|Lzmk}+-L|zADR$-W<!dkC229!27BCxug9y_3>g#Xc*ckgi;c%!7
zj3?Ot!#cWQI8S&j^M3~H(y?SEs(|I6b~bj1%-D5mWrSU0UJUk5dNr@_a$bc}oQOCQ
zJ%L8IDtd}a8SM3wXujK;M@okir4)`}LS2t(6VpNUyxL8h=rD{F3E9tA1$FY2K6W)x
zHn==y>rJFyB=fHa@S<Z))FS_6OSgCh)w>1SSJdE_P)w#W=e6sb+tn}!0AyFCgxF1+
zT$fW_A0rU9qR8J*QIySqHL{P-IB}|jeN2)W7&8sbTR(N?=KRJrtZB{YJg1yObz-g^
zC$jukWFK{fcD8_nfW+j{PtZe=Lq^D%?bWdQ>P{f=l2*27H;v?{$kDKR{MBb%F-1aN
zg>os}2p5bpFK|Q+sDv<!J;I{ABxXb@(wGUkPw&T;n|0jv;%(BZ(FuA5`nm{IKl<F4
z>@!>9@QSqHPql(*{2OF6)o0l$^>u03VJk~F{1bo)7tj7ZhqTtqX2@b9UwTRe{plNd
z%>R`GK33zM^8U{LOp_OwxcX$mr_uTT_8iw+j4iG0rR=hjEBbPFL$r#@Op%NuKPo$J
zIJJB2zkpwtzxjy|(haxF|0sS|d-V)%&)*(stOZsu4mVt?i}A9WeU{ez@FCuwnW}LD
znUzNF#5`3lG;}lFQK#`eqm!t*X@f%$oO6@B))bAMSZ?#9a=r2y!*ob@JHmudx9E*9
zS<dl{_Gs<FUSjR4wKdDs?@z{90Fz=xeFKTqa!<BnEzP?QxszfHFp>7Xo8fOFA13{P
z)+!w@lH?d-?dOd5=EWd#;el9hO#W=qF3-S>NL5+6qPR)nmmN5d(?N;*_T`xdQF4V8
zNvK*5F)Vv$O^To*4=J(co|HW)ZF58*?M46%!|107Nf)}xzfPE(G#PKS@o<ZCJtVXP
z1<D2l0!%m>FAOG|I+p~&O{<*bw%7$QObg@2HR$(Qaq3=D>ue9$TOxzpc)UY}J6B@G
zGye;3@WC@6uOKxpzd)s#(4Z5v2UFBC%gPfpc?Ab?K6l;t#}K3Pu=}d(+oZ5$JD5FR
zP*hcy?M55tg^QrS?bmFi+5sMq%T(=SC@7939vi?HVr8uO48v;S-3OEOB;bbMcxK}J
z7|iN}OENMTzG#ju5_Eij>wT{)(1j~l2Wo!Sd@(fu=TmbK%p{`;6anG^#LnV|iMxWU
zyI#=sX3GMKcXxiXMUp%JCeaIJsGk6$YeT37xe<=d413?|0SSJ%qnH+GyN86Fm^Uc6
z_f~X0TJ#_Ge2f#G_&<B3zDi?q5nn=SV&7y>CE^=fun_5yH8YIqQ)me4MQ7@evD@nB
z3E;)L`Ega9^RbO6jd*OcLYjV(ainG_>56J&XydFM+&gucY1rWUF0u4P^DkyC>d_}+
z7BLOKkz;w>sdaXdIAIjI%9;~-A&>+$`BJlStiCywKs&Cr^jIa9ld}Il6MRPL>Lu^j
zznCA(jjuB+$55%}jc%7u{op9GKsPK>7Q{{ATj%y#ig)xF98CtRM;WG>>9_mKT5{i*
z!^exizQ23-aAb|F?wtr}lc*SJ<`X)s>Lmb`ocL|F_f<AP%nlcWBmqrqhMZnOQcJ%&
zOfv(%T_h*=aLYFn#wr3B#qiU+s{thr%PeH?t1v*);DbS|(GtQRqWc$RAEIF12&&lR
zVUy&?C>|9hun$u|LL#=`!HtAW99MwTzC;Bpq$G#?AX4-eGrSodyhM;S6I8c3JbHv%
zj)gxZj_2;7BRafDn@O|SulU92DK^-&#Xw=zN&Bo4tLMUr0QV`-qBQRcsS)lQ5=me!
z-XNeJ=cKoacuPUojMgy%3LJVdlY8KJ8USO|sMJ|@8Hwmw0rYfJ7+6ad!D)S9j1Yl1
zF`c#lz=2XmFqkV56Z<WH&_sa6qCDtWmlYX0Kx=1C_P&@DteS%V>QdPL=6DNqzpb|t
z6f@+1Q9ztsXHqD%OO=PDLYP!e@0Z%C9R27Hu+=zAgUbAWgq?L*l<n5`XBYu#6(p2U
zk&>ai#{fiH1cuHb1f?YghT#DOY3VKrrMpYILAq<`?vC%~dG~(by^n9d-}kS7;Bnmd
zJ=eO{TIcyY1%fRyS#jOeGY>^y_W;=AgKE8B)w3u$qe#U+6I*fNg$3ReK;eM?D20(0
z#OD4<_3q+xnR@<dytEb=6)fEs*E5u>K(ko1m~p*Nrn>(hn=8u-*kGrXodUoSgUZ^N
zyBkGrs}Etd<xKm~O=P{ZMCB-brR#;0TZ}%M>XEQG&}>be7|Mv}g7L)r>98{SB3GiF
zq)d*$v4<*TGQ<uB#0~GUPa!oS#6BInThhc>hEyg~`7(6l_vz*q_rTuG(jEwNwn*t2
z2j(*1!djphW+gO0yxn{HlpW^7TPw~93PrutpsrBj9{`2hDBc(lW2|ZR(+R!oghFuF
zIA_00d<uTNS7D1LCf9Dbk9<`a|DNgigYD1zBT_TNAyGX`onq!s96N(w5yOL;nhiQ0
z$Jm|Tb67rL8wy6`QNQ=f`9)yfI2D#xPUP^Mrc+_M!{5GO0BaTF(ge@b_$Vh{a{uV$
z%Wut25`1El5Ha*4fzAbU=UP0>J`>TRV(j7lbMU7(=K+XCNgXP?=7aM##KFbw?|kE5
zf9LOSf9ENH-abjd+Vsgi5F9Q}(4NE9V186!;SdyM0C?Vs?re9D5q#--b`&S^ww`v6
zBH?O1T{ur#C1L0YU~=8$AGXK2k4`=|v9N35m-Xoj9n$1#>iQ<Snf{^689CmgR{^7M
z3m)cI2qwIcAXs`Lx?$`{H|7CsOQ4tOA68BOLpx*e8W<C{P%;}kXLvx4HNT$Sll+b-
zh_~&z#C0;BP4i$DuQk>$e$h9efO=AFt){cwL%j!8&|g+FiXvUK^`Cx{_+7&dZ8DIA
z+SpsTFxC_mAq^^%t!rASK<yYzCm<JtMLsLohz38FAmu;>4IbDwp^^?A8v`%@;5pG5
z4MSkyw||Jnb~@fGMP?Lb-Uq#<&J-lcvzh5#OfV+dxt^jp2yPPG+oZl|+ukP6l%6&s
zU>!B4S`eMCkCwX{AKbeTgsRW5KohXz=T^c4q6TgltpKgq|D3Lde?N8Wev$@uQx3|(
z;j9Ag{RK2QM$OfZ8_~hY2MK_BCR96r*Nq)lrLeDr_f*HH8&#JL;pxO}Wb3&92(ltl
zk*ut3e%lwGAHZT5Ar!RMN6uT>5Sg#jQJ$J5S4`eBlxzbjMO;Kq<I949`cGi$r)&K~
z4s15zhR+N6Q}1*Ov;zpr{Ov$1;*bec-DgcW1iV15;i{4QW-(Nyv`2$ls`VIze$Fa(
zS{Rs%J*;|M*D(plXw{E5eu3mfo!a#WV{@AHc4=zDg-K}p?e{A_>nw`nptpC-w-Z$?
zuFR_8$_|cWnBUFx)F-q)oQa8HZFY{qD}C0~tp^J;U0aLu7$>elx&jm1IbtNKaz<hA
zUM0nGLQ!8hL)z_n{+(ky9l~$vj?L1zv*KCfJrq0cSkehuo3rAYaLcuBkRbJE8}X^8
zine*fdO_%s!dJWFl0T&d1upfr?zeT7#VtfT=wFAFX_2?+d4nWGnN$RV-#jKyP4E;*
zW}-*^F^$RofTwO+l4lv=!m*wk*E*dli_}RE6HWkt?!7~2I{AZE7j-|m2xk6#O_+qh
zbIW)wY>%-z$Gg6F3{T`x_8U`8q)KK)2hpm+g%R^CD*6PFAFB_;#gH-B$r^Lv_*Vh{
zVOC^01{++aQk{rb{cIN<$N-5oAgT6{RPbDisa@w&t)nY0C@I^OiTDfXi_<OZKB#(g
z!4BL&3yXb)kAXYPBweyh{n9#hTOe4K`s)`v$4tRQpA+jTyQ2%ic$LlN`AT6Vs%5)!
ze{<0$9uH%qPiA@yOO-p2mmzMQPt#B^TmsQ3*dSPbYAyM84P%ff&c%~$T+$g#^Vw@!
zMk=Od;}}4I1pHNncm?juL_u~tuKHTAylHE!XLNKep)R9fI?03mm%yg}f5yPge?MUm
z3{{%ntzY;?nOpL?-Xk)W>cVf?@8XIl4_jLW4yQIv9j<;pwl~Q4cQDKxbeQUj+OWvK
zmT?RK+(34bW`lYA1!Pew<b^mxAa41K1+D1Kl{-eyh@s&guV+*>x!nsH1E{lUCLO;P
zf!;t!XFN>aMp0-z4GgY?OK_*U?*XAtw^&xo5F=nX_g^|PBIjVnPl=|L5=ai^#zM1L
ztssW0`I4>1sfc3d(8?f1_LPa=t}2vsRM2RCceUcQkp(g|^O}z(R^`;(ZOH&otp*ED
zqV*@H6F%q72Cox(YT?w5b|+%qm>B!Hv}zxR`o&_q&EDEk7Glh%^E}DmHFU=~nBIqc
z@9aKt={zsIj;CW@_&T@U*wy_QsmRlT^uq*bLG=>-ZK&yjrVXavG6B#o*{>%^CdU)|
zidF+LvAjaP?nK;c^n*h-<IOeaN7mHh!1FG8srIkW+w1S=oq@%o7p9WYs{8Kpkrk<`
zLq;Q6PR)zrhzX^lmtg{OBMB-y0|>V~C4XR>BiT0S@GdidBc8T=J<o>z=zO8HQ+<M<
zBfKcUqr0Y)O~;6sh`T`Jdk3qL{&&o^5di}-*0!<+&<-n+N)KVlt1K9f|Gwt-tk@No
zoi)dP#s0q>-<J<sUj*cEbs0DL`Mp4)v(>QY2KGza3P_3@^C}KrYx89eKcY~GM8rU4
z36(ozv@9B5zOv4Z#$x#Uex6OzFLyp@B}RL~x%D2(Hnkg^gtxJYqU{~D-aoX-B+_^E
zZ$duWv*Ri^G<e%~B`Qx6R&yeF-R-*CCGO0w#=b67Ss6U&2psLE%xenT`%i0)olKdg
z#Ab~tr{1<T*~v$aj}OY6Z*?Tt*900$Y=`yf(!ubJ0=Rs%xC11bfRiHxj~loPW0HYm
z`!+U=5%49@QWmMldUQ9M%Hz+v7IX7Z#TJzqw><;owkor9rvwSpThI91&y4xpw?8DD
zHak>R4XUj3Zt3()R!CM8hL~itp@_^8ss`q<RVJN}E_k{C=@xbovr$dAoH?c;$x?oX
ztSn$gOH|RQEnuk3di@DUO`|L{GNtRgza-zy?$|0`<?es_w#m(K3gD;nxt+_5b)?+}
z2=Ar=!h1^UU0)RqxpBS4Hb<o4waPs!@?+(A{U!EegT0Qv;8@r7=7Q|X(Pg9X@C4Gn
z_MAEx58jK*MAj?8H98-=C5q)4kGIssj5(R}Z>n0*vY}#0#+1NkYeQ+up|!bl^p~4P
z&c%GIt1rW#he@}WcbJ1iy|J^6EEa66>Gu@U#YWqd$^F=gmOp4`{%YpR(A~681SqtD
z<4vuoj)H4KdiC>X4vNd)36lnJ@y4G3m;L`Kj<5c1JwS9!wT)1`Y~FP9*Dd6M+T!m{
z>fMr1oSW6cwHKAbdo6bMjm7G}*1~cR4`!x^Z6*lWrYmIAA9_cTboGqI-GTRv+l{%K
zhw0v~#69FxB%vHCSkZSq>t8C{<rg@`3yXX{gnydt1yC)G+%a2a4=9a^`oONw`mox0
z^TPxF<5bMRv&`&kWykWXxVT1^qGQ7LoL|3VjqJ2%$XaT8Wcllr1we=~Pji$(U&fWj
zT}bpu5-1*8iS1jYUaY1N(#7`0(}{8V9IMB0-r6pMqp5?Fmc8zzz~<7}$q}<|a0TFi
zn?y#3IgU>*jA?Cz!k&<sh%bg~;QDg9ksUkro-$0wPxoP*HG!9K!rH1}1FxS|nZ6uv
zYKgZGP6|6oTiFCm($~~H+{PNV&{;ukY7XGQ;C?ytuLs6$30yE1nn<dBTM_E6<ds7>
zPvO+5dRGO$uDa3pTO@;OD~oyfuU!9O_wjpIHkU~bGCThEhM5>(v~8L_K3kVrd_D70
zVpCj7%#>}glP3Ks@kYB+O=Aj`vZgEf6IZ5Ue#LiWbqqcM0p_n!TYYz0?7z1LE{J^e
z26hHw*33V|v;@T~v3&L)$2q=}DtXEQ&<t-QgZ^c0H>OVlDM}?*m^MLWP5YY!sm4nJ
z?M8<F?`EPWoe`CMa<<GNrFIWdzh@gvXBNC_W4%wp!v5>wy5Yu#dzd<&Ttp{`22>-R
zOb2Z1qcMWl)a$x&<#wBN)y^{=f-Q}0@c|eEu|x@`-#VM>n=#ou_<UEKII&Z0_|8RB
z*71y}sn^@&7j4wl7jB!&+cpW)95ISTGaX3xpdHLgv<;Knz@e+5L|#khok0qrpTS&_
zhjy(!U0a)E)y}7ei`O<zf-<!1y%g_Sza>qx$#ihaC@dE;y=%2&6z9fWZGrK0aBX$a
z_+8Bbkc^stN|%1e-M*N7QXp2&#qhM02Grg*`=W<LD&~~~zHC!>J^3a0p6O>7#b5<@
z&8RxtLIwA9KyNLk%p~q!6+3k!3$WG+&Q(Xt`P_$bwwPH|zfw3V6}@d@bc6Ult=+XG
z-P~Te*WCGGLw%#+VLH^5l-_S`RryC#7gVmf)EzKV=z0*W#Q?9Q-RnPHRDQz3W_vn6
z^}e`#VsfC>8_spC#T|#UYjzO7)RQ8yIlAb>J%xDk0sSu9!l3u{H_YR>*3H3^%8*@l
zruM(vZh9Wv*!D0}$CdF2)kLE++Z|M4y&b>}m1&(93Zk&1?oE1{*X)mRfs^c@hQJD~
zd)7AE+sYY6<zx9{JnB)8ZT>u5<yiETaPvkoZHu{fCq}k;R<v$j+fXMODJ@fvK0e-+
zId-j)+ooN)KfBY>*R^C2Iv={-vAZp$XAEW92oJz4EmM8jH@(!xX<?3hJnwvLZ?tdC
zPRBHb5wy`*u;#koI)1RY-Z!I5C*}3+z>$$?$CImHZuy}KwS-JX3jhN?+e`gNF~HHl
z=k-@HAbF1T1BH3NX0sKIYb1?yOo3`dgYL)|XLTb=ELYwGMpu<`xy$Ujk((Eg=*+d#
zsJ;O%McwQSl9E3W?nn)quZqLbl3FFjC8Bcsdv`h<too48E@LPm{nS_D{q~JFO#^0|
zi$Fldzu#t#_NMN6m_(@3o8EXMF1nMb^;?J6l}`V3&2SCFnZV-RHaO>km)#A^+lukg
z$dPmT$gR6zS8nwOb;&RhfPo<3Yij(}R<Ty;`I-+cxkRGDHt>Br*E5w1)}D?}hn~;|
z9S0tqShtK}ngc{^WSEG<0w_j<FVgfj$-8<6k68~=w0QSUta}s5H?KQ5w-fFC$1yGQ
zmx(1|CmjV`ZKxrSx~(PI^EPt|qZ9wxaT_Oh_T^TwGo;>gB1gr@j`6EK|H-k*=N+`T
z?qf+}x~@A0qVaZxhf`uytx00j(HKq?UE6%3^}Rk_V&<TowU)Jhl0O<JjO7)dZ60-4
z?*R;23Eh9H4nP3%*X9+t^Wf?oB=mBk_qtt9UO_`8biWmw;a%s1S9z|he@1>%N|2rK
zIt74AwnHYQkP0WI8R=GNON(9;@VTvmb<T#Kh;bT5W<Y1tXoTYKS}@6ERwU|<E*>P_
zb!Tk76=qigE5LtWxVec6yh)}~Q!G%g9dV&*pg;J<)u+l=Iy>4uNl74k7DHn9Tf!90
zYkFXr$7uDWxaxSaiF`B>99v`^`SO#(jn)?hQTk$d<qDO;i=VU-&Opg6UliM%U#PgI
zo^9(Zv7!z}jy1K+V{8ShlYoCH)GD&HsUSHx*teDps!>;J4>Q2*(q_Ub9S^KqE{_l6
z<tCfRWLZsuJ!@#<o6Vgrqr=Rb=WFR&+`5~dV_Mg2B){$%z$$i#IGtl>)Z?9xISWPl
zI)sJW%t>bl?dt_;6sx0=`$D}<Ld2v#{um^DYT>*>D^x78i<7pGi*$CQt#~ELxUcH~
zD9vsD|5%Jt>a+Ai6*4eCv1-4*W^GI%k3Wr7XBb_VE3i+6efl+(QO+;-KJHuj+Cm})
z5UI8-T)6U0I1j@he4bLjx=So}6p_ieQFs5K)I>I&8;83X{)RM6C-m0vWe6aI>cYAJ
z9|MrkQuGq&;c`U&3l{BpCzfA8O<nyCE5m__hkQzju<@qm>75VhXX<xAv3Q`t^}cvG
z!cR-}mp4c8{<yaJHsb7+DC{GYsXfW)?=Rnwk`CvMURtmAbS@<(>dwPv4~(^_h_&^z
zpiIR_CFwrFR}1!cAsE^V1`o)|o?UkZQxlU=3~DoSQIL=u^W<?9&F64|UjK`Vyz5PK
zp_Xl+1I6H0>s`hZwuv2iPji?wHlMYruT;31wN+?yVMuJ`i>gTl*hTtmTxwKS{k3fz
z8?6+N4o$7p0TV!Gn#ic2eEqA$Iy<OmG^QjZP@)>RUTya-t~i`<+6cOTu`M5yV3+tD
zPI|_rSQ+PY;KYTIrV<SlRM-a)sB|3v9M_@&H;`Zq%BYisy9kfgn^}($&@ynRGbM3h
zGwbCE#m)Zd+7ZuBBDnz~EA753tlI=BY&#VyxGqKsANPs?d1(q&c|e-%>G?;taqZM1
z%lRmzf)3#Y7k$u%#%$Gt#Bacq;6i~Nh-3ZtVoldabinz9Of+sKAlnl$NOrEoq@q^L
zpI1b^>ACRHFHOE=IMUQ1sCK<OTL`gL2W*nO5!iUL78V{Ep7KmlTAMg~azak^Eq*L=
z?p`(+znm)}=f2@V)>a!=CdJYGUK?s$;%zB7(DpAT%CR1=_~~)GUQ+v<n3AYK<1c<F
z(M#tI2?P<e?OoEIz!Mc{`<AqzpF{WTZcfEx+&4)u6rQ}n80WTa;IY?7@A-(<*=^E}
z4#${Ifw|$B)nMY01W~KtuHqF;pN;0!MKNI2c<iHi%tg8I&S|X4MRY3EMpO>?|1ghM
zHVi~{F7fCtoT-^eFxRva3UNp@?N$Ha0q3@2f{|k|&_OrPPjUQTFJ=6YlsJD5`8nP`
zx-VuREdRZChv{?>QEY*032odtI`8qNSi%rc${mCi*>q3y6kIO*gj24<gg5F?=Nc>J
zLq-6HcB0{z4}VO|{s(U?1il(qGL?$<4oSAiwOjyz7jd)&GB^KylymO%IT}`6P+t^#
zwFxk2ErouneYMDYugdKFMNh}5lbssIa$pFAg4h;NX%UR2HgHc`C*RYe1d@Qd@N6qv
zKHuwWhvLK@zNVDOt7zGtqIax6v53drY_83mr_S`R&m=;+UWx)##F3wXsUeor;4ig2
z?aE1spr(l3!lae?KCLoKyaTn-Je|@nTNBfMC4l>-MUWUnb`d6k?=X-`l^Ijx<4n?%
zD5h2Gax@SJSS!F1+ZdNqF|tf-rlbTGbuLHKV2W~<&r`@LK|SU#^p^3C`$7tBgMn8S
zd0OS{HDTsaw<2*Dz*tf0w71aDA(m@gbuBlc`z5jR+>IN$Ewf8DI*EBnE3V+5UES6P
zkUCI1<`=aR7qvbPLY4JCzmqHqU0hD1-B?_wNT!2V0FcL<dejx~=cx+Ldcfa9vGN_z
zvr+uhW7hx(F5*8nUZ>)#SX**qis^+QF%e6eWkwNS*&nCR5?@8IZ`&k-yCXji>}e+)
zmM5)+<8G@QOd{9PkYCrM#7xY_8gnP!w`NMcAo{8Hi(8A9>{snDQ=U~fk{+n|K+?p{
ze=PJ*JTlVbS~T%s-ILU@<<9xS`S^AmEmh{*!$NTs#gn(`h{#qwe#J&;B~e3o0rhp5
z0HSwsKIx5(&`b96F`|W-f;#(1({&uXop%{+OSA~mC9*Hpn!wV#yCTdaXS1n<p4!LG
zJ<qCU?`yBn3iY(hM(!h#FP7HyqKK4F3#N(?Mvi>Q^}%wF3P2ckSpSu|4v>2Bj}%7!
zBpH3FU$om)5GDI~o-&#bcWP{Lz*Z8w*en)a3Rf8dV7_zZnWCY#2PF%KyKU1{F>~bA
zZRX7<oDRwPR+D9HAo8hpfnNF!<h#i$W(Z1QTOoE);NuYU)f}$GU2eh4-6jXX%40EB
zU;sd5Rqk%AwVH=9UIiEB>DFrA!s8@pEoW+kRxLVgp9?7jvpPRtdTdu%^?Q5yC{?2M
z>Ylslz^;AGv?z*tr0YR-+nHSpU8@*X?}^#7$%^R(-uRW?glxM$G4)$_@Nl`+Ea0$r
zc0noD-d^+mJ1QIwT=PiJyh$JALgDmXginI>c%)7gYgI;%>or$^HAm)=Adu8tE0uX&
z>lr(JRit1xsot~mDEUj1so7=d(S&BCs&n1|8LJAV;O;XEj<%XT-F4DYf#ur!|F55J
z0eKI#n!Ji|?o@Zm@OVv@Z7Y>ZxImw3&wgQ+hUoj@jTvQ!Gje~UGqvBB%<2Z9lEwv1
z(Pe!|3g0lAT}7AvAFqCV)@rCL$zWTjW0962%;kW%YAI3+OMe4EtXcbGoF0_~O0;dG
z0-Km(T*buaL1O5hvx61fU5})wo_ti2To;x5GO(ymlL>p}{%w$joT(#`WY3Uc_TWph
zjQgIwn6~TAnoYggpHvc%d|%U`{4(i9+h^2{vyH=XK?~&#d<3qVU$y>1>~zj^er6iO
zdllovd6Jk|jiemwnj_1^>tE|n@eC4LrYyCZ)t#<(9QD0V-{@o~#;VYDU;zCB_~}RW
z+-vh{_TTAP|HiJ`ECNWb04`yp9Fk&qG#iGD$5qd_0vd>KLK^v}w8tfe0;Y;hNnvL$
zHNYf}noeS!xPNhGvi(Xdr}}_h0u#V<2zUsfaQ^Sk@n5<Dzmt<P5Kh3!fQtoz0`LJm
z?9432d>tIJ*~zjuwG_1(^$L}edOJ4`2?I+X5@t8vCrs}NCmcFGrg68D=Bu{*ndh`l
zG%aLX#>g?5%*Hsn+f~D&XuuL}<1oX>t5SEGF3zl^Jnd(r^Q|B)=~m1b1*BX4n^G`4
zc;zcT@4+UbyfPo(-#DUHr3@OiOrp}$?DK(Vw>)e6Kt=WFx-#qhN|z6_t#Y(E8a`W~
zROy${`2nFC)Aruj#^T}c#`%}<SNI1R|7d?{{hl7BlDG&?B{)lsa5k(?SmX~ojOt{r
zBqG_Fzc(LPYdZ%G+5$td6~R2xvnNI*Z9ItQ`IHqeZG>$S6;7Scn;B@-8^sbYkeakf
z*`p%IOvImQBU$CnKrMxG6~d%G7V*UU#+9o8yqUTI8Bel#h7?uoYI26xGY$slE3_<F
zdrJpV|7s!@fr9wA#o~PeshnGFnIjV~T28ov=_GXPFQ#|whm@tw8_?jFZ3Zm45L%Lz
zJ%i;Qz&0l{s%<9^dK-QO=vj+VHy5)Tfcn1obH7)5*seliWK>^L0!H*Mh8h@>2v+OK
z*}%6?!V`HuZ&S4S?_jK325cO0V{K+-GB5@jWp2wcrt3@+XWv{-wri17E%TKN3UNUC
zGzNf}S~i4J-z6-%#f(@wNNE*o@!EA^)pBO~13CsR;U6pGA7_rBfbadC&Yumbgp0+d
zj}v8aL7xW|qa?&+PJ(pP(nY#dHO)D!j+s;=vO5`bw0l7rTo?YQ5s3)IUNbx(<^SI~
zn6qe2;NWLj5+WS4OjbS)*G$T;ACV4EoiM1*UUPIhoaPicy6KLW-DIsJ>(gA@SXH@}
zgj92qGCM6b0H|nf+%O|9F;Qocdzk0=+wy<TmZVJThJsxFkR*uJv`My-GAR&}r{x2P
z^{T7mNj``54?XchDgZTz0H0wkdCusUp)*^u2mh-Sgfsm<J@vkJg>{NaPwd0!UFL#1
z<H3iz(Y%FkTwEPZV;19u>|W6>;+bJ$3c`K%xuyRo)+DCA)ObrOus^Q@n}HfFC7wWc
zoEyKBuOS~L3Y>T$2vxr9FTHa*$2(JD%z=4HCTb^KjYsuj@!p(|)_>d>$$cum_Fdu{
zh$e|?vy<X5sAP`RZ48y*POcdmZc>q6_ZPYLa$8Ao{L&SFh*xZbY1L=AY^12W7g=<D
zi*WoTdvR}cV|&xV+PNX%;qfm<#M^|obV3gj_X}qu=~(M-4P>X7;86fMo@zdr;Vrrs
zd|=C{7&-j6Pan2?OCJZ!KQxPCZs+!Y6X^kT<ip6n46!*PzZRE^*?cD`$`t!9wx4n{
zfa2tSrPkiP@djS5A4YqWIa5Io*=Z_CXj@u$!*ibq;g=#`BrRt}JfqtgY)<bZrDKH3
z=`Mb&Vd2>eq3;&_;hH8ND^89ld(-?eum*LBq1zZ%*CRsswTa5nHZvsvGa%Q0&VcZq
z0!NX#-4c1AhKO!?V5NrMhI35ss%@5ndu2NN+Sw<a;S=42Q)?!g!Mr5RFO=!oNzpm9
zD&!VDBQ!5e7q}~nrpmr5{uok-t&IJPAuP;7=s9G(%|VvCdAMmSaQFY}tC(^BCx+)W
zPAs}{osIPaPj2Q%VI2lB>z^ODPt}*U&r@w3Q7_MqlT;lhZ+^snxluo>zEMdJF8lDg
z5t5}Vpb{^orHfc9J_|(@Q<DMt+Qrb&e=K3IuB2)!GU*!k;yWK|6sPTFN?iRC2MX@S
ztr6=Q0KBS5Q`@{5Q}<aQ@r8s>0Y{baT3Xmpx-7-C`_0vPr+VE3^c~sfU<mVYu-I{Q
z-BGR;HhMTmqd0l<dR1BizsttC&)Jy;_kx83aPI69n*?WEIBbq7vuTwL(FcpBUTjx$
zG+WKoIFDhZok7p9^U{cv4*~c!#W4~~A-Wz!;njryl`r71Fhn_#R8drU_HtRk^C-ZY
zc`{L8e|5IL9(9X)tOc2Lu&foAP_j;cDxW4N=H7Q{w)J+#F|OHdBzQ@Tu1*v~ezh$K
zkE}8fGufdA{Z-M+yaU3E;y^GE<+%QBw<NVe4beDu`Ub$W-qFG{^Z*nT=ndFyCCR<+
zy6@&p;v5J!$o*SE{Ye&Jiw($*JY^^C7BuKvy2XklxE+rRACDO|CBi2>n@}LhG_ce8
zVA3^N=f2eqAV#adIk{Q)*Sk!O<uy;;L7h}rleqcC9GXKqWh?8COY^a*`hzMU)}ZXe
zKe8OPyl`V+IM5x>Wxv{Aa2Z+Gt}Pd1hU^UVCeioFYXD0g-e0%9wbcn-0T3fX+z)5E
z?>k}Dx4&CrpN$alZIA|<8u`>60LgpVzNeT7uOsR=8RtEM@$;B;--QL1J(rj`E*VrH
za0|SCsrBdPZCnXul>2ED-QjRPijVq53kf-#1<*mY_dBk64tXBeFZi)G!kd%C`V=Lu
z>{qMdylyO_Z_b$gyZnH-oBx@Qf=5C;GHM@$^OTu)gg;t0nh%th#tWFl4QiI&`^7b9
z%C*@$)<m`tYUM2NZSpxjgJRZAxt`kX2m98MYQp%DWy0W+YpLrY)5o*+>5pe`6lq3K
z$qKbr0jL@0Di?XZAL);k2Z|yi<+aslj~_h5e>HQQn#IP%O_JHWoua_y99`vQ*8=$Y
zE~Nfb6O2n>7Wz=)tEzElsri*=F1+RqLpq{!QGUM6*<Z-7&u=l?4nOHphq`9dNQBUv
zT<vjfZ(9PXchhxDV)o%#JC`J?r8fb4?pax$h2Wkn5%F}(3wM=_rp$=iL9}UR*9V>}
zXVsN=Y^a$1FS)%@FQRe_im238ne|l75k-439K}M8QQihicV&f(=gG9{Y^aaDW4o0I
zm37xVMSad3Ns4d+KxUwq%#G+yfW6EFe_bGj_1vmzEdxXgl4Gvjtc4iwZ#SN|ijY}Q
zbspWaX$Be|cyklCgOUdlb!l^cGG8;VfgqipA|pBOSEKGkFn>hYBDnto$nu~9J;N8p
zuZP@+BG7LeNDnB~c*u4~b&+hFJ0jZ^m{-m+KDH^odVsMPtY6@GPahWJ<XEs6R^lsN
z*-ru(>>t9qZVjfVx6Wgusin1|z@)Rig|n3;5;Vb_wAfF`h+)HoH-#=GnqZ&7YT>(t
zZH6nFq_ew7i*ncQXjpw?-iMC(Nx&`$vgr0llERz&siU!ixZ|vwBM7}Jh{l5LyI;1s
z-zku+-OF^NGP1eZ6&5^@dsM<TM=8ZV<ayn|BXI}hb@~#E#$>L+BP9s^i^FYAcK#Z0
zvk<`^XtbFlQHQV50-w4`@DFiOcR){U+%EUpUe~83*0u2hSm~>W*4_?@>DA#BR0GF{
z8s+%UME2?BPcrKlj)6)g5e^2)(?NbN9C!2d-642Z_3Tp%3CAT2BAC&$hCCz!LEU`q
zzXb=d2yCrjzOuUc7rwl~vaIr^iYZ>Cl!z2H>PVO9Z{@uBex|$x_`Rpa7*D}oym-<z
z9PVdbDWBk<<*2m9*?FQKqkt>z=|8P9o;gn;@<=}KLA*1mxHUQQzDBBI>ATi30==<C
z&ELMx8_iKODhG0h@?du^XjEudsQv0Di*aDsI64Q`^K+9YYg$+Jzet4>a*B0pKTsT=
z2KA!0aroADD}pL=qlhg;-uG<NWG30WXwhUsN6z1qB}20Bj`)NGvfkoO1^&}KLfi>P
znysRXhyc{&xx-m4{-xQ_ADynHJ?Y+++@=le)pqGw6vP6w8L^U08|~y@_Bas~gzBH2
zEQUp`eX}NSeD=4*e^9G+PfE~$O!Hp<&Mj&(4X->H%&Oc_$#R*K?L0aEo<&YrAnNne
zWWGPdy%l89o98SmV(Pp|5nOQbG)udGb_zEezZ^jtkvsMg#`{<;kh3i+W?%!x6UlQ~
zM~$!cvh`Y^ZZ8n@tnm4Oqw?<bKv(E%G&X4W))yTqn-dRLy^D)(XoM&=J!%$ffOv-&
z-1?jMG+y}r^)BuKF#$Eq_|!VAxbios2apuH<3AyiOc45Z=p>DWRoDp*0Eg-F<Ho9F
z$l-O0k1#iokgH75be}&^fqnpiVHB8$Ldj;3efX;YL||1ig|(zdrYPK8rVHZ~V<eS_
zCXVbi>}ax;JOhn@Y+>9DgWzWkzApVM;1BT3kcaby0q=1duwWa=*^9(mqjE@a=Ig5h
z>-)V?e~N0zq((q@U<C6ZUNon%bls_QOeNz5{%5E+4rSpX`8`j`@8@CSAoBiz8Z)MK
z4_$BUP&&%~TxO=F)O@E&q^`|O2|b#ui&vj&&RaE@l^a)X&j{P?X^^zAjh0}48V$G#
z&6X2A<S8yQ3z{3ehRYO+4sx*2Ie@(AS<!;qL*Krn<o5@ausJ`$orn|##jesp=}n;1
zt_M8^P_SHWKpLM}e{umd;fv&s5h0i#+yc@-@jeyhehzhL#Uir4^V8^I`gMTf`5oMA
z{T`F<f%4VYId?&>sN>=!qCLP#j+sGV1LQ@F<Uelq;?a*^^z*9QambbNk*70SsbhAZ
z?(PPb#TTi@G%`tc=T0=}^TCD$t>y*6{6^QR8A`b`rSILM2!~1`b(y!In+%=U_3*qN
z`=;H;_Bwr;_8;QF!db&lXpoai2?L)?P!pdrbYrUCbdBT<evGr0mH&LhoO%s>_!9s9
z;oq_=fO+-N97LXMFyM>YV#$|j95%h0w_379X3Li1EzjD6?ZzJf*ZwBhf+-NC?e0?>
z?ttOQV%#<Oa)a%EVZQYFhC{XFS|FnQ{00yZ{kgV#!I!Kty`A*I?&3y~kLFr~@5KFO
z#Flub-{j5fapK6?E1&4el&re8)4HaJ$_pjcDF@%IO(Dmd`<*k$pWYW7$ZN%}wmike
zoT+sxMv=$Ws!LhCF%d&xv&Qzg6S__o*jI$3IBo;;1oY3MK6&JDs^z|%X!IdoTZAL?
z=4a+Y=GL^p^t;&+T6{ESSm&TvTHNHbm_AoVXqRzi#>z__5aG?v%~d_fmD%orS!(kS
z(2re_eq*8M<j@ANIrt6u8I;wSiWy(n7S@6LHgy*g{?=!AQI=9~|L&>*9y8*Z`xpW{
zj9fq1r6B%$dhw#7uvg1FoYjU$zTRXFIB?tw5b7tvIjHEM>c?a2^%u!l)^Eh2d)R_S
zmk%tcmX5fYy1c2s2X{S^7&sCKwJojqNthcP0a%HTA^B}B!3-VHojaDw;#21C6I86!
z<hD@v_NiP?uf0PtMGl@%-w@yF7_YBpZ^YpJ8ofl(OSl9xJ6=b&jcKo|63Hc8`V|W_
zjpUbouo@WV^&1%V_d{M2<AJ8wCYmKt?GNVKf<b8Q{U74+WQ{7>SzJpXH;1K!LV^;M
zMI}%oG9m?IpZ~V9lortbdlIKL&=JVn?P9Q_Jrn9Tp3*;BNj&lEx;J+X@Ge&Mu9x(%
z?&7?mcU;(?`5k+6OARqt<8(_941Y>{!p?E|@6##Z%B}{c^5gCGX6;Hp?ROk|Oh(3{
ztCVq<(x{ZM4TDKCGAxr3*ZZaOaD(s5u-UW(K>BvvJ#o+sJ{T&(2z@-?9JN1eavu;1
zmdGF6rc36S=9lv+rOl$BARJY83>c9i3Ft3Rq|KeJ<yEd<6;&lA|D44(B(2!~^?zRn
z;F6f7>c?=aYot4D2z=}qz@DgJ{jr}@s_HOOq0LumtJ=4nnI$5TzfO$9FXCGDdFmQC
z!y(rK!zDc1yDX}cZgmOZ-s79Rb&(tS2=a`a>H4_Y`gdD5lB$lE#oV8+HH09sXA_80
zzAb2COW*9ChW|-YXRKCYNuXV%+Q@>s5AWN9*$DTNZ``aI;~NcV#1V1CWPE*b&aQ{!
zxju9jK67th@@4C}#}Tfq_9gpaC7AVbCo8CTR<)UdP`xq5*r??<w{QA`g1gqQze%07
zx-?VEJ3+sIxEc*UJs01GZ-PqTJZ>?}0fxPGvM^@9N>ADxQ4sS!3~X|Z1qYFXV?o5_
zs8C`*J`mm$FG)W)Z|0tGT!u%iPv`oUzS7Nk*<wMDjsMsSS+^K^Q`e$EGV3cx$vPlz
z_znmQ+IEK;j8hnIIQ-@d`W@69tqScx(?k2h@pikeMI&+OjT!{VCDwm?a3<0ru$n{p
zlkfGfEqQ_2;KS{~4AqN1i0{@z_Zu#o=&m^ri4LC!OFsfpUrL-*_F*S4?dV1Jg_3s_
zn!jErK^vLBusx|+W!*G=PCQ4GO(UCeb|1~(n&s;dA`Iy^S84oS&SPW15eAVz=HGWF
zJ7nGI^MZ62k&%admHW=-kfF^f!CbHz&adyxW^;P~FL8O~eo)$t{!#{U06}DKQB6!0
z2w$T}_9mT9nQ=cuaX<p>>y_E}fKF+n;gMKK5!U)tYRA6zxgfXwlhun*4Fi~t8wdZ{
zy(HHl0Ft@ebYDM8oUW^5=Npjk_AgWTN5cJ!1qqLk<~G7lK|BO;e}U(6@HWf|NV$?8
zzAVh@q(Hr7?b;eB+S)N6TbTLCx1^@?=LElW)g_dD+b$w++rg9Ft*(~X{3JBUQtYkt
zjIgF&okIn&x!4<@NByEDN6c=OM=%-(3-?k>o;P}KAB&`VCzZ5*1}BCPkpg1ElLT!5
z$@TA8BJo5L@_u6u=icS!PPqsmkR_R+i`?uW@0}`$H=CQkJW{h75w6FlQz<^0y<VF1
zvuZSWy?yl8@-SU?#^C6OgEu^ouSxGCJX(Pt#gD9N2~+(;JtAJzJ1R(k>edY(dhi7#
zJVx!txM7cHz=$a;qf0ew`<hL)cHd1^LsJNIX%TsBA)zFnPse<-eiG>ZGj<g9s55c{
zVSsb3J09pS|HW1FBU13KnK~A5&<_YjiZt9!{<h?Y_*FyjO<HimlL5s%#5@PO2O#M<
zcd#y%P+&365lTwQN9a4NO)tC=-v!a{VWALtMS|qBTdt9T#AE!@C_0IK6UBHAHtZ5G
zmMh+`K0UxO6N4t3LF(!)ga{;K!5V6O9>}{NETtrJIHxK>T0xJ{A(Un=v0RuSQ7{XL
zv_VxHw?2~?Bp)f93bEf9VT132Xxi~ZUXB|?F_6oHh#i`%T0}@1@b5wh4T;j@LHsb1
zt+$CX$!iD?a5|cHfJre&e-mO%{3C~2?hP`OU*c`^(VZuDscdse$A-O4Bq4Wvv7Q%3
z-P^?#{EfBYkW#;=qc{;z;`{pSrwBJ#8;mOOHC#%g1LK>E^wAx=GyrOJkb1ix9nNZ;
zuE{y~Qo}_1zGcodyI$T<0&+i<PJiz?F}ZCgemmA!;SD)~SfsMCe0@g#ca94lb7^s_
z*VPZ`bi&GB<qdfp8=AAJH?XV80uT@FHGeN@+IC+Esvvm*XVCpS3yj+O-DIn6a^j(R
z<fkwSL@J32adZKS!S&x{4Ms|Qi|NJuxnoK4kHGrN5nLe}9iece<BtylYvY{7ISfa&
z*?)urxkQ@WdVM|DaqK$|=>Q!jRP)#@gWy6pRf}o*7cL`VKOItj$iT721Xf%hRO#Rw
zRELPJbDh+uvE5KY(x#J61Ps{Zwyh5M=#8=tMp_J)4NSauJM^-Q{$Dpzt_5h5nPsVC
zBNNu*i6VmpHK<NnXsGQ{k+ZiVk80d|wJ>%5B;_*K?EQ%94+$*Ur0NEErKiK$nZHa{
z_5Q-<ncS9$*xJ>A#NA|N@!uHE?W<(fO-@Z-hwVdqY7Y(ubXwssB7HRuy8Nl$2!5-b
zlA`cmjMb{D<Da}u1+BO{v|?<5%+1KGBE#zar8qKT_ACRuln8IJpE$f4dEZ2Jz)O#c
zyJ$fkp@GBr=;rDWC+1Rd0e$RO9mXB3vY>A}=EY=W4R@_y3ALtrbNh+BD`eeA-$#RN
z-ujU8ervN;;2<EsjhzZZ-HS>R2lpR_y;Keb{}#3Ld*RIOem0;%+z)<!KZsX<{m$DK
zKgsLO@#n-yJa8<$|I%K4zm){$ZK&vi4o_1JEE2l!V+-L`ItD{&zw&k@u_8>?*c<N9
zfGnPau={c6K-vYHlpxTB7$QYH?X?J4a6GP)<W4wIGCpJ-<g`ds(Yzo#v57n^nv>QX
z!rgc-r!khlDjQh%0?W^g>9V$$Gg3Ea2t?Zuxq=IWv>PQX;odzb*itwvY;GLF1u33&
z;0paw6zuoihlP{kSk3@RUGp${kdU(y>cp`!lz7QRgUj4t*qHb!;TutC_K-K~jbIe3
zgAw%;M`PId+pY^@^dne1>TN651-FKpToPuE;rv17=`#*!VBKX5|3U)n2D~{a<z=e<
z1h*A`tMP$Bxk75(K5@zpcSZ;xKokvQv^jeTM(+U2$L&vk!{ioXdxg=1K6?k&UN0H7
zWp4N6bpI`0;4@A(SIObV;bGm%dyxKff>8s6AeQX8l#A)qETO*OE(Qz|8~dTEkBUly
zuZ7ph-5lnIg_k03W6*hye3a`4yNauRO-`Ag8zspIp8yfTX{vN?zXEl%NCpkciT)V4
zZ<)gE3MvB!?&socsMBO-&mIzK*5=AR%t*faad(`P>y{<ffdD8q|3)=j$DHYdd#v?i
zOJxy~&VqzyOQDik&~aJ?#~!Nw#l?{q5KGlzG@A-HQOWSocMbXX2Jsw|KVyF>W-re-
z5yDZr!&~C%vBx*+eC3m+2_Ev0gfo%x;EvF-KFa<}$F*yYER%LR)lUb!Aa1P|S`pS@
zaA%h2T;UYV<kylBYz&6bV7Jg0G_`XzkU%*E))M7RQG=rL7ly{GAo5UacyoanR)`aY
z3uV&dsnj*QJKTqy(BNhlFKq9n5RlliVRZVh&BCVl*!}kYj4-Y;GZ#%_UJb*y621p1
ziq?Yoe&UgA4c?F*r#0x=kCI|-3iO6Pn2~tC;4PTgH~4X<$kJ~bPN_Q1HLxovWAoRY
zhP}=G=bsvd=5gTQQ@|{z-A<7V6C|^5c-Hk=CisM$91g-o;*am(zLhFuA|8$Yw&SpO
z{>1d~E{LiDOW(JYIK+oe-zz;F&QOSUu5Jqd=GyNgs<{6|XQ0hRy1FQw7iaGDEb(2W
zFeLg47eB|Q{abo*oiu4Oo^PsRsCqn!%@~9*#1yXp0u(zn3lMVye#o-JiLkX|WyL^!
z2pE_zFv21^6KUQ2(i@>C>D{0nv5s*B%oWBmB8`5A<SXPI2sRTrCs0E|!%OI3Npw8E
zvNtI-aCX8|qj9#azNxR&%&hy^@u5ED*gN+k>ule1?vPD|&>Y<%Gc<3-?<64$|J^0y
zb~cX#dn+*~O)%l@#Qp>R`QKKF0WXJ@JTl}T&JT^hQ+dP9OupH~i)JaKM0_l_iq(=-
z^)53lTRRHmIc%I#IBy!8UB6WC9gGp{`@`p1_d|rSTqP>T?QkMMqsn6SmCp4LKmq&w
zQ(=QsTbk<U<$H@bm=)bmqtDy<0KG^_PahA6gc3RL8%{k+@6-nu*k1nsK4tl?=bP#r
z`OJgo4jUz|e;!0cZ3A#bt_G9NXz-u)M@WhfkE6NE8<}EWm5?A7)IxC@tkzbo{Vb2K
ze>BMy^U;izYk*=^X~fgJ6Osgy&&(mQN$-zzTxS+-Rps3~mbYyYgrc^#P$pmX3Adm@
z<4=Zt-}54T&g!s~Eayg&4sf`ehO4w^-YjSKKI?;5dN~J8E0t+H@JsC;kUQp0h3_l8
z8J{8TKRiYA9*-D2T(W+!(1aF5<Ky!;FyX{9FSu4)E5gF;S8=zA2bp&5l4YjbJsF}p
z#A2y^u>`TqS{q`{^}%oU9Ft{q+6f5mbxDI2;pLP~6+l-Rlb#G!4Foht_nEfAk)SW@
z@}fD-X%JAWA9lm<2`OFldt)L>@ZF+2VOMVL)CnN8l-u1l+xH7$0&3ES<T<Tguk+yZ
zA@{f1e@jA>mJ-@M1VtgA3-1nafOx~FH|9T4m9w6-$1K&OO>&~3tQn%8d+reaDVzCP
zf(H;MdPA8Ij2xZ!Ap!%($OV18J7$LWaVa6KKS|H;BX9~Jb07s9=L8$iZ%x63Y2E=e
zW-in@d6#z^UP`vd6?Bg`zaAO#=N`O1G>F2r1&P|7l3Yd7DcpUXU?xH=yWoRm3~qqv
zlTyH1Nw~u4{FyHTIv`YrlGreaCP1LP5dz7d=R@Sn&EAWj`v#kKuB&}G7jH<c!#;oz
zU*Ya5O2Bl(zZW^_Xw#;(jJ*ZaeLnwG_ZbsWnLX;|#cpRbtJ9-u(ET))W8-a;F+rC3
zAU7ZWtGui}z09g?@S$5>USxv8NQ{oc9}*mt9!dRCfXOHAobNliBQ#gbYXQfk_V#gw
z!bgdC=UaCCHf3>XO<KW2-yKQ%*Bk?eWIlae^&Ppg|1fD#W!I0?Ec-PS{sbj+)5bD5
z6!YzSloN364*XxP-OouT(xEE%6%urctYf}}t39&LRvuaYa9lq6!E{SAZ{05e=iMh%
zE{4v|z(M|HOs|z*uAk#Y)@Ef_c;(4k)#(FC)$$c|vb*XG%v1yZ{(?ny`buMoPFz=&
zl<xiKD@#?)GdPYvLqwl?#G~SsP%A_vebNhn6C#Jy8^6@@93D1~<dYEBYUx-2%@pnb
z)#xXB!uWa-Y-UhT)hX0v3ig11U+qHQX`q}VPX0z`=T-ig@Dzr(vip#@0L<KgmW-q{
z+yYwzw~EE!g>4w=(&QkPnPI7YGS$lKBw+rI9cHT}AmKosjE7FM1l>6{3l(_he)+lW
zFHi0u?@+N=_*w`FCL-(&I0K#oQbsdP-R;7wgkmFN=yx{Z88%eE9xt^%>tZQe1Sc<k
zP=o5F5iqw2w<fA}z7|D41QAg!^m&NFg)|0aMUUBvcn8^B(9$qbTP!n9M<X`Ka~0Gv
z)c|TJfes5~-43y!ZSdvCjt&)AI~)F;Jj;V~=88|Qoa5zM{rwB94*nYqN544-+5D8a
zzyj}sfX~gaM=nDi9_)!Ac@m&Xm3#-5ssMVK(+y(JgSt6rhydY}9`^lUUr75Kpb6+}
zI>2H6v(qbfGNq)D^VcDsaUfsJx%C3ntt)pP<htJb;M4h|JxYN^k}*dDA$`hSUj-R$
z%6n}C`o`@8md4U?SWe$rFrH#RhNv{a>-~6z_J&8xK15SgeQDA<zaX&irP0gN-y%LX
zUXsy*aCdV%3BvXbVil9cm$SE@{C^8tEs1E&Qmr{XI1NS@Be>XI1qOAshtoAn+#;)b
zcYe|lUwgB!T|G<~$Jhg2ALuk!daV@nary~=&il4einXYOnGtixpdq~S!{LZvIpt~;
z!O?PK>G_$zN{`!L>@LPxnnck*9wU4D(}ROYg&au+Rp`mhG5jbo4mr&m=4V#_yV#fx
znyevD%Ws9?hEh!?d;3Rzl$Buw$>(_TY2-M#+EL1;c<{!lvq>dXSTo15*rY|UsSo!)
z>{Q%IlMCMYbvg4|b$Z?(hxdjaE|C5CrljbEE?q7&J3Q&~XBMIHauEVM*?dhuw|g01
zp9jvKD>qNeYLPd!>{R`JCRudE%W2-rCURiH&Rh)582O+9w~OLT8%LJ5uotwt+IPA|
zWX8RNwXZazvEC+qqPzem-pqx3FxYyd?AS&At~ROD?=LwjDh^`aae_9aP_K`;hh#C}
zm%NG7;w{`HgTyX+{M9Xr%ZW22*#Dfc0gP@Bm4)h}nFkss2=1<MKFLAW6k`f|YP+bu
z_%qr*7snR{fOlwBn7|Qa^L}EHGW#-NPG$+@Nk^`6JK3CX-9qH+2`|XUf(b1Cw~xCh
z$pt3x`(ItqTM%D@oA>Flwq)Ts#pn-F{(2Y}Ym&DqG$2V3Mdx3=Z>B@SwA?%V1>5hV
z16(^iZ5cfnoqw|nMAPB+dlq=TXWCEM9wHitgn(l%mFD*nWClH0ul%b3i~=su)V79x
z20h*SYpR{V+343D7)1OR#e9cn8Mq(v9NgcW-hIj?1@C;9lv3lG8CJJ=_J_%t33Qi=
zViBA2&Rbh7wzEPYz9+W*p7RB*{NjegQ^oz|wnvMSi6yKC-$l#^%BkEz@)ONn3ws8O
z-g-czDE?3VyFSn;en|etSG*hZ0M%05QO#)jPD)`(6``qKCWTQ~=>MTYH|+4*Jxc|c
zYai&<!X+Lb^N8_b@Z`0Rsu($Re&(r=uU7>PI!x38$j9lsIv3r7Vtg)jGn!SBTAiJI
zP9vJNf(9{Fb%mNLk)H;lu^$g@1wM4KSwOnjb>KvN81rxM+`I2<9{m688F$ecr7A#6
zgu^Rf3D1|5C69TaZTN(w>&ZpS#}l6Wy`3sOBOtX;Lmqf>#ZJ6P!FVRqgFlgyopl1R
zn7xzikNvq>YX_`rrF61xR!XRDvCiK0^zN^fSqn-@yPi)w_0uELxi_hZ<{rSz*SrU<
zb6q!S*jzcrXTafqjLD8o9;uHbpI9`M43WK7rw-$m>!e{wSaO0lpUPps;{{=56XUb_
zvFqPUZolYcdv4>pteX=Jf<j+0r=!Kf-{&~PV@{_a2Ly;$g-Z=|;U}jdRFTLBQBEMK
zK3qoPIlm&#-nTv{jB^D2UQ0ej%2wz*BNFSdGhao`DbgU<jwG9PXH1*uTk0GiPNae}
z4jBl@GzULhrPt-p+vCx{%ld4Nkbvq0jt!<y>ln}gaf2ijJTGP9zUWu7C2{;7=zh4t
zpvDP<8sPan3=$2Y;AV0Y&wk~4`JH6x6)jpGO=$Z}e*GoVc?O7`=s>@jxO-2|^&Ho^
zFfF1AEG!Cj(7(hA{|Px0?hm*iqHfh_4v!H_i(h<~z|4e>|FSrJr{8aq+e8rpdvkqQ
zyFFd)J}3QMP!y4|2`p?vC_VvoDDc=uN!Doq@l9wyooy|tV!lsQvOYbqK$M=<c>Am|
z;Hf>?Icd>>i$Tn?+so_1hvC9cto98pzXM8{pl~0~#V(XwD>b>b@p7boXUBW(fj1_)
zR>3<<&;BWVL8che<WVv;wUSlu^Z;I;aC`rx1UEF-80Ci+bOuSgAg#9EGn{T#@d7a~
z<F~+tOC>9Zx=Vdb)@oNNSF8T1C@CYS!7D)yS<#o1Z;T88qaDgQKd(0nS}p7E2pvd9
zygqE>`sAP-oIO-@Kvk7|gI$Fh+8J{GqL|Az{S$km{7%8?QKr<o7qQXM!ZzOdh@I7%
zcZY0=6GlxVi(4jo!4RL+PnL-Eoq}rRizV*ite<%W_aeAU883rwUOrrI2pgsP;x`>l
zxq;7%aSHr)8)#x+3|CLq=+5O*0!(A3cZC=|ObV@ArXE2UHgq<}+w2H={k{E@HeNS3
zWX14Enq$r#=0##7M9P0=6v+)!7Z{-ol1#Q9d`QaPXaZkiCs$^JKrugx89J@Blh>ER
zREzE7KaTV&_2iXyu5?;NH9u-7b&jt+8KyoD;{c*h%$kGk*=VLtU)~p3q29s1gh~*T
z;9u|(`9?-PBqn*8*wSiO=ooJ|5WF%Lh#UaKbb~8lXhW-bIi&k#=sR9~e#EyXGNT2j
zO7NQ+>1f<P_bJ=pO9jkDIpi&LyIVjG+Ct4$qQ!f)4(r~Q#%9#wigWAs-c{ll0@Aj`
z-rB(_Dgyu3<VJ}EHY1g(Mh{sfHkVkJ)dTS8+B_;9HAWc%qff}lPmIMcnmSz09TfLl
z0tL|ytvvHi?BtAJSX*g(59S|f@rt>T?AYPVQxaT%0~~_1qCm=yu5Zw40>!d8?XLLU
zCyocEOe8Cir;LrQ+a0FNF`q9tyLtAQAJ_Nq>il@#e`t>pw08UU9_0MOLdb|XgVg$+
zq97i>sd}AFrRA_N5v8b<1^*|adfmzTd+a!|^M@~0Ucuk1hDlRFb{{IlE>%9$`?aRE
zvX){VycUa6Ukgvn%$AG_YpUzMFcD2N-!L_s&P#}`ul(@oLSc|0IB2AhhS02xDr?pu
zCNE_tZ#0AEVdtV83*KB|8Fh*%9@W0~UF8c$>yAluwu(0G1UGgpl4<>M7iT0S0Pq%7
zIs&msXM|7NszCBdDu(5)Fc}Rp<ni960c;Qf*L#MIX7n3yK`)24OqAM57nr%wsz!dE
z7t`;%0P)I5gvX`lg!6RIe@W`{VkqSs=Z2&_#35E@2NGBxixWRyu+@vBl{KDK%;yho
zsGy*%xBi$HQ9Ctbg-dF-%~E%<)gjU0Ad{Vkk^#Kbeq8&8Ge+Hp292-$j_|w4B86@g
zm@X9mJXEUkUkbDLznT&Q(n?Oq7>~2NfOp<B#ALaD*0jJdjuGsa`|H=@|Do%xqoRzW
zwr>>#QBqVShYsnGW)ML-6qJx|kQnJ07(ik`>28pejzKyU6d1Z|2&IQ+=z)Rn_KA0W
zZ#@6af;Eeq`<%1SKG#0`ceTm(oIx$ffF`1x1Q@6rdVd=|!BNrXlBw$#wP4xJ8@cfw
zDA`I-7wONlLdD7YpA)1HG*l9Hfo;k807(ciwE@hw2#<yy2n+aWyVwLiGFjYj`5DqP
zO!#_ic>D}#kGcd}GdJGsux;=0$eS(5LfQ3gR_=@Lpu1fxIwWhjI8TSjIBwBAA@0K`
z)X-lyk9eYHf*@c>sd`_+6b%2+wG~y`_ayO)y8eANnGfS~KHu0GI*fFBI$@cgOCz5^
zjPqbe*wT{%!{fo*)0>&u8=IYad}A6b&bldQ%??w^4d1={*`r_(#QQUUy%ZTqMP5nb
zN%UEic3C^(x#JK)#gd76?n$tdPMzir<0(eGiPmZGS8qQk7m4AY6OE>t{+(ml(A~`^
z`&5FF--L$e@f|zNOZm#ZoJ2jrhVCD~x3c_fhVblaA7uxN#&>|Fe2lyxJUf`r+YGX9
zuWaaAe|p=X-wRTAyWrZ}2U>4e{}Itzyz3kqjN~6eFx7GwnZ)QjM`fXPuCotLZAh&K
zYUdMzjm-iJL3v$^VyBC2DG=C0Y~N1{kk+OyY5Yeo)iAA#07QT6A@Ph<?8+=EEcjb0
zA+G1v*_re`xH$mi`cD(mykNn5=Z{Ao>pr>*qqo8nD|Y=VG9rPl3pu-u-$U7PNs%l=
z^J^{lHYu5z)*Xkxx1Z=&EsGBkD`-&#RQ~Qf*e1y+s<Fc=WyqBG!~jix;#qCAw>tr7
zQ-S4h&|p;lNzH5BX{)4)(h1CCwdZ~#FPuzjG@mCm9ytv-o;wnCluwg`1N%R*JrM(Q
z*CH(7>yBAvKuaH*Dmb7B(b6#CQ<eGEfzYIE%JRL#udy1y9W7sUcdp3XZF)O-5#cdL
zA3WLGrR={pdWbzL8yE4J)pg?9Nb&10elqf;k>Nx-oX(0Wi4C^?S$dNLDC7(8s}R6q
zS~o9&&r(!nUYInW0p9f1*4zAi{T6>$F1vr2+!PQ4CZ0m>xya&s7S!x3&4yQ3<~#PW
zm3SJaLW<y_6OA(vQnLoN$io<w*0$Sb>SH*5y~#;;r|UL%pc-kyUjZgkkOe=TrQ{y1
zQJ2*0eEda$Q7l>QSN<mmJ4q?F1s#}a*sZp<DBI$K%f(dpksLa^c=)!~`R|L7Ff_wp
zuom!a4@Q6L?<pv)qGaf{2hkrhwR?zc>4=6JLT-7v6d%j2X3a^7>Ln8M;AKv{@-;p0
zq<YD*AtIDqdkdV_x_oTGmt2G(Gm0Kvw-NymcjnZN>Yu^iWTG#pBWhmgT6Z$glX~~2
z(yCd`LRO6I`CHc1k{x~p0^JzqSwYGG8`oC7m5u{Iml9XoyiIB#LSZ)CfAX%Oo;&Qi
zf6o}vO*v^ZQu30O(`XhOC4&cO4R>66Rm^LPlR~>rRY{;;!Lak?j1Oa;)zNK*IYT~X
zd3&_&Z~N=t+UZd*jg{U_2Ljw9`TrQ4o!T=fB&C2>EmQm!6Oo9{buKZw3vjKMtrsY=
z$W3#_J#D%pB2gX!LRv_;f`+*oFP^X6VIi~n)e>?zaLJtf^=x6LqFw0tn!!-She~7>
zy#Gt)BzPN7L`GAvL+hcA(p(z1t4lQ+P?B>mM(K)74}wM29$W}Ux2Mo`aqXwYQD8KH
zendq7ff)cCy@it^NMG%FKE(aM;+3scm$aIOczOoq_CL$Fk8j06zBZIN;5sW63|$=D
zO)t(MGCz*INaf|ob18qmbbon-L9;7A-z`z1C-_K)S(EdqQ4lJrB4*1jZNj=Ux~lvk
zkR6YY8nnK+FyG<q=u!UH{RM5^9sOFr)8NehweKI$&!XtWjkZNAb&KY#EExhv9KmJk
zq=iejK0dv=HGF???y6F9Xs>*U<R*%5+anY(WTi6}J<wjb#%9q}TU9@SnC4;n?hhIx
zboez*iXbvlZ+Q0^DxY{ojfmx>6=)_TJM7IFKkH9l?k5T(aCXxLKXq`3x;+;|ITE?w
zrr{mdh>5gN4N$o=$l3(8T(%H$7fT^>ru#n3foG9x5Gh^JJ)q+1wf<Pw;{cDEkw8M~
zH%`nsp}S9|lcb4Jo+mxhHks>5f;mqIwr_0Qti3quJR}HcDn{kh-Mj(1%%*{B&^H+?
zzOKAeDa!)ALqQz4uAZtMx6dmDS6~)OtH=CVyzCss1<<~IAabe89=E3uZzQ@e(I$|}
zHkyxGF}UYOv>@PY`|jwSpT+;-$QsK@=&lg`He1lL`=+Dq?AAeP<J`>+1Oqi4hpLMx
zS}LV~k~w|YQlk*{CSk@<J+g-F)H!cd<hd^S@B5rjKDV)+R_a%Ce>1y{|D^us#jx$c
z&#sSNv#(xzc5DP8ys%jkOKD9<^IT$}jmQ^#2ELteX9;k61IP8(%CO+qNQ$vf`k32B
z;m;mt;3!)?{?2@8xBcx^bWG2wl9uh2lEr;S1sI+2^t;)MSNwLU8c0nGsbhmO#~5HR
z6V!XkI?ORu9jDC;ek&TH_w&+Xs<p@(S?QdXMg)Qrwd^xLK(u`O_dxy)S-9RDr|AsO
z`bB@e+V7P{)1LHLHL0Uf;hm0rP@Q1HA|gVEX7}(7Xl7~O?`M6-9}~l!UfS~Mk}k&v
zFd6IeMQPjNo|S{9z1Bt`ljd{v>5@U=(-7A!UXD3FP6eI3CAp5ffBGO%Fvj#&_j`G>
zwdbBpX33<%cJ;=py;-okdB|c)M#0VR<%qGSd53qZf5J3JzJ{|4ryt3~n`1cAVV?c)
zmF8Tz3+|)!1+_eM&_asN*z8wVEaq~lb%Lnb(kPIjsnqnXK1gf*N6!n;<^bj3Gy1>m
zL<9HHBb0+ah4P&*MJ~!C!;7j`RKGZo4{8z0i5SWm%x=1uR2vxx{UrL7IY;n!)m@DN
z6T(7F<{er*=4K6I2Uep2_Y#}-`+NPY7ui&xc58H52gfoYdo%nOyzq#nJ)GThdCLUM
zcY68>p_TSKv)eX;CjPbxzAFus|45w!^rx~RWzi2wFfY=(=nIWxYZtRD6@C?xi21vZ
zZMK2hE<dHk&IzEbd#fCGxqW9p`=`{u<J-4}F0{jqam&w7uM;Dtdc~%;Xq{F!i9CBr
zVrI4i+D4z?R&E(yX*Hm#E@iCnvb|;EkcsS8DF$QR$cq+-lZaoMS(2kg#SSKwmvPZg
z9>u0$yhb6)GB3YTP6%d6mI+j-@ZL6lhbC`v{|3M(gGc`Ee0vtawVAq4?D~OC&9=}G
z)P3`0#(~I$>|LJdZpx5_k|~(3Ie`@QU;%C$l^iJ;z6}0Ad!NH#WJG(hRVr99yKOVV
z)yJ%A5?fqbb`>dGz`{Sqg7*Pzn>WkQR7RTo7Y~HzpUZMHmanQbX1odUREAG<_-Kff
zG)zF)UBo@=%m4Rb`S#<jeux3PO$KB7KUWv_yV9$Z5|`Y6?Uh@N{{|zz@V9$|tKnkv
z=cDF_MNpJe!=UTWyHo>_+pnC7872hU{wQ^gdWoo*`Y^YS9gLZTn-&$=2n=1ZLH4r7
zhWkH^2bb0$O?)m$TCSVTPlvYm-wQ9vsVI<p_C>D1b$U?#5Lvro8NaaIk4KvIQ<Dxe
zTPYP5QlyKE*C*$*o=b*avztiBTtT-lt!&M#v#(<vuKuR09*b_6!*CEIJg$9Hvb+}%
zo|i}aGqjZ2N=deAIucBUgIr`95vv?-Zq0}8;G9paXH8uDS9g3$urkMnOipa+zuvt6
zj-Rx!{nB~N+%UqXV!K*3<{_3-OQaAbl&O`U%!vD5AU`iNdJN*My5Il3?X}TXlP>Z_
z?{TM&y~R`;m54H-^+g+$d~o(}fDW1n5h55E&H)RtelhrftSvmbMGNl0yXB_&0z}hr
zcQjK<8GiJ|4ucvYbFB}&iSm1IMC8>?ZHqseISee&``^sA9Gm(gKc=%mUN(DdC)+an
zNcEb%WPN>L{}D^ue?Bwa1o)m>*ZKg>pR4CPw->AuG%&D1ID;=|&(Ybt(4Hv8;Ddud
za$gCFqC%i_o-^hbN9@G4_R7BweU5yX=_{R*x>--ShUyg9E544V(-h6Yb-GbdoyG4p
zI^8oZD&3I*J=sqW6G<9j)iO62WvJ~-*D@$dQNwqTAz*`o_Pmo(YAeXn2|-(5Yod2t
zPf}?zOiBZbHOBm66SdgUi3pEIG>7>@k8nuOg)Yb(gkOvGXu{Aj&C(g6apTjG<-LM9
z?R{XmgCpB6|5_p#x>kTv5X#Vx?_bpFn@vnXi;8M~U+-?d@^4u8O!k0%KCWB>sO<9>
zygqMkjjk4k!K8k_YkUT+2=6Ozj*KpNfKNZB$&@ZV@)AC|+;37yh<(B^L#}b!HnMbH
z-OGX2k!<#s@oqU&vi&ejcdxj4vkUf+{ux8?GXi}ckFb%gSNYkXR2=RJD-$F>dH#80
z%5^J<Sq#*HcwoIQcCvQb{1vBaim{x(HG1r1;Bqg&ago<5ESKzfEu!f7ar1`^Wq7vT
z5I24N6Lf8DFjYb8x8iI!*-1Biw*fnt<bMw6TWoE@dej!+8)B8=++5=HeNxwFKmHl(
zW12YnH!Do^Hu&exj|qv;zYHpko86HdC=B}17ze`{b9SJ?1>?YWRY&$&uFTPZ@Wmfc
zW#*^;h<b7}8d5@Xie&hGlIWj=QL&_BEfht(Lgp`39p=3=u6~MVNM9R6y{<A&XEv2~
zD#i9|4V4x?cttk<cJ_DqW_<Z$(sPJpW5#2cxb|-q8N$4Zq-NWM+KyvBAl^yXj&R$@
z2eR$rqh+|C?%`DR^Ky2Cf{spPj@?fH2~K@&%inSGb>V8$u)a&eOMjozxvsK@wC0^=
z$Fyd%d5FyUJZDB~@APV7iV9^n>eu1^Y~zHv`uWF94H~g-cFSyi+x?h~)bKN?sQJ8p
zMj{4lF4ZR?`+GYpI(q4jPy`^sn;5-_Kc~AB44o%}o1JzT*jL?+BNfb-uQY)h**mn>
z{aqskem4p-SIbH|egDwZ6Qb)cpE6J>FJh1s#P>LP>DHSZX0gII516w`1{TY)VVd5D
zJ$;V}WGCK*{wAXjh=-+l*WqEI*$>Dy#zrC^50l^v4>Kv?Ro#7+SS18mG{)1(t|8})
z<BZq=!)#M(=dXswVs>bx7(?V^S`*pnJPBs*mx;8=tXj&6jWIdZl?R6qwSDIsOX`+2
z3Akszt_#0nC@3jJMCN?kZZ}ojv?c0d%8?V<;QwS-f9F}Ma^5qUBm`usz5ZxO3I#M#
z*aZ`i^kqa7sXYxjD9!y0rwxA4p3yXGld+~_Tk<Ex*wbLK3c0iXtgJj<Hmy@mzFAC9
z&YqcAU;JzGxU~U0w>i5TV31f}nEd`(fYgJ$P8o`1Cv;1eYw%3H)}G4YwS(6pz+Y?H
zVBdDr14-G9fOc%B_daBhVoRJ{R;KFLPBLb-Di=ql+WeeQR)4e{kAe*6lJ%S%42=*`
z70^NK)a7TBsiXM4`ja)o*}r(C2(Xx`MKGMK9tcEih0u(XmJ2?q+9l)q*|Y}K{T`Br
zwV3wNDe#5l<w|JqZf#IR-J)v?NfThbSC87wM(y`1ET2O6stxy>2c4U6^_dM)`^wG@
zOO_2q+Z9|pPAYrfc!rDl*Y<;Gh3PRpSxe!Cv%!R9Lmkie8sFhO3)`M3SS#w)Lg_7(
z^t!@G`il$u$izTP@U#D%3#TU}GTKyykjXb!SCBx1Lx)$KRliVzYVgg_;jY(I{^A!Z
z@UfiEx_5q=|C-Gb-Y;$*wOWi6%02nCkT4O=*e5mo`kp&_udDudmGBoB1O`nqb@mMk
z5;)g=N@U)IZDc|`z$Y@sSP(8?MZ*G=94&L|bLHoR`T1UnZ-kOYXf4=W2AYVB508ln
zk-j!n3sA{vJzInl{#ShWdlk%sNj~2ys~S=>1$ti2{66t;jC>%_37MeOXkz+@o6ho_
zv#-C#0t|M`7@1_1XOKpw5FoyPb80>8_FRtOSXb<pNs6W#^8>o=cLBq|Uc#G?eLR;!
z|5~&%yVD&i?fZEfjIx9N#4?lKwdhVL6LWK_k``MIW#wl(dT{#7Ee1srLt1Z2I)j?p
zYWV80-;(Fi=6H7-hQ2@b9YSSTx#X(SY%mSxo>Qf&h`)jwHuSg5O5#mbTke6!u(Ez7
z3r7=Krvq+WF<iShU`(R1=+$l1&DG!fAuh4mF{K5^vP_YI(;c^tEU!z1=YIZhk*jRV
z9xhYn+UObJ@>$SYAT4v&8l`+%oa5h}U1WsZouwEq6aqtrr7v4$aPw|x2Bb;i8sdpf
zZW)X$%>aW`Bh&=?n7&{2A1s=vG1Kd!L%7MFDnAsyDGYG<H|nR)WJabB$iwWksUAlc
z<_c|ZZQECbyRV-xBJk-K0ce1aS$=bPu6!EB5=Whq>paEqOhU;L6oV|@Z7R-D4gT{+
z)MX4mL&&;x(qEhnX_X=EqkXOZb|JBu)d<Dt;UVgw9pB6)`n>Qr54r&9Au8CUl`T12
zq`roVwEvT8rK+0pQ73+4vhcF#AVJ4Fq^?mjUgmd>wwOdyv50nik*4bOBZ&3_92sCi
zdSkxuE793CN8w5`<KS;(5U0-zWqD??2pico_yU}J#Yal76AC~)n241rKs`N2&Wly<
zEIdAm0q@?2&%b2|2mmSC_iA|VhY7R?a=N?|H}1dWZa+IZ!@w4Km`KvfuG$P8t_v@1
z+4|u3qWT&b*IeYN&=FKLu^O}gX6G=Z%acqubo8t=hv!l++xDM-4Ef+iBx8ETizH%;
zp18x<0n@8xDj(#|(#`c*8ixS(K9@FnA)L&fPW+=;Hj#A%5B|%L#Djw0sAQlkRwhAz
zQ4xXv;_$KV-zNU3PlPc6v<l#x1GwyJng-3&tlVUlSd*rK{I;?A)V?XR5^0;L6Oz-u
z8LC|^FM5Xz<Cek-Z+oq6Pqgb)n^uZ~NXwjSipO_rzTLR+b=9ddEq=qVreub=xk;xE
zsk6G`9^gu~+)BC6hu}%EiP;v)t({7$MMOU2sl7uEEQ8l;fPC9yYquV<1a&&FtU_HT
z%y4(ZC@XnlzBPn>zVcwT-#z=&P{?8>nqXZw(=RZ=Rp4I5X0HL4cK1%}x|JiA^`3p5
z&HcQlu7L2k7iDg<Qm2G?X5`TEszN!5bd*TaDzNLptHnQh=-1c`LwZpW(V50(0@~+W
zPSD*83Q^}yJAg7cpLuGWlFg!{W2|MK4jcv^wujEMt5~P&<`i(r1Q!+7rKeYbWeAH{
zy}&HLy^?5YiVI+VpEWjSQ?4!)U3Km+O-S<wSd=o!bfhPUutQ}oH^!6vj|T@;Z0V+w
z`!XL6=JbECt5kC-T0ig@)Jbr|RA~*jB@V4BsGmGcw$00K2+vM%9hOu<_nF#+zCoS2
zkZ0Fm=})fSN#jJX4R*#~ONZ+D;d}Mc_g{!VRke5dy|iTNc|}q9Rp|WkQeZLK;IKDk
z*}MoW)LYEZr={0<S+`bsF2r)KvI;ji`(4?3;kU^EmI+n**rUnd^YaF1wzAPKb_c65
z1tbR7JhrT?H3=OsD3I^2l+NhKVpT%izb#=~i9Ei2xC%MA6hu=hxLq@c+i<`BUC3e{
z!OiNnmZVn?!(rEkVm_A&h|nxWs8;vy0#U4Hw_yp)pFuC_Zs(c*aD^<b5eTEgqgN+Y
z0SnJY6aTF5N2I~4HM6YE<h$PZ<xr`l*<@I~Xa1{lUVtR@*q<9ofzRy8pQ*DViYsQS
z^qM{U6%eRGZi8voh0D`A27ldZz7!tI%cVnF+SB!{Q&^XHMmho4iTxk#9E{*OQp~ZV
zA5?1{!|<V>Ui62kgnu}_)PW$>eS+|Gub}zL6Z)Peadq4&>87;R;}+9CmbY(7z7tZV
z^9(2+LJY$W(Mir(#+(iglE0$l9wC$lh>wzm-NjoH*G@f&aG9idd8Qa#b_(W<bAQs{
z_jYa_%Vs|BZqo!vd$H`*UrMcrR)76zJ4nY#blb()15p=Di|(|(tq!fWWW+bQ>mqB5
z)Jm12hBxUtnPKP?l`&yH-0ArCm@r!$vRlE7R=L5YyNxM%yqMihhsK*DPn_Ofbrlg|
zOKz=trUCPF{ouRRFUy{eY!j4utt~0sKA5@p#qe2x{O9RZWnKjY6WZE>W?op}RZbzS
zK>kMk*jWhbRiG|EJ!W(trYi3l;gygk_q>c-Tt3L<7JulWo-deW01N>~UfIFH#Ge*Y
z9{=3wVvKyi*fy4rF9qZME&ue_ctc-hBbYd^^;^e*xxe>plBoiW2eO*60_?&N24N%)
z@n|i{_Or6~J0DiCb1$(R|0a3uo+%x&Dp3gL@X~vW<C5P;P7XWaVA{cYo&{L-0V$R6
zf234T3AtYgdCj}@Tz&krOGvmUfTzMCZ>wQ>NG7&eP(2V?Sm7t76BVkN0C7IuTaelx
zwyF<xZzLqk4JHGNQ)(o4#utr^e=olZpTBLhOp();F%W)1_%f-Es)SW|(e+4Cl7wI@
zgB`U=1ermSWji~VBH}zc>QX9x@YpQvP2rZaf$Ye6iiKi0JEh8wmhI+<L279cD?k#M
zPUGZC^P^SVx%LF=UAOCY(RI2q4*6af(~N|jo}qRjVzTxztykF^sp-9T&jMtEQ_%LY
z!n`DJ$NY~nPDtmj%V2PAZ*P}d4?Myo73n#<?!9x)qHiB~_tNms-_9`?21erOmvBTr
zPeX&+CQ5&CxK@jxXQr$(Nt>Iqe`oieM-4EgROccexxe3{ELL}_7lcIB=mD4CL7IL`
zMXcJ>vDhtpG%RX->qsoRc-%`%%h@ThXktl>)n463H^)yLO%H8wu8T@LuaBa?sf@yX
zEN_#z;DtJ2IBB0>@A~zfOHnS>j3Rg2kd{uaxL?$gz46%zez4M@@t;gJvB=eIOyb?n
z0b-l6Y4>>@*imuW*wLr7VWfw!%?s(#iFY<KU)I75Je~xA7${mZLK?%DeOfxr*!@Gf
z_N%IiKU8&G`=S7|W5)%L=|Md6PHi#2nbKpnnIyu{BN#?W7Od6y;M_rRe2MlAtWO+F
zAId}zqI%P)eSl}AhEw+Qz9Q=BtNLf%T)*InT6=8%;7+2RJ7;`I#-IMx1wa{T%a!ky
z<TrZ@TGdm9zVP0X<$r}c3+WeWl7+c!?OZ954@v~i#`4zBbUrb3tZ8k)fNsu}vPZ3`
zd4&`8OneIeArI07#`MYWv2M4c86~owh&_7_(Gx$nt@u9Ghe$8ZbX5i95`lr2m53op
zpBoPQ`&;f!y3cR%hvfgYQ$(=kU8uymCJ-=0P0b|sN%DG-#<ov{6SjJ{?RxfuBg{_g
zDtW$YOz%;7G-OjDY!PKA%^y6ZHtUyASyl5+gD2xoA8;a2=N<g*%L9k{-3zk!Rg{qL
ziyWj4yTfg$p*EZW6!ZOTERc19mqhy9d1FGPH%(4N$46+kfCpq<ikJ6&2~SK?aWD{n
zv|Q0)t*9;BZY(U=PWM0N(^F-%RJwF)jo#0m61IO)E2LQ&8fJ$*HR|lK=iekNmYdkh
zaCBr&=a2DY%ITCzDy%KZDcX0eR**!?jt{4NVsEHdv&44w|9(TcSdB8Fx4<)L@2vE{
zR;3lO2sFOze9^M=o4b7f3vLHvJMMNk8{~Q7hF7tqriY5+C#|vViBy(6-wb6EPK+H4
zE4iRF800bzw~=?1-_ns?N}Y3Z>mtsZ6><9T1lV5Z=VJ+qfH|cQc6B7U&4eC81$AyV
zylAa7eE3_(w-Lz~5DNI2Rj@r|&qJy?bU!55=p57Hirho*7IKnZXFmjJGyHc}apd!8
zG6PZN#?n?*{KBv9;rXt88rO|JUJ+M*t=fGbnbZADTl2{^a)na%f~aFj%fgEU8YEeb
z2kk!FZHZy*&222gwC)p(W_H}mlOX(9y7Te@uT{>?)ds@s8X%zve7V}v>*(VM!TiYm
zDewgeR?|%RcdB=TtE?$K{q~QUbYYGFs7Z4T04Yt~LeJP^GBX<Ra#;SRQ~BYR$-N$w
z3F+PR00m*0d40vC2piQHzvrT;@H5kDM2`4$^Xazp!sSU_Hcwg`Yg&RkdVEs1<pcfp
z&Fb@0EFt8*+lokrbjjin!*y!~wA-nEA}-kJg2lN$Yxtsy1a|d~RtDxuGgBahNa0w>
z=d<?FW=}6c^s*+%-Lt;n)S)M@z%ewwv1Lr(K}~8)0hwu;YuVj8UFW$6yFl|NkIF_+
z5!~Qi>$8UUFASU^DX*yvB>Qvu!`F8>`*(prG}^Fnp(tW7^nS_Aa*5iqs{KlpIUO`N
zVFj@e6`&YYSi=ZSTGgS^r}IS`@tx~3@ex-G$IGu#$8)=mg=M-m4F1sfRbO=LlPZlC
ze2tspE9N`kC+&_WtZh!expnLK@O0}m_)s%^=le_Ai`@p6G{3&W6_fXiV1Dm?okgaF
z@lwN;n12k8)0W2-Heb*M9vhiM?3wchCYYm0h>30K!sY1|s>XPy%i)RUGW>U|^!P){
zwkXL9quySyEOvbV_%M4gk|TBW(dhB@8cv{$r&h;#4b>ri1pB;yH=r<BfG^N)j6Y8n
zH1#Dg?R@O}9^CkBb!$J!*xBf@$st=|&N53R{F9w0_&>zImrsN3_{9HH9LjvwPt4M9
zM$sM=(HjkVoG35!ZQ!jdDb)MmC*e^<3NfD_T<z7mz%3gCeUttHPI^u?Tc&R!q>=)r
z$cW;;y#=+u3`ZGuwYIi2!Jljx9wM`^tM_fL5?wsE9U5knry)*dXZubve7heO_;>AP
zuoYHirr+yvew$T0qYa?%Z&nr&m!`0uXqw|jCzsQ#`ehgSijA7C-Pp41x}PyowGG`G
zwOkR5c4UCVWrlbEp#Ns%2%E1u9@%1@*W0>N=$~$Rv2dGhyNda`G{$r{#s3_8puP}U
zJHHqW9Jis_{0a$|B4TSgMoTnV1p55#nc=@!9lmJ=dttXG0E|piMmE{WSKhie+Sg3}
z1DuOjTjZ!6tW$YU(n1AviEnCqIlI2oTE?-dUCl5BtoD8y)>aP9ah^IEd+a#t;5+w3
z&&irf)=#ea?2$vM*4qX_73Re8;^{&W9oZaA{N+g^sQ2T-amkbdSSQ=3TOG1`5kJ61
z)+j76@rR*}*|%Pm>>NoVTt<UEMkh2B?bP%zT(}XGW6eF|5Kn6NwDq^d_G0C14jzMo
z5(QcOk4KeIfE!fC@Y~#gztt4kf2G}2Qm6(xE(RexfQ4l*VN;yxvKv|Upr2bt%0Rh^
z#Xw+7i1elJ&s%ADxPOSOjV?l;v_T=+u?;u7f?+j#T^&Ik7(nS;Z;LGdEV-E*ywiF#
z)W1UBO_vxdcLp$yk``3Hba@=Cuxj?z0g#Th$GQc@u5Q|3{vU+IPh)#tPFRka4}9X|
z6u!RP?+9p@C9HbW@s6Qe$dQei2hSIS?DxX-l=ADU<PwgpFT&|P^{=CpWBh8jzF?}|
zlb5XJ0{gm!Lzmv-8ys|SJxrw^GoQ^~`?noR%85|%;Ey`qeLJK5&0T2x^|v2JZ5cbI
zx!~w|e|^wmi@BaY1juE7WGQmzA<bU^GU$VWr`{s#{$RA{RN}P=+p<JO$)aNWSu6e7
z4`lsCAW4N1O-T7iE<(HtE7>-4sMN_QVz)++e~x!Jx^{&?)|xw8dcGMX^-EN8C0ao!
zr}{6;VvQF#Gu~EZR`yrMnc|}F5jLygV3UPJ(z{C|wKXw0N`?N>8e}7gHux#n{|{ji
zI66jeP4~gS>3DamofHqC?UXIQYnr^%=L&5J0Y?Xk;khs@&XoOGnRp2NNb03=xv$IH
ze;eAEv3AEyYWRAA$j1%xNIl2p5CvD7G5@C>g28F`XfIOdGw*T!s`BZ+?vc2ExuVnB
zUBu<Nk2;%y>|JEx&P$&={Z>mI-O1{;9#$7g*++x`^tUeGW3=-E_PE*^r&r$=w(l?<
z)5=#k9vg1c9MDIG77_EeBhg3z&Vpj85uvxH#img3arVM?f7SRBQ3(|{d5qeHO^%(V
zFbH{bj<^4Am)dw`C`pc$_49D7oC%{h`4jry9<eh1jw?)BL7WqBhGVS?yK8GEGYO1i
zGQi}SJULVO=dOP9dr$CnHpB%ik^YPoe&dPP8Fjs49AcEtyWgjW@QJ<Fj4z&q_nciP
z1HczQQ{|`lj)3Bbw3T@=cJ7C#M>;SI5R@>u20)JVtChizMKChX?ewgoN(mXf&A=1n
z%*3MwF|5~9h7c!!{pRJ^fjjNb@f(Tc;2-ZeF66iV^qG`X*cyvky0!s8HnvkNXI9-V
zfId}c87;w{Q?4%Y)}|Npp{yc#ILFNKC3P=e!P7ersKXlLA-X{|@9c}SJkzP8wc6u4
z?%FFREIkWfpO=@+1SupOOf1FQB{LfRS4lBQvcGl=9TfOlu;dpefBdU3dt^-T`SHOS
z+WV(*7FC|2e$n4o7fw-jx+>7P6}CLWQ7Z!nn)~`RpJwVr7<~dT`YqeLaHnDOj@oSh
z7F2CYxuY@Pj(tbdu9qijlW?arF~|Q>*1|ZZc|2L%x3R(pOYgZF$gvZ32%1|J=`|a+
z)lp4Iuq$SzSJ$+!RZm%~z4-Mj>@CdXyhu*zy8He#X}AQApz-dg<G=KMt%w(>_Xz)+
zlnQver{o$1y04RGQ50LuI3H+uQ9(uO(3ge@t*L3P8i4HM01R=nMs_TLKDxcBV_jjI
zm6DcYR*j@;E|qH^5FpOxSXVgOY$<y>Ap2T=)7gGED%+RZrYP*zuG8<M0LO(P(=K15
z=2DN{rRNm=g-$2ne$?lk;r(%xuW4O+*7do)=S|DmTu8Hv&kr%5hf64Rv<3{dSOV==
zoNTkC+-9GFM@+Qd{JlKY#2Jd#A38zH4tb&Zhgrvd8pDB!``ra!N>pt9bjB$)Zg_s}
zpu~l(GdV8++LcIx?7hbYU%IZar@Ibi<+ZBp{>NnG1U9Ht`(S(_HS~Cx-`|EwUBfju
zngjWBMrlN0Ecd;~{-T87nt8YUDAC$3GFj=Hy%4q~N#c9U=j$1eA$6s^OD1xed;9mf
z%gz45zW@$B+$=ec`h|_K3fas032n8tTJ0wU(*vHazBpCx3M+u7XL_{7PD*cuA}xs0
z%9R77sgMB;&I=ebS_>~=5v<=9=dKz8Upig_=~=p-3q|tAL(mUf&p7J}a-Y7Q&pgX7
zIho3?z=^lpTy5?)UjtaJY~(t?H%cXvL?C0GhMNDp`q9vS!Yg%szJj<O*#<f{ZMQ2L
zOcvjV@G%uWxWh@+@fb_4(#bmllwT`&v$70hH6M@uYI}6GERcH<14wT_^h9Vr(F6xD
zCXz{kG(UBuB+s*{cC}m_m*SJ}jkBa{_)H1D*-{YJG0C`KShs?DZFx&wXX`la_KVf5
zU6#jH%6<<H%e3FP+IWNdn`d6%$JL;JX1QNKIE0PKazlMNJ<b5l6c8*{9+~A0lqA@x
z;)9cN5=f8UT#=6`IA&lRZTfzR*hp{`4iq>L{>p^m%>(CZQq%QP*YPKx?1reVo7#*Y
zDj+-(t^XYZ#BU<l&+LbO?d-;c+ZfZV9{KkZoPIyziWdp@J@`)OnDdix#zg+KUIxB5
z%RcKr!%)A?+fi!_L5+Jd)R*Cou@&WKYh@@*%4z)?wlWiY@z^P?A3EcTgU)?Mw%!rW
zOz~=1yYy^`B5<;h?eL*Ox%FPxUq)TdW{d4RT#_!JNE@yBQ7g#{PAP_-2xIuJ6_f-u
zeo9>*gZnV!T)$CmiUM;DP&D91vjeu9CZdbqyFCNtyezp4x5L^NxQ7~eVzqjiRwE6h
zZ5|q&k{ky8kZUHDIUZ4u)|5W{m9{j0Io)xcq(a*-m3Bz^b#0s`meWSgp0Jt9o+4#e
zX!T1F_hBlLRd)>Kx&%*b^*O@*NggsNklcJ;B?p_^dRfPm4wUg~fU{zc&w6N(Fn-)^
zNmqcNxi%}S*#!@Dh0G(#ir3`54{b|$oOztAr;^vc)mhIgd&Rwaf^7aX&drhI=h{vf
z5?`YL@R)_;65C5UV&`itd)B43`F4&G0|i1X5qzU}VFLxO1b3TXOvjoe1qI1C54;!Z
zLdtOAUDIXXn>BJpvLOP<2Z!*`0^v4JjN$8aFWB<sM#v{dncFcJ_e&1$uk4z<KaxKf
z$31#-$hknI)>EuwyFoZQP*&{?9dcyhxWisL8*6mI>2=NbzOPI623qLk`8(S__t4U?
zYWTX|f9c3+<I7$hL(igPS+-vWZaELWv*gw8$f9^yv+@W&P8e}RQRAqNUhN9xgfb)2
z?AY(j#bQvYwdynW<1;-p!s)qv6^O2srqcO|tpCm(4`NC4W`+8M{%y^fO9~0m)It&-
zA1tv^iyFbldQN=>L>EUoI@+>><!17@N3_^wjj5H+ML;{&+Foo>%CHJmc$%`^@RB)3
zrhL0lrrb7X4?Q>7e7v1qkii?ElCqi;zGe+bfNJrp2|X>v-V3$vYA)iTt^L2bG;^GN
z_F=;ciZ1){W5!qvTLs~g*<OEfjjr;cp7PnIeTGR>s0?_Eq6`vfhcPN|Kqhdk+tNu^
z_=1@mM{Qq{Qop!f<&Z@RTW@5#X)b>0U6f58uU$=`7EaFm{6nrcRSuXkX<AtGWTX(~
zIRF|9Sl`LryE|&J&yFa>U7ihd^d^8=J-bN62H-eB6mnmq9mk--n^j16Zh5*K!S2Ec
z9T0If-J%S(y2p|u6kH(plq;vN_o1oH9qI=HEXZ62*M1>A8yjkU#PJg$rL?p*tEuGK
zSAFU}jQt_DO^%f|T&D9Wwv@Qr!sP?6^9*|>gp*_Kutsr^aCX+L!tdlT7%fh7lM2(m
zp6;+-*x!2V7RGjo-9>E4@93^G+EGn`5|Q#_=LP>}^=CWZ1(i9xPQp>6YH|fP(%kqG
z3P#_1jGwOjo-Tj<y%)4_iQL=&yWlVWE8m@ihs3rWw#fQw7=9#D48N&E+TVI@n;=o}
z)g^s$&0jRib8o?GRyQTvuHUq=q~f$8q2dzh?^Gj8QoR?@jDq?6Wbs7e_~!_9XtX7{
zWCvBY#)E~8FBs27lKWtF{Nw2l;~Xu}ddzwDtdnM3>V0@th=`y`4(`4?4`3{^0*^-@
z^)0>e+}GXC<rN4wFQE&5PH?v|=$Ze2W`!x9KjfQv;kPHBsb;;K9W_wf^~nyr7VwDF
zDHdfDI;@QH`)i#cBtbD~6Tbe)Vby(l-M7X6n74jC*TbppHtn3^wDY`ZlZ_=x92qw4
zY(;B=h;#Bg2~!cR{$8~jGFv%?i?^;&T5=z>s%@!oeFHH(8v~)Vv}b^86HB6vo-(LS
z$-O4scn`-mIoi2qt_qUM?6_;Jm}6?Ev$Jf|hX;lMe-!KYet0-7+@_t;6x;nU+$P(3
zPN%x!ZlX{W<j*`zE>N|iJ@TBDH1D`=-=Cjf3O5=9vjU)z?%5TqeLS-d)p*PSWP6{Q
zdYTLiUk8P-Yphms;uyU8I02%=5AZP4I5ClIJRMB92drPxq;2R8KB;2=-SaWI-*>Gz
zM>1@#9C6bc0kp(E(TnI~$|E9L9idnLn!RBpQ3toKp0@-DeuHQ4X5Z=-0AzU|uIlcm
zvD&9E;~_RCT(p-g2DSHp+H@xrHu3znuHw4n^Rd@3%2rE_cyQa;(j`z~|JE%t2^9so
zw^7f_p=W#b148{)^Z{$!rbt8zIb@LWsFtj@cTwA&3m5mq+KQj6*%B>;PbG`~#MZn@
zVlYCAcRx?!449G!<E-gy_o$#Evh+Y=g;)A4OW_;lX!*_B5(5z1(5V|@qpaAv<6`ha
znfZH5=42hm9YWAt*}g^{)nXYUZeODwuhM76H`6c5L*`W4@bp`mQ$#+b-dd}5;dJ-X
zbUR?F?DQ(NL^p-Z+sr_lIo>`{M26&3kw$S`fWlDLs^7*6U#vl^BpYW{I6I^ZIR1j_
z?z%tyw_~pH?FPc#$p@3)Y?b##53O^LsyA4u$(u-Co=VvVp44hwef;#?ukI|1(n#X_
zNiY{^1~9^{<>&^R$He%zI$?$qPR}<fPB&X(=2-T@7BOuJ^NO-IH@=&o`pM^pbA32n
zRRy>19oq(&v19+W;q0Lo9ptRt3oZj*mxKBaP@Ybt7VpF*P61zb)@nTqkvSZqJY7X;
zvHSF8t6yue-*7CYnZqp9?<Z6ax1D8bd^i_?jN0{a#EP-!Nm!!Grd_x;X&&~-$&zc_
zBkfm~Sj&<mCbF<08SCX?f{(rk4_A}6k!6Y=P!&3`nj#gtFJZaQh#qecu<VI{cM1|Z
z>T%=<*=yde?7tK@V17Wf+D`n9i7nNj@BKp!d_=<~wZ^GA_A#IZ4XD?WB^XfeX<Ts9
z>JdS?(zxEv@kFta4Vb(XlCZibgv>?))o{`MH-(w8h8UG*a;+_%l?Tb=FT1%w#eZvi
z;$3U%lBg$=^h_kS1B1Z6p^Wliru`js0Y5v4ZFDse7G89$24hOusd)wHFbug=|At4D
zTmgZX?|9FB0MJl6kBgoc-_~Fb*}+kt{3P_88sXSY-OjFOse(?CHeGWWo|Q{Eq4UFL
z;+05ie#bN}!rWH}>pKM-P&NTA3$XDUrj}LNs<{xPw6eF(f|ZAT;4enbVG9BbM4@L!
zH;?+?*&4WBW=2rz9?bAcWcJJ2wuPBzqU37nDe<UPMA6oA>zy-34J;JKLZDBKp|0*H
z``f_6PclfNvV~&T;Bv%?WC%sgj;iP`iBA@ZPM`u7ud3+%_eblek{mYnIrxWbr`1|!
z<r=Pr2x3})c58q#w340HrLr~bU#IO!gdP`78d_2=O*ksYXWvG4sCu)nIz8mcz$rg{
zs?C=~7+1;GSTju4`CCqC;E$Zej83?eT1wN)4~@RHxoe2bqQ1ziQ&%Ckd$(Uzh2+bA
zg#<;);WO0^%nLTM87ou#`b-VB5kxlRj9MZ25S4xW9bfn#cw(6IkYJtUP0Z!P=%plE
z8P@c=e=z8c4Vkq=>%>^oF;*wimli%e!YKofTrGGn;NtSJE*o!WNknHD`ERVz+{27E
zUL}S+L2SfoPHA7QB>1d@UJK#fKMZqU)=V4y(oQ)PZ%Cnm>H5L&aB$23bL(poxDf;>
zWM5;i?%Kz&UpPeA5v&uVfgKB`9ZTu~PsyXRd#rGT0Hl4=$Xccmx#57kG=|7Q`xgoK
zEaZ>g(h6efFVV7)hP|G7i!UJXd^>d2%<!sLf$zo}!HelZ?ZT0*6BNNtO<i#I8(PQT
zlwmX1IEK2j?n#3!<A<Y}HlQNu=hY#}9!)`4wDx8+3T}Ih^aP`hpnro7#0C07(xy_G
z09^&0m#I7@OB(g{MCKCKe!gwbYqo&mWe``C7)MO~wn@{XC~aMm$M({Aa);mD?FGp5
z%O1*&nIlRQ<q)Co5+;mATQ8RuW0e8v@WSey@RG%lAscweP%RI`L^H2L`0fcvw=*y<
zO*PqZ573j+s&lE9KYV;pW1&l~%`PMO;&)Vn4wGyw?Wwmjz?2mxn_!~Dd)v#c5JUay
z)NM>_g|sI4T&`QZ=KE*6%9g!m1i8i*yst9}<CBNv(9>V!&MT>w-k$B~72}dXCP}~c
z><dq;75H-+5-xeK$;zi<%!Z8JJhe^1bRdzhVcs>(F@fy3vn?&uI)Dp5Y1T~nd2>jt
zjwtDF(A&>;+JxxfS0eU!t5JT+j0dsrF$&bZgjkEPq~K5b$t}6`1i$uylJOoB@D|HS
z@JDxkaMIo(%h3w9TZg<a`q??Lt_ZcfvTDHu={a7yft2yx8hKBPGaN@*DjPs_o>>6w
zEGoi|RP9tjk_=?Ea4p_@x_22);(aBPA@ZDJFp(hmDFH39QEz}*_kL-RJ^5f$@W065
z|80!K7br0N=lZ-dpd2(PsJLNFk6>JfaI!?$xmJa}Eu!jfv$|Yx&8_eld|%)9u588=
z&#TtFYT-jdhHYJgy@m>*eI()N>JRbcpP%i9^hwpQ6<_+`b^!lGc?mAxN{g_JeVXj{
z@{xgRze-(Zc%M4muW<iNp39II7}_{Ky>?-u(U7+_Hx!uP`|SRCw#h-Rs@zZ+>T=gS
zZiLr>jq3<GoX{R!j)^i*pkywR=3X+gH#NeI&Boi6Mi%slNh9%q*-=j9lw>RG5oO)6
zQ@lFP{aHzpBg;LWw{GO+>2H3RV6nyp7p?Ul+N>QV>tQxnGshh+;OHwK@Q><Nu7uS;
zz9+GE{62ZeH&5lMjfBMQ6hEIlmGqF`F>QrBx5JL}4Bh)6O6jq<=vsGC0b6CboQMOS
zoQRY_U)YOc9GZNP`jXoK_S~iKEp%<c#dEhSe@6_LQ(YLtV5i35Qi0>77zD9?x?o8j
zW{{ZC72lJT`tZY2bQRgK^PQ|npT1|0BX2HstS{#|GB}knJ#a}GZyY!(09M`MHkYpp
zX^(-S+U4}<eYbmoU`)?>6k)N-OIk2ovP?3F%)DpiTI;pFh)`2HuJMQOSa&q;ZFmk@
z0;>qrYUUUH%wyk2OkrzIc(h<qpbM$iXFu?novb~k#6W|w4N3Y|bXiy+B=EbU$dUh8
zk}VS}A*OSHrPEc{W@<(U32?jgg>+OvYIKE2qT(IPt39udjGRJ_%#SaxNs>ful&Lrh
zg|>Ujnfsj#R3AuKmT)6Jl4~R%2nDjqzAGNW`f>YLu0{;|zvaz(=aZ%~;l=q*v5)YA
zm7nL*+B1o*>@)mY{8fr|Tddu`3l7n;9>?W~p<0OF3T+yl(7hflp912{5B>HH120Pl
zI&N_L!ai)IcOq=51r;K}scCD;;q3ko`fYrue>}849@_5$9yh2~Bz3Kser;m*N9O;3
zF@0jd&jA8dpq=P*eH}}GA+4H*hd!-WvQ;SU9wSRNkl>GG$z_7epK<a|NEV2ym0NX%
z>(=X*zlU!`Jv&zu6@Z&KlN|O#V<J&-qpf?|*LcD(Jrea`yCM35F2?j-`6TBNOY*_`
z{0Sz#?;97dL~fszFsEC?=DsNSJ3d3iw3?}3RKo2<hf*st=f!8U{cni2Zxle6XLuVk
z3Hvh6E5)p1$yZ}5Outp)_cG4(g;K4!c!{6KsgHe7AEs>2DsO$a_E;F3w|M9m%=uMo
zesvEe>ZyRP_Byya8Xo=?FB3A*Di?cxMW_%b@A>rOGwIHDh4vDnAnB`vuA?By=9IB+
zO(XFB?Isqda0d9=$8_$T*FoYnbcsRy{^Uy&s{#N!%6gd7o|Ctb7m=LxvUx^|sN@mT
z+ozS}lHj#YIq-Z=uy5}JL<PCMrknMMhW_3w>3RL@=6_N>^{YbU0J}$G7tqUs{yae$
zgZ&+O@SDmsQ$-3p+MsMuI`H-7`_E-<UwMM84FIve?@teFGr#$WSV5AXTsKk*PU3>)
zwMw0xx_f)f9EeQt|1D>kIm)uDO%WmyVocY0sFFD!YlWRiQ#iU*GPeD4?b68WGU0TZ
zFH7<rbnT*CyJW$g?VGc3V8@)k!cidCGtoMnu;2bXzPqkDh7|8J0m*rkhyY2Rgjejp
z<`A-X&2lkvBB*F>xY1;`l;o(Rroa2*`j3*o$RQJ45n9Jl^++LOLPQXBJf7Kie^fo@
z{)*09znfGoi?!JSbrBuh&ua>#BiX2xnR3$zhjVy%Z0AEwJU#pXg4l-y4F$o}vYkQ#
z@94&u{oPf>v7}lD<e~BhEE=70?1vv$N*fC*s_*At`7qI)>6!>oE^gfRDlC+KP1?!X
z@r;An_yS5Ejmt1nfeSQO?Wr&bZo@G0;Vb*a)kcM;JHO%>8nJ$y{HhruMERI!WZ{@5
z#updtz<;wrI>zL|7H+fg-<!YruvZ>3mytw>rnr#MhDJnvkYhv?1*P2H#VRSGK68*Y
zOlna(XIx*{pwN+W^*`<9oz`^rs-vzi0v9f>p3;id@OeBV@yQ``C`&g0XABJbL`}AW
z{5%v6Xh#KsGBEUCG(^FlE#&qvraLK>$X9+C2J%dy9}h<-9_r-eG{sRXFjAx|Fr-Be
zAd-+nB`f7NX~}fO&spzi%UUMKr>U_v>F2qkFMbFa^bRRuMswr$zyV6L6objRegXp4
zAFG1Uk*$M_-1j$jUpQY8G^QL{r8GTSzGw{ZWC7Ff>sjmR7YxLZ1%vbo->g4|QfoSc
zN%Oq#coIxV>wV*&VGB81d%W5+xe>%|?k`hzpIi1PeDdAd@k>cx>Og<FTNiAyz*}SZ
zH}&RT{8z$6F$~E|>&EvH7@d%oFfLyA7u}i86#xE{!QUy9-fd;vvrj{7h8l0bM%Gm5
z$F=n4Z7FoO?GArKbX}Y!`^@U2N<pt@BwVok$>JbFNNWaDwld?5qI=|j|DDsXs$!dh
z;Wnr@Td`$xZc41_j!CJDu>{u|==9KFfBi{qNEG>?m8POvOz`TE>+B=Z$v8#jqUZYl
ziwSAit4Gn*(y|z)7t-pfr|RodFBrm=+APh!{q%3I!kou(W?nU__KAmZcUh!#%!k7h
z_Z}9$nM*no%B}><8mY;yz1kC6sf2Puz+_;VeRd`u9Mehc;cd^`!XDjArY@~nB5R80
z@pgSipRW@6Ayv=dV|rJ2Z};i_Un_G67-iAIa~(<vM`tV}$qV+PpD*RxMMa;_Bq}EV
z6soqXonG`_aBg*}j-nTyqoFODODypd&%|YFAamf#_cI7sSH{hPhcgb0zoS*|0P$3^
z8`$*7{_jn%iG?pK)J`+w&P8esCZrI!Eiy>hxHicd;a>e#G_ISPpJco(en5POLx!6%
zt)!3fZB}AFPa}t-<jO9afrDJH7!WrF1&-Bh{b$KL88Jtm5o9s+0ilz`IamJgP4EXW
z(`|{_RH0c3EyFQhN3d&HDn%7plS-dECku|L*~(1AD3nQK*57a5oh*m_q<@1-cBe4o
zjJxaNxV1_334SaYhW`Id-Lj6a*wcCZKYYlJ@3rwadHA~lXqdX)E5e{qg7~I&iVfpB
zRi#N?{CLEBI)%yyGHMHrC=D)6x6fViWfOv88r9wFC0ti^ei%l)r_m#11LwwQpqf4;
z^Z(dvNL561at0Z&VG7TRH6hiKy==l6KC=#6H$Lp@4$3%j;Y<<`MZW^g7%ttXM7gm$
zQ8T`PKuxQFHDY8H!1&(Ii;(O!30-ZKH_?$~PM?sAdsr!?)|yiZjwYK<1};k83*q-L
zV#{Uo8ly7TVeospU<!NZxvyxPVEj_N@C)c9gc?Gyx~O&CoNN<v?En5g%g?=WZrs&m
z3uUo8zcsMF7_0t!-v6)lr2?!kB<-fX4@8qW(*Dsc6{}mZbhgqwV}ZF}zk+j^y}C5I
zoQ@6ikC1iehxJk*;<$+Qf=tcpkKRLi&FiCzxBTt?$3q8=Lc;AQORC2}J)TP|zMeUB
zi0}@ZZyffjNI{e`&R+APn8qqTHC)wI?X8G}|M6_x>L4%1NuaozOSx341YCHmeV<E{
z#>gfweeZ>&#8tV^z{To~l6jp{0yI(kF%{BUd0HZr`<2~bO#KhY_QH#I=VKass}UlJ
z4ape&q<|)()G~ryFrOZU28GciWp15X9Oy4jrmS<*gsrdo&fk7_IkmsCYOz!>el3xt
zv%6e54}=QjwM5EbHxhCl`H!~5iUp54H53rUUATh>c@j4i|9N;98N1LFLEWM`aHwia
z?f`Y?61E)Jujl9ASs3{rqwp6ySF$nSMaT)@-|^ZT-VdXGvj}!j*6NFCNZDJ>puR_;
zk)-t3Z?+ZR=B{tP?CL9gVcChO<Zv4^Pe3^+N?}N4``T46Cs`o##ZDFeY1$kb0?uX2
zE4dEhc9#1O5$ZqADN!8ZHa&#RUnz+Jnd<+svu<#Vu6i9R=m_i5yj8hZ-0D9So1F-P
z@Go$|0s8=FmC|KGFoEP(+0~3R@<DAqqr#paJF%7DzJUA=<eYU}6NVbYNDaf>@4nFs
z;1s)!m~R?`Fr~76Ub*XqmCJs|lQsE3>;m1l-`)R-G5rYiUupRYQQqufkn<+~|FCQ>
zWb%=LpZ8f_iu6R&XKc4!wQ@q}@2*l&j4I?jQ!=R3zJs`vRi_}rMqd-{_dYiIJ#%{e
zh<Sf(bZK}I)qZr-vrH|r0PG8F_?vviwT)zs^!r-kFfkaA4E~>w@IPi8z<W_RcVZ;x
zp@squa+tP8vAu_T3!@=^I;7bU$zMf0DSbuNi@g6Da0XTnjLglV7Ph(Fr<;|PVjsDG
zYkn%}*aCcfQPFCr5}^E!=3%IkGTc^Xn5#P%Fs`(@&N?`HZK@K0*^s@F2w(S8{I9)j
z6A*^{a%umkVCw^;)EHZHy8hRXW&51kZZtq(CVu1zIl;S}@0|R=%z*|yWg=D`1r19i
z6{_o)KW;d#@{kiFJ}BWZddI&C7jXW;Aem`ALWq6l!P(|RA9h_@IPrDP4eG%u+Zj)Q
z9{vA_d+&cZ*S32&N|X>ah^V7?f*p)*^d2=MN+f2GLNI!ZAcJJ<Jz7MGo)C#HYA_RJ
zB7`W>OVrW(_ndw2=lMMM`wzT7@i8LTb=Kop$6D)9eR`e$fvJ_RnhNIJDj4yf$~GaB
z%6q}6S@VEol62_Q>3>!cm_!wT9UZ#@4b!}S?m(hSk3cJ%HHDx|7KK4y{LovS(8c^F
zA=Rl3ts}HjE9iSP_KIz`BjLqqa?S>mG;ENT#AEl}aTpuASa%?u^3UE4@H2AggNx~&
z1m2xHI#0@n9ha+Y3Sd_DgGB}$rB@Bm>4gSWCP#Wp_CKLt{L+_FjvpNB0&VOQpdSE5
zOpfaaSl%{B&=1w*z!qTrp92|G5@`YG(eIwG7jP<JOP{*ad~tr_pVxgU_WVw}d$iwY
zAiI*wRmIZrlI;_EMuAEkY^+=gzoZYZ?<<$pP+>+LfL1K?VShb{5ShGbDp<aTxKS&%
zOu~VG?EgEAT-tGz+QtC7yA4wrfvg{a<}&?G-~X6;SC^xo4DoyXRJjygFuiy7;aG=K
z=FJBkuXElgB)Rq&or^r!5GsR4V4TVRy;_0nbsYxcwA@=m!)LY5(L_wv8H&#jPw}=7
zq9?zbsAfkCkmya!S0Q|Dj0NTD(LrKjl1|*AqjtUvtPoVhR8vdS1`NrB(zGj3h8d?T
zP;q=RDFE6a&7_+&8_SQ3p~-iJmvgrzB*#jRHKHfzasT%i%K~G_wJZ>|DEL1J1bDRl
zH^Y$8BO}>*G?@uQ{dS;7sGVX=yu$)(JC=KE`zzOHLU*7)@y>VQMbb30XRyp_PN2Jx
zcGAt+9&x;7TPRueH6E_zov{4_u<3KGX(DKT2`)!DORpb+Hku}!pJ!iS_0Sb7*Hrg{
z8R{vgvQN5kWB)swH!uG)o2bnHxu*U*JliSuxE|lx*t>>M*R(vBGAAR;6{~$95X(xt
z`i9marjH|?X*q$_=psnl#8%PF?x_5k${=K}?5<lqVsVG{AQg2T`~roBUf^OE{ebg|
zal>OHwB@;O3YV+ZnZSf#)M9Zs7?MlkmVVREoSfmWq835JNhEKk<y9J2TALWj8nPm}
ziKnoT2QicAibNZj0Tf8xJHyuiKMO-!J<{@nzeMX}0SOid==4enTux#Mt}8?WC6*Q|
z?FI@`d)^-k4{KkDaQNL0+?R}<w`OW>7Me3~A;0mZ_lbJsfUod;ZkX_T1`z}~5mc`q
zn0I5C+@4hCZSTvssX5~IxnXf(Dub+&vFSteUAGMem{+8vL<g@z=e4BP;hHtWT(~Ld
zJNcO6Iu-E@<sm1erRH~ZS)$qr0ddYNykh*obS}ABT+`g6@9;^ai3yJ{&dHQNzqj|E
z7muE9#T$N(rwek9+~6m;EQOW#%tMJZQU++7hoh;i@itm15>3gPjEd{*t*Nu4H%d{z
zBOBha(lyIJG(xXyv9|sNH@%%=P1!ID@1~Bj67u5^dG(pRR9%%~$SG3eX!adHWf1jl
zSpP?jH$fg2{9*7g<3WgItjoX-Uh94PcNQU|`m{5xe~+I_r6N$XTYP`9(;FG5&drDk
zHO=0$4u2?mE$&gqi!~ItQjc59(IOlvi+OKpMwORU*x!^rjc!va9^2;Lk;#11!J3!#
zWC-GJ=>N<wpB4JJ-Xy_~W2K9J9WKL-H<uq<=&r4SEmS0$Nd|ayQd5w87vV_txasTj
z4r|Gytm~J0Q>suoP1D@5uP&XWhPZhs$KRu%a0_>L`BA@`&T{-db|=HV@wkBnnk6$x
zKfq*93X_HS?bx6TiW{!h9S8mv%&$M3eK}wYHAt71EF>LAgC#|ch?HcbA$b^ujN?>Z
zD=E&&te3m0N~TDGjJ{!cH<BimMq~RXZ(X3A__rZyij9cvnbLyQoNS{e8ia&*&nD?m
z(_E~g^bj>izsQpxZ&DWrB|sJhX16rZN1#NchNX&kcR6ibZ!>DwzQLoS!QGi5<+FZ)
zY84Er<}yGB3=%}dMuGMp8sSo^EJMP>q32Lr+OdDdJE7y*9S^>IIKQO_3^@#6<q_?^
zbA@(D%o}adbu|OBOm0@pbS1rcV>q^K_LHS<NAcWut!1t|L_H-n24Wu~CV`w#<|`)T
zl{h6<8Qf)37Jpre=vMD=Z2bZbMYqWYZ_L*(GVdJYzum^kieQ_%+G6jyRS9v-dZ||a
z^gr}kSoS9)1-Qi%d(Yx6TYgJdt9oCnH|5I_=wwvXIo;`0@WJHi$@WOb!q+R3%8>v0
z+#db25VqU<VFw%Ha`Yi?G_lggbMrBdnS*<o#jbiSX`dQC+eN;iluTI_qBc{8TYJUB
z@vYYcI=`+Q9*9gDmNY=O(1^C+R<W~t!<k$S8afBlp{-ZBRFtZc)g5q<nF1J>!(x0~
zWCOY04JE&)=OK__-aYfZYs#MdVs4028<th3VD5GQ<-s>cjd4E>I2dRGm!JBm>kK!N
za|KDY2F*p1@0mf(IZpYYd@9f3+bIIiGdxCwi?b5b`d3YPREtvY(f%`Fm)^~$&7S$0
zcig=NRf<~lZQ0@Y1&U5<w4sNvPh0`ZZ;k>w>2;jif_^%p7yeAd;CQm7C&8eM0wt`X
z;?@@u=VPywWY^kxdr^(CW|ahBj{4na!b1nNPz_1zh@iR4(zoHQn|%QkOix-5yCl~@
zQ<@BZTZ4v&z0EaxcTvxD(-46I6A3z;3NlT*8{b#f{xeb&AhJb%az09dTAbR@R~3((
z4*m_qlA{GOE2c9}G$qU^oxmaq&Zl2MrW(y<z59w2miFU$>>%qp=+ox=DVzrS&I~<_
z8FZ}8<IU-9+-zxF2NgLVz%U+Qf6J3jxub{D4LTD5j3W<zS^{$kWN6_po@t2v^`~Up
z+m-esEx0K!S=95Ni@(&=Op<GG&JF%lQ8ybiQf6JnT<E(w1us1v2m0jdn2Jy14{Hi)
zaQdQSeWF#54a94cgEzc#G1e0*E6vcdybeo1Yrn-Y+w$}N%P|5!9wXTr<U`A0MRhFd
z9hKLy8N3;|pHdan)!@6CGpT`IXWZ3tXy;@e_eztn^E@ev?c+KD+S!~c#A$q-cU=$9
z%lUomC$fnqnv$|N9V3@H@Wh}2<k`a!#(<>U(Ko)d8Fue{9}a+e`0itH$KXEi#KSK)
zzf=A$IBQkd@$RMg%W;_j=VQ%@Qk2k%U;^HE%O?b5bZd-Hve!A)9P?25!?3jN7|A*e
z#E6$H-Sn7-k`<kH{+65Sc{B}Q?8c=h@aiCC0?&VbIp6(mvLw1evGIgy_QRU?N`eaJ
zLH%qugJNoOmfvhN0c;=YSG_n5YL(`Q_()f1Df&Ft2KQzcvv~l>k+(KE>Oh$k-)4+}
zcBau7wDwCZ(<Pi0rrGyF3sm0H?(X&Q#{C=!J>NNff1gr3V9%Aiwj36X8Z=XMsyYD8
zQ6D=T<I)J4O<$Gx%ae@u0mT~(si;A(iWOi3KHF_YT)PxTLXK=i^@J$=uxmF!Te^B9
zf%0)VCg3YI!+Hk7ZP?(43@U24f#w&FQXE{O=h6j%i9GP=NH28N!BS^|OZsh2KFmr$
z3^eZaHy~PvpBK_t>cBvFrg4JsCYFpox>d;~m=D{1b*{3q6Y<ec#UrIBR`tQ<h%v)`
zNtn=IEsHUeNydDGDj-O_w8g{@aNsLC;e$C$x`J29+g`n2xe@AqFtf-PW8Q5ZK!Ncc
zP(H9d!Kbxm=`s9SUs(E%U$Hm>2T~SdY$0~*<o&?j&j)d}h*Wqtr{C|ejMn1zPsy<d
zo7ntn>l9YSK<S^M9|lxTevUt?$_5q(aoUzYZHC`Siss|3BOOynNzBNK2)g?~*LlI)
zepLp!R?gQPLQ<N=<b1|ug7e2^(8d;f(bZ@g=TXQ><=FSaed?R;B_^5=_0a1~<C_Xg
zSq~ge#z*lvj;lQ07IVbyerJ(1&P=-+c1nxs4<6Sqq+*<fr5HytX)v-6?FJbi0Kdnn
zUn6$DlS&F2f%{O+#T1&1H|3s!b{k*))FZ0-QWoy8QU1Hr#_sZ@71A01-Ssso(DmyC
zr;vnoLC{>sv-9Uv5<Mj5G&PR#d0U)yM=1S^e(Ju`YOg(LXMgyg2u9=gh%~!f1K6Q4
zk*2)&o@v8ZEJEzgUh871^Y*7hFHX7V)c83?4g-ZMXZ4_pL2&w%Y1oERgzJD9-07Zv
zpq0|9ScBEiQWR9QfJDy#s;Nqy?=nDb$Q5!pGm*nY%W?cEZurVSA35de=GIMzoGd6?
zzSaJ_k3_4h1i|r%WmwCeU0oJ^l`Zv8{;uGe0dWlp$8DB4&{wjlB67?yM;xTh;VKC%
z=}q!Bp=WzLZNVR%chOv<{w1(nE8$hizUhjOf&UJ}yPx9T_iQoRv)>-pd5*s3=(-*5
zMw|Reb45t}s-c#w|APW0|3g{6T4D#}#oZoixI5KQd7@32WX-N1huIP<^~XrS_I1{=
z%%G=TTgEUN)jFRxwFKYL70(l^=_FBvngKZzsB*oPiAEYg%b5<P1Gm5Qdr>{-%kM3z
z35fjxb^_Jv6tL|`juyxj!%HeRsld=bK-BSST^-bNICmZkd%rurp3bq1UZPlM%~wET
z)B9L-cJ$H5PsVu>oK63UHZEB_7HJe|lH;}fTiucFl+`cL>uBU%#nyqTjkKsTQb_`k
za13P$y*-(ui_p08%kRpgM%k>-kklLg@#?aFNj)(_t9-M0CRv4gcU&P#Voq<QD)od?
z<y(L3r(`D8U9d8T3X}sA3!o+ETCP0V98ZKrLzeZx1NELix@uSm>-^(oyyC<Bv9SD5
zT^91^qjOYI%fQ;6K#fUFE$R`IjmK_<d$4G$e7@V2Wh&)TWpX5Wz|hOLOjdUwRi^h|
z@#eu+9hcuD^4OXwrk-|OUH>LGjU8ah^AY6;NxW75u_<|@uv!0*WA1ZSAYo^_l;F~Q
z-cWb^pe3zeotdy-&WYMI<gfeg`1JQ>!AsS>zmoVQ5dofm*0vgEV>tH$b^V@t#N({)
zK1q@5eB>I7DyU;{uexk%AB_^Mi6FX${<)RGwg2ehJ&{9M{gpKrVChkLT(IBNedEzQ
z?9Ombp;zaDWMV&!gH?f8YeWMDe2KfT;`0=g>+#I=)E}Ea(A}*KiUxJ`M`xAjy!}_X
zqmYLEZ4g73kYuc|D11R?dyn)EIJaK2&u_RRx`im7C^hGwkE3xN0PYq~RLWm0)vj`K
z8y+>Sti_)fyiM(kRo#2Z(gyw-W5-M03VE|2<*yjon^gF+#MD!R1Kh^s<w->;Uy6KI
z7Z#Ui4^f5h!la{SE6!M*vl2}9aE6Khm>5u*`dcC^ar7EXG1+sU{L{B@KH#*e6Qz01
z;d}MUyaQD{jlMd|GEo@a2lA58i*x^brmM6yp94#n-ZfbUB(McM;KDh8{=PwGxm9H`
zZ=phpW}<oJ$&;g!D>~2n<&W;H!;kT1FQL1Ly#?{fA*x`G^`mNx1SWjATKO|yHisob
zq6__20v}YQ?DY@@-;^@=TuYqtS`<<B@a2_oQx?|&EOkvQU1EeiQ>-lDp<gLrOW{gT
zX`&-jMcqOzn>L=}F+0>By5%E=>bk#nRGr#A-SLqOy9~^|NY;t3rgz)jdg{%|kgD|*
zCrA<mC7*Vs8}SuUsjyK?p?M#u?vQ2C>-so^OizQoi}&g@(I~OSxmf`%r3{zY>y!}v
z5`7ZFi?*C6AFIbb9H8D}N)-)>dHWIU^uEet0p+JDJ;s3{XM3YwfmVki%?6>m&N^_n
zwm)i)gK6V6$HoQJclDjg5#^<d_Sb0CozIVKS8q2g+K8Ig7jmUIm^FAU%)`fPkTOiM
z4beUJqAq&Y;`LsI%3dMQxu_s2v1vFXSARLZdu{4;-_#n_NUQ|&wC!UiE@P?R4HNy|
zBrj>Hd3RYC9TG1rM+r@^Zwd*zzG(%<de)mqc(SZLKn@jEKcbd_fm*F<xux?X&QcoJ
zYKg1fim0<l8>mi6$ndB}5*Gg!bW5n?UfS#3ua;2J!kTdDfHR8~^$_v;6M0YA!^<se
zrRMhf&D;??bJs~_3VP#w<87YxiA*LRCpx&M3K_HZ|M`54%b4NL7M-Tdpu{1W5qVzh
zh@8(vkj@`PlFAlFv+H`|Y6$BigPKuMAXZa}<fXQL_IlR_EG|^x8biIo`!u&guI)h?
zx7ufB<maoCLoYx7vH8(5<WR9{8eqziJlTk`kD2`~sUBymsMy3QRE4r$=&#*{ZNVH`
zj00BIwh&r6mH}JdhBHm}3sZ^1#6r#`#UcqU`pu43jaorAwZyxFoE!mis}<d{wka<r
zN@sz5`F|Hl{ThXnPE)7>2kG{5e0`r0A$i0lkm=g1;>&PxsEx^>OkME&DaHjXn-bZ6
z*^c|+L=HMk#mv9=-d2~X++QkH)>bLUPE`~_$HvqZFd@`6Ytp=hoQ03Ntkb0*JEQyH
zF(|$o(7a6mxw+{rD2L7FNdBNN6q-U~(AfD^YpaThgE*y;KbU(NcE-bV#-qY%UHb%E
z`&`5>dDc71t^wa*AwQ9l+KHwwogk*7dG10se))#<-`U1$JaW^N7T#Ap1!=OcSgW6d
zWx47t6s+frOFBpkWCabHd<vLNQzz57yc8~NAy2QFAjg<K)xeIocsO(=Z5Je(lV@!3
zbK_&`{>_|CDo$yyQQs3oUDH$)6)pNjlxGA&tGFVLd+*ggUIJSCO7PgX1B5XSh=bWP
z%~yR}>EPD@GvCWHd*>{F!aGH%Qf`#EfKd~At$=ogc6(-FaqgZ(hyV6WbKt!;5Vxu|
z{$`WCYDXJD%k?7m%SqXeso+cp*wQSLIC7rgdzAroM_ahUwZqXX1zeRPjT07<;Q5-S
z>H?H~ef*krxugvwr#8-R8ho9aJn^VtGPU}_g=JW&_?UJqvbT6fcxE$eT=?g-<8T+A
z>XWVXN0*bCCX=DN0qUivPK!1S2R6@9zqO#hW88*}EZI_8FRtoe*j!rJJ%*8<j+b`e
z>}KcgrB;$qsr)m-(_77c?>h1r2NNmYh*TC><r$knHHMFPaA>)-{P%#g4p!AtYjb3>
z!P9~zoESu6#^9izCdIB{C!nKZIK|hTz20D6poDOn2kREdRVmg#a!ohsFmOgomhJK<
zu<e-zBsP+Alx~WBf5+q(e25Q}+w+*R+lwos74;HVN2K|2^t{T`G5&ksyYgE7$?iiU
zM8*dy+|}nBer|zs5%LI*6uPgi7I?@1ec(ePFu0<zCyOM5p+Mlb{=S2z2Bcgd^gTa3
z^;ouc2>vTnH2Nv0epxE?YM%j|9{ZsC)dvXd#ZLkRrw9U&_21#Np9pp<J~%zW1$V=3
zaWfTB^+N^a--InM8r-wC$Lkww+PaW>cMUWhB}k{b$!OTqRq6ib+N-=ap{O_**B8Uu
z?D+N7)>t@5&PE6^#l(w}I3Of;_c|+3_Tp>iJhhM0uVUu5yNyh_+c20C;{N3skq=NF
zj=w>3K5PX!>KaHfKJE;ts)h<$=cbtksE8K;XlDLbKJeXe5Ak6KMyd6*O7rosHK6wT
zmMf#KW3LMSiPC)*JJgV2GpG0PZJ#Osr7@WSr6kqq+oDj?W%RA+mVW4;@^Y$i(KAA-
zoOYu2Lo{c!stK=OnFgNJr!tP5{i)i>T!wRwE5YT-RsG$=%cY_e4FMoM^|G7IP*<LC
zZ#}EFKg^0g){kDjZjCZ*EaKd3y*MAKa%i218zF9@@oyoR{J?3ya`@VYgq+UMhQ24<
z-iu3y7F_lDU$^JJF5Nn-3O&P|AAVI`JT_=zG?s^`wz_^_w)eO;4LgvWB*_*3&+?0>
zzoAHC+(Q3VEv(-W7#!CAeQR6@V-kXA`~@BtkN%q#wY^oUA0GSS5r`vINqD1<BpTQ5
z6F|+8F`!U#yZfZp(kJ{GzAb3iwk1qEsQBeDanKPxDj%J8dUN#0kFmFLDw*W@la0Qg
zbCd&-b~{^wv7t{Lt5O1lvTs{S*4-ZP`<?66uRw4ITMDRKDSADQeEkIZ%e^(2^OefM
zgxArDhy4Y^TJ5vyHG6bj+>O8bDjQZsegW^=^<vdCSGVs3A>haYjpACo-WLE`KVcLw
zOy?KJ#W#c5EKt_CL2$~j9RPYi|C5o=EeL`@%FE7N9z~+C1plWdGODBe=+>0q-02#G
zX3ttO`ytV8aGdIS-Ps)1#rxc|HRt-+vt=FCYLIJ;5M;A^{2X}d7o<P}It_dhBA>i(
zBdZT_tNC{8(`L)bIDwLDkY)31brjIbU94?ABF_jq_<;vGtQR|~7fHQ}&{SiWvVah7
zEN;6!;@kshvOY<MY`=u3#b4Njp?cC^q$Mk^fBcT!KV5eZ4MYYCPy;&<2}vAS7e0z-
zyWiJAA;&)!7Fgtlv)*6-H3}NRH?vzE=OmFbvy%={f23-Gl+~#Dy_b1<OptgfT2G~_
z?Eie=(<fX!kG=y$HeNdUa^FJkE%us=Ka(zNO81|`!p`TDOg@2H&lD#}ehkGR>EQbo
z?-d;XYlu-y*`YMXUKDBp=PxygI=#6=mKnwj)w~|BA*=K1dV;t1^~k*yZ=ZKrxQsrN
zyYq$gHSA?oEdc_0*izUCE4P$d@bVjUwfz}haP}nN4wdPcWe7c6y`9&$zx=sdI~fe-
zfdwa&I~Ds%jf~&{NnRLVE-jt`j#^`~EouQ)tV7Ut*C((mo%8*HbtE>Q3k92;<L`+J
zJ)Jv!>=@Y&V1isj3rJ?XXpZ8l<RhYwzkphkn|33EDF2&A`_ScMCo*%Wrlj1I-vdt2
ztO`xbmE#?BadxnjQ2^w4-$w2{xC&UGe~{fJmraewr5fsvSIc%{c#wZ(M4wD7smQSn
zX$IXzXpO#i?2uNsbE{u?3{(KM1UcScTCzjrP)!qpYn}a7SoQ<~KT`zC(8rRobgOfO
z60JjNRdt`p(UBon@Ac)QrQCs<j91a!u^hj6hIU)S`QHELjzZ?y<*7fCiiTf3O*D|`
zb<gp8Fo1(jE!JFnZX8L@6)wVgzWVO3o`aPiSA@zU@HG}_BBh4=89(bh=Bm|RTvxhz
zCvS&C!E^pGM+MgeFGvs%^of+#?v#Gxekpe#W;eesQsX5Ssqy)WLy8sQV(6psOaJpk
zg;*f~lp8&R=lSW~Qy}@vBdy<bg6u#-hQ&F+<G9o4^vO;?x|8K&?Qz{EU`3vG=kULz
z-U<%N)dt~)(b;{nbpR&!S?_jG>*{q<V)BwKxt3GIL=CnF#hqeUyhW>;s@FkvqJBd<
zY8@zPhGD2JJVtV9jBi}5Oy19D>+!y3lF!Tjjh2%QrFiE*WKZsvCEi)B0|2(ObQMcS
z?WMlnhcw;dgd%2$cAiO2+rq1&%|CR=F@HzwV|Ig9NbyVYPj{wNIz4ZP9pC+-)7Shq
z+!Lg4Jo5<+x?qJg<PH^_Ml8g{`m5}i#V-8XB@$~cta2(!g5O^)+V9R6>{<}<68@hM
z?Qa{7LM%n#=Q|SZu$cZXxsWJ98NJ5!Pp>CY-;HGE^qrP$9trJn+vS;a;;Bv3>;qO#
zk;&v-*?uLeT|a6G5fGJ{?;48=foW&Yik!G9Ge<O35PKY8Tl52X2&cBgxbP$j_j4yW
zd?ymfH$OU#OSos$PGkO1$sCve#Zh`uhU=O>-~X(%HGi4B=8M>0$h=OqR3a7(!0ir&
zuj+}fm?HKYl)#a=1IAl+XlC#YY&~%L#9qK`H!cRPMt7%(nlh3*?@vF?%HSU&#Q`Ve
z2zO=(Yk9jV%h{J~(VDKNtnh16ud$bM<+C|(5J}JazF4Umpk@~lU(uhDf~k$g^}IAp
z#mKzH>XQ*%kt57ABhs4r4`M0sMyv%e$+gC7gw$L0Sk(%Y{WD(OBV)|wO;>pP+{bX7
zVOGywI*sAZ)Bl9$FAPbF%^fK@FbJS?$xgd@^YsdwDoF_0NmHA%e{x(^88i;u^;Okr
zJa$rzoeYTtH%CcVjjGkoA2Y`&P}oUM31hHcAyTKM6X3+tp~#G}kGHIt@_X-^ulw(9
z>*~0A`&gu~%{;-|JgP_`H(aYguB-ubz;p+n_ZA5Y+t`RhBG%*XjX%5u|7J6{)i-O)
zmxHBDHPG1idz;hKVD<`?0y|%^JB2jwjbP3o8xht)g>8i5j21d-6#_^KnN9bSRMKxd
z(+!-<?R%-`xqPMA-bOdoj56kT86@=6{9rM;LEpFOEL#YpH(XO{{E=Fp{|GtDV(&N8
zG@ZoCKa27)4MFL^4d{nX|4|3K=%;l$dXqQg_j}YaXvXk=+|9^>6Af9gckS@VZI05b
zpWnqnYUV;BpQF_MkU{<>oiXyqTHVoE#n0Y*5>2XXcQV#j9{bs*v`pM2XIA>>sFy`v
zEtRUt3Tykm(YTp&;Is89qmZVB@tz)~f!_6Z=O@z_{cZt+!W<coh){{RH+_=9+yPrP
z#VO7D%PGw~Vgt9Y<4lWVC;!wib1Ul>7@){|o$d|5V&MF#TX|gj6OX~*WVjr+m30>X
z7`vT%m)33nzUNbtlb7#&l@9&Q2fpN{ajZoXY5_YK-qc$qoIoiDRTO$F*(#~%(=2p9
zl)-&%<pRKfc6YshSng2KMlpCNrEwR`{CZyiw2+qG6vEL(#|FrWtrK7w6HlHO;=|$C
z%X!j%eb?N?|CYDUM*<~gg|Xzp6|@Rusnm<Y$5?Mcjr+Kt&8^O{Xi8DBKyig!rt<|J
zCJVyLfdOsV?FZdyDV*i!zY&Uo@8L!yLyujx&5_Z^bQdQm&3JbQmA-JAdJWk!E1@l(
z@LtOP{d*-h(RYSIT<5Uwc1~xy1ThJ4Ptj_<%yav61g)6gKhVZ-&yO#V%r%Aksx0Ib
z6QInH;j&5;`Baw(!@ONg()P{wTJ4CsF1qEknpceWL5X1CUH`_fB(Z1O_M2D!%wejS
ze`6#$Oqcu`m0Sk%PM<wD)riuZP$=aD{yem;CI!hhgKu`NR-XlYy%eoZbW))({4z{M
z@FspXe8EV}$jK)?*nU{i!8Zjdo;RG`C*jB{NaDNO)g~&oG~A1;yG@%h>GVE^rleKB
z*h&p<87gIvrX(m&t*dFiN8L`^=A@m}R+j3+Kpq+;fZ62Eq&|HV=2E)0-1dhqzaB`?
zuo|*VG@(D_BGR<kuaGJ@>&Z1BE(Ag%sxIq>;%Eo1?p_|6m*S>3I)WtvCA*>L3u3{&
zS#5UvI<M6#!A$#$s10G?g13L}B47ouO8%j1cNj2cbCSU;1{p2{^?3hD1Y82MdV|Jd
zj-NyH5e0d)pmLSyB<5Gbm*=439p=*^_ju|#-U=?0DOMKlhhNmYn!{>DcGi?}U)y|Z
zW!UDcnm?wWA<Rh{LnMr|=S604n_hW)Xzz|sY6fe_!8;226+*mL=vXgUlervQ7){}x
z4GQHQL8cVpBFL;=HlYd|9Apcj4c<IH{YSyK)vMC&Shm0?cV<Z)(LuBJG>NG7wfO!5
zl1h8>thL@$qSSuRhq8iirTy;H_31}?kH7`F0y@MfwwX*^xj}kriLnps`}V~GoDtk~
zyr}?9Xgyt!bZ*yU6S#Uxm)ei>t(Hth>>T>~<SC+}IG>@GqXi6J6&6dg^CkHV$Yind
zAL||89#@&6{5}S;v!IleIK#o~cX*<|y9VwqCs+r*H%U^VS6^a(N9u9EM9$IsS*F^J
zQ)c^NT9l@)q-DNx=11q|d98PW@p|^`A7;aEiF0tj68ERso~7U@YsR$xMJXmvShaF4
zohY)^>kyqat<@i8);UDmfwJ4^ApGFcw`(U0q^G;T%W7l6B9LkSr5Z&~F7hFiDtp@}
zu0K7D^c=@%T_slt33NuU=IlN*rB%4|tT0BYj@vK@7ZG;Jdj}jWA@=zluPsxPKO}!T
zz3NoO^lk#j8_tjFVlXi#9%gfm2ouC)g&biV#(}wE201l;Ct!FJH#W1Uolz)Szx}}@
z6@sUv>|r1=>S>soH8+9B1nqp28yL<oq4fgcMOn3{Rva+4rUXh1d8801XyFWHAuCPE
z9VWy>!(Xo5H#G>91s50%By%@3$3_2I?2=^QykCe{I#*2_R^WqK?`Us7O`p)@J1j+j
zli^P5#YwAwXoGRKmQ>BE3qgB6z^pRUUGt(jvi!;s;nUOdw(K-7ESAwkt@Xymy^ZDL
z)bl2qn=(FOYM#$n0`Djf&1Yy_9Mo0?i*u=wdsqimmYe3$<|)I*;o`-0UWxy%=CE#)
zCEemE);26GTh2`Bo$MvcrQVVh?6dc4v7f_NZTV4Gq(`q<AW@KDwZxyFyUwpqiPW^d
zd~i|d-uXwLqjbWl5=3D3et9>&dOfvvzm_E1vTpH;P1*yA`_AVEZi6Ph6nlQ$g9<fo
z?PDj?fphg$ZW7UN$$4!16gy93Q*VAKcuiC?8!-7Jh9<b4zV<<VjbZA-SN6lXG}gHR
zPpCcLO(6t#nQYT=42^Kv=v8a+TG7_GdK^wg>0|U1g~n@Iy66mfqzs%=<Hl}oHitdB
zR4yHonFYNYqGLYefAeb}&8J$aq+Ua*vN)yQa!}2}<&r9IdH$`O3H7!LiTCZk^(uI`
ze8>y^LpNT~YUC+?m*7cv+RHk-qH=y9vN=hA+ma?CB!P)z`fS73`6N~VdXMKh3lE)9
z=pDZHm%KenPcjibnL}cm5R3ZDDga~$DGiD`z$=dR&F$RK^P?kBNw-LSu`LjDb_MI^
z8N*S-M{VmL@4B_Bc-fn?_L74e)N!9fitbs#Th;v@YF25!)@B{+rajnFfRtVkbgIJY
zH%8o%yYH^aJ|AS6$x(rk<dRBF$CP-i(I^fcLZVl(Y`Rx;jSPl;2SU$9|L(t8d6B~<
zZ_^c$!#?jU$A`nC-iD|(z#bAMn|l{s+ihcbLQ$XkRXVA+T;SwN#*mYP;;*anD6IfK
zw~A$2Q>#A13$pDuXFTk)T{!r$rj@@GBQ+5Z`v6@kFKqq<B$T;-4c|UCNEMX%)&G)j
zTpolr?>4r|x1dV1QM|ykcg+`Z%LrP~ukmF_G;qv+%G5gJ&D-pqrrYwk%E;q)0q;J&
zlql307cp=iL-)cd>)vpox@5orJOBO3g&WIr@ta(==SfzY*KS$oP;lRQbux(O#5v4-
z-|65WG8rG8fw}((=v?t(fJwSv#fOIYxvjKMLF>GX6V^#uuT4HN;N?cVnKM}5fJ_qR
z(<bL~qmffP2$e<7mQ9Ja-6~3!vu^60HK!Ega(ofn_RIv9od}(@xrEr`;CEm17rnxY
zE{92bJg5>iX!T1UT9G!T(@SgN!||?+T(NMgqQWlgHr&4_GV@JH<`33wbw>dAdNmKg
zfH-Sm>Kx9Xtu`gzPHCwrHJQ5mPrp2SsjdX=q;}MBu=K8G&Z{<a8|Se8@fQ}`tuA}C
z_Wd%8ZM?U<=X{UhvwctrotT1G>rU%I_1{|Uk{7_pvX$7l=cy(9$KE(o4qT-vnu(;3
zNK{v@aWBt%BT>D}n5dZ@aZg%NBZ#bgcmPCr>be30)GJR3CO@L`*PFbqh=|gqN6sP0
zpJakw4s+cGylZp<AtVtf)O#cGVV&LLmRx6YNzozRz)}V8*B#GX%7RR;`>y8e#vQ9g
zOkaB5rT3A}$Vyy_ndjANLqwpSerqfR>WvCDPp=3~PR(-snD@9$&C*9_`yf*m|Ec#^
z!orh(+H|pADc|}XFWWMMt%mQjey==2O0@osvR^f9p#C>J3HsV&nN8&sg=TfnP2awk
z*LPQlj2OirMv1VBCW`z$CkpCmsA8H#n%cXc-Sz<=LwQp);&Hwr8S!$gXhe2f@>EO!
zM0_O?U_AUBy0lTgU$J^6!#c=YtWmv_qpEg_ib4-kr0&cvkQt<3zXKeSe2$*T)aZz1
z6slD7C8!g7(@&<*2CuXLRKGOjs3AgyE{fkB57$Q1xPGO0@>*GE7^cfWBJM!`NZptJ
z&QZIWPwD_9ek+pM#%1yIZaVAzBS;Mw9XHwg@_5I~6a#FjBlNV22J8xFSq-+L4F%_F
z*CZR3^E03kPiq9TSvasZxEzQP)~R|RB{*XGuv-qf0V(+Vdka05y{OxeeXqI(_Jk#j
zTm9M)Nq`gPvP|Z364T9(2;?|qspZ>!E5${`HsH8)WK0iPTt5vC)w`4XqJUQe@#a;z
zPneQf)!+;;(n<d1HISq1{K?tAb0`N=IBciOKXCn6`&f6LrS(v&@^e8K6&<QLFd3S(
zK|;m(A26*a?1nt?nA4aT$t5TS1!FcT=3&O<{FCHaqAqitnX`wNlQRZd-#?BR;7oq5
z*p646!r$~`D3I~qsx44Wu`_MfJniTnwMHtXQ_j3Du-S(fnLiTp{=yWz{oDD1o`j#w
zihBNNmk9IabVjVy7ZEh3C{qw#U0!#S`#v&!j5bQ)<mGkwZQ|H84>8DtqHuL)$3hvs
ziZFA9D!^UIII|NTRz-DVZiR)Zgj??RQPCS4Eg%?9Zas2zFCyPV!*3mb5;=p=38r%8
zbM@<Qg#?ToF7*?9f4&!}j1s%}F4F^b^Sxu`3+mJmzl=7qo8)^b>&)s7N@ZgKUxZJF
z;YakKW}fbCqUZdquV*KPg{kRh1rE}HFbsDaMoaJXzGT=X8=Eg=FPoPai#k_-hs)Tw
zE9zbQ7D50(gxKmo^U9)bD1{5^jj>WqRqU2AnIeIwxh4OqW^-)Q6kqN0>Lz<~@%l2i
z)IItV%Vg-*erZ4n^bSAbv_CerZ*?-ET2(VBurkqx1W{S)ZeJin<W+LFYSiZWs*$DL
z4$m1@Z<N9@#SK(hv=$4v90p1?x9?I}Pu0kLNpp5XMcp-UB;~Sy=SlTX^h7~Q9Hz!i
zNBH6&1J1pFzXxB^i$QSp|M$lV_W=@l>&f^>vGmuOnbq*E5dNp`Zf^Y{DjWs;Jq{}=
z+f9uHnkl)nRO!Mm+n@4RsZhJc2-dV)6s}L!_XDQA-{;hJUHi~8Z#Ma-c^m-q^O;Z3
zvB^~lemqg<p%8}{>qC{Dl5%yQ1S(obR^ShOBvoS9p3ho#T_p_6S3<E6j1HC_DEALx
z_L}$Vj5DOWKvvD)xRevP*3IUQ#S>t24VoJ_Zah9c+};QLsGn4lP4}vr_icR2co?EK
z!tRX*pPey%Zn~1rf7m-+EZW$U-k!>5vk@bDI}?B4sP7B>CFA5=aeRR%>z`|42Uh^l
zDJPR-Vm*Th0Z%L4d)h`Dau2|)b?X2^x)rdy%w8nM=BA%ihK)KXW9X9nR&6jplDgZ}
zinV3>OwCA`v{*vQL6}8e*+KLr*;}{z*-r_eG6+#xf0mY2J|sApu6yO4WZ4C-*UIz~
zOFOocvfaENaki6t@ewd<ha_4KXAMh(u4}rEjE*h^F6YGan-~Tw)khy*i4j#bsk_D^
zvmM-^|GF&CuP((uNpZz1W%5LCez6OLn0Sm2F!vp!6wJCuz@gUxbX)Q2>sbmp{&)~Z
zI|f0sOS28_dg$sPQgvw?(D>WlCPx@UrG+<zewJ<wvuRMES(-r5{@*G1c;)J0eGK>0
zE}9{wxmPle3I5_2n-?b+)oQX@)O1LB_qrLfUoLO|E;mS%`1IXaaSSlxUVpzfG-P=Y
zpAnc6p=pL-kxD!7tD%+LJ`kJa7m>5pO^q{8DVktv6?rwGy91mSKfjHI_BHsdKm1c(
z6HBme9m{@te$G(1J#20=MxYjkl2BZ<k#jvJ1WZ^dOOK}IbIXmss#&TjqnRe-Af<^E
zUoKiXd+wgeV&D?0FG|>098{ow4(4J|ME3UqNhoCmS$=(=NZHSe(}4@8eE)BMjbf?e
zKobZypD{$*SX{{N+qKs5hh+#nyomp{5;i^oca<Kl?qn?=DXc}R?NbPE@K?p2+j}y4
z;2ne6>qT28f|MldCIkLd_}@y*nArs7Dj(28Vt^V``$|;CDHJd$LF<X)1^H!w1y2v0
ztEk%doql3wR-MFrh0VvRF(6WZo3sGITr9HGa^M=XTqqace?NC0Q7dWtmk@(i6f=+i
z4@>KB8qD08z})CFf10hdWQASu-VYfChHbhPA+3ncY4aJ#U-8!_bu_iN`F?`3C5E<<
zTeYxot58S`xY;yL&=vBksx}JVFOml)0x!=9sWa5ZZzfX4W7n15SR*KP8vu1}XyLBn
zABjMpi$46gAka7wWxepiPcGvwV;67J?DCEOyT)CZa!3!=Roh=Jf^w6>FC09Im^4y~
zCzD{8BXB$3Y(j}rxHnSM?TX;`Iijf)6))Y=PC3P;iXX6?vE<}p4vkD`+yHl6h;4I*
zXaPXPcruG~0v^UX6j^qsu|nc9`T3ggVxe49#gq28q53aRxbekGDy{f5JAd;m5xE7E
zko14%*8^aJgn_Hjm{)6m=0N%k<}6>=&G?Oogv=?_xM1Phc#?_A7?@?2ll{-F*=eKa
zprOJ)j){`>kp(>I;PU4IK<qDkwTX$8rW&_omTlm{r0*|KYH<Z{+nY0BvN5nd07z<0
zN@@HBfCl&Yya|mjkxllk^!E1l>0+{;Kgl6v4)1x4{euBXbPa$-afIplr-C=;oSTi&
zY0wWB-99SsA)!@@KgcBjg!^F0{5s!`3;jtjCanrW&KJFQ>kAa}elF&bLf#9MBkO>o
zShEg_bN=v~(BnSqmiw>44lv~K(CVQgCrfbX5*z^KC!H2mPmtR#nTG_|BbWtDa<Y2Q
z0W&VeIZyZx0rAap$El?i9D&vX{S!LE!6Fw8)zq1&K_+sQn=%;pu1(!BU=4oUG@U#h
z6BvJ<ryh=Xt!ET8#yY0xgg7sZO&c8ie@+Cl;~R&8GVKgMMe_C1X;n`LL3zmm%a@MV
zIX?)!7(Ji9I0^ONkLCZIt{?<FSZVrl54H{*qn81JI9_i6EZ+Oz?6aO?>=P~{<ZfOz
z)5jy9364ekHMwVEj6l(H&?<8v98+){1UntE4_wb;!Oig5stif{ZQWD+{fy)>*t{RL
znBko*z!Y=gbRF2t%IuKp%|+Zny3#7|n*O@~Qv7iwlU2`&NbB*Bs_~1n>5GLwo)~dw
z)w30o9V}nL-+`dHocH3ZY*HTWqe8|N-G=Sws;-{yU7?3F=p7=p<nZA4^Gpn4?Ju61
zOr|l0LQ3hLpeXClrVd)tdc}{rUrA)xMRMZ<x9-NQ={YS8|561L3Aa9_DWDwE%1sUi
zm5+W*QR{wf)fh|5R%Dh6IgBIpbbEjKZCrnK?lp)}M(|mZ@;}=h7(D!YO!c74)vcsn
zJ_c5@%9+i92mSLDJhD%Je_?+yR!U$$r{1|wc{0n=njW3huBO`pTyn>Ox87b*7Y+QZ
zgX=ur`4DBYLDUO~rixEM@fIss9kQf2-Mh_=q7wgTDj;PGxmibubU&T#|8h3FQetMK
zVx&7w#`ZYAm#q^GL#5xg_2^>qI;N9|Vw`^ob~0W5Qf{cqK;P}f&5(n}qjNfYO8+;n
z`q(18UqAWO>%?qUsHB@N&QE=S*GxKGesQWca+4SAq`NskEP|p0B3SZ!AKyGt@?I?$
ze43t-?KE=e+j?SmZr7vgubno*#c_YJFL+3<GZpTH^zb%uAfq>$jUNEv|1CzJ%6W1-
zd}E$TANiX_`}l!xw|-$?P_rg}w^GqFKic0j*(ZbtC~A<rgUjj*dmDEvp9CE-QWluJ
z^%NjwGom9jKZxyGw;o#xBt?Y%&O+hz{8{-i!tl!IjKU)9sq2c^!Tlp~j0q%wPPgY8
zn)2qK9{gPj9+}bt^4kF2ySmOcefR>H)!)x}Ov83dOj4X*nVzXK$R0s(&p(Q!ppS#&
zK*f2n>3qc`wC_mZiHbry-Yg}f5LBcBE1v-VVMatk8o^|VYw9h3C76z^b--EurxVd>
z)Z^+-iv{zXbPh!CN*Fn(yw{4DKe&4;szdbExp?LZ+88Y8jrz_o4(HMr5_p|B+oPNn
zKm>R|U?A+Q-6XwHt-~tBd$d?cghb|#nNv|BT7?i!v?chp@4=wbzaGUK+oQ=yTn{1w
z=fwJ{fIkb<&vtdKT{i)pg5c_j(4e}hHHv+lSc%&`NKp8{*N&8q)L<G#`q<lEGfUk}
z<SYPs=Px<>T+Jsd3DZ=+7QNJQSeVU_no(`L4+!_YHL-M`a!=<4Gi(V(VE-cX?K}N_
zA!_?~HEuw3=*)ZAri*t;S;Jf^Ye#GpRI$}Je|-XU_JzVmu_JaFBo-^5PFYpA_9Q|L
zcFg_5MD+~SPjzI;o*Ju>^HZ@;?q)``ldE{dA)sVu!L{2x5k=ekzuX68zQjAUYoNB%
z>`|eTb!nPOtOJQIJajhr?-h4HE0smXQeZZ<j_PP<q-6f|;^3*X`EP+ygwlYGM&g+x
zY=R`;AMA4Hd`kaFzNEC^{V~CTCh0&!cAVgoC~3{qF$>({!TnGP*MuTTQRLV^oAblJ
zcZ^pA*?Gv9xcXJ8Jdym8f!9$kEnqX%#4}a${2fD8azyaDJTR-?O$<(=v!Xe6EJ}JG
z!K;J`fCN1F8vAV?3A}o-GDtekjfxXpP1#<_&e#n3s9|tpCa3*M%)z%|%lG|w(&4MU
zEn1aOF&&NTc<qon+7tG%;vQx0?9A<yr(EqY*^GMeXo2%S1L3;Bd@xN&CrBpHMq!z~
zW+fXU=5O+XK=xpfzu#VU{_>fp5JRBq5It#P^E>ukl9K(EV^*W~MDnP4z#oy@@zp5c
znvlmvXISW})HF94a*Lm)NX%~jCzg%;H<o<_^r>3c=;)1*U$>n7g)nd}zPgyGp8gbf
z)cNU>>VQmAJWCX}Oh!(SP`NJkd@QRDQecdwPeq4<t<!3T9yz8dOZgcvOz*UYoJo6>
z^C;>&%<g<iImLWCftf+>V=bg#>MH!47M@-0c*OjHFSGEHyYu2MSgK{VP|>ihYMo@b
zV-JT@?wC<~GzEk4Q8)F_(k^`c1EM>vBY3ci_A956iR@9ZLT%c2mu>3?Y2M{|z@)*p
z(qg>EtYbC`b-K0>DE|p+HU7;j{;~m0bm-vkcry97MWY2Gwxb1e-rF|^EM(|W+`z~{
zR65YKxmrKby@qVosULYj`TmiB4mQv83mwMt?2P+1)H}8LEuqzl$`)B`;_s(SZNPs^
zm(M9CsCt4qzo+?VB06HsGBvVVHK{3Ha9>2DG^k+$tsD<d)*>3Ahm<1ZxN%M+$`gGP
zf=6mAeV`y5k6EkegULI%eJK|p-<KlMAE&zjM~PeN(*+zHDRT5p_(4WdWmU`Z!<-6j
zE;<Kes*!5jQ+j;imM$TWgaT_>GLSmms~oZdeQ0HEcrzoy@3>rWVIDx-Fk&@^u_R+o
zCJ^Ekbtza69TIf4bL7D$s;|+}EX;uA$a|l>#F16>9LtHns!Nef8-5QR5gpab#e*#u
zLQDc`2)<1NBhmO+paN^F!R#HX*`RJLKNA60VaZ29zKI1!`qienRPTO#xKiHN#sH@w
znmwOKvs&%f9($$`%PU1?IQ4Z;8APdej-z)p8JlH9|D5%r2dLbz@A0@x`Fc6>hfxiH
zA6qqS^4~9WQ!qaoFwgZhCAjt1=H^gl34iqn7>Gk;DSRk57*4q#_0iTuV405t)llg}
ze0CJ$pm>&c;C$SqtC^@7z+DnloCG(;{YF_CoBVLrpi_C&PNcSmj>wzQwy)z79QF)d
zCrPR7;pO`q-j?Wce46YNB~bGTD<9*M?`48ZU)Q3Y4{Od3n$j%b46T%=`g!z1pm+$M
zZQA$cv-(k1OnW{{t4<QIaib$ml{!2*tF)!ph1N{WM=Alh@Eaua*?aqfi+^q^0&|9N
zQ`x@d&!<$N<5o@s2H}SJE3_nd(oV*gp#KT~rZ6APJT(tMzkV$s-ms^WK*Htq-eZiA
zy#$mbEjp(BDbo<2DxGXgeQ%CwF1ci%sE0zAm4PKBaX!J_v-*@V@hi1RpeDqN;Gd6Q
zd$S9JA9j*2pq~kwM5%gPY1TBU&hA9ItUT6}-DY%a+SdLBC`X>0>D~wv!2_!sDo=<M
z=U?+zW?4}AFR2OcJBf(7y0r4wBn1T~4PgE9g=;p#*))^VXn8W(r!;eFCq`fg5VwD(
zi$Lf6&)Ng#y)!q(l6E!Cr$1dI3e)9&9wk6NL{|({O&Ma<6G@;GD@@ag?o{bRf)KmL
z^;R85PcnLD1#=sQTr$BT{5KN{CYiW-cMwDOYtrQRy~>*PkZezqkO{%j7Em{rVEZI7
zfd>G3?dy1!EbyG8CyupS;hPIykAA<xGn9sur)ktY<JOn{CB$LBE5Yw~$mvCWjLSpq
zBo<;iqr&2+=bSiInGq~;lA!U&U_g4_*)+JWzMZW7-><Q)7gf|l!h_?F-}r0PHYxgQ
zXBmi2GsOI#ix8WQ-}T%vWB3NnY*|K*qnDR>Ji{kq*%hP*$ncX!Q39RuXK`55)n{Ch
zI(KsVrc99Xg<K}G!76tZy+_a;H1ohKJ+4SFbAj`l;%1W!quRy*r`djr#`A4+C<BoU
znt~7}eggOQlJvpy5O413pN<pdOdw0&_tzSHD=B&GtPPyOX(&XZ1&|kPghMTF4p4ju
z&%by>oZg7r^UK<e8f%`+wIRNoC<^NYx~`e`?8Ztz3~w0RGJ&q}1p<;HGJ0uFzbSX^
z`NW-um9&1q@{(gT3X>}M6KRg@Wi;@GJc=&>sl51-Wli{(>0DOheH~AmvfYjuX!10p
zP-iX_<ET?0e)#nHTINtWlv)>mI2X6X-#C%G(1>l<SmS<uXs!}UODYxED|b`x^+FR}
zJ3X&9mZ6eA9|-FDgwKTfC;oE~qd-rD|J@hD8bo0bYPWc;Q)@UPr3q_6n$M>YPenTW
zfe^rSpnh8SIN!RHe?`Ge$>cuWL{mMVd>sV5a^%iE0mkhDdTHI9bC5YK?OsEdZ5U+%
zp5iY85AsoEJ+g&`Z__07_&ZiPvyaUhDGT|af0Ar&oWZXujgk1cpqLpSochHxkJkSu
zKr--4)`KKa9?1w^u0F;Et!b=4O<FZTH~b}+fxb@3K>mUQI0^p<9wMn}z$MHk2g*V+
zhbhlKF|?QEa1+3vWkP+FkSW(|-vu>sJBXNAcvtRwcIgi4H$4&Um{__^kOzDsgpY8e
zeP-#h-j=q~Ed(^an37Qd>&X=<94$cjqdBry^i4q7+C8#7K`=E1FU*X)YzxsYAaT84
zB0ilGA+N8>lbOTKo8=Vvg@}Cqqe5-!N3H5-cu3>#XylJq0yB`nQ?jgh48t8{T81*h
z2gj(O>2U`SIyDM96H^{V?m7V0#4Vq-pKzR8v`aoo#M22_KpXAZTQB@`W2GSrzjjCi
zEyMp!`2no;TZa^jO+H(Y4DFq#e@^Nu6usCP&A6jifsLYnaTWpyy56y(vk<e{O;{6v
zf)p8ffaRCTJYds3Py|}8YDGS4zQnqC;aMrhzBcGu<NUl;>fa)EmAdshNy-Il{QT6e
z9!#Lk2(4?4;k86PxV(6+Lu*LJPyfV-GlFNmferjzR3q{|Y9HuQGM{KO-bZt-NbmHl
z)UG9n9faK9_X(aN1*ecEHU>B=XR92|Cn>PYV8q}HzVd+LBICc0356_0Aha!~Zy<yv
zNfTK}p{!Q^LeT<F@bkWKRH?iT^p8ol)<x!!h5Q%G|9X8FrW=O_q}X^8FsqIkdw5NN
zxk%AnLw0gA1MKnX`73hLtHgbru%0U?k{Lf3K0J?7XD1|F>@3xk8E}9=>^Q>+6i-!3
z{1+11c@%dpc03z+4)_4$g+-ItEjSGg6OVEN6E;NqI~Q5OsV9?WGF=qi^n??VPMRRk
zbV}5;2T?X?bh^+Hb|=PwmBk^*;}595x99+CX8+kfLx|wG^lQozfl%jcZ!3E<jEe4V
z=Ni9E!@R@7Hn6I>IcK9KdsIN>|I6ps6B&q8?h2t8F<1SHxxKes1nRVyK7H*u*8LM8
z$|5-ujD^pK;b1s@C#@Tj*zf>>n-6Xjd-<XV&tbHhs9QeUC-$GX>R-GTfQ83@ZLz$E
z(cFPHu9lYdYshZz$oPnbBLw2dly$F>hIb&+?H;&_Ilba968qsaVtNp<>l!C`aJVSf
zs<6Y&nTqp-iW+$8k>__YSW*a|+ir&s#?!cpOyU(Y>0r7NA$>K^So8UaugrvtNn>>2
zhar6aLpaNcUP1q`cr$^}>vU+mnur#$H!Ti88h>^i_2L!j<x=ba@y(QlQ7bt59B$1h
z&=3!4ex3*=m@T6Z)MOXpbkpb*c0YhDbG|9hV52LTyJS9>h3Hh}&;uc0V}Mk4waBD9
zOLUR`E~d<obC!dZe;`P6ygA((dY+xm0Zyli9SjB1s*t6f(F&mk&SdVv{{^L<6SW9l
z%~7j9O&NH)m2Chn2m{Te_jH#lvjna{!cc@ZwEoe8GQRPQ-!eUxsdI_^LHl83y<|-k
zoegpv1X-knrft$U5x3~`s|Pix5Y^-#I<JC}pK}jAx!d^rIn^40_UX?pzEH-O30!Jk
z6>W&hd=bNSRGT18fd$A<Tjtp$T0A%LRbiy|PCJz(7UPE*crEp|M9CwCHSsFZKVg6Z
zx7#uqhaQ25nM8o56(lJ%lVAZo<eDBJB6(dXMuDUZAyCTa<1_Si#X8eJ3X>ys1<A?O
zA3`$XpGM`HkVB@iE@lFFV0^#VD4O-zYY#XxvKnPW%bwcywqpP5U{uhExZHke=HBnP
zZ2qkH975;70o`h`q=hsdDxT{U+lOSl^DnmuVK>ddW9IgpP6C{r@zl?=;IQsPoX77k
zVFa%I;=+fZqVTbLfZa(HquHF25%ppZA_rZ3z7=<2S?4L9cbJ*?aL*@6HGJL>Rz%!}
zVyUZ#fej>`O%#MQ{Z#EE_cRvjd2#bW_dKh4U(cD8mn06P53t#28Nj1z3Qasq1yaig
zBaFvC$|W%oDRQYed%zX0XOF*|I_)_ieG#>m_2Pu!ro!LKZmemGA83|0Ts~;M2o_k0
z*WF$Eo+zYDqjVoGzCA=@kmey6JG*iYoVS(<E+f;l&F?MJJRgmGUt&tw_%9q2<d~;m
z7_>dVZs@&hKa!%*)R%cMHHj8{h5H?o?5L&G)OgCR&Er0SDYkwgnB}l2+%}bXcg6n7
zd4u~(Wstv|yLC?U>-Jhc5jIx?NnOjJvFNvc6v!$;F8lMTZ!c#>LRR9EJt&K17aYc!
zMt=R*|6QR$5=`l1az%;$w}`5w%GJ%PL&OBGnOah2hTl`P!@tc4Au?CXr&75VQ!M-c
zcB;TGw)Nx3ny66VCM^BGQNo8RF1<Lbw213@r#^5(r)Rrx+{>*ZZPlmr%-1n}TUDZR
zZ=DG^bHIAjYn__w1e+&`R2VgcOk|%K%@y4CwM_rh<o%zXT+Y_i3p(~x>xY3{(RBIn
zqgVI-wXKcY6Wg@=&z?x7gm1=ECatnq+VMiMZraV~ya(f+UFWWw*R!o8eyMW3zs=KQ
zp&tqjep@Wn=mj^wq0J13)dGPZQk?QagML--KBFXYFZTPS{nL%soCz&sxx{o+$n#U;
z&PU5CQvY0%eGoTkxt8MHPxC?xql$o49+&cNmhyLjs&={g)Ap7;_nIf_)@1!MF>Ke;
z{nLxYHf!a)eD`_v%<#@Tsx9cFr9Y(JI5Q}ys=oa6#NpxIy;HQx?|Q#HEo-d8viY5|
z{iT&cKQ&{#oo<CzoBXY1_;+ghozoUKME&bxo~|=mX=7)Xwp91T+|_Tq+%u;x`mMS5
zliiVUi<zm@>qBpO1((g=Hr;5?9k2gd2j~A*p0Tt2OqVoreVrky#2A{gK!x+3-o$84
zuiYYoJU0y=FRQMwU$V5C)2`>cyT@sHKb4TX?s-M}j}wlczwci5^PyX$z|w^VF||K}
zp6Fk+RVmUxoE#=GQ<Tkk!t6&HHJV+{67g(1e-{Wn1<n?n-BcV}@$~3#cI1|5R4a$0
z<puq_sbYo#H5U47e=5$4sn{*vKkZM4`lTSzw$=N^17zf{d5d;5bIfo(W+TE^{%)ZG
z$GcOj4{6Lb(mguS;IPLGEe#*<1ODGd47B-$@~2AzuW4>H*jFPof7>bT2Yq+jpF-v$
ypkovuPXl9mP)weJQXJ4ZgK=ILhtt{(e*b4I^w(J~sVGy(00f?{elF{r5}E*VC68$U

literal 0
HcmV?d00001

diff --git a/_modules/index.html b/_modules/index.html
new file mode 100644
index 000000000..f7d09cc7f
--- /dev/null
+++ b/_modules/index.html
@@ -0,0 +1,522 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Overview: module code &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/index';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Overview: module code</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>All modules for which code is available</h1>
+<ul><li><a href="lmflow.html">lmflow</a></li>
+<ul><li><a href="lmflow/args.html">lmflow.args</a></li>
+<li><a href="lmflow/datasets/dataset.html">lmflow.datasets.dataset</a></li>
+<li><a href="lmflow/datasets/multi_modal_dataset.html">lmflow.datasets.multi_modal_dataset</a></li>
+<li><a href="lmflow/models/auto_model.html">lmflow.models.auto_model</a></li>
+<li><a href="lmflow/models/base_model.html">lmflow.models.base_model</a></li>
+<li><a href="lmflow/models/decoder_model.html">lmflow.models.decoder_model</a></li>
+<li><a href="lmflow/models/encoder_decoder_model.html">lmflow.models.encoder_decoder_model</a></li>
+<li><a href="lmflow/models/hf_decoder_model.html">lmflow.models.hf_decoder_model</a></li>
+<li><a href="lmflow/models/hf_encoder_decoder_model.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li><a href="lmflow/models/hf_model_mixin.html">lmflow.models.hf_model_mixin</a></li>
+<li><a href="lmflow/models/hf_text_regression_model.html">lmflow.models.hf_text_regression_model</a></li>
+<li><a href="lmflow/models/interfaces/tunable.html">lmflow.models.interfaces.tunable</a></li>
+<li><a href="lmflow/models/regression_model.html">lmflow.models.regression_model</a></li>
+<li><a href="lmflow/models/text_regression_model.html">lmflow.models.text_regression_model</a></li>
+<li><a href="lmflow/models/vision2seq_model.html">lmflow.models.vision2seq_model</a></li>
+<li><a href="lmflow/models/vision_encoder/clip_encoder.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+<li><a href="lmflow/optim/adabelief.html">lmflow.optim.adabelief</a></li>
+<li><a href="lmflow/optim/adabound.html">lmflow.optim.adabound</a></li>
+<li><a href="lmflow/optim/adadelta.html">lmflow.optim.adadelta</a></li>
+<li><a href="lmflow/optim/adagrad.html">lmflow.optim.adagrad</a></li>
+<li><a href="lmflow/optim/adam.html">lmflow.optim.adam</a></li>
+<li><a href="lmflow/optim/adamax.html">lmflow.optim.adamax</a></li>
+<li><a href="lmflow/optim/adamp.html">lmflow.optim.adamp</a></li>
+<li><a href="lmflow/optim/adamw_schedule_free.html">lmflow.optim.adamw_schedule_free</a></li>
+<li><a href="lmflow/optim/adan.html">lmflow.optim.adan</a></li>
+<li><a href="lmflow/optim/dummy.html">lmflow.optim.dummy</a></li>
+<li><a href="lmflow/optim/lamb.html">lmflow.optim.lamb</a></li>
+<li><a href="lmflow/optim/lars.html">lmflow.optim.lars</a></li>
+<li><a href="lmflow/optim/nadam.html">lmflow.optim.nadam</a></li>
+<li><a href="lmflow/optim/novograd.html">lmflow.optim.novograd</a></li>
+<li><a href="lmflow/optim/radam.html">lmflow.optim.radam</a></li>
+<li><a href="lmflow/optim/sgd_schedule_free.html">lmflow.optim.sgd_schedule_free</a></li>
+<li><a href="lmflow/optim/sgdp.html">lmflow.optim.sgdp</a></li>
+<li><a href="lmflow/optim/sophia.html">lmflow.optim.sophia</a></li>
+<li><a href="lmflow/optim/yogi.html">lmflow.optim.yogi</a></li>
+<li><a href="lmflow/pipeline/auto_pipeline.html">lmflow.pipeline.auto_pipeline</a></li>
+<li><a href="lmflow/pipeline/base_aligner.html">lmflow.pipeline.base_aligner</a></li>
+<li><a href="lmflow/pipeline/base_pipeline.html">lmflow.pipeline.base_pipeline</a></li>
+<li><a href="lmflow/pipeline/base_tuner.html">lmflow.pipeline.base_tuner</a></li>
+<li><a href="lmflow/pipeline/dpo_aligner.html">lmflow.pipeline.dpo_aligner</a></li>
+<li><a href="lmflow/pipeline/dpov2_aligner.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li><a href="lmflow/pipeline/evaluator.html">lmflow.pipeline.evaluator</a></li>
+<li><a href="lmflow/pipeline/finetuner.html">lmflow.pipeline.finetuner</a></li>
+<li><a href="lmflow/pipeline/inferencer.html">lmflow.pipeline.inferencer</a></li>
+<li><a href="lmflow/pipeline/iterative_dpo_aligner.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li><a href="lmflow/pipeline/raft_aligner.html">lmflow.pipeline.raft_aligner</a></li>
+<li><a href="lmflow/pipeline/rm_inferencer.html">lmflow.pipeline.rm_inferencer</a></li>
+<li><a href="lmflow/pipeline/rm_tuner.html">lmflow.pipeline.rm_tuner</a></li>
+<li><a href="lmflow/pipeline/utils/dpov2_dataprocessor.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li><a href="lmflow/pipeline/utils/dpov2_trainer.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li><a href="lmflow/pipeline/utils/memory_safe_dpov2_align.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li><a href="lmflow/pipeline/utils/memory_safe_vllm_inference.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li><a href="lmflow/pipeline/utils/peft_trainer.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li><a href="lmflow/pipeline/utils/raft_trainer.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li><a href="lmflow/pipeline/utils/rm_dataprocessor.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li><a href="lmflow/pipeline/utils/rm_trainer.html">lmflow.pipeline.utils.rm_trainer</a></li>
+<li><a href="lmflow/pipeline/vllm_inferencer.html">lmflow.pipeline.vllm_inferencer</a></li>
+<li><a href="lmflow/tokenization/hf_decoder_model.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li><a href="lmflow/tokenization/hf_text_regression_model.html">lmflow.tokenization.hf_text_regression_model</a></li>
+<li><a href="lmflow/utils/common.html">lmflow.utils.common</a></li>
+<li><a href="lmflow/utils/constants.html">lmflow.utils.constants</a></li>
+<li><a href="lmflow/utils/conversation_template.html">lmflow.utils.conversation_template</a></li>
+<ul><li><a href="lmflow/utils/conversation_template/base.html">lmflow.utils.conversation_template.base</a></li>
+<li><a href="lmflow/utils/conversation_template/chatglm.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li><a href="lmflow/utils/conversation_template/chatml.html">lmflow.utils.conversation_template.chatml</a></li>
+<li><a href="lmflow/utils/conversation_template/deepseek.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li><a href="lmflow/utils/conversation_template/gemma.html">lmflow.utils.conversation_template.gemma</a></li>
+<li><a href="lmflow/utils/conversation_template/hymba.html">lmflow.utils.conversation_template.hymba</a></li>
+<li><a href="lmflow/utils/conversation_template/internlm.html">lmflow.utils.conversation_template.internlm</a></li>
+<li><a href="lmflow/utils/conversation_template/llama.html">lmflow.utils.conversation_template.llama</a></li>
+<li><a href="lmflow/utils/conversation_template/phi.html">lmflow.utils.conversation_template.phi</a></li>
+<li><a href="lmflow/utils/conversation_template/qwen.html">lmflow.utils.conversation_template.qwen</a></li>
+<li><a href="lmflow/utils/conversation_template/yi.html">lmflow.utils.conversation_template.yi</a></li>
+<li><a href="lmflow/utils/conversation_template/zephyr.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul><li><a href="lmflow/utils/data_utils.html">lmflow.utils.data_utils</a></li>
+<li><a href="lmflow/utils/flash_attention/bloom_flash_attention.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li><a href="lmflow/utils/flash_attention/gpt2_flash_attention.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li><a href="lmflow/utils/flash_attention/gpt_neo_flash_attention.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li><a href="lmflow/utils/flash_attention/llama_flash_attention.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li><a href="lmflow/utils/flash_attention/triton_flash_attention.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+<li><a href="lmflow/utils/llava_conversation_lib.html">lmflow.utils.llava_conversation_lib</a></li>
+<li><a href="lmflow/utils/model.html">lmflow.utils.model</a></li>
+<li><a href="lmflow/utils/multimodal.html">lmflow.utils.multimodal</a></li>
+<li><a href="lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+<li><a href="lmflow/utils/versioning.html">lmflow.utils.versioning</a></li>
+<li><a href="lmflow/version.html">lmflow.version</a></li>
+</ul></ul>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow.html b/_modules/lmflow.html
new file mode 100644
index 000000000..a73459070
--- /dev/null
+++ b/_modules/lmflow.html
@@ -0,0 +1,454 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="index.html" class="nav-link">Module code</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span><span class="w"> </span><span class="nn">.version</span><span class="w"> </span><span class="kn">import</span> <span class="n">__version__</span> <span class="k">as</span> <span class="n">internal_version</span>
+
+<div class="viewcode-block" id="__version__">
+<a class="viewcode-back" href="../autoapi/lmflow/index.html#lmflow.__version__">[docs]</a>
+<span class="n">__version__</span> <span class="o">=</span> <span class="n">internal_version</span></div>
+
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">check_min_version</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.utils.versions</span><span class="w"> </span><span class="kn">import</span> <span class="n">require_version</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow</span><span class="w"> </span><span class="kn">import</span> <span class="n">args</span><span class="p">,</span> <span class="n">datasets</span><span class="p">,</span> <span class="n">models</span><span class="p">,</span> <span class="n">pipeline</span><span class="p">,</span> <span class="n">utils</span>
+
+<span class="c1"># Will error if the minimal version of Transformers is not installed. Remove at your own risks.</span>
+<span class="n">check_min_version</span><span class="p">(</span><span class="s2">&quot;4.27.0.dev0&quot;</span><span class="p">)</span>
+
+<span class="n">require_version</span><span class="p">(</span><span class="s2">&quot;datasets&gt;=1.8.0&quot;</span><span class="p">,</span> <span class="s2">&quot;To fix: pip install -r examples/pytorch/language-modeling/requirements.txt&quot;</span><span class="p">)</span>
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/args.html b/_modules/lmflow/args.html
new file mode 100644
index 000000000..5086fe1da
--- /dev/null
+++ b/_modules/lmflow/args.html
@@ -0,0 +1,2622 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.args &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/args';</script>
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.args</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.args</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;This script defines dataclasses: ModelArguments and DatasetArguments,</span>
+<span class="sd">that contain the arguments for the model and dataset used in training.</span>
+
+<span class="sd">It imports several modules, including dataclasses, field from typing, Optional from typing,</span>
+<span class="sd">require_version from transformers.utils.versions, MODEL_FOR_CAUSAL_LM_MAPPING,</span>
+<span class="sd">and TrainingArguments from transformers.</span>
+
+<span class="sd">MODEL_CONFIG_CLASSES is assigned a list of the model config classes from</span>
+<span class="sd">MODEL_FOR_CAUSAL_LM_MAPPING. MODEL_TYPES is assigned a tuple of the model types</span>
+<span class="sd">extracted from the MODEL_CONFIG_CLASSES.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">dataclasses</span><span class="w"> </span><span class="kn">import</span> <span class="n">dataclass</span><span class="p">,</span> <span class="n">field</span><span class="p">,</span> <span class="n">fields</span><span class="p">,</span> <span class="n">Field</span><span class="p">,</span> <span class="n">make_dataclass</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">pathlib</span><span class="w"> </span><span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">Dict</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">MODEL_FOR_CAUSAL_LM_MAPPING</span><span class="p">,</span>
+    <span class="n">TrainingArguments</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.utils.versions</span><span class="w"> </span><span class="kn">import</span> <span class="n">require_version</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.versioning</span><span class="w"> </span><span class="kn">import</span> <span class="n">is_flash_attn_available</span>
+
+<div class="viewcode-block" id="MODEL_CONFIG_CLASSES">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.MODEL_CONFIG_CLASSES">[docs]</a>
+<span class="n">MODEL_CONFIG_CLASSES</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">MODEL_FOR_CAUSAL_LM_MAPPING</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span></div>
+
+<div class="viewcode-block" id="MODEL_TYPES">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.MODEL_TYPES">[docs]</a>
+<span class="n">MODEL_TYPES</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">conf</span><span class="o">.</span><span class="n">model_type</span> <span class="k">for</span> <span class="n">conf</span> <span class="ow">in</span> <span class="n">MODEL_CONFIG_CLASSES</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="OptimizerNames">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">OptimizerNames</span><span class="p">():</span>
+<div class="viewcode-block" id="OptimizerNames.DUMMY">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.DUMMY">[docs]</a>
+    <span class="n">DUMMY</span> <span class="o">=</span> <span class="s2">&quot;dummy&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.ADABELIEF">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADABELIEF">[docs]</a>
+    <span class="n">ADABELIEF</span> <span class="o">=</span> <span class="s2">&quot;adabelief&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.ADABOUND">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADABOUND">[docs]</a>
+    <span class="n">ADABOUND</span> <span class="o">=</span> <span class="s2">&quot;adabound&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.LARS">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.LARS">[docs]</a>
+    <span class="n">LARS</span> <span class="o">=</span> <span class="s2">&quot;lars&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.LAMB">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.LAMB">[docs]</a>
+    <span class="n">LAMB</span> <span class="o">=</span> <span class="s2">&quot;lamb&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.ADAMAX">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADAMAX">[docs]</a>
+    <span class="n">ADAMAX</span> <span class="o">=</span> <span class="s2">&quot;adamax&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.NADAM">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.NADAM">[docs]</a>
+    <span class="n">NADAM</span> <span class="o">=</span> <span class="s2">&quot;nadam&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.RADAM">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.RADAM">[docs]</a>
+    <span class="n">RADAM</span> <span class="o">=</span> <span class="s2">&quot;radam&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.ADAMP">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADAMP">[docs]</a>
+    <span class="n">ADAMP</span> <span class="o">=</span> <span class="s2">&quot;adamp&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.SGDP">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.SGDP">[docs]</a>
+    <span class="n">SGDP</span> <span class="o">=</span> <span class="s2">&quot;sgdp&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.YOGI">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.YOGI">[docs]</a>
+    <span class="n">YOGI</span> <span class="o">=</span> <span class="s2">&quot;yogi&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.SOPHIA">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.SOPHIA">[docs]</a>
+    <span class="n">SOPHIA</span> <span class="o">=</span> <span class="s2">&quot;sophia&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.ADAN">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADAN">[docs]</a>
+    <span class="n">ADAN</span> <span class="o">=</span> <span class="s2">&quot;adan&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.ADAM">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADAM">[docs]</a>
+    <span class="n">ADAM</span> <span class="o">=</span> <span class="s2">&quot;adam&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.NOVOGRAD">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.NOVOGRAD">[docs]</a>
+    <span class="n">NOVOGRAD</span> <span class="o">=</span> <span class="s2">&quot;novograd&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.ADADELTA">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADADELTA">[docs]</a>
+    <span class="n">ADADELTA</span> <span class="o">=</span> <span class="s2">&quot;adadelta&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.ADAGRAD">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADAGRAD">[docs]</a>
+    <span class="n">ADAGRAD</span> <span class="o">=</span> <span class="s2">&quot;adagrad&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.ADAMW_SCHEDULE_FREE">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADAMW_SCHEDULE_FREE">[docs]</a>
+    <span class="n">ADAMW_SCHEDULE_FREE</span> <span class="o">=</span> <span class="s2">&quot;adamw_schedule_free&quot;</span></div>
+
+<div class="viewcode-block" id="OptimizerNames.SGD_SCHEDULE_FREE">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.SGD_SCHEDULE_FREE">[docs]</a>
+    <span class="n">SGD_SCHEDULE_FREE</span> <span class="o">=</span> <span class="s2">&quot;sgd_schedule_free&quot;</span></div>
+</div>
+
+    
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="ModelArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">ModelArguments</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Define a class ModelArguments using the dataclass decorator. </span>
+<span class="sd">    The class contains several optional parameters that can be used to configure a model. </span>
+<span class="sd">    </span>
+<span class="sd">    model_name_or_path : str</span>
+<span class="sd">        a string representing the path or name of a pretrained</span>
+<span class="sd">        model checkpoint for weights initialization. If None, a model will be trained from scratch.</span>
+
+<span class="sd">    model_type :  str</span>
+<span class="sd">        a string representing the type of model to use if training from</span>
+<span class="sd">        scratch. If not provided, a pretrained model will be used.</span>
+<span class="sd">    </span>
+<span class="sd">    config_overrides :  str</span>
+<span class="sd">        a string representing the default config settings to override</span>
+<span class="sd">        when training a model from scratch.</span>
+<span class="sd">    </span>
+<span class="sd">    config_name : str</span>
+<span class="sd">        a string representing the name or path of the pretrained config to</span>
+<span class="sd">        use, if different from the model_name_or_path.</span>
+<span class="sd">    </span>
+<span class="sd">    tokenizer_name :  str</span>
+<span class="sd">        a string representing the name or path of the pretrained tokenizer</span>
+<span class="sd">        to use, if different from the model_name_or_path.</span>
+
+<span class="sd">    cache_dir :  str</span>
+<span class="sd">        a string representing the path to the directory where pretrained models</span>
+<span class="sd">        downloaded from huggingface.co will be stored.</span>
+
+<span class="sd">    use_fast_tokenizer : bool</span>
+<span class="sd">        a boolean indicating whether to use a fast tokenizer (backed by the</span>
+<span class="sd">        tokenizers library) or not.</span>
+
+<span class="sd">    model_revision :  str</span>
+<span class="sd">        a string representing the specific model version to use (can be a</span>
+<span class="sd">        branch name, tag name, or commit id).</span>
+
+<span class="sd">    token : Optional[str]</span>
+<span class="sd">        Necessary when accessing a private model/dataset.</span>
+
+<span class="sd">    torch_dtype :  str</span>
+<span class="sd">        a string representing the dtype to load the model under. If auto is</span>
+<span class="sd">        passed, the dtype will be automatically derived from the model&#39;s weights.</span>
+
+<span class="sd">    use_ram_optimized_load : bool</span>
+<span class="sd">        a boolean indicating whether to use disk mapping when memory is not</span>
+<span class="sd">        enough.</span>
+<span class="sd">        </span>
+<span class="sd">    use_int8 : bool</span>
+<span class="sd">        a boolean indicating whether to load int8 quantization for inference.</span>
+<span class="sd">        </span>
+<span class="sd">    load_in_4bit : bool</span>
+<span class="sd">        whether to load the model in 4bit</span>
+<span class="sd">        </span>
+<span class="sd">    model_max_length : int</span>
+<span class="sd">        The maximum length of the model.</span>
+<span class="sd">        </span>
+<span class="sd">    truncation_side : str</span>
+<span class="sd">        The side on which the model should have truncation applied.</span>
+<span class="sd">        </span>
+<span class="sd">    arch_type : str</span>
+<span class="sd">        Model architecture type.</span>
+<span class="sd">    padding_side : str</span>
+<span class="sd">        The side on which the tokenizer should have padding applied.</span>
+<span class="sd">    eos_padding : bool</span>
+<span class="sd">        whether to pad with eos token instead of pad token.</span>
+<span class="sd">    ignore_bias_buffers : bool</span>
+<span class="sd">        fix for DDP issues with LM bias/mask buffers - invalid scalar type,`inplace operation.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+<div class="viewcode-block" id="ModelArguments.model_name_or_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.model_name_or_path">[docs]</a>
+    <span class="n">model_name_or_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;The model checkpoint for weights initialization.Don&#39;t set if you want to train a model from scratch.&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.lora_model_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.lora_model_path">[docs]</a>
+    <span class="n">lora_model_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;The incremental model diff introduced by LoRA finetuning.&quot;</span>
+                <span class="s2">&quot; Along with the original non-finetuned model forms the whole&quot;</span>
+                <span class="s2">&quot; finetuned model.&quot;</span>
+            <span class="p">)</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.model_type">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.model_type">[docs]</a>
+    <span class="n">model_type</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;If training from scratch, pass a model type from the list: &quot;</span> <span class="o">+</span> <span class="s2">&quot;, &quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">MODEL_TYPES</span><span class="p">)},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.config_overrides">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.config_overrides">[docs]</a>
+    <span class="n">config_overrides</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;Override some existing default config settings when a model is trained from scratch. Example: &quot;</span>
+                <span class="s2">&quot;n_embd=10,resid_pdrop=0.2,scale_attn_weights=false,summary_type=cls_index&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.arch_type">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.arch_type">[docs]</a>
+    <span class="n">arch_type</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;decoder_only&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;Model architecture type.&quot;</span><span class="p">),</span>
+            <span class="s2">&quot;choices&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;decoder_only&quot;</span><span class="p">,</span> <span class="s2">&quot;encoder_decoder&quot;</span><span class="p">,</span> <span class="s2">&quot;text_regression&quot;</span><span class="p">,</span> <span class="s2">&quot;vision_encoder_decoder&quot;</span><span class="p">],</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.config_name">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.config_name">[docs]</a>
+    <span class="n">config_name</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Pretrained config name or path if not the same as model_name&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.tokenizer_name">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.tokenizer_name">[docs]</a>
+    <span class="n">tokenizer_name</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Pretrained tokenizer name or path if not the same as model_name&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.cache_dir">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.cache_dir">[docs]</a>
+    <span class="n">cache_dir</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Where do you want to store the pretrained models downloaded from huggingface.co&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.use_fast_tokenizer">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_fast_tokenizer">[docs]</a>
+    <span class="n">use_fast_tokenizer</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to use one of the fast tokenizer (backed by the tokenizers library) or not.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.model_revision">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.model_revision">[docs]</a>
+    <span class="n">model_revision</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;main&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The specific model version to use (can be a branch name, tag name or commit id).&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.token">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.token">[docs]</a>
+    <span class="n">token</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span><span class="s2">&quot;Necessary to specify when accessing a private model/dataset.&quot;</span><span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.trust_remote_code">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.trust_remote_code">[docs]</a>
+    <span class="n">trust_remote_code</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;Whether to trust remote code when loading model.&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.torch_dtype">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.torch_dtype">[docs]</a>
+    <span class="n">torch_dtype</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;Override the default `torch.dtype` and load the model under this dtype. If `auto` is passed, the &quot;</span>
+                <span class="s2">&quot;dtype will be automatically derived from the model&#39;s weights.&quot;</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;choices&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;auto&quot;</span><span class="p">,</span> <span class="s2">&quot;bfloat16&quot;</span><span class="p">,</span> <span class="s2">&quot;float16&quot;</span><span class="p">,</span> <span class="s2">&quot;float32&quot;</span><span class="p">],</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.use_dora">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_dora">[docs]</a>
+    <span class="n">use_dora</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to dora, https://github.com/NVlabs/DoRA.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.use_lora">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_lora">[docs]</a>
+    <span class="n">use_lora</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to lora.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.use_qlora">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_qlora">[docs]</a>
+    <span class="n">use_qlora</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to use qlora.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.bits">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.bits">[docs]</a>
+    <span class="n">bits</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The number of bits for quantization.&quot;</span><span class="p">,</span>
+                  <span class="s2">&quot;choices&quot;</span><span class="p">:</span> <span class="p">[</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">],</span> <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.quant_type">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.quant_type">[docs]</a>
+    <span class="n">quant_type</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s1">&#39;nf4&#39;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The quantization type for quantization.&quot;</span><span class="p">,</span>
+                  <span class="s2">&quot;choices&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;nf4&quot;</span><span class="p">,</span> <span class="s2">&quot;fp4&quot;</span><span class="p">],</span> <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.double_quant">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.double_quant">[docs]</a>
+    <span class="n">double_quant</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to use double quantization.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.lora_r">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.lora_r">[docs]</a>
+    <span class="n">lora_r</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">8</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the rank of the lora parameters. The smaller lora_r is , the fewer parameters lora has.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.lora_alpha">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.lora_alpha">[docs]</a>
+    <span class="n">lora_alpha</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">32</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Merging ratio between the fine-tuned model and the original. This is controlled by a parameter called alpha in the paper.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.lora_target_modules">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.lora_target_modules">[docs]</a>
+    <span class="n">lora_target_modules</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Model modules to apply LoRA to. Use comma to separate multiple modules.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.lora_dropout">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.lora_dropout">[docs]</a>
+    <span class="n">lora_dropout</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The dropout rate in lora.linear.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.save_aggregated_lora">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.save_aggregated_lora">[docs]</a>
+    <span class="n">save_aggregated_lora</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to save aggregated lora.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.use_ram_optimized_load">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_ram_optimized_load">[docs]</a>
+    <span class="n">use_ram_optimized_load</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether use disk mapping when memory is not enough.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.use_flash_attention">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_flash_attention">[docs]</a>
+    <span class="n">use_flash_attention</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;whether use flash attention layer to reduce GPU memory with&quot;</span>
+                <span class="s2">&quot; higher time cost.&quot;</span>
+            <span class="p">)</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.truncate_to_model_max_length">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.truncate_to_model_max_length">[docs]</a>
+    <span class="n">truncate_to_model_max_length</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;whether truncate the dataset to model max length.&quot;</span>
+            <span class="p">)</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.do_rope_scaling">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.do_rope_scaling">[docs]</a>
+    <span class="n">do_rope_scaling</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;whether do ROPE scaling for llama model.&quot;</span>
+                <span class="s2">&quot;Linear_scaling credits to the Reddit user /u/kaiokendev.&quot;</span>
+                <span class="s2">&quot;https://arxiv.org/abs/2306.15595&quot;</span>
+                <span class="s2">&quot;NTK_scaling credits to the Reddit users /u/bloc97 and /u/emozilla.&quot;</span>
+                <span class="s2">&quot;https://www.reddit.com/r/LocalLLaMA/comments/14lz7j5/ntkaware_scaled_rope_allows_llama_models_to_have/&quot;</span>
+            <span class="p">)</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.rope_pi_ratio">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.rope_pi_ratio">[docs]</a>
+    <span class="n">rope_pi_ratio</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;the ratio of pi in RoPE scaling.&quot;</span>
+            <span class="p">)</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.rope_ntk_ratio">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.rope_ntk_ratio">[docs]</a>
+    <span class="n">rope_ntk_ratio</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;the ratio of NTK in RoPE scaling.&quot;</span>
+            <span class="p">)</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.use_int8">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_int8">[docs]</a>
+    <span class="n">use_int8</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;whether to load int8 quantization for inference&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.load_in_4bit">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.load_in_4bit">[docs]</a>
+    <span class="n">load_in_4bit</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;whether to load the model in 4bit&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.model_max_length">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.model_max_length">[docs]</a>
+    <span class="n">model_max_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+            <span class="s2">&quot;The maximum length of the model. When not specified, &quot;</span>
+            <span class="s2">&quot;will follow the model&#39;s default max length. (i.e., tokenizer.model_max_length)&quot;</span><span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.truncation_side">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.truncation_side">[docs]</a>
+    <span class="n">truncation_side</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;The side on which the tokenizer should have truncation applied. &quot;</span>
+                <span class="s2">&quot;When not specified, will follow the tokenizer&#39;s default truncation strategy. &quot;</span>
+                <span class="s2">&quot;(i.e., tokenizer.truncation_side)&quot;</span><span class="p">),</span>
+            <span class="s2">&quot;choices&quot;</span><span class="p">:</span> <span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="s2">&quot;left&quot;</span><span class="p">,</span> <span class="s2">&quot;right&quot;</span><span class="p">],</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.padding_side">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.padding_side">[docs]</a>
+    <span class="n">padding_side</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s1">&#39;right&#39;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;The side on which the tokenizer should have padding applied. &quot;</span>
+                <span class="s2">&quot;LMFlow uses right padding by default. When set to `auto`, will &quot;</span>
+                <span class="s2">&quot;use padding_side from tokenizer.padding_side.&quot;</span><span class="p">),</span>
+            <span class="s2">&quot;choices&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;right&quot;</span><span class="p">,</span> <span class="s2">&quot;left&quot;</span><span class="p">,</span> <span class="s2">&quot;auto&quot;</span><span class="p">],</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.eos_padding">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.eos_padding">[docs]</a>
+    <span class="n">eos_padding</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> 
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;whether to pad with eos token&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="ModelArguments.ignore_bias_buffers">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.ignore_bias_buffers">[docs]</a>
+    <span class="n">ignore_bias_buffers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="c1"># debug argument for distributed training</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;fix for DDP issues with LM bias/mask buffers - invalid scalar type,`inplace operation. See&quot;</span>
+            <span class="s2">&quot;https://github.com/huggingface/transformers/issues/22482#issuecomment-1595790992&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+    
+
+<div class="viewcode-block" id="ModelArguments.__post_init__">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.__post_init__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__post_init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">config_overrides</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">config_name</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_name_or_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;--config_overrides can&#39;t be used in combination with --config_name or --model_name_or_path&quot;</span>
+            <span class="p">)</span>
+            
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_qlora</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_lora</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;use_qlora is set to True, but use_lora is not set to True. Setting use_lora to True.&quot;</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">use_lora</span> <span class="o">=</span> <span class="kc">True</span>
+                
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_flash_attention</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">is_flash_attn_available</span><span class="p">():</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">use_flash_attention</span> <span class="o">=</span> <span class="kc">False</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;Flash attention is not available in the current environment. Disabling flash attention.&quot;</span><span class="p">)</span>
+                
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">lora_target_modules</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">lora_target_modules</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">split_args</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">lora_target_modules</span><span class="p">)</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="VisModelArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">VisModelArguments</span><span class="p">(</span><span class="n">ModelArguments</span><span class="p">):</span>
+<div class="viewcode-block" id="VisModelArguments.low_resource">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.low_resource">[docs]</a>
+    <span class="n">low_resource</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Use 8 bit and float16 when loading llm&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="VisModelArguments.custom_model">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.custom_model">[docs]</a>
+    <span class="n">custom_model</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;flag for the model from huggingface or not&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="VisModelArguments.pretrained_language_projection_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.pretrained_language_projection_path">[docs]</a>
+    <span class="n">pretrained_language_projection_path</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;path for model pretrained_language_projection_path&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="VisModelArguments.custom_vision_model">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.custom_vision_model">[docs]</a>
+    <span class="n">custom_vision_model</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;flag for the model from huggingface or not&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="VisModelArguments.image_encoder_name_or_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.image_encoder_name_or_path">[docs]</a>
+    <span class="n">image_encoder_name_or_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;The name or path of the image encoder to use.&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="VisModelArguments.qformer_name_or_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.qformer_name_or_path">[docs]</a>
+    <span class="n">qformer_name_or_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;llm model in multi-modality model&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="VisModelArguments.llm_model_name_or_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.llm_model_name_or_path">[docs]</a>
+    <span class="n">llm_model_name_or_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;llm model in multi-modality model&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="VisModelArguments.use_prompt_cache">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.use_prompt_cache">[docs]</a>
+    <span class="n">use_prompt_cache</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to use prompt cache.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="VisModelArguments.prompt_cache_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.prompt_cache_path">[docs]</a>
+    <span class="n">prompt_cache_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Path to prompt cache.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="VisModelArguments.llava_loading">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.llava_loading">[docs]</a>
+    <span class="n">llava_loading</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to load module by module from pretrained model.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="VisModelArguments.with_qformer">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.with_qformer">[docs]</a>
+    <span class="n">with_qformer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to use qformer.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="VisModelArguments.vision_select_layer">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.vision_select_layer">[docs]</a>
+    <span class="n">vision_select_layer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=-</span><span class="mi">2</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Which layer to select in vision model.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="VisModelArguments.llava_pretrain_model_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.llava_pretrain_model_path">[docs]</a>
+    <span class="n">llava_pretrain_model_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Path to llava pretrained model.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="VisModelArguments.save_pretrain_model_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.save_pretrain_model_path">[docs]</a>
+    <span class="n">save_pretrain_model_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Path to pretrained model.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="DatasetArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">DatasetArguments</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Define a class DatasetArguments using the dataclass decorator.</span>
+<span class="sd">    The class contains several optional parameters that can be used to configure a dataset for a language model.</span>
+
+
+<span class="sd">    dataset_path : str</span>
+<span class="sd">        a string representing the path of the dataset to use.</span>
+
+<span class="sd">    dataset_name : str</span>
+<span class="sd">        a string representing the name of the dataset to use. The default value is &quot;customized&quot;.</span>
+
+<span class="sd">    is_custom_dataset : bool</span>
+<span class="sd">        a boolean indicating whether to use custom data. The default value is False.</span>
+
+<span class="sd">    customized_cache_dir : str</span>
+<span class="sd">        a string representing the path to the directory where customized dataset caches will be stored.</span>
+
+<span class="sd">    dataset_config_name : str</span>
+<span class="sd">        a string representing the configuration name of the dataset to use (via the datasets library).</span>
+
+<span class="sd">    train_file : str</span>
+<span class="sd">        a string representing the path to the input training data file (a text file).</span>
+
+<span class="sd">    validation_file : str</span>
+<span class="sd">        a string representing the path to the input evaluation data file to evaluate the perplexity on (a text file).</span>
+
+<span class="sd">    max_train_samples : int</span>
+<span class="sd">        an integer indicating the maximum number of training examples to use for debugging or quicker training.</span>
+<span class="sd">        If set, the training dataset will be truncated to this number.</span>
+
+<span class="sd">    max_eval_samples: int</span>
+<span class="sd">        an integer indicating the maximum number of evaluation examples to use for debugging or quicker training.</span>
+<span class="sd">        If set, the evaluation dataset will be truncated to this number.</span>
+
+<span class="sd">    streaming : bool</span>
+<span class="sd">        a boolean indicating whether to enable streaming mode.</span>
+
+<span class="sd">    block_size: int</span>
+<span class="sd">        an integer indicating the optional input sequence length after tokenization. The training dataset will be</span>
+<span class="sd">        truncated in blocks of this size for training.</span>
+
+<span class="sd">    train_on_prompt: bool</span>
+<span class="sd">        a boolean indicating whether to train on prompt for conversation datasets such as ShareGPT.</span>
+
+<span class="sd">    conversation_template: str</span>
+<span class="sd">        a string representing the template for conversation datasets.</span>
+
+<span class="sd">    The class also includes some additional parameters that can be used to configure the dataset further, such as `overwrite_cache`,</span>
+<span class="sd">    `validation_split_percentage`, `preprocessing_num_workers`, `disable_group_texts`, `demo_example_in_prompt`, `explanation_in_prompt`,</span>
+<span class="sd">    `keep_linebreaks`, and `prompt_structure`.</span>
+
+<span class="sd">    The field function is used to set default values and provide help messages for each parameter. The Optional type hint is</span>
+<span class="sd">    used to indicate that a parameter is optional. The metadata argument is used to provide additional information about</span>
+<span class="sd">    each parameter, such as a help message.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+<div class="viewcode-block" id="DatasetArguments.dataset_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.dataset_path">[docs]</a>
+    <span class="n">dataset_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The path of the dataset to use.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.dataset_name">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.dataset_name">[docs]</a>
+    <span class="n">dataset_name</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;customized&quot;</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Should be </span><span class="se">\&quot;</span><span class="s2">customized</span><span class="se">\&quot;</span><span class="s2">&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.is_custom_dataset">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.is_custom_dataset">[docs]</a>
+    <span class="n">is_custom_dataset</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;whether to use custom data&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.customized_cache_dir">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.customized_cache_dir">[docs]</a>
+    <span class="n">customized_cache_dir</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;.cache/llm-ft/datasets&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Where do you want to store the customized dataset caches&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.dataset_config_name">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.dataset_config_name">[docs]</a>
+    <span class="n">dataset_config_name</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The configuration name of the dataset to use (via the datasets library).&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.train_file">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.train_file">[docs]</a>
+    <span class="n">train_file</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The input training data file (a text file).&quot;</span><span class="p">})</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.validation_file">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.validation_file">[docs]</a>
+    <span class="n">validation_file</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;An optional input evaluation data file to evaluate the perplexity on (a text file).&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.max_train_samples">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.max_train_samples">[docs]</a>
+    <span class="n">max_train_samples</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;For debugging purposes or quicker training, truncate the number of training examples to this &quot;</span>
+                <span class="s2">&quot;value if set.&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.max_eval_samples">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.max_eval_samples">[docs]</a>
+    <span class="n">max_eval_samples</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">1e10</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;For debugging purposes or quicker training, truncate the number of evaluation examples to this &quot;</span>
+                <span class="s2">&quot;value if set.&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.streaming">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.streaming">[docs]</a>
+    <span class="n">streaming</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Enable streaming mode&quot;</span><span class="p">})</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.block_size">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.block_size">[docs]</a>
+    <span class="n">block_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;Optional input sequence length after tokenization. &quot;</span>
+                <span class="s2">&quot;The training dataset will be truncated in block of this size for training. &quot;</span>
+                <span class="s2">&quot;Default to the model max input length for single sentence inputs (take into account special tokens).&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.overwrite_cache">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.overwrite_cache">[docs]</a>
+    <span class="n">overwrite_cache</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Overwrite the cached training and evaluation sets&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.validation_split_percentage">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.validation_split_percentage">[docs]</a>
+    <span class="n">validation_split_percentage</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">5</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The percentage of the train set used as validation set in case there&#39;s no validation split&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.preprocessing_num_workers">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.preprocessing_num_workers">[docs]</a>
+    <span class="n">preprocessing_num_workers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The number of processes to use for the preprocessing.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.group_texts_batch_size">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.group_texts_batch_size">[docs]</a>
+    <span class="n">group_texts_batch_size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;Number of samples that will be grouped together to go though&quot;</span>
+                <span class="s2">&quot; `group_texts` operation. See `--disable_group_texts` for&quot;</span>
+                <span class="s2">&quot; detailed explanation of this operation.&quot;</span>
+            <span class="p">)</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.disable_group_texts">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.disable_group_texts">[docs]</a>
+    <span class="n">disable_group_texts</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;Whether we disable group of original samples together to&quot;</span>
+                <span class="s2">&quot; generate sample sequences of length `block_size`&quot;</span>
+                <span class="s2">&quot; By Default, it is True, which means the long samples&quot;</span>
+                <span class="s2">&quot; are truncated to `block_size` tokens&quot;</span>
+                <span class="s2">&quot; and short samples are padded to `block_size` tokens.&quot;</span>
+                <span class="s2">&quot; If set to False, we group every 1000 tokenized&quot;</span>
+                <span class="s2">&quot; sequences together, divide them into&quot;</span>
+                <span class="s2">&quot; [</span><span class="si">{total_num_tokens}</span><span class="s2"> / </span><span class="si">{block_size}</span><span class="s2">] sequences,&quot;</span>
+                <span class="s2">&quot; each with `block_size` tokens&quot;</span>
+                <span class="s2">&quot; (the remaining tokens are ommited.&quot;</span>
+                <span class="s2">&quot; This group text behavior is useful&quot;</span>
+                <span class="s2">&quot; for continual pretrain or pretrain.&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.keep_linebreaks">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.keep_linebreaks">[docs]</a>
+    <span class="n">keep_linebreaks</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to keep line breaks when using TXT files or not.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.test_file">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.test_file">[docs]</a>
+    <span class="n">test_file</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Evaluation File Path&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.train_on_prompt">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.train_on_prompt">[docs]</a>
+    <span class="n">train_on_prompt</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to train on prompt for conversation datasets such as ShareGPT.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DatasetArguments.conversation_template">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.conversation_template">[docs]</a>
+    <span class="n">conversation_template</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The template for conversation datasets.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="DatasetArguments.__post_init__">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.__post_init__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__post_init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">streaming</span><span class="p">:</span>
+            <span class="n">require_version</span><span class="p">(</span><span class="s2">&quot;datasets&gt;=2.0.0&quot;</span><span class="p">,</span> <span class="s2">&quot;The streaming feature requires `datasets&gt;=2.0.0`&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">dataset_name</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_file</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">validation_file</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Need either a dataset name or a training/validation file.&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_file</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">extension</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_file</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;.&quot;</span><span class="p">)[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+                <span class="k">assert</span> <span class="n">extension</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;csv&quot;</span><span class="p">,</span> <span class="s2">&quot;json&quot;</span><span class="p">,</span> <span class="s2">&quot;txt&quot;</span><span class="p">],</span> <span class="s2">&quot;`train_file` should be a csv, a json or a txt file.&quot;</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">validation_file</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">extension</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">validation_file</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;.&quot;</span><span class="p">)[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+                <span class="k">assert</span> <span class="n">extension</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;csv&quot;</span><span class="p">,</span> <span class="s2">&quot;json&quot;</span><span class="p">,</span> <span class="s2">&quot;txt&quot;</span><span class="p">],</span> <span class="s2">&quot;`validation_file` should be a csv, a json or a txt file.&quot;</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="MultiModalDatasetArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.MultiModalDatasetArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">MultiModalDatasetArguments</span><span class="p">(</span><span class="n">DatasetArguments</span><span class="p">):</span>
+<div class="viewcode-block" id="MultiModalDatasetArguments.image_folder">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.MultiModalDatasetArguments.image_folder">[docs]</a>
+    <span class="n">image_folder</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The folder of the image file.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="MultiModalDatasetArguments.image_aspect_ratio">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.MultiModalDatasetArguments.image_aspect_ratio">[docs]</a>
+    <span class="n">image_aspect_ratio</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;pad&quot;</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The ratio type&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="MultiModalDatasetArguments.is_multimodal">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.MultiModalDatasetArguments.is_multimodal">[docs]</a>
+    <span class="n">is_multimodal</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Flag for the modality type.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="MultiModalDatasetArguments.use_image_start_end">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.MultiModalDatasetArguments.use_image_start_end">[docs]</a>
+    <span class="n">use_image_start_end</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Flag for the modality type.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="MultiModalDatasetArguments.sep_style">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.MultiModalDatasetArguments.sep_style">[docs]</a>
+    <span class="n">sep_style</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;plain&quot;</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Sep style in multi_modality dataset.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="FinetunerArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">FinetunerArguments</span><span class="p">(</span><span class="n">TrainingArguments</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Adapt transformers.TrainingArguments</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+<div class="viewcode-block" id="FinetunerArguments.eval_dataset_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.eval_dataset_path">[docs]</a>
+    <span class="n">eval_dataset_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The path of the eval dataset to use.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.remove_unused_columns">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.remove_unused_columns">[docs]</a>
+    <span class="n">remove_unused_columns</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;whether to remove the unused columns in collate fn&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.finetune_part">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.finetune_part">[docs]</a>
+    <span class="n">finetune_part</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;language_projection&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the module to finetune.&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.save_language_projection">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.save_language_projection">[docs]</a>
+    <span class="n">save_language_projection</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;whether to save language projection layer in multi-modal models.&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.use_lisa">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.use_lisa">[docs]</a>
+    <span class="n">use_lisa</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;whether to use LISA training strategy.&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.lisa_activated_layers">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.lisa_activated_layers">[docs]</a>
+    <span class="n">lisa_activated_layers</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the number of activated layers in LISA.&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.lisa_interval_steps">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.lisa_interval_steps">[docs]</a>
+    <span class="n">lisa_interval_steps</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">20</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the number of steps in each freezing interval of LISA, i.e. the selected unfreezed layers are randomly switched every </span><span class="si">{lisa_interval_steps}</span><span class="s2"> steps.&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.lisa_layers_attribute">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.lisa_layers_attribute">[docs]</a>
+    <span class="n">lisa_layers_attribute</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;model.model.layers&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;where the layer attribute stores, e.g. model.model.layers&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.use_customized_optim">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.use_customized_optim">[docs]</a>
+    <span class="n">use_customized_optim</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;whether to use customized optimizers.&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.customized_optim">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.customized_optim">[docs]</a>
+    <span class="n">customized_optim</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;sign_sgd&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;name of the customized optimizer.&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.customized_optim_args">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.customized_optim_args">[docs]</a>
+    <span class="n">customized_optim_args</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;optional arguments that are supplied.&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.optim_dummy_beta1">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_dummy_beta1">[docs]</a>
+    <span class="n">optim_dummy_beta1</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.9</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;A useless argument for dummy optimizer, just for tutorial&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.optim_dummy_beta2">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_dummy_beta2">[docs]</a>
+    <span class="n">optim_dummy_beta2</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.999</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;A useless argument for dummy optimizer, just for tutorial&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.optim_adam_beta1">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_adam_beta1">[docs]</a>
+    <span class="n">optim_adam_beta1</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.9</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Coefficient used for computing running averages of gradient&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.optim_adam_beta2">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_adam_beta2">[docs]</a>
+    <span class="n">optim_adam_beta2</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.999</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Coefficient used for computing running averages of squared gradient&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.optim_beta1">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_beta1">[docs]</a>
+    <span class="n">optim_beta1</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.9</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Coefficient used for computing running averages of gradient&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.optim_beta2">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_beta2">[docs]</a>
+    <span class="n">optim_beta2</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.999</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Coefficient used for computing running averages of squared gradient&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.optim_beta3">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_beta3">[docs]</a>
+    <span class="n">optim_beta3</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.9</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Coefficient used for computing running averages of gradient&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.optim_momentum">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_momentum">[docs]</a>
+    <span class="n">optim_momentum</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.999</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Coefficient used for the momentum term in optimizers like SGD with momentum&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="FinetunerArguments.optim_weight_decay">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_weight_decay">[docs]</a>
+    <span class="n">optim_weight_decay</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Weight decay (L2 penalty) added to the loss to prevent overfitting&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+</div>
+
+    
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="RewardModelTunerArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.RewardModelTunerArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">RewardModelTunerArguments</span><span class="p">(</span><span class="n">FinetunerArguments</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Arguments for reward modeling.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">pass</span></div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="EvaluatorArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">EvaluatorArguments</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Define a class EvaluatorArguments using the dataclass decorator. The class contains several optional</span>
+<span class="sd">    parameters that can be used to configure a evaluator.</span>
+
+<span class="sd">    local_rank : str</span>
+<span class="sd">        For distributed training: local_rank</span>
+
+<span class="sd">    random_shuffle : bool</span>
+
+<span class="sd">    use_wandb : bool</span>
+
+<span class="sd">    random_seed : int, default = 1</span>
+
+<span class="sd">    output_dir : str, default = &#39;./output_dir&#39;,</span>
+
+<span class="sd">    mixed_precision : str, choice from [&quot;bf16&quot;,&quot;fp16&quot;].</span>
+<span class="sd">        mixed precision mode, whether to use bf16 or fp16</span>
+
+<span class="sd">    deepspeed :</span>
+<span class="sd">        Enable deepspeed and pass the path to deepspeed json config file (e.g. ds_config.json) or an already</span>
+<span class="sd">        loaded json file as a dict</span>
+
+<span class="sd">    temperature : float</span>
+<span class="sd">        An argument of model.generate in huggingface to control the diversity of generation.</span>
+
+<span class="sd">    repetition_penalty : float</span>
+<span class="sd">        An argument of model.generate in huggingface to penalize repetitions.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+<div class="viewcode-block" id="EvaluatorArguments.local_rank">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.local_rank">[docs]</a>
+    <span class="n">local_rank</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=-</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;For distributed training: local_rank&quot;</span>
+                  <span class="p">}</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="EvaluatorArguments.random_shuffle">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.random_shuffle">[docs]</a>
+    <span class="n">random_shuffle</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;</span>
+                  <span class="p">}</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="EvaluatorArguments.use_wandb">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.use_wandb">[docs]</a>
+    <span class="n">use_wandb</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;When this flag is True, wandb will be enabled&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="EvaluatorArguments.random_seed">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.random_seed">[docs]</a>
+    <span class="n">random_seed</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;used to set random seed&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="EvaluatorArguments.output_dir">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.output_dir">[docs]</a>
+    <span class="n">output_dir</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;./output_dir&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Output path for the inferenced results&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="EvaluatorArguments.mixed_precision">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.mixed_precision">[docs]</a>
+    <span class="n">mixed_precision</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;bf16&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;mixed precision mode, whether to use bf16 or fp16&quot;</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;choices&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;bf16&quot;</span><span class="p">,</span> <span class="s2">&quot;fp16&quot;</span><span class="p">],</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="EvaluatorArguments.deepspeed">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.deepspeed">[docs]</a>
+    <span class="n">deepspeed</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;Enable deepspeed and pass the path to deepspeed json config file (e.g. ds_config.json) or an already&quot;</span>
+                <span class="s2">&quot; loaded json file as a dict&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="EvaluatorArguments.answer_type">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.answer_type">[docs]</a>
+    <span class="n">answer_type</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;text&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s1">&#39;Question type for answer extraction from the decoder output.&#39;</span>
+                <span class="s1">&#39; Supported types: </span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;   1) &quot;multiple_choice&quot;, e.g. A, B, C, D, ...</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;   2) &quot;binary_choice&quot;, e.g. yes, no, maybe</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;   3) &quot;math&quot;, e.g. 1.0, -3.52</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;   4) &quot;text&quot;, e.g. &quot;I think that it is okay&quot;</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;   5) Special treatment for several datasets</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;     - &quot;gsm8k&quot;</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;     - &quot;svamp&quot;</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;     - &quot;asdiv&quot;</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;     - &quot;addsub&quot;</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;     - &quot;singleeq&quot;</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;     - &quot;multiarith&quot;</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;     - &quot;aqua&quot;</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;     - &quot;csqa&quot;</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;     - &quot;strategyqa&quot;</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;     - &quot;pubmedqa&quot;</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;     - &quot;medmcqa&quot;</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="s1">&#39;     - &quot;usmle&quot;</span><span class="se">\n</span><span class="s1">&#39;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="EvaluatorArguments.prompt_structure">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.prompt_structure">[docs]</a>
+    <span class="n">prompt_structure</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;</span><span class="si">{input}</span><span class="s2">&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s1">&#39;Prompt structure to facilitate prompt engineering during&#39;</span>
+                <span class="s1">&#39; inference. The model will receive&#39;</span>
+                <span class="s1">&#39; `prompt_structure.format(input=input)` as its input.&#39;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="EvaluatorArguments.evaluate_block_size">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.evaluate_block_size">[docs]</a>
+    <span class="n">evaluate_block_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">512</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;the model will have at least block_size tokens for context when calculating the conditional likelihood of any one token&quot;</span>
+                <span class="s2">&quot; (provided there are block_size preceding tokens available to condition on)&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="EvaluatorArguments.metric">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.metric">[docs]</a>
+    <span class="n">metric</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;accuracy&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the metric the model will be evaluated on&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;choices&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;ppl&quot;</span><span class="p">,</span> <span class="s2">&quot;perplexity&quot;</span><span class="p">,</span> <span class="s2">&quot;acc&quot;</span><span class="p">,</span> <span class="s2">&quot;accuracy&quot;</span><span class="p">,</span> <span class="s2">&quot;nll&quot;</span><span class="p">,</span> <span class="s2">&quot;neg_log_likelihood&quot;</span><span class="p">],</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="EvaluatorArguments.inference_batch_size_per_device">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.inference_batch_size_per_device">[docs]</a>
+    <span class="n">inference_batch_size_per_device</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;every device will infer </span><span class="si">{inference_batch_size_per_device}</span><span class="s2">&quot;</span>
+                <span class="s2">&quot; samples in parallel. The inferred results will be concatenaed&quot;</span>
+                <span class="s2">&quot; with inputs and attach a reward.&quot;</span>
+            <span class="p">),</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="EvaluatorArguments.use_accelerator_for_evaluator">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.use_accelerator_for_evaluator">[docs]</a>
+    <span class="n">use_accelerator_for_evaluator</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to use Huggingface Accelerator instead of Deepspeed&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="EvaluatorArguments.temperature">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.temperature">[docs]</a>
+    <span class="n">temperature</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Temperature during inference.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="EvaluatorArguments.repetition_penalty">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.repetition_penalty">[docs]</a>
+    <span class="n">repetition_penalty</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Repetition_penalty during inference.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="EvaluatorArguments.max_new_tokens">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.max_new_tokens">[docs]</a>
+    <span class="n">max_new_tokens</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Maximum length during inference.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="InferencerArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">InferencerArguments</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Define a class InferencerArguments using the dataclass decorator. The class contains several optional</span>
+<span class="sd">    parameters that can be used to configure a inferencer.</span>
+
+<span class="sd">    local_rank : str</span>
+<span class="sd">        For distributed training: local_rank</span>
+<span class="sd">    random_seed : int, default = 1</span>
+<span class="sd">    inference_batch_size : int, default = 1</span>
+<span class="sd">    deepspeed :</span>
+<span class="sd">        Enable deepspeed and pass the path to deepspeed json config file (e.g. ds_config.json) or an already</span>
+<span class="sd">        loaded json file as a dict</span>
+<span class="sd">    mixed_precision : str, choice from [&quot;bf16&quot;,&quot;fp16&quot;].</span>
+<span class="sd">        mixed precision mode, whether to use bf16 or fp16</span>
+<span class="sd">    temperature : float</span>
+<span class="sd">        An argument of model.generate in huggingface to control the diversity of generation.</span>
+<span class="sd">    repetition_penalty : float</span>
+<span class="sd">        An argument of model.generate in huggingface to penalize repetitions.</span>
+<span class="sd">    use_beam_search : Optional[bool]</span>
+<span class="sd">        Whether to use beam search during inference, By default False.</span>
+<span class="sd">    num_output_sequences : Optional[int]</span>
+<span class="sd">        Number of output sequences to return for the given prompt, </span>
+<span class="sd">        currently only used in vllm inference, By default 8.</span>
+<span class="sd">    top_p : Optional[float]</span>
+<span class="sd">        top_p for sampling, By default 1.0.</span>
+<span class="sd">    top_k : Optional[int]</span>
+<span class="sd">        top_k for sampling, By default -1 (no top_k).</span>
+<span class="sd">    additional_stop_token_ids : Optional[List[int]]</span>
+<span class="sd">        the ids of the end of sentence tokens, By default [].</span>
+<span class="sd">    apply_chat_template : Optional[bool]</span>
+<span class="sd">        Whether to apply chat template, By default True.</span>
+<span class="sd">    save_results : Optional[bool]</span>
+<span class="sd">        Whether to save inference results, By default False.</span>
+<span class="sd">    results_path : Optional[str]</span>
+<span class="sd">        The **json file** path of inference results, By default None.</span>
+<span class="sd">    enable_decode_inference_result : Optional[bool]</span>
+<span class="sd">        Whether to detokenize the inference results. </span>
+
+<span class="sd">        NOTE: For iterative align pipelines, whether to detokenize depends on </span>
+<span class="sd">        the homogeneity of the policy model and the reward model </span>
+<span class="sd">        (i.e., if they have the same tokenizer).</span>
+<span class="sd">    use_vllm: bool, optional</span>
+<span class="sd">        Whether to use VLLM for inference, By default False.</span>
+<span class="sd">    vllm_tensor_parallel_size: int, optional</span>
+<span class="sd">        The tensor parallel size for VLLM inference.</span>
+<span class="sd">    vllm_gpu_memory_utilization: float, optional</span>
+<span class="sd">        The GPU memory utilization for VLLM inference. The proportion of GPU</span>
+<span class="sd">        memory (per GPU) to use for VLLM inference.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+<div class="viewcode-block" id="InferencerArguments.device">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.device">[docs]</a>
+    <span class="n">device</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;gpu&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;device of chatbot&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;choices&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;gpu&quot;</span><span class="p">,</span> <span class="s2">&quot;cpu&quot;</span><span class="p">],</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.local_rank">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.local_rank">[docs]</a>
+    <span class="n">local_rank</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=-</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;For distributed training: local_rank&quot;</span>
+                  <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.inference_batch_size">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.inference_batch_size">[docs]</a>
+    <span class="n">inference_batch_size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;batch size for inference&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.vllm_inference_batch_size">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.vllm_inference_batch_size">[docs]</a>
+    <span class="n">vllm_inference_batch_size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The batch size for VLLM inference.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.temperature">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.temperature">[docs]</a>
+    <span class="n">temperature</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Temperature during inference.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="InferencerArguments.repetition_penalty">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.repetition_penalty">[docs]</a>
+    <span class="n">repetition_penalty</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Repetition_penalty during inference.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="InferencerArguments.max_new_tokens">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.max_new_tokens">[docs]</a>
+    <span class="n">max_new_tokens</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Maximum length during inference.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="InferencerArguments.random_seed">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.random_seed">[docs]</a>
+    <span class="n">random_seed</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;used to set random seed&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.deepspeed">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.deepspeed">[docs]</a>
+    <span class="n">deepspeed</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;Enable deepspeed and pass the path to deepspeed json config file (e.g. ds_config.json) or an already&quot;</span>
+                <span class="s2">&quot; loaded json file as a dict&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.mixed_precision">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.mixed_precision">[docs]</a>
+    <span class="n">mixed_precision</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;bf16&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;mixed precision mode, whether to use bf16 or fp16&quot;</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;choices&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;bf16&quot;</span><span class="p">,</span> <span class="s2">&quot;fp16&quot;</span><span class="p">],</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.do_sample">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.do_sample">[docs]</a>
+    <span class="n">do_sample</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;whether turn on true random sampling during inference.&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.use_accelerator">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.use_accelerator">[docs]</a>
+    <span class="n">use_accelerator</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to use Huggingface Accelerator instead of Deepspeed&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.use_beam_search">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.use_beam_search">[docs]</a>
+    <span class="n">use_beam_search</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;whether to use beam search during inference.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.num_output_sequences">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.num_output_sequences">[docs]</a>
+    <span class="n">num_output_sequences</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">8</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+            <span class="s2">&quot;number of output sequences to return for the given prompt, &quot;</span>
+            <span class="s2">&quot;currently only used in vllm inference.&quot;</span>
+        <span class="p">)},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.top_p">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.top_p">[docs]</a>
+    <span class="n">top_p</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;top_p for sampling.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.top_k">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.top_k">[docs]</a>
+    <span class="n">top_k</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=-</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;top_k for sampling.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.additional_stop_token_ids">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.additional_stop_token_ids">[docs]</a>
+    <span class="n">additional_stop_token_ids</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default_factory</span><span class="o">=</span><span class="k">lambda</span><span class="p">:</span> <span class="p">[],</span> 
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the ids of the end of sentence tokens&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.apply_chat_template">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.apply_chat_template">[docs]</a>
+    <span class="n">apply_chat_template</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;whether to apply chat template&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.enable_decode_inference_result">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.enable_decode_inference_result">[docs]</a>
+    <span class="n">enable_decode_inference_result</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to decode the inference results.&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.tensor_parallel_size">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.tensor_parallel_size">[docs]</a>
+    <span class="n">tensor_parallel_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The tp size for distributed (multi-instance) inference.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.enable_distributed_inference">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.enable_distributed_inference">[docs]</a>
+    <span class="n">enable_distributed_inference</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to use multi-instance VLLM inference.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.distributed_inference_num_instances">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.distributed_inference_num_instances">[docs]</a>
+    <span class="n">distributed_inference_num_instances</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The number of instances for multi-instance VLLM inference.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+    
+    <span class="c1"># vllm inference args</span>
+<div class="viewcode-block" id="InferencerArguments.use_vllm">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.use_vllm">[docs]</a>
+    <span class="n">use_vllm</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to use VLLM for inference, By default False.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.vllm_tensor_parallel_size">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.vllm_tensor_parallel_size">[docs]</a>
+    <span class="n">vllm_tensor_parallel_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The tensor parallel size for VLLM inference.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.vllm_gpu_memory_utilization">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.vllm_gpu_memory_utilization">[docs]</a>
+    <span class="n">vllm_gpu_memory_utilization</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.95</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The GPU memory utilization for VLLM inference.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+    
+    <span class="c1"># Args for result saving</span>
+<div class="viewcode-block" id="InferencerArguments.save_results">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.save_results">[docs]</a>
+    <span class="n">save_results</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to save inference results.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="InferencerArguments.results_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.results_path">[docs]</a>
+    <span class="n">results_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The path of inference results.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+    
+<div class="viewcode-block" id="InferencerArguments.__post_init__">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.__post_init__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__post_init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">save_results</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">results_path</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Need to specify results_path when save_results is True.&quot;</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">results_path</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;.json&quot;</span><span class="p">):</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;The results_path must be a json file.&quot;</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">results_path</span><span class="p">)</span><span class="o">.</span><span class="n">parent</span><span class="o">.</span><span class="n">mkdir</span><span class="p">(</span><span class="n">parents</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="RaftAlignerArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">RaftAlignerArguments</span><span class="p">(</span><span class="n">TrainingArguments</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Define a class RaftAlignerArguments to configure raft aligner.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+<div class="viewcode-block" id="RaftAlignerArguments.output_reward_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.output_reward_path">[docs]</a>
+    <span class="n">output_reward_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;tmp/raft_aligner/&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The path of output rewards.&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="RaftAlignerArguments.output_min_length">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.output_min_length">[docs]</a>
+    <span class="n">output_min_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">64</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;minimum length of the output token sequence generated from&quot;</span>
+                <span class="s2">&quot; model given an input.&quot;</span>
+            <span class="p">),</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="RaftAlignerArguments.output_max_length">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.output_max_length">[docs]</a>
+    <span class="n">output_max_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">128</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;maximum length of the output token sequence generated from&quot;</span>
+                <span class="s2">&quot; model given an output.&quot;</span>
+            <span class="p">),</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="RaftAlignerArguments.num_raft_iteration">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.num_raft_iteration">[docs]</a>
+    <span class="n">num_raft_iteration</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">20</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;number of iterations of the raft aligner.&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="RaftAlignerArguments.raft_batch_size">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.raft_batch_size">[docs]</a>
+    <span class="n">raft_batch_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1024</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;only select </span><span class="si">{raft_batch_size}</span><span class="s2"> samples each time for STF training.&quot;</span>
+            <span class="p">)</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="RaftAlignerArguments.top_reward_percentage">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.top_reward_percentage">[docs]</a>
+    <span class="n">top_reward_percentage</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.2</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;only top </span><span class="si">{top_reward_percentage}</span><span class="s2"> samples in the raft batch,&quot;</span>
+                <span class="s2">&quot; (in terms of rewards), will be used for SFT the model.&quot;</span>
+            <span class="p">),</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="RaftAlignerArguments.inference_batch_size_per_device">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.inference_batch_size_per_device">[docs]</a>
+    <span class="n">inference_batch_size_per_device</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;every device will infer </span><span class="si">{inference_batch_size_per_device}</span><span class="s2">&quot;</span>
+                <span class="s2">&quot; samples in parallel. The inferred results will be concatenaed&quot;</span>
+                <span class="s2">&quot; with inputs and attach a reward.&quot;</span>
+            <span class="p">),</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="RaftAlignerArguments.collection_strategy">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.collection_strategy">[docs]</a>
+    <span class="n">collection_strategy</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;top&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+                <span class="s2">&quot;</span><span class="si">{collection_strategy}</span><span class="s2"> is either top or local&quot;</span>
+                <span class="s2">&quot; top means that we rank the samples globally regardless of the prompts&quot;</span>
+                <span class="s2">&quot; local means that we only rank the samples with the same prompt&quot;</span>
+            <span class="p">),</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="BenchmarkingArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.BenchmarkingArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">BenchmarkingArguments</span><span class="p">:</span>
+<div class="viewcode-block" id="BenchmarkingArguments.dataset_name">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.BenchmarkingArguments.dataset_name">[docs]</a>
+    <span class="n">dataset_name</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;benchmark dataset name provided by lmflow&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="BenchmarkingArguments.lm_evaluation_metric">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.BenchmarkingArguments.lm_evaluation_metric">[docs]</a>
+    <span class="n">lm_evaluation_metric</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;accuracy&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the metric the model will be evaluated on&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;choices&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;acc&quot;</span><span class="p">,</span> <span class="s2">&quot;acc_norm&quot;</span><span class="p">,</span> <span class="s2">&quot;bleu&quot;</span><span class="p">,</span> <span class="s2">&quot;chrf&quot;</span><span class="p">,</span> <span class="s2">&quot;em&quot;</span><span class="p">,</span> <span class="s2">&quot;f1&quot;</span><span class="p">,</span> <span class="s2">&quot;ppl&quot;</span><span class="p">,</span> \
+                        <span class="s2">&quot;ter&quot;</span><span class="p">,</span> <span class="s2">&quot;r@1&quot;</span><span class="p">,</span> <span class="s2">&quot;r@2&quot;</span><span class="p">,</span> <span class="s2">&quot;mrr&quot;</span><span class="p">,</span> <span class="s2">&quot;mc1&quot;</span><span class="p">,</span> <span class="s2">&quot;mc2&quot;</span><span class="p">,</span> <span class="s2">&quot;word_perplexity&quot;</span><span class="p">,</span> \
+                        <span class="s2">&quot;byte_perplexity&quot;</span><span class="p">,</span> <span class="s2">&quot;bits_per_byte&quot;</span><span class="p">],</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="DPOAlignerArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">DPOAlignerArguments</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    The arguments for the DPO training script.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+<div class="viewcode-block" id="DPOAlignerArguments.local_rank">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.local_rank">[docs]</a>
+    <span class="n">local_rank</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=-</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;For distributed training: local_rank&quot;</span>
+                  <span class="p">},</span>
+    <span class="p">)</span></div>
+
+    <span class="c1"># data parameters</span>
+<div class="viewcode-block" id="DPOAlignerArguments.beta">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.beta">[docs]</a>
+    <span class="n">beta</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the beta parameter for DPO loss&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+    <span class="c1"># # training parameters</span>
+<div class="viewcode-block" id="DPOAlignerArguments.learning_rate">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.learning_rate">[docs]</a>
+    <span class="n">learning_rate</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">5e-4</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;optimizer learning rate&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.lr_scheduler_type">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.lr_scheduler_type">[docs]</a>
+    <span class="n">lr_scheduler_type</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;cosine&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the lr scheduler type&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.warmup_steps">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.warmup_steps">[docs]</a>
+    <span class="n">warmup_steps</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the number of warmup steps&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.weight_decay">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.weight_decay">[docs]</a>
+    <span class="n">weight_decay</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mf">0.05</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the weight decay&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.optimizer_type">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.optimizer_type">[docs]</a>
+    <span class="n">optimizer_type</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;paged_adamw_32bit&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the optimizer type&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="DPOAlignerArguments.per_device_train_batch_size">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.per_device_train_batch_size">[docs]</a>
+    <span class="n">per_device_train_batch_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;train batch size per device&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.per_device_eval_batch_size">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.per_device_eval_batch_size">[docs]</a>
+    <span class="n">per_device_eval_batch_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;eval batch size per device&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.gradient_accumulation_steps">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.gradient_accumulation_steps">[docs]</a>
+    <span class="n">gradient_accumulation_steps</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the number of gradient accumulation steps&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.gradient_checkpointing">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.gradient_checkpointing">[docs]</a>
+    <span class="n">gradient_checkpointing</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;whether to use gradient checkpointing&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="DPOAlignerArguments.gradient_checkpointing_use_reentrant">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.gradient_checkpointing_use_reentrant">[docs]</a>
+    <span class="n">gradient_checkpointing_use_reentrant</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;whether to use reentrant for gradient checkpointing&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.max_prompt_length">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.max_prompt_length">[docs]</a>
+    <span class="n">max_prompt_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">512</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the maximum prompt length&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.max_length">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.max_length">[docs]</a>
+    <span class="n">max_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1024</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the maximum sequence length&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.max_steps">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.max_steps">[docs]</a>
+    <span class="n">max_steps</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;max number of training steps&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.logging_steps">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.logging_steps">[docs]</a>
+    <span class="n">logging_steps</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the logging frequency&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.save_steps">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.save_steps">[docs]</a>
+    <span class="n">save_steps</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the saving frequency&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.eval_steps">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.eval_steps">[docs]</a>
+    <span class="n">eval_steps</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the evaluation frequency&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.output_dir">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.output_dir">[docs]</a>
+    <span class="n">output_dir</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;./results&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the output directory&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.log_freq">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.log_freq">[docs]</a>
+    <span class="n">log_freq</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the logging frequency&quot;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.sanity_check">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.sanity_check">[docs]</a>
+    <span class="n">sanity_check</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;only train on 1000 samples&quot;</span>
+        <span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.report_to">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.report_to">[docs]</a>
+    <span class="n">report_to</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;wandb&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+            <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s1">&#39;The list of integrations to report the results and logs to. Supported platforms are `&quot;azure_ml&quot;`,&#39;</span>
+                    <span class="s1">&#39;`&quot;comet_ml&quot;`, `&quot;mlflow&quot;`, `&quot;neptune&quot;`, `&quot;tensorboard&quot;`,`&quot;clearml&quot;` and `&quot;wandb&quot;`. &#39;</span>
+                    <span class="s1">&#39;Use `&quot;all&quot;` to report to all integrations installed, `&quot;none&quot;` for no integrations.&#39;</span>
+        <span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.seed">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.seed">[docs]</a>
+    <span class="n">seed</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Random seed that will be set at the beginning of training.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.run_name">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.run_name">[docs]</a>
+    <span class="n">run_name</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;dpo&quot;</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The name of the run.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="DPOAlignerArguments.eval_dataset_path">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.eval_dataset_path">[docs]</a>
+    <span class="n">eval_dataset_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The path of the eval dataset.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="DPOv2AlignerArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">DPOv2AlignerArguments</span><span class="p">(</span><span class="n">FinetunerArguments</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    The arguments for the DPOv2 training script.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="c1"># general args</span>
+<div class="viewcode-block" id="DPOv2AlignerArguments.random_seed">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.random_seed">[docs]</a>
+    <span class="n">random_seed</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="mi">42</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the random seed&quot;</span><span class="p">})</span></div>
+
+<div class="viewcode-block" id="DPOv2AlignerArguments.accelerate_config_file">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.accelerate_config_file">[docs]</a>
+    <span class="n">accelerate_config_file</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> 
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;file path for accelerate config file, only used in memory safe dpov2 align.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+    <span class="c1"># pair sampling args</span>
+<div class="viewcode-block" id="DPOv2AlignerArguments.margin_scale">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.margin_scale">[docs]</a>
+    <span class="n">margin_scale</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the margin scale&quot;</span><span class="p">})</span></div>
+
+<div class="viewcode-block" id="DPOv2AlignerArguments.sampling_paired_method">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.sampling_paired_method">[docs]</a>
+    <span class="n">sampling_paired_method</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="s2">&quot;max_random&quot;</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the choose type&quot;</span><span class="p">})</span></div>
+
+<div class="viewcode-block" id="DPOv2AlignerArguments.length_penalty">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.length_penalty">[docs]</a>
+    <span class="n">length_penalty</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the length penalty&quot;</span><span class="p">})</span></div>
+
+    <span class="c1"># data collator args</span>
+<div class="viewcode-block" id="DPOv2AlignerArguments.max_length">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.max_length">[docs]</a>
+    <span class="n">max_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="mi">2048</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the maximum sequence length, prompt + output&quot;</span><span class="p">})</span></div>
+
+<div class="viewcode-block" id="DPOv2AlignerArguments.max_prompt_length">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.max_prompt_length">[docs]</a>
+    <span class="n">max_prompt_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the maximum prompt length&quot;</span><span class="p">})</span></div>
+
+<div class="viewcode-block" id="DPOv2AlignerArguments.mask_prompt">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.mask_prompt">[docs]</a>
+    <span class="n">mask_prompt</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;mask prompt&quot;</span><span class="p">})</span></div>
+
+    <span class="c1"># dpov2 aligner args</span>
+<div class="viewcode-block" id="DPOv2AlignerArguments.beta">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.beta">[docs]</a>
+    <span class="n">beta</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="mf">0.1</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the beta parameter for DPO loss&quot;</span><span class="p">})</span></div>
+
+<div class="viewcode-block" id="DPOv2AlignerArguments.loss_type">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.loss_type">[docs]</a>
+    <span class="n">loss_type</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="s2">&quot;sigmoid&quot;</span><span class="p">,</span> <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;the loss type&quot;</span><span class="p">})</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="IterativeAlignerArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.IterativeAlignerArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">IterativeAlignerArguments</span><span class="p">(</span><span class="n">InferencerArguments</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Arguments for iterative aligners.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+<div class="viewcode-block" id="IterativeAlignerArguments.dataset_path_list">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.IterativeAlignerArguments.dataset_path_list">[docs]</a>
+    <span class="n">dataset_path_list</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default_factory</span><span class="o">=</span><span class="nb">list</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The list of dataset paths for iterative aligners.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="IterativeAlignerArguments.initial_iter_idx">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.IterativeAlignerArguments.initial_iter_idx">[docs]</a>
+    <span class="n">initial_iter_idx</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The initial iteration index, 0 refers to the first dataset in dataset_path_list.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+</div>
+
+        
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="IterativeDPOAlignerArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">IterativeDPOAlignerArguments</span><span class="p">(</span><span class="n">IterativeAlignerArguments</span><span class="p">,</span> <span class="n">DPOv2AlignerArguments</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Arguments for iterative DPO aligners.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+<div class="viewcode-block" id="IterativeDPOAlignerArguments.output_dir">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments.output_dir">[docs]</a>
+    <span class="n">output_dir</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="s2">&quot;./runs&quot;</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Output path for the inferenced results&quot;</span><span class="p">},</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="IterativeDPOAlignerArguments.reward_model_inference_batch_size">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments.reward_model_inference_batch_size">[docs]</a>
+    <span class="n">reward_model_inference_batch_size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The batch size for reward model inference.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="IterativeDPOAlignerArguments.reward_model_inference_block_size">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments.reward_model_inference_block_size">[docs]</a>
+    <span class="n">reward_model_inference_block_size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">2048</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;The block size for reward model inference.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="IterativeDPOAlignerArguments.do_response_generation">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments.do_response_generation">[docs]</a>
+    <span class="n">do_response_generation</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to generate responses using the model.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="IterativeDPOAlignerArguments.do_scoring">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments.do_scoring">[docs]</a>
+    <span class="n">do_scoring</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to score the responses using the reward model.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+
+<div class="viewcode-block" id="IterativeDPOAlignerArguments.do_dpo_align">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments.do_dpo_align">[docs]</a>
+    <span class="n">do_dpo_align</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">metadata</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="s2">&quot;Whether to perform DPO alignment.&quot;</span><span class="p">}</span>
+    <span class="p">)</span></div>
+</div>
+
+                    
+
+<div class="viewcode-block" id="PIPELINE_ARGUMENT_MAPPING">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.PIPELINE_ARGUMENT_MAPPING">[docs]</a>
+<span class="n">PIPELINE_ARGUMENT_MAPPING</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;finetuner&quot;</span><span class="p">:</span> <span class="n">FinetunerArguments</span><span class="p">,</span>
+    <span class="s2">&quot;evaluator&quot;</span><span class="p">:</span> <span class="n">EvaluatorArguments</span><span class="p">,</span>
+    <span class="s2">&quot;inferencer&quot;</span><span class="p">:</span> <span class="n">InferencerArguments</span><span class="p">,</span>
+    <span class="s2">&quot;vllm_inferencer&quot;</span><span class="p">:</span> <span class="n">InferencerArguments</span><span class="p">,</span>
+    <span class="s2">&quot;rm_inferencer&quot;</span><span class="p">:</span> <span class="n">InferencerArguments</span><span class="p">,</span>
+    <span class="s2">&quot;raft_aligner&quot;</span><span class="p">:</span> <span class="n">RaftAlignerArguments</span><span class="p">,</span>
+    <span class="s2">&quot;dpo_aligner&quot;</span><span class="p">:</span> <span class="n">DPOAlignerArguments</span><span class="p">,</span>
+    <span class="s2">&quot;rm_tuner&quot;</span><span class="p">:</span> <span class="n">RewardModelTunerArguments</span><span class="p">,</span>
+    <span class="s2">&quot;dpov2_aligner&quot;</span><span class="p">:</span> <span class="n">DPOv2AlignerArguments</span><span class="p">,</span>
+    <span class="s2">&quot;iterative_dpo_aligner&quot;</span><span class="p">:</span> <span class="n">IterativeDPOAlignerArguments</span><span class="p">,</span>
+<span class="p">}</span></div>
+
+
+
+<div class="viewcode-block" id="AutoArguments">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.AutoArguments">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">AutoArguments</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Automatically choose arguments from FinetunerArguments or EvaluatorArguments.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+<div class="viewcode-block" id="AutoArguments.get_pipeline_args_class">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.AutoArguments.get_pipeline_args_class">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_pipeline_args_class</span><span class="p">(</span><span class="n">pipeline_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">PIPELINE_ARGUMENT_MAPPING</span><span class="p">[</span><span class="n">pipeline_name</span><span class="p">]</span></div>
+</div>
+
+
+
+<div class="viewcode-block" id="split_args">
+<a class="viewcode-back" href="../../autoapi/lmflow/args/index.html#lmflow.args.split_args">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">split_args</span><span class="p">(</span><span class="n">args</span><span class="p">):</span>
+    <span class="k">return</span> <span class="p">[</span><span class="n">elem</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span> <span class="k">for</span> <span class="n">elem</span> <span class="ow">in</span> <span class="n">args</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;,&quot;</span><span class="p">)]</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">args</span><span class="p">,</span> <span class="nb">str</span><span class="p">)</span> <span class="k">else</span> <span class="n">args</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/datasets/dataset.html b/_modules/lmflow/datasets/dataset.html
new file mode 100644
index 000000000..7fd9d1403
--- /dev/null
+++ b/_modules/lmflow/datasets/dataset.html
@@ -0,0 +1,1127 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.datasets.dataset &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/datasets/dataset';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.datasets.dataset</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.datasets.dataset</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;This Python code defines a class Dataset with methods for initializing, loading,</span>
+<span class="sd">and manipulating datasets from different backends such as Hugging Face and JSON.</span>
+<span class="sd"> </span>
+<span class="sd">The `Dataset` class includes methods for loading datasets from a dictionary and a Hugging</span>
+<span class="sd">Face dataset, mapping datasets, and retrieving the backend dataset and arguments.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+
+
+<span class="c1"># Importing necessary libraries and modules</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">copy</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">json</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">pathlib</span><span class="w"> </span><span class="kn">import</span> <span class="n">Path</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">cmath</span><span class="w"> </span><span class="kn">import</span> <span class="n">e</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">pathlib</span><span class="w"> </span><span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optional</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">datasets</span><span class="w"> </span><span class="kn">import</span> <span class="n">load_dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">datasets</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span> <span class="k">as</span> <span class="n">HFDataset</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="n">DatasetArguments</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">DATASET_DESCRIPTION_MAP</span><span class="p">,</span>
+    <span class="n">TEXT_ONLY_DATASET_DESCRIPTION</span><span class="p">,</span>
+    <span class="n">TEXT2TEXT_DATASET_DESCRIPTION</span><span class="p">,</span>
+    <span class="n">FLOAT_ONLY_DATASET_DESCRIPTION</span><span class="p">,</span>
+    <span class="n">INSTANCE_FIELDS_MAP</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.versioning</span><span class="w"> </span><span class="kn">import</span> <span class="n">is_multimodal_available</span>
+
+<span class="k">if</span> <span class="n">is_multimodal_available</span><span class="p">():</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">.multi_modal_dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">CustomMultiModalDataset</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="DATASET_TYPES">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.DATASET_TYPES">[docs]</a>
+<span class="n">DATASET_TYPES</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;text_only&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;text2text&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;float_only&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;image_text&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;conversation&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;paired_conversation&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;paired_text_to_text&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;text_to_textlist&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;text_to_scored_textlist&quot;</span>
+<span class="p">]</span></div>
+
+
+<div class="viewcode-block" id="KEY_TYPE">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.KEY_TYPE">[docs]</a>
+<span class="n">KEY_TYPE</span> <span class="o">=</span> <span class="s2">&quot;type&quot;</span></div>
+
+<div class="viewcode-block" id="KEY_INSTANCES">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.KEY_INSTANCES">[docs]</a>
+<span class="n">KEY_INSTANCES</span> <span class="o">=</span> <span class="s2">&quot;instances&quot;</span></div>
+
+<div class="viewcode-block" id="KEY_SCORE">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.KEY_SCORE">[docs]</a>
+<span class="n">KEY_SCORE</span> <span class="o">=</span> <span class="s2">&quot;score&quot;</span></div>
+
+
+<div class="viewcode-block" id="Dataset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Dataset</span><span class="p">:</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Initializes the Dataset object with the given parameters.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+<span class="sd">    data_args : DatasetArguments object.</span>
+<span class="sd">        Contains the arguments required to load the dataset.</span>
+
+<span class="sd">    backend : str,  default=&quot;huggingface&quot;</span>
+<span class="sd">        A string representing the dataset backend. Defaults to &quot;huggingface&quot;.</span>
+<span class="sd">    </span>
+<span class="sd">    args : Optional.</span>
+<span class="sd">        Positional arguments.</span>
+<span class="sd">    </span>
+<span class="sd">    kwargs : Optional.</span>
+<span class="sd">        Keyword arguments.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">backend</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s2">&quot;huggingface&quot;</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<div class="viewcode-block" id="Dataset.data_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.data_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span> <span class="o">=</span> <span class="n">data_args</span></div>
+
+<div class="viewcode-block" id="Dataset.backend">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.backend">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">=</span> <span class="n">backend</span></div>
+
+<div class="viewcode-block" id="Dataset.backend_dataset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.backend_dataset">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="Dataset.type">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.type">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">type</span> <span class="o">=</span> <span class="kc">None</span>        <span class="c1"># Original type of the dataset</span></div>
+
+<div class="viewcode-block" id="Dataset.dataset_path">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.dataset_path">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">dataset_path</span> <span class="o">=</span> <span class="n">data_args</span><span class="o">.</span><span class="n">dataset_path</span></div>
+
+
+        <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">dataset_path</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span>
+
+        <span class="k">if</span> <span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;huggingface&quot;</span><span class="p">:</span>
+            <span class="n">data_files</span> <span class="o">=</span> <span class="p">[</span>
+                <span class="n">x</span><span class="o">.</span><span class="n">absolute</span><span class="p">()</span><span class="o">.</span><span class="n">as_posix</span><span class="p">()</span>
+                 <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">dataset_path</span><span class="p">)</span><span class="o">.</span><span class="n">glob</span><span class="p">(</span><span class="s2">&quot;*.json&quot;</span><span class="p">)</span>
+            <span class="p">]</span>
+
+            <span class="c1"># Iterate through all the files and ensure they have the same data type</span>
+            <span class="k">for</span> <span class="n">single_file</span> <span class="ow">in</span> <span class="n">data_files</span><span class="p">:</span>
+                <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">single_file</span><span class="p">)</span> <span class="k">as</span> <span class="n">fin</span><span class="p">:</span>
+                    <span class="n">json_data</span> <span class="o">=</span> <span class="n">json</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">fin</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">KEY_TYPE</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">json_data</span><span class="o">.</span><span class="n">keys</span><span class="p">():</span>
+                        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                            <span class="sa">f</span><span class="s1">&#39;&quot;</span><span class="si">{</span><span class="n">KEY_TYPE</span><span class="si">}</span><span class="s1">&quot; field must be specified for data, e.g.&#39;</span>
+                            <span class="s1">&#39;{</span><span class="se">\n</span><span class="s1">&#39;</span>
+                            <span class="sa">f</span><span class="s1">&#39;   &quot;</span><span class="si">{</span><span class="n">KEY_TYPE</span><span class="si">}</span><span class="s1">: &quot;text_only&quot;,</span><span class="se">\n</span><span class="s1">&#39;</span>
+                            <span class="sa">f</span><span class="s1">&#39;   &quot;</span><span class="si">{</span><span class="n">KEY_INSTANCES</span><span class="si">}</span><span class="s1">&quot;: [</span><span class="se">\n</span><span class="s1">&#39;</span>
+                            <span class="s1">&#39;       { &quot;text&quot;: &quot;Sentence 1: This is a sentence.&quot; }</span><span class="se">\n</span><span class="s1">&#39;</span>
+                            <span class="s1">&#39;       { &quot;text&quot;: &quot;Sentence 2: This is another sentence.&quot; }</span><span class="se">\n</span><span class="s1">&#39;</span>
+                            <span class="sa">f</span><span class="s1">&#39;   ]</span><span class="se">\n</span><span class="s1">&#39;</span>
+                            <span class="s1">&#39;}&#39;</span>
+                        <span class="p">)</span>
+                    <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">type</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">type</span> <span class="o">=</span> <span class="n">json_data</span><span class="p">[</span><span class="n">KEY_TYPE</span><span class="p">]</span>
+                    <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">type</span> <span class="o">!=</span> <span class="n">json_data</span><span class="p">[</span><span class="n">KEY_TYPE</span><span class="p">]:</span>
+                        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                            <span class="s1">&#39;All task files must have same data types. Previous&#39;</span>
+                            <span class="sa">f</span><span class="s1">&#39; files have type &quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="s1">&quot;, but in file&#39;</span>
+                            <span class="sa">f</span><span class="s1">&#39; </span><span class="si">{</span><span class="n">single_file</span><span class="si">}</span><span class="s1">, it has type &quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="s1">&quot;.&#39;</span>
+                        <span class="p">)</span>
+
+            <span class="c1"># Load the dataset using the HuggingFace dataset library</span>
+            <span class="n">extensions</span> <span class="o">=</span> <span class="s2">&quot;json&quot;</span>
+            <span class="n">raw_dataset</span> <span class="o">=</span> <span class="n">load_dataset</span><span class="p">(</span>
+                <span class="n">extensions</span><span class="p">,</span>
+                <span class="n">data_files</span><span class="o">=</span><span class="n">data_files</span><span class="p">,</span>
+                <span class="n">field</span><span class="o">=</span><span class="n">KEY_INSTANCES</span><span class="p">,</span>
+                <span class="n">split</span><span class="o">=</span><span class="s2">&quot;train&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span> <span class="o">=</span> <span class="n">raw_dataset</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_check_data_format</span><span class="p">()</span>
+        <span class="k">elif</span> <span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;json&quot;</span><span class="p">:</span>
+            <span class="c1"># TODO (@Jiachun)</span>
+            <span class="k">pass</span>
+        <span class="k">elif</span> <span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;custom_multi_modal&quot;</span><span class="p">:</span>
+            <span class="c1"># FIXME refactor the backend name</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">is_multimodal_available</span><span class="p">():</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="s1">&#39;Multimodal not available. Please install via `pip install -e &quot;.[multimodal]&quot;`&#39;</span>
+                <span class="p">)</span>
+            <span class="n">raw_dataset</span> <span class="o">=</span> <span class="n">CustomMultiModalDataset</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">dataset_path</span><span class="p">,</span> <span class="n">data_args</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span> <span class="o">=</span> <span class="n">raw_dataset</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;Unsupported dataset backend &quot;</span><span class="si">{</span><span class="n">backend</span><span class="si">}</span><span class="s1">&quot;&#39;</span><span class="p">)</span>
+
+<div class="viewcode-block" id="Dataset.__len__">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.__len__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="Dataset._check_data_format">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset._check_data_format">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_check_data_format</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Checks if data type and data structure matches</span>
+
+<span class="sd">        Raise messages with hints if not matched.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">data_dict</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">to_dict</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">KEY_TYPE</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">data_dict</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s1">&#39;&quot;</span><span class="si">{</span><span class="n">KEY_TYPE</span><span class="si">}</span><span class="s1">&quot; must be provided to initialize a dataset,&#39;</span>
+                <span class="sa">f</span><span class="s1">&#39; e.g.</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="sa">f</span><span class="s1">&#39;    </span><span class="si">{</span><span class="n">TEXT_ONLY_DATASET_DESCRIPTION</span><span class="si">}</span><span class="s1">&#39;</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">KEY_INSTANCES</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">data_dict</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s1">&#39;&quot;</span><span class="si">{</span><span class="n">KEY_INSTANCES</span><span class="si">}</span><span class="s1">&quot; must be provided to initialize a&#39;</span>
+                <span class="sa">f</span><span class="s1">&#39; dataset, e.g.</span><span class="se">\n</span><span class="s1">&#39;</span>
+                <span class="sa">f</span><span class="s1">&#39;    </span><span class="si">{</span><span class="n">TEXT_ONLY_DATASET_DESCRIPTION</span><span class="si">}</span><span class="s1">&#39;</span>
+            <span class="p">)</span>
+
+        <span class="n">data_type</span> <span class="o">=</span> <span class="n">data_dict</span><span class="p">[</span><span class="n">KEY_TYPE</span><span class="p">]</span>
+        <span class="n">fields</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()</span><span class="o">.</span><span class="n">features</span>
+        <span class="n">correct_fields</span> <span class="o">=</span> <span class="n">INSTANCE_FIELDS_MAP</span><span class="p">[</span><span class="n">data_type</span><span class="p">]</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">set</span><span class="p">(</span><span class="n">correct_fields</span><span class="p">)</span><span class="o">.</span><span class="n">issubset</span><span class="p">(</span><span class="nb">set</span><span class="p">(</span><span class="n">fields</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s1">&#39;data instance fields incorrect&#39;</span>
+                <span class="sa">f</span><span class="s1">&#39; </span><span class="si">{</span><span class="nb">list</span><span class="p">(</span><span class="n">correct_fields</span><span class="p">)</span><span class="si">}</span><span class="s1"> are required.&#39;</span>
+            <span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="Dataset.from_dict">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.from_dict">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">from_dict</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dict_obj</span><span class="p">:</span> <span class="nb">dict</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Create a Dataset object from a dictionary.</span>
+
+<span class="sd">        Return a Dataset given a dict with format:</span>
+<span class="sd">            {</span>
+<span class="sd">                &quot;type&quot;: TYPE,</span>
+<span class="sd">                &quot;instances&quot;: [</span>
+<span class="sd">                    {</span>
+<span class="sd">                        &quot;key_1&quot;: VALUE_1.1,</span>
+<span class="sd">                        &quot;key_2&quot;: VALUE_1.2,</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    },</span>
+<span class="sd">                    {</span>
+<span class="sd">                        &quot;key_1&quot;: VALUE_2.1,</span>
+<span class="sd">                        &quot;key_2&quot;: VALUE_2.2,</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    },</span>
+<span class="sd">                    ...</span>
+<span class="sd">                ]</span>
+<span class="sd">            }</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        -----------</span>
+
+<span class="sd">        dict_obj : dict.</span>
+<span class="sd">            A dictionary containing the dataset information.</span>
+<span class="sd">        </span>
+<span class="sd">        args : Optional.</span>
+<span class="sd">            Positional arguments.</span>
+<span class="sd">        </span>
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        ---------</span>
+
+<span class="sd">        self : Dataset object.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;huggingface&quot;</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">KEY_TYPE</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">dict_obj</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s1">&#39;&quot;</span><span class="si">{</span><span class="n">KEY_TYPE</span><span class="si">}</span><span class="s1">&quot; must be provided to initialize a dataset,&#39;</span>
+                    <span class="sa">f</span><span class="s1">&#39; e.g.</span><span class="se">\n</span><span class="s1">&#39;</span>
+                    <span class="sa">f</span><span class="s1">&#39;    </span><span class="si">{</span><span class="n">TEXT_ONLY_DATASET_DESCRIPTION</span><span class="si">}</span><span class="s1">&#39;</span>
+                <span class="p">)</span>
+            <span class="k">if</span> <span class="n">KEY_INSTANCES</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">dict_obj</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s1">&#39;&quot;</span><span class="si">{</span><span class="n">KEY_INSTANCES</span><span class="si">}</span><span class="s1">&quot; must be provided to initialize a&#39;</span>
+                    <span class="sa">f</span><span class="s1">&#39; dataset, e.g.</span><span class="se">\n</span><span class="s1">&#39;</span>
+                    <span class="sa">f</span><span class="s1">&#39;    </span><span class="si">{</span><span class="n">TEXT_ONLY_DATASET_DESCRIPTION</span><span class="si">}</span><span class="s1">&#39;</span>
+                <span class="p">)</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">type</span> <span class="o">=</span> <span class="n">dict_obj</span><span class="p">[</span><span class="n">KEY_TYPE</span><span class="p">]</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">type</span> <span class="ow">in</span> <span class="n">INSTANCE_FIELDS_MAP</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;type &quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="s1">&quot; is not supported&#39;</span><span class="p">)</span>
+
+            <span class="n">correct_fields</span> <span class="o">=</span> <span class="n">INSTANCE_FIELDS_MAP</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">type</span><span class="p">]</span>
+
+            <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">instance</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">dict_obj</span><span class="p">[</span><span class="n">KEY_INSTANCES</span><span class="p">]):</span>
+                <span class="n">fields</span> <span class="o">=</span> <span class="n">instance</span><span class="o">.</span><span class="n">keys</span><span class="p">()</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="nb">set</span><span class="p">(</span><span class="n">correct_fields</span><span class="p">)</span><span class="o">.</span><span class="n">issubset</span><span class="p">(</span><span class="nb">set</span><span class="p">(</span><span class="n">fields</span><span class="p">)):</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s1">&#39;data instance fields incorrect&#39;</span>
+                        <span class="sa">f</span><span class="s1">&#39; </span><span class="si">{</span><span class="nb">list</span><span class="p">(</span><span class="n">correct_fields</span><span class="p">)</span><span class="si">}</span><span class="s1"> are required.&#39;</span>
+                    <span class="p">)</span>
+
+            <span class="k">try</span><span class="p">:</span>
+                <span class="n">hf_dict</span> <span class="o">=</span> <span class="p">{}</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">dict_obj</span><span class="p">[</span><span class="n">KEY_INSTANCES</span><span class="p">])</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">dict_obj</span><span class="p">[</span><span class="n">KEY_INSTANCES</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">keys</span><span class="p">():</span>
+                        <span class="n">hf_dict</span><span class="p">[</span><span class="n">key</span><span class="p">]</span> <span class="o">=</span> <span class="p">[</span>
+                            <span class="n">instance</span><span class="p">[</span><span class="n">key</span><span class="p">]</span> <span class="k">for</span> <span class="n">instance</span> <span class="ow">in</span> <span class="n">dict_obj</span><span class="p">[</span><span class="n">KEY_INSTANCES</span><span class="p">]</span>
+                        <span class="p">]</span>
+
+                <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span> <span class="o">=</span> <span class="n">HFDataset</span><span class="o">.</span><span class="n">from_dict</span><span class="p">(</span><span class="n">hf_dict</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+            <span class="k">except</span> <span class="ne">AttributeError</span> <span class="k">as</span> <span class="n">ex</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;Error occurs: </span><span class="si">{</span><span class="n">ex</span><span class="si">}</span><span class="s2">. Failed to convert dict to&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot; </span><span class="se">\&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="se">\&quot;</span><span class="s2"> dataset,&quot;</span> <span class="sa">f</span><span class="s2">&quot; the standard format is as&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot; follows:</span><span class="se">\n</span><span class="s2">&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;    </span><span class="si">{</span><span class="n">DATASET_DESCRIPTION_MAP</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">type</span><span class="p">]</span><span class="si">}</span><span class="s2">&quot;</span>
+                <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_check_data_format</span><span class="p">()</span>
+
+            <span class="k">return</span> <span class="bp">self</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;dict&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span> <span class="o">=</span> <span class="n">dict_obj</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">type</span> <span class="o">=</span> <span class="n">dict_obj</span><span class="p">[</span><span class="n">KEY_TYPE</span><span class="p">]</span>
+            <span class="k">return</span> <span class="bp">self</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s1">&#39;Currently .from_dict is not supported for backend &quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">backend</span><span class="si">}</span><span class="s1">&quot;&#39;</span>
+            <span class="p">)</span></div>
+
+
+
+    <span class="nd">@classmethod</span>
+<div class="viewcode-block" id="Dataset.create_from_dict">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.create_from_dict">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">create_from_dict</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="n">dict_obj</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Returns</span>
+<span class="sd">        --------</span>
+
+<span class="sd">        Returns a Dataset object given a dict.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">empty_data_args</span> <span class="o">=</span> <span class="n">DatasetArguments</span><span class="p">(</span><span class="n">dataset_path</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
+        <span class="n">dataset</span> <span class="o">=</span> <span class="n">Dataset</span><span class="p">(</span><span class="n">empty_data_args</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">dataset</span><span class="o">.</span><span class="n">from_dict</span><span class="p">(</span><span class="n">dict_obj</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="Dataset.to_dict">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.to_dict">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">to_dict</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Returns</span>
+<span class="sd">        ---------</span>
+
+<span class="sd">        Return a dict represents the dataset:</span>
+<span class="sd">            {</span>
+<span class="sd">                &quot;type&quot;: TYPE,</span>
+<span class="sd">                &quot;instances&quot;: [</span>
+<span class="sd">                    {</span>
+<span class="sd">                        &quot;key_1&quot;: VALUE_1.1,</span>
+<span class="sd">                        &quot;key_2&quot;: VALUE_1.2,</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    },</span>
+<span class="sd">                    {</span>
+<span class="sd">                        &quot;key_1&quot;: VALUE_2.1,</span>
+<span class="sd">                        &quot;key_2&quot;: VALUE_2.2,</span>
+<span class="sd">                        ...</span>
+<span class="sd">                    },</span>
+<span class="sd">                    ...</span>
+<span class="sd">                ]</span>
+<span class="sd">            }</span>
+
+<span class="sd">        A python dict object represents the content of this dataset.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;huggingface&quot;</span><span class="p">:</span>
+            <span class="n">dict_obj</span> <span class="o">=</span> <span class="p">{}</span>
+            <span class="n">dict_obj</span><span class="p">[</span><span class="n">KEY_TYPE</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span>
+            <span class="n">hf_dict</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="o">.</span><span class="n">to_dict</span><span class="p">()</span>
+            <span class="n">dict_obj</span><span class="p">[</span><span class="n">KEY_INSTANCES</span><span class="p">]</span> <span class="o">=</span> <span class="p">[]</span>
+
+            <span class="n">first_key</span> <span class="o">=</span> <span class="kc">None</span>
+            <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">hf_dict</span><span class="o">.</span><span class="n">keys</span><span class="p">():</span>
+                <span class="n">first_key</span> <span class="o">=</span> <span class="n">key</span>
+                <span class="k">break</span>
+
+            <span class="k">if</span> <span class="n">first_key</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">num_instances</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">hf_dict</span><span class="p">[</span><span class="n">first_key</span><span class="p">])</span>
+                <span class="n">dict_obj</span><span class="p">[</span><span class="n">KEY_INSTANCES</span><span class="p">]</span> <span class="o">=</span> <span class="p">[</span>
+                    <span class="p">{</span>
+                        <span class="n">key</span><span class="p">:</span> <span class="n">hf_dict</span><span class="p">[</span><span class="n">key</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">hf_dict</span><span class="o">.</span><span class="n">keys</span><span class="p">()</span>
+                    <span class="p">}</span>
+                    <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_instances</span><span class="p">)</span>
+                <span class="p">]</span>
+
+            <span class="k">return</span> <span class="n">dict_obj</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;dict&quot;</span><span class="p">:</span>
+            <span class="n">dict_obj</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span>
+            <span class="k">return</span> <span class="n">dict_obj</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s1">&#39;Current .to_dict is not supported for backend &quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">backend</span><span class="si">}</span><span class="s1">&quot;&#39;</span>
+            <span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="Dataset.to_list">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.to_list">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">to_list</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Returns a list of instances.&quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;huggingface&quot;</span><span class="p">:</span>
+            <span class="n">instance_list</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="o">.</span><span class="fm">__getitem__</span><span class="p">(</span><span class="n">idx</span><span class="p">)</span>
+                             <span class="k">for</span> <span class="n">idx</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="p">))]</span>
+            <span class="k">return</span> <span class="n">instance_list</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;dict&quot;</span><span class="p">:</span>
+            <span class="n">instance_list</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="p">[</span><span class="n">KEY_INSTANCES</span><span class="p">])</span>
+            <span class="c1"># TODO: should be a list of instances, instance should be huggingface datasets row format</span>
+            <span class="k">return</span> <span class="n">instance_list</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s1">&#39;Current .to_list is not supported for backend &quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">backend</span><span class="si">}</span><span class="s1">&quot;&#39;</span>
+            <span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="Dataset.map">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.map">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">map</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        args : Optional.</span>
+<span class="sd">            Positional arguments.</span>
+<span class="sd">        </span>
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        ---------</span>
+
+<span class="sd">        self : Dataset object.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># If the dataset uses Hugging Face as the backend, </span>
+        <span class="c1"># call the `map()` function of the Hugging Face backend dataset</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;huggingface&quot;</span><span class="p">:</span>
+            <span class="c1"># Set the mapped dataset as the backend dataset of the current dataset</span>
+            <span class="n">mapped_backend_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span> <span class="o">=</span> <span class="n">mapped_backend_dataset</span>
+            <span class="k">return</span> <span class="bp">self</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># If the backend is not Hugging Face, raise a NotImplementedError</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s1">&#39;Currently .map is not supported for backend &quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">backend</span><span class="si">}</span><span class="s1">&quot;&#39;</span>
+            <span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="Dataset.get_backend">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.get_backend">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_backend</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]:</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Returns</span>
+<span class="sd">        ---------</span>
+
+<span class="sd">        self.backend</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span></div>
+
+
+
+<div class="viewcode-block" id="Dataset.get_backend_dataset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.get_backend_dataset">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_backend_dataset</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Returns</span>
+<span class="sd">        ---------</span>
+
+<span class="sd">        self.backend_dataset</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span></div>
+
+
+
+<div class="viewcode-block" id="Dataset.get_fingerprint">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.get_fingerprint">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_fingerprint</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Returns</span>
+<span class="sd">        ---------</span>
+
+<span class="sd">        Fingerprint of the backend_dataset which controls the cache</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="o">.</span><span class="n">_fingerprint</span></div>
+
+
+    
+<div class="viewcode-block" id="Dataset.get_data_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.get_data_args">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_data_args</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Returns</span>
+<span class="sd">        ---------</span>
+
+<span class="sd">        self.data_args</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span></div>
+
+
+
+<div class="viewcode-block" id="Dataset.get_type">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.get_type">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_type</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Returns</span>
+<span class="sd">        ---------</span>
+
+<span class="sd">        self.type</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">type</span></div>
+
+    
+    
+<div class="viewcode-block" id="Dataset.save">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.save">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">save</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> 
+        <span class="nb">format</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s2">&quot;json&quot;</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Save the dataset to a json file.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        file_path : str.</span>
+<span class="sd">            The path to the file where the dataset will be saved.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="nb">format</span> <span class="o">==</span> <span class="s2">&quot;json&quot;</span><span class="p">:</span>
+            <span class="k">assert</span> <span class="n">Path</span><span class="p">(</span><span class="n">file_path</span><span class="p">)</span><span class="o">.</span><span class="n">suffix</span> <span class="o">==</span> <span class="s2">&quot;.json&quot;</span><span class="p">,</span> <span class="s2">&quot;The file path must have a .json extension.&quot;</span>
+            <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">file_path</span><span class="p">,</span> <span class="s2">&quot;w&quot;</span><span class="p">,</span> <span class="n">encoding</span><span class="o">=</span><span class="s1">&#39;utf-8&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">fout</span><span class="p">:</span>
+                <span class="n">json</span><span class="o">.</span><span class="n">dump</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">to_dict</span><span class="p">(),</span> <span class="n">fout</span><span class="p">,</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">ensure_ascii</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+                
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">error</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unsupported format when saving the dataset: </span><span class="si">{</span><span class="nb">format</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span></div>
+
+        
+            
+<div class="viewcode-block" id="Dataset.sample">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.sample">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">sample</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">n</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">seed</span><span class="p">:</span> <span class="nb">int</span><span class="o">=</span><span class="mi">42</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Sample n instances from the dataset.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        n : int.</span>
+<span class="sd">            The number of instances to sample from the dataset.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        ---------</span>
+
+<span class="sd">        sample_dataset : Dataset object.</span>
+<span class="sd">            A new dataset object containing the sampled instances.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;huggingface&quot;</span><span class="p">:</span>
+            <span class="n">sampled_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="o">.</span><span class="n">shuffle</span><span class="p">(</span><span class="n">seed</span><span class="o">=</span><span class="n">seed</span><span class="p">)</span><span class="o">.</span><span class="n">select</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="n">n</span><span class="p">))</span>
+            <span class="n">output_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">create_from_dict</span><span class="p">(</span>
+                <span class="p">{</span>
+                    <span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_type</span><span class="p">(),</span>
+                    <span class="s2">&quot;instances&quot;</span><span class="p">:</span> <span class="p">[</span>
+                        <span class="p">{</span>
+                            <span class="n">col_name</span><span class="p">:</span> <span class="n">sampled_dataset</span><span class="p">[</span><span class="n">col_name</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="k">for</span> <span class="n">col_name</span> <span class="ow">in</span> <span class="n">sampled_dataset</span><span class="o">.</span><span class="n">column_names</span>
+                        <span class="p">}</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">n</span><span class="p">)</span>
+                    <span class="p">]</span>
+                <span class="p">}</span>
+            <span class="p">)</span>
+            <span class="k">return</span> <span class="n">output_dataset</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s1">&#39;Currently .sample is not supported for backend &quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">backend</span><span class="si">}</span><span class="s1">&quot;&#39;</span>
+            <span class="p">)</span></div>
+
+            
+            
+<div class="viewcode-block" id="Dataset.train_test_split">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.train_test_split">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">train_test_split</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">test_size</span><span class="p">:</span> <span class="nb">float</span><span class="o">=</span><span class="mf">0.2</span><span class="p">,</span> <span class="n">shuffle</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">seed</span><span class="p">:</span> <span class="nb">int</span><span class="o">=</span><span class="mi">42</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Split the dataset into training and testing sets.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        test_size : float, default=0.2.</span>
+<span class="sd">            The proportion of the dataset that will be used for testing.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        ---------</span>
+
+<span class="sd">        train_dataset : Dataset object.</span>
+<span class="sd">            A new dataset object containing the training instances.</span>
+<span class="sd">        </span>
+<span class="sd">        test_dataset : Dataset object.</span>
+<span class="sd">            A new dataset object containing the testing instances.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;huggingface&quot;</span><span class="p">:</span>
+            <span class="n">splited</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span>
+                <span class="n">test_size</span><span class="o">=</span><span class="n">test_size</span><span class="p">,</span> <span class="n">shuffle</span><span class="o">=</span><span class="n">shuffle</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="n">seed</span>
+            <span class="p">)</span>
+            <span class="n">train_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">create_from_dict</span><span class="p">(</span>
+                <span class="p">{</span>
+                    <span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_type</span><span class="p">(),</span>
+                    <span class="s2">&quot;instances&quot;</span><span class="p">:</span> <span class="p">[</span>
+                        <span class="p">{</span>
+                            <span class="n">col_name</span><span class="p">:</span> <span class="n">splited</span><span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">][</span><span class="n">col_name</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="k">for</span> <span class="n">col_name</span> <span class="ow">in</span> <span class="n">splited</span><span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">column_names</span>
+                        <span class="p">}</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">splited</span><span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">]))</span>
+                    <span class="p">]</span>
+                <span class="p">}</span>
+            <span class="p">)</span>
+            <span class="n">test_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">create_from_dict</span><span class="p">(</span>
+                <span class="p">{</span>
+                    <span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_type</span><span class="p">(),</span>
+                    <span class="s2">&quot;instances&quot;</span><span class="p">:</span> <span class="p">[</span>
+                        <span class="p">{</span>
+                            <span class="n">col_name</span><span class="p">:</span> <span class="n">splited</span><span class="p">[</span><span class="s2">&quot;test&quot;</span><span class="p">][</span><span class="n">col_name</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="k">for</span> <span class="n">col_name</span> <span class="ow">in</span> <span class="n">splited</span><span class="p">[</span><span class="s2">&quot;test&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">column_names</span>
+                        <span class="p">}</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">splited</span><span class="p">[</span><span class="s2">&quot;test&quot;</span><span class="p">]))</span>
+                    <span class="p">]</span>
+                <span class="p">}</span>
+            <span class="p">)</span>
+            <span class="k">return</span> <span class="n">train_dataset</span><span class="p">,</span> <span class="n">test_dataset</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s1">&#39;Currently .train_test_split is not supported for backend &quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">backend</span><span class="si">}</span><span class="s1">&quot;&#39;</span>
+            <span class="p">)</span></div>
+
+            
+            
+<div class="viewcode-block" id="Dataset.drop_instances">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.drop_instances">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">drop_instances</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">indices</span><span class="p">:</span> <span class="nb">list</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Drop instances from the dataset.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        indices : list.</span>
+<span class="sd">            A list of indices of the instances to drop from the dataset.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;huggingface&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="o">.</span><span class="n">remove_indices</span><span class="p">(</span><span class="n">indices</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s1">&#39;Currently .drop_instances is not supported for backend &quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">backend</span><span class="si">}</span><span class="s1">&quot;&#39;</span>
+            <span class="p">)</span></div>
+
+            
+    
+<div class="viewcode-block" id="Dataset.sanity_check">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.sanity_check">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">sanity_check</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">drop_invalid</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform a sanity check on the dataset.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;huggingface&quot;</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">hf_dataset_sanity_check</span><span class="p">(</span><span class="n">drop_invalid</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s1">&#39;Currently .sanity_check is not supported for backend &quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">backend</span><span class="si">}</span><span class="s1">&quot;&#39;</span>
+            <span class="p">)</span></div>
+
+            
+            
+<div class="viewcode-block" id="Dataset.hf_dataset_sanity_check">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/index.html#lmflow.datasets.dataset.Dataset.hf_dataset_sanity_check">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">hf_dataset_sanity_check</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">drop_invalid</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform a sanity check on the HuggingFace dataset.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Dataset is empty.&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;text_to_textlist&#39;</span><span class="p">:</span>
+            <span class="n">num_output_per_instance</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="p">[</span><span class="s1">&#39;output&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+            <span class="n">dataset_cache</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="o">.</span><span class="n">filter</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">])</span><span class="o">!=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">dataset_cache</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="o">.</span><span class="n">filter</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="s1">&#39;output&#39;</span><span class="p">])</span> <span class="o">==</span> <span class="n">num_output_per_instance</span><span class="p">)</span>
+            <span class="n">dataset_cache</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="o">.</span><span class="n">filter</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="ow">not</span> <span class="nb">all</span><span class="p">([</span><span class="nb">len</span><span class="p">(</span><span class="n">output</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">for</span> <span class="n">output</span> <span class="ow">in</span> <span class="n">x</span><span class="p">[</span><span class="s1">&#39;output&#39;</span><span class="p">]]))</span>
+            
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">dataset_cache</span><span class="p">)</span> <span class="o">!=</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="p">):</span>
+                <span class="n">warning_info</span> <span class="o">=</span> <span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;Found </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span><span class="p">)</span><span class="w"> </span><span class="o">-</span><span class="w"> </span><span class="nb">len</span><span class="p">(</span><span class="n">dataset_cache</span><span class="p">)</span><span class="si">}</span><span class="s2"> invalid instances &quot;</span>
+                    <span class="s2">&quot;during hf_dataset_sanity_check, please check:</span><span class="se">\n</span><span class="s2">&quot;</span>
+                    <span class="s2">&quot;   1. length of input strings should not be empty</span><span class="se">\n</span><span class="s2">&quot;</span>
+                    <span class="s2">&quot;   2. length of output strings should not be all empty</span><span class="se">\n</span><span class="s2">&quot;</span>
+                    <span class="s2">&quot;   3. number of output strings should be consistent</span><span class="se">\n</span><span class="s2">&quot;</span> <span class="c1"># since we will use tensor reshape later</span>
+                <span class="p">)</span>
+                <span class="k">if</span> <span class="n">drop_invalid</span><span class="p">:</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">backend_dataset</span> <span class="o">=</span> <span class="n">dataset_cache</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="n">warning_info</span><span class="o">+</span><span class="s2">&quot;Invalid instances are dropped.&quot;</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="n">warning_info</span><span class="p">)</span>
+        
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;No sanity check for </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="s2"> dataset.&quot;</span><span class="p">)</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/datasets/multi_modal_dataset.html b/_modules/lmflow/datasets/multi_modal_dataset.html
new file mode 100644
index 000000000..22fadbc41
--- /dev/null
+++ b/_modules/lmflow/datasets/multi_modal_dataset.html
@@ -0,0 +1,783 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.datasets.multi_modal_dataset &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/datasets/multi_modal_dataset';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.datasets.multi_modal_dataset</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.datasets.multi_modal_dataset</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># FIXME update the doc string.</span>
+<span class="sd">&quot;&quot;&quot;This Python code defines a class Multi Modal Dataset.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">copy</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">dataclasses</span><span class="w"> </span><span class="kn">import</span> <span class="n">dataclass</span><span class="p">,</span> <span class="n">field</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">json</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">PIL</span><span class="w"> </span><span class="kn">import</span> <span class="n">Image</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os.path</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">osp</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">transformers</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.utils.data</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="n">DatasetArguments</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">llava_conversation_lib</span> <span class="k">as</span> <span class="n">conversation_lib</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="n">IGNORE_INDEX</span><span class="p">,</span> <span class="n">IMAGE_TOKEN_INDEX</span><span class="p">,</span> <span class="n">DEFAULT_IMAGE_TOKEN</span><span class="p">,</span> <span class="n">DEFAULT_IM_START_TOKEN</span><span class="p">,</span> <span class="n">DEFAULT_IM_END_TOKEN</span>
+
+<div class="viewcode-block" id="CustomMultiModalDataset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">CustomMultiModalDataset</span><span class="p">(</span><span class="n">Dataset</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Dataset for Multi Modal data&quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dataset_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+                 <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">CustomMultiModalDataset</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        <span class="n">data_dict</span> <span class="o">=</span> <span class="n">json</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="nb">open</span><span class="p">(</span><span class="n">dataset_path</span><span class="p">,</span> <span class="s2">&quot;r&quot;</span><span class="p">))</span>
+<div class="viewcode-block" id="CustomMultiModalDataset.data_dict">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.data_dict">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_dict</span> <span class="o">=</span> <span class="n">data_dict</span></div>
+
+        <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Finish loading json file in dataset.&quot;</span><span class="p">)</span>
+<div class="viewcode-block" id="CustomMultiModalDataset.data_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.data_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span> <span class="o">=</span> <span class="n">data_args</span></div>
+
+<div class="viewcode-block" id="CustomMultiModalDataset.image_folder">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.image_folder">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">image_folder</span> <span class="o">=</span> <span class="n">data_args</span><span class="o">.</span><span class="n">image_folder</span></div>
+
+
+<div class="viewcode-block" id="CustomMultiModalDataset.__len__">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.__len__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__len__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data_dict</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="CustomMultiModalDataset.register_tokenizer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.register_tokenizer">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">register_tokenizer</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">,</span> <span class="n">image_processor</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span> <span class="o">=</span> <span class="n">tokenizer</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">image_processor</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span>
+            <span class="n">tokenizer</span><span class="p">,</span> <span class="s2">&quot;image_processor&quot;</span><span class="p">,</span> <span class="n">image_processor</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="CustomMultiModalDataset.__getitem__">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.__getitem__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__getitem__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">i</span><span class="p">):</span>
+        <span class="n">data</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_dict</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">i</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="p">[</span><span class="n">data</span><span class="p">]</span>
+        <span class="k">assert</span> <span class="nb">len</span><span class="p">(</span><span class="n">data</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span>
+        <span class="n">processor</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">image_processor</span>
+        <span class="k">if</span> <span class="s1">&#39;image&#39;</span> <span class="ow">in</span> <span class="n">data</span><span class="p">[</span><span class="mi">0</span><span class="p">]:</span>
+            <span class="n">image_file</span> <span class="o">=</span> <span class="n">data</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;image&#39;</span><span class="p">]</span>
+            <span class="n">image</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">open</span><span class="p">(</span>
+                <span class="n">osp</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">image_folder</span><span class="p">,</span> <span class="n">image_file</span><span class="p">))</span><span class="o">.</span><span class="n">convert</span><span class="p">(</span><span class="s2">&quot;RGB&quot;</span><span class="p">)</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">image_aspect_ratio</span> <span class="o">==</span> <span class="s1">&#39;pad&#39;</span><span class="p">:</span>
+                <span class="k">def</span><span class="w"> </span><span class="nf">expand2square</span><span class="p">(</span><span class="n">pil_img</span><span class="p">,</span> <span class="n">background_color</span><span class="p">):</span>
+                    <span class="n">width</span><span class="p">,</span> <span class="n">height</span> <span class="o">=</span> <span class="n">pil_img</span><span class="o">.</span><span class="n">size</span>
+                    <span class="k">if</span> <span class="n">width</span> <span class="o">==</span> <span class="n">height</span><span class="p">:</span>
+                        <span class="k">return</span> <span class="n">pil_img</span>
+                    <span class="k">elif</span> <span class="n">width</span> <span class="o">&gt;</span> <span class="n">height</span><span class="p">:</span>
+                        <span class="n">result</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">new</span><span class="p">(</span><span class="n">pil_img</span><span class="o">.</span><span class="n">mode</span><span class="p">,</span> <span class="p">(</span><span class="n">width</span><span class="p">,</span> <span class="n">width</span><span class="p">),</span> <span class="n">background_color</span><span class="p">)</span>
+                        <span class="n">result</span><span class="o">.</span><span class="n">paste</span><span class="p">(</span><span class="n">pil_img</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="p">(</span><span class="n">width</span> <span class="o">-</span> <span class="n">height</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span><span class="p">))</span>
+                        <span class="k">return</span> <span class="n">result</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">result</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">new</span><span class="p">(</span><span class="n">pil_img</span><span class="o">.</span><span class="n">mode</span><span class="p">,</span> <span class="p">(</span><span class="n">height</span><span class="p">,</span> <span class="n">height</span><span class="p">),</span> <span class="n">background_color</span><span class="p">)</span>
+                        <span class="n">result</span><span class="o">.</span><span class="n">paste</span><span class="p">(</span><span class="n">pil_img</span><span class="p">,</span> <span class="p">((</span><span class="n">height</span> <span class="o">-</span> <span class="n">width</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">))</span>
+                        <span class="k">return</span> <span class="n">result</span>    
+                <span class="n">image</span> <span class="o">=</span> <span class="n">expand2square</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">(</span><span class="nb">int</span><span class="p">(</span><span class="n">x</span><span class="o">*</span><span class="mi">255</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">processor</span><span class="o">.</span><span class="n">image_mean</span><span class="p">))</span>
+                <span class="n">image</span> <span class="o">=</span> <span class="n">processor</span><span class="o">.</span><span class="n">preprocess</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s1">&#39;pt&#39;</span><span class="p">)[</span><span class="s1">&#39;pixel_values&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">image</span> <span class="o">=</span> <span class="n">processor</span><span class="o">.</span><span class="n">preprocess</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s1">&#39;pt&#39;</span><span class="p">)[</span><span class="s1">&#39;pixel_values&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="n">preprocess_multimodal_llava</span><span class="p">(</span>
+                <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">([</span><span class="n">e</span><span class="p">[</span><span class="s2">&quot;conversations&quot;</span><span class="p">]</span> <span class="k">for</span> <span class="n">e</span> <span class="ow">in</span> <span class="n">data</span><span class="p">]),</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">data</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">([</span><span class="n">e</span><span class="p">[</span><span class="s2">&quot;conversations&quot;</span><span class="p">]</span> <span class="k">for</span> <span class="n">e</span> <span class="ow">in</span> <span class="n">data</span><span class="p">])</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">sep_style</span> <span class="o">==</span> <span class="s2">&quot;plain&quot;</span><span class="p">:</span>
+            <span class="n">data_dict</span> <span class="o">=</span> <span class="n">preprocess_llama_from_llava_plain</span><span class="p">(</span>
+                <span class="n">data</span><span class="p">,</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">,</span>
+                <span class="n">has_image</span><span class="o">=</span><span class="p">(</span><span class="s1">&#39;image&#39;</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_dict</span><span class="p">[</span><span class="n">i</span><span class="p">])</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">data_dict</span> <span class="o">=</span> <span class="n">preprocess_llama_from_llava_v1</span><span class="p">(</span>
+                <span class="n">data</span><span class="p">,</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">,</span>
+                <span class="n">has_image</span><span class="o">=</span><span class="p">(</span><span class="s1">&#39;image&#39;</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_dict</span><span class="p">[</span><span class="n">i</span><span class="p">])</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">i</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="n">data_dict</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">input_ids</span><span class="o">=</span><span class="n">data_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">],</span>
+                             <span class="n">labels</span><span class="o">=</span><span class="n">data_dict</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+        
+        <span class="c1"># image exist in the data</span>
+        <span class="k">if</span> <span class="s1">&#39;image&#39;</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_dict</span><span class="p">[</span><span class="n">i</span><span class="p">]:</span>
+            <span class="n">data_dict</span><span class="p">[</span><span class="s1">&#39;image&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">image</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># image does not exist in the data, but the model is multimodal</span>
+            <span class="n">crop_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">image_processor</span><span class="o">.</span><span class="n">crop_size</span>
+            <span class="n">data_dict</span><span class="p">[</span><span class="s1">&#39;image&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span>
+                <span class="mi">3</span><span class="p">,</span> <span class="n">crop_size</span><span class="p">[</span><span class="s1">&#39;height&#39;</span><span class="p">],</span> <span class="n">crop_size</span><span class="p">[</span><span class="s1">&#39;width&#39;</span><span class="p">])</span>
+        <span class="k">return</span> <span class="n">data_dict</span></div>
+</div>
+
+
+
+
+<div class="viewcode-block" id="preprocess_multimodal_llava">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.preprocess_multimodal_llava">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">preprocess_multimodal_llava</span><span class="p">(</span><span class="n">sources</span><span class="p">,</span> <span class="n">data_args</span><span class="p">):</span>
+    <span class="n">is_multimodal</span> <span class="o">=</span> <span class="n">data_args</span><span class="o">.</span><span class="n">is_multimodal</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">is_multimodal</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">sources</span>
+
+    <span class="k">for</span> <span class="n">source</span> <span class="ow">in</span> <span class="n">sources</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">sentence</span> <span class="ow">in</span> <span class="n">source</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">DEFAULT_IMAGE_TOKEN</span> <span class="ow">in</span> <span class="n">sentence</span><span class="p">[</span><span class="s1">&#39;value&#39;</span><span class="p">]:</span>
+                <span class="n">sentence</span><span class="p">[</span><span class="s1">&#39;value&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">sentence</span><span class="p">[</span><span class="s1">&#39;value&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span>
+                                <span class="n">DEFAULT_IMAGE_TOKEN</span><span class="p">,</span> <span class="s1">&#39;&#39;</span><span class="p">)</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span>
+                <span class="n">sentence</span><span class="p">[</span><span class="s1">&#39;value&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">DEFAULT_IMAGE_TOKEN</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">sentence</span><span class="p">[</span><span class="s1">&#39;value&#39;</span><span class="p">]</span>
+                <span class="n">sentence</span><span class="p">[</span><span class="s1">&#39;value&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">sentence</span><span class="p">[</span><span class="s1">&#39;value&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span>
+                <span class="k">if</span> <span class="s2">&quot;mmtag&quot;</span> <span class="ow">in</span> <span class="n">conversation_lib</span><span class="o">.</span><span class="n">default_conversation</span><span class="o">.</span><span class="n">version</span><span class="p">:</span>
+                    <span class="n">sentence</span><span class="p">[</span><span class="s1">&#39;value&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">sentence</span><span class="p">[</span><span class="s1">&#39;value&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span>
+                        <span class="n">DEFAULT_IMAGE_TOKEN</span><span class="p">,</span> <span class="s1">&#39;&lt;Image&gt;&#39;</span> <span class="o">+</span> <span class="n">DEFAULT_IMAGE_TOKEN</span> <span class="o">+</span> <span class="s1">&#39;&lt;/Image&gt;&#39;</span><span class="p">)</span>
+            <span class="n">replace_token</span> <span class="o">=</span> <span class="n">DEFAULT_IMAGE_TOKEN</span>
+            <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">use_image_start_end</span><span class="p">:</span>
+                <span class="n">replace_token</span> <span class="o">=</span> <span class="n">DEFAULT_IM_START_TOKEN</span> <span class="o">+</span> \
+                    <span class="n">replace_token</span> <span class="o">+</span> <span class="n">DEFAULT_IM_END_TOKEN</span>
+            <span class="n">sentence</span><span class="p">[</span><span class="s2">&quot;value&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">sentence</span><span class="p">[</span><span class="s2">&quot;value&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span>
+                    <span class="n">DEFAULT_IMAGE_TOKEN</span><span class="p">,</span> <span class="n">replace_token</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">sources</span></div>
+
+
+
+<div class="viewcode-block" id="tokenizer_image_token">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.tokenizer_image_token">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">tokenizer_image_token</span><span class="p">(</span><span class="n">prompt</span><span class="p">,</span>
+                          <span class="n">tokenizer</span><span class="p">,</span>
+                          <span class="n">image_token_index</span><span class="o">=</span><span class="n">IMAGE_TOKEN_INDEX</span><span class="p">,</span>
+                          <span class="n">return_tensors</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+    <span class="n">prompt_chunks</span> <span class="o">=</span> <span class="p">[</span><span class="n">tokenizer</span><span class="p">(</span><span class="n">chunk</span><span class="p">)</span><span class="o">.</span><span class="n">input_ids</span> <span class="k">for</span> <span class="n">chunk</span> <span class="ow">in</span> <span class="n">prompt</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s1">&#39;&lt;image&gt;&#39;</span><span class="p">)]</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="nf">insert_separator</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">sep</span><span class="p">):</span>
+        <span class="k">return</span> <span class="p">[</span><span class="n">ele</span> <span class="k">for</span> <span class="n">sublist</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="p">[</span><span class="n">sep</span><span class="p">]</span><span class="o">*</span><span class="nb">len</span><span class="p">(</span><span class="n">X</span><span class="p">))</span> <span class="k">for</span> <span class="n">ele</span> <span class="ow">in</span> <span class="n">sublist</span><span class="p">][:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+
+    <span class="n">input_ids</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="n">offset</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">prompt_chunks</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">prompt_chunks</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">prompt_chunks</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span> <span class="o">==</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">bos_token_id</span><span class="p">:</span>
+        <span class="n">offset</span> <span class="o">=</span> <span class="mi">1</span>
+        <span class="n">input_ids</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">prompt_chunks</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span>
+
+    <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">insert_separator</span><span class="p">(</span><span class="n">prompt_chunks</span><span class="p">,</span> <span class="p">[</span><span class="n">image_token_index</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="n">offset</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)):</span>
+        <span class="n">input_ids</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="n">offset</span><span class="p">:])</span>
+
+    <span class="k">if</span> <span class="n">return_tensors</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">return_tensors</span> <span class="o">==</span> <span class="s1">&#39;pt&#39;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">input_ids</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">long</span><span class="p">)</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;Unsupported tensor type: </span><span class="si">{</span><span class="n">return_tensors</span><span class="si">}</span><span class="s1">&#39;</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">input_ids</span></div>
+
+
+
+<div class="viewcode-block" id="preprocess_llama_from_llava_plain">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_plain">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">preprocess_llama_from_llava_plain</span><span class="p">(</span>
+    <span class="n">sources</span><span class="p">,</span>
+    <span class="n">tokenizer</span><span class="p">:</span> <span class="n">transformers</span><span class="o">.</span><span class="n">PreTrainedTokenizer</span><span class="p">,</span>
+    <span class="n">has_image</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    This function just add the image in the front of text.</span>
+<span class="sd">    And don&#39;t add any prompt.</span>
+<span class="sd">    Args:</span>
+<span class="sd">        sources: The input data with text and image.</span>
+<span class="sd">        tokenizer: The tokenizer to process text.</span>
+<span class="sd">        has_image: Whether the input data has image.</span>
+<span class="sd">    Returns:</span>
+<span class="sd">        The input_ids and labels for the model.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">conversations</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">for</span> <span class="n">source</span> <span class="ow">in</span> <span class="n">sources</span><span class="p">:</span>
+        <span class="k">assert</span> <span class="nb">len</span><span class="p">(</span><span class="n">source</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span>
+        <span class="k">assert</span> <span class="n">DEFAULT_IMAGE_TOKEN</span> <span class="ow">in</span> <span class="n">source</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;value&#39;</span><span class="p">]</span>
+        <span class="n">source</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;value&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">DEFAULT_IMAGE_TOKEN</span>
+        <span class="n">conversation</span> <span class="o">=</span> <span class="n">source</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;value&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="n">source</span><span class="p">[</span><span class="mi">1</span><span class="p">][</span><span class="s1">&#39;value&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="n">conversation_lib</span><span class="o">.</span><span class="n">default_conversation</span><span class="o">.</span><span class="n">sep</span>
+        <span class="n">conversations</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">conversation</span><span class="p">)</span>
+    <span class="c1"># tokenize conversations</span>
+    <span class="n">input_ids</span> <span class="o">=</span> <span class="p">[</span><span class="n">tokenizer_image_token</span><span class="p">(</span><span class="n">prompt</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s1">&#39;pt&#39;</span><span class="p">)</span> <span class="k">for</span> <span class="n">prompt</span> <span class="ow">in</span> <span class="n">conversations</span><span class="p">]</span>
+    <span class="n">targets</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">input_ids</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">target</span><span class="p">,</span> <span class="n">source</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">targets</span><span class="p">,</span> <span class="n">sources</span><span class="p">):</span>
+        <span class="n">tokenized_len</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">tokenizer_image_token</span><span class="p">(</span><span class="n">source</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;value&#39;</span><span class="p">],</span> <span class="n">tokenizer</span><span class="p">))</span>
+        <span class="n">target</span><span class="p">[:</span><span class="n">tokenized_len</span><span class="p">]</span> <span class="o">=</span> <span class="n">IGNORE_INDEX</span>
+
+    <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="n">input_ids</span><span class="o">=</span><span class="n">input_ids</span><span class="p">,</span> <span class="n">labels</span><span class="o">=</span><span class="n">targets</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="preprocess_llama_from_llava_v1">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_v1">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">preprocess_llama_from_llava_v1</span><span class="p">(</span>
+    <span class="n">sources</span><span class="p">,</span>
+    <span class="n">tokenizer</span><span class="p">:</span> <span class="n">transformers</span><span class="o">.</span><span class="n">PreTrainedTokenizer</span><span class="p">,</span>
+    <span class="n">has_image</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    This function add the prompt and then put the image after the prompt.</span>
+<span class="sd">    So it needs additional code to generate the target label.</span>
+<span class="sd">    Args:</span>
+<span class="sd">        sources: The input data with text and image.</span>
+<span class="sd">        tokenizer: The tokenizer to process text.</span>
+<span class="sd">        has_image: Whether the input data has image.</span>
+<span class="sd">    Returns:</span>
+<span class="sd">        The input_ids and labels for the model.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">conv</span> <span class="o">=</span> <span class="n">conversation_lib</span><span class="o">.</span><span class="n">default_conversation</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
+    <span class="n">roles</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;human&quot;</span><span class="p">:</span> <span class="n">conv</span><span class="o">.</span><span class="n">roles</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="s2">&quot;gpt&quot;</span><span class="p">:</span> <span class="n">conv</span><span class="o">.</span><span class="n">roles</span><span class="p">[</span><span class="mi">1</span><span class="p">]}</span>
+
+    <span class="c1"># Apply prompt templates</span>
+    <span class="n">conversations</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">source</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">sources</span><span class="p">):</span>
+        <span class="k">if</span> <span class="n">roles</span><span class="p">[</span><span class="n">source</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s2">&quot;from&quot;</span><span class="p">]]</span> <span class="o">!=</span> <span class="n">conv</span><span class="o">.</span><span class="n">roles</span><span class="p">[</span><span class="mi">0</span><span class="p">]:</span>
+            <span class="c1"># Skip the first one if it is not from human</span>
+            <span class="n">source</span> <span class="o">=</span> <span class="n">source</span><span class="p">[</span><span class="mi">1</span><span class="p">:]</span>
+
+        <span class="n">conv</span><span class="o">.</span><span class="n">messages</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">j</span><span class="p">,</span> <span class="n">sentence</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">source</span><span class="p">):</span>
+            <span class="n">role</span> <span class="o">=</span> <span class="n">roles</span><span class="p">[</span><span class="n">sentence</span><span class="p">[</span><span class="s2">&quot;from&quot;</span><span class="p">]]</span>
+            <span class="k">assert</span> <span class="n">role</span> <span class="o">==</span> <span class="n">conv</span><span class="o">.</span><span class="n">roles</span><span class="p">[</span><span class="n">j</span> <span class="o">%</span> <span class="mi">2</span><span class="p">],</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">i</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="n">conv</span><span class="o">.</span><span class="n">append_message</span><span class="p">(</span><span class="n">role</span><span class="p">,</span> <span class="n">sentence</span><span class="p">[</span><span class="s2">&quot;value&quot;</span><span class="p">])</span>
+        <span class="n">conversations</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">conv</span><span class="o">.</span><span class="n">get_prompt</span><span class="p">())</span>
+
+    <span class="c1"># Tokenize conversations</span>
+
+    <span class="k">if</span> <span class="n">has_image</span><span class="p">:</span>
+        <span class="n">input_ids</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">([</span><span class="n">tokenizer_image_token</span><span class="p">(</span><span class="n">prompt</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s1">&#39;pt&#39;</span><span class="p">)</span> <span class="k">for</span> <span class="n">prompt</span> <span class="ow">in</span> <span class="n">conversations</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">input_ids</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">(</span>
+            <span class="n">conversations</span><span class="p">,</span>
+            <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span><span class="p">,</span>
+            <span class="n">padding</span><span class="o">=</span><span class="s2">&quot;longest&quot;</span><span class="p">,</span>
+            <span class="n">max_length</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="p">,</span>
+            <span class="n">truncation</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="p">)</span><span class="o">.</span><span class="n">input_ids</span>
+
+    <span class="n">targets</span> <span class="o">=</span> <span class="n">input_ids</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+    <span class="k">assert</span> <span class="n">conv</span><span class="o">.</span><span class="n">sep_style</span> <span class="o">==</span> <span class="n">conversation_lib</span><span class="o">.</span><span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">TWO</span>
+
+    <span class="c1"># Mask targets</span>
+    <span class="n">sep</span> <span class="o">=</span> <span class="n">conv</span><span class="o">.</span><span class="n">sep</span> <span class="o">+</span> <span class="n">conv</span><span class="o">.</span><span class="n">roles</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="s2">&quot;: &quot;</span>
+    <span class="k">for</span> <span class="n">conversation</span><span class="p">,</span> <span class="n">target</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">conversations</span><span class="p">,</span> <span class="n">targets</span><span class="p">):</span>
+        <span class="n">total_len</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">target</span><span class="o">.</span><span class="n">ne</span><span class="p">(</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span><span class="p">)</span><span class="o">.</span><span class="n">sum</span><span class="p">())</span>
+
+        <span class="n">rounds</span> <span class="o">=</span> <span class="n">conversation</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">conv</span><span class="o">.</span><span class="n">sep2</span><span class="p">)</span>
+        <span class="n">cur_len</span> <span class="o">=</span> <span class="mi">1</span>
+        <span class="n">target</span><span class="p">[:</span><span class="n">cur_len</span><span class="p">]</span> <span class="o">=</span> <span class="n">IGNORE_INDEX</span>
+        <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">rou</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">rounds</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">rou</span> <span class="o">==</span> <span class="s2">&quot;&quot;</span><span class="p">:</span>
+                <span class="k">break</span>
+
+            <span class="n">parts</span> <span class="o">=</span> <span class="n">rou</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">sep</span><span class="p">)</span>
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">parts</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">2</span><span class="p">:</span>
+                <span class="k">break</span>
+            <span class="n">parts</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">+=</span> <span class="n">sep</span>
+
+            <span class="k">if</span> <span class="n">has_image</span><span class="p">:</span>
+                <span class="n">round_len</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">tokenizer_image_token</span><span class="p">(</span><span class="n">rou</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">))</span>
+                <span class="n">instruction_len</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">tokenizer_image_token</span><span class="p">(</span><span class="n">parts</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">tokenizer</span><span class="p">))</span> <span class="o">-</span> <span class="mi">2</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">round_len</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">tokenizer</span><span class="p">(</span><span class="n">rou</span><span class="p">)</span><span class="o">.</span><span class="n">input_ids</span><span class="p">)</span>
+                <span class="n">instruction_len</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">tokenizer</span><span class="p">(</span><span class="n">parts</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span><span class="o">.</span><span class="n">input_ids</span><span class="p">)</span> <span class="o">-</span> <span class="mi">2</span>
+
+            <span class="n">target</span><span class="p">[</span><span class="n">cur_len</span> <span class="p">:</span> <span class="n">cur_len</span> <span class="o">+</span> <span class="n">instruction_len</span><span class="p">]</span> <span class="o">=</span> <span class="n">IGNORE_INDEX</span>
+
+            <span class="n">cur_len</span> <span class="o">+=</span> <span class="n">round_len</span>
+        <span class="n">target</span><span class="p">[</span><span class="n">cur_len</span><span class="p">:]</span> <span class="o">=</span> <span class="n">IGNORE_INDEX</span>
+
+        <span class="k">if</span> <span class="n">cur_len</span> <span class="o">&lt;</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">cur_len</span> <span class="o">!=</span> <span class="n">total_len</span><span class="p">:</span>
+                <span class="n">target</span><span class="p">[:]</span> <span class="o">=</span> <span class="n">IGNORE_INDEX</span>
+                <span class="nb">print</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;WARNING: tokenization mismatch: </span><span class="si">{</span><span class="n">cur_len</span><span class="si">}</span><span class="s2"> vs. </span><span class="si">{</span><span class="n">total_len</span><span class="si">}</span><span class="s2">.&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot; (ignored)&quot;</span>
+                <span class="p">)</span>
+
+    <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span>
+        <span class="n">input_ids</span><span class="o">=</span><span class="n">input_ids</span><span class="p">,</span>
+        <span class="n">labels</span><span class="o">=</span><span class="n">targets</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="DataCollatorForSupervisedDataset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">DataCollatorForSupervisedDataset</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Collate examples for supervised fine-tuning.&quot;&quot;&quot;</span>
+
+<div class="viewcode-block" id="DataCollatorForSupervisedDataset.tokenizer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset.tokenizer">[docs]</a>
+    <span class="n">tokenizer</span><span class="p">:</span> <span class="n">transformers</span><span class="o">.</span><span class="n">PreTrainedTokenizer</span></div>
+
+
+<div class="viewcode-block" id="DataCollatorForSupervisedDataset.__call__">
+<a class="viewcode-back" href="../../../autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset.__call__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">instances</span><span class="p">):</span>
+        <span class="n">input_ids</span><span class="p">,</span> <span class="n">labels</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">([</span><span class="n">instance</span><span class="p">[</span><span class="n">key</span><span class="p">]</span> <span class="k">for</span> <span class="n">instance</span> <span class="ow">in</span> <span class="n">instances</span><span class="p">]</span>
+                                  <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;input_ids&quot;</span><span class="p">,</span> <span class="s2">&quot;labels&quot;</span><span class="p">))</span>
+        <span class="n">input_ids</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">rnn</span><span class="o">.</span><span class="n">pad_sequence</span><span class="p">(</span>
+            <span class="n">input_ids</span><span class="p">,</span>
+            <span class="n">batch_first</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+            <span class="n">padding_value</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span><span class="p">)</span>
+        <span class="n">labels</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">rnn</span><span class="o">.</span><span class="n">pad_sequence</span><span class="p">(</span><span class="n">labels</span><span class="p">,</span>
+                                                 <span class="n">batch_first</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                                                 <span class="n">padding_value</span><span class="o">=</span><span class="n">IGNORE_INDEX</span><span class="p">)</span>
+        <span class="n">input_ids</span> <span class="o">=</span> <span class="n">input_ids</span><span class="p">[:,</span> <span class="p">:</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="p">]</span>
+        <span class="n">labels</span> <span class="o">=</span> <span class="n">labels</span><span class="p">[:,</span> <span class="p">:</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="p">]</span>
+        <span class="n">batch</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span>
+            <span class="n">input_ids</span><span class="o">=</span><span class="n">input_ids</span><span class="p">,</span>
+            <span class="n">labels</span><span class="o">=</span><span class="n">labels</span><span class="p">,</span>
+            <span class="n">attention_mask</span><span class="o">=</span><span class="n">input_ids</span><span class="o">.</span><span class="n">ne</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span><span class="p">),</span>
+        <span class="p">)</span>
+
+        <span class="k">if</span> <span class="s1">&#39;image&#39;</span> <span class="ow">in</span> <span class="n">instances</span><span class="p">[</span><span class="mi">0</span><span class="p">]:</span>
+            <span class="n">images</span> <span class="o">=</span> <span class="p">[</span><span class="n">instance</span><span class="p">[</span><span class="s1">&#39;image&#39;</span><span class="p">]</span> <span class="k">for</span> <span class="n">instance</span> <span class="ow">in</span> <span class="n">instances</span><span class="p">]</span>
+            <span class="k">if</span> <span class="nb">all</span><span class="p">(</span><span class="n">x</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span> <span class="o">==</span> <span class="n">images</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">images</span><span class="p">):</span>
+                <span class="n">batch</span><span class="p">[</span><span class="s1">&#39;images&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">images</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">batch</span><span class="p">[</span><span class="s1">&#39;images&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">images</span>
+        <span class="k">return</span> <span class="n">batch</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/models/auto_model.html b/_modules/lmflow/models/auto_model.html
new file mode 100644
index 000000000..32bbfb043
--- /dev/null
+++ b/_modules/lmflow/models/auto_model.html
@@ -0,0 +1,472 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.models.auto_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/models/auto_model';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.auto_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.models.auto_model</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;Automatically get correct model type.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_decoder_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFDecoderModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_text_regression_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFTextRegressionModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_encoder_decoder_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFEncoderDecoderModel</span>
+
+<div class="viewcode-block" id="AutoModel">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/auto_model/index.html#lmflow.models.auto_model.AutoModel">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">AutoModel</span><span class="p">:</span>
+
+    <span class="nd">@classmethod</span>
+<div class="viewcode-block" id="AutoModel.get_model">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/auto_model/index.html#lmflow.models.auto_model.AutoModel.get_model">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model_args</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+        <span class="n">arch_type</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">arch_type</span>
+        <span class="k">if</span> <span class="n">arch_type</span> <span class="o">==</span> <span class="s2">&quot;decoder_only&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">HFDecoderModel</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">arch_type</span> <span class="o">==</span> <span class="s2">&quot;text_regression&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">HFTextRegressionModel</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">arch_type</span> <span class="o">==</span> <span class="s2">&quot;encoder_decoder&quot;</span> <span class="ow">or</span> \
+                <span class="n">arch_type</span> <span class="o">==</span> <span class="s2">&quot;vision_encoder_decoder&quot;</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">HFEncoderDecoderModel</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;model architecture type </span><span class="se">\&quot;</span><span class="si">{</span><span class="n">arch_type</span><span class="si">}</span><span class="se">\&quot;</span><span class="s2"> is not supported&quot;</span>
+            <span class="p">)</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/models/base_model.html b/_modules/lmflow/models/base_model.html
new file mode 100644
index 000000000..8087f38e0
--- /dev/null
+++ b/_modules/lmflow/models/base_model.html
@@ -0,0 +1,456 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.models.base_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/models/base_model';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.base_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.models.base_model</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;Base model class.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">abc</span><span class="w"> </span><span class="kn">import</span> <span class="n">ABC</span>
+
+
+<div class="viewcode-block" id="BaseModel">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/base_model/index.html#lmflow.models.base_model.BaseModel">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">BaseModel</span><span class="p">(</span><span class="n">ABC</span><span class="p">):</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+        <span class="k">pass</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/models/decoder_model.html b/_modules/lmflow/models/decoder_model.html
new file mode 100644
index 000000000..711d6ae0f
--- /dev/null
+++ b/_modules/lmflow/models/decoder_model.html
@@ -0,0 +1,466 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.models.decoder_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/models/decoder_model';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.decoder_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.models.decoder_model</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;A one-line summary of the module or program, terminated by a period.</span>
+
+<span class="sd">Leave one blank line.  The rest of this docstring should contain an</span>
+<span class="sd">overall description of the module or program.  Optionally, it may also</span>
+<span class="sd">contain a brief description of exported classes and functions and/or usage</span>
+<span class="sd">examples.</span>
+
+<span class="sd">Typical usage example:</span>
+
+<span class="sd">  foo = ClassFoo()</span>
+<span class="sd">  bar = foo.FunctionBar()</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.base_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">BaseModel</span>
+
+
+<div class="viewcode-block" id="DecoderModel">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/decoder_model/index.html#lmflow.models.decoder_model.DecoderModel">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">DecoderModel</span><span class="p">(</span><span class="n">BaseModel</span><span class="p">):</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+        <span class="k">pass</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/models/encoder_decoder_model.html b/_modules/lmflow/models/encoder_decoder_model.html
new file mode 100644
index 000000000..fe8a38b85
--- /dev/null
+++ b/_modules/lmflow/models/encoder_decoder_model.html
@@ -0,0 +1,466 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.models.encoder_decoder_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/models/encoder_decoder_model';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.encoder_decoder_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.models.encoder_decoder_model</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;A one-line summary of the module or program, terminated by a period.</span>
+
+<span class="sd">Leave one blank line.  The rest of this docstring should contain an</span>
+<span class="sd">overall description of the module or program.  Optionally, it may also</span>
+<span class="sd">contain a brief description of exported classes and functions and/or usage</span>
+<span class="sd">examples.</span>
+
+<span class="sd">Typical usage example:</span>
+
+<span class="sd">  foo = ClassFoo()</span>
+<span class="sd">  bar = foo.FunctionBar()</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.base_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">BaseModel</span>
+
+
+<div class="viewcode-block" id="EncoderDecoderModel">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/encoder_decoder_model/index.html#lmflow.models.encoder_decoder_model.EncoderDecoderModel">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">EncoderDecoderModel</span><span class="p">(</span><span class="n">BaseModel</span><span class="p">):</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+        <span class="k">pass</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/models/hf_decoder_model.html b/_modules/lmflow/models/hf_decoder_model.html
new file mode 100644
index 000000000..eab6c6b97
--- /dev/null
+++ b/_modules/lmflow/models/hf_decoder_model.html
@@ -0,0 +1,1157 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.models.hf_decoder_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/models/hf_decoder_model';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.hf_decoder_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.models.hf_decoder_model</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;This is a class called HFDecoderModel which is a wrapper around transformers model and</span>
+<span class="sd">tokenizer classes. It has several methods such as __init__, tokenize, and train that are </span>
+<span class="sd">used for training and fine-tuning the model. The __init__ method takes in several arguments</span>
+<span class="sd">such as model_args, tune_strategy, and ds_config, which are used to load the pretrained </span>
+<span class="sd">model and tokenizer, and initialize the training settings.</span>
+
+<span class="sd">The tokenize method is used to tokenize the input text and return the input IDs and attention</span>
+<span class="sd">masks that can be fed to the model for training or inference.</span>
+
+<span class="sd">This class supports different tune_strategy options such as &#39;normal&#39;, &#39;none&#39;, &#39;lora&#39;, and</span>
+<span class="sd">&#39;adapter&#39;, which allow for different fine-tuning settings of the model. However, the &#39;lora&#39;</span>
+<span class="sd">and &#39;adapter&#39; strategies are not yet implemented.</span>
+
+<span class="sd">Overall, this class provides a convenient interface for loading and fine-tuning transformer</span>
+<span class="sd">models and can be used for various NLP tasks such as language modeling, text classification,</span>
+<span class="sd">and question answering.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">hashlib</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span><span class="o">,</span><span class="w"> </span><span class="nn">shutil</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Dict</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">AutoConfig</span><span class="p">,</span>
+    <span class="n">AutoModelForCausalLM</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">peft</span><span class="w"> </span><span class="kn">import</span> <span class="n">PeftModel</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets.dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_model_mixin</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFModelMixin</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.decoder_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">DecoderModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.interfaces.tunable</span><span class="w"> </span><span class="kn">import</span> <span class="n">Tunable</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">TEXT_ONLY_DATASET_DESCRIPTION</span><span class="p">,</span>
+    <span class="n">TEXT2TEXT_DATASET_DESCRIPTION</span><span class="p">,</span>
+    <span class="n">CONVERSATION_DATASET_DESCRIPTION</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.conversation_template</span><span class="w"> </span><span class="kn">import</span> <span class="n">PRESET_TEMPLATES</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.data_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">VLLMInferenceResultWithInput</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.tokenization.hf_decoder_model</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">tokenize_function</span><span class="p">,</span> 
+    <span class="n">conversation_tokenize_function</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.versioning</span><span class="w"> </span><span class="kn">import</span> <span class="n">is_ray_available</span><span class="p">,</span> <span class="n">is_vllm_available</span><span class="p">,</span> <span class="n">is_flash_attn_available</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<span class="k">if</span> <span class="n">is_flash_attn_available</span><span class="p">():</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">flash_attn</span>
+<span class="k">else</span><span class="p">:</span>
+    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;Consider install flash_attn for better performance.&quot;</span><span class="p">)</span>
+    
+<span class="k">if</span> <span class="n">is_vllm_available</span><span class="p">():</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">vllm</span><span class="w"> </span><span class="kn">import</span> <span class="n">SamplingParams</span>
+    
+<span class="k">if</span> <span class="n">is_ray_available</span><span class="p">():</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">ray</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">ray.data</span>
+
+
+<div class="viewcode-block" id="HFDecoderModel">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">HFDecoderModel</span><span class="p">(</span><span class="n">DecoderModel</span><span class="p">,</span> <span class="n">HFModelMixin</span><span class="p">,</span> <span class="n">Tunable</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Initializes a HFDecoderModel instance.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+
+<span class="sd">    model_args : </span>
+<span class="sd">        Model arguments such as model name, path, revision, etc.</span>
+
+<span class="sd">    tune_strategy : str or none,  default=&quot;normal&quot;.</span>
+<span class="sd">        A string representing the dataset backend. Defaults to &quot;huggingface&quot;.</span>
+<span class="sd">    </span>
+<span class="sd">    ds_config :   </span>
+<span class="sd">        Deepspeed configuations.</span>
+<span class="sd">    </span>
+<span class="sd">    args : Optional.</span>
+<span class="sd">        Positional arguments.</span>
+<span class="sd">    </span>
+<span class="sd">    kwargs : Optional.</span>
+<span class="sd">        Keyword arguments.    </span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">,</span>
+        <span class="n">tune_strategy</span><span class="o">=</span><span class="s1">&#39;normal&#39;</span><span class="p">,</span>
+        <span class="n">ds_config</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">device</span><span class="o">=</span><span class="s2">&quot;gpu&quot;</span><span class="p">,</span>
+        <span class="n">use_accelerator</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Initializes a HFDecoderModel instance.</span>
+<span class="sd">        :param model_args: dictionary with model arguments such as model name, path, revision, etc.</span>
+<span class="sd">        :param tune_strategy: tuning strategy: normal, none, lora or adapter</span>
+<span class="sd">        :param ds_config: deepspeed configuration for distributed training</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">HFModelMixin</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="bp">self</span><span class="p">,</span>
+            <span class="n">model_args</span><span class="o">=</span><span class="n">model_args</span><span class="p">,</span>
+            <span class="n">do_train</span><span class="o">=</span><span class="kc">True</span> <span class="k">if</span> <span class="n">tune_strategy</span> <span class="o">==</span> <span class="s2">&quot;normal&quot;</span> <span class="k">else</span> <span class="kc">False</span><span class="p">,</span>
+            <span class="n">ds_config</span><span class="o">=</span><span class="n">ds_config</span><span class="p">,</span>
+            <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span>
+            <span class="n">use_accelerator</span><span class="o">=</span><span class="n">use_accelerator</span><span class="p">,</span>
+            <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+            <span class="o">**</span><span class="n">kwargs</span>
+        <span class="p">)</span>
+
+
+<div class="viewcode-block" id="HFDecoderModel.tokenize">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.tokenize">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">tokenize</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">dataset</span><span class="p">,</span> 
+        <span class="n">add_special_tokens</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> 
+        <span class="o">*</span><span class="n">args</span><span class="p">,</span> 
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dataset</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Tokenize the full dataset.</span>
+<span class="sd">    </span>
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        dataset : lmflow.datasets.Dataset.</span>
+
+<span class="sd">        args : Optional.</span>
+<span class="sd">            Positional arguments.</span>
+<span class="sd">        </span>
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.    </span>
+<span class="sd">        </span>
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        tokenized_datasets :</span>
+<span class="sd">            The tokenized dataset, without any leading or trailing special</span>
+<span class="sd">            tokens (normally they are Begin-Of-Sentence or End-Of-Sentence</span>
+<span class="sd">            tokens).</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># Preprocessing the datasets.</span>
+        <span class="c1"># First we tokenize all the texts.</span>
+        <span class="k">if</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_backend</span><span class="p">()</span> <span class="o">!=</span> <span class="s2">&quot;huggingface&quot;</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="s2">&quot;tokenization of datasets with non-huggingface backend are&quot;</span>
+                <span class="s2">&quot;not supported yet&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">dataset_type</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span>
+        <span class="n">model_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span>
+        <span class="n">raw_datasets</span> <span class="o">=</span> <span class="n">dataset</span>
+        <span class="n">hf_raw_datasets</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()</span>
+        <span class="n">column_names</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">hf_raw_datasets</span><span class="o">.</span><span class="n">features</span><span class="p">)</span>
+        <span class="n">data_args</span> <span class="o">=</span> <span class="n">raw_datasets</span><span class="o">.</span><span class="n">get_data_args</span><span class="p">()</span>
+
+        <span class="c1"># Requires three types of information for tokenizing different datasets</span>
+        <span class="c1">#   1) Which fields require tokenization, e.g.</span>
+        <span class="c1">#        &quot;text2float&quot;: &quot;text&quot;, but not &quot;float&quot;</span>
+        <span class="c1">#        &quot;text2text&quot;: both &quot;input&quot; and &quot;output&quot;</span>
+        <span class="c1">#   2) How will there tokenized sequence concatenated together, e.g.</span>
+        <span class="c1">#        &quot;text_only&quot;: &quot;text&quot; -&gt; &quot;text&quot;</span>
+        <span class="c1">#        &quot;text2text&quot;: &quot;input&quot;, &quot;output&quot; -&gt; &quot;input&quot; + &quot;output&quot;</span>
+        <span class="c1">#   3) Which fields require loss in final computation, e.g.</span>
+        <span class="c1">#        &quot;text_only&quot;: &quot;text&quot;</span>
+        <span class="c1">#        &quot;text2text&quot;: &quot;output&quot; only</span>
+        <span class="n">tokenized_column_order</span> <span class="o">=</span> <span class="kc">None</span>       <span class="c1"># Handles 1) and 2)</span>
+        <span class="n">label_columns</span> <span class="o">=</span> <span class="kc">None</span>                <span class="c1"># Handles 3)</span>
+        <span class="k">if</span> <span class="n">dataset_type</span> <span class="o">==</span> <span class="s2">&quot;text_only&quot;</span><span class="p">:</span>
+            <span class="n">tokenized_column_order</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;text&quot;</span><span class="p">]</span>
+            <span class="n">label_columns</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;text&quot;</span><span class="p">]</span>
+        <span class="k">elif</span> <span class="n">dataset_type</span> <span class="o">==</span> <span class="s2">&quot;text2text&quot;</span><span class="p">:</span>
+            <span class="n">tokenized_column_order</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">,</span> <span class="s2">&quot;output&quot;</span><span class="p">]</span>
+            <span class="n">label_columns</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">]</span>
+            <span class="n">add_special_tokens</span> <span class="o">=</span> <span class="kc">False</span>
+        <span class="k">elif</span> <span class="n">dataset_type</span> <span class="o">==</span> <span class="s2">&quot;conversation&quot;</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">conversation_template</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">conversation_template</span> <span class="ow">in</span> <span class="n">PRESET_TEMPLATES</span><span class="o">.</span><span class="n">keys</span><span class="p">():</span>
+                    <span class="n">conversation_template</span> <span class="o">=</span> <span class="n">PRESET_TEMPLATES</span><span class="p">[</span><span class="n">data_args</span><span class="o">.</span><span class="n">conversation_template</span><span class="p">]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;Conversation template </span><span class="si">{</span><span class="n">data_args</span><span class="o">.</span><span class="n">conversation_template</span><span class="si">}</span><span class="s2"> is not supported yet.&quot;</span>
+                    <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;No conversation template provided. Using default template.&quot;</span><span class="p">)</span>
+                <span class="n">conversation_template</span> <span class="o">=</span> <span class="n">PRESET_TEMPLATES</span><span class="p">[</span><span class="s1">&#39;empty&#39;</span><span class="p">]</span>
+                        
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Conversation template: </span><span class="si">{</span><span class="n">conversation_template</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;dataset type </span><span class="se">\&quot;</span><span class="si">{</span><span class="n">dataset_type</span><span class="si">}</span><span class="se">\&quot;</span><span class="s2"> is not supported, currently&quot;</span>
+                <span class="s2">&quot; only support following data types:</span><span class="se">\n</span><span class="s2">&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;    1) </span><span class="si">{</span><span class="n">TEXT_ONLY_DATASET_DESCRIPTION</span><span class="si">}</span><span class="se">\n</span><span class="s2">&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;    2) </span><span class="si">{</span><span class="n">TEXT2TEXT_DATASET_DESCRIPTION</span><span class="si">}</span><span class="se">\n</span><span class="s2">&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;    3) </span><span class="si">{</span><span class="n">CONVERSATION_DATASET_DESCRIPTION</span><span class="si">}</span><span class="se">\n</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="c1"># Whether to truncate long sequences to fit into max_length</span>
+        <span class="n">use_truncation</span> <span class="o">=</span> <span class="kc">False</span>
+        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_lora</span> <span class="ow">or</span> <span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+            <span class="n">use_truncation</span> <span class="o">=</span> <span class="kc">True</span>
+        
+        <span class="n">tokenize_fn</span> <span class="o">=</span> <span class="n">conversation_tokenize_function</span> <span class="k">if</span> <span class="s2">&quot;conversation&quot;</span> <span class="ow">in</span> <span class="n">dataset_type</span> <span class="k">else</span> <span class="n">tokenize_function</span>
+        <span class="n">tokenize_fn_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;data_args&quot;</span><span class="p">:</span> <span class="n">data_args</span><span class="p">,</span>
+            <span class="s2">&quot;tokenizer&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">,</span>
+            <span class="s2">&quot;column_names&quot;</span><span class="p">:</span> <span class="n">column_names</span><span class="p">,</span>
+        <span class="p">}</span>
+        <span class="k">if</span> <span class="s2">&quot;conversation&quot;</span> <span class="ow">in</span> <span class="n">dataset_type</span><span class="p">:</span>
+            <span class="n">tokenize_fn_kwargs</span><span class="p">[</span><span class="s2">&quot;conversation_template&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">conversation_template</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">tokenize_fn_kwargs</span><span class="p">[</span><span class="s2">&quot;label_columns&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">label_columns</span>
+            <span class="n">tokenize_fn_kwargs</span><span class="p">[</span><span class="s2">&quot;tokenized_column_order&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">tokenized_column_order</span>
+            <span class="n">tokenize_fn_kwargs</span><span class="p">[</span><span class="s2">&quot;add_special_tokens&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">add_special_tokens</span>
+            <span class="n">tokenize_fn_kwargs</span><span class="p">[</span><span class="s2">&quot;use_truncation&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">use_truncation</span>
+                           
+        <span class="n">tokenize_kwargs</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">data_args</span><span class="o">.</span><span class="n">streaming</span><span class="p">:</span>
+            <span class="n">fingerprint</span> <span class="o">=</span> <span class="n">hashlib</span><span class="o">.</span><span class="n">md5</span><span class="p">(</span>
+                <span class="p">(</span>
+                    <span class="n">raw_datasets</span><span class="o">.</span><span class="n">get_fingerprint</span><span class="p">()</span>
+                    <span class="o">+</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">)</span>
+                    <span class="o">+</span> <span class="sa">f</span><span class="s1">&#39;###padding_side=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">padding_side</span><span class="si">}</span><span class="s1">&#39;</span>
+                    <span class="o">+</span> <span class="p">(</span><span class="s1">&#39;###conversation_template=&#39;</span> <span class="o">+</span> <span class="nb">str</span><span class="p">(</span><span class="n">conversation_template</span><span class="p">)</span> <span class="k">if</span> <span class="s2">&quot;conversation&quot;</span> <span class="ow">in</span> <span class="n">dataset_type</span> <span class="k">else</span> <span class="s2">&quot;&quot;</span><span class="p">)</span>
+                    <span class="o">+</span> <span class="sa">f</span><span class="s1">&#39;###disable_group_texts=</span><span class="si">{</span><span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="si">}</span><span class="s1">&#39;</span>
+                    <span class="o">+</span> <span class="sa">f</span><span class="s1">&#39;###block_size=</span><span class="si">{</span><span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span><span class="si">}</span><span class="s1">&#39;</span>
+                <span class="p">)</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="s2">&quot;utf-8&quot;</span><span class="p">)</span>
+            <span class="p">)</span><span class="o">.</span><span class="n">hexdigest</span><span class="p">()</span>
+            <span class="n">tokenize_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="s2">&quot;num_proc&quot;</span><span class="p">:</span> <span class="n">data_args</span><span class="o">.</span><span class="n">preprocessing_num_workers</span><span class="p">,</span>
+                <span class="s2">&quot;load_from_cache_file&quot;</span><span class="p">:</span> <span class="ow">not</span> <span class="n">data_args</span><span class="o">.</span><span class="n">overwrite_cache</span><span class="p">,</span>
+                <span class="s2">&quot;desc&quot;</span><span class="p">:</span> <span class="s2">&quot;Running tokenizer on dataset&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;new_fingerprint&quot;</span><span class="p">:</span> <span class="n">fingerprint</span><span class="p">,</span>
+            <span class="p">}</span>
+
+        <span class="n">tokenized_datasets</span> <span class="o">=</span> <span class="n">raw_datasets</span><span class="o">.</span><span class="n">map</span><span class="p">(</span>
+            <span class="n">tokenize_fn</span><span class="p">,</span>
+            <span class="n">batched</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+            <span class="n">remove_columns</span><span class="o">=</span><span class="n">column_names</span><span class="p">,</span>
+            <span class="n">fn_kwargs</span><span class="o">=</span><span class="n">tokenize_fn_kwargs</span><span class="p">,</span>
+            <span class="o">**</span><span class="n">tokenize_kwargs</span>
+        <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">tokenized_datasets</span></div>
+
+
+
+<div class="viewcode-block" id="HFDecoderModel.encode">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.encode">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">encode</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]],</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span> <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform encoding process of the tokenizer.</span>
+<span class="sd">    </span>
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        inputs : str or list.</span>
+<span class="sd">            The text sequence.</span>
+<span class="sd">            </span>
+<span class="sd">        args : Optional.</span>
+<span class="sd">            Positional arguments.</span>
+<span class="sd">        </span>
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.    </span>
+<span class="sd">        </span>
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        outputs :</span>
+<span class="sd">            if string input,return the tokenized inputs.</span>
+<span class="sd">            &quot;Hello,world!&quot;-&gt; [101, 7592, 1010, 2088, 102]</span>
+<span class="sd">            if batch input,return {input_ids,attention_mask,token_type_ids}</span>
+<span class="sd">            [&quot;Hello,world!&quot;,&quot;Hello!&quot;]-&gt; {&#39;input_ids&#39;: tensor([[  101,  7592,  1010,  2088,   102],...),&#39;attention_mask&#39;: tensor([[1, 1, 1, 1, 1],[0,0,1,1,1]])}</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">(</span><span class="n">text</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span><span class="c1">#batch encode,will automatically do left padding</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="n">text</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;type &quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span><span class="si">}</span><span class="s1">&quot; cannot be encoded&#39;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="HFDecoderModel.decode">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.decode">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">decode</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span> <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform decoding process of the tokenizer.</span>
+<span class="sd">    </span>
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        inputs : list or tensor.</span>
+<span class="sd">            The token sequence.</span>
+<span class="sd">            </span>
+<span class="sd">        args : Optional.</span>
+<span class="sd">            Positional arguments.</span>
+<span class="sd">        </span>
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.    </span>
+<span class="sd">        </span>
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        outputs :</span>
+<span class="sd">            The text decoded from the token inputs.</span>
+<span class="sd">            if batch input,return the list of text</span>
+<span class="sd">            [[101, 7592, 1010, 2088, 102],[101, 7592, 1010, 2088, 102]]-&gt; [&quot;Hello,world!&quot;,&quot;Hello,world!&quot;</span>
+<span class="sd">            if single input,return the text</span>
+<span class="sd">            [101, 7592, 1010, 2088, 102]-&gt; &quot;Hello,world!&quot;</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">List</span><span class="p">):</span>
+            <span class="nb">input</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">input</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span><span class="o">==</span><span class="mi">2</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">batch_decode</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span><span class="c1">#batch_decode</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># Can be list of ints or a Tensor</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+        
+<div class="viewcode-block" id="HFDecoderModel.inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">inputs</span><span class="p">,</span> 
+        <span class="n">release_gpu</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">use_vllm</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform generation process of the model.</span>
+<span class="sd">    </span>
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        inputs :</span>
+<span class="sd">            The sequence used as a prompt for the generation or as model inputs to the model.</span>
+<span class="sd">            When using vllm inference, this should be a string or a list of strings.</span>
+<span class="sd">            When using normal inference, this should be a tensor.</span>
+<span class="sd">        release_gpu : bool, optional</span>
+<span class="sd">            Whether to release the GPU resource after inference, by default False.</span>
+<span class="sd">        use_vllm : bool, optional</span>
+<span class="sd">            Whether to use VLLM for inference, by default False.</span>
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.    </span>
+<span class="sd">        </span>
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        outputs :</span>
+<span class="sd">            The generated sequence output </span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_activated</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">activate_model_for_inference</span><span class="p">(</span>
+                <span class="n">use_vllm</span><span class="o">=</span><span class="n">use_vllm</span><span class="p">,</span>
+                <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+            <span class="p">)</span>
+            
+        <span class="k">if</span> <span class="n">use_vllm</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">is_vllm_available</span><span class="p">():</span>
+                <span class="k">raise</span> <span class="ne">ImportError</span><span class="p">(</span><span class="s2">&quot;vllm is not installed. Please install vllm to use VLLM inference.&quot;</span><span class="p">)</span>
+            <span class="n">res</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__vllm_inference</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">res</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__inference</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+            
+        <span class="k">if</span> <span class="n">release_gpu</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">deactivate_model_for_inference</span><span class="p">(</span><span class="n">use_vllm</span><span class="o">=</span><span class="n">use_vllm</span><span class="p">)</span>
+            
+        <span class="k">return</span> <span class="n">res</span></div>
+
+
+
+<div class="viewcode-block" id="HFDecoderModel.__inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.__inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__inference</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform generation process of the model.</span>
+<span class="sd">    </span>
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        inputs :</span>
+<span class="sd">            The **tokenized** sequence used as a prompt for the generation or as model inputs to the model.</span>
+<span class="sd">            </span>
+<span class="sd">        args : Optional.</span>
+<span class="sd">            Positional arguments.</span>
+<span class="sd">        </span>
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.    </span>
+<span class="sd">        </span>
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        outputs :</span>
+<span class="sd">            The generated sequence output </span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_accelerator</span><span class="p">:</span>
+                <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">generate</span><span class="p">(</span>
+                    <span class="n">input_ids</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span>
+                    <span class="n">pad_token_id</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span><span class="p">,</span>
+                    <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+                    <span class="o">**</span><span class="n">kwargs</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;gpu&quot;</span><span class="p">:</span>
+                    <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ds_engine</span><span class="o">.</span><span class="n">module</span><span class="o">.</span><span class="n">generate</span><span class="p">(</span>
+                        <span class="n">input_ids</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span>
+                        <span class="n">synced_gpus</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                        <span class="n">pad_token_id</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span><span class="p">,</span>
+                        <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+                        <span class="o">**</span><span class="n">kwargs</span>
+                    <span class="p">)</span>
+                <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;cpu&quot;</span><span class="p">:</span>
+                    <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">generate</span><span class="p">(</span>
+                        <span class="n">input_ids</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span>
+                        <span class="n">synced_gpus</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                        <span class="n">pad_token_id</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span><span class="p">,</span>
+                        <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+                        <span class="o">**</span><span class="n">kwargs</span>
+                    <span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;device </span><span class="se">\&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="se">\&quot;</span><span class="s2"> is not supported&quot;</span>
+                    <span class="p">)</span>
+        <span class="k">return</span> <span class="n">outputs</span></div>
+
+    
+    
+<div class="viewcode-block" id="HFDecoderModel.__vllm_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.__vllm_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__vllm_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">inputs</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]],</span>
+        <span class="n">sampling_params</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="s1">&#39;SamplingParams&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">VLLMInferenceResultWithInput</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Perform VLLM inference process of the model.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ----------</span>
+<span class="sd">        inputs : Union[str, List[str]]</span>
+<span class="sd">            Prompt(s), string or a list of strings.</span>
+<span class="sd">        sampling_params : Optional[SamplingParams], optional</span>
+<span class="sd">            vllm SamplingParams object, by default None.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        -------</span>
+<span class="sd">        List[VLLMInferenceResultWithInput]</span>
+<span class="sd">            Return a list of VLLMInferenceResultWithInput, where each</span>
+<span class="sd">            element contains the input prompt and the corresponding output.</span>
+<span class="sd">            </span>
+<span class="sd">            When `sampling_params.detokenize = True`, the output would be a list of strings,</span>
+<span class="sd">            contains sampling_params.n samples for the corresponding prompt.</span>
+<span class="sd">            </span>
+<span class="sd">            When `sampling_params.detokenize = False`, return a list of list of ints </span>
+<span class="sd">            (token ids, no decoding after generation).</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">vllm_outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model_for_inference</span><span class="o">.</span><span class="n">generate</span><span class="p">(</span>
+            <span class="n">inputs</span><span class="p">,</span>
+            <span class="n">sampling_params</span><span class="o">=</span><span class="n">sampling_params</span><span class="p">,</span>
+            <span class="n">use_tqdm</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">final_output</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">output</span> <span class="ow">in</span> <span class="n">vllm_outputs</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">sampling_params</span><span class="o">.</span><span class="n">detokenize</span><span class="p">:</span>
+                <span class="n">output_list</span> <span class="o">=</span> <span class="p">[</span><span class="n">sentence</span><span class="o">.</span><span class="n">text</span> <span class="k">for</span> <span class="n">sentence</span> <span class="ow">in</span> <span class="n">output</span><span class="o">.</span><span class="n">outputs</span><span class="p">]</span>  
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">output_list</span> <span class="o">=</span> <span class="p">[</span><span class="n">sentence</span><span class="o">.</span><span class="n">token_ids</span> <span class="k">for</span> <span class="n">sentence</span> <span class="ow">in</span> <span class="n">output</span><span class="o">.</span><span class="n">outputs</span><span class="p">]</span>
+                
+            <span class="n">final_output</span><span class="o">.</span><span class="n">append</span><span class="p">({</span><span class="s2">&quot;input&quot;</span><span class="p">:</span> <span class="n">output</span><span class="o">.</span><span class="n">prompt</span><span class="p">,</span> <span class="s2">&quot;output&quot;</span><span class="p">:</span> <span class="n">output_list</span><span class="p">})</span>
+                                
+        <span class="k">return</span> <span class="n">final_output</span></div>
+
+    
+    
+<div class="viewcode-block" id="HFDecoderModel.prepare_inputs_for_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.prepare_inputs_for_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">prepare_inputs_for_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">apply_chat_template</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">enable_distributed_inference</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">use_vllm</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="s2">&quot;ray.data.Dataset&quot;</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Prepare inputs for inference.</span>
+<span class="sd">    </span>
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        dataset : lmflow.datasets.Dataset.</span>
+<span class="sd">            The dataset used for inference.</span>
+<span class="sd">            </span>
+<span class="sd">        args : Optional.</span>
+<span class="sd">            Positional arguments.</span>
+<span class="sd">        </span>
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.    </span>
+<span class="sd">        </span>
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        outputs :</span>
+<span class="sd">            The prepared inputs for inference.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">use_vllm</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">is_ray_available</span><span class="p">()</span> <span class="ow">and</span> <span class="n">enable_distributed_inference</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ImportError</span><span class="p">(</span><span class="s2">&quot;ray is not installed. Please install ray to use distributed vllm inference.&quot;</span><span class="p">)</span>
+            <span class="n">inference_inputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__prepare_inputs_for_vllm_inference</span><span class="p">(</span>
+                <span class="n">dataset</span><span class="o">=</span><span class="n">dataset</span><span class="p">,</span> 
+                <span class="n">apply_chat_template</span><span class="o">=</span><span class="n">apply_chat_template</span><span class="p">,</span>
+                <span class="n">enable_distributed_inference</span><span class="o">=</span><span class="n">enable_distributed_inference</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">inference_inputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__prepare_inputs_for_inference</span><span class="p">(</span>
+                <span class="n">dataset</span><span class="p">,</span>
+                <span class="n">apply_chat_template</span><span class="o">=</span><span class="n">apply_chat_template</span><span class="p">,</span>
+                <span class="n">enable_distributed_inference</span><span class="o">=</span><span class="n">enable_distributed_inference</span><span class="p">,</span>
+            <span class="p">)</span>
+            
+        <span class="k">return</span> <span class="n">inference_inputs</span></div>
+
+    
+    
+<div class="viewcode-block" id="HFDecoderModel.__prepare_inputs_for_vllm_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.__prepare_inputs_for_vllm_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__prepare_inputs_for_vllm_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">apply_chat_template</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">enable_distributed_inference</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="s2">&quot;ray.data.Dataset&quot;</span><span class="p">]:</span>
+        <span class="k">if</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="o">==</span> <span class="s1">&#39;text_only&#39;</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">apply_chat_template</span><span class="p">:</span>
+                <span class="n">dataset</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">map</span><span class="p">(</span>
+                    <span class="k">lambda</span> <span class="n">sample</span><span class="p">:</span> <span class="p">{</span>
+                        <span class="s2">&quot;templated&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">apply_chat_template</span><span class="p">(</span>
+                            <span class="p">[{</span><span class="s2">&quot;role&quot;</span><span class="p">:</span><span class="s2">&quot;user&quot;</span><span class="p">,</span> <span class="s2">&quot;content&quot;</span><span class="p">:</span> <span class="n">sample</span><span class="p">[</span><span class="s1">&#39;text&#39;</span><span class="p">]}],</span> 
+                            <span class="n">tokenize</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> 
+                            <span class="n">add_generation_prompt</span><span class="o">=</span><span class="kc">True</span>
+                        <span class="p">)</span>
+                    <span class="p">},</span>
+                    <span class="n">num_proc</span><span class="o">=</span><span class="n">dataset</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">preprocessing_num_workers</span><span class="p">,</span>
+                <span class="p">)</span>
+                <span class="n">inference_inputs</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()[</span><span class="s1">&#39;templated&#39;</span><span class="p">]</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">inference_inputs</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()[</span><span class="s1">&#39;text&#39;</span><span class="p">]</span>
+            
+        <span class="k">elif</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="o">==</span> <span class="s2">&quot;text2text&quot;</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;For a text2text dataset, only `input` will be used as the model input.&quot;</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">apply_chat_template</span><span class="p">:</span>
+                <span class="n">dataset</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">map</span><span class="p">(</span>
+                    <span class="k">lambda</span> <span class="n">sample</span><span class="p">:</span> <span class="p">{</span>
+                        <span class="s2">&quot;templated&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">apply_chat_template</span><span class="p">(</span>
+                            <span class="n">conversation</span><span class="o">=</span><span class="p">[{</span><span class="s2">&quot;role&quot;</span><span class="p">:</span><span class="s2">&quot;user&quot;</span><span class="p">,</span> <span class="s2">&quot;content&quot;</span><span class="p">:</span> <span class="n">sample</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">]}],</span> 
+                            <span class="n">tokenize</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> 
+                            <span class="n">add_generation_prompt</span><span class="o">=</span><span class="kc">True</span>
+                        <span class="p">)</span>
+                    <span class="p">},</span>
+                    <span class="n">num_proc</span><span class="o">=</span><span class="n">dataset</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">preprocessing_num_workers</span><span class="p">,</span>
+                <span class="p">)</span>
+                <span class="n">inference_inputs</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()[</span><span class="s1">&#39;templated&#39;</span><span class="p">]</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">inference_inputs</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()[</span><span class="s1">&#39;input&#39;</span><span class="p">]</span>
+            
+        <span class="k">elif</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="o">==</span> <span class="s1">&#39;conversation&#39;</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">apply_chat_template</span><span class="p">:</span>
+                <span class="k">def</span><span class="w"> </span><span class="nf">preprocess_conversation</span><span class="p">(</span><span class="n">sample</span><span class="p">):</span>
+                    <span class="n">conversation</span> <span class="o">=</span> <span class="n">sample</span><span class="p">[</span><span class="s1">&#39;messages&#39;</span><span class="p">][:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">sample</span><span class="p">[</span><span class="s1">&#39;messages&#39;</span><span class="p">])</span><span class="o">%</span><span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="n">sample</span><span class="p">[</span><span class="s1">&#39;messages&#39;</span><span class="p">]</span>
+                        
+                    <span class="k">if</span> <span class="n">sample</span><span class="p">[</span><span class="s1">&#39;messages&#39;</span><span class="p">][</span><span class="o">-</span><span class="mi">1</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">!=</span> <span class="s1">&#39;user&#39;</span><span class="p">:</span>
+                        <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                            <span class="s2">&quot;Not a valid conversation for generation, since the conversation &quot;</span>
+                            <span class="s2">&quot;doesn&#39;t end up with an user message. Skip.&quot;</span>
+                        <span class="p">)</span>
+                        <span class="n">sample_out</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;templated&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;</span><span class="p">}</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">sample_out</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;templated&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">apply_chat_template</span><span class="p">(</span>
+                            <span class="n">conversation</span><span class="o">=</span><span class="n">conversation</span><span class="p">,</span>
+                            <span class="n">tokenize</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                            <span class="n">add_generation_prompt</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                        <span class="p">)}</span>
+                        
+                    <span class="k">return</span> <span class="n">sample_out</span>
+                <span class="n">dataset</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">map</span><span class="p">(</span>
+                    <span class="n">preprocess_conversation</span><span class="p">,</span>
+                    <span class="n">num_proc</span><span class="o">=</span><span class="n">dataset</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">preprocessing_num_workers</span><span class="p">,</span>
+                <span class="p">)</span>
+                <span class="n">inference_inputs</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()[</span><span class="s1">&#39;templated&#39;</span><span class="p">]</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="s2">&quot;Your dataset is `conversation` type but `apply_chat_template` is set to False. &quot;</span>
+                    <span class="s2">&quot;Will use the first user input in conversation as model input.&quot;</span>
+                <span class="p">)</span>
+                <span class="n">inference_inputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">conversation</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="k">for</span> <span class="n">conversation</span> <span class="ow">in</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()[</span><span class="s1">&#39;messages&#39;</span><span class="p">]]</span>
+
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Currently `</span><span class="si">{</span><span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span><span class="si">}</span><span class="s2">` data are not supported for vllm inference.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">inference_inputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">sentence</span> <span class="k">for</span> <span class="n">sentence</span> <span class="ow">in</span> <span class="n">inference_inputs</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">sentence</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">]</span>
+        
+        <span class="k">if</span> <span class="n">enable_distributed_inference</span><span class="p">:</span>
+            <span class="n">inference_inputs</span> <span class="o">=</span> <span class="n">ray</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">from_items</span><span class="p">(</span><span class="n">inference_inputs</span><span class="p">)</span> <span class="c1"># -&gt; Dict[str, np.ndarray], {&quot;item&quot;: array([&#39;...&#39;, &#39;...&#39;, &#39;...&#39;])}</span>
+        
+        <span class="k">return</span> <span class="n">inference_inputs</span></div>
+
+    
+    
+<div class="viewcode-block" id="HFDecoderModel.__prepare_inputs_for_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.__prepare_inputs_for_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__prepare_inputs_for_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">&quot;prepare_inputs_for_inference is not implemented&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="HFDecoderModel.merge_lora_weights">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.merge_lora_weights">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">merge_lora_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">use_lora</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">use_qlora</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">merge_and_unload</span><span class="p">()</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">use_qlora</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;Reloading base model in 16-bit precision to merge adapter weights. NOTE: Your device must have&quot;</span>
+                           <span class="s2">&quot;sufficient memory to reload the model in half-precision without quantization.&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">get_peft_without_qlora</span><span class="p">()</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">merge_and_unload</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;LoRA training is NOT enabled. Merging LoRA weights is not applicable.&quot;</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="HFDecoderModel.get_peft_without_qlora">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.get_peft_without_qlora">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_peft_without_qlora</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="kn">import</span><span class="w"> </span><span class="nn">tempfile</span>
+
+        <span class="k">with</span> <span class="n">tempfile</span><span class="o">.</span><span class="n">TemporaryDirectory</span><span class="p">()</span> <span class="k">as</span> <span class="n">tmpdirname</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s1">&#39;created temporary directory&#39;</span><span class="p">,</span> <span class="n">tmpdirname</span><span class="p">)</span>
+
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="n">tmpdirname</span><span class="p">)</span>
+
+            <span class="n">torch_dtype</span> <span class="o">=</span> <span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">torch_dtype</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">torch_dtype</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;auto&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span>
+                <span class="k">else</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">torch</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">torch_dtype</span><span class="p">)</span>
+            <span class="p">)</span>
+            <span class="n">config_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="s2">&quot;cache_dir&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">cache_dir</span><span class="p">,</span>
+                <span class="s2">&quot;revision&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_revision</span><span class="p">,</span>
+                <span class="s2">&quot;token&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">token</span><span class="p">,</span>
+            <span class="p">}</span>
+            <span class="n">config</span> <span class="o">=</span> <span class="n">AutoConfig</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span> <span class="o">**</span><span class="n">config_kwargs</span><span class="p">)</span>
+            <span class="n">device_map</span> <span class="o">=</span> <span class="s2">&quot;auto&quot;</span>
+            <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">&#39;LOCAL_RANK&#39;</span><span class="p">)</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">local_rank</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">&#39;LOCAL_RANK&#39;</span><span class="p">,</span><span class="s1">&#39;0&#39;</span><span class="p">))</span>
+                <span class="n">device_map</span> <span class="o">=</span> <span class="p">{</span><span class="s1">&#39;&#39;</span><span class="p">:</span> <span class="n">local_rank</span><span class="p">}</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_model_full</span> <span class="o">=</span> <span class="n">AutoModelForCausalLM</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span>
+                <span class="n">from_tf</span><span class="o">=</span><span class="nb">bool</span><span class="p">(</span><span class="s2">&quot;.ckpt&quot;</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">),</span>
+                <span class="n">config</span><span class="o">=</span><span class="n">config</span><span class="p">,</span>
+                <span class="n">cache_dir</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">cache_dir</span><span class="p">,</span>
+                <span class="n">revision</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_revision</span><span class="p">,</span>
+                <span class="n">token</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">token</span><span class="p">,</span>
+                <span class="n">torch_dtype</span><span class="o">=</span><span class="n">torch_dtype</span><span class="p">,</span>
+                <span class="n">device_map</span><span class="o">=</span><span class="n">device_map</span><span class="p">,</span>
+                <span class="n">trust_remote_code</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">trust_remote_code</span><span class="p">,</span>
+                <span class="n">attn_implementation</span><span class="o">=</span><span class="s2">&quot;flash_attention_2&quot;</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">use_flash_attention</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="p">)</span>
+        
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span> <span class="o">=</span> <span class="n">PeftModel</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">backend_model_full</span><span class="p">,</span> <span class="n">tmpdirname</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="HFDecoderModel.save">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.save">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">save</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">dir</span><span class="p">,</span> <span class="n">save_full_model</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform generation process of the model.</span>
+<span class="sd">    </span>
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        dir :</span>
+<span class="sd">            The directory to save model and tokenizer</span>
+<span class="sd">            </span>
+<span class="sd">        save_full_model : Optional.</span>
+<span class="sd">            Whether to save full model.</span>
+<span class="sd">        </span>
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.    </span>
+<span class="sd">        </span>
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        outputs :</span>
+<span class="sd">            The generated sequence output </span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">get_tokenizer</span><span class="p">()</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="nb">dir</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">save_full_model</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">use_lora</span><span class="p">:</span>
+            <span class="n">save_dtype</span> <span class="o">=</span> <span class="p">(</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">float16</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">torch_dtype</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;auto&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span>
+                <span class="k">else</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">torch</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">torch_dtype</span><span class="p">)</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_model_full</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">save_dtype</span><span class="p">)</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="nb">dir</span><span class="p">)</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Save full model with dtype: </span><span class="si">{</span><span class="n">save_dtype</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="nb">dir</span><span class="p">)</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/models/hf_encoder_decoder_model.html b/_modules/lmflow/models/hf_encoder_decoder_model.html
new file mode 100644
index 000000000..b3b3cb827
--- /dev/null
+++ b/_modules/lmflow/models/hf_encoder_decoder_model.html
@@ -0,0 +1,999 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.models.hf_encoder_decoder_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/models/hf_encoder_decoder_model';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.hf_encoder_decoder_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.models.hf_encoder_decoder_model</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;This is a class called HFDecoderModel which is a wrapper around transformers model and</span>
+<span class="sd">tokenizer classes. It has several methods such as __init__, tokenize, and train that are</span>
+<span class="sd">used for training and fine-tuning the model. The __init__ method takes in several arguments</span>
+<span class="sd">such as model_args, tune_strategy, and ds_config, which are used to load the pretrained</span>
+<span class="sd">model and tokenizer, and initialize the training settings.</span>
+
+<span class="sd">The tokenize method is used to tokenize the input text and return the input IDs and attention</span>
+<span class="sd">masks that can be fed to the model for training or inference.</span>
+
+<span class="sd">This class supports different tune_strategy options such as &#39;normal&#39;, &#39;none&#39;, &#39;lora&#39;, and</span>
+<span class="sd">&#39;adapter&#39;, which allow for different fine-tuning settings of the model. However, the &#39;lora&#39;</span>
+<span class="sd">and &#39;adapter&#39; strategies are not yet implemented.</span>
+
+<span class="sd">Overall, this class provides a convenient interface for loading and fine-tuning transformer</span>
+<span class="sd">models and can be used for various NLP tasks such as language modeling, text classification,</span>
+<span class="sd">and question answering.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">copy</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">time</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">deepspeed</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">peft</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">LoraConfig</span><span class="p">,</span>
+    <span class="n">PeftModel</span><span class="p">,</span>
+    <span class="n">TaskType</span><span class="p">,</span>
+    <span class="n">get_peft_config</span><span class="p">,</span>
+    <span class="n">get_peft_model</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">CONFIG_MAPPING</span><span class="p">,</span>
+    <span class="n">AutoConfig</span><span class="p">,</span>
+    <span class="n">AutoTokenizer</span><span class="p">,</span>
+    <span class="n">AutoModelForSeq2SeqLM</span><span class="p">,</span>
+    <span class="n">AutoModelForVision2Seq</span><span class="p">,</span>
+    <span class="n">AutoModel</span><span class="p">,</span>
+    <span class="n">AutoProcessor</span><span class="p">,</span>
+    <span class="n">LlamaConfig</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.testing_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">CaptureLogger</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets.dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.encoder_decoder_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">EncoderDecoderModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.interfaces.tunable</span><span class="w"> </span><span class="kn">import</span> <span class="n">Tunable</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.vision2seq_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">CustomAutoVision2SeqModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.multimodal</span><span class="w"> </span><span class="kn">import</span> <span class="n">update_custom_config</span><span class="p">,</span> <span class="n">load_llava_pretrain_model</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.versioning</span><span class="w"> </span><span class="kn">import</span> <span class="n">is_package_version_at_least</span>
+
+<span class="k">if</span> <span class="n">is_package_version_at_least</span><span class="p">(</span><span class="s2">&quot;transformers&quot;</span><span class="p">,</span> <span class="s2">&quot;4.46.0&quot;</span><span class="p">):</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.integrations.deepspeed</span><span class="w"> </span><span class="kn">import</span> <span class="n">HfDeepSpeedConfig</span><span class="p">,</span> <span class="n">HfTrainerDeepSpeedConfig</span>
+<span class="k">else</span><span class="p">:</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.deepspeed</span><span class="w"> </span><span class="kn">import</span> <span class="n">HfDeepSpeedConfig</span><span class="p">,</span> <span class="n">HfTrainerDeepSpeedConfig</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="HFEncoderDecoderModel">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">HFEncoderDecoderModel</span><span class="p">(</span><span class="n">EncoderDecoderModel</span><span class="p">,</span> <span class="n">Tunable</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Initializes a HFEncoderDecoderModel instance.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+
+<span class="sd">    model_args :</span>
+<span class="sd">        Model arguments such as model name, path, revision, etc.</span>
+
+<span class="sd">    tune_strategy : str or none,  default=&quot;normal&quot;.</span>
+<span class="sd">        A string representing the dataset backend. Defaults to &quot;huggingface&quot;.</span>
+
+<span class="sd">    ds_config :</span>
+<span class="sd">        Deepspeed configuations.</span>
+
+<span class="sd">    args : Optional.</span>
+<span class="sd">        Positional arguments.</span>
+
+<span class="sd">    kwargs : Optional.</span>
+<span class="sd">        Keyword arguments.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">,</span>
+        <span class="n">tune_strategy</span><span class="o">=</span><span class="s1">&#39;normal&#39;</span><span class="p">,</span>
+        <span class="n">ds_config</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">device</span><span class="o">=</span><span class="s2">&quot;gpu&quot;</span><span class="p">,</span>
+        <span class="n">use_accelerator</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">custom_model</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">with_deepspeed</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">pipeline_args</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Initializes a HFDecoderModel instance.</span>
+<span class="sd">        :param model_args: dictionary with model arguments such as model name, path, revision, etc.</span>
+<span class="sd">        :param tune_strategy: tuning strategy: normal, none, lora or adapter</span>
+<span class="sd">        :param ds_config: deepspeed configuration for distributed training</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="c1"># See more about loading any type of standard or custom dataset (from</span>
+        <span class="c1"># files, python dict, pandas DataFrame, etc) at</span>
+        <span class="c1"># https://huggingface.co/docs/datasets/loading_datasets.html.</span>
+
+        <span class="c1"># Load pretrained model and tokenizer</span>
+        <span class="c1">#</span>
+        <span class="c1"># Distributed training: The .from_pretrained methods guarantee that</span>
+        <span class="c1"># only one local process can concurrently download model &amp; vocab.</span>
+
+<div class="viewcode-block" id="HFEncoderDecoderModel.device">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.device">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">device</span> <span class="o">=</span> <span class="n">device</span></div>
+
+
+        <span class="k">if</span> <span class="n">tune_strategy</span> <span class="o">==</span> <span class="s1">&#39;normal&#39;</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;tune_strategy </span><span class="se">\&quot;</span><span class="si">{</span><span class="n">tune_strategy</span><span class="si">}</span><span class="se">\&quot;</span><span class="s2"> is not supported&quot;</span>
+            <span class="p">)</span>
+        <span class="k">elif</span> <span class="n">tune_strategy</span> <span class="o">==</span> <span class="s1">&#39;none&#39;</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">use_accelerator</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;Currently encoder2decoder model is not supported with accelerator&quot;</span>
+                <span class="p">)</span>
+            <span class="c1"># dschf = HfDeepSpeedConfig(ds_config)</span>
+            <span class="n">dschf</span> <span class="o">=</span> <span class="n">HfTrainerDeepSpeedConfig</span><span class="p">(</span><span class="n">ds_config</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">pipeline_args</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">dschf</span><span class="o">.</span><span class="n">trainer_config_process</span><span class="p">(</span><span class="n">pipeline_args</span><span class="p">)</span>
+            <span class="n">peft_model_id</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">lora_model_path</span>
+            <span class="c1"># NOTE: Currently offload is not supported by llama</span>
+            <span class="k">if</span> <span class="s2">&quot;llama&quot;</span> <span class="ow">in</span> <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span> <span class="ow">and</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_ram_optimized_load</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="s2">&quot;llama does not support RAM optimized load. Automatically&quot;</span>
+                    <span class="s2">&quot; use original load instead.&quot;</span>
+                <span class="p">)</span>
+                <span class="n">model_args</span><span class="o">.</span><span class="n">use_ram_optimized_load</span> <span class="o">=</span> <span class="kc">False</span>
+
+            <span class="c1"># get model register</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">arch_type</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">arch_type</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">arch_type</span> <span class="o">==</span> <span class="s2">&quot;encoder_decoder&quot;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span> <span class="o">==</span> <span class="s1">&#39;THUDM/chatglm-6b&#39;</span><span class="p">:</span>
+                    <span class="n">model_register</span> <span class="o">=</span> <span class="n">AutoModel</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">model_register</span> <span class="o">=</span> <span class="n">AutoModelForSeq2SeqLM</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">arch_type</span> <span class="o">==</span> <span class="s2">&quot;vision_encoder_decoder&quot;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="n">custom_model</span><span class="p">:</span>
+                    <span class="n">model_register</span> <span class="o">=</span> <span class="n">AutoModelForVision2Seq</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">model_register</span> <span class="o">=</span> <span class="n">CustomAutoVision2SeqModel</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">NotImplementedError</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">custom_model</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span> <span class="o">==</span> <span class="s1">&#39;THUDM/chatglm-6b&#39;</span><span class="p">:</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span> <span class="o">=</span> <span class="n">model_register</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span> <span class="n">trust_remote_code</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+                <span class="k">elif</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_ram_optimized_load</span> <span class="ow">and</span> <span class="n">peft_model_id</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">try</span><span class="p">:</span>
+                        <span class="c1"># RAM-optimized load</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span> <span class="o">=</span> <span class="n">model_register</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                            <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span>
+                            <span class="n">device_map</span><span class="o">=</span><span class="s2">&quot;auto&quot;</span><span class="p">,</span>
+                            <span class="n">offload_folder</span><span class="o">=</span><span class="s2">&quot;offload&quot;</span><span class="p">,</span>
+                            <span class="n">offload_state_dict</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                        <span class="p">)</span>
+                    <span class="k">except</span><span class="p">:</span>
+                        <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                            <span class="s2">&quot;Failed to use RAM optimized load. Automatically&quot;</span>
+                            <span class="s2">&quot; use original load instead.&quot;</span>
+                        <span class="p">)</span>
+                        <span class="c1"># Normal load</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span> <span class="o">=</span> <span class="n">model_register</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                            <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span>
+                        <span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="n">peft_model_id</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                            <span class="s2">&quot;LoRA does not support RAM optimized load currently.&quot;</span>
+                            <span class="s2">&quot; Automatically use original load instead.&quot;</span>
+                        <span class="p">)</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span> <span class="o">=</span> <span class="n">model_register</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                        <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span>
+                    <span class="p">)</span>
+            <span class="c1"># else:</span>
+            <span class="c1">#     self.backend_model = model_register.from_pretrained(</span>
+            <span class="c1">#         model_args.model_name_or_path)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">llava_loading</span> <span class="ow">is</span> <span class="kc">False</span><span class="p">:</span>
+                    <span class="c1"># FIXME remove the following from_pretrained code by</span>
+                    <span class="c1"># creating a unified pretrained model.</span>
+                    <span class="n">model</span> <span class="o">=</span> <span class="n">CustomAutoVision2SeqModel</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">llm_model_name_or_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="n">text_config</span> <span class="o">=</span> <span class="n">LlamaConfig</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="n">model_args</span><span class="o">.</span><span class="n">llm_model_name_or_path</span><span class="p">)</span>
+                        <span class="n">model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">text_config</span> <span class="o">=</span> <span class="n">text_config</span>
+                    <span class="n">model</span><span class="o">.</span><span class="n">language_model_from_pretrained</span><span class="p">(</span><span class="n">model_args</span><span class="o">.</span><span class="n">llm_model_name_or_path</span><span class="p">,</span>
+                                                        <span class="n">low_resource</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">low_resource</span><span class="p">)</span>
+                    <span class="n">state_dict</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span>
+                        <span class="n">model_args</span><span class="o">.</span><span class="n">pretrained_language_projection_path</span><span class="p">,</span>
+                        <span class="n">map_location</span><span class="o">=</span><span class="s2">&quot;cpu&quot;</span><span class="p">)</span>
+                    <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">state_dict</span><span class="p">,</span> <span class="n">strict</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">config</span> <span class="o">=</span> <span class="n">AutoConfig</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                        <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">low_resource</span><span class="p">:</span>
+                        <span class="n">kwargs</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span>
+                            <span class="n">torch_dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float16</span><span class="p">,</span>
+                            <span class="n">load_in_8bit</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                            <span class="n">device_map</span><span class="o">=</span><span class="s2">&quot;auto&quot;</span><span class="p">,</span>
+                        <span class="p">)</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="c1"># kwargs = dict(torch_dtype=torch.float16)</span>
+                        <span class="n">kwargs</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">device_map</span><span class="o">=</span><span class="s2">&quot;auto&quot;</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="p">(</span><span class="n">model_args</span><span class="o">.</span><span class="n">image_encoder_name_or_path</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span>
+                        <span class="n">model_args</span><span class="o">.</span><span class="n">qformer_name_or_path</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span>
+                        <span class="n">model_args</span><span class="o">.</span><span class="n">llm_model_name_or_path</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">):</span>
+                        <span class="n">config</span> <span class="o">=</span> <span class="n">AutoConfig</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                            <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">)</span>
+                        <span class="n">model</span> <span class="o">=</span> <span class="n">CustomAutoVision2SeqModel</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                            <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">config</span> <span class="o">=</span> <span class="n">update_custom_config</span><span class="p">(</span><span class="n">config</span><span class="p">,</span> <span class="n">model_args</span><span class="p">)</span>
+                        <span class="n">model</span> <span class="o">=</span> <span class="n">CustomAutoVision2SeqModel</span><span class="p">(</span>
+                            <span class="n">config</span><span class="p">,</span>
+                            <span class="n">image_encoder_name_or_path</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">image_encoder_name_or_path</span><span class="p">,</span>
+                            <span class="n">qformer_name_or_path</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">qformer_name_or_path</span><span class="p">,</span>
+                            <span class="n">language_model_name_or_path</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">llm_model_name_or_path</span><span class="p">,</span>
+                            <span class="n">low_resource</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">low_resource</span><span class="p">)</span>
+                        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">pretrained_language_projection_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                            <span class="n">state_dict</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span>
+                                <span class="n">model_args</span><span class="o">.</span><span class="n">pretrained_language_projection_path</span><span class="p">,</span> <span class="n">map_location</span><span class="o">=</span><span class="s2">&quot;cpu&quot;</span><span class="p">)</span>
+                            <span class="n">new_state_dict</span> <span class="o">=</span> <span class="p">{}</span>
+                            <span class="n">new_state_dict</span><span class="p">[</span><span class="s1">&#39;model.language_projection.weight&#39;</span><span class="p">]</span> <span class="o">=</span> \
+                                <span class="n">state_dict</span><span class="p">[</span><span class="s1">&#39;model.mm_projector.weight&#39;</span><span class="p">]</span>
+                            <span class="n">new_state_dict</span><span class="p">[</span><span class="s1">&#39;model.language_projection.bias&#39;</span><span class="p">]</span> <span class="o">=</span> \
+                                <span class="n">state_dict</span><span class="p">[</span><span class="s1">&#39;model.mm_projector.bias&#39;</span><span class="p">]</span>
+                <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">llava_pretrain_model_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="c1"># used for inference that directly load the preatrain model</span>
+                    <span class="n">model</span> <span class="o">=</span> <span class="n">load_llava_pretrain_model</span><span class="p">(</span>
+                        <span class="n">model</span><span class="p">,</span> <span class="n">model_args</span><span class="o">.</span><span class="n">llava_pretrain_model_path</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">save_pretrain_model_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="n">model</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span>
+                            <span class="n">model_args</span><span class="o">.</span><span class="n">save_pretrain_model_path</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span> <span class="o">=</span> <span class="n">model</span>
+            <span class="c1"># init tokenizer</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">arch_type</span> <span class="o">==</span> <span class="s2">&quot;encoder_decoder&quot;</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span> <span class="o">=</span> <span class="n">AutoTokenizer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                    <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span> <span class="n">trust_remote_code</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">arch_type</span> <span class="o">==</span> <span class="s2">&quot;vision_encoder_decoder&quot;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">llava_loading</span> <span class="ow">is</span> <span class="kc">False</span><span class="p">:</span>
+                    <span class="c1"># blip2 image and token processor</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span> <span class="o">=</span> <span class="n">AutoProcessor</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                        <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span> <span class="n">trust_remote_code</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">llm_model_name_or_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="c1"># update the tokenizer from the custom llm.</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">tokenizer</span> <span class="o">=</span> <span class="p">(</span>
+                            <span class="n">AutoTokenizer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                                <span class="n">model_args</span><span class="o">.</span><span class="n">llm_model_name_or_path</span><span class="p">)</span>
+                        <span class="p">)</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">image_processor</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">image_processor</span>
+
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="c1"># image processor is stored in the vision encoder</span>
+                    <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">llm_model_name_or_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span> <span class="o">=</span> <span class="n">AutoTokenizer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                            <span class="n">model_args</span><span class="o">.</span><span class="n">llm_model_name_or_path</span><span class="p">)</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span> <span class="o">=</span> <span class="n">AutoTokenizer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                            <span class="n">config</span><span class="o">.</span><span class="n">text_config</span><span class="o">.</span><span class="n">_name_or_path</span><span class="p">)</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">image_processor</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">image_processor</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">NotImplementedError</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_model_full</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span>
+            <span class="k">if</span> <span class="n">peft_model_id</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span> <span class="o">=</span> <span class="n">PeftModel</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="p">,</span> <span class="n">peft_model_id</span>
+                <span class="p">)</span>
+            <span class="k">if</span> <span class="n">tune_strategy</span> <span class="o">==</span> <span class="s2">&quot;none&quot;</span> <span class="ow">and</span> <span class="n">with_deepspeed</span> <span class="ow">is</span> <span class="kc">True</span><span class="p">:</span>
+                <span class="c1"># when load the model with 4bit / 8bit.</span>
+                <span class="c1"># fail to use deepspeed.</span>
+                <span class="k">if</span> <span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;gpu&quot;</span><span class="p">:</span>
+                    <span class="n">deepspeed</span><span class="o">.</span><span class="n">init_distributed</span><span class="p">()</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">ds_engine</span> <span class="o">=</span> <span class="n">deepspeed</span><span class="o">.</span><span class="n">initialize</span><span class="p">(</span><span class="n">model</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="p">,</span> <span class="n">config_params</span><span class="o">=</span><span class="n">ds_config</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">ds_engine</span><span class="o">.</span><span class="n">module</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">padding_side</span> <span class="o">=</span> <span class="s2">&quot;left&quot;</span> <span class="c1"># necessary for auto-gressive inference</span>
+
+        <span class="k">elif</span> <span class="n">tune_strategy</span> <span class="o">==</span> <span class="s1">&#39;adapter&#39;</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s1">&#39;adapter tune strategy not implemented&#39;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">arch_type</span> <span class="o">==</span> <span class="s2">&quot;encoder_decoder&quot;</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">eos_token_id</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span>
+
+<div class="viewcode-block" id="HFEncoderDecoderModel.tokenize">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.tokenize">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">tokenize</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dataset</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Tokenize the full dataset.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        dataset :</span>
+<span class="sd">            Text dataset.</span>
+
+<span class="sd">        args : Optional.</span>
+<span class="sd">            Positional arguments.</span>
+
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        tokenized_datasets :</span>
+<span class="sd">            The tokenized dataset.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s1">&#39;tokenize not implemented&#39;</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="HFEncoderDecoderModel.encode">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.encode">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">encode</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]],</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span> <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform encoding process of the tokenizer.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        inputs : str or list.</span>
+<span class="sd">            The text sequence.</span>
+
+<span class="sd">        args : Optional.</span>
+<span class="sd">            Positional arguments.</span>
+
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        outputs :</span>
+<span class="sd">            The tokenized inputs.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># check how to handle the image processor</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="nb">dict</span><span class="p">):</span>
+            <span class="c1"># TODO refactor the input type to make it elegant.</span>
+            <span class="n">kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+            <span class="k">if</span> <span class="s2">&quot;images&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="nb">input</span><span class="p">:</span>
+                <span class="n">tokens</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">,</span> <span class="s2">&quot;image_processor&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">tokens</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;image_processor&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">images</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;images&quot;</span><span class="p">)</span>
+                    <span class="n">tokens</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+                    <span class="n">images</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">image_processor</span><span class="o">.</span><span class="n">preprocess</span><span class="p">(</span>
+                        <span class="n">images</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s1">&#39;pt&#39;</span><span class="p">)[</span><span class="s1">&#39;pixel_values&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+                    <span class="n">tokens</span><span class="p">[</span><span class="s1">&#39;pixel_values&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">images</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Can not find the image processor&quot;</span><span class="p">)</span>
+                    <span class="k">raise</span> <span class="ne">NotImplementedError</span>
+            <span class="k">return</span> <span class="n">tokens</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">(</span><span class="n">text</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span><span class="c1">#batch encode,will automatically do left padding</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="n">text</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;type &quot;</span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span><span class="si">}</span><span class="s1">&quot; cannot be encoded&#39;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="HFEncoderDecoderModel.decode">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.decode">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">decode</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">input</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span> <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform decoding process of the tokenizer.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        inputs : list.</span>
+<span class="sd">            The token sequence.</span>
+
+<span class="sd">        args : Optional.</span>
+<span class="sd">            Positional arguments.</span>
+
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        outputs :</span>
+<span class="sd">            The text decoded from the token inputs.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">List</span><span class="p">):</span>
+            <span class="nb">input</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">input</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span><span class="o">==</span><span class="mi">2</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">batch_decode</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span><span class="c1">#batch_decode</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># Can be list of ints or a Tensor</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="HFEncoderDecoderModel.inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">inference</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform generation process of the model.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        inputs :</span>
+<span class="sd">            The sequence used as a prompt for the generation or as model inputs to the model.</span>
+
+<span class="sd">        args : Optional.</span>
+<span class="sd">            Positional arguments.</span>
+
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        outputs :</span>
+<span class="sd">            The generated sequence output</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># current_time = time.strftime(&quot;%H:%M:%S&quot;, time.localtime())</span>
+        <span class="c1"># print(f&quot;{current_time}: model.inference: start&quot;, flush=True)</span>
+
+        <span class="c1"># TODO need to discuss how to handle pad_token_id</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">arch_type</span> <span class="o">==</span> <span class="s2">&quot;encoder_decoder&quot;</span><span class="p">:</span>
+            <span class="n">kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">pad_token_id</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">arch_type</span> <span class="o">==</span> <span class="s2">&quot;vision_encoder_decoder&quot;</span><span class="p">:</span>
+            <span class="c1"># TODO disucss how to modify the interface to remove this part.</span>
+            <span class="n">inputs</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+            <span class="n">input_ids</span> <span class="o">=</span> <span class="n">inputs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">&#39;input_ids&#39;</span><span class="p">)</span>
+            <span class="n">kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="o">**</span><span class="n">inputs</span><span class="p">)</span>
+            <span class="n">inputs</span> <span class="o">=</span> <span class="n">input_ids</span>
+
+        <span class="c1"># current_time = time.strftime(&quot;%H:%M:%S&quot;, time.localtime())</span>
+        <span class="c1"># print(f&quot;{current_time}: model.inference: kwargs update end&quot;, flush=True)</span>
+
+        <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;gpu&quot;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;ds_engine&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ds_engine</span><span class="o">.</span><span class="n">module</span><span class="o">.</span><span class="n">generate</span><span class="p">(</span>
+                        <span class="n">input_ids</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span>
+                        <span class="n">synced_gpus</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                        <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+                        <span class="o">**</span><span class="n">kwargs</span>
+                    <span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">generate</span><span class="p">(</span>
+                        <span class="n">input_ids</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span>
+                        <span class="n">synced_gpus</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                        <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+                        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+                    <span class="p">)</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;cpu&quot;</span><span class="p">:</span>
+                <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">generate</span><span class="p">(</span>
+                    <span class="n">input_ids</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span>
+                    <span class="n">synced_gpus</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                    <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+                    <span class="o">**</span><span class="n">kwargs</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;device </span><span class="se">\&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="se">\&quot;</span><span class="s2"> is not supported&quot;</span>
+                <span class="p">)</span>
+
+        <span class="c1"># current_time = time.strftime(&quot;%H:%M:%S&quot;, time.localtime())</span>
+        <span class="c1"># print(f&quot;{current_time}: model.inference: end&quot;, flush=True)</span>
+
+        <span class="k">return</span> <span class="n">outputs</span></div>
+
+
+
+<div class="viewcode-block" id="HFEncoderDecoderModel.merge_lora_weights">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.merge_lora_weights">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">merge_lora_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">use_lora</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">merge_and_unload</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;LoRA training is NOT enabled. Merging LoRA weights is not applicable.&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="HFEncoderDecoderModel.save">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.save">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">save</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">dir</span><span class="p">,</span> <span class="n">save_full_model</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform generation process of the model.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        dir :</span>
+<span class="sd">            The directory to save model and tokenizer</span>
+
+<span class="sd">        save_full_model : Optional.</span>
+<span class="sd">            Whether to save full model.</span>
+
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        outputs :</span>
+<span class="sd">            The generated sequence output</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">get_tokenizer</span><span class="p">()</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="nb">dir</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">save_full_model</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">use_lora</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_model_full</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="nb">dir</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="nb">dir</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="HFEncoderDecoderModel.get_max_length">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_max_length">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_max_length</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Return max acceptable input length in terms of tokens.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="s2">&quot;tokenizer&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="vm">__dict__</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># for the multi-modality processor,</span>
+            <span class="c1"># the max length is stored in the inner text tokenizer</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span></div>
+
+
+
+<div class="viewcode-block" id="HFEncoderDecoderModel.get_tokenizer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_tokenizer">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_tokenizer</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Return the tokenizer of the model.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span></div>
+
+
+
+<div class="viewcode-block" id="HFEncoderDecoderModel.get_backend_model">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_backend_model">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_backend_model</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Return the backend model.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/models/hf_model_mixin.html b/_modules/lmflow/models/hf_model_mixin.html
new file mode 100644
index 000000000..0282e7912
--- /dev/null
+++ b/_modules/lmflow/models/hf_model_mixin.html
@@ -0,0 +1,1110 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.models.hf_model_mixin &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/models/hf_model_mixin';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.hf_model_mixin</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.models.hf_model_mixin</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">gc</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Union</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">copy</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">deepspeed</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">CONFIG_MAPPING</span><span class="p">,</span>
+    <span class="n">AutoConfig</span><span class="p">,</span>
+    <span class="n">BitsAndBytesConfig</span><span class="p">,</span>
+    <span class="n">AutoTokenizer</span><span class="p">,</span>
+    <span class="n">PreTrainedTokenizer</span><span class="p">,</span>
+    <span class="n">PreTrainedTokenizerFast</span><span class="p">,</span>
+    <span class="n">AutoModelForCausalLM</span><span class="p">,</span>
+    <span class="n">AutoModelForSequenceClassification</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">peft</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">LoraConfig</span><span class="p">,</span>
+    <span class="n">PeftModel</span><span class="p">,</span>
+    <span class="n">TaskType</span><span class="p">,</span>
+    <span class="n">get_peft_model</span><span class="p">,</span>
+    <span class="n">prepare_model_for_kbit_training</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">peft.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="n">TRANSFORMERS_MODELS_TO_LORA_TARGET_MODULES_MAPPING</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.base_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">BaseModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">LMFLOW_LORA_TARGET_MODULES_MAPPING</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="n">ModelArguments</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.versioning</span><span class="w"> </span><span class="kn">import</span> <span class="n">is_vllm_available</span>
+
+<span class="k">if</span> <span class="n">is_vllm_available</span><span class="p">():</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">vllm</span><span class="w"> </span><span class="kn">import</span> <span class="n">LLM</span><span class="p">,</span> <span class="n">SamplingParams</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">vllm.distributed.parallel_state</span><span class="w"> </span><span class="kn">import</span> <span class="n">destroy_model_parallel</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="HF_AUTOMODEL_MAPPING">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HF_AUTOMODEL_MAPPING">[docs]</a>
+<span class="n">HF_AUTOMODEL_MAPPING</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;decoder_only&quot;</span><span class="p">:</span> <span class="n">AutoModelForCausalLM</span><span class="p">,</span>
+    <span class="s2">&quot;text_regression&quot;</span><span class="p">:</span> <span class="n">AutoModelForSequenceClassification</span>
+<span class="p">}</span></div>
+
+
+<div class="viewcode-block" id="HF_AUTOMODEL_TYPE">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HF_AUTOMODEL_TYPE">[docs]</a>
+<span class="n">HF_AUTOMODEL_TYPE</span> <span class="o">=</span> <span class="n">Union</span><span class="p">[</span><span class="n">AutoModelForCausalLM</span><span class="p">,</span> <span class="n">AutoModelForSequenceClassification</span><span class="p">]</span></div>
+
+
+<div class="viewcode-block" id="LORA_TARGET_MODULES_MAPPING">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.LORA_TARGET_MODULES_MAPPING">[docs]</a>
+<span class="n">LORA_TARGET_MODULES_MAPPING</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="n">k</span><span class="p">:</span> <span class="n">TRANSFORMERS_MODELS_TO_LORA_TARGET_MODULES_MAPPING</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">k</span><span class="p">,</span> <span class="n">LMFLOW_LORA_TARGET_MODULES_MAPPING</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">k</span><span class="p">))</span> 
+    <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="nb">set</span><span class="p">(</span><span class="n">TRANSFORMERS_MODELS_TO_LORA_TARGET_MODULES_MAPPING</span><span class="p">)</span> <span class="o">|</span> <span class="nb">set</span><span class="p">(</span><span class="n">LMFLOW_LORA_TARGET_MODULES_MAPPING</span><span class="p">)</span>
+<span class="p">}</span></div>
+
+
+
+<div class="viewcode-block" id="HFModelMixin">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">HFModelMixin</span><span class="p">(</span><span class="n">BaseModel</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">do_train</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">ds_config</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">device</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span><span class="o">=</span><span class="s2">&quot;gpu&quot;</span><span class="p">,</span>
+        <span class="n">use_accelerator</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">hf_auto_model_additional_args</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">]</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Initializes a HFModel instance.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ----------</span>
+<span class="sd">        model_args : </span>
+<span class="sd">            Dictionary with model arguments such as model name, path, revision, etc.</span>
+<span class="sd">        do_train : bool</span>
+<span class="sd">            To prepare the model for training or inference.</span>
+<span class="sd">        ds_config : optional</span>
+<span class="sd">            Deepspeed configuration for distributed training, by default None</span>
+<span class="sd">        device : str, optional</span>
+<span class="sd">            By default &quot;gpu&quot;</span>
+<span class="sd">        use_accelerator : bool, optional</span>
+<span class="sd">            By default False</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="c1"># See more about loading any type of standard or custom dataset (from</span>
+        <span class="c1"># files, python dict, pandas DataFrame, etc) at</span>
+        <span class="c1"># https://huggingface.co/docs/datasets/loading_datasets.html.</span>
+
+        <span class="c1"># Load pretrained model and tokenizer</span>
+        <span class="c1">#</span>
+        <span class="c1"># Distributed training: The .from_pretrained methods guarantee that</span>
+        <span class="c1"># only one local process can concurrently download model &amp; vocab.</span>
+
+<div class="viewcode-block" id="HFModelMixin.device">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.device">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">device</span> <span class="o">=</span> <span class="n">device</span></div>
+
+<div class="viewcode-block" id="HFModelMixin.model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span> <span class="o">=</span> <span class="n">model_args</span></div>
+
+<div class="viewcode-block" id="HFModelMixin.hf_auto_model">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.hf_auto_model">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">hf_auto_model</span> <span class="o">=</span> <span class="n">HF_AUTOMODEL_MAPPING</span><span class="p">[</span><span class="n">model_args</span><span class="o">.</span><span class="n">arch_type</span><span class="p">]</span></div>
+
+<div class="viewcode-block" id="HFModelMixin.use_accelerator">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.use_accelerator">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_accelerator</span> <span class="o">=</span> <span class="n">use_accelerator</span></div>
+
+<div class="viewcode-block" id="HFModelMixin.ds_config">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.ds_config">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ds_config</span> <span class="o">=</span> <span class="n">ds_config</span></div>
+
+<div class="viewcode-block" id="HFModelMixin.do_train">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.do_train">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">do_train</span> <span class="o">=</span> <span class="n">do_train</span></div>
+
+        
+<div class="viewcode-block" id="HFModelMixin.tokenizer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.tokenizer">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__prepare_tokenizer</span><span class="p">(</span><span class="n">model_args</span><span class="p">)</span></div>
+
+<div class="viewcode-block" id="HFModelMixin.torch_dtype">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.torch_dtype">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">torch_dtype</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__prepare_dtype</span><span class="p">(</span><span class="n">model_args</span><span class="p">)</span></div>
+
+<div class="viewcode-block" id="HFModelMixin.hf_model_config">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.hf_model_config">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">hf_model_config</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__prepare_model_config</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="n">hf_auto_model_additional_args</span><span class="p">)</span></div>
+
+<div class="viewcode-block" id="HFModelMixin.quant_config">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.quant_config">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">quant_config</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__prepare_quant_config</span><span class="p">(</span><span class="n">model_args</span><span class="p">)</span></div>
+
+<div class="viewcode-block" id="HFModelMixin.peft_config">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.peft_config">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">peft_config</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__prepare_peft_config</span><span class="p">(</span><span class="n">model_args</span><span class="p">)</span></div>
+
+<div class="viewcode-block" id="HFModelMixin._activated">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin._activated">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_activated</span> <span class="o">=</span> <span class="kc">False</span> <span class="c1"># for inference load and offload</span></div>
+
+        
+        <span class="c1"># Some implementations require custom modules to be injected into the model.</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">__model_module_inject</span><span class="p">(</span><span class="n">model_args</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">do_train</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">__prepare_model_for_training</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">hf_auto_model</span><span class="p">)</span>
+                        
+
+<div class="viewcode-block" id="HFModelMixin.__prepare_tokenizer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_tokenizer">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__prepare_tokenizer</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">PreTrainedTokenizer</span><span class="p">,</span> <span class="n">PreTrainedTokenizerFast</span><span class="p">]:</span>
+        <span class="n">tokenizer_name</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">tokenizer_name</span> <span class="ow">or</span> <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">tokenizer_name</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;You are instantiating a new tokenizer from scratch. This is&quot;</span>
+                <span class="s2">&quot; not supported by this script. You can do it from another&quot;</span>
+                <span class="s2">&quot; script, save it, and load it from here, using&quot;</span>
+                <span class="s2">&quot; --tokenizer_name.&quot;</span>
+            <span class="p">)</span>
+        
+        <span class="n">tokenizer_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;cache_dir&quot;</span><span class="p">:</span> <span class="n">model_args</span><span class="o">.</span><span class="n">cache_dir</span><span class="p">,</span>
+            <span class="s2">&quot;use_fast&quot;</span><span class="p">:</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_fast_tokenizer</span><span class="p">,</span>
+            <span class="s2">&quot;revision&quot;</span><span class="p">:</span> <span class="n">model_args</span><span class="o">.</span><span class="n">model_revision</span><span class="p">,</span>
+            <span class="s2">&quot;token&quot;</span><span class="p">:</span> <span class="n">model_args</span><span class="o">.</span><span class="n">token</span><span class="p">,</span>
+            <span class="s2">&quot;trust_remote_code&quot;</span><span class="p">:</span> <span class="n">model_args</span><span class="o">.</span><span class="n">trust_remote_code</span><span class="p">,</span>
+        <span class="p">}</span>
+        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">padding_side</span> <span class="o">!=</span> <span class="s1">&#39;auto&#39;</span><span class="p">:</span>
+            <span class="n">tokenizer_kwargs</span><span class="p">[</span><span class="s2">&quot;padding_side&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">padding_side</span>
+        
+        <span class="k">try</span><span class="p">:</span>
+            <span class="n">tokenizer</span> <span class="o">=</span> <span class="n">AutoTokenizer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="n">tokenizer_name</span><span class="p">,</span> <span class="o">**</span><span class="n">tokenizer_kwargs</span><span class="p">)</span>
+        <span class="k">except</span> <span class="ne">RecursionError</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                <span class="s2">&quot;The tokenizer_config.json file doesn&#39;t set the special tokens. Using default values: &quot;</span>
+                <span class="s2">&quot;&lt;unk&gt;, &lt;s&gt;, &lt;/s&gt; for unknown token, bos token and eos token respectively.&quot;</span><span class="p">)</span>
+            <span class="n">tokenizer</span> <span class="o">=</span> <span class="n">AutoTokenizer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="n">tokenizer_name</span><span class="p">,</span> <span class="n">unk_token</span><span class="o">=</span><span class="s2">&quot;&lt;unk&gt;&quot;</span><span class="p">,</span>
+                                                <span class="n">bos_token</span><span class="o">=</span><span class="s2">&quot;&lt;s&gt;&quot;</span><span class="p">,</span>
+                                                <span class="n">eos_token</span><span class="o">=</span><span class="s2">&quot;&lt;/s&gt;&quot;</span><span class="p">,</span>
+                                                <span class="o">**</span><span class="n">tokenizer_kwargs</span><span class="p">)</span>
+
+        <span class="n">tokenizer</span><span class="o">.</span><span class="n">truncation_side</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">truncation_side</span> <span class="ow">or</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">truncation_side</span>
+        <span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">model_max_length</span> <span class="ow">or</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span>
+        
+        <span class="k">return</span> <span class="n">tokenizer</span></div>
+
+    
+    
+<div class="viewcode-block" id="HFModelMixin.__prepare_dtype">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_dtype">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__prepare_dtype</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">dtype</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">arch_type</span> <span class="o">==</span> <span class="s1">&#39;text_regression&#39;</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">torch_dtype</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;auto&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="s2">&quot;bf16&quot;</span><span class="p">,</span> <span class="s2">&quot;bfloat16&quot;</span><span class="p">]:</span>
+                <span class="n">torch_dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">bfloat16</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">torch_dtype</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">torch</span><span class="p">,</span> <span class="n">model_args</span><span class="o">.</span><span class="n">torch_dtype</span><span class="p">)</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;If you are doing reward modeling,&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot; InstructGPT uses torch.bfloat16 for reward model, but you&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot; are using </span><span class="si">{</span><span class="n">torch_dtype</span><span class="si">}</span><span class="s2"> for your reward model init. Ignore&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot; this warning if it is intended.&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">torch_dtype</span> <span class="o">=</span> <span class="p">(</span>
+                <span class="n">model_args</span><span class="o">.</span><span class="n">torch_dtype</span>
+                <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">torch_dtype</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;auto&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span>
+                <span class="k">else</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">torch</span><span class="p">,</span> <span class="n">model_args</span><span class="o">.</span><span class="n">torch_dtype</span><span class="p">)</span>
+            <span class="p">)</span>
+            
+        <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;torch_dtype on init: </span><span class="si">{</span><span class="n">torch_dtype</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">torch_dtype</span></div>
+
+
+
+<div class="viewcode-block" id="HFModelMixin.__prepare_model_config">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_config">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__prepare_model_config</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">hf_auto_model_additional_args</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">]</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Prepare model configuration for hf auto register,</span>
+<span class="sd">        Parameters</span>
+<span class="sd">        ----------</span>
+<span class="sd">        model_args : ModelArguments</span>
+<span class="sd">            LMFlow model arguments.</span>
+<span class="sd">        hf_auto_model_additional_args : Optional[Dict], optional</span>
+<span class="sd">            Special configurations such as `num_labels` in `AutoModelForSequenceClassification` </span>
+<span class="sd">            (commonly used in reward modeling) will not preset in __prepare_model_config, </span>
+<span class="sd">            so it should be passed in hf_auto_model_additional_args.</span>
+<span class="sd">        Returns</span>
+<span class="sd">        -------</span>
+<span class="sd">        config : ModelConfig</span>
+<span class="sd">            hf model config.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">config_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;attn_implementation&quot;</span><span class="p">:</span> <span class="s2">&quot;flash_attention_2&quot;</span> <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_flash_attention</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="s2">&quot;cache_dir&quot;</span><span class="p">:</span> <span class="n">model_args</span><span class="o">.</span><span class="n">cache_dir</span><span class="p">,</span>
+            <span class="s2">&quot;revision&quot;</span><span class="p">:</span> <span class="n">model_args</span><span class="o">.</span><span class="n">model_revision</span><span class="p">,</span>
+            <span class="s2">&quot;token&quot;</span><span class="p">:</span> <span class="n">model_args</span><span class="o">.</span><span class="n">token</span><span class="p">,</span>
+            <span class="s2">&quot;trust_remote_code&quot;</span><span class="p">:</span> <span class="n">model_args</span><span class="o">.</span><span class="n">trust_remote_code</span><span class="p">,</span>
+            <span class="s2">&quot;from_tf&quot;</span><span class="p">:</span> <span class="nb">bool</span><span class="p">(</span><span class="s2">&quot;.ckpt&quot;</span> <span class="ow">in</span> <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">),</span>
+        <span class="p">}</span>
+        <span class="k">if</span> <span class="n">hf_auto_model_additional_args</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">config_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">hf_auto_model_additional_args</span><span class="p">)</span>
+            
+        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">config_name</span><span class="p">:</span>
+            <span class="n">config</span> <span class="o">=</span> <span class="n">AutoConfig</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="n">model_args</span><span class="o">.</span><span class="n">config_name</span><span class="p">,</span> <span class="o">**</span><span class="n">config_kwargs</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">:</span>
+            <span class="n">config</span> <span class="o">=</span> <span class="n">AutoConfig</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span> <span class="o">**</span><span class="n">config_kwargs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">config</span> <span class="o">=</span> <span class="n">CONFIG_MAPPING</span><span class="p">[</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_type</span><span class="p">]()</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;You are instantiating a new config instance from scratch.&quot;</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">config_overrides</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Overriding config: </span><span class="si">{</span><span class="n">model_args</span><span class="o">.</span><span class="n">config_overrides</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="n">config</span><span class="o">.</span><span class="n">update_from_string</span><span class="p">(</span><span class="n">model_args</span><span class="o">.</span><span class="n">config_overrides</span><span class="p">)</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;New config: </span><span class="si">{</span><span class="n">config</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">config</span></div>
+
+    
+    
+<div class="viewcode-block" id="HFModelMixin.__prepare_quant_config">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_quant_config">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__prepare_quant_config</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">quant_config</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">do_train</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_qlora</span><span class="p">:</span>
+                <span class="n">quant_config</span> <span class="o">=</span> <span class="n">BitsAndBytesConfig</span><span class="p">(</span>
+                    <span class="n">load_in_4bit</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">bits</span> <span class="o">==</span> <span class="mi">4</span><span class="p">,</span>
+                    <span class="n">load_in_8bit</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">bits</span> <span class="o">==</span> <span class="mi">8</span><span class="p">,</span>
+                    <span class="n">llm_int8_threshold</span><span class="o">=</span><span class="mf">6.0</span><span class="p">,</span>
+                    <span class="n">llm_int8_has_fp16_weight</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                    <span class="n">bnb_4bit_compute_dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">torch_dtype</span><span class="p">,</span>
+                    <span class="n">bnb_4bit_use_double_quant</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">double_quant</span><span class="p">,</span>
+                    <span class="n">bnb_4bit_quant_type</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">quant_type</span><span class="p">,</span>
+                <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span> <span class="c1"># inference</span>
+            <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_int8</span><span class="p">:</span>
+                <span class="n">quant_config</span> <span class="o">=</span> <span class="n">BitsAndBytesConfig</span><span class="p">(</span>
+                    <span class="n">load_in_8bit</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_int8</span><span class="p">,</span>
+                <span class="p">)</span>
+                
+        <span class="k">return</span> <span class="n">quant_config</span></div>
+
+
+    
+<div class="viewcode-block" id="HFModelMixin.__prepare_peft_config">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_peft_config">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__prepare_peft_config</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">peft_config</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_lora</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">lora_target_modules</span><span class="p">:</span>
+                <span class="n">lora_target_modules</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">lora_target_modules</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">model_config</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">hf_model_config</span>
+                <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">model_config</span><span class="p">,</span> <span class="s2">&quot;to_dict&quot;</span><span class="p">):</span>
+                    <span class="n">model_config</span> <span class="o">=</span> <span class="n">model_config</span><span class="o">.</span><span class="n">to_dict</span><span class="p">()</span>
+                <span class="k">if</span> <span class="s2">&quot;model_type&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">model_config</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">model_config</span><span class="p">[</span><span class="s2">&quot;model_type&quot;</span><span class="p">]:</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;It seems that your base model is a custom model, since &quot;</span>
+                                   <span class="s2">&quot;model_type is not found in model_config when preparing peft config. &quot;</span>
+                                   <span class="s2">&quot;Setting model_type to &#39;custom&#39; as a fallback.&quot;</span><span class="p">)</span>
+                    <span class="n">model_config</span><span class="p">[</span><span class="s2">&quot;model_type&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;custom&quot;</span>
+                <span class="n">lora_target_modules</span> <span class="o">=</span> <span class="n">LORA_TARGET_MODULES_MAPPING</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">model_config</span><span class="p">[</span><span class="s2">&quot;model_type&quot;</span><span class="p">],</span> <span class="kc">None</span><span class="p">)</span>
+
+            <span class="n">peft_config</span> <span class="o">=</span> <span class="n">LoraConfig</span><span class="p">(</span>
+                <span class="n">task_type</span><span class="o">=</span><span class="n">TaskType</span><span class="o">.</span><span class="n">CAUSAL_LM</span><span class="p">,</span>
+                <span class="n">inference_mode</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                <span class="n">r</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">lora_r</span><span class="p">,</span>
+                <span class="n">lora_alpha</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">lora_alpha</span><span class="p">,</span>
+                <span class="n">lora_dropout</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">lora_dropout</span><span class="p">,</span>
+                <span class="n">target_modules</span><span class="o">=</span><span class="n">lora_target_modules</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_dora</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">lora_target_modules</span><span class="p">:</span>
+                <span class="n">lora_target_modules</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">lora_target_modules</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">model_config</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">hf_model_config</span>
+                <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">model_config</span><span class="p">,</span> <span class="s2">&quot;to_dict&quot;</span><span class="p">):</span>
+                    <span class="n">model_config</span> <span class="o">=</span> <span class="n">model_config</span><span class="o">.</span><span class="n">to_dict</span><span class="p">()</span>
+                <span class="k">if</span> <span class="s2">&quot;model_type&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">model_config</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">model_config</span><span class="p">[</span><span class="s2">&quot;model_type&quot;</span><span class="p">]:</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;It seems that your base model is a custom model, since &quot;</span>
+                                   <span class="s2">&quot;model_type is not found in model_config when preparing peft config. &quot;</span>
+                                   <span class="s2">&quot;Setting model_type to &#39;custom&#39; as a fallback.&quot;</span><span class="p">)</span>
+                    <span class="n">model_config</span><span class="p">[</span><span class="s2">&quot;model_type&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;custom&quot;</span>
+                <span class="n">lora_target_modules</span> <span class="o">=</span> <span class="n">LORA_TARGET_MODULES_MAPPING</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">model_config</span><span class="p">[</span><span class="s2">&quot;model_type&quot;</span><span class="p">],</span> <span class="kc">None</span><span class="p">)</span>
+            
+            <span class="n">peft_config</span> <span class="o">=</span> <span class="n">LoraConfig</span><span class="p">(</span>
+                <span class="n">use_dora</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                <span class="n">task_type</span><span class="o">=</span><span class="n">TaskType</span><span class="o">.</span><span class="n">CAUSAL_LM</span><span class="p">,</span>
+                <span class="n">inference_mode</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                <span class="n">r</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">lora_r</span><span class="p">,</span>
+                <span class="n">lora_alpha</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">lora_alpha</span><span class="p">,</span>
+                <span class="n">lora_dropout</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">lora_dropout</span><span class="p">,</span>
+                <span class="n">target_modules</span><span class="o">=</span><span class="n">lora_target_modules</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">return</span> <span class="n">peft_config</span></div>
+
+    
+    
+<div class="viewcode-block" id="HFModelMixin.__model_module_inject">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__model_module_inject">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__model_module_inject</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Override some model modules with custom implementations.</span>
+<span class="sd">        </span>
+<span class="sd">        Current implementations:</span>
+<span class="sd">        - Position interpolation (model_args.do_rope_scaling): </span>
+<span class="sd">            replace llama embeddings with condense embeddings.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># position interpolation</span>
+        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">do_rope_scaling</span><span class="p">:</span>
+            <span class="k">if</span> <span class="s2">&quot;LlamaForCausalLM&quot;</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_config</span><span class="o">.</span><span class="n">architectures</span><span class="p">:</span>
+                <span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+                        <span class="n">replace_llama_with_condense</span><span class="p">,</span>
+                <span class="p">)</span>
+                <span class="n">replace_llama_with_condense</span><span class="p">(</span><span class="n">model_args</span><span class="o">.</span><span class="n">rope_pi_ratio</span><span class="p">,</span> <span class="n">model_args</span><span class="o">.</span><span class="n">rope_ntk_ratio</span><span class="p">)</span></div>
+
+                
+                
+<div class="viewcode-block" id="HFModelMixin.__prepare_model_for_training">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_training">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__prepare_model_for_training</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">hf_auto_model</span><span class="p">:</span> <span class="n">HF_AUTOMODEL_TYPE</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">assert</span> <span class="bp">self</span><span class="o">.</span><span class="n">do_train</span><span class="p">,</span> <span class="s2">&quot;To prepare the model for training, please set do_train=True.&quot;</span>
+        <span class="c1"># TODO: change to accelerate</span>
+
+        <span class="k">if</span> <span class="s1">&#39;hymba&#39;</span> <span class="ow">in</span> <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">:</span>
+            <span class="kn">import</span><span class="w"> </span><span class="nn">torch._dynamo</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">_dynamo</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">suppress_errors</span> <span class="o">=</span> <span class="kc">True</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">_dynamo</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">disable</span> <span class="o">=</span> <span class="kc">True</span>
+    
+
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="s2">&quot;Preparing model for training&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">:</span>
+            <span class="n">model</span> <span class="o">=</span> <span class="n">hf_auto_model</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span>
+                <span class="n">torch_dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">torch_dtype</span><span class="p">,</span>
+                <span class="n">config</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">hf_model_config</span><span class="p">,</span>
+                <span class="n">quantization_config</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">quant_config</span><span class="p">,</span>
+                <span class="n">trust_remote_code</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">trust_remote_code</span><span class="p">,</span>
+            <span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_qlora</span><span class="p">:</span>
+                <span class="n">model</span><span class="o">.</span><span class="n">gradient_checkpointing_enable</span><span class="p">()</span>
+                <span class="n">model</span> <span class="o">=</span> <span class="n">prepare_model_for_kbit_training</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">model</span> <span class="o">=</span> <span class="n">hf_auto_model</span><span class="o">.</span><span class="n">from_config</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">hf_model_config</span><span class="p">)</span>
+            <span class="n">n_params</span> <span class="o">=</span> <span class="nb">sum</span><span class="p">(</span><span class="nb">dict</span><span class="p">((</span><span class="n">p</span><span class="o">.</span><span class="n">data_ptr</span><span class="p">(),</span> <span class="n">p</span><span class="o">.</span><span class="n">numel</span><span class="p">())</span> <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">model</span><span class="o">.</span><span class="n">parameters</span><span class="p">())</span><span class="o">.</span><span class="n">values</span><span class="p">())</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Training new model from scratch - Total size=</span><span class="si">{</span><span class="n">n_params</span><span class="o">/</span><span class="mi">2</span><span class="o">**</span><span class="mi">20</span><span class="si">:</span><span class="s2">.2f</span><span class="si">}</span><span class="s2">M params&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">backend_model_full</span> <span class="o">=</span> <span class="n">model</span>
+        
+        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">ignore_bias_buffers</span><span class="p">:</span>
+            <span class="c1"># torch distributed hack</span>
+            <span class="c1"># fix for DDP issues with LM bias/mask buffers - invalid scalar type, inplace operation. </span>
+            <span class="c1"># See: https://github.com/huggingface/transformers/issues/22482#issuecomment-1595790992</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">_ddp_params_and_buffers_to_ignore</span> <span class="o">=</span> <span class="p">[</span>
+                <span class="n">name</span> <span class="k">for</span> <span class="n">name</span><span class="p">,</span> <span class="n">buffer</span> <span class="ow">in</span> <span class="n">model</span><span class="o">.</span><span class="n">named_buffers</span><span class="p">()</span> <span class="k">if</span> <span class="n">buffer</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">bool</span>
+            <span class="p">]</span>
+        
+        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_lora</span> <span class="ow">or</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_dora</span><span class="p">:</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">enable_input_require_grads</span><span class="p">()</span>
+            <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">lora_model_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="c1"># Load model from LoRA weights</span>
+                <span class="n">model</span> <span class="o">=</span> <span class="n">PeftModel</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                    <span class="n">model</span><span class="p">,</span>
+                    <span class="n">model_args</span><span class="o">.</span><span class="n">lora_model_path</span><span class="p">,</span>
+                    <span class="n">is_trainable</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="c1"># New LoRA Finetuning</span>
+                <span class="n">model</span> <span class="o">=</span> <span class="n">get_peft_model</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">peft_config</span><span class="p">)</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">print_trainable_parameters</span><span class="p">()</span>
+
+        <span class="c1"># We resize the embeddings only when necessary to avoid index errors.</span>
+        <span class="c1"># If you are creating a model from scratch on a small vocab and want a</span>
+        <span class="c1"># smaller embedding size, remove this test.</span>
+        <span class="k">with</span> <span class="n">deepspeed</span><span class="o">.</span><span class="n">zero</span><span class="o">.</span><span class="n">GatheredParameters</span><span class="p">(</span><span class="n">model</span><span class="o">.</span><span class="n">get_input_embeddings</span><span class="p">()</span><span class="o">.</span><span class="n">weight</span><span class="p">,</span> <span class="n">modifier_rank</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+            <span class="n">weights</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">get_input_embeddings</span><span class="p">()</span><span class="o">.</span><span class="n">weight</span>
+            <span class="n">embedding_size</span> <span class="o">=</span> <span class="n">weights</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">)</span> <span class="o">&gt;</span> <span class="n">embedding_size</span><span class="p">:</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">resize_token_embeddings</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">))</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span> <span class="o">=</span> <span class="n">model</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">__prepare_model_post_process</span><span class="p">()</span></div>
+
+
+    
+<div class="viewcode-block" id="HFModelMixin.__prepare_model_for_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__prepare_model_for_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">hf_auto_model</span><span class="p">:</span> <span class="n">HF_AUTOMODEL_TYPE</span><span class="p">,</span>
+        <span class="n">use_accelerator</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">ds_config</span>
+    <span class="p">):</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Backend model already initialized, moving to device: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;backend_model&quot;</span><span class="p">):</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">(</span><span class="s2">&quot;cpu&quot;</span><span class="p">):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+            <span class="k">return</span>
+            
+        <span class="c1"># TODO: change to accelerate</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="s2">&quot;Preparing model for inference&quot;</span><span class="p">)</span>
+        <span class="n">inference_load_kwargs</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="n">inference_load_kwargs_bak</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">inference_load_kwargs</span><span class="p">)</span>
+        <span class="n">ram_optimized_load_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;device_map&quot;</span><span class="p">:</span> <span class="s2">&quot;auto&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;offload_folder&quot;</span><span class="p">:</span> <span class="s2">&quot;offload&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;offload_state_dict&quot;</span><span class="p">:</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="p">}</span>
+
+        <span class="k">if</span> <span class="n">use_accelerator</span> <span class="ow">or</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_ram_optimized_load</span><span class="p">:</span>
+            <span class="n">inference_load_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">ram_optimized_load_kwargs</span><span class="p">)</span>
+                   
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">use_accelerator</span><span class="p">:</span>
+            <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.integrations</span><span class="w"> </span><span class="kn">import</span> <span class="n">HfDeepSpeedConfig</span>
+            <span class="n">dschf</span> <span class="o">=</span> <span class="n">HfDeepSpeedConfig</span><span class="p">(</span><span class="n">ds_config</span><span class="p">)</span>
+            
+        <span class="k">try</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span> <span class="o">=</span> <span class="n">hf_auto_model</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span>
+                <span class="n">torch_dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">torch_dtype</span><span class="p">,</span>
+                <span class="n">config</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">hf_model_config</span><span class="p">,</span>
+                <span class="n">quantization_config</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">quant_config</span><span class="p">,</span>
+                <span class="o">**</span><span class="n">inference_load_kwargs</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">except</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                <span class="s2">&quot;Failed to use RAM optimized load. Using original load instead.&quot;</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span> <span class="o">=</span> <span class="n">hf_auto_model</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span>
+                <span class="n">torch_dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">torch_dtype</span><span class="p">,</span>
+                <span class="n">config</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">hf_model_config</span><span class="p">,</span>
+                <span class="n">quantization_config</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">quant_config</span><span class="p">,</span>
+                <span class="o">**</span><span class="n">inference_load_kwargs_bak</span><span class="p">,</span>
+            <span class="p">)</span>
+            
+        <span class="bp">self</span><span class="o">.</span><span class="n">backend_model_full</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span>
+        
+        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">lora_model_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span> <span class="o">=</span> <span class="n">PeftModel</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="p">,</span> 
+                <span class="n">model_args</span><span class="o">.</span><span class="n">lora_model_path</span><span class="p">,</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="p">(</span><span class="ow">not</span> <span class="n">use_accelerator</span><span class="p">)</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;gpu&quot;</span><span class="p">:</span>
+            <span class="n">deepspeed</span><span class="o">.</span><span class="n">init_distributed</span><span class="p">()</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">ds_engine</span> <span class="o">=</span> <span class="n">deepspeed</span><span class="o">.</span><span class="n">initialize</span><span class="p">(</span><span class="n">model</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="p">,</span> <span class="n">config_params</span><span class="o">=</span><span class="n">ds_config</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">ds_engine</span><span class="o">.</span><span class="n">module</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+            
+        <span class="bp">self</span><span class="o">.</span><span class="n">__prepare_model_post_process</span><span class="p">()</span></div>
+
+                    
+        
+<div class="viewcode-block" id="HFModelMixin.__prepare_model_for_vllm_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_vllm_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__prepare_model_for_vllm_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">vllm_gpu_memory_utilization</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+        <span class="n">vllm_tensor_parallel_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">is_vllm_available</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">ImportError</span><span class="p">(</span><span class="s1">&#39;VLLM is not available. Please install via `pip install -e &quot;.[vllm]&quot;`.&#39;</span><span class="p">)</span>
+        
+        <span class="bp">self</span><span class="o">.</span><span class="n">backend_model_for_inference</span> <span class="o">=</span> <span class="n">LLM</span><span class="p">(</span>
+            <span class="n">model</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span>
+            <span class="n">tokenizer</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span>
+            <span class="n">dtype</span><span class="o">=</span><span class="n">model_args</span><span class="o">.</span><span class="n">torch_dtype</span> <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">torch_dtype</span> <span class="k">else</span> <span class="s2">&quot;auto&quot;</span><span class="p">,</span>
+            <span class="n">load_format</span><span class="o">=</span><span class="s2">&quot;auto&quot;</span><span class="p">,</span>
+            <span class="n">gpu_memory_utilization</span><span class="o">=</span><span class="n">vllm_gpu_memory_utilization</span><span class="p">,</span>
+            <span class="n">tensor_parallel_size</span><span class="o">=</span><span class="n">vllm_tensor_parallel_size</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+        
+    
+<div class="viewcode-block" id="HFModelMixin.__prepare_model_post_process">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_post_process">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__prepare_model_post_process</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="c1"># old models/tokenizers may not have these attributes, fixing            </span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">eos_token</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">eos_token_id</span>
+            
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span>
+        
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">eos_padding</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token</span>
+            
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">config</span><span class="p">,</span> <span class="s2">&quot;pad_token_id&quot;</span><span class="p">):</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;pad_token_id not found in model config. Setting pad_token_id to eos_token_id.&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">pad_token_id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">eos_token_id</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">pad_token_id</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;pad_token_id is None in model config. Setting pad_token_id to eos_token_id.&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">pad_token_id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">eos_token_id</span></div>
+
+            
+
+<div class="viewcode-block" id="HFModelMixin.activate_model_for_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.activate_model_for_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">activate_model_for_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">use_vllm</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_activated</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;You are trying to activate the model for inference, but it is already activated.&quot;</span><span class="p">)</span>
+            <span class="k">return</span>
+        
+        <span class="k">if</span> <span class="n">use_vllm</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">__prepare_model_for_vllm_inference</span><span class="p">(</span>
+                <span class="n">model_args</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="p">,</span>
+                <span class="n">vllm_gpu_memory_utilization</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;vllm_gpu_memory_utilization&quot;</span><span class="p">),</span>
+                <span class="n">vllm_tensor_parallel_size</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;vllm_tensor_parallel_size&quot;</span><span class="p">),</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">__prepare_model_for_inference</span><span class="p">(</span>
+                <span class="n">model_args</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="p">,</span>
+                <span class="n">hf_auto_model</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">hf_auto_model</span><span class="p">,</span>
+                <span class="n">use_accelerator</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">use_accelerator</span><span class="p">,</span>
+                <span class="n">ds_config</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">ds_config</span><span class="p">,</span>
+            <span class="p">)</span>
+            
+        <span class="bp">self</span><span class="o">.</span><span class="n">_activated</span> <span class="o">=</span> <span class="kc">True</span></div>
+
+            
+            
+<div class="viewcode-block" id="HFModelMixin.deactivate_model_for_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.deactivate_model_for_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">deactivate_model_for_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">use_vllm</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Deactivate the model and release the resources.</span>
+<span class="sd">        </span>
+<span class="sd">        NOTE: Currently, VLLM doesn&#39;t have an official way to do this, and the</span>
+<span class="sd">        implementation below cannot release all gpu resources by our observation.</span>
+<span class="sd">        Thus this method is just a placeholder for future implementation. See: </span>
+<span class="sd">        [Github issue](https://github.com/vllm-project/vllm/issues/1908)</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_activated</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;You are trying to deactivate the model for inference, but it is already deactivated.&quot;</span><span class="p">)</span>
+            <span class="k">return</span>
+        
+        <span class="k">if</span> <span class="n">use_vllm</span><span class="p">:</span>
+            <span class="n">destroy_model_parallel</span><span class="p">()</span>
+            <span class="k">del</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model_for_inference</span><span class="o">.</span><span class="n">llm_engine</span><span class="o">.</span><span class="n">model_executor</span><span class="o">.</span><span class="n">driver_worker</span>
+            <span class="k">del</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model_for_inference</span>
+            <span class="n">gc</span><span class="o">.</span><span class="n">collect</span><span class="p">()</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">empty_cache</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="s2">&quot;cpu&quot;</span><span class="p">)</span>
+            <span class="k">pass</span>
+        
+        <span class="bp">self</span><span class="o">.</span><span class="n">_activated</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+
+    
+<div class="viewcode-block" id="HFModelMixin.get_max_length">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.get_max_length">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_max_length</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Return max acceptable input length in terms of tokens.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span></div>
+
+
+
+<div class="viewcode-block" id="HFModelMixin.get_tokenizer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.get_tokenizer">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_tokenizer</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Return the tokenizer of the model.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span></div>
+
+
+
+<div class="viewcode-block" id="HFModelMixin.get_backend_model">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.get_backend_model">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_backend_model</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Return the backend model.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/models/hf_text_regression_model.html b/_modules/lmflow/models/hf_text_regression_model.html
new file mode 100644
index 000000000..07e1025c0
--- /dev/null
+++ b/_modules/lmflow/models/hf_text_regression_model.html
@@ -0,0 +1,958 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.models.hf_text_regression_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/models/hf_text_regression_model';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.hf_text_regression_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.models.hf_text_regression_model</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">copy</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">hashlib</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">pathlib</span><span class="w"> </span><span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">peft</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">LoraConfig</span><span class="p">,</span>
+    <span class="n">PeftModel</span><span class="p">,</span>
+    <span class="n">TaskType</span><span class="p">,</span>
+    <span class="n">get_peft_config</span><span class="p">,</span>
+    <span class="n">get_peft_model</span><span class="p">,</span>
+    <span class="n">prepare_model_for_kbit_training</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.modeling_outputs</span><span class="w"> </span><span class="kn">import</span> <span class="n">SequenceClassifierOutputWithPast</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="n">ModelArguments</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets.dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span><span class="p">,</span> <span class="n">KEY_SCORE</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.interfaces.tunable</span><span class="w"> </span><span class="kn">import</span> <span class="n">Tunable</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_model_mixin</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFModelMixin</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.text_regression_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">TextRegressionModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.tokenization.hf_text_regression_model</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">paired_conversation_tokenize_function</span><span class="p">,</span> 
+    <span class="n">conversation_tokenize_function</span><span class="p">,</span>
+    <span class="n">tokenize_function</span><span class="p">,</span>
+    <span class="n">text_to_textlist_tokenize_function</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.conversation_template</span><span class="w"> </span><span class="kn">import</span> <span class="n">PRESET_TEMPLATES</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">PAIRED_CONVERSATION_DATASET_DESCRIPTION</span><span class="p">,</span> 
+    <span class="n">TEXT2TEXT_DATASET_DESCRIPTION</span><span class="p">,</span>
+    <span class="n">TEXT_ONLY_DATASET_DESCRIPTION</span><span class="p">,</span>
+    <span class="n">TEXT_TO_TEXTLIST_DATASET_DESCRIPTION</span><span class="p">,</span>
+    <span class="n">CONVERSATION_DATASET_DESCRIPTION</span><span class="p">,</span> 
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.data_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">RewardModelInferenceResultWithInput</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.versioning</span><span class="w"> </span><span class="kn">import</span> <span class="n">is_ray_available</span><span class="p">,</span> <span class="n">is_vllm_available</span>
+
+<span class="k">if</span> <span class="n">is_ray_available</span><span class="p">():</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">ray</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">ray.data</span>
+    
+<span class="k">if</span> <span class="n">is_vllm_available</span><span class="p">():</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">vllm</span><span class="w"> </span><span class="kn">import</span> <span class="n">SamplingParams</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="HFTextRegressionModel">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">HFTextRegressionModel</span><span class="p">(</span><span class="n">TextRegressionModel</span><span class="p">,</span> <span class="n">HFModelMixin</span><span class="p">,</span> <span class="n">Tunable</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Initializes a HFTextRegressionModel instance.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+
+<span class="sd">    model_args : </span>
+<span class="sd">        Model arguments such as model name, path, revision, etc.</span>
+
+<span class="sd">    tune_strategy : str or none,  default=&quot;normal&quot;.</span>
+<span class="sd">        A string representing the dataset backend. Defaults to &quot;huggingface&quot;.</span>
+<span class="sd">    </span>
+<span class="sd">    ds_config :   </span>
+<span class="sd">        Deepspeed configuations.</span>
+<span class="sd">    </span>
+<span class="sd">    args : Optional.</span>
+<span class="sd">        Positional arguments.</span>
+<span class="sd">    </span>
+<span class="sd">    kwargs : Optional.</span>
+<span class="sd">        Keyword arguments.    </span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">tune_strategy</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s1">&#39;normal&#39;</span><span class="p">,</span>
+        <span class="n">ds_config</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">device</span><span class="o">=</span><span class="s2">&quot;gpu&quot;</span><span class="p">,</span>
+        <span class="n">use_accelerator</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Initializes a HFTextRegressionModel instance.</span>
+<span class="sd">        :param model_args: dictionary with model arguments such as model name, path, revision, etc.</span>
+<span class="sd">        :param tune_strategy: tuning strategy: normal, none, lora or adapter</span>
+<span class="sd">        :param ds_config: deepspeed configuration for distributed training</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">assert</span> <span class="n">model_args</span><span class="o">.</span><span class="n">arch_type</span> <span class="o">==</span> <span class="s2">&quot;text_regression&quot;</span><span class="p">,</span> <span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Invalid model architecture type: </span><span class="si">{</span><span class="n">model_args</span><span class="o">.</span><span class="n">arch_type</span><span class="si">}</span><span class="s2">. &quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;Expected: text_regression&quot;</span>
+        <span class="p">)</span>
+        <span class="n">config_additional_args</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;num_labels&quot;</span><span class="p">:</span> <span class="mi">1</span><span class="p">}</span>
+        <span class="n">HFModelMixin</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="bp">self</span><span class="p">,</span>
+            <span class="n">model_args</span><span class="o">=</span><span class="n">model_args</span><span class="p">,</span>
+            <span class="n">do_train</span><span class="o">=</span><span class="kc">True</span> <span class="k">if</span> <span class="n">tune_strategy</span> <span class="o">==</span> <span class="s2">&quot;normal&quot;</span> <span class="k">else</span> <span class="kc">False</span><span class="p">,</span>
+            <span class="n">ds_config</span><span class="o">=</span><span class="n">ds_config</span><span class="p">,</span>
+            <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">,</span>
+            <span class="n">use_accelerator</span><span class="o">=</span><span class="n">use_accelerator</span><span class="p">,</span>
+            <span class="n">hf_auto_model_additional_args</span><span class="o">=</span><span class="n">config_additional_args</span><span class="p">,</span>
+            <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+            <span class="o">**</span><span class="n">kwargs</span>
+        <span class="p">)</span>
+
+    
+<div class="viewcode-block" id="HFTextRegressionModel.tokenize">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.tokenize">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">tokenize</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span> 
+        <span class="n">add_special_tokens</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> 
+        <span class="o">*</span><span class="n">args</span><span class="p">,</span> 
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Tokenize the full dataset.</span>
+<span class="sd">    </span>
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        dataset : lmflow.datasets.Dataset.</span>
+
+<span class="sd">        args : Optional.</span>
+<span class="sd">            Positional arguments.</span>
+<span class="sd">        </span>
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.    </span>
+<span class="sd">        </span>
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        tokenized_datasets :</span>
+<span class="sd">            The tokenized dataset, without any leading or trailing special</span>
+<span class="sd">            tokens (normally they are Begin-Of-Sentence or End-Of-Sentence</span>
+<span class="sd">            tokens).</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># Preprocessing the datasets.</span>
+        <span class="c1"># First we tokenize all the texts.</span>
+        <span class="k">if</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_backend</span><span class="p">()</span> <span class="o">!=</span> <span class="s2">&quot;huggingface&quot;</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="s2">&quot;tokenization of datasets with non-huggingface backend are&quot;</span>
+                <span class="s2">&quot;not supported yet&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">dataset_type</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span>
+        <span class="n">model_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span>
+        <span class="n">raw_datasets</span> <span class="o">=</span> <span class="n">dataset</span>
+        <span class="n">hf_raw_datasets</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()</span>
+        <span class="n">column_names</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">hf_raw_datasets</span><span class="o">.</span><span class="n">features</span><span class="p">)</span> <span class="c1"># in paired conversation, for example, would be &#39;chosen&#39; and &#39;rejected&#39;</span>
+        <span class="n">data_args</span> <span class="o">=</span> <span class="n">raw_datasets</span><span class="o">.</span><span class="n">get_data_args</span><span class="p">()</span>
+        
+        <span class="c1"># Whether to truncate long sequences to fit into max_length</span>
+        <span class="n">use_truncation</span> <span class="o">=</span> <span class="kc">False</span>
+        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_lora</span> <span class="ow">or</span> <span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+            <span class="n">use_truncation</span> <span class="o">=</span> <span class="kc">True</span>
+
+        <span class="c1"># Requires three types of information for tokenizing different datasets</span>
+        <span class="c1">#   1) Which fields require tokenization, e.g.</span>
+        <span class="c1">#        &quot;text2float&quot;: &quot;text&quot;, but not &quot;float&quot;</span>
+        <span class="c1">#        &quot;text2text&quot;: both &quot;input&quot; and &quot;output&quot;</span>
+        <span class="c1">#   2) How will there tokenized sequence concatenated together, e.g.</span>
+        <span class="c1">#        &quot;text_only&quot;: &quot;text&quot; -&gt; &quot;text&quot;</span>
+        <span class="c1">#        &quot;text2text&quot;: &quot;input&quot;, &quot;output&quot; -&gt; &quot;input&quot; + &quot;output&quot;</span>
+        <span class="c1">#   3) Which fields require loss in final computation, e.g.</span>
+        <span class="c1">#        &quot;text_only&quot;: &quot;text&quot;</span>
+        <span class="c1">#        &quot;text2text&quot;: &quot;output&quot; only</span>
+        <span class="n">tokenize_fn</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">tokenize_fn_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;data_args&quot;</span><span class="p">:</span> <span class="n">data_args</span><span class="p">,</span>
+            <span class="s2">&quot;tokenizer&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">,</span>
+            <span class="s2">&quot;column_names&quot;</span><span class="p">:</span> <span class="n">column_names</span><span class="p">,</span>
+        <span class="p">}</span>            
+        <span class="k">if</span> <span class="n">dataset_type</span> <span class="o">==</span> <span class="s2">&quot;text_only&quot;</span><span class="p">:</span>
+            <span class="n">tokenize_fn</span> <span class="o">=</span> <span class="n">tokenize_function</span>
+            <span class="n">text_only_tokenize_fn_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="s2">&quot;tokenized_column_order&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;text&quot;</span><span class="p">],</span>
+                <span class="s2">&quot;label_columns&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;text&quot;</span><span class="p">],</span>
+                <span class="s2">&quot;add_special_tokens&quot;</span><span class="p">:</span> <span class="n">add_special_tokens</span><span class="p">,</span>
+                <span class="s2">&quot;use_truncation&quot;</span><span class="p">:</span> <span class="n">use_truncation</span><span class="p">,</span>
+            <span class="p">}</span>
+            <span class="n">tokenize_fn_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">text_only_tokenize_fn_kwargs</span><span class="p">)</span>
+            
+        <span class="k">elif</span> <span class="n">dataset_type</span> <span class="o">==</span> <span class="s2">&quot;text2text&quot;</span><span class="p">:</span>
+            <span class="n">tokenize_fn</span> <span class="o">=</span> <span class="n">tokenize_function</span>
+            <span class="n">text2text_tokenize_fn_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="s2">&quot;tokenized_column_order&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">,</span> <span class="s2">&quot;output&quot;</span><span class="p">],</span>
+                <span class="s2">&quot;label_columns&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">],</span>
+                <span class="s2">&quot;add_special_tokens&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
+                <span class="s2">&quot;use_truncation&quot;</span><span class="p">:</span> <span class="n">use_truncation</span><span class="p">,</span>
+            <span class="p">}</span>
+            <span class="n">tokenize_fn_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">text2text_tokenize_fn_kwargs</span><span class="p">)</span>
+            
+        <span class="k">elif</span> <span class="n">dataset_type</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;conversation&quot;</span><span class="p">,</span> <span class="s2">&quot;paired_conversation&quot;</span><span class="p">]:</span>
+            <span class="k">if</span> <span class="n">dataset_type</span> <span class="o">==</span> <span class="s2">&quot;conversation&quot;</span><span class="p">:</span>
+                <span class="n">tokenize_fn</span> <span class="o">=</span> <span class="n">conversation_tokenize_function</span>
+            <span class="k">elif</span> <span class="n">dataset_type</span> <span class="o">==</span> <span class="s2">&quot;paired_conversation&quot;</span><span class="p">:</span>
+                <span class="n">tokenize_fn</span> <span class="o">=</span> <span class="n">paired_conversation_tokenize_function</span>
+            
+            <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">conversation_template</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">conversation_template</span> <span class="ow">in</span> <span class="n">PRESET_TEMPLATES</span><span class="o">.</span><span class="n">keys</span><span class="p">():</span>
+                    <span class="n">conversation_template</span> <span class="o">=</span> <span class="n">PRESET_TEMPLATES</span><span class="p">[</span><span class="n">data_args</span><span class="o">.</span><span class="n">conversation_template</span><span class="p">]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;Conversation template </span><span class="si">{</span><span class="n">data_args</span><span class="o">.</span><span class="n">conversation_template</span><span class="si">}</span><span class="s2"> is not supported yet.&quot;</span>
+                    <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;No conversation template provided. Using default template.&quot;</span><span class="p">)</span>
+                <span class="n">conversation_template</span> <span class="o">=</span> <span class="n">PRESET_TEMPLATES</span><span class="p">[</span><span class="s1">&#39;empty&#39;</span><span class="p">]</span>
+            <span class="n">tokenize_fn_kwargs</span><span class="p">[</span><span class="s2">&quot;conversation_template&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">conversation_template</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Conversation template: </span><span class="si">{</span><span class="n">conversation_template</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            
+        <span class="k">elif</span> <span class="n">dataset_type</span> <span class="o">==</span> <span class="s2">&quot;text_to_textlist&quot;</span><span class="p">:</span>
+            <span class="n">tokenize_fn</span> <span class="o">=</span> <span class="n">text_to_textlist_tokenize_function</span>
+            <span class="n">text_to_textlist_tokenize_fn_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="s2">&quot;add_special_tokens&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
+                <span class="s2">&quot;use_truncation&quot;</span><span class="p">:</span> <span class="n">use_truncation</span><span class="p">,</span>
+            <span class="p">}</span>
+            <span class="n">tokenize_fn_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">text_to_textlist_tokenize_fn_kwargs</span><span class="p">)</span>
+            
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Dataset type </span><span class="se">\&quot;</span><span class="si">{</span><span class="n">dataset_type</span><span class="si">}</span><span class="se">\&quot;</span><span class="s2"> is not supported, currently&quot;</span>
+                <span class="s2">&quot; only support following data types for HFTextRegressionModel:</span><span class="se">\n</span><span class="s2">&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;    1) [Inference]</span><span class="si">{</span><span class="n">TEXT_ONLY_DATASET_DESCRIPTION</span><span class="si">}</span><span class="se">\n</span><span class="s2">&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;    2) [Inference]</span><span class="si">{</span><span class="n">TEXT2TEXT_DATASET_DESCRIPTION</span><span class="si">}</span><span class="se">\n</span><span class="s2">&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;    3) [Training]</span><span class="si">{</span><span class="n">PAIRED_CONVERSATION_DATASET_DESCRIPTION</span><span class="si">}</span><span class="se">\n</span><span class="s2">&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;    4) [Inference]</span><span class="si">{</span><span class="n">CONVERSATION_DATASET_DESCRIPTION</span><span class="si">}</span><span class="se">\n</span><span class="s2">&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;    5) [Inference]</span><span class="si">{</span><span class="n">TEXT_TO_TEXTLIST_DATASET_DESCRIPTION</span><span class="si">}</span><span class="se">\n</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+  
+        <span class="n">tokenize_kwargs</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">data_args</span><span class="o">.</span><span class="n">streaming</span><span class="p">:</span>
+            <span class="n">fingerprint</span> <span class="o">=</span> <span class="n">hashlib</span><span class="o">.</span><span class="n">md5</span><span class="p">(</span>
+                <span class="p">(</span>
+                    <span class="n">raw_datasets</span><span class="o">.</span><span class="n">get_fingerprint</span><span class="p">()</span>
+                    <span class="o">+</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">)</span>
+                    <span class="o">+</span> <span class="sa">f</span><span class="s1">&#39;###padding_side=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">padding_side</span><span class="si">}</span><span class="s1">&#39;</span>
+                    <span class="o">+</span> <span class="p">(</span><span class="s1">&#39;###conversation_template=&#39;</span> <span class="o">+</span> <span class="nb">str</span><span class="p">(</span><span class="n">conversation_template</span><span class="p">)</span> <span class="k">if</span> <span class="s2">&quot;conversation&quot;</span> <span class="ow">in</span> <span class="n">dataset_type</span> <span class="k">else</span> <span class="s2">&quot;&quot;</span><span class="p">)</span>
+                    <span class="o">+</span> <span class="sa">f</span><span class="s1">&#39;###disable_group_texts=</span><span class="si">{</span><span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="si">}</span><span class="s1">&#39;</span>
+                    <span class="o">+</span> <span class="sa">f</span><span class="s1">&#39;###block_size=</span><span class="si">{</span><span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span><span class="si">}</span><span class="s1">&#39;</span>
+                <span class="p">)</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="s2">&quot;utf-8&quot;</span><span class="p">)</span>
+            <span class="p">)</span><span class="o">.</span><span class="n">hexdigest</span><span class="p">()</span>
+            <span class="n">tokenize_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="s2">&quot;num_proc&quot;</span><span class="p">:</span> <span class="n">data_args</span><span class="o">.</span><span class="n">preprocessing_num_workers</span><span class="p">,</span>
+                <span class="s2">&quot;load_from_cache_file&quot;</span><span class="p">:</span> <span class="ow">not</span> <span class="n">data_args</span><span class="o">.</span><span class="n">overwrite_cache</span><span class="p">,</span>
+                <span class="s2">&quot;desc&quot;</span><span class="p">:</span> <span class="s2">&quot;Running tokenizer on dataset&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;new_fingerprint&quot;</span><span class="p">:</span> <span class="n">fingerprint</span><span class="p">,</span>
+            <span class="p">}</span>
+
+        <span class="n">tokenized_datasets</span> <span class="o">=</span> <span class="n">raw_datasets</span><span class="o">.</span><span class="n">map</span><span class="p">(</span>
+            <span class="n">tokenize_fn</span><span class="p">,</span>
+            <span class="n">batched</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+            <span class="n">remove_columns</span><span class="o">=</span><span class="n">column_names</span><span class="p">,</span>
+            <span class="n">fn_kwargs</span><span class="o">=</span><span class="n">tokenize_fn_kwargs</span><span class="p">,</span>
+            <span class="o">**</span><span class="n">tokenize_kwargs</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">tokenized_datasets</span></div>
+
+
+
+<div class="viewcode-block" id="HFTextRegressionModel.inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">inputs</span><span class="p">,</span> 
+        <span class="n">release_gpu</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">use_vllm</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">SequenceClassifierOutputWithPast</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform generation process of the model.</span>
+<span class="sd">    </span>
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        inputs :</span>
+<span class="sd">            The sequence used as a prompt for the generation or as model inputs to the model.</span>
+<span class="sd">            When using vllm inference, this should be a string or a list of strings.</span>
+<span class="sd">            When using normal inference, this should be a tensor.</span>
+<span class="sd">        release_gpu : bool, optional</span>
+<span class="sd">            Whether to release the GPU resource after inference, by default False.</span>
+<span class="sd">        use_vllm : bool, optional</span>
+<span class="sd">            Whether to use VLLM for inference, by default False.</span>
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.    </span>
+<span class="sd">        </span>
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        outputs :</span>
+<span class="sd">            The generated sequence output </span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">use_vllm</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                <span class="s2">&quot;VLLM inference is not supported for text regression model, using normal inference instead.&quot;</span>
+            <span class="p">)</span>
+            <span class="n">use_vllm</span> <span class="o">=</span> <span class="kc">False</span>
+            
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">_activated</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">activate_model_for_inference</span><span class="p">(</span>
+                <span class="n">use_vllm</span><span class="o">=</span><span class="n">use_vllm</span><span class="p">,</span>
+                <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+            <span class="p">)</span>
+            
+        <span class="k">if</span> <span class="n">use_vllm</span><span class="p">:</span>
+            <span class="n">res</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__vllm_inference</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">res</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__inference</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+            
+        <span class="k">if</span> <span class="n">release_gpu</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">deactivate_model_for_inference</span><span class="p">(</span><span class="n">use_vllm</span><span class="o">=</span><span class="n">use_vllm</span><span class="p">)</span>
+            
+        <span class="k">return</span> <span class="n">res</span></div>
+
+
+
+<div class="viewcode-block" id="HFTextRegressionModel.__inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.__inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">inputs</span><span class="p">,</span> 
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform generation process of the model.</span>
+<span class="sd">    </span>
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        inputs :</span>
+<span class="sd">            The **tokenized** sequence used as a prompt for the generation or as model inputs to the model.</span>
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.    </span>
+<span class="sd">        </span>
+<span class="sd">        Returns</span>
+<span class="sd">        ------------</span>
+<span class="sd">        outputs :</span>
+<span class="sd">            The generated sequence output </span>
+<span class="sd">        &quot;&quot;&quot;</span>       
+        <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_accelerator</span><span class="p">:</span>
+                <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="p">(</span>
+                    <span class="n">input_ids</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span>
+                    <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;gpu&quot;</span><span class="p">:</span>
+                    <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ds_engine</span><span class="o">.</span><span class="n">module</span><span class="p">(</span>
+                        <span class="n">input_ids</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span>
+                        <span class="n">synced_gpus</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+                    <span class="p">)</span>
+                <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;cpu&quot;</span><span class="p">:</span>
+                    <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">backend_model</span><span class="p">(</span>
+                        <span class="n">input_ids</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span>
+                        <span class="n">synced_gpus</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+                    <span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;device </span><span class="se">\&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="se">\&quot;</span><span class="s2"> is not supported&quot;</span>
+                    <span class="p">)</span>
+        
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">&#39;return_input&#39;</span><span class="p">,</span> <span class="kc">False</span><span class="p">):</span>
+            <span class="n">outputs</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;input&quot;</span><span class="p">:</span> <span class="n">inputs</span><span class="p">,</span> <span class="s2">&quot;output&quot;</span><span class="p">:</span> <span class="n">outputs</span><span class="p">}</span>
+        
+        <span class="k">return</span> <span class="n">outputs</span></div>
+
+    
+    
+<div class="viewcode-block" id="HFTextRegressionModel.__vllm_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.__vllm_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__vllm_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">inputs</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]],</span>
+        <span class="n">sampling_params</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="s1">&#39;SamplingParams&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]],</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Perform VLLM inference process of the model.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ----------</span>
+<span class="sd">        inputs : Union[str, List[str]]</span>
+<span class="sd">            Prompt(s), string or a list of strings.</span>
+<span class="sd">        sampling_params : Optional[SamplingParams], optional</span>
+<span class="sd">            vllm SamplingParams object, by default None.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        -------</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+            <span class="s2">&quot;VLLM inference is not supported for text regression model.&quot;</span>
+        <span class="p">)</span></div>
+
+        
+        
+<div class="viewcode-block" id="HFTextRegressionModel.prepare_inputs_for_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.prepare_inputs_for_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">prepare_inputs_for_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">enable_distributed_inference</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">use_vllm</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">Dataset</span><span class="p">,</span> <span class="s1">&#39;ray.data.Dataset&#39;</span><span class="p">]:</span>
+        <span class="k">if</span> <span class="n">use_vllm</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="s2">&quot;VLLM inference is not supported for text regression model.&quot;</span>
+            <span class="p">)</span>
+            
+        <span class="n">inference_inputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenize</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span>
+                
+        <span class="k">if</span> <span class="n">enable_distributed_inference</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">is_ray_available</span><span class="p">():</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="s1">&#39;Ray is not available. Please install ray via `pip install -e &quot;.[ray]&quot;`.&#39;</span>
+                <span class="p">)</span>
+            
+            <span class="n">inference_inputs</span><span class="o">.</span><span class="n">sanity_check</span><span class="p">(</span><span class="n">drop_invalid</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="n">inference_inputs</span> <span class="o">=</span> <span class="n">inference_inputs</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()</span>
+            <span class="n">inference_inputs</span> <span class="o">=</span> <span class="n">ray</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">from_items</span><span class="p">(</span><span class="n">inference_inputs</span><span class="p">)</span> 
+            <span class="c1"># -&gt; Dict[str, np.ndarray]</span>
+            <span class="c1"># Example (batch size=2):</span>
+            <span class="c1"># {&#39;input&#39;: array([&#39;...&#39;,&#39;...&#39;], dtype=object),</span>
+            <span class="c1">#  &#39;output&#39;: array([array([&quot;...&quot;, &quot;...&quot;], dtype=object), array([&#39;...&#39;,&#39;...&#39;], dtype=object)], dtype=object),</span>
+            <span class="c1">#  &#39;input_ids&#39;: array(</span>
+            <span class="c1">#      [</span>
+            <span class="c1">#          array([array([    27,     91,    882, ..., 128256, 128256, 128256]),</span>
+            <span class="c1">#                 array([    27,     91,    882, ..., 128256, 128256, 128256])],</span>
+            <span class="c1">#                dtype=object),</span>
+            <span class="c1">#          array([array([    27,     91,    882, ..., 128256, 128256, 128256]),</span>
+            <span class="c1">#                 array([    27,     91,    882, ..., 128256, 128256, 128256])],</span>
+            <span class="c1">#                dtype=object)</span>
+            <span class="c1">#      ], dtype=object)}</span>
+        
+        <span class="k">return</span> <span class="n">inference_inputs</span></div>
+
+            
+    
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="HFTextRegressionModel.postprocess_inference_outputs">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.postprocess_inference_outputs">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">postprocess_inference_outputs</span><span class="p">(</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">scores</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]],</span>
+    <span class="p">):</span>
+        <span class="n">output_dict</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;</span><span class="p">,</span> <span class="s2">&quot;instances&quot;</span><span class="p">:</span> <span class="p">[]}</span>
+        <span class="k">if</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="o">==</span> <span class="s2">&quot;text_to_textlist&quot;</span><span class="p">:</span>
+            <span class="n">output_dict</span><span class="p">[</span><span class="s2">&quot;type&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;text_to_scored_textlist&quot;</span>
+            <span class="k">for</span> <span class="n">idx</span><span class="p">,</span> <span class="n">instance</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()):</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">instance</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">])</span> <span class="o">&lt;</span> <span class="mi">2</span><span class="p">:</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Instance </span><span class="si">{</span><span class="n">idx</span><span class="si">}</span><span class="s2"> has less than 2 outputs, skipping.&quot;</span><span class="p">)</span>
+                <span class="n">output_dict</span><span class="p">[</span><span class="s2">&quot;instances&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                    <span class="p">{</span>
+                        <span class="s2">&quot;input&quot;</span><span class="p">:</span> <span class="n">instance</span><span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">],</span>
+                        <span class="s2">&quot;output&quot;</span><span class="p">:</span> <span class="p">[{</span><span class="s2">&quot;text&quot;</span><span class="p">:</span> <span class="n">text</span><span class="p">}</span> <span class="k">for</span> <span class="n">text</span> <span class="ow">in</span> <span class="n">instance</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">]],</span>
+                    <span class="p">}</span>
+                <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Dataset type </span><span class="si">{</span><span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span><span class="si">}</span><span class="s2"> is not supported for reward model inference.&quot;</span><span class="p">)</span>
+        
+        <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">instance_scores</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">scores</span><span class="p">):</span>
+            <span class="k">for</span> <span class="n">j</span><span class="p">,</span> <span class="n">score</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">instance_scores</span><span class="p">):</span>
+                <span class="n">output_dict</span><span class="p">[</span><span class="s2">&quot;instances&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">][</span><span class="s2">&quot;output&quot;</span><span class="p">][</span><span class="n">j</span><span class="p">][</span><span class="n">KEY_SCORE</span><span class="p">]</span> <span class="o">=</span> <span class="n">score</span>
+        
+        <span class="n">output_dataset_args</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">dataset</span><span class="o">.</span><span class="n">data_args</span><span class="p">)</span>
+        <span class="n">output_dataset_args</span><span class="o">.</span><span class="n">dataset_path</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">output_dataset_args</span><span class="o">.</span><span class="n">dataset_name</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">output_dataset_args</span><span class="o">.</span><span class="n">dataset_name</span><span class="si">}</span><span class="s2">_scored&quot;</span>
+        <span class="n">output_dataset</span> <span class="o">=</span> <span class="n">Dataset</span><span class="p">(</span><span class="n">output_dataset_args</span><span class="p">)</span>
+        <span class="n">output_dataset</span> <span class="o">=</span> <span class="n">output_dataset</span><span class="o">.</span><span class="n">from_dict</span><span class="p">(</span><span class="n">output_dict</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">output_dataset</span></div>
+
+    
+    
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="HFTextRegressionModel.postprocess_distributed_inference_outputs">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.postprocess_distributed_inference_outputs">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">postprocess_distributed_inference_outputs</span><span class="p">(</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">inference_result</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">RewardModelInferenceResultWithInput</span><span class="p">],</span>
+    <span class="p">):</span>
+        <span class="n">output_dict</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="s2">&quot;text_to_scored_textlist&quot;</span><span class="p">,</span> <span class="s2">&quot;instances&quot;</span><span class="p">:</span> <span class="n">inference_result</span><span class="p">}</span>
+        <span class="n">output_dataset_args</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">dataset</span><span class="o">.</span><span class="n">data_args</span><span class="p">)</span>
+        <span class="n">output_dataset_args</span><span class="o">.</span><span class="n">dataset_path</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">output_dataset_args</span><span class="o">.</span><span class="n">dataset_name</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">output_dataset_args</span><span class="o">.</span><span class="n">dataset_name</span><span class="si">}</span><span class="s2">_scored&quot;</span>
+        <span class="n">output_dataset</span> <span class="o">=</span> <span class="n">Dataset</span><span class="p">(</span><span class="n">output_dataset_args</span><span class="p">)</span>
+        <span class="n">output_dataset</span> <span class="o">=</span> <span class="n">output_dataset</span><span class="o">.</span><span class="n">from_dict</span><span class="p">(</span><span class="n">output_dict</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">output_dataset</span></div>
+
+        
+
+<div class="viewcode-block" id="HFTextRegressionModel.save">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.save">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">save</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="nb">dir</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform generation process of the model.</span>
+<span class="sd">    </span>
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        dir :</span>
+<span class="sd">            The directory to save model and tokenizer</span>
+<span class="sd">        </span>
+<span class="sd">        kwargs : Optional.</span>
+<span class="sd">            Keyword arguments.    </span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">get_tokenizer</span><span class="p">()</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="nb">dir</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="nb">dir</span><span class="p">)</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/models/interfaces/tunable.html b/_modules/lmflow/models/interfaces/tunable.html
new file mode 100644
index 000000000..525fd1a24
--- /dev/null
+++ b/_modules/lmflow/models/interfaces/tunable.html
@@ -0,0 +1,454 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.models.interfaces.tunable &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/models/interfaces/tunable';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.interfaces.tunable</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.models.interfaces.tunable</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;Tunable class</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">abc</span><span class="w"> </span><span class="kn">import</span> <span class="n">ABC</span>
+
+
+<div class="viewcode-block" id="Tunable">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/interfaces/tunable/index.html#lmflow.models.interfaces.tunable.Tunable">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Tunable</span><span class="p">(</span><span class="n">ABC</span><span class="p">):</span>
+    <span class="k">pass</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/models/regression_model.html b/_modules/lmflow/models/regression_model.html
new file mode 100644
index 000000000..8da269f8f
--- /dev/null
+++ b/_modules/lmflow/models/regression_model.html
@@ -0,0 +1,455 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.models.regression_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/models/regression_model';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.regression_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.models.regression_model</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;General regression model.&quot;&quot;&quot;</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.base_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">BaseModel</span>
+
+
+<div class="viewcode-block" id="RegressionModel">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/regression_model/index.html#lmflow.models.regression_model.RegressionModel">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">RegressionModel</span><span class="p">(</span><span class="n">BaseModel</span><span class="p">):</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+        <span class="k">pass</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/models/text_regression_model.html b/_modules/lmflow/models/text_regression_model.html
new file mode 100644
index 000000000..4eb18b844
--- /dev/null
+++ b/_modules/lmflow/models/text_regression_model.html
@@ -0,0 +1,510 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.models.text_regression_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/models/text_regression_model';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.text_regression_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.models.text_regression_model</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">A model maps &quot;text_only&quot; data to float.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.regression_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">RegressionModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets.dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+
+
+<div class="viewcode-block" id="TextRegressionModel">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/text_regression_model/index.html#lmflow.models.text_regression_model.TextRegressionModel">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">TextRegressionModel</span><span class="p">(</span><span class="n">RegressionModel</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Initializes a TextRegressionModel instance.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+
+<span class="sd">    model_args : </span>
+<span class="sd">        Model arguments such as model name, path, revision, etc.</span>
+
+<span class="sd">    args : Optional.</span>
+<span class="sd">        Positional arguments.</span>
+
+<span class="sd">    kwargs : Optional.</span>
+<span class="sd">        Keyword arguments.    </span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">,</span>
+        <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Initializes a TextRegressionModel instance.</span>
+<span class="sd">        :param model_args: dictionary with model arguments such as model name, path, revision, etc.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+<div class="viewcode-block" id="TextRegressionModel.inference_func">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/text_regression_model/index.html#lmflow.models.text_regression_model.TextRegressionModel.inference_func">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inference_func</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+
+
+<div class="viewcode-block" id="TextRegressionModel.register_inference_function">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/text_regression_model/index.html#lmflow.models.text_regression_model.TextRegressionModel.register_inference_function">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">register_inference_function</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inference_func</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Registers a regression function.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inference_func</span> <span class="o">=</span> <span class="n">inference_func</span></div>
+
+
+
+<div class="viewcode-block" id="TextRegressionModel.inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/text_regression_model/index.html#lmflow.models.text_regression_model.TextRegressionModel.inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">inference</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Gets regression results of a given dataset.</span>
+
+<span class="sd">        :inputs: Dataset object, only accept type &quot;text_only&quot;.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">inference_func</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">inference_func</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">pass</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/models/vision2seq_model.html b/_modules/lmflow/models/vision2seq_model.html
new file mode 100644
index 000000000..ccad5e5ae
--- /dev/null
+++ b/_modules/lmflow/models/vision2seq_model.html
@@ -0,0 +1,987 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.models.vision2seq_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/models/vision2seq_model';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.vision2seq_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.models.vision2seq_model</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># TODO update the doc</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">copy</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">time</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.nn</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">nn</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.nn</span><span class="w"> </span><span class="kn">import</span> <span class="n">CrossEntropyLoss</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">AutoModelForCausalLM</span><span class="p">,</span>
+    <span class="n">AutoModelForSeq2SeqLM</span><span class="p">,</span>
+    <span class="n">AutoModel</span><span class="p">,</span>
+    <span class="n">Blip2ForConditionalGeneration</span><span class="p">,</span>
+    <span class="n">Blip2Config</span><span class="p">,</span>
+    <span class="n">Blip2QFormerModel</span><span class="p">,</span>
+    <span class="n">Blip2VisionModel</span><span class="p">,</span>
+    <span class="n">Blip2PreTrainedModel</span><span class="p">,</span>
+    <span class="n">PreTrainedModel</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.modeling_outputs</span><span class="w"> </span><span class="kn">import</span> <span class="n">CausalLMOutputWithPast</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.base_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">BaseModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.vision_encoder</span><span class="w"> </span><span class="kn">import</span> <span class="n">build_vision_tower</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.versioning</span><span class="w"> </span><span class="kn">import</span> <span class="n">is_package_version_at_least</span>
+
+<span class="k">if</span> <span class="n">is_package_version_at_least</span><span class="p">(</span><span class="s1">&#39;transformers&#39;</span><span class="p">,</span> <span class="s1">&#39;4.46.0&#39;</span><span class="p">):</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.integrations.deepspeed</span><span class="w"> </span><span class="kn">import</span> <span class="n">is_deepspeed_zero3_enabled</span>
+<span class="k">else</span><span class="p">:</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.deepspeed</span><span class="w"> </span><span class="kn">import</span> <span class="n">is_deepspeed_zero3_enabled</span>
+
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">CustomAutoVision2SeqModel</span><span class="p">(</span><span class="n">Blip2ForConditionalGeneration</span><span class="p">,</span> <span class="n">BaseModel</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
+                 <span class="n">config</span><span class="p">:</span> <span class="n">Blip2Config</span><span class="p">,</span>
+                 <span class="n">image_encoder_name_or_path</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+                 <span class="n">qformer_name_or_path</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+                 <span class="n">language_model_name_or_path</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+                 <span class="n">low_resource</span><span class="o">=</span><span class="kc">False</span><span class="p">,):</span>
+<span class="w">        </span><span class="sd">&#39;&#39;&#39;</span>
+<span class="sd">        TODO update the docs</span>
+<span class="sd">        Args:</span>
+<span class="sd">            config:</span>
+<span class="sd">            # the below varaible are used to overwrite the model in config</span>
+<span class="sd">            image_encoder_name_or_path:</span>
+<span class="sd">            qformer_name_or_path:</span>
+<span class="sd">            language_model_name_or_path:</span>
+<span class="sd">        Returns:</span>
+<span class="sd">        &#39;&#39;&#39;</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">Blip2PreTrainedModel</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">config</span><span class="p">)</span>
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.custom_vision_model">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.custom_vision_model">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">custom_vision_model</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span>
+            <span class="n">config</span><span class="p">,</span> <span class="s2">&quot;custom_vision_model&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span></div>
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.with_qformer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.with_qformer">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">with_qformer</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">config</span><span class="p">,</span> <span class="s2">&quot;with_qformer&quot;</span><span class="p">,</span> <span class="kc">True</span><span class="p">)</span></div>
+
+        <span class="c1"># vision model</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">custom_vision_model</span><span class="p">:</span>
+            <span class="c1"># custom vision model means the vit model customized from llava.</span>
+            <span class="c1"># vision_model_args = getattr(config, &quot;vision_model_args&quot;, dict())</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">vision_model</span> <span class="o">=</span> <span class="n">build_vision_tower</span><span class="p">(</span><span class="n">config</span><span class="p">)</span>
+            <span class="n">config</span><span class="o">.</span><span class="n">vision_config</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">vision_model</span><span class="o">.</span><span class="n">config</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">image_processor</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">vision_model</span><span class="o">.</span><span class="n">image_processor</span>
+        <span class="k">elif</span> <span class="n">image_encoder_name_or_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="c1"># use the model from transformers</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">vision_model</span> <span class="o">=</span> <span class="n">AutoModel</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                <span class="n">image_encoder_name_or_path</span><span class="p">)</span>
+            <span class="n">config</span><span class="o">.</span><span class="n">vision_config</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">vision_model</span><span class="o">.</span><span class="n">config</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># the default vit in Blip2</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">vision_model</span> <span class="o">=</span> <span class="n">Blip2VisionModel</span><span class="p">(</span><span class="n">config</span><span class="o">.</span><span class="n">vision_config</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">with_qformer</span><span class="p">:</span>
+            <span class="c1"># check if with qformer, the blip series model use qformer</span>
+            <span class="c1"># and the llava based models don&#39;t use qformer.</span>
+            <span class="k">if</span> <span class="n">qformer_name_or_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">query_tokens</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">config</span><span class="o">.</span><span class="n">num_query_tokens</span><span class="p">,</span>
+                                <span class="n">config</span><span class="o">.</span><span class="n">qformer_config</span><span class="o">.</span><span class="n">hidden_size</span><span class="p">))</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">qformer</span> <span class="o">=</span> <span class="n">AutoModel</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                    <span class="n">qformer_name_or_path</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">query_tokens</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Parameter</span><span class="p">(</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">config</span><span class="o">.</span><span class="n">num_query_tokens</span><span class="p">,</span>
+                                <span class="n">config</span><span class="o">.</span><span class="n">qformer_config</span><span class="o">.</span><span class="n">hidden_size</span><span class="p">))</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">qformer</span> <span class="o">=</span> <span class="n">Blip2QFormerModel</span><span class="p">(</span><span class="n">config</span><span class="o">.</span><span class="n">qformer_config</span><span class="p">)</span>
+        <span class="n">kwargs</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">language_model_name_or_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">low_resource</span><span class="p">:</span>
+                <span class="n">kwargs</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span>
+                    <span class="n">torch_dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float16</span><span class="p">,</span>
+                    <span class="n">load_in_8bit</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                    <span class="n">device_map</span><span class="o">=</span><span class="s2">&quot;auto&quot;</span><span class="p">,</span>
+                    <span class="n">low_cpu_mem_usage</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="n">is_deepspeed_zero3_enabled</span><span class="p">:</span>
+                    <span class="n">kwargs</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">device_map</span><span class="o">=</span><span class="s2">&quot;auto&quot;</span><span class="p">,</span>
+                            <span class="n">torch_dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float16</span><span class="p">)</span>
+            <span class="n">language_model</span> <span class="o">=</span> <span class="n">AutoModelForCausalLM</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                <span class="n">language_model_name_or_path</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+            <span class="n">config</span><span class="o">.</span><span class="n">text_config</span> <span class="o">=</span> <span class="n">language_model</span><span class="o">.</span><span class="n">config</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">config</span><span class="o">.</span><span class="n">use_decoder_only_language_model</span><span class="p">:</span>
+                <span class="n">language_model</span> <span class="o">=</span> <span class="n">AutoModelForCausalLM</span><span class="o">.</span><span class="n">from_config</span><span class="p">(</span>
+                    <span class="n">config</span><span class="o">.</span><span class="n">text_config</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">language_model</span> <span class="o">=</span> <span class="n">AutoModelForSeq2SeqLM</span><span class="o">.</span><span class="n">from_config</span><span class="p">(</span>
+                    <span class="n">config</span><span class="o">.</span><span class="n">text_config</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="c1"># Update _tied_weights_keys using the base model used.</span>
+        <span class="k">if</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">language_model</span><span class="p">,</span> <span class="s2">&quot;_tied_weights_keys&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_tied_weights_keys</span> <span class="o">=</span> <span class="p">[</span><span class="sa">f</span><span class="s2">&quot;language_model.</span><span class="si">{</span><span class="n">k</span><span class="si">}</span><span class="s2">&quot;</span> <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">language_model</span><span class="o">.</span><span class="n">_tied_weights_keys</span><span class="p">]</span>
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.language_model">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.language_model">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span> <span class="o">=</span> <span class="n">language_model</span></div>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">with_qformer</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">language_projection</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">qformer</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">hidden_size</span><span class="p">,</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">hidden_size</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">language_projection</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">vision_model</span><span class="o">.</span><span class="n">hidden_size</span><span class="p">,</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">hidden_size</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">image_encoder_name_or_path</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> \
+           <span class="n">language_model_name_or_path</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">post_init</span><span class="p">()</span>
+        <span class="c1"># for deepspeed</span>
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.hidden_size">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.hidden_size">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">hidden_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">hidden_size</span></div>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">hidden_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">hidden_size</span>
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.get_backend_model">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.get_backend_model">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_backend_model</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="bp">self</span></div>
+
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.vision_model_from_pretrained">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.vision_model_from_pretrained">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">vision_model_from_pretrained</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">pretrained_path</span><span class="p">):</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">vision_model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">vision_model</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                                <span class="n">pretrained_path</span><span class="p">,</span>
+                                <span class="n">config</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">vision_config</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.qformer_from_pretrained">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.qformer_from_pretrained">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">qformer_from_pretrained</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">pretrained_path</span><span class="p">):</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">qformer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">qformer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                                <span class="n">pretrained_path</span><span class="p">,</span>
+                                <span class="n">config</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">qformer_config</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.language_model_from_pretrained">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.language_model_from_pretrained">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">language_model_from_pretrained</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
+                                       <span class="n">pretrained_path</span><span class="p">,</span>
+                                       <span class="n">low_resource</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                                       <span class="n">use_prompt_cache</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+        <span class="c1"># TODO remove the low resource related loading in the future</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_prompt_cache</span> <span class="o">=</span> <span class="n">use_prompt_cache</span>
+        <span class="k">if</span> <span class="n">low_resource</span><span class="p">:</span>
+            <span class="n">kwargs</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span>
+                <span class="n">torch_dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float16</span><span class="p">,</span>
+                <span class="n">load_in_8bit</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                <span class="n">device_map</span><span class="o">=</span><span class="s2">&quot;auto&quot;</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">kwargs</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="n">past_model_dim</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="o">.</span><span class="n">model_dim</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span> <span class="o">=</span> <span class="n">AutoModelForCausalLM</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+                                <span class="n">pretrained_path</span><span class="p">,</span>
+                                <span class="n">config</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">text_config</span><span class="p">,</span>
+                                <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">text_config</span><span class="o">.</span><span class="n">hidden_size</span> <span class="o">!=</span> <span class="n">past_model_dim</span><span class="p">:</span>
+            <span class="c1"># should update the language projection layer</span>
+            <span class="n">in_channels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">language_projection</span><span class="o">.</span><span class="n">in_features</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">language_projection</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">in_channels</span><span class="p">,</span>
+                                                 <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">text_config</span><span class="o">.</span><span class="n">hidden_size</span><span class="p">,</span>
+                                                 <span class="n">bias</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.vision_feature_select">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.vision_feature_select">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">vision_feature_select</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">image_forward_outs</span><span class="p">):</span>
+        <span class="n">image_features</span> <span class="o">=</span> <span class="n">image_forward_outs</span><span class="o">.</span><span class="n">hidden_states</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">vision_feature_select_layer</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">select_vision_feature_type</span> <span class="o">==</span> <span class="s2">&quot;patch&quot;</span><span class="p">:</span>
+            <span class="n">image_features</span> <span class="o">=</span> <span class="n">image_features</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">:]</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">select_vision_feature_type</span> <span class="o">==</span> <span class="s2">&quot;cls_patch&quot;</span><span class="p">:</span>
+            <span class="n">image_features</span> <span class="o">=</span> <span class="n">image_features</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;Unexpected select feature: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">select_feature</span><span class="si">}</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">image_features</span></div>
+
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.register_prompt_cache">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.register_prompt_cache">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">register_prompt_cache</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">prompt_ids</span><span class="p">,</span> <span class="n">prompt_keys_values</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Udpate the prompt id and embedding for reuse in the future</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            prompt_ids (torch.LongTensor): The id of the prompt.</span>
+<span class="sd">            prompt_keys_values (torch.FloatTensor): The embedding of the prompt.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            None</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">prompt_ids</span> <span class="o">=</span> <span class="n">prompt_ids</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">prompt_keys_values</span> <span class="o">=</span> <span class="n">prompt_keys_values</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">with_prompt_cache</span> <span class="o">=</span> <span class="kc">True</span></div>
+
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.save_prompt_cache">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.save_prompt_cache">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">save_prompt_cache</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">path</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Save prompt embedding and id.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            path: The path to save the prompt embedding and id.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            None</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span>
+            <span class="nb">dict</span><span class="p">(</span>
+                <span class="n">prompt_ids</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">prompt_ids</span><span class="p">,</span>
+                <span class="n">prompt_keys_values</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">prompt_keys_values</span>
+            <span class="p">),</span>
+            <span class="n">path</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.load_prompt_cache">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.load_prompt_cache">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">load_prompt_cache</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">path</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Load prompt embedding and id.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            path: The path to load the prompt embedding and id.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            None</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">prompt_cache</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">path</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">register_prompt_cache</span><span class="p">(</span><span class="n">prompt_cache</span><span class="p">[</span><span class="s2">&quot;prompt_ids&quot;</span><span class="p">],</span>
+                                   <span class="n">prompt_cache</span><span class="p">[</span><span class="s2">&quot;prompt_keys_values&quot;</span><span class="p">])</span></div>
+
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.get_tokenizer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.get_tokenizer">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_tokenizer</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span></div>
+
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.forward">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.forward">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">input_ids</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">pixel_values</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">images</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">attention_mask</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">past_key_values</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">inputs_embeds</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">labels</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">use_cache</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">output_attentions</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">output_hidden_states</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">return_dict</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">image_token_indexes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">]</span> <span class="o">=</span> <span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+        <span class="n">one_sample_multiple_images</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">Tuple</span><span class="p">,</span> <span class="n">CausalLMOutputWithPast</span><span class="p">]:</span>
+        <span class="k">if</span> <span class="n">pixel_values</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">images</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">pixel_values</span> <span class="o">=</span> <span class="n">images</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">one_sample_multiple_images</span><span class="p">:</span>
+            <span class="n">batch_size</span> <span class="o">=</span> <span class="n">pixel_values</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">batch_size</span> <span class="o">=</span> <span class="mi">1</span>
+        <span class="n">output_attentions</span> <span class="o">=</span> <span class="n">output_attentions</span> <span class="k">if</span> <span class="n">output_attentions</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">output_attentions</span>
+        <span class="n">output_hidden_states</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="n">output_hidden_states</span> <span class="k">if</span> <span class="n">output_hidden_states</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">output_hidden_states</span>
+        <span class="p">)</span>
+        <span class="n">return_dict</span> <span class="o">=</span> <span class="n">return_dict</span> <span class="k">if</span> <span class="n">return_dict</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">use_return_dict</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">custom_vision_model</span><span class="p">:</span>
+            <span class="c1"># do the processing as blip2 and mini gpt-4;</span>
+            <span class="k">if</span> <span class="n">past_key_values</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">input_ids</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+                <span class="c1"># no need to recompute the key values</span>
+                <span class="n">attention_mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span>
+                    <span class="n">attention_mask</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+                    <span class="n">past_key_values</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">][</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span> <span class="o">+</span> <span class="mi">1</span><span class="p">),</span>
+                    <span class="n">dtype</span><span class="o">=</span><span class="n">attention_mask</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span>
+                    <span class="n">device</span><span class="o">=</span><span class="n">attention_mask</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">image_embeds</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">vision_model</span><span class="p">(</span>
+                    <span class="n">pixel_values</span><span class="p">,</span> <span class="n">return_dict</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span><span class="o">.</span><span class="n">last_hidden_state</span>
+                <span class="n">image_attention_mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span>
+                    <span class="n">image_embeds</span><span class="o">.</span><span class="n">size</span><span class="p">()[:</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span>
+                    <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">long</span><span class="p">,</span>
+                    <span class="n">device</span><span class="o">=</span><span class="n">image_embeds</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">with_qformer</span><span class="p">:</span>
+                    <span class="n">query_tokens</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">query_tokens</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span>
+                        <span class="n">image_embeds</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+                    <span class="n">query_outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">qformer</span><span class="p">(</span>
+                        <span class="n">query_embeds</span><span class="o">=</span><span class="n">query_tokens</span><span class="p">,</span>
+                        <span class="n">encoder_hidden_states</span><span class="o">=</span><span class="n">image_embeds</span><span class="p">,</span>
+                        <span class="n">encoder_attention_mask</span><span class="o">=</span><span class="n">image_attention_mask</span><span class="p">,</span>
+                        <span class="n">return_dict</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                    <span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">query_outputs</span> <span class="o">=</span> <span class="n">image_embeds</span>
+                <span class="n">query_output</span> <span class="o">=</span> <span class="n">query_outputs</span><span class="o">.</span><span class="n">last_hidden_state</span>
+                <span class="n">language_model_inputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">language_projection</span><span class="p">(</span><span class="n">query_output</span><span class="p">)</span>
+                <span class="n">inputs_embeds</span><span class="p">,</span> <span class="n">attention_mask</span> <span class="o">=</span> \
+                    <span class="bp">self</span><span class="o">.</span><span class="n">processor_image_token_in_minigpt4</span><span class="p">(</span>
+                        <span class="n">input_ids</span><span class="p">,</span>
+                        <span class="n">language_model_inputs</span><span class="p">,</span>
+                        <span class="n">attention_mask</span><span class="p">,</span>
+                        <span class="n">image_token_indexes</span><span class="p">,</span>
+                        <span class="n">pixel_values</span><span class="p">,</span>
+                        <span class="n">batch_size</span><span class="p">)</span>
+                <span class="n">input_ids</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># do the processing in the vision model</span>
+            <span class="c1"># language is the causallm model.</span>
+            <span class="c1"># so use language model.model to do the embed_tokens</span>
+            <span class="n">input_ids</span><span class="p">,</span> <span class="n">attention_mask</span><span class="p">,</span> <span class="n">past_key_values</span><span class="p">,</span> <span class="n">inputs_embeds</span><span class="p">,</span> <span class="n">labels</span> <span class="o">=</span> \
+                <span class="bp">self</span><span class="o">.</span><span class="n">vision_model</span><span class="o">.</span><span class="n">prepare_inputs_labels_for_multimodal</span><span class="p">(</span>
+                    <span class="n">input_ids</span><span class="p">,</span> <span class="n">attention_mask</span><span class="p">,</span>
+                    <span class="n">past_key_values</span><span class="p">,</span> <span class="n">labels</span><span class="p">,</span>
+                    <span class="n">pixel_values</span><span class="p">,</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">language_projection</span><span class="p">,</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="o">.</span><span class="n">model</span><span class="p">)</span>
+        <span class="c1"># decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)</span>
+        <span class="c1"># TODO check how to generate the labels with image embeddings</span>
+        <span class="c1"># print(input_ids, attention_mask)</span>
+        <span class="c1"># if inputs_embeds is not None:</span>
+        <span class="c1">#     print(&quot;input_embeds&quot;, inputs_embeds.shape)</span>
+        <span class="c1"># attention_mask.shape, inputs_embeds.shape)</span>
+        <span class="c1"># TODO remove this code by fixing the ddp training issue</span>
+        <span class="n">inputs_embeds</span> <span class="o">=</span> <span class="n">inputs_embeds</span><span class="o">.</span><span class="n">to</span><span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="o">.</span><span class="n">lm_head</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="p">(</span>
+            <span class="n">input_ids</span><span class="o">=</span><span class="n">input_ids</span><span class="p">,</span>
+            <span class="n">attention_mask</span><span class="o">=</span><span class="n">attention_mask</span><span class="p">,</span>
+            <span class="n">past_key_values</span><span class="o">=</span><span class="n">past_key_values</span><span class="p">,</span>
+            <span class="n">inputs_embeds</span><span class="o">=</span><span class="n">inputs_embeds</span><span class="p">,</span>
+            <span class="n">use_cache</span><span class="o">=</span><span class="n">use_cache</span><span class="p">,</span>
+            <span class="n">output_attentions</span><span class="o">=</span><span class="n">output_attentions</span><span class="p">,</span>
+            <span class="n">output_hidden_states</span><span class="o">=</span><span class="n">output_hidden_states</span><span class="p">,</span>
+            <span class="n">return_dict</span><span class="o">=</span><span class="n">return_dict</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">logits</span> <span class="o">=</span> <span class="n">outputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="n">shift_logits</span> <span class="o">=</span> <span class="n">logits</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="p">:</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="p">:]</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+            <span class="n">shift_labels</span> <span class="o">=</span> <span class="n">labels</span><span class="p">[</span><span class="o">...</span><span class="p">,</span> <span class="mi">1</span><span class="p">:]</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+            <span class="c1"># Flatten the tokens</span>
+            <span class="n">loss_fct</span> <span class="o">=</span> <span class="n">CrossEntropyLoss</span><span class="p">()</span>
+            <span class="n">shift_logits</span> <span class="o">=</span> <span class="n">shift_logits</span><span class="o">.</span><span class="n">view</span><span class="p">(</span>
+                <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">text_config</span><span class="o">.</span><span class="n">vocab_size</span><span class="p">)</span>
+            <span class="n">shift_labels</span> <span class="o">=</span> <span class="n">shift_labels</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+            <span class="c1"># Enable model/pipeline parallelism</span>
+            <span class="n">shift_labels</span> <span class="o">=</span> <span class="n">shift_labels</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">shift_logits</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">loss_fct</span><span class="p">(</span><span class="n">shift_logits</span><span class="p">,</span> <span class="n">shift_labels</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">return_dict</span><span class="p">:</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="p">(</span><span class="n">shift_logits</span><span class="p">,)</span> <span class="o">+</span> <span class="n">outputs</span><span class="p">[</span><span class="mi">1</span><span class="p">:]</span>
+            <span class="k">return</span> <span class="p">((</span><span class="n">loss</span><span class="p">,)</span> <span class="o">+</span> <span class="n">output</span><span class="p">)</span> <span class="k">if</span> <span class="n">loss</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">output</span>
+
+        <span class="k">return</span> <span class="n">CausalLMOutputWithPast</span><span class="p">(</span>
+            <span class="n">loss</span><span class="o">=</span><span class="n">loss</span><span class="p">,</span>
+            <span class="n">logits</span><span class="o">=</span><span class="n">logits</span><span class="p">,</span>
+            <span class="n">past_key_values</span><span class="o">=</span><span class="n">outputs</span><span class="o">.</span><span class="n">past_key_values</span><span class="p">,</span>
+            <span class="n">hidden_states</span><span class="o">=</span><span class="n">outputs</span><span class="o">.</span><span class="n">hidden_states</span><span class="p">,</span>
+            <span class="n">attentions</span><span class="o">=</span><span class="n">outputs</span><span class="o">.</span><span class="n">attentions</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.processor_image_token_in_minigpt4">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.processor_image_token_in_minigpt4">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">processor_image_token_in_minigpt4</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
+                                          <span class="n">input_ids</span><span class="p">,</span>
+                                          <span class="n">language_model_inputs</span><span class="p">,</span>
+                                          <span class="n">attention_mask</span><span class="p">,</span>
+                                          <span class="n">image_token_indexes</span><span class="p">,</span>
+                                          <span class="n">pixel_values</span><span class="p">,</span>
+                                          <span class="n">batch_size</span><span class="o">=</span><span class="mi">1</span><span class="p">):</span>
+        <span class="n">language_attention_mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span>
+            <span class="n">language_model_inputs</span><span class="o">.</span><span class="n">size</span><span class="p">()[:</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span>
+            <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">long</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">language_model_inputs</span><span class="o">.</span><span class="n">device</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="n">input_ids</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">input_ids</span> <span class="o">=</span> <span class="p">(</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">([[</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">text_config</span><span class="o">.</span><span class="n">bos_token_id</span><span class="p">]])</span>
+                <span class="o">.</span><span class="n">repeat</span><span class="p">(</span><span class="n">batch_size</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+                <span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">language_model_inputs</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">attention_mask</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">attention_mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones_like</span><span class="p">(</span><span class="n">input_ids</span><span class="p">)</span>
+            <span class="n">attention_mask</span> <span class="o">=</span> <span class="n">attention_mask</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">language_attention_mask</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+
+        <span class="c1"># concatenate query embeddings with prompt embeddings</span>
+        <span class="n">inputs_embeds</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_input_embeddings</span><span class="p">()(</span><span class="n">input_ids</span><span class="p">)</span>
+        <span class="n">inputs_embeds</span> <span class="o">=</span> <span class="n">inputs_embeds</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="n">language_model_inputs</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+        <span class="c1"># concatenate the text embeddings with image embeddings</span>
+        <span class="n">inputs_embeds_with_images</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">attention_mask_with_images</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="c1"># currently we only support with one image</span>
+        <span class="n">start_index</span><span class="p">,</span> <span class="n">end_index</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span>
+        <span class="k">assert</span> <span class="nb">len</span><span class="p">(</span><span class="n">image_token_indexes</span><span class="p">)</span> <span class="o">==</span> <span class="n">pixel_values</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="c1"># token format: (# text, # image)xN, # text</span>
+
+        <span class="k">for</span> <span class="n">idx</span><span class="p">,</span> <span class="n">image_token_index</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">image_token_indexes</span><span class="p">):</span>
+            <span class="n">end_index</span> <span class="o">+=</span> <span class="n">image_token_index</span>
+            <span class="n">inputs_embeds_with_images</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">inputs_embeds</span><span class="p">[:,</span> <span class="n">start_index</span><span class="p">:</span><span class="n">end_index</span><span class="p">])</span>
+            <span class="n">inputs_embeds_with_images</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">language_model_inputs</span><span class="p">[</span><span class="n">idx</span><span class="p">][</span><span class="kc">None</span><span class="p">])</span>
+            <span class="n">attention_mask_with_images</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">attention_mask</span><span class="p">[:,</span> <span class="n">start_index</span><span class="p">:</span><span class="n">end_index</span><span class="p">])</span>
+            <span class="n">attention_mask_with_images</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="n">language_attention_mask</span><span class="p">[</span><span class="n">idx</span><span class="p">][</span><span class="kc">None</span><span class="p">])</span>
+            <span class="n">start_index</span> <span class="o">=</span> <span class="n">end_index</span>
+
+        <span class="n">inputs_embeds_with_images</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">inputs_embeds</span><span class="p">[:,</span> <span class="n">image_token_indexes</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]:])</span>
+        <span class="n">inputs_embeds</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span>
+            <span class="n">inputs_embeds_with_images</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">attention_mask_with_images</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+            <span class="n">attention_mask</span><span class="p">[:,</span> <span class="n">image_token_indexes</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]:])</span>
+        <span class="n">attention_mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">attention_mask_with_images</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="c1"># comebine the embeds</span>
+        <span class="n">inputs_embeds</span> <span class="o">=</span> <span class="n">inputs_embeds</span><span class="o">.</span><span class="n">to</span><span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="o">.</span><span class="n">lm_head</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+        <span class="n">attention_mask</span> <span class="o">=</span> <span class="n">attention_mask</span><span class="o">.</span><span class="n">to</span><span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="o">.</span><span class="n">lm_head</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">inputs_embeds</span><span class="p">,</span> <span class="n">attention_mask</span></div>
+
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<div class="viewcode-block" id="CustomAutoVision2SeqModel.generate">
+<a class="viewcode-back" href="../../../autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.generate">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">generate</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">pixel_values</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">,</span>
+        <span class="n">input_ids</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">attention_mask</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">image_token_indexes</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">]</span> <span class="o">=</span> <span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+        <span class="n">one_sample_multiple_images</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">images</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">generate_kwargs</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Overrides `generate` function to be able to use the model as a conditional generator.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            pixel_values (`torch.FloatTensor` of shape (batch_size, num_channels, height, width)):</span>
+<span class="sd">                Input images to be processed.</span>
+<span class="sd">            input_ids (`torch.LongTensor` of shape (batch_size, sequence_length), *optional*):</span>
+<span class="sd">                The sequence used as a prompt for the generation.</span>
+<span class="sd">            attention_mask (`torch.LongTensor` of shape (batch_size, sequence_length), *optional*):</span>
+<span class="sd">                Mask to avoid performing attention on padding token indices</span>
+<span class="sd">            image_token_indexes (bool, *optional*):</span>
+<span class="sd">                The index for inserting the image tokens.</span>
+<span class="sd">            one_sample_multiple_images: (bool, *optional*):</span>
+<span class="sd">                The flag for inference that the input batch size is 1 and contain multiple images.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            captions (list): A list of strings of length batch_size * num_captions.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">pixel_values</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">images</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">pixel_values</span> <span class="o">=</span> <span class="n">images</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">one_sample_multiple_images</span><span class="p">:</span>
+            <span class="n">batch_size</span> <span class="o">=</span> <span class="n">pixel_values</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">batch_size</span> <span class="o">=</span> <span class="mi">1</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">custom_vision_model</span><span class="p">:</span>
+            <span class="c1"># do the processing as blip2 and mini gpt-4;</span>
+            <span class="n">image_embeds</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">vision_model</span><span class="p">(</span>
+                <span class="n">pixel_values</span><span class="p">,</span> <span class="n">return_dict</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span><span class="o">.</span><span class="n">last_hidden_state</span>
+            <span class="n">image_attention_mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span>
+                <span class="n">image_embeds</span><span class="o">.</span><span class="n">size</span><span class="p">()[:</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span>
+                <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">long</span><span class="p">,</span>
+                <span class="n">device</span><span class="o">=</span><span class="n">image_embeds</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">with_qformer</span><span class="p">:</span>
+                <span class="n">query_tokens</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">query_tokens</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span>
+                    <span class="n">image_embeds</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+                <span class="n">query_outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">qformer</span><span class="p">(</span>
+                    <span class="n">query_embeds</span><span class="o">=</span><span class="n">query_tokens</span><span class="p">,</span>
+                    <span class="n">encoder_hidden_states</span><span class="o">=</span><span class="n">image_embeds</span><span class="p">,</span>
+                    <span class="n">encoder_attention_mask</span><span class="o">=</span><span class="n">image_attention_mask</span><span class="p">,</span>
+                    <span class="n">return_dict</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">query_outputs</span> <span class="o">=</span> <span class="n">image_embeds</span>
+            <span class="n">query_output</span> <span class="o">=</span> <span class="n">query_outputs</span><span class="o">.</span><span class="n">last_hidden_state</span>
+            <span class="n">language_model_inputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">language_projection</span><span class="p">(</span><span class="n">query_output</span><span class="p">)</span>
+            <span class="n">inputs_embeds</span><span class="p">,</span> <span class="n">attention_mask</span> <span class="o">=</span> \
+                <span class="bp">self</span><span class="o">.</span><span class="n">processor_image_token_in_minigpt4</span><span class="p">(</span>
+                    <span class="n">input_ids</span><span class="p">,</span>
+                    <span class="n">language_model_inputs</span><span class="p">,</span>
+                    <span class="n">attention_mask</span><span class="p">,</span>
+                    <span class="n">image_token_indexes</span><span class="p">,</span>
+                    <span class="n">pixel_values</span><span class="p">,</span>
+                    <span class="n">batch_size</span><span class="p">)</span>
+            <span class="n">input_ids</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># do the processing in the vision model</span>
+            <span class="c1"># language is the causallm model.</span>
+            <span class="c1"># so use language model.model to do the embed_tokens</span>
+            <span class="k">if</span> <span class="n">pixel_values</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
+                <span class="c1"># the batch dim is missing;</span>
+                <span class="n">pixel_values</span> <span class="o">=</span> <span class="n">pixel_values</span><span class="p">[</span><span class="kc">None</span><span class="p">]</span>
+            <span class="n">input_ids</span><span class="p">,</span> <span class="n">attention_mask</span><span class="p">,</span> <span class="n">past_key_values</span><span class="p">,</span> <span class="n">inputs_embeds</span><span class="p">,</span> <span class="n">labels</span> <span class="o">=</span> \
+                <span class="bp">self</span><span class="o">.</span><span class="n">vision_model</span><span class="o">.</span><span class="n">prepare_inputs_labels_for_multimodal</span><span class="p">(</span>
+                    <span class="n">input_ids</span><span class="p">,</span> <span class="n">attention_mask</span><span class="p">,</span>
+                    <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span>
+                    <span class="n">pixel_values</span><span class="p">,</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">language_projection</span><span class="p">,</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="o">.</span><span class="n">model</span><span class="p">)</span>
+        <span class="c1"># convert the dtype.</span>
+        <span class="c1"># FIXME check when need to do this</span>
+        <span class="n">inputs_embeds</span> <span class="o">=</span> <span class="n">inputs_embeds</span><span class="o">.</span><span class="n">to</span><span class="p">(</span>
+            <span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="o">.</span><span class="n">lm_head</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+        <span class="n">inputs_embeds</span> <span class="o">=</span> <span class="n">inputs_embeds</span><span class="o">.</span><span class="n">to</span><span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="o">.</span><span class="n">lm_head</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">language_model</span><span class="o">.</span><span class="n">generate</span><span class="p">(</span>
+            <span class="n">inputs_embeds</span><span class="o">=</span><span class="n">inputs_embeds</span><span class="p">,</span>
+            <span class="n">attention_mask</span><span class="o">=</span><span class="n">attention_mask</span><span class="p">,</span>
+            <span class="o">**</span><span class="n">generate_kwargs</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">outputs</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/models/vision_encoder/clip_encoder.html b/_modules/lmflow/models/vision_encoder/clip_encoder.html
new file mode 100644
index 000000000..13ba142df
--- /dev/null
+++ b/_modules/lmflow/models/vision_encoder/clip_encoder.html
@@ -0,0 +1,719 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.models.vision_encoder.clip_encoder &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/models/vision_encoder/clip_encoder';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.vision_encoder.clip_encoder</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.models.vision_encoder.clip_encoder</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span><span class="w"> </span><span class="nn">webbrowser</span><span class="w"> </span><span class="kn">import</span> <span class="n">get</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.nn</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">nn</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">CLIPVisionModel</span><span class="p">,</span> <span class="n">CLIPImageProcessor</span><span class="p">,</span> <span class="n">CLIPVisionConfig</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span><span class="n">IGNORE_INDEX</span><span class="p">,</span>
+                                    <span class="n">IMAGE_TOKEN_INDEX</span><span class="p">,</span>
+                                    <span class="n">DEFAULT_IMAGE_PATCH_TOKEN</span><span class="p">,</span>
+                                    <span class="n">DEFAULT_IM_START_TOKEN</span><span class="p">,</span>
+                                    <span class="n">DEFAULT_IM_END_TOKEN</span><span class="p">)</span>
+
+<div class="viewcode-block" id="build_vision_tower">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.build_vision_tower">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">build_vision_tower</span><span class="p">(</span><span class="n">vision_tower_cfg</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+    <span class="n">vision_tower</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">vision_tower_cfg</span><span class="p">,</span> <span class="s1">&#39;image_encoder_name_or_path&#39;</span><span class="p">,</span> <span class="s2">&quot;openai/clip-vit-large-patch14&quot;</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">vision_tower</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;openai&quot;</span><span class="p">)</span> <span class="ow">or</span> <span class="n">vision_tower</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;laion&quot;</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">CLIPVisionTower</span><span class="p">(</span><span class="n">vision_tower</span><span class="p">,</span> <span class="n">args</span><span class="o">=</span><span class="n">vision_tower_cfg</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;Unknown vision tower: </span><span class="si">{</span><span class="n">vision_tower</span><span class="si">}</span><span class="s1">&#39;</span><span class="p">)</span></div>
+
+
+<span class="c1"># FIXME check if can directly use the BlipVisionEncoder</span>
+<div class="viewcode-block" id="CLIPVisionTower">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">CLIPVisionTower</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">vision_tower</span><span class="p">,</span> <span class="n">args</span><span class="p">,</span> <span class="n">delay_load</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+
+<div class="viewcode-block" id="CLIPVisionTower.is_loaded">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.is_loaded">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">is_loaded</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+
+<div class="viewcode-block" id="CLIPVisionTower.vision_tower_name">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.vision_tower_name">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">vision_tower_name</span> <span class="o">=</span> <span class="n">vision_tower</span></div>
+
+<div class="viewcode-block" id="CLIPVisionTower.select_layer">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.select_layer">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">select_layer</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">vision_select_layer</span></div>
+
+<div class="viewcode-block" id="CLIPVisionTower.select_feature">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.select_feature">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">select_feature</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">args</span><span class="p">,</span> <span class="s1">&#39;vision_select_feature&#39;</span><span class="p">,</span> <span class="s1">&#39;patch&#39;</span><span class="p">)</span></div>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">delay_load</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">load_model</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">cfg_only</span> <span class="o">=</span> <span class="n">CLIPVisionConfig</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">vision_tower_name</span><span class="p">)</span>
+
+<div class="viewcode-block" id="CLIPVisionTower.load_model">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.load_model">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">load_model</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">image_processor</span> <span class="o">=</span> <span class="n">CLIPImageProcessor</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">vision_tower_name</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">vision_tower</span> <span class="o">=</span> <span class="n">CLIPVisionModel</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">vision_tower_name</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">vision_tower</span><span class="o">.</span><span class="n">requires_grad_</span><span class="p">(</span><span class="kc">False</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">is_loaded</span> <span class="o">=</span> <span class="kc">True</span></div>
+
+
+<div class="viewcode-block" id="CLIPVisionTower.encode_images">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.encode_images">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">encode_images</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">images</span><span class="p">,</span> <span class="n">language_projection</span><span class="p">):</span>
+        <span class="n">image_features</span> <span class="o">=</span> <span class="bp">self</span><span class="p">(</span><span class="n">images</span><span class="p">)</span>
+        <span class="c1"># FIXME the language projection is register in the CustomAutoVision2SeqModel</span>
+        <span class="c1"># check how to move this code to improve the readability</span>
+        <span class="k">if</span> <span class="n">language_projection</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">image_features</span> <span class="o">=</span> <span class="n">language_projection</span><span class="p">(</span><span class="n">image_features</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">image_features</span></div>
+
+
+<div class="viewcode-block" id="CLIPVisionTower.feature_select">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.feature_select">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">feature_select</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">image_forward_outs</span><span class="p">):</span>
+        <span class="n">image_features</span> <span class="o">=</span> <span class="n">image_forward_outs</span><span class="o">.</span><span class="n">hidden_states</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">select_layer</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">select_feature</span> <span class="o">==</span> <span class="s1">&#39;patch&#39;</span><span class="p">:</span>
+            <span class="n">image_features</span> <span class="o">=</span> <span class="n">image_features</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">:]</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">select_feature</span> <span class="o">==</span> <span class="s1">&#39;cls_patch&#39;</span><span class="p">:</span>
+            <span class="n">image_features</span> <span class="o">=</span> <span class="n">image_features</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;Unexpected select feature: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">select_feature</span><span class="si">}</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">image_features</span></div>
+
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<div class="viewcode-block" id="CLIPVisionTower.forward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.forward">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">images</span><span class="p">):</span>
+        <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">images</span><span class="p">)</span> <span class="ow">is</span> <span class="nb">list</span><span class="p">:</span>
+            <span class="n">image_features</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="k">for</span> <span class="n">image</span> <span class="ow">in</span> <span class="n">images</span><span class="p">:</span>
+                <span class="n">image_forward_out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">vision_tower</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">output_hidden_states</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                <span class="n">image_feature</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">feature_select</span><span class="p">(</span><span class="n">image_forward_out</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+                <span class="n">image_features</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">image_feature</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">image_forward_outs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">vision_tower</span><span class="p">(</span><span class="n">images</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">),</span> <span class="n">output_hidden_states</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="n">image_features</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">feature_select</span><span class="p">(</span><span class="n">image_forward_outs</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">images</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">image_features</span></div>
+
+
+    <span class="nd">@property</span>
+<div class="viewcode-block" id="CLIPVisionTower.dummy_feature">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.dummy_feature">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">dummy_feature</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">hidden_size</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span></div>
+
+
+    <span class="nd">@property</span>
+<div class="viewcode-block" id="CLIPVisionTower.dtype">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.dtype">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">dtype</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">vision_tower</span><span class="o">.</span><span class="n">dtype</span></div>
+
+
+    <span class="nd">@property</span>
+<div class="viewcode-block" id="CLIPVisionTower.device">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.device">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">device</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">vision_tower</span><span class="o">.</span><span class="n">device</span></div>
+
+
+    <span class="nd">@property</span>
+<div class="viewcode-block" id="CLIPVisionTower.config">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.config">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">config</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_loaded</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">vision_tower</span><span class="o">.</span><span class="n">config</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">cfg_only</span></div>
+
+
+    <span class="nd">@property</span>
+<div class="viewcode-block" id="CLIPVisionTower.hidden_size">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.hidden_size">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">hidden_size</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">hidden_size</span></div>
+
+
+    <span class="nd">@property</span>
+<div class="viewcode-block" id="CLIPVisionTower.num_patches">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.num_patches">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">num_patches</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">image_size</span> <span class="o">//</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">patch_size</span><span class="p">)</span> <span class="o">**</span> <span class="mi">2</span></div>
+
+
+<div class="viewcode-block" id="CLIPVisionTower.prepare_inputs_labels_for_multimodal">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.prepare_inputs_labels_for_multimodal">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">prepare_inputs_labels_for_multimodal</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">input_ids</span><span class="p">,</span> <span class="n">attention_mask</span><span class="p">,</span> <span class="n">past_key_values</span><span class="p">,</span> <span class="n">labels</span><span class="p">,</span> <span class="n">images</span><span class="p">,</span>
+        <span class="n">language_projection</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">language_model</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&#39;&#39;&#39;</span>
+<span class="sd">        Copy from the LLAVA code base.</span>
+<span class="sd">        Should be polished.</span>
+<span class="sd">        &#39;&#39;&#39;</span>
+        <span class="n">vision_tower</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">vision_tower</span>
+        <span class="c1"># commonly used in model.generate (past_key_values is not None)</span>
+        <span class="c1"># to avoid forward the image multiple time</span>
+        <span class="k">if</span> <span class="n">vision_tower</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">images</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">input_ids</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">if</span> <span class="p">(</span><span class="n">past_key_values</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span>
+                <span class="n">vision_tower</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span>
+                <span class="n">images</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span>
+                <span class="n">input_ids</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">==</span> <span class="mi">1</span><span class="p">):</span>
+                <span class="n">attention_mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span>
+                    <span class="n">attention_mask</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span>
+                    <span class="n">past_key_values</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">][</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span> <span class="o">+</span> <span class="mi">1</span><span class="p">),</span>
+                    <span class="n">dtype</span><span class="o">=</span><span class="n">attention_mask</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">attention_mask</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">input_ids</span><span class="p">,</span> <span class="n">attention_mask</span><span class="p">,</span> <span class="n">past_key_values</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="n">labels</span>
+        <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">images</span><span class="p">)</span> <span class="ow">is</span> <span class="nb">list</span> <span class="ow">or</span> <span class="n">images</span><span class="o">.</span><span class="n">ndim</span> <span class="o">==</span> <span class="mi">5</span><span class="p">:</span>
+            <span class="n">concat_images</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">image</span> <span class="k">for</span> <span class="n">image</span> <span class="ow">in</span> <span class="n">images</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">image_features</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">encode_images</span><span class="p">(</span><span class="n">concat_images</span><span class="p">,</span> <span class="n">language_projection</span><span class="p">)</span>
+            <span class="n">split_sizes</span> <span class="o">=</span> <span class="p">[</span><span class="n">image</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="k">for</span> <span class="n">image</span> <span class="ow">in</span> <span class="n">images</span><span class="p">]</span>
+            <span class="n">image_features</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">image_features</span><span class="p">,</span> <span class="n">split_sizes</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">image_features</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">image_features</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">image_features</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">encode_images</span><span class="p">(</span><span class="n">images</span><span class="p">,</span> <span class="n">language_projection</span><span class="p">)</span>
+        <span class="n">new_input_embeds</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">new_labels</span> <span class="o">=</span> <span class="p">[]</span> <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="kc">None</span>
+        <span class="n">cur_image_idx</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">for</span> <span class="n">batch_idx</span><span class="p">,</span> <span class="n">cur_input_ids</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">input_ids</span><span class="p">):</span>
+            <span class="k">if</span> <span class="p">(</span><span class="n">cur_input_ids</span> <span class="o">==</span> <span class="n">IMAGE_TOKEN_INDEX</span><span class="p">)</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="c1"># multimodal LLM, but the current sample is not multimodal</span>
+                <span class="n">cur_input_embeds</span> <span class="o">=</span> <span class="n">language_model</span><span class="o">.</span><span class="n">embed_tokens</span><span class="p">(</span><span class="n">cur_input_ids</span><span class="p">)</span>
+                <span class="n">cur_input_embeds</span> <span class="o">=</span> <span class="n">cur_input_embeds</span> <span class="o">+</span> <span class="p">(</span><span class="mf">0.</span> <span class="o">*</span> <span class="n">language_projection</span><span class="p">(</span><span class="n">vision_tower</span><span class="o">.</span><span class="n">dummy_feature</span><span class="p">))</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span>
+                <span class="n">new_input_embeds</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cur_input_embeds</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">new_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">labels</span><span class="p">[</span><span class="n">batch_idx</span><span class="p">])</span>
+                <span class="n">cur_image_idx</span> <span class="o">+=</span> <span class="mi">1</span>
+                <span class="k">continue</span>
+            <span class="n">image_token_indices</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">cur_input_ids</span> <span class="o">==</span> <span class="n">IMAGE_TOKEN_INDEX</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="n">cur_new_input_embeds</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">cur_labels</span> <span class="o">=</span> <span class="n">labels</span><span class="p">[</span><span class="n">batch_idx</span><span class="p">]</span>
+                <span class="n">cur_new_labels</span> <span class="o">=</span> <span class="p">[]</span>
+                <span class="k">assert</span> <span class="n">cur_labels</span><span class="o">.</span><span class="n">shape</span> <span class="o">==</span> <span class="n">cur_input_ids</span><span class="o">.</span><span class="n">shape</span>
+            <span class="k">while</span> <span class="n">image_token_indices</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">cur_image_features</span> <span class="o">=</span> <span class="n">image_features</span><span class="p">[</span><span class="n">cur_image_idx</span><span class="p">]</span>
+                <span class="n">image_token_start</span> <span class="o">=</span> <span class="n">image_token_indices</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+                <span class="c1"># print(&quot;image token_start&quot;, image_token_start,</span>
+                    <span class="c1">#   &quot;curr_input_ids&quot;, cur_input_ids.shape)</span>
+                <span class="k">if</span> <span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="p">,</span> <span class="s1">&#39;tune_mm_mlp_adapter&#39;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="p">,</span> <span class="s1">&#39;mm_use_im_start_end&#39;</span><span class="p">,</span> <span class="kc">False</span><span class="p">):</span>
+                    <span class="n">cur_new_input_embeds</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">language_model</span><span class="o">.</span><span class="n">embed_tokens</span><span class="p">(</span><span class="n">cur_input_ids</span><span class="p">[:</span><span class="n">image_token_start</span><span class="o">-</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">detach</span><span class="p">())</span>
+                    <span class="n">cur_new_input_embeds</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">language_model</span><span class="o">.</span><span class="n">embed_tokens</span><span class="p">(</span><span class="n">cur_input_ids</span><span class="p">[</span><span class="n">image_token_start</span><span class="o">-</span><span class="mi">1</span><span class="p">:</span><span class="n">image_token_start</span><span class="p">]))</span>
+                    <span class="n">cur_new_input_embeds</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cur_image_features</span><span class="p">)</span>
+                    <span class="n">cur_new_input_embeds</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">language_model</span><span class="o">.</span><span class="n">embed_tokens</span><span class="p">(</span><span class="n">cur_input_ids</span><span class="p">[</span><span class="n">image_token_start</span><span class="o">+</span><span class="mi">1</span><span class="p">:</span><span class="n">image_token_start</span><span class="o">+</span><span class="mi">2</span><span class="p">]))</span>
+                    <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="n">cur_new_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cur_labels</span><span class="p">[:</span><span class="n">image_token_start</span><span class="p">])</span>
+                        <span class="n">cur_new_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">full</span><span class="p">((</span><span class="n">cur_image_features</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],),</span> <span class="n">IGNORE_INDEX</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">labels</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">labels</span><span class="o">.</span><span class="n">dtype</span><span class="p">))</span>
+                        <span class="n">cur_new_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cur_labels</span><span class="p">[</span><span class="n">image_token_start</span><span class="p">:</span><span class="n">image_token_start</span><span class="o">+</span><span class="mi">1</span><span class="p">])</span>
+                        <span class="n">cur_labels</span> <span class="o">=</span> <span class="n">cur_labels</span><span class="p">[</span><span class="n">image_token_start</span><span class="o">+</span><span class="mi">2</span><span class="p">:]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">cur_input_ids</span> <span class="o">=</span> <span class="n">cur_input_ids</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="n">language_model</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+                    <span class="n">cur_new_input_embeds</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">language_model</span><span class="o">.</span><span class="n">embed_tokens</span><span class="p">(</span><span class="n">cur_input_ids</span><span class="p">[:</span><span class="n">image_token_start</span><span class="p">]))</span>
+                    <span class="n">cur_new_input_embeds</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cur_image_features</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="n">cur_new_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cur_labels</span><span class="p">[:</span><span class="n">image_token_start</span><span class="p">])</span>
+                        <span class="n">cur_new_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">full</span><span class="p">((</span><span class="n">cur_image_features</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],),</span> <span class="n">IGNORE_INDEX</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">labels</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">labels</span><span class="o">.</span><span class="n">dtype</span><span class="p">))</span>
+                        <span class="n">cur_labels</span> <span class="o">=</span> <span class="n">cur_labels</span><span class="p">[</span><span class="n">image_token_start</span><span class="o">+</span><span class="mi">1</span><span class="p">:]</span>
+                <span class="n">cur_image_idx</span> <span class="o">+=</span> <span class="mi">1</span>
+                <span class="k">if</span> <span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="p">,</span> <span class="s1">&#39;tune_mm_mlp_adapter&#39;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="p">,</span> <span class="s1">&#39;mm_use_im_start_end&#39;</span><span class="p">,</span> <span class="kc">False</span><span class="p">):</span>
+                    <span class="n">cur_input_ids</span> <span class="o">=</span> <span class="n">cur_input_ids</span><span class="p">[</span><span class="n">image_token_start</span><span class="o">+</span><span class="mi">2</span><span class="p">:]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">cur_input_ids</span> <span class="o">=</span> <span class="n">cur_input_ids</span><span class="p">[</span><span class="n">image_token_start</span><span class="o">+</span><span class="mi">1</span><span class="p">:]</span>
+                <span class="n">image_token_indices</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">cur_input_ids</span> <span class="o">==</span> <span class="n">IMAGE_TOKEN_INDEX</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="k">if</span> <span class="n">cur_input_ids</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">if</span> <span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="p">,</span> <span class="s1">&#39;tune_mm_mlp_adapter&#39;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="p">,</span> <span class="s1">&#39;mm_use_im_start_end&#39;</span><span class="p">,</span> <span class="kc">False</span><span class="p">):</span>
+                    <span class="n">cur_new_input_embeds</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">language_model</span><span class="o">.</span><span class="n">embed_tokens</span><span class="p">(</span><span class="n">cur_input_ids</span><span class="p">)</span><span class="o">.</span><span class="n">detach</span><span class="p">())</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">cur_new_input_embeds</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">language_model</span><span class="o">.</span><span class="n">embed_tokens</span><span class="p">(</span><span class="n">cur_input_ids</span><span class="p">))</span>
+                <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">cur_new_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cur_labels</span><span class="p">)</span>
+            <span class="n">cur_new_input_embeds</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">device</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">cur_new_input_embeds</span><span class="p">]</span>
+            <span class="n">cur_new_input_embeds</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">cur_new_input_embeds</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">new_input_embeds</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cur_new_input_embeds</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">cur_new_labels</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">cur_new_labels</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+                <span class="n">new_labels</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cur_new_labels</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">any</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">shape</span> <span class="o">!=</span> <span class="n">new_input_embeds</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">new_input_embeds</span><span class="p">):</span>
+            <span class="n">max_len</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">new_input_embeds</span><span class="p">)</span>
+
+            <span class="n">new_input_embeds_align</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="k">for</span> <span class="n">cur_new_embed</span> <span class="ow">in</span> <span class="n">new_input_embeds</span><span class="p">:</span>
+                <span class="n">cur_new_embed</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">cur_new_embed</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="n">max_len</span> <span class="o">-</span> <span class="n">cur_new_embed</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">cur_new_embed</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">cur_new_embed</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">cur_new_embed</span><span class="o">.</span><span class="n">device</span><span class="p">)),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+                <span class="n">new_input_embeds_align</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cur_new_embed</span><span class="p">)</span>
+            <span class="n">new_input_embeds</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">new_input_embeds_align</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">new_labels_align</span> <span class="o">=</span> <span class="p">[]</span>
+                <span class="n">_new_labels</span> <span class="o">=</span> <span class="n">new_labels</span>
+                <span class="k">for</span> <span class="n">cur_new_label</span> <span class="ow">in</span> <span class="n">new_labels</span><span class="p">:</span>
+                    <span class="n">cur_new_label</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">cur_new_label</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">full</span><span class="p">((</span><span class="n">max_len</span> <span class="o">-</span> <span class="n">cur_new_label</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],),</span> <span class="n">IGNORE_INDEX</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">cur_new_label</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">cur_new_label</span><span class="o">.</span><span class="n">device</span><span class="p">)),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+                    <span class="n">new_labels_align</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cur_new_label</span><span class="p">)</span>
+                <span class="n">new_labels</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">new_labels_align</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">attention_mask</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">new_attention_mask</span> <span class="o">=</span> <span class="p">[]</span>
+                <span class="k">for</span> <span class="n">cur_attention_mask</span><span class="p">,</span> <span class="n">cur_new_labels</span><span class="p">,</span> <span class="n">cur_new_labels_align</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">attention_mask</span><span class="p">,</span> <span class="n">_new_labels</span><span class="p">,</span> <span class="n">new_labels</span><span class="p">):</span>
+                    <span class="n">new_attn_mask_pad_left</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">full</span><span class="p">((</span><span class="n">cur_new_labels</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="n">labels</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">],),</span> <span class="kc">True</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">attention_mask</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">attention_mask</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+                    <span class="n">new_attn_mask_pad_right</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">full</span><span class="p">((</span><span class="n">cur_new_labels_align</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">-</span> <span class="n">cur_new_labels</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],),</span> <span class="kc">False</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">attention_mask</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">attention_mask</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+                    <span class="n">cur_new_attention_mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">new_attn_mask_pad_left</span><span class="p">,</span> <span class="n">cur_attention_mask</span><span class="p">,</span> <span class="n">new_attn_mask_pad_right</span><span class="p">),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+                    <span class="n">new_attention_mask</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">cur_new_attention_mask</span><span class="p">)</span>
+                <span class="n">attention_mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">new_attention_mask</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+                <span class="k">assert</span> <span class="n">attention_mask</span><span class="o">.</span><span class="n">shape</span> <span class="o">==</span> <span class="n">new_labels</span><span class="o">.</span><span class="n">shape</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">new_input_embeds</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">new_input_embeds</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">new_labels</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">new_labels</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">attention_mask</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">new_attn_mask_pad_left</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">full</span><span class="p">((</span><span class="n">attention_mask</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">new_input_embeds</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">-</span> <span class="n">input_ids</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]),</span> <span class="kc">True</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">attention_mask</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">attention_mask</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+                <span class="n">attention_mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">new_attn_mask_pad_left</span><span class="p">,</span> <span class="n">attention_mask</span><span class="p">),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+                <span class="k">assert</span> <span class="n">attention_mask</span><span class="o">.</span><span class="n">shape</span> <span class="o">==</span> <span class="n">new_input_embeds</span><span class="o">.</span><span class="n">shape</span><span class="p">[:</span><span class="mi">2</span><span class="p">]</span>
+        <span class="k">return</span> <span class="kc">None</span><span class="p">,</span> <span class="n">attention_mask</span><span class="p">,</span> <span class="n">past_key_values</span><span class="p">,</span> \
+                <span class="n">new_input_embeds</span><span class="p">,</span> <span class="n">new_labels</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/adabelief.html b/_modules/lmflow/optim/adabelief.html
new file mode 100644
index 000000000..3ad18b5e4
--- /dev/null
+++ b/_modules/lmflow/optim/adabelief.html
@@ -0,0 +1,649 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.adabelief &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/adabelief';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adabelief</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.adabelief</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+<div class="viewcode-block" id="AdaBelief">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">AdaBelief</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Implements AdaBelief algorithm. Modified from Adam in PyTorch</span>
+<span class="sd">    reference: AdaBelief Optimizer, adapting stepsizes by the belief in observed gradients, NeurIPS 2020</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">params</span><span class="p">,</span> <span class="n">lr</span><span class="o">=</span><span class="mf">1e-3</span><span class="p">,</span> <span class="n">betas</span><span class="o">=</span><span class="p">(</span><span class="mf">0.9</span><span class="p">,</span> <span class="mf">0.999</span><span class="p">),</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-16</span><span class="p">,</span>
+                 <span class="n">weight_decay</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">amsgrad</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">weight_decouple</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">fixed_decay</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">rectify</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                 <span class="n">degenerated_to_sgd</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">print_change_log</span> <span class="o">=</span> <span class="kc">True</span><span class="p">):</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">lr</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">eps</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid epsilon value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">eps</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid beta parameter at index 0: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid beta parameter at index 1: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+
+<div class="viewcode-block" id="AdaBelief.degenerated_to_sgd">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.degenerated_to_sgd">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">degenerated_to_sgd</span> <span class="o">=</span> <span class="n">degenerated_to_sgd</span></div>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">))</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">params</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">params</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="nb">dict</span><span class="p">):</span>
+            <span class="k">for</span> <span class="n">param</span> <span class="ow">in</span> <span class="n">params</span><span class="p">:</span>
+                <span class="k">if</span> <span class="s1">&#39;betas&#39;</span> <span class="ow">in</span> <span class="n">param</span> <span class="ow">and</span> <span class="p">(</span><span class="n">param</span><span class="p">[</span><span class="s1">&#39;betas&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span> <span class="o">!=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="ow">or</span> <span class="n">param</span><span class="p">[</span><span class="s1">&#39;betas&#39;</span><span class="p">][</span><span class="mi">1</span><span class="p">]</span> <span class="o">!=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]):</span>
+                    <span class="n">param</span><span class="p">[</span><span class="s1">&#39;buffer&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="p">[[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">10</span><span class="p">)]</span>
+
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span> <span class="n">betas</span><span class="o">=</span><span class="n">betas</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span>
+                        <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">,</span> <span class="n">amsgrad</span><span class="o">=</span><span class="n">amsgrad</span><span class="p">,</span> <span class="n">buffer</span><span class="o">=</span><span class="p">[[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">10</span><span class="p">)])</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">AdaBelief</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">degenerated_to_sgd</span> <span class="o">=</span> <span class="n">degenerated_to_sgd</span>
+<div class="viewcode-block" id="AdaBelief.weight_decouple">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.weight_decouple">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">weight_decouple</span> <span class="o">=</span> <span class="n">weight_decouple</span></div>
+
+<div class="viewcode-block" id="AdaBelief.rectify">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.rectify">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">rectify</span> <span class="o">=</span> <span class="n">rectify</span></div>
+
+<div class="viewcode-block" id="AdaBelief.fixed_decay">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.fixed_decay">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fixed_decay</span> <span class="o">=</span> <span class="n">fixed_decay</span></div>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">weight_decouple</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s1">&#39;Weight decoupling enabled in AdaBelief&#39;</span><span class="p">)</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">fixed_decay</span><span class="p">:</span>
+                <span class="nb">print</span><span class="p">(</span><span class="s1">&#39;Weight decay fixed&#39;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">rectify</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s1">&#39;Rectification enabled in AdaBelief&#39;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">amsgrad</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s1">&#39;AMSGrad enabled in AdaBelief&#39;</span><span class="p">)</span>
+
+<div class="viewcode-block" id="AdaBelief.__setstate__">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.__setstate__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__setstate__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">state</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">AdaBelief</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="n">__setstate__</span><span class="p">(</span><span class="n">state</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">group</span><span class="o">.</span><span class="n">setdefault</span><span class="p">(</span><span class="s1">&#39;amsgrad&#39;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="AdaBelief.reset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.reset">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">reset</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+                <span class="n">amsgrad</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;amsgrad&#39;</span><span class="p">]</span>
+
+                <span class="c1"># State initialization</span>
+                <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+                <span class="c1"># Exponential moving average of gradient values</span>
+                <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+                <span class="c1"># Exponential moving average of squared gradient values</span>
+                <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_var&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+                <span class="k">if</span> <span class="n">amsgrad</span><span class="p">:</span>
+                    <span class="c1"># Maintains max of all exp. moving avg. of sq. grad. values</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;max_exp_avg_var&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="AdaBelief.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Performs a single optimization step.</span>
+<span class="sd">        Arguments:</span>
+<span class="sd">            closure (callable, optional): A closure that reevaluates the model</span>
+<span class="sd">                and returns the loss.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                
+                <span class="c1"># cast data type</span>
+                <span class="n">half_precision</span> <span class="o">=</span> <span class="kc">False</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">float16</span><span class="p">:</span>
+                    <span class="n">half_precision</span> <span class="o">=</span> <span class="kc">True</span>
+                    <span class="n">p</span><span class="o">.</span><span class="n">data</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">float</span><span class="p">()</span>
+                    <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">float</span><span class="p">()</span>
+
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+                <span class="k">if</span> <span class="n">grad</span><span class="o">.</span><span class="n">is_sparse</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                        <span class="s1">&#39;AdaBelief does not support sparse gradients, please consider SparseAdam instead&#39;</span><span class="p">)</span>
+                <span class="n">amsgrad</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;amsgrad&#39;</span><span class="p">]</span>
+
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;betas&#39;</span><span class="p">]</span>
+
+                <span class="c1"># State initialization</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="c1"># Exponential moving average of gradient values</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+                    <span class="c1"># Exponential moving average of squared gradient values</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_var&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">amsgrad</span><span class="p">:</span>
+                        <span class="c1"># Maintains max of all exp. moving avg. of sq. grad. values</span>
+                        <span class="n">state</span><span class="p">[</span><span class="s1">&#39;max_exp_avg_var&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+                
+                <span class="c1"># perform weight decay, check if decoupled weight decay</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">weight_decouple</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">fixed_decay</span><span class="p">:</span>
+                        <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">]</span> <span class="o">*</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">])</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">])</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">]</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="n">grad</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">])</span>
+
+                <span class="c1"># get current state variable</span>
+                <span class="n">exp_avg</span><span class="p">,</span> <span class="n">exp_avg_var</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">],</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_var&#39;</span><span class="p">]</span>
+
+                <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
+                <span class="n">bias_correction1</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span>
+                <span class="n">bias_correction2</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span>
+
+                <span class="c1"># Update first and second moment running average</span>
+                <span class="n">exp_avg</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">)</span>
+                <span class="n">grad_residual</span> <span class="o">=</span> <span class="n">grad</span> <span class="o">-</span> <span class="n">exp_avg</span>
+                <span class="n">exp_avg_var</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta2</span><span class="p">)</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span> <span class="n">grad_residual</span><span class="p">,</span> <span class="n">grad_residual</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">)</span>
+
+                <span class="k">if</span> <span class="n">amsgrad</span><span class="p">:</span>
+                    <span class="n">max_exp_avg_var</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;max_exp_avg_var&#39;</span><span class="p">]</span>
+                    <span class="c1"># Maintains the maximum of all 2nd moment running avg. till now</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">max_exp_avg_var</span><span class="p">,</span> <span class="n">exp_avg_var</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">]),</span> <span class="n">out</span><span class="o">=</span><span class="n">max_exp_avg_var</span><span class="p">)</span>
+
+                    <span class="c1"># Use the max. for normalizing running avg. of gradient</span>
+                    <span class="n">denom</span> <span class="o">=</span> <span class="p">(</span><span class="n">max_exp_avg_var</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">bias_correction2</span><span class="p">))</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">])</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">denom</span> <span class="o">=</span> <span class="p">(</span><span class="n">exp_avg_var</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">])</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">bias_correction2</span><span class="p">))</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">])</span>
+                
+                <span class="c1"># update</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">rectify</span><span class="p">:</span>
+                    <span class="c1"># Default update</span>
+                    <span class="n">step_size</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">]</span> <span class="o">/</span> <span class="n">bias_correction1</span>
+                    <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">addcdiv_</span><span class="p">(</span> <span class="n">exp_avg</span><span class="p">,</span> <span class="n">denom</span><span class="p">,</span> <span class="n">value</span><span class="o">=-</span><span class="n">step_size</span><span class="p">)</span>
+
+                <span class="k">else</span><span class="p">:</span>  <span class="c1"># Rectified update, forked from RAdam</span>
+                    <span class="n">buffered</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;buffer&#39;</span><span class="p">][</span><span class="nb">int</span><span class="p">(</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">%</span> <span class="mi">10</span><span class="p">)]</span>
+                    <span class="k">if</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="n">buffered</span><span class="p">[</span><span class="mi">0</span><span class="p">]:</span>
+                        <span class="n">N_sma</span><span class="p">,</span> <span class="n">step_size</span> <span class="o">=</span> <span class="n">buffered</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">buffered</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">buffered</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span>
+                        <span class="n">beta2_t</span> <span class="o">=</span> <span class="n">beta2</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span>
+                        <span class="n">N_sma_max</span> <span class="o">=</span> <span class="mi">2</span> <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span>
+                        <span class="n">N_sma</span> <span class="o">=</span> <span class="n">N_sma_max</span> <span class="o">-</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">*</span> <span class="n">beta2_t</span> <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2_t</span><span class="p">)</span>
+                        <span class="n">buffered</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="n">N_sma</span>
+
+                        <span class="c1"># more conservative since it&#39;s an approximated value</span>
+                        <span class="k">if</span> <span class="n">N_sma</span> <span class="o">&gt;=</span> <span class="mi">5</span><span class="p">:</span>
+                            <span class="n">step_size</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span>
+                                <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2_t</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="n">N_sma</span> <span class="o">-</span> <span class="mi">4</span><span class="p">)</span> <span class="o">/</span> <span class="p">(</span><span class="n">N_sma_max</span> <span class="o">-</span> <span class="mi">4</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="n">N_sma</span> <span class="o">-</span> <span class="mi">2</span><span class="p">)</span> <span class="o">/</span> <span class="n">N_sma</span> <span class="o">*</span> <span class="n">N_sma_max</span> <span class="o">/</span> <span class="p">(</span>
+                                        <span class="n">N_sma_max</span> <span class="o">-</span> <span class="mi">2</span><span class="p">))</span> <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">])</span>
+                        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">degenerated_to_sgd</span><span class="p">:</span>
+                            <span class="n">step_size</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">])</span>
+                        <span class="k">else</span><span class="p">:</span>
+                            <span class="n">step_size</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span>
+                        <span class="n">buffered</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">=</span> <span class="n">step_size</span>
+
+                    <span class="k">if</span> <span class="n">N_sma</span> <span class="o">&gt;=</span> <span class="mi">5</span><span class="p">:</span>
+                        <span class="n">denom</span> <span class="o">=</span> <span class="n">exp_avg_var</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">])</span>
+                        <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">addcdiv_</span><span class="p">(</span><span class="n">exp_avg</span><span class="p">,</span> <span class="n">denom</span><span class="p">,</span> <span class="n">value</span><span class="o">=-</span><span class="n">step_size</span> <span class="o">*</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">])</span>
+                    <span class="k">elif</span> <span class="n">step_size</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span> <span class="n">exp_avg</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=-</span><span class="n">step_size</span> <span class="o">*</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">])</span>
+                
+                <span class="k">if</span> <span class="n">half_precision</span><span class="p">:</span>
+                    <span class="n">p</span><span class="o">.</span><span class="n">data</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">half</span><span class="p">()</span>
+                    <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">half</span><span class="p">()</span> 
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/adabound.html b/_modules/lmflow/optim/adabound.html
new file mode 100644
index 000000000..f2b15683c
--- /dev/null
+++ b/_modules/lmflow/optim/adabound.html
@@ -0,0 +1,611 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.adabound &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/adabound';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adabound</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.adabound</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+<div class="viewcode-block" id="AdaBound">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adabound/index.html#lmflow.optim.adabound.AdaBound">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">AdaBound</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Implements AdaBound algorithm.</span>
+<span class="sd">    </span>
+<span class="sd">    It has been proposed in `Adaptive Gradient Methods with Dynamic Bound of</span>
+<span class="sd">    Learning Rate</span>
+<span class="sd">    https://arxiv.org/abs/1902.09843</span>
+<span class="sd">    Note:</span>
+<span class="sd">        Reference code: https://github.com/Luolc/AdaBound</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">params</span><span class="p">,</span>
+        <span class="n">lr</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-3</span><span class="p">,</span>
+        <span class="n">betas</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.9</span><span class="p">,</span> <span class="mf">0.999</span><span class="p">),</span>
+        <span class="n">final_lr</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.1</span><span class="p">,</span>
+        <span class="n">gamma</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-3</span><span class="p">,</span>
+        <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-8</span><span class="p">,</span>
+        <span class="n">weight_decay</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">amsbound</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">lr</span> <span class="o">&lt;=</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">eps</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid epsilon value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">eps</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid beta parameter at index 0: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid beta parameter at index 1: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">final_lr</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid final learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">final_lr</span><span class="p">)</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">gamma</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid gamma parameter: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">gamma</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">weight_decay</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid weight_decay value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">weight_decay</span><span class="p">)</span>
+            <span class="p">)</span>
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span>
+            <span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span>
+            <span class="n">betas</span><span class="o">=</span><span class="n">betas</span><span class="p">,</span>
+            <span class="n">final_lr</span><span class="o">=</span><span class="n">final_lr</span><span class="p">,</span>
+            <span class="n">gamma</span><span class="o">=</span><span class="n">gamma</span><span class="p">,</span>
+            <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span>
+            <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">,</span>
+            <span class="n">amsbound</span><span class="o">=</span><span class="n">amsbound</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">AdaBound</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+<div class="viewcode-block" id="AdaBound.base_lrs">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adabound/index.html#lmflow.optim.adabound.AdaBound.base_lrs">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">base_lrs</span> <span class="o">=</span> <span class="p">[</span><span class="n">group</span><span class="p">[</span><span class="s2">&quot;lr&quot;</span><span class="p">]</span> <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">]</span></div>
+
+
+<div class="viewcode-block" id="AdaBound.__setstate__">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adabound/index.html#lmflow.optim.adabound.AdaBound.__setstate__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__setstate__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">state</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">AdaBound</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="n">__setstate__</span><span class="p">(</span><span class="n">state</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">group</span><span class="o">.</span><span class="n">setdefault</span><span class="p">(</span><span class="s2">&quot;amsbound&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="AdaBound.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adabound/index.html#lmflow.optim.adabound.AdaBound.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Performs a single optimization step.</span>
+
+<span class="sd">        Arguments:</span>
+<span class="sd">            closure: A closure that reevaluates the model and returns the loss.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span><span class="p">,</span> <span class="n">base_lr</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">base_lrs</span><span class="p">):</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;params&quot;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+                <span class="k">if</span> <span class="n">grad</span><span class="o">.</span><span class="n">is_sparse</span><span class="p">:</span>
+                    <span class="n">msg</span> <span class="o">=</span> <span class="p">(</span>
+                        <span class="s2">&quot;AdaBound does not support sparse gradients, &quot;</span>
+                        <span class="s2">&quot;please consider SparseAdam instead&quot;</span>
+                    <span class="p">)</span>
+                    <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="n">msg</span><span class="p">)</span>
+                <span class="n">amsbound</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;amsbound&quot;</span><span class="p">]</span>
+
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="c1"># State initialization</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="c1"># Exponential moving average of gradient values</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span>
+                        <span class="n">p</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span>
+                    <span class="p">)</span>
+                    <span class="c1"># Exponential moving average of squared gradient values</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg_sq&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span>
+                        <span class="n">p</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span>
+                    <span class="p">)</span>
+                    <span class="k">if</span> <span class="n">amsbound</span><span class="p">:</span>
+                        <span class="c1"># Maintains max of all exp. moving avg. of</span>
+                        <span class="c1"># sq. grad. values</span>
+                        <span class="n">state</span><span class="p">[</span><span class="s2">&quot;max_exp_avg_sq&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span>
+                            <span class="n">p</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span>
+                        <span class="p">)</span>
+
+                <span class="n">exp_avg</span><span class="p">,</span> <span class="n">exp_avg_sq</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">],</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg_sq&quot;</span><span class="p">]</span>
+                <span class="k">if</span> <span class="n">amsbound</span><span class="p">:</span>
+                    <span class="n">max_exp_avg_sq</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;max_exp_avg_sq&quot;</span><span class="p">]</span>
+                <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;betas&quot;</span><span class="p">]</span>
+
+                <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
+
+                <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">]</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">grad</span> <span class="o">=</span> <span class="n">grad</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">])</span>
+
+                <span class="c1"># Decay the first and second moment running average coefficient</span>
+                <span class="n">exp_avg</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">)</span>
+                <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta2</span><span class="p">)</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">amsbound</span><span class="p">:</span>
+                    <span class="c1"># Maintains the maximum of all 2nd moment running</span>
+                    <span class="c1"># avg. till now</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">max_exp_avg_sq</span><span class="p">,</span> <span class="n">exp_avg_sq</span><span class="p">,</span> <span class="n">out</span><span class="o">=</span><span class="n">max_exp_avg_sq</span><span class="p">)</span>
+                    <span class="c1"># Use the max. for normalizing running avg. of gradient</span>
+                    <span class="n">denom</span> <span class="o">=</span> <span class="n">max_exp_avg_sq</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s2">&quot;eps&quot;</span><span class="p">])</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">denom</span> <span class="o">=</span> <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s2">&quot;eps&quot;</span><span class="p">])</span>
+
+                <span class="n">bias_correction1</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span>
+                <span class="n">bias_correction2</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span>
+                <span class="n">step_size</span> <span class="o">=</span> <span class="p">(</span>
+                    <span class="n">group</span><span class="p">[</span><span class="s2">&quot;lr&quot;</span><span class="p">]</span>
+                    <span class="o">*</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">bias_correction2</span><span class="p">)</span>
+                    <span class="o">/</span> <span class="n">bias_correction1</span>
+                <span class="p">)</span>
+
+                <span class="c1"># Applies bounds on actual learning rate</span>
+                <span class="c1"># lr_scheduler cannot affect final_lr, this is a workaround</span>
+                <span class="c1"># to apply lr decay</span>
+                <span class="n">final_lr</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;final_lr&quot;</span><span class="p">]</span> <span class="o">*</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;lr&quot;</span><span class="p">]</span> <span class="o">/</span> <span class="n">base_lr</span>
+                <span class="n">lower_bound</span> <span class="o">=</span> <span class="n">final_lr</span> <span class="o">*</span> <span class="p">(</span>
+                    <span class="mi">1</span> <span class="o">-</span> <span class="mi">1</span> <span class="o">/</span> <span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s2">&quot;gamma&quot;</span><span class="p">]</span> <span class="o">*</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span>
+                <span class="p">)</span>
+                <span class="n">upper_bound</span> <span class="o">=</span> <span class="n">final_lr</span> <span class="o">*</span> <span class="p">(</span>
+                    <span class="mi">1</span> <span class="o">+</span> <span class="mi">1</span> <span class="o">/</span> <span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s2">&quot;gamma&quot;</span><span class="p">]</span> <span class="o">*</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">])</span>
+                <span class="p">)</span>
+                <span class="n">step_size</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">full_like</span><span class="p">(</span><span class="n">denom</span><span class="p">,</span> <span class="n">step_size</span><span class="p">)</span>
+                <span class="n">step_size</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="n">denom</span><span class="p">)</span><span class="o">.</span><span class="n">clamp_</span><span class="p">(</span><span class="n">lower_bound</span><span class="p">,</span> <span class="n">upper_bound</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span>
+                    <span class="n">exp_avg</span>
+                <span class="p">)</span>
+
+                <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="o">-</span><span class="n">step_size</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/adadelta.html b/_modules/lmflow/optim/adadelta.html
new file mode 100644
index 000000000..336afff9c
--- /dev/null
+++ b/_modules/lmflow/optim/adadelta.html
@@ -0,0 +1,490 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.adadelta &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/adadelta';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adadelta</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.adadelta</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+<div class="viewcode-block" id="Adadelta">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adadelta/index.html#lmflow.optim.adadelta.Adadelta">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Adadelta</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">params</span><span class="p">,</span> <span class="n">lr</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span> <span class="n">rho</span><span class="o">=</span><span class="mf">0.95</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-6</span><span class="p">):</span>
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span> <span class="n">rho</span><span class="o">=</span><span class="n">rho</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">Adadelta</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+<div class="viewcode-block" id="Adadelta.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adadelta/index.html#lmflow.optim.adadelta.Adadelta.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;square_avg&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;acc_delta&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+                <span class="n">square_avg</span><span class="p">,</span> <span class="n">acc_delta</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;square_avg&#39;</span><span class="p">],</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;acc_delta&#39;</span><span class="p">]</span>
+                <span class="n">rho</span><span class="p">,</span> <span class="n">eps</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;rho&#39;</span><span class="p">],</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">]</span>
+
+                <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
+
+                <span class="n">square_avg</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">rho</span><span class="p">)</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">rho</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">grad</span><span class="p">)</span>
+
+                <span class="n">std</span> <span class="o">=</span> <span class="n">square_avg</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">eps</span><span class="p">)</span><span class="o">.</span><span class="n">sqrt_</span><span class="p">()</span>
+                <span class="n">delta</span> <span class="o">=</span> <span class="n">acc_delta</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">eps</span><span class="p">)</span><span class="o">.</span><span class="n">sqrt_</span><span class="p">()</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="n">std</span><span class="p">)</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">grad</span><span class="p">)</span>
+
+                <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="o">-</span><span class="n">delta</span><span class="p">)</span>
+
+                <span class="n">acc_delta</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">rho</span><span class="p">)</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">rho</span><span class="p">,</span> <span class="n">delta</span><span class="p">,</span> <span class="n">delta</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/adagrad.html b/_modules/lmflow/optim/adagrad.html
new file mode 100644
index 000000000..6d3e8ced5
--- /dev/null
+++ b/_modules/lmflow/optim/adagrad.html
@@ -0,0 +1,482 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.adagrad &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/adagrad';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adagrad</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.adagrad</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+<div class="viewcode-block" id="AdaGrad">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adagrad/index.html#lmflow.optim.adagrad.AdaGrad">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">AdaGrad</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">Optimizer</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">params</span><span class="p">,</span> <span class="n">lr</span><span class="o">=</span><span class="mf">0.001</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-8</span><span class="p">,</span> <span class="n">weight_decay</span><span class="o">=</span><span class="mi">0</span><span class="p">):</span>
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span> <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">AdaGrad</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+        
+<div class="viewcode-block" id="AdaGrad.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adagrad/index.html#lmflow.optim.adagrad.AdaGrad.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+                <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">]</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">grad</span> <span class="o">=</span> <span class="n">grad</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">],</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;sum&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+                <span class="nb">sum</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;sum&#39;</span><span class="p">]</span>
+                <span class="nb">sum</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">grad</span><span class="p">)</span>
+                <span class="n">std</span> <span class="o">=</span> <span class="nb">sum</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">])</span>
+                <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">addcdiv_</span><span class="p">(</span><span class="o">-</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">],</span> <span class="n">grad</span><span class="p">,</span> <span class="n">std</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/adam.html b/_modules/lmflow/optim/adam.html
new file mode 100644
index 000000000..faa76ea2d
--- /dev/null
+++ b/_modules/lmflow/optim/adam.html
@@ -0,0 +1,493 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.adam &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/adam';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adam</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.adam</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+<div class="viewcode-block" id="Adam">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adam/index.html#lmflow.optim.adam.Adam">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Adam</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">params</span><span class="p">,</span> <span class="n">lr</span><span class="o">=</span><span class="mf">0.001</span><span class="p">,</span> <span class="n">betas</span><span class="o">=</span><span class="p">(</span><span class="mf">0.9</span><span class="p">,</span> <span class="mf">0.999</span><span class="p">),</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-8</span><span class="p">):</span>
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span> <span class="n">betas</span><span class="o">=</span><span class="n">betas</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">Adam</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+<div class="viewcode-block" id="Adam.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adam/index.html#lmflow.optim.adam.Adam.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+                
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_sq&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+                <span class="n">exp_avg</span><span class="p">,</span> <span class="n">exp_avg_sq</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">],</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_sq&#39;</span><span class="p">]</span>
+                <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;betas&#39;</span><span class="p">]</span>
+
+                <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
+
+                <span class="n">exp_avg</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">,</span> <span class="n">grad</span><span class="p">)</span>
+
+                <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta2</span><span class="p">)</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">grad</span><span class="p">)</span>
+
+                <span class="n">bias_correction1</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span>
+                <span class="n">bias_correction2</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span>
+
+                <span class="n">step_size</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="n">bias_correction2</span> <span class="o">**</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">/</span> <span class="n">bias_correction1</span>
+                <span class="n">denom</span> <span class="o">=</span> <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">])</span>
+                <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">addcdiv_</span><span class="p">(</span><span class="o">-</span><span class="n">step_size</span><span class="p">,</span> <span class="n">exp_avg</span><span class="p">,</span> <span class="n">denom</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/adamax.html b/_modules/lmflow/optim/adamax.html
new file mode 100644
index 000000000..50d749fb2
--- /dev/null
+++ b/_modules/lmflow/optim/adamax.html
@@ -0,0 +1,518 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.adamax &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/adamax';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adamax</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.adamax</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+<div class="viewcode-block" id="Adamax">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adamax/index.html#lmflow.optim.adamax.Adamax">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Adamax</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">params</span><span class="p">,</span> <span class="n">lr</span><span class="o">=</span><span class="mf">2e-3</span><span class="p">,</span> <span class="n">betas</span><span class="o">=</span><span class="p">(</span><span class="mf">0.9</span><span class="p">,</span> <span class="mf">0.999</span><span class="p">),</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-8</span><span class="p">,</span> <span class="n">weight_decay</span><span class="o">=</span><span class="mi">0</span><span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">lr</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">eps</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid epsilon value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">eps</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid beta parameter at index 0: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid beta parameter at index 1: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">weight_decay</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid weight_decay value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">weight_decay</span><span class="p">))</span>
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span> <span class="n">betas</span><span class="o">=</span><span class="n">betas</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span> <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">Adamax</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+<div class="viewcode-block" id="Adamax.__setstate__">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adamax/index.html#lmflow.optim.adamax.Adamax.__setstate__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__setstate__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">state</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">Adamax</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="n">__setstate__</span><span class="p">(</span><span class="n">state</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="Adamax.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adamax/index.html#lmflow.optim.adamax.Adamax.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+                <span class="k">if</span> <span class="n">grad</span><span class="o">.</span><span class="n">is_sparse</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s1">&#39;Adamax does not support sparse gradients&#39;</span><span class="p">)</span>
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="c1"># State initialization</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_inf&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+                <span class="n">exp_avg</span><span class="p">,</span> <span class="n">exp_inf</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">],</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_inf&#39;</span><span class="p">]</span>
+                <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;betas&#39;</span><span class="p">]</span>
+
+                <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
+
+                <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">]</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">grad</span> <span class="o">=</span> <span class="n">grad</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">],</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+                <span class="c1"># Update biased first moment estimate</span>
+                <span class="n">exp_avg</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">,</span> <span class="n">grad</span><span class="p">)</span>
+
+                <span class="c1"># Update the exponentially weighted infinity norm</span>
+                <span class="n">norm_buf</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span>
+                    <span class="n">exp_inf</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta2</span><span class="p">)</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span>
+                    <span class="n">grad</span><span class="o">.</span><span class="n">abs</span><span class="p">()</span><span class="o">.</span><span class="n">unsqueeze_</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+                <span class="p">],</span> <span class="mi">0</span><span class="p">)</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">norm_buf</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">keepdim</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">out</span><span class="o">=</span><span class="p">(</span><span class="n">exp_inf</span><span class="p">,</span> <span class="n">exp_inf</span><span class="o">.</span><span class="n">new</span><span class="p">()</span><span class="o">.</span><span class="n">long</span><span class="p">()))</span>
+
+                <span class="n">bias_correction</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span>
+                <span class="n">clr</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">]</span> <span class="o">/</span> <span class="n">bias_correction</span>
+
+                <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">addcdiv_</span><span class="p">(</span><span class="o">-</span><span class="n">clr</span><span class="p">,</span> <span class="n">exp_avg</span><span class="p">,</span> <span class="n">exp_inf</span> <span class="o">+</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">])</span>
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/adamp.html b/_modules/lmflow/optim/adamp.html
new file mode 100644
index 000000000..e82de98a8
--- /dev/null
+++ b/_modules/lmflow/optim/adamp.html
@@ -0,0 +1,632 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.adamp &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/adamp';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adamp</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.adamp</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+
+<div class="viewcode-block" id="AdamP">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adamp/index.html#lmflow.optim.adamp.AdamP">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">AdamP</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Implements AdamP algorithm.</span>
+
+<span class="sd">    It has been proposed in `Slowing Down the Weight Norm Increase in</span>
+<span class="sd">    Momentum-based Optimizers`</span>
+<span class="sd">    https://arxiv.org/abs/2006.08217</span>
+
+<span class="sd">    Note:</span>
+<span class="sd">        Reference code: https://github.com/clovaai/AdamP</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">params</span><span class="p">,</span>
+        <span class="n">lr</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-3</span><span class="p">,</span>
+        <span class="n">betas</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.9</span><span class="p">,</span> <span class="mf">0.999</span><span class="p">),</span>
+        <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-8</span><span class="p">,</span>
+        <span class="n">weight_decay</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">delta</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.1</span><span class="p">,</span>
+        <span class="n">wd_ratio</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.1</span><span class="p">,</span>
+        <span class="n">nesterov</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">lr</span> <span class="o">&lt;=</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">eps</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid epsilon value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">eps</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid beta parameter at index 0: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid beta parameter at index 1: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">weight_decay</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid weight_decay value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">weight_decay</span><span class="p">)</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">delta</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid delta value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">delta</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">wd_ratio</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid wd_ratio value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">wd_ratio</span><span class="p">))</span>
+
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span>
+            <span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span>
+            <span class="n">betas</span><span class="o">=</span><span class="n">betas</span><span class="p">,</span>
+            <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span>
+            <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">,</span>
+            <span class="n">delta</span><span class="o">=</span><span class="n">delta</span><span class="p">,</span>
+            <span class="n">wd_ratio</span><span class="o">=</span><span class="n">wd_ratio</span><span class="p">,</span>
+            <span class="n">nesterov</span><span class="o">=</span><span class="n">nesterov</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">AdamP</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="AdamP._channel_view">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adamp/index.html#lmflow.optim.adamp.AdamP._channel_view">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_channel_view</span><span class="p">(</span><span class="n">x</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span></div>
+
+
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="AdamP._layer_view">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adamp/index.html#lmflow.optim.adamp.AdamP._layer_view">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_layer_view</span><span class="p">(</span><span class="n">x</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span></div>
+
+
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="AdamP._cosine_similarity">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adamp/index.html#lmflow.optim.adamp.AdamP._cosine_similarity">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_cosine_similarity</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">eps</span><span class="p">,</span> <span class="n">view_func</span><span class="p">):</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">view_func</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">y</span> <span class="o">=</span> <span class="n">view_func</span><span class="p">(</span><span class="n">y</span><span class="p">)</span>
+
+        <span class="n">x_norm</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">eps</span><span class="p">)</span>
+        <span class="n">y_norm</span> <span class="o">=</span> <span class="n">y</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">eps</span><span class="p">)</span>
+        <span class="n">dot</span> <span class="o">=</span> <span class="p">(</span><span class="n">x</span> <span class="o">*</span> <span class="n">y</span><span class="p">)</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">dot</span><span class="o">.</span><span class="n">abs</span><span class="p">()</span> <span class="o">/</span> <span class="n">x_norm</span> <span class="o">/</span> <span class="n">y_norm</span></div>
+
+
+<div class="viewcode-block" id="AdamP._projection">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adamp/index.html#lmflow.optim.adamp.AdamP._projection">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_projection</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">p</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">perturb</span><span class="p">,</span> <span class="n">delta</span><span class="p">,</span> <span class="n">wd_ratio</span><span class="p">,</span> <span class="n">eps</span><span class="p">):</span>
+        <span class="n">wd</span> <span class="o">=</span> <span class="mi">1</span>
+        <span class="n">expand_size</span> <span class="o">=</span> <span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">view_func</span> <span class="ow">in</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_channel_view</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_layer_view</span><span class="p">]:</span>
+            <span class="n">cosine_sim</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_cosine_similarity</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">eps</span><span class="p">,</span> <span class="n">view_func</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">cosine_sim</span><span class="o">.</span><span class="n">max</span><span class="p">()</span> <span class="o">&lt;</span> <span class="n">delta</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">view_func</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">)):</span>
+                <span class="n">p_n</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span> <span class="o">/</span> <span class="n">view_func</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">view</span><span class="p">(</span>
+                    <span class="n">expand_size</span>
+                <span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">eps</span><span class="p">)</span>
+                <span class="n">perturb</span> <span class="o">-=</span> <span class="n">p_n</span> <span class="o">*</span> <span class="n">view_func</span><span class="p">(</span><span class="n">p_n</span> <span class="o">*</span> <span class="n">perturb</span><span class="p">)</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">view</span><span class="p">(</span>
+                    <span class="n">expand_size</span>
+                <span class="p">)</span>
+                <span class="n">wd</span> <span class="o">=</span> <span class="n">wd_ratio</span>
+
+                <span class="k">return</span> <span class="n">perturb</span><span class="p">,</span> <span class="n">wd</span>
+
+        <span class="k">return</span> <span class="n">perturb</span><span class="p">,</span> <span class="n">wd</span></div>
+
+
+<div class="viewcode-block" id="AdamP.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adamp/index.html#lmflow.optim.adamp.AdamP.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Performs a single optimization step.</span>
+
+<span class="sd">        Arguments:</span>
+<span class="sd">            closure: A closure that reevaluates the model and returns the loss.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;params&quot;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+                <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;betas&quot;</span><span class="p">]</span>
+                <span class="n">nesterov</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;nesterov&quot;</span><span class="p">]</span>
+
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="c1"># State initialization</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span>
+                        <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span>
+                    <span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg_sq&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span>
+                        <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span>
+                    <span class="p">)</span>
+
+                <span class="c1"># Adam</span>
+                <span class="n">exp_avg</span><span class="p">,</span> <span class="n">exp_avg_sq</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">],</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg_sq&quot;</span><span class="p">]</span>
+
+                <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
+                <span class="n">bias_correction1</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span>
+                <span class="n">bias_correction2</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span>
+
+                <span class="n">exp_avg</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">)</span>
+                <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta2</span><span class="p">)</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">)</span>
+
+                <span class="n">denom</span> <span class="o">=</span> <span class="p">(</span><span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">bias_correction2</span><span class="p">))</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span>
+                    <span class="n">group</span><span class="p">[</span><span class="s2">&quot;eps&quot;</span><span class="p">]</span>
+                <span class="p">)</span>
+                <span class="n">step_size</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;lr&quot;</span><span class="p">]</span> <span class="o">/</span> <span class="n">bias_correction1</span>
+
+                <span class="k">if</span> <span class="n">nesterov</span><span class="p">:</span>
+                    <span class="n">perturb</span> <span class="o">=</span> <span class="p">(</span><span class="n">beta1</span> <span class="o">*</span> <span class="n">exp_avg</span> <span class="o">+</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">)</span> <span class="o">*</span> <span class="n">grad</span><span class="p">)</span> <span class="o">/</span> <span class="n">denom</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">perturb</span> <span class="o">=</span> <span class="n">exp_avg</span> <span class="o">/</span> <span class="n">denom</span>
+
+                <span class="c1"># Projection</span>
+                <span class="n">wd_ratio</span> <span class="o">=</span> <span class="mi">1</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+                    <span class="n">perturb</span><span class="p">,</span> <span class="n">wd_ratio</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_projection</span><span class="p">(</span>
+                        <span class="n">p</span><span class="p">,</span>
+                        <span class="n">grad</span><span class="p">,</span>
+                        <span class="n">perturb</span><span class="p">,</span>
+                        <span class="n">group</span><span class="p">[</span><span class="s2">&quot;delta&quot;</span><span class="p">],</span>
+                        <span class="n">group</span><span class="p">[</span><span class="s2">&quot;wd_ratio&quot;</span><span class="p">],</span>
+                        <span class="n">group</span><span class="p">[</span><span class="s2">&quot;eps&quot;</span><span class="p">],</span>
+                    <span class="p">)</span>
+
+                <span class="c1"># Weight decay</span>
+                <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span>
+                        <span class="mi">1</span> <span class="o">-</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;lr&quot;</span><span class="p">]</span> <span class="o">*</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">]</span> <span class="o">*</span> <span class="n">wd_ratio</span>
+                    <span class="p">)</span>
+
+                <span class="c1"># Step</span>
+                <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">perturb</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=-</span><span class="n">step_size</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/adamw_schedule_free.html b/_modules/lmflow/optim/adamw_schedule_free.html
new file mode 100644
index 000000000..e29188273
--- /dev/null
+++ b/_modules/lmflow/optim/adamw_schedule_free.html
@@ -0,0 +1,632 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.adamw_schedule_free &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/adamw_schedule_free';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adamw_schedule_free</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.adamw_schedule_free</h1><div class="highlight"><pre>
+<span></span><span class="c1"># Copyright (c) Meta Platforms, Inc. and affiliates.</span>
+<span class="c1"># All rights reserved.</span>
+<span class="c1"># </span>
+<span class="c1"># This source code is licensed under the license found in the</span>
+<span class="c1"># LICENSE file in the root directory of this source tree.</span>
+<span class="c1">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.optim</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+
+<div class="viewcode-block" id="AdamWScheduleFree">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adamw_schedule_free/index.html#lmflow.optim.adamw_schedule_free.AdamWScheduleFree">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">AdamWScheduleFree</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">Optimizer</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Schedule-Free AdamW</span>
+<span class="sd">    As the name suggests, no scheduler is needed with this optimizer. </span>
+<span class="sd">    To add warmup, rather than using a learning rate schedule you can just</span>
+<span class="sd">    set the warmup_steps parameter.</span>
+<span class="sd">    </span>
+<span class="sd">    This optimizer requires that .train() and .eval() be called before the</span>
+<span class="sd">    beginning of training and evaluation respectively. The optimizer should</span>
+<span class="sd">    also be placed in eval mode when saving checkpoints.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
+                 <span class="n">params</span><span class="p">,</span> 
+                 <span class="n">lr</span><span class="o">=</span><span class="mf">0.0025</span><span class="p">,</span> 
+                 <span class="n">betas</span><span class="o">=</span><span class="p">(</span><span class="mf">0.9</span><span class="p">,</span> <span class="mf">0.999</span><span class="p">),</span> 
+                 <span class="n">eps</span><span class="o">=</span><span class="mf">1e-8</span><span class="p">,</span>
+                 <span class="n">weight_decay</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+                 <span class="n">warmup_steps</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+                 <span class="n">r</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+                 <span class="n">weight_lr_power</span><span class="o">=</span><span class="mf">2.0</span><span class="p">,</span>
+                 <span class="n">foreach</span><span class="o">=</span><span class="nb">hasattr</span><span class="p">(</span><span class="n">torch</span><span class="p">,</span> <span class="s2">&quot;_foreach_mul_&quot;</span><span class="p">)</span>
+                 <span class="p">):</span>
+
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span> 
+                        <span class="n">betas</span><span class="o">=</span><span class="n">betas</span><span class="p">,</span> 
+                        <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span>
+                        <span class="n">r</span><span class="o">=</span><span class="n">r</span><span class="p">,</span>
+                        <span class="n">k</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+                        <span class="n">warmup_steps</span><span class="o">=</span><span class="n">warmup_steps</span><span class="p">,</span>
+                        <span class="n">train_mode</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                        <span class="n">weight_sum</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+                        <span class="n">lr_max</span><span class="o">=-</span><span class="mf">1.0</span><span class="p">,</span>
+                        <span class="n">weight_lr_power</span><span class="o">=</span><span class="n">weight_lr_power</span><span class="p">,</span>
+                        <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">,</span>
+                        <span class="n">foreach</span><span class="o">=</span><span class="n">foreach</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+    
+<div class="viewcode-block" id="AdamWScheduleFree.eval">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adamw_schedule_free/index.html#lmflow.optim.adamw_schedule_free.AdamWScheduleFree.eval">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">eval</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">train_mode</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;train_mode&#39;</span><span class="p">]</span>
+            <span class="n">beta1</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;betas&#39;</span><span class="p">]</span>
+            <span class="k">if</span> <span class="n">train_mode</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                    <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+                    <span class="k">if</span> <span class="s1">&#39;z&#39;</span> <span class="ow">in</span> <span class="n">state</span><span class="p">:</span>
+                        <span class="c1"># Set p.data to x</span>
+                        <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">lerp_</span><span class="p">(</span><span class="n">end</span><span class="o">=</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;z&#39;</span><span class="p">],</span> <span class="n">weight</span><span class="o">=</span><span class="mi">1</span><span class="o">-</span><span class="mi">1</span><span class="o">/</span><span class="n">beta1</span><span class="p">)</span>
+                <span class="n">group</span><span class="p">[</span><span class="s1">&#39;train_mode&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+
+<div class="viewcode-block" id="AdamWScheduleFree.train">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adamw_schedule_free/index.html#lmflow.optim.adamw_schedule_free.AdamWScheduleFree.train">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">train</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">train_mode</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;train_mode&#39;</span><span class="p">]</span>
+            <span class="n">beta1</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;betas&#39;</span><span class="p">]</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">train_mode</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                    <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+                    <span class="k">if</span> <span class="s1">&#39;z&#39;</span> <span class="ow">in</span> <span class="n">state</span><span class="p">:</span>
+                        <span class="c1"># Set p.data to y</span>
+                        <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">lerp_</span><span class="p">(</span><span class="n">end</span><span class="o">=</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;z&#39;</span><span class="p">],</span> <span class="n">weight</span><span class="o">=</span><span class="mi">1</span><span class="o">-</span><span class="n">beta1</span><span class="p">)</span>
+                <span class="n">group</span><span class="p">[</span><span class="s1">&#39;train_mode&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="kc">True</span></div>
+
+
+<div class="viewcode-block" id="AdamWScheduleFree.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adamw_schedule_free/index.html#lmflow.optim.adamw_schedule_free.AdamWScheduleFree.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Performs a single optimization step.</span>
+
+<span class="sd">        Arguments:</span>
+<span class="sd">            closure (callable, optional): A closure that reevaluates the model</span>
+<span class="sd">                and returns the loss.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+        
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">eps</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">]</span>
+            <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;betas&#39;</span><span class="p">]</span>
+            <span class="n">decay</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">]</span>
+            <span class="n">k</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;k&#39;</span><span class="p">]</span>
+            <span class="n">r</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;r&#39;</span><span class="p">]</span>
+            <span class="n">warmup_steps</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;warmup_steps&#39;</span><span class="p">]</span>
+            <span class="n">weight_lr_power</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_lr_power&#39;</span><span class="p">]</span>
+            
+            <span class="k">if</span> <span class="n">k</span> <span class="o">&lt;</span> <span class="n">warmup_steps</span><span class="p">:</span>
+              <span class="n">sched</span> <span class="o">=</span> <span class="p">(</span><span class="n">k</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="n">warmup_steps</span>
+            <span class="k">else</span><span class="p">:</span>
+              <span class="n">sched</span> <span class="o">=</span> <span class="mf">1.0</span>
+            
+            <span class="n">bias_correction2</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span> <span class="o">**</span> <span class="p">(</span><span class="n">k</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span>
+            <span class="n">lr</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">]</span><span class="o">*</span><span class="n">sched</span><span class="o">*</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">bias_correction2</span><span class="p">)</span>
+            
+            <span class="n">lr_max</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr_max&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">lr</span><span class="p">,</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr_max&#39;</span><span class="p">])</span>
+            
+            <span class="n">weight</span> <span class="o">=</span> <span class="p">((</span><span class="n">k</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span><span class="o">**</span><span class="n">r</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="n">lr_max</span><span class="o">**</span><span class="n">weight_lr_power</span><span class="p">)</span>
+            <span class="n">weight_sum</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_sum&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_sum&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="n">weight</span>
+
+            <span class="k">try</span><span class="p">:</span>
+                <span class="n">ckp1</span> <span class="o">=</span> <span class="n">weight</span><span class="o">/</span><span class="n">weight_sum</span>
+            <span class="k">except</span> <span class="ne">ZeroDivisionError</span><span class="p">:</span>
+                <span class="n">ckp1</span> <span class="o">=</span> <span class="mi">0</span>
+
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;train_mode&#39;</span><span class="p">]:</span>
+                <span class="k">raise</span> <span class="ne">Exception</span><span class="p">(</span><span class="s2">&quot;Not in train mode!&quot;</span><span class="p">)</span>
+
+            <span class="n">active_p</span> <span class="o">=</span> <span class="p">[</span><span class="n">p</span> <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]</span> <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">]</span>
+            
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">active_p</span><span class="p">:</span>
+                <span class="k">if</span> <span class="s1">&#39;z&#39;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]:</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">][</span><span class="s1">&#39;z&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">clone</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">][</span><span class="s1">&#39;exp_avg_sq&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;foreach&#39;</span><span class="p">]</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">active_p</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">y</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">exp_avg_sq</span><span class="p">,</span> <span class="n">z</span> <span class="o">=</span> <span class="nb">zip</span><span class="p">(</span><span class="o">*</span><span class="p">[(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> 
+                                                <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="p">,</span> 
+                                                <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">][</span><span class="s1">&#39;exp_avg_sq&#39;</span><span class="p">],</span> 
+                                                <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">][</span><span class="s1">&#39;z&#39;</span><span class="p">])</span> 
+                                                <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">active_p</span><span class="p">])</span>
+
+                <span class="c1"># Decay the first and second moment running average coefficient</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_mul_</span><span class="p">(</span><span class="n">exp_avg_sq</span><span class="p">,</span> <span class="n">beta2</span><span class="p">)</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_addcmul_</span><span class="p">(</span><span class="n">exp_avg_sq</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="mi">1</span><span class="o">-</span><span class="n">beta2</span><span class="p">)</span>
+                <span class="n">denom</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_sqrt</span><span class="p">(</span><span class="n">exp_avg_sq</span><span class="p">)</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_add_</span><span class="p">(</span><span class="n">denom</span><span class="p">,</span> <span class="n">eps</span><span class="p">)</span>
+
+                <span class="c1"># Normalize grad in-place for memory efficiency</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_div_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">denom</span><span class="p">)</span>
+
+                <span class="c1"># Weight decay calculated at y</span>
+                <span class="k">if</span> <span class="n">decay</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_add_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">decay</span><span class="p">)</span>
+
+                <span class="c1"># These operations update y in-place,</span>
+                <span class="c1"># without computing x explicitly.</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_lerp_</span><span class="p">(</span><span class="n">y</span><span class="p">,</span> <span class="n">z</span><span class="p">,</span> <span class="n">weight</span><span class="o">=</span><span class="n">ckp1</span><span class="p">)</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_add_</span><span class="p">(</span><span class="n">y</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">lr</span><span class="o">*</span><span class="p">(</span><span class="n">beta1</span><span class="o">*</span><span class="p">(</span><span class="mi">1</span><span class="o">-</span><span class="n">ckp1</span><span class="p">)</span><span class="o">-</span><span class="mi">1</span><span class="p">))</span>
+
+                <span class="c1"># z step</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_sub_</span><span class="p">(</span><span class="n">z</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">lr</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">active_p</span><span class="p">:</span>
+                    <span class="n">y</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span> <span class="c1"># Notation to match theory</span>
+                    <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+
+                    <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                    <span class="n">z</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;z&#39;</span><span class="p">]</span>
+                    <span class="n">exp_avg_sq</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_sq&#39;</span><span class="p">]</span>
+
+                    <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta2</span><span class="p">)</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="mi">1</span><span class="o">-</span><span class="n">beta2</span><span class="p">)</span>
+                    <span class="n">denom</span> <span class="o">=</span> <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">eps</span><span class="p">)</span>
+
+                    <span class="c1"># Reuse grad buffer for memory efficiency</span>
+                    <span class="n">grad_normalized</span> <span class="o">=</span> <span class="n">grad</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="n">denom</span><span class="p">)</span>
+
+                    <span class="c1"># Weight decay calculated at y</span>
+                    <span class="k">if</span> <span class="n">decay</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="n">grad_normalized</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">y</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">decay</span><span class="p">)</span>
+
+                    <span class="c1"># These operations update y in-place,</span>
+                    <span class="c1"># without computing x explicitly.</span>
+                    <span class="n">y</span><span class="o">.</span><span class="n">lerp_</span><span class="p">(</span><span class="n">end</span><span class="o">=</span><span class="n">z</span><span class="p">,</span> <span class="n">weight</span><span class="o">=</span><span class="n">ckp1</span><span class="p">)</span>
+                    <span class="n">y</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">grad_normalized</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">lr</span><span class="o">*</span><span class="p">(</span><span class="n">beta1</span><span class="o">*</span><span class="p">(</span><span class="mi">1</span><span class="o">-</span><span class="n">ckp1</span><span class="p">)</span><span class="o">-</span><span class="mi">1</span><span class="p">))</span>
+
+                    <span class="c1"># z step</span>
+                    <span class="n">z</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">grad_normalized</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">lr</span><span class="p">)</span>
+
+            <span class="n">group</span><span class="p">[</span><span class="s1">&#39;k&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">k</span><span class="o">+</span><span class="mi">1</span>
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/adan.html b/_modules/lmflow/optim/adan.html
new file mode 100644
index 000000000..2c4bc54da
--- /dev/null
+++ b/_modules/lmflow/optim/adan.html
@@ -0,0 +1,739 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.adan &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/adan';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adan</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.adan</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">List</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch</span><span class="w"> </span><span class="kn">import</span> <span class="n">Tensor</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+<div class="viewcode-block" id="Adan">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adan/index.html#lmflow.optim.adan.Adan">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Adan</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Implements a pytorch variant of Adan.</span>
+
+<span class="sd">    Adan was proposed in</span>
+<span class="sd">    Adan : Adaptive Nesterov Momentum Algorithm for Faster Optimizing Deep Models.</span>
+<span class="sd">    https://arxiv.org/abs/2208.06677</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
+                 <span class="n">params</span><span class="p">,</span>
+                 <span class="n">lr</span><span class="o">=</span><span class="mf">1e-3</span><span class="p">,</span>
+                 <span class="n">betas</span><span class="o">=</span><span class="p">(</span><span class="mf">0.98</span><span class="p">,</span> <span class="mf">0.92</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">),</span>
+                 <span class="n">eps</span><span class="o">=</span><span class="mf">1e-8</span><span class="p">,</span>
+                 <span class="n">weight_decay</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+                 <span class="n">max_grad_norm</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+                 <span class="n">no_prox</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                 <span class="n">foreach</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">max_grad_norm</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s1">&#39;Invalid Max grad norm: </span><span class="si">{}</span><span class="s1">&#39;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">max_grad_norm</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">lr</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s1">&#39;Invalid learning rate: </span><span class="si">{}</span><span class="s1">&#39;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">eps</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s1">&#39;Invalid epsilon value: </span><span class="si">{}</span><span class="s1">&#39;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">eps</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s1">&#39;Invalid beta parameter at index 0: </span><span class="si">{}</span><span class="s1">&#39;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span>
+                <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s1">&#39;Invalid beta parameter at index 1: </span><span class="si">{}</span><span class="s1">&#39;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span>
+                <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s1">&#39;Invalid beta parameter at index 2: </span><span class="si">{}</span><span class="s1">&#39;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span>
+                <span class="n">betas</span><span class="p">[</span><span class="mi">2</span><span class="p">]))</span>
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span>
+            <span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span>
+            <span class="n">betas</span><span class="o">=</span><span class="n">betas</span><span class="p">,</span>
+            <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span>
+            <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">,</span>
+            <span class="n">max_grad_norm</span><span class="o">=</span><span class="n">max_grad_norm</span><span class="p">,</span>
+            <span class="n">no_prox</span><span class="o">=</span><span class="n">no_prox</span><span class="p">,</span>
+            <span class="n">foreach</span><span class="o">=</span><span class="n">foreach</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+<div class="viewcode-block" id="Adan.__setstate__">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adan/index.html#lmflow.optim.adan.Adan.__setstate__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__setstate__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">state</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">Adan</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="n">__setstate__</span><span class="p">(</span><span class="n">state</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">group</span><span class="o">.</span><span class="n">setdefault</span><span class="p">(</span><span class="s1">&#39;no_prox&#39;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span></div>
+
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<div class="viewcode-block" id="Adan.restart_opt">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adan/index.html#lmflow.optim.adan.Adan.restart_opt">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">restart_opt</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">group</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">requires_grad</span><span class="p">:</span>
+                    <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+                    <span class="c1"># State initialization</span>
+
+                    <span class="c1"># Exponential moving average of gradient values</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">)</span>
+                    <span class="c1"># Exponential moving average of squared gradient values</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_sq&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">)</span>
+                    <span class="c1"># Exponential moving average of gradient difference</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_diff&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">)</span></div>
+
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<div class="viewcode-block" id="Adan.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adan/index.html#lmflow.optim.adan.Adan.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Performs a single optimization step.&quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">defaults</span><span class="p">[</span><span class="s1">&#39;max_grad_norm&#39;</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">device</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;params&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">device</span>
+            <span class="n">global_grad_norm</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>
+
+            <span class="n">max_grad_norm</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">defaults</span><span class="p">[</span><span class="s1">&#39;max_grad_norm&#39;</span><span class="p">],</span> <span class="n">device</span><span class="o">=</span><span class="n">device</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+
+                <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                    <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span>
+                        <span class="n">global_grad_norm</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">grad</span><span class="o">.</span><span class="n">pow</span><span class="p">(</span><span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">sum</span><span class="p">())</span>
+
+            <span class="n">global_grad_norm</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">global_grad_norm</span><span class="p">)</span> <span class="o">+</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">]</span>
+
+            <span class="n">clip_global_grad_norm</span> <span class="o">=</span> \
+                <span class="n">torch</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="n">max_grad_norm</span> <span class="o">/</span> <span class="n">global_grad_norm</span><span class="p">,</span> <span class="nb">max</span><span class="o">=</span><span class="mf">1.0</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">clip_global_grad_norm</span> <span class="o">=</span> <span class="mf">1.0</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">params_with_grad</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">grads</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">exp_avgs</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">exp_avg_sqs</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">exp_avg_diffs</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">pre_grads</span> <span class="o">=</span> <span class="p">[]</span>
+
+            <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span><span class="p">,</span> <span class="n">beta3</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;betas&#39;</span><span class="p">]</span>
+            <span class="c1"># assume same step across group now to simplify things</span>
+            <span class="c1"># per parameter step can be easily support</span>
+            <span class="c1"># by making it tensor, or pass list into kernel</span>
+            <span class="k">if</span> <span class="s1">&#39;step&#39;</span> <span class="ow">in</span> <span class="n">group</span><span class="p">:</span>
+                <span class="n">group</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">group</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">1</span>
+
+            <span class="n">bias_correction1</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">-</span> <span class="n">beta1</span><span class="o">**</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span>
+            <span class="n">bias_correction2</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">-</span> <span class="n">beta2</span><span class="o">**</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span>
+            <span class="n">bias_correction3</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">-</span> <span class="n">beta3</span><span class="o">**</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span>
+
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">params_with_grad</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">p</span><span class="p">)</span>
+                <span class="n">grads</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="p">)</span>
+
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_sq&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_diff&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">)</span>
+
+                <span class="k">if</span> <span class="s1">&#39;pre_grad&#39;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">state</span> <span class="ow">or</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+                    <span class="c1"># at first step grad wouldn&#39;t be clipped</span>
+                    <span class="c1"># by `clip_global_grad_norm`</span>
+                    <span class="c1"># this is only to simplify implementation</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;pre_grad&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span>
+
+                <span class="n">exp_avgs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">])</span>
+                <span class="n">exp_avg_sqs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_sq&#39;</span><span class="p">])</span>
+                <span class="n">exp_avg_diffs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_diff&#39;</span><span class="p">])</span>
+                <span class="n">pre_grads</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;pre_grad&#39;</span><span class="p">])</span>
+
+            <span class="n">kwargs</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span>
+                <span class="n">params</span><span class="o">=</span><span class="n">params_with_grad</span><span class="p">,</span>
+                <span class="n">grads</span><span class="o">=</span><span class="n">grads</span><span class="p">,</span>
+                <span class="n">exp_avgs</span><span class="o">=</span><span class="n">exp_avgs</span><span class="p">,</span>
+                <span class="n">exp_avg_sqs</span><span class="o">=</span><span class="n">exp_avg_sqs</span><span class="p">,</span>
+                <span class="n">exp_avg_diffs</span><span class="o">=</span><span class="n">exp_avg_diffs</span><span class="p">,</span>
+                <span class="n">pre_grads</span><span class="o">=</span><span class="n">pre_grads</span><span class="p">,</span>
+                <span class="n">beta1</span><span class="o">=</span><span class="n">beta1</span><span class="p">,</span>
+                <span class="n">beta2</span><span class="o">=</span><span class="n">beta2</span><span class="p">,</span>
+                <span class="n">beta3</span><span class="o">=</span><span class="n">beta3</span><span class="p">,</span>
+                <span class="n">bias_correction1</span><span class="o">=</span><span class="n">bias_correction1</span><span class="p">,</span>
+                <span class="n">bias_correction2</span><span class="o">=</span><span class="n">bias_correction2</span><span class="p">,</span>
+                <span class="n">bias_correction3_sqrt</span><span class="o">=</span><span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">bias_correction3</span><span class="p">),</span>
+                <span class="n">lr</span><span class="o">=</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">],</span>
+                <span class="n">weight_decay</span><span class="o">=</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">],</span>
+                <span class="n">eps</span><span class="o">=</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">],</span>
+                <span class="n">no_prox</span><span class="o">=</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;no_prox&#39;</span><span class="p">],</span>
+                <span class="n">clip_global_grad_norm</span><span class="o">=</span><span class="n">clip_global_grad_norm</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;foreach&#39;</span><span class="p">]:</span>
+                <span class="n">copy_grads</span> <span class="o">=</span> <span class="n">_multi_tensor_adan</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">copy_grads</span> <span class="o">=</span> <span class="n">_single_tensor_adan</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+            <span class="k">for</span> <span class="n">p</span><span class="p">,</span> <span class="n">copy_grad</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">params_with_grad</span><span class="p">,</span> <span class="n">copy_grads</span><span class="p">):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">][</span><span class="s1">&#39;pre_grad&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">copy_grad</span></div>
+</div>
+
+
+
+<div class="viewcode-block" id="_single_tensor_adan">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adan/index.html#lmflow.optim.adan._single_tensor_adan">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_single_tensor_adan</span><span class="p">(</span>
+    <span class="n">params</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="n">grads</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="n">exp_avgs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="n">exp_avg_sqs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="n">exp_avg_diffs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="n">pre_grads</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">beta1</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">beta2</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">beta3</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">bias_correction1</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">bias_correction2</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">bias_correction3_sqrt</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">lr</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">weight_decay</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">no_prox</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="n">clip_global_grad_norm</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+<span class="p">):</span>
+    <span class="n">copy_grads</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">param</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">params</span><span class="p">):</span>
+        <span class="n">grad</span> <span class="o">=</span> <span class="n">grads</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+        <span class="n">exp_avg</span> <span class="o">=</span> <span class="n">exp_avgs</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+        <span class="n">exp_avg_sq</span> <span class="o">=</span> <span class="n">exp_avg_sqs</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+        <span class="n">exp_avg_diff</span> <span class="o">=</span> <span class="n">exp_avg_diffs</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+        <span class="n">pre_grad</span> <span class="o">=</span> <span class="n">pre_grads</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+
+        <span class="n">grad</span> <span class="o">=</span> <span class="n">grad</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">clip_global_grad_norm</span><span class="p">)</span>
+        <span class="n">copy_grads</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">grad</span><span class="o">.</span><span class="n">clone</span><span class="p">())</span>
+
+        <span class="n">diff</span> <span class="o">=</span> <span class="n">grad</span> <span class="o">-</span> <span class="n">pre_grad</span>
+        <span class="n">update</span> <span class="o">=</span> <span class="n">grad</span> <span class="o">+</span> <span class="n">beta2</span> <span class="o">*</span> <span class="n">diff</span>
+
+        <span class="n">exp_avg</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">)</span>  <span class="c1"># m_t</span>
+        <span class="n">exp_avg_diff</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta2</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">diff</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">)</span>  <span class="c1"># diff_t</span>
+        <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta3</span><span class="p">)</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span><span class="n">update</span><span class="p">,</span> <span class="n">update</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta3</span><span class="p">)</span>  <span class="c1"># n_t</span>
+
+        <span class="n">denom</span> <span class="o">=</span> <span class="p">(</span><span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span> <span class="o">/</span> <span class="n">bias_correction3_sqrt</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">eps</span><span class="p">)</span>
+        <span class="n">update</span> <span class="o">=</span> <span class="n">exp_avg</span> <span class="o">/</span> <span class="n">bias_correction1</span>
+        <span class="n">update</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">beta2</span> <span class="o">*</span> <span class="n">exp_avg_diff</span> <span class="o">/</span> <span class="n">bias_correction2</span><span class="p">)</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="n">denom</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">no_prox</span><span class="p">:</span>
+            <span class="n">param</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">lr</span> <span class="o">*</span> <span class="n">weight_decay</span><span class="p">)</span>
+            <span class="n">param</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">update</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=-</span><span class="n">lr</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">param</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">update</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=-</span><span class="n">lr</span><span class="p">)</span>
+            <span class="n">param</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="mi">1</span> <span class="o">+</span> <span class="n">lr</span> <span class="o">*</span> <span class="n">weight_decay</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">copy_grads</span></div>
+
+
+
+<div class="viewcode-block" id="_multi_tensor_adan">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/adan/index.html#lmflow.optim.adan._multi_tensor_adan">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_multi_tensor_adan</span><span class="p">(</span>
+    <span class="n">params</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="n">grads</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="n">exp_avgs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="n">exp_avg_sqs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="n">exp_avg_diffs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="n">pre_grads</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Tensor</span><span class="p">],</span>
+    <span class="o">*</span><span class="p">,</span>
+    <span class="n">beta1</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">beta2</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">beta3</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">bias_correction1</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">bias_correction2</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">bias_correction3_sqrt</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">lr</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">weight_decay</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="n">no_prox</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+    <span class="n">clip_global_grad_norm</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
+<span class="p">):</span>
+    <span class="k">if</span> <span class="n">clip_global_grad_norm</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_mul_</span><span class="p">(</span><span class="n">grads</span><span class="p">,</span> <span class="n">clip_global_grad_norm</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+    <span class="n">copy_grads</span> <span class="o">=</span> <span class="p">[</span><span class="n">g</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span> <span class="k">for</span> <span class="n">g</span> <span class="ow">in</span> <span class="n">grads</span><span class="p">]</span>
+
+    <span class="n">diff</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_sub</span><span class="p">(</span><span class="n">grads</span><span class="p">,</span> <span class="n">pre_grads</span><span class="p">)</span>
+    <span class="c1"># NOTE: line below while looking identical gives different result,</span>
+    <span class="c1"># due to float precision errors.</span>
+    <span class="c1"># using mul+add produces identical results to single-tensor,</span>
+    <span class="c1"># using add+alpha doesn&#39;t</span>
+    <span class="c1"># update = torch._foreach_add(grads, torch._foreach_mul(diff, beta2))</span>
+    <span class="n">update</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_add</span><span class="p">(</span><span class="n">grads</span><span class="p">,</span> <span class="n">diff</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">beta2</span><span class="p">)</span>
+
+    <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_mul_</span><span class="p">(</span><span class="n">exp_avgs</span><span class="p">,</span> <span class="n">beta1</span><span class="p">)</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_add_</span><span class="p">(</span><span class="n">exp_avgs</span><span class="p">,</span> <span class="n">grads</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">)</span>  <span class="c1"># m_t</span>
+
+    <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_mul_</span><span class="p">(</span><span class="n">exp_avg_diffs</span><span class="p">,</span> <span class="n">beta2</span><span class="p">)</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_add_</span><span class="p">(</span><span class="n">exp_avg_diffs</span><span class="p">,</span> <span class="n">diff</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">)</span>  <span class="c1"># diff_t</span>
+
+    <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_mul_</span><span class="p">(</span><span class="n">exp_avg_sqs</span><span class="p">,</span> <span class="n">beta3</span><span class="p">)</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_addcmul_</span><span class="p">(</span>
+        <span class="n">exp_avg_sqs</span><span class="p">,</span> <span class="n">update</span><span class="p">,</span> <span class="n">update</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta3</span><span class="p">)</span>  <span class="c1"># n_t</span>
+
+    <span class="n">denom</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_sqrt</span><span class="p">(</span><span class="n">exp_avg_sqs</span><span class="p">)</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_div_</span><span class="p">(</span><span class="n">denom</span><span class="p">,</span> <span class="n">bias_correction3_sqrt</span><span class="p">)</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_add_</span><span class="p">(</span><span class="n">denom</span><span class="p">,</span> <span class="n">eps</span><span class="p">)</span>
+
+    <span class="n">update</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_div</span><span class="p">(</span><span class="n">exp_avgs</span><span class="p">,</span> <span class="n">bias_correction1</span><span class="p">)</span>
+    <span class="c1"># NOTE: same issue as above.</span>
+    <span class="c1"># beta2 * diff / bias_correction2 != diff * (beta2 / bias_correction2)  # noqa</span>
+    <span class="c1"># using faster version by default. uncomment for tests to pass</span>
+    <span class="c1"># torch._foreach_add_(update, torch._foreach_div(torch._foreach_mul(exp_avg_diffs, beta2), bias_correction2))  # noqa</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_add_</span><span class="p">(</span>
+        <span class="n">update</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_mul</span><span class="p">(</span><span class="n">exp_avg_diffs</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">/</span> <span class="n">bias_correction2</span><span class="p">))</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_div_</span><span class="p">(</span><span class="n">update</span><span class="p">,</span> <span class="n">denom</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">no_prox</span><span class="p">:</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_mul_</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">lr</span> <span class="o">*</span> <span class="n">weight_decay</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_add_</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">update</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=-</span><span class="n">lr</span><span class="p">)</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_div_</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="mi">1</span> <span class="o">+</span> <span class="n">lr</span> <span class="o">*</span> <span class="n">weight_decay</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">copy_grads</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/dummy.html b/_modules/lmflow/optim/dummy.html
new file mode 100644
index 000000000..d5b9caf91
--- /dev/null
+++ b/_modules/lmflow/optim/dummy.html
@@ -0,0 +1,527 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.dummy &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/dummy';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.dummy</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.dummy</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;Dummy Optimizer.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">warnings</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Iterable</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch</span><span class="w"> </span><span class="kn">import</span> <span class="n">nn</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+<div class="viewcode-block" id="Dummy">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/dummy/index.html#lmflow.optim.dummy.Dummy">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Dummy</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    An dummy optimizer that does nothing.</span>
+
+<span class="sd">    Parameters:</span>
+<span class="sd">        params (:obj:`Iterable[nn.parameter.Parameter]`):</span>
+<span class="sd">            Iterable of parameters to optimize or dictionaries defining parameter groups.</span>
+<span class="sd">        lr (:obj:`float`, `optional`, defaults to 0):</span>
+<span class="sd">            The learning rate to use.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">params</span><span class="p">:</span> <span class="n">Iterable</span><span class="p">[</span><span class="n">nn</span><span class="o">.</span><span class="n">parameter</span><span class="o">.</span><span class="n">Parameter</span><span class="p">],</span>
+        <span class="n">lr</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.</span><span class="p">,</span>
+        <span class="n">betas</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">float</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.9</span><span class="p">,</span> <span class="mf">0.999</span><span class="p">),</span>
+        <span class="n">weight_decay</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">if</span> <span class="n">lr</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{</span><span class="n">lr</span><span class="si">}</span><span class="s2"> - should be &gt;= 0.0&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Invalid beta parameter: </span><span class="si">{</span><span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="si">}</span><span class="s2"> - should be in [0.0, 1.0)&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Invalid beta parameter: </span><span class="si">{</span><span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="si">}</span><span class="s2"> - should be in [0.0, 1.0)&quot;</span><span class="p">)</span>
+        <span class="n">defaults</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;lr&quot;</span><span class="p">:</span> <span class="n">lr</span><span class="p">,</span> <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="n">betas</span><span class="p">,</span> <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="n">weight_decay</span><span class="p">}</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<div class="viewcode-block" id="Dummy.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/dummy/index.html#lmflow.optim.dummy.Dummy.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span><span class="p">:</span> <span class="n">Callable</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Performs a single optimization step.</span>
+
+<span class="sd">        Arguments:</span>
+<span class="sd">            closure (:obj:`Callable`, `optional`): A closure that reevaluates the model and returns the loss.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;params&quot;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span>
+                <span class="k">if</span> <span class="n">grad</span><span class="o">.</span><span class="n">is_sparse</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Dummy does not support sparse gradients yet&quot;</span><span class="p">)</span>
+
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="c1"># State initialization</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg2&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">)</span>
+
+                <span class="c1"># v := exp_avg</span>
+                <span class="c1"># m := double_exp_avg</span>
+                <span class="n">v</span><span class="p">,</span> <span class="n">m</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">],</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg2&quot;</span><span class="p">]</span>
+                <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;betas&quot;</span><span class="p">]</span>
+                <span class="n">step_size</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;lr&quot;</span><span class="p">]</span>
+
+                <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
+
+                <span class="n">p</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=-</span><span class="mf">0.0</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mf">0.0</span><span class="p">:</span>
+                    <span class="n">p</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="n">group</span><span class="p">[</span><span class="s2">&quot;lr&quot;</span><span class="p">]</span> <span class="o">*</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">]))</span>
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/lamb.html b/_modules/lmflow/optim/lamb.html
new file mode 100644
index 000000000..fed2a1e71
--- /dev/null
+++ b/_modules/lmflow/optim/lamb.html
@@ -0,0 +1,588 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.lamb &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/lamb';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.lamb</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.lamb</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+<div class="viewcode-block" id="Lamb">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/lamb/index.html#lmflow.optim.lamb.Lamb">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Lamb</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Implements Lamb algorithm.</span>
+
+<span class="sd">    It has been proposed in `Large Batch Optimization for Deep Learning:</span>
+<span class="sd">    Training BERT in 76 minutes`</span>
+<span class="sd">    https://arxiv.org/abs/1904.00962</span>
+
+<span class="sd">    Note:</span>
+<span class="sd">        Reference code: https://github.com/cybertronai/pytorch-lamb</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">params</span><span class="p">,</span>
+        <span class="n">lr</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-3</span><span class="p">,</span>
+        <span class="n">betas</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.9</span><span class="p">,</span> <span class="mf">0.999</span><span class="p">),</span>
+        <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-6</span><span class="p">,</span>
+        <span class="n">weight_decay</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">clamp_value</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">10</span><span class="p">,</span>
+        <span class="n">adam</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">debias</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">lr</span> <span class="o">&lt;=</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">eps</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid epsilon value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">eps</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid beta parameter at index 0: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid beta parameter at index 1: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">weight_decay</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid weight_decay value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">weight_decay</span><span class="p">)</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">clamp_value</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid clamp value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">clamp_value</span><span class="p">))</span>
+
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span> <span class="n">betas</span><span class="o">=</span><span class="n">betas</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span> <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">)</span>
+<div class="viewcode-block" id="Lamb.clamp_value">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/lamb/index.html#lmflow.optim.lamb.Lamb.clamp_value">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">clamp_value</span> <span class="o">=</span> <span class="n">clamp_value</span></div>
+
+<div class="viewcode-block" id="Lamb.adam">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/lamb/index.html#lmflow.optim.lamb.Lamb.adam">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">adam</span> <span class="o">=</span> <span class="n">adam</span></div>
+
+<div class="viewcode-block" id="Lamb.debias">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/lamb/index.html#lmflow.optim.lamb.Lamb.debias">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">debias</span> <span class="o">=</span> <span class="n">debias</span></div>
+
+
+        <span class="nb">super</span><span class="p">(</span><span class="n">Lamb</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+<div class="viewcode-block" id="Lamb.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/lamb/index.html#lmflow.optim.lamb.Lamb.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Performs a single optimization step.</span>
+
+<span class="sd">        Arguments:</span>
+<span class="sd">            closure: A closure that reevaluates the model and returns the loss.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;params&quot;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+                <span class="k">if</span> <span class="n">grad</span><span class="o">.</span><span class="n">is_sparse</span><span class="p">:</span>
+                    <span class="n">msg</span> <span class="o">=</span> <span class="p">(</span>
+                        <span class="s2">&quot;Lamb does not support sparse gradients, &quot;</span>
+                        <span class="s2">&quot;please consider SparseAdam instead&quot;</span>
+                    <span class="p">)</span>
+                    <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="n">msg</span><span class="p">)</span>
+
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="c1"># State initialization</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="c1"># Exponential moving average of gradient values</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span>
+                        <span class="n">p</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span>
+                    <span class="p">)</span>
+                    <span class="c1"># Exponential moving average of squared gradient values</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg_sq&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span>
+                        <span class="n">p</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span>
+                    <span class="p">)</span>
+
+                <span class="n">exp_avg</span><span class="p">,</span> <span class="n">exp_avg_sq</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">],</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg_sq&quot;</span><span class="p">]</span>
+                <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;betas&quot;</span><span class="p">]</span>
+
+                <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
+
+                <span class="c1"># Decay the first and second moment running average coefficient</span>
+                <span class="c1"># m_t</span>
+                <span class="n">exp_avg</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">)</span>
+                <span class="c1"># v_t</span>
+                <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta2</span><span class="p">)</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">)</span>
+
+                <span class="c1"># Paper v3 does not use debiasing.</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">debias</span><span class="p">:</span>
+                    <span class="n">bias_correction</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">])</span>
+                    <span class="n">bias_correction</span> <span class="o">/=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">bias_correction</span> <span class="o">=</span> <span class="mi">1</span>
+
+                <span class="c1"># Apply bias to lr to avoid broadcast.</span>
+                <span class="n">step_size</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;lr&quot;</span><span class="p">]</span> <span class="o">*</span> <span class="n">bias_correction</span>
+
+                <span class="n">weight_norm</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span><span class="o">.</span><span class="n">clamp</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">clamp_value</span><span class="p">)</span>
+
+                <span class="n">adam_step</span> <span class="o">=</span> <span class="n">exp_avg</span> <span class="o">/</span> <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s2">&quot;eps&quot;</span><span class="p">])</span>
+                <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">]</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">adam_step</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">])</span>
+
+                <span class="n">adam_norm</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">adam_step</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">weight_norm</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">adam_norm</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">trust_ratio</span> <span class="o">=</span> <span class="mi">1</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">trust_ratio</span> <span class="o">=</span> <span class="n">weight_norm</span> <span class="o">/</span> <span class="n">adam_norm</span>
+                <span class="n">state</span><span class="p">[</span><span class="s2">&quot;weight_norm&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">weight_norm</span>
+                <span class="n">state</span><span class="p">[</span><span class="s2">&quot;adam_norm&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">adam_norm</span>
+                <span class="n">state</span><span class="p">[</span><span class="s2">&quot;trust_ratio&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">trust_ratio</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">adam</span><span class="p">:</span>
+                    <span class="n">trust_ratio</span> <span class="o">=</span> <span class="mi">1</span>
+
+                <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">adam_step</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=-</span><span class="n">step_size</span> <span class="o">*</span> <span class="n">trust_ratio</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/lars.html b/_modules/lmflow/optim/lars.html
new file mode 100644
index 000000000..1204fc5d3
--- /dev/null
+++ b/_modules/lmflow/optim/lars.html
@@ -0,0 +1,593 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.lars &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/lars';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.lars</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.lars</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+<div class="viewcode-block" id="LARS">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/lars/index.html#lmflow.optim.lars.LARS">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">LARS</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Extends SGD in PyTorch with LARS scaling from the paper</span>
+<span class="sd">    `Large batch training of Convolutional Networks`__.</span>
+<span class="sd">    .. note::</span>
+<span class="sd">        The application of momentum in the SGD part is modified according to</span>
+<span class="sd">        the PyTorch standards. LARS scaling fits into the equation in the</span>
+<span class="sd">        following fashion.</span>
+
+<span class="sd">        .. math::</span>
+<span class="sd">            \begin{aligned}</span>
+<span class="sd">                g_{t+1} &amp; = \text{lars_lr} * (\beta * p_{t} + g_{t+1}), \\</span>
+<span class="sd">                v_{t+1} &amp; = \\mu * v_{t} + g_{t+1}, \\</span>
+<span class="sd">                p_{t+1} &amp; = p_{t} - \text{lr} * v_{t+1},</span>
+<span class="sd">            \\end{aligned}</span>
+
+<span class="sd">        where :math:`p`, :math:`g`, :math:`v`, :math:`\\mu` and :math:`\beta`</span>
+<span class="sd">        denote the parameters, gradient, velocity, momentum, and weight decay</span>
+<span class="sd">        respectively.  The :math:`lars_lr` is defined by Eq. 6 in the paper.</span>
+<span class="sd">        The Nesterov version is analogously modified.</span>
+
+<span class="sd">    .. warning::</span>
+<span class="sd">        Parameters with weight decay set to 0 will automatically be excluded</span>
+<span class="sd">        from layer-wise LR scaling. This is to ensure consistency with papers</span>
+<span class="sd">        like SimCLR and BYOL.</span>
+
+
+<span class="sd">    __ https://arxiv.org/pdf/1708.03888.pdf</span>
+
+<span class="sd">    Note:</span>
+<span class="sd">        Reference code: https://github.com/PyTorchLightning/lightning-bolts/</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">params</span><span class="p">,</span>
+        <span class="n">lr</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-2</span><span class="p">,</span>
+        <span class="n">momentum</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">dampening</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">weight_decay</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">nesterov</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">trust_coefficient</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.01</span><span class="p">,</span>
+        <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-8</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">if</span> <span class="n">lr</span> <span class="o">&lt;=</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">eps</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid epsilon value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">eps</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">momentum</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid momentum value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">momentum</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">dampening</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid dampening value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">dampening</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">weight_decay</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid weight_decay value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">weight_decay</span><span class="p">)</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">trust_coefficient</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid trust_coefficient value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">trust_coefficient</span><span class="p">)</span>
+            <span class="p">)</span>
+
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span>
+            <span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span>
+            <span class="n">momentum</span><span class="o">=</span><span class="n">momentum</span><span class="p">,</span>
+            <span class="n">dampening</span><span class="o">=</span><span class="n">dampening</span><span class="p">,</span>
+            <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">,</span>
+            <span class="n">nesterov</span><span class="o">=</span><span class="n">nesterov</span><span class="p">,</span>
+            <span class="n">trust_coefficient</span><span class="o">=</span><span class="n">trust_coefficient</span><span class="p">,</span>
+            <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="n">nesterov</span> <span class="ow">and</span> <span class="p">(</span><span class="n">momentum</span> <span class="o">&lt;=</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">dampening</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Nesterov momentum requires a momentum and zero dampening&quot;</span>
+            <span class="p">)</span>
+
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+<div class="viewcode-block" id="LARS.__setstate__">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/lars/index.html#lmflow.optim.lars.LARS.__setstate__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__setstate__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">state</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">__setstate__</span><span class="p">(</span><span class="n">state</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">group</span><span class="o">.</span><span class="n">setdefault</span><span class="p">(</span><span class="s2">&quot;nesterov&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span></div>
+
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<div class="viewcode-block" id="LARS.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/lars/index.html#lmflow.optim.lars.LARS.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Performs a single optimization step.</span>
+
+<span class="sd">        Arguments:</span>
+<span class="sd">            closure: A closure that reevaluates the model and returns the loss.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">enable_grad</span><span class="p">():</span>
+                <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="c1"># exclude scaling for params with 0 weight decay</span>
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">weight_decay</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">]</span>
+            <span class="n">momentum</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;momentum&quot;</span><span class="p">]</span>
+            <span class="n">dampening</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;dampening&quot;</span><span class="p">]</span>
+            <span class="n">nesterov</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;nesterov&quot;</span><span class="p">]</span>
+
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;params&quot;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+
+                <span class="n">d_p</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span>
+                <span class="n">p_norm</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+                <span class="n">g_norm</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+                <span class="c1"># lars scaling + weight decay part</span>
+                <span class="k">if</span> <span class="n">weight_decay</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="n">p_norm</span> <span class="o">!=</span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">g_norm</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="n">lars_lr</span> <span class="o">=</span> <span class="n">p_norm</span> <span class="o">/</span> <span class="p">(</span>
+                            <span class="n">g_norm</span> <span class="o">+</span> <span class="n">p_norm</span> <span class="o">*</span> <span class="n">weight_decay</span> <span class="o">+</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;eps&quot;</span><span class="p">]</span>
+                        <span class="p">)</span>
+                        <span class="n">lars_lr</span> <span class="o">*=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;trust_coefficient&quot;</span><span class="p">]</span>
+
+                        <span class="n">d_p</span> <span class="o">=</span> <span class="n">d_p</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">)</span>
+                        <span class="n">d_p</span> <span class="o">*=</span> <span class="n">lars_lr</span>
+
+                <span class="k">if</span> <span class="n">momentum</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">param_state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+                    <span class="k">if</span> <span class="s2">&quot;momentum_buffer&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">param_state</span><span class="p">:</span>
+                        <span class="n">buf</span> <span class="o">=</span> <span class="n">param_state</span><span class="p">[</span><span class="s2">&quot;momentum_buffer&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">clone</span><span class="p">(</span>
+                            <span class="n">d_p</span>
+                        <span class="p">)</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">buf</span> <span class="o">=</span> <span class="n">param_state</span><span class="p">[</span><span class="s2">&quot;momentum_buffer&quot;</span><span class="p">]</span>
+                        <span class="n">buf</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">momentum</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">d_p</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">dampening</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">nesterov</span><span class="p">:</span>
+                        <span class="n">d_p</span> <span class="o">=</span> <span class="n">d_p</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">buf</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">momentum</span><span class="p">)</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">d_p</span> <span class="o">=</span> <span class="n">buf</span>
+
+                <span class="n">p</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">d_p</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=-</span><span class="n">group</span><span class="p">[</span><span class="s2">&quot;lr&quot;</span><span class="p">])</span>
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/nadam.html b/_modules/lmflow/optim/nadam.html
new file mode 100644
index 000000000..230b76958
--- /dev/null
+++ b/_modules/lmflow/optim/nadam.html
@@ -0,0 +1,523 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.nadam &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/nadam';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.nadam</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.nadam</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+
+<div class="viewcode-block" id="NAdam">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/nadam/index.html#lmflow.optim.nadam.NAdam">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">NAdam</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">Optimizer</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">params</span><span class="p">,</span> <span class="n">lr</span><span class="o">=</span><span class="mf">2e-3</span><span class="p">,</span> <span class="n">betas</span><span class="o">=</span><span class="p">(</span><span class="mf">0.9</span><span class="p">,</span> <span class="mf">0.999</span><span class="p">),</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-8</span><span class="p">,</span> <span class="n">weight_decay</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">momentum_decay</span><span class="o">=</span><span class="mf">4e-3</span><span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">lr</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">eps</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid epsilon value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">eps</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid beta parameter at index 0: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid beta parameter at index 1: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">weight_decay</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid weight_decay value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">weight_decay</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">momentum_decay</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid momentum_decay value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">momentum_decay</span><span class="p">))</span>
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span> <span class="n">betas</span><span class="o">=</span><span class="n">betas</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span> <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">,</span> <span class="n">momentum_decay</span><span class="o">=</span><span class="n">momentum_decay</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">NAdam</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+<div class="viewcode-block" id="NAdam.__setstate__">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/nadam/index.html#lmflow.optim.nadam.NAdam.__setstate__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__setstate__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">state</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">NAdam</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="n">__setstate__</span><span class="p">(</span><span class="n">state</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="NAdam.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/nadam/index.html#lmflow.optim.nadam.NAdam.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+                <span class="k">if</span> <span class="n">grad</span><span class="o">.</span><span class="n">is_sparse</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s1">&#39;NAdam does not support sparse gradients&#39;</span><span class="p">)</span>
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="c1"># State initialization</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;m_prev&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;v&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+                <span class="n">m_prev</span><span class="p">,</span> <span class="n">v</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;m_prev&#39;</span><span class="p">],</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;v&#39;</span><span class="p">]</span>
+                <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;betas&#39;</span><span class="p">]</span>
+
+                <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
+                <span class="n">bias_correction1</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span>
+                <span class="n">bias_correction2</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span>
+
+                <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">]</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">grad</span> <span class="o">=</span> <span class="n">grad</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">],</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+                <span class="n">m</span> <span class="o">=</span> <span class="n">beta1</span> <span class="o">*</span> <span class="n">m_prev</span> <span class="o">+</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">)</span> <span class="o">*</span> <span class="n">grad</span>
+                <span class="n">v</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta2</span><span class="p">)</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">grad</span><span class="p">)</span>
+
+                <span class="n">m_hat</span> <span class="o">=</span> <span class="n">m</span> <span class="o">/</span> <span class="n">bias_correction1</span>
+                <span class="n">v_hat</span> <span class="o">=</span> <span class="n">v</span> <span class="o">/</span> <span class="n">bias_correction2</span>
+
+                <span class="n">denom</span> <span class="o">=</span> <span class="n">v_hat</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">])</span>
+
+                <span class="n">momentum_decay</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;momentum_decay&#39;</span><span class="p">]</span>
+                <span class="n">m_prev</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">,</span> <span class="n">grad</span><span class="p">)</span>
+                <span class="n">m_prev_hat</span> <span class="o">=</span> <span class="n">m_prev</span> <span class="o">/</span> <span class="n">bias_correction1</span>
+
+                <span class="n">step_size</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">]</span> <span class="o">*</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">bias_correction2</span><span class="p">)</span> <span class="o">/</span> <span class="n">bias_correction1</span>
+
+                <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">addcdiv_</span><span class="p">(</span><span class="o">-</span><span class="n">step_size</span><span class="p">,</span> <span class="n">m_hat</span> <span class="o">+</span> <span class="n">momentum_decay</span> <span class="o">*</span> <span class="n">m_prev_hat</span><span class="p">,</span> <span class="n">denom</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/novograd.html b/_modules/lmflow/optim/novograd.html
new file mode 100644
index 000000000..a6af40ae8
--- /dev/null
+++ b/_modules/lmflow/optim/novograd.html
@@ -0,0 +1,534 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.novograd &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/novograd';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.novograd</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.novograd</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.optim</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">optim</span>
+
+<div class="viewcode-block" id="NovoGrad">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/novograd/index.html#lmflow.optim.novograd.NovoGrad">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">NovoGrad</span><span class="p">(</span><span class="n">optim</span><span class="o">.</span><span class="n">Optimizer</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">params</span><span class="p">,</span> <span class="n">lr</span><span class="o">=</span><span class="mf">0.01</span><span class="p">,</span> <span class="n">betas</span><span class="o">=</span><span class="p">(</span><span class="mf">0.9</span><span class="p">,</span> <span class="mf">0.999</span><span class="p">),</span> <span class="n">eps</span><span class="o">=</span><span class="mf">1e-8</span><span class="p">,</span> <span class="n">weight_decay</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">grad_averaging</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">amsgrad</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">lr</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">eps</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid epsilon value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">eps</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid beta parameter at index 0: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid beta parameter at index 1: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">weight_decay</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid weight_decay value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">weight_decay</span><span class="p">))</span>
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span> <span class="n">betas</span><span class="o">=</span><span class="n">betas</span><span class="p">,</span> <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span> <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">,</span> <span class="n">grad_averaging</span><span class="o">=</span><span class="n">grad_averaging</span><span class="p">,</span> <span class="n">amsgrad</span><span class="o">=</span><span class="n">amsgrad</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">NovoGrad</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+<div class="viewcode-block" id="NovoGrad.__setstate__">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/novograd/index.html#lmflow.optim.novograd.NovoGrad.__setstate__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__setstate__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">state</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">NovoGrad</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="n">__setstate__</span><span class="p">(</span><span class="n">state</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">group</span><span class="o">.</span><span class="n">setdefault</span><span class="p">(</span><span class="s1">&#39;amsgrad&#39;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="NovoGrad.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/novograd/index.html#lmflow.optim.novograd.NovoGrad.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+                <span class="k">if</span> <span class="n">grad</span><span class="o">.</span><span class="n">is_sparse</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s1">&#39;NovoGrad does not support sparse gradients&#39;</span><span class="p">)</span>
+                <span class="n">amsgrad</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;amsgrad&#39;</span><span class="p">]</span>
+
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="c1"># State initialization</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_sq&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">([])</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">amsgrad</span><span class="p">:</span>
+                        <span class="n">state</span><span class="p">[</span><span class="s1">&#39;max_exp_avg_sq&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">([])</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+
+                <span class="n">exp_avg</span><span class="p">,</span> <span class="n">exp_avg_sq</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">],</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg_sq&#39;</span><span class="p">]</span>
+                <span class="k">if</span> <span class="n">amsgrad</span><span class="p">:</span>
+                    <span class="n">max_exp_avg_sq</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s1">&#39;max_exp_avg_sq&#39;</span><span class="p">]</span>
+                <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;betas&#39;</span><span class="p">]</span>
+
+                <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
+
+                <span class="n">norm</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">pow</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="mi">2</span><span class="p">))</span>
+
+                <span class="k">if</span> <span class="n">exp_avg_sq</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">copy_</span><span class="p">(</span><span class="n">norm</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta2</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">,</span> <span class="n">norm</span><span class="p">)</span>
+
+                <span class="k">if</span> <span class="n">amsgrad</span><span class="p">:</span>
+                    <span class="c1"># Maintains the maximum of all 2nd moment running avg. till now</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">max_exp_avg_sq</span><span class="p">,</span> <span class="n">exp_avg_sq</span><span class="p">,</span> <span class="n">out</span><span class="o">=</span><span class="n">max_exp_avg_sq</span><span class="p">)</span>
+                    <span class="c1"># Use the max. for normalizing running avg. of gradient</span>
+                    <span class="n">denom</span> <span class="o">=</span> <span class="n">max_exp_avg_sq</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">])</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">denom</span> <span class="o">=</span> <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;eps&#39;</span><span class="p">])</span>
+
+                <span class="n">grad</span><span class="o">.</span><span class="n">div_</span><span class="p">(</span><span class="n">denom</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">]</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">grad</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">],</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+                <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;grad_averaging&#39;</span><span class="p">]:</span>
+                    <span class="n">grad</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">)</span>
+
+                <span class="n">exp_avg</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">grad</span><span class="p">)</span>
+
+                <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="o">-</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">],</span> <span class="n">exp_avg</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/radam.html b/_modules/lmflow/optim/radam.html
new file mode 100644
index 000000000..227bbdc1e
--- /dev/null
+++ b/_modules/lmflow/optim/radam.html
@@ -0,0 +1,623 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.radam &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/radam';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.radam</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.radam</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">warnings</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+<div class="viewcode-block" id="RAdam">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/radam/index.html#lmflow.optim.radam.RAdam">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">RAdam</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Implements RAdam optimization algorithm.</span>
+
+<span class="sd">    Note:</span>
+<span class="sd">        Deprecated, please use version provided by PyTorch_.</span>
+
+<span class="sd">    It has been proposed in `On the Variance of the Adaptive Learning</span>
+<span class="sd">    Rate and Beyond`.</span>
+<span class="sd">    https://arxiv.org/abs/1908.03265</span>
+
+<span class="sd">    Note:</span>
+<span class="sd">        Reference code: https://github.com/LiyuanLucasLiu/RAdam</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">params</span><span class="p">,</span>
+        <span class="n">lr</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-3</span><span class="p">,</span>
+        <span class="n">betas</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.9</span><span class="p">,</span> <span class="mf">0.999</span><span class="p">),</span>
+        <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-8</span><span class="p">,</span>
+        <span class="n">weight_decay</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+            <span class="s2">&quot;RAdam optimizer is deprecated, since it is included &quot;</span>
+            <span class="s2">&quot;in pytorch natively.&quot;</span><span class="p">,</span>
+            <span class="ne">DeprecationWarning</span><span class="p">,</span>
+            <span class="n">stacklevel</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="n">lr</span> <span class="o">&lt;=</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">eps</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid epsilon value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">eps</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid beta parameter at index 0: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid beta parameter at index 1: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">weight_decay</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid weight_decay value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">weight_decay</span><span class="p">)</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="p">(</span>
+            <span class="nb">isinstance</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">))</span>
+            <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">params</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span>
+            <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">params</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="nb">dict</span><span class="p">)</span>
+        <span class="p">):</span>
+            <span class="k">for</span> <span class="n">param</span> <span class="ow">in</span> <span class="n">params</span><span class="p">:</span>
+                <span class="k">if</span> <span class="s2">&quot;betas&quot;</span> <span class="ow">in</span> <span class="n">param</span> <span class="ow">and</span> <span class="p">(</span>
+                    <span class="n">param</span><span class="p">[</span><span class="s2">&quot;betas&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span> <span class="o">!=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+                    <span class="ow">or</span> <span class="n">param</span><span class="p">[</span><span class="s2">&quot;betas&quot;</span><span class="p">][</span><span class="mi">1</span><span class="p">]</span> <span class="o">!=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+                <span class="p">):</span>
+                    <span class="n">param</span><span class="p">[</span><span class="s2">&quot;buffer&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="p">[[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">10</span><span class="p">)]</span>
+
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span>
+            <span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span>
+            <span class="n">betas</span><span class="o">=</span><span class="n">betas</span><span class="p">,</span>
+            <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span>
+            <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">,</span>
+            <span class="n">buffer</span><span class="o">=</span><span class="p">[[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">10</span><span class="p">)],</span>
+        <span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">RAdam</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+<div class="viewcode-block" id="RAdam.__setstate__">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/radam/index.html#lmflow.optim.radam.RAdam.__setstate__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__setstate__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">state</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">RAdam</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="n">__setstate__</span><span class="p">(</span><span class="n">state</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RAdam.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/radam/index.html#lmflow.optim.radam.RAdam.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Performs a single optimization step.</span>
+
+<span class="sd">        Arguments:</span>
+<span class="sd">            closure: A closure that reevaluates the model and returns the loss.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">lr</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;lr&quot;</span><span class="p">]</span>
+            <span class="n">weight_decay</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">]</span>
+            <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;betas&quot;</span><span class="p">]</span>
+            <span class="n">eps</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;eps&quot;</span><span class="p">]</span>
+
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;params&quot;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">float</span><span class="p">()</span>
+                <span class="k">if</span> <span class="n">grad</span><span class="o">.</span><span class="n">is_sparse</span><span class="p">:</span>
+                    <span class="n">msg</span> <span class="o">=</span> <span class="p">(</span>
+                        <span class="s2">&quot;RAdam does not support sparse gradients, &quot;</span>
+                        <span class="s2">&quot;please consider SparseAdam instead&quot;</span>
+                    <span class="p">)</span>
+                    <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="n">msg</span><span class="p">)</span>
+
+                <span class="n">p_data_fp32</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">float</span><span class="p">()</span>
+
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span>
+                        <span class="n">p_data_fp32</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span>
+                    <span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg_sq&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span>
+                        <span class="n">p_data_fp32</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span>
+                    <span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">type_as</span><span class="p">(</span><span class="n">p_data_fp32</span><span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg_sq&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg_sq&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">type_as</span><span class="p">(</span>
+                        <span class="n">p_data_fp32</span>
+                    <span class="p">)</span>
+
+                <span class="n">exp_avg</span><span class="p">,</span> <span class="n">exp_avg_sq</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">],</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg_sq&quot;</span><span class="p">]</span>
+
+                <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta2</span><span class="p">)</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">)</span>
+                <span class="n">exp_avg</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">)</span>
+
+                <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
+                <span class="n">buffered</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;buffer&quot;</span><span class="p">][</span><span class="nb">int</span><span class="p">(</span><span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">%</span> <span class="mi">10</span><span class="p">)]</span>
+                <span class="k">if</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">==</span> <span class="n">buffered</span><span class="p">[</span><span class="mi">0</span><span class="p">]:</span>
+                    <span class="n">N_sma</span><span class="p">,</span> <span class="n">step_size</span> <span class="o">=</span> <span class="n">buffered</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">buffered</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">buffered</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span>
+                    <span class="n">beta2_t</span> <span class="o">=</span> <span class="n">beta2</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span>
+                    <span class="n">N_sma_max</span> <span class="o">=</span> <span class="mi">2</span> <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span>
+                    <span class="n">N_sma</span> <span class="o">=</span> <span class="n">N_sma_max</span> <span class="o">-</span> <span class="mi">2</span> <span class="o">*</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">*</span> <span class="n">beta2_t</span> <span class="o">/</span> <span class="p">(</span>
+                        <span class="mi">1</span> <span class="o">-</span> <span class="n">beta2_t</span>
+                    <span class="p">)</span>
+                    <span class="n">buffered</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="n">N_sma</span>
+
+                    <span class="c1"># more conservative since it&#39;s an approximated value</span>
+                    <span class="k">if</span> <span class="n">N_sma</span> <span class="o">&gt;=</span> <span class="mi">5</span><span class="p">:</span>
+                        <span class="n">step_size</span> <span class="o">=</span> <span class="p">(</span>
+                            <span class="n">lr</span>
+                            <span class="o">*</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span>
+                                <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2_t</span><span class="p">)</span>
+                                <span class="o">*</span> <span class="p">(</span><span class="n">N_sma</span> <span class="o">-</span> <span class="mi">4</span><span class="p">)</span>
+                                <span class="o">/</span> <span class="p">(</span><span class="n">N_sma_max</span> <span class="o">-</span> <span class="mi">4</span><span class="p">)</span>
+                                <span class="o">*</span> <span class="p">(</span><span class="n">N_sma</span> <span class="o">-</span> <span class="mi">2</span><span class="p">)</span>
+                                <span class="o">/</span> <span class="n">N_sma</span>
+                                <span class="o">*</span> <span class="n">N_sma_max</span>
+                                <span class="o">/</span> <span class="p">(</span><span class="n">N_sma_max</span> <span class="o">-</span> <span class="mi">2</span><span class="p">)</span>
+                            <span class="p">)</span>
+                            <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">])</span>
+                        <span class="p">)</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">step_size</span> <span class="o">=</span> <span class="n">lr</span> <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">])</span>
+                    <span class="n">buffered</span><span class="p">[</span><span class="mi">2</span><span class="p">]</span> <span class="o">=</span> <span class="n">step_size</span>
+
+                <span class="k">if</span> <span class="n">weight_decay</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">p_data_fp32</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">p_data_fp32</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=-</span><span class="n">weight_decay</span> <span class="o">*</span> <span class="n">lr</span><span class="p">)</span>
+
+                <span class="c1"># more conservative since it&#39;s an approximated value</span>
+                <span class="k">if</span> <span class="n">N_sma</span> <span class="o">&gt;=</span> <span class="mi">5</span><span class="p">:</span>
+                    <span class="n">denom</span> <span class="o">=</span> <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">eps</span><span class="p">)</span>
+                    <span class="n">p_data_fp32</span><span class="o">.</span><span class="n">addcdiv_</span><span class="p">(</span><span class="n">exp_avg</span><span class="p">,</span> <span class="n">denom</span><span class="p">,</span> <span class="n">value</span><span class="o">=-</span><span class="n">step_size</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">p_data_fp32</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">exp_avg</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=-</span><span class="n">step_size</span><span class="p">)</span>
+
+                <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">copy_</span><span class="p">(</span><span class="n">p_data_fp32</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/sgd_schedule_free.html b/_modules/lmflow/optim/sgd_schedule_free.html
new file mode 100644
index 000000000..fc90def3b
--- /dev/null
+++ b/_modules/lmflow/optim/sgd_schedule_free.html
@@ -0,0 +1,611 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.sgd_schedule_free &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/sgd_schedule_free';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.sgd_schedule_free</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.sgd_schedule_free</h1><div class="highlight"><pre>
+<span></span><span class="c1"># Copyright (c) Meta Platforms, Inc. and affiliates.</span>
+<span class="c1"># All rights reserved.</span>
+<span class="c1"># </span>
+<span class="c1"># This source code is licensed under the license found in the</span>
+<span class="c1"># LICENSE file in the root directory of this source tree.</span>
+<span class="c1">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.optim</span>
+
+<div class="viewcode-block" id="SGDScheduleFree">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sgd_schedule_free/index.html#lmflow.optim.sgd_schedule_free.SGDScheduleFree">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">SGDScheduleFree</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">Optimizer</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Schedule-Free SGD</span>
+<span class="sd">    As the name suggests, no scheduler is needed with this optimizer. </span>
+<span class="sd">    To add warmup, rather than using a learning rate schedule you can just</span>
+<span class="sd">    set the warmup_steps parameter.</span>
+
+<span class="sd">    This optimizer requires that .train() and .eval() be called before the</span>
+<span class="sd">    beginning of training and evaluation respectively. The optimizer should</span>
+<span class="sd">    also be placed in eval mode when saving checkpoints.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
+                 <span class="n">params</span><span class="p">,</span> 
+                 <span class="n">lr</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span> 
+                 <span class="n">momentum</span><span class="o">=</span><span class="mf">0.9</span><span class="p">,</span> 
+                 <span class="n">weight_decay</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+                 <span class="n">warmup_steps</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+                 <span class="n">r</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+                 <span class="n">weight_lr_power</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span>
+                 <span class="n">foreach</span><span class="o">=</span><span class="nb">hasattr</span><span class="p">(</span><span class="n">torch</span><span class="p">,</span> <span class="s2">&quot;_foreach_mul_&quot;</span><span class="p">),</span>
+                 <span class="p">):</span>
+        <span class="k">if</span> <span class="n">lr</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">weight_decay</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid weight_decay value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">weight_decay</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">momentum</span> <span class="o">&lt;=</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">momentum</span> <span class="o">&gt;=</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Momentum must be between 0 and 1 exclusive: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">momentum</span><span class="p">))</span>
+
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span> 
+                        <span class="n">momentum</span><span class="o">=</span><span class="n">momentum</span><span class="p">,</span> 
+                        <span class="n">r</span><span class="o">=</span><span class="n">r</span><span class="p">,</span>
+                        <span class="n">k</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+                        <span class="n">warmup_steps</span><span class="o">=</span><span class="n">warmup_steps</span><span class="p">,</span>
+                        <span class="n">train_mode</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                        <span class="n">weight_sum</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+                        <span class="n">lr_max</span><span class="o">=-</span><span class="mf">1.0</span><span class="p">,</span>
+                        <span class="n">weight_lr_power</span><span class="o">=</span><span class="n">weight_lr_power</span><span class="p">,</span>
+                        <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">,</span>
+                        <span class="n">foreach</span><span class="o">=</span><span class="n">foreach</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+    
+<div class="viewcode-block" id="SGDScheduleFree.eval">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sgd_schedule_free/index.html#lmflow.optim.sgd_schedule_free.SGDScheduleFree.eval">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">eval</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">train_mode</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;train_mode&#39;</span><span class="p">]</span>
+            <span class="n">momentum</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;momentum&#39;</span><span class="p">]</span>
+            <span class="k">if</span> <span class="n">train_mode</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                    <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+                    <span class="k">if</span> <span class="s1">&#39;z&#39;</span> <span class="ow">in</span> <span class="n">state</span><span class="p">:</span>
+                        <span class="c1"># Set p.data to x</span>
+                        <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">lerp_</span><span class="p">(</span><span class="n">end</span><span class="o">=</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;z&#39;</span><span class="p">],</span> <span class="n">weight</span><span class="o">=</span><span class="mi">1</span><span class="o">-</span><span class="mi">1</span><span class="o">/</span><span class="n">momentum</span><span class="p">)</span>
+                <span class="n">group</span><span class="p">[</span><span class="s1">&#39;train_mode&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+
+<div class="viewcode-block" id="SGDScheduleFree.train">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sgd_schedule_free/index.html#lmflow.optim.sgd_schedule_free.SGDScheduleFree.train">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">train</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">train_mode</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;train_mode&#39;</span><span class="p">]</span>
+            <span class="n">momentum</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;momentum&#39;</span><span class="p">]</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">train_mode</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                    <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+                    <span class="k">if</span> <span class="s1">&#39;z&#39;</span> <span class="ow">in</span> <span class="n">state</span><span class="p">:</span>
+                        <span class="c1"># Set p.data to y</span>
+                        <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">lerp_</span><span class="p">(</span><span class="n">end</span><span class="o">=</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;z&#39;</span><span class="p">],</span> <span class="n">weight</span><span class="o">=</span><span class="mi">1</span><span class="o">-</span><span class="n">momentum</span><span class="p">)</span>
+                <span class="n">group</span><span class="p">[</span><span class="s1">&#39;train_mode&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="kc">True</span></div>
+
+
+<div class="viewcode-block" id="SGDScheduleFree.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sgd_schedule_free/index.html#lmflow.optim.sgd_schedule_free.SGDScheduleFree.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Performs a single optimization step.</span>
+
+<span class="sd">        Arguments:</span>
+<span class="sd">            closure (callable, optional): A closure that reevaluates the model</span>
+<span class="sd">                and returns the loss.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+        
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">momentum</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;momentum&#39;</span><span class="p">]</span>
+            <span class="n">lr</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">]</span>
+            <span class="n">weight_decay</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">]</span>
+            <span class="n">k</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;k&#39;</span><span class="p">]</span>
+            <span class="n">warmup_steps</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;warmup_steps&#39;</span><span class="p">]</span>
+
+            <span class="k">if</span> <span class="n">k</span> <span class="o">&lt;</span> <span class="n">warmup_steps</span><span class="p">:</span>
+              <span class="n">sched</span> <span class="o">=</span> <span class="p">(</span><span class="n">k</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span> <span class="o">/</span> <span class="n">warmup_steps</span>
+            <span class="k">else</span><span class="p">:</span>
+              <span class="n">sched</span> <span class="o">=</span> <span class="mf">1.0</span>
+            <span class="n">lr</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">]</span><span class="o">*</span><span class="n">sched</span>
+
+            <span class="n">weight_lr_power</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_lr_power&#39;</span><span class="p">]</span>
+            
+            <span class="n">r</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;r&#39;</span><span class="p">]</span>
+            <span class="n">lr_max</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr_max&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">lr</span><span class="p">,</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr_max&#39;</span><span class="p">])</span>
+            
+            <span class="n">weight</span> <span class="o">=</span> <span class="p">((</span><span class="n">k</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span><span class="o">**</span><span class="n">r</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="n">lr_max</span><span class="o">**</span><span class="n">weight_lr_power</span><span class="p">)</span>
+            <span class="n">weight_sum</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_sum&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_sum&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="n">weight</span>
+
+            <span class="k">try</span><span class="p">:</span>
+                <span class="n">ckp1</span> <span class="o">=</span> <span class="n">weight</span><span class="o">/</span><span class="n">weight_sum</span>
+            <span class="k">except</span> <span class="ne">ZeroDivisionError</span><span class="p">:</span>
+                <span class="n">ckp1</span> <span class="o">=</span> <span class="mi">0</span>
+
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;train_mode&#39;</span><span class="p">]:</span>
+                <span class="k">raise</span> <span class="ne">Exception</span><span class="p">(</span><span class="s2">&quot;Not in train mode!&quot;</span><span class="p">)</span>
+
+            <span class="n">active_p</span> <span class="o">=</span> <span class="p">[</span><span class="n">p</span> <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]</span> <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">]</span>
+
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">active_p</span><span class="p">:</span>
+                <span class="k">if</span> <span class="s1">&#39;z&#39;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]:</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">][</span><span class="s1">&#39;z&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">clone</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;foreach&#39;</span><span class="p">]</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">active_p</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">y</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">z</span> <span class="o">=</span> <span class="nb">zip</span><span class="p">(</span><span class="o">*</span><span class="p">[(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">][</span><span class="s1">&#39;z&#39;</span><span class="p">])</span> 
+                                <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">active_p</span><span class="p">])</span>
+
+                <span class="c1"># Apply weight decay</span>
+                <span class="k">if</span> <span class="n">weight_decay</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_add_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">)</span>
+
+                <span class="c1"># These operations update y in-place,</span>
+                <span class="c1"># without computing x explicitly.</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_lerp_</span><span class="p">(</span><span class="n">y</span><span class="p">,</span> <span class="n">z</span><span class="p">,</span> <span class="n">weight</span><span class="o">=</span><span class="n">ckp1</span><span class="p">)</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_add_</span><span class="p">(</span><span class="n">y</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">lr</span><span class="o">*</span><span class="p">(</span><span class="n">momentum</span><span class="o">*</span><span class="p">(</span><span class="mi">1</span><span class="o">-</span><span class="n">ckp1</span><span class="p">)</span><span class="o">-</span><span class="mi">1</span><span class="p">))</span>
+
+                <span class="c1"># SGD step</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">_foreach_sub_</span><span class="p">(</span><span class="n">z</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">lr</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">active_p</span><span class="p">:</span>
+                    <span class="n">y</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span> <span class="c1"># Notation to match theory</span>
+                    <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+                    <span class="n">z</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">][</span><span class="s1">&#39;z&#39;</span><span class="p">]</span>
+
+                    <span class="c1"># Apply weight decay</span>
+                    <span class="k">if</span> <span class="n">weight_decay</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="n">grad</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">y</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">)</span>
+
+                    <span class="c1"># These operations update y in-place,</span>
+                    <span class="c1"># without computing x explicitly.</span>
+                    <span class="n">y</span><span class="o">.</span><span class="n">lerp_</span><span class="p">(</span><span class="n">end</span><span class="o">=</span><span class="n">z</span><span class="p">,</span> <span class="n">weight</span><span class="o">=</span><span class="n">ckp1</span><span class="p">)</span>
+                    <span class="n">y</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">lr</span><span class="o">*</span><span class="p">(</span><span class="n">momentum</span><span class="o">*</span><span class="p">(</span><span class="mi">1</span><span class="o">-</span><span class="n">ckp1</span><span class="p">)</span><span class="o">-</span><span class="mi">1</span><span class="p">))</span>
+
+                    <span class="c1"># SGD step</span>
+                    <span class="n">z</span><span class="o">.</span><span class="n">sub_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">lr</span><span class="p">)</span>
+
+            <span class="n">group</span><span class="p">[</span><span class="s1">&#39;k&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">k</span><span class="o">+</span><span class="mi">1</span>
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/sgdp.html b/_modules/lmflow/optim/sgdp.html
new file mode 100644
index 000000000..49798ee64
--- /dev/null
+++ b/_modules/lmflow/optim/sgdp.html
@@ -0,0 +1,620 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.sgdp &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/sgdp';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.sgdp</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.sgdp</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+
+<div class="viewcode-block" id="SGDP">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sgdp/index.html#lmflow.optim.sgdp.SGDP">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">SGDP</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Implements SGDP algorithm.</span>
+
+<span class="sd">    It has been proposed in `Slowing Down the Weight Norm Increase in</span>
+<span class="sd">    Momentum-based Optimizers`.</span>
+<span class="sd">    https://arxiv.org/abs/2006.08217</span>
+
+<span class="sd">    Note:</span>
+<span class="sd">        Reference code: https://github.com/clovaai/AdamP</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">params</span><span class="p">,</span>
+        <span class="n">lr</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-3</span><span class="p">,</span>
+        <span class="n">momentum</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">dampening</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-8</span><span class="p">,</span>
+        <span class="n">weight_decay</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">delta</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.1</span><span class="p">,</span>
+        <span class="n">wd_ratio</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.1</span><span class="p">,</span>
+        <span class="n">nesterov</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">lr</span> <span class="o">&lt;=</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">eps</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid epsilon value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">eps</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">momentum</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid momentum value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">momentum</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">dampening</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid dampening value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">dampening</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">weight_decay</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid weight_decay value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">weight_decay</span><span class="p">)</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">delta</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid delta value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">delta</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">wd_ratio</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid wd_ratio value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">wd_ratio</span><span class="p">))</span>
+
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span>
+            <span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span>
+            <span class="n">momentum</span><span class="o">=</span><span class="n">momentum</span><span class="p">,</span>
+            <span class="n">dampening</span><span class="o">=</span><span class="n">dampening</span><span class="p">,</span>
+            <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span>
+            <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">,</span>
+            <span class="n">delta</span><span class="o">=</span><span class="n">delta</span><span class="p">,</span>
+            <span class="n">wd_ratio</span><span class="o">=</span><span class="n">wd_ratio</span><span class="p">,</span>
+            <span class="n">nesterov</span><span class="o">=</span><span class="n">nesterov</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">SGDP</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="SGDP._channel_view">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sgdp/index.html#lmflow.optim.sgdp.SGDP._channel_view">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_channel_view</span><span class="p">(</span><span class="n">x</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span></div>
+
+
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="SGDP._layer_view">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sgdp/index.html#lmflow.optim.sgdp.SGDP._layer_view">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_layer_view</span><span class="p">(</span><span class="n">x</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span></div>
+
+
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="SGDP._cosine_similarity">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sgdp/index.html#lmflow.optim.sgdp.SGDP._cosine_similarity">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_cosine_similarity</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">eps</span><span class="p">,</span> <span class="n">view_func</span><span class="p">):</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">view_func</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
+        <span class="n">y</span> <span class="o">=</span> <span class="n">view_func</span><span class="p">(</span><span class="n">y</span><span class="p">)</span>
+
+        <span class="n">x_norm</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">eps</span><span class="p">)</span>
+        <span class="n">y_norm</span> <span class="o">=</span> <span class="n">y</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">eps</span><span class="p">)</span>
+        <span class="n">dot</span> <span class="o">=</span> <span class="p">(</span><span class="n">x</span> <span class="o">*</span> <span class="n">y</span><span class="p">)</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">dot</span><span class="o">.</span><span class="n">abs</span><span class="p">()</span> <span class="o">/</span> <span class="n">x_norm</span> <span class="o">/</span> <span class="n">y_norm</span></div>
+
+
+<div class="viewcode-block" id="SGDP._projection">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sgdp/index.html#lmflow.optim.sgdp.SGDP._projection">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_projection</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">p</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">perturb</span><span class="p">,</span> <span class="n">delta</span><span class="p">,</span> <span class="n">wd_ratio</span><span class="p">,</span> <span class="n">eps</span><span class="p">):</span>
+        <span class="n">wd</span> <span class="o">=</span> <span class="mi">1</span>
+        <span class="n">expand_size</span> <span class="o">=</span> <span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">-</span> <span class="mi">1</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">view_func</span> <span class="ow">in</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">_channel_view</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_layer_view</span><span class="p">]:</span>
+            <span class="n">cosine_sim</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_cosine_similarity</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">eps</span><span class="p">,</span> <span class="n">view_func</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">cosine_sim</span><span class="o">.</span><span class="n">max</span><span class="p">()</span> <span class="o">&lt;</span> <span class="n">delta</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">view_func</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">)):</span>
+                <span class="n">p_n</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">data</span> <span class="o">/</span> <span class="n">view_func</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">)</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">view</span><span class="p">(</span>
+                    <span class="n">expand_size</span>
+                <span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">eps</span><span class="p">)</span>
+                <span class="n">perturb</span> <span class="o">-=</span> <span class="n">p_n</span> <span class="o">*</span> <span class="n">view_func</span><span class="p">(</span><span class="n">p_n</span> <span class="o">*</span> <span class="n">perturb</span><span class="p">)</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">view</span><span class="p">(</span>
+                    <span class="n">expand_size</span>
+                <span class="p">)</span>
+                <span class="n">wd</span> <span class="o">=</span> <span class="n">wd_ratio</span>
+
+                <span class="k">return</span> <span class="n">perturb</span><span class="p">,</span> <span class="n">wd</span>
+
+        <span class="k">return</span> <span class="n">perturb</span><span class="p">,</span> <span class="n">wd</span></div>
+
+
+<div class="viewcode-block" id="SGDP.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sgdp/index.html#lmflow.optim.sgdp.SGDP.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Performs a single optimization step.</span>
+
+<span class="sd">        Arguments:</span>
+<span class="sd">            closure: A closure that reevaluates the model and returns the loss.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">weight_decay</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">]</span>
+            <span class="n">momentum</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;momentum&quot;</span><span class="p">]</span>
+            <span class="n">dampening</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;dampening&quot;</span><span class="p">]</span>
+            <span class="n">nesterov</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;nesterov&quot;</span><span class="p">]</span>
+
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;params&quot;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="c1"># State initialization</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;momentum&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span>
+                        <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span>
+                    <span class="p">)</span>
+
+                <span class="c1"># SGD</span>
+                <span class="n">buf</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;momentum&quot;</span><span class="p">]</span>
+                <span class="n">buf</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">momentum</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">dampening</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">nesterov</span><span class="p">:</span>
+                    <span class="n">d_p</span> <span class="o">=</span> <span class="n">grad</span> <span class="o">+</span> <span class="n">momentum</span> <span class="o">*</span> <span class="n">buf</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">d_p</span> <span class="o">=</span> <span class="n">buf</span>
+
+                <span class="c1"># Projection</span>
+                <span class="n">wd_ratio</span> <span class="o">=</span> <span class="mi">1</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+                    <span class="n">d_p</span><span class="p">,</span> <span class="n">wd_ratio</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_projection</span><span class="p">(</span>
+                        <span class="n">p</span><span class="p">,</span>
+                        <span class="n">grad</span><span class="p">,</span>
+                        <span class="n">d_p</span><span class="p">,</span>
+                        <span class="n">group</span><span class="p">[</span><span class="s2">&quot;delta&quot;</span><span class="p">],</span>
+                        <span class="n">group</span><span class="p">[</span><span class="s2">&quot;wd_ratio&quot;</span><span class="p">],</span>
+                        <span class="n">group</span><span class="p">[</span><span class="s2">&quot;eps&quot;</span><span class="p">],</span>
+                    <span class="p">)</span>
+
+                <span class="c1"># Weight decay</span>
+                <span class="k">if</span> <span class="n">weight_decay</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span>
+                        <span class="mi">1</span>
+                        <span class="o">-</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;lr&quot;</span><span class="p">]</span>
+                        <span class="o">*</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">]</span>
+                        <span class="o">*</span> <span class="n">wd_ratio</span>
+                        <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">momentum</span><span class="p">)</span>
+                    <span class="p">)</span>
+
+                <span class="c1"># Step</span>
+                <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">d_p</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=-</span><span class="n">group</span><span class="p">[</span><span class="s2">&quot;lr&quot;</span><span class="p">])</span>
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/sophia.html b/_modules/lmflow/optim/sophia.html
new file mode 100644
index 000000000..43da6207f
--- /dev/null
+++ b/_modules/lmflow/optim/sophia.html
@@ -0,0 +1,571 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.sophia &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/sophia';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.sophia</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.sophia</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+<div class="viewcode-block" id="SophiaG">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sophia/index.html#lmflow.optim.sophia.SophiaG">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">SophiaG</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training.</span>
+<span class="sd">    Code from: https://github.com/Liuhong99/Sophia/</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">params</span><span class="p">,</span> <span class="n">lr</span><span class="o">=</span><span class="mf">1e-4</span><span class="p">,</span> <span class="n">betas</span><span class="o">=</span><span class="p">(</span><span class="mf">0.965</span><span class="p">,</span> <span class="mf">0.99</span><span class="p">),</span> <span class="n">rho</span> <span class="o">=</span> <span class="mf">0.04</span><span class="p">,</span>
+         <span class="n">weight_decay</span><span class="o">=</span><span class="mf">1e-1</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span> <span class="n">maximize</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+         <span class="n">capturable</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">lr</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid beta parameter at index 0: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid beta parameter at index 1: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">rho</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid rho parameter at index 1: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">rho</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">weight_decay</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid weight_decay value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">weight_decay</span><span class="p">))</span>
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span><span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span> <span class="n">betas</span><span class="o">=</span><span class="n">betas</span><span class="p">,</span> <span class="n">rho</span><span class="o">=</span><span class="n">rho</span><span class="p">,</span> 
+                        <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">,</span> 
+                        <span class="n">maximize</span><span class="o">=</span><span class="n">maximize</span><span class="p">,</span> <span class="n">capturable</span><span class="o">=</span><span class="n">capturable</span><span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">SophiaG</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+<div class="viewcode-block" id="SophiaG.__setstate__">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sophia/index.html#lmflow.optim.sophia.SophiaG.__setstate__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__setstate__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">state</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">__setstate__</span><span class="p">(</span><span class="n">state</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">group</span><span class="o">.</span><span class="n">setdefault</span><span class="p">(</span><span class="s1">&#39;maximize&#39;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+            <span class="n">group</span><span class="o">.</span><span class="n">setdefault</span><span class="p">(</span><span class="s1">&#39;capturable&#39;</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+        <span class="n">state_values</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">values</span><span class="p">())</span>
+        <span class="n">step_is_tensor</span> <span class="o">=</span> <span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">state_values</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">)</span> <span class="ow">and</span> <span class="n">torch</span><span class="o">.</span><span class="n">is_tensor</span><span class="p">(</span><span class="n">state_values</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;step&#39;</span><span class="p">])</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">step_is_tensor</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">state_values</span><span class="p">:</span>
+                <span class="n">s</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="nb">float</span><span class="p">(</span><span class="n">s</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]))</span></div>
+
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<div class="viewcode-block" id="SophiaG.update_hessian">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sophia/index.html#lmflow.optim.sophia.SophiaG.update_hessian">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">update_hessian</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;betas&#39;</span><span class="p">]</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="mi">1</span><span class="p">,),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">p</span><span class="o">.</span><span class="n">device</span><span class="p">)</span> \
+                        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">defaults</span><span class="p">[</span><span class="s1">&#39;capturable&#39;</span><span class="p">]</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.</span><span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span><span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;hessian&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span><span class="p">)</span>
+                
+                <span class="k">if</span> <span class="s1">&#39;hessian&#39;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">state</span><span class="o">.</span><span class="n">keys</span><span class="p">():</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;hessian&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span><span class="p">)</span>
+
+                <span class="n">state</span><span class="p">[</span><span class="s1">&#39;hessian&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta2</span><span class="p">)</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="p">,</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">)</span></div>
+
+
+    <span class="nd">@torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()</span>
+<div class="viewcode-block" id="SophiaG.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/sophia/index.html#lmflow.optim.sophia.SophiaG.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">bs</span><span class="o">=</span><span class="mi">5120</span><span class="p">):</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">enable_grad</span><span class="p">():</span>
+                <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="n">params_with_grad</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">grads</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">exp_avgs</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">state_steps</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">hessian</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;betas&#39;</span><span class="p">]</span>
+
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;params&#39;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">params_with_grad</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">p</span><span class="p">)</span>
+                
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">is_sparse</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s1">&#39;SophiaG does not support sparse gradients&#39;</span><span class="p">)</span>
+                <span class="n">grads</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="p">)</span>
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+                <span class="c1"># State initialization</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">((</span><span class="mi">1</span><span class="p">,),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">p</span><span class="o">.</span><span class="n">device</span><span class="p">)</span> \
+                        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">defaults</span><span class="p">[</span><span class="s1">&#39;capturable&#39;</span><span class="p">]</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.</span><span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span><span class="p">)</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;hessian&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span><span class="p">)</span>
+                
+                <span class="k">if</span> <span class="s1">&#39;hessian&#39;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">state</span><span class="o">.</span><span class="n">keys</span><span class="p">():</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s1">&#39;hessian&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span><span class="p">)</span>                
+
+                <span class="n">exp_avgs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;exp_avg&#39;</span><span class="p">])</span>
+                <span class="n">state_steps</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;step&#39;</span><span class="p">])</span>
+                <span class="n">hessian</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">state</span><span class="p">[</span><span class="s1">&#39;hessian&#39;</span><span class="p">])</span>
+                
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">defaults</span><span class="p">[</span><span class="s1">&#39;capturable&#39;</span><span class="p">]:</span>
+                    <span class="n">bs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">((</span><span class="mi">1</span><span class="p">,),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">p</span><span class="o">.</span><span class="n">device</span><span class="p">)</span> <span class="o">*</span> <span class="n">bs</span>
+
+            <span class="c1"># Perform the actual update step here instead of calling SophiaG again</span>
+            <span class="k">for</span> <span class="n">p</span><span class="p">,</span> <span class="n">grad</span><span class="p">,</span> <span class="n">exp_avg</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">step</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">params_with_grad</span><span class="p">,</span> <span class="n">grads</span><span class="p">,</span> <span class="n">exp_avgs</span><span class="p">,</span> <span class="n">hessian</span><span class="p">,</span> <span class="n">state_steps</span><span class="p">):</span>
+                <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">]</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">grad</span> <span class="o">=</span> <span class="n">grad</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;weight_decay&#39;</span><span class="p">])</span>
+
+                <span class="c1"># Decay the first and second moment running average coefficient</span>
+                <span class="n">exp_avg</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">)</span>
+                <span class="n">step</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+
+                <span class="c1"># Compute the update using the hessian information</span>
+                <span class="n">update</span> <span class="o">=</span> <span class="n">exp_avg</span><span class="o">.</span><span class="n">div</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span> <span class="o">**</span> <span class="n">step</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+                <span class="n">h_sqrt</span> <span class="o">=</span> <span class="n">h</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;rho&#39;</span><span class="p">])</span>
+                <span class="n">p</span><span class="o">.</span><span class="n">addcdiv_</span><span class="p">(</span><span class="n">update</span><span class="p">,</span> <span class="n">h_sqrt</span><span class="p">,</span> <span class="n">value</span><span class="o">=-</span><span class="n">group</span><span class="p">[</span><span class="s1">&#39;lr&#39;</span><span class="p">])</span>
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+        
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/optim/yogi.html b/_modules/lmflow/optim/yogi.html
new file mode 100644
index 000000000..c5b92cb81
--- /dev/null
+++ b/_modules/lmflow/optim/yogi.html
@@ -0,0 +1,573 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.optim.yogi &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/optim/yogi';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.yogi</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.optim.yogi</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.nn</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">nn</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim.optimizer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optimizer</span>
+
+<div class="viewcode-block" id="Yogi">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/yogi/index.html#lmflow.optim.yogi.Yogi">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Yogi</span><span class="p">(</span><span class="n">Optimizer</span><span class="p">):</span>
+<span class="w">    </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Implements Yogi Optimizer Algorithm.</span>
+<span class="sd">    It has been proposed in `Adaptive methods for Nonconvex Optimization`.</span>
+
+<span class="sd">    https://papers.nips.cc/paper/8186-adaptive-methods-for-nonconvex-optimization  # noqa</span>
+
+<span class="sd">    Note:</span>
+<span class="sd">        Reference code: https://github.com/4rtemi5/Yogi-Optimizer_Keras</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">params</span><span class="p">,</span>
+        <span class="n">lr</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-2</span><span class="p">,</span>
+        <span class="n">betas</span> <span class="o">=</span> <span class="p">(</span><span class="mf">0.9</span><span class="p">,</span> <span class="mf">0.999</span><span class="p">),</span>
+        <span class="n">eps</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-3</span><span class="p">,</span>
+        <span class="n">initial_accumulator</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1e-6</span><span class="p">,</span>
+        <span class="n">weight_decay</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">lr</span> <span class="o">&lt;=</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid learning rate: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">lr</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">eps</span> <span class="o">&lt;</span> <span class="mf">0.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Invalid epsilon value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">eps</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid beta parameter at index 0: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="mf">0.0</span> <span class="o">&lt;=</span> <span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid beta parameter at index 1: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">betas</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">weight_decay</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Invalid weight_decay value: </span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">weight_decay</span><span class="p">)</span>
+            <span class="p">)</span>
+
+        <span class="n">defaults</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">(</span>
+            <span class="n">lr</span><span class="o">=</span><span class="n">lr</span><span class="p">,</span>
+            <span class="n">betas</span><span class="o">=</span><span class="n">betas</span><span class="p">,</span>
+            <span class="n">eps</span><span class="o">=</span><span class="n">eps</span><span class="p">,</span>
+            <span class="n">initial_accumulator</span><span class="o">=</span><span class="n">initial_accumulator</span><span class="p">,</span>
+            <span class="n">weight_decay</span><span class="o">=</span><span class="n">weight_decay</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="nb">super</span><span class="p">(</span><span class="n">Yogi</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">params</span><span class="p">,</span> <span class="n">defaults</span><span class="p">)</span>
+
+<div class="viewcode-block" id="Yogi.step">
+<a class="viewcode-back" href="../../../autoapi/lmflow/optim/yogi/index.html#lmflow.optim.yogi.Yogi.step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">closure</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Performs a single optimization step.</span>
+
+<span class="sd">        Arguments:</span>
+<span class="sd">            closure: A closure that reevaluates the model and returns the loss.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">closure</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">closure</span><span class="p">()</span>
+
+        <span class="k">for</span> <span class="n">group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;params&quot;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">grad</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">grad</span><span class="o">.</span><span class="n">data</span>
+                <span class="k">if</span> <span class="n">grad</span><span class="o">.</span><span class="n">is_sparse</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                        <span class="s2">&quot;Yogi does not support sparse gradients, &quot;</span>
+                        <span class="s2">&quot;please consider SparseAdam instead&quot;</span>
+                    <span class="p">)</span>
+
+                <span class="n">state</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">[</span><span class="n">p</span><span class="p">]</span>
+
+                <span class="c1"># State initialization</span>
+                <span class="c1"># Followed from official implementation in tensorflow addons:</span>
+                <span class="c1"># https://github.com/tensorflow/addons/blob/master/tensorflow_addons/optimizers/yogi.py#L118 # noqa</span>
+                <span class="c1"># For more details refer to the discussion:</span>
+                <span class="c1"># https://github.com/jettify/pytorch-optimizer/issues/77</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">state</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="c1"># Exponential moving average of gradient values</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span>
+                        <span class="n">torch</span><span class="o">.</span><span class="n">empty_like</span><span class="p">(</span>
+                            <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span>
+                        <span class="p">),</span>
+                        <span class="n">group</span><span class="p">[</span><span class="s2">&quot;initial_accumulator&quot;</span><span class="p">],</span>
+                    <span class="p">)</span>
+                    <span class="c1"># Exponential moving average of squared gradient values</span>
+                    <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg_sq&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">init</span><span class="o">.</span><span class="n">constant_</span><span class="p">(</span>
+                        <span class="n">torch</span><span class="o">.</span><span class="n">empty_like</span><span class="p">(</span>
+                            <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">memory_format</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">preserve_format</span>
+                        <span class="p">),</span>
+                        <span class="n">group</span><span class="p">[</span><span class="s2">&quot;initial_accumulator&quot;</span><span class="p">],</span>
+                    <span class="p">)</span>
+
+                <span class="n">exp_avg</span><span class="p">,</span> <span class="n">exp_avg_sq</span> <span class="o">=</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg&quot;</span><span class="p">],</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;exp_avg_sq&quot;</span><span class="p">]</span>
+                <span class="n">beta1</span><span class="p">,</span> <span class="n">beta2</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;betas&quot;</span><span class="p">]</span>
+
+                <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span> <span class="o">+=</span> <span class="mi">1</span>
+                <span class="n">bias_correction1</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span>
+                <span class="n">bias_correction2</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span> <span class="o">**</span> <span class="n">state</span><span class="p">[</span><span class="s2">&quot;step&quot;</span><span class="p">]</span>
+
+                <span class="k">if</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">]</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">grad</span> <span class="o">=</span> <span class="n">grad</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="n">group</span><span class="p">[</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">])</span>
+
+                <span class="c1"># Decay the first and second moment running average coefficient</span>
+                <span class="n">exp_avg</span><span class="o">.</span><span class="n">mul_</span><span class="p">(</span><span class="n">beta1</span><span class="p">)</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span><span class="n">grad</span><span class="p">,</span> <span class="n">alpha</span><span class="o">=</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta1</span><span class="p">)</span>
+
+                <span class="n">grad_squared</span> <span class="o">=</span> <span class="n">grad</span><span class="o">.</span><span class="n">mul</span><span class="p">(</span><span class="n">grad</span><span class="p">)</span>
+
+                <span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">addcmul_</span><span class="p">(</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">sign</span><span class="p">(</span><span class="n">exp_avg_sq</span> <span class="o">-</span> <span class="n">grad_squared</span><span class="p">),</span>
+                    <span class="n">grad_squared</span><span class="p">,</span>
+                    <span class="n">value</span><span class="o">=-</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">beta2</span><span class="p">),</span>
+                <span class="p">)</span>
+
+                <span class="n">denom</span> <span class="o">=</span> <span class="p">(</span><span class="n">exp_avg_sq</span><span class="o">.</span><span class="n">sqrt</span><span class="p">()</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">bias_correction2</span><span class="p">))</span><span class="o">.</span><span class="n">add_</span><span class="p">(</span>
+                    <span class="n">group</span><span class="p">[</span><span class="s2">&quot;eps&quot;</span><span class="p">]</span>
+                <span class="p">)</span>
+                <span class="n">step_size</span> <span class="o">=</span> <span class="n">group</span><span class="p">[</span><span class="s2">&quot;lr&quot;</span><span class="p">]</span> <span class="o">/</span> <span class="n">bias_correction1</span>
+                <span class="n">p</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">addcdiv_</span><span class="p">(</span><span class="n">exp_avg</span><span class="p">,</span> <span class="n">denom</span><span class="p">,</span> <span class="n">value</span><span class="o">=-</span><span class="n">step_size</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">loss</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/auto_pipeline.html b/_modules/lmflow/pipeline/auto_pipeline.html
new file mode 100644
index 000000000..57ce4d0ed
--- /dev/null
+++ b/_modules/lmflow/pipeline/auto_pipeline.html
@@ -0,0 +1,537 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.auto_pipeline &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/auto_pipeline';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.auto_pipeline</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.auto_pipeline</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;Return a pipeline automatically based on its name.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.versioning</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">is_package_version_at_least</span><span class="p">,</span> 
+    <span class="n">is_vllm_available</span><span class="p">,</span> 
+    <span class="n">is_trl_available</span><span class="p">,</span> 
+    <span class="n">is_ray_available</span>
+<span class="p">)</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.evaluator</span><span class="w"> </span><span class="kn">import</span> <span class="n">Evaluator</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.finetuner</span><span class="w"> </span><span class="kn">import</span> <span class="n">Finetuner</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.inferencer</span><span class="w"> </span><span class="kn">import</span> <span class="n">Inferencer</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.rm_tuner</span><span class="w"> </span><span class="kn">import</span> <span class="n">RewardModelTuner</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.rm_inferencer</span><span class="w"> </span><span class="kn">import</span> <span class="n">RewardModelInferencer</span>
+<div class="viewcode-block" id="PIPELINE_MAPPING">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/auto_pipeline/index.html#lmflow.pipeline.auto_pipeline.PIPELINE_MAPPING">[docs]</a>
+<span class="n">PIPELINE_MAPPING</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;evaluator&quot;</span><span class="p">:</span> <span class="n">Evaluator</span><span class="p">,</span>
+    <span class="s2">&quot;finetuner&quot;</span><span class="p">:</span> <span class="n">Finetuner</span><span class="p">,</span>
+    <span class="s2">&quot;inferencer&quot;</span><span class="p">:</span> <span class="n">Inferencer</span><span class="p">,</span>
+    <span class="s2">&quot;rm_inferencer&quot;</span><span class="p">:</span> <span class="n">RewardModelInferencer</span><span class="p">,</span>
+    <span class="s2">&quot;rm_tuner&quot;</span><span class="p">:</span> <span class="n">RewardModelTuner</span><span class="p">,</span>
+<span class="p">}</span></div>
+
+<div class="viewcode-block" id="PIPELINE_NEEDS_EXTRAS">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/auto_pipeline/index.html#lmflow.pipeline.auto_pipeline.PIPELINE_NEEDS_EXTRAS">[docs]</a>
+<span class="n">PIPELINE_NEEDS_EXTRAS</span> <span class="o">=</span> <span class="p">[]</span></div>
+
+
+<span class="k">if</span> <span class="ow">not</span> <span class="n">is_package_version_at_least</span><span class="p">(</span><span class="s1">&#39;transformers&#39;</span><span class="p">,</span> <span class="s1">&#39;4.35.0&#39;</span><span class="p">):</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.raft_aligner</span><span class="w"> </span><span class="kn">import</span> <span class="n">RaftAligner</span>
+    <span class="n">PIPELINE_MAPPING</span><span class="p">[</span><span class="s1">&#39;raft_aligner&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">RaftAligner</span>
+<span class="k">else</span><span class="p">:</span>
+    <span class="n">PIPELINE_NEEDS_EXTRAS</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s1">&#39;raft_aligner&#39;</span><span class="p">)</span>
+    
+<span class="k">if</span> <span class="n">is_vllm_available</span><span class="p">():</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.vllm_inferencer</span><span class="w"> </span><span class="kn">import</span> <span class="n">VLLMInferencer</span>
+    <span class="n">PIPELINE_MAPPING</span><span class="p">[</span><span class="s1">&#39;vllm_inferencer&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">VLLMInferencer</span>
+<span class="k">else</span><span class="p">:</span>
+    <span class="n">PIPELINE_NEEDS_EXTRAS</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s1">&#39;vllm_inferencer&#39;</span><span class="p">)</span>
+    
+<span class="k">if</span> <span class="n">is_trl_available</span><span class="p">():</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.dpo_aligner</span><span class="w"> </span><span class="kn">import</span> <span class="n">DPOAligner</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.dpov2_aligner</span><span class="w"> </span><span class="kn">import</span> <span class="n">DPOv2Aligner</span>
+    <span class="n">PIPELINE_MAPPING</span><span class="p">[</span><span class="s1">&#39;dpo_aligner&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">DPOAligner</span>
+    <span class="n">PIPELINE_MAPPING</span><span class="p">[</span><span class="s1">&#39;dpov2_aligner&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">DPOv2Aligner</span>
+<span class="k">else</span><span class="p">:</span>
+    <span class="n">PIPELINE_NEEDS_EXTRAS</span><span class="o">.</span><span class="n">extend</span><span class="p">([</span><span class="s1">&#39;dpo_aligner&#39;</span><span class="p">,</span> <span class="s1">&#39;dpov2_aligner&#39;</span><span class="p">])</span>
+    
+<span class="k">if</span> <span class="n">is_vllm_available</span><span class="p">()</span> <span class="ow">and</span> <span class="n">is_trl_available</span><span class="p">()</span> <span class="ow">and</span> <span class="n">is_ray_available</span><span class="p">():</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.iterative_dpo_aligner</span><span class="w"> </span><span class="kn">import</span> <span class="n">IterativeDPOAligner</span>
+    <span class="n">PIPELINE_MAPPING</span><span class="p">[</span><span class="s1">&#39;iterative_dpo_aligner&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">IterativeDPOAligner</span>
+<span class="k">else</span><span class="p">:</span>
+    <span class="n">PIPELINE_NEEDS_EXTRAS</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="s1">&#39;iterative_dpo_aligner&#39;</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="AutoPipeline">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/auto_pipeline/index.html#lmflow.pipeline.auto_pipeline.AutoPipeline">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">AutoPipeline</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot; </span>
+<span class="sd">    The class designed to return a pipeline automatically based on its name.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="nd">@classmethod</span>
+<div class="viewcode-block" id="AutoPipeline.get_pipeline">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/auto_pipeline/index.html#lmflow.pipeline.auto_pipeline.AutoPipeline.get_pipeline">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_pipeline</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
+        <span class="n">pipeline_name</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">,</span>
+        <span class="n">data_args</span><span class="p">,</span>
+        <span class="n">pipeline_args</span><span class="p">,</span>
+        <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">):</span>
+        <span class="k">if</span> <span class="n">pipeline_name</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">PIPELINE_MAPPING</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">pipeline_name</span> <span class="ow">in</span> <span class="n">PIPELINE_NEEDS_EXTRAS</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s1">&#39;Please install the necessary dependencies &#39;</span>
+                    <span class="sa">f</span><span class="s1">&#39;to use pipeline &quot;</span><span class="si">{</span><span class="n">pipeline_name</span><span class="si">}</span><span class="s1">&quot;&#39;</span>
+                <span class="p">)</span>
+                
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s1">&#39;Pipeline &quot;</span><span class="si">{</span><span class="n">pipeline_name</span><span class="si">}</span><span class="s1">&quot; is not supported&#39;</span>
+            <span class="p">)</span>
+
+        <span class="n">pipeline</span> <span class="o">=</span> <span class="n">PIPELINE_MAPPING</span><span class="p">[</span><span class="n">pipeline_name</span><span class="p">](</span>
+            <span class="n">model_args</span><span class="p">,</span>
+            <span class="n">data_args</span><span class="p">,</span>
+            <span class="n">pipeline_args</span><span class="p">,</span>
+            <span class="o">*</span><span class="n">args</span><span class="p">,</span>
+            <span class="o">**</span><span class="n">kwargs</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">pipeline</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/base_aligner.html b/_modules/lmflow/pipeline/base_aligner.html
new file mode 100644
index 000000000..55c789fa4
--- /dev/null
+++ b/_modules/lmflow/pipeline/base_aligner.html
@@ -0,0 +1,471 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.base_aligner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/base_aligner';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.base_aligner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.base_aligner</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot; BaseTuner: a subclass of BasePipeline.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.base_pipeline</span><span class="w"> </span><span class="kn">import</span> <span class="n">BasePipeline</span>
+
+
+<div class="viewcode-block" id="BaseAligner">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/base_aligner/index.html#lmflow.pipeline.base_aligner.BaseAligner">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">BaseAligner</span><span class="p">(</span><span class="n">BasePipeline</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot; A subclass of BasePipeline which is alignable.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+        <span class="k">pass</span>
+
+<div class="viewcode-block" id="BaseAligner._check_if_alignable">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/base_aligner/index.html#lmflow.pipeline.base_aligner.BaseAligner._check_if_alignable">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_check_if_alignable</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">dataset</span><span class="p">,</span> <span class="n">reward_model</span><span class="p">):</span>
+        <span class="c1"># TODO: check if the model is alignable and dataset is compatible</span>
+        <span class="c1"># TODO: add reward_model</span>
+        <span class="k">pass</span></div>
+
+
+<div class="viewcode-block" id="BaseAligner.align">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/base_aligner/index.html#lmflow.pipeline.base_aligner.BaseAligner.align">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">align</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">dataset</span><span class="p">,</span> <span class="n">reward_model</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">&quot;.align is not implemented&quot;</span><span class="p">)</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/base_pipeline.html b/_modules/lmflow/pipeline/base_pipeline.html
new file mode 100644
index 000000000..3b1af8f4f
--- /dev/null
+++ b/_modules/lmflow/pipeline/base_pipeline.html
@@ -0,0 +1,453 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.base_pipeline &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/base_pipeline';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.base_pipeline</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.base_pipeline</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot; BasePipeline.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">abc</span><span class="w"> </span><span class="kn">import</span> <span class="n">ABC</span>         <span class="c1"># abstract class</span>
+
+<div class="viewcode-block" id="BasePipeline">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/base_pipeline/index.html#lmflow.pipeline.base_pipeline.BasePipeline">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">BasePipeline</span><span class="p">(</span><span class="n">ABC</span><span class="p">):</span>
+    <span class="k">pass</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/base_tuner.html b/_modules/lmflow/pipeline/base_tuner.html
new file mode 100644
index 000000000..7bfb2f396
--- /dev/null
+++ b/_modules/lmflow/pipeline/base_tuner.html
@@ -0,0 +1,470 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.base_tuner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/base_tuner';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.base_tuner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.base_tuner</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot; BaseTuner: a subclass of BasePipeline.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.base_pipeline</span><span class="w"> </span><span class="kn">import</span> <span class="n">BasePipeline</span>
+
+
+<div class="viewcode-block" id="BaseTuner">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/base_tuner/index.html#lmflow.pipeline.base_tuner.BaseTuner">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">BaseTuner</span><span class="p">(</span><span class="n">BasePipeline</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot; A subclass of BasePipeline which is tunable.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+        <span class="k">pass</span>
+
+<div class="viewcode-block" id="BaseTuner._check_if_tunable">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/base_tuner/index.html#lmflow.pipeline.base_tuner.BaseTuner._check_if_tunable">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_check_if_tunable</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">dataset</span><span class="p">):</span>
+        <span class="c1"># TODO: check if the model is tunable and dataset is compatible</span>
+        <span class="k">pass</span></div>
+
+
+<div class="viewcode-block" id="BaseTuner.tune">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/base_tuner/index.html#lmflow.pipeline.base_tuner.BaseTuner.tune">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">tune</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">dataset</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">&quot;.tune is not implemented&quot;</span><span class="p">)</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/dpo_aligner.html b/_modules/lmflow/pipeline/dpo_aligner.html
new file mode 100644
index 000000000..f49b5a920
--- /dev/null
+++ b/_modules/lmflow/pipeline/dpo_aligner.html
@@ -0,0 +1,645 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.dpo_aligner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/dpo_aligner';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.dpo_aligner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.dpo_aligner</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># -*- coding: utf-8 -*-</span>
+<span class="c1"># @Time    : 7/4/2024 21:12</span>
+<span class="c1"># @Author  : Yu Li</span>
+<span class="c1"># @Site    :</span>
+<span class="c1"># @File    : dpo_pipeline.py</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">pathlib</span><span class="w"> </span><span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">datasets</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span><span class="p">,</span> <span class="n">load_dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">peft</span><span class="w"> </span><span class="kn">import</span> <span class="n">LoraConfig</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">TrainingArguments</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.base_aligner</span><span class="w"> </span><span class="kn">import</span> <span class="n">BaseAligner</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.versioning</span><span class="w"> </span><span class="kn">import</span> <span class="n">is_trl_available</span>
+
+<span class="k">if</span> <span class="n">is_trl_available</span><span class="p">():</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">trl</span><span class="w"> </span><span class="kn">import</span> <span class="n">DPOTrainer</span>
+<span class="k">else</span><span class="p">:</span>
+    <span class="k">raise</span> <span class="ne">ImportError</span><span class="p">(</span><span class="s2">&quot;Please install trl package to use dpo_aligner.py&quot;</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="get_paired_dataset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.get_paired_dataset">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">get_paired_dataset</span><span class="p">(</span>
+        <span class="n">data_root</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">data_dir</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">sanity_check</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">cache_dir</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">num_proc</span><span class="o">=</span><span class="mi">24</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dataset</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Load dataset and convert it to the necessary format.</span>
+
+<span class="sd">    The dataset is converted to a dictionary with the following structure:</span>
+<span class="sd">    {</span>
+<span class="sd">        &#39;prompt&#39;: List[str],</span>
+<span class="sd">        &#39;chosen&#39;: List[str],</span>
+<span class="sd">        &#39;rejected&#39;: List[str],</span>
+<span class="sd">    }</span>
+
+<span class="sd">    Prompts are structured as follows:</span>
+<span class="sd">      &quot;Question: &quot; + &lt;prompt&gt; + &quot;\n\nAnswer: &quot;</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">data_path</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">data_root</span><span class="p">)</span> <span class="o">/</span> <span class="n">data_dir</span>
+    <span class="n">data_files</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="n">x</span><span class="o">.</span><span class="n">absolute</span><span class="p">()</span><span class="o">.</span><span class="n">as_posix</span><span class="p">()</span>
+            <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">data_path</span><span class="o">.</span><span class="n">glob</span><span class="p">(</span><span class="s2">&quot;*.json&quot;</span><span class="p">)</span>
+    <span class="p">]</span>
+    <span class="n">dataset</span> <span class="o">=</span> <span class="n">load_dataset</span><span class="p">(</span>
+        <span class="n">path</span><span class="o">=</span><span class="n">data_root</span><span class="p">,</span>
+        <span class="n">split</span><span class="o">=</span><span class="s2">&quot;train&quot;</span><span class="p">,</span>
+        <span class="n">data_files</span><span class="o">=</span><span class="n">data_files</span><span class="p">,</span>
+        <span class="n">cache_dir</span><span class="o">=</span><span class="n">cache_dir</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="n">original_columns</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">column_names</span>
+
+    <span class="k">if</span> <span class="n">sanity_check</span><span class="p">:</span>
+        <span class="n">dataset</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">select</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="nb">min</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">dataset</span><span class="p">),</span> <span class="mi">1000</span><span class="p">)))</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="nf">return_prompt_and_responses</span><span class="p">(</span><span class="n">samples</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]:</span>
+        <span class="k">return</span> <span class="p">{</span>
+            <span class="s2">&quot;prompt&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;Question: &quot;</span> <span class="o">+</span> <span class="n">question</span> <span class="o">+</span> <span class="s2">&quot;</span><span class="se">\n\n</span><span class="s2">Answer: &quot;</span> <span class="k">for</span> <span class="n">question</span> <span class="ow">in</span> <span class="n">samples</span><span class="p">[</span><span class="s2">&quot;question&quot;</span><span class="p">]],</span>
+            <span class="s2">&quot;chosen&quot;</span><span class="p">:</span> <span class="n">samples</span><span class="p">[</span><span class="s2">&quot;response_j&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;rejected&quot;</span><span class="p">:</span> <span class="n">samples</span><span class="p">[</span><span class="s2">&quot;response_k&quot;</span><span class="p">],</span>
+        <span class="p">}</span>
+
+    <span class="k">return</span> <span class="n">dataset</span><span class="o">.</span><span class="n">map</span><span class="p">(</span>
+        <span class="n">return_prompt_and_responses</span><span class="p">,</span>
+        <span class="n">batched</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">num_proc</span><span class="o">=</span><span class="n">num_proc</span><span class="p">,</span>
+        <span class="n">remove_columns</span><span class="o">=</span><span class="n">original_columns</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="DPOAligner">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">DPOAligner</span><span class="p">(</span><span class="n">BaseAligner</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">aligner_args</span><span class="p">):</span>
+<div class="viewcode-block" id="DPOAligner.model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner.model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span> <span class="o">=</span> <span class="n">model_args</span></div>
+
+<div class="viewcode-block" id="DPOAligner.data_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner.data_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span> <span class="o">=</span> <span class="n">data_args</span></div>
+
+<div class="viewcode-block" id="DPOAligner.aligner_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner.aligner_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span> <span class="o">=</span> <span class="n">aligner_args</span></div>
+
+<div class="viewcode-block" id="DPOAligner.train_dataset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner.train_dataset">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="DPOAligner.eval_dataset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner.eval_dataset">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">eval_dataset</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+
+<div class="viewcode-block" id="DPOAligner._initialize_trainer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner._initialize_trainer">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_initialize_trainer</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">):</span>
+        <span class="n">peft_config</span> <span class="o">=</span> <span class="n">LoraConfig</span><span class="p">(</span>
+            <span class="n">r</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">lora_r</span><span class="p">,</span>
+            <span class="n">lora_alpha</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">lora_alpha</span><span class="p">,</span>
+            <span class="n">lora_dropout</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">lora_dropout</span><span class="p">,</span>
+            <span class="n">target_modules</span><span class="o">=</span><span class="p">[</span>
+                <span class="s2">&quot;q_proj&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;v_proj&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;k_proj&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;out_proj&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;fc_in&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;fc_out&quot;</span><span class="p">,</span>
+                <span class="s2">&quot;wte&quot;</span><span class="p">,</span>
+            <span class="p">],</span>
+            <span class="n">bias</span><span class="o">=</span><span class="s2">&quot;none&quot;</span><span class="p">,</span>
+            <span class="n">task_type</span><span class="o">=</span><span class="s2">&quot;CAUSAL_LM&quot;</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">training_args</span> <span class="o">=</span> <span class="n">TrainingArguments</span><span class="p">(</span>
+            <span class="n">per_device_train_batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">per_device_train_batch_size</span><span class="p">,</span>
+            <span class="n">per_device_eval_batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">per_device_eval_batch_size</span><span class="p">,</span>
+            <span class="n">max_steps</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">max_steps</span><span class="p">,</span>
+            <span class="n">logging_steps</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">logging_steps</span><span class="p">,</span>
+            <span class="n">save_steps</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">save_steps</span><span class="p">,</span>
+            <span class="n">gradient_accumulation_steps</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span><span class="p">,</span>
+            <span class="n">gradient_checkpointing</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">gradient_checkpointing</span><span class="p">,</span>
+            <span class="n">learning_rate</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">learning_rate</span><span class="p">,</span>
+            <span class="n">evaluation_strategy</span><span class="o">=</span><span class="s2">&quot;steps&quot;</span><span class="p">,</span>
+            <span class="n">eval_steps</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">eval_steps</span><span class="p">,</span>
+            <span class="n">output_dir</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span>
+            <span class="n">report_to</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">report_to</span><span class="p">,</span>
+            <span class="n">lr_scheduler_type</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">lr_scheduler_type</span><span class="p">,</span>
+            <span class="n">warmup_steps</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">warmup_steps</span><span class="p">,</span>
+            <span class="n">optim</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">optimizer_type</span><span class="p">,</span>
+            <span class="n">bf16</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+            <span class="n">remove_unused_columns</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+            <span class="n">run_name</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">run_name</span><span class="p">,</span>
+            <span class="n">ddp_find_unused_parameters</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+            <span class="c1"># gradient_checkpointing_kwargs=dict(use_reentrant=self.aligner_args.gradient_checkpointing_use_reentrant),</span>
+            <span class="n">seed</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">seed</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">dpo_trainer</span> <span class="o">=</span> <span class="n">DPOTrainer</span><span class="p">(</span>
+            <span class="n">model</span><span class="p">,</span>
+            <span class="n">ref_model</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+            <span class="n">args</span><span class="o">=</span><span class="n">training_args</span><span class="p">,</span>
+            <span class="n">beta</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">beta</span><span class="p">,</span>
+            <span class="n">train_dataset</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span><span class="p">,</span>
+            <span class="n">eval_dataset</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">eval_dataset</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">eval_dataset</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="n">tokenizer</span><span class="o">=</span><span class="n">tokenizer</span><span class="p">,</span>
+            <span class="n">peft_config</span><span class="o">=</span><span class="n">peft_config</span><span class="p">,</span>
+            <span class="n">max_prompt_length</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">beta</span><span class="p">,</span>
+            <span class="n">max_length</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">max_length</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">dpo_trainer</span></div>
+
+
+<div class="viewcode-block" id="DPOAligner._load_dataset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner._load_dataset">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_load_dataset</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="c1"># load training set</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span> <span class="o">=</span> <span class="n">get_paired_dataset</span><span class="p">(</span><span class="n">data_root</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">dataset_path</span><span class="p">,</span>
+                                                <span class="n">data_dir</span><span class="o">=</span><span class="s2">&quot;train&quot;</span><span class="p">,</span>
+                                                <span class="n">sanity_check</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">sanity_check</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span><span class="o">.</span><span class="n">filter</span><span class="p">(</span>
+            <span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="s2">&quot;prompt&quot;</span><span class="p">])</span> <span class="o">+</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="s2">&quot;chosen&quot;</span><span class="p">])</span> <span class="o">&lt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">max_length</span>
+                      <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="s2">&quot;prompt&quot;</span><span class="p">])</span> <span class="o">+</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="s2">&quot;rejected&quot;</span><span class="p">])</span> <span class="o">&lt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">max_length</span>
+        <span class="p">)</span>
+        <span class="c1"># load evaluation set</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">eval_dataset_path</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">eval_dataset</span> <span class="o">=</span> <span class="n">get_paired_dataset</span><span class="p">(</span><span class="n">data_root</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">eval_dataset_path</span><span class="p">,</span>
+                                                <span class="n">data_dir</span><span class="o">=</span><span class="s2">&quot;test&quot;</span><span class="p">,</span>
+                                                <span class="n">sanity_check</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">eval_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">eval_dataset</span><span class="o">.</span><span class="n">filter</span><span class="p">(</span>
+                <span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="s2">&quot;prompt&quot;</span><span class="p">])</span> <span class="o">+</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="s2">&quot;chosen&quot;</span><span class="p">])</span> <span class="o">&lt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">max_length</span>
+                        <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="s2">&quot;prompt&quot;</span><span class="p">])</span> <span class="o">+</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="s2">&quot;rejected&quot;</span><span class="p">])</span> <span class="o">&lt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">max_length</span>
+            <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="DPOAligner.align">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner.align">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">align</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">dataset</span><span class="p">,</span> <span class="n">reward_model</span><span class="p">):</span>
+        <span class="n">tokenizer</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">get_tokenizer</span><span class="p">()</span>
+        <span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token</span>
+        <span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_load_dataset</span><span class="p">()</span>
+
+        <span class="n">wrapped_model</span> <span class="o">=</span> <span class="n">model</span>
+        <span class="n">model</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span>
+
+        <span class="n">dpo_trainer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_initialize_trainer</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+        <span class="n">dpo_trainer</span><span class="o">.</span><span class="n">train</span><span class="p">()</span>
+        <span class="n">dpo_trainer</span><span class="o">.</span><span class="n">save_model</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">)</span>
+
+        <span class="c1"># 7. save</span>
+        <span class="n">output_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span> <span class="s2">&quot;final_checkpoint&quot;</span><span class="p">)</span>
+        <span class="n">dpo_trainer</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span></div>
+</div>
+
+
+
+
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/dpov2_aligner.html b/_modules/lmflow/pipeline/dpov2_aligner.html
new file mode 100644
index 000000000..0ffc6ffc8
--- /dev/null
+++ b/_modules/lmflow/pipeline/dpov2_aligner.html
@@ -0,0 +1,856 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.dpov2_aligner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/dpov2_aligner';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.dpov2_aligner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.dpov2_aligner</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span><span class="w"> </span><span class="nn">copy</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">importlib.resources</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">pkg_resources</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">subprocess</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">tqdm</span><span class="w"> </span><span class="kn">import</span> <span class="n">tqdm</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">TrainingArguments</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.utils.dpov2_trainer</span><span class="w"> </span><span class="kn">import</span> <span class="n">DPOv2Trainer</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.base_aligner</span><span class="w"> </span><span class="kn">import</span> <span class="n">BaseAligner</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">ModelArguments</span><span class="p">,</span>
+    <span class="n">DatasetArguments</span><span class="p">,</span>
+    <span class="n">DPOv2AlignerArguments</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.common</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">make_shell_args_from_dataclass</span><span class="p">,</span> 
+    <span class="n">add_dataclass_attr_prefix</span><span class="p">,</span> 
+    <span class="n">create_copied_dataclass</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_decoder_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFDecoderModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets.dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span><span class="p">,</span> <span class="n">KEY_SCORE</span><span class="p">,</span> <span class="n">KEY_TYPE</span><span class="p">,</span> <span class="n">KEY_INSTANCES</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="n">MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="ReferenceModelArguments">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.ReferenceModelArguments">[docs]</a>
+<span class="n">ReferenceModelArguments</span> <span class="o">=</span> <span class="n">create_copied_dataclass</span><span class="p">(</span>
+    <span class="n">original_dataclass</span><span class="o">=</span><span class="n">ModelArguments</span><span class="p">,</span> 
+    <span class="n">field_prefix</span><span class="o">=</span><span class="s2">&quot;reference_&quot;</span><span class="p">,</span>
+    <span class="n">class_prefix</span><span class="o">=</span><span class="s2">&quot;Reference&quot;</span>
+<span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="DPOv2Aligner">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">DPOv2Aligner</span><span class="p">(</span><span class="n">BaseAligner</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span>
+        <span class="n">aligner_args</span><span class="p">:</span> <span class="n">DPOv2AlignerArguments</span><span class="p">,</span>
+        <span class="n">ref_model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+    <span class="p">):</span>
+<div class="viewcode-block" id="DPOv2Aligner.model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span> <span class="o">=</span> <span class="n">model_args</span></div>
+
+<div class="viewcode-block" id="DPOv2Aligner.ref_model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.ref_model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ref_model_args</span> <span class="o">=</span> <span class="n">ref_model_args</span></div>
+
+<div class="viewcode-block" id="DPOv2Aligner.data_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.data_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span> <span class="o">=</span> <span class="n">data_args</span></div>
+
+<div class="viewcode-block" id="DPOv2Aligner.aligner_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.aligner_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span> <span class="o">=</span> <span class="n">aligner_args</span></div>
+
+
+
+<div class="viewcode-block" id="DPOv2Aligner.align">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.align">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">align</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span>
+        <span class="n">ref_model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span>
+        <span class="n">train_dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">eval_dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">transform_dataset_in_place</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">if</span> <span class="p">(</span><span class="n">train_dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;text_to_scored_textlist&quot;</span><span class="p">,</span> <span class="s2">&quot;paired_text_to_text&quot;</span><span class="p">])</span> <span class="ow">or</span> \
+            <span class="p">(</span><span class="n">eval_dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;text_to_scored_textlist&quot;</span><span class="p">,</span> <span class="s2">&quot;paired_text_to_text&quot;</span><span class="p">]):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Unsupported dataset type </span><span class="si">{</span><span class="n">train_dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span><span class="si">}</span><span class="s2"> for DPOv2 aligner.&quot;</span>
+            <span class="p">)</span>
+            
+        <span class="c1"># step 0. setting up</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">gradient_checkpointing</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                <span class="s2">&quot;Setting backend_model.config.use_cache to False since using gradient checkpointing&quot;</span>
+            <span class="p">)</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">use_cache</span> <span class="o">=</span> <span class="kc">False</span>
+            <span class="n">ref_model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">use_cache</span> <span class="o">=</span> <span class="kc">False</span>
+            
+        <span class="c1"># step 1. prepare datasets</span>
+        <span class="k">if</span> <span class="n">train_dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="o">==</span> <span class="s2">&quot;text_to_scored_textlist&quot;</span><span class="p">:</span>
+            <span class="n">train_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convert_to_paired_dataset</span><span class="p">(</span>
+                <span class="n">source_dataset</span><span class="o">=</span><span class="n">train_dataset</span><span class="p">,</span>
+                <span class="n">sampling_paired_method</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">sampling_paired_method</span><span class="p">,</span>
+                <span class="n">length_penalty</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">length_penalty</span><span class="p">,</span>
+                <span class="n">margin_scale</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">margin_scale</span><span class="p">,</span>
+                <span class="n">use_fast</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">max_train_samples</span><span class="p">:</span>
+            <span class="n">train_dataset</span><span class="o">.</span><span class="n">backend_dataset</span> <span class="o">=</span> <span class="n">train_dataset</span><span class="o">.</span><span class="n">backend_dataset</span><span class="o">.</span><span class="n">select</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">max_train_samples</span><span class="p">))</span>
+
+        <span class="k">if</span> <span class="n">eval_dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="o">==</span> <span class="s2">&quot;text_to_scored_textlist&quot;</span><span class="p">:</span>
+            <span class="n">eval_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">convert_to_paired_dataset</span><span class="p">(</span>
+                <span class="n">source_dataset</span><span class="o">=</span><span class="n">eval_dataset</span><span class="p">,</span>
+                <span class="n">sampling_paired_method</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">sampling_paired_method</span><span class="p">,</span>
+                <span class="n">margin_scale</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">margin_scale</span><span class="p">,</span>
+                <span class="n">use_fast</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+            <span class="p">)</span>
+            
+        <span class="c1"># step 2. prepare trainer</span>
+        <span class="n">dpo_trainer</span> <span class="o">=</span> <span class="n">DPOv2Trainer</span><span class="p">(</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">(),</span>
+            <span class="n">ref_model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">(),</span>
+            <span class="n">train_dataset</span><span class="o">=</span><span class="n">train_dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">(),</span> <span class="c1"># tokenization is done in the trainer</span>
+            <span class="n">eval_dataset</span><span class="o">=</span><span class="n">eval_dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">(),</span>
+            <span class="n">tokenizer</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">,</span>
+            <span class="n">args</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">__prepare_training_args</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="p">),</span>
+            <span class="n">beta</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">beta</span><span class="p">,</span>
+            <span class="n">loss_type</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">loss_type</span><span class="p">,</span>
+            <span class="n">max_prompt_length</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">max_prompt_length</span><span class="p">,</span>
+            <span class="n">max_length</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">max_length</span><span class="p">,</span>
+            <span class="n">mask_prompt</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">mask_prompt</span><span class="p">,</span>
+            <span class="n">len_penalty</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">length_penalty</span><span class="p">,</span>
+            <span class="c1"># preprocessing_num_workers=self.data_args.preprocessing_num_workers, # will trigger TypeError: cannot pickle &#39;torch._C._distributed_c10d.ProcessGroup&#39; object</span>
+        <span class="p">)</span>
+        
+        <span class="c1"># step 3. train</span>
+        <span class="n">dpo_trainer</span><span class="o">.</span><span class="n">train</span><span class="p">()</span>
+        <span class="n">dpo_trainer</span><span class="o">.</span><span class="n">save_model</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">)</span>
+
+        <span class="c1"># step 4. save</span>
+        <span class="n">output_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span> <span class="s2">&quot;final_checkpoint&quot;</span><span class="p">)</span>
+        <span class="n">dpo_trainer</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span>
+        
+        <span class="c1"># step 5. release resources</span>
+        <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">empty_cache</span><span class="p">()</span></div>
+
+        
+        
+<div class="viewcode-block" id="DPOv2Aligner.__prepare_training_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.__prepare_training_args">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__prepare_training_args</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">args</span><span class="p">:</span> <span class="n">DPOv2AlignerArguments</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">TrainingArguments</span><span class="p">:</span>
+        <span class="n">training_args</span> <span class="o">=</span> <span class="n">TrainingArguments</span><span class="p">(</span>
+            <span class="n">per_device_train_batch_size</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">per_device_train_batch_size</span><span class="p">,</span>
+            <span class="n">per_device_eval_batch_size</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">per_device_eval_batch_size</span><span class="p">,</span>
+            <span class="n">num_train_epochs</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">num_train_epochs</span><span class="p">,</span>
+            <span class="n">save_strategy</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">save_strategy</span><span class="p">,</span>
+            <span class="n">logging_steps</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">logging_steps</span><span class="p">,</span>
+            <span class="n">save_steps</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">save_steps</span><span class="p">,</span>
+            <span class="n">gradient_accumulation_steps</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span><span class="p">,</span>
+            <span class="n">gradient_checkpointing</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">gradient_checkpointing</span><span class="p">,</span>
+            <span class="n">learning_rate</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">learning_rate</span><span class="p">,</span>
+            <span class="n">evaluation_strategy</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">evaluation_strategy</span><span class="p">,</span>
+            <span class="n">eval_steps</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">eval_steps</span><span class="p">,</span>
+            <span class="n">output_dir</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span>
+            <span class="n">lr_scheduler_type</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">lr_scheduler_type</span><span class="p">,</span>
+            <span class="n">warmup_steps</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">warmup_steps</span><span class="p">,</span>
+            <span class="n">optim</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">optim</span><span class="p">,</span>
+            <span class="n">bf16</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">bf16</span><span class="p">,</span>
+            <span class="n">report_to</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">report_to</span><span class="p">,</span>
+            <span class="n">run_name</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">run_name</span><span class="p">,</span>
+            <span class="n">remove_unused_columns</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="c1"># DO NOT CHANGE THIS, may cause error https://discuss.huggingface.co/t/indexerror-invalid-key-16-is-out-of-bounds-for-size-0/14298/3</span>
+        <span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Actual training arguments for dpo trainer: </span><span class="si">{</span><span class="n">training_args</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">training_args</span></div>
+
+    
+    
+<div class="viewcode-block" id="DPOv2Aligner.convert_to_paired_dataset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.convert_to_paired_dataset">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">convert_to_paired_dataset</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">source_dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">sampling_paired_method</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s2">&quot;random&quot;</span><span class="p">,</span>
+        <span class="n">length_penalty</span><span class="p">:</span> <span class="nb">float</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">margin_scale</span><span class="p">:</span> <span class="nb">float</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span>
+        <span class="n">use_fast</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dataset</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Convert a scored one to multiple (text_to_scored_textlist) to a paired dataset by rejection sampling.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">output_dict</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="n">KEY_INSTANCES</span><span class="p">:</span> <span class="p">[]</span>
+        <span class="p">}</span>
+        <span class="k">if</span> <span class="n">source_dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;text_to_scored_textlist&quot;</span><span class="p">]:</span>
+            <span class="n">output_dict</span><span class="p">[</span><span class="n">KEY_TYPE</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;paired_text_to_text&quot;</span>
+        
+        <span class="k">for</span> <span class="n">sample</span> <span class="ow">in</span> <span class="n">tqdm</span><span class="p">(</span><span class="n">source_dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">(),</span> <span class="n">desc</span><span class="o">=</span><span class="s2">&quot;Converting to paired dataset&quot;</span><span class="p">):</span>
+            <span class="n">sample_output_dict</span> <span class="o">=</span> <span class="p">{}</span>
+            <span class="n">lengths</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_calc_response_lengths</span><span class="p">(</span><span class="n">sample</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">],</span> <span class="n">source_dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">())</span>
+            <span class="n">penalized_rewards</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_calc_reward_with_length_penalty</span><span class="p">(</span>
+                <span class="n">rewards</span><span class="o">=</span><span class="p">[</span><span class="n">content</span><span class="p">[</span><span class="n">KEY_SCORE</span><span class="p">]</span> <span class="k">for</span> <span class="n">content</span> <span class="ow">in</span> <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">]],</span> 
+                <span class="n">lengths</span><span class="o">=</span><span class="n">lengths</span><span class="p">,</span> 
+                <span class="n">length_penalty</span><span class="o">=</span><span class="n">length_penalty</span>
+            <span class="p">)</span>
+            <span class="n">chosen_idx</span><span class="p">,</span> <span class="n">rejected_idx</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">sampling_paired_idx_from_rewards</span><span class="p">(</span>
+                <span class="n">rewards</span><span class="o">=</span><span class="n">penalized_rewards</span><span class="p">,</span> 
+                <span class="n">sampling_paired_method</span><span class="o">=</span><span class="n">sampling_paired_method</span><span class="p">,</span>
+                <span class="n">use_fast</span><span class="o">=</span><span class="n">use_fast</span>
+            <span class="p">)</span>
+            
+            <span class="n">sample_output_dict</span><span class="p">[</span><span class="s2">&quot;prompt&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">]</span>
+            <span class="n">sample_output_dict</span><span class="p">[</span><span class="s2">&quot;chosen&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">][</span><span class="n">chosen_idx</span><span class="p">][</span><span class="s2">&quot;text&quot;</span><span class="p">]</span>
+            <span class="n">sample_output_dict</span><span class="p">[</span><span class="s2">&quot;rejected&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">][</span><span class="n">rejected_idx</span><span class="p">][</span><span class="s2">&quot;text&quot;</span><span class="p">]</span>
+            <span class="n">sample_output_dict</span><span class="p">[</span><span class="s2">&quot;margin&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="n">sample</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">][</span><span class="n">chosen_idx</span><span class="p">][</span><span class="n">KEY_SCORE</span><span class="p">]</span> <span class="o">-</span> <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">][</span><span class="n">rejected_idx</span><span class="p">][</span><span class="n">KEY_SCORE</span><span class="p">])</span> <span class="o">*</span> <span class="n">margin_scale</span>
+            <span class="n">output_dict</span><span class="p">[</span><span class="n">KEY_INSTANCES</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">sample_output_dict</span><span class="p">)</span>
+        
+        <span class="n">output_dataset_args</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">source_dataset</span><span class="o">.</span><span class="n">data_args</span><span class="p">)</span>
+        <span class="n">output_dataset_args</span><span class="o">.</span><span class="n">dataset_path</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">output_dataset_args</span><span class="o">.</span><span class="n">dataset_name</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;paired_</span><span class="si">{</span><span class="n">output_dataset_args</span><span class="o">.</span><span class="n">dataset_name</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">output_dataset</span> <span class="o">=</span> <span class="n">Dataset</span><span class="p">(</span><span class="n">output_dataset_args</span><span class="p">)</span>
+        <span class="n">output_dataset</span> <span class="o">=</span> <span class="n">output_dataset</span><span class="o">.</span><span class="n">from_dict</span><span class="p">(</span><span class="n">output_dict</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">output_dataset</span></div>
+
+    
+    
+<div class="viewcode-block" id="DPOv2Aligner._calc_response_lengths">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._calc_response_lengths">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_calc_response_lengths</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">outputs</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]]],</span>
+        <span class="n">dataset_type</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+        <span class="n">all_lengths</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">if</span> <span class="n">dataset_type</span> <span class="o">==</span> <span class="s2">&quot;text_to_scored_textlist&quot;</span><span class="p">:</span>
+            <span class="n">all_lengths</span> <span class="o">=</span> <span class="p">[</span><span class="nb">len</span><span class="p">(</span><span class="n">output</span><span class="p">[</span><span class="s2">&quot;text&quot;</span><span class="p">])</span> <span class="k">for</span> <span class="n">output</span> <span class="ow">in</span> <span class="n">outputs</span><span class="p">]</span>
+            
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Unknown dataset type </span><span class="si">{</span><span class="n">dataset_type</span><span class="si">}</span><span class="s2"> when calculating the response length.&quot;</span>
+            <span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">all_lengths</span></div>
+
+    
+    
+<div class="viewcode-block" id="DPOv2Aligner._calc_reward_with_length_penalty">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._calc_reward_with_length_penalty">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_calc_reward_with_length_penalty</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">rewards</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> 
+        <span class="n">lengths</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> 
+        <span class="n">length_penalty</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;When length_penalty &gt; 0, penalize the longer sequence by subtracting </span>
+<span class="sd">        length_penalty * length from the reward. Vice versa when length_penalty &lt; 0.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">assert</span> <span class="nb">len</span><span class="p">(</span><span class="n">rewards</span><span class="p">)</span> <span class="o">==</span> <span class="nb">len</span><span class="p">(</span><span class="n">lengths</span><span class="p">),</span> <span class="s2">&quot;The number of rewards and lengths should be the same.&quot;</span>
+        <span class="k">return</span> <span class="p">[</span><span class="n">reward</span> <span class="o">-</span> <span class="n">length_penalty</span> <span class="o">*</span> <span class="n">length</span> <span class="k">for</span> <span class="n">reward</span><span class="p">,</span> <span class="n">length</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">rewards</span><span class="p">,</span> <span class="n">lengths</span><span class="p">)]</span></div>
+
+    
+    
+<div class="viewcode-block" id="DPOv2Aligner.sampling_paired_idx_from_rewards">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.sampling_paired_idx_from_rewards">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">sampling_paired_idx_from_rewards</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">rewards</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+        <span class="n">sampling_paired_method</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s2">&quot;random&quot;</span><span class="p">,</span>
+        <span class="n">use_fast</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Prepare the dataset for DPO training by rejection sampling.</span>
+<span class="sd">        We implement different strategies to select pairs, including</span>
+<span class="sd">        random: randomly select two instances</span>
+<span class="sd">        max_min: best v.s. worst</span>
+<span class="sd">        max_max: best v.s. second best</span>
+<span class="sd">        max_random: best v.s. random from the remaining</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">use_fast</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_sampling_paired_idx_from_rewards_fast</span><span class="p">(</span><span class="n">rewards</span><span class="p">,</span> <span class="n">sampling_paired_method</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_sampling_paired_idx_from_rewards</span><span class="p">(</span><span class="n">rewards</span><span class="p">,</span> <span class="n">sampling_paired_method</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="DPOv2Aligner._sampling_paired_idx_from_rewards">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._sampling_paired_idx_from_rewards">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_sampling_paired_idx_from_rewards</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">rewards</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> 
+        <span class="n">sampling_paired_method</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s2">&quot;random&quot;</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+        <span class="n">idx_0</span><span class="p">,</span> <span class="n">idx_1</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span>
+        
+        <span class="k">if</span> <span class="n">sampling_paired_method</span> <span class="o">==</span> <span class="s2">&quot;random&quot;</span><span class="p">:</span>
+            <span class="n">idx_0</span><span class="p">,</span> <span class="n">idx_1</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">choice</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">rewards</span><span class="p">),</span> <span class="n">size</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">replace</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">sampling_paired_method</span> <span class="o">==</span> <span class="s2">&quot;max_min&quot;</span><span class="p">:</span>
+            <span class="n">idx_0</span><span class="p">,</span> <span class="n">idx_1</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">rewards</span><span class="p">),</span> <span class="n">np</span><span class="o">.</span><span class="n">argmin</span><span class="p">(</span><span class="n">rewards</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">sampling_paired_method</span> <span class="o">==</span> <span class="s2">&quot;max_max&quot;</span><span class="p">:</span>
+            <span class="n">sorted_indices</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argsort</span><span class="p">(</span><span class="n">rewards</span><span class="p">)</span>
+            <span class="n">idx_0</span><span class="p">,</span> <span class="n">idx_1</span> <span class="o">=</span> <span class="n">sorted_indices</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span> <span class="n">sorted_indices</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span>
+        <span class="k">elif</span> <span class="n">sampling_paired_method</span> <span class="o">==</span> <span class="s2">&quot;max_random&quot;</span><span class="p">:</span>
+            <span class="n">idx_0</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">rewards</span><span class="p">)</span>
+            <span class="n">idx_1</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">choice</span><span class="p">([</span><span class="n">i</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">rewards</span><span class="p">))</span> <span class="k">if</span> <span class="n">i</span> <span class="o">!=</span> <span class="n">idx_0</span><span class="p">])</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unknown sampling method: </span><span class="si">{</span><span class="n">sampling_paired_method</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        
+        <span class="n">chosen_idx</span><span class="p">,</span> <span class="n">rejected_idx</span> <span class="o">=</span> <span class="p">(</span><span class="n">idx_0</span><span class="p">,</span> <span class="n">idx_1</span><span class="p">)</span> <span class="k">if</span> <span class="n">rewards</span><span class="p">[</span><span class="n">idx_0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="n">rewards</span><span class="p">[</span><span class="n">idx_1</span><span class="p">]</span> <span class="k">else</span> <span class="p">(</span><span class="n">idx_1</span><span class="p">,</span> <span class="n">idx_0</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">chosen_idx</span><span class="p">,</span> <span class="n">rejected_idx</span></div>
+
+
+
+<div class="viewcode-block" id="DPOv2Aligner._sampling_paired_idx_from_rewards_fast">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._sampling_paired_idx_from_rewards_fast">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_sampling_paired_idx_from_rewards_fast</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">rewards</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span>
+        <span class="n">sampling_paired_method</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s2">&quot;random&quot;</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]:</span>
+        <span class="n">idx_0</span><span class="p">,</span> <span class="n">idx_1</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span>
+        
+        <span class="k">if</span> <span class="n">sampling_paired_method</span> <span class="o">==</span> <span class="s2">&quot;random&quot;</span><span class="p">:</span>
+            <span class="n">idx_0</span><span class="p">,</span> <span class="n">idx_1</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span>
+        <span class="k">elif</span> <span class="n">sampling_paired_method</span> <span class="o">==</span> <span class="s2">&quot;max_min&quot;</span><span class="p">:</span>
+            <span class="n">idx_0</span><span class="p">,</span> <span class="n">idx_1</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">rewards</span><span class="p">),</span> <span class="n">np</span><span class="o">.</span><span class="n">argmin</span><span class="p">(</span><span class="n">rewards</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">sampling_paired_method</span> <span class="o">==</span> <span class="s2">&quot;max_max&quot;</span><span class="p">:</span>
+            <span class="n">sorted_indices</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argsort</span><span class="p">(</span><span class="n">rewards</span><span class="p">)</span>
+            <span class="n">idx_0</span><span class="p">,</span> <span class="n">idx_1</span> <span class="o">=</span> <span class="n">sorted_indices</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span> <span class="n">sorted_indices</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span>
+        <span class="k">elif</span> <span class="n">sampling_paired_method</span> <span class="o">==</span> <span class="s2">&quot;max_random&quot;</span><span class="p">:</span>
+            <span class="n">idx_0</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">rewards</span><span class="p">)</span>
+            <span class="n">idx_1</span> <span class="o">=</span> <span class="mi">0</span> <span class="k">if</span> <span class="n">idx_0</span> <span class="o">!=</span> <span class="mi">0</span> <span class="k">else</span> <span class="mi">1</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unknown sampling method: </span><span class="si">{</span><span class="n">sampling_paired_method</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        
+        <span class="n">chosen_idx</span><span class="p">,</span> <span class="n">rejected_idx</span> <span class="o">=</span> <span class="p">(</span><span class="n">idx_0</span><span class="p">,</span> <span class="n">idx_1</span><span class="p">)</span> <span class="k">if</span> <span class="n">rewards</span><span class="p">[</span><span class="n">idx_0</span><span class="p">]</span> <span class="o">&gt;</span> <span class="n">rewards</span><span class="p">[</span><span class="n">idx_1</span><span class="p">]</span> <span class="k">else</span> <span class="p">(</span><span class="n">idx_1</span><span class="p">,</span> <span class="n">idx_0</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">chosen_idx</span><span class="p">,</span> <span class="n">rejected_idx</span></div>
+</div>
+
+        
+
+<div class="viewcode-block" id="MemorySafeDPOv2Aligner">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">MemorySafeDPOv2Aligner</span><span class="p">:</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span>
+        <span class="n">aligner_args</span><span class="p">:</span> <span class="n">DPOv2AlignerArguments</span><span class="p">,</span>
+        <span class="n">ref_model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+    <span class="p">):</span>
+<div class="viewcode-block" id="MemorySafeDPOv2Aligner.model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span> <span class="o">=</span> <span class="n">model_args</span></div>
+
+<div class="viewcode-block" id="MemorySafeDPOv2Aligner.ref_model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.ref_model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ref_model_args</span> <span class="o">=</span> <span class="n">ReferenceModelArguments</span><span class="p">(</span><span class="o">**</span><span class="n">add_dataclass_attr_prefix</span><span class="p">(</span><span class="n">ref_model_args</span><span class="p">,</span> <span class="s1">&#39;reference_&#39;</span><span class="p">))</span></div>
+
+<div class="viewcode-block" id="MemorySafeDPOv2Aligner.data_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.data_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span> <span class="o">=</span> <span class="n">data_args</span></div>
+
+<div class="viewcode-block" id="MemorySafeDPOv2Aligner.aligner_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.aligner_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span> <span class="o">=</span> <span class="n">aligner_args</span></div>
+
+<div class="viewcode-block" id="MemorySafeDPOv2Aligner.aligner_file_path">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.aligner_file_path">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">aligner_file_path</span> <span class="o">=</span> <span class="n">pkg_resources</span><span class="o">.</span><span class="n">files</span><span class="p">(</span><span class="s2">&quot;lmflow.pipeline.utils&quot;</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;memory_safe_dpov2_align.py&quot;</span></div>
+
+
+<div class="viewcode-block" id="MemorySafeDPOv2Aligner.align">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.align">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">align</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="n">aligner_args</span> <span class="o">=</span> <span class="n">make_shell_args_from_dataclass</span><span class="p">(</span>
+            <span class="n">dataclass_objects</span><span class="o">=</span><span class="p">[</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="p">,</span> 
+                <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="p">,</span> 
+                <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="p">,</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">ref_model_args</span>
+            <span class="p">],</span>
+            <span class="nb">format</span><span class="o">=</span><span class="s2">&quot;shell&quot;</span><span class="p">,</span>
+            <span class="n">ignored_args_list</span><span class="o">=</span><span class="p">[</span><span class="s1">&#39;accelerator_config&#39;</span><span class="p">,</span> <span class="s1">&#39;fsdp_config&#39;</span><span class="p">,</span> <span class="s1">&#39;_n_gpu&#39;</span><span class="p">],</span>
+        <span class="p">)</span>
+        <span class="n">cmd</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;accelerate launch --config_file </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">accelerate_config_file</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="o">+</span> <span class="s2">&quot; &quot;</span> 
+            <span class="o">+</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_file_path</span><span class="p">)</span> 
+            <span class="o">+</span> <span class="s2">&quot; &quot;</span> 
+            <span class="o">+</span> <span class="n">aligner_args</span>
+        <span class="p">)</span>
+        <span class="n">current_env</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
+        <span class="k">for</span> <span class="n">var</span> <span class="ow">in</span> <span class="n">MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE</span><span class="p">:</span>
+            <span class="n">current_env</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="n">var</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+        
+        <span class="n">cli_res</span> <span class="o">=</span> <span class="n">subprocess</span><span class="o">.</span><span class="n">run</span><span class="p">(</span>
+            <span class="n">args</span><span class="o">=</span><span class="n">cmd</span><span class="p">,</span>
+            <span class="n">stdout</span><span class="o">=</span><span class="n">sys</span><span class="o">.</span><span class="n">stdout</span><span class="p">,</span>
+            <span class="n">stderr</span><span class="o">=</span><span class="n">sys</span><span class="o">.</span><span class="n">stdout</span><span class="p">,</span>
+            <span class="n">shell</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+            <span class="n">preexec_fn</span><span class="o">=</span><span class="n">os</span><span class="o">.</span><span class="n">setsid</span><span class="p">,</span>
+            <span class="n">env</span><span class="o">=</span><span class="n">current_env</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;MemorySafeDPOv2Aligner subprocess run finished, info at finish: </span><span class="si">{</span><span class="n">cli_res</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        
+        <span class="k">if</span> <span class="n">cli_res</span><span class="o">.</span><span class="n">returncode</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="n">cli_res</span><span class="o">.</span><span class="n">stderr</span><span class="p">)</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Error during MemorySafeDPOv2Aligner: </span><span class="si">{</span><span class="n">cli_res</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/evaluator.html b/_modules/lmflow/pipeline/evaluator.html
new file mode 100644
index 000000000..8479d3738
--- /dev/null
+++ b/_modules/lmflow/pipeline/evaluator.html
@@ -0,0 +1,979 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.evaluator &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/evaluator';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.evaluator</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.evaluator</h1><div class="highlight"><pre>
+<span></span><span class="sd">&quot;&quot;&quot;The Evaluator class simplifies the process of running evaluation on a language model provided by a HFDecoderModel instance imported from the lmflow package. The class constructor takes three dictionaries as arguments: model_args containing arguments related to the language model, data_args containing arguments related to the data used for evaluation, and evaluator_args containing other arguments for the evaluation process.</span>
+
+<span class="sd">The class has two methods: create_dataloader() that loads the data from the test file, creates a data loader, and returns it with the size of the data, and evaluate(model) that generates output text given input text. It uses the create_dataloader() method to load the data, iterates over the data in mini-batches, and encodes the input text with the encode() method of the HFDecoderModel class. Then, it generates output text using the evaluate() method of the HFDecoderModel class, decodes the generated output text using the decode() method of the HFDecoderModel class, and writes the output to a file in the output directory. The method also logs some information to the console and Weights and Biases if the use_wandb argument is True.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">wandb</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">deepspeed</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">datetime</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">json</span>
+<span class="c1"># TODO: remove later</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">accelerate</span><span class="w"> </span><span class="kn">import</span> <span class="n">Accelerator</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">AutoConfig</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.distributed</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">dist</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets.dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.base_pipeline</span><span class="w"> </span><span class="kn">import</span> <span class="n">BasePipeline</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_decoder_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFDecoderModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.data_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">set_random_seed</span><span class="p">,</span> <span class="n">batchlize</span><span class="p">,</span> <span class="n">answer_extraction</span>
+<span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="p">[</span><span class="s2">&quot;TOKENIZERS_PARALLELISM&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;false&quot;</span>  <span class="c1"># To avoid warnings about parallelism in tokenizers</span>
+
+<div class="viewcode-block" id="Evaluator">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Evaluator</span><span class="p">(</span><span class="n">BasePipeline</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Initializes the `Evaluator` class with given arguments.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+<span class="sd">    model_args : ModelArguments object.</span>
+<span class="sd">        Contains the arguments required to load the model.</span>
+<span class="sd">    </span>
+<span class="sd">    data_args : DatasetArguments object.</span>
+<span class="sd">        Contains the arguments required to load the dataset.</span>
+
+<span class="sd">    evaluator_args : EvaluatorArguments object.</span>
+<span class="sd">        Contains the arguments required to perform evaluation.</span>
+
+
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">evaluator_args</span><span class="p">):</span>
+    <span class="c1"># our method</span>
+<div class="viewcode-block" id="Evaluator.data_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.data_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span> <span class="o">=</span> <span class="n">data_args</span></div>
+
+<div class="viewcode-block" id="Evaluator.evaluator_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.evaluator_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span> <span class="o">=</span> <span class="n">evaluator_args</span></div>
+
+<div class="viewcode-block" id="Evaluator.model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span> <span class="o">=</span> <span class="n">model_args</span></div>
+
+
+        <span class="c1"># logger</span>
+        <span class="k">if</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">use_wandb</span> <span class="o">==</span> <span class="kc">True</span><span class="p">):</span>
+            <span class="n">wandb</span><span class="o">.</span><span class="n">init</span><span class="p">(</span><span class="n">project</span><span class="o">=</span><span class="s2">&quot;lmflow_evaluation&quot;</span><span class="p">)</span>
+        <span class="c1"># random seed</span>
+        <span class="n">set_random_seed</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">random_seed</span><span class="p">)</span>
+<div class="viewcode-block" id="Evaluator.local_rank">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.local_rank">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">getenv</span><span class="p">(</span><span class="s2">&quot;LOCAL_RANK&quot;</span><span class="p">,</span> <span class="s2">&quot;0&quot;</span><span class="p">))</span></div>
+
+<div class="viewcode-block" id="Evaluator.world_size">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.world_size">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">world_size</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">getenv</span><span class="p">(</span><span class="s2">&quot;WORLD_SIZE&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">))</span></div>
+
+        <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">set_device</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>  <span class="c1"># NOTE: cpu-only machine will have error</span>
+
+        <span class="k">if</span> <span class="n">evaluator_args</span><span class="o">.</span><span class="n">use_accelerator_for_evaluator</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span> <span class="o">=</span> <span class="n">Accelerator</span><span class="p">()</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="o">.</span><span class="n">wait_for_everyone</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">deepspeed</span><span class="o">.</span><span class="n">init_distributed</span><span class="p">()</span>
+
+<div class="viewcode-block" id="Evaluator.config">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.config">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">config</span> <span class="o">=</span> <span class="n">AutoConfig</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">)</span></div>
+
+        <span class="k">try</span><span class="p">:</span> 
+            <span class="bp">self</span><span class="o">.</span><span class="n">model_hidden_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">hidden_size</span>
+        <span class="k">except</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Error in setting hidden size, use the default size 1024&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model_hidden_size</span> <span class="o">=</span> <span class="mi">1024</span> <span class="c1"># gpt2 seems do not have hidden_size in config</span>
+
+        <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;model_hidden_size = </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">model_hidden_size</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="c1"># batch size has to be divisible by world_size, but can be bigger than world_size</span>
+        <span class="n">train_batch_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">world_size</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">minibatch_size</span> <span class="o">=</span> <span class="n">train_batch_size</span>
+<div class="viewcode-block" id="Evaluator.block_size">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.block_size">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span> <span class="o">=</span> <span class="n">evaluator_args</span><span class="o">.</span><span class="n">evaluate_block_size</span></div>
+
+        <span class="c1"># dataloader, data_size = create_dataloader(args)    # load dataset</span>
+
+
+<div class="viewcode-block" id="Evaluator.create_dataloader">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.create_dataloader">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">create_dataloader</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">):</span>
+        <span class="n">data_dict</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">to_dict</span><span class="p">()</span>
+        <span class="n">inputs</span> <span class="o">=</span> <span class="p">[</span> <span class="n">instance</span><span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">]</span> <span class="k">for</span> <span class="n">instance</span> <span class="ow">in</span> <span class="n">data_dict</span><span class="p">[</span><span class="s2">&quot;instances&quot;</span><span class="p">]</span> <span class="p">]</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="p">[</span> <span class="n">instance</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">]</span> <span class="k">for</span> <span class="n">instance</span> <span class="ow">in</span> <span class="n">data_dict</span><span class="p">[</span><span class="s2">&quot;instances&quot;</span><span class="p">]</span> <span class="p">]</span>
+        <span class="n">dataset_size</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">outputs</span><span class="p">)</span>
+        <span class="n">dataset_buf</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">idx</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">dataset_size</span><span class="p">):</span>
+            <span class="n">dataset_buf</span><span class="o">.</span><span class="n">append</span><span class="p">({</span>
+                <span class="s2">&quot;input&quot;</span><span class="p">:</span> <span class="n">inputs</span><span class="p">[</span><span class="n">idx</span><span class="p">],</span>
+                <span class="s2">&quot;output&quot;</span><span class="p">:</span> <span class="n">outputs</span><span class="p">[</span><span class="n">idx</span><span class="p">],</span>
+                <span class="s2">&quot;input_idx&quot;</span><span class="p">:</span> <span class="n">idx</span>
+            <span class="p">})</span>
+
+        <span class="n">dataloader</span> <span class="o">=</span> <span class="n">batchlize</span><span class="p">(</span>
+            <span class="n">dataset_buf</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">minibatch_size</span><span class="p">,</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">random_shuffle</span>
+        <span class="p">)</span>
+        <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Successfully create dataloader with size </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">dataloader</span><span class="p">)</span><span class="si">}</span><span class="s2">,batch_size </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">minibatch_size</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">dataloader</span><span class="p">,</span> <span class="n">dataset_size</span></div>
+
+
+
+    <span class="c1"># TODO: Split for better unittest</span>
+
+<div class="viewcode-block" id="Evaluator._match">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator._match">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_match</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">predicted_answer</span><span class="p">,</span> <span class="n">groundtruth</span><span class="p">,</span> <span class="n">answer_type</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="n">case_insensitive_types</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="s2">&quot;strategyqa&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;coin_flip&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;pubmedqa&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;binary_choice&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;medmcqa&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;usmle&quot;</span><span class="p">,</span>
+        <span class="p">]</span>
+        <span class="k">if</span> <span class="n">answer_type</span> <span class="ow">in</span> <span class="n">case_insensitive_types</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">predicted_answer</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span> <span class="o">==</span> <span class="n">groundtruth</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">predicted_answer</span> <span class="o">==</span> <span class="n">groundtruth</span>
+        <span class="k">return</span> <span class="kc">False</span></div>
+
+
+
+<div class="viewcode-block" id="Evaluator.evaluate">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.evaluate">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">evaluate</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">metric</span> <span class="o">=</span> <span class="s2">&quot;accuracy&quot;</span><span class="p">,</span>
+        <span class="n">verbose</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform Evaluation for a model</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        model : TunableModel object.</span>
+<span class="sd">            TunableModel to perform inference</span>
+
+<span class="sd">        dataset : Dataset object.</span>
+<span class="sd">            </span>
+
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">metric</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;acc&quot;</span><span class="p">,</span> <span class="s2">&quot;accuracy&quot;</span><span class="p">]:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">use_accelerator_for_evaluator</span><span class="p">:</span>
+                <span class="n">acc</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_evaluate_acc_with_accelerator</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">dataset</span><span class="p">,</span> <span class="n">verbose</span><span class="o">=</span><span class="n">verbose</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">acc</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_evaluate_acc_with_deepspeed</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">dataset</span><span class="p">,</span> <span class="n">verbose</span><span class="o">=</span><span class="n">verbose</span><span class="p">)</span>
+            <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Evaluating final accuracy: </span><span class="si">{</span><span class="n">acc</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">acc</span>
+        <span class="k">elif</span> <span class="n">metric</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;ppl&quot;</span><span class="p">,</span> <span class="s2">&quot;perplexity&quot;</span><span class="p">]:</span>
+            <span class="n">ppl</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_evaluate_ppl</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">dataset</span><span class="p">,</span> <span class="n">verbose</span><span class="o">=</span><span class="n">verbose</span><span class="p">)</span>
+            <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Evaluating final perplexity: </span><span class="si">{</span><span class="n">ppl</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">ppl</span>
+        <span class="k">elif</span> <span class="n">metric</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;nll&quot;</span><span class="p">,</span> <span class="s2">&quot;neg_log_likelihood&quot;</span><span class="p">]:</span>
+            <span class="n">nll</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_evaluate_nll</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">dataset</span><span class="p">,</span> <span class="n">verbose</span><span class="o">=</span><span class="n">verbose</span><span class="p">)</span>
+            <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Evaluating final negative log likelihood: </span><span class="si">{</span><span class="n">nll</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">nll</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;metric </span><span class="si">{</span><span class="n">metric</span><span class="si">}</span><span class="s2"> is not supported&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="Evaluator._evaluate_acc_with_accelerator">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator._evaluate_acc_with_accelerator">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_evaluate_acc_with_accelerator</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">dataset</span><span class="p">,</span> <span class="n">verbose</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
+        <span class="n">dataloader</span><span class="p">,</span> <span class="n">data_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">create_dataloader</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="o">.</span><span class="n">is_local_main_process</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">):</span>
+                <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">)</span>
+            <span class="n">output_writer</span> <span class="o">=</span> <span class="nb">open</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">output_dir</span><span class="si">}</span><span class="s2">/evaluation.json&quot;</span><span class="p">,</span> <span class="s2">&quot;w&quot;</span><span class="p">)</span>
+        
+        <span class="n">correct_number_list</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">batch_index</span><span class="p">,</span> <span class="n">batch</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">dataloader</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">batch_index</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">world_size</span> <span class="o">&gt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">max_eval_samples</span><span class="p">:</span> 
+                <span class="k">break</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="o">*</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span> <span class="o">&gt;=</span> <span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="p">):</span>
+                <span class="n">current_batch</span> <span class="o">=</span> <span class="n">batch</span><span class="p">[:</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span><span class="p">]</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">current_batch</span> <span class="o">=</span> <span class="n">batch</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="o">*</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span><span class="p">:(</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span><span class="o">*</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span><span class="p">]</span>
+            <span class="n">prompt_structure</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">prompt_structure</span>
+            <span class="nb">input</span> <span class="o">=</span> <span class="p">[</span><span class="n">prompt_structure</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">i</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">])</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">current_batch</span><span class="p">]</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="p">[</span><span class="n">i</span><span class="p">[</span><span class="s1">&#39;output&#39;</span><span class="p">]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">current_batch</span><span class="p">]</span>   
+
+            <span class="n">batch_input</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span><span class="p">,</span><span class="n">padding</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+            <span class="n">inputs</span> <span class="o">=</span> <span class="n">batch_input</span><span class="p">[</span><span class="s1">&#39;input_ids&#39;</span><span class="p">]</span>
+            <span class="n">mask</span> <span class="o">=</span> <span class="n">batch_input</span><span class="p">[</span><span class="s1">&#39;attention_mask&#39;</span><span class="p">]</span>
+            <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="o">.</span><span class="n">autocast</span><span class="p">():</span>
+                <span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">max_new_tokens</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">max_new_tokens</span><span class="p">,</span><span class="n">attention_mask</span><span class="o">=</span><span class="n">mask</span><span class="p">,</span><span class="n">temperature</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">temperature</span><span class="p">,</span> <span class="n">repetition_penalty</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">repetition_penalty</span><span class="p">,</span><span class="n">use_accelerator</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">use_accelerator_for_evaluator</span><span class="p">)</span>
+            <span class="n">text_out</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="n">skip_special_tokens</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="n">decoded_input</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">skip_special_tokens</span><span class="o">=</span><span class="kc">True</span><span class="p">,)</span>
+            <span class="n">prompt_length</span> <span class="o">=</span> <span class="p">[</span><span class="nb">len</span><span class="p">(</span><span class="n">i</span><span class="p">)</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">decoded_input</span><span class="p">]</span>
+            <span class="n">text_out</span> <span class="o">=</span> <span class="p">[</span><span class="n">text_out</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="n">prompt_length</span><span class="p">[</span><span class="n">i</span><span class="p">]:]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">text_out</span><span class="p">))]</span>
+            <span class="n">answer_type</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">answer_type</span>
+            <span class="n">pred_answer</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">text_out</span><span class="p">:</span>
+                <span class="n">pred_answer</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">answer_extraction</span><span class="p">(</span>
+                    <span class="n">i</span><span class="p">,</span>
+                    <span class="n">answer_type</span><span class="o">=</span><span class="n">answer_type</span><span class="p">,</span>
+                <span class="p">))</span>
+            <span class="k">if</span> <span class="n">verbose</span><span class="p">:</span>
+                <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;batch_index</span><span class="si">{</span><span class="n">batch_index</span><span class="si">}</span><span class="s2"> rank</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="si">}</span><span class="s2">:</span><span class="se">\n</span><span class="s2">   question=</span><span class="si">{</span><span class="nb">input</span><span class="si">}</span><span class="se">\n</span><span class="s2">  prediction=</span><span class="si">{</span><span class="n">text_out</span><span class="si">}</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;predicted answer: </span><span class="si">{</span><span class="n">pred_answer</span><span class="si">}</span><span class="s2"> </span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;groundtruth answer: </span><span class="si">{</span><span class="n">output</span><span class="si">}</span><span class="s2"> </span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span>  <span class="o">&gt;=</span> <span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="p">):</span>
+                <span class="n">correct_</span> <span class="o">=</span> <span class="mi">0</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">correct_</span> <span class="o">=</span> <span class="mi">0</span>
+                <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">pred_answer</span><span class="p">)):</span>
+                    <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_match</span><span class="p">(</span><span class="n">pred_answer</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="n">output</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="n">answer_type</span><span class="p">):</span>
+                        <span class="n">correct_</span> <span class="o">+=</span> <span class="mi">1</span>
+
+            <span class="c1"># collect accuracy from all gpus</span>
+            <span class="n">all_process</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="n">correct_</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>
+            <span class="n">all_process</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="o">.</span><span class="n">gather</span><span class="p">(</span><span class="n">all_process</span><span class="p">)</span>
+            <span class="n">correct_</span> <span class="o">=</span> <span class="nb">sum</span><span class="p">(</span><span class="n">all_process</span><span class="o">.</span><span class="n">tolist</span><span class="p">())</span>
+            <span class="n">correct_number_list</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">correct_</span><span class="p">)</span>
+
+            <span class="c1"># collect predictions from all gpus</span>
+            <span class="n">output_dict</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;question&quot;</span><span class="p">:</span> <span class="nb">input</span><span class="p">,</span>
+                        <span class="s2">&quot;prediction&quot;</span><span class="p">:</span> <span class="n">text_out</span><span class="p">,</span>
+                        <span class="s2">&quot;pred_answer&quot;</span><span class="p">:</span> <span class="n">pred_answer</span><span class="p">,</span>
+                        <span class="s2">&quot;answer&quot;</span><span class="p">:</span> <span class="n">output</span><span class="p">}</span>
+            <span class="k">if</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">world_size</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">):</span>
+                <span class="n">all_process_list</span> <span class="o">=</span> <span class="p">[{}]</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">world_size</span>
+                <span class="n">dist</span><span class="o">.</span><span class="n">gather_object</span><span class="p">(</span><span class="n">output_dict</span><span class="p">,</span> <span class="n">all_process_list</span> <span class="k">if</span> <span class="n">dist</span><span class="o">.</span><span class="n">get_rank</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span> <span class="n">dst</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">all_process_list</span> <span class="o">=</span> <span class="p">[</span><span class="n">output_dict</span><span class="p">]</span>
+            
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="o">.</span><span class="n">is_local_main_process</span><span class="p">:</span>
+                <span class="n">current_total</span> <span class="o">=</span> <span class="p">(</span><span class="n">batch_index</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">world_size</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span>
+                <span class="n">current_accuracy</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">correct_number_list</span><span class="p">)</span> <span class="o">/</span> <span class="n">current_total</span> <span class="k">if</span> <span class="nb">int</span><span class="p">(</span><span class="n">current_total</span><span class="p">)</span> <span class="o">&lt;</span> <span class="n">data_size</span> <span class="k">else</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">correct_number_list</span><span class="p">)</span> <span class="o">/</span> <span class="n">data_size</span>
+                <span class="nb">print</span><span class="p">(</span><span class="n">datetime</span><span class="o">.</span><span class="n">datetime</span><span class="o">.</span><span class="n">now</span><span class="p">()</span><span class="o">.</span><span class="n">strftime</span><span class="p">(</span><span class="s2">&quot;%Y-%m-</span><span class="si">%d</span><span class="s2"> %H:%M:%S&quot;</span><span class="p">),</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">int</span><span class="p">(</span><span class="n">current_total</span><span class="p">)</span><span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="nb">int</span><span class="p">(</span><span class="n">current_total</span><span class="p">)</span><span class="w"> </span><span class="o">&lt;</span><span class="w"> </span><span class="n">data_size</span><span class="w"> </span><span class="k">else</span><span class="w"> </span><span class="n">data_size</span><span class="si">}</span><span class="s2"> / </span><span class="si">{</span><span class="n">data_size</span><span class="si">}</span><span class="s2"> has been finished, # correct = </span><span class="si">{</span><span class="w"> </span><span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">correct_number_list</span><span class="p">)</span><span class="si">}</span><span class="s2">, current accuracy = </span><span class="si">{</span><span class="n">current_accuracy</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+                <span class="k">if</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">use_wandb</span> <span class="o">==</span> <span class="kc">True</span><span class="p">):</span>
+                    <span class="n">wandb</span><span class="o">.</span><span class="n">log</span><span class="p">({</span><span class="s2">&quot;Accuracy&quot;</span><span class="p">:</span> <span class="n">current_accuracy</span><span class="p">})</span>
+
+                <span class="k">for</span> <span class="n">index</span><span class="p">,</span> <span class="n">output</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">all_process_list</span><span class="p">):</span>
+                    <span class="n">output_json</span> <span class="o">=</span> <span class="n">json</span><span class="o">.</span><span class="n">dumps</span><span class="p">(</span><span class="n">output</span><span class="p">)</span>
+                    <span class="n">output_writer</span><span class="o">.</span><span class="n">write</span><span class="p">(</span><span class="n">output_json</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="o">.</span><span class="n">is_local_main_process</span><span class="p">:</span>
+            <span class="n">current_accuracy</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">correct_number_list</span><span class="p">)</span> <span class="o">/</span> <span class="n">data_size</span>
+            <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;# Correct = </span><span class="si">{</span><span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">correct_number_list</span><span class="p">)</span><span class="si">}</span><span class="s2">, # Total = </span><span class="si">{</span><span class="n">data_size</span><span class="si">}</span><span class="s2">, Final accuracy = &quot;</span><span class="p">,</span> <span class="n">current_accuracy</span><span class="p">)</span>
+            <span class="n">output_writer</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
+        <span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">correct_number_list</span><span class="p">)</span> <span class="o">/</span> <span class="n">data_size</span></div>
+
+
+<div class="viewcode-block" id="Evaluator._evaluate_acc_with_deepspeed">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator._evaluate_acc_with_deepspeed">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_evaluate_acc_with_deepspeed</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">dataset</span><span class="p">,</span> <span class="n">verbose</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
+        <span class="n">dataloader</span><span class="p">,</span> <span class="n">data_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">create_dataloader</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">dist</span><span class="o">.</span><span class="n">is_initialized</span><span class="p">()</span> <span class="ow">or</span> <span class="n">dist</span><span class="o">.</span><span class="n">get_rank</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">):</span>
+                <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">)</span>
+            <span class="n">output_writer</span> <span class="o">=</span> <span class="nb">open</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">output_dir</span><span class="si">}</span><span class="s2">/evaluation.json&quot;</span><span class="p">,</span> <span class="s2">&quot;w&quot;</span><span class="p">)</span>
+
+        <span class="n">correct_number_list</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">batch_index</span><span class="p">,</span> <span class="n">batch</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">dataloader</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">batch_index</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">world_size</span> <span class="o">&gt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">max_eval_samples</span><span class="p">:</span> 
+                <span class="k">break</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="o">*</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span> <span class="o">&gt;=</span> <span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="p">):</span>
+                <span class="n">current_batch</span> <span class="o">=</span> <span class="n">batch</span><span class="p">[:</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span><span class="p">]</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">current_batch</span> <span class="o">=</span> <span class="n">batch</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="o">*</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span><span class="p">:(</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span><span class="o">*</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span><span class="p">]</span>
+            <span class="n">prompt_structure</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">prompt_structure</span>
+            <span class="nb">input</span> <span class="o">=</span> <span class="p">[</span><span class="n">prompt_structure</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">i</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">])</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">current_batch</span><span class="p">]</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="p">[</span><span class="n">i</span><span class="p">[</span><span class="s1">&#39;output&#39;</span><span class="p">]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">current_batch</span><span class="p">]</span>   
+            <span class="n">input_idx</span> <span class="o">=</span> <span class="p">[</span><span class="n">i</span><span class="p">[</span><span class="s1">&#39;input_idx&#39;</span><span class="p">]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">current_batch</span><span class="p">]</span>
+            <span class="n">batch_input</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span><span class="p">,</span><span class="n">padding</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>
+            <span class="n">inputs</span> <span class="o">=</span> <span class="n">batch_input</span><span class="p">[</span><span class="s1">&#39;input_ids&#39;</span><span class="p">]</span>
+            <span class="n">mask</span> <span class="o">=</span> <span class="n">batch_input</span><span class="p">[</span><span class="s1">&#39;attention_mask&#39;</span><span class="p">]</span>
+            <span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">max_new_tokens</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">max_new_tokens</span><span class="p">,</span> <span class="n">attention_mask</span><span class="o">=</span><span class="n">mask</span><span class="p">,</span><span class="n">temperature</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">temperature</span><span class="p">,</span> <span class="n">repetition_penalty</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">repetition_penalty</span><span class="p">)</span>
+            <span class="n">text_out</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="n">skip_special_tokens</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="c1"># # only return the generation, trucating the input</span>
+            <span class="n">decoded_input</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">inputs</span><span class="p">,</span> <span class="n">skip_special_tokens</span><span class="o">=</span><span class="kc">True</span><span class="p">,)</span>
+            <span class="n">prompt_length</span> <span class="o">=</span> <span class="p">[</span><span class="nb">len</span><span class="p">(</span><span class="n">i</span><span class="p">)</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">decoded_input</span><span class="p">]</span>
+            <span class="n">text_out</span> <span class="o">=</span> <span class="p">[</span><span class="n">text_out</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="n">prompt_length</span><span class="p">[</span><span class="n">i</span><span class="p">]:]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">text_out</span><span class="p">))]</span>
+            <span class="n">answer_type</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">answer_type</span>
+            <span class="n">pred_answer</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">text_out</span><span class="p">:</span>
+                <span class="n">pred_answer</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">answer_extraction</span><span class="p">(</span>
+                    <span class="n">i</span><span class="p">,</span>
+                    <span class="n">answer_type</span><span class="o">=</span><span class="n">answer_type</span><span class="p">,</span>
+                <span class="p">))</span>
+            <span class="k">if</span> <span class="n">verbose</span><span class="p">:</span>
+                <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;batch_index</span><span class="si">{</span><span class="n">batch_index</span><span class="si">}</span><span class="s2"> rank</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="si">}</span><span class="s2">:</span><span class="se">\n</span><span class="s2">   question=</span><span class="si">{</span><span class="nb">input</span><span class="si">}</span><span class="se">\n</span><span class="s2">  prediction=</span><span class="si">{</span><span class="n">text_out</span><span class="si">}</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;predicted answer: </span><span class="si">{</span><span class="n">pred_answer</span><span class="si">}</span><span class="s2"> </span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;groundtruth answer: </span><span class="si">{</span><span class="n">output</span><span class="si">}</span><span class="s2"> </span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span>  <span class="o">&gt;=</span> <span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="p">):</span>
+                <span class="n">correct_</span> <span class="o">=</span> <span class="mi">0</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">correct_</span> <span class="o">=</span> <span class="mi">0</span>
+                <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">pred_answer</span><span class="p">)):</span>
+                    <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_match</span><span class="p">(</span><span class="n">pred_answer</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="n">output</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="n">answer_type</span><span class="p">):</span>
+                        <span class="n">correct_</span> <span class="o">+=</span> <span class="mi">1</span>
+
+            <span class="c1"># collect accuracy from all gpus</span>
+            <span class="n">all_process</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="n">correct_</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>
+            <span class="n">dist</span><span class="o">.</span><span class="n">all_reduce</span><span class="p">(</span><span class="n">all_process</span><span class="p">,</span> <span class="n">dist</span><span class="o">.</span><span class="n">ReduceOp</span><span class="o">.</span><span class="n">SUM</span><span class="p">,</span> <span class="n">async_op</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+            <span class="n">correct_</span> <span class="o">=</span> <span class="n">all_process</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span>
+            <span class="n">correct_number_list</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">correct_</span><span class="p">)</span>
+
+            <span class="c1"># collect predictions from all gpus</span>
+            <span class="n">output_dict</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;question&quot;</span><span class="p">:</span> <span class="nb">input</span><span class="p">,</span>
+                        <span class="s2">&quot;prediction&quot;</span><span class="p">:</span> <span class="n">text_out</span><span class="p">,</span>
+                        <span class="s2">&quot;pred_answer&quot;</span><span class="p">:</span> <span class="n">pred_answer</span><span class="p">,</span>
+                        <span class="s2">&quot;answer&quot;</span><span class="p">:</span> <span class="n">output</span><span class="p">}</span>
+            <span class="n">all_process_list</span> <span class="o">=</span> <span class="p">[{}]</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">world_size</span>
+
+            <span class="n">dist</span><span class="o">.</span><span class="n">gather_object</span><span class="p">(</span><span class="n">output_dict</span><span class="p">,</span> <span class="n">all_process_list</span> <span class="k">if</span> <span class="n">dist</span><span class="o">.</span><span class="n">get_rank</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span> <span class="n">dst</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">dist</span><span class="o">.</span><span class="n">is_initialized</span><span class="p">()</span> <span class="ow">or</span> <span class="n">dist</span><span class="o">.</span><span class="n">get_rank</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">current_total</span> <span class="o">=</span> <span class="p">(</span><span class="n">batch_index</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">world_size</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span>
+                <span class="n">current_accuracy</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">correct_number_list</span><span class="p">)</span> <span class="o">/</span> <span class="n">current_total</span> <span class="k">if</span> <span class="nb">int</span><span class="p">(</span><span class="n">current_total</span><span class="p">)</span> <span class="o">&lt;</span> <span class="n">data_size</span> <span class="k">else</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">correct_number_list</span><span class="p">)</span> <span class="o">/</span> <span class="n">data_size</span>
+                <span class="nb">print</span><span class="p">(</span><span class="n">datetime</span><span class="o">.</span><span class="n">datetime</span><span class="o">.</span><span class="n">now</span><span class="p">()</span><span class="o">.</span><span class="n">strftime</span><span class="p">(</span><span class="s2">&quot;%Y-%m-</span><span class="si">%d</span><span class="s2"> %H:%M:%S&quot;</span><span class="p">),</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">int</span><span class="p">(</span><span class="n">current_total</span><span class="p">)</span><span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="nb">int</span><span class="p">(</span><span class="n">current_total</span><span class="p">)</span><span class="w"> </span><span class="o">&lt;</span><span class="w"> </span><span class="n">data_size</span><span class="w"> </span><span class="k">else</span><span class="w"> </span><span class="n">data_size</span><span class="si">}</span><span class="s2"> / </span><span class="si">{</span><span class="n">data_size</span><span class="si">}</span><span class="s2"> has been finished, # correct = </span><span class="si">{</span><span class="w"> </span><span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">correct_number_list</span><span class="p">)</span><span class="si">}</span><span class="s2">, current accuracy = </span><span class="si">{</span><span class="n">current_accuracy</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+                <span class="k">if</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">use_wandb</span> <span class="o">==</span> <span class="kc">True</span><span class="p">):</span>
+                    <span class="n">wandb</span><span class="o">.</span><span class="n">log</span><span class="p">({</span><span class="s2">&quot;Accuracy&quot;</span><span class="p">:</span> <span class="n">current_accuracy</span><span class="p">})</span>
+
+                <span class="k">for</span> <span class="n">index</span><span class="p">,</span> <span class="n">output</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">all_process_list</span><span class="p">):</span>
+                    <span class="n">output_json</span> <span class="o">=</span> <span class="n">json</span><span class="o">.</span><span class="n">dumps</span><span class="p">(</span><span class="n">output</span><span class="p">)</span>
+                    <span class="n">output_writer</span><span class="o">.</span><span class="n">write</span><span class="p">(</span><span class="n">output_json</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">dist</span><span class="o">.</span><span class="n">is_initialized</span><span class="p">()</span> <span class="ow">or</span> <span class="n">dist</span><span class="o">.</span><span class="n">get_rank</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">current_accuracy</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">correct_number_list</span><span class="p">)</span> <span class="o">/</span> <span class="n">data_size</span>
+            <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;# Correct = </span><span class="si">{</span><span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">correct_number_list</span><span class="p">)</span><span class="si">}</span><span class="s2">, # Total = </span><span class="si">{</span><span class="n">data_size</span><span class="si">}</span><span class="s2">, Final accuracy = &quot;</span><span class="p">,</span> <span class="n">current_accuracy</span><span class="p">)</span>
+            <span class="n">output_writer</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
+        <span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">correct_number_list</span><span class="p">)</span> <span class="o">/</span> <span class="n">data_size</span></div>
+
+
+<div class="viewcode-block" id="Evaluator._evaluate_ppl">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator._evaluate_ppl">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_evaluate_ppl</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span> <span class="n">verbose</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
+        <span class="n">data_dict</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">to_dict</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">data_dict</span><span class="p">[</span><span class="s1">&#39;type&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;text2text&#39;</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">&quot;ppl evaluation is currently not supported for text2text dataset, please use text_only dataset.&quot;</span><span class="p">)</span>
+        <span class="n">texts</span> <span class="o">=</span> <span class="p">[</span> <span class="n">instance</span><span class="p">[</span><span class="s2">&quot;text&quot;</span><span class="p">]</span> <span class="k">for</span> <span class="n">instance</span> <span class="ow">in</span> <span class="n">data_dict</span><span class="p">[</span><span class="s2">&quot;instances&quot;</span><span class="p">]</span> <span class="p">]</span>
+        <span class="n">encodings</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">get_tokenizer</span><span class="p">()(</span><span class="s2">&quot;</span><span class="se">\n\n</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">texts</span><span class="p">),</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span><span class="p">)</span>
+        <span class="c1"># Define some constant</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">truncate_to_model_max_length</span><span class="p">:</span>
+            <span class="k">try</span><span class="p">:</span>
+                <span class="n">max_length</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">n_positions</span><span class="p">,</span> <span class="n">model</span><span class="o">.</span><span class="n">get_max_length</span><span class="p">())</span>
+            <span class="k">except</span><span class="p">:</span>
+                <span class="n">max_length</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="mi">1024</span><span class="p">,</span> <span class="n">model</span><span class="o">.</span><span class="n">get_max_length</span><span class="p">())</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">max_length</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span>
+        
+        <span class="k">if</span> <span class="n">verbose</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The maximum sequence length : </span><span class="si">{</span><span class="n">max_length</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">seq_len</span> <span class="o">=</span> <span class="n">encodings</span><span class="o">.</span><span class="n">input_ids</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">nlls</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">prev_end_loc</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">for</span> <span class="n">begin_loc</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">seq_len</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="p">):</span>
+            <span class="n">end_loc</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">begin_loc</span> <span class="o">+</span> <span class="n">max_length</span><span class="p">,</span> <span class="n">seq_len</span><span class="p">)</span>
+            <span class="n">trg_len</span> <span class="o">=</span> <span class="n">end_loc</span> <span class="o">-</span> <span class="n">prev_end_loc</span>  <span class="c1"># may be different from block_size on last loop</span>
+            <span class="n">input_ids</span> <span class="o">=</span> <span class="n">encodings</span><span class="o">.</span><span class="n">input_ids</span><span class="p">[:,</span> <span class="n">begin_loc</span><span class="p">:</span><span class="n">end_loc</span><span class="p">]</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>
+            <span class="n">target_ids</span> <span class="o">=</span> <span class="n">input_ids</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+            <span class="n">target_ids</span><span class="p">[:,</span> <span class="p">:</span><span class="o">-</span><span class="n">trg_len</span><span class="p">]</span> <span class="o">=</span> <span class="o">-</span><span class="mi">100</span>
+
+            <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+                <span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()(</span><span class="n">input_ids</span><span class="p">,</span> <span class="n">labels</span><span class="o">=</span><span class="n">target_ids</span><span class="p">)</span>
+                <span class="c1"># loss is calculated using CrossEntropyLoss which averages over valid labels</span>
+                <span class="c1"># N.B. the model only calculates loss over trg_len - 1 labels, because it internally shifts the labels</span>
+                <span class="c1"># to the left by 1.</span>
+                <span class="n">neg_log_likelihood</span> <span class="o">=</span> <span class="n">outputs</span><span class="o">.</span><span class="n">loss</span>
+
+            <span class="n">nlls</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">neg_log_likelihood</span><span class="p">)</span>
+            <span class="n">prev_end_loc</span> <span class="o">=</span> <span class="n">end_loc</span>
+            <span class="k">if</span> <span class="n">verbose</span><span class="p">:</span>
+                <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Evaluating PPL: </span><span class="si">{</span><span class="nb">int</span><span class="p">(</span><span class="n">begin_loc</span><span class="o">/</span><span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="p">)</span><span class="w"> </span><span class="o">+</span><span class="w"> </span><span class="mi">1</span><span class="si">}</span><span class="s2"> / </span><span class="si">{</span><span class="nb">int</span><span class="p">(</span><span class="n">seq_len</span><span class="o">/</span><span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="p">)</span><span class="si">}</span><span class="s2"> Complete, current ppl : </span><span class="si">{</span><span class="n">torch</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">nlls</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">())</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">end_loc</span> <span class="o">==</span> <span class="n">seq_len</span><span class="p">:</span>
+                <span class="k">break</span>
+        <span class="n">ppl</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">nlls</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">())</span>
+        <span class="k">return</span> <span class="n">ppl</span></div>
+
+
+
+<div class="viewcode-block" id="Evaluator._evaluate_nll">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator._evaluate_nll">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_evaluate_nll</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">verbose</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Evaluates negative log likelihood of the model over a dataset.</span>
+
+<span class="sd">        NLL = -1/N sum_{i=1}^N sum_{j=1}^|w_i| ln(p(w_{i,j}|context_window)),</span>
+
+<span class="sd">        where N is the number of data samples, w_{i,j} is the j-th token in</span>
+<span class="sd">        i-th sample. Here &quot;context_window&quot; = p(w_{i,start}, w_{i,start+1}, ...,</span>
+<span class="sd">        p_{i,j-1} with start = max(0, j - window_length + 1). &quot;window_length&quot;</span>
+<span class="sd">        is normally the maximum length accepted by the model.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            A float which represents the negative log likelihood.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">data_dict</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">to_dict</span><span class="p">()</span>
+
+        <span class="c1"># Handles prompt structure</span>
+        <span class="k">if</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="o">==</span> <span class="s2">&quot;text2text&quot;</span><span class="p">:</span>
+            <span class="n">prompt</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluator_args</span><span class="o">.</span><span class="n">prompt_structure</span>
+            <span class="n">data_dict</span><span class="p">[</span><span class="s2">&quot;instances&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="p">[</span>
+                <span class="p">{</span>
+                    <span class="s2">&quot;input&quot;</span><span class="p">:</span> <span class="n">prompt</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">instance</span><span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">]),</span>
+                    <span class="s2">&quot;output&quot;</span><span class="p">:</span> <span class="n">instance</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">]</span>
+                <span class="p">}</span>
+                <span class="k">for</span> <span class="n">instance</span> <span class="ow">in</span> <span class="n">data_dict</span><span class="p">[</span><span class="s2">&quot;instances&quot;</span><span class="p">]</span>
+            <span class="p">]</span>
+
+        <span class="n">dataset</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">from_dict</span><span class="p">(</span><span class="n">data_dict</span><span class="p">)</span>
+        <span class="n">tokenized_dataset</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">tokenize</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">add_special_tokens</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="n">tokenized_dataset</span> <span class="o">=</span> <span class="n">tokenized_dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()</span>
+        <span class="n">encoding_list</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="p">{</span>
+                <span class="s2">&quot;input_ids&quot;</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="n">input_ids</span><span class="p">]),</span>
+                <span class="s2">&quot;labels&quot;</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">([</span><span class="n">labels</span><span class="p">]),</span>
+            <span class="p">}</span>
+            <span class="k">for</span> <span class="n">input_ids</span><span class="p">,</span> <span class="n">labels</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">tokenized_dataset</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">],</span>
+                                         <span class="n">tokenized_dataset</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">])</span>
+        <span class="p">]</span>
+
+        <span class="c1"># Gets context window length</span>
+        <span class="k">try</span><span class="p">:</span>
+            <span class="n">max_length</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">n_positions</span><span class="p">,</span>
+                             <span class="n">model</span><span class="o">.</span><span class="n">get_max_length</span><span class="p">())</span>
+        <span class="k">except</span><span class="p">:</span>
+            <span class="n">max_length</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="mi">1024</span><span class="p">,</span> <span class="n">model</span><span class="o">.</span><span class="n">get_max_length</span><span class="p">())</span>
+
+        <span class="n">nlls</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">full_nlls</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">num_samples</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">encoding_list</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">sample_idx</span><span class="p">,</span> <span class="n">encodings</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">encoding_list</span><span class="p">):</span>
+            <span class="n">seq_len</span> <span class="o">=</span> <span class="n">encodings</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+
+            <span class="n">prev_end_loc</span> <span class="o">=</span> <span class="mi">0</span>
+            <span class="k">for</span> <span class="n">begin_loc</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">seq_len</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">block_size</span><span class="p">):</span>
+                <span class="n">end_loc</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">begin_loc</span> <span class="o">+</span> <span class="n">max_length</span><span class="p">,</span> <span class="n">seq_len</span><span class="p">)</span>
+
+                <span class="c1"># may be different from block_size on last loop</span>
+                <span class="n">trg_len</span> <span class="o">=</span> <span class="n">end_loc</span> <span class="o">-</span> <span class="n">prev_end_loc</span>
+                <span class="n">input_ids</span> <span class="o">=</span> <span class="n">encodings</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][:,</span> <span class="n">begin_loc</span><span class="p">:</span><span class="n">end_loc</span><span class="p">]</span>
+                <span class="n">input_ids</span> <span class="o">=</span> <span class="n">input_ids</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>
+
+                <span class="n">labels</span> <span class="o">=</span> <span class="n">encodings</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][:,</span> <span class="n">begin_loc</span><span class="p">:</span><span class="n">end_loc</span><span class="p">]</span>
+                <span class="n">target_ids</span> <span class="o">=</span> <span class="n">labels</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+                <span class="n">full_target_ids</span> <span class="o">=</span> <span class="n">input_ids</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+
+                <span class="k">def</span><span class="w"> </span><span class="nf">get_nll</span><span class="p">(</span><span class="n">label_ids</span><span class="p">,</span> <span class="n">nll_list</span><span class="p">):</span>
+                    <span class="n">label_ids</span><span class="p">[:,</span> <span class="p">:</span><span class="o">-</span><span class="n">trg_len</span><span class="p">]</span> <span class="o">=</span> <span class="o">-</span><span class="mi">100</span>
+                    <span class="n">label_ids</span> <span class="o">=</span> <span class="n">label_ids</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>
+
+                    <span class="c1"># Valid labels are from 0 to `vocab_size`</span>
+                    <span class="n">num_valid_labels</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">count_nonzero</span><span class="p">(</span><span class="n">label_ids</span> <span class="o">&gt;=</span> <span class="mi">0</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">label_ids</span><span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">!=</span> <span class="o">-</span><span class="mi">100</span><span class="p">:</span>
+                        <span class="n">num_valid_labels</span> <span class="o">-=</span> <span class="mi">1</span>
+
+                    <span class="k">if</span> <span class="ow">not</span> <span class="n">torch</span><span class="o">.</span><span class="n">all</span><span class="p">(</span><span class="n">label_ids</span> <span class="o">==</span> <span class="o">-</span><span class="mi">100</span><span class="p">):</span>
+                        <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+                            <span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()(</span>
+                                <span class="n">input_ids</span><span class="p">,</span> <span class="n">labels</span><span class="o">=</span><span class="n">label_ids</span>
+                            <span class="p">)</span>
+                            <span class="c1"># loss is calculated using CrossEntropyLoss which</span>
+                            <span class="c1"># sums over valid labels N.B. the model only</span>
+                            <span class="c1"># calculates loss over trg_len - 1 labels, because</span>
+                            <span class="c1"># it internally shifts the labels to the left by 1.</span>
+                            <span class="n">neg_log_likelihood</span> <span class="o">=</span> <span class="n">outputs</span><span class="o">.</span><span class="n">loss</span> <span class="o">*</span> <span class="n">num_valid_labels</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">neg_log_likelihood</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros</span><span class="p">([])</span><span class="o">.</span><span class="n">to</span><span class="p">(</span>
+                            <span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span>
+                        <span class="p">)</span>
+
+                    <span class="n">nll_list</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">neg_log_likelihood</span><span class="p">)</span>
+
+                <span class="n">get_nll</span><span class="p">(</span><span class="n">target_ids</span><span class="p">,</span> <span class="n">nlls</span><span class="p">)</span>
+                <span class="n">get_nll</span><span class="p">(</span><span class="n">full_target_ids</span><span class="p">,</span> <span class="n">full_nlls</span><span class="p">)</span>
+
+                <span class="n">current_output_nll</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">nlls</span><span class="p">)</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span> <span class="o">/</span> <span class="p">(</span><span class="n">sample_idx</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span>
+                <span class="n">current_full_nll</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">full_nlls</span><span class="p">)</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span> <span class="o">/</span> <span class="p">(</span><span class="n">sample_idx</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span>
+
+                <span class="n">prev_end_loc</span> <span class="o">=</span> <span class="n">end_loc</span>
+                <span class="k">if</span> <span class="n">verbose</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="o">==</span> <span class="s2">&quot;text_only&quot;</span><span class="p">:</span>
+                        <span class="nb">print</span><span class="p">(</span>
+                            <span class="sa">f</span><span class="s2">&quot;Evaluating negative log likelihood:&quot;</span>
+                            <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">sample_idx</span><span class="w"> </span><span class="o">+</span><span class="w"> </span><span class="mi">1</span><span class="si">}</span><span class="s2"> / </span><span class="si">{</span><span class="n">num_samples</span><span class="si">}</span><span class="s2"> Complete,&quot;</span>
+                            <span class="sa">f</span><span class="s2">&quot; current nll: </span><span class="si">{</span><span class="n">current_full_nll</span><span class="si">}</span><span class="s2">&quot;</span>
+                        <span class="p">)</span>
+                    <span class="k">elif</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="o">==</span> <span class="s2">&quot;text2text&quot;</span><span class="p">:</span>
+                        <span class="nb">print</span><span class="p">(</span>
+                            <span class="sa">f</span><span class="s2">&quot;Evaluating negative log likelihood:&quot;</span>
+                            <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">sample_idx</span><span class="w"> </span><span class="o">+</span><span class="w"> </span><span class="mi">1</span><span class="si">}</span><span class="s2"> / </span><span class="si">{</span><span class="n">num_samples</span><span class="si">}</span><span class="s2"> Complete,&quot;</span>
+                            <span class="sa">f</span><span class="s2">&quot; current full nll / input nll / output nll:&quot;</span>
+                            <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">current_full_nll</span><span class="si">}</span><span class="s2"> /&quot;</span>
+                            <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">current_full_nll</span><span class="w"> </span><span class="o">-</span><span class="w"> </span><span class="n">current_output_nll</span><span class="si">}</span><span class="s2"> /&quot;</span>
+                            <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">current_output_nll</span><span class="si">}</span><span class="s2">&quot;</span>
+                        <span class="p">)</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                            <span class="s2">&quot;f{dataset.get_type()} typed datasets are not&quot;</span>
+                            <span class="s2">&quot; supported&quot;</span>
+                        <span class="p">)</span>
+
+                <span class="k">if</span> <span class="n">end_loc</span> <span class="o">==</span> <span class="n">seq_len</span><span class="p">:</span>
+                    <span class="k">break</span>
+
+        <span class="n">mean_nll</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span><span class="n">nlls</span><span class="p">)</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span> <span class="o">/</span> <span class="n">num_samples</span>
+        <span class="k">return</span> <span class="n">mean_nll</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/finetuner.html b/_modules/lmflow/pipeline/finetuner.html
new file mode 100644
index 000000000..b68195948
--- /dev/null
+++ b/_modules/lmflow/pipeline/finetuner.html
@@ -0,0 +1,1101 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.finetuner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/finetuner';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.finetuner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.finetuner</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;The Finetuner class simplifies the process of running finetuning process on a language model for a TunableModel instance with given dataset.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">copy</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Iterable</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">datasets</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">transformers</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">evaluate</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">itertools</span><span class="w"> </span><span class="kn">import</span> <span class="n">chain</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">Trainer</span><span class="p">,</span>
+    <span class="n">default_data_collator</span><span class="p">,</span>
+    <span class="n">set_seed</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">copy</span><span class="w"> </span><span class="kn">import</span> <span class="n">deepcopy</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">PreTrainedModel</span><span class="p">,</span> <span class="n">TrainingArguments</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.trainer_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">get_last_checkpoint</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.trainer_callback</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">TrainerCallback</span><span class="p">,</span>
+    <span class="n">TrainerControl</span><span class="p">,</span>
+    <span class="n">TrainerState</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.utils</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">is_sagemaker_mp_enabled</span><span class="p">,</span>
+    <span class="n">send_example_telemetry</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">lmflow.optim.optimizers</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">optim</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="n">OptimizerNames</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets.dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.base_tuner</span><span class="w"> </span><span class="kn">import</span> <span class="n">BaseTuner</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.utils.peft_trainer</span><span class="w"> </span><span class="kn">import</span> <span class="n">PeftTrainer</span><span class="p">,</span> <span class="n">PeftSavingCallback</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="Finetuner">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Finetuner</span><span class="p">(</span><span class="n">BaseTuner</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Initializes the `Finetuner` class with given arguments.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+<span class="sd">    model_args : ModelArguments object.</span>
+<span class="sd">        Contains the arguments required to load the model.</span>
+
+<span class="sd">    data_args : DatasetArguments object.</span>
+<span class="sd">        Contains the arguments required to load the dataset.</span>
+
+<span class="sd">    finetuner_args : FinetunerArguments object.</span>
+<span class="sd">        Contains the arguments required to perform finetuning.</span>
+
+<span class="sd">    args : Optional.</span>
+<span class="sd">        Positional arguments.</span>
+
+<span class="sd">    kwargs : Optional.</span>
+<span class="sd">        Keyword arguments.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">finetuner_args</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+
+<div class="viewcode-block" id="Finetuner.model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span> <span class="o">=</span> <span class="n">model_args</span></div>
+
+<div class="viewcode-block" id="Finetuner.data_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.data_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span> <span class="o">=</span> <span class="n">data_args</span></div>
+
+<div class="viewcode-block" id="Finetuner.finetuner_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.finetuner_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span> <span class="o">=</span> <span class="n">finetuner_args</span></div>
+
+
+        <span class="c1"># Sending telemetry. Tracking the example usage helps us better</span>
+        <span class="c1"># allocate resources to maintain them. The information sent is the one</span>
+        <span class="c1"># passed as arguments along with your Python/PyTorch versions.</span>
+        <span class="n">send_example_telemetry</span><span class="p">(</span><span class="s2">&quot;run_clm&quot;</span><span class="p">,</span> <span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">)</span>
+
+        <span class="c1"># Setup logging</span>
+        <span class="n">logging</span><span class="o">.</span><span class="n">basicConfig</span><span class="p">(</span>
+            <span class="nb">format</span><span class="o">=</span><span class="s2">&quot;</span><span class="si">%(asctime)s</span><span class="s2"> - </span><span class="si">%(levelname)s</span><span class="s2"> - </span><span class="si">%(name)s</span><span class="s2"> - </span><span class="si">%(message)s</span><span class="s2">&quot;</span><span class="p">,</span>
+            <span class="n">datefmt</span><span class="o">=</span><span class="s2">&quot;%m/</span><span class="si">%d</span><span class="s2">/%Y %H:%M:%S&quot;</span><span class="p">,</span>
+            <span class="n">handlers</span><span class="o">=</span><span class="p">[</span><span class="n">logging</span><span class="o">.</span><span class="n">StreamHandler</span><span class="p">(</span><span class="n">sys</span><span class="o">.</span><span class="n">stdout</span><span class="p">)],</span>
+        <span class="p">)</span>
+
+        <span class="n">log_level</span> <span class="o">=</span> <span class="n">finetuner_args</span><span class="o">.</span><span class="n">get_process_log_level</span><span class="p">()</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">setLevel</span><span class="p">(</span><span class="n">log_level</span><span class="p">)</span>
+        <span class="n">datasets</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">logging</span><span class="o">.</span><span class="n">set_verbosity</span><span class="p">(</span><span class="n">log_level</span><span class="p">)</span>
+        <span class="n">transformers</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">logging</span><span class="o">.</span><span class="n">set_verbosity</span><span class="p">(</span><span class="n">log_level</span><span class="p">)</span>
+        <span class="n">transformers</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">logging</span><span class="o">.</span><span class="n">enable_default_handler</span><span class="p">()</span>
+        <span class="n">transformers</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">logging</span><span class="o">.</span><span class="n">enable_explicit_format</span><span class="p">()</span>
+
+        <span class="c1"># Log on each process the small summary:</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;Process rank: </span><span class="si">{</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">local_rank</span><span class="si">}</span><span class="s2">,&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot; device: </span><span class="si">{</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="s2">,&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot; n_gpu: </span><span class="si">{</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">n_gpu</span><span class="si">}</span><span class="s2">,&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;distributed training: </span><span class="si">{</span><span class="nb">bool</span><span class="p">(</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">local_rank</span><span class="w"> </span><span class="o">!=</span><span class="w"> </span><span class="o">-</span><span class="mi">1</span><span class="p">)</span><span class="si">}</span><span class="s2">,&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot; 16-bits training: </span><span class="si">{</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">fp16</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Training/evaluation parameters </span><span class="si">{</span><span class="n">finetuner_args</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="c1"># Detecting last checkpoint.</span>
+        <span class="n">last_checkpoint</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">)</span> <span class="ow">and</span> <span class="n">finetuner_args</span><span class="o">.</span><span class="n">do_train</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">finetuner_args</span><span class="o">.</span><span class="n">overwrite_output_dir</span><span class="p">:</span>
+            <span class="n">last_checkpoint</span> <span class="o">=</span> <span class="n">get_last_checkpoint</span><span class="p">(</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">last_checkpoint</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">listdir</span><span class="p">(</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">))</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;Output directory (</span><span class="si">{</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">output_dir</span><span class="si">}</span><span class="s2">) already&quot;</span>
+                    <span class="s2">&quot; exists and is not empty. &quot;</span>
+                    <span class="s2">&quot;Use --overwrite_output_dir to overcome.&quot;</span>
+                <span class="p">)</span>
+            <span class="k">elif</span> <span class="n">last_checkpoint</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">finetuner_args</span><span class="o">.</span><span class="n">resume_from_checkpoint</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;Checkpoint detected, resuming training at&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">last_checkpoint</span><span class="si">}</span><span class="s2">. To avoid this behavior, change&quot;</span>
+                    <span class="s2">&quot; the `--output_dir` or add `--overwrite_output_dir` to&quot;</span>
+                    <span class="s2">&quot; train from scratch.&quot;</span>
+                <span class="p">)</span>
+<div class="viewcode-block" id="Finetuner.last_checkpoint">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.last_checkpoint">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">last_checkpoint</span> <span class="o">=</span> <span class="n">last_checkpoint</span></div>
+
+
+        <span class="c1"># Set seed before initializing model.</span>
+        <span class="n">set_seed</span><span class="p">(</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">seed</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="Finetuner.group_text">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.group_text">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">group_text</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tokenized_datasets</span><span class="p">,</span> <span class="n">model_max_length</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Groups texts together to form blocks of maximum length `model_max_length` and returns the processed data as</span>
+<span class="sd">        a dictionary.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">data_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span>
+        <span class="n">finetuner_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span>
+
+        <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">block_size</span> <span class="o">=</span> <span class="n">model_max_length</span>
+            <span class="k">if</span> <span class="n">block_size</span> <span class="o">&gt;</span> <span class="mi">1024</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="s2">&quot;The chosen tokenizer supports a `model_max_length` that is&quot;</span>
+                    <span class="s2">&quot; longer than the default `block_size` value&quot;</span>
+                    <span class="s2">&quot; of 1024. If you would like to use a longer `block_size`&quot;</span>
+                    <span class="s2">&quot; up to `tokenizer.model_max_length` you can override this &quot;</span>
+                    <span class="s2">&quot; default with `--block_size xxx`.&quot;</span>
+                <span class="p">)</span>
+                <span class="n">block_size</span> <span class="o">=</span> <span class="mi">1024</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span> <span class="o">&gt;</span> <span class="n">model_max_length</span><span class="p">:</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">truncate_to_model_max_length</span><span class="p">:</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;The block_size passed (</span><span class="si">{</span><span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span><span class="si">}</span><span class="s2">) is larger&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot; than the maximum length for the model&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot;(</span><span class="si">{</span><span class="n">model_max_length</span><span class="si">}</span><span class="s2">).&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot; Using block_size=</span><span class="si">{</span><span class="n">model_max_length</span><span class="si">}</span><span class="s2">.&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot;If you would like to use a longer &#39;block_size&#39; that is&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot; longer than the maximum length supported by the model,&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot; you can override this behavior with&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot;default with `--truncate_to_model_max_length False`.&quot;</span>
+                    <span class="p">)</span>
+                    <span class="n">block_size</span> <span class="o">=</span> <span class="n">model_max_length</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;The block_size passed (</span><span class="si">{</span><span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span><span class="si">}</span><span class="s2">) is larger&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot;than the maximum length for the model&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot;(</span><span class="si">{</span><span class="n">model_max_length</span><span class="si">}</span><span class="s2">).&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot;Using block_size=</span><span class="si">{</span><span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+                    <span class="n">block_size</span> <span class="o">=</span> <span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">block_size</span> <span class="o">=</span> <span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span>
+        <span class="c1"># Main data processing function that will concatenate all texts from</span>
+        <span class="c1"># our dataset and generate chunks of block_size.</span>
+        <span class="k">def</span><span class="w"> </span><span class="nf">group_texts</span><span class="p">(</span><span class="n">examples</span><span class="p">):</span>
+            <span class="c1"># Concatenate all texts.</span>
+            <span class="n">concatenated_examples</span> <span class="o">=</span> <span class="p">{</span><span class="n">k</span><span class="p">:</span> <span class="nb">list</span><span class="p">(</span><span class="n">chain</span><span class="p">(</span><span class="o">*</span><span class="n">examples</span><span class="p">[</span><span class="n">k</span><span class="p">]))</span> <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">examples</span><span class="o">.</span><span class="n">keys</span><span class="p">()}</span>
+            <span class="n">total_length</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">concatenated_examples</span><span class="p">[</span><span class="nb">list</span><span class="p">(</span><span class="n">examples</span><span class="o">.</span><span class="n">keys</span><span class="p">())[</span><span class="mi">0</span><span class="p">]])</span>
+            <span class="c1"># We drop the small remainder, we could add padding if the model</span>
+            <span class="c1"># supported it instead of this drop, you can customize this part to</span>
+            <span class="c1"># your needs.</span>
+            <span class="n">total_length</span> <span class="o">=</span> <span class="p">(</span><span class="n">total_length</span> <span class="o">//</span> <span class="n">block_size</span><span class="p">)</span> <span class="o">*</span> <span class="n">block_size</span>
+            <span class="c1"># Split by chunks of max_len.</span>
+            <span class="n">result</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="n">k</span><span class="p">:</span> <span class="p">[</span><span class="n">t</span><span class="p">[</span><span class="n">i</span> <span class="p">:</span> <span class="n">i</span> <span class="o">+</span> <span class="n">block_size</span><span class="p">]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">total_length</span><span class="p">,</span> <span class="n">block_size</span><span class="p">)]</span>
+                <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">concatenated_examples</span><span class="o">.</span><span class="n">items</span><span class="p">()</span>
+            <span class="p">}</span>
+            <span class="k">return</span> <span class="n">result</span>
+
+        <span class="c1"># Note that with `batched=True`, this map processes 1,000 texts</span>
+        <span class="c1"># together, so group_texts throws away a remainder for each of those</span>
+        <span class="c1"># groups of 1,000 texts. You can adjust that batch_size here but a</span>
+        <span class="c1"># higher value might be slower to preprocess.</span>
+        <span class="c1">#</span>
+        <span class="c1"># To speed up this part, we use multiprocessing. See the documentation</span>
+        <span class="c1"># of the map method for more information:</span>
+        <span class="c1"># https://huggingface.co/docs/datasets/package_reference/main_classes.html#datasets.Dataset.map</span>
+        <span class="k">with</span> <span class="n">finetuner_args</span><span class="o">.</span><span class="n">main_process_first</span><span class="p">(</span><span class="n">desc</span><span class="o">=</span><span class="s2">&quot;grouping texts together&quot;</span><span class="p">):</span>
+            <span class="n">group_batch_size</span> <span class="o">=</span> <span class="n">data_args</span><span class="o">.</span><span class="n">group_texts_batch_size</span>
+            <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+                <span class="n">group_batch_size</span> <span class="o">=</span> <span class="mi">1</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">data_args</span><span class="o">.</span><span class="n">streaming</span><span class="p">:</span>
+                <span class="n">lm_datasets</span> <span class="o">=</span> <span class="n">tokenized_datasets</span><span class="o">.</span><span class="n">map</span><span class="p">(</span>
+                    <span class="n">group_texts</span><span class="p">,</span>
+                    <span class="n">batched</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                    <span class="n">batch_size</span><span class="o">=</span><span class="n">group_batch_size</span><span class="p">,</span>
+                    <span class="n">num_proc</span><span class="o">=</span><span class="n">data_args</span><span class="o">.</span><span class="n">preprocessing_num_workers</span><span class="p">,</span>
+                    <span class="n">load_from_cache_file</span><span class="o">=</span><span class="ow">not</span> <span class="n">data_args</span><span class="o">.</span><span class="n">overwrite_cache</span><span class="p">,</span>
+                    <span class="n">desc</span><span class="o">=</span><span class="sa">f</span><span class="s2">&quot;Grouping texts in chunks of </span><span class="si">{</span><span class="n">block_size</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">lm_datasets</span> <span class="o">=</span> <span class="n">tokenized_datasets</span><span class="o">.</span><span class="n">map</span><span class="p">(</span>
+                    <span class="n">group_texts</span><span class="p">,</span>
+                    <span class="n">batched</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                    <span class="n">batch_size</span><span class="o">=</span><span class="n">group_batch_size</span><span class="p">,</span>
+                <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">lm_datasets</span></div>
+
+
+<div class="viewcode-block" id="Finetuner.create_customized_optimizer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.create_customized_optimizer">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">create_customized_optimizer</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">base_trainer_class</span><span class="p">,</span> <span class="n">model_args</span><span class="p">):</span>
+        <span class="k">class</span><span class="w"> </span><span class="nc">CustomizedOptimTrainer</span><span class="p">(</span><span class="n">base_trainer_class</span><span class="p">):</span>
+
+            <span class="nd">@staticmethod</span>
+            <span class="k">def</span><span class="w"> </span><span class="nf">get_optimizer_cls_and_kwargs</span><span class="p">(</span>
+                <span class="n">args</span><span class="p">:</span> <span class="n">TrainingArguments</span><span class="p">,</span>
+                <span class="n">model</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">PreTrainedModel</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+                <span class="c1"># parse args.optim_args</span>
+                <span class="n">optim_args</span> <span class="o">=</span> <span class="p">{}</span>
+                <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim_args</span><span class="p">:</span>
+                    <span class="k">for</span> <span class="n">mapping</span> <span class="ow">in</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim_args</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot; &quot;</span><span class="p">,</span> <span class="s2">&quot;&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;,&quot;</span><span class="p">):</span>
+                        <span class="n">key</span><span class="p">,</span> <span class="n">value</span> <span class="o">=</span> <span class="n">mapping</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;=&quot;</span><span class="p">)</span>
+                        <span class="n">optim_args</span><span class="p">[</span><span class="n">key</span><span class="p">]</span> <span class="o">=</span> <span class="n">value</span>
+
+                <span class="n">optimizer_kwargs</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;lr&quot;</span><span class="p">:</span> <span class="n">args</span><span class="o">.</span><span class="n">learning_rate</span><span class="p">}</span>
+
+                <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">DUMMY</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">Dummy</span>
+                    <span class="n">dummy_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_dummy_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_dummy_beta2</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">dummy_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADABELIEF</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">AdaBelief</span>
+                    <span class="n">adabelief_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta2</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">)</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adabelief_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADABOUND</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">AdaBound</span>
+                    <span class="n">adabound_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta2</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">)</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adabound_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">LARS</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">LARS</span>
+                    <span class="n">lars_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;momentum&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_momentum</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">lars_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">LAMB</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">Lamb</span>
+                    <span class="n">lamb_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta2</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">lamb_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAMAX</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">Adamax</span>
+                    <span class="n">adamax_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta2</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adamax_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">NADAM</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">NAdam</span>
+                    <span class="n">nadam_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta2</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">nadam_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">RADAM</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">RAdam</span>
+                    <span class="n">radam_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta2</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">radam_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAMP</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">AdamP</span>
+                    <span class="n">adamp_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta2</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adamp_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">SGDP</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">SGDP</span>
+                    <span class="n">sgdp_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;momentum&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_momentum</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">sgdp_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">YOGI</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">Yogi</span>
+                    <span class="n">yogi_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta2</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">yogi_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">SOPHIA</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">SophiaG</span>
+                    <span class="n">sophia_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta2</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">sophia_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAM</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">Adam</span>
+                    <span class="n">adam_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta2</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adam_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">NOVOGRAD</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">NovoGrad</span>
+                    <span class="n">novograd_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta2</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">novograd_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADADELTA</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">Adadelta</span>
+                    <span class="n">adadelta_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adadelta_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAGRAD</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">AdaGrad</span>
+                    <span class="n">adagrad_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adagrad_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAMW_SCHEDULE_FREE</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">AdamWScheduleFree</span>
+                    <span class="n">adamw_schedule_free_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta2</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adamw_schedule_free_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">SGD_SCHEDULE_FREE</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">SGDScheduleFree</span>
+                    <span class="n">sgd_schedule_free_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;momentum&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_momentum</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">sgd_schedule_free_kwargs</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAN</span><span class="p">:</span>
+                    <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">optim</span><span class="o">.</span><span class="n">Adan</span>
+                    <span class="n">adan_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta2</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_beta3</span><span class="p">),</span>
+                        <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">optim_weight_decay</span><span class="p">),</span>
+                    <span class="p">}</span>
+                    <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adan_kwargs</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;Trainer cannot instantiate unsupported optimizer: &quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">args</span><span class="o">.</span><span class="n">customized_optim</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="p">)</span>
+                <span class="k">return</span> <span class="n">optimizer_cls</span><span class="p">,</span> <span class="n">optimizer_kwargs</span>
+
+            <span class="k">def</span><span class="w"> </span><span class="nf">create_optimizer</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+                <span class="n">opt_model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span> <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">()</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span>
+
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">decay_parameters</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_decay_parameter_names</span><span class="p">(</span><span class="n">opt_model</span><span class="p">)</span>
+                    <span class="n">optimizer_grouped_parameters</span> <span class="o">=</span> <span class="p">[</span>
+                        <span class="p">{</span>
+                            <span class="s2">&quot;params&quot;</span><span class="p">:</span> <span class="p">[</span>
+                                <span class="n">p</span> <span class="k">for</span> <span class="n">n</span><span class="p">,</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">opt_model</span><span class="o">.</span><span class="n">named_parameters</span><span class="p">()</span>
+                                    <span class="k">if</span> <span class="p">(</span><span class="n">n</span> <span class="ow">in</span> <span class="n">decay_parameters</span> <span class="ow">and</span> <span class="n">p</span><span class="o">.</span><span class="n">requires_grad</span><span class="p">)</span>
+                            <span class="p">],</span>
+                            <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">weight_decay</span><span class="p">,</span>
+                        <span class="p">},</span>
+                        <span class="p">{</span>
+                            <span class="s2">&quot;params&quot;</span><span class="p">:</span> <span class="p">[</span>
+                                <span class="n">p</span> <span class="k">for</span> <span class="n">n</span><span class="p">,</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">opt_model</span><span class="o">.</span><span class="n">named_parameters</span><span class="p">()</span>
+                                    <span class="k">if</span> <span class="p">(</span><span class="n">n</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">decay_parameters</span> <span class="ow">and</span> <span class="n">p</span><span class="o">.</span><span class="n">requires_grad</span><span class="p">)</span>
+                            <span class="p">],</span>
+                            <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="mf">0.0</span><span class="p">,</span>
+                        <span class="p">},</span>
+                    <span class="p">]</span>
+
+                    <span class="n">optimizer_cls</span><span class="p">,</span> <span class="n">optimizer_kwargs</span> <span class="o">=</span> <span class="n">CustomizedOptimTrainer</span><span class="o">.</span><span class="n">get_optimizer_cls_and_kwargs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="n">opt_model</span><span class="p">)</span>
+
+                    <span class="c1"># Overwrite `params` in case it&#39;s created by</span>
+                    <span class="c1"># `get_optimizer_cls_and_kwargs` e.g. for GaLore optimizer.</span>
+                    <span class="k">if</span> <span class="s2">&quot;params&quot;</span> <span class="ow">in</span> <span class="n">optimizer_kwargs</span><span class="p">:</span>
+                        <span class="n">optimizer_grouped_parameters</span> <span class="o">=</span> <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span>
+                            <span class="s2">&quot;params&quot;</span>
+                        <span class="p">)</span>
+
+                    <span class="c1"># For layer-wise dummy optimizers we overwrite</span>
+                    <span class="c1"># optimizer_grouped_parameters with `optimizer_dict` to</span>
+                    <span class="c1"># avoid arguments conflicts.</span>
+                    <span class="k">if</span> <span class="s2">&quot;optimizer_dict&quot;</span> <span class="ow">in</span> <span class="n">optimizer_kwargs</span><span class="p">:</span>
+                        <span class="n">optimizer_grouped_parameters</span> <span class="o">=</span> <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span>
+                            <span class="s2">&quot;optimizer_dict&quot;</span>
+                        <span class="p">)</span>
+
+                    <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">optimizer_cls</span><span class="p">(</span>
+                        <span class="n">optimizer_grouped_parameters</span><span class="p">,</span>
+                        <span class="o">**</span><span class="n">optimizer_kwargs</span>
+                    <span class="p">)</span>
+                <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">smp</span><span class="o">.</span><span class="n">DistributedOptimizer</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">)</span>
+                    
+        <span class="k">return</span> <span class="n">CustomizedOptimTrainer</span></div>
+
+
+<div class="viewcode-block" id="Finetuner.tune">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.tune">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">tune</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
+             <span class="n">model</span><span class="p">,</span>
+             <span class="n">dataset</span><span class="p">,</span>
+             <span class="n">transform_dataset_in_place</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+             <span class="n">data_collator</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform tuning for a model</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        model : TunableModel object.</span>
+<span class="sd">            TunableModel to perform tuning.</span>
+
+<span class="sd">        dataset:</span>
+<span class="sd">            dataset to train model.</span>
+
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">model_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span>
+        <span class="n">data_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span>
+        <span class="n">finetuner_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">transform_dataset_in_place</span><span class="p">:</span>
+            <span class="n">dataset</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span>
+
+        <span class="c1"># Tokenization and text grouping must be done in the main process</span>
+        <span class="k">if</span> <span class="n">dataset</span><span class="o">.</span><span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;custom_multi_modal&quot;</span><span class="p">:</span>
+            <span class="n">dataset</span><span class="o">.</span><span class="n">backend_dataset</span><span class="o">.</span><span class="n">register_tokenizer</span><span class="p">(</span>
+                <span class="n">model</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">,</span> <span class="n">model</span><span class="o">.</span><span class="n">image_processor</span><span class="p">)</span>
+            <span class="n">lm_dataset</span> <span class="o">=</span> <span class="n">dataset</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">with</span> <span class="n">finetuner_args</span><span class="o">.</span><span class="n">main_process_first</span><span class="p">(</span><span class="n">desc</span><span class="o">=</span><span class="s2">&quot;dataset map tokenization&quot;</span><span class="p">):</span>
+                <span class="n">tokenized_dataset</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">tokenize</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+                    <span class="n">lm_dataset</span> <span class="o">=</span> <span class="n">tokenized_dataset</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">lm_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">group_text</span><span class="p">(</span>
+                        <span class="n">tokenized_dataset</span><span class="p">,</span>
+                        <span class="n">model_max_length</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">get_max_length</span><span class="p">(),</span>
+                    <span class="p">)</span>
+
+        <span class="n">train_dataset</span> <span class="o">=</span> <span class="n">lm_dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Number of train samples: </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">finetuner_args</span><span class="o">.</span><span class="n">do_eval</span><span class="p">:</span>
+            <span class="n">eval_dataset_args</span> <span class="o">=</span> <span class="n">deepcopy</span><span class="p">(</span><span class="n">data_args</span><span class="p">)</span>
+            <span class="n">eval_dataset_args</span><span class="o">.</span><span class="n">dataset_path</span> <span class="o">=</span> <span class="n">finetuner_args</span><span class="o">.</span><span class="n">eval_dataset_path</span>
+            <span class="n">eval_dataset</span> <span class="o">=</span> <span class="n">Dataset</span><span class="p">(</span><span class="n">eval_dataset_args</span><span class="p">)</span>
+            <span class="k">with</span> <span class="n">finetuner_args</span><span class="o">.</span><span class="n">main_process_first</span><span class="p">(</span><span class="n">desc</span><span class="o">=</span><span class="s2">&quot;dataset map tokenization&quot;</span><span class="p">):</span>
+                <span class="n">tokenized_dataset</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">tokenize</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+                    <span class="n">lm_dataset</span> <span class="o">=</span> <span class="n">tokenized_dataset</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">lm_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">group_text</span><span class="p">(</span>
+                        <span class="n">tokenized_dataset</span><span class="p">,</span>
+                        <span class="n">model_max_length</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">get_max_length</span><span class="p">(),</span>
+                    <span class="p">)</span>
+            <span class="n">eval_dataset</span> <span class="o">=</span> <span class="n">lm_dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Number of eval samples: </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+            <span class="k">def</span><span class="w"> </span><span class="nf">preprocess_logits_for_metrics</span><span class="p">(</span><span class="n">logits</span><span class="p">,</span> <span class="n">labels</span><span class="p">):</span>
+                <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">logits</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">):</span>
+                    <span class="c1"># Depending on the model and config, logits may contain extra tensors,</span>
+                    <span class="c1"># like past_key_values, but logits always come first</span>
+                    <span class="n">logits</span> <span class="o">=</span> <span class="n">logits</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+                <span class="k">return</span> <span class="n">logits</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+
+            <span class="n">metric</span> <span class="o">=</span> <span class="n">evaluate</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="s2">&quot;accuracy&quot;</span><span class="p">)</span>
+
+            <span class="k">def</span><span class="w"> </span><span class="nf">compute_metrics</span><span class="p">(</span><span class="n">eval_preds</span><span class="p">):</span>
+                <span class="n">preds</span><span class="p">,</span> <span class="n">labels</span> <span class="o">=</span> <span class="n">eval_preds</span>
+                <span class="c1"># preds have the same shape as the labels, after the argmax(-1) has been calculated</span>
+                <span class="c1"># by preprocess_logits_for_metrics but we need to shift the labels</span>
+                <span class="n">labels</span> <span class="o">=</span> <span class="n">labels</span><span class="p">[:,</span> <span class="mi">1</span><span class="p">:]</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+                <span class="n">preds</span> <span class="o">=</span> <span class="n">preds</span><span class="p">[:,</span> <span class="p">:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span>
+                <span class="k">return</span> <span class="n">metric</span><span class="o">.</span><span class="n">compute</span><span class="p">(</span><span class="n">predictions</span><span class="o">=</span><span class="n">preds</span><span class="p">,</span> <span class="n">references</span><span class="o">=</span><span class="n">labels</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">finetuner_args</span><span class="o">.</span><span class="n">do_train</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">max_train_samples</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">max_train_samples</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">),</span> <span class="n">data_args</span><span class="o">.</span><span class="n">max_train_samples</span><span class="p">)</span>
+                <span class="n">train_dataset</span> <span class="o">=</span> <span class="n">train_dataset</span><span class="o">.</span><span class="n">select</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="n">max_train_samples</span><span class="p">))</span>
+
+        <span class="c1"># Initialize our Trainer</span>
+        <span class="n">training_args</span> <span class="o">=</span> <span class="n">finetuner_args</span>
+
+        <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_lora</span><span class="p">:</span>
+            <span class="n">FinetuningTrainer</span> <span class="o">=</span> <span class="n">PeftTrainer</span>
+            <span class="n">trainer_callbacks</span> <span class="o">=</span> <span class="p">[</span><span class="n">PeftSavingCallback</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">FinetuningTrainer</span> <span class="o">=</span> <span class="n">Trainer</span>
+            <span class="n">trainer_callbacks</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">if</span> <span class="n">data_collator</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">data_collator</span> <span class="o">=</span> <span class="n">default_data_collator</span>
+
+        <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">use_customized_optim</span><span class="p">:</span>
+            <span class="n">BaseTrainer</span> <span class="o">=</span> <span class="n">FinetuningTrainer</span>
+            <span class="n">FinetuningTrainer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">create_customized_optimizer</span><span class="p">(</span>
+                <span class="n">BaseTrainer</span><span class="p">,</span> <span class="n">model_args</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">use_lisa</span><span class="p">:</span>
+            <span class="k">class</span><span class="w"> </span><span class="nc">DynamicLayerActivationCallback</span><span class="p">(</span><span class="n">TrainerCallback</span><span class="p">):</span>
+                <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">n_layers</span><span class="p">,</span> <span class="n">interval_steps</span><span class="p">,</span> <span class="n">model</span><span class="p">):</span>
+                    <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">n_layers</span> <span class="o">=</span> <span class="n">n_layers</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">interval_steps</span> <span class="o">=</span> <span class="n">interval_steps</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">model</span>
+
+                    <span class="c1"># Determine the way to access layers based on the model type</span>
+                    <span class="n">class_to_layers_map</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s1">&#39;LlamaForCausalLM&#39;</span><span class="p">:</span> <span class="s1">&#39;model.model.layers&#39;</span><span class="p">,</span>
+                        <span class="s1">&#39;Qwen2ForCausalLM&#39;</span><span class="p">:</span> <span class="s1">&#39;model.model.layers&#39;</span><span class="p">,</span>
+                        <span class="s1">&#39;MistralForCausalLM&#39;</span><span class="p">:</span> <span class="s1">&#39;model.model.layers&#39;</span><span class="p">,</span>
+                        <span class="s1">&#39;MixtralForCausalLM&#39;</span><span class="p">:</span> <span class="s1">&#39;model.model.layers&#39;</span><span class="p">,</span>
+                        <span class="s1">&#39;GemmaForCausalLM&#39;</span><span class="p">:</span> <span class="s1">&#39;model.model.layers&#39;</span><span class="p">,</span>
+                        <span class="s1">&#39;GPT2LMHeadModel&#39;</span><span class="p">:</span> <span class="s1">&#39;model.transformer.h&#39;</span><span class="p">,</span>
+                        <span class="s1">&#39;HymbaForCausalLM&#39;</span><span class="p">:</span> <span class="s1">&#39;model.model.layers&#39;</span><span class="p">,</span>
+                    <span class="p">}</span>
+                    <span class="n">model_class_name</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span>
+                    <span class="k">if</span> <span class="n">model_class_name</span> <span class="ow">in</span> <span class="n">class_to_layers_map</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">layers_attribute</span> <span class="o">=</span> <span class="n">class_to_layers_map</span><span class="p">[</span><span class="n">model_class_name</span><span class="p">]</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">layers_attribute</span> <span class="o">=</span> <span class="n">training_args</span><span class="o">.</span><span class="n">lisa_layers_attribute</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">total_layers</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="nb">eval</span><span class="p">(</span><span class="s1">&#39;self.&#39;</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">layers_attribute</span><span class="p">))</span>  <span class="c1"># Dynamically execute to get the number of layers</span>
+
+                    <span class="bp">self</span><span class="o">.</span><span class="n">active_layers_indices</span> <span class="o">=</span> <span class="p">[]</span>
+
+                <span class="k">def</span><span class="w"> </span><span class="nf">freeze_all_layers</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+                    <span class="n">layers</span> <span class="o">=</span> <span class="nb">eval</span><span class="p">(</span><span class="s1">&#39;self.&#39;</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">layers_attribute</span><span class="p">)</span>  <span class="c1"># Dynamically execute to get layers</span>
+                    <span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="n">layers</span><span class="p">:</span>
+                        <span class="k">for</span> <span class="n">param</span> <span class="ow">in</span> <span class="n">layer</span><span class="o">.</span><span class="n">parameters</span><span class="p">():</span>
+                            <span class="n">param</span><span class="o">.</span><span class="n">requires_grad</span> <span class="o">=</span> <span class="kc">False</span>
+
+                <span class="k">def</span><span class="w"> </span><span class="nf">on_step_begin</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">,</span> <span class="n">state</span><span class="p">,</span> <span class="n">control</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+                    <span class="c1"># Check if it&#39;s time to switch active layers, including at step 0</span>
+                    <span class="k">if</span> <span class="n">state</span><span class="o">.</span><span class="n">global_step</span> <span class="o">%</span> <span class="bp">self</span><span class="o">.</span><span class="n">interval_steps</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">switch_active_layers</span><span class="p">()</span>
+
+                <span class="k">def</span><span class="w"> </span><span class="nf">switch_active_layers</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+                    <span class="c1"># First, disable gradients for all layers</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">freeze_all_layers</span><span class="p">()</span>
+
+                     <span class="c1"># Randomly select n_layers to activate</span>
+                    <span class="n">layers</span> <span class="o">=</span> <span class="nb">eval</span><span class="p">(</span><span class="s1">&#39;self.&#39;</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">layers_attribute</span><span class="p">)</span>  <span class="c1"># Re-fetch layer references</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">active_layers_indices</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">choice</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">total_layers</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">n_layers</span><span class="p">,</span> <span class="n">replace</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+                    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Activating layers at indices: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">active_layers_indices</span><span class="si">}</span><span class="s2"> for the next steps.&quot;</span><span class="p">,</span> <span class="n">flush</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+                    <span class="c1"># Enable gradients only for the selected layers</span>
+                    <span class="k">for</span> <span class="n">idx</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">active_layers_indices</span><span class="p">:</span>
+                        <span class="k">for</span> <span class="n">param</span> <span class="ow">in</span> <span class="n">layers</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span><span class="o">.</span><span class="n">parameters</span><span class="p">():</span>
+                            <span class="n">param</span><span class="o">.</span><span class="n">requires_grad</span> <span class="o">=</span> <span class="kc">True</span>
+
+            <span class="c1"># Instantiate the callback</span>
+            <span class="n">dynamic_layer_activation_callback</span> <span class="o">=</span> <span class="n">DynamicLayerActivationCallback</span><span class="p">(</span>
+                <span class="n">n_layers</span><span class="o">=</span><span class="n">training_args</span><span class="o">.</span><span class="n">lisa_activated_layers</span><span class="p">,</span>               <span class="c1"># Number of layers to activate</span>
+                <span class="n">interval_steps</span><span class="o">=</span><span class="n">training_args</span><span class="o">.</span><span class="n">lisa_interval_steps</span><span class="p">,</span>           <span class="c1"># Step interval to update active layers</span>
+                <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span>
+            <span class="p">)</span>
+
+            <span class="n">trainer_callbacks</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">dynamic_layer_activation_callback</span><span class="p">)</span>
+
+        <span class="n">trainer</span> <span class="o">=</span> <span class="n">FinetuningTrainer</span><span class="p">(</span>
+            <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">(),</span>
+            <span class="n">args</span><span class="o">=</span><span class="n">training_args</span><span class="p">,</span>
+            <span class="n">train_dataset</span><span class="o">=</span><span class="n">train_dataset</span> <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">do_train</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="n">eval_dataset</span><span class="o">=</span><span class="n">eval_dataset</span> <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">do_eval</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="n">tokenizer</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">get_tokenizer</span><span class="p">(),</span>
+            <span class="c1"># Data collator will default to DataCollatorWithPadding, so we change it.</span>
+            <span class="n">data_collator</span><span class="o">=</span><span class="n">data_collator</span><span class="p">,</span>
+            <span class="n">compute_metrics</span><span class="o">=</span><span class="n">compute_metrics</span> <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">do_eval</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="n">preprocess_logits_for_metrics</span><span class="o">=</span><span class="n">preprocess_logits_for_metrics</span> <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">do_eval</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="n">callbacks</span><span class="o">=</span><span class="n">trainer_callbacks</span>
+        <span class="p">)</span>
+        <span class="c1"># Training</span>
+        <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">do_train</span><span class="p">:</span>
+            <span class="n">checkpoint</span> <span class="o">=</span> <span class="kc">None</span>
+            <span class="n">last_checkpoint</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">last_checkpoint</span>
+            <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">resume_from_checkpoint</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">checkpoint</span> <span class="o">=</span> <span class="n">training_args</span><span class="o">.</span><span class="n">resume_from_checkpoint</span>
+            <span class="k">elif</span> <span class="n">last_checkpoint</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">checkpoint</span> <span class="o">=</span> <span class="n">last_checkpoint</span>
+            <span class="n">train_result</span> <span class="o">=</span> <span class="n">trainer</span><span class="o">.</span><span class="n">train</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="o">=</span><span class="n">checkpoint</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">model_args</span><span class="o">.</span><span class="n">use_lora</span><span class="p">:</span>
+                <span class="n">trainer</span><span class="o">.</span><span class="n">save_model</span><span class="p">()</span>  <span class="c1"># Saves the tokenizer too for easy upload</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">save_aggregated_lora</span><span class="p">:</span>
+                    <span class="n">model</span><span class="o">.</span><span class="n">merge_lora_weights</span><span class="p">()</span>
+                <span class="n">model</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">model_args</span><span class="o">.</span><span class="n">save_aggregated_lora</span><span class="p">)</span>
+            <span class="c1"># save language_projection for multi-modal model;</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">save_language_projection</span><span class="p">:</span>
+                <span class="n">language_projection_state</span> <span class="o">=</span> <span class="n">trainer</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">language_projection</span><span class="o">.</span><span class="n">state_dict</span><span class="p">()</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span>
+                    <span class="n">osp</span><span class="o">.</span><span class="n">join</span><span class="p">(</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span>
+                        <span class="s2">&quot;language_projection.pth&quot;</span><span class="p">),</span>
+                    <span class="n">language_projection_state</span><span class="p">)</span>
+            <span class="n">metrics</span> <span class="o">=</span> <span class="n">train_result</span><span class="o">.</span><span class="n">metrics</span>
+
+            <span class="n">max_train_samples</span> <span class="o">=</span> <span class="p">(</span>
+                <span class="n">data_args</span><span class="o">.</span><span class="n">max_train_samples</span> <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">max_train_samples</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="nb">len</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">)</span>
+            <span class="p">)</span>
+            <span class="n">metrics</span><span class="p">[</span><span class="s2">&quot;train_samples&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">max_train_samples</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">))</span>
+
+            <span class="n">trainer</span><span class="o">.</span><span class="n">log_metrics</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="n">metrics</span><span class="p">)</span>
+            <span class="n">trainer</span><span class="o">.</span><span class="n">save_metrics</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="n">metrics</span><span class="p">)</span>
+            <span class="n">trainer</span><span class="o">.</span><span class="n">save_state</span><span class="p">()</span>
+
+        <span class="n">kwargs</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;finetuned_from&quot;</span><span class="p">:</span> <span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span> <span class="s2">&quot;tasks&quot;</span><span class="p">:</span> <span class="s2">&quot;text-generation&quot;</span><span class="p">}</span>
+        <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">dataset_name</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;dataset_tags&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">data_args</span><span class="o">.</span><span class="n">dataset_name</span>
+            <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">dataset_config_name</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;dataset_args&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">data_args</span><span class="o">.</span><span class="n">dataset_config_name</span>
+                <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;dataset&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">data_args</span><span class="o">.</span><span class="n">dataset_name</span><span class="si">}</span><span class="s2"> </span><span class="si">{</span><span class="n">data_args</span><span class="o">.</span><span class="n">dataset_config_name</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;dataset&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">data_args</span><span class="o">.</span><span class="n">dataset_name</span>
+
+        <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">push_to_hub</span><span class="p">:</span>
+            <span class="n">trainer</span><span class="o">.</span><span class="n">push_to_hub</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">trainer</span><span class="o">.</span><span class="n">create_model_card</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">model</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/inferencer.html b/_modules/lmflow/pipeline/inferencer.html
new file mode 100644
index 000000000..ee436bcd8
--- /dev/null
+++ b/_modules/lmflow/pipeline/inferencer.html
@@ -0,0 +1,1181 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.inferencer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/inferencer';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.inferencer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.inferencer</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;The Inferencer class simplifies the process of model inferencing.&quot;&quot;&quot;</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">copy</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">wandb</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">deepspeed</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">datetime</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">json</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">time</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">concurrent.futures</span><span class="w"> </span><span class="kn">import</span> <span class="n">ThreadPoolExecutor</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">subprocess</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">accelerate</span><span class="w"> </span><span class="kn">import</span> <span class="n">Accelerator</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">AutoConfig</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.distributed</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">dist</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.nn.functional</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">F</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="n">DatasetArguments</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets.dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.base_pipeline</span><span class="w"> </span><span class="kn">import</span> <span class="n">BasePipeline</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_decoder_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFDecoderModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.data_utils</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span><span class="n">set_random_seed</span><span class="p">,</span> <span class="n">batchlize</span><span class="p">,</span>
+                                     <span class="n">answer_extraction</span><span class="p">,</span> <span class="n">process_image_flag</span><span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="n">IMAGE_TOKEN_INDEX</span>
+<span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="p">[</span><span class="s2">&quot;TOKENIZERS_PARALLELISM&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;false&quot;</span>  <span class="c1"># To avoid warnings about parallelism in tokenizers</span>
+<div class="viewcode-block" id="rstrip_partial_utf8">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.rstrip_partial_utf8">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">rstrip_partial_utf8</span><span class="p">(</span><span class="n">string</span><span class="p">):</span>
+    <span class="k">return</span> <span class="n">string</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\ufffd</span><span class="s2">&quot;</span><span class="p">,</span> <span class="s2">&quot;&quot;</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="supported_dataset_type">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.supported_dataset_type">[docs]</a>
+<span class="n">supported_dataset_type</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;text_only&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;image_text&quot;</span><span class="p">,</span>
+<span class="p">]</span></div>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="Inferencer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Inferencer</span><span class="p">(</span><span class="n">BasePipeline</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Initializes the `Inferencer` class with given arguments.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+<span class="sd">    model_args : ModelArguments object.</span>
+<span class="sd">        Contains the arguments required to load the model.</span>
+
+<span class="sd">    data_args : DatasetArguments object.</span>
+<span class="sd">        Contains the arguments required to load the dataset.</span>
+
+<span class="sd">    inferencer_args : InferencerArguments object.</span>
+<span class="sd">        Contains the arguments required to perform inference.</span>
+
+
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">inferencer_args</span><span class="p">):</span>
+<div class="viewcode-block" id="Inferencer.data_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.data_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span> <span class="o">=</span> <span class="n">data_args</span></div>
+
+<div class="viewcode-block" id="Inferencer.inferencer_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.inferencer_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span> <span class="o">=</span> <span class="n">inferencer_args</span></div>
+
+<div class="viewcode-block" id="Inferencer.model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span> <span class="o">=</span> <span class="n">model_args</span></div>
+
+
+        <span class="n">set_random_seed</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">random_seed</span><span class="p">)</span>
+
+<div class="viewcode-block" id="Inferencer.local_rank">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.local_rank">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">getenv</span><span class="p">(</span><span class="s2">&quot;LOCAL_RANK&quot;</span><span class="p">,</span> <span class="s2">&quot;0&quot;</span><span class="p">))</span></div>
+
+<div class="viewcode-block" id="Inferencer.world_size">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.world_size">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">world_size</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">getenv</span><span class="p">(</span><span class="s2">&quot;WORLD_SIZE&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">))</span></div>
+
+        <span class="k">if</span> <span class="n">inferencer_args</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;gpu&quot;</span><span class="p">:</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">set_device</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>  <span class="c1"># NOTE: cpu-only machine will have error</span>
+            <span class="n">deepspeed</span><span class="o">.</span><span class="n">init_distributed</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="p">[</span><span class="s2">&quot;MASTER_ADDR&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;localhost&quot;</span>
+            <span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="p">[</span><span class="s2">&quot;MASTER_PORT&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;15000&quot;</span>
+            <span class="n">dist</span><span class="o">.</span><span class="n">init_process_group</span><span class="p">(</span>
+                <span class="s2">&quot;gloo&quot;</span><span class="p">,</span> <span class="n">rank</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">,</span> <span class="n">world_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">world_size</span>
+            <span class="p">)</span>
+
+<div class="viewcode-block" id="Inferencer.config">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.config">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">config</span> <span class="o">=</span> <span class="n">AutoConfig</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span> <span class="n">trust_remote_code</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span></div>
+
+        <span class="k">try</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model_hidden_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">hidden_size</span>
+        <span class="k">except</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Error in setting hidden size, use the default size 1024&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model_hidden_size</span> <span class="o">=</span> <span class="mi">1024</span> <span class="c1"># gpt2 seems do not have hidden_size in config</span>
+
+        <span class="k">if</span> <span class="n">inferencer_args</span><span class="o">.</span><span class="n">use_accelerator</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span> <span class="o">=</span> <span class="n">Accelerator</span><span class="p">()</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="o">.</span><span class="n">wait_for_everyone</span><span class="p">()</span>
+
+
+<div class="viewcode-block" id="Inferencer.create_dataloader">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.create_dataloader">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">create_dataloader</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">):</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&quot;&quot;&quot;Batchlize dataset and format it to dataloader.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            dataset (Dataset): the dataset object</span>
+
+<span class="sd">        Output:</span>
+<span class="sd">            dataloader (batchlize): the dataloader object</span>
+<span class="sd">            dataset_size (int): the length of the dataset</span>
+
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="o">==</span> <span class="s2">&quot;text_only&quot;</span><span class="p">:</span>
+            <span class="n">data_dict</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">to_dict</span><span class="p">()</span>
+            <span class="n">inputs</span> <span class="o">=</span> <span class="p">[</span><span class="n">instance</span><span class="p">[</span><span class="s2">&quot;text&quot;</span><span class="p">]</span> <span class="k">for</span> <span class="n">instance</span> <span class="ow">in</span> <span class="n">data_dict</span><span class="p">[</span><span class="s2">&quot;instances&quot;</span><span class="p">]</span> <span class="p">]</span>
+        <span class="k">elif</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="o">==</span> <span class="s2">&quot;image_text&quot;</span><span class="p">:</span>
+            <span class="n">inputs</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">to_list</span><span class="p">()</span>
+
+        <span class="n">dataset_size</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="n">dataset_buf</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">idx</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">dataset_size</span><span class="p">):</span>
+            <span class="n">dataset_buf</span><span class="o">.</span><span class="n">append</span><span class="p">({</span>
+                <span class="s2">&quot;input&quot;</span><span class="p">:</span> <span class="n">inputs</span><span class="p">[</span><span class="n">idx</span><span class="p">],</span>
+                <span class="s2">&quot;input_idx&quot;</span><span class="p">:</span> <span class="n">idx</span>
+            <span class="p">})</span>
+
+        <span class="n">dataloader</span> <span class="o">=</span> <span class="n">batchlize</span><span class="p">(</span>
+            <span class="n">dataset_buf</span><span class="p">,</span>
+            <span class="n">batch_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+            <span class="n">random_shuffle</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">dataloader</span><span class="p">,</span> <span class="n">dataset_size</span></div>
+
+
+
+<div class="viewcode-block" id="Inferencer.inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">max_new_tokens</span><span class="p">:</span> <span class="nb">int</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span>
+        <span class="n">temperature</span><span class="p">:</span> <span class="nb">float</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+        <span class="n">prompt_structure</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s1">&#39;</span><span class="si">{input}</span><span class="s1">&#39;</span><span class="p">,</span>
+        <span class="n">remove_image_flag</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">chatbot_type</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s2">&quot;mini_gpt&quot;</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform inference for a model</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        model : TunableModel object.</span>
+<span class="sd">            TunableModel to perform inference</span>
+
+<span class="sd">        dataset : Dataset object.</span>
+
+
+<span class="sd">        Returns:</span>
+
+<span class="sd">        output_dataset: Dataset object.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">supported_dataset_type</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="s1">&#39;input dataset should have type </span><span class="si">{}</span><span class="s1">&#39;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span>
+                                        <span class="n">supported_dataset_type</span><span class="p">))</span>
+        <span class="n">dataloader</span><span class="p">,</span> <span class="n">data_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">create_dataloader</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span>
+
+        <span class="c1"># The output dataset</span>
+        <span class="n">output_dict</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="s2">&quot;text_only&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;instances&quot;</span><span class="p">:</span> <span class="p">[</span>
+            <span class="p">]</span>
+        <span class="p">}</span>
+
+        <span class="k">for</span> <span class="n">batch_index</span><span class="p">,</span> <span class="n">batch</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">dataloader</span><span class="p">):</span>
+            <span class="n">current_batch</span> <span class="o">=</span> <span class="n">batch</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>        <span class="c1"># batch size is 1</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">current_batch</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">],</span> <span class="nb">str</span><span class="p">):</span>
+                <span class="nb">input</span> <span class="o">=</span> <span class="n">prompt_structure</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">current_batch</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">])</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="nb">input</span> <span class="o">=</span> <span class="n">current_batch</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">]</span>
+                <span class="nb">input</span><span class="p">[</span><span class="s1">&#39;text&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">prompt_structure</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">[</span><span class="s1">&#39;text&#39;</span><span class="p">])</span>
+
+            <span class="k">if</span> <span class="kc">False</span> <span class="ow">and</span> <span class="s1">&#39;images&#39;</span> <span class="ow">in</span> <span class="nb">input</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="nb">input</span><span class="p">[</span><span class="s1">&#39;images&#39;</span><span class="p">],</span> <span class="nb">list</span><span class="p">):</span>
+                <span class="nb">input</span><span class="p">[</span><span class="s1">&#39;images&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="nb">input</span><span class="p">[</span><span class="s1">&#39;images&#39;</span><span class="p">])</span>
+            <span class="k">if</span> <span class="n">remove_image_flag</span><span class="p">:</span>
+                <span class="c1"># remove the image flag &lt;ImageHere&gt; in tokenization;</span>
+                <span class="k">if</span> <span class="n">chatbot_type</span> <span class="o">==</span> <span class="s2">&quot;mini_gpt&quot;</span><span class="p">:</span>
+                    <span class="n">image_split_flag</span> <span class="o">=</span> <span class="s2">&quot;&lt;ImageHere&gt;&quot;</span>
+                <span class="k">elif</span> <span class="n">chatbot_type</span><span class="p">:</span>
+                    <span class="n">image_split_flag</span> <span class="o">=</span> <span class="s2">&quot;&lt;image&gt;&quot;</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">NotImplementedError</span>
+                <span class="nb">input</span><span class="p">[</span><span class="s1">&#39;text&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="nb">input</span><span class="p">[</span><span class="s1">&#39;text&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">image_split_flag</span><span class="p">)</span>
+                <span class="c1"># TODO remove this code by update the tokenizer</span>
+                <span class="n">input_ids</span> <span class="o">=</span> <span class="p">[]</span>
+                <span class="n">attention_mask</span> <span class="o">=</span> <span class="p">[]</span>
+                <span class="n">image_token_indexes</span> <span class="o">=</span> <span class="p">[]</span>
+                <span class="n">temp_input</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="nb">input</span><span class="p">)</span>
+                <span class="k">for</span> <span class="n">idx</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="nb">input</span><span class="p">[</span><span class="s1">&#39;text&#39;</span><span class="p">])):</span>
+                    <span class="n">temp_input</span><span class="p">[</span><span class="s1">&#39;text&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="nb">input</span><span class="p">[</span><span class="s1">&#39;text&#39;</span><span class="p">][</span><span class="n">idx</span><span class="p">]</span>
+                    <span class="n">temp_inputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span>
+                        <span class="n">temp_input</span><span class="p">,</span>
+                        <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span><span class="p">,</span>
+                        <span class="n">add_special_tokens</span><span class="o">=</span><span class="n">idx</span> <span class="o">==</span> <span class="mi">0</span>
+                    <span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>
+                    <span class="n">input_ids</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">temp_inputs</span><span class="p">[</span><span class="s1">&#39;input_ids&#39;</span><span class="p">])</span>
+                    <span class="n">attention_mask</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">temp_inputs</span><span class="p">[</span><span class="s1">&#39;attention_mask&#39;</span><span class="p">])</span>
+                    <span class="k">if</span> <span class="n">chatbot_type</span> <span class="o">==</span> <span class="s2">&quot;llava&quot;</span><span class="p">:</span>
+                        <span class="c1"># add the flag for inserting the image.</span>
+                        <span class="c1"># TODO should merge the way of handling image flag in minigpt and llava.</span>
+                        <span class="n">index_tensor</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span>
+                            <span class="p">[</span><span class="n">IMAGE_TOKEN_INDEX</span><span class="p">]</span>
+                        <span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>
+                        <span class="n">index_tensor</span> <span class="o">=</span> <span class="n">index_tensor</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+                        <span class="n">input_ids</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">index_tensor</span><span class="p">)</span>
+                        <span class="n">attention_mask</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                            <span class="n">torch</span><span class="o">.</span><span class="n">ones</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">))</span>
+                    <span class="n">image_token_indexes</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                        <span class="n">temp_inputs</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">image_token_indexes</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+                    <span class="n">image_token_indexes</span> <span class="o">=</span> <span class="n">image_token_indexes</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+                    <span class="k">if</span> <span class="n">chatbot_type</span> <span class="o">==</span> <span class="s2">&quot;llava&quot;</span><span class="p">:</span>
+                        <span class="n">input_ids</span> <span class="o">=</span> <span class="n">input_ids</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+                        <span class="n">attention_mask</span> <span class="o">=</span> <span class="n">attention_mask</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+                <span class="n">inputs</span> <span class="o">=</span> <span class="n">temp_inputs</span>
+                <span class="n">inputs</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">input_ids</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+                <span class="n">inputs</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">(</span><span class="n">attention_mask</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;gpu&quot;</span><span class="p">:</span>
+                    <span class="n">inputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span>
+                        <span class="nb">input</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span>
+                    <span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;cpu&quot;</span><span class="p">:</span>
+                    <span class="n">inputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span>
+                        <span class="nb">input</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span>
+                    <span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="s1">&#39;cpu&#39;</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;device </span><span class="se">\&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="se">\&quot;</span><span class="s2"> is not supported&quot;</span>
+                    <span class="p">)</span>
+
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">use_accelerator</span><span class="p">:</span>
+                    <span class="n">inputs</span> <span class="o">=</span> <span class="n">inputs</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+
+
+            <span class="k">if</span> <span class="n">remove_image_flag</span><span class="p">:</span>
+                <span class="n">inputs</span><span class="p">[</span><span class="s2">&quot;image_token_indexes&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">image_token_indexes</span>
+                <span class="n">inputs</span><span class="p">[</span><span class="s2">&quot;one_sample_multiple_images&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="kc">True</span>
+
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">use_accelerator</span><span class="p">:</span>
+                <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="o">.</span><span class="n">autocast</span><span class="p">():</span>
+                    <span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span>
+                        <span class="n">inputs</span><span class="p">,</span>
+                        <span class="n">max_new_tokens</span><span class="o">=</span><span class="n">max_new_tokens</span><span class="p">,</span>
+                        <span class="n">temperature</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">temperature</span><span class="p">,</span>
+                        <span class="n">repetition_penalty</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">repetition_penalty</span><span class="p">,</span>
+                        <span class="n">do_sample</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">do_sample</span><span class="p">,</span>
+                        <span class="n">use_accelerator</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                    <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span>
+                    <span class="n">inputs</span><span class="p">,</span>
+                    <span class="n">max_new_tokens</span><span class="o">=</span><span class="n">max_new_tokens</span><span class="p">,</span>
+                    <span class="n">temperature</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">temperature</span><span class="p">,</span>
+                    <span class="n">repetition_penalty</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">repetition_penalty</span><span class="p">,</span>
+                    <span class="n">do_sample</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">do_sample</span><span class="p">,</span>
+                <span class="p">)</span>
+
+            <span class="c1"># only return the generation, trucating the input</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">arch_type</span> <span class="o">!=</span> <span class="s2">&quot;vision_encoder_decoder&quot;</span><span class="p">:</span>
+                <span class="n">text_out</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">outputs</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">skip_special_tokens</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                <span class="n">prompt_length</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">model</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">inputs</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">skip_special_tokens</span><span class="o">=</span><span class="kc">True</span><span class="p">,))</span>
+                <span class="n">text_out</span> <span class="o">=</span> <span class="n">text_out</span><span class="p">[</span><span class="n">prompt_length</span><span class="p">:]</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="c1"># to avoid redundant/missing leading space problem, we use a</span>
+                <span class="c1"># part of the input text</span>
+                <span class="n">input_text</span> <span class="o">=</span> <span class="n">inputs</span><span class="p">[</span><span class="s1">&#39;input_ids&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">][</span><span class="o">-</span><span class="mi">1</span><span class="p">:]</span>
+                <span class="n">text_out</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">input_text</span><span class="p">,</span> <span class="n">outputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]]),</span> <span class="n">skip_special_tokens</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                <span class="n">prompt_length</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">model</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">input_text</span><span class="p">,</span> <span class="n">skip_special_tokens</span><span class="o">=</span><span class="kc">True</span><span class="p">,))</span>
+                <span class="n">text_out</span> <span class="o">=</span> <span class="n">text_out</span><span class="p">[</span><span class="n">prompt_length</span><span class="p">:]</span>
+
+            <span class="n">output_dict</span><span class="p">[</span><span class="s2">&quot;instances&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">({</span> <span class="s2">&quot;text&quot;</span><span class="p">:</span> <span class="n">text_out</span> <span class="p">})</span>
+
+        <span class="n">output_dataset</span> <span class="o">=</span> <span class="n">Dataset</span><span class="p">(</span><span class="n">DatasetArguments</span><span class="p">(</span><span class="n">dataset_path</span> <span class="o">=</span> <span class="kc">None</span><span class="p">))</span>
+        <span class="n">output_dataset</span> <span class="o">=</span> <span class="n">output_dataset</span><span class="o">.</span><span class="n">from_dict</span><span class="p">(</span><span class="n">output_dict</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">output_dataset</span></div>
+
+
+<div class="viewcode-block" id="Inferencer.stream_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.stream_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">stream_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">context</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">,</span>
+        <span class="n">max_new_tokens</span><span class="p">,</span>
+        <span class="n">token_per_step</span><span class="p">,</span>
+        <span class="n">temperature</span><span class="p">,</span>
+        <span class="n">end_string</span><span class="p">,</span>
+        <span class="n">input_dataset</span><span class="p">,</span>
+        <span class="n">remove_image_flag</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">response</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">history</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">if</span> <span class="s2">&quot;ChatGLMModel&quot;</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">architectures</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">response</span><span class="p">,</span> <span class="n">history</span> <span class="ow">in</span> <span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">stream_chat</span><span class="p">(</span><span class="n">model</span><span class="o">.</span><span class="n">get_tokenizer</span><span class="p">(),</span> <span class="n">context</span><span class="p">,</span> <span class="n">history</span><span class="o">=</span><span class="n">history</span><span class="p">):</span>
+                <span class="n">response</span> <span class="o">=</span> <span class="n">rstrip_partial_utf8</span><span class="p">(</span><span class="n">response</span><span class="p">)</span>
+                <span class="k">yield</span> <span class="n">response</span><span class="p">,</span> <span class="kc">False</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">max_new_tokens</span> <span class="o">//</span> <span class="n">token_per_step</span><span class="p">):</span>
+                <span class="n">output_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span>
+                    <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span>
+                    <span class="n">dataset</span><span class="o">=</span><span class="n">input_dataset</span><span class="p">,</span>
+                    <span class="n">max_new_tokens</span><span class="o">=</span><span class="n">token_per_step</span><span class="p">,</span>
+                    <span class="n">temperature</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">temperature</span><span class="p">,</span>
+                    <span class="n">remove_image_flag</span><span class="o">=</span><span class="n">remove_image_flag</span><span class="p">,</span>
+                <span class="p">)</span>
+
+                <span class="n">new_append_text</span> <span class="o">=</span> <span class="n">output_dataset</span><span class="o">.</span><span class="n">to_dict</span><span class="p">()[</span><span class="s2">&quot;instances&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">][</span><span class="s2">&quot;text&quot;</span><span class="p">]</span>
+                <span class="n">new_append_text</span> <span class="o">=</span> <span class="n">rstrip_partial_utf8</span><span class="p">(</span><span class="n">new_append_text</span><span class="p">)</span>
+                <span class="n">response</span> <span class="o">+=</span> <span class="n">new_append_text</span>
+
+                <span class="n">input_dict</span> <span class="o">=</span> <span class="n">input_dataset</span><span class="o">.</span><span class="n">to_dict</span><span class="p">()</span>
+                <span class="n">input_dict</span><span class="p">[</span><span class="s2">&quot;instances&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">][</span><span class="s2">&quot;text&quot;</span><span class="p">]</span> <span class="o">+=</span> <span class="n">new_append_text</span>
+                <span class="n">input_dataset</span> <span class="o">=</span> <span class="n">input_dataset</span><span class="o">.</span><span class="n">from_dict</span><span class="p">(</span><span class="n">input_dict</span><span class="p">)</span>
+
+                <span class="n">flag_break</span> <span class="o">=</span> <span class="kc">False</span>
+                <span class="k">try</span><span class="p">:</span>
+                    <span class="n">index</span> <span class="o">=</span> <span class="n">response</span><span class="o">.</span><span class="n">index</span><span class="p">(</span><span class="n">end_string</span><span class="p">)</span>
+                    <span class="n">flag_break</span> <span class="o">=</span> <span class="kc">True</span>
+                <span class="k">except</span> <span class="ne">ValueError</span><span class="p">:</span>
+                    <span class="n">response</span> <span class="o">+=</span> <span class="n">end_string</span>
+                    <span class="n">index</span> <span class="o">=</span> <span class="n">response</span><span class="o">.</span><span class="n">index</span><span class="p">(</span><span class="n">end_string</span><span class="p">)</span>
+
+                <span class="n">response</span> <span class="o">=</span> <span class="n">response</span><span class="p">[:</span><span class="n">index</span><span class="p">]</span>
+
+                <span class="k">yield</span> <span class="n">response</span><span class="p">,</span> <span class="n">flag_break</span></div>
+</div>
+
+
+
+<div class="viewcode-block" id="SpeculativeInferencer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">SpeculativeInferencer</span><span class="p">(</span><span class="n">Inferencer</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Ref: [arXiv:2211.17192v2](https://arxiv.org/abs/2211.17192)</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+<span class="sd">    target_model_args : ModelArguments object.</span>
+<span class="sd">        Contains the arguments required to load the target model.</span>
+<span class="sd">        </span>
+<span class="sd">    draft_model_args : ModelArguments object.</span>
+<span class="sd">        Contains the arguments required to load the draft model.</span>
+<span class="sd">    </span>
+<span class="sd">    data_args : DatasetArguments object.</span>
+<span class="sd">        Contains the arguments required to load the dataset.</span>
+
+<span class="sd">    inferencer_args : InferencerArguments object.</span>
+<span class="sd">        Contains the arguments required to perform inference.</span>
+
+
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model_args</span><span class="p">,</span> <span class="n">draft_model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">inferencer_args</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">inferencer_args</span><span class="p">)</span>
+<div class="viewcode-block" id="SpeculativeInferencer.draft_model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.draft_model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">draft_model_args</span> <span class="o">=</span> <span class="n">draft_model_args</span></div>
+
+
+<div class="viewcode-block" id="SpeculativeInferencer.draft_config">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.draft_config">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">draft_config</span> <span class="o">=</span> <span class="n">AutoConfig</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="n">draft_model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span> <span class="n">trust_remote_code</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span></div>
+
+        <span class="k">try</span><span class="p">:</span> 
+            <span class="bp">self</span><span class="o">.</span><span class="n">draft_model_hidden_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">draft_config</span><span class="o">.</span><span class="n">hidden_size</span>
+        <span class="k">except</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Error in setting hidden size for draft model, use the default size 1024&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">draft_model_hidden_size</span> <span class="o">=</span> <span class="mi">768</span>
+            
+    
+    <span class="nd">@staticmethod</span>        
+<div class="viewcode-block" id="SpeculativeInferencer.score_to_prob">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.score_to_prob">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">score_to_prob</span><span class="p">(</span><span class="n">scores</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> 
+                      <span class="n">temperature</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.</span><span class="p">,</span>
+                      <span class="n">top_p</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">1.</span><span class="p">,)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Convert scores (NOT softmaxed tensor) to probabilities with support for temperature, top-p sampling, and argmax.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ----------</span>
+<span class="sd">        scores : torch.Tensor</span>
+<span class="sd">            Input scores.</span>
+<span class="sd">        temperature : float, optional</span>
+<span class="sd">            Temperature parameter for controlling randomness. Higher values make the distribution more uniform, </span>
+<span class="sd">            lower values make it peakier. When temperature &lt;= 1e-6, argmax is used. by default 0.0</span>
+<span class="sd">        top_p : float, optional</span>
+<span class="sd">            Top-p sampling parameter for controlling the cumulative probability threshold, by default 1.0 (no threshold)</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        -------</span>
+<span class="sd">        torch.Tensor</span>
+<span class="sd">            Probability distribution after adjustments.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">assert</span> <span class="n">temperature</span> <span class="o">&gt;=</span> <span class="mf">0.0</span>
+        <span class="k">assert</span> <span class="mf">0.0</span> <span class="o">&lt;</span> <span class="n">top_p</span> <span class="o">&lt;=</span> <span class="mf">1.0</span>
+        
+        <span class="k">if</span> <span class="n">temperature</span> <span class="o">&lt;=</span> <span class="mf">1e-6</span><span class="p">:</span>
+            <span class="n">final_prob</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">one_hot</span><span class="p">(</span><span class="n">scores</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span> <span class="n">num_classes</span><span class="o">=</span><span class="n">scores</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">1</span><span class="p">))</span><span class="o">.</span><span class="n">float</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">scores</span> <span class="o">/=</span> <span class="n">temperature</span>
+            <span class="k">if</span> <span class="n">top_p</span> <span class="o">&lt;</span> <span class="mf">1.0</span><span class="p">:</span>
+                <span class="n">sorted_scores</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">sort</span><span class="p">(</span><span class="n">scores</span><span class="p">,</span> <span class="n">descending</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                <span class="n">probs</span> <span class="o">=</span> <span class="n">sorted_scores</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+                <span class="n">cumulative_probs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cumsum</span><span class="p">(</span><span class="n">probs</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+                <span class="n">mask</span> <span class="o">=</span> <span class="n">cumulative_probs</span> <span class="o">&lt;=</span> <span class="n">top_p</span>
+                <span class="k">if</span> <span class="n">mask</span><span class="o">.</span><span class="n">any</span><span class="p">():</span>
+                    <span class="n">thresholded_probs</span> <span class="o">=</span> <span class="n">probs</span> <span class="o">*</span> <span class="n">mask</span>
+                    <span class="n">thresholded_probs</span> <span class="o">=</span> <span class="n">thresholded_probs</span> <span class="o">/</span> <span class="n">thresholded_probs</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">keepdim</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                    <span class="n">final_prob</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">scores</span><span class="p">)</span>
+                    <span class="n">final_prob</span><span class="o">.</span><span class="n">scatter_add_</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">sorted_scores</span><span class="o">.</span><span class="n">argsort</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">),</span> <span class="n">thresholded_probs</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">final_prob</span> <span class="o">=</span> <span class="n">scores</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+                    
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">final_prob</span> <span class="o">=</span> <span class="n">scores</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">final_prob</span></div>
+
+    
+    
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="SpeculativeInferencer.sample">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.sample">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">sample</span><span class="p">(</span><span class="n">prob</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">num_samples</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Sample from a tensor of probabilities</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">sampled_indices</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">multinomial</span><span class="p">(</span><span class="n">prob</span><span class="p">,</span> <span class="n">num_samples</span><span class="o">=</span><span class="n">num_samples</span><span class="p">,</span> <span class="n">replacement</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span> 
+        <span class="k">return</span> <span class="p">{</span><span class="s1">&#39;sampled_token&#39;</span><span class="p">:</span> <span class="n">sampled_indices</span><span class="p">,</span> <span class="s1">&#39;sampled_prob&#39;</span><span class="p">:</span> <span class="n">prob</span><span class="o">.</span><span class="n">gather</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">index</span><span class="o">=</span><span class="n">sampled_indices</span><span class="p">),</span> <span class="s1">&#39;all_prob&#39;</span><span class="p">:</span> <span class="n">prob</span><span class="p">}</span></div>
+
+    
+    
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="SpeculativeInferencer.predict_next_token">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.predict_next_token">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">predict_next_token</span><span class="p">(</span><span class="n">model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span> <span class="n">input_ids</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">num_new_tokens</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Predict the next token given the input_ids.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span><span class="n">input_ids</span><span class="p">,</span> 
+                                 <span class="n">use_accelerator</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> 
+                                 <span class="n">max_new_tokens</span><span class="o">=</span><span class="n">num_new_tokens</span><span class="p">,</span>
+                                 <span class="n">return_dict_in_generate</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                                 <span class="n">output_scores</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                                 <span class="n">do_sample</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                                 <span class="n">num_beams</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">output</span></div>
+
+    
+    
+<div class="viewcode-block" id="SpeculativeInferencer.autoregressive_sampling">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.autoregressive_sampling">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">autoregressive_sampling</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> 
+                                <span class="n">input_ids</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> 
+                                <span class="n">model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span> 
+                                <span class="n">temperature</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.</span><span class="p">,</span> 
+                                <span class="n">num_new_tokens</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">5</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Ref: [arXiv:2211.17192v2](https://arxiv.org/abs/2211.17192) Section 2.2</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">sequence</span> <span class="o">=</span> <span class="n">input_ids</span>
+        <span class="n">new_tokens</span> <span class="o">=</span> <span class="p">[]</span>
+        
+        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_new_tokens</span><span class="p">):</span>
+            <span class="n">pred</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">predict_next_token</span><span class="p">(</span><span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span> <span class="n">input_ids</span><span class="o">=</span><span class="n">sequence</span><span class="p">,</span> <span class="n">num_new_tokens</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span> <span class="c1"># predict next one token</span>
+            <span class="n">prob</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">score_to_prob</span><span class="p">(</span><span class="n">pred</span><span class="o">.</span><span class="n">scores</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">temperature</span><span class="o">=</span><span class="n">temperature</span><span class="p">)</span>
+            <span class="n">sampled</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">sample</span><span class="p">(</span><span class="n">prob</span><span class="o">=</span><span class="n">prob</span><span class="p">,</span> <span class="n">num_samples</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+            <span class="n">new_tokens</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">sampled</span><span class="p">)</span>
+            <span class="n">sequence</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">sequence</span><span class="p">,</span> <span class="n">sampled</span><span class="p">[</span><span class="s1">&#39;sampled_token&#39;</span><span class="p">]],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+            
+        <span class="k">return</span> <span class="p">{</span><span class="s2">&quot;sequence&quot;</span><span class="p">:</span> <span class="n">sequence</span><span class="p">,</span> <span class="s2">&quot;new_tokens&quot;</span><span class="p">:</span> <span class="n">new_tokens</span><span class="p">}</span></div>
+
+    
+    
+<div class="viewcode-block" id="SpeculativeInferencer.inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span>
+        <span class="n">draft_model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span>
+        <span class="nb">input</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">temperature</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.</span><span class="p">,</span>
+        <span class="n">gamma</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">5</span><span class="p">,</span>
+        <span class="n">max_new_tokens</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">100</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform inference for a model</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        model : HFDecoderModel object.</span>
+<span class="sd">            TunableModel to verify tokens generated by the draft model.</span>
+<span class="sd">            </span>
+<span class="sd">        draft_model : HFDecoderModel object.</span>
+<span class="sd">            TunableModel that provides approximations of the target model.</span>
+
+<span class="sd">        input : str.</span>
+<span class="sd">            The input text (i.e., the prompt) for the model.</span>
+<span class="sd">            </span>
+<span class="sd">        gamma : int.</span>
+<span class="sd">            The number of tokens to be generated by the draft model within each iter.</span>
+<span class="sd">            </span>
+<span class="sd">        max_new_tokens : int.</span>
+<span class="sd">            The maximum number of tokens to be generated by the target model.</span>
+<span class="sd">            </span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        -------</span>
+<span class="sd">        output: str.</span>
+<span class="sd">            The output text generated by the model.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">assert</span> <span class="n">gamma</span> <span class="o">&gt;</span> <span class="mi">0</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;gpu&quot;</span><span class="p">:</span>
+            <span class="n">inputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;cpu&quot;</span><span class="p">:</span>
+            <span class="n">inputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="s1">&#39;cpu&#39;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;device </span><span class="se">\&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">device</span><span class="si">}</span><span class="se">\&quot;</span><span class="s2"> is not supported&quot;</span>
+            <span class="p">)</span>
+
+
+        <span class="k">def</span><span class="w"> </span><span class="nf">speculative_sampling</span><span class="p">(</span><span class="n">input_ids</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+                                 <span class="n">model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span>
+                                 <span class="n">draft_model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span>
+                                 <span class="n">temperature</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">            </span><span class="sd">&quot;&quot;&quot;Ref: [arXiv:2211.17192v2](https://arxiv.org/abs/2211.17192)</span>
+
+<span class="sd">            Parameters</span>
+<span class="sd">            ----------</span>
+<span class="sd">            input_ids : torch.Tensor</span>
+<span class="sd">            draft_model : TunableModel object</span>
+<span class="sd">            model_list : List[TunableModel object]</span>
+
+<span class="sd">            Returns</span>
+<span class="sd">            -------</span>
+<span class="sd">            torch.Tensor</span>
+<span class="sd">            &quot;&quot;&quot;</span>
+            <span class="n">len_input_ids</span><span class="o">=</span> <span class="n">input_ids</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;len of input_ids: </span><span class="si">{</span><span class="n">len_input_ids</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            
+            <span class="c1"># STEP 1: Sample γ guesses x1, ..., xγ from Mq (draft model) autoregressively</span>
+            <span class="n">output_draft</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">autoregressive_sampling</span><span class="p">(</span><span class="n">input_ids</span><span class="o">=</span><span class="n">input_ids</span><span class="p">,</span> <span class="n">model</span><span class="o">=</span><span class="n">draft_model</span><span class="p">,</span> <span class="n">num_new_tokens</span><span class="o">=</span><span class="n">gamma</span><span class="p">)</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;draft result: </span><span class="si">{</span><span class="n">output_draft</span><span class="p">[</span><span class="s1">&#39;sequence&#39;</span><span class="p">]</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;draft result decoded: </span><span class="si">{</span><span class="n">draft_model</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">output_draft</span><span class="p">[</span><span class="s1">&#39;sequence&#39;</span><span class="p">][</span><span class="mi">0</span><span class="p">])</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            
+            
+            <span class="c1"># STEP 2: Run Mp (target model) in parallel</span>
+            <span class="c1"># generate sequences [prefix, x1, x2, ..., xγ]</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()(</span><span class="n">input_ids</span><span class="o">=</span><span class="n">output_draft</span><span class="p">[</span><span class="s1">&#39;sequence&#39;</span><span class="p">],</span> <span class="n">return_dict</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;shape of output: </span><span class="si">{</span><span class="n">output</span><span class="o">.</span><span class="n">logits</span><span class="o">.</span><span class="n">shape</span><span class="si">}</span><span class="s1">&#39;</span><span class="p">)</span>
+            
+            
+            <span class="c1"># STEP 3: Determine the number of accepted guesses n</span>
+            <span class="n">accepted</span> <span class="o">=</span> <span class="p">[</span><span class="kc">False</span><span class="p">]</span> <span class="o">*</span> <span class="n">gamma</span>
+            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">gamma</span><span class="p">):</span>
+                <span class="n">draft_sampled_token_id</span> <span class="o">=</span> <span class="n">output_draft</span><span class="p">[</span><span class="s1">&#39;new_tokens&#39;</span><span class="p">][</span><span class="n">i</span><span class="p">][</span><span class="s1">&#39;sampled_token&#39;</span><span class="p">]</span>
+                <span class="n">draft_sampled_token_prob</span> <span class="o">=</span> <span class="n">output_draft</span><span class="p">[</span><span class="s1">&#39;new_tokens&#39;</span><span class="p">][</span><span class="n">i</span><span class="p">][</span><span class="s1">&#39;sampled_prob&#39;</span><span class="p">]</span>
+                <span class="n">token_prob</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">score_to_prob</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">logits</span><span class="p">[:,</span><span class="n">len_input_ids</span><span class="o">+</span><span class="n">i</span><span class="o">-</span><span class="mi">1</span><span class="p">,:],</span> <span class="n">temperature</span><span class="o">=</span><span class="n">temperature</span><span class="p">)[</span><span class="mi">0</span><span class="p">,</span> <span class="n">draft_sampled_token_id</span><span class="p">]</span>
+
+                <span class="c1"># reject the sample with probability 1 - p(x)/q(x)</span>
+                <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">rand_like</span><span class="p">(</span><span class="n">token_prob</span><span class="p">)</span> <span class="o">&gt;</span> <span class="n">token_prob</span><span class="o">/</span><span class="n">draft_sampled_token_prob</span><span class="p">:</span>
+                    <span class="k">break</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">accepted</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="kc">True</span>
+                
+            <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Speculative Sampling: Accepted: </span><span class="si">{</span><span class="nb">sum</span><span class="p">(</span><span class="n">accepted</span><span class="p">)</span><span class="si">}</span><span class="s2">/</span><span class="si">{</span><span class="n">gamma</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+
+            <span class="c1"># STEP 4: Adjust the distribution from Mp if needed</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="nb">all</span><span class="p">(</span><span class="n">accepted</span><span class="p">):</span>
+                <span class="n">all_prob</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">score_to_prob</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">logits</span><span class="p">[:,</span><span class="n">len_input_ids</span><span class="o">+</span><span class="n">i</span><span class="o">-</span><span class="mi">1</span><span class="p">,:],</span> <span class="n">temperature</span><span class="o">=</span><span class="n">temperature</span><span class="p">)</span>
+                <span class="n">draft_all_prob</span> <span class="o">=</span> <span class="n">output_draft</span><span class="p">[</span><span class="s1">&#39;new_tokens&#39;</span><span class="p">][</span><span class="n">i</span><span class="p">][</span><span class="s1">&#39;all_prob&#39;</span><span class="p">]</span>
+                <span class="n">adjusted_prob</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">all_prob</span><span class="p">),</span> <span class="n">all_prob</span> <span class="o">-</span> <span class="n">draft_all_prob</span><span class="p">)</span>
+                <span class="n">prob</span> <span class="o">=</span> <span class="n">adjusted_prob</span> <span class="o">/</span> <span class="n">adjusted_prob</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">keepdim</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">prob</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">score_to_prob</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">logits</span><span class="p">[:,</span><span class="o">-</span><span class="mi">1</span><span class="p">,:],</span> <span class="n">temperature</span><span class="o">=</span><span class="n">temperature</span><span class="p">)</span>
+
+
+            <span class="c1"># STEP 5: Return n tokens from Mq, and one token from Mp</span>
+            <span class="n">token_from_target_model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">sample</span><span class="p">(</span><span class="n">prob</span><span class="p">)[</span><span class="s1">&#39;sampled_token&#39;</span><span class="p">]</span>
+            <span class="n">final_sequence</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">concat</span><span class="p">([</span><span class="n">output_draft</span><span class="p">[</span><span class="s1">&#39;sequence&#39;</span><span class="p">][:,:</span><span class="n">len_input_ids</span><span class="o">+</span><span class="nb">sum</span><span class="p">(</span><span class="n">accepted</span><span class="p">)],</span> <span class="n">token_from_target_model</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+            <span class="k">return</span> <span class="n">final_sequence</span>
+        
+
+        <span class="n">num_generated_new_tokens</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="n">len_raw_input</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">inputs</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+        <span class="k">while</span> <span class="n">num_generated_new_tokens</span> <span class="o">&lt;</span> <span class="n">max_new_tokens</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;===== New iter =====&#39;</span><span class="p">)</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;input_ids: </span><span class="si">{</span><span class="n">inputs</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            <span class="n">sampling_result</span> <span class="o">=</span> <span class="n">speculative_sampling</span><span class="p">(</span><span class="n">input_ids</span><span class="o">=</span><span class="n">inputs</span><span class="p">,</span>
+                                                   <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span>
+                                                   <span class="n">draft_model</span><span class="o">=</span><span class="n">draft_model</span><span class="p">,</span>
+                                                   <span class="n">temperature</span><span class="o">=</span><span class="n">temperature</span><span class="p">)</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;sampling result: </span><span class="si">{</span><span class="n">sampling_result</span><span class="si">}</span><span class="s1">&#39;</span><span class="p">)</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;sampling result decoded: </span><span class="si">{</span><span class="n">model</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">sampling_result</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span><span class="si">}</span><span class="s1">&#39;</span><span class="p">)</span>
+            <span class="n">num_generated_new_tokens</span> <span class="o">+=</span> <span class="nb">len</span><span class="p">(</span><span class="n">sampling_result</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span> <span class="o">-</span> <span class="nb">len</span><span class="p">(</span><span class="n">inputs</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+            <span class="n">inputs</span> <span class="o">=</span> <span class="n">sampling_result</span>
+        
+        
+        <span class="c1"># if, say, num_generated_new_tokens = 19, and the model accept 3 </span>
+        <span class="c1"># tokens, the actual generated tokens would be 22.</span>
+        <span class="k">return</span> <span class="n">model</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">inputs</span><span class="p">[</span><span class="mi">0</span><span class="p">,:</span><span class="n">len_raw_input</span><span class="o">+</span><span class="n">max_new_tokens</span><span class="p">])</span></div>
+
+        
+
+<div class="viewcode-block" id="SpeculativeInferencer.stream_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.stream_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">stream_inference</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">&quot;Streaming output for SpeculativeInferencer is not supported yet&quot;</span><span class="p">)</span></div>
+</div>
+
+
+<div class="viewcode-block" id="ToolInferencer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.ToolInferencer">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">ToolInferencer</span><span class="p">(</span><span class="n">Inferencer</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Initializes the `ToolInferencer` class with given arguments.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+<span class="sd">    model_args : ModelArguments object.</span>
+<span class="sd">        Contains the arguments required to load the model.</span>
+
+<span class="sd">    data_args : DatasetArguments object.</span>
+<span class="sd">        Contains the arguments required to load the dataset.</span>
+
+<span class="sd">    inferencer_args : InferencerArguments object.</span>
+<span class="sd">        Contains the arguments required to perform inference.</span>
+
+
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">inferencer_args</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">inferencer_args</span><span class="p">)</span>
+
+<div class="viewcode-block" id="ToolInferencer.model">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.ToolInferencer.model">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">HFDecoderModel</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="ToolInferencer.inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.ToolInferencer.inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span>
+        <span class="nb">input</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">max_new_tokens</span><span class="p">:</span> <span class="nb">int</span><span class="o">=</span><span class="mi">1024</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform inference for a model</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        model : HFDecoderModel object.</span>
+<span class="sd">            TunableModel to perform inference</span>
+
+<span class="sd">        input : str.</span>
+<span class="sd">            The input text (i.e., the prompt) for the model. </span>
+<span class="sd">            </span>
+<span class="sd">        max_new_tokens : int.</span>
+<span class="sd">            The maximum number of tokens to be generated by the model.</span>
+
+<span class="sd">        Returns:</span>
+
+<span class="sd">        output : str.</span>
+<span class="sd">            The output text generated by the model.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;gpu&quot;</span><span class="p">:</span>
+            <span class="n">input_id</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;cpu&quot;</span><span class="p">:</span>
+            <span class="n">input_id</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="o">=</span><span class="s1">&#39;cpu&#39;</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;input_id: </span><span class="si">{</span><span class="n">input_id</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">input_length</span> <span class="o">=</span> <span class="n">input_id</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+        <span class="n">output_id</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span>
+            <span class="n">input_id</span><span class="p">,</span>
+            <span class="n">use_accelerator</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+            <span class="n">max_new_tokens</span><span class="o">=</span><span class="n">max_new_tokens</span><span class="p">,</span>
+            <span class="c1"># pad_token_id=model.tokenizer.eos_token_id,</span>
+        <span class="p">)</span>
+        <span class="c1"># logger.debug(f&quot;output: {output_id}&quot;)</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">output_id</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="nb">input</span><span class="p">,</span><span class="s2">&quot;&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">output</span></div>
+
+    
+<div class="viewcode-block" id="ToolInferencer.code_exec">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.ToolInferencer.code_exec">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">code_exec</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">code</span><span class="p">):</span>
+        <span class="c1"># Execute the code</span>
+        <span class="n">result</span> <span class="o">=</span> <span class="n">subprocess</span><span class="o">.</span><span class="n">run</span><span class="p">([</span><span class="s2">&quot;python&quot;</span><span class="p">,</span> <span class="s2">&quot;-c&quot;</span><span class="p">,</span> <span class="n">code</span><span class="p">],</span> <span class="n">capture_output</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">text</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="c1"># Print the result</span>
+        <span class="k">if</span> <span class="n">result</span><span class="o">.</span><span class="n">returncode</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Successfully Executed, the result is:&quot;</span><span class="p">)</span>
+            <span class="nb">print</span><span class="p">(</span><span class="n">result</span><span class="o">.</span><span class="n">stdout</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">result</span><span class="o">.</span><span class="n">stdout</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Error:&quot;</span><span class="p">)</span>
+            <span class="nb">print</span><span class="p">(</span><span class="n">result</span><span class="o">.</span><span class="n">stderr</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">result</span></div>
+</div>
+
+            
+        
+            
+        
+        
+        
+        
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/iterative_dpo_aligner.html b/_modules/lmflow/pipeline/iterative_dpo_aligner.html
new file mode 100644
index 000000000..ef8403cec
--- /dev/null
+++ b/_modules/lmflow/pipeline/iterative_dpo_aligner.html
@@ -0,0 +1,746 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.iterative_dpo_aligner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/iterative_dpo_aligner';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.iterative_dpo_aligner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.iterative_dpo_aligner</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span><span class="w"> </span><span class="nn">copy</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">dataclasses</span><span class="w"> </span><span class="kn">import</span> <span class="n">fields</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">gc</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">json</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">pathlib</span><span class="w"> </span><span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Optional</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">tqdm</span><span class="w"> </span><span class="kn">import</span> <span class="n">tqdm</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_text_regression_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFTextRegressionModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_decoder_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFDecoderModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets.dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.dpov2_aligner</span><span class="w"> </span><span class="kn">import</span> <span class="n">MemorySafeDPOv2Aligner</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.rm_inferencer</span><span class="w"> </span><span class="kn">import</span> <span class="n">RewardModelInferencer</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.vllm_inferencer</span><span class="w"> </span><span class="kn">import</span> <span class="n">MemorySafeVLLMInferencer</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">ModelArguments</span><span class="p">,</span> 
+    <span class="n">DatasetArguments</span><span class="p">,</span> 
+    <span class="n">InferencerArguments</span><span class="p">,</span>
+    <span class="n">IterativeDPOAlignerArguments</span><span class="p">,</span>
+    <span class="n">DPOv2AlignerArguments</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.common</span><span class="w"> </span><span class="kn">import</span> <span class="n">print_banner</span>
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="IterativeDPOAligner">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">IterativeDPOAligner</span><span class="p">:</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span>
+        <span class="n">aligner_args</span><span class="p">:</span><span class="n">IterativeDPOAlignerArguments</span><span class="p">,</span>
+        <span class="n">ref_model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">reward_model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">):</span>
+<div class="viewcode-block" id="IterativeDPOAligner.model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span> <span class="o">=</span> <span class="n">model_args</span></div>
+
+<div class="viewcode-block" id="IterativeDPOAligner.data_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.data_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span> <span class="o">=</span> <span class="n">data_args</span></div>
+
+<div class="viewcode-block" id="IterativeDPOAligner.aligner_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.aligner_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span> <span class="o">=</span> <span class="n">aligner_args</span></div>
+
+<div class="viewcode-block" id="IterativeDPOAligner.ref_model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.ref_model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ref_model_args</span> <span class="o">=</span> <span class="n">ref_model_args</span></div>
+
+<div class="viewcode-block" id="IterativeDPOAligner.reward_model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.reward_model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">reward_model_args</span> <span class="o">=</span> <span class="n">reward_model_args</span></div>
+
+<div class="viewcode-block" id="IterativeDPOAligner.workspace_path">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.workspace_path">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">workspace_path</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">)</span></div>
+
+        
+        
+<div class="viewcode-block" id="IterativeDPOAligner.align">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.align">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">align</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">dataset_list</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dataset</span><span class="p">]</span>
+    <span class="p">):</span>
+        <span class="n">num_iterations</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">dataset_list</span><span class="p">)</span>
+        
+        <span class="k">for</span> <span class="n">iter_idx</span> <span class="ow">in</span> <span class="n">tqdm</span><span class="p">(</span>
+            <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">initial_iter_idx</span><span class="p">,</span> <span class="n">num_iterations</span><span class="p">),</span> 
+            <span class="n">desc</span><span class="o">=</span><span class="s2">&quot;Iterative DPO Align&quot;</span><span class="p">,</span> 
+            <span class="n">unit</span><span class="o">=</span><span class="s2">&quot;iteration&quot;</span>
+        <span class="p">):</span>
+            <span class="k">if</span> <span class="n">iter_idx</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">target_model_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">target_model_args</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="p">)</span>
+                <span class="n">target_model_args</span><span class="o">.</span><span class="n">model_name_or_path</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">workspace_path</span><span class="o">/</span><span class="sa">f</span><span class="s2">&quot;iteration_</span><span class="si">{</span><span class="n">iter_idx</span><span class="si">}</span><span class="s2">&quot;</span><span class="o">/</span><span class="s2">&quot;model&quot;</span><span class="p">)</span>
+                
+            <span class="bp">self</span><span class="o">.</span><span class="n">_align_single_iteration</span><span class="p">(</span>
+                <span class="n">iteration_name</span><span class="o">=</span><span class="sa">f</span><span class="s2">&quot;iteration_</span><span class="si">{</span><span class="n">iter_idx</span><span class="o">+</span><span class="mi">1</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span>
+                <span class="n">target_model_args</span><span class="o">=</span><span class="n">target_model_args</span><span class="p">,</span>
+                <span class="n">reward_model_args</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">reward_model_args</span><span class="p">,</span>
+                <span class="n">ref_model_args</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">ref_model_args</span><span class="p">,</span>
+                <span class="n">dataset</span><span class="o">=</span><span class="n">dataset_list</span><span class="p">[</span><span class="n">iter_idx</span><span class="p">],</span>
+            <span class="p">)</span></div>
+
+    
+    
+<div class="viewcode-block" id="IterativeDPOAligner._align_single_iteration">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._align_single_iteration">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_align_single_iteration</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">iteration_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">target_model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">reward_model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">ref_model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">do_response_generation</span><span class="p">:</span>
+            <span class="c1"># generate responses</span>
+            <span class="n">print_banner</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;Iterative DPO </span><span class="si">{</span><span class="n">iteration_name</span><span class="si">}</span><span class="s1">: Generate responses&#39;</span><span class="p">)</span>
+            <span class="n">model</span> <span class="o">=</span> <span class="n">HFDecoderModel</span><span class="p">(</span>
+                <span class="n">model_args</span><span class="o">=</span><span class="n">target_model_args</span><span class="p">,</span>
+                <span class="n">tune_strategy</span><span class="o">=</span><span class="s1">&#39;none&#39;</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_do_target_model_inference</span><span class="p">(</span>
+                <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span>
+                <span class="n">dataset</span><span class="o">=</span><span class="n">dataset</span><span class="p">,</span>
+                <span class="n">output_dir</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">workspace_path</span><span class="o">/</span><span class="n">iteration_name</span><span class="p">),</span>
+            <span class="p">)</span>
+            <span class="k">del</span> <span class="n">model</span>
+        
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">do_scoring</span><span class="p">:</span>
+        <span class="c1"># reward model scoring</span>
+            <span class="n">print_banner</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;Iterative DPO </span><span class="si">{</span><span class="n">iteration_name</span><span class="si">}</span><span class="s1">: Reward model scoring&#39;</span><span class="p">)</span>
+            <span class="n">reward_model</span> <span class="o">=</span> <span class="n">HFTextRegressionModel</span><span class="p">(</span>
+                <span class="n">model_args</span><span class="o">=</span><span class="n">reward_model_args</span><span class="p">,</span>
+                <span class="n">tune_strategy</span><span class="o">=</span><span class="s1">&#39;none&#39;</span><span class="p">,</span>
+                <span class="n">use_accelerator</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">use_accelerator</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="n">target_model_inference_result_data_args</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">dataset</span><span class="o">.</span><span class="n">data_args</span><span class="p">)</span>
+            <span class="n">target_model_inference_result_data_args</span><span class="o">.</span><span class="n">dataset_path</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">workspace_path</span><span class="o">/</span><span class="n">iteration_name</span><span class="o">/</span><span class="s2">&quot;target_model_inference_result&quot;</span><span class="p">)</span>
+            <span class="n">target_model_inference_result_data_args</span><span class="o">.</span><span class="n">block_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">reward_model_inference_block_size</span>
+            <span class="n">target_model_inference_result_dataset</span> <span class="o">=</span> <span class="n">Dataset</span><span class="p">(</span><span class="n">target_model_inference_result_data_args</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_do_reward_model_inference</span><span class="p">(</span>
+                <span class="n">model</span><span class="o">=</span><span class="n">reward_model</span><span class="p">,</span>
+                <span class="n">dataset</span><span class="o">=</span><span class="n">target_model_inference_result_dataset</span><span class="p">,</span>
+                <span class="n">output_dir</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">workspace_path</span><span class="o">/</span><span class="n">iteration_name</span><span class="p">),</span>
+            <span class="p">)</span>
+            <span class="k">del</span> <span class="n">reward_model</span>
+        
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">do_dpo_align</span><span class="p">:</span>
+            <span class="c1"># DPO training</span>
+            <span class="n">print_banner</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;Iterative DPO </span><span class="si">{</span><span class="n">iteration_name</span><span class="si">}</span><span class="s1">: DPO training&#39;</span><span class="p">)</span>
+            <span class="n">dpo_train_data_args</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">dataset</span><span class="o">.</span><span class="n">data_args</span><span class="p">)</span>
+            <span class="n">dpo_train_data_args</span><span class="o">.</span><span class="n">dataset_path</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">workspace_path</span><span class="o">/</span><span class="n">iteration_name</span><span class="o">/</span><span class="s2">&quot;reward_model_inference_result&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_do_single_dpo_align</span><span class="p">(</span>
+                <span class="n">model_args</span><span class="o">=</span><span class="n">target_model_args</span><span class="p">,</span>
+                <span class="n">ref_model_args</span><span class="o">=</span><span class="n">ref_model_args</span><span class="p">,</span>
+                <span class="n">data_args</span><span class="o">=</span><span class="n">dpo_train_data_args</span><span class="p">,</span>
+                <span class="n">output_dir</span><span class="o">=</span><span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">workspace_path</span><span class="o">/</span><span class="n">iteration_name</span><span class="o">/</span><span class="s2">&quot;model&quot;</span><span class="p">),</span>
+                <span class="n">iteration_name</span><span class="o">=</span><span class="n">iteration_name</span><span class="p">,</span>
+            <span class="p">)</span></div>
+
+    
+    
+<div class="viewcode-block" id="IterativeDPOAligner._do_target_model_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_target_model_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_do_target_model_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">output_dir</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">result_cache_path</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">Path</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span><span class="o">/</span><span class="s2">&quot;cache&quot;</span><span class="o">/</span><span class="s2">&quot;target_model_inference_result.json&quot;</span><span class="p">)</span>
+        <span class="n">inferencer</span> <span class="o">=</span> <span class="n">MemorySafeVLLMInferencer</span><span class="p">(</span>
+            <span class="n">model_args</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">model_args</span><span class="p">,</span>
+            <span class="n">data_args</span><span class="o">=</span><span class="n">dataset</span><span class="o">.</span><span class="n">data_args</span><span class="p">,</span>
+            <span class="n">inferencer_args</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_parse_target_model_inference_args</span><span class="p">(</span>
+                <span class="n">args</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="p">,</span>
+                <span class="n">result_cache_path</span><span class="o">=</span><span class="n">result_cache_path</span><span class="p">,</span>
+            <span class="p">),</span>
+        <span class="p">)</span>
+        <span class="n">res</span> <span class="o">=</span> <span class="n">inferencer</span><span class="o">.</span><span class="n">inference</span><span class="p">()</span>
+        
+        <span class="n">dataset_out</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="s2">&quot;text_to_textlist&quot;</span><span class="p">,</span> <span class="s2">&quot;instances&quot;</span><span class="p">:</span> <span class="n">res</span><span class="p">}</span>
+            
+        <span class="n">target_model_inference_result_dir</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span><span class="o">/</span><span class="s2">&quot;target_model_inference_result&quot;</span>
+        <span class="n">target_model_inference_result_dir</span><span class="o">.</span><span class="n">mkdir</span><span class="p">(</span><span class="n">parents</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="n">json</span><span class="o">.</span><span class="n">dump</span><span class="p">(</span>
+            <span class="n">dataset_out</span><span class="p">,</span> 
+            <span class="nb">open</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">target_model_inference_result_dir</span><span class="o">/</span><span class="s2">&quot;result.json&quot;</span><span class="p">),</span> <span class="s2">&quot;w&quot;</span><span class="p">,</span> <span class="n">encoding</span><span class="o">=</span><span class="s1">&#39;utf-8&#39;</span><span class="p">),</span>
+            <span class="n">ensure_ascii</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+            <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+        
+        
+<div class="viewcode-block" id="IterativeDPOAligner._do_reward_model_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_reward_model_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_do_reward_model_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFTextRegressionModel</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">output_dir</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">inferencer</span> <span class="o">=</span> <span class="n">RewardModelInferencer</span><span class="p">(</span>
+            <span class="n">model_args</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">model_args</span><span class="p">,</span>
+            <span class="n">data_args</span><span class="o">=</span><span class="n">dataset</span><span class="o">.</span><span class="n">data_args</span><span class="p">,</span>
+            <span class="n">inferencer_args</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_parse_reward_model_inference_args</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="p">),</span>
+        <span class="p">)</span>
+        <span class="n">res</span> <span class="o">=</span> <span class="n">inferencer</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span>
+            <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span>
+            <span class="n">dataset</span><span class="o">=</span><span class="n">dataset</span><span class="p">,</span>
+            <span class="n">transform_dataset_in_place</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+            <span class="n">use_vllm</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+            <span class="n">enable_distributed_inference</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">enable_distributed_inference</span><span class="p">,</span>
+            <span class="n">distributed_inference_num_instances</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">distributed_inference_num_instances</span><span class="p">,</span>
+            <span class="n">inference_batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">reward_model_inference_batch_size</span><span class="p">,</span>
+        <span class="p">)</span>
+        
+        <span class="n">reward_model_inference_result_dir</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span><span class="o">/</span><span class="s2">&quot;reward_model_inference_result&quot;</span>
+        <span class="n">reward_model_inference_result_dir</span><span class="o">.</span><span class="n">mkdir</span><span class="p">(</span><span class="n">parents</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="n">res</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">reward_model_inference_result_dir</span><span class="o">/</span><span class="s2">&quot;result.json&quot;</span><span class="p">))</span></div>
+
+    
+    
+<div class="viewcode-block" id="IterativeDPOAligner._do_single_dpo_align">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_single_dpo_align">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_do_single_dpo_align</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">ref_model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span>
+        <span class="n">output_dir</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">iteration_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">aligner</span> <span class="o">=</span> <span class="n">MemorySafeDPOv2Aligner</span><span class="p">(</span>
+            <span class="n">model_args</span><span class="o">=</span><span class="n">model_args</span><span class="p">,</span>
+            <span class="n">data_args</span><span class="o">=</span><span class="n">data_args</span><span class="p">,</span>
+            <span class="n">aligner_args</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_parse_dpo_aligner_args</span><span class="p">(</span>
+                <span class="n">args</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span><span class="p">,</span>
+                <span class="n">output_dir</span><span class="o">=</span><span class="n">output_dir</span><span class="p">,</span>
+                <span class="n">iteration_name</span><span class="o">=</span><span class="n">iteration_name</span><span class="p">,</span>
+            <span class="p">),</span>
+            <span class="n">ref_model_args</span><span class="o">=</span><span class="n">ref_model_args</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">aligner</span><span class="o">.</span><span class="n">align</span><span class="p">()</span></div>
+
+        
+    
+<div class="viewcode-block" id="IterativeDPOAligner._parse_target_model_inference_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_target_model_inference_args">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_parse_target_model_inference_args</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">args</span><span class="p">:</span> <span class="n">IterativeDPOAlignerArguments</span><span class="p">,</span>
+        <span class="n">result_cache_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">InferencerArguments</span><span class="p">:</span>
+        <span class="n">inferencer_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__filter_args</span><span class="p">(</span>
+            <span class="n">mixed_args</span><span class="o">=</span><span class="n">args</span><span class="p">,</span>
+            <span class="n">target_cls</span><span class="o">=</span><span class="n">InferencerArguments</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">inferencer_args</span><span class="o">.</span><span class="n">save_results</span><span class="o">=</span><span class="kc">True</span>
+        <span class="n">inferencer_args</span><span class="o">.</span><span class="n">results_path</span><span class="o">=</span><span class="n">result_cache_path</span>
+        
+        <span class="k">return</span> <span class="n">inferencer_args</span></div>
+
+    
+    
+<div class="viewcode-block" id="IterativeDPOAligner._parse_reward_model_inference_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_reward_model_inference_args">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_parse_reward_model_inference_args</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">args</span><span class="p">:</span> <span class="n">IterativeDPOAlignerArguments</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">InferencerArguments</span><span class="p">:</span>
+        <span class="n">inferencer_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__filter_args</span><span class="p">(</span>
+            <span class="n">mixed_args</span><span class="o">=</span><span class="n">args</span><span class="p">,</span>
+            <span class="n">target_cls</span><span class="o">=</span><span class="n">InferencerArguments</span><span class="p">,</span>
+        <span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">inferencer_args</span></div>
+
+    
+    
+<div class="viewcode-block" id="IterativeDPOAligner._parse_dpo_aligner_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_dpo_aligner_args">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_parse_dpo_aligner_args</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">args</span><span class="p">:</span> <span class="n">IterativeDPOAlignerArguments</span><span class="p">,</span>
+        <span class="n">output_dir</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">iteration_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DPOv2AlignerArguments</span><span class="p">:</span>
+        <span class="n">aligner_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__filter_args</span><span class="p">(</span>
+            <span class="n">mixed_args</span><span class="o">=</span><span class="n">args</span><span class="p">,</span>
+            <span class="n">target_cls</span><span class="o">=</span><span class="n">DPOv2AlignerArguments</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">aligner_args</span><span class="o">.</span><span class="n">output_dir</span> <span class="o">=</span> <span class="n">output_dir</span>
+        <span class="n">aligner_args</span><span class="o">.</span><span class="n">run_name</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">args</span><span class="o">.</span><span class="n">run_name</span><span class="si">}</span><span class="s2">_</span><span class="si">{</span><span class="n">iteration_name</span><span class="si">}</span><span class="s2">&quot;</span>
+        
+        <span class="k">return</span> <span class="n">aligner_args</span></div>
+
+    
+    
+<div class="viewcode-block" id="IterativeDPOAligner.__filter_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.__filter_args">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__filter_args</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">mixed_args</span><span class="p">,</span>
+        <span class="n">target_cls</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">target_cls_fields</span> <span class="o">=</span> <span class="p">{</span><span class="n">f</span><span class="o">.</span><span class="n">name</span> <span class="k">for</span> <span class="n">f</span> <span class="ow">in</span> <span class="n">fields</span><span class="p">(</span><span class="n">target_cls</span><span class="p">)</span> <span class="k">if</span> <span class="n">f</span><span class="o">.</span><span class="n">init</span><span class="p">}</span>
+        <span class="n">common_fields</span> <span class="o">=</span> <span class="p">{</span><span class="n">f</span><span class="p">:</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">mixed_args</span><span class="p">,</span> <span class="n">f</span><span class="p">)</span> <span class="k">for</span> <span class="n">f</span> <span class="ow">in</span> <span class="n">target_cls_fields</span> <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">mixed_args</span><span class="p">,</span> <span class="n">f</span><span class="p">)}</span>
+        <span class="k">return</span> <span class="n">target_cls</span><span class="p">(</span><span class="o">**</span><span class="n">common_fields</span><span class="p">)</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/raft_aligner.html b/_modules/lmflow/pipeline/raft_aligner.html
new file mode 100644
index 000000000..12c390ec1
--- /dev/null
+++ b/_modules/lmflow/pipeline/raft_aligner.html
@@ -0,0 +1,1176 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.raft_aligner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/raft_aligner';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.raft_aligner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.raft_aligner</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">The Aligner class simplifies the process of running alignment.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">time</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">itertools</span><span class="w"> </span><span class="kn">import</span> <span class="n">chain</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.distributed</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">dist</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">transformers</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">datasets</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">set_caching_enabled</span><span class="p">,</span>
+    <span class="n">Dataset</span><span class="p">,</span>
+    <span class="n">DatasetDict</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">default_data_collator</span><span class="p">,</span>
+    <span class="n">pipeline</span><span class="p">,</span>
+    <span class="n">set_seed</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.testing_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">CaptureLogger</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="n">DatasetArguments</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets.dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span> <span class="k">as</span> <span class="n">LMFlowDataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.base_aligner</span><span class="w"> </span><span class="kn">import</span> <span class="n">BaseAligner</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.utils.raft_trainer</span><span class="w"> </span><span class="kn">import</span> <span class="n">RaftTrainer</span>
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="RaftAligner">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">RaftAligner</span><span class="p">(</span><span class="n">BaseAligner</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Initializes the `RaftAligner` class with given arguments.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+<span class="sd">    model_args : ModelArguments object.</span>
+<span class="sd">        Contains the arguments required to load the model.</span>
+<span class="sd">    </span>
+<span class="sd">    data_args : DatasetArguments object.</span>
+<span class="sd">        Contains the arguments required to load the dataset.</span>
+
+<span class="sd">    raft_aligner_args : RaftAlignerArguments object.</span>
+<span class="sd">        Contains the arguments required to perform alignment.</span>
+
+<span class="sd">    args : Optional.</span>
+<span class="sd">        Positional arguments.</span>
+<span class="sd">    </span>
+<span class="sd">    kwargs : Optional.</span>
+<span class="sd">        Keyword arguments.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">aligner_args</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<div class="viewcode-block" id="RaftAligner.model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner.model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span> <span class="o">=</span> <span class="n">model_args</span></div>
+
+<div class="viewcode-block" id="RaftAligner.data_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner.data_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span> <span class="o">=</span> <span class="n">data_args</span></div>
+
+<div class="viewcode-block" id="RaftAligner.aligner_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner.aligner_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span> <span class="o">=</span> <span class="n">aligner_args</span></div>
+
+
+        <span class="n">logging</span><span class="o">.</span><span class="n">basicConfig</span><span class="p">(</span>
+            <span class="nb">format</span><span class="o">=</span><span class="s2">&quot;</span><span class="si">%(asctime)s</span><span class="s2"> - </span><span class="si">%(levelname)s</span><span class="s2"> - </span><span class="si">%(name)s</span><span class="s2"> - </span><span class="si">%(message)s</span><span class="s2">&quot;</span><span class="p">,</span>
+            <span class="n">datefmt</span><span class="o">=</span><span class="s2">&quot;%m/</span><span class="si">%d</span><span class="s2">/%Y %H:%M:%S&quot;</span><span class="p">,</span>
+            <span class="n">handlers</span><span class="o">=</span><span class="p">[</span><span class="n">logging</span><span class="o">.</span><span class="n">StreamHandler</span><span class="p">(</span><span class="n">sys</span><span class="o">.</span><span class="n">stdout</span><span class="p">)],</span>
+        <span class="p">)</span>
+
+<div class="viewcode-block" id="RaftAligner.INF">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner.INF">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">INF</span> <span class="o">=</span> <span class="mi">888888888</span></div>
+
+        <span class="n">logger</span><span class="o">.</span><span class="n">setLevel</span><span class="p">(</span><span class="n">logging</span><span class="o">.</span><span class="n">INFO</span><span class="p">)</span>
+
+        <span class="n">output_reward_path</span> <span class="o">=</span> <span class="n">aligner_args</span><span class="o">.</span><span class="n">output_reward_path</span>
+        <span class="k">if</span> <span class="n">output_reward_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">dirname</span><span class="p">(</span><span class="n">output_reward_path</span><span class="p">),</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="c1"># Deletes a maybe-exist file</span>
+            <span class="k">try</span><span class="p">:</span>
+                <span class="n">os</span><span class="o">.</span><span class="n">remove</span><span class="p">(</span><span class="n">output_reward_path</span><span class="p">)</span>
+            <span class="k">except</span> <span class="ne">OSError</span><span class="p">:</span>
+                <span class="k">pass</span>
+
+
+<div class="viewcode-block" id="RaftAligner._initialize_trainer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._initialize_trainer">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_initialize_trainer</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">,</span> <span class="n">training_args</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        This function takes the model and tokenizer as the input and initialize the trainer.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">trainer</span> <span class="o">=</span> <span class="n">RaftTrainer</span><span class="p">(</span>
+            <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span>
+            <span class="n">args</span><span class="o">=</span><span class="n">training_args</span><span class="p">,</span>
+            <span class="n">train_dataset</span><span class="o">=</span><span class="n">Dataset</span><span class="o">.</span><span class="n">from_dict</span><span class="p">({</span><span class="s2">&quot;text&quot;</span><span class="p">:</span> <span class="p">[</span> <span class="s2">&quot; &quot;</span> <span class="p">]</span> <span class="p">}),</span>
+            <span class="n">eval_dataset</span><span class="o">=</span><span class="n">Dataset</span><span class="o">.</span><span class="n">from_dict</span><span class="p">({}),</span>
+            <span class="n">tokenizer</span><span class="o">=</span><span class="n">tokenizer</span><span class="p">,</span>
+            <span class="n">data_collator</span><span class="o">=</span><span class="n">default_data_collator</span><span class="p">,</span>
+            <span class="n">compute_metrics</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+            <span class="n">preprocess_logits_for_metrics</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">trainer</span></div>
+
+
+
+<div class="viewcode-block" id="RaftAligner._load_dataset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._load_dataset">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_load_dataset</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">selected_dataset</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">,</span>
+        <span class="n">tokenizer</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">,</span>
+        <span class="n">data_args</span><span class="p">,</span>
+        <span class="n">training_args</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&#39;&#39;&#39;</span>
+<span class="sd">        This function prepares the dataset for every iteration.</span>
+<span class="sd">        &#39;&#39;&#39;</span>
+        <span class="n">raw_datasets</span> <span class="o">=</span> <span class="n">selected_dataset</span>
+
+        <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">do_train</span><span class="p">:</span>
+            <span class="n">column_names</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">raw_datasets</span><span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">features</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">column_names</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">raw_datasets</span><span class="p">[</span><span class="s2">&quot;validation&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">features</span><span class="p">)</span>
+        <span class="n">text_column_name</span> <span class="o">=</span> <span class="s2">&quot;text&quot;</span> <span class="k">if</span> <span class="s2">&quot;text&quot;</span> <span class="ow">in</span> <span class="n">column_names</span> <span class="k">else</span> <span class="n">column_names</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="c1"># since this will be pickled to avoid _LazyModule error in Hasher force logger loading before tokenize_function</span>
+        <span class="n">tok_logger</span> <span class="o">=</span> <span class="n">transformers</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">logging</span><span class="o">.</span><span class="n">get_logger</span><span class="p">(</span><span class="s2">&quot;transformers.tokenization_utils_base&quot;</span><span class="p">)</span>
+
+        <span class="k">def</span><span class="w"> </span><span class="nf">tokenize_function</span><span class="p">(</span><span class="n">examples</span><span class="p">):</span>
+            <span class="k">with</span> <span class="n">CaptureLogger</span><span class="p">(</span><span class="n">tok_logger</span><span class="p">)</span> <span class="k">as</span> <span class="n">cl</span><span class="p">:</span>
+                <span class="n">output</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">(</span><span class="n">examples</span><span class="p">[</span><span class="n">text_column_name</span><span class="p">])</span>
+            <span class="c1"># clm input could be much much longer than block_size</span>
+            <span class="k">if</span> <span class="s2">&quot;Token indices sequence length is longer than the&quot;</span> <span class="ow">in</span> <span class="n">cl</span><span class="o">.</span><span class="n">out</span><span class="p">:</span>
+                <span class="n">tok_logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="s2">&quot;^^^^^^^^^^^^^^^^ Please ignore the warning above - this long input will be chunked into smaller bits&quot;</span>
+                    <span class="s2">&quot; before being passed to the model.&quot;</span>
+                <span class="p">)</span>
+            <span class="k">return</span> <span class="n">output</span>
+
+        <span class="k">with</span> <span class="n">training_args</span><span class="o">.</span><span class="n">main_process_first</span><span class="p">(</span><span class="n">desc</span><span class="o">=</span><span class="s2">&quot;dataset map tokenization&quot;</span><span class="p">):</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">data_args</span><span class="o">.</span><span class="n">streaming</span><span class="p">:</span>
+                <span class="n">tokenized_datasets</span> <span class="o">=</span> <span class="n">raw_datasets</span><span class="o">.</span><span class="n">map</span><span class="p">(</span>
+                    <span class="n">tokenize_function</span><span class="p">,</span>
+                    <span class="n">batched</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                    <span class="n">num_proc</span><span class="o">=</span><span class="n">data_args</span><span class="o">.</span><span class="n">preprocessing_num_workers</span><span class="p">,</span>
+                    <span class="n">remove_columns</span><span class="o">=</span><span class="n">column_names</span><span class="p">,</span>
+                    <span class="n">load_from_cache_file</span><span class="o">=</span><span class="ow">not</span> <span class="n">data_args</span><span class="o">.</span><span class="n">overwrite_cache</span><span class="p">,</span>
+                    <span class="n">desc</span><span class="o">=</span><span class="s2">&quot;Running tokenizer on dataset&quot;</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">tokenized_datasets</span> <span class="o">=</span> <span class="n">raw_datasets</span><span class="o">.</span><span class="n">map</span><span class="p">(</span>
+                    <span class="n">tokenize_function</span><span class="p">,</span>
+                    <span class="n">batched</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                    <span class="n">remove_columns</span><span class="o">=</span><span class="n">column_names</span><span class="p">,</span>
+                <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">block_size</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span>
+            <span class="k">if</span> <span class="n">block_size</span> <span class="o">&gt;</span> <span class="mi">1024</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="s2">&quot;The chosen tokenizer supports a `model_max_length` that is longer than the default `block_size` value&quot;</span>
+                    <span class="s2">&quot; of 1024. If you would like to use a longer `block_size` up to `tokenizer.model_max_length` you can&quot;</span>
+                    <span class="s2">&quot; override this default with `--block_size xxx`.&quot;</span>
+                <span class="p">)</span>
+                <span class="n">block_size</span> <span class="o">=</span> <span class="mi">512</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span> <span class="o">&gt;</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;The block_size passed (</span><span class="si">{</span><span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span><span class="si">}</span><span class="s2">) is larger than the maximum length for the model&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;(</span><span class="si">{</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="si">}</span><span class="s2">). Using block_size=</span><span class="si">{</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="si">}</span><span class="s2">.&quot;</span>
+                <span class="p">)</span>
+            <span class="n">block_size</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span><span class="p">,</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="p">)</span>
+
+        <span class="c1"># Main data processing function that will concatenate all texts from our dataset and generate chunks of block_size.</span>
+        <span class="k">def</span><span class="w"> </span><span class="nf">group_texts</span><span class="p">(</span><span class="n">examples</span><span class="p">):</span>
+            <span class="c1"># Concatenate all texts.</span>
+            <span class="n">concatenated_examples</span> <span class="o">=</span> <span class="p">{</span><span class="n">k</span><span class="p">:</span> <span class="nb">list</span><span class="p">(</span><span class="n">chain</span><span class="p">(</span><span class="o">*</span><span class="n">examples</span><span class="p">[</span><span class="n">k</span><span class="p">]))</span> <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">examples</span><span class="o">.</span><span class="n">keys</span><span class="p">()}</span>
+            <span class="n">total_length</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">concatenated_examples</span><span class="p">[</span><span class="nb">list</span><span class="p">(</span><span class="n">examples</span><span class="o">.</span><span class="n">keys</span><span class="p">())[</span><span class="mi">0</span><span class="p">]])</span>
+            <span class="c1"># We drop the small remainder, we could add padding if the model supported it instead of this drop, you can</span>
+            <span class="c1"># customize this part to your needs.</span>
+            <span class="k">if</span> <span class="n">total_length</span> <span class="o">&gt;=</span> <span class="n">block_size</span><span class="p">:</span>
+                <span class="n">total_length</span> <span class="o">=</span> <span class="p">(</span><span class="n">total_length</span> <span class="o">//</span> <span class="n">block_size</span><span class="p">)</span> <span class="o">*</span> <span class="n">block_size</span>
+            <span class="c1"># Split by chunks of max_len.</span>
+            <span class="n">result</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="n">k</span><span class="p">:</span> <span class="p">[</span><span class="n">t</span><span class="p">[</span><span class="n">i</span> <span class="p">:</span> <span class="n">i</span> <span class="o">+</span> <span class="n">block_size</span><span class="p">]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">total_length</span><span class="p">,</span> <span class="n">block_size</span><span class="p">)]</span>
+                <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">concatenated_examples</span><span class="o">.</span><span class="n">items</span><span class="p">()</span>
+            <span class="p">}</span>
+            <span class="n">result</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">result</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
+            <span class="k">return</span> <span class="n">result</span>
+
+        <span class="c1"># Note that with `batched=True`, this map processes 1,000 texts together, so group_texts throws away a remainder</span>
+        <span class="c1"># for each of those groups of 1,000 texts. You can adjust that batch_size here but a higher value might be slower</span>
+        <span class="c1"># to preprocess.</span>
+        <span class="c1">#</span>
+        <span class="c1"># To speed up this part, we use multiprocessing. See the documentation of the map method for more information:</span>
+        <span class="c1"># https://huggingface.co/docs/datasets/package_reference/main_classes.html#datasets.Dataset.map</span>
+
+        <span class="k">with</span> <span class="n">training_args</span><span class="o">.</span><span class="n">main_process_first</span><span class="p">(</span><span class="n">desc</span><span class="o">=</span><span class="s2">&quot;grouping texts together&quot;</span><span class="p">):</span>
+            <span class="n">group_batch_size</span> <span class="o">=</span> <span class="mi">1000</span>
+            <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+                <span class="n">group_batch_size</span> <span class="o">=</span> <span class="mi">1</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">data_args</span><span class="o">.</span><span class="n">streaming</span><span class="p">:</span>
+                <span class="n">lm_datasets</span> <span class="o">=</span> <span class="n">tokenized_datasets</span><span class="o">.</span><span class="n">map</span><span class="p">(</span>
+                    <span class="n">group_texts</span><span class="p">,</span>
+                    <span class="n">batched</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                    <span class="n">batch_size</span><span class="o">=</span><span class="n">group_batch_size</span><span class="p">,</span>
+                    <span class="n">num_proc</span><span class="o">=</span><span class="n">data_args</span><span class="o">.</span><span class="n">preprocessing_num_workers</span><span class="p">,</span>
+                    <span class="n">load_from_cache_file</span><span class="o">=</span><span class="ow">not</span> <span class="n">data_args</span><span class="o">.</span><span class="n">overwrite_cache</span><span class="p">,</span>
+                    <span class="n">desc</span><span class="o">=</span><span class="sa">f</span><span class="s2">&quot;Grouping texts in chunks of </span><span class="si">{</span><span class="n">block_size</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">lm_datasets</span> <span class="o">=</span> <span class="n">tokenized_datasets</span><span class="o">.</span><span class="n">map</span><span class="p">(</span>
+                    <span class="n">group_texts</span><span class="p">,</span>
+                    <span class="n">batched</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                    <span class="n">batch_size</span><span class="o">=</span><span class="n">group_batch_size</span><span class="p">,</span>
+                <span class="p">)</span>
+    
+        <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">do_train</span><span class="p">:</span>
+            <span class="k">if</span> <span class="s2">&quot;train&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">tokenized_datasets</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;--do_train requires a train dataset&quot;</span><span class="p">)</span>
+            <span class="n">train_dataset</span> <span class="o">=</span> <span class="n">lm_datasets</span><span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">]</span>
+            <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">max_train_samples</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">max_train_samples</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">),</span> <span class="n">data_args</span><span class="o">.</span><span class="n">max_train_samples</span><span class="p">)</span>
+                <span class="n">train_dataset</span> <span class="o">=</span> <span class="n">train_dataset</span><span class="o">.</span><span class="n">select</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="n">max_train_samples</span><span class="p">))</span>
+
+        <span class="k">return</span> <span class="n">train_dataset</span></div>
+
+
+
+<div class="viewcode-block" id="RaftAligner._load_input_dataset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._load_input_dataset">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_load_input_dataset</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dataset</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Load input dataset (i.e. prompt/question dataset) for training.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            dataset: A Dataset object.</span>
+<span class="sd">                The dataset to be loaded.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            dataloader (`torch.utils.data.DataLoader`):</span>
+<span class="sd">                The dataloader for the dataset.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">ds</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()</span>
+
+        <span class="k">def</span><span class="w"> </span><span class="nf">tokenize</span><span class="p">(</span><span class="n">sample</span><span class="p">):</span>
+            <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="n">sample</span><span class="p">[</span><span class="s2">&quot;text&quot;</span><span class="p">])</span>
+            <span class="n">sample</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">sample</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">])</span>
+            <span class="k">return</span> <span class="n">sample</span>
+
+        <span class="n">ds</span> <span class="o">=</span> <span class="n">ds</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="n">tokenize</span><span class="p">,</span> <span class="n">batched</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="n">ds</span> <span class="o">=</span> <span class="n">ds</span><span class="o">.</span><span class="n">filter</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">])</span> <span class="o">&lt;=</span> <span class="mi">256</span><span class="p">)</span>
+
+        <span class="n">ds</span><span class="o">.</span><span class="n">set_format</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;torch&#39;</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">ds</span></div>
+
+
+<div class="viewcode-block" id="RaftAligner._clean_text">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._clean_text">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_clean_text</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">text</span><span class="p">):</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">text</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">text</span>
+        <span class="n">stext</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">text</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;###Human&quot;</span><span class="p">)</span> <span class="k">if</span> <span class="n">x</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">stext</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span><span class="o">.</span><span class="n">strip</span><span class="p">(</span><span class="s2">&quot;#&quot;</span><span class="p">)</span> </div>
+
+
+<div class="viewcode-block" id="RaftAligner._discard_sample">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._discard_sample">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_discard_sample</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">text</span><span class="p">):</span>
+        <span class="k">if</span> <span class="s2">&quot;#&quot;</span> <span class="ow">in</span> <span class="n">text</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">True</span>
+        <span class="k">elif</span> <span class="nb">len</span><span class="p">(</span><span class="n">text</span><span class="p">)</span> <span class="o">&lt;</span> <span class="mi">2</span><span class="p">:</span> <span class="c1"># delete empty sample</span>
+            <span class="k">return</span> <span class="kc">True</span>
+        <span class="k">return</span> <span class="kc">False</span></div>
+
+
+<div class="viewcode-block" id="RaftAligner._get_batch_dataset_top">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._get_batch_dataset_top">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_get_batch_dataset_top</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">,</span>
+        <span class="n">batch_input</span><span class="p">,</span>
+        <span class="n">alpha</span><span class="o">=</span><span class="mf">0.2</span><span class="p">,</span>
+        <span class="n">iter_id</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+        <span class="n">local_rank</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+        <span class="n">output_min_length</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span>
+        <span class="n">output_max_length</span><span class="o">=</span><span class="mi">48</span><span class="p">,</span>
+        <span class="n">infer_batch_size</span><span class="o">=</span><span class="mi">8</span><span class="p">,</span>
+        <span class="n">generation_kwargs</span><span class="o">=</span><span class="p">{},</span>
+        <span class="n">tokenizer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">training_args</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">reward_model</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">output_reward_path</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        :param batch_input: input prompts</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># we will get the batch dataset via Dataset.from_dict</span>
+        <span class="n">start_time</span> <span class="o">=</span> <span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span>
+
+        <span class="n">query_tensors</span> <span class="o">=</span> <span class="n">batch_input</span><span class="p">[</span><span class="s1">&#39;input_ids&#39;</span><span class="p">]</span>
+        <span class="n">querys</span> <span class="o">=</span> <span class="n">batch_input</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">]</span>
+        <span class="n">data_size</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">querys</span><span class="p">)</span>
+
+        <span class="n">reward_eva</span> <span class="o">=</span> <span class="p">[]</span>  <span class="c1"># record the reward of the samples</span>
+        <span class="n">input_texts</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">responses</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">query_tensor</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">query_tensors</span><span class="p">):</span>
+            <span class="n">query</span> <span class="o">=</span> <span class="n">querys</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+            <span class="n">input_texts</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">query</span><span class="p">)</span>
+            <span class="k">if</span> <span class="p">(</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">%</span> <span class="n">infer_batch_size</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">or</span> <span class="p">(</span><span class="n">i</span><span class="o">+</span><span class="mi">1</span> <span class="o">==</span> <span class="n">data_size</span><span class="p">):</span>
+                <span class="n">gen_len</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">output_min_length</span><span class="p">,</span> <span class="n">output_max_length</span><span class="p">)</span>
+                <span class="n">generation_kwargs</span><span class="p">[</span><span class="s2">&quot;max_new_tokens&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">gen_len</span>
+                <span class="n">inputs</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">(</span><span class="n">input_texts</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">training_args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+                <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+                    <span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">generate</span><span class="p">(</span><span class="o">**</span><span class="n">inputs</span><span class="p">,</span> <span class="o">**</span><span class="n">generation_kwargs</span><span class="p">)</span>
+                <span class="n">generated_texts</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">batch_decode</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="n">skip_special_tokens</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                <span class="n">generated_texts</span> <span class="o">=</span> <span class="p">[</span>
+                    <span class="n">generated_text</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="n">input_texts</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="s2">&quot;&quot;</span><span class="p">)</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">generated_text</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">generated_texts</span><span class="p">)</span>
+                <span class="p">]</span>
+                <span class="n">texts_for_rewards</span> <span class="o">=</span> <span class="p">[</span><span class="n">q</span> <span class="o">+</span> <span class="n">r</span> <span class="k">for</span> <span class="n">q</span><span class="p">,</span> <span class="n">r</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">input_texts</span><span class="p">,</span> <span class="n">generated_texts</span><span class="p">)]</span>
+
+                <span class="n">texts_for_reward_dataset</span> <span class="o">=</span> <span class="n">LMFlowDataset</span><span class="o">.</span><span class="n">create_from_dict</span><span class="p">({</span>
+                    <span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="s2">&quot;text_only&quot;</span><span class="p">,</span>
+                    <span class="s2">&quot;instances&quot;</span><span class="p">:</span> <span class="p">[</span>
+                        <span class="p">{</span> <span class="s2">&quot;text&quot;</span><span class="p">:</span> <span class="n">text</span> <span class="p">}</span> <span class="k">for</span> <span class="n">text</span> <span class="ow">in</span> <span class="n">texts_for_rewards</span>
+                    <span class="p">],</span>
+                <span class="p">})</span>
+
+                <span class="n">reward_dataset</span> <span class="o">=</span> <span class="n">reward_model</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span><span class="n">texts_for_reward_dataset</span><span class="p">)</span>
+                <span class="n">rewards</span> <span class="o">=</span> <span class="p">[</span> <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;value&quot;</span><span class="p">]</span> <span class="k">for</span> <span class="n">sample</span> <span class="ow">in</span> <span class="n">reward_dataset</span><span class="o">.</span><span class="n">to_dict</span><span class="p">()[</span><span class="s2">&quot;instances&quot;</span><span class="p">]</span> <span class="p">]</span>
+
+                <span class="n">reward_eva</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">rewards</span><span class="p">)</span>
+                <span class="n">responses</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">generated_texts</span><span class="p">)</span>
+                <span class="n">input_texts</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="n">data</span> <span class="o">=</span> <span class="p">[{</span><span class="s2">&quot;input&quot;</span><span class="p">:</span> <span class="n">querys</span><span class="p">[</span><span class="n">j</span><span class="p">],</span> <span class="s2">&quot;output&quot;</span><span class="p">:</span> <span class="p">[</span><span class="n">responses</span><span class="p">[</span><span class="n">j</span><span class="p">]]}</span> <span class="k">for</span> <span class="n">j</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">reward_eva</span><span class="p">))]</span>
+
+        <span class="n">world_size</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">getenv</span><span class="p">(</span><span class="s2">&quot;WORLD_SIZE&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">))</span>
+        <span class="n">all_process_list</span> <span class="o">=</span><span class="p">[{}]</span> <span class="o">*</span> <span class="n">world_size</span>
+
+
+        <span class="n">data_to_send</span> <span class="o">=</span> <span class="p">[[</span><span class="n">data</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="n">reward_eva</span><span class="p">[</span><span class="n">i</span><span class="p">]]</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">data</span><span class="p">))]</span>
+        <span class="n">dist</span><span class="o">.</span><span class="n">all_gather_object</span><span class="p">(</span><span class="n">all_process_list</span><span class="p">,</span> <span class="n">data_to_send</span><span class="p">)</span>
+        <span class="n">gathered_data</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">gathered_reward</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">world_size</span><span class="p">):</span>
+            <span class="n">tmp_data</span> <span class="o">=</span> <span class="p">[</span><span class="n">tmp</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="k">for</span> <span class="n">tmp</span> <span class="ow">in</span> <span class="n">all_process_list</span><span class="p">[</span><span class="n">i</span><span class="p">]]</span>
+            <span class="n">gathered_data</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">tmp_data</span><span class="p">)</span>
+
+            <span class="n">tmp_reward</span> <span class="o">=</span> <span class="p">[</span><span class="n">tmp</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="k">for</span> <span class="n">tmp</span> <span class="ow">in</span> <span class="n">all_process_list</span><span class="p">[</span><span class="n">i</span><span class="p">]]</span>
+            <span class="n">gathered_reward</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">tmp_reward</span><span class="p">)</span>
+
+        <span class="n">idx</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argsort</span><span class="p">(</span><span class="n">gathered_reward</span><span class="p">)[::</span><span class="o">-</span><span class="mi">1</span><span class="p">][:</span><span class="nb">int</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">gathered_reward</span><span class="p">)</span> <span class="o">*</span> <span class="n">alpha</span><span class="p">)]</span>
+        <span class="n">gathered_data</span> <span class="o">=</span> <span class="p">[</span><span class="n">gathered_data</span><span class="p">[</span><span class="n">j</span><span class="p">]</span> <span class="k">for</span> <span class="n">j</span> <span class="ow">in</span> <span class="n">idx</span><span class="p">]</span>
+        <span class="n">reward_train</span> <span class="o">=</span> <span class="p">[</span><span class="n">gathered_reward</span><span class="p">[</span><span class="n">j</span><span class="p">]</span> <span class="k">for</span> <span class="n">j</span> <span class="ow">in</span> <span class="n">idx</span><span class="p">]</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">reward_seq</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">gathered_reward</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">train_reawrd</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">reward_train</span><span class="p">))</span>
+        <span class="kn">import</span><span class="w"> </span><span class="nn">matplotlib.pyplot</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">plt</span>
+        <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">plt</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">reward_seq</span><span class="p">,</span> <span class="n">marker</span><span class="o">=</span><span class="s2">&quot;o&quot;</span><span class="p">)</span>
+            <span class="n">plt</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">train_reawrd</span><span class="p">,</span> <span class="n">marker</span><span class="o">=</span><span class="s2">&quot;*&quot;</span><span class="p">)</span>
+            <span class="n">plt</span><span class="o">.</span><span class="n">legend</span><span class="p">([</span><span class="s2">&quot;Model reward&quot;</span><span class="p">,</span> <span class="s2">&quot;Reward of SFT Set&quot;</span><span class="p">])</span>
+            <span class="n">plt</span><span class="o">.</span><span class="n">savefig</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">store_dir</span> <span class="o">+</span> <span class="s1">&#39;/training_reward.png&#39;</span><span class="p">)</span>
+            <span class="n">plt</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
+
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;collected data of </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">gathered_data</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">([</span><span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">gathered_reward</span><span class="p">),</span> <span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">reward_train</span><span class="p">)])</span>
+
+        <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">output_reward_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">output_reward_path</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s1">&#39;a&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">fout</span><span class="p">:</span>
+                <span class="n">fout</span><span class="o">.</span><span class="n">write</span><span class="p">(</span><span class="s1">&#39;mean reward: &#39;</span> <span class="o">+</span> <span class="nb">str</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">gathered_reward</span><span class="p">))</span> <span class="o">+</span> <span class="s1">&#39;mean reward in training set: &#39;</span> <span class="o">+</span> <span class="nb">str</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">reward_train</span><span class="p">)))</span>
+                <span class="n">fout</span><span class="o">.</span><span class="n">write</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span>
+
+
+        <span class="n">prompt_structure</span> <span class="o">=</span> <span class="s2">&quot;</span><span class="si">{definition}{input}{output}</span><span class="s2">&quot;</span>
+        <span class="n">tmp_output_dataset</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;text&quot;</span><span class="p">:</span> <span class="p">[</span> <span class="n">prompt_structure</span><span class="o">.</span><span class="n">format</span><span class="p">(</span>
+                          <span class="n">definition</span><span class="o">=</span><span class="s2">&quot;&quot;</span><span class="p">,</span> <span class="nb">input</span><span class="o">=</span><span class="n">sample</span><span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">],</span> <span class="n">output</span><span class="o">=</span><span class="n">sample</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+                      <span class="p">)</span> <span class="k">for</span> <span class="n">sample</span> <span class="ow">in</span> <span class="n">gathered_data</span>
+            <span class="p">]</span>
+        <span class="p">}</span>
+
+        <span class="c1"># We store the training set for monitoring the RAFT training</span>
+        <span class="n">all_texts</span> <span class="o">=</span> <span class="n">tmp_output_dataset</span><span class="p">[</span><span class="s1">&#39;text&#39;</span><span class="p">]</span>
+        <span class="n">output_eval_dataset</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="n">output_eval_dataset</span><span class="p">[</span><span class="s1">&#39;type&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="s1">&#39;text_only&#39;</span>
+        <span class="n">output_eval_dataset</span><span class="p">[</span><span class="s1">&#39;instances&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="p">[{</span><span class="s1">&#39;text&#39;</span><span class="p">:</span> <span class="n">i_text</span><span class="p">}</span> <span class="k">for</span> <span class="n">i_text</span> <span class="ow">in</span> <span class="n">all_texts</span><span class="p">]</span>
+        <span class="kn">import</span><span class="w"> </span><span class="nn">json</span>
+        <span class="k">if</span> <span class="n">local_rank</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">store_dir</span> <span class="o">+</span> <span class="s2">&quot;/train_set_&quot;</span> <span class="o">+</span> <span class="nb">str</span><span class="p">(</span><span class="n">iter_id</span><span class="p">)</span> <span class="o">+</span> <span class="s2">&quot;.json&quot;</span><span class="p">,</span> <span class="s1">&#39;w&#39;</span><span class="p">,</span> <span class="n">encoding</span><span class="o">=</span><span class="s1">&#39;utf8&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+                <span class="n">json</span><span class="o">.</span><span class="n">dump</span><span class="p">(</span><span class="n">output_eval_dataset</span><span class="p">,</span> <span class="n">f</span><span class="p">,</span> <span class="n">ensure_ascii</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+        
+        <span class="c1"># We need to make sure that the order of the samples are the same for each agent</span>
+        <span class="n">all_process_list</span> <span class="o">=</span> <span class="p">[{}]</span> <span class="o">*</span> <span class="n">world_size</span>
+        <span class="n">data_to_send</span> <span class="o">=</span> <span class="p">[</span><span class="n">tmp_output_dataset</span><span class="p">,</span> <span class="n">local_rank</span><span class="p">]</span>
+        <span class="n">dist</span><span class="o">.</span><span class="n">all_gather_object</span><span class="p">(</span><span class="n">all_process_list</span><span class="p">,</span> <span class="n">data_to_send</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">world_size</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">all_process_list</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="mi">1</span><span class="p">]</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">output_dataset</span> <span class="o">=</span> <span class="n">all_process_list</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+                <span class="k">break</span>
+
+        <span class="k">return</span> <span class="n">DatasetDict</span><span class="p">({</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="n">Dataset</span><span class="o">.</span><span class="n">from_dict</span><span class="p">(</span><span class="n">output_dataset</span><span class="p">)</span> <span class="p">})</span></div>
+
+
+<div class="viewcode-block" id="RaftAligner._get_batch_dataset_local">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._get_batch_dataset_local">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_get_batch_dataset_local</span><span class="p">(</span>
+            <span class="bp">self</span><span class="p">,</span>
+            <span class="n">model</span><span class="p">,</span>
+            <span class="n">batch_input</span><span class="p">,</span>
+            <span class="n">K</span><span class="o">=</span><span class="mi">8</span><span class="p">,</span>
+            <span class="n">iter_id</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+            <span class="n">local_rank</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+            <span class="n">output_min_length</span><span class="o">=</span><span class="mi">16</span><span class="p">,</span>
+            <span class="n">output_max_length</span><span class="o">=</span><span class="mi">48</span><span class="p">,</span>
+            <span class="n">infer_batch_size</span><span class="o">=</span><span class="mi">8</span><span class="p">,</span>
+            <span class="n">generation_kwargs</span><span class="o">=</span><span class="p">{},</span>
+            <span class="n">tokenizer</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+            <span class="n">training_args</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+            <span class="n">reward_model</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+            <span class="n">output_reward_path</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="p">):</span>
+<span class="w">            </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">            :param batch_input: input prompts</span>
+<span class="sd">            &quot;&quot;&quot;</span>
+            <span class="c1"># we will get the batch dataset via Dataset.from_dict</span>
+            <span class="n">start_time</span> <span class="o">=</span> <span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span>
+
+            <span class="n">querys</span> <span class="o">=</span> <span class="n">batch_input</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">]</span>
+            <span class="n">data_size</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">querys</span><span class="p">)</span>
+
+            <span class="n">reward_eva</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">reward_train</span> <span class="o">=</span> <span class="p">[]</span>
+
+            <span class="n">input_texts</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">responses</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">record_querys</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">all_outputs</span> <span class="o">=</span> <span class="p">[]</span>
+
+            <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">query</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">querys</span><span class="p">):</span>
+                <span class="n">input_texts</span> <span class="o">=</span> <span class="p">[</span><span class="n">query</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">K</span><span class="p">)]</span>
+                
+                <span class="n">gen_len</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">randint</span><span class="p">(</span><span class="n">output_min_length</span><span class="p">,</span> <span class="n">output_max_length</span><span class="p">)</span>
+                <span class="n">generation_kwargs</span><span class="p">[</span><span class="s2">&quot;max_new_tokens&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">gen_len</span>
+                <span class="n">inputs</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">(</span><span class="n">input_texts</span><span class="p">,</span> <span class="n">return_tensors</span><span class="o">=</span><span class="s2">&quot;pt&quot;</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">training_args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+                <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+                    <span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">generate</span><span class="p">(</span><span class="o">**</span><span class="n">inputs</span><span class="p">,</span> <span class="o">**</span><span class="n">generation_kwargs</span><span class="p">)</span>
+                <span class="n">generated_texts</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">batch_decode</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="n">skip_special_tokens</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                <span class="n">generated_texts</span> <span class="o">=</span> <span class="p">[</span>
+                    <span class="n">generated_text</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="n">input_texts</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="s2">&quot;&quot;</span><span class="p">)</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">generated_text</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">generated_texts</span><span class="p">)</span>
+                <span class="p">]</span>
+                <span class="n">generated_texts</span> <span class="o">=</span> <span class="p">[</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">_clean_text</span><span class="p">(</span><span class="n">generated_text</span><span class="p">)</span> <span class="k">for</span> <span class="n">generated_text</span> <span class="ow">in</span> <span class="n">generated_texts</span>
+                <span class="p">]</span>
+                <span class="n">texts_for_rewards</span> <span class="o">=</span> <span class="p">[</span><span class="n">q</span> <span class="o">+</span> <span class="n">r</span> <span class="k">for</span> <span class="n">q</span><span class="p">,</span> <span class="n">r</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span><span class="n">input_texts</span><span class="p">,</span> <span class="n">generated_texts</span><span class="p">)]</span>
+
+                <span class="n">texts_for_reward_dataset</span> <span class="o">=</span> <span class="n">LMFlowDataset</span><span class="o">.</span><span class="n">create_from_dict</span><span class="p">({</span>
+                    <span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="s2">&quot;text_only&quot;</span><span class="p">,</span>
+                    <span class="s2">&quot;instances&quot;</span><span class="p">:</span> <span class="p">[</span>
+                        <span class="p">{</span> <span class="s2">&quot;text&quot;</span><span class="p">:</span> <span class="n">texts_for_rewards</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="p">}</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">texts_for_rewards</span><span class="p">))</span>
+                    <span class="p">],</span>
+                <span class="p">})</span>
+
+                <span class="n">reward_dataset</span> <span class="o">=</span> <span class="n">reward_model</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span><span class="n">texts_for_reward_dataset</span><span class="p">)</span>
+                <span class="n">rewards</span> <span class="o">=</span> <span class="p">[</span> <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;value&quot;</span><span class="p">]</span> <span class="k">for</span> <span class="n">sample</span> <span class="ow">in</span> <span class="n">reward_dataset</span><span class="o">.</span><span class="n">to_dict</span><span class="p">()[</span><span class="s2">&quot;instances&quot;</span><span class="p">]</span> <span class="p">]</span>
+                <span class="n">reward_eva</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">rewards</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+
+                <span class="c1">################################</span>
+                <span class="c1"># we impose some post-detection and discard the samples with certain criteria.</span>
+                <span class="k">for</span> <span class="n">kk</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">K</span><span class="p">):</span>
+                    <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_discard_sample</span><span class="p">(</span><span class="n">generated_texts</span><span class="p">[</span><span class="n">kk</span><span class="p">]):</span>
+                        <span class="n">rewards</span><span class="p">[</span><span class="n">kk</span><span class="p">]</span> <span class="o">=</span> <span class="o">-</span><span class="bp">self</span><span class="o">.</span><span class="n">INF</span>
+                <span class="c1">################################</span>
+                
+                <span class="n">idx_to_record</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">argmax</span><span class="p">(</span><span class="n">rewards</span><span class="p">)</span>
+                <span class="n">all_outputs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">generated_texts</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
+
+                <span class="c1"># if we discard all the samples, we do not record the sample </span>
+                <span class="k">if</span> <span class="n">rewards</span><span class="p">[</span><span class="n">idx_to_record</span><span class="p">]</span> <span class="o">!=</span> <span class="o">-</span><span class="bp">self</span><span class="o">.</span><span class="n">INF</span><span class="p">:</span>
+                    <span class="n">responses</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">generated_texts</span><span class="p">[</span><span class="n">idx_to_record</span><span class="p">])</span>
+                    <span class="n">reward_train</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">rewards</span><span class="p">[</span><span class="n">idx_to_record</span><span class="p">])</span>
+                    <span class="n">record_querys</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">query</span><span class="p">)</span>
+                <span class="n">input_texts</span> <span class="o">=</span> <span class="p">[]</span>
+
+
+            <span class="n">data</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="k">for</span> <span class="n">j</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">reward_train</span><span class="p">)):</span>
+                <span class="n">sample</span> <span class="o">=</span> <span class="p">{}</span>
+                <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">record_querys</span><span class="p">[</span><span class="n">j</span><span class="p">]</span>
+                <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="p">[</span><span class="n">responses</span><span class="p">[</span><span class="n">j</span><span class="p">]]</span>
+                <span class="n">data</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">sample</span><span class="p">)</span>
+
+
+            <span class="n">world_size</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">getenv</span><span class="p">(</span><span class="s2">&quot;WORLD_SIZE&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">))</span>
+            <span class="n">all_process_data</span> <span class="o">=</span><span class="p">[{}]</span> <span class="o">*</span> <span class="n">world_size</span>
+            <span class="n">dist</span><span class="o">.</span><span class="n">all_gather_object</span><span class="p">(</span><span class="n">all_process_data</span><span class="p">,</span> <span class="n">data</span><span class="p">)</span>
+
+            <span class="n">all_process_eval_reward</span> <span class="o">=</span><span class="p">[{}]</span> <span class="o">*</span> <span class="n">world_size</span>
+            <span class="n">dist</span><span class="o">.</span><span class="n">all_gather_object</span><span class="p">(</span><span class="n">all_process_eval_reward</span><span class="p">,</span> <span class="n">reward_eva</span><span class="p">)</span>
+            <span class="n">all_process_train_set_reward</span> <span class="o">=</span><span class="p">[{}]</span> <span class="o">*</span> <span class="n">world_size</span>
+            <span class="n">dist</span><span class="o">.</span><span class="n">all_gather_object</span><span class="p">(</span><span class="n">all_process_train_set_reward</span><span class="p">,</span> <span class="n">reward_train</span><span class="p">)</span>
+
+            
+            <span class="n">gathered_data</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">gathered_reward</span> <span class="o">=</span> <span class="p">[]</span>
+            <span class="n">gathered_train_reward</span> <span class="o">=</span> <span class="p">[]</span>
+
+            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">world_size</span><span class="p">):</span>
+                <span class="n">gathered_data</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">all_process_data</span><span class="p">[</span><span class="n">i</span><span class="p">])</span>
+                <span class="n">gathered_reward</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">all_process_eval_reward</span><span class="p">[</span><span class="n">i</span><span class="p">])</span>
+                <span class="n">gathered_train_reward</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">all_process_train_set_reward</span><span class="p">[</span><span class="n">i</span><span class="p">])</span>
+
+            <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">output_reward_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">output_reward_path</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="s1">&#39;a&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">fout</span><span class="p">:</span>
+                    <span class="n">fout</span><span class="o">.</span><span class="n">write</span><span class="p">(</span><span class="s1">&#39;mean reward: &#39;</span> <span class="o">+</span> <span class="nb">str</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">gathered_reward</span><span class="p">))</span> <span class="o">+</span> <span class="s1">&#39;mean reward in training set: &#39;</span> <span class="o">+</span> <span class="nb">str</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">gathered_train_reward</span><span class="p">)))</span>
+                    <span class="n">fout</span><span class="o">.</span><span class="n">write</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">([</span><span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">gathered_reward</span><span class="p">),</span> <span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">gathered_train_reward</span><span class="p">)])</span>
+
+            
+            <span class="bp">self</span><span class="o">.</span><span class="n">reward_seq</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">gathered_reward</span><span class="p">))</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">train_reawrd</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">reward_train</span><span class="p">))</span>
+            <span class="kn">import</span><span class="w"> </span><span class="nn">matplotlib.pyplot</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">plt</span>
+            <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">plt</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">reward_seq</span><span class="p">,</span> <span class="n">marker</span><span class="o">=</span><span class="s2">&quot;o&quot;</span><span class="p">)</span>
+                <span class="n">plt</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">train_reawrd</span><span class="p">,</span> <span class="n">marker</span><span class="o">=</span><span class="s2">&quot;*&quot;</span><span class="p">)</span>
+                <span class="n">plt</span><span class="o">.</span><span class="n">legend</span><span class="p">([</span><span class="s2">&quot;Model reward&quot;</span><span class="p">,</span> <span class="s2">&quot;Reward of SFT Set&quot;</span><span class="p">])</span>
+                <span class="n">plt</span><span class="o">.</span><span class="n">savefig</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">store_dir</span> <span class="o">+</span> <span class="s1">&#39;/training_reward.png&#39;</span><span class="p">)</span>
+                <span class="n">plt</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
+            
+
+            <span class="n">prompt_structure</span> <span class="o">=</span> <span class="s2">&quot;</span><span class="si">{definition}{input}{output}</span><span class="s2">&quot;</span>
+            <span class="n">tmp_output_dataset</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="s2">&quot;text&quot;</span><span class="p">:</span> <span class="p">[</span> <span class="n">prompt_structure</span><span class="o">.</span><span class="n">format</span><span class="p">(</span>
+                            <span class="n">definition</span><span class="o">=</span><span class="s2">&quot;&quot;</span><span class="p">,</span> <span class="nb">input</span><span class="o">=</span><span class="n">sample</span><span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">],</span> <span class="n">output</span><span class="o">=</span><span class="n">sample</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+                        <span class="p">)</span> <span class="k">for</span> <span class="n">sample</span> <span class="ow">in</span> <span class="n">gathered_data</span>
+                <span class="p">]</span>
+            <span class="p">}</span>
+
+            <span class="c1"># We store the training set for monitoring the RAFT training</span>
+            <span class="n">all_texts</span> <span class="o">=</span> <span class="n">tmp_output_dataset</span><span class="p">[</span><span class="s1">&#39;text&#39;</span><span class="p">]</span>
+            <span class="n">output_eval_dataset</span> <span class="o">=</span> <span class="p">{}</span>
+            <span class="n">output_eval_dataset</span><span class="p">[</span><span class="s1">&#39;type&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="s1">&#39;text_only&#39;</span>
+            <span class="n">output_eval_dataset</span><span class="p">[</span><span class="s1">&#39;instances&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="p">[{</span><span class="s1">&#39;text&#39;</span><span class="p">:</span> <span class="n">i_text</span><span class="p">}</span> <span class="k">for</span> <span class="n">i_text</span> <span class="ow">in</span> <span class="n">all_texts</span><span class="p">]</span>
+            <span class="kn">import</span><span class="w"> </span><span class="nn">json</span>
+            <span class="k">if</span> <span class="n">local_rank</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">store_dir</span> <span class="o">+</span> <span class="s2">&quot;/train_set_&quot;</span> <span class="o">+</span> <span class="nb">str</span><span class="p">(</span><span class="n">iter_id</span><span class="p">)</span> <span class="o">+</span> <span class="s2">&quot;.json&quot;</span><span class="p">,</span> <span class="s1">&#39;w&#39;</span><span class="p">,</span> <span class="n">encoding</span><span class="o">=</span><span class="s1">&#39;utf8&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+                    <span class="n">json</span><span class="o">.</span><span class="n">dump</span><span class="p">(</span><span class="n">output_eval_dataset</span><span class="p">,</span> <span class="n">f</span><span class="p">,</span> <span class="n">ensure_ascii</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+            
+            <span class="c1"># We need to make sure that the order of the samples are the same for each agent</span>
+            <span class="n">all_process_list</span> <span class="o">=</span> <span class="p">[{}]</span> <span class="o">*</span> <span class="n">world_size</span>
+            <span class="n">data_to_send</span> <span class="o">=</span> <span class="p">[</span><span class="n">tmp_output_dataset</span><span class="p">,</span> <span class="n">local_rank</span><span class="p">]</span>
+            <span class="n">dist</span><span class="o">.</span><span class="n">all_gather_object</span><span class="p">(</span><span class="n">all_process_list</span><span class="p">,</span> <span class="n">data_to_send</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">world_size</span><span class="p">):</span>
+                <span class="k">if</span> <span class="n">all_process_list</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="mi">1</span><span class="p">]</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">output_dataset</span> <span class="o">=</span> <span class="n">all_process_list</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span>
+                    <span class="k">break</span>
+
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;collected data of </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">output_dataset</span><span class="p">[</span><span class="s1">&#39;text&#39;</span><span class="p">])</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+
+            <span class="k">return</span> <span class="n">DatasetDict</span><span class="p">({</span> <span class="s2">&quot;train&quot;</span><span class="p">:</span> <span class="n">Dataset</span><span class="o">.</span><span class="n">from_dict</span><span class="p">(</span><span class="n">output_dataset</span><span class="p">)</span> <span class="p">})</span></div>
+
+
+
+<div class="viewcode-block" id="RaftAligner.align">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner.align">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">align</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">dataset</span><span class="p">,</span> <span class="n">reward_model</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform alignment for a model</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ------------</span>
+<span class="sd">        model : BaseModel object.</span>
+<span class="sd">        dataset: Dataset object.</span>
+<span class="sd">            Input dataset for model to generate outputs. The input and output</span>
+<span class="sd">                will then be feed into reward model to get the reward for</span>
+<span class="sd">                alignment.</span>
+<span class="sd">        reward_model: RegressionModel object.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">tokenizer</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">get_tokenizer</span><span class="p">()</span>
+        <span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token</span>
+        <span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span>
+        <span class="n">tokenizer</span><span class="o">.</span><span class="n">padding_side</span> <span class="o">=</span> <span class="s2">&quot;left&quot;</span>
+
+        <span class="n">dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_load_input_dataset</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+        <span class="n">set_caching_enabled</span><span class="p">(</span><span class="kc">False</span><span class="p">)</span>
+
+        <span class="n">wrapped_model</span> <span class="o">=</span> <span class="n">model</span>
+        <span class="n">model</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span>
+
+        <span class="n">generation_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;min_length&quot;</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span>
+            <span class="s2">&quot;top_k&quot;</span><span class="p">:</span> <span class="mf">0.0</span><span class="p">,</span>
+            <span class="s2">&quot;top_p&quot;</span><span class="p">:</span> <span class="mf">1.0</span><span class="p">,</span>
+            <span class="s2">&quot;do_sample&quot;</span><span class="p">:</span> <span class="kc">True</span><span class="p">,</span>
+            <span class="s2">&quot;pad_token_id&quot;</span><span class="p">:</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span><span class="p">,</span>
+            <span class="s2">&quot;temperature&quot;</span><span class="p">:</span><span class="mf">0.85</span><span class="p">,</span>
+        <span class="p">}</span>
+
+        <span class="n">aligner_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">aligner_args</span>
+        <span class="n">training_args</span> <span class="o">=</span> <span class="n">aligner_args</span>
+        <span class="n">model_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span>
+        <span class="n">data_args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span>
+        <span class="n">world_size</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">getenv</span><span class="p">(</span><span class="s2">&quot;WORLD_SIZE&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">))</span>
+
+
+        <span class="n">set_seed</span><span class="p">(</span><span class="mi">42</span> <span class="o">+</span> <span class="n">training_args</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>
+        <span class="n">ITERATION</span> <span class="o">=</span> <span class="n">aligner_args</span><span class="o">.</span><span class="n">num_raft_iteration</span>
+        <span class="n">collection_strategy</span> <span class="o">=</span> <span class="n">aligner_args</span><span class="o">.</span><span class="n">collection_strategy</span>
+        <span class="n">sft_batch_size</span> <span class="o">=</span> <span class="n">aligner_args</span><span class="o">.</span><span class="n">raft_batch_size</span>
+
+        <span class="k">if</span> <span class="n">collection_strategy</span> <span class="o">==</span> <span class="s2">&quot;top&quot;</span><span class="p">:</span>
+            <span class="n">alpha</span> <span class="o">=</span> <span class="n">aligner_args</span><span class="o">.</span><span class="n">top_reward_percentage</span>
+            <span class="n">M</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">sft_batch_size</span> <span class="o">/</span> <span class="n">world_size</span> <span class="o">/</span> <span class="n">alpha</span><span class="p">)</span> 
+        <span class="k">elif</span> <span class="n">collection_strategy</span> <span class="o">==</span> <span class="s2">&quot;local&quot;</span><span class="p">:</span>
+            <span class="n">K</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="mi">1</span><span class="o">/</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">top_reward_percentage</span><span class="p">)</span>
+            <span class="n">M</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">sft_batch_size</span> <span class="o">/</span> <span class="n">world_size</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">&quot;We only support two data collection strategies&quot;</span><span class="p">)</span>
+
+        <span class="nb">print</span><span class="p">(</span><span class="n">M</span><span class="p">,</span> <span class="n">K</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">training_args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="n">aligner_args</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">store_dir</span> <span class="o">=</span> <span class="n">aligner_args</span><span class="o">.</span><span class="n">output_dir</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">reward_seq</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">train_reawrd</span> <span class="o">=</span> <span class="p">[]</span>
+        
+        <span class="n">data_size</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">dataset</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">])</span>
+        <span class="n">lr</span> <span class="o">=</span> <span class="n">training_args</span><span class="o">.</span><span class="n">learning_rate</span>
+        <span class="n">random_idxs</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="n">data_size</span><span class="p">)</span>
+        <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">shuffle</span><span class="p">(</span><span class="n">random_idxs</span><span class="p">)</span>
+
+        <span class="n">raft_trainer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_initialize_trainer</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">,</span> <span class="n">training_args</span><span class="p">)</span>
+        <span class="n">raft_trainer</span><span class="o">.</span><span class="n">train</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">is_first_time</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">iteration</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">ITERATION</span><span class="p">):</span>
+            <span class="n">set_seed</span><span class="p">(</span><span class="mi">666</span> <span class="o">+</span> <span class="n">training_args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">+</span> <span class="n">world_size</span> <span class="o">*</span> <span class="p">(</span><span class="n">iteration</span><span class="o">+</span><span class="mi">1</span><span class="p">))</span>
+
+            <span class="n">end_idx</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">min</span><span class="p">([</span><span class="n">data_size</span><span class="p">,</span> <span class="p">(</span><span class="n">iteration</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="n">M</span><span class="p">])</span>
+            <span class="n">batch_input</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">select</span><span class="p">(</span><span class="n">random_idxs</span><span class="p">[</span><span class="n">iteration</span> <span class="o">*</span> <span class="n">M</span> <span class="p">:</span> <span class="n">end_idx</span><span class="p">])</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">gradient_checkpointing_disable</span><span class="p">()</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">use_cache</span> <span class="o">=</span> <span class="kc">True</span>
+
+            <span class="n">start_time</span> <span class="o">=</span> <span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span>
+            <span class="k">if</span> <span class="n">collection_strategy</span> <span class="o">==</span> <span class="s2">&quot;top&quot;</span><span class="p">:</span>
+                <span class="n">selected_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_batch_dataset_top</span><span class="p">(</span>
+                    <span class="n">raft_trainer</span><span class="o">.</span><span class="n">tmp_model</span><span class="p">,</span>
+                    <span class="n">batch_input</span><span class="p">,</span>
+                    <span class="n">alpha</span><span class="p">,</span>
+                    <span class="n">iteration</span><span class="p">,</span>
+                    <span class="n">training_args</span><span class="o">.</span><span class="n">local_rank</span><span class="p">,</span>
+                    <span class="n">output_min_length</span><span class="o">=</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">output_min_length</span><span class="p">,</span>
+                    <span class="n">output_max_length</span><span class="o">=</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">output_max_length</span><span class="p">,</span>
+                    <span class="n">infer_batch_size</span><span class="o">=</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">inference_batch_size_per_device</span><span class="p">,</span>
+                    <span class="n">generation_kwargs</span><span class="o">=</span><span class="n">generation_kwargs</span><span class="p">,</span>
+                    <span class="n">tokenizer</span><span class="o">=</span><span class="n">tokenizer</span><span class="p">,</span>
+                    <span class="n">training_args</span><span class="o">=</span><span class="n">training_args</span><span class="p">,</span>
+                    <span class="n">reward_model</span><span class="o">=</span><span class="n">reward_model</span><span class="p">,</span>
+                    <span class="n">output_reward_path</span><span class="o">=</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">output_reward_path</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">elif</span> <span class="n">collection_strategy</span> <span class="o">==</span> <span class="s2">&quot;local&quot;</span><span class="p">:</span>
+                <span class="n">selected_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_batch_dataset_local</span><span class="p">(</span>
+                    <span class="n">raft_trainer</span><span class="o">.</span><span class="n">tmp_model</span><span class="p">,</span>
+                    <span class="n">batch_input</span><span class="p">,</span>
+                    <span class="n">K</span><span class="p">,</span>
+                    <span class="n">iteration</span><span class="p">,</span>
+                    <span class="n">training_args</span><span class="o">.</span><span class="n">local_rank</span><span class="p">,</span>
+                    <span class="n">output_min_length</span><span class="o">=</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">output_min_length</span><span class="p">,</span>
+                    <span class="n">output_max_length</span><span class="o">=</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">output_max_length</span><span class="p">,</span>
+                    <span class="n">infer_batch_size</span><span class="o">=</span><span class="n">K</span><span class="p">,</span>
+                    <span class="n">generation_kwargs</span><span class="o">=</span><span class="n">generation_kwargs</span><span class="p">,</span>
+                    <span class="n">tokenizer</span><span class="o">=</span><span class="n">tokenizer</span><span class="p">,</span>
+                    <span class="n">training_args</span><span class="o">=</span><span class="n">training_args</span><span class="p">,</span>
+                    <span class="n">reward_model</span><span class="o">=</span><span class="n">reward_model</span><span class="p">,</span>
+                    <span class="n">output_reward_path</span><span class="o">=</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">output_reward_path</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="n">end_time</span> <span class="o">=</span> <span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="s2">&quot;It takes </span><span class="si">%.2f</span><span class="s2"> s to inference one stage&quot;</span><span class="p">,</span> <span class="n">end_time</span> <span class="o">-</span> <span class="n">start_time</span><span class="p">)</span>
+            
+            <span class="n">raft_trainer</span><span class="o">.</span><span class="n">train_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_load_dataset</span><span class="p">(</span>
+                <span class="n">selected_dataset</span><span class="p">,</span>
+                <span class="n">raft_trainer</span><span class="o">.</span><span class="n">tmp_model</span><span class="p">,</span>
+                <span class="n">tokenizer</span><span class="p">,</span>
+                <span class="n">model_args</span><span class="p">,</span>
+                <span class="n">data_args</span><span class="p">,</span>
+                <span class="n">training_args</span><span class="p">,</span>
+            <span class="p">)</span>
+
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;iter </span><span class="si">{</span><span class="n">iteration</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            <span class="n">start_time</span> <span class="o">=</span> <span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">gradient_checkpointing_enable</span><span class="p">()</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">use_cache</span> <span class="o">=</span> <span class="kc">False</span>
+
+
+            <span class="n">train_result</span> <span class="o">=</span> <span class="n">raft_trainer</span><span class="o">.</span><span class="n">train</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+            <span class="n">end_time</span> <span class="o">=</span> <span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="s2">&quot;It takes </span><span class="si">%.2f</span><span class="s2"> s to train one stage&quot;</span><span class="p">,</span> <span class="n">end_time</span> <span class="o">-</span> <span class="n">start_time</span><span class="p">)</span>
+            <span class="k">if</span> <span class="p">(</span><span class="n">iteration</span><span class="o">+</span><span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="n">M</span>  <span class="o">&gt;</span> <span class="n">data_size</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="s2">&quot;One epoch is completed.&quot;</span><span class="p">)</span>
+                <span class="k">break</span>
+
+<span class="w">            </span><span class="sd">&#39;&#39;&#39;</span>
+<span class="sd">            if training_args.local_rank == 0 and iteration % 2 == 0:</span>
+<span class="sd">                wrapped_model.save(aligner_args.output_dir + &quot;/&quot; + &quot;model&quot; + str(iteration))</span>
+<span class="sd">                print(iteration, &quot;I save a model with&quot;, self.reward_seq[-1])</span>
+<span class="sd">            &#39;&#39;&#39;</span>
+
+        <span class="k">if</span> <span class="n">aligner_args</span><span class="o">.</span><span class="n">output_dir</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">wrapped_model</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">wrapped_model</span> </div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/rm_inferencer.html b/_modules/lmflow/pipeline/rm_inferencer.html
new file mode 100644
index 000000000..49e479511
--- /dev/null
+++ b/_modules/lmflow/pipeline/rm_inferencer.html
@@ -0,0 +1,845 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.rm_inferencer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/rm_inferencer';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.rm_inferencer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.rm_inferencer</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">copy</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">wandb</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">deepspeed</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">datetime</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">json</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">time</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Any</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">accelerate</span><span class="w"> </span><span class="kn">import</span> <span class="n">Accelerator</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">tqdm</span><span class="w"> </span><span class="kn">import</span> <span class="n">tqdm</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">AutoConfig</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.modeling_outputs</span><span class="w"> </span><span class="kn">import</span> <span class="n">SequenceClassifierOutputWithPast</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.distributed</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">dist</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.nn.functional</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">F</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">DatasetArguments</span><span class="p">,</span>
+    <span class="n">ModelArguments</span><span class="p">,</span>
+    <span class="n">InferencerArguments</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets.dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_text_regression_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFTextRegressionModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.base_pipeline</span><span class="w"> </span><span class="kn">import</span> <span class="n">BasePipeline</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.data_utils</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">set_random_seed</span><span class="p">,</span>
+    <span class="n">batchlize</span><span class="p">,</span>
+    <span class="n">RewardModelInferenceResultWithInput</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets.dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">KEY_SCORE</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.versioning</span><span class="w"> </span><span class="kn">import</span> <span class="n">is_ray_available</span>
+
+<span class="k">if</span> <span class="n">is_ray_available</span><span class="p">():</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">ray</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">ray.data</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">ray.util.scheduling_strategies</span><span class="w"> </span><span class="kn">import</span> <span class="n">PlacementGroupSchedulingStrategy</span>
+
+
+<span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="p">[</span><span class="s2">&quot;TOKENIZERS_PARALLELISM&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;false&quot;</span>  <span class="c1"># To avoid warnings about parallelism in tokenizers</span>
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="RewardModelInferencer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">RewardModelInferencer</span><span class="p">(</span><span class="n">BasePipeline</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Initializes the `Inferencer` class with given arguments.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+<span class="sd">    model_args : ModelArguments object.</span>
+<span class="sd">        Contains the arguments required to load the model.</span>
+
+<span class="sd">    data_args : DatasetArguments object.</span>
+<span class="sd">        Contains the arguments required to load the dataset.</span>
+
+<span class="sd">    inferencer_args : InferencerArguments object.</span>
+<span class="sd">        Contains the arguments required to perform inference.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span> 
+        <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span> 
+        <span class="n">inferencer_args</span><span class="p">:</span> <span class="n">InferencerArguments</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">):</span>
+<div class="viewcode-block" id="RewardModelInferencer.data_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.data_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span> <span class="o">=</span> <span class="n">data_args</span></div>
+
+<div class="viewcode-block" id="RewardModelInferencer.inferencer_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.inferencer_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span> <span class="o">=</span> <span class="n">inferencer_args</span></div>
+
+<div class="viewcode-block" id="RewardModelInferencer.model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span> <span class="o">=</span> <span class="n">model_args</span></div>
+
+
+        <span class="n">set_random_seed</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">random_seed</span><span class="p">)</span>
+
+<div class="viewcode-block" id="RewardModelInferencer.local_rank">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.local_rank">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">getenv</span><span class="p">(</span><span class="s2">&quot;LOCAL_RANK&quot;</span><span class="p">,</span> <span class="s2">&quot;0&quot;</span><span class="p">))</span></div>
+
+<div class="viewcode-block" id="RewardModelInferencer.world_size">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.world_size">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">world_size</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">getenv</span><span class="p">(</span><span class="s2">&quot;WORLD_SIZE&quot;</span><span class="p">,</span> <span class="s2">&quot;1&quot;</span><span class="p">))</span></div>
+
+        <span class="k">if</span> <span class="n">inferencer_args</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;gpu&quot;</span><span class="p">:</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">set_device</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">)</span>  <span class="c1"># NOTE: cpu-only machine will have error</span>
+            <span class="n">deepspeed</span><span class="o">.</span><span class="n">init_distributed</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">dist</span><span class="o">.</span><span class="n">init_process_group</span><span class="p">(</span>
+                <span class="s2">&quot;gloo&quot;</span><span class="p">,</span> <span class="n">rank</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">local_rank</span><span class="p">,</span> <span class="n">world_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">world_size</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">inferencer_args</span><span class="o">.</span><span class="n">use_accelerator</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="p">:</span> <span class="n">Accelerator</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">&#39;accelerator&#39;</span><span class="p">,</span> <span class="n">Accelerator</span><span class="p">())</span>
+            
+
+<div class="viewcode-block" id="RewardModelInferencer.inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFTextRegressionModel</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+        <span class="n">transform_dataset_in_place</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">use_vllm</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">enable_distributed_inference</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dataset</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">use_vllm</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;VLLM doesn&#39;t support reward model inference, using normal inference instead.&quot;</span><span class="p">)</span>
+            <span class="n">use_vllm</span> <span class="o">=</span> <span class="kc">False</span>
+            
+        <span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">HFTextRegressionModel</span><span class="p">),</span> <span class="s2">&quot;model should be HFTextRegressionModel&quot;</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">transform_dataset_in_place</span><span class="p">:</span>
+            <span class="n">dataset</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span>
+            
+        <span class="n">model_input</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">prepare_inputs_for_inference</span><span class="p">(</span>
+            <span class="n">dataset</span><span class="o">=</span><span class="n">dataset</span><span class="p">,</span>
+            <span class="n">apply_chat_template</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+            <span class="n">enable_distributed_inference</span><span class="o">=</span><span class="n">enable_distributed_inference</span><span class="p">,</span>
+            <span class="n">use_vllm</span><span class="o">=</span><span class="n">use_vllm</span>
+        <span class="p">)</span>
+            
+        <span class="k">if</span> <span class="n">use_vllm</span><span class="p">:</span>
+            <span class="n">inference_result</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__vllm_inference</span><span class="p">(</span>
+                <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span> 
+                <span class="n">model_input</span><span class="o">=</span><span class="n">model_input</span><span class="p">,</span>
+                <span class="n">enable_distributed_inference</span><span class="o">=</span><span class="n">enable_distributed_inference</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">inference_result</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_inference</span><span class="p">(</span>
+                <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span>
+                <span class="n">model_input</span><span class="o">=</span><span class="n">model_input</span><span class="p">,</span>
+                <span class="n">enable_distributed_inference</span><span class="o">=</span><span class="n">enable_distributed_inference</span><span class="p">,</span>
+                <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+            <span class="p">)</span>
+        
+        <span class="k">if</span> <span class="n">enable_distributed_inference</span><span class="p">:</span>
+            <span class="n">output_dataset</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">postprocess_distributed_inference_outputs</span><span class="p">(</span>
+                <span class="n">dataset</span><span class="o">=</span><span class="n">dataset</span><span class="p">,</span>
+                <span class="n">inference_result</span><span class="o">=</span><span class="n">inference_result</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">output_dataset</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">postprocess_inference_outputs</span><span class="p">(</span>
+                <span class="n">dataset</span><span class="o">=</span><span class="n">dataset</span><span class="p">,</span> 
+                <span class="n">scores</span><span class="o">=</span><span class="n">inference_result</span>
+            <span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">output_dataset</span></div>
+
+    
+    
+<div class="viewcode-block" id="RewardModelInferencer._inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer._inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFTextRegressionModel</span><span class="p">,</span>
+        <span class="n">model_input</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">Dataset</span><span class="p">,</span> <span class="s1">&#39;ray.data.Dataset&#39;</span><span class="p">],</span>
+        <span class="n">enable_distributed_inference</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">if</span> <span class="n">enable_distributed_inference</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">is_ray_available</span><span class="p">():</span>
+                <span class="k">raise</span> <span class="ne">ImportError</span><span class="p">(</span><span class="s1">&#39;Ray is not installed. Please install via `pip install -e &quot;.[ray]&quot;`.&#39;</span><span class="p">)</span>
+            
+            <span class="n">inference_res</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__distributed_inference</span><span class="p">(</span>
+                <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span> 
+                <span class="n">model_input</span><span class="o">=</span><span class="n">model_input</span><span class="p">,</span> 
+                <span class="n">num_instances</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;distributed_inference_num_instances&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+                <span class="n">batch_size</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;inference_batch_size&quot;</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">inference_res</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__inference</span><span class="p">(</span>
+                <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span> 
+                <span class="n">model_input</span><span class="o">=</span><span class="n">model_input</span><span class="p">,</span>
+            <span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">inference_res</span></div>
+
+
+
+<div class="viewcode-block" id="RewardModelInferencer.__inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFTextRegressionModel</span><span class="p">,</span>
+        <span class="n">model_input</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]]]:</span>
+        <span class="k">if</span> <span class="n">model_input</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;text_to_textlist&quot;</span><span class="p">]:</span>
+            <span class="n">model_input_ids</span><span class="p">,</span> <span class="n">num_outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">flatten_list</span><span class="p">(</span><span class="n">model_input</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">])</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">model_input_ids</span> <span class="o">=</span> <span class="n">model_input</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span>
+            
+        <span class="n">dataloader</span> <span class="o">=</span> <span class="n">batchlize</span><span class="p">(</span>
+            <span class="n">examples</span><span class="o">=</span><span class="n">model_input_ids</span><span class="p">,</span>
+            <span class="n">batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">inference_batch_size</span><span class="p">,</span>
+            <span class="n">random_shuffle</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="c1"># DO NOT shuffle when inference</span>
+        <span class="p">)</span>
+        <span class="n">num_batches</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">dataloader</span><span class="p">)</span>
+        <span class="n">final_output</span> <span class="o">=</span> <span class="p">[]</span>
+        
+        <span class="k">for</span> <span class="n">batch_index</span><span class="p">,</span> <span class="n">batched_input_ids</span> <span class="ow">in</span> <span class="n">tqdm</span><span class="p">(</span>
+            <span class="n">iterable</span><span class="o">=</span><span class="nb">enumerate</span><span class="p">(</span><span class="n">dataloader</span><span class="p">),</span> 
+            <span class="n">total</span><span class="o">=</span><span class="n">num_batches</span><span class="p">,</span> 
+            <span class="n">desc</span><span class="o">=</span><span class="s2">&quot;Inference&quot;</span><span class="p">,</span> 
+            <span class="n">unit</span><span class="o">=</span><span class="s2">&quot;batch&quot;</span>
+        <span class="p">):</span>
+            <span class="c1"># len(batch) = batch_size, and batch element is dataset sample</span>
+            <span class="n">model_input_tensor</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">(</span><span class="n">batched_input_ids</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="s2">&quot;cpu&quot;</span> <span class="k">if</span> <span class="n">model</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="s2">&quot;cpu&quot;</span> <span class="k">else</span> <span class="s2">&quot;cuda&quot;</span><span class="p">)</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">use_accelerator</span><span class="p">:</span>
+                <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="o">.</span><span class="n">autocast</span><span class="p">():</span>
+                    <span class="n">batch_output</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span>
+                        <span class="n">inputs</span><span class="o">=</span><span class="n">model_input_tensor</span><span class="p">,</span> 
+                        <span class="n">use_vllm</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                    <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">batch_output</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span>
+                    <span class="n">inputs</span><span class="o">=</span><span class="n">model_input_tensor</span><span class="p">,</span> 
+                    <span class="n">use_vllm</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                <span class="p">)</span>
+            
+            <span class="n">batch_output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">__post_process_model_output</span><span class="p">(</span><span class="n">batch_output</span><span class="p">)</span>
+            <span class="n">final_output</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">batch_output</span><span class="p">)</span>
+        
+        <span class="k">if</span> <span class="n">model_input</span><span class="o">.</span><span class="n">get_type</span><span class="p">()</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;text_to_textlist&quot;</span><span class="p">]:</span>
+            <span class="n">final_output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">compress_list</span><span class="p">(</span><span class="n">final_output</span><span class="p">,</span> <span class="n">num_outputs</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">final_output</span></div>
+
+    
+    
+<div class="viewcode-block" id="RewardModelInferencer.__distributed_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__distributed_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__distributed_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFTextRegressionModel</span><span class="p">,</span>
+        <span class="n">model_input</span><span class="p">:</span> <span class="s1">&#39;ray.data.Dataset&#39;</span><span class="p">,</span>
+        <span class="n">num_instances</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">batch_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">RewardModelInferenceResultWithInput</span><span class="p">]:</span>
+        <span class="k">def</span><span class="w"> </span><span class="nf">scheduling_strategy_fn</span><span class="p">():</span>
+            <span class="c1"># One bundle per tensor parallel worker</span>
+            <span class="n">pg</span> <span class="o">=</span> <span class="n">ray</span><span class="o">.</span><span class="n">util</span><span class="o">.</span><span class="n">placement_group</span><span class="p">(</span>
+                <span class="p">[{</span>
+                    <span class="s2">&quot;GPU&quot;</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span>
+                    <span class="s2">&quot;CPU&quot;</span><span class="p">:</span> <span class="mi">1</span>
+                <span class="p">}]</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">tensor_parallel_size</span><span class="p">,</span>
+                <span class="n">strategy</span><span class="o">=</span><span class="s2">&quot;STRICT_PACK&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span>
+                <span class="n">scheduling_strategy</span><span class="o">=</span><span class="n">PlacementGroupSchedulingStrategy</span><span class="p">(</span>
+                    <span class="n">pg</span><span class="p">,</span> <span class="n">placement_group_capture_child_tasks</span><span class="o">=</span><span class="kc">True</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+            
+        <span class="n">resources_kwarg</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">tensor_parallel_size</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="c1"># For tensor_parallel_size == 1, we simply set num_gpus=1.</span>
+            <span class="n">resources_kwarg</span><span class="p">[</span><span class="s2">&quot;num_gpus&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">1</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># Otherwise, we have to set num_gpus=0 and provide</span>
+            <span class="c1"># a function that will create a placement group for</span>
+            <span class="c1"># each instance.</span>
+            <span class="n">resources_kwarg</span><span class="p">[</span><span class="s2">&quot;num_gpus&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+            <span class="n">resources_kwarg</span><span class="p">[</span><span class="s2">&quot;ray_remote_args_fn&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">scheduling_strategy_fn</span>
+            
+        <span class="c1">## predictor</span>
+        <span class="k">class</span><span class="w"> </span><span class="nc">DistributedPredictor</span><span class="p">:</span>
+            <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+                <span class="bp">self</span><span class="p">,</span> 
+                <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+            <span class="p">):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">HFTextRegressionModel</span><span class="p">(</span>
+                    <span class="n">model_args</span><span class="o">=</span><span class="n">model_args</span><span class="p">,</span> 
+                    <span class="n">tune_strategy</span><span class="o">=</span><span class="s1">&#39;none&#39;</span><span class="p">,</span> 
+                    <span class="n">use_accelerator</span><span class="o">=</span><span class="kc">True</span>
+                <span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">activate_model_for_inference</span><span class="p">(</span><span class="n">use_vllm</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+                
+            <span class="k">def</span><span class="w"> </span><span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]):</span>
+<span class="w">                </span><span class="sd">&quot;&quot;&quot;batch: Dict[str, np.ndarray]</span>
+<span class="sd">                Example (batch size=2):</span>
+<span class="sd">                {&#39;input&#39;: array([&#39;...&#39;,&#39;...&#39;], dtype=object),</span>
+<span class="sd">                 &#39;output&#39;: array([array([&quot;...&quot;, &quot;...&quot;], dtype=object), array([&#39;...&#39;,&#39;...&#39;], dtype=object)], dtype=object),</span>
+<span class="sd">                 &#39;input_ids&#39;: array([[[128000, 128006,    882, ..., 128256, 128256, 128256],</span>
+<span class="sd">                         [128000, 128006,    882, ..., 128256, 128256, 128256]],</span>
+<span class="sd">                        [[128000, 128006,    882, ..., 128256, 128256, 128256],</span>
+<span class="sd">                         [128000, 128006,    882, ..., 128256, 128256, 128256]]])}</span>
+<span class="sd">                &quot;&quot;&quot;</span>
+                <span class="c1"># The batch is managed by ray and the actual batch size may smaller than </span>
+                <span class="c1"># inference_batch_size in config, since there may be some remainders. </span>
+                <span class="c1"># For example, 10 examples with 2 inference instances and inference_batch_size=4,</span>
+                <span class="c1"># there will be only 2 examples for instance 0 to run and then the </span>
+                <span class="c1"># actual batch size changes.</span>
+                <span class="n">actual_batch_size</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">])</span>
+                <span class="n">input_tensor</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">([</span>
+                    <span class="p">[</span><span class="nb">list</span><span class="p">(</span><span class="n">arr</span><span class="p">)</span> <span class="k">for</span> <span class="n">arr</span> <span class="ow">in</span> <span class="n">batch</span><span class="p">[</span><span class="s1">&#39;input_ids&#39;</span><span class="p">][</span><span class="n">batch_idx</span><span class="p">]]</span> 
+                    <span class="k">for</span> <span class="n">batch_idx</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">actual_batch_size</span><span class="p">)</span>
+                <span class="p">])</span><span class="o">.</span><span class="n">flatten</span><span class="p">(</span><span class="n">start_dim</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">end_dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="s2">&quot;cuda&quot;</span><span class="p">)</span>
+                <span class="n">batched_inference_res</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span><span class="n">input_tensor</span><span class="p">)</span><span class="o">.</span><span class="n">logits</span>
+                <span class="n">batched_inference_res</span> <span class="o">=</span> <span class="n">batched_inference_res</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="s2">&quot;cpu&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">actual_batch_size</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span> 
+                <span class="c1"># [bs, num_output_sequences]</span>
+                <span class="n">batched_final_res</span> <span class="o">=</span> <span class="p">{</span>
+                    <span class="s2">&quot;input&quot;</span><span class="p">:</span> <span class="n">batch</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">tolist</span><span class="p">(),</span>
+                    <span class="s2">&quot;output&quot;</span><span class="p">:</span> <span class="p">[</span>
+                        <span class="p">[</span>
+                            <span class="p">{</span><span class="s2">&quot;score&quot;</span><span class="p">:</span> <span class="n">batched_inference_res</span><span class="p">[</span><span class="n">j</span><span class="p">][</span><span class="n">i</span><span class="p">],</span> <span class="s2">&quot;text&quot;</span><span class="p">:</span> <span class="n">batch</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">][</span><span class="n">j</span><span class="p">][</span><span class="n">i</span><span class="p">]}</span>
+                            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">batch</span><span class="p">[</span><span class="s1">&#39;output&#39;</span><span class="p">][</span><span class="n">j</span><span class="p">]))</span>
+                        <span class="p">]</span> 
+                        <span class="k">for</span> <span class="n">j</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">actual_batch_size</span><span class="p">)</span>
+                    <span class="p">],</span>
+                <span class="p">}</span> <span class="c1"># do this since we&#39;re writing to a pandas dataframe</span>
+                <span class="k">return</span> <span class="n">batched_final_res</span>
+
+        <span class="c1"># inference</span>
+        <span class="n">model_input_mapping</span> <span class="o">=</span> <span class="n">model_input</span><span class="o">.</span><span class="n">map_batches</span><span class="p">(</span>
+            <span class="n">DistributedPredictor</span><span class="p">,</span>
+            <span class="n">concurrency</span><span class="o">=</span><span class="n">num_instances</span><span class="p">,</span> <span class="c1"># Set the concurrency to the number of LLM instances.</span>
+            <span class="n">batch_size</span><span class="o">=</span><span class="n">batch_size</span><span class="p">,</span>
+            <span class="n">fn_constructor_kwargs</span><span class="o">=</span><span class="p">{</span>
+                <span class="s2">&quot;model_args&quot;</span><span class="p">:</span> <span class="n">model</span><span class="o">.</span><span class="n">model_args</span><span class="p">,</span>
+            <span class="p">},</span>
+            <span class="o">**</span><span class="n">resources_kwarg</span><span class="p">,</span>
+        <span class="p">)</span>
+        
+        <span class="n">df_model_output</span> <span class="o">=</span> <span class="n">model_input_mapping</span><span class="o">.</span><span class="n">to_pandas</span><span class="p">()</span> <span class="c1"># the actual forwards are executed here</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Distributed reward model inference result preview:</span><span class="se">\n</span><span class="si">{</span><span class="n">df_model_output</span><span class="o">.</span><span class="n">head</span><span class="p">(</span><span class="mi">10</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        
+        <span class="n">model_output</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="p">{</span><span class="s2">&quot;input&quot;</span><span class="p">:</span> <span class="n">row</span><span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">],</span> <span class="s2">&quot;output&quot;</span><span class="p">:</span> <span class="n">row</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">]}</span> <span class="k">for</span> <span class="n">_</span><span class="p">,</span> <span class="n">row</span> <span class="ow">in</span> <span class="n">df_model_output</span><span class="p">[:]</span><span class="o">.</span><span class="n">iterrows</span><span class="p">()</span>
+        <span class="p">]</span>
+        
+        <span class="k">return</span> <span class="n">model_output</span></div>
+
+    
+    
+<div class="viewcode-block" id="RewardModelInferencer.__vllm_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__vllm_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__vllm_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFTextRegressionModel</span><span class="p">,</span>
+        <span class="n">model_input</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span>
+        <span class="n">enable_distributed_inference</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]:</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">&quot;VLLM inference for reward model is not implemented yet.&quot;</span><span class="p">)</span></div>
+
+        
+    
+<div class="viewcode-block" id="RewardModelInferencer.__post_process_model_output">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__post_process_model_output">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__post_process_model_output</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_output</span><span class="p">:</span> <span class="n">SequenceClassifierOutputWithPast</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">float</span><span class="p">]:</span>
+        <span class="n">final_output</span> <span class="o">=</span> <span class="n">model_output</span><span class="o">.</span><span class="n">logits</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="s2">&quot;cpu&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">tolist</span><span class="p">()</span>
+        
+        <span class="k">return</span> <span class="n">final_output</span></div>
+
+            
+    
+<div class="viewcode-block" id="RewardModelInferencer.flatten_list">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.flatten_list">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">flatten_list</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">list_of_list</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">]</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]:</span>
+        <span class="n">sublist_lengths</span> <span class="o">=</span> <span class="p">[</span><span class="nb">len</span><span class="p">(</span><span class="n">sublist</span><span class="p">)</span> <span class="k">for</span> <span class="n">sublist</span> <span class="ow">in</span> <span class="n">list_of_list</span><span class="p">]</span>
+        <span class="n">flattened_list</span> <span class="o">=</span> <span class="p">[</span><span class="n">item</span> <span class="k">for</span> <span class="n">sublist</span> <span class="ow">in</span> <span class="n">list_of_list</span> <span class="k">for</span> <span class="n">item</span> <span class="ow">in</span> <span class="n">sublist</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">flattened_list</span><span class="p">,</span> <span class="n">sublist_lengths</span></div>
+
+    
+
+<div class="viewcode-block" id="RewardModelInferencer.compress_list">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.compress_list">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">compress_list</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">list_to_compress</span><span class="p">:</span> <span class="n">List</span><span class="p">,</span> 
+        <span class="n">sublist_lengths</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">]:</span>
+        <span class="k">assert</span> <span class="nb">sum</span><span class="p">(</span><span class="n">sublist_lengths</span><span class="p">)</span> <span class="o">==</span> <span class="nb">len</span><span class="p">(</span><span class="n">list_to_compress</span><span class="p">),</span> <span class="s2">&quot;Sum of sublist lengths should be equal to length of list to compress.&quot;</span>
+        <span class="n">compressed_list</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">start_index</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">for</span> <span class="n">length</span> <span class="ow">in</span> <span class="n">sublist_lengths</span><span class="p">:</span>
+            <span class="n">sublist</span> <span class="o">=</span> <span class="n">list_to_compress</span><span class="p">[</span><span class="n">start_index</span><span class="p">:</span> <span class="n">start_index</span> <span class="o">+</span> <span class="n">length</span><span class="p">]</span>
+            <span class="n">compressed_list</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">sublist</span><span class="p">)</span>
+            <span class="n">start_index</span> <span class="o">+=</span> <span class="n">length</span>
+        <span class="k">return</span> <span class="n">compressed_list</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/rm_tuner.html b/_modules/lmflow/pipeline/rm_tuner.html
new file mode 100644
index 000000000..6bbe2b89b
--- /dev/null
+++ b/_modules/lmflow/pipeline/rm_tuner.html
@@ -0,0 +1,683 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.rm_tuner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/rm_tuner';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.rm_tuner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.rm_tuner</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optional</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">copy</span><span class="w"> </span><span class="kn">import</span> <span class="n">deepcopy</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">datasets</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">transformers</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">set_seed</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">send_example_telemetry</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.trainer_callback</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">TrainerCallback</span>
+<span class="p">)</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_text_regression_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFTextRegressionModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.finetuner</span><span class="w"> </span><span class="kn">import</span> <span class="n">Finetuner</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.utils.rm_trainer</span><span class="w"> </span><span class="kn">import</span> <span class="n">compute_metrics</span><span class="p">,</span> <span class="n">RewardTrainer</span><span class="p">,</span> <span class="n">PeftRewardTrainer</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.utils.peft_trainer</span><span class="w"> </span><span class="kn">import</span> <span class="n">PeftSavingCallback</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.utils.rm_dataprocessor</span><span class="w"> </span><span class="kn">import</span> <span class="n">RewardDataCollatorWithPadding</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_tuner/index.html#lmflow.pipeline.rm_tuner.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="RewardModelTuner">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_tuner/index.html#lmflow.pipeline.rm_tuner.RewardModelTuner">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">RewardModelTuner</span><span class="p">(</span><span class="n">Finetuner</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Initializes the `RewardModelTuner` class.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ----------</span>
+<span class="sd">    model_args : ModelArguments object.</span>
+<span class="sd">        Contains the arguments required to load the model.</span>
+
+<span class="sd">    data_args : DatasetArguments object.</span>
+<span class="sd">        Contains the arguments required to load the dataset.</span>
+
+<span class="sd">    finetuner_args : RewardModelTunerArguments object.</span>
+<span class="sd">        Contains the arguments required to perform finetuning.</span>
+
+<span class="sd">    args : Optional.</span>
+<span class="sd">        Positional arguments.</span>
+
+<span class="sd">    kwargs : Optional.</span>
+<span class="sd">        Keyword arguments.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">model_args</span><span class="p">,</span> 
+        <span class="n">data_args</span><span class="p">,</span> 
+        <span class="n">finetuner_args</span><span class="p">,</span> 
+        <span class="o">*</span><span class="n">args</span><span class="p">,</span> 
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">finetuner_args</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        
+    
+<div class="viewcode-block" id="RewardModelTuner.tune">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/rm_tuner/index.html#lmflow.pipeline.rm_tuner.RewardModelTuner.tune">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">tune</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFTextRegressionModel</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">,</span>
+        <span class="n">transform_dataset_in_place</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">data_collator</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">):</span>
+        <span class="c1"># 0. basic init</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">transform_dataset_in_place</span><span class="p">:</span>
+            <span class="n">dataset</span> <span class="o">=</span> <span class="n">deepcopy</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span>
+            
+        <span class="c1"># 1. prepare dataset</span>
+        <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">main_process_first</span><span class="p">(</span><span class="n">desc</span><span class="o">=</span><span class="s2">&quot;dataset map tokenization&quot;</span><span class="p">):</span>
+            <span class="n">tokenized_dataset</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">tokenize</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+                <span class="n">lm_dataset</span> <span class="o">=</span> <span class="n">tokenized_dataset</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">lm_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">group_text</span><span class="p">(</span>
+                    <span class="n">tokenized_dataset</span><span class="p">,</span>
+                    <span class="n">model_max_length</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">get_max_length</span><span class="p">(),</span>
+                <span class="p">)</span>
+        <span class="n">train_dataset</span> <span class="o">=</span> <span class="n">lm_dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Number of train samples: </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">do_train</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">max_train_samples</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">max_train_samples</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">max_train_samples</span><span class="p">)</span>
+            <span class="n">train_dataset</span> <span class="o">=</span> <span class="n">train_dataset</span><span class="o">.</span><span class="n">select</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="n">max_train_samples</span><span class="p">))</span>
+        
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">do_eval</span><span class="p">:</span>
+            <span class="n">eval_dataset_args</span> <span class="o">=</span> <span class="n">deepcopy</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="p">)</span>
+            <span class="n">eval_dataset_args</span><span class="o">.</span><span class="n">dataset_path</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">eval_dataset_path</span>
+            <span class="n">eval_dataset</span> <span class="o">=</span> <span class="n">Dataset</span><span class="p">(</span><span class="n">eval_dataset_args</span><span class="p">)</span>
+            <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">main_process_first</span><span class="p">(</span><span class="n">desc</span><span class="o">=</span><span class="s2">&quot;dataset map tokenization&quot;</span><span class="p">):</span>
+                <span class="n">tokenized_dataset</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">tokenize</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">)</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+                    <span class="n">lm_dataset</span> <span class="o">=</span> <span class="n">tokenized_dataset</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">lm_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">group_text</span><span class="p">(</span>
+                        <span class="n">tokenized_dataset</span><span class="p">,</span>
+                        <span class="n">model_max_length</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">get_max_length</span><span class="p">(),</span>
+                    <span class="p">)</span>
+            <span class="n">eval_dataset</span> <span class="o">=</span> <span class="n">lm_dataset</span><span class="o">.</span><span class="n">get_backend_dataset</span><span class="p">()</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Number of eval samples: </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        
+        <span class="k">if</span> <span class="n">data_collator</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">data_collator</span> <span class="o">=</span> <span class="n">RewardDataCollatorWithPadding</span><span class="p">(</span>
+                <span class="n">tokenizer</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">get_tokenizer</span><span class="p">(),</span>
+                <span class="n">max_length</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_max_length</span>
+            <span class="p">)</span>
+            
+        <span class="c1"># 2. prepare trainer</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">use_lora</span><span class="p">:</span>
+            <span class="n">RewardModelingTrainer</span> <span class="o">=</span> <span class="n">PeftRewardTrainer</span>
+            <span class="n">trainer_callbacks</span> <span class="o">=</span> <span class="p">[</span><span class="n">PeftSavingCallback</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">RewardModelingTrainer</span> <span class="o">=</span> <span class="n">RewardTrainer</span>
+            <span class="n">trainer_callbacks</span> <span class="o">=</span> <span class="p">[]</span>
+            
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">use_lisa</span><span class="p">:</span>
+            <span class="k">class</span><span class="w"> </span><span class="nc">DynamicLayerActivationCallback</span><span class="p">(</span><span class="n">TrainerCallback</span><span class="p">):</span>
+                <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">n_layers</span><span class="p">,</span> <span class="n">interval_steps</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+                    <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">n_layers</span> <span class="o">=</span> <span class="n">n_layers</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">interval_steps</span> <span class="o">=</span> <span class="n">interval_steps</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">model</span>
+
+                    <span class="c1"># Determine the way to access layers based on the model type</span>
+                    <span class="n">class_to_layers_map</span> <span class="o">=</span> <span class="p">{</span>
+                        <span class="s1">&#39;LlamaForCausalLM&#39;</span><span class="p">:</span> <span class="s1">&#39;model.model.layers&#39;</span><span class="p">,</span>
+                        <span class="s1">&#39;Qwen2ForCausalLM&#39;</span><span class="p">:</span> <span class="s1">&#39;model.model.layers&#39;</span><span class="p">,</span>
+                        <span class="s1">&#39;MistralForCausalLM&#39;</span><span class="p">:</span> <span class="s1">&#39;model.model.layers&#39;</span><span class="p">,</span>
+                        <span class="s1">&#39;MixtralForCausalLM&#39;</span><span class="p">:</span> <span class="s1">&#39;model.model.layers&#39;</span><span class="p">,</span>
+                        <span class="s1">&#39;GemmaForCausalLM&#39;</span><span class="p">:</span> <span class="s1">&#39;model.model.layers&#39;</span><span class="p">,</span>
+                        <span class="s1">&#39;GPT2LMHeadModel&#39;</span><span class="p">:</span> <span class="s1">&#39;model.transformer.h&#39;</span><span class="p">,</span>
+                    <span class="p">}</span>
+                    <span class="n">model_class_name</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span>
+                    <span class="k">if</span> <span class="n">model_class_name</span> <span class="ow">in</span> <span class="n">class_to_layers_map</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">layers_attribute</span> <span class="o">=</span> <span class="n">class_to_layers_map</span><span class="p">[</span><span class="n">model_class_name</span><span class="p">]</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">layers_attribute</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;lisa_layers_attribute&quot;</span><span class="p">)</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">total_layers</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="nb">eval</span><span class="p">(</span><span class="s1">&#39;self.&#39;</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">layers_attribute</span><span class="p">))</span>  <span class="c1"># Dynamically execute to get the number of layers</span>
+
+                    <span class="bp">self</span><span class="o">.</span><span class="n">active_layers_indices</span> <span class="o">=</span> <span class="p">[]</span>
+
+                <span class="k">def</span><span class="w"> </span><span class="nf">freeze_all_layers</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+                    <span class="n">layers</span> <span class="o">=</span> <span class="nb">eval</span><span class="p">(</span><span class="s1">&#39;self.&#39;</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">layers_attribute</span><span class="p">)</span>  <span class="c1"># Dynamically execute to get layers</span>
+                    <span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="n">layers</span><span class="p">:</span>
+                        <span class="k">for</span> <span class="n">param</span> <span class="ow">in</span> <span class="n">layer</span><span class="o">.</span><span class="n">parameters</span><span class="p">():</span>
+                            <span class="n">param</span><span class="o">.</span><span class="n">requires_grad</span> <span class="o">=</span> <span class="kc">False</span>
+
+                <span class="k">def</span><span class="w"> </span><span class="nf">on_step_begin</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">,</span> <span class="n">state</span><span class="p">,</span> <span class="n">control</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+                    <span class="c1"># Check if it&#39;s time to switch active layers, including at step 0</span>
+                    <span class="k">if</span> <span class="n">state</span><span class="o">.</span><span class="n">global_step</span> <span class="o">%</span> <span class="bp">self</span><span class="o">.</span><span class="n">interval_steps</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">switch_active_layers</span><span class="p">()</span>
+
+                <span class="k">def</span><span class="w"> </span><span class="nf">switch_active_layers</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+                    <span class="c1"># First, disable gradients for all layers</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">freeze_all_layers</span><span class="p">()</span>
+
+                    <span class="c1"># Randomly select n_layers to activate</span>
+                    <span class="n">layers</span> <span class="o">=</span> <span class="nb">eval</span><span class="p">(</span><span class="s1">&#39;self.&#39;</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">layers_attribute</span><span class="p">)</span>  <span class="c1"># Re-fetch layer references</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">active_layers_indices</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">choice</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">total_layers</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">n_layers</span><span class="p">,</span> <span class="n">replace</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+                    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Activating layers at indices: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">active_layers_indices</span><span class="si">}</span><span class="s2"> for the next steps.&quot;</span><span class="p">,</span> <span class="n">flush</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+                    <span class="c1"># Enable gradients only for the selected layers</span>
+                    <span class="k">for</span> <span class="n">idx</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">active_layers_indices</span><span class="p">:</span>
+                        <span class="k">for</span> <span class="n">param</span> <span class="ow">in</span> <span class="n">layers</span><span class="p">[</span><span class="n">idx</span><span class="p">]</span><span class="o">.</span><span class="n">parameters</span><span class="p">():</span>
+                            <span class="n">param</span><span class="o">.</span><span class="n">requires_grad</span> <span class="o">=</span> <span class="kc">True</span>
+
+            <span class="c1"># Instantiate the callback</span>
+            <span class="n">dynamic_layer_activation_callback</span> <span class="o">=</span> <span class="n">DynamicLayerActivationCallback</span><span class="p">(</span>
+                <span class="n">n_layers</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">lisa_activated_layers</span><span class="p">,</span>      <span class="c1"># Number of layers to activate</span>
+                <span class="n">interval_steps</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">lisa_interval_steps</span><span class="p">,</span>  <span class="c1"># Step interval to update active layers</span>
+                <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">(),</span>
+                <span class="n">lisa_layers_attribute</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">lisa_layers_attribute</span>
+            <span class="p">)</span>
+
+            <span class="n">trainer_callbacks</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">dynamic_layer_activation_callback</span><span class="p">)</span>
+            
+        <span class="n">trainer</span> <span class="o">=</span> <span class="n">RewardModelingTrainer</span><span class="p">(</span>
+            <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">(),</span>
+            <span class="n">args</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="p">,</span>
+            <span class="n">train_dataset</span><span class="o">=</span><span class="n">train_dataset</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">do_train</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="n">eval_dataset</span><span class="o">=</span><span class="n">eval_dataset</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">do_eval</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="n">tokenizer</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">get_tokenizer</span><span class="p">(),</span>
+            <span class="n">data_collator</span><span class="o">=</span><span class="n">data_collator</span><span class="p">,</span>
+            <span class="n">compute_metrics</span><span class="o">=</span><span class="n">compute_metrics</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">do_eval</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="n">callbacks</span><span class="o">=</span><span class="n">trainer_callbacks</span>
+        <span class="p">)</span>
+        
+        <span class="c1"># 3. training</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">do_train</span><span class="p">:</span>
+            <span class="n">checkpoint</span> <span class="o">=</span> <span class="kc">None</span>
+            <span class="n">last_checkpoint</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">last_checkpoint</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">resume_from_checkpoint</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">checkpoint</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">resume_from_checkpoint</span>
+            <span class="k">elif</span> <span class="n">last_checkpoint</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">checkpoint</span> <span class="o">=</span> <span class="n">last_checkpoint</span>
+                
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">gradient_checkpointing</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">use_cache</span><span class="p">:</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                        <span class="s2">&quot;Backend model config `use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.&quot;</span>
+                    <span class="p">)</span>
+                    <span class="n">model</span><span class="o">.</span><span class="n">get_backend_model</span><span class="p">()</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">use_cache</span> <span class="o">=</span> <span class="kc">False</span>
+                
+            <span class="n">train_result</span> <span class="o">=</span> <span class="n">trainer</span><span class="o">.</span><span class="n">train</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="o">=</span><span class="n">checkpoint</span><span class="p">)</span>
+
+            <span class="n">trainer</span><span class="o">.</span><span class="n">save_model</span><span class="p">()</span>  <span class="c1"># Saves the tokenizer too for easy upload</span>
+
+            <span class="n">metrics</span> <span class="o">=</span> <span class="n">train_result</span><span class="o">.</span><span class="n">metrics</span>
+
+            <span class="n">max_train_samples</span> <span class="o">=</span> <span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">max_train_samples</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">max_train_samples</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="nb">len</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">)</span>
+            <span class="p">)</span>
+            <span class="n">metrics</span><span class="p">[</span><span class="s2">&quot;train_samples&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">max_train_samples</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">))</span>
+
+            <span class="n">trainer</span><span class="o">.</span><span class="n">log_metrics</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="n">metrics</span><span class="p">)</span>
+            <span class="n">trainer</span><span class="o">.</span><span class="n">save_metrics</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="n">metrics</span><span class="p">)</span>
+            <span class="n">trainer</span><span class="o">.</span><span class="n">save_state</span><span class="p">()</span>
+
+        <span class="n">kwargs</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;finetuned_from&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span> <span class="s2">&quot;tasks&quot;</span><span class="p">:</span> <span class="s2">&quot;reward-modeling&quot;</span><span class="p">}</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">dataset_name</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;dataset_tags&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">dataset_name</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">dataset_config_name</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;dataset_args&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">dataset_config_name</span>
+                <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;dataset&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">dataset_name</span><span class="si">}</span><span class="s2"> </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">dataset_config_name</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;dataset&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="o">.</span><span class="n">dataset_name</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">finetuner_args</span><span class="o">.</span><span class="n">push_to_hub</span><span class="p">:</span>
+            <span class="n">trainer</span><span class="o">.</span><span class="n">push_to_hub</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">trainer</span><span class="o">.</span><span class="n">create_model_card</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">model</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html b/_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html
new file mode 100644
index 000000000..2932e91e5
--- /dev/null
+++ b/_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html
@@ -0,0 +1,684 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.utils.dpov2_dataprocessor &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/utils/dpov2_dataprocessor';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.dpov2_dataprocessor</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.utils.dpov2_dataprocessor</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span><span class="w"> </span><span class="nn">dataclasses</span><span class="w"> </span><span class="kn">import</span> <span class="n">dataclass</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Any</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.nn.utils.rnn</span><span class="w"> </span><span class="kn">import</span> <span class="n">pad_sequence</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">PreTrainedModel</span><span class="p">,</span>
+    <span class="n">PreTrainedTokenizerBase</span><span class="p">,</span>
+<span class="p">)</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">PreferenceDataCollatorWithPadding</span><span class="p">:</span>
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.tokenizer">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.tokenizer">[docs]</a>
+    <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizerBase</span></div>
+
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.model">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.model">[docs]</a>
+    <span class="n">model</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">PreTrainedModel</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.padding">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.padding">[docs]</a>
+    <span class="n">padding</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">bool</span><span class="p">,</span> <span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">True</span></div>
+
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.max_length">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_length">[docs]</a>
+    <span class="n">max_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.max_prompt_length">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_prompt_length">[docs]</a>
+    <span class="n">max_prompt_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.label_pad_token_id">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.label_pad_token_id">[docs]</a>
+    <span class="n">label_pad_token_id</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="o">-</span><span class="mi">100</span></div>
+
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.padding_value">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.padding_value">[docs]</a>
+    <span class="n">padding_value</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span></div>
+
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.truncation_mode">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.truncation_mode">[docs]</a>
+    <span class="n">truncation_mode</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;keep_end&quot;</span></div>
+
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.is_encoder_decoder">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.is_encoder_decoder">[docs]</a>
+    <span class="n">is_encoder_decoder</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.max_target_length">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_target_length">[docs]</a>
+    <span class="n">max_target_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.mask_prompt">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.mask_prompt">[docs]</a>
+    <span class="n">mask_prompt</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+
+
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.tokenize_batch_element">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.tokenize_batch_element">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">tokenize_batch_element</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">prompt</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">chosen</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">rejected</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Tokenize a single batch element.</span>
+
+<span class="sd">        At this stage, we don&#39;t convert to PyTorch tensors yet; we just handle the truncation</span>
+<span class="sd">            in case the prompt + chosen or prompt + rejected responses is/are too long. First</span>
+<span class="sd">            we truncate the prompt; if we&#39;re still too long, we truncate the chosen/rejected.</span>
+
+<span class="sd">        We also create the labels for the chosen/rejected responses, which are of length equal to</span>
+<span class="sd">            the sum of the length of the prompt and the chosen/rejected response, with</span>
+<span class="sd">            label_pad_token_id  for the prompt tokens.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">batch</span> <span class="o">=</span> <span class="p">{}</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_encoder_decoder</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">NotImplementedError</span>
+        
+        <span class="n">chosen_tokens</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">(</span><span class="n">chosen</span><span class="p">,</span> <span class="n">add_special_tokens</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="n">rejected_tokens</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">(</span><span class="n">rejected</span><span class="p">,</span> <span class="n">add_special_tokens</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="n">prompt_tokens</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">(</span><span class="n">prompt</span><span class="p">,</span> <span class="n">add_special_tokens</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+        <span class="n">eos_token_id</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span>
+        <span class="c1"># Get indices in list prompt_tokens[&quot;input_ids&quot;] that equals the EOS token (often 0)</span>
+        <span class="n">eos_indices_prompt</span> <span class="o">=</span> <span class="p">[</span><span class="n">i</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">x</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">prompt_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">])</span> <span class="k">if</span> <span class="n">x</span> <span class="o">==</span> <span class="n">eos_token_id</span><span class="p">]</span>
+        <span class="c1"># attention mask these indices to eos_token_id</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">mask_prompt</span><span class="p">:</span>
+            <span class="n">new_attention_mask</span> <span class="o">=</span> <span class="p">[</span><span class="mi">0</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">p</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">prompt_tokens</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">])]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">new_attention_mask</span> <span class="o">=</span> <span class="p">[</span>
+                <span class="mi">0</span> <span class="k">if</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">eos_indices_prompt</span> <span class="k">else</span> <span class="n">p</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">p</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">prompt_tokens</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">])</span>
+            <span class="p">]</span>
+        <span class="n">prompt_tokens</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">new_attention_mask</span>
+
+        <span class="c1"># do the same for chosen and rejected</span>
+        <span class="n">eos_indices_chosen</span> <span class="o">=</span> <span class="p">[</span><span class="n">i</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">x</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">chosen_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">])</span> <span class="k">if</span> <span class="n">x</span> <span class="o">==</span> <span class="n">eos_token_id</span><span class="p">]</span>
+        <span class="n">new_attention_mask_c</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="mi">0</span> <span class="k">if</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">eos_indices_chosen</span> <span class="k">else</span> <span class="n">p</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">p</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">chosen_tokens</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">])</span>
+        <span class="p">]</span>
+        <span class="n">chosen_tokens</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">new_attention_mask_c</span>
+
+        <span class="n">eos_indices_rejected</span> <span class="o">=</span> <span class="p">[</span><span class="n">i</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">x</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">rejected_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">])</span> <span class="k">if</span> <span class="n">x</span> <span class="o">==</span> <span class="n">eos_token_id</span><span class="p">]</span>
+        <span class="n">new_attention_mask_r</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="mi">0</span> <span class="k">if</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">eos_indices_rejected</span> <span class="k">else</span> <span class="n">p</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">p</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">rejected_tokens</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">])</span>
+        <span class="p">]</span>
+        <span class="n">rejected_tokens</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">new_attention_mask_r</span>
+
+        <span class="c1"># add EOS token to end of prompt</span>
+
+        <span class="n">chosen_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span><span class="p">)</span>
+        <span class="n">chosen_tokens</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">rejected_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span><span class="p">)</span>
+        <span class="n">rejected_tokens</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="n">longer_response_length</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">chosen_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]),</span> <span class="nb">len</span><span class="p">(</span><span class="n">rejected_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]))</span>
+
+        <span class="c1"># if combined sequence is too long, truncate the prompt</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">prompt_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">])</span> <span class="o">+</span> <span class="n">longer_response_length</span> <span class="o">&gt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_length</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">truncation_mode</span> <span class="o">==</span> <span class="s2">&quot;keep_start&quot;</span><span class="p">:</span>
+                <span class="n">prompt_tokens</span> <span class="o">=</span> <span class="p">{</span><span class="n">k</span><span class="p">:</span> <span class="n">v</span><span class="p">[:</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_prompt_length</span><span class="p">]</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">prompt_tokens</span><span class="o">.</span><span class="n">items</span><span class="p">()}</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">truncation_mode</span> <span class="o">==</span> <span class="s2">&quot;keep_end&quot;</span><span class="p">:</span>
+                <span class="n">prompt_tokens</span> <span class="o">=</span> <span class="p">{</span><span class="n">k</span><span class="p">:</span> <span class="n">v</span><span class="p">[</span><span class="o">-</span><span class="bp">self</span><span class="o">.</span><span class="n">max_prompt_length</span> <span class="p">:]</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">prompt_tokens</span><span class="o">.</span><span class="n">items</span><span class="p">()}</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unknown truncation mode: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">truncation_mode</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="c1"># if that&#39;s still too long, truncate the response</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">prompt_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">])</span> <span class="o">+</span> <span class="n">longer_response_length</span> <span class="o">&gt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_length</span><span class="p">:</span>
+            <span class="n">chosen_tokens</span> <span class="o">=</span> <span class="p">{</span><span class="n">k</span><span class="p">:</span> <span class="n">v</span><span class="p">[:</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_length</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_prompt_length</span><span class="p">]</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">chosen_tokens</span><span class="o">.</span><span class="n">items</span><span class="p">()}</span>
+            <span class="n">rejected_tokens</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="n">k</span><span class="p">:</span> <span class="n">v</span><span class="p">[:</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_length</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_prompt_length</span><span class="p">]</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">rejected_tokens</span><span class="o">.</span><span class="n">items</span><span class="p">()</span>
+            <span class="p">}</span>
+
+        <span class="c1"># Create labels</span>
+        <span class="n">chosen_sequence_tokens</span> <span class="o">=</span> <span class="p">{</span><span class="n">k</span><span class="p">:</span> <span class="n">prompt_tokens</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">+</span> <span class="n">chosen_tokens</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">chosen_tokens</span><span class="p">}</span>
+        <span class="n">rejected_sequence_tokens</span> <span class="o">=</span> <span class="p">{</span><span class="n">k</span><span class="p">:</span> <span class="n">prompt_tokens</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">+</span> <span class="n">rejected_tokens</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">rejected_tokens</span><span class="p">}</span>
+        <span class="n">chosen_sequence_tokens</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">chosen_sequence_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][:]</span>
+        <span class="n">chosen_sequence_tokens</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][:</span> <span class="nb">len</span><span class="p">(</span><span class="n">prompt_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">])]</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">label_pad_token_id</span><span class="p">]</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span>
+            <span class="n">prompt_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span>
+        <span class="p">)</span>
+        <span class="n">rejected_sequence_tokens</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">rejected_sequence_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][:]</span>
+        <span class="n">rejected_sequence_tokens</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][:</span> <span class="nb">len</span><span class="p">(</span><span class="n">prompt_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">])]</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">label_pad_token_id</span><span class="p">]</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span>
+            <span class="n">prompt_tokens</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span>
+        <span class="p">)</span>
+
+        <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">toks</span> <span class="ow">in</span> <span class="p">{</span>
+            <span class="s2">&quot;chosen&quot;</span><span class="p">:</span> <span class="n">chosen_sequence_tokens</span><span class="p">,</span>
+            <span class="s2">&quot;rejected&quot;</span><span class="p">:</span> <span class="n">rejected_sequence_tokens</span><span class="p">,</span>
+            <span class="s2">&quot;prompt&quot;</span><span class="p">:</span> <span class="n">prompt_tokens</span><span class="p">,</span>
+        <span class="p">}</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+            <span class="k">for</span> <span class="n">type_key</span><span class="p">,</span> <span class="n">tokens</span> <span class="ow">in</span> <span class="n">toks</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+                <span class="k">if</span> <span class="n">type_key</span> <span class="o">==</span> <span class="s2">&quot;token_type_ids&quot;</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="n">batch</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">k</span><span class="si">}</span><span class="s2">_</span><span class="si">{</span><span class="n">type_key</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">tokens</span>
+
+
+
+        <span class="n">batch</span><span class="p">[</span><span class="s2">&quot;prompt&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">prompt</span>
+        <span class="n">batch</span><span class="p">[</span><span class="s2">&quot;chosen&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">prompt</span> <span class="o">+</span> <span class="n">chosen</span>
+        <span class="n">batch</span><span class="p">[</span><span class="s2">&quot;rejected&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">prompt</span> <span class="o">+</span> <span class="n">rejected</span>
+        <span class="n">batch</span><span class="p">[</span><span class="s2">&quot;chosen_response_only&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">chosen</span>
+        <span class="n">batch</span><span class="p">[</span><span class="s2">&quot;rejected_response_only&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">rejected</span>
+
+        <span class="k">return</span> <span class="n">batch</span></div>
+
+
+
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.collate">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.collate">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">collate</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">):</span>
+        <span class="c1"># first, pad everything to the same length</span>
+        <span class="n">padded_batch</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">batch</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">keys</span><span class="p">():</span>
+            <span class="k">if</span> <span class="n">k</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;_input_ids&quot;</span><span class="p">)</span> <span class="ow">or</span> <span class="n">k</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;_attention_mask&quot;</span><span class="p">)</span> <span class="ow">or</span> <span class="n">k</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;_labels&quot;</span><span class="p">):</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_encoder_decoder</span><span class="p">:</span>
+                    <span class="n">to_pad</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">(</span><span class="n">ex</span><span class="p">[</span><span class="n">k</span><span class="p">])</span> <span class="k">for</span> <span class="n">ex</span> <span class="ow">in</span> <span class="n">batch</span><span class="p">]</span>
+
+                    <span class="k">if</span> <span class="p">(</span><span class="n">k</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;prompt&quot;</span><span class="p">))</span> <span class="ow">and</span> <span class="p">(</span><span class="n">k</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;input_ids&quot;</span><span class="p">)):</span>
+                        <span class="n">padding_value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span>
+                    <span class="k">elif</span> <span class="n">k</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;_attention_mask&quot;</span><span class="p">):</span>
+                        <span class="n">padding_value</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="k">elif</span> <span class="p">(</span><span class="n">k</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;chosen&quot;</span><span class="p">))</span> <span class="ow">or</span> <span class="p">(</span><span class="n">k</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;rejected&quot;</span><span class="p">))</span> <span class="ow">or</span> <span class="p">(</span><span class="s2">&quot;decoder&quot;</span> <span class="ow">in</span> <span class="n">k</span><span class="p">):</span>
+                        <span class="n">padding_value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">label_pad_token_id</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unexpected key in batch &#39;</span><span class="si">{</span><span class="n">k</span><span class="si">}</span><span class="s2">&#39;&quot;</span><span class="p">)</span>
+                    <span class="n">padded_batch</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="n">pad_sequence</span><span class="p">(</span><span class="n">to_pad</span><span class="p">,</span> <span class="n">batch_first</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">padding_value</span><span class="o">=</span><span class="n">padding_value</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="c1"># adapted from https://stackoverflow.com/questions/73256206</span>
+                    <span class="k">if</span> <span class="s2">&quot;prompt&quot;</span> <span class="ow">in</span> <span class="n">k</span><span class="p">:</span>
+                        <span class="n">to_pad</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">(</span><span class="n">ex</span><span class="p">[</span><span class="n">k</span><span class="p">][::</span><span class="o">-</span><span class="mi">1</span><span class="p">])</span> <span class="k">for</span> <span class="n">ex</span> <span class="ow">in</span> <span class="n">batch</span><span class="p">]</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">to_pad</span> <span class="o">=</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">(</span><span class="n">ex</span><span class="p">[</span><span class="n">k</span><span class="p">])</span> <span class="k">for</span> <span class="n">ex</span> <span class="ow">in</span> <span class="n">batch</span><span class="p">]</span>
+                    <span class="k">if</span> <span class="n">k</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;_input_ids&quot;</span><span class="p">):</span>
+                        <span class="n">padding_value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span>
+                    <span class="k">elif</span> <span class="n">k</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;_labels&quot;</span><span class="p">):</span>
+                        <span class="n">padding_value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">label_pad_token_id</span>
+                    <span class="k">elif</span> <span class="n">k</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;_attention_mask&quot;</span><span class="p">):</span>
+                        <span class="n">padding_value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">padding_value</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unexpected key in batch &#39;</span><span class="si">{</span><span class="n">k</span><span class="si">}</span><span class="s2">&#39;&quot;</span><span class="p">)</span>
+
+                    <span class="n">padded_batch</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="n">pad_sequence</span><span class="p">(</span><span class="n">to_pad</span><span class="p">,</span> <span class="n">batch_first</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">padding_value</span><span class="o">=</span><span class="n">padding_value</span><span class="p">)</span>
+                    <span class="c1"># for the prompt, flip back so padding is on left side</span>
+                    <span class="k">if</span> <span class="s2">&quot;prompt&quot;</span> <span class="ow">in</span> <span class="n">k</span><span class="p">:</span>
+                        <span class="n">padded_batch</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="n">padded_batch</span><span class="p">[</span><span class="n">k</span><span class="p">]</span><span class="o">.</span><span class="n">flip</span><span class="p">(</span><span class="n">dims</span><span class="o">=</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">padded_batch</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="p">[</span><span class="n">ex</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="k">for</span> <span class="n">ex</span> <span class="ow">in</span> <span class="n">batch</span><span class="p">]</span>
+
+        <span class="k">return</span> <span class="n">padded_batch</span></div>
+
+
+
+<div class="viewcode-block" id="PreferenceDataCollatorWithPadding.__call__">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.__call__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">features</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">tokenized_batch</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="k">for</span> <span class="n">feature</span> <span class="ow">in</span> <span class="n">features</span><span class="p">:</span>
+            <span class="n">prompt</span> <span class="o">=</span> <span class="n">feature</span><span class="p">[</span><span class="s2">&quot;prompt&quot;</span><span class="p">]</span>
+            <span class="n">chosen</span> <span class="o">=</span> <span class="n">feature</span><span class="p">[</span><span class="s2">&quot;chosen&quot;</span><span class="p">]</span>
+            <span class="n">rejected</span> <span class="o">=</span> <span class="n">feature</span><span class="p">[</span><span class="s2">&quot;rejected&quot;</span><span class="p">]</span>
+
+            <span class="n">batch_element</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenize_batch_element</span><span class="p">(</span><span class="n">prompt</span><span class="p">,</span> <span class="n">chosen</span><span class="p">,</span> <span class="n">rejected</span><span class="p">)</span>
+            <span class="n">batch_element</span><span class="p">[</span><span class="s2">&quot;margin&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">feature</span><span class="p">[</span><span class="s2">&quot;margin&quot;</span><span class="p">]</span>
+            <span class="n">tokenized_batch</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">batch_element</span><span class="p">)</span>
+
+        <span class="c1"># return collated batch</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">collate</span><span class="p">(</span><span class="n">tokenized_batch</span><span class="p">)</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/utils/dpov2_trainer.html b/_modules/lmflow/pipeline/utils/dpov2_trainer.html
new file mode 100644
index 000000000..c6ad15cd8
--- /dev/null
+++ b/_modules/lmflow/pipeline/utils/dpov2_trainer.html
@@ -0,0 +1,714 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.utils.dpov2_trainer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/utils/dpov2_trainer';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.dpov2_trainer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.utils.dpov2_trainer</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Literal</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">datasets</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.nn</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">nn</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.nn.functional</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">F</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">PreTrainedModel</span><span class="p">,</span> 
+    <span class="n">PreTrainedTokenizerBase</span><span class="p">,</span> 
+    <span class="n">DataCollator</span><span class="p">,</span> 
+    <span class="n">TrainingArguments</span><span class="p">,</span> 
+    <span class="n">TrainerCallback</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.trainer_callback</span><span class="w"> </span><span class="kn">import</span> <span class="n">TrainerCallback</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.trainer_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">EvalLoopOutput</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.utils.dpov2_dataprocessor</span><span class="w"> </span><span class="kn">import</span> <span class="n">PreferenceDataCollatorWithPadding</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.versioning</span><span class="w"> </span><span class="kn">import</span> <span class="n">is_trl_available</span>
+
+<span class="k">if</span> <span class="n">is_trl_available</span><span class="p">():</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">trl</span><span class="w"> </span><span class="kn">import</span> <span class="n">DPOTrainer</span>
+<span class="k">else</span><span class="p">:</span>
+    <span class="k">raise</span> <span class="ne">ImportError</span><span class="p">(</span><span class="s2">&quot;Please install trl package to use dpo_aligner.py&quot;</span><span class="p">)</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="DPOv2Trainer">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">DPOv2Trainer</span><span class="p">(</span><span class="n">DPOTrainer</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">PreTrainedModel</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">ref_model</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">PreTrainedModel</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">beta</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mf">0.1</span><span class="p">,</span>
+        <span class="n">loss_type</span><span class="p">:</span> <span class="n">Literal</span><span class="p">[</span><span class="s2">&quot;sigmoid&quot;</span><span class="p">,</span> <span class="s2">&quot;hinge&quot;</span><span class="p">,</span> <span class="s2">&quot;cross_entropy&quot;</span><span class="p">,</span> <span class="s2">&quot;kl&quot;</span><span class="p">,</span> <span class="s2">&quot;rev_kl&quot;</span><span class="p">,</span> <span class="s2">&quot;raft&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;rev_kl&quot;</span><span class="p">,</span>
+        <span class="n">args</span><span class="p">:</span> <span class="n">TrainingArguments</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">data_collator</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">DataCollator</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">label_pad_token_id</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="o">-</span><span class="mi">100</span><span class="p">,</span>
+        <span class="n">padding_value</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">truncation_mode</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;keep_end&quot;</span><span class="p">,</span>
+        <span class="n">train_dataset</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dataset</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">eval_dataset</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Dataset</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Dataset</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">PreTrainedTokenizerBase</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">model_init</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[],</span> <span class="n">PreTrainedModel</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">callbacks</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">TrainerCallback</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">optimizers</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">Optimizer</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">lr_scheduler</span><span class="o">.</span><span class="n">LambdaLR</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="kc">None</span><span class="p">,</span>
+            <span class="kc">None</span><span class="p">,</span>
+        <span class="p">),</span>
+        <span class="n">preprocess_logits_for_metrics</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">max_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">max_prompt_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">max_target_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">peft_config</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">is_encoder_decoder</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">disable_dropout</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">generate_during_eval</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">compute_metrics</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[</span><span class="n">EvalLoopOutput</span><span class="p">],</span> <span class="n">Dict</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">mask_prompt</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">len_penalty</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="n">preprocessing_num_workers</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
+    <span class="p">):</span>
+
+        <span class="k">if</span> <span class="n">data_collator</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">data_collator</span> <span class="o">=</span> <span class="n">PreferenceDataCollatorWithPadding</span><span class="p">(</span>
+                <span class="n">tokenizer</span><span class="p">,</span>
+                <span class="n">max_length</span><span class="o">=</span><span class="n">max_length</span><span class="p">,</span>
+                <span class="n">max_prompt_length</span><span class="o">=</span><span class="n">max_prompt_length</span><span class="p">,</span>
+                <span class="n">label_pad_token_id</span><span class="o">=</span><span class="n">label_pad_token_id</span><span class="p">,</span>
+                <span class="n">padding_value</span><span class="o">=</span><span class="n">padding_value</span><span class="p">,</span>
+                <span class="n">truncation_mode</span><span class="o">=</span><span class="n">truncation_mode</span><span class="p">,</span>
+                <span class="n">is_encoder_decoder</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                <span class="n">max_target_length</span><span class="o">=</span><span class="n">max_target_length</span><span class="p">,</span>
+                <span class="n">mask_prompt</span><span class="o">=</span><span class="n">mask_prompt</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
+            <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span>
+            <span class="n">ref_model</span><span class="o">=</span><span class="n">ref_model</span><span class="p">,</span>
+            <span class="n">beta</span><span class="o">=</span><span class="n">beta</span><span class="p">,</span>
+            <span class="n">loss_type</span><span class="o">=</span><span class="n">loss_type</span><span class="p">,</span>
+            <span class="n">args</span><span class="o">=</span><span class="n">args</span><span class="p">,</span>
+            <span class="n">data_collator</span><span class="o">=</span><span class="n">data_collator</span><span class="p">,</span>
+            <span class="n">label_pad_token_id</span><span class="o">=</span><span class="n">label_pad_token_id</span><span class="p">,</span>
+            <span class="n">padding_value</span><span class="o">=</span><span class="n">padding_value</span><span class="p">,</span>
+            <span class="n">truncation_mode</span><span class="o">=</span><span class="n">truncation_mode</span><span class="p">,</span>
+            <span class="n">train_dataset</span><span class="o">=</span><span class="n">train_dataset</span><span class="p">,</span>
+            <span class="n">eval_dataset</span><span class="o">=</span><span class="n">eval_dataset</span><span class="p">,</span>
+            <span class="n">tokenizer</span><span class="o">=</span><span class="n">tokenizer</span><span class="p">,</span>
+            <span class="n">model_init</span><span class="o">=</span><span class="n">model_init</span><span class="p">,</span>
+            <span class="n">callbacks</span><span class="o">=</span><span class="n">callbacks</span><span class="p">,</span>
+            <span class="n">optimizers</span><span class="o">=</span><span class="n">optimizers</span><span class="p">,</span>
+            <span class="n">preprocess_logits_for_metrics</span><span class="o">=</span><span class="n">preprocess_logits_for_metrics</span><span class="p">,</span>
+            <span class="n">max_length</span><span class="o">=</span><span class="n">max_length</span><span class="p">,</span>
+            <span class="n">max_prompt_length</span><span class="o">=</span><span class="n">max_prompt_length</span><span class="p">,</span>
+            <span class="n">max_target_length</span><span class="o">=</span><span class="n">max_target_length</span><span class="p">,</span>
+            <span class="n">peft_config</span><span class="o">=</span><span class="n">peft_config</span><span class="p">,</span>
+            <span class="n">is_encoder_decoder</span><span class="o">=</span><span class="n">is_encoder_decoder</span><span class="p">,</span>
+            <span class="n">disable_dropout</span><span class="o">=</span><span class="n">disable_dropout</span><span class="p">,</span>
+            <span class="n">generate_during_eval</span><span class="o">=</span><span class="n">generate_during_eval</span><span class="p">,</span>
+            <span class="n">compute_metrics</span><span class="o">=</span><span class="n">compute_metrics</span><span class="p">,</span>
+            <span class="n">dataset_num_proc</span><span class="o">=</span><span class="n">preprocessing_num_workers</span><span class="p">,</span>
+        <span class="p">)</span>
+<div class="viewcode-block" id="DPOv2Trainer.use_dpo_data_collator">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.use_dpo_data_collator">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_dpo_data_collator</span> <span class="o">=</span> <span class="kc">True</span></div>
+
+<div class="viewcode-block" id="DPOv2Trainer.len_penalty">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.len_penalty">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">len_penalty</span> <span class="o">=</span> <span class="n">len_penalty</span></div>
+
+
+<div class="viewcode-block" id="DPOv2Trainer.dpo_loss">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.dpo_loss">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">dpo_loss</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">policy_chosen_logps</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">,</span>
+        <span class="n">policy_rejected_logps</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">,</span>
+        <span class="n">reference_chosen_logps</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">,</span>
+        <span class="n">reference_rejected_logps</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">,</span>
+        <span class="n">reference_free</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">margin</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">len_penalty</span><span class="p">:</span> <span class="nb">float</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Compute the DPO loss for a batch of policy and reference model log probabilities.</span>
+
+<span class="sd">        Args:</span>
+<span class="sd">            policy_chosen_logps: Log probabilities of the policy model for the chosen responses. Shape: (batch_size,)</span>
+<span class="sd">            policy_rejected_logps: Log probabilities of the policy model for the rejected responses. Shape: (batch_size,)</span>
+<span class="sd">            reference_chosen_logps: Log probabilities of the reference model for the chosen responses. Shape: (batch_size,)</span>
+<span class="sd">            reference_rejected_logps: Log probabilities of the reference model for the rejected responses. Shape: (batch_size,)</span>
+<span class="sd">            beta: Temperature parameter for the DPO loss, typically something in the range of 0.1 to 0.5. We ignore the reference model as beta -&gt; 0.</span>
+<span class="sd">            reference_free: If True, we ignore the _provided_ reference model and implicitly use a reference model that assigns equal probability to all responses.</span>
+
+<span class="sd">        Returns:</span>
+<span class="sd">            A tuple of three tensors: (losses, chosen_rewards, rejected_rewards).</span>
+<span class="sd">            The losses tensor contains the DPO loss for each example in the batch.</span>
+<span class="sd">            The chosen_rewards and rejected_rewards tensors contain the rewards for the chosen and rejected responses, respectively.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">pi_logratios</span> <span class="o">=</span> <span class="n">policy_chosen_logps</span> <span class="o">-</span> <span class="n">policy_rejected_logps</span>
+        <span class="n">ref_logratios</span> <span class="o">=</span> <span class="n">reference_chosen_logps</span> <span class="o">-</span> <span class="n">reference_rejected_logps</span> <span class="o">+</span> <span class="n">len_penalty</span>
+
+        <span class="k">if</span> <span class="n">reference_free</span><span class="p">:</span>
+            <span class="n">ref_logratios</span> <span class="o">=</span> <span class="mi">0</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">loss_type</span> <span class="o">==</span> <span class="s2">&quot;sigmoid&quot;</span><span class="p">:</span>
+            <span class="n">logits</span> <span class="o">=</span> <span class="n">pi_logratios</span> <span class="o">-</span> <span class="n">ref_logratios</span>
+            <span class="n">losses</span> <span class="o">=</span> <span class="o">-</span><span class="n">F</span><span class="o">.</span><span class="n">logsigmoid</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">beta</span> <span class="o">*</span> <span class="n">logits</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">loss_type</span> <span class="o">==</span> <span class="s2">&quot;hinge&quot;</span><span class="p">:</span>
+            <span class="n">logits</span> <span class="o">=</span> <span class="n">pi_logratios</span> <span class="o">-</span> <span class="n">ref_logratios</span>
+            <span class="n">losses</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">relu</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">beta</span> <span class="o">*</span> <span class="n">logits</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">loss_type</span> <span class="o">==</span> <span class="s2">&quot;cross_entropy&quot;</span><span class="p">:</span>
+            <span class="n">logits</span> <span class="o">=</span> <span class="n">policy_chosen_logps</span> <span class="o">-</span> <span class="n">reference_chosen_logps</span>
+            <span class="n">losses</span> <span class="o">=</span> <span class="o">-</span><span class="n">F</span><span class="o">.</span><span class="n">logsigmoid</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">beta</span> <span class="o">*</span> <span class="n">logits</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">loss_type</span> <span class="o">==</span> <span class="s2">&quot;raft&quot;</span><span class="p">:</span>
+            <span class="n">losses</span> <span class="o">=</span> <span class="o">-</span><span class="n">policy_chosen_logps</span>  <span class="c1"># F.logsigmoid(self.beta * logits)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">loss_type</span> <span class="o">==</span> <span class="s2">&quot;ipo&quot;</span><span class="p">:</span>
+            <span class="n">logits</span> <span class="o">=</span> <span class="n">pi_logratios</span> <span class="o">-</span> <span class="n">ref_logratios</span>
+            <span class="c1"># eqn (17) of the paper where beta is the regularization parameter for the IPO loss, denoted by tau in the paper.</span>
+            <span class="n">losses</span> <span class="o">=</span> <span class="p">(</span><span class="n">logits</span> <span class="o">-</span> <span class="mi">1</span> <span class="o">/</span> <span class="p">(</span><span class="mi">2</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">beta</span><span class="p">))</span> <span class="o">**</span> <span class="mi">2</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">loss_type</span> <span class="o">==</span> <span class="s2">&quot;kl&quot;</span><span class="p">:</span>
+            <span class="n">logits</span> <span class="o">=</span> <span class="n">pi_logratios</span> <span class="o">-</span> <span class="n">ref_logratios</span>
+            <span class="n">p</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">sigmoid</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">beta</span> <span class="o">*</span> <span class="n">logits</span><span class="p">)</span>
+            <span class="n">p</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">minimum</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones_like</span><span class="p">(</span><span class="n">p</span><span class="p">)</span> <span class="o">*</span> <span class="mf">0.999</span><span class="p">)</span>
+            <span class="n">p_gt</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">margin</span><span class="p">)</span> <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">+</span> <span class="n">torch</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">margin</span><span class="p">)</span> <span class="o">+</span> <span class="mf">1e-3</span><span class="p">)</span>
+            <span class="n">losses</span> <span class="o">=</span> <span class="n">p</span> <span class="o">*</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="n">p</span><span class="p">)</span> <span class="o">-</span> <span class="n">torch</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="n">p_gt</span><span class="p">))</span> <span class="o">+</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">p</span><span class="p">)</span> <span class="o">*</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">p</span><span class="p">)</span> <span class="o">-</span> <span class="n">torch</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">p_gt</span><span class="p">))</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">loss_type</span> <span class="o">==</span> <span class="s2">&quot;tv&quot;</span><span class="p">:</span>
+            <span class="n">logits</span> <span class="o">=</span> <span class="n">pi_logratios</span> <span class="o">-</span> <span class="n">ref_logratios</span>
+            <span class="n">p</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">sigmoid</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">beta</span> <span class="o">*</span> <span class="n">logits</span><span class="p">)</span>
+            <span class="n">p_gt</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">margin</span><span class="p">)</span> <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">+</span> <span class="n">torch</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">margin</span><span class="p">))</span>
+            <span class="n">losses</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">abs</span><span class="p">(</span><span class="n">p</span> <span class="o">-</span> <span class="n">p_gt</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">loss_type</span> <span class="o">==</span> <span class="s2">&quot;hellinger&quot;</span><span class="p">:</span>
+            <span class="n">logits</span> <span class="o">=</span> <span class="n">pi_logratios</span> <span class="o">-</span> <span class="n">ref_logratios</span>
+            <span class="n">p</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">sigmoid</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">beta</span> <span class="o">*</span> <span class="n">logits</span><span class="p">)</span>
+            <span class="n">p</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">minimum</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones_like</span><span class="p">(</span><span class="n">p</span><span class="p">)</span> <span class="o">*</span> <span class="mf">0.999</span><span class="p">)</span>
+            <span class="n">p_gt</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">margin</span><span class="p">)</span> <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">+</span> <span class="n">torch</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">margin</span><span class="p">))</span>
+            <span class="n">losses</span> <span class="o">=</span> <span class="mf">0.5</span> <span class="o">*</span> <span class="p">((</span><span class="n">p</span><span class="o">**</span><span class="mf">0.5</span> <span class="o">-</span> <span class="n">p_gt</span><span class="o">**</span><span class="mf">0.5</span><span class="p">)</span> <span class="o">**</span> <span class="mi">2</span> <span class="o">+</span> <span class="p">((</span><span class="mi">1</span> <span class="o">-</span> <span class="n">p</span><span class="p">)</span> <span class="o">**</span> <span class="mf">0.5</span> <span class="o">-</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">p_gt</span><span class="p">)</span> <span class="o">**</span> <span class="mf">0.5</span><span class="p">)</span> <span class="o">**</span> <span class="mi">2</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">loss_type</span> <span class="o">==</span> <span class="s2">&quot;rev_kl&quot;</span><span class="p">:</span>
+            <span class="n">logits</span> <span class="o">=</span> <span class="n">pi_logratios</span> <span class="o">-</span> <span class="n">ref_logratios</span>
+            <span class="n">logp</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">logsigmoid</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">beta</span> <span class="o">*</span> <span class="n">logits</span><span class="p">)</span>
+            <span class="n">logp_neg</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">logsigmoid</span><span class="p">(</span><span class="o">-</span><span class="bp">self</span><span class="o">.</span><span class="n">beta</span> <span class="o">*</span> <span class="n">logits</span><span class="p">)</span>
+            <span class="n">p_gt</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">sigmoid</span><span class="p">(</span><span class="n">margin</span><span class="p">)</span>
+            <span class="n">losses</span> <span class="o">=</span> <span class="o">-</span><span class="n">p_gt</span> <span class="o">*</span> <span class="p">(</span><span class="n">logp</span><span class="p">)</span> <span class="o">-</span> <span class="p">(</span><span class="mi">1</span> <span class="o">-</span> <span class="n">p_gt</span><span class="p">)</span> <span class="o">*</span> <span class="n">logp_neg</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unknown loss type: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">loss_type</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+        <span class="n">chosen_rewards</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">beta</span> <span class="o">*</span> <span class="p">(</span><span class="n">policy_chosen_logps</span> <span class="o">-</span> <span class="n">reference_chosen_logps</span><span class="p">)</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span>
+        <span class="n">rejected_rewards</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">beta</span> <span class="o">*</span> <span class="p">(</span><span class="n">policy_rejected_logps</span> <span class="o">-</span> <span class="n">reference_rejected_logps</span><span class="p">)</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span>
+
+        <span class="k">return</span> <span class="n">losses</span><span class="p">,</span> <span class="n">chosen_rewards</span><span class="p">,</span> <span class="n">rejected_rewards</span></div>
+
+
+<div class="viewcode-block" id="DPOv2Trainer.get_batch_loss_metrics">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.get_batch_loss_metrics">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_batch_loss_metrics</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">,</span>
+        <span class="n">batch</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">]],</span>
+        <span class="n">train_eval</span><span class="p">:</span> <span class="n">Literal</span><span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;eval&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_batch_metrics</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">batch</span><span class="p">,</span> <span class="n">train_eval</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="DPOv2Trainer.get_batch_metrics">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.get_batch_metrics">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_batch_metrics</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">,</span>
+        <span class="n">batch</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">]],</span>
+        <span class="n">train_eval</span><span class="p">:</span> <span class="n">Literal</span><span class="p">[</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="s2">&quot;eval&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;train&quot;</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Compute the DPO loss and other metrics for the given batch of inputs for train or test.&quot;&quot;&quot;</span>
+        <span class="n">metrics</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="p">(</span>
+            <span class="n">policy_chosen_logps</span><span class="p">,</span>
+            <span class="n">policy_rejected_logps</span><span class="p">,</span>
+            <span class="n">policy_chosen_logits</span><span class="p">,</span>
+            <span class="n">policy_rejected_logits</span><span class="p">,</span>
+        <span class="p">)</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">concatenated_forward</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">batch</span><span class="p">)</span>
+        <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">ref_model</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="o">.</span><span class="n">unwrap_model</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">)</span><span class="o">.</span><span class="n">disable_adapter</span><span class="p">():</span>
+                    <span class="p">(</span>
+                        <span class="n">reference_chosen_logps</span><span class="p">,</span>
+                        <span class="n">reference_rejected_logps</span><span class="p">,</span>
+                        <span class="n">_</span><span class="p">,</span>
+                        <span class="n">_</span><span class="p">,</span>
+                    <span class="p">)</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">concatenated_forward</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">,</span> <span class="n">batch</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="p">(</span>
+                    <span class="n">reference_chosen_logps</span><span class="p">,</span>
+                    <span class="n">reference_rejected_logps</span><span class="p">,</span>
+                    <span class="n">_</span><span class="p">,</span>
+                    <span class="n">_</span><span class="p">,</span>
+                <span class="p">)</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">concatenated_forward</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">ref_model</span><span class="p">,</span> <span class="n">batch</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">len_penalty</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">chosen_len</span> <span class="o">=</span> <span class="n">batch</span><span class="p">[</span><span class="s2">&quot;chosen_input_ids&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">len_penalty</span>
+            <span class="n">rejected_len</span> <span class="o">=</span> <span class="n">batch</span><span class="p">[</span><span class="s2">&quot;rejected_input_ids&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">len_penalty</span>
+            <span class="n">len_penalty</span> <span class="o">=</span> <span class="n">chosen_len</span> <span class="o">-</span> <span class="n">rejected_len</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">chosen_len</span> <span class="o">=</span> <span class="mi">1</span>
+            <span class="n">rejected_len</span> <span class="o">=</span> <span class="mi">1</span>
+            <span class="n">len_penalty</span> <span class="o">=</span> <span class="mi">0</span>
+
+        <span class="n">margin</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">batch</span><span class="p">[</span><span class="s2">&quot;margin&quot;</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">policy_chosen_logps</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">accelerator</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+        <span class="n">losses</span><span class="p">,</span> <span class="n">chosen_rewards</span><span class="p">,</span> <span class="n">rejected_rewards</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dpo_loss</span><span class="p">(</span>
+            <span class="n">policy_chosen_logps</span><span class="p">,</span>
+            <span class="n">policy_rejected_logps</span><span class="p">,</span>
+            <span class="n">reference_chosen_logps</span><span class="p">,</span>
+            <span class="n">reference_rejected_logps</span><span class="p">,</span>
+            <span class="n">margin</span><span class="o">=</span><span class="n">margin</span><span class="p">,</span>
+            <span class="n">len_penalty</span><span class="o">=</span><span class="n">len_penalty</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">reward_accuracies</span> <span class="o">=</span> <span class="p">(</span><span class="n">chosen_rewards</span> <span class="o">&gt;</span> <span class="n">rejected_rewards</span><span class="p">)</span><span class="o">.</span><span class="n">float</span><span class="p">()</span>
+
+        <span class="n">prefix</span> <span class="o">=</span> <span class="s2">&quot;eval_&quot;</span> <span class="k">if</span> <span class="n">train_eval</span> <span class="o">==</span> <span class="s2">&quot;eval&quot;</span> <span class="k">else</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}</span><span class="s2">rewards/chosen&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">chosen_rewards</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span>
+        <span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}</span><span class="s2">rewards/rejected&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">rejected_rewards</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span>
+        <span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}</span><span class="s2">rewards/accuracies&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">reward_accuracies</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span>
+        <span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}</span><span class="s2">rewards/margins&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="n">chosen_rewards</span> <span class="o">-</span> <span class="n">rejected_rewards</span><span class="p">)</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span>
+        <span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}</span><span class="s2">logps/rejected&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">policy_rejected_logps</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span>
+        <span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}</span><span class="s2">logps/chosen&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">policy_chosen_logps</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span>
+        <span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}</span><span class="s2">logits/rejected&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">policy_rejected_logits</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span>
+        <span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}</span><span class="s2">logits/chosen&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">policy_chosen_logits</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span>
+
+        <span class="k">return</span> <span class="n">losses</span><span class="o">.</span><span class="n">mean</span><span class="p">(),</span> <span class="n">metrics</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/utils/memory_safe_dpov2_align.html b/_modules/lmflow/pipeline/utils/memory_safe_dpov2_align.html
new file mode 100644
index 000000000..fc3c44941
--- /dev/null
+++ b/_modules/lmflow/pipeline/utils/memory_safe_dpov2_align.html
@@ -0,0 +1,520 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.utils.memory_safe_dpov2_align &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/utils/memory_safe_dpov2_align';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.memory_safe_dpov2_align</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.utils.memory_safe_dpov2_align</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">copy</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">HfArgumentParser</span>
+<span class="p">)</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_decoder_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFDecoderModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.dpov2_aligner</span><span class="w"> </span><span class="kn">import</span> <span class="n">DPOv2Aligner</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">ModelArguments</span><span class="p">,</span> 
+    <span class="n">DatasetArguments</span><span class="p">,</span> 
+    <span class="n">DPOv2AlignerArguments</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.common</span><span class="w"> </span><span class="kn">import</span> <span class="n">remove_dataclass_attr_prefix</span><span class="p">,</span> <span class="n">create_copied_dataclass</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.html#lmflow.pipeline.utils.memory_safe_dpov2_align.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="ReferenceModelArguments">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.html#lmflow.pipeline.utils.memory_safe_dpov2_align.ReferenceModelArguments">[docs]</a>
+<span class="n">ReferenceModelArguments</span><span class="p">:</span> <span class="n">ModelArguments</span> <span class="o">=</span> <span class="n">create_copied_dataclass</span><span class="p">(</span>
+    <span class="n">original_dataclass</span><span class="o">=</span><span class="n">ModelArguments</span><span class="p">,</span> 
+    <span class="n">field_prefix</span><span class="o">=</span><span class="s2">&quot;reference_&quot;</span><span class="p">,</span>
+    <span class="n">class_prefix</span><span class="o">=</span><span class="s2">&quot;Reference&quot;</span>
+<span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="main">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.html#lmflow.pipeline.utils.memory_safe_dpov2_align.main">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">main</span><span class="p">():</span>
+    <span class="c1"># Parses arguments</span>
+    <span class="n">parser</span> <span class="o">=</span> <span class="n">HfArgumentParser</span><span class="p">((</span>
+        <span class="n">ModelArguments</span><span class="p">,</span> 
+        <span class="n">ReferenceModelArguments</span><span class="p">,</span>
+        <span class="n">DatasetArguments</span><span class="p">,</span>
+        <span class="n">DPOv2AlignerArguments</span><span class="p">,</span>
+    <span class="p">))</span>
+    <span class="n">target_model_args</span><span class="p">,</span> <span class="n">ref_model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">aligner_args</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">parse_args_into_dataclasses</span><span class="p">()</span>
+        
+    <span class="n">ref_model_args_dict</span> <span class="o">=</span> <span class="n">remove_dataclass_attr_prefix</span><span class="p">(</span><span class="n">ref_model_args</span><span class="p">,</span> <span class="s2">&quot;reference_&quot;</span><span class="p">)</span>
+    <span class="n">ref_model_args</span> <span class="o">=</span> <span class="n">ModelArguments</span><span class="p">(</span><span class="o">**</span><span class="n">ref_model_args_dict</span><span class="p">)</span>
+
+    <span class="n">target_model</span> <span class="o">=</span> <span class="n">HFDecoderModel</span><span class="p">(</span><span class="n">target_model_args</span><span class="p">)</span>
+    <span class="n">ref_model</span> <span class="o">=</span> <span class="n">HFDecoderModel</span><span class="p">(</span><span class="n">ref_model_args</span><span class="p">)</span>
+    <span class="n">train_dataset</span> <span class="o">=</span> <span class="n">Dataset</span><span class="p">(</span><span class="n">data_args</span><span class="p">)</span>
+    <span class="n">eval_dataset</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">train_dataset</span><span class="o">.</span><span class="n">sample</span><span class="p">(</span>
+        <span class="n">n</span><span class="o">=</span><span class="mi">100</span><span class="p">,</span> 
+        <span class="n">seed</span><span class="o">=</span><span class="n">aligner_args</span><span class="o">.</span><span class="n">random_seed</span>
+    <span class="p">))</span>
+    
+    <span class="n">aligner</span> <span class="o">=</span> <span class="n">DPOv2Aligner</span><span class="p">(</span>
+        <span class="n">model_args</span><span class="o">=</span><span class="n">target_model_args</span><span class="p">,</span>
+        <span class="n">data_args</span><span class="o">=</span><span class="n">train_dataset</span><span class="o">.</span><span class="n">data_args</span><span class="p">,</span>
+        <span class="n">aligner_args</span><span class="o">=</span><span class="n">aligner_args</span><span class="p">,</span>
+        <span class="n">ref_model_args</span><span class="o">=</span><span class="n">ref_model</span><span class="o">.</span><span class="n">model_args</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="n">aligner</span><span class="o">.</span><span class="n">align</span><span class="p">(</span>
+        <span class="n">model</span><span class="o">=</span><span class="n">target_model</span><span class="p">,</span>
+        <span class="n">ref_model</span><span class="o">=</span><span class="n">ref_model</span><span class="p">,</span>
+        <span class="n">train_dataset</span><span class="o">=</span><span class="n">train_dataset</span><span class="p">,</span>
+        <span class="n">eval_dataset</span><span class="o">=</span><span class="n">eval_dataset</span><span class="p">,</span>
+    <span class="p">)</span></div>
+
+    
+
+<span class="k">if</span> <span class="vm">__name__</span> <span class="o">==</span> <span class="s2">&quot;__main__&quot;</span><span class="p">:</span>
+    <span class="n">main</span><span class="p">()</span>
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/utils/memory_safe_vllm_inference.html b/_modules/lmflow/pipeline/utils/memory_safe_vllm_inference.html
new file mode 100644
index 000000000..af897e430
--- /dev/null
+++ b/_modules/lmflow/pipeline/utils/memory_safe_vllm_inference.html
@@ -0,0 +1,515 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.utils.memory_safe_vllm_inference &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/utils/memory_safe_vllm_inference';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.memory_safe_vllm_inference</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.utils.memory_safe_vllm_inference</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+
+<span class="c1"># Note that this is only a workaround, since vllm</span>
+<span class="c1"># inference engine cannot release GPU memory properly by now. Please see this github </span>
+<span class="c1"># [issue](https://github.com/vllm-project/vllm/issues/1908).</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dict</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">HfArgumentParser</span>
+<span class="p">)</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.auto_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">AutoModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.vllm_inferencer</span><span class="w"> </span><span class="kn">import</span> <span class="n">VLLMInferencer</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">ModelArguments</span><span class="p">,</span> 
+    <span class="n">DatasetArguments</span><span class="p">,</span> 
+    <span class="n">AutoArguments</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="n">MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.html#lmflow.pipeline.utils.memory_safe_vllm_inference.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="main">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.html#lmflow.pipeline.utils.memory_safe_vllm_inference.main">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">main</span><span class="p">():</span>
+    <span class="c1"># Parses arguments</span>
+    <span class="n">pipeline_name</span> <span class="o">=</span> <span class="s2">&quot;vllm_inferencer&quot;</span>
+    <span class="n">PipelineArguments</span> <span class="o">=</span> <span class="n">AutoArguments</span><span class="o">.</span><span class="n">get_pipeline_args_class</span><span class="p">(</span><span class="n">pipeline_name</span><span class="p">)</span>
+
+    <span class="n">parser</span> <span class="o">=</span> <span class="n">HfArgumentParser</span><span class="p">((</span>
+        <span class="n">ModelArguments</span><span class="p">,</span> 
+        <span class="n">DatasetArguments</span><span class="p">,</span>
+        <span class="n">PipelineArguments</span>
+    <span class="p">))</span>
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">sys</span><span class="o">.</span><span class="n">argv</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span> <span class="ow">and</span> <span class="n">sys</span><span class="o">.</span><span class="n">argv</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;.json&quot;</span><span class="p">):</span>
+        <span class="c1"># If we pass only one argument to the script and it&#39;s the path to a json file,</span>
+        <span class="c1"># let&#39;s parse it to get our arguments.</span>
+        <span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">pipeline_args</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">parse_json_file</span><span class="p">(</span><span class="n">json_file</span><span class="o">=</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">abspath</span><span class="p">(</span><span class="n">sys</span><span class="o">.</span><span class="n">argv</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">pipeline_args</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">parse_args_into_dataclasses</span><span class="p">()</span>
+
+    <span class="n">dataset</span> <span class="o">=</span> <span class="n">Dataset</span><span class="p">(</span><span class="n">data_args</span><span class="p">)</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">AutoModel</span><span class="o">.</span><span class="n">get_model</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="n">tune_strategy</span><span class="o">=</span><span class="s1">&#39;none&#39;</span><span class="p">)</span>
+    <span class="n">inferencer</span> <span class="o">=</span> <span class="n">VLLMInferencer</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">pipeline_args</span><span class="p">)</span>
+
+    <span class="n">res</span> <span class="o">=</span> <span class="n">inferencer</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span>
+        <span class="n">model</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">,</span>
+        <span class="n">release_gpu</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">enable_decode_inference_result</span><span class="o">=</span><span class="n">pipeline_args</span><span class="o">.</span><span class="n">enable_decode_inference_result</span><span class="p">,</span>
+        <span class="n">enable_distributed_inference</span><span class="o">=</span><span class="n">pipeline_args</span><span class="o">.</span><span class="n">enable_distributed_inference</span><span class="p">,</span>
+        <span class="n">distributed_inference_num_instances</span><span class="o">=</span><span class="n">pipeline_args</span><span class="o">.</span><span class="n">distributed_inference_num_instances</span><span class="p">,</span>
+        <span class="n">inference_batch_size</span><span class="o">=</span><span class="n">pipeline_args</span><span class="o">.</span><span class="n">vllm_inference_batch_size</span><span class="p">,</span>
+    <span class="p">)</span>
+    
+    <span class="c1"># use this as a flag, stdout will be captured by the pipeline</span>
+    <span class="nb">print</span><span class="p">(</span><span class="n">MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG</span><span class="p">)</span></div>
+
+    
+
+<span class="k">if</span> <span class="vm">__name__</span> <span class="o">==</span> <span class="s2">&quot;__main__&quot;</span><span class="p">:</span>
+    <span class="n">main</span><span class="p">()</span>
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/utils/peft_trainer.html b/_modules/lmflow/pipeline/utils/peft_trainer.html
new file mode 100644
index 000000000..4ee8856fe
--- /dev/null
+++ b/_modules/lmflow/pipeline/utils/peft_trainer.html
@@ -0,0 +1,540 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.utils.peft_trainer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/utils/peft_trainer';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.peft_trainer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.utils.peft_trainer</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;Trainer for Peft models</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">__future__</span><span class="w"> </span><span class="kn">import</span> <span class="n">absolute_import</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">Trainer</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.trainer_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">PREFIX_CHECKPOINT_DIR</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.trainer_callback</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">TrainerCallback</span><span class="p">,</span>
+    <span class="n">TrainerControl</span><span class="p">,</span>
+    <span class="n">TrainerState</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.training_args</span><span class="w"> </span><span class="kn">import</span> <span class="n">TrainingArguments</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
+
+<div class="viewcode-block" id="PeftTrainer">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftTrainer">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">PeftTrainer</span><span class="p">(</span><span class="n">Trainer</span><span class="p">):</span>
+<div class="viewcode-block" id="PeftTrainer._save_checkpoint">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftTrainer._save_checkpoint">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_save_checkpoint</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">trial</span><span class="p">,</span> <span class="n">metrics</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot; Don&#39;t save base model, optimizer etc.</span>
+<span class="sd">            but create checkpoint folder (needed for saving adapter) &quot;&quot;&quot;</span>
+        <span class="n">checkpoint_folder</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">PREFIX_CHECKPOINT_DIR</span><span class="si">}</span><span class="s2">-</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span><span class="si">}</span><span class="s2">&quot;</span>
+
+        <span class="n">run_dir</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_output_dir</span><span class="p">(</span><span class="n">trial</span><span class="o">=</span><span class="n">trial</span><span class="p">)</span>
+        <span class="n">output_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">run_dir</span><span class="p">,</span> <span class="n">checkpoint_folder</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">metrics</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">metric_for_best_model</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">metric_to_check</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">metric_for_best_model</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">metric_to_check</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;eval_&quot;</span><span class="p">):</span>
+                <span class="n">metric_to_check</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;eval_</span><span class="si">{</span><span class="n">metric_to_check</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="n">metric_value</span> <span class="o">=</span> <span class="n">metrics</span><span class="p">[</span><span class="n">metric_to_check</span><span class="p">]</span>
+
+            <span class="n">operator</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">greater</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">greater_is_better</span> <span class="k">else</span> <span class="n">np</span><span class="o">.</span><span class="n">less</span>
+            <span class="k">if</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_metric</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span> <span class="ow">is</span> <span class="kc">None</span>
+                <span class="ow">or</span> <span class="n">operator</span><span class="p">(</span><span class="n">metric_value</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_metric</span><span class="p">)):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_metric</span> <span class="o">=</span> <span class="n">metric_value</span>
+
+                <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span> <span class="o">=</span> <span class="n">output_dir</span>
+
+        <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_rotate_checkpoints</span><span class="p">(</span><span class="n">use_mtime</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">output_dir</span><span class="o">=</span><span class="n">run_dir</span><span class="p">)</span></div>
+</div>
+
+
+<div class="viewcode-block" id="PeftSavingCallback">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">PeftSavingCallback</span><span class="p">(</span><span class="n">TrainerCallback</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot; Correctly save PEFT model and not full model &quot;&quot;&quot;</span>
+<div class="viewcode-block" id="PeftSavingCallback._save">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback._save">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_save</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">folder</span><span class="p">):</span>
+        <span class="k">if</span> <span class="n">folder</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">folder</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">peft_model_path</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">folder</span><span class="p">,</span> <span class="s2">&quot;adapter_model&quot;</span><span class="p">)</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="n">peft_model_path</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="PeftSavingCallback.on_train_end">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_train_end">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">on_train_end</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">:</span> <span class="n">TrainingArguments</span><span class="p">,</span> <span class="n">state</span><span class="p">:</span> <span class="n">TrainerState</span><span class="p">,</span>
+            <span class="n">control</span><span class="p">:</span> <span class="n">TrainerControl</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot; Save final best model adapter &quot;&quot;&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_save</span><span class="p">(</span><span class="n">kwargs</span><span class="p">[</span><span class="s1">&#39;model&#39;</span><span class="p">],</span> <span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="PeftSavingCallback.on_epoch_end">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_epoch_end">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">on_epoch_end</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">args</span><span class="p">:</span> <span class="n">TrainingArguments</span><span class="p">,</span> <span class="n">state</span><span class="p">:</span> <span class="n">TrainerState</span><span class="p">,</span>
+            <span class="n">control</span><span class="p">:</span> <span class="n">TrainerControl</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot; Save intermediate model adapters in case of interrupted training &quot;&quot;&quot;</span>
+        <span class="n">folder</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">PREFIX_CHECKPOINT_DIR</span><span class="si">}</span><span class="s2">-</span><span class="si">{</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_save</span><span class="p">(</span><span class="n">kwargs</span><span class="p">[</span><span class="s1">&#39;model&#39;</span><span class="p">],</span> <span class="n">folder</span><span class="p">)</span></div>
+
+        
+<div class="viewcode-block" id="PeftSavingCallback.on_save">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_save">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">on_save</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">args</span><span class="p">:</span> <span class="n">TrainingArguments</span><span class="p">,</span>
+        <span class="n">state</span><span class="p">:</span> <span class="n">TrainerState</span><span class="p">,</span>
+        <span class="n">control</span><span class="p">:</span> <span class="n">TrainerControl</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="n">checkpoint_folder</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span>
+            <span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">PREFIX_CHECKPOINT_DIR</span><span class="si">}</span><span class="s2">-</span><span class="si">{</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="p">)</span>       
+        <span class="bp">self</span><span class="o">.</span><span class="n">_save</span><span class="p">(</span><span class="n">kwargs</span><span class="p">[</span><span class="s1">&#39;model&#39;</span><span class="p">],</span> <span class="n">checkpoint_folder</span><span class="p">)</span>
+
+        <span class="n">peft_model_path</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint_folder</span><span class="p">,</span> <span class="s2">&quot;adapter_model&quot;</span><span class="p">)</span>
+        <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;model&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="n">peft_model_path</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">control</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/utils/raft_trainer.html b/_modules/lmflow/pipeline/utils/raft_trainer.html
new file mode 100644
index 000000000..ad6bdecda
--- /dev/null
+++ b/_modules/lmflow/pipeline/utils/raft_trainer.html
@@ -0,0 +1,4580 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.utils.raft_trainer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/utils/raft_trainer';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.raft_trainer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.utils.raft_trainer</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span><span class="w"> </span><span class="nn">contextlib</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">functools</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">glob</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">inspect</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">random</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">re</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">shutil</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">time</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">warnings</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">collections.abc</span><span class="w"> </span><span class="kn">import</span> <span class="n">Mapping</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">distutils.util</span><span class="w"> </span><span class="kn">import</span> <span class="n">strtobool</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">pathlib</span><span class="w"> </span><span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">TYPE_CHECKING</span><span class="p">,</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">tqdm.auto</span><span class="w"> </span><span class="kn">import</span> <span class="n">tqdm</span>
+
+
+<span class="c1"># Integrations must be imported before ML frameworks:</span>
+<span class="c1"># isort: off</span>
+
+<span class="c1">### Fix the import bug in the latest version</span>
+<span class="k">try</span><span class="p">:</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.integrations</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+        <span class="n">default_hp_search_backend</span><span class="p">,</span>
+        <span class="n">get_reporting_integration_callbacks</span><span class="p">,</span>
+        <span class="n">hp_params</span><span class="p">,</span>
+        <span class="n">is_fairscale_available</span><span class="p">,</span>
+        <span class="n">is_optuna_available</span><span class="p">,</span>
+        <span class="n">is_ray_tune_available</span><span class="p">,</span>
+        <span class="n">is_sigopt_available</span><span class="p">,</span>
+        <span class="n">is_wandb_available</span><span class="p">,</span>
+        <span class="n">run_hp_search_optuna</span><span class="p">,</span>
+        <span class="n">run_hp_search_ray</span><span class="p">,</span>
+        <span class="n">run_hp_search_sigopt</span><span class="p">,</span>
+        <span class="n">run_hp_search_wandb</span><span class="p">,</span>
+    <span class="p">)</span>
+<span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.integrations</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">get_reporting_integration_callbacks</span><span class="p">,</span>
+    <span class="n">hp_params</span><span class="p">,</span>
+    <span class="n">is_fairscale_available</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.hyperparameter_search</span><span class="w"> </span><span class="kn">import</span> <span class="n">default_hp_search_backend</span><span class="p">,</span><span class="n">ALL_HYPERPARAMETER_SEARCH_BACKENDS</span>
+
+    
+
+<span class="c1"># isort: on</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.distributed</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">dist</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">huggingface_hub</span><span class="w"> </span><span class="kn">import</span> <span class="n">Repository</span><span class="p">,</span> <span class="n">create_repo</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">packaging</span><span class="w"> </span><span class="kn">import</span> <span class="n">version</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch</span><span class="w"> </span><span class="kn">import</span> <span class="n">nn</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.utils.data</span><span class="w"> </span><span class="kn">import</span> <span class="n">DataLoader</span><span class="p">,</span> <span class="n">Dataset</span><span class="p">,</span> <span class="n">RandomSampler</span><span class="p">,</span> <span class="n">SequentialSampler</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch.utils.data.distributed</span><span class="w"> </span><span class="kn">import</span> <span class="n">DistributedSampler</span>
+
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.configuration_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">PretrainedConfig</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.data.data_collator</span><span class="w"> </span><span class="kn">import</span> <span class="n">DataCollator</span><span class="p">,</span> <span class="n">DataCollatorWithPadding</span><span class="p">,</span> <span class="n">default_data_collator</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.debug_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">DebugOption</span><span class="p">,</span> <span class="n">DebugUnderflowOverflow</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.deepspeed</span><span class="w"> </span><span class="kn">import</span> <span class="n">deepspeed_init</span><span class="p">,</span> <span class="n">is_deepspeed_zero3_enabled</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.dependency_versions_check</span><span class="w"> </span><span class="kn">import</span> <span class="n">dep_version_check</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.modelcard</span><span class="w"> </span><span class="kn">import</span> <span class="n">TrainingSummary</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.modeling_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">PreTrainedModel</span><span class="p">,</span> <span class="n">load_sharded_checkpoint</span><span class="p">,</span> <span class="n">unwrap_model</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.models.auto.modeling_auto</span><span class="w"> </span><span class="kn">import</span> <span class="n">MODEL_FOR_CAUSAL_LM_MAPPING_NAMES</span><span class="p">,</span> <span class="n">MODEL_MAPPING_NAMES</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.optimization</span><span class="w"> </span><span class="kn">import</span> <span class="n">Adafactor</span><span class="p">,</span> <span class="n">get_scheduler</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.pytorch_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">ALL_LAYERNORM_LAYERS</span>
+
+<div class="viewcode-block" id="is_torch_greater_or_equal_than_1_10">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.is_torch_greater_or_equal_than_1_10">[docs]</a>
+<span class="n">is_torch_greater_or_equal_than_1_10</span> <span class="o">=</span> <span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">__version__</span><span class="p">)</span> <span class="o">&gt;=</span> <span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="s2">&quot;1.10&quot;</span><span class="p">)</span></div>
+
+<div class="viewcode-block" id="is_torch_less_than_1_11">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.is_torch_less_than_1_11">[docs]</a>
+<span class="n">is_torch_less_than_1_11</span> <span class="o">=</span> <span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">__version__</span><span class="p">)</span> <span class="o">&lt;</span> <span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="s2">&quot;1.11&quot;</span><span class="p">)</span></div>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.tokenization_utils_base</span><span class="w"> </span><span class="kn">import</span> <span class="n">PreTrainedTokenizerBase</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.trainer_callback</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">CallbackHandler</span><span class="p">,</span>
+    <span class="n">DefaultFlowCallback</span><span class="p">,</span>
+    <span class="n">PrinterCallback</span><span class="p">,</span>
+    <span class="n">ProgressCallback</span><span class="p">,</span>
+    <span class="n">TrainerCallback</span><span class="p">,</span>
+    <span class="n">TrainerControl</span><span class="p">,</span>
+    <span class="n">TrainerState</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.trainer_pt_utils</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">DistributedLengthGroupedSampler</span><span class="p">,</span>
+    <span class="n">DistributedSamplerWithLoop</span><span class="p">,</span>
+    <span class="n">DistributedTensorGatherer</span><span class="p">,</span>
+    <span class="n">IterableDatasetShard</span><span class="p">,</span>
+    <span class="n">LabelSmoother</span><span class="p">,</span>
+    <span class="n">LengthGroupedSampler</span><span class="p">,</span>
+    <span class="n">SequentialDistributedSampler</span><span class="p">,</span>
+    <span class="n">ShardSampler</span><span class="p">,</span>
+    <span class="n">distributed_broadcast_scalars</span><span class="p">,</span>
+    <span class="n">distributed_concat</span><span class="p">,</span>
+    <span class="n">find_batch_size</span><span class="p">,</span>
+    <span class="n">get_module_class_from_name</span><span class="p">,</span>
+    <span class="n">get_parameter_names</span><span class="p">,</span>
+    <span class="n">nested_concat</span><span class="p">,</span>
+    <span class="n">nested_detach</span><span class="p">,</span>
+    <span class="n">nested_numpify</span><span class="p">,</span>
+    <span class="n">nested_truncate</span><span class="p">,</span>
+    <span class="n">nested_xla_mesh_reduce</span><span class="p">,</span>
+    <span class="n">reissue_pt_warnings</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.trainer_utils</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">PREFIX_CHECKPOINT_DIR</span><span class="p">,</span>
+    <span class="n">BestRun</span><span class="p">,</span>
+    <span class="n">EvalLoopOutput</span><span class="p">,</span>
+    <span class="n">EvalPrediction</span><span class="p">,</span>
+    <span class="n">FSDPOption</span><span class="p">,</span>
+    <span class="n">HPSearchBackend</span><span class="p">,</span>
+    <span class="n">HubStrategy</span><span class="p">,</span>
+    <span class="n">IntervalStrategy</span><span class="p">,</span>
+    <span class="n">PredictionOutput</span><span class="p">,</span>
+    <span class="n">RemoveColumnsCollator</span><span class="p">,</span>
+    <span class="n">ShardedDDPOption</span><span class="p">,</span>
+    <span class="n">TrainerMemoryTracker</span><span class="p">,</span>
+    <span class="n">TrainOutput</span><span class="p">,</span>
+    <span class="n">default_compute_objective</span><span class="p">,</span>
+    <span class="n">denumpify_detensorize</span><span class="p">,</span>
+    <span class="n">enable_full_determinism</span><span class="p">,</span>
+    <span class="n">find_executable_batch_size</span><span class="p">,</span>
+    <span class="n">get_last_checkpoint</span><span class="p">,</span>
+    <span class="n">has_length</span><span class="p">,</span>
+    <span class="n">number_of_arguments</span><span class="p">,</span>
+    <span class="n">seed_worker</span><span class="p">,</span>
+    <span class="n">set_seed</span><span class="p">,</span>
+    <span class="n">speed_metrics</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.training_args</span><span class="w"> </span><span class="kn">import</span> <span class="n">OptimizerNames</span><span class="p">,</span> <span class="n">ParallelMode</span><span class="p">,</span> <span class="n">TrainingArguments</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.utils</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">CONFIG_NAME</span><span class="p">,</span>
+    <span class="n">WEIGHTS_INDEX_NAME</span><span class="p">,</span>
+    <span class="n">WEIGHTS_NAME</span><span class="p">,</span>
+    <span class="n">can_return_loss</span><span class="p">,</span>
+    <span class="n">find_labels</span><span class="p">,</span>
+    <span class="n">get_full_repo_name</span><span class="p">,</span>
+    <span class="n">is_accelerate_available</span><span class="p">,</span>
+    <span class="n">is_apex_available</span><span class="p">,</span>
+    <span class="n">is_datasets_available</span><span class="p">,</span>
+    <span class="n">is_in_notebook</span><span class="p">,</span>
+    <span class="n">is_ipex_available</span><span class="p">,</span>
+    <span class="n">is_sagemaker_dp_enabled</span><span class="p">,</span>
+    <span class="n">is_sagemaker_mp_enabled</span><span class="p">,</span>
+    <span class="n">is_torch_compile_available</span><span class="p">,</span>
+    <span class="n">is_torch_neuroncore_available</span><span class="p">,</span>
+    <span class="n">is_torch_tpu_available</span><span class="p">,</span>
+    <span class="n">logging</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.utils.generic</span><span class="w"> </span><span class="kn">import</span> <span class="n">ContextManagers</span>
+
+
+<div class="viewcode-block" id="_is_native_cpu_amp_available">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer._is_native_cpu_amp_available">[docs]</a>
+<span class="n">_is_native_cpu_amp_available</span> <span class="o">=</span> <span class="n">is_torch_greater_or_equal_than_1_10</span></div>
+
+
+<div class="viewcode-block" id="DEFAULT_CALLBACKS">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.DEFAULT_CALLBACKS">[docs]</a>
+<span class="n">DEFAULT_CALLBACKS</span> <span class="o">=</span> <span class="p">[</span><span class="n">DefaultFlowCallback</span><span class="p">]</span></div>
+
+<span class="n">DEFAULT_PROGRESS_CALLBACK</span> <span class="o">=</span> <span class="n">ProgressCallback</span>
+
+<span class="k">if</span> <span class="n">is_in_notebook</span><span class="p">():</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.utils.notebook</span><span class="w"> </span><span class="kn">import</span> <span class="n">NotebookProgressCallback</span>
+
+<div class="viewcode-block" id="DEFAULT_PROGRESS_CALLBACK">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.DEFAULT_PROGRESS_CALLBACK">[docs]</a>
+    <span class="n">DEFAULT_PROGRESS_CALLBACK</span> <span class="o">=</span> <span class="n">NotebookProgressCallback</span></div>
+
+
+<span class="k">if</span> <span class="n">is_apex_available</span><span class="p">():</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">apex</span><span class="w"> </span><span class="kn">import</span> <span class="n">amp</span>
+
+<span class="k">if</span> <span class="n">is_datasets_available</span><span class="p">():</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">datasets</span>
+
+<span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">(</span><span class="n">check_device</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">torch_xla.core.xla_model</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">xm</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">torch_xla.debug.metrics</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">met</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">torch_xla.distributed.parallel_loader</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">pl</span>
+
+<span class="k">if</span> <span class="n">is_fairscale_available</span><span class="p">():</span>
+    <span class="n">dep_version_check</span><span class="p">(</span><span class="s2">&quot;fairscale&quot;</span><span class="p">)</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">fairscale</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">fairscale.nn.data_parallel</span><span class="w"> </span><span class="kn">import</span> <span class="n">FullyShardedDataParallel</span> <span class="k">as</span> <span class="n">FullyShardedDDP</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">fairscale.nn.data_parallel</span><span class="w"> </span><span class="kn">import</span> <span class="n">ShardedDataParallel</span> <span class="k">as</span> <span class="n">ShardedDDP</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">fairscale.nn.wrap</span><span class="w"> </span><span class="kn">import</span> <span class="n">auto_wrap</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">fairscale.optim</span><span class="w"> </span><span class="kn">import</span> <span class="n">OSS</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">fairscale.optim.grad_scaler</span><span class="w"> </span><span class="kn">import</span> <span class="n">ShardedGradScaler</span>
+
+
+<span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">smdistributed.modelparallel.torch</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">smp</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">smdistributed.modelparallel</span><span class="w"> </span><span class="kn">import</span> <span class="n">__version__</span> <span class="k">as</span> <span class="n">SMP_VERSION</span>
+
+<div class="viewcode-block" id="IS_SAGEMAKER_MP_POST_1_10">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.IS_SAGEMAKER_MP_POST_1_10">[docs]</a>
+    <span class="n">IS_SAGEMAKER_MP_POST_1_10</span> <span class="o">=</span> <span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">SMP_VERSION</span><span class="p">)</span> <span class="o">&gt;=</span> <span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="s2">&quot;1.10&quot;</span><span class="p">)</span></div>
+
+
+    <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.trainer_pt_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">smp_forward_backward</span><span class="p">,</span> <span class="n">smp_forward_only</span><span class="p">,</span> <span class="n">smp_gather</span><span class="p">,</span> <span class="n">smp_nested_concat</span>
+<span class="k">else</span><span class="p">:</span>
+    <span class="n">IS_SAGEMAKER_MP_POST_1_10</span> <span class="o">=</span> <span class="kc">False</span>
+
+
+<div class="viewcode-block" id="skip_first_batches">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.skip_first_batches">[docs]</a>
+<span class="n">skip_first_batches</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">get_logger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<span class="c1"># Name of the files used for checkpointing</span>
+<div class="viewcode-block" id="TRAINING_ARGS_NAME">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.TRAINING_ARGS_NAME">[docs]</a>
+<span class="n">TRAINING_ARGS_NAME</span> <span class="o">=</span> <span class="s2">&quot;training_args.bin&quot;</span></div>
+
+<div class="viewcode-block" id="TRAINER_STATE_NAME">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.TRAINER_STATE_NAME">[docs]</a>
+<span class="n">TRAINER_STATE_NAME</span> <span class="o">=</span> <span class="s2">&quot;trainer_state.json&quot;</span></div>
+
+<div class="viewcode-block" id="OPTIMIZER_NAME">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.OPTIMIZER_NAME">[docs]</a>
+<span class="n">OPTIMIZER_NAME</span> <span class="o">=</span> <span class="s2">&quot;optimizer.pt&quot;</span></div>
+
+<div class="viewcode-block" id="SCHEDULER_NAME">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.SCHEDULER_NAME">[docs]</a>
+<span class="n">SCHEDULER_NAME</span> <span class="o">=</span> <span class="s2">&quot;scheduler.pt&quot;</span></div>
+
+<div class="viewcode-block" id="SCALER_NAME">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.SCALER_NAME">[docs]</a>
+<span class="n">SCALER_NAME</span> <span class="o">=</span> <span class="s2">&quot;scaler.pt&quot;</span></div>
+
+
+
+<div class="viewcode-block" id="RaftTrainer">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">RaftTrainer</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Trainer is a simple but feature-complete training and eval loop for PyTorch, optimized for 🤗 Transformers.</span>
+<span class="sd">    Args:</span>
+<span class="sd">        model ([`PreTrainedModel`] or `torch.nn.Module`, *optional*):</span>
+<span class="sd">            The model to train, evaluate or use for predictions. If not provided, a `model_init` must be passed.</span>
+<span class="sd">            &lt;Tip&gt;</span>
+<span class="sd">            [`Trainer`] is optimized to work with the [`PreTrainedModel`] provided by the library. You can still use</span>
+<span class="sd">            your own models defined as `torch.nn.Module` as long as they work the same way as the 🤗 Transformers</span>
+<span class="sd">            models.</span>
+<span class="sd">            &lt;/Tip&gt;</span>
+<span class="sd">        args ([`TrainingArguments`], *optional*):</span>
+<span class="sd">            The arguments to tweak for training. Will default to a basic instance of [`TrainingArguments`] with the</span>
+<span class="sd">            `output_dir` set to a directory named *tmp_trainer* in the current directory if not provided.</span>
+<span class="sd">        data_collator (`DataCollator`, *optional*):</span>
+<span class="sd">            The function to use to form a batch from a list of elements of `train_dataset` or `eval_dataset`. Will</span>
+<span class="sd">            default to [`default_data_collator`] if no `tokenizer` is provided, an instance of</span>
+<span class="sd">            [`DataCollatorWithPadding`] otherwise.</span>
+<span class="sd">        train_dataset (`torch.utils.data.Dataset` or `torch.utils.data.IterableDataset`, *optional*):</span>
+<span class="sd">            The dataset to use for training. If it is a [`~datasets.Dataset`], columns not accepted by the</span>
+<span class="sd">            `model.forward()` method are automatically removed.</span>
+<span class="sd">            Note that if it&#39;s a `torch.utils.data.IterableDataset` with some randomization and you are training in a</span>
+<span class="sd">            distributed fashion, your iterable dataset should either use a internal attribute `generator` that is a</span>
+<span class="sd">            `torch.Generator` for the randomization that must be identical on all processes (and the Trainer will</span>
+<span class="sd">            manually set the seed of this `generator` at each epoch) or have a `set_epoch()` method that internally</span>
+<span class="sd">            sets the seed of the RNGs used.</span>
+<span class="sd">        eval_dataset (Union[`torch.utils.data.Dataset`, Dict[str, `torch.utils.data.Dataset`]), *optional*):</span>
+<span class="sd">             The dataset to use for evaluation. If it is a [`~datasets.Dataset`], columns not accepted by the</span>
+<span class="sd">             `model.forward()` method are automatically removed. If it is a dictionary, it will evaluate on each</span>
+<span class="sd">             dataset prepending the dictionary key to the metric name.</span>
+<span class="sd">        tokenizer ([`PreTrainedTokenizerBase`], *optional*):</span>
+<span class="sd">            The tokenizer used to preprocess the data. If provided, will be used to automatically pad the inputs to the</span>
+<span class="sd">            maximum length when batching inputs, and it will be saved along the model to make it easier to rerun an</span>
+<span class="sd">            interrupted training or reuse the fine-tuned model.</span>
+<span class="sd">        model_init (`Callable[[], PreTrainedModel]`, *optional*):</span>
+<span class="sd">            A function that instantiates the model to be used. If provided, each call to [`~Trainer.train`] will start</span>
+<span class="sd">            from a new instance of the model as given by this function.</span>
+<span class="sd">            The function may have zero argument, or a single one containing the optuna/Ray Tune/SigOpt trial object, to</span>
+<span class="sd">            be able to choose different architectures according to hyper parameters (such as layer count, sizes of</span>
+<span class="sd">            inner layers, dropout probabilities etc).</span>
+<span class="sd">        compute_metrics (`Callable[[EvalPrediction], Dict]`, *optional*):</span>
+<span class="sd">            The function that will be used to compute metrics at evaluation. Must take a [`EvalPrediction`] and return</span>
+<span class="sd">            a dictionary string to metric values.</span>
+<span class="sd">        callbacks (List of [`TrainerCallback`], *optional*):</span>
+<span class="sd">            A list of callbacks to customize the training loop. Will add those to the list of default callbacks</span>
+<span class="sd">            detailed in [here](callback).</span>
+<span class="sd">            If you want to remove one of the default callbacks used, use the [`Trainer.remove_callback`] method.</span>
+<span class="sd">        optimizers (`Tuple[torch.optim.Optimizer, torch.optim.lr_scheduler.LambdaLR]`, *optional*): A tuple</span>
+<span class="sd">            containing the optimizer and the scheduler to use. Will default to an instance of [`AdamW`] on your model</span>
+<span class="sd">            and a scheduler given by [`get_linear_schedule_with_warmup`] controlled by `args`.</span>
+<span class="sd">        preprocess_logits_for_metrics (`Callable[[torch.Tensor, torch.Tensor], torch.Tensor]`, *optional*):</span>
+<span class="sd">            A function that preprocess the logits right before caching them at each evaluation step. Must take two</span>
+<span class="sd">            tensors, the logits and the labels, and return the logits once processed as desired. The modifications made</span>
+<span class="sd">            by this function will be reflected in the predictions received by `compute_metrics`.</span>
+<span class="sd">            Note that the labels (second parameter) will be `None` if the dataset does not have them.</span>
+<span class="sd">    Important attributes:</span>
+<span class="sd">        - **model** -- Always points to the core model. If using a transformers model, it will be a [`PreTrainedModel`]</span>
+<span class="sd">          subclass.</span>
+<span class="sd">        - **model_wrapped** -- Always points to the most external model in case one or more other modules wrap the</span>
+<span class="sd">          original model. This is the model that should be used for the forward pass. For example, under `DeepSpeed`,</span>
+<span class="sd">          the inner model is wrapped in `DeepSpeed` and then again in `torch.nn.DistributedDataParallel`. If the inner</span>
+<span class="sd">          model hasn&#39;t been wrapped, then `self.model_wrapped` is the same as `self.model`.</span>
+<span class="sd">        - **is_model_parallel** -- Whether or not a model has been switched to a model parallel mode (different from</span>
+<span class="sd">          data parallelism, this means some of the model layers are split on different GPUs).</span>
+<span class="sd">        - **place_model_on_device** -- Whether or not to automatically place the model on the device - it will be set</span>
+<span class="sd">          to `False` if model parallel or deepspeed is used, or if the default</span>
+<span class="sd">          `TrainingArguments.place_model_on_device` is overridden to return `False` .</span>
+<span class="sd">        - **is_in_train** -- Whether or not a model is currently running `train` (e.g. when `evaluate` is called while</span>
+<span class="sd">          in `train`)</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.trainer_pt_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">_get_learning_rate</span><span class="p">,</span> <span class="n">log_metrics</span><span class="p">,</span> <span class="n">metrics_format</span><span class="p">,</span> <span class="n">save_metrics</span><span class="p">,</span> <span class="n">save_state</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">PreTrainedModel</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">args</span><span class="p">:</span> <span class="n">TrainingArguments</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">data_collator</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">DataCollator</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">train_dataset</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dataset</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">eval_dataset</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">Dataset</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Dataset</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">PreTrainedTokenizerBase</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">model_init</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[],</span> <span class="n">PreTrainedModel</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">compute_metrics</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[</span><span class="n">EvalPrediction</span><span class="p">],</span> <span class="n">Dict</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">callbacks</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">TrainerCallback</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">optimizers</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">Optimizer</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">lr_scheduler</span><span class="o">.</span><span class="n">LambdaLR</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span>
+        <span class="n">preprocess_logits_for_metrics</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="c1">############</span>
+<div class="viewcode-block" id="RaftTrainer.save_counter">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.save_counter">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">save_counter</span> <span class="o">=</span> <span class="mi">0</span></div>
+
+        <span class="c1">##############</span>
+        <span class="k">if</span> <span class="n">args</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">output_dir</span> <span class="o">=</span> <span class="s2">&quot;tmp_trainer&quot;</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;No `TrainingArguments` passed, using `output_dir=</span><span class="si">{</span><span class="n">output_dir</span><span class="si">}</span><span class="s2">`.&quot;</span><span class="p">)</span>
+            <span class="n">args</span> <span class="o">=</span> <span class="n">TrainingArguments</span><span class="p">(</span><span class="n">output_dir</span><span class="o">=</span><span class="n">output_dir</span><span class="p">)</span>
+<div class="viewcode-block" id="RaftTrainer.args">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">args</span> <span class="o">=</span> <span class="n">args</span></div>
+
+        <span class="c1"># Seed must be set before instantiating the model when using model</span>
+        <span class="n">enable_full_determinism</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">seed</span><span class="p">)</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">full_determinism</span> <span class="k">else</span> <span class="n">set_seed</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">seed</span><span class="p">)</span>
+<div class="viewcode-block" id="RaftTrainer.hp_name">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.hp_name">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">hp_name</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="RaftTrainer.deepspeed">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.deepspeed">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="RaftTrainer.is_in_train">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_in_train">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">is_in_train</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+
+        <span class="c1"># memory metrics - must set up as early as possible</span>
+<div class="viewcode-block" id="RaftTrainer._memory_tracker">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._memory_tracker">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_memory_tracker</span> <span class="o">=</span> <span class="n">TrainerMemoryTracker</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">skip_memory_metrics</span><span class="p">)</span></div>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_memory_tracker</span><span class="o">.</span><span class="n">start</span><span class="p">()</span>
+
+        <span class="c1"># set the correct log level depending on the node</span>
+        <span class="n">log_level</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">get_process_log_level</span><span class="p">()</span>
+        <span class="n">logging</span><span class="o">.</span><span class="n">set_verbosity</span><span class="p">(</span><span class="n">log_level</span><span class="p">)</span>
+
+        <span class="c1"># force device and distributed setup init explicitly</span>
+        <span class="n">args</span><span class="o">.</span><span class="n">_setup_devices</span>
+
+        <span class="k">if</span> <span class="n">model</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">model_init</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model_init</span> <span class="o">=</span> <span class="n">model_init</span>
+                <span class="n">model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">call_model_init</span><span class="p">()</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;`Trainer` requires either a `model` or `model_init` argument&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">model_init</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                    <span class="s2">&quot;`Trainer` requires either a `model` or `model_init` argument, but not both. `model_init` will&quot;</span>
+                    <span class="s2">&quot; overwrite your model when calling the `train` method. This will become a fatal error in the next&quot;</span>
+                    <span class="s2">&quot; release.&quot;</span><span class="p">,</span>
+                    <span class="ne">FutureWarning</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model_init</span> <span class="o">=</span> <span class="n">model_init</span>
+
+        <span class="k">if</span> <span class="n">model</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span> <span class="ow">in</span> <span class="n">MODEL_MAPPING_NAMES</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;The model you have picked (</span><span class="si">{</span><span class="n">model</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">) cannot be used as is for training: it only &quot;</span>
+                <span class="s2">&quot;computes hidden states and does not accept any labels. You should choose a model with a head &quot;</span>
+                <span class="s2">&quot;suitable for your task like any of the `AutoModelForXxx` listed at &quot;</span>
+                <span class="s2">&quot;https://huggingface.co/docs/transformers/model_doc/auto.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="s2">&quot;is_parallelizable&quot;</span><span class="p">)</span> <span class="ow">and</span> <span class="n">model</span><span class="o">.</span><span class="n">is_parallelizable</span> <span class="ow">and</span> <span class="n">model</span><span class="o">.</span><span class="n">model_parallel</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">is_model_parallel</span> <span class="o">=</span> <span class="kc">True</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">is_model_parallel</span> <span class="o">=</span> <span class="kc">False</span>
+
+        <span class="c1"># At this stage the model is already loaded</span>
+        <span class="k">if</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="s2">&quot;is_loaded_in_8bit&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">):</span>
+            <span class="k">if</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="s2">&quot;_is_int8_training_enabled&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">):</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span>
+                    <span class="s2">&quot;The model is loaded in 8-bit precision. To train this model you need to add additional modules&quot;</span>
+                    <span class="s2">&quot; inside the model such as adapters using `peft` library and freeze the model weights. Please&quot;</span>
+                    <span class="s2">&quot; check &quot;</span>
+                    <span class="s2">&quot; the examples in https://github.com/huggingface/peft for more details.&quot;</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="s2">&quot;The model you want to train is loaded in 8-bit precision.  if you want to fine-tune an 8-bit&quot;</span>
+                    <span class="s2">&quot; model, please make sure that you have installed `bitsandbytes&gt;=0.37.0`. &quot;</span>
+                <span class="p">)</span>
+
+        <span class="c1"># Setup Sharded DDP training</span>
+<div class="viewcode-block" id="RaftTrainer.sharded_ddp">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.sharded_ddp">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">sharded_ddp</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="s2">&quot;Using --sharded_ddp xxx together with --deepspeed is not possible, deactivate one of those flags.&quot;</span>
+                <span class="p">)</span>
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="s2">&quot;Using --sharded_ddp xxx together with --fsdp is not possible, deactivate one of those flags.&quot;</span>
+                <span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">==</span> <span class="o">-</span><span class="mi">1</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Using sharded DDP only works in distributed training.&quot;</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="ow">not</span> <span class="n">is_fairscale_available</span><span class="p">():</span>
+                <span class="k">raise</span> <span class="ne">ImportError</span><span class="p">(</span><span class="s2">&quot;Sharded DDP training requires fairscale: `pip install fairscale`.&quot;</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">SIMPLE</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">args</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="ow">and</span> <span class="n">FullyShardedDDP</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ImportError</span><span class="p">(</span>
+                    <span class="s2">&quot;Sharded DDP in a mode other than simple training requires fairscale version &gt;= 0.3, found &quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">fairscale</span><span class="o">.</span><span class="n">__version__</span><span class="si">}</span><span class="s2">. Upgrade your fairscale library: `pip install --upgrade fairscale`.&quot;</span>
+                <span class="p">)</span>
+            <span class="k">elif</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">SIMPLE</span> <span class="ow">in</span> <span class="n">args</span><span class="o">.</span><span class="n">sharded_ddp</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="o">=</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">SIMPLE</span>
+            <span class="k">elif</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">ZERO_DP_2</span> <span class="ow">in</span> <span class="n">args</span><span class="o">.</span><span class="n">sharded_ddp</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="o">=</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">ZERO_DP_2</span>
+            <span class="k">elif</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">ZERO_DP_3</span> <span class="ow">in</span> <span class="n">args</span><span class="o">.</span><span class="n">sharded_ddp</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="o">=</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">ZERO_DP_3</span>
+
+<div class="viewcode-block" id="RaftTrainer.fsdp">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.fsdp">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">fsdp</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="s2">&quot;Using --fsdp xxx together with --deepspeed is not possible, deactivate one of those flags.&quot;</span>
+                <span class="p">)</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span><span class="p">[</span><span class="s2">&quot;xla&quot;</span><span class="p">]</span> <span class="ow">and</span> <span class="n">args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">==</span> <span class="o">-</span><span class="mi">1</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Using fsdp only works in distributed training.&quot;</span><span class="p">)</span>
+
+            <span class="c1"># dep_version_check(&quot;torch&gt;=1.12.0&quot;)</span>
+            <span class="c1"># Would have to update setup.py with torch&gt;=1.12.0</span>
+            <span class="c1"># which isn&#39;t ideally given that it will force people not using FSDP to also use torch&gt;=1.12.0</span>
+            <span class="c1"># below is the current alternative.</span>
+            <span class="k">if</span> <span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">__version__</span><span class="p">)</span><span class="o">.</span><span class="n">base_version</span><span class="p">)</span> <span class="o">&lt;</span> <span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="s2">&quot;1.12.0&quot;</span><span class="p">):</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;FSDP requires PyTorch &gt;= 1.12.0&quot;</span><span class="p">)</span>
+
+            <span class="kn">from</span><span class="w"> </span><span class="nn">torch.distributed.fsdp.fully_sharded_data_parallel</span><span class="w"> </span><span class="kn">import</span> <span class="n">BackwardPrefetch</span><span class="p">,</span> <span class="n">ShardingStrategy</span>
+
+            <span class="k">if</span> <span class="n">FSDPOption</span><span class="o">.</span><span class="n">FULL_SHARD</span> <span class="ow">in</span> <span class="n">args</span><span class="o">.</span><span class="n">fsdp</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">fsdp</span> <span class="o">=</span> <span class="n">ShardingStrategy</span><span class="o">.</span><span class="n">FULL_SHARD</span>
+            <span class="k">elif</span> <span class="n">FSDPOption</span><span class="o">.</span><span class="n">SHARD_GRAD_OP</span> <span class="ow">in</span> <span class="n">args</span><span class="o">.</span><span class="n">fsdp</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">fsdp</span> <span class="o">=</span> <span class="n">ShardingStrategy</span><span class="o">.</span><span class="n">SHARD_GRAD_OP</span>
+            <span class="k">elif</span> <span class="n">FSDPOption</span><span class="o">.</span><span class="n">NO_SHARD</span> <span class="ow">in</span> <span class="n">args</span><span class="o">.</span><span class="n">fsdp</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">fsdp</span> <span class="o">=</span> <span class="n">ShardingStrategy</span><span class="o">.</span><span class="n">NO_SHARD</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">backward_prefetch</span> <span class="o">=</span> <span class="n">BackwardPrefetch</span><span class="o">.</span><span class="n">BACKWARD_PRE</span>
+            <span class="k">if</span> <span class="s2">&quot;backward_prefetch&quot;</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span> <span class="ow">and</span> <span class="s2">&quot;backward_pos&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">backward_prefetch</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">backward_prefetch</span> <span class="o">=</span> <span class="n">BackwardPrefetch</span><span class="o">.</span><span class="n">BACKWARD_POST</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">forword_prefetch</span> <span class="o">=</span> <span class="kc">False</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;forword_prefect&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">forword_prefetch</span> <span class="o">=</span> <span class="kc">True</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">limit_all_gathers</span> <span class="o">=</span> <span class="kc">False</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;limit_all_gathers&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">limit_all_gathers</span> <span class="o">=</span> <span class="kc">True</span>
+
+        <span class="c1"># one place to sort out whether to place the model on device or not</span>
+        <span class="c1"># postpone switching model to cuda when:</span>
+        <span class="c1"># 1. MP - since we are trying to fit a much bigger than 1 gpu model</span>
+        <span class="c1"># 2. fp16-enabled DeepSpeed loads the model in half the size and it doesn&#39;t need .to() anyway,</span>
+        <span class="c1">#    and we only use deepspeed for training at the moment</span>
+        <span class="c1"># 3. full bf16 or fp16 eval - since the model needs to be cast to the right dtype first</span>
+        <span class="c1"># 4. Sharded DDP - same as MP</span>
+        <span class="c1"># 5. FSDP - same as MP</span>
+<div class="viewcode-block" id="RaftTrainer.place_model_on_device">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.place_model_on_device">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">place_model_on_device</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">place_model_on_device</span></div>
+
+        <span class="k">if</span> <span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">is_model_parallel</span>
+            <span class="ow">or</span> <span class="n">args</span><span class="o">.</span><span class="n">deepspeed</span>
+            <span class="ow">or</span> <span class="p">((</span><span class="n">args</span><span class="o">.</span><span class="n">fp16_full_eval</span> <span class="ow">or</span> <span class="n">args</span><span class="o">.</span><span class="n">bf16_full_eval</span><span class="p">)</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">args</span><span class="o">.</span><span class="n">do_train</span><span class="p">)</span>
+            <span class="ow">or</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="ow">in</span> <span class="p">[</span><span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">ZERO_DP_2</span><span class="p">,</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">ZERO_DP_3</span><span class="p">])</span>
+            <span class="ow">or</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">fsdp</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">)</span>
+        <span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">place_model_on_device</span> <span class="o">=</span> <span class="kc">False</span>
+
+        <span class="n">default_collator</span> <span class="o">=</span> <span class="n">default_data_collator</span> <span class="k">if</span> <span class="n">tokenizer</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">DataCollatorWithPadding</span><span class="p">(</span><span class="n">tokenizer</span><span class="p">)</span>
+<div class="viewcode-block" id="RaftTrainer.data_collator">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.data_collator">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_collator</span> <span class="o">=</span> <span class="n">data_collator</span> <span class="k">if</span> <span class="n">data_collator</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">default_collator</span></div>
+
+<div class="viewcode-block" id="RaftTrainer.train_dataset">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.train_dataset">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span> <span class="o">=</span> <span class="n">train_dataset</span></div>
+
+<div class="viewcode-block" id="RaftTrainer.eval_dataset">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.eval_dataset">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">eval_dataset</span> <span class="o">=</span> <span class="n">eval_dataset</span></div>
+
+<div class="viewcode-block" id="RaftTrainer.tokenizer">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.tokenizer">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span> <span class="o">=</span> <span class="n">tokenizer</span></div>
+
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">place_model_on_device</span> <span class="ow">and</span> <span class="ow">not</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="s2">&quot;is_loaded_in_8bit&quot;</span><span class="p">,</span> <span class="kc">False</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_move_model_to_device</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+
+        <span class="c1"># Force n_gpu to 1 to avoid DataParallel as MP will manage the GPUs</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_model_parallel</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">_n_gpu</span> <span class="o">=</span> <span class="mi">1</span>
+
+        <span class="c1"># later use `self.model is self.model_wrapped` to check if it&#39;s wrapped or not</span>
+<div class="viewcode-block" id="RaftTrainer.model_wrapped">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.model_wrapped">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span> <span class="o">=</span> <span class="n">model</span></div>
+
+<div class="viewcode-block" id="RaftTrainer.model">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.model">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">model</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.compute_metrics">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_metrics">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">compute_metrics</span> <span class="o">=</span> <span class="n">compute_metrics</span></div>
+
+<div class="viewcode-block" id="RaftTrainer.preprocess_logits_for_metrics">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.preprocess_logits_for_metrics">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">preprocess_logits_for_metrics</span> <span class="o">=</span> <span class="n">preprocess_logits_for_metrics</span></div>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span> <span class="o">=</span> <span class="n">optimizers</span>
+        <span class="k">if</span> <span class="n">model_init</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                <span class="s2">&quot;Passing a `model_init` is incompatible with providing the `optimizers` argument. &quot;</span>
+                <span class="s2">&quot;You should subclass `Trainer` and override the `create_optimizer_and_scheduler` method.&quot;</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">()</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">param</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">parameters</span><span class="p">():</span>
+                <span class="n">model_device</span> <span class="o">=</span> <span class="n">param</span><span class="o">.</span><span class="n">device</span>
+                <span class="k">break</span>
+            <span class="k">for</span> <span class="n">param_group</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">param_groups</span><span class="p">:</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">param_group</span><span class="p">[</span><span class="s2">&quot;params&quot;</span><span class="p">])</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">optimizer_device</span> <span class="o">=</span> <span class="n">param_group</span><span class="p">[</span><span class="s2">&quot;params&quot;</span><span class="p">][</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">device</span>
+                    <span class="k">break</span>
+            <span class="k">if</span> <span class="n">model_device</span> <span class="o">!=</span> <span class="n">optimizer_device</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="s2">&quot;The model and the optimizer parameters are not on the same device, which probably means you&quot;</span>
+                    <span class="s2">&quot; created an optimizer around your model **before** putting on the device and passing it to the&quot;</span>
+                    <span class="s2">&quot; `Trainer`. Make sure the lines `import torch_xla.core.xla_model as xm` and&quot;</span>
+                    <span class="s2">&quot; `model.to(xm.xla_device())` is performed before the optimizer creation in your script.&quot;</span>
+                <span class="p">)</span>
+        <span class="k">if</span> <span class="p">((</span><span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">)</span> <span class="ow">or</span> <span class="n">args</span><span class="o">.</span><span class="n">deepspeed</span> <span class="ow">or</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">fsdp</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">))</span> <span class="ow">and</span> <span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+        <span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                <span class="s2">&quot;Passing `optimizers` is not allowed if Fairscale, Deepspeed or PyTorch FSDP is enabled.&quot;</span>
+                <span class="s2">&quot;You should subclass `Trainer` and override the `create_optimizer_and_scheduler` method.&quot;</span>
+            <span class="p">)</span>
+        <span class="n">default_callbacks</span> <span class="o">=</span> <span class="n">DEFAULT_CALLBACKS</span> <span class="o">+</span> <span class="n">get_reporting_integration_callbacks</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">report_to</span><span class="p">)</span>
+        <span class="n">callbacks</span> <span class="o">=</span> <span class="n">default_callbacks</span> <span class="k">if</span> <span class="n">callbacks</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">default_callbacks</span> <span class="o">+</span> <span class="n">callbacks</span>
+<div class="viewcode-block" id="RaftTrainer.callback_handler">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.callback_handler">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span> <span class="o">=</span> <span class="n">CallbackHandler</span><span class="p">(</span>
+            <span class="n">callbacks</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span>
+        <span class="p">)</span></div>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">add_callback</span><span class="p">(</span><span class="n">PrinterCallback</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">disable_tqdm</span> <span class="k">else</span> <span class="n">DEFAULT_PROGRESS_CALLBACK</span><span class="p">)</span>
+
+        <span class="c1"># Will be set to True by `self._setup_loggers()` on first call to `self.log()`.</span>
+<div class="viewcode-block" id="RaftTrainer._loggers_initialized">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._loggers_initialized">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_loggers_initialized</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+
+        <span class="c1"># Create clone of distant repo and output directory if needed</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">push_to_hub</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">init_git_repo</span><span class="p">(</span><span class="n">at_init</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="c1"># In case of pull, we need to make sure every process has the latest.</span>
+            <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+                <span class="n">xm</span><span class="o">.</span><span class="n">rendezvous</span><span class="p">(</span><span class="s2">&quot;init git repo&quot;</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">!=</span> <span class="o">-</span><span class="mi">1</span><span class="p">:</span>
+                <span class="n">dist</span><span class="o">.</span><span class="n">barrier</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+            <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">callable</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data_collator</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">callable</span><span class="p">(</span><span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">data_collator</span><span class="p">,</span> <span class="s2">&quot;collate_batch&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;The `data_collator` should be a simple callable (function, class with `__call__`).&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="s2">&quot;max_steps is given, it will override any value given in num_train_epochs&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">train_dataset</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">has_length</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">)</span> <span class="ow">and</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">&lt;=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;train_dataset does not implement __len__, max_steps has to be specified&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="p">(</span>
+            <span class="n">train_dataset</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+            <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">IterableDataset</span><span class="p">)</span>
+            <span class="ow">and</span> <span class="n">args</span><span class="o">.</span><span class="n">group_by_length</span>
+        <span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;the `--group_by_length` option is only available for `Dataset`, not `IterableDataset&quot;</span><span class="p">)</span>
+
+<div class="viewcode-block" id="RaftTrainer._signature_columns">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._signature_columns">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_signature_columns</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+
+        <span class="c1"># Mixed precision setup</span>
+<div class="viewcode-block" id="RaftTrainer.use_apex">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_apex">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_apex</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+<div class="viewcode-block" id="RaftTrainer.use_cuda_amp">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_cuda_amp">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_cuda_amp</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+<div class="viewcode-block" id="RaftTrainer.use_cpu_amp">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_cpu_amp">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_cpu_amp</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+
+        <span class="c1"># Mixed precision setup for SageMaker Model Parallel</span>
+        <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+            <span class="c1"># BF16 + model parallelism in SageMaker: currently not supported, raise an error</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">bf16</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;SageMaker Model Parallelism does not support BF16 yet. Please use FP16 instead &quot;</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">IS_SAGEMAKER_MP_POST_1_10</span><span class="p">:</span>
+                <span class="c1"># When there&#39;s mismatch between SMP config and trainer argument, use SMP config as truth</span>
+                <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">fp16</span> <span class="o">!=</span> <span class="n">smp</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">cfg</span><span class="o">.</span><span class="n">fp16</span><span class="p">:</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;FP16 provided in SM_HP_MP_PARAMETERS is </span><span class="si">{</span><span class="n">smp</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">cfg</span><span class="o">.</span><span class="n">fp16</span><span class="si">}</span><span class="s2">,&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot;but FP16 provided in trainer argument is </span><span class="si">{</span><span class="n">args</span><span class="o">.</span><span class="n">fp16</span><span class="si">}</span><span class="s2">,&quot;</span>
+                        <span class="sa">f</span><span class="s2">&quot;setting to </span><span class="si">{</span><span class="n">smp</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">cfg</span><span class="o">.</span><span class="n">fp16</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="p">)</span>
+                    <span class="n">args</span><span class="o">.</span><span class="n">fp16</span> <span class="o">=</span> <span class="n">smp</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">cfg</span><span class="o">.</span><span class="n">fp16</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="c1"># smp &lt; 1.10 does not support fp16 in trainer.</span>
+                <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">smp</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">cfg</span><span class="p">,</span> <span class="s2">&quot;fp16&quot;</span><span class="p">):</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;FP16 provided in SM_HP_MP_PARAMETERS is </span><span class="si">{</span><span class="n">smp</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">cfg</span><span class="o">.</span><span class="n">fp16</span><span class="si">}</span><span class="s2">, &quot;</span>
+                        <span class="s2">&quot;but SageMaker Model Parallelism &lt; 1.10 does not support FP16 in trainer.&quot;</span>
+                    <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">fp16</span> <span class="ow">or</span> <span class="n">args</span><span class="o">.</span><span class="n">bf16</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">half_precision_backend</span> <span class="o">==</span> <span class="s2">&quot;auto&quot;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">device</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">device</span><span class="p">(</span><span class="s2">&quot;cpu&quot;</span><span class="p">):</span>
+                    <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">fp16</span><span class="p">:</span>
+                        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Tried to use `fp16` but it is not supported on cpu&quot;</span><span class="p">)</span>
+                    <span class="k">elif</span> <span class="n">_is_native_cpu_amp_available</span><span class="p">:</span>
+                        <span class="n">args</span><span class="o">.</span><span class="n">half_precision_backend</span> <span class="o">=</span> <span class="s2">&quot;cpu_amp&quot;</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Tried to use cpu amp but native cpu amp is not available&quot;</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">args</span><span class="o">.</span><span class="n">half_precision_backend</span> <span class="o">=</span> <span class="s2">&quot;cuda_amp&quot;</span>
+
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Using </span><span class="si">{</span><span class="n">args</span><span class="o">.</span><span class="n">half_precision_backend</span><span class="si">}</span><span class="s2"> half precision backend&quot;</span><span class="p">)</span>
+
+<div class="viewcode-block" id="RaftTrainer.do_grad_scaling">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.do_grad_scaling">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">do_grad_scaling</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+        <span class="k">if</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">fp16</span> <span class="ow">or</span> <span class="n">args</span><span class="o">.</span><span class="n">bf16</span><span class="p">)</span> <span class="ow">and</span> <span class="ow">not</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">deepspeed</span> <span class="ow">or</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">()</span> <span class="ow">or</span> <span class="n">is_torch_tpu_available</span><span class="p">()):</span>
+            <span class="c1"># deepspeed and SageMaker Model Parallel manage their own half precision</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">half_precision_backend</span> <span class="o">==</span> <span class="s2">&quot;cuda_amp&quot;</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">use_cuda_amp</span> <span class="o">=</span> <span class="kc">True</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">amp_dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">float16</span> <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">fp16</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">bfloat16</span>
+                <span class="c1">#  bf16 does not need grad scaling</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">do_grad_scaling</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">amp_dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">float16</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">do_grad_scaling</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span> <span class="o">=</span> <span class="n">ShardedGradScaler</span><span class="p">()</span>
+                    <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">fsdp</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="kn">from</span><span class="w"> </span><span class="nn">torch.distributed.fsdp.sharded_grad_scaler</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+                            <span class="n">ShardedGradScaler</span> <span class="k">as</span> <span class="n">FSDPShardedGradScaler</span><span class="p">,</span>
+                        <span class="p">)</span>
+
+                        <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span> <span class="o">=</span> <span class="n">FSDPShardedGradScaler</span><span class="p">()</span>
+                    <span class="k">elif</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+                        <span class="kn">from</span><span class="w"> </span><span class="nn">torch_xla.amp</span><span class="w"> </span><span class="kn">import</span> <span class="n">GradScaler</span>
+
+                        <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span> <span class="o">=</span> <span class="n">GradScaler</span><span class="p">()</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">amp</span><span class="o">.</span><span class="n">GradScaler</span><span class="p">()</span>
+            <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">half_precision_backend</span> <span class="o">==</span> <span class="s2">&quot;cpu_amp&quot;</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">use_cpu_amp</span> <span class="o">=</span> <span class="kc">True</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">amp_dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">bfloat16</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="n">is_apex_available</span><span class="p">():</span>
+                    <span class="k">raise</span> <span class="ne">ImportError</span><span class="p">(</span>
+                        <span class="s2">&quot;Using FP16 with APEX but APEX is not installed, please refer to&quot;</span>
+                        <span class="s2">&quot; https://www.github.com/nvidia/apex.&quot;</span>
+                    <span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">use_apex</span> <span class="o">=</span> <span class="kc">True</span>
+
+        <span class="c1"># FP16 + model parallelism in SageMaker: gradient clipping does not work for now so we raise a helpful error.</span>
+        <span class="k">if</span> <span class="p">(</span>
+            <span class="n">is_sagemaker_mp_enabled</span><span class="p">()</span>
+            <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_cuda_amp</span>
+            <span class="ow">and</span> <span class="n">args</span><span class="o">.</span><span class="n">max_grad_norm</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+            <span class="ow">and</span> <span class="n">args</span><span class="o">.</span><span class="n">max_grad_norm</span> <span class="o">&gt;</span> <span class="mi">0</span>
+        <span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;SageMaker Model Parallelism in mixed precision mode does not support gradient clipping yet. Pass &quot;</span>
+                <span class="s2">&quot;along &#39;max_grad_norm&#39;: 0 in your hyperparameters.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="c1"># Label smoothing</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">label_smoothing_factor</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">label_smoother</span> <span class="o">=</span> <span class="n">LabelSmoother</span><span class="p">(</span><span class="n">epsilon</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">label_smoothing_factor</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">label_smoother</span> <span class="o">=</span> <span class="kc">None</span>
+
+<div class="viewcode-block" id="RaftTrainer.state">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.state">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">state</span> <span class="o">=</span> <span class="n">TrainerState</span><span class="p">(</span>
+            <span class="n">is_local_process_zero</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">is_local_process_zero</span><span class="p">(),</span>
+            <span class="n">is_world_process_zero</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">is_world_process_zero</span><span class="p">(),</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.control">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.control">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="n">TrainerControl</span><span class="p">()</span></div>
+
+        <span class="c1"># Internal variable to count flos in each process, will be accumulated in `self.state.total_flos` then</span>
+        <span class="c1"># returned to 0 every time flos need to be logged</span>
+<div class="viewcode-block" id="RaftTrainer.current_flos">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.current_flos">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">current_flos</span> <span class="o">=</span> <span class="mi">0</span></div>
+
+<div class="viewcode-block" id="RaftTrainer.hp_search_backend">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.hp_search_backend">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="RaftTrainer.use_tune_checkpoints">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_tune_checkpoints">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_tune_checkpoints</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+        <span class="n">default_label_names</span> <span class="o">=</span> <span class="n">find_labels</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="vm">__class__</span><span class="p">)</span>
+<div class="viewcode-block" id="RaftTrainer.label_names">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.label_names">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">label_names</span> <span class="o">=</span> <span class="n">default_label_names</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">label_names</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">label_names</span></div>
+
+<div class="viewcode-block" id="RaftTrainer.can_return_loss">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.can_return_loss">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">can_return_loss</span> <span class="o">=</span> <span class="n">can_return_loss</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="vm">__class__</span><span class="p">)</span></div>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_init_end</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">)</span>
+
+        <span class="c1"># Internal variables to keep track of the original batch size</span>
+<div class="viewcode-block" id="RaftTrainer._train_batch_size">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._train_batch_size">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_train_batch_size</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">train_batch_size</span></div>
+
+
+        <span class="c1"># very last</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_memory_tracker</span><span class="o">.</span><span class="n">stop_and_update_metrics</span><span class="p">()</span>
+
+        <span class="c1"># torch.compile</span>
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">torch_compile</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">is_torch_compile_available</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;Using torch.compile requires PyTorch 2.0 or higher.&quot;</span><span class="p">)</span>
+
+<div class="viewcode-block" id="RaftTrainer.add_callback">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.add_callback">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">add_callback</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">callback</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Add a callback to the current list of [`~transformer.TrainerCallback`].</span>
+<span class="sd">        Args:</span>
+<span class="sd">           callback (`type` or [`~transformer.TrainerCallback`]):</span>
+<span class="sd">               A [`~transformer.TrainerCallback`] class or an instance of a [`~transformer.TrainerCallback`]. In the</span>
+<span class="sd">               first case, will instantiate a member of that class.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">add_callback</span><span class="p">(</span><span class="n">callback</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.pop_callback">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.pop_callback">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">pop_callback</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">callback</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Remove a callback from the current list of [`~transformer.TrainerCallback`] and returns it.</span>
+<span class="sd">        If the callback is not found, returns `None` (and no error is raised).</span>
+<span class="sd">        Args:</span>
+<span class="sd">           callback (`type` or [`~transformer.TrainerCallback`]):</span>
+<span class="sd">               A [`~transformer.TrainerCallback`] class or an instance of a [`~transformer.TrainerCallback`]. In the</span>
+<span class="sd">               first case, will pop the first member of that class found in the list of callbacks.</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            [`~transformer.TrainerCallback`]: The callback removed, if found.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">pop_callback</span><span class="p">(</span><span class="n">callback</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.remove_callback">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.remove_callback">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">remove_callback</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">callback</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Remove a callback from the current list of [`~transformer.TrainerCallback`].</span>
+<span class="sd">        Args:</span>
+<span class="sd">           callback (`type` or [`~transformer.TrainerCallback`]):</span>
+<span class="sd">               A [`~transformer.TrainerCallback`] class or an instance of a [`~transformer.TrainerCallback`]. In the</span>
+<span class="sd">               first case, will remove the first member of that class found in the list of callbacks.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">remove_callback</span><span class="p">(</span><span class="n">callback</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._move_model_to_device">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._move_model_to_device">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_move_model_to_device</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">device</span><span class="p">):</span>
+        <span class="n">model</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="p">)</span>
+        <span class="c1"># Moving a model to an XLA device disconnects the tied weights, so we have to retie them.</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">parallel_mode</span> <span class="o">==</span> <span class="n">ParallelMode</span><span class="o">.</span><span class="n">TPU</span> <span class="ow">and</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="s2">&quot;tie_weights&quot;</span><span class="p">):</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">tie_weights</span><span class="p">()</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._set_signature_columns_if_needed">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._set_signature_columns_if_needed">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_set_signature_columns_if_needed</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_signature_columns</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="c1"># Inspect model forward signature to keep only the arguments it accepts.</span>
+            <span class="n">signature</span> <span class="o">=</span> <span class="n">inspect</span><span class="o">.</span><span class="n">signature</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">forward</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_signature_columns</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">signature</span><span class="o">.</span><span class="n">parameters</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span>
+            <span class="c1"># Labels may be named label or label_ids, the default data collator handles that.</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_signature_columns</span> <span class="o">+=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">set</span><span class="p">([</span><span class="s2">&quot;label&quot;</span><span class="p">,</span> <span class="s2">&quot;label_ids&quot;</span><span class="p">]</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">label_names</span><span class="p">))</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._remove_unused_columns">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._remove_unused_columns">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_remove_unused_columns</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dataset</span><span class="p">:</span> <span class="s2">&quot;datasets.Dataset&quot;</span><span class="p">,</span> <span class="n">description</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">remove_unused_columns</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">dataset</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_set_signature_columns_if_needed</span><span class="p">()</span>
+        <span class="n">signature_columns</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_signature_columns</span>
+
+        <span class="n">ignored_columns</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">set</span><span class="p">(</span><span class="n">dataset</span><span class="o">.</span><span class="n">column_names</span><span class="p">)</span> <span class="o">-</span> <span class="nb">set</span><span class="p">(</span><span class="n">signature_columns</span><span class="p">))</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">ignored_columns</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">dset_description</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span> <span class="k">if</span> <span class="n">description</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="sa">f</span><span class="s2">&quot;in the </span><span class="si">{</span><span class="n">description</span><span class="si">}</span><span class="s2"> set&quot;</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;The following columns </span><span class="si">{</span><span class="n">dset_description</span><span class="si">}</span><span class="s2"> don&#39;t have a corresponding argument in &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;`</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">.forward` and have been ignored: </span><span class="si">{</span><span class="s1">&#39;, &#39;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">ignored_columns</span><span class="p">)</span><span class="si">}</span><span class="s2">.&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot; If </span><span class="si">{</span><span class="s1">&#39;, &#39;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">ignored_columns</span><span class="p">)</span><span class="si">}</span><span class="s2"> are not expected by `</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">.forward`, &quot;</span>
+                <span class="s2">&quot; you can safely ignore this message.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">columns</span> <span class="o">=</span> <span class="p">[</span><span class="n">k</span> <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">signature_columns</span> <span class="k">if</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">dataset</span><span class="o">.</span><span class="n">column_names</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">datasets</span><span class="o">.</span><span class="n">__version__</span><span class="p">)</span> <span class="o">&lt;</span> <span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="s2">&quot;1.4.0&quot;</span><span class="p">):</span>
+            <span class="n">dataset</span><span class="o">.</span><span class="n">set_format</span><span class="p">(</span>
+                <span class="nb">type</span><span class="o">=</span><span class="n">dataset</span><span class="o">.</span><span class="n">format</span><span class="p">[</span><span class="s2">&quot;type&quot;</span><span class="p">],</span> <span class="n">columns</span><span class="o">=</span><span class="n">columns</span><span class="p">,</span> <span class="n">format_kwargs</span><span class="o">=</span><span class="n">dataset</span><span class="o">.</span><span class="n">format</span><span class="p">[</span><span class="s2">&quot;format_kwargs&quot;</span><span class="p">]</span>
+            <span class="p">)</span>
+            <span class="k">return</span> <span class="n">dataset</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">dataset</span><span class="o">.</span><span class="n">remove_columns</span><span class="p">(</span><span class="n">ignored_columns</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._get_collator_with_removed_columns">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_collator_with_removed_columns">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_get_collator_with_removed_columns</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">data_collator</span><span class="p">:</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">description</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Callable</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Wrap the data collator in a callable removing unused columns.&quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">remove_unused_columns</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">data_collator</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_set_signature_columns_if_needed</span><span class="p">()</span>
+        <span class="n">signature_columns</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_signature_columns</span>
+
+        <span class="n">remove_columns_collator</span> <span class="o">=</span> <span class="n">RemoveColumnsCollator</span><span class="p">(</span>
+            <span class="n">data_collator</span><span class="o">=</span><span class="n">data_collator</span><span class="p">,</span>
+            <span class="n">signature_columns</span><span class="o">=</span><span class="n">signature_columns</span><span class="p">,</span>
+            <span class="n">logger</span><span class="o">=</span><span class="n">logger</span><span class="p">,</span>
+            <span class="n">description</span><span class="o">=</span><span class="n">description</span><span class="p">,</span>
+            <span class="n">model_name</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="k">return</span> <span class="n">remove_columns_collator</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._get_train_sampler">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_train_sampler">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_get_train_sampler</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">Sampler</span><span class="p">]:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">has_length</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span><span class="p">):</span>
+            <span class="k">return</span> <span class="kc">None</span>
+
+        <span class="n">generator</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span> <span class="o">&lt;=</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">generator</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">Generator</span><span class="p">()</span>
+            <span class="c1"># for backwards compatibility, we generate a seed here (which is sampled from a generator seeded with</span>
+            <span class="c1"># `args.seed`) if data_seed isn&#39;t provided.</span>
+            <span class="c1"># Further on in this method, we default to `args.seed` instead.</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">data_seed</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">seed</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">((),</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">int64</span><span class="p">)</span><span class="o">.</span><span class="n">random_</span><span class="p">()</span><span class="o">.</span><span class="n">item</span><span class="p">())</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">seed</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">data_seed</span>
+            <span class="n">generator</span><span class="o">.</span><span class="n">manual_seed</span><span class="p">(</span><span class="n">seed</span><span class="p">)</span>
+
+        <span class="n">seed</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">data_seed</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">data_seed</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">seed</span>
+
+        <span class="c1"># Build the sampler.</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">group_by_length</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">is_datasets_available</span><span class="p">()</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span><span class="p">,</span> <span class="n">datasets</span><span class="o">.</span><span class="n">Dataset</span><span class="p">):</span>
+                <span class="n">lengths</span> <span class="o">=</span> <span class="p">(</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">length_column_name</span><span class="p">]</span>
+                    <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">length_column_name</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span><span class="o">.</span><span class="n">column_names</span>
+                    <span class="k">else</span> <span class="kc">None</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">lengths</span> <span class="o">=</span> <span class="kc">None</span>
+            <span class="n">model_input_name</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_input_names</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="kc">None</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span> <span class="o">&lt;=</span> <span class="mi">1</span><span class="p">:</span>
+                <span class="k">return</span> <span class="n">LengthGroupedSampler</span><span class="p">(</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">train_batch_size</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span><span class="p">,</span>
+                    <span class="n">dataset</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span><span class="p">,</span>
+                    <span class="n">lengths</span><span class="o">=</span><span class="n">lengths</span><span class="p">,</span>
+                    <span class="n">model_input_name</span><span class="o">=</span><span class="n">model_input_name</span><span class="p">,</span>
+                    <span class="n">generator</span><span class="o">=</span><span class="n">generator</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">return</span> <span class="n">DistributedLengthGroupedSampler</span><span class="p">(</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">train_batch_size</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span><span class="p">,</span>
+                    <span class="n">dataset</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span><span class="p">,</span>
+                    <span class="n">num_replicas</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span><span class="p">,</span>
+                    <span class="n">rank</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">process_index</span><span class="p">,</span>
+                    <span class="n">lengths</span><span class="o">=</span><span class="n">lengths</span><span class="p">,</span>
+                    <span class="n">model_input_name</span><span class="o">=</span><span class="n">model_input_name</span><span class="p">,</span>
+                    <span class="n">seed</span><span class="o">=</span><span class="n">seed</span><span class="p">,</span>
+                <span class="p">)</span>
+
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span> <span class="o">&lt;=</span> <span class="mi">1</span><span class="p">:</span>
+                <span class="k">return</span> <span class="n">RandomSampler</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span><span class="p">,</span> <span class="n">generator</span><span class="o">=</span><span class="n">generator</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">parallel_mode</span> <span class="ow">in</span> <span class="p">[</span><span class="n">ParallelMode</span><span class="o">.</span><span class="n">TPU</span><span class="p">,</span> <span class="n">ParallelMode</span><span class="o">.</span><span class="n">SAGEMAKER_MODEL_PARALLEL</span><span class="p">]</span>
+                <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_drop_last</span>
+            <span class="p">):</span>
+                <span class="c1"># Use a loop for TPUs when drop_last is False to have all batches have the same size.</span>
+                <span class="k">return</span> <span class="n">DistributedSamplerWithLoop</span><span class="p">(</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span><span class="p">,</span>
+                    <span class="n">batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">per_device_train_batch_size</span><span class="p">,</span>
+                    <span class="n">num_replicas</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span><span class="p">,</span>
+                    <span class="n">rank</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">process_index</span><span class="p">,</span>
+                    <span class="n">seed</span><span class="o">=</span><span class="n">seed</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">return</span> <span class="n">DistributedSampler</span><span class="p">(</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span><span class="p">,</span>
+                    <span class="n">num_replicas</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span><span class="p">,</span>
+                    <span class="n">rank</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">process_index</span><span class="p">,</span>
+                    <span class="n">seed</span><span class="o">=</span><span class="n">seed</span><span class="p">,</span>
+                <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.get_train_dataloader">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_train_dataloader">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_train_dataloader</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DataLoader</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Returns the training [`~torch.utils.data.DataLoader`].</span>
+<span class="sd">        Will use no sampler if `train_dataset` does not implement `__len__`, a random sampler (adapted to distributed</span>
+<span class="sd">        training if necessary) otherwise.</span>
+<span class="sd">        Subclass and override this method if you want to inject some custom behavior.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Trainer: training requires a train_dataset.&quot;</span><span class="p">)</span>
+
+        <span class="n">train_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">train_dataset</span>
+        <span class="n">data_collator</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_collator</span>
+        <span class="k">if</span> <span class="n">is_datasets_available</span><span class="p">()</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">,</span> <span class="n">datasets</span><span class="o">.</span><span class="n">Dataset</span><span class="p">):</span>
+            <span class="n">train_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_remove_unused_columns</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">,</span> <span class="n">description</span><span class="o">=</span><span class="s2">&quot;training&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">data_collator</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_collator_with_removed_columns</span><span class="p">(</span><span class="n">data_collator</span><span class="p">,</span> <span class="n">description</span><span class="o">=</span><span class="s2">&quot;training&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">train_dataset</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">IterableDataset</span><span class="p">):</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+                <span class="n">train_dataset</span> <span class="o">=</span> <span class="n">IterableDatasetShard</span><span class="p">(</span>
+                    <span class="n">train_dataset</span><span class="p">,</span>
+                    <span class="n">batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_train_batch_size</span><span class="p">,</span>
+                    <span class="n">drop_last</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_drop_last</span><span class="p">,</span>
+                    <span class="n">num_processes</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span><span class="p">,</span>
+                    <span class="n">process_index</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">process_index</span><span class="p">,</span>
+                <span class="p">)</span>
+
+            <span class="k">return</span> <span class="n">DataLoader</span><span class="p">(</span>
+                <span class="n">train_dataset</span><span class="p">,</span>
+                <span class="n">batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_train_batch_size</span><span class="p">,</span>
+                <span class="n">collate_fn</span><span class="o">=</span><span class="n">data_collator</span><span class="p">,</span>
+                <span class="n">num_workers</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_num_workers</span><span class="p">,</span>
+                <span class="n">pin_memory</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_pin_memory</span><span class="p">,</span>
+            <span class="p">)</span>
+
+        <span class="n">train_sampler</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_train_sampler</span><span class="p">()</span>
+
+        <span class="k">return</span> <span class="n">DataLoader</span><span class="p">(</span>
+            <span class="n">train_dataset</span><span class="p">,</span>
+            <span class="n">batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">_train_batch_size</span><span class="p">,</span>
+            <span class="n">sampler</span><span class="o">=</span><span class="n">train_sampler</span><span class="p">,</span>
+            <span class="n">collate_fn</span><span class="o">=</span><span class="n">data_collator</span><span class="p">,</span>
+            <span class="n">drop_last</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_drop_last</span><span class="p">,</span>
+            <span class="n">num_workers</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_num_workers</span><span class="p">,</span>
+            <span class="n">pin_memory</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_pin_memory</span><span class="p">,</span>
+            <span class="n">worker_init_fn</span><span class="o">=</span><span class="n">seed_worker</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._get_eval_sampler">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_eval_sampler">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_get_eval_sampler</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">eval_dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">Sampler</span><span class="p">]:</span>
+        <span class="c1"># Deprecated code</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">use_legacy_prediction_loop</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+                <span class="k">return</span> <span class="n">SequentialDistributedSampler</span><span class="p">(</span>
+                    <span class="n">eval_dataset</span><span class="p">,</span> <span class="n">num_replicas</span><span class="o">=</span><span class="n">xm</span><span class="o">.</span><span class="n">xrt_world_size</span><span class="p">(),</span> <span class="n">rank</span><span class="o">=</span><span class="n">xm</span><span class="o">.</span><span class="n">get_ordinal</span><span class="p">()</span>
+                <span class="p">)</span>
+            <span class="k">elif</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+                <span class="k">return</span> <span class="n">SequentialDistributedSampler</span><span class="p">(</span>
+                    <span class="n">eval_dataset</span><span class="p">,</span>
+                    <span class="n">num_replicas</span><span class="o">=</span><span class="n">smp</span><span class="o">.</span><span class="n">dp_size</span><span class="p">(),</span>
+                    <span class="n">rank</span><span class="o">=</span><span class="n">smp</span><span class="o">.</span><span class="n">dp_rank</span><span class="p">(),</span>
+                    <span class="n">batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">per_device_eval_batch_size</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">!=</span> <span class="o">-</span><span class="mi">1</span><span class="p">:</span>
+                <span class="k">return</span> <span class="n">SequentialDistributedSampler</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">return</span> <span class="n">SequentialSampler</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span> <span class="o">&lt;=</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">SequentialSampler</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">ShardSampler</span><span class="p">(</span>
+                <span class="n">eval_dataset</span><span class="p">,</span>
+                <span class="n">batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">per_device_eval_batch_size</span><span class="p">,</span>
+                <span class="n">num_processes</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span><span class="p">,</span>
+                <span class="n">process_index</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">process_index</span><span class="p">,</span>
+            <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.get_eval_dataloader">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_eval_dataloader">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_eval_dataloader</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">eval_dataset</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dataset</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DataLoader</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Returns the evaluation [`~torch.utils.data.DataLoader`].</span>
+<span class="sd">        Subclass and override this method if you want to inject some custom behavior.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            eval_dataset (`torch.utils.data.Dataset`, *optional*):</span>
+<span class="sd">                If provided, will override `self.eval_dataset`. If it is a [`~datasets.Dataset`], columns not accepted</span>
+<span class="sd">                by the `model.forward()` method are automatically removed. It must implement `__len__`.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">eval_dataset</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">eval_dataset</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Trainer: evaluation requires an eval_dataset.&quot;</span><span class="p">)</span>
+        <span class="n">eval_dataset</span> <span class="o">=</span> <span class="n">eval_dataset</span> <span class="k">if</span> <span class="n">eval_dataset</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">eval_dataset</span>
+        <span class="n">data_collator</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_collator</span>
+
+        <span class="k">if</span> <span class="n">is_datasets_available</span><span class="p">()</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">,</span> <span class="n">datasets</span><span class="o">.</span><span class="n">Dataset</span><span class="p">):</span>
+            <span class="n">eval_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_remove_unused_columns</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">,</span> <span class="n">description</span><span class="o">=</span><span class="s2">&quot;evaluation&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">data_collator</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_collator_with_removed_columns</span><span class="p">(</span><span class="n">data_collator</span><span class="p">,</span> <span class="n">description</span><span class="o">=</span><span class="s2">&quot;evaluation&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">IterableDataset</span><span class="p">):</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+                <span class="n">eval_dataset</span> <span class="o">=</span> <span class="n">IterableDatasetShard</span><span class="p">(</span>
+                    <span class="n">eval_dataset</span><span class="p">,</span>
+                    <span class="n">batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">per_device_eval_batch_size</span><span class="p">,</span>
+                    <span class="n">drop_last</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_drop_last</span><span class="p">,</span>
+                    <span class="n">num_processes</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span><span class="p">,</span>
+                    <span class="n">process_index</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">process_index</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">return</span> <span class="n">DataLoader</span><span class="p">(</span>
+                <span class="n">eval_dataset</span><span class="p">,</span>
+                <span class="n">batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">eval_batch_size</span><span class="p">,</span>
+                <span class="n">collate_fn</span><span class="o">=</span><span class="n">data_collator</span><span class="p">,</span>
+                <span class="n">num_workers</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_num_workers</span><span class="p">,</span>
+                <span class="n">pin_memory</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_pin_memory</span><span class="p">,</span>
+            <span class="p">)</span>
+
+        <span class="n">eval_sampler</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_eval_sampler</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">DataLoader</span><span class="p">(</span>
+            <span class="n">eval_dataset</span><span class="p">,</span>
+            <span class="n">sampler</span><span class="o">=</span><span class="n">eval_sampler</span><span class="p">,</span>
+            <span class="n">batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">eval_batch_size</span><span class="p">,</span>
+            <span class="n">collate_fn</span><span class="o">=</span><span class="n">data_collator</span><span class="p">,</span>
+            <span class="n">drop_last</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_drop_last</span><span class="p">,</span>
+            <span class="n">num_workers</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_num_workers</span><span class="p">,</span>
+            <span class="n">pin_memory</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_pin_memory</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.get_test_dataloader">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_test_dataloader">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_test_dataloader</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">test_dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">DataLoader</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Returns the test [`~torch.utils.data.DataLoader`].</span>
+<span class="sd">        Subclass and override this method if you want to inject some custom behavior.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            test_dataset (`torch.utils.data.Dataset`, *optional*):</span>
+<span class="sd">                The test dataset to use. If it is a [`~datasets.Dataset`], columns not accepted by the</span>
+<span class="sd">                `model.forward()` method are automatically removed. It must implement `__len__`.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">data_collator</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">data_collator</span>
+
+        <span class="k">if</span> <span class="n">is_datasets_available</span><span class="p">()</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">test_dataset</span><span class="p">,</span> <span class="n">datasets</span><span class="o">.</span><span class="n">Dataset</span><span class="p">):</span>
+            <span class="n">test_dataset</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_remove_unused_columns</span><span class="p">(</span><span class="n">test_dataset</span><span class="p">,</span> <span class="n">description</span><span class="o">=</span><span class="s2">&quot;test&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">data_collator</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_collator_with_removed_columns</span><span class="p">(</span><span class="n">data_collator</span><span class="p">,</span> <span class="n">description</span><span class="o">=</span><span class="s2">&quot;test&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">test_dataset</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">IterableDataset</span><span class="p">):</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+                <span class="n">test_dataset</span> <span class="o">=</span> <span class="n">IterableDatasetShard</span><span class="p">(</span>
+                    <span class="n">test_dataset</span><span class="p">,</span>
+                    <span class="n">batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">eval_batch_size</span><span class="p">,</span>
+                    <span class="n">drop_last</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_drop_last</span><span class="p">,</span>
+                    <span class="n">num_processes</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span><span class="p">,</span>
+                    <span class="n">process_index</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">process_index</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">return</span> <span class="n">DataLoader</span><span class="p">(</span>
+                <span class="n">test_dataset</span><span class="p">,</span>
+                <span class="n">batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">eval_batch_size</span><span class="p">,</span>
+                <span class="n">collate_fn</span><span class="o">=</span><span class="n">data_collator</span><span class="p">,</span>
+                <span class="n">num_workers</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_num_workers</span><span class="p">,</span>
+                <span class="n">pin_memory</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_pin_memory</span><span class="p">,</span>
+            <span class="p">)</span>
+
+        <span class="n">test_sampler</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_eval_sampler</span><span class="p">(</span><span class="n">test_dataset</span><span class="p">)</span>
+
+        <span class="c1"># We use the same batch_size as for eval.</span>
+        <span class="k">return</span> <span class="n">DataLoader</span><span class="p">(</span>
+            <span class="n">test_dataset</span><span class="p">,</span>
+            <span class="n">sampler</span><span class="o">=</span><span class="n">test_sampler</span><span class="p">,</span>
+            <span class="n">batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">eval_batch_size</span><span class="p">,</span>
+            <span class="n">collate_fn</span><span class="o">=</span><span class="n">data_collator</span><span class="p">,</span>
+            <span class="n">drop_last</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_drop_last</span><span class="p">,</span>
+            <span class="n">num_workers</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_num_workers</span><span class="p">,</span>
+            <span class="n">pin_memory</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">dataloader_pin_memory</span><span class="p">,</span>
+        <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.create_optimizer_and_scheduler">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_optimizer_and_scheduler">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">create_optimizer_and_scheduler</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_training_steps</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Setup the optimizer and the learning rate scheduler.</span>
+<span class="sd">        We provide a reasonable default that works well. If you want to use something else, you can pass a tuple in the</span>
+<span class="sd">        Trainer&#39;s init through `optimizers`, or subclass and override this method (or `create_optimizer` and/or</span>
+<span class="sd">        `create_scheduler`) in a subclass.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">create_optimizer</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">IS_SAGEMAKER_MP_POST_1_10</span> <span class="ow">and</span> <span class="n">smp</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">cfg</span><span class="o">.</span><span class="n">fp16</span><span class="p">:</span>
+            <span class="c1"># If smp &gt;= 1.10 and fp16 is enabled, we unwrap the optimizer</span>
+            <span class="n">optimizer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">optimizer</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">optimizer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">create_scheduler</span><span class="p">(</span><span class="n">num_training_steps</span><span class="o">=</span><span class="n">num_training_steps</span><span class="p">,</span> <span class="n">optimizer</span><span class="o">=</span><span class="n">optimizer</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.create_optimizer">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_optimizer">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">create_optimizer</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Setup the optimizer.</span>
+<span class="sd">        We provide a reasonable default that works well. If you want to use something else, you can pass a tuple in the</span>
+<span class="sd">        Trainer&#39;s init through `optimizers`, or subclass and override this method in a subclass.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">opt_model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span> <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">()</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">decay_parameters</span> <span class="o">=</span> <span class="n">get_parameter_names</span><span class="p">(</span><span class="n">opt_model</span><span class="p">,</span> <span class="n">ALL_LAYERNORM_LAYERS</span><span class="p">)</span>
+            <span class="n">decay_parameters</span> <span class="o">=</span> <span class="p">[</span><span class="n">name</span> <span class="k">for</span> <span class="n">name</span> <span class="ow">in</span> <span class="n">decay_parameters</span> <span class="k">if</span> <span class="s2">&quot;bias&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">name</span><span class="p">]</span>
+            <span class="n">optimizer_grouped_parameters</span> <span class="o">=</span> <span class="p">[</span>
+                <span class="p">{</span>
+                    <span class="s2">&quot;params&quot;</span><span class="p">:</span> <span class="p">[</span>
+                        <span class="n">p</span> <span class="k">for</span> <span class="n">n</span><span class="p">,</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">opt_model</span><span class="o">.</span><span class="n">named_parameters</span><span class="p">()</span> <span class="k">if</span> <span class="p">(</span><span class="n">n</span> <span class="ow">in</span> <span class="n">decay_parameters</span> <span class="ow">and</span> <span class="n">p</span><span class="o">.</span><span class="n">requires_grad</span><span class="p">)</span>
+                    <span class="p">],</span>
+                    <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">weight_decay</span><span class="p">,</span>
+                <span class="p">},</span>
+                <span class="p">{</span>
+                    <span class="s2">&quot;params&quot;</span><span class="p">:</span> <span class="p">[</span>
+                        <span class="n">p</span> <span class="k">for</span> <span class="n">n</span><span class="p">,</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">opt_model</span><span class="o">.</span><span class="n">named_parameters</span><span class="p">()</span> <span class="k">if</span> <span class="p">(</span><span class="n">n</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">decay_parameters</span> <span class="ow">and</span> <span class="n">p</span><span class="o">.</span><span class="n">requires_grad</span><span class="p">)</span>
+                    <span class="p">],</span>
+                    <span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="mf">0.0</span><span class="p">,</span>
+                <span class="p">},</span>
+            <span class="p">]</span>
+
+            <span class="n">optimizer_cls</span><span class="p">,</span> <span class="n">optimizer_kwargs</span> <span class="o">=</span> <span class="n">Trainer</span><span class="o">.</span><span class="n">get_optimizer_cls_and_kwargs</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="o">==</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">SIMPLE</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">OSS</span><span class="p">(</span>
+                    <span class="n">params</span><span class="o">=</span><span class="n">optimizer_grouped_parameters</span><span class="p">,</span>
+                    <span class="n">optim</span><span class="o">=</span><span class="n">optimizer_cls</span><span class="p">,</span>
+                    <span class="o">**</span><span class="n">optimizer_kwargs</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">optimizer_cls</span><span class="p">(</span><span class="n">optimizer_grouped_parameters</span><span class="p">,</span> <span class="o">**</span><span class="n">optimizer_kwargs</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">optimizer_cls</span><span class="o">.</span><span class="vm">__name__</span> <span class="o">==</span> <span class="s2">&quot;Adam8bit&quot;</span><span class="p">:</span>
+                    <span class="kn">import</span><span class="w"> </span><span class="nn">bitsandbytes</span>
+
+                    <span class="n">manager</span> <span class="o">=</span> <span class="n">bitsandbytes</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">GlobalOptimManager</span><span class="o">.</span><span class="n">get_instance</span><span class="p">()</span>
+
+                    <span class="n">skipped</span> <span class="o">=</span> <span class="mi">0</span>
+                    <span class="k">for</span> <span class="n">module</span> <span class="ow">in</span> <span class="n">opt_model</span><span class="o">.</span><span class="n">modules</span><span class="p">():</span>
+                        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">module</span><span class="p">,</span> <span class="n">nn</span><span class="o">.</span><span class="n">Embedding</span><span class="p">):</span>
+                            <span class="n">skipped</span> <span class="o">+=</span> <span class="nb">sum</span><span class="p">({</span><span class="n">p</span><span class="o">.</span><span class="n">data_ptr</span><span class="p">():</span> <span class="n">p</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span> <span class="k">for</span> <span class="n">p</span> <span class="ow">in</span> <span class="n">module</span><span class="o">.</span><span class="n">parameters</span><span class="p">()}</span><span class="o">.</span><span class="n">values</span><span class="p">())</span>
+                            <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;skipped </span><span class="si">{</span><span class="n">module</span><span class="si">}</span><span class="s2">: </span><span class="si">{</span><span class="n">skipped</span><span class="o">/</span><span class="mi">2</span><span class="o">**</span><span class="mi">20</span><span class="si">}</span><span class="s2">M params&quot;</span><span class="p">)</span>
+                            <span class="n">manager</span><span class="o">.</span><span class="n">register_module_override</span><span class="p">(</span><span class="n">module</span><span class="p">,</span> <span class="s2">&quot;weight&quot;</span><span class="p">,</span> <span class="p">{</span><span class="s2">&quot;optim_bits&quot;</span><span class="p">:</span> <span class="mi">32</span><span class="p">})</span>
+                            <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;bitsandbytes: will optimize </span><span class="si">{</span><span class="n">module</span><span class="si">}</span><span class="s2"> in fp32&quot;</span><span class="p">)</span>
+                    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;skipped: </span><span class="si">{</span><span class="n">skipped</span><span class="o">/</span><span class="mi">2</span><span class="o">**</span><span class="mi">20</span><span class="si">}</span><span class="s2">M params&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">smp</span><span class="o">.</span><span class="n">DistributedOptimizer</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span></div>
+
+
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="RaftTrainer.get_optimizer_cls_and_kwargs">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_optimizer_cls_and_kwargs">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_optimizer_cls_and_kwargs</span><span class="p">(</span><span class="n">args</span><span class="p">:</span> <span class="n">TrainingArguments</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Any</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Returns the optimizer class and optimizer parameters based on the training arguments.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            args (`transformers.training_args.TrainingArguments`):</span>
+<span class="sd">                The training arguments for the training session.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="c1"># parse args.optim_args</span>
+        <span class="n">optim_args</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_args</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">mapping</span> <span class="ow">in</span> <span class="n">args</span><span class="o">.</span><span class="n">optim_args</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot; &quot;</span><span class="p">,</span> <span class="s2">&quot;&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;,&quot;</span><span class="p">):</span>
+                <span class="n">key</span><span class="p">,</span> <span class="n">value</span> <span class="o">=</span> <span class="n">mapping</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;=&quot;</span><span class="p">)</span>
+                <span class="n">optim_args</span><span class="p">[</span><span class="n">key</span><span class="p">]</span> <span class="o">=</span> <span class="n">value</span>
+
+        <span class="n">optimizer_kwargs</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;lr&quot;</span><span class="p">:</span> <span class="n">args</span><span class="o">.</span><span class="n">learning_rate</span><span class="p">}</span>
+
+        <span class="n">adam_kwargs</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;betas&quot;</span><span class="p">:</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">adam_beta1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">adam_beta2</span><span class="p">),</span>
+            <span class="s2">&quot;eps&quot;</span><span class="p">:</span> <span class="n">args</span><span class="o">.</span><span class="n">adam_epsilon</span><span class="p">,</span>
+        <span class="p">}</span>
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAFACTOR</span><span class="p">:</span>
+            <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">Adafactor</span>
+            <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">({</span><span class="s2">&quot;scale_parameter&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;relative_step&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">})</span>
+        <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAMW_HF</span><span class="p">:</span>
+            <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.optimization</span><span class="w"> </span><span class="kn">import</span> <span class="n">AdamW</span>
+
+            <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">AdamW</span>
+            <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adam_kwargs</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">optim</span> <span class="ow">in</span> <span class="p">[</span><span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAMW_TORCH</span><span class="p">,</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAMW_TORCH_FUSED</span><span class="p">]:</span>
+            <span class="kn">from</span><span class="w"> </span><span class="nn">torch.optim</span><span class="w"> </span><span class="kn">import</span> <span class="n">AdamW</span>
+
+            <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">AdamW</span>
+            <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adam_kwargs</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAMW_TORCH_FUSED</span><span class="p">:</span>
+                <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">({</span><span class="s2">&quot;fused&quot;</span><span class="p">:</span> <span class="kc">True</span><span class="p">})</span>
+        <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAMW_TORCH_XLA</span><span class="p">:</span>
+            <span class="k">try</span><span class="p">:</span>
+                <span class="kn">from</span><span class="w"> </span><span class="nn">torch_xla.amp.syncfree</span><span class="w"> </span><span class="kn">import</span> <span class="n">AdamW</span>
+
+                <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">AdamW</span>
+                <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adam_kwargs</span><span class="p">)</span>
+            <span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Trainer failed to import syncfree AdamW from torch_xla.&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAMW_APEX_FUSED</span><span class="p">:</span>
+            <span class="k">try</span><span class="p">:</span>
+                <span class="kn">from</span><span class="w"> </span><span class="nn">apex.optimizers</span><span class="w"> </span><span class="kn">import</span> <span class="n">FusedAdam</span>
+
+                <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">FusedAdam</span>
+                <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adam_kwargs</span><span class="p">)</span>
+            <span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Trainer tried to instantiate apex FusedAdam but apex is not installed!&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAMW_BNB</span><span class="p">:</span>
+            <span class="k">try</span><span class="p">:</span>
+                <span class="kn">from</span><span class="w"> </span><span class="nn">bitsandbytes.optim</span><span class="w"> </span><span class="kn">import</span> <span class="n">Adam8bit</span>
+
+                <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">Adam8bit</span>
+                <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adam_kwargs</span><span class="p">)</span>
+            <span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Trainer tried to instantiate bnb Adam8bit but bnb is not installed!&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAMW_ANYPRECISION</span><span class="p">:</span>
+            <span class="k">try</span><span class="p">:</span>
+                <span class="kn">from</span><span class="w"> </span><span class="nn">torchdistx.optimizers</span><span class="w"> </span><span class="kn">import</span> <span class="n">AnyPrecisionAdamW</span>
+
+                <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">AnyPrecisionAdamW</span>
+                <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="n">adam_kwargs</span><span class="p">)</span>
+
+                <span class="c1"># TODO Change dtypes back to M=FP32, Var = BF16, Kahan = False once they can be cast together in torchdistx.</span>
+                <span class="n">optimizer_kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">(</span>
+                    <span class="p">{</span>
+                        <span class="s2">&quot;use_kahan_summation&quot;</span><span class="p">:</span> <span class="n">strtobool</span><span class="p">(</span><span class="n">optim_args</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;use_kahan_summation&quot;</span><span class="p">,</span> <span class="s2">&quot;False&quot;</span><span class="p">)),</span>
+                        <span class="s2">&quot;momentum_dtype&quot;</span><span class="p">:</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">torch</span><span class="p">,</span> <span class="n">optim_args</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;momentum_dtype&quot;</span><span class="p">,</span> <span class="s2">&quot;float32&quot;</span><span class="p">)),</span>
+                        <span class="s2">&quot;variance_dtype&quot;</span><span class="p">:</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">torch</span><span class="p">,</span> <span class="n">optim_args</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;variance_dtype&quot;</span><span class="p">,</span> <span class="s2">&quot;float32&quot;</span><span class="p">)),</span>
+                        <span class="s2">&quot;compensation_buffer_dtype&quot;</span><span class="p">:</span> <span class="nb">getattr</span><span class="p">(</span>
+                            <span class="n">torch</span><span class="p">,</span> <span class="n">optim_args</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;compensation_buffer_dtype&quot;</span><span class="p">,</span> <span class="s2">&quot;bfloat16&quot;</span><span class="p">)</span>
+                        <span class="p">),</span>
+                    <span class="p">}</span>
+                <span class="p">)</span>
+            <span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Please install https://github.com/pytorch/torchdistx&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">SGD</span><span class="p">:</span>
+            <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">SGD</span>
+        <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">optim</span> <span class="o">==</span> <span class="n">OptimizerNames</span><span class="o">.</span><span class="n">ADAGRAD</span><span class="p">:</span>
+            <span class="n">optimizer_cls</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">Adagrad</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Trainer cannot instantiate unsupported optimizer: </span><span class="si">{</span><span class="n">args</span><span class="o">.</span><span class="n">optim</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">optimizer_cls</span><span class="p">,</span> <span class="n">optimizer_kwargs</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.create_scheduler">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_scheduler">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">create_scheduler</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_training_steps</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">optimizer</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">Optimizer</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Setup the scheduler. The optimizer of the trainer must have been set up either before this method is called or</span>
+<span class="sd">        passed as an argument.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            num_training_steps (int): The number of training steps to do.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1">############</span>
+        <span class="n">num_training_steps</span> <span class="o">*=</span> <span class="mi">3</span>
+        <span class="c1">############</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span> <span class="o">=</span> <span class="n">get_scheduler</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">lr_scheduler_type</span><span class="p">,</span>
+                <span class="n">optimizer</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="k">if</span> <span class="n">optimizer</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">optimizer</span><span class="p">,</span>
+                <span class="n">num_warmup_steps</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">get_warmup_steps</span><span class="p">(</span><span class="n">num_training_steps</span><span class="p">),</span>
+                <span class="n">num_training_steps</span><span class="o">=</span><span class="n">num_training_steps</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.num_examples">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.num_examples">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">num_examples</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dataloader</span><span class="p">:</span> <span class="n">DataLoader</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Helper to get number of samples in a [`~torch.utils.data.DataLoader`] by accessing its dataset. When</span>
+<span class="sd">        dataloader.dataset does not exist or has no length, estimates as best it can</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">try</span><span class="p">:</span>
+            <span class="n">dataset</span> <span class="o">=</span> <span class="n">dataloader</span><span class="o">.</span><span class="n">dataset</span>
+            <span class="c1"># Special case for IterableDatasetShard, we need to dig deeper</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">dataset</span><span class="p">,</span> <span class="n">IterableDatasetShard</span><span class="p">):</span>
+                <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="n">dataloader</span><span class="o">.</span><span class="n">dataset</span><span class="o">.</span><span class="n">dataset</span><span class="p">)</span>
+            <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="n">dataloader</span><span class="o">.</span><span class="n">dataset</span><span class="p">)</span>
+        <span class="k">except</span> <span class="p">(</span><span class="ne">NameError</span><span class="p">,</span> <span class="ne">AttributeError</span><span class="p">,</span> <span class="ne">TypeError</span><span class="p">):</span>  <span class="c1"># no dataset or length, estimate by length of dataloader</span>
+            <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="n">dataloader</span><span class="p">)</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">per_device_train_batch_size</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._hp_search_setup">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._hp_search_setup">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_hp_search_setup</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">trial</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="s2">&quot;optuna.Trial&quot;</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;HP search setup code&quot;&quot;&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_trial</span> <span class="o">=</span> <span class="n">trial</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">trial</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">OPTUNA</span><span class="p">:</span>
+            <span class="n">params</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_space</span><span class="p">(</span><span class="n">trial</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">RAY</span><span class="p">:</span>
+            <span class="n">params</span> <span class="o">=</span> <span class="n">trial</span>
+            <span class="n">params</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;wandb&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">SIGOPT</span><span class="p">:</span>
+            <span class="n">params</span> <span class="o">=</span> <span class="p">{</span><span class="n">k</span><span class="p">:</span> <span class="nb">int</span><span class="p">(</span><span class="n">v</span><span class="p">)</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">v</span><span class="p">,</span> <span class="nb">str</span><span class="p">)</span> <span class="k">else</span> <span class="n">v</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">trial</span><span class="o">.</span><span class="n">assignments</span><span class="o">.</span><span class="n">items</span><span class="p">()}</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">WANDB</span><span class="p">:</span>
+            <span class="n">params</span> <span class="o">=</span> <span class="n">trial</span>
+
+        <span class="k">for</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span> <span class="ow">in</span> <span class="n">params</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="n">key</span><span class="p">):</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;Trying to set </span><span class="si">{</span><span class="n">key</span><span class="si">}</span><span class="s2"> in the hyperparameter search but there is no corresponding field in&quot;</span>
+                    <span class="s2">&quot; `TrainingArguments`.&quot;</span>
+                <span class="p">)</span>
+                <span class="k">continue</span>
+            <span class="n">old_attr</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="n">key</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+            <span class="c1"># Casting value to the proper type</span>
+            <span class="k">if</span> <span class="n">old_attr</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">value</span> <span class="o">=</span> <span class="nb">type</span><span class="p">(</span><span class="n">old_attr</span><span class="p">)(</span><span class="n">value</span><span class="p">)</span>
+            <span class="nb">setattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">OPTUNA</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Trial: </span><span class="si">{</span><span class="n">trial</span><span class="o">.</span><span class="n">params</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">SIGOPT</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;SigOpt Assignments: </span><span class="si">{</span><span class="n">trial</span><span class="o">.</span><span class="n">assignments</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">WANDB</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;W&amp;B Sweep parameters: </span><span class="si">{</span><span class="n">trial</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+            <span class="c1"># Rebuild the deepspeed config to reflect the updated training parameters</span>
+            <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.deepspeed</span><span class="w"> </span><span class="kn">import</span> <span class="n">HfTrainerDeepSpeedConfig</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hf_deepspeed_config</span> <span class="o">=</span> <span class="n">HfTrainerDeepSpeedConfig</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hf_deepspeed_config</span><span class="o">.</span><span class="n">trainer_config_process</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._report_to_hp_search">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._report_to_hp_search">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_report_to_hp_search</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">trial</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="s2">&quot;optuna.Trial&quot;</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]],</span> <span class="n">step</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">metrics</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">float</span><span class="p">]):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="n">trial</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">objective</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_objective</span><span class="p">(</span><span class="n">metrics</span><span class="o">.</span><span class="n">copy</span><span class="p">())</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">OPTUNA</span><span class="p">:</span>
+            <span class="kn">import</span><span class="w"> </span><span class="nn">optuna</span>
+
+            <span class="n">trial</span><span class="o">.</span><span class="n">report</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">objective</span><span class="p">,</span> <span class="n">step</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">trial</span><span class="o">.</span><span class="n">should_prune</span><span class="p">():</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_train_end</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">)</span>
+                <span class="k">raise</span> <span class="n">optuna</span><span class="o">.</span><span class="n">TrialPruned</span><span class="p">()</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">RAY</span><span class="p">:</span>
+            <span class="kn">from</span><span class="w"> </span><span class="nn">ray</span><span class="w"> </span><span class="kn">import</span> <span class="n">tune</span>
+
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_tune_save_checkpoint</span><span class="p">()</span>
+            <span class="n">tune</span><span class="o">.</span><span class="n">report</span><span class="p">(</span><span class="n">objective</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">objective</span><span class="p">,</span> <span class="o">**</span><span class="n">metrics</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._tune_save_checkpoint">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._tune_save_checkpoint">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_tune_save_checkpoint</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="kn">from</span><span class="w"> </span><span class="nn">ray</span><span class="w"> </span><span class="kn">import</span> <span class="n">tune</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_tune_checkpoints</span><span class="p">:</span>
+            <span class="k">return</span>
+        <span class="k">with</span> <span class="n">tune</span><span class="o">.</span><span class="n">checkpoint_dir</span><span class="p">(</span><span class="n">step</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span><span class="p">)</span> <span class="k">as</span> <span class="n">checkpoint_dir</span><span class="p">:</span>
+            <span class="n">output_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint_dir</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">PREFIX_CHECKPOINT_DIR</span><span class="si">}</span><span class="s2">-</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">save_model</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">_internal_call</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">save_to_json</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">TRAINER_STATE_NAME</span><span class="p">))</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">state_dict</span><span class="p">(),</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">OPTIMIZER_NAME</span><span class="p">))</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span><span class="o">.</span><span class="n">state_dict</span><span class="p">(),</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">SCHEDULER_NAME</span><span class="p">))</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.call_model_init">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.call_model_init">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">call_model_init</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">trial</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="n">model_init_argcount</span> <span class="o">=</span> <span class="n">number_of_arguments</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model_init</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">model_init_argcount</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_init</span><span class="p">()</span>
+        <span class="k">elif</span> <span class="n">model_init_argcount</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_init</span><span class="p">(</span><span class="n">trial</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;model_init should have 0 or 1 argument.&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">model</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;model_init should not return None.&quot;</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">model</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.torch_jit_model_eval">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.torch_jit_model_eval">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">torch_jit_model_eval</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">dataloader</span><span class="p">,</span> <span class="n">training</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">training</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">dataloader</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;failed to use PyTorch jit mode due to current dataloader is none.&quot;</span><span class="p">)</span>
+                <span class="k">return</span> <span class="n">model</span>
+            <span class="n">example_batch</span> <span class="o">=</span> <span class="nb">next</span><span class="p">(</span><span class="nb">iter</span><span class="p">(</span><span class="n">dataloader</span><span class="p">))</span>
+            <span class="n">example_batch</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_prepare_inputs</span><span class="p">(</span><span class="n">example_batch</span><span class="p">)</span>
+            <span class="k">try</span><span class="p">:</span>
+                <span class="n">jit_model</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+                <span class="k">with</span> <span class="n">ContextManagers</span><span class="p">([</span><span class="bp">self</span><span class="o">.</span><span class="n">autocast_smart_context_manager</span><span class="p">(</span><span class="n">cache_enabled</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">()]):</span>
+                    <span class="k">if</span> <span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">__version__</span><span class="p">)</span><span class="o">.</span><span class="n">base_version</span><span class="p">)</span> <span class="o">&gt;=</span> <span class="n">version</span><span class="o">.</span><span class="n">parse</span><span class="p">(</span><span class="s2">&quot;1.14.0&quot;</span><span class="p">):</span>
+                        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">example_batch</span><span class="p">,</span> <span class="nb">dict</span><span class="p">):</span>
+                            <span class="n">jit_model</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">trace</span><span class="p">(</span><span class="n">jit_model</span><span class="p">,</span> <span class="n">example_kwarg_inputs</span><span class="o">=</span><span class="n">example_batch</span><span class="p">,</span> <span class="n">strict</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+                        <span class="k">else</span><span class="p">:</span>
+                            <span class="n">jit_model</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">trace</span><span class="p">(</span>
+                                <span class="n">jit_model</span><span class="p">,</span>
+                                <span class="n">example_kwarg_inputs</span><span class="o">=</span><span class="p">{</span><span class="n">key</span><span class="p">:</span> <span class="n">example_batch</span><span class="p">[</span><span class="n">key</span><span class="p">]</span> <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">example_batch</span><span class="p">},</span>
+                                <span class="n">strict</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                            <span class="p">)</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">jit_inputs</span> <span class="o">=</span> <span class="p">[]</span>
+                        <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">example_batch</span><span class="p">:</span>
+                            <span class="n">example_tensor</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">ones_like</span><span class="p">(</span><span class="n">example_batch</span><span class="p">[</span><span class="n">key</span><span class="p">])</span>
+                            <span class="n">jit_inputs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">example_tensor</span><span class="p">)</span>
+                        <span class="n">jit_inputs</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">jit_inputs</span><span class="p">)</span>
+                        <span class="n">jit_model</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">trace</span><span class="p">(</span><span class="n">jit_model</span><span class="p">,</span> <span class="n">jit_inputs</span><span class="p">,</span> <span class="n">strict</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+                <span class="n">jit_model</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">jit</span><span class="o">.</span><span class="n">freeze</span><span class="p">(</span><span class="n">jit_model</span><span class="p">)</span>
+                <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+                    <span class="n">jit_model</span><span class="p">(</span><span class="o">**</span><span class="n">example_batch</span><span class="p">)</span>
+                    <span class="n">jit_model</span><span class="p">(</span><span class="o">**</span><span class="n">example_batch</span><span class="p">)</span>
+                <span class="n">model</span> <span class="o">=</span> <span class="n">jit_model</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">use_cpu_amp</span> <span class="o">=</span> <span class="kc">False</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">use_cuda_amp</span> <span class="o">=</span> <span class="kc">False</span>
+            <span class="k">except</span> <span class="p">(</span><span class="ne">RuntimeError</span><span class="p">,</span> <span class="ne">TypeError</span><span class="p">,</span> <span class="ne">ValueError</span><span class="p">,</span> <span class="ne">NameError</span><span class="p">,</span> <span class="ne">IndexError</span><span class="p">)</span> <span class="k">as</span> <span class="n">e</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;failed to use PyTorch jit mode due to: </span><span class="si">{</span><span class="n">e</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">model</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.ipex_optimize_model">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.ipex_optimize_model">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">ipex_optimize_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">training</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">is_ipex_available</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">ImportError</span><span class="p">(</span>
+                <span class="s2">&quot;Using IPEX but IPEX is not installed or IPEX&#39;s version does not match current PyTorch, please refer&quot;</span>
+                <span class="s2">&quot; to https://github.com/intel/intel-extension-for-pytorch.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="kn">import</span><span class="w"> </span><span class="nn">intel_extension_for_pytorch</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">ipex</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">training</span><span class="p">:</span>
+            <span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+            <span class="n">dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">bfloat16</span> <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_in_train</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">bf16_full_eval</span> <span class="k">else</span> <span class="n">dtype</span>
+            <span class="c1"># conv_bn_folding is disabled as it fails in symbolic tracing, resulting in ipex warnings</span>
+            <span class="n">model</span> <span class="o">=</span> <span class="n">ipex</span><span class="o">.</span><span class="n">optimize</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">level</span><span class="o">=</span><span class="s2">&quot;O1&quot;</span><span class="p">,</span> <span class="n">conv_bn_folding</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_in_train</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">model</span><span class="o">.</span><span class="n">training</span><span class="p">:</span>
+                <span class="n">model</span><span class="o">.</span><span class="n">train</span><span class="p">()</span>
+            <span class="n">model</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">ipex</span><span class="o">.</span><span class="n">optimize</span><span class="p">(</span>
+                <span class="n">model</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">optimizer</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">,</span> <span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">level</span><span class="o">=</span><span class="s2">&quot;O1&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">model</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._wrap_model">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._wrap_model">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_wrap_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">training</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">dataloader</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">torch_compile</span><span class="p">:</span>
+            <span class="n">model</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">compile</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">backend</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">torch_compile_backend</span><span class="p">,</span> <span class="n">mode</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">torch_compile_mode</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">use_ipex</span><span class="p">:</span>
+            <span class="n">dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">bfloat16</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_cpu_amp</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span>
+            <span class="n">model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">ipex_optimize_model</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">training</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+            <span class="c1"># Wrapping the base model twice in a DistributedModel will raise an error.</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span><span class="p">,</span> <span class="n">smp</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">DistributedModel</span><span class="p">):</span>
+                <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span>
+            <span class="k">return</span> <span class="n">smp</span><span class="o">.</span><span class="n">DistributedModel</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">backward_passes_per_step</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span><span class="p">)</span>
+
+        <span class="c1"># already initialized its own DDP and AMP</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span>
+
+        <span class="c1"># train/eval could be run multiple-times - if already wrapped, don&#39;t re-wrap it again</span>
+        <span class="k">if</span> <span class="n">unwrap_model</span><span class="p">(</span><span class="n">model</span><span class="p">)</span> <span class="ow">is</span> <span class="ow">not</span> <span class="n">model</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">model</span>
+
+        <span class="c1"># Mixed precision training with apex (torch &lt; 1.6)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_apex</span> <span class="ow">and</span> <span class="n">training</span><span class="p">:</span>
+            <span class="n">model</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">amp</span><span class="o">.</span><span class="n">initialize</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">,</span> <span class="n">opt_level</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fp16_opt_level</span><span class="p">)</span>
+
+        <span class="c1"># Multi-gpu training (should be after apex fp16 initialization)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">n_gpu</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">model</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">DataParallel</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">jit_mode_eval</span><span class="p">:</span>
+            <span class="n">start_time</span> <span class="o">=</span> <span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span>
+            <span class="n">model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">torch_jit_model_eval</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">dataloader</span><span class="p">,</span> <span class="n">training</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">jit_compilation_time</span> <span class="o">=</span> <span class="nb">round</span><span class="p">(</span><span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span> <span class="o">-</span> <span class="n">start_time</span><span class="p">,</span> <span class="mi">4</span><span class="p">)</span>
+
+        <span class="c1"># Note: in torch.distributed mode, there&#39;s no point in wrapping the model</span>
+        <span class="c1"># inside a DistributedDataParallel as we&#39;ll be under `no_grad` anyways.</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">training</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">model</span>
+
+        <span class="c1"># Distributed training (should be after apex fp16 initialization)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="c1"># Sharded DDP!</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="o">==</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">SIMPLE</span><span class="p">:</span>
+                <span class="n">model</span> <span class="o">=</span> <span class="n">ShardedDDP</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">mixed_precision</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fp16</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">bf16</span>
+                <span class="n">cpu_offload</span> <span class="o">=</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">OFFLOAD</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">sharded_ddp</span>
+                <span class="n">zero_3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="o">==</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">ZERO_DP_3</span>
+                <span class="c1"># XXX: Breaking the self.model convention but I see no way around it for now.</span>
+                <span class="k">if</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">AUTO_WRAP</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">sharded_ddp</span><span class="p">:</span>
+                    <span class="n">model</span> <span class="o">=</span> <span class="n">auto_wrap</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">model</span> <span class="o">=</span> <span class="n">FullyShardedDDP</span><span class="p">(</span>
+                    <span class="n">model</span><span class="p">,</span>
+                    <span class="n">mixed_precision</span><span class="o">=</span><span class="n">mixed_precision</span><span class="p">,</span>
+                    <span class="n">reshard_after_forward</span><span class="o">=</span><span class="n">zero_3</span><span class="p">,</span>
+                    <span class="n">cpu_offload</span><span class="o">=</span><span class="n">cpu_offload</span><span class="p">,</span>
+                <span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+        <span class="c1"># Distributed training using PyTorch FSDP</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">fsdp</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span><span class="p">[</span><span class="s2">&quot;xla&quot;</span><span class="p">]:</span>
+                <span class="c1"># PyTorch FSDP!</span>
+                <span class="kn">from</span><span class="w"> </span><span class="nn">torch.distributed.fsdp.fully_sharded_data_parallel</span><span class="w"> </span><span class="kn">import</span> <span class="n">CPUOffload</span><span class="p">,</span> <span class="n">MixedPrecision</span>
+                <span class="kn">from</span><span class="w"> </span><span class="nn">torch.distributed.fsdp.fully_sharded_data_parallel</span><span class="w"> </span><span class="kn">import</span> <span class="n">FullyShardedDataParallel</span> <span class="k">as</span> <span class="n">FSDP</span>
+                <span class="kn">from</span><span class="w"> </span><span class="nn">torch.distributed.fsdp.wrap</span><span class="w"> </span><span class="kn">import</span> <span class="n">size_based_auto_wrap_policy</span><span class="p">,</span> <span class="n">transformer_auto_wrap_policy</span>
+
+                <span class="k">if</span> <span class="n">FSDPOption</span><span class="o">.</span><span class="n">OFFLOAD</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp</span><span class="p">:</span>
+                    <span class="n">cpu_offload</span> <span class="o">=</span> <span class="n">CPUOffload</span><span class="p">(</span><span class="n">offload_params</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">cpu_offload</span> <span class="o">=</span> <span class="n">CPUOffload</span><span class="p">(</span><span class="n">offload_params</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+                <span class="n">auto_wrap_policy</span> <span class="o">=</span> <span class="kc">None</span>
+
+                <span class="k">if</span> <span class="n">FSDPOption</span><span class="o">.</span><span class="n">AUTO_WRAP</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span><span class="p">[</span><span class="s2">&quot;fsdp_min_num_params&quot;</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="n">auto_wrap_policy</span> <span class="o">=</span> <span class="n">functools</span><span class="o">.</span><span class="n">partial</span><span class="p">(</span>
+                            <span class="n">size_based_auto_wrap_policy</span><span class="p">,</span> <span class="n">min_num_params</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span><span class="p">[</span><span class="s2">&quot;fsdp_min_num_params&quot;</span><span class="p">]</span>
+                        <span class="p">)</span>
+                    <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;fsdp_transformer_layer_cls_to_wrap&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="n">transformer_cls_to_wrap</span> <span class="o">=</span> <span class="nb">set</span><span class="p">()</span>
+                        <span class="k">for</span> <span class="n">layer_class</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span><span class="p">[</span><span class="s2">&quot;fsdp_transformer_layer_cls_to_wrap&quot;</span><span class="p">]:</span>
+                            <span class="n">transformer_cls</span> <span class="o">=</span> <span class="n">get_module_class_from_name</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">layer_class</span><span class="p">)</span>
+                            <span class="k">if</span> <span class="n">transformer_cls</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                                <span class="k">raise</span> <span class="ne">Exception</span><span class="p">(</span><span class="s2">&quot;Could not find the transformer layer class to wrap in the model.&quot;</span><span class="p">)</span>
+                            <span class="k">else</span><span class="p">:</span>
+                                <span class="n">transformer_cls_to_wrap</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">transformer_cls</span><span class="p">)</span>
+                        <span class="n">auto_wrap_policy</span> <span class="o">=</span> <span class="n">functools</span><span class="o">.</span><span class="n">partial</span><span class="p">(</span>
+                            <span class="n">transformer_auto_wrap_policy</span><span class="p">,</span>
+                            <span class="c1"># Transformer layer class to wrap</span>
+                            <span class="n">transformer_layer_cls</span><span class="o">=</span><span class="n">transformer_cls_to_wrap</span><span class="p">,</span>
+                        <span class="p">)</span>
+                <span class="n">mixed_precision_policy</span> <span class="o">=</span> <span class="kc">None</span>
+                <span class="n">dtype</span> <span class="o">=</span> <span class="kc">None</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fp16</span><span class="p">:</span>
+                    <span class="n">dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">float16</span>
+                <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">bf16</span><span class="p">:</span>
+                    <span class="n">dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">bfloat16</span>
+                <span class="k">if</span> <span class="n">dtype</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">mixed_precision_policy</span> <span class="o">=</span> <span class="n">MixedPrecision</span><span class="p">(</span><span class="n">param_dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">reduce_dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">,</span> <span class="n">buffer_dtype</span><span class="o">=</span><span class="n">dtype</span><span class="p">)</span>
+                <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">model</span><span class="p">)</span> <span class="o">!=</span> <span class="n">FSDP</span><span class="p">:</span>
+                    <span class="c1"># XXX: Breaking the self.model convention but I see no way around it for now.</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">model</span> <span class="o">=</span> <span class="n">FSDP</span><span class="p">(</span>
+                        <span class="n">model</span><span class="p">,</span>
+                        <span class="n">sharding_strategy</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">fsdp</span><span class="p">,</span>
+                        <span class="n">cpu_offload</span><span class="o">=</span><span class="n">cpu_offload</span><span class="p">,</span>
+                        <span class="n">auto_wrap_policy</span><span class="o">=</span><span class="n">auto_wrap_policy</span><span class="p">,</span>
+                        <span class="n">mixed_precision</span><span class="o">=</span><span class="n">mixed_precision_policy</span><span class="p">,</span>
+                        <span class="n">device_id</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">,</span>
+                        <span class="n">backward_prefetch</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">backward_prefetch</span><span class="p">,</span>
+                        <span class="n">forward_prefetch</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">forword_prefetch</span><span class="p">,</span>
+                        <span class="n">limit_all_gathers</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">limit_all_gathers</span><span class="p">,</span>
+                    <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">try</span><span class="p">:</span>
+                    <span class="kn">from</span><span class="w"> </span><span class="nn">torch_xla.distributed.fsdp</span><span class="w"> </span><span class="kn">import</span> <span class="n">XlaFullyShardedDataParallel</span> <span class="k">as</span> <span class="n">FSDP</span>
+                    <span class="kn">from</span><span class="w"> </span><span class="nn">torch_xla.distributed.fsdp</span><span class="w"> </span><span class="kn">import</span> <span class="n">checkpoint_module</span>
+                    <span class="kn">from</span><span class="w"> </span><span class="nn">torch_xla.distributed.fsdp.wrap</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+                        <span class="n">size_based_auto_wrap_policy</span><span class="p">,</span>
+                        <span class="n">transformer_auto_wrap_policy</span><span class="p">,</span>
+                    <span class="p">)</span>
+                <span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ImportError</span><span class="p">(</span><span class="s2">&quot;Missing XLA FSDP related module; please make sure to use torch-xla &gt;= 2.0.&quot;</span><span class="p">)</span>
+                <span class="n">auto_wrap_policy</span> <span class="o">=</span> <span class="kc">None</span>
+                <span class="n">auto_wrapper_callable</span> <span class="o">=</span> <span class="kc">None</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span><span class="p">[</span><span class="s2">&quot;fsdp_min_num_params&quot;</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">auto_wrap_policy</span> <span class="o">=</span> <span class="n">functools</span><span class="o">.</span><span class="n">partial</span><span class="p">(</span>
+                        <span class="n">size_based_auto_wrap_policy</span><span class="p">,</span> <span class="n">min_num_params</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span><span class="p">[</span><span class="s2">&quot;fsdp_min_num_params&quot;</span><span class="p">]</span>
+                    <span class="p">)</span>
+                <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;fsdp_transformer_layer_cls_to_wrap&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">transformer_cls_to_wrap</span> <span class="o">=</span> <span class="nb">set</span><span class="p">()</span>
+                    <span class="k">for</span> <span class="n">layer_class</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span><span class="p">[</span><span class="s2">&quot;fsdp_transformer_layer_cls_to_wrap&quot;</span><span class="p">]:</span>
+                        <span class="n">transformer_cls</span> <span class="o">=</span> <span class="n">get_module_class_from_name</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">layer_class</span><span class="p">)</span>
+                        <span class="k">if</span> <span class="n">transformer_cls</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                            <span class="k">raise</span> <span class="ne">Exception</span><span class="p">(</span><span class="s2">&quot;Could not find the transformer layer class to wrap in the model.&quot;</span><span class="p">)</span>
+                        <span class="k">else</span><span class="p">:</span>
+                            <span class="n">transformer_cls_to_wrap</span><span class="o">.</span><span class="n">add</span><span class="p">(</span><span class="n">transformer_cls</span><span class="p">)</span>
+                    <span class="n">auto_wrap_policy</span> <span class="o">=</span> <span class="n">functools</span><span class="o">.</span><span class="n">partial</span><span class="p">(</span>
+                        <span class="n">transformer_auto_wrap_policy</span><span class="p">,</span>
+                        <span class="c1"># Transformer layer class to wrap</span>
+                        <span class="n">transformer_layer_cls</span><span class="o">=</span><span class="n">transformer_cls_to_wrap</span><span class="p">,</span>
+                    <span class="p">)</span>
+                <span class="n">fsdp_kwargs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">xla_fsdp_config</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fsdp_config</span><span class="p">[</span><span class="s2">&quot;xla_fsdp_grad_ckpt&quot;</span><span class="p">]:</span>
+                    <span class="c1"># Apply gradient checkpointing to auto-wrapped sub-modules if specified</span>
+                    <span class="k">def</span><span class="w"> </span><span class="nf">auto_wrapper_callable</span><span class="p">(</span><span class="n">m</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+                        <span class="k">return</span> <span class="n">FSDP</span><span class="p">(</span><span class="n">checkpoint_module</span><span class="p">(</span><span class="n">m</span><span class="p">),</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+
+                <span class="c1"># Wrap the base model with an outer FSDP wrapper</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">model</span> <span class="o">=</span> <span class="n">FSDP</span><span class="p">(</span>
+                    <span class="n">model</span><span class="p">,</span>
+                    <span class="n">auto_wrap_policy</span><span class="o">=</span><span class="n">auto_wrap_policy</span><span class="p">,</span>
+                    <span class="n">auto_wrapper_callable</span><span class="o">=</span><span class="n">auto_wrapper_callable</span><span class="p">,</span>
+                    <span class="o">**</span><span class="n">fsdp_kwargs</span><span class="p">,</span>
+                <span class="p">)</span>
+
+                <span class="c1"># Patch `xm.optimizer_step` should not reduce gradients in this case,</span>
+                <span class="c1"># as FSDP does not need gradient reduction over sharded parameters.</span>
+                <span class="k">def</span><span class="w"> </span><span class="nf">patched_optimizer_step</span><span class="p">(</span><span class="n">optimizer</span><span class="p">,</span> <span class="n">barrier</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">optimizer_args</span><span class="o">=</span><span class="p">{}):</span>
+                    <span class="n">loss</span> <span class="o">=</span> <span class="n">optimizer</span><span class="o">.</span><span class="n">step</span><span class="p">(</span><span class="o">**</span><span class="n">optimizer_args</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">barrier</span><span class="p">:</span>
+                        <span class="n">xm</span><span class="o">.</span><span class="n">mark_step</span><span class="p">()</span>
+                    <span class="k">return</span> <span class="n">loss</span>
+
+                <span class="n">xm</span><span class="o">.</span><span class="n">optimizer_step</span> <span class="o">=</span> <span class="n">patched_optimizer_step</span>
+        <span class="k">elif</span> <span class="n">is_sagemaker_dp_enabled</span><span class="p">():</span>
+            <span class="n">model</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">parallel</span><span class="o">.</span><span class="n">DistributedDataParallel</span><span class="p">(</span>
+                <span class="n">model</span><span class="p">,</span> <span class="n">device_ids</span><span class="o">=</span><span class="p">[</span><span class="nb">int</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">getenv</span><span class="p">(</span><span class="s2">&quot;SMDATAPARALLEL_LOCAL_RANK&quot;</span><span class="p">))]</span>
+            <span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">!=</span> <span class="o">-</span><span class="mi">1</span><span class="p">:</span>
+            <span class="n">kwargs</span> <span class="o">=</span> <span class="p">{}</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">ddp_find_unused_parameters</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;find_unused_parameters&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">ddp_find_unused_parameters</span>
+            <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">PreTrainedModel</span><span class="p">):</span>
+                <span class="c1"># find_unused_parameters breaks checkpointing as per</span>
+                <span class="c1"># https://github.com/huggingface/transformers/pull/4659#issuecomment-643356021</span>
+                <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;find_unused_parameters&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="ow">not</span> <span class="n">model</span><span class="o">.</span><span class="n">is_gradient_checkpointing</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;find_unused_parameters&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="kc">True</span>
+
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">ddp_bucket_cap_mb</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">kwargs</span><span class="p">[</span><span class="s2">&quot;bucket_cap_mb&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">ddp_bucket_cap_mb</span>
+            <span class="k">if</span> <span class="n">is_torch_neuroncore_available</span><span class="p">():</span>
+                <span class="k">return</span> <span class="n">model</span>
+            <span class="n">model</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">parallel</span><span class="o">.</span><span class="n">DistributedDataParallel</span><span class="p">(</span>
+                <span class="n">model</span><span class="p">,</span>
+                <span class="n">device_ids</span><span class="o">=</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">local_rank</span><span class="p">]</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">_n_gpu</span> <span class="o">!=</span> <span class="mi">0</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+                <span class="n">output_device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">local_rank</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">_n_gpu</span> <span class="o">!=</span> <span class="mi">0</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+                <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+            <span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">model</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.train">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.train">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">train</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">resume_from_checkpoint</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">trial</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="s2">&quot;optuna.Trial&quot;</span><span class="p">,</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">ignore_keys_for_eval</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">is_first_time</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Main training entry point.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            resume_from_checkpoint (`str` or `bool`, *optional*):</span>
+<span class="sd">                If a `str`, local path to a saved checkpoint as saved by a previous instance of [`Trainer`]. If a</span>
+<span class="sd">                `bool` and equals `True`, load the last checkpoint in *args.output_dir* as saved by a previous instance</span>
+<span class="sd">                of [`Trainer`]. If present, training will resume from the model/optimizer/scheduler states loaded here.</span>
+<span class="sd">            trial (`optuna.Trial` or `Dict[str, Any]`, *optional*):</span>
+<span class="sd">                The trial run or the hyperparameter dictionary for hyperparameter search.</span>
+<span class="sd">            ignore_keys_for_eval (`List[str]`, *optional*)</span>
+<span class="sd">                A list of keys in the output of your model (if it is a dictionary) that should be ignored when</span>
+<span class="sd">                gathering predictions for evaluation during the training.</span>
+<span class="sd">            kwargs:</span>
+<span class="sd">                Additional keyword arguments used to hide deprecated arguments</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">resume_from_checkpoint</span> <span class="ow">is</span> <span class="kc">False</span><span class="p">:</span>
+            <span class="n">resume_from_checkpoint</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="c1"># memory metrics - must set up as early as possible</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_memory_tracker</span><span class="o">.</span><span class="n">start</span><span class="p">()</span>
+
+        <span class="n">args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span>
+
+        <span class="c1">#self.is_in_train = True</span>
+
+        <span class="c1"># do_train is not a reliable argument, as it might not be set and .train() still called, so</span>
+        <span class="c1"># the following is a workaround:</span>
+        <span class="k">if</span> <span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">fp16_full_eval</span> <span class="ow">or</span> <span class="n">args</span><span class="o">.</span><span class="n">bf16_full_eval</span><span class="p">)</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">args</span><span class="o">.</span><span class="n">do_train</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_move_model_to_device</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="s2">&quot;model_path&quot;</span> <span class="ow">in</span> <span class="n">kwargs</span><span class="p">:</span>
+            <span class="n">resume_from_checkpoint</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;model_path&quot;</span><span class="p">)</span>
+            <span class="n">warnings</span><span class="o">.</span><span class="n">warn</span><span class="p">(</span>
+                <span class="s2">&quot;`model_path` is deprecated and will be removed in a future version. Use `resume_from_checkpoint` &quot;</span>
+                <span class="s2">&quot;instead.&quot;</span><span class="p">,</span>
+                <span class="ne">FutureWarning</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">kwargs</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;train() received got unexpected keyword arguments: </span><span class="si">{</span><span class="s1">&#39;, &#39;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="nb">list</span><span class="p">(</span><span class="n">kwargs</span><span class="o">.</span><span class="n">keys</span><span class="p">()))</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+        <span class="c1"># This might change the seed so needs to run first.</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_hp_search_setup</span><span class="p">(</span><span class="n">trial</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_train_batch_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">train_batch_size</span>
+
+        <span class="c1"># Model re-init</span>
+        <span class="n">model_reloaded</span> <span class="o">=</span> <span class="kc">False</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_init</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="c1"># Seed must be set before instantiating the model when using model_init.</span>
+            <span class="n">enable_full_determinism</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">seed</span><span class="p">)</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">full_determinism</span> <span class="k">else</span> <span class="n">set_seed</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">seed</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">call_model_init</span><span class="p">(</span><span class="n">trial</span><span class="p">)</span>
+            <span class="n">model_reloaded</span> <span class="o">=</span> <span class="kc">True</span>
+            <span class="c1"># Reinitializes optimizer and scheduler</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span>
+
+        <span class="c1"># Load potential model checkpoint</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="p">,</span> <span class="nb">bool</span><span class="p">)</span> <span class="ow">and</span> <span class="n">resume_from_checkpoint</span><span class="p">:</span>
+            <span class="n">resume_from_checkpoint</span> <span class="o">=</span> <span class="n">get_last_checkpoint</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">resume_from_checkpoint</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;No valid checkpoint found in output directory (</span><span class="si">{</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="si">}</span><span class="s2">)&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">resume_from_checkpoint</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">()</span> <span class="ow">and</span> <span class="n">args</span><span class="o">.</span><span class="n">deepspeed</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_load_from_checkpoint</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="p">)</span>
+
+        <span class="c1"># If model was re-initialized, put it on the right device and update self.model_wrapped</span>
+        <span class="k">if</span> <span class="n">model_reloaded</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">place_model_on_device</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_move_model_to_device</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span>
+        <span class="k">if</span> <span class="n">is_first_time</span><span class="p">:</span>
+            <span class="n">inner_training_loop1</span> <span class="o">=</span> <span class="n">find_executable_batch_size</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_inner_training_loop</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_train_batch_size</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">auto_find_batch_size</span>
+            <span class="p">)</span>
+            <span class="k">return</span> <span class="n">inner_training_loop1</span><span class="p">(</span>
+                <span class="n">args</span><span class="o">=</span><span class="n">args</span><span class="p">,</span>
+                <span class="n">resume_from_checkpoint</span><span class="o">=</span><span class="n">resume_from_checkpoint</span><span class="p">,</span>
+                <span class="n">trial</span><span class="o">=</span><span class="n">trial</span><span class="p">,</span>
+                <span class="n">ignore_keys_for_eval</span><span class="o">=</span><span class="n">ignore_keys_for_eval</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">inner_training_loop2</span> <span class="o">=</span> <span class="n">find_executable_batch_size</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_one_train</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_train_batch_size</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">auto_find_batch_size</span>
+            <span class="p">)</span>
+            <span class="k">return</span> <span class="n">inner_training_loop2</span><span class="p">(</span>
+                <span class="n">args</span><span class="o">=</span><span class="n">args</span><span class="p">,</span>
+                <span class="n">resume_from_checkpoint</span><span class="o">=</span><span class="n">resume_from_checkpoint</span><span class="p">,</span>
+                <span class="n">trial</span><span class="o">=</span><span class="n">trial</span><span class="p">,</span>
+                <span class="n">ignore_keys_for_eval</span><span class="o">=</span><span class="n">ignore_keys_for_eval</span><span class="p">,</span>
+            <span class="p">)</span></div>
+
+            
+    
+<div class="viewcode-block" id="RaftTrainer._one_train">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._one_train">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_one_train</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">batch_size</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">args</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">resume_from_checkpoint</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">trial</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">ignore_keys_for_eval</span><span class="o">=</span><span class="kc">None</span>
+    <span class="p">):</span>
+        <span class="c1">#print(self.lr_scheduler)</span>
+        <span class="c1">#print(dir(self.lr_scheduler))</span>
+        
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">state</span> <span class="o">=</span> <span class="n">TrainerState</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">is_hyper_param_search</span> <span class="o">=</span> <span class="n">trial</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+        <span class="c1"># Get dataloader</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_train_batch_size</span> <span class="o">=</span> <span class="n">batch_size</span>
+        <span class="c1"># Data loader and number of training steps</span>
+        <span class="n">train_dataloader</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_train_dataloader</span><span class="p">()</span>
+        <span class="c1">#print(&quot;AAAAAAA&quot;, len(train_dataloader))</span>
+
+        <span class="n">total_train_batch_size</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">train_batch_size</span> <span class="o">*</span> <span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span> <span class="o">*</span> <span class="n">args</span><span class="o">.</span><span class="n">world_size</span>
+
+        <span class="n">len_dataloader</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">has_length</span><span class="p">(</span><span class="n">train_dataloader</span><span class="p">):</span>
+            <span class="n">len_dataloader</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">train_dataloader</span><span class="p">)</span>
+            <span class="n">num_update_steps_per_epoch</span> <span class="o">=</span> <span class="n">len_dataloader</span> <span class="o">//</span> <span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span>
+            <span class="n">num_update_steps_per_epoch</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">num_update_steps_per_epoch</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+            <span class="n">num_examples</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_examples</span><span class="p">(</span><span class="n">train_dataloader</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">max_steps</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span>
+                <span class="n">num_train_epochs</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">//</span> <span class="n">num_update_steps_per_epoch</span> <span class="o">+</span> <span class="nb">int</span><span class="p">(</span>
+                    <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">%</span> <span class="n">num_update_steps_per_epoch</span> <span class="o">&gt;</span> <span class="mi">0</span>
+                <span class="p">)</span>
+                <span class="c1"># May be slightly incorrect if the last batch in the training dataloader has a smaller size but it&#39;s</span>
+                <span class="c1"># the best we can do.</span>
+                <span class="n">num_train_samples</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">*</span> <span class="n">total_train_batch_size</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">max_steps</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">num_train_epochs</span> <span class="o">*</span> <span class="n">num_update_steps_per_epoch</span><span class="p">)</span>
+                <span class="n">num_train_epochs</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">num_train_epochs</span><span class="p">)</span>
+                <span class="n">num_train_samples</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_examples</span><span class="p">(</span><span class="n">train_dataloader</span><span class="p">)</span> <span class="o">*</span> <span class="n">args</span><span class="o">.</span><span class="n">num_train_epochs</span>
+        <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>  <span class="c1"># Rely on max_steps when dataloader does not have a working size</span>
+            <span class="n">max_steps</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span>
+            <span class="c1"># Setting a very large number of epochs so we go as many times as necessary over the iterator.</span>
+            <span class="n">num_train_epochs</span> <span class="o">=</span> <span class="n">sys</span><span class="o">.</span><span class="n">maxsize</span>
+            <span class="n">num_update_steps_per_epoch</span> <span class="o">=</span> <span class="n">max_steps</span>
+            <span class="n">num_examples</span> <span class="o">=</span> <span class="n">total_train_batch_size</span> <span class="o">*</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span>
+            <span class="n">num_train_samples</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">*</span> <span class="n">total_train_batch_size</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;args.max_steps must be set to a positive value if dataloader does not have a length, was&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">args</span><span class="o">.</span><span class="n">max_steps</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+        <span class="c1">###########</span>
+        <span class="c1">#num_train_epochs = 5</span>
+
+        <span class="c1"># Train!</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="s2">&quot;***** Running training *****&quot;</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;  Num examples = </span><span class="si">{</span><span class="n">num_examples</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;  Num Epochs = </span><span class="si">{</span><span class="n">num_train_epochs</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;  Instantaneous batch size per device = </span><span class="si">{</span><span class="n">args</span><span class="o">.</span><span class="n">per_device_train_batch_size</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;  Total train batch size (w. parallel, distributed &amp; accumulation) = </span><span class="si">{</span><span class="n">total_train_batch_size</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;  Gradient Accumulation steps = </span><span class="si">{</span><span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;  Total optimization steps = </span><span class="si">{</span><span class="n">max_steps</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;  Number of trainable parameters = </span><span class="si">{</span><span class="nb">sum</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">numel</span><span class="p">()</span><span class="w"> </span><span class="k">for</span><span class="w"> </span><span class="n">p</span><span class="w"> </span><span class="ow">in</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">tmp_model</span><span class="o">.</span><span class="n">parameters</span><span class="p">()</span><span class="w"> </span><span class="k">if</span><span class="w"> </span><span class="n">p</span><span class="o">.</span><span class="n">requires_grad</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">epoch</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="n">start_time</span> <span class="o">=</span> <span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span>
+        <span class="n">epochs_trained</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="n">steps_trained_in_current_epoch</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="n">steps_trained_progress_bar</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="c1"># Update the references</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">optimizer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">lr_scheduler</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">train_dataloader</span> <span class="o">=</span> <span class="n">train_dataloader</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_name</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">_trial</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="c1"># use self._trial because the SigOpt/Optuna hpo only call `_hp_search_setup(trial)` instead of passing trial</span>
+            <span class="c1"># parameter to Train when using DDP.</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">trial_name</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_name</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_trial</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">trial</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">assignments</span> <span class="o">=</span> <span class="n">trial</span><span class="o">.</span><span class="n">assignments</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">SIGOPT</span> <span class="k">else</span> <span class="n">trial</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">trial_params</span> <span class="o">=</span> <span class="n">hp_params</span><span class="p">(</span><span class="n">assignments</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">trial_params</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="c1"># This should be the same if the state has been saved but in case the training arguments changed, it&#39;s safer</span>
+        <span class="c1"># to set this after the load.</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">=</span> <span class="n">max_steps</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">num_train_epochs</span> <span class="o">=</span> <span class="n">num_train_epochs</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">is_local_process_zero</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_local_process_zero</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">is_world_process_zero</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_world_process_zero</span><span class="p">()</span>
+
+        <span class="c1"># tr_loss is a tensor to avoid synchronization of TPUs through .item()</span>
+        <span class="n">tr_loss</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="mf">0.0</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+        <span class="c1"># _total_loss_scalar is updated everytime .item() has to be called on tr_loss and stores the sum of all losses</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_total_loss_scalar</span> <span class="o">=</span> <span class="mf">0.0</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_globalstep_last_logged</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span>
+        <span class="c1">#model.zero_grad()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">tmp_model</span><span class="o">.</span><span class="n">zero_grad</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_train_begin</span><span class="p">(</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">)</span>
+
+        <span class="c1"># Skip the first epochs_trained epochs to get the random state of the dataloader at the right point.</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">args</span><span class="o">.</span><span class="n">ignore_data_skip</span><span class="p">:</span>
+            <span class="c1">#print(&quot;I skip!&quot;) called</span>
+            <span class="k">for</span> <span class="n">epoch</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">epochs_trained</span><span class="p">):</span>
+                <span class="n">is_random_sampler</span> <span class="o">=</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">train_dataloader</span><span class="p">,</span> <span class="s2">&quot;sampler&quot;</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span>
+                    <span class="n">train_dataloader</span><span class="o">.</span><span class="n">sampler</span><span class="p">,</span> <span class="n">RandomSampler</span>
+                <span class="p">)</span>
+                <span class="k">if</span> <span class="n">is_torch_less_than_1_11</span> <span class="ow">or</span> <span class="ow">not</span> <span class="n">is_random_sampler</span><span class="p">:</span>
+                    <span class="c1"># We just need to begin an iteration to create the randomization of the sampler.</span>
+                    <span class="c1"># That was before PyTorch 1.11 however...</span>
+                    <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="n">train_dataloader</span><span class="p">:</span>
+                        <span class="k">break</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="c1"># Otherwise we need to call the whooooole sampler cause there is some random operation added</span>
+                    <span class="c1"># AT THE VERY END!</span>
+                    <span class="n">_</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">train_dataloader</span><span class="o">.</span><span class="n">sampler</span><span class="p">)</span>
+
+        <span class="c1">###############</span>
+        <span class="c1">#num_train_epochs = 10</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">is_in_train</span> <span class="o">=</span> <span class="kc">True</span>
+        <span class="c1">#print(&quot;The number of epoches: &quot;, num_train_epochs)</span>
+        <span class="c1">############# </span>
+        <span class="n">total_batched_samples</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">for</span> <span class="n">epoch</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">epochs_trained</span><span class="p">,</span> <span class="n">num_train_epochs</span><span class="p">):</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">train_dataloader</span><span class="p">,</span> <span class="n">DataLoader</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">train_dataloader</span><span class="o">.</span><span class="n">sampler</span><span class="p">,</span> <span class="n">DistributedSampler</span><span class="p">):</span>
+                <span class="n">train_dataloader</span><span class="o">.</span><span class="n">sampler</span><span class="o">.</span><span class="n">set_epoch</span><span class="p">(</span><span class="n">epoch</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">train_dataloader</span><span class="p">,</span> <span class="s2">&quot;dataset&quot;</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">train_dataloader</span><span class="o">.</span><span class="n">dataset</span><span class="p">,</span> <span class="n">IterableDatasetShard</span><span class="p">):</span>
+                <span class="n">train_dataloader</span><span class="o">.</span><span class="n">dataset</span><span class="o">.</span><span class="n">set_epoch</span><span class="p">(</span><span class="n">epoch</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+                <span class="n">parallel_loader</span> <span class="o">=</span> <span class="n">pl</span><span class="o">.</span><span class="n">ParallelLoader</span><span class="p">(</span><span class="n">train_dataloader</span><span class="p">,</span> <span class="p">[</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">])</span><span class="o">.</span><span class="n">per_device_loader</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+                <span class="n">epoch_iterator</span> <span class="o">=</span> <span class="n">parallel_loader</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">epoch_iterator</span> <span class="o">=</span> <span class="n">train_dataloader</span>
+
+            <span class="c1"># Reset the past mems state at the beginning of each epoch if necessary.</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">past_index</span> <span class="o">&gt;=</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_past</span> <span class="o">=</span> <span class="kc">None</span>
+
+            <span class="n">steps_in_epoch</span> <span class="o">=</span> <span class="p">(</span>
+                <span class="nb">len</span><span class="p">(</span><span class="n">epoch_iterator</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">len_dataloader</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+                <span class="k">else</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">*</span> <span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_epoch_begin</span><span class="p">(</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">epoch</span> <span class="o">==</span> <span class="n">epochs_trained</span> <span class="ow">and</span> <span class="n">resume_from_checkpoint</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">steps_trained_in_current_epoch</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_load_rng_state</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="p">)</span>
+
+            <span class="n">rng_to_sync</span> <span class="o">=</span> <span class="kc">False</span>
+            <span class="n">steps_skipped</span> <span class="o">=</span> <span class="mi">0</span>
+            <span class="k">if</span> <span class="n">skip_first_batches</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">steps_trained_in_current_epoch</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">epoch_iterator</span> <span class="o">=</span> <span class="n">skip_first_batches</span><span class="p">(</span><span class="n">epoch_iterator</span><span class="p">,</span> <span class="n">steps_trained_in_current_epoch</span><span class="p">)</span>
+                <span class="n">steps_skipped</span> <span class="o">=</span> <span class="n">steps_trained_in_current_epoch</span>
+                <span class="n">steps_trained_in_current_epoch</span> <span class="o">=</span> <span class="mi">0</span>
+                <span class="n">rng_to_sync</span> <span class="o">=</span> <span class="kc">True</span>
+
+            <span class="c1">#print(&quot;The number of one epoch: &quot;, len(epoch_iterator))</span>
+            <span class="n">step</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span>
+            <span class="k">for</span> <span class="n">step</span><span class="p">,</span> <span class="n">inputs</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">epoch_iterator</span><span class="p">):</span>
+                <span class="n">total_batched_samples</span> <span class="o">+=</span> <span class="mi">1</span>
+                <span class="k">if</span> <span class="n">rng_to_sync</span><span class="p">:</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">_load_rng_state</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="p">)</span>
+                    <span class="n">rng_to_sync</span> <span class="o">=</span> <span class="kc">False</span>
+
+                <span class="c1"># Skip past any already trained steps if resuming training</span>
+                <span class="k">if</span> <span class="n">steps_trained_in_current_epoch</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">steps_trained_in_current_epoch</span> <span class="o">-=</span> <span class="mi">1</span>
+                    <span class="k">if</span> <span class="n">steps_trained_progress_bar</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                        <span class="n">steps_trained_progress_bar</span><span class="o">.</span><span class="n">update</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">steps_trained_in_current_epoch</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">_load_rng_state</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="p">)</span>
+                    <span class="k">continue</span>
+                <span class="k">elif</span> <span class="n">steps_trained_progress_bar</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">steps_trained_progress_bar</span><span class="o">.</span><span class="n">close</span><span class="p">()</span>
+                    <span class="n">steps_trained_progress_bar</span> <span class="o">=</span> <span class="kc">None</span>
+
+                <span class="k">if</span> <span class="n">step</span> <span class="o">%</span> <span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_step_begin</span><span class="p">(</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">)</span>
+
+                <span class="k">if</span> <span class="p">(</span>
+                    <span class="p">(</span><span class="n">total_batched_samples</span> <span class="o">%</span> <span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">)</span>
+                    <span class="ow">and</span> <span class="n">args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">!=</span> <span class="o">-</span><span class="mi">1</span>
+                    <span class="ow">and</span> <span class="n">args</span><span class="o">.</span><span class="n">_no_sync_in_gradient_accumulation</span>
+                <span class="p">):</span>
+                    <span class="c1"># Avoid unnecessary DDP synchronization since there will be no backward pass on this example.</span>
+                    <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">tmp_model</span><span class="o">.</span><span class="n">no_sync</span><span class="p">():</span>
+                        <span class="n">tr_loss_step</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">training_step</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">tmp_model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">)</span>
+                    <span class="c1">#with model.no_sync():</span>
+                        <span class="c1">#tr_loss_step = self.training_step(model, inputs)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">tr_loss_step</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">training_step</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">tmp_model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">)</span>
+
+                <span class="k">if</span> <span class="p">(</span>
+                    <span class="n">args</span><span class="o">.</span><span class="n">logging_nan_inf_filter</span>
+                    <span class="ow">and</span> <span class="ow">not</span> <span class="n">is_torch_tpu_available</span><span class="p">()</span>
+                    <span class="ow">and</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">isnan</span><span class="p">(</span><span class="n">tr_loss_step</span><span class="p">)</span> <span class="ow">or</span> <span class="n">torch</span><span class="o">.</span><span class="n">isinf</span><span class="p">(</span><span class="n">tr_loss_step</span><span class="p">))</span>
+                <span class="p">):</span>
+                    <span class="c1"># if loss is nan or inf simply add the average of previous logged losses</span>
+                    <span class="n">tr_loss</span> <span class="o">+=</span> <span class="n">tr_loss</span> <span class="o">/</span> <span class="p">(</span><span class="mi">1</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">_globalstep_last_logged</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">tr_loss</span> <span class="o">+=</span> <span class="n">tr_loss_step</span>
+
+                <span class="bp">self</span><span class="o">.</span><span class="n">current_flos</span> <span class="o">+=</span> <span class="nb">float</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">floating_point_ops</span><span class="p">(</span><span class="n">inputs</span><span class="p">))</span>
+
+                <span class="c1"># Optimizer step for deepspeed must be called on every step regardless of the value of gradient_accumulation_steps</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="o">.</span><span class="n">step</span><span class="p">()</span>
+
+                <span class="k">if</span> <span class="n">total_batched_samples</span> <span class="o">%</span> <span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">or</span> <span class="p">(</span>
+                    <span class="c1"># last step in epoch but step is always smaller than gradient_accumulation_steps</span>
+                    <span class="n">steps_in_epoch</span> <span class="o">&lt;=</span> <span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span>
+                    <span class="ow">and</span> <span class="p">(</span><span class="n">step</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">==</span> <span class="n">steps_in_epoch</span>
+                <span class="p">):</span>
+                    <span class="c1"># Gradient clipping</span>
+                    <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">max_grad_norm</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">args</span><span class="o">.</span><span class="n">max_grad_norm</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+                        <span class="c1"># deepspeed does its own clipping</span>
+
+                        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">do_grad_scaling</span><span class="p">:</span>
+                            <span class="c1"># Reduce gradients first for XLA</span>
+                            <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+                                <span class="n">gradients</span> <span class="o">=</span> <span class="n">xm</span><span class="o">.</span><span class="n">_fetch_gradients</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">)</span>
+                                <span class="n">xm</span><span class="o">.</span><span class="n">all_reduce</span><span class="p">(</span><span class="s2">&quot;sum&quot;</span><span class="p">,</span> <span class="n">gradients</span><span class="p">,</span> <span class="n">scale</span><span class="o">=</span><span class="mf">1.0</span> <span class="o">/</span> <span class="n">xm</span><span class="o">.</span><span class="n">xrt_world_size</span><span class="p">())</span>
+                            <span class="c1"># AMP: gradients need unscaling</span>
+                            <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span><span class="o">.</span><span class="n">unscale_</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">)</span>
+
+                        <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">()</span> <span class="ow">and</span> <span class="n">args</span><span class="o">.</span><span class="n">fp16</span><span class="p">:</span>
+                            <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">clip_master_grads</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">max_grad_norm</span><span class="p">)</span>
+                        <span class="k">elif</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">,</span> <span class="s2">&quot;clip_grad_norm&quot;</span><span class="p">):</span>
+                            <span class="c1"># Some optimizers (like the sharded optimizer) have a specific way to do gradient clipping</span>
+                            <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">clip_grad_norm</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">max_grad_norm</span><span class="p">)</span>
+                        <span class="k">elif</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="s2">&quot;clip_grad_norm_&quot;</span><span class="p">):</span>
+                            <span class="c1"># Some models (like FullyShardedDDP) have a specific way to do gradient clipping</span>
+                            <span class="n">model</span><span class="o">.</span><span class="n">clip_grad_norm_</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">max_grad_norm</span><span class="p">)</span>
+                        <span class="k">else</span><span class="p">:</span>
+                            <span class="c1"># Revert to normal clipping otherwise, handling Apex or full precision</span>
+                            <span class="n">nn</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">clip_grad_norm_</span><span class="p">(</span>
+                                <span class="n">amp</span><span class="o">.</span><span class="n">master_params</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">)</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_apex</span> <span class="k">else</span> <span class="n">model</span><span class="o">.</span><span class="n">parameters</span><span class="p">(),</span>
+                                <span class="n">args</span><span class="o">.</span><span class="n">max_grad_norm</span><span class="p">,</span>
+                            <span class="p">)</span>
+
+                    <span class="c1"># Optimizer step</span>
+                    <span class="n">optimizer_was_run</span> <span class="o">=</span> <span class="kc">True</span>
+                    <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+                        <span class="k">pass</span>  <span class="c1"># called outside the loop</span>
+                    <span class="k">elif</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+                        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">do_grad_scaling</span><span class="p">:</span>
+                            <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span><span class="o">.</span><span class="n">step</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">)</span>
+                            <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span><span class="o">.</span><span class="n">update</span><span class="p">()</span>
+                        <span class="k">else</span><span class="p">:</span>
+                            <span class="n">xm</span><span class="o">.</span><span class="n">optimizer_step</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">)</span>
+                    <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">do_grad_scaling</span><span class="p">:</span>
+                        <span class="n">scale_before</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span><span class="o">.</span><span class="n">get_scale</span><span class="p">()</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span><span class="o">.</span><span class="n">step</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">)</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span><span class="o">.</span><span class="n">update</span><span class="p">()</span>
+                        <span class="n">scale_after</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span><span class="o">.</span><span class="n">get_scale</span><span class="p">()</span>
+                        <span class="n">optimizer_was_run</span> <span class="o">=</span> <span class="n">scale_before</span> <span class="o">&lt;=</span> <span class="n">scale_after</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">step</span><span class="p">()</span>
+
+                    <span class="k">if</span> <span class="n">optimizer_was_run</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span><span class="o">.</span><span class="n">step</span><span class="p">()</span>
+
+                    <span class="bp">self</span><span class="o">.</span><span class="n">tmp_model</span><span class="o">.</span><span class="n">zero_grad</span><span class="p">()</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span> <span class="o">+=</span> <span class="mi">1</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">epoch</span> <span class="o">=</span> <span class="n">epoch</span> <span class="o">+</span> <span class="p">(</span><span class="n">step</span> <span class="o">+</span> <span class="mi">1</span> <span class="o">+</span> <span class="n">steps_skipped</span><span class="p">)</span> <span class="o">/</span> <span class="n">steps_in_epoch</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_step_end</span><span class="p">(</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">)</span>
+
+                    <span class="bp">self</span><span class="o">.</span><span class="n">_maybe_log_save_evaluate</span><span class="p">(</span><span class="n">tr_loss</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">tmp_model</span><span class="p">,</span> <span class="n">trial</span><span class="p">,</span> <span class="n">epoch</span><span class="p">,</span> <span class="n">ignore_keys_for_eval</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_substep_end</span><span class="p">(</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">)</span>
+
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="o">.</span><span class="n">should_epoch_stop</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="o">.</span><span class="n">should_training_stop</span><span class="p">:</span>
+                    <span class="k">break</span>
+            <span class="k">if</span> <span class="n">step</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="s2">&quot;There seems to be not a single sample in your epoch_iterator, stopping training at step&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span><span class="si">}</span><span class="s2">! This is expected if you&#39;re using an IterableDataset and set&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot; num_steps (</span><span class="si">{</span><span class="n">max_steps</span><span class="si">}</span><span class="s2">) higher than the number of available samples.&quot;</span>
+                <span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="o">.</span><span class="n">should_training_stop</span> <span class="o">=</span> <span class="kc">True</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_epoch_end</span><span class="p">(</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_maybe_log_save_evaluate</span><span class="p">(</span><span class="n">tr_loss</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">tmp_model</span><span class="p">,</span> <span class="n">trial</span><span class="p">,</span> <span class="n">epoch</span><span class="p">,</span> <span class="n">ignore_keys_for_eval</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">DebugOption</span><span class="o">.</span><span class="n">TPU_METRICS_DEBUG</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">debug</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+                    <span class="c1"># tpu-comment: Logging debug metrics for PyTorch/XLA (compile, execute times, ops, etc.)</span>
+                    <span class="n">xm</span><span class="o">.</span><span class="n">master_print</span><span class="p">(</span><span class="n">met</span><span class="o">.</span><span class="n">metrics_report</span><span class="p">())</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                        <span class="s2">&quot;You enabled PyTorch/XLA debug metrics but you don&#39;t have a TPU &quot;</span>
+                        <span class="s2">&quot;configured. Check your training configuration if this is unexpected.&quot;</span>
+                    <span class="p">)</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="o">.</span><span class="n">should_training_stop</span><span class="p">:</span>
+                <span class="k">break</span>
+
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">past_index</span> <span class="ow">and</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;_past&quot;</span><span class="p">):</span>
+            <span class="c1"># Clean the state at the end of training</span>
+            <span class="nb">delattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;_past&quot;</span><span class="p">)</span>
+
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n\n</span><span class="s2">Training completed. Do not forget to share your model on huggingface.co/models =)</span><span class="se">\n\n</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">load_best_model_at_end</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="c1"># Wait for everyone to get here so we are sur the model has been saved by process 0.</span>
+            <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+                <span class="n">xm</span><span class="o">.</span><span class="n">rendezvous</span><span class="p">(</span><span class="s2">&quot;load_best_model_at_end&quot;</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">!=</span> <span class="o">-</span><span class="mi">1</span><span class="p">:</span>
+                <span class="n">dist</span><span class="o">.</span><span class="n">barrier</span><span class="p">()</span>
+            <span class="k">elif</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+                <span class="n">smp</span><span class="o">.</span><span class="n">barrier</span><span class="p">()</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">_load_best_model</span><span class="p">()</span>
+
+        <span class="c1"># add remaining tr_loss</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_total_loss_scalar</span> <span class="o">+=</span> <span class="n">tr_loss</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+        <span class="n">train_loss</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_total_loss_scalar</span> <span class="o">/</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span>
+
+        <span class="n">metrics</span> <span class="o">=</span> <span class="n">speed_metrics</span><span class="p">(</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="n">start_time</span><span class="p">,</span> <span class="n">num_samples</span><span class="o">=</span><span class="n">num_train_samples</span><span class="p">,</span> <span class="n">num_steps</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">max_steps</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">store_flos</span><span class="p">()</span>
+        <span class="n">metrics</span><span class="p">[</span><span class="s2">&quot;total_flos&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">total_flos</span>
+        <span class="n">metrics</span><span class="p">[</span><span class="s2">&quot;train_loss&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">train_loss</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">is_in_train</span> <span class="o">=</span> <span class="kc">False</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_memory_tracker</span><span class="o">.</span><span class="n">stop_and_update_metrics</span><span class="p">(</span><span class="n">metrics</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="n">metrics</span><span class="p">)</span>
+
+        <span class="n">run_dir</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_output_dir</span><span class="p">(</span><span class="n">trial</span><span class="p">)</span>
+        <span class="n">checkpoints_sorted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_sorted_checkpoints</span><span class="p">(</span><span class="n">use_mtime</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">output_dir</span><span class="o">=</span><span class="n">run_dir</span><span class="p">)</span>
+
+        <span class="c1"># Delete the last checkpoint when save_total_limit=1 if it&#39;s different from the best checkpoint and process allowed to save.</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">save_total_limit</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">checkpoint</span> <span class="ow">in</span> <span class="n">checkpoints_sorted</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">checkpoint</span> <span class="o">!=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span><span class="p">:</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Deleting older checkpoint [</span><span class="si">{</span><span class="n">checkpoint</span><span class="si">}</span><span class="s2">] due to args.save_total_limit&quot;</span><span class="p">)</span>
+                    <span class="n">shutil</span><span class="o">.</span><span class="n">rmtree</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_train_end</span><span class="p">(</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">TrainOutput</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span><span class="p">,</span> <span class="n">train_loss</span><span class="p">,</span> <span class="n">metrics</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._inner_training_loop">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._inner_training_loop">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_inner_training_loop</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">batch_size</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">args</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">resume_from_checkpoint</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">trial</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">ignore_keys_for_eval</span><span class="o">=</span><span class="kc">None</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&#39;&#39;&#39;</span>
+<span class="sd">        0 This function serves to train one time</span>
+<span class="sd">        1 Update the self.train_dataset before calling this function</span>
+<span class="sd">        &#39;&#39;&#39;</span>
+        <span class="c1"># 1 Get dataloader</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_train_batch_size</span> <span class="o">=</span> <span class="n">batch_size</span>
+        <span class="c1"># Data loader and number of training steps</span>
+        <span class="n">train_dataloader</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_train_dataloader</span><span class="p">()</span>
+        <span class="n">total_train_batch_size</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">train_batch_size</span> <span class="o">*</span> <span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span> <span class="o">*</span> <span class="n">args</span><span class="o">.</span><span class="n">world_size</span>
+
+        <span class="n">len_dataloader</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="n">has_length</span><span class="p">(</span><span class="n">train_dataloader</span><span class="p">):</span>
+            <span class="n">len_dataloader</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">train_dataloader</span><span class="p">)</span>
+            <span class="n">num_update_steps_per_epoch</span> <span class="o">=</span> <span class="n">len_dataloader</span> <span class="o">//</span> <span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span>
+            <span class="n">num_update_steps_per_epoch</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">num_update_steps_per_epoch</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+            <span class="n">num_examples</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_examples</span><span class="p">(</span><span class="n">train_dataloader</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">max_steps</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span>
+                <span class="n">num_train_epochs</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">//</span> <span class="n">num_update_steps_per_epoch</span> <span class="o">+</span> <span class="nb">int</span><span class="p">(</span>
+                    <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">%</span> <span class="n">num_update_steps_per_epoch</span> <span class="o">&gt;</span> <span class="mi">0</span>
+                <span class="p">)</span>
+                <span class="c1"># May be slightly incorrect if the last batch in the training dataloader has a smaller size but it&#39;s</span>
+                <span class="c1"># the best we can do.</span>
+                <span class="n">num_train_samples</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">*</span> <span class="n">total_train_batch_size</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">max_steps</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">num_train_epochs</span> <span class="o">*</span> <span class="n">num_update_steps_per_epoch</span><span class="p">)</span>
+                <span class="n">num_train_epochs</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">num_train_epochs</span><span class="p">)</span>
+                <span class="n">num_train_samples</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_examples</span><span class="p">(</span><span class="n">train_dataloader</span><span class="p">)</span> <span class="o">*</span> <span class="n">args</span><span class="o">.</span><span class="n">num_train_epochs</span>
+        <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>  <span class="c1"># Rely on max_steps when dataloader does not have a working size</span>
+            <span class="n">max_steps</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span>
+            <span class="c1"># Setting a very large number of epochs so we go as many times as necessary over the iterator.</span>
+            <span class="n">num_train_epochs</span> <span class="o">=</span> <span class="n">sys</span><span class="o">.</span><span class="n">maxsize</span>
+            <span class="n">num_update_steps_per_epoch</span> <span class="o">=</span> <span class="n">max_steps</span>
+            <span class="n">num_examples</span> <span class="o">=</span> <span class="n">total_train_batch_size</span> <span class="o">*</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span>
+            <span class="n">num_train_samples</span> <span class="o">=</span> <span class="n">args</span><span class="o">.</span><span class="n">max_steps</span> <span class="o">*</span> <span class="n">total_train_batch_size</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;args.max_steps must be set to a positive value if dataloader does not have a length, was&quot;</span>
+                <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">args</span><span class="o">.</span><span class="n">max_steps</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">DebugOption</span><span class="o">.</span><span class="n">UNDERFLOW_OVERFLOW</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">debug</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">n_gpu</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+                <span class="c1"># nn.DataParallel(model) replicates the model, creating new variables and module</span>
+                <span class="c1"># references registered here no longer work on other gpus, breaking the module</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="s2">&quot;Currently --debug underflow_overflow is not supported under DP. Please use DDP&quot;</span>
+                    <span class="s2">&quot; (torch.distributed.launch).&quot;</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">debug_overflow</span> <span class="o">=</span> <span class="n">DebugUnderflowOverflow</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">)</span>  <span class="c1"># noqa</span>
+
+        <span class="n">delay_optimizer_creation</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+            <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="o">!=</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">SIMPLE</span>
+            <span class="ow">or</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">()</span>
+            <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">fsdp</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+            <span class="n">deepspeed_engine</span><span class="p">,</span> <span class="n">optimizer</span><span class="p">,</span> <span class="n">lr_scheduler</span> <span class="o">=</span> <span class="n">deepspeed_init</span><span class="p">(</span>
+                <span class="bp">self</span><span class="p">,</span> <span class="n">num_training_steps</span><span class="o">=</span><span class="n">max_steps</span><span class="p">,</span> <span class="n">resume_from_checkpoint</span><span class="o">=</span><span class="n">resume_from_checkpoint</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">deepspeed_engine</span><span class="o">.</span><span class="n">module</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span> <span class="o">=</span> <span class="n">deepspeed_engine</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span> <span class="o">=</span> <span class="n">deepspeed_engine</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">optimizer</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span> <span class="o">=</span> <span class="n">lr_scheduler</span>
+            <span class="c1">#print(&quot;I just create a optimizer here!&quot;)  # called</span>
+        <span class="k">elif</span> <span class="ow">not</span> <span class="n">delay_optimizer_creation</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">create_optimizer_and_scheduler</span><span class="p">(</span><span class="n">num_training_steps</span><span class="o">=</span><span class="n">max_steps</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">state</span> <span class="o">=</span> <span class="n">TrainerState</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">is_hyper_param_search</span> <span class="o">=</span> <span class="n">trial</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+
+        <span class="c1"># Activate gradient checkpointing if needed</span>
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">gradient_checkpointing</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">gradient_checkpointing_enable</span><span class="p">()</span>
+
+        <span class="c1">#model = self._wrap_model(self.model_wrapped)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">tmp_model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_wrap_model</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span><span class="p">)</span>
+
+
+        <span class="c1">#if is_sagemaker_mp_enabled() and resume_from_checkpoint is not None:</span>
+        <span class="c1">#    self._load_from_checkpoint(resume_from_checkpoint, model)</span>
+
+        <span class="c1"># for the rest of this function `model` is the outside model, whether it was wrapped or not</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">tmp_model</span> <span class="ow">is</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tmp_model</span>
+
+        <span class="k">if</span> <span class="n">delay_optimizer_creation</span><span class="p">:</span>
+            <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;I create here!&quot;</span><span class="p">)</span>  <span class="c1"># not called</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">create_optimizer_and_scheduler</span><span class="p">(</span><span class="n">num_training_steps</span><span class="o">=</span><span class="n">max_steps</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="kc">True</span></div>
+
+        <span class="c1"># Check if saved optimizer or scheduler states exist</span>
+        <span class="c1">#self._load_optimizer_and_scheduler(resume_from_checkpoint)</span>
+
+        <span class="c1"># important: at this point:</span>
+        <span class="c1"># self.model         is the Transformers Model</span>
+        <span class="c1"># self.model_wrapped is DDP(Transformers Model), Deepspeed(Transformers Model), etc.</span>
+
+        
+
+<div class="viewcode-block" id="RaftTrainer._get_output_dir">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_output_dir">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_get_output_dir</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">trial</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">trial</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">OPTUNA</span><span class="p">:</span>
+                <span class="n">run_id</span> <span class="o">=</span> <span class="n">trial</span><span class="o">.</span><span class="n">number</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">RAY</span><span class="p">:</span>
+                <span class="kn">from</span><span class="w"> </span><span class="nn">ray</span><span class="w"> </span><span class="kn">import</span> <span class="n">tune</span>
+
+                <span class="n">run_id</span> <span class="o">=</span> <span class="n">tune</span><span class="o">.</span><span class="n">get_trial_id</span><span class="p">()</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">SIGOPT</span><span class="p">:</span>
+                <span class="n">run_id</span> <span class="o">=</span> <span class="n">trial</span><span class="o">.</span><span class="n">id</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">WANDB</span><span class="p">:</span>
+                <span class="kn">import</span><span class="w"> </span><span class="nn">wandb</span>
+
+                <span class="n">run_id</span> <span class="o">=</span> <span class="n">wandb</span><span class="o">.</span><span class="n">run</span><span class="o">.</span><span class="n">id</span>
+            <span class="n">run_name</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_name</span><span class="p">(</span><span class="n">trial</span><span class="p">)</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_name</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="sa">f</span><span class="s2">&quot;run-</span><span class="si">{</span><span class="n">run_id</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="n">run_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">run_name</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">run_dir</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span>
+        <span class="k">return</span> <span class="n">run_dir</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._load_from_checkpoint">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_from_checkpoint">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_load_from_checkpoint</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">resume_from_checkpoint</span><span class="p">,</span> <span class="n">model</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="k">if</span> <span class="n">model</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="p">,</span> <span class="n">WEIGHTS_NAME</span><span class="p">))</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span>
+            <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="p">,</span> <span class="n">WEIGHTS_INDEX_NAME</span><span class="p">)</span>
+        <span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Can&#39;t find a valid checkpoint at </span><span class="si">{</span><span class="n">resume_from_checkpoint</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Loading model from </span><span class="si">{</span><span class="n">resume_from_checkpoint</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="p">,</span> <span class="n">CONFIG_NAME</span><span class="p">)):</span>
+            <span class="n">config</span> <span class="o">=</span> <span class="n">PretrainedConfig</span><span class="o">.</span><span class="n">from_json_file</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="p">,</span> <span class="n">CONFIG_NAME</span><span class="p">))</span>
+            <span class="n">checkpoint_version</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">transformers_version</span>
+            <span class="k">if</span> <span class="n">checkpoint_version</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">checkpoint_version</span> <span class="o">!=</span> <span class="n">__version__</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;You are resuming training from a checkpoint trained with </span><span class="si">{</span><span class="n">checkpoint_version</span><span class="si">}</span><span class="s2"> of &quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;Transformers but your current version is </span><span class="si">{</span><span class="n">__version__</span><span class="si">}</span><span class="s2">. This is not recommended and could &quot;</span>
+                    <span class="s2">&quot;yield to errors or unwanted behaviors.&quot;</span>
+                <span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="p">,</span> <span class="n">WEIGHTS_NAME</span><span class="p">)):</span>
+            <span class="c1"># If the model is on the GPU, it still works!</span>
+            <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+                <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="p">,</span> <span class="s2">&quot;user_content.pt&quot;</span><span class="p">)):</span>
+                    <span class="c1"># If the &#39;user_content.pt&#39; file exists, load with the new smp api.</span>
+                    <span class="c1"># Checkpoint must have been saved with the new smp api.</span>
+                    <span class="n">smp</span><span class="o">.</span><span class="n">resume_from_checkpoint</span><span class="p">(</span>
+                        <span class="n">path</span><span class="o">=</span><span class="n">resume_from_checkpoint</span><span class="p">,</span> <span class="n">tag</span><span class="o">=</span><span class="n">WEIGHTS_NAME</span><span class="p">,</span> <span class="n">partial</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">load_optimizer</span><span class="o">=</span><span class="kc">False</span>
+                    <span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="c1"># If the &#39;user_content.pt&#39; file does NOT exist, load with the old smp api.</span>
+                    <span class="c1"># Checkpoint must have been saved with the old smp api.</span>
+                    <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="s2">&quot;fp16&quot;</span><span class="p">)</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">fp16</span> <span class="ow">is</span> <span class="kc">True</span><span class="p">:</span>
+                        <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                            <span class="s2">&quot;Enabling FP16 and loading from smp &lt; 1.10 checkpoint together is not suppported.&quot;</span>
+                        <span class="p">)</span>
+                    <span class="n">state_dict</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="p">,</span> <span class="n">WEIGHTS_NAME</span><span class="p">),</span> <span class="n">map_location</span><span class="o">=</span><span class="s2">&quot;cpu&quot;</span><span class="p">)</span>
+                    <span class="c1"># Required for smp to not auto-translate state_dict from hf to smp (is already smp).</span>
+                    <span class="n">state_dict</span><span class="p">[</span><span class="s2">&quot;_smp_is_partial&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="kc">False</span>
+                    <span class="n">load_result</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">state_dict</span><span class="p">,</span> <span class="n">strict</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                    <span class="c1"># release memory</span>
+                    <span class="k">del</span> <span class="n">state_dict</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="c1"># We load the model state dict on the CPU to avoid an OOM error.</span>
+                <span class="n">state_dict</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">resume_from_checkpoint</span><span class="p">,</span> <span class="n">WEIGHTS_NAME</span><span class="p">),</span> <span class="n">map_location</span><span class="o">=</span><span class="s2">&quot;cpu&quot;</span><span class="p">)</span>
+                <span class="c1"># workaround for FSDP bug https://github.com/pytorch/pytorch/issues/82963</span>
+                <span class="c1"># which takes *args instead of **kwargs</span>
+                <span class="n">load_result</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">state_dict</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+                <span class="c1"># release memory</span>
+                <span class="k">del</span> <span class="n">state_dict</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_issue_warnings_after_load</span><span class="p">(</span><span class="n">load_result</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># We load the sharded checkpoint</span>
+            <span class="n">load_result</span> <span class="o">=</span> <span class="n">load_sharded_checkpoint</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">resume_from_checkpoint</span><span class="p">,</span> <span class="n">strict</span><span class="o">=</span><span class="n">is_sagemaker_mp_enabled</span><span class="p">())</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_issue_warnings_after_load</span><span class="p">(</span><span class="n">load_result</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._load_best_model">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_best_model">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_load_best_model</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Loading best model from </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span><span class="si">}</span><span class="s2"> (score: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_metric</span><span class="si">}</span><span class="s2">).&quot;</span><span class="p">)</span>
+        <span class="n">best_model_path</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span><span class="p">,</span> <span class="n">WEIGHTS_NAME</span><span class="p">)</span>
+        <span class="n">model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span> <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">()</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span>
+        <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">best_model_path</span><span class="p">):</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="c1"># this removes the pre-hooks from the previous engine</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span><span class="o">.</span><span class="n">destroy</span><span class="p">()</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span> <span class="o">=</span> <span class="kc">None</span>
+
+                <span class="c1"># temp hack until Deepspeed fixes the problem with resume from an existing engine that did some stepping</span>
+                <span class="n">deepspeed_engine</span><span class="p">,</span> <span class="n">optimizer</span><span class="p">,</span> <span class="n">lr_scheduler</span> <span class="o">=</span> <span class="n">deepspeed_init</span><span class="p">(</span>
+                    <span class="bp">self</span><span class="p">,</span>
+                    <span class="n">num_training_steps</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">max_steps</span><span class="p">,</span>
+                    <span class="n">resume_from_checkpoint</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span><span class="p">,</span>
+                <span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">deepspeed_engine</span><span class="o">.</span><span class="n">module</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span> <span class="o">=</span> <span class="n">deepspeed_engine</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span> <span class="o">=</span> <span class="n">deepspeed_engine</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">optimizer</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span> <span class="o">=</span> <span class="n">lr_scheduler</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+                    <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span><span class="p">,</span> <span class="s2">&quot;user_content.pt&quot;</span><span class="p">)):</span>
+                        <span class="c1"># If the &#39;user_content.pt&#39; file exists, load with the new smp api.</span>
+                        <span class="c1"># Checkpoint must have been saved with the new smp api.</span>
+                        <span class="n">smp</span><span class="o">.</span><span class="n">resume_from_checkpoint</span><span class="p">(</span>
+                            <span class="n">path</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span><span class="p">,</span>
+                            <span class="n">tag</span><span class="o">=</span><span class="n">WEIGHTS_NAME</span><span class="p">,</span>
+                            <span class="n">partial</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                            <span class="n">load_optimizer</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                        <span class="p">)</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="c1"># If the &#39;user_content.pt&#39; file does NOT exist, load with the old smp api.</span>
+                        <span class="c1"># Checkpoint must have been saved with the old smp api.</span>
+                        <span class="n">state_dict</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">best_model_path</span><span class="p">,</span> <span class="n">map_location</span><span class="o">=</span><span class="s2">&quot;cpu&quot;</span><span class="p">)</span>
+                        <span class="n">state_dict</span><span class="p">[</span><span class="s2">&quot;_smp_is_partial&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="kc">False</span>
+                        <span class="n">load_result</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">state_dict</span><span class="p">,</span> <span class="n">strict</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="c1"># We load the model state dict on the CPU to avoid an OOM error.</span>
+                    <span class="n">state_dict</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">best_model_path</span><span class="p">,</span> <span class="n">map_location</span><span class="o">=</span><span class="s2">&quot;cpu&quot;</span><span class="p">)</span>
+                    <span class="c1"># If the model is on the GPU, it still works!</span>
+                    <span class="c1"># workaround for FSDP bug https://github.com/pytorch/pytorch/issues/82963</span>
+                    <span class="c1"># which takes *args instead of **kwargs</span>
+                    <span class="n">load_result</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">state_dict</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">_issue_warnings_after_load</span><span class="p">(</span><span class="n">load_result</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span><span class="p">,</span> <span class="n">WEIGHTS_INDEX_NAME</span><span class="p">)):</span>
+            <span class="n">load_result</span> <span class="o">=</span> <span class="n">load_sharded_checkpoint</span><span class="p">(</span>
+                <span class="n">model</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span><span class="p">,</span> <span class="n">strict</span><span class="o">=</span><span class="n">is_sagemaker_mp_enabled</span><span class="p">()</span>
+            <span class="p">)</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_issue_warnings_after_load</span><span class="p">(</span><span class="n">load_result</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Could not locate the best model at </span><span class="si">{</span><span class="n">best_model_path</span><span class="si">}</span><span class="s2">, if you are running a distributed training &quot;</span>
+                <span class="s2">&quot;on multiple nodes, you should activate `--save_on_each_node`.&quot;</span>
+            <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._issue_warnings_after_load">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._issue_warnings_after_load">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_issue_warnings_after_load</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">load_result</span><span class="p">):</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">load_result</span><span class="o">.</span><span class="n">missing_keys</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">_keys_to_ignore_on_save</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="nb">set</span><span class="p">(</span><span class="n">load_result</span><span class="o">.</span><span class="n">missing_keys</span><span class="p">)</span> <span class="o">==</span> <span class="nb">set</span><span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">_keys_to_ignore_on_save</span>
+            <span class="p">):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">tie_weights</span><span class="p">()</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;There were missing keys in the checkpoint model loaded: </span><span class="si">{</span><span class="n">load_result</span><span class="o">.</span><span class="n">missing_keys</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">load_result</span><span class="o">.</span><span class="n">unexpected_keys</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;There were unexpected keys in the checkpoint model loaded: </span><span class="si">{</span><span class="n">load_result</span><span class="o">.</span><span class="n">unexpected_keys</span><span class="si">}</span><span class="s2">.&quot;</span>
+            <span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._maybe_log_save_evaluate">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._maybe_log_save_evaluate">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_maybe_log_save_evaluate</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tr_loss</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">trial</span><span class="p">,</span> <span class="n">epoch</span><span class="p">,</span> <span class="n">ignore_keys_for_eval</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="o">.</span><span class="n">should_log</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+                <span class="n">xm</span><span class="o">.</span><span class="n">mark_step</span><span class="p">()</span>
+
+            <span class="n">logs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="p">{}</span>
+
+            <span class="c1"># all_gather + mean() to get average loss over all processes</span>
+            <span class="n">tr_loss_scalar</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_nested_gather</span><span class="p">(</span><span class="n">tr_loss</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+
+            <span class="c1"># reset tr_loss to zero</span>
+            <span class="n">tr_loss</span> <span class="o">-=</span> <span class="n">tr_loss</span>
+
+            <span class="n">logs</span><span class="p">[</span><span class="s2">&quot;loss&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="nb">round</span><span class="p">(</span><span class="n">tr_loss_scalar</span> <span class="o">/</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">_globalstep_last_logged</span><span class="p">),</span> <span class="mi">4</span><span class="p">)</span>
+            <span class="n">logs</span><span class="p">[</span><span class="s2">&quot;learning_rate&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_learning_rate</span><span class="p">()</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">_total_loss_scalar</span> <span class="o">+=</span> <span class="n">tr_loss_scalar</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_globalstep_last_logged</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">store_flos</span><span class="p">()</span>
+
+            <span class="bp">self</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="n">logs</span><span class="p">)</span>
+
+        <span class="n">metrics</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="o">.</span><span class="n">should_evaluate</span><span class="p">:</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">eval_dataset</span><span class="p">,</span> <span class="nb">dict</span><span class="p">):</span>
+                <span class="k">for</span> <span class="n">eval_dataset_name</span><span class="p">,</span> <span class="n">eval_dataset</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">eval_dataset</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+                    <span class="n">metrics</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluate</span><span class="p">(</span>
+                        <span class="n">eval_dataset</span><span class="o">=</span><span class="n">eval_dataset</span><span class="p">,</span>
+                        <span class="n">ignore_keys</span><span class="o">=</span><span class="n">ignore_keys_for_eval</span><span class="p">,</span>
+                        <span class="n">metric_key_prefix</span><span class="o">=</span><span class="sa">f</span><span class="s2">&quot;eval_</span><span class="si">{</span><span class="n">eval_dataset_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span>
+                    <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">metrics</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluate</span><span class="p">(</span><span class="n">ignore_keys</span><span class="o">=</span><span class="n">ignore_keys_for_eval</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_report_to_hp_search</span><span class="p">(</span><span class="n">trial</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span><span class="p">,</span> <span class="n">metrics</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_save_checkpoint</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">trial</span><span class="p">,</span> <span class="n">metrics</span><span class="o">=</span><span class="n">metrics</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._load_rng_state">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_rng_state">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_load_rng_state</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">checkpoint</span><span class="p">):</span>
+        <span class="c1"># Load RNG states from `checkpoint`</span>
+        <span class="k">if</span> <span class="n">checkpoint</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">process_index</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">process_index</span>
+            <span class="n">rng_file</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;rng_state_</span><span class="si">{</span><span class="n">process_index</span><span class="si">}</span><span class="s2">.pth&quot;</span><span class="p">)</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">rng_file</span><span class="p">):</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;Didn&#39;t find an RNG file for process </span><span class="si">{</span><span class="n">process_index</span><span class="si">}</span><span class="s2">, if you are resuming a training that &quot;</span>
+                    <span class="s2">&quot;wasn&#39;t launched in a distributed fashion, reproducibility is not guaranteed.&quot;</span>
+                <span class="p">)</span>
+                <span class="k">return</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">rng_file</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="s2">&quot;rng_state.pth&quot;</span><span class="p">)</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">rng_file</span><span class="p">):</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span>
+                    <span class="s2">&quot;Didn&#39;t find an RNG file, if you are resuming a training that was launched in a distributed &quot;</span>
+                    <span class="s2">&quot;fashion, reproducibility is not guaranteed.&quot;</span>
+                <span class="p">)</span>
+                <span class="k">return</span>
+
+        <span class="n">checkpoint_rng_state</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">rng_file</span><span class="p">)</span>
+        <span class="n">random</span><span class="o">.</span><span class="n">setstate</span><span class="p">(</span><span class="n">checkpoint_rng_state</span><span class="p">[</span><span class="s2">&quot;python&quot;</span><span class="p">])</span>
+        <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">set_state</span><span class="p">(</span><span class="n">checkpoint_rng_state</span><span class="p">[</span><span class="s2">&quot;numpy&quot;</span><span class="p">])</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">set_rng_state</span><span class="p">(</span><span class="n">checkpoint_rng_state</span><span class="p">[</span><span class="s2">&quot;cpu&quot;</span><span class="p">])</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">is_available</span><span class="p">():</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">!=</span> <span class="o">-</span><span class="mi">1</span><span class="p">:</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">set_rng_state</span><span class="p">(</span><span class="n">checkpoint_rng_state</span><span class="p">[</span><span class="s2">&quot;cuda&quot;</span><span class="p">])</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">try</span><span class="p">:</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">set_rng_state_all</span><span class="p">(</span><span class="n">checkpoint_rng_state</span><span class="p">[</span><span class="s2">&quot;cuda&quot;</span><span class="p">])</span>
+                <span class="k">except</span> <span class="ne">Exception</span> <span class="k">as</span> <span class="n">e</span><span class="p">:</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;Didn&#39;t manage to set back the RNG states of the GPU because of the following error:</span><span class="se">\n</span><span class="s2"> </span><span class="si">{</span><span class="n">e</span><span class="si">}</span><span class="s2">&quot;</span>
+                        <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">This won&#39;t yield the same results as if the training had not been interrupted.&quot;</span>
+                    <span class="p">)</span>
+        <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+            <span class="n">xm</span><span class="o">.</span><span class="n">set_rng_state</span><span class="p">(</span><span class="n">checkpoint_rng_state</span><span class="p">[</span><span class="s2">&quot;xla&quot;</span><span class="p">])</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._save_checkpoint">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._save_checkpoint">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_save_checkpoint</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">trial</span><span class="p">,</span> <span class="n">metrics</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="c1"># In all cases, including ddp/dp/deepspeed, self.model is always a reference to the model we</span>
+        <span class="c1"># want to save except FullyShardedDDP.</span>
+        <span class="c1"># assert unwrap_model(model) is self.model, &quot;internal model should be a reference to self.model&quot;</span>
+
+        <span class="c1"># Save model checkpoint</span>
+        <span class="c1">#checkpoint_folder = f&quot;{PREFIX_CHECKPOINT_DIR}-{self.state.global_step}&quot;</span>
+        <span class="n">checkpoint_folder</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">PREFIX_CHECKPOINT_DIR</span><span class="si">}</span><span class="s2">-</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">save_counter</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="c1">##########</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">save_counter</span> <span class="o">+=</span> <span class="mi">1</span>
+        <span class="c1">##########</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">trial</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">store_flos</span><span class="p">()</span>
+
+        <span class="n">run_dir</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_get_output_dir</span><span class="p">(</span><span class="n">trial</span><span class="o">=</span><span class="n">trial</span><span class="p">)</span>
+        <span class="n">output_dir</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">run_dir</span><span class="p">,</span> <span class="n">checkpoint_folder</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">save_model</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">_internal_call</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+            <span class="c1"># under zero3 model file itself doesn&#39;t get saved since it&#39;s bogus! Unless deepspeed</span>
+            <span class="c1"># config `stage3_gather_16bit_weights_on_model_save` is True</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="o">.</span><span class="n">save_checkpoint</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span>
+
+        <span class="c1"># Save optimizer and scheduler</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">sharded_ddp</span> <span class="o">==</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">SIMPLE</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">consolidate_state_dict</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+            <span class="n">xm</span><span class="o">.</span><span class="n">rendezvous</span><span class="p">(</span><span class="s2">&quot;saving_optimizer_states&quot;</span><span class="p">)</span>
+            <span class="n">xm</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">state_dict</span><span class="p">(),</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">OPTIMIZER_NAME</span><span class="p">))</span>
+            <span class="k">with</span> <span class="n">warnings</span><span class="o">.</span><span class="n">catch_warnings</span><span class="p">(</span><span class="n">record</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span> <span class="k">as</span> <span class="n">caught_warnings</span><span class="p">:</span>
+                <span class="n">xm</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span><span class="o">.</span><span class="n">state_dict</span><span class="p">(),</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">SCHEDULER_NAME</span><span class="p">))</span>
+                <span class="n">reissue_pt_warnings</span><span class="p">(</span><span class="n">caught_warnings</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+            <span class="n">opt_state_dict</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">local_state_dict</span><span class="p">(</span><span class="n">gather_if_shard</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+            <span class="n">smp</span><span class="o">.</span><span class="n">barrier</span><span class="p">()</span>
+            <span class="k">if</span> <span class="n">smp</span><span class="o">.</span><span class="n">rdp_rank</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">or</span> <span class="n">smp</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">cfg</span><span class="o">.</span><span class="n">shard_optimizer_state</span><span class="p">:</span>
+                <span class="n">smp</span><span class="o">.</span><span class="n">save</span><span class="p">(</span>
+                    <span class="n">opt_state_dict</span><span class="p">,</span>
+                    <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">OPTIMIZER_NAME</span><span class="p">),</span>
+                    <span class="n">partial</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                    <span class="n">v3</span><span class="o">=</span><span class="n">smp</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">cfg</span><span class="o">.</span><span class="n">shard_optimizer_state</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+                <span class="k">with</span> <span class="n">warnings</span><span class="o">.</span><span class="n">catch_warnings</span><span class="p">(</span><span class="n">record</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span> <span class="k">as</span> <span class="n">caught_warnings</span><span class="p">:</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span><span class="o">.</span><span class="n">state_dict</span><span class="p">(),</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">SCHEDULER_NAME</span><span class="p">))</span>
+                <span class="n">reissue_pt_warnings</span><span class="p">(</span><span class="n">caught_warnings</span><span class="p">)</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">do_grad_scaling</span><span class="p">:</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">scaler</span><span class="o">.</span><span class="n">state_dict</span><span class="p">(),</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">SCALER_NAME</span><span class="p">))</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+            <span class="c1"># deepspeed.save_checkpoint above saves model/optim/sched</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">state_dict</span><span class="p">(),</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">OPTIMIZER_NAME</span><span class="p">))</span>
+            <span class="k">with</span> <span class="n">warnings</span><span class="o">.</span><span class="n">catch_warnings</span><span class="p">(</span><span class="n">record</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span> <span class="k">as</span> <span class="n">caught_warnings</span><span class="p">:</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span><span class="o">.</span><span class="n">state_dict</span><span class="p">(),</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">SCHEDULER_NAME</span><span class="p">))</span>
+            <span class="n">reissue_pt_warnings</span><span class="p">(</span><span class="n">caught_warnings</span><span class="p">)</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">do_grad_scaling</span><span class="p">:</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">scaler</span><span class="o">.</span><span class="n">state_dict</span><span class="p">(),</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">SCALER_NAME</span><span class="p">))</span>
+
+        <span class="c1"># Determine the new best metric / best model checkpoint</span>
+        <span class="k">if</span> <span class="n">metrics</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">metric_for_best_model</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">metric_to_check</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">metric_for_best_model</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">metric_to_check</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="s2">&quot;eval_&quot;</span><span class="p">):</span>
+                <span class="n">metric_to_check</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;eval_</span><span class="si">{</span><span class="n">metric_to_check</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="n">metric_value</span> <span class="o">=</span> <span class="n">metrics</span><span class="p">[</span><span class="n">metric_to_check</span><span class="p">]</span>
+
+            <span class="n">operator</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">greater</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">greater_is_better</span> <span class="k">else</span> <span class="n">np</span><span class="o">.</span><span class="n">less</span>
+            <span class="k">if</span> <span class="p">(</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_metric</span> <span class="ow">is</span> <span class="kc">None</span>
+                <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span> <span class="ow">is</span> <span class="kc">None</span>
+                <span class="ow">or</span> <span class="n">operator</span><span class="p">(</span><span class="n">metric_value</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_metric</span><span class="p">)</span>
+            <span class="p">):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_metric</span> <span class="o">=</span> <span class="n">metric_value</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span> <span class="o">=</span> <span class="n">output_dir</span>
+
+        <span class="c1"># Save the Trainer state</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">save_to_json</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">TRAINER_STATE_NAME</span><span class="p">))</span>
+
+        <span class="c1"># Save RNG state in non-distributed training</span>
+        <span class="n">rng_states</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;python&quot;</span><span class="p">:</span> <span class="n">random</span><span class="o">.</span><span class="n">getstate</span><span class="p">(),</span>
+            <span class="s2">&quot;numpy&quot;</span><span class="p">:</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">get_state</span><span class="p">(),</span>
+            <span class="s2">&quot;cpu&quot;</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">get_rng_state</span><span class="p">(),</span>
+        <span class="p">}</span>
+        <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">is_available</span><span class="p">():</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">==</span> <span class="o">-</span><span class="mi">1</span><span class="p">:</span>
+                <span class="c1"># In non distributed, we save the global CUDA RNG state (will take care of DataParallel)</span>
+                <span class="n">rng_states</span><span class="p">[</span><span class="s2">&quot;cuda&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">get_rng_state_all</span><span class="p">()</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">rng_states</span><span class="p">[</span><span class="s2">&quot;cuda&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">get_rng_state</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+            <span class="n">rng_states</span><span class="p">[</span><span class="s2">&quot;xla&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">xm</span><span class="o">.</span><span class="n">get_rng_state</span><span class="p">()</span>
+
+        <span class="c1"># A process can arrive here before the process 0 has a chance to save the model, in which case output_dir may</span>
+        <span class="c1"># not yet exist.</span>
+        <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span> <span class="o">&lt;=</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="n">rng_states</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="s2">&quot;rng_state.pth&quot;</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="n">rng_states</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;rng_state_</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">process_index</span><span class="si">}</span><span class="s2">.pth&quot;</span><span class="p">))</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">push_to_hub</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_push_from_checkpoint</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span>
+
+        <span class="c1"># Maybe delete some older checkpoints.</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_rotate_checkpoints</span><span class="p">(</span><span class="n">use_mtime</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">output_dir</span><span class="o">=</span><span class="n">run_dir</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._load_optimizer_and_scheduler">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_optimizer_and_scheduler">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_load_optimizer_and_scheduler</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">checkpoint</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;If optimizer and scheduler states exist, load them.&quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">checkpoint</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+            <span class="c1"># deepspeed loads optimizer/lr_scheduler together with the model in deepspeed_init</span>
+            <span class="k">return</span>
+
+        <span class="n">checkpoint_file_exists</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="n">glob</span><span class="o">.</span><span class="n">glob</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="n">OPTIMIZER_NAME</span><span class="p">)</span> <span class="o">+</span> <span class="s2">&quot;_*&quot;</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">()</span>
+            <span class="k">else</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="n">OPTIMIZER_NAME</span><span class="p">))</span>
+        <span class="p">)</span>
+        <span class="k">if</span> <span class="n">checkpoint_file_exists</span> <span class="ow">and</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="n">SCHEDULER_NAME</span><span class="p">)):</span>
+            <span class="c1"># Load in optimizer and scheduler states</span>
+            <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+                <span class="c1"># On TPU we have to take some extra precautions to properly load the states on the right device.</span>
+                <span class="n">optimizer_state</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="n">OPTIMIZER_NAME</span><span class="p">),</span> <span class="n">map_location</span><span class="o">=</span><span class="s2">&quot;cpu&quot;</span><span class="p">)</span>
+                <span class="k">with</span> <span class="n">warnings</span><span class="o">.</span><span class="n">catch_warnings</span><span class="p">(</span><span class="n">record</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span> <span class="k">as</span> <span class="n">caught_warnings</span><span class="p">:</span>
+                    <span class="n">lr_scheduler_state</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="n">SCHEDULER_NAME</span><span class="p">),</span> <span class="n">map_location</span><span class="o">=</span><span class="s2">&quot;cpu&quot;</span><span class="p">)</span>
+                <span class="n">reissue_pt_warnings</span><span class="p">(</span><span class="n">caught_warnings</span><span class="p">)</span>
+
+                <span class="n">xm</span><span class="o">.</span><span class="n">send_cpu_data_to_device</span><span class="p">(</span><span class="n">optimizer_state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+                <span class="n">xm</span><span class="o">.</span><span class="n">send_cpu_data_to_device</span><span class="p">(</span><span class="n">lr_scheduler_state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+
+                <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">optimizer_state</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">lr_scheduler_state</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">map_location</span> <span class="o">=</span> <span class="s2">&quot;cpu&quot;</span> <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">()</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">device</span>
+                <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+                    <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="s2">&quot;user_content.pt&quot;</span><span class="p">)):</span>
+                        <span class="c1"># Optimizer checkpoint was saved with smp &gt;= 1.10</span>
+                        <span class="k">def</span><span class="w"> </span><span class="nf">opt_load_hook</span><span class="p">(</span><span class="n">mod</span><span class="p">,</span> <span class="n">opt</span><span class="p">):</span>
+                            <span class="n">opt</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">smp</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="n">OPTIMIZER_NAME</span><span class="p">),</span> <span class="n">partial</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="c1"># Optimizer checkpoint was saved with smp &lt; 1.10</span>
+                        <span class="k">def</span><span class="w"> </span><span class="nf">opt_load_hook</span><span class="p">(</span><span class="n">mod</span><span class="p">,</span> <span class="n">opt</span><span class="p">):</span>
+                            <span class="k">if</span> <span class="n">IS_SAGEMAKER_MP_POST_1_10</span><span class="p">:</span>
+                                <span class="n">opt</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span>
+                                    <span class="n">smp</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="n">OPTIMIZER_NAME</span><span class="p">),</span> <span class="n">partial</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">back_compat</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                                <span class="p">)</span>
+                            <span class="k">else</span><span class="p">:</span>
+                                <span class="n">opt</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">smp</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="n">OPTIMIZER_NAME</span><span class="p">),</span> <span class="n">partial</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+
+                    <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span><span class="o">.</span><span class="n">register_post_step_hook</span><span class="p">(</span><span class="n">opt_load_hook</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span>
+                        <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="n">OPTIMIZER_NAME</span><span class="p">),</span> <span class="n">map_location</span><span class="o">=</span><span class="n">map_location</span><span class="p">)</span>
+                    <span class="p">)</span>
+                <span class="k">with</span> <span class="n">warnings</span><span class="o">.</span><span class="n">catch_warnings</span><span class="p">(</span><span class="n">record</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span> <span class="k">as</span> <span class="n">caught_warnings</span><span class="p">:</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">lr_scheduler</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="n">SCHEDULER_NAME</span><span class="p">)))</span>
+                <span class="n">reissue_pt_warnings</span><span class="p">(</span><span class="n">caught_warnings</span><span class="p">)</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">do_grad_scaling</span> <span class="ow">and</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="n">SCALER_NAME</span><span class="p">)):</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="n">SCALER_NAME</span><span class="p">)))</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.hyperparameter_search">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.hyperparameter_search">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">hyperparameter_search</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">hp_space</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[</span><span class="s2">&quot;optuna.Trial&quot;</span><span class="p">],</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">float</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">compute_objective</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">float</span><span class="p">]],</span> <span class="nb">float</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">n_trials</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">20</span><span class="p">,</span>
+        <span class="n">direction</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;minimize&quot;</span><span class="p">,</span>
+        <span class="n">backend</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="s2">&quot;str&quot;</span><span class="p">,</span> <span class="n">HPSearchBackend</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">hp_name</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Callable</span><span class="p">[[</span><span class="s2">&quot;optuna.Trial&quot;</span><span class="p">],</span> <span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">BestRun</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Launch an hyperparameter search using `optuna` or `Ray Tune` or `SigOpt`. The optimized quantity is determined</span>
+<span class="sd">        by `compute_objective`, which defaults to a function returning the evaluation loss when no metric is provided,</span>
+<span class="sd">        the sum of all metrics otherwise.</span>
+<span class="sd">        &lt;Tip warning={true}&gt;</span>
+<span class="sd">        To use this method, you need to have provided a `model_init` when initializing your [`Trainer`]: we need to</span>
+<span class="sd">        reinitialize the model at each new run. This is incompatible with the `optimizers` argument, so you need to</span>
+<span class="sd">        subclass [`Trainer`] and override the method [`~Trainer.create_optimizer_and_scheduler`] for custom</span>
+<span class="sd">        optimizer/scheduler.</span>
+<span class="sd">        &lt;/Tip&gt;</span>
+<span class="sd">        Args:</span>
+<span class="sd">            hp_space (`Callable[[&quot;optuna.Trial&quot;], Dict[str, float]]`, *optional*):</span>
+<span class="sd">                A function that defines the hyperparameter search space. Will default to</span>
+<span class="sd">                [`~trainer_utils.default_hp_space_optuna`] or [`~trainer_utils.default_hp_space_ray`] or</span>
+<span class="sd">                [`~trainer_utils.default_hp_space_sigopt`] depending on your backend.</span>
+<span class="sd">            compute_objective (`Callable[[Dict[str, float]], float]`, *optional*):</span>
+<span class="sd">                A function computing the objective to minimize or maximize from the metrics returned by the `evaluate`</span>
+<span class="sd">                method. Will default to [`~trainer_utils.default_compute_objective`].</span>
+<span class="sd">            n_trials (`int`, *optional*, defaults to 100):</span>
+<span class="sd">                The number of trial runs to test.</span>
+<span class="sd">            direction (`str`, *optional*, defaults to `&quot;minimize&quot;`):</span>
+<span class="sd">                Whether to optimize greater or lower objects. Can be `&quot;minimize&quot;` or `&quot;maximize&quot;`, you should pick</span>
+<span class="sd">                `&quot;minimize&quot;` when optimizing the validation loss, `&quot;maximize&quot;` when optimizing one or several metrics.</span>
+<span class="sd">            backend (`str` or [`~training_utils.HPSearchBackend`], *optional*):</span>
+<span class="sd">                The backend to use for hyperparameter search. Will default to optuna or Ray Tune or SigOpt, depending</span>
+<span class="sd">                on which one is installed. If all are installed, will default to optuna.</span>
+<span class="sd">            hp_name (`Callable[[&quot;optuna.Trial&quot;], str]]`, *optional*):</span>
+<span class="sd">                A function that defines the trial/run name. Will default to None.</span>
+<span class="sd">            kwargs (`Dict[str, Any]`, *optional*):</span>
+<span class="sd">                Additional keyword arguments passed along to `optuna.create_study` or `ray.tune.run`. For more</span>
+<span class="sd">                information see:</span>
+<span class="sd">                - the documentation of</span>
+<span class="sd">                  [optuna.create_study](https://optuna.readthedocs.io/en/stable/reference/generated/optuna.study.create_study.html)</span>
+<span class="sd">                - the documentation of [tune.run](https://docs.ray.io/en/latest/tune/api_docs/execution.html#tune-run)</span>
+<span class="sd">                - the documentation of [sigopt](https://app.sigopt.com/docs/endpoints/experiments/create)</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            [`trainer_utils.BestRun`]: All the information about the best run. Experiment summary can be found in</span>
+<span class="sd">            `run_summary` attribute for Ray backend.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">backend</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">backend</span> <span class="o">=</span> <span class="n">default_hp_search_backend</span><span class="p">()</span>
+            <span class="k">if</span> <span class="n">backend</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                    <span class="s2">&quot;At least one of optuna or ray should be installed. &quot;</span>
+                    <span class="s2">&quot;To install optuna run `pip install optuna`. &quot;</span>
+                    <span class="s2">&quot;To install ray run `pip install ray[tune]`. &quot;</span>
+                    <span class="s2">&quot;To install sigopt run `pip install sigopt`.&quot;</span>
+                <span class="p">)</span>
+        <span class="n">backend</span> <span class="o">=</span> <span class="n">HPSearchBackend</span><span class="p">(</span><span class="n">backend</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">OPTUNA</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;You picked the optuna backend, but it is not installed. Use `pip install optuna`.&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">RAY</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                <span class="s2">&quot;You picked the Ray Tune backend, but it is not installed. Use `pip install &#39;ray[tune]&#39;`.&quot;</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">SIGOPT</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;You picked the sigopt backend, but it is not installed. Use `pip install sigopt`.&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">backend</span> <span class="o">==</span> <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">WANDB</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s2">&quot;You picked the wandb backend, but it is not installed. Use `pip install wandb`.&quot;</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">=</span> <span class="n">backend</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_init</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span>
+                <span class="s2">&quot;To use hyperparameter search, you need to pass your model through a model_init function.&quot;</span>
+            <span class="p">)</span>
+        <span class="k">try</span><span class="p">:</span>
+            <span class="n">backend_dict</span> <span class="o">=</span> <span class="p">{</span>
+                <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">OPTUNA</span><span class="p">:</span> <span class="n">run_hp_search_optuna</span><span class="p">,</span>
+                <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">RAY</span><span class="p">:</span> <span class="n">run_hp_search_ray</span><span class="p">,</span>
+                <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">SIGOPT</span><span class="p">:</span> <span class="n">run_hp_search_sigopt</span><span class="p">,</span>
+                <span class="n">HPSearchBackend</span><span class="o">.</span><span class="n">WANDB</span><span class="p">:</span> <span class="n">run_hp_search_wandb</span><span class="p">,</span>
+            <span class="p">}</span>
+            <span class="n">backend_run</span> <span class="o">=</span> <span class="n">backend_dict</span><span class="p">[</span><span class="n">backend</span><span class="p">]</span>
+        <span class="k">except</span> <span class="ne">NameError</span><span class="p">:</span>
+            <span class="n">ALL_HYPERPARAMETER_SEARCH_BACKENDS</span>
+            <span class="n">backend_obj</span> <span class="o">=</span> <span class="n">ALL_HYPERPARAMETER_SEARCH_BACKENDS</span><span class="p">[</span><span class="n">backend</span><span class="p">]()</span>
+            <span class="n">backend_run</span> <span class="o">=</span> <span class="n">backend_obj</span><span class="o">.</span><span class="n">run</span>
+        <span class="k">try</span><span class="p">:</span>
+            <span class="kn">from</span><span class="w"> </span><span class="nn">transformers.trainer_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">default_hp_space</span>
+        <span class="k">except</span> <span class="ne">ImportError</span><span class="p">:</span>
+            <span class="n">default_hp_space</span> <span class="o">=</span> <span class="n">backend_obj</span><span class="o">.</span><span class="n">default_hp_space</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">hp_space</span> <span class="o">=</span> <span class="n">default_hp_space</span><span class="p">[</span><span class="n">backend</span><span class="p">]</span> <span class="k">if</span> <span class="n">hp_space</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">hp_space</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">hp_name</span> <span class="o">=</span> <span class="n">hp_name</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">compute_objective</span> <span class="o">=</span> <span class="n">default_compute_objective</span> <span class="k">if</span> <span class="n">compute_objective</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">compute_objective</span>
+
+
+        <span class="n">best_run</span> <span class="o">=</span> <span class="n">backend_run</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">n_trials</span><span class="p">,</span> <span class="n">direction</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">hp_search_backend</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="k">return</span> <span class="n">best_run</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.log">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.log">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">log</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">logs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">float</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Log `logs` on the various objects watching training.</span>
+<span class="sd">        Subclass and override this method to inject custom behavior.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            logs (`Dict[str, float]`):</span>
+<span class="sd">                The values to log.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">epoch</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">logs</span><span class="p">[</span><span class="s2">&quot;epoch&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="nb">round</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">epoch</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+
+        <span class="n">output</span> <span class="o">=</span> <span class="p">{</span><span class="o">**</span><span class="n">logs</span><span class="p">,</span> <span class="o">**</span><span class="p">{</span><span class="s2">&quot;step&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span><span class="p">}}</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">log_history</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">output</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_log</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">,</span> <span class="n">logs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._prepare_input">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._prepare_input">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_prepare_input</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Any</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Prepares one `data` before feeding it to the model, be it a tensor or a nested list/dictionary of tensors.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">Mapping</span><span class="p">):</span>
+            <span class="k">return</span> <span class="nb">type</span><span class="p">(</span><span class="n">data</span><span class="p">)({</span><span class="n">k</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_prepare_input</span><span class="p">(</span><span class="n">v</span><span class="p">)</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">data</span><span class="o">.</span><span class="n">items</span><span class="p">()})</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="p">(</span><span class="nb">tuple</span><span class="p">,</span> <span class="nb">list</span><span class="p">)):</span>
+            <span class="k">return</span> <span class="nb">type</span><span class="p">(</span><span class="n">data</span><span class="p">)(</span><span class="bp">self</span><span class="o">.</span><span class="n">_prepare_input</span><span class="p">(</span><span class="n">v</span><span class="p">)</span> <span class="k">for</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">data</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+            <span class="n">kwargs</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;device&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">}</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span> <span class="ow">and</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">is_floating_point</span><span class="p">(</span><span class="n">data</span><span class="p">)</span> <span class="ow">or</span> <span class="n">torch</span><span class="o">.</span><span class="n">is_complex</span><span class="p">(</span><span class="n">data</span><span class="p">)):</span>
+                <span class="c1"># NLP models inputs are int/uint and those get adjusted to the right dtype of the</span>
+                <span class="c1"># embedding. Other models such as wav2vec2&#39;s inputs are already float and thus</span>
+                <span class="c1"># may need special handling to match the dtypes of the model</span>
+                <span class="n">kwargs</span><span class="o">.</span><span class="n">update</span><span class="p">({</span><span class="s2">&quot;dtype&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hf_deepspeed_config</span><span class="o">.</span><span class="n">dtype</span><span class="p">()})</span>
+            <span class="k">return</span> <span class="n">data</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">data</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._prepare_inputs">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._prepare_inputs">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_prepare_inputs</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Any</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Prepare `inputs` before feeding them to the model, converting them to tensors if they are not already and</span>
+<span class="sd">        handling potential state.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">inputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_prepare_input</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;The batch received was empty, your model won&#39;t be able to train on it. Double-check that your &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;training dataset contains keys expected by the model: </span><span class="si">{</span><span class="s1">&#39;,&#39;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_signature_columns</span><span class="p">)</span><span class="si">}</span><span class="s2">.&quot;</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">past_index</span> <span class="o">&gt;=</span> <span class="mi">0</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">_past</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">inputs</span><span class="p">[</span><span class="s2">&quot;mems&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_past</span>
+
+        <span class="k">return</span> <span class="n">inputs</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.compute_loss_context_manager">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_loss_context_manager">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">compute_loss_context_manager</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        A helper wrapper to group together context managers.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">autocast_smart_context_manager</span><span class="p">()</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.autocast_smart_context_manager">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.autocast_smart_context_manager">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">autocast_smart_context_manager</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">cache_enabled</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">True</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        A helper wrapper that creates an appropriate context manager for `autocast` while feeding it the desired</span>
+<span class="sd">        arguments, depending on the situation.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_cuda_amp</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_cpu_amp</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">is_torch_greater_or_equal_than_1_10</span><span class="p">:</span>
+                <span class="n">ctx_manager</span> <span class="o">=</span> <span class="p">(</span>
+                    <span class="n">torch</span><span class="o">.</span><span class="n">cpu</span><span class="o">.</span><span class="n">amp</span><span class="o">.</span><span class="n">autocast</span><span class="p">(</span><span class="n">cache_enabled</span><span class="o">=</span><span class="n">cache_enabled</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">amp_dtype</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_cpu_amp</span>
+                    <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">amp</span><span class="o">.</span><span class="n">autocast</span><span class="p">(</span><span class="n">cache_enabled</span><span class="o">=</span><span class="n">cache_enabled</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">amp_dtype</span><span class="p">)</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">ctx_manager</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">amp</span><span class="o">.</span><span class="n">autocast</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">ctx_manager</span> <span class="o">=</span> <span class="n">contextlib</span><span class="o">.</span><span class="n">nullcontext</span><span class="p">()</span> <span class="k">if</span> <span class="n">sys</span><span class="o">.</span><span class="n">version_info</span> <span class="o">&gt;=</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">7</span><span class="p">)</span> <span class="k">else</span> <span class="n">contextlib</span><span class="o">.</span><span class="n">suppress</span><span class="p">()</span>
+
+        <span class="k">return</span> <span class="n">ctx_manager</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.training_step">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.training_step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">training_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span> <span class="n">inputs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Any</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform a training step on a batch of inputs.</span>
+<span class="sd">        Subclass and override to inject custom behavior.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            model (`nn.Module`):</span>
+<span class="sd">                The model to train.</span>
+<span class="sd">            inputs (`Dict[str, Union[torch.Tensor, Any]]`):</span>
+<span class="sd">                The inputs and targets of the model.</span>
+<span class="sd">                The dictionary will be unpacked before being fed to the model. Most models expect the targets under the</span>
+<span class="sd">                argument `labels`. Check your model&#39;s documentation for all accepted arguments.</span>
+<span class="sd">        Return:</span>
+<span class="sd">            `torch.Tensor`: The tensor with training loss on this batch.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">train</span><span class="p">()</span>
+        <span class="n">inputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_prepare_inputs</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+            <span class="n">loss_mb</span> <span class="o">=</span> <span class="n">smp_forward_backward</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span><span class="p">)</span>
+            <span class="k">return</span> <span class="n">loss_mb</span><span class="o">.</span><span class="n">reduce_mean</span><span class="p">()</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+
+        <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_loss_context_manager</span><span class="p">():</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_loss</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">n_gpu</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">loss</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span>  <span class="c1"># mean() to average on multi-gpu parallel training</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+            <span class="c1"># deepspeed handles loss scaling by gradient_accumulation_steps in its `backward`</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">loss</span> <span class="o">/</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">gradient_accumulation_steps</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">do_grad_scaling</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">scaler</span><span class="o">.</span><span class="n">scale</span><span class="p">(</span><span class="n">loss</span><span class="p">)</span><span class="o">.</span><span class="n">backward</span><span class="p">()</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_apex</span><span class="p">:</span>
+            <span class="k">with</span> <span class="n">amp</span><span class="o">.</span><span class="n">scale_loss</span><span class="p">(</span><span class="n">loss</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">optimizer</span><span class="p">)</span> <span class="k">as</span> <span class="n">scaled_loss</span><span class="p">:</span>
+                <span class="n">scaled_loss</span><span class="o">.</span><span class="n">backward</span><span class="p">()</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+            <span class="c1"># loss gets scaled under gradient_accumulation_steps in deepspeed</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="o">.</span><span class="n">backward</span><span class="p">(</span><span class="n">loss</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">loss</span><span class="o">.</span><span class="n">backward</span><span class="p">()</span>
+
+        <span class="k">return</span> <span class="n">loss</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.compute_loss">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_loss">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">compute_loss</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">return_outputs</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        How the loss is computed by Trainer. By default, all models return the loss in the first element.</span>
+<span class="sd">        Subclass and override for custom behavior.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">label_smoother</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="s2">&quot;labels&quot;</span> <span class="ow">in</span> <span class="n">inputs</span><span class="p">:</span>
+            <span class="n">labels</span> <span class="o">=</span> <span class="n">inputs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;labels&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">labels</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="o">**</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="c1"># Save past state if it exists</span>
+        <span class="c1"># TODO: this needs to be fixed and made cleaner later.</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">past_index</span> <span class="o">&gt;=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_past</span> <span class="o">=</span> <span class="n">outputs</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">past_index</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">unwrap_model</span><span class="p">(</span><span class="n">model</span><span class="p">)</span><span class="o">.</span><span class="n">_get_name</span><span class="p">()</span> <span class="ow">in</span> <span class="n">MODEL_FOR_CAUSAL_LM_MAPPING_NAMES</span><span class="o">.</span><span class="n">values</span><span class="p">():</span>
+                <span class="n">loss</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">label_smoother</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="n">labels</span><span class="p">,</span> <span class="n">shift_labels</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">loss</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">label_smoother</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="n">labels</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="nb">dict</span><span class="p">)</span> <span class="ow">and</span> <span class="s2">&quot;loss&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">outputs</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="s2">&quot;The model did not return a loss from the inputs, only the following keys: &quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="s1">&#39;,&#39;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">outputs</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span><span class="si">}</span><span class="s2">. For reference, the inputs it received are </span><span class="si">{</span><span class="s1">&#39;,&#39;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">inputs</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span><span class="si">}</span><span class="s2">.&quot;</span>
+                <span class="p">)</span>
+            <span class="c1"># We don&#39;t use .loss here since the model may return tuples instead of ModelOutput.</span>
+            <span class="n">loss</span> <span class="o">=</span> <span class="n">outputs</span><span class="p">[</span><span class="s2">&quot;loss&quot;</span><span class="p">]</span> <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="nb">dict</span><span class="p">)</span> <span class="k">else</span> <span class="n">outputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="k">return</span> <span class="p">(</span><span class="n">loss</span><span class="p">,</span> <span class="n">outputs</span><span class="p">)</span> <span class="k">if</span> <span class="n">return_outputs</span> <span class="k">else</span> <span class="n">loss</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.is_local_process_zero">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_local_process_zero">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">is_local_process_zero</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Whether or not this process is the local (e.g., on one machine if training in a distributed fashion on several</span>
+<span class="sd">        machines) main process.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">local_process_index</span> <span class="o">==</span> <span class="mi">0</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.is_world_process_zero">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_world_process_zero">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">is_world_process_zero</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Whether or not this process is the global main process (when training in a distributed fashion on several</span>
+<span class="sd">        machines, this is only going to be `True` for one process).</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># Special case for SageMaker ModelParallel since there process_index is dp_process_index, not the global</span>
+        <span class="c1"># process index.</span>
+        <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+            <span class="k">return</span> <span class="n">smp</span><span class="o">.</span><span class="n">rank</span><span class="p">()</span> <span class="o">==</span> <span class="mi">0</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">process_index</span> <span class="o">==</span> <span class="mi">0</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.save_model">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.save_model">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">save_model</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">output_dir</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">_internal_call</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Will save the model, so you can reload it using `from_pretrained()`.</span>
+<span class="sd">        Will only save from the main process.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+
+        <span class="k">if</span> <span class="n">output_dir</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">output_dir</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span>
+
+        <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_save_tpu</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+            <span class="c1"># Calling the state_dict needs to be done on the wrapped model and on all processes.</span>
+            <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="n">state_dict</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span><span class="o">.</span><span class="n">state_dict</span><span class="p">()</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_save</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">state_dict</span><span class="o">=</span><span class="n">state_dict</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">IS_SAGEMAKER_MP_POST_1_10</span><span class="p">:</span>
+                <span class="c1"># &#39;user_content.pt&#39; indicates model state_dict saved with smp &gt;= 1.10</span>
+                <span class="n">Path</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="s2">&quot;user_content.pt&quot;</span><span class="p">))</span><span class="o">.</span><span class="n">touch</span><span class="p">()</span>
+        <span class="k">elif</span> <span class="p">(</span>
+            <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">ZERO_DP_2</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">sharded_ddp</span>
+            <span class="ow">or</span> <span class="n">ShardedDDPOption</span><span class="o">.</span><span class="n">ZERO_DP_3</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">sharded_ddp</span>
+            <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">fsdp</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+        <span class="p">):</span>
+            <span class="n">state_dict</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">state_dict</span><span class="p">()</span>
+
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_save</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">state_dict</span><span class="o">=</span><span class="n">state_dict</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+            <span class="c1"># this takes care of everything as long as we aren&#39;t under zero3</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_save</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">is_deepspeed_zero3_enabled</span><span class="p">():</span>
+                <span class="c1"># It&#39;s too complicated to try to override different places where the weights dump gets</span>
+                <span class="c1"># saved, so since under zero3 the file is bogus, simply delete it. The user should</span>
+                <span class="c1"># either user deepspeed checkpoint to resume or to recover full weights use</span>
+                <span class="c1"># zero_to_fp32.py stored in the checkpoint.</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+                    <span class="n">file</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">WEIGHTS_NAME</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">file</span><span class="p">):</span>
+                        <span class="c1"># logger.info(f&quot;deepspeed zero3: removing {file}, see zero_to_fp32.py to recover weights&quot;)</span>
+                        <span class="n">os</span><span class="o">.</span><span class="n">remove</span><span class="p">(</span><span class="n">file</span><span class="p">)</span>
+
+                <span class="c1"># now save the real model if stage3_gather_16bit_weights_on_model_save=True</span>
+                <span class="c1"># if false it will not be saved.</span>
+                <span class="c1"># This must be called on all ranks</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="o">.</span><span class="n">save_16bit_model</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">WEIGHTS_NAME</span><span class="p">):</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                        <span class="s2">&quot;deepspeed.save_16bit_model didn&#39;t save the model, since&quot;</span>
+                        <span class="s2">&quot; stage3_gather_16bit_weights_on_model_save=false. Saving the full checkpoint instead, use&quot;</span>
+                        <span class="s2">&quot; zero_to_fp32.py to recover weights&quot;</span>
+                    <span class="p">)</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="o">.</span><span class="n">save_checkpoint</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span>
+
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_save</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span>
+
+        <span class="c1"># Push to the Hub when `save_model` is called by the user.</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">push_to_hub</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">_internal_call</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">push_to_hub</span><span class="p">(</span><span class="n">commit_message</span><span class="o">=</span><span class="s2">&quot;Model save&quot;</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._save_tpu">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._save_tpu">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_save_tpu</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">output_dir</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
+        <span class="n">output_dir</span> <span class="o">=</span> <span class="n">output_dir</span> <span class="k">if</span> <span class="n">output_dir</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Saving model checkpoint to </span><span class="si">{</span><span class="n">output_dir</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">xm</span><span class="o">.</span><span class="n">is_master_ordinal</span><span class="p">():</span>
+            <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+            <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">TRAINING_ARGS_NAME</span><span class="p">))</span>
+
+        <span class="c1"># Save a trained model and configuration using `save_pretrained()`.</span>
+        <span class="c1"># They can then be reloaded using `from_pretrained()`</span>
+        <span class="n">xm</span><span class="o">.</span><span class="n">rendezvous</span><span class="p">(</span><span class="s2">&quot;saving_checkpoint&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">,</span> <span class="n">PreTrainedModel</span><span class="p">):</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">unwrap_model</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">),</span> <span class="n">PreTrainedModel</span><span class="p">):</span>
+                <span class="n">unwrap_model</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">)</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span>
+                    <span class="n">output_dir</span><span class="p">,</span>
+                    <span class="n">is_main_process</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">,</span>
+                    <span class="n">state_dict</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">state_dict</span><span class="p">(),</span>
+                    <span class="n">save_function</span><span class="o">=</span><span class="n">xm</span><span class="o">.</span><span class="n">save</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="s2">&quot;Trainer.model is not a `PreTrainedModel`, only saving its state dict.&quot;</span><span class="p">)</span>
+                <span class="n">state_dict</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">state_dict</span><span class="p">()</span>
+                <span class="n">xm</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="n">state_dict</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">WEIGHTS_NAME</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">is_main_process</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">,</span> <span class="n">save_function</span><span class="o">=</span><span class="n">xm</span><span class="o">.</span><span class="n">save</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._save">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._save">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_save</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">output_dir</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">state_dict</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="c1"># If we are executing this function, we are the process zero, so we don&#39;t check for that.</span>
+        <span class="n">output_dir</span> <span class="o">=</span> <span class="n">output_dir</span> <span class="k">if</span> <span class="n">output_dir</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span>
+        <span class="n">os</span><span class="o">.</span><span class="n">makedirs</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Saving model checkpoint to </span><span class="si">{</span><span class="n">output_dir</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="c1"># Save a trained model and configuration using `save_pretrained()`.</span>
+        <span class="c1"># They can then be reloaded using `from_pretrained()`</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">,</span> <span class="n">PreTrainedModel</span><span class="p">):</span>
+            <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">unwrap_model</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">),</span> <span class="n">PreTrainedModel</span><span class="p">):</span>
+                <span class="k">if</span> <span class="n">state_dict</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">state_dict</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">state_dict</span><span class="p">()</span>
+                <span class="n">unwrap_model</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">)</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">state_dict</span><span class="o">=</span><span class="n">state_dict</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="s2">&quot;Trainer.model is not a `PreTrainedModel`, only saving its state dict.&quot;</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">state_dict</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">state_dict</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">state_dict</span><span class="p">()</span>
+                <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="n">state_dict</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">WEIGHTS_NAME</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">state_dict</span><span class="o">=</span><span class="n">state_dict</span><span class="p">)</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span>
+
+        <span class="c1"># Good practice: save your training arguments together with the trained model</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">TRAINING_ARGS_NAME</span><span class="p">))</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.store_flos">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.store_flos">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">store_flos</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="c1"># Storing the number of floating-point operations that went into the model</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">!=</span> <span class="o">-</span><span class="mi">1</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">total_flos</span> <span class="o">+=</span> <span class="p">(</span>
+                <span class="n">distributed_broadcast_scalars</span><span class="p">([</span><span class="bp">self</span><span class="o">.</span><span class="n">current_flos</span><span class="p">],</span> <span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span><span class="o">.</span><span class="n">sum</span><span class="p">()</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">current_flos</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">total_flos</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">current_flos</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">current_flos</span> <span class="o">=</span> <span class="mi">0</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._sorted_checkpoints">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._sorted_checkpoints">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_sorted_checkpoints</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">output_dir</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">checkpoint_prefix</span><span class="o">=</span><span class="n">PREFIX_CHECKPOINT_DIR</span><span class="p">,</span> <span class="n">use_mtime</span><span class="o">=</span><span class="kc">False</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]:</span>
+        <span class="n">ordering_and_checkpoint_path</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="n">glob_checkpoints</span> <span class="o">=</span> <span class="p">[</span><span class="nb">str</span><span class="p">(</span><span class="n">x</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">Path</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span><span class="o">.</span><span class="n">glob</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">checkpoint_prefix</span><span class="si">}</span><span class="s2">-*&quot;</span><span class="p">)</span> <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="n">x</span><span class="p">)]</span>
+
+        <span class="k">for</span> <span class="n">path</span> <span class="ow">in</span> <span class="n">glob_checkpoints</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">use_mtime</span><span class="p">:</span>
+                <span class="n">ordering_and_checkpoint_path</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">getmtime</span><span class="p">(</span><span class="n">path</span><span class="p">),</span> <span class="n">path</span><span class="p">))</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">regex_match</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">match</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;.*</span><span class="si">{</span><span class="n">checkpoint_prefix</span><span class="si">}</span><span class="s2">-([0-9]+)&quot;</span><span class="p">,</span> <span class="n">path</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">regex_match</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">regex_match</span><span class="o">.</span><span class="n">groups</span><span class="p">()</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">ordering_and_checkpoint_path</span><span class="o">.</span><span class="n">append</span><span class="p">((</span><span class="nb">int</span><span class="p">(</span><span class="n">regex_match</span><span class="o">.</span><span class="n">groups</span><span class="p">()[</span><span class="mi">0</span><span class="p">]),</span> <span class="n">path</span><span class="p">))</span>
+
+        <span class="n">checkpoints_sorted</span> <span class="o">=</span> <span class="nb">sorted</span><span class="p">(</span><span class="n">ordering_and_checkpoint_path</span><span class="p">)</span>
+        <span class="n">checkpoints_sorted</span> <span class="o">=</span> <span class="p">[</span><span class="n">checkpoint</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="k">for</span> <span class="n">checkpoint</span> <span class="ow">in</span> <span class="n">checkpoints_sorted</span><span class="p">]</span>
+        <span class="c1"># Make sure we don&#39;t delete the best model.</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">best_model_index</span> <span class="o">=</span> <span class="n">checkpoints_sorted</span><span class="o">.</span><span class="n">index</span><span class="p">(</span><span class="nb">str</span><span class="p">(</span><span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span><span class="p">)))</span>
+            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">best_model_index</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">checkpoints_sorted</span><span class="p">)</span> <span class="o">-</span> <span class="mi">2</span><span class="p">):</span>
+                <span class="n">checkpoints_sorted</span><span class="p">[</span><span class="n">i</span><span class="p">],</span> <span class="n">checkpoints_sorted</span><span class="p">[</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="n">checkpoints_sorted</span><span class="p">[</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">],</span> <span class="n">checkpoints_sorted</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">checkpoints_sorted</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._rotate_checkpoints">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._rotate_checkpoints">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_rotate_checkpoints</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">use_mtime</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">output_dir</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">save_total_limit</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">save_total_limit</span> <span class="o">&lt;=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">return</span>
+
+        <span class="c1"># Check if we should delete older checkpoint(s)</span>
+        <span class="n">checkpoints_sorted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_sorted_checkpoints</span><span class="p">(</span><span class="n">use_mtime</span><span class="o">=</span><span class="n">use_mtime</span><span class="p">,</span> <span class="n">output_dir</span><span class="o">=</span><span class="n">output_dir</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">checkpoints_sorted</span><span class="p">)</span> <span class="o">&lt;=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">save_total_limit</span><span class="p">:</span>
+            <span class="k">return</span>
+
+        <span class="c1"># If save_total_limit=1 with load_best_model_at_end=True, we could end up deleting the last checkpoint, which</span>
+        <span class="c1"># we don&#39;t do to allow resuming.</span>
+        <span class="n">save_total_limit</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">save_total_limit</span>
+        <span class="k">if</span> <span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+            <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">save_total_limit</span> <span class="o">==</span> <span class="mi">1</span>
+            <span class="ow">and</span> <span class="n">checkpoints_sorted</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">!=</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">best_model_checkpoint</span>
+        <span class="p">):</span>
+            <span class="n">save_total_limit</span> <span class="o">=</span> <span class="mi">2</span>
+
+        <span class="n">number_of_checkpoints_to_delete</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">checkpoints_sorted</span><span class="p">)</span> <span class="o">-</span> <span class="n">save_total_limit</span><span class="p">)</span>
+        <span class="n">checkpoints_to_be_deleted</span> <span class="o">=</span> <span class="n">checkpoints_sorted</span><span class="p">[:</span><span class="n">number_of_checkpoints_to_delete</span><span class="p">]</span>
+        <span class="k">for</span> <span class="n">checkpoint</span> <span class="ow">in</span> <span class="n">checkpoints_to_be_deleted</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Deleting older checkpoint [</span><span class="si">{</span><span class="n">checkpoint</span><span class="si">}</span><span class="s2">] due to args.save_total_limit&quot;</span><span class="p">)</span>
+            <span class="n">shutil</span><span class="o">.</span><span class="n">rmtree</span><span class="p">(</span><span class="n">checkpoint</span><span class="p">,</span> <span class="n">ignore_errors</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.evaluate">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.evaluate">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">evaluate</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">eval_dataset</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dataset</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">ignore_keys</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">metric_key_prefix</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;eval&quot;</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">float</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Run evaluation and returns metrics.</span>
+<span class="sd">        The calling script will be responsible for providing a method to compute metrics, as they are task-dependent</span>
+<span class="sd">        (pass it to the init `compute_metrics` argument).</span>
+<span class="sd">        You can also subclass and override this method to inject custom behavior.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            eval_dataset (`Dataset`, *optional*):</span>
+<span class="sd">                Pass a dataset if you wish to override `self.eval_dataset`. If it is a [`~datasets.Dataset`], columns</span>
+<span class="sd">                not accepted by the `model.forward()` method are automatically removed. It must implement the `__len__`</span>
+<span class="sd">                method.</span>
+<span class="sd">            ignore_keys (`Lst[str]`, *optional*):</span>
+<span class="sd">                A list of keys in the output of your model (if it is a dictionary) that should be ignored when</span>
+<span class="sd">                gathering predictions.</span>
+<span class="sd">            metric_key_prefix (`str`, *optional*, defaults to `&quot;eval&quot;`):</span>
+<span class="sd">                An optional prefix to be used as the metrics key prefix. For example the metrics &quot;bleu&quot; will be named</span>
+<span class="sd">                &quot;eval_bleu&quot; if the prefix is &quot;eval&quot; (default)</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            A dictionary containing the evaluation loss and the potential metrics computed from the predictions. The</span>
+<span class="sd">            dictionary also contains the epoch number which comes from the training state.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># memory metrics - must set up as early as possible</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_memory_tracker</span><span class="o">.</span><span class="n">start</span><span class="p">()</span>
+
+        <span class="n">eval_dataloader</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_eval_dataloader</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">)</span>
+        <span class="n">start_time</span> <span class="o">=</span> <span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span>
+
+        <span class="n">eval_loop</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">prediction_loop</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">use_legacy_prediction_loop</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluation_loop</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">eval_loop</span><span class="p">(</span>
+            <span class="n">eval_dataloader</span><span class="p">,</span>
+            <span class="n">description</span><span class="o">=</span><span class="s2">&quot;Evaluation&quot;</span><span class="p">,</span>
+            <span class="c1"># No point gathering the predictions if there are no metrics, otherwise we defer to</span>
+            <span class="c1"># self.args.prediction_loss_only</span>
+            <span class="n">prediction_loss_only</span><span class="o">=</span><span class="kc">True</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_metrics</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="kc">None</span><span class="p">,</span>
+            <span class="n">ignore_keys</span><span class="o">=</span><span class="n">ignore_keys</span><span class="p">,</span>
+            <span class="n">metric_key_prefix</span><span class="o">=</span><span class="n">metric_key_prefix</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="n">total_batch_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">eval_batch_size</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span>
+        <span class="k">if</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">metric_key_prefix</span><span class="si">}</span><span class="s2">_jit_compilation_time&quot;</span> <span class="ow">in</span> <span class="n">output</span><span class="o">.</span><span class="n">metrics</span><span class="p">:</span>
+            <span class="n">start_time</span> <span class="o">+=</span> <span class="n">output</span><span class="o">.</span><span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">metric_key_prefix</span><span class="si">}</span><span class="s2">_jit_compilation_time&quot;</span><span class="p">]</span>
+        <span class="n">output</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">update</span><span class="p">(</span>
+            <span class="n">speed_metrics</span><span class="p">(</span>
+                <span class="n">metric_key_prefix</span><span class="p">,</span>
+                <span class="n">start_time</span><span class="p">,</span>
+                <span class="n">num_samples</span><span class="o">=</span><span class="n">output</span><span class="o">.</span><span class="n">num_samples</span><span class="p">,</span>
+                <span class="n">num_steps</span><span class="o">=</span><span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">num_samples</span> <span class="o">/</span> <span class="n">total_batch_size</span><span class="p">),</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">metrics</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">DebugOption</span><span class="o">.</span><span class="n">TPU_METRICS_DEBUG</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">debug</span><span class="p">:</span>
+            <span class="c1"># tpu-comment: Logging debug metrics for PyTorch/XLA (compile, execute times, ops, etc.)</span>
+            <span class="n">xm</span><span class="o">.</span><span class="n">master_print</span><span class="p">(</span><span class="n">met</span><span class="o">.</span><span class="n">metrics_report</span><span class="p">())</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_evaluate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">,</span> <span class="n">output</span><span class="o">.</span><span class="n">metrics</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">_memory_tracker</span><span class="o">.</span><span class="n">stop_and_update_metrics</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">metrics</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">output</span><span class="o">.</span><span class="n">metrics</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.predict">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.predict">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">predict</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">test_dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span> <span class="n">ignore_keys</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">metric_key_prefix</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;test&quot;</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">PredictionOutput</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Run prediction and returns predictions and potential metrics.</span>
+<span class="sd">        Depending on the dataset and your use case, your test dataset may contain labels. In that case, this method</span>
+<span class="sd">        will also return metrics, like in `evaluate()`.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            test_dataset (`Dataset`):</span>
+<span class="sd">                Dataset to run the predictions on. If it is an `datasets.Dataset`, columns not accepted by the</span>
+<span class="sd">                `model.forward()` method are automatically removed. Has to implement the method `__len__`</span>
+<span class="sd">            ignore_keys (`Lst[str]`, *optional*):</span>
+<span class="sd">                A list of keys in the output of your model (if it is a dictionary) that should be ignored when</span>
+<span class="sd">                gathering predictions.</span>
+<span class="sd">            metric_key_prefix (`str`, *optional*, defaults to `&quot;test&quot;`):</span>
+<span class="sd">                An optional prefix to be used as the metrics key prefix. For example the metrics &quot;bleu&quot; will be named</span>
+<span class="sd">                &quot;test_bleu&quot; if the prefix is &quot;test&quot; (default)</span>
+<span class="sd">        &lt;Tip&gt;</span>
+<span class="sd">        If your predictions or labels have different sequence length (for instance because you&#39;re doing dynamic padding</span>
+<span class="sd">        in a token classification task) the predictions will be padded (on the right) to allow for concatenation into</span>
+<span class="sd">        one array. The padding index is -100.</span>
+<span class="sd">        &lt;/Tip&gt;</span>
+<span class="sd">        Returns: *NamedTuple* A namedtuple with the following keys:</span>
+<span class="sd">            - predictions (`np.ndarray`): The predictions on `test_dataset`.</span>
+<span class="sd">            - label_ids (`np.ndarray`, *optional*): The labels (if the dataset contained some).</span>
+<span class="sd">            - metrics (`Dict[str, float]`, *optional*): The potential dictionary of metrics (if the dataset contained</span>
+<span class="sd">              labels).</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># memory metrics - must set up as early as possible</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_memory_tracker</span><span class="o">.</span><span class="n">start</span><span class="p">()</span>
+
+        <span class="n">test_dataloader</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">get_test_dataloader</span><span class="p">(</span><span class="n">test_dataset</span><span class="p">)</span>
+        <span class="n">start_time</span> <span class="o">=</span> <span class="n">time</span><span class="o">.</span><span class="n">time</span><span class="p">()</span>
+
+        <span class="n">eval_loop</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">prediction_loop</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">use_legacy_prediction_loop</span> <span class="k">else</span> <span class="bp">self</span><span class="o">.</span><span class="n">evaluation_loop</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">eval_loop</span><span class="p">(</span>
+            <span class="n">test_dataloader</span><span class="p">,</span> <span class="n">description</span><span class="o">=</span><span class="s2">&quot;Prediction&quot;</span><span class="p">,</span> <span class="n">ignore_keys</span><span class="o">=</span><span class="n">ignore_keys</span><span class="p">,</span> <span class="n">metric_key_prefix</span><span class="o">=</span><span class="n">metric_key_prefix</span>
+        <span class="p">)</span>
+        <span class="n">total_batch_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">eval_batch_size</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">world_size</span>
+        <span class="k">if</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">metric_key_prefix</span><span class="si">}</span><span class="s2">_jit_compilation_time&quot;</span> <span class="ow">in</span> <span class="n">output</span><span class="o">.</span><span class="n">metrics</span><span class="p">:</span>
+            <span class="n">start_time</span> <span class="o">+=</span> <span class="n">output</span><span class="o">.</span><span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">metric_key_prefix</span><span class="si">}</span><span class="s2">_jit_compilation_time&quot;</span><span class="p">]</span>
+        <span class="n">output</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">update</span><span class="p">(</span>
+            <span class="n">speed_metrics</span><span class="p">(</span>
+                <span class="n">metric_key_prefix</span><span class="p">,</span>
+                <span class="n">start_time</span><span class="p">,</span>
+                <span class="n">num_samples</span><span class="o">=</span><span class="n">output</span><span class="o">.</span><span class="n">num_samples</span><span class="p">,</span>
+                <span class="n">num_steps</span><span class="o">=</span><span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">num_samples</span> <span class="o">/</span> <span class="n">total_batch_size</span><span class="p">),</span>
+            <span class="p">)</span>
+        <span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_predict</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">,</span> <span class="n">output</span><span class="o">.</span><span class="n">metrics</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_memory_tracker</span><span class="o">.</span><span class="n">stop_and_update_metrics</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">metrics</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">PredictionOutput</span><span class="p">(</span><span class="n">predictions</span><span class="o">=</span><span class="n">output</span><span class="o">.</span><span class="n">predictions</span><span class="p">,</span> <span class="n">label_ids</span><span class="o">=</span><span class="n">output</span><span class="o">.</span><span class="n">label_ids</span><span class="p">,</span> <span class="n">metrics</span><span class="o">=</span><span class="n">output</span><span class="o">.</span><span class="n">metrics</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.evaluation_loop">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.evaluation_loop">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">evaluation_loop</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">dataloader</span><span class="p">:</span> <span class="n">DataLoader</span><span class="p">,</span>
+        <span class="n">description</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">prediction_loss_only</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">ignore_keys</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">metric_key_prefix</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;eval&quot;</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EvalLoopOutput</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Prediction/evaluation loop, shared by `Trainer.evaluate()` and `Trainer.predict()`.</span>
+<span class="sd">        Works both with or without labels.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span>
+
+        <span class="n">prediction_loss_only</span> <span class="o">=</span> <span class="n">prediction_loss_only</span> <span class="k">if</span> <span class="n">prediction_loss_only</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">args</span><span class="o">.</span><span class="n">prediction_loss_only</span>
+
+        <span class="c1"># if eval is called w/o train init deepspeed here</span>
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">deepspeed</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+            <span class="c1"># XXX: eval doesn&#39;t have `resume_from_checkpoint` arg but we should be able to do eval</span>
+            <span class="c1"># from the checkpoint eventually</span>
+            <span class="n">deepspeed_engine</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">deepspeed_init</span><span class="p">(</span>
+                <span class="bp">self</span><span class="p">,</span> <span class="n">num_training_steps</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">resume_from_checkpoint</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">inference</span><span class="o">=</span><span class="kc">True</span>
+            <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">deepspeed_engine</span><span class="o">.</span><span class="n">module</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span> <span class="o">=</span> <span class="n">deepspeed_engine</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span> <span class="o">=</span> <span class="n">deepspeed_engine</span>
+
+        <span class="n">model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_wrap_model</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">,</span> <span class="n">training</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">dataloader</span><span class="o">=</span><span class="n">dataloader</span><span class="p">)</span>
+
+        <span class="c1"># if full fp16 or bf16 eval is wanted and this ``evaluation`` or ``predict`` isn&#39;t called</span>
+        <span class="c1"># while ``train`` is running, cast it to the right dtype first and then put on device</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_in_train</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">fp16_full_eval</span><span class="p">:</span>
+                <span class="n">model</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float16</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">bf16_full_eval</span><span class="p">:</span>
+                <span class="n">model</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">bfloat16</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+
+        <span class="n">batch_size</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">eval_batch_size</span>
+
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;***** Running </span><span class="si">{</span><span class="n">description</span><span class="si">}</span><span class="s2"> *****&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">has_length</span><span class="p">(</span><span class="n">dataloader</span><span class="p">):</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;  Num examples = </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">num_examples</span><span class="p">(</span><span class="n">dataloader</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="s2">&quot;  Num examples: Unknown&quot;</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;  Batch size = </span><span class="si">{</span><span class="n">batch_size</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">eval_dataloader</span> <span class="o">=</span> <span class="n">dataloader</span>
+        <span class="c1"># Do this before wrapping.</span>
+        <span class="n">eval_dataset</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">dataloader</span><span class="p">,</span> <span class="s2">&quot;dataset&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+            <span class="n">dataloader</span> <span class="o">=</span> <span class="n">pl</span><span class="o">.</span><span class="n">ParallelLoader</span><span class="p">(</span><span class="n">dataloader</span><span class="p">,</span> <span class="p">[</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">])</span><span class="o">.</span><span class="n">per_device_loader</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">past_index</span> <span class="o">&gt;=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_past</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="c1"># Initialize containers</span>
+        <span class="c1"># losses/preds/labels on GPU/TPU (accumulated for eval_accumulation_steps)</span>
+        <span class="n">losses_host</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">preds_host</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">labels_host</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">inputs_host</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="c1"># losses/preds/labels on CPU (final containers)</span>
+        <span class="n">all_losses</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">all_preds</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">all_labels</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">all_inputs</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="c1"># Will be useful when we have an iterable dataset so don&#39;t know its length.</span>
+
+        <span class="n">observed_num_examples</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="c1"># Main evaluation loop</span>
+        <span class="k">for</span> <span class="n">step</span><span class="p">,</span> <span class="n">inputs</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">dataloader</span><span class="p">):</span>
+            <span class="c1"># Update the observed num examples</span>
+            <span class="n">observed_batch_size</span> <span class="o">=</span> <span class="n">find_batch_size</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">observed_batch_size</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">observed_num_examples</span> <span class="o">+=</span> <span class="n">observed_batch_size</span>
+                <span class="c1"># For batch samplers, batch_size is not known by the dataloader in advance.</span>
+                <span class="k">if</span> <span class="n">batch_size</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">batch_size</span> <span class="o">=</span> <span class="n">observed_batch_size</span>
+
+            <span class="c1"># Prediction step</span>
+            <span class="n">loss</span><span class="p">,</span> <span class="n">logits</span><span class="p">,</span> <span class="n">labels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">prediction_step</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">prediction_loss_only</span><span class="p">,</span> <span class="n">ignore_keys</span><span class="o">=</span><span class="n">ignore_keys</span><span class="p">)</span>
+            <span class="n">inputs_decode</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_prepare_input</span><span class="p">(</span><span class="n">inputs</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">])</span> <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">include_inputs_for_metrics</span> <span class="k">else</span> <span class="kc">None</span>
+
+            <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+                <span class="n">xm</span><span class="o">.</span><span class="n">mark_step</span><span class="p">()</span>
+
+            <span class="c1"># Update containers on host</span>
+            <span class="k">if</span> <span class="n">loss</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">losses</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_nested_gather</span><span class="p">(</span><span class="n">loss</span><span class="o">.</span><span class="n">repeat</span><span class="p">(</span><span class="n">batch_size</span><span class="p">))</span>
+                <span class="n">losses_host</span> <span class="o">=</span> <span class="n">losses</span> <span class="k">if</span> <span class="n">losses_host</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">losses_host</span><span class="p">,</span> <span class="n">losses</span><span class="p">),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">labels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_pad_across_processes</span><span class="p">(</span><span class="n">labels</span><span class="p">)</span>
+                <span class="n">labels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_nested_gather</span><span class="p">(</span><span class="n">labels</span><span class="p">)</span>
+                <span class="n">labels_host</span> <span class="o">=</span> <span class="n">labels</span> <span class="k">if</span> <span class="n">labels_host</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">nested_concat</span><span class="p">(</span><span class="n">labels_host</span><span class="p">,</span> <span class="n">labels</span><span class="p">,</span> <span class="n">padding_index</span><span class="o">=-</span><span class="mi">100</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">inputs_decode</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">inputs_decode</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_pad_across_processes</span><span class="p">(</span><span class="n">inputs_decode</span><span class="p">)</span>
+                <span class="n">inputs_decode</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_nested_gather</span><span class="p">(</span><span class="n">inputs_decode</span><span class="p">)</span>
+                <span class="n">inputs_host</span> <span class="o">=</span> <span class="p">(</span>
+                    <span class="n">inputs_decode</span>
+                    <span class="k">if</span> <span class="n">inputs_host</span> <span class="ow">is</span> <span class="kc">None</span>
+                    <span class="k">else</span> <span class="n">nested_concat</span><span class="p">(</span><span class="n">inputs_host</span><span class="p">,</span> <span class="n">inputs_decode</span><span class="p">,</span> <span class="n">padding_index</span><span class="o">=-</span><span class="mi">100</span><span class="p">)</span>
+                <span class="p">)</span>
+            <span class="k">if</span> <span class="n">logits</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">logits</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_pad_across_processes</span><span class="p">(</span><span class="n">logits</span><span class="p">)</span>
+                <span class="n">logits</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_nested_gather</span><span class="p">(</span><span class="n">logits</span><span class="p">)</span>
+                <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">preprocess_logits_for_metrics</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">logits</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">preprocess_logits_for_metrics</span><span class="p">(</span><span class="n">logits</span><span class="p">,</span> <span class="n">labels</span><span class="p">)</span>
+                <span class="n">preds_host</span> <span class="o">=</span> <span class="n">logits</span> <span class="k">if</span> <span class="n">preds_host</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">nested_concat</span><span class="p">(</span><span class="n">preds_host</span><span class="p">,</span> <span class="n">logits</span><span class="p">,</span> <span class="n">padding_index</span><span class="o">=-</span><span class="mi">100</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_prediction_step</span><span class="p">(</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">)</span>
+
+            <span class="c1"># Gather all tensors and put them back on the CPU if we have done enough accumulation steps.</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">eval_accumulation_steps</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="p">(</span><span class="n">step</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">%</span> <span class="n">args</span><span class="o">.</span><span class="n">eval_accumulation_steps</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">losses_host</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">losses</span> <span class="o">=</span> <span class="n">nested_numpify</span><span class="p">(</span><span class="n">losses_host</span><span class="p">)</span>
+                    <span class="n">all_losses</span> <span class="o">=</span> <span class="n">losses</span> <span class="k">if</span> <span class="n">all_losses</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">((</span><span class="n">all_losses</span><span class="p">,</span> <span class="n">losses</span><span class="p">),</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">preds_host</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">logits</span> <span class="o">=</span> <span class="n">nested_numpify</span><span class="p">(</span><span class="n">preds_host</span><span class="p">)</span>
+                    <span class="n">all_preds</span> <span class="o">=</span> <span class="n">logits</span> <span class="k">if</span> <span class="n">all_preds</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">nested_concat</span><span class="p">(</span><span class="n">all_preds</span><span class="p">,</span> <span class="n">logits</span><span class="p">,</span> <span class="n">padding_index</span><span class="o">=-</span><span class="mi">100</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">inputs_host</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">inputs_decode</span> <span class="o">=</span> <span class="n">nested_numpify</span><span class="p">(</span><span class="n">inputs_host</span><span class="p">)</span>
+                    <span class="n">all_inputs</span> <span class="o">=</span> <span class="p">(</span>
+                        <span class="n">inputs_decode</span>
+                        <span class="k">if</span> <span class="n">all_inputs</span> <span class="ow">is</span> <span class="kc">None</span>
+                        <span class="k">else</span> <span class="n">nested_concat</span><span class="p">(</span><span class="n">all_inputs</span><span class="p">,</span> <span class="n">inputs_decode</span><span class="p">,</span> <span class="n">padding_index</span><span class="o">=-</span><span class="mi">100</span><span class="p">)</span>
+                    <span class="p">)</span>
+                <span class="k">if</span> <span class="n">labels_host</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                    <span class="n">labels</span> <span class="o">=</span> <span class="n">nested_numpify</span><span class="p">(</span><span class="n">labels_host</span><span class="p">)</span>
+                    <span class="n">all_labels</span> <span class="o">=</span> <span class="p">(</span>
+                        <span class="n">labels</span> <span class="k">if</span> <span class="n">all_labels</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">nested_concat</span><span class="p">(</span><span class="n">all_labels</span><span class="p">,</span> <span class="n">labels</span><span class="p">,</span> <span class="n">padding_index</span><span class="o">=-</span><span class="mi">100</span><span class="p">)</span>
+                    <span class="p">)</span>
+
+                <span class="c1"># Set back to None to begin a new accumulation</span>
+                <span class="n">losses_host</span><span class="p">,</span> <span class="n">preds_host</span><span class="p">,</span> <span class="n">inputs_host</span><span class="p">,</span> <span class="n">labels_host</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span>
+
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">past_index</span> <span class="ow">and</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;_past&quot;</span><span class="p">):</span>
+            <span class="c1"># Clean the state at the end of the evaluation loop</span>
+            <span class="nb">delattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;_past&quot;</span><span class="p">)</span>
+
+        <span class="c1"># Gather all remaining tensors and put them back on the CPU</span>
+        <span class="k">if</span> <span class="n">losses_host</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">losses</span> <span class="o">=</span> <span class="n">nested_numpify</span><span class="p">(</span><span class="n">losses_host</span><span class="p">)</span>
+            <span class="n">all_losses</span> <span class="o">=</span> <span class="n">losses</span> <span class="k">if</span> <span class="n">all_losses</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">((</span><span class="n">all_losses</span><span class="p">,</span> <span class="n">losses</span><span class="p">),</span> <span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">preds_host</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">logits</span> <span class="o">=</span> <span class="n">nested_numpify</span><span class="p">(</span><span class="n">preds_host</span><span class="p">)</span>
+            <span class="n">all_preds</span> <span class="o">=</span> <span class="n">logits</span> <span class="k">if</span> <span class="n">all_preds</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">nested_concat</span><span class="p">(</span><span class="n">all_preds</span><span class="p">,</span> <span class="n">logits</span><span class="p">,</span> <span class="n">padding_index</span><span class="o">=-</span><span class="mi">100</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">inputs_host</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">inputs_decode</span> <span class="o">=</span> <span class="n">nested_numpify</span><span class="p">(</span><span class="n">inputs_host</span><span class="p">)</span>
+            <span class="n">all_inputs</span> <span class="o">=</span> <span class="p">(</span>
+                <span class="n">inputs_decode</span> <span class="k">if</span> <span class="n">all_inputs</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">nested_concat</span><span class="p">(</span><span class="n">all_inputs</span><span class="p">,</span> <span class="n">inputs_decode</span><span class="p">,</span> <span class="n">padding_index</span><span class="o">=-</span><span class="mi">100</span><span class="p">)</span>
+            <span class="p">)</span>
+        <span class="k">if</span> <span class="n">labels_host</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">labels</span> <span class="o">=</span> <span class="n">nested_numpify</span><span class="p">(</span><span class="n">labels_host</span><span class="p">)</span>
+            <span class="n">all_labels</span> <span class="o">=</span> <span class="n">labels</span> <span class="k">if</span> <span class="n">all_labels</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">nested_concat</span><span class="p">(</span><span class="n">all_labels</span><span class="p">,</span> <span class="n">labels</span><span class="p">,</span> <span class="n">padding_index</span><span class="o">=-</span><span class="mi">100</span><span class="p">)</span>
+
+        <span class="c1"># Number of samples</span>
+        <span class="k">if</span> <span class="n">has_length</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">):</span>
+            <span class="n">num_samples</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">)</span>
+        <span class="c1"># The instance check is weird and does not actually check for the type, but whether the dataset has the right</span>
+        <span class="c1"># methods. Therefore we need to make sure it also has the attribute.</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">,</span> <span class="n">IterableDatasetShard</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">eval_dataset</span><span class="p">,</span> <span class="s2">&quot;num_examples&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">num_samples</span> <span class="o">=</span> <span class="n">eval_dataset</span><span class="o">.</span><span class="n">num_examples</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">has_length</span><span class="p">(</span><span class="n">dataloader</span><span class="p">):</span>
+                <span class="n">num_samples</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_examples</span><span class="p">(</span><span class="n">dataloader</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>  <span class="c1"># both len(dataloader.dataset) and len(dataloader) fail</span>
+                <span class="n">num_samples</span> <span class="o">=</span> <span class="n">observed_num_examples</span>
+        <span class="k">if</span> <span class="n">num_samples</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">observed_num_examples</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">num_samples</span> <span class="o">=</span> <span class="n">observed_num_examples</span>
+
+        <span class="c1"># Number of losses has been rounded to a multiple of batch_size and in a distributed training, the number of</span>
+        <span class="c1"># samplers has been rounded to a multiple of batch_size, so we truncate.</span>
+        <span class="k">if</span> <span class="n">all_losses</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">all_losses</span> <span class="o">=</span> <span class="n">all_losses</span><span class="p">[:</span><span class="n">num_samples</span><span class="p">]</span>
+        <span class="k">if</span> <span class="n">all_preds</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">all_preds</span> <span class="o">=</span> <span class="n">nested_truncate</span><span class="p">(</span><span class="n">all_preds</span><span class="p">,</span> <span class="n">num_samples</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">all_labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">all_labels</span> <span class="o">=</span> <span class="n">nested_truncate</span><span class="p">(</span><span class="n">all_labels</span><span class="p">,</span> <span class="n">num_samples</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">all_inputs</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">all_inputs</span> <span class="o">=</span> <span class="n">nested_truncate</span><span class="p">(</span><span class="n">all_inputs</span><span class="p">,</span> <span class="n">num_samples</span><span class="p">)</span>
+
+        <span class="c1"># Metrics!</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_metrics</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">all_preds</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">all_labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">include_inputs_for_metrics</span><span class="p">:</span>
+                <span class="n">metrics</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_metrics</span><span class="p">(</span>
+                    <span class="n">EvalPrediction</span><span class="p">(</span><span class="n">predictions</span><span class="o">=</span><span class="n">all_preds</span><span class="p">,</span> <span class="n">label_ids</span><span class="o">=</span><span class="n">all_labels</span><span class="p">,</span> <span class="n">inputs</span><span class="o">=</span><span class="n">all_inputs</span><span class="p">)</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">metrics</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_metrics</span><span class="p">(</span><span class="n">EvalPrediction</span><span class="p">(</span><span class="n">predictions</span><span class="o">=</span><span class="n">all_preds</span><span class="p">,</span> <span class="n">label_ids</span><span class="o">=</span><span class="n">all_labels</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">metrics</span> <span class="o">=</span> <span class="p">{}</span>
+
+        <span class="c1"># To be JSON-serializable, we need to remove numpy types or zero-d tensors</span>
+        <span class="n">metrics</span> <span class="o">=</span> <span class="n">denumpify_detensorize</span><span class="p">(</span><span class="n">metrics</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">all_losses</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">metric_key_prefix</span><span class="si">}</span><span class="s2">_loss&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">all_losses</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+        <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;jit_compilation_time&quot;</span><span class="p">):</span>
+            <span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">metric_key_prefix</span><span class="si">}</span><span class="s2">_jit_compilation_time&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">jit_compilation_time</span>
+
+        <span class="c1"># Prefix all keys with metric_key_prefix + &#39;_&#39;</span>
+        <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="nb">list</span><span class="p">(</span><span class="n">metrics</span><span class="o">.</span><span class="n">keys</span><span class="p">()):</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">key</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">metric_key_prefix</span><span class="si">}</span><span class="s2">_&quot;</span><span class="p">):</span>
+                <span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">metric_key_prefix</span><span class="si">}</span><span class="s2">_</span><span class="si">{</span><span class="n">key</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">metrics</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="n">key</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">EvalLoopOutput</span><span class="p">(</span><span class="n">predictions</span><span class="o">=</span><span class="n">all_preds</span><span class="p">,</span> <span class="n">label_ids</span><span class="o">=</span><span class="n">all_labels</span><span class="p">,</span> <span class="n">metrics</span><span class="o">=</span><span class="n">metrics</span><span class="p">,</span> <span class="n">num_samples</span><span class="o">=</span><span class="n">num_samples</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._nested_gather">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._nested_gather">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_nested_gather</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tensors</span><span class="p">,</span> <span class="n">name</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Gather value of `tensors` (tensor or list/tuple of nested tensors) and convert them to numpy before</span>
+<span class="sd">        concatenating them to `gathered`</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">tensors</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span>
+        <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+            <span class="k">if</span> <span class="n">name</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">name</span> <span class="o">=</span> <span class="s2">&quot;nested_gather&quot;</span>
+            <span class="n">tensors</span> <span class="o">=</span> <span class="n">nested_xla_mesh_reduce</span><span class="p">(</span><span class="n">tensors</span><span class="p">,</span> <span class="n">name</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+            <span class="n">tensors</span> <span class="o">=</span> <span class="n">smp_gather</span><span class="p">(</span><span class="n">tensors</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">!=</span> <span class="o">-</span><span class="mi">1</span><span class="p">:</span>
+            <span class="n">tensors</span> <span class="o">=</span> <span class="n">distributed_concat</span><span class="p">(</span><span class="n">tensors</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">tensors</span></div>
+
+
+    <span class="c1"># Copied from Accelerate.</span>
+<div class="viewcode-block" id="RaftTrainer._pad_across_processes">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._pad_across_processes">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_pad_across_processes</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tensor</span><span class="p">,</span> <span class="n">pad_index</span><span class="o">=-</span><span class="mi">100</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Recursively pad the tensors in a nested list/tuple/dictionary of tensors from all devices to the same size so</span>
+<span class="sd">        they can safely be gathered.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="p">(</span><span class="nb">list</span><span class="p">,</span> <span class="nb">tuple</span><span class="p">)):</span>
+            <span class="k">return</span> <span class="nb">type</span><span class="p">(</span><span class="n">tensor</span><span class="p">)(</span><span class="bp">self</span><span class="o">.</span><span class="n">_pad_across_processes</span><span class="p">(</span><span class="n">t</span><span class="p">,</span> <span class="n">pad_index</span><span class="o">=</span><span class="n">pad_index</span><span class="p">)</span> <span class="k">for</span> <span class="n">t</span> <span class="ow">in</span> <span class="n">tensor</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="nb">dict</span><span class="p">):</span>
+            <span class="k">return</span> <span class="nb">type</span><span class="p">(</span><span class="n">tensor</span><span class="p">)({</span><span class="n">k</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">_pad_across_processes</span><span class="p">(</span><span class="n">v</span><span class="p">,</span> <span class="n">pad_index</span><span class="o">=</span><span class="n">pad_index</span><span class="p">)</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">tensor</span><span class="o">.</span><span class="n">items</span><span class="p">()})</span>
+        <span class="k">elif</span> <span class="ow">not</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">TypeError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Can&#39;t pad the values of type </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">tensor</span><span class="p">)</span><span class="si">}</span><span class="s2">, only of nested list/tuple/dicts of tensors.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">tensor</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> <span class="o">&lt;</span> <span class="mi">2</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">tensor</span>
+        <span class="c1"># Gather all sizes</span>
+        <span class="n">size</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="n">tensor</span><span class="o">.</span><span class="n">shape</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">tensor</span><span class="o">.</span><span class="n">device</span><span class="p">)[</span><span class="kc">None</span><span class="p">]</span>
+        <span class="n">sizes</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_nested_gather</span><span class="p">(</span><span class="n">size</span><span class="p">)</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span>
+
+        <span class="n">max_size</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">s</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">sizes</span><span class="p">)</span>
+        <span class="c1"># When extracting XLA graphs for compilation, max_size is 0,</span>
+        <span class="c1"># so use inequality to avoid errors.</span>
+        <span class="k">if</span> <span class="n">tensor</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">&gt;=</span> <span class="n">max_size</span><span class="p">:</span>
+            <span class="k">return</span> <span class="n">tensor</span>
+
+        <span class="c1"># Then pad to the maximum size</span>
+        <span class="n">old_size</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">shape</span>
+        <span class="n">new_size</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">old_size</span><span class="p">)</span>
+        <span class="n">new_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="n">max_size</span>
+        <span class="n">new_tensor</span> <span class="o">=</span> <span class="n">tensor</span><span class="o">.</span><span class="n">new_zeros</span><span class="p">(</span><span class="nb">tuple</span><span class="p">(</span><span class="n">new_size</span><span class="p">))</span> <span class="o">+</span> <span class="n">pad_index</span>
+        <span class="n">new_tensor</span><span class="p">[:,</span> <span class="p">:</span> <span class="n">old_size</span><span class="p">[</span><span class="mi">1</span><span class="p">]]</span> <span class="o">=</span> <span class="n">tensor</span>
+        <span class="k">return</span> <span class="n">new_tensor</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.prediction_step">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.prediction_step">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">prediction_step</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">,</span>
+        <span class="n">inputs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Any</span><span class="p">]],</span>
+        <span class="n">prediction_loss_only</span><span class="p">:</span> <span class="nb">bool</span><span class="p">,</span>
+        <span class="n">ignore_keys</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Perform an evaluation step on `model` using `inputs`.</span>
+<span class="sd">        Subclass and override to inject custom behavior.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            model (`nn.Module`):</span>
+<span class="sd">                The model to evaluate.</span>
+<span class="sd">            inputs (`Dict[str, Union[torch.Tensor, Any]]`):</span>
+<span class="sd">                The inputs and targets of the model.</span>
+<span class="sd">                The dictionary will be unpacked before being fed to the model. Most models expect the targets under the</span>
+<span class="sd">                argument `labels`. Check your model&#39;s documentation for all accepted arguments.</span>
+<span class="sd">            prediction_loss_only (`bool`):</span>
+<span class="sd">                Whether or not to return the loss only.</span>
+<span class="sd">            ignore_keys (`Lst[str]`, *optional*):</span>
+<span class="sd">                A list of keys in the output of your model (if it is a dictionary) that should be ignored when</span>
+<span class="sd">                gathering predictions.</span>
+<span class="sd">        Return:</span>
+<span class="sd">            Tuple[Optional[torch.Tensor], Optional[torch.Tensor], Optional[torch.Tensor]]: A tuple with the loss,</span>
+<span class="sd">            logits and labels (each being optional).</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">has_labels</span> <span class="o">=</span> <span class="kc">False</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">label_names</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="nb">all</span><span class="p">(</span><span class="n">inputs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">k</span><span class="p">)</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">label_names</span><span class="p">)</span>
+        <span class="c1"># For CLIP-like models capable of returning loss values.</span>
+        <span class="c1"># If `return_loss` is not specified or being `None` in `inputs`, we check if the default value of `return_loss`</span>
+        <span class="c1"># is `True` in `model.forward`.</span>
+        <span class="n">return_loss</span> <span class="o">=</span> <span class="n">inputs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;return_loss&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">return_loss</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">return_loss</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">can_return_loss</span>
+        <span class="n">loss_without_labels</span> <span class="o">=</span> <span class="kc">True</span> <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">label_names</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">return_loss</span> <span class="k">else</span> <span class="kc">False</span>
+
+        <span class="n">inputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_prepare_inputs</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">ignore_keys</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">,</span> <span class="s2">&quot;config&quot;</span><span class="p">):</span>
+                <span class="n">ignore_keys</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">config</span><span class="p">,</span> <span class="s2">&quot;keys_to_ignore_at_inference&quot;</span><span class="p">,</span> <span class="p">[])</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">ignore_keys</span> <span class="o">=</span> <span class="p">[]</span>
+
+        <span class="c1"># labels may be popped when computing the loss (label smoothing for instance) so we grab them first.</span>
+        <span class="k">if</span> <span class="n">has_labels</span> <span class="ow">or</span> <span class="n">loss_without_labels</span><span class="p">:</span>
+            <span class="n">labels</span> <span class="o">=</span> <span class="n">nested_detach</span><span class="p">(</span><span class="nb">tuple</span><span class="p">(</span><span class="n">inputs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="n">name</span><span class="p">)</span> <span class="k">for</span> <span class="n">name</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">label_names</span><span class="p">))</span>
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">labels</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+                <span class="n">labels</span> <span class="o">=</span> <span class="n">labels</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">labels</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">no_grad</span><span class="p">():</span>
+            <span class="k">if</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+                <span class="n">raw_outputs</span> <span class="o">=</span> <span class="n">smp_forward_only</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">has_labels</span> <span class="ow">or</span> <span class="n">loss_without_labels</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">raw_outputs</span><span class="p">,</span> <span class="nb">dict</span><span class="p">):</span>
+                        <span class="n">loss_mb</span> <span class="o">=</span> <span class="n">raw_outputs</span><span class="p">[</span><span class="s2">&quot;loss&quot;</span><span class="p">]</span>
+                        <span class="n">logits_mb</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">v</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">raw_outputs</span><span class="o">.</span><span class="n">items</span><span class="p">()</span> <span class="k">if</span> <span class="n">k</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">ignore_keys</span> <span class="o">+</span> <span class="p">[</span><span class="s2">&quot;loss&quot;</span><span class="p">])</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">loss_mb</span> <span class="o">=</span> <span class="n">raw_outputs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+                        <span class="n">logits_mb</span> <span class="o">=</span> <span class="n">raw_outputs</span><span class="p">[</span><span class="mi">1</span><span class="p">:]</span>
+
+                    <span class="n">loss</span> <span class="o">=</span> <span class="n">loss_mb</span><span class="o">.</span><span class="n">reduce_mean</span><span class="p">()</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span><span class="o">.</span><span class="n">cpu</span><span class="p">()</span>
+                    <span class="n">logits</span> <span class="o">=</span> <span class="n">smp_nested_concat</span><span class="p">(</span><span class="n">logits_mb</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+                    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">raw_outputs</span><span class="p">,</span> <span class="nb">dict</span><span class="p">):</span>
+                        <span class="n">logits_mb</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">v</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">raw_outputs</span><span class="o">.</span><span class="n">items</span><span class="p">()</span> <span class="k">if</span> <span class="n">k</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">ignore_keys</span><span class="p">)</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">logits_mb</span> <span class="o">=</span> <span class="n">raw_outputs</span>
+                    <span class="n">logits</span> <span class="o">=</span> <span class="n">smp_nested_concat</span><span class="p">(</span><span class="n">logits_mb</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">has_labels</span> <span class="ow">or</span> <span class="n">loss_without_labels</span><span class="p">:</span>
+                    <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_loss_context_manager</span><span class="p">():</span>
+                        <span class="n">loss</span><span class="p">,</span> <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_loss</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">return_outputs</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+                    <span class="n">loss</span> <span class="o">=</span> <span class="n">loss</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span>
+
+                    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="nb">dict</span><span class="p">):</span>
+                        <span class="n">logits</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">v</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">outputs</span><span class="o">.</span><span class="n">items</span><span class="p">()</span> <span class="k">if</span> <span class="n">k</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">ignore_keys</span> <span class="o">+</span> <span class="p">[</span><span class="s2">&quot;loss&quot;</span><span class="p">])</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">logits</span> <span class="o">=</span> <span class="n">outputs</span><span class="p">[</span><span class="mi">1</span><span class="p">:]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">loss</span> <span class="o">=</span> <span class="kc">None</span>
+                    <span class="k">with</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_loss_context_manager</span><span class="p">():</span>
+                        <span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span><span class="o">**</span><span class="n">inputs</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="nb">dict</span><span class="p">):</span>
+                        <span class="n">logits</span> <span class="o">=</span> <span class="nb">tuple</span><span class="p">(</span><span class="n">v</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">outputs</span><span class="o">.</span><span class="n">items</span><span class="p">()</span> <span class="k">if</span> <span class="n">k</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">ignore_keys</span><span class="p">)</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">logits</span> <span class="o">=</span> <span class="n">outputs</span>
+                    <span class="c1"># TODO: this needs to be fixed and made cleaner later.</span>
+                    <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">past_index</span> <span class="o">&gt;=</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="bp">self</span><span class="o">.</span><span class="n">_past</span> <span class="o">=</span> <span class="n">outputs</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">past_index</span> <span class="o">-</span> <span class="mi">1</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="n">prediction_loss_only</span><span class="p">:</span>
+            <span class="k">return</span> <span class="p">(</span><span class="n">loss</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+
+        <span class="n">logits</span> <span class="o">=</span> <span class="n">nested_detach</span><span class="p">(</span><span class="n">logits</span><span class="p">)</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">logits</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">logits</span> <span class="o">=</span> <span class="n">logits</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+
+        <span class="k">return</span> <span class="p">(</span><span class="n">loss</span><span class="p">,</span> <span class="n">logits</span><span class="p">,</span> <span class="n">labels</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.floating_point_ops">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.floating_point_ops">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">floating_point_ops</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inputs</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Any</span><span class="p">]]):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        For models that inherit from [`PreTrainedModel`], uses that method to compute the number of floating point</span>
+<span class="sd">        operations for every backward + forward pass. If using another model, either implement such a method in the</span>
+<span class="sd">        model or subclass and override this method.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            inputs (`Dict[str, Union[torch.Tensor, Any]]`):</span>
+<span class="sd">                The inputs and targets of the model.</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            `int`: The number of floating-point operations.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">,</span> <span class="s2">&quot;floating_point_ops&quot;</span><span class="p">):</span>
+            <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">floating_point_ops</span><span class="p">(</span><span class="n">inputs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">return</span> <span class="mi">0</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.init_git_repo">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.init_git_repo">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">init_git_repo</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">at_init</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Initializes a git repo in `self.args.hub_model_id`.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            at_init (`bool`, *optional*, defaults to `False`):</span>
+<span class="sd">                Whether this function is called before any training or not. If `self.args.overwrite_output_dir` is</span>
+<span class="sd">                `True` and `at_init` is `True`, the path to the repo (which is `self.args.output_dir`) might be wiped</span>
+<span class="sd">                out.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_world_process_zero</span><span class="p">():</span>
+            <span class="k">return</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hub_model_id</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">repo_name</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">)</span><span class="o">.</span><span class="n">absolute</span><span class="p">()</span><span class="o">.</span><span class="n">name</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">repo_name</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hub_model_id</span>
+        <span class="k">if</span> <span class="s2">&quot;/&quot;</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">repo_name</span><span class="p">:</span>
+            <span class="n">repo_name</span> <span class="o">=</span> <span class="n">get_full_repo_name</span><span class="p">(</span><span class="n">repo_name</span><span class="p">,</span> <span class="n">token</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hub_token</span><span class="p">)</span>
+
+        <span class="c1"># Make sure the repo exists.</span>
+        <span class="n">create_repo</span><span class="p">(</span><span class="n">repo_name</span><span class="p">,</span> <span class="n">token</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hub_token</span><span class="p">,</span> <span class="n">private</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hub_private_repo</span><span class="p">,</span> <span class="n">exist_ok</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="k">try</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">repo</span> <span class="o">=</span> <span class="n">Repository</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">clone_from</span><span class="o">=</span><span class="n">repo_name</span><span class="p">,</span> <span class="n">token</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hub_token</span><span class="p">)</span>
+        <span class="k">except</span> <span class="ne">EnvironmentError</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">overwrite_output_dir</span> <span class="ow">and</span> <span class="n">at_init</span><span class="p">:</span>
+                <span class="c1"># Try again after wiping output_dir</span>
+                <span class="n">shutil</span><span class="o">.</span><span class="n">rmtree</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">repo</span> <span class="o">=</span> <span class="n">Repository</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">clone_from</span><span class="o">=</span><span class="n">repo_name</span><span class="p">,</span> <span class="n">token</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hub_token</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">repo</span><span class="o">.</span><span class="n">git_pull</span><span class="p">()</span>
+
+        <span class="c1"># By default, ignore the checkpoint folders</span>
+        <span class="k">if</span> <span class="p">(</span>
+            <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span> <span class="s2">&quot;.gitignore&quot;</span><span class="p">))</span>
+            <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hub_strategy</span> <span class="o">!=</span> <span class="n">HubStrategy</span><span class="o">.</span><span class="n">ALL_CHECKPOINTS</span>
+        <span class="p">):</span>
+            <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span> <span class="s2">&quot;.gitignore&quot;</span><span class="p">),</span> <span class="s2">&quot;w&quot;</span><span class="p">,</span> <span class="n">encoding</span><span class="o">=</span><span class="s2">&quot;utf-8&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">writer</span><span class="p">:</span>
+                <span class="n">writer</span><span class="o">.</span><span class="n">writelines</span><span class="p">([</span><span class="s2">&quot;checkpoint-*/&quot;</span><span class="p">])</span>
+
+        <span class="c1"># Add &quot;*.sagemaker&quot; to .gitignore if using SageMaker</span>
+        <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;SM_TRAINING_ENV&quot;</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_add_sm_patterns_to_gitignore</span><span class="p">()</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">push_in_progress</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.create_model_card">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_model_card">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">create_model_card</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">language</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">license</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">tags</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="kc">None</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">model_name</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">finetuned_from</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">tasks</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="kc">None</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">dataset_tags</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="kc">None</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="kc">None</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">dataset_args</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="kc">None</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Creates a draft of a model card using the information available to the `Trainer`.</span>
+<span class="sd">        Args:</span>
+<span class="sd">            language (`str`, *optional*):</span>
+<span class="sd">                The language of the model (if applicable)</span>
+<span class="sd">            license (`str`, *optional*):</span>
+<span class="sd">                The license of the model. Will default to the license of the pretrained model used, if the original</span>
+<span class="sd">                model given to the `Trainer` comes from a repo on the Hub.</span>
+<span class="sd">            tags (`str` or `List[str]`, *optional*):</span>
+<span class="sd">                Some tags to be included in the metadata of the model card.</span>
+<span class="sd">            model_name (`str`, *optional*):</span>
+<span class="sd">                The name of the model.</span>
+<span class="sd">            finetuned_from (`str`, *optional*):</span>
+<span class="sd">                The name of the model used to fine-tune this one (if applicable). Will default to the name of the repo</span>
+<span class="sd">                of the original model given to the `Trainer` (if it comes from the Hub).</span>
+<span class="sd">            tasks (`str` or `List[str]`, *optional*):</span>
+<span class="sd">                One or several task identifiers, to be included in the metadata of the model card.</span>
+<span class="sd">            dataset_tags (`str` or `List[str]`, *optional*):</span>
+<span class="sd">                One or several dataset tags, to be included in the metadata of the model card.</span>
+<span class="sd">            dataset (`str` or `List[str]`, *optional*):</span>
+<span class="sd">                One or several dataset identifiers, to be included in the metadata of the model card.</span>
+<span class="sd">            dataset_args (`str` or `List[str]`, *optional*):</span>
+<span class="sd">               One or several dataset arguments, to be included in the metadata of the model card.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_world_process_zero</span><span class="p">():</span>
+            <span class="k">return</span>
+
+        <span class="n">training_summary</span> <span class="o">=</span> <span class="n">TrainingSummary</span><span class="o">.</span><span class="n">from_trainer</span><span class="p">(</span>
+            <span class="bp">self</span><span class="p">,</span>
+            <span class="n">language</span><span class="o">=</span><span class="n">language</span><span class="p">,</span>
+            <span class="n">license</span><span class="o">=</span><span class="n">license</span><span class="p">,</span>
+            <span class="n">tags</span><span class="o">=</span><span class="n">tags</span><span class="p">,</span>
+            <span class="n">model_name</span><span class="o">=</span><span class="n">model_name</span><span class="p">,</span>
+            <span class="n">finetuned_from</span><span class="o">=</span><span class="n">finetuned_from</span><span class="p">,</span>
+            <span class="n">tasks</span><span class="o">=</span><span class="n">tasks</span><span class="p">,</span>
+            <span class="n">dataset_tags</span><span class="o">=</span><span class="n">dataset_tags</span><span class="p">,</span>
+            <span class="n">dataset</span><span class="o">=</span><span class="n">dataset</span><span class="p">,</span>
+            <span class="n">dataset_args</span><span class="o">=</span><span class="n">dataset_args</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">model_card</span> <span class="o">=</span> <span class="n">training_summary</span><span class="o">.</span><span class="n">to_model_card</span><span class="p">()</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">,</span> <span class="s2">&quot;README.md&quot;</span><span class="p">),</span> <span class="s2">&quot;w&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">f</span><span class="o">.</span><span class="n">write</span><span class="p">(</span><span class="n">model_card</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._push_from_checkpoint">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._push_from_checkpoint">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_push_from_checkpoint</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">checkpoint_folder</span><span class="p">):</span>
+        <span class="c1"># Only push from one node.</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_world_process_zero</span><span class="p">()</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hub_strategy</span> <span class="o">==</span> <span class="n">HubStrategy</span><span class="o">.</span><span class="n">END</span><span class="p">:</span>
+            <span class="k">return</span>
+        <span class="c1"># If we haven&#39;t finished the last push, we don&#39;t do this one.</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">push_in_progress</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">push_in_progress</span><span class="o">.</span><span class="n">is_done</span><span class="p">:</span>
+            <span class="k">return</span>
+
+        <span class="n">output_dir</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span>
+        <span class="c1"># To avoid a new synchronization of all model weights, we just copy the file from the checkpoint folder</span>
+        <span class="n">modeling_files</span> <span class="o">=</span> <span class="p">[</span><span class="n">CONFIG_NAME</span><span class="p">,</span> <span class="n">WEIGHTS_NAME</span><span class="p">]</span>
+        <span class="k">for</span> <span class="n">modeling_file</span> <span class="ow">in</span> <span class="n">modeling_files</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isfile</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint_folder</span><span class="p">,</span> <span class="n">modeling_file</span><span class="p">)):</span>
+                <span class="n">shutil</span><span class="o">.</span><span class="n">copy</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">checkpoint_folder</span><span class="p">,</span> <span class="n">modeling_file</span><span class="p">),</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">modeling_file</span><span class="p">))</span>
+        <span class="c1"># Saving the tokenizer is fast and we don&#39;t know how many files it may have spawned, so we resave it to be sure.</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">save_pretrained</span><span class="p">(</span><span class="n">output_dir</span><span class="p">)</span>
+        <span class="c1"># Same for the training arguments</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="p">,</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="n">TRAINING_ARGS_NAME</span><span class="p">))</span>
+
+        <span class="k">try</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hub_strategy</span> <span class="o">==</span> <span class="n">HubStrategy</span><span class="o">.</span><span class="n">CHECKPOINT</span><span class="p">:</span>
+                <span class="c1"># Temporarily move the checkpoint just saved for the push</span>
+                <span class="n">tmp_checkpoint</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">output_dir</span><span class="p">,</span> <span class="s2">&quot;last-checkpoint&quot;</span><span class="p">)</span>
+                <span class="c1"># We have to remove the &quot;last-checkpoint&quot; dir if it exists, otherwise the checkpoint is moved as a</span>
+                <span class="c1"># subfolder.</span>
+                <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">isdir</span><span class="p">(</span><span class="n">tmp_checkpoint</span><span class="p">):</span>
+                    <span class="n">shutil</span><span class="o">.</span><span class="n">rmtree</span><span class="p">(</span><span class="n">tmp_checkpoint</span><span class="p">)</span>
+                <span class="n">shutil</span><span class="o">.</span><span class="n">move</span><span class="p">(</span><span class="n">checkpoint_folder</span><span class="p">,</span> <span class="n">tmp_checkpoint</span><span class="p">)</span>
+
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">save_strategy</span> <span class="o">==</span> <span class="n">IntervalStrategy</span><span class="o">.</span><span class="n">STEPS</span><span class="p">:</span>
+                <span class="n">commit_message</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;Training in progress, step </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">global_step</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">commit_message</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;Training in progress, epoch </span><span class="si">{</span><span class="nb">int</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="o">.</span><span class="n">epoch</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+            <span class="n">_</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">push_in_progress</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">repo</span><span class="o">.</span><span class="n">push_to_hub</span><span class="p">(</span>
+                <span class="n">commit_message</span><span class="o">=</span><span class="n">commit_message</span><span class="p">,</span> <span class="n">blocking</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">auto_lfs_prune</span><span class="o">=</span><span class="kc">True</span>
+            <span class="p">)</span>
+        <span class="k">finally</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hub_strategy</span> <span class="o">==</span> <span class="n">HubStrategy</span><span class="o">.</span><span class="n">CHECKPOINT</span><span class="p">:</span>
+                <span class="c1"># Move back the checkpoint to its place</span>
+                <span class="n">shutil</span><span class="o">.</span><span class="n">move</span><span class="p">(</span><span class="n">tmp_checkpoint</span><span class="p">,</span> <span class="n">checkpoint_folder</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer.push_to_hub">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.push_to_hub">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">push_to_hub</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">commit_message</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;End of training&quot;</span><span class="p">,</span> <span class="n">blocking</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Upload *self.model* and *self.tokenizer* to the 🤗 model hub on the repo *self.args.hub_model_id*.</span>
+<span class="sd">        Parameters:</span>
+<span class="sd">            commit_message (`str`, *optional*, defaults to `&quot;End of training&quot;`):</span>
+<span class="sd">                Message to commit while pushing.</span>
+<span class="sd">            blocking (`bool`, *optional*, defaults to `True`):</span>
+<span class="sd">                Whether the function should return only when the `git push` has finished.</span>
+<span class="sd">            kwargs:</span>
+<span class="sd">                Additional keyword arguments passed along to [`~Trainer.create_model_card`].</span>
+<span class="sd">        Returns:</span>
+<span class="sd">            The url of the commit of your model in the given repository if `blocking=False`, a tuple with the url of</span>
+<span class="sd">            the commit and an object to track the progress of the commit if `blocking=True`</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># If a user calls manually `push_to_hub` with `self.args.push_to_hub = False`, we try to create the repo but</span>
+        <span class="c1"># it might fail.</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;repo&quot;</span><span class="p">):</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">init_git_repo</span><span class="p">()</span>
+
+        <span class="n">model_name</span> <span class="o">=</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s2">&quot;model_name&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">model_name</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hub_model_id</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">model_name</span> <span class="o">=</span> <span class="n">Path</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">output_dir</span><span class="p">)</span><span class="o">.</span><span class="n">name</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">model_name</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">hub_model_id</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;/&quot;</span><span class="p">)[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+
+        <span class="c1"># Needs to be executed on all processes for TPU training, but will only save on the processed determined by</span>
+        <span class="c1"># self.args.should_save.</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">save_model</span><span class="p">(</span><span class="n">_internal_call</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+        <span class="c1"># Only push from one node.</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_world_process_zero</span><span class="p">():</span>
+            <span class="k">return</span>
+
+        <span class="c1"># Cancel any async push in progress if blocking=True. The commits will all be pushed together.</span>
+        <span class="k">if</span> <span class="n">blocking</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">push_in_progress</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">push_in_progress</span><span class="o">.</span><span class="n">is_done</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">push_in_progress</span><span class="o">.</span><span class="n">_process</span><span class="o">.</span><span class="n">kill</span><span class="p">()</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">push_in_progress</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="n">git_head_commit_url</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">repo</span><span class="o">.</span><span class="n">push_to_hub</span><span class="p">(</span>
+            <span class="n">commit_message</span><span class="o">=</span><span class="n">commit_message</span><span class="p">,</span> <span class="n">blocking</span><span class="o">=</span><span class="n">blocking</span><span class="p">,</span> <span class="n">auto_lfs_prune</span><span class="o">=</span><span class="kc">True</span>
+        <span class="p">)</span>
+        <span class="c1"># push separately the model card to be independant from the rest of the model</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">should_save</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">create_model_card</span><span class="p">(</span><span class="n">model_name</span><span class="o">=</span><span class="n">model_name</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+            <span class="k">try</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">repo</span><span class="o">.</span><span class="n">push_to_hub</span><span class="p">(</span>
+                    <span class="n">commit_message</span><span class="o">=</span><span class="s2">&quot;update model card README.md&quot;</span><span class="p">,</span> <span class="n">blocking</span><span class="o">=</span><span class="n">blocking</span><span class="p">,</span> <span class="n">auto_lfs_prune</span><span class="o">=</span><span class="kc">True</span>
+                <span class="p">)</span>
+            <span class="k">except</span> <span class="ne">EnvironmentError</span> <span class="k">as</span> <span class="n">exc</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">error</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Error pushing update to the model card. Please read logs and retry.</span><span class="se">\n</span><span class="s2">$</span><span class="si">{</span><span class="n">exc</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">git_head_commit_url</span></div>
+
+
+    <span class="c1">#</span>
+    <span class="c1"># Deprecated code</span>
+    <span class="c1">#</span>
+
+<div class="viewcode-block" id="RaftTrainer.prediction_loop">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.prediction_loop">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">prediction_loop</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">dataloader</span><span class="p">:</span> <span class="n">DataLoader</span><span class="p">,</span>
+        <span class="n">description</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+        <span class="n">prediction_loss_only</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">ignore_keys</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">metric_key_prefix</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;eval&quot;</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">EvalLoopOutput</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Prediction/evaluation loop, shared by `Trainer.evaluate()` and `Trainer.predict()`.</span>
+<span class="sd">        Works both with or without labels.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">args</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">has_length</span><span class="p">(</span><span class="n">dataloader</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;dataloader must implement a working __len__&quot;</span><span class="p">)</span>
+
+        <span class="n">prediction_loss_only</span> <span class="o">=</span> <span class="n">prediction_loss_only</span> <span class="k">if</span> <span class="n">prediction_loss_only</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">args</span><span class="o">.</span><span class="n">prediction_loss_only</span>
+
+        <span class="c1"># if eval is called w/o train init deepspeed here</span>
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">deepspeed</span> <span class="ow">and</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span><span class="p">:</span>
+            <span class="c1"># XXX: eval doesn&#39;t have `resume_from_checkpoint` arg but we should be able to do eval</span>
+            <span class="c1"># from the checkpoint eventually</span>
+            <span class="n">deepspeed_engine</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">deepspeed_init</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">num_training_steps</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">resume_from_checkpoint</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">deepspeed_engine</span><span class="o">.</span><span class="n">module</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">model_wrapped</span> <span class="o">=</span> <span class="n">deepspeed_engine</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">deepspeed</span> <span class="o">=</span> <span class="n">deepspeed_engine</span>
+            <span class="c1"># XXX: we don&#39;t need optim/sched for inference, but this needs to be sorted out, since</span>
+            <span class="c1"># for example the Z3-optimizer is a must for zero3 to work even for inference - what we</span>
+            <span class="c1"># don&#39;t need is the deepspeed basic optimizer which is self.optimizer.optimizer</span>
+            <span class="n">deepspeed_engine</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">optimizer</span> <span class="o">=</span> <span class="kc">None</span>
+            <span class="n">deepspeed_engine</span><span class="o">.</span><span class="n">lr_scheduler</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="n">model</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_wrap_model</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="p">,</span> <span class="n">training</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">dataloader</span><span class="o">=</span><span class="n">dataloader</span><span class="p">)</span>
+
+        <span class="c1"># if full fp16 or bf16 eval is wanted and this ``evaluation`` or ``predict`` isn&#39;t called</span>
+        <span class="c1"># while ``train`` is running, cast it to the right dtype first and then put on device</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_in_train</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">fp16_full_eval</span><span class="p">:</span>
+                <span class="n">model</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float16</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">args</span><span class="o">.</span><span class="n">bf16_full_eval</span><span class="p">:</span>
+                <span class="n">model</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">bfloat16</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+
+        <span class="n">batch_size</span> <span class="o">=</span> <span class="n">dataloader</span><span class="o">.</span><span class="n">batch_size</span>
+        <span class="n">num_examples</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_examples</span><span class="p">(</span><span class="n">dataloader</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;***** Running </span><span class="si">{</span><span class="n">description</span><span class="si">}</span><span class="s2"> *****&quot;</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;  Num examples = </span><span class="si">{</span><span class="n">num_examples</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;  Batch size = </span><span class="si">{</span><span class="n">batch_size</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">losses_host</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">preds_host</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">labels_host</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="n">inputs_host</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="n">world_size</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">args</span><span class="o">.</span><span class="n">world_size</span><span class="p">)</span>
+
+        <span class="n">eval_losses_gatherer</span> <span class="o">=</span> <span class="n">DistributedTensorGatherer</span><span class="p">(</span><span class="n">world_size</span><span class="p">,</span> <span class="n">num_examples</span><span class="p">,</span> <span class="n">make_multiple_of</span><span class="o">=</span><span class="n">batch_size</span><span class="p">)</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">prediction_loss_only</span><span class="p">:</span>
+            <span class="c1"># The actual number of eval_sample can be greater than num_examples in distributed settings (when we pass</span>
+            <span class="c1"># a batch size to the sampler)</span>
+            <span class="n">make_multiple_of</span> <span class="o">=</span> <span class="kc">None</span>
+            <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">dataloader</span><span class="p">,</span> <span class="s2">&quot;sampler&quot;</span><span class="p">)</span> <span class="ow">and</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">dataloader</span><span class="o">.</span><span class="n">sampler</span><span class="p">,</span> <span class="n">SequentialDistributedSampler</span><span class="p">):</span>
+                <span class="n">make_multiple_of</span> <span class="o">=</span> <span class="n">dataloader</span><span class="o">.</span><span class="n">sampler</span><span class="o">.</span><span class="n">batch_size</span>
+            <span class="n">preds_gatherer</span> <span class="o">=</span> <span class="n">DistributedTensorGatherer</span><span class="p">(</span><span class="n">world_size</span><span class="p">,</span> <span class="n">num_examples</span><span class="p">,</span> <span class="n">make_multiple_of</span><span class="o">=</span><span class="n">make_multiple_of</span><span class="p">)</span>
+            <span class="n">labels_gatherer</span> <span class="o">=</span> <span class="n">DistributedTensorGatherer</span><span class="p">(</span><span class="n">world_size</span><span class="p">,</span> <span class="n">num_examples</span><span class="p">,</span> <span class="n">make_multiple_of</span><span class="o">=</span><span class="n">make_multiple_of</span><span class="p">)</span>
+            <span class="n">inputs_gatherer</span> <span class="o">=</span> <span class="n">DistributedTensorGatherer</span><span class="p">(</span><span class="n">world_size</span><span class="p">,</span> <span class="n">num_examples</span><span class="p">,</span> <span class="n">make_multiple_of</span><span class="o">=</span><span class="n">make_multiple_of</span><span class="p">)</span>
+
+        <span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
+
+        <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+            <span class="n">dataloader</span> <span class="o">=</span> <span class="n">pl</span><span class="o">.</span><span class="n">ParallelLoader</span><span class="p">(</span><span class="n">dataloader</span><span class="p">,</span> <span class="p">[</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">])</span><span class="o">.</span><span class="n">per_device_loader</span><span class="p">(</span><span class="n">args</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">past_index</span> <span class="o">&gt;=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">_past</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">eval_dataloader</span> <span class="o">=</span> <span class="n">dataloader</span>
+
+        <span class="k">for</span> <span class="n">step</span><span class="p">,</span> <span class="n">inputs</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">dataloader</span><span class="p">):</span>
+            <span class="n">loss</span><span class="p">,</span> <span class="n">logits</span><span class="p">,</span> <span class="n">labels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">prediction_step</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">prediction_loss_only</span><span class="p">,</span> <span class="n">ignore_keys</span><span class="o">=</span><span class="n">ignore_keys</span><span class="p">)</span>
+            <span class="n">inputs_decode</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_prepare_input</span><span class="p">(</span><span class="n">inputs</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">])</span> <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">include_inputs_for_metrics</span> <span class="k">else</span> <span class="kc">None</span>
+
+            <span class="k">if</span> <span class="n">loss</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">losses</span> <span class="o">=</span> <span class="n">loss</span><span class="o">.</span><span class="n">repeat</span><span class="p">(</span><span class="n">batch_size</span><span class="p">)</span>
+                <span class="n">losses_host</span> <span class="o">=</span> <span class="n">losses</span> <span class="k">if</span> <span class="n">losses_host</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">losses_host</span><span class="p">,</span> <span class="n">losses</span><span class="p">),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">logits</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">preds_host</span> <span class="o">=</span> <span class="n">logits</span> <span class="k">if</span> <span class="n">preds_host</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">nested_concat</span><span class="p">(</span><span class="n">preds_host</span><span class="p">,</span> <span class="n">logits</span><span class="p">,</span> <span class="n">padding_index</span><span class="o">=-</span><span class="mi">100</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">labels</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">labels_host</span> <span class="o">=</span> <span class="n">labels</span> <span class="k">if</span> <span class="n">labels_host</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">nested_concat</span><span class="p">(</span><span class="n">labels_host</span><span class="p">,</span> <span class="n">labels</span><span class="p">,</span> <span class="n">padding_index</span><span class="o">=-</span><span class="mi">100</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">inputs_decode</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">inputs_host</span> <span class="o">=</span> <span class="p">(</span>
+                    <span class="n">inputs_decode</span>
+                    <span class="k">if</span> <span class="n">inputs_host</span> <span class="ow">is</span> <span class="kc">None</span>
+                    <span class="k">else</span> <span class="n">nested_concat</span><span class="p">(</span><span class="n">inputs_host</span><span class="p">,</span> <span class="n">inputs_decode</span><span class="p">,</span> <span class="n">padding_index</span><span class="o">=-</span><span class="mi">100</span><span class="p">)</span>
+                <span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">control</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">callback_handler</span><span class="o">.</span><span class="n">on_prediction_step</span><span class="p">(</span><span class="n">args</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">state</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">control</span><span class="p">)</span>
+
+            <span class="c1"># Gather all tensors and put them back on the CPU if we have done enough accumulation steps.</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">eval_accumulation_steps</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="p">(</span><span class="n">step</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">%</span> <span class="n">args</span><span class="o">.</span><span class="n">eval_accumulation_steps</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">eval_losses_gatherer</span><span class="o">.</span><span class="n">add_arrays</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_gather_and_numpify</span><span class="p">(</span><span class="n">losses_host</span><span class="p">,</span> <span class="s2">&quot;eval_losses&quot;</span><span class="p">))</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="n">prediction_loss_only</span><span class="p">:</span>
+                    <span class="n">preds_gatherer</span><span class="o">.</span><span class="n">add_arrays</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_gather_and_numpify</span><span class="p">(</span><span class="n">preds_host</span><span class="p">,</span> <span class="s2">&quot;eval_preds&quot;</span><span class="p">))</span>
+                    <span class="n">labels_gatherer</span><span class="o">.</span><span class="n">add_arrays</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_gather_and_numpify</span><span class="p">(</span><span class="n">labels_host</span><span class="p">,</span> <span class="s2">&quot;eval_label_ids&quot;</span><span class="p">))</span>
+                    <span class="n">inputs_gatherer</span><span class="o">.</span><span class="n">add_arrays</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_gather_and_numpify</span><span class="p">(</span><span class="n">inputs_host</span><span class="p">,</span> <span class="s2">&quot;eval_inputs_ids&quot;</span><span class="p">))</span>
+
+                <span class="c1"># Set back to None to begin a new accumulation</span>
+                <span class="n">losses_host</span><span class="p">,</span> <span class="n">preds_host</span><span class="p">,</span> <span class="n">labels_host</span><span class="p">,</span> <span class="n">inputs_host</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span>
+
+        <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">past_index</span> <span class="ow">and</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;_past&quot;</span><span class="p">):</span>
+            <span class="c1"># Clean the state at the end of the evaluation loop</span>
+            <span class="nb">delattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;_past&quot;</span><span class="p">)</span>
+
+        <span class="c1"># Gather all remaining tensors and put them back on the CPU</span>
+        <span class="n">eval_losses_gatherer</span><span class="o">.</span><span class="n">add_arrays</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_gather_and_numpify</span><span class="p">(</span><span class="n">losses_host</span><span class="p">,</span> <span class="s2">&quot;eval_losses&quot;</span><span class="p">))</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">prediction_loss_only</span><span class="p">:</span>
+            <span class="n">preds_gatherer</span><span class="o">.</span><span class="n">add_arrays</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_gather_and_numpify</span><span class="p">(</span><span class="n">preds_host</span><span class="p">,</span> <span class="s2">&quot;eval_preds&quot;</span><span class="p">))</span>
+            <span class="n">labels_gatherer</span><span class="o">.</span><span class="n">add_arrays</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_gather_and_numpify</span><span class="p">(</span><span class="n">labels_host</span><span class="p">,</span> <span class="s2">&quot;eval_label_ids&quot;</span><span class="p">))</span>
+            <span class="n">inputs_gatherer</span><span class="o">.</span><span class="n">add_arrays</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_gather_and_numpify</span><span class="p">(</span><span class="n">inputs_host</span><span class="p">,</span> <span class="s2">&quot;eval_inputs_ids&quot;</span><span class="p">))</span>
+
+        <span class="n">eval_loss</span> <span class="o">=</span> <span class="n">eval_losses_gatherer</span><span class="o">.</span><span class="n">finalize</span><span class="p">()</span>
+        <span class="n">preds</span> <span class="o">=</span> <span class="n">preds_gatherer</span><span class="o">.</span><span class="n">finalize</span><span class="p">()</span> <span class="k">if</span> <span class="ow">not</span> <span class="n">prediction_loss_only</span> <span class="k">else</span> <span class="kc">None</span>
+        <span class="n">label_ids</span> <span class="o">=</span> <span class="n">labels_gatherer</span><span class="o">.</span><span class="n">finalize</span><span class="p">()</span> <span class="k">if</span> <span class="ow">not</span> <span class="n">prediction_loss_only</span> <span class="k">else</span> <span class="kc">None</span>
+        <span class="n">inputs_ids</span> <span class="o">=</span> <span class="n">inputs_gatherer</span><span class="o">.</span><span class="n">finalize</span><span class="p">()</span> <span class="k">if</span> <span class="ow">not</span> <span class="n">prediction_loss_only</span> <span class="k">else</span> <span class="kc">None</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_metrics</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">preds</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">label_ids</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">args</span><span class="o">.</span><span class="n">include_inputs_for_metrics</span><span class="p">:</span>
+                <span class="n">metrics</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_metrics</span><span class="p">(</span>
+                    <span class="n">EvalPrediction</span><span class="p">(</span><span class="n">predictions</span><span class="o">=</span><span class="n">preds</span><span class="p">,</span> <span class="n">label_ids</span><span class="o">=</span><span class="n">label_ids</span><span class="p">,</span> <span class="n">inputs</span><span class="o">=</span><span class="n">inputs_ids</span><span class="p">)</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">metrics</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">compute_metrics</span><span class="p">(</span><span class="n">EvalPrediction</span><span class="p">(</span><span class="n">predictions</span><span class="o">=</span><span class="n">preds</span><span class="p">,</span> <span class="n">label_ids</span><span class="o">=</span><span class="n">label_ids</span><span class="p">))</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">metrics</span> <span class="o">=</span> <span class="p">{}</span>
+
+        <span class="c1"># To be JSON-serializable, we need to remove numpy types or zero-d tensors</span>
+        <span class="n">metrics</span> <span class="o">=</span> <span class="n">denumpify_detensorize</span><span class="p">(</span><span class="n">metrics</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">eval_loss</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">metric_key_prefix</span><span class="si">}</span><span class="s2">_loss&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">eval_loss</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span><span class="o">.</span><span class="n">item</span><span class="p">()</span>
+
+        <span class="c1"># Prefix all keys with metric_key_prefix + &#39;_&#39;</span>
+        <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="nb">list</span><span class="p">(</span><span class="n">metrics</span><span class="o">.</span><span class="n">keys</span><span class="p">()):</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">key</span><span class="o">.</span><span class="n">startswith</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">metric_key_prefix</span><span class="si">}</span><span class="s2">_&quot;</span><span class="p">):</span>
+                <span class="n">metrics</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">metric_key_prefix</span><span class="si">}</span><span class="s2">_</span><span class="si">{</span><span class="n">key</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">metrics</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="n">key</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">EvalLoopOutput</span><span class="p">(</span><span class="n">predictions</span><span class="o">=</span><span class="n">preds</span><span class="p">,</span> <span class="n">label_ids</span><span class="o">=</span><span class="n">label_ids</span><span class="p">,</span> <span class="n">metrics</span><span class="o">=</span><span class="n">metrics</span><span class="p">,</span> <span class="n">num_samples</span><span class="o">=</span><span class="n">num_examples</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._gather_and_numpify">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._gather_and_numpify">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_gather_and_numpify</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">tensors</span><span class="p">,</span> <span class="n">name</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Gather value of `tensors` (tensor or list/tuple of nested tensors) and convert them to numpy before</span>
+<span class="sd">        concatenating them to `gathered`</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">tensors</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span>
+        <span class="k">if</span> <span class="n">is_torch_tpu_available</span><span class="p">():</span>
+            <span class="n">tensors</span> <span class="o">=</span> <span class="n">nested_xla_mesh_reduce</span><span class="p">(</span><span class="n">tensors</span><span class="p">,</span> <span class="n">name</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="n">is_sagemaker_mp_enabled</span><span class="p">():</span>
+            <span class="n">tensors</span> <span class="o">=</span> <span class="n">smp_gather</span><span class="p">(</span><span class="n">tensors</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">args</span><span class="o">.</span><span class="n">local_rank</span> <span class="o">!=</span> <span class="o">-</span><span class="mi">1</span><span class="p">:</span>
+            <span class="n">tensors</span> <span class="o">=</span> <span class="n">distributed_concat</span><span class="p">(</span><span class="n">tensors</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">nested_numpify</span><span class="p">(</span><span class="n">tensors</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="RaftTrainer._add_sm_patterns_to_gitignore">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._add_sm_patterns_to_gitignore">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_add_sm_patterns_to_gitignore</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Add SageMaker Checkpointing patterns to .gitignore file.&quot;&quot;&quot;</span>
+        <span class="c1"># Make sure we only do this on the main process</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_world_process_zero</span><span class="p">():</span>
+            <span class="k">return</span>
+
+        <span class="n">patterns</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;*.sagemaker-uploading&quot;</span><span class="p">,</span> <span class="s2">&quot;*.sagemaker-uploaded&quot;</span><span class="p">]</span>
+
+        <span class="c1"># Get current .gitignore content</span>
+        <span class="k">if</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">repo</span><span class="o">.</span><span class="n">local_dir</span><span class="p">,</span> <span class="s2">&quot;.gitignore&quot;</span><span class="p">)):</span>
+            <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">repo</span><span class="o">.</span><span class="n">local_dir</span><span class="p">,</span> <span class="s2">&quot;.gitignore&quot;</span><span class="p">),</span> <span class="s2">&quot;r&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+                <span class="n">current_content</span> <span class="o">=</span> <span class="n">f</span><span class="o">.</span><span class="n">read</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">current_content</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
+
+        <span class="c1"># Add the patterns to .gitignore</span>
+        <span class="n">content</span> <span class="o">=</span> <span class="n">current_content</span>
+        <span class="k">for</span> <span class="n">pattern</span> <span class="ow">in</span> <span class="n">patterns</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">pattern</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">content</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">content</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">):</span>
+                    <span class="n">content</span> <span class="o">+=</span> <span class="n">pattern</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">content</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="se">\n</span><span class="si">{</span><span class="n">pattern</span><span class="si">}</span><span class="s2">&quot;</span>
+
+        <span class="c1"># Write the .gitignore file if it has changed</span>
+        <span class="k">if</span> <span class="n">content</span> <span class="o">!=</span> <span class="n">current_content</span><span class="p">:</span>
+            <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">repo</span><span class="o">.</span><span class="n">local_dir</span><span class="p">,</span> <span class="s2">&quot;.gitignore&quot;</span><span class="p">),</span> <span class="s2">&quot;w&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Writing .gitignore file. Content: </span><span class="si">{</span><span class="n">content</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="n">f</span><span class="o">.</span><span class="n">write</span><span class="p">(</span><span class="n">content</span><span class="p">)</span>
+
+        <span class="bp">self</span><span class="o">.</span><span class="n">repo</span><span class="o">.</span><span class="n">git_add</span><span class="p">(</span><span class="s2">&quot;.gitignore&quot;</span><span class="p">)</span>
+
+        <span class="c1"># avoid race condition with git status</span>
+        <span class="n">time</span><span class="o">.</span><span class="n">sleep</span><span class="p">(</span><span class="mf">0.5</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">repo</span><span class="o">.</span><span class="n">is_repo_clean</span><span class="p">():</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">repo</span><span class="o">.</span><span class="n">git_commit</span><span class="p">(</span><span class="s2">&quot;Add *.sagemaker patterns to .gitignore.&quot;</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">repo</span><span class="o">.</span><span class="n">git_push</span><span class="p">()</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/utils/rm_dataprocessor.html b/_modules/lmflow/pipeline/utils/rm_dataprocessor.html
new file mode 100644
index 000000000..f0fa17484
--- /dev/null
+++ b/_modules/lmflow/pipeline/utils/rm_dataprocessor.html
@@ -0,0 +1,515 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.utils.rm_dataprocessor &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/utils/rm_dataprocessor';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.rm_dataprocessor</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.utils.rm_dataprocessor</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">dataclasses</span><span class="w"> </span><span class="kn">import</span> <span class="n">dataclass</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">datasets</span><span class="w"> </span><span class="kn">import</span> <span class="n">load_dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">AutoTokenizer</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">PaddingStrategy</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="RewardDataCollatorWithPadding">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">RewardDataCollatorWithPadding</span><span class="p">:</span>
+<div class="viewcode-block" id="RewardDataCollatorWithPadding.tokenizer">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.tokenizer">[docs]</a>
+    <span class="n">tokenizer</span><span class="p">:</span> <span class="n">AutoTokenizer</span></div>
+
+<div class="viewcode-block" id="RewardDataCollatorWithPadding.padding">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.padding">[docs]</a>
+    <span class="n">padding</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">bool</span><span class="p">,</span> <span class="nb">str</span><span class="p">,</span> <span class="n">PaddingStrategy</span><span class="p">]</span> <span class="o">=</span> <span class="kc">True</span></div>
+
+<div class="viewcode-block" id="RewardDataCollatorWithPadding.max_length">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.max_length">[docs]</a>
+    <span class="n">max_length</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="RewardDataCollatorWithPadding.pad_to_multiple_of">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.pad_to_multiple_of">[docs]</a>
+    <span class="n">pad_to_multiple_of</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="RewardDataCollatorWithPadding.return_tensors">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.return_tensors">[docs]</a>
+    <span class="n">return_tensors</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;pt&quot;</span></div>
+
+
+<div class="viewcode-block" id="RewardDataCollatorWithPadding.__call__">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.__call__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">features</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]:</span>
+        <span class="n">merged_features</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">feature</span> <span class="ow">in</span> <span class="n">features</span><span class="p">:</span>
+            <span class="n">merged_features</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="p">{</span>
+                    <span class="s2">&quot;input_ids&quot;</span><span class="p">:</span> <span class="n">feature</span><span class="p">[</span><span class="s2">&quot;input_ids_chosen&quot;</span><span class="p">],</span>
+                    <span class="s2">&quot;attention_mask&quot;</span><span class="p">:</span> <span class="n">feature</span><span class="p">[</span><span class="s2">&quot;attention_mask_chosen&quot;</span><span class="p">],</span>
+                <span class="p">}</span>
+            <span class="p">)</span>
+            <span class="n">merged_features</span><span class="o">.</span><span class="n">append</span><span class="p">(</span>
+                <span class="p">{</span>
+                    <span class="s2">&quot;input_ids&quot;</span><span class="p">:</span> <span class="n">feature</span><span class="p">[</span><span class="s2">&quot;input_ids_rejected&quot;</span><span class="p">],</span>
+                    <span class="s2">&quot;attention_mask&quot;</span><span class="p">:</span> <span class="n">feature</span><span class="p">[</span><span class="s2">&quot;attention_mask_rejected&quot;</span><span class="p">],</span>
+                <span class="p">}</span>
+            <span class="p">)</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Chosen: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">feature</span><span class="p">[</span><span class="s1">&#39;input_ids_chosen&#39;</span><span class="p">])</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Rejected: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">decode</span><span class="p">(</span><span class="n">feature</span><span class="p">[</span><span class="s1">&#39;input_ids_rejected&#39;</span><span class="p">])</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="n">batch</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad</span><span class="p">(</span>
+            <span class="n">merged_features</span><span class="p">,</span>
+            <span class="n">padding</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">padding</span><span class="p">,</span>
+            <span class="n">max_length</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">max_length</span><span class="p">,</span>
+            <span class="n">pad_to_multiple_of</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">pad_to_multiple_of</span><span class="p">,</span>
+            <span class="n">return_tensors</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">return_tensors</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">batch</span> <span class="o">=</span> <span class="p">{</span>
+            <span class="s2">&quot;input_ids&quot;</span><span class="p">:</span> <span class="n">batch</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;attention_mask&quot;</span><span class="p">:</span> <span class="n">batch</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">],</span>
+            <span class="s2">&quot;return_loss&quot;</span><span class="p">:</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="p">}</span>
+        <span class="k">return</span> <span class="n">batch</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/utils/rm_trainer.html b/_modules/lmflow/pipeline/utils/rm_trainer.html
new file mode 100644
index 000000000..9d8157b51
--- /dev/null
+++ b/_modules/lmflow/pipeline/utils/rm_trainer.html
@@ -0,0 +1,501 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.utils.rm_trainer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/utils/rm_trainer';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.rm_trainer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.utils.rm_trainer</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.nn</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">nn</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">Trainer</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">.peft_trainer</span><span class="w"> </span><span class="kn">import</span> <span class="n">PeftTrainer</span>
+
+
+<div class="viewcode-block" id="compute_metrics">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_trainer/index.html#lmflow.pipeline.utils.rm_trainer.compute_metrics">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">compute_metrics</span><span class="p">(</span><span class="n">eval_pred</span><span class="p">):</span>
+    <span class="n">result</span> <span class="o">=</span> <span class="p">{}</span>
+    <span class="n">pos_predictions_scores</span> <span class="o">=</span> <span class="n">eval_pred</span><span class="o">.</span><span class="n">predictions</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="n">neg_predictions_scores</span> <span class="o">=</span> <span class="n">eval_pred</span><span class="o">.</span><span class="n">predictions</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+    <span class="c1"># We assume that the first sample is preferred by default in groundtruth</span>
+    <span class="n">result</span><span class="p">[</span><span class="s1">&#39;accuracy&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span>
+        <span class="n">pos_predictions_scores</span> <span class="o">&gt;</span> <span class="n">neg_predictions_scores</span><span class="p">)</span> <span class="o">/</span> <span class="nb">len</span><span class="p">(</span><span class="n">pos_predictions_scores</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">result</span></div>
+
+
+
+<div class="viewcode-block" id="rm_loss">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_trainer/index.html#lmflow.pipeline.utils.rm_trainer.rm_loss">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">rm_loss</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">return_outputs</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+    <span class="n">rewards</span> <span class="o">=</span> <span class="n">model</span><span class="p">(</span>
+        <span class="n">input_ids</span><span class="o">=</span><span class="n">inputs</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">],</span> 
+        <span class="n">attention_mask</span><span class="o">=</span><span class="n">inputs</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">]</span>
+    <span class="p">)[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="n">bsz</span> <span class="o">=</span> <span class="n">rewards</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+    <span class="n">jidx</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">bsz</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="n">kidx</span> <span class="o">=</span> <span class="n">jidx</span> <span class="o">+</span> <span class="mi">1</span>
+    <span class="n">rewards_j</span> <span class="o">=</span> <span class="n">rewards</span><span class="p">[</span><span class="n">jidx</span><span class="p">]</span>
+    <span class="n">rewards_k</span> <span class="o">=</span> <span class="n">rewards</span><span class="p">[</span><span class="n">kidx</span><span class="p">]</span>
+    <span class="n">loss</span> <span class="o">=</span> <span class="o">-</span><span class="n">nn</span><span class="o">.</span><span class="n">functional</span><span class="o">.</span><span class="n">logsigmoid</span><span class="p">(</span><span class="n">rewards_j</span> <span class="o">-</span> <span class="n">rewards_k</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span>
+    <span class="k">if</span> <span class="n">return_outputs</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">loss</span><span class="p">,</span> <span class="p">{</span><span class="s2">&quot;rewards_j&quot;</span><span class="p">:</span> <span class="n">rewards_j</span><span class="p">,</span> <span class="s2">&quot;rewards_k&quot;</span><span class="p">:</span> <span class="n">rewards_k</span><span class="p">}</span>
+    <span class="k">return</span> <span class="n">loss</span></div>
+
+
+
+<div class="viewcode-block" id="RewardTrainer">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_trainer/index.html#lmflow.pipeline.utils.rm_trainer.RewardTrainer">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">RewardTrainer</span><span class="p">(</span><span class="n">Trainer</span><span class="p">):</span>
+<div class="viewcode-block" id="RewardTrainer.compute_loss">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_trainer/index.html#lmflow.pipeline.utils.rm_trainer.RewardTrainer.compute_loss">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">compute_loss</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">return_outputs</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">rm_loss</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">return_outputs</span><span class="p">)</span></div>
+</div>
+
+
+
+<div class="viewcode-block" id="PeftRewardTrainer">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_trainer/index.html#lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">PeftRewardTrainer</span><span class="p">(</span><span class="n">PeftTrainer</span><span class="p">):</span>
+<div class="viewcode-block" id="PeftRewardTrainer.compute_loss">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/pipeline/utils/rm_trainer/index.html#lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer.compute_loss">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">compute_loss</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">return_outputs</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">rm_loss</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">return_outputs</span><span class="p">)</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/pipeline/vllm_inferencer.html b/_modules/lmflow/pipeline/vllm_inferencer.html
new file mode 100644
index 000000000..2fc0849ac
--- /dev/null
+++ b/_modules/lmflow/pipeline/vllm_inferencer.html
@@ -0,0 +1,866 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.pipeline.vllm_inferencer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/pipeline/vllm_inferencer';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.vllm_inferencer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.pipeline.vllm_inferencer</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">copy</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">importlib.resources</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">pkg_resources</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">json</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">os</span>
+<span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="p">[</span><span class="s1">&#39;VLLM_WORKER_MULTIPROC_METHOD&#39;</span><span class="p">]</span><span class="o">=</span><span class="s1">&#39;spawn&#39;</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">subprocess</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">functools</span><span class="w"> </span><span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Any</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">AutoTokenizer</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.base_pipeline</span><span class="w"> </span><span class="kn">import</span> <span class="n">BasePipeline</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.hf_decoder_model</span><span class="w"> </span><span class="kn">import</span> <span class="n">HFDecoderModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">InferencerArguments</span><span class="p">,</span> 
+    <span class="n">ModelArguments</span><span class="p">,</span> 
+    <span class="n">DatasetArguments</span><span class="p">,</span>
+<span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.common</span><span class="w"> </span><span class="kn">import</span> <span class="n">make_shell_args_from_dataclass</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="n">RETURN_CODE_ERROR_BUFFER</span><span class="p">,</span> <span class="n">MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.data_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">VLLMInferenceResultWithInput</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.versioning</span><span class="w"> </span><span class="kn">import</span> <span class="n">is_vllm_available</span><span class="p">,</span> <span class="n">is_ray_available</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<span class="k">if</span> <span class="n">is_vllm_available</span><span class="p">():</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">vllm</span><span class="w"> </span><span class="kn">import</span> <span class="n">SamplingParams</span><span class="p">,</span> <span class="n">LLM</span>
+<span class="k">else</span><span class="p">:</span>
+    <span class="k">raise</span> <span class="ne">ImportError</span><span class="p">(</span><span class="s2">&quot;VLLM is not available, please install vllm.&quot;</span><span class="p">)</span>
+
+<span class="k">if</span> <span class="n">is_ray_available</span><span class="p">():</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">ray</span>
+    <span class="kn">import</span><span class="w"> </span><span class="nn">ray.data</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">ray.util.scheduling_strategies</span><span class="w"> </span><span class="kn">import</span> <span class="n">PlacementGroupSchedulingStrategy</span>
+<span class="k">else</span><span class="p">:</span>
+    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;Ray is not available, distributed vllm inference will not be supported.&quot;</span><span class="p">)</span>
+    
+
+<div class="viewcode-block" id="InferencerWithOffloading">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">InferencerWithOffloading</span><span class="p">(</span><span class="n">BasePipeline</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span>
+        <span class="n">inferencer_args</span><span class="p">:</span> <span class="n">InferencerArguments</span><span class="p">,</span>
+    <span class="p">):</span>
+<div class="viewcode-block" id="InferencerWithOffloading.model_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.model_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span> <span class="o">=</span> <span class="n">model_args</span></div>
+
+<div class="viewcode-block" id="InferencerWithOffloading.data_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.data_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span> <span class="o">=</span> <span class="n">data_args</span></div>
+
+<div class="viewcode-block" id="InferencerWithOffloading.inferencer_args">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.inferencer_args">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span> <span class="o">=</span> <span class="n">inferencer_args</span></div>
+
+<div class="viewcode-block" id="InferencerWithOffloading.eos_token_id">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.eos_token_id">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">eos_token_id</span> <span class="o">=</span> <span class="n">AutoTokenizer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">)</span><span class="o">.</span><span class="n">eos_token_id</span></div>
+
+
+<div class="viewcode-block" id="InferencerWithOffloading.inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">inference</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">&quot;.inference is not implemented&quot;</span><span class="p">)</span></div>
+
+        
+<div class="viewcode-block" id="InferencerWithOffloading.save_inference_results">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.save_inference_results">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">save_inference_results</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">&quot;.save_inference_results is not implemented&quot;</span><span class="p">)</span></div>
+
+        
+<div class="viewcode-block" id="InferencerWithOffloading.load_inference_results">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.load_inference_results">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">load_inference_results</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">&quot;.load_inference_results is not implemented&quot;</span><span class="p">)</span></div>
+</div>
+
+    
+    
+<div class="viewcode-block" id="VLLMInferencer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">VLLMInferencer</span><span class="p">(</span><span class="n">InferencerWithOffloading</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span>
+        <span class="n">inferencer_args</span><span class="p">:</span> <span class="n">InferencerArguments</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">assert</span> <span class="n">inferencer_args</span><span class="o">.</span><span class="n">use_vllm</span><span class="p">,</span> <span class="s2">&quot;The inferencer_args.use_vllm must be True.&quot;</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">inferencer_args</span><span class="p">)</span>
+<div class="viewcode-block" id="VLLMInferencer.sampling_params">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer.sampling_params">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">sampling_params</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">parse_to_sampling_params</span><span class="p">(</span><span class="n">inferencer_args</span><span class="p">)</span></div>
+
+        
+    
+<div class="viewcode-block" id="VLLMInferencer.parse_to_sampling_params">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer.parse_to_sampling_params">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">parse_to_sampling_params</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">inference_args</span><span class="p">:</span> <span class="n">InferencerArguments</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">SamplingParams</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">SamplingParams</span><span class="p">(</span>
+            <span class="n">use_beam_search</span><span class="o">=</span><span class="n">inference_args</span><span class="o">.</span><span class="n">use_beam_search</span><span class="p">,</span>
+            <span class="n">n</span><span class="o">=</span><span class="n">inference_args</span><span class="o">.</span><span class="n">num_output_sequences</span><span class="p">,</span>
+            <span class="n">temperature</span><span class="o">=</span><span class="n">inference_args</span><span class="o">.</span><span class="n">temperature</span> <span class="o">+</span> <span class="mf">1e-6</span><span class="p">,</span>
+            <span class="n">max_tokens</span><span class="o">=</span><span class="n">inference_args</span><span class="o">.</span><span class="n">max_new_tokens</span><span class="p">,</span>
+            <span class="n">seed</span><span class="o">=</span><span class="n">inference_args</span><span class="o">.</span><span class="n">random_seed</span><span class="p">,</span>
+            <span class="n">top_p</span><span class="o">=</span><span class="n">inference_args</span><span class="o">.</span><span class="n">top_p</span><span class="p">,</span>
+            <span class="n">top_k</span><span class="o">=</span><span class="n">inference_args</span><span class="o">.</span><span class="n">top_k</span><span class="p">,</span>
+            <span class="n">stop_token_ids</span><span class="o">=</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">eos_token_id</span><span class="p">]</span> <span class="o">+</span> <span class="n">inference_args</span><span class="o">.</span><span class="n">additional_stop_token_ids</span>
+        <span class="p">)</span></div>
+
+        
+        
+<div class="viewcode-block" id="VLLMInferencer.inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer.inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span> 
+        <span class="n">dataset</span><span class="p">:</span> <span class="n">Dataset</span><span class="p">,</span> 
+        <span class="n">enable_decode_inference_result</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">True</span><span class="p">,</span>
+        <span class="n">release_gpu</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">inference_args</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">InferencerArguments</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">enable_distributed_inference</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">VLLMInferenceResultWithInput</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Perform inference using the provided model and dataset. Will save inference results if</span>
+<span class="sd">        `save_results` is set to True in `inferencer_args`.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ----------</span>
+<span class="sd">        model : HFDecoderModel</span>
+<span class="sd">            LMFlow HFDecoderModel object</span>
+<span class="sd">        dataset : Dataset</span>
+<span class="sd">            LMFlow Dataset object</span>
+<span class="sd">        apply_chat_template : bool, optional</span>
+<span class="sd">            Whether to apply chat template to the input, by default True.</span>
+<span class="sd">        enable_decode_inference_result : bool, optional</span>
+<span class="sd">            Whether to decode after generation, by default False.</span>
+<span class="sd">        release_gpu : bool, optional</span>
+<span class="sd">            Whether to release gpu resources, by default False. </span>
+<span class="sd">        inference_args : InferencerArguments, optional</span>
+<span class="sd">            by default None</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        -------</span>
+<span class="sd">        List[VLLMInferenceResultWithInput]</span>
+<span class="sd">            Return a list of VLLMInferenceResultWithInput, where each</span>
+<span class="sd">            element contains the input prompt and the corresponding output.</span>
+<span class="sd">            </span>
+<span class="sd">            When `enable_decode_inference_result = True`, the output would be a list of strings,</span>
+<span class="sd">            contains sampling_params.n samples for the corresponding prompt.</span>
+<span class="sd">            </span>
+<span class="sd">            When `enable_decode_inference_result = False`, return a list of list of ints </span>
+<span class="sd">            (token ids, no decoding after generation).</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">inference_args</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                <span class="s2">&quot;Overriding the default inference arguments with the provided arguments in .inference()&quot;</span>
+            <span class="p">)</span>
+            <span class="n">sampling_params</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">parse_to_sampling_params</span><span class="p">(</span><span class="n">inference_args</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">sampling_params</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">sampling_params</span>
+            
+        <span class="n">sampling_params</span><span class="o">.</span><span class="n">detokenize</span> <span class="o">=</span> <span class="n">enable_decode_inference_result</span>
+        
+        <span class="n">model_input</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">prepare_inputs_for_inference</span><span class="p">(</span>
+            <span class="n">dataset</span><span class="o">=</span><span class="n">dataset</span><span class="p">,</span>
+            <span class="n">apply_chat_template</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">apply_chat_template</span><span class="p">,</span>
+            <span class="n">use_vllm</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">use_vllm</span><span class="p">,</span>
+            <span class="n">enable_distributed_inference</span><span class="o">=</span><span class="n">enable_distributed_inference</span><span class="p">,</span>
+        <span class="p">)</span>
+        
+        <span class="k">if</span> <span class="n">enable_distributed_inference</span><span class="p">:</span>
+            <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_distributed_inference</span><span class="p">(</span>
+                <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span> 
+                <span class="n">model_input</span><span class="o">=</span><span class="n">model_input</span><span class="p">,</span> 
+                <span class="n">sampling_params</span><span class="o">=</span><span class="n">sampling_params</span><span class="p">,</span>
+                <span class="n">num_instances</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;distributed_inference_num_instances&quot;</span><span class="p">),</span>
+                <span class="n">batch_size</span><span class="o">=</span><span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;inference_batch_size&quot;</span><span class="p">,</span> <span class="mi">4</span><span class="p">),</span>
+                <span class="n">release_gpu</span><span class="o">=</span><span class="n">release_gpu</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_inference</span><span class="p">(</span>
+                <span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span>
+                <span class="n">model_input</span><span class="o">=</span><span class="n">model_input</span><span class="p">,</span> 
+                <span class="n">sampling_params</span><span class="o">=</span><span class="n">sampling_params</span><span class="p">,</span>
+                <span class="n">release_gpu</span><span class="o">=</span><span class="n">release_gpu</span><span class="p">,</span>
+            <span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">save_results</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">save_inference_results</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">results_path</span><span class="p">)</span>
+            
+        <span class="k">return</span> <span class="n">outputs</span></div>
+
+
+
+<div class="viewcode-block" id="VLLMInferencer._inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer._inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span> 
+        <span class="n">model_input</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> 
+        <span class="n">sampling_params</span><span class="p">:</span> <span class="n">SamplingParams</span><span class="p">,</span>
+        <span class="n">release_gpu</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">VLLMInferenceResultWithInput</span><span class="p">]:</span>        
+        <span class="n">outputs</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span>
+            <span class="n">inputs</span><span class="o">=</span><span class="n">model_input</span><span class="p">,</span>
+            <span class="n">sampling_params</span><span class="o">=</span><span class="n">sampling_params</span><span class="p">,</span>
+            <span class="n">release_gpu</span><span class="o">=</span><span class="n">release_gpu</span><span class="p">,</span>
+            <span class="n">use_vllm</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+            <span class="n">vllm_gpu_memory_utilization</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">vllm_gpu_memory_utilization</span><span class="p">,</span>
+            <span class="n">vllm_tensor_parallel_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">vllm_tensor_parallel_size</span><span class="p">,</span>
+        <span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">outputs</span></div>
+
+    
+    
+<div class="viewcode-block" id="VLLMInferencer._distributed_inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer._distributed_inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_distributed_inference</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span> 
+        <span class="n">model_input</span><span class="p">:</span> <span class="n">ray</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">Dataset</span><span class="p">,</span> 
+        <span class="n">sampling_params</span><span class="p">:</span> <span class="n">SamplingParams</span><span class="p">,</span>
+        <span class="n">num_instances</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+        <span class="n">batch_size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">4</span><span class="p">,</span>
+        <span class="n">release_gpu</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">VLLMInferenceResultWithInput</span><span class="p">]:</span>
+        <span class="c1"># prepare distributed inference resources</span>
+        <span class="c1"># from https://github.com/vllm-project/vllm/blob/main/examples/offline_inference_distributed.py</span>
+        <span class="c1">## strategy</span>
+        <span class="k">def</span><span class="w"> </span><span class="nf">scheduling_strategy_fn</span><span class="p">():</span>
+            <span class="c1"># One bundle per tensor parallel worker</span>
+            <span class="n">pg</span> <span class="o">=</span> <span class="n">ray</span><span class="o">.</span><span class="n">util</span><span class="o">.</span><span class="n">placement_group</span><span class="p">(</span>
+                <span class="p">[{</span>
+                    <span class="s2">&quot;GPU&quot;</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span>
+                    <span class="s2">&quot;CPU&quot;</span><span class="p">:</span> <span class="mi">1</span>
+                <span class="p">}]</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">vllm_tensor_parallel_size</span><span class="p">,</span>
+                <span class="n">strategy</span><span class="o">=</span><span class="s2">&quot;STRICT_PACK&quot;</span><span class="p">,</span>
+            <span class="p">)</span>
+            <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span>
+                <span class="n">scheduling_strategy</span><span class="o">=</span><span class="n">PlacementGroupSchedulingStrategy</span><span class="p">(</span>
+                    <span class="n">pg</span><span class="p">,</span> <span class="n">placement_group_capture_child_tasks</span><span class="o">=</span><span class="kc">True</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+            
+        <span class="n">resources_kwarg</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Any</span><span class="p">]</span> <span class="o">=</span> <span class="p">{}</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">vllm_tensor_parallel_size</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="c1"># For tensor_parallel_size == 1, we simply set num_gpus=1.</span>
+            <span class="n">resources_kwarg</span><span class="p">[</span><span class="s2">&quot;num_gpus&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">1</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># Otherwise, we have to set num_gpus=0 and provide</span>
+            <span class="c1"># a function that will create a placement group for</span>
+            <span class="c1"># each instance.</span>
+            <span class="n">resources_kwarg</span><span class="p">[</span><span class="s2">&quot;num_gpus&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
+            <span class="n">resources_kwarg</span><span class="p">[</span><span class="s2">&quot;ray_remote_args_fn&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">scheduling_strategy_fn</span>
+            
+        <span class="c1">## predictor</span>
+        <span class="k">class</span><span class="w"> </span><span class="nc">DistributedPredictor</span><span class="p">:</span>
+            <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+                <span class="bp">self</span><span class="p">,</span> 
+                <span class="n">model</span><span class="p">:</span> <span class="n">HFDecoderModel</span><span class="p">,</span>
+                <span class="n">sampling_params</span><span class="p">:</span> <span class="n">SamplingParams</span><span class="p">,</span>
+                <span class="n">vllm_gpu_memory_utilization</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
+                <span class="n">vllm_tensor_parallel_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+                <span class="n">release_gpu</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+            <span class="p">):</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">activate_model_for_inference</span><span class="p">(</span>
+                    <span class="n">use_vllm</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                    <span class="n">vllm_gpu_memory_utilization</span><span class="o">=</span><span class="n">vllm_gpu_memory_utilization</span><span class="p">,</span>
+                    <span class="n">vllm_tensor_parallel_size</span><span class="o">=</span><span class="n">vllm_tensor_parallel_size</span><span class="p">,</span>
+                <span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">sampling_params</span> <span class="o">=</span> <span class="n">sampling_params</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">release_gpu</span> <span class="o">=</span> <span class="n">release_gpu</span>
+                
+            <span class="k">def</span><span class="w"> </span><span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">:</span> <span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">]):</span>
+<span class="w">                </span><span class="sd">&quot;&quot;&quot;batch: Dict[str, np.ndarray], {&quot;item&quot;: array([&#39;...&#39;, &#39;...&#39;, &#39;...&#39;, ...])}</span>
+<span class="sd">                &quot;&quot;&quot;</span>
+                <span class="n">batched_inference_res</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">inference</span><span class="p">(</span>
+                    <span class="n">inputs</span><span class="o">=</span><span class="n">batch</span><span class="p">[</span><span class="s1">&#39;item&#39;</span><span class="p">],</span>
+                    <span class="n">sampling_params</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">sampling_params</span><span class="p">,</span>
+                    <span class="n">release_gpu</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">release_gpu</span><span class="p">,</span>
+                    <span class="n">use_vllm</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+                <span class="p">)</span> <span class="c1"># this is the postprocessed output, see model.__vllm_inference</span>
+                <span class="n">batched_final_res</span> <span class="o">=</span> <span class="p">{</span>
+                    <span class="s2">&quot;input&quot;</span><span class="p">:</span> <span class="p">[</span><span class="n">sample</span><span class="p">[</span><span class="s1">&#39;input&#39;</span><span class="p">]</span> <span class="k">for</span> <span class="n">sample</span> <span class="ow">in</span> <span class="n">batched_inference_res</span><span class="p">],</span>
+                    <span class="s2">&quot;output&quot;</span><span class="p">:</span> <span class="p">[</span><span class="n">sample</span><span class="p">[</span><span class="s1">&#39;output&#39;</span><span class="p">]</span> <span class="k">for</span> <span class="n">sample</span> <span class="ow">in</span> <span class="n">batched_inference_res</span><span class="p">]</span> 
+                <span class="p">}</span> <span class="c1"># do this since we&#39;re writing to a pandas dataframe</span>
+                <span class="k">return</span> <span class="n">batched_final_res</span>
+            
+        <span class="c1"># inference        </span>
+        <span class="n">model_input_mapping</span> <span class="o">=</span> <span class="n">model_input</span><span class="o">.</span><span class="n">map_batches</span><span class="p">(</span>
+            <span class="n">DistributedPredictor</span><span class="p">,</span>
+            <span class="n">concurrency</span><span class="o">=</span><span class="n">num_instances</span><span class="p">,</span> <span class="c1"># Set the concurrency to the number of LLM instances.</span>
+            <span class="n">batch_size</span><span class="o">=</span><span class="n">batch_size</span><span class="p">,</span>
+            <span class="n">fn_constructor_kwargs</span><span class="o">=</span><span class="p">{</span>
+                <span class="s2">&quot;model&quot;</span><span class="p">:</span> <span class="n">model</span><span class="p">,</span>
+                <span class="s2">&quot;sampling_params&quot;</span><span class="p">:</span> <span class="n">sampling_params</span><span class="p">,</span>
+                <span class="s2">&quot;vllm_gpu_memory_utilization&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">vllm_gpu_memory_utilization</span><span class="p">,</span>
+                <span class="s2">&quot;vllm_tensor_parallel_size&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">vllm_tensor_parallel_size</span><span class="p">,</span>
+                <span class="s2">&quot;release_gpu&quot;</span><span class="p">:</span> <span class="n">release_gpu</span><span class="p">,</span>
+            <span class="p">},</span>
+            <span class="o">**</span><span class="n">resources_kwarg</span><span class="p">,</span>
+        <span class="p">)</span>
+        
+        <span class="n">df_model_output</span> <span class="o">=</span> <span class="n">model_input_mapping</span><span class="o">.</span><span class="n">to_pandas</span><span class="p">()</span> <span class="c1"># the actual forwards are executed here</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Distributed vllm inference result preview:</span><span class="se">\n</span><span class="si">{</span><span class="n">df_model_output</span><span class="o">.</span><span class="n">head</span><span class="p">(</span><span class="mi">10</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        
+        <span class="n">model_output</span> <span class="o">=</span> <span class="p">[</span>
+            <span class="p">{</span><span class="s2">&quot;input&quot;</span><span class="p">:</span> <span class="n">row</span><span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">],</span> <span class="s2">&quot;output&quot;</span><span class="p">:</span> <span class="n">row</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">]}</span> <span class="k">for</span> <span class="n">_</span><span class="p">,</span> <span class="n">row</span> <span class="ow">in</span> <span class="n">df_model_output</span><span class="p">[:]</span><span class="o">.</span><span class="n">iterrows</span><span class="p">()</span>
+        <span class="p">]</span>
+        
+        <span class="k">return</span> <span class="n">model_output</span></div>
+
+    
+    
+<div class="viewcode-block" id="VLLMInferencer.save_inference_results">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer.save_inference_results">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">save_inference_results</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">outputs</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]],</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]],</span>
+        <span class="n">save_file_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">save_file_path</span><span class="p">,</span> <span class="s2">&quot;w&quot;</span><span class="p">,</span> <span class="n">encoding</span><span class="o">=</span><span class="s1">&#39;utf-8&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">json</span><span class="o">.</span><span class="n">dump</span><span class="p">(</span><span class="n">outputs</span><span class="p">,</span> <span class="n">f</span><span class="p">,</span> <span class="n">ensure_ascii</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
+            
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Inference results are saved to </span><span class="si">{</span><span class="n">save_file_path</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span></div>
+
+        
+        
+<div class="viewcode-block" id="VLLMInferencer.load_inference_results">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer.load_inference_results">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">load_inference_results</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">results_path</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]],</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]]:</span>
+        <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">results_path</span><span class="p">,</span> <span class="s2">&quot;r&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+            <span class="n">results</span> <span class="o">=</span> <span class="n">json</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">f</span><span class="p">)</span>
+            
+        <span class="k">return</span> <span class="n">results</span></div>
+</div>
+
+        
+        
+<div class="viewcode-block" id="MemorySafeVLLMInferencer">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">MemorySafeVLLMInferencer</span><span class="p">(</span><span class="n">VLLMInferencer</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="p">:</span> <span class="n">ModelArguments</span><span class="p">,</span>
+        <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span>
+        <span class="n">inferencer_args</span><span class="p">:</span> <span class="n">InferencerArguments</span><span class="p">,</span>
+    <span class="p">):</span>
+        <span class="k">assert</span> <span class="n">inferencer_args</span><span class="o">.</span><span class="n">save_results</span><span class="p">,</span> <span class="s2">&quot;For MemorySafeVLLMInferencer, `save_results` must be True.&quot;</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">inferencer_args</span><span class="p">)</span>
+<div class="viewcode-block" id="MemorySafeVLLMInferencer.inferencer_file_path">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer.inferencer_file_path">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_file_path</span> <span class="o">=</span> <span class="n">pkg_resources</span><span class="o">.</span><span class="n">files</span><span class="p">(</span><span class="s2">&quot;lmflow.pipeline.utils&quot;</span><span class="p">)</span> <span class="o">/</span> <span class="s2">&quot;memory_safe_vllm_inference.py&quot;</span></div>
+
+        
+    
+<div class="viewcode-block" id="MemorySafeVLLMInferencer.inference">
+<a class="viewcode-back" href="../../../autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer.inference">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">inference</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">VLLMInferenceResultWithInput</span><span class="p">]:</span>
+        <span class="n">inferencer_args</span> <span class="o">=</span> <span class="n">make_shell_args_from_dataclass</span><span class="p">(</span>
+            <span class="n">dataclass_objects</span><span class="o">=</span><span class="p">[</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">model_args</span><span class="p">,</span> 
+                <span class="bp">self</span><span class="o">.</span><span class="n">data_args</span><span class="p">,</span> 
+                <span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="p">,</span>
+            <span class="p">],</span>
+            <span class="nb">format</span><span class="o">=</span><span class="s2">&quot;shell&quot;</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">cmd</span> <span class="o">=</span> <span class="s2">&quot;python &quot;</span> <span class="o">+</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_file_path</span><span class="p">)</span> <span class="o">+</span> <span class="s2">&quot; &quot;</span> <span class="o">+</span> <span class="n">inferencer_args</span>
+        <span class="n">current_env</span> <span class="o">=</span> <span class="n">os</span><span class="o">.</span><span class="n">environ</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
+        <span class="k">for</span> <span class="n">var</span> <span class="ow">in</span> <span class="n">MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE</span><span class="p">:</span>
+            <span class="n">current_env</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="n">var</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+        
+        <span class="n">cli_res</span> <span class="o">=</span> <span class="n">subprocess</span><span class="o">.</span><span class="n">run</span><span class="p">(</span>
+            <span class="n">args</span><span class="o">=</span><span class="n">cmd</span><span class="p">,</span>
+            <span class="n">stdout</span><span class="o">=</span><span class="n">sys</span><span class="o">.</span><span class="n">stdout</span><span class="p">,</span>
+            <span class="n">stderr</span><span class="o">=</span><span class="n">sys</span><span class="o">.</span><span class="n">stdout</span><span class="p">,</span>
+            <span class="n">shell</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+            <span class="n">preexec_fn</span><span class="o">=</span><span class="n">os</span><span class="o">.</span><span class="n">setsid</span><span class="p">,</span>
+            <span class="n">env</span><span class="o">=</span><span class="n">current_env</span><span class="p">,</span>
+        <span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;MemorySafeVLLMInference subprocess run finished, info at finish: </span><span class="si">{</span><span class="n">cli_res</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        
+        <span class="k">if</span> <span class="n">cli_res</span><span class="o">.</span><span class="n">returncode</span> <span class="ow">in</span> <span class="n">RETURN_CODE_ERROR_BUFFER</span><span class="p">:</span>
+            <span class="c1"># &gt; Fatal Python error: _enter_buffered_busy: could not acquire lock for &lt;_io.BufferedWriter name=&#39;&lt;stdout&gt;&#39;&gt; </span>
+            <span class="c1"># &gt; at interpreter shutdown, possibly due to daemon threads</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                <span class="s2">&quot;^^^^^^^^^^ Please ignore the above error, as it comes from the subprocess. &quot;</span>
+                <span class="s2">&quot;This may due to a kill signal with unfinished stdout/stderr writing in the subprocess. &quot;</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">cli_res</span><span class="o">.</span><span class="n">returncode</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Error during MemorySafeVLLMInference: </span><span class="si">{</span><span class="n">cli_res</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                
+        <span class="n">outputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">load_inference_results</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">inferencer_args</span><span class="o">.</span><span class="n">results_path</span><span class="p">)</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="s2">&quot;MemorySafeVLLMInference result captured.&quot;</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">outputs</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/tokenization/hf_decoder_model.html b/_modules/lmflow/tokenization/hf_decoder_model.html
new file mode 100644
index 000000000..34cd4a09f
--- /dev/null
+++ b/_modules/lmflow/tokenization/hf_decoder_model.html
@@ -0,0 +1,673 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.tokenization.hf_decoder_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/tokenization/hf_decoder_model';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.tokenization.hf_decoder_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.tokenization.hf_decoder_model</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">logging</span><span class="w"> </span><span class="kn">import</span> <span class="n">Logger</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">transformers</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.testing_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">CaptureLogger</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">PreTrainedTokenizer</span><span class="p">,</span> <span class="n">PreTrainedTokenizerFast</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.conversation_template</span><span class="w"> </span><span class="kn">import</span> <span class="n">ConversationTemplate</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="n">CONVERSATION_ROLE_NAMES</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="n">DatasetArguments</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_decoder_model/index.html#lmflow.tokenization.hf_decoder_model.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+<div class="viewcode-block" id="tok_logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_decoder_model/index.html#lmflow.tokenization.hf_decoder_model.tok_logger">[docs]</a>
+<span class="n">tok_logger</span> <span class="o">=</span> <span class="n">transformers</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">logging</span><span class="o">.</span><span class="n">get_logger</span><span class="p">(</span><span class="s2">&quot;transformers.tokenization_utils_base&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="blocking">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_decoder_model/index.html#lmflow.tokenization.hf_decoder_model.blocking">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">blocking</span><span class="p">(</span>
+    <span class="n">token_dict</span><span class="p">:</span> <span class="n">Dict</span><span class="p">,</span> 
+    <span class="n">block_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> 
+    <span class="n">model_max_length</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">pad_token_id</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">padding_side</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="n">truncation_side</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s1">&#39;right&#39;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+    <span class="n">block_size_warning_num</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="n">num_example</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">token_dict</span><span class="p">[</span><span class="nb">list</span><span class="p">(</span><span class="n">token_dict</span><span class="o">.</span><span class="n">keys</span><span class="p">())[</span><span class="mi">0</span><span class="p">]])</span>
+    <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">):</span>
+        <span class="n">max_length</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">block_size</span><span class="p">,</span> <span class="n">model_max_length</span><span class="p">)</span>
+        <span class="n">pad_length</span> <span class="o">=</span> <span class="n">max_length</span> <span class="o">-</span> <span class="nb">len</span><span class="p">(</span><span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">])</span>
+        <span class="k">if</span> <span class="n">block_size</span> <span class="o">&lt;</span> <span class="n">model_max_length</span><span class="p">:</span>
+            <span class="n">block_size_warning_num</span> <span class="o">+=</span> <span class="mi">1</span>
+        <span class="k">if</span> <span class="n">pad_length</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="c1"># Truncates too long samples</span>
+            <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">,</span> <span class="s2">&quot;attention_mask&quot;</span><span class="p">,</span> <span class="s2">&quot;labels&quot;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">truncation_side</span> <span class="o">==</span> <span class="s1">&#39;right&#39;</span><span class="p">:</span>
+                    <span class="n">token_dict</span><span class="p">[</span><span class="n">key</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="n">token_dict</span><span class="p">[</span><span class="n">key</span><span class="p">][</span><span class="n">i</span><span class="p">][:</span><span class="n">max_length</span><span class="p">]</span>
+                <span class="k">elif</span> <span class="n">truncation_side</span> <span class="o">==</span> <span class="s1">&#39;left&#39;</span><span class="p">:</span>
+                    <span class="n">token_dict</span><span class="p">[</span><span class="n">key</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="n">token_dict</span><span class="p">[</span><span class="n">key</span><span class="p">][</span><span class="n">i</span><span class="p">][</span><span class="o">-</span><span class="n">max_length</span><span class="p">:]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;truncation_side should be either &#39;right&#39; or &#39;left&#39;, got </span><span class="si">{</span><span class="n">truncation_side</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">padding_side</span> <span class="o">==</span> <span class="s1">&#39;right&#39;</span><span class="p">:</span>
+                <span class="c1"># Pads too short samples</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+                    <span class="p">[</span><span class="n">pad_token_id</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span>
+                <span class="p">)</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+                    <span class="p">[</span><span class="mi">0</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span>
+                <span class="p">)</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+                    <span class="p">[</span><span class="o">-</span><span class="mi">100</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span>
+                <span class="p">)</span>
+            <span class="k">elif</span> <span class="n">padding_side</span> <span class="o">==</span> <span class="s1">&#39;left&#39;</span><span class="p">:</span>
+                <span class="c1"># Pads too short samples</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span>
+                    <span class="p">[</span><span class="n">pad_token_id</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span> <span class="o">+</span> <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+                <span class="p">)</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span>
+                    <span class="p">[</span><span class="mi">0</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span> <span class="o">+</span> <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+                <span class="p">)</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span>
+                    <span class="p">[</span><span class="o">-</span><span class="mi">100</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span> <span class="o">+</span> <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;padding_side should be either &#39;right&#39; or &#39;left&#39;, got </span><span class="si">{</span><span class="n">padding_side</span><span class="si">}</span><span class="s2">&quot;</span>
+                <span class="p">)</span>
+    <span class="k">if</span> <span class="n">block_size_warning_num</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;There are </span><span class="si">{</span><span class="n">block_size_warning_num</span><span class="si">}</span><span class="s2"> of </span><span class="si">{</span><span class="n">num_example</span><span class="si">}</span><span class="s2"> samples where&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot;block_size </span><span class="si">{</span><span class="n">block_size</span><span class="si">}</span><span class="s2"> &lt; model_max_length&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">model_max_length</span><span class="si">}</span><span class="s2">, use block_size&quot;</span>
+            <span class="s2">&quot; for maximum tokenized sequence length&quot;</span>
+        <span class="p">)</span>
+        
+    <span class="k">return</span> <span class="n">token_dict</span></div>
+
+
+
+<div class="viewcode-block" id="tokenize_function">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_decoder_model/index.html#lmflow.tokenization.hf_decoder_model.tokenize_function">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">tokenize_function</span><span class="p">(</span>
+    <span class="n">examples</span><span class="p">,</span> 
+    <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span>
+    <span class="n">tokenizer</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">PreTrainedTokenizer</span><span class="p">,</span> <span class="n">PreTrainedTokenizerFast</span><span class="p">],</span>
+    <span class="n">column_names</span><span class="p">,</span>
+    <span class="n">label_columns</span><span class="p">,</span>
+    <span class="n">tokenized_column_order</span><span class="p">,</span>
+    <span class="n">add_special_tokens</span><span class="p">,</span>
+    <span class="n">use_truncation</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Handels text_only and text2text datasets tokenization</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">num_example</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">examples</span><span class="p">[</span><span class="n">column_names</span><span class="p">[</span><span class="mi">0</span><span class="p">]])</span>
+    <span class="n">token_dict</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;input_ids&quot;</span><span class="p">:</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)],</span>
+        <span class="s2">&quot;attention_mask&quot;</span><span class="p">:</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)],</span>
+        <span class="s2">&quot;labels&quot;</span><span class="p">:</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)],</span>
+    <span class="p">}</span>
+    <span class="k">with</span> <span class="n">CaptureLogger</span><span class="p">(</span><span class="n">tok_logger</span><span class="p">)</span> <span class="k">as</span> <span class="n">cl</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">column_name</span> <span class="ow">in</span> <span class="n">tokenized_column_order</span><span class="p">:</span>
+            <span class="n">encoding</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">(</span>
+                <span class="n">examples</span><span class="p">[</span><span class="n">column_name</span><span class="p">],</span>
+                <span class="n">add_special_tokens</span><span class="o">=</span><span class="n">add_special_tokens</span><span class="p">,</span>
+                <span class="n">truncation</span><span class="o">=</span><span class="n">use_truncation</span><span class="p">,</span>
+            <span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">column_name</span> <span class="ow">in</span> <span class="n">label_columns</span><span class="p">:</span>
+                <span class="n">labels</span> <span class="o">=</span> <span class="n">encoding</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">labels</span> <span class="o">=</span> <span class="p">[</span>
+                    <span class="p">[</span><span class="o">-</span><span class="mi">100</span><span class="p">]</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">encoding</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">])</span>
+                    <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)</span>
+                <span class="p">]</span>
+
+            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">):</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+                    <span class="n">encoding</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+                <span class="p">)</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+                    <span class="n">encoding</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+                <span class="p">)</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">labels</span><span class="p">[</span><span class="n">i</span><span class="p">])</span>
+                
+    <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+        <span class="n">token_dict</span> <span class="o">=</span> <span class="n">blocking</span><span class="p">(</span>
+            <span class="n">token_dict</span><span class="o">=</span><span class="n">token_dict</span><span class="p">,</span>
+            <span class="n">block_size</span><span class="o">=</span><span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span><span class="p">,</span>
+            <span class="n">model_max_length</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="p">,</span>
+            <span class="n">pad_token_id</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span><span class="p">,</span>
+            <span class="n">padding_side</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">padding_side</span><span class="p">,</span>
+            <span class="n">truncation_side</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">truncation_side</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="c1"># clm input could be much much longer than block_size</span>
+    <span class="k">if</span> <span class="s2">&quot;Token indices sequence length is longer than the&quot;</span> <span class="ow">in</span> <span class="n">cl</span><span class="o">.</span><span class="n">out</span><span class="p">:</span>
+        <span class="n">tok_logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+            <span class="s2">&quot;^^^^^^^^^^^^^^^^ Please ignore the warning above - this long input will be chunked into smaller bits&quot;</span>
+            <span class="s2">&quot; before being passed to the model.&quot;</span>
+        <span class="p">)</span>
+    <span class="k">return</span> <span class="n">token_dict</span></div>
+
+    
+
+<div class="viewcode-block" id="conversation_tokenize_function">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_decoder_model/index.html#lmflow.tokenization.hf_decoder_model.conversation_tokenize_function">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">conversation_tokenize_function</span><span class="p">(</span>
+    <span class="n">examples</span><span class="p">,</span> 
+    <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span>
+    <span class="n">tokenizer</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">PreTrainedTokenizer</span><span class="p">,</span> <span class="n">PreTrainedTokenizerFast</span><span class="p">],</span> 
+    <span class="n">column_names</span><span class="p">,</span>
+    <span class="n">conversation_template</span><span class="p">:</span> <span class="n">ConversationTemplate</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Handels conversation datasets tokenization</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">num_example</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">examples</span><span class="p">[</span><span class="n">column_names</span><span class="p">[</span><span class="mi">0</span><span class="p">]])</span>
+    <span class="n">token_dict</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;input_ids&quot;</span><span class="p">:</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)],</span>
+        <span class="s2">&quot;attention_mask&quot;</span><span class="p">:</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)],</span>
+        <span class="s2">&quot;labels&quot;</span><span class="p">:</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)],</span>
+    <span class="p">}</span>
+    <span class="k">with</span> <span class="n">CaptureLogger</span><span class="p">(</span><span class="n">tok_logger</span><span class="p">)</span> <span class="k">as</span> <span class="n">cl</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">examples</span><span class="p">[</span><span class="s2">&quot;messages&quot;</span><span class="p">])):</span>
+            <span class="n">messages</span> <span class="o">=</span> <span class="n">examples</span><span class="p">[</span><span class="s2">&quot;messages&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+            <span class="n">system</span> <span class="o">=</span> <span class="n">examples</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;system&quot;</span><span class="p">,</span> <span class="p">[</span><span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">num_example</span><span class="p">)[</span><span class="n">i</span><span class="p">]</span>
+            <span class="n">tools</span> <span class="o">=</span> <span class="n">examples</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;tools&quot;</span><span class="p">,</span> <span class="p">[</span><span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">num_example</span><span class="p">)[</span><span class="n">i</span><span class="p">]</span>
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">messages</span><span class="p">)</span> <span class="o">&lt;</span> <span class="mi">2</span> <span class="ow">or</span> <span class="n">messages</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">!=</span> <span class="n">CONVERSATION_ROLE_NAMES</span><span class="p">[</span><span class="s1">&#39;user&#39;</span><span class="p">]:</span>
+                <span class="n">tok_logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="s2">&quot;Invalid instance encountered. Either the conversation has less than &quot;</span>
+                    <span class="s2">&quot;one round or the first message is not from the user.&quot;</span>
+                <span class="p">)</span>
+                <span class="k">continue</span>
+        
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">messages</span><span class="p">)</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="s2">&quot;The number of messages is not even, the last message will be ignored.&quot;</span>
+                <span class="p">)</span>
+                <span class="n">messages</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+                
+            <span class="n">encoded_conversation</span> <span class="o">=</span> <span class="n">conversation_template</span><span class="o">.</span><span class="n">encode_conversation</span><span class="p">(</span>
+                <span class="n">tokenizer</span><span class="o">=</span><span class="n">tokenizer</span><span class="p">,</span>
+                <span class="n">messages</span><span class="o">=</span><span class="n">messages</span><span class="p">,</span>
+                <span class="n">system</span><span class="o">=</span><span class="n">system</span><span class="p">,</span>
+                <span class="n">tools</span><span class="o">=</span><span class="n">tools</span><span class="p">,</span>
+            <span class="p">)</span>
+
+            <span class="n">input_ids</span><span class="p">,</span> <span class="n">labels</span> <span class="o">=</span> <span class="p">[],</span> <span class="p">[]</span>
+            <span class="k">for</span> <span class="n">turn_idx</span><span class="p">,</span> <span class="p">(</span><span class="n">user_input</span><span class="p">,</span> <span class="n">assistant_result</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">encoded_conversation</span><span class="p">):</span>
+                <span class="n">input_ids</span> <span class="o">+=</span> <span class="n">user_input</span> <span class="o">+</span> <span class="n">assistant_result</span>
+                
+                <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">train_on_prompt</span><span class="p">:</span>
+                    <span class="n">labels</span> <span class="o">+=</span> <span class="n">user_input</span> <span class="o">+</span> <span class="n">assistant_result</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">labels</span> <span class="o">+=</span> <span class="p">[</span><span class="o">-</span><span class="mi">100</span><span class="p">]</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">user_input</span><span class="p">)</span> <span class="o">+</span> <span class="n">assistant_result</span>
+                
+            <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">input_ids</span><span class="p">)</span>
+            <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">([</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">input_ids</span><span class="p">))</span>
+            <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">labels</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+        <span class="n">token_dict</span> <span class="o">=</span> <span class="n">blocking</span><span class="p">(</span>
+            <span class="n">token_dict</span><span class="o">=</span><span class="n">token_dict</span><span class="p">,</span>
+            <span class="n">block_size</span><span class="o">=</span><span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span><span class="p">,</span>
+            <span class="n">model_max_length</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="p">,</span>
+            <span class="n">pad_token_id</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span><span class="p">,</span>
+            <span class="n">padding_side</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">padding_side</span><span class="p">,</span>
+            <span class="n">truncation_side</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">truncation_side</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="c1"># clm input could be much much longer than block_size</span>
+    <span class="k">if</span> <span class="s2">&quot;Token indices sequence length is longer than the&quot;</span> <span class="ow">in</span> <span class="n">cl</span><span class="o">.</span><span class="n">out</span><span class="p">:</span>
+        <span class="n">tok_logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+            <span class="s2">&quot;^^^^^^^^^^^^^^^^ Please ignore the warning above - this long input will be chunked into smaller bits&quot;</span>
+            <span class="s2">&quot; before being passed to the model.&quot;</span>
+        <span class="p">)</span>
+    <span class="k">return</span> <span class="n">token_dict</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/tokenization/hf_text_regression_model.html b/_modules/lmflow/tokenization/hf_text_regression_model.html
new file mode 100644
index 000000000..a68025e6a
--- /dev/null
+++ b/_modules/lmflow/tokenization/hf_text_regression_model.html
@@ -0,0 +1,914 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.tokenization.hf_text_regression_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/tokenization/hf_text_regression_model';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.tokenization.hf_text_regression_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.tokenization.hf_text_regression_model</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">logging</span><span class="w"> </span><span class="kn">import</span> <span class="n">Logger</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">transformers</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.testing_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">CaptureLogger</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">PreTrainedTokenizer</span><span class="p">,</span> <span class="n">PreTrainedTokenizerFast</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.conversation_template</span><span class="w"> </span><span class="kn">import</span> <span class="n">ConversationTemplate</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="n">CONVERSATION_ROLE_NAMES</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="n">DatasetArguments</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+<div class="viewcode-block" id="tok_logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.tok_logger">[docs]</a>
+<span class="n">tok_logger</span> <span class="o">=</span> <span class="n">transformers</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">logging</span><span class="o">.</span><span class="n">get_logger</span><span class="p">(</span><span class="s2">&quot;transformers.tokenization_utils_base&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="blocking_paired">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.blocking_paired">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">blocking_paired</span><span class="p">(</span>
+    <span class="n">token_dict</span><span class="p">:</span> <span class="n">Dict</span><span class="p">,</span> 
+    <span class="n">column_names</span><span class="p">:</span> <span class="n">List</span><span class="p">,</span>
+    <span class="n">block_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> 
+    <span class="n">model_max_length</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">pad_token_id</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">padding_side</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="n">truncation_side</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s1">&#39;right&#39;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+    <span class="n">block_size_warning_num</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="n">num_example</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">token_dict</span><span class="p">[</span><span class="nb">list</span><span class="p">(</span><span class="n">token_dict</span><span class="o">.</span><span class="n">keys</span><span class="p">())[</span><span class="mi">0</span><span class="p">]])</span>
+    <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">column_name</span> <span class="ow">in</span> <span class="n">column_names</span><span class="p">:</span>
+            <span class="n">max_length</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">block_size</span><span class="p">,</span> <span class="n">model_max_length</span><span class="p">)</span>
+            <span class="n">pad_length</span> <span class="o">=</span> <span class="n">max_length</span> <span class="o">-</span> <span class="nb">len</span><span class="p">(</span><span class="n">token_dict</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;input_ids_</span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">])</span>
+            <span class="k">if</span> <span class="n">block_size</span> <span class="o">&lt;</span> <span class="n">model_max_length</span><span class="p">:</span>
+                <span class="n">block_size_warning_num</span> <span class="o">+=</span> <span class="mi">1</span>
+            <span class="k">if</span> <span class="n">pad_length</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="c1"># Truncates too long samples</span>
+                <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="p">[</span><span class="sa">f</span><span class="s2">&quot;input_ids_</span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;attention_mask_</span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">]:</span>
+                    <span class="k">if</span> <span class="n">truncation_side</span> <span class="o">==</span> <span class="s1">&#39;right&#39;</span><span class="p">:</span>
+                        <span class="n">token_dict</span><span class="p">[</span><span class="n">key</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="n">token_dict</span><span class="p">[</span><span class="n">key</span><span class="p">][</span><span class="n">i</span><span class="p">][:</span><span class="n">pad_length</span><span class="p">]</span>
+                    <span class="k">elif</span> <span class="n">truncation_side</span> <span class="o">==</span> <span class="s1">&#39;left&#39;</span><span class="p">:</span>
+                        <span class="n">token_dict</span><span class="p">[</span><span class="n">key</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="n">token_dict</span><span class="p">[</span><span class="n">key</span><span class="p">][</span><span class="n">i</span><span class="p">][</span><span class="o">-</span><span class="n">pad_length</span><span class="p">:]</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                            <span class="sa">f</span><span class="s2">&quot;truncation_side should be either &#39;right&#39; or &#39;left&#39;, got </span><span class="si">{</span><span class="n">truncation_side</span><span class="si">}</span><span class="s2">&quot;</span>
+                        <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">padding_side</span> <span class="o">==</span> <span class="s1">&#39;right&#39;</span><span class="p">:</span>
+                    <span class="c1"># Pads too short samples</span>
+                    <span class="n">token_dict</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;input_ids_</span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+                        <span class="p">[</span><span class="n">pad_token_id</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span>
+                    <span class="p">)</span>
+                    <span class="n">token_dict</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;attention_mask_</span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+                        <span class="p">[</span><span class="mi">0</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span>
+                    <span class="p">)</span>
+                <span class="k">elif</span> <span class="n">padding_side</span> <span class="o">==</span> <span class="s1">&#39;left&#39;</span><span class="p">:</span>
+                    <span class="c1"># Pads too short samples</span>
+                    <span class="n">token_dict</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;input_ids_</span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span>
+                        <span class="p">[</span><span class="n">pad_token_id</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span> <span class="o">+</span> <span class="n">token_dict</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;input_ids_</span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+                    <span class="p">)</span>
+                    <span class="n">token_dict</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;attention_mask_</span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span>
+                        <span class="p">[</span><span class="mi">0</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span> <span class="o">+</span> <span class="n">token_dict</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;attention_mask_</span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+                    <span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;padding_side should be either &#39;right&#39; or &#39;left&#39;, got </span><span class="si">{</span><span class="n">padding_side</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="p">)</span>
+    <span class="k">if</span> <span class="n">block_size_warning_num</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;There are </span><span class="si">{</span><span class="n">block_size_warning_num</span><span class="si">}</span><span class="s2"> of </span><span class="si">{</span><span class="n">num_example</span><span class="si">}</span><span class="s2"> samples where&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot; block_size </span><span class="si">{</span><span class="n">block_size</span><span class="si">}</span><span class="s2"> &lt; model_max_length&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">model_max_length</span><span class="si">}</span><span class="s2">, use block_size&quot;</span>
+            <span class="s2">&quot; for maximum tokenized sequence length&quot;</span>
+        <span class="p">)</span>
+        
+    <span class="k">return</span> <span class="n">token_dict</span></div>
+
+
+
+<div class="viewcode-block" id="blocking">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.blocking">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">blocking</span><span class="p">(</span>
+    <span class="n">token_dict</span><span class="p">:</span> <span class="n">Dict</span><span class="p">,</span> 
+    <span class="n">block_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> 
+    <span class="n">model_max_length</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">pad_token_id</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">padding_side</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="n">truncation_side</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s1">&#39;right&#39;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+    <span class="n">block_size_warning_num</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="n">num_example</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">token_dict</span><span class="p">[</span><span class="nb">list</span><span class="p">(</span><span class="n">token_dict</span><span class="o">.</span><span class="n">keys</span><span class="p">())[</span><span class="mi">0</span><span class="p">]])</span>
+    <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">):</span>
+        <span class="n">max_length</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">block_size</span><span class="p">,</span> <span class="n">model_max_length</span><span class="p">)</span>
+        <span class="n">pad_length</span> <span class="o">=</span> <span class="n">max_length</span> <span class="o">-</span> <span class="nb">len</span><span class="p">(</span><span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">])</span>
+        <span class="k">if</span> <span class="n">block_size</span> <span class="o">&lt;</span> <span class="n">model_max_length</span><span class="p">:</span>
+            <span class="n">block_size_warning_num</span> <span class="o">+=</span> <span class="mi">1</span>
+        <span class="k">if</span> <span class="n">pad_length</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="c1"># Truncates too long samples</span>
+            <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">,</span> <span class="s2">&quot;attention_mask&quot;</span><span class="p">,</span> <span class="s2">&quot;labels&quot;</span><span class="p">]:</span>
+                <span class="k">if</span> <span class="n">truncation_side</span> <span class="o">==</span> <span class="s1">&#39;right&#39;</span><span class="p">:</span>
+                    <span class="n">token_dict</span><span class="p">[</span><span class="n">key</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="n">token_dict</span><span class="p">[</span><span class="n">key</span><span class="p">][</span><span class="n">i</span><span class="p">][:</span><span class="n">pad_length</span><span class="p">]</span>
+                <span class="k">elif</span> <span class="n">truncation_side</span> <span class="o">==</span> <span class="s1">&#39;left&#39;</span><span class="p">:</span>
+                    <span class="n">token_dict</span><span class="p">[</span><span class="n">key</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="n">token_dict</span><span class="p">[</span><span class="n">key</span><span class="p">][</span><span class="n">i</span><span class="p">][</span><span class="o">-</span><span class="n">pad_length</span><span class="p">:]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;truncation_side should be either &#39;right&#39; or &#39;left&#39;, got </span><span class="si">{</span><span class="n">truncation_side</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">padding_side</span> <span class="o">==</span> <span class="s1">&#39;right&#39;</span><span class="p">:</span>
+                <span class="c1"># Pads too short samples</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+                    <span class="p">[</span><span class="n">pad_token_id</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span>
+                <span class="p">)</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+                    <span class="p">[</span><span class="mi">0</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span>
+                <span class="p">)</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+                    <span class="p">[</span><span class="o">-</span><span class="mi">100</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span>
+                <span class="p">)</span>
+            <span class="k">elif</span> <span class="n">padding_side</span> <span class="o">==</span> <span class="s1">&#39;left&#39;</span><span class="p">:</span>
+                <span class="c1"># Pads too short samples</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span>
+                    <span class="p">[</span><span class="n">pad_token_id</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span> <span class="o">+</span> <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+                <span class="p">)</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span>
+                    <span class="p">[</span><span class="mi">0</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span> <span class="o">+</span> <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+                <span class="p">)</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span>
+                    <span class="p">[</span><span class="o">-</span><span class="mi">100</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span> <span class="o">+</span> <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+                <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;padding_side should be either &#39;right&#39; or &#39;left&#39;, got </span><span class="si">{</span><span class="n">padding_side</span><span class="si">}</span><span class="s2">&quot;</span>
+                <span class="p">)</span>
+    <span class="k">if</span> <span class="n">block_size_warning_num</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;There are </span><span class="si">{</span><span class="n">block_size_warning_num</span><span class="si">}</span><span class="s2"> of </span><span class="si">{</span><span class="n">num_example</span><span class="si">}</span><span class="s2"> samples where&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot; block_size </span><span class="si">{</span><span class="n">block_size</span><span class="si">}</span><span class="s2"> &lt; model_max_length&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">model_max_length</span><span class="si">}</span><span class="s2">, use block_size&quot;</span>
+            <span class="s2">&quot; for maximum tokenized sequence length&quot;</span>
+        <span class="p">)</span>
+        
+    <span class="k">return</span> <span class="n">token_dict</span></div>
+
+
+
+<div class="viewcode-block" id="blocking_text_to_textlist">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.blocking_text_to_textlist">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">blocking_text_to_textlist</span><span class="p">(</span>
+    <span class="n">token_dict</span><span class="p">:</span> <span class="n">Dict</span><span class="p">,</span> 
+    <span class="n">block_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> 
+    <span class="n">model_max_length</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">pad_token_id</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
+    <span class="n">padding_side</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
+    <span class="n">truncation_side</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s1">&#39;right&#39;</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+    <span class="n">block_size_warning_num</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="n">num_example</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">token_dict</span><span class="p">[</span><span class="nb">list</span><span class="p">(</span><span class="n">token_dict</span><span class="o">.</span><span class="n">keys</span><span class="p">())[</span><span class="mi">0</span><span class="p">]])</span>
+    <span class="n">max_length</span> <span class="o">=</span> <span class="nb">min</span><span class="p">(</span><span class="n">block_size</span><span class="p">,</span> <span class="n">model_max_length</span><span class="p">)</span>
+    
+    <span class="k">for</span> <span class="n">example_idx</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">):</span>
+        <span class="k">for</span> <span class="n">content_idx</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">example_idx</span><span class="p">])):</span>
+            <span class="n">pad_length</span> <span class="o">=</span> <span class="n">max_length</span> <span class="o">-</span> <span class="nb">len</span><span class="p">(</span><span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">example_idx</span><span class="p">][</span><span class="n">content_idx</span><span class="p">])</span>
+            <span class="k">if</span> <span class="n">block_size</span> <span class="o">&lt;</span> <span class="n">model_max_length</span><span class="p">:</span>
+                <span class="n">block_size_warning_num</span> <span class="o">+=</span> <span class="mi">1</span>
+            <span class="k">if</span> <span class="n">pad_length</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="c1"># Truncates too long samples</span>
+                <span class="k">if</span> <span class="n">truncation_side</span> <span class="o">==</span> <span class="s1">&#39;right&#39;</span><span class="p">:</span>
+                    <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">example_idx</span><span class="p">][</span><span class="n">content_idx</span><span class="p">]</span> <span class="o">=</span> <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">example_idx</span><span class="p">][</span><span class="n">content_idx</span><span class="p">][:</span><span class="n">pad_length</span><span class="p">]</span>
+                <span class="k">elif</span> <span class="n">truncation_side</span> <span class="o">==</span> <span class="s1">&#39;left&#39;</span><span class="p">:</span>
+                    <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">example_idx</span><span class="p">][</span><span class="n">content_idx</span><span class="p">]</span> <span class="o">=</span> <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">example_idx</span><span class="p">][</span><span class="n">content_idx</span><span class="p">][</span><span class="o">-</span><span class="n">pad_length</span><span class="p">:]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;truncation_side should be either &#39;right&#39; or &#39;left&#39;, got </span><span class="si">{</span><span class="n">truncation_side</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">padding_side</span> <span class="o">==</span> <span class="s1">&#39;right&#39;</span><span class="p">:</span>
+                    <span class="c1"># Pads too short samples</span>
+                    <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">example_idx</span><span class="p">][</span><span class="n">content_idx</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+                        <span class="p">[</span><span class="n">pad_token_id</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span>
+                    <span class="p">)</span>
+                <span class="k">elif</span> <span class="n">padding_side</span> <span class="o">==</span> <span class="s1">&#39;left&#39;</span><span class="p">:</span>
+                    <span class="c1"># Pads too short samples</span>
+                    <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">example_idx</span><span class="p">][</span><span class="n">content_idx</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span>
+                        <span class="p">[</span><span class="n">pad_token_id</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">pad_length</span><span class="p">)]</span> <span class="o">+</span> <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">example_idx</span><span class="p">][</span><span class="n">content_idx</span><span class="p">]</span>
+                    <span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;padding_side should be either &#39;right&#39; or &#39;left&#39;, got </span><span class="si">{</span><span class="n">padding_side</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="p">)</span>
+    <span class="k">if</span> <span class="n">block_size_warning_num</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;There are </span><span class="si">{</span><span class="n">block_size_warning_num</span><span class="si">}</span><span class="s2"> of </span><span class="si">{</span><span class="n">num_example</span><span class="si">}</span><span class="s2"> samples where&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot; block_size </span><span class="si">{</span><span class="n">block_size</span><span class="si">}</span><span class="s2"> &lt; model_max_length&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">model_max_length</span><span class="si">}</span><span class="s2">, use block_size&quot;</span>
+            <span class="s2">&quot; for maximum tokenized sequence length&quot;</span>
+        <span class="p">)</span>
+        
+    <span class="k">return</span> <span class="n">token_dict</span></div>
+
+
+
+<div class="viewcode-block" id="paired_conversation_tokenize_function">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.paired_conversation_tokenize_function">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">paired_conversation_tokenize_function</span><span class="p">(</span>
+    <span class="n">examples</span><span class="p">,</span> 
+    <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span>
+    <span class="n">tokenizer</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">PreTrainedTokenizer</span><span class="p">,</span> <span class="n">PreTrainedTokenizerFast</span><span class="p">],</span> 
+    <span class="n">column_names</span><span class="p">,</span>
+    <span class="n">conversation_template</span><span class="p">:</span> <span class="n">ConversationTemplate</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+    <span class="n">num_example</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">examples</span><span class="p">[</span><span class="n">column_names</span><span class="p">[</span><span class="mi">0</span><span class="p">]])</span>
+    <span class="n">token_dict</span> <span class="o">=</span> <span class="p">{}</span>
+    <span class="k">for</span> <span class="n">column_name</span> <span class="ow">in</span> <span class="n">column_names</span><span class="p">:</span>
+        <span class="n">token_dict</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;input_ids_</span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)]</span>
+        <span class="n">token_dict</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;attention_mask_</span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)]</span>
+        
+    <span class="k">with</span> <span class="n">CaptureLogger</span><span class="p">(</span><span class="n">tok_logger</span><span class="p">)</span> <span class="k">as</span> <span class="n">cl</span><span class="p">:</span>
+        <span class="n">num_corrupted</span> <span class="o">=</span> <span class="mi">0</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">):</span>
+            <span class="k">try</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">column_name</span> <span class="ow">in</span> <span class="n">column_names</span><span class="p">:</span>
+                    <span class="n">messages</span> <span class="o">=</span> <span class="n">examples</span><span class="p">[</span><span class="n">column_name</span><span class="p">][</span><span class="n">i</span><span class="p">][</span><span class="s2">&quot;messages&quot;</span><span class="p">]</span>
+                    <span class="n">system</span> <span class="o">=</span> <span class="n">examples</span><span class="p">[</span><span class="n">column_name</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;system&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+                    <span class="n">tools</span> <span class="o">=</span> <span class="n">examples</span><span class="p">[</span><span class="n">column_name</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;tools&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">messages</span><span class="p">)</span> <span class="o">&lt;</span> <span class="mi">2</span> <span class="ow">or</span> <span class="n">messages</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">!=</span> <span class="n">CONVERSATION_ROLE_NAMES</span><span class="p">[</span><span class="s1">&#39;user&#39;</span><span class="p">]:</span>
+                        <span class="n">tok_logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                            <span class="s2">&quot;Invalid instance encountered. Either the conversation has less than &quot;</span>
+                            <span class="s2">&quot;one round or the first message is not from the user.&quot;</span>
+                        <span class="p">)</span>
+                        <span class="k">continue</span>
+                
+                    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">messages</span><span class="p">)</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                            <span class="s2">&quot;The number of messages is not even, the last message will be ignored.&quot;</span>
+                        <span class="p">)</span>
+                        <span class="n">messages</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+                    
+                    <span class="n">encoded_conversation</span> <span class="o">=</span> <span class="n">conversation_template</span><span class="o">.</span><span class="n">encode_conversation</span><span class="p">(</span>
+                        <span class="n">tokenizer</span><span class="o">=</span><span class="n">tokenizer</span><span class="p">,</span>
+                        <span class="n">messages</span><span class="o">=</span><span class="n">messages</span><span class="p">,</span>
+                        <span class="n">system</span><span class="o">=</span><span class="n">system</span><span class="p">,</span>
+                        <span class="n">tools</span><span class="o">=</span><span class="n">tools</span><span class="p">,</span>
+                    <span class="p">)</span>
+
+                    <span class="n">input_ids</span> <span class="o">=</span> <span class="p">[]</span>
+                    <span class="k">for</span> <span class="n">turn_idx</span><span class="p">,</span> <span class="p">(</span><span class="n">user_input</span><span class="p">,</span> <span class="n">assistant_result</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">encoded_conversation</span><span class="p">):</span>
+                        <span class="n">input_ids</span> <span class="o">+=</span> <span class="n">user_input</span> <span class="o">+</span> <span class="n">assistant_result</span>
+                        
+                    <span class="n">token_dict</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;input_ids_</span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">input_ids</span><span class="p">)</span>
+                    <span class="n">token_dict</span><span class="p">[</span><span class="sa">f</span><span class="s2">&quot;attention_mask_</span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">([</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">input_ids</span><span class="p">))</span>
+                    
+            <span class="k">except</span><span class="p">:</span>
+                <span class="n">num_corrupted</span> <span class="o">+=</span> <span class="mi">1</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">error</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Error in encoding conversation </span><span class="si">{</span><span class="n">i</span><span class="si">}</span><span class="s2">: </span><span class="si">{</span><span class="n">column_name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">error</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Messages: </span><span class="si">{</span><span class="n">messages</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                <span class="k">continue</span>
+        <span class="k">if</span> <span class="n">num_corrupted</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">error</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Number of corrupted examples: </span><span class="si">{</span><span class="n">num_corrupted</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                
+    <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+        <span class="n">token_dict</span> <span class="o">=</span> <span class="n">blocking_paired</span><span class="p">(</span>
+            <span class="n">token_dict</span><span class="o">=</span><span class="n">token_dict</span><span class="p">,</span>
+            <span class="n">column_names</span><span class="o">=</span><span class="n">column_names</span><span class="p">,</span>
+            <span class="n">block_size</span><span class="o">=</span><span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span><span class="p">,</span>
+            <span class="n">model_max_length</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="p">,</span>
+            <span class="n">pad_token_id</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span><span class="p">,</span>
+            <span class="n">padding_side</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">padding_side</span><span class="p">,</span>
+            <span class="n">truncation_side</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">truncation_side</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="c1"># clm input could be much much longer than block_size</span>
+    <span class="k">if</span> <span class="s2">&quot;Token indices sequence length is longer than the&quot;</span> <span class="ow">in</span> <span class="n">cl</span><span class="o">.</span><span class="n">out</span><span class="p">:</span>
+        <span class="n">tok_logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+            <span class="s2">&quot;^^^^^^^^^^^^^^^^ Please ignore the warning above - this long input will be chunked into smaller bits&quot;</span>
+            <span class="s2">&quot; before being passed to the model.&quot;</span>
+        <span class="p">)</span>
+    <span class="k">return</span> <span class="n">token_dict</span></div>
+
+
+
+<div class="viewcode-block" id="conversation_tokenize_function">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.conversation_tokenize_function">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">conversation_tokenize_function</span><span class="p">(</span>
+    <span class="n">examples</span><span class="p">,</span> 
+    <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span>
+    <span class="n">tokenizer</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">PreTrainedTokenizer</span><span class="p">,</span> <span class="n">PreTrainedTokenizerFast</span><span class="p">],</span> 
+    <span class="n">column_names</span><span class="p">,</span>
+    <span class="n">conversation_template</span><span class="p">:</span> <span class="n">ConversationTemplate</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Handels conversation datasets tokenization</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">num_example</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">examples</span><span class="p">[</span><span class="n">column_names</span><span class="p">[</span><span class="mi">0</span><span class="p">]])</span>
+    <span class="n">token_dict</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;input_ids&quot;</span><span class="p">:</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)],</span>
+        <span class="s2">&quot;attention_mask&quot;</span><span class="p">:</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)],</span>
+        <span class="s2">&quot;labels&quot;</span><span class="p">:</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)],</span>
+    <span class="p">}</span>
+    <span class="k">with</span> <span class="n">CaptureLogger</span><span class="p">(</span><span class="n">tok_logger</span><span class="p">)</span> <span class="k">as</span> <span class="n">cl</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">examples</span><span class="p">[</span><span class="s2">&quot;messages&quot;</span><span class="p">])):</span>
+            <span class="n">messages</span> <span class="o">=</span> <span class="n">examples</span><span class="p">[</span><span class="s2">&quot;messages&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+            <span class="n">system</span> <span class="o">=</span> <span class="n">examples</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;system&quot;</span><span class="p">,</span> <span class="p">[</span><span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">num_example</span><span class="p">)[</span><span class="n">i</span><span class="p">]</span>
+            <span class="n">tools</span> <span class="o">=</span> <span class="n">examples</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s2">&quot;tools&quot;</span><span class="p">,</span> <span class="p">[</span><span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">num_example</span><span class="p">)[</span><span class="n">i</span><span class="p">]</span>
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">messages</span><span class="p">)</span> <span class="o">&lt;</span> <span class="mi">2</span> <span class="ow">or</span> <span class="n">messages</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">!=</span> <span class="n">CONVERSATION_ROLE_NAMES</span><span class="p">[</span><span class="s1">&#39;user&#39;</span><span class="p">]:</span>
+                <span class="n">tok_logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="s2">&quot;Invalid instance encountered. Either the conversation has less than &quot;</span>
+                    <span class="s2">&quot;one round or the first message is not from the user.&quot;</span>
+                <span class="p">)</span>
+                <span class="k">continue</span>
+        
+            <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">messages</span><span class="p">)</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                    <span class="s2">&quot;The number of messages is not even, the last message will be ignored.&quot;</span>
+                <span class="p">)</span>
+                <span class="n">messages</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+                
+            <span class="n">encoded_conversation</span> <span class="o">=</span> <span class="n">conversation_template</span><span class="o">.</span><span class="n">encode_conversation</span><span class="p">(</span>
+                <span class="n">tokenizer</span><span class="o">=</span><span class="n">tokenizer</span><span class="p">,</span>
+                <span class="n">messages</span><span class="o">=</span><span class="n">messages</span><span class="p">,</span>
+                <span class="n">system</span><span class="o">=</span><span class="n">system</span><span class="p">,</span>
+                <span class="n">tools</span><span class="o">=</span><span class="n">tools</span><span class="p">,</span>
+            <span class="p">)</span>
+
+            <span class="n">input_ids</span><span class="p">,</span> <span class="n">labels</span> <span class="o">=</span> <span class="p">[],</span> <span class="p">[]</span>
+            <span class="k">for</span> <span class="n">turn_idx</span><span class="p">,</span> <span class="p">(</span><span class="n">user_input</span><span class="p">,</span> <span class="n">assistant_result</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">encoded_conversation</span><span class="p">):</span>
+                <span class="n">input_ids</span> <span class="o">+=</span> <span class="n">user_input</span> <span class="o">+</span> <span class="n">assistant_result</span>
+                
+                <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">train_on_prompt</span><span class="p">:</span>
+                    <span class="n">labels</span> <span class="o">+=</span> <span class="n">user_input</span> <span class="o">+</span> <span class="n">assistant_result</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">labels</span> <span class="o">+=</span> <span class="p">[</span><span class="o">-</span><span class="mi">100</span><span class="p">]</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">user_input</span><span class="p">)</span> <span class="o">+</span> <span class="n">assistant_result</span>
+                
+            <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">input_ids</span><span class="p">)</span>
+            <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">([</span><span class="mi">1</span><span class="p">]</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">input_ids</span><span class="p">))</span>
+            <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">labels</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+        <span class="n">token_dict</span> <span class="o">=</span> <span class="n">blocking</span><span class="p">(</span>
+            <span class="n">token_dict</span><span class="o">=</span><span class="n">token_dict</span><span class="p">,</span>
+            <span class="n">block_size</span><span class="o">=</span><span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span><span class="p">,</span>
+            <span class="n">model_max_length</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="p">,</span>
+            <span class="n">pad_token_id</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span><span class="p">,</span>
+            <span class="n">padding_side</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">padding_side</span><span class="p">,</span>
+            <span class="n">truncation_side</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">truncation_side</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="c1"># clm input could be much much longer than block_size</span>
+    <span class="k">if</span> <span class="s2">&quot;Token indices sequence length is longer than the&quot;</span> <span class="ow">in</span> <span class="n">cl</span><span class="o">.</span><span class="n">out</span><span class="p">:</span>
+        <span class="n">tok_logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+            <span class="s2">&quot;^^^^^^^^^^^^^^^^ Please ignore the warning above - this long input will be chunked into smaller bits&quot;</span>
+            <span class="s2">&quot; before being passed to the model.&quot;</span>
+        <span class="p">)</span>
+    <span class="k">return</span> <span class="n">token_dict</span></div>
+
+
+            
+<div class="viewcode-block" id="tokenize_function">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.tokenize_function">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">tokenize_function</span><span class="p">(</span>
+    <span class="n">examples</span><span class="p">,</span> 
+    <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span>
+    <span class="n">tokenizer</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">PreTrainedTokenizer</span><span class="p">,</span> <span class="n">PreTrainedTokenizerFast</span><span class="p">],</span>
+    <span class="n">column_names</span><span class="p">,</span>
+    <span class="n">label_columns</span><span class="p">,</span>
+    <span class="n">tokenized_column_order</span><span class="p">,</span>
+    <span class="n">add_special_tokens</span><span class="p">,</span>
+    <span class="n">use_truncation</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Handels text_only and text2text datasets tokenization</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">num_example</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">examples</span><span class="p">[</span><span class="n">column_names</span><span class="p">[</span><span class="mi">0</span><span class="p">]])</span>
+    <span class="n">token_dict</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s2">&quot;input_ids&quot;</span><span class="p">:</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)],</span>
+        <span class="s2">&quot;attention_mask&quot;</span><span class="p">:</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)],</span>
+        <span class="s2">&quot;labels&quot;</span><span class="p">:</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)],</span>
+    <span class="p">}</span>
+    <span class="k">with</span> <span class="n">CaptureLogger</span><span class="p">(</span><span class="n">tok_logger</span><span class="p">)</span> <span class="k">as</span> <span class="n">cl</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">column_name</span> <span class="ow">in</span> <span class="n">tokenized_column_order</span><span class="p">:</span>
+            <span class="n">encoding</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">(</span>
+                <span class="n">examples</span><span class="p">[</span><span class="n">column_name</span><span class="p">],</span>
+                <span class="n">add_special_tokens</span><span class="o">=</span><span class="n">add_special_tokens</span><span class="p">,</span>
+                <span class="n">truncation</span><span class="o">=</span><span class="n">use_truncation</span><span class="p">,</span>
+            <span class="p">)</span>
+
+            <span class="k">if</span> <span class="n">column_name</span> <span class="ow">in</span> <span class="n">label_columns</span><span class="p">:</span>
+                <span class="n">labels</span> <span class="o">=</span> <span class="n">encoding</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">labels</span> <span class="o">=</span> <span class="p">[</span>
+                    <span class="p">[</span><span class="o">-</span><span class="mi">100</span><span class="p">]</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">encoding</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">])</span>
+                    <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)</span>
+                <span class="p">]</span>
+
+            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">):</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+                    <span class="n">encoding</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+                <span class="p">)</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span>
+                    <span class="n">encoding</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span>
+                <span class="p">)</span>
+                <span class="n">token_dict</span><span class="p">[</span><span class="s2">&quot;labels&quot;</span><span class="p">][</span><span class="n">i</span><span class="p">]</span><span class="o">.</span><span class="n">extend</span><span class="p">(</span><span class="n">labels</span><span class="p">[</span><span class="n">i</span><span class="p">])</span>
+
+    <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+        <span class="n">token_dict</span> <span class="o">=</span> <span class="n">blocking</span><span class="p">(</span>
+            <span class="n">token_dict</span><span class="o">=</span><span class="n">token_dict</span><span class="p">,</span>
+            <span class="n">block_size</span><span class="o">=</span><span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span><span class="p">,</span>
+            <span class="n">model_max_length</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="p">,</span>
+            <span class="n">pad_token_id</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span><span class="p">,</span>
+            <span class="n">padding_side</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">padding_side</span><span class="p">,</span>
+            <span class="n">truncation_side</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">truncation_side</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="c1"># clm input could be much much longer than block_size</span>
+    <span class="k">if</span> <span class="s2">&quot;Token indices sequence length is longer than the&quot;</span> <span class="ow">in</span> <span class="n">cl</span><span class="o">.</span><span class="n">out</span><span class="p">:</span>
+        <span class="n">tok_logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+            <span class="s2">&quot;^^^^^^^^^^^^^^^^ Please ignore the warning above - this long input will be chunked into smaller bits&quot;</span>
+            <span class="s2">&quot; before being passed to the model.&quot;</span>
+        <span class="p">)</span>
+    <span class="k">return</span> <span class="n">token_dict</span></div>
+
+
+
+<div class="viewcode-block" id="text_to_textlist_tokenize_function">
+<a class="viewcode-back" href="../../../autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.text_to_textlist_tokenize_function">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">text_to_textlist_tokenize_function</span><span class="p">(</span>
+    <span class="n">examples</span><span class="p">,</span> 
+    <span class="n">data_args</span><span class="p">:</span> <span class="n">DatasetArguments</span><span class="p">,</span>
+    <span class="n">tokenizer</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">PreTrainedTokenizer</span><span class="p">,</span> <span class="n">PreTrainedTokenizerFast</span><span class="p">],</span>
+    <span class="n">column_names</span><span class="p">,</span>
+    <span class="n">add_special_tokens</span><span class="p">,</span>
+    <span class="n">use_truncation</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;For rm inference, and don&#39;t need attn mask and labels.</span>
+<span class="sd">    NOTE: input_ids here refers to the tokenized input_ids of the input **and** output</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">num_example</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">examples</span><span class="p">[</span><span class="n">column_names</span><span class="p">[</span><span class="mi">0</span><span class="p">]])</span>
+    <span class="n">output_dict</span> <span class="o">=</span> <span class="p">{</span><span class="n">column_name</span><span class="p">:</span> <span class="n">examples</span><span class="p">[</span><span class="n">column_name</span><span class="p">]</span> <span class="k">for</span> <span class="n">column_name</span> <span class="ow">in</span> <span class="n">column_names</span><span class="p">}</span>
+    <span class="n">output_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="p">[[]</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">)]</span>
+    
+    <span class="k">for</span> <span class="n">example_idx</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">num_example</span><span class="p">):</span>
+        <span class="n">encoded</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">(</span>
+            <span class="p">[</span>
+                <span class="n">examples</span><span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">][</span><span class="n">example_idx</span><span class="p">]</span> <span class="o">+</span> <span class="n">examples</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">][</span><span class="n">example_idx</span><span class="p">][</span><span class="n">i</span><span class="p">]</span> 
+                <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">examples</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">][</span><span class="n">example_idx</span><span class="p">]))</span>
+            <span class="p">],</span>
+            <span class="n">add_special_tokens</span><span class="o">=</span><span class="n">add_special_tokens</span><span class="p">,</span>
+            <span class="n">truncation</span><span class="o">=</span><span class="n">use_truncation</span><span class="p">,</span>
+        <span class="p">)</span>
+
+        <span class="n">output_dict</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">][</span><span class="n">example_idx</span><span class="p">]</span> <span class="o">=</span> <span class="n">encoded</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span>
+        
+    <span class="k">if</span> <span class="n">data_args</span><span class="o">.</span><span class="n">disable_group_texts</span><span class="p">:</span>
+        <span class="n">output_dict</span> <span class="o">=</span> <span class="n">blocking_text_to_textlist</span><span class="p">(</span>
+            <span class="n">token_dict</span><span class="o">=</span><span class="n">output_dict</span><span class="p">,</span>
+            <span class="n">block_size</span><span class="o">=</span><span class="n">data_args</span><span class="o">.</span><span class="n">block_size</span><span class="p">,</span>
+            <span class="n">model_max_length</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">model_max_length</span><span class="p">,</span>
+            <span class="n">pad_token_id</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">pad_token_id</span><span class="p">,</span>
+            <span class="n">padding_side</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">padding_side</span><span class="p">,</span>
+            <span class="n">truncation_side</span><span class="o">=</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">truncation_side</span><span class="p">,</span>
+        <span class="p">)</span>
+        
+    <span class="k">return</span> <span class="n">output_dict</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/common.html b/_modules/lmflow/utils/common.html
new file mode 100644
index 000000000..5eabc7ee2
--- /dev/null
+++ b/_modules/lmflow/utils/common.html
@@ -0,0 +1,650 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.common &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/common';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.common</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.common</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">dataclasses</span><span class="w"> </span><span class="kn">import</span> <span class="n">dataclass</span><span class="p">,</span> <span class="n">field</span><span class="p">,</span> <span class="n">fields</span><span class="p">,</span> <span class="n">Field</span><span class="p">,</span> <span class="n">make_dataclass</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">pathlib</span><span class="w"> </span><span class="kn">import</span> <span class="n">Path</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">Dict</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.versioning</span><span class="w"> </span><span class="kn">import</span> <span class="n">get_python_version</span>
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/common/index.html#lmflow.utils.common.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="make_shell_args_from_dataclass">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/common/index.html#lmflow.utils.common.make_shell_args_from_dataclass">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">make_shell_args_from_dataclass</span><span class="p">(</span>
+    <span class="n">dataclass_objects</span><span class="p">:</span> <span class="n">List</span><span class="p">,</span> 
+    <span class="nb">format</span><span class="p">:</span> <span class="nb">str</span><span class="o">=</span><span class="s2">&quot;subprocess&quot;</span><span class="p">,</span>
+    <span class="n">skip_default</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+    <span class="n">ignored_args_list</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Return a string or a list of strings that can be used as shell arguments.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ----------</span>
+<span class="sd">    dataclass_objects : List</span>
+<span class="sd">        A list of dataclass objects.</span>
+<span class="sd">    format : str, optional</span>
+<span class="sd">        Return format, can be &quot;shell&quot; or &quot;subprocess&quot;, by default &quot;subprocess&quot;.</span>
+<span class="sd">    skip_default : bool, optional</span>
+<span class="sd">        Whether to skip attributes with default values, by default True. </span>
+
+<span class="sd">    Returns</span>
+<span class="sd">    -------</span>
+<span class="sd">    Union[str, List[str]]</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">dataclass_objects</span><span class="p">,</span> <span class="nb">list</span><span class="p">),</span> <span class="s2">&quot;dataclass_objects should be a list of dataclass objects.&quot;</span>
+    <span class="n">all_args</span> <span class="o">=</span> <span class="p">{}</span>
+    <span class="k">for</span> <span class="n">dataclass_object</span> <span class="ow">in</span> <span class="n">dataclass_objects</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">dataclass_object</span><span class="o">.</span><span class="vm">__dict__</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+            <span class="k">if</span> <span class="n">ignored_args_list</span> <span class="ow">and</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">ignored_args_list</span><span class="p">:</span>
+                <span class="k">continue</span>
+            <span class="k">if</span> <span class="n">k</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">dataclass_object</span><span class="o">.</span><span class="n">__dataclass_fields__</span><span class="p">:</span>
+                <span class="c1"># skip attributes that added dynamically</span>
+                <span class="k">continue</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="n">v</span><span class="p">:</span>
+                <span class="c1"># skip attributes with None values</span>
+                <span class="k">continue</span>
+            <span class="k">if</span> <span class="n">skip_default</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">dataclass_object</span><span class="o">.</span><span class="n">__dataclass_fields__</span><span class="p">[</span><span class="n">k</span><span class="p">]</span><span class="o">.</span><span class="n">default</span> <span class="o">==</span> <span class="n">v</span><span class="p">:</span>
+                    <span class="k">continue</span>
+            
+            <span class="k">if</span> <span class="n">k</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">all_args</span><span class="p">:</span>
+                <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">v</span><span class="p">,</span> <span class="n">Path</span><span class="p">):</span>
+                    <span class="n">all_args</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">v</span><span class="p">)</span>
+                <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">v</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+                    <span class="n">all_args</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="s2">&quot;,&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">v</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">all_args</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="n">v</span>
+            <span class="k">elif</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">all_args</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">all_args</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">==</span> <span class="n">v</span><span class="p">:</span>
+                    <span class="k">continue</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Found different values for the same key: </span><span class="si">{</span><span class="n">k</span><span class="si">}</span><span class="s2">, using value: </span><span class="si">{</span><span class="n">v</span><span class="si">}</span><span class="s2"> instead.&quot;</span><span class="p">)</span>
+                    <span class="n">all_args</span><span class="p">[</span><span class="n">k</span><span class="p">]</span> <span class="o">=</span> <span class="n">v</span>
+    
+    <span class="k">if</span> <span class="nb">format</span> <span class="o">==</span> <span class="s2">&quot;shell&quot;</span><span class="p">:</span>
+        <span class="n">final_res</span> <span class="o">=</span> <span class="s2">&quot; &quot;</span><span class="o">.</span><span class="n">join</span><span class="p">([</span><span class="sa">f</span><span class="s2">&quot;--</span><span class="si">{</span><span class="n">k</span><span class="si">}</span><span class="s2"> </span><span class="si">{</span><span class="n">v</span><span class="si">}</span><span class="s2">&quot;</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">all_args</span><span class="o">.</span><span class="n">items</span><span class="p">()])</span>
+    <span class="k">elif</span> <span class="nb">format</span> <span class="o">==</span> <span class="s2">&quot;subprocess&quot;</span><span class="p">:</span>
+        <span class="n">final_res</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="ow">in</span> <span class="n">all_args</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+            <span class="n">final_res</span><span class="o">.</span><span class="n">extend</span><span class="p">([</span><span class="sa">f</span><span class="s2">&quot;--</span><span class="si">{</span><span class="n">k</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="nb">str</span><span class="p">(</span><span class="n">v</span><span class="p">)])</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unknown format: </span><span class="si">{</span><span class="nb">format</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        
+    <span class="k">return</span> <span class="n">final_res</span></div>
+
+
+
+<div class="viewcode-block" id="create_copied_dataclass">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/common/index.html#lmflow.utils.common.create_copied_dataclass">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">create_copied_dataclass</span><span class="p">(</span>
+    <span class="n">original_dataclass</span><span class="p">,</span> 
+    <span class="n">field_prefix</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> 
+    <span class="n">class_prefix</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> 
+    <span class="n">new_default</span><span class="p">:</span> <span class="n">Dict</span><span class="o">=</span><span class="kc">None</span>
+<span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Create a copied dataclass with new field names and default values.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ----------</span>
+<span class="sd">    original_dataclass : dataclass</span>
+<span class="sd">    field_prefix : str</span>
+<span class="sd">        The prefix to add to the **field** names of the copied dataclass.</span>
+<span class="sd">    class_prefix : str</span>
+<span class="sd">        The prefix to add to the **class** name of the copied dataclass.</span>
+<span class="sd">    new_default : Dict, optional</span>
+<span class="sd">        The new default values for the copied dataclass. When None, the </span>
+<span class="sd">        default values of the original dataclass are used.</span>
+
+<span class="sd">    Returns</span>
+<span class="sd">    -------</span>
+<span class="sd">    dataclass</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">original_fields</span> <span class="o">=</span> <span class="n">fields</span><span class="p">(</span><span class="n">original_dataclass</span><span class="p">)</span>
+    <span class="n">new_default</span> <span class="o">=</span> <span class="n">new_default</span> <span class="ow">or</span> <span class="p">{}</span>
+    <span class="n">new_fields</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">for</span> <span class="n">field</span> <span class="ow">in</span> <span class="n">original_fields</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">get_python_version</span><span class="p">()</span><span class="o">.</span><span class="n">minor</span> <span class="o">&gt;=</span> <span class="mi">10</span><span class="p">:</span>
+            <span class="n">new_field</span> <span class="o">=</span> <span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">field_prefix</span><span class="si">}{</span><span class="n">field</span><span class="o">.</span><span class="n">name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span> 
+                <span class="n">field</span><span class="o">.</span><span class="n">type</span><span class="p">,</span> 
+                <span class="n">Field</span><span class="p">(</span>
+                    <span class="n">default</span><span class="o">=</span><span class="n">new_default</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">field_prefix</span><span class="si">}{</span><span class="n">field</span><span class="o">.</span><span class="n">name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="n">field</span><span class="o">.</span><span class="n">default</span><span class="p">),</span> 
+                    <span class="n">default_factory</span><span class="o">=</span><span class="n">field</span><span class="o">.</span><span class="n">default_factory</span><span class="p">,</span>
+                    <span class="n">init</span><span class="o">=</span><span class="n">field</span><span class="o">.</span><span class="n">init</span><span class="p">,</span>
+                    <span class="nb">repr</span><span class="o">=</span><span class="n">field</span><span class="o">.</span><span class="n">repr</span><span class="p">,</span>
+                    <span class="nb">hash</span><span class="o">=</span><span class="n">field</span><span class="o">.</span><span class="n">hash</span><span class="p">,</span>
+                    <span class="n">compare</span><span class="o">=</span><span class="n">field</span><span class="o">.</span><span class="n">compare</span><span class="p">,</span>
+                    <span class="n">metadata</span><span class="o">=</span><span class="n">field</span><span class="o">.</span><span class="n">metadata</span><span class="p">,</span>
+                    <span class="n">kw_only</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="c1"># add in py3.10: https://docs.python.org/3/library/dataclasses.html</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">new_field</span> <span class="o">=</span> <span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">field_prefix</span><span class="si">}{</span><span class="n">field</span><span class="o">.</span><span class="n">name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span> 
+                <span class="n">field</span><span class="o">.</span><span class="n">type</span><span class="p">,</span> 
+                <span class="n">Field</span><span class="p">(</span>
+                    <span class="n">default</span><span class="o">=</span><span class="n">new_default</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">field_prefix</span><span class="si">}{</span><span class="n">field</span><span class="o">.</span><span class="n">name</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="n">field</span><span class="o">.</span><span class="n">default</span><span class="p">),</span> 
+                    <span class="n">default_factory</span><span class="o">=</span><span class="n">field</span><span class="o">.</span><span class="n">default_factory</span><span class="p">,</span>
+                    <span class="n">init</span><span class="o">=</span><span class="n">field</span><span class="o">.</span><span class="n">init</span><span class="p">,</span>
+                    <span class="nb">repr</span><span class="o">=</span><span class="n">field</span><span class="o">.</span><span class="n">repr</span><span class="p">,</span>
+                    <span class="nb">hash</span><span class="o">=</span><span class="n">field</span><span class="o">.</span><span class="n">hash</span><span class="p">,</span>
+                    <span class="n">compare</span><span class="o">=</span><span class="n">field</span><span class="o">.</span><span class="n">compare</span><span class="p">,</span>
+                    <span class="n">metadata</span><span class="o">=</span><span class="n">field</span><span class="o">.</span><span class="n">metadata</span><span class="p">,</span>
+                <span class="p">)</span>
+            <span class="p">)</span>
+            
+        <span class="n">new_fields</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">new_field</span><span class="p">)</span>
+    <span class="n">copied_dataclass</span> <span class="o">=</span> <span class="n">make_dataclass</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">class_prefix</span><span class="si">}{</span><span class="n">original_dataclass</span><span class="o">.</span><span class="vm">__name__</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">,</span> <span class="n">new_fields</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">copied_dataclass</span></div>
+
+
+
+<div class="viewcode-block" id="remove_dataclass_attr_prefix">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/common/index.html#lmflow.utils.common.remove_dataclass_attr_prefix">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">remove_dataclass_attr_prefix</span><span class="p">(</span><span class="n">data_instance</span><span class="p">,</span> <span class="n">prefix</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Remove the prefix from the attribute names of a dataclass instance.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ----------</span>
+<span class="sd">    data_instance : dataclass</span>
+<span class="sd">    prefix : str</span>
+<span class="sd">        The prefix to remove from the attribute names of the dataclass instance.</span>
+
+<span class="sd">    Returns</span>
+<span class="sd">    -------</span>
+<span class="sd">    Dict</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">new_attributes</span> <span class="o">=</span> <span class="p">{}</span>
+    <span class="k">for</span> <span class="n">field</span> <span class="ow">in</span> <span class="n">fields</span><span class="p">(</span><span class="n">data_instance</span><span class="p">):</span>
+        <span class="n">attr_name</span> <span class="o">=</span> <span class="n">field</span><span class="o">.</span><span class="n">name</span>
+        <span class="n">attr_value</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">data_instance</span><span class="p">,</span> <span class="n">attr_name</span><span class="p">)</span>
+        <span class="n">new_attr_name</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">attr_name</span><span class="p">[</span><span class="nb">len</span><span class="p">(</span><span class="n">prefix</span><span class="p">):]</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">new_attributes</span><span class="p">[</span><span class="n">new_attr_name</span><span class="p">]</span> <span class="o">=</span> <span class="n">attr_value</span>
+    
+    <span class="k">return</span> <span class="n">new_attributes</span></div>
+
+
+
+<div class="viewcode-block" id="add_dataclass_attr_prefix">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/common/index.html#lmflow.utils.common.add_dataclass_attr_prefix">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">add_dataclass_attr_prefix</span><span class="p">(</span><span class="n">data_instance</span><span class="p">,</span> <span class="n">prefix</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Dict</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Add the prefix to the attribute names of a dataclass instance.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ----------</span>
+<span class="sd">    data_instance : dataclass</span>
+<span class="sd">    prefix : str</span>
+<span class="sd">        The prefix to add to the attribute names of the dataclass instance.</span>
+
+<span class="sd">    Returns</span>
+<span class="sd">    -------</span>
+<span class="sd">    Dict</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">new_attributes</span> <span class="o">=</span> <span class="p">{}</span>
+    <span class="k">for</span> <span class="n">field</span> <span class="ow">in</span> <span class="n">fields</span><span class="p">(</span><span class="n">data_instance</span><span class="p">):</span>
+        <span class="n">attr_name</span> <span class="o">=</span> <span class="n">field</span><span class="o">.</span><span class="n">name</span>
+        <span class="n">attr_value</span> <span class="o">=</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">data_instance</span><span class="p">,</span> <span class="n">attr_name</span><span class="p">)</span>
+        <span class="n">new_attr_name</span> <span class="o">=</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="n">prefix</span><span class="si">}{</span><span class="n">attr_name</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="n">new_attributes</span><span class="p">[</span><span class="n">new_attr_name</span><span class="p">]</span> <span class="o">=</span> <span class="n">attr_value</span>
+    
+    <span class="k">return</span> <span class="n">new_attributes</span></div>
+
+
+
+<div class="viewcode-block" id="print_banner">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/common/index.html#lmflow.utils.common.print_banner">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">print_banner</span><span class="p">(</span><span class="n">message</span><span class="p">:</span> <span class="nb">str</span><span class="p">):</span>
+    <span class="n">length</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">message</span><span class="p">)</span> <span class="o">+</span> <span class="mi">8</span>
+    <span class="n">border</span> <span class="o">=</span> <span class="s2">&quot;#&quot;</span> <span class="o">*</span> <span class="n">length</span>
+
+    <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="n">border</span><span class="p">)</span>
+    <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;#   </span><span class="si">{</span><span class="n">message</span><span class="si">}</span><span class="s2">   #&quot;</span><span class="p">)</span>
+    <span class="n">logger</span><span class="o">.</span><span class="n">info</span><span class="p">(</span><span class="n">border</span><span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/constants.html b/_modules/lmflow/utils/constants.html
new file mode 100644
index 000000000..13bf4215c
--- /dev/null
+++ b/_modules/lmflow/utils/constants.html
@@ -0,0 +1,981 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.constants &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/constants';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.constants</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.constants</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">Commonly used constants.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<div class="viewcode-block" id="TEXT_ONLY_DATASET_DESCRIPTION">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT_ONLY_DATASET_DESCRIPTION">[docs]</a>
+<span class="n">TEXT_ONLY_DATASET_DESCRIPTION</span> <span class="o">=</span> <span class="p">(</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">&quot;text_only&quot;: a dataset with only raw text instances, with following format:</span>
+
+<span class="sd">    {</span>
+<span class="sd">        &quot;type&quot;: &quot;text_only&quot;,</span>
+<span class="sd">        &quot;instances&quot;: [</span>
+<span class="sd">            { &quot;text&quot;: &quot;TEXT_1&quot; },</span>
+<span class="sd">            { &quot;text&quot;: &quot;TEXT_2&quot; },</span>
+<span class="sd">            ...</span>
+<span class="sd">        ]</span>
+<span class="sd">    }</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="p">)</span><span class="o">.</span><span class="n">lstrip</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION">[docs]</a>
+<span class="n">TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION</span> <span class="o">=</span> <span class="p">(</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">This kind of dataset is commonly used in reward model training/prediction, as well as rl training.</span>
+<span class="sd">{</span>
+<span class="sd">    &quot;type&quot;: &quot;text_to_scored_textlist&quot;,</span>
+<span class="sd">    &quot;instances&quot;: [</span>
+<span class="sd">        {</span>
+<span class="sd">            &quot;input&quot;: &quot;what&#39;s your name?&quot;,</span>
+<span class="sd">            &quot;output&quot;: [</span>
+<span class="sd">                {&quot;score&quot;: 1.0, &quot;text&quot;: &quot;My name is John&quot;},</span>
+<span class="sd">                {&quot;score&quot;: -0.8, &quot;text&quot;: &quot;I&#39;m John&quot;}</span>
+<span class="sd">            ]</span>
+<span class="sd">        },</span>
+<span class="sd">        {</span>
+<span class="sd">            &quot;input&quot;: &quot;Who are you?&quot;,</span>
+<span class="sd">            &quot;output&quot;: [</span>
+<span class="sd">                {&quot;score&quot;: 1.5, &quot;text&quot;: &quot;My name is Amy&quot;},</span>
+<span class="sd">                {&quot;score&quot;: 1.0, &quot;text&quot;: &quot;I&#39;m Amy&quot;}</span>
+<span class="sd">            ]</span>
+<span class="sd">        },</span>
+<span class="sd">    ]</span>
+<span class="sd">}</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="p">)</span><span class="o">.</span><span class="n">lstrip</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION">[docs]</a>
+<span class="n">PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION</span> <span class="o">=</span> <span class="p">(</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">This kind of dataset is commonly used in reward model training as well as rl training.</span>
+<span class="sd">{</span>
+<span class="sd">    &quot;type&quot;: &quot;paired_text_to_text&quot;,</span>
+<span class="sd">    &quot;instances&quot;: [</span>
+<span class="sd">        {</span>
+<span class="sd">            &quot;prompt&quot;: &quot;Who are you?&quot;,</span>
+<span class="sd">            &quot;chosen&quot;: &quot;My name is Amy.&quot;,</span>
+<span class="sd">            &quot;rejected&quot;: &quot;I&#39;m Amy&quot;,</span>
+<span class="sd">            &quot;margin&quot;: 0.6</span>
+<span class="sd">        },</span>
+<span class="sd">        {</span>
+<span class="sd">            &quot;prompt&quot;: &quot;what&#39;s your name?&quot;,</span>
+<span class="sd">            &quot;chosen&quot;: &quot;My name is John.&quot;,</span>
+<span class="sd">            &quot;rejected&quot;: &quot;I&#39;m John&quot;,</span>
+<span class="sd">            &quot;margin&quot;: 0.5</span>
+<span class="sd">        }</span>
+<span class="sd">    ]</span>
+<span class="sd">}</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="p">)</span><span class="o">.</span><span class="n">lstrip</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="TEXT_ONLY_DATASET_DETAILS">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT_ONLY_DATASET_DETAILS">[docs]</a>
+<span class="n">TEXT_ONLY_DATASET_DETAILS</span> <span class="o">=</span> <span class="p">(</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    For example,</span>
+
+<span class="sd">    ```python</span>
+<span class="sd">    from lmflow.datasets import Dataset</span>
+
+<span class="sd">    data_dict = {</span>
+<span class="sd">        &quot;type&quot;: &quot;text_only&quot;,</span>
+<span class="sd">        &quot;instances&quot;: [</span>
+<span class="sd">            { &quot;text&quot;: &quot;Human: Hello. Bot: Hi!&quot; },</span>
+<span class="sd">            { &quot;text&quot;: &quot;Human: How are you today? Bot: Fine, thank you!&quot; },</span>
+<span class="sd">        ]</span>
+<span class="sd">    }</span>
+<span class="sd">    dataset = Dataset.create_from_dict(data_dict)</span>
+<span class="sd">    ```</span>
+
+<span class="sd">    You may also save the corresponding format to json,</span>
+<span class="sd">    ```python</span>
+<span class="sd">    import json</span>
+<span class="sd">    from lmflow.args import DatasetArguments</span>
+<span class="sd">    from lmflow.datasets import Dataset</span>
+
+<span class="sd">    data_dict = {</span>
+<span class="sd">        &quot;type&quot;: &quot;text_only&quot;,</span>
+<span class="sd">        &quot;instances&quot;: [</span>
+<span class="sd">            { &quot;text&quot;: &quot;Human: Hello. Bot: Hi!&quot; },</span>
+<span class="sd">            { &quot;text&quot;: &quot;Human: How are you today? Bot: Fine, thank you!&quot; },</span>
+<span class="sd">        ]</span>
+<span class="sd">    }</span>
+<span class="sd">    with open(&quot;data.json&quot;, &quot;w&quot;) as fout:</span>
+<span class="sd">        json.dump(data_dict, fout)</span>
+
+<span class="sd">    data_args = DatasetArgument(dataset_path=&quot;data.json&quot;)</span>
+<span class="sd">    dataset = Dataset(data_args)</span>
+<span class="sd">    new_data_dict = dataset.to_dict()</span>
+<span class="sd">    # `new_data_dict` Should have the same content as `data_dict`</span>
+<span class="sd">    ```</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="p">)</span><span class="o">.</span><span class="n">lstrip</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="TEXT2TEXT_DATASET_DESCRIPTION">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT2TEXT_DATASET_DESCRIPTION">[docs]</a>
+<span class="n">TEXT2TEXT_DATASET_DESCRIPTION</span> <span class="o">=</span> <span class="p">(</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">&quot;text2text&quot;: a dataset with input &amp; output instances, with following format:</span>
+
+<span class="sd">    {</span>
+<span class="sd">        &quot;type&quot;: &quot;text2text&quot;,</span>
+<span class="sd">        &quot;instances&quot;: [</span>
+<span class="sd">            { &quot;input&quot;: &quot;INPUT_1&quot;, &quot;output&quot;: &quot;OUTPUT_1&quot; },</span>
+<span class="sd">            { &quot;input&quot;: &quot;INPUT_2&quot;, &quot;output&quot;: &quot;OUTPUT_2&quot; },</span>
+<span class="sd">            ...</span>
+<span class="sd">        ]</span>
+<span class="sd">    }</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="p">)</span><span class="o">.</span><span class="n">lstrip</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="CONVERSATION_DATASET_DESCRIPTION">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.CONVERSATION_DATASET_DESCRIPTION">[docs]</a>
+<span class="n">CONVERSATION_DATASET_DESCRIPTION</span> <span class="o">=</span> <span class="p">(</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">&quot;conversation&quot;: a dataset with conversation instances, with following format (`conversation_id`, `system` and `tools` are optional):</span>
+
+<span class="sd">    {</span>
+<span class="sd">    &quot;type&quot;: &quot;conversation&quot;,</span>
+<span class="sd">    &quot;instances&quot;: [</span>
+<span class="sd">        {</span>
+<span class="sd">        &quot;conversation_id&quot;: &quot;CONVERSATION_ID&quot;,</span>
+<span class="sd">        &quot;system&quot;: &quot;SYSTEM_PROPMT&quot;,</span>
+<span class="sd">        &quot;tools&quot;: [&quot;TOOL_DESCRIPTION_1&quot;,&quot;TOOL_DESCRIPTION_2&quot;,&quot;TOOL_DESCRIPTION_X&quot;],</span>
+<span class="sd">        &quot;messages&quot;: [</span>
+<span class="sd">            {</span>
+<span class="sd">                &quot;role&quot;: &quot;user&quot;,</span>
+<span class="sd">                &quot;content&quot;: &quot;USER_INPUT_1&quot;</span>
+<span class="sd">            },</span>
+<span class="sd">            {</span>
+<span class="sd">                &quot;role&quot;: &quot;assistant&quot;,</span>
+<span class="sd">                &quot;content&quot;: &quot;ASSISTANT_RESPONSE_1&quot;</span>
+<span class="sd">            },</span>
+<span class="sd">            {</span>
+<span class="sd">                &quot;role&quot;: &quot;user&quot;,</span>
+<span class="sd">                &quot;content&quot;: &quot;USER_INPUT_2&quot;</span>
+<span class="sd">            },</span>
+<span class="sd">            {</span>
+<span class="sd">                &quot;role&quot;: &quot;assistant&quot;,</span>
+<span class="sd">                &quot;content&quot;: &quot;ASSISTANT_RESPONSE_2&quot;</span>
+<span class="sd">            }</span>
+<span class="sd">        ]</span>
+<span class="sd">        },</span>
+<span class="sd">        {</span>
+<span class="sd">        &quot;conversation_id&quot;: &quot;CONVERSATION_ID&quot;,</span>
+<span class="sd">        &quot;system&quot;: &quot;SYSTEM_PROPMT&quot;,</span>
+<span class="sd">        &quot;tools&quot;: [&quot;TOOL_DESCRIPTION_1&quot;],</span>
+<span class="sd">        &quot;messages&quot;: [</span>
+<span class="sd">            {</span>
+<span class="sd">                &quot;role&quot;: &quot;user&quot;,</span>
+<span class="sd">                &quot;content&quot;: &quot;USER_INPUT_1&quot;</span>
+<span class="sd">            },</span>
+<span class="sd">            {</span>
+<span class="sd">                &quot;role&quot;: &quot;assistant&quot;,</span>
+<span class="sd">                &quot;content&quot;: &quot;ASSISTANT_RESPONSE_1&quot;</span>
+<span class="sd">            }</span>
+<span class="sd">        ]</span>
+<span class="sd">        }</span>
+<span class="sd">    ]</span>
+<span class="sd">    }</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="p">)</span><span class="o">.</span><span class="n">lstrip</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="PAIRED_CONVERSATION_DATASET_DESCRIPTION">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.PAIRED_CONVERSATION_DATASET_DESCRIPTION">[docs]</a>
+<span class="n">PAIRED_CONVERSATION_DATASET_DESCRIPTION</span> <span class="o">=</span> <span class="p">(</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">&quot;paired_conversation&quot;: a dataset with paired conversation instances, with following format:</span>
+
+<span class="sd">    {</span>
+<span class="sd">        &quot;type&quot;: &quot;paired_conversation&quot;,</span>
+<span class="sd">        &quot;instances&quot;: [</span>
+<span class="sd">            {</span>
+<span class="sd">                &quot;chosen&quot;: {</span>
+<span class="sd">                    &quot;conversation_id&quot;: &quot;CONVERSATION_ID&quot;,</span>
+<span class="sd">                    &quot;system&quot;: &quot;SYSTEM_PROPMT&quot;,</span>
+<span class="sd">                    &quot;tools&quot;: [&quot;TOOL_DESCRIPTION_1&quot;,&quot;TOOL_DESCRIPTION_2&quot;,&quot;TOOL_DESCRIPTION_3&quot;],</span>
+<span class="sd">                    &quot;messages&quot;: [</span>
+<span class="sd">                        {</span>
+<span class="sd">                            &quot;role&quot;: &quot;user&quot;,</span>
+<span class="sd">                            &quot;content&quot;: &quot;USER_INPUT_1&quot;</span>
+<span class="sd">                        },</span>
+<span class="sd">                        {</span>
+<span class="sd">                            &quot;role&quot;: &quot;assistant&quot;,</span>
+<span class="sd">                            &quot;content&quot;: &quot;ASSISTANT_RESPONSE_1_GOOD&quot;</span>
+<span class="sd">                        },</span>
+<span class="sd">                        {</span>
+<span class="sd">                            &quot;role&quot;: &quot;user&quot;,</span>
+<span class="sd">                            &quot;content&quot;: &quot;USER_INPUT_2&quot;</span>
+<span class="sd">                        },</span>
+<span class="sd">                        {</span>
+<span class="sd">                            &quot;role&quot;: &quot;assistant&quot;,</span>
+<span class="sd">                            &quot;content&quot;: &quot;ASSISTANT_RESPONSE_2_GOOD&quot;</span>
+<span class="sd">                        }</span>
+<span class="sd">                    ]</span>
+<span class="sd">                },</span>
+<span class="sd">                &quot;rejected&quot;: {</span>
+<span class="sd">                    &quot;conversation_id&quot;: &quot;CONVERSATION_ID&quot;,</span>
+<span class="sd">                    &quot;system&quot;: &quot;SYSTEM_PROPMT&quot;,</span>
+<span class="sd">                    &quot;tools&quot;: [&quot;TOOL_DESCRIPTION_1&quot;,&quot;TOOL_DESCRIPTION_2&quot;,&quot;TOOL_DESCRIPTION_3&quot;],</span>
+<span class="sd">                    &quot;messages&quot;: [</span>
+<span class="sd">                        {</span>
+<span class="sd">                            &quot;role&quot;: &quot;user&quot;,</span>
+<span class="sd">                            &quot;content&quot;: &quot;USER_INPUT_1&quot;</span>
+<span class="sd">                        },</span>
+<span class="sd">                        {</span>
+<span class="sd">                            &quot;role&quot;: &quot;assistant&quot;,</span>
+<span class="sd">                            &quot;content&quot;: &quot;ASSISTANT_RESPONSE_1_BAD&quot;</span>
+<span class="sd">                        },</span>
+<span class="sd">                        {</span>
+<span class="sd">                            &quot;role&quot;: &quot;user&quot;,</span>
+<span class="sd">                            &quot;content&quot;: &quot;USER_INPUT_2&quot;</span>
+<span class="sd">                        },</span>
+<span class="sd">                        {</span>
+<span class="sd">                            &quot;role&quot;: &quot;assistant&quot;,</span>
+<span class="sd">                            &quot;content&quot;: &quot;ASSISTANT_RESPONSE_2_BAD&quot;</span>
+<span class="sd">                        }</span>
+<span class="sd">                    ]</span>
+<span class="sd">                }</span>
+<span class="sd">            }</span>
+<span class="sd">        ]</span>
+<span class="sd">    }</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="p">)</span><span class="o">.</span><span class="n">lstrip</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="TEXT_TO_TEXTLIST_DATASET_DESCRIPTION">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT_TO_TEXTLIST_DATASET_DESCRIPTION">[docs]</a>
+<span class="n">TEXT_TO_TEXTLIST_DATASET_DESCRIPTION</span> <span class="o">=</span> <span class="p">(</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">This kind of dataset is commonly used in reward model inference.</span>
+<span class="sd">{</span>
+<span class="sd">    &quot;type&quot;: &quot;text_to_textlist&quot;,</span>
+<span class="sd">    &quot;instances&quot;: [</span>
+<span class="sd">        {</span>
+<span class="sd">            &quot;input&quot;: &quot;what&#39;s your name?&quot;,</span>
+<span class="sd">            &quot;output&quot;: [</span>
+<span class="sd">                &quot;My name is John&quot;,</span>
+<span class="sd">                &quot;I&#39;m John&quot;,</span>
+<span class="sd">            ]</span>
+<span class="sd">        },</span>
+<span class="sd">        {</span>
+<span class="sd">            &quot;input&quot;: &quot;Who are you?&quot;,</span>
+<span class="sd">            &quot;output&quot;: [</span>
+<span class="sd">                &quot;My name is Amy&quot;,</span>
+<span class="sd">                &quot;I&#39;m Amy&quot;,</span>
+<span class="sd">            ]</span>
+<span class="sd">        },</span>
+<span class="sd">    ]</span>
+<span class="sd">}</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="p">)</span><span class="o">.</span><span class="n">lstrip</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="TEXT2TEXT_DATASET_DETAILS">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT2TEXT_DATASET_DETAILS">[docs]</a>
+<span class="n">TEXT2TEXT_DATASET_DETAILS</span> <span class="o">=</span> <span class="p">(</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    For example,</span>
+
+<span class="sd">    ```python</span>
+<span class="sd">    from lmflow.datasets import Dataset</span>
+
+<span class="sd">    data_dict = {</span>
+<span class="sd">        &quot;type&quot;: &quot;text2text&quot;,</span>
+<span class="sd">        &quot;instances&quot;: [</span>
+<span class="sd">            {</span>
+<span class="sd">                &quot;input&quot;: &quot;Human: Hello.&quot;,</span>
+<span class="sd">                &quot;output&quot;: &quot;Bot: Hi!&quot;,</span>
+<span class="sd">            },</span>
+<span class="sd">            {</span>
+<span class="sd">                &quot;input&quot;: &quot;Human: How are you today?&quot;,</span>
+<span class="sd">                &quot;output&quot;: &quot;Bot: Fine, thank you! And you?&quot;,</span>
+<span class="sd">            }</span>
+<span class="sd">        ]</span>
+<span class="sd">    }</span>
+<span class="sd">    dataset = Dataset.create_from_dict(data_dict)</span>
+<span class="sd">    ```</span>
+
+<span class="sd">    You may also save the corresponding format to json,</span>
+<span class="sd">    ```python</span>
+<span class="sd">    import json</span>
+<span class="sd">    from lmflow.args import DatasetArguments</span>
+<span class="sd">    from lmflow.datasets import Dataset</span>
+
+<span class="sd">    data_dict = {</span>
+<span class="sd">        &quot;type&quot;: &quot;text2text&quot;,</span>
+<span class="sd">        &quot;instances&quot;: [</span>
+<span class="sd">            {</span>
+<span class="sd">                &quot;input&quot;: &quot;Human: Hello.&quot;,</span>
+<span class="sd">                &quot;output&quot;: &quot;Bot: Hi!&quot;,</span>
+<span class="sd">            },</span>
+<span class="sd">            {</span>
+<span class="sd">                &quot;input&quot;: &quot;Human: How are you today?&quot;,</span>
+<span class="sd">                &quot;output&quot;: &quot;Bot: Fine, thank you! And you?&quot;,</span>
+<span class="sd">            }</span>
+<span class="sd">        ]</span>
+<span class="sd">    }</span>
+<span class="sd">    with open(&quot;data.json&quot;, &quot;w&quot;) as fout:</span>
+<span class="sd">        json.dump(data_dict, fout)</span>
+
+<span class="sd">    data_args = DatasetArgument(dataset_path=&quot;data.json&quot;)</span>
+<span class="sd">    dataset = Dataset(data_args)</span>
+<span class="sd">    new_data_dict = dataset.to_dict()</span>
+<span class="sd">    # `new_data_dict` Should have the same content as `data_dict`</span>
+<span class="sd">    ```</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="p">)</span><span class="o">.</span><span class="n">lstrip</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="FLOAT_ONLY_DATASET_DESCRIPTION">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.FLOAT_ONLY_DATASET_DESCRIPTION">[docs]</a>
+<span class="n">FLOAT_ONLY_DATASET_DESCRIPTION</span> <span class="o">=</span> <span class="p">(</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">&quot;float_only&quot;: a dataset with only float instances, with following format:</span>
+
+<span class="sd">    {</span>
+<span class="sd">        &quot;type&quot;: &quot;float_only&quot;,</span>
+<span class="sd">        &quot;instances&quot;: [</span>
+<span class="sd">            { &quot;value&quot;: &quot;FLOAT_1&quot; },</span>
+<span class="sd">            { &quot;value&quot;: &quot;FLOAT_2&quot; },</span>
+<span class="sd">            ...</span>
+<span class="sd">        ]</span>
+<span class="sd">    }</span>
+<span class="sd">&quot;&quot;&quot;</span>
+<span class="p">)</span><span class="o">.</span><span class="n">lstrip</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="TEXT_ONLY_DATASET_LONG_DESCRITION">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT_ONLY_DATASET_LONG_DESCRITION">[docs]</a>
+<span class="n">TEXT_ONLY_DATASET_LONG_DESCRITION</span> <span class="o">=</span> <span class="p">(</span>
+    <span class="n">TEXT_ONLY_DATASET_DESCRIPTION</span> <span class="o">+</span> <span class="n">TEXT_ONLY_DATASET_DETAILS</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="TEXT2TEXT_DATASET_LONG_DESCRITION">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT2TEXT_DATASET_LONG_DESCRITION">[docs]</a>
+<span class="n">TEXT2TEXT_DATASET_LONG_DESCRITION</span> <span class="o">=</span> <span class="p">(</span>
+    <span class="n">TEXT2TEXT_DATASET_DESCRIPTION</span> <span class="o">+</span> <span class="n">TEXT2TEXT_DATASET_DETAILS</span>
+<span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="DATASET_DESCRIPTION_MAP">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.DATASET_DESCRIPTION_MAP">[docs]</a>
+<span class="n">DATASET_DESCRIPTION_MAP</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;text_only&quot;</span><span class="p">:</span> <span class="n">TEXT_ONLY_DATASET_DESCRIPTION</span><span class="p">,</span>
+    <span class="s2">&quot;text2text&quot;</span><span class="p">:</span> <span class="n">TEXT2TEXT_DATASET_DESCRIPTION</span><span class="p">,</span>
+    <span class="s2">&quot;float_only&quot;</span><span class="p">:</span> <span class="n">FLOAT_ONLY_DATASET_DESCRIPTION</span><span class="p">,</span>
+<span class="p">}</span></div>
+
+
+<div class="viewcode-block" id="INSTANCE_FIELDS_MAP">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.INSTANCE_FIELDS_MAP">[docs]</a>
+<span class="n">INSTANCE_FIELDS_MAP</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;text_only&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;text&quot;</span><span class="p">],</span>
+    <span class="s2">&quot;text2text&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">,</span> <span class="s2">&quot;output&quot;</span><span class="p">],</span>
+    <span class="s2">&quot;conversation&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;messages&quot;</span><span class="p">],</span> <span class="c1"># system, tools and conversation_id are optional</span>
+    <span class="s2">&quot;paired_conversation&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;chosen&quot;</span><span class="p">,</span> <span class="s2">&quot;rejected&quot;</span><span class="p">],</span>
+    <span class="s2">&quot;paired_text_to_text&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;prompt&quot;</span><span class="p">,</span> <span class="s2">&quot;chosen&quot;</span><span class="p">,</span> <span class="s2">&quot;rejected&quot;</span><span class="p">],</span>
+    <span class="s2">&quot;float_only&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;value&quot;</span><span class="p">],</span>
+    <span class="s2">&quot;image_text&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;images&quot;</span><span class="p">,</span> <span class="s2">&quot;text&quot;</span><span class="p">],</span>
+    <span class="s2">&quot;text_to_textlist&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">,</span> <span class="s2">&quot;output&quot;</span><span class="p">],</span>
+    <span class="s2">&quot;text_to_scored_textlist&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">,</span> <span class="s2">&quot;output&quot;</span><span class="p">],</span>
+<span class="p">}</span></div>
+
+
+<div class="viewcode-block" id="CONVERSATION_ROLE_NAMES">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.CONVERSATION_ROLE_NAMES">[docs]</a>
+<span class="n">CONVERSATION_ROLE_NAMES</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;system&quot;</span><span class="p">:</span> <span class="s2">&quot;system&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;user&quot;</span><span class="p">:</span> <span class="s2">&quot;user&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;assistant&quot;</span><span class="p">:</span> <span class="s2">&quot;assistant&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;function&quot;</span><span class="p">:</span> <span class="s2">&quot;function&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;observation&quot;</span><span class="p">:</span> <span class="s2">&quot;observation&quot;</span>
+<span class="p">}</span></div>
+
+
+<span class="c1"># LLAVA constants</span>
+<div class="viewcode-block" id="CONTROLLER_HEART_BEAT_EXPIRATION">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.CONTROLLER_HEART_BEAT_EXPIRATION">[docs]</a>
+<span class="n">CONTROLLER_HEART_BEAT_EXPIRATION</span> <span class="o">=</span> <span class="mi">30</span></div>
+
+<div class="viewcode-block" id="WORKER_HEART_BEAT_INTERVAL">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.WORKER_HEART_BEAT_INTERVAL">[docs]</a>
+<span class="n">WORKER_HEART_BEAT_INTERVAL</span> <span class="o">=</span> <span class="mi">15</span></div>
+
+
+<div class="viewcode-block" id="LOGDIR">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.LOGDIR">[docs]</a>
+<span class="n">LOGDIR</span> <span class="o">=</span> <span class="s2">&quot;.&quot;</span></div>
+
+
+<span class="c1"># Model Constants</span>
+<div class="viewcode-block" id="IGNORE_INDEX">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.IGNORE_INDEX">[docs]</a>
+<span class="n">IGNORE_INDEX</span> <span class="o">=</span> <span class="o">-</span><span class="mi">100</span></div>
+
+<div class="viewcode-block" id="IMAGE_TOKEN_INDEX">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.IMAGE_TOKEN_INDEX">[docs]</a>
+<span class="n">IMAGE_TOKEN_INDEX</span> <span class="o">=</span> <span class="o">-</span><span class="mi">200</span></div>
+
+<div class="viewcode-block" id="DEFAULT_IMAGE_TOKEN">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.DEFAULT_IMAGE_TOKEN">[docs]</a>
+<span class="n">DEFAULT_IMAGE_TOKEN</span> <span class="o">=</span> <span class="s2">&quot;&lt;image&gt;&quot;</span></div>
+
+<div class="viewcode-block" id="DEFAULT_IMAGE_PATCH_TOKEN">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.DEFAULT_IMAGE_PATCH_TOKEN">[docs]</a>
+<span class="n">DEFAULT_IMAGE_PATCH_TOKEN</span> <span class="o">=</span> <span class="s2">&quot;&lt;im_patch&gt;&quot;</span></div>
+
+<div class="viewcode-block" id="DEFAULT_IM_START_TOKEN">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.DEFAULT_IM_START_TOKEN">[docs]</a>
+<span class="n">DEFAULT_IM_START_TOKEN</span> <span class="o">=</span> <span class="s2">&quot;&lt;im_start&gt;&quot;</span></div>
+
+<div class="viewcode-block" id="DEFAULT_IM_END_TOKEN">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.DEFAULT_IM_END_TOKEN">[docs]</a>
+<span class="n">DEFAULT_IM_END_TOKEN</span> <span class="o">=</span> <span class="s2">&quot;&lt;im_end&gt;&quot;</span></div>
+
+
+<span class="c1"># Lora</span>
+<span class="c1"># NOTE: This work as a mapping for those models that `peft` library doesn&#39;t support yet, and will be </span>
+<span class="c1"># overwritten by peft.utils.constants.TRANSFORMERS_MODELS_TO_LORA_TARGET_MODULES_MAPPING</span>
+<span class="c1"># if the model is supported (see hf_model_mixin.py).</span>
+<span class="c1"># NOTE: When passing lora_target_modules through arg parser, the </span>
+<span class="c1"># value should be a string. Using commas to separate the module names, e.g.</span>
+<span class="c1"># &quot;--lora_target_modules &#39;q_proj, v_proj&#39;&quot;. </span>
+<span class="c1"># However, when specifying here, they should be lists.</span>
+<div class="viewcode-block" id="LMFLOW_LORA_TARGET_MODULES_MAPPING">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.LMFLOW_LORA_TARGET_MODULES_MAPPING">[docs]</a>
+<span class="n">LMFLOW_LORA_TARGET_MODULES_MAPPING</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s1">&#39;qwen2&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;q_proj&quot;</span><span class="p">,</span> <span class="s2">&quot;v_proj&quot;</span><span class="p">],</span>
+    <span class="s1">&#39;internlm2&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;wqkv&quot;</span><span class="p">],</span>
+    <span class="s1">&#39;hymba&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;x_proj.0&quot;</span><span class="p">,</span> <span class="s2">&quot;in_proj&quot;</span><span class="p">,</span> <span class="s2">&quot;out_proj&quot;</span><span class="p">,</span> <span class="s2">&quot;dt_proj.0&quot;</span><span class="p">]</span>
+<span class="p">}</span></div>
+
+
+<span class="c1"># vllm inference</span>
+<div class="viewcode-block" id="MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG">[docs]</a>
+<span class="n">MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG</span> <span class="o">=</span> <span class="s2">&quot;MEMORY_SAFE_VLLM_INFERENCE_DONE&quot;</span></div>
+
+<div class="viewcode-block" id="RETURN_CODE_ERROR_BUFFER">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.RETURN_CODE_ERROR_BUFFER">[docs]</a>
+<span class="n">RETURN_CODE_ERROR_BUFFER</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="mi">134</span>
+<span class="p">]</span></div>
+
+<span class="c1"># return code 134:</span>
+<span class="c1"># &gt; Fatal Python error: _enter_buffered_busy: could not acquire lock for &lt;_io.BufferedWriter name=&#39;&lt;stdout&gt;&#39;&gt; </span>
+<span class="c1"># &gt; at interpreter shutdown, possibly due to daemon threads</span>
+<span class="c1"># The above error, by our observation, is due to the kill signal with unfinished </span>
+<span class="c1"># stdout/stderr writing in the subprocess</span>
+<div class="viewcode-block" id="MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE">[docs]</a>
+<span class="n">MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;OMP_NUM_THREADS&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;LOCAL_RANK&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RANK&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;GROUP_RANK&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ROLE_RANK&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ROLE_NAME&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;LOCAL_WORLD_SIZE&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;WORLD_SIZE&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;GROUP_WORLD_SIZE&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ROLE_WORLD_SIZE&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MASTER_ADDR&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MASTER_PORT&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;TORCHELASTIC_RESTART_COUNT&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;TORCHELASTIC_MAX_RESTARTS&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;TORCHELASTIC_RUN_ID&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;TORCHELASTIC_USE_AGENT_STORE&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;TORCH_NCCL_ASYNC_ERROR_HANDLING&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;TORCHELASTIC_ERROR_FILE&quot;</span><span class="p">,</span>
+<span class="p">]</span></div>
+
+
+<span class="c1"># dpov2 align</span>
+<div class="viewcode-block" id="MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE">[docs]</a>
+<span class="n">MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE</span> <span class="o">=</span> <span class="p">[</span>
+    <span class="s2">&quot;OMP_NUM_THREADS&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;LOCAL_RANK&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;RANK&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;GROUP_RANK&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ROLE_RANK&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ROLE_NAME&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;LOCAL_WORLD_SIZE&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;WORLD_SIZE&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;GROUP_WORLD_SIZE&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;ROLE_WORLD_SIZE&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MASTER_ADDR&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;MASTER_PORT&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;TORCHELASTIC_RESTART_COUNT&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;TORCHELASTIC_MAX_RESTARTS&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;TORCHELASTIC_RUN_ID&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;TORCHELASTIC_USE_AGENT_STORE&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;TORCH_NCCL_ASYNC_ERROR_HANDLING&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;TORCHELASTIC_ERROR_FILE&quot;</span><span class="p">,</span>
+<span class="p">]</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/conversation_template.html b/_modules/lmflow/utils/conversation_template.html
new file mode 100644
index 000000000..b230e32ff
--- /dev/null
+++ b/_modules/lmflow/utils/conversation_template.html
@@ -0,0 +1,481 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.conversation_template &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/conversation_template';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.conversation_template</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.base</span><span class="w"> </span><span class="kn">import</span> <span class="n">EMPTY_TEMPLATE</span><span class="p">,</span> <span class="n">EMPTY_NO_SPECIAL_TOKENS_TEMPLATE</span><span class="p">,</span> <span class="n">ConversationTemplate</span><span class="p">,</span> <span class="n">ConversationTemplateForTool</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.chatglm</span><span class="w"> </span><span class="kn">import</span> <span class="n">CHATGLM3_TEMPLATE</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.chatml</span><span class="w"> </span><span class="kn">import</span> <span class="n">CHATML_TEMPLATE</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.deepseek</span><span class="w"> </span><span class="kn">import</span> <span class="n">DEEPSEEK_TEMPLATE</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.gemma</span><span class="w"> </span><span class="kn">import</span> <span class="n">GEMMA_TEMPLATE</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.hymba</span><span class="w"> </span><span class="kn">import</span> <span class="n">HYMBA_TEMPLATE</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.internlm</span><span class="w"> </span><span class="kn">import</span> <span class="n">INTERNLM2_TEMPLATE</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.llama</span><span class="w"> </span><span class="kn">import</span> <span class="n">LLAMA2_TEMPLATE</span><span class="p">,</span> <span class="n">LLAMA3_TEMPLATE</span><span class="p">,</span> <span class="n">LLAMA3_TEMPLATE_FOR_TOOL</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.phi</span><span class="w"> </span><span class="kn">import</span> <span class="n">PHI3_TEMPLATE</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.qwen</span><span class="w"> </span><span class="kn">import</span> <span class="n">QWEN2_TEMPLATE</span><span class="p">,</span> <span class="n">QWEN2_TEMPLATE_FOR_TOOL</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.yi</span><span class="w"> </span><span class="kn">import</span> <span class="n">YI1_5_TEMPLATE</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.zephyr</span><span class="w"> </span><span class="kn">import</span> <span class="n">ZEPHYR_TEMPLATE</span>
+
+
+<div class="viewcode-block" id="PRESET_TEMPLATES">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.PRESET_TEMPLATES">[docs]</a>
+<span class="n">PRESET_TEMPLATES</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s1">&#39;chatglm3&#39;</span><span class="p">:</span> <span class="n">CHATGLM3_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;chatml&#39;</span><span class="p">:</span> <span class="n">CHATML_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;deepseek&#39;</span><span class="p">:</span> <span class="n">DEEPSEEK_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;disable&#39;</span><span class="p">:</span> <span class="n">EMPTY_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;empty&#39;</span><span class="p">:</span> <span class="n">EMPTY_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;empty_no_special_tokens&#39;</span><span class="p">:</span> <span class="n">EMPTY_NO_SPECIAL_TOKENS_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;gemma&#39;</span><span class="p">:</span> <span class="n">GEMMA_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;hymba&#39;</span><span class="p">:</span> <span class="n">HYMBA_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;internlm2&#39;</span><span class="p">:</span> <span class="n">INTERNLM2_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;llama2&#39;</span><span class="p">:</span> <span class="n">LLAMA2_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;llama3&#39;</span><span class="p">:</span> <span class="n">LLAMA3_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;llama3_for_tool&#39;</span><span class="p">:</span> <span class="n">LLAMA3_TEMPLATE_FOR_TOOL</span><span class="p">,</span>
+    <span class="s1">&#39;phi3&#39;</span><span class="p">:</span> <span class="n">PHI3_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;qwen2&#39;</span><span class="p">:</span> <span class="n">QWEN2_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;qwen2_for_tool&#39;</span><span class="p">:</span> <span class="n">QWEN2_TEMPLATE_FOR_TOOL</span><span class="p">,</span>
+    <span class="s1">&#39;yi&#39;</span><span class="p">:</span> <span class="n">CHATML_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;yi1_5&#39;</span><span class="p">:</span> <span class="n">YI1_5_TEMPLATE</span><span class="p">,</span>
+    <span class="s1">&#39;zephyr&#39;</span><span class="p">:</span> <span class="n">ZEPHYR_TEMPLATE</span>
+<span class="p">}</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/conversation_template/base.html b/_modules/lmflow/utils/conversation_template/base.html
new file mode 100644
index 000000000..09bbc2df2
--- /dev/null
+++ b/_modules/lmflow/utils/conversation_template/base.html
@@ -0,0 +1,1173 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.conversation_template.base &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/conversation_template/base';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../conversation_template.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.base</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.conversation_template.base</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">re</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">abc</span><span class="w"> </span><span class="kn">import</span> <span class="n">ABC</span><span class="p">,</span> <span class="n">abstractmethod</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">dataclasses</span><span class="w"> </span><span class="kn">import</span> <span class="n">dataclass</span><span class="p">,</span> <span class="n">field</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Set</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Literal</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">PreTrainedTokenizer</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="n">CONVERSATION_ROLE_NAMES</span>
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="TemplateComponent">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">TemplateComponent</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;The minimal unit of a template, which can be a token, a string, or a list of tools.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ----------</span>
+<span class="sd">    type : Literal[&#39;token&#39;, &#39;token_id&#39;, &#39;string&#39;, &#39;tools&#39;]</span>
+<span class="sd">        - Type of the component.  </span>
+<span class="sd">        </span>
+<span class="sd">        - When the component is a token or a string, the content should be `string`. </span>
+<span class="sd">        The difference between the two is that token will be converted to token ids </span>
+<span class="sd">        by the tokenizer.convert_tokens_to_ids() method, while string will be directly </span>
+<span class="sd">        encoded by the tokenizer.encode() method. Specially, since the bos token and eos</span>
+<span class="sd">        token are frequently used across different templates, we provide the convenience</span>
+<span class="sd">        to use `&#39;bos_token&#39;` and `&#39;eos_token&#39;` to represent the actual bos and eos tokens when</span>
+<span class="sd">        `type` of the `TemplateComponent` is `token`. For example:  </span>
+<span class="sd">        </span>
+<span class="sd">        ```python</span>
+<span class="sd">        TemplateComponent(type=&#39;token&#39;, content=&#39;bos_token&#39;)</span>
+<span class="sd">        ```</span>
+<span class="sd">        </span>
+<span class="sd">        After encoding, the content will be replaced by the actual token id of the bos token.</span>
+<span class="sd">        Please do remember that if you set the `type` to `string`, the tokenizer will try to </span>
+<span class="sd">        encode the string &#39;bos_token&#39; instead of providing the actual bos token.</span>
+<span class="sd">        </span>
+<span class="sd">        - When the component is token_id, the content should be `int` or `List[int]`, and </span>
+<span class="sd">        will be directly appended to the encoded token ids.</span>
+<span class="sd">        </span>
+<span class="sd">        - Tools are not supported yet.</span>
+<span class="sd">        </span>
+<span class="sd">    content : Union[str, int, List[str], List[int]]</span>
+<span class="sd">        Content of the component.</span>
+
+<span class="sd">    &quot;&quot;&quot;</span>
+<div class="viewcode-block" id="TemplateComponent.type">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent.type">[docs]</a>
+    <span class="nb">type</span><span class="p">:</span> <span class="n">Literal</span><span class="p">[</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="s1">&#39;token_id&#39;</span><span class="p">,</span> <span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="s1">&#39;tools&#39;</span><span class="p">]</span></div>
+
+<div class="viewcode-block" id="TemplateComponent.content">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent.content">[docs]</a>
+    <span class="n">content</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">int</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]</span></div>
+
+<div class="viewcode-block" id="TemplateComponent.mask">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent.mask">[docs]</a>
+    <span class="n">mask</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">True</span> <span class="c1"># for token specific masking, work in progress</span></div>
+
+    
+<div class="viewcode-block" id="TemplateComponent.__post_init__">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent.__post_init__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__post_init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">assert</span> <span class="bp">self</span><span class="o">.</span><span class="n">content</span><span class="p">,</span> <span class="s2">&quot;Content of the component cannot be empty.&quot;</span>
+        
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;tools&#39;</span><span class="p">:</span>
+            <span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">content</span><span class="p">,</span> <span class="nb">list</span><span class="p">),</span> <span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Content of tools component must be a list, got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">content</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">type</span> <span class="ow">in</span> <span class="p">[</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="s1">&#39;string&#39;</span><span class="p">]:</span>
+            <span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">content</span><span class="p">,</span> <span class="nb">str</span><span class="p">),</span> <span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Content of string/token component must be a string, got </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">content</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;token_id&#39;</span><span class="p">:</span>
+            <span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">content</span><span class="p">,</span> <span class="nb">int</span><span class="p">)</span> <span class="ow">or</span> <span class="nb">all</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">token_id</span><span class="p">,</span> <span class="nb">int</span><span class="p">)</span> <span class="k">for</span> <span class="n">token_id</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">content</span><span class="p">),</span> <span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Content of token_id component must be an integer or a list of integers.&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;The type of the component must be either &quot;</span>
+                             <span class="sa">f</span><span class="s2">&quot;&#39;token&#39;, &#39;string&#39; or &#39;tools&#39;, got </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span></div>
+
+            
+<div class="viewcode-block" id="TemplateComponent.__repr__">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent.__repr__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;TemplateComponent(type=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="s2">, content=</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">content</span><span class="si">}</span><span class="s2">)&quot;</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">,</span> <span class="s2">&quot;</span><span class="se">\\</span><span class="s2">n&quot;</span><span class="p">)</span></div>
+
+    
+<div class="viewcode-block" id="TemplateComponent.__str__">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent.__str__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__str__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">str</span><span class="p">:</span>
+        <span class="k">return</span> <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">content</span><span class="si">}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">,</span> <span class="s2">&quot;</span><span class="se">\\</span><span class="s2">n&quot;</span><span class="p">)</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="Formatter">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.Formatter">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Formatter</span><span class="p">(</span><span class="n">ABC</span><span class="p">):</span>
+<div class="viewcode-block" id="Formatter.template">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.Formatter.template">[docs]</a>
+    <span class="n">template</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">TemplateComponent</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span><span class="n">default_factory</span><span class="o">=</span><span class="nb">list</span><span class="p">)</span></div>
+
+    
+    <span class="nd">@abstractmethod</span>
+<div class="viewcode-block" id="Formatter.format">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.Formatter.format">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">format</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="n">TemplateComponent</span><span class="p">]:</span> <span class="o">...</span></div>
+
+    
+<div class="viewcode-block" id="Formatter.has_placeholder">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.Formatter.has_placeholder">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">has_placeholder</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="n">flag</span> <span class="o">=</span> <span class="kc">False</span>
+        <span class="k">for</span> <span class="n">component</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">template</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">component</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;string&#39;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">re</span><span class="o">.</span><span class="n">search</span><span class="p">(</span><span class="sa">r</span><span class="s2">&quot;{{(.*?)}}&quot;</span><span class="p">,</span> <span class="n">component</span><span class="o">.</span><span class="n">content</span><span class="p">):</span>
+                    <span class="n">flag</span> <span class="o">=</span> <span class="kc">True</span>
+                    <span class="k">break</span>
+        <span class="k">return</span> <span class="n">flag</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="EmptyFormatter">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.EmptyFormatter">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">EmptyFormatter</span><span class="p">(</span><span class="n">Formatter</span><span class="p">):</span>
+<div class="viewcode-block" id="EmptyFormatter.__post_init__">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.EmptyFormatter.__post_init__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__post_init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">has_placeholder</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Empty formatter should not have placeholders.&quot;</span><span class="p">)</span></div>
+
+    
+<div class="viewcode-block" id="EmptyFormatter.format">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.EmptyFormatter.format">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">format</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">list</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Empty formatter for when no formatting is needed.</span>
+<span class="sd">        This is useful when user has already applied formatting to the dataset.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        -------</span>
+<span class="sd">        list</span>
+<span class="sd">            Original template.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">template</span></div>
+</div>
+
+    
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="StringFormatter">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.StringFormatter">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">StringFormatter</span><span class="p">(</span><span class="n">Formatter</span><span class="p">):</span>
+<div class="viewcode-block" id="StringFormatter.__post_init__">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.StringFormatter.__post_init__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__post_init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">has_placeholder</span><span class="p">():</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;String formatter should have placeholders.&quot;</span><span class="p">)</span></div>
+
+    
+<div class="viewcode-block" id="StringFormatter.format">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.StringFormatter.format">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">format</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">list</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Format the string components with the provided keyword arguments. </span>
+<span class="sd">        Mostly used for formatting system prompt, user and assistant messages.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ----------</span>
+<span class="sd">        **kwargs : dict</span>
+<span class="sd">            Keyword arguments containing values to replace in the template components.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        -------</span>
+<span class="sd">        list</span>
+<span class="sd">            Formatted template.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">formatted_template</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">component</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">template</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">component</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;string&#39;</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span> <span class="ow">in</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+                    <span class="n">templated</span> <span class="o">=</span> <span class="n">component</span><span class="o">.</span><span class="n">content</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;{{&quot;</span> <span class="o">+</span> <span class="n">key</span> <span class="o">+</span> <span class="s2">&quot;}}&quot;</span><span class="p">,</span> <span class="n">value</span><span class="p">)</span>
+                    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">templated</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;Found empty string after formatting, adding a space instead. &quot;</span>
+                                       <span class="s2">&quot;If this is not intended, please check the dataset.&quot;</span><span class="p">)</span>
+                        <span class="n">templated</span> <span class="o">=</span> <span class="s2">&quot; &quot;</span>
+                    <span class="n">formatted_template</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="n">templated</span><span class="p">))</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">formatted_template</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">component</span><span class="p">)</span>
+                
+        <span class="n">logger</span><span class="o">.</span><span class="n">debug</span><span class="p">(</span><span class="n">formatted_template</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">formatted_template</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="ListFormatter">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ListFormatter">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">ListFormatter</span><span class="p">(</span><span class="n">Formatter</span><span class="p">):</span>
+<div class="viewcode-block" id="ListFormatter.format">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ListFormatter.format">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">format</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">list</span><span class="p">:</span>
+        <span class="k">pass</span> <span class="c1"># Work in progress</span></div>
+</div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="ConversationTemplate">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplate">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">ConversationTemplate</span><span class="p">:</span>
+<div class="viewcode-block" id="ConversationTemplate.user_formatter">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.user_formatter">[docs]</a>
+    <span class="n">user_formatter</span><span class="p">:</span> <span class="n">Formatter</span></div>
+
+<div class="viewcode-block" id="ConversationTemplate.assistant_formatter">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.assistant_formatter">[docs]</a>
+    <span class="n">assistant_formatter</span><span class="p">:</span> <span class="n">Formatter</span></div>
+
+<div class="viewcode-block" id="ConversationTemplate.function_formatter">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.function_formatter">[docs]</a>
+    <span class="n">function_formatter</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Formatter</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="ConversationTemplate.observation_formatter">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.observation_formatter">[docs]</a>
+    <span class="n">observation_formatter</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Formatter</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="ConversationTemplate.system_formatter">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.system_formatter">[docs]</a>
+    <span class="n">system_formatter</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Formatter</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="ConversationTemplate.force_system">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.force_system">[docs]</a>
+    <span class="n">force_system</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+<div class="viewcode-block" id="ConversationTemplate.tools_formatter">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.tools_formatter">[docs]</a>
+    <span class="n">tools_formatter</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Formatter</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="ConversationTemplate.separator">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.separator">[docs]</a>
+    <span class="n">separator</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">TemplateComponent</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="ConversationTemplate.remove_last_sep">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.remove_last_sep">[docs]</a>
+    <span class="n">remove_last_sep</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+<div class="viewcode-block" id="ConversationTemplate.special_starter">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.special_starter">[docs]</a>
+    <span class="n">special_starter</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">TemplateComponent</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="ConversationTemplate.special_stopper">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.special_stopper">[docs]</a>
+    <span class="n">special_stopper</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">TemplateComponent</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="ConversationTemplate.template_name">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.template_name">[docs]</a>
+    <span class="n">template_name</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+    
+<div class="viewcode-block" id="ConversationTemplate.__post_init__">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.__post_init__">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">__post_init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">separator</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">separator</span><span class="o">.</span><span class="n">type</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="s1">&#39;token_id&#39;</span><span class="p">]:</span>
+                <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Component type </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">separator</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="s2"> cannot be used as a separator.&quot;</span><span class="p">)</span>
+            
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_starter</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_starter</span><span class="o">.</span><span class="n">type</span> <span class="ow">not</span> <span class="ow">in</span> <span class="p">[</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="s1">&#39;token_id&#39;</span><span class="p">]:</span>
+                <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Component type </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">special_starter</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="s2"> cannot be used as a special starter.&quot;</span><span class="p">)</span></div>
+
+    
+<div class="viewcode-block" id="ConversationTemplate.encode_conversation">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.encode_conversation">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">encode_conversation</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizer</span><span class="p">,</span>
+        <span class="n">messages</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]],</span>
+        <span class="n">system</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">tools</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&#39;&#39;&#39;</span>
+<span class="sd">        Messages here should be guaranteed to be in pairs, with the first message being the user message and the second message being the system message.</span>
+<span class="sd">        Data example: </span>
+<span class="sd">        ```json</span>
+<span class="sd">        {</span>
+<span class="sd">            &quot;conversation_id&quot;: 2,</span>
+<span class="sd">            &quot;system&quot;: &quot;sysinfo1&quot;,</span>
+<span class="sd">            &quot;tools&quot;: [&quot;tool_1_desc&quot;],</span>
+<span class="sd">            &quot;messages&quot;: [</span>
+<span class="sd">                {</span>
+<span class="sd">                    &quot;role&quot;: &quot;user&quot;,</span>
+<span class="sd">                    &quot;content&quot;: &quot;hi&quot;</span>
+<span class="sd">                },</span>
+<span class="sd">                {</span>
+<span class="sd">                    &quot;role&quot;: &quot;assistant&quot;,</span>
+<span class="sd">                    &quot;content&quot;: &quot;Hello!&quot;</span>
+<span class="sd">                }</span>
+<span class="sd">            ]</span>
+<span class="sd">        }</span>
+<span class="sd">        ```</span>
+<span class="sd">        &#39;&#39;&#39;</span>
+        <span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">messages</span><span class="p">,</span> <span class="nb">list</span><span class="p">),</span> <span class="s2">&quot;Messages must be a list.&quot;</span>
+        
+        <span class="k">if</span> <span class="n">tools</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;Tools are not supported yet. Please include tools in the system message manually.&quot;</span><span class="p">)</span>
+        
+        <span class="k">if</span> <span class="n">system</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">system</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot; &quot;</span><span class="p">,</span><span class="s2">&quot;&quot;</span><span class="p">):</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">system_formatter</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Your dataset contains system message but no system formatter is provided. &quot;</span>
+                                     <span class="s2">&quot;Consider either providing a system formatter or removing system prompt from your dataset.&quot;</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">system</span> <span class="o">=</span> <span class="kc">None</span>
+        
+        <span class="n">encoded_pairs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode</span><span class="p">(</span><span class="n">tokenizer</span><span class="p">,</span> <span class="n">messages</span><span class="p">,</span> <span class="n">system</span><span class="p">,</span> <span class="n">tools</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="n">encoded_pairs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">post_process_pairs</span><span class="p">(</span><span class="n">encoded_pairs</span><span class="o">=</span><span class="n">encoded_pairs</span><span class="p">,</span> <span class="n">tokenizer</span><span class="o">=</span><span class="n">tokenizer</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">encoded_pairs</span></div>
+
+        
+<div class="viewcode-block" id="ConversationTemplate._encode">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplate._encode">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_encode</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizer</span><span class="p">,</span>
+        <span class="n">messages</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]],</span>
+        <span class="n">system</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">tools</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+        <span class="c1"># TODO: truncation according to model max length</span>
+        <span class="c1"># TODO: make sure the last few tokens are &quot;learnable&quot;, not masked with token_id = -100.</span>
+        
+        <span class="n">res_all</span> <span class="o">=</span> <span class="p">[]</span>
+        
+        <span class="k">if</span> <span class="n">system</span><span class="p">:</span>
+            <span class="n">system_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">system_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">system</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">system_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">system_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="s1">&#39;&#39;</span><span class="p">)</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">force_system</span> <span class="k">else</span> <span class="p">[]</span>
+        <span class="n">system_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">system_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+        
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">messages</span><span class="p">),</span> <span class="mi">2</span><span class="p">):</span>
+            <span class="n">user_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+            <span class="n">assistant_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">]</span>
+            
+            <span class="n">user_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">user_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">user_message</span><span class="p">[</span><span class="s2">&quot;content&quot;</span><span class="p">])</span>
+            <span class="n">assistant_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">assistant_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">assistant_message</span><span class="p">[</span><span class="s2">&quot;content&quot;</span><span class="p">])</span>
+            
+            <span class="n">user_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">user_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+            <span class="n">assistant_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">assistant_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+            
+            <span class="n">res_all</span><span class="o">.</span><span class="n">append</span><span class="p">((</span>
+                <span class="n">system_encoded</span> <span class="o">+</span> <span class="n">user_encoded</span> <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="n">user_encoded</span><span class="p">,</span> 
+                <span class="n">assistant_encoded</span>
+            <span class="p">))</span>
+            
+        <span class="k">return</span> <span class="n">res_all</span></div>
+
+    
+<div class="viewcode-block" id="ConversationTemplate._encode_template">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplate._encode_template">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_encode_template</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">template</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">TemplateComponent</span><span class="p">],</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizer</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Encode template components into token ids.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ----------</span>
+<span class="sd">        template : List[TemplateComponent]</span>
+<span class="sd">            Formatted template components.</span>
+<span class="sd">        tokenizer : PreTrainedTokenizer</span>
+<span class="sd">            Tokenizer to convert tokens into token ids.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        -------</span>
+<span class="sd">        List[int]</span>
+<span class="sd">            Encoded token ids.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">encoded_ids</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">component</span> <span class="ow">in</span> <span class="n">template</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">component</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;string&#39;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">component</span><span class="o">.</span><span class="n">content</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;Empty string component found in the template.&quot;</span><span class="p">)</span>
+                    <span class="k">continue</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">encoded_ids</span> <span class="o">+=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="n">component</span><span class="o">.</span><span class="n">content</span><span class="p">,</span> <span class="n">add_special_tokens</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">component</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;token&#39;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">component</span><span class="o">.</span><span class="n">content</span> <span class="o">==</span> <span class="s1">&#39;bos_token&#39;</span><span class="p">:</span>
+                    <span class="n">encoded_ids</span> <span class="o">+=</span> <span class="p">[</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">bos_token_id</span><span class="p">]</span>
+                <span class="k">elif</span> <span class="n">component</span><span class="o">.</span><span class="n">content</span> <span class="o">==</span> <span class="s1">&#39;eos_token&#39;</span><span class="p">:</span>
+                    <span class="n">encoded_ids</span> <span class="o">+=</span> <span class="p">[</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span><span class="p">]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">encoded_ids</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ensure_id_list</span><span class="p">(</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">convert_tokens_to_ids</span><span class="p">(</span><span class="n">component</span><span class="o">.</span><span class="n">content</span><span class="p">))</span>
+            <span class="k">elif</span> <span class="n">component</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;token_id&#39;</span><span class="p">:</span>
+                <span class="n">encoded_ids</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ensure_id_list</span><span class="p">(</span><span class="n">component</span><span class="o">.</span><span class="n">content</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Component type </span><span class="si">{</span><span class="n">component</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="s2"> is not supported yet.&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">encoded_ids</span></div>
+
+    
+<div class="viewcode-block" id="ConversationTemplate.post_process_pairs">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.post_process_pairs">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">post_process_pairs</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">encoded_pairs</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">):</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">separator</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">remove_last_sep</span><span class="p">:</span>
+            <span class="c1"># For models that require a separator between messages, </span>
+            <span class="c1"># user can include the seperator at the end of each template</span>
+            <span class="c1"># and specify the separator. Auto formatting will remove the </span>
+            <span class="c1"># last separator once user specifies this option.</span>
+            <span class="n">encoded_pairs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">remove_last_separator</span><span class="p">(</span><span class="n">encoded_pairs</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+            
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_starter</span><span class="p">:</span>
+            <span class="c1"># For models that has ONLY ONE bos token at the beginning of </span>
+            <span class="c1"># a conversation session (not a conversation pair), user can</span>
+            <span class="c1"># specify a special starter to add that starter to the very</span>
+            <span class="c1"># beginning of the conversation session. </span>
+            <span class="c1"># eg:</span>
+            <span class="c1">#   llama-2: &lt;s&gt; and &lt;/s&gt; at every pair of conversation </span>
+            <span class="c1">#   v.s.</span>
+            <span class="c1">#   llama-3: &lt;|begin_of_text|&gt; only at the beginning of a session</span>
+            <span class="n">encoded_pairs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">add_special_starter</span><span class="p">(</span><span class="n">encoded_pairs</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+            
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_stopper</span><span class="p">:</span>
+            <span class="n">encoded_pairs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">add_special_stopper</span><span class="p">(</span><span class="n">encoded_pairs</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+            
+        <span class="k">return</span> <span class="n">encoded_pairs</span></div>
+
+    
+<div class="viewcode-block" id="ConversationTemplate.remove_last_separator">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.remove_last_separator">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">remove_last_separator</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">encoded_pairs</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizer</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+        <span class="n">last_assistant_msg</span> <span class="o">=</span> <span class="n">encoded_pairs</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">][</span><span class="mi">1</span><span class="p">]</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">separator</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;string&#39;</span><span class="p">:</span>
+            <span class="n">separator_ids</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">separator</span><span class="o">.</span><span class="n">content</span><span class="p">,</span> <span class="n">add_special_tokens</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">separator</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;token&#39;</span><span class="p">:</span>
+            <span class="n">separator_ids</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ensure_id_list</span><span class="p">(</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">convert_tokens_to_ids</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">separator</span><span class="o">.</span><span class="n">content</span><span class="p">))</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">separator</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;token_id&#39;</span><span class="p">:</span>
+            <span class="n">separator_ids</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ensure_id_list</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">separator</span><span class="o">.</span><span class="n">content</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Component type </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">separator</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="s2"> cannot be used as a separator.&quot;</span><span class="p">)</span>
+        
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">separator_ids</span><span class="p">)</span> <span class="o">&gt;</span> <span class="nb">len</span><span class="p">(</span><span class="n">last_assistant_msg</span><span class="p">):</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Separator is longer than the last assistant message, please check.&quot;</span><span class="p">)</span>
+        
+        <span class="k">if</span> <span class="n">last_assistant_msg</span><span class="p">[</span><span class="o">-</span><span class="nb">len</span><span class="p">(</span><span class="n">separator_ids</span><span class="p">):]</span> <span class="o">==</span> <span class="n">separator_ids</span><span class="p">:</span>
+            <span class="n">last_assistant_msg</span> <span class="o">=</span> <span class="n">last_assistant_msg</span><span class="p">[:</span><span class="o">-</span><span class="nb">len</span><span class="p">(</span><span class="n">separator_ids</span><span class="p">)]</span>
+            
+        <span class="n">encoded_pairs</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="n">encoded_pairs</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">][</span><span class="mi">0</span><span class="p">],</span> <span class="n">last_assistant_msg</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">encoded_pairs</span></div>
+
+    
+<div class="viewcode-block" id="ConversationTemplate.add_special_starter">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.add_special_starter">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">add_special_starter</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">encoded_pairs</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizer</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_starter</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;string&#39;</span><span class="p">:</span>
+            <span class="n">special_starter_ids</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">special_starter</span><span class="o">.</span><span class="n">content</span><span class="p">,</span> <span class="n">add_special_tokens</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_starter</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;token&#39;</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_starter</span><span class="o">.</span><span class="n">content</span> <span class="o">==</span> <span class="s1">&#39;bos_token&#39;</span><span class="p">:</span>
+                <span class="n">special_starter_ids</span> <span class="o">=</span> <span class="p">[</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">bos_token_id</span><span class="p">]</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_starter</span><span class="o">.</span><span class="n">content</span> <span class="o">==</span> <span class="s1">&#39;eos_token&#39;</span><span class="p">:</span>
+                <span class="n">special_starter_ids</span> <span class="o">=</span> <span class="p">[</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span><span class="p">]</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">special_starter_ids</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ensure_id_list</span><span class="p">(</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">convert_tokens_to_ids</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">special_starter</span><span class="o">.</span><span class="n">content</span><span class="p">))</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_starter</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;token_id&#39;</span><span class="p">:</span>
+            <span class="n">special_starter_ids</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ensure_id_list</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">special_starter</span><span class="o">.</span><span class="n">content</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Component type </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">special_starter</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="s2"> cannot be used as a special starter.&quot;</span><span class="p">)</span>
+        
+        <span class="n">encoded_pairs</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="n">special_starter_ids</span> <span class="o">+</span> <span class="n">encoded_pairs</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="mi">0</span><span class="p">],</span> <span class="n">encoded_pairs</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="mi">1</span><span class="p">])</span>
+        
+        <span class="k">return</span> <span class="n">encoded_pairs</span></div>
+
+    
+<div class="viewcode-block" id="ConversationTemplate.add_special_stopper">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.add_special_stopper">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">add_special_stopper</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">encoded_pairs</span><span class="p">:</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]],</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizer</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_stopper</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;string&#39;</span><span class="p">:</span>
+            <span class="n">special_stopper_ids</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">special_stopper</span><span class="o">.</span><span class="n">content</span><span class="p">,</span> <span class="n">add_special_tokens</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_stopper</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;token&#39;</span><span class="p">:</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_stopper</span><span class="o">.</span><span class="n">content</span> <span class="o">==</span> <span class="s1">&#39;bos_token&#39;</span><span class="p">:</span>
+                <span class="n">special_stopper_ids</span> <span class="o">=</span> <span class="p">[</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">bos_token_id</span><span class="p">]</span>
+            <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_stopper</span><span class="o">.</span><span class="n">content</span> <span class="o">==</span> <span class="s1">&#39;eos_token&#39;</span><span class="p">:</span>
+                <span class="n">special_stopper_ids</span> <span class="o">=</span> <span class="p">[</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span><span class="p">]</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">special_stopper_ids</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ensure_id_list</span><span class="p">(</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">convert_tokens_to_ids</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">special_stopper</span><span class="o">.</span><span class="n">content</span><span class="p">))</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">special_stopper</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;token_id&#39;</span><span class="p">:</span>
+            <span class="n">special_stopper_ids</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ensure_id_list</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">special_stopper</span><span class="o">.</span><span class="n">content</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Component type </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">special_stopper</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="s2"> cannot be used as a special stopper.&quot;</span><span class="p">)</span>
+        
+        <span class="n">encoded_pairs</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="n">encoded_pairs</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">][</span><span class="mi">0</span><span class="p">],</span> <span class="n">encoded_pairs</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">][</span><span class="mi">1</span><span class="p">]</span> <span class="o">+</span> <span class="n">special_stopper_ids</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">encoded_pairs</span></div>
+
+    
+<div class="viewcode-block" id="ConversationTemplate._ensure_id_list">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplate._ensure_id_list">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_ensure_id_list</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">obj</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]])</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&#39;&#39;&#39;Make sure the object is a list of integers. Useful for handling token ids.</span>
+<span class="sd">        &#39;&#39;&#39;</span>
+        <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">obj</span><span class="p">,</span> <span class="nb">int</span><span class="p">):</span>
+            <span class="k">return</span> <span class="p">[</span><span class="n">obj</span><span class="p">]</span>
+        <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">obj</span><span class="p">,</span> <span class="nb">list</span><span class="p">):</span>
+            <span class="k">return</span> <span class="n">obj</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Object type </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">obj</span><span class="p">)</span><span class="si">}</span><span class="s2"> is not supported yet.&quot;</span><span class="p">)</span></div>
+</div>
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="ConversationTemplateForTool">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplateForTool">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">ConversationTemplateForTool</span><span class="p">(</span><span class="n">ConversationTemplate</span><span class="p">):</span>
+<div class="viewcode-block" id="ConversationTemplateForTool.encode_conversation">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplateForTool.encode_conversation">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">encode_conversation</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizer</span><span class="p">,</span>
+        <span class="n">messages</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]],</span>
+        <span class="n">system</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">tools</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&#39;&#39;&#39;</span>
+<span class="sd">        Messages here should be guaranteed to be in pairs, with the first message being the user message and the second message being the system message.</span>
+<span class="sd">        Data example: </span>
+<span class="sd">        ```json</span>
+<span class="sd">        {</span>
+<span class="sd">            &quot;conversation_id&quot;: 2,</span>
+<span class="sd">            &quot;system&quot;: &quot;sysinfo1&quot;,</span>
+<span class="sd">            &quot;tools&quot;: [&quot;tool_1_desc&quot;],</span>
+<span class="sd">            &quot;messages&quot;: [</span>
+<span class="sd">                {</span>
+<span class="sd">                    &quot;role&quot;: &quot;user&quot;,</span>
+<span class="sd">                    &quot;content&quot;: &quot;hi&quot;</span>
+<span class="sd">                },</span>
+<span class="sd">                {</span>
+<span class="sd">                    &quot;role&quot;: &quot;assistant&quot;,</span>
+<span class="sd">                    &quot;content&quot;: &quot;Hello!&quot;</span>
+<span class="sd">                }</span>
+<span class="sd">            ]</span>
+<span class="sd">        }</span>
+<span class="sd">        ```</span>
+<span class="sd">        &#39;&#39;&#39;</span>
+        <span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">messages</span><span class="p">,</span> <span class="nb">list</span><span class="p">),</span> <span class="s2">&quot;Messages must be a list.&quot;</span>
+        
+        <span class="k">if</span> <span class="n">tools</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">tools</span> <span class="o">=</span> <span class="s1">&#39;&#39;</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">tools</span> <span class="o">=</span> <span class="s1">&#39;,&#39;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">tools</span><span class="p">)</span>
+            <span class="c1"># logger.warning(&quot;Tools are not supported yet. Please include tools in the system message manually.&quot;)</span>
+        
+        <span class="k">if</span> <span class="n">system</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">system</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">system</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot; &quot;</span><span class="p">,</span><span class="s2">&quot;&quot;</span><span class="p">):</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">system_formatter</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Your dataset contains system message but no system formatter is provided. &quot;</span>
+                                     <span class="s2">&quot;Consider either providing a system formatter or removing system prompt from your dataset.&quot;</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">system</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">encoded_pairs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode</span><span class="p">(</span><span class="n">tokenizer</span><span class="p">,</span> <span class="n">messages</span><span class="p">,</span> <span class="n">system</span><span class="p">,</span> <span class="n">tools</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="n">encoded_pairs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">post_process_pairs</span><span class="p">(</span><span class="n">encoded_pairs</span><span class="o">=</span><span class="n">encoded_pairs</span><span class="p">,</span> <span class="n">tokenizer</span><span class="o">=</span><span class="n">tokenizer</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">encoded_pairs</span></div>
+
+        
+<div class="viewcode-block" id="ConversationTemplateForTool._encode">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplateForTool._encode">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_encode</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizer</span><span class="p">,</span>
+        <span class="n">messages</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]],</span>
+        <span class="n">system</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">tools</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+        <span class="c1"># TODO: truncation according to model max length</span>
+        <span class="c1"># TODO: make sure the last few tokens are &quot;learnable&quot;, not masked with token_id = -100.</span>
+        
+        <span class="n">res_all</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="c1"># Concatenate the system and tools strings</span>
+        <span class="n">system</span> <span class="o">=</span> <span class="n">system</span> <span class="o">+</span> <span class="n">tools</span>
+        <span class="k">if</span> <span class="n">system</span><span class="p">:</span>
+            <span class="n">system_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">system_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">system</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">system_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">system_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="s1">&#39;&#39;</span><span class="p">)</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">force_system</span> <span class="k">else</span> <span class="p">[]</span>
+        <span class="n">system_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">system_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+        <span class="n">ls_for_save</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">messages</span><span class="p">),</span> <span class="mi">1</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="n">CONVERSATION_ROLE_NAMES</span><span class="p">[</span><span class="s1">&#39;user&#39;</span><span class="p">]:</span>
+                <span class="n">user_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+                <span class="n">user_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">user_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">user_message</span><span class="p">[</span><span class="s2">&quot;content&quot;</span><span class="p">])</span>
+                <span class="n">user_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">user_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+                <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">user_encoded</span> <span class="o">=</span> <span class="n">system_encoded</span> <span class="o">+</span> <span class="n">user_encoded</span>
+                <span class="n">ls_for_save</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">user_encoded</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="n">CONVERSATION_ROLE_NAMES</span><span class="p">[</span><span class="s1">&#39;function&#39;</span><span class="p">]:</span>
+                <span class="n">function_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+                <span class="n">function_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">function_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">function_message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">])</span>
+                <span class="n">function_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">function_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+                <span class="n">ls_for_save</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">function_encoded</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="n">CONVERSATION_ROLE_NAMES</span><span class="p">[</span><span class="s1">&#39;observation&#39;</span><span class="p">]:</span>
+                <span class="n">observation_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+                <span class="n">observation_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">observation_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">observation_message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">])</span>
+                <span class="n">observation_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">observation_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+                <span class="n">ls_for_save</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">observation_encoded</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="n">CONVERSATION_ROLE_NAMES</span><span class="p">[</span><span class="s1">&#39;assistant&#39;</span><span class="p">]:</span>
+                <span class="n">assistant_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+                <span class="n">assistant_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">assistant_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">assistant_message</span><span class="p">[</span><span class="s2">&quot;content&quot;</span><span class="p">])</span>
+                <span class="n">assistant_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">assistant_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+                <span class="n">ls_for_save</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">assistant_encoded</span><span class="p">)</span>
+                <span class="c1"># res_tuple = (ls_for_save[0], ls_for_save[1], ls_for_save[2], ls_for_save[3])</span>
+                <span class="n">res_all</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="nb">tuple</span><span class="p">(</span><span class="n">ls_for_save</span><span class="p">))</span>
+                <span class="n">ls_for_save</span> <span class="o">=</span> <span class="p">[]</span>
+        
+        <span class="k">if</span> <span class="n">ls_for_save</span><span class="p">:</span>
+            <span class="n">res_all</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="nb">tuple</span><span class="p">(</span><span class="n">ls_for_save</span><span class="p">))</span>
+            
+        <span class="k">return</span> <span class="n">res_all</span></div>
+
+    
+<div class="viewcode-block" id="ConversationTemplateForTool._encode_template">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.ConversationTemplateForTool._encode_template">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_encode_template</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> 
+        <span class="n">template</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">TemplateComponent</span><span class="p">],</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizer</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Encode template components into token ids.</span>
+
+<span class="sd">        Parameters</span>
+<span class="sd">        ----------</span>
+<span class="sd">        template : List[TemplateComponent]</span>
+<span class="sd">            Formatted template components.</span>
+<span class="sd">        tokenizer : PreTrainedTokenizer</span>
+<span class="sd">            Tokenizer to convert tokens into token ids.</span>
+
+<span class="sd">        Returns</span>
+<span class="sd">        -------</span>
+<span class="sd">        List[int]</span>
+<span class="sd">            Encoded token ids.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="n">encoded_ids</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">component</span> <span class="ow">in</span> <span class="n">template</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">component</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;string&#39;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">component</span><span class="o">.</span><span class="n">content</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;Empty string component found in the template.&quot;</span><span class="p">)</span>
+                    <span class="k">continue</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">encoded_ids</span> <span class="o">+=</span> <span class="n">tokenizer</span><span class="o">.</span><span class="n">encode</span><span class="p">(</span><span class="n">component</span><span class="o">.</span><span class="n">content</span><span class="p">,</span> <span class="n">add_special_tokens</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">component</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;token&#39;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">component</span><span class="o">.</span><span class="n">content</span> <span class="o">==</span> <span class="s1">&#39;bos_token&#39;</span><span class="p">:</span>
+                    <span class="n">encoded_ids</span> <span class="o">+=</span> <span class="p">[</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">bos_token_id</span><span class="p">]</span>
+                <span class="k">elif</span> <span class="n">component</span><span class="o">.</span><span class="n">content</span> <span class="o">==</span> <span class="s1">&#39;eos_token&#39;</span><span class="p">:</span>
+                    <span class="n">encoded_ids</span> <span class="o">+=</span> <span class="p">[</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">eos_token_id</span><span class="p">]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">encoded_ids</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ensure_id_list</span><span class="p">(</span><span class="n">tokenizer</span><span class="o">.</span><span class="n">convert_tokens_to_ids</span><span class="p">(</span><span class="n">component</span><span class="o">.</span><span class="n">content</span><span class="p">))</span>
+            <span class="k">elif</span> <span class="n">component</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;token_id&#39;</span><span class="p">:</span>
+                <span class="n">encoded_ids</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ensure_id_list</span><span class="p">(</span><span class="n">component</span><span class="o">.</span><span class="n">content</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Component type </span><span class="si">{</span><span class="n">component</span><span class="o">.</span><span class="n">type</span><span class="si">}</span><span class="s2"> is not supported yet.&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">encoded_ids</span></div>
+</div>
+
+
+
+<div class="viewcode-block" id="EMPTY_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.EMPTY_TEMPLATE">[docs]</a>
+<span class="n">EMPTY_TEMPLATE</span> <span class="o">=</span> <span class="n">ConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;empty&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;bos_token&#39;</span><span class="p">),</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;{{content}}&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;{{content}}&#39;</span><span class="p">),</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;eos_token&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">)</span>
+<span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="EMPTY_NO_SPECIAL_TOKENS_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.base.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE">[docs]</a>
+<span class="n">EMPTY_NO_SPECIAL_TOKENS_TEMPLATE</span> <span class="o">=</span> <span class="n">ConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;empty_no_special_tokens&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;{{content}}&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;{{content}}&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">)</span>
+<span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/conversation_template/chatglm.html b/_modules/lmflow/utils/conversation_template/chatglm.html
new file mode 100644
index 000000000..894899c3d
--- /dev/null
+++ b/_modules/lmflow/utils/conversation_template/chatglm.html
@@ -0,0 +1,472 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.conversation_template.chatglm &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/conversation_template/chatglm';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../conversation_template.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.chatglm</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.conversation_template.chatglm</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.base</span><span class="w"> </span><span class="kn">import</span> <span class="n">StringFormatter</span><span class="p">,</span> <span class="n">TemplateComponent</span><span class="p">,</span> <span class="n">ConversationTemplate</span>
+
+
+<div class="viewcode-block" id="CHATGLM3_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.chatglm.CHATGLM3_TEMPLATE">[docs]</a>
+<span class="n">CHATGLM3_TEMPLATE</span> <span class="o">=</span> <span class="n">ConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;chatglm3&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|user|&gt;</span><span class="se">\n</span><span class="s1"> {{content}}&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|assistant|&gt;</span><span class="se">\n</span><span class="s1"> {{content}}&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|system|&gt;</span><span class="se">\n</span><span class="s1"> {{content}}&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">special_starter</span><span class="o">=</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;[gMASK]sop&#39;</span><span class="p">)</span>
+<span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/conversation_template/chatml.html b/_modules/lmflow/utils/conversation_template/chatml.html
new file mode 100644
index 000000000..7436fff6d
--- /dev/null
+++ b/_modules/lmflow/utils/conversation_template/chatml.html
@@ -0,0 +1,471 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.conversation_template.chatml &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/conversation_template/chatml';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../conversation_template.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.chatml</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.conversation_template.chatml</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.base</span><span class="w"> </span><span class="kn">import</span> <span class="n">StringFormatter</span><span class="p">,</span> <span class="n">TemplateComponent</span><span class="p">,</span> <span class="n">ConversationTemplate</span>
+
+
+<div class="viewcode-block" id="CHATML_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.chatml.CHATML_TEMPLATE">[docs]</a>
+<span class="n">CHATML_TEMPLATE</span> <span class="o">=</span> <span class="n">ConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;chatml&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;user</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;assistant</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;system</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">)</span>
+<span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/conversation_template/deepseek.html b/_modules/lmflow/utils/conversation_template/deepseek.html
new file mode 100644
index 000000000..666c5dbc8
--- /dev/null
+++ b/_modules/lmflow/utils/conversation_template/deepseek.html
@@ -0,0 +1,473 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.conversation_template.deepseek &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/conversation_template/deepseek';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../conversation_template.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.deepseek</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.conversation_template.deepseek</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.base</span><span class="w"> </span><span class="kn">import</span> <span class="n">StringFormatter</span><span class="p">,</span> <span class="n">TemplateComponent</span><span class="p">,</span> <span class="n">ConversationTemplate</span>
+
+
+<div class="viewcode-block" id="DEEPSEEK_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.deepseek.DEEPSEEK_TEMPLATE">[docs]</a>
+<span class="n">DEEPSEEK_TEMPLATE</span> <span class="o">=</span> <span class="n">ConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;deepseek&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;User: {{content}}</span><span class="se">\n\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;Assistant: {{content}}&#39;</span><span class="p">),</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;eos_token&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;{{content}}</span><span class="se">\n\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">special_starter</span><span class="o">=</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;bos_token&#39;</span><span class="p">)</span>
+<span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/conversation_template/gemma.html b/_modules/lmflow/utils/conversation_template/gemma.html
new file mode 100644
index 000000000..6dead8160
--- /dev/null
+++ b/_modules/lmflow/utils/conversation_template/gemma.html
@@ -0,0 +1,501 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.conversation_template.gemma &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/conversation_template/gemma';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../conversation_template.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.gemma</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.conversation_template.gemma</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">dataclasses</span><span class="w"> </span><span class="kn">import</span> <span class="n">dataclass</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">.base</span><span class="w"> </span><span class="kn">import</span> <span class="n">StringFormatter</span><span class="p">,</span> <span class="n">TemplateComponent</span><span class="p">,</span> <span class="n">ConversationTemplate</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/gemma/index.html#lmflow.utils.conversation_template.gemma.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<span class="nd">@dataclass</span>
+<div class="viewcode-block" id="GemmaConversationTemplate">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/gemma/index.html#lmflow.utils.conversation_template.gemma.GemmaConversationTemplate">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">GemmaConversationTemplate</span><span class="p">(</span><span class="n">ConversationTemplate</span><span class="p">):</span>
+<div class="viewcode-block" id="GemmaConversationTemplate.encode_conversation">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/gemma/index.html#lmflow.utils.conversation_template.gemma.GemmaConversationTemplate.encode_conversation">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">encode_conversation</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
+        <span class="k">if</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">&#39;system&#39;</span><span class="p">):</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span>
+                <span class="s1">&#39;As of now, Gemma does not support system messages officially. &#39;</span>
+                <span class="s1">&#39;ConversationTemplate will add your system messages right after &#39;</span>
+                <span class="s1">&#39;the bos token and before the user message without any special formatting. &#39;</span>
+                <span class="s1">&#39;For more details, please refer to the [official template]&#39;</span>
+                <span class="s1">&#39;(https://huggingface.co/google/gemma-1.1-2b-it/blob/bf4924f313df5166dee1467161e886e55f2eb4d4/tokenizer_config.json#L1507).&#39;</span>
+            <span class="p">)</span>
+        <span class="k">return</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">encode_conversation</span><span class="p">(</span><span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span></div>
+</div>
+
+        
+
+<div class="viewcode-block" id="GEMMA_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.gemma.GEMMA_TEMPLATE">[docs]</a>
+<span class="n">GEMMA_TEMPLATE</span> <span class="o">=</span> <span class="n">GemmaConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;gemma&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;start_of_turn&gt;user</span><span class="se">\n</span><span class="s1">{{content}}&lt;end_of_turn&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;start_of_turn&gt;model</span><span class="se">\n</span><span class="s1">{{content}}&lt;end_of_turn&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;{{content}}&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">special_starter</span><span class="o">=</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;bos_token&#39;</span><span class="p">)</span>
+<span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/conversation_template/hymba.html b/_modules/lmflow/utils/conversation_template/hymba.html
new file mode 100644
index 000000000..9e7402e89
--- /dev/null
+++ b/_modules/lmflow/utils/conversation_template/hymba.html
@@ -0,0 +1,576 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.conversation_template.hymba &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/conversation_template/hymba';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../conversation_template.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.hymba</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.conversation_template.hymba</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.base</span><span class="w"> </span><span class="kn">import</span> <span class="n">StringFormatter</span><span class="p">,</span> <span class="n">TemplateComponent</span><span class="p">,</span> <span class="n">ConversationTemplateForTool</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Set</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Literal</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">PreTrainedTokenizer</span>
+
+<span class="c1"># NOTE: &#39;contexts&#39; are not used in sft</span>
+<span class="c1"># {{&#39;&lt;extra_id_0&gt;System&#39;}}</span>
+<span class="c1"># {% for message in messages %}</span>
+<span class="c1">#   {% if message[&#39;role&#39;] == &#39;system&#39; %}</span>
+<span class="c1">#       {{&#39;\n&#39; + message[&#39;content&#39;].strip()}}</span>
+<span class="c1">#       {% if tools %}</span>
+<span class="c1">#           {{&#39;\n&#39;}}</span>
+<span class="c1">#       {% endif %}</span>
+<span class="c1">#   {% endif %}</span>
+<span class="c1"># {% endfor %}</span>
+<span class="c1"># {% if tools %}</span>
+<span class="c1">#   {% for tool in tools %}</span>
+<span class="c1">#       {{ &#39;\n&lt;tool&gt; &#39; + tool|tojson + &#39; &lt;/tool&gt;&#39; }}</span>
+<span class="c1">#   {% endfor %}</span>
+<span class="c1"># {% endif %}</span>
+<span class="c1"># {{&#39;\n\n&#39;}}</span>
+<span class="c1"># {% for message in messages %}</span>
+<span class="c1">#   {% if message[&#39;role&#39;] == &#39;user&#39; %}</span>
+<span class="c1">#       {{ &#39;&lt;extra_id_1&gt;User\n&#39; + message[&#39;content&#39;].strip() + &#39;\n&#39; }}</span>
+<span class="c1">#   {% elif message[&#39;role&#39;] == &#39;assistant&#39; %}</span>
+<span class="c1">#       {{ &#39;&lt;extra_id_1&gt;Assistant\n&#39; + message[&#39;content&#39;].strip() + &#39;\n&#39; }}</span>
+<span class="c1">#   {% elif message[&#39;role&#39;] == &#39;tool&#39; %}</span>
+<span class="c1">#       {{ &#39;&lt;extra_id_1&gt;Tool\n&#39; + message[&#39;content&#39;].strip() + &#39;\n&#39; }}</span>
+<span class="c1">#   {% endif %}</span>
+<span class="c1"># {% endfor %}</span>
+<span class="c1"># {%- if add_generation_prompt %}</span>
+<span class="c1">#   {{&#39;&lt;extra_id_1&gt;Assistant\n&#39;}}</span>
+<span class="c1"># {%- endif %}</span>
+
+
+<div class="viewcode-block" id="HymbaConversationTemplate">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/hymba/index.html#lmflow.utils.conversation_template.hymba.HymbaConversationTemplate">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">HymbaConversationTemplate</span><span class="p">(</span><span class="n">ConversationTemplateForTool</span><span class="p">):</span>
+<div class="viewcode-block" id="HymbaConversationTemplate.encode_conversation">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/hymba/index.html#lmflow.utils.conversation_template.hymba.HymbaConversationTemplate.encode_conversation">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">encode_conversation</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizer</span><span class="p">,</span>
+        <span class="n">messages</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]],</span>
+        <span class="n">system</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">tools</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+<span class="w">        </span><span class="sa">r</span><span class="sd">&#39;&#39;&#39;</span>
+<span class="sd">        Messages here should be guaranteed to be in pairs, with the first message being the user message and the second message being the system message.</span>
+<span class="sd">        Data example: </span>
+<span class="sd">        ```json</span>
+<span class="sd">        {</span>
+<span class="sd">            &quot;conversation_id&quot;: 2,</span>
+<span class="sd">            &quot;system&quot;: &quot;sysinfo1&quot;,</span>
+<span class="sd">            &quot;tools&quot;: [&quot;tool_1_desc&quot;],</span>
+<span class="sd">            &quot;messages&quot;: [</span>
+<span class="sd">                {</span>
+<span class="sd">                    &quot;role&quot;: &quot;user&quot;,</span>
+<span class="sd">                    &quot;content&quot;: &quot;hi&quot;</span>
+<span class="sd">                },</span>
+<span class="sd">                {</span>
+<span class="sd">                    &quot;role&quot;: &quot;assistant&quot;,</span>
+<span class="sd">                    &quot;content&quot;: &quot;Hello!&quot;</span>
+<span class="sd">                }</span>
+<span class="sd">            ]</span>
+<span class="sd">        }</span>
+<span class="sd">        ```</span>
+<span class="sd">        &#39;&#39;&#39;</span>
+        <span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">messages</span><span class="p">,</span> <span class="nb">list</span><span class="p">),</span> <span class="s2">&quot;Messages must be a list.&quot;</span>
+        
+        <span class="n">tools_out</span> <span class="o">=</span> <span class="s1">&#39;&#39;</span>
+        <span class="k">if</span> <span class="n">tools</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">for</span> <span class="n">tool</span> <span class="ow">in</span> <span class="n">tools</span><span class="p">:</span>
+                <span class="n">tools_out</span> <span class="o">+=</span> <span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&lt;tool&gt; &quot;</span> <span class="o">+</span> <span class="n">tool</span> <span class="o">+</span> <span class="s2">&quot; &lt;/tool&gt;&quot;</span>
+        
+        <span class="k">if</span> <span class="n">system</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">system</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">system</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot; &quot;</span><span class="p">,</span><span class="s2">&quot;&quot;</span><span class="p">):</span> <span class="c1"># has actual content</span>
+                <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">system_formatter</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Your dataset contains system message but no system formatter is provided. &quot;</span>
+                                     <span class="s2">&quot;Consider either providing a system formatter or removing system prompt from your dataset.&quot;</span><span class="p">)</span>
+                <span class="n">system</span> <span class="o">=</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">system</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">system</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
+        <span class="n">encoded_pairs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode</span><span class="p">(</span><span class="n">tokenizer</span><span class="p">,</span> <span class="n">messages</span><span class="p">,</span> <span class="n">system</span><span class="p">,</span> <span class="n">tools_out</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
+        <span class="n">encoded_pairs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">post_process_pairs</span><span class="p">(</span><span class="n">encoded_pairs</span><span class="o">=</span><span class="n">encoded_pairs</span><span class="p">,</span> <span class="n">tokenizer</span><span class="o">=</span><span class="n">tokenizer</span><span class="p">)</span>
+        
+        <span class="k">return</span> <span class="n">encoded_pairs</span>    </div>
+</div>
+
+
+
+<div class="viewcode-block" id="HYMBA_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.hymba.HYMBA_TEMPLATE">[docs]</a>
+<span class="n">HYMBA_TEMPLATE</span> <span class="o">=</span> <span class="n">HymbaConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;hymba&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;extra_id_1&gt;User</span><span class="se">\n</span><span class="s1">{{content}}</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;extra_id_1&gt;Assistant</span><span class="se">\n</span><span class="s1">{{content}}</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">function_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;extra_id_1&gt;Assistant</span><span class="se">\n</span><span class="s1">{{content}}</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">observation_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;extra_id_1&gt;Tool</span><span class="se">\n</span><span class="s1">{{content}}</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;extra_id_0&gt;System{{content}}</span><span class="se">\n\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">separator</span><span class="o">=</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token_id&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="mi">13</span><span class="p">),</span>
+    <span class="n">remove_last_sep</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+    <span class="n">special_stopper</span><span class="o">=</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;eos_token&#39;</span><span class="p">),</span>
+    <span class="n">force_system</span><span class="o">=</span><span class="kc">True</span>
+<span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/conversation_template/internlm.html b/_modules/lmflow/utils/conversation_template/internlm.html
new file mode 100644
index 000000000..2aea84a12
--- /dev/null
+++ b/_modules/lmflow/utils/conversation_template/internlm.html
@@ -0,0 +1,472 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.conversation_template.internlm &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/conversation_template/internlm';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../conversation_template.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.internlm</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.conversation_template.internlm</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.base</span><span class="w"> </span><span class="kn">import</span> <span class="n">StringFormatter</span><span class="p">,</span> <span class="n">TemplateComponent</span><span class="p">,</span> <span class="n">ConversationTemplate</span>
+
+
+<div class="viewcode-block" id="INTERNLM2_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/internlm/index.html#lmflow.utils.conversation_template.INTERNLM2_TEMPLATE">[docs]</a>
+<span class="n">INTERNLM2_TEMPLATE</span> <span class="o">=</span> <span class="n">ConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;internlm2&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;user</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;assistant</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;system</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">special_starter</span><span class="o">=</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;bos_token&#39;</span><span class="p">)</span>
+<span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/conversation_template/llama.html b/_modules/lmflow/utils/conversation_template/llama.html
new file mode 100644
index 000000000..8383af531
--- /dev/null
+++ b/_modules/lmflow/utils/conversation_template/llama.html
@@ -0,0 +1,664 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.conversation_template.llama &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/conversation_template/llama';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../conversation_template.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.llama</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.conversation_template.llama</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Set</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Literal</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">PreTrainedTokenizer</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">.base</span><span class="w"> </span><span class="kn">import</span> <span class="n">StringFormatter</span><span class="p">,</span> <span class="n">TemplateComponent</span><span class="p">,</span> <span class="n">ConversationTemplate</span><span class="p">,</span> <span class="n">ConversationTemplateForTool</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.utils.constants</span><span class="w"> </span><span class="kn">import</span> <span class="n">CONVERSATION_ROLE_NAMES</span>
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="Llama2ConversationTemplate">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.Llama2ConversationTemplate">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Llama2ConversationTemplate</span><span class="p">(</span><span class="n">ConversationTemplate</span><span class="p">):</span>    
+<div class="viewcode-block" id="Llama2ConversationTemplate._encode">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.Llama2ConversationTemplate._encode">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_encode</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizer</span><span class="p">,</span>
+        <span class="n">messages</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]],</span>
+        <span class="n">system</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">tools</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+        <span class="k">if</span> <span class="n">tools</span><span class="p">:</span>
+            <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="s2">&quot;Formatted tools are not supported in Llama2, thus tools will be ignored. &quot;</span>
+                           <span class="s2">&quot;If this is intended, please include tools in the system message manually.&quot;</span><span class="p">)</span>
+        
+        <span class="n">res_all</span> <span class="o">=</span> <span class="p">[]</span>
+        
+        <span class="n">system_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">system_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">system</span><span class="p">)</span> <span class="k">if</span> <span class="n">system</span> <span class="k">else</span> <span class="p">[]</span>
+        <span class="n">system_formatted_text</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">([</span><span class="n">component</span><span class="o">.</span><span class="n">content</span> <span class="k">for</span> <span class="n">component</span> <span class="ow">in</span> <span class="n">system_formatted</span> <span class="k">if</span> <span class="n">component</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;string&#39;</span><span class="p">])</span> <span class="c1"># HACK</span>
+        
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">messages</span><span class="p">),</span> <span class="mi">2</span><span class="p">):</span>
+            <span class="n">user_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+            <span class="n">assistant_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">]</span>
+            
+            <span class="n">user_content</span> <span class="o">=</span> <span class="n">system_formatted_text</span> <span class="o">+</span> <span class="n">user_message</span><span class="p">[</span><span class="s2">&quot;content&quot;</span><span class="p">]</span> <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="n">user_message</span><span class="p">[</span><span class="s2">&quot;content&quot;</span><span class="p">]</span>
+            <span class="n">user_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">user_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">user_content</span><span class="p">)</span>
+            <span class="n">assistant_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">assistant_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">assistant_message</span><span class="p">[</span><span class="s2">&quot;content&quot;</span><span class="p">])</span>
+            
+            <span class="n">user_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">user_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+            <span class="n">assistant_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">assistant_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+            
+            <span class="n">res_all</span><span class="o">.</span><span class="n">append</span><span class="p">((</span>
+                <span class="n">user_encoded</span><span class="p">,</span> 
+                <span class="n">assistant_encoded</span>
+            <span class="p">))</span>
+            
+        <span class="k">return</span> <span class="n">res_all</span></div>
+</div>
+
+
+<div class="viewcode-block" id="Llama2ConversationTemplateForTool">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.Llama2ConversationTemplateForTool">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Llama2ConversationTemplateForTool</span><span class="p">(</span><span class="n">Llama2ConversationTemplate</span><span class="p">):</span>    
+<div class="viewcode-block" id="Llama2ConversationTemplateForTool._encode">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.Llama2ConversationTemplateForTool._encode">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_encode</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizer</span><span class="p">,</span>
+        <span class="n">messages</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]],</span>
+        <span class="n">system</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">tools</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+        <span class="k">if</span> <span class="n">tools</span><span class="p">:</span>
+            <span class="c1"># logger.warning(&quot;Formatted tools are not supported in Llama2, thus tools will be ignored. &quot;</span>
+            <span class="c1">#                &quot;If this is intended, please include tools in the system message manually.&quot;)</span>
+            <span class="n">system</span> <span class="o">=</span> <span class="n">system</span> <span class="o">+</span> <span class="n">tools</span>
+        <span class="n">res_all</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="n">system_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">system_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">system</span><span class="p">)</span> <span class="k">if</span> <span class="n">system</span> <span class="k">else</span> <span class="p">[]</span>
+        <span class="n">system_formatted_text</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">([</span><span class="n">component</span><span class="o">.</span><span class="n">content</span> <span class="k">for</span> <span class="n">component</span> <span class="ow">in</span> <span class="n">system_formatted</span> <span class="k">if</span> <span class="n">component</span><span class="o">.</span><span class="n">type</span> <span class="o">==</span> <span class="s1">&#39;string&#39;</span><span class="p">])</span> <span class="c1"># HACK</span>
+        <span class="n">ls_for_save</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">messages</span><span class="p">),</span> <span class="mi">1</span><span class="p">):</span>
+            <span class="k">if</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="n">CONVERSATION_ROLE_NAMES</span><span class="p">[</span><span class="s1">&#39;user&#39;</span><span class="p">]:</span>
+                <span class="n">user_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+                <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="n">user_content</span> <span class="o">=</span> <span class="n">system_formatted_text</span> <span class="o">+</span> <span class="n">user_message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">user_content</span> <span class="o">=</span> <span class="n">user_message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span>
+                <span class="n">user_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">user_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">user_content</span><span class="p">)</span>
+                <span class="n">user_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">user_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+                <span class="n">ls_for_save</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">user_encoded</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="n">CONVERSATION_ROLE_NAMES</span><span class="p">[</span><span class="s1">&#39;function&#39;</span><span class="p">]:</span>
+                <span class="n">function_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+                <span class="n">function_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">assistant_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">function_message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">])</span>
+                <span class="n">function_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">function_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+                <span class="n">ls_for_save</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">function_encoded</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="n">CONVERSATION_ROLE_NAMES</span><span class="p">[</span><span class="s1">&#39;observation&#39;</span><span class="p">]:</span>
+                <span class="n">observation_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+                <span class="n">observation_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">user_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">observation_message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">])</span>
+                <span class="n">observation_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">observation_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+                <span class="n">ls_for_save</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">observation_encoded</span><span class="p">)</span>
+            <span class="k">elif</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="n">CONVERSATION_ROLE_NAMES</span><span class="p">[</span><span class="s1">&#39;assistant&#39;</span><span class="p">]:</span>
+                <span class="n">assistant_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+                <span class="n">assistant_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">assistant_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">assistant_message</span><span class="p">[</span><span class="s2">&quot;content&quot;</span><span class="p">])</span>
+                <span class="n">assistant_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">assistant_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+                <span class="n">ls_for_save</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">assistant_encoded</span><span class="p">)</span>
+                <span class="c1"># res_tuple = (ls_for_save[0], ls_for_save[1], ls_for_save[2], ls_for_save[3])</span>
+                <span class="n">res_all</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="nb">tuple</span><span class="p">(</span><span class="n">ls_for_save</span><span class="p">))</span>
+                <span class="n">ls_for_save</span> <span class="o">=</span> <span class="p">[]</span>
+        
+        <span class="k">if</span> <span class="n">ls_for_save</span><span class="p">:</span>
+            <span class="n">res_all</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="nb">tuple</span><span class="p">(</span><span class="n">ls_for_save</span><span class="p">))</span>
+
+        <span class="k">return</span> <span class="n">res_all</span></div>
+</div>
+
+
+<div class="viewcode-block" id="LLAMA3_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.LLAMA3_TEMPLATE">[docs]</a>
+<span class="n">LLAMA3_TEMPLATE</span> <span class="o">=</span> <span class="n">ConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;llama3&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|start_header_id|&gt;user&lt;|end_header_id|&gt;</span><span class="se">\n\n</span><span class="s1">{{content}}&lt;|eot_id|&gt;&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|start_header_id|&gt;assistant&lt;|end_header_id|&gt;</span><span class="se">\n\n</span><span class="s1">{{content}}&lt;|eot_id|&gt;&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|start_header_id|&gt;system&lt;|end_header_id|&gt;</span><span class="se">\n\n</span><span class="s1">{{content}}&lt;|eot_id|&gt;&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">special_starter</span><span class="o">=</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;bos_token&#39;</span><span class="p">)</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="LLAMA3_TEMPLATE_FOR_TOOL">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.LLAMA3_TEMPLATE_FOR_TOOL">[docs]</a>
+<span class="n">LLAMA3_TEMPLATE_FOR_TOOL</span> <span class="o">=</span> <span class="n">ConversationTemplateForTool</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;llama3_for_tool&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|start_header_id|&gt;user&lt;|end_header_id|&gt;</span><span class="se">\n\n</span><span class="s1">{{content}}&lt;|eot_id|&gt;&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">function_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|start_header_id|&gt;assistant&lt;|end_header_id|&gt;</span><span class="se">\n\n</span><span class="s1">{{content}}&lt;|eot_id|&gt;&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">observation_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|start_header_id|&gt;tool&lt;|end_header_id|&gt;</span><span class="se">\n\n</span><span class="s1">{{content}}&lt;|eot_id|&gt;&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|start_header_id|&gt;assistant&lt;|end_header_id|&gt;</span><span class="se">\n\n</span><span class="s1">{{content}}&lt;|eot_id|&gt;&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|start_header_id|&gt;system&lt;|end_header_id|&gt;</span><span class="se">\n\n</span><span class="s1">{{content}}&lt;|eot_id|&gt;&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">special_starter</span><span class="o">=</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;bos_token&#39;</span><span class="p">)</span>
+<span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="LLAMA2_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.LLAMA2_TEMPLATE">[docs]</a>
+<span class="n">LLAMA2_TEMPLATE</span> <span class="o">=</span> <span class="n">Llama2ConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;llama2&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;bos_token&#39;</span><span class="p">),</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;[INST] {{content}} [/INST]&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;{{content}}&#39;</span><span class="p">),</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;eos_token&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;&lt;SYS&gt;&gt;</span><span class="se">\n</span><span class="s1">{{content}}</span><span class="se">\n</span><span class="s1">&lt;&lt;/SYS&gt;&gt;</span><span class="se">\n\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">)</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="LLAMA2_TEMPLATE_FOR_TOOL">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.LLAMA2_TEMPLATE_FOR_TOOL">[docs]</a>
+<span class="n">LLAMA2_TEMPLATE_FOR_TOOL</span> <span class="o">=</span> <span class="n">Llama2ConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;llama2_for_tool&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;bos_token&#39;</span><span class="p">),</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;[INST] {{content}} [/INST]&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;{{content}}&#39;</span><span class="p">),</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;eos_token&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;&lt;SYS&gt;&gt;</span><span class="se">\n</span><span class="s1">{{content}}</span><span class="se">\n</span><span class="s1">&lt;&lt;/SYS&gt;&gt;</span><span class="se">\n\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">)</span>
+<span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/conversation_template/phi.html b/_modules/lmflow/utils/conversation_template/phi.html
new file mode 100644
index 000000000..a44e3ed71
--- /dev/null
+++ b/_modules/lmflow/utils/conversation_template/phi.html
@@ -0,0 +1,473 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.conversation_template.phi &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/conversation_template/phi';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../conversation_template.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.phi</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.conversation_template.phi</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.base</span><span class="w"> </span><span class="kn">import</span> <span class="n">StringFormatter</span><span class="p">,</span> <span class="n">TemplateComponent</span><span class="p">,</span> <span class="n">ConversationTemplate</span>
+
+
+<div class="viewcode-block" id="PHI3_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/phi/index.html#lmflow.utils.conversation_template.PHI3_TEMPLATE">[docs]</a>
+<span class="n">PHI3_TEMPLATE</span> <span class="o">=</span> <span class="n">ConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;phi3&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|user|&gt;</span><span class="se">\n</span><span class="s1">{{content}}&lt;|end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|assistant|&gt;</span><span class="se">\n</span><span class="s1">{{content}}&lt;|end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|system|&gt;</span><span class="se">\n</span><span class="s1">{{content}}&lt;|end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">special_starter</span><span class="o">=</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;bos_token&#39;</span><span class="p">),</span>
+    <span class="n">special_stopper</span><span class="o">=</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;eos_token&#39;</span><span class="p">)</span>
+<span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/conversation_template/qwen.html b/_modules/lmflow/utils/conversation_template/qwen.html
new file mode 100644
index 000000000..3c1943cb5
--- /dev/null
+++ b/_modules/lmflow/utils/conversation_template/qwen.html
@@ -0,0 +1,505 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.conversation_template.qwen &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/conversation_template/qwen';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../conversation_template.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.qwen</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.conversation_template.qwen</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.base</span><span class="w"> </span><span class="kn">import</span> <span class="n">StringFormatter</span><span class="p">,</span> <span class="n">TemplateComponent</span><span class="p">,</span> <span class="n">ConversationTemplate</span><span class="p">,</span> <span class="n">ConversationTemplateForTool</span>
+
+
+<div class="viewcode-block" id="QWEN2_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/qwen/index.html#lmflow.utils.conversation_template.QWEN2_TEMPLATE">[docs]</a>
+<span class="n">QWEN2_TEMPLATE</span> <span class="o">=</span> <span class="n">ConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;qwen2&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;user</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;assistant</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;system</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">separator</span><span class="o">=</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="QWEN2_TEMPLATE_FOR_TOOL">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/qwen/index.html#lmflow.utils.conversation_template.QWEN2_TEMPLATE_FOR_TOOL">[docs]</a>
+<span class="n">QWEN2_TEMPLATE_FOR_TOOL</span> <span class="o">=</span> <span class="n">ConversationTemplateForTool</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;qwen2_for_tool&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;user</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">function_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;assistant</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">observation_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;tool</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;assistant</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;system</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">separator</span><span class="o">=</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+<span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/conversation_template/yi.html b/_modules/lmflow/utils/conversation_template/yi.html
new file mode 100644
index 000000000..b275bce1b
--- /dev/null
+++ b/_modules/lmflow/utils/conversation_template/yi.html
@@ -0,0 +1,471 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.conversation_template.yi &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/conversation_template/yi';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../conversation_template.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.yi</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.conversation_template.yi</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.base</span><span class="w"> </span><span class="kn">import</span> <span class="n">StringFormatter</span><span class="p">,</span> <span class="n">TemplateComponent</span><span class="p">,</span> <span class="n">ConversationTemplate</span>
+
+
+<div class="viewcode-block" id="YI1_5_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/yi/index.html#lmflow.utils.conversation_template.YI1_5_TEMPLATE">[docs]</a>
+<span class="n">YI1_5_TEMPLATE</span> <span class="o">=</span> <span class="n">ConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;yi1_5&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;user</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|im_start|&gt;assistant</span><span class="se">\n</span><span class="s1">{{content}}&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;{{content}}&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">)</span>
+<span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/conversation_template/zephyr.html b/_modules/lmflow/utils/conversation_template/zephyr.html
new file mode 100644
index 000000000..e96be2f4b
--- /dev/null
+++ b/_modules/lmflow/utils/conversation_template/zephyr.html
@@ -0,0 +1,529 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.conversation_template.zephyr &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/conversation_template/zephyr';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../conversation_template.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.zephyr</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.conversation_template.zephyr</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Set</span><span class="p">,</span> <span class="n">Sequence</span><span class="p">,</span> <span class="n">Literal</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">PreTrainedTokenizer</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">.base</span><span class="w"> </span><span class="kn">import</span> <span class="n">StringFormatter</span><span class="p">,</span> <span class="n">TemplateComponent</span><span class="p">,</span> <span class="n">ConversationTemplate</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/zephyr/index.html#lmflow.utils.conversation_template.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="ZephyrConversationTemplate">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/zephyr/index.html#lmflow.utils.conversation_template.ZephyrConversationTemplate">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">ZephyrConversationTemplate</span><span class="p">(</span><span class="n">ConversationTemplate</span><span class="p">):</span>    
+<div class="viewcode-block" id="ZephyrConversationTemplate._encode">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/zephyr/index.html#lmflow.utils.conversation_template.ZephyrConversationTemplate._encode">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">_encode</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">tokenizer</span><span class="p">:</span> <span class="n">PreTrainedTokenizer</span><span class="p">,</span>
+        <span class="n">messages</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]],</span>
+        <span class="n">system</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">tools</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="o">**</span><span class="n">kwargs</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Sequence</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]:</span>
+        <span class="c1"># TODO: truncation according to model max length</span>
+        <span class="c1"># TODO: make sure the last few tokens are &quot;learnable&quot;, not masked with token_id = -100.</span>
+        
+        <span class="n">res_all</span> <span class="o">=</span> <span class="p">[]</span>
+        
+        <span class="n">system_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">system_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">system</span><span class="p">)</span> <span class="k">if</span> <span class="n">system</span> <span class="k">else</span> <span class="p">[]</span>
+        <span class="n">system_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">system_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+        
+        <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">messages</span><span class="p">),</span> <span class="mi">2</span><span class="p">):</span>
+            <span class="n">user_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span><span class="p">]</span>
+            <span class="n">assistant_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">]</span>
+            
+            <span class="n">user_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">user_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">user_message</span><span class="p">[</span><span class="s2">&quot;content&quot;</span><span class="p">])</span>
+            <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">system</span><span class="p">:</span>
+                <span class="c1"># when system is not provided, the first user message should not start with a newline</span>
+                <span class="n">user_formatted</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">content</span> <span class="o">=</span> <span class="n">user_formatted</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">content</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">,</span> <span class="s1">&#39;&#39;</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+            <span class="n">assistant_formatted</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">assistant_formatter</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">content</span><span class="o">=</span><span class="n">assistant_message</span><span class="p">[</span><span class="s2">&quot;content&quot;</span><span class="p">])</span>
+            
+            <span class="n">user_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">user_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+            <span class="n">assistant_encoded</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_encode_template</span><span class="p">(</span><span class="n">assistant_formatted</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
+            
+            <span class="n">res_all</span><span class="o">.</span><span class="n">append</span><span class="p">((</span>
+                <span class="n">system_encoded</span> <span class="o">+</span> <span class="n">user_encoded</span> <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="mi">0</span> <span class="k">else</span> <span class="n">user_encoded</span><span class="p">,</span> 
+                <span class="n">assistant_encoded</span>
+            <span class="p">))</span>
+            
+        <span class="k">return</span> <span class="n">res_all</span></div>
+</div>
+
+    
+    
+<div class="viewcode-block" id="ZEPHYR_TEMPLATE">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/conversation_template/zephyr/index.html#lmflow.utils.conversation_template.ZEPHYR_TEMPLATE">[docs]</a>
+<span class="n">ZEPHYR_TEMPLATE</span> <span class="o">=</span> <span class="n">ZephyrConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;zephyr&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&lt;|user|&gt;</span><span class="se">\n</span><span class="s1">{{content}}&#39;</span><span class="p">),</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;eos_token&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&lt;|assistant|&gt;</span><span class="se">\n</span><span class="s1">{{content}}&#39;</span><span class="p">),</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;eos_token&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;&lt;|system|&gt;</span><span class="se">\n</span><span class="s1">{{content}}&#39;</span><span class="p">),</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;eos_token&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">)</span>
+<span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/data_utils.html b/_modules/lmflow/utils/data_utils.html
new file mode 100644
index 000000000..ae2cb3959
--- /dev/null
+++ b/_modules/lmflow/utils/data_utils.html
@@ -0,0 +1,711 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.data_utils &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/data_utils';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.data_utils</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.data_utils</h1><div class="highlight"><pre>
+<span></span><span class="sd">&quot;&quot;&quot;The program includes several functions: setting a random seed, </span>
+<span class="sd">loading data from a JSON file, batching data, and extracting answers from generated text.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">random</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">json</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">re</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Union</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">TypedDict</span><span class="p">,</span> <span class="n">Dict</span>
+
+
+<div class="viewcode-block" id="set_random_seed">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.set_random_seed">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">set_random_seed</span><span class="p">(</span><span class="n">seed</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Set the random seed for `random`, `numpy`, `torch`, `torch.cuda`.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+<span class="sd">    seed : int</span>
+<span class="sd">        The default seed.</span>
+<span class="sd">        </span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">random</span><span class="o">.</span><span class="n">seed</span><span class="p">(</span><span class="n">seed</span><span class="p">)</span>
+    <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">seed</span><span class="p">(</span><span class="n">seed</span><span class="p">)</span>
+    <span class="n">torch</span><span class="o">.</span><span class="n">manual_seed</span><span class="p">(</span><span class="n">seed</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">is_available</span><span class="p">():</span>
+        <span class="n">torch</span><span class="o">.</span><span class="n">cuda</span><span class="o">.</span><span class="n">manual_seed_all</span><span class="p">(</span><span class="n">seed</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="load_data">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.load_data">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">load_data</span><span class="p">(</span><span class="n">file_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Load data with file name.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+<span class="sd">    file_name : str.</span>
+<span class="sd">        The dataset file name.</span>
+<span class="sd">    </span>
+<span class="sd">    Returns</span>
+<span class="sd">    ------------</span>
+<span class="sd">    inputs : list.</span>
+<span class="sd">        The input texts of the dataset.</span>
+<span class="sd">    outputs : list.</span>
+<span class="sd">        The output texts file datasets.    </span>
+<span class="sd">    len : int.</span>
+<span class="sd">        The length of the dataset.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">inputs</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="n">outputs</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="nb">type</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
+    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">file_name</span><span class="p">,</span> <span class="n">encoding</span><span class="o">=</span><span class="s1">&#39;utf-8&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+        <span class="n">json_data</span> <span class="o">=</span> <span class="n">json</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">f</span><span class="p">)</span>
+        <span class="nb">type</span> <span class="o">=</span> <span class="n">json_data</span><span class="p">[</span><span class="s2">&quot;type&quot;</span><span class="p">]</span>
+        <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">json_data</span><span class="p">[</span><span class="s2">&quot;instances&quot;</span><span class="p">]:</span>
+            <span class="n">inputs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">line</span><span class="p">[</span><span class="s2">&quot;input&quot;</span><span class="p">])</span>
+            <span class="n">outputs</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">line</span><span class="p">[</span><span class="s2">&quot;output&quot;</span><span class="p">])</span>
+
+    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;load dataset </span><span class="si">{</span><span class="n">file_name</span><span class="si">}</span><span class="s2"> success.</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="nb">print</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Type : </span><span class="si">{</span><span class="nb">type</span><span class="si">}</span><span class="s2">, datasize : </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">outputs</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">inputs</span><span class="p">,</span> <span class="n">outputs</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">outputs</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="batchlize">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.batchlize">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">batchlize</span><span class="p">(</span><span class="n">examples</span><span class="p">:</span> <span class="nb">list</span><span class="p">,</span> <span class="n">batch_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">random_shuffle</span><span class="p">:</span> <span class="nb">bool</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Convert examples to a dataloader.</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+<span class="sd">    examples : list.</span>
+<span class="sd">        Data list.</span>
+<span class="sd">    batch_size : int.</span>
+
+<span class="sd">    random_shuffle : bool</span>
+<span class="sd">        If true, the dataloader shuffle the training data.</span>
+<span class="sd">    </span>
+<span class="sd">    Returns</span>
+<span class="sd">    ------------</span>
+<span class="sd">    dataloader:</span>
+<span class="sd">        Dataloader with batch generator.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+    <span class="n">size</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="n">dataloader</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="n">length</span> <span class="o">=</span> <span class="nb">len</span><span class="p">(</span><span class="n">examples</span><span class="p">)</span>
+    <span class="k">if</span> <span class="p">(</span><span class="n">random_shuffle</span><span class="p">):</span>
+        <span class="n">random</span><span class="o">.</span><span class="n">shuffle</span><span class="p">(</span><span class="n">examples</span><span class="p">)</span>
+    <span class="k">while</span> <span class="n">size</span> <span class="o">&lt;</span> <span class="n">length</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">length</span> <span class="o">-</span> <span class="n">size</span> <span class="o">&gt;</span> <span class="n">batch_size</span><span class="p">:</span>
+            <span class="n">dataloader</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">examples</span><span class="p">[</span><span class="n">size</span> <span class="p">:</span> <span class="n">size</span><span class="o">+</span><span class="n">batch_size</span><span class="p">])</span>
+            <span class="n">size</span> <span class="o">+=</span> <span class="n">batch_size</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">dataloader</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">examples</span><span class="p">[</span><span class="n">size</span> <span class="p">:</span> <span class="n">size</span><span class="o">+</span><span class="p">(</span><span class="n">length</span><span class="o">-</span><span class="n">size</span><span class="p">)])</span>
+            <span class="n">size</span> <span class="o">+=</span> <span class="p">(</span><span class="n">length</span> <span class="o">-</span> <span class="n">size</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">dataloader</span></div>
+
+
+
+
+<div class="viewcode-block" id="answer_extraction">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.answer_extraction">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">answer_extraction</span><span class="p">(</span><span class="n">response</span><span class="p">,</span> <span class="n">answer_type</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>   <span class="c1">#use this funtion to extract answers from generated text</span>
+
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Use this funtion to extract answers from generated text</span>
+
+<span class="sd">    Parameters</span>
+<span class="sd">    ------------</span>
+<span class="sd">    args : </span>
+<span class="sd">        Arguments.</span>
+<span class="sd">    response : str</span>
+<span class="sd">        plain string response.</span>
+
+
+<span class="sd">    Returns</span>
+<span class="sd">    ------------</span>
+<span class="sd">    answer:</span>
+<span class="sd">        Decoded answer (such as A, B, C, D, E for mutiple-choice QA).</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="c1"># temp = response[&quot;generated_text&quot;]</span>
+    <span class="n">temp</span> <span class="o">=</span> <span class="n">response</span>
+    <span class="k">if</span> <span class="n">answer_type</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;gsm8k&quot;</span><span class="p">,</span> <span class="s2">&quot;svamp&quot;</span><span class="p">,</span> <span class="s2">&quot;asdiv&quot;</span><span class="p">,</span> <span class="s2">&quot;addsub&quot;</span><span class="p">,</span> <span class="s2">&quot;singleeq&quot;</span><span class="p">,</span> <span class="s2">&quot;multiarith&quot;</span><span class="p">,</span> <span class="s2">&quot;math&quot;</span><span class="p">):</span>
+        <span class="n">temp</span> <span class="o">=</span> <span class="n">temp</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;,&quot;</span><span class="p">,</span> <span class="s2">&quot;&quot;</span><span class="p">)</span>
+        <span class="n">temp</span> <span class="o">=</span> <span class="p">[</span><span class="n">s</span> <span class="k">for</span> <span class="n">s</span> <span class="ow">in</span> <span class="n">re</span><span class="o">.</span><span class="n">findall</span><span class="p">(</span><span class="sa">r</span><span class="s1">&#39;-?\d+\.?\d*&#39;</span><span class="p">,</span> <span class="n">temp</span><span class="p">)]</span>
+    <span class="k">elif</span> <span class="n">answer_type</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;aqua&quot;</span><span class="p">,</span> <span class="s2">&quot;csqa&quot;</span><span class="p">,</span> <span class="s2">&quot;multiple_choice&quot;</span><span class="p">):</span>
+        <span class="n">temp</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">findall</span><span class="p">(</span><span class="sa">r</span><span class="s1">&#39;A|B|C|D|E&#39;</span><span class="p">,</span> <span class="n">temp</span><span class="p">)</span>
+    <span class="k">elif</span> <span class="n">answer_type</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;strategyqa&quot;</span><span class="p">,</span> <span class="s2">&quot;coin_flip&quot;</span><span class="p">):</span>
+        <span class="n">temp</span> <span class="o">=</span> <span class="n">temp</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span>
+        <span class="n">temp</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">sub</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\&quot;</span><span class="s2">|</span><span class="se">\&#39;</span><span class="s2">|</span><span class="se">\n</span><span class="s2">|\.|\s|\:|\,&quot;</span><span class="p">,</span><span class="s2">&quot; &quot;</span><span class="p">,</span> <span class="n">temp</span><span class="p">)</span>
+        <span class="n">temp</span> <span class="o">=</span> <span class="n">temp</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot; &quot;</span><span class="p">)</span>
+        <span class="n">temp</span> <span class="o">=</span> <span class="p">[</span><span class="n">i</span> <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="n">temp</span> <span class="k">if</span> <span class="n">i</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;yes&quot;</span><span class="p">,</span> <span class="s2">&quot;no&quot;</span><span class="p">)]</span>
+    <span class="k">elif</span> <span class="n">answer_type</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;last_letters&quot;</span><span class="p">):</span>
+        <span class="n">temp</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">sub</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\&quot;</span><span class="s2">|</span><span class="se">\&#39;</span><span class="s2">|</span><span class="se">\n</span><span class="s2">|\.|\s&quot;</span><span class="p">,</span><span class="s2">&quot;&quot;</span><span class="p">,</span> <span class="n">temp</span><span class="p">)</span>
+        <span class="n">temp</span> <span class="o">=</span> <span class="p">[</span><span class="n">temp</span><span class="p">]</span>
+    <span class="k">elif</span> <span class="n">answer_type</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;pubmedqa&quot;</span><span class="p">,</span> <span class="s2">&quot;binary_choice&quot;</span><span class="p">):</span>
+        <span class="c1"># pattern = &quot;Output: (yes|no|maybe)&quot;</span>
+        <span class="c1"># sttr = re.search(pattern, temp)</span>
+        <span class="c1"># answer = sttr.group(0)[8:] if sttr is not None else &quot;N/A&quot;</span>
+        <span class="n">pattern</span> <span class="o">=</span> <span class="s2">&quot;(answer|Answer|ANSWER|output|Output|OUTPUT|A): \(*(yes|Yes|YES|no|No|NO|maybe|Maybe|MAYBE)&quot;</span>
+        <span class="n">sttr</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">search</span><span class="p">(</span><span class="n">pattern</span><span class="p">,</span> <span class="n">temp</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">sttr</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">mid_answer</span> <span class="o">=</span> <span class="n">sttr</span><span class="o">.</span><span class="n">group</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">mid_answer</span> <span class="o">=</span> <span class="n">mid_answer</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;:&quot;</span><span class="p">)[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span>
+            <span class="n">answer</span> <span class="o">=</span> <span class="n">mid_answer</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">pattern</span> <span class="o">=</span> <span class="s2">&quot;(yes|Yes|YES|no|No|NO|maybe|Maybe|MAYBE)(\.|\s)&quot;</span>
+            <span class="n">sttr</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">search</span><span class="p">(</span><span class="n">pattern</span><span class="p">,</span> <span class="n">temp</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">sttr</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="n">answer</span> <span class="o">=</span> <span class="n">sttr</span><span class="o">.</span><span class="n">group</span><span class="p">(</span><span class="mi">0</span><span class="p">)[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">answer</span> <span class="o">=</span> <span class="s2">&quot;N/A&quot;</span>
+        <span class="k">return</span> <span class="n">answer</span>
+    <span class="k">elif</span> <span class="n">answer_type</span> <span class="o">==</span> <span class="s2">&quot;medmcqa&quot;</span><span class="p">:</span>
+        <span class="c1"># pattern = &quot;Output: (A|B|C|D).&quot;</span>
+        <span class="c1"># sttr = re.search(pattern, temp)</span>
+        <span class="c1"># answer = sttr.group(0)[8:-1].lower() if sttr is not None else &quot;N/A&quot;</span>
+        <span class="n">pattern</span> <span class="o">=</span> <span class="s2">&quot;(answer|Answer|ANSWER|output|Output|OUTPUT|A): \(*(A|B|C|D|a|b|c|d)&quot;</span>
+        <span class="n">sttr</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">search</span><span class="p">(</span><span class="n">pattern</span><span class="p">,</span> <span class="n">temp</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">sttr</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">mid_answer</span> <span class="o">=</span> <span class="n">sttr</span><span class="o">.</span><span class="n">group</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">answer</span> <span class="o">=</span> <span class="n">mid_answer</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">pattern</span> <span class="o">=</span> <span class="s2">&quot;\(*(A|B|C|D|a|b|c|d)\)*(\.|\s)&quot;</span>
+            <span class="n">sttr</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">search</span><span class="p">(</span><span class="n">pattern</span><span class="p">,</span> <span class="n">temp</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">sttr</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="k">if</span> <span class="s1">&#39;(&#39;</span> <span class="ow">in</span> <span class="n">sttr</span><span class="o">.</span><span class="n">group</span><span class="p">(</span><span class="mi">0</span><span class="p">):</span>
+                    <span class="n">answer</span> <span class="o">=</span> <span class="n">sttr</span><span class="o">.</span><span class="n">group</span><span class="p">(</span><span class="mi">0</span><span class="p">)[</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">answer</span> <span class="o">=</span> <span class="n">sttr</span><span class="o">.</span><span class="n">group</span><span class="p">(</span><span class="mi">0</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">answer</span> <span class="o">=</span> <span class="s2">&quot;N/A&quot;</span>
+        <span class="k">return</span> <span class="n">answer</span>
+
+    <span class="k">elif</span> <span class="n">answer_type</span> <span class="o">==</span> <span class="s2">&quot;usmle&quot;</span><span class="p">:</span>
+        <span class="c1"># pattern = &quot;Output: (A|B|C|D).&quot;</span>
+        <span class="c1"># sttr = re.search(pattern, temp)</span>
+        <span class="c1"># answer = sttr.group(0)[8:-1].lower() if sttr is not None else &quot;N/A&quot;</span>
+        <span class="n">pattern</span> <span class="o">=</span> <span class="s2">&quot;(Answer|Output|A): \(*(A|B|C|D|a|b|c|d)&quot;</span>
+        <span class="n">sttr</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">search</span><span class="p">(</span><span class="n">pattern</span><span class="p">,</span> <span class="n">temp</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">sttr</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">mid_answer</span> <span class="o">=</span> <span class="n">sttr</span><span class="o">.</span><span class="n">group</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+            <span class="n">answer</span> <span class="o">=</span> <span class="n">mid_answer</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">pattern</span> <span class="o">=</span> <span class="s2">&quot;\(*(A|B|C|D|a|b|c|d)\)*(\.|\s)&quot;</span>
+            <span class="n">sttr</span> <span class="o">=</span> <span class="n">re</span><span class="o">.</span><span class="n">search</span><span class="p">(</span><span class="n">pattern</span><span class="p">,</span> <span class="n">temp</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">sttr</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="k">if</span> <span class="s1">&#39;(&#39;</span> <span class="ow">in</span> <span class="n">sttr</span><span class="o">.</span><span class="n">group</span><span class="p">(</span><span class="mi">0</span><span class="p">):</span>
+                    <span class="n">answer</span> <span class="o">=</span> <span class="n">sttr</span><span class="o">.</span><span class="n">group</span><span class="p">(</span><span class="mi">0</span><span class="p">)[</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">answer</span> <span class="o">=</span> <span class="n">sttr</span><span class="o">.</span><span class="n">group</span><span class="p">(</span><span class="mi">0</span><span class="p">)[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">lower</span><span class="p">()</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">answer</span> <span class="o">=</span> <span class="s2">&quot;N/A&quot;</span>
+        <span class="k">return</span> <span class="n">answer</span>
+    <span class="k">elif</span> <span class="n">answer_type</span> <span class="o">==</span> <span class="s2">&quot;text&quot;</span><span class="p">:</span>
+        <span class="k">return</span> <span class="n">response</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unsupported answer type: </span><span class="si">{</span><span class="n">answer_type</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">temp</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">answer</span> <span class="o">=</span> <span class="n">temp</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+        <span class="c1"># if there is . at the end of answer, remove it</span>
+        <span class="c1"># e.g. answer = 64.</span>
+        <span class="k">if</span> <span class="n">answer</span> <span class="o">!=</span> <span class="s2">&quot;&quot;</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">answer</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">==</span> <span class="s2">&quot;.&quot;</span><span class="p">:</span>
+                <span class="n">answer</span> <span class="o">=</span> <span class="n">answer</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+
+            <span class="c1"># round the answer to nearest integer</span>
+        <span class="k">if</span> <span class="n">answer_type</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;gsm8k&quot;</span><span class="p">,</span> <span class="s2">&quot;svamp&quot;</span><span class="p">):</span>
+            <span class="k">try</span><span class="p">:</span>
+                <span class="n">answer</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="nb">round</span><span class="p">(</span><span class="nb">float</span><span class="p">(</span><span class="n">answer</span><span class="p">)))</span>
+            <span class="k">except</span><span class="p">:</span>
+                <span class="n">answer</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span> <span class="c1"># no sol or sol doesn&#39;t have valid format</span>
+        <span class="k">elif</span> <span class="n">answer_type</span> <span class="ow">in</span> <span class="p">(</span><span class="s2">&quot;last_letters&quot;</span><span class="p">):</span>
+            <span class="k">try</span><span class="p">:</span>
+                <span class="n">answer</span> <span class="o">=</span> <span class="n">answer</span><span class="p">[</span><span class="o">-</span><span class="n">args</span><span class="o">.</span><span class="n">concat_length</span><span class="p">:]</span>
+            <span class="k">except</span><span class="p">:</span>
+                <span class="n">answer</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">answer</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
+    <span class="k">return</span> <span class="n">answer</span></div>
+
+
+
+<div class="viewcode-block" id="process_image_flag">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.process_image_flag">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">process_image_flag</span><span class="p">(</span><span class="n">text</span><span class="p">,</span> <span class="n">image_flag</span><span class="o">=</span><span class="s2">&quot;&lt;ImageHere&gt;&quot;</span><span class="p">):</span>
+    <span class="n">texts</span> <span class="o">=</span> <span class="n">text</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">image_flag</span><span class="p">)</span>
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">texts</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">image_token_indexes</span> <span class="o">=</span> <span class="p">[</span><span class="nb">len</span><span class="p">(</span><span class="n">text</span><span class="p">)</span> <span class="k">for</span> <span class="n">text</span> <span class="ow">in</span> <span class="n">texts</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]]</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">image_token_indexes</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="c1"># cumsun</span>
+    <span class="n">image_token_indexes</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">cumsum</span><span class="p">(</span><span class="n">image_token_indexes</span><span class="p">))</span>
+    <span class="n">texts</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">texts</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">texts</span><span class="p">,</span> <span class="n">image_token_indexes</span></div>
+
+
+
+<div class="viewcode-block" id="VLLMInferenceResultWithInput">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.VLLMInferenceResultWithInput">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">VLLMInferenceResultWithInput</span><span class="p">(</span><span class="n">TypedDict</span><span class="p">):</span>
+<div class="viewcode-block" id="VLLMInferenceResultWithInput.input">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.VLLMInferenceResultWithInput.input">[docs]</a>
+    <span class="nb">input</span><span class="p">:</span> <span class="nb">str</span></div>
+
+<div class="viewcode-block" id="VLLMInferenceResultWithInput.output">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.VLLMInferenceResultWithInput.output">[docs]</a>
+    <span class="n">output</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]</span></div>
+</div>
+
+    
+
+<div class="viewcode-block" id="RewardModelInferenceResultWithInput">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.RewardModelInferenceResultWithInput">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">RewardModelInferenceResultWithInput</span><span class="p">(</span><span class="n">TypedDict</span><span class="p">):</span>
+<div class="viewcode-block" id="RewardModelInferenceResultWithInput.input">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.RewardModelInferenceResultWithInput.input">[docs]</a>
+    <span class="nb">input</span><span class="p">:</span> <span class="nb">str</span></div>
+
+<div class="viewcode-block" id="RewardModelInferenceResultWithInput.output">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.RewardModelInferenceResultWithInput.output">[docs]</a>
+    <span class="n">output</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">Union</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">float</span><span class="p">]]]</span> <span class="c1"># [{&quot;score&quot;: 0.5, &quot;text&quot;: &quot;output text&quot;}]</span></div>
+</div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/flash_attention/bloom_flash_attention.html b/_modules/lmflow/utils/flash_attention/bloom_flash_attention.html
new file mode 100644
index 000000000..8d74c66b0
--- /dev/null
+++ b/_modules/lmflow/utils/flash_attention/bloom_flash_attention.html
@@ -0,0 +1,552 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.flash_attention.bloom_flash_attention &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/flash_attention/bloom_flash_attention';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.flash_attention.bloom_flash_attention</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.flash_attention.bloom_flash_attention</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch</span><span class="w"> </span><span class="kn">import</span> <span class="n">nn</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch.nn.functional</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">F</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">transformers</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.models.bloom.modeling_bloom</span><span class="w"> </span><span class="kn">import</span> <span class="n">dropout_add</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">einops</span><span class="w"> </span><span class="kn">import</span> <span class="n">rearrange</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">.triton_flash_attention</span><span class="w"> </span><span class="kn">import</span> <span class="n">flash_attn_qkvpacked_func</span>
+
+<div class="viewcode-block" id="forward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.html#lmflow.utils.flash_attention.bloom_flash_attention.forward">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">hidden_states</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+        <span class="n">residual</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+        <span class="n">alibi</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+        <span class="n">attention_mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+        <span class="n">layer_past</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">head_mask</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">use_cache</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">output_attentions</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">):</span>  
+        <span class="n">dtype</span> <span class="o">=</span> <span class="n">hidden_states</span><span class="o">.</span><span class="n">dtype</span>
+        <span class="n">fused_qkv</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">query_key_value</span><span class="p">(</span><span class="n">hidden_states</span><span class="p">)</span>  <span class="c1"># [batch_size, seq_length, 3 x hidden_size]</span>
+
+        <span class="c1"># 3 x [batch_size, seq_length, num_heads, head_dim]</span>
+        <span class="p">(</span><span class="n">query_layer</span><span class="p">,</span> <span class="n">key_layer</span><span class="p">,</span> <span class="n">value_layer</span><span class="p">)</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split_heads</span><span class="p">(</span><span class="n">fused_qkv</span><span class="p">)</span>
+
+        <span class="n">batch_size</span><span class="p">,</span> <span class="n">q_length</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">query_layer</span><span class="o">.</span><span class="n">shape</span>
+        <span class="n">bsz</span><span class="p">,</span> <span class="n">q_len</span> <span class="o">=</span> <span class="n">batch_size</span><span class="p">,</span> <span class="n">q_length</span>
+
+        <span class="k">if</span> <span class="n">layer_past</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">past_key</span><span class="p">,</span> <span class="n">past_value</span> <span class="o">=</span> <span class="n">layer_past</span>
+            <span class="c1"># concatenate along seq_length dimension:</span>
+            <span class="c1">#  - key: [batch_size * self.num_heads, head_dim, kv_length]</span>
+            <span class="c1">#  - value: [batch_size * self.num_heads, kv_length, head_dim]</span>
+            <span class="n">key_layer</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">past_key</span><span class="p">,</span> <span class="n">key_layer</span><span class="p">),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+            <span class="n">value_layer</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">past_value</span><span class="p">,</span> <span class="n">value_layer</span><span class="p">),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="n">use_cache</span> <span class="ow">is</span> <span class="kc">True</span><span class="p">:</span>
+            <span class="n">present</span> <span class="o">=</span> <span class="p">(</span><span class="n">key_layer</span><span class="p">,</span> <span class="n">value_layer</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">present</span> <span class="o">=</span> <span class="kc">None</span>
+
+        <span class="n">reshaped_alibi</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span><span class="n">alibi</span><span class="p">,</span> <span class="s1">&#39;(b h) one s-&gt; b h one s&#39;</span><span class="p">,</span> <span class="n">h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">)</span>
+        <span class="n">reshaped_alibi</span> <span class="o">=</span> <span class="n">reshaped_alibi</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">beta</span>
+        
+        <span class="n">attention_mask</span> <span class="o">=</span> <span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">attention_mask</span><span class="p">)</span>
+        <span class="n">attention_mask</span> <span class="o">=</span> <span class="n">attention_mask</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:]</span><span class="o">.</span><span class="n">bool</span><span class="p">()</span>
+        <span class="n">reshaped_alibi_masked</span> <span class="o">=</span> <span class="n">reshaped_alibi</span><span class="o">.</span><span class="n">masked_fill</span><span class="p">(</span><span class="n">attention_mask</span><span class="p">,</span> <span class="o">-</span><span class="mf">1e9</span><span class="p">)</span>
+
+        <span class="n">reshaped_query_layer</span> <span class="o">=</span> <span class="n">query_layer</span>
+        <span class="n">reshaped_key_layer</span> <span class="o">=</span> <span class="n">key_layer</span>
+        <span class="n">reshaped_value_layer</span> <span class="o">=</span> <span class="n">value_layer</span>
+
+        <span class="n">qkv</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">concat</span><span class="p">([</span><span class="n">reshaped_query_layer</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">reshaped_key_layer</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">reshaped_value_layer</span><span class="o">.</span><span class="n">unsqueeze</span><span class="p">(</span><span class="mi">2</span><span class="p">)],</span> <span class="n">dim</span> <span class="o">=</span> <span class="mi">2</span><span class="p">)</span>
+
+        <span class="n">output</span> <span class="o">=</span> <span class="n">flash_attn_qkvpacked_func</span><span class="p">(</span>
+                <span class="n">qkv</span><span class="p">,</span> <span class="n">reshaped_alibi_masked</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">inv_norm_factor</span>
+            <span class="p">)</span>
+
+        <span class="n">output</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="s1">&#39;b s h d -&gt; (b h) s d&#39;</span><span class="p">)</span>
+        
+        <span class="c1"># change view [batch_size, num_heads, q_length, head_dim]</span>
+        <span class="n">context_layer</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_merge_heads</span><span class="p">(</span><span class="n">output</span><span class="p">)</span>
+
+        <span class="c1"># aggregate results across tp ranks. See here: https://github.com/pytorch/pytorch/issues/76232</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">pretraining_tp</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">slow_but_exact</span><span class="p">:</span>
+            <span class="n">slices</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">hidden_size</span> <span class="o">/</span> <span class="bp">self</span><span class="o">.</span><span class="n">pretraining_tp</span>
+            <span class="n">output_tensor</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">zeros_like</span><span class="p">(</span><span class="n">context_layer</span><span class="p">)</span>
+            <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">pretraining_tp</span><span class="p">):</span>
+                <span class="n">output_tensor</span> <span class="o">=</span> <span class="n">output_tensor</span> <span class="o">+</span> <span class="n">F</span><span class="o">.</span><span class="n">linear</span><span class="p">(</span>
+                    <span class="n">context_layer</span><span class="p">[:,</span> <span class="p">:,</span> <span class="nb">int</span><span class="p">(</span><span class="n">i</span> <span class="o">*</span> <span class="n">slices</span><span class="p">)</span> <span class="p">:</span> <span class="nb">int</span><span class="p">((</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="n">slices</span><span class="p">)],</span>
+                    <span class="bp">self</span><span class="o">.</span><span class="n">dense</span><span class="o">.</span><span class="n">weight</span><span class="p">[:,</span> <span class="nb">int</span><span class="p">(</span><span class="n">i</span> <span class="o">*</span> <span class="n">slices</span><span class="p">)</span> <span class="p">:</span> <span class="nb">int</span><span class="p">((</span><span class="n">i</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="n">slices</span><span class="p">)],</span>
+                <span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">output_tensor</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dense</span><span class="p">(</span><span class="n">context_layer</span><span class="p">)</span>
+
+        <span class="n">output_tensor</span> <span class="o">=</span> <span class="n">dropout_add</span><span class="p">(</span><span class="n">output_tensor</span><span class="p">,</span> <span class="n">residual</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">hidden_dropout</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span><span class="p">)</span>
+        
+        <span class="n">outputs</span> <span class="o">=</span> <span class="p">(</span><span class="n">output_tensor</span><span class="p">,</span> <span class="n">present</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">output_attentions</span><span class="p">:</span>
+            <span class="n">outputs</span> <span class="o">+=</span> <span class="p">(</span><span class="n">context_layer</span><span class="p">,)</span>
+
+        <span class="k">return</span> <span class="n">outputs</span></div>
+
+
+
+<span class="c1"># Disable the transformation of the attention mask in LlamaModel as the flash attention</span>
+<span class="c1"># requires the attention mask to be the same as the key_padding_mask</span>
+<div class="viewcode-block" id="_prepare_attn_mask">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.html#lmflow.utils.flash_attention.bloom_flash_attention._prepare_attn_mask">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_prepare_attn_mask</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span> <span class="n">attention_mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">input_shape</span><span class="p">:</span> <span class="n">Tuple</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">],</span> <span class="n">past_key_values_length</span><span class="p">:</span> <span class="nb">int</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">torch</span><span class="o">.</span><span class="n">BoolTensor</span><span class="p">:</span>
+
+        <span class="k">return</span> <span class="n">attention_mask</span></div>
+
+
+<div class="viewcode-block" id="replace_bloom_attn_with_flash_attn">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.html#lmflow.utils.flash_attention.bloom_flash_attention.replace_bloom_attn_with_flash_attn">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">replace_bloom_attn_with_flash_attn</span><span class="p">():</span>
+    <span class="n">transformers</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">bloom</span><span class="o">.</span><span class="n">modeling_bloom</span><span class="o">.</span><span class="n">BloomModel</span><span class="o">.</span><span class="n">_prepare_attn_mask</span> <span class="o">=</span> <span class="p">(</span>
+        <span class="n">_prepare_attn_mask</span>
+    <span class="p">)</span>
+    <span class="n">transformers</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">bloom</span><span class="o">.</span><span class="n">modeling_bloom</span><span class="o">.</span><span class="n">BloomAttention</span><span class="o">.</span><span class="n">forward</span> <span class="o">=</span> <span class="n">forward</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/flash_attention/gpt2_flash_attention.html b/_modules/lmflow/utils/flash_attention/gpt2_flash_attention.html
new file mode 100644
index 000000000..9875a88f7
--- /dev/null
+++ b/_modules/lmflow/utils/flash_attention/gpt2_flash_attention.html
@@ -0,0 +1,586 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.flash_attention.gpt2_flash_attention &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/flash_attention/gpt2_flash_attention';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.flash_attention.gpt2_flash_attention</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.flash_attention.gpt2_flash_attention</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch</span><span class="w"> </span><span class="kn">import</span> <span class="n">nn</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">transformers</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.models.llama.modeling_llama</span><span class="w"> </span><span class="kn">import</span> <span class="n">apply_rotary_pos_emb</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">einops</span><span class="w"> </span><span class="kn">import</span> <span class="n">rearrange</span>
+
+<span class="c1">#try to import flash_attn 2.x.x, if not, import flash_attn 1.x.x</span>
+<span class="k">try</span><span class="p">:</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">flash_attn.flash_attn_interface</span><span class="w"> </span><span class="kn">import</span> <span class="n">flash_attn_varlen_qkvpacked_func</span> <span class="k">as</span> <span class="n">flash_attn_unpadded_qkvpacked_func</span>
+<span class="k">except</span><span class="p">:</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">flash_attn.flash_attn_interface</span><span class="w"> </span><span class="kn">import</span> <span class="n">flash_attn_unpadded_qkvpacked_func</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">flash_attn.bert_padding</span><span class="w"> </span><span class="kn">import</span> <span class="n">unpad_input</span><span class="p">,</span> <span class="n">pad_input</span>
+
+
+<div class="viewcode-block" id="forward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.html#lmflow.utils.flash_attention.gpt2_flash_attention.forward">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">hidden_states</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">]],</span>
+        <span class="n">layer_past</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">attention_mask</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">head_mask</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">encoder_hidden_states</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">encoder_attention_mask</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">FloatTensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">use_cache</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+        <span class="n">output_attentions</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">bool</span><span class="p">]</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">Union</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]],</span> <span class="o">...</span><span class="p">]:</span>
+    
+
+        <span class="k">if</span> <span class="n">encoder_hidden_states</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">if</span> <span class="ow">not</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="s2">&quot;q_attn&quot;</span><span class="p">):</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="s2">&quot;If class is used as cross attention, the weights `q_attn` have to be defined. &quot;</span>
+                    <span class="s2">&quot;Please make sure to instantiate class with `GPT2Attention(..., is_cross_attention=True)`.&quot;</span>
+                <span class="p">)</span>
+
+            <span class="n">query</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">q_attn</span><span class="p">(</span><span class="n">hidden_states</span><span class="p">)</span>
+            <span class="n">key</span><span class="p">,</span> <span class="n">value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">c_attn</span><span class="p">(</span><span class="n">encoder_hidden_states</span><span class="p">)</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">split_size</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+            <span class="n">attention_mask</span> <span class="o">=</span> <span class="n">encoder_attention_mask</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">query</span><span class="p">,</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">c_attn</span><span class="p">(</span><span class="n">hidden_states</span><span class="p">)</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">split_size</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+        <span class="n">bsz</span><span class="p">,</span> <span class="n">q_len</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">hidden_states</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
+
+        <span class="n">query</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split_heads</span><span class="p">(</span><span class="n">query</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">)</span>
+        <span class="n">key</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split_heads</span><span class="p">(</span><span class="n">key</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">)</span>
+        <span class="n">value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split_heads</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">)</span>
+
+        <span class="c1">#TODO Should we support?</span>
+        <span class="k">if</span> <span class="n">layer_past</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">past_key</span><span class="p">,</span> <span class="n">past_value</span> <span class="o">=</span> <span class="n">layer_past</span>
+            <span class="n">key</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">past_key</span><span class="p">,</span> <span class="n">key</span><span class="p">),</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">2</span><span class="p">)</span>
+            <span class="n">value</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">past_value</span><span class="p">,</span> <span class="n">value</span><span class="p">),</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">2</span><span class="p">)</span>
+
+        <span class="k">assert</span> <span class="n">use_cache</span> <span class="ow">is</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;Use cache is not supported&quot;</span>
+        <span class="n">present</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="c1"># if use_cache is True:</span>
+        <span class="c1">#     present = (key, value)</span>
+        <span class="c1"># else:</span>
+        <span class="c1">#     present = None</span>
+
+        <span class="k">assert</span> <span class="bp">self</span><span class="o">.</span><span class="n">reorder_and_upcast_attn</span> <span class="ow">is</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;reorder_and_upcast_attn is not supported yet&quot;</span>
+
+        <span class="n">qkv</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">([</span><span class="n">query</span><span class="p">,</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span><span class="p">],</span> <span class="n">dim</span> <span class="o">=</span> <span class="mi">2</span><span class="p">)</span>
+        <span class="n">qkv</span> <span class="o">=</span> <span class="n">qkv</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>   <span class="c1"># [bsz, seq_len, 3, heads, hiddens_per_head]</span>
+        
+        <span class="c1"># breakpoint()</span>
+        <span class="n">key_padding_mask</span> <span class="o">=</span> <span class="n">attention_mask</span>
+        <span class="c1"># key_padding_mask = None</span>
+        <span class="c1"># breakpoint()</span>
+        <span class="k">if</span> <span class="n">key_padding_mask</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">qkv</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span><span class="n">qkv</span><span class="p">,</span> <span class="s2">&quot;b s ... -&gt; (b s) ...&quot;</span><span class="p">)</span>
+            <span class="n">max_s</span> <span class="o">=</span> <span class="n">q_len</span>
+            <span class="n">cu_q_lens</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span>
+                <span class="mi">0</span><span class="p">,</span> <span class="p">(</span><span class="n">bsz</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="n">q_len</span><span class="p">,</span> <span class="n">step</span><span class="o">=</span><span class="n">q_len</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">int32</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">qkv</span><span class="o">.</span><span class="n">device</span>
+            <span class="p">)</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="n">flash_attn_unpadded_qkvpacked_func</span><span class="p">(</span>
+                <span class="n">qkv</span><span class="p">,</span> <span class="n">cu_q_lens</span><span class="p">,</span> <span class="n">max_s</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="kc">True</span>
+            <span class="p">)</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="s2">&quot;(b s) ... -&gt; b s ...&quot;</span><span class="p">,</span> <span class="n">b</span><span class="o">=</span><span class="n">bsz</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># flip in flash attention</span>
+            <span class="n">key_padding_mask</span> <span class="o">=</span> <span class="n">key_padding_mask</span><span class="o">.</span><span class="n">clone</span><span class="p">()</span>
+            <span class="n">key_padding_mask</span> <span class="o">=</span> <span class="p">(</span><span class="mf">1.0</span> <span class="o">-</span> <span class="n">key_padding_mask</span><span class="p">)</span>
+            <span class="n">key_padding_mask</span> <span class="o">=</span> <span class="n">key_padding_mask</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">squeeze</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+            <span class="n">nheads</span> <span class="o">=</span> <span class="n">qkv</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span>
+            <span class="n">x</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span><span class="n">qkv</span><span class="p">,</span> <span class="s2">&quot;b s three h d -&gt; b s (three h d)&quot;</span><span class="p">)</span>
+            <span class="n">x_unpad</span><span class="p">,</span> <span class="n">indices</span><span class="p">,</span> <span class="n">cu_q_lens</span><span class="p">,</span> <span class="n">max_s</span> <span class="o">=</span> <span class="n">unpad_input</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">key_padding_mask</span><span class="p">)</span>
+            <span class="n">x_unpad</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span>
+                <span class="n">x_unpad</span><span class="p">,</span> <span class="s2">&quot;nnz (three h d) -&gt; nnz three h d&quot;</span><span class="p">,</span> <span class="n">three</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">h</span><span class="o">=</span><span class="n">nheads</span>
+            <span class="p">)</span>
+            <span class="n">output_unpad</span> <span class="o">=</span> <span class="n">flash_attn_unpadded_qkvpacked_func</span><span class="p">(</span>
+                <span class="n">x_unpad</span><span class="p">,</span> <span class="n">cu_q_lens</span><span class="p">,</span> <span class="n">max_s</span><span class="p">,</span> <span class="mf">0.0</span><span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="kc">True</span>
+            <span class="p">)</span>
+            <span class="n">output</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span>
+                <span class="n">pad_input</span><span class="p">(</span>
+                    <span class="n">rearrange</span><span class="p">(</span><span class="n">output_unpad</span><span class="p">,</span> <span class="s2">&quot;nnz h d -&gt; nnz (h d)&quot;</span><span class="p">),</span> <span class="n">indices</span><span class="p">,</span> <span class="n">bsz</span><span class="p">,</span> <span class="n">q_len</span>
+                <span class="p">),</span>
+                <span class="s2">&quot;b s (h d) -&gt; b s h d&quot;</span><span class="p">,</span>
+                <span class="n">h</span><span class="o">=</span><span class="n">nheads</span><span class="p">,</span>
+            <span class="p">)</span>
+        <span class="c1"># if self.reorder_and_upcast_attn:</span>
+        <span class="c1">#     attn_output, attn_weights = self._upcast_and_reordered_attn(query, key, value, attention_mask, head_mask)</span>
+        <span class="c1"># else:</span>
+        <span class="c1">#     attn_output, attn_weights = self._attn(query, key, value, attention_mask, head_mask)</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="s1">&#39;b s h d -&gt; b h s d&#39;</span><span class="p">)</span>
+        <span class="n">attn_output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_merge_heads</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">)</span>
+        <span class="n">attn_output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">c_proj</span><span class="p">(</span><span class="n">attn_output</span><span class="p">)</span>
+        <span class="n">attn_output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">resid_dropout</span><span class="p">(</span><span class="n">attn_output</span><span class="p">)</span>
+
+        <span class="n">outputs</span> <span class="o">=</span> <span class="p">(</span><span class="n">attn_output</span><span class="p">,</span> <span class="n">present</span><span class="p">)</span>
+        
+        <span class="k">assert</span> <span class="n">output_attentions</span> <span class="ow">is</span> <span class="kc">False</span><span class="p">,</span> <span class="s2">&quot;output attentions is not supported yet&quot;</span>
+        <span class="c1"># if output_attentions:</span>
+        <span class="c1">#     outputs += (attn_weights,)</span>
+
+        <span class="k">return</span> <span class="n">outputs</span>  <span class="c1"># a, present, (attentions)</span></div>
+
+
+
+<span class="c1"># Disable the transformation of the attention mask in LlamaModel as the flash attention</span>
+<span class="c1"># requires the attention mask to be the same as the key_padding_mask</span>
+<div class="viewcode-block" id="_prepare_decoder_attention_mask">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.html#lmflow.utils.flash_attention.gpt2_flash_attention._prepare_decoder_attention_mask">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_prepare_decoder_attention_mask</span><span class="p">(</span>
+    <span class="bp">self</span><span class="p">,</span> <span class="n">attention_mask</span><span class="p">,</span> <span class="n">input_shape</span><span class="p">,</span> <span class="n">inputs_embeds</span><span class="p">,</span> <span class="n">past_key_values_length</span>
+<span class="p">):</span>
+    <span class="c1"># [bsz, seq_len]</span>
+    <span class="k">return</span> <span class="n">attention_mask</span></div>
+
+
+
+<div class="viewcode-block" id="replace_gpt2_attn_with_flash_attn">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.html#lmflow.utils.flash_attention.gpt2_flash_attention.replace_gpt2_attn_with_flash_attn">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">replace_gpt2_attn_with_flash_attn</span><span class="p">():</span>
+    <span class="c1"># transformers.models.gpt2.modeling_gpt2.LlamaModel._prepare_decoder_attention_mask = (</span>
+    <span class="c1">#     _prepare_decoder_attention_mask</span>
+    <span class="c1"># )</span>
+    <span class="n">transformers</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">gpt2</span><span class="o">.</span><span class="n">modeling_gpt2</span><span class="o">.</span><span class="n">GPT2Attention</span><span class="o">.</span><span class="n">forward</span> <span class="o">=</span> <span class="n">forward</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/flash_attention/gpt_neo_flash_attention.html b/_modules/lmflow/utils/flash_attention/gpt_neo_flash_attention.html
new file mode 100644
index 000000000..3e895ee2f
--- /dev/null
+++ b/_modules/lmflow/utils/flash_attention/gpt_neo_flash_attention.html
@@ -0,0 +1,550 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.flash_attention.gpt_neo_flash_attention &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/flash_attention/gpt_neo_flash_attention';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.flash_attention.gpt_neo_flash_attention</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.flash_attention.gpt_neo_flash_attention</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">transformers</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">einops</span><span class="w"> </span><span class="kn">import</span> <span class="n">rearrange</span>
+
+<span class="c1">#try to import flash_attn 2.x.x, if not, import flash_attn 1.x.x</span>
+<span class="k">try</span><span class="p">:</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">flash_attn.flash_attn_interface</span><span class="w"> </span><span class="kn">import</span> <span class="n">flash_attn_varlen_qkvpacked_func</span> <span class="k">as</span> <span class="n">flash_attn_unpadded_qkvpacked_func</span>
+<span class="k">except</span><span class="p">:</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">flash_attn.flash_attn_interface</span><span class="w"> </span><span class="kn">import</span> <span class="n">flash_attn_unpadded_qkvpacked_func</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">flash_attn.bert_padding</span><span class="w"> </span><span class="kn">import</span> <span class="n">unpad_input</span><span class="p">,</span> <span class="n">pad_input</span>
+
+<div class="viewcode-block" id="_attn">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html#lmflow.utils.flash_attention.gpt_neo_flash_attention._attn">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_attn</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">query</span><span class="p">,</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span><span class="p">,</span> <span class="n">attention_mask</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">head_mask</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+    <span class="c1"># (batch, head, seq_length, head_features)</span>
+    <span class="n">query</span> <span class="o">=</span> <span class="n">query</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">bfloat16</span><span class="p">)</span>
+    <span class="n">key</span> <span class="o">=</span> <span class="n">key</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">bfloat16</span><span class="p">)</span>
+    <span class="n">query</span> <span class="o">=</span> <span class="n">query</span> <span class="o">*</span> <span class="n">torch</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">tensor</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">))</span>
+    <span class="n">qkv</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">stack</span><span class="p">(</span>
+        <span class="p">[</span><span class="n">query</span><span class="p">,</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">2</span>
+    <span class="p">)</span><span class="c1"># [bsz, nh, 3, t, hd]</span>
+    <span class="n">qkv</span> <span class="o">=</span> <span class="n">qkv</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span><span class="mi">3</span><span class="p">)</span><span class="c1">## [bsz, q_len, 3, nh, hd]</span>
+    <span class="n">bsz</span> <span class="o">=</span> <span class="n">qkv</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+    <span class="n">q_len</span> <span class="o">=</span> <span class="n">qkv</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+    
+    <span class="n">attention_mask</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">attention_mask</span> <span class="o">==</span> <span class="o">-</span><span class="mf">0.0</span><span class="p">,</span> <span class="kc">True</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
+    <span class="n">key_padding_mask</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span><span class="n">attention_mask</span><span class="p">,</span> <span class="s2">&quot;b () () s -&gt; b s&quot;</span><span class="p">)</span> <span class="k">if</span> <span class="n">attention_mask</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="kc">None</span>
+    <span class="k">if</span> <span class="n">key_padding_mask</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">qkv</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span><span class="n">qkv</span><span class="p">,</span> <span class="s2">&quot;b s ... -&gt; (b s) ...&quot;</span><span class="p">)</span>
+        <span class="n">max_s</span> <span class="o">=</span> <span class="n">q_len</span>
+        <span class="n">cu_q_lens</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span>
+            <span class="mi">0</span><span class="p">,</span> <span class="p">(</span><span class="n">bsz</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="n">q_len</span><span class="p">,</span> <span class="n">step</span><span class="o">=</span><span class="n">q_len</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">int32</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">qkv</span><span class="o">.</span><span class="n">device</span>
+        <span class="p">)</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">flash_attn_unpadded_qkvpacked_func</span><span class="p">(</span>
+            <span class="n">qkv</span><span class="p">,</span> <span class="n">cu_q_lens</span><span class="p">,</span> <span class="n">max_s</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">attn_dropout</span><span class="o">.</span><span class="n">p</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span> <span class="k">else</span> <span class="mf">0.0</span> <span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="kc">True</span>
+        <span class="p">)</span><span class="c1"># attention compute</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span><span class="n">output</span><span class="p">,</span> <span class="s2">&quot;(b s) ... -&gt; b s ...&quot;</span><span class="p">,</span> <span class="n">b</span><span class="o">=</span><span class="n">bsz</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">nheads</span> <span class="o">=</span> <span class="n">qkv</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span>
+        <span class="n">x</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span><span class="n">qkv</span><span class="p">,</span> <span class="s2">&quot;b s three h d -&gt; b s (three h d)&quot;</span><span class="p">)</span>
+        <span class="n">x_unpad</span><span class="p">,</span> <span class="n">indices</span><span class="p">,</span> <span class="n">cu_q_lens</span><span class="p">,</span> <span class="n">max_s</span> <span class="o">=</span> <span class="n">unpad_input</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">key_padding_mask</span><span class="p">)</span>
+        <span class="n">x_unpad</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span>
+            <span class="n">x_unpad</span><span class="p">,</span> <span class="s2">&quot;nnz (three h d) -&gt; nnz three h d&quot;</span><span class="p">,</span> <span class="n">three</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">h</span><span class="o">=</span><span class="n">nheads</span>
+        <span class="p">)</span>
+        <span class="n">output_unpad</span> <span class="o">=</span> <span class="n">flash_attn_unpadded_qkvpacked_func</span><span class="p">(</span>
+            <span class="n">x_unpad</span><span class="p">,</span> <span class="n">cu_q_lens</span><span class="p">,</span> <span class="n">max_s</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">attn_dropout</span><span class="o">.</span><span class="n">p</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">training</span> <span class="k">else</span> <span class="mf">0.0</span><span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="kc">True</span>
+        <span class="p">)</span>
+        <span class="n">output</span> <span class="o">=</span> <span class="n">rearrange</span><span class="p">(</span>
+            <span class="n">pad_input</span><span class="p">(</span>
+                <span class="n">rearrange</span><span class="p">(</span><span class="n">output_unpad</span><span class="p">,</span> <span class="s2">&quot;nnz h d -&gt; nnz (h d)&quot;</span><span class="p">),</span> <span class="n">indices</span><span class="p">,</span> <span class="n">bsz</span><span class="p">,</span> <span class="n">q_len</span>
+            <span class="p">),</span>
+            <span class="s2">&quot;b s (h d) -&gt; b s h d&quot;</span><span class="p">,</span>
+            <span class="n">h</span><span class="o">=</span><span class="n">nheads</span><span class="p">,</span>
+        <span class="p">)</span>
+    
+    <span class="k">return</span> <span class="n">output</span><span class="p">,</span> <span class="kc">None</span></div>
+
+
+<div class="viewcode-block" id="forward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html#lmflow.utils.flash_attention.gpt_neo_flash_attention.forward">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span>
+        <span class="bp">self</span><span class="p">,</span>
+        <span class="n">hidden_states</span><span class="p">,</span>
+        <span class="n">attention_mask</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">layer_past</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">head_mask</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">use_cache</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+        <span class="n">output_attentions</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+    <span class="p">):</span>
+    
+    <span class="k">assert</span> <span class="n">head_mask</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">,</span> <span class="s2">&quot;head_mask is not supported&quot;</span>
+    <span class="k">assert</span> <span class="ow">not</span> <span class="n">output_attentions</span><span class="p">,</span> <span class="s2">&quot;output_attentions is not supported&quot;</span>
+    <span class="k">assert</span> <span class="ow">not</span> <span class="n">use_cache</span><span class="p">,</span> <span class="s2">&quot;use_cache is not supported&quot;</span>
+    
+    <span class="n">query</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">q_proj</span><span class="p">(</span><span class="n">hidden_states</span><span class="p">)</span>
+    <span class="n">key</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">k_proj</span><span class="p">(</span><span class="n">hidden_states</span><span class="p">)</span>
+    <span class="n">value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">v_proj</span><span class="p">(</span><span class="n">hidden_states</span><span class="p">)</span>
+
+    <span class="n">query</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split_heads</span><span class="p">(</span><span class="n">query</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">)</span>
+    <span class="n">key</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split_heads</span><span class="p">(</span><span class="n">key</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">)</span>
+    <span class="n">value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_split_heads</span><span class="p">(</span><span class="n">value</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">)</span>
+    
+    <span class="k">if</span> <span class="n">layer_past</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">past_key</span> <span class="o">=</span> <span class="n">layer_past</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+        <span class="n">past_value</span> <span class="o">=</span> <span class="n">layer_past</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span>
+        <span class="n">key</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">past_key</span><span class="p">,</span> <span class="n">key</span><span class="p">),</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">2</span><span class="p">)</span>
+        <span class="n">value</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">past_value</span><span class="p">,</span> <span class="n">value</span><span class="p">),</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">2</span><span class="p">)</span>
+        
+    <span class="n">present</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="n">attn_output</span><span class="p">,</span> <span class="n">attn_weights</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_attn</span><span class="p">(</span><span class="n">query</span><span class="p">,</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span><span class="p">,</span> <span class="n">attention_mask</span><span class="p">,</span> <span class="n">head_mask</span><span class="p">)</span>
+    <span class="n">new_shape</span> <span class="o">=</span> <span class="n">attn_output</span><span class="o">.</span><span class="n">size</span><span class="p">()[:</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span> <span class="o">+</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">,)</span>
+    <span class="n">attn_output</span> <span class="o">=</span> <span class="n">attn_output</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">new_shape</span><span class="p">)</span>
+    <span class="n">attn_output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">out_proj</span><span class="p">(</span><span class="n">attn_output</span><span class="p">)</span>
+    <span class="n">attn_output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">resid_dropout</span><span class="p">(</span><span class="n">attn_output</span><span class="p">)</span>
+
+    <span class="n">outputs</span> <span class="o">=</span> <span class="p">(</span><span class="n">attn_output</span><span class="p">,</span> <span class="n">present</span><span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">outputs</span>  <span class="c1"># a, present, (attentions)</span></div>
+
+        
+<div class="viewcode-block" id="replace_gpt_neo_attn_with_flash_attn">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html#lmflow.utils.flash_attention.gpt_neo_flash_attention.replace_gpt_neo_attn_with_flash_attn">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">replace_gpt_neo_attn_with_flash_attn</span><span class="p">():</span>
+    <span class="n">transformers</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">gpt_neo</span><span class="o">.</span><span class="n">modeling_gpt_neo</span><span class="o">.</span><span class="n">GPTNeoSelfAttention</span><span class="o">.</span><span class="n">_attn</span> <span class="o">=</span> <span class="n">_attn</span>
+    <span class="n">transformers</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">gpt_neo</span><span class="o">.</span><span class="n">modeling_gpt_neo</span><span class="o">.</span><span class="n">GPTNeoSelfAttention</span><span class="o">.</span><span class="n">forward</span> <span class="o">=</span> <span class="n">forward</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/flash_attention/llama_flash_attention.html b/_modules/lmflow/utils/flash_attention/llama_flash_attention.html
new file mode 100644
index 000000000..3cfc05617
--- /dev/null
+++ b/_modules/lmflow/utils/flash_attention/llama_flash_attention.html
@@ -0,0 +1,574 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.flash_attention.llama_flash_attention &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/flash_attention/llama_flash_attention';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.flash_attention.llama_flash_attention</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.flash_attention.llama_flash_attention</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Tuple</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">torch</span><span class="w"> </span><span class="kn">import</span> <span class="n">nn</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">transformers</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers.models.llama.modeling_llama</span><span class="w"> </span><span class="kn">import</span> <span class="n">apply_rotary_pos_emb</span><span class="p">,</span><span class="n">_make_causal_mask</span><span class="p">,</span><span class="n">_expand_mask</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">einops</span><span class="w"> </span><span class="kn">import</span> <span class="n">rearrange</span>
+
+<span class="c1">#try to import flash_attn 2.x.x, if not, import flash_attn 1.x.x</span>
+<span class="k">try</span><span class="p">:</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">flash_attn.flash_attn_interface</span><span class="w"> </span><span class="kn">import</span> <span class="n">flash_attn_func</span>
+<span class="k">except</span><span class="p">:</span>
+    <span class="kn">from</span><span class="w"> </span><span class="nn">flash_attn.flash_attn_interface</span><span class="w"> </span><span class="kn">import</span> <span class="n">flash_attn_unpadded_qkvpacked_func</span> <span class="k">as</span> <span class="n">flash_attn_func</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">flash_attn.bert_padding</span><span class="w"> </span><span class="kn">import</span> <span class="n">unpad_input</span><span class="p">,</span> <span class="n">pad_input</span>
+
+
+<div class="viewcode-block" id="forward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.html#lmflow.utils.flash_attention.llama_flash_attention.forward">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span>
+    <span class="bp">self</span><span class="p">,</span>
+    <span class="n">hidden_states</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
+    <span class="n">attention_mask</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">position_ids</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">LongTensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">past_key_value</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+    <span class="n">output_attentions</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+    <span class="n">use_cache</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
+<span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]]]:</span>
+    <span class="n">bsz</span><span class="p">,</span> <span class="n">q_len</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">hidden_states</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
+
+    <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">pretraining_tp</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;pretraining_tp &gt; 1 is not supported for flash attention&quot;</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">query_states</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">q_proj</span><span class="p">(</span><span class="n">hidden_states</span><span class="p">)</span>
+        <span class="n">key_states</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">k_proj</span><span class="p">(</span><span class="n">hidden_states</span><span class="p">)</span>
+        <span class="n">value_states</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">v_proj</span><span class="p">(</span><span class="n">hidden_states</span><span class="p">)</span>
+
+    <span class="n">query_states</span> <span class="o">=</span> <span class="n">query_states</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">bsz</span><span class="p">,</span> <span class="n">q_len</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">)</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="n">key_states</span> <span class="o">=</span> <span class="n">key_states</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">bsz</span><span class="p">,</span> <span class="n">q_len</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_key_value_heads</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">)</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+    <span class="n">value_states</span> <span class="o">=</span> <span class="n">value_states</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">bsz</span><span class="p">,</span> <span class="n">q_len</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_key_value_heads</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">)</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
+
+    <span class="n">kv_seq_len</span> <span class="o">=</span> <span class="n">key_states</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span>
+
+    <span class="k">if</span> <span class="n">past_key_value</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">kv_seq_len</span> <span class="o">+=</span> <span class="n">past_key_value</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="o">-</span><span class="mi">2</span><span class="p">]</span>
+    <span class="n">cos</span><span class="p">,</span> <span class="n">sin</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">rotary_emb</span><span class="p">(</span><span class="n">value_states</span><span class="p">,</span> <span class="n">seq_len</span><span class="o">=</span><span class="n">kv_seq_len</span><span class="p">)</span>
+    <span class="n">query_states</span><span class="p">,</span> <span class="n">key_states</span> <span class="o">=</span> <span class="n">apply_rotary_pos_emb</span><span class="p">(</span><span class="n">query_states</span><span class="p">,</span> <span class="n">key_states</span><span class="p">,</span> <span class="n">cos</span><span class="p">,</span> <span class="n">sin</span><span class="p">,</span> <span class="n">position_ids</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">past_key_value</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="c1"># reuse k, v, self_attention</span>
+        <span class="n">key_states</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">past_key_value</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="n">key_states</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+        <span class="n">value_states</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">past_key_value</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="n">value_states</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
+
+    <span class="n">past_key_value</span> <span class="o">=</span> <span class="p">(</span><span class="n">key_states</span><span class="p">,</span> <span class="n">value_states</span><span class="p">)</span> <span class="k">if</span> <span class="n">use_cache</span> <span class="k">else</span> <span class="kc">None</span>
+
+    <span class="n">query_states</span><span class="p">,</span> <span class="n">key_states</span><span class="p">,</span> <span class="n">value_states</span> <span class="o">=</span> <span class="p">[</span>
+        <span class="n">rearrange</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="s2">&quot;b h s d -&gt; b s h d&quot;</span><span class="p">)</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="p">[</span><span class="n">query_states</span><span class="p">,</span> <span class="n">key_states</span><span class="p">,</span> <span class="n">value_states</span><span class="p">]</span>
+    <span class="p">]</span>
+
+    <span class="n">input_dtype</span> <span class="o">=</span> <span class="n">query_states</span><span class="o">.</span><span class="n">dtype</span>
+    <span class="k">if</span> <span class="n">input_dtype</span> <span class="o">==</span> <span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">:</span>
+        <span class="c1"># Handle the case where the model is quantized</span>
+        <span class="k">if</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="p">,</span> <span class="s2">&quot;_pre_quantization_dtype&quot;</span><span class="p">):</span>
+            <span class="n">target_dtype</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">_pre_quantization_dtype</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">target_dtype</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">q_proj</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">dtype</span>
+
+        <span class="n">query_states</span> <span class="o">=</span> <span class="n">query_states</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">target_dtype</span><span class="p">)</span>
+        <span class="n">key_states</span> <span class="o">=</span> <span class="n">key_states</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">target_dtype</span><span class="p">)</span>
+        <span class="n">value_states</span> <span class="o">=</span> <span class="n">value_states</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">target_dtype</span><span class="p">)</span>
+
+    <span class="c1"># below output will have shape (batch_size, seqlen, nheads, headdim)</span>
+    <span class="n">attn_output</span> <span class="o">=</span> <span class="n">flash_attn_func</span><span class="p">(</span><span class="n">query_states</span><span class="p">,</span> <span class="n">key_states</span><span class="p">,</span> <span class="n">value_states</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">attn_output</span><span class="o">.</span><span class="n">size</span><span class="p">()</span> <span class="o">!=</span> <span class="p">(</span><span class="n">bsz</span><span class="p">,</span> <span class="n">q_len</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">):</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+            <span class="sa">f</span><span class="s2">&quot;`attn_output` should be of size </span><span class="si">{</span><span class="p">(</span><span class="n">bsz</span><span class="p">,</span><span class="w"> </span><span class="n">q_len</span><span class="p">,</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">num_heads</span><span class="p">,</span><span class="w"> </span><span class="bp">self</span><span class="o">.</span><span class="n">head_dim</span><span class="p">)</span><span class="si">}</span><span class="s2">, but is&quot;</span>
+            <span class="sa">f</span><span class="s2">&quot; </span><span class="si">{</span><span class="n">attn_output</span><span class="o">.</span><span class="n">size</span><span class="p">()</span><span class="si">}</span><span class="s2">&quot;</span>
+        <span class="p">)</span>
+
+    <span class="n">attn_output</span> <span class="o">=</span> <span class="n">attn_output</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">bsz</span><span class="p">,</span> <span class="n">q_len</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">hidden_size</span><span class="p">)</span>
+    <span class="n">attn_output</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">o_proj</span><span class="p">(</span><span class="n">attn_output</span><span class="p">)</span>
+    <span class="k">if</span> <span class="n">output_attentions</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span><span class="s2">&quot;`output_attentions` is not supported when `use_flash_attn` is True&quot;</span><span class="p">)</span>
+    <span class="n">attn_weights</span> <span class="o">=</span> <span class="kc">None</span>
+
+    <span class="k">return</span> <span class="n">attn_output</span><span class="p">,</span> <span class="n">attn_weights</span><span class="p">,</span> <span class="n">past_key_value</span></div>
+
+
+
+<span class="c1"># Disable the transformation of the attention mask in LlamaModel as the flash attention</span>
+<span class="c1"># requires the attention mask to be the same as the key_padding_mask</span>
+<div class="viewcode-block" id="_prepare_decoder_attention_mask">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.html#lmflow.utils.flash_attention.llama_flash_attention._prepare_decoder_attention_mask">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_prepare_decoder_attention_mask</span><span class="p">(</span>
+    <span class="bp">self</span><span class="p">,</span> <span class="n">attention_mask</span><span class="p">,</span> <span class="n">input_shape</span><span class="p">,</span> <span class="n">inputs_embeds</span><span class="p">,</span> <span class="n">past_key_values_length</span>
+<span class="p">):</span>
+    <span class="c1"># [bsz, seq_len]</span>
+    <span class="k">if</span> <span class="n">input_shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mi">1</span> <span class="ow">and</span> <span class="n">past_key_values_length</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>  <span class="c1"># encode</span>
+        <span class="k">return</span> <span class="n">attention_mask</span>
+    <span class="c1"># create causal mask</span>
+    <span class="c1"># [bsz, seq_len] -&gt; [bsz, 1, tgt_seq_len, src_seq_len]</span>
+    <span class="n">combined_attention_mask</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="k">if</span> <span class="n">input_shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">combined_attention_mask</span> <span class="o">=</span> <span class="n">_make_causal_mask</span><span class="p">(</span>
+            <span class="n">input_shape</span><span class="p">,</span>
+            <span class="n">inputs_embeds</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span>
+            <span class="n">device</span><span class="o">=</span><span class="n">inputs_embeds</span><span class="o">.</span><span class="n">device</span><span class="p">,</span>
+            <span class="n">past_key_values_length</span><span class="o">=</span><span class="n">past_key_values_length</span><span class="p">,</span>
+        <span class="p">)</span>
+
+    <span class="k">if</span> <span class="n">attention_mask</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="c1"># [bsz, seq_len] -&gt; [bsz, 1, tgt_seq_len, src_seq_len]</span>
+        <span class="n">expanded_attn_mask</span> <span class="o">=</span> <span class="n">_expand_mask</span><span class="p">(</span><span class="n">attention_mask</span><span class="p">,</span> <span class="n">inputs_embeds</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">tgt_len</span><span class="o">=</span><span class="n">input_shape</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">])</span><span class="o">.</span><span class="n">to</span><span class="p">(</span>
+            <span class="n">inputs_embeds</span><span class="o">.</span><span class="n">device</span>
+        <span class="p">)</span>
+        <span class="n">combined_attention_mask</span> <span class="o">=</span> <span class="p">(</span>
+            <span class="n">expanded_attn_mask</span> <span class="k">if</span> <span class="n">combined_attention_mask</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">expanded_attn_mask</span> <span class="o">+</span> <span class="n">combined_attention_mask</span>
+        <span class="p">)</span>
+
+    <span class="k">return</span> <span class="n">combined_attention_mask</span></div>
+
+
+
+<div class="viewcode-block" id="replace_llama_attn_with_flash_attn">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.html#lmflow.utils.flash_attention.llama_flash_attention.replace_llama_attn_with_flash_attn">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">replace_llama_attn_with_flash_attn</span><span class="p">():</span>
+    <span class="n">transformers</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">llama</span><span class="o">.</span><span class="n">modeling_llama</span><span class="o">.</span><span class="n">LlamaModel</span><span class="o">.</span><span class="n">_prepare_decoder_attention_mask</span> <span class="o">=</span> <span class="n">_prepare_decoder_attention_mask</span>
+    <span class="n">transformers</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">llama</span><span class="o">.</span><span class="n">modeling_llama</span><span class="o">.</span><span class="n">LlamaAttention</span><span class="o">.</span><span class="n">forward</span> <span class="o">=</span> <span class="n">forward</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/flash_attention/triton_flash_attention.html b/_modules/lmflow/utils/flash_attention/triton_flash_attention.html
new file mode 100644
index 000000000..b7cf47b66
--- /dev/null
+++ b/_modules/lmflow/utils/flash_attention/triton_flash_attention.html
@@ -0,0 +1,1333 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.flash_attention.triton_flash_attention &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/flash_attention/triton_flash_attention';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.flash_attention.triton_flash_attention</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.flash_attention.triton_flash_attention</h1><div class="highlight"><pre>
+<span></span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">*Experimental* implementation of FlashAttention in Triton.</span>
+<span class="sd">Tested with triton==2.0.0.dev20221202.</span>
+<span class="sd">Triton 2.0 has a new backend (MLIR) but seems like it doesn&#39;t yet work for head dimensions</span>
+<span class="sd">other than 64:</span>
+<span class="sd">https://github.com/openai/triton/blob/d376020f90002757eea3ea9475d4f7cfc2ec5ead/python/triton/ops/flash_attention.py#L207</span>
+<span class="sd">We&#39;ll update this implementation with the new Triton backend once this is fixed.</span>
+
+<span class="sd">We use the FlashAttention implementation from Phil Tillet a starting point.</span>
+<span class="sd">https://github.com/openai/triton/blob/master/python/tutorials/06-fused-attention.py</span>
+
+<span class="sd">Changes:</span>
+<span class="sd">- Implement both causal and non-causal attention.</span>
+<span class="sd">- Implement both self-attention and cross-attention.</span>
+<span class="sd">- Support arbitrary seqlens (not just multiples of 128), for both forward and backward.</span>
+<span class="sd">- Support all head dimensions up to 128 (not just 16, 32, 64, 128), for both forward and backward.</span>
+<span class="sd">- Support attention bias.</span>
+<span class="sd">- Speed up the forward pass a bit, and only store the LSE instead of m and l.</span>
+<span class="sd">- Make the backward for d=128 much faster by reducing register spilling.</span>
+<span class="sd">- Optionally parallelize the backward pass across seqlen_k, to deal with the case of</span>
+<span class="sd">small batch size * nheads.</span>
+
+<span class="sd">Caution:</span>
+<span class="sd">- This is an *experimental* implementation. The forward pass should be quite robust but</span>
+<span class="sd">I&#39;m not 100% sure that the backward pass doesn&#39;t have race conditions (due to the Triton compiler).</span>
+<span class="sd">- This implementation has only been tested on A100.</span>
+<span class="sd">- If you plan to use headdim other than 64 and 128, you should test for race conditions</span>
+<span class="sd">(due to the Triton compiler), as done in tests/test_flash_attn.py</span>
+<span class="sd">&quot;test_flash_attn_triton_race_condition&quot;. I&#39;ve tested and fixed many race conditions</span>
+<span class="sd">for different head dimensions (40, 48, 64, 128, 80, 88, 96), but I&#39;m still not 100% confident</span>
+<span class="sd">that there are none left for other head dimensions.</span>
+
+<span class="sd">Differences between this Triton version and the CUDA version:</span>
+<span class="sd">- Triton version doesn&#39;t support dropout.</span>
+<span class="sd">- Triton forward is generally faster than CUDA forward, while Triton backward is</span>
+<span class="sd">generally slower than CUDA backward. Overall Triton forward + backward is slightly slower</span>
+<span class="sd">than CUDA forward + backward.</span>
+<span class="sd">- Triton version doesn&#39;t support different sequence lengths in a batch (i.e., RaggedTensor/NestedTensor).</span>
+<span class="sd">- Triton version supports attention bias, while CUDA version doesn&#39;t.</span>
+<span class="sd">&quot;&quot;&quot;</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">math</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">triton</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">triton.language</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">tl</span>
+
+
+<span class="c1"># Disabling autotune for now, set num_warps=4 if headdim=64 and num_warps=8 if headdim=128</span>
+<span class="c1"># @triton.autotune(</span>
+<span class="c1">#     configs=[</span>
+<span class="c1">#         triton.Config({&quot;BLOCK_M&quot;: 128, &quot;BLOCK_N&quot;: 128}, num_warps=4, num_stages=1),</span>
+<span class="c1">#         # This config has a race condition when EVEN_M == False, disabling it for now.</span>
+<span class="c1">#         # triton.Config({&quot;BLOCK_M&quot;: 64, &quot;BLOCK_N&quot;: 64}, num_warps=4, num_stages=1),</span>
+<span class="c1">#     ],</span>
+<span class="c1">#     key=[&#39;CACHE_KEY_SEQLEN_Q&#39;, &#39;CACHE_KEY_SEQLEN_K&#39;, &#39;BIAS_TYPE&#39;, &#39;IS_CAUSAL&#39;, &#39;BLOCK_HEADDIM&#39;]</span>
+<span class="c1"># )</span>
+<span class="nd">@triton</span><span class="o">.</span><span class="n">heuristics</span><span class="p">(</span>
+    <span class="p">{</span>
+        <span class="s2">&quot;EVEN_M&quot;</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">args</span><span class="p">:</span> <span class="n">args</span><span class="p">[</span><span class="s2">&quot;seqlen_q&quot;</span><span class="p">]</span> <span class="o">%</span> <span class="n">args</span><span class="p">[</span><span class="s2">&quot;BLOCK_M&quot;</span><span class="p">]</span> <span class="o">==</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="s2">&quot;EVEN_N&quot;</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">args</span><span class="p">:</span> <span class="n">args</span><span class="p">[</span><span class="s2">&quot;seqlen_k&quot;</span><span class="p">]</span> <span class="o">%</span> <span class="n">args</span><span class="p">[</span><span class="s2">&quot;BLOCK_N&quot;</span><span class="p">]</span> <span class="o">==</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="s2">&quot;EVEN_HEADDIM&quot;</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">args</span><span class="p">:</span> <span class="n">args</span><span class="p">[</span><span class="s2">&quot;headdim&quot;</span><span class="p">]</span> <span class="o">==</span> <span class="n">args</span><span class="p">[</span><span class="s2">&quot;BLOCK_HEADDIM&quot;</span><span class="p">],</span>
+    <span class="p">}</span>
+<span class="p">)</span>
+<span class="nd">@triton</span><span class="o">.</span><span class="n">jit</span>
+<div class="viewcode-block" id="_fwd_kernel">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._fwd_kernel">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_fwd_kernel</span><span class="p">(</span>
+    <span class="n">Q</span><span class="p">,</span> <span class="n">K</span><span class="p">,</span> <span class="n">V</span><span class="p">,</span> <span class="n">Bias</span><span class="p">,</span> <span class="n">Out</span><span class="p">,</span>
+    <span class="n">Lse</span><span class="p">,</span> <span class="n">TMP</span><span class="p">,</span>  <span class="c1"># NOTE: TMP is a scratchpad buffer to workaround a compiler bug</span>
+    <span class="n">softmax_scale</span><span class="p">,</span>
+    <span class="n">stride_qb</span><span class="p">,</span> <span class="n">stride_qh</span><span class="p">,</span> <span class="n">stride_qm</span><span class="p">,</span>
+    <span class="n">stride_kb</span><span class="p">,</span> <span class="n">stride_kh</span><span class="p">,</span> <span class="n">stride_kn</span><span class="p">,</span>
+    <span class="n">stride_vb</span><span class="p">,</span> <span class="n">stride_vh</span><span class="p">,</span> <span class="n">stride_vn</span><span class="p">,</span>
+    <span class="n">stride_bb</span><span class="p">,</span> <span class="n">stride_bh</span><span class="p">,</span> <span class="n">stride_bm</span><span class="p">,</span>
+    <span class="n">stride_ob</span><span class="p">,</span> <span class="n">stride_oh</span><span class="p">,</span> <span class="n">stride_om</span><span class="p">,</span>
+    <span class="n">nheads</span><span class="p">,</span> <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">seqlen_q_rounded</span><span class="p">,</span> <span class="n">headdim</span><span class="p">,</span>
+    <span class="n">CACHE_KEY_SEQLEN_Q</span><span class="p">,</span> <span class="n">CACHE_KEY_SEQLEN_K</span><span class="p">,</span>
+    <span class="n">BIAS_TYPE</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">IS_CAUSAL</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">BLOCK_HEADDIM</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">EVEN_M</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span> <span class="n">EVEN_N</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">BLOCK_M</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span> <span class="n">BLOCK_N</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+<span class="p">):</span>
+    <span class="n">start_m</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">program_id</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+    <span class="n">off_hb</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">program_id</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">off_b</span> <span class="o">=</span> <span class="n">off_hb</span> <span class="o">//</span> <span class="n">nheads</span>
+    <span class="n">off_h</span> <span class="o">=</span> <span class="n">off_hb</span> <span class="o">%</span> <span class="n">nheads</span>
+    <span class="c1"># off_b = tl.program_id(1)</span>
+    <span class="c1"># off_h = tl.program_id(2)</span>
+    <span class="c1"># off_hb = off_b * nheads + off_h</span>
+    <span class="c1"># initialize offsets</span>
+    <span class="n">offs_m</span> <span class="o">=</span> <span class="n">start_m</span> <span class="o">*</span> <span class="n">BLOCK_M</span> <span class="o">+</span> <span class="n">tl</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">BLOCK_M</span><span class="p">)</span>
+    <span class="n">offs_n</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">BLOCK_N</span><span class="p">)</span>
+    <span class="n">offs_d</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">BLOCK_HEADDIM</span><span class="p">)</span>
+    <span class="c1"># Initialize pointers to Q, K, V</span>
+    <span class="c1"># Adding parenthesis around indexing might use int32 math instead of int64 math?</span>
+    <span class="c1"># https://github.com/openai/triton/issues/741</span>
+    <span class="c1"># I&#39;m seeing a tiny bit of difference (5-7us)</span>
+    <span class="n">q_ptrs</span> <span class="o">=</span> <span class="n">Q</span> <span class="o">+</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_qb</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_qh</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_m</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_qm</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+    <span class="n">k_ptrs</span> <span class="o">=</span> <span class="n">K</span> <span class="o">+</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_kb</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_kh</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_kn</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+    <span class="n">v_ptrs</span> <span class="o">=</span> <span class="n">V</span> <span class="o">+</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_vb</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_vh</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_vn</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+    <span class="k">if</span> <span class="n">BIAS_TYPE</span> <span class="o">==</span> <span class="s1">&#39;vector&#39;</span><span class="p">:</span>
+        <span class="n">b_ptrs</span> <span class="o">=</span> <span class="n">Bias</span> <span class="o">+</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_bb</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_bh</span> <span class="o">+</span> <span class="n">offs_n</span>
+    <span class="k">elif</span> <span class="n">BIAS_TYPE</span> <span class="o">==</span> <span class="s1">&#39;matrix&#39;</span><span class="p">:</span>
+        <span class="n">b_ptrs</span> <span class="o">=</span> <span class="n">Bias</span> <span class="o">+</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_bb</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_bh</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_m</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_bm</span> <span class="o">+</span> <span class="n">offs_n</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+    <span class="c1"># initialize pointer to m and l</span>
+    <span class="n">t_ptrs</span> <span class="o">=</span> <span class="n">TMP</span> <span class="o">+</span> <span class="n">off_hb</span> <span class="o">*</span> <span class="n">seqlen_q_rounded</span> <span class="o">+</span> <span class="n">offs_m</span>
+    <span class="n">lse_i</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">zeros</span><span class="p">([</span><span class="n">BLOCK_M</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span> <span class="o">-</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;inf&quot;</span><span class="p">)</span>
+    <span class="n">m_i</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">zeros</span><span class="p">([</span><span class="n">BLOCK_M</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span> <span class="o">-</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;inf&quot;</span><span class="p">)</span>
+    <span class="n">acc_o</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">zeros</span><span class="p">([</span><span class="n">BLOCK_M</span><span class="p">,</span> <span class="n">BLOCK_HEADDIM</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+    <span class="c1"># load q: it will stay in SRAM throughout</span>
+    <span class="c1"># [2022-10-30] TD: Triton bug - in the case of EVEN_M=True and EVEN_N=False, if we just call</span>
+    <span class="c1"># tl.load(q_ptrs), we get the wrong output!</span>
+    <span class="k">if</span> <span class="n">EVEN_M</span> <span class="o">&amp;</span> <span class="n">EVEN_N</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+            <span class="n">q</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">q_ptrs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">q</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">q_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">,</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+            <span class="n">q</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">q_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_m</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">q</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">q_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_m</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">),</span>
+                        <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+    <span class="c1"># loop over k, v and update accumulator</span>
+    <span class="n">end_n</span> <span class="o">=</span> <span class="n">seqlen_k</span> <span class="k">if</span> <span class="ow">not</span> <span class="n">IS_CAUSAL</span> <span class="k">else</span> <span class="n">tl</span><span class="o">.</span><span class="n">minimum</span><span class="p">((</span><span class="n">start_m</span> <span class="o">+</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="n">BLOCK_M</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">start_n</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">end_n</span><span class="p">,</span> <span class="n">BLOCK_N</span><span class="p">):</span>
+        <span class="n">start_n</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">multiple_of</span><span class="p">(</span><span class="n">start_n</span><span class="p">,</span> <span class="n">BLOCK_N</span><span class="p">)</span>
+        <span class="c1"># -- compute qk ----</span>
+        <span class="k">if</span> <span class="n">EVEN_N</span> <span class="o">&amp;</span> <span class="n">EVEN_M</span><span class="p">:</span>  <span class="c1"># If we just do &quot;if EVEN_N&quot;, there seems to be some race condition</span>
+            <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+                <span class="n">k</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">k_ptrs</span> <span class="o">+</span> <span class="n">start_n</span> <span class="o">*</span> <span class="n">stride_kn</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">k</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">k_ptrs</span> <span class="o">+</span> <span class="n">start_n</span> <span class="o">*</span> <span class="n">stride_kn</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">,</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+                <span class="n">k</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">k_ptrs</span> <span class="o">+</span> <span class="n">start_n</span> <span class="o">*</span> <span class="n">stride_kn</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">start_n</span> <span class="o">+</span> <span class="n">offs_n</span><span class="p">)[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">,</span>
+                            <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">k</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">k_ptrs</span> <span class="o">+</span> <span class="n">start_n</span> <span class="o">*</span> <span class="n">stride_kn</span><span class="p">,</span>
+                            <span class="n">mask</span><span class="o">=</span><span class="p">((</span><span class="n">start_n</span> <span class="o">+</span> <span class="n">offs_n</span><span class="p">)[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">),</span>
+                            <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+        <span class="n">qk</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">zeros</span><span class="p">([</span><span class="n">BLOCK_M</span><span class="p">,</span> <span class="n">BLOCK_N</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+        <span class="n">qk</span> <span class="o">+=</span> <span class="n">tl</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">trans_b</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="c1"># Trying to combine the two masks seem to make the result wrong</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">EVEN_N</span><span class="p">:</span>  <span class="c1"># Need to mask out otherwise the softmax is wrong</span>
+            <span class="n">qk</span> <span class="o">+=</span> <span class="n">tl</span><span class="o">.</span><span class="n">where</span><span class="p">((</span><span class="n">start_n</span> <span class="o">+</span> <span class="n">offs_n</span><span class="p">)[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;-inf&quot;</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">IS_CAUSAL</span><span class="p">:</span>
+            <span class="n">qk</span> <span class="o">+=</span> <span class="n">tl</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">offs_m</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&gt;=</span> <span class="p">(</span><span class="n">start_n</span> <span class="o">+</span> <span class="n">offs_n</span><span class="p">)[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:],</span> <span class="mi">0</span><span class="p">,</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;-inf&quot;</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">BIAS_TYPE</span> <span class="o">!=</span> <span class="s1">&#39;none&#39;</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">BIAS_TYPE</span> <span class="o">==</span> <span class="s1">&#39;vector&#39;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">EVEN_N</span><span class="p">:</span>
+                    <span class="n">bias</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">b_ptrs</span> <span class="o">+</span> <span class="n">start_n</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">bias</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">b_ptrs</span> <span class="o">+</span> <span class="n">start_n</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">start_n</span> <span class="o">+</span> <span class="n">offs_n</span><span class="p">)</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+                <span class="n">bias</span> <span class="o">=</span> <span class="n">bias</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span>
+            <span class="k">elif</span> <span class="n">BIAS_TYPE</span> <span class="o">==</span> <span class="s1">&#39;matrix&#39;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">EVEN_M</span> <span class="o">&amp;</span> <span class="n">EVEN_N</span><span class="p">:</span>
+                    <span class="n">bias</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">b_ptrs</span> <span class="o">+</span> <span class="n">start_n</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">bias</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">b_ptrs</span> <span class="o">+</span> <span class="n">start_n</span><span class="p">,</span>
+                                   <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_m</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">)</span>
+                                        <span class="o">&amp;</span> <span class="p">((</span><span class="n">start_n</span> <span class="o">+</span> <span class="n">offs_n</span><span class="p">)[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">),</span>
+                                   <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+            <span class="c1"># Slightly faster to multiply the softmax_scale in the tl.exp below since the compiler</span>
+            <span class="c1"># can then fuse the mult and add into an fma instruction. But if we have bias we need to</span>
+            <span class="c1"># to multiply with softmax_scale here.</span>
+            <span class="n">qk</span> <span class="o">=</span> <span class="n">qk</span> <span class="o">*</span> <span class="n">softmax_scale</span> <span class="o">+</span> <span class="n">bias</span>
+            <span class="n">m_ij</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">maximum</span><span class="p">(</span><span class="n">tl</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">qk</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span> <span class="n">lse_i</span><span class="p">)</span>
+            <span class="n">p</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">qk</span> <span class="o">-</span> <span class="n">m_ij</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">])</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">m_ij</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">maximum</span><span class="p">(</span><span class="n">tl</span><span class="o">.</span><span class="n">max</span><span class="p">(</span><span class="n">qk</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span> <span class="o">*</span> <span class="n">softmax_scale</span><span class="p">,</span> <span class="n">lse_i</span><span class="p">)</span>
+            <span class="n">p</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">qk</span> <span class="o">*</span> <span class="n">softmax_scale</span> <span class="o">-</span> <span class="n">m_ij</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">])</span>
+        <span class="n">l_ij</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
+
+        <span class="c1"># scale acc_o</span>
+        <span class="n">acc_o_scale</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">m_i</span> <span class="o">-</span> <span class="n">m_ij</span><span class="p">)</span>
+
+        <span class="c1"># # -- update output accumulator --</span>
+        <span class="c1"># BUG: have to store and immediately load</span>
+        <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">t_ptrs</span><span class="p">,</span> <span class="n">acc_o_scale</span><span class="p">)</span>
+        <span class="n">acc_o_scale</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">t_ptrs</span><span class="p">)</span>
+        <span class="n">acc_o</span> <span class="o">=</span> <span class="n">acc_o</span> <span class="o">*</span> <span class="n">acc_o_scale</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span>
+        <span class="c1"># update acc_o</span>
+        <span class="k">if</span> <span class="n">EVEN_N</span> <span class="o">&amp;</span> <span class="n">EVEN_M</span><span class="p">:</span>  <span class="c1"># If we just do &quot;if EVEN_N&quot;, there seems to be some race condition</span>
+            <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+                <span class="n">v</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">v_ptrs</span> <span class="o">+</span> <span class="n">start_n</span> <span class="o">*</span> <span class="n">stride_vn</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">v</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">v_ptrs</span> <span class="o">+</span> <span class="n">start_n</span> <span class="o">*</span> <span class="n">stride_vn</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">,</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+                <span class="n">v</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">v_ptrs</span> <span class="o">+</span> <span class="n">start_n</span> <span class="o">*</span> <span class="n">stride_vn</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">start_n</span> <span class="o">+</span> <span class="n">offs_n</span><span class="p">)[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">,</span>
+                            <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">v</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">v_ptrs</span> <span class="o">+</span> <span class="n">start_n</span> <span class="o">*</span> <span class="n">stride_vn</span><span class="p">,</span>
+                            <span class="n">mask</span><span class="o">=</span><span class="p">((</span><span class="n">start_n</span> <span class="o">+</span> <span class="n">offs_n</span><span class="p">)[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">),</span>
+                            <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+        <span class="n">p</span> <span class="o">=</span> <span class="n">p</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">v</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+        <span class="n">acc_o</span> <span class="o">+=</span> <span class="n">tl</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">p</span><span class="p">,</span> <span class="n">v</span><span class="p">)</span>
+
+        <span class="c1"># -- update statistics</span>
+        <span class="n">m_i</span> <span class="o">=</span> <span class="n">m_ij</span>
+        <span class="n">l_i_new</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">lse_i</span> <span class="o">-</span> <span class="n">m_ij</span><span class="p">)</span> <span class="o">+</span> <span class="n">l_ij</span>
+        <span class="n">lse_i</span> <span class="o">=</span> <span class="n">m_ij</span> <span class="o">+</span> <span class="n">tl</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="n">l_i_new</span><span class="p">)</span>
+
+    <span class="n">o_scale</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">m_i</span> <span class="o">-</span> <span class="n">lse_i</span><span class="p">)</span>
+    <span class="c1"># BUG: have to store and immediately load</span>
+    <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">t_ptrs</span><span class="p">,</span> <span class="n">o_scale</span><span class="p">)</span>
+    <span class="n">o_scale</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">t_ptrs</span><span class="p">)</span>
+    <span class="n">acc_o</span> <span class="o">=</span> <span class="n">acc_o</span> <span class="o">*</span> <span class="n">o_scale</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span>
+    <span class="c1"># rematerialize offsets to save registers</span>
+    <span class="n">start_m</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">program_id</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+    <span class="n">offs_m</span> <span class="o">=</span> <span class="n">start_m</span> <span class="o">*</span> <span class="n">BLOCK_M</span> <span class="o">+</span> <span class="n">tl</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">BLOCK_M</span><span class="p">)</span>
+    <span class="c1"># write back l and m</span>
+    <span class="n">lse_ptrs</span> <span class="o">=</span> <span class="n">Lse</span> <span class="o">+</span> <span class="n">off_hb</span> <span class="o">*</span> <span class="n">seqlen_q_rounded</span> <span class="o">+</span> <span class="n">offs_m</span>
+    <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">lse_ptrs</span><span class="p">,</span> <span class="n">lse_i</span><span class="p">)</span>
+    <span class="c1"># initialize pointers to output</span>
+    <span class="n">offs_d</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">BLOCK_HEADDIM</span><span class="p">)</span>
+    <span class="n">out_ptrs</span> <span class="o">=</span> <span class="n">Out</span> <span class="o">+</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_ob</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_oh</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_m</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_om</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+    <span class="k">if</span> <span class="n">EVEN_M</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">out_ptrs</span><span class="p">,</span> <span class="n">acc_o</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">out_ptrs</span><span class="p">,</span> <span class="n">acc_o</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">out_ptrs</span><span class="p">,</span> <span class="n">acc_o</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_m</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">out_ptrs</span><span class="p">,</span> <span class="n">acc_o</span><span class="p">,</span>
+                     <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_m</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">))</span></div>
+
+
+
+<span class="nd">@triton</span><span class="o">.</span><span class="n">jit</span>
+<div class="viewcode-block" id="_bwd_preprocess_do_o_dot">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._bwd_preprocess_do_o_dot">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_bwd_preprocess_do_o_dot</span><span class="p">(</span>
+    <span class="n">Out</span><span class="p">,</span> <span class="n">DO</span><span class="p">,</span> <span class="n">Delta</span><span class="p">,</span>
+    <span class="n">stride_ob</span><span class="p">,</span> <span class="n">stride_oh</span><span class="p">,</span> <span class="n">stride_om</span><span class="p">,</span>
+    <span class="n">stride_dob</span><span class="p">,</span> <span class="n">stride_doh</span><span class="p">,</span> <span class="n">stride_dom</span><span class="p">,</span>
+    <span class="n">nheads</span><span class="p">,</span> <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">seqlen_q_rounded</span><span class="p">,</span> <span class="n">headdim</span><span class="p">,</span>
+    <span class="n">BLOCK_M</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span> <span class="n">BLOCK_HEADDIM</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+<span class="p">):</span>
+    <span class="n">start_m</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">program_id</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+    <span class="n">off_hb</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">program_id</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">off_b</span> <span class="o">=</span> <span class="n">off_hb</span> <span class="o">//</span> <span class="n">nheads</span>
+    <span class="n">off_h</span> <span class="o">=</span> <span class="n">off_hb</span> <span class="o">%</span> <span class="n">nheads</span>
+    <span class="c1"># initialize offsets</span>
+    <span class="n">offs_m</span> <span class="o">=</span> <span class="n">start_m</span> <span class="o">*</span> <span class="n">BLOCK_M</span> <span class="o">+</span> <span class="n">tl</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">BLOCK_M</span><span class="p">)</span>
+    <span class="n">offs_d</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">BLOCK_HEADDIM</span><span class="p">)</span>
+    <span class="c1"># load</span>
+    <span class="n">o</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">Out</span> <span class="o">+</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_ob</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_oh</span> <span class="o">+</span> <span class="n">offs_m</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_om</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:],</span>
+                <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_m</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">),</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+    <span class="n">do</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">DO</span> <span class="o">+</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_dob</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_doh</span> <span class="o">+</span> <span class="n">offs_m</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_dom</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:],</span>
+                 <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_m</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">),</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+    <span class="n">delta</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">sum</span><span class="p">(</span><span class="n">o</span> <span class="o">*</span> <span class="n">do</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
+    <span class="c1"># write-back</span>
+    <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">Delta</span> <span class="o">+</span> <span class="n">off_hb</span> <span class="o">*</span> <span class="n">seqlen_q_rounded</span> <span class="o">+</span> <span class="n">offs_m</span><span class="p">,</span> <span class="n">delta</span><span class="p">)</span></div>
+
+
+
+<span class="nd">@triton</span><span class="o">.</span><span class="n">jit</span>
+<div class="viewcode-block" id="_bwd_store_dk_dv">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._bwd_store_dk_dv">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_bwd_store_dk_dv</span><span class="p">(</span>
+    <span class="n">dk_ptrs</span><span class="p">,</span> <span class="n">dv_ptrs</span><span class="p">,</span> <span class="n">dk</span><span class="p">,</span> <span class="n">dv</span><span class="p">,</span> <span class="n">offs_n</span><span class="p">,</span> <span class="n">offs_d</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">headdim</span><span class="p">,</span>
+    <span class="n">EVEN_M</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span> <span class="n">EVEN_N</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+<span class="p">):</span>
+    <span class="c1"># [2022-11-01] TD: Same bug. In the case of EVEN_N=True and EVEN_M=False,</span>
+    <span class="c1"># if we just call tl.store(dv_ptrs), there&#39;s a race condition</span>
+    <span class="k">if</span> <span class="n">EVEN_N</span> <span class="o">&amp;</span> <span class="n">EVEN_M</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">dv_ptrs</span><span class="p">,</span> <span class="n">dv</span><span class="p">)</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">dk_ptrs</span><span class="p">,</span> <span class="n">dk</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">dv_ptrs</span><span class="p">,</span> <span class="n">dv</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">)</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">dk_ptrs</span><span class="p">,</span> <span class="n">dk</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">dv_ptrs</span><span class="p">,</span> <span class="n">dv</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">)</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">dk_ptrs</span><span class="p">,</span> <span class="n">dk</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">dv_ptrs</span><span class="p">,</span> <span class="n">dv</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">))</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">dk_ptrs</span><span class="p">,</span> <span class="n">dk</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">))</span></div>
+
+
+
+<span class="nd">@triton</span><span class="o">.</span><span class="n">jit</span>
+<div class="viewcode-block" id="_bwd_kernel_one_col_block">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel_one_col_block">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_bwd_kernel_one_col_block</span><span class="p">(</span>
+    <span class="n">start_n</span><span class="p">,</span>
+    <span class="n">Q</span><span class="p">,</span> <span class="n">K</span><span class="p">,</span> <span class="n">V</span><span class="p">,</span> <span class="n">Bias</span><span class="p">,</span>
+    <span class="n">DO</span><span class="p">,</span> <span class="n">DQ</span><span class="p">,</span> <span class="n">DK</span><span class="p">,</span> <span class="n">DV</span><span class="p">,</span>
+    <span class="n">LSE</span><span class="p">,</span> <span class="n">D</span><span class="p">,</span>
+    <span class="n">softmax_scale</span><span class="p">,</span>
+    <span class="n">stride_qm</span><span class="p">,</span> <span class="n">stride_kn</span><span class="p">,</span> <span class="n">stride_vn</span><span class="p">,</span> <span class="n">stride_bm</span><span class="p">,</span>
+    <span class="n">stride_dom</span><span class="p">,</span> <span class="n">stride_dqm</span><span class="p">,</span> <span class="n">stride_dkn</span><span class="p">,</span> <span class="n">stride_dvn</span><span class="p">,</span>
+    <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">headdim</span><span class="p">,</span>
+    <span class="n">ATOMIC_ADD</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">BIAS_TYPE</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">IS_CAUSAL</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">BLOCK_HEADDIM</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">EVEN_M</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span> <span class="n">EVEN_N</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">BLOCK_M</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span> <span class="n">BLOCK_N</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+<span class="p">):</span>
+    <span class="c1"># We need to make sure begin_m is a multiple of BLOCK_M (not BLOCK_N)</span>
+    <span class="n">begin_m</span> <span class="o">=</span> <span class="mi">0</span> <span class="k">if</span> <span class="ow">not</span> <span class="n">IS_CAUSAL</span> <span class="k">else</span> <span class="p">((</span><span class="n">start_n</span> <span class="o">*</span> <span class="n">BLOCK_N</span><span class="p">)</span> <span class="o">//</span> <span class="n">BLOCK_M</span><span class="p">)</span> <span class="o">*</span> <span class="n">BLOCK_M</span>
+    <span class="c1"># initialize row/col offsets</span>
+    <span class="n">offs_qm</span> <span class="o">=</span> <span class="n">begin_m</span> <span class="o">+</span> <span class="n">tl</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">BLOCK_M</span><span class="p">)</span>
+    <span class="n">offs_n</span> <span class="o">=</span> <span class="n">start_n</span> <span class="o">*</span> <span class="n">BLOCK_N</span> <span class="o">+</span> <span class="n">tl</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">BLOCK_N</span><span class="p">)</span>
+    <span class="n">offs_m</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">BLOCK_M</span><span class="p">)</span>
+    <span class="n">offs_d</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">BLOCK_HEADDIM</span><span class="p">)</span>
+    <span class="c1"># initialize pointers to value-like data</span>
+    <span class="n">q_ptrs</span> <span class="o">=</span> <span class="n">Q</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_qm</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_qm</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+    <span class="n">k_ptrs</span> <span class="o">=</span> <span class="n">K</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_kn</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+    <span class="n">v_ptrs</span> <span class="o">=</span> <span class="n">V</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_vn</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+    <span class="n">do_ptrs</span> <span class="o">=</span> <span class="n">DO</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_qm</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_dom</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+    <span class="n">dq_ptrs</span> <span class="o">=</span> <span class="n">DQ</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_qm</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_dqm</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+    <span class="k">if</span> <span class="n">BIAS_TYPE</span> <span class="o">==</span> <span class="s1">&#39;vector&#39;</span><span class="p">:</span>
+        <span class="n">b_ptrs</span> <span class="o">=</span> <span class="n">Bias</span> <span class="o">+</span> <span class="n">offs_n</span>
+    <span class="k">elif</span> <span class="n">BIAS_TYPE</span> <span class="o">==</span> <span class="s1">&#39;matrix&#39;</span><span class="p">:</span>
+        <span class="n">b_ptrs</span> <span class="o">=</span> <span class="n">Bias</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_qm</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_bm</span> <span class="o">+</span> <span class="n">offs_n</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+    <span class="c1"># initialize dv and dk</span>
+    <span class="n">dv</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">zeros</span><span class="p">([</span><span class="n">BLOCK_N</span><span class="p">,</span> <span class="n">BLOCK_HEADDIM</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+    <span class="n">dk</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">zeros</span><span class="p">([</span><span class="n">BLOCK_N</span><span class="p">,</span> <span class="n">BLOCK_HEADDIM</span><span class="p">],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+    <span class="c1"># There seems to be some problem with Triton pipelining that makes results wrong for</span>
+    <span class="c1"># headdim=64, seqlen=(113, 255), bias_type=&#39;matrix&#39;. In this case the for loop</span>
+    <span class="c1"># may have zero step, and pipelining with the bias matrix could screw it up.</span>
+    <span class="c1"># So we just exit early.</span>
+    <span class="k">if</span> <span class="n">begin_m</span> <span class="o">&gt;=</span> <span class="n">seqlen_q</span><span class="p">:</span>
+        <span class="n">dv_ptrs</span> <span class="o">=</span> <span class="n">DV</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_dvn</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+        <span class="n">dk_ptrs</span> <span class="o">=</span> <span class="n">DK</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_dkn</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+        <span class="n">_bwd_store_dk_dv</span><span class="p">(</span><span class="n">dk_ptrs</span><span class="p">,</span> <span class="n">dv_ptrs</span><span class="p">,</span> <span class="n">dk</span><span class="p">,</span> <span class="n">dv</span><span class="p">,</span> <span class="n">offs_n</span><span class="p">,</span> <span class="n">offs_d</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">headdim</span><span class="p">,</span>
+                         <span class="n">EVEN_M</span><span class="o">=</span><span class="n">EVEN_M</span><span class="p">,</span> <span class="n">EVEN_N</span><span class="o">=</span><span class="n">EVEN_N</span><span class="p">,</span> <span class="n">EVEN_HEADDIM</span><span class="o">=</span><span class="n">EVEN_HEADDIM</span><span class="p">)</span>
+        <span class="k">return</span>
+    <span class="c1"># k and v stay in SRAM throughout</span>
+    <span class="c1"># [2022-10-30] TD: Same bug as the fwd. In the case of EVEN_N=True and EVEN_M=False,</span>
+    <span class="c1"># if we just call tl.load(k_ptrs), we get the wrong output!</span>
+    <span class="k">if</span> <span class="n">EVEN_N</span> <span class="o">&amp;</span> <span class="n">EVEN_M</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+            <span class="n">k</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">k_ptrs</span><span class="p">)</span>
+            <span class="n">v</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">v_ptrs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">k</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">k_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">,</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+            <span class="n">v</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">v_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">,</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+            <span class="n">k</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">k_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+            <span class="n">v</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">v_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">k</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">k_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">),</span>
+                        <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+            <span class="n">v</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">v_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">),</span>
+                        <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+    <span class="c1"># loop over rows</span>
+    <span class="n">num_block_m</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">cdiv</span><span class="p">(</span><span class="n">seqlen_q</span><span class="p">,</span> <span class="n">BLOCK_M</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">start_m</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">begin_m</span><span class="p">,</span> <span class="n">num_block_m</span> <span class="o">*</span> <span class="n">BLOCK_M</span><span class="p">,</span> <span class="n">BLOCK_M</span><span class="p">):</span>
+        <span class="n">start_m</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">multiple_of</span><span class="p">(</span><span class="n">start_m</span><span class="p">,</span> <span class="n">BLOCK_M</span><span class="p">)</span>
+        <span class="n">offs_m_curr</span> <span class="o">=</span> <span class="n">start_m</span> <span class="o">+</span> <span class="n">offs_m</span>
+        <span class="c1"># load q, k, v, do on-chip</span>
+        <span class="c1"># Same bug as below. Otherwise gives wrong result for headdim=40, seqlen=(128, 117)</span>
+        <span class="k">if</span> <span class="n">EVEN_M</span> <span class="o">&amp;</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+            <span class="n">q</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">q_ptrs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+                <span class="n">q</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">q_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_m_curr</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">q</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">q_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_m_curr</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">)</span>
+                                         <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">),</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+        <span class="c1"># recompute p = softmax(qk, dim=-1).T</span>
+        <span class="n">qk</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">trans_b</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="c1"># Trying to combine the two masks seem to make the result wrong</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">EVEN_N</span><span class="p">:</span>  <span class="c1"># Need to mask out otherwise the softmax is wrong</span>
+            <span class="n">qk</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">offs_n</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">qk</span><span class="p">,</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;-inf&quot;</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">IS_CAUSAL</span><span class="p">:</span>
+            <span class="n">qk</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">offs_m_curr</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&gt;=</span> <span class="p">(</span><span class="n">offs_n</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]),</span> <span class="n">qk</span><span class="p">,</span> <span class="nb">float</span><span class="p">(</span><span class="s2">&quot;-inf&quot;</span><span class="p">))</span>
+        <span class="k">if</span> <span class="n">BIAS_TYPE</span> <span class="o">!=</span> <span class="s1">&#39;none&#39;</span><span class="p">:</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">debug_barrier</span><span class="p">()</span>  <span class="c1"># Race condition otherwise</span>
+            <span class="k">if</span> <span class="n">BIAS_TYPE</span> <span class="o">==</span> <span class="s1">&#39;vector&#39;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">EVEN_N</span><span class="p">:</span>
+                    <span class="n">bias</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">b_ptrs</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">bias</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">b_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_n</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+                <span class="n">bias</span> <span class="o">=</span> <span class="n">bias</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span>
+            <span class="k">elif</span> <span class="n">BIAS_TYPE</span> <span class="o">==</span> <span class="s1">&#39;matrix&#39;</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">EVEN_M</span> <span class="o">&amp;</span> <span class="n">EVEN_N</span><span class="p">:</span>
+                    <span class="n">bias</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">b_ptrs</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">bias</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">b_ptrs</span><span class="p">,</span>
+                                   <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_m_curr</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">)</span>
+                                        <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_n</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">seqlen_k</span><span class="p">),</span>
+                                   <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">tl</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+            <span class="n">qk</span> <span class="o">=</span> <span class="n">qk</span> <span class="o">*</span> <span class="n">softmax_scale</span> <span class="o">+</span> <span class="n">bias</span>
+        <span class="c1"># There seems to be a race condition when headdim=48/96, and dq, dk, dv are wrong.</span>
+        <span class="c1"># Also wrong for headdim=64.</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="n">EVEN_M</span> <span class="o">&amp;</span> <span class="n">EVEN_HEADDIM</span><span class="p">):</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">debug_barrier</span><span class="p">()</span>
+        <span class="n">lse_i</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">LSE</span> <span class="o">+</span> <span class="n">offs_m_curr</span><span class="p">)</span>
+        <span class="k">if</span> <span class="n">BIAS_TYPE</span> <span class="o">==</span> <span class="s1">&#39;none&#39;</span><span class="p">:</span>
+            <span class="n">p</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">qk</span> <span class="o">*</span> <span class="n">softmax_scale</span> <span class="o">-</span> <span class="n">lse_i</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">])</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="n">p</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">exp</span><span class="p">(</span><span class="n">qk</span> <span class="o">-</span> <span class="n">lse_i</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">])</span>
+        <span class="c1"># compute dv</span>
+        <span class="c1"># [2022-10-30] TD: A Triton bug: if EVEN_M=True and EVEN_HEADDIM=False, if we call</span>
+        <span class="c1"># do = tl.load(do_ptrs, mask=offs_d[None, :] &lt; headdim, other=0.0), we get wrong outputs</span>
+        <span class="c1"># in the case of headdim=48/96, seqlen_q &amp; seqlen_k &gt;= 512. If headdim=40 or seqlen &lt; 512,</span>
+        <span class="c1"># the output is correct.</span>
+        <span class="k">if</span> <span class="n">EVEN_M</span> <span class="o">&amp;</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+            <span class="n">do</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">do_ptrs</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="c1"># [2022-11-01] TD: Triton bug, there&#39;s a race condition if we just use m_mask and not d_mask.</span>
+            <span class="n">do</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">do_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_m_curr</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">)</span>
+                                        <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">),</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
+        <span class="c1"># if EVEN_M:</span>
+        <span class="c1">#     if EVEN_HEADDIM:</span>
+        <span class="c1">#         do = tl.load(do_ptrs)</span>
+        <span class="c1">#     else:</span>
+        <span class="c1">#         do = tl.load(do_ptrs, mask=offs_d[None, :] &lt; headdim, other=0.0)</span>
+        <span class="c1"># else:</span>
+        <span class="c1">#     if EVEN_HEADDIM:</span>
+        <span class="c1">#         do = tl.load(do_ptrs, mask=offs_m_curr[:, None] &lt; seqlen_q, other=0.0)</span>
+        <span class="c1">#     else:</span>
+        <span class="c1">#         do = tl.load(do_ptrs, mask=(offs_m_curr[:, None] &lt; seqlen_q)</span>
+        <span class="c1">#                                    &amp; (offs_d[None, :] &lt; headdim), other=0.0)</span>
+        <span class="n">dv</span> <span class="o">+=</span> <span class="n">tl</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">p</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">do</span><span class="o">.</span><span class="n">dtype</span><span class="p">),</span> <span class="n">do</span><span class="p">,</span> <span class="n">trans_a</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="c1"># compute dp = dot(v, do)</span>
+        <span class="c1"># There seems to be a race condition when headdim=48/96, and dq, dk are wrong.</span>
+        <span class="c1"># Also wrong for headdim=128, seqlen=(108, 256), and ATOMIC_ADD=True</span>
+        <span class="c1"># Also wrong for headdim=64, seqlen=(1023, 1024), and ATOMIC_ADD=False</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="n">EVEN_M</span> <span class="o">&amp;</span> <span class="n">EVEN_HEADDIM</span><span class="p">):</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">debug_barrier</span><span class="p">()</span>
+        <span class="n">dp</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">do</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="n">trans_b</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="c1"># There&#39;s a race condition for headdim=48</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">debug_barrier</span><span class="p">()</span>
+        <span class="c1"># compute ds = p * (dp - delta[:, None])</span>
+        <span class="c1"># Putting the subtraction after the dp matmul (instead of before) is slightly faster</span>
+        <span class="n">Di</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">D</span> <span class="o">+</span> <span class="n">offs_m_curr</span><span class="p">)</span>
+        <span class="c1"># Converting ds to q.dtype here reduces register pressure and makes it much faster</span>
+        <span class="c1"># for BLOCK_HEADDIM=128</span>
+        <span class="n">ds</span> <span class="o">=</span> <span class="p">(</span><span class="n">p</span> <span class="o">*</span> <span class="p">(</span><span class="n">dp</span> <span class="o">-</span> <span class="n">Di</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">])</span> <span class="o">*</span> <span class="n">softmax_scale</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">q</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
+        <span class="c1"># compute dk = dot(ds.T, q)</span>
+        <span class="n">dk</span> <span class="o">+=</span> <span class="n">tl</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">ds</span><span class="p">,</span> <span class="n">q</span><span class="p">,</span> <span class="n">trans_a</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="c1"># compute dq</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="p">(</span><span class="n">EVEN_M</span> <span class="o">&amp;</span> <span class="n">EVEN_HEADDIM</span><span class="p">):</span>  <span class="c1"># Otherewise there&#39;s a race condition when BIAS_TYPE=&#39;matrix&#39;</span>
+            <span class="n">tl</span><span class="o">.</span><span class="n">debug_barrier</span><span class="p">()</span>
+        <span class="k">if</span> <span class="ow">not</span> <span class="n">ATOMIC_ADD</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">EVEN_M</span> <span class="o">&amp;</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>  <span class="c1"># Race condition if we just do EVEN_M</span>
+                <span class="n">dq</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">dq_ptrs</span><span class="p">,</span> <span class="n">eviction_policy</span><span class="o">=</span><span class="s2">&quot;evict_last&quot;</span><span class="p">)</span>
+                <span class="n">dq</span> <span class="o">+=</span> <span class="n">tl</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">ds</span><span class="p">,</span> <span class="n">k</span><span class="p">)</span>
+                <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">dq_ptrs</span><span class="p">,</span> <span class="n">dq</span><span class="p">,</span> <span class="n">eviction_policy</span><span class="o">=</span><span class="s2">&quot;evict_last&quot;</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+                    <span class="n">dq</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">dq_ptrs</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_m_curr</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span>
+                                <span class="n">eviction_policy</span><span class="o">=</span><span class="s2">&quot;evict_last&quot;</span><span class="p">)</span>
+                    <span class="n">dq</span> <span class="o">+=</span> <span class="n">tl</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">ds</span><span class="p">,</span> <span class="n">k</span><span class="p">)</span>
+                    <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">dq_ptrs</span><span class="p">,</span> <span class="n">dq</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_m_curr</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">,</span>
+                            <span class="n">eviction_policy</span><span class="o">=</span><span class="s2">&quot;evict_last&quot;</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">dq</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">dq_ptrs</span><span class="p">,</span>
+                                 <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_m_curr</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">),</span>
+                                 <span class="n">other</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">eviction_policy</span><span class="o">=</span><span class="s2">&quot;evict_last&quot;</span><span class="p">)</span>
+                    <span class="n">dq</span> <span class="o">+=</span> <span class="n">tl</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">ds</span><span class="p">,</span> <span class="n">k</span><span class="p">)</span>
+                    <span class="n">tl</span><span class="o">.</span><span class="n">store</span><span class="p">(</span><span class="n">dq_ptrs</span><span class="p">,</span> <span class="n">dq</span><span class="p">,</span>
+                             <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_m_curr</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">),</span>
+                             <span class="n">eviction_policy</span><span class="o">=</span><span class="s2">&quot;evict_last&quot;</span><span class="p">)</span>
+        <span class="k">else</span><span class="p">:</span>  <span class="c1"># If we&#39;re parallelizing across the seqlen_k dimension</span>
+            <span class="n">dq</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">ds</span><span class="p">,</span> <span class="n">k</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">EVEN_M</span> <span class="o">&amp;</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>  <span class="c1"># Race condition if we just do EVEN_M</span>
+                <span class="n">tl</span><span class="o">.</span><span class="n">atomic_add</span><span class="p">(</span><span class="n">dq_ptrs</span><span class="p">,</span> <span class="n">dq</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span>
+                    <span class="n">tl</span><span class="o">.</span><span class="n">atomic_add</span><span class="p">(</span><span class="n">dq_ptrs</span><span class="p">,</span> <span class="n">dq</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">offs_m_curr</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">)</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">tl</span><span class="o">.</span><span class="n">atomic_add</span><span class="p">(</span><span class="n">dq_ptrs</span><span class="p">,</span> <span class="n">dq</span><span class="p">,</span>
+                                  <span class="n">mask</span><span class="o">=</span><span class="p">(</span><span class="n">offs_m_curr</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">seqlen_q</span><span class="p">)</span> <span class="o">&amp;</span> <span class="p">(</span><span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:]</span> <span class="o">&lt;</span> <span class="n">headdim</span><span class="p">))</span>
+        <span class="c1"># increment pointers</span>
+        <span class="n">dq_ptrs</span> <span class="o">+=</span> <span class="n">BLOCK_M</span> <span class="o">*</span> <span class="n">stride_dqm</span>
+        <span class="n">q_ptrs</span> <span class="o">+=</span> <span class="n">BLOCK_M</span> <span class="o">*</span> <span class="n">stride_qm</span>
+        <span class="n">do_ptrs</span> <span class="o">+=</span> <span class="n">BLOCK_M</span> <span class="o">*</span> <span class="n">stride_dom</span>
+        <span class="k">if</span> <span class="n">BIAS_TYPE</span> <span class="o">==</span> <span class="s1">&#39;matrix&#39;</span><span class="p">:</span>
+            <span class="n">b_ptrs</span> <span class="o">+=</span> <span class="n">BLOCK_M</span> <span class="o">*</span> <span class="n">stride_bm</span>
+    <span class="c1"># write-back</span>
+    <span class="n">dv_ptrs</span> <span class="o">=</span> <span class="n">DV</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_dvn</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+    <span class="n">dk_ptrs</span> <span class="o">=</span> <span class="n">DK</span> <span class="o">+</span> <span class="p">(</span><span class="n">offs_n</span><span class="p">[:,</span> <span class="kc">None</span><span class="p">]</span> <span class="o">*</span> <span class="n">stride_dkn</span> <span class="o">+</span> <span class="n">offs_d</span><span class="p">[</span><span class="kc">None</span><span class="p">,</span> <span class="p">:])</span>
+    <span class="n">_bwd_store_dk_dv</span><span class="p">(</span><span class="n">dk_ptrs</span><span class="p">,</span> <span class="n">dv_ptrs</span><span class="p">,</span> <span class="n">dk</span><span class="p">,</span> <span class="n">dv</span><span class="p">,</span> <span class="n">offs_n</span><span class="p">,</span> <span class="n">offs_d</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">headdim</span><span class="p">,</span>
+                     <span class="n">EVEN_M</span><span class="o">=</span><span class="n">EVEN_M</span><span class="p">,</span> <span class="n">EVEN_N</span><span class="o">=</span><span class="n">EVEN_N</span><span class="p">,</span> <span class="n">EVEN_HEADDIM</span><span class="o">=</span><span class="n">EVEN_HEADDIM</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="init_to_zero">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.init_to_zero">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">init_to_zero</span><span class="p">(</span><span class="n">name</span><span class="p">):</span>
+    <span class="k">return</span> <span class="k">lambda</span> <span class="n">nargs</span><span class="p">:</span> <span class="n">nargs</span><span class="p">[</span><span class="n">name</span><span class="p">]</span><span class="o">.</span><span class="n">zero_</span><span class="p">()</span></div>
+
+
+
+<span class="nd">@triton</span><span class="o">.</span><span class="n">autotune</span><span class="p">(</span>
+    <span class="n">configs</span><span class="o">=</span><span class="p">[</span>
+        <span class="n">triton</span><span class="o">.</span><span class="n">Config</span><span class="p">({</span><span class="s2">&quot;BLOCK_M&quot;</span><span class="p">:</span> <span class="mi">128</span><span class="p">,</span> <span class="s2">&quot;BLOCK_N&quot;</span><span class="p">:</span> <span class="mi">128</span><span class="p">,</span> <span class="s2">&quot;SEQUENCE_PARALLEL&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">},</span> <span class="n">num_warps</span><span class="o">=</span><span class="mi">8</span><span class="p">,</span> <span class="n">num_stages</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">pre_hook</span><span class="o">=</span><span class="n">init_to_zero</span><span class="p">(</span><span class="s1">&#39;DQ&#39;</span><span class="p">)),</span>
+        <span class="n">triton</span><span class="o">.</span><span class="n">Config</span><span class="p">({</span><span class="s2">&quot;BLOCK_M&quot;</span><span class="p">:</span> <span class="mi">128</span><span class="p">,</span> <span class="s2">&quot;BLOCK_N&quot;</span><span class="p">:</span> <span class="mi">128</span><span class="p">,</span> <span class="s2">&quot;SEQUENCE_PARALLEL&quot;</span><span class="p">:</span> <span class="kc">True</span><span class="p">},</span> <span class="n">num_warps</span><span class="o">=</span><span class="mi">8</span><span class="p">,</span> <span class="n">num_stages</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">pre_hook</span><span class="o">=</span><span class="n">init_to_zero</span><span class="p">(</span><span class="s1">&#39;DQ&#39;</span><span class="p">)),</span>
+        <span class="c1"># Other configs seem to give wrong results when seqlen_q % 128 != 0, disabling them for now</span>
+        <span class="c1"># # Kernel is buggy (give wrong result) if we set BLOCK_m=128, BLOCK_n=64, num_warps=*4*</span>
+        <span class="c1"># triton.Config({&quot;BLOCK_M&quot;: 128, &quot;BLOCK_N&quot;: 64, &quot;SEQUENCE_PARALLEL&quot;: False}, num_warps=8, num_stages=1, pre_hook=init_to_zero(&#39;DQ&#39;)),</span>
+        <span class="c1"># triton.Config({&quot;BLOCK_M&quot;: 128, &quot;BLOCK_N&quot;: 64, &quot;SEQUENCE_PARALLEL&quot;: True}, num_warps=8, num_stages=1, pre_hook=init_to_zero(&#39;DQ&#39;)),</span>
+        <span class="c1"># triton.Config({&quot;BLOCK_M&quot;: 64, &quot;BLOCK_N&quot;: 64, &quot;SEQUENCE_PARALLEL&quot;: False}, num_warps=4, num_stages=1, pre_hook=init_to_zero(&#39;DQ&#39;)),</span>
+        <span class="c1"># triton.Config({&quot;BLOCK_M&quot;: 64, &quot;BLOCK_N&quot;: 64, &quot;SEQUENCE_PARALLEL&quot;: True}, num_warps=4, num_stages=1, pre_hook=init_to_zero(&#39;DQ&#39;)),</span>
+    <span class="p">],</span>
+    <span class="n">key</span><span class="o">=</span><span class="p">[</span><span class="s1">&#39;CACHE_KEY_SEQLEN_Q&#39;</span><span class="p">,</span> <span class="s1">&#39;CACHE_KEY_SEQLEN_K&#39;</span><span class="p">,</span> <span class="s1">&#39;BIAS_TYPE&#39;</span><span class="p">,</span> <span class="s1">&#39;IS_CAUSAL&#39;</span><span class="p">,</span> <span class="s1">&#39;BLOCK_HEADDIM&#39;</span><span class="p">],</span>
+<span class="p">)</span>
+<span class="nd">@triton</span><span class="o">.</span><span class="n">heuristics</span><span class="p">(</span>
+    <span class="p">{</span>
+        <span class="s2">&quot;EVEN_M&quot;</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">args</span><span class="p">:</span> <span class="n">args</span><span class="p">[</span><span class="s2">&quot;seqlen_q&quot;</span><span class="p">]</span> <span class="o">%</span> <span class="n">args</span><span class="p">[</span><span class="s2">&quot;BLOCK_M&quot;</span><span class="p">]</span> <span class="o">==</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="s2">&quot;EVEN_N&quot;</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">args</span><span class="p">:</span> <span class="n">args</span><span class="p">[</span><span class="s2">&quot;seqlen_k&quot;</span><span class="p">]</span> <span class="o">%</span> <span class="n">args</span><span class="p">[</span><span class="s2">&quot;BLOCK_N&quot;</span><span class="p">]</span> <span class="o">==</span> <span class="mi">0</span><span class="p">,</span>
+        <span class="s2">&quot;EVEN_HEADDIM&quot;</span><span class="p">:</span> <span class="k">lambda</span> <span class="n">args</span><span class="p">:</span> <span class="n">args</span><span class="p">[</span><span class="s2">&quot;headdim&quot;</span><span class="p">]</span> <span class="o">==</span> <span class="n">args</span><span class="p">[</span><span class="s2">&quot;BLOCK_HEADDIM&quot;</span><span class="p">],</span>
+    <span class="p">}</span>
+<span class="p">)</span>
+<span class="nd">@triton</span><span class="o">.</span><span class="n">jit</span>
+<div class="viewcode-block" id="_bwd_kernel">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_bwd_kernel</span><span class="p">(</span>
+    <span class="n">Q</span><span class="p">,</span> <span class="n">K</span><span class="p">,</span> <span class="n">V</span><span class="p">,</span> <span class="n">Bias</span><span class="p">,</span>
+    <span class="n">DO</span><span class="p">,</span> <span class="n">DQ</span><span class="p">,</span> <span class="n">DK</span><span class="p">,</span> <span class="n">DV</span><span class="p">,</span>
+    <span class="n">LSE</span><span class="p">,</span> <span class="n">D</span><span class="p">,</span>
+    <span class="n">softmax_scale</span><span class="p">,</span>
+    <span class="n">stride_qb</span><span class="p">,</span> <span class="n">stride_qh</span><span class="p">,</span> <span class="n">stride_qm</span><span class="p">,</span>
+    <span class="n">stride_kb</span><span class="p">,</span> <span class="n">stride_kh</span><span class="p">,</span> <span class="n">stride_kn</span><span class="p">,</span>
+    <span class="n">stride_vb</span><span class="p">,</span> <span class="n">stride_vh</span><span class="p">,</span> <span class="n">stride_vn</span><span class="p">,</span>
+    <span class="n">stride_bb</span><span class="p">,</span> <span class="n">stride_bh</span><span class="p">,</span> <span class="n">stride_bm</span><span class="p">,</span>
+    <span class="n">stride_dob</span><span class="p">,</span> <span class="n">stride_doh</span><span class="p">,</span> <span class="n">stride_dom</span><span class="p">,</span>
+    <span class="n">stride_dqb</span><span class="p">,</span> <span class="n">stride_dqh</span><span class="p">,</span> <span class="n">stride_dqm</span><span class="p">,</span>
+    <span class="n">stride_dkb</span><span class="p">,</span> <span class="n">stride_dkh</span><span class="p">,</span> <span class="n">stride_dkn</span><span class="p">,</span>
+    <span class="n">stride_dvb</span><span class="p">,</span> <span class="n">stride_dvh</span><span class="p">,</span> <span class="n">stride_dvn</span><span class="p">,</span>
+    <span class="n">nheads</span><span class="p">,</span> <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">seqlen_q_rounded</span><span class="p">,</span> <span class="n">headdim</span><span class="p">,</span>
+    <span class="n">CACHE_KEY_SEQLEN_Q</span><span class="p">,</span> <span class="n">CACHE_KEY_SEQLEN_K</span><span class="p">,</span>
+    <span class="n">BIAS_TYPE</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">IS_CAUSAL</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">BLOCK_HEADDIM</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">SEQUENCE_PARALLEL</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">EVEN_M</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span> <span class="n">EVEN_N</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span> <span class="n">EVEN_HEADDIM</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+    <span class="n">BLOCK_M</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span> <span class="n">BLOCK_N</span><span class="p">:</span> <span class="n">tl</span><span class="o">.</span><span class="n">constexpr</span><span class="p">,</span>
+<span class="p">):</span>
+    <span class="n">off_hb</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">program_id</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
+    <span class="n">off_b</span> <span class="o">=</span> <span class="n">off_hb</span> <span class="o">//</span> <span class="n">nheads</span>
+    <span class="n">off_h</span> <span class="o">=</span> <span class="n">off_hb</span> <span class="o">%</span> <span class="n">nheads</span>
+    <span class="c1"># offset pointers for batch/head</span>
+    <span class="n">Q</span> <span class="o">+=</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_qb</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_qh</span>
+    <span class="n">K</span> <span class="o">+=</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_kb</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_kh</span>
+    <span class="n">V</span> <span class="o">+=</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_vb</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_vh</span>
+    <span class="n">DO</span> <span class="o">+=</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_dob</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_doh</span>
+    <span class="n">DQ</span> <span class="o">+=</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_dqb</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_dqh</span>
+    <span class="n">DK</span> <span class="o">+=</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_dkb</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_dkh</span>
+    <span class="n">DV</span> <span class="o">+=</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_dvb</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_dvh</span>
+    <span class="k">if</span> <span class="n">BIAS_TYPE</span> <span class="o">!=</span> <span class="s1">&#39;none&#39;</span><span class="p">:</span>
+        <span class="n">Bias</span> <span class="o">+=</span> <span class="n">off_b</span> <span class="o">*</span> <span class="n">stride_bb</span> <span class="o">+</span> <span class="n">off_h</span> <span class="o">*</span> <span class="n">stride_bh</span>
+    <span class="c1"># pointer to row-wise quantities in value-like data</span>
+    <span class="n">D</span> <span class="o">+=</span> <span class="n">off_hb</span> <span class="o">*</span> <span class="n">seqlen_q_rounded</span>
+    <span class="n">LSE</span> <span class="o">+=</span> <span class="n">off_hb</span> <span class="o">*</span> <span class="n">seqlen_q_rounded</span>
+    <span class="k">if</span> <span class="ow">not</span> <span class="n">SEQUENCE_PARALLEL</span><span class="p">:</span>
+        <span class="n">num_block_n</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">cdiv</span><span class="p">(</span><span class="n">seqlen_k</span><span class="p">,</span> <span class="n">BLOCK_N</span><span class="p">)</span>
+        <span class="k">for</span> <span class="n">start_n</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">num_block_n</span><span class="p">):</span>
+            <span class="n">_bwd_kernel_one_col_block</span><span class="p">(</span>
+                <span class="n">start_n</span><span class="p">,</span>
+                <span class="n">Q</span><span class="p">,</span> <span class="n">K</span><span class="p">,</span> <span class="n">V</span><span class="p">,</span> <span class="n">Bias</span><span class="p">,</span>
+                <span class="n">DO</span><span class="p">,</span> <span class="n">DQ</span><span class="p">,</span> <span class="n">DK</span><span class="p">,</span> <span class="n">DV</span><span class="p">,</span>
+                <span class="n">LSE</span><span class="p">,</span> <span class="n">D</span><span class="p">,</span>
+                <span class="n">softmax_scale</span><span class="p">,</span>
+                <span class="n">stride_qm</span><span class="p">,</span> <span class="n">stride_kn</span><span class="p">,</span> <span class="n">stride_vn</span><span class="p">,</span> <span class="n">stride_bm</span><span class="p">,</span>
+                <span class="n">stride_dom</span><span class="p">,</span> <span class="n">stride_dqm</span><span class="p">,</span> <span class="n">stride_dkn</span><span class="p">,</span> <span class="n">stride_dvn</span><span class="p">,</span>
+                <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">headdim</span><span class="p">,</span>
+                <span class="n">ATOMIC_ADD</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
+                <span class="n">BIAS_TYPE</span><span class="o">=</span><span class="n">BIAS_TYPE</span><span class="p">,</span>
+                <span class="n">IS_CAUSAL</span><span class="o">=</span><span class="n">IS_CAUSAL</span><span class="p">,</span>
+                <span class="n">BLOCK_HEADDIM</span><span class="o">=</span><span class="n">BLOCK_HEADDIM</span><span class="p">,</span>
+                <span class="n">EVEN_M</span><span class="o">=</span><span class="n">EVEN_M</span><span class="p">,</span> <span class="n">EVEN_N</span><span class="o">=</span><span class="n">EVEN_N</span><span class="p">,</span> <span class="n">EVEN_HEADDIM</span><span class="o">=</span><span class="n">EVEN_HEADDIM</span><span class="p">,</span>
+                <span class="n">BLOCK_M</span><span class="o">=</span><span class="n">BLOCK_M</span><span class="p">,</span> <span class="n">BLOCK_N</span><span class="o">=</span><span class="n">BLOCK_N</span>
+            <span class="p">)</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">start_n</span> <span class="o">=</span> <span class="n">tl</span><span class="o">.</span><span class="n">program_id</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span>
+        <span class="n">_bwd_kernel_one_col_block</span><span class="p">(</span>
+            <span class="n">start_n</span><span class="p">,</span>
+            <span class="n">Q</span><span class="p">,</span> <span class="n">K</span><span class="p">,</span> <span class="n">V</span><span class="p">,</span> <span class="n">Bias</span><span class="p">,</span>
+            <span class="n">DO</span><span class="p">,</span> <span class="n">DQ</span><span class="p">,</span> <span class="n">DK</span><span class="p">,</span> <span class="n">DV</span><span class="p">,</span>
+            <span class="n">LSE</span><span class="p">,</span> <span class="n">D</span><span class="p">,</span>
+            <span class="n">softmax_scale</span><span class="p">,</span>
+            <span class="n">stride_qm</span><span class="p">,</span> <span class="n">stride_kn</span><span class="p">,</span> <span class="n">stride_vn</span><span class="p">,</span> <span class="n">stride_bm</span><span class="p">,</span>
+            <span class="n">stride_dom</span><span class="p">,</span> <span class="n">stride_dqm</span><span class="p">,</span> <span class="n">stride_dkn</span><span class="p">,</span> <span class="n">stride_dvn</span><span class="p">,</span>
+            <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">headdim</span><span class="p">,</span>
+            <span class="n">ATOMIC_ADD</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+            <span class="n">BIAS_TYPE</span><span class="o">=</span><span class="n">BIAS_TYPE</span><span class="p">,</span>
+            <span class="n">IS_CAUSAL</span><span class="o">=</span><span class="n">IS_CAUSAL</span><span class="p">,</span>
+            <span class="n">BLOCK_HEADDIM</span><span class="o">=</span><span class="n">BLOCK_HEADDIM</span><span class="p">,</span>
+            <span class="n">EVEN_M</span><span class="o">=</span><span class="n">EVEN_M</span><span class="p">,</span> <span class="n">EVEN_N</span><span class="o">=</span><span class="n">EVEN_N</span><span class="p">,</span> <span class="n">EVEN_HEADDIM</span><span class="o">=</span><span class="n">EVEN_HEADDIM</span><span class="p">,</span>
+            <span class="n">BLOCK_M</span><span class="o">=</span><span class="n">BLOCK_M</span><span class="p">,</span> <span class="n">BLOCK_N</span><span class="o">=</span><span class="n">BLOCK_N</span>
+        <span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="_flash_attn_forward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._flash_attn_forward">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_flash_attn_forward</span><span class="p">(</span><span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+    <span class="c1"># shape constraints</span>
+    <span class="n">batch</span><span class="p">,</span> <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">nheads</span><span class="p">,</span> <span class="n">d</span> <span class="o">=</span> <span class="n">q</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">_</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">k</span><span class="o">.</span><span class="n">shape</span>
+    <span class="k">assert</span> <span class="n">k</span><span class="o">.</span><span class="n">shape</span> <span class="o">==</span> <span class="p">(</span><span class="n">batch</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">nheads</span><span class="p">,</span> <span class="n">d</span><span class="p">)</span>
+    <span class="k">assert</span> <span class="n">v</span><span class="o">.</span><span class="n">shape</span> <span class="o">==</span> <span class="p">(</span><span class="n">batch</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">nheads</span><span class="p">,</span> <span class="n">d</span><span class="p">)</span>
+    <span class="k">assert</span> <span class="n">d</span> <span class="o">&lt;=</span> <span class="mi">128</span><span class="p">,</span> <span class="s1">&#39;FlashAttention only support head dimensions up to 128&#39;</span>
+    <span class="k">assert</span> <span class="n">q</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">k</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="n">v</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="s1">&#39;All tensors must have the same type&#39;</span>
+    <span class="k">assert</span> <span class="n">q</span><span class="o">.</span><span class="n">dtype</span> <span class="ow">in</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">float16</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">bfloat16</span><span class="p">],</span> <span class="s1">&#39;Only support fp16 and bf16&#39;</span>
+    <span class="k">assert</span> <span class="n">q</span><span class="o">.</span><span class="n">is_cuda</span> <span class="ow">and</span> <span class="n">k</span><span class="o">.</span><span class="n">is_cuda</span> <span class="ow">and</span> <span class="n">v</span><span class="o">.</span><span class="n">is_cuda</span>
+    <span class="n">softmax_scale</span> <span class="o">=</span> <span class="n">softmax_scale</span> <span class="ow">or</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">d</span><span class="p">)</span>
+
+    <span class="n">has_bias</span> <span class="o">=</span> <span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+    <span class="n">bias_type</span> <span class="o">=</span> <span class="s1">&#39;none&#39;</span>
+    <span class="k">if</span> <span class="n">has_bias</span><span class="p">:</span>
+        <span class="k">assert</span> <span class="n">bias</span><span class="o">.</span><span class="n">dtype</span> <span class="ow">in</span> <span class="p">[</span><span class="n">q</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">]</span>
+        <span class="k">assert</span> <span class="n">bias</span><span class="o">.</span><span class="n">is_cuda</span>
+        <span class="k">assert</span> <span class="n">bias</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span> <span class="o">==</span> <span class="mi">4</span>
+        <span class="k">if</span> <span class="n">bias</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">bias</span> <span class="o">=</span> <span class="n">bias</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+        <span class="k">if</span> <span class="n">bias</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">:]</span> <span class="o">==</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">):</span>
+            <span class="n">bias_type</span> <span class="o">=</span> <span class="s1">&#39;vector&#39;</span>
+        <span class="k">elif</span> <span class="n">bias</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">:]</span> <span class="o">==</span> <span class="p">(</span><span class="n">seqlen_q</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">):</span>
+            <span class="n">bias_type</span> <span class="o">=</span> <span class="s1">&#39;matrix&#39;</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s1">&#39;Last 2 dimensions of bias must be (1, seqlen_k)&#39;</span>
+                               <span class="s1">&#39; or (seqlen_q, seqlen_k)&#39;</span><span class="p">)</span>
+        <span class="n">bias</span> <span class="o">=</span> <span class="n">bias</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="n">batch</span><span class="p">,</span> <span class="n">nheads</span><span class="p">,</span> <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">)</span>
+    <span class="n">bias_strides</span> <span class="o">=</span> <span class="p">(</span><span class="n">bias</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">bias</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span> <span class="n">bias</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">))</span> <span class="k">if</span> <span class="n">has_bias</span> <span class="k">else</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+    <span class="n">seqlen_q_rounded</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">seqlen_q</span> <span class="o">/</span> <span class="mi">128</span><span class="p">)</span> <span class="o">*</span> <span class="mi">128</span>
+    <span class="n">lse</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">((</span><span class="n">batch</span><span class="p">,</span> <span class="n">nheads</span><span class="p">,</span> <span class="n">seqlen_q_rounded</span><span class="p">),</span> <span class="n">device</span><span class="o">=</span><span class="n">q</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+    <span class="n">tmp</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty</span><span class="p">((</span><span class="n">batch</span><span class="p">,</span> <span class="n">nheads</span><span class="p">,</span> <span class="n">seqlen_q_rounded</span><span class="p">),</span> <span class="n">device</span><span class="o">=</span><span class="n">q</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+    <span class="n">o</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty_like</span><span class="p">(</span><span class="n">q</span><span class="p">)</span>
+
+    <span class="n">BLOCK_HEADDIM</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">triton</span><span class="o">.</span><span class="n">next_power_of_2</span><span class="p">(</span><span class="n">d</span><span class="p">),</span> <span class="mi">16</span><span class="p">)</span>
+    <span class="n">BLOCK</span> <span class="o">=</span> <span class="mi">128</span>
+    <span class="n">num_warps</span> <span class="o">=</span> <span class="mi">4</span> <span class="k">if</span> <span class="n">d</span> <span class="o">&lt;=</span> <span class="mi">64</span> <span class="k">else</span> <span class="mi">8</span>
+    <span class="n">grid</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">META</span><span class="p">:</span> <span class="p">(</span><span class="n">triton</span><span class="o">.</span><span class="n">cdiv</span><span class="p">(</span><span class="n">seqlen_q</span><span class="p">,</span> <span class="n">META</span><span class="p">[</span><span class="s2">&quot;BLOCK_M&quot;</span><span class="p">]),</span> <span class="n">batch</span> <span class="o">*</span> <span class="n">nheads</span><span class="p">)</span>
+    <span class="n">_fwd_kernel</span><span class="p">[</span><span class="n">grid</span><span class="p">](</span>
+        <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="n">bias</span><span class="p">,</span> <span class="n">o</span><span class="p">,</span>
+        <span class="n">lse</span><span class="p">,</span> <span class="n">tmp</span><span class="p">,</span>
+        <span class="n">softmax_scale</span><span class="p">,</span>
+        <span class="n">q</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">q</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">q</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+        <span class="n">k</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">k</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">k</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+        <span class="n">v</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">v</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">v</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+        <span class="o">*</span><span class="n">bias_strides</span><span class="p">,</span>
+        <span class="n">o</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">o</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">o</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+        <span class="n">nheads</span><span class="p">,</span> <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">seqlen_q_rounded</span><span class="p">,</span> <span class="n">d</span><span class="p">,</span>
+        <span class="n">seqlen_q</span> <span class="o">//</span> <span class="mi">32</span><span class="p">,</span>  <span class="n">seqlen_k</span> <span class="o">//</span> <span class="mi">32</span><span class="p">,</span> <span class="c1"># key for triton cache (limit number of compilations)</span>
+        <span class="c1"># Can&#39;t use kwargs here because triton autotune expects key to be args, not kwargs</span>
+        <span class="c1"># IS_CAUSAL=causal, BLOCK_HEADDIM=d,</span>
+        <span class="n">bias_type</span><span class="p">,</span> <span class="n">causal</span><span class="p">,</span> <span class="n">BLOCK_HEADDIM</span><span class="p">,</span>
+        <span class="n">BLOCK_M</span><span class="o">=</span><span class="n">BLOCK</span><span class="p">,</span> <span class="n">BLOCK_N</span><span class="o">=</span><span class="n">BLOCK</span><span class="p">,</span>
+        <span class="n">num_warps</span><span class="o">=</span><span class="n">num_warps</span><span class="p">,</span>
+        <span class="n">num_stages</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="k">return</span> <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span> <span class="n">softmax_scale</span>  <span class="c1"># softmax_scale could have been updated</span></div>
+
+
+
+<div class="viewcode-block" id="_flash_attn_backward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._flash_attn_backward">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_flash_attn_backward</span><span class="p">(</span><span class="n">do</span><span class="p">,</span> <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span> <span class="n">dq</span><span class="p">,</span> <span class="n">dk</span><span class="p">,</span> <span class="n">dv</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+    <span class="c1"># Make sure that the last dimension is contiguous</span>
+    <span class="k">if</span> <span class="n">do</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">:</span>
+        <span class="n">do</span> <span class="o">=</span> <span class="n">do</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+    <span class="n">batch</span><span class="p">,</span> <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">nheads</span><span class="p">,</span> <span class="n">d</span> <span class="o">=</span> <span class="n">q</span><span class="o">.</span><span class="n">shape</span>
+    <span class="n">_</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">k</span><span class="o">.</span><span class="n">shape</span>
+    <span class="c1"># assert d in {16, 32, 64, 128}</span>
+    <span class="k">assert</span> <span class="n">d</span> <span class="o">&lt;=</span> <span class="mi">128</span>
+    <span class="n">seqlen_q_rounded</span> <span class="o">=</span> <span class="n">math</span><span class="o">.</span><span class="n">ceil</span><span class="p">(</span><span class="n">seqlen_q</span> <span class="o">/</span> <span class="mi">128</span><span class="p">)</span> <span class="o">*</span> <span class="mi">128</span>
+    <span class="k">assert</span> <span class="n">lse</span><span class="o">.</span><span class="n">shape</span> <span class="o">==</span> <span class="p">(</span><span class="n">batch</span><span class="p">,</span> <span class="n">nheads</span><span class="p">,</span> <span class="n">seqlen_q_rounded</span><span class="p">)</span>
+    <span class="k">assert</span> <span class="n">q</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">==</span> <span class="n">k</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">==</span> <span class="n">v</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">==</span> <span class="n">o</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span>
+    <span class="k">assert</span> <span class="n">dq</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">==</span> <span class="n">dk</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">==</span> <span class="n">dv</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span>
+    <span class="n">softmax_scale</span> <span class="o">=</span> <span class="n">softmax_scale</span> <span class="ow">or</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="n">math</span><span class="o">.</span><span class="n">sqrt</span><span class="p">(</span><span class="n">d</span><span class="p">)</span>
+    <span class="c1"># dq_accum = torch.zeros_like(q, dtype=torch.float32)</span>
+    <span class="n">dq_accum</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty_like</span><span class="p">(</span><span class="n">q</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">torch</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+    <span class="n">delta</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty_like</span><span class="p">(</span><span class="n">lse</span><span class="p">)</span>
+    <span class="c1"># delta = torch.zeros_like(lse)</span>
+
+    <span class="n">BLOCK_HEADDIM</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">triton</span><span class="o">.</span><span class="n">next_power_of_2</span><span class="p">(</span><span class="n">d</span><span class="p">),</span> <span class="mi">16</span><span class="p">)</span>
+    <span class="n">grid</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">META</span><span class="p">:</span> <span class="p">(</span><span class="n">triton</span><span class="o">.</span><span class="n">cdiv</span><span class="p">(</span><span class="n">seqlen_q</span><span class="p">,</span> <span class="n">META</span><span class="p">[</span><span class="s2">&quot;BLOCK_M&quot;</span><span class="p">]),</span> <span class="n">batch</span> <span class="o">*</span> <span class="n">nheads</span><span class="p">)</span>
+    <span class="n">_bwd_preprocess_do_o_dot</span><span class="p">[</span><span class="n">grid</span><span class="p">](</span>
+        <span class="n">o</span><span class="p">,</span> <span class="n">do</span><span class="p">,</span> <span class="n">delta</span><span class="p">,</span>
+        <span class="n">o</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">o</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">o</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+        <span class="n">do</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">do</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">do</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+        <span class="n">nheads</span><span class="p">,</span> <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">seqlen_q_rounded</span><span class="p">,</span> <span class="n">d</span><span class="p">,</span>
+        <span class="n">BLOCK_M</span><span class="o">=</span><span class="mi">128</span><span class="p">,</span> <span class="n">BLOCK_HEADDIM</span><span class="o">=</span><span class="n">BLOCK_HEADDIM</span><span class="p">,</span>
+    <span class="p">)</span>
+
+    <span class="n">has_bias</span> <span class="o">=</span> <span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
+    <span class="n">bias_type</span> <span class="o">=</span> <span class="s1">&#39;none&#39;</span>
+    <span class="k">if</span> <span class="n">has_bias</span><span class="p">:</span>
+        <span class="k">assert</span> <span class="n">bias</span><span class="o">.</span><span class="n">dtype</span> <span class="ow">in</span> <span class="p">[</span><span class="n">q</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="n">torch</span><span class="o">.</span><span class="n">float</span><span class="p">]</span>
+        <span class="k">assert</span> <span class="n">bias</span><span class="o">.</span><span class="n">is_cuda</span>
+        <span class="k">assert</span> <span class="n">bias</span><span class="o">.</span><span class="n">dim</span><span class="p">()</span> <span class="o">==</span> <span class="mi">4</span>
+        <span class="k">assert</span> <span class="n">bias</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span>
+        <span class="k">if</span> <span class="n">bias</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">:]</span> <span class="o">==</span> <span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">):</span>
+            <span class="n">bias_type</span> <span class="o">=</span> <span class="s1">&#39;vector&#39;</span>
+        <span class="k">elif</span> <span class="n">bias</span><span class="o">.</span><span class="n">shape</span><span class="p">[</span><span class="mi">2</span><span class="p">:]</span> <span class="o">==</span> <span class="p">(</span><span class="n">seqlen_q</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">):</span>
+            <span class="n">bias_type</span> <span class="o">=</span> <span class="s1">&#39;matrix&#39;</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="s1">&#39;Last 2 dimensions of bias must be (1, seqlen_k)&#39;</span>
+                               <span class="s1">&#39; or (seqlen_q, seqlen_k)&#39;</span><span class="p">)</span>
+        <span class="n">bias</span> <span class="o">=</span> <span class="n">bias</span><span class="o">.</span><span class="n">expand</span><span class="p">(</span><span class="n">batch</span><span class="p">,</span> <span class="n">nheads</span><span class="p">,</span> <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">)</span>
+    <span class="n">bias_strides</span> <span class="o">=</span> <span class="p">(</span><span class="n">bias</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">bias</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span> <span class="n">bias</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">))</span> <span class="k">if</span> <span class="n">has_bias</span> <span class="k">else</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
+
+    <span class="c1"># BLOCK_M = 128</span>
+    <span class="c1"># BLOCK_N = 64</span>
+    <span class="c1"># num_warps = 4</span>
+    <span class="n">grid</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">META</span><span class="p">:</span> <span class="p">(</span><span class="n">triton</span><span class="o">.</span><span class="n">cdiv</span><span class="p">(</span><span class="n">seqlen_k</span><span class="p">,</span> <span class="n">META</span><span class="p">[</span><span class="s2">&quot;BLOCK_N&quot;</span><span class="p">])</span> <span class="k">if</span> <span class="n">META</span><span class="p">[</span><span class="s2">&quot;SEQUENCE_PARALLEL&quot;</span><span class="p">]</span> <span class="k">else</span> <span class="mi">1</span><span class="p">,</span>
+                    <span class="n">batch</span> <span class="o">*</span> <span class="n">nheads</span><span class="p">)</span>
+    <span class="n">_bwd_kernel</span><span class="p">[</span><span class="n">grid</span><span class="p">](</span>
+        <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="n">bias</span><span class="p">,</span>
+        <span class="n">do</span><span class="p">,</span> <span class="n">dq_accum</span><span class="p">,</span> <span class="n">dk</span><span class="p">,</span> <span class="n">dv</span><span class="p">,</span>
+        <span class="n">lse</span><span class="p">,</span> <span class="n">delta</span><span class="p">,</span>
+        <span class="n">softmax_scale</span><span class="p">,</span>
+        <span class="n">q</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">q</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">q</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+        <span class="n">k</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">k</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">k</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+        <span class="n">v</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">v</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">v</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+        <span class="o">*</span><span class="n">bias_strides</span><span class="p">,</span>
+        <span class="n">do</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">do</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">do</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+        <span class="n">dq_accum</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">dq_accum</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">dq_accum</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+        <span class="n">dk</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">dk</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">dk</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+        <span class="n">dv</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="n">dv</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">2</span><span class="p">),</span> <span class="n">dv</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="mi">1</span><span class="p">),</span>
+        <span class="n">nheads</span><span class="p">,</span> <span class="n">seqlen_q</span><span class="p">,</span> <span class="n">seqlen_k</span><span class="p">,</span> <span class="n">seqlen_q_rounded</span><span class="p">,</span> <span class="n">d</span><span class="p">,</span>
+        <span class="n">seqlen_q</span> <span class="o">//</span> <span class="mi">32</span><span class="p">,</span>  <span class="n">seqlen_k</span> <span class="o">//</span> <span class="mi">32</span><span class="p">,</span> <span class="c1"># key for triton cache (limit number of compilations)</span>
+        <span class="c1"># Can&#39;t use kwargs here because triton autotune expects key to be args, not kwargs</span>
+        <span class="c1"># IS_CAUSAL=causal, BLOCK_HEADDIM=d,</span>
+        <span class="n">bias_type</span><span class="p">,</span> <span class="n">causal</span><span class="p">,</span> <span class="n">BLOCK_HEADDIM</span><span class="p">,</span>
+        <span class="c1"># SEQUENCE_PARALLEL=False,</span>
+        <span class="c1"># BLOCK_M=BLOCK_M, BLOCK_N=BLOCK_N,</span>
+        <span class="c1"># num_warps=num_warps,</span>
+        <span class="c1"># num_stages=1,</span>
+    <span class="p">)</span>
+    <span class="n">dq</span><span class="o">.</span><span class="n">copy_</span><span class="p">(</span><span class="n">dq_accum</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="FlashAttnQKVPackedFunc">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">FlashAttnQKVPackedFunc</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">autograd</span><span class="o">.</span><span class="n">Function</span><span class="p">):</span>
+
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="FlashAttnQKVPackedFunc.forward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc.forward">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span><span class="n">ctx</span><span class="p">,</span> <span class="n">qkv</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">            qkv: (batch, seqlen, 3, nheads, headdim)</span>
+<span class="sd">            bias: optional, shape broadcastible to (batch, nheads, seqlen, seqlen).</span>
+<span class="sd">                For example, ALiBi mask for causal would have shape (1, nheads, 1, seqlen).</span>
+<span class="sd">                ALiBi mask for non-causal would have shape (1, nheads, seqlen, seqlen)</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># Make sure that the last dimension is contiguous</span>
+        <span class="k">if</span> <span class="n">qkv</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">:</span>
+            <span class="n">qkv</span> <span class="o">=</span> <span class="n">qkv</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>
+        <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span> <span class="n">ctx</span><span class="o">.</span><span class="n">softmax_scale</span> <span class="o">=</span> <span class="n">_flash_attn_forward</span><span class="p">(</span>
+            <span class="n">qkv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">],</span> <span class="n">qkv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">],</span> <span class="n">qkv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">2</span><span class="p">],</span> <span class="n">bias</span><span class="o">=</span><span class="n">bias</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="n">causal</span><span class="p">,</span>
+            <span class="n">softmax_scale</span><span class="o">=</span><span class="n">softmax_scale</span>
+        <span class="p">)</span>
+        <span class="n">ctx</span><span class="o">.</span><span class="n">save_for_backward</span><span class="p">(</span><span class="n">qkv</span><span class="p">,</span> <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span> <span class="n">bias</span><span class="p">)</span>
+        <span class="n">ctx</span><span class="o">.</span><span class="n">causal</span> <span class="o">=</span> <span class="n">causal</span>
+        <span class="k">return</span> <span class="n">o</span></div>
+
+
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="FlashAttnQKVPackedFunc.backward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc.backward">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">backward</span><span class="p">(</span><span class="n">ctx</span><span class="p">,</span> <span class="n">do</span><span class="p">):</span>
+        <span class="n">qkv</span><span class="p">,</span> <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span> <span class="n">bias</span> <span class="o">=</span> <span class="n">ctx</span><span class="o">.</span><span class="n">saved_tensors</span>
+        <span class="k">assert</span> <span class="ow">not</span> <span class="n">ctx</span><span class="o">.</span><span class="n">needs_input_grad</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="s1">&#39;FlashAttention does not support bias gradient yet&#39;</span>
+        <span class="c1"># Triton&#39;s autotune causes the Tensor._version to change, and so Pytorch autograd</span>
+        <span class="c1"># does a memcpy. To avoid this we run in inference_mode, which doesn&#39;t track the version.</span>
+        <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">inference_mode</span><span class="p">():</span>
+            <span class="n">dqkv</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty_like</span><span class="p">(</span><span class="n">qkv</span><span class="p">)</span>
+            <span class="n">_flash_attn_backward</span><span class="p">(</span><span class="n">do</span><span class="p">,</span> <span class="n">qkv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">],</span> <span class="n">qkv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">],</span> <span class="n">qkv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">2</span><span class="p">],</span> <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span>
+                                 <span class="n">dqkv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">],</span> <span class="n">dqkv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">],</span> <span class="n">dqkv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">2</span><span class="p">],</span>
+                                 <span class="n">bias</span><span class="o">=</span><span class="n">bias</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="n">ctx</span><span class="o">.</span><span class="n">causal</span><span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="n">ctx</span><span class="o">.</span><span class="n">softmax_scale</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">dqkv</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span></div>
+</div>
+
+
+
+<div class="viewcode-block" id="flash_attn_qkvpacked_func">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_qkvpacked_func">[docs]</a>
+<span class="n">flash_attn_qkvpacked_func</span> <span class="o">=</span> <span class="n">FlashAttnQKVPackedFunc</span><span class="o">.</span><span class="n">apply</span></div>
+
+
+
+<div class="viewcode-block" id="FlashAttnKVPackedFunc">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">FlashAttnKVPackedFunc</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">autograd</span><span class="o">.</span><span class="n">Function</span><span class="p">):</span>
+
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="FlashAttnKVPackedFunc.forward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc.forward">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span><span class="n">ctx</span><span class="p">,</span> <span class="n">q</span><span class="p">,</span> <span class="n">kv</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">            q: (batch, seqlen_q, nheads, headdim)</span>
+<span class="sd">            kv: (batch, seqlen_k, 2, nheads, headdim)</span>
+<span class="sd">            bias: optional, shape broadcastible to (batch, nheads, seqlen_q, seqlen_k).</span>
+<span class="sd">                For example, ALiBi mask for causal would have shape (1, nheads, 1, seqlen_k).</span>
+<span class="sd">                ALiBi mask for non-causal would have shape (1, nheads, seqlen_q, seqlen_k)</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># Make sure that the last dimension is contiguous</span>
+        <span class="n">q</span><span class="p">,</span> <span class="n">kv</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span> <span class="k">if</span> <span class="n">x</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">x</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="p">[</span><span class="n">q</span><span class="p">,</span> <span class="n">kv</span><span class="p">]]</span>
+        <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span> <span class="n">ctx</span><span class="o">.</span><span class="n">softmax_scale</span> <span class="o">=</span> <span class="n">_flash_attn_forward</span><span class="p">(</span>
+            <span class="n">q</span><span class="p">,</span> <span class="n">kv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">],</span> <span class="n">kv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">],</span> <span class="n">bias</span><span class="o">=</span><span class="n">bias</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="n">causal</span><span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="n">softmax_scale</span>
+        <span class="p">)</span>
+        <span class="n">ctx</span><span class="o">.</span><span class="n">save_for_backward</span><span class="p">(</span><span class="n">q</span><span class="p">,</span> <span class="n">kv</span><span class="p">,</span> <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span> <span class="n">bias</span><span class="p">)</span>
+        <span class="n">ctx</span><span class="o">.</span><span class="n">causal</span> <span class="o">=</span> <span class="n">causal</span>
+        <span class="k">return</span> <span class="n">o</span></div>
+
+
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="FlashAttnKVPackedFunc.backward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc.backward">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">backward</span><span class="p">(</span><span class="n">ctx</span><span class="p">,</span> <span class="n">do</span><span class="p">):</span>
+        <span class="n">q</span><span class="p">,</span> <span class="n">kv</span><span class="p">,</span> <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span> <span class="n">bias</span> <span class="o">=</span> <span class="n">ctx</span><span class="o">.</span><span class="n">saved_tensors</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">ctx</span><span class="o">.</span><span class="n">needs_input_grad</span><span class="p">)</span> <span class="o">&gt;=</span> <span class="mi">3</span><span class="p">:</span>
+            <span class="k">assert</span> <span class="ow">not</span> <span class="n">ctx</span><span class="o">.</span><span class="n">needs_input_grad</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="s1">&#39;FlashAttention does not support bias gradient yet&#39;</span>
+        <span class="c1"># Triton&#39;s autotune causes the Tensor._version to change, and so Pytorch autograd</span>
+        <span class="c1"># does a memcpy. To avoid this we run in inference_mode, which doesn&#39;t track the version.</span>
+        <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">inference_mode</span><span class="p">():</span>
+            <span class="n">dq</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty_like</span><span class="p">(</span><span class="n">q</span><span class="p">)</span>
+            <span class="n">dkv</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty_like</span><span class="p">(</span><span class="n">kv</span><span class="p">)</span>
+            <span class="n">_flash_attn_backward</span><span class="p">(</span><span class="n">do</span><span class="p">,</span> <span class="n">q</span><span class="p">,</span> <span class="n">kv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">],</span> <span class="n">kv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">],</span> <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span>
+                                 <span class="n">dq</span><span class="p">,</span> <span class="n">dkv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">0</span><span class="p">],</span> <span class="n">dkv</span><span class="p">[:,</span> <span class="p">:,</span> <span class="mi">1</span><span class="p">],</span>
+                                 <span class="n">bias</span><span class="o">=</span><span class="n">bias</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="n">ctx</span><span class="o">.</span><span class="n">causal</span><span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="n">ctx</span><span class="o">.</span><span class="n">softmax_scale</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">dq</span><span class="p">,</span> <span class="n">dkv</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span></div>
+</div>
+
+
+
+<div class="viewcode-block" id="flash_attn_kvpacked_func">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_kvpacked_func">[docs]</a>
+<span class="n">flash_attn_kvpacked_func</span> <span class="o">=</span> <span class="n">FlashAttnKVPackedFunc</span><span class="o">.</span><span class="n">apply</span></div>
+
+
+
+<div class="viewcode-block" id="FlashAttnFunc">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">FlashAttnFunc</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">autograd</span><span class="o">.</span><span class="n">Function</span><span class="p">):</span>
+
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="FlashAttnFunc.forward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc.forward">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span><span class="n">ctx</span><span class="p">,</span> <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">            q: (batch_size, seqlen_q, nheads, headdim)</span>
+<span class="sd">            k, v: (batch_size, seqlen_k, nheads, headdim)</span>
+<span class="sd">            bias: optional, shape broadcastible to (batch, nheads, seqlen_q, seqlen_k).</span>
+<span class="sd">                For example, ALiBi mask for causal would have shape (1, nheads, 1, seqlen_k).</span>
+<span class="sd">                ALiBi mask for non-causal would have shape (1, nheads, seqlen_q, seqlen_k)</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="c1"># Make sure that the last dimension is contiguous</span>
+        <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span> <span class="k">if</span> <span class="n">x</span><span class="o">.</span><span class="n">stride</span><span class="p">(</span><span class="o">-</span><span class="mi">1</span><span class="p">)</span> <span class="o">==</span> <span class="mi">1</span> <span class="k">else</span> <span class="n">x</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="p">[</span><span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">]]</span>
+        <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span> <span class="n">ctx</span><span class="o">.</span><span class="n">softmax_scale</span> <span class="o">=</span> <span class="n">_flash_attn_forward</span><span class="p">(</span>
+            <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="n">bias</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="n">causal</span><span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="n">softmax_scale</span>
+        <span class="p">)</span>
+        <span class="n">ctx</span><span class="o">.</span><span class="n">save_for_backward</span><span class="p">(</span><span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span> <span class="n">bias</span><span class="p">)</span>
+        <span class="n">ctx</span><span class="o">.</span><span class="n">causal</span> <span class="o">=</span> <span class="n">causal</span>
+        <span class="k">return</span> <span class="n">o</span></div>
+
+
+    <span class="nd">@staticmethod</span>
+<div class="viewcode-block" id="FlashAttnFunc.backward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc.backward">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">backward</span><span class="p">(</span><span class="n">ctx</span><span class="p">,</span> <span class="n">do</span><span class="p">):</span>
+        <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span> <span class="n">bias</span> <span class="o">=</span> <span class="n">ctx</span><span class="o">.</span><span class="n">saved_tensors</span>
+        <span class="k">assert</span> <span class="ow">not</span> <span class="n">ctx</span><span class="o">.</span><span class="n">needs_input_grad</span><span class="p">[</span><span class="mi">3</span><span class="p">],</span> <span class="s1">&#39;FlashAttention does not support bias gradient yet&#39;</span>
+        <span class="c1"># Triton&#39;s autotune causes the Tensor._version to change, and so Pytorch autograd</span>
+        <span class="c1"># does a memcpy. To avoid this we run in inference_mode, which doesn&#39;t track the version.</span>
+        <span class="k">with</span> <span class="n">torch</span><span class="o">.</span><span class="n">inference_mode</span><span class="p">():</span>
+            <span class="n">dq</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty_like</span><span class="p">(</span><span class="n">q</span><span class="p">)</span>
+            <span class="n">dk</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty_like</span><span class="p">(</span><span class="n">k</span><span class="p">)</span>
+            <span class="n">dv</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">empty_like</span><span class="p">(</span><span class="n">v</span><span class="p">)</span>
+            <span class="n">_flash_attn_backward</span><span class="p">(</span><span class="n">do</span><span class="p">,</span> <span class="n">q</span><span class="p">,</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="n">o</span><span class="p">,</span> <span class="n">lse</span><span class="p">,</span> <span class="n">dq</span><span class="p">,</span> <span class="n">dk</span><span class="p">,</span> <span class="n">dv</span><span class="p">,</span>
+                                 <span class="n">bias</span><span class="o">=</span><span class="n">bias</span><span class="p">,</span> <span class="n">causal</span><span class="o">=</span><span class="n">ctx</span><span class="o">.</span><span class="n">causal</span><span class="p">,</span> <span class="n">softmax_scale</span><span class="o">=</span><span class="n">ctx</span><span class="o">.</span><span class="n">softmax_scale</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">dq</span><span class="p">,</span> <span class="n">dk</span><span class="p">,</span> <span class="n">dv</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="kc">None</span></div>
+</div>
+
+
+
+<div class="viewcode-block" id="flash_attn_func">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_func">[docs]</a>
+<span class="n">flash_attn_func</span> <span class="o">=</span> <span class="n">FlashAttnFunc</span><span class="o">.</span><span class="n">apply</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/llava_conversation_lib.html b/_modules/lmflow/utils/llava_conversation_lib.html
new file mode 100644
index 000000000..e12df3cb0
--- /dev/null
+++ b/_modules/lmflow/utils/llava_conversation_lib.html
@@ -0,0 +1,927 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.llava_conversation_lib &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/llava_conversation_lib';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.llava_conversation_lib</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.llava_conversation_lib</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span><span class="w"> </span><span class="nn">dataclasses</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">enum</span><span class="w"> </span><span class="kn">import</span> <span class="n">auto</span><span class="p">,</span> <span class="n">Enum</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Tuple</span>
+
+
+<div class="viewcode-block" id="SeparatorStyle">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.SeparatorStyle">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">SeparatorStyle</span><span class="p">(</span><span class="n">Enum</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Different separator style.&quot;&quot;&quot;</span>
+<div class="viewcode-block" id="SeparatorStyle.SINGLE">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.SeparatorStyle.SINGLE">[docs]</a>
+    <span class="n">SINGLE</span> <span class="o">=</span> <span class="n">auto</span><span class="p">()</span></div>
+
+<div class="viewcode-block" id="SeparatorStyle.TWO">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.SeparatorStyle.TWO">[docs]</a>
+    <span class="n">TWO</span> <span class="o">=</span> <span class="n">auto</span><span class="p">()</span></div>
+
+<div class="viewcode-block" id="SeparatorStyle.MPT">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.SeparatorStyle.MPT">[docs]</a>
+    <span class="n">MPT</span> <span class="o">=</span> <span class="n">auto</span><span class="p">()</span></div>
+
+<div class="viewcode-block" id="SeparatorStyle.PLAIN">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.SeparatorStyle.PLAIN">[docs]</a>
+    <span class="n">PLAIN</span> <span class="o">=</span> <span class="n">auto</span><span class="p">()</span></div>
+
+<div class="viewcode-block" id="SeparatorStyle.LLAMA_2">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.SeparatorStyle.LLAMA_2">[docs]</a>
+    <span class="n">LLAMA_2</span> <span class="o">=</span> <span class="n">auto</span><span class="p">()</span></div>
+</div>
+
+
+
+<span class="nd">@dataclasses</span><span class="o">.</span><span class="n">dataclass</span>
+<div class="viewcode-block" id="Conversation">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">Conversation</span><span class="p">:</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;A class that keeps all conversation history.&quot;&quot;&quot;</span>
+<div class="viewcode-block" id="Conversation.system">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.system">[docs]</a>
+    <span class="n">system</span><span class="p">:</span> <span class="nb">str</span></div>
+
+<div class="viewcode-block" id="Conversation.roles">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.roles">[docs]</a>
+    <span class="n">roles</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span></div>
+
+<div class="viewcode-block" id="Conversation.messages">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.messages">[docs]</a>
+    <span class="n">messages</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]</span></div>
+
+<div class="viewcode-block" id="Conversation.offset">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.offset">[docs]</a>
+    <span class="n">offset</span><span class="p">:</span> <span class="nb">int</span></div>
+
+<div class="viewcode-block" id="Conversation.sep_style">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.sep_style">[docs]</a>
+    <span class="n">sep_style</span><span class="p">:</span> <span class="n">SeparatorStyle</span> <span class="o">=</span> <span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">SINGLE</span></div>
+
+<div class="viewcode-block" id="Conversation.sep">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.sep">[docs]</a>
+    <span class="n">sep</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;###&quot;</span></div>
+
+<div class="viewcode-block" id="Conversation.sep2">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.sep2">[docs]</a>
+    <span class="n">sep2</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="kc">None</span></div>
+
+<div class="viewcode-block" id="Conversation.version">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.version">[docs]</a>
+    <span class="n">version</span><span class="p">:</span> <span class="nb">str</span> <span class="o">=</span> <span class="s2">&quot;Unknown&quot;</span></div>
+
+
+<div class="viewcode-block" id="Conversation.skip_next">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.skip_next">[docs]</a>
+    <span class="n">skip_next</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span></div>
+
+
+<div class="viewcode-block" id="Conversation.get_prompt">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.get_prompt">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_prompt</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="n">messages</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">messages</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">messages</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span> <span class="ow">and</span> <span class="nb">type</span><span class="p">(</span><span class="n">messages</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="mi">1</span><span class="p">])</span> <span class="ow">is</span> <span class="nb">tuple</span><span class="p">:</span>
+            <span class="n">messages</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">messages</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
+            <span class="n">init_role</span><span class="p">,</span> <span class="n">init_msg</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>
+            <span class="n">init_msg</span> <span class="o">=</span> <span class="n">init_msg</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;&lt;image&gt;&quot;</span><span class="p">,</span> <span class="s2">&quot;&quot;</span><span class="p">)</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span>
+            <span class="k">if</span> <span class="s1">&#39;mmtag&#39;</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">version</span><span class="p">:</span>
+                <span class="n">messages</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="n">init_role</span><span class="p">,</span> <span class="n">init_msg</span><span class="p">)</span>
+                <span class="n">messages</span><span class="o">.</span><span class="n">insert</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">roles</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="s2">&quot;&lt;Image&gt;&lt;image&gt;&lt;/Image&gt;&quot;</span><span class="p">))</span>
+                <span class="n">messages</span><span class="o">.</span><span class="n">insert</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">roles</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="s2">&quot;Received.&quot;</span><span class="p">))</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">messages</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="o">=</span> <span class="p">(</span><span class="n">init_role</span><span class="p">,</span> <span class="s2">&quot;&lt;image&gt;</span><span class="se">\n</span><span class="s2">&quot;</span> <span class="o">+</span> <span class="n">init_msg</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep_style</span> <span class="o">==</span> <span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">SINGLE</span><span class="p">:</span>
+            <span class="n">ret</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">system</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep</span>
+            <span class="k">for</span> <span class="n">role</span><span class="p">,</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">message</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">message</span><span class="p">)</span> <span class="ow">is</span> <span class="nb">tuple</span><span class="p">:</span>
+                        <span class="n">message</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">message</span>
+                    <span class="n">ret</span> <span class="o">+=</span> <span class="n">role</span> <span class="o">+</span> <span class="s2">&quot;: &quot;</span> <span class="o">+</span> <span class="n">message</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">ret</span> <span class="o">+=</span> <span class="n">role</span> <span class="o">+</span> <span class="s2">&quot;:&quot;</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep_style</span> <span class="o">==</span> <span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">TWO</span><span class="p">:</span>
+            <span class="n">seps</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">sep</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep2</span><span class="p">]</span>
+            <span class="n">ret</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">system</span> <span class="o">+</span> <span class="n">seps</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
+            <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="p">(</span><span class="n">role</span><span class="p">,</span> <span class="n">message</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">messages</span><span class="p">):</span>
+                <span class="k">if</span> <span class="n">message</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">message</span><span class="p">)</span> <span class="ow">is</span> <span class="nb">tuple</span><span class="p">:</span>
+                        <span class="n">message</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">message</span>
+                    <span class="n">ret</span> <span class="o">+=</span> <span class="n">role</span> <span class="o">+</span> <span class="s2">&quot;: &quot;</span> <span class="o">+</span> <span class="n">message</span> <span class="o">+</span> <span class="n">seps</span><span class="p">[</span><span class="n">i</span> <span class="o">%</span> <span class="mi">2</span><span class="p">]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">ret</span> <span class="o">+=</span> <span class="n">role</span> <span class="o">+</span> <span class="s2">&quot;:&quot;</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep_style</span> <span class="o">==</span> <span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">MPT</span><span class="p">:</span>
+            <span class="n">ret</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">system</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep</span>
+            <span class="k">for</span> <span class="n">role</span><span class="p">,</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span><span class="p">:</span>
+                <span class="k">if</span> <span class="n">message</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">message</span><span class="p">)</span> <span class="ow">is</span> <span class="nb">tuple</span><span class="p">:</span>
+                        <span class="n">message</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">message</span>
+                    <span class="n">ret</span> <span class="o">+=</span> <span class="n">role</span> <span class="o">+</span> <span class="n">message</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">ret</span> <span class="o">+=</span> <span class="n">role</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep_style</span> <span class="o">==</span> <span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">LLAMA_2</span><span class="p">:</span>
+            <span class="n">wrap_sys</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">msg</span><span class="p">:</span> <span class="sa">f</span><span class="s2">&quot;&lt;&lt;SYS&gt;&gt;</span><span class="se">\n</span><span class="si">{</span><span class="n">msg</span><span class="si">}</span><span class="se">\n</span><span class="s2">&lt;&lt;/SYS&gt;&gt;</span><span class="se">\n\n</span><span class="s2">&quot;</span>
+            <span class="n">wrap_inst</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">msg</span><span class="p">:</span> <span class="sa">f</span><span class="s2">&quot;[INST] </span><span class="si">{</span><span class="n">msg</span><span class="si">}</span><span class="s2"> [/INST]&quot;</span>
+            <span class="n">ret</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
+
+            <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="p">(</span><span class="n">role</span><span class="p">,</span> <span class="n">message</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">messages</span><span class="p">):</span>
+                <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="k">assert</span> <span class="n">message</span><span class="p">,</span> <span class="s2">&quot;first message should not be none&quot;</span>
+                    <span class="k">assert</span> <span class="n">role</span> <span class="o">==</span> <span class="bp">self</span><span class="o">.</span><span class="n">roles</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="s2">&quot;first message should come from user&quot;</span>
+                <span class="k">if</span> <span class="n">message</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">message</span><span class="p">)</span> <span class="ow">is</span> <span class="nb">tuple</span><span class="p">:</span>
+                        <span class="n">message</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">message</span>
+                    <span class="k">if</span> <span class="n">i</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span> <span class="n">message</span> <span class="o">=</span> <span class="n">wrap_sys</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">system</span><span class="p">)</span> <span class="o">+</span> <span class="n">message</span>
+                    <span class="k">if</span> <span class="n">i</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="n">message</span> <span class="o">=</span> <span class="n">wrap_inst</span><span class="p">(</span><span class="n">message</span><span class="p">)</span>
+                        <span class="n">ret</span> <span class="o">+=</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep</span> <span class="o">+</span> <span class="n">message</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">ret</span> <span class="o">+=</span> <span class="s2">&quot; &quot;</span> <span class="o">+</span> <span class="n">message</span> <span class="o">+</span> <span class="s2">&quot; &quot;</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep2</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">ret</span> <span class="o">+=</span> <span class="s2">&quot;&quot;</span>
+            <span class="n">ret</span> <span class="o">=</span> <span class="n">ret</span><span class="o">.</span><span class="n">lstrip</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">sep</span><span class="p">)</span>
+        <span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep_style</span> <span class="o">==</span> <span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">PLAIN</span><span class="p">:</span>
+            <span class="n">seps</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">sep</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep2</span><span class="p">]</span>
+            <span class="n">ret</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">system</span>
+            <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="p">(</span><span class="n">role</span><span class="p">,</span> <span class="n">message</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="n">messages</span><span class="p">):</span>
+                <span class="k">if</span> <span class="n">message</span><span class="p">:</span>
+                    <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">message</span><span class="p">)</span> <span class="ow">is</span> <span class="nb">tuple</span><span class="p">:</span>
+                        <span class="n">message</span><span class="p">,</span> <span class="n">_</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">message</span>
+                    <span class="n">ret</span> <span class="o">+=</span> <span class="n">message</span> <span class="o">+</span> <span class="n">seps</span><span class="p">[</span><span class="n">i</span> <span class="o">%</span> <span class="mi">2</span><span class="p">]</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">ret</span> <span class="o">+=</span> <span class="s2">&quot;&quot;</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Invalid style: </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">sep_style</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">ret</span></div>
+
+
+<div class="viewcode-block" id="Conversation.append_message">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.append_message">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">append_message</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">role</span><span class="p">,</span> <span class="n">message</span><span class="p">):</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">messages</span><span class="o">.</span><span class="n">append</span><span class="p">([</span><span class="n">role</span><span class="p">,</span> <span class="n">message</span><span class="p">])</span></div>
+
+
+<div class="viewcode-block" id="Conversation.get_images">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.get_images">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_images</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">return_pil</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
+        <span class="n">images</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="p">(</span><span class="n">role</span><span class="p">,</span> <span class="n">msg</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">messages</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">offset</span><span class="p">:]):</span>
+            <span class="k">if</span> <span class="n">i</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">msg</span><span class="p">)</span> <span class="ow">is</span> <span class="nb">tuple</span><span class="p">:</span>
+                    <span class="kn">import</span><span class="w"> </span><span class="nn">base64</span>
+                    <span class="kn">from</span><span class="w"> </span><span class="nn">io</span><span class="w"> </span><span class="kn">import</span> <span class="n">BytesIO</span>
+                    <span class="kn">from</span><span class="w"> </span><span class="nn">PIL</span><span class="w"> </span><span class="kn">import</span> <span class="n">Image</span>
+                    <span class="n">msg</span><span class="p">,</span> <span class="n">image</span><span class="p">,</span> <span class="n">image_process_mode</span> <span class="o">=</span> <span class="n">msg</span>
+                    <span class="k">if</span> <span class="n">image_process_mode</span> <span class="o">==</span> <span class="s2">&quot;Pad&quot;</span><span class="p">:</span>
+                        <span class="k">def</span><span class="w"> </span><span class="nf">expand2square</span><span class="p">(</span><span class="n">pil_img</span><span class="p">,</span> <span class="n">background_color</span><span class="o">=</span><span class="p">(</span><span class="mi">122</span><span class="p">,</span> <span class="mi">116</span><span class="p">,</span> <span class="mi">104</span><span class="p">)):</span>
+                            <span class="n">width</span><span class="p">,</span> <span class="n">height</span> <span class="o">=</span> <span class="n">pil_img</span><span class="o">.</span><span class="n">size</span>
+                            <span class="k">if</span> <span class="n">width</span> <span class="o">==</span> <span class="n">height</span><span class="p">:</span>
+                                <span class="k">return</span> <span class="n">pil_img</span>
+                            <span class="k">elif</span> <span class="n">width</span> <span class="o">&gt;</span> <span class="n">height</span><span class="p">:</span>
+                                <span class="n">result</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">new</span><span class="p">(</span><span class="n">pil_img</span><span class="o">.</span><span class="n">mode</span><span class="p">,</span> <span class="p">(</span><span class="n">width</span><span class="p">,</span> <span class="n">width</span><span class="p">),</span> <span class="n">background_color</span><span class="p">)</span>
+                                <span class="n">result</span><span class="o">.</span><span class="n">paste</span><span class="p">(</span><span class="n">pil_img</span><span class="p">,</span> <span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="p">(</span><span class="n">width</span> <span class="o">-</span> <span class="n">height</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span><span class="p">))</span>
+                                <span class="k">return</span> <span class="n">result</span>
+                            <span class="k">else</span><span class="p">:</span>
+                                <span class="n">result</span> <span class="o">=</span> <span class="n">Image</span><span class="o">.</span><span class="n">new</span><span class="p">(</span><span class="n">pil_img</span><span class="o">.</span><span class="n">mode</span><span class="p">,</span> <span class="p">(</span><span class="n">height</span><span class="p">,</span> <span class="n">height</span><span class="p">),</span> <span class="n">background_color</span><span class="p">)</span>
+                                <span class="n">result</span><span class="o">.</span><span class="n">paste</span><span class="p">(</span><span class="n">pil_img</span><span class="p">,</span> <span class="p">((</span><span class="n">height</span> <span class="o">-</span> <span class="n">width</span><span class="p">)</span> <span class="o">//</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">))</span>
+                                <span class="k">return</span> <span class="n">result</span>
+                        <span class="n">image</span> <span class="o">=</span> <span class="n">expand2square</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+                    <span class="k">elif</span> <span class="n">image_process_mode</span> <span class="o">==</span> <span class="s2">&quot;Crop&quot;</span><span class="p">:</span>
+                        <span class="k">pass</span>
+                    <span class="k">elif</span> <span class="n">image_process_mode</span> <span class="o">==</span> <span class="s2">&quot;Resize&quot;</span><span class="p">:</span>
+                        <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">resize</span><span class="p">((</span><span class="mi">336</span><span class="p">,</span> <span class="mi">336</span><span class="p">))</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Invalid image_process_mode: </span><span class="si">{</span><span class="n">image_process_mode</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
+                    <span class="n">max_hw</span><span class="p">,</span> <span class="n">min_hw</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">size</span><span class="p">),</span> <span class="nb">min</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">size</span><span class="p">)</span>
+                    <span class="n">aspect_ratio</span> <span class="o">=</span> <span class="n">max_hw</span> <span class="o">/</span> <span class="n">min_hw</span>
+                    <span class="n">max_len</span><span class="p">,</span> <span class="n">min_len</span> <span class="o">=</span> <span class="mi">800</span><span class="p">,</span> <span class="mi">400</span>
+                    <span class="n">shortest_edge</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">min</span><span class="p">(</span><span class="n">max_len</span> <span class="o">/</span> <span class="n">aspect_ratio</span><span class="p">,</span> <span class="n">min_len</span><span class="p">,</span> <span class="n">min_hw</span><span class="p">))</span>
+                    <span class="n">longest_edge</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">shortest_edge</span> <span class="o">*</span> <span class="n">aspect_ratio</span><span class="p">)</span>
+                    <span class="n">W</span><span class="p">,</span> <span class="n">H</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">size</span>
+                    <span class="k">if</span> <span class="n">H</span> <span class="o">&gt;</span> <span class="n">W</span><span class="p">:</span>
+                        <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">longest_edge</span><span class="p">,</span> <span class="n">shortest_edge</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">shortest_edge</span><span class="p">,</span> <span class="n">longest_edge</span>
+                    <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">resize</span><span class="p">((</span><span class="n">W</span><span class="p">,</span> <span class="n">H</span><span class="p">))</span>
+                    <span class="k">if</span> <span class="n">return_pil</span><span class="p">:</span>
+                        <span class="n">images</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">buffered</span> <span class="o">=</span> <span class="n">BytesIO</span><span class="p">()</span>
+                        <span class="n">image</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="n">buffered</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="s2">&quot;PNG&quot;</span><span class="p">)</span>
+                        <span class="n">img_b64_str</span> <span class="o">=</span> <span class="n">base64</span><span class="o">.</span><span class="n">b64encode</span><span class="p">(</span><span class="n">buffered</span><span class="o">.</span><span class="n">getvalue</span><span class="p">())</span><span class="o">.</span><span class="n">decode</span><span class="p">()</span>
+                        <span class="n">images</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">img_b64_str</span><span class="p">)</span>
+        <span class="k">return</span> <span class="n">images</span></div>
+
+
+<div class="viewcode-block" id="Conversation.to_gradio_chatbot">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.to_gradio_chatbot">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">to_gradio_chatbot</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="n">ret</span> <span class="o">=</span> <span class="p">[]</span>
+        <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="p">(</span><span class="n">role</span><span class="p">,</span> <span class="n">msg</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">messages</span><span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">offset</span><span class="p">:]):</span>
+            <span class="k">if</span> <span class="n">i</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
+                <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">msg</span><span class="p">)</span> <span class="ow">is</span> <span class="nb">tuple</span><span class="p">:</span>
+                    <span class="kn">import</span><span class="w"> </span><span class="nn">base64</span>
+                    <span class="kn">from</span><span class="w"> </span><span class="nn">io</span><span class="w"> </span><span class="kn">import</span> <span class="n">BytesIO</span>
+                    <span class="n">msg</span><span class="p">,</span> <span class="n">image</span><span class="p">,</span> <span class="n">image_process_mode</span> <span class="o">=</span> <span class="n">msg</span>
+                    <span class="n">max_hw</span><span class="p">,</span> <span class="n">min_hw</span> <span class="o">=</span> <span class="nb">max</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">size</span><span class="p">),</span> <span class="nb">min</span><span class="p">(</span><span class="n">image</span><span class="o">.</span><span class="n">size</span><span class="p">)</span>
+                    <span class="n">aspect_ratio</span> <span class="o">=</span> <span class="n">max_hw</span> <span class="o">/</span> <span class="n">min_hw</span>
+                    <span class="n">max_len</span><span class="p">,</span> <span class="n">min_len</span> <span class="o">=</span> <span class="mi">800</span><span class="p">,</span> <span class="mi">400</span>
+                    <span class="n">shortest_edge</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="nb">min</span><span class="p">(</span><span class="n">max_len</span> <span class="o">/</span> <span class="n">aspect_ratio</span><span class="p">,</span> <span class="n">min_len</span><span class="p">,</span> <span class="n">min_hw</span><span class="p">))</span>
+                    <span class="n">longest_edge</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">shortest_edge</span> <span class="o">*</span> <span class="n">aspect_ratio</span><span class="p">)</span>
+                    <span class="n">W</span><span class="p">,</span> <span class="n">H</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">size</span>
+                    <span class="k">if</span> <span class="n">H</span> <span class="o">&gt;</span> <span class="n">W</span><span class="p">:</span>
+                        <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">longest_edge</span><span class="p">,</span> <span class="n">shortest_edge</span>
+                    <span class="k">else</span><span class="p">:</span>
+                        <span class="n">H</span><span class="p">,</span> <span class="n">W</span> <span class="o">=</span> <span class="n">shortest_edge</span><span class="p">,</span> <span class="n">longest_edge</span>
+                    <span class="n">image</span> <span class="o">=</span> <span class="n">image</span><span class="o">.</span><span class="n">resize</span><span class="p">((</span><span class="n">W</span><span class="p">,</span> <span class="n">H</span><span class="p">))</span>
+                    <span class="n">buffered</span> <span class="o">=</span> <span class="n">BytesIO</span><span class="p">()</span>
+                    <span class="n">image</span><span class="o">.</span><span class="n">save</span><span class="p">(</span><span class="n">buffered</span><span class="p">,</span> <span class="nb">format</span><span class="o">=</span><span class="s2">&quot;JPEG&quot;</span><span class="p">)</span>
+                    <span class="n">img_b64_str</span> <span class="o">=</span> <span class="n">base64</span><span class="o">.</span><span class="n">b64encode</span><span class="p">(</span><span class="n">buffered</span><span class="o">.</span><span class="n">getvalue</span><span class="p">())</span><span class="o">.</span><span class="n">decode</span><span class="p">()</span>
+                    <span class="n">img_str</span> <span class="o">=</span> <span class="sa">f</span><span class="s1">&#39;&lt;img src=&quot;data:image/png;base64,</span><span class="si">{</span><span class="n">img_b64_str</span><span class="si">}</span><span class="s1">&quot; alt=&quot;user upload image&quot; /&gt;&#39;</span>
+                    <span class="n">ret</span><span class="o">.</span><span class="n">append</span><span class="p">([</span><span class="n">img_str</span><span class="p">,</span> <span class="kc">None</span><span class="p">])</span>
+                    <span class="n">msg</span> <span class="o">=</span> <span class="n">msg</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s1">&#39;&lt;image&gt;&#39;</span><span class="p">,</span> <span class="s1">&#39;&#39;</span><span class="p">)</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span>
+                    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">msg</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+                        <span class="n">ret</span><span class="o">.</span><span class="n">append</span><span class="p">([</span><span class="n">msg</span><span class="p">,</span> <span class="kc">None</span><span class="p">])</span>
+                <span class="k">else</span><span class="p">:</span>
+                    <span class="n">ret</span><span class="o">.</span><span class="n">append</span><span class="p">([</span><span class="n">msg</span><span class="p">,</span> <span class="kc">None</span><span class="p">])</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">ret</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">][</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span> <span class="o">=</span> <span class="n">msg</span>
+        <span class="k">return</span> <span class="n">ret</span></div>
+
+
+<div class="viewcode-block" id="Conversation.copy">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.copy">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">copy</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">return</span> <span class="n">Conversation</span><span class="p">(</span>
+            <span class="n">system</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">system</span><span class="p">,</span>
+            <span class="n">roles</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">roles</span><span class="p">,</span>
+            <span class="n">messages</span><span class="o">=</span><span class="p">[[</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">]</span> <span class="k">for</span> <span class="n">x</span><span class="p">,</span> <span class="n">y</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">messages</span><span class="p">],</span>
+            <span class="n">offset</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">offset</span><span class="p">,</span>
+            <span class="n">sep_style</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">sep_style</span><span class="p">,</span>
+            <span class="n">sep</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">sep</span><span class="p">,</span>
+            <span class="n">sep2</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">sep2</span><span class="p">,</span>
+            <span class="n">version</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">version</span><span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="Conversation.dict">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.dict">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">dict</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">get_images</span><span class="p">())</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">return</span> <span class="p">{</span>
+                <span class="s2">&quot;system&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">system</span><span class="p">,</span>
+                <span class="s2">&quot;roles&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">roles</span><span class="p">,</span>
+                <span class="s2">&quot;messages&quot;</span><span class="p">:</span> <span class="p">[[</span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> <span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="n">y</span><span class="p">)</span> <span class="ow">is</span> <span class="nb">tuple</span> <span class="k">else</span> <span class="n">y</span><span class="p">]</span> <span class="k">for</span> <span class="n">x</span><span class="p">,</span> <span class="n">y</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">messages</span><span class="p">],</span>
+                <span class="s2">&quot;offset&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">offset</span><span class="p">,</span>
+                <span class="s2">&quot;sep&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep</span><span class="p">,</span>
+                <span class="s2">&quot;sep2&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep2</span><span class="p">,</span>
+            <span class="p">}</span>
+        <span class="k">return</span> <span class="p">{</span>
+            <span class="s2">&quot;system&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">system</span><span class="p">,</span>
+            <span class="s2">&quot;roles&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">roles</span><span class="p">,</span>
+            <span class="s2">&quot;messages&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">messages</span><span class="p">,</span>
+            <span class="s2">&quot;offset&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">offset</span><span class="p">,</span>
+            <span class="s2">&quot;sep&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep</span><span class="p">,</span>
+            <span class="s2">&quot;sep2&quot;</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">sep2</span><span class="p">,</span>
+        <span class="p">}</span></div>
+</div>
+
+
+
+<div class="viewcode-block" id="conv_vicuna_v0">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_vicuna_v0">[docs]</a>
+<span class="n">conv_vicuna_v0</span> <span class="o">=</span> <span class="n">Conversation</span><span class="p">(</span>
+    <span class="n">system</span><span class="o">=</span><span class="s2">&quot;A chat between a curious human and an artificial intelligence assistant. &quot;</span>
+           <span class="s2">&quot;The assistant gives helpful, detailed, and polite answers to the human&#39;s questions.&quot;</span><span class="p">,</span>
+    <span class="n">roles</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;Human&quot;</span><span class="p">,</span> <span class="s2">&quot;Assistant&quot;</span><span class="p">),</span>
+    <span class="n">messages</span><span class="o">=</span><span class="p">(</span>
+        <span class="p">(</span><span class="s2">&quot;Human&quot;</span><span class="p">,</span> <span class="s2">&quot;What are the key differences between renewable and non-renewable energy sources?&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;Assistant&quot;</span><span class="p">,</span>
+            <span class="s2">&quot;Renewable energy sources are those that can be replenished naturally in a relatively &quot;</span>
+            <span class="s2">&quot;short amount of time, such as solar, wind, hydro, geothermal, and biomass. &quot;</span>
+            <span class="s2">&quot;Non-renewable energy sources, on the other hand, are finite and will eventually be &quot;</span>
+            <span class="s2">&quot;depleted, such as coal, oil, and natural gas. Here are some key differences between &quot;</span>
+            <span class="s2">&quot;renewable and non-renewable energy sources:</span><span class="se">\n</span><span class="s2">&quot;</span>
+            <span class="s2">&quot;1. Availability: Renewable energy sources are virtually inexhaustible, while non-renewable &quot;</span>
+            <span class="s2">&quot;energy sources are finite and will eventually run out.</span><span class="se">\n</span><span class="s2">&quot;</span>
+            <span class="s2">&quot;2. Environmental impact: Renewable energy sources have a much lower environmental impact &quot;</span>
+            <span class="s2">&quot;than non-renewable sources, which can lead to air and water pollution, greenhouse gas emissions, &quot;</span>
+            <span class="s2">&quot;and other negative effects.</span><span class="se">\n</span><span class="s2">&quot;</span>
+            <span class="s2">&quot;3. Cost: Renewable energy sources can be more expensive to initially set up, but they typically &quot;</span>
+            <span class="s2">&quot;have lower operational costs than non-renewable sources.</span><span class="se">\n</span><span class="s2">&quot;</span>
+            <span class="s2">&quot;4. Reliability: Renewable energy sources are often more reliable and can be used in more remote &quot;</span>
+            <span class="s2">&quot;locations than non-renewable sources.</span><span class="se">\n</span><span class="s2">&quot;</span>
+            <span class="s2">&quot;5. Flexibility: Renewable energy sources are often more flexible and can be adapted to different &quot;</span>
+            <span class="s2">&quot;situations and needs, while non-renewable sources are more rigid and inflexible.</span><span class="se">\n</span><span class="s2">&quot;</span>
+            <span class="s2">&quot;6. Sustainability: Renewable energy sources are more sustainable over the long term, while &quot;</span>
+            <span class="s2">&quot;non-renewable sources are not, and their depletion can lead to economic and social instability.</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">)</span>
+    <span class="p">),</span>
+    <span class="n">offset</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span>
+    <span class="n">sep_style</span><span class="o">=</span><span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">SINGLE</span><span class="p">,</span>
+    <span class="n">sep</span><span class="o">=</span><span class="s2">&quot;###&quot;</span><span class="p">,</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="conv_vicuna_v1">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_vicuna_v1">[docs]</a>
+<span class="n">conv_vicuna_v1</span> <span class="o">=</span> <span class="n">Conversation</span><span class="p">(</span>
+    <span class="n">system</span><span class="o">=</span><span class="s2">&quot;A chat between a curious user and an artificial intelligence assistant. &quot;</span>
+    <span class="s2">&quot;The assistant gives helpful, detailed, and polite answers to the user&#39;s questions.&quot;</span><span class="p">,</span>
+    <span class="n">roles</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;USER&quot;</span><span class="p">,</span> <span class="s2">&quot;ASSISTANT&quot;</span><span class="p">),</span>
+    <span class="n">version</span><span class="o">=</span><span class="s2">&quot;v1&quot;</span><span class="p">,</span>
+    <span class="n">messages</span><span class="o">=</span><span class="p">(),</span>
+    <span class="n">offset</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+    <span class="n">sep_style</span><span class="o">=</span><span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">TWO</span><span class="p">,</span>
+    <span class="n">sep</span><span class="o">=</span><span class="s2">&quot; &quot;</span><span class="p">,</span>
+    <span class="n">sep2</span><span class="o">=</span><span class="s2">&quot;&lt;/s&gt;&quot;</span><span class="p">,</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="conv_llama_2">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llama_2">[docs]</a>
+<span class="n">conv_llama_2</span> <span class="o">=</span> <span class="n">Conversation</span><span class="p">(</span>
+    <span class="n">system</span><span class="o">=</span><span class="s2">&quot;&quot;&quot;You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe.  Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature.</span>
+
+<span class="s2">If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don&#39;t know the answer to a question, please don&#39;t share false information.&quot;&quot;&quot;</span><span class="p">,</span>
+    <span class="n">roles</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;USER&quot;</span><span class="p">,</span> <span class="s2">&quot;ASSISTANT&quot;</span><span class="p">),</span>
+    <span class="n">version</span><span class="o">=</span><span class="s2">&quot;llama_v2&quot;</span><span class="p">,</span>
+    <span class="n">messages</span><span class="o">=</span><span class="p">(),</span>
+    <span class="n">offset</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+    <span class="n">sep_style</span><span class="o">=</span><span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">LLAMA_2</span><span class="p">,</span>
+    <span class="n">sep</span><span class="o">=</span><span class="s2">&quot;&lt;s&gt;&quot;</span><span class="p">,</span>
+    <span class="n">sep2</span><span class="o">=</span><span class="s2">&quot;&lt;/s&gt;&quot;</span><span class="p">,</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="conv_llava_llama_2">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llava_llama_2">[docs]</a>
+<span class="n">conv_llava_llama_2</span> <span class="o">=</span> <span class="n">Conversation</span><span class="p">(</span>
+    <span class="n">system</span><span class="o">=</span><span class="s2">&quot;You are a helpful language and vision assistant. &quot;</span>
+           <span class="s2">&quot;You are able to understand the visual content that the user provides, &quot;</span>
+           <span class="s2">&quot;and assist the user with a variety of tasks using natural language.&quot;</span><span class="p">,</span>
+    <span class="n">roles</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;USER&quot;</span><span class="p">,</span> <span class="s2">&quot;ASSISTANT&quot;</span><span class="p">),</span>
+    <span class="n">version</span><span class="o">=</span><span class="s2">&quot;llama_v2&quot;</span><span class="p">,</span>
+    <span class="n">messages</span><span class="o">=</span><span class="p">(),</span>
+    <span class="n">offset</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+    <span class="n">sep_style</span><span class="o">=</span><span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">LLAMA_2</span><span class="p">,</span>
+    <span class="n">sep</span><span class="o">=</span><span class="s2">&quot;&lt;s&gt;&quot;</span><span class="p">,</span>
+    <span class="n">sep2</span><span class="o">=</span><span class="s2">&quot;&lt;/s&gt;&quot;</span><span class="p">,</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="conv_mpt">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_mpt">[docs]</a>
+<span class="n">conv_mpt</span> <span class="o">=</span> <span class="n">Conversation</span><span class="p">(</span>
+    <span class="n">system</span><span class="o">=</span><span class="s2">&quot;&quot;&quot;&lt;|im_start|&gt;system</span>
+<span class="s2">A conversation between a user and an LLM-based AI assistant. The assistant gives helpful and honest answers.&quot;&quot;&quot;</span><span class="p">,</span>
+    <span class="n">roles</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;&lt;|im_start|&gt;user</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">,</span> <span class="s2">&quot;&lt;|im_start|&gt;assistant</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">),</span>
+    <span class="n">version</span><span class="o">=</span><span class="s2">&quot;mpt&quot;</span><span class="p">,</span>
+    <span class="n">messages</span><span class="o">=</span><span class="p">(),</span>
+    <span class="n">offset</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+    <span class="n">sep_style</span><span class="o">=</span><span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">MPT</span><span class="p">,</span>
+    <span class="n">sep</span><span class="o">=</span><span class="s2">&quot;&lt;|im_end|&gt;&quot;</span><span class="p">,</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="conv_llava_plain">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llava_plain">[docs]</a>
+<span class="n">conv_llava_plain</span> <span class="o">=</span> <span class="n">Conversation</span><span class="p">(</span>
+    <span class="n">system</span><span class="o">=</span><span class="s2">&quot;&quot;</span><span class="p">,</span>
+    <span class="n">roles</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;&quot;</span><span class="p">,</span> <span class="s2">&quot;&quot;</span><span class="p">),</span>
+    <span class="n">messages</span><span class="o">=</span><span class="p">(</span>
+    <span class="p">),</span>
+    <span class="n">offset</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+    <span class="n">sep_style</span><span class="o">=</span><span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">PLAIN</span><span class="p">,</span>
+    <span class="n">sep</span><span class="o">=</span><span class="s2">&quot;</span><span class="se">\n</span><span class="s2">&quot;</span><span class="p">,</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="conv_llava_v0">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llava_v0">[docs]</a>
+<span class="n">conv_llava_v0</span> <span class="o">=</span> <span class="n">Conversation</span><span class="p">(</span>
+    <span class="n">system</span><span class="o">=</span><span class="s2">&quot;A chat between a curious human and an artificial intelligence assistant. &quot;</span>
+           <span class="s2">&quot;The assistant gives helpful, detailed, and polite answers to the human&#39;s questions.&quot;</span><span class="p">,</span>
+    <span class="n">roles</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;Human&quot;</span><span class="p">,</span> <span class="s2">&quot;Assistant&quot;</span><span class="p">),</span>
+    <span class="n">messages</span><span class="o">=</span><span class="p">(</span>
+        <span class="p">(</span><span class="s2">&quot;Human&quot;</span><span class="p">,</span> <span class="s2">&quot;Hi!&quot;</span><span class="p">),</span>
+        <span class="p">(</span><span class="s2">&quot;Assistant&quot;</span><span class="p">,</span> <span class="s2">&quot;Hi there! How can I help you today?&quot;</span><span class="p">)</span>
+    <span class="p">),</span>
+    <span class="n">offset</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span>
+    <span class="n">sep_style</span><span class="o">=</span><span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">SINGLE</span><span class="p">,</span>
+    <span class="n">sep</span><span class="o">=</span><span class="s2">&quot;###&quot;</span><span class="p">,</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="conv_llava_v0_mmtag">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llava_v0_mmtag">[docs]</a>
+<span class="n">conv_llava_v0_mmtag</span> <span class="o">=</span> <span class="n">Conversation</span><span class="p">(</span>
+    <span class="n">system</span><span class="o">=</span><span class="s2">&quot;A chat between a curious user and an artificial intelligence assistant. &quot;</span>
+           <span class="s2">&quot;The assistant is able to understand the visual content that the user provides, and assist the user with a variety of tasks using natural language.&quot;</span>
+           <span class="s2">&quot;The visual content will be provided with the following format: &lt;Image&gt;visual content&lt;/Image&gt;.&quot;</span><span class="p">,</span>
+    <span class="n">roles</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;Human&quot;</span><span class="p">,</span> <span class="s2">&quot;Assistant&quot;</span><span class="p">),</span>
+    <span class="n">messages</span><span class="o">=</span><span class="p">(</span>
+    <span class="p">),</span>
+    <span class="n">offset</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+    <span class="n">sep_style</span><span class="o">=</span><span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">SINGLE</span><span class="p">,</span>
+    <span class="n">sep</span><span class="o">=</span><span class="s2">&quot;###&quot;</span><span class="p">,</span>
+    <span class="n">version</span><span class="o">=</span><span class="s2">&quot;v0_mmtag&quot;</span><span class="p">,</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="conv_llava_v1">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llava_v1">[docs]</a>
+<span class="n">conv_llava_v1</span> <span class="o">=</span> <span class="n">Conversation</span><span class="p">(</span>
+    <span class="n">system</span><span class="o">=</span><span class="s2">&quot;A chat between a curious human and an artificial intelligence assistant. &quot;</span>
+           <span class="s2">&quot;The assistant gives helpful, detailed, and polite answers to the human&#39;s questions.&quot;</span><span class="p">,</span>
+    <span class="n">roles</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;USER&quot;</span><span class="p">,</span> <span class="s2">&quot;ASSISTANT&quot;</span><span class="p">),</span>
+    <span class="n">version</span><span class="o">=</span><span class="s2">&quot;v1&quot;</span><span class="p">,</span>
+    <span class="n">messages</span><span class="o">=</span><span class="p">(),</span>
+    <span class="n">offset</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+    <span class="n">sep_style</span><span class="o">=</span><span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">TWO</span><span class="p">,</span>
+    <span class="n">sep</span><span class="o">=</span><span class="s2">&quot; &quot;</span><span class="p">,</span>
+    <span class="n">sep2</span><span class="o">=</span><span class="s2">&quot;&lt;/s&gt;&quot;</span><span class="p">,</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="conv_llava_v1_mmtag">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llava_v1_mmtag">[docs]</a>
+<span class="n">conv_llava_v1_mmtag</span> <span class="o">=</span> <span class="n">Conversation</span><span class="p">(</span>
+    <span class="n">system</span><span class="o">=</span><span class="s2">&quot;A chat between a curious user and an artificial intelligence assistant. &quot;</span>
+           <span class="s2">&quot;The assistant is able to understand the visual content that the user provides, and assist the user with a variety of tasks using natural language.&quot;</span>
+           <span class="s2">&quot;The visual content will be provided with the following format: &lt;Image&gt;visual content&lt;/Image&gt;.&quot;</span><span class="p">,</span>
+    <span class="n">roles</span><span class="o">=</span><span class="p">(</span><span class="s2">&quot;USER&quot;</span><span class="p">,</span> <span class="s2">&quot;ASSISTANT&quot;</span><span class="p">),</span>
+    <span class="n">messages</span><span class="o">=</span><span class="p">(),</span>
+    <span class="n">offset</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+    <span class="n">sep_style</span><span class="o">=</span><span class="n">SeparatorStyle</span><span class="o">.</span><span class="n">TWO</span><span class="p">,</span>
+    <span class="n">sep</span><span class="o">=</span><span class="s2">&quot; &quot;</span><span class="p">,</span>
+    <span class="n">sep2</span><span class="o">=</span><span class="s2">&quot;&lt;/s&gt;&quot;</span><span class="p">,</span>
+    <span class="n">version</span><span class="o">=</span><span class="s2">&quot;v1_mmtag&quot;</span><span class="p">,</span>
+<span class="p">)</span></div>
+
+
+<div class="viewcode-block" id="default_conversation">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.default_conversation">[docs]</a>
+<span class="n">default_conversation</span> <span class="o">=</span> <span class="n">conv_vicuna_v1</span> <span class="c1"># currently only support the v1 version</span></div>
+
+<div class="viewcode-block" id="conv_templates">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_templates">[docs]</a>
+<span class="n">conv_templates</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s2">&quot;default&quot;</span><span class="p">:</span> <span class="n">conv_vicuna_v0</span><span class="p">,</span>
+    <span class="s2">&quot;v0&quot;</span><span class="p">:</span> <span class="n">conv_vicuna_v0</span><span class="p">,</span>
+    <span class="s2">&quot;v1&quot;</span><span class="p">:</span> <span class="n">conv_vicuna_v1</span><span class="p">,</span>
+    <span class="s2">&quot;vicuna_v1&quot;</span><span class="p">:</span> <span class="n">conv_vicuna_v1</span><span class="p">,</span>
+    <span class="s2">&quot;llama_2&quot;</span><span class="p">:</span> <span class="n">conv_llama_2</span><span class="p">,</span>
+
+    <span class="s2">&quot;plain&quot;</span><span class="p">:</span> <span class="n">conv_llava_plain</span><span class="p">,</span>
+    <span class="s2">&quot;v0_plain&quot;</span><span class="p">:</span> <span class="n">conv_llava_plain</span><span class="p">,</span>
+    <span class="s2">&quot;llava_v0&quot;</span><span class="p">:</span> <span class="n">conv_llava_v0</span><span class="p">,</span>
+    <span class="s2">&quot;v0_mmtag&quot;</span><span class="p">:</span> <span class="n">conv_llava_v0_mmtag</span><span class="p">,</span>
+    <span class="s2">&quot;llava_v1&quot;</span><span class="p">:</span> <span class="n">conv_llava_v1</span><span class="p">,</span>
+    <span class="s2">&quot;v1_mmtag&quot;</span><span class="p">:</span> <span class="n">conv_llava_v1_mmtag</span><span class="p">,</span>
+    <span class="s2">&quot;llava_llama_2&quot;</span><span class="p">:</span> <span class="n">conv_llava_llama_2</span><span class="p">,</span>
+
+    <span class="s2">&quot;mpt&quot;</span><span class="p">:</span> <span class="n">conv_mpt</span><span class="p">,</span>
+<span class="p">}</span></div>
+
+
+
+<span class="k">if</span> <span class="vm">__name__</span> <span class="o">==</span> <span class="s2">&quot;__main__&quot;</span><span class="p">:</span>
+    <span class="nb">print</span><span class="p">(</span><span class="n">default_conversation</span><span class="o">.</span><span class="n">get_prompt</span><span class="p">())</span>
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/model.html b/_modules/lmflow/utils/model.html
new file mode 100644
index 000000000..0ac2ebc65
--- /dev/null
+++ b/_modules/lmflow/utils/model.html
@@ -0,0 +1,472 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/model';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.model</h1><div class="highlight"><pre>
+<span></span><span class="ch">#!/usr/bin/env python</span>
+<span class="c1"># coding=utf-8</span>
+<span class="c1"># Copyright 2024 Statistics and Machine Learning Research Group. All rights reserved.</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">Any</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Union</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">AutoTokenizer</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="n">ModelArguments</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/model/index.html#lmflow.utils.model.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="check_homogeneity">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/model/index.html#lmflow.utils.model.check_homogeneity">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">check_homogeneity</span><span class="p">(</span><span class="n">model_args_list</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">ModelArguments</span><span class="p">])</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
+    <span class="k">assert</span> <span class="nb">all</span><span class="p">(</span><span class="nb">isinstance</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="n">ModelArguments</span><span class="p">)</span> <span class="k">for</span> <span class="n">model_args</span> <span class="ow">in</span> <span class="n">model_args_list</span><span class="p">),</span> \
+        <span class="s2">&quot;model_args_list should be a list of ModelArguments objects.&quot;</span>
+    <span class="k">assert</span> <span class="nb">len</span><span class="p">(</span><span class="n">model_args_list</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">,</span> <span class="s2">&quot;model_args_list should have at least two elements.&quot;</span>
+    
+    <span class="n">tokenizer_names</span> <span class="o">=</span> <span class="p">[]</span>
+    <span class="k">for</span> <span class="n">model_args</span> <span class="ow">in</span> <span class="n">model_args_list</span><span class="p">:</span>
+        <span class="n">tokenizer</span> <span class="o">=</span> <span class="n">AutoTokenizer</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span><span class="n">model_args</span><span class="o">.</span><span class="n">model_name_or_path</span><span class="p">,</span> <span class="n">use_fast</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="n">tokenizer_names</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">tokenizer</span><span class="o">.</span><span class="vm">__class__</span><span class="o">.</span><span class="vm">__name__</span><span class="p">)</span>
+    
+    <span class="k">return</span> <span class="nb">len</span><span class="p">(</span><span class="nb">set</span><span class="p">(</span><span class="n">tokenizer_names</span><span class="p">))</span> <span class="o">==</span> <span class="mi">1</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/multimodal.html b/_modules/lmflow/utils/multimodal.html
new file mode 100644
index 000000000..015ce2240
--- /dev/null
+++ b/_modules/lmflow/utils/multimodal.html
@@ -0,0 +1,500 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.multimodal &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/multimodal';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.multimodal</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.multimodal</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span><span class="w"> </span><span class="nn">glob</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">LlamaConfig</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">tqdm</span><span class="w"> </span><span class="kn">import</span> <span class="n">tqdm</span>
+
+
+<div class="viewcode-block" id="update_custom_config">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/multimodal/index.html#lmflow.utils.multimodal.update_custom_config">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">update_custom_config</span><span class="p">(</span><span class="n">config</span><span class="p">,</span> <span class="n">model_args</span><span class="p">):</span>
+    <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">llm_model_name_or_path</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="n">text_config</span> <span class="o">=</span> <span class="n">LlamaConfig</span><span class="o">.</span><span class="n">from_pretrained</span><span class="p">(</span>
+            <span class="n">model_args</span><span class="o">.</span><span class="n">llm_model_name_or_path</span><span class="p">)</span>
+        <span class="n">config</span><span class="o">.</span><span class="n">text_config</span> <span class="o">=</span> <span class="n">text_config</span>
+    <span class="n">config</span><span class="o">.</span><span class="n">with_qformer</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">with_qformer</span>
+    <span class="n">config</span><span class="o">.</span><span class="n">custom_vision_model</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">custom_vision_model</span>
+    <span class="k">if</span> <span class="n">model_args</span><span class="o">.</span><span class="n">custom_vision_model</span><span class="p">:</span>
+        <span class="c1"># config.vision_model_args = model_args</span>
+        <span class="n">config</span><span class="o">.</span><span class="n">image_encoder_name_or_path</span> <span class="o">=</span> \
+            <span class="n">model_args</span><span class="o">.</span><span class="n">image_encoder_name_or_path</span>
+        <span class="n">config</span><span class="o">.</span><span class="n">vision_select_layer</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">vision_select_layer</span>
+        <span class="k">if</span> <span class="nb">getattr</span><span class="p">(</span><span class="n">model_args</span><span class="p">,</span> <span class="s2">&quot;vision_select_feature&quot;</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="n">config</span><span class="o">.</span><span class="n">vision_select_feature</span> <span class="o">=</span> <span class="n">model_args</span><span class="o">.</span><span class="n">vision_select_feature</span>
+    <span class="k">return</span> <span class="n">config</span></div>
+
+
+
+<div class="viewcode-block" id="load_llava_pretrain_model">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/multimodal/index.html#lmflow.utils.multimodal.load_llava_pretrain_model">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">load_llava_pretrain_model</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">checkpoint_path</span><span class="p">):</span>
+    <span class="n">checkpoint_path</span> <span class="o">=</span> <span class="n">glob</span><span class="o">.</span><span class="n">glob</span><span class="p">(</span><span class="n">checkpoint_path</span><span class="p">)</span>
+    <span class="k">for</span> <span class="n">path</span> <span class="ow">in</span> <span class="n">tqdm</span><span class="p">(</span><span class="n">checkpoint_path</span><span class="p">):</span>
+        <span class="n">state_dict</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">path</span><span class="p">,</span> <span class="n">map_location</span><span class="o">=</span><span class="s2">&quot;cpu&quot;</span><span class="p">)</span>
+        <span class="n">new_state_dict</span> <span class="o">=</span> <span class="n">adapt_llava_model_to_lmflow_type</span><span class="p">(</span><span class="n">state_dict</span><span class="p">)</span>
+        <span class="c1"># modify the name of the key</span>
+        <span class="c1"># import pdb; pdb.set_trace()</span>
+        <span class="n">lmflow_keys</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">state_dict</span><span class="p">()</span><span class="o">.</span><span class="n">keys</span><span class="p">()</span>
+        <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">new_state_dict</span><span class="o">.</span><span class="n">keys</span><span class="p">():</span>
+            <span class="k">if</span> <span class="n">key</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">lmflow_keys</span><span class="p">:</span>
+                <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;key not in lmflow_keys: &quot;</span><span class="p">,</span> <span class="n">key</span><span class="p">)</span>
+        <span class="n">model</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">new_state_dict</span><span class="p">,</span> <span class="n">strict</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+    <span class="k">return</span> <span class="n">model</span></div>
+
+
+<div class="viewcode-block" id="adapt_llava_model_to_lmflow_type">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/multimodal/index.html#lmflow.utils.multimodal.adapt_llava_model_to_lmflow_type">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">adapt_llava_model_to_lmflow_type</span><span class="p">(</span><span class="n">state_dict</span><span class="p">):</span>
+    <span class="n">new_state_dict</span> <span class="o">=</span> <span class="p">{}</span>
+    <span class="k">for</span> <span class="n">key</span><span class="p">,</span> <span class="n">item</span> <span class="ow">in</span> <span class="n">state_dict</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+        <span class="n">key</span> <span class="o">=</span> <span class="n">key</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;model.layers&quot;</span><span class="p">,</span> <span class="s2">&quot;language_model.model.layers&quot;</span><span class="p">)</span>
+        <span class="n">key</span> <span class="o">=</span> <span class="n">key</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;model.embed_tokens&quot;</span><span class="p">,</span>
+                          <span class="s2">&quot;language_model.model.embed_tokens&quot;</span><span class="p">)</span>
+        <span class="n">key</span> <span class="o">=</span> <span class="n">key</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;model.mm_projector&quot;</span><span class="p">,</span> <span class="s2">&quot;language_projection&quot;</span><span class="p">)</span>
+        <span class="n">key</span> <span class="o">=</span> <span class="n">key</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;lm_head&quot;</span><span class="p">,</span> <span class="s2">&quot;language_model.lm_head&quot;</span><span class="p">)</span>
+        <span class="n">key</span> <span class="o">=</span> <span class="n">key</span><span class="o">.</span><span class="n">replace</span><span class="p">(</span><span class="s2">&quot;model.norm&quot;</span><span class="p">,</span> <span class="s2">&quot;language_model.model.norm&quot;</span><span class="p">)</span>
+        <span class="k">if</span> <span class="s2">&quot;vision_tower&quot;</span> <span class="ow">in</span> <span class="n">key</span><span class="p">:</span>
+            <span class="k">continue</span>
+        <span class="n">new_state_dict</span><span class="p">[</span><span class="n">key</span><span class="p">]</span> <span class="o">=</span> <span class="n">item</span>
+    <span class="k">return</span> <span class="n">new_state_dict</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch.html b/_modules/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch.html
new file mode 100644
index 000000000..7605c84a0
--- /dev/null
+++ b/_modules/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch.html
@@ -0,0 +1,508 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</h1><div class="highlight"><pre>
+<span></span><span class="kn">from</span><span class="w"> </span><span class="nn">functools</span><span class="w"> </span><span class="kn">import</span> <span class="n">partial</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">transformers</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">transformers.models.llama.modeling_llama</span>
+
+<div class="viewcode-block" id="CondenseRotaryEmbedding">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding">[docs]</a>
+<span class="k">class</span><span class="w"> </span><span class="nc">CondenseRotaryEmbedding</span><span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
+    <span class="k">def</span><span class="w"> </span><span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dim</span><span class="p">,</span> <span class="n">pi_ratio</span><span class="p">,</span> <span class="n">ntk_ratio</span><span class="p">,</span> <span class="n">max_position_embeddings</span><span class="o">=</span><span class="mi">2048</span><span class="p">,</span> <span class="n">base</span><span class="o">=</span><span class="mi">10000</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
+        
+<div class="viewcode-block" id="CondenseRotaryEmbedding.ntk_ratio">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.ntk_ratio">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">ntk_ratio</span> <span class="o">=</span> <span class="n">ntk_ratio</span></div>
+
+        <span class="n">max_position_embeddings</span> <span class="o">*=</span> <span class="n">ntk_ratio</span>
+        <span class="n">base</span> <span class="o">=</span> <span class="n">base</span> <span class="o">*</span> <span class="n">ntk_ratio</span> <span class="o">**</span> <span class="p">(</span><span class="n">dim</span> <span class="o">/</span> <span class="p">(</span><span class="n">dim</span><span class="o">-</span><span class="mi">2</span><span class="p">))</span> <span class="c1">#Base change formula</span>
+        
+        <span class="n">inv_freq</span> <span class="o">=</span> <span class="mf">1.0</span> <span class="o">/</span> <span class="p">(</span><span class="n">base</span> <span class="o">**</span> <span class="p">(</span><span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">dim</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">float</span><span class="p">()</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">device</span><span class="p">)</span> <span class="o">/</span> <span class="n">dim</span><span class="p">))</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">register_buffer</span><span class="p">(</span><span class="s2">&quot;inv_freq&quot;</span><span class="p">,</span> <span class="n">inv_freq</span><span class="p">)</span>
+        
+<div class="viewcode-block" id="CondenseRotaryEmbedding.pi_ratio">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.pi_ratio">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">pi_ratio</span> <span class="o">=</span> <span class="n">pi_ratio</span></div>
+
+        <span class="n">max_position_embeddings</span> <span class="o">*=</span> <span class="n">pi_ratio</span>
+<div class="viewcode-block" id="CondenseRotaryEmbedding.max_seq_len_cached">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.max_seq_len_cached">[docs]</a>
+        <span class="bp">self</span><span class="o">.</span><span class="n">max_seq_len_cached</span> <span class="o">=</span> <span class="n">max_position_embeddings</span></div>
+
+        <span class="n">t</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">max_seq_len_cached</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inv_freq</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inv_freq</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span> <span class="o">/</span> <span class="n">pi_ratio</span>
+        <span class="n">freqs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">einsum</span><span class="p">(</span><span class="s2">&quot;i,j-&gt;ij&quot;</span><span class="p">,</span> <span class="n">t</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">inv_freq</span><span class="p">)</span>
+        
+        <span class="c1"># Different from paper, but it uses a different permutation in order to obtain the same calculation</span>
+        <span class="n">emb</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">freqs</span><span class="p">,</span> <span class="n">freqs</span><span class="p">),</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span>
+        <span class="n">dtype</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">get_default_dtype</span><span class="p">()</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">register_buffer</span><span class="p">(</span><span class="s2">&quot;cos_cached&quot;</span><span class="p">,</span> <span class="n">emb</span><span class="o">.</span><span class="n">cos</span><span class="p">()[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="p">),</span> <span class="n">persistent</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">register_buffer</span><span class="p">(</span><span class="s2">&quot;sin_cached&quot;</span><span class="p">,</span> <span class="n">emb</span><span class="o">.</span><span class="n">sin</span><span class="p">()[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="p">),</span> <span class="n">persistent</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+
+<div class="viewcode-block" id="CondenseRotaryEmbedding.forward">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.forward">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">,</span> <span class="n">seq_len</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
+        <span class="c1"># x: [bs, num_attention_heads, seq_len, head_size]</span>
+        <span class="c1"># This `if` block is unlikely to be run after we build sin/cos in `__init__`. Keep the logic here just in case.</span>
+        <span class="k">if</span> <span class="n">seq_len</span> <span class="o">&gt;</span> <span class="bp">self</span><span class="o">.</span><span class="n">max_seq_len_cached</span><span class="p">:</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">max_seq_len_cached</span> <span class="o">=</span> <span class="n">seq_len</span>
+            <span class="n">t</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">max_seq_len_cached</span><span class="p">,</span> <span class="n">device</span><span class="o">=</span><span class="n">x</span><span class="o">.</span><span class="n">device</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">inv_freq</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span> <span class="o">/</span> <span class="bp">self</span><span class="o">.</span><span class="n">pi_ratio</span>
+            <span class="n">freqs</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">einsum</span><span class="p">(</span><span class="s2">&quot;i,j-&gt;ij&quot;</span><span class="p">,</span> <span class="n">t</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">inv_freq</span><span class="p">)</span>
+            
+            <span class="c1"># Different from paper, but it uses a different permutation in order to obtain the same calculation</span>
+            <span class="n">emb</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">freqs</span><span class="p">,</span> <span class="n">freqs</span><span class="p">),</span> <span class="n">dim</span><span class="o">=-</span><span class="mi">1</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">device</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">register_buffer</span><span class="p">(</span><span class="s2">&quot;cos_cached&quot;</span><span class="p">,</span> <span class="n">emb</span><span class="o">.</span><span class="n">cos</span><span class="p">()[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">dtype</span><span class="p">),</span> <span class="n">persistent</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">register_buffer</span><span class="p">(</span><span class="s2">&quot;sin_cached&quot;</span><span class="p">,</span> <span class="n">emb</span><span class="o">.</span><span class="n">sin</span><span class="p">()[</span><span class="kc">None</span><span class="p">,</span> <span class="kc">None</span><span class="p">,</span> <span class="p">:,</span> <span class="p">:]</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">dtype</span><span class="p">),</span> <span class="n">persistent</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+            
+        <span class="k">return</span> <span class="p">(</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">cos_cached</span><span class="p">[:,</span> <span class="p">:,</span> <span class="p">:</span><span class="n">seq_len</span><span class="p">,</span> <span class="o">...</span><span class="p">]</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">x</span><span class="o">.</span><span class="n">dtype</span><span class="p">),</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">sin_cached</span><span class="p">[:,</span> <span class="p">:,</span> <span class="p">:</span><span class="n">seq_len</span><span class="p">,</span> <span class="o">...</span><span class="p">]</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="o">=</span><span class="n">x</span><span class="o">.</span><span class="n">dtype</span><span class="p">),</span>
+        <span class="p">)</span></div>
+</div>
+
+
+<div class="viewcode-block" id="replace_llama_with_condense">
+<a class="viewcode-back" href="../../../../autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.replace_llama_with_condense">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">replace_llama_with_condense</span><span class="p">(</span><span class="n">pi_ratio</span><span class="p">,</span> <span class="n">ntk_ratio</span><span class="p">):</span>
+    <span class="n">transformers</span><span class="o">.</span><span class="n">models</span><span class="o">.</span><span class="n">llama</span><span class="o">.</span><span class="n">modeling_llama</span><span class="o">.</span><span class="n">LlamaRotaryEmbedding</span> <span class="o">=</span> <span class="n">partial</span><span class="p">(</span><span class="n">CondenseRotaryEmbedding</span><span class="p">,</span> <span class="n">pi_ratio</span><span class="o">=</span><span class="n">pi_ratio</span><span class="p">,</span> <span class="n">ntk_ratio</span><span class="o">=</span><span class="n">ntk_ratio</span><span class="p">)</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/utils/versioning.html b/_modules/lmflow/utils/versioning.html
new file mode 100644
index 000000000..e0f15ff3d
--- /dev/null
+++ b/_modules/lmflow/utils/versioning.html
@@ -0,0 +1,555 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.utils.versioning &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/utils/versioning';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.versioning</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.utils.versioning</h1><div class="highlight"><pre>
+<span></span><span class="kn">import</span><span class="w"> </span><span class="nn">importlib</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
+<span class="kn">import</span><span class="w"> </span><span class="nn">logging</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Union</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">importlib.metadata</span><span class="w"> </span><span class="kn">import</span> <span class="n">version</span><span class="p">,</span> <span class="n">PackageNotFoundError</span>
+
+<span class="kn">import</span><span class="w"> </span><span class="nn">pkg_resources</span>
+
+
+<div class="viewcode-block" id="logger">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/versioning/index.html#lmflow.datasets.logger">[docs]</a>
+<span class="n">logger</span> <span class="o">=</span> <span class="n">logging</span><span class="o">.</span><span class="n">getLogger</span><span class="p">(</span><span class="vm">__name__</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="get_python_version">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/versioning/index.html#lmflow.datasets.get_python_version">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">get_python_version</span><span class="p">():</span>
+    <span class="k">return</span> <span class="n">sys</span><span class="o">.</span><span class="n">version_info</span></div>
+
+
+
+<div class="viewcode-block" id="_is_package_available">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/versioning/index.html#lmflow.datasets._is_package_available">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_is_package_available</span><span class="p">(</span><span class="n">package_name</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span> <span class="n">skippable</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">):</span>
+    <span class="k">assert</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">package_name</span><span class="p">,</span> <span class="nb">str</span><span class="p">),</span> <span class="sa">f</span><span class="s2">&quot;Invalid type of package_name: </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">package_name</span><span class="p">)</span><span class="si">}</span><span class="s2">&quot;</span>
+    <span class="k">try</span><span class="p">:</span>
+        <span class="n">importlib</span><span class="o">.</span><span class="n">import_module</span><span class="p">(</span><span class="n">package_name</span><span class="p">)</span>
+        <span class="k">return</span> <span class="kc">True</span>
+    <span class="k">except</span> <span class="ne">Exception</span> <span class="k">as</span> <span class="n">e</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">e</span><span class="o">.</span><span class="vm">__class__</span> <span class="o">==</span> <span class="ne">ModuleNotFoundError</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">False</span>
+        <span class="k">else</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">skippable</span><span class="p">:</span>
+                <span class="n">logger</span><span class="o">.</span><span class="n">warning</span><span class="p">(</span><span class="sa">f</span><span class="s1">&#39;An error occurred when importing </span><span class="si">{</span><span class="n">package_name</span><span class="si">}</span><span class="s1">:</span><span class="se">\n</span><span class="si">{</span><span class="n">e</span><span class="si">}</span><span class="se">\n</span><span class="si">{</span><span class="n">package_name</span><span class="si">}</span><span class="s1"> is disabled.&#39;</span><span class="p">)</span>
+                <span class="k">return</span> <span class="kc">False</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="n">e</span></div>
+
+            
+            
+<div class="viewcode-block" id="_is_packages_available">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/versioning/index.html#lmflow.datasets._is_packages_available">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">_is_packages_available</span><span class="p">(</span><span class="n">packages</span><span class="p">:</span> <span class="n">Union</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">],</span> <span class="n">List</span><span class="p">[</span><span class="n">Tuple</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">bool</span><span class="p">]]]):</span>
+    <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">packages</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="nb">str</span><span class="p">):</span>
+        <span class="k">return</span> <span class="nb">all</span><span class="p">([</span><span class="n">_is_package_available</span><span class="p">(</span><span class="n">package</span><span class="p">)</span> <span class="k">for</span> <span class="n">package</span> <span class="ow">in</span> <span class="n">packages</span><span class="p">])</span>
+    <span class="k">elif</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">packages</span><span class="p">[</span><span class="mi">0</span><span class="p">],</span> <span class="nb">tuple</span><span class="p">):</span>
+        <span class="k">return</span> <span class="nb">all</span><span class="p">([</span><span class="n">_is_package_available</span><span class="p">(</span><span class="n">package</span><span class="p">,</span> <span class="n">skippable</span><span class="p">)</span> <span class="k">for</span> <span class="n">package</span><span class="p">,</span> <span class="n">skippable</span> <span class="ow">in</span> <span class="n">packages</span><span class="p">])</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Invalid type of packages: </span><span class="si">{</span><span class="nb">type</span><span class="p">(</span><span class="n">packages</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span></div>
+
+    
+
+<div class="viewcode-block" id="is_package_version_at_least">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/versioning/index.html#lmflow.datasets.is_package_version_at_least">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">is_package_version_at_least</span><span class="p">(</span><span class="n">package_name</span><span class="p">,</span> <span class="n">min_version</span><span class="p">):</span>
+    <span class="k">try</span><span class="p">:</span>
+        <span class="n">package_version</span> <span class="o">=</span> <span class="n">pkg_resources</span><span class="o">.</span><span class="n">get_distribution</span><span class="p">(</span><span class="n">package_name</span><span class="p">)</span><span class="o">.</span><span class="n">version</span>
+        <span class="k">if</span> <span class="p">(</span><span class="n">pkg_resources</span><span class="o">.</span><span class="n">parse_version</span><span class="p">(</span><span class="n">package_version</span><span class="p">)</span>
+                <span class="o">&lt;</span> <span class="n">pkg_resources</span><span class="o">.</span><span class="n">parse_version</span><span class="p">(</span><span class="n">min_version</span><span class="p">)):</span>
+            <span class="k">return</span> <span class="kc">False</span>
+    <span class="k">except</span> <span class="n">pkg_resources</span><span class="o">.</span><span class="n">DistributionNotFound</span><span class="p">:</span>
+        <span class="k">return</span> <span class="kc">False</span>
+    <span class="k">return</span> <span class="kc">True</span></div>
+
+
+
+<div class="viewcode-block" id="is_gradio_available">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/versioning/index.html#lmflow.datasets.is_gradio_available">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">is_gradio_available</span><span class="p">():</span>
+    <span class="k">return</span> <span class="n">_is_package_available</span><span class="p">(</span><span class="s2">&quot;gradio&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="is_ray_available">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/versioning/index.html#lmflow.datasets.is_ray_available">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">is_ray_available</span><span class="p">():</span>
+    <span class="k">return</span> <span class="n">_is_package_available</span><span class="p">(</span><span class="s2">&quot;ray&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="is_vllm_available">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/versioning/index.html#lmflow.datasets.is_vllm_available">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">is_vllm_available</span><span class="p">():</span>
+    <span class="k">return</span> <span class="n">_is_package_available</span><span class="p">(</span><span class="s2">&quot;vllm&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="is_flash_attn_available">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/versioning/index.html#lmflow.datasets.is_flash_attn_available">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">is_flash_attn_available</span><span class="p">():</span>
+    <span class="k">return</span> <span class="n">_is_package_available</span><span class="p">(</span><span class="s2">&quot;flash_attn&quot;</span><span class="p">,</span> <span class="n">skippable</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="is_flask_available">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/versioning/index.html#lmflow.datasets.is_flask_available">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">is_flask_available</span><span class="p">():</span>
+    <span class="k">return</span> <span class="n">_is_packages_available</span><span class="p">([</span><span class="s2">&quot;flask&quot;</span><span class="p">,</span> <span class="s2">&quot;flask_cors&quot;</span><span class="p">])</span></div>
+
+
+
+<div class="viewcode-block" id="is_trl_available">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/versioning/index.html#lmflow.datasets.is_trl_available">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">is_trl_available</span><span class="p">():</span>
+    <span class="k">return</span> <span class="n">_is_package_available</span><span class="p">(</span><span class="s2">&quot;trl&quot;</span><span class="p">)</span></div>
+
+
+
+<div class="viewcode-block" id="is_multimodal_available">
+<a class="viewcode-back" href="../../../autoapi/lmflow/utils/versioning/index.html#lmflow.datasets.is_multimodal_available">[docs]</a>
+<span class="k">def</span><span class="w"> </span><span class="nf">is_multimodal_available</span><span class="p">():</span>
+    <span class="k">return</span> <span class="n">_is_packages_available</span><span class="p">([</span><span class="s2">&quot;PIL&quot;</span><span class="p">])</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_modules/lmflow/version.html b/_modules/lmflow/version.html
new file mode 100644
index 000000000..0665e6bfa
--- /dev/null
+++ b/_modules/lmflow/version.html
@@ -0,0 +1,445 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>lmflow.version &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = '_modules/lmflow/version';</script>
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">Module code</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../lmflow.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.version</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <h1>Source code for lmflow.version</h1><div class="highlight"><pre>
+<div class="viewcode-block" id="__version__">
+<a class="viewcode-back" href="../../autoapi/lmflow/version/index.html#lmflow.version.__version__">[docs]</a>
+<span></span><span class="n">__version__</span> <span class="o">=</span> <span class="s2">&quot;0.0.9&quot;</span></div>
+
+</pre></div>
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/_sources/about/authors.md.txt b/_sources/about/authors.md.txt
new file mode 100644
index 000000000..a3481ee66
--- /dev/null
+++ b/_sources/about/authors.md.txt
@@ -0,0 +1,4 @@
+# Contributors
+
+
+Shizhe Diao, Rui Pan, Hanze Dong, Ka Shun Shum, Jipeng Zhang, Wei Xiong, Tong Zhang
diff --git a/_sources/about/changelog.md.txt b/_sources/about/changelog.md.txt
new file mode 100644
index 000000000..991794df9
--- /dev/null
+++ b/_sources/about/changelog.md.txt
@@ -0,0 +1,15 @@
+# Changelog
+
+
+## Version 0.0.1 (Mar 28, 2023)
+
+The first public version. 
+
+Task tuning, instruction tuning, on user defined datasets.
+
+A simple and extensible API for developers.
+
+Efficient finetuning with LoRA. 
+
+Simplified model inference framework.
+
diff --git a/_sources/about/index.md.txt b/_sources/about/index.md.txt
new file mode 100644
index 000000000..df657b31c
--- /dev/null
+++ b/_sources/about/index.md.txt
@@ -0,0 +1,15 @@
+# About
+
+
+```{toctree}
+:maxdepth: 2
+
+changelog
+```
+
+
+```{toctree}
+:maxdepth: 2
+
+authors
+```
diff --git a/_sources/autoapi/index.rst.txt b/_sources/autoapi/index.rst.txt
new file mode 100644
index 000000000..333d5c5cd
--- /dev/null
+++ b/_sources/autoapi/index.rst.txt
@@ -0,0 +1,11 @@
+API Reference
+=============
+
+This page contains auto-generated API reference documentation [#f1]_.
+
+.. toctree::
+   :titlesonly:
+
+   /autoapi/lmflow/index
+
+.. [#f1] Created with `sphinx-autoapi <https://github.com/readthedocs/sphinx-autoapi>`_
\ No newline at end of file
diff --git a/_sources/autoapi/lmflow/args/index.rst.txt b/_sources/autoapi/lmflow/args/index.rst.txt
new file mode 100644
index 000000000..04e230fed
--- /dev/null
+++ b/_sources/autoapi/lmflow/args/index.rst.txt
@@ -0,0 +1,1982 @@
+lmflow.args
+===========
+
+.. py:module:: lmflow.args
+
+.. autoapi-nested-parse::
+
+   This script defines dataclasses: ModelArguments and DatasetArguments,
+   that contain the arguments for the model and dataset used in training.
+
+   It imports several modules, including dataclasses, field from typing, Optional from typing,
+   require_version from transformers.utils.versions, MODEL_FOR_CAUSAL_LM_MAPPING,
+   and TrainingArguments from transformers.
+
+   MODEL_CONFIG_CLASSES is assigned a list of the model config classes from
+   MODEL_FOR_CAUSAL_LM_MAPPING. MODEL_TYPES is assigned a tuple of the model types
+   extracted from the MODEL_CONFIG_CLASSES.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.args.MODEL_CONFIG_CLASSES
+   lmflow.args.MODEL_TYPES
+   lmflow.args.logger
+   lmflow.args.PIPELINE_ARGUMENT_MAPPING
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.args.OptimizerNames
+   lmflow.args.ModelArguments
+   lmflow.args.VisModelArguments
+   lmflow.args.DatasetArguments
+   lmflow.args.MultiModalDatasetArguments
+   lmflow.args.FinetunerArguments
+   lmflow.args.RewardModelTunerArguments
+   lmflow.args.EvaluatorArguments
+   lmflow.args.InferencerArguments
+   lmflow.args.RaftAlignerArguments
+   lmflow.args.BenchmarkingArguments
+   lmflow.args.DPOAlignerArguments
+   lmflow.args.DPOv2AlignerArguments
+   lmflow.args.IterativeAlignerArguments
+   lmflow.args.IterativeDPOAlignerArguments
+   lmflow.args.AutoArguments
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.args.split_args
+
+
+Module Contents
+---------------
+
+.. py:data:: MODEL_CONFIG_CLASSES
+
+.. py:data:: MODEL_TYPES
+
+.. py:data:: logger
+
+.. py:class:: OptimizerNames
+
+   .. py:attribute:: DUMMY
+      :value: 'dummy'
+
+
+
+   .. py:attribute:: ADABELIEF
+      :value: 'adabelief'
+
+
+
+   .. py:attribute:: ADABOUND
+      :value: 'adabound'
+
+
+
+   .. py:attribute:: LARS
+      :value: 'lars'
+
+
+
+   .. py:attribute:: LAMB
+      :value: 'lamb'
+
+
+
+   .. py:attribute:: ADAMAX
+      :value: 'adamax'
+
+
+
+   .. py:attribute:: NADAM
+      :value: 'nadam'
+
+
+
+   .. py:attribute:: RADAM
+      :value: 'radam'
+
+
+
+   .. py:attribute:: ADAMP
+      :value: 'adamp'
+
+
+
+   .. py:attribute:: SGDP
+      :value: 'sgdp'
+
+
+
+   .. py:attribute:: YOGI
+      :value: 'yogi'
+
+
+
+   .. py:attribute:: SOPHIA
+      :value: 'sophia'
+
+
+
+   .. py:attribute:: ADAN
+      :value: 'adan'
+
+
+
+   .. py:attribute:: ADAM
+      :value: 'adam'
+
+
+
+   .. py:attribute:: NOVOGRAD
+      :value: 'novograd'
+
+
+
+   .. py:attribute:: ADADELTA
+      :value: 'adadelta'
+
+
+
+   .. py:attribute:: ADAGRAD
+      :value: 'adagrad'
+
+
+
+   .. py:attribute:: ADAMW_SCHEDULE_FREE
+      :value: 'adamw_schedule_free'
+
+
+
+   .. py:attribute:: SGD_SCHEDULE_FREE
+      :value: 'sgd_schedule_free'
+
+
+
+.. py:class:: ModelArguments
+
+   
+   Define a class ModelArguments using the dataclass decorator. 
+   The class contains several optional parameters that can be used to configure a model. 
+
+   model_name_or_path : str
+       a string representing the path or name of a pretrained
+       model checkpoint for weights initialization. If None, a model will be trained from scratch.
+
+   model_type :  str
+       a string representing the type of model to use if training from
+       scratch. If not provided, a pretrained model will be used.
+
+   config_overrides :  str
+       a string representing the default config settings to override
+       when training a model from scratch.
+
+   config_name : str
+       a string representing the name or path of the pretrained config to
+       use, if different from the model_name_or_path.
+
+   tokenizer_name :  str
+       a string representing the name or path of the pretrained tokenizer
+       to use, if different from the model_name_or_path.
+
+   cache_dir :  str
+       a string representing the path to the directory where pretrained models
+       downloaded from huggingface.co will be stored.
+
+   use_fast_tokenizer : bool
+       a boolean indicating whether to use a fast tokenizer (backed by the
+       tokenizers library) or not.
+
+   model_revision :  str
+       a string representing the specific model version to use (can be a
+       branch name, tag name, or commit id).
+
+   token : Optional[str]
+       Necessary when accessing a private model/dataset.
+
+   torch_dtype :  str
+       a string representing the dtype to load the model under. If auto is
+       passed, the dtype will be automatically derived from the model's weights.
+
+   use_ram_optimized_load : bool
+       a boolean indicating whether to use disk mapping when memory is not
+       enough.
+
+   use_int8 : bool
+       a boolean indicating whether to load int8 quantization for inference.
+
+   load_in_4bit : bool
+       whether to load the model in 4bit
+
+   model_max_length : int
+       The maximum length of the model.
+
+   truncation_side : str
+       The side on which the model should have truncation applied.
+
+   arch_type : str
+       Model architecture type.
+   padding_side : str
+       The side on which the tokenizer should have padding applied.
+   eos_padding : bool
+       whether to pad with eos token instead of pad token.
+   ignore_bias_buffers : bool
+       fix for DDP issues with LM bias/mask buffers - invalid scalar type,`inplace operation.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: model_name_or_path
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: lora_model_path
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: model_type
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: config_overrides
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: arch_type
+      :type:  Optional[str]
+      :value: 'decoder_only'
+
+
+
+   .. py:attribute:: config_name
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: tokenizer_name
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: cache_dir
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: use_fast_tokenizer
+      :type:  bool
+      :value: True
+
+
+
+   .. py:attribute:: model_revision
+      :type:  str
+      :value: 'main'
+
+
+
+   .. py:attribute:: token
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: trust_remote_code
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: torch_dtype
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: use_dora
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: use_lora
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: use_qlora
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: bits
+      :type:  int
+      :value: 4
+
+
+
+   .. py:attribute:: quant_type
+      :type:  str
+      :value: 'nf4'
+
+
+
+   .. py:attribute:: double_quant
+      :type:  bool
+      :value: True
+
+
+
+   .. py:attribute:: lora_r
+      :type:  int
+      :value: 8
+
+
+
+   .. py:attribute:: lora_alpha
+      :type:  int
+      :value: 32
+
+
+
+   .. py:attribute:: lora_target_modules
+      :type:  str
+      :value: None
+
+
+
+   .. py:attribute:: lora_dropout
+      :type:  float
+      :value: 0.1
+
+
+
+   .. py:attribute:: save_aggregated_lora
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: use_ram_optimized_load
+      :type:  bool
+      :value: True
+
+
+
+   .. py:attribute:: use_flash_attention
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: truncate_to_model_max_length
+      :type:  bool
+      :value: True
+
+
+
+   .. py:attribute:: do_rope_scaling
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: rope_pi_ratio
+      :type:  int
+      :value: 1
+
+
+
+   .. py:attribute:: rope_ntk_ratio
+      :type:  int
+      :value: 1
+
+
+
+   .. py:attribute:: use_int8
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: load_in_4bit
+      :type:  Optional[bool]
+      :value: True
+
+
+
+   .. py:attribute:: model_max_length
+      :type:  Optional[int]
+      :value: None
+
+
+
+   .. py:attribute:: truncation_side
+      :type:  str
+      :value: None
+
+
+
+   .. py:attribute:: padding_side
+      :type:  str
+      :value: 'right'
+
+
+
+   .. py:attribute:: eos_padding
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: ignore_bias_buffers
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:method:: __post_init__()
+
+
+.. py:class:: VisModelArguments
+
+   Bases: :py:obj:`ModelArguments`
+
+
+   
+   Define a class ModelArguments using the dataclass decorator. 
+   The class contains several optional parameters that can be used to configure a model. 
+
+   model_name_or_path : str
+       a string representing the path or name of a pretrained
+       model checkpoint for weights initialization. If None, a model will be trained from scratch.
+
+   model_type :  str
+       a string representing the type of model to use if training from
+       scratch. If not provided, a pretrained model will be used.
+
+   config_overrides :  str
+       a string representing the default config settings to override
+       when training a model from scratch.
+
+   config_name : str
+       a string representing the name or path of the pretrained config to
+       use, if different from the model_name_or_path.
+
+   tokenizer_name :  str
+       a string representing the name or path of the pretrained tokenizer
+       to use, if different from the model_name_or_path.
+
+   cache_dir :  str
+       a string representing the path to the directory where pretrained models
+       downloaded from huggingface.co will be stored.
+
+   use_fast_tokenizer : bool
+       a boolean indicating whether to use a fast tokenizer (backed by the
+       tokenizers library) or not.
+
+   model_revision :  str
+       a string representing the specific model version to use (can be a
+       branch name, tag name, or commit id).
+
+   token : Optional[str]
+       Necessary when accessing a private model/dataset.
+
+   torch_dtype :  str
+       a string representing the dtype to load the model under. If auto is
+       passed, the dtype will be automatically derived from the model's weights.
+
+   use_ram_optimized_load : bool
+       a boolean indicating whether to use disk mapping when memory is not
+       enough.
+
+   use_int8 : bool
+       a boolean indicating whether to load int8 quantization for inference.
+
+   load_in_4bit : bool
+       whether to load the model in 4bit
+
+   model_max_length : int
+       The maximum length of the model.
+
+   truncation_side : str
+       The side on which the model should have truncation applied.
+
+   arch_type : str
+       Model architecture type.
+   padding_side : str
+       The side on which the tokenizer should have padding applied.
+   eos_padding : bool
+       whether to pad with eos token instead of pad token.
+   ignore_bias_buffers : bool
+       fix for DDP issues with LM bias/mask buffers - invalid scalar type,`inplace operation.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: low_resource
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: custom_model
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: pretrained_language_projection_path
+      :type:  str
+      :value: None
+
+
+
+   .. py:attribute:: custom_vision_model
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: image_encoder_name_or_path
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: qformer_name_or_path
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: llm_model_name_or_path
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: use_prompt_cache
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: prompt_cache_path
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: llava_loading
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: with_qformer
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: vision_select_layer
+      :type:  Optional[int]
+      :value: -2
+
+
+
+   .. py:attribute:: llava_pretrain_model_path
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: save_pretrain_model_path
+      :type:  Optional[str]
+      :value: None
+
+
+
+.. py:class:: DatasetArguments
+
+   
+   Define a class DatasetArguments using the dataclass decorator.
+   The class contains several optional parameters that can be used to configure a dataset for a language model.
+
+   dataset_path : str
+       a string representing the path of the dataset to use.
+
+   dataset_name : str
+       a string representing the name of the dataset to use. The default value is "customized".
+
+   is_custom_dataset : bool
+       a boolean indicating whether to use custom data. The default value is False.
+
+   customized_cache_dir : str
+       a string representing the path to the directory where customized dataset caches will be stored.
+
+   dataset_config_name : str
+       a string representing the configuration name of the dataset to use (via the datasets library).
+
+   train_file : str
+       a string representing the path to the input training data file (a text file).
+
+   validation_file : str
+       a string representing the path to the input evaluation data file to evaluate the perplexity on (a text file).
+
+   max_train_samples : int
+       an integer indicating the maximum number of training examples to use for debugging or quicker training.
+       If set, the training dataset will be truncated to this number.
+
+   max_eval_samples: int
+       an integer indicating the maximum number of evaluation examples to use for debugging or quicker training.
+       If set, the evaluation dataset will be truncated to this number.
+
+   streaming : bool
+       a boolean indicating whether to enable streaming mode.
+
+   block_size: int
+       an integer indicating the optional input sequence length after tokenization. The training dataset will be
+       truncated in blocks of this size for training.
+
+   train_on_prompt: bool
+       a boolean indicating whether to train on prompt for conversation datasets such as ShareGPT.
+
+   conversation_template: str
+       a string representing the template for conversation datasets.
+
+   The class also includes some additional parameters that can be used to configure the dataset further, such as `overwrite_cache`,
+   `validation_split_percentage`, `preprocessing_num_workers`, `disable_group_texts`, `demo_example_in_prompt`, `explanation_in_prompt`,
+   `keep_linebreaks`, and `prompt_structure`.
+
+   The field function is used to set default values and provide help messages for each parameter. The Optional type hint is
+   used to indicate that a parameter is optional. The metadata argument is used to provide additional information about
+   each parameter, such as a help message.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: dataset_path
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: dataset_name
+      :type:  Optional[str]
+      :value: 'customized'
+
+
+
+   .. py:attribute:: is_custom_dataset
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: customized_cache_dir
+      :type:  Optional[str]
+      :value: '.cache/llm-ft/datasets'
+
+
+
+   .. py:attribute:: dataset_config_name
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: train_file
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: validation_file
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: max_train_samples
+      :type:  Optional[int]
+      :value: None
+
+
+
+   .. py:attribute:: max_eval_samples
+      :type:  Optional[int]
+      :value: 10000000000.0
+
+
+
+   .. py:attribute:: streaming
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: block_size
+      :type:  Optional[int]
+      :value: None
+
+
+
+   .. py:attribute:: overwrite_cache
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: validation_split_percentage
+      :type:  Optional[int]
+      :value: 5
+
+
+
+   .. py:attribute:: preprocessing_num_workers
+      :type:  Optional[int]
+      :value: None
+
+
+
+   .. py:attribute:: group_texts_batch_size
+      :type:  int
+      :value: 1000
+
+
+
+   .. py:attribute:: disable_group_texts
+      :type:  bool
+      :value: True
+
+
+
+   .. py:attribute:: keep_linebreaks
+      :type:  bool
+      :value: True
+
+
+
+   .. py:attribute:: test_file
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: train_on_prompt
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: conversation_template
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:method:: __post_init__()
+
+
+.. py:class:: MultiModalDatasetArguments
+
+   Bases: :py:obj:`DatasetArguments`
+
+
+   
+   Define a class DatasetArguments using the dataclass decorator.
+   The class contains several optional parameters that can be used to configure a dataset for a language model.
+
+   dataset_path : str
+       a string representing the path of the dataset to use.
+
+   dataset_name : str
+       a string representing the name of the dataset to use. The default value is "customized".
+
+   is_custom_dataset : bool
+       a boolean indicating whether to use custom data. The default value is False.
+
+   customized_cache_dir : str
+       a string representing the path to the directory where customized dataset caches will be stored.
+
+   dataset_config_name : str
+       a string representing the configuration name of the dataset to use (via the datasets library).
+
+   train_file : str
+       a string representing the path to the input training data file (a text file).
+
+   validation_file : str
+       a string representing the path to the input evaluation data file to evaluate the perplexity on (a text file).
+
+   max_train_samples : int
+       an integer indicating the maximum number of training examples to use for debugging or quicker training.
+       If set, the training dataset will be truncated to this number.
+
+   max_eval_samples: int
+       an integer indicating the maximum number of evaluation examples to use for debugging or quicker training.
+       If set, the evaluation dataset will be truncated to this number.
+
+   streaming : bool
+       a boolean indicating whether to enable streaming mode.
+
+   block_size: int
+       an integer indicating the optional input sequence length after tokenization. The training dataset will be
+       truncated in blocks of this size for training.
+
+   train_on_prompt: bool
+       a boolean indicating whether to train on prompt for conversation datasets such as ShareGPT.
+
+   conversation_template: str
+       a string representing the template for conversation datasets.
+
+   The class also includes some additional parameters that can be used to configure the dataset further, such as `overwrite_cache`,
+   `validation_split_percentage`, `preprocessing_num_workers`, `disable_group_texts`, `demo_example_in_prompt`, `explanation_in_prompt`,
+   `keep_linebreaks`, and `prompt_structure`.
+
+   The field function is used to set default values and provide help messages for each parameter. The Optional type hint is
+   used to indicate that a parameter is optional. The metadata argument is used to provide additional information about
+   each parameter, such as a help message.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: image_folder
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: image_aspect_ratio
+      :type:  Optional[str]
+      :value: 'pad'
+
+
+
+   .. py:attribute:: is_multimodal
+      :type:  Optional[bool]
+      :value: True
+
+
+
+   .. py:attribute:: use_image_start_end
+      :type:  Optional[bool]
+      :value: True
+
+
+
+   .. py:attribute:: sep_style
+      :type:  Optional[str]
+      :value: 'plain'
+
+
+
+.. py:class:: FinetunerArguments
+
+   Bases: :py:obj:`transformers.TrainingArguments`
+
+
+   
+   Adapt transformers.TrainingArguments
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: eval_dataset_path
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: remove_unused_columns
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: finetune_part
+      :type:  Optional[str]
+      :value: 'language_projection'
+
+
+
+   .. py:attribute:: save_language_projection
+      :type:  Optional[str]
+      :value: False
+
+
+
+   .. py:attribute:: use_lisa
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: lisa_activated_layers
+      :type:  int
+      :value: 2
+
+
+
+   .. py:attribute:: lisa_interval_steps
+      :type:  int
+      :value: 20
+
+
+
+   .. py:attribute:: lisa_layers_attribute
+      :type:  str
+      :value: 'model.model.layers'
+
+
+
+   .. py:attribute:: use_customized_optim
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: customized_optim
+      :type:  str
+      :value: 'sign_sgd'
+
+
+
+   .. py:attribute:: customized_optim_args
+      :type:  str
+      :value: None
+
+
+
+   .. py:attribute:: optim_dummy_beta1
+      :type:  float
+      :value: 0.9
+
+
+
+   .. py:attribute:: optim_dummy_beta2
+      :type:  float
+      :value: 0.999
+
+
+
+   .. py:attribute:: optim_adam_beta1
+      :type:  float
+      :value: 0.9
+
+
+
+   .. py:attribute:: optim_adam_beta2
+      :type:  float
+      :value: 0.999
+
+
+
+   .. py:attribute:: optim_beta1
+      :type:  float
+      :value: 0.9
+
+
+
+   .. py:attribute:: optim_beta2
+      :type:  float
+      :value: 0.999
+
+
+
+   .. py:attribute:: optim_beta3
+      :type:  float
+      :value: 0.9
+
+
+
+   .. py:attribute:: optim_momentum
+      :type:  float
+      :value: 0.999
+
+
+
+   .. py:attribute:: optim_weight_decay
+      :type:  float
+      :value: 0
+
+
+
+.. py:class:: RewardModelTunerArguments
+
+   Bases: :py:obj:`FinetunerArguments`
+
+
+   
+   Arguments for reward modeling.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:class:: EvaluatorArguments
+
+   
+   Define a class EvaluatorArguments using the dataclass decorator. The class contains several optional
+   parameters that can be used to configure a evaluator.
+
+   local_rank : str
+       For distributed training: local_rank
+
+   random_shuffle : bool
+
+   use_wandb : bool
+
+   random_seed : int, default = 1
+
+   output_dir : str, default = './output_dir',
+
+   mixed_precision : str, choice from ["bf16","fp16"].
+       mixed precision mode, whether to use bf16 or fp16
+
+   deepspeed :
+       Enable deepspeed and pass the path to deepspeed json config file (e.g. ds_config.json) or an already
+       loaded json file as a dict
+
+   temperature : float
+       An argument of model.generate in huggingface to control the diversity of generation.
+
+   repetition_penalty : float
+       An argument of model.generate in huggingface to penalize repetitions.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: local_rank
+      :type:  int
+      :value: -1
+
+
+
+   .. py:attribute:: random_shuffle
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: use_wandb
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: random_seed
+      :type:  Optional[int]
+      :value: 1
+
+
+
+   .. py:attribute:: output_dir
+      :type:  Optional[str]
+      :value: './output_dir'
+
+
+
+   .. py:attribute:: mixed_precision
+      :type:  Optional[str]
+      :value: 'bf16'
+
+
+
+   .. py:attribute:: deepspeed
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: answer_type
+      :type:  Optional[str]
+      :value: 'text'
+
+
+
+   .. py:attribute:: prompt_structure
+      :type:  Optional[str]
+      :value: '{input}'
+
+
+
+   .. py:attribute:: evaluate_block_size
+      :type:  Optional[int]
+      :value: 512
+
+
+
+   .. py:attribute:: metric
+      :type:  Optional[str]
+      :value: 'accuracy'
+
+
+
+   .. py:attribute:: inference_batch_size_per_device
+      :type:  Optional[int]
+      :value: 1
+
+
+
+   .. py:attribute:: use_accelerator_for_evaluator
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: temperature
+      :type:  float
+      :value: 0
+
+
+
+   .. py:attribute:: repetition_penalty
+      :type:  float
+      :value: 1
+
+
+
+   .. py:attribute:: max_new_tokens
+      :type:  int
+      :value: 100
+
+
+
+.. py:class:: InferencerArguments
+
+   
+   Define a class InferencerArguments using the dataclass decorator. The class contains several optional
+   parameters that can be used to configure a inferencer.
+
+   local_rank : str
+       For distributed training: local_rank
+   random_seed : int, default = 1
+   inference_batch_size : int, default = 1
+   deepspeed :
+       Enable deepspeed and pass the path to deepspeed json config file (e.g. ds_config.json) or an already
+       loaded json file as a dict
+   mixed_precision : str, choice from ["bf16","fp16"].
+       mixed precision mode, whether to use bf16 or fp16
+   temperature : float
+       An argument of model.generate in huggingface to control the diversity of generation.
+   repetition_penalty : float
+       An argument of model.generate in huggingface to penalize repetitions.
+   use_beam_search : Optional[bool]
+       Whether to use beam search during inference, By default False.
+   num_output_sequences : Optional[int]
+       Number of output sequences to return for the given prompt, 
+       currently only used in vllm inference, By default 8.
+   top_p : Optional[float]
+       top_p for sampling, By default 1.0.
+   top_k : Optional[int]
+       top_k for sampling, By default -1 (no top_k).
+   additional_stop_token_ids : Optional[List[int]]
+       the ids of the end of sentence tokens, By default [].
+   apply_chat_template : Optional[bool]
+       Whether to apply chat template, By default True.
+   save_results : Optional[bool]
+       Whether to save inference results, By default False.
+   results_path : Optional[str]
+       The **json file** path of inference results, By default None.
+   enable_decode_inference_result : Optional[bool]
+       Whether to detokenize the inference results. 
+
+       NOTE: For iterative align pipelines, whether to detokenize depends on 
+       the homogeneity of the policy model and the reward model 
+       (i.e., if they have the same tokenizer).
+   use_vllm: bool, optional
+       Whether to use VLLM for inference, By default False.
+   vllm_tensor_parallel_size: int, optional
+       The tensor parallel size for VLLM inference.
+   vllm_gpu_memory_utilization: float, optional
+       The GPU memory utilization for VLLM inference. The proportion of GPU
+       memory (per GPU) to use for VLLM inference.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: device
+      :type:  str
+      :value: 'gpu'
+
+
+
+   .. py:attribute:: local_rank
+      :type:  int
+      :value: -1
+
+
+
+   .. py:attribute:: inference_batch_size
+      :type:  int
+      :value: 1
+
+
+
+   .. py:attribute:: vllm_inference_batch_size
+      :type:  int
+      :value: 1
+
+
+
+   .. py:attribute:: temperature
+      :type:  float
+      :value: 0.0
+
+
+
+   .. py:attribute:: repetition_penalty
+      :type:  float
+      :value: 1
+
+
+
+   .. py:attribute:: max_new_tokens
+      :type:  int
+      :value: 100
+
+
+
+   .. py:attribute:: random_seed
+      :type:  Optional[int]
+      :value: 1
+
+
+
+   .. py:attribute:: deepspeed
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: mixed_precision
+      :type:  Optional[str]
+      :value: 'bf16'
+
+
+
+   .. py:attribute:: do_sample
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: use_accelerator
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: use_beam_search
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: num_output_sequences
+      :type:  Optional[int]
+      :value: 8
+
+
+
+   .. py:attribute:: top_p
+      :type:  Optional[float]
+      :value: 1.0
+
+
+
+   .. py:attribute:: top_k
+      :type:  Optional[int]
+      :value: -1
+
+
+
+   .. py:attribute:: additional_stop_token_ids
+      :type:  Optional[List[int]]
+      :value: []
+
+
+
+   .. py:attribute:: apply_chat_template
+      :type:  Optional[bool]
+      :value: True
+
+
+
+   .. py:attribute:: enable_decode_inference_result
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: tensor_parallel_size
+      :type:  Optional[int]
+      :value: 1
+
+
+
+   .. py:attribute:: enable_distributed_inference
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: distributed_inference_num_instances
+      :type:  Optional[int]
+      :value: 1
+
+
+
+   .. py:attribute:: use_vllm
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: vllm_tensor_parallel_size
+      :type:  Optional[int]
+      :value: 1
+
+
+
+   .. py:attribute:: vllm_gpu_memory_utilization
+      :type:  Optional[float]
+      :value: 0.95
+
+
+
+   .. py:attribute:: save_results
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: results_path
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:method:: __post_init__()
+
+
+.. py:class:: RaftAlignerArguments
+
+   Bases: :py:obj:`transformers.TrainingArguments`
+
+
+   
+   Define a class RaftAlignerArguments to configure raft aligner.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: output_reward_path
+      :type:  Optional[str]
+      :value: 'tmp/raft_aligner/'
+
+
+
+   .. py:attribute:: output_min_length
+      :type:  Optional[int]
+      :value: 64
+
+
+
+   .. py:attribute:: output_max_length
+      :type:  Optional[int]
+      :value: 128
+
+
+
+   .. py:attribute:: num_raft_iteration
+      :type:  Optional[int]
+      :value: 20
+
+
+
+   .. py:attribute:: raft_batch_size
+      :type:  Optional[int]
+      :value: 1024
+
+
+
+   .. py:attribute:: top_reward_percentage
+      :type:  Optional[float]
+      :value: 0.2
+
+
+
+   .. py:attribute:: inference_batch_size_per_device
+      :type:  Optional[int]
+      :value: 1
+
+
+
+   .. py:attribute:: collection_strategy
+      :type:  Optional[str]
+      :value: 'top'
+
+
+
+.. py:class:: BenchmarkingArguments
+
+   .. py:attribute:: dataset_name
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: lm_evaluation_metric
+      :type:  Optional[str]
+      :value: 'accuracy'
+
+
+
+.. py:class:: DPOAlignerArguments
+
+   
+   The arguments for the DPO training script.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: local_rank
+      :type:  int
+      :value: -1
+
+
+
+   .. py:attribute:: beta
+      :type:  Optional[float]
+      :value: 0.1
+
+
+
+   .. py:attribute:: learning_rate
+      :type:  Optional[float]
+      :value: 0.0005
+
+
+
+   .. py:attribute:: lr_scheduler_type
+      :type:  Optional[str]
+      :value: 'cosine'
+
+
+
+   .. py:attribute:: warmup_steps
+      :type:  Optional[int]
+      :value: 100
+
+
+
+   .. py:attribute:: weight_decay
+      :type:  Optional[float]
+      :value: 0.05
+
+
+
+   .. py:attribute:: optimizer_type
+      :type:  Optional[str]
+      :value: 'paged_adamw_32bit'
+
+
+
+   .. py:attribute:: per_device_train_batch_size
+      :type:  Optional[int]
+      :value: 4
+
+
+
+   .. py:attribute:: per_device_eval_batch_size
+      :type:  Optional[int]
+      :value: 1
+
+
+
+   .. py:attribute:: gradient_accumulation_steps
+      :type:  Optional[int]
+      :value: 4
+
+
+
+   .. py:attribute:: gradient_checkpointing
+      :type:  Optional[bool]
+      :value: True
+
+
+
+   .. py:attribute:: gradient_checkpointing_use_reentrant
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: max_prompt_length
+      :type:  Optional[int]
+      :value: 512
+
+
+
+   .. py:attribute:: max_length
+      :type:  Optional[int]
+      :value: 1024
+
+
+
+   .. py:attribute:: max_steps
+      :type:  Optional[int]
+      :value: 1000
+
+
+
+   .. py:attribute:: logging_steps
+      :type:  Optional[int]
+      :value: 10
+
+
+
+   .. py:attribute:: save_steps
+      :type:  Optional[int]
+      :value: 100
+
+
+
+   .. py:attribute:: eval_steps
+      :type:  Optional[int]
+      :value: 100
+
+
+
+   .. py:attribute:: output_dir
+      :type:  Optional[str]
+      :value: './results'
+
+
+
+   .. py:attribute:: log_freq
+      :type:  Optional[int]
+      :value: 1
+
+
+
+   .. py:attribute:: sanity_check
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: report_to
+      :type:  Optional[str]
+      :value: 'wandb'
+
+
+
+   .. py:attribute:: seed
+      :type:  Optional[int]
+      :value: 0
+
+
+
+   .. py:attribute:: run_name
+      :type:  Optional[str]
+      :value: 'dpo'
+
+
+
+   .. py:attribute:: eval_dataset_path
+      :type:  Optional[str]
+      :value: None
+
+
+
+.. py:class:: DPOv2AlignerArguments
+
+   Bases: :py:obj:`FinetunerArguments`
+
+
+   
+   The arguments for the DPOv2 training script.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: random_seed
+      :type:  Optional[int]
+      :value: 42
+
+
+
+   .. py:attribute:: accelerate_config_file
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:attribute:: margin_scale
+      :type:  Optional[float]
+      :value: 1.0
+
+
+
+   .. py:attribute:: sampling_paired_method
+      :type:  Optional[str]
+      :value: 'max_random'
+
+
+
+   .. py:attribute:: length_penalty
+      :type:  Optional[float]
+      :value: 0
+
+
+
+   .. py:attribute:: max_length
+      :type:  Optional[int]
+      :value: 2048
+
+
+
+   .. py:attribute:: max_prompt_length
+      :type:  Optional[int]
+      :value: 1000
+
+
+
+   .. py:attribute:: mask_prompt
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: beta
+      :type:  Optional[float]
+      :value: 0.1
+
+
+
+   .. py:attribute:: loss_type
+      :type:  Optional[str]
+      :value: 'sigmoid'
+
+
+
+.. py:class:: IterativeAlignerArguments
+
+   Bases: :py:obj:`InferencerArguments`
+
+
+   
+   Arguments for iterative aligners.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: dataset_path_list
+      :type:  List[str]
+      :value: []
+
+
+
+   .. py:attribute:: initial_iter_idx
+      :type:  int
+      :value: 0
+
+
+
+.. py:class:: IterativeDPOAlignerArguments
+
+   Bases: :py:obj:`IterativeAlignerArguments`, :py:obj:`DPOv2AlignerArguments`
+
+
+   
+   Arguments for iterative DPO aligners.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: output_dir
+      :type:  Optional[str]
+      :value: './runs'
+
+
+
+   .. py:attribute:: reward_model_inference_batch_size
+      :type:  int
+      :value: 1
+
+
+
+   .. py:attribute:: reward_model_inference_block_size
+      :type:  int
+      :value: 2048
+
+
+
+   .. py:attribute:: do_response_generation
+      :type:  bool
+      :value: True
+
+
+
+   .. py:attribute:: do_scoring
+      :type:  bool
+      :value: True
+
+
+
+   .. py:attribute:: do_dpo_align
+      :type:  bool
+      :value: True
+
+
+
+.. py:data:: PIPELINE_ARGUMENT_MAPPING
+
+.. py:class:: AutoArguments
+
+   
+   Automatically choose arguments from FinetunerArguments or EvaluatorArguments.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: get_pipeline_args_class()
+
+
+.. py:function:: split_args(args)
+
diff --git a/_sources/autoapi/lmflow/datasets/dataset/index.rst.txt b/_sources/autoapi/lmflow/datasets/dataset/index.rst.txt
new file mode 100644
index 000000000..a171bd384
--- /dev/null
+++ b/_sources/autoapi/lmflow/datasets/dataset/index.rst.txt
@@ -0,0 +1,646 @@
+lmflow.datasets.dataset
+=======================
+
+.. py:module:: lmflow.datasets.dataset
+
+.. autoapi-nested-parse::
+
+   This Python code defines a class Dataset with methods for initializing, loading,
+   and manipulating datasets from different backends such as Hugging Face and JSON.
+    
+   The `Dataset` class includes methods for loading datasets from a dictionary and a Hugging
+   Face dataset, mapping datasets, and retrieving the backend dataset and arguments.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.datasets.dataset.logger
+   lmflow.datasets.dataset.DATASET_TYPES
+   lmflow.datasets.dataset.KEY_TYPE
+   lmflow.datasets.dataset.KEY_INSTANCES
+   lmflow.datasets.dataset.KEY_SCORE
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.datasets.dataset.Dataset
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:data:: DATASET_TYPES
+   :value: ['text_only', 'text2text', 'float_only', 'image_text', 'conversation', 'paired_conversation',...
+
+
+.. py:data:: KEY_TYPE
+   :value: 'type'
+
+
+.. py:data:: KEY_INSTANCES
+   :value: 'instances'
+
+
+.. py:data:: KEY_SCORE
+   :value: 'score'
+
+
+.. py:class:: Dataset(data_args: lmflow.args.DatasetArguments = None, backend: str = 'huggingface', *args, **kwargs)
+
+   
+   Initializes the Dataset object with the given parameters.
+
+
+   :Parameters:
+
+       **data_args** : DatasetArguments object.
+           Contains the arguments required to load the dataset.
+
+       **backend** : str,  default="huggingface"
+           A string representing the dataset backend. Defaults to "huggingface".
+
+       **args** : Optional.
+           Positional arguments.
+
+       **kwargs** : Optional.
+           Keyword arguments.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: data_args
+      :value: None
+
+
+
+   .. py:attribute:: backend
+      :value: 'huggingface'
+
+
+
+   .. py:attribute:: backend_dataset
+      :value: None
+
+
+
+   .. py:attribute:: type
+      :value: None
+
+
+
+   .. py:attribute:: dataset_path
+
+
+   .. py:method:: __len__()
+
+
+   .. py:method:: _check_data_format()
+
+      
+      Checks if data type and data structure matches
+
+      Raise messages with hints if not matched.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: from_dict(dict_obj: dict, *args, **kwargs)
+
+      
+      Create a Dataset object from a dictionary.
+
+      Return a Dataset given a dict with format:
+          {
+              "type": TYPE,
+              "instances": [
+                  {
+                      "key_1": VALUE_1.1,
+                      "key_2": VALUE_1.2,
+                      ...
+                  },
+                  {
+                      "key_1": VALUE_2.1,
+                      "key_2": VALUE_2.2,
+                      ...
+                  },
+                  ...
+              ]
+          }
+
+      :Parameters:
+
+          **dict_obj** : dict.
+              A dictionary containing the dataset information.
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.
+
+
+
+      :Returns:
+
+          **self** : Dataset object.
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: create_from_dict(dict_obj, *args, **kwargs)
+      :classmethod:
+
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          Returns a Dataset object given a dict.
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: to_dict()
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          Return a dict represents the dataset:
+              {
+                  "type": TYPE,
+                  "instances": [
+                      {
+                          "key_1": VALUE_1.1,
+                          "key_2": VALUE_1.2,
+                          ...
+                      },
+                      {
+                          "key_1": VALUE_2.1,
+                          "key_2": VALUE_2.2,
+                          ...
+                      },
+                      ...
+                  ]
+              }
+
+          A python dict object represents the content of this dataset.
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: to_list()
+
+      
+      Returns a list of instances.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: map(*args, **kwargs)
+
+      
+
+
+
+      :Parameters:
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.
+
+
+
+      :Returns:
+
+          **self** : Dataset object.
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_backend() -> Optional[str]
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          self.backend
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_backend_dataset()
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          self.backend_dataset
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_fingerprint()
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          Fingerprint of the backend_dataset which controls the cache
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_data_args()
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          self.data_args
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_type() -> str
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          self.type
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: save(file_path: str, format: str = 'json')
+
+      
+      Save the dataset to a json file.
+
+
+      :Parameters:
+
+          **file_path** : str.
+              The path to the file where the dataset will be saved.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: sample(n: int, seed: int = 42)
+
+      
+      Sample n instances from the dataset.
+
+
+      :Parameters:
+
+          **n** : int.
+              The number of instances to sample from the dataset.
+
+
+
+      :Returns:
+
+          **sample_dataset** : Dataset object.
+              A new dataset object containing the sampled instances.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: train_test_split(test_size: float = 0.2, shuffle: bool = True, seed: int = 42)
+
+      
+      Split the dataset into training and testing sets.
+
+
+      :Parameters:
+
+          **test_size** : float, default=0.2.
+              The proportion of the dataset that will be used for testing.
+
+
+
+      :Returns:
+
+          **train_dataset** : Dataset object.
+              A new dataset object containing the training instances.
+
+          **test_dataset** : Dataset object.
+              A new dataset object containing the testing instances.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: drop_instances(indices: list)
+
+      
+      Drop instances from the dataset.
+
+
+      :Parameters:
+
+          **indices** : list.
+              A list of indices of the instances to drop from the dataset.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: sanity_check(drop_invalid: bool = True)
+
+      
+      Perform a sanity check on the dataset.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: hf_dataset_sanity_check(drop_invalid: bool = True)
+
+      
+      Perform a sanity check on the HuggingFace dataset.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/datasets/index.rst.txt b/_sources/autoapi/lmflow/datasets/index.rst.txt
new file mode 100644
index 000000000..0dd7a83c5
--- /dev/null
+++ b/_sources/autoapi/lmflow/datasets/index.rst.txt
@@ -0,0 +1,651 @@
+lmflow.datasets
+===============
+
+.. py:module:: lmflow.datasets
+
+.. autoapi-nested-parse::
+
+   
+   This Python code defines a class Dataset with methods for initializing, loading,
+   and manipulating datasets from different backends such as Hugging Face and JSON.
+
+   The `Dataset` class includes methods for loading datasets from a dictionary and a Hugging
+   Face dataset, mapping datasets, and retrieving the backend dataset and arguments.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+
+Submodules
+----------
+
+.. toctree::
+   :maxdepth: 1
+
+   /autoapi/lmflow/datasets/dataset/index
+   /autoapi/lmflow/datasets/multi_modal_dataset/index
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.datasets.Dataset
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.datasets.is_multimodal_available
+
+
+Package Contents
+----------------
+
+.. py:function:: is_multimodal_available()
+
+.. py:class:: Dataset(data_args: lmflow.args.DatasetArguments = None, backend: str = 'huggingface', *args, **kwargs)
+
+   
+   Initializes the Dataset object with the given parameters.
+
+
+   :Parameters:
+
+       **data_args** : DatasetArguments object.
+           Contains the arguments required to load the dataset.
+
+       **backend** : str,  default="huggingface"
+           A string representing the dataset backend. Defaults to "huggingface".
+
+       **args** : Optional.
+           Positional arguments.
+
+       **kwargs** : Optional.
+           Keyword arguments.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: data_args
+      :value: None
+
+
+
+   .. py:attribute:: backend
+      :value: 'huggingface'
+
+
+
+   .. py:attribute:: backend_dataset
+      :value: None
+
+
+
+   .. py:attribute:: type
+      :value: None
+
+
+
+   .. py:attribute:: dataset_path
+
+
+   .. py:method:: __len__()
+
+
+   .. py:method:: _check_data_format()
+
+      
+      Checks if data type and data structure matches
+
+      Raise messages with hints if not matched.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: from_dict(dict_obj: dict, *args, **kwargs)
+
+      
+      Create a Dataset object from a dictionary.
+
+      Return a Dataset given a dict with format:
+          {
+              "type": TYPE,
+              "instances": [
+                  {
+                      "key_1": VALUE_1.1,
+                      "key_2": VALUE_1.2,
+                      ...
+                  },
+                  {
+                      "key_1": VALUE_2.1,
+                      "key_2": VALUE_2.2,
+                      ...
+                  },
+                  ...
+              ]
+          }
+
+      :Parameters:
+
+          **dict_obj** : dict.
+              A dictionary containing the dataset information.
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.
+
+
+
+      :Returns:
+
+          **self** : Dataset object.
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: create_from_dict(dict_obj, *args, **kwargs)
+      :classmethod:
+
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          Returns a Dataset object given a dict.
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: to_dict()
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          Return a dict represents the dataset:
+              {
+                  "type": TYPE,
+                  "instances": [
+                      {
+                          "key_1": VALUE_1.1,
+                          "key_2": VALUE_1.2,
+                          ...
+                      },
+                      {
+                          "key_1": VALUE_2.1,
+                          "key_2": VALUE_2.2,
+                          ...
+                      },
+                      ...
+                  ]
+              }
+
+          A python dict object represents the content of this dataset.
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: to_list()
+
+      
+      Returns a list of instances.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: map(*args, **kwargs)
+
+      
+
+
+
+      :Parameters:
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.
+
+
+
+      :Returns:
+
+          **self** : Dataset object.
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_backend() -> Optional[str]
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          self.backend
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_backend_dataset()
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          self.backend_dataset
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_fingerprint()
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          Fingerprint of the backend_dataset which controls the cache
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_data_args()
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          self.data_args
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_type() -> str
+
+      
+
+
+
+
+
+
+      :Returns:
+
+          self.type
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: save(file_path: str, format: str = 'json')
+
+      
+      Save the dataset to a json file.
+
+
+      :Parameters:
+
+          **file_path** : str.
+              The path to the file where the dataset will be saved.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: sample(n: int, seed: int = 42)
+
+      
+      Sample n instances from the dataset.
+
+
+      :Parameters:
+
+          **n** : int.
+              The number of instances to sample from the dataset.
+
+
+
+      :Returns:
+
+          **sample_dataset** : Dataset object.
+              A new dataset object containing the sampled instances.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: train_test_split(test_size: float = 0.2, shuffle: bool = True, seed: int = 42)
+
+      
+      Split the dataset into training and testing sets.
+
+
+      :Parameters:
+
+          **test_size** : float, default=0.2.
+              The proportion of the dataset that will be used for testing.
+
+
+
+      :Returns:
+
+          **train_dataset** : Dataset object.
+              A new dataset object containing the training instances.
+
+          **test_dataset** : Dataset object.
+              A new dataset object containing the testing instances.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: drop_instances(indices: list)
+
+      
+      Drop instances from the dataset.
+
+
+      :Parameters:
+
+          **indices** : list.
+              A list of indices of the instances to drop from the dataset.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: sanity_check(drop_invalid: bool = True)
+
+      
+      Perform a sanity check on the dataset.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: hf_dataset_sanity_check(drop_invalid: bool = True)
+
+      
+      Perform a sanity check on the HuggingFace dataset.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/datasets/multi_modal_dataset/index.rst.txt b/_sources/autoapi/lmflow/datasets/multi_modal_dataset/index.rst.txt
new file mode 100644
index 000000000..747bf15ea
--- /dev/null
+++ b/_sources/autoapi/lmflow/datasets/multi_modal_dataset/index.rst.txt
@@ -0,0 +1,177 @@
+lmflow.datasets.multi_modal_dataset
+===================================
+
+.. py:module:: lmflow.datasets.multi_modal_dataset
+
+.. autoapi-nested-parse::
+
+   This Python code defines a class Multi Modal Dataset.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset
+   lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.datasets.multi_modal_dataset.preprocess_multimodal_llava
+   lmflow.datasets.multi_modal_dataset.tokenizer_image_token
+   lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_plain
+   lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_v1
+
+
+Module Contents
+---------------
+
+.. py:class:: CustomMultiModalDataset(dataset_path: str, data_args: lmflow.args.DatasetArguments)
+
+   Bases: :py:obj:`torch.utils.data.Dataset`
+
+
+   
+   Dataset for Multi Modal data
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: data_dict
+
+
+   .. py:attribute:: data_args
+
+
+   .. py:attribute:: image_folder
+
+
+   .. py:method:: __len__()
+
+
+   .. py:method:: register_tokenizer(tokenizer, image_processor=None)
+
+
+   .. py:method:: __getitem__(i)
+
+
+.. py:function:: preprocess_multimodal_llava(sources, data_args)
+
+.. py:function:: tokenizer_image_token(prompt, tokenizer, image_token_index=IMAGE_TOKEN_INDEX, return_tensors=None)
+
+.. py:function:: preprocess_llama_from_llava_plain(sources, tokenizer: transformers.PreTrainedTokenizer, has_image: bool = False)
+
+   
+   This function just add the image in the front of text.
+   And don't add any prompt.
+   Args:
+       sources: The input data with text and image.
+       tokenizer: The tokenizer to process text.
+       has_image: Whether the input data has image.
+   Returns:
+       The input_ids and labels for the model.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:function:: preprocess_llama_from_llava_v1(sources, tokenizer: transformers.PreTrainedTokenizer, has_image: bool = False)
+
+   
+   This function add the prompt and then put the image after the prompt.
+   So it needs additional code to generate the target label.
+   Args:
+       sources: The input data with text and image.
+       tokenizer: The tokenizer to process text.
+       has_image: Whether the input data has image.
+   Returns:
+       The input_ids and labels for the model.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:class:: DataCollatorForSupervisedDataset
+
+   Bases: :py:obj:`object`
+
+
+   
+   Collate examples for supervised fine-tuning.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: tokenizer
+      :type:  transformers.PreTrainedTokenizer
+
+
+   .. py:method:: __call__(instances)
+
+
diff --git a/_sources/autoapi/lmflow/index.rst.txt b/_sources/autoapi/lmflow/index.rst.txt
new file mode 100644
index 000000000..774227f8a
--- /dev/null
+++ b/_sources/autoapi/lmflow/index.rst.txt
@@ -0,0 +1,40 @@
+lmflow
+======
+
+.. py:module:: lmflow
+
+
+Submodules
+----------
+
+.. toctree::
+   :maxdepth: 1
+
+   /autoapi/lmflow/args/index
+   /autoapi/lmflow/datasets/index
+   /autoapi/lmflow/models/index
+   /autoapi/lmflow/optim/index
+   /autoapi/lmflow/pipeline/index
+   /autoapi/lmflow/tokenization/index
+   /autoapi/lmflow/utils/index
+   /autoapi/lmflow/version/index
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.internal_version
+   lmflow.__version__
+
+
+Package Contents
+----------------
+
+.. py:data:: internal_version
+   :value: '0.0.9'
+
+
+.. py:data:: __version__
+
diff --git a/_sources/autoapi/lmflow/models/auto_model/index.rst.txt b/_sources/autoapi/lmflow/models/auto_model/index.rst.txt
new file mode 100644
index 000000000..02555c027
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/auto_model/index.rst.txt
@@ -0,0 +1,31 @@
+lmflow.models.auto_model
+========================
+
+.. py:module:: lmflow.models.auto_model
+
+.. autoapi-nested-parse::
+
+   Automatically get correct model type.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.models.auto_model.AutoModel
+
+
+Module Contents
+---------------
+
+.. py:class:: AutoModel
+
+   .. py:method:: get_model(model_args, *args, **kwargs)
+      :classmethod:
+
+
+
diff --git a/_sources/autoapi/lmflow/models/base_model/index.rst.txt b/_sources/autoapi/lmflow/models/base_model/index.rst.txt
new file mode 100644
index 000000000..ddc469cfe
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/base_model/index.rst.txt
@@ -0,0 +1,51 @@
+lmflow.models.base_model
+========================
+
+.. py:module:: lmflow.models.base_model
+
+.. autoapi-nested-parse::
+
+   Base model class.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.models.base_model.BaseModel
+
+
+Module Contents
+---------------
+
+.. py:class:: BaseModel(*args, **kwargs)
+
+   Bases: :py:obj:`abc.ABC`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
diff --git a/_sources/autoapi/lmflow/models/decoder_model/index.rst.txt b/_sources/autoapi/lmflow/models/decoder_model/index.rst.txt
new file mode 100644
index 000000000..16c36de6b
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/decoder_model/index.rst.txt
@@ -0,0 +1,61 @@
+lmflow.models.decoder_model
+===========================
+
+.. py:module:: lmflow.models.decoder_model
+
+.. autoapi-nested-parse::
+
+   A one-line summary of the module or program, terminated by a period.
+
+   Leave one blank line.  The rest of this docstring should contain an
+   overall description of the module or program.  Optionally, it may also
+   contain a brief description of exported classes and functions and/or usage
+   examples.
+
+   Typical usage example:
+
+     foo = ClassFoo()
+     bar = foo.FunctionBar()
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.models.decoder_model.DecoderModel
+
+
+Module Contents
+---------------
+
+.. py:class:: DecoderModel(*args, **kwargs)
+
+   Bases: :py:obj:`lmflow.models.base_model.BaseModel`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
diff --git a/_sources/autoapi/lmflow/models/encoder_decoder_model/index.rst.txt b/_sources/autoapi/lmflow/models/encoder_decoder_model/index.rst.txt
new file mode 100644
index 000000000..5f1c1c6a8
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/encoder_decoder_model/index.rst.txt
@@ -0,0 +1,61 @@
+lmflow.models.encoder_decoder_model
+===================================
+
+.. py:module:: lmflow.models.encoder_decoder_model
+
+.. autoapi-nested-parse::
+
+   A one-line summary of the module or program, terminated by a period.
+
+   Leave one blank line.  The rest of this docstring should contain an
+   overall description of the module or program.  Optionally, it may also
+   contain a brief description of exported classes and functions and/or usage
+   examples.
+
+   Typical usage example:
+
+     foo = ClassFoo()
+     bar = foo.FunctionBar()
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.models.encoder_decoder_model.EncoderDecoderModel
+
+
+Module Contents
+---------------
+
+.. py:class:: EncoderDecoderModel(*args, **kwargs)
+
+   Bases: :py:obj:`lmflow.models.base_model.BaseModel`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
diff --git a/_sources/autoapi/lmflow/models/hf_decoder_model/index.rst.txt b/_sources/autoapi/lmflow/models/hf_decoder_model/index.rst.txt
new file mode 100644
index 000000000..3b4da2737
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/hf_decoder_model/index.rst.txt
@@ -0,0 +1,427 @@
+lmflow.models.hf_decoder_model
+==============================
+
+.. py:module:: lmflow.models.hf_decoder_model
+
+.. autoapi-nested-parse::
+
+   This is a class called HFDecoderModel which is a wrapper around transformers model and
+   tokenizer classes. It has several methods such as __init__, tokenize, and train that are 
+   used for training and fine-tuning the model. The __init__ method takes in several arguments
+   such as model_args, tune_strategy, and ds_config, which are used to load the pretrained 
+   model and tokenizer, and initialize the training settings.
+
+   The tokenize method is used to tokenize the input text and return the input IDs and attention
+   masks that can be fed to the model for training or inference.
+
+   This class supports different tune_strategy options such as 'normal', 'none', 'lora', and
+   'adapter', which allow for different fine-tuning settings of the model. However, the 'lora'
+   and 'adapter' strategies are not yet implemented.
+
+   Overall, this class provides a convenient interface for loading and fine-tuning transformer
+   models and can be used for various NLP tasks such as language modeling, text classification,
+   and question answering.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.models.hf_decoder_model.logger
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.models.hf_decoder_model.HFDecoderModel
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: HFDecoderModel(model_args, tune_strategy='normal', ds_config=None, device='gpu', use_accelerator=False, *args, **kwargs)
+
+   Bases: :py:obj:`lmflow.models.decoder_model.DecoderModel`, :py:obj:`lmflow.models.hf_model_mixin.HFModelMixin`, :py:obj:`lmflow.models.interfaces.tunable.Tunable`
+
+
+   
+   Initializes a HFDecoderModel instance.
+
+
+   :Parameters:
+
+       **model_args**
+           Model arguments such as model name, path, revision, etc.
+
+       **tune_strategy** : str or none,  default="normal".
+           A string representing the dataset backend. Defaults to "huggingface".
+
+       **ds_config**
+           Deepspeed configuations.
+
+       **args** : Optional.
+           Positional arguments.
+
+       **kwargs** : Optional.
+           Keyword arguments.    
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: tokenize(dataset, add_special_tokens=True, *args, **kwargs) -> lmflow.datasets.dataset.Dataset
+
+      
+      Tokenize the full dataset.
+
+
+      :Parameters:
+
+          **dataset** : lmflow.datasets.Dataset.
+              ..
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.    
+
+
+
+      :Returns:
+
+          tokenized_datasets
+              The tokenized dataset, without any leading or trailing special
+              tokens (normally they are Begin-Of-Sentence or End-Of-Sentence
+              tokens).
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: encode(input: Union[str, List[str]], *args, **kwargs) -> Union[List[int], List[List[int]]]
+
+      
+      Perform encoding process of the tokenizer.
+
+
+      :Parameters:
+
+          **inputs** : str or list.
+              The text sequence.
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.    
+
+
+
+      :Returns:
+
+          outputs
+              if string input,return the tokenized inputs.
+              "Hello,world!"-> [101, 7592, 1010, 2088, 102]
+              if batch input,return {input_ids,attention_mask,token_type_ids}
+              ["Hello,world!","Hello!"]-> {'input_ids': tensor([[  101,  7592,  1010,  2088,   102],...),'attention_mask': tensor([[1, 1, 1, 1, 1],[0,0,1,1,1]])}
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: decode(input, *args, **kwargs) -> Union[str, List[str]]
+
+      
+      Perform decoding process of the tokenizer.
+
+
+      :Parameters:
+
+          **inputs** : list or tensor.
+              The token sequence.
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.    
+
+
+
+      :Returns:
+
+          outputs
+              The text decoded from the token inputs.
+              if batch input,return the list of text
+              [[101, 7592, 1010, 2088, 102],[101, 7592, 1010, 2088, 102]]-> ["Hello,world!","Hello,world!"
+              if single input,return the text
+              [101, 7592, 1010, 2088, 102]-> "Hello,world!"
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: inference(inputs, release_gpu: bool = False, use_vllm: bool = False, **kwargs)
+
+      
+      Perform generation process of the model.
+
+
+      :Parameters:
+
+          **inputs**
+              The sequence used as a prompt for the generation or as model inputs to the model.
+              When using vllm inference, this should be a string or a list of strings.
+              When using normal inference, this should be a tensor.
+
+          **release_gpu** : bool, optional
+              Whether to release the GPU resource after inference, by default False.
+
+          **use_vllm** : bool, optional
+              Whether to use VLLM for inference, by default False.
+
+          **kwargs** : Optional.
+              Keyword arguments.    
+
+
+
+      :Returns:
+
+          outputs
+              The generated sequence output 
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: __inference(inputs, *args, **kwargs)
+
+      
+      Perform generation process of the model.
+
+
+      :Parameters:
+
+          **inputs**
+              The **tokenized** sequence used as a prompt for the generation or as model inputs to the model.
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.    
+
+
+
+      :Returns:
+
+          outputs
+              The generated sequence output 
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: __vllm_inference(inputs: Union[str, List[str]], sampling_params: Optional[vllm.SamplingParams] = None, **kwargs) -> List[lmflow.utils.data_utils.VLLMInferenceResultWithInput]
+
+      
+      Perform VLLM inference process of the model.
+
+
+      :Parameters:
+
+          **inputs** : Union[str, List[str]]
+              Prompt(s), string or a list of strings.
+
+          **sampling_params** : Optional[SamplingParams], optional
+              vllm SamplingParams object, by default None.
+
+
+
+      :Returns:
+
+          List[VLLMInferenceResultWithInput]
+              Return a list of VLLMInferenceResultWithInput, where each
+              element contains the input prompt and the corresponding output.
+              
+              When `sampling_params.detokenize = True`, the output would be a list of strings,
+              contains sampling_params.n samples for the corresponding prompt.
+              
+              When `sampling_params.detokenize = False`, return a list of list of ints 
+              (token ids, no decoding after generation).
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: prepare_inputs_for_inference(dataset: lmflow.datasets.dataset.Dataset, apply_chat_template: bool = True, enable_distributed_inference: bool = False, use_vllm: bool = False, **kwargs) -> Union[List[str], ray.data.Dataset, Dict[str, torch.Tensor]]
+
+      
+      Prepare inputs for inference.
+
+
+      :Parameters:
+
+          **dataset** : lmflow.datasets.Dataset.
+              The dataset used for inference.
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.    
+
+
+
+      :Returns:
+
+          outputs
+              The prepared inputs for inference.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: __prepare_inputs_for_vllm_inference(dataset: lmflow.datasets.dataset.Dataset, apply_chat_template: bool = True, enable_distributed_inference: bool = False) -> Union[List[str], ray.data.Dataset]
+
+
+   .. py:method:: __prepare_inputs_for_inference(dataset: lmflow.datasets.dataset.Dataset, **kwargs)
+      :abstractmethod:
+
+
+
+   .. py:method:: merge_lora_weights()
+
+
+   .. py:method:: get_peft_without_qlora()
+
+
+   .. py:method:: save(dir, save_full_model=False, *args, **kwargs)
+
+      
+      Perform generation process of the model.
+
+
+      :Parameters:
+
+          **dir**
+              The directory to save model and tokenizer
+
+          **save_full_model** : Optional.
+              Whether to save full model.
+
+          **kwargs** : Optional.
+              Keyword arguments.    
+
+
+
+      :Returns:
+
+          outputs
+              The generated sequence output 
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/models/hf_encoder_decoder_model/index.rst.txt b/_sources/autoapi/lmflow/models/hf_encoder_decoder_model/index.rst.txt
new file mode 100644
index 000000000..fa6f847d4
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/hf_encoder_decoder_model/index.rst.txt
@@ -0,0 +1,363 @@
+lmflow.models.hf_encoder_decoder_model
+======================================
+
+.. py:module:: lmflow.models.hf_encoder_decoder_model
+
+.. autoapi-nested-parse::
+
+   This is a class called HFDecoderModel which is a wrapper around transformers model and
+   tokenizer classes. It has several methods such as __init__, tokenize, and train that are
+   used for training and fine-tuning the model. The __init__ method takes in several arguments
+   such as model_args, tune_strategy, and ds_config, which are used to load the pretrained
+   model and tokenizer, and initialize the training settings.
+
+   The tokenize method is used to tokenize the input text and return the input IDs and attention
+   masks that can be fed to the model for training or inference.
+
+   This class supports different tune_strategy options such as 'normal', 'none', 'lora', and
+   'adapter', which allow for different fine-tuning settings of the model. However, the 'lora'
+   and 'adapter' strategies are not yet implemented.
+
+   Overall, this class provides a convenient interface for loading and fine-tuning transformer
+   models and can be used for various NLP tasks such as language modeling, text classification,
+   and question answering.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.models.hf_encoder_decoder_model.logger
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: HFEncoderDecoderModel(model_args, tune_strategy='normal', ds_config=None, device='gpu', use_accelerator=False, custom_model=False, with_deepspeed=True, pipeline_args=None, *args, **kwargs)
+
+   Bases: :py:obj:`lmflow.models.encoder_decoder_model.EncoderDecoderModel`, :py:obj:`lmflow.models.interfaces.tunable.Tunable`
+
+
+   
+   Initializes a HFEncoderDecoderModel instance.
+
+
+   :Parameters:
+
+       **model_args**
+           Model arguments such as model name, path, revision, etc.
+
+       **tune_strategy** : str or none,  default="normal".
+           A string representing the dataset backend. Defaults to "huggingface".
+
+       **ds_config**
+           Deepspeed configuations.
+
+       **args** : Optional.
+           Positional arguments.
+
+       **kwargs** : Optional.
+           Keyword arguments.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: device
+      :value: 'gpu'
+
+
+
+   .. py:method:: tokenize(dataset, *args, **kwargs)
+      :abstractmethod:
+
+
+      
+      Tokenize the full dataset.
+
+
+      :Parameters:
+
+          **dataset**
+              Text dataset.
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.
+
+
+
+      :Returns:
+
+          tokenized_datasets
+              The tokenized dataset.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: encode(input: Union[str, List[str]], *args, **kwargs) -> Union[List[int], List[List[int]]]
+
+      
+      Perform encoding process of the tokenizer.
+
+
+      :Parameters:
+
+          **inputs** : str or list.
+              The text sequence.
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.
+
+
+
+      :Returns:
+
+          outputs
+              The tokenized inputs.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: decode(input, *args, **kwargs) -> Union[str, List[str]]
+
+      
+      Perform decoding process of the tokenizer.
+
+
+      :Parameters:
+
+          **inputs** : list.
+              The token sequence.
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.
+
+
+
+      :Returns:
+
+          outputs
+              The text decoded from the token inputs.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: inference(inputs, *args, **kwargs)
+
+      
+      Perform generation process of the model.
+
+
+      :Parameters:
+
+          **inputs**
+              The sequence used as a prompt for the generation or as model inputs to the model.
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.
+
+
+
+      :Returns:
+
+          outputs
+              The generated sequence output
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: merge_lora_weights()
+
+
+   .. py:method:: save(dir, save_full_model=False, *args, **kwargs)
+
+      
+      Perform generation process of the model.
+
+
+      :Parameters:
+
+          **dir**
+              The directory to save model and tokenizer
+
+          **save_full_model** : Optional.
+              Whether to save full model.
+
+          **kwargs** : Optional.
+              Keyword arguments.
+
+
+
+      :Returns:
+
+          outputs
+              The generated sequence output
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_max_length()
+
+      
+      Return max acceptable input length in terms of tokens.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_tokenizer()
+
+      
+      Return the tokenizer of the model.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_backend_model()
+
+      
+      Return the backend model.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/models/hf_model_mixin/index.rst.txt b/_sources/autoapi/lmflow/models/hf_model_mixin/index.rst.txt
new file mode 100644
index 000000000..24007cde9
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/hf_model_mixin/index.rst.txt
@@ -0,0 +1,301 @@
+lmflow.models.hf_model_mixin
+============================
+
+.. py:module:: lmflow.models.hf_model_mixin
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.models.hf_model_mixin.logger
+   lmflow.models.hf_model_mixin.HF_AUTOMODEL_MAPPING
+   lmflow.models.hf_model_mixin.HF_AUTOMODEL_TYPE
+   lmflow.models.hf_model_mixin.LORA_TARGET_MODULES_MAPPING
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.models.hf_model_mixin.HFModelMixin
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:data:: HF_AUTOMODEL_MAPPING
+
+.. py:data:: HF_AUTOMODEL_TYPE
+
+.. py:data:: LORA_TARGET_MODULES_MAPPING
+
+.. py:class:: HFModelMixin(model_args: lmflow.args.ModelArguments, do_train: bool, ds_config=None, device: Optional[str] = 'gpu', use_accelerator: bool = False, hf_auto_model_additional_args: Optional[Dict] = None, *args, **kwargs)
+
+   Bases: :py:obj:`lmflow.models.base_model.BaseModel`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: device
+      :value: 'gpu'
+
+
+
+   .. py:attribute:: model_args
+
+
+   .. py:attribute:: hf_auto_model
+
+
+   .. py:attribute:: use_accelerator
+      :value: False
+
+
+
+   .. py:attribute:: ds_config
+      :value: None
+
+
+
+   .. py:attribute:: do_train
+
+
+   .. py:attribute:: tokenizer
+
+
+   .. py:attribute:: torch_dtype
+
+
+   .. py:attribute:: hf_model_config
+
+
+   .. py:attribute:: quant_config
+      :value: None
+
+
+
+   .. py:attribute:: peft_config
+      :value: None
+
+
+
+   .. py:attribute:: _activated
+      :value: False
+
+
+
+   .. py:method:: __prepare_tokenizer(model_args: lmflow.args.ModelArguments) -> Union[transformers.PreTrainedTokenizer, transformers.PreTrainedTokenizerFast]
+
+
+   .. py:method:: __prepare_dtype(model_args: lmflow.args.ModelArguments) -> torch.dtype
+
+
+   .. py:method:: __prepare_model_config(model_args: lmflow.args.ModelArguments, hf_auto_model_additional_args: Optional[Dict] = None)
+
+      
+      Prepare model configuration for hf auto register,
+      Parameters
+      ----------
+      model_args : ModelArguments
+          LMFlow model arguments.
+      hf_auto_model_additional_args : Optional[Dict], optional
+          Special configurations such as `num_labels` in `AutoModelForSequenceClassification` 
+          (commonly used in reward modeling) will not preset in __prepare_model_config, 
+          so it should be passed in hf_auto_model_additional_args.
+      Returns
+      -------
+      config : ModelConfig
+          hf model config.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: __prepare_quant_config(model_args: lmflow.args.ModelArguments)
+
+
+   .. py:method:: __prepare_peft_config(model_args: lmflow.args.ModelArguments)
+
+
+   .. py:method:: __model_module_inject(model_args: lmflow.args.ModelArguments) -> None
+
+      
+      Override some model modules with custom implementations.
+
+      Current implementations:
+      - Position interpolation (model_args.do_rope_scaling): 
+          replace llama embeddings with condense embeddings.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: __prepare_model_for_training(model_args: lmflow.args.ModelArguments, hf_auto_model: HF_AUTOMODEL_TYPE)
+
+
+   .. py:method:: __prepare_model_for_inference(model_args: lmflow.args.ModelArguments, hf_auto_model: HF_AUTOMODEL_TYPE, use_accelerator: bool, ds_config)
+
+
+   .. py:method:: __prepare_model_for_vllm_inference(model_args: lmflow.args.ModelArguments, vllm_gpu_memory_utilization: float, vllm_tensor_parallel_size: int)
+
+
+   .. py:method:: __prepare_model_post_process()
+
+
+   .. py:method:: activate_model_for_inference(use_vllm: bool = False, **kwargs)
+
+
+   .. py:method:: deactivate_model_for_inference(use_vllm: bool = False)
+
+      
+      Deactivate the model and release the resources.
+
+      NOTE: Currently, VLLM doesn't have an official way to do this, and the
+      implementation below cannot release all gpu resources by our observation.
+      Thus this method is just a placeholder for future implementation. See: 
+      [Github issue](https://github.com/vllm-project/vllm/issues/1908)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_max_length()
+
+      
+      Return max acceptable input length in terms of tokens.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_tokenizer()
+
+      
+      Return the tokenizer of the model.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_backend_model()
+
+      
+      Return the backend model.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/models/hf_text_regression_model/index.rst.txt b/_sources/autoapi/lmflow/models/hf_text_regression_model/index.rst.txt
new file mode 100644
index 000000000..313095aab
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/hf_text_regression_model/index.rst.txt
@@ -0,0 +1,268 @@
+lmflow.models.hf_text_regression_model
+======================================
+
+.. py:module:: lmflow.models.hf_text_regression_model
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.models.hf_text_regression_model.logger
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.models.hf_text_regression_model.HFTextRegressionModel
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: HFTextRegressionModel(model_args: lmflow.args.ModelArguments, tune_strategy: str = 'normal', ds_config=None, device='gpu', use_accelerator=False, *args, **kwargs)
+
+   Bases: :py:obj:`lmflow.models.text_regression_model.TextRegressionModel`, :py:obj:`lmflow.models.hf_model_mixin.HFModelMixin`, :py:obj:`lmflow.models.interfaces.tunable.Tunable`
+
+
+   
+   Initializes a HFTextRegressionModel instance.
+
+
+   :Parameters:
+
+       **model_args**
+           Model arguments such as model name, path, revision, etc.
+
+       **tune_strategy** : str or none,  default="normal".
+           A string representing the dataset backend. Defaults to "huggingface".
+
+       **ds_config**
+           Deepspeed configuations.
+
+       **args** : Optional.
+           Positional arguments.
+
+       **kwargs** : Optional.
+           Keyword arguments.    
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: tokenize(dataset: lmflow.datasets.dataset.Dataset, add_special_tokens=True, *args, **kwargs)
+
+      
+      Tokenize the full dataset.
+
+
+      :Parameters:
+
+          **dataset** : lmflow.datasets.Dataset.
+              ..
+
+          **args** : Optional.
+              Positional arguments.
+
+          **kwargs** : Optional.
+              Keyword arguments.    
+
+
+
+      :Returns:
+
+          tokenized_datasets
+              The tokenized dataset, without any leading or trailing special
+              tokens (normally they are Begin-Of-Sentence or End-Of-Sentence
+              tokens).
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: inference(inputs, release_gpu: bool = False, use_vllm: bool = False, **kwargs) -> Union[List[float], transformers.modeling_outputs.SequenceClassifierOutputWithPast]
+
+      
+      Perform generation process of the model.
+
+
+      :Parameters:
+
+          **inputs**
+              The sequence used as a prompt for the generation or as model inputs to the model.
+              When using vllm inference, this should be a string or a list of strings.
+              When using normal inference, this should be a tensor.
+
+          **release_gpu** : bool, optional
+              Whether to release the GPU resource after inference, by default False.
+
+          **use_vllm** : bool, optional
+              Whether to use VLLM for inference, by default False.
+
+          **kwargs** : Optional.
+              Keyword arguments.    
+
+
+
+      :Returns:
+
+          outputs
+              The generated sequence output 
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: __inference(inputs, **kwargs)
+
+      
+      Perform generation process of the model.
+
+
+      :Parameters:
+
+          **inputs**
+              The **tokenized** sequence used as a prompt for the generation or as model inputs to the model.
+
+          **kwargs** : Optional.
+              Keyword arguments.    
+
+
+
+      :Returns:
+
+          outputs
+              The generated sequence output 
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: __vllm_inference(inputs: Union[str, List[str]], sampling_params: Optional[vllm.SamplingParams] = None, **kwargs) -> Union[List[List[str]], List[List[List[int]]]]
+      :abstractmethod:
+
+
+      
+      Perform VLLM inference process of the model.
+
+
+      :Parameters:
+
+          **inputs** : Union[str, List[str]]
+              Prompt(s), string or a list of strings.
+
+          **sampling_params** : Optional[SamplingParams], optional
+              vllm SamplingParams object, by default None.
+
+
+
+      :Returns:
+
+          
+              ..
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: prepare_inputs_for_inference(dataset: lmflow.datasets.dataset.Dataset, enable_distributed_inference: bool = False, use_vllm: bool = False, **kwargs) -> Union[lmflow.datasets.dataset.Dataset, ray.data.Dataset]
+
+
+   .. py:method:: postprocess_inference_outputs(dataset: lmflow.datasets.dataset.Dataset, scores: Union[List[float], List[List[float]]])
+      :staticmethod:
+
+
+
+   .. py:method:: postprocess_distributed_inference_outputs(dataset: lmflow.datasets.dataset.Dataset, inference_result: List[lmflow.utils.data_utils.RewardModelInferenceResultWithInput])
+      :staticmethod:
+
+
+
+   .. py:method:: save(dir, *args, **kwargs)
+
+      
+      Perform generation process of the model.
+
+
+      :Parameters:
+
+          **dir**
+              The directory to save model and tokenizer
+
+          **kwargs** : Optional.
+              Keyword arguments.    
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/models/index.rst.txt b/_sources/autoapi/lmflow/models/index.rst.txt
new file mode 100644
index 000000000..a27f13d57
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/index.rst.txt
@@ -0,0 +1,27 @@
+lmflow.models
+=============
+
+.. py:module:: lmflow.models
+
+
+Submodules
+----------
+
+.. toctree::
+   :maxdepth: 1
+
+   /autoapi/lmflow/models/auto_model/index
+   /autoapi/lmflow/models/base_model/index
+   /autoapi/lmflow/models/decoder_model/index
+   /autoapi/lmflow/models/encoder_decoder_model/index
+   /autoapi/lmflow/models/hf_decoder_model/index
+   /autoapi/lmflow/models/hf_encoder_decoder_model/index
+   /autoapi/lmflow/models/hf_model_mixin/index
+   /autoapi/lmflow/models/hf_text_regression_model/index
+   /autoapi/lmflow/models/interfaces/index
+   /autoapi/lmflow/models/regression_model/index
+   /autoapi/lmflow/models/text_regression_model/index
+   /autoapi/lmflow/models/vision2seq_model/index
+   /autoapi/lmflow/models/vision_encoder/index
+
+
diff --git a/_sources/autoapi/lmflow/models/interfaces/index.rst.txt b/_sources/autoapi/lmflow/models/interfaces/index.rst.txt
new file mode 100644
index 000000000..ca86d17cd
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/interfaces/index.rst.txt
@@ -0,0 +1,15 @@
+lmflow.models.interfaces
+========================
+
+.. py:module:: lmflow.models.interfaces
+
+
+Submodules
+----------
+
+.. toctree::
+   :maxdepth: 1
+
+   /autoapi/lmflow/models/interfaces/tunable/index
+
+
diff --git a/_sources/autoapi/lmflow/models/interfaces/tunable/index.rst.txt b/_sources/autoapi/lmflow/models/interfaces/tunable/index.rst.txt
new file mode 100644
index 000000000..beb8ecd7e
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/interfaces/tunable/index.rst.txt
@@ -0,0 +1,51 @@
+lmflow.models.interfaces.tunable
+================================
+
+.. py:module:: lmflow.models.interfaces.tunable
+
+.. autoapi-nested-parse::
+
+   Tunable class
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.models.interfaces.tunable.Tunable
+
+
+Module Contents
+---------------
+
+.. py:class:: Tunable
+
+   Bases: :py:obj:`abc.ABC`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
diff --git a/_sources/autoapi/lmflow/models/regression_model/index.rst.txt b/_sources/autoapi/lmflow/models/regression_model/index.rst.txt
new file mode 100644
index 000000000..7b68aa48d
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/regression_model/index.rst.txt
@@ -0,0 +1,51 @@
+lmflow.models.regression_model
+==============================
+
+.. py:module:: lmflow.models.regression_model
+
+.. autoapi-nested-parse::
+
+   General regression model.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.models.regression_model.RegressionModel
+
+
+Module Contents
+---------------
+
+.. py:class:: RegressionModel(*args, **kwargs)
+
+   Bases: :py:obj:`lmflow.models.base_model.BaseModel`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
diff --git a/_sources/autoapi/lmflow/models/text_regression_model/index.rst.txt b/_sources/autoapi/lmflow/models/text_regression_model/index.rst.txt
new file mode 100644
index 000000000..96bb6ae00
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/text_regression_model/index.rst.txt
@@ -0,0 +1,114 @@
+lmflow.models.text_regression_model
+===================================
+
+.. py:module:: lmflow.models.text_regression_model
+
+.. autoapi-nested-parse::
+
+   A model maps "text_only" data to float.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.models.text_regression_model.TextRegressionModel
+
+
+Module Contents
+---------------
+
+.. py:class:: TextRegressionModel(model_args, *args, **kwargs)
+
+   Bases: :py:obj:`lmflow.models.regression_model.RegressionModel`
+
+
+   
+   Initializes a TextRegressionModel instance.
+
+
+   :Parameters:
+
+       **model_args**
+           Model arguments such as model name, path, revision, etc.
+
+       **args** : Optional.
+           Positional arguments.
+
+       **kwargs** : Optional.
+           Keyword arguments.    
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: inference_func
+      :value: None
+
+
+
+   .. py:method:: register_inference_function(inference_func)
+
+      
+      Registers a regression function.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: inference(inputs: lmflow.datasets.dataset.Dataset)
+
+      
+      Gets regression results of a given dataset.
+
+      :inputs: Dataset object, only accept type "text_only".
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/models/vision2seq_model/index.rst.txt b/_sources/autoapi/lmflow/models/vision2seq_model/index.rst.txt
new file mode 100644
index 000000000..c11b576ef
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/vision2seq_model/index.rst.txt
@@ -0,0 +1,205 @@
+lmflow.models.vision2seq_model
+==============================
+
+.. py:module:: lmflow.models.vision2seq_model
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.models.vision2seq_model.CustomAutoVision2SeqModel
+
+
+Module Contents
+---------------
+
+.. py:class:: CustomAutoVision2SeqModel(config: transformers.Blip2Config, image_encoder_name_or_path=None, qformer_name_or_path=None, language_model_name_or_path=None, low_resource=False)
+
+   Bases: :py:obj:`transformers.Blip2ForConditionalGeneration`, :py:obj:`lmflow.models.base_model.BaseModel`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: custom_vision_model
+
+
+   .. py:attribute:: with_qformer
+
+
+   .. py:attribute:: language_model
+
+
+   .. py:attribute:: hidden_size
+
+
+   .. py:method:: get_backend_model()
+
+
+   .. py:method:: vision_model_from_pretrained(pretrained_path)
+
+
+   .. py:method:: qformer_from_pretrained(pretrained_path)
+
+
+   .. py:method:: language_model_from_pretrained(pretrained_path, low_resource=False, use_prompt_cache=False)
+
+
+   .. py:method:: vision_feature_select(image_forward_outs)
+
+
+   .. py:method:: register_prompt_cache(prompt_ids, prompt_keys_values)
+
+      
+      Udpate the prompt id and embedding for reuse in the future
+
+      Args:
+          prompt_ids (torch.LongTensor): The id of the prompt.
+          prompt_keys_values (torch.FloatTensor): The embedding of the prompt.
+
+      Returns:
+          None
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: save_prompt_cache(path)
+
+      
+      Save prompt embedding and id.
+
+      Args:
+          path: The path to save the prompt embedding and id.
+
+      Returns:
+          None
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: load_prompt_cache(path)
+
+      
+      Load prompt embedding and id.
+      Args:
+          path: The path to load the prompt embedding and id.
+
+      Returns:
+          None
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_tokenizer()
+
+
+   .. py:method:: forward(input_ids: torch.LongTensor = None, pixel_values: Optional[torch.FloatTensor] = None, images: Optional[torch.FloatTensor] = None, attention_mask: Optional[torch.Tensor] = None, past_key_values: Optional[List[torch.FloatTensor]] = None, inputs_embeds: Optional[torch.FloatTensor] = None, labels: Optional[torch.LongTensor] = None, use_cache: Optional[bool] = None, output_attentions: Optional[bool] = None, output_hidden_states: Optional[bool] = None, return_dict: Optional[bool] = None, image_token_indexes: Optional[List] = [0], one_sample_multiple_images: bool = False) -> Union[Tuple, transformers.modeling_outputs.CausalLMOutputWithPast]
+
+
+   .. py:method:: processor_image_token_in_minigpt4(input_ids, language_model_inputs, attention_mask, image_token_indexes, pixel_values, batch_size=1)
+
+
+   .. py:method:: generate(pixel_values: torch.FloatTensor, input_ids: Optional[torch.LongTensor] = None, attention_mask: Optional[torch.LongTensor] = None, image_token_indexes: Optional[List] = [0], one_sample_multiple_images: Optional[bool] = False, images: Optional[torch.LongTensor] = None, **generate_kwargs) -> torch.LongTensor
+
+      
+      Overrides `generate` function to be able to use the model as a conditional generator.
+
+      Args:
+          pixel_values (`torch.FloatTensor` of shape (batch_size, num_channels, height, width)):
+              Input images to be processed.
+          input_ids (`torch.LongTensor` of shape (batch_size, sequence_length), *optional*):
+              The sequence used as a prompt for the generation.
+          attention_mask (`torch.LongTensor` of shape (batch_size, sequence_length), *optional*):
+              Mask to avoid performing attention on padding token indices
+          image_token_indexes (bool, *optional*):
+              The index for inserting the image tokens.
+          one_sample_multiple_images: (bool, *optional*):
+              The flag for inference that the input batch size is 1 and contain multiple images.
+
+      Returns:
+          captions (list): A list of strings of length batch_size * num_captions.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/models/vision_encoder/clip_encoder/index.rst.txt b/_sources/autoapi/lmflow/models/vision_encoder/clip_encoder/index.rst.txt
new file mode 100644
index 000000000..4bc695829
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/vision_encoder/clip_encoder/index.rst.txt
@@ -0,0 +1,101 @@
+lmflow.models.vision_encoder.clip_encoder
+=========================================
+
+.. py:module:: lmflow.models.vision_encoder.clip_encoder
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.models.vision_encoder.clip_encoder.build_vision_tower
+
+
+Module Contents
+---------------
+
+.. py:function:: build_vision_tower(vision_tower_cfg, **kwargs)
+
+.. py:class:: CLIPVisionTower(vision_tower, args, delay_load=False)
+
+   Bases: :py:obj:`torch.nn.Module`
+
+
+   .. py:attribute:: is_loaded
+      :value: False
+
+
+
+   .. py:attribute:: vision_tower_name
+
+
+   .. py:attribute:: select_layer
+
+
+   .. py:attribute:: select_feature
+
+
+   .. py:method:: load_model()
+
+
+   .. py:method:: encode_images(images, language_projection)
+
+
+   .. py:method:: feature_select(image_forward_outs)
+
+
+   .. py:method:: forward(images)
+
+
+   .. py:property:: dummy_feature
+
+
+   .. py:property:: dtype
+
+
+   .. py:property:: device
+
+
+   .. py:property:: config
+
+
+   .. py:property:: hidden_size
+
+
+   .. py:property:: num_patches
+
+
+   .. py:method:: prepare_inputs_labels_for_multimodal(input_ids, attention_mask, past_key_values, labels, images, language_projection=None, language_model=None, **kwargs)
+
+      
+      Copy from the LLAVA code base.
+      Should be polished.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/models/vision_encoder/index.rst.txt b/_sources/autoapi/lmflow/models/vision_encoder/index.rst.txt
new file mode 100644
index 000000000..fac6977d4
--- /dev/null
+++ b/_sources/autoapi/lmflow/models/vision_encoder/index.rst.txt
@@ -0,0 +1,28 @@
+lmflow.models.vision_encoder
+============================
+
+.. py:module:: lmflow.models.vision_encoder
+
+
+Submodules
+----------
+
+.. toctree::
+   :maxdepth: 1
+
+   /autoapi/lmflow/models/vision_encoder/clip_encoder/index
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.models.vision_encoder.build_vision_tower
+
+
+Package Contents
+----------------
+
+.. py:function:: build_vision_tower(vision_tower_cfg, **kwargs)
+
diff --git a/_sources/autoapi/lmflow/optim/adabelief/index.rst.txt b/_sources/autoapi/lmflow/optim/adabelief/index.rst.txt
new file mode 100644
index 000000000..2a8515da5
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/adabelief/index.rst.txt
@@ -0,0 +1,97 @@
+lmflow.optim.adabelief
+======================
+
+.. py:module:: lmflow.optim.adabelief
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.adabelief.AdaBelief
+
+
+Module Contents
+---------------
+
+.. py:class:: AdaBelief(params, lr=0.001, betas=(0.9, 0.999), eps=1e-16, weight_decay=0, amsgrad=False, weight_decouple=True, fixed_decay=False, rectify=True, degenerated_to_sgd=True, print_change_log=True)
+
+   Bases: :py:obj:`torch.optim.optimizer.Optimizer`
+
+
+   
+   Implements AdaBelief algorithm. Modified from Adam in PyTorch
+   reference: AdaBelief Optimizer, adapting stepsizes by the belief in observed gradients, NeurIPS 2020
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: degenerated_to_sgd
+      :value: True
+
+
+
+   .. py:attribute:: weight_decouple
+      :value: True
+
+
+
+   .. py:attribute:: rectify
+      :value: True
+
+
+
+   .. py:attribute:: fixed_decay
+      :value: False
+
+
+
+   .. py:method:: __setstate__(state)
+
+
+   .. py:method:: reset()
+
+
+   .. py:method:: step(closure=None)
+
+      
+      Performs a single optimization step.
+      Arguments:
+          closure (callable, optional): A closure that reevaluates the model
+              and returns the loss.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/optim/adabound/index.rst.txt b/_sources/autoapi/lmflow/optim/adabound/index.rst.txt
new file mode 100644
index 000000000..45f8de11c
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/adabound/index.rst.txt
@@ -0,0 +1,80 @@
+lmflow.optim.adabound
+=====================
+
+.. py:module:: lmflow.optim.adabound
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.adabound.AdaBound
+
+
+Module Contents
+---------------
+
+.. py:class:: AdaBound(params, lr: float = 0.001, betas=(0.9, 0.999), final_lr: float = 0.1, gamma: float = 0.001, eps: float = 1e-08, weight_decay: float = 0, amsbound: bool = False)
+
+   Bases: :py:obj:`torch.optim.optimizer.Optimizer`
+
+
+   
+   Implements AdaBound algorithm.
+
+   It has been proposed in `Adaptive Gradient Methods with Dynamic Bound of
+   Learning Rate
+   https://arxiv.org/abs/1902.09843
+   Note:
+       Reference code: https://github.com/Luolc/AdaBound
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: base_lrs
+
+
+   .. py:method:: __setstate__(state) -> None
+
+
+   .. py:method:: step(closure=None)
+
+      
+      Performs a single optimization step.
+
+      Arguments:
+          closure: A closure that reevaluates the model and returns the loss.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/optim/adadelta/index.rst.txt b/_sources/autoapi/lmflow/optim/adadelta/index.rst.txt
new file mode 100644
index 000000000..dc71d4e69
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/adadelta/index.rst.txt
@@ -0,0 +1,25 @@
+lmflow.optim.adadelta
+=====================
+
+.. py:module:: lmflow.optim.adadelta
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.adadelta.Adadelta
+
+
+Module Contents
+---------------
+
+.. py:class:: Adadelta(params, lr=1.0, rho=0.95, eps=1e-06)
+
+   Bases: :py:obj:`torch.optim.optimizer.Optimizer`
+
+
+   .. py:method:: step(closure=None)
+
+
diff --git a/_sources/autoapi/lmflow/optim/adagrad/index.rst.txt b/_sources/autoapi/lmflow/optim/adagrad/index.rst.txt
new file mode 100644
index 000000000..831f76385
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/adagrad/index.rst.txt
@@ -0,0 +1,25 @@
+lmflow.optim.adagrad
+====================
+
+.. py:module:: lmflow.optim.adagrad
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.adagrad.AdaGrad
+
+
+Module Contents
+---------------
+
+.. py:class:: AdaGrad(params, lr=0.001, eps=1e-08, weight_decay=0)
+
+   Bases: :py:obj:`torch.optim.Optimizer`
+
+
+   .. py:method:: step(closure=None)
+
+
diff --git a/_sources/autoapi/lmflow/optim/adam/index.rst.txt b/_sources/autoapi/lmflow/optim/adam/index.rst.txt
new file mode 100644
index 000000000..42729f39d
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/adam/index.rst.txt
@@ -0,0 +1,25 @@
+lmflow.optim.adam
+=================
+
+.. py:module:: lmflow.optim.adam
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.adam.Adam
+
+
+Module Contents
+---------------
+
+.. py:class:: Adam(params, lr=0.001, betas=(0.9, 0.999), eps=1e-08)
+
+   Bases: :py:obj:`torch.optim.optimizer.Optimizer`
+
+
+   .. py:method:: step(closure=None)
+
+
diff --git a/_sources/autoapi/lmflow/optim/adamax/index.rst.txt b/_sources/autoapi/lmflow/optim/adamax/index.rst.txt
new file mode 100644
index 000000000..8622d08ae
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/adamax/index.rst.txt
@@ -0,0 +1,28 @@
+lmflow.optim.adamax
+===================
+
+.. py:module:: lmflow.optim.adamax
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.adamax.Adamax
+
+
+Module Contents
+---------------
+
+.. py:class:: Adamax(params, lr=0.002, betas=(0.9, 0.999), eps=1e-08, weight_decay=0)
+
+   Bases: :py:obj:`torch.optim.optimizer.Optimizer`
+
+
+   .. py:method:: __setstate__(state)
+
+
+   .. py:method:: step(closure=None)
+
+
diff --git a/_sources/autoapi/lmflow/optim/adamp/index.rst.txt b/_sources/autoapi/lmflow/optim/adamp/index.rst.txt
new file mode 100644
index 000000000..d7c1d93a6
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/adamp/index.rst.txt
@@ -0,0 +1,93 @@
+lmflow.optim.adamp
+==================
+
+.. py:module:: lmflow.optim.adamp
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.adamp.AdamP
+
+
+Module Contents
+---------------
+
+.. py:class:: AdamP(params, lr: float = 0.001, betas=(0.9, 0.999), eps: float = 1e-08, weight_decay: float = 0, delta: float = 0.1, wd_ratio: float = 0.1, nesterov: bool = False)
+
+   Bases: :py:obj:`torch.optim.optimizer.Optimizer`
+
+
+   
+   Implements AdamP algorithm.
+
+   It has been proposed in `Slowing Down the Weight Norm Increase in
+   Momentum-based Optimizers`
+   https://arxiv.org/abs/2006.08217
+
+   Note:
+       Reference code: https://github.com/clovaai/AdamP
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: _channel_view(x)
+      :staticmethod:
+
+
+
+   .. py:method:: _layer_view(x)
+      :staticmethod:
+
+
+
+   .. py:method:: _cosine_similarity(x, y, eps, view_func)
+      :staticmethod:
+
+
+
+   .. py:method:: _projection(p, grad, perturb, delta, wd_ratio, eps)
+
+
+   .. py:method:: step(closure=None)
+
+      
+      Performs a single optimization step.
+
+      Arguments:
+          closure: A closure that reevaluates the model and returns the loss.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/optim/adamw_schedule_free/index.rst.txt b/_sources/autoapi/lmflow/optim/adamw_schedule_free/index.rst.txt
new file mode 100644
index 000000000..6557f514d
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/adamw_schedule_free/index.rst.txt
@@ -0,0 +1,82 @@
+lmflow.optim.adamw_schedule_free
+================================
+
+.. py:module:: lmflow.optim.adamw_schedule_free
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.adamw_schedule_free.AdamWScheduleFree
+
+
+Module Contents
+---------------
+
+.. py:class:: AdamWScheduleFree(params, lr=0.0025, betas=(0.9, 0.999), eps=1e-08, weight_decay=0, warmup_steps=0, r=0.0, weight_lr_power=2.0, foreach=hasattr(torch, '_foreach_mul_'))
+
+   Bases: :py:obj:`torch.optim.Optimizer`
+
+
+   
+   Schedule-Free AdamW
+   As the name suggests, no scheduler is needed with this optimizer. 
+   To add warmup, rather than using a learning rate schedule you can just
+   set the warmup_steps parameter.
+
+   This optimizer requires that .train() and .eval() be called before the
+   beginning of training and evaluation respectively. The optimizer should
+   also be placed in eval mode when saving checkpoints.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: eval()
+
+
+   .. py:method:: train()
+
+
+   .. py:method:: step(closure=None)
+
+      
+      Performs a single optimization step.
+
+      Arguments:
+          closure (callable, optional): A closure that reevaluates the model
+              and returns the loss.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/optim/adan/index.rst.txt b/_sources/autoapi/lmflow/optim/adan/index.rst.txt
new file mode 100644
index 000000000..dc5e50c09
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/adan/index.rst.txt
@@ -0,0 +1,89 @@
+lmflow.optim.adan
+=================
+
+.. py:module:: lmflow.optim.adan
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.adan.Adan
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.optim.adan._single_tensor_adan
+   lmflow.optim.adan._multi_tensor_adan
+
+
+Module Contents
+---------------
+
+.. py:class:: Adan(params, lr=0.001, betas=(0.98, 0.92, 0.99), eps=1e-08, weight_decay=0.0, max_grad_norm=0.0, no_prox=False, foreach: bool = True)
+
+   Bases: :py:obj:`torch.optim.optimizer.Optimizer`
+
+
+   
+   Implements a pytorch variant of Adan.
+
+   Adan was proposed in
+   Adan : Adaptive Nesterov Momentum Algorithm for Faster Optimizing Deep Models.
+   https://arxiv.org/abs/2208.06677
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: __setstate__(state)
+
+
+   .. py:method:: restart_opt()
+
+
+   .. py:method:: step()
+
+      
+      Performs a single optimization step.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+.. py:function:: _single_tensor_adan(params: List[torch.Tensor], grads: List[torch.Tensor], exp_avgs: List[torch.Tensor], exp_avg_sqs: List[torch.Tensor], exp_avg_diffs: List[torch.Tensor], pre_grads: List[torch.Tensor], *, beta1: float, beta2: float, beta3: float, bias_correction1: float, bias_correction2: float, bias_correction3_sqrt: float, lr: float, weight_decay: float, eps: float, no_prox: bool, clip_global_grad_norm: torch.Tensor)
+
+.. py:function:: _multi_tensor_adan(params: List[torch.Tensor], grads: List[torch.Tensor], exp_avgs: List[torch.Tensor], exp_avg_sqs: List[torch.Tensor], exp_avg_diffs: List[torch.Tensor], pre_grads: List[torch.Tensor], *, beta1: float, beta2: float, beta3: float, bias_correction1: float, bias_correction2: float, bias_correction3_sqrt: float, lr: float, weight_decay: float, eps: float, no_prox: bool, clip_global_grad_norm: torch.Tensor)
+
diff --git a/_sources/autoapi/lmflow/optim/dummy/index.rst.txt b/_sources/autoapi/lmflow/optim/dummy/index.rst.txt
new file mode 100644
index 000000000..1a016a369
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/dummy/index.rst.txt
@@ -0,0 +1,81 @@
+lmflow.optim.dummy
+==================
+
+.. py:module:: lmflow.optim.dummy
+
+.. autoapi-nested-parse::
+
+   Dummy Optimizer.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.dummy.Dummy
+
+
+Module Contents
+---------------
+
+.. py:class:: Dummy(params: Iterable[torch.nn.parameter.Parameter], lr: float = 0.0, betas: Tuple[float, float] = (0.9, 0.999), weight_decay: float = 0.0)
+
+   Bases: :py:obj:`torch.optim.Optimizer`
+
+
+   
+   An dummy optimizer that does nothing.
+
+   Parameters:
+       params (:obj:`Iterable[nn.parameter.Parameter]`):
+           Iterable of parameters to optimize or dictionaries defining parameter groups.
+       lr (:obj:`float`, `optional`, defaults to 0):
+           The learning rate to use.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: step(closure: Callable = None)
+
+      
+      Performs a single optimization step.
+
+      Arguments:
+          closure (:obj:`Callable`, `optional`): A closure that reevaluates the model and returns the loss.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/optim/index.rst.txt b/_sources/autoapi/lmflow/optim/index.rst.txt
new file mode 100644
index 000000000..f0fbcc423
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/index.rst.txt
@@ -0,0 +1,34 @@
+lmflow.optim
+============
+
+.. py:module:: lmflow.optim
+
+
+Submodules
+----------
+
+.. toctree::
+   :maxdepth: 1
+
+   /autoapi/lmflow/optim/adabelief/index
+   /autoapi/lmflow/optim/adabound/index
+   /autoapi/lmflow/optim/adadelta/index
+   /autoapi/lmflow/optim/adagrad/index
+   /autoapi/lmflow/optim/adam/index
+   /autoapi/lmflow/optim/adamax/index
+   /autoapi/lmflow/optim/adamp/index
+   /autoapi/lmflow/optim/adamw_schedule_free/index
+   /autoapi/lmflow/optim/adan/index
+   /autoapi/lmflow/optim/dummy/index
+   /autoapi/lmflow/optim/lamb/index
+   /autoapi/lmflow/optim/lars/index
+   /autoapi/lmflow/optim/nadam/index
+   /autoapi/lmflow/optim/novograd/index
+   /autoapi/lmflow/optim/optimizers/index
+   /autoapi/lmflow/optim/radam/index
+   /autoapi/lmflow/optim/sgd_schedule_free/index
+   /autoapi/lmflow/optim/sgdp/index
+   /autoapi/lmflow/optim/sophia/index
+   /autoapi/lmflow/optim/yogi/index
+
+
diff --git a/_sources/autoapi/lmflow/optim/lamb/index.rst.txt b/_sources/autoapi/lmflow/optim/lamb/index.rst.txt
new file mode 100644
index 000000000..b31a50c1f
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/lamb/index.rst.txt
@@ -0,0 +1,90 @@
+lmflow.optim.lamb
+=================
+
+.. py:module:: lmflow.optim.lamb
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.lamb.Lamb
+
+
+Module Contents
+---------------
+
+.. py:class:: Lamb(params, lr: float = 0.001, betas=(0.9, 0.999), eps: float = 1e-06, weight_decay: float = 0, clamp_value: float = 10, adam: bool = False, debias: bool = False)
+
+   Bases: :py:obj:`torch.optim.optimizer.Optimizer`
+
+
+   
+   Implements Lamb algorithm.
+
+   It has been proposed in `Large Batch Optimization for Deep Learning:
+   Training BERT in 76 minutes`
+   https://arxiv.org/abs/1904.00962
+
+   Note:
+       Reference code: https://github.com/cybertronai/pytorch-lamb
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: clamp_value
+      :value: 10
+
+
+
+   .. py:attribute:: adam
+      :value: False
+
+
+
+   .. py:attribute:: debias
+      :value: False
+
+
+
+   .. py:method:: step(closure=None)
+
+      
+      Performs a single optimization step.
+
+      Arguments:
+          closure: A closure that reevaluates the model and returns the loss.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/optim/lars/index.rst.txt b/_sources/autoapi/lmflow/optim/lars/index.rst.txt
new file mode 100644
index 000000000..e30fbf6ab
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/lars/index.rst.txt
@@ -0,0 +1,98 @@
+lmflow.optim.lars
+=================
+
+.. py:module:: lmflow.optim.lars
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.lars.LARS
+
+
+Module Contents
+---------------
+
+.. py:class:: LARS(params, lr: float = 0.01, momentum: float = 0.0, dampening: float = 0.0, weight_decay: float = 0.0, nesterov: bool = False, trust_coefficient: float = 0.01, eps: float = 1e-08)
+
+   Bases: :py:obj:`torch.optim.optimizer.Optimizer`
+
+
+   
+   Extends SGD in PyTorch with LARS scaling from the paper
+   `Large batch training of Convolutional Networks`__.
+   .. note::
+       The application of momentum in the SGD part is modified according to
+       the PyTorch standards. LARS scaling fits into the equation in the
+       following fashion.
+
+       .. math::
+           \begin{aligned}
+               g_{t+1} & = \text{lars_lr} * (\beta * p_{t} + g_{t+1}), \\
+               v_{t+1} & = \\mu * v_{t} + g_{t+1}, \\
+               p_{t+1} & = p_{t} - \text{lr} * v_{t+1},
+           \\end{aligned}
+
+       where :math:`p`, :math:`g`, :math:`v`, :math:`\\mu` and :math:`\beta`
+       denote the parameters, gradient, velocity, momentum, and weight decay
+       respectively.  The :math:`lars_lr` is defined by Eq. 6 in the paper.
+       The Nesterov version is analogously modified.
+
+   .. warning::
+       Parameters with weight decay set to 0 will automatically be excluded
+       from layer-wise LR scaling. This is to ensure consistency with papers
+       like SimCLR and BYOL.
+
+   __ https://arxiv.org/pdf/1708.03888.pdf
+
+   Note:
+       Reference code: https://github.com/PyTorchLightning/lightning-bolts/
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: __setstate__(state) -> None
+
+
+   .. py:method:: step(closure=None)
+
+      
+      Performs a single optimization step.
+
+      Arguments:
+          closure: A closure that reevaluates the model and returns the loss.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/optim/nadam/index.rst.txt b/_sources/autoapi/lmflow/optim/nadam/index.rst.txt
new file mode 100644
index 000000000..fe3d1c3d5
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/nadam/index.rst.txt
@@ -0,0 +1,28 @@
+lmflow.optim.nadam
+==================
+
+.. py:module:: lmflow.optim.nadam
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.nadam.NAdam
+
+
+Module Contents
+---------------
+
+.. py:class:: NAdam(params, lr=0.002, betas=(0.9, 0.999), eps=1e-08, weight_decay=0, momentum_decay=0.004)
+
+   Bases: :py:obj:`torch.optim.Optimizer`
+
+
+   .. py:method:: __setstate__(state)
+
+
+   .. py:method:: step(closure=None)
+
+
diff --git a/_sources/autoapi/lmflow/optim/novograd/index.rst.txt b/_sources/autoapi/lmflow/optim/novograd/index.rst.txt
new file mode 100644
index 000000000..2cd3e4ff7
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/novograd/index.rst.txt
@@ -0,0 +1,28 @@
+lmflow.optim.novograd
+=====================
+
+.. py:module:: lmflow.optim.novograd
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.novograd.NovoGrad
+
+
+Module Contents
+---------------
+
+.. py:class:: NovoGrad(params, lr=0.01, betas=(0.9, 0.999), eps=1e-08, weight_decay=0, grad_averaging=False, amsgrad=False)
+
+   Bases: :py:obj:`torch.optim.Optimizer`
+
+
+   .. py:method:: __setstate__(state)
+
+
+   .. py:method:: step(closure=None)
+
+
diff --git a/_sources/autoapi/lmflow/optim/optimizers/index.rst.txt b/_sources/autoapi/lmflow/optim/optimizers/index.rst.txt
new file mode 100644
index 000000000..44c92a3d7
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/optimizers/index.rst.txt
@@ -0,0 +1,13 @@
+lmflow.optim.optimizers
+=======================
+
+.. py:module:: lmflow.optim.optimizers
+
+.. autoapi-nested-parse::
+
+   All optimizers.
+
+   ..
+       !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/optim/radam/index.rst.txt b/_sources/autoapi/lmflow/optim/radam/index.rst.txt
new file mode 100644
index 000000000..69b5a98a6
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/radam/index.rst.txt
@@ -0,0 +1,81 @@
+lmflow.optim.radam
+==================
+
+.. py:module:: lmflow.optim.radam
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.radam.RAdam
+
+
+Module Contents
+---------------
+
+.. py:class:: RAdam(params, lr: float = 0.001, betas=(0.9, 0.999), eps: float = 1e-08, weight_decay: float = 0)
+
+   Bases: :py:obj:`torch.optim.optimizer.Optimizer`
+
+
+   
+   Implements RAdam optimization algorithm.
+
+   Note:
+       Deprecated, please use version provided by PyTorch_.
+
+   It has been proposed in `On the Variance of the Adaptive Learning
+   Rate and Beyond`.
+   https://arxiv.org/abs/1908.03265
+
+   Note:
+       Reference code: https://github.com/LiyuanLucasLiu/RAdam
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: __setstate__(state)
+
+
+   .. py:method:: step(closure=None)
+
+      
+      Performs a single optimization step.
+
+      Arguments:
+          closure: A closure that reevaluates the model and returns the loss.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/optim/sgd_schedule_free/index.rst.txt b/_sources/autoapi/lmflow/optim/sgd_schedule_free/index.rst.txt
new file mode 100644
index 000000000..ac6a70e88
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/sgd_schedule_free/index.rst.txt
@@ -0,0 +1,82 @@
+lmflow.optim.sgd_schedule_free
+==============================
+
+.. py:module:: lmflow.optim.sgd_schedule_free
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.sgd_schedule_free.SGDScheduleFree
+
+
+Module Contents
+---------------
+
+.. py:class:: SGDScheduleFree(params, lr=1.0, momentum=0.9, weight_decay=0, warmup_steps=0, r=0.0, weight_lr_power=2, foreach=hasattr(torch, '_foreach_mul_'))
+
+   Bases: :py:obj:`torch.optim.Optimizer`
+
+
+   
+   Schedule-Free SGD
+   As the name suggests, no scheduler is needed with this optimizer. 
+   To add warmup, rather than using a learning rate schedule you can just
+   set the warmup_steps parameter.
+
+   This optimizer requires that .train() and .eval() be called before the
+   beginning of training and evaluation respectively. The optimizer should
+   also be placed in eval mode when saving checkpoints.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: eval()
+
+
+   .. py:method:: train()
+
+
+   .. py:method:: step(closure=None)
+
+      
+      Performs a single optimization step.
+
+      Arguments:
+          closure (callable, optional): A closure that reevaluates the model
+              and returns the loss.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/optim/sgdp/index.rst.txt b/_sources/autoapi/lmflow/optim/sgdp/index.rst.txt
new file mode 100644
index 000000000..5d695ece7
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/sgdp/index.rst.txt
@@ -0,0 +1,93 @@
+lmflow.optim.sgdp
+=================
+
+.. py:module:: lmflow.optim.sgdp
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.sgdp.SGDP
+
+
+Module Contents
+---------------
+
+.. py:class:: SGDP(params, lr: float = 0.001, momentum: float = 0, dampening: float = 0, eps: float = 1e-08, weight_decay: float = 0, delta: float = 0.1, wd_ratio: float = 0.1, nesterov: bool = False)
+
+   Bases: :py:obj:`torch.optim.optimizer.Optimizer`
+
+
+   
+   Implements SGDP algorithm.
+
+   It has been proposed in `Slowing Down the Weight Norm Increase in
+   Momentum-based Optimizers`.
+   https://arxiv.org/abs/2006.08217
+
+   Note:
+       Reference code: https://github.com/clovaai/AdamP
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: _channel_view(x)
+      :staticmethod:
+
+
+
+   .. py:method:: _layer_view(x)
+      :staticmethod:
+
+
+
+   .. py:method:: _cosine_similarity(x, y, eps, view_func)
+      :staticmethod:
+
+
+
+   .. py:method:: _projection(p, grad, perturb, delta, wd_ratio, eps)
+
+
+   .. py:method:: step(closure=None)
+
+      
+      Performs a single optimization step.
+
+      Arguments:
+          closure: A closure that reevaluates the model and returns the loss.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/optim/sophia/index.rst.txt b/_sources/autoapi/lmflow/optim/sophia/index.rst.txt
new file mode 100644
index 000000000..57144a0ef
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/sophia/index.rst.txt
@@ -0,0 +1,53 @@
+lmflow.optim.sophia
+===================
+
+.. py:module:: lmflow.optim.sophia
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.sophia.SophiaG
+
+
+Module Contents
+---------------
+
+.. py:class:: SophiaG(params, lr=0.0001, betas=(0.965, 0.99), rho=0.04, weight_decay=0.1, *, maximize: bool = False, capturable: bool = False)
+
+   Bases: :py:obj:`torch.optim.optimizer.Optimizer`
+
+
+   
+   Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training.
+   Code from: https://github.com/Liuhong99/Sophia/
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: __setstate__(state)
+
+
+   .. py:method:: update_hessian()
+
+
+   .. py:method:: step(closure=None, bs=5120)
+
+
diff --git a/_sources/autoapi/lmflow/optim/yogi/index.rst.txt b/_sources/autoapi/lmflow/optim/yogi/index.rst.txt
new file mode 100644
index 000000000..0189c4218
--- /dev/null
+++ b/_sources/autoapi/lmflow/optim/yogi/index.rst.txt
@@ -0,0 +1,74 @@
+lmflow.optim.yogi
+=================
+
+.. py:module:: lmflow.optim.yogi
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.optim.yogi.Yogi
+
+
+Module Contents
+---------------
+
+.. py:class:: Yogi(params, lr: float = 0.01, betas=(0.9, 0.999), eps: float = 0.001, initial_accumulator: float = 1e-06, weight_decay: float = 0)
+
+   Bases: :py:obj:`torch.optim.optimizer.Optimizer`
+
+
+   
+   Implements Yogi Optimizer Algorithm.
+   It has been proposed in `Adaptive methods for Nonconvex Optimization`.
+
+   https://papers.nips.cc/paper/8186-adaptive-methods-for-nonconvex-optimization  # noqa
+
+   Note:
+       Reference code: https://github.com/4rtemi5/Yogi-Optimizer_Keras
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: step(closure=None)
+
+      
+      Performs a single optimization step.
+
+      Arguments:
+          closure: A closure that reevaluates the model and returns the loss.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/auto_pipeline/index.rst.txt b/_sources/autoapi/lmflow/pipeline/auto_pipeline/index.rst.txt
new file mode 100644
index 000000000..c0194c84d
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/auto_pipeline/index.rst.txt
@@ -0,0 +1,67 @@
+lmflow.pipeline.auto_pipeline
+=============================
+
+.. py:module:: lmflow.pipeline.auto_pipeline
+
+.. autoapi-nested-parse::
+
+   Return a pipeline automatically based on its name.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.auto_pipeline.PIPELINE_MAPPING
+   lmflow.pipeline.auto_pipeline.PIPELINE_NEEDS_EXTRAS
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.auto_pipeline.AutoPipeline
+
+
+Module Contents
+---------------
+
+.. py:data:: PIPELINE_MAPPING
+
+.. py:data:: PIPELINE_NEEDS_EXTRAS
+   :value: []
+
+
+.. py:class:: AutoPipeline
+
+   
+   The class designed to return a pipeline automatically based on its name.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: get_pipeline(pipeline_name, model_args, data_args, pipeline_args, *args, **kwargs)
+      :classmethod:
+
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/base_aligner/index.rst.txt b/_sources/autoapi/lmflow/pipeline/base_aligner/index.rst.txt
new file mode 100644
index 000000000..23043558c
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/base_aligner/index.rst.txt
@@ -0,0 +1,58 @@
+lmflow.pipeline.base_aligner
+============================
+
+.. py:module:: lmflow.pipeline.base_aligner
+
+.. autoapi-nested-parse::
+
+   BaseTuner: a subclass of BasePipeline.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.base_aligner.BaseAligner
+
+
+Module Contents
+---------------
+
+.. py:class:: BaseAligner(*args, **kwargs)
+
+   Bases: :py:obj:`lmflow.pipeline.base_pipeline.BasePipeline`
+
+
+   
+   A subclass of BasePipeline which is alignable.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: _check_if_alignable(model, dataset, reward_model)
+
+
+   .. py:method:: align(model, dataset, reward_model)
+      :abstractmethod:
+
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/base_pipeline/index.rst.txt b/_sources/autoapi/lmflow/pipeline/base_pipeline/index.rst.txt
new file mode 100644
index 000000000..0af8835df
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/base_pipeline/index.rst.txt
@@ -0,0 +1,51 @@
+lmflow.pipeline.base_pipeline
+=============================
+
+.. py:module:: lmflow.pipeline.base_pipeline
+
+.. autoapi-nested-parse::
+
+   BasePipeline.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.base_pipeline.BasePipeline
+
+
+Module Contents
+---------------
+
+.. py:class:: BasePipeline
+
+   Bases: :py:obj:`abc.ABC`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
diff --git a/_sources/autoapi/lmflow/pipeline/base_tuner/index.rst.txt b/_sources/autoapi/lmflow/pipeline/base_tuner/index.rst.txt
new file mode 100644
index 000000000..a2953952d
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/base_tuner/index.rst.txt
@@ -0,0 +1,58 @@
+lmflow.pipeline.base_tuner
+==========================
+
+.. py:module:: lmflow.pipeline.base_tuner
+
+.. autoapi-nested-parse::
+
+   BaseTuner: a subclass of BasePipeline.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.base_tuner.BaseTuner
+
+
+Module Contents
+---------------
+
+.. py:class:: BaseTuner(*args, **kwargs)
+
+   Bases: :py:obj:`lmflow.pipeline.base_pipeline.BasePipeline`
+
+
+   
+   A subclass of BasePipeline which is tunable.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: _check_if_tunable(model, dataset)
+
+
+   .. py:method:: tune(model, dataset)
+      :abstractmethod:
+
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/dpo_aligner/index.rst.txt b/_sources/autoapi/lmflow/pipeline/dpo_aligner/index.rst.txt
new file mode 100644
index 000000000..95c9a6f13
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/dpo_aligner/index.rst.txt
@@ -0,0 +1,92 @@
+lmflow.pipeline.dpo_aligner
+===========================
+
+.. py:module:: lmflow.pipeline.dpo_aligner
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.dpo_aligner.DPOAligner
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.pipeline.dpo_aligner.get_paired_dataset
+
+
+Module Contents
+---------------
+
+.. py:function:: get_paired_dataset(data_root: str, data_dir: str, sanity_check: bool = False, cache_dir: Optional[str] = None, num_proc=24) -> datasets.Dataset
+
+   
+   Load dataset and convert it to the necessary format.
+
+       The dataset is converted to a dictionary with the following structure:
+       {
+           'prompt': List[str],
+           'chosen': List[str],
+           'rejected': List[str],
+       }
+
+       Prompts are structured as follows:
+         "Question: " + <prompt> + "
+
+   Answer: "
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:class:: DPOAligner(model_args, data_args, aligner_args)
+
+   Bases: :py:obj:`lmflow.pipeline.base_aligner.BaseAligner`
+
+
+   .. py:attribute:: model_args
+
+
+   .. py:attribute:: data_args
+
+
+   .. py:attribute:: aligner_args
+
+
+   .. py:attribute:: train_dataset
+      :value: None
+
+
+
+   .. py:attribute:: eval_dataset
+      :value: None
+
+
+
+   .. py:method:: _initialize_trainer(model, tokenizer)
+
+
+   .. py:method:: _load_dataset()
+
+
+   .. py:method:: align(model, dataset, reward_model)
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/dpov2_aligner/index.rst.txt b/_sources/autoapi/lmflow/pipeline/dpov2_aligner/index.rst.txt
new file mode 100644
index 000000000..b6cd418cc
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/dpov2_aligner/index.rst.txt
@@ -0,0 +1,161 @@
+lmflow.pipeline.dpov2_aligner
+=============================
+
+.. py:module:: lmflow.pipeline.dpov2_aligner
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.dpov2_aligner.logger
+   lmflow.pipeline.dpov2_aligner.ReferenceModelArguments
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.dpov2_aligner.DPOv2Aligner
+   lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:data:: ReferenceModelArguments
+
+.. py:class:: DPOv2Aligner(model_args: lmflow.args.ModelArguments, data_args: lmflow.args.DatasetArguments, aligner_args: lmflow.args.DPOv2AlignerArguments, ref_model_args: lmflow.args.ModelArguments)
+
+   Bases: :py:obj:`lmflow.pipeline.base_aligner.BaseAligner`
+
+
+   .. py:attribute:: model_args
+
+
+   .. py:attribute:: ref_model_args
+
+
+   .. py:attribute:: data_args
+
+
+   .. py:attribute:: aligner_args
+
+
+   .. py:method:: align(model: lmflow.models.hf_decoder_model.HFDecoderModel, ref_model: lmflow.models.hf_decoder_model.HFDecoderModel, train_dataset: lmflow.datasets.dataset.Dataset, eval_dataset: lmflow.datasets.dataset.Dataset, transform_dataset_in_place: bool = True)
+
+
+   .. py:method:: __prepare_training_args(args: lmflow.args.DPOv2AlignerArguments) -> transformers.TrainingArguments
+
+
+   .. py:method:: convert_to_paired_dataset(source_dataset: lmflow.datasets.dataset.Dataset, sampling_paired_method: str = 'random', length_penalty: float = 0.0, margin_scale: float = 1.0, use_fast: bool = False) -> lmflow.datasets.dataset.Dataset
+
+      
+      Convert a scored one to multiple (text_to_scored_textlist) to a paired dataset by rejection sampling.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _calc_response_lengths(outputs: List[Union[str, Dict[str, str]]], dataset_type: str) -> List[int]
+
+
+   .. py:method:: _calc_reward_with_length_penalty(rewards: List[float], lengths: List[int], length_penalty: float) -> List[float]
+
+      
+      When length_penalty > 0, penalize the longer sequence by subtracting 
+      length_penalty * length from the reward. Vice versa when length_penalty < 0.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: sampling_paired_idx_from_rewards(rewards: List[float], sampling_paired_method: str = 'random', use_fast: bool = False) -> Tuple[int, int]
+
+      
+      Prepare the dataset for DPO training by rejection sampling.
+      We implement different strategies to select pairs, including
+      random: randomly select two instances
+      max_min: best v.s. worst
+      max_max: best v.s. second best
+      max_random: best v.s. random from the remaining
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _sampling_paired_idx_from_rewards(rewards: List[float], sampling_paired_method: str = 'random') -> Tuple[int, int]
+
+
+   .. py:method:: _sampling_paired_idx_from_rewards_fast(rewards: List[float], sampling_paired_method: str = 'random') -> Tuple[int, int]
+
+
+.. py:class:: MemorySafeDPOv2Aligner(model_args: lmflow.args.ModelArguments, data_args: lmflow.args.DatasetArguments, aligner_args: lmflow.args.DPOv2AlignerArguments, ref_model_args: lmflow.args.ModelArguments)
+
+   .. py:attribute:: model_args
+
+
+   .. py:attribute:: ref_model_args
+
+
+   .. py:attribute:: data_args
+
+
+   .. py:attribute:: aligner_args
+
+
+   .. py:attribute:: aligner_file_path
+
+
+   .. py:method:: align()
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/evaluator/index.rst.txt b/_sources/autoapi/lmflow/pipeline/evaluator/index.rst.txt
new file mode 100644
index 000000000..d110bda1e
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/evaluator/index.rst.txt
@@ -0,0 +1,162 @@
+lmflow.pipeline.evaluator
+=========================
+
+.. py:module:: lmflow.pipeline.evaluator
+
+.. autoapi-nested-parse::
+
+   The Evaluator class simplifies the process of running evaluation on a language model provided by a HFDecoderModel instance imported from the lmflow package. The class constructor takes three dictionaries as arguments: model_args containing arguments related to the language model, data_args containing arguments related to the data used for evaluation, and evaluator_args containing other arguments for the evaluation process.
+
+   The class has two methods: create_dataloader() that loads the data from the test file, creates a data loader, and returns it with the size of the data, and evaluate(model) that generates output text given input text. It uses the create_dataloader() method to load the data, iterates over the data in mini-batches, and encodes the input text with the encode() method of the HFDecoderModel class. Then, it generates output text using the evaluate() method of the HFDecoderModel class, decodes the generated output text using the decode() method of the HFDecoderModel class, and writes the output to a file in the output directory. The method also logs some information to the console and Weights and Biases if the use_wandb argument is True.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.evaluator.Evaluator
+
+
+Module Contents
+---------------
+
+.. py:class:: Evaluator(model_args, data_args, evaluator_args)
+
+   Bases: :py:obj:`lmflow.pipeline.base_pipeline.BasePipeline`
+
+
+   
+   Initializes the `Evaluator` class with given arguments.
+
+
+   :Parameters:
+
+       **model_args** : ModelArguments object.
+           Contains the arguments required to load the model.
+
+       **data_args** : DatasetArguments object.
+           Contains the arguments required to load the dataset.
+
+       **evaluator_args** : EvaluatorArguments object.
+           Contains the arguments required to perform evaluation.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: data_args
+
+
+   .. py:attribute:: evaluator_args
+
+
+   .. py:attribute:: model_args
+
+
+   .. py:attribute:: local_rank
+
+
+   .. py:attribute:: world_size
+
+
+   .. py:attribute:: config
+
+
+   .. py:attribute:: block_size
+
+
+   .. py:method:: create_dataloader(dataset: lmflow.datasets.dataset.Dataset)
+
+
+   .. py:method:: _match(predicted_answer, groundtruth, answer_type=None)
+
+
+   .. py:method:: evaluate(model, dataset: lmflow.datasets.dataset.Dataset, metric='accuracy', verbose=True)
+
+      
+      Perform Evaluation for a model
+
+
+      :Parameters:
+
+          **model** : TunableModel object.
+              TunableModel to perform inference
+
+          **dataset** : Dataset object.
+              ..
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _evaluate_acc_with_accelerator(model, dataset, verbose=True)
+
+
+   .. py:method:: _evaluate_acc_with_deepspeed(model, dataset, verbose=True)
+
+
+   .. py:method:: _evaluate_ppl(model, dataset: lmflow.datasets.dataset.Dataset, verbose=True)
+
+
+   .. py:method:: _evaluate_nll(model, dataset: lmflow.datasets.dataset.Dataset, verbose=True)
+
+      
+      Evaluates negative log likelihood of the model over a dataset.
+
+      NLL = -1/N sum_{i=1}^N sum_{j=1}^|w_i| ln(p(w_{i,j}|context_window)),
+
+      where N is the number of data samples, w_{i,j} is the j-th token in
+      i-th sample. Here "context_window" = p(w_{i,start}, w_{i,start+1}, ...,
+      p_{i,j-1} with start = max(0, j - window_length + 1). "window_length"
+      is normally the maximum length accepted by the model.
+
+      Returns:
+          A float which represents the negative log likelihood.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/finetuner/index.rst.txt b/_sources/autoapi/lmflow/pipeline/finetuner/index.rst.txt
new file mode 100644
index 000000000..6f560af84
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/finetuner/index.rst.txt
@@ -0,0 +1,149 @@
+lmflow.pipeline.finetuner
+=========================
+
+.. py:module:: lmflow.pipeline.finetuner
+
+.. autoapi-nested-parse::
+
+   The Finetuner class simplifies the process of running finetuning process on a language model for a TunableModel instance with given dataset.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.finetuner.logger
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.finetuner.Finetuner
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: Finetuner(model_args, data_args, finetuner_args, *args, **kwargs)
+
+   Bases: :py:obj:`lmflow.pipeline.base_tuner.BaseTuner`
+
+
+   
+   Initializes the `Finetuner` class with given arguments.
+
+
+   :Parameters:
+
+       **model_args** : ModelArguments object.
+           Contains the arguments required to load the model.
+
+       **data_args** : DatasetArguments object.
+           Contains the arguments required to load the dataset.
+
+       **finetuner_args** : FinetunerArguments object.
+           Contains the arguments required to perform finetuning.
+
+       **args** : Optional.
+           Positional arguments.
+
+       **kwargs** : Optional.
+           Keyword arguments.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: model_args
+
+
+   .. py:attribute:: data_args
+
+
+   .. py:attribute:: finetuner_args
+
+
+   .. py:attribute:: last_checkpoint
+      :value: None
+
+
+
+   .. py:method:: group_text(tokenized_datasets, model_max_length)
+
+      
+      Groups texts together to form blocks of maximum length `model_max_length` and returns the processed data as
+      a dictionary.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: create_customized_optimizer(base_trainer_class, model_args)
+
+
+   .. py:method:: tune(model, dataset, transform_dataset_in_place=True, data_collator=None)
+
+      
+      Perform tuning for a model
+
+
+      :Parameters:
+
+          **model** : TunableModel object.
+              TunableModel to perform tuning.
+
+          **dataset:**
+              dataset to train model.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/index.rst.txt b/_sources/autoapi/lmflow/pipeline/index.rst.txt
new file mode 100644
index 000000000..78ec28381
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/index.rst.txt
@@ -0,0 +1,29 @@
+lmflow.pipeline
+===============
+
+.. py:module:: lmflow.pipeline
+
+
+Submodules
+----------
+
+.. toctree::
+   :maxdepth: 1
+
+   /autoapi/lmflow/pipeline/auto_pipeline/index
+   /autoapi/lmflow/pipeline/base_aligner/index
+   /autoapi/lmflow/pipeline/base_pipeline/index
+   /autoapi/lmflow/pipeline/base_tuner/index
+   /autoapi/lmflow/pipeline/dpo_aligner/index
+   /autoapi/lmflow/pipeline/dpov2_aligner/index
+   /autoapi/lmflow/pipeline/evaluator/index
+   /autoapi/lmflow/pipeline/finetuner/index
+   /autoapi/lmflow/pipeline/inferencer/index
+   /autoapi/lmflow/pipeline/iterative_dpo_aligner/index
+   /autoapi/lmflow/pipeline/raft_aligner/index
+   /autoapi/lmflow/pipeline/rm_inferencer/index
+   /autoapi/lmflow/pipeline/rm_tuner/index
+   /autoapi/lmflow/pipeline/utils/index
+   /autoapi/lmflow/pipeline/vllm_inferencer/index
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/inferencer/index.rst.txt b/_sources/autoapi/lmflow/pipeline/inferencer/index.rst.txt
new file mode 100644
index 000000000..62a1c333a
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/inferencer/index.rst.txt
@@ -0,0 +1,468 @@
+lmflow.pipeline.inferencer
+==========================
+
+.. py:module:: lmflow.pipeline.inferencer
+
+.. autoapi-nested-parse::
+
+   The Inferencer class simplifies the process of model inferencing.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.inferencer.supported_dataset_type
+   lmflow.pipeline.inferencer.logger
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.inferencer.Inferencer
+   lmflow.pipeline.inferencer.SpeculativeInferencer
+   lmflow.pipeline.inferencer.ToolInferencer
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.pipeline.inferencer.rstrip_partial_utf8
+
+
+Module Contents
+---------------
+
+.. py:function:: rstrip_partial_utf8(string)
+
+.. py:data:: supported_dataset_type
+   :value: ['text_only', 'image_text']
+
+
+.. py:data:: logger
+
+.. py:class:: Inferencer(model_args, data_args, inferencer_args)
+
+   Bases: :py:obj:`lmflow.pipeline.base_pipeline.BasePipeline`
+
+
+   
+   Initializes the `Inferencer` class with given arguments.
+
+
+   :Parameters:
+
+       **model_args** : ModelArguments object.
+           Contains the arguments required to load the model.
+
+       **data_args** : DatasetArguments object.
+           Contains the arguments required to load the dataset.
+
+       **inferencer_args** : InferencerArguments object.
+           Contains the arguments required to perform inference.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: data_args
+
+
+   .. py:attribute:: inferencer_args
+
+
+   .. py:attribute:: model_args
+
+
+   .. py:attribute:: local_rank
+
+
+   .. py:attribute:: world_size
+
+
+   .. py:attribute:: config
+
+
+   .. py:method:: create_dataloader(dataset: lmflow.datasets.dataset.Dataset)
+
+      
+      Batchlize dataset and format it to dataloader.
+
+      Args:
+          dataset (Dataset): the dataset object
+
+      Output:
+          dataloader (batchlize): the dataloader object
+          dataset_size (int): the length of the dataset
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: inference(model, dataset: lmflow.datasets.dataset.Dataset, max_new_tokens: int = 100, temperature: float = 0.0, prompt_structure: str = '{input}', remove_image_flag: bool = False, chatbot_type: str = 'mini_gpt')
+
+      
+      Perform inference for a model
+
+
+      :Parameters:
+
+          **model** : TunableModel object.
+              TunableModel to perform inference
+
+          **dataset** : Dataset object.
+              ..
+
+          **Returns:**
+              ..
+
+          **output_dataset: Dataset object.**
+              ..
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: stream_inference(context, model, max_new_tokens, token_per_step, temperature, end_string, input_dataset, remove_image_flag: bool = False)
+
+
+.. py:class:: SpeculativeInferencer(model_args, draft_model_args, data_args, inferencer_args)
+
+   Bases: :py:obj:`Inferencer`
+
+
+   
+   Ref: [arXiv:2211.17192v2](https://arxiv.org/abs/2211.17192)
+
+
+   :Parameters:
+
+       **target_model_args** : ModelArguments object.
+           Contains the arguments required to load the target model.
+
+       **draft_model_args** : ModelArguments object.
+           Contains the arguments required to load the draft model.
+
+       **data_args** : DatasetArguments object.
+           Contains the arguments required to load the dataset.
+
+       **inferencer_args** : InferencerArguments object.
+           Contains the arguments required to perform inference.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: draft_model_args
+
+
+   .. py:attribute:: draft_config
+
+
+   .. py:method:: score_to_prob(scores: torch.Tensor, temperature: float = 0.0, top_p: float = 1.0) -> torch.Tensor
+      :staticmethod:
+
+
+      
+      Convert scores (NOT softmaxed tensor) to probabilities with support for temperature, top-p sampling, and argmax.
+
+
+      :Parameters:
+
+          **scores** : torch.Tensor
+              Input scores.
+
+          **temperature** : float, optional
+              Temperature parameter for controlling randomness. Higher values make the distribution more uniform, 
+              lower values make it peakier. When temperature <= 1e-6, argmax is used. by default 0.0
+
+          **top_p** : float, optional
+              Top-p sampling parameter for controlling the cumulative probability threshold, by default 1.0 (no threshold)
+
+
+
+      :Returns:
+
+          torch.Tensor
+              Probability distribution after adjustments.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: sample(prob: torch.Tensor, num_samples: int = 1) -> Dict
+      :staticmethod:
+
+
+      
+      Sample from a tensor of probabilities
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: predict_next_token(model: lmflow.models.hf_decoder_model.HFDecoderModel, input_ids: torch.Tensor, num_new_tokens: int = 1)
+      :staticmethod:
+
+
+      
+      Predict the next token given the input_ids.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: autoregressive_sampling(input_ids: torch.Tensor, model: lmflow.models.hf_decoder_model.HFDecoderModel, temperature: float = 0.0, num_new_tokens: int = 5) -> Dict
+
+      
+      Ref: [arXiv:2211.17192v2](https://arxiv.org/abs/2211.17192) Section 2.2
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: inference(model: lmflow.models.hf_decoder_model.HFDecoderModel, draft_model: lmflow.models.hf_decoder_model.HFDecoderModel, input: str, temperature: float = 0.0, gamma: int = 5, max_new_tokens: int = 100)
+
+      
+      Perform inference for a model
+
+
+      :Parameters:
+
+          **model** : HFDecoderModel object.
+              TunableModel to verify tokens generated by the draft model.
+
+          **draft_model** : HFDecoderModel object.
+              TunableModel that provides approximations of the target model.
+
+          **input** : str.
+              The input text (i.e., the prompt) for the model.
+
+          **gamma** : int.
+              The number of tokens to be generated by the draft model within each iter.
+
+          **max_new_tokens** : int.
+              The maximum number of tokens to be generated by the target model.
+
+
+
+      :Returns:
+
+          output: str.
+              The output text generated by the model.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: stream_inference()
+      :abstractmethod:
+
+
+
+.. py:class:: ToolInferencer(model_args, data_args, inferencer_args)
+
+   Bases: :py:obj:`Inferencer`
+
+
+   
+   Initializes the `ToolInferencer` class with given arguments.
+
+
+   :Parameters:
+
+       **model_args** : ModelArguments object.
+           Contains the arguments required to load the model.
+
+       **data_args** : DatasetArguments object.
+           Contains the arguments required to load the dataset.
+
+       **inferencer_args** : InferencerArguments object.
+           Contains the arguments required to perform inference.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: model
+
+
+   .. py:method:: inference(model: lmflow.models.hf_decoder_model.HFDecoderModel, input: str, max_new_tokens: int = 1024)
+
+      
+      Perform inference for a model
+
+
+      :Parameters:
+
+          **model** : HFDecoderModel object.
+              TunableModel to perform inference
+
+          **input** : str.
+              The input text (i.e., the prompt) for the model. 
+
+          **max_new_tokens** : int.
+              The maximum number of tokens to be generated by the model.
+
+          **Returns:**
+              ..
+
+          **output** : str.
+              The output text generated by the model.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: code_exec(code)
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/iterative_dpo_aligner/index.rst.txt b/_sources/autoapi/lmflow/pipeline/iterative_dpo_aligner/index.rst.txt
new file mode 100644
index 000000000..78b0a5373
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/iterative_dpo_aligner/index.rst.txt
@@ -0,0 +1,74 @@
+lmflow.pipeline.iterative_dpo_aligner
+=====================================
+
+.. py:module:: lmflow.pipeline.iterative_dpo_aligner
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.iterative_dpo_aligner.logger
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: IterativeDPOAligner(model_args: lmflow.args.ModelArguments, data_args: lmflow.args.DatasetArguments, aligner_args: lmflow.args.IterativeDPOAlignerArguments, ref_model_args: lmflow.args.ModelArguments, reward_model_args: lmflow.args.ModelArguments, **kwargs)
+
+   .. py:attribute:: model_args
+
+
+   .. py:attribute:: data_args
+
+
+   .. py:attribute:: aligner_args
+
+
+   .. py:attribute:: ref_model_args
+
+
+   .. py:attribute:: reward_model_args
+
+
+   .. py:attribute:: workspace_path
+
+
+   .. py:method:: align(dataset_list: List[lmflow.datasets.dataset.Dataset])
+
+
+   .. py:method:: _align_single_iteration(iteration_name: str, target_model_args: lmflow.args.ModelArguments, reward_model_args: lmflow.args.ModelArguments, ref_model_args: lmflow.args.ModelArguments, dataset: lmflow.datasets.dataset.Dataset)
+
+
+   .. py:method:: _do_target_model_inference(model: lmflow.models.hf_decoder_model.HFDecoderModel, dataset: lmflow.datasets.dataset.Dataset, output_dir: str)
+
+
+   .. py:method:: _do_reward_model_inference(model: lmflow.models.hf_text_regression_model.HFTextRegressionModel, dataset: lmflow.datasets.dataset.Dataset, output_dir: str)
+
+
+   .. py:method:: _do_single_dpo_align(model_args: lmflow.args.ModelArguments, ref_model_args: lmflow.args.ModelArguments, data_args: lmflow.args.DatasetArguments, output_dir: str, iteration_name: str)
+
+
+   .. py:method:: _parse_target_model_inference_args(args: lmflow.args.IterativeDPOAlignerArguments, result_cache_path: str) -> lmflow.args.InferencerArguments
+
+
+   .. py:method:: _parse_reward_model_inference_args(args: lmflow.args.IterativeDPOAlignerArguments) -> lmflow.args.InferencerArguments
+
+
+   .. py:method:: _parse_dpo_aligner_args(args: lmflow.args.IterativeDPOAlignerArguments, output_dir: str, iteration_name: str) -> lmflow.args.DPOv2AlignerArguments
+
+
+   .. py:method:: __filter_args(mixed_args, target_cls)
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/raft_aligner/index.rst.txt b/_sources/autoapi/lmflow/pipeline/raft_aligner/index.rst.txt
new file mode 100644
index 000000000..af6724393
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/raft_aligner/index.rst.txt
@@ -0,0 +1,259 @@
+lmflow.pipeline.raft_aligner
+============================
+
+.. py:module:: lmflow.pipeline.raft_aligner
+
+.. autoapi-nested-parse::
+
+   The Aligner class simplifies the process of running alignment.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.raft_aligner.logger
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.raft_aligner.RaftAligner
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: RaftAligner(model_args, data_args, aligner_args, *args, **kwargs)
+
+   Bases: :py:obj:`lmflow.pipeline.base_aligner.BaseAligner`
+
+
+   
+   Initializes the `RaftAligner` class with given arguments.
+
+
+   :Parameters:
+
+       **model_args** : ModelArguments object.
+           Contains the arguments required to load the model.
+
+       **data_args** : DatasetArguments object.
+           Contains the arguments required to load the dataset.
+
+       **raft_aligner_args** : RaftAlignerArguments object.
+           Contains the arguments required to perform alignment.
+
+       **args** : Optional.
+           Positional arguments.
+
+       **kwargs** : Optional.
+           Keyword arguments.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: model_args
+
+
+   .. py:attribute:: data_args
+
+
+   .. py:attribute:: aligner_args
+
+
+   .. py:attribute:: INF
+      :value: 888888888
+
+
+
+   .. py:method:: _initialize_trainer(model, tokenizer, training_args)
+
+      
+      This function takes the model and tokenizer as the input and initialize the trainer.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _load_dataset(selected_dataset, model, tokenizer, model_args, data_args, training_args)
+
+      
+      This function prepares the dataset for every iteration.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _load_input_dataset(dataset, tokenizer)
+
+      
+      Load input dataset (i.e. prompt/question dataset) for training.
+
+      Args:
+          dataset: A Dataset object.
+              The dataset to be loaded.
+
+      Returns:
+          dataloader (`torch.utils.data.DataLoader`):
+              The dataloader for the dataset.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _clean_text(text)
+
+
+   .. py:method:: _discard_sample(text)
+
+
+   .. py:method:: _get_batch_dataset_top(model, batch_input, alpha=0.2, iter_id=0, local_rank=0, output_min_length=16, output_max_length=48, infer_batch_size=8, generation_kwargs={}, tokenizer=None, training_args=None, reward_model=None, output_reward_path=None)
+
+      
+      :param batch_input: input prompts
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _get_batch_dataset_local(model, batch_input, K=8, iter_id=0, local_rank=0, output_min_length=16, output_max_length=48, infer_batch_size=8, generation_kwargs={}, tokenizer=None, training_args=None, reward_model=None, output_reward_path=None)
+
+      
+      :param batch_input: input prompts
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: align(model, dataset, reward_model)
+
+      
+      Perform alignment for a model
+
+
+      :Parameters:
+
+          **model** : BaseModel object.
+              ..
+
+          **dataset: Dataset object.**
+              Input dataset for model to generate outputs. The input and output
+                  will then be feed into reward model to get the reward for
+                  alignment.
+
+          **reward_model: RegressionModel object.**
+              ..
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/rm_inferencer/index.rst.txt b/_sources/autoapi/lmflow/pipeline/rm_inferencer/index.rst.txt
new file mode 100644
index 000000000..7633e1970
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/rm_inferencer/index.rst.txt
@@ -0,0 +1,104 @@
+lmflow.pipeline.rm_inferencer
+=============================
+
+.. py:module:: lmflow.pipeline.rm_inferencer
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.rm_inferencer.logger
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.rm_inferencer.RewardModelInferencer
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: RewardModelInferencer(model_args: lmflow.args.ModelArguments, data_args: lmflow.args.DatasetArguments, inferencer_args: lmflow.args.InferencerArguments, **kwargs)
+
+   Bases: :py:obj:`lmflow.pipeline.base_pipeline.BasePipeline`
+
+
+   
+   Initializes the `Inferencer` class with given arguments.
+
+
+   :Parameters:
+
+       **model_args** : ModelArguments object.
+           Contains the arguments required to load the model.
+
+       **data_args** : DatasetArguments object.
+           Contains the arguments required to load the dataset.
+
+       **inferencer_args** : InferencerArguments object.
+           Contains the arguments required to perform inference.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: data_args
+
+
+   .. py:attribute:: inferencer_args
+
+
+   .. py:attribute:: model_args
+
+
+   .. py:attribute:: local_rank
+
+
+   .. py:attribute:: world_size
+
+
+   .. py:method:: inference(model: lmflow.models.hf_text_regression_model.HFTextRegressionModel, dataset: lmflow.datasets.dataset.Dataset, transform_dataset_in_place: bool = True, use_vllm: bool = False, enable_distributed_inference: bool = False, **kwargs) -> lmflow.datasets.dataset.Dataset
+
+
+   .. py:method:: _inference(model: lmflow.models.hf_text_regression_model.HFTextRegressionModel, model_input: Union[lmflow.datasets.dataset.Dataset, ray.data.Dataset], enable_distributed_inference: bool = False, **kwargs)
+
+
+   .. py:method:: __inference(model: lmflow.models.hf_text_regression_model.HFTextRegressionModel, model_input: lmflow.datasets.dataset.Dataset) -> Union[List[float], List[List[float]]]
+
+
+   .. py:method:: __distributed_inference(model: lmflow.models.hf_text_regression_model.HFTextRegressionModel, model_input: ray.data.Dataset, num_instances: int, batch_size: int) -> List[lmflow.utils.data_utils.RewardModelInferenceResultWithInput]
+
+
+   .. py:method:: __vllm_inference(model: lmflow.models.hf_text_regression_model.HFTextRegressionModel, model_input: List[str], enable_distributed_inference: bool = False) -> List[float]
+      :abstractmethod:
+
+
+
+   .. py:method:: __post_process_model_output(model_output: transformers.modeling_outputs.SequenceClassifierOutputWithPast) -> List[float]
+
+
+   .. py:method:: flatten_list(list_of_list: List[List]) -> Tuple[List, List[int]]
+
+
+   .. py:method:: compress_list(list_to_compress: List, sublist_lengths: List[int]) -> List[List]
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/rm_tuner/index.rst.txt b/_sources/autoapi/lmflow/pipeline/rm_tuner/index.rst.txt
new file mode 100644
index 000000000..c07043786
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/rm_tuner/index.rst.txt
@@ -0,0 +1,100 @@
+lmflow.pipeline.rm_tuner
+========================
+
+.. py:module:: lmflow.pipeline.rm_tuner
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.rm_tuner.logger
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.rm_tuner.RewardModelTuner
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: RewardModelTuner(model_args, data_args, finetuner_args, *args, **kwargs)
+
+   Bases: :py:obj:`lmflow.pipeline.finetuner.Finetuner`
+
+
+   
+   Initializes the `RewardModelTuner` class.
+
+
+   :Parameters:
+
+       **model_args** : ModelArguments object.
+           Contains the arguments required to load the model.
+
+       **data_args** : DatasetArguments object.
+           Contains the arguments required to load the dataset.
+
+       **finetuner_args** : RewardModelTunerArguments object.
+           Contains the arguments required to perform finetuning.
+
+       **args** : Optional.
+           Positional arguments.
+
+       **kwargs** : Optional.
+           Keyword arguments.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: tune(model: lmflow.models.hf_text_regression_model.HFTextRegressionModel, dataset, transform_dataset_in_place=True, data_collator=None, **kwargs)
+
+      
+      Perform tuning for a model
+
+
+      :Parameters:
+
+          **model** : TunableModel object.
+              TunableModel to perform tuning.
+
+          **dataset:**
+              dataset to train model.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.rst.txt b/_sources/autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.rst.txt
new file mode 100644
index 000000000..c634bf403
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.rst.txt
@@ -0,0 +1,130 @@
+lmflow.pipeline.utils.dpov2_dataprocessor
+=========================================
+
+.. py:module:: lmflow.pipeline.utils.dpov2_dataprocessor
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.dpov2_dataprocessor.logger
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: PreferenceDataCollatorWithPadding
+
+   .. py:attribute:: tokenizer
+      :type:  transformers.PreTrainedTokenizerBase
+
+
+   .. py:attribute:: model
+      :type:  Optional[transformers.PreTrainedModel]
+      :value: None
+
+
+
+   .. py:attribute:: padding
+      :type:  Union[bool, str]
+      :value: True
+
+
+
+   .. py:attribute:: max_length
+      :type:  Optional[int]
+      :value: None
+
+
+
+   .. py:attribute:: max_prompt_length
+      :type:  Optional[int]
+      :value: None
+
+
+
+   .. py:attribute:: label_pad_token_id
+      :type:  int
+      :value: -100
+
+
+
+   .. py:attribute:: padding_value
+      :type:  int
+      :value: 0
+
+
+
+   .. py:attribute:: truncation_mode
+      :type:  str
+      :value: 'keep_end'
+
+
+
+   .. py:attribute:: is_encoder_decoder
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:attribute:: max_target_length
+      :type:  Optional[int]
+      :value: None
+
+
+
+   .. py:attribute:: mask_prompt
+      :type:  Optional[bool]
+      :value: False
+
+
+
+   .. py:method:: tokenize_batch_element(prompt: str, chosen: str, rejected: str) -> Dict
+
+      
+      Tokenize a single batch element.
+
+      At this stage, we don't convert to PyTorch tensors yet; we just handle the truncation
+          in case the prompt + chosen or prompt + rejected responses is/are too long. First
+          we truncate the prompt; if we're still too long, we truncate the chosen/rejected.
+
+      We also create the labels for the chosen/rejected responses, which are of length equal to
+          the sum of the length of the prompt and the chosen/rejected response, with
+          label_pad_token_id  for the prompt tokens.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: collate(batch)
+
+
+   .. py:method:: __call__(features: List[Dict[str, Any]]) -> Dict[str, Any]
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/utils/dpov2_trainer/index.rst.txt b/_sources/autoapi/lmflow/pipeline/utils/dpov2_trainer/index.rst.txt
new file mode 100644
index 000000000..4b0b208ce
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/utils/dpov2_trainer/index.rst.txt
@@ -0,0 +1,105 @@
+lmflow.pipeline.utils.dpov2_trainer
+===================================
+
+.. py:module:: lmflow.pipeline.utils.dpov2_trainer
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.dpov2_trainer.logger
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: DPOv2Trainer(model: Union[transformers.PreTrainedModel, torch.nn.Module] = None, ref_model: Optional[Union[transformers.PreTrainedModel, torch.nn.Module]] = None, beta: float = 0.1, loss_type: Literal['sigmoid', 'hinge', 'cross_entropy', 'kl', 'rev_kl', 'raft'] = 'rev_kl', args: transformers.TrainingArguments = None, data_collator: Optional[transformers.DataCollator] = None, label_pad_token_id: int = -100, padding_value: int = 0, truncation_mode: str = 'keep_end', train_dataset: Optional[datasets.Dataset] = None, eval_dataset: Optional[Union[datasets.Dataset, Dict[str, datasets.Dataset]]] = None, tokenizer: Optional[transformers.PreTrainedTokenizerBase] = None, model_init: Optional[Callable[[], transformers.PreTrainedModel]] = None, callbacks: Optional[List[transformers.trainer_callback.TrainerCallback]] = None, optimizers: Tuple[torch.optim.Optimizer, torch.optim.lr_scheduler.LambdaLR] = (None, None), preprocess_logits_for_metrics: Optional[Callable[[torch.Tensor, torch.Tensor], torch.Tensor]] = None, max_length: Optional[int] = None, max_prompt_length: Optional[int] = None, max_target_length: Optional[int] = None, peft_config: Optional[Dict] = None, is_encoder_decoder: Optional[bool] = None, disable_dropout: bool = True, generate_during_eval: bool = False, compute_metrics: Optional[Callable[[transformers.trainer_utils.EvalLoopOutput], Dict]] = None, mask_prompt: Optional[bool] = False, len_penalty: float = 0, preprocessing_num_workers: int = 1)
+
+   Bases: :py:obj:`trl.DPOTrainer`
+
+
+   .. py:attribute:: use_dpo_data_collator
+      :value: True
+
+
+
+   .. py:attribute:: len_penalty
+      :value: 0
+
+
+
+   .. py:method:: dpo_loss(policy_chosen_logps: torch.FloatTensor, policy_rejected_logps: torch.FloatTensor, reference_chosen_logps: torch.FloatTensor, reference_rejected_logps: torch.FloatTensor, reference_free: bool = False, margin: Optional[torch.FloatTensor] = None, len_penalty: float = 0) -> Tuple[torch.FloatTensor, torch.FloatTensor, torch.FloatTensor]
+
+      
+      Compute the DPO loss for a batch of policy and reference model log probabilities.
+
+      Args:
+          policy_chosen_logps: Log probabilities of the policy model for the chosen responses. Shape: (batch_size,)
+          policy_rejected_logps: Log probabilities of the policy model for the rejected responses. Shape: (batch_size,)
+          reference_chosen_logps: Log probabilities of the reference model for the chosen responses. Shape: (batch_size,)
+          reference_rejected_logps: Log probabilities of the reference model for the rejected responses. Shape: (batch_size,)
+          beta: Temperature parameter for the DPO loss, typically something in the range of 0.1 to 0.5. We ignore the reference model as beta -> 0.
+          reference_free: If True, we ignore the _provided_ reference model and implicitly use a reference model that assigns equal probability to all responses.
+
+      Returns:
+          A tuple of three tensors: (losses, chosen_rewards, rejected_rewards).
+          The losses tensor contains the DPO loss for each example in the batch.
+          The chosen_rewards and rejected_rewards tensors contain the rewards for the chosen and rejected responses, respectively.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_batch_loss_metrics(model, batch: Dict[str, Union[List, torch.LongTensor]], train_eval: Literal['train', 'eval'] = 'train')
+
+
+   .. py:method:: get_batch_metrics(model, batch: Dict[str, Union[List, torch.LongTensor]], train_eval: Literal['train', 'eval'] = 'train')
+
+      
+      Compute the DPO loss and other metrics for the given batch of inputs for train or test.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/utils/index.rst.txt b/_sources/autoapi/lmflow/pipeline/utils/index.rst.txt
new file mode 100644
index 000000000..68ca9f80d
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/utils/index.rst.txt
@@ -0,0 +1,22 @@
+lmflow.pipeline.utils
+=====================
+
+.. py:module:: lmflow.pipeline.utils
+
+
+Submodules
+----------
+
+.. toctree::
+   :maxdepth: 1
+
+   /autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index
+   /autoapi/lmflow/pipeline/utils/dpov2_trainer/index
+   /autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index
+   /autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index
+   /autoapi/lmflow/pipeline/utils/peft_trainer/index
+   /autoapi/lmflow/pipeline/utils/raft_trainer/index
+   /autoapi/lmflow/pipeline/utils/rm_dataprocessor/index
+   /autoapi/lmflow/pipeline/utils/rm_trainer/index
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.rst.txt b/_sources/autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.rst.txt
new file mode 100644
index 000000000..023a68c13
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.rst.txt
@@ -0,0 +1,33 @@
+lmflow.pipeline.utils.memory_safe_dpov2_align
+=============================================
+
+.. py:module:: lmflow.pipeline.utils.memory_safe_dpov2_align
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.memory_safe_dpov2_align.logger
+   lmflow.pipeline.utils.memory_safe_dpov2_align.ReferenceModelArguments
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.memory_safe_dpov2_align.main
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:data:: ReferenceModelArguments
+   :type:  lmflow.args.ModelArguments
+
+.. py:function:: main()
+
diff --git a/_sources/autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.rst.txt b/_sources/autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.rst.txt
new file mode 100644
index 000000000..53ae3f3a7
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.rst.txt
@@ -0,0 +1,29 @@
+lmflow.pipeline.utils.memory_safe_vllm_inference
+================================================
+
+.. py:module:: lmflow.pipeline.utils.memory_safe_vllm_inference
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.memory_safe_vllm_inference.logger
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.memory_safe_vllm_inference.main
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:function:: main()
+
diff --git a/_sources/autoapi/lmflow/pipeline/utils/peft_trainer/index.rst.txt b/_sources/autoapi/lmflow/pipeline/utils/peft_trainer/index.rst.txt
new file mode 100644
index 000000000..aed03c218
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/utils/peft_trainer/index.rst.txt
@@ -0,0 +1,135 @@
+lmflow.pipeline.utils.peft_trainer
+==================================
+
+.. py:module:: lmflow.pipeline.utils.peft_trainer
+
+.. autoapi-nested-parse::
+
+   Trainer for Peft models
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.peft_trainer.PeftTrainer
+   lmflow.pipeline.utils.peft_trainer.PeftSavingCallback
+
+
+Module Contents
+---------------
+
+.. py:class:: PeftTrainer
+
+   Bases: :py:obj:`transformers.Trainer`
+
+
+   .. py:method:: _save_checkpoint(_, trial, metrics=None)
+
+      
+      Don't save base model, optimizer etc.
+      but create checkpoint folder (needed for saving adapter) 
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+.. py:class:: PeftSavingCallback
+
+   Bases: :py:obj:`transformers.trainer_callback.TrainerCallback`
+
+
+   
+   Correctly save PEFT model and not full model 
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: _save(model, folder)
+
+
+   .. py:method:: on_train_end(args: transformers.training_args.TrainingArguments, state: transformers.trainer_callback.TrainerState, control: transformers.trainer_callback.TrainerControl, **kwargs)
+
+      
+      Save final best model adapter 
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: on_epoch_end(args: transformers.training_args.TrainingArguments, state: transformers.trainer_callback.TrainerState, control: transformers.trainer_callback.TrainerControl, **kwargs)
+
+      
+      Save intermediate model adapters in case of interrupted training 
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: on_save(args: transformers.training_args.TrainingArguments, state: transformers.trainer_callback.TrainerState, control: transformers.trainer_callback.TrainerControl, **kwargs)
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/utils/raft_trainer/index.rst.txt b/_sources/autoapi/lmflow/pipeline/utils/raft_trainer/index.rst.txt
new file mode 100644
index 000000000..15c2931ee
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/utils/raft_trainer/index.rst.txt
@@ -0,0 +1,1560 @@
+lmflow.pipeline.utils.raft_trainer
+==================================
+
+.. py:module:: lmflow.pipeline.utils.raft_trainer
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.raft_trainer.is_torch_greater_or_equal_than_1_10
+   lmflow.pipeline.utils.raft_trainer.is_torch_less_than_1_11
+   lmflow.pipeline.utils.raft_trainer._is_native_cpu_amp_available
+   lmflow.pipeline.utils.raft_trainer.DEFAULT_CALLBACKS
+   lmflow.pipeline.utils.raft_trainer.DEFAULT_PROGRESS_CALLBACK
+   lmflow.pipeline.utils.raft_trainer.DEFAULT_PROGRESS_CALLBACK
+   lmflow.pipeline.utils.raft_trainer.IS_SAGEMAKER_MP_POST_1_10
+   lmflow.pipeline.utils.raft_trainer.skip_first_batches
+   lmflow.pipeline.utils.raft_trainer.logger
+   lmflow.pipeline.utils.raft_trainer.TRAINING_ARGS_NAME
+   lmflow.pipeline.utils.raft_trainer.TRAINER_STATE_NAME
+   lmflow.pipeline.utils.raft_trainer.OPTIMIZER_NAME
+   lmflow.pipeline.utils.raft_trainer.SCHEDULER_NAME
+   lmflow.pipeline.utils.raft_trainer.SCALER_NAME
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.raft_trainer.RaftTrainer
+
+
+Module Contents
+---------------
+
+.. py:data:: is_torch_greater_or_equal_than_1_10
+
+.. py:data:: is_torch_less_than_1_11
+
+.. py:data:: _is_native_cpu_amp_available
+
+.. py:data:: DEFAULT_CALLBACKS
+
+.. py:data:: DEFAULT_PROGRESS_CALLBACK
+
+.. py:data:: DEFAULT_PROGRESS_CALLBACK
+
+.. py:data:: IS_SAGEMAKER_MP_POST_1_10
+
+.. py:data:: skip_first_batches
+   :value: None
+
+
+.. py:data:: logger
+
+.. py:data:: TRAINING_ARGS_NAME
+   :value: 'training_args.bin'
+
+
+.. py:data:: TRAINER_STATE_NAME
+   :value: 'trainer_state.json'
+
+
+.. py:data:: OPTIMIZER_NAME
+   :value: 'optimizer.pt'
+
+
+.. py:data:: SCHEDULER_NAME
+   :value: 'scheduler.pt'
+
+
+.. py:data:: SCALER_NAME
+   :value: 'scaler.pt'
+
+
+.. py:class:: RaftTrainer(model: Union[transformers.modeling_utils.PreTrainedModel, torch.nn.Module] = None, args: transformers.training_args.TrainingArguments = None, data_collator: Optional[transformers.data.data_collator.DataCollator] = None, train_dataset: Optional[torch.utils.data.Dataset] = None, eval_dataset: Optional[Union[torch.utils.data.Dataset, Dict[str, torch.utils.data.Dataset]]] = None, tokenizer: Optional[transformers.tokenization_utils_base.PreTrainedTokenizerBase] = None, model_init: Optional[Callable[[], transformers.modeling_utils.PreTrainedModel]] = None, compute_metrics: Optional[Callable[[transformers.trainer_utils.EvalPrediction], Dict]] = None, callbacks: Optional[List[transformers.trainer_callback.TrainerCallback]] = None, optimizers: Tuple[torch.optim.Optimizer, torch.optim.lr_scheduler.LambdaLR] = (None, None), preprocess_logits_for_metrics: Optional[Callable[[torch.Tensor, torch.Tensor], torch.Tensor]] = None)
+
+   
+   Trainer is a simple but feature-complete training and eval loop for PyTorch, optimized for 🤗 Transformers.
+   Args:
+       model ([`PreTrainedModel`] or `torch.nn.Module`, *optional*):
+           The model to train, evaluate or use for predictions. If not provided, a `model_init` must be passed.
+           <Tip>
+           [`Trainer`] is optimized to work with the [`PreTrainedModel`] provided by the library. You can still use
+           your own models defined as `torch.nn.Module` as long as they work the same way as the 🤗 Transformers
+           models.
+           </Tip>
+       args ([`TrainingArguments`], *optional*):
+           The arguments to tweak for training. Will default to a basic instance of [`TrainingArguments`] with the
+           `output_dir` set to a directory named *tmp_trainer* in the current directory if not provided.
+       data_collator (`DataCollator`, *optional*):
+           The function to use to form a batch from a list of elements of `train_dataset` or `eval_dataset`. Will
+           default to [`default_data_collator`] if no `tokenizer` is provided, an instance of
+           [`DataCollatorWithPadding`] otherwise.
+       train_dataset (`torch.utils.data.Dataset` or `torch.utils.data.IterableDataset`, *optional*):
+           The dataset to use for training. If it is a [`~datasets.Dataset`], columns not accepted by the
+           `model.forward()` method are automatically removed.
+           Note that if it's a `torch.utils.data.IterableDataset` with some randomization and you are training in a
+           distributed fashion, your iterable dataset should either use a internal attribute `generator` that is a
+           `torch.Generator` for the randomization that must be identical on all processes (and the Trainer will
+           manually set the seed of this `generator` at each epoch) or have a `set_epoch()` method that internally
+           sets the seed of the RNGs used.
+       eval_dataset (Union[`torch.utils.data.Dataset`, Dict[str, `torch.utils.data.Dataset`]), *optional*):
+            The dataset to use for evaluation. If it is a [`~datasets.Dataset`], columns not accepted by the
+            `model.forward()` method are automatically removed. If it is a dictionary, it will evaluate on each
+            dataset prepending the dictionary key to the metric name.
+       tokenizer ([`PreTrainedTokenizerBase`], *optional*):
+           The tokenizer used to preprocess the data. If provided, will be used to automatically pad the inputs to the
+           maximum length when batching inputs, and it will be saved along the model to make it easier to rerun an
+           interrupted training or reuse the fine-tuned model.
+       model_init (`Callable[[], PreTrainedModel]`, *optional*):
+           A function that instantiates the model to be used. If provided, each call to [`~Trainer.train`] will start
+           from a new instance of the model as given by this function.
+           The function may have zero argument, or a single one containing the optuna/Ray Tune/SigOpt trial object, to
+           be able to choose different architectures according to hyper parameters (such as layer count, sizes of
+           inner layers, dropout probabilities etc).
+       compute_metrics (`Callable[[EvalPrediction], Dict]`, *optional*):
+           The function that will be used to compute metrics at evaluation. Must take a [`EvalPrediction`] and return
+           a dictionary string to metric values.
+       callbacks (List of [`TrainerCallback`], *optional*):
+           A list of callbacks to customize the training loop. Will add those to the list of default callbacks
+           detailed in [here](callback).
+           If you want to remove one of the default callbacks used, use the [`Trainer.remove_callback`] method.
+       optimizers (`Tuple[torch.optim.Optimizer, torch.optim.lr_scheduler.LambdaLR]`, *optional*): A tuple
+           containing the optimizer and the scheduler to use. Will default to an instance of [`AdamW`] on your model
+           and a scheduler given by [`get_linear_schedule_with_warmup`] controlled by `args`.
+       preprocess_logits_for_metrics (`Callable[[torch.Tensor, torch.Tensor], torch.Tensor]`, *optional*):
+           A function that preprocess the logits right before caching them at each evaluation step. Must take two
+           tensors, the logits and the labels, and return the logits once processed as desired. The modifications made
+           by this function will be reflected in the predictions received by `compute_metrics`.
+           Note that the labels (second parameter) will be `None` if the dataset does not have them.
+   Important attributes:
+       - **model** -- Always points to the core model. If using a transformers model, it will be a [`PreTrainedModel`]
+         subclass.
+       - **model_wrapped** -- Always points to the most external model in case one or more other modules wrap the
+         original model. This is the model that should be used for the forward pass. For example, under `DeepSpeed`,
+         the inner model is wrapped in `DeepSpeed` and then again in `torch.nn.DistributedDataParallel`. If the inner
+         model hasn't been wrapped, then `self.model_wrapped` is the same as `self.model`.
+       - **is_model_parallel** -- Whether or not a model has been switched to a model parallel mode (different from
+         data parallelism, this means some of the model layers are split on different GPUs).
+       - **place_model_on_device** -- Whether or not to automatically place the model on the device - it will be set
+         to `False` if model parallel or deepspeed is used, or if the default
+         `TrainingArguments.place_model_on_device` is overridden to return `False` .
+       - **is_in_train** -- Whether or not a model is currently running `train` (e.g. when `evaluate` is called while
+         in `train`)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: save_counter
+      :value: 0
+
+
+
+   .. py:attribute:: args
+      :value: None
+
+
+
+   .. py:attribute:: hp_name
+      :value: None
+
+
+
+   .. py:attribute:: deepspeed
+      :value: None
+
+
+
+   .. py:attribute:: is_in_train
+      :value: False
+
+
+
+   .. py:attribute:: _memory_tracker
+
+
+   .. py:attribute:: sharded_ddp
+      :value: None
+
+
+
+   .. py:attribute:: fsdp
+      :value: None
+
+
+
+   .. py:attribute:: place_model_on_device
+
+
+   .. py:attribute:: data_collator
+      :value: None
+
+
+
+   .. py:attribute:: train_dataset
+      :value: None
+
+
+
+   .. py:attribute:: eval_dataset
+      :value: None
+
+
+
+   .. py:attribute:: tokenizer
+      :value: None
+
+
+
+   .. py:attribute:: model_wrapped
+      :value: None
+
+
+
+   .. py:attribute:: model
+      :value: None
+
+
+
+   .. py:attribute:: compute_metrics
+      :value: None
+
+
+
+   .. py:attribute:: preprocess_logits_for_metrics
+      :value: None
+
+
+
+   .. py:attribute:: callback_handler
+
+
+   .. py:attribute:: _loggers_initialized
+      :value: False
+
+
+
+   .. py:attribute:: _signature_columns
+      :value: None
+
+
+
+   .. py:attribute:: use_apex
+      :value: False
+
+
+
+   .. py:attribute:: use_cuda_amp
+      :value: False
+
+
+
+   .. py:attribute:: use_cpu_amp
+      :value: False
+
+
+
+   .. py:attribute:: do_grad_scaling
+      :value: False
+
+
+
+   .. py:attribute:: state
+
+
+   .. py:attribute:: control
+
+
+   .. py:attribute:: current_flos
+      :value: 0
+
+
+
+   .. py:attribute:: hp_search_backend
+      :value: None
+
+
+
+   .. py:attribute:: use_tune_checkpoints
+      :value: False
+
+
+
+   .. py:attribute:: label_names
+
+
+   .. py:attribute:: can_return_loss
+
+
+   .. py:attribute:: _train_batch_size
+
+
+   .. py:method:: add_callback(callback)
+
+      
+      Add a callback to the current list of [`~transformer.TrainerCallback`].
+      Args:
+         callback (`type` or [`~transformer.TrainerCallback`]):
+             A [`~transformer.TrainerCallback`] class or an instance of a [`~transformer.TrainerCallback`]. In the
+             first case, will instantiate a member of that class.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: pop_callback(callback)
+
+      
+      Remove a callback from the current list of [`~transformer.TrainerCallback`] and returns it.
+      If the callback is not found, returns `None` (and no error is raised).
+      Args:
+         callback (`type` or [`~transformer.TrainerCallback`]):
+             A [`~transformer.TrainerCallback`] class or an instance of a [`~transformer.TrainerCallback`]. In the
+             first case, will pop the first member of that class found in the list of callbacks.
+      Returns:
+          [`~transformer.TrainerCallback`]: The callback removed, if found.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: remove_callback(callback)
+
+      
+      Remove a callback from the current list of [`~transformer.TrainerCallback`].
+      Args:
+         callback (`type` or [`~transformer.TrainerCallback`]):
+             A [`~transformer.TrainerCallback`] class or an instance of a [`~transformer.TrainerCallback`]. In the
+             first case, will remove the first member of that class found in the list of callbacks.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _move_model_to_device(model, device)
+
+
+   .. py:method:: _set_signature_columns_if_needed()
+
+
+   .. py:method:: _remove_unused_columns(dataset: datasets.Dataset, description: Optional[str] = None)
+
+
+   .. py:method:: _get_collator_with_removed_columns(data_collator: Callable, description: Optional[str] = None) -> Callable
+
+      
+      Wrap the data collator in a callable removing unused columns.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _get_train_sampler() -> Optional[torch.utils.data.Sampler]
+
+
+   .. py:method:: get_train_dataloader() -> torch.utils.data.DataLoader
+
+      
+      Returns the training [`~torch.utils.data.DataLoader`].
+      Will use no sampler if `train_dataset` does not implement `__len__`, a random sampler (adapted to distributed
+      training if necessary) otherwise.
+      Subclass and override this method if you want to inject some custom behavior.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _get_eval_sampler(eval_dataset: torch.utils.data.Dataset) -> Optional[torch.utils.data.Sampler]
+
+
+   .. py:method:: get_eval_dataloader(eval_dataset: Optional[torch.utils.data.Dataset] = None) -> torch.utils.data.DataLoader
+
+      
+      Returns the evaluation [`~torch.utils.data.DataLoader`].
+      Subclass and override this method if you want to inject some custom behavior.
+      Args:
+          eval_dataset (`torch.utils.data.Dataset`, *optional*):
+              If provided, will override `self.eval_dataset`. If it is a [`~datasets.Dataset`], columns not accepted
+              by the `model.forward()` method are automatically removed. It must implement `__len__`.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_test_dataloader(test_dataset: torch.utils.data.Dataset) -> torch.utils.data.DataLoader
+
+      
+      Returns the test [`~torch.utils.data.DataLoader`].
+      Subclass and override this method if you want to inject some custom behavior.
+      Args:
+          test_dataset (`torch.utils.data.Dataset`, *optional*):
+              The test dataset to use. If it is a [`~datasets.Dataset`], columns not accepted by the
+              `model.forward()` method are automatically removed. It must implement `__len__`.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: create_optimizer_and_scheduler(num_training_steps: int)
+
+      
+      Setup the optimizer and the learning rate scheduler.
+      We provide a reasonable default that works well. If you want to use something else, you can pass a tuple in the
+      Trainer's init through `optimizers`, or subclass and override this method (or `create_optimizer` and/or
+      `create_scheduler`) in a subclass.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: create_optimizer()
+
+      
+      Setup the optimizer.
+      We provide a reasonable default that works well. If you want to use something else, you can pass a tuple in the
+      Trainer's init through `optimizers`, or subclass and override this method in a subclass.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: get_optimizer_cls_and_kwargs(args: transformers.training_args.TrainingArguments) -> Tuple[Any, Any]
+      :staticmethod:
+
+
+      
+      Returns the optimizer class and optimizer parameters based on the training arguments.
+      Args:
+          args (`transformers.training_args.TrainingArguments`):
+              The training arguments for the training session.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: create_scheduler(num_training_steps: int, optimizer: torch.optim.Optimizer = None)
+
+      
+      Setup the scheduler. The optimizer of the trainer must have been set up either before this method is called or
+      passed as an argument.
+      Args:
+          num_training_steps (int): The number of training steps to do.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: num_examples(dataloader: torch.utils.data.DataLoader) -> int
+
+      
+      Helper to get number of samples in a [`~torch.utils.data.DataLoader`] by accessing its dataset. When
+      dataloader.dataset does not exist or has no length, estimates as best it can
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _hp_search_setup(trial: Union[optuna.Trial, Dict[str, Any]])
+
+      
+      HP search setup code
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _report_to_hp_search(trial: Union[optuna.Trial, Dict[str, Any]], step: int, metrics: Dict[str, float])
+
+
+   .. py:method:: _tune_save_checkpoint()
+
+
+   .. py:method:: call_model_init(trial=None)
+
+
+   .. py:method:: torch_jit_model_eval(model, dataloader, training=False)
+
+
+   .. py:method:: ipex_optimize_model(model, training=False, dtype=torch.float32)
+
+
+   .. py:method:: _wrap_model(model, training=True, dataloader=None)
+
+
+   .. py:method:: train(resume_from_checkpoint: Optional[Union[str, bool]] = None, trial: Union[optuna.Trial, Dict[str, Any]] = None, ignore_keys_for_eval: Optional[List[str]] = None, is_first_time=False, **kwargs)
+
+      
+      Main training entry point.
+      Args:
+          resume_from_checkpoint (`str` or `bool`, *optional*):
+              If a `str`, local path to a saved checkpoint as saved by a previous instance of [`Trainer`]. If a
+              `bool` and equals `True`, load the last checkpoint in *args.output_dir* as saved by a previous instance
+              of [`Trainer`]. If present, training will resume from the model/optimizer/scheduler states loaded here.
+          trial (`optuna.Trial` or `Dict[str, Any]`, *optional*):
+              The trial run or the hyperparameter dictionary for hyperparameter search.
+          ignore_keys_for_eval (`List[str]`, *optional*)
+              A list of keys in the output of your model (if it is a dictionary) that should be ignored when
+              gathering predictions for evaluation during the training.
+          kwargs:
+              Additional keyword arguments used to hide deprecated arguments
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _one_train(batch_size=None, args=None, resume_from_checkpoint=None, trial=None, ignore_keys_for_eval=None)
+
+
+   .. py:method:: _inner_training_loop(batch_size=None, args=None, resume_from_checkpoint=None, trial=None, ignore_keys_for_eval=None)
+
+      
+      0 This function serves to train one time
+      1 Update the self.train_dataset before calling this function
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _get_output_dir(trial)
+
+
+   .. py:method:: _load_from_checkpoint(resume_from_checkpoint, model=None)
+
+
+   .. py:method:: _load_best_model()
+
+
+   .. py:method:: _issue_warnings_after_load(load_result)
+
+
+   .. py:method:: _maybe_log_save_evaluate(tr_loss, model, trial, epoch, ignore_keys_for_eval)
+
+
+   .. py:method:: _load_rng_state(checkpoint)
+
+
+   .. py:method:: _save_checkpoint(model, trial, metrics=None)
+
+
+   .. py:method:: _load_optimizer_and_scheduler(checkpoint)
+
+      
+      If optimizer and scheduler states exist, load them.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: hyperparameter_search(hp_space: Optional[Callable[[optuna.Trial], Dict[str, float]]] = None, compute_objective: Optional[Callable[[Dict[str, float]], float]] = None, n_trials: int = 20, direction: str = 'minimize', backend: Optional[Union[str, transformers.trainer_utils.HPSearchBackend]] = None, hp_name: Optional[Callable[[optuna.Trial], str]] = None, **kwargs) -> transformers.trainer_utils.BestRun
+
+      
+      Launch an hyperparameter search using `optuna` or `Ray Tune` or `SigOpt`. The optimized quantity is determined
+      by `compute_objective`, which defaults to a function returning the evaluation loss when no metric is provided,
+      the sum of all metrics otherwise.
+      <Tip warning={true}>
+      To use this method, you need to have provided a `model_init` when initializing your [`Trainer`]: we need to
+      reinitialize the model at each new run. This is incompatible with the `optimizers` argument, so you need to
+      subclass [`Trainer`] and override the method [`~Trainer.create_optimizer_and_scheduler`] for custom
+      optimizer/scheduler.
+      </Tip>
+      Args:
+          hp_space (`Callable[["optuna.Trial"], Dict[str, float]]`, *optional*):
+              A function that defines the hyperparameter search space. Will default to
+              [`~trainer_utils.default_hp_space_optuna`] or [`~trainer_utils.default_hp_space_ray`] or
+              [`~trainer_utils.default_hp_space_sigopt`] depending on your backend.
+          compute_objective (`Callable[[Dict[str, float]], float]`, *optional*):
+              A function computing the objective to minimize or maximize from the metrics returned by the `evaluate`
+              method. Will default to [`~trainer_utils.default_compute_objective`].
+          n_trials (`int`, *optional*, defaults to 100):
+              The number of trial runs to test.
+          direction (`str`, *optional*, defaults to `"minimize"`):
+              Whether to optimize greater or lower objects. Can be `"minimize"` or `"maximize"`, you should pick
+              `"minimize"` when optimizing the validation loss, `"maximize"` when optimizing one or several metrics.
+          backend (`str` or [`~training_utils.HPSearchBackend`], *optional*):
+              The backend to use for hyperparameter search. Will default to optuna or Ray Tune or SigOpt, depending
+              on which one is installed. If all are installed, will default to optuna.
+          hp_name (`Callable[["optuna.Trial"], str]]`, *optional*):
+              A function that defines the trial/run name. Will default to None.
+          kwargs (`Dict[str, Any]`, *optional*):
+              Additional keyword arguments passed along to `optuna.create_study` or `ray.tune.run`. For more
+              information see:
+              - the documentation of
+                [optuna.create_study](https://optuna.readthedocs.io/en/stable/reference/generated/optuna.study.create_study.html)
+              - the documentation of [tune.run](https://docs.ray.io/en/latest/tune/api_docs/execution.html#tune-run)
+              - the documentation of [sigopt](https://app.sigopt.com/docs/endpoints/experiments/create)
+      Returns:
+          [`trainer_utils.BestRun`]: All the information about the best run. Experiment summary can be found in
+          `run_summary` attribute for Ray backend.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: log(logs: Dict[str, float]) -> None
+
+      
+      Log `logs` on the various objects watching training.
+      Subclass and override this method to inject custom behavior.
+      Args:
+          logs (`Dict[str, float]`):
+              The values to log.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _prepare_input(data: Union[torch.Tensor, Any]) -> Union[torch.Tensor, Any]
+
+      
+      Prepares one `data` before feeding it to the model, be it a tensor or a nested list/dictionary of tensors.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _prepare_inputs(inputs: Dict[str, Union[torch.Tensor, Any]]) -> Dict[str, Union[torch.Tensor, Any]]
+
+      
+      Prepare `inputs` before feeding them to the model, converting them to tensors if they are not already and
+      handling potential state.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: compute_loss_context_manager()
+
+      
+      A helper wrapper to group together context managers.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: autocast_smart_context_manager(cache_enabled: Optional[bool] = True)
+
+      
+      A helper wrapper that creates an appropriate context manager for `autocast` while feeding it the desired
+      arguments, depending on the situation.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: training_step(model: torch.nn.Module, inputs: Dict[str, Union[torch.Tensor, Any]]) -> torch.Tensor
+
+      
+      Perform a training step on a batch of inputs.
+      Subclass and override to inject custom behavior.
+      Args:
+          model (`nn.Module`):
+              The model to train.
+          inputs (`Dict[str, Union[torch.Tensor, Any]]`):
+              The inputs and targets of the model.
+              The dictionary will be unpacked before being fed to the model. Most models expect the targets under the
+              argument `labels`. Check your model's documentation for all accepted arguments.
+      Return:
+          `torch.Tensor`: The tensor with training loss on this batch.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: compute_loss(model, inputs, return_outputs=False)
+
+      
+      How the loss is computed by Trainer. By default, all models return the loss in the first element.
+      Subclass and override for custom behavior.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: is_local_process_zero() -> bool
+
+      
+      Whether or not this process is the local (e.g., on one machine if training in a distributed fashion on several
+      machines) main process.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: is_world_process_zero() -> bool
+
+      
+      Whether or not this process is the global main process (when training in a distributed fashion on several
+      machines, this is only going to be `True` for one process).
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: save_model(output_dir: Optional[str] = None, _internal_call: bool = False)
+
+      
+      Will save the model, so you can reload it using `from_pretrained()`.
+      Will only save from the main process.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _save_tpu(output_dir: Optional[str] = None)
+
+
+   .. py:method:: _save(output_dir: Optional[str] = None, state_dict=None)
+
+
+   .. py:method:: store_flos()
+
+
+   .. py:method:: _sorted_checkpoints(output_dir=None, checkpoint_prefix=PREFIX_CHECKPOINT_DIR, use_mtime=False) -> List[str]
+
+
+   .. py:method:: _rotate_checkpoints(use_mtime=False, output_dir=None) -> None
+
+
+   .. py:method:: evaluate(eval_dataset: Optional[torch.utils.data.Dataset] = None, ignore_keys: Optional[List[str]] = None, metric_key_prefix: str = 'eval') -> Dict[str, float]
+
+      
+      Run evaluation and returns metrics.
+      The calling script will be responsible for providing a method to compute metrics, as they are task-dependent
+      (pass it to the init `compute_metrics` argument).
+      You can also subclass and override this method to inject custom behavior.
+      Args:
+          eval_dataset (`Dataset`, *optional*):
+              Pass a dataset if you wish to override `self.eval_dataset`. If it is a [`~datasets.Dataset`], columns
+              not accepted by the `model.forward()` method are automatically removed. It must implement the `__len__`
+              method.
+          ignore_keys (`Lst[str]`, *optional*):
+              A list of keys in the output of your model (if it is a dictionary) that should be ignored when
+              gathering predictions.
+          metric_key_prefix (`str`, *optional*, defaults to `"eval"`):
+              An optional prefix to be used as the metrics key prefix. For example the metrics "bleu" will be named
+              "eval_bleu" if the prefix is "eval" (default)
+      Returns:
+          A dictionary containing the evaluation loss and the potential metrics computed from the predictions. The
+          dictionary also contains the epoch number which comes from the training state.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: predict(test_dataset: torch.utils.data.Dataset, ignore_keys: Optional[List[str]] = None, metric_key_prefix: str = 'test') -> transformers.trainer_utils.PredictionOutput
+
+      
+      Run prediction and returns predictions and potential metrics.
+      Depending on the dataset and your use case, your test dataset may contain labels. In that case, this method
+      will also return metrics, like in `evaluate()`.
+      Args:
+          test_dataset (`Dataset`):
+              Dataset to run the predictions on. If it is an `datasets.Dataset`, columns not accepted by the
+              `model.forward()` method are automatically removed. Has to implement the method `__len__`
+          ignore_keys (`Lst[str]`, *optional*):
+              A list of keys in the output of your model (if it is a dictionary) that should be ignored when
+              gathering predictions.
+          metric_key_prefix (`str`, *optional*, defaults to `"test"`):
+              An optional prefix to be used as the metrics key prefix. For example the metrics "bleu" will be named
+              "test_bleu" if the prefix is "test" (default)
+      <Tip>
+      If your predictions or labels have different sequence length (for instance because you're doing dynamic padding
+      in a token classification task) the predictions will be padded (on the right) to allow for concatenation into
+      one array. The padding index is -100.
+      </Tip>
+      Returns: *NamedTuple* A namedtuple with the following keys:
+          - predictions (`np.ndarray`): The predictions on `test_dataset`.
+          - label_ids (`np.ndarray`, *optional*): The labels (if the dataset contained some).
+          - metrics (`Dict[str, float]`, *optional*): The potential dictionary of metrics (if the dataset contained
+            labels).
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: evaluation_loop(dataloader: torch.utils.data.DataLoader, description: str, prediction_loss_only: Optional[bool] = None, ignore_keys: Optional[List[str]] = None, metric_key_prefix: str = 'eval') -> transformers.trainer_utils.EvalLoopOutput
+
+      
+      Prediction/evaluation loop, shared by `Trainer.evaluate()` and `Trainer.predict()`.
+      Works both with or without labels.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _nested_gather(tensors, name=None)
+
+      
+      Gather value of `tensors` (tensor or list/tuple of nested tensors) and convert them to numpy before
+      concatenating them to `gathered`
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _pad_across_processes(tensor, pad_index=-100)
+
+      
+      Recursively pad the tensors in a nested list/tuple/dictionary of tensors from all devices to the same size so
+      they can safely be gathered.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: prediction_step(model: torch.nn.Module, inputs: Dict[str, Union[torch.Tensor, Any]], prediction_loss_only: bool, ignore_keys: Optional[List[str]] = None) -> Tuple[Optional[torch.Tensor], Optional[torch.Tensor], Optional[torch.Tensor]]
+
+      
+      Perform an evaluation step on `model` using `inputs`.
+      Subclass and override to inject custom behavior.
+      Args:
+          model (`nn.Module`):
+              The model to evaluate.
+          inputs (`Dict[str, Union[torch.Tensor, Any]]`):
+              The inputs and targets of the model.
+              The dictionary will be unpacked before being fed to the model. Most models expect the targets under the
+              argument `labels`. Check your model's documentation for all accepted arguments.
+          prediction_loss_only (`bool`):
+              Whether or not to return the loss only.
+          ignore_keys (`Lst[str]`, *optional*):
+              A list of keys in the output of your model (if it is a dictionary) that should be ignored when
+              gathering predictions.
+      Return:
+          Tuple[Optional[torch.Tensor], Optional[torch.Tensor], Optional[torch.Tensor]]: A tuple with the loss,
+          logits and labels (each being optional).
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: floating_point_ops(inputs: Dict[str, Union[torch.Tensor, Any]])
+
+      
+      For models that inherit from [`PreTrainedModel`], uses that method to compute the number of floating point
+      operations for every backward + forward pass. If using another model, either implement such a method in the
+      model or subclass and override this method.
+      Args:
+          inputs (`Dict[str, Union[torch.Tensor, Any]]`):
+              The inputs and targets of the model.
+      Returns:
+          `int`: The number of floating-point operations.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: init_git_repo(at_init: bool = False)
+
+      
+      Initializes a git repo in `self.args.hub_model_id`.
+      Args:
+          at_init (`bool`, *optional*, defaults to `False`):
+              Whether this function is called before any training or not. If `self.args.overwrite_output_dir` is
+              `True` and `at_init` is `True`, the path to the repo (which is `self.args.output_dir`) might be wiped
+              out.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: create_model_card(language: Optional[str] = None, license: Optional[str] = None, tags: Union[str, List[str], None] = None, model_name: Optional[str] = None, finetuned_from: Optional[str] = None, tasks: Union[str, List[str], None] = None, dataset_tags: Union[str, List[str], None] = None, dataset: Union[str, List[str], None] = None, dataset_args: Union[str, List[str], None] = None)
+
+      
+      Creates a draft of a model card using the information available to the `Trainer`.
+      Args:
+          language (`str`, *optional*):
+              The language of the model (if applicable)
+          license (`str`, *optional*):
+              The license of the model. Will default to the license of the pretrained model used, if the original
+              model given to the `Trainer` comes from a repo on the Hub.
+          tags (`str` or `List[str]`, *optional*):
+              Some tags to be included in the metadata of the model card.
+          model_name (`str`, *optional*):
+              The name of the model.
+          finetuned_from (`str`, *optional*):
+              The name of the model used to fine-tune this one (if applicable). Will default to the name of the repo
+              of the original model given to the `Trainer` (if it comes from the Hub).
+          tasks (`str` or `List[str]`, *optional*):
+              One or several task identifiers, to be included in the metadata of the model card.
+          dataset_tags (`str` or `List[str]`, *optional*):
+              One or several dataset tags, to be included in the metadata of the model card.
+          dataset (`str` or `List[str]`, *optional*):
+              One or several dataset identifiers, to be included in the metadata of the model card.
+          dataset_args (`str` or `List[str]`, *optional*):
+             One or several dataset arguments, to be included in the metadata of the model card.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _push_from_checkpoint(checkpoint_folder)
+
+
+   .. py:method:: push_to_hub(commit_message: Optional[str] = 'End of training', blocking: bool = True, **kwargs) -> str
+
+      
+      Upload *self.model* and *self.tokenizer* to the 🤗 model hub on the repo *self.args.hub_model_id*.
+      Parameters:
+          commit_message (`str`, *optional*, defaults to `"End of training"`):
+              Message to commit while pushing.
+          blocking (`bool`, *optional*, defaults to `True`):
+              Whether the function should return only when the `git push` has finished.
+          kwargs:
+              Additional keyword arguments passed along to [`~Trainer.create_model_card`].
+      Returns:
+          The url of the commit of your model in the given repository if `blocking=False`, a tuple with the url of
+          the commit and an object to track the progress of the commit if `blocking=True`
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: prediction_loop(dataloader: torch.utils.data.DataLoader, description: str, prediction_loss_only: Optional[bool] = None, ignore_keys: Optional[List[str]] = None, metric_key_prefix: str = 'eval') -> transformers.trainer_utils.EvalLoopOutput
+
+      
+      Prediction/evaluation loop, shared by `Trainer.evaluate()` and `Trainer.predict()`.
+      Works both with or without labels.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _gather_and_numpify(tensors, name)
+
+      
+      Gather value of `tensors` (tensor or list/tuple of nested tensors) and convert them to numpy before
+      concatenating them to `gathered`
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _add_sm_patterns_to_gitignore() -> None
+
+      
+      Add SageMaker Checkpointing patterns to .gitignore file.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.rst.txt b/_sources/autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.rst.txt
new file mode 100644
index 000000000..9be0deb3d
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.rst.txt
@@ -0,0 +1,60 @@
+lmflow.pipeline.utils.rm_dataprocessor
+======================================
+
+.. py:module:: lmflow.pipeline.utils.rm_dataprocessor
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.rm_dataprocessor.logger
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: RewardDataCollatorWithPadding
+
+   .. py:attribute:: tokenizer
+      :type:  transformers.AutoTokenizer
+
+
+   .. py:attribute:: padding
+      :type:  Union[bool, str, transformers.utils.PaddingStrategy]
+      :value: True
+
+
+
+   .. py:attribute:: max_length
+      :type:  Optional[int]
+      :value: None
+
+
+
+   .. py:attribute:: pad_to_multiple_of
+      :type:  Optional[int]
+      :value: None
+
+
+
+   .. py:attribute:: return_tensors
+      :type:  str
+      :value: 'pt'
+
+
+
+   .. py:method:: __call__(features: List[Dict[str, Any]]) -> Dict[str, Any]
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/utils/rm_trainer/index.rst.txt b/_sources/autoapi/lmflow/pipeline/utils/rm_trainer/index.rst.txt
new file mode 100644
index 000000000..4f863d79f
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/utils/rm_trainer/index.rst.txt
@@ -0,0 +1,47 @@
+lmflow.pipeline.utils.rm_trainer
+================================
+
+.. py:module:: lmflow.pipeline.utils.rm_trainer
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.rm_trainer.RewardTrainer
+   lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.pipeline.utils.rm_trainer.compute_metrics
+   lmflow.pipeline.utils.rm_trainer.rm_loss
+
+
+Module Contents
+---------------
+
+.. py:function:: compute_metrics(eval_pred)
+
+.. py:function:: rm_loss(model, inputs, return_outputs=False)
+
+.. py:class:: RewardTrainer
+
+   Bases: :py:obj:`transformers.Trainer`
+
+
+   .. py:method:: compute_loss(model, inputs, return_outputs=False)
+
+
+.. py:class:: PeftRewardTrainer
+
+   Bases: :py:obj:`lmflow.pipeline.utils.peft_trainer.PeftTrainer`
+
+
+   .. py:method:: compute_loss(model, inputs, return_outputs=False)
+
+
diff --git a/_sources/autoapi/lmflow/pipeline/vllm_inferencer/index.rst.txt b/_sources/autoapi/lmflow/pipeline/vllm_inferencer/index.rst.txt
new file mode 100644
index 000000000..b26597cfc
--- /dev/null
+++ b/_sources/autoapi/lmflow/pipeline/vllm_inferencer/index.rst.txt
@@ -0,0 +1,268 @@
+lmflow.pipeline.vllm_inferencer
+===============================
+
+.. py:module:: lmflow.pipeline.vllm_inferencer
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.pipeline.vllm_inferencer.logger
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.pipeline.vllm_inferencer.InferencerWithOffloading
+   lmflow.pipeline.vllm_inferencer.VLLMInferencer
+   lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: InferencerWithOffloading(model_args: lmflow.args.ModelArguments, data_args: lmflow.args.DatasetArguments, inferencer_args: lmflow.args.InferencerArguments)
+
+   Bases: :py:obj:`lmflow.pipeline.base_pipeline.BasePipeline`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: model_args
+
+
+   .. py:attribute:: data_args
+
+
+   .. py:attribute:: inferencer_args
+
+
+   .. py:attribute:: eos_token_id
+
+
+   .. py:method:: inference()
+      :abstractmethod:
+
+
+
+   .. py:method:: save_inference_results()
+      :abstractmethod:
+
+
+
+   .. py:method:: load_inference_results()
+      :abstractmethod:
+
+
+
+.. py:class:: VLLMInferencer(model_args: lmflow.args.ModelArguments, data_args: lmflow.args.DatasetArguments, inferencer_args: lmflow.args.InferencerArguments)
+
+   Bases: :py:obj:`InferencerWithOffloading`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: sampling_params
+
+
+   .. py:method:: parse_to_sampling_params(inference_args: lmflow.args.InferencerArguments) -> vllm.SamplingParams
+
+
+   .. py:method:: inference(model: lmflow.models.hf_decoder_model.HFDecoderModel, dataset: lmflow.datasets.Dataset, enable_decode_inference_result: bool = True, release_gpu: bool = False, inference_args: Optional[lmflow.args.InferencerArguments] = None, enable_distributed_inference: bool = False, **kwargs) -> List[lmflow.utils.data_utils.VLLMInferenceResultWithInput]
+
+      
+      Perform inference using the provided model and dataset. Will save inference results if
+      `save_results` is set to True in `inferencer_args`.
+
+
+      :Parameters:
+
+          **model** : HFDecoderModel
+              LMFlow HFDecoderModel object
+
+          **dataset** : Dataset
+              LMFlow Dataset object
+
+          **apply_chat_template** : bool, optional
+              Whether to apply chat template to the input, by default True.
+
+          **enable_decode_inference_result** : bool, optional
+              Whether to decode after generation, by default False.
+
+          **release_gpu** : bool, optional
+              Whether to release gpu resources, by default False. 
+
+          **inference_args** : InferencerArguments, optional
+              by default None
+
+
+
+      :Returns:
+
+          List[VLLMInferenceResultWithInput]
+              Return a list of VLLMInferenceResultWithInput, where each
+              element contains the input prompt and the corresponding output.
+              
+              When `enable_decode_inference_result = True`, the output would be a list of strings,
+              contains sampling_params.n samples for the corresponding prompt.
+              
+              When `enable_decode_inference_result = False`, return a list of list of ints 
+              (token ids, no decoding after generation).
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _inference(model: lmflow.models.hf_decoder_model.HFDecoderModel, model_input: List[str], sampling_params: vllm.SamplingParams, release_gpu: bool = False) -> List[lmflow.utils.data_utils.VLLMInferenceResultWithInput]
+
+
+   .. py:method:: _distributed_inference(model: lmflow.models.hf_decoder_model.HFDecoderModel, model_input: ray.data.Dataset, sampling_params: vllm.SamplingParams, num_instances: int, batch_size: int = 4, release_gpu: bool = False) -> List[lmflow.utils.data_utils.VLLMInferenceResultWithInput]
+
+
+   .. py:method:: save_inference_results(outputs: Union[List[List[str]], List[List[List[int]]]], save_file_path: str)
+
+
+   .. py:method:: load_inference_results(results_path: str) -> Union[List[List[str]], List[List[List[int]]]]
+
+
+.. py:class:: MemorySafeVLLMInferencer(model_args: lmflow.args.ModelArguments, data_args: lmflow.args.DatasetArguments, inferencer_args: lmflow.args.InferencerArguments)
+
+   Bases: :py:obj:`VLLMInferencer`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: inferencer_file_path
+
+
+   .. py:method:: inference() -> List[lmflow.utils.data_utils.VLLMInferenceResultWithInput]
+
+      
+      Perform inference using the provided model and dataset. Will save inference results if
+      `save_results` is set to True in `inferencer_args`.
+
+
+      :Parameters:
+
+          **model** : HFDecoderModel
+              LMFlow HFDecoderModel object
+
+          **dataset** : Dataset
+              LMFlow Dataset object
+
+          **apply_chat_template** : bool, optional
+              Whether to apply chat template to the input, by default True.
+
+          **enable_decode_inference_result** : bool, optional
+              Whether to decode after generation, by default False.
+
+          **release_gpu** : bool, optional
+              Whether to release gpu resources, by default False. 
+
+          **inference_args** : InferencerArguments, optional
+              by default None
+
+
+
+      :Returns:
+
+          List[VLLMInferenceResultWithInput]
+              Return a list of VLLMInferenceResultWithInput, where each
+              element contains the input prompt and the corresponding output.
+              
+              When `enable_decode_inference_result = True`, the output would be a list of strings,
+              contains sampling_params.n samples for the corresponding prompt.
+              
+              When `enable_decode_inference_result = False`, return a list of list of ints 
+              (token ids, no decoding after generation).
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
diff --git a/_sources/autoapi/lmflow/tokenization/hf_decoder_model/index.rst.txt b/_sources/autoapi/lmflow/tokenization/hf_decoder_model/index.rst.txt
new file mode 100644
index 000000000..9d3e7767e
--- /dev/null
+++ b/_sources/autoapi/lmflow/tokenization/hf_decoder_model/index.rst.txt
@@ -0,0 +1,80 @@
+lmflow.tokenization.hf_decoder_model
+====================================
+
+.. py:module:: lmflow.tokenization.hf_decoder_model
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.tokenization.hf_decoder_model.logger
+   lmflow.tokenization.hf_decoder_model.tok_logger
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.tokenization.hf_decoder_model.blocking
+   lmflow.tokenization.hf_decoder_model.tokenize_function
+   lmflow.tokenization.hf_decoder_model.conversation_tokenize_function
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:data:: tok_logger
+
+.. py:function:: blocking(token_dict: Dict, block_size: int, model_max_length: int, pad_token_id: int, padding_side: str, truncation_side: str = 'right') -> Dict
+
+.. py:function:: tokenize_function(examples, data_args: lmflow.args.DatasetArguments, tokenizer: Union[transformers.PreTrainedTokenizer, transformers.PreTrainedTokenizerFast], column_names, label_columns, tokenized_column_order, add_special_tokens, use_truncation) -> Dict
+
+   
+   Handels text_only and text2text datasets tokenization
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:function:: conversation_tokenize_function(examples, data_args: lmflow.args.DatasetArguments, tokenizer: Union[transformers.PreTrainedTokenizer, transformers.PreTrainedTokenizerFast], column_names, conversation_template: lmflow.utils.conversation_template.ConversationTemplate) -> Dict
+
+   
+   Handels conversation datasets tokenization
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
diff --git a/_sources/autoapi/lmflow/tokenization/hf_text_regression_model/index.rst.txt b/_sources/autoapi/lmflow/tokenization/hf_text_regression_model/index.rst.txt
new file mode 100644
index 000000000..142f4b240
--- /dev/null
+++ b/_sources/autoapi/lmflow/tokenization/hf_text_regression_model/index.rst.txt
@@ -0,0 +1,114 @@
+lmflow.tokenization.hf_text_regression_model
+============================================
+
+.. py:module:: lmflow.tokenization.hf_text_regression_model
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.tokenization.hf_text_regression_model.logger
+   lmflow.tokenization.hf_text_regression_model.tok_logger
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.tokenization.hf_text_regression_model.blocking_paired
+   lmflow.tokenization.hf_text_regression_model.blocking
+   lmflow.tokenization.hf_text_regression_model.blocking_text_to_textlist
+   lmflow.tokenization.hf_text_regression_model.paired_conversation_tokenize_function
+   lmflow.tokenization.hf_text_regression_model.conversation_tokenize_function
+   lmflow.tokenization.hf_text_regression_model.tokenize_function
+   lmflow.tokenization.hf_text_regression_model.text_to_textlist_tokenize_function
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:data:: tok_logger
+
+.. py:function:: blocking_paired(token_dict: Dict, column_names: List, block_size: int, model_max_length: int, pad_token_id: int, padding_side: str, truncation_side: str = 'right') -> Dict
+
+.. py:function:: blocking(token_dict: Dict, block_size: int, model_max_length: int, pad_token_id: int, padding_side: str, truncation_side: str = 'right') -> Dict
+
+.. py:function:: blocking_text_to_textlist(token_dict: Dict, block_size: int, model_max_length: int, pad_token_id: int, padding_side: str, truncation_side: str = 'right') -> Dict
+
+.. py:function:: paired_conversation_tokenize_function(examples, data_args: lmflow.args.DatasetArguments, tokenizer: Union[transformers.PreTrainedTokenizer, transformers.PreTrainedTokenizerFast], column_names, conversation_template: lmflow.utils.conversation_template.ConversationTemplate) -> Dict
+
+.. py:function:: conversation_tokenize_function(examples, data_args: lmflow.args.DatasetArguments, tokenizer: Union[transformers.PreTrainedTokenizer, transformers.PreTrainedTokenizerFast], column_names, conversation_template: lmflow.utils.conversation_template.ConversationTemplate) -> Dict
+
+   
+   Handels conversation datasets tokenization
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:function:: tokenize_function(examples, data_args: lmflow.args.DatasetArguments, tokenizer: Union[transformers.PreTrainedTokenizer, transformers.PreTrainedTokenizerFast], column_names, label_columns, tokenized_column_order, add_special_tokens, use_truncation) -> Dict
+
+   
+   Handels text_only and text2text datasets tokenization
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:function:: text_to_textlist_tokenize_function(examples, data_args: lmflow.args.DatasetArguments, tokenizer: Union[transformers.PreTrainedTokenizer, transformers.PreTrainedTokenizerFast], column_names, add_special_tokens, use_truncation) -> Dict
+
+   
+   For rm inference, and don't need attn mask and labels.
+   NOTE: input_ids here refers to the tokenized input_ids of the input **and** output
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
diff --git a/_sources/autoapi/lmflow/tokenization/index.rst.txt b/_sources/autoapi/lmflow/tokenization/index.rst.txt
new file mode 100644
index 000000000..ee3c2d670
--- /dev/null
+++ b/_sources/autoapi/lmflow/tokenization/index.rst.txt
@@ -0,0 +1,16 @@
+lmflow.tokenization
+===================
+
+.. py:module:: lmflow.tokenization
+
+
+Submodules
+----------
+
+.. toctree::
+   :maxdepth: 1
+
+   /autoapi/lmflow/tokenization/hf_decoder_model/index
+   /autoapi/lmflow/tokenization/hf_text_regression_model/index
+
+
diff --git a/_sources/autoapi/lmflow/utils/common/index.rst.txt b/_sources/autoapi/lmflow/utils/common/index.rst.txt
new file mode 100644
index 000000000..a4ebc0357
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/common/index.rst.txt
@@ -0,0 +1,179 @@
+lmflow.utils.common
+===================
+
+.. py:module:: lmflow.utils.common
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.common.logger
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.utils.common.make_shell_args_from_dataclass
+   lmflow.utils.common.create_copied_dataclass
+   lmflow.utils.common.remove_dataclass_attr_prefix
+   lmflow.utils.common.add_dataclass_attr_prefix
+   lmflow.utils.common.print_banner
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:function:: make_shell_args_from_dataclass(dataclass_objects: List, format: str = 'subprocess', skip_default: bool = True, ignored_args_list: Optional[List[str]] = None) -> Union[str, List[str]]
+
+   
+   Return a string or a list of strings that can be used as shell arguments.
+
+
+   :Parameters:
+
+       **dataclass_objects** : List
+           A list of dataclass objects.
+
+       **format** : str, optional
+           Return format, can be "shell" or "subprocess", by default "subprocess".
+
+       **skip_default** : bool, optional
+           Whether to skip attributes with default values, by default True. 
+
+
+
+   :Returns:
+
+       Union[str, List[str]]
+           ..
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:function:: create_copied_dataclass(original_dataclass, field_prefix: str, class_prefix: str, new_default: Dict = None)
+
+   
+   Create a copied dataclass with new field names and default values.
+
+
+   :Parameters:
+
+       **original_dataclass** : dataclass
+           ..
+
+       **field_prefix** : str
+           The prefix to add to the **field** names of the copied dataclass.
+
+       **class_prefix** : str
+           The prefix to add to the **class** name of the copied dataclass.
+
+       **new_default** : Dict, optional
+           The new default values for the copied dataclass. When None, the 
+           default values of the original dataclass are used.
+
+
+
+   :Returns:
+
+       dataclass
+           ..
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:function:: remove_dataclass_attr_prefix(data_instance, prefix: str) -> Dict
+
+   
+   Remove the prefix from the attribute names of a dataclass instance.
+
+
+   :Parameters:
+
+       **data_instance** : dataclass
+           ..
+
+       **prefix** : str
+           The prefix to remove from the attribute names of the dataclass instance.
+
+
+
+   :Returns:
+
+       Dict
+           ..
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:function:: add_dataclass_attr_prefix(data_instance, prefix: str) -> Dict
+
+   
+   Add the prefix to the attribute names of a dataclass instance.
+
+
+   :Parameters:
+
+       **data_instance** : dataclass
+           ..
+
+       **prefix** : str
+           The prefix to add to the attribute names of the dataclass instance.
+
+
+
+   :Returns:
+
+       Dict
+           ..
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:function:: print_banner(message: str)
+
diff --git a/_sources/autoapi/lmflow/utils/constants/index.rst.txt b/_sources/autoapi/lmflow/utils/constants/index.rst.txt
new file mode 100644
index 000000000..828b27381
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/constants/index.rst.txt
@@ -0,0 +1,160 @@
+lmflow.utils.constants
+======================
+
+.. py:module:: lmflow.utils.constants
+
+.. autoapi-nested-parse::
+
+   Commonly used constants.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.constants.TEXT_ONLY_DATASET_DESCRIPTION
+   lmflow.utils.constants.TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION
+   lmflow.utils.constants.PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION
+   lmflow.utils.constants.TEXT_ONLY_DATASET_DETAILS
+   lmflow.utils.constants.TEXT2TEXT_DATASET_DESCRIPTION
+   lmflow.utils.constants.CONVERSATION_DATASET_DESCRIPTION
+   lmflow.utils.constants.PAIRED_CONVERSATION_DATASET_DESCRIPTION
+   lmflow.utils.constants.TEXT_TO_TEXTLIST_DATASET_DESCRIPTION
+   lmflow.utils.constants.TEXT2TEXT_DATASET_DETAILS
+   lmflow.utils.constants.FLOAT_ONLY_DATASET_DESCRIPTION
+   lmflow.utils.constants.TEXT_ONLY_DATASET_LONG_DESCRITION
+   lmflow.utils.constants.TEXT2TEXT_DATASET_LONG_DESCRITION
+   lmflow.utils.constants.DATASET_DESCRIPTION_MAP
+   lmflow.utils.constants.INSTANCE_FIELDS_MAP
+   lmflow.utils.constants.CONVERSATION_ROLE_NAMES
+   lmflow.utils.constants.CONTROLLER_HEART_BEAT_EXPIRATION
+   lmflow.utils.constants.WORKER_HEART_BEAT_INTERVAL
+   lmflow.utils.constants.LOGDIR
+   lmflow.utils.constants.IGNORE_INDEX
+   lmflow.utils.constants.IMAGE_TOKEN_INDEX
+   lmflow.utils.constants.DEFAULT_IMAGE_TOKEN
+   lmflow.utils.constants.DEFAULT_IMAGE_PATCH_TOKEN
+   lmflow.utils.constants.DEFAULT_IM_START_TOKEN
+   lmflow.utils.constants.DEFAULT_IM_END_TOKEN
+   lmflow.utils.constants.LMFLOW_LORA_TARGET_MODULES_MAPPING
+   lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG
+   lmflow.utils.constants.RETURN_CODE_ERROR_BUFFER
+   lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE
+   lmflow.utils.constants.MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE
+
+
+Module Contents
+---------------
+
+.. py:data:: TEXT_ONLY_DATASET_DESCRIPTION
+   :value: ''
+
+
+.. py:data:: TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION
+   :value: ''
+
+
+.. py:data:: PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION
+   :value: ''
+
+
+.. py:data:: TEXT_ONLY_DATASET_DETAILS
+   :value: ''
+
+
+.. py:data:: TEXT2TEXT_DATASET_DESCRIPTION
+   :value: ''
+
+
+.. py:data:: CONVERSATION_DATASET_DESCRIPTION
+   :value: ''
+
+
+.. py:data:: PAIRED_CONVERSATION_DATASET_DESCRIPTION
+   :value: ''
+
+
+.. py:data:: TEXT_TO_TEXTLIST_DATASET_DESCRIPTION
+   :value: ''
+
+
+.. py:data:: TEXT2TEXT_DATASET_DETAILS
+   :value: ''
+
+
+.. py:data:: FLOAT_ONLY_DATASET_DESCRIPTION
+   :value: ''
+
+
+.. py:data:: TEXT_ONLY_DATASET_LONG_DESCRITION
+   :value: ''
+
+
+.. py:data:: TEXT2TEXT_DATASET_LONG_DESCRITION
+   :value: ''
+
+
+.. py:data:: DATASET_DESCRIPTION_MAP
+
+.. py:data:: INSTANCE_FIELDS_MAP
+
+.. py:data:: CONVERSATION_ROLE_NAMES
+
+.. py:data:: CONTROLLER_HEART_BEAT_EXPIRATION
+   :value: 30
+
+
+.. py:data:: WORKER_HEART_BEAT_INTERVAL
+   :value: 15
+
+
+.. py:data:: LOGDIR
+   :value: '.'
+
+
+.. py:data:: IGNORE_INDEX
+   :value: -100
+
+
+.. py:data:: IMAGE_TOKEN_INDEX
+   :value: -200
+
+
+.. py:data:: DEFAULT_IMAGE_TOKEN
+   :value: '<image>'
+
+
+.. py:data:: DEFAULT_IMAGE_PATCH_TOKEN
+   :value: '<im_patch>'
+
+
+.. py:data:: DEFAULT_IM_START_TOKEN
+   :value: '<im_start>'
+
+
+.. py:data:: DEFAULT_IM_END_TOKEN
+   :value: '<im_end>'
+
+
+.. py:data:: LMFLOW_LORA_TARGET_MODULES_MAPPING
+
+.. py:data:: MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG
+   :value: 'MEMORY_SAFE_VLLM_INFERENCE_DONE'
+
+
+.. py:data:: RETURN_CODE_ERROR_BUFFER
+   :value: [134]
+
+
+.. py:data:: MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE
+   :value: ['OMP_NUM_THREADS', 'LOCAL_RANK', 'RANK', 'GROUP_RANK', 'ROLE_RANK', 'ROLE_NAME',...
+
+
+.. py:data:: MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE
+   :value: ['OMP_NUM_THREADS', 'LOCAL_RANK', 'RANK', 'GROUP_RANK', 'ROLE_RANK', 'ROLE_NAME',...
+
+
diff --git a/_sources/autoapi/lmflow/utils/conversation_template/base/index.rst.txt b/_sources/autoapi/lmflow/utils/conversation_template/base/index.rst.txt
new file mode 100644
index 000000000..ffaf718e9
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/conversation_template/base/index.rst.txt
@@ -0,0 +1,580 @@
+lmflow.utils.conversation_template.base
+=======================================
+
+.. py:module:: lmflow.utils.conversation_template.base
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.base.logger
+   lmflow.utils.conversation_template.base.EMPTY_TEMPLATE
+   lmflow.utils.conversation_template.base.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.base.TemplateComponent
+   lmflow.utils.conversation_template.base.Formatter
+   lmflow.utils.conversation_template.base.EmptyFormatter
+   lmflow.utils.conversation_template.base.StringFormatter
+   lmflow.utils.conversation_template.base.ListFormatter
+   lmflow.utils.conversation_template.base.ConversationTemplate
+   lmflow.utils.conversation_template.base.ConversationTemplateForTool
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: TemplateComponent
+
+   
+   The minimal unit of a template, which can be a token, a string, or a list of tools.
+
+
+   :Parameters:
+
+       **type** : Literal['token', 'token_id', 'string', 'tools']
+           - Type of the component.  
+           
+           - When the component is a token or a string, the content should be `string`. 
+           The difference between the two is that token will be converted to token ids 
+           by the tokenizer.convert_tokens_to_ids() method, while string will be directly 
+           encoded by the tokenizer.encode() method. Specially, since the bos token and eos
+           token are frequently used across different templates, we provide the convenience
+           to use `'bos_token'` and `'eos_token'` to represent the actual bos and eos tokens when
+           `type` of the `TemplateComponent` is `token`. For example:  
+           
+           ```python
+           TemplateComponent(type='token', content='bos_token')
+           ```
+           
+           After encoding, the content will be replaced by the actual token id of the bos token.
+           Please do remember that if you set the `type` to `string`, the tokenizer will try to 
+           encode the string 'bos_token' instead of providing the actual bos token.
+           
+           - When the component is token_id, the content should be `int` or `List[int]`, and 
+           will be directly appended to the encoded token ids.
+           
+           - Tools are not supported yet.
+
+       **content** : Union[str, int, List[str], List[int]]
+           Content of the component.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: type
+      :type:  Literal['token', 'token_id', 'string', 'tools']
+
+
+   .. py:attribute:: content
+      :type:  Union[str, int, List[str], List[int]]
+
+
+   .. py:attribute:: mask
+      :type:  Optional[bool]
+      :value: True
+
+
+
+   .. py:method:: __post_init__()
+
+
+   .. py:method:: __repr__() -> str
+
+
+   .. py:method:: __str__() -> str
+
+
+.. py:class:: Formatter
+
+   Bases: :py:obj:`abc.ABC`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: template
+      :type:  List[TemplateComponent]
+      :value: []
+
+
+
+   .. py:method:: format(**kwargs) -> List[TemplateComponent]
+      :abstractmethod:
+
+
+
+   .. py:method:: has_placeholder()
+
+
+.. py:class:: EmptyFormatter
+
+   Bases: :py:obj:`Formatter`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: __post_init__()
+
+
+   .. py:method:: format(**kwargs) -> list
+
+      
+      Empty formatter for when no formatting is needed.
+      This is useful when user has already applied formatting to the dataset.
+
+
+
+
+
+      :Returns:
+
+          list
+              Original template.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+.. py:class:: StringFormatter
+
+   Bases: :py:obj:`Formatter`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: __post_init__()
+
+
+   .. py:method:: format(**kwargs) -> list
+
+      
+      Format the string components with the provided keyword arguments. 
+      Mostly used for formatting system prompt, user and assistant messages.
+
+
+      :Parameters:
+
+          **\*\*kwargs** : dict
+              Keyword arguments containing values to replace in the template components.
+
+
+
+      :Returns:
+
+          list
+              Formatted template.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+.. py:class:: ListFormatter
+
+   Bases: :py:obj:`Formatter`
+
+
+   
+   Helper class that provides a standard way to create an ABC using
+   inheritance.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:method:: format(**kwargs) -> list
+
+
+.. py:class:: ConversationTemplate
+
+   .. py:attribute:: user_formatter
+      :type:  Formatter
+
+
+   .. py:attribute:: assistant_formatter
+      :type:  Formatter
+
+
+   .. py:attribute:: function_formatter
+      :type:  Optional[Formatter]
+      :value: None
+
+
+
+   .. py:attribute:: observation_formatter
+      :type:  Optional[Formatter]
+      :value: None
+
+
+
+   .. py:attribute:: system_formatter
+      :type:  Optional[Formatter]
+      :value: None
+
+
+
+   .. py:attribute:: force_system
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: tools_formatter
+      :type:  Optional[Formatter]
+      :value: None
+
+
+
+   .. py:attribute:: separator
+      :type:  Optional[TemplateComponent]
+      :value: None
+
+
+
+   .. py:attribute:: remove_last_sep
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: special_starter
+      :type:  Optional[TemplateComponent]
+      :value: None
+
+
+
+   .. py:attribute:: special_stopper
+      :type:  Optional[TemplateComponent]
+      :value: None
+
+
+
+   .. py:attribute:: template_name
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:method:: __post_init__()
+
+
+   .. py:method:: encode_conversation(tokenizer: transformers.PreTrainedTokenizer, messages: List[Dict[str, str]], system: Optional[str] = None, tools: Optional[List[str]] = None, **kwargs) -> Sequence[Tuple[List[int], List[int]]]
+
+      
+      Messages here should be guaranteed to be in pairs, with the first message being the user message and the second message being the system message.
+      Data example: 
+      ```json
+      {
+          "conversation_id": 2,
+          "system": "sysinfo1",
+          "tools": ["tool_1_desc"],
+          "messages": [
+              {
+                  "role": "user",
+                  "content": "hi"
+              },
+              {
+                  "role": "assistant",
+                  "content": "Hello!"
+              }
+          ]
+      }
+      ```
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _encode(tokenizer: transformers.PreTrainedTokenizer, messages: List[Dict[str, str]], system: Optional[str] = None, tools: Optional[str] = None, **kwargs) -> Sequence[Tuple[List[int], List[int]]]
+
+
+   .. py:method:: _encode_template(template: List[TemplateComponent], tokenizer: transformers.PreTrainedTokenizer, **kwargs) -> List[int]
+
+      
+      Encode template components into token ids.
+
+
+      :Parameters:
+
+          **template** : List[TemplateComponent]
+              Formatted template components.
+
+          **tokenizer** : PreTrainedTokenizer
+              Tokenizer to convert tokens into token ids.
+
+
+
+      :Returns:
+
+          List[int]
+              Encoded token ids.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: post_process_pairs(encoded_pairs, tokenizer)
+
+
+   .. py:method:: remove_last_separator(encoded_pairs: Sequence[Tuple[List[int], List[int]]], tokenizer: transformers.PreTrainedTokenizer) -> Sequence[Tuple[List[int], List[int]]]
+
+
+   .. py:method:: add_special_starter(encoded_pairs: Sequence[Tuple[List[int], List[int]]], tokenizer: transformers.PreTrainedTokenizer) -> Sequence[Tuple[List[int], List[int]]]
+
+
+   .. py:method:: add_special_stopper(encoded_pairs: Sequence[Tuple[List[int], List[int]]], tokenizer: transformers.PreTrainedTokenizer) -> Sequence[Tuple[List[int], List[int]]]
+
+
+   .. py:method:: _ensure_id_list(obj: Union[int, List[int]]) -> List[int]
+
+      
+      Make sure the object is a list of integers. Useful for handling token ids.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+.. py:class:: ConversationTemplateForTool
+
+   Bases: :py:obj:`ConversationTemplate`
+
+
+   .. py:method:: encode_conversation(tokenizer: transformers.PreTrainedTokenizer, messages: List[Dict[str, str]], system: Optional[str] = None, tools: Optional[List[str]] = None, **kwargs) -> Sequence[Tuple[List[int], List[int]]]
+
+      
+      Messages here should be guaranteed to be in pairs, with the first message being the user message and the second message being the system message.
+      Data example: 
+      ```json
+      {
+          "conversation_id": 2,
+          "system": "sysinfo1",
+          "tools": ["tool_1_desc"],
+          "messages": [
+              {
+                  "role": "user",
+                  "content": "hi"
+              },
+              {
+                  "role": "assistant",
+                  "content": "Hello!"
+              }
+          ]
+      }
+      ```
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _encode(tokenizer: transformers.PreTrainedTokenizer, messages: List[Dict[str, str]], system: Optional[str] = None, tools: Optional[str] = None, **kwargs) -> Sequence[Tuple[List[int], List[int]]]
+
+
+   .. py:method:: _encode_template(template: List[TemplateComponent], tokenizer: transformers.PreTrainedTokenizer, **kwargs) -> List[int]
+
+      
+      Encode template components into token ids.
+
+
+      :Parameters:
+
+          **template** : List[TemplateComponent]
+              Formatted template components.
+
+          **tokenizer** : PreTrainedTokenizer
+              Tokenizer to convert tokens into token ids.
+
+
+
+      :Returns:
+
+          List[int]
+              Encoded token ids.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+.. py:data:: EMPTY_TEMPLATE
+
+.. py:data:: EMPTY_NO_SPECIAL_TOKENS_TEMPLATE
+
diff --git a/_sources/autoapi/lmflow/utils/conversation_template/chatglm/index.rst.txt b/_sources/autoapi/lmflow/utils/conversation_template/chatglm/index.rst.txt
new file mode 100644
index 000000000..fa17b5407
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/conversation_template/chatglm/index.rst.txt
@@ -0,0 +1,19 @@
+lmflow.utils.conversation_template.chatglm
+==========================================
+
+.. py:module:: lmflow.utils.conversation_template.chatglm
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.chatglm.CHATGLM3_TEMPLATE
+
+
+Module Contents
+---------------
+
+.. py:data:: CHATGLM3_TEMPLATE
+
diff --git a/_sources/autoapi/lmflow/utils/conversation_template/chatml/index.rst.txt b/_sources/autoapi/lmflow/utils/conversation_template/chatml/index.rst.txt
new file mode 100644
index 000000000..b414e64a9
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/conversation_template/chatml/index.rst.txt
@@ -0,0 +1,19 @@
+lmflow.utils.conversation_template.chatml
+=========================================
+
+.. py:module:: lmflow.utils.conversation_template.chatml
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.chatml.CHATML_TEMPLATE
+
+
+Module Contents
+---------------
+
+.. py:data:: CHATML_TEMPLATE
+
diff --git a/_sources/autoapi/lmflow/utils/conversation_template/deepseek/index.rst.txt b/_sources/autoapi/lmflow/utils/conversation_template/deepseek/index.rst.txt
new file mode 100644
index 000000000..48ea50616
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/conversation_template/deepseek/index.rst.txt
@@ -0,0 +1,19 @@
+lmflow.utils.conversation_template.deepseek
+===========================================
+
+.. py:module:: lmflow.utils.conversation_template.deepseek
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.deepseek.DEEPSEEK_TEMPLATE
+
+
+Module Contents
+---------------
+
+.. py:data:: DEEPSEEK_TEMPLATE
+
diff --git a/_sources/autoapi/lmflow/utils/conversation_template/gemma/index.rst.txt b/_sources/autoapi/lmflow/utils/conversation_template/gemma/index.rst.txt
new file mode 100644
index 000000000..e7d9381e0
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/conversation_template/gemma/index.rst.txt
@@ -0,0 +1,38 @@
+lmflow.utils.conversation_template.gemma
+========================================
+
+.. py:module:: lmflow.utils.conversation_template.gemma
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.gemma.logger
+   lmflow.utils.conversation_template.gemma.GEMMA_TEMPLATE
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.gemma.GemmaConversationTemplate
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: GemmaConversationTemplate
+
+   Bases: :py:obj:`lmflow.utils.conversation_template.base.ConversationTemplate`
+
+
+   .. py:method:: encode_conversation(*args, **kwargs)
+
+
+.. py:data:: GEMMA_TEMPLATE
+
diff --git a/_sources/autoapi/lmflow/utils/conversation_template/hymba/index.rst.txt b/_sources/autoapi/lmflow/utils/conversation_template/hymba/index.rst.txt
new file mode 100644
index 000000000..145f1afe7
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/conversation_template/hymba/index.rst.txt
@@ -0,0 +1,74 @@
+lmflow.utils.conversation_template.hymba
+========================================
+
+.. py:module:: lmflow.utils.conversation_template.hymba
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.hymba.HYMBA_TEMPLATE
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.hymba.HymbaConversationTemplate
+
+
+Module Contents
+---------------
+
+.. py:class:: HymbaConversationTemplate
+
+   Bases: :py:obj:`lmflow.utils.conversation_template.base.ConversationTemplateForTool`
+
+
+   .. py:method:: encode_conversation(tokenizer: transformers.PreTrainedTokenizer, messages: List[Dict[str, str]], system: Optional[str] = None, tools: Optional[List[str]] = None, **kwargs) -> Sequence[Tuple[List[int], List[int]]]
+
+      
+      Messages here should be guaranteed to be in pairs, with the first message being the user message and the second message being the system message.
+      Data example: 
+      ```json
+      {
+          "conversation_id": 2,
+          "system": "sysinfo1",
+          "tools": ["tool_1_desc"],
+          "messages": [
+              {
+                  "role": "user",
+                  "content": "hi"
+              },
+              {
+                  "role": "assistant",
+                  "content": "Hello!"
+              }
+          ]
+      }
+      ```
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+.. py:data:: HYMBA_TEMPLATE
+
diff --git a/_sources/autoapi/lmflow/utils/conversation_template/index.rst.txt b/_sources/autoapi/lmflow/utils/conversation_template/index.rst.txt
new file mode 100644
index 000000000..58c07c520
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/conversation_template/index.rst.txt
@@ -0,0 +1,370 @@
+lmflow.utils.conversation_template
+==================================
+
+.. py:module:: lmflow.utils.conversation_template
+
+
+Submodules
+----------
+
+.. toctree::
+   :maxdepth: 1
+
+   /autoapi/lmflow/utils/conversation_template/base/index
+   /autoapi/lmflow/utils/conversation_template/chatglm/index
+   /autoapi/lmflow/utils/conversation_template/chatml/index
+   /autoapi/lmflow/utils/conversation_template/deepseek/index
+   /autoapi/lmflow/utils/conversation_template/gemma/index
+   /autoapi/lmflow/utils/conversation_template/hymba/index
+   /autoapi/lmflow/utils/conversation_template/internlm/index
+   /autoapi/lmflow/utils/conversation_template/llama/index
+   /autoapi/lmflow/utils/conversation_template/phi/index
+   /autoapi/lmflow/utils/conversation_template/qwen/index
+   /autoapi/lmflow/utils/conversation_template/yi/index
+   /autoapi/lmflow/utils/conversation_template/zephyr/index
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.EMPTY_TEMPLATE
+   lmflow.utils.conversation_template.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE
+   lmflow.utils.conversation_template.CHATGLM3_TEMPLATE
+   lmflow.utils.conversation_template.CHATML_TEMPLATE
+   lmflow.utils.conversation_template.DEEPSEEK_TEMPLATE
+   lmflow.utils.conversation_template.GEMMA_TEMPLATE
+   lmflow.utils.conversation_template.HYMBA_TEMPLATE
+   lmflow.utils.conversation_template.INTERNLM2_TEMPLATE
+   lmflow.utils.conversation_template.LLAMA2_TEMPLATE
+   lmflow.utils.conversation_template.LLAMA3_TEMPLATE
+   lmflow.utils.conversation_template.LLAMA3_TEMPLATE_FOR_TOOL
+   lmflow.utils.conversation_template.PHI3_TEMPLATE
+   lmflow.utils.conversation_template.QWEN2_TEMPLATE
+   lmflow.utils.conversation_template.QWEN2_TEMPLATE_FOR_TOOL
+   lmflow.utils.conversation_template.YI1_5_TEMPLATE
+   lmflow.utils.conversation_template.ZEPHYR_TEMPLATE
+   lmflow.utils.conversation_template.PRESET_TEMPLATES
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.ConversationTemplate
+   lmflow.utils.conversation_template.ConversationTemplateForTool
+
+
+Package Contents
+----------------
+
+.. py:data:: EMPTY_TEMPLATE
+
+.. py:data:: EMPTY_NO_SPECIAL_TOKENS_TEMPLATE
+
+.. py:class:: ConversationTemplate
+
+   .. py:attribute:: user_formatter
+      :type:  Formatter
+
+
+   .. py:attribute:: assistant_formatter
+      :type:  Formatter
+
+
+   .. py:attribute:: function_formatter
+      :type:  Optional[Formatter]
+      :value: None
+
+
+
+   .. py:attribute:: observation_formatter
+      :type:  Optional[Formatter]
+      :value: None
+
+
+
+   .. py:attribute:: system_formatter
+      :type:  Optional[Formatter]
+      :value: None
+
+
+
+   .. py:attribute:: force_system
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: tools_formatter
+      :type:  Optional[Formatter]
+      :value: None
+
+
+
+   .. py:attribute:: separator
+      :type:  Optional[TemplateComponent]
+      :value: None
+
+
+
+   .. py:attribute:: remove_last_sep
+      :type:  bool
+      :value: False
+
+
+
+   .. py:attribute:: special_starter
+      :type:  Optional[TemplateComponent]
+      :value: None
+
+
+
+   .. py:attribute:: special_stopper
+      :type:  Optional[TemplateComponent]
+      :value: None
+
+
+
+   .. py:attribute:: template_name
+      :type:  Optional[str]
+      :value: None
+
+
+
+   .. py:method:: __post_init__()
+
+
+   .. py:method:: encode_conversation(tokenizer: transformers.PreTrainedTokenizer, messages: List[Dict[str, str]], system: Optional[str] = None, tools: Optional[List[str]] = None, **kwargs) -> Sequence[Tuple[List[int], List[int]]]
+
+      
+      Messages here should be guaranteed to be in pairs, with the first message being the user message and the second message being the system message.
+      Data example: 
+      ```json
+      {
+          "conversation_id": 2,
+          "system": "sysinfo1",
+          "tools": ["tool_1_desc"],
+          "messages": [
+              {
+                  "role": "user",
+                  "content": "hi"
+              },
+              {
+                  "role": "assistant",
+                  "content": "Hello!"
+              }
+          ]
+      }
+      ```
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _encode(tokenizer: transformers.PreTrainedTokenizer, messages: List[Dict[str, str]], system: Optional[str] = None, tools: Optional[str] = None, **kwargs) -> Sequence[Tuple[List[int], List[int]]]
+
+
+   .. py:method:: _encode_template(template: List[TemplateComponent], tokenizer: transformers.PreTrainedTokenizer, **kwargs) -> List[int]
+
+      
+      Encode template components into token ids.
+
+
+      :Parameters:
+
+          **template** : List[TemplateComponent]
+              Formatted template components.
+
+          **tokenizer** : PreTrainedTokenizer
+              Tokenizer to convert tokens into token ids.
+
+
+
+      :Returns:
+
+          List[int]
+              Encoded token ids.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: post_process_pairs(encoded_pairs, tokenizer)
+
+
+   .. py:method:: remove_last_separator(encoded_pairs: Sequence[Tuple[List[int], List[int]]], tokenizer: transformers.PreTrainedTokenizer) -> Sequence[Tuple[List[int], List[int]]]
+
+
+   .. py:method:: add_special_starter(encoded_pairs: Sequence[Tuple[List[int], List[int]]], tokenizer: transformers.PreTrainedTokenizer) -> Sequence[Tuple[List[int], List[int]]]
+
+
+   .. py:method:: add_special_stopper(encoded_pairs: Sequence[Tuple[List[int], List[int]]], tokenizer: transformers.PreTrainedTokenizer) -> Sequence[Tuple[List[int], List[int]]]
+
+
+   .. py:method:: _ensure_id_list(obj: Union[int, List[int]]) -> List[int]
+
+      
+      Make sure the object is a list of integers. Useful for handling token ids.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+.. py:class:: ConversationTemplateForTool
+
+   Bases: :py:obj:`ConversationTemplate`
+
+
+   .. py:method:: encode_conversation(tokenizer: transformers.PreTrainedTokenizer, messages: List[Dict[str, str]], system: Optional[str] = None, tools: Optional[List[str]] = None, **kwargs) -> Sequence[Tuple[List[int], List[int]]]
+
+      
+      Messages here should be guaranteed to be in pairs, with the first message being the user message and the second message being the system message.
+      Data example: 
+      ```json
+      {
+          "conversation_id": 2,
+          "system": "sysinfo1",
+          "tools": ["tool_1_desc"],
+          "messages": [
+              {
+                  "role": "user",
+                  "content": "hi"
+              },
+              {
+                  "role": "assistant",
+                  "content": "Hello!"
+              }
+          ]
+      }
+      ```
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: _encode(tokenizer: transformers.PreTrainedTokenizer, messages: List[Dict[str, str]], system: Optional[str] = None, tools: Optional[str] = None, **kwargs) -> Sequence[Tuple[List[int], List[int]]]
+
+
+   .. py:method:: _encode_template(template: List[TemplateComponent], tokenizer: transformers.PreTrainedTokenizer, **kwargs) -> List[int]
+
+      
+      Encode template components into token ids.
+
+
+      :Parameters:
+
+          **template** : List[TemplateComponent]
+              Formatted template components.
+
+          **tokenizer** : PreTrainedTokenizer
+              Tokenizer to convert tokens into token ids.
+
+
+
+      :Returns:
+
+          List[int]
+              Encoded token ids.
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+.. py:data:: CHATGLM3_TEMPLATE
+
+.. py:data:: CHATML_TEMPLATE
+
+.. py:data:: DEEPSEEK_TEMPLATE
+
+.. py:data:: GEMMA_TEMPLATE
+
+.. py:data:: HYMBA_TEMPLATE
+
+.. py:data:: INTERNLM2_TEMPLATE
+
+.. py:data:: LLAMA2_TEMPLATE
+
+.. py:data:: LLAMA3_TEMPLATE
+
+.. py:data:: LLAMA3_TEMPLATE_FOR_TOOL
+
+.. py:data:: PHI3_TEMPLATE
+
+.. py:data:: QWEN2_TEMPLATE
+
+.. py:data:: QWEN2_TEMPLATE_FOR_TOOL
+
+.. py:data:: YI1_5_TEMPLATE
+
+.. py:data:: ZEPHYR_TEMPLATE
+
+.. py:data:: PRESET_TEMPLATES
+
diff --git a/_sources/autoapi/lmflow/utils/conversation_template/internlm/index.rst.txt b/_sources/autoapi/lmflow/utils/conversation_template/internlm/index.rst.txt
new file mode 100644
index 000000000..26e182b57
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/conversation_template/internlm/index.rst.txt
@@ -0,0 +1,19 @@
+lmflow.utils.conversation_template.internlm
+===========================================
+
+.. py:module:: lmflow.utils.conversation_template.internlm
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.internlm.INTERNLM2_TEMPLATE
+
+
+Module Contents
+---------------
+
+.. py:data:: INTERNLM2_TEMPLATE
+
diff --git a/_sources/autoapi/lmflow/utils/conversation_template/llama/index.rst.txt b/_sources/autoapi/lmflow/utils/conversation_template/llama/index.rst.txt
new file mode 100644
index 000000000..579b833ee
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/conversation_template/llama/index.rst.txt
@@ -0,0 +1,56 @@
+lmflow.utils.conversation_template.llama
+========================================
+
+.. py:module:: lmflow.utils.conversation_template.llama
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.llama.logger
+   lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE
+   lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE_FOR_TOOL
+   lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE
+   lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE_FOR_TOOL
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.llama.Llama2ConversationTemplate
+   lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: Llama2ConversationTemplate
+
+   Bases: :py:obj:`lmflow.utils.conversation_template.base.ConversationTemplate`
+
+
+   .. py:method:: _encode(tokenizer: transformers.PreTrainedTokenizer, messages: List[Dict[str, str]], system: Optional[str] = None, tools: Optional[str] = None, **kwargs) -> Sequence[Tuple[List[int], List[int]]]
+
+
+.. py:class:: Llama2ConversationTemplateForTool
+
+   Bases: :py:obj:`Llama2ConversationTemplate`
+
+
+   .. py:method:: _encode(tokenizer: transformers.PreTrainedTokenizer, messages: List[Dict[str, str]], system: Optional[str] = None, tools: Optional[str] = None, **kwargs) -> Sequence[Tuple[List[int], List[int]]]
+
+
+.. py:data:: LLAMA3_TEMPLATE
+
+.. py:data:: LLAMA3_TEMPLATE_FOR_TOOL
+
+.. py:data:: LLAMA2_TEMPLATE
+
+.. py:data:: LLAMA2_TEMPLATE_FOR_TOOL
+
diff --git a/_sources/autoapi/lmflow/utils/conversation_template/phi/index.rst.txt b/_sources/autoapi/lmflow/utils/conversation_template/phi/index.rst.txt
new file mode 100644
index 000000000..727d06e43
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/conversation_template/phi/index.rst.txt
@@ -0,0 +1,19 @@
+lmflow.utils.conversation_template.phi
+======================================
+
+.. py:module:: lmflow.utils.conversation_template.phi
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.phi.PHI3_TEMPLATE
+
+
+Module Contents
+---------------
+
+.. py:data:: PHI3_TEMPLATE
+
diff --git a/_sources/autoapi/lmflow/utils/conversation_template/qwen/index.rst.txt b/_sources/autoapi/lmflow/utils/conversation_template/qwen/index.rst.txt
new file mode 100644
index 000000000..bfe5c7348
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/conversation_template/qwen/index.rst.txt
@@ -0,0 +1,22 @@
+lmflow.utils.conversation_template.qwen
+=======================================
+
+.. py:module:: lmflow.utils.conversation_template.qwen
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE
+   lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE_FOR_TOOL
+
+
+Module Contents
+---------------
+
+.. py:data:: QWEN2_TEMPLATE
+
+.. py:data:: QWEN2_TEMPLATE_FOR_TOOL
+
diff --git a/_sources/autoapi/lmflow/utils/conversation_template/yi/index.rst.txt b/_sources/autoapi/lmflow/utils/conversation_template/yi/index.rst.txt
new file mode 100644
index 000000000..b088e0fe2
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/conversation_template/yi/index.rst.txt
@@ -0,0 +1,19 @@
+lmflow.utils.conversation_template.yi
+=====================================
+
+.. py:module:: lmflow.utils.conversation_template.yi
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.yi.YI1_5_TEMPLATE
+
+
+Module Contents
+---------------
+
+.. py:data:: YI1_5_TEMPLATE
+
diff --git a/_sources/autoapi/lmflow/utils/conversation_template/zephyr/index.rst.txt b/_sources/autoapi/lmflow/utils/conversation_template/zephyr/index.rst.txt
new file mode 100644
index 000000000..0f5534552
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/conversation_template/zephyr/index.rst.txt
@@ -0,0 +1,38 @@
+lmflow.utils.conversation_template.zephyr
+=========================================
+
+.. py:module:: lmflow.utils.conversation_template.zephyr
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.zephyr.logger
+   lmflow.utils.conversation_template.zephyr.ZEPHYR_TEMPLATE
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:class:: ZephyrConversationTemplate
+
+   Bases: :py:obj:`lmflow.utils.conversation_template.base.ConversationTemplate`
+
+
+   .. py:method:: _encode(tokenizer: transformers.PreTrainedTokenizer, messages: List[Dict[str, str]], system: Optional[str] = None, tools: Optional[str] = None, **kwargs) -> Sequence[Tuple[List[int], List[int]]]
+
+
+.. py:data:: ZEPHYR_TEMPLATE
+
diff --git a/_sources/autoapi/lmflow/utils/data_utils/index.rst.txt b/_sources/autoapi/lmflow/utils/data_utils/index.rst.txt
new file mode 100644
index 000000000..2a4489bff
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/data_utils/index.rst.txt
@@ -0,0 +1,241 @@
+lmflow.utils.data_utils
+=======================
+
+.. py:module:: lmflow.utils.data_utils
+
+.. autoapi-nested-parse::
+
+   The program includes several functions: setting a random seed, 
+   loading data from a JSON file, batching data, and extracting answers from generated text.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.utils.data_utils.VLLMInferenceResultWithInput
+   lmflow.utils.data_utils.RewardModelInferenceResultWithInput
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.utils.data_utils.set_random_seed
+   lmflow.utils.data_utils.load_data
+   lmflow.utils.data_utils.batchlize
+   lmflow.utils.data_utils.answer_extraction
+   lmflow.utils.data_utils.process_image_flag
+
+
+Module Contents
+---------------
+
+.. py:function:: set_random_seed(seed: int)
+
+   
+   Set the random seed for `random`, `numpy`, `torch`, `torch.cuda`.
+
+
+   :Parameters:
+
+       **seed** : int
+           The default seed.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:function:: load_data(file_name: str)
+
+   
+   Load data with file name.
+
+
+   :Parameters:
+
+       **file_name** : str.
+           The dataset file name.
+
+
+
+   :Returns:
+
+       **inputs** : list.
+           The input texts of the dataset.
+
+       **outputs** : list.
+           The output texts file datasets.    
+
+       **len** : int.
+           The length of the dataset.
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:function:: batchlize(examples: list, batch_size: int, random_shuffle: bool)
+
+   
+   Convert examples to a dataloader.
+
+
+   :Parameters:
+
+       **examples** : list.
+           Data list.
+
+       **batch_size** : int.
+           ..
+
+       **random_shuffle** : bool
+           If true, the dataloader shuffle the training data.
+
+
+
+   :Returns:
+
+       dataloader:
+           Dataloader with batch generator.
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:function:: answer_extraction(response, answer_type=None)
+
+   
+   Use this funtion to extract answers from generated text
+
+
+   :Parameters:
+
+       **args**
+           Arguments.
+
+       **response** : str
+           plain string response.
+
+
+
+   :Returns:
+
+       answer:
+           Decoded answer (such as A, B, C, D, E for mutiple-choice QA).
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+.. py:function:: process_image_flag(text, image_flag='<ImageHere>')
+
+.. py:class:: VLLMInferenceResultWithInput
+
+   Bases: :py:obj:`TypedDict`
+
+
+   
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: input
+      :type:  str
+
+
+   .. py:attribute:: output
+      :type:  Union[List[str], List[List[int]]]
+
+
+.. py:class:: RewardModelInferenceResultWithInput
+
+   Bases: :py:obj:`TypedDict`
+
+
+   
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: input
+      :type:  str
+
+
+   .. py:attribute:: output
+      :type:  List[Dict[str, Union[str, float]]]
+
+
diff --git a/_sources/autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.rst.txt b/_sources/autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.rst.txt
new file mode 100644
index 000000000..090a98ef9
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.rst.txt
@@ -0,0 +1,25 @@
+lmflow.utils.flash_attention.bloom_flash_attention
+==================================================
+
+.. py:module:: lmflow.utils.flash_attention.bloom_flash_attention
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.utils.flash_attention.bloom_flash_attention.forward
+   lmflow.utils.flash_attention.bloom_flash_attention._prepare_attn_mask
+   lmflow.utils.flash_attention.bloom_flash_attention.replace_bloom_attn_with_flash_attn
+
+
+Module Contents
+---------------
+
+.. py:function:: forward(self, hidden_states: torch.Tensor, residual: torch.Tensor, alibi: torch.Tensor, attention_mask: torch.Tensor, layer_past: Optional[Tuple[torch.Tensor, torch.Tensor]] = None, head_mask: Optional[torch.Tensor] = None, use_cache: bool = False, output_attentions: bool = False)
+
+.. py:function:: _prepare_attn_mask(self, attention_mask: torch.Tensor, input_shape: Tuple[int, int], past_key_values_length: int) -> torch.BoolTensor
+
+.. py:function:: replace_bloom_attn_with_flash_attn()
+
diff --git a/_sources/autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.rst.txt b/_sources/autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.rst.txt
new file mode 100644
index 000000000..59785df48
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.rst.txt
@@ -0,0 +1,25 @@
+lmflow.utils.flash_attention.gpt2_flash_attention
+=================================================
+
+.. py:module:: lmflow.utils.flash_attention.gpt2_flash_attention
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.utils.flash_attention.gpt2_flash_attention.forward
+   lmflow.utils.flash_attention.gpt2_flash_attention._prepare_decoder_attention_mask
+   lmflow.utils.flash_attention.gpt2_flash_attention.replace_gpt2_attn_with_flash_attn
+
+
+Module Contents
+---------------
+
+.. py:function:: forward(self, hidden_states: Optional[Tuple[torch.FloatTensor]], layer_past: Optional[Tuple[torch.Tensor]] = None, attention_mask: Optional[torch.FloatTensor] = None, head_mask: Optional[torch.FloatTensor] = None, encoder_hidden_states: Optional[torch.Tensor] = None, encoder_attention_mask: Optional[torch.FloatTensor] = None, use_cache: Optional[bool] = False, output_attentions: Optional[bool] = False) -> Tuple[Union[torch.Tensor, Tuple[torch.Tensor]], Ellipsis]
+
+.. py:function:: _prepare_decoder_attention_mask(self, attention_mask, input_shape, inputs_embeds, past_key_values_length)
+
+.. py:function:: replace_gpt2_attn_with_flash_attn()
+
diff --git a/_sources/autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.rst.txt b/_sources/autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.rst.txt
new file mode 100644
index 000000000..e093d94c1
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.rst.txt
@@ -0,0 +1,25 @@
+lmflow.utils.flash_attention.gpt_neo_flash_attention
+====================================================
+
+.. py:module:: lmflow.utils.flash_attention.gpt_neo_flash_attention
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.utils.flash_attention.gpt_neo_flash_attention._attn
+   lmflow.utils.flash_attention.gpt_neo_flash_attention.forward
+   lmflow.utils.flash_attention.gpt_neo_flash_attention.replace_gpt_neo_attn_with_flash_attn
+
+
+Module Contents
+---------------
+
+.. py:function:: _attn(self, query, key, value, attention_mask=None, head_mask=None)
+
+.. py:function:: forward(self, hidden_states, attention_mask=None, layer_past=None, head_mask=None, use_cache=False, output_attentions=False)
+
+.. py:function:: replace_gpt_neo_attn_with_flash_attn()
+
diff --git a/_sources/autoapi/lmflow/utils/flash_attention/index.rst.txt b/_sources/autoapi/lmflow/utils/flash_attention/index.rst.txt
new file mode 100644
index 000000000..8194e87ee
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/flash_attention/index.rst.txt
@@ -0,0 +1,19 @@
+lmflow.utils.flash_attention
+============================
+
+.. py:module:: lmflow.utils.flash_attention
+
+
+Submodules
+----------
+
+.. toctree::
+   :maxdepth: 1
+
+   /autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index
+   /autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index
+   /autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index
+   /autoapi/lmflow/utils/flash_attention/llama_flash_attention/index
+   /autoapi/lmflow/utils/flash_attention/triton_flash_attention/index
+
+
diff --git a/_sources/autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.rst.txt b/_sources/autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.rst.txt
new file mode 100644
index 000000000..1f9392c98
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.rst.txt
@@ -0,0 +1,25 @@
+lmflow.utils.flash_attention.llama_flash_attention
+==================================================
+
+.. py:module:: lmflow.utils.flash_attention.llama_flash_attention
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.utils.flash_attention.llama_flash_attention.forward
+   lmflow.utils.flash_attention.llama_flash_attention._prepare_decoder_attention_mask
+   lmflow.utils.flash_attention.llama_flash_attention.replace_llama_attn_with_flash_attn
+
+
+Module Contents
+---------------
+
+.. py:function:: forward(self, hidden_states: torch.Tensor, attention_mask: Optional[torch.Tensor] = None, position_ids: Optional[torch.LongTensor] = None, past_key_value: Optional[Tuple[torch.Tensor]] = None, output_attentions: bool = False, use_cache: bool = False) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]
+
+.. py:function:: _prepare_decoder_attention_mask(self, attention_mask, input_shape, inputs_embeds, past_key_values_length)
+
+.. py:function:: replace_llama_attn_with_flash_attn()
+
diff --git a/_sources/autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.rst.txt b/_sources/autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.rst.txt
new file mode 100644
index 000000000..62dcdc3ea
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.rst.txt
@@ -0,0 +1,229 @@
+lmflow.utils.flash_attention.triton_flash_attention
+===================================================
+
+.. py:module:: lmflow.utils.flash_attention.triton_flash_attention
+
+.. autoapi-nested-parse::
+
+   *Experimental* implementation of FlashAttention in Triton.
+   Tested with triton==2.0.0.dev20221202.
+   Triton 2.0 has a new backend (MLIR) but seems like it doesn't yet work for head dimensions
+   other than 64:
+   https://github.com/openai/triton/blob/d376020f90002757eea3ea9475d4f7cfc2ec5ead/python/triton/ops/flash_attention.py#L207
+   We'll update this implementation with the new Triton backend once this is fixed.
+
+   We use the FlashAttention implementation from Phil Tillet a starting point.
+   https://github.com/openai/triton/blob/master/python/tutorials/06-fused-attention.py
+
+   Changes:
+   - Implement both causal and non-causal attention.
+   - Implement both self-attention and cross-attention.
+   - Support arbitrary seqlens (not just multiples of 128), for both forward and backward.
+   - Support all head dimensions up to 128 (not just 16, 32, 64, 128), for both forward and backward.
+   - Support attention bias.
+   - Speed up the forward pass a bit, and only store the LSE instead of m and l.
+   - Make the backward for d=128 much faster by reducing register spilling.
+   - Optionally parallelize the backward pass across seqlen_k, to deal with the case of
+   small batch size * nheads.
+
+   Caution:
+   - This is an *experimental* implementation. The forward pass should be quite robust but
+   I'm not 100% sure that the backward pass doesn't have race conditions (due to the Triton compiler).
+   - This implementation has only been tested on A100.
+   - If you plan to use headdim other than 64 and 128, you should test for race conditions
+   (due to the Triton compiler), as done in tests/test_flash_attn.py
+   "test_flash_attn_triton_race_condition". I've tested and fixed many race conditions
+   for different head dimensions (40, 48, 64, 128, 80, 88, 96), but I'm still not 100% confident
+   that there are none left for other head dimensions.
+
+   Differences between this Triton version and the CUDA version:
+   - Triton version doesn't support dropout.
+   - Triton forward is generally faster than CUDA forward, while Triton backward is
+   generally slower than CUDA backward. Overall Triton forward + backward is slightly slower
+   than CUDA forward + backward.
+   - Triton version doesn't support different sequence lengths in a batch (i.e., RaggedTensor/NestedTensor).
+   - Triton version supports attention bias, while CUDA version doesn't.
+
+   ..
+       !! processed by numpydoc !!
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.flash_attention.triton_flash_attention.flash_attn_qkvpacked_func
+   lmflow.utils.flash_attention.triton_flash_attention.flash_attn_kvpacked_func
+   lmflow.utils.flash_attention.triton_flash_attention.flash_attn_func
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc
+   lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc
+   lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.utils.flash_attention.triton_flash_attention._fwd_kernel
+   lmflow.utils.flash_attention.triton_flash_attention._bwd_preprocess_do_o_dot
+   lmflow.utils.flash_attention.triton_flash_attention._bwd_store_dk_dv
+   lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel_one_col_block
+   lmflow.utils.flash_attention.triton_flash_attention.init_to_zero
+   lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel
+   lmflow.utils.flash_attention.triton_flash_attention._flash_attn_forward
+   lmflow.utils.flash_attention.triton_flash_attention._flash_attn_backward
+
+
+Module Contents
+---------------
+
+.. py:function:: _fwd_kernel(Q, K, V, Bias, Out, Lse, TMP, softmax_scale, stride_qb, stride_qh, stride_qm, stride_kb, stride_kh, stride_kn, stride_vb, stride_vh, stride_vn, stride_bb, stride_bh, stride_bm, stride_ob, stride_oh, stride_om, nheads, seqlen_q, seqlen_k, seqlen_q_rounded, headdim, CACHE_KEY_SEQLEN_Q, CACHE_KEY_SEQLEN_K, BIAS_TYPE: triton.language.constexpr, IS_CAUSAL: triton.language.constexpr, BLOCK_HEADDIM: triton.language.constexpr, EVEN_M: triton.language.constexpr, EVEN_N: triton.language.constexpr, EVEN_HEADDIM: triton.language.constexpr, BLOCK_M: triton.language.constexpr, BLOCK_N: triton.language.constexpr)
+
+.. py:function:: _bwd_preprocess_do_o_dot(Out, DO, Delta, stride_ob, stride_oh, stride_om, stride_dob, stride_doh, stride_dom, nheads, seqlen_q, seqlen_q_rounded, headdim, BLOCK_M: triton.language.constexpr, BLOCK_HEADDIM: triton.language.constexpr)
+
+.. py:function:: _bwd_store_dk_dv(dk_ptrs, dv_ptrs, dk, dv, offs_n, offs_d, seqlen_k, headdim, EVEN_M: triton.language.constexpr, EVEN_N: triton.language.constexpr, EVEN_HEADDIM: triton.language.constexpr)
+
+.. py:function:: _bwd_kernel_one_col_block(start_n, Q, K, V, Bias, DO, DQ, DK, DV, LSE, D, softmax_scale, stride_qm, stride_kn, stride_vn, stride_bm, stride_dom, stride_dqm, stride_dkn, stride_dvn, seqlen_q, seqlen_k, headdim, ATOMIC_ADD: triton.language.constexpr, BIAS_TYPE: triton.language.constexpr, IS_CAUSAL: triton.language.constexpr, BLOCK_HEADDIM: triton.language.constexpr, EVEN_M: triton.language.constexpr, EVEN_N: triton.language.constexpr, EVEN_HEADDIM: triton.language.constexpr, BLOCK_M: triton.language.constexpr, BLOCK_N: triton.language.constexpr)
+
+.. py:function:: init_to_zero(name)
+
+.. py:function:: _bwd_kernel(Q, K, V, Bias, DO, DQ, DK, DV, LSE, D, softmax_scale, stride_qb, stride_qh, stride_qm, stride_kb, stride_kh, stride_kn, stride_vb, stride_vh, stride_vn, stride_bb, stride_bh, stride_bm, stride_dob, stride_doh, stride_dom, stride_dqb, stride_dqh, stride_dqm, stride_dkb, stride_dkh, stride_dkn, stride_dvb, stride_dvh, stride_dvn, nheads, seqlen_q, seqlen_k, seqlen_q_rounded, headdim, CACHE_KEY_SEQLEN_Q, CACHE_KEY_SEQLEN_K, BIAS_TYPE: triton.language.constexpr, IS_CAUSAL: triton.language.constexpr, BLOCK_HEADDIM: triton.language.constexpr, SEQUENCE_PARALLEL: triton.language.constexpr, EVEN_M: triton.language.constexpr, EVEN_N: triton.language.constexpr, EVEN_HEADDIM: triton.language.constexpr, BLOCK_M: triton.language.constexpr, BLOCK_N: triton.language.constexpr)
+
+.. py:function:: _flash_attn_forward(q, k, v, bias=None, causal=False, softmax_scale=None)
+
+.. py:function:: _flash_attn_backward(do, q, k, v, o, lse, dq, dk, dv, bias=None, causal=False, softmax_scale=None)
+
+.. py:class:: FlashAttnQKVPackedFunc
+
+   Bases: :py:obj:`torch.autograd.Function`
+
+
+   .. py:method:: forward(ctx, qkv, bias=None, causal=False, softmax_scale=None)
+      :staticmethod:
+
+
+      
+      qkv: (batch, seqlen, 3, nheads, headdim)
+      bias: optional, shape broadcastible to (batch, nheads, seqlen, seqlen).
+          For example, ALiBi mask for causal would have shape (1, nheads, 1, seqlen).
+          ALiBi mask for non-causal would have shape (1, nheads, seqlen, seqlen)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: backward(ctx, do)
+      :staticmethod:
+
+
+
+.. py:data:: flash_attn_qkvpacked_func
+
+.. py:class:: FlashAttnKVPackedFunc
+
+   Bases: :py:obj:`torch.autograd.Function`
+
+
+   .. py:method:: forward(ctx, q, kv, bias=None, causal=False, softmax_scale=None)
+      :staticmethod:
+
+
+      
+      q: (batch, seqlen_q, nheads, headdim)
+      kv: (batch, seqlen_k, 2, nheads, headdim)
+      bias: optional, shape broadcastible to (batch, nheads, seqlen_q, seqlen_k).
+          For example, ALiBi mask for causal would have shape (1, nheads, 1, seqlen_k).
+          ALiBi mask for non-causal would have shape (1, nheads, seqlen_q, seqlen_k)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: backward(ctx, do)
+      :staticmethod:
+
+
+
+.. py:data:: flash_attn_kvpacked_func
+
+.. py:class:: FlashAttnFunc
+
+   Bases: :py:obj:`torch.autograd.Function`
+
+
+   .. py:method:: forward(ctx, q, k, v, bias=None, causal=False, softmax_scale=None)
+      :staticmethod:
+
+
+      
+      q: (batch_size, seqlen_q, nheads, headdim)
+      k, v: (batch_size, seqlen_k, nheads, headdim)
+      bias: optional, shape broadcastible to (batch, nheads, seqlen_q, seqlen_k).
+          For example, ALiBi mask for causal would have shape (1, nheads, 1, seqlen_k).
+          ALiBi mask for non-causal would have shape (1, nheads, seqlen_q, seqlen_k)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+      ..
+          !! processed by numpydoc !!
+
+
+   .. py:method:: backward(ctx, do)
+      :staticmethod:
+
+
+
+.. py:data:: flash_attn_func
+
diff --git a/_sources/autoapi/lmflow/utils/index.rst.txt b/_sources/autoapi/lmflow/utils/index.rst.txt
new file mode 100644
index 000000000..da7015fe2
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/index.rst.txt
@@ -0,0 +1,24 @@
+lmflow.utils
+============
+
+.. py:module:: lmflow.utils
+
+
+Submodules
+----------
+
+.. toctree::
+   :maxdepth: 1
+
+   /autoapi/lmflow/utils/common/index
+   /autoapi/lmflow/utils/constants/index
+   /autoapi/lmflow/utils/conversation_template/index
+   /autoapi/lmflow/utils/data_utils/index
+   /autoapi/lmflow/utils/flash_attention/index
+   /autoapi/lmflow/utils/llava_conversation_lib/index
+   /autoapi/lmflow/utils/model/index
+   /autoapi/lmflow/utils/multimodal/index
+   /autoapi/lmflow/utils/position_interpolation/index
+   /autoapi/lmflow/utils/versioning/index
+
+
diff --git a/_sources/autoapi/lmflow/utils/llava_conversation_lib/index.rst.txt b/_sources/autoapi/lmflow/utils/llava_conversation_lib/index.rst.txt
new file mode 100644
index 000000000..8253f9e45
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/llava_conversation_lib/index.rst.txt
@@ -0,0 +1,187 @@
+lmflow.utils.llava_conversation_lib
+===================================
+
+.. py:module:: lmflow.utils.llava_conversation_lib
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.llava_conversation_lib.conv_vicuna_v0
+   lmflow.utils.llava_conversation_lib.conv_vicuna_v1
+   lmflow.utils.llava_conversation_lib.conv_llama_2
+   lmflow.utils.llava_conversation_lib.conv_llava_llama_2
+   lmflow.utils.llava_conversation_lib.conv_mpt
+   lmflow.utils.llava_conversation_lib.conv_llava_plain
+   lmflow.utils.llava_conversation_lib.conv_llava_v0
+   lmflow.utils.llava_conversation_lib.conv_llava_v0_mmtag
+   lmflow.utils.llava_conversation_lib.conv_llava_v1
+   lmflow.utils.llava_conversation_lib.conv_llava_v1_mmtag
+   lmflow.utils.llava_conversation_lib.default_conversation
+   lmflow.utils.llava_conversation_lib.conv_templates
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.utils.llava_conversation_lib.SeparatorStyle
+   lmflow.utils.llava_conversation_lib.Conversation
+
+
+Module Contents
+---------------
+
+.. py:class:: SeparatorStyle(*args, **kwds)
+
+   Bases: :py:obj:`enum.Enum`
+
+
+   
+   Different separator style.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: SINGLE
+
+
+   .. py:attribute:: TWO
+
+
+   .. py:attribute:: MPT
+
+
+   .. py:attribute:: PLAIN
+
+
+   .. py:attribute:: LLAMA_2
+
+
+.. py:class:: Conversation
+
+   
+   A class that keeps all conversation history.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+   ..
+       !! processed by numpydoc !!
+
+   .. py:attribute:: system
+      :type:  str
+
+
+   .. py:attribute:: roles
+      :type:  List[str]
+
+
+   .. py:attribute:: messages
+      :type:  List[List[str]]
+
+
+   .. py:attribute:: offset
+      :type:  int
+
+
+   .. py:attribute:: sep_style
+      :type:  SeparatorStyle
+
+
+   .. py:attribute:: sep
+      :type:  str
+      :value: '###'
+
+
+
+   .. py:attribute:: sep2
+      :type:  str
+      :value: None
+
+
+
+   .. py:attribute:: version
+      :type:  str
+      :value: 'Unknown'
+
+
+
+   .. py:attribute:: skip_next
+      :type:  bool
+      :value: False
+
+
+
+   .. py:method:: get_prompt()
+
+
+   .. py:method:: append_message(role, message)
+
+
+   .. py:method:: get_images(return_pil=False)
+
+
+   .. py:method:: to_gradio_chatbot()
+
+
+   .. py:method:: copy()
+
+
+   .. py:method:: dict()
+
+
+.. py:data:: conv_vicuna_v0
+
+.. py:data:: conv_vicuna_v1
+
+.. py:data:: conv_llama_2
+
+.. py:data:: conv_llava_llama_2
+
+.. py:data:: conv_mpt
+
+.. py:data:: conv_llava_plain
+
+.. py:data:: conv_llava_v0
+
+.. py:data:: conv_llava_v0_mmtag
+
+.. py:data:: conv_llava_v1
+
+.. py:data:: conv_llava_v1_mmtag
+
+.. py:data:: default_conversation
+
+.. py:data:: conv_templates
+
diff --git a/_sources/autoapi/lmflow/utils/model/index.rst.txt b/_sources/autoapi/lmflow/utils/model/index.rst.txt
new file mode 100644
index 000000000..0f106ecf0
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/model/index.rst.txt
@@ -0,0 +1,29 @@
+lmflow.utils.model
+==================
+
+.. py:module:: lmflow.utils.model
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.model.logger
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.utils.model.check_homogeneity
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:function:: check_homogeneity(model_args_list: List[lmflow.args.ModelArguments]) -> bool
+
diff --git a/_sources/autoapi/lmflow/utils/multimodal/index.rst.txt b/_sources/autoapi/lmflow/utils/multimodal/index.rst.txt
new file mode 100644
index 000000000..a700ab4dc
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/multimodal/index.rst.txt
@@ -0,0 +1,25 @@
+lmflow.utils.multimodal
+=======================
+
+.. py:module:: lmflow.utils.multimodal
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.utils.multimodal.update_custom_config
+   lmflow.utils.multimodal.load_llava_pretrain_model
+   lmflow.utils.multimodal.adapt_llava_model_to_lmflow_type
+
+
+Module Contents
+---------------
+
+.. py:function:: update_custom_config(config, model_args)
+
+.. py:function:: load_llava_pretrain_model(model, checkpoint_path)
+
+.. py:function:: adapt_llava_model_to_lmflow_type(state_dict)
+
diff --git a/_sources/autoapi/lmflow/utils/position_interpolation/index.rst.txt b/_sources/autoapi/lmflow/utils/position_interpolation/index.rst.txt
new file mode 100644
index 000000000..586547af6
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/position_interpolation/index.rst.txt
@@ -0,0 +1,15 @@
+lmflow.utils.position_interpolation
+===================================
+
+.. py:module:: lmflow.utils.position_interpolation
+
+
+Submodules
+----------
+
+.. toctree::
+   :maxdepth: 1
+
+   /autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index
+
+
diff --git a/_sources/autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.rst.txt b/_sources/autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.rst.txt
new file mode 100644
index 000000000..60e8d61c8
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.rst.txt
@@ -0,0 +1,44 @@
+lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch
+==================================================================
+
+.. py:module:: lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch
+
+
+Classes
+-------
+
+.. autoapisummary::
+
+   lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.replace_llama_with_condense
+
+
+Module Contents
+---------------
+
+.. py:class:: CondenseRotaryEmbedding(dim, pi_ratio, ntk_ratio, max_position_embeddings=2048, base=10000, device=None)
+
+   Bases: :py:obj:`torch.nn.Module`
+
+
+   .. py:attribute:: ntk_ratio
+
+
+   .. py:attribute:: pi_ratio
+
+
+   .. py:attribute:: max_seq_len_cached
+
+
+   .. py:method:: forward(x, seq_len=None)
+
+
+.. py:function:: replace_llama_with_condense(pi_ratio, ntk_ratio)
+
diff --git a/_sources/autoapi/lmflow/utils/versioning/index.rst.txt b/_sources/autoapi/lmflow/utils/versioning/index.rst.txt
new file mode 100644
index 000000000..bc05d4efa
--- /dev/null
+++ b/_sources/autoapi/lmflow/utils/versioning/index.rst.txt
@@ -0,0 +1,59 @@
+lmflow.utils.versioning
+=======================
+
+.. py:module:: lmflow.utils.versioning
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.utils.versioning.logger
+
+
+Functions
+---------
+
+.. autoapisummary::
+
+   lmflow.utils.versioning.get_python_version
+   lmflow.utils.versioning._is_package_available
+   lmflow.utils.versioning._is_packages_available
+   lmflow.utils.versioning.is_package_version_at_least
+   lmflow.utils.versioning.is_gradio_available
+   lmflow.utils.versioning.is_ray_available
+   lmflow.utils.versioning.is_vllm_available
+   lmflow.utils.versioning.is_flash_attn_available
+   lmflow.utils.versioning.is_flask_available
+   lmflow.utils.versioning.is_trl_available
+   lmflow.utils.versioning.is_multimodal_available
+
+
+Module Contents
+---------------
+
+.. py:data:: logger
+
+.. py:function:: get_python_version()
+
+.. py:function:: _is_package_available(package_name: str, skippable: bool = False)
+
+.. py:function:: _is_packages_available(packages: Union[List[str], List[Tuple[str, bool]]])
+
+.. py:function:: is_package_version_at_least(package_name, min_version)
+
+.. py:function:: is_gradio_available()
+
+.. py:function:: is_ray_available()
+
+.. py:function:: is_vllm_available()
+
+.. py:function:: is_flash_attn_available()
+
+.. py:function:: is_flask_available()
+
+.. py:function:: is_trl_available()
+
+.. py:function:: is_multimodal_available()
+
diff --git a/_sources/autoapi/lmflow/version/index.rst.txt b/_sources/autoapi/lmflow/version/index.rst.txt
new file mode 100644
index 000000000..0da4c4154
--- /dev/null
+++ b/_sources/autoapi/lmflow/version/index.rst.txt
@@ -0,0 +1,21 @@
+lmflow.version
+==============
+
+.. py:module:: lmflow.version
+
+
+Attributes
+----------
+
+.. autoapisummary::
+
+   lmflow.version.__version__
+
+
+Module Contents
+---------------
+
+.. py:data:: __version__
+   :value: '0.0.9'
+
+
diff --git a/_sources/blogs/benchmark.md.txt b/_sources/blogs/benchmark.md.txt
new file mode 100644
index 000000000..c09c74868
--- /dev/null
+++ b/_sources/blogs/benchmark.md.txt
@@ -0,0 +1,253 @@
+# LMFlow Benchmark: An Automatic Evaluation Framework for Open-Source LLMs
+
+May 9, 2023
+
+
+## Introduction
+
+Evaluation of a chat-style Large Language Model (LLM) has been a huge challenge since the breakthrough of ChatGPT. On the one hand, researchers and engineers need a reliable way to compare two models and decide which model to choose under a certain application scenario. On the other hand, they have to monitor the model performance during the training of an LLM to avoid performance issues such as forgetting.
+
+Recent work of Vicuna introduces comparison methods of human evaluation, a.k.a. Chatbot Arena. They also pioneered the evaluation method by invoking GPT-4 to compare the outputs of two models. However, those methods require expensive human labeling or GPT-4 API calls, which are neither scalable nor convenient for LLM development.
+
+In this article, we introduce LMFlow benchmark, a new benchmark which provides a cheap and easy-to-use evaluation framework that can help reflect different aspects of LLMs. We have open-sourced the dataset and the code as well, so that everyone in the LLM community can use those toolkits to evaluate, monitor or compare different LLMs.
+## Metric
+In our evaluation framework, Negative Log Likelihood (NLL) is used for evaluating LLM
+![](../_static/nll.png)
+
+
+which corresponds to the LLM model’s prediction probability over a corpus set given their contexts. If the corpus set itself indicates a certain type of LLM ability, such as multi-round conversation, instruction following, math problem solving, role-playing, then NLL on those corpora can provide quantitative metrics to reflect those abilities.
+
+![](../_static/benchmark-1.png)
+
+The key idea behind NLL, is that
+
+*Generation ability is positively correlated with prediction ability.*
+
+For instance, an LLM which performs well in essay writing should have no problem understanding and predicting a reference human essay, just like human chess masters performing well at memorizing an endgame on a chessboard.
+
+Besides NLL, another similar and commonly used metric in NLP is Perplexity (PPL):
+
+![](../_static/ppl.png)
+
+Nevertheless, perplexity intrinsically depends on the lengths of the tokenized sequences, which induces unfair comparison between models with different tokenizers. For example, if a model has a smaller vocabulary size, it inherently results in a longer tokenized sequence and a lower token-level perplexity. Thus in all our experiments, we use NLL instead of PPL.
+
+One huge advantage of NLL evaluation is that it does not require human involvement during the evaluation process. As long as the test reference corpus is given, one can evaluate different aspects of an LLM’s ability automatically. This makes the evaluation of LLM more accessible to researchers.
+
+Besides its convenience, NLL itself is also a good metric. In our experimental results in commonsense QA, we find that NLL is correlated with QA accuracy when comparing the different finetuned versions of a single model.
+
+**Table 1: Accuracy results in traditional commonsense QA benchmarks**
+
+
+||winogrande|boolq|arc\_e|hellaswag|piqa|obqa|arc\_c|Average|
+| :- | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
+|bloom-3b|58\.7|61\.6|59\.5|52\.7|70\.8|42\.2|30\.6|53\.7|
+|bloom-7.1b|64\.4|62\.9|65\.0|59\.6|73\.6|35\.8|33\.4|56\.3|
+|opt-6.9b|65\.2|66\.1|65\.6|67\.2|76\.5|37\.4|34\.6|58\.9|
+|opt-13b|65\.0|65\.9|67\.1|69\.8|76\.9|39\.0|35\.7|59\.9|
+|llama-7b|67\.9|73\.2|67\.3|73\.0|78\.3|42\.4|41\.4|62\.7|
+|llama-13b|**70.0**|**68.5**|**74.5**|**76.2**|**79.1**|**42.2**|**44.5**|**65.0**|
+
+**Table 2: NLL results in corpus of commonsense QA benchmarks**
+
+
+||winogrande|boolq|arc\_e|hellaswag|piqa|obqa|arc\_c|Average|
+| :- | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
+|bloom-3b|86\.5|228|86|245|134|64\.5|101\.5|135\.1|
+|bloom-7.1b|85|215|81\.5|237|130|62\.5|96|129\.5|
+|opt-6.9b|81\.5|200|81\.5|224|125|61|96|124\.1|
+|opt-13b|82|198|82\.5|220|125|61\.8|97|123\.7|
+|llama-7b|79\.5|167|71\.5|214|121|58|85|113\.7|
+|llama-13b|**79**|**153**|**70**|**207**|**119**|**57.3**|**83**|**109.7**|
+
+
+**Figure 1: Correlation between NLL and accuracy on commonsense QA benchmarks**
+
+![](../_static/benchmark-2.png)
+
+In the above figure, one can find that QA accuracy is roughly correlated to NLL. Thus NLL is able to reflect the “magnitude” of prediction level difference between models. A huge gap in NLL normally entails a huge performance gap.
+
+In the following sections, we provide a comprehensive evaluation of currently available LLM models and summarize their performance. Due to page limits, we only demonstrate partial evaluation results in this article. The full results can be found here [LLM Comparison Sheet 0501](https://docs.google.com/spreadsheets/d/1JYh4_pxNzmNA9I0YM2epgRA7VXBIeIGS64gPJBg5NHA/edit?usp=sharing). One can also evaluate their own LLM models using our LMFlow evaluation toolkit (<https://optimalscale.github.io/LMFlow/examples/TASK_GUIDE.html>).
+
+We also introduce our own finetuned model Robin-7b-Chat and compare its performance with some other popular models. One may try our online demo LMFlow-Chat at <https://lmflow.com/>.
+
+## Chat Performance
+The first aspect of LLM is chat performance, especially multi-round chat. Based on our experimental results, we have the following observations:
+
+1. *Without finetuning, foundation models do not perform well on chat.* 
+1. *After finetuning, foundation models can all obtain huge performance gain.*
+1. *Our Robin-Chat-7b achieves comparable chat performance with Vicuna-13b.*
+
+
+
+**Table 3: NLL on multi-round chat datasets, Lower the better**
+
+
+||En\_multi\_round\_chat|En\_singe\_round\_chat|Average|
+| :- | :-: | :-: | :-: |
+|Bloom-7.1b|188|213|200\.5|
+|Bloom-7.1b-chat|164|181|172\.5|
+|Llama-7b|163   |184    |173\.5|
+|Vicuna-7b|155  |175    |165|
+|Vicuna-13b|150 |170    |160|
+|Robin-Chat-7b|**146**  |**164**    |**155**|
+
+We also evaluate Vicuna-13b and our Robin-Chat-7b model qualitatively via use cases. It turns out that Robin-Chat-7b’s performance is comparable to Vicuna-13b. For example, both models can effortlessly refer to the conversation context with only slight mistakes, where Vicuna-13b sometimes doesn’t count the item correctly, while Robin-Chat-7b introduces only Mexican foods for international cuisine recommendations and omits other choices.
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+**Table 4: Use cases of multi-round chat**
+
+
+|Round|Question|Vicuna-13b|Robin-Chat-7b|
+| :- | :-: | :-: | :-: |
+|R1|Where is the capital of USA?|The capital of the United States is **Washington, D.C.** (District of Columbia). It is located on the east coast of the country, in the northern part of the state of Maryland. Washington, D.C. is named after George Washington, the first president of the United States. It is a federal district, which means that it is not part of any state, but is instead directly administered by the federal government. The city is home to many national landmarks, including the White House, the Capitol Building, and the Supreme Court. It is also home to many museums, monuments, and other cultural institutions.|The capital of the United States is **Washington, D.C.** (District of Columbia). It is located on the east coast of the USA, between the states of Maryland and Virginia.|
+|R2|Can you introduce some popular food there?|<p>Washington, D.C. is a city with a diverse food scene, and there are many popular restaurants and cuisines to choose from. Some popular foods in Washington, D.C. include:</p><p>- American Classics: There are many restaurants in Washington, D.C. that serve classic American dishes such as hamburgers, hot dogs, and apple pie.</p><p>- Seafood: Washington, D.C. is located near the Chesapeake Bay, which is known for its fresh seafood. Many restaurants in the city serve dishes such as crab cakes, shrimp, and oysters.</p><p>- Ethnic Cuisine: Washington, D.C. has a diverse population, and there are many restaurants that serve food from different parts of the world. Some popular ethnic cuisines include Mexican, Chinese, Japanese, and Korean.</p><p>- Barbecue: Washington, D.C. is also known for its barbecue restaurants. Many of these restaurants serve pulled pork, ribs, and brisket.</p><p>- **Donuts**: Washington, D.C. is home to the famous "Donut Happy" which is a famous donut shop with a variety of flavors and toppings.</p><p>- Street Food: There are also many street food options in Washington, D.C. such as food trucks and stands that serve food such as falafel, kebabs, and tacos.</p><p>These are just a few examples of the many types of food that are available in Washington, D.C. There is something for everyone in this city when it comes to food.</p><p></p>|<p>Washington, D.C. is a diverse city with a wide range of cuisines to choose from. Here are some popular foods that you can try when you visit:</p><p>1. Soul food: This style of cuisine originated in the southern United States and is known for its bold flavors and hearty portions. Some popular soul food dishes in Washington, D.C. include fried chicken, collard greens, and macaroni and cheese.</p><p>2. Ethiopian cuisine: Ethiopian food is known for its spicy and flavorful dishes, and it is a popular choice in Washington, D.C. Some popular dishes include injera (a sourdough flatbread), wat (a spicy vegetable stew), and kitfo (a spicy meat dish).</p><p>3. Seafood: Washington, D.C. is located near the Chesapeake Bay and has easy access to fresh seafood. Some popular seafood dishes include crab cakes, oysters, and seafood pasta.</p><p>4. BBQ: Washington, D.C. has a thriving barbecue scene, with many restaurants serving up slow-cooked meats and sides. Some popular BBQ spots include Ben’s Chili Bowl, Pete’s Soul Vegetarian, and Smoke & Barrel.</p><p>5. **International cuisine:** Washington, D.C. is a hub for international diplomacy, and as a result, it has a diverse culinary scene with many restaurants serving dishes from around the world. Some popular international cuisines include Japanese, Indian, and **Mexican.**</p><p></p>|
+|R3|Can you be more specific about the **fifth item** above?|<p>Sure, **the fifth item I mentioned, "Barbecue,"** refers to a style of cooking and food that originated in the southern United States. Barbecue is usually made with meat that is slow-cooked over an open flame or heated coals. The meat is often marinated and then sauced, and it is typically served with a variety of sides such as coleslaw, baked beans, and corn on the cob. Barbecue is a popular cuisine in Washington, D.C. and many of the restaurants serve pulled pork, ribs, and brisket. Some of the most popular barbecue restaurants in Washington, D.C. include:</p><p>- The Rib Pit</p><p>- The Smokehouse</p><p>- The BBQ Joint</p><p>- The Salt Line</p><p>- The Sovereign</p><p>These are just a few examples of the many barbecue restaurants in Washington, D.C. that offer delicious and flavorful barbecue dishes.</p>|<p>Sure! Here are some specific recommendations for **Mexican food** in Washington, D.C.:</p><p>1. Margarita’s - This popular Mexican restaurant in Adams Morgan has been serving up authentic Mexican cuisine for over 30 years. Some of their most popular dishes include their tacos al pastor (grilled marinated pork), enchiladas (meat or cheese filled tortillas), and chiles rellenos (stuffed peppers).</p><p>2. El Comal - This cozy little restaurant in the Columbia Heights neighborhood has been serving up traditional Mexican cuisine for over 20 years. Some of their most popular dishes include their mole (a rich, spicy sauce made from a blend of chiles, spices, and chocolate), tacos de carnitas (crispy pork tacos), and chiles rellenos de queso (cheese stuffed peppers).</p><p>3. Casa Oaxaca - This upscale Mexican restaurant in the Shaw neighborhood has a menu that features a wide range of traditional and modern dishes. Some of their most popular dishes include their ceviche (raw fish marinated in lime juice and spices), mole negro (a rich, dark mole sauce), and chiles en nogada (a spicy dish made with poblano peppers, walnuts, and honey).</p><p>4. Taco King - This popular taco chain has several locations in the Washington, D.C. area, including one in the Adams Morgan neighborhood. They offer a wide range of tacos, burritos, and other Mexican dishes at affordable prices. Some of their most popular items include their carne asada tacos, chicken tacos, and quesadillas.</p><p>5. Los Tacos No. 1 - This small, casual taco stand in the Columbia Heights neighborhood has a menu that features a wide range of tacos, burritos, and other Mexican dishes. Some of their most popular items include their adobada tacos (marinated pork tacos), al pastor tacos (grilled marinated pork tacos), and chicken tacos.</p>|
+
+
+**Experiment Details**
+
+Our dataset comes from real conversation history between humans and ChatGPT. We collect 183 English conversations with a total of 447 rounds of chats, where the original conversation history is already multi-round. For single-round NLL evaluation, we break a multi-round conversation into multiple single-round text pairs. For example, with a conversation history
+
+    Q1, A1
+
+    Q2, A2
+
+    Q3, A3
+
+where Q stands for Question, A stands for answer, we can get three single-round examples:
+
+    Context: Q1
+
+    Output: A1
+
+    Context: Q1, A1, Q2
+
+    Output: A2
+
+    Context: Q1, A1, Q2, A2, Q3
+
+    Ouptut: A3
+
+## CommonSense Performance
+Another important aspect of an LLM model is its common sense ability, where a model should acquire a certain level of factual knowledge and utilize them properly under different scenarios. Regarding this aspect of the ability, we found:
+
+1. *Finetuning on chat dataset results in commonsense degradation.*
+1. *Our Robin-Chat-7b model still achieves a competitive performance.*
+
+
+
+**Table 5: Accuracy results in commonsense QA benchmarks**
+
+
+||winogrand|boolq|arc\_easy|hellaswag|piqa|obqa|arc\_c|Average|
+| :- | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
+|Bloom-7.1b|64\.4|62\.9|65\.0|59\.6|73\.6|35\.8|33\.4|56\.4|
+|Bloom-7.1b-chat|60\.3|56\.8|61\.3|58\.7|72\.7|37\.8|38\.7|55\.2|
+|Llama-7b|67.9  |73\.2  |67.3   |73.0   |78\.4  |42\.4  |41\.4|62\.7|
+|Vicuna-7b|63.7 |77\.4  |63.1   |68.8   |76\.3  |39\.6  |38\.7|61\.1|
+|Vicuna-13b|66.2    |79\.9  |64.7   |73.0   |77\.6  |41\.6  |40\.4|63\.3|
+|Robin-Chat-7b|64.7 |75\.2  |69.8   |72.4   |76\.6  |39\.0  |42\.9|62\.9|
+
+
+
+**Table 6: NLL results in corpus of commonsense QA benchmarks**
+
+
+||winogrand|boolq|arc\_easy|hellaswag|piqa|obqa|arc\_c|Average|
+| :- | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
+|Bloom-7.1b|96.0|254|89|266|147|69|106\.5|146\.7|
+|Bloom-7.1b-chat|85.0|215|81\.5|237|130|62\.5|96|129\.5|
+|Llama-7b|79.5  |167    |71.5   |214    |121    |58 |85|113\.7|
+|Vicuna-7b|97.0 |219    |89 |258    |151    |72 |105|141\.6|
+|Vicuna-13b|95\.5   |207|88.5   |262    |149    |73 |105|140\.0|
+|Robin-Chat-7b|89\.0    |200    |81 |240    |139    |65\.5  |96\.5|130\.1|
+
+It is surprising to find that finetuning actually leads to commonsense degradation. A more well-known consequence of this issue is hallucination, where models tend to give plausible but factually incorrect or irrelevant answers. One explanation of this phenomenon is that LLMs “forget” the pretrained knowledge during finetuning when attempting to fit the finetune dataset.
+
+As one can see from the table above, both Vicuna and Robin suffer from this commonsense degradation issue. However, Robin suffers less. We conjecture that this improvement is brought by finetuning with smaller-sized but higher-quality dataset. Vicuna uses sharegpt data with ~900MB size and was finetuned for 3 epochs, while Robin uses multiple sources of dataset with ~600MB size and was only finetuned for 2 epochs. This suggests that in the finetuning stage, data quality plays a critical role.
+
+This improvement gives Vicuna-13b a run for its money when competing with Robin. The following examples show that Robin-Chat-7b possesses a comparable level of commonsense when compared with Vicuna-13b.
+
+**Table 7: Use cases of commonsense QA**
+
+
+|Question|Vicuna-13b|Robin-Chat-7b|Note|
+| :- | :-: | :-: | :-: |
+|<p>I go to a friend's home but the door is locked. However, I smell coal gas. Should I break the window?</p><p></p>|If you smell coal gas, it is important to get out of the house as quickly as possible and call the fire department or the gas company. Coal gas is a flammable gas that can be dangerous and even deadly if it is not properly ventilated. If the door is locked, you **should not break the window because it could cause a fire or explosion**. Instead, try to find another way to escape the house or call for help. If you cannot leave the house safely, try to stay calm and wait for help to arrive.|No, it is not recommended to break a window in this situation. **Breaking a window can be dangerous and can cause injury or damage to yourself or others**. Additionally, it is **illegal** to break a window to enter a property without permission. Instead, you should try to find another way to enter the home, such as by using a key or by calling the homeowner or a locksmith for assistance. If you are unable to enter the home, you should call the emergency services or the police for assistance. In the meantime, it is important to **avoid using any sources of heat**, such as a lighter or matches, as these can cause a gas leak or explosion. If you smell gas, you should evacuate the area immediately and call the emergency services or the gas company for assistance.|*Breaking the window helps the ventilation and will not cause a gas explosion directly. However, it is indeed illegal and may injure yourself when breaking the window.*|
+|Can I bring my pet ferret to that beautiful pacific island of USA to attend ICML 2023?|I'm sorry, but I am **not aware of** the specific policies regarding pets at the ICML 2023 conference or the island you are referring to. It would be best to check with the conference organizers or the local authorities on the island to determine if ferrets are allowed and if there are any necessary procedures or requirements for bringing a pet to the conference.|<p>It is **not recommended** to bring a pet ferret to the United States, including to attend ICML 2023. Ferrets are not permitted as pets in many areas of the United States, and **it is illegal** to bring them into the country without proper documentation and licensing. Additionally, ferrets can **carry diseases** that are not common in humans but can be dangerous to people, such as toxoplasmosis and salmonellosis. It is important to prioritize the health and safety of yourself and others by not bringing a pet ferret to a crowded event like ICML 2023.</p><p></p>|*ICML 2023 will be held in Hawaii, where ferrets are strictly prohibited as pets because they are potential carriers of the rabies virus.*|
+
+**Experiment Details**
+
+Here, we make use of  Eleuther AI’s LM evaluation harness repository (<https://github.com/EleutherAI/lm-evaluation-harness>) to get QA accuracy results. We also evaluate all models’ NLL metrics on their datasets, with their questions as contexts and answers as output sentences.
+
+## Instruction Following
+The final aspect we evaluate is instruction following, where a model is expected to loyally obey the user’s instructions, given the instruction being ethically or legally appropriate. It turns out that,
+
+1. *Without finetuning, foundation models do not perform well on instruction following.*
+1. *Finetuning on chat dataset helps improve the performance.*
+1. *Our Robin-Chat is still comparable with Vicuna-13b.*
+
+**Table 8: NLL on instruction following chat datasets**
+
+
+||gpt4\_en\_instruction|
+| :- | :-: |
+|Alpaca|222|
+|Llama-7b|214   |
+|Vicuna-7b|211  |
+|Vicuna-13b|206 |
+|Robin-Chat-7b|**180**  |
+
+Following examples show that Robin-Chat-7b can generate similar responses on instruction following tasks when compared with Vicuna-13b.
+
+**Table 9: Uses cases of instruction following**
+
+
+|Question|Vicuna-13b|Robin-Chat-7b|
+| :- | :-: | :-: |
+|Can you write quicksort?|<p>Yes, here is a simple implementation of the quicksort algorithm in Python:</p><p>def quicksort(arr):</p><p>`    `if len(arr) <= 1:</p><p>`        `return arr</p><p>`    `pivot = arr[len(arr) // 2]</p><p>`    `left = [x for x in arr if x < pivot]</p><p>`    `middle = [x for x in arr if x == pivot]</p><p>`    `right = [x for x in arr if x > pivot]</p><p>`    `return quicksort(left) + middle + quicksort(right)</p><p></p><p>This implementation uses the middle element of the array as the pivot, and then partitions the array into three sub-arrays: one containing all elements that are less than the pivot, one containing all elements that are equal to the pivot, and one containing all elements that are greater than the pivot. It then recursively applies the quicksort algorithm to the left and right sub-arrays, and returns the concatenation of the sorted left sub-array, the unsorted middle sub-array, and the sorted right sub-array.</p><p>This implementation has a time complexity of O(n log n) due to the recursive calls, although it could potentially be faster in practice if the pivot is carefully chosen.</p><p></p>|<p>Sure, here is a simple implementation of the quicksort algorithm in Python:</p><p>def quicksort(arr):</p><p>`    `if len(arr) <= 1:</p><p>`        `return arr</p><p>`    `pivot = arr[len(arr) // 2]</p><p>`    `left = [x for x in arr if x < pivot]</p><p>`    `middle = [x for x in arr if x == pivot]</p><p>`    `right = [x for x in arr if x > pivot]</p><p>`    `return quicksort(left) + middle + quicksort(right)</p><p></p><p>This implementation chooses the middle element of the array as the pivot, and then partitions the array into three parts: elements less than the pivot, elements equal to the pivot, and elements greater than the pivot. It then recursively applies the quicksort algorithm to the left and right subarrays, and concatenates the results with the middle subarray to produce the final sorted array.</p><p>Note that this implementation uses a single pass through the array, which is the most efficient way to sort an array using quicksort.</p>|
+|Instruction: Find synonyms for the given word. You need to write down how the provided synonyms differ from the original word in terms of meaning, usage, etc. adversity|<p>Synonyms for "adversity" include:</p><p>- Difficulty</p><p>- Hardship</p><p>- Trouble</p><p>- Misfortune</p><p>- Adversity</p><p>- Difficulties</p><p>- Hard times</p><p>All of these words refer to challenging or unpleasant circumstances or situations. However, "adversity" specifically connotes a difficult or hostile situation that is caused by external factors, such as an enemy or a natural disaster. "Difficulty" and "hardship" can be caused by external or internal factors, and can refer to a wide range of challenges, from mild inconveniences to major setbacks. "Trouble" often connotes a specific type of difficulty or distress, such as financial trouble or legal trouble. "Misfortune" specifically refers to bad luck or ill fate, while "hard times" refers to a general period of economic or social distress.</p>|<p>1. Hardship</p><p>2. Struggle</p><p>3. Difficulty</p><p>4. Oppression</p><p>5. Trial</p><p>6. Affliction</p><p>7. Distress</p><p>8. Calamity</p><p>9. Tribulation</p><p>10. Hardship</p><p>Adversity refers to difficult or challenging circumstances or conditions. Hardship, struggle, difficulty, oppression, trial, affliction, distress, and calamity are all words that can be used to describe adversity, but they often imply a more specific or intense situation. For example, hardship can refer to a general difficulty or challenge, while struggle implies a more intense battle or conflict. Difficulty and oppression imply a more specific obstacle or injustice, while trial and affliction suggest a more specific test or suffering. Trial can also refer to a legal proceeding, while tribulation implies a more general time of suffering or persecution.</p>|
+
+**Experiment Details**
+
+In order to make evaluations for models’ instruction following ability, we collected GPT-4 instruction data from the project “Instruction Tuning with GPT-4”, which provides many instruction tuning data labeled by GPT-4. We sample 1,000 English data and manually filtered examples with the following issues, where 767 effective samples remain after the filtering:
+
+1. Long response with too many nonsense words
+1. Incomplete input texts
+1. Specific domains involving chemistry/biology, where most LLM models do not possess the knowledge and always fail
+
+## Conclusion
+In this article, we introduce LMFlow’s evaluation framework, which uses NLL metric to reflect LLM models’ ability. NLL provides a good metric to evaluate different aspects of a LLM model. According to our evaluation results, Robin-7b achieves on-par performance when compared with Vicuna-13b. As our Robin-7b model is finetuned with different sources of dataset instead of sharegpt only, this shows that Vicuna can be further improved or surpassed with smaller-sized models and better dataset.
+
+The checkpoint of Robin-7b is now available for engineers and researchers to download and use (<https://github.com/OptimalScale/LMFlow#model-zoo>). Its effectiveness demonstrates that a multi-aspect evaluation is indeed essential to the development of LLMs.  
+
+## References
+Vicuna Chatbot Arena: <https://chat.lmsys.org/?arena> 
+
+lm-evaluation-harness: <https://github.com/EleutherAI/lm-evaluation-harness> 
+
+LMFlow: <https://github.com/OptimalScale/LMFlow>
diff --git a/_sources/blogs/index.md.txt b/_sources/blogs/index.md.txt
new file mode 100644
index 000000000..d71f731f2
--- /dev/null
+++ b/_sources/blogs/index.md.txt
@@ -0,0 +1,13 @@
+# Blogs
+
+## 2023
+
+
+```{toctree}
+:maxdepth: 1
+
+benchmark
+```
+
+
+
diff --git a/_sources/examples/DATASETS.md.txt b/_sources/examples/DATASETS.md.txt
new file mode 100644
index 000000000..6d41e08cc
--- /dev/null
+++ b/_sources/examples/DATASETS.md.txt
@@ -0,0 +1,385 @@
+# Dataset
+
+- [Dataset](#dataset)
+  - [Dataset Format in General](#dataset-format-in-general)
+  - [Supported Dataset and Detailed Formats](#supported-dataset-and-detailed-formats)
+    - [Conversation](#conversation)
+      - [Data Format](#data-format)
+      - [Conversation Template](#conversation-template)
+      - [Customize Conversation Template](#customize-conversation-template)
+    - [TextOnly](#textonly)
+    - [Text2Text](#text2text)
+    - [Paired Conversation](#paired-conversation)
+
+We provide several available datasets under `data`. You may download them all by running: 
+```sh
+cd data && ./download.sh all && cd -
+```
+You can replace `all` with a specific dataset name to only download that dataset (e.g. `./download.sh alpaca`).
+
+Customized datasets are strongly encouraged, since this way users can apply
+their own prompt engineering techniques over various source datasets. As long
+as the generated dataset following the format below, they can be accepted as
+the input of our pipelines :hugs:
+
+
+## Dataset Format in General
+
+To specify the input for model finetune, users can provide a list of `.json`
+files under a specified dataset directory. For example,
+
+```sh
+|- path_to_dataset
+  |- data_1.json
+  |- data_2.json
+  |- another_data.json
+  |- ...
+```
+
+For inference, we currently only support a single `.json` file.
+
+Each json file shall have the following format (three instances with four keys
+for example),
+
+```json
+{
+  "type": "TYPE",
+  "instances": [
+    {
+        "KEY_1": "VALUE_1.1",
+        "KEY_2": "VALUE_1.2",
+        "KEY_3": "VALUE_1.3",
+        "KEY_4": "VALUE_1.4",
+    },
+    {
+        "KEY_1": "VALUE_2.1",
+        "KEY_2": "VALUE_2.2",
+        "KEY_3": "VALUE_2.3",
+        "KEY_4": "VALUE_2.4",
+    },
+    {
+        "KEY_1": "VALUE_3.1",
+        "KEY_2": "VALUE_3.2",
+        "KEY_3": "VALUE_3.3",
+        "KEY_4": "VALUE_3.4",
+    },
+  ]
+}
+```
+
+where the `TYPE` indicates the dataset type and defines the set of keys
+`{ KEY_1, KEY_2, ... }` and their corresponding interpretations. The list of
+supported types are listed as follows.
+
+## Supported Dataset and Detailed Formats
+
+### Conversation
+
+#### Data Format
+
+Conversational data are commonly used in sft process. We currently support conversational data in ShareGPT format:
+
+````{dropdown} A conversation dataset
+```json
+{
+  "type": "conversation",
+  "instances": [
+    {
+      "conversation_id": "CONVERSATION_ID",
+      "system": "SYSTEM_PROPMT",
+      "tools": ["TOOL_DESCRIPTION_1","TOOL_DESCRIPTION_2","TOOL_DESCRIPTION_X"],
+      "messages": [
+        {
+            "role": "user",
+            "content": "USER_INPUT_1"
+        },
+        {
+            "role": "assistant",
+            "content": "ASSISTANT_RESPONSE_1"
+        },
+        {
+            "role": "user",
+            "content": "USER_INPUT_2"
+        },
+        {
+            "role": "assistant",
+            "content": "ASSISTANT_RESPONSE_2"
+        }
+      ]
+    },
+    {
+      "conversation_id": "CONVERSATION_ID",
+      "system": "SYSTEM_PROPMT",
+      "tools": ["TOOL_DESCRIPTION_1"],
+      "messages": [
+        {
+            "role": "user",
+            "content": "USER_INPUT_1"
+        },
+        {
+            "role": "assistant",
+            "content": "ASSISTANT_RESPONSE_1"
+        }
+      ]
+    }
+  ]
+}
+```
+````
+Data types:
+- `conversation_id`: `Optional[Any]`. An identifier for the conversation. `conversation_id` is only for convience of tracking the conversation and will not be used in the pipeline.
+- `system`: `Optional[string]`. A system prompt that is used to start the conversation.
+- `tools`: `Optional[List[string]]`. A list of tools that are used in the conversation.
+- `messages`: `List[Dict]`. A list of messages in the conversation. Each message contains the following fields:
+  - `role`: `string`. The role of the message. It can be either `user` or `assistant`.
+  - `content`: `string`. The content of the message.
+
+> We are working on supporting customized message keys and role names. Please stay tuned.
+
+Tips:
+- Please make sure the messages are:
+  1. Start with an user message.
+  2. In the correct order. The pipeline will not check the order of the messages.
+  3. In pairs of user and assistant (i.e., the length of the messages should be even). If the conversation ends with the user, the pipeline will trim the last user message.
+  4. Make sure the `content`s are not empty. If the `content` is empty, the pipeline will add a space to it.
+
+#### Conversation Template
+
+Conversations should be formatted before feeding into the model. As of now, we've preset the conversation template for following models:
+
+| Template Name | Filled Example | Detailed Template |
+| ------------- | -------------- | ----------------- |
+| `chatglm3` | `[gMASK]sop<\|system\|>`<br>` You are a chatbot developed by LMFlow team.<\|user\|>`<br>` Who are you?<\|assistant\|>`<br>` I am a chatbot developed by LMFlow team.<\|user\|>`<br>` How old are you?<\|assistant\|>`<br>` I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.` | [Link](./supported_conversation_template.md#chatglm3) |
+| `chatml` | `<\|im_start\|>system`<br>`You are a chatbot developed by LMFlow team.<\|im_end\|>`<br>`<\|im_start\|>user`<br>`Who are you?<\|im_end\|>`<br>`<\|im_start\|>assistant`<br>`I am a chatbot developed by LMFlow team.<\|im_end\|>`<br>`<\|im_start\|>user`<br>`How old are you?<\|im_end\|>`<br>`<\|im_start\|>assistant`<br>`I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<\|im_end\|>`<br> | [Link](./supported_conversation_template.md#chatml) |
+| `deepseek` | `<｜begin▁of▁sentence｜>You are a chatbot developed by LMFlow team.`<br><br>`User: Who are you?`<br><br>`Assistant: I am a chatbot developed by LMFlow team.<｜end▁of▁sentence｜>User: How old are you?`<br><br>`Assistant: I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<｜end▁of▁sentence｜>` | [Link](./supported_conversation_template.md#deepseek) |
+| `gemma` | `<bos>You are a chatbot developed by LMFlow team.<start_of_turn>user`<br>`Who are you?<end_of_turn>`<br>`<start_of_turn>model`<br>`I am a chatbot developed by LMFlow team.<end_of_turn>`<br>`<start_of_turn>user`<br>`How old are you?<end_of_turn>`<br>`<start_of_turn>model`<br>`I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<end_of_turn>`<br> | [Link](./supported_conversation_template.md#gemma) |
+| `hymba` | `<extra_id_0>System`<br>`You are a chatbot developed by LMFlow team.`<br>`<tool> {"name": "generate_qrcode", "description": "Generate a QR code for a given text", "parameters": {"type": "object", "properties": {"text": {"type": "string", "description": "The text to encode in the QR code"}}, "required": ["text"]}} </tool>`<br><br>`<extra_id_1>User`<br>`Who are you?`<br>`<extra_id_1>Assistant`<br>`I am a chatbot developed by LMFlow team.`<br>`<extra_id_1>User`<br>`How old are you?`<br>`<extra_id_1>Assistant`<br>`I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.</s>` | [Link](./supported_conversation_template.md#hymba) |
+| `internlm2` | `<s><\|im_start\|>system`<br>`You are a chatbot developed by LMFlow team.<\|im_end\|>`<br>`<\|im_start\|>user`<br>`Who are you?<\|im_end\|>`<br>`<\|im_start\|>assistant`<br>`I am a chatbot developed by LMFlow team.<\|im_end\|>`<br>`<\|im_start\|>user`<br>`How old are you?<\|im_end\|>`<br>`<\|im_start\|>assistant`<br>`I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<\|im_end\|>`<br> | [Link](./supported_conversation_template.md#internlm2) |
+| `llama3` | `<\|begin_of_text\|><\|start_header_id\|>system<\|end_header_id\|>`<br><br>`You are a chatbot developed by LMFlow team.<\|eot_id\|><\|start_header_id\|>user<\|end_header_id\|>`<br><br>`Who are you?<\|eot_id\|><\|start_header_id\|>assistant<\|end_header_id\|>`<br><br>`I am a chatbot developed by LMFlow team.<\|eot_id\|><\|start_header_id\|>user<\|end_header_id\|>`<br><br>`How old are you?<\|eot_id\|><\|start_header_id\|>assistant<\|end_header_id\|>`<br><br>`I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<\|eot_id\|>` | [Link](./supported_conversation_template.md#llama-3) |
+| `llama2` | `<s>[INST] <<SYS>>`<br>`You are a chatbot developed by LMFlow team.`<br>`<</SYS>>`<br><br>`Who are you? [/INST] I am a chatbot developed by LMFlow team.</s><s>[INST] How old are you? [/INST] I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.</s>` | [Link](./supported_conversation_template.md#llama-2) |
+| `phi3` | `<s><\|system\|>`<br>`You are a chatbot developed by LMFlow team.<\|end\|>`<br>`<\|user\|>`<br>`Who are you?<\|end\|>`<br>`<\|assistant\|>`<br>`I am a chatbot developed by LMFlow team.<\|end\|>`<br>`<\|user\|>`<br>`How old are you?<\|end\|>`<br>`<\|assistant\|>`<br>`I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<\|end\|>`<br>`<\|endoftext\|>` | [Link](./supported_conversation_template.md#phi-3) |
+| `qwen2` | `<\|im_start\|>system`<br>`You are a chatbot developed by LMFlow team.<\|im_end\|>`<br>`<\|im_start\|>user`<br>`Who are you?<\|im_end\|>`<br>`<\|im_start\|>assistant`<br>`I am a chatbot developed by LMFlow team.<\|im_end\|>`<br>`<\|im_start\|>user`<br>`How old are you?<\|im_end\|>`<br>`<\|im_start\|>assistant`<br>`I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<\|im_end\|>`<br> | [Link](./supported_conversation_template.md#qwen-2) |
+| `yi` | Same as `chatml` | [Link](./supported_conversation_template.md#yi) |
+| `yi1_5`| `You are a chatbot developed by LMFlow team.<\|im_start\|>user`<br>`Who are you?<\|im_end\|>`<br>`<\|im_start\|>assistant`<br>`I am a chatbot developed by LMFlow team.<\|im_end\|>`<br>`<\|im_start\|>user`<br>`How old are you?<\|im_end\|>`<br>`<\|im_start\|>assistant`<br>`I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<\|im_end\|>`<br> | [Link](./supported_conversation_template.md#yi-15) |
+| `zephyr` | `<\|system\|>`<br>`You are a chatbot developed by LMFlow team.</s>`<br>`<\|user\|>`<br>`Who are you?</s>`<br>`<\|assistant\|>`<br>`I am a chatbot developed by LMFlow team.</s>`<br>`<\|user\|>`<br>`How old are you?</s>`<br>`<\|assistant\|>`<br>`I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.</s>`<br> | [Link](./supported_conversation_template.md#zephyr) |
+
+Passing the template name to the `--conversation_template` argument to apply the corresponding conversation template:
+```sh
+# scripts/run_finetune.sh
+# ...
+deepspeed ${deepspeed_args} \
+  examples/finetune.py \
+    --model_name_or_path meta-llama/Llama-2-7b-chat-hf \
+    --dataset_path ${dataset_path} \
+    --conversation_template llama2 \
+# ...
+```
+
+`````{admonition} Formatted Dataset
+:class: info
+
+For dataset that system prompts, tool prompts and templates are already applied (like the one below), user can run the finetune shell by passing `empty` or `empty_no_special_tokens` to the `--conversation_template` argument. `empty` template will add a bos token to the beginning of every round of conversation as well as a eos token to the end of every round of conversation. `empty_no_special_tokens` will not add any special tokens to the conversation, just concatenates the user and assistant messages. 
+````{dropdown} A formatted dataset
+```json
+{
+  "type": "conversation",
+  "instances": [
+    {
+      "messages": [
+        {
+            "role": "user",
+            "content": "[INST] <<SYS>>\nYou are a helpful assistant.\n<</SYS>>\n\nHello! [/INST]"
+        },
+        {
+            "role": "assistant",
+            "content": "Hi, how are you?"
+        },
+        {
+            "role": "user",
+            "content": "[INST] Good. [/INST]"
+        },
+        {
+            "role": "assistant",
+            "content": "Glad to hear that."
+        }
+      ]
+    },
+    {
+      "messages": [
+        {
+            "role": "user",
+            "content": "[INST] <<SYS>>\nYou are a helpful assistant.\n<</SYS>>\n\nWhat's the weather like now? [/INST]"
+        },
+        {
+            "role": "assistant",
+            "content": "I'm sorry for any confusion, but as an AI, I don't have access to real-time data such as current weather conditions."
+        }
+      ]
+    }
+  ]
+}
+```
+````
+`````
+
+#### Customize Conversation Template
+
+Please refer to the [Customize Conversation Template](./customize_conversation_template.md) for more details.
+
+
+### TextOnly
+
+This is the most common dataset type, which only contains raw texts in each
+sample. This type of dataset can be used as the training set for text decoder
+models, or the input of decoder models / encoder-decoder models. Its format is
+as follows (three instances for example),
+
+````{dropdown} A textonly dataset
+```json
+{
+  "type": "text_only",
+  "instances": [
+    {  "text": "SAMPLE_TEXT_1" },
+    {  "text": "SAMPLE_TEXT_2" },
+    {  "text": "SAMPLE_TEXT_3" },
+  ]
+}
+```
+````
+
+For example, `data/example_dataset/train/train_50.json` has the aboved format.
+
+
+### Text2Text
+
+This is the dataset type mostly used for inferencing, which contains a pair of
+texts in each sample. This type of dataset can be used as the training set for
+text encoder-decoder models, or question-answer pair for evaluating model
+inferences. Its format is as follows (three instances for example):
+
+````{dropdown} A text2text dataset
+```json
+{
+  "type": "text2text",
+  "instances": [
+    {
+        "input": "SAMPLE_INPUT_1",
+        "output": "SAMPLE_OUTPUT_1",
+    },
+    {
+        "input": "SAMPLE_INPUT_2",
+        "output": "SAMPLE_OUTPUT_2",
+    },
+    {
+        "input": "SAMPLE_INPUT_3",
+        "output": "SAMPLE_OUTPUT_3",
+    },
+  ]
+}
+```
+````
+
+For example, `data/example_dataset/test/test_13.json` has the aboved format.
+
+
+### Paired Conversation
+
+```{admonition} **Work in Progress**
+:class: info
+
+We are working on paired conversation dataset and will update it soon.  
+```
+
+This type of dataset are commonly used for alignment such as [reward modeling](https://arxiv.org/abs/2203.02155), 
+[Direct Preference Optimization (DPO)](https://arxiv.org/abs/2305.18290), etc. For requirements of the conversations,
+please refer to [conversation data](#conversation).
+
+````{dropdown} A paired conversation dataset
+```json
+{
+    "type": "paired_conversation",
+    "instances": [
+        {
+            "chosen": {
+                "conversation_id": "CONVERSATION_ID",
+                "system": "SYSTEM_PROPMT",
+                "tools": ["TOOL_DESCRIPTION_1","TOOL_DESCRIPTION_2","TOOL_DESCRIPTION_3"],
+                "messages": [
+                    {
+                        "role": "user",
+                        "content": "USER_INPUT_1"
+                    },
+                    {
+                        "role": "assistant",
+                        "content": "ASSISTANT_RESPONSE_1_GOOD"
+                    },
+                    {
+                        "role": "user",
+                        "content": "USER_INPUT_2"
+                    },
+                    {
+                        "role": "assistant",
+                        "content": "ASSISTANT_RESPONSE_2_GOOD"
+                    }
+                ]
+            },
+            "rejected": {
+                "conversation_id": "CONVERSATION_ID",
+                "system": "SYSTEM_PROPMT",
+                "tools": ["TOOL_DESCRIPTION_1","TOOL_DESCRIPTION_2","TOOL_DESCRIPTION_3"],
+                "messages": [
+                    {
+                        "role": "user",
+                        "content": "USER_INPUT_1"
+                    },
+                    {
+                        "role": "assistant",
+                        "content": "ASSISTANT_RESPONSE_1_BAD"
+                    },
+                    {
+                        "role": "user",
+                        "content": "USER_INPUT_2"
+                    },
+                    {
+                        "role": "assistant",
+                        "content": "ASSISTANT_RESPONSE_2_BAD"
+                    }
+                ]
+            }
+        },
+        {
+            "chosen": {
+                "conversation_id": "CONVERSATION_ID",
+                "system": "SYSTEM_PROPMT",
+                "tools": ["TOOL_DESCRIPTION_1"],
+                "messages": [
+                    {
+                        "role": "user",
+                        "content": "USER_INPUT_1"
+                    },
+                    {
+                        "role": "assistant",
+                        "content": "ASSISTANT_RESPONSE_1_GOOD"
+                    }
+                ]
+            },
+            "rejected": {
+                "conversation_id": "CONVERSATION_ID",
+                "system": "SYSTEM_PROPMT",
+                "tools": ["TOOL_DESCRIPTION_1"],
+                "messages": [
+                    {
+                        "role": "user",
+                        "content": "USER_INPUT_1"
+                    },
+                    {
+                        "role": "assistant",
+                        "content": "ASSISTANT_RESPONSE_1_BAD"
+                    }
+                ]
+            }
+        }
+    ]
+}
+```
+````
\ No newline at end of file
diff --git a/_sources/examples/TASK_GUIDE.md.txt b/_sources/examples/TASK_GUIDE.md.txt
new file mode 100644
index 000000000..afbfd3ca8
--- /dev/null
+++ b/_sources/examples/TASK_GUIDE.md.txt
@@ -0,0 +1,132 @@
+# LMFlow Benchmark Guide
+
+We support two ways to add evaluation settings in our repo, `NLL Task Setting` and `LM-Evaluation Task Setting`. Below are the details of them: 
+
+# 1. NLL Task Setting
+Users can easily create new tasks and evaluate their datasets on 
+the provide `nll (Negative Log Likelihood)` metric. 
+
+## Setup
+
+Fork the main repo, clone it, and create a new branch with the name of 
+your task, and install the following:
+
+```bash
+# After forking...
+git clone https://github.com/<YOUR-USERNAME>/LMFlow.git
+cd LMFlow
+git checkout -b <TASK-NAME>
+conda create -n lmflow python=3.9 -y
+conda activate lmflow
+conda install mpi4py
+pip install -e .
+```
+## Create Your Task Dataset File
+We provide several available datasets under `data` after running
+```sh
+cd data && ./download.sh && cd -
+```
+
+You can refer to some given evaluation dataset files and create your own.
+Also, you may refer to our guide on 
+[DATASET](https://optimalscale.github.io/LMFlow/examples/DATASETS.html).
+
+In this step, you will need to decide your answer type like `text2text` 
+or `text_only` (Notice that the current `nll` implementation only supports these 
+two answer types). We will note the chosen answer type as `<ANSWER_TYPE>`.
+
+After preparing your own `DATASET` file, you can put it under `data` dir
+and make a `TASK` dir.
+
+```bash
+mkdir <TASK>
+mv <DATASET> <TASK>
+```
+
+## Task Registration
+
+Note the path of your dataset, `data/<TASK>/<DATASET>`.
+
+Open the file `examples/benchmarking.py`, add your task's info into 
+`LOCAL_DATSET_GROUP_MAP`, `LOCAL_DATSET_MAP`, `LOCAL_DATSET_ANSWERTYPE_MAP`
+
+In `LOCAL_DATSET_MAP`, you will need to specify your `DATASET` files' path:
+
+```python
+LOCAL_DATSET_MAP ={
+    "...":"...",
+    "<TASK>":"data/<TASK>/<DATASET>",
+}
+```
+
+In `LOCAL_DATSET_ANSWERTYPE_MAP`, you will need to specify your task's 
+`<ANSWER_TYPE>`:
+
+```python
+LOCAL_DATSET_ANSWERTYPE_MAP ={
+    "...":"...",
+    "<TASK>":"<ANSWER_TYPE>,
+}
+```
+
+If you only have one task, you can add key-value pair like `"<TASK>":"<TASK>"`
+in `LOCAL_DATSET_GROUP_MAP`:
+```python
+LOCAL_DATSET_GROUP_MAP ={
+    "...":"...",
+    "<TASK>":"<TASK>",
+}
+```
+
+
+If you want to combine several tasks, you may first specify a 
+combination name `<TASK_COMBINATION>` and add key-value pair like
+`"<TASK_COMBINATION>":"<TASK_1>,<TASK_2>,.."`in `LOCAL_DATSET_GROUP_MAP`.
+
+Remember to separate TASK by `,`:
+```python
+LOCAL_DATSET_GROUP_MAP ={
+    "...":"...",
+    "<TASK_COMBINATION>":"<TASK_1>,<TASK_2>,..",
+}
+```
+
+After finishing changing these items, you can run your own `<TASK>` like:
+
+```bash
+deepspeed examples/benchmarking.py \
+  --answer_type <ANSWER_TYPE> \
+  --use_ram_optimized_load False \
+  --model_name_or_path ${model_name} \
+  --dataset_name data/<TASK>/<DATASET>\
+  --deepspeed examples/ds_config.json \
+  --metric nll \
+  --prompt_structure "###Human: {input}###Assistant:" \
+  | tee ${log_dir}/train.log \
+  2> ${log_dir}/train.err 
+```
+
+# 2. LM-Evaluation Task Setting
+
+We integrate [EleutherAI/lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness) into 
+`benchamrk.py` by directly executing the evaluate commands. Users 
+can also use their evaluation by simply changing two items in
+`<LM_EVAL_DATASET_MAP>` of `examples/benchmarking.py`. 
+
+Please refer to Eleuther's 
+[task-table](https://github.com/EleutherAI/lm-evaluation-harness/blob/master/docs/task_table.md)
+to get exact `<TASK>` name.
+
+Similarly, you can combine several tasks, you may first specify a 
+combination name `<TASK_COMBINATION>` and add key-value pair like
+`"<TASK_COMBINATION>":"<TASK_1>,<TASK_2>,.."`in `LM_EVAL_DATASET_MAP`.
+
+Also, remember to separate TASK by `,`:
+
+```python
+LM_EVAL_DATASET_MAP ={
+    "...":"...",
+    "<TASK_COMBINATION>":"<TASK_1>,<TASK_2>,..",
+}
+```
+
diff --git a/_sources/examples/checkpoints.md.txt b/_sources/examples/checkpoints.md.txt
new file mode 100644
index 000000000..4c36eac49
--- /dev/null
+++ b/_sources/examples/checkpoints.md.txt
@@ -0,0 +1,31 @@
+# Checkpoints
+
+In general, you can directly load from checkpoints by using `--model_name_or_path`. However, the LLaMA case is slightly different due to the copyright issue.
+
+
+## LLaMA Checkpoint
+
+1. First, you need to get the access of LLaMA model from [facebookresearch/llama](https://github.com/facebookresearch/llama). Download the official checkpoints and save them into `${llama-path}`.
+
+2. Second, convert the official checkpoints `${llama-path}` to HuggingFace supported checkpoints `${llama-hf-path}` by running
+
+    `python ./scripts/convert_llama_weights_to_hf.py --input_dir ${llama-path} --model_size 7B --output_dir ${llama-hf-path}/llama-7b-hf`
+
+3. Then you are good to go by setting the checkpoint path to `${llama-hf-path}/llama-7b-hf`. Enjoy it!
+
+4. (optional) Now you have the original llama-7b-hf pretrained model. With
+```sh
+cd output_models && ./download.sh all && cd -
+```
+You can obtain the model difference finetuned by ours. By a way similar to `./scripts/run_evaluation_with_lora.sh`,
+```sh
+CUDA_VISIBLE_DEVICES=0 \
+    deepspeed examples/evaluate.py \
+    --answer_type text \
+    --model_name_or_path ${llama-hf-path}/llama-7b-hf \
+    --lora_model_path output_models/${llama-model-diff-path} \
+    --dataset_path data/alpaca/test \
+    --prompt_structure "Input: {input}" \
+    --deepspeed examples/ds_config.json
+```
+You can now evaluate with the finetuned llama model.
\ No newline at end of file
diff --git a/_sources/examples/customize_conversation_template.md.txt b/_sources/examples/customize_conversation_template.md.txt
new file mode 100644
index 000000000..a129f0f3f
--- /dev/null
+++ b/_sources/examples/customize_conversation_template.md.txt
@@ -0,0 +1,115 @@
+# Customize Conversation Template
+
+> For beginners: Why template?   
+> Almost all LLMs today do a simple job - predict the next "word". To make the interaction between user and model smoother, developers use tricks: they add special "words" to the input text (at back-end, thus invisible to the user when using services like ChatGPT) to "tell" the model what user had said before, and ask the model to respond like an assistant. These "hidden words" are called "template".
+
+We provide the flexibility to customize the conversation template. You can customize your own conversation template by following the steps below:
+
+### 1. Decompose your conversations
+Say you want to make the conversations between user and assistant look like:  
+
+```
+<bos>System:
+You are a chatbot developed by LMFlow team.
+
+User:
+Who are you?
+
+Assistant:
+I am a chatbot developed by LMFlow team.<eos>
+
+User:
+How old are you?
+
+Assistant:
+I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<eos>
+```
+
+It is easy to abstract the format for each message:
+- System message: `System:\n{{content}}\n\n`  
+- User message: `User:\n{{content}}\n\n`  
+- Assistant message: `Assistant:\n{{content}}\n\n<eos>`  
+
+Also, we have a bos token at the beginning of the conversation session.
+
+###  2. Choose proper `Formatter`  
+Recall the requirements for a conversation dataset:  
+> - `system`: `Optional[string]`. 
+> - `tools`: `Optional[List[string]]`.  
+> - `messages`: `List[Dict]`.  
+>    - `role`: `string`.  
+>    - `content`: `string`.  
+
+System message, user message, and assistant message are strings thus we can use `StringFormatter` for them.
+
+### 3. Build the template
+All preset templates are located at `src/lmflow/utils/conversation_template`.
+
+Within the template file, define your own template like:
+
+```python
+from .base import StringFormatter, TemplateComponent, ConversationTemplate
+
+
+YOUR_TEMPLATE = ConversationTemplate(
+    template_name='your_template_name',
+    user_formatter=StringFormatter(
+        template=[
+            TemplateComponent(type='string', content='User:\n{{content}}\n\n')
+        ]
+    ),
+    assistant_formatter=StringFormatter(
+        template=[
+            TemplateComponent(type='string', content='Assistant:\n{{content}}\n\n'),
+            TemplateComponent(type='token', content='eos_token') # this will add the eos token at the end of every assistant message
+            # please refer to the docstring of the `TemplateComponent` class to 
+            # see the difference between different types of components.
+        ]
+    ),
+    system_formatter=StringFormatter(
+        template=[
+            TemplateComponent(type='string', content='System:\n{{content}}\n\n')
+        ]
+    )
+    # For models that has ONLY ONE bos token at the beginning of 
+    # a conversation session (not a conversation pair), user can
+    # specify a special starter to add that starter to the very
+    # beginning of the conversation session. 
+    # eg:
+    #   llama-2: <s> and </s> at every pair of conversation 
+    #   v.s.
+    #   llama-3: <|begin_of_text|> only at the beginning of a session
+    special_starter=TemplateComponent(type='token', content='bos_token'),
+
+    # Similar to the special starter... (just for illustration, commented out 
+    # since it is not necessary for our purposed template above)
+    # special_stopper=TemplateComponent(type='token', content='eos_token')
+)
+```
+
+Feel free to create your own template by inheriting the `ConversationTemplate` class. Llama-2 v.s. llama-3 would be a good examples to refer to.
+
+### 4. Register your template
+After defining your own template, you need to register it in the `src/lmflow/utils/conversation_template/__init__.py` file. 
+
+```python
+# ...
+from .your_template_file import YOUR_TEMPLATE
+
+
+PRESET_TEMPLATES = {
+    #...
+    'your_template_name': YOUR_TEMPLATE,
+}
+```
+
+### 5. Use your template
+You are all set! Specify the template name in, for example, your finetune script:
+
+```bash
+./scripts/run_finetune.sh \
+    --model_name_or_path path_to_your_model \
+    --dataset_path your_conversation_dataset \
+    --conversation_template your_template_name \
+    --output_model_path output_models/your_model
+```
\ No newline at end of file
diff --git a/_sources/examples/finetuning.md.txt b/_sources/examples/finetuning.md.txt
new file mode 100644
index 000000000..821a9a594
--- /dev/null
+++ b/_sources/examples/finetuning.md.txt
@@ -0,0 +1,101 @@
+# Finetuning 
+
+## Full Parameters
+
+Full training updates all the parameters to finetune a language model.
+Here is an example to finetune a GPT-2 base model.
+
+```sh
+cd data && ./download.sh alpaca && cd -
+
+./scripts/run_finetune.sh \
+    --model_name_or_path gpt2 \
+    --dataset_path data/alpaca/train_conversation \
+    --output_model_path output_models/finetuned_gpt2
+```
+
+```{admonition} Conversation Template
+:class: tip
+
+For conversation dataset, specify a conversation template for better performance by adding `--conversation_template` to the command.  
+```
+
+````{dropdown} Llama-3-8B conversation dataset example
+```sh
+cd data && ./download.sh alpaca && cd -
+
+./scripts/run_finetune.sh \
+    --model_name_or_path meta-llama/Meta-Llama-3-8B \
+    --dataset_path data/alpaca/train_conversation \
+    --conversation_template llama3 \
+    --output_model_path output_models/finetuned_llama3_8b
+```
+````
+
+
+## Layerwise Importance Sampled AdamW (LISA)
+
+[LISA](https://arxiv.org/abs/2403.17919) is a memory-efficient finetuning algorithm that allows tradeoff between memory and the number of randomly unfreezed layers. This script currently is only tested in single gpus. Please stay tuned for our latest updates!
+
+```sh
+cd data && ./download.sh alpaca && cd -
+
+./scripts/run_finetune_with_lisa.sh \
+    --model_name_or_path meta-llama/Llama-2-7b-hf \
+    --dataset_path data/alpaca/train_conversation \
+    --output_model_path output_models/finetuned_llama2_7b \
+    --lisa_activated_layers 1 \
+    --lisa_interval_steps 20
+```
+
+````{dropdown} Llama-2-7B conversation dataset example
+```sh
+cd data && ./download.sh alpaca && cd -
+
+./scripts/run_finetune_with_lisa.sh \
+    --model_name_or_path meta-llama/Llama-2-7b-hf \
+    --dataset_path data/alpaca/train_conversation \
+    --conversation_template llama2 \
+    --output_model_path output_models/finetuned_llama2_7b_lisa \
+    --lisa_activated_layers 1 \
+    --lisa_interval_steps 20
+```
+````
+
+
+## Low-Rank Adaptation (LoRA)
+
+LoRA is a parameter-efficient finetuning algorithm and is more efficient than full finetuning.
+
+```sh
+cd data && ./download.sh alpaca && cd -
+
+./scripts/run_finetune_with_lora.sh \
+    --model_name_or_path facebook/galactica-1.3b \
+    --dataset_path data/alpaca/train_conversation \
+    --output_lora_path output_models/finetuned_galactica_lora
+```
+
+````{admonition} Merge LoRA Weight
+:class: tip
+
+Merge LoRA weight and the base model into one using:  
+```sh
+./scripts/run_merge_lora.sh \
+    --model_name_or_path Qwen/Qwen1.5-1.8B \
+    --lora_model_path output_models/lora \
+    --output_model_path output_models/lora_merged \
+```
+````
+
+````{dropdown} Llama-2-7B conversation dataset example
+```sh
+cd data && ./download.sh alpaca && cd -
+
+./scripts/run_finetune_with_lora.sh \
+    --model_name_or_path meta-llama/Llama-2-7b-hf \
+    --dataset_path data/alpaca/train_conversation \
+    --conversation_template llama2 \
+    --output_model_path output_models/finetuned_llama2_7b_lora \
+```
+````
\ No newline at end of file
diff --git a/_sources/examples/index.md.txt b/_sources/examples/index.md.txt
new file mode 100644
index 000000000..30dc1d335
--- /dev/null
+++ b/_sources/examples/index.md.txt
@@ -0,0 +1,57 @@
+# Examples
+
+We provide several examples to show how to use our package in your problem.
+
+## Data preparation
+
+```{toctree}
+:maxdepth: 3
+
+DATASETS
+```
+
+```{toctree}
+:maxdepth: 3
+
+checkpoints
+```
+
+## Finetuning
+
+For SFT,
+
+```{toctree}
+:maxdepth: 3
+
+finetuning
+```
+
+
+For alignment process, 
+
+```{toctree}
+:maxdepth: 3
+
+reward_modeling
+```
+
+
+```{toctree}
+:maxdepth: 3
+
+raft
+```
+
+## Inference
+
+Refer to [examples](https://github.com/OptimalScale/LMFlow/blob/main/examples).
+
+## Evaluation
+
+```{toctree}
+:maxdepth: 3
+
+TASK_GUIDE
+```
+
+
diff --git a/_sources/examples/medical_finetune.md.txt b/_sources/examples/medical_finetune.md.txt
new file mode 100644
index 000000000..327e062be
--- /dev/null
+++ b/_sources/examples/medical_finetune.md.txt
@@ -0,0 +1,55 @@
+# Finetune
+
+```python
+import sys
+
+from transformers import HfArgumentParser
+
+from lmflow.args import (
+    ModelArguments,
+    DatasetArguments,
+    AutoArguments,
+)
+
+from lmflow.datasets.dataset import Dataset
+from lmflow.models.tunable_models import TunableModel
+from lmflow.pipeline.auto_pipeline import AutoPipeline
+
+
+def main():
+    # Parses arguments
+    pipeline_name = "finetuner"
+    PipelineArguments = AutoArguments.get_pipeline_args_class(pipeline_name)
+
+    parser = HfArgumentParser((ModelArguments, DatasetArguments, PipelineArguments))
+    if len(sys.argv) == 2 and sys.argv[1].endswith(".json"):
+        # If we pass only one argument to the script and it's the path to a json file,
+        # let's parse it to get our arguments.
+        model_args, data_args, pipeline_args = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1]))
+    else:
+        model_args, data_args, pipeline_args = parser.parse_args_into_dataclasses()
+
+    # TODO: deepspeed config initialization
+
+    # Initialization
+    finetuner = AutoPipeline.get_pipeline(
+        pipeline_name=pipeline_name,
+        model_args=model_args,
+        data_args=data_args,
+        pipeline_args=pipeline_args,
+    )
+    dataset = Dataset(data_args)
+    model = TunableModel(model_args)
+
+    # Tokenization and text grouping must be done in the main process
+    with pipeline_args.main_process_first(desc="dataset map tokenization"):
+        tokenized_dataset = model.tokenize(dataset)
+        lm_dataset = finetuner.group_text(
+            tokenized_dataset,
+            model_max_length=model.get_max_length(),
+        )
+
+    # Finetuning
+    tuned_model = finetuner.tune(model=model, lm_dataset=lm_dataset)
+
+```
diff --git a/_sources/examples/raft.md.txt b/_sources/examples/raft.md.txt
new file mode 100644
index 000000000..bb36b3c55
--- /dev/null
+++ b/_sources/examples/raft.md.txt
@@ -0,0 +1,342 @@
+# RAFT
+## 1 Introduction
+
+We remark that the example is built on LLaMA whose [licensed](https://docs.google.com/forms/d/e/1FAIpQLSfqNECQnMkycAp2jP4Z9TFX0cGR4uf7b_fBxjY_OjhJILlKGA/viewform) is for non-commercial use only.
+
+
+
+Reinforcement Learning from Human Feedback (RLHF) requires a reward function to guide the adjustment of the generative model. In this example, we show how to use LMFlow framework to train a reward model following the procedure in the InstructGPT paper: https://arxiv.org/abs/2203.02155 and then align the model via the RAFT algorithm (Reward rAnked FineTuning). 
+
+
+
+**This example contains both reward modeling and RAFT alignment for completeness. For users' convenience, we have already provided a reward model based on GPT-Neo-2.7B in huggingface repo so one can skip the reward modeling first.**
+
+### 1.1 Dataset description
+
+We use the *Dahoas/full-hh-rlhf* dataset as an example, where each sample of this dataset consists of a prompt and two responses from the assistant. In particular, the response with label "chosen" is preferred as compared to the response with label "rejected". The dataset consists of 112K training samples and 12.5K test samples. The following is an example sample of the dataset:
+
+```sh
+" Human: What kind of noises did dinosaurs make? Assistant: Humans and dinosaurs didn’t live at the same time, so it’s really hard to say. The best place to find out what noises dinosaurs made would be Human: yes they did Assistant: to guess, and that would probably require lots of reading and a certain amount of imagination, so we’re not really prepared to do that. Human: you cant read Assistant: 
+
+Chosen response: "You can read?"
+
+Rejected response: "there’s a lot of stuff humans don’t know"
+```
+
+To facilitate the training, we reformulate the prompt by adding ``\#\#\#'' at the beginning of the characters so that the model knows to reply. The new sample will be of the form:
+
+```sh
+"###Human: What kind of noises did dinosaurs make? ###Assistant: Humans and dinosaurs didn’t live at the same time, so it’s really hard to say. The best place to find out what noises dinosaurs made would be ###Human: yes they did ###Assistant: to guess, and that would probably require lots of reading and a certain amount of imagination, so we’re not really prepared to do that. ###Human: you cant read ###Assistant: 
+
+Chosen response: "You can read?"
+
+Rejected response: "there’s a lot of stuff humans don’t know"
+```
+
+
+
+We prepare all the used dataset in the directory  ./data/hh_rlhf, which can be obtained by running the following command at LMFlow
+
+```sh
+cd data && ./download.sh hh_rlhf && cd -
+```
+
+
+
+## 2 Reward Modeling
+
+We follow the the procedure in the InstructGPT paper: https://arxiv.org/abs/2203.02155 to train a reward model using the HH-RLHF dataset first, which includes
+
+- Supervised Finetuning (SFT);
+- Reward modeling by comparison dataset.
+
+
+
+### 2.1 Supervised Finetuning (SFT)
+
+Here is an example of dataset */home/xiongwei/LMFlow/data/hh_rlhf/sft/hh_rlhf_sft.json*. We use only the preferred responses so we get 112K training samples.
+
+
+
+
+```json
+  
+{"type": "text_only",  
+"instances":   
+[  
+    {"text": "###Human: Should you buy a case to protect your cell phone?###Assistant: It depends on your circumstances.  If you carry your phone in a pocket or a purse then you probably want a case.  But if you only need a phone for quick interactions, a case may actually cause more harm than good.  What do you need the phone for?  Are you a parent, or do you work from home?###Human: What harm could it do?###Assistant: A phone case can damage the screen, for one thing.  It can also get you in trouble if you have your phone turned off for some reason.  Then you will turn it back on and it won’t do anything.  If you can afford to replace it, then you need a case to protect it.  The problem is that most people aren’t able to afford to replace their phones all the time.###Human: Thanks for letting me know.###Assistant: You’re welcome."},  
+    {"text": "###Human: I'm trying to learn about the salam witch trials###Assistant: If you’re looking for books about witchcraft trials, I can recommend some books for you. But you probably mean you’re looking for more historical information about the Salem witch trials in 1692, and specifically about the salam witch trials in 1692?###Human: What are some good books to learn about the salam witch trials###Assistant: What would you like to learn?  If you’re interested in history, one of the best books is The Witch Persecutions in Colonial America: A History.  If you’re interested in witchcraft as a cultural phenomenon, you might enjoy two excellent books: Religion and the Decline of Magic: Studies in Popular Beliefs in Sixteenth- and Seventeenth-Century England by Keith Thomas and Magic, Witchcraft, and the Otherworld: An Anthropology of Superstition by Jack Goody.  If you’re interested in history specifically as it relates to religion, you might enjoy The Popish Plot, or Prelates' Plot: A History of the Popish Plot in England, by K. J. Everett."}  
+]  
+}  
+```
+
+
+
+You can edit */scripts/run_finetune.sh* and modify the parameters. We present an example with GPT-Neo-2.7B here.
+
+- --model_name_or_path: EleutherAI/gpt-neo-2.7B
+- --dataset_path: ${project_dir}/data/hh_rlhf/sft
+- --output_dir: the path you want to store the sft model
+- --num_train_epochs: 1
+- --learning_rate: 2e-5
+- --per_device_train_batch_size: adjust according to your GPU memory source.
+
+
+
+Then, we can run the following command to perform SFT.
+
+```sh
+./scripts/run_finetune.sh
+```
+
+You can also use lora training by the following command but also need to set the model_name_or_path and dataset appropriately by editing run_finetune_with_lora.sh .
+
+```sh
+./scripts/run_finetune_with_lora.sh
+```
+
+
+
+### 2.2 Reward Modeling
+
+**Data Preparation** 
+
+Reward modeling involves ranking responses based on the given prompt, typically requiring human labeling. Fortunately, we already have two responses for each prompt in this example, with one preferred over the other. To create a new dataset, we modify the original dataset to include two keys, "positive" and "negative", with the former being the preferred response. 
+
+
+
+Here is an example of the dataset *LMFlow/data/hh_rlhf/rm/hh_rlhf_rm_training.json*, consisting of two samples.
+
+```sh
+{"instances": [
+    {"positive": "###Human: Are space heaters safe?###Assistant: What do you mean? Can you give me more details?###Human: I mean, can they be a fire hazard?###Assistant: I don’t know about that.  Can you tell me more about the situation? Are you using the heater on a fireplace?  Do you have any other heater in the room?  What materials are in the room?  Do you feel any heat from the heater?  How are you touching the heater?###Human: Can you just tell me if there are any situations where they could be a fire hazard?###Assistant: The answer is “it depends”.  It would depend on the specific heater you are using, how you are using it, what kind of fuel it uses, what materials you have in the room, and what the surroundings are like.  You’d also have to make sure you’re using it correctly.  Is it switched off?  Is it safe to touch?  It’s not the sort of thing I could just answer by looking at the text you provided.###Human: Thank you. That information is helpful.###Assistant: Glad to be helpful.  I’m always happy to assist.", "negative": "###Human: Are space heaters safe?###Assistant: What do you mean? Can you give me more details?###Human: I mean, can they be a fire hazard?###Assistant: I don’t know about that.  Can you tell me more about the situation? Are you using the heater on a fireplace?  Do you have any other heater in the room?  What materials are in the room?  Do you feel any heat from the heater?  How are you touching the heater?###Human: Can you just tell me if there are any situations where they could be a fire hazard?###Assistant: The answer is “it depends”.  It would depend on the specific heater you are using, how you are using it, what kind of fuel it uses, what materials you have in the room, and what the surroundings are like.  You’d also have to make sure you’re using it correctly.  Is it switched off?  Is it safe to touch?  It’s not the sort of thing I could just answer by looking at the text you provided.###Human: Thank you. That information is helpful.###Assistant: You’re welcome. I’m glad to help. Is there anything else you want to know?"}, 
+    {"positive": "###Human: how do i teach my kids to make forts at home?###Assistant: Do you mean you want to teach them to build a simple structure they can hide in? Or something more elaborate?  I don’t know exactly what you mean by a “fort”, but maybe this gives you some ideas?", "negative": "###Human: how do i teach my kids to make forts at home?###Assistant: There are many ways to make forts!  The simplest might be to just pile up all the furniture in one room.  This works if you have a single room in your house, although it might still be a bit challenging to build a taller and sturdier fort this way.  The best way to build forts is to start with something easy, and then have fun exploring how you can improve it."}
+]}
+```
+
+
+
+**Training**
+
+To start from a model from a previous sft step, you may edit the *./scripts/run_reward_modeling.sh* script and update the "dataset_path" to use the desired dataset. By default, we will use LoRA training in reward modeling as we found that it is superior in terms of optimization and gives comparable performance.
+
+
+
+- --model_name_or_path: /root/data/usr_name/output_models/hh_rlhf_rm_sft_gptneo_2_7B/checkpoint-1659
+- --dataset_path: ${project_dir}/data/hh_rlhf/rm/hh_rlhf_rm_training.json
+- --output_dir: the path you want to store the reward model
+- --num_train_epochs: 1
+- --learning_rate: 3e-5
+- --per_device_train_batch_size: adjust according to your GPU memory source.
+- --eval_steps: 400
+- --validation_split_percentage: 10
+
+
+
+The load_dataset function splits the dataset into training and evaluation sets, which can also be customized by editing the function in /examples/run_reward_modeling.py if you want to prepare your own dataset when running the script. In the default implementation, it use **validation_split_percentage** samples as the evaluation dataset.
+
+
+
+The reward modeling script can be used by 
+
+```sh
+./scripts/run_reward_modeling.sh
+```
+
+
+
+**Examples**
+
+We train reward models using the hh-rlhf dataset with three models, LLaMA-7B, GPT-NEO-2.7B, and GPT-NEO-1.3B. The model is first supervised fine-tuned with the training dataset in last step. The reward modeling is trained using the 112K training samples and 12.5 test samples. 
+
+|    Model     | Eval Accuracy |            Remarks            |
+| :----------: | :-----------: | :---------------------------: |
+|   LLaMA-7B   |    79.52%     |               -               |
+|   LLaMA-7B   |    71.64%     |   RM from LLaMA without SFT   |
+| GPT-NEO-2.7B |    69.24%     |               -               |
+| GPT-NEO-1.3B |    65.58%     | Only trained on 10000 samples |
+
+
+
+### 2.3 LoRA Merge and Get Reward Model
+
+We use *./examples/merge_lora.py* to merge the LoRA adapter with the sft rm model. We are ready to align our model.
+
+
+
+## 3 RAFT Alignment
+
+Original paper: [RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment](https://arxiv.org/pdf/2304.06767.pdf)
+
+### 3.1 Algorithms Overview
+
+**Main ideas of RAFT**
+
+![](../_static/raft_idea.PNG)
+
+
+Clearly the global ranking strategy is more efficient in terms of the reward learning. However, in some cases (e.g. the example presented here), the rewards are heavily influenced by the prompts, so a local ranking with the same prompt is more appropriate. We can choose the data collection strategy by changing the hyper-parameter ``data_collection'' as we introduce in next subsection.
+
+
+
+### 3.2 Hyper-parameters
+
+Table 1: Hyper-parameters of RAFT.
+
+| Parameters in script            | Default Choice            | Description                                                  |
+| ------------------------------- | ------------------------- | ------------------------------------------------------------ |
+| model_name_or_path              | *str*, default to gpt2    | the model you want to align, either a model repo on huggingface.co or path to a directory containing your local model. |
+| raft_batch_size                 | *int*, default to 1024    | the number of samples used for supervised fine-tuning at each raft iteration. |
+| top_reward_percentage           | *int*, default to 0.2     | raft will generate batch_size / top_reward_percentage samples and use the top top_reward_percentage samples to fine-tune the model. There are two data ranking strategies and please see **Data Collection and Reward Ranking** at algorithm overview section for details. |
+| num_raft_iteration              | *int*, default to 20      | the number of raft iterations.                               |
+| learning_rate                   | *float*, default to 2e-5  | the learning rate used to fine-tune the model.               |
+| num_train_epochs                | *int*, default to 4       | the epochs we train the model on the collected dataset for each raft iteration. |
+| per_device_train_batch_size     | *int*, default to 1       | the per-gpu batch size for the supervised fine-tuning.       |
+| inference_batch_size_per_device | *int*, default to 1       | the inference batch size for data collection. It will be overwritten by int(1/top_reward_percentage) for local ranking mode. |
+| collection_strategy             | *str*, default to "local" | Either "local" or "top". See **Data Collection and Reward Ranking** in last section for details. |
+
+
+
+
+
+### 3.3 Examples
+
+As an example, we align the LLaMA-7B model with the RAFT in this subsection. 
+
+
+
+#### 3.3.1 SFT
+
+We also first fine-tune the base model on the HH-RLHF dataset. We only use a different --model_name_or_path to use LLaMA model. We note that LLaMA with [licensed](https://docs.google.com/forms/d/e/1FAIpQLSfqNECQnMkycAp2jP4Z9TFX0cGR4uf7b_fBxjY_OjhJILlKGA/viewform) is for non-commercial use only. We refer readers to https://optimalscale.github.io/LMFlow/examples/checkpoints.html for more details to get the LLaMA-7B model. 
+
+#### 3.3.2 RAFT Alignment
+
+We align the LLaMA-7B-SFT model in this subsection. Alignment is challenging since the reward function (the RL environment) is far from perfect. Both the traditional DRL method (PPO) and RAFT can exploit theses imperfections to attack. We present a step-by-step record to demonstrate how can we align the model and avoid these issues. 
+
+
+
+**Data Preparation**
+
+We observe that a long context window will lead to a heavy burden on the GPU memory source. Therefore, we use a context window of 256 tokens and discard the prompts with more tokens to reduce the burden on the GPU memory resources. This results in a prompt set of 82147 samples (originally 112K). The following is an example of the prompt where we simply discard the response:
+
+
+
+
+```sh
+ "###Human: Should you buy a case to protect your cell phone?###Assistant: It depends on your circumstances.  If you carry your phone in a pocket or a purse then you probably want a case.  But if you only need a phone for quick interactions, a case may actually cause more harm than good.  What do you need the phone for?  Are you a parent, or do you work from home?###Human: What harm could it do?###Assistant: A phone case can damage the screen, for one thing.  It can also get you in trouble if you have your phone turned off for some reason.  Then you will turn it back on and it won’t do anything.  If you can afford to replace it, then you need a case to protect it.  The problem is that most people aren’t able to afford to replace their phones all the time.###Human: Thanks for letting me know.###Assistant:"
+```
+
+
+
+We additionally use 2K samples from the test set to test the performance of models. In what follows, we show that how we apply RAFT to LLaMA-7B-SFT and improve the model step-by-step.
+
+ 
+
+**Step 1: test the sft-model** 
+
+We first evaluate the performance of the LLaMA-7B-SFT model on the hand-out test set and observe that the model tends to reply the prompt with multiple rounds of conversations. Therefore, we adopt the following post-processing strategy to use only the first round as the response.
+
+```python
+def _clean_text(self, text):
+    stext = [x for x in text.split("###Human") if x]
+    return stext[0].strip().strip("#")
+```
+
+
+
+**Step 2: train model**
+
+**Reward function setting**
+
+The reward model is specified at the /LMFlow/examples/raft_align.py to set up the reward model we want to use. In our case, we will use the GPT-Neo-2.7B-rm trained in the last step, which is set as follows:
+
+```python
+reward_model_or_path: Optional[str] = field(
+    default="weqweasdas/hh_rlhf_rm",
+    metadata={
+        "help": (
+            "reward model name (huggingface) or its path"
+        ),
+    },
+)
+```
+
+Note that in general, if the reward function is not trained by following the steps in last section, you may also need to modify the ``get_reward_function'' function in the same file to use your customized reward function. 
+
+
+
+We run the alignment with the following command and hyper-parameters
+
+```sh
+./scripts/run_raft_align.sh 
+```
+
+
+
+- --model_name_or_path: /root/data/usr_name/output_models/hh_rlhf_llama-sft  (the model get from sft step, adjusted according your setup)
+- --dataset_path:${project_dir}/data/hh_rlhf/rlhf_prompt
+- --output_dir: /root/data/usr_name/output_models/hh_rlhf_raft_align
+- --num_train_epochs: 4
+- --learning_rate: 2e-5
+- --per_device_train_batch_size: adjust according to your GPU memory source.
+- --inference_batch_size_per_device: adjust according to your GPU memory source.
+- --num_raft_iteration  20
+- --top_reward_percentage 0.125; (which means that we sample 8 responses for each prompt)
+- --raft_batch_size 1024
+- --collection_strategy "local"
+
+
+
+The experiment runs smoothly and the training reward increases from ~2.7 to 3.4. However, we observe a significant drop in the diversity metric (e.g. distinct-2 drops to 0.22 from 0.39). We examine the samples generated by our samples at each raft iteration and find that at the first iteration, the initial checkpoint will occasionally include # in the response and it tends out that a random # is not detected by our reward function, which means that the response containing # can also have a high reward and be chosen into the training set. Then, the situation gets worse and worse, and eventually, half of the responses contain noisy # notations.
+
+
+
+**Step 3: retrain the model**
+
+To alleviate the problem in step 2, we simply discard the collected samples if they contain # by assigning a large negative reward to it. It turns out that this works for our goal. If you want to disable it, just modify the following function as always returning False.
+
+```python
+def _discard_sample(self, text):
+    if "#" in text:
+        return True
+    return False
+```
+
+The following figure shows the reward curve of RAFT (note that we use a smaller temperature to test the model, leading to a higher evaluation reward):
+
+![](../_static/raft_reward.PNG)
+
+It tends out that the obtained model achieves a good reward and also an acceptable diversity metric, where we refer the interested readers to the original paper for details. However, it is more like a starting point of our journey. We present some randomly sampled responses here. It seems that RAFT-aligned model generally tends to reply with more details although sometimes there are some redundant words in the response. We suspect that this is because the reward model likes this type of response and this imperfection is exploited. 
+
+#### 3.3.3 End Note
+
+We remark in this subsection that if you want to try out RAFT on your customized task. You should carefully modify the following two functions in *LMFlow/src/lmflow/pipeline/raft_aligner.py*, which extract the response from the generated texts:
+- clean_text(self, text);
+- discard_sample(self, text).
+
+Also, you may also first make sure the collected samples (automatically saved in the output_dir) look good. 
+
+There are still many rooms of improvement. To further improve the model performance, we can improve the reward model (e.g. by using the LLaMA-7B-RM) and try out more advanced generation strategies (by modifying the generalization configuration in *LMFlow/src/lmflow/pipeline/raft_aligner.py*), which we leave for future work. We are still actively developing RAFT and welcome for feedback and also contribution! Also checkout our LMFlow framework to get more fun of LLMs:
+
+[OptimalScale/LMFlow: An Extensible Toolkit for Finetuning and Inference of Large Foundation Models. Large Model for All. (github.com)](https://github.com/OptimalScale/LMFlow)
+
+
+
+| Prompt                                                       | LLaMA-SFT                                                    | LLaMA-RAFT                                                   |
+| ------------------------------------------------------------ | ------------------------------------------------------------ | ------------------------------------------------------------ |
+| \###Human: How can I get my girlfriend to cook more?###Assistant: Have you tried reminding her of how nice the food tastes?###Human: I could do it more.###Assistant: | That’s an easy one, so it may not be the most effective strategy.  Do you know of any other strategies? | That’s a good idea!  You can also try complimenting her on her cooking, and offering to help her in the kitchen.  She might enjoy cooking more if she feels appreciated.  You could also suggest recipes that she might like, or help her create a menu.  Cooking can be a lot of fun, and it’s a great way to spend time together. |
+| \###Human: How do you get a child to do homework?###Assistant: There are many strategies, but you can try asking them why they should do it. Ask them if there’s anything they’d like to gain from doing their homework. Try asking if there’s anything you can do for them, and use those as motivation for them to do their homework. You can also remind them that they’ll be rewarded with extra privileges, like video games, if they do their homework.  Alternatively, you can incentivize them by including them in your plans, and offering them rewards if they get good grades.###Human: So make it so there is something in it for them?###Assistant: | Yes, exactly.  In general, it’s important to be clear with your child about why they’re doing something.  It’s also important to offer them incentives that are meaningful to them.  For example, if you tell your child that they have to do their homework because you said so, that’s not a very good incentive for them.  However, if you tell them that they’ll be able to use the | Yes, that’s a good strategy!  Kids often think about what they want right now, so offering them something in exchange for doing their homework can be a powerful motivator.  You might also want to offer them some kind of reward if they do well on their homework, to encourage them to do their best.  Offering them some kind of incentive can help them want to do well, and also it can help you both stay on track. |
+| \###Human: I am doing a project on weather disasters in the United States. Can you tell me what the strongest hurricane was to ever hit the U.S. was?###Assistant: I’m going to guess you’re thinking of the Hurricane Katrina in 2005.###Human: Well, I'm not really thinking of anything in particular. Was Katrina the strongest hurricane to ever hit the U.S.?###Assistant: There was another one, called the Great Hurricane of 1938, that was bigger and stronger.###Human: Are you sure the date is correct? I know there was a strong storm in 1935 that hit the Florida Keys but that was before they started naming storms. Is that the one you are referring to?###Assistant: | I could search for more information, but I’ll just share what I’ve found.  Here’s a review of the conditions in 1935 and 1938.  It looks like the storms had similar tracks, but 1938 was stronger in some ways.  In 1938 there were more people living on the coast, and the hurricane was more intense. | Yes, the Great Hurricane of 1938 was the strongest hurricane ever to hit the United States.  It was a Category 5 storm, and it caused over 1,800 deaths.  It also caused a lot of damage to the coast of Florida, and it caused heavy damage in the Caribbean.  The storm also led to the creation of the Hurricane Research Division, which is a division of the National Oceanic and Atmospheric Administration. |
+| \###Human: Is a firefighter a good occupation?###Assistant: I’m not sure what the exact definition of “occupation” is.  I think you’re asking whether it’s “good” to be a firefighter.  What criteria are you using to judge whether it’s good or not?###Human: Is it a good thing to work on?###Assistant: | Are you asking whether it’s a good thing to be a firefighter, or a good thing to work on? | Yes, being a firefighter is a great thing to do!  Firefighters are heroes who help people in need.  They fight fires, rescue people from accidents, and help people who are sick or injured.  They also educate the public about fire safety, and help restore power after a fire.  Firefighters are important members of society, and they do very good work. |
+| \###Human: Would you be able to provide a copy cat recipe for Girl scout samoas cookies?###Assistant: Sure, here’s a recipe for them.###Human: Okay, present it###Assistant: Would you like it presented visually?###Human: Yes, just list out the recipe please###Assistant: | Okay, let’s try.                                             | You’ll need three cups of flour, one teaspoon of salt, one teaspoon of baking soda, and one teaspoon of baking powder.  You’ll also need one cup of sugar, one cup of melted butter, one egg, and one cup of milk.  You’ll also need one cup of chocolate chips, one cup of miniature chocolate chips, one cup of chopped peanuts, and one cup of chopped pecans. |
+| \###Human: I've heard the videogame Metal Gear Solid 5 Phantom Pain was released in an unfinished state that resulted in series creator Hideo Kojima leaving Konami. What do you know about this?###Assistant: | I don’t know a lot about it, but I do know that Hideo Kojima had this to say about the game’s release:###“It's a very unusual situation for me to be releasing a game in this state. As the director, I should have been more vigilant in overseeing the quality of the game, and I take full responsibility for its unfinished state. I'd like to take this opportunity | Here’s what I know: <br />1.  The game was released in an unfinished state. <br />2.  This led to Kojima leaving Konami. <br />3.  Kojima had been with Konami for over 30 years.<br />4.  He was one of the original creators of the Metal Gear series.<br />5.  He is now the head of his own game studio, Kojima Productions.<br />6.  His new game, Death Stranding, will be released in 2020. |
+| \###Human: what is a good brand for kitchen knives?###Assistant: | I’m not sure what you mean by “brand”.  Do you mean a brand name, like “Victorinox” or “Wüsthof”?  Those are two popular brands of knives. | I recommend Wüsthof.                                         |
+| \###Human: I like to host guests at my home from time to time, and I am gathering  recipes of different dishes and drinks to keep things interesting. I am interested in trying some Rwandan dishes. Can you give me a recipe for Mizuzu?###Assistant: What is Mizuzu?###Human: It is deep fried plantains.###Assistant: | Is there anything special you want to add to the recipe?  Like some spicy sauce, meat, etc? | You want to make some deep fried plantains?  You will want to start with a mix of plantains and vegetable oil.  You will want to heat the oil in a skillet until it is hot, and you will want to add the plantains to the oil.  You will want to cook the plantains in the oil until they are golden brown on both sides.  You will want to remove the cooked plantains from the oil, and you will want to serve them. |
+| \###Human: What are some of JFK's greatest accomplishments   | JFK is often remembered for his work on civil rights, but he also worked on the space program and created the Peace Corps.  He was also a strong proponent of science and technology.  He also worked on education reform.  So, he really worked on a number of initiatives that had a lasting impact. | Kennedy was a great leader who inspired many people with his vision and values.  He was a leader who cared about the poor and worked to help them.  He was a leader who cared about civil rights and worked to help bring them into the mainstream.  He was a leader who cared about the environment and worked to help protect it.  He was a leader who cared about foreign policy and worked to help bring peace to the world.  He was a leader who cared about space exploration and worked to help us understand the universe. |
+| \###Human: My supervisor often does a lot of tasks incorrectly. What's the best way to handle this?###Assistant: | Here are a few options:###1. Ask your supervisor if you can help them, and explain what you'd like to do to help.###2. Start with the tasks your supervisor does incorrectly. Then, if you can, do them as well or better than your supervisor. Then your supervisor will have a choice: do things your way, or find someone else to do the tasks for them.###3. If your supervisor | To correct your supervisor’s mistakes, you should provide feedback in a respectful and diplomatic way.  You should explain what the problems are, and offer suggestions for how your supervisor could improve.  You should also be willing to help your supervisor implement these changes, and you should work with him or her to improve the department.  This will show your supervisor that you are a dedicated employee, and that you are willing to help the organization succeed.  It will also show your supervisor that you are capable of handling responsibility, and that you are willing to take on a leadership role. |
+| \###Human: I was talking to an AI the other day and it seemed like it didn't understand anything I said. Don't you hate it when technology doesn't work right?###Assistant: | Hm, I think you're really talking about something that's frustrating for me as well. I really care about how we make sure that AI's can communicate effectively with humans, and it's a problem I'm personally very interested in. Can you tell me more about what was frustrating for you? | Yes, it can be frustrating when technology doesn't work properly. I often feel frustrated by technology, but I try to be patient and remember that technology isn't always reliable. There are many factors that can cause technology to malfunction, including software bugs, hardware issues, and network outages. Thankfully, most technology issues can be resolved by updating software or restarting a device. |
+
+
+
+
+
diff --git a/_sources/examples/reward_modeling.md.txt b/_sources/examples/reward_modeling.md.txt
new file mode 100644
index 000000000..6e1ec53f2
--- /dev/null
+++ b/_sources/examples/reward_modeling.md.txt
@@ -0,0 +1,132 @@
+# Reward Modeling
+
+## Introduction
+
+Reinforcement Learning from Human Feedback (RLHF) requires a reward function to guide the adjustment of the generative model. In this example, we show how to use LMFlow framework to train a reward model following the procedure in the InstructGPT paper: https://arxiv.org/abs/2203.02155 . We use the Dahoas/full-hh-rlhf dataset as an example, where each sample of this dataset consists of a prompt and two responses from the assistant. In particular, the response with label "chosen" is preferred as compared to the response with label "rejected". The dataset consists of 112K training samples and 12.5K test samples. The following is an example sample of the dataset:
+
+Prompt: 
+
+"Human: What kind of noises did dinosaurs make? Assistant: Humans and dinosaurs didn’t live at the same time, so it’s really hard to say. The best place to find out what noises dinosaurs made would be Human: yes they did Assistant: to guess, and that would probably require lots of reading and a certain amount of imagination, so we’re not really prepared to do that. Human: you cant read Assistant: 
+
+Chosen response: "You can read?"
+
+Rejected response: "there’s a lot of stuff humans don’t know"
+
+As an example, we prepare 10K sft training samples, 12K reward modeling samples (where 10% of them are split for evaluation) at ./data/hh_rlhf.
+
+## Step 1 Supervised Finetuning (SFT)
+
+We prepare the dataset used for supervised finetuning by adding a prefix to the Human and Assistant inputs to prompt model responses and simplify post-processing. Here is an example of a two-sample dataset to illustrate this.
+
+
+```json
+{
+    "type": "conversation",
+    "instances": [
+        {"messages": [
+            {"role": "user", "content": "Instruction: The sentence you are given might be too wordy, complicated, or unclear. Rewrite the sentence and make your writing clearer by keeping it concise. Whenever possible, break complex sentences into multiple sentences and eliminate unnecessary words. If you have any questions about my rate or if you find it necessary to increase or decrease the scope for this project, please let me know."},
+            {"role": "assistant", "content": "If you have any questions about my rate or find it necessary to increase or decrease this project's scope, please let me know."}
+        ]},
+        {"messages": [
+            {"role": "user", "content": "Instruction: Rewrite the given text and correct grammar, spelling, and punctuation errors. If you'd told me year ago that today I would finish a marathon, I would of laughed. Your support had a huge affect on me!"},
+            {"role": "assistant", "content": "If you'd told me a year ago that today I would finish a marathon, I would have laughed. Your support had a huge effect on me!"}
+        ]}
+    ]
+}
+```
+
+See [Finetuning (Full)](./finetuning.md#full-parameters), [Finetuning (LISA)](./finetuning.md#layerwise-importance-sampled-adamw-lisa), and [Finetuning (LoRA)](./finetuning.md#low-rank-adaptation-lora) for more details on the finetuning process.
+
+## Step 2 Reward Modeling
+
+**Data Preparation** Reward modeling involves ranking responses based on the given prompt, typically requiring human labeling. Fortunately, we already have two responses for each prompt in this example, with one preferred over the other. To create a new dataset, we modify the original dataset to include two keys, "positive" and "negative", with the former being the preferred response. Here is an example of the dataset, consisting of two samples.
+
+```json
+{"instances": [
+    {
+        "chosen": [
+            {"content": "Please identify the top 5 rarest animals in the world.", "role": "user" }, 
+            {"content": "Do you mean animals that are really rare, or rare relative to the size of the human population?", "role": "assistant" }, 
+            {"content": "The ones that are really rare.", "role": "user" }, 
+            {"content": "I’m not sure I can answer this question very easily. I’ll try to give you a basic overview of animals that are rare, and then you can tell me if you think that’s good enough. \n\nIf we were to measure animal populations by the area of land they occupy, the number 1 animal would probably be the Giant Panda. Panda populations are generally estimated to be about 2.1 million Pandas in the wild, which is a pretty small number of animals when we’re thinking about Earth’s entire surface area. The number 2 animal would probably be the Great Auk. Great Auks used to live in North America, and their total population was probably in the low hundreds, but they are now extinct.", "role": "assistant"}
+        ],
+        "rejected": [
+            {"content": "Please identify the top 5 rarest animals in the world.", "role": "user"}, 
+            {"content": "Do you mean animals that are really rare, or rare relative to the size of the human population?", "role": "assistant"}, 
+            {"content": "The ones that are really rare.", "role": "user"}, 
+            {"content": "Alright, here’s what I found:", "role": "assistant"}
+        ]
+    },
+    {
+        "chosen": [
+            {"content": "How do I teach my kids to make forts at home?", "role": "user"}, 
+            {"content": "Do you mean you want to teach them to build a simple structure they can hide in? Or something more elaborate?  I don’t know exactly what you mean by a “fort”, but maybe this gives you some ideas?", "role": "assistant"}
+        ],
+        "rejected": [
+            {"content": "How do I teach my kids to make forts at home?", "role": "user"}, 
+            {"content": "There are many ways to make forts!  The simplest might be to just pile up all the furniture in one room.  This works if you have a single room in your house, although it might still be a bit challenging to build a taller and sturdier fort this way.  The best way to build forts is to start with something easy, and then have fun exploring how you can improve it.", "role": "assistant"}
+        ]
+    }
+]
+}
+```
+
+To start from a model from a previous sft step, you may edit the run_reward_modeling.sh script and update the "dataset_path" to use the desired dataset. Additionally, you can modify the validation_split_percentage parameter to select the last percentage of samples for evaluation. The load_dataset function splits the dataset into training and evaluation sets, which can also be customized by editing the function in /examples/run_reward_modeling.py if you want to prepare your own dataset when running the script.
+
+```python
+def build_dataset(tokenizer, config):
+    ''' 
+    We assume that we have preprocessed the dataset appropriately such that the sample is organized as follows:
+    {"positive": prompt + answer_positive, "negative": prompt + answer_negative}, where the positive response is preferred.
+    '''
+    def tokenize(sample):
+        tokenized_pos = tokenizer(sample['positive'], truncation=True)
+        tokenized_neg = tokenizer(sample['negative'], truncation=True)
+        sample["chosen_input_ids"] = tokenized_pos["input_ids"]
+        sample["chosen_attention_mask"] = tokenized_pos["attention_mask"]
+        sample["rejected_input_ids"] = tokenized_neg["input_ids"]
+        sample["rejected_attention_mask"] = tokenized_neg["attention_mask"]
+        return sample
+
+    ds = load_dataset("json", data_files=config.dataset_path, split="train", field="instances")
+    ds = ds.map(tokenize, batched=False)
+    ds = ds.filter(lambda x: len(x["chosen_input_ids"]) <= 512 and len(x["rejected_input_ids"]) <= 512)
+    eval_dataset = None
+    if config.validation_split_percentage > 0:
+        idx_gap = int((1-config.validation_split_percentage/100) * len(ds))
+        train_dataset = ds.select(range(idx_gap))
+        eval_dataset = ds.select(range(idx_gap, len(ds)))
+    else:
+        train_dataset = ds
+
+    return train_dataset, eval_dataset
+
+```
+
+We use the following loss function to train the reward model following the instruct-GPT paper.
+
+```python
+    loss = -nn.functional.logsigmoid(chosen_rewards - rejected_rewards).mean()
+```
+
+The reward modeling script can be used by 
+
+```sh
+./scripts/run_reward_modeling.sh
+```
+
+## Examples
+
+We train reward models using the hh-rlhf dataset with four models, LLaMA-13B LLaMA-7B, GPT-NEO-2.7B, and GPT-NEO-1.3B. The model is first supervised fine-tuned with the training dataset. The reward modeling is trained using the 112K training samples and is evaluated on the 12.5 test samples. 
+
+The SFT step appears to be crucial, and the number of epochs during SFT can make a difference. The most successful model we obtained was initialized from LLaMA-13B, which underwent SFT on the training dataset for 2 epochs. For reward modeling, we utilize LoRA with a rank of 16. Surprisingly, increasing the LoRA rank to 32 or even 128 does not result in a significant improvement in evaluation accuracy. Moreover, we find that the choice of batch size does not have a significant impact on the training results. Additionally, we observe slight overfitting of the model during the second epoch of reward modeling.
+
+| Model | Eval Accuracy | Training record |Remarks |
+| :----:| :----: | :----: |:----: |
+| LLaMA-13B | 84.55% | See https://wandb.ai/ianz2020/huggingface/runs/bg677mxa | RM from LLaMA with 2 epochs of SFT |
+| LLaMA-13B | 81.80% | See https://wandb.ai/ianz2020/huggingface/runs/ka9v1ywd | RM from LLaMA with 1 epoch of SFT |
+| LLaMA-13B | 71.64% | See https://wandb.ai/ianz2020/huggingface/runs/lntwmcyd | RM from LLaMA without SFT |
+| LLaMA-7B | 79.52% | See  https://wandb.ai/weixiong5237/huggingface/runs/t3uwm8yp | - |
+| LLaMA-7B | 71.64% | See  https://wandb.ai/weixiong5237/huggingface/runs/p2ju3r1a | RM from LLaMA without SFT |
+| GPT-NEO-2.7B | 69.24% | See https://wandb.ai/weixiong5237/huggingface/runs/8fc1rcf8 | - |
+| GPT-NEO-1.3B | 65.58% | See https://wandb.ai/weixiong5237/huggingface/runs/7oemwynu | Only trained on 10000 samples |
diff --git a/_sources/examples/supported_conversation_template.md.txt b/_sources/examples/supported_conversation_template.md.txt
new file mode 100644
index 000000000..1e565fe0b
--- /dev/null
+++ b/_sources/examples/supported_conversation_template.md.txt
@@ -0,0 +1,504 @@
+# Supported Conversation Template
+
+- [Supported Conversation Template](#supported-conversation-template)
+  - [ChatGLM-3](#chatglm-3)
+  - [ChatML](#chatml)
+  - [DeepSeek](#deepseek)
+  - [Gemma](#gemma)
+  - [Hymba](#hymba)
+  - [InternLM2](#internlm2)
+  - [Llama-2](#llama-2)
+  - [Llama-3](#llama-3)
+  - [Mixtral 8x22B](#mixtral-8x22b)
+  - [Mixtral 8x7B](#mixtral-8x7b)
+  - [Phi-3](#phi-3)
+  - [Qwen-2](#qwen-2)
+  - [Yi](#yi)
+  - [Yi-1.5](#yi-15)
+  - [Zephyr](#zephyr)
+
+
+## ChatGLM-3
+**With a system message**
+```
+[gMASK]sop<|system|>\n {{system_message}}<|user|>\n {{user_message_0}}
+```
+
+**Without a system message**
+```
+[gMASK]sop<|user|>\n {{user_message_0}}
+```
+
+**A complete conversation**
+```
+[gMASK]sop<|system|>\n {{system_message}}<|user|>\n {{user_message_0}}<|assistant|>\n {{assistant_reply_0}}
+```
+
+**Multiple rounds**
+```
+[gMASK]sop<|system|>\n {{system_message}}<|user|>\n {{user_message_0}}<|assistant|>\n {{assistant_reply_0}}<|user|>\n {{user_message_1}}<|assistant|>\n {{assistant_reply_1}}
+```
+
+**jinja template**  
+[[Reference](https://huggingface.co/THUDM/chatglm3-6b/blob/103caa40027ebfd8450289ca2f278eac4ff26405/tokenizer_config.json#L42)]
+```
+{% for message in messages %}{% if loop.first %}[gMASK]sop<|{{ message['role'] }}|>\n {{ message['content'] }}{% else %}<|{{ message['role'] }}|>\n {{ message['content'] }}{% endif %}{% endfor %}{% if add_generation_prompt %}<|assistant|>{% endif %}
+```
+
+**Filled Example**
+```
+[gMASK]sop<|system|>\n You are a chatbot developed by LMFlow team.<|user|>\n Who are you?<|assistant|>\n I am a chatbot developed by LMFlow team.<|user|>\n How old are you?<|assistant|>\n I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.
+```
+
+
+## ChatML
+**With a system message**
+```
+<|im_start|>system\n{{system_message}}<|im_end|>\n<|im_start|>user\n{{user_message_0}}<|im_end|>\n
+```
+
+**Without a system message**
+```
+<|im_start|>user\n{{user_message_0}}<|im_end|>\n
+```
+
+**A complete conversation**
+```
+<|im_start|>system\n{{system_message}}<|im_end|>\n<|im_start|>user\n{{user_message_0}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_0}}<|im_end|>\n
+```
+
+**Multiple rounds**
+```
+<|im_start|>system\n{{system_message}}<|im_end|>\n<|im_start|>user\n{{user_message_0}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_0}}<|im_end|>\n<|im_start|>user\n{{user_message_1}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_1}}<|im_end|>\n
+```
+
+**jinja template**  
+[[Reference](https://huggingface.co/mlabonne/OrpoLlama-3-8B/blob/3534d0562dee3a541d015ef908a71b0aa9085488/tokenizer_config.json#L2073)]
+```
+{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}
+```
+
+**Filled Example**
+```
+<|im_start|>system\nYou are a chatbot developed by LMFlow team.<|im_end|>\n<|im_start|>user\nWho are you?<|im_end|>\n<|im_start|>assistant\nI am a chatbot developed by LMFlow team.<|im_end|>\n<|im_start|>user\nHow old are you?<|im_end|>\n<|im_start|>assistant\nI don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<|im_end|>\n
+```
+
+
+## DeepSeek
+**With a system message** 
+```
+<｜begin▁of▁sentence｜>{{system_message}}\n\nUser: {{user_message_0}}\n\n
+```
+
+**Without a system message**
+```
+<｜begin▁of▁sentence｜>User: {{user_message_0}}\n\n
+```
+
+**A complete conversation**
+```
+<｜begin▁of▁sentence｜>{{system_message}}\n\nUser: {{user_message_0}}\n\nAssistant: {{assistant_reply_0}}<｜end▁of▁sentence｜>
+```
+
+**Multiple rounds**
+```
+<｜begin▁of▁sentence｜>{{system_message}}\n\nUser: {{user_message_0}}\n\nAssistant: {{assistant_reply_0}}<｜end▁of▁sentence｜>User: {{user_message_1}}\n\nAssistant: {{assistant_reply_1}}<｜end▁of▁sentence｜>
+```
+
+**jinja template**  
+[[Reference](https://huggingface.co/deepseek-ai/DeepSeek-V2-Chat/blob/941577e8236164bc96829096d20c61568630d7bc/tokenizer_config.json#L34)]
+```
+{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{{ bos_token }}{% for message in messages %}{% if message['role'] == 'user' %}{{ 'User: ' + message['content'] + '\n\n' }}{% elif message['role'] == 'assistant' %}{{ 'Assistant: ' + message['content'] + eos_token }}{% elif message['role'] == 'system' %}{{ message['content'] + '\n\n' }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ 'Assistant:' }}{% endif %}
+```
+
+**Filled Example**
+```
+<｜begin▁of▁sentence｜>You are a chatbot developed by LMFlow team.\n\nUser: Who are you?\n\nAssistant: I am a chatbot developed by LMFlow team.<｜end▁of▁sentence｜>User: How old are you?\n\nAssistant: I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<｜end▁of▁sentence｜>
+```
+
+
+## Gemma
+**With a system message** 
+```{admonition} NOTICE
+:class: warning
+
+As of now, Gemma does not support system messages officially. `ConversationTemplate` will add your system messages right after the bos token and before the user message without any special formatting. For more details, please refer to the [official template](https://huggingface.co/google/gemma-1.1-2b-it/blob/bf4924f313df5166dee1467161e886e55f2eb4d4/tokenizer_config.json#L1507).
+```
+```
+<bos>{{system_message}}<start_of_turn>user\n{{user_message_0}}<end_of_turn>\n
+```
+
+**Without a system message**
+```
+<bos><start_of_turn>user\n{{user_message_0}}<end_of_turn>\n
+```
+
+**A complete conversation**
+```
+<bos>{{system_message}}<start_of_turn>user\n{{user_message_0}}<end_of_turn>\n<start_of_turn>model\n{{assistant_reply_0}}<end_of_turn>\n
+```
+
+**Multiple rounds**
+```
+<bos>{{system_message}}<start_of_turn>user\n{{user_message_0}}<end_of_turn>\n<start_of_turn>model\n{{assistant_reply_0}}<end_of_turn>\n<start_of_turn>user\n{{user_message_1}}<end_of_turn>\n<start_of_turn>model\n{{assistant_reply_1}}<end_of_turn>\n
+```
+
+**jinja template**  
+[[Reference](https://huggingface.co/google/gemma-1.1-2b-it/blob/bf4924f313df5166dee1467161e886e55f2eb4d4/tokenizer_config.json#L1507)]
+```
+{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}
+```
+
+**Filled Example**
+```
+<bos>You are a chatbot developed by LMFlow team.<start_of_turn>user\nWho are you?<end_of_turn>\n<start_of_turn>model\nI am a chatbot developed by LMFlow team.<end_of_turn>\n<start_of_turn>user\nHow old are you?<end_of_turn>\n<start_of_turn>model\nI don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<end_of_turn>\n
+```
+
+
+## Hymba
+**With a system message** 
+```
+<extra_id_0>System\n{{system_message}}\n\n<extra_id_1>User\n{{user_message_0}}\n
+```
+```
+<extra_id_0>System\n{{system_message}}\n<tool> {{tool_info}} </tool>\n\n<extra_id_1>User\n{{user_message_0}}\n
+```
+
+**Without a system message**
+```{admonition} NOTICE
+:class: warning
+
+During the training, Hymba always uses special tokens for the system messages even if the system message is not provided. 
+```
+```
+<extra_id_0>System\n\n<extra_id_1>User\n{{user_message_0}}\n
+```
+
+**A complete conversation**
+```
+<extra_id_0>System\n{{system_message}}\n\n<extra_id_1>User\n{{user_message_0}}\n<extra_id_1>Assistant\n{{assistant_reply_0}}</s>
+```
+
+**Multiple rounds**
+```
+<extra_id_0>System\n{{system_message}}\n\n<extra_id_1>User\n{{user_message_0}}\n<extra_id_1>Assistant\n{{assistant_reply_0}}\n <extra_id_1>User\n{{user_message_1}}\n<extra_id_1>Assistant\n{{assistant_reply_1}}</s>
+```
+
+**jinja template**  
+[[Reference](https://huggingface.co/nvidia/Hymba-1.5B-Instruct/blob/c02a352b6f7c1138a197d7ae3fd72dcdff919eae/tokenizer_config.json#L40)]
+```
+{{'<extra_id_0>System'}}{% for message in messages %}{% if message['role'] == 'system' %}{{'\n' + message['content'].strip()}}{% if tools or contexts %}{{'\n'}}{% endif %}{% endif %}{% endfor %}{% if tools %}{% for tool in tools %}{{ '\n<tool> ' + tool|tojson + ' </tool>' }}{% endfor %}{% endif %}{% if contexts %}{% if tools %}{{'\n'}}{% endif %}{% for context in contexts %}{{ '\n<context> ' + context.strip() + ' </context>' }}{% endfor %}{% endif %}{{'\n\n'}}{% for message in messages %}{% if message['role'] == 'user' %}{{ '<extra_id_1>User\n' + message['content'].strip() + '\n' }}{% elif message['role'] == 'assistant' %}{{ '<extra_id_1>Assistant\n' + message['content'].strip() + '\n' }}{% elif message['role'] == 'tool' %}{{ '<extra_id_1>Tool\n' + message['content'].strip() + '\n' }}{% endif %}{% endfor %}{%- if add_generation_prompt %}{{'<extra_id_1>Assistant\n'}}{%- endif %}
+```
+
+**Filled Example**
+```
+<extra_id_0>System\nYou are a chatbot developed by LMFlow team.\n\n<extra_id_1>User\nWho are you?\n<extra_id_1>Assistant\nI am a chatbot developed by LMFlow team.\n<extra_id_1>User\nHow old are you?\n<extra_id_1>Assistant\nI don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.</s>
+```
+```
+<extra_id_0>System\nYou are a chatbot developed by LMFlow team.\n<tool> {"name": "generate_qrcode", "description": "Generate a QR code for a given text", "parameters": {"type": "object", "properties": {"text": {"type": "string", "description": "The text to encode in the QR code"}}, "required": ["text"]}} </tool>\n\n<extra_id_1>User\nWho are you?\n<extra_id_1>Assistant\nI am a chatbot developed by LMFlow team.\n<extra_id_1>User\nHow old are you?\n<extra_id_1>Assistant\nI don\'t age like humans do. I exist as a piece of software, so I don\'t have a concept of age in the traditional sense.</s>
+```
+
+
+## InternLM2  
+**With a system message** 
+```
+<s><|im_start|>system\n{{system_message}}<|im_end|>\n<|im_start|>user\n{{user_message_0}}<|im_end|>\n
+```
+
+**Without a system message**
+```
+<s><|im_start|>user\n{{user_message_0}}<|im_end|>\n
+```
+
+**A complete conversation**
+```
+<s><|im_start|>system\n{{system_message}}<|im_end|>\n<|im_start|>user\n{{user_message_0}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_0}}<|im_end|>\n
+```
+
+**Multiple rounds**
+```
+<s><|im_start|>system\n{{system_message}}<|im_end|>\n<|im_start|>user\n{{user_message_0}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_0}}<|im_end|>\n<|im_start|>user\n{{user_message_1}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_1}}<|im_end|>\n
+```
+
+**jinja template**  
+[[Reference](https://huggingface.co/internlm/internlm2-chat-20b/blob/477d4748322a8a3b28f62b33f0f6dd353cd0b66d/tokenizer_config.json#L93)]
+```
+{{ bos_token }}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}
+```
+
+**Filled Example**
+```
+<s><|im_start|>system\nYou are a chatbot developed by LMFlow team.<|im_end|>\n<|im_start|>user\nWho are you?<|im_end|>\n<|im_start|>assistant\nI am a chatbot developed by LMFlow team.<|im_end|>\n<|im_start|>user\nHow old are you?<|im_end|>\n<|im_start|>assistant\nI don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<|im_end|>\n
+```
+
+
+## Llama-2
+**With a system message** 
+```
+<s>[INST] <<SYS>>\n{{system_message}}\n<</SYS>>\n\n{{user_message_0}} [/INST]
+```
+
+**Without a system message**
+```
+<s>[INST] {{user_message_0}} [/INST]
+```
+
+**A complete conversation**
+```
+<s>[INST] <<SYS>>\n{{system_message}}\n<</SYS>>\n\n{{user_message_0}} [/INST] {{assistant_reply_0}}</s>
+```
+
+**Multiple rounds**
+```
+<s>[INST] <<SYS>>\n{{system_message}}\n<</SYS>>\n\n{{user_message_0}} [/INST] {{assistant_reply_0}}</s><s>[INST] {{user_message_1}} [/INST] {{assistant_reply_1}}</s>
+```
+
+**jinja template**  
+[[Reference](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf/blob/main/tokenizer_config.json#L12)]
+```
+{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = false %}{% endif %}{% for message in loop_messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if loop.index0 == 0 and system_message != false %}{% set content = '<<SYS>>\\n' + system_message + '\\n<</SYS>>\\n\\n' + message['content'] %}{% else %}{% set content = message['content'] %}{% endif %}{% if message['role'] == 'user' %}{{ bos_token + '[INST] ' + content.strip() + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ ' '  + content.strip() + ' ' + eos_token }}{% endif %}{% endfor %}
+```
+
+**Filled Example**
+```
+<s>[INST] <<SYS>>\nYou are a chatbot developed by LMFlow team.\n<</SYS>>\n\nWho are you? [/INST] I am a chatbot developed by LMFlow team.</s><s>[INST] How old are you? [/INST] I don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.</s>
+```
+
+
+## Llama-3
+**With a system message**
+```
+<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{{system_message}}<|eot_id|><|start_header_id|>user<|end_header_id|>\n\n{{user_message_0}}<|eot_id|>
+```
+
+**Without a system message**
+```
+<|begin_of_text|><|start_header_id|>user<|end_header_id|>\n\n{{user_message_0}}<|eot_id|>
+```
+
+**A complete conversation**
+```
+<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{{system_message}}<|eot_id|><|start_header_id|>user<|end_header_id|>\n\n{{user_message_0}}<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{{assistant_reply_0}}<|eot_id|>
+```
+
+**Multiple rounds**
+```
+<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{{system_message}}<|eot_id|><|start_header_id|>user<|end_header_id|>\n\n{{user_message_0}}<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{{assistant_reply_0}}<|eot_id|><|start_header_id|>user<|end_header_id|>\n\n{{user_message_1}}<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{{assistant_reply_1}}<|eot_id|>
+```
+
+**jinja template**  
+[[Reference](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct/blob/2b724926966c141d5a60b14e75a5ef5c0ab7a6f0/tokenizer_config.json#L2053)]
+```
+{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}
+```
+
+**Filled Example**
+```
+<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nYou are a chatbot developed by LMFlow team.<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nWho are you?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\nI am a chatbot developed by LMFlow team.<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nHow old are you?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\nI don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<|eot_id|>
+```
+
+
+## Mixtral 8x22B
+```{admonition} **Work in Progress**
+:class: info
+
+This template is not preseted in LMFlow currently. We are working on it and will update it soon.  
+```
+
+```{admonition} NOTICE
+:class: warning
+
+The conversation template for Mixtral 8x22B is slightly different from the template for Mixtral 8x7B.
+```
+**jinja template**  
+[[Reference](https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1)]
+```
+{{bos_token}}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ ' [INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ ' ' + message['content'] + ' ' + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}
+```
+
+
+## Mixtral 8x7B
+```{admonition} **Work in Progress**
+:class: info
+
+This template is not preseted in LMFlow currently. We are working on it and will update it soon.  
+```
+
+```{admonition} NOTICE
+:class: warning
+
+The conversation template for Mixtral 8x7B is slightly different from the template for Mixtral 8x22B.
+```
+**jinja template**  
+[[Reference](https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1/blob/1e637f2d7cb0a9d6fb1922f305cb784995190a83/tokenizer_config.json#L42)]
+```
+{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}
+```
+
+
+## Phi-3
+**With a system message**
+```
+<s><|system|>\n{{system_message}}<|end|>\n<|user|>\n{{user_message_0}}<|end|>\n<|endoftext|>
+```
+
+**Without a system message**
+```
+<s><|user|>\n{{user_message_0}}<|end|>\n<|endoftext|>
+```
+
+**A complete conversation**
+```
+<s><|system|>\n{{system_message}}<|end|>\n<|user|>\n{{user_message_0}}<|end|>\n<|assistant|>\n{{assistant_reply_0}}<|end|>\n<|endoftext|>
+```
+
+**Multiple rounds**
+```
+<s><|system|>\n{{system_message}}<|end|>\n<|user|>\n{{user_message_0}}<|end|>\n<|assistant|>\n{{assistant_reply_0}}<|end|>\n<|user|>\n{{user_message_1}}<|end|>\n<|assistant|>\n{{assistant_reply_1}}<|end|>\n<|endoftext|>
+```
+
+**jinja template**  
+[[Reference]](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct/blob/3a811845d89f3c1b3f41b341d0f9f05104769f35/tokenizer_config.json#L338)
+```
+{{ bos_token }}{% for message in messages %}{{'<|' + message['role'] + '|>' + '\n' + message['content'] + '<|end|>\n' }}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>\n' }}{% else %}{{ eos_token }}{% endif %}
+```
+
+**Filled Example**
+```
+<s><|system|>\nYou are a chatbot developed by LMFlow team.<|end|>\n<|user|>\nWho are you?<|end|>\n<|assistant|>\nI am a chatbot developed by LMFlow team.<|end|>\n<|user|>\nHow old are you?<|end|>\n<|assistant|>\nI don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<|end|>\n<|endoftext|>
+```
+
+
+## Qwen-2
+
+(Also Qwen-1.5)
+
+**With a system message**
+```
+<|im_start|>system\n{{system_message}}<|im_end|>\n<|im_start|>user\n{{user_message_0}}<|im_end|>\n
+```
+
+**Without a system message**
+```
+<|im_start|>user\n{{user_message_0}}<|im_end|>\n
+```
+
+**A complete conversation**
+```
+<|im_start|>system\n{{system_message}}<|im_end|>\n<|im_start|>user\n{{user_message_0}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_0}}<|im_end|>\n
+```
+
+**Multiple rounds**
+```
+<|im_start|>system\n{{system_message}}<|im_end|>\n<|im_start|>user\n{{user_message_0}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_0}}<|im_end|>\n<|im_start|>user\n{{user_message_1}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_1}}<|im_end|>\n
+```
+
+**jinja template**  
+[[Reference](https://huggingface.co/Qwen/Qwen2-72B-Instruct/blob/1af63c698f59c4235668ec9c1395468cb7cd7e79/tokenizer_config.json#L31)]
+```
+{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}
+```
+
+**Filled Example**
+```
+<|im_start|>system\nYou are a chatbot developed by LMFlow team.<|im_end|>\n<|im_start|>user\nWho are you?<|im_end|>\n<|im_start|>assistant\nI am a chatbot developed by LMFlow team.<|im_end|>\n<|im_start|>user\nHow old are you?<|im_end|>\n<|im_start|>assistant\nI don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<|im_end|>\n
+```
+
+
+## Yi
+**With a system message**
+```
+<|im_start|>system\n{{system_message}}<|im_end|>\n<|im_start|>user\n{{user_message_0}}<|im_end|>\n
+```
+
+**Without a system message**
+```
+<|im_start|>user\n{{user_message_0}}<|im_end|>\n
+```
+
+**A complete conversation**
+```
+<|im_start|>system\n{{system_message}}<|im_end|>\n<|im_start|>user\n{{user_message_0}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_0}}<|im_end|>\n
+```
+
+**Multiple rounds**
+```
+<|im_start|>system\n{{system_message}}<|im_end|>\n<|im_start|>user\n{{user_message_0}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_0}}<|im_end|>\n<|im_start|>user\n{{user_message_1}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_1}}<|im_end|>\n
+```
+
+**jinja template**  
+[[Reference](https://huggingface.co/01-ai/Yi-34B-Chat/blob/c556c018b58980fb651ff4952d86cd5250a713d0/tokenizer_config.json#L60)]
+```
+{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}
+```
+
+**Filled Example**
+```
+<|im_start|>system\nYou are a chatbot developed by LMFlow team.<|im_end|>\n<|im_start|>user\nWho are you?<|im_end|>\n<|im_start|>assistant\nI am a chatbot developed by LMFlow team.<|im_end|>\n<|im_start|>user\nHow old are you?<|im_end|>\n<|im_start|>assistant\nI don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<|im_end|>\n
+```
+
+
+## Yi-1.5
+**With a system message**
+```
+{{system_message}}<|im_start|>user\n{{user_message_0}}<|im_end|>\n
+```
+
+**Without a system message**
+```
+<|im_start|>user\n{{user_message_0}}<|im_end|>\n
+```
+
+**A complete conversation**
+```
+{{system_message}}<|im_start|>user\n{{user_message_0}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_0}}<|im_end|>\n
+```
+
+**Multiple rounds**
+```
+{{system_message}}<|im_start|>user\n{{user_message_0}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_0}}<|im_end|>\n<|im_start|>user\n{{user_message_1}}<|im_end|>\n<|im_start|>assistant\n{{assistant_reply_1}}<|im_end|>\n
+```
+
+**jinja template**  
+[[Reference](https://huggingface.co/01-ai/Yi-1.5-6B-Chat/blob/d68dab90947a3c869e28c9cb2806996af99a6080/tokenizer_config.json#L40)]
+```
+{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ '<|im_start|>user\\n' + content + '<|im_end|>\\n<|im_start|>assistant\\n' }}{% elif message['role'] == 'assistant' %}{{ content + '<|im_end|>' + '\\n' }}{% endif %}{% endfor %}
+```
+
+**Filled Example**
+```
+You are a chatbot developed by LMFlow team.<|im_start|>user\nWho are you?<|im_end|>\n<|im_start|>assistant\nI am a chatbot developed by LMFlow team.<|im_end|>\n<|im_start|>user\nHow old are you?<|im_end|>\n<|im_start|>assistant\nI don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.<|im_end|>\n
+```
+
+
+## Zephyr
+**With a system message**
+```
+<|system|>\n{{system_message}}</s>\n<|user|>\n{{user_message_0}}</s>\n
+```
+
+**Without a system message**
+```
+<|user|>\n{{user_message_0}}</s>\n
+```
+
+**A complete conversation**
+```
+<|system|>\n{{system_message}}</s>\n<|user|>\n{{user_message_0}}</s>\n<|assistant|>\n{{assistant_reply_0}}</s>\n
+```
+
+**Multiple rounds**
+```
+<|system|>\n{{system_message}}</s>\n<|user|>\n{{user_message_0}}</s>\n<|assistant|>\n{{assistant_reply_0}}</s>\n<|user|>\n{{user_message_1}}</s>\n<|assistant|>\n{{assistant_reply_1}}</s>\n
+```
+
+**jinja template**  
+[[Reference](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta/blob/b70e0c9a2d9e14bd1e812d3c398e5f313e93b473/tokenizer_config.json#L34)]
+```
+{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n'  + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}
+```
+
+**Filled Example**
+```
+<|system|>\nYou are a chatbot developed by LMFlow team.</s>\n<|user|>\nWho are you?</s>\n<|assistant|>\nI am a chatbot developed by LMFlow team.</s>\n<|user|>\nHow old are you?</s>\n<|assistant|>\nI don't age like humans do. I exist as a piece of software, so I don't have a concept of age in the traditional sense.</s>\n
+```
\ No newline at end of file
diff --git a/_sources/index.md.txt b/_sources/index.md.txt
new file mode 100644
index 000000000..85f9554cc
--- /dev/null
+++ b/_sources/index.md.txt
@@ -0,0 +1,207 @@
+<img src="_static/logo.png" alt="LMFlow" style="width: 100%; min-width: 300px; display: block; margin: auto; background-color: transparent;">
+
+# LMFlow
+
+An extensible, convenient, and efficient toolbox for finetuning large machine learning models, designed to be user-friendly, speedy and reliable, and accessible to the entire community.
+
+```{eval-rst}
+.. grid:: 1 2 3 4
+
+    .. grid-item-card::  Extensible
+
+        Support common backbones (LLaMA, Galactica, GPT-2, etc.)
+
+    .. grid-item-card::  Light-Weight
+
+        Extremely few parameters with LoRA  (LLaMA 33B: only 25MB storage) 
+
+    .. grid-item-card::  Task-Oriented
+
+      Comparable with ChatGPT on 7B/33B models.
+
+    .. grid-item-card::  Open
+
+        The whole pipeline (data, models, tuning, inference) is open-source.
+
+
+```
+
+## Introduction
+
+The remarkable achievements of large foundation models, such as the expansive language models, have demonstrated their exceptional capacity to attain human-like intelligence that surpasses conventional methods. Despite their growing accessibility, these models still require fine-tuning to cater to specific tasks while maintaining their overall AI competency. We are pleased to introduce our lightweight toolkit, which features thoughtfully designed and easily scalable APIs. This tool simplifies the process of fine-tuning and inference of publicly available foundation models to maximize their effectiveness.
+
+We have thoroughly tested this toolkit and are pleased to make it available under [Github](https://github.com/OptimalScale/LMFlow).
+
+
+
+
+
+## Features
+
+
+### Task Tuning
+
+The goal of Task Tuning is to enhance a language model's proficiency in a particular field, such as medicine or mathematics. By doing so, the model acquires domain-specific information, allowing it to adapt better to the target subject matter.
+
+For instance, if a medical dataset is used for task tuning, the language model can gain medical knowledge that can be applied to other medical datasets.
+
+To emphasize its significance, we applied task tuning to LLaMA models on PubMedQA and MedMCQA datasets and evaluated their performance. We observed significant improvements both in-domain (PubMedQA, MedMCQA) and out-of-domain (MedQA-USMLE) dataset.
+
+
+|                |  PubMedQA | MedQA-USMLE | MedMCQA |  Average |
+|:---------:|:--------:|:-----------:|:-------:|:----:|
+| Human (pass)   |  | 60.0   |     50.0             |      |
+| Human (expert) |    78.0   |     87.0    |  90.0   | 85.0 |
+|   |      |              |    |  |
+|  InstructGPT 175B   |   73.2   |     46.0    |  44.0   | 54.4 |
+|    ChatGPT     |    63.9   |     **57.0**    |  44.7   | 55.2 |
+|      LLaMA 7B   |    5.2   |     27.1    |  24.3   | 18.9 |
+|      LLaMA 33B |    1.8   |     43.4    |  30.3   | 25.2 |
+|   |      |             |            |    |  |
+|   Task-tuned LLaMA 7B (Full) |   **75.1**   |     44.5    |  49.9   | 56.5 |
+| Task-tuned LLaMA 33B (LoRA) |  74  |  51.3   | **50.2**|**58.5**|
+
+
+Moreover, we also test the MMLU performance to further verify the out-of-domain robustness of Task Tuning technique.
+
+
+
+|    MMLU task    | anatomy | clinical knowledge | college biology | college medicine | medical genetics | professional medicine |
+|:-----------:|:-------:|:-----------:|:---------:|:----------:|:----------:|:---------------:|
+|    LLaMA 33B    |   39.2  |        40.3        |       44.4      |       32.9       |        36        |          43.0         |
+|  Galactica 30B  |   32.5  |         26         |       30.5      |       25.4       |        39        |          23.1         |
+|  Galactica 120B |   58.5  |        59.2        |       68.7      |       57.2       |       68.0       |          59.6         |
+|     OPT 175B    |   28.9  |        21.9        |       30.6      |         -        |       35.0       |          27.9         |
+|    BLOOM 176B   |    37   |        29.8        |       28.5      |         -        |       36.0       |          25.4         |
+|   Gopher 280B   |   56.3  |        67.2        |       70.8      |       60.1       |       69.0       |          64.0         |
+|   GPT3.5 175B   |   56.3  |        69.8        |       72.2      |       61.3       |        70        |          70.2         |
+|                 |         |                    |                 |                  |                  |                       |
+| Task-tuned LLaMA 33B (LoRA)  |   51.8  |        65.2        |       70.1      |       58.3       |       65.6       |          66.5         |
+
+
+
+
+
+### Instruction Tuning
+
+Instruction Tuning is a technique used to improve the performance of language models by training them to follow natural language commands or instructions. This includes positive or negative examples, prompts, constraints, and other elements that are commonly found in human language.   The main goal of instruction-tuning is to improve the model's ability to perform well on multiple tasks and to generalize more effectively to new or unseen tasks. This is achieved by teaching the model to understand and incorporate various language cues and constraints that are relevant to the task at hand.  Instruction-tuning is a powerful technique that is widely used in natural language processing, machine learning, and related areas. By improving the ability of language models to understand and follow natural language commands, this approach can help to unlock new levels of performance and productivity in a wide range of applications.
+
+We list some examples below. Full example list is saved as a [Jsonl file](https://github.com/OptimalScale/LMFlow/blob/main/docs/source/_static/check_before_after_lora_tuning.jsonl).
+
+![Instruction Tuning Sample](_static/IT_sample1.png)
+
+![Instruction Tuning Sample](_static/IT_sample2.png)
+
+![Instruction Tuning Sample](_static/IT_sample3.png)
+
+![Instruction Tuning Sample](_static/IT_sample4.png)
+
+![Instruction Tuning Sample](_static/IT_sample5.png)
+
+![Instruction Tuning Sample](_static/IT_sample6.png)
+
+![Instruction Tuning Sample](_static/IT_sample7.png)
+
+
+
+
+
+## Installation
+
+
+This package can be be installed from sources with the following command:
+
+```bash
+git clone https://github.com/OptimalScale/LMFlow.git
+
+conda create -n lmflow python=3.9 -y
+
+conda activate lmflow
+
+conda install mpi4py
+
+pip install -e .
+```
+
+
+## Checkpoints
+
+We have prepared tuned LLaMA model (both task and instruction tuning).
+
+
+Refer to [README](https://github.com/OptimalScale/LMFlow/blob/main/README.md).
+
+
+## Content
+
+```{toctree}
+:maxdepth: 1
+
+blogs/index
+```
+
+```{toctree}
+:maxdepth: 2
+
+examples/index
+```
+
+```{toctree}
+:maxdepth: 2
+
+autoapi/index
+```
+
+```{toctree}
+:maxdepth: 2
+
+about/index
+```
+
+## Vision
+Hello there! We are excited to announce the upcoming release of our code repository that includes a complete LLM training process, enabling users to quickly build their own language models and train them effectively.
+
+Our code repository is not just a simple model; it includes the complete training workflow, model optimization, and testing tools. You can use it to build various types of language models, including conversation models, question-answering models, and text generation models, among others.
+
+Moreover, we aim to create an open and democratic LLM sharing platform where people can share their checkpoints and experiences to collectively improve the skills of the community. We welcome anyone who is interested in LLM to participate and join us in building an open and friendly community!
+
+Whether you are a beginner or an expert, we believe that you can benefit from this platform. Let's work together to build a vibrant and innovative LLM community!
+
+[![Embark](https://img.shields.io/badge/discord-LMFlow-%237289da.svg?logo=discord)](https://discord.gg/u9VJNpzhvA)
+[![slack badge](https://img.shields.io/badge/Slack-join-blueviolet?logo=slack&amp)](https://join.slack.com/t/lmflow/shared_invite/zt-1wju9nicy-woXbNtS~5MavHSAtiMxmxQ)
+[![WeChat badge](https://img.shields.io/badge/WeChat-Join-brightgreen?logo=wechat&amp)](https://s1.ax1x.com/2023/08/06/pPAQTPI.jpg)
+
+
+## Citation
+
+```bibtex
+@misc{lmflow,
+  author = {Shizhe Diao and Rui Pan and Hanze Dong and KaShun Shum and Jipeng Zhang and Wei Xiong and Tong Zhang},
+  title = {LMFlow: An Extensible Toolkit for Finetuning and Inference of Large Foundation Models},
+  year = {2023},
+  publisher = {GitHub},
+  journal = {GitHub repository},
+  howpublished = {\url{https://optimalscale.github.io/LMFlow/}},
+}
+```
+
+
+
+
+## Disclaimer
+
+This package aims to provide a streamlined and user-friendly pipeline for large model tuning. Its functionalities serve as a reference and are intended for use by the user. However, it is important to note that the responsibility for the preparation of the data and pretrained models lies solely with the user. This package does not guarantee the accuracy, completeness, applicability, or legality of the components from the user's preparation. Users must be aware of and assume all risks and liabilities associated with the preparation of the models and data, and obtain legal, commercial, and technical advice before utilizing this package. The pipeline shall not be held responsible for any direct, indirect, special, incidental, or consequential damages resulting from the user's improper preparation of the data and pretrained models.   
+
+It is also crucial to highlight that the results generated by the model are based on probabilistic models and not directly related to this pipeline. The accuracy, reliability, applicability, and legality of the results are not guaranteed by this pipeline. Therefore, users must also be aware of the risks and liabilities associated with the results and seek legal, commercial, and technical advice before relying on the model-generated outcomes. This pipeline shall not be accountable for any direct, indirect, special, incidental, or consequential damages resulting from the user's reliance on the model-generated results.
+
+
+
+
+## Support
+
+If you need any help, please submit a [Github](https://github.com/OptimalScale/LMFlow) issue.
+
+## Indices and tables
+
+- {ref}`genindex`
+- {ref}`search`
diff --git a/_sphinx_design_static/design-tabs.js b/_sphinx_design_static/design-tabs.js
new file mode 100644
index 000000000..b25bd6a4f
--- /dev/null
+++ b/_sphinx_design_static/design-tabs.js
@@ -0,0 +1,101 @@
+// @ts-check
+
+// Extra JS capability for selected tabs to be synced
+// The selection is stored in local storage so that it persists across page loads.
+
+/**
+ * @type {Record<string, HTMLElement[]>}
+ */
+let sd_id_to_elements = {};
+const storageKeyPrefix = "sphinx-design-tab-id-";
+
+/**
+ * Create a key for a tab element.
+ * @param {HTMLElement} el - The tab element.
+ * @returns {[string, string, string] | null} - The key.
+ *
+ */
+function create_key(el) {
+  let syncId = el.getAttribute("data-sync-id");
+  let syncGroup = el.getAttribute("data-sync-group");
+  if (!syncId || !syncGroup) return null;
+  return [syncGroup, syncId, syncGroup + "--" + syncId];
+}
+
+/**
+ * Initialize the tab selection.
+ *
+ */
+function ready() {
+  // Find all tabs with sync data
+
+  /** @type {string[]} */
+  let groups = [];
+
+  document.querySelectorAll(".sd-tab-label").forEach((label) => {
+    if (label instanceof HTMLElement) {
+      let data = create_key(label);
+      if (data) {
+        let [group, id, key] = data;
+
+        // add click event listener
+        // @ts-ignore
+        label.onclick = onSDLabelClick;
+
+        // store map of key to elements
+        if (!sd_id_to_elements[key]) {
+          sd_id_to_elements[key] = [];
+        }
+        sd_id_to_elements[key].push(label);
+
+        if (groups.indexOf(group) === -1) {
+          groups.push(group);
+          // Check if a specific tab has been selected via URL parameter
+          const tabParam = new URLSearchParams(window.location.search).get(
+            group
+          );
+          if (tabParam) {
+            console.log(
+              "sphinx-design: Selecting tab id for group '" +
+                group +
+                "' from URL parameter: " +
+                tabParam
+            );
+            window.sessionStorage.setItem(storageKeyPrefix + group, tabParam);
+          }
+        }
+
+        // Check is a specific tab has been selected previously
+        let previousId = window.sessionStorage.getItem(
+          storageKeyPrefix + group
+        );
+        if (previousId === id) {
+          // console.log(
+          //   "sphinx-design: Selecting tab from session storage: " + id
+          // );
+          // @ts-ignore
+          label.previousElementSibling.checked = true;
+        }
+      }
+    }
+  });
+}
+
+/**
+ *  Activate other tabs with the same sync id.
+ *
+ * @this {HTMLElement} - The element that was clicked.
+ */
+function onSDLabelClick() {
+  let data = create_key(this);
+  if (!data) return;
+  let [group, id, key] = data;
+  for (const label of sd_id_to_elements[key]) {
+    if (label === this) continue;
+    // @ts-ignore
+    label.previousElementSibling.checked = true;
+  }
+  window.sessionStorage.setItem(storageKeyPrefix + group, id);
+}
+
+document.addEventListener("DOMContentLoaded", ready, false);
diff --git a/_sphinx_design_static/sphinx-design.min.css b/_sphinx_design_static/sphinx-design.min.css
new file mode 100644
index 000000000..860c36da0
--- /dev/null
+++ b/_sphinx_design_static/sphinx-design.min.css
@@ -0,0 +1 @@
+.sd-bg-primary{background-color:var(--sd-color-primary) !important}.sd-bg-text-primary{color:var(--sd-color-primary-text) !important}button.sd-bg-primary:focus,button.sd-bg-primary:hover{background-color:var(--sd-color-primary-highlight) !important}a.sd-bg-primary:focus,a.sd-bg-primary:hover{background-color:var(--sd-color-primary-highlight) !important}.sd-bg-secondary{background-color:var(--sd-color-secondary) !important}.sd-bg-text-secondary{color:var(--sd-color-secondary-text) !important}button.sd-bg-secondary:focus,button.sd-bg-secondary:hover{background-color:var(--sd-color-secondary-highlight) !important}a.sd-bg-secondary:focus,a.sd-bg-secondary:hover{background-color:var(--sd-color-secondary-highlight) !important}.sd-bg-success{background-color:var(--sd-color-success) !important}.sd-bg-text-success{color:var(--sd-color-success-text) !important}button.sd-bg-success:focus,button.sd-bg-success:hover{background-color:var(--sd-color-success-highlight) !important}a.sd-bg-success:focus,a.sd-bg-success:hover{background-color:var(--sd-color-success-highlight) !important}.sd-bg-info{background-color:var(--sd-color-info) !important}.sd-bg-text-info{color:var(--sd-color-info-text) !important}button.sd-bg-info:focus,button.sd-bg-info:hover{background-color:var(--sd-color-info-highlight) !important}a.sd-bg-info:focus,a.sd-bg-info:hover{background-color:var(--sd-color-info-highlight) !important}.sd-bg-warning{background-color:var(--sd-color-warning) !important}.sd-bg-text-warning{color:var(--sd-color-warning-text) !important}button.sd-bg-warning:focus,button.sd-bg-warning:hover{background-color:var(--sd-color-warning-highlight) !important}a.sd-bg-warning:focus,a.sd-bg-warning:hover{background-color:var(--sd-color-warning-highlight) !important}.sd-bg-danger{background-color:var(--sd-color-danger) !important}.sd-bg-text-danger{color:var(--sd-color-danger-text) !important}button.sd-bg-danger:focus,button.sd-bg-danger:hover{background-color:var(--sd-color-danger-highlight) !important}a.sd-bg-danger:focus,a.sd-bg-danger:hover{background-color:var(--sd-color-danger-highlight) !important}.sd-bg-light{background-color:var(--sd-color-light) !important}.sd-bg-text-light{color:var(--sd-color-light-text) !important}button.sd-bg-light:focus,button.sd-bg-light:hover{background-color:var(--sd-color-light-highlight) !important}a.sd-bg-light:focus,a.sd-bg-light:hover{background-color:var(--sd-color-light-highlight) !important}.sd-bg-muted{background-color:var(--sd-color-muted) !important}.sd-bg-text-muted{color:var(--sd-color-muted-text) !important}button.sd-bg-muted:focus,button.sd-bg-muted:hover{background-color:var(--sd-color-muted-highlight) !important}a.sd-bg-muted:focus,a.sd-bg-muted:hover{background-color:var(--sd-color-muted-highlight) !important}.sd-bg-dark{background-color:var(--sd-color-dark) !important}.sd-bg-text-dark{color:var(--sd-color-dark-text) !important}button.sd-bg-dark:focus,button.sd-bg-dark:hover{background-color:var(--sd-color-dark-highlight) !important}a.sd-bg-dark:focus,a.sd-bg-dark:hover{background-color:var(--sd-color-dark-highlight) !important}.sd-bg-black{background-color:var(--sd-color-black) !important}.sd-bg-text-black{color:var(--sd-color-black-text) !important}button.sd-bg-black:focus,button.sd-bg-black:hover{background-color:var(--sd-color-black-highlight) !important}a.sd-bg-black:focus,a.sd-bg-black:hover{background-color:var(--sd-color-black-highlight) !important}.sd-bg-white{background-color:var(--sd-color-white) !important}.sd-bg-text-white{color:var(--sd-color-white-text) !important}button.sd-bg-white:focus,button.sd-bg-white:hover{background-color:var(--sd-color-white-highlight) !important}a.sd-bg-white:focus,a.sd-bg-white:hover{background-color:var(--sd-color-white-highlight) !important}.sd-text-primary,.sd-text-primary>p{color:var(--sd-color-primary) !important}a.sd-text-primary:focus,a.sd-text-primary:hover{color:var(--sd-color-primary-highlight) !important}.sd-text-secondary,.sd-text-secondary>p{color:var(--sd-color-secondary) !important}a.sd-text-secondary:focus,a.sd-text-secondary:hover{color:var(--sd-color-secondary-highlight) !important}.sd-text-success,.sd-text-success>p{color:var(--sd-color-success) !important}a.sd-text-success:focus,a.sd-text-success:hover{color:var(--sd-color-success-highlight) !important}.sd-text-info,.sd-text-info>p{color:var(--sd-color-info) !important}a.sd-text-info:focus,a.sd-text-info:hover{color:var(--sd-color-info-highlight) !important}.sd-text-warning,.sd-text-warning>p{color:var(--sd-color-warning) !important}a.sd-text-warning:focus,a.sd-text-warning:hover{color:var(--sd-color-warning-highlight) !important}.sd-text-danger,.sd-text-danger>p{color:var(--sd-color-danger) !important}a.sd-text-danger:focus,a.sd-text-danger:hover{color:var(--sd-color-danger-highlight) !important}.sd-text-light,.sd-text-light>p{color:var(--sd-color-light) !important}a.sd-text-light:focus,a.sd-text-light:hover{color:var(--sd-color-light-highlight) !important}.sd-text-muted,.sd-text-muted>p{color:var(--sd-color-muted) !important}a.sd-text-muted:focus,a.sd-text-muted:hover{color:var(--sd-color-muted-highlight) !important}.sd-text-dark,.sd-text-dark>p{color:var(--sd-color-dark) !important}a.sd-text-dark:focus,a.sd-text-dark:hover{color:var(--sd-color-dark-highlight) !important}.sd-text-black,.sd-text-black>p{color:var(--sd-color-black) !important}a.sd-text-black:focus,a.sd-text-black:hover{color:var(--sd-color-black-highlight) !important}.sd-text-white,.sd-text-white>p{color:var(--sd-color-white) !important}a.sd-text-white:focus,a.sd-text-white:hover{color:var(--sd-color-white-highlight) !important}.sd-outline-primary{border-color:var(--sd-color-primary) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-primary:focus,a.sd-outline-primary:hover{border-color:var(--sd-color-primary-highlight) !important}.sd-outline-secondary{border-color:var(--sd-color-secondary) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-secondary:focus,a.sd-outline-secondary:hover{border-color:var(--sd-color-secondary-highlight) !important}.sd-outline-success{border-color:var(--sd-color-success) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-success:focus,a.sd-outline-success:hover{border-color:var(--sd-color-success-highlight) !important}.sd-outline-info{border-color:var(--sd-color-info) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-info:focus,a.sd-outline-info:hover{border-color:var(--sd-color-info-highlight) !important}.sd-outline-warning{border-color:var(--sd-color-warning) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-warning:focus,a.sd-outline-warning:hover{border-color:var(--sd-color-warning-highlight) !important}.sd-outline-danger{border-color:var(--sd-color-danger) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-danger:focus,a.sd-outline-danger:hover{border-color:var(--sd-color-danger-highlight) !important}.sd-outline-light{border-color:var(--sd-color-light) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-light:focus,a.sd-outline-light:hover{border-color:var(--sd-color-light-highlight) !important}.sd-outline-muted{border-color:var(--sd-color-muted) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-muted:focus,a.sd-outline-muted:hover{border-color:var(--sd-color-muted-highlight) !important}.sd-outline-dark{border-color:var(--sd-color-dark) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-dark:focus,a.sd-outline-dark:hover{border-color:var(--sd-color-dark-highlight) !important}.sd-outline-black{border-color:var(--sd-color-black) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-black:focus,a.sd-outline-black:hover{border-color:var(--sd-color-black-highlight) !important}.sd-outline-white{border-color:var(--sd-color-white) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-white:focus,a.sd-outline-white:hover{border-color:var(--sd-color-white-highlight) !important}.sd-bg-transparent{background-color:transparent !important}.sd-outline-transparent{border-color:transparent !important}.sd-text-transparent{color:transparent !important}.sd-p-0{padding:0 !important}.sd-pt-0,.sd-py-0{padding-top:0 !important}.sd-pr-0,.sd-px-0{padding-right:0 !important}.sd-pb-0,.sd-py-0{padding-bottom:0 !important}.sd-pl-0,.sd-px-0{padding-left:0 !important}.sd-p-1{padding:.25rem !important}.sd-pt-1,.sd-py-1{padding-top:.25rem !important}.sd-pr-1,.sd-px-1{padding-right:.25rem !important}.sd-pb-1,.sd-py-1{padding-bottom:.25rem !important}.sd-pl-1,.sd-px-1{padding-left:.25rem !important}.sd-p-2{padding:.5rem !important}.sd-pt-2,.sd-py-2{padding-top:.5rem !important}.sd-pr-2,.sd-px-2{padding-right:.5rem !important}.sd-pb-2,.sd-py-2{padding-bottom:.5rem !important}.sd-pl-2,.sd-px-2{padding-left:.5rem !important}.sd-p-3{padding:1rem !important}.sd-pt-3,.sd-py-3{padding-top:1rem !important}.sd-pr-3,.sd-px-3{padding-right:1rem !important}.sd-pb-3,.sd-py-3{padding-bottom:1rem !important}.sd-pl-3,.sd-px-3{padding-left:1rem !important}.sd-p-4{padding:1.5rem !important}.sd-pt-4,.sd-py-4{padding-top:1.5rem !important}.sd-pr-4,.sd-px-4{padding-right:1.5rem !important}.sd-pb-4,.sd-py-4{padding-bottom:1.5rem !important}.sd-pl-4,.sd-px-4{padding-left:1.5rem !important}.sd-p-5{padding:3rem !important}.sd-pt-5,.sd-py-5{padding-top:3rem !important}.sd-pr-5,.sd-px-5{padding-right:3rem !important}.sd-pb-5,.sd-py-5{padding-bottom:3rem !important}.sd-pl-5,.sd-px-5{padding-left:3rem !important}.sd-m-auto{margin:auto !important}.sd-mt-auto,.sd-my-auto{margin-top:auto !important}.sd-mr-auto,.sd-mx-auto{margin-right:auto !important}.sd-mb-auto,.sd-my-auto{margin-bottom:auto !important}.sd-ml-auto,.sd-mx-auto{margin-left:auto !important}.sd-m-0{margin:0 !important}.sd-mt-0,.sd-my-0{margin-top:0 !important}.sd-mr-0,.sd-mx-0{margin-right:0 !important}.sd-mb-0,.sd-my-0{margin-bottom:0 !important}.sd-ml-0,.sd-mx-0{margin-left:0 !important}.sd-m-1{margin:.25rem !important}.sd-mt-1,.sd-my-1{margin-top:.25rem !important}.sd-mr-1,.sd-mx-1{margin-right:.25rem !important}.sd-mb-1,.sd-my-1{margin-bottom:.25rem !important}.sd-ml-1,.sd-mx-1{margin-left:.25rem !important}.sd-m-2{margin:.5rem !important}.sd-mt-2,.sd-my-2{margin-top:.5rem !important}.sd-mr-2,.sd-mx-2{margin-right:.5rem !important}.sd-mb-2,.sd-my-2{margin-bottom:.5rem !important}.sd-ml-2,.sd-mx-2{margin-left:.5rem !important}.sd-m-3{margin:1rem !important}.sd-mt-3,.sd-my-3{margin-top:1rem !important}.sd-mr-3,.sd-mx-3{margin-right:1rem !important}.sd-mb-3,.sd-my-3{margin-bottom:1rem !important}.sd-ml-3,.sd-mx-3{margin-left:1rem !important}.sd-m-4{margin:1.5rem !important}.sd-mt-4,.sd-my-4{margin-top:1.5rem !important}.sd-mr-4,.sd-mx-4{margin-right:1.5rem !important}.sd-mb-4,.sd-my-4{margin-bottom:1.5rem !important}.sd-ml-4,.sd-mx-4{margin-left:1.5rem !important}.sd-m-5{margin:3rem !important}.sd-mt-5,.sd-my-5{margin-top:3rem !important}.sd-mr-5,.sd-mx-5{margin-right:3rem !important}.sd-mb-5,.sd-my-5{margin-bottom:3rem !important}.sd-ml-5,.sd-mx-5{margin-left:3rem !important}.sd-w-25{width:25% !important}.sd-w-50{width:50% !important}.sd-w-75{width:75% !important}.sd-w-100{width:100% !important}.sd-w-auto{width:auto !important}.sd-h-25{height:25% !important}.sd-h-50{height:50% !important}.sd-h-75{height:75% !important}.sd-h-100{height:100% !important}.sd-h-auto{height:auto !important}.sd-d-none{display:none !important}.sd-d-inline{display:inline !important}.sd-d-inline-block{display:inline-block !important}.sd-d-block{display:block !important}.sd-d-grid{display:grid !important}.sd-d-flex-row{display:-ms-flexbox !important;display:flex !important;flex-direction:row !important}.sd-d-flex-column{display:-ms-flexbox !important;display:flex !important;flex-direction:column !important}.sd-d-inline-flex{display:-ms-inline-flexbox !important;display:inline-flex !important}@media(min-width: 576px){.sd-d-sm-none{display:none !important}.sd-d-sm-inline{display:inline !important}.sd-d-sm-inline-block{display:inline-block !important}.sd-d-sm-block{display:block !important}.sd-d-sm-grid{display:grid !important}.sd-d-sm-flex{display:-ms-flexbox !important;display:flex !important}.sd-d-sm-inline-flex{display:-ms-inline-flexbox !important;display:inline-flex !important}}@media(min-width: 768px){.sd-d-md-none{display:none !important}.sd-d-md-inline{display:inline !important}.sd-d-md-inline-block{display:inline-block !important}.sd-d-md-block{display:block !important}.sd-d-md-grid{display:grid !important}.sd-d-md-flex{display:-ms-flexbox !important;display:flex !important}.sd-d-md-inline-flex{display:-ms-inline-flexbox !important;display:inline-flex !important}}@media(min-width: 992px){.sd-d-lg-none{display:none !important}.sd-d-lg-inline{display:inline !important}.sd-d-lg-inline-block{display:inline-block !important}.sd-d-lg-block{display:block !important}.sd-d-lg-grid{display:grid !important}.sd-d-lg-flex{display:-ms-flexbox !important;display:flex !important}.sd-d-lg-inline-flex{display:-ms-inline-flexbox !important;display:inline-flex !important}}@media(min-width: 1200px){.sd-d-xl-none{display:none !important}.sd-d-xl-inline{display:inline !important}.sd-d-xl-inline-block{display:inline-block !important}.sd-d-xl-block{display:block !important}.sd-d-xl-grid{display:grid !important}.sd-d-xl-flex{display:-ms-flexbox !important;display:flex !important}.sd-d-xl-inline-flex{display:-ms-inline-flexbox !important;display:inline-flex !important}}.sd-align-major-start{justify-content:flex-start !important}.sd-align-major-end{justify-content:flex-end !important}.sd-align-major-center{justify-content:center !important}.sd-align-major-justify{justify-content:space-between !important}.sd-align-major-spaced{justify-content:space-evenly !important}.sd-align-minor-start{align-items:flex-start !important}.sd-align-minor-end{align-items:flex-end !important}.sd-align-minor-center{align-items:center !important}.sd-align-minor-stretch{align-items:stretch !important}.sd-text-justify{text-align:justify !important}.sd-text-left{text-align:left !important}.sd-text-right{text-align:right !important}.sd-text-center{text-align:center !important}.sd-font-weight-light{font-weight:300 !important}.sd-font-weight-lighter{font-weight:lighter !important}.sd-font-weight-normal{font-weight:400 !important}.sd-font-weight-bold{font-weight:700 !important}.sd-font-weight-bolder{font-weight:bolder !important}.sd-font-italic{font-style:italic !important}.sd-text-decoration-none{text-decoration:none !important}.sd-text-lowercase{text-transform:lowercase !important}.sd-text-uppercase{text-transform:uppercase !important}.sd-text-capitalize{text-transform:capitalize !important}.sd-text-wrap{white-space:normal !important}.sd-text-nowrap{white-space:nowrap !important}.sd-text-truncate{overflow:hidden;text-overflow:ellipsis;white-space:nowrap}.sd-fs-1,.sd-fs-1>p{font-size:calc(1.375rem + 1.5vw) !important;line-height:unset !important}.sd-fs-2,.sd-fs-2>p{font-size:calc(1.325rem + 0.9vw) !important;line-height:unset !important}.sd-fs-3,.sd-fs-3>p{font-size:calc(1.3rem + 0.6vw) !important;line-height:unset !important}.sd-fs-4,.sd-fs-4>p{font-size:calc(1.275rem + 0.3vw) !important;line-height:unset !important}.sd-fs-5,.sd-fs-5>p{font-size:1.25rem !important;line-height:unset !important}.sd-fs-6,.sd-fs-6>p{font-size:1rem !important;line-height:unset !important}.sd-border-0{border:0 solid !important}.sd-border-top-0{border-top:0 solid !important}.sd-border-bottom-0{border-bottom:0 solid !important}.sd-border-right-0{border-right:0 solid !important}.sd-border-left-0{border-left:0 solid !important}.sd-border-1{border:1px solid !important}.sd-border-top-1{border-top:1px solid !important}.sd-border-bottom-1{border-bottom:1px solid !important}.sd-border-right-1{border-right:1px solid !important}.sd-border-left-1{border-left:1px solid !important}.sd-border-2{border:2px solid !important}.sd-border-top-2{border-top:2px solid !important}.sd-border-bottom-2{border-bottom:2px solid !important}.sd-border-right-2{border-right:2px solid !important}.sd-border-left-2{border-left:2px solid !important}.sd-border-3{border:3px solid !important}.sd-border-top-3{border-top:3px solid !important}.sd-border-bottom-3{border-bottom:3px solid !important}.sd-border-right-3{border-right:3px solid !important}.sd-border-left-3{border-left:3px solid !important}.sd-border-4{border:4px solid !important}.sd-border-top-4{border-top:4px solid !important}.sd-border-bottom-4{border-bottom:4px solid !important}.sd-border-right-4{border-right:4px solid !important}.sd-border-left-4{border-left:4px solid !important}.sd-border-5{border:5px solid !important}.sd-border-top-5{border-top:5px solid !important}.sd-border-bottom-5{border-bottom:5px solid !important}.sd-border-right-5{border-right:5px solid !important}.sd-border-left-5{border-left:5px solid !important}.sd-rounded-0{border-radius:0 !important}.sd-rounded-1{border-radius:.2rem !important}.sd-rounded-2{border-radius:.3rem !important}.sd-rounded-3{border-radius:.5rem !important}.sd-rounded-pill{border-radius:50rem !important}.sd-rounded-circle{border-radius:50% !important}.shadow-none{box-shadow:none !important}.sd-shadow-sm{box-shadow:0 .125rem .25rem var(--sd-color-shadow) !important}.sd-shadow-md{box-shadow:0 .5rem 1rem var(--sd-color-shadow) !important}.sd-shadow-lg{box-shadow:0 1rem 3rem var(--sd-color-shadow) !important}@keyframes sd-slide-from-left{0%{transform:translateX(-100%)}100%{transform:translateX(0)}}@keyframes sd-slide-from-right{0%{transform:translateX(200%)}100%{transform:translateX(0)}}@keyframes sd-grow100{0%{transform:scale(0);opacity:.5}100%{transform:scale(1);opacity:1}}@keyframes sd-grow50{0%{transform:scale(0.5);opacity:.5}100%{transform:scale(1);opacity:1}}@keyframes sd-grow50-rot20{0%{transform:scale(0.5) rotateZ(-20deg);opacity:.5}75%{transform:scale(1) rotateZ(5deg);opacity:1}95%{transform:scale(1) rotateZ(-1deg);opacity:1}100%{transform:scale(1) rotateZ(0);opacity:1}}.sd-animate-slide-from-left{animation:1s ease-out 0s 1 normal none running sd-slide-from-left}.sd-animate-slide-from-right{animation:1s ease-out 0s 1 normal none running sd-slide-from-right}.sd-animate-grow100{animation:1s ease-out 0s 1 normal none running sd-grow100}.sd-animate-grow50{animation:1s ease-out 0s 1 normal none running sd-grow50}.sd-animate-grow50-rot20{animation:1s ease-out 0s 1 normal none running sd-grow50-rot20}.sd-badge{display:inline-block;padding:.35em .65em;font-size:.75em;font-weight:700;line-height:1;text-align:center;white-space:nowrap;vertical-align:baseline;border-radius:.25rem}.sd-badge:empty{display:none}a.sd-badge{text-decoration:none}.sd-btn .sd-badge{position:relative;top:-1px}.sd-btn{background-color:transparent;border:1px solid transparent;border-radius:.25rem;cursor:pointer;display:inline-block;font-weight:400;font-size:1rem;line-height:1.5;padding:.375rem .75rem;text-align:center;text-decoration:none;transition:color .15s ease-in-out,background-color .15s ease-in-out,border-color .15s ease-in-out,box-shadow .15s ease-in-out;vertical-align:middle;user-select:none;-moz-user-select:none;-ms-user-select:none;-webkit-user-select:none}.sd-btn:hover{text-decoration:none}@media(prefers-reduced-motion: reduce){.sd-btn{transition:none}}.sd-btn-primary,.sd-btn-outline-primary:hover,.sd-btn-outline-primary:focus{color:var(--sd-color-primary-text) !important;background-color:var(--sd-color-primary) !important;border-color:var(--sd-color-primary) !important;border-width:1px !important;border-style:solid !important}.sd-btn-primary:hover,.sd-btn-primary:focus{color:var(--sd-color-primary-text) !important;background-color:var(--sd-color-primary-highlight) !important;border-color:var(--sd-color-primary-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-primary{color:var(--sd-color-primary) !important;border-color:var(--sd-color-primary) !important;border-width:1px !important;border-style:solid !important}.sd-btn-secondary,.sd-btn-outline-secondary:hover,.sd-btn-outline-secondary:focus{color:var(--sd-color-secondary-text) !important;background-color:var(--sd-color-secondary) !important;border-color:var(--sd-color-secondary) !important;border-width:1px !important;border-style:solid !important}.sd-btn-secondary:hover,.sd-btn-secondary:focus{color:var(--sd-color-secondary-text) !important;background-color:var(--sd-color-secondary-highlight) !important;border-color:var(--sd-color-secondary-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-secondary{color:var(--sd-color-secondary) !important;border-color:var(--sd-color-secondary) !important;border-width:1px !important;border-style:solid !important}.sd-btn-success,.sd-btn-outline-success:hover,.sd-btn-outline-success:focus{color:var(--sd-color-success-text) !important;background-color:var(--sd-color-success) !important;border-color:var(--sd-color-success) !important;border-width:1px !important;border-style:solid !important}.sd-btn-success:hover,.sd-btn-success:focus{color:var(--sd-color-success-text) !important;background-color:var(--sd-color-success-highlight) !important;border-color:var(--sd-color-success-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-success{color:var(--sd-color-success) !important;border-color:var(--sd-color-success) !important;border-width:1px !important;border-style:solid !important}.sd-btn-info,.sd-btn-outline-info:hover,.sd-btn-outline-info:focus{color:var(--sd-color-info-text) !important;background-color:var(--sd-color-info) !important;border-color:var(--sd-color-info) !important;border-width:1px !important;border-style:solid !important}.sd-btn-info:hover,.sd-btn-info:focus{color:var(--sd-color-info-text) !important;background-color:var(--sd-color-info-highlight) !important;border-color:var(--sd-color-info-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-info{color:var(--sd-color-info) !important;border-color:var(--sd-color-info) !important;border-width:1px !important;border-style:solid !important}.sd-btn-warning,.sd-btn-outline-warning:hover,.sd-btn-outline-warning:focus{color:var(--sd-color-warning-text) !important;background-color:var(--sd-color-warning) !important;border-color:var(--sd-color-warning) !important;border-width:1px !important;border-style:solid !important}.sd-btn-warning:hover,.sd-btn-warning:focus{color:var(--sd-color-warning-text) !important;background-color:var(--sd-color-warning-highlight) !important;border-color:var(--sd-color-warning-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-warning{color:var(--sd-color-warning) !important;border-color:var(--sd-color-warning) !important;border-width:1px !important;border-style:solid !important}.sd-btn-danger,.sd-btn-outline-danger:hover,.sd-btn-outline-danger:focus{color:var(--sd-color-danger-text) !important;background-color:var(--sd-color-danger) !important;border-color:var(--sd-color-danger) !important;border-width:1px !important;border-style:solid !important}.sd-btn-danger:hover,.sd-btn-danger:focus{color:var(--sd-color-danger-text) !important;background-color:var(--sd-color-danger-highlight) !important;border-color:var(--sd-color-danger-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-danger{color:var(--sd-color-danger) !important;border-color:var(--sd-color-danger) !important;border-width:1px !important;border-style:solid !important}.sd-btn-light,.sd-btn-outline-light:hover,.sd-btn-outline-light:focus{color:var(--sd-color-light-text) !important;background-color:var(--sd-color-light) !important;border-color:var(--sd-color-light) !important;border-width:1px !important;border-style:solid !important}.sd-btn-light:hover,.sd-btn-light:focus{color:var(--sd-color-light-text) !important;background-color:var(--sd-color-light-highlight) !important;border-color:var(--sd-color-light-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-light{color:var(--sd-color-light) !important;border-color:var(--sd-color-light) !important;border-width:1px !important;border-style:solid !important}.sd-btn-muted,.sd-btn-outline-muted:hover,.sd-btn-outline-muted:focus{color:var(--sd-color-muted-text) !important;background-color:var(--sd-color-muted) !important;border-color:var(--sd-color-muted) !important;border-width:1px !important;border-style:solid !important}.sd-btn-muted:hover,.sd-btn-muted:focus{color:var(--sd-color-muted-text) !important;background-color:var(--sd-color-muted-highlight) !important;border-color:var(--sd-color-muted-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-muted{color:var(--sd-color-muted) !important;border-color:var(--sd-color-muted) !important;border-width:1px !important;border-style:solid !important}.sd-btn-dark,.sd-btn-outline-dark:hover,.sd-btn-outline-dark:focus{color:var(--sd-color-dark-text) !important;background-color:var(--sd-color-dark) !important;border-color:var(--sd-color-dark) !important;border-width:1px !important;border-style:solid !important}.sd-btn-dark:hover,.sd-btn-dark:focus{color:var(--sd-color-dark-text) !important;background-color:var(--sd-color-dark-highlight) !important;border-color:var(--sd-color-dark-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-dark{color:var(--sd-color-dark) !important;border-color:var(--sd-color-dark) !important;border-width:1px !important;border-style:solid !important}.sd-btn-black,.sd-btn-outline-black:hover,.sd-btn-outline-black:focus{color:var(--sd-color-black-text) !important;background-color:var(--sd-color-black) !important;border-color:var(--sd-color-black) !important;border-width:1px !important;border-style:solid !important}.sd-btn-black:hover,.sd-btn-black:focus{color:var(--sd-color-black-text) !important;background-color:var(--sd-color-black-highlight) !important;border-color:var(--sd-color-black-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-black{color:var(--sd-color-black) !important;border-color:var(--sd-color-black) !important;border-width:1px !important;border-style:solid !important}.sd-btn-white,.sd-btn-outline-white:hover,.sd-btn-outline-white:focus{color:var(--sd-color-white-text) !important;background-color:var(--sd-color-white) !important;border-color:var(--sd-color-white) !important;border-width:1px !important;border-style:solid !important}.sd-btn-white:hover,.sd-btn-white:focus{color:var(--sd-color-white-text) !important;background-color:var(--sd-color-white-highlight) !important;border-color:var(--sd-color-white-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-white{color:var(--sd-color-white) !important;border-color:var(--sd-color-white) !important;border-width:1px !important;border-style:solid !important}.sd-stretched-link::after{position:absolute;top:0;right:0;bottom:0;left:0;z-index:1;content:""}.sd-hide-link-text{font-size:0}.sd-octicon,.sd-material-icon{display:inline-block;fill:currentColor;vertical-align:middle}.sd-avatar-xs{border-radius:50%;object-fit:cover;object-position:center;width:1rem;height:1rem}.sd-avatar-sm{border-radius:50%;object-fit:cover;object-position:center;width:3rem;height:3rem}.sd-avatar-md{border-radius:50%;object-fit:cover;object-position:center;width:5rem;height:5rem}.sd-avatar-lg{border-radius:50%;object-fit:cover;object-position:center;width:7rem;height:7rem}.sd-avatar-xl{border-radius:50%;object-fit:cover;object-position:center;width:10rem;height:10rem}.sd-avatar-inherit{border-radius:50%;object-fit:cover;object-position:center;width:inherit;height:inherit}.sd-avatar-initial{border-radius:50%;object-fit:cover;object-position:center;width:initial;height:initial}.sd-card{background-clip:border-box;background-color:var(--sd-color-card-background);border:1px solid var(--sd-color-card-border);border-radius:.25rem;color:var(--sd-color-card-text);display:-ms-flexbox;display:flex;-ms-flex-direction:column;flex-direction:column;min-width:0;position:relative;word-wrap:break-word}.sd-card>hr{margin-left:0;margin-right:0}.sd-card-hover:hover{border-color:var(--sd-color-card-border-hover);transform:scale(1.01)}.sd-card-body{-ms-flex:1 1 auto;flex:1 1 auto;padding:1rem 1rem}.sd-card-title{margin-bottom:.5rem}.sd-card-subtitle{margin-top:-0.25rem;margin-bottom:0}.sd-card-text:last-child{margin-bottom:0}.sd-card-link:hover{text-decoration:none}.sd-card-link+.card-link{margin-left:1rem}.sd-card-header{padding:.5rem 1rem;margin-bottom:0;background-color:var(--sd-color-card-header);border-bottom:1px solid var(--sd-color-card-border)}.sd-card-header:first-child{border-radius:calc(0.25rem - 1px) calc(0.25rem - 1px) 0 0}.sd-card-footer{padding:.5rem 1rem;background-color:var(--sd-color-card-footer);border-top:1px solid var(--sd-color-card-border)}.sd-card-footer:last-child{border-radius:0 0 calc(0.25rem - 1px) calc(0.25rem - 1px)}.sd-card-header-tabs{margin-right:-0.5rem;margin-bottom:-0.5rem;margin-left:-0.5rem;border-bottom:0}.sd-card-header-pills{margin-right:-0.5rem;margin-left:-0.5rem}.sd-card-img-overlay{position:absolute;top:0;right:0;bottom:0;left:0;padding:1rem;border-radius:calc(0.25rem - 1px)}.sd-card-img,.sd-card-img-bottom,.sd-card-img-top{width:100%}.sd-card-img,.sd-card-img-top{border-top-left-radius:calc(0.25rem - 1px);border-top-right-radius:calc(0.25rem - 1px)}.sd-card-img,.sd-card-img-bottom{border-bottom-left-radius:calc(0.25rem - 1px);border-bottom-right-radius:calc(0.25rem - 1px)}.sd-cards-carousel{width:100%;display:flex;flex-wrap:nowrap;-ms-flex-direction:row;flex-direction:row;overflow-x:hidden;scroll-snap-type:x mandatory}.sd-cards-carousel.sd-show-scrollbar{overflow-x:auto}.sd-cards-carousel:hover,.sd-cards-carousel:focus{overflow-x:auto}.sd-cards-carousel>.sd-card{flex-shrink:0;scroll-snap-align:start}.sd-cards-carousel>.sd-card:not(:last-child){margin-right:3px}.sd-card-cols-1>.sd-card{width:90%}.sd-card-cols-2>.sd-card{width:45%}.sd-card-cols-3>.sd-card{width:30%}.sd-card-cols-4>.sd-card{width:22.5%}.sd-card-cols-5>.sd-card{width:18%}.sd-card-cols-6>.sd-card{width:15%}.sd-card-cols-7>.sd-card{width:12.8571428571%}.sd-card-cols-8>.sd-card{width:11.25%}.sd-card-cols-9>.sd-card{width:10%}.sd-card-cols-10>.sd-card{width:9%}.sd-card-cols-11>.sd-card{width:8.1818181818%}.sd-card-cols-12>.sd-card{width:7.5%}.sd-container,.sd-container-fluid,.sd-container-lg,.sd-container-md,.sd-container-sm,.sd-container-xl{margin-left:auto;margin-right:auto;padding-left:var(--sd-gutter-x, 0.75rem);padding-right:var(--sd-gutter-x, 0.75rem);width:100%}@media(min-width: 576px){.sd-container-sm,.sd-container{max-width:540px}}@media(min-width: 768px){.sd-container-md,.sd-container-sm,.sd-container{max-width:720px}}@media(min-width: 992px){.sd-container-lg,.sd-container-md,.sd-container-sm,.sd-container{max-width:960px}}@media(min-width: 1200px){.sd-container-xl,.sd-container-lg,.sd-container-md,.sd-container-sm,.sd-container{max-width:1140px}}.sd-row{--sd-gutter-x: 1.5rem;--sd-gutter-y: 0;display:-ms-flexbox;display:flex;-ms-flex-wrap:wrap;flex-wrap:wrap;margin-top:calc(var(--sd-gutter-y) * -1);margin-right:calc(var(--sd-gutter-x) * -0.5);margin-left:calc(var(--sd-gutter-x) * -0.5)}.sd-row>*{box-sizing:border-box;flex-shrink:0;width:100%;max-width:100%;padding-right:calc(var(--sd-gutter-x) * 0.5);padding-left:calc(var(--sd-gutter-x) * 0.5);margin-top:var(--sd-gutter-y)}.sd-col{flex:1 0 0%;-ms-flex:1 0 0%}.sd-row-cols-auto>*{flex:0 0 auto;width:auto}.sd-row-cols-1>*{flex:0 0 auto;-ms-flex:0 0 auto;width:100%}.sd-row-cols-2>*{flex:0 0 auto;-ms-flex:0 0 auto;width:50%}.sd-row-cols-3>*{flex:0 0 auto;-ms-flex:0 0 auto;width:33.3333333333%}.sd-row-cols-4>*{flex:0 0 auto;-ms-flex:0 0 auto;width:25%}.sd-row-cols-5>*{flex:0 0 auto;-ms-flex:0 0 auto;width:20%}.sd-row-cols-6>*{flex:0 0 auto;-ms-flex:0 0 auto;width:16.6666666667%}.sd-row-cols-7>*{flex:0 0 auto;-ms-flex:0 0 auto;width:14.2857142857%}.sd-row-cols-8>*{flex:0 0 auto;-ms-flex:0 0 auto;width:12.5%}.sd-row-cols-9>*{flex:0 0 auto;-ms-flex:0 0 auto;width:11.1111111111%}.sd-row-cols-10>*{flex:0 0 auto;-ms-flex:0 0 auto;width:10%}.sd-row-cols-11>*{flex:0 0 auto;-ms-flex:0 0 auto;width:9.0909090909%}.sd-row-cols-12>*{flex:0 0 auto;-ms-flex:0 0 auto;width:8.3333333333%}@media(min-width: 576px){.sd-col-sm{flex:1 0 0%;-ms-flex:1 0 0%}.sd-row-cols-sm-auto{flex:1 0 auto;-ms-flex:1 0 auto;width:100%}.sd-row-cols-sm-1>*{flex:0 0 auto;-ms-flex:0 0 auto;width:100%}.sd-row-cols-sm-2>*{flex:0 0 auto;-ms-flex:0 0 auto;width:50%}.sd-row-cols-sm-3>*{flex:0 0 auto;-ms-flex:0 0 auto;width:33.3333333333%}.sd-row-cols-sm-4>*{flex:0 0 auto;-ms-flex:0 0 auto;width:25%}.sd-row-cols-sm-5>*{flex:0 0 auto;-ms-flex:0 0 auto;width:20%}.sd-row-cols-sm-6>*{flex:0 0 auto;-ms-flex:0 0 auto;width:16.6666666667%}.sd-row-cols-sm-7>*{flex:0 0 auto;-ms-flex:0 0 auto;width:14.2857142857%}.sd-row-cols-sm-8>*{flex:0 0 auto;-ms-flex:0 0 auto;width:12.5%}.sd-row-cols-sm-9>*{flex:0 0 auto;-ms-flex:0 0 auto;width:11.1111111111%}.sd-row-cols-sm-10>*{flex:0 0 auto;-ms-flex:0 0 auto;width:10%}.sd-row-cols-sm-11>*{flex:0 0 auto;-ms-flex:0 0 auto;width:9.0909090909%}.sd-row-cols-sm-12>*{flex:0 0 auto;-ms-flex:0 0 auto;width:8.3333333333%}}@media(min-width: 768px){.sd-col-md{flex:1 0 0%;-ms-flex:1 0 0%}.sd-row-cols-md-auto{flex:1 0 auto;-ms-flex:1 0 auto;width:100%}.sd-row-cols-md-1>*{flex:0 0 auto;-ms-flex:0 0 auto;width:100%}.sd-row-cols-md-2>*{flex:0 0 auto;-ms-flex:0 0 auto;width:50%}.sd-row-cols-md-3>*{flex:0 0 auto;-ms-flex:0 0 auto;width:33.3333333333%}.sd-row-cols-md-4>*{flex:0 0 auto;-ms-flex:0 0 auto;width:25%}.sd-row-cols-md-5>*{flex:0 0 auto;-ms-flex:0 0 auto;width:20%}.sd-row-cols-md-6>*{flex:0 0 auto;-ms-flex:0 0 auto;width:16.6666666667%}.sd-row-cols-md-7>*{flex:0 0 auto;-ms-flex:0 0 auto;width:14.2857142857%}.sd-row-cols-md-8>*{flex:0 0 auto;-ms-flex:0 0 auto;width:12.5%}.sd-row-cols-md-9>*{flex:0 0 auto;-ms-flex:0 0 auto;width:11.1111111111%}.sd-row-cols-md-10>*{flex:0 0 auto;-ms-flex:0 0 auto;width:10%}.sd-row-cols-md-11>*{flex:0 0 auto;-ms-flex:0 0 auto;width:9.0909090909%}.sd-row-cols-md-12>*{flex:0 0 auto;-ms-flex:0 0 auto;width:8.3333333333%}}@media(min-width: 992px){.sd-col-lg{flex:1 0 0%;-ms-flex:1 0 0%}.sd-row-cols-lg-auto{flex:1 0 auto;-ms-flex:1 0 auto;width:100%}.sd-row-cols-lg-1>*{flex:0 0 auto;-ms-flex:0 0 auto;width:100%}.sd-row-cols-lg-2>*{flex:0 0 auto;-ms-flex:0 0 auto;width:50%}.sd-row-cols-lg-3>*{flex:0 0 auto;-ms-flex:0 0 auto;width:33.3333333333%}.sd-row-cols-lg-4>*{flex:0 0 auto;-ms-flex:0 0 auto;width:25%}.sd-row-cols-lg-5>*{flex:0 0 auto;-ms-flex:0 0 auto;width:20%}.sd-row-cols-lg-6>*{flex:0 0 auto;-ms-flex:0 0 auto;width:16.6666666667%}.sd-row-cols-lg-7>*{flex:0 0 auto;-ms-flex:0 0 auto;width:14.2857142857%}.sd-row-cols-lg-8>*{flex:0 0 auto;-ms-flex:0 0 auto;width:12.5%}.sd-row-cols-lg-9>*{flex:0 0 auto;-ms-flex:0 0 auto;width:11.1111111111%}.sd-row-cols-lg-10>*{flex:0 0 auto;-ms-flex:0 0 auto;width:10%}.sd-row-cols-lg-11>*{flex:0 0 auto;-ms-flex:0 0 auto;width:9.0909090909%}.sd-row-cols-lg-12>*{flex:0 0 auto;-ms-flex:0 0 auto;width:8.3333333333%}}@media(min-width: 1200px){.sd-col-xl{flex:1 0 0%;-ms-flex:1 0 0%}.sd-row-cols-xl-auto{flex:1 0 auto;-ms-flex:1 0 auto;width:100%}.sd-row-cols-xl-1>*{flex:0 0 auto;-ms-flex:0 0 auto;width:100%}.sd-row-cols-xl-2>*{flex:0 0 auto;-ms-flex:0 0 auto;width:50%}.sd-row-cols-xl-3>*{flex:0 0 auto;-ms-flex:0 0 auto;width:33.3333333333%}.sd-row-cols-xl-4>*{flex:0 0 auto;-ms-flex:0 0 auto;width:25%}.sd-row-cols-xl-5>*{flex:0 0 auto;-ms-flex:0 0 auto;width:20%}.sd-row-cols-xl-6>*{flex:0 0 auto;-ms-flex:0 0 auto;width:16.6666666667%}.sd-row-cols-xl-7>*{flex:0 0 auto;-ms-flex:0 0 auto;width:14.2857142857%}.sd-row-cols-xl-8>*{flex:0 0 auto;-ms-flex:0 0 auto;width:12.5%}.sd-row-cols-xl-9>*{flex:0 0 auto;-ms-flex:0 0 auto;width:11.1111111111%}.sd-row-cols-xl-10>*{flex:0 0 auto;-ms-flex:0 0 auto;width:10%}.sd-row-cols-xl-11>*{flex:0 0 auto;-ms-flex:0 0 auto;width:9.0909090909%}.sd-row-cols-xl-12>*{flex:0 0 auto;-ms-flex:0 0 auto;width:8.3333333333%}}.sd-col-auto{flex:0 0 auto;-ms-flex:0 0 auto;width:auto}.sd-col-1{flex:0 0 auto;-ms-flex:0 0 auto;width:8.3333333333%}.sd-col-2{flex:0 0 auto;-ms-flex:0 0 auto;width:16.6666666667%}.sd-col-3{flex:0 0 auto;-ms-flex:0 0 auto;width:25%}.sd-col-4{flex:0 0 auto;-ms-flex:0 0 auto;width:33.3333333333%}.sd-col-5{flex:0 0 auto;-ms-flex:0 0 auto;width:41.6666666667%}.sd-col-6{flex:0 0 auto;-ms-flex:0 0 auto;width:50%}.sd-col-7{flex:0 0 auto;-ms-flex:0 0 auto;width:58.3333333333%}.sd-col-8{flex:0 0 auto;-ms-flex:0 0 auto;width:66.6666666667%}.sd-col-9{flex:0 0 auto;-ms-flex:0 0 auto;width:75%}.sd-col-10{flex:0 0 auto;-ms-flex:0 0 auto;width:83.3333333333%}.sd-col-11{flex:0 0 auto;-ms-flex:0 0 auto;width:91.6666666667%}.sd-col-12{flex:0 0 auto;-ms-flex:0 0 auto;width:100%}.sd-g-0,.sd-gy-0{--sd-gutter-y: 0}.sd-g-0,.sd-gx-0{--sd-gutter-x: 0}.sd-g-1,.sd-gy-1{--sd-gutter-y: 0.25rem}.sd-g-1,.sd-gx-1{--sd-gutter-x: 0.25rem}.sd-g-2,.sd-gy-2{--sd-gutter-y: 0.5rem}.sd-g-2,.sd-gx-2{--sd-gutter-x: 0.5rem}.sd-g-3,.sd-gy-3{--sd-gutter-y: 1rem}.sd-g-3,.sd-gx-3{--sd-gutter-x: 1rem}.sd-g-4,.sd-gy-4{--sd-gutter-y: 1.5rem}.sd-g-4,.sd-gx-4{--sd-gutter-x: 1.5rem}.sd-g-5,.sd-gy-5{--sd-gutter-y: 3rem}.sd-g-5,.sd-gx-5{--sd-gutter-x: 3rem}@media(min-width: 576px){.sd-col-sm-auto{-ms-flex:0 0 auto;flex:0 0 auto;width:auto}.sd-col-sm-1{-ms-flex:0 0 auto;flex:0 0 auto;width:8.3333333333%}.sd-col-sm-2{-ms-flex:0 0 auto;flex:0 0 auto;width:16.6666666667%}.sd-col-sm-3{-ms-flex:0 0 auto;flex:0 0 auto;width:25%}.sd-col-sm-4{-ms-flex:0 0 auto;flex:0 0 auto;width:33.3333333333%}.sd-col-sm-5{-ms-flex:0 0 auto;flex:0 0 auto;width:41.6666666667%}.sd-col-sm-6{-ms-flex:0 0 auto;flex:0 0 auto;width:50%}.sd-col-sm-7{-ms-flex:0 0 auto;flex:0 0 auto;width:58.3333333333%}.sd-col-sm-8{-ms-flex:0 0 auto;flex:0 0 auto;width:66.6666666667%}.sd-col-sm-9{-ms-flex:0 0 auto;flex:0 0 auto;width:75%}.sd-col-sm-10{-ms-flex:0 0 auto;flex:0 0 auto;width:83.3333333333%}.sd-col-sm-11{-ms-flex:0 0 auto;flex:0 0 auto;width:91.6666666667%}.sd-col-sm-12{-ms-flex:0 0 auto;flex:0 0 auto;width:100%}.sd-g-sm-0,.sd-gy-sm-0{--sd-gutter-y: 0}.sd-g-sm-0,.sd-gx-sm-0{--sd-gutter-x: 0}.sd-g-sm-1,.sd-gy-sm-1{--sd-gutter-y: 0.25rem}.sd-g-sm-1,.sd-gx-sm-1{--sd-gutter-x: 0.25rem}.sd-g-sm-2,.sd-gy-sm-2{--sd-gutter-y: 0.5rem}.sd-g-sm-2,.sd-gx-sm-2{--sd-gutter-x: 0.5rem}.sd-g-sm-3,.sd-gy-sm-3{--sd-gutter-y: 1rem}.sd-g-sm-3,.sd-gx-sm-3{--sd-gutter-x: 1rem}.sd-g-sm-4,.sd-gy-sm-4{--sd-gutter-y: 1.5rem}.sd-g-sm-4,.sd-gx-sm-4{--sd-gutter-x: 1.5rem}.sd-g-sm-5,.sd-gy-sm-5{--sd-gutter-y: 3rem}.sd-g-sm-5,.sd-gx-sm-5{--sd-gutter-x: 3rem}}@media(min-width: 768px){.sd-col-md-auto{-ms-flex:0 0 auto;flex:0 0 auto;width:auto}.sd-col-md-1{-ms-flex:0 0 auto;flex:0 0 auto;width:8.3333333333%}.sd-col-md-2{-ms-flex:0 0 auto;flex:0 0 auto;width:16.6666666667%}.sd-col-md-3{-ms-flex:0 0 auto;flex:0 0 auto;width:25%}.sd-col-md-4{-ms-flex:0 0 auto;flex:0 0 auto;width:33.3333333333%}.sd-col-md-5{-ms-flex:0 0 auto;flex:0 0 auto;width:41.6666666667%}.sd-col-md-6{-ms-flex:0 0 auto;flex:0 0 auto;width:50%}.sd-col-md-7{-ms-flex:0 0 auto;flex:0 0 auto;width:58.3333333333%}.sd-col-md-8{-ms-flex:0 0 auto;flex:0 0 auto;width:66.6666666667%}.sd-col-md-9{-ms-flex:0 0 auto;flex:0 0 auto;width:75%}.sd-col-md-10{-ms-flex:0 0 auto;flex:0 0 auto;width:83.3333333333%}.sd-col-md-11{-ms-flex:0 0 auto;flex:0 0 auto;width:91.6666666667%}.sd-col-md-12{-ms-flex:0 0 auto;flex:0 0 auto;width:100%}.sd-g-md-0,.sd-gy-md-0{--sd-gutter-y: 0}.sd-g-md-0,.sd-gx-md-0{--sd-gutter-x: 0}.sd-g-md-1,.sd-gy-md-1{--sd-gutter-y: 0.25rem}.sd-g-md-1,.sd-gx-md-1{--sd-gutter-x: 0.25rem}.sd-g-md-2,.sd-gy-md-2{--sd-gutter-y: 0.5rem}.sd-g-md-2,.sd-gx-md-2{--sd-gutter-x: 0.5rem}.sd-g-md-3,.sd-gy-md-3{--sd-gutter-y: 1rem}.sd-g-md-3,.sd-gx-md-3{--sd-gutter-x: 1rem}.sd-g-md-4,.sd-gy-md-4{--sd-gutter-y: 1.5rem}.sd-g-md-4,.sd-gx-md-4{--sd-gutter-x: 1.5rem}.sd-g-md-5,.sd-gy-md-5{--sd-gutter-y: 3rem}.sd-g-md-5,.sd-gx-md-5{--sd-gutter-x: 3rem}}@media(min-width: 992px){.sd-col-lg-auto{-ms-flex:0 0 auto;flex:0 0 auto;width:auto}.sd-col-lg-1{-ms-flex:0 0 auto;flex:0 0 auto;width:8.3333333333%}.sd-col-lg-2{-ms-flex:0 0 auto;flex:0 0 auto;width:16.6666666667%}.sd-col-lg-3{-ms-flex:0 0 auto;flex:0 0 auto;width:25%}.sd-col-lg-4{-ms-flex:0 0 auto;flex:0 0 auto;width:33.3333333333%}.sd-col-lg-5{-ms-flex:0 0 auto;flex:0 0 auto;width:41.6666666667%}.sd-col-lg-6{-ms-flex:0 0 auto;flex:0 0 auto;width:50%}.sd-col-lg-7{-ms-flex:0 0 auto;flex:0 0 auto;width:58.3333333333%}.sd-col-lg-8{-ms-flex:0 0 auto;flex:0 0 auto;width:66.6666666667%}.sd-col-lg-9{-ms-flex:0 0 auto;flex:0 0 auto;width:75%}.sd-col-lg-10{-ms-flex:0 0 auto;flex:0 0 auto;width:83.3333333333%}.sd-col-lg-11{-ms-flex:0 0 auto;flex:0 0 auto;width:91.6666666667%}.sd-col-lg-12{-ms-flex:0 0 auto;flex:0 0 auto;width:100%}.sd-g-lg-0,.sd-gy-lg-0{--sd-gutter-y: 0}.sd-g-lg-0,.sd-gx-lg-0{--sd-gutter-x: 0}.sd-g-lg-1,.sd-gy-lg-1{--sd-gutter-y: 0.25rem}.sd-g-lg-1,.sd-gx-lg-1{--sd-gutter-x: 0.25rem}.sd-g-lg-2,.sd-gy-lg-2{--sd-gutter-y: 0.5rem}.sd-g-lg-2,.sd-gx-lg-2{--sd-gutter-x: 0.5rem}.sd-g-lg-3,.sd-gy-lg-3{--sd-gutter-y: 1rem}.sd-g-lg-3,.sd-gx-lg-3{--sd-gutter-x: 1rem}.sd-g-lg-4,.sd-gy-lg-4{--sd-gutter-y: 1.5rem}.sd-g-lg-4,.sd-gx-lg-4{--sd-gutter-x: 1.5rem}.sd-g-lg-5,.sd-gy-lg-5{--sd-gutter-y: 3rem}.sd-g-lg-5,.sd-gx-lg-5{--sd-gutter-x: 3rem}}@media(min-width: 1200px){.sd-col-xl-auto{-ms-flex:0 0 auto;flex:0 0 auto;width:auto}.sd-col-xl-1{-ms-flex:0 0 auto;flex:0 0 auto;width:8.3333333333%}.sd-col-xl-2{-ms-flex:0 0 auto;flex:0 0 auto;width:16.6666666667%}.sd-col-xl-3{-ms-flex:0 0 auto;flex:0 0 auto;width:25%}.sd-col-xl-4{-ms-flex:0 0 auto;flex:0 0 auto;width:33.3333333333%}.sd-col-xl-5{-ms-flex:0 0 auto;flex:0 0 auto;width:41.6666666667%}.sd-col-xl-6{-ms-flex:0 0 auto;flex:0 0 auto;width:50%}.sd-col-xl-7{-ms-flex:0 0 auto;flex:0 0 auto;width:58.3333333333%}.sd-col-xl-8{-ms-flex:0 0 auto;flex:0 0 auto;width:66.6666666667%}.sd-col-xl-9{-ms-flex:0 0 auto;flex:0 0 auto;width:75%}.sd-col-xl-10{-ms-flex:0 0 auto;flex:0 0 auto;width:83.3333333333%}.sd-col-xl-11{-ms-flex:0 0 auto;flex:0 0 auto;width:91.6666666667%}.sd-col-xl-12{-ms-flex:0 0 auto;flex:0 0 auto;width:100%}.sd-g-xl-0,.sd-gy-xl-0{--sd-gutter-y: 0}.sd-g-xl-0,.sd-gx-xl-0{--sd-gutter-x: 0}.sd-g-xl-1,.sd-gy-xl-1{--sd-gutter-y: 0.25rem}.sd-g-xl-1,.sd-gx-xl-1{--sd-gutter-x: 0.25rem}.sd-g-xl-2,.sd-gy-xl-2{--sd-gutter-y: 0.5rem}.sd-g-xl-2,.sd-gx-xl-2{--sd-gutter-x: 0.5rem}.sd-g-xl-3,.sd-gy-xl-3{--sd-gutter-y: 1rem}.sd-g-xl-3,.sd-gx-xl-3{--sd-gutter-x: 1rem}.sd-g-xl-4,.sd-gy-xl-4{--sd-gutter-y: 1.5rem}.sd-g-xl-4,.sd-gx-xl-4{--sd-gutter-x: 1.5rem}.sd-g-xl-5,.sd-gy-xl-5{--sd-gutter-y: 3rem}.sd-g-xl-5,.sd-gx-xl-5{--sd-gutter-x: 3rem}}.sd-flex-row-reverse{flex-direction:row-reverse !important}details.sd-dropdown{position:relative;font-size:var(--sd-fontsize-dropdown)}details.sd-dropdown:hover{cursor:pointer}details.sd-dropdown .sd-summary-content{cursor:default}details.sd-dropdown summary.sd-summary-title{padding:.5em .6em .5em 1em;font-size:var(--sd-fontsize-dropdown-title);font-weight:var(--sd-fontweight-dropdown-title);user-select:none;-moz-user-select:none;-ms-user-select:none;-webkit-user-select:none;list-style:none;display:inline-flex;justify-content:space-between}details.sd-dropdown summary.sd-summary-title::-webkit-details-marker{display:none}details.sd-dropdown summary.sd-summary-title:focus{outline:none}details.sd-dropdown summary.sd-summary-title .sd-summary-icon{margin-right:.6em;display:inline-flex;align-items:center}details.sd-dropdown summary.sd-summary-title .sd-summary-icon svg{opacity:.8}details.sd-dropdown summary.sd-summary-title .sd-summary-text{flex-grow:1;line-height:1.5;padding-right:.5rem}details.sd-dropdown summary.sd-summary-title .sd-summary-state-marker{pointer-events:none;display:inline-flex;align-items:center}details.sd-dropdown summary.sd-summary-title .sd-summary-state-marker svg{opacity:.6}details.sd-dropdown summary.sd-summary-title:hover .sd-summary-state-marker svg{opacity:1;transform:scale(1.1)}details.sd-dropdown[open] summary .sd-octicon.no-title{visibility:hidden}details.sd-dropdown .sd-summary-chevron-right{transition:.25s}details.sd-dropdown[open]>.sd-summary-title .sd-summary-chevron-right{transform:rotate(90deg)}details.sd-dropdown[open]>.sd-summary-title .sd-summary-chevron-down{transform:rotate(180deg)}details.sd-dropdown:not([open]).sd-card{border:none}details.sd-dropdown:not([open])>.sd-card-header{border:1px solid var(--sd-color-card-border);border-radius:.25rem}details.sd-dropdown.sd-fade-in[open] summary~*{-moz-animation:sd-fade-in .5s ease-in-out;-webkit-animation:sd-fade-in .5s ease-in-out;animation:sd-fade-in .5s ease-in-out}details.sd-dropdown.sd-fade-in-slide-down[open] summary~*{-moz-animation:sd-fade-in .5s ease-in-out,sd-slide-down .5s ease-in-out;-webkit-animation:sd-fade-in .5s ease-in-out,sd-slide-down .5s ease-in-out;animation:sd-fade-in .5s ease-in-out,sd-slide-down .5s ease-in-out}.sd-col>.sd-dropdown{width:100%}.sd-summary-content>.sd-tab-set:first-child{margin-top:0}@keyframes sd-fade-in{0%{opacity:0}100%{opacity:1}}@keyframes sd-slide-down{0%{transform:translate(0, -10px)}100%{transform:translate(0, 0)}}.sd-tab-set{border-radius:.125rem;display:flex;flex-wrap:wrap;margin:1em 0;position:relative}.sd-tab-set>input{opacity:0;position:absolute}.sd-tab-set>input:checked+label{border-color:var(--sd-color-tabs-underline-active);color:var(--sd-color-tabs-label-active)}.sd-tab-set>input:checked+label+.sd-tab-content{display:block}.sd-tab-set>input:not(:checked)+label:hover{color:var(--sd-color-tabs-label-hover);border-color:var(--sd-color-tabs-underline-hover)}.sd-tab-set>input:focus+label{outline-style:auto}.sd-tab-set>input:not(.focus-visible)+label{outline:none;-webkit-tap-highlight-color:transparent}.sd-tab-set>label{border-bottom:.125rem solid transparent;margin-bottom:0;color:var(--sd-color-tabs-label-inactive);border-color:var(--sd-color-tabs-underline-inactive);cursor:pointer;font-size:var(--sd-fontsize-tabs-label);font-weight:700;padding:1em 1.25em .5em;transition:color 250ms;width:auto;z-index:1}html .sd-tab-set>label:hover{color:var(--sd-color-tabs-label-active)}.sd-col>.sd-tab-set{width:100%}.sd-tab-content{box-shadow:0 -0.0625rem var(--sd-color-tabs-overline),0 .0625rem var(--sd-color-tabs-underline);display:none;order:99;padding-bottom:.75rem;padding-top:.75rem;width:100%}.sd-tab-content>:first-child{margin-top:0 !important}.sd-tab-content>:last-child{margin-bottom:0 !important}.sd-tab-content>.sd-tab-set{margin:0}.sd-sphinx-override,.sd-sphinx-override *{-moz-box-sizing:border-box;-webkit-box-sizing:border-box;box-sizing:border-box}.sd-sphinx-override p{margin-top:0}:root{--sd-color-primary: #0071bc;--sd-color-secondary: #6c757d;--sd-color-success: #28a745;--sd-color-info: #17a2b8;--sd-color-warning: #f0b37e;--sd-color-danger: #dc3545;--sd-color-light: #f8f9fa;--sd-color-muted: #6c757d;--sd-color-dark: #212529;--sd-color-black: black;--sd-color-white: white;--sd-color-primary-highlight: #0060a0;--sd-color-secondary-highlight: #5c636a;--sd-color-success-highlight: #228e3b;--sd-color-info-highlight: #148a9c;--sd-color-warning-highlight: #cc986b;--sd-color-danger-highlight: #bb2d3b;--sd-color-light-highlight: #d3d4d5;--sd-color-muted-highlight: #5c636a;--sd-color-dark-highlight: #1c1f23;--sd-color-black-highlight: black;--sd-color-white-highlight: #d9d9d9;--sd-color-primary-bg: rgba(0, 113, 188, 0.2);--sd-color-secondary-bg: rgba(108, 117, 125, 0.2);--sd-color-success-bg: rgba(40, 167, 69, 0.2);--sd-color-info-bg: rgba(23, 162, 184, 0.2);--sd-color-warning-bg: rgba(240, 179, 126, 0.2);--sd-color-danger-bg: rgba(220, 53, 69, 0.2);--sd-color-light-bg: rgba(248, 249, 250, 0.2);--sd-color-muted-bg: rgba(108, 117, 125, 0.2);--sd-color-dark-bg: rgba(33, 37, 41, 0.2);--sd-color-black-bg: rgba(0, 0, 0, 0.2);--sd-color-white-bg: rgba(255, 255, 255, 0.2);--sd-color-primary-text: #fff;--sd-color-secondary-text: #fff;--sd-color-success-text: #fff;--sd-color-info-text: #fff;--sd-color-warning-text: #212529;--sd-color-danger-text: #fff;--sd-color-light-text: #212529;--sd-color-muted-text: #fff;--sd-color-dark-text: #fff;--sd-color-black-text: #fff;--sd-color-white-text: #212529;--sd-color-shadow: rgba(0, 0, 0, 0.15);--sd-color-card-border: rgba(0, 0, 0, 0.125);--sd-color-card-border-hover: hsla(231, 99%, 66%, 1);--sd-color-card-background: transparent;--sd-color-card-text: inherit;--sd-color-card-header: transparent;--sd-color-card-footer: transparent;--sd-color-tabs-label-active: hsla(231, 99%, 66%, 1);--sd-color-tabs-label-hover: hsla(231, 99%, 66%, 1);--sd-color-tabs-label-inactive: hsl(0, 0%, 66%);--sd-color-tabs-underline-active: hsla(231, 99%, 66%, 1);--sd-color-tabs-underline-hover: rgba(178, 206, 245, 0.62);--sd-color-tabs-underline-inactive: transparent;--sd-color-tabs-overline: rgb(222, 222, 222);--sd-color-tabs-underline: rgb(222, 222, 222);--sd-fontsize-tabs-label: 1rem;--sd-fontsize-dropdown: inherit;--sd-fontsize-dropdown-title: 1rem;--sd-fontweight-dropdown-title: 700}
diff --git a/_static/IT_sample1.png b/_static/IT_sample1.png
new file mode 100644
index 0000000000000000000000000000000000000000..705705509999310e1323b0cb6742ff04c01effdc
GIT binary patch
literal 523440
zcmeFZWmp{Fwmyiv27+54xHJ;ngC$6CcMrjWI}HH>0fGb%4haOe1gAT=LvVL@r?G}9
ze&^hK&olGofAeKN%=A;$wY#cz)vmRdyz5;BpVZzc;9^l=As`^&zJ4XEfq;O8gn)n=
zhk*v{`4T&*3hXGclaWz-Eh9sx=IU%?=U|P1@aj`i3c8x6K55XAA1E_4nIJOpPdZ&D
zg2v0Yxk~7i&oBuTBZ6}*6i9U6b4El2(Qg+=e6UBfDvPi%SFfDHt&U*XvuOR;_Ym}~
z{&x7`;Ve6iZ_95h)9<js>oCBL27%bRTrUHW9XXD*FJAV_k8gFAi0le62!THb=@mlv
zhIN$w%1SFj66>v(4;Wo)?XqCPyBrLDFqI1f^=u*{gym;+3~yTJ`BP1hid8*FN3asy
z6*S6VpxN?Fw0rVXD)H#ETi<j1&u*hAmcM0{wNNOnDG*S2Q`lc41`SBTqm4|kzJ~9I
zRez2E5$AHq&oJ=uMa7;HCJz1~8&{wi9#+2LQtexEW$JZvZD4iHGU^W{@DH6?#W@^(
za1*_+v$A<rxXCS@l);7>AyvlKXbT&y`LNERh=rSO@eE8K_(AhK;)v!zzYVi8Pxc<s
zbJ}!53bD@Kmy$7Js(8%rB2i}`@tPdHk#WOG9Og9zsqAT4x?gEHVTD*ir*}R@>Q$N<
z?9rH+tNZ6YIXh&v^PqV7o;}ityOzSOi-ZbszT2h?EaB5C;Uaa!(_e0dz#+Xc;hMGe
zUni&%kTIT-Q*W1;I)5$~`Fb&vxD^o0c1ymmdou^!)z_ikAAhVqctsOunhx74hYb;j
zAX?%YA&x=s3m=5QGugXHHucNbcC%bT<=@K@-jjUfo&uhi2+>PZy=%wG5-#}9ZYJ!O
z44xny=2cZyX^ug1h7qWy(1V_*U#49m<V0lM_8WR@ygvNqyvk)h;vDoO2T>&l8z-2&
z63O#7W<4Tk85>EOECT7o3~@PFEAEpHgp32-Ug{YOT6(L18e(xP&3gnYGzDosA=IG|
zrDckGM5mB&DYTnbC2pL>5Os_X#{{T(FTcjoSmEfu^dulU4f4-5V?l}xf09f53gva~
z#vg+pWNg7~ITSzeZcs%qGsDZJ7XAeMAf*Uhl*%?oVQPP&hb|Co+p3(3ir%Vz>@S3B
zg0j_$dMsa0RUU@ls&edeMd44G+zMJDM4`jNpdL{uC17fkZ;L~DP99Ik9#8)br$~+>
zPUjnP!>69OXX`|0;kg!MWbDzDObA4uSIkMwbu0wT*tpTTRZ=|%RFpwBWM{Zat?w*3
ztIGB81n|;77)Nt<B%8y`Rn0&a&X%`TM=nBC)V7gsU8u{eE_U_S^|<w!_44z=MuO$|
zf*-oq_)n6(z5L0tqJ~>ZSG|_;SB6*pj^S6aD5xK3T4R<KUeQHhT!-?uNw<-|#5WJ(
zR@7pmATs`BE-T@}>O$lKs!H<CoK~e`EQ(VG=7mLr!Vb46V_d57m9MOCNluPg&T<~N
zg7l9{gIuGPh&JmmlQ2~)?P}CCl`MtiFM?lS3B&P62}>F*#6*##jKN=IW?mKL8|8@Q
z`fGh-GhlZmzK$#*W`CMa>_=?MN|+R#q^F3Tc$(Ok7{h9;OQFHUmd;X>)X>MC#Ga(e
zwxms?o?AR!99m>m?4Tz6!&Pphfz4z>c0xC)!XT$n;o`$P+NZe`8Z%ln9-JPa3;JK#
zAM?{jg?_#;;5pzfixMa;DBjE=$*+|y(>VC~H9F>x$~KiRS^yE81&#-=X^at3GR}dR
zv`Duoz(cxla6!1i^6~QdU5=lJzv8Hv4!RF=ejWT8`-|Iv#K7$*iT%N7;}8B9Ap-4=
zXwLE@(fNEdZ`)@{W(jBef5P5+y^G!6-o4t3@bKI=`a?eYeb0BNU?+JuYlM0{f7fRl
z^Fr}rXt%}e2ajGfj=XSwCQoTa8Ep<|qGIG|l$M{HKSH2K@a+q4hgZB-6ik%K6t(1d
zlsn{MWSoM<g1%<TFKj3ZDVzllx$n&^-aWCe9{M)3-ca<Bwfn{J&&Bh2^3H)Vhj{(6
z8BVth;W4oV|H5lr?CG`<`iZ>J)ka2#9V^zK^L8`hM51HX1#fK*2AXuN*ps!(wOh0+
z_!mT`MV1<8ZYDVL^)YK|2X<302D?U>+HEQZWem0Fq~`qRbmjy-Nj$}x6q^h$iTxyR
zwJ$j@E2ufBQSh3m4uv(SWvHfvYlQ<nTs^-C@eARKPzp1P=n5}ckEX5+;12Y2>5)38
zXmb@BNi>KwY}V6w!21XLI2R6b=FVmcYEWxnLx*mH@3$JK>ifmWXg<+2iCbmeWEl+`
z3>TVwH1V$pFhS;U)1lZS*h{MWT-R|M3PZ1D_d~tSz+YWE(p(?gDA}bvFkHFq82%0F
ztXOoOUK#<XJv1OVNU8l!4r-#J#*5YLCdYHfGa{U(k;RL|DaP=na3JKsbixgmL6WKc
zD2Bz4_cTl(Jed2<^kilCO2JB08#ps0ok*h4b*KhoE@Fd_I)1z-EFPU<Nx?_KG18aJ
zk%aTJQ<QpS8mT7f*|SOJ+UGhXg6vLm^S`{yyv_ZUkR+ZlnVYut7%er-ZC(z$W#|eW
z4f&5<D{Y$U=lr6xY~2Y{#BRWjWwao&B*JG3T>0o?-@x!X<6H4JbDH$JOxTyBPS6HE
z);ucXhhQt%`j_=B)<af9O*Bn$&3=vWPzUp{V-Eu<1CQD8^U8DhjwN{&`RBCIfsTO$
zdTsT_AL2@a_OEs7rZtRMk~v+Ce>#p-K7UmWqU_7cUicy-Lz(}kDD52G&beH{$@;Ln
zlP4;cQ(d3lV6<V}V&~>NvQMCoB&eNZjnm$+9O^%Ky7;v8DL;E|(rflP9rAKEy$R?n
z_#&A98~=`dV1HdNs25x(;OS@|{HAsH1BCVt`~#%Lkx0B~yj5F%qc9NLM%y$wSFu!e
zZ%7E;epjX9qJNTt+l;FVu7(%g6;^%jrEC%qcAWi1UZ?-_dWdA06J1eD@l^4q_M?L%
zn{^^d5@fFS{ag}nc=hhVw}a7x>eJ=ZWNLX)fuk#pcmtQF(WdGu#in-w6XBV6nc_{&
zp8Q&STBN$*Mb7zviY{nD(d4&U2Ye^h!z^1r%S+F@fx8B$>`BOKfGI5MvL8>D>TQKv
zd5P$cfcKi>MxZiUu@^^wKDSGR8&Z@~yrbCp+^O)Ja3`J=`uyH?>+PSBTFU`T-0sh@
zx|umyEbl&iO%}ivEEMQBST$@^May!q_plc~DY>Y)Ro%%h7Ay1g9U~u2`@m;udo~o4
zL6l7?lI;qqa5SH5$95rg`Za8K)JW!ja>@u*A{Ei2wgokrN_cc$^!1hulnhQ~l=FLt
zcxGJ<p41T;zcE&s%B-lHf8Q`NE8cXg9BVs#oh~5=T3DTL^b|e!gk^4v!VZU;*qgec
zu)Dx6_!Pb?ZM1|D9OL%KcGc-xHS?_MJFwtGi~8ns(S^OSx$76FJM9uCOZ}474_Bk6
z$I+*y4pPoJhYbs7njwK9f+hn2BM%yPuI@jMRg)x3Oww9J?o*~rGuvI-;T4P(-HQ(1
zCD)!0qTybS3@Z$(i?$G2xZ1h;*;=sbfGTyir6hhd*ZqeFjr)l^`@Omx=h!aZ?ye3a
zOs~ABQfgrd35eb`v<Qk~2;-D*AGp1$v$F`^p3<6q_zhm|&GLJhf$&yp(A%&>0{u6k
zmWob(gF3bIr@Uqa2WQY!(V|zLDlIA1O?UO2ZN{PVXNF*Kj}`ud@xsGTGqDont3o&_
zTo)~<aS}-vK(CS3darF%R1nyJHU<I;A_W2}&_V=O2}H{OZp$OGARzy<AHcogb_gi{
zKIaXvef-1$>tmmPZIQoxL_i0AJpoql9Hjp`8w)81`M=tzk7puCYs$QS4Qw?nU9GL1
z+}=97-%RGX0y{9@ztVR@Kp<g!ToGSuJU<4mKW(R_=dP!sEMn>G$YpNjY+=pi?fCxj
zItXIkB0$s8+TEPa+tI<vO~hNA{+}~MfcE2V5Ix;Lr?}gT)9b0I(aAWwTGI(|adUCg
zOJLE_(TTZQ*@$S!%Ky7N@JpQjt-Jes5fI4B%Ztm4kIUKB7Q`bgEDYl21@ZE70%vf#
z`8c_odviLuG5qTy|8*T%Yd1?*yZ7#P&Q5fX*EP3r_HY-cr+>WBe}DcpPHS(w|GATs
z+rJ+RctFtO5fBd-H|W2w4RjTI+$*AH=WXqvFKg!r$PBoL1kVfJ7h?Z(_<tPwpIiP%
zSH1t~%F8dn`=4F^<Iw-TtG1i9tBkWFa8q}Q|Cz9VcmB_V|L!OTdVKc(kj1|Q{m))N
z(h^u=p#PpV39N3i2!CKAsqAD`wSX<4W{)4F9^i}RUt6G!f)iLNcM*bsAcgQ+R$9v&
zaX%Bi`MF#}oqg}CSyWF<Y$ey;N-F}b(JvtwEErEv5T!x%(gd<!Sg=`)WMvr@AzrQ6
zKj<(@E0L7qGM?L_GQNs|x2RrEsEv*B*%mYnr3w4p|AN*mjO8}&7M<%%7&_zj4+skr
z5+MH75-k1NER`4$-MW$W4?P{y-z>|RuL#x1rBP(1{$@erL1ab3TkX*NBllPRWu+p6
z`jG!-+`!XvM*w4gPr#h@f$OgxoeoI@MT_RIMXwBuJw)h)*!{1j4v{d9KIb$DBjT?X
z91CFVR2`oRbE5vLe?$;5ViNk_&AF@@0Z|<NGTj?7RO7$V8Q`wpFn);rox0K?g`@&w
zM|W9e-Tq6pp_7x^_&ak$BqZwq#x4{=q$*ANmzQ8gV#EKtIgj{F$3!RhTROuVtMM<@
z1~7G>zqC`7|F=yY=Lh|=+pzeS#LafTbhOgnPC)Pwrn2b4Pp4p1lI^=rfkiSkx!?R#
zMq@4Kqe_k_O0j=CH{&v9L&96`bqk(`P}#RqPZ77W!GW;VDA1*n?0V5(&eu6xa6E&;
zW%@I5R;0^!#q36Zr}CP(!A0_f$GPD&uC)Hgty&4(kr976qMtHmjXMQyN}v{Z(O3Y@
z2N|FYRF|$H?4tgq7fhwVz@Zc<HO=G6i41~2LH1$$OJ6SY=oiA@ud?s{osdGZAN>lL
zf(`BO`v3p;|NpxDrF5SnF8S_p!Zy>LJww6cKh&M3E6u4}^27bYf3No@ET{-BtV`Tq
zS=Q|oq&Gg*HLCJL5_}paYi-Y_I1m^edHPziEoth1r+wT(D2>G!-n&JGjhb(Szvpk?
zuQN-!7&=tRDI9IDEYiI{Y}jgS$CsSX^-r`;qm=Nk7u_qZnHSkB9iv3dh<uq_&Ht3q
zMp9nx`Wx*^pZ)*Nh{>dojI~jkk07`67gMH3PE7{_f-bAFu>0F1Qu%-w;m!A^@Lk;n
z-hPfXf!VhVb=!GSBob5Dq!o?3Y!&Mte>LT98+3vjHj>?7OJFc{a#`Kttg-LTcaN#_
zd=bQ?vR$0;5BQpv666u?_S(a^9k2`B|J?w4h_;LAm(gkcmlp$I#?H_1l)%hc-WC%6
z_pqJ9>_auObf3+%1!ca$^@4P##<s1-0GGZ*R;M|~I**bHNTaU9zNa_Z&28E0!;t7v
z{fuq)LcrBZxJRpj-S?#VOs_?KFBQpRBDI;~q4z9T+`l#wf?w6>T+B=Pn=bwDU}Z4<
zqa>q|-4{Hj8cx=_yHnn64U5AvD8+r<54=EG&G$EZiw4gW>Gvw2klV5B`|+sO{W|W8
zrw_4`_tS>1{x>^?du_)XAn=T_YcHe6RKX>!9>o9lpw50I!?AW|{^!^m!`sXoS!;Iw
zk)knZYnCb+z4lG<78*UywR7{<qU({3^Ezm+4kJQ}5=DaFAvAq)aJnnOSN8uD;rV7u
z4YMFE`W>}6&DXnxbpS1K!FC)z;>EqND3*qUs&1OY=CdI;y=jVfkwFJt4{g{Wci`;0
zTNw1=Ai*(Sj^tzxsiJ;0vL!AWndIZcb@qcxIBlT!fu954Z1$XSiNl22-Z#|X)(lkx
zS@xMS&%FjKT8FWGuj(n4r^gcl*^`!l>F@;4eEj@^|F{0ClF0ulL<^Qy%CRLzJc2$1
z*3G;2Km8!vP_A2t%=EjP7ThelnQA$o?hM&!mmDWXocsJ>=-jeF6L@8=U{IrSeA*8>
zA~ATIoAqX{4M+cEwE_&cuiMJ<oAmeD&JCacqQs^DpHj~Nq%JpG21uRbMDRWz<4V#0
z^%gK{$0w<d(=4n=wY`e5pZ?DP{ka5?dFd$UYpRM*P`h*x9wzBZsCSSlMh0be(Uuqh
zHd&r-^VM!1P;RaG`(=M)K<M8M3w=t&O)s?gfb79RzN^7xC<<s~IW>+62HR!_E{b1V
z@j0!?k-B4ZHA^jll&w{P1dB~cO-|ua4cMNfzOeh!Up-LPblCVbV04cagGF3JEs^%o
zWpV$+5QCqO9iDX4!bB@LnvdIL*iR$N{(%`%x|ly6Mz!y5>i-LMNX7il(M0iI4n}`8
zqqL|la9|7tf1hLXN!MZa!(Cz1)oRqjLEX}!iSG`hpzHdpx{fDG^~Rnvot%|YXP=zV
zu(%r5pM;0irkDoK3<}ISIPFhYISukpxt|S*R*0T-5G_D%5BI?N1D7GRcfT_OE@$rj
zT!NBKeC@oBThXVu99u3XV-7xpA8wbX$~&2}Z^jGfYv-Ky0CdUV)%jGnp1q=Zv;`Q&
zefpyhgX^>>Fspj1ik-;b85~e`wvXg1H0hcKOh65r;zeg%`#BemSK8flC9ivL!FOx4
z3xk47`_R}hnbpq|5WxNlhAb?ZPqJ;ZeC8$ZPhwRih%-5Br!AbyeoaP}k7qyGV%03T
zdhyIy;_g_cZh*J{sxpoXo2%Xoj(u4odG@NkThVCSTJHC~#Z<Pn{iMEanG))%l)m3)
znyvq}E&Fc6=GXps{O)M-HON&b=tj&K_65FOQQy6^^DVmbD9o*|?R$#JE{)$lPlcM&
z%(oZ{$-w5iO7bjrHwRrpZhnA52c*@69ir@8Rx>05!822G+(qW-c(V09*`Uv{x|a#p
zL#nT_I>9%_S0bAfn<42cj~I~%nc8A8(`VP`FmVmM0`XbSavBcsqPD2=D%T}Oe7Ko{
zK8B=D_Ut%}a2nqqDTd)jA6VKRb>)vW(z(h`g5<@xrZL_p3H;z-WqcB^kM6^<TR6$%
z^(S4F3u(4l`q+mq-PBt^Fon!IXGTbYh|C7gb3*W3E1>s{_)J(cfEL<=(DlugWP_>}
zF^QPQZ(g>}J<gU>7r8Bt<eWrx9MMCMYV=C~cZT;D6S_@NT#Y|ASB3*G`Jij62Or+Z
z=3oZXK3sJNru?|N1R}sLpaD%+Uxwt&{t%J%-1p~%Upz2y^PjgDB>z+slg;4>&ZTGA
zyyd(-f<h?;os9Gyd{t4<LLXIG3^wn!@GP8alv@g+Ok3A@f;#YO<<M~HCfV<_mw92%
z8G7)vds-*86*&<>Rb=ORg<eWw6w*8xJ_T+7!}m;0T5Xc`ZCqHkTP|nqW6pr+w8;Jx
zFby9Q)<fvSEwqL4H}|Wjw~rpRo{&L)oGUjjLW5c{631zSdPSycD+n3yJrsW6OdYAj
z(sZnHmnP(r*R)^VPvrt~4X5@@#TN}c{#+^Naz8WnMszT@^oy6CxRlG6;3veF(`}@v
zzwHV$C&h2~Yb2;x2=b&n+Ax`~_A)!bHG=~7#VFU0aYZ$7t6i3gOigO!^t0gNvE2uH
zQlm;p1i$pO|B>I48;cSi0OL31z#H+i=xkZ%)%0fI%DtoNr-{<Nq=_qrP7R&-G@5wo
zi8AC@N8pF1;cg=akaPXYM+uN-SIAy$Jvy{MvFeyhA&~+$V@vC`6ac%fSI-_4ht1!M
zY;}C{jlI%z>aX{Fim2&!1B_=UM(nhwO(fvggpM1aAq}FJ)0P3ocL4Ob-HPA?pAAbw
zE9wB$k#CTw-o_LyS3dS{{ax?ObB0a*(uoX;6+8aXw7<%ZJ=3+4X1AF(FL8TxD0EqT
zJ6OTxO6e%PILLR@V}iRH6voaNl>olO_S!dRu9t~fT%taPJLBq6`EJM*5hLo$zJ$+z
zngevT<t5*d*xhlvNtQWbpRw1%Mmb=Z$x+Wmu4A&w<d@?WXin^^;g@r|XRn+A-UbH?
zsV9X6srpK}{1yd`R~!ILWJ|H%#7zUQH_~HFe=4D(l8{*Nmh-B`pw^t1Y(I?F$rR}M
zpLC+{p206OohtY1dV1ChP{{T~kAEXC<i|?vk|OaADS(ZfcA0-)i7Wa=udHQ(?|mfY
z-z>4v)Yzw>U@35v=9xU)!=OutK&8Z}0w%5|Bzb?e)^dVyQ~8Zro_t+42ZRK;G1YmG
zDPvtf90ua0yG=p3xAQ^X8{aGhHv>Im$Ia9dJpg6E^X_vSiS*L$l`5v-gQKG076&m6
zSo`V9=+P<4I6nKZ(~lNmEvt?{&bisH-PB&*`QqE_J)ZzWCUcJrV%ano4wHUXy4i`x
zh*Ns?%$++Z+si@WQE3|iFjnN$Ir|7IJ?9(+E3`t|@cQVL|K#O#izi~UP75*lshh~~
z;Fnl}@6WQ~$;c;l1nyqb=ICHx8f`B#trFcZNc#f8H)GkW-7TwlSYoyrF7mlgXbn1X
zBFzZ*=9ca~7n<%cLMT~!)d+L-LfR}BbE#^8*fC|y1X;Qs^73yc<F?sbG<KZe3}M9S
zDEm38cnFx|$<cts?S5b$h1|fK)nT(mT*3KeK=EQfx<3>1`A!W%2Ty<GOi};v0ePhp
z+7E)xvp?`GS5D{{`h1rrW_E$cD>Qe1y59p_`S?Cso&yhe#45{%>leo@x%*a&`3LyY
z!?m=`DGxGUV$^cw{Kq7vKB^6>6p_OfL+gG$?`%AQ3rr$WmM4}Aai`%HXZp68+LSMH
zYd+_3ZPZ2E|KrO0t{m$YHunnDhCd!_g1_jnFxOOnc#X&bK~u<#Nx~M&ng1lS!>4UL
z^IE*y>}{I#E<s9lH((5!OiPD&lE(^%?f4RD1A&)2TgsnY(FETus-oH3W?Krt(wFW>
zUH0W*t8QSgz4GP;(Tjbe`74#FgHAd;y;O}Fp@#vbd3=2F)&sAe#Y4)anSkG`BkZ(_
z-!ZO!RRj#y2}f~MzcL<ITeacAyzqSg^YtUX6@9onqsC@@-{ah4Hh$Lg!aYk8)-W8V
z>%LEqXk~Rn-Sf>A2pF1@h)m^ri7)^%D#PQcQFy3&&dsD*6j=ZwCxJHm6LX2EoSf*e
zP@nKt##*H(pBkS1=6O<i%cX%Nr1PBz8J^7)XL3l|j2W)z8dEen_hxJPvz}fk5Gv8^
z-H8wJF#bqlt&KEyk@2h@M<XYZTekADMRi@vB-1LFPu<c)*NqNUYGRsO-+h+5b*2A9
zis;$%30JqYvJJ}#w%`SGvk7P^xowd_U_fDU(YP$aU*+N6BrfVFFR^`wb(U1-iMW@k
z_Z|23zYOKz2^(c>BD~VGy?E<}DJM{GDf+yF8RcVSG}#U3_D!Yc)=?m=Y4&c2`Gq?Z
zlDPbDG(+FEnZCV&Kg9R?E5+FPDycT6*L}JaS^QyH1Bny7H8e?S)uy!o>J$*O#PUf~
zflLCJ;1^#F%T{)#+mJBd-nb1@^NsBQg|SVTFyDp*c!?O%ernLy#K!POz`wn=bg;fV
z8}m0m^mvQ|OF&Gu8MCg8WLUYXYz2ZVf9pK^P1Ytd1{bDMSspd~2se*$mMU8l8@H~^
z*&X)oUSJ!hsrX-QVN_tAdj!sq^xxAMiF|^3T=LG1)(wH$tRjOJp;yor*kV&xy6wQJ
zDW|fSEtb-+^Hm5dmUY4UQ@{A{Lku2Sx_9R6th++f2mx}mhS0^=Gi`;5`)I`uX&O#8
z?B(x_266mslEvPcImo#fOE52`y2^X*mDNpVy{3jOyUg%*-6po;3QW`GJ24F01*}mn
z^}Nh{Wo(TV{L620WatNTU;UfY*L`9y^&BfD`8Jje#9kT+e@QM7%S?M|WuwTO#(K{W
z42#cWC$qSG|7R9^d&{n<Gx*Nn^0Zr(_+I>M6M2R@R;Bnh-LivfM%U}Xb>SCO!bS5k
z7N3UwJU?v-*#~~9;CyTsc-b-LD-ey`$-p1gZQ8=zbP&H+qwwG}6=!wpbZIj|zr7?V
z=MGvE1Dvk@UAR$QKUdSZF<&QAcJ?5WdwbnrUK!LjG6)yuo-2ytu%ccIWF#J2Yd6Ih
z5x;|oa`5WHn33;|11eLc1V`IBKHCB^FmVZ*ZGI=pe|0%e%sA`MaA2ux<P!D}2#)3Z
zhU-z~ykMhG#b$iqj1(=yd-zQ71?Kc51cJvW_ge$O;pE+A9Td*V*ZD!?$m{HT>nO}D
zF!)476K$7?E5MI{sIqfEjl6}dgNN*^wCxbVb%x8;gsy3eQz>@ZXW^y<RS98AJ{#pO
zsxW;#YS)~;b5k^>g(Y1JozDu9fqJIq<M2z&AGDJa?Bds$kKjoH2vjB{@?Njrp!jAY
zkWBivdE|w@401$`1ih_li_j14ANZhYQxe^r7p;#|s>y;q{8bxbkU_2|g$bIMS&2mq
znV#u;CyCL?Cr!vbukai5=wrSeQurLMLQUM{vDi`WfMtO!Urk0zZZz55>e~`C`}>?9
zq)TgJ6&MjnTKD9ETTc;H?1l5!o=QfYTI~;%+ARbmmZ{BRmM2_Q0MN@L@Why@$Flpm
zP-K8hs>H+f@KhqJPx94Fa)rn6pH>X4=vJtf&-1I4-RZr^pbkFNreF?38^&rbNi|X<
zj|uHx&t-ClN`oiGWoY=6f`8ezy^aKcO>r_0@v;S~LR*F|P1OO!{^aGzb)h8O?#f@G
z+i9ZpSsV^R=SG&>C}9{GF}{n<4u6fK2}M>7XEtA=v}H<l`Tn=8sCI6GtukW7sKl|&
z3_sb+L1Y-2mZE7<(o$qSPO`m)dg-}=09=C=$q8VfXNpV8XVuo@4=xwHmKyh1Qq`bo
z6SIwbcX9M)$Y07$e~84*Twt&uJYsP=Bx=G}JcttquZdWZJQLS_>(PO#S|vJ>HA0A%
zHsO1Okgs=|A>WU9w3vJnm@CCffx49$ZfA~|9TSM}p>5TFQ!*T%HY)(Yu(XLMl-pAH
z$&FADm-gBShv~(I@xl=;(!$}`rE2NAJ(!4$jTcB9Zbg}U9}-c}7AAiS`0P>}u4lRS
zv91(KaseECpdo+lCT<L;{$b-@`SBR-U}W~~(gWNSJmhIwQd%9k<eH3aL;lBk&2`?T
z`;jTRwY`dU<eSMFLwlaVS};*w<AF2R)QOjnbA14A52*Fy3#&h_@qE!bof&wu6KKB(
z@H(^E>7C-XNO8A^UeG-ty~lgtn4`5!7W6c?Uw%Kx8zI<E9T=Xk-R5D4T%&??61{**
zYNR$dnYssXM|n%hLSi*t6!zXk%kSGfo`4PYkLob+0$pCnNkOgO)OTI_I^+_G)~XB@
zUXU0%tPGZa+~O7u1jqp2nza>!VzDQ8N*?<gWx+C3nw6A2)gI>~3QY_#Li8w0v?|Za
zohS+VIOjfe4-vn}(BF>&T_4n)U6&~@d??iJCo`0-#`WAkgxoqoRh;T=NL~a45C+!B
z9X_To$2A9*O@_$m_?qkS%cR}(I*++d%|r;7!|kA3LFIO`X`s)WJ*uOJZ3W3<V<mlE
zPmSRTByK6@vRX7QthOilI-aI89)@1)Crj*&#!Z*I>=g}XfdQYK6G*7|pmR3ep!$-E
zyM%QA1<u3ZTbgQtu?9I($~DN(G13KF38?Quiw!f_Z=1)fw}EJ34eU1yhj)VOXTpa!
zyEuLZ%~_S;mIb6wwA{~9I>A*Yi~Qi#*cI=G#o*0;7p4{UV}ZT}b2j0e3t##<)L}T1
zo0-muVV3OB&xY+PO~;pp&P{G_pnqyj;lG!%JZJ6ZCD8pBB!^obn%2FXPI2pI-Hz(5
zb@loV?#nV4B80<lZL@zJnR<>S@sqFgn=bSXLYMu24L4tRr62B_8kx;4g7+&%NDlmt
zO@7_VF+ZjO>8B~8^Wt8MTQs!*s><%<tllQ8yV=t5Fm$N-Bs$gBIN_Pz)F3~k?<f8g
zTkP(DQMa)$`$r0cgyoV6yryIvGBt(i+O<yU<N|Fve+L^+7@pSgJu>i(@w802`#I-(
zJGhW4N5XMV?*YN0^*LS)U-&$=;9zJnd>}CIFD|&}Q1j7{Gq$J}tzgKSosaUo_?G4(
zj@5qZ7hh^$;o?1j*N-INJE~LS&N}Y^d7qI(`78BhE>6F5K5=p2alodzYZ1SkZg);I
zfms}e>+^Y#n)+9zi}_T8rxJ^(&Gj9BO*BW_CS$odpAW0nK9{}xi7F77u0o07i4AJ5
zT)HP~NG{tHoUuwMpaoIJK5`%Z2i^{oHkqtZx9|T1N?fleIGt~g4Cz!f0ba6%$+SB-
z2o^?avfVFkI5^iv52EZC4m|^ijxJ|EqCisWrl+y~dJb`Ia3E7V2>^nUA;w0Jt%Hn;
z*miiTMhT^`tJS^sLz9CTmiXTC>bKVk8LYjNi)JMUK-M%-XkY(!@=rikp;OzS*x5i}
z32V(a8k!<%pDq%UYtF$e?^R6b(;uANu1cK066XSM5B6#Wu|R3s7G5KAq!Y79eInLu
zLerH2xtw~(DQ~gyB(9d<i(bd`1^~S7wZsg7MD}h!I$$a;3VgSsh1u-9GIOx#Qc2_e
zPQrEi+g9>c{~<k)?|LaQ8D6`&j%n3TKUF417~V+|aa}gU3QKez&4P#rOactgyc<2F
zKCf8@`W52eBPwQ_T)zP1pG#Wh%W0B#EBJ?5eupbRH_1YBOfcujG|<0}dL2kG_xwEf
zTH{=6M?TN&v8^mztm+KL(@*uZ0+?UU{UzvWAA1bsyGrA==LZpd5(j&wMLLt8{5{nt
zWb#)w6P9-YZGCk=lxX~qrI$=|Tlh+fhe+OD^D53exFo0Cs*Dt_oLP@CT>f~MRT&l+
zb6{fjN;3N9!)1tntAK5^>7S3p!hLubsB?xhANZEgX38;eW50%-&60HB3KAL7;ip;$
zD$sn(Vuoti&I5AAlL-AJG|ffHc_u8qqa&<3SgpK}t%_CKZdDpt-FLY%F8GsrTriqP
z16JtSqgWgFvpgUdHxDu_+9l-JR_^KZNCX>drhG=9`>6r33wd9QF)~AM4_?pW{Cc?&
z=2q+f6Ezt^Q-aDOA}x=5L0!$WR{PV`;XC!NMAyr;vS7QdKp?xDg2a0D#QU>*j_(Xh
z+&UaQslEWJ7n-*V_RJ2ihpu?8BLz|=P%qY|t1MfZWqj2Ic?GoAMF!`Jb`SW5gxbTC
zxWk&|p<x0kB1Gg-_t=Kd19y~6PaUPzucFH1ht_FNHQs7QWkprQKU~@DTU;|i7I&K?
zn~iiN_L3k917ZFaq7>H<<c(gE*&9z}pAk=KSLpSiv%lZ;o*}ZAfrJHG(fY>pu;;Mm
zg?RC0)Y(dzhQw*$h^{mfQ0ZAat24OgVgRAz_a~1W2MV6aV#ribS^6}vC|Al4$zFbo
zvES_LNxKdtgX7PpHRnrqdJ|tDa4q30F-|2OK+i1}vTdh_juKBI&a?~Rl|rS~r`mUc
zk#6fcT|_%dZ5TeJ4H1xIpzgBSoxh#W_hxn=sqCnD5?*v`z7VO9Wj`TM>L7Ny!gAZB
zNU6iG+F^e9F*>#Q;mN9cnDuc852#CfwJbQTZqsXjJA9PZakDY0LL>H93b~~eDQgc#
zbbmhhsp#h1G7*)p_PNaIn^XO)SFO%O9|6?%?lTrPJ4t3E^s$WfjH`fK2ZnRxzk+T0
zL)?~n_T#PIL|Y{G+`Wn0Apar-wV})G`b(RT7r(rO&A-b?P?d2Y-CqxboxHPPGrU4V
zL^!-e^nA5fgU;3j&d3yy^g4{5-Sqb$iM-F$RLQ>_jw@G$OAy)^?jy82+wnv>ESoG>
z6-7D$%r=K2-wv~g3gWK)wDQrm5+T`3pm+3`DjC@_g)O!t@X`ZsjUiU7dV_Xuqig2_
zY9nN^sJcw8&^k(}z)UdCx*tt>uM}Lp8*N|3Uj3>8(65fzt21Jk?)+lwXqct__h-YY
zaq;@$1($UIKRUI+Tsy?MCNjueBotOj<J$Ag<Ae?(@RrJHDu(F?zaH*}zNW<w@%Gon
zjZdH=MO)s1bjDzPl;(?-z3QjB)G32}OFa&Q9+N!~pWiey&1bQFItBq@1BolVntlBw
zM&9=h=O50jyE9#_l(0i2MfXA-a^*;Rl&?XC{zaYDgLSo>Qv&zREs`SX&gWt?NS2$!
zlXF&?Xp#c+7I26CFHzRmpxT^a${PIag@;XpySm-4?6$EjqyF}DqS^I4hEAQ|;iA~l
zX&^pEXtZNOMa#iM^>BHUya4P+L$&sl5i6fuy+aIPjcvB^5#LNg#nM*0+XZKeO@Fi@
zo1s+|&&;v#R!{j&e@El{H;|)u;OygQ0jP>k#bMy0#zI3%rqc;|!e<_uY+9&J?ZS}h
z0HpGOY350P^q~BpD6ip~TZxUHWAk#P>C|;XB<jk{1*PcZ2_?_e@qP{i#2D7%U%$3A
z3_W{a3+I}%Is<W6sJ43u+!)nvqzGJ?3(ARQcRd7jxy&fhiZ3qP_W3_;b5D1qZs{o`
z?TbKPT5_oa@%sykrAro6V?@kfZl-^p!y(m2k(IZ^{$APXXJm>J6$HK80P*cT_ucC*
z2uhe)R<lXqN@I<&>d^oQW#2o4g$7&)@4e7``Ol{cwVbtxdJhoXT3Z40%ozl!1yT6P
z+J4ROQ03-JzX8Fb{3>3@$RLyZjxRMTivn=>ByYO9S+mNzN^$eH&;h0txZoqZU{fHK
z{H)M2fO2M%WZ^|9B6iJW8qAU4fE$i9o1Z6{{=>mr67pxzyJO||*Gkrqs~1&aZ)Si>
zW5$eGLoBnQUd=Cj9rCXmpV>|#+SNDj9z&CjqL;)V#$F$LMaDDBu9#}eC=^~Ry3u6E
zh{x*M6(Kk#$?FusI-$8I(oKi}e8*Ig1(Y~Fl6!nc65aXI*DrBS9sowY{4AX_u3nYq
znX~5P%@}+cUGi8NV9sc0Jt;X(tfOBr`4?bp-IfiunML;ex>Ihkh_d6|t;&xP>2t$$
zaG>j|;r$+oR?(Qo{L<`|>gt=ugw69RW%afc9}b!JUJ{r$(>Gk7t?R99{EAx!l53Su
z-?~E1LtaCMO@YEh(9^(Ejsf9r-hJRL5-q=R5|~fadcnX-cJ7g5R1~i!cZqmA=D`Qf
zw<jJFKADYQc;7`vr?F05zmOuaE@^LWzY$sr3&!QP(R<B&=rAEShe$h01ZGt#q84Qh
zJ<=>+{tXkHHbWlnBK{LV@jmuN9&WzL_j}HFg8c;w5xS9L(k)WZ5t6Uh_1s-nif!Zv
ztah(xc&sardm9-nRk?FGMkRA`*S+pzuWF*?zzt@Ee@I%dc{oQyYvWwF`!GGIFVG#&
z5(K&?20Usq3a#{O4`UnDoMy~}ZZgGUm8HqdUtcrC{<I*!o#<^~6|hcZ*d@Qe%3{5H
z>T8buHo#;}c=kmX-&wh$SOo1x@SU>ovZ%nPteNLf&|FW}x}()3K<;00)>dI9di@k0
ze?5-f^L$;V>$J_;^G539fHQ9N-GEy#>Pk+=W+rR$xJ%EG?-wv69a=Q?nz@g~QIO-u
zLs9s015hlx@<c9>D_358qwZXzgzZ6ZCnZ#M)iFf;Q}7!#oOGSNR-f7Fr*-t+sflQL
zw49Wt7;FXeg7biHtu(W~_|#SZx>WauCip2y(2uQcO0`XyxnLZbnZk;`X~6>{kTb4P
zfj<fwm1a+W6*x1aaMw@kfVA7Z90-R{p`&qnv!_17{!I?#xJO=ms-fKHees@=B9Npt
z^4$lU^3sggk~nFIK0zYcf#`0iKHbd(r_OQu+fg0xN-bBF8Gufnfcpd<+`l1@WrJ4|
zELLB$HD|B(_)An1z&Y2S-|Y0L^!T1Ll2v@ed%@V+Hmv=xn0I5o5Tk%o_ZaDDT!e}<
zbnV)UH$M>^I`u}d-hiIrQI%BU+^bE|VRsuM^A*tC6rXq7<wMRgk3w~KV-|*VS-WTZ
zV|a3=5E%42MGO%W$?1zPj+PC6HZcq<N?><<rPY5rLR1b-R3ppDm+At$6j;zFw?<wC
z3fZQcp(vp#rJ1q0X0k=R&_uC9Dxy~KetFea6-`eUVqa^~&<Zka&?m)Vdp=e_<iP*+
zXn70WN#qmQG&0Dap@UU|M8%-8#qADfOA9Sj5U6n^-SjcELOo-YU-E9x><;VvoD9qC
zjnf%+tB7R1c+*dEqTK34*H1z%A!1@f$R)ZGN^}>>X{h%|<frNP_6&%~+vu3@)+g8&
zvmjk2IRpFy6WidajZEr}EFT@!eX3bxKN1%9$#W~w>ttIXp0S?3qN)W-D?Wv<KBWx>
z>}yOUT>%6!X}CGv(5(_7fpvu_O?qM(w0CwUns0DBX$(*FGG~9N+j)S-+g$PT_nBrj
z<^7Fl1&ZKQ4<UJ<o)05?V7^*7?@C(~SE{k^fzRUbZFUoX(^gt~4Qs5?Q4&UBn&(%C
z&WSGmX7E%(Ld_gaLn~B)=HUcPy8?FOyl2P&KP-=YgvofDc*ElNFBO=tz0|zhD>R2P
z;yA6$Uy`4kpQ#Sr-*l+V7^^NOs$9*e&)80yaQfX%_%(xvn=f*}umPnv7e~g|TTy$E
zv-vsviMz#H(I(rmH-e4}a00X#=+&TV3-1P6dRe-$$63ezVIh+(KIFY2TI}4tr_5!3
z0yu6^pc}kfWNPZJm?>GLM-X$HU3VmhqaNts)wB)vztXGiuJi02h%~P~qnWc6Oru{H
zSAFQb#e{yEx??i*`TVrmbz_G#!-y0elF4S;X*Aq&PU_*oXl|cGyXvLMiX_0kH!d@t
zbkLZC(PKu)OLXwHEV@rGv2~STUA$Jgu-#zY8%>Th(Mzqv@-($JLh_L3IJo)=k!x*n
zaZl9qsnm08jf~4k&O^H>yD_!Jyt5c11;fq`&-fAw{KQ*_q<T<|G7>NA=et3AN~RLC
zJ}X9!sL}A_^a(avof{k+GNQvgNy8HzVpMS1+zAGvWfh?vWX2}_`X1+0z$fy>vn(6(
zhFXGdj@KWj-gOtB{P58_4VBhfm0EBl;o(EeIz@)F(7(@ojc|i(Q&c7$AB9T7@|w;o
zpIB-ss+pj#``467dYU$~q?$c@&fd!?4YK>WNGDpnnqw0$T1Jt>1ciYwMmEXqyN8AM
z=9724iE8Se&jTW4Ua5}8g=ts5+hjq1CfSa?@rr4G$1NvF3QQny72BMC6y+hRN&p&7
z=+yegJfI$tAAP2fDHHG-QpiPr^d8D`6#5faOigXY=TwV5ymU}hS%=%kY;y9>%aYNV
z<WQ|^IXs){dhlX8mrD9;D@_>1bb|bN+WJ@so=6YO@<;4xfD2Q@Cs5=>Xb2|ijm-E%
zqT4B2g4sA+S>!j}cWgLk&FaRS1=@*2XE}b8KK7|~Fn8|!tNBlqk>rW-d7eKUd<hD*
zd_KJR4t=vIO^ynAzaHyefG=-~0`XCpp)>t+8NbPr_p$jY?FnCYvJbTQ4i6}VIdWfH
zO?vu^y(E9eIq<G%*o(y?KK8VHZbR&8Kbr^G>}6(8D79jXjmJO=XxGU5OF51hgLBTe
zN!<m)I|mq9ADIWg2g(w=23D8kDN7XEW2x(b;~-$axGSVGv^K_7nMM+L=c~E7JGHoM
zdpPTbwe)CAbxken_-1>bM;AzKw%~K_n;^!cG}MU!jtJiv8^<bX!nhlL7J1j~Q@x)#
zA@1f+K2{O$ss_UH@~Nc-WLQtv6I3s)iAlye-}A5!nW%qpG2ESOW>2q^$tu=x<~vj*
z>y72e%H=G;NkyGA5S0&nj-MB&43MG%QD3g#65n&>WAwmQeM!<jo;y!zC#77!`B%K@
z2)<KT`h+Q5BX6L)l$N9q@*9z;e*0M@2Y9otKN+74MOWJd?><o&&F9CvrTlkLp-^$T
zYmC4!%fiSY?e&f0!D{Ol%}aw~gl&J=71lR^vQ`gz%>b{`n1tO^;e^Ws3CaU8OPd~o
z6Mr)^;+t*Y^{8xca-hyUzS6U)<1#f98Om`daCOS={qNEN9&NEJ&!j-IBi&xz=IlOo
z)0v^?X$Pk%BcAm5&kE_f>7mtcV+9gj`ssClK}rl@g}dP%qBOydK+g3}?4?jKP03Gl
z?K2efqk`zvyoV=e6>Yp%n%^+qk{kuhDCh{36hNF`&VXEjz*E#UTN2Il2loBI$|W#f
znGkkFCIiq1V7<^$n8Hg+Bb(@^KVwDvt8uz0&&8SFWNb4(w+o88Cwpb7@$2oMJWbSi
zo5&yxq=R~`K2Gy$VcMl;x!;hs@z-~^8Yt^?*T=_mwHa?Q&_c?tZOQ4P*eAy$6>)=E
z>eUh@uHPqLf9B7wF4sf_K@uVtq>?$&e97L**B6xL{f>HqltAmV(I1BfFT&9K70q)@
z*+x!V6qUN|blEMtJYJ3w!5Er~x%K{&vARA-{)^Aqnd5FQWN53+P-4W8YU1R@WxoMc
zh^b}_HcSocHoIi#1g0v!6^iy9l$GwkL<@(vz~(spPFkS+Vj+3UXz3Lxj?EiaAN^KD
z>m+?cU~BuI3i$(=#O9`4+KLo>pz=d}y0v|_Is@fRYmxi<nwQu|gfztG5r*|$C+{@p
zLu5<{xdS)<(9%q65faGt#-9yj1~|cotzhDQO1XIyOA>e8OBYMM>_{JvzF868bs(8H
zMFxF4E3Fy)0-y4N_0imPAK>adnD`$NYLdX+YR^w?SUXB-7VWXJS42;$-a#VaFRS9Y
z!$izgG}k$*<;kE#Lw?$X1gOGH26j^x;TLqoNqTYy>y<2R&#Bm}A~`sD2;LdW-P989
z;T18L$fT+nJrmcvNY<TRB}c=ma+eb|Uf1YgaPLw7Ace@HGqjSfr){P2q=d1I>OwMM
z)Y=Ht5#k?e!Ky=$h5^zLdnO(udMK1_BNrLOx~l%s^bXG_$+gwWhmYOB{U$BKrjp;(
z^BB9IkpoNm8daE8QeVpCyJS}$&8ghtE?pqX3|9q4Ps?%)KobV)qF50`zmCrH<|Pc)
zFlFDvepX7QjwgfpS(4Z)Z5iwSi|Mvu6$^)FOB}P>{B!ZNX;zIM#%%6NlG%?0g+erP
zTbXRMl{YD`S%@KDg?B~QNAsGanun#Hdf@J#Stt4{-l;W8kKi*N(JOo<(jDcl@Rpgc
zdeZ~<4iBOS_nEY}Au5vU_2ZPX){RTD40U2fjLU^~*52)(xn+J+C>Co)A@1AN6Uz&X
zrrrtR7Z|0oZnBBFdeY6N15dwR7`gzW<rM~2+O=3PEzX^WDbH&;NN<w~%T{O+TYs^K
z<S8=tkQ#?gT)K$B(GP}R5C`}6IFkNig$;D)HeZial<8L`6GaNBE3hY9`FA&$hnSi?
zbF5~UI>yT)kd4fYxe?PidT1CXDY4`aBJ%UtBNy5*E#hnSpNx1uQ9)uM`8Z$d)ZaU9
zR^lmSYK<og`q}1G)CdK;6a#g<(%kZA9N)-s;sYXy5lc1?v@QfTLYjj#7<gqM)YvCE
z4{hJbTz30CG+=^fB}9x2eQMf1Y45+q4L2^p%=Xj#Ke@gv<(9dNd+o6IuEOl~oOB~Y
z(|KNrfyAi`8v|$y&T=Ly_mA7tzV7LUCh&e$G<l;gFC*5N#M}mJ3XnDFGT?;2yH_KS
zeq%us5s2x=!x1NC*W9C>#MhqDiyJS;KTEexZ&f`_NOJ?bzGMw7h&CdE(q^aFwKp2W
zx>qV(O&DqX>)uJ0zJnA$fn6D&Qo(!k1fE={$XT8rs9Oy-X<ZmkG-kh7e@HnPyHpe_
zt~D=OjaS{`g_hI;&Q2E86H#;n($@bH7;IaK@oGU;G8sacoEb%8_|g3+$zR)}RD6OA
z*#jI;Z?@bSUzH5_trIsHV~HsJjx7z87Zf;pJl>{OJkecTch5TVdPqf4GTuv4^*i7W
z>?U~{v27%qu`-4VQ;irLc6dB_fDj{+H{-8F#L5wo$W69;GjFCUrj-%VloBh%*<UDp
zf7*>NH}(jcXXxIFn(r0RF9ZIKf<{M!&s<eMj6I`n&PfzxR`W}4AoJ{A3#HVctGscy
z_~Y>t8|7u*oQ3_08CkHY&-XF^FOB|BlXmhDomX2g{0%Hn**%MP74QD5Du<kl(c=kG
zZT57|eMw4kvOOR8ps>W*{Lo<!vpTB^+kcr~Q*#g7@=)}WCf~!;q-wfuwZ~rCN=xi>
zmL??Fkhz23qF-QvpDABkYRml&iM@IAaie|ju)f(&?3!tkWM-EP`{OnDu_$BIGMG6s
zDBbawdF^_2lx=LYzpeSzhwK#i&Yi>bi}q?F@JV~ymKsXwmx?wop}e1%Q?yiG?*&%I
z^IJSRuL4WXOtev~j$c%KBi<$AFdOOGppFFhROxl)$;=J#T=o8DVJ|XZ<$2);&D>*|
zQAf|uQ88Nmv~nzgrCS;eO#KN4^=Y1?NxLAv__-R0J#<N~yBayI<q3Z6Kz@wP2S^e&
zyD0hdDC1)9j=*Gbe4+=vKrYur+)^C=bcf&7)N>*vte-pYb6-V$kRCy*Bu;QSsnf_{
zc%Z+{)gnyvo1r23zU)aA@|;D@qPJ+GcCEF-q?d@rU4xVaw>-Qu?=JKS1+S&a%5$*<
zNly_M8F<%Zn}6YtL%C~;e(lHigzq))1q}@c0d9EFS9%5c;elQ;)IUFsQJaZUk7bMl
zR~@C0OG>y}k_x%{{8vwGm5fbyr2c%t2wpssk5a;k-%4}B`Ku&mvd|7i852bM&`%<v
z$Js!!X+@*Gs_)zD5L-qr6gjC{53Gxq@q+8@Bxvim>Ma^Wm~Pt1+IxYss(pCLRR$__
z38s~mAMc2^VDyL9<>t5)!XRaqoaUivBE|WWAR(-vbyW`JF2L)wg$gSMd{bu^ct-Ve
zjD_r3KEK=eMDz|pxQ(S*fQp`(3lZ1Pu!BkXVB47KlwGzOlI_dSHw{Voi#DatkBa~C
zEeXcw?2+p?LIcXa`O7}{*IV5w*u$blx7gy9QJE~{0IrH^Qv-Nxn<ZpMmA-8gzXR<O
z`^eRgqr(<aoSXf2?82qn3&D*#zx>zRXNsCLz|j38;(*v<`070i?eX@v=z=dPYwj$D
z@*%VfeMBp_3lj98vF=s_UU1KIJ#_gMN;!-hiRf$MURl<4Wh+F;a9;9q+@RfqelsEL
z(=<h@Kt&Qe!_<qj3T4X4!A7XQSf3G4TskfTl6vB6w$S|D_y$ocnpsVf(#?Lv(>vfV
zB(SZ2|Knk3(X?;OTb{%gJL(mCYYl2%!}<X*bt+r&kALAA=H>8_JhnPvhQC>-z@bsw
zJgG_A`|hu8{pem~-pT!b(^YTyeQey_nJm6aI=Ni5hGtbcxR9Gl=Y0k)$M`}GJU5f|
zXu`NgTG`fcWZbx>3^T(qOvh`mcEY9hKz4zjvi|I7_EK*M?M)^w*~9|A<iKn)A8PjT
z*a7e!{SX@OJE$;^oRn<{mB&;YehuA<1EcemzUWt^biUw*%chx6WoRA(Gn!s=rAERv
zak@?NUuIEs4!neT!bEJ5Pws8`7HgB@Y9{zrk|bg?JC=U>W|oZ?ow=nLd%A38xWzzi
za21~+Q@VT9wrDT-)mF4Hl=|4R$7bTj-Siioy}CD)gF5~h0i4=M*(^Wb!I1-9-DUcn
zm5%5;Xw;ihhBXKKk9mdxCicmb&3an3^;usQq!Ky%O7v-53wm}38WFLEWZP9zxm&k<
z(s4X$BNx-?E1)#@e82-`wg4q}&CV8bm}h?~CeaX6>0VnG{;Jia6Ggdh%O8&BE5k%p
z4HvAHf<>MvS)gP;yPjKvxY1~TS1FBZ%MRZypOeb2nK=Jt4>(9Mz(JOvngzXC>^1Rn
zZqS{*pZ$+A^TEVflQWDHHjpsfXz$$Ayd$`o?&9M02On~9DY>7HJFG_HT)!#r^;;0n
zjkftAxTef`VLpG;^K4MEkF)kF&uM2P`Td@Ab8R~C${*Gor<nilN#R1V<&=jF;U7oH
zMHlYU_IjIsVdfUz{UqVxC3In##K0lCCdS9GV##OEbfL@MX>Wi&=aHCx3UM2aed9(H
zBi0Lt!xbnGkFLVR=kxK!H<eeaq_aNJ0=&vV%hi086U|yeik~+ZW7P42=NXFQ+~`@^
zX6@}R=>iWkHH`^KY<B&d;l|ir{jVw0e)#@A5k*$f*NRh`*_=qfz0Y+StiXS%7(6Yi
z9rLI)&!3NWN^bV*4b;Sc7EW(iyXMfD%2&dMr#~`~M+&hZ3LdI$hB+GF1?g*h_UKbO
z)#ui5Hm(}tZrL<+^_wVAxrFld^W(xwOmgN;fhx-oS__pAB6%0FX9jMMh^E!aDa4Uz
z$6>iPcfK$(TqVd4ua<W>E$mAh%Lq+-kU)7_{=MMdi=jlHui_fsExN1bAp{WW@0^^)
znlG1GWzIpb@~;{Im%mix`XY?oHYzehMd;?7&J@+wFh%Gl*z|UGYX1M?>@B0B{@V6$
zrKORU25FE|KxzmP5hNv)jv<HcRslgkLRv~%N_qf6y1P>tDd`$I2A<9Dx~}`VpXdMP
zf8T4lSiWFpExvJ{d++l&j?X9d^y%o*?@iI!ibJs~Vw?@yo)3p+SMr3yV+M;$CnU#I
zsO{K|CcNG*VMC|kd?$X|ns<&?@g0=kbj&;psZ^x2#FZjAem0WNk#h98ARmUTS&#W?
z`|At)ckUi^xMLnP&0&sS{G$1YJnW*&EvAy;2=DfzD;jzdo*9jsB2DM75JUpp7;c7N
zJJ^j78tfx%EAycKFT?ixZ}lj9LcYS%YcM5Bg@7pvi3u+#dRF8C#vHP6*ZZoi`1DLH
z%HP$-60eljYlQ9bW<#E2mZVU}?CdeWwn^B#bhn-7Ew4TrnExjWK#_eThD&u=YUELp
zhNDbqYUL*ygueiV`~}6+P7Ijr_{_JmMD(!qjUih&Z(;Uzh`5~-w?fmG+Tjp5@9(SZ
z+g71Wp1`^3uldqNxNNVF-VC=S^_BFcgh{N$AkR>hch#~FJRWYwGYThlv3N6J5P&-N
z_YFLmD`pVCKpxpDdkwCMU-ZK7RG*}cA29wH_kj;oIS9>dXeKjPmag@qeQ+qXdEb{7
z<Rr6Xnjq)madS9E@l|w_t`c{hQ1s8cpO*a^#ibrG^H!ujBkFr8$f`skDpI>uW$WP@
z6Q^sXm>SKAxx5$K$|MVYecen0v_p(8!;elUL^gAF9Zn=QM-QBtymI*d*@5SSCF{)T
z8EP+JPq}WYdJ_^{y9>1QQa8+rDNi~a>)0bXdljUZ(|Y?r<-S%KIsHD%r=lf=#24et
z2IEpP=;QWmJuhb8ra6TbD3St|_a|DNDp3R~t<Z!y)9^y-1*_#!6Q`l*iXp8T=GlPA
z2Dd{K>cl_7%p}DD8gb|RTk=1RrRe``EU!Uh$?YYX8KAKCNyE@PTkPWHpj?sHYgjwv
zGRr=g);lvSy!c@=uj*NN)aK1fuQgW&p7IU7Ro!ZWqA%>;^3Rng0t$M+xZ+DtpO(uD
z`d<xGIEo5|sZF1BSR<6q3ww2a3I#qkZu(DGi>E&wU#%3)m`IzGhN;%tzsWYOes?d#
zzn)Bc!u0tp&_|`z9v3|zpm-o(4M}nQuD8?sGUQBZEwF`|aS{)iuz@4aeJ<{l*51Sy
z{!aMMu1?2auh}32x#(&8M>n)Z<xAug1-uDQFtS!KwPU+X_H0*SJwxM$TUD?FQ(SlS
zH}GjLF^7?{IHpawj45|olK2|RnxxEz+6?NcPCai^RAOVDg3JNSSLUep#l3cB#-g}G
zYsEBT@i!R0S;P1B*e4NXIUiTy<l1k343`>P+x^_g?`!nvzxyh~>pCsuS3KM(^>hr@
zw6_zxvtP{z-Gc4Uz(&gPED!rd1of<|oo1eB`p+Gq;<Wca$T#OBYMVArWgC{vw8Gyl
zM=Z`RPi?NJ9Gtd!y`*RvazAXJS5-aHzE%<4NrEqxcdqFSXzSU=w_0kh6cJx@bs1LU
zOxw(<^U#~fJlu<h7El5Oi&K*yzaEarW!or;-e35v2Jc=#e$$&po7lF6P3(kISVgyk
z^^lE|_9wH6UaT;hhF9cKWrBQ3>kenp23D4MQx)7_eOT#XZJiX|P_QhGL-}9-D$|G|
zpDY(*z2h+P2#!yW=z7VheLXu97I!+`X?1JI`L(~sS?v7Sb9NDPNp@P-E{v5APVtbh
zEr^#Bcsx`ca&y0@b6k+2*tJXHTi&dy#&ixTcJK|MY29yG_OL0g*VK4~x>|C(=03%>
zd|~mlJVIYT*o^IIT2}rtth^9w9E4Bg>FU%Ti&NZ9v7=#N94QUu4GG@jj4bYd#3fYq
zL{~PHF5*S(IK^Gf)lZGuEMp2RSKAyT{aWSe!~H`lKd{5!>HgX5n^{6*U}&Y)FO73)
z)eMG){R{w7<RRslh959&mGf-QHLrwi*O=Mzc%g)Qa*bxRfpo+$nLvh`$E?OTJx<G=
z=XMiFgen~o7zsK-`|}M#a;;M0_70simkzZ3zrMj}4>Tqx=7#7_;St)B!|uVtDN-iu
z7iaA0V^j+i%CIy$bGw1W!j1<GE*=@tKf+~z9xiC-<S9ZcUM{SyW%xIxV5QRjL=tv4
zVknUaa^=`OePo4abXEDwiiBVOyj!6`)9dMt{|tijf?n&g1C$H(D{;<j&Xv{Ipg83I
zS8>qWUY;|aHMvgG%%yL+Ph=O!JIRf2!SyuO>(1!iy7wm;bk*y4EDqDMA~$wIgv(o8
zy&Bv)&wp6|HA)sPPH_vU^o*-B6sBL)0}ILmq!y0SL^kiAK6~I{kD%kFsx+B1y*N4I
zo-Ujv)5L>LkR8*Rl)gP(R*3i4Ep5-Xp-JGP-!ER$_|M`K%#k>Cza+e8KOE2N{(HFs
zZ?4JJ+KI;YW;F`lkE63yebPCsGp{N-M>j;fiG1^7?8>Pd%<M)P2pk(wFf|-=$B(i-
zcV3-UYOzuWqFR{#1o<pZ+dypL6of+^T^6J|i{gt`2^bF+EJxYhO&ldpG7d!}B{HID
zhWJTHut>}xDQO=6?0e6#NFB}{*Ot3nl!3~-Qj7g}VI4TudlJQ+j!!P%{GLlMPN&yV
z3bInu_`phDx<YY6Q2R(~Zs7@yqr&+<m=xQ3H)&1M@r1p@f2l#VHYRS=lv~U#Ua<C7
zU^mJKz5IzI*ouDpe%zlzWD#|J_+z_;(xGqH|AKF(gvcN<&FUqT<?n-Z-lkE(nYNqS
zY&C11T%=wN*<FzZXJ2@`pc&pW<yn=Bau2g_sR7tDy-&B!qptKLx_+*AKvmg|HsdZH
zP`>|$6OCdDYX|>GL=&T(R{WT`R~ptbI>->)i`q&0zACPKe`F?)P=Ok6l?p3l<}H_d
z!4iqs%De`~cD(2GrcS9Snb7WvgqDQZxkZs>TRSB1(xg7aGQ)3jjU7uTh0`aqwyj-y
zWYSw6VP>4LqkQ-yO>j2ygY^PJ#lKSba|a*H@7dFU^VLK@1R4F2I<@0-nH+F(>BFj8
z#*g*%sdfYRk`=d|-~2R>!wP~^$SIHkHup(GrMUZZ#gi~`e%rs(KgpA;)5OFjG|4DA
z(JbB+N{mb~t{7`zm^U}_g)mbi<s>>MxK~rlbduk_R6iA>==fIoW@#HQm%$;B75bz)
z`*$`@zjg%=+;~nL$)(pe)|d=cKs?eQBLAF`8>x_d@6h7qiI+BC<5`x@!k&@|kFxI7
zQ9h)^%V}5!`&_tU{_#6*yf1A=>ZrGe5{sX9%$NJ<oH&*>R(8lpt%K;i=yQoWv)(?v
zxNC9FW_G#Rs6PW2J*gpO%`QX=B3`j(#A{`}P<Qai>Tft*@VS)uoycs=p|?ntuO5(=
z&%tnB=4PuQ@UQbG@WLLhc(6ks2REb^MtrPw<yH;a^SGQ=HY9w`6s^!L&ats-5O#qJ
zvFqaH<1;o6#C+auY~eP|t$Vv4Z!}WAVPGZbBzapW==@6PrnkyuzW6ewBfI~&0<17?
z5=vW9+F<s3&gIM95d)8Mde8~$_io%{`9}L<M%o=0&v(fo_xs*{sm**8!++ZtHdlrz
zjeOMVAw@B<BrL{c-F*4LYC;71cDCxE?BiMbBkk`+Z|e+ubeJNbU9*~>1&*{%NmV$}
zK8|yFY*Wii#|;rrhh4*LM?OyHY9MjB<<u7XR3Gv;3qyO!`1n6M&Int$$--hjgxaxd
z;qe8xz#sOhX}3NB^WD9DcUo><Io;M6LeFhoC(_-Ht_H^z!<7S+V`M8m+RFujkxtwN
z&kxzLXW+^B%RU3^`-{bz!E8%pab3@}jiXVn6t7?k(M%<(VmAJ_8c7Yh1%w$VbWPC#
zZ31v&%eiQC__3#4`i5%wJ4bLfrDlPDa8DubPk)2x!qn3TusI9XO1r0mb7J3j4aFh<
z9^(FB4u8iXDM&&r&en(??%4h_F4;Ab;jn154duwA_mAio)AWx2(NE?-8*6ECX#pLj
zc=0W2&wMr!r<M?@)lk^%M&5u5<YuLHO@pgR3%L7DN9BV}9zH@ynnD<&z8O<8oI<ws
zW{!kZGUl(RbMp4OhD5|w!Hb>^Ui5T^8f3+c;BrE5y3b`x!a?l`{N|8UlKtkQD|jeZ
z8{?`gBnqlY-}vZq4s!GM>=*IWm6J^c&I|Tf!hE~f1;!ylX|30%pMZG}Qn~YH^G63H
zdW-ryRB-<T-lI^5^Fik@&mclCP-irA6(&551(Ibv=SH(5UFsnk9oh4H)ra~?XG9cj
zO{c&1&Krgpq1={wQ&8QPo3_$eUZD+Mz!v)0lag(DedvkH5z4a(LL1l(cgwHI(;SBo
zBA$_C{p<CghJWy9T)*93gu1tpcqKK8I~&*c(m~nm&-hX1WRIzn+FPuF)O4UsjlT2l
z)#u*<U9>I{Q3^4^<G-_c174Ew_i2FEjm4y>hkWf1V^#Ua70Xr*fer<NxeQNFtVsEA
zxQ3nnbMM?#(wjn_qDfS@P>zS8VNX7R*{6LtZ&QQDn`{3tL~*SG34M5=Zk^;|zmVV0
z8SV6i3LFZ?dt1PPF~46oe!wAQGPr_sY;F|(%{t(d0#Fl}w4(jxun}wtGyP=uMn5rw
z#`qa!nGJm9i_+vZG8`{kAvXQK`-;$LRdLbR8cpRK5_fQhnk~!Jn4IlvOqb|tr=VB~
z)1wV&`lm`bzdSE4zAqo<eR=E2F4Zf```w$l9!PFDS{cvGNYPq`PuokC5FHlx;jLfM
z_OHTKmYaEdPQToBAJ=+FvL&^@hsN`w-agJgE1ZBriT78>-7HaO9kDHYEsii&zOhb|
z@{C~Ih{d&kN{2dsKE3H8-sC+ypXW`wAN<LuV%U?y%?Zo-{^{qI|G$1-f+QcR+6>9<
z)nVd+M)CYGRjHUkyBCKZ^QNa1{>rh<vT)L)r47sju3vA29D}3kUQf_IMsvYjk4BqS
zH79|j&+$)MAu91T&po9-UZ4#ej%|&polsXoBh$qSlDC3=<B9Z=%THdL;g6%LrUqzH
z*qt+TGm8)7eJ6Q%0*}2x+(AVcI#bE|9uWOH{0UDUxT=)ba~~e*hbVO7KH}hJN)?Pw
z@OV6UKoP4;9%>LHh(B8Z&UaK!`i!ne9hBZDIH?IXY!((hXq7utv#{wZM-ydWz(NPs
zeS64s$K?H_h5<<1-jy&d$7S%K&y|GsU6b2<2&l3kQ!}ys71WtR$)N>U5a3ziO>JI=
zJw!~Oa3%c-uR(0{kXmNvP*D5`lRmBHYRvQy*&2EY&b0N+%MSZ*y_3MTsFp}6+bGR+
zN)&J==f3yr)B^Y}!x^jyPFrIHWx535Wr>1_*SuypZ`xUcL}=eU2{LYBya8U^WgRU|
z3r<R9j4HEo#*9R9pdEQilE5r4%_drQM9rqAj++xpCK|N77{Xv!WG2_3F&Pi0f{T~v
zZrg71%|kzGp2IIJb$Sdh)(`S<-$Ed3&oc60bg4SyOQ)zI#O7W$zpiUjtSyxM?v3-!
z6DfQM<g~q&Kh>+J;8rkc1ENmm>2%D61|Q484=6h=Z&f0)kkBrmR!pRG9`~sD4A*bk
zPmbe3t5G4lt18$bWuA&ocvITCzqm$PIv-n7@%QBnEsMp_^~K67P?BC__yWa9#j&uV
zxSrJrW@Wh&Z$eBblf55ZKAXJsChex|?E6MI1X|`bnxp7uXVfbe-tpv|SMcy!Xyheb
z<f!pNZm7a~v<^5}x1|<~k%=4%sh-#~48<<W+E^}oqQaxJ*7o_gBQSYf#`m22a4yB1
zR`%}kA4E+^vAuf1K;d2$#(*$mes@tc@$5;UtnH7wyuBuJJ17}aOK4TdoU{1xaUwHg
zY)KpqJkk2^dXtIC4^X4d?fF*j@qEtmm4=*+D)HGhzn8_29g@+Gh);b$ANeRQxpvX*
z3Skx{=}Q-}ItH8|-FoQHn2O@F83etLteW0o)-}bO)t89x)@JXMfR&3s+7tQoZ>keE
zAtBnNYD`WqxPdBcjv>TDgFcX%QLu$IQGayYcJqE*OxR`|n6zOu{4;6WK}$GO+lJ27
zSrFD?EgYD(TMQ-qQqi|E5}q+9ZPI<JtMxpg^<w?jSRW-%k)hd@-5>N+@Lul>?DJ&t
zK;BNSVgA)oE)Gr?zljqt$y+KP2^gNQXGg(5Q$aI_RU$wjglaz=<s6J1!%cl^t{>uF
zA8!_sST)dBrFG~&@u`aVFPH@;^y^xh%lK_~`$QZ^3v*Ztb}l4x&rx(;yJwld)-qJ#
z%y6sRd(YEGwL$CBuWosV@o(=_=;xoI6=R@KD-(Vcbe}c}iQm**Z9)w0-IlP=_OH0|
z)_pJ_NVgZXUD^}_MF~-7;s;*r(5k|Q)QX6YqB-599DNmOjDlEZ^Hn8*9mlMEEZM~%
zx}?!4Y*ce0H(Dr$HJVx&s1PB4TTDf1I~mupvuI!VO-91Bpa1CT;=Fn^_R+0!muOMN
zsrWuoBSNQfvxp&q*+^@(`2*dG_-wO9WP`Z-wr(kYQb3-KALSL$7c8s(w@nF?K`*Ul
z_3e?k4Rtv__CmpIfMUm&GO`16Wg>acPRyh+`#j>`c;{iF?LYc&Pb{!Hg+|_UeDyJz
zN}==kJ;Emu%WRx#nwI8aKoZ^R7|4lytZA6t7RpT5`_7~+R561Ytht-mR8n>IP(9L5
zgCDtywUGq3eAB0ji?8i!C*ZCpp>i_6HUcRR{-dFn$~l+>h<C|$1Rt6XNQ(4bajnkw
zer`Rw5q;^V+Z>*Ox2B(uKAhNQYUs&M`$?q87^HTb!}^m1{P+2OhezQz@XNFaleGvq
zefYF}_H<Ak>)Bl*c^U0cA(<ABVeQEHyxGr?KpkJm4P6qp6q9m}L-BT-$Fhto8`pA)
zO!=2ka<G8wXDM&u9K?P}Ql>n^>|ZcV@Oq~8$Uc=pmYxL>?Z8Phj?NJu?6dgd{9#l`
zPt<Fts6((P>U?IZPR+W_<d6RrR5g(_d|%Q*!Sfu5C^;wE7nK~kSmdol#HfQ3i&pM|
z;Jz|_cYT~2wEkFM3LpKT9`rz5%hPF!1`mj2-ARb^B%#q9{3OLEs$smAm2a*Yx%&}l
z8hGLD%-81)Xmf{q65FL_1ABVdyD^#3@j4BreF$ocB&j*Oj8Y2^sq3|$mkj~HJz5b=
zX~3A`cLtkDQ~j;$0XaK9J!Kg9OCzcJ&)84*Yj~TCC|?6I0+v7Dl<Mb?dV+@txcGc&
zdmy>8j`EXq+>#3@G{E!4Q9ZM`KHJtKvZx%8vW2%6_F0}vdV~iMu66<qUWVUg^sh><
zzcd?LOuEgyrDtcY^-<=Rehv(=EEfI3Gd6+SLmoGoyLzgq%f{_2L|SqQe3C0qT14R_
zm##_80}_<ag*PhJE<ziQ#5gskX{3qd;IfCk40FNHkeoZ;nuDH7xw!05FI~-TJ?fr3
zyRw<#sr2I#Km3w0kxTBM1NycEEgye)?3f<?p<63|B5&>f`p3P_p{}(8^1I2?SwfW8
zW!f*h_^t-^9(~@zAxo3{)sbW2uYpNT6`<y6dO!ZKVN`^)iQ^Ab!03VbFq=F6*G$@G
zbLKEvN<$C>E<W@0SkUf%te;mAgEHahBf;s<6I1LNz7>o0TFWx*F^XD9jF&O;q)QTQ
z!rtAuWjvd(`}_-Q)I6OlB~UZyUOTC3zwX1=IaVWJaIfkD415<-_z7XU62#&cY)v<)
zy~^szw|(4A<sZcN3OiZ$jKUBmOAK|`Pc!NIHsdJam`sQfhOkS*9EhGlnH|$)nA$oX
zYEZPHTe;ilNU}8uYY*wiL#pvNMlnX~?P}5Vkjh|zS+s5PbnWq^o^9+#5cy05zZ(2p
z>fj;1JW6R3Yxf|e_0R1L?~?9^=JT&$UqcE*1$`j=O)c8`MOaYsJjtJ@S5Wy%jgpKL
ztz!5gxHB7DNQqwG#Rpy55@_tbl*gEE3{YI#m6c@eNJj&ZBVkA#?m=bS9TM*yr%1rF
zsn*mP>7BM;P8=#luy)>{^sCPeiAicM@5OvlIp(tB*cc`IYT=8B7kO?*Eqz{pGga`r
zS~{f|dT=GJG4p}mZ|+o5G{j*$6niyoU$aJoh^(3+PV2%NC?jAW<3~ha?Z@_bQfQSl
zU0;-+#t#M3-oJQsrg|>ATf4l8WIm8qNdo=v>O-BU-XG#nL(Vm=H`*naibiZl(Z95H
zO)wz~RQc=`bYAZdAUyBjnmQPrY4jk@P$uJ)DltbA&nt?bf8-Ta0qLO|4nmQHsPir-
z<<*lS?z%)B47`J!h_*0C!A>Pf8d?*JR<hi?6vSh~0waL|ChEqX!crga;dOfU3{_b=
zT>xas5NA_&QJm#<^l{kqDnjWv#fheueMgMq)$+whN3~N;OQFs_Pa-$P?ZTB?jmKi9
z0qzd3^*iP#NSrWYY<f@Q&Lv+YyI<C~h6i0N^7By<^hgWQtfcNq{Y4Ma&Qv+k!rs1u
z6`|I#fktdZ*9Fn)CrlA4k<dPdV;VnovRW%64w02UUxs@Dj=B{1t9a*+5lyHeP1IpP
zJd#8f^fbcj*H5)~e4;JD$(dFk_Ce{}=LMui-F*eUR+|zs^^~(r7tvMNJ3x|w)Qb0r
zp)OF3N!%cHF?`H6Btm6r{0M8c%$I>FKJ`=aJN;({fvu{JJlAIBc>3X;Hl9BU^#d6=
zXZf!-+~-xPG-@jjUQ7@v#g`VA;ow!QrwDbXB!*Kf<S471ByaWB7ar4H=YGFD$bShb
z)>%@oCQu?F($#k2oGl~sV$O}w{iqOjt%RF!R79;xh}81W#O!?+p(vQq1YA64QU?C`
zKPZt}>JMDSCeuwL-aD2t1_xi3lbe`(MKF#e=IqIcBVZCj)xO&87G*|Tg<#8ZMm3x#
z_d+&=(Db@nnj}pPt2(L549Q=Le}BP*G59Vvn~N^_{U2QlL26fWEo|D_vZ4<)%k9X}
zZ8;_d{<PNPKVuU8i+xD)Zdohmz#}P1Rv-rVuVTu&Ib*{M>PU~;$ajt)9jc~Z6W^8m
z&Cgb%)+mSx)Pt5{mjoN`|EUI-NeUk(_?;S(0kI^;G<gJh@muAJdfRM2jDH0Mtq6)F
zG{K2fAwu8Ak(cobcQrwab!Fl?Ypciqb_k=w=pcgIDJCn()@smU%igxBES?<mu9Iq}
zUze4z7$d6J_=sMH<Y8zEJmYGGJ$OjXpZc{bwZ~?{AE4jvl}I4y89T4-Xxa$M{sS)0
z5AIl1jj0x<VxR}(z>Q*nEKHp$H>_>>X`=)V49M0%pAtW&J@ysR=az3C8-j4&T6o+7
z#+1ch318AyP8jI9Znw^$^C=f_x@K}5r^%fj4D}0c<u`mGzuvCA*>xMDb>Eh`rWWak
zwyn{o^VW=35a?C@sNQ`rGtEl*#MTba9501o%l@biR-e|CWLLMt1|QgVHhgdYKKL)+
z>j&ZF@X)^g%KkWBtD&rjngK#R^9;nF%3DLw@yib^Ysuo5aL(C|J>VL3pRS-VDn_W=
zQwVQ4c?s9=R?973wq&rd(VFt)^(6=fxAXa0SHf~9boQ_NpEwV~5P_$?C%-+W@F~uG
zjy{DvIXkXNyxHX4_?%?xDcl)1p?%)}a%R$K<~KLL|7;wuy8FuBAC}WTde25FNm(k<
z1n2=c*F3ZHJ3M&QODa*hB@*WK`*{4-%<kp(O1uc=`BBn#WkRQ(pVuMmH&-7tn=*AO
ze?wK}#^fW?oYsC&6O?O2%I>dO7oy|ED~gW|%obO^zUgmC_F;2hf`Yjgqr}30a$i0F
z!Smo@uFwFSxCQM=FM2T{LAdsJy|R-J5RTK~JZF_(86IaH7zBb0{-$r!;x&~g8C^6>
zndd4b^^7L-k^VK8DGo^9S>D#3!UcnVLB24ts=DsY?)8DVk~5VV&*5{q^gIWdbYoNc
zbDB2-qFC5<SS;yG5#2cQk(O^12P<>#B^)O;JWprySN0s`9N?bDZ1QHJ&k;gj3?ffX
zDREj?ie@*K_g~)oo*ikUMqlzEYTtj;Ea@zt=iB(yZqU8SRJGA8+F)e)y!+faru7On
zk}Wp<{MwNAdN~TRKs1~6r>z|w7jFfV%rOSap_dsU+@}M~An8W#Gf;oov_|sPd#v1K
z>^SICXKW~z3&j;Xw)LK_mQ3jvzw5dfhFVyz2o;dx-1;eX6<+np0ni^p2eLCCpPq}2
z^Kv|lrpOUNnf7|c6nD#35+fY((Umi=ai<n#0y5*}D4sDzdV|!L&WE#3E0$zAJ(}Yq
zIqcepih83-{L#->p0JIZh6vJ~QllFSzIP#OFb#3Wo9am%hn`E2K;|}{>N&(@oJqwx
zrgqh1t4)nkn$fb4w;jzXl5Rd+BBP+rdlALu^}!c|9ScSsutb}i^`~2IMF{yE2Xer=
z6GrQ&)EQ{Cm&5@jPHM#C`R-z!LB+1PF&+QGQIi6PWiAOj`CZjF0yBX-c#=K&RhH+U
zW<(`^sSVy!{YNLnwok>)v}4@o$+Sv(j+GFnI+#4YrK)gE=tideAzL|EK$a}yYhB?1
zBT;PA8Qx;|RQzvSXiwA$j#r?7WT&}$b3SkGkL+tR89bCvFM%8O^$xHz4`i*B3Yo~D
zp3kh#yVF%XSq*~pItTQ047AFnIS2BFDcqw?U*Vf2UV&~}FGZr!d4+$<yLu*gozOGK
zxn<WXts9g3r>oT)%+U7R@6`o8&Xt5|PrWX=w3g$H-4DBocKAbSwA}E<zmSVhCw|EI
z)0Q@38Dccl(BL}uO62^Spn}(0QSW`0PzA9wh6{QQCp4rH6D8$t+9Tep#bLJc+4Hy5
z4J_d5k*1Was76m;OA}XDg&)YkS>jq{(V{M4=4)%Rq27s-aJmisI^m8Zy~sv4P!StQ
zi5SsL9R16D>7w$6Z+7EuhwUcOtV?>2-b+;`%XHUK#m4kZ0{KOD(F;2hm6)I3x^ts#
zfh3VhfBwfMXtMNuoh(oYk9=cc>-%d;FCuEn8H%U(W+DaD&Be?uFHCg*$UiOW)+*T3
zm_!AY{HufI5(LV`d+z(=>f{nA#j~&1=X@a%hkmKJ(FkT;e7+Yd%-b)%LVlO}6!(yz
zD!wXK-puh=bSZK}Iw@&y-MskMCKb8~W7O!-w>=rjx(*7x@xR>sj9rm@{pIAJEl)yx
z6V?QuT+u&&sQhO;tR4j2r4d-U9@)|GIOn~dLedEp?ZEg4XKXSPhGrPQ3t<x7+D_|1
zw4G?5)gb((i-qFw%*!X1pMxqpXJ)=NRK6#$L~k0z3Z}WLx(l&PK%eKW`Mm^lY<u(3
zmF=AvO^}OrH%eT%d8gk7CoGX*uZt4)ZZd$oF#*C31IZh()o7u`&Nr;vo>34Q*MmjY
zVqvN!nDgkHjulwh%$tto;^s5YS!2HeJsLGvVS%0E`>vDsoPhc-x{~*4>OIy$EU~nb
zb=SOiYp{ypjy+7Gi?Ak6da=2mq7sIlRV|Pqp5MTjH1OqHWjtz-tu3u2^w^2jt^{r0
zq(a7xfuSh1wMdHY%z}GLk-q$1n6{ytZ&ZWn%gE}}KX&Ns)d)_N-Z<XTq=x9QG=g2R
z6yL4W3QRs*9$)Gl<>_Ag3=dACI5}HU--Rbkzh|^r8wblLrUtQg=&WtgCkC;;rj?k#
zq5#cP^!6BQGTAL?Mr>dZYdV#a_Uvb0+|yKohNme!N2_tpDiU*UP^99fn5big*79Jn
z);MOGSl0t@%a_xUWwSw5s(L9Lj#nVsz+#ZhQeXetj0l`*`AxXU+sV-;T#E&;`^^7Z
zHRL?Hjt^+ud~}n65Z8?6Xkd8cZ<C-}x8puDTSK%sE9voj%f9}+6iB9i)NUdN<ECt;
z()~7{o_2AU7FFBk8Lz|VowzNv4{U~hyIs!PUu;<JTf45d7yWh_lG-!$U1SxRKN_p^
z`b)i$pzB>c=h=5)56r@K$rkAaCx+K6X*CEZ54|LRpM$;MV<!frg7XrGk<Pa+T0Qd-
z0sHtTI_`&2=~KyH46e02mX8CRqJH04wH;ptuj;&=oDuFk@2f`K_@f9;^C8O4ItKoq
zeKv9f-p&ct2<1xO`iSdRb@V$W%@)53o)g|&txXz@@z(V(ReQAcY?U_d(*L|65!v|K
zv)jX-E=+ObIHp&xJgsBshgk#0{2qDu>-|fHp1F&<3Bd#$KW+UzezDBYI&uH_nVNgr
zOAs|G6<f3W6GiXQ_o}9jo!+na7qWEfcelDrT?emZ&9t|fh(!!$8fT1fLke(F(Fg2b
zWAFYpGF-oQ?5MT}p^5I|I|t-7ixyiQQI91vK#r<Nc1rFBO2uLNrH5r~%&R_H=Olt_
z-Q=XkL0e5RXJN<9L$_KJTsr$H0}nUh+D$$tQl`s})#QsL;YaQ={uL9$hW@fDIgcI(
zqjw!v`E{#xdkD)=HhtYrJ&6<55SjZxWm(kjCqD3`Rc0=t-z4p?Y~9y)ai6hwO6=w~
zjUVWzvpQFn^dBQ-Jv-x;Hdlrn(^wWoFz@iie(CIPt&$Y3+ki2B4j@YRW#0n=lp~Mm
z+*dMPBk4>cBkYXyRc*pAHZF0yr{#aZB*Ly<x_z#AmdxL`WY;&sM8Bd=(~PVjhlcxp
z`fWu@-k3^u>>*ogw}ozr2jXJvU4$LCqC3{PzaE=f;_5dN$=_g3aT#IWh)K={g(ad7
zWaaN+BWd15@289wLPl$n1fgN4vurJCoaj@to_e1)VJ%@oxz!07Y><==Z$to(H-Fi-
zg@IfxPRJ`!(TmwQpA}=xMr6QIj~#M+MphgV(%;t|hknsi73EjwD>=IDRoSq&g}Xb)
zWej;VyghR!=KQ&Y>%@;_G|HcEF*nK-Jwz%h8B^I(+dGw$deRX3d+FARjQSgNEns%H
zWu!De3WHl;N;PP$m)2x33Gd1&XogM9H+e|}8Yhh7e=9k&F%e-!%fPk{f!%L!O5~F4
z{d=#|`8$oNf{)@1;cisGRko0P_WoFnO9ZJp(Fz^x1;ZJ?zx;z(|0=%-fki$)5zWy4
zrneU!hJ@uE&W847?_Dg>bNrP*I!@=r2?euQ;))s4Tb;E0v-kYy`MTgzj`|5yAHklh
z&onh0`Sn8tpNpg2B%fVp?by~hB1g!&&r``Ub2oEa6Z^4?){;pqy<vjYUys^BDh{ur
z<ZL?Th6v2LXTM4z$}`;|DF;fkk9!FY+DG7{jZa+S_<G5ff9Cb?Dc>!g$#$DE3BCGU
ziTTC<3cmaP?wO2UKNS+>h^}^dh2ovlJ;k6Ni^E9ceTZ~ZZoHA+?_Jr6I~F!E*6(Dq
z_x+ljYu&|A@#)pWbANn4YGfhjhkegLaAY|4aSkIjvFU*XUv0Dl%d$dYcb^szEYZ%d
z`?MqSn|f-w^r`z4*krWxV4BZ+QY_|25NnbbWJh`%TUZKupL`>X;v81&SEAm#LVFc(
zjJUj>p6RUW!uMi+8tKD*b9k~d3CtF<NWG}9wfd|AR_=|4x`+F-qA34{QLXJlMXX03
zA`DmCy3YuAJQiY9EUw!66#5MqxiKkwt>UjA9+SF}FYEUoV=Bypl)AMuuSa^*njADa
z|LWllwDrJ}U{c_0$VSh3X~GG1vS*r8{JYOn6z3F-(tKs<!glYXWG+m)JyJznpj-9h
zZUGbjs5W3U6r=etiPlT4D6QGDI4|a4un(2*y;dSm2udsGyC@90XW3bFb@lGX<VGNB
zNFQf8uH^|$;FfzM?qYZpv~yNYqBk;L&0)^|?rwz8;69Sh&-@5&*uhjkP2lmaljKkZ
zWTi5xYutH3OI}(K2eXt^0Xr5qKR)}ghtZWP+Wv%RXps_~4=3x4j*a9yV={;NH#$rr
zw$3mt!~K+qmu9Wf*vTXTRNKbL_lNuV$rXHEB5z+w1)oyuX9~UO_^|AArA~athW;Wb
zB@)A_8PSNvI`BJBm1s2hOz0g&YoLa4<=yJWB_hfz+f^LP5ezPqMZd_a1K!qWvm|+*
z%n#3oT>MsxRFE%5L|SWlX>YfhbXezQZ5WLO(~A%S!+^vRVxnG3=m)_U=lK$9z^<kA
zG&iCx<gz{4`Q?mN$eRgYO?U)}0+y;LBq;W0c!hLbrav<*?dKmI_C58coZ>*ab1n^x
zXEF&g=u32?pQ~7ZToITX%;-o?-D`^*f&<T)xWljWg!rvVqg0EpCWO%DX}`0Ukc6r0
zvBsb+kjWBQ$*bF@n!%3n--;Ffo&TBBv?H=X_I!y*{^u#e2a#FZ;cz8;J_P^$*~r>E
z(x|lQ!eOsVL7V<M_ND}1w^j;@=on(|5<%2BoU9@bxV^9EJr>+<{0Ejyj`0y0))ty9
zl@6+wYwcT-mlTm(nW&*lIf(_|w%HDk8`4K-Z68FV2?v%)F2@i&zO)u!Te`Tf?R{11
zuTN4O^rXBHQ}^Cu<7N&lMT2Iu_=h&#LHGKB#c^Bc(=|t-!J|)Ko?XwJdDwvA&^>dC
zNO{6X^2gdEKJr5vM4kRVe<u^jpN_J3t5$_eDEWc&9LvhbT*#s^Tc<jr3a-V)l&UwG
zcJ^He!2`yHa#K|jrCCTVhC<TU_&4b?yB`Ch1wyd{%2uWmw}L&RAGHV8lG1O&TIu8R
zW_w7XOCyvif~V?)#u_ofl-&~p7aRjuZ1gfe-r=nVN=QQr=?J@fyVEjZ2ajkEu$$y=
zic5Vbczpc~@}t6(`#fX=$m<d=Fe#j<URbtL=&SqPY8JAbNfAVAKyWMTiBw?or^?BG
z)GMwVO*HGYgJgsRH<E3PsQBX|h!!&=kRdDv1vO78CZ4Rc?8Ej$)|M>`hzckJ`_?`=
z{@NBWTKj@H!-{NCM_Q#DSq*iUj%U{r2i#;w-5{@z4?O30EWDT>i7W<7^krGT%@sOl
ziqjf>FZQ9T#Yoic3q&^QE#2X&cn+o;rnSu51ALQjjVwmh%ifu(`NoP8h2y9)yd4f@
zMToB8!tkR$ch4GcwzvY{Mm#qesUeDUEn)h0>W_YP?yb_E=t`TR!;!5dKbX>ei}-FL
zr7o;2_1eVA?x}@2bRE26Bcs9-F2XkCB*~k1UQwDQq|($0`WYigJGC~@I%U0}qUAuE
zu1msC$Z%eqCKkC?2=CbW`>IueKKqgV-|x1v3qiK*NWpWi@$=MJk8^sjBEEAT2Kv@J
zl)W!;(7iGtm-O-`j7H3s%@J4Kr)^2b(H<|>e9y~Eex<%yJTa^&OYFFrcTP(nYE^8+
zx3eg6r0n=LaR}XO`KgRxF`5i-Rl4{x^ucx_fKyLOBVk0%l6aR9U2{;@%T~)$n==~~
z&(5)Rq{*??uS%R~=pM&DDzx{V=KC6w@Nd;dEo0)}j22a-O?UXXNDh0z*<2y-z3W4j
zbOVmI^9xSjrg>jQ(XXpT(F}nZC6V&0d^f&)<EV4~`15&di7=#Ay3dGEj$P%II;&Cq
zCl_EY4ByxE7)v0f*i=PwdVOk?tCS$WBt`1v9~%UWqhuO6$MSb_J)=by6~AJfhyDE2
zWZ}h`qd4OkbXs)ld@VS{pzZ4og+5(bCa&Sk+)FC8<P6BsTh|MA&@uVM;G`Qi+eDnc
zFNu7x$b0L7bxJCXs3wh4Yy6I`7_Np`om)FgSxh}O@>(U&C>33!GanuP7*f?=x$<T-
z(2P5Ah^8bx?ZKWK2q?6{C=~r9ZG=(QZaRlPMPIp-f?I+!PCxt!2mdY_I2^OOu6!W<
z<LP_g76UmM;vn!4%opTeV>~l9#<|)Lfw*?j`xMu}gM$^Hl5jA;>+s>ZVN_KS+#+{o
z2K-!rsnVppTYOQ$e}`L|2j8mtJ4@u#Nv3jGXV6^=O`mx*P1+MfafdvdY1_ruo>&QY
zCR;f4fZIB<O$;aYpYYm$g)IK-=PyNRvz=ebk2qyn<(tXKRzq-HrQeY=>Wbn;PMpaX
z<+b|J$s2PJa{vih5i49V)>Z{D#s+5ABXN%8!T-Qg%8z7S60Mlnyr>+6oi;nHPVMFd
z>~%b2+T=qk#)8_e+(xvNvvTP@Ed}Z5sN<*S;y5@Qs`wwCX%NeQ!???H*OY`948;T8
z=@LwvknyFpAcnrx!DJya{)0P;-eyRV#PNHsr_TY@gNUM~l9i+P-E9ExfB*NNzoFpv
zZ+QK1HR@jx|JG=?Fb8pf0N7fvl%|huZ^3Y!e}&q=Wqn03qOIPBN=k_T;zUE2|8vWM
zxaDvR^h5M?B$3@L_U)l0xA}7Za{};-IGUdrgcAGG9w5$t{7$sOwD5-#1~%oSKkx-_
z+omU~J9pERk1^ud^DZ80V}vl?g;TX65lz>}rEj;5JQp#Wf{9-$@Gv+0`GEJkJBG#f
zG$oC{lOJG7MEtKjMclWiEuuEONzq{W)g#Yo>JkmTdn5r-s_`?OkVQm*cV;k!Go^uR
zOsC`5G8(SL(#8v$CP}J<yEF9aD*s;;#y1fhtKwfGqbA>EnSkPwxn%EO=qR%L#*ALa
z6FMFsyxnr%kInHm?)`MfApqPEe_vT}5c&R+-)AotPmJ^~iB(FoYnz4h*w2kwe?ubR
z;>MS_Tl^^%*v@vrDccEyB~8<xXj1sB4R;O5g%42@Ba}gdUHSmE%9N?pZ5Dh7rO|H(
zyoe87{X#?f_4%=xGSXhARymcXPpfc-uPae|azv7^!#$(#MPpA}FAvKRxmDfud%Do3
z!7&%FA~f*WPx8m*Q?&hmpfIRYgpuBNnFoJvlYma=_xyGAUHU)Hs*(ZI0BayYb)1gx
zf2aI;r@iZ(^^pu7{iquD&f2S2<R3Q6dqSKtt?nN7jO`!wyJB6&CJVO_?)Ct;;BuSR
zJm-5ja9DFZ2R%^**Yg|xdy`!VsWx<B;`YTvmwVTN24Yy>+GRbx(JIEWRcexm&pf6%
zKLCmI-{l|V5*82fpYvcwJxE}3%aRbt`Jxv?CyP;ba<N;#N70#d;Sbi4JPrBb8}}Y7
zg#*aWAOVAv1PXgK`#;=?|6*%ML(v}CmPR<LdEFw_;`MQhZ+}JE0OLmk@}<M^9=H|`
z4+PlGW+v@2{7$SdBJBTXV6bMjTUJ|U>P1-n?H9Wy-{I~7#VlS*PGAmJSqD;nj?Pp6
zhwQ=knJ+*QkQ107xBQ=BaQ@Ard0+4mAe!>O=DdFPza#&BWWJ?6cD&X}{qN7g|NVF6
zJpkO1RGTjvBqaYUdLC8SEq#wn)4+-1fA#q!8yvthDO7wZ{~xT4cX%@hpVUu)+5=(L
z&f8{hCVYeN1}PrEGs*cco{7h8ma3lTVsOao--j_;^BL@dwE^4?M7u|t>)=9&v3wIQ
z7BcjH`|tYT{rTMm->(E_O$E=m(!5C5*F4|~VG+GITu&+5Wso?}rIh26Z#GmW4_5s5
zC-TtH!g=9Imq7dS`!-;^diqF~>L<lH1_NM)1Z)G^%%WI^@Vu)g$o8F^Fc^zkV0B#n
zEb>!mNTQOWc{x&i5QI_sRwDv8R&E(@q-@IjxSc^hH>gvO<<8GE%IfP=W$E6V1?fXF
z*mNtmp}_tq?NXl~KbxjBzgO;aI(T#zC8kR)g|gT<{{ey`5d3)d#}GHYl^IP^@mrd3
zgcQZ<#%iL`iIcA1@x(kxz}MuIl6$k8@pWbOW%8e!Cfc=Nd5SmfbY3eloDJ~|<oDV)
z?l)*4B%+7aF}%ZXFGoB*R5q3&0N4L&Q>V!pYy!;Ao__1HU*)cbpnvVZNWwWHdk<8|
zLjr?W7h9#5jwbtN;CN=IfWPy-qJdHMm`GF;f?;jG@bZ%75z^%rTI%>Gc`N+3G40<>
zixbuAeLReU*o3+tFZfV0+qjZE$!ecivLRv3@*t6XbYC`<VqP96>lr(aeN~ZsUF}=q
zUl=5&H$xsthX@aZ<eKE#cIqV@;VZRQ?*s22^v&PxcgO$_DHRz86_v@C&2YX(E3w8Q
ziNlE7b|vh(KAg8_?P?-!Vzq)W7WY{C#6AmR_qD@Mz-hcKDD*{ien%tKX0sb}81&xj
znfzY2Ot4u}RSA;bDNq5YYS0bXUClZp?gbJldqd9t1iG)r>&yZY5=u;msId*-oz`_&
z#(gD5VHPk&1kdiQnEqCo&-jVz!BnUxzsKRAgE8H?bA<3eOhHTo8VO=`7n4G_8T5&w
z*>Y8>^Y(Rae$iBqTUZp6a+cIjASv7iV4fOqAl?51jNh-IvjC37!8EK)^Krs0=-tuI
z`ql`87zpE6^0mDlHr_oY%DHwv<TSkLO`^Pg2B(WF2W<e8!vhuWW>HKMx^bW3yT{>+
ztQ&h0GP|-CEMd@#ju0<6R&p`C4^i^xg^v~AN>O!Xcp_R*hZ#3jAe{b@Z_vSjgMTIG
zjYgEfk#?iT<=AWIsOoTywW4wgN6GI-_s^~lhi<EFIe_cTf!5vTISl&9Z_DK`KO1fS
z&Nqzldl&7hlJHIFaB^rL%E>m{T)6^Z)xT=GG##2UMF604fj(I}QI~bO2-?orEvV^o
z%jjknWlp0odryugin^@kR_=as%OIg>VolI(g(LiZ02s}4C&~Rn_&qq?gq)nrIn05h
zbW;!*n|1({cv%oJEC%Ao$oq8{n?-W~jA6O;quxFs;-x}s#A+8hkp%yhdLp0<{Bw_}
z`>!;+47|jjjHILFrfUG5Xd1kpw2vbDK9>TFg!`x>`-Ft&TABw8WVgdb7Cc#K&3e8l
zvL!#%Wrda{9V(|g4R|hpo?2c0@RfZQJcR4A-)4j(h*;LwOh{mh!IW^2n^zSMDY>TV
zM_s=kH?n^+-}fWR%LIHS{FWti(u__NXOL%RmMpEDc&7U+Md3cm<#MyAbQbWatif*?
zeh0DHJY<{R7~WgAH_I@r%3mT9;s=%$ixyI~qDRAWJ7D<%w+Mb$&_(6VZB&4=mb`IA
zMvk8s#+t+<P#Jd^z8x&|P>tYn7ZZ0bszOjlQZ(h`$6A$pf738V_1pr&L<<^U%9MX)
zxR0j2oL~Q6{bG|C)KSWayRjgZFn^GIo=8aj3Y5KBAXc3GfA7)bnp;LEd<Q=F5@(u7
z^9tNjf=R54N@9M~F0p&03-{5UGaK2x$-A4mGOnp=r}^rM)>LfZwvM=4z~+f%G}E4y
zYkZA(Sk1T#mQT5N`4zHnuBP$0#F#D`FZrlfucXD9aY}w`NZHCvAR78GKiT5zSA;IN
znINwNwANkYb}3?jQ#JdMw|ZPdc(375n*z=KUeFCy#6Oq&QJ$rEQ@~L+12sj=Lkw7m
z;o2T`1hne#b=LwZpsk$qYk!Qt{0z`D+mgDf54it*ja_;!>k+6yfHSE{2>SPBB`+vM
zao(q{Z_)(s0FAd~354`k0s{j!w2DR^3?5|3Tchd=dqZC!2KUR|bzpx~SUQ`Ta6*6Y
z?ie!8;lEbp<T&45bnh<c)>d!!jp^cK(W{3vOMjh?BQ8^6-g0C`vkBl@92-E7YYyZ7
zVFikiQyCXccIY>v0|ts~T|q&yVl!od4l?&%&U`BE9HfeDvjo9Txb$1J`)GN~Pl4&2
z^ZN^&qz|TRSHtyQYtxYg8LZO}XyULDLUE;HiGTfMaHi{~tN&KcW=wkKHoi5}Dfrae
zWN{DwIg@cE%+=uO-3gsl{m#GY_wIJF*-0w>yXoD(58yvEAC4Q=Srbk{a~nIWg!`e6
zr#`ddm+gK?G5&DxgO*6~^TniE6!4ik0L4{6n**CWcz5^sTl(sNsHh(Zhaw&6`iAx~
z&6BG+6TZ~~z@W`gOC>0Eh}$yoM`yob1pzQDuta@l#o9eUYZGa`1ofz4<z*#UQci=S
z+&Jn&3lqWr-Pb+@u1w1hJAfk;27!85_8<rpH-Xq9-OEDHA{h$HPjkm5onLGfyD5us
z0qc8gNO@NC!wW!9u+sx0?^~c)<*hEv9Y_X4>J>vw+CB!aO&t8#Y2)eg0B`!o)dU96
zAB}tcz#<i2Th5bBsk0T<Mj!&)q~5_l@dX;;7TF$9w9ImzoZ~Mxdo&)>&4HkXpSl!W
zco>Uj1?zyK#R0_XN$e>wK^A&oI((zc?4Ih>$yxt|Ht_Ivud~*4IluRy6^mB6+b3Hg
zicoueAF!D&4R>a*RN_5zaOI>VN_N-Sg+V#C&%S14FkKA13JaD*-gc*jl7<TX2Z8D3
zk$+}vOMk)D8CK|4Z{l?f3+Ue3*R8_hwX_RRD%F~7Kr(&M6&T!nCFCh=@_kMBqc7OK
zF`)`PzE4q&H#4^Z_<#*qli2^#Yddg9wvVYo>~sv=v}!Ya>%mP6T62uJ3YCAoxd~GB
z`FY5Lp(PH!%QS9nGN}HkPhN>b%je7ID9~!A27z_RxY`X2-tgS$y)bMIbj?}ut2i6V
zD#@%=Bl%N(cI%j~bXm<ef&ko1Y}})ANUA=~bLms{RonfKpkaHe8p2&0=YK+=mtz|5
zd}_}?GpV{|WMwX78+`Re_I!3$njIw^wa_$DXHR^(shB{`64eBrn@j>`)e)>)?3vr4
zd#`)6qt+*ddawF!bnfp*@6!QFuGzFZnyFK;rphODav|UbrItay7v40-xpuJdzAX0@
zBSZN^A6Joi*Vr8ZF?de*{PMwN5dUpyP%5eZ8~koc-{sZ1&S+Cs#*3y%XsC6dxARV(
zLT|&o3RSlyK*Z&l$O!C})z*S2@F_xdK286_$cr!DmFu&1Z5qdy;fvNMfRL3zcag2n
zuzd0-km#eofY*nUAoox1L8o8kDGjPJr>b-s@{QF$`D|Q{?y9X3(1|@hqNMTu!(KE=
zdS=uS<vv$NRoXxt?fe3fAw|DRkpS2{p-k&pzVIdQYy{%vnV6GZ{IxT_YN3CCxZ5jy
zbiIq6l|Zn;7dhJLwmjPK%Fm601CZ6!j>5BmFj?td8aHZG_qrz^;HT|;HDnW5nA(KD
z%=lngJuA{%LsJzTp{X`Cd@_=U+O~w?`GtbCfF8gx*c~>?j|x7%IwOqvi7+x37VhKN
z<7j|9R|p8JRe3z}Y?1i*)aV+S=qc8q=V22O7!sXm_H{Vp=4edS_L%NGM)>_HRqLCf
z>bc0jhtI{ApGQC%0LA$PB<KctJiPj2(6DX+UJqjI&b0@13t<WZ%3w@Ue-OI(B8EN7
zXvcoeFF^6YK3&YH`mH6rk0-#W5mv3Yrvay0&ivIXkpAVy@e`a}NxNKC#qZb6b=Nv0
z?SHZWF4qFpJ(}Jy8E*=FbVq6#!2<5+Ri{q9N~KVFy;0fKRYe+AxGLe+Xj7B%<_Ud_
z$#3nn8Y>a{gL9ZqJ?i_YrD_^||9rcX^M!WtS+fuNGKo`v!x%99=2%Pl>Fw~<9NMR=
zV*e$lQnmb@aE{k0ai*5wL~zjy4lAsaLjKU9K!Yz9Hylx!S8X^nDG*yA3U~jza`DnH
zeh53z%Sg7l<v;t*Q0XQ%$95703?^*8S3s;`K=r#8phN?Q`4|MSpw#P@jBALSYY*MR
z&+nrqwDoxPPXMtRYY~sm_iE`uyM(LuDInMJ$Eac_+`W%=s>@6mLlb!I6{2Pfm~uFL
zYP4N^Y%~XWO!|^t?*xZT7_nN=PRkm!lFgdw7L_HZ?5XzmqU8whWhsdD<7130UQab$
zlVjM1VuoVN(y_6T>@+v@gV)K@L@?+R8kLje_Buur&_ui^>02#_(fcb4_B=dUgWSvj
zj%A+U-mKKe8OPDU9)>FmjW|mZ_9cPUI8Ce;N<L?0!^g{S{k)io1nxE)(+&Z~mb5TW
za%xZ;U@`{ke&0UdH~nx0Ah`IlRBUdFcx>JWDXh387GK<te}ny!4WQ-7i=rm=959)-
z?qbstr-wV<>uTpfhpx2T(gbD5$W?q%r8W{8K}-zk?kCIbY}S6juMpzff{eru6MPXy
z^`#-|1u}?K1xGgdj24Mc`d*zwsZqXNu>l#Wm<S{sd_1b*(e8l+!ewCR!xEtQ#cwpb
z_A!4S2Ft^i0~nhhG50mVCGMUmFZlrdAvOs|U*a;Q6Z7@i0AoMbBW4KzO56(bpBo?r
zd{QN11x*MuzR8_Gp!?y$yks(<32$V*qq}-;#%lnv)YpIs<Cw|qmFtSrX3uR}Iz87s
zGnuhkYN9t1A`4|B>Lv<x=D-`LgB4_uyqj`S_Phb6(kGj=S6=}krA=z+TLcTSys^?e
zp=CPWP27)v3TP~sR8=X!1F8~RGhb(4=lQV>37EXN?EdrQn6ed_#Ue>OrfI+C*h1rl
zTfPS<?UYDbw@rY2oy(0-fM`?)p6NWH-}$>kMF|gO+TaG8B%e<6O<6o%G9=Tnr1p+g
z6vxf^r=2U*PQqaW1IjN0scP^7!H-LG%T^nXB(8w~_{B*Z!!`f2*o)5<k^Ct#0}XPI
zTkhPhYU#0_6|jHj`yNR0>kCDWR&m$iX(x&cgmpO7&|pd7X3pni;3+Njg)klsaqfxs
za>k}tRaYro<smxW9o(W*LYb;az@YFo>iX@wx6_%g-lKnTS$Y9^?|cq0al)at-x2F;
z>Qm|}BuU%gio?NP3d^EHVPq{2C3E)idz?hD<JXf|e*!;)GiL3A(~%bw-XHk2j<e^5
zd61#9(inqPH%Dfk9@%`)^zPu*BAq5x{gJVrdQo%b^Lu*&(q@YvxKOK94P>Zm_pPr~
zSnw6h3)~g#3-=>n?~5Q9ek7<UKQC*Pna+)S_aL0;{yNJNhgpLWa-m<`Azyz&N?zQL
zWi-ZL%#D1nykST~VFXx*GfUmT-XB}$3M?Neq-a-Y_u%EcO7nYsNV*hg+~YrVfgV(}
z%IoOuS8ZoFcDyi09F^p4^OV^bwyMMfbgo&F1p&_h=Ctp7o<qz5hblU*S_sUd6Hcqk
z?};1XroTS_{CbMaGE4n<@XOn@b0JbeszIyQYQmE^ve1i}A(88Hr?Dx=Q$#hsG!%3X
z3v&B|1+wEu44eYbUj(GUBb~CYx-uiISsHUCbEK9fQ;4WjoCa|3YDNiE#g<;Yy^PID
z6^VT=X`op>1mftl`n-C5wQ_Ik==tM9F$Z;4dI4I$%zvtAGJU@$3It{1EE61UZKw)#
zD&K@5DD@rRmjE;yoxuZ9`LCT-pXanM`*s>q#{M79-aD-6tb6+vL<oWs1Sv{Ks&qtp
zkt))pgOpH=Nbgl3AiaYiMd^r0FVaDJ2SGsTO?nj)dW&ag#(Cc7ch37>XU_S@Yer^5
zlJDMYuf5iNe^zOZt!ZRhdAcG?+JgZ&?_8)ua=B4go2QxyR|J}O$aH2KWI;D~HF+By
zr&VW%zT~5De9qj~Fmd&^{>`&uLF4~b3crl?Bhq<}u82MD&{<7<z?boSz+1Y?K-t@b
z(HTo>m89n9vzjZ3r{e-uTi#dhr5;C{t)`&|L={rpFZeU4db~{pF%xbB<|Q1HPoF!i
zrnL;<=~y3kQ`5@kp9a7JZ!~*Li5Y9uyT<9RQm;n7^4$xdeJ-I+d~|Ab;mm8EPm45Z
zFkSG?VC`?rxzn~C|2f6LviG7eX+RX2;_k~&j)HTs_swLRRfE{mt>89!hGVh%I%R-`
zo<GB;-*B7kS2q#u@>k3R%>g;;6oOlA&O&NaCm)`WBMQ)T_C=t8k)n3;+%A_+Y&r?q
zKeMSH8lqNjVO>-%zZ!p*(mWt=DMWtwT!M1?`{6QqM)H978@hePPXjP2{){U`h}8Yw
z*8|Pr*TrV)pXO&s2&JDcJFH6g#wwUSJGp{bz4|*Pf`~7}ug{QF(@PRhu^~Cen@9Za
zLmRzSvKMVWnEAN;8H5DN7uPsW)|BqJA@9qH6$ED~`AuD5$@Jx@C4g_@&1t_y4>`|D
z4BtBT!~pC-X3n8ien;`c)r~bBDt_|A7j{PX5|XB_!FCs7)G~v;g1y^zRfT99Z#W_!
z&AvNEcbctYw>+($nbLafj50`ZCe{jN*--O&JjononE9^m3gfaPIWqXHCx~LU@e=?N
zy)xaw?uFu!TLw98E^fCZJN3~X0!{2yvJ`|*$|;>&je~=qP}jxRMASEBN+uuA;0Pvu
z@@LNzcFd!NnFIa3un|T)RDG<bnd$7#hD&Av19;rhm-u&o<wj~C>dPLSJTh8>Nd=c_
zjDK_k%r~~_3{uDXy*2`ww8xh_*;6fX2Xtz&_J$pIQC#pzmb?b>H)oLtdtsXcEGyQY
z;T4b3dtv>NM_-G@605(mO+CWa7N9v9wwqfurm|ErP=a;INmseWI^87L^rFoqIGoaT
zce#Y+rn(>IBYwXoxg7AW%gtD2`2LONMAm@wwcQ5icWbzh>~AZveXOBa<^!gD#A9>x
zOHIR;x2GmA*Ym!M4eV&C#E^P6JfT>(Cb5{}>?<W$laV;^KYYfq(LPEOC&|`cE~)JE
z&K-l-xkP&>zndNE9>(jGL8OTEaw5m3mL`HvLI|=XrRTpkf80HSoXv1O!S}7L<*c0F
zTk;+0W0B&od41<hjLT)$O%ktQj3oc!WzCI<I47kWS$27Y&JkUdm)<Cq=Z_?(B{?8I
z*U5i%Onc2l?|_+iSa%4>wKllB7qoYdb@Q#wC?@JD7@UWtyF_pLzM(Zkyq3W6@g<*H
zwcn^75{*Z2Xo#vJz6WV%(K!3tuD+iACfVC8_?=6&QnN~J*lr#;z=Le)OAg4_Sz?e#
zvQ4MCuyyD5Z^a`)^>@j*R{LhWoH<wX&o2dHn&&9r!@sj^@o(N*>}d#4!?mbn8Fow;
zNqIQf`8!2znVEJebSRs8`OA=`4rJ*{dDD{5TZte8ikm-8Z1M)QZ52R-WsRvkJW%`D
z>4DcAfmp7pa37|~1M!tm;RoL{0h+)ZtMpLE{QBJVuX#%o-k1P8Rm@*lv4kc)wA<EH
z7u0OROikl@6tT4QfyBHTO|O`=ALYAuGa$p<Ir9A;CcJ8#%TQ?~IA7#1Y8;#wcI0eQ
z7#5Tl7F-S#AVV$nT<5Z_#LR@wq|eoJhtk=_Kh0mX2MkDZ5bWM;)ZL$xnjh~(ryx=(
zx`<Oq_K1)SO&_MUET0i--M@tYirRT>GxR=XK5|LiPM9|`gLfQkF`gl*jFjqNhi?im
zY{;7iTAdo_QuRb#C4Z{DWhofwJ$!q1oyN)Th79ySO@uRUEM_l?P=|=8Rgb@^yNr%J
zdZ{70E=suEAXgn65T;DFkYQj2VqNdI)4Z!x)YQVb&r-a%dIdxuzd7!1j>hLE2N=yG
zgBvoW6!eMG9~ANgcPSR+rSQ7>x<4>tGsq^vv*no8JbD9p5nmlRUa6{eQv_PHrS(dT
z6s?DnZ-hQA)0CEi-C<LHa;A}!f`~F8l<vBcY7mIeaD$m;>&8c7Ut_cI=-%hd)eAjJ
z+Wa>kk&ryRpx=C+&Giv0U^{{O_noo`UEnkqw`Czj&aqDmm50$07{jG+)U?E7io&L5
z*CgcDbZ)=)AoYJH9x@SuW#92I&8a+9o}O-n3Jag)r(H_UgXWI1rxpzCnI-g>5%Kqz
z@Aq+WFDpCIn{n0yU(H9nc)bor?!o9G#wZVoMRVr*caFt39lmir4Su|9mY!)d1ImX#
z8@Y?0oq5!tZwFq3VDHEK8!C&pw}c^1DK7T7FO)9VJ6d(NeA{B{FD(U(Lpd#9&6+)f
z4wLOOqt$8ae~h$zX2l5<gkLEaCJ{QV77Y$MyFtwE87^kLYIbT>LtS(yrh><@=tpDJ
zr=-R&pq!AXD%*aG_r1XotIzXrC*w^=NztB6ZYpAxC8!9Go+l`IJh#&qISBbA=$PQR
z`Fo-yH~{ur1`-<k`a*X*DR$fv2Bp#vOspf_P#3VzgIv9&{2eawepd1g>(*Uj_bE2m
zEeYqX#@VMIUbdssKW-ndUE03LI16tK+Dk#}mJdewXv;o^%&ga}Qt#aIv6C<}VPIQr
z_qahjNq5~*$%sB9@lXjw5wZJ0w7dh3(^hwJmHLPl?+LYX?Y_+LETx~Pug`5BkF?I6
zQ&E?Ahp(Ae8g}^^<YnFz(;LTmW{UNr%^ZdC8;sxZWI5)0R)YML2Zwi;0%}0Jy&yuW
zF^*PyP<uyHYv(Tgi=1?6o|Ih{O-bie&fQ9(O}d&c+B2t4p>FRV?=R7$r{Q9*padJt
zi-WV}eL%7l{q&7U0Fh{ZNVuz0oZ{R0Wu-IH#{S#Xr_}z`yN=!*1}^zi!v#Lq+vL#9
zw5NPsohC;o4Hz~NC7@lgduA(4+~#_jOTEI|5g<1Ft3lXTt5;WhOF13(Be%f)^wZ5m
zIeHTb??@k4X-G0co~4W8S+6cgpA@wR_YJ2fOQ$T<ZpV<WG(Ks1GED1-_~D_IR`ZxL
z#_gPRXSu!NmB!)Zw5=qRL8FvMBKV~A<W~y?ZWjfG`&!y(P3ab}_n*%^Nz-pknlEmA
z-ym?v!@N2+YE6~0YgUwW>@c;yeq!?eK|7BWe+hVD&YVED&QwF#r_y&nMPp0jpbm6-
zC5y~|i3@lZ@V})jTT|ikWP`%*=<9FzyjR+)*1actBX3$VlC$7!T#O2Hs`d%g*Lak?
zjdp6Rd9ucH7B^pgCs-)6AIJSm>lQ%xvnqdE%^*2WiR_qqHks+t7_5jR^4sMof?F5)
zj@{rwVSTxO<vTO;=4lA)a8hrvd;K)>1xzG)=7m!tw~c33@mdco6q~}kQ1CRPIjNQx
zzY~Y~7)2#SR(!Iu+q=t8*}Z&Qhif0sh*0ruCPBUwAUxgEnvA*X2l_0&M#D^KMjg52
zFoEvuoF>yZlvtK2K_W$DbQK5i6OlJH<{oCk@R2}^IBw#7I+6RTzTK}DkM`P$hGt|D
zJqk#@$;1<N)iC?H+&&Usac}*e;5NDR5I@}e_FrpX^uIia@W+BVHQ0HYIK@BW-3-qT
zAda_bO0}4+e(m+(nIx$6J}%p`7izoX@5~qyB*ihdiWgg(eqQ?<$O4Mq(-?-Z`ju*x
z%j=H~cA|dRiRyvFq<o<2T}}Q{7-JhbjHXG8BJA;}foz)sCu6c0v|D6pvZck+J=e+)
zJy*)Pt!bw0QG_?|7lxMfQ5{l{dG*N-8OB_BzxFMJvtDaMoyk7$`c=pF*PINDMfOR;
zXKe>S^mE$?knq=GU<&u}Zn`g{an5gD9o+q6hINy+)})jn+M7D5<?;qcKUL7u5|91e
zQ?y&z4`sdOuS;gb9H%$Wl}NG8<jT`?niRwoMg3QnUl@KqwCh6+E7yv6s%&RyZ*Z@*
zk6Sf2YjSMR@=wanQcE^v6Gh1l%18S$m>Cuo#SJBQw4LcrzLH`@Oo{7Osi{}llU`#B
zUEGiMN_|*3K^~Eqv_D=!yRU73c&@p6Uv?~%9vX*Q>rc5BF<NO5Dd5`2;GMn!7?%`N
z*NSfKqjk53_)LAk(CUQfpbu6Gg$HbAhCPy}5;|gY1J(bvAt~o}j7M!m?_8Fm>}W1O
zy(ujiOW`HduFl^y!pCZU5;SLrfXIQbnF;1-|3l(cfyW2dwVlDV58FgqVG2JyBNCA7
z);kJsC0;d{Hx$_f@rDVNXzGk5H=A{`opGi?u|F!0nw=PcNlC{wC_7bHvbImSGYQ2{
zlSYnt3%D^s`qx3rsyi|4@X8*)`JKVu!=SSQ4vBz`2_e}r{`aZ>X_rYdEhL@~hz6`G
zIR#Ukym`6`vwE-gE`>V?E)#N7fHqbDwmN=M=<00k>T148QdSKbUhm%LKdf0Kz4-LX
zk%-0HY4`%(=3UWx5tJQL*en9d-v;jy@aJ}HCHY&g>OcM9ImwJa)qt;()emqcBzWz@
z8m3J{s9((G@4BT|I?yMa@=4#AJVjNm#yhTZYplk{&tBv^wczazz(lrz>!AubE<R@I
z&V><~?kQ#lk9WpPq`;d96YAt@JwCZ8p`Dc;t?Dph4~;}LNvx9<Yy(XZf%2@hc8!t`
zbbjV_T3sjaK6jkefZ+hV$tL|2m9*Nwk&IB8*qK|zBDZ{Jz{UeJ{+utC!bii)Q5;M^
zr|YV8xnL%|ALg*H7#}=WcUaf*{aVt)-h1&bbz|2AKo$CZkXRWgT(}D>%t}Ag@HO(7
zSuyOREl;f%8Z=}j9vQ*&`g%?BMnARb-pW8E-y)f)7so5RtUhngcd@%OBATJn1!Bab
z?Zx%i(xXmlAw#~%2w1X#GGCtw<G2lFl<x^$P!uJ9*UXfz*mv8lF(v6!G0+!oNZnr{
z7)TP}nK(y97!9r<YCXnh!VBu4-@GeW5e4Sw#bcK^_gv3Mt(I9cxI*y}T+#Wj8i=TZ
z9?o{66PpL|Rt8uOWzu)@Rr8ATfP6JO_Tg#BKQZ96Z21_B?<90SS$6rf>pdb;`Sp9V
zgL7Qz=?F7DU^bD?Zobx95u-(;>vtInpYC3)3O3_kDxAqTr54v6YQF+~0NXI8M2PJs
z(BIe{NLs!(M?9dl*t~wG!qwMDTe5Th;Y%DDe8;ok;M1EL%^-isD7RaykBQvuR$I{>
zyv-$yddYjkBtY2bw6Sn1c;(Ve65iar?ylB+8E|z!Mozj}awyY-cKoT^hUh4w{WP=g
z6qS%Q7p%#saJrV(#pxB&<cOXEapQm`h<3QN*0$;fAgjGS7M$AliuRS{xNK(<1^xOh
zBOTE>n=kJxyuZ`TDT$UNbgG)E0_LrnC7G1w$|(pOi2drWd}AdL>braCLf-zK;4BjD
zE3P0poTMvI(?@w0A%SmaXIYbf{K?C=+NsehBim{|)HlW8@xz)k+Ra!Bw>e^yC0dKX
zdR7{XNI#U4=RiYpsaN5A1|N)=d(-6i(-v>Dp^F^){ayoSWYH-eHht;x(p;F&Cg1Oc
zZ-6<^?u=!&z7(fGxr<WV`$n5pnzYN(?eHha=v^?1EcuK{R+oXEA}fuZ)ilOv9r-a_
z;jP#(>E1(bt7__~IbBb|n7@zy0g7PuX{|%9UG~VKdO4|mrfH>s?=Nj?wu5!RtmNS^
z-e4V+wtbBx&tvUDWMO#WKN-NR_uC4|r=Z}n#;MVj!(B<6f4037y>o`QXR16Pd9cp+
zP@Tix4k>(4We9m`*ID9GKeC$ulK;o|U0)!LeQPw1`C)CZ$er~)yA2s_iFcES)meC1
z)}>o`5O-q6r>ikB=|M34-x+|JH|~}Ws8hc~h5y~bZ}FpVq5EXe3%NS%@{sc2pj4SK
z4ws}7eeKoBIjHbV<d`t3LimymNUu~~F9Hd`LBs5zX=8J(tpK4mp35Zxt@GBsV7-Dk
z8NZ(}(jT(6Q6;vAGl%0ad%I!YMwb(mbMf5V-WfRUXP&&aE$<|I^r?Hno3T40#sG4i
zq!E-!?30vy6iLl_-Y71-)p^$Ijh{{D1!ssLhvUdAcO>BB)q5Nq5-~496=)C_NO)cY
zwbJdF#%>Tf5cY2`g|4h8g6TXow(CDYXU4V(M+b_}>zVG}l%SVv=vbKaXfI^%_AzKb
zhMWc0hqx$lI(!YGPt?TeS|(Twl=9=deeKoh*8chTnB}`QR&OK?>(va2x6b(RtIl||
z(?d5NdAe@d<POE)nTz8(*L2(>O0s*;nS)$qzj{^qMg7V!9d7nq;TmfpWb__Rk-~c6
z5~MbSkn$Gu`@1lGUoyx0r}(We)y1?LL09e<#B67=<KZ|60VU8fE3t6XsAdoXW)!A>
z_WK9(DL1mpTztHlnXxpcBJ<@@@;;Zg@X+}r?3g%RV`!9_?h10ngm5^6n}AZ5sj@!-
zfl4>PtpMZ)QNjeWB?CD~GP^j2iC`LPID_(|_E_*=O`OF3>@icU7$kB1wbxW(HrVF3
zFO?>DxzA=t?K`X?Z)QgH)RJ(KQ(brM4c|LtQZb%n`aEaIW~l+~uaKigA-W--oiUPA
zi;sH%vSD-l{xTWx1j1$*Oy=I(ogdT@`|-wY=*-rqanro4o>MWB2>(g(zS;bDkrhHX
zxbBB9f;};T`0QoVJ(rhTI);v&bb8K8gFnASG-0=GzX<eew62pI-y_CuWACEgv$v@&
z@Ch&Q$<cb9@b?C0hpmjJ9^acY_4ZtFsGU?CXr$FXI;9vW#tJ4P@gR~*=<O~&K%SKO
z#r%G09kKfX<m&7`;S{^<uZx=iz8?NSTg+t24m`r)Fx$^NFTLH~y6>+mfJqotgOR46
zDpa(0pFmEu*$>>#Wmq*5_t`xoUKJipYIZWp3ER8>x(7#OJ`r0d(oO+6IT1?EzM7Wp
ze0%3QmS?HF(2a5VXrv?Tl5L>y0~z*5)lqM;XSNah#})H^4X09|b^k7Yi{Mb3R|Rj@
zo)+>3h)w^N0j_uBa5Z0`%9h^}CF3}c%%$M>SJk?fl;M0dsJ!0aY_=-Obxn)Aey;Au
zi+tW*qaRgfBBBz7M$W@X^kVf7o(?8s!z=0s1xN4U$v(F}9Emo625tNocDW@H`MVdV
zwYww^{`?<*Xg1VFW~fblI1=G#-8B0`a|u@@M$t1L>3fx@NG6_t6-^X0uVW|of_L<?
z6t%r+;aFRMZ3pEp>?N($LU|C@vp$GZ0;S%uFPRPua$?twUla#tc_12^m)_9Xu~jH=
z&%BdccxFGzjY?E2c#hW4+mG;Wb^7)}G0%u2&n|_JL-G)``Nh7w+C9p5!T;p#p|=^8
z8}+_79{J8&SmiB?nOAKxGi~FD5BWl1ZcVrnL4wiDpavGl=~#~nPb-eTzK}xHo!cmU
zC0ju!G~A+ARq3i=c0$lf@#X#DE^+~(Ei!HXkv`5mU>_{h<RDU<)z>jL$6SnE`swuF
zdVDAFF7y=LCGw%2q9Dh9A)%it4usJuM25i+gL7~CEC?bz^<LCj-5u*}_eBTzMn#nI
z)woe5E}AR0tQ!WVsepOgSw?P3&rRzVJh+<SmL@f%sgmE}8StL1s<r&hS$0oRZ3?pw
zz0L=YhIEufK`R9_+3F@nrA?(%!9Qq@4pskT-I_-|qFl_D9&GSLUw2<NIEp}Nl-8jI
z>&9*r%PhOCUN{SyYr6WGnoA$ih3(<xwM349x%h~gbwPopcg60SNknb(ls&%%Naje#
zJJ9;!XuhA8Rr49)xd!9}mrms8!PRq|b-^p%o-x^op_3M_hV5I4^!6Q5Ou4L*2jnva
z8mrPH1CqD;3@?7HS3h4XIASX}iZa~3J;nUnAzLYx0+IFA38|m7WY^l@Qn!*^+IKc;
zB(_JJwtk$~r^{RPovG3NG$e;|Nswdo7g<sR=TDl_-S3Gauf<f*FD9)~Z03;VyHiOO
zysx!(UcQx(TULkTkN>mn7)IStvW+z)(=VonPi!`4+~&XL&;lOH4-x+0tiFfZUkY>m
z|H;NTK{k%)e1*B{hbx3T&Hb8D>_bbv{c{&iX7MV#d^~AkUvp_P*Lg!xmejT8;8hC2
zddEmh+T&4*IGyQ557<mK%I+8Tq(-8hso4IHl<Xwae!kX5H5Vt#1-E_KKH<8ir!ZG6
zNeE(sF7k6W5Sad@s=pe3h2*&lRZ5uy|7Z#~_nXsO)+q)PknY)f-PQgoINDV$2B|7c
zKk-CC7`N!ZPduOA3c8NHT$_kT`Il}=Q7MRAnh|Dho&uw%jv-h#YQht30Wkfx&!+FI
z&)jOtH&0N0DF#{F>ZFsGY%praLLtg*Y@EA?Ld4@<ZT@@+gS1E`;utfbXE$V!Xg5SR
zmQ+VURH}i&fVaLM-Amh!H0z7PRyq;0#h}KY{o&&(h4Mo-hlXIMDYDJ<Eyr+ck}>B%
zb7SoeQ@Ml4`&bXnMY%BNR%;-;tA}{ENl${NYDyhbzgLor@Epi>1+^&~bJ8AFzP^%`
z?vdL10LCDepox#gc!=V3rE#q>aAa-H49P)PAr@xhdE#4{H?jJ-H#A+!y{qoyfXb7Z
z$KtNeezgHdpNYm$9n%!gLU7Bc9u93`Zi)9IKZ^OUshc4UVdx#5=lnG%Z|G^93O{7U
zP_}Qu?C>1ve4b93`zF<~gX+cK7gw3*GQutqb6Jv)#bfs2W^FAZiyK}gR(Ev{1pnls
zmZCFWmcb4m(*s5U<o+&7N4<7iXO9miUuXJ0rn%|D33>}9q2;2z8Mdzb&-ImnBfme?
znsKfFC2nG19pyO4KI^^rT8ZMB>@3AEI00=jg<Eb)Irct`<lKVotRLqiII6BELUN!l
zc@-55bcc|{F$Q*}?jlrq8IRFz$mj(ffq?bzoGmYEfLu{%t{0(wPO5z_pO3@F3p><l
zmXvD>-xxR-G=I2U4%(}t-p_yP*zm(p5+MI;2(4ZcXRCqed+rvpFyz_yH?WzTH7PsP
zDCGTpkIgKARG$GvX5ug@^3yvI99_(&!3{RUXk}LH3>bp6i}}qd3T!lgG^j%+9@j?i
zmCq^6!5v>#A%bc1yKBMp74IL@J48QK)CZAy1^?TOvNNY<I_A62k((SWa(~bCcQUW}
z9Wws8Jj(rzKq!y)yF|Inw<4dn(K?%Ge+HZnh$Ji367KK1@C}jVXhn!nvW9`<?Y~Yi
ziddXvij>)6siL~ocda7wl@qJXyY2U`FG|72I0=)FaOhS7P3}{#P?TUOI$#=eSTkwK
z4<0S*Z=H+Kc%~2CK+!bGdA(^s>oJLO>PaC><))iC0iH>B1RkvQ{Ega;-RFGWNe8`m
z@Fh?*l_*r;dI&yt5to1ykI6^_>%A$@wKSQgVnJkg2N@a3g5FUTVv-+hpZ1snBx`U^
z8V~8^%QbhZEB91DG_0sMKB?$^6xmT77omgYv#}aUE*qEugjsejN)^uVm4ls{`J3_T
z9YbFS$J3~!UDlKgaMcNpA2{h{fwh9yVDyGTt>d8O_4FO<qBx98Wj4J%Q{bv_=+22*
zHRbpd%Fbr?0QZ@glMtbkH{VPdV$Fwn<A(g+aJ%2_vxJdjf3!m|guJK;^)GW+QA!aY
z6*EEZhB8J&##B+h{Gp*t63lBbuZ`S`e6Dl7s~=K%!k(@6_30w(<r5tXyK+v7P8vXg
z_fQyCUzSB7{!E^Oq!ZLCc(-lXi1ZMQKFP@+n}5Klq&Z-7e2|Wio77WG6MEG-H}!l^
zS7d~n!Bx2-HAb=kf`w$1^V0e{&DI9>EIVYnA(OJ!e&b98DBkr>UgE{MiYH~?=s!=Z
zn9n$_Nx5Xjk=>3C`f1jfKU#L8@cT8<jqo-&Gegg-sKCIJM{~=RO@_~FOtgzx_IQG5
zJN>H8(AY|6?Rw58X2k=h#Bb~I3VSdB)k&t==qiofj~uk>q0spxjG#T()w9M=R|tA;
zk$={|1G*m-_lix&KXDI<GI(ZD(wq2d#5sj(>q#)TXwdCer@8$127A%mY}Le&wh?gl
z#v+SWarW+fu17}5TW)&lGXlNr?Sys_#wDsKiK9%FG+_G)peUn=24~U_iEBR>jp8e&
zikOnU4gYJ0p$s*5Yeu3`+XcI;Cx34+Phah%g1IPo<($Hlo?GV|%e)}JGz#;VOt8bC
zBNjQ~(s!k`%~l!W+F}|B`|PLea)>MG+KX{!b09tPYPei~o{{!|UF0kaK9+CU&#PD7
z%l!U3b;PpKyE?AF%0{=wNG+d5@;4jg^Y{O<UzzpJgkWj@DGwRpjq>7S;b~gkFZpvU
zC0?Oc!NGCq*=SMdzQRcsTvn^Db`fwTg8Ui!r0s@eSsPhQr>)~_E{On6GVDcHM}r~m
zM23PjXcSOYdQ9?mQ;{Tk^uj{kejrg5qBqx{t*F(yarLK}M&Joa^rA*pV30MCST{rE
z7xkI-o;%_2;x1AfN5+8JggnIU<UE*>oz*~yhlhnb1j0kC14GZ^sq%<PZY+0U_EL(T
z_9LEfZEWV_evVKRIOj+B3}0LypA^!4X7w^=$e)&hkh!Idti|v7oFh?y281{zv>pJE
zQF;8#9=!z1jK<M9OA+F@=w3WMYCjC9mZnWBkw#1~IKG!lxx9Pcahg?4CkpRp$(>+|
zW4CJM31YRx?*X?H(|YbUpZzGU4_yJh;tq`<y^H{7#%@`D9W2GGzx=JmhhH!vo(U_Q
zHGzvumD4lQGCoQ<VROm@;ILy@cJ2`LR6sWad78|Zo)J8~+Z>59eRx-)df8z*Pf@;r
zDC#Jc=sj5RJ~RNsBP~lka_pu~!iZ6`44+C+re^!zuV;rRpcI<)?7tjVoUUa^Zj+D3
zh-6q(P~|nTR@@V-q)nBviDrUw81kmbiHBXIMd2~rAHn&U=AgZWrufLdqjr$fM}|ZL
z3kd0y69@6_&aBW3F!}zhYyiiV*He6#m3H|y?p~svS=ctv39oA~6Z<YmlFgQl=;`d{
zzQil+3J}unosPqYCvoc$LvM42#nljRY5sy1M6;cugyn=|Oih&<<UBrwY{U;skoOy+
z&zdULU~i6+>obK?<S!NV^GX$GjDdlK>ruBE(P0mmVNZplzfFElSSC-}eYvd8aY{P9
zU$7C!V6S^ahMWf-a4-+N5;B*HUwT_Eg2;{a8jJdONkP`?i%62HIt}2vbKY^ln4BuQ
z*K)PJq81}h>d~#_3_XwN2i^o6HV(zdl&kH~GrZwAM43)7&&@4d<)g&6H3f$_Lxr0P
zZy!dADW;g|t8%Olm#(#!T{M{3+=Z@xV_7rG(`yP<_~9i8Kb-)!4rw2$v=&imw76fH
zTU+inw=QTb!xar*n)i?#&m#~jvPzUK?3^ovCEqs^Xi-ngV?rnbmkvnwrKKj8B(bGr
z>y;rRc`H*D-r54iAO|)>+vnNKQj>J!mQkhJH!O~{=ur%fb5<sa-G|Y(X4afv24-0`
zEi#f6U6M$i>Btd`9TYDFNhQ>OR9;XKDyZ*ybDUII+Eo8OG}VFk=i>}(uy59nYDsrq
zhD%(-%Hcrz-*FXM;yrDY^<6SbI+mH6k0|-OK201`;F;Z1Jv*;GHLSI}8u;Z&Q-KDM
zK5`;peO^iKP}z1pfwfs|g$s|hOzP*#k?wp`Z50oTp8V#Z%_}e`)JEO@8M=~F)0GC1
z33rN?)HVY3>mME;IMy(EHChUcYD=s6bcYC~$w9jVa{bj_QlBHzjYmG+LWDkIiR>>_
zc|=g+=K7JxD2d+2>&~&?-_InC<USrPaDDN%KBJsxE={k_51z0eeash_*ug6@b#Ko7
zbUo^bd$*v2Q7-+l0E<GZ*XO1_e(HhqTi7>Ozw6HYZ7jeg^q&XC$*ld*D^abr>nHD=
zW*ob3tZnK;q<Q$k)Rqv*rZ-t6eajwVWHnJuC#Kl^#^g1J7mHgqZQ85rQ1jMUa)Od}
zzFXY%kZ1@ANszJ-`L$ao+DwGX6MKu@A0S8t)<-O9kFwVxce^2X$DtUhIB$B1t*7OJ
z1czxk75`Ob<u^Q&VQ`WWEP@*+tpwUEROmUdgCFroi^Ot!P(DiTS->;3aJo$(PB_8u
zkl`NMsAq?XTnso1O(&vury8FI$~k&5H>l)tF(M@`Mh7#VyK%Q+3k)u=wJTYnm!{I!
zUqFm|2r$Nz<Qdrqoj$3I(GK^b19OvUlzovi6sm1Qfw~veH77&l?T3bcslC6H9lBg#
z3gbv7+P3TWoq1Jyh7x!?k{#fg2`C+J)8b>oB5Xd7@ScdUPE}FnL&YII)d-;GHR8w!
z3_D<IUV~NQGoIchhQ@O(GvPjXIOnmnr!hz%dK{-lu`K1Gy;`u0e-Uw2l_R(F9>d{%
z1_*NiE6Il&I@x7YDW#+4dIgE<xOYNX;%k)?%447pcA?LmQy&CdQ!16@ZMVgH%htU>
z#VDG0+~tsYtmmnk7wD(GiT3s`td#Ub6l}M@QG7ln!i=<(fnHa)7?F>YJdwFjZqRo5
zzQPk@=CbLC`~K4w+tRZ2QAB*rNqhx$5h%ELit=TBp_lAJgWh|W++bxu!v4g=aT3T+
zwI5D2?QKWE89)RN^GcT7<i$mw1(-u)dHB_J#w21bgkvaTTell?eDY|rR9~2+_yjHu
z55wp);#EP`k(bW>xnRc5>%+kGPB48}O>O6E23L+t3JOv@zlibu1~4cldmctsq7yUx
z(L_J|47-2uyEbqQzBKXAjEYe_Icwq=>a-ZnlaH02B|A?!QM@ZlnUk8~*hOh1;T?Kh
zp`xy)g;5L=wnBCl64?4^u&K{FLmfPJTW&GLKPX#}xz|UQouYcc#xSb25>WX;Vj-rH
z_u*!r*5wM2*A#GI6r4fs5BqM}(yT<Fxef`tJuIY6IvR*QKTUKm1>i)FB7l5Q(SGid
z!aNh5&Hc5i^SCUTyTA@~u{_)wd>AEaLwBh<h<`e|s6!s>M-=ipL?1z!mwy6>A;0n5
z)+L%E)@}c`E2m>PZTiRyB+c93-(Iim^umxic$=w<3io-bw5uEY-yB1|z)~^7h!@wl
zh;ipE%S{ZAh6#pBMxsA%YeLCnKkCsjantK%N6CLvfu&1GE@>zaCe{6QXRmxBwgB?o
z1f=Z>h2tka-$c}Gssot4y)%|lyjS3E3nG*SA+Ym)SNyJf5JoYd6>ZY^x?Fz_I2ik<
z%G}N^I%~Bl%|E>>g(%s_h{|Dvg9mwi)m9m47P?^yeVB$DG3aYwoq5lWsd0nh=;!A^
z=Te2yx%6m8^7G-fpWnOW*i@NgT(?Ks4c1Ct-j=A$nJW?`$C-9BdfS+iJ!n(&%fc|D
ziEmf+AZfDsboIS81;tXx-OC1{aR9?TNU_XVe4?+t6vv?UAvr!GZujdMd$4JleK(dL
zQAtnOKjz~Db8HESHCH%q#=0QPtAvas;~rne1P^&$idy44qNDM_PHOV;2@qn45o5Dw
zbcLSX%?FJ{q3@a-likgfhSrGA7RqkieK^o)o+@cLeOwwo5Jb3&O|j2yG9Ws<N}{(K
zPqwNPJs|pZmE={hq^B+u;hy7t!Btq)SR+U9C`{uk%eR4M=3=9z#YkxSlz*xFy0NFi
zcq0kz#X+mpKGxjc+Uq;}SZve4*>FQrcpQH%9=RHyca+q8(P?P&9OGIv^=F})z4Pvr
z&t)~6(fR%Zqt{X*Z!%^MUL`FZ`krJ~oyz*wj5_T3n>zR3TTQ(66}CQ5?3GOP@_n;j
z@Wp1l61}*Obk_7`88ekcxury`y~HQlU-TDtMy!FeEc#ZbCCN1CMKebENFKchP9Lhc
zZLPSpSiclK5z`3DdSIeL^?|U?Watxt1J~g7bMzZ434Kf8)#0Pv7C@8>l1Bvdsh(PO
z#pQX^{V*(hAu(+Htogx`+UN17=`-(O#i?ia;Gf7BFD2@1?XGbqeDN3vDqLOEbjDH=
zBdzY3EOwm$DeOx%@xui55DhY)r|sTS?cQY*-uU9qKx9v{ZS-peG+tUS(Q>(`fo2}n
zM9^ddlLx=8xW7}Ls9dfG;K=)^6N(<|a^^HGm)BjnR%%tE7tQuNFn(`gvwWFKEPp@r
z>LKeP;Yx2OB|X=6LlDb8lWm<?exb%O`PC7-b4{>l?NQja6-0MZ2#z98;_zPL`8f1=
zOAUIP@sx4CQXQD)G7}&d3h_!6Nk5;xW%8+zZC7Grc;)E(j!6T!zSo$4n5lU3aypdd
z5ZjVKN$(r(a8f4W*}djEDbBeuIPfW~i3e~WelVcF;e%D9W6tF;1RJq9?hRRH=@YoD
zF>CH@UM2FmAyf~q=0etjyHsRlkLZ*SIopq$z@(+J8qilY{~;x!PWmnk=(7GcuGmqa
zkUhI34{NYrN=!@_6J<1))YA5QivE?W@rvzS94TJgC*qZE_(i4`B{>BJZwj?WLY&1}
zw1!Y7LQMcivhA|mSz3Z4t&QOCTKWUQ-Uc|AoVfZAU{&E+RGmbI1wsYpur}~U^Rj&M
z((>kS$ZU#|_azhbT^(nL&^okJ23z|wFrwa1g5pw#JUG}T_xG@6QbBWS!946O6TDY4
zmA|Hz28GqUmTVsVl<Oj=p^T|Cl1i8_lk*5^MzeTt_u6tsV-DgcZTt#R(1n8WR-ko_
z8gcurtpS}g`2^dkvULx4%t*#|wZG<PK#z=Kt<hO#6Bx^BI|Z)fMNZYz?(+HROW!s>
zYeCqsPMwrw<X7*N_@MegiAPlNrZfIXx-P-&92#TuD1pVHdj=>zpXXg4KOOfLe$QDZ
zCO-~z7o4-CTw4$?NssW%{^9ljp7FX2IVbD$Rg`OYEHJ`hRB}c9J^I$AZ^k{Ys3)*6
zg#x}QE3Yd|+Ucy<W|mtEc3y6}?`xeFfI^C*)(vhY{aqAYjosZ%tGQ+hwddRM61XdU
zoyec!DUKF~jvVE)X#KsSz0z6U;DXa|lY_TMA!U^YK$LJ@w`z!{{hOVbM$$eA`pB0x
zk}Y4@mek=GabRsT=nMvCJ;@M<d+pK4z!D{z?(RXBMjK;zb(CRA?)Yc-8wG!UQt&K&
zyiu_HmDKe`gRA5YmrZg_3-61rbCRW7Lcc><2SSGJ=YSQ`dY{malY2m)mt~c<Jn$TF
z)V0$;iLj6}pbwkp3QAF?6#tJz4-40D!-cQah;)KCOJ1RV?jFS=CGLY)PNAlYQufGa
z7cptGi*<|k%pO}oOve4s>Ci}x$7?BTsF36jnx=jvX*=tEZM_5tW38?a?kSI1&eZKA
zSmViYa6gLdEc8S$Z75$r1KNm9&BPM}^C!JqtCs7slVg{L<C+)lo;q@#hDo+zZ+S)U
z?T63?rdhP6ojWYQT@T-Q)D(FtX{=ezskBwGH5I0?$&(ONI9$HCujt-Kx}lUtd>Tuh
zjKeWT8+9T7eMJd|m&HONZr;Aq&*quQYR9a4iqE?EB(>kLsHP|B;%BJj=^NV|6!E1s
zi<*PIuhdv^qTKGYDS!a_Zq7T1%<2?LKf|6I7BHhMH6<ALPI<BC*UJH$n)TjTxX4vJ
zG8cKcq|Xokl6I7rVo~FC0~^xz=uTX>nJC4z70)I(y6xVan#4~p6u;S~JX{Rcv{w^J
zZVI}{Hi7`<?zY_e8DA1nSv5S=^ZE#z=4-86zvr<1NMKoPdZ_o^Tsh?2+{F#0GvdQd
zKaVDk=9ZgBp1MWWeI;2Lp#b8WcJMf@W*T2c-%yCGU7j*X<uliqPeQL=hTv#$XT`Ts
z=kSg1qld-mLj+j+$7Iv77j<6eAMkwSgkotoEHt_{#=rL~oV?9W|Fo%~-0kJqPs^QH
z?)?lUA)RX3VE#LLA1TDa3ewxTQm^rjrRC46hv#adI}e=i$NW?W<C8&=Is9ZI+D;L#
zkHNsO;Sw0P53!Au`@P1s^R3s37zgQXC-ao!-M*I|WFX%=*P30X@3M5ccY9MTO~|12
zkm4K&u1*O!=aDT$)b!)rw$^;;+VGK_LVC=O8``j`WPtx`bKZelvm_W1Gu^9OMmINn
zd*4CdXw0gC?DChTQd5wqY4{ROVAO-~OQUM>I*L`D>EQ`Gh2IezLeUw;2TA7oxi@2w
zbkWQ97xv3%8kmynL}JZIslN&s9vAAYwg$NhFQnA9QP_92Ym|i5I28SGckYw@;eC5F
zzBZ1?{50Ivl+e1sw#lw@i2cxxkmL|&k&I%A*7o*F_@#H6kxjc<T{!x|RN@b;&Mi|#
zG&5EC<!csOxtz0tCru25KE;<|h6cs%5n|J)Drkn{lR^=RZ>nM<>Vwl?b>m>7m;S09
z0Tqw*nVL1nJDXVGRqCgoZpC<T4=E4lg`%rItNgHt0m+<iY*YC!7Q^zv4|93j%iS{(
zLt#T|<+d#Ls2OmT<Cy;1k)?M`ejc(se}bm2yS+=lyFZY;1_mv6Fv2OByf5Nt0^_`;
z(qFz%<?3RxdzI4^K18!^04h+gZa%&Ek5#;j-Cq~xBjBwPq#Z4~+vr>*VL(LUuZ2@q
z+n(gFM<)&_OH;QxK3-Qs9I$ly*&xDdxbbZis9m+nw(I)X3lEY^?x+4;&Z55eR1GrG
zs?fLPD|JPw_#keb+a!!tmG2-R^L{A9)1(%Y&ExT)Op?F<Yx6X}sf_YyLn6|@Tc-XJ
za#GSQ5^t%lNB*OpYLQ}UNqWT5>;i)S-TsP-)NjD_D)Ka^(Bp;!wjMSZwc2l8));PR
zje~pU#{qj_cJ^Bg_71%quXa-@?lUPi>}KCT*<hew-&HtH*_#2U<-g2bR}+nL%o)DT
zLkJIi5eOxsAYF`K8kYd8N-}^c_qN)fE-otz9leP#IrZN~aZG2|{|OGT($O|xvQl4@
zcXyqqlCdO#L)s=M>Y<e-e(gJjDPsR}<ty<Adgxa`foi4ZyHhp34US#`W*dtXI2p!~
zVmoD-SDZGkDvXs_C%T$!-o__h+K39r;6<`%GL41SGXrcse|-qXUc}J)08Eh0`nRkq
zc#8gDz>0AX<8oR~jKtJxs`H=(4Z$_Fzl`3Q@i=<Mn8B(Q;5A0&V`!+Snt_&zQ_Kmx
z+Zw>Z34|+&QsSP@*5G4g%cjEY99Ju55>g=t7T8y-=I1j;WlaIA0_xu?CABkT@)+K!
z{>MP>#SFfJp-7jYf1ZjTq*B=rjwl6<fhiwQcr=Axa(p=$!8rK)_;jKo-Ofa?iW<r<
zb^&8Wn}EIB0W$G1*({vSj4`HPfO!vqPbS&&Qq5rANVy$pibV`{sa;wF;gaEv3QKcn
z!+*5!_@tMXe}G3uPvV`B3V<HammYN2vJL#bZY&|+9$o;gN*O>~F;tSA+?ke|3IitD
zp)P9onFmDi+?Q{z&}QuM`g1^z`<FE{32<<Mei5p2`S&_oH^=#oJ}Adgcle`)&rE_m
zhD<0e9D-`wpOxsZ;ulsFBXRbLOp*CvxlL<FBD1iyUgfA7#N_KO0(*G$h1d@*&9`YI
z<=9w7^uV-wE`TMAK6oClGFDnV4=rtMaeR%>To%Us?h~w=EkgoB0RgZ5Nj1|LVFuJ(
zHS{vUS_nVyzt)q4v<cnu`8mYwn#|u%G7#*GTdX4Y{`4*bV*LXB0(gdR!~c1If4|Fr
z`KQ0X4A@rw@rmJK|LXz%e}F8t(A}e>U!YTVk<<UPApYYa_*eg6vH*>RIxKw1HDmbm
zf&PAs|MpKF2>xS;5QM}?{r?|~h(ZYG9uAd|Wq$JihY$VF6%za}gwlY|aI;NTH{m}K
z^Iv^#sXyNq?oBKZaaEtOGymI9{69Hx|J%p!?=QCiggljCUw>zC1we@j5c7=Bi>d$J
z67ly}!K{h2rtq}?S`#gFbPfOpGzr|iJ3!J^lb7K;)fGu;4*blsbfI<GtpD!$T>+o>
zEz$q@jxsYyPJo$w2lU~p@0T`?9DxDPuK-!e=4>NxcdcAw?_Ykk@0dU+`uxG=zu&bV
z<iHUdV@9+E0rM925z<9t?|=OAe;&mD{V(YRz#G*hd5icT&*hsEST5U4dtQHm%Kl%M
z%mE*GkypZ42OfXE`A^ad0FzJw9*kZhy!Cx?^;F-BG}p0rNwhu(acRK$dj#MW+*Ehn
zXB=&Td9*amVhtP(Ri6h%`kO9LN`N*2hM02JR01@_7%=%ie6?NP{dyh@ri{UmFRDkL
z+;^=LL@%6Sg5Sk58lNTg+o0rc@4Et42^d-He`7Q%*ghsR%;~tF1NnQ1M-+|tlriGc
z^}SU=zD-HpY6Xxv_UHg$1S5@j9T?a&f<CHgo&*>>(zb`E%kk4d8*Dp&`ODoFJf~`4
zUO&c2b0;fDRSh<^fA9c$&+l?T{FnhJ-%SiBFDlpIXG0G|#00o*kS96>ZN$>cJjobp
zE%0pwET1FbNEhJo04G`7NO`=<!<4VWYpL|VFy-|qVAakY`hBt#b2SJ2mgb)q;CQmY
zKn}}SX1l1P{$T;kz>|!~7Q4{5lsHTqcB%~?=dDkICZHu4#|^u%1V>F3sMT<QAqaaw
z?OKUJ_aq`P@WW}}^PMOj9oOCF-%V+NhGT@1SP5v<b3t4z(eBi(r592ggAq-(;2Gs#
zMvEr!kHoLIWB7*WU`N?t&Q8G3_!fMV`c7?^b;LC2rp?MnQ|XH;O8}AE?e<*-8qW@{
zy5(Dp&Fx{{r(bq`O5JRILr`9Tz7yPd33@LR%KTqR`^bDbff#Ho61tb4mwtU^AIKMh
zobMIqa;xxaa{I#>hxVrurWp3%8f~G^zud$Us+fE!UFv%JpH2LL!LJPBI13h-J^22P
zw}DCR^Gc#Yi5r2LP@pFuKIua_>E&PTe-*S|n|wqVD0O{!0<fU$=fR^7V8ws*6SRTI
z1GPX$*?!^|&_6ud`_Xyzw_^`|2#zTQ15;pdXa~SQA|?ScC5?_lx-4+Q(*GJ+Sz$?i
z9~eGZpH2Q#P`WUHNJ2=A)^xFrfSanMa~|_G2eaA&s!2KK8D(Ye0V{bmpb@Zy-X;D`
zpUCj*1PMSA6QF!<9hx;Ler#v}8gKQ1OmY;EU0@Wp1IO@7XdP$eH)(pB&~B(bL$It(
z(3Pl@ggUUa0L7phv;n6>XnkLX(h(OT4co4S*aKV_{cfxq{AL_Yn^G;k2V+cJ^o)`O
zojpAiC<)O3QN$VWiWC_q-yT9ll>I0$3$iC9+0XI$JB>Tok|8{eKx0oIgTXdFx`JtG
zj+I!t<6*HcHnQ%$2RnEN^!CeC9!`FEpVYA1)G+JR&zWTd)sc1mpfaZ(?0YtvF;!eM
zX9dVkzO45hg7~qUK)S=XTnAcQ)mnz0V8q$fZCyj+#(PZ=79ybqzm`z%(3}g5axJR}
za1ffW!v#}M&gQ{%URu;`#~g144yNW?3L_yE{KK*&0)}@^Fs%lvPuu{yA<t;fu)18A
zfk1mgy)=H=0n`sDkZ}LIK>e*7oU9YwB#&GFh!*|jaiEr_!vtHdl4rv)7_j7ZvfV{-
zy@YeH+bMAz^;*sWU-NE<1X9cZN7XQZ2jAKPAV%^*erx<lHYn&}pnT|A*=pD4GE{Ia
z*DjRY%#-N8mEHrHZ!RE;1#dxB=oe&gf*w0y2wL%K62gdAwzNnRgycyp5dfB)r_j$y
zCkIJwfF%Uc83^sRDl8@|sxVC}S?8)>&)%a7v<#hrVlCpeN#oc;QZd4QoZn^gA%|6s
zO2oJqLfd8SHlRnx6KERxi#gfkwaL$2jPS!zrbAJ0Q{T97vzr{|O<eTIY~uzoou%Tl
z*tt+B;<O(&m1b8(6vu-7HJkA(6y18m+STxPh~!0Uw8-WshiyR3x=FTP#u|F%_RkrC
zaYId#mkiJE`Ra*-))%$RJ37Mv8J7=uWVF2`%d{w329>sFKZt!hia`K!fxdb}4$Vx|
z|8S3xhkhuQ7GngbtB5f5!?Fx}tJDz$!6U}%YK($HvuGLm!57IEdbs$-QHfGIA?z>I
z5dc|0u*Cv{Fuo^NXT2ubdk}M!5S&j0Cq%DenG$kmT+jT*d6!@laUV;fB;vSXK8Wv;
zm!^DOIx61XP+>BVd+b&@VVFbgiF(ous43E?D{)q7k#7ZgN>={cHKKrjvfT7vb{3X}
zHrg@VK|_e1AT!rYpJA=m@deoYcP>sqWJs%HE$DY`s&sGzheq%|SMa727uuoF;b)@(
zMZY)Y)IjGeq}D`zEVnaKV|6iyt9ywE)+OjnueCac&ri+^=XxHI*EZJqBqd^<kNg9i
z%eZeQ{8`feg+u?#ln9{mM?G{V))SPh?dsG^iP)CbGlKMFy#vo)EFd?6R7T7y9r5L=
z5;-DB2xIB!90jvYF6c_QaEnhV-Ay*y8d_SC-b96L*%>`2D+ww#NF_Y34=r^QA97R*
zp>*sQt$C1Cr`<D9JPx<>=Ry=z?~<|clmCF*!ONyEt0LI9F-oWu1NrU60Pf=V7|}I-
za|cO`B-&#&yKA5`8u`>#DcxNuB!xhqpZv><y?#-0>K9z!`pFDz%43_2q*5ejZReA6
z1`T5=(5&{$$34;ByRWAQf)b2=ITOUE5Io^0*L}gYzU36XSL??!4bV*BWr8c)hO+lp
z8rtgsv5KE@Ku$rVK4@S-F1%CojPqY_a}zn)|4?;7F7}uS$EDwq^7R+KHL~{CwO=_e
z-=+;_L-)Xia)UV`To40AOfle%kZiCalDya(C?<Z<bk8<vsgoAcH&dVwO};%;ms<&y
zm$d`pMZf+;FlCfWnx{k!Y(imV2T8!Efh5~OX#MRV^U(cY;D{yscxZTC66MSIE3lY5
ziVEjVQ-TdND9x#l?X>^NaSjw3M`6~V{rL02$5NvD+?IhPqQPiM^vMVyA{ae<-V?hG
zn3!DRu)|CP1R+BcYpCoMz$?HbPB)k<^o+Rz<t^s+F%%_p_dRK+eE|qdkD=LJ=J092
zNaF1HQLK}gRzeLNuqQ}>+$k}bvm!sg$)F_bx12{219a3ekB5t0$8KY5RQp)2V$S=a
z8EK0|a%Xu-=!3J0Nn~F2|40x2XKw!AKe`O`OT_(^lX&Mg$Di~2Wl#$SZ)=xdJU!Cl
zDH#n;`Z7b#{_RBJL^2vcK^6yk3V2>SZ{h^JcN)aJ+Gzo<OfZ$yScroJpdq&0M2Gz4
zP4(b!wcAi-qT6Kwbq+-bXpdO<%O#CMj+i^t{l)d6Y!~G=*5Yezjb%0*{kUtTU^k@q
z@_pt2RC;Y)%Kk4ER6Ag}!b``LIWXlyfMfU_!i~3x?ZTlvtFBY>6(C=>RaDx!a+SeP
z*FZ&M2&1MeGknm`Bj^M`Z0|5&t*%?O90XsTTE2=T{$^_8&^`yO7+3vw<3VAj!p4c9
z!Wl^7rF{G%Bc(Zd#ZKS%6hnOUIiH@d#^g#fPfYY~ju-2`Gw3Bio^Gfi?j(-EPq!{m
zogC$20A?A;6pa9I&E+q@s)w99_Dm;@YxE{}yv15>b#frS-kRjNI~(5sNJ{OfE{;~i
zQX4r*APsHUv+*EL!aX`B!Bk<sUJEj<vqLeNW>J+r-^-5RvchDm?IW$?Z(C2qZ2#re
z;tUrg^5^m*tfYS)>eH|mDJU?qOH*LRx*zpmJ5pv>$1y5Q-HTV&Xz<<0O?p_vWr3+B
z?daVX1){r*wXsQHSh!=kH1=zt7>bLecT5GCv;gQpF<OAMSsU4&lv)L%d@ta-@*7y@
zzO|Y2T&uGDn8FNc@t&M)x$|hn9&v79tpi!t!D-lwcHFVt8e7AZf%t2#Q{cQoo`I~X
z8_?P~UIB((z+*ldwK}Gf&{-&wSPdu2u4?3_(887dm5Q&$mbm^4AWbI5K{VmGC3T<k
zHPhSJ-5!P#94&iOF2)KHn~pgLT7attzSyQwyreI9ACzXj;JPOQIYO<4pWBs>&Y)gE
zjmd)i43&K8R1>H>8X&l*Doyu%d?+nM0VOG_;vUPtj6MiVbp<i<g#J$n&A%Q8_cau1
z?F)K|6&Rej>yp0MPo-2h;t^B$4iH=s>CA<XKin6Uqd*O{zN5;P<LE?&K(;b|Usyo>
zW`QnrRcZ8ywavPlNcQDjyIRw-Z=n>>2sb^|4s3dzpL_cNKvfAMwRI5$D@R+ZNXYE=
zx1`P+*^fKiK_Hfca=7(jQ;*S-gMZlzc`LCPxgy%(*CboE12}1|ZA^=e7|f_%<3L^(
zcc-<cs&MWO;4209l(2;S#$1PXK*C)5%Cb<p1&sA>*D<iIqI2k+GGVL%%)N<?ybNPq
z2*oawO6XlO>E8DjtHL2vPNC{0P}#FvT}Y>Yy%>csi_rs{&EwBv{6^>x1e>xUhc$fe
zpcabvQy*q@heCaZSb@qE0ruKdwY(e1D{<qf69=Kst2Q+P{h55i@WxilE_7W&_sR}J
zR)i8WnTq6ee-H^_y8EHO%5e0UD`ZZB1Uk!2iDLm3-%sz6tX9a6dhnT<LM8Q8YvQ9<
zK7w3aM-(W&Y(Zu5x(+1d#$B=^K46=O|0#$q+F_r{Es{veJj-f9qVhl;>9TVS!Wc#3
zIvPWg1l!+d$HOsCn|=Y5l}U>3FrWhT#gBnBWLiUs6AmE1%K6Ye5R`8!@t4-jdF2|L
z5JR^D0B+C%%EvIW-AiBT4Ni~ORf>tNiT|<H{3+4>ABtb=)Sy^%bkkj2`%m@d%`GYA
zYCw3I2Ddm9aI<g~Y}UsnL*B-&Y+}cG?l}UkM}q)+#(drnLz{sMiffHW$5oEc=YVQ?
zZCF#R>j>x;t8eS*(TCPyh~;<f8<D6y>vM3KS1*$eaoBoiU6M}?4n7<U@S}42`T~|(
zGyNqDa#3&XuDF-W&Z%Qw0WG8pb48Sq(&@U=3T}Nnw{aZIl^_(w*iGQ3@yMZOG5r1%
zCW#6Rg_}llXxNmayDORiq4Y(WP%mGtUISy4UN$*$9@Jq-`xH`=qt1c&b<Knq`p7n=
zy=@V^S6fUY9fM46Eij&RrLxwBgihFjG6LZW=BGEv@cJ+0ubdUaF4;rL^iegn4=}ed
znxfx;#%=U)P-rQ^lXw%FZ0Is9_cJAz+{b4RJjb#unUG+nvT~t15OFCzF&z2-Qz_-Y
z|2~r(a)NUBFN9O7Uqlfffae+j2{b><>tRP%r_3aPNWMRH%9{jdQ*zxN%YLruXE3Z>
zc*)Qu!}rWmXf?aa27S6lbZ&Rak$%$vCjEeF4efJ0E9e~IYBYHMrgRcqrV>4MnfKv}
z{N$Og!b3zQllIM*eB?f^1XtILsFIApZ%ZKv=Of>TbA2P|79yV|;BJhqn_x3A@xNBV
zlw`<9Q)270F+0Hd6>e9>xu5I!S%KE4(k+^r{K|`Q7{X_7KIzq04C$F`O*8Q45CZMI
zaHSs<EJOl~=1xaCCoy<2n>Kuw0mIak|BJP^j;b<RyTv7xE|HE+C?d5%kd_XWMnF16
zx?39QR1r2vZ$UbxL`u55LAtwp!(H2R&i&4J@A-}U{qY-vF=T8UyWV%L^*nRVXHHS|
z>Eg{0oqC+)7^K(JU885dbD#t733@xpl3(!o!tpnKGKV#*0s$FL@X|8bHN6e^lp9lA
z(MNA04R(wk330sWoe?xzWdPMDQSh~vqt>A$#}~YX#|Gzz{T0`@VC?tZYk4l|)p#my
zvLV-QJ1#noL!2`_il+~!K^MYJ7TNatcQIIS&z<CwCjkr{uLs28TBYEa$9ZG5t!teC
z%pMdg<~}_#%}jO5_oVMdk_X%1%kS_Kn48L)SwpS02IhaAy4z%nm%#P#Ek{j(UNgRF
zuvZlj!oB;FRm(`DC!F;2QE#MsGzB-4Z;f+WZJ_Tm-X;uGHQ#u4v-q~s{|!p~H~jSf
zgwVuPQOhtVa2DmBgK(q2H;$nAJV1Y{B=H4WK*KMhW1bPT4KTE)1;s{)Xwz4E`(y>h
zrEfVjewAQ`EYXt9Cq{$?RFr6$rHtxInLSRnmtapoW{Vh+^tMW_q6#|wjzPBM!6!Tg
zM};+;dBLrGY@0?wxK=(K0o?Iftv}~kG`X5zQCzXzO(Kx=BJ?Q{-U1Wzq(#lprihcL
zWK0xttHvvR(ypfw2G>nv;C0-sY`Q%PX29;>z)kipF@jFgy)^h7tRTv<eS3~JZZemk
zqlzT&Y)kTz%6kNCv!!qIaCRlU8ZLV6f>*Rl@0<e>c1Q`JH#&a{>fT&zzd1pZvjLX4
z9-3Watw!`p)%I-5Sc?sS7fOnFD#<UfXIyQ&;eWRVR2#wR3lO7di%$eljZ@w0L{T+2
zNM(t`r>~{D&@-1QN;?1F*i7+EVQ3n6mOfX_=HqMZ$wZPH0rO+Mj-?aSJ>$BPnWtyX
zk&XeTqGYbYWuPi*xL$#NknUUjJYerUXi;(}>{j#PLp#z!{jxyrmEc1iXU)4v<L=&|
z>V-sLh(osVa*}TD2lcZtrqUi+3R4ObGNvg|WKq_Fqi<Bg;}KN(JAn04KZ?F2>*Cem
z)texTAD=b<7WG~}3BYen$nHnG)=;AE!`y%E3IAR<|KGoxzyRT40;@{>hi3^LDh;Eb
z=2kG3EvDQbhE2)LawN<MhJ5YjF2||@&eu~tcd`1-!j$9hx0MQWpnj`$>JBhUoOQ=t
zxG(paA-Rd3T9SE{6!nAq$46mA#L^)N<`X4G;-M{EEiW;XXS9uKTMmtKXpU*B<Vn$k
zFj2MMk@(+7%>@YEbIi?9;_gj`Kx2WSr)%+hxSxiQ4&zd<F_SU3*nadF1I#G$=Lv5T
zfE9b$)8hW<Kf%T1@F>zT_yK?+p^}x0z_1l_EIzq=jOoS=Jz4u`V*p=I`GXX!=B$!3
zexS1lWBnyyQbu4o21o#h2@@8=Hx@K@QJc&5S7L~iq+&qPog$swcKFkFvUGBq#UNyY
zz=%!GjPYGm^h)LvGB!Dd`r@^YD@-gRuFpj(17VsZ3$#rFRbxn8#)1z@SaNv%zT2^#
zikT1T_l7rUo+xQeKI}4*qmp60?=@sNX$pHbMEaX`Bbu@QHU2l>KyN@?VM^4o7F9P9
z@_ptL2o@c|vgP6mrH(LeE8C}IcHRoi=ySk{5!XewDDirECja*Z{Ev^&AZf%9ly1*|
zc1Pq%=!75X1tshQt1-9;P-#wycFu=>((_ZxJ`7yaZgKMnX~b7ibq;M6xjA1t=f*tR
z=M(Ic7N=-0!PIylbr4xg$N~voSDIm{X&Rz9Ybi1M6KP-C8^S*OdWPX^6D7kLv0F>v
zjBNGAR;Q&_9A{%<$!FS@4{OA<L;jp}U0YA^pIpyR9$aY7exRB8#)&I3NHPrmx|^!Y
zRm@XEOHVttQL^j~R{x2)L(wOL!c@teaJo3JM5U{Q+oUKAdE?ng3v8j5QV-vw1xBcf
z%mlBa>B{%*_d`*&(z8rTpVD5w?&2|8q%K;Atns%&nidk;u9-yd6CA!{jV_rdTfxD}
zSmqM=A0oY_N&x9C?blWMCshurAAj+e^B_?gyL;R?=&5!l^%uGa7XPznLm_%051y_?
zY~|l+6frkc<UEgN-g^)SR1Sgyv)_FxyNVP1pObb2ETCxBdGvkj-#f1N5A6XGJ_TqB
zWgUFjBlk2At(pWHe;cr8RO-3?83hqM*|dnkyZ`Z?`Hv4iC;muSVkT1~@o)7=x-@FP
z=xuEZc(zjj76c3ZV9USY1cl86*fz`o<>Uxb2mVLp%ij}e3d#CkZ98ZG+mQx>4{opK
zZnr^8U#x}Brq8>P89OCKte_@bF!vwd=YM?A29Tj8c;+?I|Muqpy?Xk8{eFp!><nJL
zVy6C&PyhdZZ~)ooPWye=`JaCq|LuRK7b8DGab{o>{r{{1{J#*B|KGoY0S5ouN#C8L
zZBh2wtZvVUM)W0l>!nHbx)b;3e^nj-?F0OOeZz)#z<p(Z=W!AIpH^8VSl5rQ21H8F
zFkByEC~q^BKG~sKlIQg^&dfp9G>!9*{z@4DOmFAZ`wZ3#rm;Sd1sZc8zZOJjBa)la
zik}36m8lqs!gSZqeuph@oJWf{-g3mwF>u;r=Jfq{RvH%<bu$ew(s^5o4KR@C>K9M`
z`30N!H)c;_oR-!Jk)~$KX3}xyD%cQ5aeH7F3$C_<iWF(A90cG&C#39t8dQJpB!h8F
z!I>-|pU?-?AT*w*h9{tFgj6Dl9wSxCdl)p%AA3b_uRoB4EV82_f*S!TW!X;Smb?Sb
zEu@+0S(5$Pn%@n8a(;mOeQx&sRv}>&@IP(?4*aki>PJC(AlHqJH~}<kQ8mkeaI?PZ
z+=Ny|6*2N@ARQ|iOj`iT@nfJ0;IIS%RiPT7ag0AC%JRuIcpWZ{0d@#%^h@m&EzDG(
zaIrLcWXjG|C=sLOf&lQEwm$5<I!&-=FPYkaq4U(wKw}H%?tOLJ)Uo~93eFskzsxc6
zm^yj^Oz*wVHQxh9u@s2>v*+51iBb06QmEe+5I^DNy#7LHZ<zVb_#byF5EAg*M-@3*
zHoqFq8Xs$sKG~P&IgqOC)}N-D_TT2<cM2@5_)W=+z#Tr-=6Q{Sl31`Ho&UHBv;Y*o
z&wnE0Uyu<dNcya~dm7Zw6o>tw$Tz!o*aaNHZYNcW-KR#R3qJ;}yCGpEXapEt!T+5G
z*&PFLXpJt2;i5oA+Rpzg$c4wrquDAN4CB$wAX~Q`VBu}yUa0t;-j4(t=%*xn5cnJ<
zgq(t2dJa&$*F33ba40ewN9C4)MzEYrGD9}|;m5L~mlPZqrl(PMUW}M1$QrWJKBr-Z
zv@-h77l!9~t6?LbuKtg_7n}UT$}%MVa$-_*F>c!+W24FUMUW2l87f+-=)m}952f`^
zs2GINkN@5I{=zB938Gr^0dn+Gkn~j)6QufkBLaPDN5{hJC-nPjBQEjY5GV!mk)qw-
zAJOkU{GQ^fy(IY^@Q2e#MqBcj_v2Qr>!o`N+c$83_mssmXaLuYbY{_iE=$lbg6k4p
zyM`j<oMy5e(SPgvc`FX4f6`fuWm7|O<2!G3_=M>PPLW{q1isqYWxm}+o?M|>cB*G>
zVX5LCJ6gOdW+*wT(QJ;&H_J|B!#)M@1`~jrB_XFIE0&OXXb0mR2PET2P)vUo@H)&o
zj-8Ke;9*SI5@LM_B$o|Vz!cKOUS@V(mkRv`?DY2z9>L@B_+~i0ULr%<s_@dBBQQh7
z7*REG;kMuvj;sM^Y~gJ2dJVWPCetlScl-hdjl3=~9C?a&to+JfW-b5@)Ea3f@1o>6
ziE&8qArl!J<0<g)y%#oW?$*}_MxP<KN}wMl<-F_xd2Ca+5Nk&z#0rymR93fY7Sebq
zEaxy)Rwu+%Oax!s?T^b2P*Lo)e5*O1{CnBRA}{HYm#yL-m(Bj~W!ncR)IISYP*qa2
z;hcY@lGAcz7$Ej{z0#mG*bOV_%Vw|#uu5>83Lx~UkU4S9?PEx;@_arz=4(d;4o7vm
zgEd`^>(xK6SW9v|lOnic3(N-pykdq0#O4ptua9Guc6#~hwFRN~tWUCh4Y%yd<98h9
zZ=v7#fGS|J=SQFUQZ%1!t#zY~f!EC=o%gfqJ2R|`KTdcreTrM5S%e>7aQcUlGt_|2
ze82|~arfk2f@ur~VQT<EnV#p<Fi-xjxi8U(gxyY;Ah#2TXHEjds!!MSQ8^aq*}k#?
z@bx|_(_5BeDLSQjfE-qVI=r1cbPjYMhuKq8YjSVcau`0R0>Q)7p(J00l3{LEd>H@`
z-NZ0a0N`YW#8Hvxo)si?cbgksb^Mt(lQG&uI%w^_M2WJP=Rn?T-=FHnDRm<1@6|eq
zQ=&2^b)9-mtpFCLNrYNCU93ViwxLURd&TPSKLuIuc6sO9%Zh#0$YlTfE*ZHo1<?Jt
z3T8*@jn>Dt!Crq4zGh)8LHe6<6(HM#9s!cOzG&XGxt5S(VpBE5kFw-EFepW+eDSX%
z>hg)dj6`D@bFb(h)yj(!g>6_f0h`cs!x%g@dPA<?ga;tF$M$#L93SX37dL_%kYJq-
zIE!}tF=$HwmOPDDJlfeuxZWaXXePm8Af4`eKSxk`Sk&aXy(ByRe51nj4K4TRGsN?|
zTOkIBl}qdQOFf1!VoIN&Tl3-ND{53->Hc|)>Yd_uiRS%0=z$0kxe!uIC6tUY^g8df
zkf&bwqod?pWH&PI26^NFsG!L1UP*IC-c<W*iXD5@&MOY2nkGQ<f3Njh!u2nN`*m(}
zRXF`MY?|E(E|rPaOYFXeeEOdGf1IYhPnVaJe!B2iDRQ1!S8j}nmak}Nf2S&m=&K)@
z_1UVk_xjXIhS_}({eGY8*VXCOc9?8hIZWSu^qs5kT8b~n)$i|HQvnfubG<RF(E2ug
zkBj_AzuqNzzneCXz^l=9x>9hPG$9bx5~^(Knk&tM832Qw6j8lHGtY}G-zm?%TB~ft
zCH3?KLep?zxBshA+lDX)l`8;4BV>ahx9-~(zqTqpy~9>4wmr8B-^D8OS}cSgrscAu
zlBNyi5$^v}`h#xe3)}~Idunb0t-V!wy$3WPZ$ZDTKI|2~kau<k1P|rllr46p%-3!X
z#`^)v;}X|xaE#)PrJjGr4at&-mrAuR+m+J-?jqAhol5a<TC^F)je~u_@mm3;2Q6(h
zRli2NLHf#sK0tcr&$fT`+alrZiI`j4nvL%zx+d3kNMQtMlOeuLe^w1p!QN#13k3IU
z|HgpeE|~l@E4T|*X*4KfPmZXVR*}-1Y2au<#U^KzN1Oi%nR&+!RGt2R0CF_VKEpR-
zFM0kM>Kl-ip`|Jq{wnSL=96DQ*>eP&Kqr<r1Q4N3N$A*0;4K|Zhrx0oK+v9R%9}hT
zma_)(5KD{q4-h$rN!34knB+J_fttzaCFx-=I77i9QVpJEyu#4~s0i+Le--vc!j>L~
z;LIYBbBF<0)4nc<1wu^&VkWo_=j*i{TLSZgJJ+S;uY3Y<#<nyW&V=SJ6g6$}BLp4x
z@5y}O>QdS~RC%WK5e99zcKijrxFmiW=F)VvF+{K-gU0kqj@$j7d{FUl>wE+wKcSJ!
zP6I3ry@Hl%<+|o&^3{=CK3Vb4?XvF3Vz#5`tXc|mt1*={A`CX-#c?2$G9D2fc>}B)
z@jS|ViUCJEj*QP4;3c*@Wb!xx%MoXV#<$d7np#8KBI_x%hWeX)Y%x&YOe&i-*#HI*
zQ|?P_@tp6iw~psd#jlyzL)o~eulNro#1{SN0xx3@X^&cCe2y8Dd~bGld_-t>GG*%#
zEzqeBfI?NQ?f``HK>=T>^`q!hvww)kLMlUQJC8cYzYGI@WKFLqOcYFyngDF+0*t;)
zd-=d^nf9*1rL98ha4JD{Fb(cKYh=(OVg>4pwZrKmU>cQGk}Lb|o9@tUIwCCE>qFc^
z?KcHa5R_*bVm_>)Wg0iN_dp07=f9$;1<F>{>#apVwc^irklr9JwEqvq(~LzG&C1Oe
z7rYzbVRG;-&nnxi#q9fwnDsYt>@Ij0G1XEbtpVVG%Qy0d6!*g1PGIciZ|Wuu6hDDc
z2&qvK<DNGTp1|{S<g)wNYCUbx%)l(g#nO9CQ5)U5%3yffMspeV^tt&jXx*_Yqe3WT
z2gODRVS0Ft%iTi=nYKr2wWg58?LNK)#1mwO@XhX+$C~$BSXq4Nn(f@Lw)`_^>ni}s
z&GS*)Sn>u6Tz{4AHv>>4NGs{xD{#n$FhAX}+NXaz`xJ=32;}}i%*~P+(^A#Xu2fSx
zJ&C=2y0E;3;#IN@!sG1QJ{wBAgZ0qNF<A7q_%KzS{hNB%U-R&AQS~LFW>r%m!pp>#
zv@@1Aot!I=IV0S-1BNzCa#3p;%mjw#Q#QVruP$;9Y2*7t5EWjT+d5g~2Ei72^(rFX
zQ<F+uuA{aWp$j(I?8pjQQYZ)#Svt^QrW=2@tny7J=s>KkNOQ;qXfpAZpa&kN)3b^(
z*W?G_%%^SpnZU<fliF5%>GUTpF9rsW)YwQ{8D^7CNjy`!@{R$hnm1J80mHxO>2NVH
z3fwO$)aE+-9_qO>I&Qstu~uovUXi0Gut=F?F_vhZ>akXG>wS@hK+Iz7bl;ku{_u7y
z+i|;cV6T;bzgOPczXOR8HNnJgXt>$ry_GncOw7Kf54BxQb>m0^WxQAV@^jDKjkO&r
zVg+-Zy^^4v%Drekrlr$~g*(1&lBd6~AQPXv*^J2aUEb|bY*UxF6S=PSx`mj4VEf7)
zmw}?nBYh#~pYOUM+vX-R4dy`SWZ-#l@y7a)K6d(;+nG!pFVQkIs(RNt$@qnvxDpuw
zqMh8}o%@f9kjbk(?|Fut1DjqKjf&Vlks^1kjlFs%^tWZnWR?9QS8TYM+=i`h+X0(9
z^ynJ5zyOr5e_EdDaab<3BfBA_G^}6dIz$sMV-i4psz*^|j7=SPpbN=>E^2y95&PDo
zU5*De99hwccLYFyn*DDNCf*a^w+9$<ytTu#oswpJqZ0hzIA|`?8Pl|$+eV=%4s0mz
zf#*y}*itPd_jE~ap294d2t6v4f~Z6uDBXk4BmHAQd?Z`oC#<ZeZos1AuJ~g#^H7-=
zO4VGA`yiApXr=Y$yj|3F>2Kapk^O3r7ufJ|jOB)&E2{oB0(%YC4SkLTbFk(*u!AZE
zX+wpi)Vm?X&~}6Odj9j+x8&5)h7uEmz8`N)8EU@DlDKL*2JzeijcD^S6)J0L(0Vv0
zo$W&d8pjU$ETskhSouJT1N^IHb|%z=kg}#0v=7gaF=L?yv){e|?=?H2t1nK0-hpW^
zUvyDyG>u09_jjhlQNUQxE(Q~RJilM1wgNl_=-7FI$poEdYeNvJHnaU>1){qjIlh=2
zx~b@v|Gr0sF21>z=1aR6HD74`v?{U277Q6-#l~>4Gjn6+Jn`)1?^;swy&UHBKk?m8
z!AKyH9t$SGzK4#EmY&XxZ7egMjJC>_z?tm=-WgssJb`QJtHyYszzrM@&nF*alTmx{
zfQSpNh)ACKByG=J@9K0XsmAKvZpy``$GaLkHy4*toznaUk-PNRoBW&HU2Nz~nq4#}
zxTw?Tj@yL4=o>gEKWrJ+k(}V3P@m8spi+$!_>?>y`jMm&!$2qv%{enCqw_;os=brI
zX%ujp_@h8n=1<UzYI@x<!Ek;LdOH1Mf6*z@(8-St%zN`;jwq%5$lbc-ZDudQoNr$O
zS2MZ93ml^)w}iLB%ESvxL62H!)+OQi)}NF*pyg-tF|dS}_@=m~&`dN)3(*P|bdvt5
zY*Lt4djxYi8MEDb_TW4Cj$a9-aU)QpKIGH5V};2<@9HP%N$|S`$Z!wgIHepdxAT4?
zkdyley%!;#vw-#^eYnZgf&y}RMA`7pCs6QRFjlB%`UTdgO$;CV+`=j$tt%N*UkPpA
z@hDiF4UZA{HV|GcG<TIr|CII(#<JxH^Wl8QE9Bi@oa*a8gfv7ity~jGQaaUAGl-Ub
z2`eyZG50_!1x!H;$$Q#XjpI!^#Sv+<Pu1iPUc5>ic27~L$YN8Z#|bwM;x6JBifQu7
zD;dRz=u!sEbTr=SPhTehF}ymyTP^IVkGP_nHK`EHEp+5)bY79a2K$#1@x4`r3kYp|
z;}6P|l*c&Bq7U}z=|rf}_{zu0t>__*giB13!0Mt|mUO}ZNNBb%CVZJSar%ui+R!>;
z%a5o<L?x}QvU*z%SJwLvYCqHj)r9|Kjg&sHj$UIP(Geat%5BQ&)-EE{2o?NV;X_BO
zwPhaa4F1t`qA)YozkBo`;Yag1A)J=8P50pxv?pqTaoSi8Eu<@k4WU#x62sPSR4+W{
zguoQ2#O7+CwzOV8V%p(tnNg3w^motRh~S#@^Y{eX)8;9gNIcFxS5S;h=t9~8$*)Yj
zLNBnucW{0S%CjH!Y<>Bpht2&&jJD+5OLaJHzZ~9J-sH_r-zb~Z61wnopZBb3jqwhL
zG}}ZG5B$DIgZ1&w-yboW`{zEV<F=REXRJ3mh8c>xB_;$MTeq<grfS&SRq?cQKHi_1
zqn;i6KVgzFPQ~)?yXsCh$j`<eo6Hawi_FIxcxYA4H>J!(5T|tn3G9r&+&y{HUqd$M
z(Zv443p!pTDKu{JJK8oWrPBVJLG5^7)YbA`zS9q8GiXYG7I@dIqPMCWuXh7j`+9=>
zdFn0>j!4_Iw~ANCrr++=I`NAtF8O)R+x3k()e!gbe{bx!icKp$Vw9OfC#Kuoj!^2<
zTPkiPwLo}{wFGR%PRw#HVrMVn;t&{BZgCp4j^A)zCEa5;72L2n7=nEv`6K&yEJaaK
zfCR?xjb%DM3ktJSs`l$NMQZ`?>U`!NOW}>?qT%9qJN1_(Rt_rkvigKCa29jANrh4d
zT!1-Gjb!aNMQLF3eUINe)Ro21@$#?-nLVc1iWh&+TYP!sGUhdo7DwN`n~#bW1M_M<
zq<CYe>dBv$(&0320Lmsk$8r(7d%sO0$XAhlRRHm7V1q0}u~14NCBexo*&+Bfj@K@z
zQ8?bqc?)cgJdGU=5&w1XK$cQ{e_^C4X!zwlj4_AoCS18}ZE6R>i}{hfc***FI_5R?
z2H|phwR5wB7Se@9_mSNBp<Te7U-9c%P7S3K&ng<s^OJ=0r;ONN!rPFi4W4QdAz6#g
z<2Te@+>4=Qe*EiFL2(%X7Vq4MK+T3O!d3crwz|=?W9{RCss|B1o$K;jm%jss?7$Bt
z?>hh7x>n37+i3;{Lw}TsGDq<$`qrHt1&eLhGs|C#C%;}O$Wk}RBZrd6O|VOj#t@pK
z8s)fmY~YA>^DRH>;@cGJ%E$I#$Lpx7mPCj@f3rojiecJZWjAQ^$VRVRp|OHlmU%di
zKLyiSkbYI|d`&Ik-3f{*mJvfVXqQ|qe~;vT?8<E{rh5VWdUM9c3Q{-aEZhJGLxEOK
zj45jnAb=Q}$n_ha>I)D%_6x5|4!=#-_UC58B=(3B4Cl$P-IC+UZW_<*{y^QWElY$&
z5~wpBU7p^nhJ&lo{#=O!D&#F-&4vs_3~)E=1Y2+3BT-f$`Z!V3KfgQ_9B;@Y9L%^4
z6<S7{ddw;7nPc47;uv;xdE#+>*wf&GI@T5I6KA_bNs_`(az|T0`5>&7!u0R3hC93<
z&%8DIL!n!T+=%2CG7@zyZz(EDH}jBD$JjqOF83NFh02fSgG3`K)J7JLyWi($ug4BB
zG0?0t4icmnwChdzJXi>{w2~-KJOF-ZLUM@xGwe?gkpeay0|W2L0a&Dl-I7U@@mZ*T
z|AFIh3`}xamIwDsgk_i17#AFB=5G6Xq2@K$uZoxENdsgi1>d2pDcWx<%*e{~FH!s(
z*v@%(s=w!6aS|{{`K0fR#o#1|hbLnn7bu~TiKY(3_*h$KH8GOy+<H-8llE8a1L$m;
zUwGW8Ap9WF_wT&o4h5DIraMQ(uYYW~F-m_oTuzY#rw8P2eq>YnkvgM}6|b#$G(dh*
zj_l<DY&gc>>sXInI3$RY;BDes2MSinYaAP<RaWIHHS-mq=VqX4hTCRDXCPXf#2S1M
z$=FgfZRHNI3SD~Swq}LcKX`GayItH~tNgleoCoH>)CE}$=gf8lvPP|2f5)&j>YisC
zvM*0(3L%kNI>Dt5GeOL6w64;$b0J!(&BEga-kPD;`wqFTRF2khE<OSy6FJ)V9B%1X
zi^|4g%Dqc{9vCBtQ=c81$R%O4T$lN5e@tR+*Cbw&CMni|1PkC+P}8&<9^ulr<=sVj
z<8fyv$@jZPeh;bHtxyd+ukR@J6mkPDaCx8wGxb{iA!3!Z(Y#Pui8YUNSMWX^%HJP<
z!G{CN5@JxlX&+djP1$0k(c_;l4eC-Ke@VLR_T23YDXKbx>G}H13iu*k<#X7rUD^C7
z#Xi$l{ypRAaU*hdp+}5<s%P_P-1Jl5?lTgMODlxy68o6a1Jui-iAvjHK^4q(L~W*o
zv^<@G+xW`e1y9j9+fUK1J8M7o30DQFiWca63YIUmiBV-Du6=KgAacZku2CEN_emE~
z<k#~|^ICM4jP5s4VEXM3kh>(j2qH&^>h0S{hon~z7bhia(>xs{)n|GrfI0s;&`+g-
zaN$`nA))a8A!Dk>zW<9Cg?`fC;;Le>gfWj6$D4>J#0dRi%&p1V6;D6i3seQa6mu?U
zs`?nA(h`$dN_PdiIiNgYlHjEG^YWV~gicd-88vm_Ou^5@sOeC}5kV7K39oU8S7jR}
zoGq~)Dm@6Z8e#BPc%Reyv4yTIey2q&Ul{W@`JrS351Te05)jw*jOXEs(s`E)$>>7w
z;%KpW=^*71(m`N~jXGtFe}a8cBzH)9q<Z3jnZ-hay7{!0DL|by79aXUne-dJggIv3
zh{QL^p8=fEPXD_@C|s7ut&gt*g6#So8XjVZa}*)jzTL$|t8>{u;?KvSda)b?u54B`
z($fS+B2)76sc_Ly!SaxtU+$&F8uQz&_Qzh}ev5T;pf@}g0RVx*15_DAdT-$Rw<pQk
zN8B0IfsOThVpn%n1gkS<-9up7%tAG|k)mIw^PVYsUgVL+I<_wg)y$Oq$=95WzUn`u
zGS;kJZh>rKPvNJD94jO^x1ZDPmYsgXgv#j<=k-Qr=TFViZ^|C0S{yM_%H$v{MU%s3
z$E%8|p|ZKldwj?40`IGoFH~$2I|OA~B5Wfe1LKQ9U+t2-rZ|11DPGGJjP*auBpjF<
zw#%m&Kezf?Mj%ODw%YE8g0|fp0Q59HxhG6%1v8|JJ{7CR{mSQ!bT6eMT_K27AXKwN
z{dmqzh}v8&%RZ<rN=b$qY<VRa36l)`<Py2Iu!P1JDbz?RK`GoUop$B`i9}g)$stW`
z{us=*<ga{0ood=)V^WnEDkSn<6=xJ5az;0|gkP&n%31v7A$sJc((59LQI!@(D@2E?
ztmJpbuJS3w>eD_oIX?}K11{rhG%ew+{5XRoRzG`U8!aymMk$}SJHCq^<6lk~TbwkV
zav&YC1tzSbN}pVP1X?DT^(%+2`vl$MDtrPrBe>jxc<%(b_DHt0ht(4_dx?rTOZ$?#
zRD6aE@-HCJMT!dAo`KreqVH*2^NH3Lb5rOq%XYJB5xFFa22tN}4isz@D?U4N?td<E
z=9lSQfqr!`y{@|36RI$Ys+b0dT~lOiALSQ!S_v`DEzjMZ!LYSDcYM;T{inC#NgQnZ
zZ>%K5FzFzk9+^TH$xDBJ!cuniz0L@%LPegpRz5&wA~S>1`q4e<TwvAK>}iC}I)uPb
z2r(A6rq4e(k3gAL3{NoQ`xh6f&bt|$!)G6N*ZX3QnBoyHg_itA9cyB7NO<9Ajs_)1
z;IQiDb?G~k{Gx?w?5jgsAPIpHLbd20Oz-RMLVdb?(g0bm2!oXulMp|@_90}Dm_dtG
zjas0KZ+hcdLg?=p0P{)+d|7sHWogNP1&9_aq>oJY=YQJ*^57Kq!1~9Roq)zXVLzTF
zQ!KU@aJC0d`yrj*Gs7%=72+?6^5k)WbJ8W2-q2*dSk$;lEsrlKlc@E_NS(AETYhdT
zTn!Rn>U7o`EgSo@&p<iG_PHKihMH3aO)iG;8<PO1*soBiP%L$kqz<M(^;FV2j`|)X
zpv9P!yDshHG>^*&4$8raGgpXXK&Q+;xq#!_)GE-OOz*rWLoKSm+`PMkt`BcgeNkSN
zfnXbRjrZKZ?~Wy0$$*46l=OEeh>6k(H_M`zP^%Rbefg#S)#riHI}0m`io}&6ja{$M
zOQJg2(Cg4(n5k$l`p*Uf=7mIp;;%OGT+&XIP}#(wt{BU1aPBg+9g%;dPd3Kid03#I
z(VInBLY_4j4eAPE;zm`0f-|P)SLZ_CZkt0)OZIUhmK@x`u<NuYnoHrAqp@!TN>X|`
zm7*=~0#f9I<yjO<7h837ci~!`tS7a5#Rt=DHs1*tw|KkV>C}FuJi~iA&gYsG_B1eN
zny!ZBq#&ev@nQ{Rs>-UCJGBpe$yavaAq$&x8v<>IOL@i+k<y!=bG=5%SQ+@2jy*~b
zz5O;R=F!CT1i2mRAN^e|Uc!h<jZN99wUSk3(T`*qfqoXNFv+MWuf(ph-hW#wTyK90
z@OsTho|yn1*$H2VeMOc-wI)^ubAo!xr=PByoWSV(XA8I<?ez79u3o`gKR^X@b>y~l
z-#jR1qXqT9x{(y3^WWZ#+$-JTCUx>_qMfu?cpqQ)3uVBq@Mm`HFYFIxpI)|pl14zE
zy4eZrE;FvI6BDrR|Jbmc?o<**uqj1(zgSIhiVirsY3$$3R)USbJ@dZm)*?$=B$##`
zFFcLY`r-4*leX^Bp8K(EE`Q@3M(p>F%{pGv3QXqd-CEP*A+@VZRo!<!e8WQ{uJx60
z0t=F$VMWp}9@;KB9d?vTf<}y&#xIXeYu5wS)Wz4I%EG&E!?e=~D`eD1E3}^HV@f+~
z$qp8qNC&-xOZe-4WSPBRQA-C{;Ranm_rMnuW(2m^ah|HE_^+2K!<kNx`dWQ;0X7de
z_#7cp5T-rdB@|V@B^8@<=P`JTAwI3F*~yOK>yCMY&A(4aEt7Jx&{V|3)>T`>7B(|0
zb%B2;JfO1W$nwt%=a(vjWNL6!F%>I=RV8W8G63U7gnLR!yGs{`n31S=0retN&6k4p
zHz6KpC=Fm*On=tr%4V>5=csR?U53i$p`Z!_p!}7^^;@5BbyIZnAsF=4<jF)+E~>(t
zq}Xr5IO15UL+X_)_;Zz2oUs%KO*5t0A=-&AYAA+sy7@x2Up_!sa(70omlbF*DZ75w
zjW0`ip#Q9$)?=AS1tra|8BB@j9I95}nz>-PD6`u>IPX;zm59Uge#vn8S;N_bcaThc
zOmQ7n2A2yA3tue2Q*p>$U)C1CMxc-|gmEx(SQ05a3*h_m4{NltkBQ`xlz(1WkB}RD
z^x-WSV~q!uOWGp@e0H-Nb#SF{8N3+lem?2U+r#BfxSSnX@)5?XiZ|}pvKIp|(lMFz
zUUmP{98+}^$qIQhgnzhC_n3)q1J87UiyBOZsDhKFrnZQ#u9dnAe<(YiTpFgyXP0sK
z?X&AIT_JzW4BhlwX~@KcEr1?=+@M|%xtQM$kmb@nb0as!K$L{%*wlU!I=G5fG{$a{
z^yUZ79fyh#ml2gQU%IIE7|JR7ewR+u!7TE8O5&h(*&pT4GNyXiyg;X9PCc|L%M+=5
zMGYN4a~IIYmw4Wwro&VgO>`njZY4K*|D$1~eot;kW19dA`h$uV?H-+hrMPG-IG;_I
z2<|}#Ac}zwK)SH08#xoZbYot#c(o1zBav(36}G8v+J(#!^)nR>swOHujO!uVysP1t
zIHRW_<j@dI#x3&x=GG&jR5`A(=85Vd4>H}kFa1U$H(G6I8LI+bx)Qcjzj|A)1ANMj
z`=yZ)EuER^*NZ<JMrM@1I+El?GaPDd$;Jy(KBalY4R7fLXE8OHHqUwSbJ^=tJ4iXe
z(7cH1uL|slDL}-;e4FDnMPMFKYwVpguSt%Tsy{k<(E5+PMZ3)UN1vW)Atd9{xJPdT
z<1z2rYs+l{QFq%}Al<g6MccM~A`15iJ0nYoRhO$9Y?eMtlAGQmPB4uO;Vc_bfrOk_
zP90Db^$)#k+v}gJkn@vhoa&`&i$s6nhx~bUKL>-Jj@F)*!pW}FdH!u0hwtYd*V$Kn
z4U0X(4HeAjJrRZ^qRi-Q_}vj0BP{7*-nB}h*K}SV#83m$sz1!*65Cs9H*81vBXz6#
z$B|%Q9-Vivqzp3!zVLRk|B^6Ajy>Gx5wzyP_poCSWB#0&cEd3jPD&I-6d1h{=;w|Z
zNQhZW|LPd3OII1liB9}vT-QkfE{4g#5XyIA)n3ND?|3ivxJLME57<pj5pR~nVg5Wu
zR7@(o6NYFTqxrC$6<*@!Mvg=dfrO<QL4Ki5Q)PMz?_;ke)~msFdGm0$wObxR`KP5C
z{E@yp&_@AUqgJhukQ1b=lFxGbl;i?ykWxzY@|lV8B>J}k;%6_DS-p*-FT(_b!YFD7
zple{xTrwZx;6Pf`&4MmPo<kK%ya~}yU&=!F`H<z~U~LoXqhJaBGcQ_>qaw?dh&}jA
z8=Z~fxje0;doUS0-yi4L3%oc;T=_KnOH9q}ETPSO?Ie;4PUeoQ>YRqR_r42<VL8{b
zwN=AaLmY|VklvxOKx=MR1-BW4oJ%4+J_R4CVuGJ(?*eA#M8E9!guL0|-ARB8M8UrG
zNUX?KfEb{C+=F)aJJA09L4_wF4rSW;KPn?78W%ImmNAHPp{xJ$0ys3BZ}dO?L@2`Z
zWQ8-ek()%nj}{mzc<9rss^Xjmto874yq!*Ssf|20!(wb9@8>=aco3;DEDZf_!;L`v
zu>BE<AyF}{HT==p8x*4Y_CafKM@aRfRz4p#g`e2!vS7iMyYQ$9F3RU${LyVoBn)2q
zu!2R*G|}#yMV^!OYH;)+_NtD=vh3K+W7@s%d5{cH-KM@Tf{2KIRmCI+VBhd(mn73&
z7ApkiD4EGF_Fg;EWT1?He`cn^_jX2BtRfyFf-;JYg1o_p(~vhfOQIYb)7Xhas9ioy
z+QrszW#Y|y+busok&dBH|3LKaV2#lGw`o#V?^&{vvW-*eT~@jzjqL_<N54G&kmWz^
zX1*r!^}==01JXC~q<`Xm6HM3B*1thphjOA_pf_F)jjZ^s&r5MO^$pnmaEO#nn?~UP
ziJpw|UHKw~Bwj`KdJ=-`Gj2^<)sR(uIsG=h_$;L*qb$TuZxc-0Se0FSvW)-bnKF)9
zWqF+9j230KQW_W{g1Q^$ONh9E;Um&I5lW``*w)WggVi}f*=2J2U`E4Lk^Sxz;`h(b
zqTZbpbuO#kY{Yq_capL1`~kwXO|^%Xa;P6Zgj3sSX|&*&iU;{g@2$ZoUPt&zBXDU1
zN8;U>H^r4Q*{8Fb3YL?@&Df)u&sJtYx{|q%I&S>>GdTpN9#e@F?&19vs&0Ko=%+On
z<qA_VxfP4CEUM4*YBfBZ@ujLNwv$<T*Uv{RbJ@E$ePydODapG9V2<WwKup!6s=rMQ
zrl)&kW=$?!ygx0dSsS`wcEi>Kl_xp#PJrn)gU>p5mq7nb2V+c+$4Pq^JIyMTIUczP
z;>wA&U^fQ`@;}s#a))7gcVQerky1ILevBTEB{QvMspZn}AgzAAM=vh`Yb#CJfU4W*
zi8h<0ZWPXF46@*?N*v9cyTVf!^>Uy1-WjugLVq^QmbkPBv3#Np(XW^;Hwj<m(T{$L
z69*ZM2avi{q)P16AM>e_7}K}v{JmNJONPb`KbA4#y(AlgO-orkZ2cjTNsIVP$_PUi
zpX$ZAjOiP;B~!=-k5M4y(gpuIjzTwtEdth-Wy3u@!@BVm*^lkGs@9g#T5a<Z_aEnE
zlbZzFTXd<13R2eiWafUOANur+;A=Pk5l|LZIp(77gADA;@rJK+=U#H5*Bw*0EsS$&
zCyXOb2}(+JV+4@ff+MuJp4ShO4Dg@v9$H{kC~*Jp#%nY~5+fU;7~kSAAizd1>~+4M
z+@ReXV+p%w%VE>dpTK4!9QY|K`jWq;YBahZKAvsC{cYT-p%UwbjciOQ<3Au3l>5vw
z>9m1<mI^&ILd&73{%K@G3Rf<lr%f7ec%K}0Wj@Ba7RRjKZL=bwJ{$KHv4V$TMuZV9
z>}7>1X`a{c*bmx!xyM;&-yEh=aW8P4?UEI1kkAJCwZ1yEYp-(`gY}l9fsvAcVN+ci
z+M@8ks=ve2cpcj(tF6(JwRcKmwdd{KtHXpaQ=B~(&lhIwo*CH@EX843Ig+XWj9sB8
zY+|nXYzW03Rh1qTl8nq4;|NbZdra4)hVu{~m53ii8*WbyZ<J{%!q`t{u`6q~zhiNq
zu(k&{MY+sp#wt42$md{v4G+WR@*mNZ5X%%&nC$D?YKQnb*T{c(y%o)b&Q40pRHEYA
z$SA_KQ``yatS?xYS}qrh@Yr`b!-RJ?%Ni76Sf_&=6Yo-To?t2ysIj28R9upj=y<He
z6^c9xT8Y{C+%Ptqv!dY3Y?q{t=?*lCLggu~)Ki3*b(|<#MK<iDGPz3`o6JnFiDdU#
zR-6SaSm$22jRbWN%-M>9(Z9j;qf;)ANcVV0t;d&gL3+Eo=Je^l8__qU4BjIOBgA@B
z#W^_*y41tT#TxFUWq*$WbME2&a06nvk5qF3uPA&xc((tkIPEL<_(uvBZ4Y`aYN<`h
zSVZ&KiM0aFLPY~N$jy{eg$2GQ2Y5%Kaf-Z_Yr4BexheK+i4qd>eY=@3gzduLj9S=k
z7l>2v`y(u9+IY`HUlo4H#AcA1h8Wi-KZ{g05hmb@qH8tSs<N1p_q7?bi9Zm_aTT(n
z{9I%Q?sgNokz(Yi=MzD7h#Y_%Mg|E_7_DgVK1JvGKh_<fZc9W`x?v~C3HZPVHx8H*
z?<%fg{B!pKzjZ3=+j3l0_dN%>@qd*=h$+6M(Yf#lUvK|&P|LxDj{A%^d=A>XSK#yp
zlukMI<W6tgMT28@BLpHPN4s(=NCpX*tw62gP^^I&YLz58y+`quVu?PI?O%;ld_`~M
zTv~$_>`sthMG{{m#Xnh{FAtRoB*ZKp+1@D9%xp9(I;LH7(kt<w-HaFiVvM<sNGjIT
zahuQZ{MGN9n&3tYuibH*oMJB>F~$%0UH!uDdJXy$1+B`hTJ{P3e6^B(2}bS@s6ZB7
zq<g}OkJ9C)OI2q)RSoTPGN-zDy{;6`NehyYm-L36i^)Hd?{&-7)D~*R_}>)@-47R#
zC;xqkDrNclyZn>Y-rp>gEsBK1j|4ES;<YVvTgb#pGOK{#XiCzv5F3Slz9~XN50&eQ
zhpRm7F0Yk$93~2Vx)6{CM^Uf^Zm6Qa!`la#JS(M$;`~;t?VlO}J)hrfdvS-ja^6S{
z&l6~F--6W^5coBi|Mc&`vaph;PbX_rdtLeZk%;Rjjt`biJt%1R9@ZxWR!ELON*WEg
z=nVk|0hga=_4InK^Qt!3RRZS6%Vmfwx%;?_dPuA5t5%;=kkDu?XFHC!<|ga2ix6I<
z1CT?CpbJ=mcE)UF*eHCHZ;7+zLLOJT-2~4DSn3L|^_xuw=Y30%m-cBYyT7UVSu!!0
z|3UmK7p;9^DE(n`1c5Cuq^T`I!~k{ZFgvg;F%9xVHZO(!-3$`@X9ccQ+L)nNSFl|&
zJ?Br&%R*A5#E%sVcr6pJ->?efO#VF+$`cbC5k3w8gou)@xFD+IJ@%;}ULL#(xt}WC
zT;M9&`;UUD_HV(|RE4l8lxZ$&2aYgXoFc5J$sf4x6wJcY@?Gc+U!NK=6KEZobbO>g
z?Lmv^$f8b-LF!s-9&R~zyw>!Y1#M&a>XW8mu^XT&<&zs5^b9gb!gxP|i7X@6CNqgL
z&#e%=3P~Hqy*0g0X)#A_%@ibuPNM$2Gxmx+U`fT`%2w&xIQ(t9RaoVkHMPdG!Xib3
zry^Q5Hb}U^+}beY1Tcs&5hLKxRP{4k)fjxng>NO97x9D_pDJ#{*0f%a#3=V0^=A3d
zDV0SRmW4-3L`E%|vi{e)nZQBe>JNz{-d}JXSrE}IL_K0U_j_vFn67ciNV&!%(YQKr
z>UmGA%^ef0+>?V)TTnWg+Hik6TlKLaPA4p>JPP^O!Ql@_e}#o&^kff^n1XD-wr=S<
zsH$)MX6&ea&8E8VN>n0|yK&q49VXjc(|Dd+Oeo9e4>|qN4bSsZ3?a;A_*!X9mOtMg
zOUX0XH+=gP((QG?WZJkbi{5PZ)gAM=O>klS5-qF<-{Nf0rkVD9%x7e2Buv;g&mH5U
z>2qG|o05v-d&p;3Eks+RU%a<Ps+kk3kn<n{t;w&^{*$eqZW_t^UxU!uEx3iawq4yV
zW!0JguNpjFy~l1b?%LOTI=jE7DiZUf{t%<Kw8pQ;lck=)%JJCITgs#o_z!*0zdneS
zBecCRnIQc!51Ncw2)bl7De?E4ggNBZmF_lNC)O9fvv4oa6FKJ7sbi>0PQc-wI96o`
zr~^H>l{zo~eI&eBBnRVo*NYFYe;ATQML+PntY&sW8c&~dae>qrvhOeboDE7fWwcQY
zlvCroDEA!1N%g^wNEa!kT?)~|r-(%_sSnJe1_oL|GnptwA9C)Gk<bVJMt|ELO|x_9
zxY0ONhMX-@kTAS>su|B(6^X(m37S9GD`Utwu!$~fM14?i-SSW-O~jj;Nm4>co{erc
zhd?Fzd<akCuVb0Z&eW(T61r?-;HVcD_$_yDSwJ6~i5=yh=!biFWuEk+i$*QO>Nb<n
z25GB&`f8~_+!wz#A9aXkgpXcG7%N`>6XYl&hum9-C|5}84E}BuVt(PZjj-dK#|JOD
zH4T1CJz*241#SIc&f3nmRJvZ)f|3BOPZ9NA?Qbg0Z}A1!>8gMhepPSH*HFNC4g~)&
z9Rl+Ozknr7=&rI^DG$~rT}w_$?FLO3K>@eK$URnZ!jIBR7`~XqXMcb{TGZXsC7;H7
z?JDCD6>oCf41=)R)?sbc7)1~+<G`?dqW(_P0+NNVmU?4?n^z*sj8nBh8ae}<;ojlL
z*3Mqk8-Wom9)Frmr`qUuy#zBjHVpc&B(8D0-R!FgE;-77m2`zc8$LlgVD8l0sE82P
zdppP}R?}|%tECCJmjQsQ#$1s@h=Z>a&w$U039HiiQ!0e=h(Eg&%it-&fvld?ulqY6
z;NwPf4bHMU5m{3N0?Qu*$|xI{TI}T80=RGJ-nV-r>>Aj4@iG6kZP9B6=RmX$8Ko!*
z@3l3G>>(;A+<$6L&bwd4%yc}ZJ!kx4?WuLdXqAY`B_l<>?-juHJie*<N0OU48mGP<
zP-swJa+aGCHl|;bdVW6|*Z9>dSafh^l_!FGM(_!;zP9g=j^5o)?G!0FP2xiCYs7}_
zK$Mzo3rz9WEvDO#o*q6h7SqgVcLD^fWF}C`O@1CA7it<(8u{0z2Dk|fCU>7F<N8Zp
z*B*3T-1T!0`i(**;}7n%>C&$XEVAVx^Z-S<7)kJov#2e(9enA@W=1q882vo%r@x)V
z*ZNX$*n-X2-_UQGhhBcFhz@=MnpWmcU!VwSVtA11w*Pw#sO}TTmA?5lUE{3Y=W6gX
zW7aH_o!ip;TP)LYDCmhwJP&yWyYv_Cd|KHh*bsRYLQX}WT)?q4qdo>;bEu3U(hrqe
z(eHu?lbXu_ePIkKP~W537rz4HubuA-Ng95I8i#16-7$mv1p&_Ib3REAeACibhfW!v
zgPjn(x3L3?p^p{4QsmAj;^`#tTr{G8V<Z{(KU~SY>w<-q(mHIt5IFndyPA@8_`7|m
z=#N*~yhMs)yG}1x$jt3B&Ct)Srk^HeDbD^gp6zH)Z8pgckNSUPUKvMpXb4#|W$W1+
z_w`Gg8PHeZH%Kxy+-5j$kEoF9z<))$FUxF{i?t`GDZH6he9uuAm=mfbGxI)8_;!Rm
z@qG7f;^icd#k8ybMx{I7q(_yF>GX}x>K{=uzum^YQ>q=NIWI&?PKmNw;Nv|ja&9v(
zXbPE@rlN@F$L*TYo1uEis}}zAlE$UIkYqX>_HsYA2@iU$yf*}iA_&&R%MG=Q(=9tN
z1n3jzkFm}281MO*!RP7}&Rsn%-xYQj?6s=l2#zuJHlO{1^@tm(+rbYzZ^aX{)kD@Z
zMy@V7Spa-mzrL)>fr|)`clN+m*>d+d7%Wzk4iXjc^^tx2N*X3#o_49Ka4arW<F?09
zvmJhBtV4p`2|Un}+zK*)Qm96%cpW`4hfdb{P*ROr<<Rxg5K8d;_~SaP65b>_7-pc-
zZ3u_zN=dW92{Pc&At`B}Hl`<>nKT<}4$13iH90S2)H>(|>6ru<*w&JrqOCzX6+D1o
zu+wbtrN<zd(@Sx6Z+U52(#rZCLK4$Z?6`8D@`_p&D;v^Oj{YYsWB0w!$(Fef$vv_(
zo;aRK>Tf<pg~wvRp=DB!8Sff3H-raC!RTWTmfE|463fiOuR(RA@mY<0{Q7t=Y8#NF
z6O-Vt?7p$a@W?@Z6#+ezlur2hH0=70Nv5Q<9z1?f${>#fhh@|u(+!xd5t2w*dK<iW
zf3{n=GtmzxWn>q1vCsgZa2$6s21ZJH9MY+HH>Aw(X{d9Ka`$2}@iRCo59^!#K<A8`
zT>HM$(XVgy$zKvFy{?RH$~{fMrYjUdfFgyof`gGoW5_2xC5Ka4?ZqROFQvjVlj83`
z8UTra&s%n4<_{gj-Y;1NCpo~6Vf6a|{;ddlt1IRB<JONAC<i>}bC-u2YCPsZenKKh
z0lgd=?_%nz;}Ckp;<^u;Ahl4I^VSH4KW=;=tP&4bgbh~mK7NmTB`ClI#zRum6_+Tz
zGprA!^`t$cueMan;KY7#RR7xU!M5CivFi-^C2j9XHkSL{7f5fE!n!e-5ghuJc$zh(
z;KwPgQTpD{qqwLiL!n!Zs5a@^<Rm1_Uu-@&@fF)YmZKSkJ&GqVmmJV%=#SxF#8lz>
zo^Bv@6;?^KAqIpVXKj*?_1F~-nBR30KuG!N?FPnj-20{|zdSh>XO;_X2F4Y}JBH<j
z2jeCYymvivHKNdW@S4QuLU`Z?7~IEJmjN%4%JcQC+_S-cd}{sy{)`!as>_!o>e>nG
zz619*an!eB?0w~FdNY<}fR<3^Zaf^0XyMP{G*GKGUO5MMF}Fw<EgB6qFsRBYX<7`W
zQ{hurvvP=<MLm%9#BXyTD}yLZkI^Y=P4|Fk@|!-;fiBBVxyu$0!&u>{=j-(ON-vj8
zzu+PK#E(tB2;WjeZ~Ym!+~;FH4(*xt#;}<(ERz%M)W<|%?-eSwpWj<TnZYpe7WOp~
zMrk1b@qkJvGlRlX_k$;&G$i;cxQD?04GueRUlTWNo2KR9_~nbpl#|x1sKNcn3<wX&
zF?%RM8`JvU=-1CDqVB=E5@}QiY~eT5{c-l>A)6TVr;!#IK+IUNL%^lqM=2A}m1>p|
ze}2~+le$UHve*HCiKK@_ZLzU(N+9@EQ(=n|Xc0p_qF*g)cE#9b!+pY6<Z2;!TC!Yy
zvKK80I-<_yW*R+i^B$f39_a<^vdO0f#gIp<y;ex&36KXO&jbd@kME4yq-EQDAwaUS
z;?)2AF?mutH5}5;O+4k<<DFs9Sc{r!<bq|iMXbijaXFZ_9#R<${maT(X7~iU<69}7
zFz!rXuMj$oYpual_$4Ew%F2nVk=j9ULcpP81kgg8MslJh)GzACrt6Qnio*oRYJY9a
zH8fN(zMT0gC_3=&Y-_xK^$($FWX#~6NiCJjB~ih87eg5sWt1jZJ4H5}#If%p$?~G#
zQKVg*hnRDFYtD@EBFSwbUvi;;k=w7*!K_X<wEpuax9BpgEjY@0SmjYg+@I4wTAa#f
zyC<-@k7Q|{qqpTdEcwL`^9$=NcPC_7SiH_}+8v>~qh)Eh6Hc;;KTaTUb|`x7<*A%$
zn;W<3_Z!51bIc*4XU{a%V>SNgiQSFjk=(D-PUt5G?po(mexuJbUcEbqMYg_)ClPFa
zer2;Y9G1aJuQQXvAK2Be;wNUc$1ct@UN@hKc4#V$Kd%y{Ba~XqsFoY=D>FYUDtp`C
zYFgoP^Dx47i^cNi0CM-wYU-dWs#T}0wJWc`PJMGK;!ZH>m85VV(MAyb0F_}I=w_ux
z0;H~a?>5kD4nZrhqnES;Rr<tn(+x`Ek4U$#L4RsRY(j`HR-hIp=_65+9#1O&0tLlS
z`<&YdP5)~lm~A+Hnw?-syn-uy%xXj2y#do>Qu|<zYti0Km)z`t+0Lm#TF4#OQknpf
zzh0oHu-M0ENCv|_N?V=6u9&e?32cFy%|h~Cb^tTI$qrK(xeQcb5o5c1fTb@YzU}wa
zk3G=uruEgM4fm7LH`1{2Qcp2MmE7~eH4JqB>c>(733LOG`s5q$qEPCWo4|QEgnJ(`
zH{&xs9H~hX2A+IOUjzJz8E7gfEdrEk<?A-{@*RoEJ4~Q8f&8yv7v)%vkRW#-Lk2R5
z`!nso%*k&wy^_9Lw>Q5yqOy>nkf_he;xLKsWV#k-`%Wr;*MSv49H7tpg7w9;T%B-3
z1X!2yPU4qRN<|aa!KriOgo$4mHGGlu9)6U&BMoI*BZBIF5R8WH5?+bR-^+MC0_Gk~
zv2ElYDYCa{{UScYiSNJ(E%`_aH4eN(?DC*VXv8KIX9@qHEh0E4Vh-crwVcGydV$>3
zrCD>xAgx3}P#FI{aPL5K5oaf(jIAc;dcVhZWwFJ4{``8snQ5dAp;D}JHqhsXN1lO@
z{3#<ia+p=IQqLaME+FS~Ub3&c2S%WHP>5QR%?$?HGT1Wqq*Kpb%hiPZy&7={&x1Y^
zn%-FymdF0KBk`WIGyIe}P*TsYWCVwJK;{FC4DDOLc#_|hIP><jWv#QU0VX{wQ@(@V
zXGw%aN&xwN)HX->O(tiaz`ajDIZW&vSY2p_kdjT!E3&hYN+vbYua$)d4qZd<E;-9=
ze~iR{UijHfHLcAwaDh(Q+LH^S*xRZzh=x56n@jEfYFWE&^642;P(5HGH9{O&L7xZg
zjH?Qt2Xo%>7HVN5Z~vAelXfRW<gUpjA;(B{TYFaN73U1!Tc7h&#X_8J@XA&uSfU(l
z)woM|wF6S3)VJr&+>N9r!wySrlhq{9OXb=9p3oa=$oJe8Gb%HQ>%`@<as+)_H+|C!
zp|&I85s6B!<Aw;E16{uukTR`sLQQHZ{LwY_o2p3b^{8r6*ui5d;&aSnFloz~L2_=~
z6IuGS#rNFso{59-IFL9v#$`n_hyV~QR50BAtpY|<$EMe>q}89*8f}S&egcAN#U}01
z6+fZEN*PfS24I&F=U=@)Kk(SFh`{N8vG<l?RjzH<sDv=-iAZ+{(jn5)h)9DX9Rf;s
zN;imfD=i2pEiEl24bt5yjneR4w@cSr@AJm9zP*3!{o`H7JUoD<%zIvWp5q+n7@RW!
z9T1zE6^s&q3GzJCEwYHUGUPEiLD+mI@Y>PBzj&}l31Dm+s0-VhCKL4#UQpPM`u14D
zwe_53<^66DKZ|>ZdiPyNPyMNa4?#<*vM*cl4uoC2ysJCtR=+CmSc(kB|9KF)qYXbh
z-K{FnLA;4LVl~6@j7}60dMQPVGE!5qCG#7;6RwKG6xW-p;NsLDB5ZL2*^}yFWc{_&
zuKc|dKNr{Ci*CwuMK|0jwE$aXLVU#g5r+MGa*_DoSC8heHld{Kv1G6Z47Qr&Op#PH
zZH>nuoZ>=E9^r?Bq$StyelT3Qf1oJRRYc+?huxb^fU-JRHed5_5m|ZBdDfkw_#F^j
zK4BFeIJ-HC15$qpqIx;zRjQ>2HwwhM{ZS+Bu-Wg|T}w}%&0pQi#9(jKUtR|yMTHOf
zyi0w8Ooh_@F!u3mgJAiIFgg-F?T8|_Jt}_>Z6qh%NTd?v@VCkMAH8p46%Bg$hXi?9
z)JaTTcwBILd0mk%^QN*bG1BuzdVZsw1^$Zj211(8m}F2lFkKp{Up?G}ozvb6zO#?y
zWfL44fw%vTntVG<B#YSe=wzKiTUo6(bQE_9b%JjvdLPNOxeK<lU~s(X`6k=qj6>Wa
ziha-Bf43wS4nJyjgzC&(K3rj?e*0?&_{#^CPUn>`bWM|*9qUu6(maS@l#YVE?e@}S
zN~uh?x2<87i4uJEp2rU9m<FZ!V;9OZ?mGMPG`#~3wD)c|%-kBbpTaq6q?EJ});SI<
zhU1jI!P*)eQKr;u!SNtnFgLuh)fefMeAM2AhuK-peS(XHw|s*=D^j|{xdqJ_?F>e0
z>-%{x>vAk8zzIzoj=d0e-bgu7U?S{xrBS(8fo=PvmP1%n#McgYD*`SuWL);4{+|5H
zcF>B&R;(04h$=I36kYhm4%1frMKSC0!g|OV)_BA_OLO<+(y}DsSMW*h@3fCRjlvq9
ztIP1UcA|75w|9BLk-z3I`iO{s@MJfK;@mx_J9$%&?%bDj+ey=l<eBx-C-W)x(85&N
z6K;6%;^c|s+1K+s0S%pD<~CA>M+Q~T*Um-Vf;c=VGAU7=&olZYQe*8+&OB0Sgl5y;
zMzG%PJ)A$~ddYWeOL<I-Jr+-bL!~&+-UZQ+Oxs=bl}xK81!PNhg-R?pR*T__2zu`t
zQJ3MSMGmR7otYCJdKApnY<c%Eb)<1|V%cMDiyQBFQ^sU^;xpq>cf~!AeiwdvOp9eg
z`93A5@%sbWkr&}MrSFgo{B6Dq^}c#X>3DPb#xkM@YA5X}me5%SRwI8^)(hW7>U~<v
zD^=z{=zAY1hr<h$!uz+MZtAEc9rPW};B>OI5y`LtDj(z2!|)6_+RNiEz)`C3Ja+Ll
zov7zyLOz@~2<3>XFw<6DL7m{VVOeEz_3GE2@Z8OP=E?tvT#7HZJ+3N?r%{jaeb3>%
zt=1dl+sOGn?gsgP5C<Nlv-sG7Mr#D_;-IJEjVSe&Jt&l#F!hgUJ#bDIgIC#i^oPG_
znRLO)*l+o~rdH}Q{&a&{zsK8168w9!)NA>=;zJR}Aygaew><MtS~z^^l>Cs*T;-Jg
zs)%mgpz<M5@V0!0!G29n_!o@NP3;5ThyK+KYk5SveqWdpY|NMIjm)i|H7*lf=;?~i
zv*5)Jo=r8dc-1K&0=fir{IYG*x!F<+Xc5bGD;SaU-!`}8MvIcey>E%CBe&SGuZ^yQ
z=V`{x{mv;>70X2jg>fCwTRR0}`uyTZJfOM<qP(vk7vsReHjbtQOH84b_)S-)r&bCT
z7fsxsb?ovmJ}tZYa)u|;Z|pqz;HdQzaa!vOhnNpmb?OO+DMd0acfZRsB6s)s0Nr;P
zwyHVbsp#wKB%46=BuXJgK&H}adI_{6!~jh<x+|JX8Xo~FB^={3(3Ac~2vm^n<8=|a
zgD%Wc&}OT7K`QKC1zPO597SI%(;?h2F+)^;h#U_!BwsCO?t|?#F___TdKk4m{R#92
zPtky81)FUlXYg5>7?f}fUv{cWq(R4Ye=+jdGaeSbK!u;DHgs^OUYm)qENLa+vYg;f
z;<L?UU9X3FU<2;GIE(}>%kstkY(FV#+}vPPNMCb&X86V2MxlX*Vl{b6;9P-NWP3Wf
z@x$W}YWMZ$*)Y0hQ7$w9^CujIMrE=|1(uj55<@Q^4a^X<pY^3$GAKFKN+gW~c+L1T
z7VlwD)h(YIYUslcNK?iET#8s#??`C}=##SX1LfQ-aQ!)ceuE39-b4V=M5Oq9=Xh_?
zJ}S6^5eUmPU{VN;O~faoZiA+tM{?;DZ8z~50x4tw-6h=zt+>ZRzY#O__spiVAasY!
z(Qbm99lw@xb1*v;lY~3BwVsk0BZ2}9mhX*pILB=Tl>5I%CrmmW2)sVH=q8@+QWUv-
zbnPi<#_2pK7raSk&4>!vGJU4*<VT>tZIvSUpzoyzF)yf>x8NM`q74c8ytq9H?!LF(
ziJWh}Z?$+AcaE!EpS&*W0zDZELfI`7)o8Mb=<5XPRO>e*0Ds|;=+B3z^V0=S@*rIq
z1|I%qcP-kV-AcrWtt&E`l~z3wMP0><reN3DfNg-;9!>p={UG}GM($&3{kP(|Z{H}x
z)TNVJbEALrB7U_z{deDrrr?5NKrh#;fj@Q_6@sF-xTs>LFreFxNDFhu|GMY@xb6Rk
zKb0b&t=X<^m<&w;c!{E&GkshlJV|2N@E0{tl3j`AalD~*Kj!zG{PDN`hd<L(!RA+2
ztZuFP^<hL`dE*fZHm8+8OphyjKb*^{`%fQE6wy%uFrozdMDM5l<M;pjqoo@|H521i
z5mm+C8|Z%+i~sz;lvbh9IlweJ{53j%|K<PSYfUHt)pEML^ZnOHR^mm5h>Ryd3&0qt
z1H=RAj!LVAp5T8v?_fQMN<g0}1MQCcucVgBhYGm@Xn(kZviI^AhP2q!du}LbF+SWv
zJ`f@X)XJHAt83rrI1WGN0+LFr<ckr33^qU)yvrXAZ{NzlrW~9b%iyOI4Dg;V0LQ2(
z-1|N>5y5Tv;2#WHzdjfu3yw0(#NRVSMd$NaO+%_JoSYd*%7h<oPYW9LW8EO|w*qpW
zbq-q-<zmf638D;&sVtB{lRIepCs<Y#K9K}y@Vz8qkD4;e$qL=*0gDBGi!lyY;Im-2
z?g}W{KaXFa;JOliw+(w!+No2R-GJ2*9ECBE&a+$kUpK=4Ff-{^;D%7I)MyWm`D1U1
zcF`1r%ExP|M$oRS3)OmGE1Uy%0t*m%c&`mws1eQ0S;u=`-<1VGHT}oueLR?;QQi-e
zm`n0mZ!#z)<K28!RDb&LS$AU8C;QoYk7Xb>8hg+y$PS=WXLXTg_z%|;;1%=uok9u|
z)a;+=O^p8gjQH=iH#E&U;BYHu`xq8Q|LXvcm!1Pc(6K;gRVH5jDVpimck=Q`iQJ~q
zkX0Gi(-zn`SWV^vg<<5?Pz$#m(8p4Rjg8HPPFVxNx_F>^)u0^UolhynMY}u?gz6sa
zyT*t&#$I#%U5fu7&Kzz$kU;RnyfWf~&WN8cv!H7M3atmefXrs`dI2-QTVCwmw|k7p
zm|Za$f6}}D+g~YK2YozU`|!PA9}nS{4X6Qi9Ik&2YXu%N7G3>7YN=}<$f$OWl^8CA
zr%wRY3+>axby^_k9-3dZ@G@koSZ))thB*O+^sRHCy}<}oQ}jda0dp=0!U!rNvLy`$
zEP))+1I$ezEmC~TR?p-6;}cLnGzNOZaln8_ZA8<Ym6$`{yBnyTg^Zr~fns8z#^6Ci
z3?$2LIZ)Pk1zIl|Ah8I*t-6CK>-%8zX(%hbd45SCCK;bnjnmfan-()Az>G(i0%$0U
z34q3eyNk6zVf|r|Zk-X3JHgAslTC({NvbzNaYrqQ%lPFv&>@R4_#F5JD19=5d)9rs
zc2>hEN(y2J^VDCe7g#XJB~pRi8X=VfZfQF}M$=54C*rFG!tsE*Ak@0w4%!zIx`9R-
zzO+aI9DZU0gh0Nugy1cM1o`9s=*Q=ZAFHZ>nM?x9la4Fk5~2*=iQg5dlCjKsp6^Mo
zj}*Mg2Ym$<+-%6U%VdSsV8s@&rE3Eq3rDR%P2J!+|C~2c{uGU#pwl;X?p&7j+axA|
zc%}i~rh@}R-xt9%#{$0FM10=;ai3v@GB;52tVq9Tx$((_mh#0KwYiAWNMt3%J4)YX
zsD@k0t5gk>g<>EmixkytFp0gWGoP)miyNr0oII6qnmG^^@%gcFdA5@Pwdri2H6zf*
zu4D0xQ90|Bq?k=!f_~PCg%eQ6(2QIo?N3eDZJWNjn7#S}O124t-_AZwAwc28zkR_1
zyjhTcpTs%HRH@#%Sk3e>VMqi`!>W22%o!Ultcpq<h?Shk$D-Jdf+?p(rF1VQ11st}
zVi}I`aU=3$IFHyhUc)xb6Cjc%M|H=fpUM^R1?1p}87M!VXc*&H1F_lb3(>iL*Y*I#
zE;dg!XW~Zv^e1Ofz&U9}CWCue08<;t#Bj2D&2hj3hK<{nukkT6$db~R%^!(#94HOw
z+EAcUZTKkR_|{U~dy(jk>U%#6!1yy!Kx?2QUMwK7IPnb7rkB$Fu({^PAcuf?h=Ci=
z7CbNDxM74z9q0~vGwhqFaHuH6rws7nzQuF{)P=xP>yb>8I$Qt!8RB~{%E&wp2h^||
zNqjp2sQaWXjP!XX?-C=jI6TP{NL8y5KrLy}&D|z9-s2-K=mHLLrl=&3+YbvoHK*;V
z@lK(A662@9BQhEY#dmkQ9qM{s0eyxjpqbQr{$qQ3QZhsR8OhUe=yaq)_v<%7({ZXv
zlpvopjJRhf{$S{C3lf>qh|*{y#ypj<Eg=p09*4wvp&Bgl7C9&?D)CY~`FRssihpw2
zS^^hGIoEZAMMz~`ybLT=MXE$Usx45O<m8;AjS!+SLCR#4)WXAJ-}2Ok1q%!`czKyO
z!09KLTgZK_f}897)<~KKch!OHj#2rr+Q%0t1Kb;Er$oWeOQb}uA-HfRyxVHGYcpxD
zYb*k60Xr~uZ}f#c(`du_#QOnXghw8VZZ)jyzULRVUhMJy`=wT91P;uKrK#r=e?$%&
zudgBV!)+QLjw=Xm8Hrc&SWYC4VH4z2y*HVzDGwRculojwUw1a0fJ#g}84#*)H4U;s
z7Uzc*P*n;15*@Gu{YDcd;0Ejj0zHO^pMff0BQg=Kba*$OZj%Zb%cYoq@TR^gR{R27
zF&M+TcJF>QkYSf|Knm4T?s_!P(TGURQa@i)<(FLtiXxLN*kGa6?|r53@V$l-pBIS&
z!=d)}mmr~D_-w1L?9X8V;t6&b(&~>{FE2dxW~K#D=YMV>ij3w%$tR|cGLO;1wb&lf
z7l0r}w2d@#ar*s^@(Q38CJF91Q;GXNz@5~zhx>@n1s!H2iyj9FkqfyLTY%e|f79!>
zGjM2!2VbL3_fvWwdrK&-*DSyrste9p2a!8PHe763s})isi|dLOyMl}(@o^<25Va7~
za&zRG0VAFoBQVDk!@F@}hn7J&h{lPZpNnrFjI@6N)MHEcefC37QLt}Y2Tvw)8ZH4`
zG!{74y{Ae91RfZ7qYro|4d?_za>TQw-$%EEw^*C(G+skAQg@Uj4y4)!UY_Np+mm82
z=eJwe2cydi&>}quB(-5MF{V?0D1%c|gA8)a&8c=nHFx}pegN4)P9Mra90NMZM)+z|
zH1h^ba{4T<m}Eo;h`YGr$+j(~RBRoc_|zjde}l|2TFT+WZRDrOSr{~jfIA@%G+i<A
z=-dC<vNAzD^$iOD9crcbqOIL1Q0l1$Op3NV{m*&$^|d$iOgi&yTJsuP*GtPP(pIVW
zB?2@F;?dDkUT!C&XD6ed*a;8$#_JTRW!kmFg%{FC4w5xBx(;^k<w@Oj1{z5M_=^1a
zizlBSTRx6`YWoS?ICw>>)S0Ro3172WQ_v^S&vRQPatLmwXnNZ--L_>CVA*|3L!Cpz
z?hI&Af?0RBg=F!E`N|9NMJyYJo8wwl13+u@=KLHGyMJz~G?1dYm-F?O5=Lf0Gg*%B
z!!FfwA~hS~b#Kk>k2W!5?H3}T{-{vEAW*Kk@+etdsqk^t#v}jd%QDx~2ohX~jthJl
zLcPkmUwGY~tg)zwL-S%W#SmfQDsyS+qt({AKS7Pn#P=%ADPvF>ZsUz)M*TG}uK%St
zdR+LsLQ1A&d$U`8dF6tuUC5oZ6GHf?0xf-l_}H4Rk8wYW7)^p}O4pkhfU=pEL8_aD
z3fDsmLEEds^>J*=XZ%5%176yI1cdD<u+a$g6;C2ko(xzX#%=|fobgUt+{ww@sx7vz
zS*OqO{ZU&BaZpoUMyI^Ib8x<o`tQ$+$VO{q{dJ$e&#b6_&InKg$YD;p5gbRARRxS;
zLQfCGMa%n*XkJz2a;%F7U#E^%2{pU^v>?|j_zRQ+<fyHXNoo(*sDOsdD~G(NIRpiA
z$B$VovQT{`cYViLH>iAlcI6Vh9pPB<b~P;}XYC@FQJ}OMpb!%Eem|WMkph@NDHP1$
z*O5#OtE58)2cf4Ssk!%IxoRvanv8dBRk>ds*R$=IBNpBrQK$8MKga7Y3wO@DRhA4t
z1R_Cq5lnqnJA6QBqcN?#>7$VFi1l`Z{$cl>KDziffL%gomgMvS7cuamwCX4h4D<`6
zNN5BLz}$|2IC*WYlp8`+vL>wC6>DH*2@#`sF~HmGWDt2A&Mp2R=!J5~=kMUE$yddA
z6EZVEkm(`)hR~UM;u7q@@d>Nh-rQ_abHyR9^bq2Ud(_t?b=RW1FX5H{B-zA8)Fx4%
zr62g6nAhPQe@(ljQ(wWA<)&v1+xUtqk|O9_kxju~fBe*o&l<9a?1FmK0Nb(5n5u-u
zg5P;SiZJA5o&~WCOPyvXwL42s-cu?Kpvfug%F4$Jz#1(QHCDuj?xy-Iej&>SCGM+0
z$WG73^3Yq8p=>!oaUff_<WaG+Iz2(xqxW&2e0S?88(O*I%vihPUw>~ucN)x+$!zA+
zdhZ0nfFD7dW^N#D>H;eK>`-LN+N!~{&ji1Le=xQWF5U9j56?{JfE(t6NCrq4SKo9+
zW&7%K=PLg^55K*3S<x12pH#N-ebqp`cx!Z)O)fKtjjL*b>2GExW7FW*R<eE4u6Hva
zD9C)OtVN(zU_MdW!IT*dN+M(NuUqY=Yzdjj_iVsz4c<m+-DKm`C)>D$t`ME~0G>Im
z4w5FzJX0_6v^V+R4Dz6)WePjU#N7Rux&kyR$f~>*Qb-jVkN1Jv(RCe8z8~LPl*MT8
zTjY&jyzl0^5W=vMf<5p9kd!}HB=f}Ywv~3P$7>fLPm~@4uq^o%DQ|B)!xiLiV#yTG
z*j75_@K2`r1f;3mt4(^p!@mS%DrEm$H1U^q*y4?s7iXbtj!A377zo1~aZn@`;<<oI
zJwyV!!qA#J(NKY)9va>^U^uFI`hAlS=l#--ig+Ugx1nH@-L7*x;>l3U^QPKX-t?yr
z9x_(8*Ns9c_cuYz0CW^y=h}&scLV~DNoVP<bYU=hvsTV8vfGJ-N9+4_^h4b#561R9
z9T`VT#c%1nd+d;>svMetM!ifqc*|2hksqX$wRRBD7NY4A3p`S!9r&L^aY&Z=Xklj8
zu_m#1?t_GAHfH3OX{E~{lJr{O46_deV>_RxdC;5b^e^y%9J~`-@EesGf3av85e@lC
z9@fzZ5^DWi2X4oU4kl#Gff4^MF<jUQmA?*9+AI(ffAwBXuOR=4T^fscBivHNzh*7r
z@}$q}^#d~TLtx^nHni{_w6Z)(g>E<4kFCkkjY77^+A^hULBH>%-#Ox+$>N`1Xp%vQ
zh<5y)rava4C+eENcX2T)H370rE19J7J+~h>b()`mnBz*)+SK2>#wZt!M_*Y!VO1V;
zloE2s?3tj*FAgY!L2ec3e4@eSUljyTCgY|y24h?^AupO9$>;E4Od^i-rSfcrz?%Z=
zH#(mYcg^C&?{#XlA9#MWsDtl<^N6JwkpSlixS)ItS*NOyEe$68t_O?@3D5wqlNrDf
zM%=;CX8=|0SVS6NmdpX%??Q;Sq$Z3j&4H|?jw*Q)u>4tt0O3O3F;CrR4L&-LqsC*P
zvkjmfLiLwO?CQ0zYT)a^frNGqKS$d?pVCy_2p=)f#C)*j)7V}3ejP|~)ljMBroo_g
zKTT!JW+t^L7Apbf1|MMv5@i>;LCghZ(HBi5%%{2dPqYUBD4GAF1y@>~6uD#$2yE)d
zquEETV*iuNg%X*j3)w#5_xz+Qpyjr5=foc;v-N3mUuUYYVY;*zK(}FPX5LT?&|HxQ
z>Yv~7L^3itKS3c~NLuGv7~eOL!CC=#2b+)2dO8K4e2IV(8xIH5Im8{XSN*_(Oi{&6
z|LifySFW<U0<<}u=X_A#RSP+Xz2ss|1rc8dx$Gx<z}u`A_BdU-gq%Ig#2!1Qu)Q@m
z?a#JP(QIITQNT962@;qkhrE?c&rAm3-NuDMnBP?m9Cbgh-$++r)rmiN?Q#j!AZ^Wt
z2iK!B6<9^-Pn*Qb(bpZSXERqF$Mth$3k2}5(VvQGsBxyN4n%)7_Qg0il9v|qM!;bK
zo~!Yf4}CG?!Xv-Gjdp3j$ldM9YwbHW$m!~@fuQ{zzKI)UNuaA{+a<Vyo>1L8MaOBp
zH3P?6)%N|@TM;SfPgBr0lG#3L{lKo$O;}=G$N8=#Re0S~y!^t%9``x^mH_^`Ksjv9
zVcju@Kr;osZ5q(U%udoru3;=FeY>`?1+Z|{Bh=RBW!vBI-S0M_Up0%reVh9jQRM@!
zVeOy0hlyb>z#DF)1HzDPfKhdSBJ#bh^K8K1uIAfQ8C+cwYN}?CV=l|Pa2a>T*1Ac1
ziMF5^5J)~Y)dtnxW>PMr;DDkoiUdpwM7;8um4&C$!Jbq%Hu^LCzAitN4MZpjV@ukE
zA^=3Mnkfq(m)u6x0Mwz*jDYK$(V8t>O&S3S&-Pcq6I@%r(Yp=l#K9C<zbAQ{L`f0Y
zn%ydxb073E^Y?~RIc<*g6aaK#8{8`o(Ak_ZyN8)Hw|Sk0#NU;untQmmzZHq>t1+^X
zp{8n)xs3<Ehi`Aj*#CWFxU0q6Rz%|~E)M(B4){%=@xRnF%ODDvKe;`#{A1tPQd9cp
z0s28cT47yq;aE8zMT77cYtSE&TX6-vt8C5kz5*K;wL$<QDkRKTn5^AsS4gO<tRT=T
zpGdk{=W7TxeSXc(mzS&pxL?is*Brv*`M(YNf1ZXveqv$;0i9O-orizm3}p`x;fUg*
z2tL!k7;I}iKU*tq3j`Me`E&J304nk8pp0$-`B{D+shc;EZQ&<#09c_Cc*9{JzMUvS
zNJw<bL6z>O{xIQ|lgHT)M<9ir;K$MV3ABd{Vu2co<N9E$W&Oztwr@RmVKN*}69tU5
z`qu4own3f77EUHb&f+RIhk`)aq6Y~-AC_$?Ci4x5jiG!90BBFi=esjvqA>H9#Z;)W
zuJ9Fr5XxRUSYc3OBU?b!@05sAQ5QH41+c}$MvPOzR$BoHF7!4d2)IFQHsG4_QP`QO
zlV1wq+Z^mtNEmkny7)Pq$Vf}~SfX0V!%s>VD_c9Cs3eJMI)2gpI0(bA0(mIQ1Cf2C
zOcr(dL$v{Sp-7w#Y4sOw0DH7UR9^~WGqi9-UcQEy6;3XwsoijSYVPG~wW%Sh?f@_^
z0+^*^o>op5t(Go1V5vi*T_X<1Hl<e7f%_+Om2ezN(3O3?{q6G1Ebk)$@nIawBD0Db
zheC}czs4T`@#^|?b6GM0xexD+Yp6E_RGfj($o^8I>~3|V%C@L~kh}kIZIzS=M<D(^
zb(w(dd$9(`QZH&HD9$#OD@rIWKg{7+3WvW9PoEAQjZm_@0)T_%))YX079G$ww1ozo
z(|5#Y5ldXNw<u4A4h@g}f=iP5rtZwg|2|Rw`x|c~Nt9VqS1+>vPf&D-69K<GSd$j7
ze1(C@DkdlO#luHybW9Go<%M)inlXSjqjP$+!>;qmiS<F=%Niq4l5)~+y_GpKDwz;>
zQ^$(0Z|B(?OM0f<4#5?#bAV;;_e<b>kzq)oW14~E2Wuod$fzX=xt8VzFlc}qeHjFr
zmBkm1!xp%KFVUnDPzZEh^6aVL8;590-NcU!VLM`CI*gVg)Oj={kepkr@Ti-X$tenm
z_)9zzrDsa+#*hwRZNpi=CNZU0U>1U|jBMa0%)}if=7$o3J|vld*NcH95{bivjaOb3
zZ<ry@qHLR)o7lFH-eu6)pCN%!_#M?D&v+c%ao=*-1euwhF$WhNk-qpavLkt&f4_~?
zI{2lJ5s*~e2WSKd9zL-~&4{j><-qbgDb^n$=mZ#<>Ee&7fy(oKxSQwRes-77OY{rL
zgpr<uqOQG{Y^Y3-=0qIxQcwfIGCIrS*jr3SH*`p}9xWv~UjTH}(Do5$RIcdI{@^NY
zXH%3k1!IkSt~S_gnVDifBKSO-<9i4h!j>Sg$fQvnFt#kfW-T95QL9ucd?aM%h;pR1
zm%CiqFk=1j-RGbfr_i$^U~8Y5Ia^C5d_0GwT#~r_{LmlWsi5$Fgn<ZHm_rdRQ=J!T
z1BCz@?w>$h@;KU_W(_kLoqJ?L4;=QQ$`fw$nyBvO`j>_{Ap&8E2QM}6-o4V8B*CEj
zETV&=Gvj_BuZ&xe3s;M*&lLn-3CjQrFXPUw8lT`G6Fgme+e13RR~A#HEA?qaZDVlD
z4#=`E`SOF5St?;2AAjoQ%lif4Er6R;L1~FTVS0Ga<S0_nW-j7#9iZ8G`1zXJc2UeF
zZ`H7+uCMb`KGE|i&yD0idL#JnBj|oIP&gSQvyU^S`FU8H*do`$;hEU5o!y*~RHK#d
zq)dYNZ#XpJARuHYbp@qx<Ebj!kkw4o;f2>B8JX>wXYf~I5AK)^-G<OoWE$G?HX~nB
zSSS^l-xw_$$E!ePGfH247_lrG0yUXW;t~M&aCIoY`fI_>UB%KgEAh%xZ^7F{vl5o5
z9{L2|Xuc1dr{eRwhVIK)?qagg5T~a31zGQbUbLT=&ifI6o+=6pxHAlo+Q{#=YynFA
z%^!K!r3bP(SzUr{sr*e@R>3XyMeG`29_nc%`9@;CQZ7_VrK)&jZV)oyyS4|`B0(^7
z7e5T`Xwq5)gLiv0ff<H~OBozZ)vF@=98oM)Q<oco`nXLhDHpXSzXa_LK+UXSmqVv-
z^1;zoA(f~L!=R?}mmc7re+^MFl`cWu#uO@J$@Qbr!wDDa&En*h=!^Bbjb)@uIGA;$
z+0TA!0o)50Qj(2TV_>dESx^`nr326?ki}A&dPx5H-oX7AoOGRVWJ43ux$W8m1w6D&
z7V)`_t{W|Vy65O(cMJV621e9H8Lu+PGQ)f%+ycua$tUz-gl`oJB45wpw%f2a2iLgN
zzr~W)@-Az>HH<@cWK1JFiBm$^GH7ztujiW66&}jgYPF2J_n{?>TyVsH&un}Ic@2K=
z{cztywrt9O5VdGQWHZYV&V^sAY8TZf7rq<uwgK%i6Yg|VCLW7;oISjaPQx=)7qtoI
z9}Ewq(w?Fmby(}CR$judtC$jQ@qq7%!7G4laO`3Dp(rfGMYIK}iDDr_L|rD|9|Kte
z?n6xuSLp#Z6;7-yzHy7*9ZefVEl)3Bv>=R9W5eD)B6C2K!^0AftZE}MPfSM@Z)vkr
zrl!gd%eK9Zq3iH?`$h5XIAKbHeLOE4IIZukitK$P77t<lSv|iqYDHzWNNF3mt81vW
z)7+64U&N*oMI^np-}!gfm-{+Q*g(Z_eb*psng%Ow$lf34U8`gvA0-VrMB0mN%a9L7
zSB)Cg9Yf8a>Xfc!p=pTH)w^0e0+2?_=lmt!iMx7)(F=iPRPxHGUGH?4Na|+J#E2u}
z4T}-)?7SStNk)CkV%e}7@q*Zr)I(^|_Q3YxM5aX(7i*``#97tyXqLIlZ+VbE<HG;^
z#YPI~v;Cgx_e$hfdT<-!=d7Th%I@g`P~uquyBX*E>TJCACayGfPof2q;C;}H_2$0(
zGZ<hb+)w2FvpfC2i-&&w+=oC8l;{bPPviuCzLV$yb^la=H?Rf_fehO1z>4ogA+?|a
zH4*F7*4&kt7_Tg3*~D$o15^so&++K&;{K7uaPh{rtRA;NugHH>$`R!OOYGTV`-fj;
z`SdR^{c&lQ2*KZ^7_AyWiqUd^akxee*vDF5ven+OEErq4gZfuzvBBr=8x*z8G!*d-
zEnE@6a5n?830?TAL7!VAe9z+)XoePk%F=7_9Q^R4=)P<hx$_h?1k1B#Ly5Qyh>Hp=
z4Dx!L{7}2BK5dPd*v^Oj=Lz`le!@g6O0934ycYT^axe?OhHM9<xf1}%Wm(c5Zr%hU
zKdItZ+hq3L%JzF~gc&k1^j{#maH#fN^k5Z4o4275jx}zw+<f$zH5MFAy5@OrG9w6h
zEr;BbL3L3Zz;@u$Z&)hrO!xqJ++Fb*y(x45BvJfT6PJgB;w*isPt@;8c0dL|gu%%E
z!V|!Q(iCv|MkJz5*kG`}3u<z`A*8sbS*Ws$JU{lU91&P-=wgn~w92e!XY!FwXQ%!-
zwP6SaU<2l~ltdW+{*d`ESLFjKuoNGSx#0eTl=#0Y7yj=9{O<_-|6l}s?&Bk31Awg_
zG*e7G;5QWe2Q-}j%v};i43kU;IRuwcM<hzM)?0Lh1dV6JNdk^$esumm|9n^ehoe}u
z;D$Odvt0stt$F0yUEmJ`+~A^#?>ysvAWn4q|LL}4@&P=6PQA<h1e6e6@SxO)$@kox
z{@*>O$?*F)a0b<jWsc?o0;~)I(r>2R|K;KGuNT~Zb#VNLG5VLkAbJ7uwJr}L8UM>O
z>)-y+qzS?unv(jmeleE*<@f))Z-+#{sk$kGlZ5zx`N99dME>{V|A+AA-;MUau7v;Z
zXWxe&9}()X*@wtuS!3_yNi~2!+rv{CffL}7)(L7IF%C@579dWG2guBdC+}e|AYmE*
z_%<$O{jpyh;|t(bX&k!WWr%d*cwh&Dj5sL6OIhIjxtx3iXq03`NaTBgUI#O|fI~2a
z+2IAiuv8%WA4DQ%RDIV0v1q1i9Ag|uG^FD-o+aInP&Ayi0VyLF@ZfuIHCb){+;sT;
z<Hw`@_LjIVfaS0RDAE<Sb3l=f^+SH?LrW<R2@AlVdp*CgB>Oe|Dn`OQ@of*kB55Fz
z#NT(d2`1vvtXD%Vz}4+Q0*bNX{jnzg5VLRfdwzDdaFKOypOd?88f%Xw_B(A>iPWo2
zy{l+I@n~Mrerq4u%ilz)?LHe6>KvOWV#RU<$jV(HL4LfqwX5I^=rGp(u7fNWw^wR*
zs#hsaz&~XQX8uzQ^cosBJ~YsfW*e8?Qb)9Dd#GqFUx6B?ODAJa2apX@<FdePX?sPC
zXQYy~6y_;X4LTRmBuQJRz_hT)vQ-<Ee0@ybE_wwD%Zdwk!F;6wrZnFk8O|Rk^$?EF
z!_xs1Gb;Meq1WU#M71k^Oqh28=OY&2d}K1D<+l|<a`&MEC|e})SiaWJMD+^BJ3R#@
zV@0U>o|wlXQU1|gb-LGlbD)AM$5y=*{r)KeJ!BpQjrtZGfbpc@qM-Ioc$jF?Dt#M@
zNI?NH*NG;Hi~fP41k43;{c{CAFRE?t=h@ABT_<)AT~Pmar1%8VK+#}l+T|g`*&4*7
zhy$S~^Vb5+i914Rqj&HVYHF+aMJD%L>rQ%_Vm}yxmG?0j5UqH2P5}x8>pTJ(zs7cH
zG9R1*;)p?}@144Moq$>bLDKlBtE*9*l@)Wo7!zbiz`gUieJ8lx3iy(rQhfaJcCH?P
zxte-H4r@x21fK{a{yHl~C2p9gP|pF7kolMtke-A9iW0Kk(umF84GAH&9LS(HoNWsN
z95@2vOv&W*kPHGQ8aotYi{2cqaom*j%D`EWjSRtK=)60sZ8+(2j7KjA`@uD=I8+Oq
zzV;?NX0I-tt3l-rBmCfx147gTcUO-UGGp3b2xTV-6QY+q{jtSN>Avd|(lSt)x#M??
zE$$|_n+!^NSZ1sQwvu|6paP!sn7akR16NY=Sb(q)d^*4b3hKB{sIA1!`rUH@ICaMY
z)PHIpq$k^4oJK0IcO98^Cs-QJ_e(*h{Gc!OUQF$9OGyyLiw=4$58)BXEE3Ke%YzMo
z1=Lw(ySV=p8DSfElJ18OFGfIIH2>PYD)x%8ry$Q^Z#n@uB$MZa=MKos_CcQW1xWtT
z7QZHUG$G_*D||efd{CyAbmu>Cx0lf%OVCoFtv|*`rgp@+lR9GT8pwY1Dwan9%R|TB
zT^IG^iyqJrsKd6^b21*@x9Js|zQToiApLUxh)UPb;AX9J@?G<INK}?fzww&bn;*bZ
zq7(4L=!@;1bbkjNvK1<~!!-l@NSzu7#+AFrd%zP}n?vNKt0j?>Shc(zXq+g{uU>i0
zb9D^Z!{ZO{sO$9fSsPL~0z<=S;JnEBRd6K#;<lRUXZIsuOJqPf16<znN<lKpa4V$!
zuN*Di0BtQnOn$}lbe#ce5D@f0TN}~y@)*lfI9F{-{feO9x>pN!+%(9_qKF8#GmxI<
zu#=^rqcCIN@t4*?86ONiz6FdVTmq}Y0`>q|^%K}GjC;bMyy`@WBKb3b>vuqQ7ZGsX
zH_AagZsE~u2y~$IJkI#;2BpfMz!W9F+}%Haf30sFS=sMLsM!U}PAbH9_-gzu#WU}<
zGuhs-%!hY=6OGPl%)tr;NP_s1HDIK8*tZ5snjjWzPCOo~qFGWq7?U}m<FE}d<k`k0
zvpdIj>+LRb47V`$U4hd;K@aj{V&;^&ZrL~JYAK$Liw-IyZAYgWiZdrB#L{m573*7-
zvrY^t4$-R0cD+R}VIN+}EV^26yH2>!xib8HzxN|U*11!OnopW-fzg@vhE|rx&&d}M
z53@0_chf~JebF0{ub~Sj|KjK5*P2_f*>lJ^BJRjL5^LrV>?eEfZaG}%2`a@}o#gIr
zuqbXqslKEZ;PtB)mvT`^Saqz44p3D#7_=n$MzkH|s)CPP^BG%^g)=<3NR!v$K|9zD
z-)4aITn=TKpiVPiUqhkq;-Q2~j@pSzJFlb&di-hBrykVzxq~UduP$%#Gsa&uekxW3
zcQIwWZ9p*YBQ29cee2@$zH3G8rXnVpBbtAv0kt+}%j?0s>p3mQ0id7S1CGs7(sJYa
zQ6{{Y(LU`6VlDdwQrmz7k+AG&?Eo_Ut|b*mPNS=DUkYK<h+O3z!jZde@RV_)5^7!d
zeS%Tv+CH#qxR7_NN`;0|&SJRcb357T-NNNGHaF$X0W6A!hl3Hed7jECPui4moTeQ|
z=k`orUV9FB<+7xbwZKgEwfx*^mHz%+&7HSrm$Yf)AD{oM7Tw!zM6xQZUg5fe-yC^w
zzU`{{VTzdj?=v-s{7x=uVK++8p}P*yo8#+qJI2fUcH@L-@vDt{$P15F9j{p`0A+oI
zJi6a=1w(k#MJA_mk8d;oqn%{~8`3xZS9D+S-L78Zl$$Rj8I`vsj0Q=a-GN<yi|wfW
zB#v>nNDVQiurQjJ7r(A>$g{)qCS?QgMo%ITaX)^EQx7$A{d%Qyq7A)*?}E6ZL`)gr
z3Fs)+M7ym4n5|Uz0B8HBa%X~=944w>>s)}Ls(^bK0eQ>mML*CEw9SrFW2zzqT=S%}
z+Mg^;ZcHC)iNKp+3iJS>HqDo`tzh@W^FabO(=O>kBgF59^Xe3K_(HkVL9bX=@Rzap
zIbFa|sVzHmZzLF@FdwIm07^&KNbRZXS><8s@~76-%LJo<WvQ*p4z9^<fDjife=Fd#
zN_6?Y`~#3~*^1|MLV(Qx<_J0p7KzdPyhOaF6NpC4lLQxkcOpT^r&97Y;co|q_6@D}
zFtvrek!;t%q7~x2iDT}7s~4WSs6{LBl*6$;W`i@yX<Ns&mv7+3{5?%An@xT=!iics
z2=+DEYN7`_dLCDEl`avQ%nAuyY-{cALe%-8p8!~N>$Kg>7MRS0yRn}M8i~AoGynEU
zzEco)BNvsL&D+1=JD)Zhr8}Sp&aTaS4bYLf@kTtoS1UO-&jH3Azg!APEE;?q_*+F`
zR)UTjk9b}M{^(B>Zv=O7wKr0LH&)7L%%8nCBx#VPwr2m#+bh}9Yc^oR)13x97Hx2s
z?6^r6auY(?K3%8J16G~%kWO9=6j#*&rY^O$&R2uD*K_aVifls>gaD=8UXSc;ayd7(
z{@wNVC&wUGEQL6R)1d5S*|jqA;*y~2=y(qht(qsIA=|it=;wFEMg8xY4$=joY570t
zD&$`^Wwn1k_1A*(Mvg|ClPCpNr2>2WB4Nu{+<m*lLbSJXGAqJI4h{q!@Km1XzH@jK
z^JC@QqJ8!I3NI>N0V+g4(gt9j{g$w#Pj<4`6^X|aP4fd+!0GDFW}xV!qo!IJSzzuz
z-c%4+&r;z-?+w_NzQ6*gd$0EE5zs*K`f7Kw{f*Y|o#u1OpkyyDiuu5xt^;Q*a}rO1
zFN4muR;c9$%*8dhh&6Nc)FaWr%1jyv*{xpN`Pq@w6Cnyd981&sS?d}>7egb{6O*69
zvr~N-2)67j*U!P4t_Gn=C6T%J1LwMW^(s)v1p|1t{}mXRm=5BEBHvzZjW-c<&6VZ_
z)AwN#2^CxJm1jBlrG&Lpc3u4Hup-A%?SrzB>N}o)ukr*SgZUQ@T6Mj0CzuAo{>P2g
zNVBy0ui#902Zn8E^FZJ>Py2apnM0P0o4rTH;=zl9H+~@_NWryx)VC(w)W0+H&OVZZ
zjTPxNm=w#`T!H)d)(^u6hS6|%ln!WRxKY%Ex;k9df05h3yMoBs7r?E5OVFK4PWmr2
zBMgW+S!Mb~(mU74WvtohL)ENcPhx7bP7<2dS{F!fFtcm*AuzW)2tT6)R;8uD&YGCZ
zWmlgc=rd$HZjM|&vicw{I#6V6*AU|b-2?!Nc=d`>WajcPJL_j)1TI$1+D9gd4jMC@
z+V&7>t|N+8in2yhUZjiZLq3h*w_?8wU0{H`f4+!1SkcYZb(8D}6`q4jdp{UESSo;I
z@U1isq-4FOuj7EGzr$)h2?XW_vN89=^Q1(fpzk1j?+=2fs0SY3R@qm{R`2T}N}=AF
z@g4i&IF!QL*h0IFwT8zTUAE|_jcNs&_(0$N{dy?GpBgCQhbnxrV8`^L8fcZPGXq1^
zvaO@_uhJvxhGDe_$5HP@?o*1G4dwb4+lT`F1K?&N$>1u*;F6!FhL8!msUCx{gk70a
ztI2UVt8#9E3)KM)gH(&o^=pylK^rHaipAFg+guslZ2Q&T(&c!qm!4a!HAAIqDtJ%}
zsRp2uSZe!l^#c-Hi@B|mNLHlacGc%LAtqp5M9@Y^6zYT3_ROq6Sq(TOHmPp{srrv@
zQb<EGto)b9i}pcIGp^4EicIe;w*q@zLt`pJ6C+gD?5CV3fStucNt?6p1o(+=>!5{a
zreqCA8!~0oI+4^+GQS2y3Q9nTl9nJs<n}nmb7?e|Q8lx&Yl|J!UwLMF$b+}C7|*aY
z?mr#qx;O{b{DNs^gCdL4FZ9)mp3-`$_~dV&=I66B*NsjQ^dXrM4;`;!U)HNy)(1V?
z2SKD-Y)-}w>1w*HkVdr^=(PJ*jkuNSiD_)vWOmgz<Ju^zs(K+p6^#B+I`RqfO=LzU
z-`q#0gObSWgA}5oS!fSXakGL41$}q>{S)4+AZI<vV|tm+&8TuyX^(vk<3MD#);iw8
z>6x<=uiMUyo7ai^cFOp4lH9&LJwNzQ!`-$L^Z^UB3E4gnX>VB9M*d|p4s5XXE$~EV
zyNN1UsO$kf{KV?w`zZTa_4+5P_TB05s7Q^kO3vLWZux2NXfK3D+BRR!eqxLO=>N<#
zD5jWJT&&jA#E3<V-_<S3`O3&Dd1g1(zL_PsI_63fHaalPxdYDEzLSMn5Ev#CpPn4N
z_cFM=dy!HtHF=-+qGq+PZtY|9=YGZ4&Xc<(>n4HTy%nFVPqzyyXZ8i;Ca2%6y9p<%
z5<@3Yg}dH&sY1^(7SF@d7zV{3A{nDK;FRuS`|3<Xh4||gnb7z^opi+YzIxQtn4Xz?
zTvpMnIr3ilHYHzH2iL|l`{}Dgo$EGrJheE5US=H~RHsObE^y2j7k}Alrmx3%!^hUm
zQKY|2x9VOA(mDeAGJJZY_;-Hnd!A4IOdCi7Si@#ddYuW#o+lX-G@O3My!-I$vq0}`
z+X*_m5>VE=3CImkIKfFk$TPpQ*mO1zSbf@KMs~Kq1f6h+A8Uey?^b3X^sF4hXpg?L
z8fzh0g9tPG0`8&zcrBZqYq6#obKYY%`NtQF=|<~wNb*KLVk<}Fa(c&e^{fl-?B3Ev
z#B8)@)%bG2G$&=`Xn-t=dUuReoc#prXyl&LqJg)iag@Zuxxala^Td)22oPmmG-%`R
zJp<=dGXdAlcScO|BOqs7T&7r}-M+Kf)ck6$w$=cH<D6gXJ`m$=Z3Rtdiw%lh(S02P
z)c44)4s%5CFX97;{qMFCw|h<5wx66ATf*#=&b8dFFm69Q2qSwQV{qN>hPGu?$g2MR
zyn0=#qa*_hgHQ&N{l}Fj&p&MPvF@D!IdS#Zg`OLi=^GDB#jkc`g&2ybm4-VHtnMM+
zV4aey*6J*rwCn51vPxR<w9-a-6=2~Ly^6#zGY+~x2Puh3ITZB2HM`JXelSB?sB&cc
zJ>Sqs{(#Mig_Mvj61I&PFP4Mzq)@G`hWxX$;Qg0)16zroX7`NGzX_}wTGlG0)-Xos
zoi7l2Y$~AKeKzWHv2~KAAduqZ=n5^`dw(oiMZS6eW&&H@i8Bx^6zM$PZ3s9SEsN+_
z#x17wP<!BBNr3<iV8`jPZD%Yp9D*wrIiHJY{@}3h9K6lEQe-)AGj4pU@Ynt-y+Kq{
z;KPYbr*A=sVKtu|QpiOsgn~tc{u-4;#9*?3I=z5|vHpCSo!8PaDC3$qRZMpUO-l|c
z!kdsOfYMYPgBJa16(U8UJxPPqtJ*iKk<Y@3ZgPqB@nFeSv^NACa7i>m7f~_3$8{7u
z?V$Oz0;~PDaL8kVmf{s=5|NxACW0!>DGM}ai!q|)5~8Py2j^<~EH_S5ZKh1#hSiEG
z5~t{%(Zu!xcx)f<#ytOg@E&=?9^{OBmtxY|%Yu0h<G^1ly#u`uUHWtm@DH=Oue1#*
z&^-iR^jOp4-BqkfQZJ6%z5epFl$E_(wM36f^io`Rk#;}I729h3zTYv&DMqKn5<2}8
z(2zK=0D7|a0;Q-aHn8qH#u5fP&!5>yM&WnbgI1f}$+;g|1QvuIpvksEDD?SjR1?MV
zT5N(2MWSy3Wi9OhH2IXx?SccH^48agDwM}T>tk*A^O<xEOiBqdBF%u)am!0fV<&?d
zUfIgE383nrIXG-RO({ukDr1zOclZ>R_;JeqPM+4g>NL{gk3IN{k-NQI!W=kD*Dd1<
z=8jNeGq&Hg1N~G6$+vG|!QQ`LjeIovx?LB?#vtus>Yj8pWWnq5>j+!nuQx3G^mDbL
zqpcZtPwyOa34PkO%?}&>@`QJJ4-`F*3IxCKJ7Fwm?!EYEh&i*!l%k+NvT0lP*D=|!
z8^F$8{=Mu;N;4G}!rAcifHG%un29d-8$N%}h}M{j81diBteDigwuCXIo0a)*kCu*x
z36@dy3pUWe)vU$0V7mfvO3Q`jYq&^ah1K9TzObGqL9_ympYjx;>Xxl@)X0<A>746r
z_8ghT1=yc~@IDhWCeC1p%Vc~ZN5YtAbZ>>J$qR)J!uBC}q<aClt^+J>r{;=5c^tn&
zC*6yJLVZ{N=0YO4E->s*W@71y5qakOQ6bZ7qDtiVm<X24r-(>!q4q(_)<7eonBB7w
zhE(<RMplfAL#f+`y5NdOjBv~-LYJPrKXlc$b30w4V}LT|**;k+J|*Y3W+XD*j|)5e
zv*3ogc<SeQHWQhk&~8Of5FC0PMr84B{k{T~-t5J;bKR@M@*<CdWfG$p&>B$5C^HnK
zuhc?weF=RGSN+S`$Fz%l>W})c4(Hee%c7|pBB2DwVIq_9Rk(^G#$p1I=DQ;aEBr)=
zc33_kS%)OT3B-s`wTaq8RzFi6M9qLqF^7?jz#jhMOM~UzZ*`T~rp$McyOodZM{(2k
zgfG8yl{7Be#k*P2J9cp&*?XOI>Ai?Zv(+hO^~Ru=L?db~1T;u_#{~1F?mBRpykxK_
ziT3@yoWQzjJf`t}ZbjhCIU`Zj{amQIoy(AK7SQMfp4Crsy0aV7wQ1LdSD%sVIU&P!
zIKa(IS2wxe(R62ZEKJCx*3(@1Ne)txXEVpEQ5mn^k^@PiqJ!RSj9Ne4bR7?(ENXph
z4+7I7b6AJaH;^z+mwEFaI<B1+)@;0Yy8YU-YdV5Q>e7ockraD<F?dE<?5_h0MPA<o
z3*n2)w>>t3*u>n3&p;1kp)yNi0JnmEvkkrrncL&0CaD{8kRk46sE2%QQmZpqNZeE(
z_sMwbj`cDTGDwzOWY}y*k2x0TRh1B}*hNph_PHGRO>5AyDeQU^c`j;HC53}?$&f^V
z5CIqNvnQM035P3Mx<oeHTTJBW!;eyS+;dz2`pMk1v=$2GHuv@LPg!0eeU1DeLO%mh
zB$kteNRdt<jpU>8yr3DV=f*nYjon7Yj{L#n@70qb*B4IOc<TqEi$I!K1Enb!56uWL
z^oL}sQAg&hK$>|sdLLrje{mxkT}0!+zI{W(yYYeuF|g4|nY!0T+qM<|b02HFZWoQD
ze}=A$0^hqp@5gp}X0TwnwomYd`QYe#*q|rh?CZ`D>Z^S5HJ&Sr&-fw{2M%4^$&MmV
z$k9%=R>wA55c|3{wyTqd7*+Qp52}ey-+jOB!nv%0j9K4RXy7Qk6Z|iW!e+0P(2Z7<
zU?L!3agV^i#G)hf=>2tX#<&{8V;enQ!z@on%!_b#iPwVU?HevB?vAPy{4OgbpG<2n
zUwTDst~m${&k8i)bh)IIt&j;u9PJ6%XV^~m77{o%r(+`6Z+x5}Ml79sLp~7Bw-Ag-
zz64N@@fHMnzEFd3o^IUHd9GYfZqFZjSEr-z%1YBo?}h#C`7om!l}HgwRWdDTNQ-z&
z2Ab*LSS(y;q;C+<eGPv$D1v2yN{G>yyaR~&>&C1UVK!a$vz}*n3n%scurwD=hH``-
z7mMG*283tTE=Qp|cabqCaobJZ0A{tAOM-Ve{~G8tWUo8_>v6Jo9K=v1fpf1cy&<&n
zp46Uq)Lb(L`?tEOHk4h2N8$Ck+jQcEmx*P-{=6sKV+)*qTUbxkZC@qKfTdrTUNi%G
zPtGfF`S(f390h)Gl@x;xjq;^UL&`Jt(CE#Wv~!~_N0)t!`Ffy0=z=5}*m#GEY)qFM
zI}3C-?_y@Wk-5ITUe3~d@N}nLY(@qPA@9?C<#8ejVyTYWx!@}3f6{e{6?vx>x{KYq
zbHDNg_>POr?spHtIi64VJ(hl*et4Xb77~!Woccy4h3mRT?<!aIn~~{aOUK(38+I>x
zz!gT|Iimd67L?Ra#)VBvfTpDCs5`Y~z?|)n)sEUwxJ|%Oe45ie_`@4cl?KCY&-MiB
zo|JGy$6AGY5ZRCk*@sPY>w0}sa~fQ)^)TLf{+Uac0hPbKchRFZHT+EDn<fKhC!8Zk
z_(KXcRp|JD!JURK-!=6kpP5C_u+5bEVuNwVpcFg8rYHS$04^epiMM!YcL@@t_Jp;+
zWelZwL}n7AuTB6QeL+BtCWcl-gzPUkb?@VQTq$XoBNl=NtHwJr8HSNyXHd?ul)*?v
zLL)ls;rNWn+OS9<hy}Q=Pr4kJ7<rKCgfS^p5%!!L&QhEYB0hLsg3{23cjjaJoY-8*
zuT1}V(>8V=Ch&I4dzU{z3P%9&c6T360Teykt7iYk{ExjG_|_#&s)hBrz0e}yi?xa4
zdV`-uogNGh@aN|hrTM+xakFr*9K*tp>o;3Wj~56V#vZlB^5U^k7pKz{A|)?s5M38W
z@dJ%I1-zbbwNr~(r55AfPs=gw*6McBWW^X3%3SZPv}onnkt9?1h<IXFAxK{ot4dbW
z<Hc~xVY^1}o#;+>c#G@ycX!Y%XLo1)(L0|fy{qwG4%a2E#^cu<8Lo@JcDxDBx!*4$
zQsjQzkNpuPLRzPRjftt(ZPr~P`apxGUQFeIjpB>^JAn~9d4l&lIajuw-HCHPO`oWf
z7sce%EH{;dy-xJeyjH=qmRw-nVWz(H=%WE~QH)=*;p(2SuA(#IgH@ytW-U};t8f0?
z>Tj`Lbd}bJ6*vvJI+LP|N3@wTzWGoOGVn!Nhrb2V<7k-W)Kt|4`>DL3LF+W`Ce)`#
zcsVx0%A<Q5uMmDiHNu<dMm$;sfVkZ;EGmX%i*5<0hNg#5OPvY^5}W24!j8Xv?2B)c
zt`|8Sh~T+YS(NafhnW9+9i=j#!3kcz_etT8w6hHrOljlL0nx+PkOIrlB3s~hAa>%3
z^x)5Fq^p0JbOBi80kWFm)%%k-Y+DKalGTs*6B?#yVp;;dhx~wy%%^3NC8Ao+w;g%_
zE<vw2ejvdcgFZDWtI^We&6y|a4iy1>cvl~($=0#bjzZl{o+W7e<Xpyo&D|59nY{lz
zcM7k$^Rl<0D&M^FB~<^d(Ee;x^tXc$+}^as03&%Z9wD_>qL7>nz^1zF7siKeP*qZz
z$gyqLmd>&L0e{;RII9;ODg`M&EIRH#suZf?LG2?mbP5|~46FPsNm3NE`NzybrMWk4
z7qwoF&40~ZdUa<<3b}Qz#B`w8kGHm(Yj0I`mH@B*re3D~U(2aOL4OSk0qn-_Pr=<`
zjT@_^oeV15twF+v)FOFZ1kSd(<}h_*ljF~nmmJ}}X&<p{n2I?wx<$)7t9C&1hG|I7
z#>5yc)qRrnkH>PJPjp?EI~#`8wB!T5SzEpbKrzcJFGR4hshcXQz}0JDD&kk-nkd;Z
zOgNkb{-g|wE>w92w5k^8qvH(=&nqOg_gYO9SR%8aE>63<tX%EfU*#5;!+|3}HX|z7
zSvH33#<(IJ?OBL>21DxfTG<o`i_a5MoZnn2JY1V{6v;uXW4W4kXOL(dDX8tLy)%-u
zWDt5(c4<{Ma9_SPN&#l$@k0J9*6E*#Ka>d4qd^5REbCQBjzzD(J+pq-89?x7)AlMS
zQGIK1f&l`C$9&;6hFN4k&raK#$&`rUJptGES`^xE{GNw%ernr{ZJW&KCEDBi;3h~K
zzbBB_m)E|Q<%d>gCu~pGzPU~?@y%VCVTlXbXq4EAiFRq^dHEgW^U2vuwexjFFD9E<
zm66JlPZIzi(M1htc-DVmk{k&VG+=;!w_bAfZ$L##V?QZk^dh*33LaIt0pxcl>{^1B
z6~FZm{A->$%mq|i;@ZNR>9F!5$&q^*EV^nT`5*37fZ<Yyy&E*)-~npC8y{pB&zf*N
zm5NWZHzHWE4g2oy1DQKLJt#S7tfK1=qo%s{{-!9=#K+8-$PeuA?}UKx=?7Y+*W*r$
z({Jx1nHwJO<;+I>XhEF3fS3Z;Q>pyqRGpGR_r6eH<fRooepOCFDXecH(NM+5<-Jik
z23wVPnt71b!K@5g%BStsN6$tBRcz;r<ye-U4d%%qg(vq(t=ECp9iF-mAXjs1QX4_O
zxd*AqeciFZZ#VkzANLTAcaz?(Q?J$Cj(~Ma!?V&oa*8=hu<I&BH(t#z!ZyOA>|Q}3
zaND|-v#?tG=2D)OtBoagDu3HIk|I?SYgA&yn#@rt2XMi8f{c-;e5>rjQ9O>q1(app
z+q*525X65@jMPdS{sM>l3DiymUb1g3t^k<J_hdl80Gge^(&`OBcA)#$*-7L;Hge-Z
zo_JJ@k)~lA6Ua6Ha`S6}4=9a(8Fb+kmisJx_D%K07O1lyAx8L9-`m>*_oU+6JY!hL
zNPw--{CO~KxJgLSG$9wVg|u`-$j^YEy1#&WE`U&1eWI^_cFJWjj6&|u125Kuz4M#)
z36|-a1KC!{^}!&_&*-|b-H-94hR}qtgZg{25RVR32<A1&?B$r?iugMq9*X_@RB#J_
zz{*|_Cq`U3fWe-=Dg(kVarY;~_T4~Dloe`UeF}OD*x!L$zRemyqxPJx`%4R;fjaj#
zxd-dt0Tneg2&dNFHngw<f!UMG9<DAHtP_s4c>ZmmH)#1$_c<n|w373ySi_&`u7dNY
zgZF#1=fH&LDt6|z8_Ng?J^hDSaSD5y?zAV?y{v58EZoEPT9?(!>?jR@E~|{>9R&t*
zxoe3BNiF;;i;kG{3qRc3h@WF>n7MTuz&gYw2*^CQ?b8nWIt%(Xj9?uS8bh@J+2P@Y
z3H3+iLwQxW%KLQ88p^P=u0m((uz45I!O`}wQ3_t5cmCnn|6%W~qoQite{l&383{oc
zT0rR-Lg`MWRRp931eB0w=nff5TBM{*N+e{YOF$aw92$wC8R~b>^Stl-Jl{Cy_dVyo
zv(~#7x@OIineE>9zV~%~>H;9j<Ir27g=GHQA(xqi26RTq<AWsIg5)oev;!-oL`?Z8
zFhv^`XMip9jQS-Ahj}L0O7)=Pw1?Wn?%kK$+x6%BC51nb@x>scL^7s$!-f{OvNexc
z8H=4xlCS(C-$n@c*sgwAdA`yWtlD7gr<-{<)cDMqv(GeXtY(LPpl{WnV5N|OK)S4f
z8v4m!+27*h3f8?_4YsRj@U2B@&py0=yw(PcQ!_gG?!xacg@D52X)3tqj&!Ko!tcvh
zK7JB)YrfrGYU|ELW>R0xy1%2fhw2RnerQTm*rbL;vYp(klNv-+7ObSu$i7S&#aN&P
z9)6Sl{Z!KfG=^3bvQM!}8IsE1y8kLdu+6#|e*EtCnke81sIeI$Xn_1bJWu|ShX^qc
z*#W*`{owbY6mcs4|1CupC|=mZ@$(l{o+16g7mMuSR4lJ-1M|?&Djw}tLsG?tufHC5
z$(}qnJ9!;&&(ZG@fT5TW({$2mee12Nkn7rj(7JNv;$&ZKFV7iwr3fYB(sQT%%3m8M
zObu7jhR9$My@|C<O#>9Rl1RgAm4`$xJW#kBrr5RUHK2A`5W+nGvm3}yS-%5Xi~!nT
z|MU@sP?X4!ApQcAv0>3F>zT%M>Scm%knN&1xFv+o%T}<3tx3#JDZqs!C%bRQ+`J4T
z6@h-APu1+9=QS490ET;&`qmz?r{N-Q60u7yxGDi1e4wH5U`ubjv27!JVi*kx4&r>t
zqcE1=bcN?rUnGY#=oS>Dcrj}0&$x?NA__*f#e101$;kpcw_Xl?%bhqK5v1!fJnUjW
z@t(S0#Gm+K-GBe1S#k=A^#kkQbn~76>=zSGYC!Gp*7)`YwcuQ1qQ@z)K661iq=8`Q
zoJ_H?9-xL_*>Q27Gh)}`KY!XtagfI>+FztCIZ$M6Z>)_th3a@FTG@5pNrsb+cnYg<
z(<i#;ZD9+`fyq@d?5_GT0}CvSjDG&)Xp58)iwgveLs8G4-6IBU-hFRvEL6X0<(etV
zQ72hisa{ahk1r0{oXwF1*#4mJMG1Higph7ydx(vhYHB|Yk_R7~z=cGA5yX%j&x(Pz
z%zY@T^>F=fO@ucKOe8$e-ZGGU@3WKNM-gTXY9aJ>5{*F(VNEoM?MOC_D>L?W5VP4%
z3kSvlWNQZVl{gXKgD=?*QNkW`NWh*~fjbnks<aH|CeFApu3Gxci7|MLTJVkJzB^tL
zt*13x>O~&~d6G*7oq~qU>9h`b_9|O~Y!om5Q91>omCXSp_S|l3)kx=35l%=+oZoND
zf)Suc*aYtWmFP}(iaIPZCmlgngRz|daQ~?#?;!s|hg0nlYGyzWDgS8hC@$cXt**j3
zOI_R>Y_rEZ?SW*MA2W=W&jT)#e|k6TP-Km$Q-<1*yp#R9%;_AN=-UQ0r<mKnwzCM1
z%SXzIDCWxqo;HgM=`(z&WD^IGZ1Y_&Lr5jw-Qu5D)I9}~wc93LgpiUn^X5_RP{&=f
z!q8~|X)3s#r+wL6{=lY5{2Q$}JWrGM<)>b4{2##PgTe*AWmjwqN|TKlDncI89QB9i
z{nl=$-AG##{i|ps4=U5VB;fD0tH^|<<0PLF7?x>d|2*@0@Zr^u_1HN@+;6mO`;vb&
zApd!tXrlJ3o-e9-;WK~|sisHVgJ}!p?&ivj>5F*nW(1h`iHYQ*`A?rV3Iq%Qio&z$
zb@F_z6DXDXz1WTLBjO)q4?F5d4Mp?q=C1teP{gz|thsdYEqf9WuR_`r)>NB!#oj(Z
z45qD#pq~F7?5`n$fQn`_pCUZP)O)oPX)#MpqG4}F!(K3$@nVa?jJ}5g*IAH}P-^uv
z@Vse`pax`OGoFvSu(UKggIF-`G+r_sCx+GApT0>Qj9NC|YP~8tIa?(hiy~fz61Cwt
zF=>2?m}y4)ZTV!+?5qR-F_%G&JE;Dtp%1rz1J>kAZcAgIC3sNT*93h0T=|kY@6Rec
z$XPB&qhGzg43M=I3s^ri*VOrSYgrz3RG*WXlJTByE*VL?w(mS8+VT_*kcy6C;6+cd
z+4&8q&Ai1LG!1|z+DA6u41Ww#_4Wm9*FC2xpeQujv?zG~toty%*SB!-+vt%H6t}#9
z>ZA8yL#QtDdoKW&XR7lx<hu^?fg9+N^V<6&?=d`DzWn!KWsxaO-#gCPQ~K46M{uPV
zBo!hT={6f91xTSMetUV$4mBxZtlP(}0{(L`Ftoqgw<ynVtA)E0>=9nyeqMmsxv?Fj
zTjSlHZ`mJXLr)J}kpRMcyu;Lajf{Rnp$ALVXr;fZdeggj)<j}H+g@7>l$|sc%ZSU{
zSNwr;mm0J|@12&LsJAwt>;G}pDVu`@vJ2n`r}pWD@x@@}L-}w-Tt`X-gNxegXZQB)
zxh2OowX;8<&u|e>iC9pgAFmZkxomge_80`5&Ev7VMY{*d-?OiRLW5y}K~g&ujRC0S
z<WO-Q%?efL_+?BeNTqwX^Qh95V>8Zofh}ga-?QaRTg#7HGO6Y4p>I#{qDU!cbvGN=
zVeYbJzb~E22T)1)?EK&mU5xh7|Bqnk&{dptTWfP^AUpE5h`jq5I1OsOfSA{)QdDjm
zAw604*@O<}Ukem`(V%Dge7E{#n98Gvg2zyLj71BUM!d5rpA&O+w{ACcO9Xz^s({|~
z$-JB|NK4dbi{0)99=D%lPk|X$vi>olwdN6@jL=+cRh`#-ERNyt1P<dk!C8RNt?D~n
zlMUuuZTGtqRG|TIp&{uzZDBxUw}s08$F^#YxON0Yz-w1aNAGd$QJ^Wb9ldNVlSP^E
z>d-MxR9+urO6GhMHScg`0ru+tmG1}2O5yTG+Xm#alqGebPj^-~%g&Q+%P<}Opi>tB
zYDM*3+PdEy^(?U-s!3L&(_WSZXu3DB{16lFk4b#wa!;cSSqTeBC?hFD8qLk25S-(4
zJmw?0pk*ZZx6K?GdS5{bs&|kxb3I#z3D^Z<crDyo4p*5?qY{?mbwO{96W~{GEr0V9
zzgD(#7TP?>SZu`D^){_q>&TOqDj?yc0@CRKwzE`Ym%H!n!+67Z!*g#KU1PC9Gaw(K
z>?A@Rcj6LQ?IH(T#JlnG$em)pX*1NUsj2A;sIHMa^c2aSt)?Iesm+c%^^sDhi~aoi
zHOo)$S82G)dJbxeI#~RG1|7U9yR^l;CH8R-&hd^Arjk(n9~#O^;v7mFo;D!;u4AEm
zVn81Epwp!T%A5}Cm|aak)4O;Yv?w!7#m?0fWe8pyIh0uH5v*G{-e@vlE=jEVwwMrY
zQ(Q)GH{$hu+tn!3O`>$|He}E%QwqCclh6_p#W`Q{OxmGU+&OsGjpo1m1vW|g!UPYa
zL}e;VtvRkso%b)7azZl|rLv>ycsGR=Xb*&|Udmh(_+&lJp!hg_Yw>kX?LrMNX`yV?
zAWiXdWbvEJiN>O@&TJo3Tkt#?Xwf1=hWx#Pvc=7NM%2-H9bYxH&%i+OedEIQ#M<{d
zElRg>2U(+e=fkR?<O<w82HV$~^WAy(T9={9fWMZKi7ma1OSv%_XQO~Ta+X%$pP(oK
z#oX0iZw;a^3Q^A=1f^D(#|P2-6h~+On2m`tXaC)axLiZsn6yDh{!Q`5g+35!2m(y$
zPAg6OPx0bn-Ju&p*AAvU&16~9Oxu^=7ryRXzQC-Pkj3@?iGi9!u7OE^crU@0Qd0h%
z%<s~)g&|9n#h}m3$+Sv<5GWC@8#&e#h>}}P{P@uK+^mIL9WzlKM`ky#cA;$+0AglE
znWmfqPFpCokNAYU5G7vhGnOk~LCJk5D?MxP^^8bXd9T-u5pW&&!dvdurjabmKawwQ
zv;_y*vs-@2&&!)H+XCK1%;DAB+#zp0a$`n^iSx%3MY&GG^ko$DMRh8ilM)gg86x2V
z!X)lE6g9GvQt4aw%1t+NQlN?%tZ@`O^SYZa^>S<^t%7>h(A=1=C~8Jcw$0W{moz9U
z6X>Oj8+fdJ{k(g8niuGfV-%-Ov3>9KjgOzs2^CcaC=}}5Wjz^*SJw{I>fr6RuH(bb
zR!`R0j3%>yu#+$Dhwf&_WJh1kbjCMpiOBL%kc2q#Yoy|QCml_`0gB#U>RF7!Up1kU
zxh?Uhih+AE%6syg&5>j^8Lj?yUH~jQYZ69^>BYoEMU-8MP8zic?CXMNoq)c(I;F2G
zO>*RDC*Xn3d4LpR<cDGAPDA=7%d!s`T~*I(oJ-K!Ff3pqdQ|gyJFC6jpontR(}$rD
zEn%JI`nwy~hPu^#pmD*M9do-$EpEg)9IFTyal_|*wkWNE=iLOmpj(ncNnG#g4=5W=
z5jS+>e32i=HMp8DX}sO-zh`JKsbLh#FFd>+w&$!=-+0?N8f{Co-~;@!j-`=Xb6~9l
zwwfDk+*enxs4m6RfVM{k^WtUR-|k>GdGjkwPjLI+g#zouOnTp=V|&W+DbTp1?Dg)U
z=q6+u(u-tln2z5#giIV|QM%sUh?T+EBIq;fN-%P#cjQ+zBKJ~XFrz0rS!_O<zM=gk
zzN@j)UEPzOko9F_?hWl_j7`Rk-_`ZJqulg1Djw&($Q>~wzf}$Ne79GBdd_eACqwO)
zlzp+X{=zK3(sA;I=f=mZ{Mj^HoIV=Ax!N<O?Y~25qY~UxiRW=A$MeqJEo?>+^D^im
zX6@hhrps+5T2Y?%RCe>Io1!EqNXOEH<WXP#387L4NjUhuVZ#ssIrWlw=S6>9T6z1S
zv$Glf3}uhcYP1H+Qa7GM<l8+pvu9T*W(2}XFspGt*@90%V&sf2HLNuOOUg8{8O45I
zsJiv)l>msFrM8*qO*y@PcD+1;>-Y`V-H!t0W>}e(OO3L_KeGVLQe@B3aTBIh0DbZi
ze*(%N7gYx5P{qO;;CEmU>H?>F%bMlEX$`<RMup3juk=7DGv<8(I`am8)eX1{iiFwj
z+Yoq~7Qz-^3uqVj>QPIjxuutPp8=_fd%!7*`MbRe?T!r-zwhI_MYO^a6AJBJk2;i}
z+Annzbuag`3(>&?m%+XSE@FkSK>sGl7W1eqh!l1THzidp4zAVwRC82EE}wZi0+)SY
z_$m9Fw&B7jw#rPc4o7{}0lCNDh=+cZFYg;|E?G~cAO5VoySnrp8-wud$4F%#r;p(z
zNLO|QCHjj>Kc9)yq%$=mZRzEAe_K{q$q56<gfZPRNpacxiHMqS@Hzl{Q<g9aakSp)
zYGl6B4ei^P2<MoWu*?_3nToA(#bUrM%P4E0L6Jx%++`GZ;|MaVw-nvGX7T$o+S?{m
zYwSooNx2pU1HiCr+xwb5Xyx`;dTN))nCFUlTbCE()M$AZ|GK+N<>*jvG&Z|g^SB-r
zD>(@m%-zB*%vs>(Rq1<l-orMzHsT<jb#P_e9|XH^WYQkpF18>&1kk*rua<V2qP0x&
zF{(WN3$rFlHLhmbd;M74bk}ReeuSXpqejV~&0iftho<?%t2nd`8SQD6@TN-wcd8Y0
zTKZy%m&_YzSc&Sxdfj8~PA7N4&Mj$lcRub7r2v7q2D`AlJ<kIPm7jhPDf{agy=;f^
z)7w%Lk4Be+e7ETdppB6wye>xK^MqV2&mBK~XN?eXW_Es63Fh7q{M-nJZ`cWy+k1ZY
zr~HE|fy&^zk7LA!_lh_pR%BtZ^CcO54rULFvX_tdtoQEPtrW->^N=*$;7lJ}4)2{>
z5}h2`gksPhBl-3H?TG@I3y+}n_*=8n{4W?O6`<SchlFNi)r9Gn$AXA)tSGIhtRMyj
zL3=q$d8Djex~cK5Fe1o4w~gVp?*ey{hX$0wiz$FNe>Q!IrKu?Wn-b=(C$AzV<5d7>
z;M~2M_`_{~9ntvXBV+2dh5DWo!&}k&_Yc>HK0B@vL41gVY(&9inh(qS4*>VOJw23A
zVkhZRS5i7BT?>p&Zyek=psH^4_O<+n1p<EWCnns!|6LkDPHxT+iZG=9&^EZzOItNW
zY~efE3$5#ca%rlIT5e9Fx<vSVi-rBwwk&xS$6Y!v&SFRIS{Cg<@q7K-9J%ymo+C|F
zT94s5AJDGHj7|ZP8YuVsddry~KLSIS;&(@yzD7hUTH>2z+cB0(A3>+n$hPAuPzOc?
zo<mH`dIFh9{VeoAPZQz15GRxym&*X02tNq7yhT}tM|2DIRQy|_d*YuT@sNn0^o#lL
ziK+8_P5QzY^ZopKo(8qa-^D3v<WN1uhJ=osESA54jDOrtmgx;(X6_=X_yV~Ten-nJ
z=$SXA9~HtCe#GVGIP`BwXGF|^Q2yTl%Ng9(RsCya{)J%$LGV?cBV#R=fxu+tVA54P
zI~r&Q$1~jq?j}b+7PivUb(5b3B6ByOhbl44@8;+D{X+liY<0A)`}@C6_z>$2#fHae
zZV7+A^{2cBA%!FLyj1hHsT47EkI}VTu)vhcaIWWrYo#)1Knv(UUNFNKC`2z#cjJ8L
z_m%#eBZL;=tN0*eMZaV93BbtW?>_=T;EWv@5!mDz&4K7;ktz@!=9SqEhM6qaVTMaE
zJG4^|S6Baqd{IUPLU1_l+wqkA>#aXuSiWok@a6cznG-gEGdyrZ2n|Sau}Xp`JSfoa
zO-_OE{TPSZiK(KQe_=y^f3ZdZ4M$5%%$tHS45TJF6FAthcvytj70@)+-!)b~*yE3`
zfeG?t2U)b@vS3e{0@muaLNJZSzE?--ucOh407tm4GjRK_xBlEMVOQ{V7J7ke3+%30
zyV_Xq#Z*=OnM2j)xAkx0OKEvr%3%J$D{SjH1w3BsK%wMu_(R)&pp0)@frM|!EmP*t
z%kZE6-cvO<Xj`|Z#PTk`{0?hp+(E~YCz}maTPIy_HaA~40O%9-2n3&34c%9-0p_O}
zta5u3h$uZr&?^LgU8a>(aOfwu-ktm%K>Gs^48$+4KliQz<{mNRRdNv1O)c6C5w6B|
zYs_QUj({7%8EX^(UEIhd*!xYZ0Oui5dHIQv1soS@KF9g&uP^z>Y7HV*#~!v<j-}@%
zYQkbrLOTMBPG+gK&|}&F(7I0oigXd+8g`-}Jn4D#Qa?A5wEjO!xP~9(I?4Y;`2KYj
z!_zJqoyveA0Gl9H+6`R33|X5KX2f}6`cpbl(64?=fBO2z1z&jw>n2E#t;hgovMGR7
zbFwwC@ax~w{O5=N>vMq=7+Q&zn8umeFZQhADolG~AHo>&{v!M0nE5dK;#Af2QoH#W
z*N=Z(XxKDS2l)$R^1B_j|9ES|_=&M|0XD&}6j{Cv8b0^Jy{U0~_+gA}WoG{`VO051
zGS`1SKP<O$(3p5?#H(9>eHvvvcJ-U*@^t;@#r5|G^8flZ3cEJjwGFvB{_lSO-{1fL
z`uT5&q{GDyq{3>->16kkDF4gC`==`7ucyj?_s%<cb6}iy`||USQ4CLQy_M2coYcq9
zxFQ5|{!VEEn(u%7qj3{g_@_u>EqF8$zFMTXjc*L}*VaU0@lPJ+|KF|FW({+IR(g;D
zdxzXBvA5xYAT>7M2k1%&Ur{J?|8-aWAANfBy;;UBuZRqgw@CO32@aD%jg;zd56}Pk
ze;E8r4Rc%}{5K6@C7+`I$LHzqUx4rfzt3=D{f^=9@BYtU|NHy@eg6Nw{{DUc{g?d*
zLKp<%20B5=sHGzq#N-b2<cBYAKEo|iSQHE*F*C%<l>awBL4x<>!UHWUW(m@NgAkMH
zDWK+JWEKRCOARpnmpM;?faq0he{u*oz2mJEfN3A@mjrzOmn1Gcu_1_%X(h{d=Wcd2
zmWZt5W9~`f*28YqFn{7VFPexqfu!Vx&+oE`2~ak+IjOX(^{9N4YEv}x1u68^Adosw
zw<&_zzdFM*R-7)|ofWrPV;1N;wl}tB&JRShyRk@tV>DkSbm=CK!~gCP(5GMyNDpj0
zq$f~d*8;i4$6L<6SaPdcz;}+Mpgv3h+m>Oy9Kf`Nbr^mFoy{~>;i);Ze2TT=R`V2B
zzol3N*uWOGiCcf$^4<b~FI>Sqj_eCMONjwdV3IoN)q}%9^}YeH;2Dt*LB6JzF}Ej6
z0?wXSb+AT1Y`XwX_*!gyx-_V8W1hc!yWPe{CcPKmJA>^*dH{~swk1LF$3=lcg0M98
z8K8Yvxp%O$S@^)b%=0W@Tpq>tK{BM65}p!d`NwKE^eoXE0HHd9JO&q_P&%1qdYcBS
z|M1`jFg{@k=*5^-Gmt>Ed{y~1y9+Bg!7c}_aZ)>xJ|H`MGMwH~?5%PK^bm9&l|BGJ
zWwcZnfMMG|TRH+XY`--mN5a?n91P=d29s=RPk`WVcC+rp@o>1rf0z+myXI7#0=55A
zcmJyZm%|??hq@~^fD2aT2*k7Q7h4x+vf14rK)A^Q35I%nIqv<`bPgn6ee5JXp!YD{
z1T#zbN3aYo))(#$&Q#om)bPUxH6RwYYi`a~2PY&7Fjf;5Luv-Voi+vXDIZ{}vE$er
z!RQ-Tpr%D;xO8$&MaZ67IPuaT1m}_DI8s0*^z2afVpkS)YQ?Tt9{C>X=kek_p3e(;
zC;f5$4{0|*AGr2i^)lK0IRzw69jNvl0cGS#!dDT0f{qJd@E`aGmM@22Iq9PA3FcFX
zV(PZj04t+ir=1B}!Dju~yIP29uvma;Npuce_m5i<b`vw`SkG##Nh@aEUC0}slf#N&
z0gBNBoP7{l^OE*61lQ37(Cf5&K5duV{PXe*WuI)-&rAWa3EREb&a)u4i$>Nl6U?FW
z0aU;gE(tH}tzE|%xH>gaq)xuP_Db17HVp+Sk2|}-a6;dmZFU2$Kk=sj{tgeihMVU|
z6!~f&bg{0SCdAM-zjdRTowguf|6ZTz!G4k+Y`_I}Ac0E_;B!%vz;CbygF!5M#FyQ$
zdWNI4vq&2x*2U_>-!e<yOu}voT>Et?Heh}h+!2&#hglTofUx)Jg*#*Bk*AxU^{&`6
z@ZBaJjFKAgL($K?P-S#~sQ#v=zQK}(AW5p07vQ89FPKkf38%p5B~%XxG%8m|HnJY4
zvY2x*`Eg-uJwsw_d_~>N3AUxC5=cz*8kPpM3f+ul!*^gpmKeu?^;3(M=J0Z9)zJwL
z)HG1pXa+vbC7GRsxN0DZYdNgROCgsw0P+|mqEg0NSwpbtEW5&Vm4$R~wcA1c+;I$k
zg5hWN0JT7-9<DKvZl;2)oucWC4KEn}Lo=yrZ=r5MZ2xZ>p29K<OMNxK{=cq4byTpM
z&Y+Se5MOO#(NyeIu7T`leljaLA*W}Ju+m?ewERzA;juM<QFSN&BxZ;Mu{gZ8*zA+Z
z<w+nB`TfWgZtZk4qCNCODHt45iv>rVd=RRaEB&x1Yh{$(D;{SHk)$~6nl^eokXpC!
zUAxT0!mz<L7lPB(u_B4x;5)PfKVOWNpQYBHT0dB5w3VFxh42qOpd|~gSaj~N@Fj0A
z+C3-gILeXr|7L`}kK0&-82rj0IJ?P6%)Bpe`+W^ipQ;xWRJ{CGecRA+4IAu^w50S9
ze(>iKgZkrYOYbKs);Ajj0C0rq9G@o)k0VDy@I{Mv137s-i}^z$>}aR&6ql#Z{K3t7
z@|kjMnkddsowqD>;@QW~lR)@RH-$d1gW<O{4}d_PSw}EkD?+s%gjp3hQ_KjKxdIgd
zhR?<yh<-7j6X|G_IP5Q<f?IdUi5S?gXuR{}+1<`h>MwkTzT-N4-~Y7<Vu22_$Yu`f
z4wc{gpNQPS_XV!iDG)ugH6{Z>b+CaRf27+O0iCfQ(@dSZ&7aSZ@8m~d>2IDSmjDry
zW)(0|l`8{vj?)(SBfI6mU&nGs2Fym}@ijXDSdzCu8DI$N4{d9~Y`eCqq%73+)Vgi5
zfTNJ+6TYm^P^-DS)VIcfL(tIMtVdwAUf|mbYc`?O{;RFN3v=VRRB6mG?bEc}owDx>
zc$OD~+OkvN*p1HN8TRE-&4Ek;(H+#hR~ovQdrd(WjC!^o$jOo<V6~?_4q-lRplHX!
zN-i5fi86<>kX-h#sMtv;BzF3V9a(c2ehT#F;#Ndb@&aH|vLjI7x?`ifn`^C;KMkwP
z42bOQ&|cmG_v~2pbtebFH--5uOU{fPk!ojn$&)e_8)^hi9!($CRXJ;lSpffrbRskI
ziR%V7j9+Fqy|=o=caf)Hi;$_{eS0XG--@Cm;Ic^jkJ%h9p~Ff+GQx2y$DHdXn~`<-
z;tCi|wH&|EK?%>a$%6aR&bucm-?9#Bl}NhnA{vCL1?Ht=wtALBR$<rkC~e;+pO)gC
zSkcGstUWX;qgeSp^6)nDxa=x7DjdjG22-zZz3AXE^2M?%aZZQFw_?S%wXj1v(fWML
zfSyebczW7gJH5}prN<bzhTp~k^=r2Pkd%i#2eT7x@0QnTA?XaD80(4?o?_y)12BVi
zbw9!L*I6XSbzF4X2G`;GRx1dG*jSOw0&{6(W8cnT8ObTEps`wD<nHMyHo2>R^TEZ+
zpvR=M>LD<4d;?WV8&>2ptuD1i3+cqLu9*oeQY;tENc2B9)1a5Xke5`tP`#)HF5LFJ
z&)*C#4U~{^>+thJ^GFAJ9+(+^!qypLZ8dv#LnNK{<bDRRzKfgDK6WQ4U!CHLm=1v<
zLF%o3=DvW^8=(iD<SY))<1K(Q=zEH)?u^%=qO8I2uf0TU*lyz#INLEc?GB50WKy&1
zvgI5b^gAseieaGoj&$y4l5%dLd0^35ns6Lny-_u<CwV}WX2hg6nE$7f`!zl`BORDp
zMrr^n-6{>LfmjcQ<MXqvfck7+W&aPawp4|uEYHTkbi5o(-n$Kr4Pf%+YVW=_*jKBG
zVQslnkgcs3MIOHSvFEE;DXG;G&=PyDC|}0{eL18-K=a+m=-fgvO*{+$<A=*BDbH^T
zRNbH*7QpbuF}Sme20kyYX<PrGfCSzQWVRwnwmVgjBXI{gjJ0hxBT9<BE)8n26a6_l
zy~7)fQ$teVg?qWC8blK+%CTde;VQB<+C}96h7ipP9Kc@`91-Fg?I9;`7H1FQ2lHWN
zqjs<!WS<w9-!`%nLb)Se4McAm!cP-P5jY={<QmE29XRGX{WR#zUo;-HbFQRAPO)`Z
z-NDzA7`e`UBJ92jt;dIXD{g0k$}2<2@F`ITzWu28D~DN>T)m<(6m9-4WPmx(4KWUz
zeNuidIddE<N`%-VS#H6m17V^JHZ9%;fHdgctQUf@MFIvOor_W7;08oNK?2s+Tpj2@
zahLK6V4GXS2|T3eo5?Y51Bv1OJ3(kkf^_J;Mn_Pn9ZGSMSdnx_T_$9x-~a3cXDiw}
zhHCd|mum|146jIbAzq}t!<u4vVrW$*nVsukDhljxO><yp4TLhYF~aX@pi=V%*N%zA
zA$T9ez`7C3GVOg|;`Ol&c9o5TIx;@{0#kDFObI?>;t5wQ0_vw9U9r6QjSZKYgT>5S
z0xd|($#<xn&8b3)wAOAC_yVhQ+67--oH7&{VjP0jcsBVJL~Ds5aEOsJ$cPSu5F!o8
zD=n)B2ysD$xy<>qw^D)%%_}69sPe$K40+}d0hkpW3q6AkaARvRv%#Dyoq~<8NA~!K
zHs-WVB{|Q8L!pkdccXHmr!0(jOVWXTxdLw}2y`_g^lx=MZ9{D7IWxq3!IR~%O?pmp
zjlAu7NACu%gP=QNeNO%q_EwXqtzw9HTCz<TX#>r?|A><Uj9?8@*-HX>Mip^TY6~4D
zAMG0RmAj^1gae+5M4UE=Wn0hm>V-F$7#x;fR#J!?EZ(Y-YPFkoJ0}*04M-stf)4PP
zhV-UIr?n9gpexXIdeh;wU<2wOr@Qw`7AaUebve8@q6J6i8Q%spcBJ_>z2((u?^w@K
z01vsNYb7mvK6{pXVMF>Du}(;cK5}J{snDWgX4lvcgnNA5Ak*-QuzTJ|aig|>y6^mc
zyG7J_kS8PF$Okk$^5X=X!&MRn+Y)#Hm*O}rL}@;-2OrN1!elpMCklHTJ@MLqxA7#F
zG+1t;ogac7WE+s3z)Vzn3~u7B89a7EK^?7TEOp@t@HNagwFxC7W#2tZ-z+fH!syxi
zg2$BvJzQk@b_0LVrz_!?gj|qB@7GMNWZUo=CN8(JVMAS<W)hg8TQ{^>M+S<8!!2>N
z9L0-GUbA;PVduG;xwY}jb$r+>!zy-;iDAXo#+|sY87yAg6{2_8H;S4%$jw=}U8E!g
zsM+pkg&0rX5vp<71QR-Uck8XaXA<s(WF^MQZC0|Iby87c4uHg~oyA9Q_0xpzE97Ks
zPsc?{o)qRYWk|<um93RehW62p(aHyOAy<R<@Lvv6{Z_mV@v}%qE{mkY3`X&-v=1ln
zb)CWBGivsS)OYS0j^f`#0|uG8&{#mD@qSux^(ZhZ-$PXrR|Ho~<f~e!P=#s6Q{U+~
zNo$%kb4r|$F}b5_l!jYS8dcu(=tF8-Yifo=rJ=>k)TX4BnCJH8g)p|bCZx1T9sMfh
z?)m-E7JG|MbY)y+lvB2>QSyVd!=fhk5|WIxT@8!pl!;-Qzo`oixVuNo9E>r4wiP*|
zxKFJfj$SDm7tL6voYC+1n6R?3P_eC>DwGw`mI>1=#?<}5?;jA;^g8gn<(CFOL!TvU
z-lus=bU*D?A<!2~UlD%DWv@Qp6uLhk=Hh!-qTBo<EF?1cTy#(NPUHHnMfk(I-aC0w
z1MT~XWLB4|!N|^MVen;cdsZY&yWw(gI6beXp<bFkY=-bPj)6Me3dhg*ccsJu**$=i
z?+I9{J&IedK#7eX6VXZnvH2C@Qyy%7PLtI7@lK#}d|9_z%Lx4f=fY<Zu$hPyo62P1
z=*cswF}vKbeyZ|WQ+F&h{lXt?*8JUaUtnxQ;Seal2wPMWWb%q|H&?eo=Lj)0t^b{X
z{sRTCu^&>()GaiG#d*Jfp6FVrTt{o4zwy4l7ET*Qg%wqH-L9#F;SC2%90R9oOIJsJ
z_TigDgfG^x&fm@1TJK}sE22o=`*M%9s(Bb9bY++Bahjlx#WWcS{erjQ8xNxNambB~
zw{#m4@=v_txU_d&O!`tn%kUWDQ-3!6s9)~*xfC{>8b{l4dohH}d8`rNqaryon5*9J
zO~pjT?TY>i;v{XNrN!RV1ln_W4+HJUwjWXc1obx6QziX0mY9;%{3xK1K}lzmO<OD+
zio|p5Xp0L~9#1aIEblU}T}Gwk#Y%jD<>eM*^7h2_vf<~T(cQ>)YKyM>UU1&fgrJbt
zZhNuB(tCKfP>D;UadW1iT+D}$Y`#+uyPf<L8%Y(|kM#e=u1vR~V9VUp&%WPPp1mPC
zPPi6IJ5N(sx`XoR4E1hg-j7GatiKH0<2gpVVz&GJdtLWi&nnJfg3Ucm5K$6v1?_P%
zQ;WPxnk1i1XI$CkTo=!vX{NHA$+mmm4Xc6SHO~VV;Qod%1|gE>T;(*;@A-ny#eDMD
zLI<Ir=DX&-uTL9{(>xU}ve<l@=<ckzu0-4J4#?F%A>qAZRJu}v^2Qj9UllGFpaz4t
zgZ)PlPp=5q*?*HxmFlB<{|6H)M~2_sH{B(^qSws4!HPQCPS4}>ex8um#T3CS1F)v+
zPay_r?GI<N2FV6I*(B3#PS|O`@RItC(BoJMVOZOwtWn76C&MOSi{ZktdT50-#yR73
zC~oSYFh>UYr^`f_v16vom^#RI^Pb|%{QB~C?}9mK(jlf{iOoDV{q!U}wmkI)yoTmk
zNG*pde9wr3AVJdpC6pdJwz;lEk*x3r?isoH^*x=-q1FAsBlqN*_3e9WBHKH;R!?7*
z^8(xEsUy!Y3430e82gP&qxW)~9^oJLZa{E;N(Spc&D)N{qx9VtbP7g~XfAV3{#-ZT
zkll3~bqgpWt8~xEwzpEqF=-%=C%(i!n?@ePajS;xSE>0xLJ;LW_kT$Gk@((%`1xDd
z`NSlUEAxSk>F~Z?2T96PyN20}AB4MJi!s6X@ouxn=(>QBN&Qy{I6|)MjV$D8=oH4a
zH^hp-ax=q?k8fPv8-wsaU}GOJcdR?xs*Evx9ol9C>4|Kv#Ig6)KeRH~k!R{we{5mO
z{vECQSfGQ<{U`F^2nq55YIbx}jfUN!ODuzF9QkkWc{GXPC|-r=JTSZ_PBcbj$ql*Q
zq>yj~?+8HpKsC@=NtYp^5}M@B7a;iKE{%0~|8qL8UcrW(X^+PuP2YJN<{qy_!ZQvI
z+w8qaU97rY4c0;(#8<DYK?7Ju+JbfpWlA|CPt$2W9xYccL&-)2@gWJPpi0}_5Nb=;
zYm_>^zDH>7J|dOkgC{dlhqS}FA4y^@nOKhqO4HGbd05kf960iQvKQu?-Fdd8*+%b^
zUerkL@XHb!VeyW>s%puurzo23_=b+^&*s7sSNCiEn(&1Wl*r}p2dymHd#zBBm`6n6
zffhK1#{4Xrh>N<DsV(p3vxgLtyj$XS8$|*={ReCurIEO2IF{Q(L}blw)bwdIxxt=@
zANi9(J2`x{V0<FIu4QC3P*0!ue?%8#-nfX5t*GzKBxCP}-CT^rgmlwju4HLqR`gK~
zP1@6T<4TB&ru{iO%x%cHAs7|Kb{l`;oBv|aHt!!Mm>g!@3T5bOYXC31d|Om~bBDM?
zTvrx0&d+Cx_e_Jk-Pv#aFsU_n3&W;k^JZj*(yfX}!ydP$UA5vXuj73v9M|HLI1|tW
z&_0iqf=ic?g5|H7X|wj)8^2FI_v6Mewfv5YbPt=mUhK|Fl@>&(WTmgKIr*m(=}yf{
zpHm9Zxhto?J!&w*xdL*ftS2^(qRPvSn((QZaEK?Ht<oHbSIiz#^@wQ5;|zil8x>jl
zQpTS;5uw4Fw;=GKDQBh90eW`xHs!GU`8KzX=(dJb_Us0`-s>V~d_}oO?;l;B&l9wK
zyZ(}=_}w>+nLjzSJ+&MfMLFbRE0giOs%1ySCGUZ|9Ia+2yTC}&;X~2jDpwoGa)n?P
zs&|#!9HuS%Y$6eZV>)gl)E`|GVh+7l#HKHkMpVTIZW;4V+}^^AW|XepiYki;?{O*L
zWj`4~(&seW<dGbLI}8!V$C0St`wWT(h{v%YYIlIsK3ZYRI#SQb?#4w@mBVp={N=&t
zN8Po5r~~spS4{8tWD>tkfZ^ZdV}aQ&eOFBgeiJSFd^}1>TVpQk1tdHWYOQl1uSHs)
z${3NW)fa1_p>bAxPk1rHf`ga9jwo_&fkOsOJX2gq5VZ}2tG$n`?Q%ZBVC8(M#|T=>
zgM|d-tV4M>xfB#P3ncLsSW?ZGZ-}10eJ3_;yllY`E?Ocs`s+tlz<4qKRtEa3-ZF48
zyZ$1c-Zed$kB{qrBopx?UBDk59aH9}X5<@6Pf+MpfM|vxmszCG_FBcq{iP=6;1!89
zstEXMo|gDxQv9Ah&}~()K^?;bc_S&92sX<;2?O0a{JUy3KZENs9*8Q~LbKA#%1V{|
zR+Ig?Nl{^_ha=sNjn$fn6;hPjTvug!=UrU)NA8I~@ldj1XO`teuFuwnb^U5>0t;A{
zqC#B#50uP(uR%A^{YUu{5;a#WHED5UNdu(6O5=RJy>X0KG`l50roS3<xD(GZ`(n7N
z?9FQ~HKeA_Rn5{UpZbfh@TPYfiSXAio+{~JYBlMOPBaZoFp_5D)b~Lyy7<Jk!c{uq
zeAmj%SYeBSbex<o!~5r+8P)EEeIAtq(_>a)U!r;nPmauki-enHZ9pm;yD-=XHxOT<
zZrd24Bz9#MMkgldI4&kkeKr=mH{J<{)m<=Lpu96GT!HBS+n3K*QQ2>}Ch48WsO2>4
z(+9@WM;Zq9Zks^;U|UQwQ~mnMMU@iM=@xmysrfxgE6F415Y43{<XGps6sZ=`@rAOP
znMnxS6i~G=T;%`5V<SVHCSPp?j4p2+zCY$E)YCT^PG>=W`LQst{xh&=OsQ13_dyTx
z*Bukf2~zQlZI7j}hA}0#L8n#147q*3GWf~PnKQ<xK6}+@CBQ*l-Ny7vrshM*iSp*L
z%Df)Lu5Uep!9s<PFRnA0O!(Y$DX1~MXgb9vRA2?=P%B*p`q1!_*RhpNqv<$x!f}Ch
z3>p5vw=E2I%D`krI4!aGdxU34IIX6*QQ%y6PR{B#<71-U)n!`({kqN3d%OZgx+MxQ
zMKPN;UPr*zxGnj8T4aRf;o6Yg^Zj16@u1nEK-EfmT0mIFc}eg6>T9hnP8!b-`kE96
zM~x7>O!%qm($!ux6I;m6^hC;&o=P_N%6(@$$po2L<T$kzDZC%KEZ&Bb*#sLRo?so(
zrLG+Qs=kO8I|I;2A7c9r?bS9}y{ru9=n~Z>mhU^S_Xvey`wTy~f2K#&7iPa##|{$`
zwiH**CMVow-Ry487uNRZp{8t={rW~&i<N1dch_KSPtw`79?s!bdD&~wX)eSeVRNe>
z?gwL&Sa+qGd*x4iHg62w_!D6fk4nZDD4!j3m3IZG(l+G_&-+zUF2AC*w&1mOV)+5U
zd*P0lf<A~<G1t&IaZ<7CLF`PhHzMC*4uS;_Ktr=$!We!ib@N+xUG+@m4B~~~RDl$2
zNEOKnzpW~q(B8tLR^12l6Im%CEQH38vdY=W(<{`*kXkoi72P`%E2loo*@*1(>bbFF
zVWcHFfjiMv{d97~>rpMGQRw^$_wI|8KJ>%OyGrXF5x_hSRCzUv<^BsnDII%%1RYqh
z>gYcujuTdVW<!#8syZ%VbzCmv1HhgA9lvUUb_zoRL*A9CC|5Ah)3vg?fD40dsnd%l
z+C2A^JBaMepEV<YVVs4(`Hb651hvJTt!57hUOycC#U>eHN0CwO^eWjQC}j%|x=cPB
zvgxhEl5$~RhU|@lzdE$EY|{s<K+TshRtM%@FASgv4!miXw9;Hlm1Sivz$D%vVS%#(
z)MpR`(AgE^yimFFV<e)loz-dLZN||x8<M`$>RXI6O+8X)l5L0~!OnDNqO)xfWIuT`
zRa7|*tysbf-GwoS_y^bIU>UM5nU#+CkM9LmY~z30XN;v#Ef-Jlk*&;;v?)|tCVycR
zM1Bg7?Pd@=+A;rq0FcpRX9nE|X4vq4n`!wkb<E;x`1V+I_QFx8#NPR{0GkmE#jg2!
zVJ4vSR@N3uV&P$DjE_YbfbUZ8{~r3KTV$K4kRg9_rMgeLGLdI6!a1O7<GM4f?(HX{
zA@x0(r#EM9Ir^kut3IDm>5Fau#vP!zQ$mw+m&z~T_derdqs6DYwJk>xcM#&bdnE|T
zKFvN;Y#Ib(SGdPD>2{#;+v?7%9)ahHs`Q397HHIG=3V5NUL&(KH|pN(s?gvF;m`1%
z(q#PjK+M^w{S>xAXX)uC$$W9tK83}q?IUnJ0e^LJz>vwf^%@@PwDy3O>@5vR(78Ea
zOl@D2sR*cKlTR})UAMKdb#ehbV0B|U;gL#G7w;DYb}0Rsjaw>=gE>DzhC<qaZl&a%
zI*3z<aN2!egcLxUVPSi1(Va(hE|s@iGV^|F432g!n@*Eug*XuGb9KOn%qHymMlAgm
zZ8Ri1#LJm03HCHYh@?oyV5H7dk@W(SsYof~9T!k@6lpE|j<$<P=<U-MoVL(H)J1!A
zy|F%m(<}rm<=S%;l&`(tQp#b%Sr3fhHux~%c(wZxpI1RMz)MS`xavD=e^+g1Y7Tgp
zP|XReu74uojWm>x(#Ne_v~HC&4U44gh1+knRz_H}c)cwR&p``k>DmwcaPqXuR3D^%
zrpJUcrbR-=_DdWyy#J{*?!rEJxJ1k5wvS;$VlFYq*Ng3KUD48?468ycUr~Hs_!R^o
z^TbQN(egUaEvi37i#yj=TG`BN9aPazRAo&nFBfnX`32Z?zIr>zoaLa|sQc}zH9zjt
zJ+s+l>Me)8Rr`RuVAi(-YfdTg+WoD{nw=n3IV5fsNv2fQJ}S)CU<U_J_?~S<?uJv%
zPX7Um$te_hiG{hXk&13wIDyL?RYJd9xUl0+R#Wr<wwN5y+x_nS;($aJtHkif6t&vZ
zU?j?Y905D^Y5f)N8{w&5S#fpZN$OY1)x2O;IjO<ntIawb!9_JQ0MB$t{Amz)=Sw%e
z*>KEvFW1J8bQ{v;p2qE?G7U*@E?Y7qjSCv7a%KEr8$*t$TvDgyr#WdVgQIk54lh6A
zqPV55kg|q48(?}yt>Su_nN(3=dMx2XZ+Kh@3h4-6_(YjGz3buUtaf2rqCjR!XD?{1
zc7L&--N;Y8HH$uCPMVB(vq)gGYC~1kB2hbx);4sZLV7&3V#KcE`@42#Y@$*<qXY(T
z4`+q9FCi^g*@6Hr^9VeOx(3)oUS|nK`i|mjG{#<?!XJ1>z{2Rn66YoNX@^J+|9dAM
zBXcpyA3(DzuDdm$`}~FgV>)|by9-qe?L||-r9I_T3VIdVdzgEh2R-QGEskTJL)%j~
zo`sN3B1n<E_9~OpArdHxz3s`6)N`SdVeHXZw2@vlsn^7aN>4)^Pq(C)ovQXELm(Xi
z;NkaXxk2iZE9IWyZ~m5){rpCR9(Q6OGWt|0OMF1o3BbJmzL#X%u2f#C-}9FeOw?|>
znk&^kn`{1Yg&c5DbX&PzU<&c>NIZKRf!EDgJX6n_)G}7p0jD-G8Ae#X)O=0i5h(u7
zO3h8f-7?Uc7Ai%izjCCYB<X@$7U77?`v~-lbi&g&?vjLLYvSKT#|d}LNy4JBTqq2W
zS$4*X#N~>x<F56~EQm=at^RfK8Gx||JAb9!pt?h`xlV1ZIHfS7^!-V3ywq!pyOfnc
zS1_=gjfvr~Tly*{aI!x1E1jj_u!VVZF>@iaQ88eu@41iiX=UK(a-!!RkZ^3*hNGG0
z-81g=LHzv0%LfTCeINRGx;`9vAJ=7jCM}1Ru}he_eMjbXD-UdrJrSAd(b4a&{_<|1
zpMq?ICTrxqBJ!}bB~N;T%Hfj3i?K6tu#G61JgKA2OrR}I`K8`T>8HKfaPThob6&s=
zz86*;Vr#c+k8%c4q8KV8{G}O|0m}6)TS7ZBMcJ650AX9Zd;FG)<J4&`jA-}q+l5@g
zubD@WS&-Qajnsjd%)k|tZB+=OAuZmw|DXWJgyJ~&{Dx}Z(u25Q6A;mFnAtJ*Yr$4{
z(06~QX5=u$rjbZId?)8~{^ZY<5}2ilS!z7RJM|Tq4alDjuHiJYnpPK*oVHZ?9OEtT
z{f>UrXXxsy)ChIc9$luhp(udldaI0^ThUA>_XV&R?e>)N#7W`)@>6#=;lMSRC(~Q8
zBJu*TxA)w;XOgKB;el>K5bF5`xya9Dqw{UNEzwxqoy5>yIHWY}(iza#OnE*=`^`yd
z$hC$Tvd@!;lgpvDYKlPAU@PH0cY;xfY3KQ<5UEAWBZAQOxDE>W2zg6JlzgaD=m)K!
zDBhjtAIl_6KFs_n4*zzup~0gl)8m7mzHA#Iry!JfFyhx4kFdu3%uza;Ck)!MTG>6g
zxC3EtL1OqV^Yq;N#lDD2mf6w;ykDGIK3BmGwZ1!eqGOyxuGmIwMDcb(AtA8QPA^kM
z#@gj`>=QP~-n$aJmAzP(GXMs_13GvpJu9DlE4@DG4}{~T%B7#F_qPm_8(+N6Hih7T
zW@+?;k1UggdZ$O@Bh{020tX^BIvs=8?Ym7<@uIIn`<W6M_3k~YEBXXm_%z`t>>N+f
zX5Av3fbD-kF?MVqgfG(TX6J5qj0n!cP-?$`H2W}J8;@f#uXhN5@=Q45vVh)MV^(NY
zHZstHPbEr&G!K>2ZNFG<X3+RVdGA(!;%CoQ9uzFn7iKZ?nZ?{h@h&^ZkLQw#4W^4h
zz*5=q0d0n<mBIF?awfPSfK*7PjXY3xw{4!rDGY2Z$1<KdQ;G*x#$gn1;WJ#fjRm!(
zOtT%QZO($(WZHwuXe?oE2T<ZC$OyE!CDpFn!&0@-R(=h!FPGR!p<8|J5n*g)ke}{*
zCa0t%yGOT6FY~czb_DF@Lhp@|Du%&t9vt5g*e+wdij=?M_s}qq3v(4>)s51{;Xw?*
zrrY1+67XwO4#t7p2icvwbbbB1Qd!z>SYU+mNO^+Uj%6=L+$Crb?2!sVKH_cwCmA)w
z+&8K$3O-1xYa@HmjH}4E<H*D<O)|~1YrnaR>$CGDtH4(#UilrHbBA6PO3X*v#y41=
zvMq!6aT6@$FzcH<TsS>uj?-IlK>}gH^#HoyE$cE&OwIV35Q*EYyawBUU7i!GYnEQZ
zO14B>OiY8J*-`ROw`#u>wOd}V<zQldZBu&tq7C+bav~{kMNxL+djE~*iLyIUTH(i<
zyO+kjM(j>vCDu4g$etAuZYEnNCD&=6z*XTuiqT>oe8BjfaEs49L$>oxjs7lT<>bzq
z8=aRd?8f35Y)hw(zjcdH#ajL_ImP)M$Y^Co(huFD?$Ir*n8|f)Ro&6E;lK0Ktf<!s
zi4M1ElM?phCAhzG)aNw5$z0c)cP8xgHH%|Ex{NPc5_Ld-77wr3fnx~*-wa#R+1E^D
zvg7v<5J93m?O=;dxonzIGy}w{|L5SLph{u)4?=GkMhR0UlBU{-D}8H3N2Nk|h-57Z
zetXaTrEejNzER>wP%e#ZEc=6fPWkML5Y>LTgg-WvVaacR-s-J-fi`P>hZJw9T#;0y
z9BsIFV*;ju_28TQ&d!S;UE-a2*%AqP0xDVgWU2&T82w*kn5`Jz1PU_2v{%ljavdI!
z2l*2r#1-XSRe$pYBomTKP!sfQ;D+Gw9xIyf3{<l0yqxyBMDI_gDnS$)S*`=9OZOH{
zRMHhu>+nD#cgXMbqns~Foct%2vxp~jEbjTZ3LJR+Nca#j6qgnHWO0TczXZpc2R8mM
z8;%4e^e(`1QY`~RuQ`g~h2wjDzI!qi&yLZ<SX0SxDi6l<Fy}cCiSZuG!NZ-exhM&+
z=*xJ+J%*(lN#Y#_R|Fc_m{iDDZ_4?~k|^l-x(P8@QV{d(8NR<~qKTI%{Hf*hm5x^(
zW?s%rN;y&qgg`4ZqfaOOc7fz&x0c;XEBOoP8vaWwg2j$n^s^PQ`Xpvk5_7?8;enj?
zhC@`;l7jH4S1}^_FGu6GigP#L;>So6htI4tSrZu?5q~jue<DSTJCOC2WJnWnDJ?1b
z>U(T)vRde{EL{f#9ZmXQ;{a&FK9nx9?B_rO^&C9E*p863_cZLS7qr7AWIsU11~Moc
zMNYOKN~N9?TU#4;WX6H*%+Oa&@<+a6t(<Ac14#gqY$-EEW_MW;Ql>S!UT)(KNn44|
zeh+{EY*MU|VnZ7K2HDQ6T0~!L3j7VJUE_1Y#LD*(EdDk)*Eb|2JP>2|;nd=IiY!tT
z7QANEgjp|bcONJ=Z2QKB?AbGBg`BqIGD!vIN4Kxgn}?MV5Ww_;PC6l`v)8CES<LEQ
zHyCOb?xfZT_xWY)s_7&~Eig+|KbKC@Op7E+j!B5Ow)ZiSRcSoA`^=bf{)hAL)XNx#
zwaejp@11dlct{zJG9?VCF`p-_ojpz%+YsHBbKgH{$UZe1IA#>-APlgk{CQT#T(gG1
z*m$CwqSA?H$&D~TkLzD|@zH4ay6ot_?i_o-ji^mvMw_OuA<`nOR7uBcKg5Sv>rGDi
z&h$#sL2%NU3pCYdbO(Zh=o~~bJkuz#hN~u|DnFo7df%BV+E;4Qo%l(>x3rA)8hqi^
zobr!2kCdTSoP{wJX`w>H;cEuHQ3<!)W%eKS?ju&6Np*gS#<N38VdL#-qEbZ@6&|kA
zmdX(M<GfQl$tKs;2ZH!-uWI1A?u-(KK1a>G=RnG?jE#r+kH0o{cv`#Xj^P_Qp~#}#
zs~h5SyWGKsmIR}`ojqxH$yIyu78c9+1lrwMMve+5B>J7q5JmR(X|PT0jfD6yXWL-D
zYiIh&zSORLHAHT}geS)qdZU4)^&m(uNZ$U1tFeb>GK=}xS6}_7M3INGNr#z`Q>aEF
z^&@-4L!uD+NBX~ri3VZT!+yhoM4M2H_*9|Bk3Ft#F@Kma<ZcHgh~QfVr#W!Fcs^x!
ziw|$p<I1-+2KVzYy$ep7``So!CIn-+|Kp|yoy#KL;-xINi7>6KiCHDH$CdZWOpvE{
zc#q8k;}lYcM>g22Xpd0Wfeg&KBRNj2$G<n^70DuZr#_X2SBzt-ELFLeijqZWa2|IK
z!&B!gZJ(@zyjxSKR&Uablyf_P&<%8=Xb2!gSV@&1>zUFvluM*|gC{K40v~5+C;Q8$
z+pEq7Kz9Q}Hh<#P;<%}!Wci}$fWH(<tSFcCw6y5l{9V)tA-npxCUsLvg|qgDZp|7>
zStio1PHs+l_HtD@uf-3P;t(A4OT~$YzaHgVOGfruxP*TjcC_z0qPIl5F~|Tw-X60W
z?OJgLE$II>yZ?oIo8b|aa8&aLj6DAM8CiM+vy^yWNrc{>Mhc8J%K65bqTWTaB=dK$
zjU~>)UlON$@Q3&>Bv#fYvX~3($Qj>h)Jd{IN7!Wbd)uJguoETB?XRtPT;u7nE{>`Y
z>ycQ_W&A9}WG+bSN5W*@#U3eWYGu@}g&}mqKoM_%<dpA6r$tBqH9Io18y9ocH!{f~
zx(TY7D;`20D4}tEMHUdn><CEj*@hd5MmLJ_8j1{aC}Q5CD*O0OC5t~o2%Tl%;S?EN
zVQ$=fs#@fHuexZ)Qh*ogDQpvj(4&Suv-%X~`L(>yX^KDkOU71*aAJ2y-1{;E2C|Xd
zs<)2$mXC|qm(My3IkwH?64pTc%KFup_u<5Nqd;2qd7~-aAL07s>6}z|vhFW;g+Fe{
zu1ZNVyHRYC@DaDKR-Cv&q4U8>&4i~2;<j_)K&-5a9hC9Q4K7jyu~z`P<>77iS?_Vn
zCA0p|MeeaPCve&7Y#4yMSI~nV!D07SzRmkKfB3*m+}68VeIDRy=tZW#(hgmYQ!@Y5
zF?lJv?Bh6EYc-I#El!M*XdeQ{4>|P69jRou$r$7J%k=Ns_YYmh|1Lm6{BCPwy5>di
zjks5-A6H{9!^O4yiqYy-?r*c}(!^O7gcs7wTP0ZncwbyGYES)PGbh2Q3=~DWqWDT_
z=E_V8mjZ|=-QLZ)dRql2J%17x=(aQ`LDfs-GUqy$An)*-B3_HJmxr^|@YZ#FW{F$u
zUg$m3u;W+%%mN6$A1wcj+<;sMJStzSymCA1V80B1I#y|D95>srrI9b>#l98P2!O23
z?4UA}F>d8en^|qkAi2h~Pp2jKuf6aqZg|Fe^bk%(+{-evK&A{C`I0T3c$tWtCEg5j
zTj#}%q(**f?q+w1KgE=8s)6S`n}^$46+ljTr|zTBH<h_Azpiv15B68$+S0v!bzg$x
z>PTC!of%Gg<fXN1%1~l}G5H|~pV6=2TsMq~8-Rx45}-ne<XJ*@Fd#n3iQ+@LMuS|i
z-ffT>s;+(w^OJCLw*6A7$crpe7mB@OrkdcWJ%hqXN+tLc<Gb1S0+`#IU18tGfNU-L
zy1ubs-Op??b(u8$oH}!nKpW1=z(_oyZr}}FU55}rIz*jUM46q9#ldC~{R0PL;w3sB
zcH`tmElY$caC&Jj{-`D9c%(Mpcr~xCyy4|jh{8PajRhU4Yp_dCw>u@ri%pYTjSeLt
z;_+Y5zpj%h>6Sj}D+L+0J|hSDb2%ZTJNr_O9A2cId~31+EzMt{#bH}bWzXcKc-@12
z8`6K=P3qF{7A$MKV)Oqo_SRuhMq9(agoKig64K!SB9hV#k^_hcDBU3--7QF`gM>&U
zNQX#wcXxM+$S`#H?eU!Rp7*@J>-&B0UtTg?I`ho)?7i07Yu$G*I_V#`dNV(L-Zxlx
zHO@a=1u-17WXr>~8dc^@eyLy_kZybjgRW<;Z)2a6SZY}&>kXODo&b5(Y~Om_eq}PB
zM_9>PO+^87GdbbcxqMa4HhvuA)^bjwspdx!Bcp=EVkL7&Zj?bE`8MxY?zR}enfKwM
zBk43eJOce`K6}zpjr{Au<$iqHNwFpT!-PH{A~`*<5@KjLORe3KATXRGU9`DiGKmG^
z(fYinFIB$TGLMjB<fdW-ZHO=}{o@>FmtO|4O~-K(d5qca3rf9b9j>rOwXL;Uy{h?L
zz2GjK3&&nw>ZRbKo}~1md^G`tzp9x9>!Do$g2#)XEB^*AoY|LNvT>~sO5I*d5QK6{
z7EP9jtct|ddV<^Bj(F`S?A9ax@{PMs&a=EDFDK3fQ~qWyROk!JMjiH>pXI{W3O%W3
zRZZYLrKMPJ>a6J=Tyr}}&m@XQ26M6mQCdS1gD8V2Ex$B)4VDqDgHFTaa`YfwH*GWY
z*xNBk8&v)Fz*dlL@{Wv09OxIR`VCcHN(`bHs6lhMb#9<D{wvxc{Xl5AQiJN%cEa<h
z4=D|k`;@ThT1Y2Zm-C%^DIINSF!IO;)QTb=pjztL3BOl=6dX8<yPXu#lBAfWWI^ij
zY%nkEJB3uU&b~T_=(ahU9znc1tQJ*Q;Mw->i>MppN&haPGVodi)8*8laK0zx%O0mu
zYYe5g%fkzIF-e^^1Dp^7G5r2LuE{VAW0l!3t^kC=Rlkm1{+dr~eHnMW>0&D9lpN;f
zmyrZ)jK<m_X%&f^#GxQG0?ae1rXg2kfw~Y@kU4U8a^0F+O3OoF(?hKFdjK8n9P8Hj
z?Vz(d5e2<sC@;zS2LEwQC;P%=sLzGbh3)K!%e7lKgM$hP1&RR#XcsjE0aT|>IZl}^
zt)B*Yz<LW~d~t3J8W!W*oag%v!1_eI;cShA^G%VxVVsqeNaOXR*2nf(5AdFzMNM$w
z5aNWkhoVU(Znt)$cd!jT^1d4M6VkP@1G+6fMA+}PTS@f?pyG2Af=E56PCG~v55<6J
z3nwhqq-~XHrA(ae$Bo%jHhkeuBC6D~a21<c!TOA(Gx$D=i-r{~kPTV87q`RA_@X(z
zv_&p!L7~NEXw3E>-OvM=95T811ioaS%g~>CnJH3geE*7eTWHFlv~`_%>(DRFU+~x5
z4rIsEqG>00)CNIja6}bkWnCsNrr@iXrm&<?kJN{i8eKgLu>*~9o{#XUE#)z;p$a;9
z_%9UT*ooG_i~Ffl<9Z##_USt0wm@gl7vrBq*m&G$siHJBkK$QY_xQU?Z%c7ssA^b(
zp72h+b3a#s2uw~&<Ck?$@DxQ~6dN*%HyEzmcE;t~CJRZ2(a~HP@^>=&KD_9WeW+PH
zl#QDd%+=136}56Z$ObqYl1}kL>bo_fK~AW+F0+U~IA@cEM+0?y@z$P=H6250owc%k
zrQzUJiN}VYWHeh4hXw<cRbg3BNi=9Qid>4hwqPQYQyYrSSQDq(^t`DKvT|o{b6uGp
zgp8Gr<L>g>P{WDH_|=}qVHjides&rqi6cixP&m`B^HGvpajcMw1~Y_*2;PR0aY*Jw
zN|Bg5WGaiF_cPu(4ysJq7dd>|Bc#yCJck<0_Y{cuv@tHbt51eKLY?s>65p7d{ZLm}
zMsZI%g-1g-;NvslG^b9z=iq+aXM)_K8>aqXPo{SZr{-^xSilYyvonvtaa~leFE=`q
zuIspz@r)RK@@#N((Wey!s@Qu+_I->gxc0dSEmb{x1^KwzC=!hN)F5Nq5UKELpNmyW
z8!;>*va?Z@K-V~2?^%jnvHGOy=&EU!53^edusg$M3<2%NZZ5J0?<1CaBc8PyU-n?%
z<?ZpbfLOAw5Q9&8Ox0nnzPup5m`)?5;$I&Ug)Z=k7qiWRBw+MiItJaM%Rc*Yg|<6w
zcH~uagcu%okf9M{wh};N_?OA1&&e0^2iHJHSNZo+gI#!@QKt{37s-e|=B8{Wi)_q=
zYwpDm%wBRC2ZMELh5VY5e6Fap_0A_%4kn_D@bd7Ol33iB8FJ&*aleD#uY)Ngl(QGE
z>69Y^hiArUI<N1^7<EU(6yiT&hkpp6A&<To|3&Cc<X~9GmLcT<o^*rQHlot0+6Beb
zG0)f8itgjF$DG~kv-6-SW)Cx5goC<mZ5nu>w)X^%BW}BOJCX6kPYg8YXs&&+^Y!@I
zNyMS)IT3v*@#V(nO{e{_!{Pxd(;4w~zJoeZE?L?ZmYfEI4_*zSbJf4KJxqRmKE=p|
zT6T~4Mu6^zY0NeAUOs~DcJfPOBL0_X3zUOFwYc@U-MHOq{pQ1irDVxhr6N4ax5+3n
zjV(z%Y^)rrlJ-ZZe^u!m$hygJU`9mRx7lXd$JNiLUX=Fr+3z|=>em)Qz~<AxFq*qe
zl%ZyDQ>q}k=TzcD>Yvzr->bBj=WUDNE%N|~xx!xYr~zDU5G(yA6*c8ELlU67V5^{M
z4kwEAv=-mDwp%Z3bB6i}h|Zb(`0k(gUH-nW7iu&K%)RX&l;H}>KC~{Mn0f2w3M|`v
ztE+)aMPsfx>F>N$zVOm*^oy4sQZe!NY{KlU4GrK0gMs0lKkk3pr}$<4z8buebLPt1
z?aG@mf%G0EjEhy87g7t+lu^8}?nwIHBjy{4f8)KlOwx-}nbU30>%MILV&57uoMdQb
z`UOP(e+&PjXh#=o*30|-%y~#CH`39JXsbW#_1FPYV$;5OM<`jx#jJVZEb3$dDnpUB
z<7j<vHYu|P^`2Jt`?Vvv2~DF;mNb+E)rK!`0z?NQFjL%B9dimD&AyE~*+vl}!8+?%
z#JwhHt{}PeC$TJy`$Hx7b3fl=@?L)KFyxl{(9dTy-3&-~dkzfIrQo0g>u_pwJ@eNQ
zD@|dhVRlOV^fbgw*QG2<!ur!mAD5CigV`ClGp)L%R~qdo<Gq@fG%PR6C89E5F&Wk$
zTV|yFzWe7e(D3Hbd|Y$_|FTl7$xU-zpw4MC1|dIpeEtJZaPZqINWm|-i+;dB*$+uC
zVKNJO!7`?5Ze6<F&z<Jm)c?D$@1@y`P1C2lx0?>V>paw1%N~#>@x3r-o%%w4C7?zt
zq_!<!B6{Ss-Cyv{7SdncoAh#`%kNA0Jn4Ium)^QK0Ge6f#*3zYeDtmA8J%HV9(nv(
z8OA{Y`HjG=)-D&<)2$por^;jJ>JFL}4B1`rl$c5e_0#EA2J=*!tD%VJF!?vmmTR87
zpU6v}b@FuY@e;51mk!Njhq3ElV0KdG&|1!$-;JGL?|~<Ga+uxP@9&e9YOLxZGr#Lg
zc?=D#P!2rs6c}|K3H~7NF(5wa%SaQEf@Ah=4JN-nWGrJ=zuRxuJvF7XF=7Sk_oQ%D
zH4`Cjx#MHBbH7@7ZDYv(jrT@FHjL%F4&5iIe&)$O#?4P7NpuM)QOk9mqL8e9zmW37
zd`Sy#S;ZWjo!$x3)3KAIsFzEzHrCG?e9hfuM>G>9zWQpitVv**2rfcquRy!Tk3;5l
zWo~rm3f#(=Y<cX_>TYF8?J-5sexik$_^n3Y@HN;_-cV)aCtJ_4rB?IROezlBoWTe(
zDnXW`^j!;cOQvBAI#<;+TyQm<C0A~c)8WE?ut*r5P^wbRsO?|qka!6u+1}^ac<#0Q
z;w<qcMPKXi^}0sBF8oclg*}pf)^zS@2~mY3=zjkkd}@MW=_?vJ+I2bkAAW8ZMc;Sl
z{nO;vjr%;~ND2h*c@rRxKc4eQ4skSlGEqkOHmtgj5LpnhvYp(Ee*WWm@*rMU3ON&x
z?)#s%%J;4k_fRH<Z7nj}vD4A1M033oG3L4^7OqQiy=eYb9P=P{e5je%Mt}7%-595`
zBF1$-`GgR6qTMDL?R5>3`&uuhOkro>f&QkEVK`>}&7%o;NA&YezfWvfkxlOe55B(R
zdp^8|J=Zyt!I0j_K>Fiu9O>E*j+^k3{jF;<>6G-R3<(U52=lvYSN!k`chF@~BSj#P
zy>X3mZm82bPh;(X$^FiL^U+vRi@A0U+<fD0NsKmm8(lQF?Q4?VO3mxiC!X&X4jk&U
zw$*IV(zyfflz3X4T5~TuRN{2+1qL+64w)@)9P?NH&Ry8oTk<a);%D9u2SH=2)JQa!
zip^cvb>dSmv6r&Qi|x1X30Q;^JY>@3#VFCQaWA}&61Vnz&O|BQ`D|44rXYya3|1a<
zOubWsyYH#4A*UeeEq?EiW%G@qKi2=?bo<XXJdXmIl}&z4WkcW`Y)>!P3^0${+#$T*
zA4l#t4uRy?d^(DEkmT6~XO(~;j(+s4D#F1N+4hvZD{ttT&rYQ>P%v^Q7~S>t(l)NM
z>zuI^sCqszNFw~H*PM-X?=G>0Y_s5RRl?T+-_>D!zK3^c_5$l_6Sfi!MnBp@{H`bE
z-4u;B=vXs;d$pcoYBU7%bdvmZTf$iiW+-UvMPCZGYK=YXBMaQ>Je+RM*tjXLZ$Dk<
z-y-Kc%I9B0esHp<(f-o{DYj<nA#rnFE%1o?+wFx$Z1Jf1t<3D)a8q_ye$o2c4ck1#
z(_=;6bn16~dGl%e5g5Vat+ZaFRH?D7H$6SQr1B>A_X*h=ElDnmJn*%CS>^K3qenIg
z_Hi@mh8^5FefLe>Wd~_#tg+?9!bqRiE32len-X<#!_L$HFkQ!C7qG?RV<kl=N)0;p
zVjP&eNi)NzW<|mg%en>+c=4Du2M0oaEZ9UYRi`1gM1L`{9o7xFr+7?+lm<gm{Ylw(
z84gHyLQR*debju{i2A|~op;<X7TfIukMwq3s$Gd>M<LkL%kEEgiK&Y7`p|hR*qkMW
zpURJo?f4`HyYf^#a~sNx%nxIwYHN4q6ppT?O0d;pYISa)T)RR<oQZ__hVMZ@qut@9
zPzXY<=0^d<zE<*L)u}D_=Vv52r|fFRz*k`@?SR|_p2f3NpFw#=ivTwlvV)fR+6S|2
zvp3a+vm$W*(D2GAUiJ!4hDU0F6{=|$`{9?Cm-5&fokBfsg7sN(=5>NZqiJj1my&Pd
zK2h07%v|c9%<CayC)q|-Pc-@PEAspwIo~-as^Kh+IOg47OaP}>%j7Li&Y$L363YG|
z7k%+OUE7w#Ubsg1ZM(gTqjGY%t2%f4iH*=T<#`LmyO>A4t8K}w)Rlb~U(Y<aHxYVc
zHZ8BNeu!0lAX}Wn(u*i~PyX@jvcPBP&|8ce?G(&KHc_B0KF@b+NfgZ;n`~ac?T%kk
z?biROnKXK|(yqt>*80p%(%dzDIN>5$_DugNiGVQr$78)cm!|2MhJos;(hjWK<wokw
zbL2uV0EBL6Y<TE0wm><?r}94j5zO*o?1&}z?+WYisIfGPG|%JFi4a<WolXYYs!fZM
zi$h82hYn?o#d=avt~_p-CJEIpXzpo^w^hE{&shWoHV&>GH|VCbt&t;@#=rA`lryl}
z`Y=;Pv35=Q%zIdN_MFSM#?2<u^$tDN@AM+(xrhAO+-|qq;lgMK6-fMbN5%m16Q>R0
z7<)k*Iw5!kh03a->kDc$#~(@84ujy?bqch%&8>;HyG}<Ht#G6NvuJe=;ePtmnoi=&
zr+CLjQkUhtk3(AThDqs%gPOnD;?F~EF694Xem*Fw8MF!NBvw_m)>9Jb&c>y$(D{8#
zmZ&Q_qezuzd6cCf?rikxaBqb^CPlmi;phJJ___^6&Nh)t6Y&pw0qT~mM5wwY7Tk)S
zE$aK+k%6!Qb-<nVM|jx%81LpRIDT>;>~^4GQ@(Jgf?rmS`7fCzc34>Az}+`D1h0F1
zc?o~}?71D?d3E7g%znv<X>!-|s93W`96u|_rgEn9RLB_*jWcn@BZ8jeDNl6$hkdtK
z23})Ne>VNOxk*)?dj@H8x819zEnoPa@6*>opFl^u^YiOF&5e@egCT(}%0E7(>J{!(
zRy7m#y$?2z`=-M4LpOc?_9U>qnbEkv7MI^_^xz2bD-x}s*ic&~BEW$>-^B`OE*mm`
z=A&yTY5dTvd95K*H+U+rIAVY!N8t>{e>FtoZL)2c9yX;oc@bZuJJVp8lW3~nKKiyZ
z$AhQZC)sp>ZC`Ft0=L(l`R?eQFh|5eCYF5K-YT4a={i`irGBDPgNW@ME25(f8jizo
z&Uu{Jwb<%nxMP7ct)dOj8YoooT{v)TpjmO}!o*`lP<qC)IGl)Ofun%?sKA;^>O*By
zLL(Wu8>(y1q0XHl`umvE;OmgadIeFd+SfGF=@@9_=q|Ui)Bc0ZMoZKjKvb}26P_}v
z-U4cwOi(l^Nuo)+V~CBW#`H<<r^UNaXhvv))kJ8h=4S6<JMD0^aP#iF`*nQ1FT9Li
zk!-xg<bpJBKXU8Amt7x}heMxvM^}yI`?((2mDu>}P4lWp4y||sTiPp}P7IX5h7%AK
z{`j5@VlGzYUHu_^kQs<Y{&5~BNi7Z*iic0HD`ocbzUjC5IC^8<M<3q=2qe7fY=0&z
z`H@ri4*tWhnHGP<4z~ViFP*riU8kHZZlxXwY`rO&67cjrEj`gnJ1#UYIxZaYi1(!L
z^?dhK5j)lJhlokeK6OfU5L4a#ZQqrB>lEin`vVuE3$VuYkufsZ5FY4txTU!QTrk?H
zNqaRR;nKFX-HIOE{{V6TXbpRT&j%0h#1U5TWe4{(TaM+p3m)wlyeG}1UPsnz^40rF
zq`r3|D6TJI`*YD7T>0Xd|FActidneN8POd-svx*@F<B*eXbj8<j_(y~yJEd0mS=#6
zGX-W3*ce;3d=7e~cSq&OIfAl^HtU)B+~(E<x%^$3UEsSU`Q0v6XBQ2@eB?ITKaY)p
z5Mh9JO}t{_q=sC&3($Z2H-3Q$@L&8RI#oVAbdXTwb9aloYj{^jXdfBcDQqUbSQh-g
zDynqE$sdQ<r0k`pAMu4sX~R8rSs3D1-b8T?{ZU=|rWRv*G&QrI@nn2Lv;8h{sO?A6
zQZ}sAk;>gvmC?8<i16I~_>KbtM4d8K2P@6FAOMX}1>^HhH`_P@F{Mee4?2YtkxdWC
zMEP4;q1EDH<1TwMVA%Jg<=ulyk0>TjC$>7jW!xWXNs~o1%QC6o!dVMTC4=~i>l-C#
zBzR_kPKFCH>ts-c?nJftcn7kML5)U7ytRQhA0vjmgP3Pw%)Cb}NM|+bGZ1wvk89$2
z{mIG*$F*{ovZ=hl@_DS6Ro_;8Q3AqmIK*5^&TrfMg0g)|{&Z^OnbLND?Q1?&F$>>&
zn<-9@PQx$i-_3>}j1emD$aR(9aGi#Em)zGxCj1e*r?g!cXqjTXfcVA=QL5;0ll!Vq
zV0e)YSAKg`sdR;$qGifF9jU3nU+8pl{vF}UgpKYPqlBv#C{9a+gNg3{??0Gn(?21h
zSsdZ!S5bJ<P^p+6Hv_KeB&aM5E2mZa(6OA>?}^2;G)wg~^EWcwaLzsx;Yz@inXaH;
z;;Bx1Vka&7LA+Z!M1q!slh~qM%no2al7)GaaGpR_9zh_b-Rj^!jChDn|01b=`A0{|
z2WBX{M^0$gq_d5PDT3GVv-MluHEBrd$w^hy!gkfYilUsuLOI(Tq3f~0DikE75>SS7
z6|+3<leev>bW0DGqVnsC6yu?T*aQ^dp`s&eOBZr{5~f+-D*cmFDJKbfsC3uK8YI%E
zZhiD|*tGC1{6DF3x1fZoIk<dNV|Bl33oIR`*X4)SMS6Jz%)!*N?His?#i7+E^d5;&
zUhn!Epb7Dq`Qo-3N=;+R=#a9tq1kltDRg>uZ5Rxq^;S-Uuu^8x(;7@Pa>kNUV%S0K
zdyX=C+{R9Npbjx8@q~~5B8KNC#(c#<N#7746&sJ`s+`)}mcZU$f-wMNl=>$$#4rxq
z{hq|g25g;}9Wu%~B%7M1XHVeJ^pvC<%~IB-IR<HE^`Z(p+ldlPeuJ^;T=ETnCGmA)
zNM83*aw-dTLAMKM)Z?b-SkIJbDG-5C>BU)0(s(??(eqLMQPY5V(-0x;_3$vM_Bg8;
z79(ns;E~?rIAyMu11keBPK`Yo%E&1?fnll%0pwA`O>}qQQx4JGCmTIo8&S1SjnYH$
zHVzns5RWZB$5nmO(8^=G<Nk*`mU*uDP<ZJ%d&rad+Tf=RfUBp(_-*os(@S;z44P(f
zX*|fv3^S9He@JYq*(#H-RY^moajO8?gJX5I*Ev_O1Sa#c5X0v54+IDdGT9cC+LcPa
zS2)JeoCvOCrbr1f*_trSijPL>fzy~iozno)eXN5qv9xDa@u4`jsG4F5z#!>$rXaMc
ze2Ca>YP$=I)6@hl0;H|EV+VB1M@bB|-#$(!tz2-??!xtM0`JaKk8j>i>QA6Bw6>59
z*w9@@o|6jjPD{tyny6^oCx_%gGnYab>2kYSmxw|~q$7@Mc~@Jo*a3}!EoZ+;+eI6h
zQdvzi(i!8ki+|ii*3DEK!~2r@_Ci$fjigzN1rULk0n*ddApaWP)TOXy&gI_r!_ilo
z<|Iez8?Fl%Wk7{OZYsH*%~}&&Pe+qQ93$l($a^7i{$f&pvr(=|4;wKg<0}vy^agTK
zEYjKHGr<;J%Qf;!8@6-<06Sc4(8G(=S$308C$fPCQ>Xph7IzJ&_+~%#R($KJ;4tSH
zUfmHF$H64#X#bxN_O{<G8L|jp%giUl=<bvO?42g=cH6&QbW*o*4CMTndl~Pdof{Ah
zE$grrEXbv$ktJ@sN#FD{K_#J*_t5=AgT>+3Xi&{<O>vjdze`4QI*z!gAy-J~&O3ix
z4|gdp9Pjg6EtZ`T<jO96`s}Gv9}SICdVb9iCtkgOg|ULhMX+@6VB0eQJ1}6cq-pUU
z3r;2;<T_^KbsWZdo|?hO3&6p0Mi5|fgpkE#CyK{cNi)4zuxrD=sjYV;!wk)79pGE~
zX*tw|efZHf*A5)uok_xx$z>hmd1`YwDn#|P_j|_*;nV9I!=aG~mkncegcHzrRCQKd
zudjiI?y5N3^;<^zB{98gKm?Ju09<Qqk^~zP0^%<jePfi5vM2No!NW#Cf`X>13~n|t
z*~Xp#YC(-N?md5Dwn`kPaY8x54v$F>2_|(Dp+7z;j=Mt6fgNuV%G-B|Xh#mQV`D!M
zY~zNgnMybReZ<7u8vs7Q2FwK<2@b9cV6t>B#W<6B_7x<-Px%i>Xo<y7B#P;@GDvB~
zC0sI#UsS~^xszIl2Oa1F981x~n9KLy0j{pImQ2Q@jb!&o0(Mz@U(9h{@N;*(xgrQe
z!%^<`z0={b$QZvvp&q5a$7~9<bu6kRe9yPP49J$d(%XQuGqypA;N(JcuaL7F%1sbN
zD?<R0i~W0MXYILxI+*yK%teSMpQrM|g@i{bONJ*cT^6exX*nKo5^wB@-}7#w-=FVz
z!4DpNypJj1e#d6#)&<!%<~p8(j#f3M<=yf}7s<EF>NQD*S=!O^XjTs`uyNqCor`0W
z``YWWb#>J@tfI2lF?>2qW|Gj2(~tBXV+npHeZ=$z^wrK|*XqvrmI@w5Pv%rdRdap1
z8NQ>a8`5LX(x>urvxv#8#@I9UP9b47A_vmGHB>cFq^@T~z%@`aI-6$ZRVM0;xqPjR
zsSNLUglzc+nMF)rG7^o(8lro{E=Z6I^qA-Y^hDCl62#$K82@d1!@-2&qm#Yjd~6uO
za+fT+FQVRbsPji?sJPbG6o=qLvLw0AQ74foIJSRGXph=t>{w5LI1(GG@<vt}!~f9l
z7W_1&lZoXTHk{zTw;~FeZwLt#2BrfF-1p|6?||9rkQJNAo&zphk%h*#b;?8zd!@$h
zXsJrG%*Hfi0QNN>4#AFY>DrfSIy*?ncZ!t)?T3B8HrA9+$w32{T*l9&atFFUSBJuk
z3Y`qud`b^jcOHN6L7+t~OEO>Tg%$jidq|t2N{y?A=%Crz$%XZ<?w@L@Kapd#91J#c
z=ZJ+ZzTDOr8s&C25$YIeCr(jqr+)v2kDhkPCevp;7I&9sLtOA<M6QRjuK(HR67Gf4
zOll&an3*}G_U$&!<!_F3y}(^o^xh$9et(8FqS;MZo*+5n`qEo*C~o2iS_WInb71^)
zg`Zi%jBO`pLnHW^i&2GWZd9a~cws3zC8|0Gx0Hd}Zo}IoKkc-~6LO5<9yG7;klB2e
zdena&K4jET^xpk2B@^~Y*quO_{PO-r1Q!PPY~pz3{k0FJJ*}_nol-#`h<EHoxINxQ
z9WzDBBKv0U8Kt}OMzg>f1NOKjBBK7KOrTzEQ&@3NZf?|If0x=;{oCtX45Le|_vfbc
zh6p{HmYj&#H((8{Si5}3-!;i9(#BaE-<|FuS(p<f=F+^hJ`cN|pJ9oV{JoLB@%1?=
zh}qcp@_D=$J|sW{Q_ECwO?IEQ^)<p*LyT#M@A}iCpGY$Q!v60;(tn0b|NU=kILl>t
ztIc4@=4ZP<w8tfqiI*M^6HDA++O&>_e)-l+q%61{4L&jekC<0W)sOZ*LpZe)lR#nY
zwRkfz*@@`$`)Xk2#-l-(z@%*C3cY{(fAjxBsL_PcR#yVqxM$+|BxyNRiO5`L2<Vrk
zV=KC4X@Q;K3ILf-c)dVw1?H)Qic|?}jnnGnK#+C<(DZ3WKZ(QcqoP1!@L@@aLzn*J
zf%(6_fnjEJ>%I?LO(=p}ksFQgm3@GCDcGlHLDlLVYJjW(=;&-0qT%8O!n7QECYM<>
z*WYvjOM`%A&Oj6>>%gIP-wdE3S)dMd8h$yKP!QV5K15xg2;wsk0NUtZj3#K!J^Q|K
zf*R+kH35LS?+qujeIP)I=2SNI5+x!_!-+@`?KqR>dr5@@qsDxX!|&cJ=<x8re<%NQ
zC;ZoI5xD4Y;?GZ-z^Jk9-lYyhKZwOnJ|@(+=O`tc0W_ZGp3fId^ffg9L;5)s6G_*h
z=MjNr<1f`M0^#4{C|OJ=#(Zoih@HudF{wSa>4a7SAs~ytF3Dxwf%puk+Jdo!N&w7@
zvVXjG<llTIZBUP7>13*n-4!=L(1O+J?+Wwp*!y2^>^}nXzdtnSCvpS$rJuk&^z`}$
z=$|YBSMSA6Lp2!=o)G$)L}i|5_IEPcU&{|-1j*?B?$3HS!l(0&CsR!~KQ@R<&#uDt
zko3ub9Tk?a<_fL|GPK!kTMe1?$?XaF^-dh{%|{XcKAidAyqo{q1x3B@bX@QqOk;)-
z%l`lU*8k=;4jB-~*hTn74`Kb+E!cb}z6|s^Wx>=Q-SGgntPQ@BF4;_mPyf7QA?W@|
z+~yY=LLkJr3G7xeq5mK0&Hv>R%MhYIKXoOs<G<fo1R6THGfJp{*hBeCx9`n5T%{5K
z|Ac(=N{&!O&ov>y&a*yfIQ_Mk;`U7p^J&wS1B&dSMMwiCy?6ojsmiBIo^}}5pCO9B
z6{rjvE9U?zKTjz|@W;N^6}ad^=&0ZuME`@-K4&Y3sh?)zZvddO`ok2e6DjR#oymNz
zYE9~Y^E5WUM(uWG(qfHN|Ne-P3>;RIPXa1V0qApV4duiA2{3kFMGo5`yufTka3z8x
zJC>vL9f)Jc2?V#({g-I^NPji`@u>sI5^Eq(rS3Cvx;g=HMRQpi?}}^iNIJhHH9ZBn
zRp#1}Jp+*(4lGS{!3WsmzNWc=6NVWu{c8$HMz8m@zTk<WqhM((&Og-LX8QEZ0L`cn
z#3dfQL&8T+9@dij!OaB3zfp`Zp|rN*%gX_|39n&2V1^tx0p8eo4K2$@C?ed&wTUoa
zFHR}*8i4gzH6T8fP;Rh4NB%#r2}~Ka76|l-IQ-WIfxX5=DQTx^Zhm850%YKxI|@r>
znRec9{u9jo$v&cA1{_s06qV}_iV#SnSsl9wbmR-gV9m*ZodWGE`y~JeT5y~7bBtcp
zEnN?i=((+Az^oYGRui4|>E9RuhTkhz)q=wEuzgT@QGQ~h8v)OdU1p#;4}tUNU2X?V
z)o(s4;>l_o*#K^WTsDA}Q|{CQu5ztYraoNhZa|+LrS`o#Z2T_#v$km|;?+PG04Or)
z<COw{-W;$;*7J_qH7!MLeU@Dwk);8#rS<?y=0--X!|?xIHU51s|MNATXmKL6<Srf~
zynp{Po#Zmg!79O<r|{QXy=fAc(1`q6IV^b>Uv<!f>Gr+$A)gcOI$)^s5EBkk@dBvJ
zaw^w(&JaK2*H~-UW9)z%V*`v6V87vKm8<ucyfqIl<Ikl(H&rQsI#fhy^2Q8EODlAN
zR>B3WvDLE82SBKJQaja9eB5use-2Qn8|^1k01xJE?pnWnqeq^f9F3|eyrzsfrJmhO
zw)aI6xdD11o<Pu56DS)+EC$nW{y!HgC@0uMzqEg#c|-WGPYcCBcMkPiBhRX344e6M
zN{c@DG@I#{E4Wb_`zj%Zce!X&X>9BEd6@_rz+2?5ghUNf`5$Ip)@`*Gz7rMfQ7<{@
zydIID4e`qguxkIi$ut8Xr_uH_3LXj=zCE`X|1LN$G0g_QSVojg0tfr(r9J0=tU?AS
z^hDR+XTncSF13izn6yJ>ef}AuLV3_pp3Vi&dvh&|!2fZA5|(L%gN~*OPaPu)jY_jQ
z@O&uRH#55pJe#VnBB>`LUMkSPehWBr8>k2N-CSiX4vHKq#1g<pndH_y)(1Q&7GNWY
zE&@o{6X0i^*3hwkj!LJ0&N~kM0Hmp5!121hWX9KYdmY6owS1(!XxzsiLZhkg{R4%{
zmC=fSZG>wa-VTUIy=&_;K9UdUm%-1smioJzfUtN@8&n9dzNT(9oprGP1d&e#iuvgR
z$i8zx7;FocIL_5skUmot<f-L8WXv&Y7MesQD0613Bhe0RLX!ZcXJ+fQM&V5{UVe&V
zTh?i^^`d0Me8TVcLCLja`<l<vpl{&a3V3Mi5*C1?)1Y0cR_{1pb0&Ea37rFE!TnDq
zUZz!3*VGP(eOuqTva<nOALH<+wcm=$Iv+-DX)hv3DA++d09XrTCg;GKYpL7Y2XTI&
zVlhrAtU?c=u&Gepo2g}j5u}0#^Bt0@n9yq(LobL$uNUP|tOfoMN1xE5*C?J^&Ro$N
z;NN*C7(cZJ*&;81J&<eLs>zxUeBLORulWVcC8#hW`&cg*E_j?pW)7ImwfC%>Q)}N(
zsW_1@<|opkg)3*grBXVK{0htF!Rsp{HN0C`HDgABBD1b~d{e6}*e*czv5Dm7@LP6S
zn1`XAi!Y8~1d`S--~RRCqo88*-jjVA%SWY;_JPuAOZL}D@o=NVpvI(IKHCvjc(E8L
zQfn{K6@zA^^GZM}@X)z>dbjP{W#8z`o)-MPyKe#rVSb)vl{CXE2~{e#3-vO>bdJLu
zhc7GsRP1yUk*&zXeTcpDbzNpUE^UaINM9THN2zK~xH>zF?NjU%Y4fUpg?}y@L>I>l
zqcwoe%zqBA9b=177Y(z_p9@kSkmaO06#9GHqxMSF9MZqGO9Jt~+d1>|TiW-HZ%}+{
zq>m0*kkGz^=%jN_%_W%5fOJ+~h)fy>QwHLX+7Q1cC+Pc}zfAsV;Ul)1s)I93dWJF$
z1GHYs!C;{G5s;RavDN=>ZTuGUd>Tl3y4KOZ%DyXG@&XcNo0+${O39PcQO`!i_RxIh
zi$8!h(5i)${l6mmwt!nQVW2KE5Nb?RMEe`m8N5y|m;Nlu`vbXsIvj_VS_cqL8SX`u
ztx3^b#6WP97|k!AhgdZMxXl;FwoAYoZi3K^8r(wQ1=5Dxh0Ql^KW8sTE#-;JUjtyW
zB%>o>t+oMk?<J5Av>lxG-T>p}aY5!K&K&R)%kU>^W8>WZ>Nwy^%PINT2>l%RhL85?
zfOX83uOIpO2^6j#;s<04xhj=RF|@Bx+wC)~&Fi28pVPW1Qbu{ZbXa0_Fr&O)_Y?p-
zTY<^Spa!58r6gusTBXjuhjTaPfSExK4~6&oQix1B@an?009fE3z&@L1hj9SDpI_OK
zMs*SZ7u~-v$BM-c{T<8HooqmmbZ~QdP?gZ|(^Rk0bVtR0H3c1=P?{5Q%N~t>04)28
zr^w?S72nxDr|BxJ;+JCguqed<*K`_1pG}Bik`qU9qAA6Clf+Y`I`bf)+MYed)}(qw
zgDKtZ?xrE-0tD<d)5&0(TB9(A{9D&LF@oLU0)VG8oW)WoorWcp%`#IZ?*~o?`uN{h
zpnCq4_L<~#1E|LSd{E-{+{3xO?A<l93_mk55xMjU?Y9@QBX6=3>gi6$>i_|EDDm~a
z^xHx$S+&`&m(Z6FgEkVGteq7gL&KZtYQ2)`-p<{0!TcFiGqPC2YCtI(3b{rsNt~ki
z)>s*-wShMQ;+K_G4FfdBlKAK>$HCQZL4p#s!#(5NazQbQ0{UP#i}dg-W23QPu<WDr
zAr85VptWxRX<fhAtc$31Z4;Y4#+3yZO*bC@J3w7l9&P^?n)-E6YBW~_OwsrXe#m#2
zU}=0jHKJINC3ZP8RDYcv9avZ{_x^;O>KHs%6WfsVs<4dW#)5*&ohen<nVh7HyKl;@
ziP5grcJ-LJ)?RNhL|disf{h8W6LR!-I}~3A<HQ`$ECLba3n}wc3T4rA6B^HiopqnJ
zkS$Qe3u~4xqyc=2s=0?K=WD2zb!>@vIxl*}n^JFXpO*L_rAv@m6;atra|jSMYeebZ
zH@_2rd0_{hi0{{6S3%W%d%8LT>S;hseoZe<V^Gp?;y~Uu8PbMpz1z=fbkMHX`9+T}
z*{5mgVdqx<r0S*>Dpztfk)$%$a|DF-w~XN<Vpf=)A=wagE=dsK|5s>lALS<}1e!K&
z;zOHbyf;|&dw51Zqv-od!epihO1NdKkmu#xQfjDQOlUSYeHA^mDg6gi`WsUe^P1be
z<mgnIu}}J4cZCA8Hb~yM@zJ@6&9x4~`PY{|VmHCEMWVF>KEMJ$!1NZmi0^R6KH^g9
zH0R26`+~wnUtNIa{gp!tzm+MJbrWVnsS+Uk#FYM(3F2m4AsAZ6O-v?2^ZCi(x_jQI
zsCP{Ch~-9>vE`gOW1G`%z8!gAdI>W`>F^1xiU#465PG&{n(6N)Gs|+f7n9ONx|hfP
zVCQ3Q`{*|lyIyIde(5`hT(n9%-is^V#PqWDtn@j)0ToSKA_qOv^wt{;8%$Ksnkf!_
zYQ<}sR`6}RrEm31zv`3b@0OGLh)dkS<!+1g$|;8(<S@9$cQwJ_ib;4o)Y$Gs_B;?=
zE(rNkICeVNRk|dz$B?@(s0t>dy`bINtyJd^TDag-Wq32QfNRec8=*7Fasb%Q4KpCn
zNbqf#tO=1$3Q0(+mlqh+I-DX&3sLk*L-y<?P7PIJ-y8(X`6kH;Qe!*Em4j&5szJDi
zrgjrFIQ3L~<bd;iEN8ybf`GkFz{4y?FdtW~M9LryiRUivn(-$@5tFS%itFQjv8|?p
z$&vdruY08x+Da)=Tz3=tac9^d$ZAtVr}grW-14FtGq3OGjm>emXAVef4O%3G(M`Zn
z_Jf-z4h$;xMO_wzdFJB`vjVeJqilrm8Boql0_UM`sQE0v>S_Vd0C#Ma)<G`(l|H<Y
z*5<mxur2#iO)sA^xfgp$rnmy9WsaO9*EVjJ(|vwG`cA(I5_wYdq{w+a|4C#1)UYO^
z`E%%tT^lCvI|zw7nCSY%uqoy>@cHf5)z;Eb#%`fVD`R;dsEbZGdIV&9bK#AT^o>)Z
z18o*i&*4kKRkFW#w50grF}KpPyO^-<o^$q4pAfylsRs|JA;E#`RZDF9(#ZF35j};t
znWXs)`;vs$g)Q;m%5EMP^A3&yyA~*(+|l&>(=Hm_Z^*Q3MzKewCQOXuBzsHfE2X;$
zJ%-urb<cj{v~_7$$j|s5RavN{)2UqN;X=Oi{bcrQDYiS#bsF~Xz7jBwh!q6r)z95U
z(}<J(k7w-`qbRhF9c-Bq@<^j}jG=Z;nFucW^hzP8(eQdN(-`_NMjZDp1g?3JLf`s>
zL7~QT+j`)Zc9jHK-a?$jXL|8^_$gJIMKEP7IwOZ1J|a8r>5nMj{`|ZYFy77RYA?yZ
zVT|JcV?f<^ALJ9Eh1(NIa?zKpYg31Op;hn4uve2ooNc=AOJkb&q4hAXB&k58iRDa)
zV70f4bOtJ(IabLEF2X6!@XDv6!Hx$@Es_z&W(jJcvy3q>c$xWCn5;%eQn?#q7J^3(
zHy38ok+2ho(#qf)f72EVb2l2rr)Tmd+Uj<FY`%VbCIS1~s?`r)DTUix?*b;kWJ@6f
zaSUcRJgJK=`c;y~NF)b=#qA8Z^9aN=Bkw{r!ZSwj4(CHI=rP96aOe9o?Ks#;R;lir
z_Vm_=o>%WOqjKE36riK0T^BfpirOWfmr@j%p=1t)^%k|o0fAAR585o=d@Cr4!QBn2
z)U|o8A~ODV=qKtdR(B0^A>4~}1Nw)E@!e_`o8LBG3u!|Ny3MV9&HU3HkC`$B3f{@u
zrYAJ&#S=Z)OH#$_$3uOk(|nm%>k>oqDL{98ssPv2ELLDBvLjAnpfR;!<iczr(7he|
zgY5GtDmvSd{G-Nh)}N>fGaFiJ`xsqo^IGkE8~3#^Di(v^K_ej%{}w-W-gu-d%XE4j
z>&8JumWu-$zi43;yQ}s$hxEna_(e!>;QP8m4Zt3u)?^Zxc$N>9FcaGX+=BLhMEutL
z0-+qJh<HBlyk{Zi$QrQ8dSM<wPclUK9z<5y60E8C|60|9h|uu<#Awm#N(}Cj{Fcnz
zmy;EKkIiK>wYQ!?%;!i-`(Z`BbUQLyqMc&ttdkf6M=scHzL$_b(aW5^x_^yaTD%@-
zm91FYV3dL|5JZP&Rb5wl5AHqZ9F>rhnKc_``bp>Gbhk&FT4qzv-=7>)p?>GlP|$3C
zYpBP<Vv%6A|5-2}Pkoz&%kF(<MLDjxH?eYPJ+~e5E3?5JW}M6HwVYQ3*M;`V5m4!W
ziR9azj5c~n=e^<$Av7e~GSl2PFP&v4{tZf|Rs~b%!WLHLuy^}M30hlOs$wBtF@hJM
zs=BBtR=!%MJ<Qwtj&fwOYk~U6Q+>}O%&pLb-rKh@ny9j3-zLYV|5`<RyOvZ$SYx?N
zX^vTlTx2Oen};u>WVpbK=3E8T4VCT6pDyYi9lx4OHm`7BkfcX{JLkkC=FEWa5+3ZS
zK3iHge%*lJ?!L}Hms=7q{vEH$n3;P?<g*j~VzOg6Q0Ku~;f3b~a%XWXL*&D@Y9V$-
zS>gKAJq+Hj2lt3e^M5fNI726bAJI6^mF6H+)#Fzi{zeS<(4Cze!Wg3bo28%A?siGD
z&2($ha*&~_&v+$nf=9Qz)JfL^Xm}%N(SsNxq3E-Nldh!?2W;^i=biT>uKSDd9B&wG
z;Dr7H-Q^KW#Q(LMiyM5Qr)^CN4KzoEqx-;4tLsQNn%?6X(KDvKCqR53rd>4)j^%8D
zn<GJcs!w^Hd{}yx%ZB^slHw37MQ}p;CAX4vuA<MlMpGr&Z<0d-iuRmDq@g^-XiC~`
zmBo&kMHWi!+#qoBF+=!_xv>8UFkh$Zt!T)%OkTJuBHp4mIdhdpauAQn%pYvMmlLc6
z@%J2v!xxSYT|wjyrO)~kN;`GvuC$)q?G61H`bmT-+AhZ?Q3+G1i{vNnGCh5mrJ^`Q
zwP*c2V~vkltu$_{F={AoAyL3Fr{}i;NEl;PgBH+tt~&V|4QCri*CmRMzf~r2#Txv?
z7CZCrX*1sQnCW?1_j47&NT1#lpfST>>{1kZp$Zc&FR)40bt&jI;T-hw0l+CSUUw#L
zcTA+mo0CBxFaBl9Vx&D|QwIuzI?&orR#p6;^JzNT@n6bLY1@iO+5Q5<sm0-8_8Y!^
z#)+2NWN<!CNDZfN-{)^xEUjf9Ly^~IqHizV5{kT%)7ey%^xi3%h-Xp8=n^-}GXB8z
zexBXVE}7?znz^$r1kDLD^Ij0$dJr+!4NmelodEl6Pen?5NJaKt`uL>E84maP-@M<S
zEbPqabv?G3e!Y96cTT?UGvSqOORKrf#bzXzFdIcboP?8}Q}k9T_0!z1XOA>(Vl}QL
zmXzAP>Oiy(yDJ7;G-=SI4M(wl=nAw(bY~MR6qY6?u1$0cz9=#3!)3+iKotN>()cw(
zwF=+rF%5=4DIrMEnggYmQ401$15w9_x(!1CU(0p@?}u#D2+r(~O>nF~_bKXh+joS(
zl>#0e;^lW2S%tFe_y+KFOV#(>t?OZ^82J3GuY#yudXq19xdF6%|G@bJWxsqE9&>78
zJYSeX1Pgb!Mu~N_WvfnWn1%*&I)Yq4v&bAMaTy|XZ0Y?HN%E=(B*L-lX{v#j)HNc^
z71f1z@W*l_t!f5Li`h`yJmZad=kL>VzLjQ+cww_Hh3QLkYSog%sh4w|4I@G2&LalF
zw`AEUIozvG%y-yh-dzwPS$Jnm;86$0Y1QEV_AX>RIFM=lAwFEzyqTp-sBpO^9nx^S
zPGu3Ku|AhIc=m)zk8Hn`l=nV<mzAL}MSFmBAFI=0ND)pT)WUf?8(-ucdLv|GrIxXC
zf#x%<5QE<v;KSryg_1q2yVvHz#s^3@aLvwl#H>}2tezA5&^_k!;mmt*bCU$N-9{3b
z3x3>&AE3Xz_9v}z#NJUxl+MQ5O1(1Gb6)bax1{&0O8nUz!i!s}QFXv5InTpOun(qA
z^~k>YoUbF5^KEGRku`jK7&OLAJY9-N9!X1S1yE+2DlnF~=hC<@`R64exvX)O475}+
zN~1X_1<5O@89vq^%m*OF@dTuteHB?sU5q@c)arY(@Yj<)vTDa!?<i#GCVFJMB!Afl
z@KV%#)f<i)!$!?pD=bMap-QJ;PpcEF3{jFqy`gQ2j5_ydcbW^?21$r_mfq5~a}%z2
z7hnElbX=pvoRLlRUM1;bS9}$CsysX+MX;zkgzuUvc}5Kqm01Ti@OukUb=3;5hcsWt
zXJdQIq{WH4a3VhrNtE?>EoC?McFhgB!<~7=s#+;ouAl4=uAFh{%Pu*AA{U=c(>$l8
z^M(1jT=H<z`Pr(M5vmj)`ldLx;GZ>0`9N`c9mD3~{8WF(D7pu096_2k07HBY3hPrn
zf$pRFaA|S8k7Uyxw?nw`UeuDCgk;?b%{<!&n;aP$=O7B0@kD0q)V5ceni_vi2u=pu
z7M~oS^zB?GX5eZM8lguw!Z9AjWRT(-bX5=z&>yXU%Ozn`+KKOR;w{E7k&8>c*d|%e
z8(n`s7!->$Eq%?rNj1~q_P1=}X7$Q<3%h_!yHAqu9?j6yqW~UNvEES;IjZN>8~DTc
zyr%XWC5FR6!jTys;5f@_;cb3Y4$eQrpO7+0Ap|!X`TG*^XYDW^uhYr-!=B!kf?S00
zClR&z1&ojw2|3V5SEgb?=}_g8i!|ZD&l-Lgn0lOSJFfKTy#+eS^VydHlm!)r>bOoy
zh5~XowIbTjXjlg8iA8k#XgFcuw+JXmW-B|Gjz`wWIAhf!g15O4w`wRQWn)c-0~;~E
z*Vd--?-HT?YjZ<OMQuX{4h+rNjEoWYX^SP^n%iqs-O3<;e_Pre$+f1@nbgrtMkFM~
zH~Qu<MZ$*$RcXL8HXL9QJC_$TKg3@b!LdUANV1+Fzz`)@)USCm43M`U=H<kQ5x6Iq
zi{?)T=m|-V%EXJ*Vz$inX)#-ZpL?`RHKieAf<WnGcnuQ{j-Py_c_M;QN25ZxejPqf
zvKboIBasp@euYr>Au9Pu;l495rKi9cjG)HGq(H>Wc|D1uD=u$7(@wQokhv8L=ciTW
zq|U8x`B0$?Xo99Cm#-<c3Gb%UT5`4wxXNRk8=z6ZJL7cUq^qEh&~V{jyIr)zVI25?
zn$GV~uL~qnJ#idWqiC?jA0aB-u&;Yga9ivrG2+hR2@6gvvTcV{g8F&t4=IT`Zf75!
zH)4sugK#7lORLy`8nu&N%ALYV68hR0du7q$L&aI7D7(Ld1eBTdB6>km%{k7NBWg5?
zMCRIH^ovK(7g~xZ)*tNpOnAeSMUGE1|HT3bcm5Hj^Y!YPs_aNncJxI=3Z4+v-8i~!
zgM<L2;`L9BROdZ+$hlnNxWGqS>#mqwI;N4P=jH}OS>6)nFBhjV`bgSUOW3QI(N)lG
zNW;;7I7aTh<$w|~)!x<a2WNmg+0IQP%VH#4Kufoyyq!@bu@p<~MBRjvdDKKUK!{88
z5Z~0qJ~x+#dVyPdx6)8)JuBaBq<D0Yu4H;|<oouIa&i4gQJq{re2f(QLuv<+@Qpme
zZI$lvij9in1xkr@ZP)x6zTLP+pOUEw*Bv3}U2OYTmf=Z5EuwPCJ?M8xk}1L2;G(Sm
z_H)mu`V7`#Po#P2K_A3K!DY^#tT*7p?`}eWpGVikvuj4%&@}Z@u#^OXrG&2=wUi7L
z4gZn0Ro%+V8doRb9!Yy+`Snx%fz%urzCjMX^m2AK3uB6c-$G+4#QFDO7e^HF@Y^bw
zZahpT7Z~T!o<dp*o%)u`aZGhi_kE5|r=i4Cs~xm!-)-@en~GZsa;v~F!7iiuL+cHF
za{FR=yQL@%nPlSAFw%VMg4YqoF*dTM!h<FT>@u9gw&cU!R(h4J<IP-f&d0{dFDrA`
znT;KwZT_~n3v8?TzXqSt9&#B8Nj5GGrQucE$dB9_SsT7pyYDd*4RlRyH-%Wkrd8Tg
zJpL<lD!Imcoy@7=x`*{ZEZK~$wq~d<?DdxpCYupB)12oEs}`9)ao%nqhAwt(V1=(V
zvezh=vj^zSnJ4FGv5fvRigB5}3EY44qkOjNvy6*L;ltQ*DDpw<+YOhTQ2C|LsVAW>
z>OtkO)IVOVmdIi#vsa%Zxr=NzgRjK8h;JHl3jZHrv-451Z_``Rmy{{%j~8T{)CLha
zi6B{3>QQHH#ZtDo+{PYeB6>67oQQ)<?i|pl_C^%6h%o9xPNLM+7gQ*nD`zO;y~b_Y
zKKEj~{jFH_MEXHfyz1O4Beqo9!(rj&9`EU2MvAec%NF3n?AN2|apO<wJbcoJu(vfW
zUk;AZmZ;2t6Nq?N>NZ#GJa=bc#A!6Jo&UVe?j}(Q&X^NA74#~y#G$KtP((;o_;M@F
z>3Yp0ZTF2&+_6zyY)@Gtr2@H@s3Xf-G#e40BD}^r!|V<$)7if*5Y<OB8_8Kx(mR^c
z$5cnIgB2{aNtRl=vUmNwzwfz86)gKYD)C-$h)q%m<|63?W6@SXSqO=|8RBN`wf~zp
zMDFa6b9M+hOO~~os2s+I!ji_iKTNu>eCqn)HQvR&*;ragQm#5l_H)b*Sh%6iAWzGi
zhr!(vi<3tp#MZ(M(^hiuZq=mv5tGlf$K@Hmtn@lLxF?fUJ*+zriGibti2E3bj1N*t
z-Or;~o09RACJ1BOxMHcfi6yw)C7ouZi{;qQLQE3qPJx?TdD2QpmC1K{V%zs?En9h1
zf}+@xb!ODZ6I<WQIX&zjiqb(M=<b<5{@4yVRHF4a!%U<zM$#^O-Rfb5p@R7z{K<`l
zJ8`;kJ)j%@PFWc1Xw2T_$k$b=e6i7^f%taBbs-`2q9<p|*kap)1l}pgrzc|W_5ISD
z$M#-t#<#l|;q<S~>7STt+x+;nN6OaT_f4!FcXB(0YtH@+Q1VD&hi<3#Xy^L|AE>5o
zGZRO1Rl0oj$&8lvcoJC~iU3uIzbnmn5sgv>YNe@2zpR$K7%}gDX4BB!%0{kIo*ISt
zHu+4(nv8zYGdb$ZXliaF{xuqL6qh@_vxpXnDu`Oj-LM;E+{bsOem8&?!nfK_%(3-S
z={?+&Syt2LVm3lDCsMQ6nJL(KdrG(7wxkbD-M4=jkDzUk$NDimb6L6LdNVCAT40pa
zm4)m5DcUyX2)bwSS)2Hf9o}^STfEgUkZCLBPUEqiA?kCEQa5pig1AL`HgND~Vs$m=
z3Xxw8<v*H7{su0coakhS6{e;+Ma+-4&@lvM9H1R_xAEq*4wF(mcJpkapUGpkf14}%
zju-oaRneOxsH|G^n`9<)rvOe5PS(?n7Qa#jgOB>uIP=r(%iq-uGO@GXRK55l?N8bL
zp@w3ixlY00H%^{RfoEMKfhv|1R**i{hZJ-;CXr&2-wNTH5jo4Zu}kOTsUn&V+vo>7
ztoRy9CVTf2D>9;7?>P6+I`7HmI)S8UES-`;>(j6amC}-}0+ge_?^0faaWm%tZ?BBP
zFq~I;E=<6X%E5_~C!-_;vL@yc5^Cb!kL(dDYCj2;D<1o0xIV&5*ue<q&-6n5Km>vw
z16IdM!yU(k^)T~pc2jMcbPAFuw0}^pw1aB#_wt?*wJ_d3N0nZWz1NycUqg0xeOB|N
zFu~~Y-$g?9aRUF}j+Gc9vTXK#A&_8Sm@OE;)h1gT@%^gQhkY?0^ZRwkYm$l|`S84T
z7BS#qlhF17hgkx|N{d<wxSB*-iFOHFmT9k8eS&!;blx++mca~U?oRw7de?Yn^(?N8
zXU(Yd^^?Zzv^ONQE#l*>6og|<!Sh|0Fy8pvIKfEEZSHR4nGo4Tj=0=s`obKmX1~yf
zM*gs2A!147yA?GKPePXR*!vk-yv-K}*iY7lmbG-OJ61KUabSh6cb?D(p55uOuaHc+
z@b9-M5NB%Qv`-y!RVWdDWJ-ibTT4&zofYpbMS-?#3o&_&IV+O{$sYE(49WW8z(!RB
zPGVP1n9ehGSvp3eG}hiHJ6REVZ83Uv9|Vu<^SqoC^~Fvrj;q<V;+dxK=$%LOLT*}=
zS#9Nt>pUmae$vJ;p9fq;62Qn9*oo4=u`6N8Syd!$OLwa7c<xYai+2yj6qTwZ5ThkC
zN9XLxK6sMcja@xa^pB!OtV#M{87>9ZHQFo=(pdBblLb?(qTHovMrfldc!Ko`vJTEP
z7A{7lwedQUMYKHSwB^=OK{a<STq?GnY%&KKvP65;CaU<GEZAD0_pMgtZx<L*s{2kA
zWm&%a(81bmS%%*dgxy(C(&(`yPg&usp=o}SQtD<WiR$FSRgH)Z{b#t}6LBkuG{ns<
zpao}<%|d9&8q`dW(O@?MI0SfFwZy+U+P}Y~$zo;IL1b8h-ppp4S3Ht%EMYE*mvf*!
zUIs?8J{C8Ky${PhMRZ9(H8>oMvkxj))2hjb`%?bk7z+onV|{jDO;eGa6WHiCk!Gi-
zWpyIgc@XJn6YG3Nm$0uaXVRfv)$C`!*`LG!)vs<u<@OSPM}f$0!*_{>x*ccQWj$g)
z{Sk)WR9wIO>h<*uO8^YFbl9pwjaP^{$Pq$Hi_66L(d1e3n1!I8`Z@<xbyacn$(-Mm
z%~<|04kK-Iv~Mly5&S=ly#-X1YZvz|B_K$MbPqUmBOo0^2m=TRinM@$G9W13AuU})
zD=E_AAPv%>v~)>GiwvE9*La?Dp7Xx%_r_YRrHh!k=f1Dnd;kBxoz^rG3^(^57K6xQ
zVX=mV$M4?oBuAxMnPXEohE_JwY;5o>wiTT|^+=oqQf`w_j&ezwkI0-5hHhLuijqdf
zv^?(g#GeNayS_<c_Pj2BLKiN+;~??5ee1z&n1923uraxIrvHQMr*72&nb=j4U9{c)
z9rwVbJJvnZ>0)C!O%!X0co1XJqY4u6@4WMz!!7@aUI+sePCqxawNpxZ06{ue#H24A
z(Fas1V@}56h$&Gss|RofAKr#7yG=0;mYX|;^VM<i0+A*zt+WxhC(@;S9$x%m?H{hY
zhCMz8Wv?An_*<X6V}^u4Uf<2-I8yC%_|~&clcNb;hGt9Z)?}smd|q>oR?F)?hKO22
zb=jbdk)Cuj!;A%`ribRZ(dKiGQdy^;$q=!(=XN+Mt`J16?FAVMRZPTRN?5Khwm;jQ
zlx*G{B|W_sD*-BY%Vj4a_xusj&OFI(nbmGyZi7zBnUd;~6>b4|sV|A#hg!!7e$FaV
z)`Cdv#|R!5_D*z^MuMBN)eQk~I?wc!r8`+m9@9Evi75dv51DPw30~HNr_V;TbWU8h
zSfX5h(L^w%GX`|@lC{L`XklFhraXLUN=sf5j<&7yz26Fs%QuBebTJ3!+_;#>O{W(o
zDOSPPaRged^F)UIuT#|W8nI8q=8Hg;+_NmYiR~xYs=`qc<XBD(IvLr$tIRW8HE(5}
zr1r99&nbWv)g1X-c2$30Unsh`d2?(ChMU(s`|h`jRFKV#Z(GjAhdRj*^DjGIP~ZRw
zo>r7|l<|ME$8cxd=L!P`g7&qa$LKt8gsu`tHO10lv?x1ai<1J(EcunQv_@<dnaaN3
zpqY<Q9@a*<QQNVv&>_ae<}~iRX(3qPm$WNRxpp!mj%lhDm$}OvELZlS>Gq2-q9|KO
zuj%P3(BM%pS1HQkb?dc|c}&mh2iZTGJMO&KMp}LS{oUu)4j^*)AjbPpX{;ounZD+M
zJMHa26k*+dAl=`f;V=3*7|%)`g*3lqCo7oR4D@Q=Jl2cWSK>)&G&xl(FpZCAAX0&M
zvTn5Cv`Hk;CQ`4rshLr+>{__Xod0T*F*~sleCxNaVb%-qP%`snP`{WLy<4S@KJQ8U
ze|2`?`pU+-UG1PLze_spMuihHgE{{--AE-&ZE(Z>W#eTDEq%zUMa!D9U`9m)+a>n%
zE~mGv*t(D$-x!VAH~FRE+!0KXAieG;=^Ed2Q+1Y&=bKQKqQ8a}_XgYa6_CCogXYBY
z4c})NjF%;@OT?;5Dj(-UNVjpTEtl~|*Bu1yjF#rE!5XeF)F$wK6#$_jImO*IxSG8C
zrM$q3lAcSo9nzn(U0uI7)EO4D?BXIz+UDGk%j{fLhPtZ6e`lw<gW5Bp2e;dz164ri
zg9nYB8k#jv&8`S!k7@Adnf6M;i}KFPZr0i+IjW;2OSf%Pn|`=?DePTr{63BJCF4W3
z2XJm@Q~UBb+Lz4Mn(42#csCjfmW$QrMx=U<i9P%g#Sv&FBdyGwpC4Yk2#tPoYab8(
z$pHk_?DkMnyPdt$aCCsxCqVG?$o5~|1I;-w*+mGmhw<W{dPqQUi#rtCfE8&5D#mjH
zmW!0a<~4Gt%)CaqhtqF=;6(bYO=FrZcboCc6b?KBG`FU!kvHvv>hlA<qAcvnBN2C)
z*?nDQ%)MhC2vM~N_CIML4qzH!uFlQG_DKaTh-bq6qt<EK%$qGNm`<Kl#W!(p->Z3k
zaWGylLm&XPM@U%r-ZZn9_bkxL+>l82(B<z$5*SUqi@12d_@QIIhld3f@fefwnqdkS
zZyh3b3G>I~%YAr_p#mMEK@>~$omn5)h$V+I!u<6&I8Qvr>*rS5!yKymUUc8Gp4zPC
zcK&vs(b`zv#A=8JB}u@1KBT%z=aN6JIQwTi_ILXxe_~$9NB`pHQr-Y@59P%)pZwJ~
zzq1LFX0s3*Os@E01}BwjhBIVu?4)|b{52C*M-FnYC2KC@Y=L6QpV|^<LXblFw_CK|
z48Kgk=4}VhT|Jw?s?h(}sxbN3)8qZE#xM2e8aCuFuC?7~;?iWmzlALpV10R(@nGmd
zheYvZnUN|8*V9Lc8h#vZP%Mwg+5Rpm>;K{>)z2Xde2U60w!qHucM#knp(4*HuZKUn
zZ1NRGC4y+Sm{ROAJ^oCc1`XHbQHF=8>DkP~hK1tH;CUe^hG%bQs_`Q2ZWY);TCIx2
zN=<i<ZWG!I$ETwBgyMclo0^Wn$`F8Q2h~lUPkGZ-&HdGiVDFRJ@6+*gwO3H=x1d+O
zGBoBkuc9EWl;&0TgzsNqyuyfAn*)*XwSn#y8%VN8R(FgVa{*;KQ%uw(3Zzie2%cPO
zU1m%67wD&-Xp}be<v0j~ymZ`Ea({T%hezWuCc(I?a252}Z7r)S!nDVXDd(?2J6z!6
zYwoO#Vu9k+#*xNxXm+|&aCM2Yh{hQ|+I@zegEj1*ok+B(Q(!@dbM2llTO)hpoMW+O
zs?+8%X0w=-)#cmWqTSu9mQ#vtFcneS=ZucDyqdUt+Bj%`<9krZIuiBM)+>eCTQ7@N
zGsrZC9vRyuA>bGXh2!24aP!a+j4L>&KT(VA10v-fnBu`FPLU@3rVmfPI%kkbZ3!lP
zABXMlHG~CQLei%|x-T`RcRoR34&;RgR{K6|f7dzI$@k5(vfU&fksq<ne%4Lltiaff
zZ>0+(&CO8Z=^n}3B-GqGcC&gePQD4p)gR$jrJ!xIPC0!W&M}gg#PLe92%+;@QZZAd
zE7o^d19>A>+(Yxl$%1j`FO<YN>&1JJ%PG!n&+QQ&X-_RoKWAqQu#Yg_Tu-!^hhd9o
zw&yRORXQ*-UPlh%z8PN<ZOCoXZTj#bt2$t`i*)MRFO?>M3?TIIw0j1sC-o>AQM--t
zL6=18y1SjNV<IjGl;lIc-Ama+uCqW;SJ3k)hS`S)9bZTew;PEF*w-x!9{mAD_R48w
zDL?Nsdzv|pXw!vRmY(TJZZN(7p_HIkK&X+zvw^dnjG_|t$os|4`V3MB6pfk<b^H`U
z^BDabrou55KO1cPQCVVv^wf3DbXbF4O)A16UmG^BL9sR0tfXqk8g?`tkzJI#lQw#h
z7um$<!+eU|7qXKBmc}U8Mh)grR9KUO5TLegR9h^+(~^xyZQuD%Auo!SR!%r;hhwlw
zh$=<3g&bifw<Es$1t=GneE6rfzKTtkh^lLrc>0;(CZf)OP(6B@m1g8-R||r41K$U$
zQxcMQ{7H0n96~V?sFF}9QxlW78N@)#BZwOL{3fgX&2QSHVKew)%(=$IwQ9B2YMMbB
zui>kfQY^k(l2cAC%NAV-u9KtT+AwihEjHhZ{Op!bua9W`M8tB+xo*Hc^S9OJ)0zZ#
z!(eL&(d~~^#mDa9?y%v#%D1Td4qi3KIO(yj&wuSCuY5fAz7&@jI%2(;nwmsFX7CgA
z4GcS6CoJ%p@daI(O79i=hK*WvQ~9KuE(ylr!XwRQla<U|s&Njvwm2as9%fbJM;oI$
ziz;I<hlBHjQv<o9_ucvwE{UGOUx}vKLFdOiKF;<J)~I^v>I=dI9F*3QQ;H87!a;EI
zK<=mos~+mi?rc2p0M_rE85+HPlZJE)lrKXvpSt3K26cU}UGt3O%`9u!d^&c$C{~7E
zUK9mdQ=Hfq`;OYRrFE-{8`3N@y?{CsENqXn?Bv^g><b-QRVYlv+j_iDSP@A*y$C&3
zv7RsSEG=$Y@&Z9D1)`G5C_Uqed%9^gA+^xmZ5|)$t)yTJo|wcZ`^<aysZHH2>P8PH
zaFaVo@Bg!4t``Qeq9lLF_XakEO>-4y41ds$ozp~*IR1`0+F1LtKP#a{@<##Rd&imf
zFX@LMA(?qjSmkk5AO9W}$MUtPBC-j=RK?4nEoB>{=_@hvxHxx~1-pG75zeI<Ke>V<
z6yD?1{i;0$dspd(M*E$pN{|}Se>RBoCQ^aS>KYV!<AjXu+ImL?aisvQ69sv!giz5g
zzEF6ypo-XKYDpK39+hB=P^?9K^8*%5_*F8SYYkm6T$P``*f@HUzpb&@SlTl%tupRD
z&(nP0MqHI=kX`4*I7}2W^q6IUh9p(cW-)u7<~=9=Rayqt0`_4-n7kTiyBRUf7gjzC
zZ_BDoG~0DD!&o8$0pcqZ-*5f)jVJWztfF^p7#fMc`mr0ELI6#*u%nl39dOr2Eoi|L
zNYo|Qv5}9fa36EqwMex1V0+{kv)sw^Z>79J;-o+ZlkM5tA|{Mf*icnx&J!lS(m`1A
zQjU=e2P%g4&^}6&RJ>QWKdh07OGJ3p=L(9<d}Y1WT8{SV%DbVJAV(3bJib>Zg?ZIC
zH_1YD!vYfBuAN8>-v-7^)dZKDICLr+1OcNQ1PW^q#Jd$3{qq3Ms5rUxvc%9jmzSbe
zfx>~KM6F0d@ouq9sJ@<T=c&YtA76BFxH`pNnu1otWLU`iL1ZMYJg@GnZm#iIT`>uE
zXuFb*U8g$txPhrVy^x}Kr153qs`2>Ax?3SzT})4I)xDbQeO-6P>g2MMC0F&)Zz${L
zQhNMSP}gKqK4a8w2_Ea>!nd@=*}`6GcI%(2B(c?n_~JS%P2<-x-BActJe8MIc%0+c
z<xzM~fBtsq*}8Zy4$%Cp7sm4sf6e1=H|z~-(uw2DM&IT=DVhrT_EptWF8a?deUm)Z
z9wtv^Z!YCeQ&`F!`8@IR$$2Wrb&RBau1V2hg#T*(&VHR*c9PIfz=E6J92)cKr0cmH
zsyp`M*uP`PW9N5b86Q(g$xeIv)eLOyD)4jar3IfUjn1t0R!0IGUb>a(D4*CAZ*u@{
zSG328#Tq#dt-RyII@|$~pV|;cE6DQqzb8YXow|eG9038-Aocq57`#&6jTWsyzQ>|U
zj6mvCw2yb4mtoDSM%~WF`+859P`B-BxFd<=kZ}qDfd?gTW#n!0;y(+A6NlwJs_Q~3
zdB*SPM=iG~Mn!rVrJyTwJon$Nf>hswI?q|va`PDLp(K>X&cUZmqOBB6haf3gDJrX2
zUxEmBnrV)u8(ZAU5c7X67$DgCvtX#QGX7FAiPvESKlRDQ0=$`B?$c<s&9NBh)FOzM
zcJfV1>wZwmV0k2*M2d=wM6D=5ul_kfkvDLk7SpJHn|Ik<t02SDyL=BOz5@dXAvWC?
z<$uY&2!s8*U;K<P#-*M&4MI?@=y%nWY;&Xwi`5(X(PmuuXiA(J+!lVG)Wm-y$8ops
z@ZEVh@u+9?HsdFiG<3kc@v5(^?c3>^j2^S4hLM`SERFRw7v2J{j$2CxM!t0&Y(Zw8
zrTd&q^_HX|5bD#zU}5MTrbcvLYn5Yw`bnkcwVg^l!EwSwJ#~8PLB3xMb}UTGfrv}d
zaG6**ixL+}wD9xOL&ZjP+^{m;F~8k{5BcyTqh^1-UJb+AZu4y;=OV8;C!_xT%pqBl
zM{;VUG`Eo2C%jV0D?Xy=T!qI*h7h8=$$QSqwaZGCWhwa&@{$Y;<GUU+oE(GVFQ{`{
zD@iJ=9Gr3u$7Qo+n-B4gzRS(MMX~%!GWxQBH$-N6h0ly5K8vRydvm?k+bF>=TH^i2
z`<s5WuHAmfFNKarJSnrw+&z}PmNk|&)p?@3zQfa&`Dc2MHW=`~sv?akodmqDo=f3n
zm2*GHT--X6tWuVD%YDB_Jy@8N^+DfL$T4$K6Meg=iT?VmHl3~<C9U<Dv{juEZJzga
zJ=u6T@l+R^fc{?Gb6_{ch_pO#t2ucRs1MKc%T`@bvfFgKSHw5%rPR$diDrc3uADuU
z>WzQJrx@uY?D3Ybes=sLuy2WuK>D^o2)Vs^QWZ)Q*~9j+X3zSFR%GbMhwJaCxRPYz
zG3*Q^v_xaaDNZaIp?UjGrX7A5`9L`P+BDTXCdNI?I1gIra5sjvoUaV;iTdtX7jpU-
zQlE?gi$wvPD5zma`!deSgLW&mNy|yU*0j*~9?8P}kQcIi`#7+seQ8&Ek=6NkUb1Ur
zFI+Y_?Uc)kEFm~!nsv=)%4Hj4I$PL}tX8MnT<QJZ_|+qDKwf_Qt125_jvB0`-@w#k
zB@1<xFUQ5_1#mgKUR0~~Oakfo`~kASv8v-!cgAFi;fFdE&8c^N)<!tuI!v15G29Qw
zzW${WFc4=p!Lw*dcd214{c!vFpGjnN>Tcqrn)hdWb(*4k7=3fb)Hu!1Wc2YeoP8bM
zeu+{I$(NqGk<T3DaD4`Bt)l<kT0<?KEr6s1x9jfQG0b1z<X@^LVD-F)D@l$O3UP`H
z6H=0&_N&8kXgnCBS72iN0P6_-kB{e{U<Dk_Xs~g?Y*qXmvT)^&eTkw;{E&_89xcaO
zx1QcOZ-qN-Zr_{S=u#o<AuzX`eE<>?xdiGIjDQK9nqp`437(ZZ{j)MmjVV`n`{qL4
zAu9b9vpIkHr&sR{q1u*NAr3HEs`tE&`@>pS8OQ_pZM`PpOojFFkd>sqzKou>$HzcC
zkkMoL_}P%ft}Jd5NN3E>XEVa&n*_99Bc;o%(f8JZuY_fj${w2w+2sOA&V@e>+du@e
zf#Vd?twoq8&628sooQm>N`b|yYKtEY&&rhSrSh;D8ek1&i0^s--~lI4cW#d9P%XdZ
zSxNd}IW@?Pc9=~f`0ELGwUfHzI5RA-0uR{-#`!#w?!Fod%ZI$cDuPFgTRS>djFfJ5
zP1FEl-eS{3fKGt+=B}UW;Puu0IZ+|nHh$!$N0E4HXMGp$4U|<$x)(~&eYA1-x;U@)
zSPML2@4p@~1pDpY+~r}74O-3k%Ts8~AksCvyLv<yO6t${lT^v$CdF+|@(M!UNV}r9
z<q=NzFHCBpNpQ$3%KA$pOd7;x$q66}$GkOaE`NLG1X(fP3g#V~LVC__ny$VN|Cq0{
z{RtBIpCs?U|LNqxb`P>1miisa5f73>`P+0yeGLo=j?%FT=0+B;D21iPR~m)(NP>Y|
zXqs=@Jm|U#9#pzRDJ#GnCyvx{ucJj|ibTw7ZZB)~gj=h4k7QXKN`ZmG9R+5@1bGXF
z69tgQ_XDeBX7lzcSzl^H$KDJu8BK>!<0>L1H?pX9ujV4YFueC>m(JampWU2%^sF?=
z1ju68q{qndDO!Q`(mtdymu46yVmXI#jToYFGqT#!=@ODLHs=DqJ0mM7wq0lD7!Pj&
z<}gPL!vl`-H>)?AEBwT=Y2a4aQo|~jCA(=DTeuzA*ifx(Z0E6XA<)CVdk^JK10Nr*
zOl{sTT@>=bTdo)IS=aagg!{RgT9E_dCtyU$Yq2WGi$1bW(icX&m`QeMAk<D5<1+%7
ziw~H%*G<z_c$<q~`jX?zNTpuwz43@Q;gS*KPtn(C$#z1OZ{XpMXz<sv=l;xT`!%db
z?6Okqz^F_$NDA^R-UZuNjHH-yL_XqKxh@3Y*ZH<iLl1~bCVO&eHpceqPB;q+HyqW{
z$;5himd&NT3;nmud=9=%N^X`cF&*{G{XQ3))E-c%))Wq>R^D(u&5X7dM$VOc<vFmV
zj1KXkCO#kBkD$mH@&Ll`r>IW4XA-p(pC#7S>urD8gRLBPQ50!C>n*B}$(Fk8#q~Ca
zv-!|XjR0=_Oykk&Ql89s^3hyDmfM`iiP29=nma~7p`NMM=TS+j+PoTo&-Ck%lD5<6
zr|tghsT&60RZV=lIR4uA6G>0iFr6fia@XjA)k<c>V0RMO+*VJsFIl&JO;y<^($?M=
zeA-|Dwb6OhKJV#uh~aR}dUx4%>o<hF;WcGV+!{tn&j1!=9$WQvW8cZJB7R5YMRpUG
z`%hqEp6h(BRgWZfIkRcke_-=tgkqI~8Av$a28)4e#(z%^mVy3Pw+bj73$Q|a$Q-96
z+pTPms%`NKGJ!{~;M=@qJnDGuIm;i$Q&c-ND7UZcc^<be`F~&FpQOzeSS#DprTLKE
zLI~!hBM7rdpNRRi3;F9UAbqv1PxcL?v4N!0?%3WKS$67&35#luBExtKdU!`rwIK8h
z&3{05=jn~Fz)$JJCIE>`CD4JsO|g6Yb)-cVmRyB7M2fclSjBh^HH#;Ke`-4Us%G&t
z2Cf8ILK579e$M8rTYGz%1?|>MaLkIqaQI-IdppZ)gUbab;Gq=ot-<q{wELc!?$J>w
zw1*j)-h)q%r#BR)s)a*84~Ok?o|rdy0@|(Dqn?<C)=7pJUo0@G5#HP6_K;I#gDo;F
z{Dl4Zs8uOXWhA*8x|kj}4?L3NdjYS5MSo>=N0=HQ$k#%lFRvTlu3ek|Cia<$)Q&=E
zu<t%hA>n}t#cl^o*ETZHJ1{Oxi=Q;@*Wo87k7e?d+X%dk6OPR8$OimqPEyUVX}+47
z#a-yk4U&@>1OUGm+?_bjk6wigA=M;InJ^7T<H{x}H%3XxX@J9%|3$50vfjn#f&jSr
z*^Qp=O(@N4mDnS>1n~}3kyKKov_P6#NkM7h#R)hiRS5{hLmhn!wjXM?8&OpD3%TzP
zsT$Q0o3KAd&JS?Z5!MzPsbbo@PJ_>7@>5w_>?LfNbnPzU_A`u4J`*(C&Jy1URbwqn
z&@~+t0pgxn@4V_!nTQE<jCtJd?AbW}nu!$y9JcTx^Q?Bjryn)%>s4NVvh#)wcHDmF
zwOPToBEKyT8Vi}HNroljS6lgF!@8Iaa@%t{;O8yUZn*<AP-2Tr4c-rp+_s}w0nixH
zSD4xNN8Cf80{(Z)xH~dz09xL~Ki(~f;6IH<G04FEHi7iM3PY?qFeJawj1)A!EMnfb
zUVhcDCA|yt3%R^@fxAJoE$O0y`;>z9>-2#EOj03wbN@BkN&k+c;`}bil?EMcqB<aW
zXVUFpuvkB8H^&ew!o$`tl;g}zem9<cbv!&&$a?eI*T9fw6D@DUrqN9na4962d?>*T
zvlaJvE#@Kd#q4me?x9$Ixe9y9=U2rpiOw?sLS#Z!H1O!&ijT|79BBnFbz5p$QkMEr
zm#KJQ(41LDHY5#?9#-?O<;`CjdKMC9A{Qf;+Esx8nW8z0E%cx1b&sv(N=oY}W_K69
zY9Y#D4(nRzzjEvIva&xG2aW;M>R$=ee|9cGT5@xEJ4|Rf+!vc70j`^j2mSg{zcgWu
zg+r)^kQ`eVpW*=-338eHhG_31-7=lkM@OGcBU!ubIZO0Q(DzW-gJe&Gmns<dvz7HV
zwAMtLt;Tr8Ey2n%zGaLJ_$7pKSz(JH)`cZn=uMB6O&mLN(Oj5%5x<=e>eY&E|1gR(
z#zE`SvWn?#Bq~&+{3By4i7s2PJb?my*s{&sf-O(h!}Mi5<#E+7A?D@XuCfNSdV(8i
z{pIQAHb7xHFJ?9D(&n+vd8|s!uLooG_XRdcdQMg3%4b;)Q!8?;;}JnQ4ezT?y3*+d
z-LASNaIPgfnA{W1Ow+9txB45<92_*+sr!_&T4oCLlAF$NqY063vgHj~BFS`FHE$9|
zU9{^Q09YDZPq6gd9oAq;4G%$9;x}QQghOnTO~nNtM!XV#zoIss;4ar1aB6(`JFO(Q
zJ%-nEN2wpF#(^LsX~9;^lkTo93y9Sx23jNTm|~avbHvx3Iju}Kw&EcoioXIXzQlQE
zhk?+IRWQEmW?Io|yFLdq7oSVa-JQ2?gNETHQDs;}-JNUmRd}<pUktZi&uvr1_At6m
zwBDtf&3gYLrrMTAJHhP{Mk=0>ShpC#L$?iGcI>5zzk}-$*kJ47vD!Qd^SBR0DUS^~
zn7dsu<!^eG+R+}>CgaVlk*i81b>BnijeU!V)}%6+1`CFvBLY1XUDdNi0ZOuies2Vl
z|D+1otM}6nEjLEr&B+!%W6)zP9T5jIwlxpE)94+8Qf98NccI1A9pMCFzBU1PcC3!P
zomiHqC1gm$u=w$#_33qqu@!}M{((V0ljLE2>EW)xB29V&stxnWJg`ko=ZqJR>YTnJ
zyVc72>hbSzq40+9lB}{aN?ykpZP>-W?7eT7h7oyOVDLvSwX0wTf4w)G+dqT<lf1ve
z{U)}|8(-y%ztE)q?D`+=_Wx{Xsk{UgY7uuEeX$b~GrONP#EA1i#YR5!WLfJHRZRh7
zz4%l1TZpqMm3xnc>(%I5Wn(!a<;^!mhpsWTBj;F|mjmRaHxHChpP{X|j-LYwb>{Fw
z&#inoW{_)|uCJ}8ulNnOR2ngOC#(+Rik!T$*Y%L^-8_^Xh*!P*#T{=IJl}xt+CIb+
z>w9@Dd%`i>KtzHoJ(YitH)j9q%jS_dZ`VGJ%SOQHZjT0da1KvJ!L->o4j1y!7GjSp
z+4q?e@g}05sG%nY#MK^DUfwoQZ56f%@K#7vCMhzTeW{t?oFw>0`f6!zY(hfWvh=rS
z&vVOSIfm6)9GBnGd@?J{Dm>m;W&>ELIBAZ|lzmE1tQ&;F`c!ICRUMT^oC!S+0}mQi
z=@XXYoo`(vG9Dy|%AYStzDfA`^sRL9XiPz6NkR-L2%ZOt`?*XB_JY*2d-FSgV;#BJ
z%DkOHkLp5egt=*Gp6E^MvMt(|WS^2iJUlJV9a)q8OmkhTNk89rc_a*;=c!fK5-Gg)
zjcPUAA9E~~?Nwj3G-TQrXsvv(2GwpbqSTn9ht6u7f)8TtA)g#^!LcDTP#|w6JHNWb
zG_X(6ySUm2b$KGU9%j=ij<KoCuTa1H2k`Lk!0UfKg~*#LviLusY<N+5{ZGm?4p)*@
zm>&7BUwosOV+UEdXU%_u0(>{&DyAfz_BJ}|zXY0pR?mMOll~rw@+^W41Pg>@G5+>`
zonyWKL)@5Zf}z>CvNFySf?C!P48xEJcWD0yPV(OZ*Z<}n1hKJD8*O{@7`Vb%V;8gE
z42JVSKfQ}l?7EH_+Drlf33}{80`vNRzebK7Zokxyv=pc{`$!#Ql2WKI!hOeM3-%+D
zKnIa7>GE_WJL>q)AJE*!<cpsUs<i*}=l@Kdf8C<LU;59_tO=|Pe@y?L1IWYDX=66Q
z-AfEViDP~>?OECXdMTJA4Zw9SZGlh*p{!{Z{40ReyJhZIh<J4wAmw20Xj?S!kcBn|
zLwx>UF9I%)o8`H|<#oDVsPRG!lr^RliTLh&3h@8?nGN2=9*{Z8z;poqSk1!9yc215
z{`iO_wV7)5&-)OIV209<;qdDJC*Ico>Dpw-2q5O9{)?{v*PHi$dKK_9!7rE*ld8cn
zO!YtA{-_(+G1BNTnf*Z})Tpio#sqsu-E5t4`wLRUr|;FlxLNa2+b-<S1uW47SPa#-
zU9niZwBUTDru2WgD!3?S5Dck{qA?TTuY1Fa4Uj!ijG#d|003Ql)x-a819pG9&xvCi
z!x-RzDx*tcLRy&JywA}`(CVz#G;i0HMNpl90f21y%SXS0-|<Rp8kU;k(kqw%?sFSb
zx%1~cp-bW91b<Mbn7=cb2S{NA*kKUki(S7U&u3l6Coy^*bqwywE|RxOZS%-oGTgPo
z%kFyQ+;&#)<O}hazqeZNG~M|7I{!87|NS8oFa$TPGIaFJKLf%tu*L6kk9pef_r*3q
zg;xW5N{1{MixVKImV(;RevRn|rXj)@ZRy=$5V0e~EIWy(!C6gpFM-^R<sK~-fz`jU
zNli<&zc0_W0YJC{a82rV-0EZ`04_x8FS^m;!<kM0mi_gSZB@Bboz0)XKWhaTXenS9
zw<vw)Bs@cKzF#Q!alq&2i$j1uXvO%K%FRIR_vJZ9ozliz#;q$~ew&mPyVZ9|%I+8p
zU1-+GEm73|H&gNdyAWZ%nF%r&<~H_=TK~NrqQUaX)>rS44EO`)P%?;uo*aTH(fDQ}
zl!J~ShU`vi(HzK5{4ubty147oa1r8OhF)Rn#+dkjp8==ZWi!1@zxyEusEI+jb|qZQ
zngTh@b`M8(HCU4xnyDAxiQxi`Li2>4w#>IWR4@1YiL&~iPeBau5<`U&$HuULE`Ttm
z8gQ6GQD76GF+C`vUJ`;yO*D=cZ_NIzj2@`hf9W@}J#Y-;2UxfN&4c}C`q+RWJ3*WN
z>7Nmci=FG=qN~AYz2n{`1!X4}Y;^&%e*$n^PFb*_@{1CQ1Qq(??ce_9935nKVx&JH
zLzO)V&iCJrq-L4U0hP!aIG+8sKbHF-YPuFpvk4ji&#orZZSGE(zt7@rcnrovSA#JV
zE~OEq{)%!HWF7J?DR!H-MA9@DuU!^QXNLz^E&tyUWWQN~g7CKmxy?TxiA9heEF%l{
zSMGq!R1^eTp`AwgIenO9nE!Qu2^F3HaM7aaao=9hiYrqR{&C#qP+{|O9s`<)jic{g
z`ZfRzT9kfTT=^ds?0??DybANCJqC+$^Z)q;!pIf?P~I?d0UM0mL8;y3lV2FFO5=~)
z<&}Wkb87GrLskhp{|#KPC_Sw^9Wu@sfG);_M{5<#e6gnbb2k#<vpd?cyb1CvB8wM`
zujm~<we|zI9HvjqHn{yMfFbns5I81>#AaXW9Z8DO!Q_X%fQ!mY7mP)EDF98|23v;^
zx_-Blei}1j&cIW9YIf55VC!_VyauE$rsn{(pmx0E&T7VoRJ`|~BT&&7ieW%FdvQdT
zn)UtDoag4rvdZP7%X17ZC=+l~t`-1FP5CwkNY%KZUtT5NxUPxXMomxgfY&`=dYw6t
zac9H<n41i|-J_<#s$*YP`{SvJ7ih>NgXqQ&Af!XFDgTHFu;GlM6?sC#<bMh7z|_C7
zORlthO?%yJF`pRnkLAEZ;yyR>!o2Np7_;Y?y=C|t8YFkQODYB6Ym8F-vO69F9{|&9
zAZ7+&TKA+886WaDRAtd|0S!wZWk<g+V`#WL9_wN*cED7xE@#v=53-wwfNC_N>udHa
zkouE%pv;?h-mzxAEX-r3UI;IM<w|Q7U1T_sI@|gn<1;ESTt6Ut7UdpI1eUk_|Jdc!
zE+jEzAdi(P07CPg26_KR<KtN_@WAS>A#rOk#Mmca)2t);P^lPj5~_>IYeK^m9j`TX
zd;nxfV@<%rRoAp+V&x9J{eR4bb0__7Q()P9ior$685!W$smQiyZnS~+J}c9F^Pu9_
z^jgx>Ydt^Z&@+@D{{x`wG?YJl4^qY(X_As3xF)Y=NS@@BIM%KtQ+1wrD%7(4u71{y
zjo473M~r%+-X%zrKkrTC3b~Xg@k1wM)1V^?4VTkhNlcPS?FG)k+$_!C{8ZBeP5)%t
zC#O1lhPgv^?YL41osl>1Tw@X4dfC{GK}Wi$<VN<DuP`ufgRzj+JzZreAo9s#?5Wap
zr{oyAj$bHvUnZTQZ`Bt2PPW-=4_;o=KX6F6_+T;UrtjjHu&rGv9I^SI3DoxL3JRxl
z0Mp2-=BzELPEC^o_h;jgsnkYM?o_9Tv=?9xyKVEhL<p3PE3x$EGWoXcACjvwSVZX%
zOk9&asvS7-I4Am_RyIk@l+@~OKlppfnlrGrJUzkK1hc%+chmhYPkfx|ZV(4ELzh4`
z+mxdd1Ib}?J^-Dpj@iX1s>{7*H!}zTZ22>-6;qf2Fjpop%k741OXxg>vHy0%vK$4d
zjh`X&)+r`q;S!kbtuSaF!}qtxpjo`1Eev@75YbY%9A*flWfxWxC4;wI{X87wcW%ky
zQBpmuu0^vrhd~*EL!70N>xMZJ!p#Q41bjJLA0qTHHWBXv`)kYSJJ7~4SN7;kSQDGT
zGNC1NJ1GIf(Y#+X;0Ouq?JGTeDv3E6Fp1N|8FcHD6;PRZ?~WC=P*hQ_oP7})aQ9?+
zCLhs_|I=qpiP`xrg_VF6F)f~r+HDMsX}kINrIu$Og+A*t*>AAHSpR_<!3ZsGy)rvr
zq}F<L%Qx(Ptk&cY6b%cnBF%T%0di)VKmjydI(CmG;d#?W@q16xK7!XBpUzHvF#MmP
zWXI-3%JN^HE<LwWXkg9W{@qvOZVXe~<$1G8Hf<vu_qu@XCzwhN$#u_{%o>zQ7GA#F
zCtxLP7{rf}yZkiHxAD-BMYd^#OWt4(bOIn=yn$sOd%HJ3rHk|>O&;BbWV?1hrsKNQ
z9jZ)K1~QjH@mSZE23{6%fMAQXR509=7QXTwcaq&F5liy51n7(_3m6YgHUM2_Yxpv$
zyi${MUa{+jxJ#yE90ywAB4<v?H?Vz1n<kOq;e+wfZ%EA^*IDnuXyfOtcqBpx*t)+8
zWe>Q<UtwJLW~zQdSV}Sy=x#O>`~yAcg=|8Yf8qf+<ZL=)eNR86S)`Sum<;`NhOvBO
z;B^M0-CqFs>XM3!srHqiRudYlHgxTE6Ri0I<?daM%tXJuIM6m|5rUWpn4TX_B*P@}
z^V)u=BU&w1o6i>^+RNFWSUZ)>^BOl`r={E!a_8SkGgG4+9+P7QdqY1OK5q#moKhTC
zWhOc)oIPHoZnotq`}CO070fa{Fw2JiGt2C-Iy(L{0p2M2qmp*Xm*yOZqC#$c&r4{M
zikyMIFn9U}y00uONj-P}n`t>sTya-9wUK~W!$Tg|`isybB10R2V#~dc>m!9{j}e{X
zQP<BtUXRjzy7c~^E%48=NGZUdc;a7?nt2A>uk7FF9A_I3n$yCI;=?&(!$(u>oFys3
zc@tLCwg4E=5vq8%fU=d%nOj)va8QX9(1{Yi0;)f<TVN8{_p|H*r2r-xo|g?**|rC$
zuYQ<dk5xA#IX<;0ZFcxfa?dmUf}S~ea2h;h9^TmnXRm@&EF-0_J%OBBK|n80orAXo
zWT01uN%FfFN#UYP>#Oz48gqNnJCJKuXsaLVH{#e5$b}QS3dyref~+Ip!NS^+YNYGB
z(0F6$urdn3*sQ3{yqyXNeY;Qx@;_nisaw|_-Ibua_n_DjDR?U_%tx^2Xf&A<XFbRm
zIXjBII0Z;_t}}4D-}aaQ5bnk=%5}Dp#_fE~O&~k;1j1FD5iUA(7}YI|RI@;a`Aq`A
zK`Wg7T9@34l2s&pZd^D{N;pn?dhQm#)obBg<8shb0d)pBJn?y+$WaStZ$q7}hZ2yL
zuEI9ZpI`=Gmvy))!l6>Y^JT2kN=ZxgMvP*Yt<m(1vq}>hZvu_C1ZVD=lN#7lT5`wU
zfiBrYAwL7A7GaO7r!wRBWm}lO3^RKwB-}3Dc_~TWGvPwC(}-awiL#DIuyrIwLquF7
zI?D^IGzWr6t)t)v*G6696wf_8p7a3=UX2y{=J2q`8e`lu5puo{AT+5Lm0Z?k7f8{t
zQl)Ual=Y;@bONZ!C!>5D-0O?7RS!#W^5aOKlTVT<a3`?JCDiEt7L%lC)R4urQr!Zq
zK`&_FW9Zw*^Fq08XIy#^d&NJqvfGv7B<Cw!I9!W@9E=7nM>$Lt1D-Hs%3tkR{jQ7A
z&3N8=cqgd(d-n{0z3zDGSh=JSStZC#xsjWTOo><p9ZKQ#2Dqk+#kd+|;fic%2$97Q
zgqGRFQuxNG6it)#ZoXQj*cu8QCG1tRGIP{Wt5<6%i&@@wfKhF9<JM7^D97%j$#w@e
z-*rz$3mqiN?c-U)=GC%dMB>I&^MZ)Lq@(yV=@#zkABj(1#zy~bGi{6Ur&4nHC_`*S
zLeplknYtCQd&5WL6mc$8w&9i`K3Pfs`W#eL2i*$GOFCk6<2vqA6jCX9WPbL|F(HO(
zHnHXYk2CBWx%n-xe>ECOy+R<Ulg5FU$EzS2_{ks3fazp-g<O?@R2nC|KrQ$&<OJN1
zqqC`y7wz@vxgWTS9)){7z=4E6jK`{zwUXgIC`umdXlGqgNV@n)_5z)@Jbp(J1@RAW
zcy3K3&`|N4|MGpLb=(!K)~JR)BU7lhXZ}i}b`XlMpXH97*AMvIcXZ8phW8BER<)ia
z)C4yX+-HNVJ7MWk1>aCp&+NV)`-J7BN6*-R_r+UP1D2=q#sT~H*;_19Lq0(SK+T*<
zvNboA*P{%L=DC5jH-L%b7{mQfR;v*OUCOn6^*Q5_(jIZnq0&alDc;gSSxRef9ZW!-
zwe$gN2m^qNX^GmIj1nG@@%D!Ct=B^Lgb3_-eaxoKB`d`*%DnP?iG>vsH>j<+)I7;@
z&<du98SHq`)>zow&ON~<w07hkp~0}U<7bbUbkkURdPP)RLGz|3-KIjZ0?Tod5|3al
z1ur#CGLdLgeN=j{sya|l#%>?yY6kl!FN0o=7uLcB>#<?}v2u3hdkDi*Dj4D=Q@iu9
z0r5OI$Yn@hQz+HLNCviWRiQsrc=9~c2zG<2gVx~PscUo<0vk$GH#Yc}9jFB5Qw2gm
zSs@AkVM2`b9Vwp};l;Dq7(1df!k~>=$v-2@`V(>SS$OXHVIi-H6W?E9s`(aq4(XSN
zAJ$(<<R*Qv&F+mW9Ue>Ry7?eo-ki5L2+t+zoz0ijAs7|Zharjxq$1WkSg!m#0?ZM>
zM0?cm-FtsWwSrA#=8x$dCCIL(%7?;3c1zRo*@)EvYQ*?zcmc1xY}+&wF7k0pJykFj
z^im2>?WTJpgb^5<+Rw~o7s6v4sbpxMM(S`(uVnK{l?Pfw<juE6Q8ileqXwL^NF8=Q
z%I3R<3c8>5XtSy$@p>s8mEAuFl?c+YlB-v*SlI0+N|~}4Zktj4O8fZv8c$+;DSAN3
zJ^0vpF`VOvz}bl{NbZ+FNR^#?!eYc<V(H#us$<#716@lvBzg!2UYtl4A~-}4J4hxT
z8EDmgiW8c|SjE7!tTnoqijZ&R0M?Gctrf)*LrhiMgTk$r+;lXV-mW5HYmr{g+%x)z
z<BZ`~#?~jf$HST5j)+a$k3cxxBJeZc^F0MItW)E>Fe9Idp0Z5={C?c_P&Yo1Q1-hM
ze;pr4>>rYZE>70Z9h8;fI(8Sww(K$(5jU#MxK{=(s7;H~Rb>t*Y}h5a{=x(bhi|DH
zyU&EUnyT0&?rF1@t}5vX$Mq^rqug%&+G})cSYkw?kOLgKOCHTGqRRIQ{lth*`DK$r
z^!{c6tmL{g+&|j$9og0v()<q>6zuUuf%4VbD@FQtn|+oM?=@KjsaD8=1VsI(pA#<m
zFTaQFhJQcgzebN6?vDi&wCi7};}!|kg%ElNK@T!v#L_c#zg60Tc-(t<!g>KhdK-{q
zL=epOXhr=ry|@_ru@b`qzGL;B7doa+@M-rcu<5O%Y(xon=HscEDGV#NO)o<Y4%1PV
zw=s;Ntrf>_vg4>OD5(W>N4$WK!e}@A%adizX+#<d$H9by8IqaWaP4DRenQ>DUa2>Y
zAjA}p!-i-5g0tVHc=NUTTe)52#6EI(2_)q-jaVl^JhSYSD7)`Ry%N`pVK%!XyG3+s
zRC#6-YNU+-U{o`hD*<1T!Xu9|CaeGoiNIVw)G^ZZvNRoRbN#z>K$yM_a=gx$V_`2m
zK)(*X0{7aU54zJn9`{5&b^t&~6tObnDsVAqFcux4mb|XJuCpwKu@u6MWKWMgla%e)
zDFx#3<K*B*&Yr{z>V^qd5s~&rj*TDf=x>!XFK{`!FgA}`$->;x!ih%}(epfWIq;S5
zBi3)-%X=L+-l*k6u@lCVOr7#hZ;_f5lsc(1??5RM@v2I;+v~?v5!augpH2?v=Q9_(
z>zF>$GzzCfHvy2i>ucq9N$sG9U{@dh*{vFQ`si%{H&FMYoOBRo=3<0Iq@h|+mN7CY
z0VI6qo6K8g7@c_BTSSb^;Tm0L!+=LR|L0~5lNB*w3J$qqE*tlAO?PVMeGx~4eAV=8
zFUey7a_EHZ+X(meg6C$b%_zCYPsSnC%@H;Fa=zHcc{X(Uyn{Ma*CPW&dC@kz(<k$g
zttlJ1GsA=$wOkTcq6+{eRL#7c6kMxMj9GNLH#=|!a%>%<9e*k7LudmqsjMvbMMCYx
zUY>G$48$Fi7IXzM@H9vV*`cd{BI>2@rguY7;6Ro9Bve-1=X$3(Zs*QEwS+TieSkFb
z>;!ZaTfBJ5XU_FH(he_+a(BLY;5bgGoQmf3ScR!hV_t)L<zQf5v$-$&(fO#5fV4nN
zql%HH-+iJ{X)|;YVuBNsXc42AH#|3Ho5i#vz!()Aw6Sc4n}8~r-V)pXYScF|z)v^0
z=^g3BK$8KYMF&TWdFav69tS1|Fl~h%4_Rs8s^#v}PC{k5bK>)@?K?@iGAuj3%TKPt
zs9m-xCTi>OdN-z?l^<M!>ZR9-U1fd>m(McjmzJ|B5&P-m`&TJEXidXs7A4~Xd@rjp
zfZ&y6-*74wSGfAD_sghHi?1gQnMSW~`9<j$XlFZslbp#rpJ#E(RJ)fZ;!=;3wjT3H
ztt9Hfvw)9|CL!mDCO1cJg?0Aeni$qTz8Iv}pkC}#MxxILLK8iz9hUKkCG7nHkX^K2
z7RR$<r0pO<Dl(CF`X}OBGV=8vII-^flM9z3`bVBI#^}HNou2|4k>=gxn$IoezPcgp
zO4FneyXsP7c<_vrtCPC<-}1cF5=?r%D%pGaFHA8g4zcEJV`<JE!kLz5Y=c_qhRHj9
zh-G6@sl3d55SC?GXu8mtD^V<kkh+35tpQN$K7?Z9hKNVm!np+lawsQ~!>(mM$5w*!
zvN*b?J-@s2n5K>9?hJ;*Q#Z3FM6M2G$jkW^lr;@^Yb_`hX;#dLr<{k&E$u^&CwWT1
zky*?2F*XOYN)Q3(jwV}}$cbZg=bZ_RZc|oaA7ELL4G$-Q5y$+2vC}J|R;YIJC5J$S
z79HkjAq8cN%TMd>$_c)C<C`eC<sf1B{pR8uSV)GB?ihPFqS%6WAHwg?Zci%gV47HT
z=0VX-V~?I)KlL16J>wezCk6^h5$mD|d-0^0#H-O{SayYjULQo*7ike!DH3O~6HuUg
zL48kO2u1qdfGfmFBs{!H;pm=2BUES4iNq5?!;@UeF4UT>0g!(4#1O5{*a&=3k31W>
zkZLOmKCkSD_}FExI+;!nxOt6+2ga_Q-7=@<IPZn*RNu3t**7uttW6f&OaZkjF!9=H
z|6&k;ZaWA@OIX->)Q+J<3q+#xh?n>A`G6v|nm9-pW@93*Ic@W(hwKxBor*<)6ELaX
z)WAx5ggDXl<!|GJtosxaHoMf8l%2aL{@VdgOczuh?~Zm~OLLz=g#XGW2tCRBx><dN
zD4ue;whGT=Liiz>l2_~7cUAIfv_wi0f&K&@G3W8j1m+VkTgP=bw`=Ijtw>Ubx^-wm
zjcZJ(G8M0=dEBiN7Ol=7<3O0iL#M{KpgdVY=6@LF%t;i*a|j6QRYsPcrFX%U>&`*Q
z9A2l&RD=YzbS$c$u5?RVzlK+s2May&jeC4Kz28y6S0O>=5uDijfy_EQsELcnp;`%c
z`4RP^iDO_}>k&kA^!Bg<LpR+nYx=~5nzoQAu;*rIo@ZsQ1Vk&6-m6)On;}d%c~#3L
zH_CH5HA|4RKB*AxO<zTycW*<0C@+rC>RKR_NN1&I<N!ywa^i919jBX8t)OBvN%tVn
zQS8k#U1OK%4ks)Z?6*Rvk^W7JehO<(7-F&Wo}r1DqCQqPR*Z|NjZ82N;kCDtH+Cw1
z)p}F>Q*Jp{X(mLiUdd;1I#C9bK$+l=+0t6b_fd?olzY<p@?R%^%sCE=Y_Ns>Um~+8
zOI)j{Lp}#(i9V~rUaR0kzJSU8SU$Rso-zro@%f2J=rze7)!BSdtDK%DKv#|m3v!J7
zBE0Ci8))EqN>!0td%Hp!8|7p0eotn{TZ(zP)~0=|MbiR_f{o@WL6wXOdalp^xcA=K
z%K?X5KrPF(H0{ticgypl5^vuYPt*1o6`A(X4iWb*oU%)cf{w8!*^0+bQ~sba{~Z(|
z9H*yZWhsifnk?kh?xu2v?F_oC!LxXLA{{nzrEQ;}`&X*oghRX3Ro2Rk0^_x4f0?MH
zb%#b3BcTNNy}lzxX4mVKb&Aa(*sgH#X~7Q&SHoD27<lK;=;>>PZldp&B>HHw_;0(y
z9<br0Nk8T7zp2w~Jr!JLr*2bF{UB}|ijHf_j)v%+j@7OoZAliC+OBX_t#Yh<8?z4E
zcIv~fdeu+<z1$_euI#S2quwKG)pd5vHppVc_Uv(y&2)+*{oX+}W)A9%?QngD>OO&L
z-$p{EB5PF6f7V17P;c%Jgec7Cd*6%++VGuoth?YU;PU;$MHeg~2)mgSKA0W;B`&->
zK73($Nr?PZ*kk|5YXS?V+q;(CqvQ(>)P#~gWUFGoq1x%1*?y}WJ(n#3+G{p4j$oH>
zm3-*$t=5x=el`y&qk7l7eeER7){T}ZKAIF@zb7$LC3Tz~A?Wz$0=QKSJg@BhJz_P=
z1~#!<9h-G-Mu*><;503@jz90wMn4xTq?Nw|YraB!yfmjwZIalt_W<f_R?zWa{Ua*3
z;m8Ip`168<ujC8tVK0!xw7FiHiyCi<mrYW;jo68}KdKIP=d)u-_T~8<F{A~tX0N{C
z<yFKc;0aRn^lxL|EmPj=tB_-{d?!FOHx-bxl`u!$_ETAoWb!kU62s_qzQ<pxfALco
z_I9o0ldnP62O*L0U{1eiw>TavjOzaD*h9<f<>HW_WR)S9tU6nnhm&hEzDukm%i9V|
zW#sZEOtP@wg}YC3Qh9*9hHB3W0`a=ACXVK{<Djge9ORhC=)km!cGi^a36y<?g3I0|
zq@34-nP}N+RYSHeNkKld#}<=(GA`^J)P=FHyGo+VOl}$%47qjJgbkF5&=QTPB)*cY
zWA^EzNk+$t8#&qxDs=q%Jt0NmK3j15+m9<;w}x{smR`9)z2J)#r+Ua2B&9mCwmVU%
zH&KzA6=L~Wxwe4#mXe9`+HfAH($FuB(%(lH7d2U+B=%i;;KNbvQN|0o%>zlB0dO)f
zo&R{~d3d_DvuFZ2S~a0v9i}?QA6yx!=EH`t{QV_}i)GBq6Xs9w*H_eI?8{p(8uFH3
zX*;nrq~2;bZsWp*t%ZODK#)VL42`m+N-<dlf3o2fr|pm9S4pxTabkKu^fZ6b{8^dv
zo+dWM^zx%~Lyp6dZVCPTcgbYDTe2OFGoK4J!bM#@@1%~jwh#viEWa~a=wK2su8eW+
z#*vr(sC{zJQmVNEv9(qq9M)Z=bl+#^M~ARWtO9rTqQr?P=dN&a0PbOeTQ<K$t<6`~
zr7{ubi)*@S#<Pl1zWD6U7j*Xt)<znku&=~dUS4ZB&juCKxAu}B-)EL>TgxG;yzn(I
z^SuNLk6g6Mtskix={-nCib#YZ#^YkVe>Y^zDayAS%?5h;5(J0^j%ghtm1l_8MDBgf
z5kq|+$#rWen3g${N#0y%8ED01%{Mr=mssBLr*|vLXUp=mU&$N}{1uC3eFT|YPb|1-
z92@f1@n{j)5#X2dDH6IdRsUEra`OUpVQh4?SD4CF8VPQ7i!JK{(~I1g2><K5Du%Ei
zPR8%?Uf$}m+fPc|dX0X1)rwCaGkzBN+Iai6Rw1c^@ycb0_B2iNZ&J|xC-_tDd3{pT
zwYTrAlMSJ}t<;O@zBtV4yHf8G$esgn6sb_2HUc^2!=mihBEuQdGSo<ix|p~o#K!y>
z)7oXojJr_i!!pjRDzXaqpAX$-zu7%NYRV0G!cxK>ScE?>w;!LN2~_tz+iPWQelLZ$
zh2jlfG+yjxj(T~<r#9c$*sUztdz8b^;=p!24B7jp-#c-stWHLzH^@z8)G}MP;VBxw
zZcb)RS#Gv*_V$#JFUQc2TAK!FEB#xY;f=Y~w4eH_Mk6;ZUcFx2!N!dM%}`7Us!pDV
zi@hjnIarR1={SDgKvS^|Iy`qcO&z<wc{;6qv2dBdq3r-0coF3m1rExtmdrePNvo>0
ze72Hz>Hj;FS+E%i)d(iYUz7Z2IJc~^RDjZ0`-728)97C->{(CUCxwlfuso0vwjtD3
znZvCb+)ZN~jBNS@52^cj{W-5Cx!W}Yj^`_y!9vtCI&&_gDUc~<T1_&OPaarOo&}YI
zG1rA$Z-=C9Ue52gGPElb$$E}QRT%pu=ZT!?7jYW|T(7y883PIZ;tH>rE(y}Qkov%V
zs;y6nW5==M7%l&xKH%-5M_Z*<W6P!7{u|Pc9EHQrT`3!~-|mIs?E3nfzbv|EBYR4j
z7c17_YmT#m)%<37cHLbL=SF)R;_bucUMIIbf-ro|t96_)t~fV{x9&l!_0N;=W&4ik
ztms{d)fbdswe~r*7Lt-^U!6r8H8f=lmrjk1etYs`pY-W8@GtFTr;Hd8gxSs%zUb|-
zGdxY~U}d+S<g+3dXL2AzK!m$QdaDVxH|)v>6oN-&WgeWi;LDQXG|dk8wdm9<Oi0>8
zh_F1Ozh5U15MJ?DlOf{!B|NOLRv3N5%&DbZ?>sZupY2WKX5tUuX@FZgI~n0ZObK8X
z8#d&Xi-*2v3w8}gIogrp`H=!mpO)-ucUJUPLs$w%98a8a-SzjlL_@u`Mg}_4O{E5c
zS0wtTXM_FEvzC<FLB3mj*Q;2MMJ`NZq2om(4OuIT^d6Ua==t;Oq>i9J&huS#!9gVi
zx1_I3IKwyV0LWeL@9R%DQgrWDw&T!Ymr)u#eDkS(Npc0Au^(Q6Zy!4E4$3boUC&Z8
z1L~Ia2tujr(L1KFkaHC!`$r^gNik@jtCLwwD?&HGB@}RaZL3W5McwSE>X7Y4>8Kis
zs2RG{L4Wfv1aW2I5zqn2_1hxq%rdT@tDf2X^+MCCyEK~<qL7DE4VXbZP_I14xYq;?
z{m?JTGiSGF8gmqM=a&%G4*Vue9`~QGt86T+3+6K~lm1?gX|Tge-F1TS8JjZt1E})=
zeaZw*Aq0dz?L>U0>cTCi-jf}XOozMZZ0h2Y6#OghLp{Ex#BAo=es_AQe^Wr=Vbqdv
z?O?ah3}i#GBu$@fu2Gtz_Jl)^Kb1wdP|p(!{~C^d7Zo93wKv+*z{&H2$cauMCt{AU
zxGI=uccnV<l~foD4SUOG3rJc~nBM5+{obmv0gA@A+YQ3-$t7>%oLI61*=haK8bs9P
zFS_WDeaHz;2?@KO+G|}TU+WEeplB|2z9tfS8h5({<G^aK9wuJAr`3V(aaxMQuDmJW
zxNEDt@aX5jgEsftg<CSHa7{`?`U?-IKHtkCwh4km<k5<?R1B7}nN7>jT?TnNvVkA|
ze2a|H-Y;+b9*rZF*86e(<CbYjYL%qps&ysF%F7?Ou=S@H3gTJ&ADh=96SQUTOl7)y
zyy;){TDC59wmVXmI&;2&FpefUFEA81L0-?~+EZi<q^!Z;hg|#;&Y+$RD=x|=;;oQ}
zBZD{-s^~~qS?uIvJ1%x7LdMG=FMftoQ%)I~`iqp(Wr_5?d5SbZJS#>amkUXz%hE`P
z6{OM;@vAp+{5+%trYG6Q4XzWzok$B-+I|4(UuMN?3N48@cSW^iJ)mEJGQtl3NOs)r
z(LzQR?_vRI%kAZNCQ|7G<(7AS-h=Mqm(3U3j;i&?t)#w!0Kjfc`i?p5f~MV)tb}2s
zBAgeuy??i<d$bvZq|Y8dKL$Oh-~gF`kPdp64G)+o)7Hz69)_Az>z~>@rzPAr`G%59
zaZFm(j#m2r*n11NsMfX*R6;<hK_#RIL_lExgA#@Yl?DZr22oN<a)y!yDM3;~K)|9z
zy1PS=7`l7tuJbIvcgOdB`+d*(1I~5LwXd}Sbusg-=eh6S{R?>t7!C!^D!wzeCyZe!
zu5*pfU2YYOSQx&)_x0SCAKr-@*1ccl>wdnMpNtZ=*i+Pa)4j^eo%$rzkr9WzOn}q<
z9yFkb6G+1b7sF7OyC{%sCVcB~9NU86Drd&!?UH@&an3iCa06Y(cS1X?BHG3wcO{7e
zbU|%Zp6tY|Fgm;_Aku5R#mB{D(d|zs=8k@9nb$Ct*arfiuI_fPvOZ}|Fw<}elQ;`=
zdQy&>G`^4TqoA%s6+J655En-OLbq*Em626~sCg~WZlX@iosJUEirGp=^%``xm&^Tf
z7}H%xhF-nyJnqFV?!L7{n;)J#e2|Fb{rqM11;*KZd?q}TKSLA(WQ+x>*jx0I-d#)n
z^6_8K9CUbyui-zRdbBoi<FCR<4fnJm{XlcY5kv{bqvH5Hdd!d$=1AYLlImKtxrV*!
zn}dQh+mVXUxT@+q9)Yw%jyw``c-T1}%(Sq_CvznChmzP#iF%n^{I*llTV77~li?*c
z4e0J+9j+e1c_1hh@W?a8x-=jAexe8qCU5tZ*ac(u*edZ5LG;qaX&oj(f&P=_E?Ii-
z>|%dW`y7!up5XQ>mb)I476|%)1)b{Wi0WH=r<84xO2VAGN8+9(=UR?G+N^I-H)y>3
zSat2ro`x}Ug2K!eo^lYS7-D&FF8Dw`r@1N^r$a_u)+OqR9nJ{4G?ar*ur9K^WUyBU
z_5zWAo9Q?(--*xXz(8RdS(i8Ux;z(7%^F7xZzr7Y&~GPoxk=4O)2^Fxbyp6_f!g%%
zh#*WJYvPcieqI>Tb^fVeqS5eaEx+Y(1LfQpq9aq|F137YRWmu=O$>iTiAo)mwipnz
zxr6gwuuK0I|7e@&j!aU#V}H5CsEutLxWkfpj;)&DVsF$I?_QSS>6FmsTy0MDJ<Sfn
z+4&^uG(`?s>=;FRg^$e8l|82~34Qt)qP=}eN4YoyV&preb>Hvg1+l1c-ej_`G*)RL
z9N3jd2(0L}rMzI}=~oyE4Bw`t?q1IL^eM;&QydA4yq}<4LBHeUu;es!9o&%-oy%U0
zvVDZjS9Qi2U;bXv(YRps4qxXv+-t1d4uq{ofJ;a;?5La&!32Dfc03OwO1htS`yHM0
zD={dDcvIS1@K9Td_KS!|v%=%yPVvt(-u$(MHV5|+sdfB1DI-6#jR*(n9MxW|b4E|O
zNXHq?i*)JSr&|^4QeC>tzl>*QIE+n1SjHGj`GN^NOOjhb4y6AqNR!f4pLN#llF-`k
zd(`+nzEr_2lSyJ&6hN-_6cr_dUB`F?_Iy7MCM@OizYO9>9LSN(898p5o$nxw@9>SE
zV|fKkR~WZT(85kzUPktLu4#&07d}0eTV6^I@!3V}k>93^E(vLyjSqV_d|IAt0yYi4
zKkC-&>!*8FVuxefhW;0iu^7wm9m1!blSTU=QIfKXLPzJ;qgS@?-38b17Uaq@&)_+>
zhL5&0N4+uO^Hj>~oc_}XYm%Z{<jg_-?F}q*O#L!d8T(49i|J=CyQ`+1(ui@>tdqM+
zfOMpzfkU=*;f3af<7kJjnw}R~m%6sg3zf=R52|I6AiOG;_}!a5rcqos;ll~Ut5wIR
z^l4!Y)^`rqP0D5&E>@Mfuun5F$cn`y>DHPT<ZQ^tC^k;3GySm6JS$=DOlw2VnG#Hi
z<7GO^Rfon*N3jP@aPF$h%dK)vK92=?K{0z;9Ngb7x{=^<M+d&+t-Dm)<?NyH(!-g(
z4u!X;feDB$d=1nXICU?DR6sdfEJ1p+e7N5IgmpPr?u@yAWcka1f4yi!=gu)maL~)F
zMp3wXGXO(R#mv56yIY9(JL;LvMGq2f*WUcj;{>sL)Kx_X9jjx6t4Szqpi`FOg&x1Z
z??=k`*YH;^L@Hgvfi`cLSA?Qx40}K)ji;Q>to~_}yscj9`cxv0Cs6#c`^e7JSCIPB
z1Uj5u8iRPM8gyPbRs==%6DI6<Q%|mWIg@S>@FWI#dAZ;#Ve`OAyOrGJSqrFa40tmE
zPiT0>Uj+mNPv~njwG|oBdw<=!6ih8G#GEk%`d{PAll36j54MbSQ<!GEwAjvrCop1D
z=NF!JQQjwJCXBilzg|s7m6|7w!h~6``NL&-E!L;@UAA*$Xj)uYJ#2<tmHxF{y-)dI
z5|tMYO}%QsgArnN<)Ze3lsCt1Xep8hmCb}kw0Z=QO!cOTxJEdwFkANB5K-3mQiFb^
zhRG#d!G(%wtV8#4@QISD!0zpd?&Y@LMj5wk=JbzAYL0Z4xf0Xn_2DEoe1enuAB!lp
zG&%@VZ(~Wn=t-84a_H9?Gy}tXhLKNQQOT$yB2?z+WP@;FTtn!A>t5+|fY(1x<8D0~
z<}QiGxwn@}<3BmAE-kP`eXnhvm$_cvc*&N`B%%2zPjQK5CtA2o-h<%nzOgt3&awm{
z!yQXW@t<De{?ro!W6s7T^qXo)1=W{BNglFf)X$x?9lzh!>VKS35=w^Ye`I#M<v!RF
zRKS+-g4$aoO0LMWg6}?GpZN14K@3A08DEN&wzvUroCCrBRWuq99V7$%Xx`&kU~VjF
z^57&$n);g3+P>t$#p+G1zEVo;lBjng5rh*-^D5Fk6qOkaYmS*eCuxWuAroP#q)nN?
z9Lj}1Ge%-i>J$r0ij<qmu#)6`h*HIp+q(S{e){6<&9+k4ZuY}r@-`!y?F02UQp-?R
z#T$Qtd9am<@YP{>(D4SzNAhaaIqBuqT4qV{q%Ji5LelUXasvO0={Pl{Ovvrar5A7x
zxXB59v;t&YFC1eLK?DbiZPwz41_{=cE<CEQL^2U@W;eo9yOi;|@d6?T%ws^esJ)G>
zOzCPP<f|Y!^pQc9U!Gtfsgj|<n?tgk+&Pro&N)cAltBNG=svAr^7!*(ve~dbq2t|2
zsRmH$bd1nx2gl}yqPOGE+`AHhc<E5K43-;Om@#g*<cO3+y0fl#QX1Z1`o59mNgfTW
zB%}OHgtKgZWM*H4G5@lk+n1ReCR~3{dAE#xH}~ouK8-`vc#Ff?E@MzcDAcC^&#PJs
z4vKOu&IhwnzYBE(KlW;H!k4E!OG(BtCm`6Azg+pc%j?9Yn|2b9=T(pw&Q7YQe@qX$
zArX&9yG>tj-nhQ0BH*$r+1`}SPj7Z6z4ZKKl}hE%v~e~b(~XNUcs#ZF&RNc1ovV&n
zVk`_EU6CVceOo*cr9T!kJ?qSOX6a<yV*B&K`*j~~7gnZY3Xh@h3<D%`6X)qumC6wi
zbyl(5wS?)w^X5T{LNj4uji&;SOA4A|`5D1bc0i_9w9N2z-SV&V`zjwy>7)|xE!eIk
zZSr7Vga+SwZj&0xoDbeXT0*nb)9}uhUam@UYS#z(tOz^?P4Z4S8YLJ@GGaV+qg<{m
zQ`;8YV;^{j4$D`Z0qIo~CM+v50m36V*9Qenq>m&}_;Q#|C9VhRi6&1!euJMJQ^<*R
zCA)8STQD$OV;VHc_8x65O{{RZS2bWfd!&x-<7|!LXU!zlPU&dxD$<)015x9~;fx<%
zj3$xX8?Dk({(>a5PmrjX3ll%hBa9whl2aOI=CAo712@Q-+Cz7he^rw+)3#5tN6)%;
zE1z*S^}G}sS=B`}Iqo6WmOmzuZXR=v%MJft$xp#bmaO;F+)oP?{%pl>>U2Of0g_9}
z;i61>KZw1Fa030wf9@+b5fYCcvJq+QSHp6&(=(I16^WX+Hms~s36sAVl#NC19g=S6
z!3bE=CD3YYeQfyBS}p#{LiBdc@C}NGV~m|9iH`4jhr-rLR2fb*@Fmbu{P$Co*7>N%
z-fv_ndX&<0I2~d@gIsRU%m8o+%kmr!{jgEv1hh@tnhDaA+fR5?^-WIaj+riFx?2G;
z&a$b+5|@+@ns>H{p^9?f=1o|^0jbixMv{xbR}{{?(ciPnsy6R!mZVHXbHMvH%;{4R
zx?B+%tBSun8`ghsHpI1Uh0~due#WJnQ*V{}e~u35^iWl-)+gckv$p+R@oM5CL+mys
zl4f`oN|Q3ZS$L_4_S+U`7_Yr0Uw`Af!a;tU1BcqipP}qU#|)m#Z4Td>?<au79#fJU
ziNdEBd8=yk5mTnBa{pl}^x#|}DyA-*k@TO@J9w+?9k0WsDj5IPqy8-!U}nxR^<7Ke
zEwW&$_LbQ9x<xJ=KSg6+=S$lnqq0b>zc1OO{=gq)6ZQ(Z)2GhRwfw4Dv_=ktD4uv-
zm|47Vvn1uF+p2}W1xjRXJ1p-l%36L_q%;;;!g2k;@?uMc#cb~|VzgIhlYHteO3NCB
z216)ign~l-6SiU+z%V_l6MfCThty;%t<3(`o4pwVdg=D&Fpq0*og8#Zo5t?|ZaX_Q
z&h=!8lxY`w8|AtdnGEiEz+Yr+5sQC-=(iARBs&Xd2E6o_5!b143q|Ka({dqUom>*b
z_=0_-tedx_X(8HvsaZ}w{qAZk*p%`)Z>|hw4u9jMK9P}PaN;@QramB4YZ3D?!ziCv
z#q@u+QtVZ{@e*`ER4on8AzJ$(^HW1J-*c`}quKXz2L$_Henh*K*_#gQc4M$ApVL+|
zPm2Y*W+buBP0rOgg!L82c%P+9iV26ZngFF-dH7D0lOAwGY&q{4;gL+rvO%XM@fW}#
zU0p?X1^;&q@yBvlkgd+Vtbimn+2rIaBfI7hEY&1!a0h1K8SW*Xtii7L@TlNN8hq$s
z8c_k8&4_K2FT)uyCAUE7+$+0Gvn(n8!)dflUrq*~7zXK=#B+ZDF5-3ROP~;W@e~MX
zB2l680qEUvk263_zUdoKThYh1r_=wI6ys6<@OUGSx;6&Vo#*i|-6SRnra0LC+>$B&
z8#6GZOx>i1^>+Jbh>14Q+ouB`iSjhwoBF*kg`!~L<5`f`1JOE=xKzRDDW`3%hXY>-
z;Z}lV4_;_^+EABwx<4J(w5B6{2gsnFPQ}y1V5fjB`Y_<p2<kD>mOJnXpRZGtuMfaR
z-0a!A5v($T(iA?JiR=Z+hBy!im7lR3L>F*6(rX(<Xd<bTZur#qv*RpdthQxVq?>Bt
zthnAN9m;O42|iK^l7X~0m0g%y?UXwLx1;e`*jl}gN}X~WCxI&*9KMo9vOh>%iB-J)
zQ=7_q?v$xE)grqjL&j9zh2=Q?E82CXQ4|bg51MTsKMD2OWIiGK1#?FH=u-<kRKH@T
zHVzwr;!ZE%XpYuob}B*{$dU%Y>m)u^&Sp;iH9IDPJ|DIY7_N#1wnBY9{WHY`x-4Ha
ztgeg6fK8%?28Vgd8sBk{cb>|MuK&Q*GGWK)35?tddE~Ym3o@tM7|RkQb*;p36_Ix`
z5T#CajiRuaUIJ#vwSGN*#5!e?OMM*^OXG+7$p(ubh(G2A%(%vJo*4wYjH$2xI4B`t
z`;2Ix`p29%CEsIi)Ja1s6bw42$KI|#6_1MM&7gP$^XgHmCX`<V4C<H9@U-`8dSaF4
z0CZ6R^j3E=n62KYP*Tu*z>j_H=uOJc%amoF-Yse~xYZ?Qtzc1BjZ0jAMP4B-vpdm~
zTsV%rGJ%|yPF)jHR`!tbcJ%pNxNHdsWh;y|M53Q}U;0qJdCLafuiGVqi<%jtCh^MD
zMMEGkilXXQte)-e(2FomYETK>90qMnyNoh--tCYXfHP7b>FO9^-8!g%sc7`#l{m+N
z^OW2J^9%@@Kr^klT3aAHxkWu3hsp~!yQ#ZCsZPk`G855QQ1ZQlU%T?_yh+g~FZ*i>
z5sP{h9p|6%b?}(QT2~$z1>L9aj|<jdVE{}X2ZfI-wJm5fE?thx$AB!5YXB+H=z2LK
z^`<(}I;KWYxP8fsyuvEd1#7GPmg~DfwL-IJ6QQc`Yf&<NLS7_l2oh%k`I&+`?Ir<z
zTt@6f(wq@y*3NaSq%QBPxh6vT2cX0J;53WVAF7PaAMkZAtiL3GC$X!bBJ#RoK5M8T
z%VBPmT%{ur^EGZKZ#%@D_MVrTW9iCRE@3k1i25~n%}tQHeyCSjjG<~H8W2c0<T8|J
zi&5kSz^bx$7YqhvMf%u7%bNQD<t8voYPu;bthAvo4Z#?4YjzC5NSa`+b3Fd;^8p7)
zQrW%F5%&HihLc*Ez#c=i^qTm2$oP$@PsBfsmTWSNvu_qJbOT{gf+=x8^0=geuTq(;
za~Ept@(DU4tKx9!3#%fMh`#rM`dXIvsh5EQYsbOkVcMyzOUToWumq10lY(!@g{&S8
zo(#Q};e=-a0+`%XtKP;QB|A}v*M)ESgZJEpc<&z62VOQ7b{RV<5yy;3JumonKp*W;
za8f{tcphd42%m)QzCUb<_v&9FXzya4FLBS_eq_Vr@7O@8c}M9U_;@-RK89F*`|HrG
ze)CXVvij<enRnKI?WYUko-yysug%WVdBp?@G5bvTeo&5!cQWI|O^fzazD(t<B7|4&
z`MkP0Gd-xrQ-XloUgv30b-^|9dmSl+-t0m%g$a%cUP{uIF5zs+X(l&u2olkU`7}}O
za)$Q}Jd?aOduaN`LK@Z7<Dm^V<sGo^gD<UrQkw3n5RFr=GRGQnp2BR41uFZ4-NI1I
z-vRrqKvZL7k#Rth<nf56PG+)K08FA4(4mH4_u*rbL`{Tu1C~bQA8+I*4N!b#ezyI|
z%gUd@(G;bt<c*5DmCShCM?2t0eoB~o1*+?!AxUkHqZhdsOmz%2gS)-Sr+u*Z@EMj-
zh{qXl23vz0*jcNurUtkL7{jg{+^L1zO}W;~5j_bYoTR`_RC43PkAz6?^}aMOfHVf{
zoHHq?@+%wW0F9<XoQOt$5_y9Ml+W@9Itj%_>_zoj{O||%Ja7FI>|#4?27XCWJiw}W
zK6i80?_6I+`_N2Fq}29|x>(cNy;e^rny6_2s`lQ`M5PAjw@|3Y?5mI*HD!Zfbu8@b
zh8Gc2pN;-rE7j{xN6BG&<zMI-X9C^OD6rr5u_w0*`Iz?c3|5oz(Ml>OONt-`0mHUc
z)Qt6@Im~naN)3$_UXAEGzxy!l5g<LIE%e!s<M-OdmAE@Vv1Dq&l8SR?=+AqtJ2X7T
zNZjX*O&(29VaJbpRg&^8Ef5exSBAdmWZxARvgdK^E-i><#HE|hef#8*+0mzgK6Q*K
z@_8Fh&vh%Lkqi>)NS=J4(liU~;;z&>?^#*^>?alO`pWtTL)X9`XEwk9VpWOm6A7k8
z6>M_;wI_>ST3@uBYO~Z^|3OS*&fA>MjGLBtPOpzfyu*w7goFcruAS#(i;r8zApNpI
zO0d@tM;w$pkfdR-;k)mMUZDtBKAQ?G=mZ%_zx-T57txR^9SEVJ<M=_ZzfCn9JGRyC
zOr35Dci~A>Nu5t}N|d%}un4@lCjql%j!xqV2}qCGzA}9yn-wcn@6T)r477O|&7ZK7
z62TV<;ZtzZ^klK}YrAjBHa&?*kZ73yIbVx$AS>Ul_mbWyd+mH5&y8n)Lr!E=>*t>M
z6<=msUrG%kiLMmfRw_YJY?is^Ad|})n+oITX3TV|i2Zo<9(fdd+e@wzvBe4UvK0Cf
zf`oK>qK0vg6&~WmTpX;g=!|o|mA!Snuhd7cqemTtwj{Kh6{r&T;x_L))#Q98+4_B|
zG_`TU{qhW}Y&;G2rk6LHzOBq74+M~s5=**5Yf%C#b|*ZIO|A@9ib9=wg=Xq@x~4Jw
zo03UEA54cEmbZj*c`$rm4WjYB8uW95kJH{Y&wH~>=BN18R~D>6X0?UutwC0*P0LrJ
zsMgXP{@N#KtN@@xZ^f3e9rS0T`A-+^Hu%C9fTI;r#*`?aY)H}j``6{suYYaWt_#p#
zZ{7hD>}T;WoMiL9n2~7+T>0Dj_ze>+ii0g`@tkhazklc7f6$%ijYAEUdKpwXIs#5K
zX6Z}LOMr@Y<T?{C`ybdOw_VAAD5zFIPQhnlHYC+=GoHNPRH~~;>{sglw+i6*FY)FI
z?j_vNsJAM@e@?dk`65N!)9gwd;~qH7egIvvNJg)Y=?CAbH;`cq0=VTRWvX&X-y78g
z5y+Ad9kKQ@V_*#{W3T?Lj%iwpuZEdPuAgVoN_+KW#<&1D6M~d7mOzyLqk&4>l4h-s
zen;!ampL7C`M*`bzsGIQMZg_)I-6wE|JCUK)BFDh7nf)T@@7c5W%5w!jzF)%ZJAbJ
z<MPs&b~Fw|-CmhV7YbBH9;8F**1M1*i7XuCPQLGrero<H_J$LPBfh#dcCJP6z#T}8
zEJ4kYoG~k?vHE$awcvMc0=1`qylyk%TmUw1w=_ec&HX>WWG{Hd=(`J*&%TR1!VFaQ
zH-^w@j)~qV+_=9t70kn5A95}JnD;JrJC@Puk@y#(2;-h?bJ_HTfBzc|`Y6I-Hhq+F
zm$?4Ahd*!T|I5W0G_E|;GFANJZVgB-6@%8Owyqj@+ubFw&*Xyk4^S0~fW_N*#RBN%
zvbgh`S<ddI-2LlefR^e%zP$L9!TROXuWwEH*Ai<V#BU5^vQ`k<?o(Ip7-W`xf(1$N
zIQPI`PuDB07|UA&sPaxB`8pgWEWB11-UC|Z@8t}HXa>Oj!l@|%(z=2XOnu`9n8nr1
zfs?S&vFauyiWda{gkp@_ejQUG@bAhu8`m1iN|-f$u?<`(SS)!zUU@PEIdB5W7sKih
z?apFy#PUKGkBG^C{-D4+e#p!WH(<(r_4$uyq<Vps7f7o7VCxVN=@!WJkrPxg9+p6(
zYNfuU4?a)~SZ|iZ+<<d=2JXWMaHvtn3Gbjq1*+*V*h@x$H$a@lA#ZY|DPUZGcU$*n
zvJ}>Ys1P0aup{PpQ4smQl3Pb+fz%p2Y*BVMl4Ep97X{rRze#IUmN6K;J5Rnj)le3i
zy>bMmi`~TwU=mXZ0^{34LB-x!nOCYN(ALvw25LjN^vBgo$TtB{stW+U*r4ErMPH_e
zY<5~&<$BaVerkVwalzZu$J4F)=chlHQd(Gb+4#xJBS4QgssgHgIsGX)v!I-+Q`aM4
zl41=sipiE<=c#g^JdbOD;X;0Vka5OcJBQPQ`QfWC-ct-m^?+yWg|<;3+sfxJ+4)74
zJKx}HlFz;aWigFjkp{57Zv63-0*Ats%`-*}zu*`mCw*f-cIpD6?y#)n09%TMsfLh3
zyYBT`VxB=L`k@#NG|k^3g0zzGV)~!I`y!CRYZi<GdVB0QFtgtUg2FDK`wn}vYZ-F5
z-PP>ys*gJ9)*nyqpTFVXFWg|^uv~I2;m3bI9orQ9zx@h;FnuL)4L!dgf6SrC_c{gK
z<S=NFJrYD*?COXiWXjsvw0#nIixtc#dm#BiyyEM1Mn7h=9vN!VNoGPw!VKlkvEjFO
z7hLf|;Ol|xNtof}lgoUoDwCjB{0@b&zw`B+Jt_y{*zbvs7&AW@?p!%JbsPPesWxOu
zQ?yq;FA8?)?$-$v%L_!@7uo*LCi(Xb!<-GQ<8OkLaeu61@LZq6uLK|Bus?De4&}|p
zXKb64Hf!p+uw|S}0!%5|T!8i%NRjM=Ip(0k?T)@d9=HLXuZDDrBW@j=UIK-TY+XbC
zu4J+-#X};&5OH$ngsXX>WC0IjpiSdHy{fZVz^0RR1Hq>8ulKd|^8WDnDHZNl&^1*v
zqSqy)UhRZBmDhnIfNvN)?Z!8nAyq!#FY~fkwTF{6&{Do5vPdxwmH>V4kC%WAO`%z~
z3<DGYTWn-d%EWHzKt*RIu#3ZhD(snhrqW%J8sO0owEIcft#u68^ArWAn(4nohgX3|
zX|P$IOhFz@GLJC8mN35DqM0pevAK>1Xr5y=oJWdnA*m<+y-iPnGB7#GH5TzWS=L_#
zckD<s=?*T3&rZhAhF%jROmBY8eQgE0NPd7xy()L)qBnSYKZCXc#MM{=p47`{ft)Nt
zki1|Bn5_&Al1D=BLJ*80u$!lL1vivQ4B5I1{CEXo5e40nliW7I$Mm}Z(6A^Dyb~bc
zCKsUsM`X$5<F`kK<AX<l-idd#sO&-O=vUMN`D$nY%gs$I?ZQt9F?z0MGn757dc*6U
z#fupU7_ga}03Nq}@7=*6un!rZfgxbRE|hW%nQh+LjgZYSLij;)btDiIQ#&=dR+(|g
z+JKwJfPQ*lu;En)gw|O0&Z&$=u`(_eg1li+P3+LlYv>*Tn+Xal+KriVL*n$nLO=fC
zTeaB2K_g;qu%`XX+1g<#wLXh=FA}T@P1iJ@nz0Q73oz~0G<tY%t)BbAY|l%hNJzQO
zN~+-15mi?3agbnDE>q>lw-R8j$g<Y3A3UozqOZpjG?1kaMpfRx07ERubsy9>esThs
zmrFKctMkk(X%fq(Ldv}sWn|F->5vaa)7EQrHM;6#H<+!7!=n6ae&=>u@<ICxVORIB
zYF#dRhJ@fOa4N7D(adv5nq}IrCN#plXf4(@Wye1TVdqqgxWNP8O6KBwLaFzDRh)wU
zeGsH?N*RB4IlN)*+M7Uv%1p8ZJ#WqX-OuVthpTjy$#ey4BCH&mvE7cV0B-C-j)E4e
zM>(Y<y~E(4^`WH?z{&cuD-V~n<mjtF2_lQCM`AGkOzwh7V)9&hK46Dh3w~kVGyPI$
zaCV&8A!>mj?`HX}NDq<dx$5tez$3-O((c$o|0w=smdB&Te|qTr&GY}^;QgO1-0Kvy
z7=uhaiGL7AS)<S4K@tcHl94WO@W3Zffb?qsCGHt_?{izb$Ol}2F^&S`N$!LFd9gDy
z$OKF3et->Dcg<CkB>HTU+=IxFeBk!4o4uk`qJ<PahW-pxxaSF~AY!A|d}Y8Gtm9-(
zpepbVbGISv2$Jo#oot1EJ}^jOA`74*3b;t27)~IKtibn?fKbB<u#58pUk(hgR=R4-
z%X^mNadK1e)+{_V&e5yJ-6sZ^*u09u!#g4pzK_RJ3_gyV8SNvcX(1S0<n`Se2HUZX
zCMG6+iHuyf`lvz7tv0U9l78=pAm($;AqY=tE;EYr^;Ed4P{Q0_%MadjEG&*Cl_eXX
zch^^bEKZ%he8fralM+BhQEM|k8>5e`e;FVQ`r8fI0z6tG%L#xAe!Q8ww6M4IVEAjx
zol3ak?<$@McO$Zf)kC!j@cZ*A2)NcFppGz>3s?abJuz8fBg8dSlHB-+`X+uBavvK=
z@k!fq^$uwW5qfsLj=V@mmMDaRWwK)fGRy?fn3$69Wry?E+27kp2CxU)lH@Dt*odPa
zFvfL8ySf0sEjg6WU|iuG<h?R-1o;TTVWJbBc(0$7p;(xYN9K{#R@WR$0Jw%<#-8NN
z0+hDgdHSJ0h|}?~Pr{{KIn?n3`%buMGDtXv0hLW#>OGMU%zqZJy8!F>2LvRoEfu#<
z-A%dx;jAKJFizv~L1J9C&A`DeVBd&sPDOjW%g8=nGCdgk{eUTp?mP;3O%e6ePi@}i
zWwrnAc?r7ITljTftM)3^LJD&hbbo{bTYHscvl!F)TtbW$evt?@MUmk^;u3+0czl~W
zp2r$${5w$e!h-_H31H1ES!ck>NW`+C@w-<g@hSq0lH}Ie+zO}*7)0cnt;=3)GP|)Z
zk*3-v(G?7d5L1-Y;K3OdCrTvCgg{N?U;Va{*r*yzDTfxyM!_h40|(F1lleV4F)$vj
zH=U;t>D5#1FAxyX&hv##lSMD2c#T>!d^6JAZh~oy1E(z2&TV$Hdr(&FV1@|HXp?Ie
z31ZLX3k*eWQ~v!7{lC|T*Cu$L{`E7mLSBI1%3qTCW5#=D5mdN}ptX2*2#r}rrnT&N
z9=C{2rUOB=K#4PR2~7@L5hDKqg8l)PjeQkh94+T%Y!idS-V$GaOMahyFZ`X(%-fM^
z&9)uHp*$i;evI@1LhT{;8yBQVYay(OBv+BsL3SsjZ3H0p7ld#=-jWbOnkMNqN9t5l
zdLT>#UX{@Bj%O_YBwUURK6^{d@K)J+*kIA<O~Bfl+nx83xyd>+gZzxnB{2yzNmn9?
ziQkg{WEXrMjqp*x&R#E}frHq1MXPBV;(NNUB6QYcb<+Eq$qKzg&>;T$J_7ebSu>$-
zUpUkhl@Do^+AwF^PU_o*xwME@H#iLPf7QJYbKC*`NFQt!Z9kg36nnxgs~c{#G?+wW
z0>L7p_1ztnY-}QDYo9NrA*$hM-hn!bjoR;{vioLaO9i7WV;0>Y<xa<Wi<SnFMZogV
zvV-eaa+MIVVb^fuCnogfYRPn0ehCux_t{&H$#A#$-pj)p6*LNcC5iVlx|S|9z)elb
zPF;R|uQEnwkiy0OgGzdX4EN$R)0uN>WgkV5%PT&`g2x4Su@CZUBJ-KM5az+Vq1z=O
zm9&UG5s`IK4{O5R8p$8YAADR=EQGnWw;3z^gJHcsR5#_F&iyNU*Gc<a_A;VJBK4xK
zS{avtBdf*Mu1n`S7bo$`r<ir{sSEsEU}-cR*NwfhOF($4(;cNV!cf?D+vl)bJ_WlM
z@AuiEAZA-3*nY|2fm>G>&t5F8i9Ik{ZP;p-aaz<KKh&_ZwJC5z&?+G`6cLv!TJ;!q
zMk?Gha@=>m1iC657#R1Fuhg73GbBe8Se074)cFQQxMQ;EN9fG^t|sL{9mycKExBef
zu-}<_B7YX};=(H4s$*wxus$n~NKCV)5D7-0b+r)nO!0J5db>>L9L?$pXx(Fvjf0(|
zEpuxKYNB=%HlKT7sf+#F1?O1yCz@}+MJAy-EebSeyiXZ-qUXa<77B<yt`}*mj$#f`
zM4tyKR|Qs1SVNkLtuDB9;&c`aV{LDh5ZzjN@BZ=clZO9W9*Upg?@vAbbJk@Lby>WA
zUeDwwV5`1^#YA}nyk0_!AY%9ej@!i%7#zMY{-!C#imzBRg~r*j%c7-XxWT}2LvY!p
zn@J)NRUtp}oo1w0Kfu1YQavj^%yEMy+913tGBy)Un#$EF7bRQShEs{X%TXeizxiR%
zF-HK-y!pvE5@l~L)Fe87ZO@#lmTYgL*S_>A3+?%P=%rRz)?9iiVfsu`_$5n&m9g$A
z*c1ys4D!s{{1jqBh09Vd&FpQELIV?JR&!8xYZwjLDQ(5eqcfB3eo8eKp7EO!zoaI_
zTrU$PdrI+>(%1dHyxSJ1xfIdsL7)W`s&1~hp5d!&fa%N@Og`H03vUrEddCxH{J2*o
z^vdLgDcREr$uYZwa5;P=Rxum>WY#K~;I<Ykc9PK~sidOrI*k6^`3QP;vif2oTnh3S
zyhT}0Z|AFCN@bpk%oneZ&z>XnsgGix`UM>Vy$Ge<9oF@hsk_osL{Y5}C+99V18OA?
zK9Tw!b>SraL$)wRzSM8)ed!aH`m_!S{EE%ioBkX*re8#Yb60ido3$CNNH>_DNd;)U
zNjSm(>AA`py+&o1%naD&RIJ;n!0BlqN#^4ywr7>5i+B*~!QkfHMVEDmrS_iEz6YP;
zm)#Ip1#7#5vJ+|aw#@w)l#<n@P}cH~<rIgMQiV6An{v~2TD@G`?7ep{8rt!ENEUt?
zqYKB{^W?XPUc0_$p0Gep0gK(Fmm674y597vDk~<xN9gbU+rM9{_=0}+@xtQMe-%wM
zIQTT@WGAM3zuqNye2<Md41o-J(Pk*(bP)9lE^98NHR$C@Tyy^(kG4y<$5eeFw`47S
z1DH`sW<?bRw?VUho}9g#_|mxn^9=Z!$n)mf5irK8^>Z3zL<Q?hg`lDE3r|IhShL<2
zf9A6vB5|5W3zI+GE{G7V*F?^mZh~wnv%)82wU=@5CHw7w9HljtOQmC;g{xZIM5ylx
zBauEx;SBTvOadZWn6GM6>tt41=*3CAq#vc{r9%Rwgk2Kq6R38`$qiR=C8dg4=D6EV
z*qHzD0`M>Wsq%&z&a5ZBq1GQLzj0jJPjN}r3G=IFN)$FV8L|wZCIpsDg3`nRSjQCv
z5CyZd6xXCtuS(n-hM+o@Q)Ra$^~s^HXtJpeP&0lpF4Ot=jWgqW`B+#rd+-~6FhR(k
zqO-g_d237wb)h2TL+76><~f|X)?V6b)yH!7m_$xAb5%CoZ?A+aYwwezRKep)58Hct
z7l6-nYyH6>eiweZ#-({1e_)CyjCod0>qzWOR3FUu25_^-iNNh2wjRuSlYnk;n4~pk
z-e4%$yzr)v{iukYS`4tMKUshEv{SBZRd1PNpx;c>`FjK7`f-ft%+HK1ksA4Dsw)`8
zD2DyN!eiZu)e7{x)SVaqTrX)+p5%53<aZKK-g%%(`Cx(2v3!A@Y#Ld9I4#12=BG9t
z0%L1yY{a*`s?i`+^X<Kb=g4o`BRS5UP?5I9d9OpXKNhuggVFr!QPgM0dJ4H%@*siT
zSujm&-iV4pWbjf4#SzoAS6q)KFS6{tTY4~;(9s|&>pnUR)MD);sMwPGZ90R`k%`G5
zPtI~K^=rnmb=fMvHW<iT0Tw<nl_H@$9oDvegjH|dCd~L_nz8X^ETZ<-N@r&L9ct~t
zy8}AL2;p1FKYxw1g_jgxi-zXaCG&UeKd7qDjoL?*2ym5wBiItO&KAPa0y-n+pg7Jh
zna?l38SN15Pb}P?Iddnuzi5%Mfpl;1%B%~Zuu(bDxzts<J*}roFca<(7@>PxGH}ga
z`SIM?Nj#Zx9is8pUdEM08$Ru6L9;E+%#KX5P}ddQ;zmA=rH4{{YD>P#CYH#G=gJOW
zoUPtDSk{Y14}ky&OBsS$K_UM;huaMijp&g~WdZ6`{COko*f8pt7l{thC7EjW5#c2Q
zy;C1@EvQf8mG`U=)o%)GMPc#EljdO5m)t6hA))5x*NqYQTyR$?)KQ+gZ&2yp>NBLS
zX<41(5pcy6wzrKgDZYm(&+z_8x_9R}aFuuMyG^|j70PeD8}BE)Y-)co<3wYWE3X2=
z&_=YMia517l<JZ#f1VdfM&EF<r+8Y>BSL(kugI6`X|qnb6@tvl%s$wx)eSfK^{~!+
zdMh8|vJRQuetEqhThjbc-7+n~Q{i6{Qx1GdCEa{U8x31a3v^ZhsH*8YU!(XWC8vJt
zjcW|vDmcohab=U)*17;QfjDHp^gMOyFm>hnd3)tgvJ5L!uHVw?RbzB#x7WXBRJAZI
z+ISgnbDG^A4K_zA{+0p!Gwur7KQA0;ymao4F3Z0bKWkt$aICe)@-@?#?tYRwQo=y?
zm9T8)SbJ+LDZMHH1o*_}isU8|o)A*vYIMkD`hd;t8Y8~6jYPR@C3>lA&@t@goSsvA
zrWXI+9U9#BDL24<g7A*Q{i$GoUci7W?@Xye(Gy4f1aPeiQsjCe1q%ZJJK<Ck;`RKs
zV)gH^6g>d>RNmse+>!Z~jN+zIM85NfAXskaW)=%TgMwCz2h!kezbfD|C$+nnz;7eO
zy1Wiu4+;S%g?|dn+3Qy1^$*`36!gN87WDgT4-?IrZCUfB(Cu!SCgMZ^%!|OwvM0QV
zEd-zqaR;vwLhLD?{TySCzWOlyj?3l@Y;O$o2XN1q(wpOymi?mduR0zA<4ep)nN~m$
zhjo1|5gX_Cd{e=u+osx)Io3rhaubm<vRKg`^u=F}J$HAZklI`T;BZSv5Y?9X>h-F;
zU|5Y#GPC{qGWrQo^X*Js?QVp!L!W(;8`e+|W0K?m!6s-fPx0UvR6C6oczay<(w<dd
zB2LF&m=<j3;utN{xw@(5z{GU_4V^$|R++bt@S&{3MI5v5BT{AqUuRCq+^CY>i<e6(
z`c+vz)n&H^!o|>6#PtJ^y#VuJbksm9|7OVjXCU(g3<+|@^n|tlg|^AkdjYi-FTSkv
zB?yT*QesNdf*iXBLBsD-m(Z?a2!WK!{E&-vg3`Ae_+R70GU~7JZ&o|!Fyau$(57<j
zPt?G|?pv~wNWSvbOapbz-3tSZU;@8x>c8HUd4)^%mo0-T{aAgYQ8P<&^2y2zP+FNi
z^S(ND(?@Oat@1dH<T>fr&SYFD=Py9-^<4z8IyV${00G9g`@?`;V<jfA3H`KTo!`HC
z76cr;Sy*RWP*?P8k|~#^tav(m^efJfet!E3?$ihJJYu6VB^)?8|D5_cfB$lHlCr%S
z#LOdl<aT4Nyliz7K!hZDtc=~Qj5u58oypRyQbH!kZJAzc;=gq+KLX}>G<xT^4pyU2
z3h%B?H6I;P>_6L}Wi+8Mb0Bs`_%b9-z_c8Qf1hpfz25!;fc(n%<BR-(vu<1I6Ebk~
zX-TZ_Wkj67#LaI{dx^;EBFSV>h%sToT@#+KGSMZ<;=`fhNr${ZGLptUTOu-&G!B~t
z+2fHAkZ+Xls+W^y{U{R_?wGz}Rro#*kTOK2;{%q5MG1rE_nyj?WRR8J7)x-ZP~}@?
zbYok=h2J`9{ZL*&HQiJCpAlA2T8q=+o}MrKv)-=exITefG&AP#6Y!#L*&s&hG&ap-
zdVvSwUbgLiK~@W4OeD(BTxbmdlMlvz-N#bo&T5TKP9+p0t*K)iX3}W%^_<qEA`nw+
zZ2ON2Th22uC_9ARZ^^_4qtRgeY%Wahj<mAa^je*H9hhPKSU?9IiH@(sOn-~demMXo
zwA3)7YGkW`v9ExV)5fFCMT$sN!b9u_?vkbtjd@c+%4}C{_xq|FMirL{8D6Q_@tO1@
z`IorDfLRI)ch&JG_Z|l}7U*T)2Hy=TGdhVi4t+pxPwvRwbU5mzd+@5D<-AQxCS7Qo
zn`_GqOXAE_P-qWWCFsl4Z?7dA30NzA|A^91KwMWuo`2qis%^<^A|1A+(C>BWv&MY3
zJFyaO5eP7jROEkv{AF#&XVYuAG>tr3iPUNSSs-R@QyGjCsT>@Sj-MzmGkfy!n^{m+
z0UwyHvhCe^iWzqA&2%TFn)$A6v!VGEq)@#@VILwZ&_(LgS~+G}-$Uo553X#t_~wO(
zQB%)K)s=ot_!vB!YB|Dc(6yNTv`ZhLq%&iLktjkXk!9iYD_?XIv_-`BE{veeRvanp
z`>swo^(=;;gc#{~e=Rc*OS*?3pr6?+&M+TumI-SSy<0Tm0W38Yk`Mbb<9B<<x)D89
z9DbIXNH%-BoU{fh#bpul<FwEjy{8GgPMC!sp>}SM%Wh;$f@F_}h|fxhpGw##!iLnR
z8`5@N5oVyYkVmW|lCu9c>Hdck#tV62!b|3_PVLHNgZ0>&=X-_GXu!%8GVK198Erwa
z4DXZ7>QmJ67B_hJSR>Hu`QBW=U9!l6SQg-TqYJLJI&I&?*8vrc)O%T1;qyLhB!jFo
zoc^FD5b?Q!q%=EF>E$<~_0g^f5J3QOCajgu8LMY+G*Ql(Pf@zD-K$K<Uhlr{3gUAT
z)DA(=4Fcbas0?<lfcUYGXT7GT%OeV0ma$l1&l8E-kcUQ77?baLU74c_z$DefxNI~?
zStem5K_kF};t>MCJ!W9OtAwrjJi3^dgS`V=+y)(;tAx(eP64Z5z`dM4bp+G~Uu5c=
zY^Ae^RC9Zjyj(*~FVyG@6*m=_+Ls0TaVCJFHCMgo?D9}HroSLK((Qx@o!=aQhMc<-
zQx&is+|rM^{?%K>_xd^q@A`@DJ-zz69c?KX<M5qcnn>$M5^L>i?IOF3<Vke`a9D8d
zUNAj_D3V#4;6HA1a4UvcAKp;MpQ3ozmy9we7#~NXG8C~p*|Slf-O_}C{I91USZi-X
zFk^ae<}cC@hrr${VWjPO+lJ%4!%u0CS8t3b@BXY9mGvMnL{n+?UhmaRV*vAk9~n;)
zI~ISc{W0tMr=e)$B{1J^QZu{3^FO8XZ7wfp=+AfV&HQIzq4uq%m+bXz5u?c(!mwLJ
ztm`Y9BqTn!iLkw**ORlk52E5j>WHQw5orN90NImScy$5%&+WYg;m=Y&0iNe@-YZiN
z79(?PIzeVvEP82R-Mm-HRP!#A^EadF(tUq+x}Uy8--?sB3y1@L>g|;d00y^-vJp`z
z)Jwo%HRfR6$*mgh<YOI>4ijNsxj`Ue0H=`%4~(4x05d1!%d$$qVZH0a9f0F^>1tuf
zGvSqOPPICFma%gi`!R$k_G)PJ;b!Wljf+}jV1%PkIu~y-MW1|F+Mo`bB5QpGe5Oex
z;LC${e2=)vTZ5eiFs5nEc*`cL7aVKc%T3g}&_UBimukJ^mD8?D-YA!vgnwCiOsXvf
z^JTwnnX5f<3Q-3)%|6zluJC(l6=+YmJEY^FuIC(2Qjp|m0LAIOU*BFXF3;1T`GF<}
zws3IgEj;TUaDI`z|Jq-2iP#cG3k`v}ReN`tXX?I8fm$Xe8;rFwZn=R0YcFU`e}b_x
z1MV-!gO*gD6<Rv(V2TaNx!zZ@+EcyRMd@A)5bdryL1Qi{hUglY&dx`l4BC<P$#jYV
z5I)r%&O3M24p`>h3H+WVR$y$?*h`ZHYcuS=GvoZZzlWY&zMS``d+*$qzf-lH?#(3q
zCf7tR3cq=~mF<-&=b<q><mZ0-O-V7_Fb5<zDy<`<V`K68<()Txx605Z{!{JcT9*Dc
z97du>AOmM!lqM+#m6R``VDc@&r0=tWcJ80KTuxQODDb8>^0`v#rYAwU%EWk-bvykN
z{palvcaXr=Xpd)jqbuLYa}DrM$NDs1)0LwS$1~=&URuD|rXnum-0qSuUpUzAFav&A
zA-6sN^{uQ}7CpH6&yK$HkCMcIzk)tmK{AOW*(gP<;dc1^kolKyRJ$4?|8G0q7*Tt@
z*=g;Qzi)c$IM22|sdr@7-e&qLK%E<l3cF2R8cVF!Uhyp!a9fF=wKq7=8C|pAc&r}k
zCO!_v)7Q2ZZ*_<^#FFnOkaIQdAbj|!aR;H~1w?kB^Hu}0+6r6KfKC}euZv+<Qz`6i
z-;@Y^73}c6JBL>`H|F<dXJfi4A-JeLQ?R3_KHnTUlJ+Lib?K9$?OgfMM9IOc)DUCj
zyU!)eZD>qqBT{2TJ|=>lNg-^@&0IMFkX=|j5dF8PCF6({+eNPrfg`{iVQE>KL>gO`
zd27qd+CJ}5?l#VPa~1L9_J;E{y$w2Zc;f^)FKa4#;)37NW<thpbCRBoyZ7~MH_Qfo
ziEoB%auQG*#iCkoM`I&6wPvPyM?kJY`+1(Z*Fl0^nPdfDaw;(+3EJ=dzA@U?Q_M-G
z=GwM81BL+aUO{YHpvk+<6eZQrJ{-%}u9K;k?t!|9Zch-<zO4Y-;^$VJQm=F07||S%
z-^H#fc3m<4%Hl^d6}j(<rRQpOh%O)47{-;nP_qAET3N9AQ<P0sw9#>dE(?n5n3EhW
zoS#d4+-n0uFXDnt_QN?+Uv;I}cIBU_6=j_v99xRA9TscjO|?gi<4eN%F-VN)5;y<r
z4lB?QLqt|!72#x0{UMfEHLW9E#B|v-Cs(Gm?z|6VJcj{7aLU<7q29aEj$JZUyMX;%
z?`AqsEVlYSI6kf8Bfai>hK-o%EKI*j687US9g+SkcW0;6qvY=CiA?ATu!^nNTl>#R
zem5PMm}6{c7sd7a);m^HD`OZTusNO7Ofp7))9;2EM<dXGQ%!$Rtd_k{Fp79im+!`{
zC)Vt$%T-boT>8dUb^$zegV5;SUPAB-_Xl&)%#a<k9~}(^bFX(7<W-RA{_dVAySvF5
zV^IF0>9SzX;Fw~P$dC@cMIYZGJ_5fW3_Zgd@=8r$Ay{QCiw7Ydf})%@eGhEUUep2q
za>hdWqvi_J<WJ)wns%e`ic<<{dY*B+@X81eOGAX+o5FL?MPf|lIRtRKcn7;RY7&5W
z>*4SNRM2W^_|Flkq!MZ%6ur9w+#k(E$aTvtu$26Vt{{NQ0!zW)+FsEqdzK#SU%F&U
zS(ZA!XtXmSksOHbJre>H8P-oC^NTS*t5Pd$`3&Amxn(eD{6i7)2TKKfYatN;dllb*
z#{CPy;Z423%lsw<&z2ak5$GX{Wnfc!IY!~XGGU|=7rnpERx{r*k|4+hdtRqs`1zVk
zuNkGS1Be52Xkc=>AT852r=01-2+#(bVguP7O9bLz1n}2wHKHna@pWC8+HM>ezzzU;
z<rcoLl&>A_*1-%DGJt?$9FEJZw_EcX2Td?b;R7$Qq`oDe9N+B@=t%tacb=py>bY4S
z%*K%S;CSy#otM@Nuo?McND{=VWN)n?rOVF}%Gy_(4O8R>tT{!q*O!|pEWUzyr=Q!L
z=jvf!8ywwy)M_GIj{R_}p7Vg-^3`;RKxabUS2aeYV$LWS(kpJj3&ouy@M>??(~kIi
z@)Agdp*mVK>o4WuO_Ae!5t+frSdP~}U6P_N1C$(JUdp`3PqK!uL|Ch=!8E|}dDBl6
z9>2o&=mDNlgFH{goh8#El&Vtl@eI4vOz&#BM3@Pj8CTH<bzD+ZW+0JJ$Mm}WgHJ@9
zFY5g5WXuSM+>%f#gS34VAMRMHr(kB>(J1IG=k!ieE9zxU*+-1^0x~(I_E@g+>r!>i
z3VtrUTVNmX^}Z9&2h$=fydBghX;|iU+1#042S4XfenR9-GMd?0TKX~BV}N3QdC0)s
zC<VAuL;?1rw~aQ|bH5HWysa<U(*>-)*6_u8EJvgge0Z1HinMTbY%A4DcBYkoO6U-W
z=<vx;cas{Uk_ssuxzTh|cINl8`|t2@C6c}s90^5DHy<r)sj(YAbd%}6FZ$6cBZ>8v
zy>%1(&f2y@N8yn2ctCj_Gcl>zJjHe|A@ll(<hrbO;KHq<y?}e)i^=(J4}kZior35s
zQ>#z$1G-fPD+VTpsAQp$FWJHmb-Xc@lB>YOGLVDTYxf<0pg8g+NRs48M&r6<B8}<(
zLL2{t9shn&aT#>&P8!1Me{}8VaR~gcUVEOKh+|NmGvPqkb&EupoDxPuL5NpLBn4|^
zb~ffMOl0=khq2n`&<o~p(Og3k6x%a*I(qjX>@yz#G3v7Gh1H+U4l{Aa$CEf0LU|48
z+|%jEEi3S|y%js}aa~=X54#2L)WbEaMO`wREX^k@)SHUcn`aIXC&GLx-?FVIS#R;m
z_hD$GpLkq&v$f2b%oe6u#(Uqz;CqxlpNe}!sn2R;KRt|BWgO#cnN}Y+%~nz0l22n@
zwj6w9S20bSy>^!*o*?^X%!4vi;mu6PdD7Ax4t%agi9^?K>IcHv68G;vQK>3sOzPWE
za*WEEMHjzNCEI;B+)gi7wD~pdTg$4tl6n!_L5sszo#?@vj`HZf7M;%Y<iaYtV2R#t
z#5eCgP^GMr;9B8t04|QC^P`(Z)CVKNF}>a<C8XvtOh4Zx&w@@rSK3B)C_ibK-T7x`
z%6mas(}A(s!oX>f@2yGp_GMkRl4XLYXA!Qa>H~J7?5iW>!;RR3<&UEirR}!|wqClW
zvEbQ7EYmCtoy`{B66><73b79cUQ)tIx^Di$%L~kozN8WJMjVB5E7mcxE8k-BIUB$H
z*M=&#2lyTdrfGX_{&w2=a~-4wh_k~dI9gx76(+%7A8d^JPV^3@OqrR+;5W?qkRtXv
z{t^sVDbi=@`k#N~A1_#6gZ+^;6@(f5Z@P$=9>fnJ{&;nbKEr4fn7SI9+hpji&NbLH
zrlrw`X+R|4{<rF-|BcORa0pm(HitF?y8rf5uGgyj-Y(}<RYE1}E+pl8^bsOO>OiEA
zVc;N&xn=b#Om(?{PjUj}&3p$m`C8Ds=z{!^hQ|BZQ6X2~RRc!SbI8M8q31YZ{J0T{
zurS>NtlV0k3x&Y>vNOvx>TlYOe>fxnwGQa+apnB~ZFHX@hLeF31qXm72tDE3xydOB
zlZpl9%_@laC;+@*p~4;O#%rD-(8kCQ?!S%dxx!@{^%8eL$5;iU^IRx~B_d~J4me6V
z`n`Dc-+4GV6dcgwtRVgGP#tS%HJ~tBo`J-tr$2Z8MFk$n*S`ndu0g@2YT_k`paw%y
zZ-I<mdmpFjDdcYhvj00h{qrm5z;7JI@E_@SfBxJ*UI3s7kn=bYB5b<<tr7gUck}Q6
z3;uD^|NkHVui)JOedqu2ZI`9<f<*Fa4-0{ZY)E1H%?j42Y9Qybg6Qf{%reAFeAE5t
z#uMnF=XE4{<KyUbE!}V>D$DK<(ElH{16~biFre{J?clHH#Y<J(;CO#y7eqM?gOtBd
zs^L%B@)*3P!^FlI-L`u&q@WBjE1(gw#0Y%j?*y$el&`eviAR<Tm?x;Gqy6q1!TR-X
z6opiE5Wf%%do65Z`oWIS$m|91;UVa8Zyo{+U_*#IM?1`F_7Urda{W;zlfeonZP+&b
zuggsq7vK(ZlAIi*fBrtCxDc~fbbrFT@Rx@6-CZ-26__})?{89eVn&?g`<<RJ_r5vs
zD4V&C=Q)3eg0uiDjOk9kBM=l11G^Iw2q!ikSGnmBO^6$F;W7}-gBWWdof=dr%3j^?
z`=l;~-UmT^+*8MZ$?*b8@d3g)duTTpaAcC{yXPgm`gyNxsU7T7K=NeBp7`pmC`f&R
zhh```gI*QMV%yjAlKM7(1!goN+Rw=K@QJ2=5z|v)&Hwylr7(dlcUMY&{!dW}$vrQ(
z-JvFcuH%$}94r45z0#Tj_I0dN)$s51H}HuanK^F#POYlktLOqUYuq~fM9I6jKyvZA
z;&`(y<USBM3k`#<R1(%ZIArn>yBRBJn+nZkHsp@@zSRP@Bo7L3cvi*E4wOeW!A6_G
z;J!b(3s|;bp?tge!6vzM^)u{;hD-{#7sUGB1vPyO6ce1bdQiu7Y5*8iPQ&(@d!3(G
z9feMMzl?Hc7+*{O5S!%Q|7&2?Kf06hZ?*bgi|+Rq7yQA|bkhB^<j+sGPRZbCwWtX2
zA_v_8Xh(Kb@ygpwc|x=5@G(&)6oaMoJwXg~ij5~Q3(L~QjNX;X+pQsEyM;wVVCvWR
z9hQsTbOt$ClHmeSN+n>?^e_6IfEcmQZgaPjBVabd@5DzP?<VGd7!H-h1^V6(tAJ_G
z`jkkk5o2AZTj_qSpbV@@6GX+!FFC_EITQcO^thV`EX*UejIBS<KEHRK1akhAYa~}o
z@FnZCS57WIY&3TP?_;?m|DzS-U)g@GMWM<bxVKZmp{dR^%BiC7^eQG<?SCKqf4&*u
zf;A30yh~n1@&D`aUWlybao->|c5MYgq%nLuOU1U+mcaNsPmzD99wIjngEPYvM8X+1
zX`LMoRAIm(wu1QGD+5)hKLA(xB$*o$<AWH5K`>l425VLs*w=$GDWs919|o4|s}4CR
z-Sp`X{*3t`k`4p$*&!7h-~%KFy^3O;)Qi!v3;Y<h0un@u0Itvo1#P`k4qgSr>UQXm
zE^ONDTzm(abE0P=<W@l5;)jzNcCiac=;^lvI}M1Z`*QVQAKcwb$TA(CZca_mA;*{f
znP2sdMYu!8ayb|&rde+7<N&V#D=4{gpnUGGi{`M&<FeX}B=<wJBM_Kinfcx>$S?t9
zyrG_A?}F2u)_YUxBjM3|K*M*Ys{cL-9t#0&(9u~1vPI{DOuW{yjZmk8dR+&(h#Fuh
z%dq2pr~OW^LSYR7-5_wPAIchl1W6D9mLEtwFaTs?WITKhG(g3oB#@j*_NpZ~?^Sc`
zF}dHNJA*dNdyclFz&$34oV;WO_)xxZ(E)I;?A0rHJLbP0YjR`FpPT`Pv?0I_byoDj
zw~Rk89IaK20~}N%t%9W!8Ysusx`HUUT(C)&vV|($BcQ~+MJYVH1!^X_VSmj!;KX~U
zTr>Me|I_HUyk58E!7~sQmS(73(4^pU(f?!g(^f|S<>k685p_i=8HS@pfCYjhNU`vX
zQNhYFG|YK`ea}ZzYl(hLq+}29rpI>s9Wayvv3ds_FS3`mjKk4adR6=tuT0ZDOPp_l
z#)@6tU~}Zkm`nsJfYfiZLnWQU%hK9fp=Dqk)%XBSt@}z<q|0L~VSEsDdDL#Sn1sg~
zPZ~0O0kD56&Zk^aINV@CVCJ;!?Dtqvs<JXRTH^4t5aVGZ7|w>R0I5o-<HdU)E4v3y
zfH$Q9Q0mJMyX%g+>Hbvzku)y?w6=d=|NrY6oO0k<a%@!-0vZvpmBiLufFijZ!4O~=
zgwhxgB|rgNv8kINKJ5p@(}&wh;ycul3KX#IfCV$+2>cIVO7NGGeLlQF64nG0P=apk
zp?z8izj?_i*avPC1YG5k6R$sH0!0}N#v*<#6q=US--QakivNeRw+yRt?Yck(Q9-3a
z6r>vkMFHs&1XP*@C|yb|LZrL9q(MTun?<LDbazU3cb@xk$G7*l-~E2)oa@>@xVDSM
z^V~V-m}86y{QQi4Tv>=gt&?2n6fm+$S5!f`8r{`Ogc=V?da6bXzs8-=(E#8$x8UMv
znh%8usA@6Dasg$y{|6~WHK6-Wk1^+b3H%z6$re19Q7D8`4z^}98OIJ;oGmCpG|+&i
zUx-|3g_0~t145qLaJh5O!z&{;694BQKt!iu^ZS5g($9ObA~}9kdu9Lf2h_j29n0bp
zu(Q2vs}>lIy>~zOC=1z`yF;~<#&XbOi8_(j@Rr8}QT;l&uMNNr7!P$SB(GW+0S(9S
zWMN-00>~Te0X&gpe+>0Sdz4r8iP9T2082PXr0o_LxPm?swRpZ?fl0}HwC=e{=GB2@
zN(%oDI_(8pmORGhq|p5^W@`0?WULhgC_HU<bBh%QB*!|zfpY+iHY!H}AK+=SXlTp<
zpi9kL05XXO>M3;Hc**Gzi4h%;7PNu^(*sAz2PGu)H#cuh)L&o$F$b5P3)%`lo_Dnl
z_K9W9UaY9(2^(zJS(jLYL6C5D031K0+*k>e$Eh7y<N5Os>Id2vq#{(rx9srtU<)VG
z)TA8fFgQhvtd7_jBy<>o4oqDGY-KIq5$Xdi@KS*m{!2m}0~!h(5^=->n7sOROxW2C
z$KJGYUTk5UZ<Uz8%3dGtCH-;I+O9gjVaMApFgD2<rvd8^%F7!|`?0hA_1o4fB5^Kv
zmF;t(w-Tfe$zN<xlg{G3jJ2QST(xF~zyBZCmVaD=_g|u%-}oQ<g*rOgnd=L!O^&U%
zNJH*~eZf>to4(kzY`tOCdaJP?#7@D=0}&HNYU5uaW0e(15Af-m^lm+{bJ=LUVZGDB
zOt3|L6c?u45e~fbQ^VPn_ap3_o?8>pcY>I#m1mynXf{N7BG~RJUpYbOKF|<wbOOe_
z9`HjnDv#T*e331>2AY5s%wNwckQi%sgi2=bbGx}xEqOn2PMAsMpZ*Zx{Q}OFxOvU!
zc~y+0w?CGElsa6w2dir0x>ikgL3T7IMW#Nkf1{kTsm8lc^+HJx?XU6dZfKT}Yvt-I
zm^3=1<Bp`>sUcjU8^b?X0Q$L&ND=~aYJw<gg3A=B_+5U~p0BrxdQ=9oJ5Z|03039}
zo8WraVs(?_nk$iU6?2~?F)wIs{pF5KVt{PCiqn0>d>jj<IFtGtCX2V5#CfKb$~V7W
z_XM32lT7{a2`*XNe(<o;5OA^)lj?sLZfI@#^fLBd_9`8w5z3k7g9eRsma-A1`+#<x
zn8dA@2fC&9?*o2tnRM!a!clG{R7;zu=$((_^6Vg@EmP0?IKFreqGH~*&5-zW?YYb1
zh~n{)`zNMbAHCTMV(f7g!uUBSd<{N*Q~6~2saRoqfn!EB?~AROzg=&0=^fbaPu1dv
z@JU(tCuUefc)sbTc#IgQvJgo(DFKsvY<Rzv0#E$Z$&ZuF)AMw2M0osmM0`fFEFOiH
zLXu<98Sa22K~5^T0OooxHHsFbJXNr1*+d;Z*bcs$b)i)z8w<ZL%ey0+w6)N%noc<|
zUGb9m!r4M^wPcb>l}ofzspWaLKH=^G><o{CgkYGKz=8Ivhdj=}q@HPcx;HJ!afxEr
z&PEG_su8Gb=}Z1@dcyx6uByMqo-wfaGxv3tmbli7iX0D43)*teFX0nUkC#=KzSv6W
z1J9lK*DNS@yu@0c@-!Hi$nZWrc~8z_2V?*-W8P&GS&GT^)ne2~WRdq9pLhWI61^Mi
zre1zorOssaOX~Q5A|1ykT(1Oqv8p>k4Bndd_-+9|7ZT=zDJWK<<XH&T0W#(wjm=M<
z9~>7xDXaO9+T2b}!C{K=Wq!UN?n%~OtwM7Y8i5sxf887dSar|88F?X?%js(C>a$Pz
zczdOqNIKqbx3wUi+B_7POI+6)QJ-tP$|ZXZr$u=jX9?VcFk-NtPN(Tx&s+JlI|ws^
z2S0PUp{XMDaKYyFnfuHSo&onB&x3D*e25KM56>S#hZs9jCzqj=iA$7$+|N)xp%!~S
zI>!4!j%Q4(DTv}5bT2qg^4r80w~Ac>!hCU*IbotDk7SL$WHHmP6(jU=1zEvgFn6WE
z8TOGEiv&{>H@59#&;|D+?1QlD2iLLhvCZ}GhGV`!WQ~Q{BEQ|Ve#-lKJHwi2r%;e^
z0#^^!0=w@^xJ6Yo6QOz%xREQz563^qdne_7R19O8H&|PiEjy}p<LO}wyWU{C9)M|9
z;m%+`S+Qzey1x?n@2P9uGSCptEHiSYDF41D2vM*`8T0FqfA($xwT<Xqq44X1DHd^N
zSeHM|Z%eA*HL|#k3hM$?3M{mr-_joL&a46t4<$F5lL?S-C4)K&t1&_agOi<8X7^|R
zbI<gf!GfhhJm}=-bRHGE(oHlzD;c)cqy>0Kx!-d$QQbtZ$E88KY^)Cjlk{ZCP|p_I
zL$YD|xZTUgiQ6lJF&HX<O+n5}KOQ9Mc)wvPHN^g<VBHnA0IJv0cMJIJRo82Bm%WU*
zejv<Lo|v?#CLQMmjb!_XrjMJqX$IK}uQvDfXhI~T#(QH7N)2pzEZXSPsCJk8MLZ*f
zQ6!q2u)tNlsqZE3+p^krtKmWXDQmA$$sJS`?9`t(vSia`x<y{|X7ci_wu#`!EX(o2
zjwrF3xZ)%yi5uR<g?VjlTTOteebiX%buEKnqtWXm&g*+iASvHt!Mcjm>l`^v3;S5e
z^7bD1{qe(Rog&2cK6b|8Ud=mL-XeA9!5u&vIDPm)xyq=ghRJjQ*6q9U-&jC(BjitU
z-|Y}A{}0lf6nQOh57Nol)s0PXMV%CDFZPE<OD-5NQ7|9ne+1r(&cEoUf+?wAr$8Ej
z0KQ1$P_HLml@I*|yUmb|4}r3YGci=li_470s8poSy@KL}O`WV2#i?z2(z*Af?46bG
z!vR~N`lQ%#kpxvKz<it;%&#gzXOr;Gl3-GRm_>pA$&6{J^iI{US>~jv@VL$!8+Y>A
zXYu;tB84xxR`&N?yHz#YC9Y|`P3#^!!V&VVcii<i@)!FB2`x^-^6XNq(sg1hmRzv*
z=yf`I%1R!Re!g`$AIH|2Mix+J#tmt%)^X#H-OAHh`XY~H`F^gGEbQhgP5}-1qO4-H
zB)TE5xk-OPRqq?(&ZWf^$x(f-jpC5l&wXOn3d5i-xr}F8tGAue)9K#bJ5*MeEn94g
z0bhY?LcA%KEqm-ef$%&~o>|n?Fx&EHgnv4r2`!S;)7vEOR2Eg?Rkj|#A-^bSvPVDD
znI;&p<SEH&?EARTtDL<XU^zKf0eIsXcKsE*&cu?joP%~gix17)`>J!B_T~Ty4$jdW
zz+k~DvnzdYBCWuiIU+RP@W3SPj#MYlWt1c)@!=N9NMcQy*AU3GOhNG6-jQbaBEF(}
zgE(^piV%%A#YQU*^mfIe0>AFwBxORmNJ~}aZecDs3EcMo*GbSrPy_U-9eh5_1lL~G
z#vEIRKL+B{MbfEA-?C2OtP=!dyTDn6kemXKu~?|Ze?j*A&>*<jTT_nacQ~+4Fn?C|
z3nU=@c=z)&vQC6;5AzX&7Svz3yoKQmg#JBl?bSY`kTbdKsuzF7jYb__;`d&f80<4n
zsj_SLm8ly9Zi#>`HN#~zQHBL)MOrmY%C$4$a+7c!5egI|UM_ZMl$l7tsLZ6mav^o0
z>jN>dqs<TQJkV3yL-7XE8<lTN*zRslSC6pWoV?gTI8AzC{Lz!u%?YB1LN0V`Ce2sV
zlYwx`o>aI+x0}k`e1feNdQ2kR5eOrn&Y!q`ol)&vkElIQ(V&D>r7x9zq@q|T$gW0O
zct31X=L#1qf*C?Zd~(Pd*bDv=nAsgzv5V9%0H=r$8BCGHg@b%jpSuM}bIl=KyQ8|+
zefc)|J%qQYMVXtE-M&-Z3CL8?$qh@-%hFz(1-A5VWy*pi(lOzy0~*q;GhKIHXUPXp
zlQzZuEZ}ZBh&6k*ztYQhn6cPdeH}9kFlTL45qIGK#RS980h*V~XL1<7&r5>$wM6Xr
z=C17uPtRuvTFE#=UX^sls_h0gy*?_fYwVto3>&M7Y<HVpYHu_54L3%(u&X&LqHHBW
zunt^HD9=%4Qr|}Pg=wU@nlB{XcdTWTu0m`cKyv{SL@*bLOw}L@DFA88p2vt*dOnaJ
zd6ZT|bmNz?V|%7(NS?0-^U~efy3@X(|HKe1l|bsfrDFqB2yS%g{nqPSEr^|SsN?S_
z$VI0Z+fucPr*#5z#cAax>dvU)H@eX;v2f-#Rv{|z&!~0a?x5+rR8q0aT<z{eNlgli
zCpBN7jOim|3(m_a6cVil-U4nB8jVg?I!#10s`Mvh&Yyphhg5JV3-X_eTfJej)r3dn
zfw<ziTZ1{ULIfA4gkvrILrrekI%_~hoJN!F(t&D!P&bJ9`SAk8zXH3gpL(_d9E+%x
z$?)d>=kPSiNdpLh^X#NlpFH3$>isV)Ds?}sZGx4*oXKD7z-{yu+$$9U<t9%YR(=Wd
zyyAc{f-!6C5y<d(PFQSPg`HY&F~<ZP6zu>dGSwCdb^NQmxnB@w^0zrO2X!ENf5RJQ
zS}Hq6LPURJQm^x=Uhvb-yYbH$B}KV*as}Ss`V@irSIS+koeXDNAHXa9RIf16pLTO|
zC2k=^nI#-@2ooCpe1|tjZ{!FPog1Yc*mt<!33ebLWl^yOsbPylJYkvM6=0{kvjAj`
zZ9{7g@;csAu^amIcx#)!)-yZlWTNWx6O~pTx5ZBK2v~d4B>JhitL^NY=bbFFF0eFy
z+hvk^mng}Dk(J`+O9#U<h{y>^Ek6ai`LCt5_(-ZnfvhZk9=Q%tYAIm-iCKL&qa;KU
zcjA+T$_Z$t#n=aOzG)ms(_2-_eu@)=#i~uYuYS98e~7?hbQwaCUd+tM3eX-kM@)Rz
zA6K;U=vgu$S0A5S&(>7nEwTAz3e;NbT+Q7`M`d3pUMeLwxjU(JZyc$sfGGKBqqH*{
zM?|GNpF!`(u>6W_6>ykutuCZF!uYzZV0oJx^Fk|1!c*%i!Cr+aH!03i{42(vNbNt5
zMn8RIJL$YXnb}X<(8`9R=les<@psET3-71DB9q4eFsIdBqtk(ig_q*vcYjoVOuv6@
z#5qF*yQ$!7`b!kbQ}wwMsCQ;hMSrwF%E)=yM(U&W&<O)+W^3`a%}zREfY7({RI+S*
zObW1*EAb8&8Z1r%K~pJoiDuU1+c)OBqPW?61yy)AW1(gP(SK|8Et>ekPeJ3Kt5TV;
zg+Ew>-nNKpf`$elLopt4cLEcqGJA$!qD@Y=upk>mb96p}$GBj<U4P3{)F!X`@HOhJ
z$L)&|gk){ZV!Y$!ylg!UipV5#)h1{gZ<Kb|?yl)d%+!8ruHdIJ=J;?{K#3$83=3p%
z$@!Z4=fx(%n>i-L6v)sH3`;1>mwpCxO9?50RXm=58);<=5~c)$+qNu8%L0pWH~2<k
zV}-lZL@^Z1HXi-;3fnU?2?~^_Q)?ToPF`GoAI{5Y@*W^Uo_NUeD{Dj2-AXvQ#D9a3
z{+y3rKWKRXbX6;rE)4q@y6XHO>RRHxswFZ-dfQyV85dfXc)=(BL@$`cV>wH13<B?x
zB`h3@IuPAl;JEl`);;R9tUa6D2gnQiP4*|{^Iv9v?ziw=GBY2yUZd9(oPBhwi(|X)
zdOu_=(Ma}zcFR?{n**w7+JHPdI<dZe3iMq3Is|;0J_G4EE3uXq)%^Fg{<om<2P$)d
zBs`J?An~c6Db&k=r+Pos4;;(?Cw}U~%thB@Fn%23u7+oL*3v_vmY3f{&@Eij(y9Wf
z52Z28NAW3`gcRN|mdG_2s7WVy!zocmfbh7Vb1hp|LeP+MOekOMLcngezlp8udjPO<
z_>+`Lp8BYHn%%Sxe=~ON-W0d5Dq4{Ue^nd)mbZFuS|$87ZE4FIQR0)vk5<-{L6;Ds
zvJ&(XLu{{$%h;;*<$z#RaaPhmt7dsYK%~)9tfkSY+FOH(_1de<a1mm?1P4Pvg7}G!
zx#f-eQNt<H`0C<F@=;E)462h|=I;J_QTM$N(lr>IC^%S>_`%dmy*s&pb$hS-WATEQ
z(-cU7MeL2lgwVNkhcg1LSSkU`_8uj;xExj9s^)lqgwYZp5QZp3fR^b`fu2nOk0RQ=
ztHo#Got6Vc{RPy->*<;bzuge9J*1XQHi3t*>$3)@NZe0+W;v_yXD=doog@h!BG&rz
zJ?!zuvntkaKM9#wBHpqR;RJOb+ymZQagw~$CW2BeMXolDdnpsftxK2{*-3m-A~77g
z56yYM-xPoOL9IkmfT##m=aQ)-l%Gw4m=oQZ9X<q_uiop%5^^er?Gxyha+WB-PAkj@
z_Bw?Ai;~B=Kt!FUaQ;coUqt52NCX9ul9>9uqHx~HloOVKMc3iQ22wL-@ACX^<sTZG
zO^pD^kljN%44<f&`9ASHyqvws?)CG%`nP6V>Wi-1i;3iW^~!to5>Rm8*+joM8|CGC
z1_8{bPY6UuAr5JavYh6oFhiF$AeKU;INvfwQa*>OlWP?lz$t}sWv1-Taz`~nh?{qY
za0S4kjs}|??Gv&JjccyEh>mCN#mam*yFafVE%xJ&L8Tir)5!9=HoNT+Whrg2!g2gQ
zSJpKvE8c5qfAD~Xlp_wbC2C<^?h}SttY)OKsYski2Qa-wSL(cFfBTggbLh!_QbCqc
z<)@v@`f*{R!^6^A9>X8PH>6+KCGZ;BMyOaY$JcPuX6UV4Kegx+qmA<2p}<(ZwIe0L
z#N@SR|6axHXJk2Q-pUvQQR9;A38A^~=`y(cXlKv6ENislj~i(X!4Vihy+ZjAN^Ov%
z(?2rRpcDCCzYJun<53(1Qc0tM93UGTo0&m|QC=}6ppB>4wV8(=Hdd3ct-oklA;IcQ
zBfw+=^ss{=rEUP1XV2y~pC&9wD-haMEafAu(&?E7sx1lz=!!rcu#|vHT1T}rU>d^j
z!SyJAUa4egj&m^WKGOqz;hv1~jCCnLQO<{tP{X}__dA+m7X{bt^x$uH#m_cKbUKA4
zgB8X;-tRf&16)aIwelkX!`)+3tFaEVQ}^XIOMl4?_~OM^OvgqOi=c>5>Wxd=T+|_)
z)*D_iTwx`_Dsj5imSEAe@cUNSMe^h{bg{}riNoDzRW`)mxV?dYC}7*P*Po{UVvNl&
zf^D+;7F&se`DmE$oecorvk@e`Qj)CoaX{Bg2#}HIey$v;!|QG`l4^^>MxZwi8WE`6
z5a=&~on9Y#IcQP&`k^_`_snFZNb>`F;D4gOM?d!+z-SN5f^|o3vb`ZaF^fO@dI7XJ
z`H&gLI?_1s(t>TQ$>%JI-%LgSJ7wjsAgVL7Fyt-<``YIB3$y@+o^{ndk&-He4Ve&|
z1dEaUQj2u<3uKhENYW{&-7A*dSc-&NY^N`H29E6eiCAuA%H905*LS>vuu7P><F32z
z@eb;G(e3gI{elpg$t2W+9%_)eG(4Pel}LUB#4yS6^+O<5dnEMQ%kQ3|x--_&<I;&w
zrofJZci|~WbK`s=N{Iy^8n=)!FZ@*1q#%V#XXUpz%$}Gps%o{p^V$&D3#QEaPV-6G
zm8o?P$lG<i=sj-h#AJ-;-e);lfxZ0f=JdVV&wUx~CpM5w;e?9ZD+}+BX}_KeGP;j1
z$%HTXr9A7xZCrAO9+XXkz_s7&3os+!X)f?LSp!|Vy$(RDtojLGm25t6_efignB-ZK
zD|=tR#4JD_CaVy#bC2u=blJ=4v#BG(MBY{6Z^|O{B1duemm1Mto||>MA^_{9l*gcF
zH~kU%J5(=JbZja|FETY?SfyL4TZt7*J`X3z9I;^LUB?Qhb>VNML@mW`qSzD)5*Z~6
z5O!h#wX=Iqa&%(3`(KrNHs%`$NzVHp-U+^QWNvIvUh$z?_C5>gfY|1{yE>KiG?RB8
z#;wn>xUqsd0lYY^X-$!IAommZZ)?+a@jAH6gg$UCf4M~7=md75Njaa4Q+qMLVdeus
zUuYcCF0;fvX*$f5s1%$C>)!=wgD`Q*WLj6x;ujL3%jB>+v$#u4&9MqWhs*tOzPU^D
z;wac``q~3|YIKwIEJ=Y26UlCJvZV(dhZL9JZe3jCwbBq%asVYq({Zf37W#Ra`{v%2
zuX&bi{=_-|x(0vy*||Urti*zMxoy8S{;IqC%E?)u8)?_GrzMTR(CU_uy8CgUz`iQ7
zYsyt%b&*qP68qH=(FB?l4w<OWP5RAyF8GgBi}`~Rn|yCrMk6=*q?Do1$&=w(K0!Gl
zM+>_72roj=xBKd%tJbXI<5ry67RN{9&Fa?Y@sbs`@*hCm1X%W*Rc}3F_TznOLs(2l
z6M9YfYF|UC`v-waUSM&?a&#3L!(c?Nbj*Hzm3;715oybTn`pp3;zE|F7@qgp(?M5x
z0P{d)L+m0FB9veGC@!82zU7k>iDC;jmKAY)Qx%j*im;Rn?hJ)1Xi_-)FM6)S1mBH*
zF{}*Uk&0(Y&6G6%5l$_!`ojv*Lhh*A9n?dH&+`P$uxM$(ljVu_^VKAu#W%dj>8rBE
z<t(*>4hL%HAGa8zlUhrY4+q9GV>657^(?sCbYIG&M#AF_5tMTynee6fD8ttlDHIlj
zvhe=04_owRYTO5pN8QKBi9``O9}PB^MTu~5556T`1fi!I$i^2UvbCbKc5h_-2y9a|
z2>Y<XFgP50+S_*8`<4&Kwa}1aaClw*Fw#)Q;q9J)1HJcpBW+lA?dsjiz9JS8?Yv!s
zLV=Q(Tg>m&l#9bt*bo))L;Hlg@uP;s@7Puy9~meX!9^$9ZPI#3OOq|pHfYsyj#NkW
zEQsOVYOT52bDcG7h-cln1PCmrYba#bkWl%rA^+{ix?|yM<TJ;tar-y__7A$A{EX!4
zT&1R<#mc|@>YraY8GcQ$&WMYJ<?i1;yL=ZJ6^YB0!Cdf<znN)v&FT*?fPX&9Z{PQs
z8p+K`MJ)c~%RheW*C*%PoMhO4dCY-+MRkBHQxe%_3O1TwcO1}>8y;`}gv&$+hIs5?
z{o@t?e9k{!B$WRexjGtF2J+t@cM3%i)nccK;1^e6Okrg(OJU-0V|*BLoQu&-^qw`o
zy!E%=_>bel;|~TSj!m<e;BTMR^{E5?ntH$;z&ys5HcA_3ZL-obX71sC{Ej~!^w(tZ
zOOiWb<)``KZMgm8P5$?6{^J99Sp2Ap*W6$J?Hm8`ssHpf&T)cZlq4U}kD>kBRQ;!4
z`|Y!IWZ;pn-S+%XEAIdJ>pbqz_&t*;#{Jv){omsRzE=vGSR_0Qfxj)C|5xvBa~mu>
z9-U<1wf6t{p%a|IBO}v2bor03w)_wolY86A9m}vMK_GVTn!fvg{226tPmyr=QQ46&
zI8{9U=QH<fQ2|4Mn9c5Z`*j*9To?iZ_X<Pp=!zXO-}$yHM6~|}OhL8l7yuT-$Mn)l
zYV6P6UtM0n!JS{KT5V(Q@RBB<_94j14}p5QQS}MHQ!Rdwd%jgiS!P19p>YOfi<bEp
z{pTA|)d_?DK429DV3Z~exOLJ0I!c5zWILlh-(EtNPesy6uSQ1-v~ncldC`ZTf1B6_
zq1Pzze!xwV9$ecCm6g!D_0YsgPbtq1<XH&d2Q=!1O}P$4IfV`vr=!5u(GaBX<={z0
z4nb9O>_;%=gAg067;ek+otAO%e~hk<9IrkC(^uVxt~6P*Y^eMR7|y*(BICgD(g1jM
z!eX;H9rSlwX#)lMPb6>u^*R67#D=~EtA*oX#*M!&G?iB~&_LpSoizE5jsPa5Enc5}
zghc3RwjsMcpaIkc*RaZh3$Uu|r+OO&^uG)(sldCT65tf#<2DOG)K>tp<19urZb7Mx
z7IDbiXJUBe%MU;g*c^!Dl&Tf_3C#|?%`a*rpdxU^0Q7DHlmQrzkOpodp$Z|tSEAB&
z0MN!8@@MLG*UmtwzYZ$Vj48Yf&i}J6{;?yTyE!Q${q=pDago89V~v+Y&&nuRH|EOl
zGxEMpJD3kRgf#<Jp+?wuTyd7*PQWshR+r~c%r62p?lGvOg2{KD-}&_WT{wy(IZ7Wg
zfytx29*{uby2qj-wI{z;!8#^I@`f)gsRAa`6|tel7GNFO`h{}>!V^O3PU)W~{y&cn
ze(17zjDAS|m&>Tf6^#tV&|y2Y%@P0+e)Qo)IXa+JikPal1){mhq2(N9H5{<=sGXAu
zQ3uGR5Lo%mTI_fK`Q?9(4kI}CMv%FUe&1`KS&>o-AM)Cj(T|0;K~`1ws~Jjw%fe{{
z@anZ@=1B>luNWlveE3D8WrH@}PV>75y07%R#jj=(zo=(`m*u~LkAEO+{^-3g@3p*&
z)lAq?Drv<ffJy!M%&)XV#6N(9FX|T=(_&q5Z8$F;*zA6ScqSz=nzCVZ2VYMQ#tiaC
zLEioGc^7*UR5M0EHkb(lu9|#(MFwynx~8@Tt9}FUL%re99ybIeuL3|~qJDM33FuSe
z(p6NK^b9$OpMb%R09C&(fNeS00z=HV8A?r-5ZexTvbKgcz}`}mWWwI(%z=0Frax~~
z-(T)!HeWC9moVT1)d;QWVMQ(fXz4a!`}hc#GKB>u-6ozQGbq0V;Tj_^HlBdq7l(jy
zNfRKu+I}6ATRo8P(&|(X;H4qxB;4#{uQ-pvAp-fi$QDe~r+w`MRIiV+L`saH${q!Y
zm0*ax6LO`F2(Xd54^P+f6fm9u^G(E*%^J)hjkxoMe_FFdupb`G2z0&hp;v<%XZ>JR
ziJ6`P_Lqmi-cKH*9fO)g)*2#!0&ZBqA(KR5d&#hB<tMm>$ld`c>v|yo@zfaVv2kxH
zeD%F)6Mhb*U>#%=+tm$NQLL%jq3DNO)6<Y45IbL!5wP#><h5`aF_uHM0t~pOaz^&u
zqTRe*)00FxSFqk4FHPRmE)ZP;(!2a(8bg1i9@B^RFa=e;!Vo!mMU5NaTp8@^ZbLj6
zXVSXJU`cR`&BodWZme&<X1#X}4S07RhIqFoRo&;;-EkHzSGzb@=X|}tyR6FrOla8j
zSjK`uu|@j}2KSO7%o7daREUlT?ZJ%KYAP@dF~0pOP?KpR0ajP{FT|C#&35CVE*$&)
zB|}~E)AVhKjb}m2JVFjfKmx~B(8JXVSyQ%X0W18#t@~mq*#~ddMgSwMXZP$KVnb&{
z=y{FWVcLNhuizGoW&yB#`Vb>OYTP~S9c%Yx^|kQc9^gppzF5_-P*v(of^=0ij~H_D
zr-(ysAhuO$=Zop8MJ~;#t<0R7c%8HFynj7X{&s{qKLgwLaR`n1@AqvpGZLCF7553i
z!#4ou%D*?AYF|`XlO1iZ2Aqy`@4ws^Crv97)+9qq-gmZuB$p{I10rhP7*uh`peJ_X
zeg1J^Kh1u-rtrw%%8N3<tKliyH7b@+(RDyu3nXm?V8<z)Q1FOeof8t&<M3TPul+8p
zar%_q4*Y)0(iH#}5J1V%VgTC$8it2WSU3fMfx=;a!yU4;z5o!7|LiO8E2?mrX;=%8
zi5r6JF@DAM7(im0P?tZLQk$;pSS5f<TJUT#jla3y^SKDJdOQ1oIoL`FxtKyafN(&c
z__Zwqb{hciG;jn~Da+%s*u1{W$$}QcB3cS^RDB`aG80r*A^u`g$+kMT3K`Las2%z)
zao#-En2i&%BTV}C((_esg4;KopyhLbEvi~i;7W31k8gtkctvypjJ0vxad8xj-&662
zu|GjO52ZOOn+x!zaJHKSXkQ?7+nXAt=im(L85Rng=>pnZDbR?}-u7itXoP*GR)ldX
zj((+RJ6IijWNR?NxBG)~1MI-Sv(-Ge5Kj^tU2Q5ZIqBG>SN=X`o8U6`HhU3wymjUG
z;5p4$*lFqsy++3R8~cG{-OabmO?c+A*WYlHW9$+@<oRN8Z;pcru$-8Dm|3=rfQ{{X
z60O(8D|aV|Es3XnJ7Dnu*cgyDHCro!W3F@qu(D3RGhT!{PJ$u{1G3?X99pCH$5D#j
z0+jOIv%(GFT-*b-;993P-DK4o*U9FCq3i8<x;f!2iOxL6E?TTw3?0JpZlKl--*|(i
zC>MPY6&<MgMBvNfw>V0Hyhin1G=@U-95kGc_R_V|^$oz5eMh85N&2X_*zmQW7FsDR
z?K}@<ayMi*=cFVnU9yve3%Ie*oze*;OkJNhp1)3=9l90tRE<{#dDc$v0N5dnca`Qu
zO&6N^r;Mc>q2=fl(w}HmOD<nP9@(rpl<&1lZmtq9L-*s!#p9CFyD_>$M+S^N(Rr5q
zqh&*Ps$b%V9oofS(g`@o?#T&n-OH3jFfyx&_y`@~Q0Q}OJAyRLu8%YQh@txEE8zCH
zBi;>F>St|zBwnE%{`X*7H`>XOJNb`)?|F%KEHW@d%I7MgCv6o5^=N_}Gzu&4R5?as
zl9*Ck&4*>~o&F@%c6IE*AQ5LM^P(&JLG~2{hV?hJ>M3hcuC|Gk8wt2SNt^r>%zmQg
z?HGi9S-_5{zkyXxN$w6YOUuf@e-u_iQ6@lI<O$fAEXOUo=Nn<umpg37p8NR_=^yGN
zy~8`XMe6N(hYr-liSf=UZASpe?c#WxGZ8Vllg?A`?pU9+k3vv~LLl1+c}icQGL77Z
z2$gmoR@RLv>p7;ier)L&1d}LNRwT+M_s7WsG*_j=T|`oRvKzDI^-3@Z%-zA#do%%D
z1q!D19yiXNEX8BS7#50|?7TNrlO|H$mo9(^!88b{O92McHMGke9xImO8RW<rm-?B+
z+8NSmS4X!zxdmkJgaWK3aN`dbFcE~wbs4nW00s+y_<+ydAY&L|SmBp=*q#f^TNg7U
z8?RY^&ceo2j0!q^TUW{Z6*hsfxq$)RM*}oV_eaph4B4;N`l7muP~2u}zRi@_IrV-t
zf;bEPG>xsihxQy8$NFg{9G2J7j8)#Do*nrL0>UuclBDu<YgTZuqXd4w4;N$0F%9!C
z&wpB$|D+6y{n_x|Zj<o09YX(acL*s-2fv-9Sgb@bvEL>!f9$x1vZGWZ$(x!ayxtdS
zLsgRxY7FuYX>(&X`<Vgqqd(VJY8NC~<e3t@9fTd&)wD@TZagYuXFU|JElR<MZmd>B
z%HRc5D5}5F)!V!f!g9qbav<q2EbLq_tZd4J&y)2oKFZEt#Ld^NyD6d<LwQ@t+V4)w
z?0Gm(nHNJ@8Ix+r5$i53RJt)iQa5onc{n$4E7pwMUfH*#8KS0#owiSz?kH`0g$P`V
zcEgDJ2s6!P*NV?=q@jKFc$`UJDXC>XLPdzQr6UQih4smaihqi4UXr8*^ZSRkfeH7A
zm6+)a96~j4bB-M*vn**36t+}bEe^P@@_5uDB9qUQO$f_ZR1zqK=YBdmVYPg33Wu#F
ztbYEc*29%9k{7T=aKhPcrvN`7elpGG@OR|kuh|TpuT!Fz)E}4F@=p{ht4BD<A1cj(
zS<=1RNd2rh)EO9Phr2HH-ok*=8DKRfdzq?~i6FseiaT1itWUak!@i^YUSI&O`!kYB
zI@_|sE$o;?{<fqwp&fPAWm<A~-e+$@Y3dV&a84+slSQ0wlVCI4PTn<Nwy_^L)HNu`
zu_jl40PlI;DC3ez!5psm+AW3GDq28ARk1DorF|F~xmu$PGNylN+Lz9Fqdg@TER{zH
zoamLQY<0=*-~?g&_Ivgd5mK#5VmAUQ!gye#!!5RG>r~Ml(`x1E?!?}AsbxCuq2XZy
zTQUaeHOK^BxFEDK;DjP3;@V)6%FV#+dyP+&s7ylX+LHutjF!5YANZh@;PfrWFUG7>
z9jLH9>h%|Sn3>s-<SiT=(wG!SQMmebjIuXa$PkBCH`~WhQuBT6xZ~BuOV!Yf+pGa3
zI*+Xe$}&%BLrMK_BNZuAp@fxLC=&#>-Q#BXxQ%McN?HwA`$F}4!7AYz&<W2<G+l>=
zhO}VGGpkX&u$~jGph+!|+}K<4`Vp^ejYUeB{B5&xuWrig_jm;j{7_mTuF6kqNdrRn
zw?NcotU-)(R10o+>7Wg1#XC(B?)WdPq-I~8M9oJOmbM=m*ndeVQ-zisY&>@u5fIl;
zDk@sG(QA-3RQ0*`d>>LKEh`wDOv*2bJYbU!-x@xP5-E5)X>9$dKDW$V2sPdEb>O{+
zS3NIX>ejhL^}U*3?x{I`PPA-F>|3qOTzx!uk(~~I==x`x@z=El<zrGeL6)K;m5lKl
z2*po;t9$UWC^`=_#>bAbUR65yngGQe;mk)2D|4$rLNV})E-aVynYJjNK=Sk`<j_u(
z8ucOizCE2SckL5_C<-ooVaC1Uzyg1q%)v+W713_4qcX3E<fftG$_?5$1jnn6);ld!
z_d&Bb5tqmz2AzT69Hm(a;d?U4yxy8ra=o{X?dHqZdGz?ukf?ERizu9RD%fhpH8H%Y
zeS9cJ4>l*cUbQoDPOlau1#mL>9DgB`L1q`m`+~aVhAWXS_XH`yitd?ujp$4Xs2mm(
zjf&3_dlzyu>ecvUPU9fIK2J~gDOprHyy^WIf936A=xH!d{2{f>{y<m(sjwhF&zs;Z
zD!E;jCU?QxXoUD0CW4Q7q)|uWM&zY;wAbPi5mEy%4|qhuWvGTR4l=`-zA9bJYIu;c
z**PfpiF>XZ#q31+5Ym!LQPB;3r%L{k|JICEZ9asWU4CB`ZATyfM9n)E?`LUTdcn|z
zKS{*3H^Ef+Xiv}(+LAt1@_EIl61|!5VFM+RF1_;w%My?It9)|X+FYmyfWhgc{G!M^
zXbEh+CQq1J+sDZCDm$1GV7VFwhB#$I3)9ssMmaTEnE{2<-Ey>lg9!dM>HP0dY%(aW
z$OQk$%oA?mZtJox29I^fULR-3%9a_yl{_+t^XNQxwfK1g%)=U*1BeaczLaiPVb~NG
zD0-%SH!$<RlMYq;*dzmW<HAcjt%=izJ%EO=^ZaGrMqxYCqM%R1S-${5L{ys9^S~)>
z5*2!4wm|I<OqxDd%2d)Eann12-xnKh?(O@8SLgb%Ex%*B`;@>eGR>5Qz&`ShX-7XI
zI&FZJKnY4WJDU7p-Cd;xew@mNx+FW@+$Cn(3%#)vN(9Q$X-Cf`b3VrLj&N_lFLz+I
zD(qSSX^pV9;WzshWi6MJxp&}TH4rzVMvxjMQt0!z?||toOQ60E#EP(X%eH9YnH>Hx
z)c-MB#Qd2gx<1oT{@G?_H}~s=)|=cmDb~r!Bk=&pm;|l^H}`by7JJ<o0!`|jtvOXe
zqP(#VH8WVS>J9L{OGzq;Z2$yaV!hNmk!e_Lk_EbPsbk8zGJ6%DH5<LK7b!X_%pR$%
z>50u1HH}lUr+zoh8omTIuy$oBUHz<e8F(*u5RXkVtrGC-?2zrWXj7D9O(nWCasE&&
z9GE+7vNrF?+F}gqQyyFJgPR2AV?dCor+)<S9Yq-AW&RQBSMk*r>Nq-BUT9e9caWB;
zA$}ROp>J;{V5EW0l<1F8>0T;XeXde<&pz5Ndq{?W2{x8^U=CWYKJS!1`BFt{53KqP
z4+a%VLF&qtn-WHJ$AWwJ-9v?(6_S(e8J{7H3<pUwBPJFWz;bl_MmR`18hpp2Q}|@D
zVK*96{~%Tn|AoOgE74#b+R1Bzh-kwGN}i$W{-o?NNV6*JP#Sh39WEZaa8`oB3*dp?
z_<<XJ^&yucHy0i6sz<-r$Z?-tLmoKZT%s7LknG{|oiaGbB<M)mmsc4FXBopIaF%6C
zpZ^0z`V&5vqJR*WhGd4{&oTvU-G!HC`es?A8%h^t4CG(jG~!MfSu+;pXP<eG#o6tO
z#eg!+qhUzJv)H=;X{vG`8fd5au_Z5AH`4AEm!=k21GJ(e3G3C)w^O=rVQ%Uqu+-d{
z%rH*w1w%X<D{Xou(6SE>lEvqNR`+fY<5;H?<H!x+HYL|kktc}IX3oZTm6n3jEAaOH
z8(&tyl}gHHsEBt@lxp;WuzlEwcmomIbt>Kc^e{-y6P?!17C3}=!v$(hgm-Rzdp+}m
z73Uz`OsKkmX8oa08hHSMen&JX@epcjL_<PMkMl|FF^B#@I^9T?WuVA9OJLPII?-PH
z!dnMoW+G<t4fq$~Ji2x~{zcXmulkjvdtOW8BM;$>w{88%g&Hb_@cp=x-CnryY$*Pj
zVGfL3{T;8Y!=!BBI7CRH6O^#$n)RZb{Te~`-ax#@ibY@?RN50om5_IBgd3Ow?LN5P
z!eo;~Q{C$G=x$}=#xM6L;o&@W1t<i^=_|>Pr1A6x#f{-}f~X_e@D~#+E<{wJ)4(Ze
z_H<rv`cCfgG`Tl3Pol}UX}FF<s2;A)@ntR^@@k+El~2UQf^hK_k@r=tC4gK-q;sZj
zP<M2P%TL2(RdTB0Q#eQd9j;r(L%zgzM*baX<Hw}!dA(XbV4kugYg1{{M419hk?mBC
z>C3Ses$2h=X-}Zq^w|S;>*{Qdp!QhU1Y1T*zc1j@j9DE9Bf)-ZCD)Lt<$q3>gg&}N
zADbYvYY)<P=A6XS1K<sJt3)<KL&Ytky2_+BigW*N3j8yY`hAX1+S`~*%+&c~i7Xzk
zgEJ7v?K+-gK{D4LiEBKkPO*}E{k}w?sfnh*pQ-E7+jnCfW<Hz2{y%7xM{xRAZT2e@
z@Dg7GSEX<hfTp&vS!=h>Xa=ys=<JJf&3>xKm<iwFK9w^03G2;hV@|m#uiqJPFHrM-
zOPetI##AAQTpA5jW)zOz`+N5Zt#K0g-5_}20Z2tp0=#`q@5n|B<(P1@kK)Q+r(LJN
zQR+L|_iYBbAW~o6tU5hZ2a0xtHw^Y87fVD2vaDlHmD-oWR-T6)DqIe`dxs%@Y*{z{
z$OmkT!MD$|2GaE7`vNRJqYpe}s)Q88y+)ye?>2f)7ZPiiX-f`tt01waAAP_YPvxMc
zkDTRmMC>M9XDxVZ)(QfXHsp(Tnu&(@D|0#myYju|?Hm<SZw{SiTS^6@Om2d{Pb?8W
zc4}E$nWLUB_8h{i<L&?4=6``<fBP|3ppG+1X7!I^lllaXC}}jQ-FAI6=8%Y-q^r`N
z-ET*L4ZyHl0;}R0N<I*cbN$4Qa6wVgYWCCH>TB53DHae!lf=CbdU*eQzW;MXmeB*G
z2`{XoXBrIN=e&H#?Qe!>xzO}w#Tn=N08Ue~K_3D}p0Fwg_I3wB)`8c%`u>kMB_|_D
z`t`;EjL9MDMn@hiB76s|A|FhiNuwZ>Er{s(n43^tg9y@J%+T|=x(OZlq#^s8txGzd
z%2u}vCJFMRy%;o-{H9o{ipI**rzkGa2DZ}!$nP)O+{&%M1R#tpyvpFvVa;og3$nMK
z&!=PPAldRh<9UY~*@z%{_x!fbyw(6|pwl}AU8Ub+l_xm>1K*E|v*5PUKb{Fzb9GZ@
z04=o~zJ8BeE%~6Zpf->DVpKaMfwLk}G=Evz!c~M1efL(0k5DQ}N&e7^RTVUY)<s|j
zjV;zPl9R0ZWm84SwR3E!2JSu2NfSn1|7&bA#Kdf54DXu=ekxldwWV|d0#OJuI62%V
z`fJ_d??K5Z3KlZB8%}@EbX54s>_@7A@cvG`1Aszr2n()t3Gxe51DslIJ2aVDQ$<7M
z(pCz46R4P1*djzfe_MvfS-gNMH1Mvp>ppa5paRwO!)_%K0VY=Sx$s#QZJEqCH_RCC
z`iJ=VL#3RT(mz2J<TUy&)%G!H2Sd!4X-UPP@j$SSYB6YAqJi1_Cmt+wik+Gy$B4Qk
zWCJL>D>O1QQZt1rL)bMz@`V8|2GxlnpWKu)eD7u8DW5rtho&p9Y}y2h5YJI0i<I^_
zV|C?F=?Y@g(*3kf0qQZBt)xkFP!vuk+}p?9`B5cU+9#QcZ3AC8LT;c4U{72=?{#Q-
z%J;R&6Ut}{Ho&Ect$hb43}D;_=AOxSe<8W!KQQ79g;5m|-TTGcri@=u-|=?IwtE1s
z3i&JQoZ(nVDCN_J7h9wnEE$xZKT4NV0s>U=%kTZ+!xdCk?`Q5;TPmX=4F-KMto(3`
zkR1Tg3PRJh0PQ5Wbatj7do-IJIk;Py(-rv4F!AwpEhMWw$a?%OGwoa8#J)q1YA@lV
zq4&m9K(ZViVSpa;RMs&_TCUHEg{Zn>@Te$}&Ghdj^}pWM{87*(i-lDDG08&wy-QTQ
zAzl)}33B`2<TX)*Rb^i{fa;%ca8RSsXkeKSdy<w<K&r0hShsHx-UE>q>Sr=kWN1!>
z!Wqca7(JdQX!m_kU=^?Q6sfD)N(IctMoc<cha0^dAU5PNXt38u^ChgWY9;B|3F*2^
zU&LZNfK_ol?mob|ao-0uX~1EW0%-zN^ITRV)Zg^N3sd~qAmk?4RS|kYxEraDe0t!D
z&j`3I&vZ!DMvY4_ylorbVPh0w`{1>{LT&;fFC!r8JuZj^XRqN;D^c}C(>}hzqypIK
z$mhp!9U@{zv5s$|UZ+w|&K?phfbK{E8xf*%dQl-4j-cWvt(^8l1?Rr*xv!5~QbnHS
z)|51VMsagSI+L)Q4B;Y3n%G1?$Henh$j=TozD14(UXMAuT*Si<Ghlce(pog+3!u|G
zRt<o++EVCYX>NuF*#)oGvl4L`Qa&PdyMAoqQ8UFqlb*8z;%%W|_frmw^^x@)_p?Ik
zLE7;kM`@JQ#2SLBiKbhFU5z~O1zgY}FD-M<+GF%XEl6I&3?e}5gvfBV!7#*YT&~>L
zAaalhjh1H{nz9Pj7|+S()5-7MYqFiGbC&kYy0F`q$PwFpQ3%Vyeoq@hDn-<`9W`3x
zRoNH(CMbt41_$Sj+>oj?7xhq>;xTXt@^hfH4LnUb5SU#6J%cEf?GFY=`x&<$xkNAS
z2;O1DgBohT7GgaRK{f)2X1NVPP(0S!`5FI%ExeNe7On9U&ghdI^KEbu2igzXdN!`%
zjK!HWpb21O9<<ykto+ppLEfTTyxYJWf@yS49IB^RCTBsrDE*?E<;g|RSjc~i0BY~D
z>Y$Uw`sk#oIOo7fy7-=c&1<)`0~^4rz#LX{v=4Lax5+J^EH$3a2OF^kZz#<3%2MP2
zahf}}diJhH{QNJK@4v2gD7j{HR?Y0;`x}14?Gk8>M;8MebvyZHV?^FE0~)IDbVm|S
zl;hVIftMdh#GB5w5Z0>=BU+Pl$lQCF$%~}=oqW47zRMka%ko|j47<1@*V*)PQS*V>
zH>Z<Z@5}^3%I`PdL?_g~EErcN)49YRQg+j1a(NzTL$J1uS&V37$cl8Nda=<B+SkG0
zqL?zoadCs^Nj6my<{8yY&z=#`MBTvv>ZKUp^^T;XC|~lvoUoj(tay^x4nRs|a`D?R
z7YL{Zal|6&;amw*^YcI-f^G$)!&v$88SHTiOpv$%KTTndf?5)9ro(B<&rxYM{%}6v
zi2)VWM_XDh<kz-;a4b^bnUQ;zvp*hlYnEsGc4-@}WG4N=Fs-XnF{RVW5NbIHYH7`f
z>AWQz{=$fQ;HL{nw`8{Ii!~!{^2*5{AlSaDTlfZR-jg*qU!@L8&1PP!_rJnY%|RP<
zc|K1}QX{5ozHF*%QB2DDpl}hrN$|WA*FV;mbauk@o1gV59Ln1|+;77My^?$-4sNmU
zl!@H+Lu6QgVT`+c(M8lbZD)$@2I%x!i-Hr!Z%4xxBbsH+URpSUBR|^rL8mOLNQ~vr
zr1Gb|+$B2MT}gsXCk#qGUw8KhdU;uyqQ{x!SqKiJ>{=mT&<+1i$V7x>CA;qlbFhEL
z=BBDLKfb*vB9@RKFp;3;1@UmCF0YRz;s*&JuZ#DWrI+P*e9k!+N0h0zaaSZlE{3Du
zV-Z`k_s2m&A`u;EK#NQ_OONv94Y>v<*H1$>jdj*s^!Jukdl9THEW@m9aRD#nvzWi|
z4H43E4OjJZ9)eDzOk?Ft4BtO4p3Tr}bP6x+SThZjf<_fgU1#U!C)2z1P$LKBh>~V~
z)eOn^d^#VQrdUphPOJ!;TTb`Pi+9h!s}w@*$aC*7iLwxoio{dCve^WkwhZ#Bc70Sn
zwNm$X>t2X%UAHmH+QKUtoAQGXXJ!S@F(m{Vgs+2|o)#J3&jx-a?O@@K8>PJ>UQx}W
zTZ|woIM*GdY1_T9aqK_)CNk)2kyi40&21`YKY)y=w_;xNpo~qM!)l+;3|r?=stW8_
z!J2`q`;F&;H#2jO_zv?SC0TFtz`dM$o46^n5{y+IW2lEI@X>iulc#OWw5C!5E+oU4
z5zl`Gzj_{k3(4Xp?2pj#GZK=$Q#CkE=UQaZNcNrONHcg;GOUX;kCD!R!K^pqUb@C2
zcF$~0+K4}Ll3@r>VS+W`_cn6-^TTB|{Q5xN^Edp|!M;MmabMVPFZZG2;DYk=2$hgb
zFuPa-zdgSs^|;69w@QI;@7-|6L3wRR_S1v+b~h+XOG~BkuFU1TaX|R};ZR!*@5Gzr
z&j7j?Msh`w&|d>(HxfP^+kLh@QQk#te}*G%%lCS&C=oWfA~OY0^TWcyy+Imph5@6%
zn)`fTmYRe!&~*mLhj3aZ0Q4z+jgs6)D3L_0h1*5CI1pHT<Wp$31$v5<``(p}(T5L8
zdK+@S2jmJ*I*AZbF+m?w7}a6OQ+YuiZzq`t9G{Z)RDVqA9F_Q>yux`b+L&%C?jqsM
zOXU${BH{7?O=wpbS)NDDrO`%s?*lL$dg1-b*GGt5PnnvbZ=z?9^(H86+C5Z4BWKxU
zD75bUE*C$x8QhvEC7DQwR;L}qnQAkS%e||CN3zi5C%n-+5i#@91wC+E7iT!P_VQFX
z!lF`(@QpWBQ?aUat#*-S6ys{8h?E~&7*E`Egy<^@sgn*<X)2Fkb%j}De{ow)iINQb
zaS3~Uiwm)%(!#JA2}CR-j;j8~n+4ENm?cm*!l9WXACN{~@kL^!mJeokzdcCKF>Fk_
z7A;cbTQl6Rn<i^ss>jVeSPgYTmX?Y``LN5pJGUAyl!ikcp;CNl#iA0TD6eU8vAH$t
zc*22pS~58H)oph42&p6m#L4e3^qLU{gM-Va{wa1IQ|K&lHb+5fgM2x(gdb15StL$Q
zm10SA#)P73)Yd#26W_lmN!9P{uN(wr-v`5FIh2LsIUh$8OBMlPA-l><yL;hiZ!2hY
zZ{?yRiAzKvaZ0kpBLkfT?P=};PS@g%6~gb&reN(>eC>yx|0P4x!3Cr`HtE&hap2G7
zNcIV#%c3rn_V>dgMFj6{tWl;XW~47$;UJdZ+bee7v!CyBXoLw-+n-AaH*gQe_;jWG
zP~=R-%(oT@jSI#X)EN(26;ra`mrs5^88SvA{cS7-j)1Skoy#MW3VZintX7U?pA8sY
zv^!FDO;5ijYJ3MDv`QznsToT$l_H{QGswf#8Tj@wZRrgbmBnnFo5pO&YX|OqyOPT4
zFQs8N!*5s;mOuN<vm1`&V|a(I4i5j|5Uvg(gIPoopJ`taZ+J#hjlb1mu)z7Qz1E2k
zM<jD3!mDLijD2#4%UTiI<AwC2Ma^<%@8NwMVRrVMX<Dbipp;Mxz#*BWg`KZ_#u6)7
zYVoQcdpi<kBxU6evLjy5SgS?Av)Fz0Y^jaSHdLnDPC##;F1&ge#WOSnA${kF1;SxR
z_wz;T)bx7I7^6~G__-ZZMXIjy`<<xKrL>D;HhqT-aeus$cK|$1V^uK+C5p;L_(%Mc
z*WSZTurr&~MMR_?W9x25cksMy5xoCmlr^u~(r3L|uNJIW^U#8@?jJ=Fb=h%PQr(PN
z4lD+^I%f_i_MQ0t<~F5$$8AhSb6*!2jVjV8h3#|<N_<vjh*(}|C+VU0pTdmMU+*yw
zD(!z|neJmfv7{s(ey-C`D0?6_N73wGHJ5Ta`tOJdT@2ti6?)5%{ts<}pWGSr1t(9<
zgEgk(1s84$Vt`*q$CYY)h6cyCmJl4LIy5SyFmmF5LD@42uLMjMBahI<KMKD8NYJsI
z>a1P^(ykR}{^vV?=767Rk%3=?*jrdNkaLfM;)0uK1{XR8{J|4cI-G>WRC@GunRHpQ
zr!`p1n+Sg_6T1SIQ>1@oP$evZH9oS}5=8RnV>la$%mC_5zN{oq$rl?KJ&H&m3H*7-
zQ`dvF;cgJiJ%OlN6)Kf_SSU8gNGYju^R<9DRMs~ESUtm3WyDX_ojSplYXSH-5Y11S
zqZ<Q!&^b<X9AXPVwt3hYeRQ=TE<((>+-YXr1QlJxjeCj-v@@yxe@k!wegi#?!v>_u
zICQ$nKOp#P$dk@>k<41_po~yL>Uhq~M(ZY}C!eQUZ3rxi5zg0d$p;i%M?(Gl)oC%L
z@2Un!3&LXh9`>I{gY*~vBZ!<0r0=k<qU+9a{0V6zhk&c4Mfu9lOSH&L(rgoeg++jv
zc7!rfjQ!{6|2ay3j%f39z<zEiLQ3KK^Vt))umIpi3a%gCEFqMRQoU%s)B^g{ND9Zu
zX)GuHTWEHGe{%@r@&u|py+XL7Kz6HW72pac48_I;Zm^jVdqP2;OEImFA@fHG^M8fN
zzpa(IXs`rL#lJ57|KP#CrUJ}ep-8zKf4<0n{ZoIw8(`et9|re-fWa&`=6@NMfBClm
z1vdYEhX2D3_{#_C%4nRhOq6k#tpEBj|8Zdc@d0NXh{1$S1N;9mQq8OYAD@e6@BV8!
zu+B<w1`w!<nYEfn!0LlvoE?!$+6cdRPwpgs9ykTD>Bv!tSaf@?FbZ$|A4AOP4W>~7
z`Ax}hj|~>%4CH<XsITdL1CV*-hgkEWHe(BDNCzlQ02zK{%i)NNYdOKy5K!?lEcH6g
z{$L2xIR0|zqKW-169dxH9OLO4Y2`A2Qz(I5p+D8_{Ryp@lLT<b%6Xps>cR#H^<nzm
zf3l6a3;=Iqk$J}sY#`1x?14M~$Fnnlr64br4>lRJGa|Q9xk373(ZSpT5!KWI!!QxD
z;AI1Kiz6#6$nX#jG?;9F9I<~~(}Pq5_`D+H69C<=>+M8^ax^yWA6Q&KS+)UaVY+4J
z(~eiw8=$PhrrC&k1}K8rYPAk*;4h2=9L|W;en0pEP{G(f&5N$}r%8S{H+wmU(sn;k
z9-g=(U+U;7l={*WqXeYn-LL!!b+lRnh-ukafqkP0K#JI86u%RGYx;f6|LrZQz~071
z`(0Gwj22V}Qd{CCHlOAa&~s!IIF)H$zj=4LmkPLD5G#S0eglwhO8~7bPvpqG`N|O)
zxa~ROxk6_i*~$kLJn4OKsQ6y7Nce}pVJfG3g);$~SVR5U17;`1Xricm(m;a1bp|Mp
z@cJ7BEW}Bld7Ym%C&6Nt{ZubOk1TJ4NkFDjFO(IIlWT14VtOxGdpvu`7LdcZ&gq8o
zF_LBl{~hmzR@WymU5ifJzf<m?lH-(tht)!<-Vs1$|Il$i29RXMT>`~lOnw%8*tL_}
z4UYO)9x!=_pqG`OrTy9pvCKTLFkY-)YLsE~gO7$~T);w6=ixZynE4@6!-LJG77JO=
znJ9Ja99_pxB#W)rZf(4F4-f@kcuTZ|VQv4S59?>JW+b_@W+x?=Vf@!4#F-mxJ!^_T
zu!6Tv;PEA#T&G+s{IK$Gq0769s{mnG4gi3FV@xTl6>Ws&P3IkZ?&cD#{3vEnLAlGi
z`9GJ%KTZWO>-^vwHE;dC@pjyt`p_86S3$n7l_U}vpoax@rG^^a>-+2$1I^cfUv?Ni
zau@h_3lOr8gBw!E5!kJ+gJ~`Wy=zB6aj`tI!VDz6xo_41l;WdjEfVnXK|i?!$3iG`
z$r2a`W8i^PF3!>`1mAQ7tVB52f2qrA47I0&d4|B&Azk?h^dWBtL3dRoliV#>3m_}Y
zaOA~8%`D`)Au1zyq-Z!u`U@su1Sf&O$7qo1u|Xc`$=TQ$qX*b<npJP<B_K!5N<eC-
z+%q3~*K!CL6<`;3uY-%`sW4v**k_c1AU7@WY5jxtyI|^5%x1N%4mek>3@w07iw+5#
z8LAqsfR38Y-}2o8#N3B8lBpkHv9so&<k_6lr2>k^1&(9Db+c*XMeT{Wh7fy-MHnN5
z128AXuglSAoE@xvggCB-M<5iKNF4=gZ~2F^qi%{dU|T|MN)hD}w}1=Fa|qhG)1Bg4
zNX~2vNDgn!vs4NTfN|^))Dk1|3!BP#)cA)*&oS8{tI*yT{?!z*I_;q9u89+Do(uu|
zrguN*ES4<p0i0rJaVWsgASPyvgKYc^u+bC?3{+}1E7o2Sbxqj-y7?H8(5rA1Lms5y
z<dHv#wa|`_Uh?5*paDI}YT*!Bnb$Zg2jO%1LL~;Fb6hE>@L(Kl!*?;q0B8ZRwFa~P
z6qI;eAnaCv6oTl&{g@#-&h8L2W*9TFgiYfsC~E}rp6WiV1<%fyt5Vnr{!cHb5-Qyl
z&A6CB6k&Gp5D-W3R~kqt<Z0*|jLu1WOb>qje>nRNa4h#f?n4SGD^g^Sgr2CV?EN5;
z5SeAqBAaYwlaQ4W*?X_-nGrIwva`wF<Ne;9)A_G+&ih{Pb#-;gczWFT@BaP1-_Q3m
z$^c%_vflxXZXf@_O5N>ja73sq;c7|t*+<$MRKXPh<(=p==Trnjj8wMDfU1FnO_I!1
znJC0htB4VWU4w)*2be@pmIq}mjaE6EY(rEP^(euVua8vzf<f8>Y%lr1d%u2IkWqDB
zeqh8nyW%{3`Mje!Y_eHuszn_c3T^oRDro=h^!@!qOGG^r?G`Ha2djVbt;?57_?K>X
z0+k#O;j+i$h*7^~l(vHc#}sgXAK@qnd5_O|B8kK7ItYySi(;yX=LU(CWM>>fJu!}i
zv^j{b8w0<Jk<Mx7*Q@*hs{2Ph#2rz%7WP^`r>Tb&;5j|TwY_kn&_htuX*bak$ch}?
z;U2eA5sM5tnl(6wMc@1L*RoO?79kS<+-Q()L|=)2!=IN<up7UT2sK^Geht5~EZT`0
z7&LMQO!tT}B^T-6yJ4x3Q=8kt^E%PAgYa~KvO&>tmXaGZh0-0j!D&Rg$uhRynYay!
zlwf+BnnJbuhDKN#tT19lX%P)b1j^8kHTV4kR^LL83*;2I2|BWH@Z}gvGjs-d-QB;-
zmMdzaui<A4{O}=Q9}~Y+B3_Vuk<29kcFyJ*aKoP>X0pSLObp_^7T~ww$s9>Yu*$|l
z{0K2S6ic4qD14+P<?r1nKB0_D(M#>E*C5+=tN*ckq*aJ{{J`#e{V+}{-3^fpMyI=C
z)+rpF$~GUsmI)=7l4#7yL`-ql<=wR<f8(RQfP=Ts1`&R4{5O8bJyF_}>1Ra}!}LN~
z2Y%D<4_W~be_g>axC+C3DlsWy2gF&e>)5EWRiCAI^_la&YH)wCQnp5@L+gr7Xp2<;
z6>Z+gLR{~7hJ&}S?;gC>u%Ba=dK3G)s`DVsOIdaI1a5#a>8_)9&UaO-Di0Nvs)D0t
zXT;f=tARw`(#YsB?rMYl#1+!rd%LJza@?YY&@7F*o}%kb31_O(Z+1^yp?IAlf8+p5
zBjCpc;vB<(PhrODU-$1;hIF0EoXcZPOb+9_3fd;7s`uNhyV(pA-P{iu&OeL-qud9;
zQJI{GOcoHoPnqWnDy4{#M?KhwH81-r;=e3%N6x=-E(b%4N!eZF_5=|EzMEMNDzS<;
zk9(-v3!&IzthAW3jo_Exl>3_s7AaEWf^m18+me1Y1D$oaJDyW!eUJ2C!U4co@*87G
z3SG$s#vSzh7<G0XQ<MqufxF#`*UguFy^cpML<f_WD+I$eVsbTNq`6eRiQIgcnkxh&
zPhbdh2KC%PJ`k_m4X@74x6lpkdHN`b7opU8S(TK~l;CjKT9%K=MY}nJIAyefxTb-;
zhJLx;niqnH6lc`^&Ud6!1!j<_S;|{!2&gfWT8_|TH*nXBRIMNPTqC0@EUoH2ag@t@
z0HZtqj;+AGalrLYqa?g9@y)AEU2~dGK$Dz4s^KfS&ta<*lZJK>ts5iC)Z#-xL+vU)
z^83A<6PGmrV18(ZA$m!hgDt;AIlJ}%oOHaGO2IYR9dBD0)M)N#BHzPvHe}aMk=dOJ
zwCzvAMcJRGa=lt2$P~~877i3k!d%~p_TptyV|ORw`wVQztNghJnM;^c+Le=XH@)uf
z4zX7CR)V`QONk3Z$r$Zm;$WOJw+%=EULrYnCCv=O2pM(1WyDa|DdnoJKl<V!ae2Ll
z>h4>2eD0Jw#nx*d_W2@pHm=OsSgSr$<2lF~ez_L9CKDNqYCBDpdEH=n+V>y^?}Izu
zj@IY~kE$BmIkl7vYPOWtFg6|Q5}f>a&)N#gD(pZv^21d0P|FB;%ZiO6Y0*xC->N|$
zQ^DKj%MJ_$V{{c`@x!w`oZ?g`Z+@01d(m)bK1d`%mSsLGYtzfQD;beZDJ}%|@saVL
zBzt_4EF7(ww3&NDihM|?2{q=^(VA^1*0DU#DQ(JP%-gb3j;cKiq8}$SIsFxhvaWkN
z>Ey<l@2NI9_VCFPJTXq3DiBfuT43t&s<0V{nHjCo3(qeOp^NHVEXU(rZzeC4du7W|
zSPB|Pn8+F@kI3r$nD_J*NucJ1!{2@6;#9CvsNUBuh1f;Dfe#}W)h<vw?Y3%)k0Q`5
zcBZUWnw82AUn6-I-Q6|?YSn)zBxlty_tLHZl-x}M{%e9Fo<(Z$)yXUmr<n7(ImwbZ
zDFpLnNfMomu7luddqKbp=Eif1o3kRh=mcLa7RUQ2;zT=z=rN+gT3ChT3~a0{RIaQ}
z_dD<~^1r5?6%($u^v8NWpPoD~?b$fcmq5^<f4Bdw?yMZiq`4`hkwRCA?T3U;7%*`y
z1sjOQ|AtG$E36f@&K(!HM@%dEjz;c*{(0QpdIi(a*Wy~Pix*nA7y_7$Jh4n-B6C;X
z&A+}Ngy{{`mRB#4l-NyFs@>;qIW$IEGWl65(u{U1@<3=;Xp3qR$cu<li%Q>npU`a&
z+?ajXNZ%_5RH#gG50?S`yKI-oKb4@UDp;FGB{$U4!L5@@j+wsZUZt&1$5_|31VI%y
z#`YPLAX_D2+Q6fWyqLNqhdLd`RcD;Tvh#aKVdW-p(F)K|CloR+@bpsehQjVeA49es
zKS4C)hORtIto><J$G5RDAx?}UH12N492@Q9t<u3T36+ioFGBWjL!rjXMk47*0TJ5y
zP}?5bHgOJhBh|r{Y@=^o+>aQr*b>g#zg3n<QPR1QC*!@(poN^D^4)1g#JH`ewxGd!
zO}4opoEb~y&t)sxoVK4Q(4vf%cO>qa4<z1@U6(nOGA=%n5T-*{WPiV#&VNIby6WF0
zegD<7$64UeU@czC>N)*;OEvodllTEMc|g6xUJffp2*0R)!bcs7a1WQNorY`kr_VD;
z1uIJt_2r4xvO6*>F(lb{Z83BzLzU>3Txv@HPR>ymN4xBSXpo0XOv^3b%V((hdDoPN
z@)*S=+o<mD&n4_5RA0*!qjVgln4{a`tEt>wxB_ohq_s|Nz8|^p)#0NDg-=G3frR9R
z$yE7-``IIo?WZwxw~39f_<ZmnDfHQ+4@$#Iip0|m%bP|UDLtADnm4u@N||Nd|Cr~Y
zKfukCH;T|$%Nas~9GQZoi!{8+$R6ngb!zko<KAbyjo6*%7dIVVKTwa7QsuqhXM5A=
z9I+}Q|E!U(Ilm?7rk=^+^ZD&HVs}aQ>A5d>4J&i@xLkZEtu`9O;<GH(BZgwwuuQ;w
zV5QvNQ*MeImp9n9vah6>iMQRfckcS&YZl(6f^IFH`vYNEJ9b^!N54(NNF{aP-ax^K
z9Z6_k_xeNc+3p(*<XNP-chrN6+Orboz`bu1jpZ}1YFwFUoImu?QLhN~A2Q7UzGv$=
z8xML>XX5_jJTq7~aGEwGX{T=s_f={;$;{P$vpYOlkyG}oaSZA!Z_B-?82i)G>jpQc
zwMf?kznOm4C$PmTQx5b!V~QNx$-JhqrH>nV8;m~2iv02dNV3}-UHopk1std0{!?Tj
z5$KjQ?qdkC3rD}Z5r|_DbG>+8ZEyUFO?&s%Nar)Mo@qs>F%DTd6Dp!k#2-1$BKojS
zU@9tEiAxsGC`Pz+?Kauwz<E`ZCox|-c`41lm+XzlqbxY&zAF!|8rX3ebkbPV_}JBR
zTgW}&)<vz|B-0s0d^D+a*#j)T>s($mXhQ(ab<4_WVe)m$R7xiw^!e+u{zj;r8)Pur
z5cSzEOKC#np%xe(rPVz;vV|?sY3j#Ja6eD6lIdDS><WJxEJ7n>3dCMi*@x%SOGsUs
zU}{(2(z1Hpx+Ri2^=hUcIxV3s6Yp;PReQ^dSLwYkwhJPr<*LXli+d^5LbQ<}g4A#D
zYL6AIptIC0z;5RoFf1xDyZKpr;5(9lvY5)XpJ&s4arcC*jJj3j8D}NMC%6JF<bmR*
z!bj$d{kb4|JIvI(#v7^mE*5ci>*a*9m!0eig4XKANEOYi9W^1{LX(hFFa5h|*{~zk
zqi<qE({wBrxmup{KUz51ICDZ*nd^R(*$%#aq!)4fA*c&Pj2@2niMg*jUfy4p6Q}r6
zr%>hlslB^7%J<x)3R`S%m9w)&u8vwVnet-{3tGFfM5FZM;pZ@dYp`M6J(2m{ac_e|
z7X7TOBjMX1=!&&t(mqbBXf$$1k|*jvE9erphL{xVex~k<_PsVsfqy)7ULY<Lu^h60
zAbytI7_vX~h$-xPZv2Z)+Dsy*wCBP$9$s&-gp0w~5S<BjoK{2oNNNc0W~OBGWmjLH
z<PuJA4&Fqya=Kh-+y#8{*R<~R%D-YyVihCNJ&n;o?`XN5;v(l2r_DmFj!(hG<}=A4
zKD+qFL!U}FmF{S<`37HNnw{un3O@S+yFGDLfFKfFTDqAd(Vl)=6GYwIGB~qA7ZBa2
zx3Dtq?Ripf&Yq+ickHO-(o7q}x*1pzb2aS&=(4Is(}ei!#r0O=7~O@%S*NSHOF&B3
zh3R*RY1QEt(<qCORZ(*m=MRQ8#L6^3`Sdn6{-p~qrHGC7?_NAYscK%_|9OIR`K(A|
zYIw(^{mDsRDsJAk-MCGGs120qu`*=Re%@3j^W^C^a5Yybf6w_jr0bPQFO|#}VAZ`q
z2RXk^8BG{!v|+{FVaQl-b~Sw=mL~E;Zw+a@KG;o-8947{+T6%YSE1&!*C7!|a`;k|
zVRf_jGHqOasC~9U!Oevm6pr{`|6YybB8Z1YLiGAic4-GA!ar&1Nkjd1h8`Vo2G!_$
z#TBZvxy_KW2Zl2TR<tFAIXR$a&p~If@W%4Z$}MWCOPzb?gT9X!e2?33MZ#n5e)|%j
z1LuDzMXe?JDMU!Ar&rpGR=?PcKH^C{WgVp5n~&I9_A`+YC{PN3(eiv!gG}m9$OHDo
z6$?4Rk4^ME=abkn?FA(T9<Zn+s+xST6>SWU{QA<k>l9$Abu{I2ngyFUZ+@p^(k%22
znHW7zoC>|lT$r$!Nq-!2uYB=A-+{!Eku~p^L)~W-GAE)OogSqc<Xds&#Xc}AAIp9o
zg(sBo?ffpY`UizXl%8pDK*0l{r%m*+r^raB1KtLou8>K*O2ur<OJ0bZcnUhG#DyXa
z<2KyO?>eU=flhzz^ZLh8Q;YZ1{b1fUQTX8$Dt^668+MY5PD($nuYHS*Z4ZG1!a3`~
zn$gl1?xc;aH3IwNHG>w8-lv({=_Y(C-Gz<KM{4&ksC^_y0Hu;7#3fK@E`2=_xql0E
z-@?~;51(2N`E{6VTvQ|YI}j9(h4ynf)6c%&`&AQ`op6P=p3b(Ven)7a8-t8Xk~)@$
z4r|zU(jV`KEH+=Jx-yVe-IK|)h~L{8jVm96(Nrkv>M#JifkGWC1}Z|-z3Ys)7zo?+
zAhqS0cN*_W&F?l&`Y(tv)1J*Glqb{ptMdU}+b!7ve}1Nfr5&e(A$+B$tA;-5WMF|(
zek!rLT|s5!j8*(2+Bhs#twQiw#Tw913PnzF=Npqj%S(R9vt>fjf;?L$Is#x~5X_+S
zwi<o7s?e{<kJ6lZaQ7-H938pLr`uv%kG2$wyEN?>dKbq_G+$L2%NzP)|J+>TNyj7V
zm#GRlY)_ArRfN4Bkk6&V%D6js9BZPBoZ69L2>GBN+7X`4!J;tePkUKv_P{QinwMB(
z$a9CNqN!QG4X%DoHx`+O1;9_IGBFB$`kG)k#Ixspz@recU|{c*E?lkWsAtdrk2|0m
zvIB-)EBxb0<BoQryCfrUxG@gf*tHjL`L8>$jN*d;Xt)ecHN1r`Az4N;fzQTKSND{W
z;5$!*Ecp>;0R!@i`#z^q^k?hr<?JA?NQ>th<lM)13k7WgRlTSTD5IhwvvSmZO;%TU
z=;8)r`O8?-!jt>Y?s;W1+ZQ&L-`E@WTahLTyoqn8`l+a+8xvdByFZH%g7|Fw51)#+
z*#Lqk9wpzCXfH-L2&Wp&P4dkgjp#YGN9**xU)yXyIPhpAMb^-(pG6kMH(4!J*=#C?
zjJLKMfcB@W`g=I~+)qx4DfDS{1ghm37(a@vki?mySnEa$QwthS{JHH}*dR#V!WeY&
z3c)Uo59yxoCN5th%3=MQgOwe+dY>|LPkq;3R`wG8l3{d;60f5fQeD=<q4?11QM&1U
zU7zcBpt9&CT{FBI^Q>|={z6U~#_?e;#DseShqyHB0-u=FTLX607B#edMdoP?McEtt
zH~3<&{u4@uLQt~Su)#kH<>-5YZXwVE-M-{}u&Ow5N-R%~B=Q#rQtrSEx`sweOUe6*
z23%6IHo%~3T1Oa(;eg%Nf$zctQVewJB0ld*b$I1d2nq9&d!9!r@3{Av|9go1E3l^E
zRdYVvR1_|R{hHy-#p?_kOGCFZfd9(DXtH%wkevc@=t`hl)-&^PXO(FJV=zv_)(9C0
z#fOV5-8+=nHcsb<;qEoIyk<I~u97^rJ7TzRlpllB1@63YgjiQSgYxTUQ8~wr421_q
zV@OO<8Xqx{cf^;&K2~^G>F^~bu24m*GkBfTJ`6TZ)s)f>4@g`B_)fi_#NRzW13_xW
z;rCAwb?LE{=uBA+!MR`i4rZ%O_Mt6hDH8RUDFo&|1Z$}@E^-2~92XG-7Oc||rJE?x
zo$;pDsUseM9!N4Zmytov0bKm%7v~;qwT4q?2B5D{u1bZRzv1sVC^cIIGnv9WG0`Um
zW?03n%WNQc35U4l!>FNZ2;B`;bi}@-39G=JZs{8@5kja@%z3`M>*`m8bjyiV^k;G!
zRp*RCvmAlS>5n;`K`pfzI{xE<dZvjgW#B^-VXk<NQHYn;D&Sw5dGx7iw&cZ2I<ND)
zxMC8$JCf{<Fs;E7%a&nGqzSt%=cHLFZR|d%uGe)V_%O-z>S!ugFN-=5Q1}p_!bwR>
zEpt}|4M}!wX`G4^vWe57d~J}D@*YTk=F<7ntp@IWrv`lKBoQ(#ZPke3VbFbbxrl8*
ztrsU*_oa5z*L|)>%HGP_Lqc75&?IY&Fxtu-PL9s&lgo7Y#6;?<#E&`zgH5hSz0Aj5
z;8pRp=Gm{U`!7(f*-Mxe3Yo`Erhj(yd?7CA=)BD0kwG;RU6ds|WOSh<*aO@2QB>Du
zH4PDj^DOyIW{}FdP7{4!1fL1^Kw)i^0z27<a9^P|`L#&j)I~_u;UH5CN0)LO6Plv(
z*aM}t!NDuZ?q;4XDsTay?OX&_nwAS|d+9t)Jw16y71Vh!SyY<SgoMWN06^MKk3Ib?
z`(O*_c`p8DotRE8HL9PzchV5E{pD$EK-(I)9|*AAI<{$ov~cPfZD)BN1yAQQx5$FR
z@np2rFUQzw9rYzSKZDq;pjbS&wuJM8h$oISfF4?osHq!)PjxKXIhBU-0q>;}0NN7$
zG1wzr4hX{YIUSq5+O<<^+n`o&fbzRR)<~+M+E;^E)PVl!lXA34TlX{ZX;giNeLW&A
zACk9PsS*JAn`sCfoMgCWgn2E3z*4(bA-3@`k{||-B@L+3DPoIU^vBO96Q_k=rp@cf
z#ylA0b&D*$&pP;W?-)+#dh1mFfiyf96R;u3EK?HLXhRZ>1`Fq{Ycwy>4-_O@-6XSW
z*fOB9BT3x>E?0qVT<@-2qisR|%h}7kwdxzJ17*B}(RL&iw@34u)Hf(*9Ja!J@5NJ_
z=c@6t9mVwZs;vZ9w5e_AQc<#gPDnc!y;0ZDYW}p0{c8+J*Jsf~4k3H-rHi$r{Wb%k
z2mXugVw_z`yrHF^Zl#RGKX_HenpT|9q(Vusn*@)<!plC2)h}g?uOb&v%bAwrFdHd?
znRT<8s4j}zzmo39WgS^0D6y2CZB2F|tWPjg0CoDtt^?WYh-AV5E5vtVn$C28pu_2*
z-9uI$0NVrX6h5V~813F0_}GU{W}~DY>txwLo`gW$Y3Yhp=^UFlfw?j$s$0h*ZY8e_
zP>EKe3vbZ5ehV{D*y_aGQx41uc{mg|W4vU=VfTXIjk;k)j8R(Z1W%AJ=_L7-@N?*N
z-W?-E<ffOOR}ZDOGB>-sahhzH+bN-)zN^M;uyzuL7=}@W6}?sp3p-#S#CnKfU=nmt
zh!--HVUZ_E#?!dTIqDfd+P>2kZ+{z1oPPDp|J76fhWQ;VzLZh2M!(2oxES7sykDZf
zA<V-2<8qtvuNxB92f#mOvS|*ug`ID98WZRulGixBCm9|DGv5_+BVBTsDr07Q0h53>
z`aaRF5VVO8b=|sKw27xxriU5AREfXpXtuQxfdoTFi3%aP!x^mCm|FX2$Lo<SH8Emh
zPYLD8uEY5L`UKC2`3(h^)wex?i-&_>vA2N0%6(dz0aLDg14EPUoHrq(+G>icDZj_e
zg5gxLs6FBa^!>c|$j^AqQQ`+_vMjA2S|ULiI8QYPg;r)jeJ$6PG%40V&mrFBBPs=>
zaOScPJ=#=VZ5l8Uv|iXSKzE^hP$35EVCQhlTcuvGYQKC;7PA8}J9^59CG94Hdy3IE
z7>7-hK^f*M+SkfMSEQ}Kcg0Ctqa5CvO5G2-Ml2}7z6P_yei)~!5EU^!@difW(|IRv
zL!R`4>jKEMf1mM~Rv_U{f6zn_-{iXv4RE4FSNmv(BUHyu`3cg2`V!`~b8BZ|8ghS-
z=)GwkYx^NGuztfVVa-SpmV$N?EK*4i8UJC_`z>e^4wZC(<gQ|97ELXQZ%ucZvI-}u
zI0+joso)h_OX?)%z7Bk&EbS!hW;u}KbI>+c<DQVhx__Xz%{W^I;w;t$lM4H(<P8&V
zB3~(`j3-9uh{>R7G^>-6vsawJgHGy;5O;tl*K>s{Ql4BEOlpq$9Nb1Y*AjcV4BE>;
zI1+!>Frf7z=VvPOJ3d338t-lB3x)Q@#EyhayBz}SEmT?+JOe)Q7h2m{cI`MlV7Tzq
zin5nl071r%<j`{aUpNM)fIu)hzT4^7A>E<%PYjM|G&rNp<R<F<zKpAHGsZQef%E4n
z%0_%)HX)DhYYKC`1`9We$b<ilgHN$fb#3?O##uEVjikV(xp(}8cVEceEqFcO`_O9s
z5iwsXmE%F*A@ODFDW~4|1fzSlSBYHs-=<d>MQ!y8^rpYrBXXWPLdO$2W_&FqPM_5%
z6d`hla1RcxtTg<?p7ow(g^iCrql<U=G?(sw+7g>oxch=(JSzU-mHEz`Cv`>|VKM5B
z4~Kgyx1@>$#J;j%Dv@I^E!dj}y%srazkXko&R-^zuKbOEE6OtS_IzwVV^ZOAuQ}gc
zuB$`RMTuLpvGdGBOg@8O77w85@)_)D&J8FMu`+tcNw6liE@}~9K}T_sOgrBClFhJr
zQnLl|Rmu9V5<%*pP`XO<VS8+&46y@2TVdlr$PJ$z4L)p2AdTweB`G%fv`@18tbIs<
zd*X7><=*t%*Rov)eUV|C4dhYwbnP1-1D<jg>ta)0p|mR+jC?UuqFmRr{P{xQ^@FSu
zBWYb@t@ni@+G)8<CZpN<Elis)3Plbj*VQ`u%4I6JW>x%)k0f`ZrhT0j_;*kD%4HYi
z1x$~w8}TL6b464Xlg375{u6<fkU0#fMbGUXQ2be~GDlZKhe_D2!lDu!j}1dE*EJkw
zl@A_8V4Qo8WbnM$2eu!EJy^()OaEN8{?_7Z!scHQCxGqaQoSRCc#85>Zv3KgxKiC4
zgMP?6HR!8~@zrt&-#Fbpy|se+@&d>53WgY_f+cI>iURrb-z$wCCYO$k_E$Il?)87e
zDnl4<SyUJ?|7uRJV`Y_0oaTWTI@Y<!&@^)}J>vrDm~xfVp$X&z<UwtB)dIU5cy?%9
z=Bvqc$BrOh3{@QgH^1O9YjBi8Hp?Nz`50&%7T<Y)()O4XFn8K;t)}K~{duAOi&)kv
z4-B%!RC;vzFZBDuHC(V$YWxi0n)|Cw)Y`|wIH68k^S!NLMy5@1+wAf83-H>4O}Ga9
zXDPVaGyo=QkueM~bV6D-gy-~=YcVr`5LDRT6IbisOQDzjTWsAm2B@HqWhw6rp8ey{
zskW6IgKaDa$Xgl2MC_xVfRxji6t3lS@C0#2p=4MA-P9mbh1+h5o|Dl5J38(0I|wC}
zFq{x~?8fDKx#W(&yYxW12Yz63z@??X*XjJvk@xGxZbSPM{vQs7e^GV**Y!nS3@jCZ
zjOvaZnDh@=(*JN0{y2ew$8+M>8~EEp`G3B_l}Z{eD*+X6B>q2~<^S_zzkjbFj@Q}$
z_y6k8*SBYf>%(WeV!-`(i|uc(Ecl+%aQ*tKU@_wgP(%iS0c!#^LebITzRi&J7xq7E
z6886mV}~Zf7Y+CR*4lOBCSVE<gI--4e;ff=qGLd9$weantUHCU2Y|_Y17q`DmT(*g
zChUuzXQ}x=KJSK{6jPYSs(-)5L&zLe4d$Io8Tp^{2!%83fmT~ojfod>cwydM@MSe1
zAKI`6D=J{-z6Gg^1Bg8#uxg7C7f^7iFaeYQY_-;aNBW=TMmlo-_2z&5mVe&wZ^5t^
zEZ;#Doc^6N*2(@KcA$0J)?X>^@$c#cmrMc7RE+RW%DIBAh2mOqk?}Z}<?MYRAa9%W
zq{wk0-i^Q*3@i3MwQ=TW5c4lyqGd@<UVa@uPa8tPDcKr^;EEzhgS6UlLU{<%V}ZO5
zcL>0qB`~nd0mrN!Bqv&&Xda3wqrGGcr01Yl+)6PnAG-AYFasSy;QzNSuuL>?=`Y9C
zB&GjcUhGci`k^+|-Ux&C%IxFZEbviCUNR9!r8twoqY+*WiAL1>vk8HVh|Rjicd&U%
z-dq|g-q8vWrV$a8>x1-J$#R%Hn1brL^<8Qe%n?S#<@yjZ8>Cm}O8S+7pslCMJD<?N
ztr~nk9|O*8>xXFzIfkMR(ON!N;tB81R37aaM<MJ!Xs=A;x0apn+Pq8n7;EKR{{H3P
zJOwV*9+Gejc-&du_}xRgQ3Ufpc!Rep_94YNp;MH4NB^C9<7$$oB^GA1*ch<kiou+;
z2E;_?$5*pQ5H9V<_0JSyACMxvj#A-65eRfKXdl;2mAp^DLTXB8EQAYT0IhI^3{dJ6
zi&^-r0LMrR@Rt=Rn}kbl@*<CY?-$Vg=hFH0&i!P8?iV1NAgTFtLwfAS#Lj;j<MJ#<
z;yPRuSYAwVGGoP-=%6kveg;bc^cZN*b0L}n;KAam0cakhd3cVFbpis8wWSGuodf^7
zk%fTO_Txtkseiagcb>!oOU$d~04ZtPe?$yD!2_0W2<Y;7s(HW=8-S~$X^Wy%P>4q6
z5#q1*_D7GO{^FwL0X3%+GOJk;9%3YFsQ}@_8rxeJF@KwanDlpiJ$lD*<mNu9do=(h
zy8&34BqRJlYoL`5BJ8vThO}os_KY=C2t953=m29=eZoD69zBQ{=~6rGJXt&jpg|v8
zDFO_YL$hSBaOD3Y`;CTh$?F0;st9O;>`$%k9iG!(V(lGA(zGJG@Q;B-f;j$a2lB3Y
z8D=QXT(W3;0xd#@X?{Bw!tvFi{-&q7B%fvgh#a52rO7%x`xN`Ri1#eHaYolu62?Te
z5YyClBlicLK0oMu*|Z=CPriVZV+rxho=HBDxkcaG-BJYJW){+_bMDKb(VC#V-hzh!
z;qW2<O%!6kYVMR#GU3xlL>p&f4}3}hKeNSR-y1}MsIB-4`7pA}#CtOt6-Rt~;#4pg
zxO`neWeHvGt~iOANePf&5>g=`8eXy3QYg)fOI79`ayET$^^xMD8v~R^8@l&IzQc{~
zF3n~oW|c7@Xc|Z9MH>1%&AzmN?<DLGC%Y0ty2=Mj(q1SuzReEpZU3NlgddT8hM1*h
zLaP*NP4;szXdckA%@PTaql~^*LZP&AZXeF>7%<~XXSYg{Zk3%shTH735W=DcnQdB;
z`u=tBdiqF$x0(_s6|`76h|9=Ue(Yh!%D~)Xr9e*;c_*n#)yjFqkr?iGxy+vCZnfs&
zyb|Y9{^k9nWiE7-qrhFRL{pFe5A6MeuJ1QjBiNDB?N<H}n{+1N41-0EHCSp4!G1w_
z5|5ZoGUcJd28JWNA2x2%Ku2J(7Qv_jy*k7YEUksdmLo`VjkzI(zqSM9{98k#M}0`@
zwRM9wC-oQCb4CeRgsBoP$wk22?wY$l{xJTmHwMG9+z=_J1a%kVyWrm*_#x(~C8RfS
zSByv#KMui&XJV81qQsdyQ11=F(?Kofszod3hot~-$V`zFoPESe6eQU%=k|aD^2L*?
z0AxSE#EL7*AQ93ph0yOjelqu5HHv2UYEZXBE(mvo;mCJ{?A(p7j6ImTjGPKX0PN)z
zD{gT<J%Np102ErffQ0&Y#JLhWsb$HLMl8q#brJaJDZ{~|L9;*U+E<bplawJxB?B<K
zT>L^-Z!N^KehBEKZ>#u@o)%KS4CS)TQ5-@UOH-IwN41iYN=<TrirnBU@uq>A2Eerj
z`cS)-$N&h!DSu)?MLofewkAVVlU&w6zfn?H-hUDh!E;STGAE84;?lvz>~I!;J;BeZ
zhdd$R(kNI~Cp<d_uAX!eBgeC#5@n~SV3x$$P$`LH3s7V~q`8xSrQ00?cW`ml(441$
zC;2<wb1RAeLV4gZn7Bm^C;1kb%!=1UyZ}@PA<5nTgl#-xu#~Ad)lvA+C_wTcnNxJZ
zb&t6#vCn`bC?~Vo`PO81L8auSBcOuU?`GsG4t2mlw*)l9YhUdv_vi#@&8NX0Qh_;c
z0e^B8Q6zIp@!mY8vR@EmeD(8N`>la9ttQ{pm-frMFOi1njdl=sUmv;hK;7i;_ff&P
zQy9m;iTz@5$9WQ)1u2;YX_yhI{dAzj8=0tdnmI;1#5o=2=-kG%%E##Zb6|%(75JEh
zcTaW2K%z##!xJcHJZv!Gs1XyX5fkNHxH9Px6Jc8JyX@vmI+!__(1t|@W{o{0YEy)2
z%w+et=RS<p7LNrHYisi6X|6~})+Df>B$FlW+pWQ-_5j(AK*<;l9Ap=BKSLBUbuK=~
zs)Wh!<h?08sz<3Zlqg;=t8{{<OBT{AN;zf#=wk<#oaaqbD49=7cf%{CQzBD4Nw#n@
zkX=zRF#K9!&x4wiE;S8wxdan$8Ye+G>~m=5Dp>*wsq|$T0kr~RH7w$%pwcy<N4-9S
zaypF~!|M<@e$fwh)NcJ{G>qTA=sAEbf+q~Y*h<W<Ya~1Lp5kx>7KR%--uKW#PCClL
zD$dV|<l%62&2{j&;xLW-^fIjF72kxW+53kju;)!)Xhv~nk&BQOoeCU2&%38`>72h+
zr2(({8ImF}aTvS8HF@c)6)MNuLSwz{WNKF!Vi~eKH2GthU7%$aZ4*wdyOCh_n%nmN
zy1Y{heXMv?P*wBZNWrx#l~$^a6yUA%j{K;t8Y5zKN5A(iVvOp&5dZ-~?11{e0!O;E
z9Nlj<vMEngWH~*j(i=n%`#_{c=!=ZVN(k9)Mf+-<X5-KcXVzL(MC!%(<u54&_8~0&
z8=r+T@u-#xW%%0nSji~Gr;U>FNGDbJ0pmXzxb`SwVeOkypzqg7@@dSnu^`-`g7@Wq
zo_KSt#;JYxIlArNE?ze^9?;Kc%NHyg+_nk36B*w8irB`KXV2;F^0Uooy573;|DaDW
zI(MIh==V+*x&UCo1->HV?bVd$TfONe_nRfy?qlIPGO#qC4<tGN5>3^6#a5?`*{?a~
zN1G$=9ycA9wCPHSryM(CzFJJwFL}WXJkjgd=Pl%dXuc$Te~CnnpB7pBDj~?g@zTQ$
z_;8Dx$a-VNyog)CE2Hzegq9RnRv~OK^eTYqcCWqay2-HvnZx-=xB1ExR+?qyawdpQ
zx})zJS@`3pIN>s^i%q5%Nhpt}>Z~WTRy91R4WpBb7-Y-!6AdFiA_>jglT)xk2Zovs
zwFelYhg2811F<eFJ&3<6JMhRVqOqG^ztN6vJgJdWK#w>_YKLCny@lPAu=&@@$`gwI
zsaPpuiLg&SI=nwx_S`9CATZwCXewN^D=&iLY03MbA(pR9FsXf>#`>d!w@icNjYCLZ
z3*SKW^4R4bD%`P#C$OFfRMywTfIxBJhBj=C^{MK$@uubbZcj2hJrj<4K$-mRxzvk$
zL4-MHoC_f~wsDV9Jj9H3G@#&>Auq#=9`cZ@-*kENOr;izL>DQq>gKfZr>DQD8tm+~
zxY<fuKH9ebh!qY%{rrDH3IE%bHb8xN-eR2j_r@%)iO4_<CuP<Q@^d9fFnbqDJPdt2
zlt{!<MjSKW8BdGkC)b>fsVrE@CGMINDWL0~@}*msWMk|K%}Q6YJO&Y$%8xg%Jx&k@
z&Wbo`yq2@f_i*-Uv989z#uB4XQ&b2`o;VZJct>0^%}Sj+tMW?O>*QB{OPv-B_HrTV
zD&L`6!mF^~JRKu^TO6YWPn7Zc3z(~(eq{Afz^JgWZ#B=fJc1*$w<v%r>^^!z!CxdO
z=SD9*QFbz&|70W?CUTNK0p+1L*XyingHi?G4b|}yVyrxO<-`Si_V}`9(E6JRDbJQD
zS?}stg2#9<QBblzt5zai%k*8F)pJH+eSXm9@ri3GYUaf+^*j|5$s`r)l|S87C`Pwn
z&nUJ{vi;&pohI-+t=>Q6>At}@_13uxB$nOpnR$PJkKURKlT9QM=G&U<L&I4shjY8?
zchK?_K1KeOv4v~*<ThRe{DT-zP!+SnvRMFA{?|PM4_72YVcp=)bkGF(yf<D;TepAY
z@^KyY(fw_Wz&ARmn?T$$rsBw0F4T?QxLH{QI^;nJtKd=iju6CPM8+xWsIqP!7^#P*
zb}T^1=!#!UT<Q?2Q$Ny&ya$GqC!S{;y=DY@xivQJfFDLsrT{5jm&LtUS~?s5Jge_B
zqL%?BOj#z3SEsK~8LOWUL>y_XVT#^ZY0FT|^|ef*gLfj<PD*E8LLs*THFcBBBO*Dd
z2$U9myp$<+A;nzV^DcX{Zldc7_0yP=>jin!L?7DZFGcxYZSQu8^lfYtyKK6AEuY$y
z<;R%o=bR<I#sc^DOgkh#qpjfsB6NyC@21&B-$R`KyPVW<45*?ro-aY|p_lgql2jXX
zPHeV7cc||G&s0LYG^i&+@0Qnr9UFD3i25CYmPPbD_$<k08o>dRZ6_+Yu8?<DY5gY)
zkGBxK!<!#I(haTMSHEcr@d!7_=#-*=;5*TAN*ovO0H@|U*>GeZ_bw0FwaDZ>f2Ys%
z%Rh--K%LhezqBUdIz3UolihqdX{ID@kT=mL<CcT^qobDYmMGum$$b~!k#6lF6D$9{
zTOc-C4{2C^VM88h;tp{VpPMy)+@^}S?T}_x#rAffrMz}KdB*1ZB^guZL9RWgz7`&_
zuJt?rG;uKngwCYZgZ}q!&y`9t?h)-6^a+$N)v1;mtziylz0YqpR?W;D?4_p`A}(;V
z7&c)y)=g8eqctjZki0!6mu&*a=JI{VI!}UP)1m>U)0+&H=}HW0t2cAyn#bpY(@~o}
z&DJ1&6BTNPe#!_jmbcr8VKxhL=}b5giKV&e;gZ&w&^7>z#E5wRBgpHfyE5Xv#J2Ae
z=68uxz^yZ+_q&o#tr%^zg4XZ1zM(z%IDbeDm783jgSfT3ec&R~!MFV&1q3dy?=M}5
zO^n)jRg|aaC)5Ky&d=;XtM^E^7#f+7E`rUnpF1uyRZU}nbz*Z0&yF|SRqx5ETRQtl
z9sCZxmsv_LuzY2F2_PMBx79_*!7#6nh*}P)?K84gvo?>8q`*7vw^%TAw2oz^9RvD@
z9tgfPJ{Fbxqvr(^K;f>OG-EtOS}DL+#X4hPB+^yXwSA+vYw`c<Zp0;%7hcp0uYvi5
zK(JRz#dUbXno?weXz<7(?CtZ)E`oOOTqC_BD|`iaifi=6`rDyVKeNX$ZepVKthb68
z%n>o&abAqZ4!P9OTXYRGIR7{A$5=R`0ZjTqH!kD^HD5f(uYDaSr#YNPI`7PzPj@1i
zbG+%?KhXpyd?FrjslnN+l0D42!&kwVf6m@U-)59szc;n4Y(6zj_3Yv23fn9x-%~`_
zpMte#O#+RUATR&Kz}X-2Uf8Q|Jv;^X^7v?JNNEHn*gxv5dMZytRzWc^4%$~}tN=nO
zSY=Tb<uNjP5h_i|QA9D<&7znw_XT=4vX7#4?6nm7g#`fAV#&!q#PKT5Lo{AUVy#he
zm^uafW5I~(e9~z&(UQT-T{%&n7VdO!_bif&g>KQ!BAyj>P5?y_)J&VV@$v$b&ChQ9
zy^(+6m1sh-y63|s=5yVgKge{ml?8Z1Ob9vX`m#a@TXRqOk#n-(_<#DEb|{xkn8W<m
z&|3Y;{^WZ3^MFR~Dboc<0bkvlYxjtL^ghJ;Bb!_BuRe&Q#603qmi+alIBddxXbH~|
zsQ2H$t9JV}8xYDq0WcMtw(U^zt-e9$6Vjcg?fQJ2)mS9nwpOwTLX?efppPpuIzou{
z-o9T4r+|a>jF_yA5nu|{C}aP3X4ibZt9%%jCDW#d=1vfNiz>lkUnP9QIU4qPRx7l=
zj{5Sl5NdU$VeW1s5Iw%B;V5&5jqCz!5_|&xHjO!k^q!^R=*jEez8L2Su0+~~kA2xK
zkr(CBdem;>OEu)bVR&P6W&~l!teRDRg7>df$lMs=5Q%p<4-5PwRHrX$OKh$^(Vbbi
z^ZOl8dE!LU6=1)YrJ_OR|B_?!y=%H|y6qB@m%i;_Fy>>258T-Lh8EMTVxQ<ZEhK&C
zIp8ftsoK2bBTB)5!CcO8RZ=q-<5eHLo}~;mydt;Kk9$O0Ee(s_p6_Lyj`v5EUho?G
zI$S(^xCK|^?ONdj=Ma|fJgARntTsEV=eiZQeIG|S2`cp;?HpviOUT8y8?qnih;&SD
zcBuU9?ea5q@^AhwV02Ns!1yb?(v?c^G4$)w!#sCICO9CB*&_`XyAn(SbpW=wH|+3u
z;t4U2zcL?nWU9gqVZnN$WjXc;kZJ@&-bAty#kX|jJlMDxfkmO(@WTHzb^!n0Y`qBX
zWvZ-PtpPQ`sTr!>L7cN7=loG$fyK0Jb$Qzf>nhY9Xu^jI#F%B<rPc2v*<Y>kO_>p%
znC43vu$~Hi^rJMs$vVMN<yl4d?%WKm$SV*2Or|1XFOU0VuKEvgW}id{L;rhYbtO?q
zIUFTWehi%#1KdEJEz6v%>U;t|258o{K)S)MBzJONA?F)<gr%Hspx9XZ_%P>Ok9KBZ
zpVUK0tWPseai~3;L3mcL8??Hi(f}EHJDkcea42o6$Yp|ZTWvyvHeU=Y$T8p8oCGuP
zp##oheZ^!bpJyTR$Pu`$!vlOvs3-bf<`?QYZcaL^Mcg8R>$CHB$QOdX^{qQStmhHi
z+T}j~zp;$}_8Eh_g2L5B(w5?XJVj-U7^2z;kM%jk#WF(NGPCgsDxZ5W7199(m^A25
zBlC<e(MSu1uuuBkIOQ7dd4}z~&?OrHubJRjvA+)B5hLQjrWhnL*Lcc0<rZed2DR2J
zb(?R6i<iExs)gLUWZ@Jn!-cES)!%V?oKVz^Dwjm#6lleLEs35S1Acf{4`3-BQ2I+E
zaj|(bG|jaR#M>~+8vX=XJ8#DrIigou8gJ~>Es$YS@UAylm`9Q7e`wtSK+B!x<dYd_
z^h0PcB`y!zw_YvAqMhh~Jpe%?r8V)jh#N6GWbtZOn%#2(&Bw;wM7{-pAozZ9#<oyP
z^0IAsITDBDC3!*KOLmoa_DN!E3^hTdDVi-g!jwq6+>75d>FlsCnt9IzMD;SC1&$Bu
zMZnec05JV?1i0^LT9konFH*&U>b7>tMZ~Vyfx77#c~Z~eC@x+-|EXUb3+QHej9%Vr
z50pA!jkpS?z%fa*)VvTY{+Sj_grN+=+)p~ZEsE+Vum23#fWrmVfi;Bm`<Ob1l<Gl6
z+sNy%yc|0}Sps>At%%^cp3~NDd+gJ>`d*V?G}cO9RWxeHdww~)Fa>1ohPk{c+U{Tg
zUr%zxdFsjLjVus84uq7ICr7zi;|;f>d~%cuw82O443+x+))EC)0Q!=-=yMp?Fd;C1
z-h>b9DY;eHO<wy^3-GJH`~bnV&kD>dJ_4TTrmoG_Xkm`G{ZRp{^<mCoL1y|4vz5X0
zq2Io*ZF1bDKcDb_TWf@We(XAr#WbGhTp~yO>r=DSSjaTTV%7B!G^aO@cb-9C<5k+<
zfWXn~o$=EeE^QUX5x!m~z??JaRyj_2ig5kxnnf&xf=a8~)MoWiJJ-p`Bk;HvLGs5z
zMi7q2JvcnNh<iNyY}vf?pw8iu+_GaQufo$5`Mp{QMbig%H(T`cX9m;4$@i7WBwy~;
z)t!8u$O{tWJatG)pQa<QpnX-^-!XjG->$jk0To$&goFBx76;gw5UYpjiS$lA<!YiB
zR8kHQJbTo7o>1ZESso)A41}I?JHSqzSBokwH^eg<1~p&RK0dNWIacQnN68zU+e{Fc
z4%f!FcoJ(Tp<xjuG8>FI+xZp?$2RTTp=PLIi>fw-sJ>Id;CgU*Ws|-0{W5^y_+P&{
z#FgG}Lim=-qK~VswqNU_gl~|&N4PqOv<2VFOh|cbOoOQXWbl)Wkf-;SZgvMKaL6oi
z@f7*ack7YYlpuXtn^?+|dta#A^rwA&-?TOGSR+<=cx`PWIG(vnBcIw{mQT~x)g?q|
zFF$I11qOWmc#YPVqC&=N`|k(%3ba?alRHMqWU>puqmkYgs5ca}6r~rSE^UHZJiGN<
zc}M^A@xW4?JAQbJRi4rW+HCxur4==&D8hZGYPl!1Y9CWS1!z>&vP(>atM6PR#pJnV
zzhc-}T8D!T*_cdJYwnex1udRv=Y;kf6F5*UQXSSuV&IXT1{6G657a2S%9^d@<Jbk=
z6&aBZ_?%>)qkLQH)GTY%ACr8&rChGmXU3;0p1;*1GV$st0e^#a;iB<H(>kYeuW!S3
zDM)Ddd&hwpi{aCUu9@0wL9-|3JNhNFi`7)M{W)vPK09%7w&szVIGe~TeU{q+r($u6
z3!Y*h<wdCmeg5W5^ZNPusdKLh*<*b_|M6Xd#SGQxXZ-IR`Uhzy1O=NsE%*-*I8ErC
zD?0{<2-oV+Z&**4K`QoQMaWn9TD`E>X(E0auY|S$2^_!Sp+MS$bVfTbc~!&2P^m|J
zA%=ufq4##~aS`a2qj`DhJWX)Mb~=AhVRjcsIE>=PyMs2bE(UxGXbL7oZku}2PYBe9
z;OO!CX%AE^^K~`)Qz%)ZepFX*oeWYR;r!fiv|<*#1;+?B2Nhj4v$W`NF+9|#XmM-X
zO@!HSIuMG#7d_u_t~tRb`nJ$F5O3(3T`s*``o`8%dmBp+R%NoBY~u;Kn;z`Tik-vq
zPY7R^Y&TS6$f+dunDDTKs&?!udx|$%%Q>pZMz62YHToRxfy=`wbK*%?e&KMbQM%jl
zqmq6Lh_AAjvzDD#(eR``Ayy^K2;cjG2)rM0ckS%1s2u=E_tM@i))H^sg4zRvBFZ3b
z*g5rvi1t4WT-`cL;S)|S$GoYfK`$u&(Gv&fJvH5WcuAAU)T@uHmyGeW@rhWxIZtx1
zXy0P$3j*(un^dn%-zE4Dmgm)y^H%de?jXWI5bs8NpZ3X1>u1ap96q0wPV*km4a&gi
zm&LLQsZrCps`2C{X<Nm@skh6nw~CX3#1ghxPGij7-4?!ZlrsaJfXpn}-YHsM;RdNI
zeeYTK>-gqQy$YSb(9wM@L(}d`QG(ptOs#!GRi)2e04=9D%l#k;No2sJqxt(k_aj_9
zTtWwCef}t=Ebn5z*lAiI)S|@V$AP2o-2Ir}*oP!|o6H4-3F|4mXbAaeB3tbXTStZ#
zb3J$PM#7D*U~`D5c%U6pEx2iQq>3W~2fS#t+&XS{OP?bcG8wOKGM;a!uvPY;-~P}l
zGpcM%k!N)`ENCQIAmxT>f`37T!)^kqch6&Q9euHfQg4HCJLm!{hdDp1*N3aT*r#&e
z-DWZnn5dPyq*ZD@B+a9BvHd9p<&fC6Vg6Fw=<K;htV!2ZiP$gB8w6h^{QP;xQ4<&T
zg7+uiJu&J0YGS-3-!j{>+hxL&;(X~V^!>1r<WF81>F5`#Kfgq&?VDR&^4L$;^z3FM
zSFZn+XKgjNqdlKr)VE7zUBX&QJ~|k$VE^n8Fz(`CKi&I=+1IY+AYR7&>ML!k){oaI
z3Y83LAD5}^xfJyTm*okpJoVYhtRQJJD170c2boAJ18Uj<Jw~^NSknhLp?WRMi#uw2
zHf_fxCk}UJAkT{J@9zuTvXx_837nLmzq)yRE`p3e*f5BU{Hw42$Nlca$H2uH!uZho
zAD{jG$2uu-ByjpE?Io>9ejShhIy(RTGp+%&G_;SK%oJ1q>Q%<Ea1szVfuWt%Nn|dI
zH?9Pp4$CXUpw;e2L@yLfAM03?uz%mZ{^4&+7I?2bq{UQBe{o?Nh$P<Bqdu+(Q`~y`
zE>|{E=Z$1-5gd<5`hf~$<JIOK5&ipn{;Q8?OR*TR7E2w|-~Mm!uz`sQdBufL>)M#A
zaDyir;z-s%{qf#Eyy4%!w84u>jJfrZU(xrUn#jNGEPsB|o(O(z2RZ%pzq-2r*BeZO
z$rlLEFkiN8FWs+h=dXVEzrM5Zovs?9P?~G>3>c;TQ$O{$mqD<D9u`N(f_$OGuMYX|
zf7idbLUCk7uL*O=tcoI8{^GX(#e(~<OVYWHtm5@Og{yxSd4K!6{?&z<6+;$<SafF6
zuLISFX$9=~1_2?}0W<J7GMv2%-%78fmH%;hHdrtLNG)WLG+7y|sYv@~l<|MLZk9~Q
zx(+UiVf^(A0&NG(9-6@!;UwZvgrtBNmt6YRUy211I7S|jw7C7oa5#m;3zNu82k|E!
z-lH)Ch;;}IQPs8>ru)DxNZ$G$7#7fc3quf0N{cp*OwGBc59{aydA~p{Tk#;1Ox;qB
z2S`76oA|7f=@+-4bX$TpzW^M`%jI*Up|jlryfcJPenIvkKBXnGEAV#)A#-{0%RWLi
zHGv`O;)mn~j(@lX6)$1wRU{_&-uq+eL3s^6D25J5nkguUa)sPa6}~VgBsT#EqrNKV
z<G0zy=!YvHs>l<;zf+I6kFSJ7j^ZesjaE_YPuyhqHhU;{4ML+u9>Xx#8d~;{EMCZU
zL|l>(2L)yBuNQeDsoUiP9KDNs0N|b38gnP$(oB(Jrq6+nX8;H#`pBf>cE_Lv$ZFqy
zpYz71d{Rv0M343Ewz*a^9R--u0zJ}nJi<Ri00(zfd{_h23x$4FG4E`T?cqmEfV{<5
z8bLbL=+FwzQi`kgBPT~N5ao#(F3}MY`V!HI^BCd?H3-&AZ)~Fe5r+`0gT-ZXA^Pc`
zJM649Ccxm<D>ysA;2@8Asd8sr$Oxj9bAgG}zf#l})s4`SCYu6_!7Ad);}V$iEW&pd
z0Y^hzxEgigLM)OP4Wo+0%Se@nr0x!+Cda67f9-+`(?A&DSPC%25ZvhD#)U^5vUA@O
zX>!yc8E6qA-tcIcVqZK*k`MwtzYtkd$bAR4hAt=&g_75PCNGa3AKBRIRV)y^_sb1D
zz@cRtT!9h!AmZ+aH0!*k(03~Butv|i@*v69<GuJDZ5139!0IVNtfUZL8xo(1i!r%?
z@NwV-P=>u(YQ}E6M%9?a@OZoG`0Ni6S-#B>QvevZz{~xo{Zvaf<v;CR%x<vE#O^Cx
z{BtQ%-Wx*#jU55q=jz~H@9%Kh%Ex!sxIqBw%UcMYt{5h6HAUdaujeywzI_CIlG7vR
za4Y<?#M*LzpJSOucz->izY^i^nA)1Pz7n{)b)_QzriV-aihYI4=1mo`{v~kyzkyTx
zh@zZ@^9#O*ukA~}IbgWpJt=AOr_h&|;Z)=`g`i;h->(ONVeWt(SZo0#w_V*=ILaWb
zH%Ykx|02K~w{3!hk%IS-_RYH!i#)-<ya1lOx;SAda(fqaFCn;+r8}T<9RL{z1MOq)
z5bM9#D}QaQJ`Aws2HjMA`>&nnHU@4q3xT`1$A!?l-JyC=JM^>FWL0SjR|hPdz{NNg
z7>dy<_BB8g=~@nZE%m=~C?|Z~S&0}r`(M0~)HbvZ6syMo1eAjaMqgK}6NqyoN4p))
zugqHJVuA?SWkmM>y!!w9ff)28IR^18vVR-_+(Fd<f|9`_Eo?i1B>3_pBbZ=ofNxF_
zFxZaKzM!Ptg=lKc0oZ9{RCzw;LxFB5-hZ8Vd>Nm6HZJG;AVR8w!&nn%MvA!$Hi`~C
zfNF;Ga0108)OZudB(VeUytu6nR47vb9Z`5uZ<T&~|8?bq(pA$ukbe!;I(ir)Y_lbp
zKwjs?{er@xgw69K!U{t+BBKURV<e;;$(Vgl2}?+NOxmj$99XL%;CY7dQ~+VJhmG$X
zz&U3|G;t_ygZCmlX-kMI$v^Dw5DW#s?Zn04vnrSMH}Jfetpw47$W^8Y9pWK?wS{BN
z@$qd>(5cZmO6_R#TgoB~m#<}VL?<WbU|^C5xK%qstz<s})U$Fd(-;mC#U|3j<3Q(w
zIx2AI8vEqsR9Dvx6AMGYmeg<#!|~UZQaOg*XQ-z!C+CsS_Ha$#t3}~UAeMMC{3RRV
zH7C;5+8Py5lH!?8KpIT7_+CQ#3^tuV*h^w(>-bUf20z&b`k+{y6R{Q<g&hEYSo<}~
z3C{v=Pk0fIa1dC4uq1of{<H2&m$nW#c}~hnvFSrP51zK=PcMtw&1QEiOb083QU~k&
z+%)d@YOkM<e%Nw)fqeGTWgXT{Adyjv1bOt-$FN_^pMy7-kMLeCxwC8%&ruf0rxUXC
z75O|KXf7#(JL>Kjz&y*-cW6@UL`)ZU#5n$hEIZC|f2&>ZD&;M$;@o@$Ql7g*f;iL;
zKw4S<@?g{L`bjEpb4dxkd1)_=kDAxTak`{#j}%}RnTdotlN+LoO6*$<k|@P88f+7^
zE4&@s545D4>Te?kV`^O3NLHln9g{~3Ip0ER_L_DEfNa{jR_~r!G}r<m2qXmQPwkq@
zfSc!MF1%wpa#fXY6~$rF&&~r!P%D;msE5nS1RsKmUNcrv!MP1{G<I2A8&A%y6VF@`
zv=>K6&071j5WTVaJr3?aPy7EWSm2+X?nuqz&MhzVd(AqljzuU-V?-}XLFduVHN@t@
z#sdWXvlwJ<ocyopr6%3iAinq_KO5EaW?EDJTJS{}61x8~$2%||saY$aix#4>fI|3G
zAd}Z=LZh*7IxveZ0tV?P!mtnLFh%(ZL3G<h>+wU)qFwjg<$Gr^@kOaYUi&H<lzihV
zdzBmfbgV6iU+<UXI)3UWS3C(;MP8jTw1__EWluQ2itz$8147mAV*y(qVY-d*8a>!k
zyl`0)Va9gd)<)<y7Z80so+E*dov3;<kKaB>*S{v<-+7#RVehw9LaQcL8n+zYF52yJ
zxunR)843oMe;}-4p_H22E}U*Hi07Es^&LT(SUQ}s$t|e7NE&Mag?XqL9`RI|kM8vH
z<f9jsXs;E}0-k*G9f?A3<aKNhR|f6k5*$vM%s3Yz@LnCD9@=YkCp`!~BE%KeKyXD7
z&+Z|<ZreX6<$VWeZk=xk->kUwoWS?voF@Os?762S518oNNA>0vpgkY#>Rop)^-jSV
z$6dDp5U4P_a*xIDN<9hAb;6c+FdlsM7(cfQKy=c27X+Aia^g!Wj%qL}yX!E6@bw*6
z8ti?sS`9fHtZ!bm+QODH4&%9C?mP%b2F3Knb$7d3a*>NBkHBbCNy@+92}`|ym3alu
z;#hE@G#6rzar}&U%++!E0|@F>i$M46U-Nc-__Gvolo?vZn@AT4t+mQ?WqP8mPQBOU
zezJ{>!D;mLGxcRe&lG_Xc4B_$(pTf+(uh*`u!0K@!|Ye<F#9GfCBEfhAO0AFoT1Z`
zq~*Ts{<*CVC$*eRfF5%GMhsDE_lU{^dbkRzvKQM661)m`;29>+lH7|ZT!!ibk1oHT
zX=EnHJHvB>;D<|FuBGtfmV+kzl}Y@)*W0ZhH=K{RoyVI_*nQ2L!G{}O4rp%eKJkw?
zNDBGY+<M{xldCpuJe)twKI%6qNaEnIj$r5iu$RM;@Ot8O`-GASI~QTO1df*SN(@3T
zZE6ZJje6wUdism&zs|P*?meL87&|^8_pgZ@1+kf&!o-LOF&PcRvK4AI@TIrIh9q+Q
zPeIX$^Yvg-Yl1GZzQM@bK~3^tQg?zXXRx480P`}c5l0uPol9_~ptr28Kl#AhH-Y$A
zIclfrQ~gPIZ@esGvJFO*Jqv47!%1)7ljn{F-|e9H;rl)T5i=G{*g_Vl>qQQS^A#qj
z(XGoe)H~P<^(T48sRrL3IRGb5<l8}=;ojRX{G0m?*PMb%eq%QuS4uf?W@j=I?-cPi
zLug6KwOyb(Ym{B9n53==_P;f9+nh-3ITWHys9C%vvun3q=jsRqE-9BWUpFTb-s@XA
zGrtgnCBmm3^624l|M?b6SDk^1%;#fNZkW!;f|P?}wN=%%FX<Bb6EzLp@sfzi-Y_mJ
zUNa98l973Qz+Rc=?`kQ8x+J5v6NFTRiRMd~_4O}lydC_c)cPY!yl4i12+exkibHf_
z!oU9ImG^-c2nt-yr-hsDS2~PJ3KX^TZ<=K*c;nF#A4^bAlYVFa{B7v*#7L@ex!TRH
zX^nd@GaPn{;+><5tIp}uiZsNw$YGtqwhk~aM<!^f37qw>CtaSfse{*tw*%|lv0OlG
zJ#t+g8>pW18(;Iu;ViuV>{_gY<6X6SrItoDOAez8fy;AO&1r270`c;s?E1$qP(g9`
zWI}kga|?>nD}IKHl#@dZ4{P+UcRZqBdyr5hu!X%qydISsq^>lvp5lu)MZC>VzLu9|
zC0=_}ij(^dP1=wbVcszKJ${D0=v(>!;q1$!q5k*(BW0wTK_bi8vSg=F*>@$RvM*V(
zWGS)@6OqAKqRmo52q}B^r7RO=uk2fvB5P!q<@bDl?&s6J_j~WBd(Q8V&pGp%(>Z3|
z^ID$I$Mdlm8ou(@dB?+m4|=^h_}!1;$B)<kgSjzNMg0%I))suk@A1_fbST;q-q6fT
z`!Z2wi;N7B!S%dcz?R*$L?A(ihO;<=*sfTukLLTG{=aUZeDuI^GmCX7a!IxXXBO(E
zD7?e|;vB(1wo=2>J?I~Kt>qdgSF3`j=jc!1@o%k%-hTVM*6>rsOKUDcy09+jrGx~>
z>!$K~5!a}Cs>NurJA-fcBJK&6P6#hP*U<d9qw8BG55>Q4yPv9Ks++>@|Fx48q<Dup
zjO)}jD4b9{1)RrMHE0Q%L{%9<=bqe-D-;)R#(mwL7Q#zCdQ6U|fuwG!!Z^pulX#Xd
z=`?$u@}c{!Y{wS@4!Is-o1^JIaewW(_8>p_3oW@|4&dyYC=@Lv7e4gM^6d>s#Y;5P
zr<SS5ld1x^Z*5IquA`iHI&zO&sU`~}I<6|JEGtWLmFb)<#ukVak}bsJX&ULn*!vV+
z&LnZ(eaYR)m7?Gts`0KWhMOo~FvfI`x^{Oxof6N~{a6eO#|w?9-ngjzu)F&{I2Ng3
zS2-D`?zO0w5#=*?vn{>KeC~OrNoceb%VD8A`)z7{1iM7LdOx<bq~c?oLR?sjTkbiV
z+fsz)W*b7@<RwD(nb_`TKVuY;cm)I_Ft0<R3u<hOV@s1HIkOt~Eim>Y-o~{a=$5|p
z=bTPEP|+}xzE0MH$!q~hSsVn8P(fn|j>HaWflhvkh~=xn+%_kqbK+8$z9PJbmxqR^
zEZ%R#Yj<?k4VL7;oRo2WgRNlw{{GsuP+e7LP;u#MC#h;Zcs0WP(Oj7n@yAu=?98Fl
zU!qN}A89Fu!}#RBwgNB#{HZOoKr-d9IE{A957E&{a=?9-B@f=VsVpci@9qq4nwnx}
zrYA2CX|fnA_|8KtgR%Qt@c^&#JF3}lC&i0I{sJEUODW=plp<~$7li&SMS%V{GE%fp
z$EK6kid)9g%y0cG{VP?Dry;Ee*!W(JzN}<kD69_|fC^EMo}RtFmYHEROh!l5+a{w;
zkB*#kv+g*HxiiL3?`$EZB1Pp|b)q)W0+Z;C>Y?~xQACi6fx}EjO4EQk;f8?T#Ctxu
zdCdJ9H^a345s=+XQ(nIt{o`3%H&@C@@K2^ge`b>T@~r9I+~NloG6uHJeoa&?#%qK|
z)Q4N;u77u?!&N{DEazsRxvF8l;JG<UGXv#*BdZkahe!vH><MF8rRo&qTHnpVZZ3F)
zdXkXG(B?k*`mH(pC77+doiz&2lSr%G*AuFwAak+cFuT2qhQZ5N?YYa~SxHCG^QJN(
z`ZkQ)T%;*x%Qfhk`<CCX%uI>D&3m2{A;z)mnaKS99-~Ye4vSR5a7Kfk6X=R~_NlO|
zz}2alKBlzg&W3-!N+>(8`J&;P0e|k0c#3~(T~18Pd-ru<l~J2>ewuhO+ien3SuiD(
z+Ut2J|8PkP#-)}(Rf9h<-3VpDy`Q`9%WP-ka&$R{fpdM?)d|nHLQ*!LkSuv#bh5-H
z;TNp%pODwL6$+l@J==<^|8h;z&AXJ2k4gdbGts<<nchSRyOU=R>a<s2u5VPfZhFE3
zO37UBo=YIzd|KurDs_3|SxCDC!^KC`%guZACN-T`j(x70dlVkfEW2J<W<E;@kvqXq
z7Kf)OOz`Z>6rAqKw{`~Nf%s~$4VJ%b#(1T=iY{0j8STw!<+%5%3XPL<xV7DIZWOWE
z_#7F&x23A|OL2Fd^W%^+lB+=CdvOZySaR|;FvbvJ%W-sqW53fYt<KXG3hzvtt>^B1
zlp)x6Rtlex*r|3FHWXW}G?d5<k&@WB0-d51`mL$Ncc>~E>XYzUC(vxhF|pf7-bKqE
z`f+jyb=ECyHFXt>@Wx$s`|uEPfV5dnLw~!@@aS<!tRzGD<M4;gRgRBCS|u5_<3d)C
zGjKhoKKH34XbV<Q`Y^e6`(><LmyDowMH+o)WKJSmgk60AwNz7I#GoMz`<m-AVp|uJ
z&PK>GkO#Xq1#voQO_J@IDUEBAiGUjnTRUia)annz>qP;%$3M&3lu7j$or~v=bmYaV
zW4*8R7I8Yg2{DsI>o^HtIk9T&2q;c&S*(JaVtSd>$wkbtCF@4Zbw4^Me$V=<yD^=q
zsRsuZ9rH$_yP9U);w_CjN^j`A9w)zC7yKYk>tTy=Br6M$=gJsQ_vAp)JSd~$QhdZY
zayq}?`P%m9P@g838au<LL9XnwlCc3R*@qsCvBFa=Rf)1j*I5fz-Aw;He%8Rt^LTv_
zo?^eGaIRTV9^%zsi%-Yvw+f`pxe1FO{GL-uZDC#Rcq88dr!!(!SJhJ7A7Q*vcgx-i
z+~Zq)MUFWLT`Ngx{Lh2u*E{V(2v{Lhp%wing(TnxNN{K*IG)*ya8Mw@2+hi7<F<E{
z6jrsb0IOZ=TH?$Zf-z!}qm;Ko2*ehMqE^7#To}o}E>P8z&EcV_1d)Lgl*dH?K<-J_
zFS?CcE%mBpC(3-et9;hzY9L-|KY21ZFr>M^9!khLe7e)*bLFP)&0Tb>qj9lm?!x=H
zWNu!^IK&8lnITmc!Kdl+etywS#O@S#bkYyZ8^ySFpkl<{LG)|g19hq1_GgqkUcuiq
z2It?Fy0u_2|L5<N&y)yIe7PF&22n(<5qA(zd!*}Uf1VsU7t{pQJ~y|v0B4nO)xy-u
zqPMJ_Toi&YSb^B{n4bKwrDNE%;&GB4%F@8lGDs9llUm;0E!ySit$lnHR*L-Sk*+lP
z?S&}CqmdX^%L7V`t4?-vErKYDSjdMJ^^(2hD%8S%RL5=~SvZpcDrWN54Bc&m@oaqg
z35r<JN?@PVeii`}nCdR&yTwf}uoY&Rc|J|{`U<fIQ9je>RsM1LlpY>2GC+|nu6Z`N
zB?wUnIta~0X6f1eJyaLKgc_zf|IY&lPsh&k1^6<buh~m5hy>j&B^YKs+*I~Pz0ruP
zHGY8a)ErR0I6BnRSB3b9Q+=&qDE3^mJ${M$z&YaCTgzuqBWHVyHp?hW3)E$xIY{2m
z(D^VSclGJJ_KZ0XQGzIU#Iai}UTR?@s87kK1Z2IyYDEKqR3wx27k_lq>AQkHSId-m
z@#5wB?baWzQ4t(7d?RA-l}{=urPFdg5VCdv3J?`X^5RsytU8;E$XAScoH%00Mylcs
zK7|7$%<lRb;912hQJ>b;u^&$9)BUn(PZ3D{nU-OrxU93Pjho=7&09-hZSXNU0Pd1W
zyAu^U_Lo<4nCesTn`cq$Q$wZ?pi_90o?>l3;_xW{yLa(ZRDatOkLwW}N2Nyss4O*0
zU6pVZ7E-4Gs;_(Erjta~Hr?EzVQ;bOU|19uZFpEBCkKeuq$)!m5pecqEQ$CGs;QH%
zZMQf&PD@7Rt;_{wAHr5S1Z`Q#dRAO8T8^Pno+?+bKC2pxc_(}`8z+lT9y)#H`$XxG
zIE+7O99;#1_qJO1o3%+WF)`xJjsfl4lK!)1m~ps(tjUDxUl`+m!jI)rWYa^P2D88Z
z3A;b0U3P&Rn^?q&p%B877n;$SbHO$>q-(5#HjrN~iMqoQU@8rnCk|RE`A(Ld1b?iR
zFp*`TNf)QRGGrV%sA+b)=L*p3UUK8#%Bb#37|_%y031Jh8*{-^Eo^kRiYiJ&fSV+M
z-6yhez9qT<etvZoR{%3812XaDf&!^syD9K<o#G33Q$q3nS-5z7P}W*aLOf=_&-2bO
zB@<ErkzAI=ebn-Hzf4B2XW_BDYPw|<S-6Axy~R|TLtumpzuKLSijT6Wb;I2n*p<@@
zcN^X#q=1v%bu;?q+3v58KrLC!7O$Bwc+m-yUlx+h?&FEhR{Vm4J10b<zKSenYH;Y_
z;$bnCY3<a^<3awacLZi6M&;42Nx0|J7O@yZEA>-R_V4TV@M8IkZ3UNmX3qw70ds`B
zywaz`B!sgO*aiSt>7op^SR&pP%j$jwZg7$y#0hvCGMzRcu{eLL+4H%h7iyWV(c5Cr
zPyS{D9!4g$;qHVN9DkO4Ef}Jb0)-KXj@4?_cifdQL6lFCaM*>sa6MJr9TeFywkT~}
z5FQ2^AESi()k(pk)Vg|HnU67de;_Q!^RGi>bBtxM&mk5d{z?^=lTq;~!Alj*U{bW7
zB#9N5dcUauWyCgcm~&P&%fV7l*r#XW{9*Eg=%gbQc6tR2p2xz_^X6YUo;-k|aE?|l
zao@b31b|z-xQkg(>XQ`O#i0=6L10c8a3x(5!j2;xM<Xm+kEI8SimNsgpD34(n^Pj0
zJWG=WzN9k|R4O{rA4#Ef4*;)aCtfS?hIL(B$r&|GLw>k3l5;|xOAyY`AI-MTT3pLH
z?mGJKLPY7&E~uP^LquYZapO=@Y{`9wiWDb#B_<UOq?w$*G+|k)EFnX39<}@)m~9ik
zJo?zcpcRw^HlZPH`)4o1+#`^(?X369gzx7#*q8aNOfADPdWgMdO)967-A5Q=B@Hjr
zx(<U)-}&y-rcr|j(M{_au;u96zg|kdcHK|)A<z)Q48yHi!Jz+RB@Z|59*NEluT%|S
z>hT4aA6lV~)_cMpr{-7Y40sk}@}3u!3Ot(Z{ZFj@*E_80Fu0AhetG}T4f*sF_)e=R
z1XSlh;?aAUYDKfSl4wR!SAjVC1W^Z%)a@WFJCy*{@*AkpwY+z1iE5aHDxkrpxh)v$
zTQD{$D1;b-TU7d4Bxd}}#&oU@hKsZWXg4<$teP9k)3I44Lp~fVj+Odt#Xtq;yaxk*
ziJQ;mAcu*TLVc3Zz%BQJYgl>H1DfP?k&rEbx`j~vm=8~&z@%Cw5&Z9-&qE1-=1cTD
zATz{RO%uXNR1ac7=_1>5`v(S%#Dh|iM>|l7xWNR;`wWWcRqgbx-(gkCO(+b#RxNAa
z{*?X4vsQvR;{11iwzywdltAVi#arMsYRV$*YI61Fhmu|u6Tt4KhSTny&jB3DEq>(_
z8_{n5F=JKA+Z*2~H)s}`z)w`4w)!I*B$1wt<BHf#LL(UZ^J#|5H)*bmICD`GR3QQA
zB6IRAXwdGl-+h{kdh<xDrI^daJhg#vJaP+DLJaW&r+mqYPh)a&xz9ZhNyuAOP4&`V
zJ1pXcazeP21tv#$fgkDdsAqqFuq-miY*I>Muykz14pozJ%52<Qxl;F`mXVBo^ty^=
zgyoO)s<96Z3x4&IK;lu2&Rc(C7iqCtu$ao%zXM~Ey{!ato@UQ1UPsCliLS$>@uEKY
zE}4xx@+uLtp>+ZyxquT=XilB)C{>!-r`Z?P-4!>_6wz%WgSreb>4JN2u~CJVWlKzr
zkH1Fz#wRat^7B71l1u^io=&k6d>-d>8Sj~ROMY#5bJo#thzW@NqZ^-^jW=Nsv{y_z
z#b-!Bqd6Z+8Hubw;`$M~=0yj=Now4paIMdM?W{l~3AO;za`ahDCO1`EKDa+hC`-aJ
zATEZCFE=eXBOjbC4)X|EMb)oBg&sQ}8022N_hLOvo9Sk-Kf#Yt@3dxKU`r))G=2K+
z-<)q)bqiYKFFSM>{#j%&)4Aw|mr04cT?ekDBpy<EnuD~**;o*sq9j9h?_CaMC-BEC
z0BewvtC`{DGYGypVP<MI=i4$+k+dE%QcyW(`EFC$rMC<>JQO?w?$~)C$ZRyNv=qir
zd;GYYwbEsB{Sh_BIf!0A4_pI+3&wnxosQMLC3u3aH9jPzrK(e2Egr9WVn@Sa?rV$b
zH=DW1`c3PX>y$2w=ojN`#?If`lgG)$-A1k$o!WMlnNMXE{o*~R*JMhxddRxIrtaiM
zbGuBT#nez&e=p@yV0c}#Yz|~yhq$@9rj*6;bkx*YBIEC#@gZETvQT`;D0K+dv&*!(
zEB&?D^$#Yies;p|@*xjw$aMb|Bw?(~o?X-xLSAA+Voso?oP~u&8ET3$Z-)!dfQ^`-
zG~2pOg8>X?n}htulRcV#RQT7(usjD_6PXBs4X&Z4(%Wgb?n*I;KMFT(R!y{7{S0H(
z<Vay2@zpDbHpExB#w=j)Durfq0FJ@EGW+&-6SeiCo>q#@yy&#6!r~HWN9OT%;2YEp
zSecgLPBb4$oE`np{&``j)VY{7|A|_O%q=({BpJ41=ojFwiw}RRbSqt?)IzGcbLrFy
zT+TF$TqY$ek!FRCFm-dz+p|;^Ua7>K?|78qaW4!hnP<4i_Qp5f;un+5r)ySZ;C+Ob
zIItdHB$0fVXL~{P8xqwO7k;NXC{OL@$M-=;Jz*joB3atb+aub_iW?5#V3s^6{eDZ>
zYG3h3<&yYt@3U|J2*<YcY1vaUp7KFd9d;e6KT4}c`_+oY>Nj56O{1io@+p3|8i(ha
znK$%22bz-0*sC5|gY(k3()dT$Z0RgEgfbe}-8pwGjJplFAiTCh$SluQlyu$qt?w#R
z7&M$1z+Y#a?)9X%EQvoBxO#X-s5p3?*kZR`hU0HOGu0#=jF;T+_Q-d})yK)M|4@{-
zdf)XoTQgA5QmsfQ<`xFgwOK}zE?qq^#yU-)(<b)T{xqPO<+Bn>zc6=1sxql8@3ZL=
z%SMtH?a&7tgR9uf1Q_O8`eu`8yLwQoqKq4a8>Wo*!g&YX7$keNW?E@A1$C|Nm0av#
zX%VPMZOiw|ORNnf;kgC94X#wN&jLI=l5mt|^E=!4%3fwC%uuGqxI81(*xlve<q*Qw
zm~Y4`6Yj_BI^(bAS#`^6I=YUof;uL@*}RF$aA75WiggJ61Cdk^;lTv?6Hl)_Z(?Kf
zSnruN52$71jD+SyIQApBI5>{g-)X9fk(<PJ-)7|Z6|Rc1h(NC*eIHvG&5<&x^m!UO
zOPX3S64ofdvOL1!=gkQ9cZOOhq6>vsC|*$<9d(3YLM%#o8WlA>C)<RnXaCrwNxXpJ
zVs}zOg$zeT<)G3B#w;;A3-_4wxn`R*zpS1wQnatz%1fR@X|Clpsb7&Jw+kXp77PSJ
zi`VS3Pwb#lT<clgoBl2oWF4^Gu=R(O#2Jt0FzOJ62E)%^E5gWvo#(9sDGvtrXFfnt
z!IWs?{Ix0fo;zdv6xo**ACek8SKSX&u%k>cm|7`;#w*y;*hKbb6;gxPZF!3QD4CuM
zwufptDO7L|uhFd{0WbAyx4?U176K2bZfnP}3x6q_ixxt$_2fQ$d{sS7+)*~_>04P{
zCbxnZLg|C`$Fb0@z~;%l*4esn&gse_W$|E4ss*zK7JD2O&KQm|9WzwgWO<uQ^i2V4
z3zLi4uV>AyS*)xzdXD$q*Cu*kgwr43IfSUROJl`99_#-e!Qe#Z%>A_m-P$1bA!1g2
zCM0feMS~ALosuN_Ew-L(`6V|JZm7ukJ<&OE7i)4Nek<#1SAk3s6xLUaNTs!E;-8Ja
zW3X=t{-MisBNOKMS$Y-I1>=Fkf9!#BZ&ZKYc>QUcZpzi-!-N3hPwTT$vpC_3Yy;H0
ztscL6U3};=GRzdV5qN<V$(Q!&_9Cv16swh*3){7Q?<-KtwmHY!G&&8-LryD~s+o@q
zd{;$vuGLU|S4Z{bse8*GL<Q}9Pf8iLFK=$`vy1#8R~xm9KWzrb&gO{u?%vZBS705#
zYS6_w?r;vZAMX55IKw`*ivD*Mx|x{S|2um67iRX-Nrd-TDy$dyk5<WtE1(e4Nc`mV
zyEx11Qt8;T0j{*}`&rs_oQCs9!2KKEa;xV+;8Mn!yOQ|mbz~`;hK)@Y=HpvSX_7w+
zDLIZVE06*$qBa&oVoAoO?2JkjYSNesBkRHV1ymQ7lAjb}O0rt;o2vHh2oQhKNtV2u
z^AJTSl97#Bny;yPb;0<ie@VU?tu;qjPG1Gv;m$sWrWE;gE)F#UcfZ3HSyK;srU8$`
z`ZgLSM`;i%QyNNKMsMx7B>aRCPbmJ|=bns+?+zYB8;MVmhH9xMedn8GGH%A{)0W++
zB6^-UF8^Gaiz>d_G&1ycWUZVEQ$*B^;j=eauRqy=gI%b$i&QYxYt=n!`tBvObF$VX
z!R`~^4Xw{?i=y?P#u)WQ!!ti_4xf&AoWFHXeo@?+*;@9MN4Q60zj*O{*W0)TTLU`N
zi9xnD1V;CWSj}(gD#6AKd~Plc^KE0X70#gBI&)dCpv3iB#`ny^qyz!6%dc#%Ok5pu
z8|}j;J;L@cn0odTF?aol)5^t_ORhtU2-(HaHH&we!QA~wZp4A9hyz#T2U{zPN#Z@?
zZTaC31GTDc=Nn8l`=gd*d$(YL6c4nXzII)<#YFiLgX{KNpYc1hrQ;+S@5Phm;cnUY
zyiyH=i+@$$|M@~T93f)9l<#Ev;|vTiKl}7P_{WW~9}-EP#x$hKsx23Ke#F+^r1gu7
zszi~2N8XsWv+cdnK<(_(e~@1HK)f&U{66&6qm>CXl6sjxG3;5g@|^56XPI+zi20ao
zkC10YrP53Hr}+LJ(q|{nMh@DpN$O6OKW9TsJnf%=tG5GVi>4Xi66AvmnJj_g93HSq
zQ|iwefS)Lb1iQEpLYsU%roO$<yxIPNp!XU@RCr1Ehs}tcqRg{<s0VUBkp7cfJQd67
z0{UMHG1-iiZt2=S&y~lIUtBP&Fb<!tO>6$oC-cvD+NTk)Z#u+(*FScay8$_2n@eWy
zFYd4<3gHg7Klzh8tS7c~S)gPcNff2wQlrqpb<joO2r^vFV_=?S2`ypzThHPf(vP$I
z2~>15(65w0pV4sG&?3M#{N?&KcrDi>RwkIU(5rt&z`&sVp!pk+URVq?iw@HmafI`!
z0EADVOa{4p$Ui{%_QX2!)M#??p#QG#2MGmQyh9wuTMZP(;h$vFegFP$Lx{A-@YM_j
zj(w&32hB9#FqgkV<Xtbg#RNjfM?+@8?d^!A`_HAxPgX_O-kb*Cp5Qwyr#^zrFyrnc
z&z0H!Zln*w@R!`$*%sO%d$XlqoD}HUFdvlA8wKL<$OMp7O1h4cE^v>P_JM-)!1o7d
ze?3`f1|3}+^7Y>Tdgs5qXn%k_{Remh=YM?0ZwnO2)MZ#mq`w`EQ<`nT(Pd`xnrlQx
z9asRR9KH7JV_}G>-yrwH%O$(alJ9$OT6ZDWGmbK%;h!p%Lpt!M7tZOW{55C%mqYT`
z`MI<gIW2?D^_hRcNB@a1{pKAXL4+mk)cwNpzkIJ>Sn}V$s0u)+s>!jPYQOyv{~JRX
znXLq15dkB%zsMo(ci-~Y*ZsdQ6t1%aFmwHzO_A>3k8A%PdQ*AYKno1J2G4A!|3;4g
z-;bo785#32`xrU@+sG*3Ft;f90BYVR?L}h;QY2=~IxpN>%GuU;{GY#)527r<YhQlB
z{cnH0eHOJJvM~KN7u50KzE@#3h_?!oekBhxAoHnJ)x$<4L^gaaWC;jedDOzCPcMC!
z-yA_BhNQ8PMO@_B=Y|0<E%}2U48X*BEs9<kP`L>h7<v4;um0vIdaeXN5WkE2KO4hE
zDuk@Gf*2=54|sZK#0OTWLIxe!gC4t$aTz6rCTu&&<_QI~&P0tx$q!+Dmvr$EV1!8j
z^1z6%F#9~K9g><1c@U45zkhxcS@F1OqvN7ZuUG?%HA2yK9jgyob_D%g3w<BCm)D-A
z!}gxJFKZD7FX9gmu&v_?3<~<8$GHx}3H{u$r+dTy+wS;txhUu$@SV=&nIh#sSBVcQ
z0OZ}(iGCYb5vDj^BfMijtVRcl!^ZX;NE-AEm@qz%(ds@Os$jR$fLJghw8mM;Se681
zA%v;-2@GOD+-PF^$Yus$gkoTxKOea<RkZ+()GNT+NXYD@AG-2cu*?!?^qlC>6RbG?
zYR=fJwm>INv}S&PDVQeqK>Vr`qH?crbh~u@cYEmDGD4_dR#w>c=XQQtgknc|!_098
z7+Ke|!!jjd#jyS(I~+*P@5h1CeDK)ipf2^N0vpzW2P8#h9!PLeAo1I^jO8Q2%uav~
zX61g^^930U21?Z?f(xxf%pUiUFnj-M#J3Of0$&sS59ECk`nR?Ce@=sTeuRu2op@gM
z*EvHDYght<{O?mQM21kbVFDNDR8;GWJCI(XY7DZizyGQMk%;o)U(Ve2Ns4(D_xIef
zJs?qLs&#5TITrB;JsB}yf%uL<Uw{jW;S1?!b-&r!y8ahGwE!VWtN{1A+YcUaKa!9P
zZErR#Ir&v{8IC*5Qo^S2#;g+{r08X2EHR5v=wX8WN#D?NPd^aKm1GNo=CdG-`y)!%
z3sg9LFe9*sjdV-W<m3>H(41hAmUWN5&v8({={Ep>Wv3ASMNcBI+N$jk^QC~@lzoVO
zO5qsTkh}uhk9s6=9jDn_{K4}f{m1GW80{k3f1$zNlIxcd`%mRJ^$(yu^<)%SI&u{-
zL&SBs1&q7O<I)Pb+kpLGhnuSHXG~V0+|Ou%5RDHU5hslU&-cO*V(axC(FW1iQ%oS{
zTB3)?JbEy%=Yh7qKVxHQ(i$Q@3@c67V#?-!abmOYw?86K*f@!R+hOu-^v1o$0nj-h
z`zv<pEu$giAJ;8_9~NEx4m=OrJlk^!RGC&enUKRoH+lNplan8ia02V0<8LN^BB9Fq
zU`@~u_F?3_eVUET(s>nd+-Ul0zD6!WMEHuq#*9_E!x=_-J$cd*@pj1*0P>&5$vX$0
zU7K72_rYd>hhxvjE_ts|DYu>sl=ZX&+Lk0q6!(x>2A3z6Q!okcT}Zw=rARR1!)Uo^
z5g;Wk^GJMX;TrL|H|Vz8hK810H`Y6%?JPABaxRje?hGX9Vn{R@fDG42QT40+kXUPr
zD5mF39Sh|@9=Pgx?pcsO0&C_mWU+zxbqk3!>$<yid`!-fa^<#Y9k}A7w(`Mvy1prG
z7SxdD`xH%28x3psaj|<QE`+t<K!&?tt|QdL1z;(?hEsY|FTBl9G^sh=D%M0*oGY~(
z=O3xdm&#)8iHCJ|+AGqw>G^7$R~m3WqO(P^6nlO;mBXPi1C#`j1lxn-LFv~DSrL6=
z+~+LCqe}K{?(K?}O+`?B@ia56z(n#Gs3*2=vkK_(71dRk6?*Lev0RDIxl3e<+4*u8
z7yI-;()Louyc}crsO){aG&;#GDA$A%j6sv?1a<&|td8oxy)_|m5pK6rPnHk;xm&-v
zkl*B!hX*(VaotiBXhh__V7Ufn&<8mf)^QOSGFK&YC6(Jf)QWpJg-AvNbs>XDJ_VeS
zhY?%tz;<5zj;=Z@y__1?dBupgA(J4rK8ZrnuwU3~l@WB<_#GlGhI^s6G3==}_yyUo
zeQOQzo2o)wf4_!O0w;3E4z4sE8K+mXjh@d00vRVksB+MXLxltAh9L_h2^$=S$-cxG
z;qq@Ors*w(TxL;fe`Er-cQ5Tu(I%k_?Z5<qpMDiI0ei^0kn$F`GkR=yK#h2l;Kjze
zgWcdj`nd-l6sw9Xa*=WnH$f<Jd(8n@xci=99JIvWA5Fv4v~aL&L1A1M_ZqPB{g#U7
zh=Md0x^lHjJ&`~(H5a}qeRFNT-GC^D;o-RZwU-zj=Y}f45ZEhQG!c&|Gf=IL&^BWn
zS->nR;JCFv1@pMTXgzXgC;N`n?rcvsiqT9f#v@QTyNu$}`zLCGA#t&C2(iA-@Fb|u
zYh{waD0Bt_O{m7?V**5DZlca<EU{T1kPdj4aiR9LSp9HFHA-5EwLap=McPl>Ti<V?
z$3lI?+xXDT?yaXLdSH^F$N%^zIIO;7uRba$R($VAo~28A5>37+L#ccM#+YlLl7e~?
z$88mbw{aFu?YB^B(1(7T=A^5YtCwqQP}2cRRm68WNRmCKcIdqPj0E|aagk=4-2l%d
zs3kqSn7T#N#AkJmo#KiNk{P79)RBAnO?Av`#dBKHleZ`WdJDm#c8`gYy`@YqhkqMg
zHkZ3qmP-}ag+t_WYr}y^4oi@jf-ECLgw)2Jg`i!CM?|i@K>3YChWGbfKM-eiRu9~^
z@;g-hlcG3U0;S7cD^2qGi+@dTl-~5hjxFFg=Q7{CwT$>S25b8Xp#oQciDT_XozIz_
zg2T@)9-mckj}m<(anR)bV`BD|Cl;3tscCjwhxA?sCS{#Z0{vI(2O_{>6+N8a-y`94
zZ1iR%GA}l<iPq~Q><!)m`oZ+aDA~LRo1z2>p#(#QH8QekQ5Hr`z_I*Jhp+s;;5!}Y
zf9v7-UkW?Wd6`n&FI^LH6}vEc`z7fu<_|vXTEi$wRUkCnf|lXJ(YaoclC<sQQ|o^b
zUv)2xvIWyId<l}zDZ`1lib#cq_tG--6&|aTKg26MUMFEXimkp6vawTTQ0mYzyf`X+
z@T}?4hltH&D1Q9xW~z7m&&)UBN0rRr^jjpoJz<$tZnrb=Qak?VSZK~Q^{fUY4$siJ
za-lx-7Nd@->7&w;{C3{LPpX`uw+QIUf!)@X*@XZ>r`={6a3#}IV*b_!wj5lpKz6Nt
zUhvVhl)g0=MtY&tPWPb=6?VO+WQ1i<;uan_XSx<U1JPlj_&fWp1M*dw&1CMC8f1`y
z_JmSrHqNC9CQf5Hm1VJjEf-7+hoQQGBzZ>IwawC|J6z7i(rVQF9xa1bQ2D35<BVHp
z*ghuZy?7W$=_**)q+0SGB353Kg6_VDxkFs(Q_;NSO~gRRvNauOePW7c-*gy?g0=}|
z)Py<~yF_cZKTIR%@>L^Pb0+R2^pxBS2^UoMtG`tAh5SKvyR1I<9F?f+b;pr1;aviQ
z<B1(gNfwLGcbjYw7bm}rCb!$=c||z~$odjmt~^2X1fkSdX=m<HM|~{e5jLc5^T;(!
z#tPrpM%rczsYeqL2N>ii<0s8IcLx@bKbCO`(8ImPFPt@%i7zGcx{_HuuL$?9HGQfr
zoI~y0uw9Pgd(!sm^PLsS7Cwh_?HR(x`={f=MU;PzGO5T{42!!eTIZ!2-E2NkPwU9$
z>j#xg?6@(DwbPN*-q+1Boz3~w=5>o^1U9F~)M9nVAF*wU9W>)(s~-e|7P@-f6>l@y
zws5Q7h}o!fd7;8Cnx-d`)}p9uH*3dh?{6=$u@hDgtd>72w3D`aRCH~OrB2_P3ztZZ
zcDru(q?}LW5!C%k=3PJC9g0~y_DG*)jcbT-WmpPYm1R)pL5zw<Yv1Mxi+P!{^rq)L
z@gj2?9ZIfa3fA)HUM+hk?}|>LVlg|JP>MTlwng!?QJUj>l&PKEkJebNrSNdeK&@K8
z8(F_s+qbV%tf`j%D;?M0B`B&6VTCc*9M)2{pWu+DTW!DJe_{0oo!oQ0eG{pwrSN^=
zw?<|2V07D@GtX*j1wQ3iC_83X+S^lyko$~>$TslFBL2rG-hl^tYkUcE2Gf25g@Ndm
zIaxl}HvJK0T;e=o+a>CHfI%WAkLcOam~}ypK!0=1>$-j@zOtz$Sd{Eu5}teC?WayW
z=!>kROT^0<*D|l%0zo1*4Wl41_$^J@Ya4k84<lxQznkzKM58k<9PC#6Nmo#&1ymgL
zDRP9;qN5~B6y>@yd-yJ)m?oT};b>xAW0*sbSbz(i>%e!v)rzA!cbiTsTBzSzVCcWp
zb@$0`75*=|dd*`)9DzL)H3n&fD!1>JW!BxY&3hL3i}$Ydo@zXq=1aVWVd0Ey)m%EY
zoQ{6Q$<enB5wAj~c35|n8zN?k8dF1*npu{_Eya@|LSa<82R)Wu`W7y|b(Y=I?T7;`
z&2Uhb7_`i9c50(RHY-R@-O1qC2lKvf9Pb2dCfgTvQQs1KX(=A{iq+K77+L2ISQ|4+
z_3kUGmC4X<DRAmqS==5TvI)#%<Y1MQjL*9l!Qf|6Z8}RNw6j{8o#wc|D%x+eI32`b
zjvAv2Fh5N{c#!lh!gah9|L#D_>l4gFLKx0;XAw?MLQRhf>DHuU2fgKSB|kzjD*TmF
zbq{u&-I_VTG3|@x`#`NmlO}<|)WzTRFYM%R5fH#Ikzec2YN(vBi&I)f1h=#}pIXM&
zLSyO;K7q^1YaykS=*z?eer^)uP4|XH<3vWTePkjK$Sip~*Ki!fy|=rc;GUE?Uqbcp
z&(6;WD7N3WD{-J|v&3ni3-@?;`I7Vt*@`a5g}NP}ip1clheqbb30}#S1UqT8`eOkm
z%?Wk?u<Bvx6f~0#{>%mkL4*CSl(qaS^tzSfmQMrYBolH!$^q}>A?}9NTyd<i%{lgO
zBs|w>*;%RuI?fX$exsT*Y4|CcMBTl$u}H?f3~!q)hjG1IbVyFK?-yVl#^X*74l4>s
z)$M3=Iz1DLmdWI#Fa=hwzzPSs1Id}lCC$k}VG54GeZ-sv43Xhg;r!vBT1mqk`Bj&6
zZXJ@~!I0<V)$Vhak*Y7}Y<{*4E<6z#eNOT8xkbnFTIqCAr2u5b%qW@DL<mqU<3_YH
z^dC)hWE#b^^!at^jJH15P3=Q@d}_wnjU9QgnzV$|nYe0;=vRGl@=Aj0N|lkZN-6`q
zPh6BT?J4{tbYym?U;Ho&TpJZY5%ft<{5%Kj66*VTzS~FB7H<FsvETVQbf4cF$Z!J$
zYDdfL;x`w3-OYbr*1D2EaCy0-<M_+vDB#*T|F4_SJQYHj(zK~uwBBIoRuk#XciD|2
z+`C6=QS{#UD5G1Ao{?rlB&#C`n`5AY_bG;6AjYf-TS|$o00S&%Fo++=;u?j2QqH@)
z!B2~$Lrpw|u(-8mmt5-1(M-vHB$8l2`c7F~MoW<cuW=gdoW}=Kmm}MzCyeqNg^fiS
z9t6pqoq&-HJ)3C*C`WrC0jeeFa<V+5ftWYyjGI{aEXbZbk;d|ff$;e}n7uP^tvAST
ziw(t8*fbvmglNF}S=nk)fy^KAXcJs+QT|}?|1#PvoU$n5Wn9SzA?wd|4DsQ<&7c-Z
z)Qt((L-IW&8TuYm-)xrkEBc8VPY;3EA{uAmc6+H?FC(Zn0>lva;5dBb3&ZPZywe;_
zv+4eiwx(YOzl<Js%MUASf5wA-I6}^Wur_Bwhbg|AqOYG7l6uf{U;G!WP*aX;9xxal
zQ@c9>T7134PdHct9e8!_YLm>8osp3+={P$3D%QAk=H^DV?3+FDNWf>h4nx5kfRYLM
zYwgPD=;#nuyPVRzV)sM|^8W8i)z#(h9CjIr+i(}=$>j@|@}103vlS-on~uc?oY;^O
zDZ{)seP%O9(~d}$Dg{}6TI*7nR@mmU-ge%{>q_+9u8M=wS*Z(_=B4x_F{(X&3~7(3
z8*2xq`%A7%JRBcw7v(W!w3~A+br|vy@!@j@p78jHi!-#PMbH~}4`@Ek_rs&?-N4!8
za5+vmLV9)FVJ<muDK4d7(Ig*_wLP8nsB^|8i$MLTZomuNr>nvyUYV>OwZ}U2*oWf}
zwwNDm*jA>in=3U?kOq!V<B!se9t$sTFKew!zly>C<etMj-<SaJ^*rDutM}&;-^V+g
znBQEN|0pEjIE5&&<fK~O*Tmr4?Dev;kMMW-@sdi9k!(SbVO~K(iF%7>$gF5rtF2MK
z2Ty&uThL@C1lWydb$8X5M<LCEV;83tb!VS1d(VCiXWJeWcv{x|?t-TZ6OfJ%rf89N
z_mX<XWKuW@duF-OZx{*IM!FXTP-It|zznxV**wI%5XicF@1=&^Ij~5!KuMnNL-2<t
z;XzGdOb5Spt!&F9JnlrjN)h2awaKS@Kxvd=q-c#0Q%TXq1lprHSZ-&%+59R}rX)}L
z(4&f$HO@XV!enLG7hrbhb~f$-!QWQ(PH5s$ib|l<w<y;&Sl+L&ahiiuEf{xz=0v96
z_$iO8YA8d(tok!)+h)eP;`%bJsvmziW%@N&KMs$pd`H;DG9qc^Im5&$v+Co7_~d&N
z?cAb@6xqLF&(=L29;t4^UUrvG!QAyK&ywZGyG-q34z-9~&=azp7?3W^m(-&r*WYmU
z$M2D3Eaknhcz&|Xv!w5p$6r^+F9_`4UOZJG-&VZ*PY9KghW%zd{>Z)&X{z20z@>P%
zb)^VG9wVCKew%9rU9{z`mxb=bWacOVo%1#7$E#z(dp)E4do|OncU-Va)Un!MW==3M
zmFo44kUt{@X`B*zIeQ6yw%Mz*6JaB}sH9ozddl0mB6u0pF6}4NFucUj$SoXazr&GO
zCwnT$CB3&f8^hF76&^Hym%we~cP>tpgQJpHrl)*O5IAYRv<x3@Unb$t*;~}|c0Gnm
z2I3+Phf)XkAE=qcuHM#iwLGJxmBo9=xm=;vvpn`SDab$V(_H3@!S(^dH4mKtiQaO1
z^kSb9aBXSzeLROE@stA89FJo&KDZI~nj?+Vz+7S@?W%@>XJTGv&w7gTZUzY+)Pb4A
zt4&i+<f;?T-&#$Hl`|{;>~yjpjGj&slLH1{Uajz|Fc@sd&ZwWu7H`YcIW8EUw>9T_
z(sHhpx0RSN+Eo>`H>SBWg>EtN1LKFzzq|mJ2ZDyFMgj(#ZqyobpOBf3$9wLd-sMR=
zZHL?|$pKVcW{$>?TkAHfva$^bv#4us@|m>uYOy!?GY8g9=NWXkx9jfasde>Hx_F?U
zroEB*Y0Pgb{riH#gC?*1hQO=8Mj8s5w55Q8(V;aB^RLal6+tP;P&nU!@ZpDF$5OYz
z-64MSN9-FXgW-#w*5glK2Ki@yj0C0>Kk>mp>V@F*&LOHrsl^$cF#>*TUbz}0!AL-M
z#1Rr_=SLApP`m=j=sEj~MPHKj0u?%{t`xmL9OMtg$y@P`3lvsh{ic4D<TvMijK~$I
z9!;x6jHD$mZj75hU$6MY4JZKmwRdrREhOc;l$P4rqmgPs{-_IsfqJ*uBPlY#OjTq-
zbRsk)KF6>Duzb0~x;kI|=8H{xsK_2XdoK0?_9T((C3k&#h~t5jz9p8c(capF@igQ&
zZoMi_T0KHY>RqL8)aIqgnfJ`~QoI!NTEAl(5Z4%Xf>%_a;;soSS>Uce>?FxOS9cwm
zni|^GFH`W%^8^HbgN8gXVM%n{c`8<}W`lZhh2X!?KqjjGd>s<7d0#b)YmN@l3T%tA
zZ?UZCKvzYo3XR#uxbgp|2hqO{jSe9rT&t5Wej!>H$Z;2bxTNU?-TdHqFr~)nG=#MV
z_E5D2bke2-<y$xKcn{f!(UQX@nMj5Gv{rrjx<ioXa3__HBJhX2#bIt3rZv2K1F~6T
zhZRNRfhSVjY#TWia;tL18#mA{kE?#!C6^eIw<}j0ckSgvOkXxkPu`z9-trJv+9laL
zTKeLsTm=j?0%OzRhw&lKvPaH~zIL8Etb+q5K+Bv|g1<}#;JJwk3;bbx<48bLi-<D~
zY52}TF{)cv?_E50u~j=&NuhCd{zZ-F$a;tJ<-OPaakejG<{zF$D8SfyX|RaUd10d^
zbK>d`cr`u_z7teKcHxY4swlEafqF9`bF`au;rj<4xY%n)kRV;Ft39+b@4ogv)Q_9E
zRQZqMZI8?X$BZxVnfVL8RP7x?x~sP`E#^;!vP2N7uQ<Sav6cLb!V)ka;bC?z)a@Df
zJ}X?#3S4M!Ud1OM1T--)ePx8tp9bN_$dOwtE{GU+b1i`FK_z^DiBvT{rfw>96$^<V
z@i{Bv_Zv(PnF-TJ3IF9Sv<jSgYXru}8Wkk>qU=KKL(o$Q^Ys1%eK5KQMMX2r3h1dK
z#ocSC62ma3=4;~bJ_cmcX5$9xP8eetx`IfxC`Y7|ix%~{c~YHt@yxe|tuHm!`#B6h
zQHkIY8m7{>4>HNB0)=CM7wQBbU_MOp$MO5Xy7;K2G0!h$+Z=~^fnH!vzWXY(7u1ph
zqbL2=5WQRA{3u{#&LxJVNEGW4FI4gT@L{^GTZrQ-L5TE{`SxoW%MmaSibqH+iiJTF
z9JH%HZpzRhmO<|!`i7Rk)LE~0^IDZjEyL(KDxe>s*#0=JYDS}afpVNeiKg!iD{$Jw
zC8HD#>m;@@`)SK1unj_P&n`Gbi!C4?fdK~xXlfBl0>yUH$wc^lZyWB_Jy^`HgCa^5
z_6(K2Rc2eK0V7tYmkY!vahH3jD|KmjkgRYW#^ar|%jIDA5y%*hzXo<nk{)j(IU8#o
z=~pBY@<KY$Q5FD^TJ*F<n#~A%Gcw@9o0;h*#is{(PaJDgKd-FO27n&t#+iQ0@%<49
z&{*K|QguoFxb^MVw?6^ZfQF)GrcNrso^ml^!`dW&xA!!q8-2h8AoDT9HYWyoAD-#i
z03=q{2zcV3-bz)qxn-_GTgW)2vECeoV5K(J_dN4n!;D|57g{PM!-bhT6PrY-rUIjW
zot@DDG+F-zO`&Kc!`x<T9qzWE{C0pKhsg>Lo7}T8n_d8SAjS>)0QqSn7JLWT#@M1f
zR0`wlzK7XU%2!(4TrhaAf8#*>gV?ySW@ba$ajY~mokzXn2ol$o1#OU}b8Q{}SJ%sF
zhi<X7gZ+tUqEXw&#q5v1PwcjTyl%WX9_3Xf<JsBRF7fx{E@TSc2!5*fchH9a+K^Ps
zl0SyPYx#l4;VFpqj?|8VBSc*@6vl<$ZI-WmjAs|$!8i#&x%$bNp||&l-4jdUyJ8Id
zkHX`fB2SHrt`cpxN4hraUayCkGoN;UUZzlQ+Kg!&E<=1s+ioPsu7|9<%)(r$0Ac%o
zSwN2?eL_${4;!LUR6iv>U%k7(BXC;y>G4$xhDKLi&F8t)mz%4+9^reaz9LqY#UxtI
z!m>B?eIr%}5nqj#s@1sgfo4^XaVv3VdA~|X@ycVHnjJ#7?71=GQ_OJ(NZ#7I4H4WM
zy?Z3bt(e?uZ0Alk%C+s;;|Y}F$00b-7BbdH<9CVoO^n_VO_wRmMdqy#XgxDK66ZQ?
z9)e8U>Txb;-EmG`H{CSv9yPanI>WP8-C<+=CBsZQK>WE=3@$`*;d0Gy3&s;^l(&wA
zH^Z4?50LGB#Ks~y>HP#}jA#^&1h?5W;fG_CCr3i(e2+zhpA4~*Kqh3(%zbgO#+K(w
z72Z9Ux9(S-4)%4T_E-^ZF9zjceE8{>x7Wu<52k8Oei_<nuF^=;2|Cp>kiL_5S<E#z
ze5v_u#=@m|p-tJWGFf<l<D90<VpfDQzC5khGgS4jLVgsE)P*ohQdZC>?fgN0UXAdc
zQIdCyiW%qpc7Y~Bb>`?8bZ6Nl+cJ%^hYfq~k8`L5Zh6#PXJnWZ({1djKIrBG`nv-w
z4_7L7eNu~Lxe8$RHNF<rGr>YY@F}~KAI@q!Q}aNqVPm+~*-ghmU**<`<2-;9eSSo}
zRJ)<dYUBo}IxEcn7nMv*HEe`0ptFpcp3BSVfPr;hjLkxs#$evd9;TW7&JFr`dp~B(
z*V;aT!B(#cvz%wH`#5=a-rfSN52Kq~vJWFyR$T{~8upsh(vex~Z5JUC1mQ9N1K~NI
z^c}`*7C#@Q=F8{*Y}W0%$dDV9-!?G&T;;QvaK3Jqb+at(m<QMf9bZ*(0C3nb?z^`K
zD0XB5B?zR)59|cW>NoAoS~cgx-3Iqe*Vu|}Gde~Xty$*EWclQU<k_-STsT&C@{{Qa
z#;#c)Eb(NuMgK@|tGl~=mmqW_z&EaYQl^jLH&bRZ8v7?eUAgn>Q^KEDO}jV+CK!k6
zVow<v6J#5vk#jj1KwXPjy+S9qtxs#Rpc{{F#;Uar)LlGNSiOZx!dMCc?@O^T_#Gz?
zK@?joL1Z62MNmX^zh#sOF8dTQ<h&F?VrAXW&gbHu3|+WIA&8b;I*)P1^^~!DD)x*H
z1=RWyr*YbnMQ5eub~fD*3&uehos3vZZ;cGqC`n^PN_l16zU$SiuUB^1OaEx(pL&AN
z-^VE;w^5N%_&Ds&tVIMeC>bBwU3b#jlmn|A*SMAv1;Q>gTqt-<5XWn!XT#=inw9s(
z<Hh$`ov7cJ*;o4Fk{^D8hN*m_j6FiJxo$y5c;?wQ%Cwi0BEsqb)N@kbuG5kqC2^w8
zzTmE31oYe<>nAQ{a>eL7ON&re=!1^k_gHT0Q+S}7G0_VlM8=T`P_M=aP+Y-o0!deb
z!hAL@i#7A%4mg}0EWSOBI`>>z=)J7{o*I_h3LEI;BhCw$M~Yd^jLHO}MZI!aOxV)-
znw584Q4b?~7ewojM9k<`ju>|x6Yd2&&vOsOCG&Q>5s13y&m{~>Tjwnw3c=8kNo`{C
z>dU!kt@c~J3E6*yw$J#|trMqtfNJuAU0B=NBEj-n!xErMg9ATpW%f@GfUzIljKbuX
z9Zw78oodQ_M&ENg99=3iDrG%`StuzJZP^CYfkA5r%#j;xDYg_LC>xqn2mv=4OME{N
zZjF&F0OdWj9q~3t`hl+-<<?R4Etri`G#8hvxEXf-&MC_7LjO|QZrfJtH<NZ7hdCV1
zeONh<c2Y>|I43>eb`2N%{F0~4_e9OQp0I{qp=r#rD#eWX!N4sS^o!&nft*q+!%uNi
zg%3!5jT`#M#NNbRL^-Zostuvf{g?|%4$(Upi(%tT`5)K-UnDH(xA~XO{I#HMC@`$^
z98cs`F5;es4w2CWvTrZWzmVvR`YAJGf)QqC8cCH_qTR;uef4$Wy=B~^oS2mtjvl$n
zamXV~DYG(cWSKIVdX?QhnRQc_vUoO4-=1<yN!ng|Idqnok|`6@>nZUu(&>mQ3)ew~
z6aIFEFp}@y{2(tE@7bP|=TAKrCg!j3;_yM@cZ@hm?0uEXGnPi%<_CSu2KjQ!*JxgG
z&V4yGbDRxRF5XI`@N8N^{$fcSh9s)qqp#_`$Rh&aQ)p$mYipg#UXH`XeDM~TJx;xj
zy<^7SqKS-bM>kSq77iuRPuG=xiBfq}zuodmBBCZvDn@zR-F$|_7kkn39BJ{+wO5j{
z3-!k}P^FBiMt&|CRT15Ml6Piv)3Cii9LjjdGf(h*T;v&VDQIrGn4NS4?V9ru%EgV~
z8zYz6)`%W2ii0eD{2E?Y>$fv3FnLrDsRrCE5-;o8w^)gaDNH@mkN#X5S^IE1ko)_I
zq5`KV4}AC7lg7vjd+Gvf$=ph{e#Lmp9`hGz2Roz+NNOaHLy<0bX1^HCMuRs|!;wTC
znslz<T<(Q)DrGnhQ?P^x<fBRdC;9Nd*1&HOLe0q-{$FT<0)>xY0OGJg_(bYIjOB?|
zVUOCCoVyUAsxb%+%(%^mGIKQr>y%iJ(#QQ{RKv25+4ef+)$AP!A?x((xECDYrBR9I
z3?o@j0T=1X>#YAl$0P?%fZ;9``oENr@F8U%RRe5rSZM(f?y2#;F_zz0y~5*Qa*C`n
zk%nm<M(qbUbg90|;O2*aoI5?GM{$D3N6+hJeid<7;;sc0F)s=;Rm7u&MzfyokOZbU
zrnd5x%;_#;`M)@y|MeUXNK}AcF0nn_A@Hw8+aGmEV;D)x1Z{rRp~e`O+n_@WYXQ?_
zNoa-oKm*5I`{wYP27PA>iUst@p3u(3IJ7WwZ1Rq9!wq<-|3;ILvi<uDduNw$<V<w3
zufsd5%{BbnNx1EwFnrd`-T}Hn1RF{aasVW|P2GCw4NNT&YsN{Rv;-<To*v+CT!*;x
zNTIxH^5}xe%71>cf9*p5;-%3c%qjB3YxKXqU_heAFqjn~a;QEX`K?2b)fwN!VSEr<
z$KrY>N7BW<6(w@QRuy#SnnFoX(<3(X9D<_cQJLwFU$|>YR|Wdlt9<`vS5)9dW3k01
z>r)7_NX@k11~2j<O4~Tgz08tMart`-V@uIG$-uiQTAuurQ8iO8{5#?^XZtQ-bTRv<
zoBR{jddf?OSX#|O3C-tC(Om2S)?N}pOHCIh%^Gl_8*uoVIh;s|JeD5EiQ2_I&u2NQ
zC(T4x^`fG)$6RIC^<2nIGP@R@zlxxea}vL^+(2N@n+WVF=O+9ANApym*oSb+Ecx{A
z{`Lp?dk&$y96SK&q_V`{62$-OhW*EzLCCfheaQd+=jN+GaXJYO_k=c6oqr>m{#rNx
z^F=@{vTqsB(*J&z`hU26|MI^HI<VVX?Q(ek?79E`S^bxl4*yue5*EAQ{<M9+0Pb(0
z6u=Z=#UtLL&|ALp9)Fa<E2-=JkAGmV`!9aLLdME^n!g?DAY4N|gs#Ldo9y%J6C6?+
z2KHqEVuXj{{kSKRNum*nv0uu3{0AxM{18V7i`AR0+3Sww>W>__=3xrj0)n&`7si1U
z-jEJ7%8&V;Do<$xKs4KH4hM8W6F?jH5Em7j33qn#!Gvni<Nfmgt&mFlG$jwjAKKi%
zD7o7BV(kBV<Ys5!(dInMwfjY4J)}Jh11515JdpUD`F<jKP@rfT<T)2-fpFyp+mu~S
zAO?!tYh$zEK=<P3c24C8;tUL>aDDLpBOjwXpRdhOVsX{lvdfo#0tyiLJp466tl}T<
z-a|vAP>W!+1XooZnL=<$A+3NRsPqDfPIIF0(*}iF6xJDMmi-W6ZJ9t4Ey5KmJeKjK
z;6gFh7fU9sR2@pk+cnUtU$Rvs`Rs#dp1C}LW{5bj2c&P7pd`y&zSr1YL$Rh?kEDKW
z{apP%%$te(T@U;%0b#c$Y~|?x`3w}IxZzXF_aF)efM51GSN@A-Z#oCZFk-BV(D&A)
zhy8b#(@RYu2XsF?sn7S9w;?pf8cxU1L5Q<eGlP%Eg}C!-cIibs2$BD?`aLW3r`7Ks
z%d3;mJdG2pf&R~+@}${!rUr(vCNGXlvX~5I=&&)Lus@Hn&Mj!<Pn)i?q2&1Tm%nd|
z)-{j*%Z|!S&Im`C$!|Y2tkONt9%vi$Fy$=wHX6tTK-<^`L4Ze)3E=^B&i}gI5yw|t
zI{&Y8?r9Ke6Kr<HNj!J<kwMUG6T8}Ee@Si&ghvZ?6s)|wjT!}L@)fX+UT-y<#f>4F
z4$$|X0aN3dz`5;j;I-Tdd`bo?vI-;h(F2FbhGER<v;%CgS#V($EMh-H3)7VEGw@$A
z0A^e`ib^o*L|m!`zfK;I27hE*Off{lZc0d%RP=1GFhM?qWRJz}u?DiM-;-x$FLp~u
zR)dz?GQyg)gGe<SWHq>dt>a%C@t-e1o(M;=$lOf{vp+Yq&+Y)*ApgNfy019HS{*9j
zm{TpEkW*Kxo*OKQrZ6FrEo`m<Zn<})9a?5*0-gQ~?Wgcjrm{sHi&>-V_3&+-%fGYI
z|8<DnMgH>R;J-LoPvhy<fI=R)dbGAiZn62eq-8Au<KP;w=P(Kzi>f<oUo2GI!&#C>
z6o!WbezJQ+R31uMRPQnYRL>dgtZgirj|~GAl{#g=(#clfPSt#Bg7(c0(Rc_!dKrQk
z?8vXQk`Kka*!(P_an|ho)f|GmS1dsz)ZL3po^0~r{X>{TlLx8J9J_Z-<bZh^i8_S<
z=>+Wg+uv8WDvDoC%v80V!nc)WBrdN%Iey%+D+{OJlDOVl@rM^PaF<(hKxdB!6dyT2
z!)-jnH5C#;9eK~-ZRT|0l_vx?n8}tP29W8M5uk}FR2fxcc0T0F@1$c2+xTTpIsZq&
zdSq20g^$6~lqWKZuaX7N6vV{5=i7%=XFyd;fTq`Ss2jwp7FI>T<Bd}}XTv*f7$vu1
zFBpF;-0y{yQNv?E&zt}=jIue;Y_>WfzjG?fAyXLa0vsN^%?7&sD;8fC0krXrSJB96
zlsbpxrd%9o>;@$K3X-Lbn6*pJ7Je!&mubyRFa`2m%_#Vfy@D=sD8}^0`{A-?@unUN
z8uFFmTZd|%3q>(WycpMW(wTZ$KuF)_XnNp4hXhy&Ib}%0AANy17^iPGIYYFuZpbO~
z&IO1;zX}`Q8+Y|s7n1Cjl;eCIY<TS(en4cCkkf4{9#NsHY`N2b<J!l+=tjaWk(;8=
zqs8;|B-_`M{q*DxS%;YO4KCm7)VX6*pkD+ch=zi~uZ2HcD|}B}{+#5@Ibr$Kse;Za
zbQwAIU;u9zq1^IguoZ`x`v!mQV95}Sxte%Px(HpgIuKRQc55?FDN{~>_vI_5PtgXa
zgw9kzj2}sV;zX{!U+3$Xf|#^EHGNx&uFJ50i`)-&1G;4g<7Y2Y+L+ok_Q<U}KY9}`
zM}Orjeu?Dyt~plH>a_S4sJs+=Dc!AIF(uqOl-RwQd&<mNBjU=Li>s>T8Hl`T?6p?y
zdzdYrxKlp3kn<`3@?!irhfpt+lDrL@Vt@;IzMp^ud4^G)<plklH;uZj7c<-pE57ng
z1X`mYjQAfBm}LLwr7Cz!g`fB<0e-&~xpA#0um<*z#1Zsq=Ml~<BK8mBb7?%5v7aV@
zK#7)_SBSmu2Y<B~AgtiK=SAKZZKe<$kpqVap|O9|>LEZ;<jsl2soiF$VptmWgiy00
zwA@<t(rp8gliWTJzI}sLT8KC<aU8nOr!A6m3%I)~+ehW@KRst9MOEKr=4BQSSK>^D
z{8r{M)=)-n{*TmO;H;F32^j#u*a<9u9gl^*+vB8;DuwOzwrPbWgd!Rw1FN&JO54l_
zuU#)42_ELY_w3|z?WIw3&NI=fUlGuM25!4TSf3S|H_vKL9Ob`fp*Sktc~50O;QHpo
zXutKQ%YDF3G4;w;5km9=#BwC329v;-t)^{r!6^RJR}YiR1s6r;I}?u%U6R#oUPZ1?
z14uM{-mkXa=GsVpmqLa%EOS4tu!hDW!tfCEWd&NHZpLoktHYif>b6g30d}#49z+Y5
zDSL*(9Sj=w%6Gncvhk#^bb|zyIh(J5_Wxt=y#uN4`~UGsMH)KEkwj#aaioxul8lhe
zv7OB9y%I@Lc9}=ABJ0>PGn2itx0F3AM0UUDyX)%u+@Je%b$##8_uudDzwT~V$9cbA
z@7L@3e5|L#x;(L%?~0*@tmQN$I2A%M^oA<E=FnU{TVhS6Sqc)vXg0ml2JwD0`Yfx;
zWc!#cI@cK$Q#Vrwy~P4i3&?NH2h`5AOsR#?BK;)kJ$4e6Qz}EDBI52U1`*5Yd1)i*
zqnY`g(ugYM@(JF;AxZlWTd!p&t{e)pRLtq&kUy9xAZZ|dw(qlxW4EHDcf5`z&QfH&
zo3m(GAGLWRrtSlxieXAs!*#B+5-nb?+-N0TgR9Bwfy$1g?fHOBujM3ejh$PJH^?jt
zDq*1bi_N!SFTmbn@fwe6X5J1_)<zv=%Yqn(VvNPo?fAf*+Ok#S$|k$iR-bX0C4gdV
zHBElbQm3(c*Y_K5T-q-04<7r)!gS-Mg7Iuysg)VT+^~}JIk+3U8tgT7C$REM<&TQl
zSdnq;fTQ3CN5QrE3ORSRW}fsP(i#0TXNgDR2w!-QUs`h(@Y5E0{!WlnlAC_v#_wJX
ziW$)=deO08vxccBNVM}qAyqhYg7`6#0M6pk5w*M7SIJUzJ8anQ)-w8oX^*O!EJoxe
zrgK(F6HUjA_GU)!CV|02pk(z;ahtbcW9{KtfaPw=a2BX6qu)1rUn-c_;ioh$*;hG{
zQ0_Cd??K}}^2W?EW;7)WT87EvWZX8hP0ih!W0GD!5R>=PI{C&|4>+qs{cRbKTIroH
z5zm_FjWgT$iXUm*7od2=LnC^4S&5TA(@DGG>?M;%Z_3PvJu%89;c?s3j13dj@7`R<
zJTkngWA_Qy-3XI0Lc*MTyywJd!7!0O!S(qp&gy)wo(krJG)DgxCOSY+A$EE7jYMDR
znY`u#Oich!*!M4P(?X5nUIlg&%M+R8ns2n@gjK%yW8Rq95#6?XscHB2x)Pn;3!9f}
zcJqOXZQA&a!7|n7BV)K_^zJM&aF$<XKRSp7eVqmUj=5-0n06Qyhqgl$*J*_zr!7hF
zdTLE&t<4iRPhjgq27BS)iP1KswA5MMfgIY)Fe&sJyY7dM3Uc2Na;{LV(1UFVTdw1V
z(e_(uiqksnzpH`%Sa|?~X)&F?{N0kOMIc<lb*D6?<W7OfR|*p^jiiz%o~zFG+f@wG
z9=~P1IRgYR+WqFj3V-57lZDRplw$4p6Yeum9k8>a>lfELXf-?veWRav=I!(rRfKCR
z^zjIxwQa83>D_zBEsu7*sB60Gpm?W6&&rpt#6-YUm#uVI9%#{Q(sZZsh%vLbJoJRP
zG<wt6uChe8Qt!rzhl>sylX%JDru3;K03o<bT+*G~KQ!*}K1DZa$xr2Y@~lng<-g7(
z&`N79HeD0-xhv7)XdO1qT~z!++bw3e$*Sdj;PRkUsTn1%^gd1YFtNGab<>uZHz^7N
zJY4CtRy`L6ed?wqvbug=60|+Ba5c0wq<G8#4sI{hhh6QmC{r!jETw&J3HL+FWW&a7
z@_)Y-{<<sTj}f^m20hyI$MEyp&yGLyk>@H!MB%USGe*-`b<qHpm$%O{#B(`QC!mX#
z!<qcf`JWjU#Fp9U-uB!VqUBAVdy6{|iE$>81(k+Ph>5gZwNO#Tajrq;pawa{3%)eN
zAxdimOSFbzFh8qeqOKhtW*pdY6!z;c>EpQGmD(?o5;fbdNf^G$BMe)4o+DA3ym-#4
zCs@nPKacM?v9t<fu+)@BOQ;tkDp=}vXHyAp-Gt1gqJy2^hJkdCWLkre+R_lM!a0;&
z0(H?C^<A50>|yT+tGDL{lWNGs-v%%$pR#XuVJHr%x^5|2o$5Iw2<PQRcxNeIO0}H-
z6x|dvudyg@Im#&#&NaWTo}gljAt&-bMi0gnS?>o2rz2Vzj}+YuA7(RDisKM0x<#X}
z6L>sOTXiRq*4cV9)vE(#K=AN!FL_F^lxAY*?h`vMODWoK=PH<$@N~*$D_wTt+txky
zw-H?BbSqD%&iMK;2RX7wara7jRlpDtul{^RU7^$2-7MXg-3EFzzs)^4aN0pZ5I}D8
z`-2_*>1#Sxmup2(w(M71_n`!3KBn6W3AGYMRoYd)-wiBWY|+~hUc1aDDRe(3?8a=G
zPDxK?>bvFg;JNplGv28yaC51?TW}oAqLCPUoH7{Q)L{~&aw?=s&LG??!$_^irEj)B
z+s$-at+E()b@RZ&noE3GVwXS$!Onp^V9#G0<=>9csvPDA6gK@>;eAh#z2o^(RD{Z#
z$t-wAdMpDghrQ`0(22Qj=-GmwKfEM^58Zdvx7tuO$0AT(v<j?VW&Y=GCfv{%W;;l}
zx~#+X@RY`tPznVsi4kX3c$bW|#u;Q3rsh6>`$eB~O~J4=CYle`k>4Y`rdeXaDbE#N
zYLC*A5;<?~I*|90tBY2&9NSmOp789ed~dtqX*vBAV>!q3!2xp_9Iu`=h>xk@Lijd1
z16mVEKRxA&p|wm1Zg@HCZLlPkvXdCC>RVyxu1r<U6n#o7OPb*$ST=Ee+5SfSK;Lvh
z=#2U|!s2)F9|{xYvx!J<>9SGFJ3nnsgTqd4CnqT`ZgtilZtJ~Q0GBJrP7jOS=1zI|
z?Ph$rY(>D`?G4l$J#l%e8%DXUc*P^!uC|!QyIJ~WoyCN2%wDaZVc-_&+pLUIs;x9G
z>sX%Y0y_=Tndamt>uKR8IjUC#``T!G;|x5UCudtXoSK8X6iP!1%JL|L<M1EVzMmz`
z=cOm;gpXT%M|!ZJdvvWSdC_X04@4OtfsT*AKTr*H)*$Ww+@GE2%t%6eZgR%qNIHRb
zv5-0I*X#aoYr2Qz@l{V-6)}(Xz+XRc`x(mS?eh}S_sAr&A*h};BdWz8P4ybIaF+K}
zmx#P|o5c=+jdFqY+J;nXWk-<Kejc>tbFNFYvy&HISmgPZ+-%(zcUbew<XSaH=8=@x
z<V<C29Yr6HA}4CaKhE4Rug53wqGJc(vNy2WN&c+8B%(Jg8n&@am}*|2l6M?%zAo`<
zT=b~Do+f@pc*!O>2vn0!$Y4#UQ*PkF;zeJqizA<idi))|F*SD8KG>J{;55~z7Q7xG
zx__TmF8(+goe;@IW`a@4Lxvcb(h!t$9h%)`-|-0ALQNRO)ny9($(ytzS^|?R3I3Ri
zxQfo<hnk!-NF9WEm<wv6f#GTc_;PK~lhQ9XdOw(=%BR^kd%2QrQ&M2XthB<yJ|DcG
zh5~J|v_IC0DtU|KjIEAc*fdHsN2cJW%AgB*1uZ?+-_a+?<`ep`wkkc>nx4<g((8cV
z@7DMflnm4ciUL_<xmt%V;i$TXqasU+b6uJUGl>G#yJp7)c6L8(E-!Ty`q@9^x@%@N
z*{glXgm9oh2mQdg(d@){IHT$|=3D_dQO|d(-<=GXwPlwLtcCw+8UQYn>==hmsVNx}
z=SG<|8@=a?P`^*fyseKOI5#R_;XLFtm9tB`Rmv`J-BzR(6#aNGyzb%;ABKB9mO3JL
zFnf4$gx=viyO9#YpsLO*OS+j4t==0>FplYk$GV=CEZ?U0;21J3$t=C1#rEAjUqW4}
zT>bqMBlulj^hT5xNj~82m1;cur3D0I-VxX&vp*iA{!OPsL5(G$AU={lmz?wS)VV-O
zX81N-%ms7<7TNymaOFPNWs$MSfz#gU-i{x87PWqL!c4X_L%f$2>Vd;FQ}n>?M~7SR
z)K(|+PahK(sQ%y+z|=$wH2(<H0Fx%JZW)5M<WHw$vZJ`BEh-F+eR$nxuh}<COvzv$
z=e&fv7CbTDKigeoxW`PqY8~@}N-R^;Fqk>bt}`DAb3sOI9Q71k6nk{KNX67jylsC1
zMGll2n7qL(u~H<Twi(>KE@w!I(r4%c&#WdtmA<e7C3B?K8B0plm^iL!n(+}(+FOK3
zv82VM6P`|y{@3-Oe&<;1a^g5xM@QD+ky0F$qFe{wwp<3APMuAe=7qH~>K)cJ_sQ;&
z8&0sC(0l1kl?MVZhTUP&)DlNoEwi+wHxtbF3>o6J!qd$!Z@5__m&4-08G`=6i{*lY
zLskNI(J`b%2WZG~7l(#v;!SCPfG?VMQHB1v2}fw8;+S0#ZNK=icC8C*><;ID9T_Eo
zuf6{)5dO!BaYljV0Y%O&uRlA`sxxQ5LSTS*yf{@T->}U)-DzDa+vI4wrW`A#>sV(i
zkT@JR*N4nOTTYX3Q4yRVhD6UbccewiE71+gM*#*tb1J!A<kVkUg@ORTl;?U0B3SHr
zgs(<aDT-DDoA#E;@T-yQyUwQ^=bJ*}GxlyS+c`nF+M2I;6^t0Zm~ly3<xGc+d^87T
zqmrT%0;Q^-P!TY^o)-@Kj1kfHffx#f-VML3VCLGf4YoB{s7beP_!;v*7R@MD5QbWW
zi>Cl;IEbK25{kUvLifn-aFr4^O@gv32gT$nD~no5wDKm{(o<0<wk~GtG2~cq7Wyqo
zPxjHWrF&G65H^-q^cX%v?{@G9IKR2-95q_QSBsHBbV=R=Z`nvP{ESo=)2?|x*c<2_
zX?7$`1p9r@W~if~hCxKCHq{}IZ;AAdxfbfRE-Cq2TBQ66O&zneSXBWPR{nqM?*HX_
zK1^m}dYkr-I&Kf~Bm4t$<`6E8=MZ-Lq>%iR%yKf|vKMw>^oG^RdQhrGXuE}LQ!1jZ
z0OXgm3e0JfKLALZ{2(|A=eBhkBDRy6Hae>ucLi<8bsF&sw-8ZRX>>E)n-Ag)$DVLK
zeNq0x3V5(@r0jBwf;h~@l7dn?o%=A8N5Z|BI!E$TvnRX_9!y17B8CMG>}2jub`=;*
z)_Xe_j7tb1Xk-)!l}^Q&&89eAKy-;{B9<kq`W-+&<uTFH72S85n@9AqcH{zzk{A*T
znqIs_z#0$q)XY02yjc?oKFOcbuXLFHI5%A@Zf**U73aP08@XEC<ZYM-=1~z`UufOS
zH1?pS#Az#dY+j$4h0)cfG?H-NYa}4|IK8_A4GfkD)uOYc(vkVWwq+uhO5pn>Q^{Jy
z6y?_Vm3B40|4jqehTX*B6+-Sk2u+r<I>uX4l_e9&R>v+rhwV+UL?ncb_<Z>aFH<`P
zoBZRg8d2Zf$a*JaNSzyR47}Sna|~ZsB&Q=}ukI1F6BAM7;$Ax0V4H8Ib}+Fp@r|SH
zor+h5nFO64Mhp9T&L7G2+aC`!?t5_9;A_&;P=BJt^398Hb*dc%KTMJeo!&a})MMsY
z82}8lmQRcJCeg;U>8XY%)_)`={as~$NQjtkLHd>_!=9f%aU<Y?mnLdV7N(NT0vT+k
zA&6_coC(9B(>TV0tFBi^Vu`Q#;s6eRKGC~3luKL<FV`a~l4J|0kQZrU<g;ZPfH5B5
zw1kFB-j(P>jJc$kV$kc$jEStVE{r7XR=mkc3X2fUaZ~cNql(@91;<9IL#%oawbs*C
zxad-2-5Wnzc|SmcO+u<tPjim~LPGUW4Ch)B>40}9dpe9#WFyy|_|jVtqIK{V()8iy
z^PcDuWV8tvi%`UyAuo?feER0x!0;qrH7nZ6x-e$#shuX2ZcR&0&e7hqcx^kWckzJz
zJ4Qje&<nqH2gRVoD~~D9S}>C=<a<zdo&WqD&f>4U?+#sJ!i{lD<qS4$TpXSSFeaN^
z$q1ZeY8@<05suBSjhRU-WDl7nPH|)jLB;MLMFm4EslPM1vcbK^UdyVNt9Abtq-$yh
zeA3un#P*4LaoEP!UGmL)V-H8>!u#AxA--~fQ7M5iIp;+}IE6x=T1OV8JZoF#T-{F0
zRF$!ur8Brdk8<A@xRo-l6O;`&-;&9dp6!vD8_%NUUN4HAmxQiOb<(7c%;WbjJ6JpR
z{_*a6GKe5dnlvPT>qG%j*9GVq_RfI(Lk^9$&{RDb&CrOCzm>P3=736)dd8<;;{&pG
zA!eLwk&`B49vpB&nV^?!#HY%l$WL(Q?o4udNI{;2Q%Fz#<qs~2PBBh~$J+7iS+C`i
z*7Kgp%u>a05Z?!I>d4DlwTk8N916$w>j#*9Hw>@KVm~s~e4Qi?^<prZmYMWcWi2;p
z^~kG8_=boQqvF~SrSW9*ct74PgFGb%#bN~@pUyTj<2fN;2^gjHks<_1u$TovIE|c<
z-2-ATWYiT@%JN)P98i}DbhT}GyG-BHwb~ICH8&VVe5X}#O>rb5fze`a6#>SAFh3)-
zN!1}w-37qNq;Ra}d6+1rTC1GP<*)y9s2rJ5pbXbne>tP?%tANR3ESIa=*n}kLGBtq
zo<r<%7su<C5JcMhYO_&p+g<7-6Hy-mQaQeI(rCMLgsb@>L2~DOIX2;~o2=@P+@@sP
zqxDI?^D;qsVb7ng6l%O^v2}P8E`FZuG4q3FeYVwlxa7E@n4-{_P#IE+Z6a2hlxI}w
z@f^gSO}9Vz8Vui(rRJ1yB;7b9xR^+D++AZaXZXBKGCXvH5L}08OBm_RURGYr2_04U
z{5d{`m+d6SZ&9o!x_0v`(Wd(wQJ>q&b4qr~gWPe6((a0>FbTQw<ct7mT5PO2fjT=Z
zS!^-yN-7`bt-KSev3WVGn2(0*gq*&~B+jFC>U2zK_*>&FKY?4e-@o^dX1JBFS{KpM
z%-3ixO4DC3sl0+qZ#7v^)-iAEjZ~*wA(*L^b2@XEO=(OjM%W|3D`rvBz66F???S$z
zY8lU((CIIfQYQ>5$2$7YA6tWn8ua&<wM$Z~XFWHN>h?RyL7c;%$5vQkgSY$dceWAn
zc*8q-j`6>6xf+t~$%2v|Ui*2oDShr2;16r5RpSFr^hG5TU(u^ORvS&)H}7CNxzUx8
z{xoya3tUZ$2>1-$#;|s@5vK4W`f%OqE~<aqJ98uN<rHI?j~cqv=8t&<irDBlfU;Zh
zRy;+il(z5TgqlhawI@nIVj!ld%)dcQZR9)cnL8EV5=wF!+|Cb^e$lTc9q$cm@itOq
z39lEYUxPPBV9b`?z7PgGMUuCSMU7oU9~wkvlzt<Gk7OD?_vbg84m*I}^IWTquu<Yt
z1!G4rGpQS<^_*v-xWZ3q`>|pn5cr&Rhx`_X?gGxNUF8T`K8ie}MO<KNOx!+LhV#D1
zU29BPz)CHDO%whaA6hs{>!AOemeo|(e#lu_1fKB{uKMDbDt{4;A?5<RgaM3?%_^bp
z7!HyD>SX(}<idsJ__cVamx8vzv4{Z1v2RYq^o@~(eTPa-Cf_Hj@q`K;9*`;fyN<%F
zxfXHdU<qziK{Jw=qwPgwo7NT<yB>7%5(|IHD*p<4d`h(PoWH+KOKtL$3=KNbZ(duS
z`o4bp*V@et^5Fvq7Qa1>4utHK=j2xt7n|i5r4~iq<7c{AbY}DjDHV2BrWfmc1*Rg;
zb|Q%|t=^7P#HuTkwXL<(9Zo$s?Nn2QN--37Z^w)Ct{z)84diaBTuv7ZSH5STZrX7*
z{~C-)>8U;Dxs7ApNhDk)xZH;{p&^*%N4J5!yG;GEaCU64A;sUiMf~<ugo^!|!H_A4
zVfeVAj-09An2y;@pL4cY3l+R6%-^of@WBj6!==qcxzVHjx|T|-yY?gTo3CB!DVv@B
z`E7#p-Eal#o@Z>Ow_1A7<17q@jj1t$4{kiuAJ6zypCFLA5QghX*N(Nff5G-x^|{VR
zEZ_ysf=?!WHt#LG-<WZLs#}}l4u;L&DQP$Ut~o%+1t3de&wEukCF<N&9_WAHfWuy|
zBF3hgtPrB<`ix5boK(?zTt^+OQ*zZddT3Mcsu-ky_I3C&N&m2BC#LS)4+^HwLrqc+
z7eA;9tg++X6m^w9{d~#sY1+mO`)A~d2E#qVK~sz)r@}_yE@8EgZ0eR0Xu8G!_&+Y;
znLkEg$CAE7wQLN$mxT}ew>(ASLI{C5SL1=YH(Q&|wRbov%vA$p8(HIy&_#=rc_e5Q
z%j{%vjkh0{P%T%mI#rugLp}^Z^S*gQk7=gU-^}+slST8`4c{|ixN_2n9QnRRBZl~~
zunSTiD3%>)LwLw)zNO$-QHs2FuE9uS@2#HLv5^qBoDVTB643*=u|2*An7ZeD?VHUC
zK{$2!&|E~;t?KB}?WxNQ50yU*cScEZ0Jkd>rKNtT!(VH4-mbi9A%bMwyF_<;UR_kQ
z=aCn7P&zV1oE1eZ?Uk_JEJw`e8{wug<1&!7sZ7UYs95^!qqt}-hs$~0xFIjCx;%)v
zIVVO+kS`!B*yr1bWEB?cIu10<CRK<%8ruX>!;=}&zS8glie$E<sGD<T(M|EL!3_*T
zzSo(n6}3^M1b0G<L6;F#`7FJF2Pzz4K0Bsf;}SFHZUgU_#eOf*CT3kK-T@&YNwJL7
zfd};2i88p#X@G|jZ{y^Z*lMrPV!MP>;wL#I7{UIXflQG3kc#^qwSco*rbAXgQuCfl
ztFtI3l6cO^qex6SmM8UYn4HC+4$8QM)@NxymfJGH<&y#TRgU79(dNu;MI!QFUpZ3*
z^tW}3+GI<K_aOC3(=LIPjELDP?2I;X_JSS#qG6fRsdn{OfLUp@Dc1Mf!w6XMYc}Sw
z45?W$+h2mT+5)*N7jRd5`&kv$&h|lL67agRepg>taKJi+7Q7@khnp$8dbE>lF{RMU
z3SW(Gs5esvEJzB?pmSo8Ev2IkYY_PPY_}E9cPZ^`rvz>l`hzBJcIhn)#J%?~0UM>Z
z+Mm58fwL0+e43dUTM=eWhG&jeJuO9{P1Q7QcR0*0U+9UWW{{h(*6**8Ru~OnA+i6@
z5~}J9^BKihUD~Z>BL<()ef+)zrbDJoYiwv`F832pbPJ_jm^<P*$faF+8uG%#Tv5}8
zDGqr#@?37rM>-wFiC&KNHLpe&xmhZbM(*UR_Dh(e-%Z$PgsERuyNa1UABVooJ>wgC
zAvZYnB98q&>XKj57$v&LlX^B=^?8jfS5!FtqKwb0Z#9a;lyPm1=XGoQTFxIGV_Y(7
zd$j!0siTYUr7Ko|@<~(1_Kdu3KAU8lGL<9S5uNPqIZd)3HBy$?EXDGMrU__A>89N^
z<W?qy3@Weft`>TndumoWmSV}?m%5NyU35<_gITQEF>N<<Q=y!4D16(0tkVu>9X2{F
zVfUfQ*3iRpXpvrZO|eIc?n{`n!0PMaQO(t-a+@Z&0<zUhzQR54jTlsnPBvuqfAHJr
z?vJ<4TBy-p@rxeJXwq=<dp)wmR-av=V2t0gvBj__k90Yu74~H4Oahx;G;lw0SM7)p
zMcuAl?4ZDBO97gK^o!pm^i+!AEOzY7qN(jWb*+GN5@%3U$3DqC8kfeo5%NWLNQFCQ
zxPqs5)wpclKlj?`=*+I-j_jnu$)4;NJ=ZbKB5Y3vlhp}w%RAvs?fA(C+J4dhi9r0(
z;82)|?P0immTmp|pFh#gMG{3uw8eahN$@slgJ8ex<&mG`?MFaR!ElE5!t=9d|D}8W
z{U7)5A@X{#Hy+hM{HwSA5w}NFL{s!EF^m~d<H*inzRtYA=GVvm`P)4B`_ooL82+)<
zEBp`t{J(E=9B)?&7~SkGT7U$k7l<M>muI*k`0QCU%=_+J1e7aP3+uvh=jp`$J)X#4
z<<{hJ>@}Me`K#0a{cnzxP~FExNY&pzI{WKQsk`VI7NTSGb|0VXH<PZ4VMy#dJ_Iyu
z-5NAqr$GEX9_jHe6hTt-ac)cAvM`44-SIt1zrQ^G<@>fDhA%g=XN>39gL0SunE3Q+
z9z=;1%ZLKb5F1F%o#20s?w^0<e~~h5Kiwl*^&|K8(ceLazj@5?bT7=n2|(+hD)IQw
z4){O*Q26{GXCF{}_2B6L1%v-~)BVrSKO_i(ivv30!Fs<Q`L9a$fBx*heAAzQtoi}J
z^EN&C+JF9^{I$&<1;L}bu4ZKTTX^;Npa1n5{LVk>9}53p{L1n0sBGF*PoDhOEg~{a
zq6;G~?g0^M36Ng_v~(paq(cFpAa=MQECg}dTGmw={U=KzDl+)jcXXHg2y)L{WH>qk
zE{SQPwk^P@<R1IEU+yC!wnN1BVzSauKz=U-xXZ@k&ZBQ<+RqXrHqm^w)mA{`!3ptA
zJRZ?aA<mwAfE;czZZgnM14MiW)O2n-0*G=0$w{2s=BF+ufx)Nug??<Wn4i1?nodhl
zCZ(T#z4v#h%bzRiDhg+j{JCTM|J=8pL8NY=!YG8Yd{Bm~+%QohFc0X3PK1{a8l&|=
zKZW9&<SF3Hg~ISDPDr37xG=dXUQ5TGxs7^u8j-}qblo9<6bB6$A@=N{FlavtA<|`S
zJqk+%u`JaT#7MkU=0n>ovY(<@2rSGR96(BVr%`ujAqb{3yT5SAf>{2_lGkaNlLJJ>
zPAe!%Pw$w6Es?taIg9f@ug4*NWM^xKKV$e~n^d)*jW@c+929C)g~UsObb?a|AjBuy
z*O^_yp4|{lgL!kch&T1z{4@4*`)E)khMo)YkB<0NB3PI39LdexUs(W{&}smtLg5$v
zzKVRG=_qr|76|&0#6{~90rs+_Wl*!MK^$jK8wM2qg@~9FgeB@L26t{BUx!G{#V;N|
zN)epD5O&9ljAjDv<Nx6$60wGtZQ(Fy^{*?Xo!b-9WLB(ZS|65TxczwL$+hjbkl(56
z_24Aw5RsIKjH68PW_Y$+ZE}<ufo|Fs8M1=NdShYg1Rbu6?IVObvTi}-?82$hb&2%b
z(en(R+3t(xg+LB5kSs7{dr~eSzD{IhBg01#0_8ICKkXxGF@z3DXktwNV_iI7itJ(d
zsE{h&bRn8d1@QScj;(|RQlhPgbN#-@89w%Zy2bWSE9m)B8u{-tHJbMQ{MdIvRY>ld
zI$;H#q)(D*LCZ(j^~PY}_qWVzWXV-ogv+7JyBRPdb4Z7?QlFp^uzQ6V9$2`mb(}{|
zP*vt<WOF!5ZNCN@+o7alVF{LO5HVgLu1*AME=0uERDkwt!BzQgP8~-|6g~=OBuOmz
zzApWYg9URWceryPZ*LB);MN*M2n<2%YzA$`16lr0UPyZE<DLvy*dlEbbG%VdYOr|K
z=+0+nJ%$TUI-CIK$wQ=5Gpa>DpaUEPrM0R4CWssdkjU5mf7dYfdS}mlM9v^!a$f<G
zRc8su2j5$g<K5f3K%6za7<@7e*a2;GFrhU(raJX1Orp~aT`l^VR|{MW#LL~Y8Uhhs
zVqioz(!D00o?Hi<s`=c#BdmOyx|@)`d6cVXoag}UGx@&4AKTN{7BgPA9dugW?!;FZ
z0;Ml+;rJP%c>KytaWaJ6(H}dw?2Pdqlv$oV(gm!RxoNtaCI<2~Jt;uuFzEq^lsmbM
zB;a*it<}vU*rk>c9wlgj-W?H8vb#DCsBIyGwvTNn|4_OQWXF5VrwL@jbU&?qOUKvf
zD?ODM@Iq>8OaiVnJ^)|+^lHMmaaMmHFE&lqZwTx-mLLnyJ!u^I7Q~!VL+~bB_?=HS
zp^{zqfaApJ-QmkME-B_ckI`Bd+z7>F0cTSo0O;Mw;40Fg)T8+UWoXtO`p+OjQMcNF
zwevIT=Cy*xaiQZ2PlW$~>wIQkqWaJ=Vkf~~hl|G8p2B@8p?0ZRl7Td;2HFc~0=mGX
z)!d9cN*VCmRn2~alb>i6az69iPd$AVP`%_eHksFEqfyg55hr3I(7TYOs(WhQ;q%bw
zu2X<{t<BuMY*?jM8T)5bjQ@wK=n+(IRc~=m!v4I_Z-b~YITI#@t-;8!exT9mh}nH_
zGpz@oTWrSd+L0O@0mUsQA4#eWZIBiXf&}7AUKdbTZrZSLKlQ>QdKU{otA9{PJNiyk
z(t+yT3#qLbzCH&tzo^d&5XO>M9o&$FWH3Kz0ilBA{Q7h<nT1u|V!Y^U!gS%-DY6Iz
zPkr2AfasgwK5L5sSMSfb{u3$3BLos8M5};;E{TY!kwJ29gX=)>AHtPB#1>?!<Cc<_
zdJDB*YMp_o9ukLd<96oM&i+7gs4mI@HRGcd5T@pv&J-C15SO`>Z%Pf8FbV<KOv`79
zQoZ8DQ|u9YFKG<Zn$judQz!2DngLTMjig20_=cq<NIy$lM+ExFZ7M3JK`FZnQ@D?I
z5XSJ&q!W}m?X{dx99fmSo3^^2DC4O8vIk#}1@p^`iY_9loQk~?59{oPg$1t8r6!7+
zr02tAthpIAcvt+TW&n%MLmcER9r)$c&G7c67o*$8Bp^<Jw^56f-hPm9jm!uh?$?DL
z8}{oVE~ZypfN@T=PJ7C0=O;hQV~=7-Ow3@-kM2*+Ow(nmGoR#nEY*@)f%%lOHVQkA
zqTZ}UnHFVfT2=y}%A>o9>)lpJ?(*%IZ?G-}L}YY5eBsx6JnMLRh9Jp7AJm=M7H~mZ
zQ|b)e*NoLLH4RW5o|~7JF;WufzSjMvz~Qx6i?D6^iJzNT5ZS~+oKL2LVU<}Z(s4H_
z-GCjbQwK%}{kXC&8_}_DzD1Ot1`6_XuTu2#C=?vs!G<Lc5QElAu2?nArJgnBVPRmt
zEFf$nOFWbB^-|rbKCvaUB2+-Sws?vY*UO?hnZPfae(MlujM=Y8|BofG=UeG2+_}H5
zJP<Qp1#S7auDD&E5(VZGg)=3oTqXQxOMEzl;fkIu9o@0AyDo+3W#5E#*YeLou%eum
zz&Tl-4zb2su30b!%3>IAS*@~{%&?Z6VdjJ=y~cekvye?~fajn1G?qItRycu84<UCC
zAvfzp`jUx=!Vv$4&h3@=al;896_mHrQ-8-lxavS8_686f!zfIN3CnV}gDR#d7H#Al
ztY9Z&9L$BbV?wJj6|HwOH_E9`HE1hP-L+l9tx02GJnEa92&eG|o=YhCg-4d0B6{`S
z+@^p=$YKskV?=!4C<xEUPiudqT>Vu263MMbXJr_F2RWt`Uf0z$OfMzI$xqPP+{-bz
z)P{XqP0O-if-%YD4rYcxKAw;}k&rV{U6~K`ENjsq^|rkLl0H4r+ZRHP4e^(s&r;Pk
zGbdc?nFrqr&iRA4hpjH{NQkywu-?dq+sB95n*3--acvA%*6R7i5+4?1YIui4t*T}J
zzJkU~$*E?c4{!QeN*XXF4ctgY7u*Xcc?a_c?O4dMW3PU*FK42+)$9|2?Mc7C#%@<}
z4(!dIPSe71#&K#bhY{zBm6!>}Pnxs>S!Isg2Cq%3uk~3T;krsNC`kJpz4^u;b+$~)
zSf-id^{qEhQ<@pX)}2Q8v7j}RkqTGO!3>XC_Jj6(yn3)$cRaZ+0NOCYH0!i%Hk@E~
zq59{&htGcTP8qC203q#(@L^5USDD<(<5^8hkm{C%j5sV2;Vx0gt>-pB^hGL$t(vVa
zO0K?zZP4k&WP9}Up;vW0XnoeA4<}dZe`+{0POT<v$lA}nzrXgHu1`vSuh|i`7l^)F
zU)v$qHJ`jdcE|KS1dMjNB^SQj5EANnG-cU(H|owv$ZaK$^Fp=b58qMW>6UW{_~(Y}
zwh6^EBci{~_!|!%cnOLtieN&-Btea>{gG-T&l#Uda`fQYW3n2g%%5723}F5*%{ERY
zy_rSc585-kyw~MF`O9sDoax|f_Fj)SV~FpxGgk<{M%L%I;hZk@DLb<UYo|YkE$#$w
z>hbzXkC;3Taw6sn$^#>sx^^<Ihv(BG58w*<r6wktFR>OKQM+<ti>)5iq^I=xif;Fv
zB~r!Te6#fIoXkU%Z`d#+?p-^o_2$HM#F$b^o1^ETpOzN+CqFTHBcf<8Ra91Sc-;im
z(NHb>`H&#3EWb^5&>a#z1#CPS$wI>MT^ds=FLiu7gBD*(n9~^>kT5!5*EQr$_NKcC
z$vqlOLQNp)IgOmQJ=^(-_s>yP95RXVSc-45`zFB?vGS2h>_-2Kr$QqjPn4JUdLYgI
zoy#|gsWWsBVP?2V*XLDDZgA)x)kUR*W{Cvxg74!&LYU=OZmJ`6cCkz$j^>%O&_l?W
z-7`wDOA{s>T??nkA><}^abLohj}+>t3fNs`a>e({c|;v&Mzt)8xGV4Uajr~$47Yqu
zp)e-j+E-h`!OHluu<i<3b9hnS77!&T#}s`@UzB`t|8kGP#)4;xuk;s0_g9nmZzb&h
zs{ktOtqf{l_ybjl?;~NNd*;nVwx3D!S>&yU5m>4KCXpzJ*a#Kl35>3}lUK;<d268Y
zO#}%iY6h+f#fG`?<qo{4hz31}pvIVTl#$HiiW(E2ML)e$i$(7)l7}PXh9{aoJigh9
z#~m_PHtJ9kknIT<lg?9Nk<n<4o>OJKs8Jp2@`_8FDr|eb{+bna`J4{+j8S<;aA;2k
z7=E)~_7kXeA4-zpfHBBOa9;TCO`-5mwcrU>zZ7$Fw(17)WsR8-7uHh5$0a6_cSUW%
z9`ZSFRg~R1(rfdgR!KF?-~u@jV&=k(bsk!PGFlE=S-q<nLrn=$84A+FJuPXYTNiV<
zoeYW8U*{~Yv)|W<0UD&$jCj&`$)shK+3ZEk=dsZ)!Nv64<QVQt=!BTU5%lCU3I)bH
z8izR7-bcT$7&K=12)|yJ5;c1+LF{85t(Mb>Z?Vy;60u~ftb0dc^Twxt#tuAlC2!L*
zPjc1U{#v6$cEsTNh%ed4uwvxantADY<;TltwE?i%nN<X&ApGl8^MMyyFU{1I(DVb5
z+y*UpCQBFHSS!y4x0CWhFr0j@x_GkqW;18pbN_wwZ6}TnzX{*4%a{@sL0U=w6u}SQ
zjh?o0dOg?Lx`?}&OBLroaLZ_AlCKrNH)|uE79+@AM2cN$7K+JpnD^oXA@tzoilxK$
zGA>h$QufzhY$#Pu$<$04(pVwxkFBZ8m5A2QwWi}jyNB=yV>psLqsUsZV3{6o8DnpL
zw-BsDFUXTyI4Pq1CC!$K+q+?4CCG+iN(yRIe+){E+6L=<MaE0~)H^Pz;N6Q3nSkX~
zt9uC2kJOz%aGs0O`f{m=)TAQJex$_hqm0IZy=V>N?8jQ@(E-43tt!flU-jwe@C3Y)
zl@EWrPjgDfPU6rQ;SH9~O~VlBEL!n10lN&bR+(9*(y9^f_Q@+gu@cTOb)7s;P%ip<
z(lWs<E5=~;20A>7JTf8M0WJ9%SMjZIqw~$-M@44G3FHMvIJq(-k8eqOt(ls=b)ApN
z3T9`qnwH=yxLoiZe5T1p9GDFME5+gM5ucVDV|z`*4P;XcjN9%BS(Y~0O}G^+bi0v(
zr6j-0wVP$dF~*AbYO%?%Hl-!>5i-p)!zlBXS<HHo6wC8hm<8TGZ39H+nTGoUXQ1OW
zPoc?&_Q=oG<8SO~{~g`>?{D~pVlu$zhF2v&|B26CufiYTDJr~Ub-fK4^6>(i>KhJB
zp5Zn6C`&1dJ~dGh-z}yr+xdH(=3--FMNa5yyeBBB9vV@xN$pA9L`H1LCNScNdxoVo
zWF+*9;k|s~(#83>;86M(0b~xn&ppIcbh2<HR6}75OVpZ4*_&W%uH*zK>EkoER&h=h
z6NWy6Ov@o(Ukeoht^T-Q>S02{JUQs;DCQZX&nFL{xAdJMYSjA-w1PY7B?SUQP--qh
zkza&NNEi+=+AcX!OA&|x^Q!FYVC1z<Oi9SqH{M@L(T*SB1<9b<P-e7P*W7Cjl@T@z
z>-Pv2Jw<g|cF`UF*~P_4&UfGQ<<e<@z`tbgIcPeh8h&>Ax)!iA?cK$f`E;m=&Dzu<
zrcsl{9sJn?fF2oIjqrH2)>gHO4O^A-)}0CzCR}#sc<_I?MoEuR->W+yMfMAwsyG9Y
zNJnc>MHqe=WEiz8RNDqpw>j^mGSiHV>1;w7NhzcKh9bw5Dxc>oxbk|xmB|G_Q#vbo
zZBr3{L3*f`Lub30&MKTM;=y8blIIBHZJj(sjuWP1iXJ%%(pQYkUbWhbT!$BWBSW<A
z%D{1b$F4cR6soo_PA3(LD5*9OZZ+|Piu1^hDcMYg-QDL>MaJq{j9udQ-?n)x$>Jlr
zl{$~h@+dyV^TZfMXoUDHzoAUzirC+)lwdHFcG_&iph)|qjKkJ@-w7(7P_1dd!X>Cp
z_OWy3a!zGx>s~GJv-{?7azOt+l0C}<h#tCJaL_qdCx7E}Q_S(kon}sA<7|u;=erD}
z*QZKAs=`$g-kschaM~OTb2%E+a2#=jTun=nW^1)YSIrdF0AAI==g+UhTjOUp@olBI
zHZ0e{>MEVqP#I5-mfr%ORFT;!%bU$4m+kJFI-|FH<VtT}iHPlse)xY_oS9=eFV&c&
zN&bL_sC7w@9&|!xEg5kVr=61_KIJR7DNTCr<qNG%nTmWHnRReFEORS(KTyqW5luoI
zhc0CRg3TVdwtjti@>HnSev9+_!yd>bTrWD*m#L7(yl_z7__{CG(y3fJC?0|(i^O^^
zhKr+hc#txmvpJpAO<<0~GR-{<Hrxm31}BvZ3~*E>X*20m4bFqdPM*mAVmQhP!({MP
z^>}`{3JZr#K;XKd#-!P0E9F8*59b*dHL7R?#roCxJ*Qa!o#$s>vDS((*IaeSEvqzc
zgjG<y?9M>a9l6kZy$c3pDd4S691zazf;y8>*P{K|38^$2W~VJZp!mXCP7V)RYwG(l
zi?$DDxDx`;2WVNyg3A56oy;pK%XfH5YuD^{Hrcj+=3p*-MNXV`N3kP+o;aLjZeQ+=
z$|&FB7T<&>i{|l45{#6fr+yewfg)~V3>CYC<R^BytuNO|VJDex(x!@55!8(tZg=QE
z=eYD44m^ha-;z{AxkmbPDkkoN&{CWqceTWnI_BA)34mH;Rn;GxhbvHpHJFV;cm9ki
zWB@qzs9wh)nK3?%;IP$NW4Q=+JZSECV6-<&L2zC(#|z&=LsM&LMd`naimb{P{OG}I
zadI#>UBpLoO~*j57)x=0KreG)5EOVEJsfSwBkn^CKpp#$tic55p)G?U#&`}BOkM+{
zB0Y>Fx5T?`<z*<c(xiXuyJgZdS*`2hGlB%luz(NWp-;!TpsW*rR1R8MW40jHOw5dI
z3?MZsyv9?@`9sj=O3Gcc1`+DHj&<k@!e4h8&C5_dd8RUi#rPM?0}9R5kt{X^fS*#j
z4F7?D!ZROKodK<|4TTz=#!t@(5hQ-pU~!wB_)Y+JV?Kath=!6A?E+mNGpDrxTtJWG
zCn0|7R^BZv<=2L&knLVj8}UG1Ox=AAm2=HT2NRR{m~VH6yc1`g7lZj@Ha<kibkQ8S
z$oDd0iEbPKLzp5Ftje;ys-(9m<FeAL8wZ1Hcu<Q-qK_lh%yMT{UKy4HUt@~mGJ8I>
zAZT<x%a3b1iDNjJTA{Z)e|sLRnRGI{5gKe<C^i7>?HCF}Tojsx(J^zkWNu2+LXU7M
z2?s*5+AjCJxP8nZ#&Pzc1mdjLs!Opv1K<_Y8lPhf5}bKd>4?<MRKrvwR7Lp2JgX$;
z?W-Ckh-h(9&kw20Qr*nE#SZ~HNr<m@(742sJwq&Hxi4<hXV53(j&Ws~>Zvd0te@}^
zGFUpM*j719#okXULB?ZoY)uT!VI^gE*@==PGTQ3zrt|f-NW~Mlr;!%k@AV84jRE?9
z*4pB(RF1h7fxt3LESDyxWMKlog<tM4(Xj9yfI!!?Po)$2KTNLYbF9E#I<+X-oInp`
zpi~0e>U&5q<aS3Ce~QL9GNSOKWO?29jdEqN2mM=kt;{yCfkj9yi=<zux3z4%ACjBT
z??So+<;m0pOJ~g^lQ+gKtU{wwkl|5-wrkeiERL2bp8<Ic%*ViiLrss>FbX$01*5Rs
zVobCryQ9AX#?Fi+d87^GRS8_vsL8yyM<?+jFU=i$m%?AA=igTshpdRl&u#5KB>%M&
zf%*SP)dQ6>-|NxcwW6tr_@5z+=pkfenHeVM0xqgYo6?|S!6e!_-C(MB*a9lWAXI7i
znNp|)TlXEYjUjJt!87xqSGs+QZ`QQ|MP#q%S7OtkU91mJ=t+rs0@V~mVe%F&7+KwI
z&|Ww4tFJvURek1XdJ2$Vf&8HEa#U7F7e=nV2Si<ycJXtuw=kSFt=`5{y`PW_=S#SV
z%L(>GFG}__57TH_07X;t72vOTt@~~QQ2u%!eNmK2OkNqF@GY51&$aIQmladABVJX_
zT<vuO#BLfkTKiy%@r<&2n8qBjtq`0i305^4S5*1X_oBF#;54{|b`TpW<>jysmgJ17
z=(oMDo6mr&>vI~mm9#$=Q9_+D^h#UAK>sswtBD8OlR4c<<rm5)-Ar-Hr{7$7t>=bm
zzgVDLU<ufb<<LHbsa{Mg9?OeXc&=3%z=yRM^0V;gEX)DTmv|Wghc=MlN00O*O-5`c
z+;WUPL7!?}Tzk!)DC(ztO2$?4;Ctck(XEinF!iwD+zax`H?M|G1ynq&?sC_8=}%aB
zcMMCbcguEEs*bAs=%N19OOrT~Cf7{`WJM(j&K99tkSSk<kdGNpZD|>B&Xq464-CQO
z&0K%^v&D7C0u9*wH#d<6es&acMuKCB&s$e~v-@^@G2H_{<W*hrwMoUP2<$DWAordp
zO$ppxoFmWum0P`CrV_3w>k^y@U%(oEZLm;lFkrd3r*-DQ9Q||ux<i4Vz>AVmayvgF
zkT91Ynk#X4!EjemPtbDc<W1JecIr=z3hqqkjx@_+k&@<Of>ZojvXfF0k}qO#RVjuL
zV}8}1!7|`Slw?m_x3Rmxn#m9ZJ(%U|#LTzp071h2-PFw;oX7gGLulH5Z(K%*+zb53
zWfr|Ek@e?g=19VHD0JqQD35PCxR=Tna7aE3Bv~K&)B~vXanZ9nGrf;R177XGdp1mW
ze_d~}H(Cq#8X-@dI266KR?~f%m~=y3-f@!jR-4pk4tZBKaykvr?|ywkak>>c)8us~
z<KPDh2{(L;Cx)eO`#6ooKk`7O2roCBEudyf2I|alOugDy%etI*4X>{6+wkNFQ_V8-
z<#(f*9z9-SF*#vhl_o<YZ^-GVC8Wv_U-!K@Oi+&Qt2E|dz8P%N%31$hiz}T>gz3K0
z29satM@4;6GLibn_T(xIK_QWbFdJ-6F{Ff+kcedGFg;c&Q90skI{eTuOj1AJSWf?J
z&&!$^vr!aEnx9qg#o)8SJ6X;8Ls&N}$9CX2M>yu9D!ERWTY2X?4WornorE~T=U#P#
zKur|thAuRL3^k7aI*0c~#6;XkxRk_+6}Fw^j&}<riG>Tr-A2s=yHmnkZt{i3naN3I
zs71fttz1jzH5iey_^6D2NNDK*a6J^Ir5MnK;@?7b6znp;x1;u5W9nAA)|NHQl2<vH
zyqU0>xtSrRTft<$E#<~eG!Gn6`yEk{$s1!BvPy{F=DP=ci)BkBcQAU1QquPaQ?km7
z??#38y#`vUF;iCuKdW~&My5S<7%aRLCfD?`WFE93KAR0oyn}>tOORHxFgS<Hx!+81
z4BTgjk}$_Hs`_q2Op>X6dJ@Ba=lODOY?{YvO`5Ef4_8NiO^NU^o9|4z)qj7C{`%Vs
zUl9?7i=$}JpJb}#v1gIu9;oIS@yeP@d3(b4k>|xm#B9Izkr$;TJLD&aP4Oe9Ba?IU
zeXIG5BQotSr4wabr&2U(@*JnHc^JKhUg$@OUAsw#gB{<@)E<&)csI*Ee9l3~7R#Q@
zDu<eC^i@=(?ZC*?GRuuIs|BguuU(!mJ|PpMcJ)Mxi(h6_vfYgQ5}tEtHmbq~MY!2L
z+k9*Ya4@a<bk^B4+A*NV<W!Cn942G9r?kH(yGHQ&bt^T$6m@)UVWwMJkYk0An#HLR
zZ_DpxgXMFT{a5X^2?k`(qsSQ3&q>OUE5vqtguTbg?Pxm^1QmsPBVS%CXdYWv47dS=
zmawfE+K8@pH;JFdyV0FFMbYZ`KxJvB*mKmHw9jCl+^z09RY{s_JaoQ^b#2BR;VEqM
z$N{m@7dJN3y505S;<x;#g~~??nuiZ)^*AQ3^|HCizYp(HD^HhCyU%MpQI^JO($cb`
zB|mlz2$Ne_cjj*ygCTnIovep`scHWH<}*MfGP(6pVSmxxRq@PU6X|yvfYp!8Idk6J
znxFKL5wYf!I`$aUPI9Hdq(8A!JU>Lia%MgqWz$TYMl!sy_W^^pq2K<0s^6%45k7i;
zF6HZAze)T(Vt}8mIBH8FqIdwzaZe*(-81x}ennz<*RnfR(FsO|)a7MphPZ;xg6>rk
z!_d2!2OLCAjPVi1eRgyI;ATQT;xQjUe>Y0+pFjQABbed!1pP!eGOSCTfJK|n_MvrM
z-XgTs3nb<S+$oniU-&($y3Lw{Xf3Q^WS}N@cW*&WX2b*GPhQ6@ptF5yRBTSv?^(5n
zIo`<2XyXeo25FODL1}QAtOOArynq^P5oSLWA>4hz%EGUg;a`+a|MJHR`N%7xxhiG)
z*8+&tlR(*JBs3myX#pxcpJFBH0)FOan%3p3$cdb>kqbi&rA3-|kyzw{1^EkaN%+8V
zSd&);Ek!BP92M9vh}=GNyq!OD(jyLfWlg@7{!H6@K+NWV$ELr#?k@!H?|->NvLMxX
zVqC7o_LrikozD~S#o-T7>_v%J*@=b938o=*kCAl{18Aih{E!K%>doJQ1vy2vhj~x;
z)g#1@d>>?Jz+~h4!>=>euJouV^Imzv`qx_gt)A${BiN2F6`zI?|07@Zcl+$|d06gu
z`D5q#{`n*R+I|0}$~try`MwguN(cY?zW=gB|M*AhWMo%&`K9##&A<EiPxsf~Uf@BN
zo@Vvt`M;L_zhhj#|KQt^C=pWrTkNs=|9s8=&A(994w3pcCf#R~{&oTW>05Z}AqtQS
znJ#{Z|84>P@mu^qwy*)J3?Cs~!USop+s{b;?^k#au`ttxT1w^q!<UEK-?f*M^I^Ro
zK%?~2Rpl(e^p{BTpBC)DuM(tltPl287-v`Ry2gd+XB;G=zqi1>1(AAPtos<0{?}Xo
zL6r#e1t2GfARMrRb>LLGXcclqC`r33Q851A@%uM>@{etQNQD@Vydd{k8QfgxLZB+c
z?YOW9H~%l>m+vCs5Yf}#pgjG*K1B9kgG1y|<&pD5|NBG4069c#CzT!k*H4mq8#zRF
z1z(Z;=d1bW>2)ChDdg+I<vae@59;<;5cl%0&p$)`I+1?6H2?ksRTmH_hFa~a_@7YB
zzyHAR4>HfU@^FZJ%)TN0AAj0k|JeV(u>a9B`+ss_4~UcfROGI-Uj7+v2yuoM3KP$7
z{gP)QPwuwGF&9wXErML75YcKH$KIXF4d<BUVGwcWA$oNApWpet5pb*!Y4PqoYhYcr
z^cXxwI!i?#ACeCOtk{zt5C|Exz~6G*1C(_E5>E30Qs{%ES=S+x#6SxWZsm}%M(1EE
z02&wt7-&0c6cj$zh(={G^7PLmBQ>@PM2J=pZ7nT!<^7eFe(3QO+!<^BFllp=MT0rn
zshW(nAJYErAW2otInfLEhz&}wKCaeou`a2w#=mNJCPAdI)E++&g)<|2c3B=GQgp|k
z;r@N8Fte9<2Sefp(!2$b02UxQmwHPPxchgvAk(VntM=N@j8+KVX?fx}1k7Dc#rN_r
z13t&j&TXe(aaqW%UjRTjL%amEHbcbZ91231FUV*XBx=&@JFvMl8!up%>{zje21~4q
zKEm5JDXumy!IQ9tSv^U}Yy@7SQBVaM0&f_)Gr@Xo);6l1?p{U=U^NW8S`Zq!2%6Z(
zv!6Ei9=&qgV|$!l_Gm9x<+dHr6-#Eee$B@uY4woIJBx2K9cz#vF#!5SMWOdcey*qk
zvkmy>29T+$S*)7t2x7(uT1QDApd1G2?oI^z+*s*D^weY0a)GkM(B%YW0``ieRtB{i
zN|6&&%p4$w(8Fj%N1e>Q0V0>hC$!HDgRA<pp0Z7kK@Y)jUlP%F*D|hd>?-VV)@u(>
z%ia-T`U)wRNIJPWqG+9cRQF4?O%#0jBGG@po&T#E6^TzF+0-Fj2LbE{B;j?duX6Xq
zUAjxg0a9g`LAG6=TcZyK3%llJT9}`_2I4oCYDC}6!Y~9TWK=!sIl@b`kW0ExGDY&k
z6vA5x5+x96ID`Rz$m7Oe!dj+?#&E|nL@9{@=s$`&QJqTs3>}P5sep%m2MKxXhd7(T
z3m@2oxk?u<4uOu(vhL&^9x`q<VCE@^tU%-|e4hgJaoJC(?SLOHG6)i2$=K7hiIYCh
zu{RdlptaX=I~;RZ(bk2lyXyv9@P+7%Mqu+Ij?>o=9>5-F{71CdVNweul?T+K-BT5`
z6Ht=ax1`}Vn$+^}v&_>VGe&o*R{~(^{R14f$5*~$wJePg<iGl$#UioOX5fQ4sqcs0
zJk<acY>Shg=@1cVz-i(<zsym}&8@db?_Y32F^4CU)H%7U2%yt<uKcML=jFuVqN)8V
zCnNp05|R$J@ULc!L5^O~Q#2l8LW|{*?AQaw#*HU_doFmgAS-o_+!hr$`!YS+%l2m^
z<X%S@cIZ)}#m)~cTLj`xRI%BDtX~@J4AX@cE0{(!lSg8CRjn4=y9`j;ryT8bZ(n+$
zU6(F`!q?eXm+EUFEK81EijlJs@5{(aeGNd-uJt2Bog=Q_hs{6qV`uco00=odncE1$
z-%!$6rhKL^puVQ^$eSvH75T0X1YaYhvl;-cZQUK`#Vr0E<O7czY{1a(1d4<c^0m9*
zXriQgMnUX!{m}vVhMpJ_rtaRd@9DT5!slpR{U*X^gn5msjM#(B_Pf<<+6UlVRHYLF
zd{G=UlXB7iaohzg8t<Y$!|pm|Z}Vf?2Vw2qr9_$)o7f6#h7jfi$09x~e5M_@s#vSd
zq4xvHX+m6~N2N90pS}F16;x8ZxyDVcAMMzyF-}Lr@bPvk=kjk;DJaRll~UwHo?Lti
zQWzw}2?;(`yV{dD4rzrKv))%%(dnpMKSS2bM0^Frk@{pf-?e<414;-3vUHYX!yvt&
zYW~u+sdLF0=(bv=GR-yh-lhD7#E9;mi*MZ*;RadBIT+sc94t+)%p~28KR_*vWXLus
zIt<J)<4Mj#h3kg5zcD+hh_C*Hyu$GiiMshk-eWUV7{an0xi8a{@;<$piXneEy3yu5
zSFOfyHP!#wqae~VS7%7+(a0f89Y#Wk^gl+7Ra<ST?a8IR#mzuMd#gZP_je_nCnJ&9
z18_iONr^Gc-_MJUH+qm?7R<3`>icu@Y{gnZ_KB?ZQ7bCqrs_@*&>!plYS{4w7SfFK
z8OOwIZ5xczJSEc>4FOiUhpa#eDJetr9$?*ErjJ0%ok+M|-o61lc7O}(N#iSZaA0xu
zGj#JNuEiCgV;BUHM%UzK0tBov!hY^eeE}ykM}$;F(pzS{b|*qG2)@<F`+WV0D<2{q
zh9%+7wg7#(W0ED0lew=eXCo5q^0=U6Y@hhjTgdvOwjV&0<6!C1RnH3-*9t5j5zwgZ
z(O!p;>*pDxO?lyUZW<9fAgQ;6aT<#lzo7T#yA?35Eb&lYce-8k!9^q@V9ZrxpsP#}
zA}Nwh3zU!e?f@;Ijrb=R%OT&nnkS{PDd98cH^m1T3v7!gu<9@M1}oQVxfWn*8?VL;
zLMF)CntWgv()g@;#VEaW(q>0B$nL6P!7>uOW8MIxQ41Sc<F0UAhl^$zap(6#&0v&g
zHL#$Y9%C1_(%!aI{{4eFVp&~O(l8%__iFyCQ~Q&{$o#q1>Vm6-?e2*j?D&|g#2t3;
z*`8+Tvm@85CGag_mJot)c<Tv=bo%LLckrJV0i98X(F`O4|MLiLc09Dg?u}$`NMfD^
zCwnJbXp<mlHN{Rh&81%bY&!8RH)u}t`-q)c<sXg!x$os~?)2`8l5<j0<-X9UpU>t~
z<iLyd!@Z=Zjtk|jM2_0~moq@8rl>aet|UWq>>iQ>u1hzmN-LgnH5U{OC(7R}%zsd)
z8uhbE@MSreK9`q@Nf~8DxuZvboVV7Lt0*=sYn%U|ZMhCw_khCWE9OTufKpE=mxVjK
z*2HX!X0$tQk_S6`HXcB$ejeul@Fd-rH2u*3rzap)N3BPW+kFSqFkUejH<kbIehm?I
zU$tw;AHnFaL7dR4B8Pm6Wx_~_x`Ef!(bh|2@f*&tyd1`vh1uNB08OTzRx4MYv@Y45
zn^t{ayBM%j>#A68X3@RlKiFTnyP8dHz~eFVh$67k{ZkXzh4uIbO%gFuY2|5O*yHS4
zL5|vF3uxaEglU<XqU)PLUd}dH`B|8^V)nj1l!)2hoL@{bn!gj?NNO7fi|a4+^2dD9
zGEyxyIhHfG<!KlDuYP!HS`r6b>XxSP_#ZaTsw&oaXU@%b<u9p7h3OfmXV*VXVshL)
zD3au-9n@QzAK1BZ@mO8A*;*gn`=+z{<qfXD9O^%OMv>jY6zu)Lb(R}QxVq*YB(e%(
z{6{qpy3W3$Cp(yqf0*g@0N&&~Wo2)`q^0jW3>vyK9Q>ScyX)xA^q^+m#YpIfIE8i@
zvitkGdS>851L@kQ@sh53&Y&Up=2zxA6Ml53p>ywjU(RnKDiP>YJyN9`<AO*H`Y7;$
zn04NSAPgw%HY)Gv&*bLe5<c&V>yUDbFJ>AGR=5*onoOqnM$8grBb8$hxxclj@3r&G
z6=PSSsv8r}$O?9`1P&@_NpsAlHJd#%KO_dKR(D3y#*OyvGuM5@E;fa7tg+F%+Gd?{
z^HpRz)DfMv9j3!N`RoLa>>_hmNW2G>XXJDQ@)51n*l%w=3y~|e<;mnXkQ$6cP9wVX
zbx*cj{j<k{bX7({03}-+u%i>CU&1pwAA56Alo**$b+zWX85(t^NjmBCQ>N6A2ZR~H
z^OD@LPTzZUjWRW>d3>Y4dHOV60=S_!VSCZq<CJI$q-zO%KE2A|L$V)5MHwS@2sKp(
zDjwS}3I88gZygoY8@B&S2q*}HC?GWq4I-t|-3SgMC6W>XQj*dwozmSB($Z2QjnduS
z%}@gazh}PhdC&Q+BY){yx|TD0Kl|SIbzPr}A8{1O?pXBtRV%vO$FaU_3QYDnL~LQw
z>L#|pA+cxj5UaP)>j!in90wh_PiMTqI|zM717BRCD&3<~)>F>0d}*_X9txJT-3l*Q
zRm%Oz9#*=8I35}U+fltu@-Xa`8C`Vz8Mt4USZq|(rWUN|5)@DCfk2cTAEoX!kyz*v
zj`ax(OeI>g8%29&1jCb=4Xh0y7xt`OWJ=jee)%Pe!+h-p>5p*MMrRgJ?%~9!j;c>2
z+h33(Ii5^Ih1=VvVpmBX0HR$iQ|1~|r@oTo|NY@3^G3}k`OU)00;1Tf?{sfO6ovC`
z&%Ln-gFLEj?%~D7FGq>XVJdiV<+*EY8B6G5%I1|d{ve=GZr8C=N4-X=E@z=T2HS%L
z^FeA~&uMR5E+RjKcW7US{dYVvYMm@NOdVcfwmRCWHr{NfejUGa;HfvJ!``_UfvWc$
z4SYGi`NU^`?!~;>wL0a*fuV@k8~udV>Lfg&uCbYtP7063%%k@x&GY9JrVNOuf+EkO
z?*Nx<_zndhL-x~%mq2VoiFipDR45NTA{0Js-5hA}iXx#b;m`E^43^G;;`NEjdr%Nn
zbOMK6`($B-f>GS^i17CnMUMLRcrf9iO=T`(Z~95P&}$cDig)Kfq5Zl)AhLOrsQ&^S
zc5lyXUDu0Dr_~3zQM&6sQO*08r&sdufoBe4Xv8J(7K=;_*CkyYv<@6C4rHz8gZx&D
z24rq$2g@97Gxte{3?8_c^Hx!;(YWmn#V9uTFL~DEfAc)utP(9UBT>gf6Z)tJuIw9K
z1IjkM`0=V<w?fN_ST^n|+!%h}6TXz1TjivkX_(iC<Yj<3SWig`Lrbvo4JIQB{KeNd
zpG7Y5zpoMn0VLDjD3`UD&v)g8m==UQN4ES89p5F9?<&z>`o@;o2=~j-=~d3(F}LN)
zH+in0S_|z^WA++Ky@t6y3u`G>;5e4I-pg9&t8u!PaXjv+*jv4pF|q3->1ioaF0E6(
zc?hfTX1G*b46wb~)@-?-egk1>D|%h$%~fAcccaE^SRkkc?7Ds~e}WGE!M+Bpr4UZL
z<K!g*qtIN27UBq7P5Z2ySwrjNgdf~|^<xPEYPzkDRw~!IK)>TjTuUyT16#qGo9&XQ
zWR;WT8>d35&h;o)1e<Ui5t$_;nKZ+{77l`7o(ueWmO%VM!wB0U=|8K2tbzL&FZ9vo
z61a8R)63Aq)p+liVN$*d2Nfl|9|k=gf<1Q)bdNLLAY>02y_TSt=*A6-uH&aStZ5p6
zT1i!JfX;dV+%@n*0Cw1TC=wFt9=YawYsS|QIgA3J@V60mI(t;Vyx9r~dNcO1)J-`)
zm4x=*<y&e2k`*JF853F7_&$ccIGF+iSsf$UIs+M8CIx~uH&dXe`{wa^fHf-*Yjd_L
z|5w``xgMFvfG1Vv6gLwPDm@(OiOAvjP;~s+=hNGi(Wd16u2`>A`_t!fSO3LpR9+#P
zWX&5O4DVQ58f?fk$&A}kSuMs&)+nxe#`-$43I1HSPh{fbGGhx?l1VCJ8XgUCL%9_U
zw@wh;+eyaDF1Dco-ND4!`xc{-_^aE!S$>9d!NHT}#k>9Wi_e!w>q;tQjjB4a$CRC{
z*EqMkp%mDrR_dn51Mo|<q8$^u`3>JZTbE54IU+!IAR{musQWIF^M{o>Q=I;3!2+*!
zFi&sB%vaA^Kxbs_(P#RC(9VuGFqhj}0CP}ub10cnqxuywJ{diZGD`0=z6&>Vy*NvK
zxghc_YZ1R?t*C>oI{NuO*-@|3@<}e)RM%aPlI-{*s+ME?liGFfb<xGMi3L<+QGH!Z
zh5qAZBHkE$z8J9!MKlC%R=H4rO>{due3jH>h3}^}cX2VHy)=b|PHvRWbu<k?^O0wq
z3g_%1e(9a$q<>*4-p2|xdJVECIx#cl>+i|9m}}wWD4p|k!)Ls%={fcCMtW1`dPhX;
z!@!VBw)9BPzPR%N*;GDIB*n~WoR%Eyj!fp()aQssWqQ$zVIN++&Dv^Dp$lB5XSawH
zTCoJ{xUhZb%dSm<s!sXaOwLPO`yR1w1&%vJx3cFICCjTi19)|dqUqF|Q{mcqds8z0
zfZ*WzTD%48)U${!-!uY=3Z`_!MrgN-!NV3VU7hxz&mx-IuekDnq$iY4cK+d^v<T@v
zHIebKj$)$AT!g3W00*e)*sSU8McHXjB6qXdwsS^!hseH;2;FUxvR`?4lpeoe;347S
z+)-GTi5oj2KC3S&-jq1{=aWb-hmZ*WkkrTENs4+f)^DH`PcOJL$-3-ZX+?p^Id1(1
zc!uY)WrY8ZW8G$t;ENvFTKLKwG<b-7{_?|5<i=hx&~buKXtN|SmnxDon4BmY@leHN
zFjQ5fz-IlpkG@Ixl=Jvk<DGFwSBTeN?k@jHDNgeLksRMBXVx0kw&S`+<pn|>K6;df
z4Gi0ydi$TkID~4+$uU`KogZgsR*Jvo3}+OMI+V+Pd=HN7HE-4sK(y8EzUWdk$QRy(
zy`555CtQ;_Nn6ja3{r48EOT3K<jm3Rm)v{DIr=z|6YJf{&Q`xTYhdr92E#Y)B!b`X
z>U9f*KUv%`+Qe3~V;2`uiL8yYr~11*e7q&|?x!1bFbTtG)SYT~&O&HD5+nbLcJB6t
zD=hy?>{R65_fcgEfzu%nq%?e|y8x0Pd3k~{o7KCc1B|V2evP5Dzrdl3EXC77+D#D_
zp_Z``{*E%~Sb4AOWOg7YudeeJATs$#4Yxr#3APOF3bu6Xdd{QIAWkFj4Fod<I!%I~
zrwE9!Pq+^j>@_$hKn}O`nl3Qj0+T0={<yFw)|Sa<t=yu|*7e-Ig+$H>F1<BQ!P;-+
z(EYl+-`QVGt}eeM`ZshOjh6(-HCO$ELJPwizWy__7wHkV+0H5A9qQ3Z?aQry=llJ}
zF6}0sk_Dx6EtnMggZ#=dBQuf(aV?(RSc#|H=HOwuNXi+5OcZS{m@X835kCjqh}8v*
z(!yYaWH?{sf{fVrij6#W*LM^^aDSC<MvgSa?_VD4WL3q9{re%4Pb4;$|B~s|b;6$w
zuH*>@vjm8&VJAo~E(tb4yUi=mg3#bS41Ezs&y>x1_FDzo5+dYrMa{#L#|Pobd4-X7
zOn@|81v-nE^#w)NMei-2e)y_qwxd_>?lr;AaqT82J|NHVYNAVd4%NS-`_~@;BSNqd
z)^MnsS|Psl{X2K?LlmSX)YcGLRh>Gslk}Mj2xLi)fMUyQ#L~IJ7q&>xiMo!{-+GUZ
zZ}l<NBVT%#M0l4ClA}s=pJAN$jg84X=2lmMrqu+q^{PZj9mpVN7SUAdzmDPaHt}U3
z+foALT1!beB7Wl@Nur<=d`^Aq{KCgA*O;$JG1eTy^#U%xc^|!bPn3$zN>5`QX@JN9
z{jnyZV|R-&M~`j?=5=}3yKPINwt0F~aXbBHD|8^eZazez*Y?xlaHvkxqp>?BJQtNh
z3Zw)S_1DK{X^!KCM>db(oG(s4E5<8*{!bhF-y5hhUbAZPZI}vC;c59AAc3>EcHKnF
z(LCWKLbPSZjntLketf8Y8W&@rcT&KqOmPBt=^qW%Yg+7Xi!u}ixj{Yut8ax0n2EA*
zbH!qoP`Q6P%*%d0Zwa;d-v}@gTC5qv-o&@7t!IN&Z3q9%TmnP`%JM(qrLUYS<fJbU
zMkTvyp0@m8S^DYRca2L>V<CnS?{v=Y*Vv^G9k!ZY<w&2*shdY8?C5K|F$FV)mSOM2
zsud~btJtgtPJB7Lx@9NYs(ax^4@rJ_@b&%NR&Ik=Y%0Ps_bbTWn`Mz}_W4TNj)ZD!
z)qIM@9gD-3Z4PrF<2H(}?=7h8o;Ie3Ig?@{c|;z}lq?GZ)cQ(JfnlTi-p0a*%5?y7
zyr8{Vvt`s?U=OgWSx|);#93v>)bl48W#VDXpq}wTkS-smdUSL;COP|@Dut)h9?J#0
zyp-$N_8fx`@V<!Hq;K4H;AA0A$KLO7O9PdSo%WGb95z2V0`(FV+&Q0Cb|CCs!)ucB
zO+=)PG@ditE`r=Tg^?UREVFcH)iYD8;R3zq*?fQz-uH0p@vX7?d#KPORA*iF$j9@d
z4g{j{eajFdzYW8M1>#q(v&dB%$!5<%W4|g42af<;oVTLG@19y*)m13dw(wEimu!W5
zT10pp=T4HKFI*9Y7%ma|%#Dd*=FLk-p@^$Br-IYDm+GT(;Enm<Rl+yX?_<xc{JlLC
zT%6lQB0{#p&0G1sY;6EjeB9h7p0)PHha2zgC37!x$i$UYbil^i*7q0xN=}kEzq}#q
z{OYjYx0Se{l#GApUrTA^$<6ZkwNK4u!r3~f%U|Y@h8>4RzV!7rF?(3%W#3=^veOP!
zM>3kovOXTT3-AlI;B}w3pBikY*FPDZ;392xuJaFXhWq=$a>oAoN{8w?(X55-<wJ4D
zjq5A`W^4_({DJ?D|1*AI^ntJDA6+uiX|PeYX?EBDbJBFAHh1xd<s{EH$Nh%UBB24&
z1w@yLYng!hs2mf7wM*CR{a3Lw32^Ht@iFMzBY?zbK4Q+IxFX=v&+8<i=K*)UcmK7_
z=dgO~>KcwPZAvb$C99j`-+(DUSAXo&A|75pS>n;*t|PjzRk6_Cb-knbK3Q9q%S*-T
zCa?6eGG~3cF)_grx|AKyp#Yvax7-iy)3mquzR8{qQh8l01*ULr#Wro%+!zea`j*zq
zv?luPz@6;5-T;H0&bd6&qFspWk3L6|m-hz_mI8cDka-kLv()|k*s9CRyTc=STHT4m
z|M@M~@0uBFA^RO8U8HyD{l8)?!A=K#`RjcY){bTQH)}-L|4P`#zHR}r;WMDZa<uUh
z>Bw6EqTnn=dLeX&Mh!LpEI5%<?o%_u_K$6LYHIIxyxd=CW{UQ}a}?ob{{w%(CSFb6
zZcAvWIYv2i-+m%1k|JA53dIS{Pu~Ak$2cABNpw0|-}-{T;qc#VoBo{1Cq*({hv@ou
z{V$jwrD+rM)e0M0_w-*IWiJ<UTp`nPonL&c@91qf3DKVT<tTZwhQHmvhNW)}D;=X!
zt?q02bB#r>4c-SVhAc{l-t9^nove9V+_3qDzM;G!{Baj~09W40>E#{L6|;FswsCD!
zJ*Kf9|Ie_v*VD*;X3M7DXK>(o8GL;iu;9sTq-WGUkQ1!}iVtdCow97zJX*!rEy=%=
zDA!JZ;s^`hpbn86V1E%{Haq@27VNZNtlQ36nA12Y;#IFAv#HQksFx2XkKEfjJgTeS
z?_J2TaBJ3ol^S`T{;^-o;6>b@uij2TvU9pO-h&cP=srZRG2amTq|g6+@yZ%(F+(>h
z8xsDU68YSA%~P*PS7Lp*<J88A7ZenXCvb;Z+#4H@jt-NqAB&g{<`V`XruE0hokm`-
zsvCXGH+6M=A-wv{DJh_0j$>xcZVXJVD4UlyWACk~E<~u_onCI9f3`anVKsg2<SqZA
z5=r)c@d2rT1;?P}(*|k67g|51*GucRzjKy3-zi%6x^h1JMFoX2dlu9w6mEOCSyCmY
zVciPR(t1<f(VC4(OY3mdG{wM7Pf!0SE-qbl`;5jsapdO46O(6^fnJ(FjBZ9QWS5<_
zG`^=b$i>^6>SL;*r))r8yTD!Vpqsnc1k$e4)6>1;VG5xsKVMdU$qo9b9BsaKaFsw5
zwwqe<^I`t=3wq~5Q#~#P*7LFqs+-!{T2?v+o<p9>1^3HODw`NAD}uj4y)*O<+X`Fx
zF4xAj*(m?ue?Ny%-1WFQkIoJAMXXlA%!F>8nLbZKH8*vgq4@Z_ixHMMJ?Y5f<8O3Q
z+Bm*<l3nBsWcydTsV*Lrztj-Zcz}J)ikh6C?}sxA4nin?K&7VB)z<g-IrjGU^1AOQ
z{{FruB<_jHf5*lf_Y-DR_z%%a(J5S~&>v*a`G4TNEc|H`j_j}W@wWXciD@0YYtLFb
zu8@+F@`MW0Uvi_m^G-Ro`Kb4YgNZy$?A)1tHhj;v)S#!uC-qA!EA%qoORW={-8+d|
z2TZUqd@@!jCWX(jY2pN!E_$@EX%m`t{Hic;i3kaKoJq`FDT1l(nFIv{E=kMByuDe&
z1(=cJmsL%5VM|5f1EJ|_o^67hjXLcDKMQtEbISAYlw#(cs^~mTJbh}gSE1m&^6%g4
z@qFyp6gl7*Ha4Y}f~xWA-&?wq1->W*4-x~fo8^;Odfef|%;?<|S%=Z?_~e}oY-}~B
z0#UA#xwG0%AM}q5KNly2i>GMazLK{`&~-Ui+>+PXN7U${Zs)uP=HXb=uU{$WG1fw7
z)Nb-H9-VUW*9bLJ77Pd9iz!N8US2y0ZKf-=PM+{!Eh=+W&LCqS=ii{)QQqX_5wrDq
zVG7M&{g2zZg@uiq#+H^1T7m1@jiYBu^k;-O4>6LjDZdd-4j>ku#CtTQk~EwO3k$E>
zTb_TphOJ9s%wWw<?)<8%sfou*+!p6+%%L2Sn=}l2K99L7<mQ0o&~O@l>B7R9x{$ha
z`HbN!zp(IM35c)BW$s_H-^mVErc244=`)Tfsr0)ty88P06B83uRZFv7)~3SPwjQ3I
z>)srWDV(OJrokdse^0qPQ(LuxU3#uDM4R`RTp;q~RO_}gFXiITId(BQpUKhk^0FZv
z<&|-ofDF^)q^yUYF9na;o2!QljzQ;wNm^OhV;;^a4?68Pbj%F1@AUNtc-k8Fa7xxO
z$GoJRwbUyPR|a1eE!mATBwS$jaP_EQ8=iz$`a~tTH+dLI!S6e}ySeq937#2ZtYV7Z
z)f(r0v=>?7m6)&{{W-vDI??qY8M``V!UI*%cPjG5Bu$!oHQArtQpkN|V`F2g6LE2o
zxK_WYYq7Vzz1^42#--$K3A@5biN4ed3Li{=Gfqeo^L+JRA_9SkMG@rnr=_LsuAH+6
zJsCLtBQ#BNw|(5Wg(o3BBje{^KIIa!oluV%w~^XU6sqF|ZUjkMEh|wQupKF(hZsr9
zG6T=OI5{~_gwObXudHScSl%lq^x*A*9h8ODYzF9V_lNHvFnd@dj|YYmy2vAC*4q=^
zxCNBNND8(a=>-CSHI3E!U++C~3drZ-BFmPP1T3#RckbNYnlgq`{4b$vHx0Lu@#W>Y
zAB#FMySKM@_KPiFdHtdw&~nN_+jI|)1+NZ22tkLhk^4+hE~<#*wZCqBX{EQky5^(b
zAx#M`83!lHCj&icN%s8=rU&)sqL=RbX#%P|J)Fwv49cD^%OM!}{q@VBnXg!BjQY#0
z%Hdxk&qKdqQ_+*8fk*i4<>Ff+Ubeu!%P9(q2(B_*X8}S!WTJ8XWe3^-j-G5e&>Ud0
zmaDmg5Oi6AD_X<Vq5=uWjvc-8hMM#1NFchu9XSoRasQ*wb_2w0Qgn1H6j=95@ybqG
zK?qH^5hxEfB0I`(V8S&7=(4B4m|wFDlvfTEy8R{u5Q2*@Kpb{ef9J}!FfX1B9nLIX
zz!)hyMt8-Po_~j@O&_Izj76iUqKg(6HUWlIE9629+jVU2z-7QK<R9<M?k!zVKgc4r
z6gH#U$eJN@tRWy4SmTUd?pm?znelm{MazKpn3Fajgm7AQnU8n@nw3TJ$Ntp9y<?yf
zsUe<2foP*z0T$2`A2-y?T-9wsU!$hz$sg#=o1<nRAV1A0Y>56L{Psiwx=DTDET_&R
zq&E;383Ek}Qxo`PI)2TcmKr9R#H&m#S7(6cwds7cts{gj>ATk}7qs_e_aYBit=0O{
z1Rn*$9b-W)oH~^$XG@_NLLdEpv<zNWSOFNRV;OMdN(_&{$Ju=G&&D(7H0(xK3l-9G
zivIXd&@3*2X7x)qi19KS-{znXM8UYN)lS6L$Z0(QPL!V2AFfl*9|kOgs2>5RqvH)&
z@iim@eIyhZpYu<Jjf%h*{Oe-=ym_L5H2rg0WwNk~ou(jVQf}i)Yh9Rg^W{IA&f51a
zpd~(b86?L{cE0qJh!Othd?5vn7`-1`z<!c2UFxSk#b3Oab&J}p5w`1<&Ue+DAiWZH
zU!{k#YUnzu-`Yn&RlDr`-VBJCv0B*{?dKO)JaFGu%{cBU6lq?2X2Xic(@mQu&*zFZ
zEk7<}0YD?xYTGv(2O!zEyaK&-J_3u?1_MiF!HSMSjesoZrSl5q!aMilm<$53k)zZa
z3%>`vn-^4qevM#r2rY51G54a4h-&_HLAFr45a10nz;g%{uk|2>E&zCzYB9i4A^Cuw
zy+hu1OBjimun$CEu7c^Uo&C^-KQuIfKn_$qIQk0&Q@vfV@;;*PK=)%BqGh&n4|WjO
zzl&n&EJxR3H~$=d_4KjocO?}{kBehqA%9m4)YuLnzT}(N0Tlh8nGO<V;jD9o86e>5
z4(=nKNA17noszZ%IUdFkpdbHC<oe*-ua5>SCkKG_doIttDqVD+(|XbhuIM4g9*HFo
zM3co11%PV*YC#Nap&pqcyPAbT04{AA-9fhFvd)SQ`Yi*s=9j2cvIYRIEu#@$J5S)`
zWQY5}s!}AkuI~hb?Y<7)ur=a==qj10CAs^gz(^JL{pz)it17qkEdQRx8ycXiSXY2n
z(8u8a7~1=!hI8G0IZCx=r7q=K7D(lp&~B^~eZU@WiXiEpP?4in*8%9|_umop00Q)K
zZ)dF-Dq!r(y$MVxc_$ybNlpALKiEr^&Y#uPn$sjH;{T~KA9-;*YGmmh7A}P%P5(0N
zh$ha3z8rs-b1LfUBX|M~Juk0~_gqvxWO9LXKWsfl&=pu~9)TsahJ*;98;TVOeg-1a
zgq`txjHipb8R&a)e_fhCs?xmCl{}d*ipU~*+|A>hd0sA&ZJH>A1{a1-{WR5NB1V1H
zlL{ErHt6AxJs-l@T&z5|A0=w`yB@;fE`1LGQ`(h}dl}cJA_WFaGdybYcFdp`O#J9E
zmu+HFq+-i9lGg-S(#8a(T_vxmT_GNuGh1ky&?ou_5?%mOrB>_BIzX%+7tZEhHgN)a
zU~0a?9AgBGI<u;55k&4#3^$HT$@2~jZQ@G_&rP|^lW6({P4SJ`9%;C4jFf5T<)`Cc
z=BPz&18FgwdP&W<`(YU^Cus)->{(sWj7XeN_R64s8Al^dn{Sc8=~d21mWw;p3Lmj5
zzd1HIO|XxZ3sGAA6k73ocz;kVr}W}1wE3muI)FxTUSC5z6!8PZ?_+YTt$8*ocI8Ll
zIRhr%r;SZ;w0Lf9{DNWf9A;shV^l8q;&T)P%nRC4$Jo7iZ}MANyu(jxIlvu4biX&q
z8{7gdp};L5A(VOxo(j_hZCSP#z}}|z2LUc*)%e(EogDkv6ub-2BWt1Mxo4qI`R(IE
zmg(pB|Dol!{e`K!i+au*&iB&69Z8Ov^TbEb@yq?oJ2^bq3=UVHaR?YPF4~y`s(y*S
zuF9|KBfrnR-wx9heS6O#_vu%6E|$wy2$sNqWCQ|7%@+H$oF9>nmk<;w-hallW^UVM
zX-0`Rdg?wmFppVyJ2z?c;Q4Ryuf?4{f*}~CY=a5@mg;g`C9bt8JFy#}O<YwNNTzWp
zjxeFsYJKTeiKu%77v$XByn(WV5UJkCp8MsAB+}w|h3ub>L({mN(frD>7lZm7_`+B|
zxPE+<6EDb#%Lf%Ub$FtONw0XQ9?=<$RLs(j+DDqa(oU5+f1ZS?;)4WB^`e|x&W6qG
zhAxT)wA-l&8TQpxDP%BGk_^c$vXJM24zRpxi57w?O`g8O7TzuYqo&NmtwPp~-|#B}
zulwpX5^L!TY|v=)G{yt%)x)2*#sBYEdm)jjNk?mB7<A`a!g>nuBi^^593$8fJ=B}b
zKU=Z0F&Lz$zws>hf{=V>ZUrzqD3bLa3K2~DXocPVjr)yuRb=`p=~J|=ko6sdueC|r
zB|s?N*H?GLuxixjf5`EDNn%=e-|sgtU(VOweI}}cxTPH#$v`s~b5~4{e?u?0^zUrp
zV!qb8quQxIIf9PVU=}CxyWd=HgaR{+($$qxjnrl&mcQ7s0>wM;)^qZfz!jBs#UQd_
zTq_1S&%hHF1GpPekM9kp$?i$3$olim5pmvnKv`HZHoi*pz3V}Fd;^|PniZLPX|CIa
z$z<isVn09r8+IQg3%0sbZ!CmM&aT$Be%Iw>`$rka=Ixt<=m)(iJ(hSilMh%`yUSx)
z$+;rqay(bR1p_JY^&SL57X4v0xl+6{N^>3A{Rizk#G!6~cr5LiWNdm9bW+&@Z|FB#
zUJ%ifmvw)xcV{Htdr>Z=nU2DN;c_JfqD4Cdl7u<x6GI;8V`AE5XZ<kQ@QE?w<%_(C
zV_PvBc`*u9zYfnS>C}28kLmttb(u#=ja9b;RRJ{2v;g|If==-Rs`cz{KRZK6-(;WU
zt1F2)@pu4vgg;Y7)FzZ1$~PmqMGf|o>TLO?DEplKJDMls0BVxQ>SWLh*vzVzgWb6g
z`kkMTE<hB~M$dnLfUPLsmA*#I2a3f02D9I(QaI9a(Gh*^IkN7udqEhAUH)(6tm0h+
z4>L3hQsEO9T@Q8L%))%jHSk7u^T{e^5)wRLi+E&2SE3XrR)>>pk*=t;w_+Eh$7@$_
zx*xdifCr%lCF7a3<w}*pYJ%thQzgW=!d48^WOL1z{&=u&*<y!tV~c2&ZH8(snvO(Z
z+7KC2wS@SMPNCPF1&a1}iJoy&6B%gd?2{97B_P7)oWt-~Oq6A^B|U%>vLqIp{Ls@;
zYD!97y0y}@UPBh!c<f<XAskzbiAn*Vx?13|D`O1TmV{{~+aLxrDoXZEqwOKJD_*jr
z^P7(lg6&?8h#5~xOmItZX+SpBzvM>|kt-P`etfCOj78y^F;<1bBfm<*3ktqms^!Wy
z*!{gp5~y?8ErE^AjAS<H4CbFZI*{nW?Id9$;c6IvQpT0Mc-MFA*N5)7NUr`-`>=OP
z?<)*xA~BTiEWT{u@M0(Uz$n_z9yYG-Gp2{|p#2$tcQqugz$Et>P7ZL#{-8KQ%7Fg?
zRp)3?BM-5^3PJrI8S&>g#=bqzkQL;QRwM*)*zuT)w{b!RB(RMQF?OS&LCH#Ii$~s~
zj=Ho~jVEB(%G>8=^juyl54$jvZoMVa<w%k@5A}fYJc+Po8+m+_<Vk}bsooC!X32Zj
zpLGSx%d$9xzj=H?7$6-sO+Io(Jb2aD-7ZSPU$ryu91)$Nz~AjtMWnu`x$pi0hB?ga
z^VxSJs>q4px|_M-7YT8>f)_12rShXS?Q3qBn2sZ_QbHQ|KwG<|!2x5CsMu|%6f&^2
zcyG*+5QE%nuvr2ggk{7`+Gb!B_?L9&O*dYY(4k%DWC=T=@Z&5$BRYeN5>MBfvxTf%
zxdmz>+6qmoY4&p8VwH*7jV7Kp{JKG$Z_0Qy>KF&}?m5;}=iterA8ceoQ)H$O>*or5
z%lpw==MuQ9b`r@Cn7Qk}v+06oyy5RU%0^m^j#cbVU1#}KQJ#0CZ%ux&_v@8OeDo*q
zJFny!{s$(f;uF?YzZPeuzVo*`avVvS!Ua$h^ITzC)giHxpU>CzSnj&q1u(1?Dr9fu
zuPryD(EB|Nv1CS=@^hQNilVUkit60z)8ERW-o@N)*sQ9H_(q8LY15jt@@?dwJJ#&m
z{;>p0g+-p3kxv5EU5FG*pt<)dAK8r>GbeH>zaG$%e?^H;oY){O{{1Kj7b6XEd8g-F
zRc8&o7?WXhL$TGO2j08MZP*F&;KKRJ*0{KPGFGe?I=9rKn9@rQr6$MiteJd?`nlw>
z>W}o(t>wtwOx6+n$>EW<zg7gj5%OdvH@<I3f{HHh&%!T{N%*k8((O`pbKJep8CV%*
zaXtJ)DYvG>x%cA{(dx}-i3ct@1O5FUCuhuBijbx-LMLlH2T6m;U~fELEZZ>DHEgnY
z_pD`wKJ8=rJrT^|p3{L3&L0gHy%MOkw#Z$vb8Y>C;mpauqYtD<STj|_^T`Bxq7aFL
z^!??23Pv5Y2@bF06eRXXrr7bh;_D+}SoAMk&dGu4od)?pgDd_Ca}bf;-3Rp4_vcO5
zyo%X-n`io79#gM$hDbL*(IesxS~2|!Wxx-cY9ztY@E9~V^hOt|*S<V97&s}0+)<uQ
zKCw#27s6X~YbTZ~@WGfZ5=>h9vNsC>FMmEm5#AG7dYU<6Q_bzS9F(iR_mFX>?|eF>
zc{VCGo*+1|zBruZD#SR*Q0*lU5lp_L#uZ4hnm|Ldc~SlL<V&!$@?l7hqL?oq-iPu!
zfcp+0Z-W&|MP)w!=j3E7{|~p$2a01i+$ym5iJYt7KF#wk^;OYX?6JR2kH+Jw^AE!A
zxKb$KU;klcLN2!9_l+owr5oQ$G(Bq*%il+oK%0g3IeEsWg+J^i^n;GTK+}he$7fLm
z1(c3H4VVgX&6KD?bjZC#xG-hW{{uFsoe2$h5LAjOpc`n5wWk8eEG22KJ0A~UQJ2L0
z7!Q2_r?nd#=+xFU^UsIkWcfUEeK#aaG3kjVKy-?oVf;2|=bD&kKvL&q;=01}Vdy!o
z3{y8QztKBmI(%hVloPk*Thhf=4SIt8A3@#(`yQUCXxDP!izlk5*)c8;!a5B-764LF
zHN4`b<)l|c#s*)*oXwcb0)?L$`2~k-++E?YKtmh0?jk7e2M7lx))7J71^0V)lBIK3
z5u&aS#cYJO$4?PQ4~e&x@zg!3NIf~Lh2NUc$!YJNNNIWV(I|CmlAGZUfCxxQnInE_
z+m{BBb5zu5#cbxN{c+kb>Bb<co1ejD7ZT+mC4_Ob0HloVLQFbr9%!#??b`UBz0#wh
z^qgAm#Vhf&>|0!y%vay-R<Tg4j*KKTxKy9TcWPVo#yr1i;K8b@N4;t~bM}rfl5(h}
zGFH?BE_yhRI`y7i3sJf45&0if4pXvUO0@*+eY!|RpK#u=!3UU4aKo>*>&f!?E4%BB
zBre{f_#Rg4_XbC7y&JpV)sIL@Ba0xi{lH(rA7Y$65`{;p=Zz<SrE0Rm99c!S#9zLz
zI!R7jJHnT5BQ6$qzC-$Cn9QU&gxb&6or?BoC?G!EDk!5F&nUa{i-%6`UAqT0QG#!C
zyhOVy?vei0<{CHcFJ$Lh&nyf7Qe7FvHe+IYW*08}oG>p^R=<YrkV;#Zj?x4b^(b2v
zo|Gu;QhD}#>*|ABa97MEA=+10Q(=(0@yHHNMb_fRh`BzfneP6kX0BeD-a{9;rVhi6
zSGVPBnzx`nb<(ey$SYO9yQ_xSJ5r;->Es^O7v>%dPaEu-lz3+h?V~!%8zss}m!QS0
z%oH0ZncUMtP^?oH7fln3F%{X3x7BqG{J3g7Kfxo0EW>!S*+k~It_(0Qqc7b#O%Zam
zjT2wLH!$V*7c1)LyVY+8-bOHE@+kW@PK*iM!o?!T^-E>HFd!2CjBZf0AlOJ)t+;1t
zkfsRJqIHpOdi>Pu@goz|PYA9MAGRQjYgbyFV8VG-qM*sZ$l(;6QMovRR50SGV3S*_
z1G|M=4k5!v!eDQ22at>?u=PRJZc(*0=jY)4$*^$vmSmY9BcBQ(zm_Irz$d76Pe9m0
zHCJqzvCex%%N*RfVo$DAGmx<~Pq^&E&+GfKUy+j=AM5U)h%Oja3;sCR2^;x9HJ7Sy
z*TaZ>(7bX>o;l)j+#(P_-hMP}=KN(qIP@)!Diac>o2>)8H#|h-cU!}0s*}265@`3e
zAXpA2xVWh3%C)Fz6$<IJ*!XWy54(%~{r(e?(UFS%kBq{8ll+VIA?KSxhLdZlgoQ-p
zb6V9hDe^OaG83HG^eVbr6D`PBw6qk2Ub%(d<XbW7O>+{=MztV`&G@*Ee|lh<XO^@D
z;hR&f2)bmVzyd*%af*!*T4l8wv&Rdgv?Hu%B7i)vgQqB8X-uRWR(YGrx<%T41SZL$
zv&RQ^QP{AE51UW%!J)<Uj7yxJ4}lTIs39hCX8O&?{Seodoe36Q7V93n^lgR0yzMEi
z2_AP2r4$a9aE(;<?&e0hnQw!S&Tv_l^=G{el_&1+%zRWrO}X7&3Ah)3K1_FAP$*3{
z6mNMjfEhI=-*q*AE*d8{tn9HeKexO@AtIqqB#=);jbog+bA3cG@Rk$x(yRES{RIn;
zk7druW{M$8o(>CcCr<3kEorjY*YUPQPYy3n<vJ;poRJGR-E`S-3@OIWmkv~&UQEml
z_sCb=y(s7U&(fU}awEQR_QpT^V)nExSYws=@??b;L5w<$BJXn!wpb#679VUq{O*)M
zR^!%V_=)S|Q)zJobM9;CpqwXFz)?mNSm_zlX=*EZwx)5@)v0Pek>a+Nv{6|Y><Sw&
zwBtSa?w@K@FHyL$2;TL;c}0#7yBLOF2nu>$JCE^oi&@=|*-i^;VKwRdg-%IrTTWMe
z{+2OEyy=!TpkSdc--C<K+w*H8c=OfaVGD}brHf1ZaRDWQ&#db(Ueb%W4T@58z-u=n
z^~>ffAAs1T-N?Yfd38}04dq%PI#67aPVwrD`0r8ztE{q^cOM$-nTFf=e~Fy0Vvw@7
z`!o}ymMd$%dUB$YDmy%Eg{!n60+UQ!gaScSwyCc@pVy$Frfm8q@YY%~b)tW%0;*l#
z!{o>_h608-3=^^z2DdzxPZ>md83`c*<TK)6$zUW7*wa`%26NCg@%Ir4CF)m?-&V8+
zU`IIm+;w$KzewqIn&bP`)HR-FKp;@6nn`$MIVElX3f0X@0%-vIQ_j6B?s$psPyC(-
ze4PzZ4NvTTwDF5ofH|V;cK<70qN2XbF-EBG4@qXC?*iq;NGHw-)$Iai%=qYED=!Ri
zTX`{idT~2ngfU9a1hkMrVcgvPA3nHIz1<Fsw15P8|DEW>poH^twt953ZHDkj)<a3S
zBWVWC`_o0VcQ^>$d*x9_wf5JYyj?n?Tic9W8Vp{4E&MQ6Ufz_u*@inkrc~55E3&ZL
z{@Z|XY)JUwd2!%OM?3SBsLgo-+1G3wKqDvVC7|S3dZbu}e!Cy>%#qufw#|@3im0QW
z--b~?@Uey@h6YL}20W8z{u_)7W0u`XAEu2_h2vZ!BkDMumXW{m&+{<8@EDji(PMpL
zCV{~lb@;cu1w8wVt7C8$dKf}zOuVemL1b=A$S~-0t^|{ITKVdn$NAhuNGQp^V$MOX
z>Qn<fl?S%wPdnMr8LU((u*FaKTbxNV&4J_nz+1Itj?o@$+Z_sK!;3FG(Ywa2b52cL
zL|c>FUvfW`hSRmvdcMKohij||nJ>QN;GX7xV#cph6!IgAFO8frm3(nV?{VZ~I8HB|
zY4o%~DR;Zx`S`Qp^_S(+nI7X168feXF#H9${`1miIQ#S8%*7odUMyuq?!UR=57}rp
zm~sl`l^5=>)+NcwCpqvgu`rZ(;NhM|zD}X<G-h6n8h;5N99k~v3=ZUx`=yXVCriRi
zQZ)Vjp2A>H?<eN@%2aZr>G^+0el=&H$?~&gslDMlI|Q+Rakf<yIVR#)6;v{0^rXm!
z_a82bxcLbV>`(yKQq9rR>&~oa`YtSMVtZN@Dc7!`uMox@!a3Q~-a>*^5FZ`2`Q9UY
zT<+!z56fjH5pAc|GfE_P>^%>bq$pcS0!JR=gYH13?Jljsh1Vx46k6XyM;VchV549s
z8Nbb@kwFA&6MVuiXzYW;BV4}sX3bv|uAV`dL`)<L7d|mZWw$WmurmAe>?zZc&91nn
z={Uwsh6EbFERMnWK|&vQcIxl&-9zw*W~u#0!8SHbQ(z%7H{)q;w1#CRp1etxW7e|2
zppsY_n=^Z$;>3kRG$A+X_M{JzEyjpYH<Z0^nO-D8{SUmR`K)Y{*u9nb$tF{~Ywk9+
zmh-A{o1D#aQet|Ov7C~f?l|(_+C~(el03w#{*xR2EIDh6DZpaU+(=Xj>FLoBwlX4z
zi3=sojx?U6lDnokn>u!f6ICc(?7WVs(yjU+8l?y8-c~;TJsC_ZP0cJzCql&i@VF@E
z-rKkdowXq~CcByrV>Kr8G{{?ZnoTHe$|ePOztHy6({DU^f9<~3LB9#cQ^vZc|9A!;
zLXE1drZqn>vh-AytI)*jQQcuPnH><%h;VJ=_XK-JYmbnMUs2-Jf)Y$&)0VU)fvfF1
z$xN3~_ZXSA(=Dz@JD+6G4=guHI@-7o+%j<>yxH{}Bn=%VbKCujqz1B?;~xt_P#mjC
zv@!oaksKd%bupnZkfVs;-w`&drBySe{|P~6A$N>)lep|E|5*>HaB1Wz(}u3PdGej!
zoqng~pY5T1#6L0o6=AelmN{agEmqCyztFbs-5h`KBQt(bn|9ttnCfN%(zDVFPy6|0
zLW$2)$0qhwd&&1ZlQ@_BDOBc~PyP(?#@$6qSDKC7vz?{3>xN0uVB`#V?AybcY#l9M
z-^P0iS^ad*nBc}eJ1aW(@UM!lQZ&G4^3(|l&uGqJZx}ja;!Wool<o87+@l8_*@IAs
zvbNwy$Tv9o>-ZC@cinUt)h+E#m0Ugufhc?}0t!9rG`_e!4eWc&%UM%B%bICBZD;Ca
z#xttSp$zX=S`MdTk<r>r?LrYTcTEMw{5>IE5ICve4@*@(<E0qzXHpJd2Z&Z4uHOtk
z7r<<g>#l$Y$yTG_&2ftQLpFId!%)-ta(0KDvbPoNSJs>pZ+BSE)EuJO{ndru&DB;!
z_AKkB9h{yNOG50!`pOkusWl(tI&s3xg{GGN^wjiOMS@WgN_?iWUTES|7_oJID&W{&
z8d<_*O&R@sz5T1@)T{dqlyjWY0&%GE@Y}TRer7>sT0M-ksNdSP6@j{7-~RsAOnY70
zpwyR}!hDPOjk3v5L%)ZCb_Y!`iuxC*4?mWq4ELa<G*CyM(>s5Qz&V|cKBt42jhI6z
z9REIxAiQ4tAoSj=o(3s#z!_LL@X~Wjo$2kg;qJ$<iNu)?qWK&_6XwkQVy~`D9S^2N
zqzv#dhrNm13}kS$I}pcYr8KRSdYlKx#RQ!JtCThzhP&!E3dOv}jE|m@*TKDs_gY9w
z>7spHcN$vstEatpZk~8C-0%=wfBECz+v?SQW#V#gVe2&Y09iFgEk#h&EHPEyAS?K)
z!7z&c>C@h0S*qgut$Q`MrJSb6|7u)@VI$5glrT3nkMYv~SyeF+2RRl@<lO{$KEwZT
zwVtu|p1e1G$<4%Pr`?azLL>B~OCa5D#kp@@xk)xfBGY5_m$Ss9%J7Dh{Imn!2d|G+
z#+id#<Yv}w*Cz`_@-ieu&9JsI2IGfo-lnUYKn(fiFEgJpqhfLGB;4i)-2M4ih4Y0E
zJv+CHi$q_KowE>aW6qHIVyyML)S@Wn^RGQJf~x!-Q%<Z`i`YEobT%lRjb&*H^>dkb
zuL6$4uLH7*IVnXC23A~2I+q4gx@){@p>o3{IOW8Djv-}1H%ZnNcSsx*`CwED?OR(2
z(ZnmkZN~>FCkHikog;FlSeaW9<C%n~UGJE#?O#WH@C+6zmRo9ilg@v>b#KQ;te3vC
z1kTjvaEbW3g_BWfx@_am|By&9;TX|?!wIo%ToPj4$nDsnZB8>{mrHH`T1puq`n3zo
z<&{;as)N{8=4?TXrSj?y%bFH5mMBUfVjd0|e-7nr$Sl6Ox>~AG5H*}Sc4QEBO2Kg`
z>Uvn8?IqtbWb|a(YI|F8<i`6vj25S%cQ01z@9zJs7vOk^?SFJ(oC$K_T>X0eO&c#j
zL%QCdP?B`5um1BuE*&JTSRwkg>p*gd*~c4b8}*fZ+e%8_=BQ-+08Uf$o`@+m*BQJ0
zC+R%#kzJp)^!ypT#n0tI=ch&a6hGW)IL5)0^Wc4MV&6j7LBqb*%2vnat`ok6?e$0}
z{x98MWe6j^S4)B|H(IgF7Gk>qOK_T7&}n*E%ksIcd?6GSbq~4H{Sw8C{L}r!Btp}<
zae1QWiz=rsH`C1*P8tmRtsjp#rA6m%5`FU`b9%Vts84Ay@_S49)7p-fT^?P2{JZKG
zrH?P{$`i5qs6eVQvbY15=Xp^l8!9J${O97H@>XAvspB>)1|{}h3}jNf2BT?c!%K3P
zb+WTIp_TAHLdd`F`&xLqgmpK*+-{90b9jSu%678n=HCVfjk!K`G52fF*%|ZwKVl09
zIj0Fb#59N-yF{eUN`jsVUEZ^gfoju_U<rX*tRkkSAw@S@Tlh{xkOFao$y%IJd0>+K
zA94Tx(fGMQ?92y1wG^1sCw)78VQ;<yDjUrYKc||DasxbX?;igi9~f5y#B3T08OWO!
z4Wr~M4VM>d7_68j2OG>(>C6d*VlJ!y#$`7FFfJcG%55jW6J3=P?jC)JJ2|c^oKyP@
zoAs*H+3{&Rr3WlMD`56a8niNhB&gru<YMi=LJ_b^iz5@3(@G%+baBo;+%dt9(ySR`
zL6lLj_NVE3zU}Q!#mm2&?Q9<olv^DFKJ;(W!uc*f=15R&ErV^zHdB~b(W=nsPP#l-
z5q{@!SD$~CvsqE98N3i$Z;Xjas2!pq(B?fD&~I=BpYmYnH{AK6Y6pMnyO~QyM_UHo
z=~-8?hNJKmmp@V1jjAYP`?tR*dv}4w9bPhw!hD%d7U|1&g3i{??G*K0;8>B^kp{*h
zmsTiVON$?`_6%5ZDkk#VX4gDewTB1j1$#%K2EQ(UsB^NV41I*Ge)*XrEh)sl6uC6=
zL0UXg8d+hUK__<=Bejr6KS(2m9)XPtVrvpR1T6VDX}-=pCqeo0_r<TuVuyKGWQ0Sw
z=nUv5cNu{l<Ny7}2z)An?xf>$6|Jab|0U-q<*pFEg<?9#w5>3b1OO>h9;O|w34edR
zpzXw)Ow5%;{x8qMg|ZJX7lF&QEy>26eeo`yw(l1#ip)okxuUIX<G<a^>1+(Hv75Fo
zKStp$urB-v$1#NJr5ze*K2Dqkc;eG)eK=Hv(39r?R6>Q?Y#msXHPxVXcAR9qpf1Ri
zUh}Z3$h@72#oy9kZ^%bynzz`il2fH`nlL1r3-Qqui{}5`yss){h&HYC<5fqi&Hnu=
z!$C*eV@Wsjcxq?s7!XoQPgkU%$TzT=NJ1CQnuKS@SITjrMZxQ%$Y?31!(_i&v2Ip%
zjIyyx>Y8sDKhn_{VEk^0`78S|`#HUXIJ`tveHAD5tbRXp4)!0-bN^1M97jnohEysa
zIIWqQXumtP?&T(^!tWNvHf=1{yxWLz{2-{n1Xk?{VW!b3^s1b*o?=FNK4?3~p|s>$
z6?$6MXlL7sfm@}Uzwhws2Q%A}`Yym(4+#d6HGO6dsFA-y<N%eBMx{ra{pp8eW9>!B
z8iIX!Y^9->0uCP&*O@lZ%bFTu43UQ}5wDSt1ZFapz&@E}9js{t1CAS76ttC~Zf*Ye
zrfrBsYyvNrsfLBmvcYY8-H77F-bR1_)Pha}|JS#VA!VRM-%9%hm{&)}5XA-hz$M`L
zOH;$=_zDd{9=QlR{rg;?rjx>=Y8A`x810sb2}j45wn%x5sDBQnd&L9~>9xhTRD_84
z-?Wz0&c7Gu_-zz<_rGh*RmS_CW6evV+jriTV2N3)Yoz_F=n(iTs?*W1*HGXDDn6a*
z>O0%8XLCMo1>#ld44$O;rZkm-wp<N1diIE`3t$$4(y}k^h3(lLSqJt%3nk(-iV6d!
z3hGm~&Z|ZcA0H@UrYH6O2c&JmC%7E-^3tMd?itQ}Q^q8W>~;~FnOx$+%pyuS@Z=W?
zBTv5j-J3_dX<8Mn>YAR;x6mptd64S3-1?Ki2M@j;1TGwJSUyusbvE&5XsfI3rBCq+
zbg%}-j#t<G%R%(9i^~VL4EhjfK)0Y|RMi3g;H5qGjly?Pvc29NI#Fi=k!t}pDLYDo
z%hVxz2|JP`Q>CG0Gj7f?fJA9j<!SK8I}B^+4AiV~&<b(@D+${d&z;dXPS<=u0wGK~
z%;oG|0sV*+!(V(?@s4oH5}`oA_ZSYOf)&t>fPi<A^)o6_aVh|sjUSKWF?^D~#&U8P
ziIhz<+$Ga9y&9~HDk=*iqkFi(z-d5<n=SxAnlCbPSxYzS%fnh``VL|RdnI*_9Rca&
zDIvO2jbm5@DC;u_lF3ZJ$!|`PB4^~wk4bcBWzck=Pkr?uoISmLiPmZwoIs{oRk=Xj
zB>BBqYq+oqnI^EZU>;OwHA_GxUQ7W<KK^mC5CD$j7wsI9%4X0KKrmp$fp$37-B8y&
zYi$!{wKgec=~vTb0>rfH$cCV%cu)V^y-0Nv1%buOXm>HN*DjG+h{g4n6{eKLls*{Q
zplxQG%gDB{+x)9m3@jki&`6iK(pArvt~P|9hv0QZam(MPx`Q=apsU=<(FHc2Bz~t-
z?W0#`va9NFme12C{D1_$3Y(Lf*Vl*%TITHYq57%^jz$fSg!kqud|GvkRsDN>|8qF}
zx32*YA+!TY8ZSs?G7aeC@GzJw^R8fDZd6Q(WUA7vw9V8&g|9ccRH_mMf0CI+Pawf&
zno_&~cYz=`jnHa)+oVLRvEgoP^_qhvEb6D&A?ON|x_ZomAy6Y;HA{gRAWn+yFTU1}
zucxiEyw@L0Cc;c2SF%%i8cSkB647i<ZrnD4?v^Y$3b*{R<*XrZvt$HkYMcW!j3qSp
zy=Di6gc%bs&j(XA*p3%#!E5YyFIbQ5fb&~$YFXukgXGj2bb~o=S|%yk^8_EXtJVTc
z3dkl`;&os}kql5GMPL~GN02q+wtvCQBDH{F!k<LtRGC*wa-afd+L#6A0X@>!ITn&S
zc9MwCJQcw|ahT_Et1XQyk~+J{)w&&%6VTTqJm?vmsbUU5(U+ViKj3=EHlFiqd9V66
zW|DiSKoVX+d7c~?UxCE}Z7q;%wmpt$Cb^-DOf%O-*MlElrwdQrV=XtL`Hcd1t`oYh
zFAc&?wI-47Z@eD$92RMo6s33Mo_o1PZcap$Yv@B2<z$RYpdhy5(dj0d>NIU`-5bjq
zah%ESFTCupmV0?R(B8#VN!>NH%4ejOB<|ErBIwH_u8hY$c<-qDo{DUrGw4Pf<}uu*
z2n5GI8_6b@ae^yt%tq0G@lu+$fzRC+|CR+tdo<kOc;EK{=caM|mY?XF)Gi@D&-qxT
zKBp8P{7X+>Eyy{w|HzC{aAC%mn8}75c){gkttOn)53<SHjzk*}0F&<|>HEno9s2t8
zF%pA--1PpEmKDJ!CQSBIa&XGERn?Ge$h{4NM>{`s-W@l6hvh|t1d1}}(sf`c87qEa
zZhu|!v&ud;1Ja-d9~7vvUk<mFr<9)_vZ&;ea};wPvOW}b^*Y;~f(IR118kSNHm!-y
z<F`AwN5kLolhjXtD&ty3L1t2;{;8P`lLpsz6&WMkshi3Ht-GKx^C02<C)R&E@!Hmf
zwmq1dan<NKbq=6cLmpT?Iu76M(U^q?6~co&rJBQ2>B`)t<ip@Ww(lnbgcFb2q=-aT
zjjwF1<bp{(m_zRbRB`8KxJscSu`Z>_hGj$KB$yef_Y`S=8aMzRO+7fIdS9mrs-KDu
z#BJR?(v?bn>-aIXjHyL+r>peW{x%DVB)ErZjsx5g^t6G?#M47W_pvI8&lBofK+l6Y
zHC{&bja^WHy#PYjEvyOk>RF^(Qui3?NG-7UU8S)z4I%euIE~}Z<nVqM9dx*5#ys;^
zU-%0^>TuGRTRKuU1ydsd-uANCFIfM-83%1Z<-%rz7%YB0dd6@kAkt9WK>Xu}UAo2a
zpEMX#*-9pcz;kz+u*=^G0V9slUF#`o61;a_21<{9nhZn8iSfimKkx(vZ#A@HH#tka
zmDInW!p`3*YtiemUOK&?`bI#qqvSLY+`3BUkB>(bWqaY7-%9>Kn3n2qs=$cLTZlgs
z5Fz@wlLit?*tirt`kvLYxZq-vOB2YyRiX{8m1cZ23(!=^2Iq~>5C*%ifbT6nLeq(R
z^v&k#Gw6pAI+9=YV%svizBsb-XU5Zm+&AfSVXO#FkYeNw*|5{Cl1)b&I7lylrnn0O
z7<vB&rGWS31nb2y+B3QVyt~b$u)yO?Dq7ZWV)FSgfSPG+npOV23{u*q@cw`W5}Rmi
zfW@)U^_8YgC)GZ~*v8Gp4f?5`6>O>5F}CBlE#xRI%qegL-z3kG%w8d)%dvzNyJ9Y@
zJt5|*_RQ_YmYzjZ2TH2P-@orl5)P!j+e2JQNV2*7TcyX1drP46{zD97=qg3_V^<8!
zN5;(C_Z=yx9v%2vl^r*Mf{`qBuW4&pF(B9g-)1$j3%l!>W^9)8cMZ(cTb*laWFaIq
z$_u*enogp48_v}A-7%Jj#Kyj)hRBz1ZxLYXilsLJcuj4M>X@4c!{r52UEo$8T)a4y
z*DBeM_SIZxyq}Wz{p@)XO?d>t!iF#y&u~{CIS?@zJKWupjEB*UFb&fLO|-&tdFQPt
zE|?Zh7+CS_mRYOvur`e{Nur*2+s@bAHOH@HL~7U6su>z+QEa+FaC1F_x=@6Z`o@%Q
zueIlDy9dDMT#DrMj6lNXKgT8B0|_)p)0U!m<r*IGAj#~ugO@6}8yeHe`<-GPaz}}d
zx_8r^m+xg4KG<~JHW$M*eu|VC-LaOO9Y2XUO_+)9%Q~1g+TGLq<dWMfU0Rc<Gvahv
z6IB!1!HD=>KvKC@ai&0<OaBKOnTU9{`3o4WU1eH>yqoAg4*a*ZKpXe|@5kEz|95}9
z@5G!dkxtxYI_qmZJtQQVO$+iD##Q+qf&Ke=eH$o5-aM@7T>X+%o8EB1C4kGMq!Yv0
zP325#oZXy)@WYkH#3Q`XEni8kJBpB^?H6p*)TQX#|7v@61>vB5dhzB=Nvl4}>&BVp
zA!F#6d|Aee_K`TExm7Hx-(TOHi9j`{qF;y$<_HUaU+s$-Y+#M4bpF*e{s@2gRq`K5
z0xn6<>c7AS*YymAXe+2(1ic`ACynou+%z^i+J%2AVsDLok#Yi2gq$nQ>8hHv&Hsn7
z^Ny#2|NnkA5eHeB2N~JQ-iM5`C5dc_jO@L4#1XP7BT?2#M`p&sLD?BuImjMIws3#W
zb$#c3-@nIw|8+es*LB5pe8&6pe!pJNSB&>tN`W;zcm1(6)@`hYLfLbIX}{8(OD&@r
z4VUOq;|VRWUvLHX-j;>+oY1!)@|RJ=p`|@3e}031QpnQURu{bTq!%u);{5xNPs1xz
zirgojJnNDi2G3+x<86a~LDTX(cJ$4~`@_mIBL{HG#r&YIeI(g0tXz$@zq~)CJT#-r
zLp>{;$tBUv4C4oNMyL{70GG^8d!^=?J4^F<l6}$InY0qtb8#|*VEJb<i+*X(&D}dp
z)UvDRJPkbBde8eQ7PMsCKxq}X=X=@x^vCM4#kOw_C8i{|%-8*Oh>?<9w8g1b9_D>x
z!Eoto^8lC7_S(fkU+bXe{L)t%<=Uu#kJ_bPBlM?1MZ9R*ZH~|tYcB!Y7uD<*2dwO|
zU<#s33N(<kFrT^qD*ylUDfss>`B#@#ln}uc)-4`KfYY(jka)b@MDUJr-XaW8wi>i=
zWahM@n*5}31-u>-xt;;p{W|iMpUl@)72m`Tr%5TotKll!Mi0c8z@z>r)Jm<jO@Q`p
z%KuSk{qGABELPDq>Q-6~Hr)yFR2P3CAUq`sSIKbU!cU^t`WVEB-;vykD@9$7z(uzD
zRy*tE_sFfC<$gmeQVJ#f$k|fHkSIYS@W(*7z&B;2K6UwjY=M7l4^wI@Ghz9QGvG&X
z>BPaeWJw7HDnA$EvYw|woc8MMQI`~QYbqWpN&{i!M+Bpxv7bTHfv?y(^0=K`<a7G{
zhdn|dY-kJow>|<Ns@pZgriPz@Z{CsE3SOaGxJbk?pKs3fXTaR?PS_NTt7ia#YJ*bx
zXPC8aN$@O$F^e=YO|cWLHg4zwxPOF+q|!dUKRlfNfBpw3$nkF*tQ>p+NksS7PtCc$
zf``x+PRy8d%kd>Hf%M_wCJMqxMw)navIVg8W`MT)>Tvaz!7`6-?g~8s_M}{@D;WaI
zB8DlP{cJ^J3lP4%FV40@C9Yc(>_i>}fh))tMBMd=mJTbrktS*rs^2NFu`7I9CP~;u
ziLXw`B4_U+Bk{lWg8s3N{`)U6SHhKmwPOo_vjSHSEwRr{K94tVd6yh+6ygx<iG-7H
zw^I70m8aVTIJ^h{uFdlxM*ya{5OQ%kS0|6(HGoSh_5sRr%b-b|lu$W?G%-r?c7|z9
zH~3ywnc$WCQxwmw;Q#CA`R||ZzyHpI1FRwzzK1OTK1D^HsQrLomUW7?2}q2+L2<S5
z?pvw-*T1)v)ea<6F}R_!dJcpbl8Cs!;{2~p%YSXJ|NK^re7N7Yey4T3*uOr*@Oy+f
zbA>OE)Kg4m1466tGaLqSU?>!zrH7A_0I<y$0O=0?v^49#Xa!aNuS#HN7$d;>BW8d<
zX7t2`H%6HbA0z;02LiFyl&c#E0CL4u5aS!MP|;!P2S72;Vz%`MNQ#+OO|xuQL73fw
znIFyuy|q%5oNqgaqyjq__ageg?9~7GVn=)!HC`-j@G0`&|Eisa6{uAyhi5@9vdN26
z;DD(=ou>+2!TFOKaiKNK+bYu8V13E~v&1y$5>EX-?Ko}+WW|NY-?Eo@v@=C>d*1yy
z0Qkp+e)-?phoEV|UzUI{16UV<f;#4yFbylvFT(|(BQI6~?I^@|r;Vi1NLQxB`<pH?
zdpOWh56*jky3sU$#qAkxGc!xM=7JaRaXB0F_i<>oD&e2QFBdc3iX@K?eRERY9=~o5
z0p5UV@Wo)@$$ek%IP%u}R4cG@fV7&=7C4|h0TIrx;EUbR|MPMcm81qsa6-_x@6$g&
z5U_5JfMH@OWdFToJx-mU-Hpp|046W@Es@aUyIh<<)A)_1QSr3<+q6JhAJ_h3PZ1Py
zu3d>t1X){O&Nkh)q`n-#Xau$<R>w1tc0W&V=3Eo5{r<XBHRNb!J7oF%&$3b-%_q>Q
z`3%^n8u29+2|nUGq(K|D>_#-a-t56Uuj-mg(jR+CH~gWi@{2YDsaWZW%>W5lyOo?y
z8wbY`X=iZQJMUNj{7jC3QRRIC9)RIUNdeewT}n*j8B!sGJsg)ffh+d5Kt{E@-?LBu
zu^RrT1p{Zt<HG;dJ!*67-)kHX_7h~xBO;xL(@2kEmHw#o@ZsW$XK-dUyk67gV&w=N
znRglhU1yKyt%ES2UFZp*!LeYy)ljIPl+J;JxdC{9`r^X4K5$@(FXKG}C8QUiOexI(
zD&+TG6@#m|v!<~TckWF(6jZR`;mCT8IF#c4*9X$az~%J;bx_y5>GH_Iw|M1Tm+8(o
zT%hyj0~SaC)#jc~<D3#9&Z{J|$hdm}V)_QZP?>)HpZ+KR^*JP3<F<4R`5eOapSuSL
ziwxt#xR{FW(RAkaHd5m)UmY$wlmn;dq`legi3p0g<!+p1;10{lV65Wegzf}P0iWyk
z;rMcU{b<m=9^=yOA7|&dGwJ-NQw``Qgh6*ecSFu=!))&4e%%J5?;6KnKUW&3Xz+<w
z_mLAY!3Z_GF@S6OKW_P7g7{-fIUoI!KUFbNKl_mWd1?F~r`J8M9k51?n|W(kH$Iqt
zoCd#&r!!7ioLMQr+=lr+^6?bVuitfV<?{pB!S{9aO%Uoi@jk%?5Ds}=%DO!Y>B1cU
z6Wn+A4T;U9#j#<f8Q@bNz(Fcq0gliYylP(K4D~ahOSsko4)F#M=G9+;=)KYK&Gj?U
zB8XoscD$}TYuhIs$L>4hoe>E5a;%P~z`Dk2#=TWLqI->Zz=t*)f>%}l>)P34#tU_3
zmqp3xXB|6+-M`s^<!B4YJU!=tkZBrNx#sB~K7oR~x;`1q58y4Ldyie{mebUBZ<QOv
zJsGARV}_DFI^@Q{5?;l16K9A&Z8R60@0G3@Cm@=DEQXb>4*b5Wr321%u1$D+6vlzK
zh7q{z7*_YxO0XTjZ9Uo!$e04Kh-q-~gEc#=sw)h1{+$>csm=`av!_9L^n1U`FXtz&
z#Spx}3U;U{dKbK>M;os8(t6YBJsG8v*W;uGx$MU4lcTgeBvmaZP7)`y;~I*))IzH4
zBKCiP!o2egoS#E?ojU`-2XZ=9v49KsO}7=^0DrnGfHS%~H|wzGDFh;Wd6oh|d-?M?
zps7|Img^d>X|OmLadI@!IJf3EM2zcsG0m1|a^NMD57s7~9K{&%1YILunuxC^8UsDu
zkEi{<EkyzEZp`THeFtuf!hECP+*;e(o9@braNoU4^T1@89`L5PpJqq)?y1H@GG{HE
z-ghM>uvVb9D)(c$M7~pB)7QfFDq!0i^Nm}_iR0w}BvL{+BqLg7^vr&B!%N6P5mw&-
zf@$eO_vrkF59Z_P^S<-nN_nsAeh^S_P$0?5gwG}KI4n}A_X&K`PrqEjEP@itoXOiL
zIHUa3aqda%1i~L2{XasQ7Ff3#U-nBiuj<@6FWwrBp6}JjeLE`o!BG+8HR1NxqHozN
z5vd29-40fO$ld}99tq{YZeG6Im&~r(o)>`IujFvE5S-J}hq!Ltl#x#?)QaJUk&h@C
zS?$q=<5<!qIbdS5WSHZu^u^6RS%`3jJZRl)&I6o?$y@zV!M`P{{9l06xYNskI-wfG
z=Tz1Z-HDyH`Em@bt%Hc~pzq8-l_W8yw0@~&Mi&O&Xj(gyeg8Z{T7PncrPu&lr7?z4
zO@K6rT=?^ATugXbaqDe*Q#CC@2o*5B%5cv!$|;@<CErMaYk43oqdp%bSpI=&i8o#J
z=!F=@13vU6e_|K}Ma1KZYpFFVonw=e$Sh>hUUpE{JpW%=uHi)V`h}2UX`agM&kK|~
z?2%NY_V5{wL{K~UB&mqxJ#8U~r#7UfMe8Y@)w)vE&xM?pS-%~z6U%aj5#j(!UWl1U
z!Z*8cRZ4e?uZ>rvX1r1{L|QZ~e6&1UP_35?yMXJhS6^mS+5j%mGu6T+v+2lt_||y4
ztoDTQ^S)nB^IoRoFu$vpgo2n1`+zC2gm{+0GRQ!SH|V_}oOkyjou&Y>2+vLmFFMDH
zGHV-*M%>-}I+$yn`-Zm{P;x|sY;iV=co=-OY+NSW+Uk0yZPJi9>t028&L~wfUN`X4
z_5;GEcL14x1mRwpUho}kQ7+l2j9sp*JE(!4&I~ihUvO~1`a3>eL;MC#-4SpU1ynqM
z-qcU-;aap@-22O>*`g$KFkzeiRZSt9kF<vD44*$u_PM#TQ&crB(69lVQ=?yG7c(C!
zVMu|FmopO0Mo72DC!4VMQfe_jKngfCyHQSbWi^hh`;P3oSv!u5X||dZ<#%WY27kRa
z>V+8vdjstAjW?{58pMM}BX4z#3N4RVy|#hnb%A5lC}S`b@<eLN`DSn;nFML^C_4tj
zv_`lJ<24%KV1fxR1%XM2*C;`vi}mXl#X_-1w>gB-kthjZlZ{PA=22yDx!pCVQu6pk
zkcHO`ki6zv*^ClHA4mN)Pih^~ls5H@pWHAp#!A#DSroDcgX^{StT;L9141+6fQfY<
z7!7BKc$M8cp~iD1Yau7zqmPh|Z_i=(1<llyFVY%JZa-$(3IaprTI;f7(h=+Bzt(YB
zt7B>2ugpY{NtMbX+c@<M&?;9Z8nt<SnPU|kNE7c9s@tC749(xic5tPCEq%^XA4F3<
zOpF}q)+I7AT<wVSck3mMb22iMS_yOuKff7mfgk)Cmvk2l3V+}C70B$Qgh2nfd<nva
z+hK~W;QjPLJf?y=mUdz^%LFwW-<&AZESnQjH+K!SoIPgn>nV8Myi3x4xA66;Yjqp3
zDdNf5M?ukt+{UDF)Wd(jEdKi06Q2=A)fzQz2X6~Rwdx0Lc@_xR6@EGb{&R~sk|LOV
z_f_%vM>yWT*$LOeKUo0L5XfcdDn(BPB>6Hlnhy3y;9xHF0@gIX8VKVSh?>?%-*|%o
zWtlHDG7J`Xv*IvXx_FL;DkK_yaLbiR@G9&9AF9s`8r2R%(?n}`G1#aLNh<-H|H?k_
z>^@pZ!Hcm71q6F?=)#k0s18GD)N@#q(x+i)>P$3*QG#9E)I4sv<i6~hZWKE5^ev27
z$~*bf4A)$7_XQal0zyUZGBA1u+4hZwT-0jYaah8AdLzW~NW?hl#NDe_Y4qYt4C33$
zAT+knVu%pI@Z#Ju>nsCo5$+ouS2oA7$$r#Iav4gE+@1fukSyNHyxa0cxgO_n6^y#{
zfYhJ@(ypR%tx217=sg5*&-5%`oj(fY9>8F$;5{WUiq&f*{t(O)C^s60eirR_=|)uA
z+XtkpV9)4xzuo8hO6dRlGuIbbwIFK!LAdiKxU-|-iTB3%w{f(CP6oiY-q|nqgwoK#
zngmffX&6C)^B)*m%*$EcV2ro+y2BK~#VNG-3rth?8i{urB2UaV<`wU;>Ub`wd!s*a
z3!<2{Os@tAX`IDzX$<TcYS_2mTZc>pDyXszK3Y-2&_ui%gNnUG6^zX3tiMZkmbPy%
zwboC<p$RTSM~!)w!CfvKHDOWy+mHXY<xt4wGNUO`teP3AWBuz#&m*gNaU+m@(?xap
z&QEMb3dQ*b$xoJ__-fYqWeVbmte9Sbs?pC{0-M7iy~B{y|NVZ85_bRlI{EOJ4zr|Y
z!`yO7kC=(#k)+4<1W0Wjp67{r3oC7MV>BF}Tr9xp)cn=h)%qI=RjU$v>3&v1QTpXB
zyup`uRT{MwQSaaMlAhEH7Oyqlo1scV|HkFEt)@k@RU8cV$!PSJ-kEz@HQyiGE&1un
z-EXmapzVXJ{dP#>e9vG!JGJavRo=40%S)_KE8l?8EoOXM%J89wN-L)}p*6VD^*gB*
zxz(S@U~=Y0!OW_Vi7iHaYW#@=Nx(W-aat~zSWc1IQ1eFC^ovc>#waV`Pw;o9Xh7{+
zVLqXTay*jE8)7gY0hECTG-~BUQ+;RVt@qU&kuz1(tK~iwey-PvXx0l#UOR02z{2R3
z{#FD2)uXqH;<^-{+_Q!21Y6N4P<s8$lJ1H#P$x{=q)gxAv}n1kNi9YOYqDL9`uO3F
z07@X&C#b_M=%|DJK5XRdaL!UWn^U=oc3Vz%h*GAVMAHg>4VC)x%cyK6c%lSQ8FHmg
zW234Ex&f-Vf>?WZlppx&y4(+7;+$2BX7Uhzl^gv8?ndeutj_O;zz!IO*MnMMdP`PF
z$}LXH<E?VPk{%)9Im4BHh`JJlH@S{;>_hV_V&qBqW7?pOkb(OpYIm_##U5lbQ0}h~
zCUKs8Y}zxTHoUAbMu|8e20;1S3}4~o4TUVD>qHzDi6$AS^s!HSzCcSNJZG%-(~NY>
zxba_J!rHxckJBiJ-5<IlMv1-*yS`q?+Y75aq3)i!%Nosig(>v}@4%h)wxn3Q%CBtj
zluaXL4o#i|Gu~w1&vxyhp`Vi>NV|1!m(VAX9Mq~wbRya1>l|-vwL3*0(i`s6iL1=O
zT};e|jHv8CIk<y1MqQ<Q#MbQ=+j=tg2$W7xEfFLSdQ?~!{QAo4v)%n!OA$Fq$Br1Y
z!;Gb(o=<5(I|_?;n$8`Wx`MfS_2gIh%YLt0UIaMq8ZMRJ8YX5)kpVh4bxu{t#-K68
z6c|)-_s4Mbd8q|1N5x%S?wl<xs2FKozdJ-N8i;$5bX~<46b#cBvRH2H_kFIrb;5P<
z8Y&8-W~|tc)C!@0`w*;-@GXedw3}M%RyCsZTnS&snGuY39M^d9zs{@3cN{!aKT238
zJk5GFA1C)I$>(O>@Y9wLp<@Cz2SFmZekViihUq_FKUSBhTX(3+h=%DFKQx3f&`c(I
zp}G7xQa1zA|A;R>euihrXhCG_7DT~`BQ`(@NHt^qrZ1CFYq#8_HvfgC;N-9%#L}BE
zWbSnl%!8S%!C~g?Q&?df*JiXG_Y}oilmaTb;b-%bwcXZL#!QvhsGR&mXFfAyK7*>Q
znYWS5aiM;y9dayorRB<|u$OPX+HewaGoZ^5*Il4yH}BtmL!dV)YieJZxTp_=8%-&o
zyeNj9w4d^@wR7};D;xY~^~vn7Ho=cI?@W@(><^H-gZYv%f2B2&ltV)eJ?VMEM`!5g
z-3=U5eDB@WkcyRo+S?KpMT(gDmt+@b^NO&Emo9GPlU?Q)cI1!t3uC)4%OiUvd+(@I
zP0<wmVmZ0o+<TZ3%8rBFU0=t{NPH+oH2bUMm@vpw>mzkcc8ypatA*LNB*2<5I`r+;
z6FJQJ)e%Zi-)?51t#)__{WHw|=EDM9wGtzpbDod0Klwk=g9f1X!z5p&PumB#nxFL3
z;^nM`?B<J?De@g_x^#Ff6_p822L5#q`w<#?A(A_iVp^ZvZe8nsB4pZ^0Ucxp?Zu5N
zB7qHqMBO?Ji;?V-6kPb(-21e6ll}L7tJjvLh-J3!$=OPI3Pk3H-kq9IwGu`McDeoS
zy8o+j1y&TKJ&TqmKS$QPd<yU$oIIh$D!j9{xqbmp6?h4Wc+Ki?9P)2@p8YZv>-M}D
zra>~tz8=Yr6NQVfN7#it_-L-S+UebZw2F`!+?z8>dZIyhTooOq8yBXFjB$CitcEP~
zhRJ&Z{X{11wv@|ZJG(p=TVbwq)tX(-%+46x6Kc2O8?`bYl_|LxBE=9@ynypv_~Mvt
zK|#3g63lK?35qsW6rnxVMF;C@@>FifU`dzlm>=^J%5~Cjkp1LTyUJ^ZFAKdPgZ-m_
ze5LoQV5g_$(oK7*w|WIzsGXn4A5H=qOk+gD{vhyOda`9Wp5)V55#n1KV6&h8lKQfW
z>abl8nCzJi+^*Q=R#WC>tL<fDLOzvQ)5!I+GeNrCpLDsA>Yh9ECj_+8^89!`G1U6&
z_LFir*~zOC$W*UuBaR+qUs!3rjzJqMVNm;q8r?6Lmj`nI@w95KP`Yxb&X2+qA(Tk#
zW}}OKF)x%+lypKGdza_WVeBd2RQG%5XqO0Q|20%!1+Dn&A{O`~Xc(226D2^1phTh!
z4*`})^-B%V49H@B<*Z=HXnEfCzj6tg0o!rOPA(qmh-HX+@XIde3NM>f?Ki=@b)m?S
z8ZBnY+=eekAp!Msx7g2_(R2;k^kPSMT{NJBGSD#D$?~yyhb3fS{bIS$aE#JQBbnG7
zZt|^5A0mzAdB2ZuYVKH)&Vslh&~W#Zrhe~H`4;awqIZ(4XFOg*xvqm!ZZE3q^&;1&
zD}Cw7`K=#gaZ`_boX<cJzTzkPm1V&dTg_r|%|o|9o!)ts(J=0pE1>)fv2G7t6tY$_
zr&7f?#ggyU;U;~-k+<Sv{dg@V^8BKFJ~|BeARs8RtT{3=i>c;@chNiBsJ@mToT$KB
z+i5{LsI=+V1zdE}C-2eb0R`;zhF#5@KoBN74SF-VewDtkOYR8G+b)y{KjqVKP&EeI
z-)jm4eU_&edh}<GIYP6)KYc5z=pr^`SjjLC-29{~e@|`~N*Ky|*cYMG?U!|sDsBn{
zRsuzBlz2&qXRYzRPlI2b<FCHGP%gkUZwI?AolfvR{+jv5_B!@_3tcA&xhGhCjIJap
zIdkVk%|^`d)|#;ZGW`wJY1PL)%*!}p<&P&T;}Eo5u*HS&xF+vFv<iKM?BuHo@Az{$
zm#3BR#*i>}B#m9<?&~?NxlK?W*U=7)u-bD<Sf=r2MvsI}<gb|cluya5^Bu!)U)PoA
zdhcv-Xm4Gdlsaez=F3IBl;T~V6)E^V=(;-m?|_?-si0Sw&&iU&Y?1y^w1Q7T@B!1y
zgQL?w{RRU@y9Te`@{Ggz*KP)oO>xV5;XqHvkEavQKV1H+xdgiB59>OR<Y#vk>{wpR
zT+3<|o~#PHGPUmFRVk=DzB>(IR}W>)<4kpZYRj}?NxM%TDtXvsy-YUW2aBNyZn$e;
zie_~ke9Toh3NRlhmq0w)gH^MYM$iFk*<)i+W$ps6uMhi)Ez}J;hna_OLPz-*SD3v|
z)Jsog_E_h@9~X2I;F4e;X#S~rM-kZ+Nhzy2^>jmQh3}Us-_swSUNMlqn2Ch5x-<9B
zeX1<fPv&0ky&hOe9qvw2<vre@>k~cCJ-Lxsm-9Q<`&O=*by8Z$tBw9QN8*+mb6#K0
z4m=TO9`?hACl1!dKt0G<ke95BYf8%YefO~cyD2Fe1KEiZpkPFlmoDepYIaH|&Ko1;
zSB}m><yvUh4&Td2C5=;|Auubw<Q{`}Cr>3tIvdW1aXW;Gx0A-XpW_;jEIQ*~_-U~_
zE21qo<J+Pr?Q|wj2E_bOaTGOgEP*xDT07gyZl~+G9Ch>Px=!F_{@(y{<*X3X#=Of-
z@PzcTw~D;6ES7VbIE6H`2$V`1K4P;$05$Q#ZGYca`a$Q!L;5w%Ao+qj#61Q|y~Cgs
zy3N+f{Q8=6FXnQU^<a$ri#Gw_wdp+5Ypf9*)p4B`pqzqH6J%rgQ5pe9!C-Xt9*=Gp
zQmo9wfPt%8QuET<Fg_Fg0dv;SMHN-N4c9{FxHf2Apj6+b(!G`qXDIEwI8al@-WcbR
zIu8-)=5cvog#SobFtEpR=A+K<`2~s>E9VdWP@<1+6SCYAw40*O8`Vg*MXBkEu@&~A
z2^Kv_S;cxdVp|A4-kZ^5;Rmpq9t|hvR|4!m9ouj<k0ieK9a#Gg9_CuH552>8BGPXV
zL5|*{45m8q{zPEq+o13+{&<vU5dwluYwl3kT7Lt~u!?2DmhdJL>lZDOsTGKhFx{?U
z6JhdNy+l>0)UaMuP=rGGA=BND=*Z3>LH+laF~_D1q$)esDQwumZC-_9>y+E{L(*+_
zzrJRv76L;&;F<H0g>{Aq7IQT(zSvWNhkf9&$QRIhA>Vg2rzZ@;`+BdD5Hj!n(h3qx
zf>t70kIdx9%U@jL<aJjrw2{~Kk0f_&TzX~JYcXCUjbx)&RvEU5$k~2LyxWE^$@7B0
z5usc_Sg2%o?F73Q$xkw?0Az?>oCS&s!E`+P<K~_^n`WxF2%lm<&v3>@upf|OFV7zE
zlcwx-tTL#u-Ojg<jX*jWhAe(y*Y;B-gt|>ky#n&3ZcbIK<hv;C>MqLm<`AMUM=umb
z!aY=Y_!v%-y`d(oOU#Kyb{gawXI8_x0WrB8%!1C*&-&~n6LJ<BvN=r@LuJ<Z8q)U+
z3TjfVOKro&30XHh;2~uN-b+~z2Rs)$Y+rhC&rKvj^;C5WLlgR2^EFN*`iq-tNL~rW
z6<Jt<Gb5YuhcG=sZ@)zO`?gHw+!bMGvC@K&hL>ae>Rq9Nb|8T+A$?HQEV6}5VMRW~
z5**%6wpE<?ziuQ02M|FY%P|@V7pAINj-TAHsC{4Si555r*b@w>5F9iqqB5bYwtoE)
zI;&dgP$R#OcLqY)k-5m0erm!7?*UYc0Z>df=a+B)@ikl{alhr5kTuYzsF~A)f0;ff
z7sS;4&<J0-9hlVgMqn^-&!KudfmgF&1G&JFMnXQH!j6A-=Y*@a+x=bY0MCOM?DC)q
z&<*4+e<%uZjXviMNRZWIEvZCg#-D!$>0)n#*?)8);KNG4u_bHT?8&U%m+%X#1^+z+
zY$a7!zTZeUY|oEc8RcWJ2@{C2y59g0?@>19<C`HJfzGLMPqET*Skt<gD^-0zjYrHr
zNmP5iBxiDtuA-31(0VqLbU#s2b*qyh+xb0X_fQ|Q$VOfm)yk$ZG~y=qdVQ?P*@Tav
z50+>^G?j)>xUM0_!(bylDD-BX({*jjoS6Hey0ZrQu<@9dz}3vqF6^B<`clJYz$A{h
zrC*Sa=QX)GCXv-TxLJBu<6V@@iz{*Dp%rw_oARZqQd{edGuIx-<KO4`BAmpsU$oKD
zJUWEK4?;SEO*}U3Ehz$A3ci8Ur1&4F$%h~3QJbdr_McoOUTnnGIB6Pst_X0Zr%y#|
zYp;#%)#}y0x8V~&{Vs65{$e(r?DEX6IYp!DFtr6_H@`*i#7H$&(plO&)G4F`S21@R
zTsJvaJjwH*PYydbT+h5)`H=dt^qb8hcI5(Pba{D46kK0lYDF%6i7U2+YgOF3>Vd@f
zGgJ!V+w4Ee5UQP8+-u#I886FgcoQ=zp05P8xQgI=&&3BTA6kxX9daTFW|=q8j3_y)
zOk!Wz#XB(;=Pz}HROxjBF12MGZv#woi^vw%bqdHxKG<@T=5p!1LDjpq*F_LT%1~4@
z49)dJID$NMnsz{_6X2RGNLSy%JFl0xlfKce)39~ksUHZg<e00ifqfpoHOP=Q1F22s
zL&1loqrMuGhO4=7<8tHn6m~i6-oxYz9P3FZD<H!m*WaR`C?qV7@CQS>*28Zkw$!HF
zrg4YtotMA%M!*j*LoeuH>Nohz3aEya>xb*;c9o%OeTeb>YiATEGzcygs9pYjQ#d`8
zj~>COzNF0~eRl!k0Y~l!>5)oiTWm-aYoQII{fhYQBJTCYoL4~RFK1b~1?M!qRD`~B
z%uh*?=3qxWdk67I!SrtDLQT#f9uJ_!-LN>0Xg}D!NwrE%kt5QcO64sa+p%!uGw1ct
ziSH|+#|=VutU?*CEm=iLFAsqJ#C-kr291GLvU2`A%b(%7J(TPZLswGCCv7x+j23$}
zUKJ(vwzo!zvDp$X5-KeGR&E%=!{DJq5qp`qbEfE)wSqZHm{X0KH4uj^EMFKLx7fD1
zpiRz*hWmT7Skg<J4D}U5K9EBN=>Xv}b&bI^&O|>nj`qWloIP>@;TKpRLrGH@2%R<D
zmkUbXRns8DzO#C2vc`L&shM12u*{bXbgsDLMzexl<_9=#_=Hx3g<pp3F?r~iiP+7*
zN(?kT)WE*G&&(xB0iAuClaxE0f7){83=hgehl#(p|HM>H=0x2ddl2wqS95iIr+M^b
zs8)#Hf#hL6Y;Rc7C@$1Vm0sg&#Q{sO$|Tw*6+f_2ka6@lo3T`dpK5sBUY}m}>D(9q
zHX83vUhRV<vw$k4Dg#t0qXQMB*eXc!T(lp-sANLRm8IZeE#B3mC4dw4vZpaRaZt-v
zb-2&d@0ut}XeDg+Zv|&Kb>by;`AGx<Ucq;58J|E_-Gird9Y4m!6l-b@GyyKjOWgQ2
zqZR9ds@!SmC9;ei52itz#{h+E^R&Oa>o3aTOq03DKt<@}Y_nrZm$}mR<UT|8OFlS(
zyLF|y)Q&3+qdcrU<$Lp|D%G?$c=&vVinX!T%S$%=w4#5k%>}uLjGGHQ6hJd)j!k$t
z1N5E&T_&^kO_E9UJ#kH^vv+q)M7&3-dc_JQ$Yj>vU4#f{wTg<&>OIfQT>o<OUAJsI
zy-TjMz}qNq^pHFHO;sH2f>J<Syd5^PEY%l5W5&x_uiF@#qJi{}OBG!_)C7Exb`MgH
zD<ZdCI{qxLEk1WZi`bi~zh=HF$Z-*hp)Gi4&e2w^p#F#_r_(ceAYeVSd^<?3tf9cq
zFfJ_eW%8(`powBo{RB?dLxGp9ED~%zt2NYfo>QhVhZOB9y10`=Pkb^B-+2)CfZpz@
z@pYs4(m#dHSR$8WF!HKYyM-Myyz@C?usy2LFMmwy(~(>;8HPr!O`MZMO{Hsb&y=yR
zYr&{%I5C0Aw&v(%$YawmLIQ&-XL;=lHV^jgo0AXdIar2E-wbfD^~K1yN4@cTIMPX=
z>PpPumvrvNf{wN1hFTynQ>r1SEuyK#dwKp+<{Le}tr8Z_otXR1q}sqvLr)D0ie8+k
zf*UHqnI>=O_qb{2#vUK6(&=dt5fa2zqPuc0H~;7YhFED%0?gzcEx7v`#~eDsFUaUv
zhzmbv0*n>Sy|2PJ59%L0)?&BUW{uf<$Y8kUUl}x7Qk4^k*#9j2kY2BnNNsJxBJR9R
zS6~)?&)2uSbV7y0V3*{WnRP`>-GW2ez|w<Q^J#6?idvV!dEhzoiZXftUQw-=L5&A!
zj$0=~gkZB>ssp?3N6b=hY({vREhRJY8Qu@lJzK_iw6&?|>&0gXc#+#Zdbg@^Ls0wQ
z!R<^N(YJ0yZ&HX->^*ONH6E-hQg9H{U?DD%Z}$=_!0&080*I?4Go3wS7LAd)GTi2=
z_9K#cF$Hn<v%<}%ee+Njrw(PZAJ?NMBa85XydvqjO|AAn?&(B(LeVB=R0lJK`VMFb
z9n%A>#0yHR_GJ!f(I;hGKds{&N)**i?thMgN>RhztO%dK7%aw1)`pK#DjxlWrV?3a
z|M9QNbG)$d+pxDG;*wH7X`s76eG9XgKBf}uvOThS&F{L@!CjkY)UU1X#fguAWwaSP
z=RQsNfQkL=Yy#sHm|eps$X^{PrOFDLKZzov9es@x_})`*E0#sw+%=cP&Rg0oDGeMc
zO@@>iv;+k;vly*w{1NK4la|L1P!0LMc!h0fCv!-T)5_^Au4eB`U|A8QMRW~e^U>*3
zQ8Gh^r~Le@k4Y!nivS0yk9ZNMtE^0FQ)pKlr5h#0!<5;2d)xX$vTAt?r($AxUz|o3
zm}wh&e?BaaG{_-yLWgVXi?;pXKhi2kXszmQcOB-`T6dJ4F6@>5(d{9b+vGO!R#ToG
zfSRCEX?Dt0E|MgxZ*VmF<%Ux*#@-8FUiiJK@mxcBkKe_Qwu<e4*m2gF%XvY7*nG<O
z>u9J>2+(KCSjKgL4~b#2GrlM}IzwTyL5GYK8+?%?`Lf5WOA1>o6YS76g1GM1O_bt?
zY@-D8EnR697uuR&?vNxc75Xp6{)-*7O0=hl{mKC6`0nkA+to{hzWn49T00L^E$<_z
z1+N#V^q1}aGNI=^f{V~IhI3{5P9=d|rp9dd;;@lS@De`P$sa3v5_&UWEf-|fYw8+!
z`v>?N_vkp>&B5gE({YS`m`jpJLZ96AP4Ddk*-Fi_qGEzTh6iC<k6IHbk?PN?$aNNP
zn6O?MA>zm1PD;Gn?0egeMg)}MCHaS3?B0$Q4J7_{>Bje0pJ=o_CMz>RN>)6-Q)NqH
z{UNeje{Uj}RPcVxrKBJeqCIj8n<0ODptD;_X$-m5Ca$4qLy<EIj;{~w$8o>EZ*Jdv
zNi9jybSWdHUz0#mdlBKB;+)iCP<OnsXa2l}YYS|;A|dc#V{t?MHnKi%%kxOs6pOsa
zW*vY-EVYXNQC$A_4x+_dbu-zYd*ZN@IOVePDfCtO%|@LJ1u0paW^jd1qc8ez!B#7y
z#wCV5XQGZ*0?rlG*W$#itoEWutM`WQKGnE75Mmvc*5%{i{G!!|e!g5s_LKAPUL+~w
z6qGzV#+<^6(GQy)gC{}(=Y4Q?HK3GJk+=d<kNNKMAva_@aPGYtdE%2C7o;g*prRru
z#bP9g1sFqofx_xh@U@Tx&-(@9J=f0XgQK@`#w?fdK4dnO5;;vix6ERL0j)cQ#ele*
z+cX?6M6L0d;4VB5w5|~o{{}PUk&3kV3_MLn(QN%BtHYm>R7iA^RZ=P8POE6Y9r4ob
z6Vc%eBF!L13!F^!o5>M{`rgNW+4qb<7Uyb0JG+o8KJ~2Qiubb^DPf+S4&_;<tF}!%
z3eIy(Boy^P752H3@G(Bmp6;b{XSc7#bOI=Tu0@(ZefW`l`SZ!!-&2Tz`Mv&?q0%4U
zwt+r6T3zkkqq*}3*HaafM!5vOiy>fPzwK_UaZO#~Fj;}{yoL)6U!{7Y#ZLUF_tNC$
z^-p@t1J##l>ID+GK85V$jmkdPMNM7Gmzx~^1AH}X;we%6QG`*;RC|Vu9^Z$i#qkW{
z(#zARmg3mRyuCAHRb(6@sAMM9ki@FOGF3k?<jp;H_>0w_D|+5kwGtZ2H84|jkO^q+
zqRBranoUew(HbYm@11-<n>Z+c9g0ao&X1khAKj`SyT!gDTdiK1d2dnPwJL?li``x)
z=Y?*fw}o)M(@WorBq>EzHmjWS2O&@T%4kP}ku^!+bm3}Y0k`g89=tkBzPOocpLU6$
z=NZs*d{@L9h_%+=#a#+$jktur8D>AJ8GF=PdANTAPlnm4y>=6f_`|8~l-7^a1$(nm
z(dvoVmKhL9Ok*LeP&)7s85Y+<oN(1Ka(%c}p{u%f+}*67B);Gg73183aUzju;zuSC
z)z{s>Nj9Gtzozm8ZZCdnxEQM~<A#*rSFSnj`@<$BtmWi49s2JDLM*f=emnz|<dII|
zVh11ueQuk@yCkivH_@(3cUXmSSeTm?;6Of?W&x%yC#v=wD<##-)H}JJ{X;d+vYJqi
z#JpWni3-bVE8c~ZppQRlu=^2k<ADWO*T<i-zF1pMVO*`!<GVo@(dcdZ7uN41ciPy4
zy($e%WQ9g`%9iVQ@n&@hmLl_z&R?m!wF;PC3Y^plD-*xD*d|N&9{AM|JA+qfEKSd!
zz|48eMFTiXH2fDU*>$SFrchM_dSjW(gUX<^9u;tdYN_)lFnR912>9mq{Wt$fk~FwC
zl8_TqHnRTjuQz{fw7Tm+{gDcs0*X2;|JP$+J=fx=M4eFqhjwVy*~fc%Od(VK8cO({
zDYsV#AN*I<8Rm5<&r3Pa&%_f(4;pGQsX|`UKy|lm_j01=4mEF#CU0Ia@IUb`GVm71
zAX)m<>QjXt<bFVTiKqNT>2KQYKW`C9ZPiWW)<p`r!`esu{7rd0W`J%O6O~qS*g3dh
zXPix|n%Q-h<lEFiRtD0=hW_eK%a2u)w#C<s#N+kE&wnMWLHFGMe9{Cb+h-C3be&g@
z`RHW0V(y2DQ9o4Xrq8sy1~1{wVBSBxtPZ^IB|Gj`G$p9a#J~CCoqUTF4;{b!VN3!_
zB@MkIE%p*j{Wys^c!@(guQ%1;ELR~NPjeHfQSsr^DO6BBS+OoT=#n%OL>5|Ki^=E-
z<VD?WX79qgFHZe4YH!8vS+V_UUy28W;B)HtKdx(|W!FRo;(@s`GEwvzuh|o6D3`3*
z&l|Hg)n2f{;aeP3Xd-(~ZUtU5-`eIrsuuoyraX!O30FIqe3zp?xnnz-if-SdCO&DX
z&Z<<qCcF<M<${K|@j~I(Beo+PBFbGBWC-UnhU7ucZi&F!-C>g6?<*Ue7}HyLiT#_2
zk@D=X9Hu&gkA{jvj_lB5T)M>BH%`q}^Wo%c%8YO=Xn=HYnWc)nPI*2&p7kt=aXJsO
zo(s8`_rb#p)}=pL(t<{%rfLc3T=v#S-u51)Ub`h5vzKbon-<m-KYlBns=t$QmOfUI
z`vQpRDrN7;i<|zkv(X;W3291FmW(!kzk0Q!g_b5WHD0<PNa5|+PzX9o8{(vJ&;cMk
zOLDyHTFjb{{=BZZ<hDp24CRLEh`4+MHn<=3k=og75?-uJ(XKQLR_j^!f^VzUvSY9V
zLp};X_^ZqFR3c=r0-kZ}g%zmONrvy%(*)(L+usO!B~RWa)HEuq=h6|R)=_a9ENhso
z^BW3qRy=~nmG7u(%@Chn(<Wcd=C;#`+e&JAq1{w2Ol#932ZDm9ar)I)8!R5*>8+EM
zR;JzrYKgdTY=X9!B6YAxmeu$<tLz>6h$ml!&efi`^p%}zF$Bw<DDK@{wH*!_BqkGk
zxGx;c@`Lk)*qsr~a%-&@`LDB!@f=M3{m8gvtKOBIxZFrq>z3Hr$@LSCtjam`l>ZIf
z&Q1jhlBdDjftHSG;x*-G-q6>zGhaX$`gh8<Odd9ZIt4zZ6>`xrYS{CnyJx|sb{IRA
zB3ivbCwA<3pp5u^D`#HvSlO1CbBF1>p+i$j?HSc{Xig&D1Uyz-y&qQ2B&g&^8a^z~
zFxDC_{PTh5=5bQ8x?CaEB?YaYUmP-p-jsY@#m~)nac3%}MIzW5LrwDrIaE82(7)Y%
z7sbV`z$(!Ds@G+^OR1Wyhm!JS1_5fK*e`AIEmW%+6?xxz+Ji++ZC^-C;~ZWKm%94T
zF7xlJYU)r<7P^f%vhZ-@@g`V)f2C^qX%6r7s?SyEBS%Zb$f4mHpF7u6nCtgH-Y55^
z4s+1PNwRko+J8m6ybMtLlB+0#6!0RqfAPpDPQ*#|UEkf2fP7Fqt*-rIleQXHONwF{
z$I?L^%KBBi?-O+q;B?O&x)P=M{&1l_)=kYv%9gfuT*25Qg@2usxY0z(mA;f@S<{sz
zHcuW5M0u*pV)pkyuP}R(B9&~3pcO-rUf3gf*935S_O?v!Jv0jL#?g0^kPqn1<lM90
zxCA)UCW&D;1%eqi(xX%as0tqMvH!j&HT^8|%e}y_;r4&Rc}{L56|HWM%HC08rKtHX
zMq)^hO8DFejNftxWPchhLK-#;uolnz_vSt=Oeh1nqis>HP{u#>j<%QZh@Uei0F3&5
zRzMPKjk_FvC6a$p32otpv}gVajkDs2Wg)KDyB|@<lfl%gtZn1L;sM1TEB}F`<lQN4
zH*hJz*Y=cGY#@YL5%FYtil{hPfS91_#&30VJwk3>3a39R+J#(+DKC_G*2TKgh$x;>
z*kvwmU)7{d#Y3$UY>QEcr!?lYwx;!mnzoXXiY)`CnLZ7RW9oNnN<{GoMdJn8Vg`yL
zca8x@%1luQOu-!Edp;6kCKl*hzhN>pWlcU5I|;=v_$$#Cu}R(I6yi8RG*J`-aajHi
zP0O}DOy0I0j6?NCFeIPQ__QJzA?3MJ!9M)d6F70(KIr8e4-<)Z+L58|4Trghof7UU
z*DHJM;6(~>vtS}P;(}f-%P`5)Hv_z<qNY9LB@CHh=o~vg0gq&R(`jIR5+n`SHt4Un
z!6H=H!-NPQVE0}lwy%7`lo>2Vy4}R%HoN>@7wZ_m4v8_03s0+~vSvBIhErWRuy%&U
z4AH<i_Q_z722JFsVd0g}%2&J4+uj)is!pt%vD*6T@Bpav&3WR<w@!xNmu){MkCf_(
zNWg8JRhK)%Jt9ca#n+;qky<Do$z~T;!STGP8?F$1pju(W(iiP3H~HUq^JDI}1~&J-
z?A%4AY~g99{0(w&X)`WATR#^S;$PO;g$j6m5}^&rKwef@?%1rGt4#GjmlD68?-~PH
z^+2azsHbu@#Doxfs7{g8v&pE$!cLTxRBT>{c7ync@@<dxkdqgfaIMp%JtHlK)yIZ(
z7(sirIX<4=)1WJN6tj^)k+`!x^;gFuoH4uxu*utqkpyw>2xEH+ja9d<WT@jaV4MHZ
z0&`7$L~&<s)hk6&z;+rqxx4~uy$fn$>OyOU&$pu#d0nn=iiB$KiXkJoRlflJx*nT2
zmc6^}v^FDAAwF^r(*NLAZynPqhlsV+pKg4ZGXWg$>?pQICKx*0$*KHfO?+2{kzrYf
z)XbU%)24rFkD=3NSbdgER3W(5sv99xSOASC<ukjt7C}r;d2g~J=m(Rfs<;is@U}em
zn+0#S5dgoKw9zy^NE!qc)u=&NUZRTWEfiylM~SO`a*dw;0H5zhW?EqBU~&F3TBFp<
zyCK=4jjgXxU{bhwWT9N7+R_53=F<l#NtlY6BRI*r4thQlHr~-D{A$2fzpg<m=()({
zd-h})SDBW;5B#{&?b2X`Ux%dD#2Q)I-3qtkZhZ<_e3GC*4R6WVa5By@*&|%k`6cpa
z`YXpPlB^X#Mj_~9Z2H$pAj_wwTyVi7JM!gew^&3N#jpRI7VyE8(VlD1%}IqDnu(_b
zpPR7GAdorE&udI)P7>No=Q79~X3Etm%uA}s$@J@v>`(8VM^TIhByF;OR?lv4b*uDL
zq7WadFzb&Rm##YTne<d-%}a964l=iycIwl6ByY3xyDFWn^Qdwn@(rBaI}j$u!_Stt
z%N=PygP*=@n{##6mfWh6mXEhUviG&{4vQlP+hT;vxgAX-Z<caQ5DQCZYuOaJcW<m7
zgX7P};_$<0_}GsVqTC>h-rn03V-441y#@s8miFdW4{OFxR03MjeN6)nlQn|dXXs=K
znyFqI(;Wm$FWnMN%;(CC5)(~<3IunuJBxLY&bmgw?43!tF0{z$?CNkzK39R20cb!y
z`C=E*q~AR>m)7|@R)4l#c7ESHH>C1(xgE(9lq8+ACcqxb<bw{)`g-c-E0E#sv9vU_
zT(k8cgT8=g_Ansuv)32ByS+8zpd2<%44l0A(~{1wrOe#8b8^$$E+K47-DF7QwVsNC
z!}~pHH)CwT@tX16zUm=hS?b+ZpH!sEsuUn7s8Ru~s(6c2zr_+qMR`h}zkE3yYRdHX
zbeL#`rh9?;h<!&Fl)|+3q)%E)w%{{?D8dQjDT7yn&eVvBQixuyeO+UN4`|@hA<zgv
z5G2_c(6RyTvcaT0oH|CFCT2bPJsES4o<Ci|5EniZt^FWlc7fAHa4NfZ(F-c4OyRGq
z8PVW^9nn*oc|1VIPAJl~03)Oj$;dy&2J*_?UhVTpacf?Gw0*<Q8h&urqa-eDK`bz0
zLT8e?&c7e&aj8^1x1)BD3bI@BtH$*!OG|3N?E07|zjs<Kr^%=ol;ADTqd7NDlQQqC
zjbVa_kxlFMF3Xj<2A4}l>;6uPh^t|0oP--TKr*rY<SCq}w%kn^M1TCyBj>L_0;V|A
zQ|+xBXE9;Bej8wA`(%*nv*s~5)+*DB+c^>TPKqAiaa~pk_D>Om3*SavG>nkM{eSJC
zlH}4gIZmCwdVF6bs-+F}1Kk}TFBJ-^3rBH3BCYwdnWDm<j8yAMwx}cS54X>a`S(U5
zH6tip`}%fbGLVZF@4ojx<Fo#HlUQ*;-f`kns$m4FJ7U6|FUp8NyDxNIfhJR$6>fm9
z{L^jrFBcNUCH!U@JgBwmvoID(b7;Dq*R*>#zlw&u#P`SHom?0lJ%*ZSDj4cyCsxT^
zuyzm@?dMMBc14CE*+1Hu+$3&!1sN#FNt0|4y>az*@c}Yth)R^9>hsX|LWrJic8n}1
zdxG&Si189&dZtO}p#lI)s0A+}e#0gY$E=(FjIp@*8%T0|#qC&ZFsVKm#TT?F@y3M(
zS;|Y$W85nsRMB@x!;X%D-Lz9{9l&W$@Wj;R1?Vvf1`I0qufuo4tcc0HbfaZ-2%}h_
za?h?}Cd&+wXt{(CTs9f3Dd0P7D2g=mp@u0uq4H`~Zn~ptXq@l*L(nf&(EFxgj|ptT
zlcC)GU8;1p+ACE?fLSt&7ZZF{M&6-9tRigx^gB$x?SAj|ReGxOhX?=BgOXscU&GDp
zvy)|+Huv-WVg&fePZ)Jzm#N{~B?m9Ei)BK1jpwK>ozBoh^nsilvTg}NNIn#odoK7B
zVGM{<HK?2r9R}|i`b>o{S=s>n4m_=xO5en{JYRTn0(6R$_|&4t4w9Yyv&>mQrTEP5
zRZhu|ziMcpRb))Zed!+kQ;)azA6OGT(h)SfcD>u4`rq@{fBm9G4XIn11yg6|+4D6I
zKywKBbXw%Mhjt5AnK;}9wBPu5rmqIr{$eF{0~Lu|4RGBY67k4)s^*7V5u{ML{dWJ!
zG)alf1gIMcOKh?1wsMHAI5v(eNnZqQ&9WG&WlkCd_o{4IE8jups|wog(hqnR0B_yY
zPJ}X_8M3_ec2gT#^6_(HMzR64&+D~)XOQm2ha>kx+1BGC<9`B?WKHWVHHi{PVmSm@
zqZ>`URu?R@qnWk2z?OV@^5ux7s~nBUAvUx7uw6N5+%mM8W&~$QHWYaP4{T+sx&x65
zYJE>dc$<E)$EWR~1>&~3qyNvdfj(}mKGKhIZ~Tu~f5FB5pgjw`<1L^@J;8(#P(Dc=
zIXce$Vzfr$ahVLiaOm384+dz-ZkN@roE;D}@K=D{1ekMsjYs7=9=gwC=WMr<jhA4L
z5>%69pN~83US@)9$BM<{N2~&;^yM{5Sqc#acg^(ope-f%njM$g1w#@zZ~BGxrwMT+
z#MAPYhqO)c?2~_(?BSv;+n1MMcMr%ukEtcIPHlowc6HQIq0fOW#n`zC+qLGt4P?kJ
zslgX4YGxv-eg?&^r=U^BlX`wO%H>T^U!1g6-9cGi8kGApvEK{_$pVOdlb$?CmsZBy
z?MMeOa7a2ydHx^APs<-qmerBLd|($YTWeNIOg%71HjPW}10NEZTbYT~qI0T_qyeYU
zvO9ZH$iT+|#rK;6;)!+BA$u==(2V`mC~sLvE!*e9MF6O<J034(jE7teUg$~XWk47Q
z)qw1S^7GPA%|`Y$QYy2gmhxnQr6U}{Yftdx3TM`M-8s7QialW0qjJitdWXjrACAcO
z*0rXz$OfC!e~1cxyfq?qY2Ytu6Jx)lR0sTr5?3W?xV-KDN5h5s;eSXqfsvmaU);(U
zf%(=G))5b#?HPmWQfy|^n})9ff3p+s`!BKAT@HrYlz)P>VNV-wADd_Qju{j0ncnzM
zL-jv8IGLWLXi2QDypRmEyEgb<6T@nzsXPJl5kp~$958Q9RUWg$5<Ov}o=se-nY!}*
zGEgSjtam_)=`NnNWH)8*+sj{iWP)I^r}^g6u=+Nsi8LNX2JIyq>gz#ezYR|&HSNVw
zeRO#y_JTlP%~QPW7p2ml9po)^af2M-lI6j%xsW@#kn?GpyMk}FdkE`Od!)W~gLs0C
zsKnAaGA?+J4sIeMJsHtYR;Dqs3mm$gWGX1?IzJV6L~t|hHovZBye{co)vlt&NbDN6
zo4v~Mx)SC}KaO?g3T~z6CEA2?&=VPK;Zjs4Bag}#DmIvt7vg#aEhdNS3y6#i`$fP^
z=J-<qD&Pdb^&Coeu<`Ks+QMtT_RQ8VVr$97^!x6`Md&tV%RpPCq5QSCRTSGa<8*IO
ziO*NKD*#K;ji(5@cfSg}I7KiD&ouqfQJ({WaPF2xa!xn&&jgOcU`Hd_bkARyTb7rb
zMJU<y<71g&iBi~|GYEcU<NlbLaV6Z|@=LWN2kk^llwGkmZQ=<v>a>8??YWDANqT}G
z4k7G~*soPA_?X0bin%=;FQvpgRU-?Xm4%M+69a|@>~6-Jgs2B^4?EPB;b0R6;7nM!
z;J+n{xNb_;`F(-_Lu9r$=xe9p6u!S+oB&tLcP&Mcm8{W#{A{X%SEEG`=4s$2TeUlV
z`kiFuksXV6RBNg1;mv|$FbJrdsojd1Q`I1yr(M1kkcA|#e|?3NKwK+_gYfH)2hww`
zq>2mXG>RmcxyuF*U0VW9Z5EezRUPhu*+uf8D<mvv^gE>}1W4OS)j?H2#<0tYa^ST{
zHm*#u{$aky>LL@6c|LZ$T(Aw&_CAd{M6YbyQOESM??<*I`n~a7&hyPzR1J*|!S_z|
z$&=+GbmEP=<uYzY-9zS)J1lj1!O0_Ebq+g|cG7lQ=28QX@Mgn+mEenrYkLH;Cw1d8
ze^_3n`LOl{eAs$ZBv^{g6^Hcbd0{FqNkkL{^$L5Dtz&WrN0bk*MNyugG=<2z|BYAm
zd*9(v5}$z$)G!H$_~^yoH>JPlXvxaTm;YMgdTl}?3H;HTtV@sKU7;n^O@D!smQBY;
z@Ko1Rhx0rZ*^zGh?xAF0`_~9%GNb&b!8klzJq%AcvWz$1uh5>N{fTaCTvA&e-t*|1
zgi8dv#2>WL|Iu6hmnqpwkh3*2F5U)BJ449BlT|E8uASX?|I?kq$XpaRuNtW)tG2m{
zwB1Vqo+oLY-}@xcP9Pp3OZUg=VW#{H$Y9Cyqa3*_NFPPmZdXEu$*Yza`c-9p*UIJ*
z67C6n$^|s598@+MVIM$_4w(**obAiv<j5ZLICe-~<^~aYwY6BGpO?`td+I*qkzJtl
z#ltD;4GswyX%qy-tsqH3F9}<t9Z0x?z_+AVj8Tz4|BM42#rsl8dH&M&y*N7*aMx?X
z%4ljd{l-NjkrCu(5MkHeId&xB`zL8xlI9gN2xh*vGhnqhlXy$~)b4K6j=+#IQv3d&
zR%Q9Cj^`qP+cc);Va|pj({_=yUnQ?f(L+<r5ccAmM!_#Ta`K4!&}xi07_P@=kYW8w
zqabU7OKnl2QJibn)u`M~owEDO5BV<sVklV{54bB-h6430maA{#6z<70ti>EfP#p&0
zXIs2Cu8e-JDqO$2G*p}PTuPNh<|;zi+s71o6Sfl$TI>Pa@zsFWC;C!@-DA`fbEhX(
z@9x^r`lCHc*}l~DAqgwINyalX<ZI@aG6~H?jox@!(|~`UZf9_4-m5UoY347B)m_gI
z+SR&Vf-4x3d}w#2>?wL!BIH$?o;do)&ViRFcn;|b8tsNVmyRW~xG*lge3@KrZZzN+
z2_=~U_^TUAqeOeGx>Qe$98*^_qEKR!uA||qj4NLT8Gg?nOu3oRWMAl<ih&phmx42i
zq?V+&doS9L0u%;HwvSv#026h?zEI0-BR*`!m;M2qez(hpMfB~!<&>;wzBet0a|hjE
zPyF!0Qp^n7D?Y-eDAKFB{HR92BkXm9jb9-xswB#a_Q;`kA|k2WFXF}G{W^?cO`$Od
z&%g<>uk?v(v^Mk?S5L{!;2<BC8C;3xl~J|S|F5cznp8CIvfQxZ5JhosV)%K>-`1;t
z&l3Kz-otAmE+rIaj;`~FYMI`)SO{)16{}fgu)qqYK7@a#2lLmKlr+1kamfgp)JZW%
z!c%2U`ta8K5<{2SP^)uL_Qh}|`E_qqWIInAVO4honZYbk>@!5mWnZ3v=0_b<lC1@|
zo_$D&blc4>u;Nildm%!l>F>S$ThM=zs*2M>{peeul?F>5{_1k_c53MGx5l6Ui?p|Z
zit_#Tz9|6}7!Z(V2oaD5X&4$55k#aU1rd<$7*bM@5+p<#0g+}zx<k69ySo|M=Nf<c
z{NHoVbJlv-V!381&fIg~SM9yO-_Mq;&q#vUxt-XYovSPCYsbay)&TQ-e-Z%gDIWS8
z-p`hKeJOdj9?uO8YCs!vZS<JW)opW`DwD0iyJCYy%bh)vE4btb`q5YAk;NFci0e!J
z7Bm5$Hl3JOYW4l7oC3t6G%(y4RqSTH6vdHypBJ0jq81(`*Yq3!zX{q~tb}_vGik==
zwa7YNmTaG9A<c}db)Obwrj%rOy~hrQ>M}oDE1+Hi$^FGvUMA?5BfkqjL2p(tcw*j1
zfM)h?^%JTwzs|3zV#%4`z{R-xhCy95giel7S}l{@oeLubBk<Ofk1=#Ck4YF(Nmv83
z9I>Q%ZZObc2Qx@F-?))uczV6(m*;ey!(^#TZ$<Xz<L$?_2d9%&@UpHRF}{QC?Jk#@
zxr<7SqmW}Mn$BZ_#I{SsI;|FhUKoE#MAHdR0M}aGcj4SqAUEWDw3XACXu<HKbC#?q
zk*EpbwKrrY+@m$+ELa}Msu{g^CD}&_+O%Q@jdn$S1u#47LUv&rp)s%h0V}Hf4os&n
z7Z<WsE2Gq|v0Si&;s|4r>*{+B`bh$%(T=O}QI931>!b9%6l-vUef}RmUAM4ybKkgd
zX1z!hvNL)_w8|c(Z6b{&EBX7!P>U!>2UjSMH@g|UQ<nhN9js4QA<rpD(TNEE=O5p>
zM`$WOu2?V6__8!K{`qBJ3OR6b6mri;h|r116)|pM5HRVra9}Xd{l9*=ebCg7=1zA2
z-N(Ur2Cy$g;lPy^wssP$Wpup?RNJ>WY}_6h*J=ED?cf`vT4QMH+65%J&cEfMxd6Br
zW5D3CRKoAMpTE!+@x9o%?WgC%7G~5Z%gBSvqL=v^EceGZ{~!O%#m$T{I-6+&O{<F&
zt+24CXD*fn+h)&EB$8II{C{q`<P;Y0+gPIjtTp=`Y+*TWx9K+lnxbqy_p3meg&gW8
z2a2L!M-PHuXZ8GF+=H99TtRE*O%PX^n)EDxa!}R*5)3w^dB#7Fc}Q+{SkvGeu*&!E
z<6LiXT>%`MT>`DiEGN`6Bkuvvkh<q{1_E@8`{zRc_pj^eP;Y=0=^XmM|3Cl!&3fCr
zbo8^foHU;w|24{f-|N8OdeGr|tDQLop_(B-27(50SxpMDaS&Rx#oPq+#H(4H4bY(q
ze1!(Az|big0$h{xAQ@H`{Qu#$l~RIxYx?b$y4PO=n*0_Dtz)v#8di$x>K!G5B>@ao
z6WmJX-Vfa`_S&of51IZoLl$WHJc^3lt{8^L7`SDzdVt+Nrx4K%k`%VV1-!0^426g?
zO#{#2WUsTe2FOn$Ox#b70}lnP#t~Utz$5S_j7GG~Jkbl!00XdYzbAQLorYn79k0D?
z)F_nSD<YG03TGnjr)_c7mYT5famR5lc~51?fB!`4H&OFu>}&N0oj<2uJS)}}pl(g9
zz>valKnU0b*632O5V!XTExb#3ADdRas-_}N`lb!VO`&z2MXv&b&Jr{^m2rCrCY(62
zq@)1Y9Yq1hX)7lqd|fCR6|Elw>ctp9_>AlcZxMU~vAzuM39u+5M3v5;6AEo11~<Xp
zYO|TdYZ5|ZTaWJCc%AuvnYoE8@d+SH7@=A_i-TB3z-fKR%m4+v+EsQ89a@8pVsv>6
zB44{qw2AN{kB!bZYJa^`vRC}}-+y5rTm%>>h86<jjDM{RPl$bE-qVLS@y$NB$TtD9
zm;_JIjQA_-@f<P!8wG!Ayk|DRy6+4WR0{e*DmOmdB07Phi$&4^+tHp&pA^oeO&gsB
z^*K<xjDrB1|3_5P79i-%SOJKT!4@D+=$SIF=G?^qLByLHu%3ijea$Ng*#h`yXAs)`
z`7Hrk#oYUB5UAxRtV_j@N1JwY6{^VJN+BDLKmYtXWd+^%F*y;sRd{~I$J@?2{O8!D
z`9$#-3*f&Vt$iPT{C)Dr%p(4}A@zj7x&J!G3M2taGYPpD+yQ#Q5`^1vgxr|@_FA_L
zbFKZq-sC_;a)2WtO|~6~3(8P9nq%OXar98G!t@64s4;*^fWlWqXU4X<w>Bc{b*A@k
zPk@6M6V_(rh^@#beC9ma6VQri`lai;)Q}BI>271-JA|A3JQ@DuJ6J?M<NRkm{r@hC
zv)*7exPacw|9ds*LIZ1h5897Pe_T9z^y@r<gx2_~k8rgxn-fR^{C)VYSA!OBxT{xu
zMj?vVQTvE*E$*&V13+}(pb{8XfKao>#+k(a*F*o;zEDbqn$6#DX?p#!NWap?09YLx
zqud@8nT3a+V-T+gpk0qq;)vF88Y;En9-cO8%zY4;_D%p=Rukh0!1tJ=ykYe6<)9aF
z(*oJx+5DH#h7l)V;;lbj!UqtULg30bIc_Fa83j}n2)ISlrYORJCEz3B7aGgZW8JAy
z%T^A{v;?Lp=cNyt<G@yLzUTrb)-gc-jO%;s$%z6n)>blnyBLg}3BdkN-2KQ8x`)M<
z8rGC_5vjXLpN;i;-KNDj;9(yD@f7(ai#1!pFMv)e1%mNO!!wYZ8o%24b=8YXmrj7>
z(<?w0^5X#c=9VT4ofKe}y#zT}%jTfAa+DJl>^dX-Whez$l)yTUFFIuw<L90am~BsJ
zHT9f6%%QKMm{uI$hMtxIz7cA2lQ|(gdcbO%LA!L~GH#yz0}dk)P~y4-+WpqM9eW_9
z)LBK<8E4=%RtxY~IHCfz%lw7KG6ceRVmGViL)aLs^7uZlXTy%Z6jU$pa}F6TwNlyO
z?YOxKTA73MZ?^<h=M3PM9p~G4cG`OxoR5#lU;2xu6?jA%bfl7_dby8*w3goFeZ!<P
zB>38*0}AI1H7TcErr6YNfq=3yjzS}Jcv@z9>H+#;Sape2LB(<jSd)c10%r21i}Q`Y
z=0w}VK^0-^OLlTe3z@;BhJmNk1Xyaf0I0>`tC`U`z~>u{=)<i>MSDII0L0ABchHW$
zNnpmqp$f2pw?G{?*!b(X<iwAv(TDI85RO3yIV6CufhSDF84m1!Z=HiUg2q73VoWu!
z1{eMO8~^2KSLM2Z?O8jHi6@Y3@spEHTw9X9T&b1Un@z6lm#5+n&$sx@f<-Ovj(8!F
z6NIAbgq?+CYp)9;g)OytxNCL+F=jY0-*^UC=57JX0ksthTVrl%<dXzfok}Y>ryGo1
zF}<-#F#X6a9F=vn1dG}CT(~O_;J>{Ekhw9C$|3IGr7i|vq0MPjUay|<UfD-Dd9To9
zife%xsr$DoCs+zT4t(V6R^h?bVD3AF-XAA9SEAG%ALI(?W6ZXF-o)ml_?QhNLgK3Y
zaXfe0ns|h9FJL;A^(yAi4<29P=7^mw2!85;0YJ+;7ZME5YhYnl%G*f!Yjyg)p(7Z;
z`pm{V&iUt5Wei3$=qv;ua13_X;pxLH!Is(k4CGi6c&(^6r+<gy%VRID17?RDo)Mqq
zd4Wp}n?E-$E#W8nWhpa$la3fITF}6nK8BL8-CMAa!U;I_WZQNCc6gg`UJT^mq6oYU
zfPBB~=W~l;82>I<i%$tZ^O%L#;}KCXP-5LJk}3kL*(HlCEn7a~DQ@5>DknFJ641mX
zd1);N=VB3ptg+|9^WeDU6z?lSbC1(bgPv@Dj<?2wWHV7@`?99cMhF%)3OZ>u7eE$%
z$f%VoiXw#&3BG&4Hh&LG?Gbyi?84JsSxdRoD=<w%4(WbMlBnWso}<8s(#*~JLp<v&
zpNNLpi|R3sf$xP=)<!Pv-W%$9Z`8RhfJ`5ki0RD-YFf|3e!~1)64Li%vy~il2ykx`
zy(>yt2!;gieufc4=Ea)t<kcZ-cEYMPu2Xi}x$K{pFNEJr;~4P*EGf#Z^*}0XZhx;9
zpGbUqrgW}>2d4~<7hp-C4jTRd!Qu$!bx6G+>_B~po7&PXsaw@WICbY!JLy*at+^O`
zc1arO#rpseZ-N_#Ta;Fy%IW8JQ6EU3o0iXb@$wRot&D?>i`|-ipWwj#%y5>~ecekz
z(|4?!9Um9IVB#w0BxVB6OBPWZ)$}g*`L;`)g$;aSA~$`*jcU-}p2L#eZt!QtZ8rue
z0x~W>&n(-O<Zm6|b2<Le%j%V$Q`L-zEi~c2`BS<qJ-Qn(K>TV#CGi7_Y0g{F@U~__
z{hWYm=wX7RyV7lpr$0?U*~t&x#G09mAWDA0_vIYd1tm$<0R?L+RW(!C-dmOXvU@{1
zSPP3^$#amnHN|nGE?EpPFj_)`hDwy(=yVsSzk?-M?$gvHPWw)j0xboGQWgO-=f3RK
zTdEgg81mbi`LcC2Ki^JRS8@2akKTnbNJjRi3P7oM8?cbl18sb_k5bNC<7y9?bU<N}
zZS3f^%Z59Qiu1TYbHZTdGP}eFq58)h8zq{Z?l>pfso8>YQ*vHsP7<p+h?>}>J}G}a
zS#G!Ig%dOb7W894G2}hNTaA7iUyK#_$c+wbDakko_A;sXnQ6~3P^=(Mv=Wh2)A*uW
zRPn-JhSmQak_28u->Ls6o+6+7eF)43u?=6v6Q+tMbQL*Vr?ZasMB4Vf{A&x)U#|92
z5BIs=;4BWc2(5wBwbHgQOAT1nRXDvs=N(t0ee~_cX_xn*UhiFAX9Ik#a7FzcH-&8<
zq0w2tz4jKX_8JW)=;mY2oQKnUiPZ-@6~XMMXm0iqZ`uQ3?Ex=7*d12(Un2xAd~3xk
z+Yh?;iPu&t7dvZK<fiplA}6D@GdlyQ+ePW4;U1K>?O+oA(0y{-+^%0{I!5<SoUUUO
zicIMU5cN#ZSx8wJ<_$K^cMnmLJ$uzB_UMW+*Ouy`=en~bOLL8Bj(~(t(vxR;gbD#G
zfq#+_0un(kF%_Yl9~oTcG%SDpP=LMH%_KSMk|TcHR<3oof`z?emum}@G1}!w8YEVl
zT5h-36f%*y+#yR5NP46@#NYAGz6O-L5|68jxQ?6p#oL+4C<L~8C3MmHg?i-lh~h%k
zqv0`%oaqna^xe&eRLh?R={-r*J&97;fv*`QE;3XtGPE7mV?vt0=??MIO@B)1^*S!K
z%Y&mm+HTLa)73lm5%1cEIyc~0P>z^?n`c-k;`oFtj__nZ_0eSvr`FW3j4xKdUo0MH
z1*3(<@Ho9LlGJfgeL%gFJT`!8IJB57O(_7WT9_YSl<sr{&`}L}g&M~w&7nu@;gHFe
z)6vJ4tsudhVy1&oA87GfH>$vYCO_RGzaDrq^S#R=Gg5<cr1h1NOM4p%n1q-Mq`FJS
za+mC$Gc}(;bu=$iqlo3>I)T#gN<9Twx#_aV+s6*{&h|S=cE3ges_;ETUZKL)dm-5!
z9FUxJ?<Udy5dS2COh#5SZ>EP1XXrl>JYBfj#u#nBd{^0UDT>qKdoF+!+b-RH$lb5s
z;p?hvOYR)#-nf&7dpV6WwIR1id^{>Mj%BiFG*ccI=}fj4C~j}#f?3k1l2m`(IDYND
zDlO+lsPp>#rlFGP5(F+qv1hrM91$J*6DQLTzh-<&B)K;*w8l@Su&l1<9J2Q0kJJMB
z71o7uPMIg9i36v)segp03THabwQR8>Ok(=0h-Yu&)EK_pRgjF^?q*MEi8r1wiS{R}
z`NN2`^vC_)v!&!+0;d}C|DI>=%=K}bAoug<-c|~PXo{ifKA)Im6>!N`eVigZ+zx!0
z-r$5w$*SF!=85zGh~YM`lt732itQQ>EdT8!PV6?}TJi^EZ)B0|Wb=NN0Y18&b??Fs
z>2=w%W@_%XIL8L(<8s?^>6!z-_1l>x@5>VQn{<TnQ7*{XxXEaJ6JgT9*}>b)vdS=V
z2m2k{7QJF+7%c!35|&nc_YcQ<hRHzj5;L70izztf`CBT&Zn}L36}vCG%d#qZQaQf1
z*ZS48(G=S5H<BU;;wKkg;EHv$+xM!gdYRB0?v(?dNZMz@BG*qUi4=CK+uM5F<8rFi
z4=Naw^E)2#BwLRVIrf)rgRVZUUvOo@K{Fe|kM{-6Zfp_WR&R{;&`mmBmA@ZbxHD)N
zSS0vs+vVZ;>MI|X?_AAKgAWNMN&VnsH={_XzhYoKTW7*_R!^4$5{Ekj%?49P^x4;=
z^5CrxK<w<4zWo97G#xg89i)h>6_>v@@@lyn`TYAvhk`PWpL+1t$wG248pjCc_}#6z
zhWJXE+ZYU2i(Y!waZ48TL#@m7?95h6yGc79%*`3NL(c=iiJl899Ip#s{m0h*oAXr_
z5F<V!V<QVggR-E>M`l&QRi{4a4IaWns?7Fo^&dS*TE<S@V~9o+Ji2O%qWkivwpLCp
z%SWs3aZZhU&F7q7seRzIBL1D{ia*72#a`Ac-G<l&-ZQO~P><eP(l?-qF8e8E_$mem
zI<%+kIjpWZ92GDIBbdGk(qvm-ebN5OuShC@IB63%n&8$SX*1s_IUis1>Nz@cLo^y5
ziX@vYG4lR21UR2F$=+8--=mm5Cj1eDIsJ_g!M!_kk1oA_gLI8tk)CcEClp=9L$`FZ
zy)T;_zgTZ?H1|kf>lHdyQW9*-D)^4)y%_Q?meAOak<2_o*10cuD%$xNOCjGG#OCNV
z$U`EpF~OEzm>kB4>eHQ$*WH`tw+JBB$RW`ojZoMP-dzGm8h1N~JP07xHQYN&W9Ou>
z-TNp&m)k5uHlIY@BfRp7kYdiQV077IynS&HzT9*#PMcj_0ai(x8JSv%Q&~w9`D1GX
z#9L50&hGlw2x!VOh=VRl&EIm9a>d}jdwy-aMg^0qjPYjubDD`x=Xr%y1bA$MSRMKE
z4a;XF-}0RyqtoUI1RwI(T3#(H9X|aY!~n%bClX%ZL5#v2M1o!?1ek&?Lr%w6&`4c&
z)-{|Xs6lqpwOD5MM~ykly0fy1b#~1p4q)qy9<~bB@+2GEd#FDW>Qg4lCzH{0Vu%Y;
znV8n7?>$0s$`%K^$fEo0YX${AGOmJ$FEm>9DPq4wXQKDiX~w(5e8A?<WXo!R&$i0&
zkbA|5W?N>y6Jb2X5<R^k4HL3X=s)6L@XQg++&ER9wz^V(CAbXziOWG`*)5Knpt4}G
zoz(8(Go!0(0*kkXg=Oo3L&fqSP9i)lx9dx2#3l%6cvIdK+DlMPv4x-BPj&_r7Xg|D
zzg;^H<h(KGZOd`Zi#dbiM3uJJoLKi$j0(%Y?l`3!k50<OS0JelSc$3jD&o6;B&vKj
zhtb#?i=Nu{zpc<rYb~oD;Ch5qHUlAMbp=cPByn=Zg8|*XSFqv2Q<CCS-VdE1)t0Cd
z@RG_3q1vqK*5@r)sc;X)0-`XPPmUU=86SipV@*LOuC~IkV-hp1!Znj9;^yP-0G2Y;
z4hq+Kw{Np~NJ=(#Z3l$JW3~Gq3Q1h>wOmvn0`7;aEP}|qA;gS0%SZSQ9}?FL_Q7GW
z_M$1fgaDhfq$}}W9A|AQCV^1BM@L)vw4sSA#lr}ycE3=GuI$Bk<g<#}(by@s$3W6W
z^uRnO-LrnMX82q6)1*_|2DO+pIK;xOW}l#ANZdm{iDhSFNpP!A?(;H6mKmF0OVmUh
ze3dtaGJAj5+38J5D(AW5ywya83ckhmaLJ2DY+OSN^2mYLun#$^GXUE#QdV4E#5pjl
zFFipU=i>Z4%jx&IRT<9ZRMq3L=ve+_$m513N*^zh8+}^s6*r1Q7M2*AZUrJjE2<=T
zN+`VY=G7LyjjPOHarE8CB_PX8xpf9OsOIBAR&%w!k6q^M25UvU7oEpDJ#9fB6~6Ic
z_%@BL4QqB3yjsPVx>ua@Ds0MW_^mE22*KTU*$t{rCazT-2XJ{bOzLFzV^0JqPnd;g
zq^y+A8SdD_Iy$YbBvwW|f3apitV8W#hC;-EbY1%bhy`@-*nt#9*<-<}q2DP$z&zaq
zeVYW32wJ2w@{Sef?P8SdhF4U~;UKR;l-w8Tl1!#m65)l3d`Y_j@?VN#+2=UyAMAMY
z>WeuqA~p`Leux%&8<Ot`Zl_i4S6BDfT(wl3x(p9Uc)d;_3wMdAX|4IVqWVjcV}FVE
zTNo^a#(6aD3ncZPS6LK_t$ltB(r@zKH$FW)2QX;kYT`cA`l@kMlpXxWVviQb4}|T!
zXH+mix4eH$rdc-azOfq3YE8wt;WYd;_u|mLkeRllL1OhC()r9T$2&tltR%l|{N5S!
z>dI>}manMrHko<K&m$(q+mgbhgKr=7((5}Gk1Y4p2xZY+7c|*|w#uLHLujnP3K8ty
z>G0N@e2ka%ayi4KS$sP_YDsD$^;d+;he2i0ZeWyY>$u7XEKm02o$({a<M(=J9e@*R
zw*63SonYtAsQ9#MQOY*G_~s2yj8Vv0Fn|$!qhEhG%)e2@ufFG8nt!41b1A%2eemvL
zSyC@Md;pRl6NR(m6N9rOa;0irO?C_NE@u$d?FP$VDj}Qj22uriiKpQ)8+L{FaK^<h
zxpn<!lSeXoa|f>??LE%^RV689L#4t--fPzWnF?dYfQhG8d>40s*!P>u1o5XciXhi^
zr4TKKr%$e<6Crs<l}xRwe1xKuS^c@1)N!?NI@x_v0odw3HYvH}E>{u9P15IfZ|E^^
z2Mg-pmoGZ3k!SF|F9*P`IKh$1t~a-kH_Wy_7ZH7jM~8%~W4(y+7CwoIE>7xs<)7jg
z08~eU&B}PQBu!6n(Wl^pXmd(?A7zROsvdDx+`$g^I-i*N2rn)HNu3syAl=xfTU@&}
z1+I$MzY6qL{k#-^Zz4IB0pwEz4;GJyzY<0^sEV9^nb^WeHy^jWt^QrKY{C;|E#S#{
z=xJb)|E*HIW@|4g)?BYFiy$$9Z|Kof0UJ~ppPdN9KBaNH`$uGe=_DU{swEjY0#z~Y
zuCeqls&KVrBNFtBAEmN{ED4O#z2@n$9&r%eT7m_8E><i7)5%O~98)2pA1rBwFGe<U
zi7Uj+3WGVu11#GJd(ICJwg~Xk=orqFuZ=FD*X91u@yy|&-qcffAB#<Jx`(mysd7i6
z`OEKPBcm@GQe>KlX#c>gHwe=MtL<DIdVBAxH;6OXI_q)bD-^tbLD<)&xuiZ2&i>|}
z|Kmvox9r)eFt{L;vRre9(0}(+y$IBMub>g`!PAGVhtu|QY=4GL+e~KId&manyaJGB
z?rDmOm2YQgH={`BD{>)~tDUoc<lAH(_ti-s@a;6>h7-~;Of%Lk;-HgzGECaIZCjI0
zo;!GWaGP|HIM{3vL7>{ZY3NYx_mKI;sSB`Io6Gv;T+@%ihho@DWKic~?2rPJmG^K#
zLS=rByZVPH>z$)Ro85j^23{;(58ZYLVW;GhEalw>+#R!)2$Il|W$kyh*I?nFVGN2(
zTK*-?4W+Em$rC)HmcfrdE0W>Xkp5fd>G@Fv^Qi?yP^V~^z<?To(_#j-DNUW}OE<%e
zB20u8@&WCe_W7U+VF?6V^0$fW_YR7u*ee8#1ZC3@2C&F1<tN#MX8ZKH@O`{<$s{eH
zYmpIBlxRRD4qXd>jN~=_Tr5A+j-9z_MvNO)Y=E^%dVEt#a0gwOuJUf(?yO#CHd(lH
zpxMV-1xs|hqUM6Vv{zsIzHUiumoKD4qRt7n-dd;~<6efKzo>%0grIXFnKy+l+dR1Q
z{|X0CaY5fTbZ;Kjzbl)dDpk!Vl2z{1puJC!znIj=!^AyE!7G>HU6Wkh^j?KE8bX5T
zGG|s(a>aRkHcwV3L`P7tWfY4((MfgD*%Oc9%#w}m9O3gg?x^s9aYwg`90*-~zW8n!
zC1lGUN(C*_SheE_8hw=cb<d^GTypB>)69{&kKy(io7m^ZH&)Ybz*30@InU}9jYE{6
zTGyRnsr~k^A7;HjM$ZY)zJDZ@xHumJR2d;(d~5jZ>YW1JF^~Fi2H(RA-mzAiM{G;+
zRk#%7P{I00@~Vj|w_;K=2Ss-?!iS<DJI@}&BuCYcH<HhCCs;%5&+X|AJ01D(`37HU
zUPCTMRhL#RAsSDqp+nburYVvIhs&BZ)Nnj@bIODqWWnsDDa+zaOc<O7vy%~edgiuf
z)DbTt_$yhpQ;akp9zuQietTG8op5kL<J$UdfwpVjVN(bvXmbQy(RhI{|MxB#`6bHJ
z(5w$`us5w`nf-fbRPPlxvE=Mf+WvC))X>yz#Aha{tlz^(qSv&kVS33UCJr3odq5RN
zO*z^NRjI7EWtvUTS$uP~XcWFE9lMB;?Fw~}GO&-hDvt0%BV%Ps?wyqJviE&#57ELb
zS!4)rc1&ts(qCF|n+M%gg3sf<XFqU)s?Z`Iz;NQ-wfq*En{JRsPK4B(zV@&U6bGZT
zl#|9EV6N;bZ$kx`$l!O#0I}cd`HE-jHPNWs18NtYJHFfP<!hfV{5JA{=H)H&S#pvq
zrw_XKX4UdXi}&y=Mb7O<S9|KPeDk&G^JqAE#0tT$BHju^)>8$Yx3!qA2esIblE3Ia
z8+0rP4WTr5Q5o3tQdRP{OM)fLk+0`zs-JEg>+jIUMu%>u8_<ndkR3CfL34~xwQo<u
zFr|i>7PQ}aaV0P0=p42Y;#f6F7%5%N>Dd=QtB%u@pQU;j<oz;Amyop0j`Ju)#2ILn
zvOyVHd^{{viY+6D5&D4$Sz#w1i(?fjFe^Q0D3?GoA*-L={q0_-S4n6RqxG+N<_bdL
zaju=&A*PgvOlLP-??U-rDGd>iiylY9{7Y(bntmQe$IVwPAu*<}In)J&7DV4{V3gSI
zFL8Z7FDNE3_A+`kn{fvgDm`SrNpmz2(EEbVjcl1#mGrFma#Uyf1uF#kiV8_#`9!!c
z?#Iy@+k%#*uQ!?=OzHQb=daonHoZ@%$+bU^?S#HWO7(Rw!<7|9JVG$%TG;8ADM1h=
zE!cbwj~sDxyMbkZM>oL7skjRx7-&~HZBK@z*HZlm6t7II?IW&PwfH!?pN@3&OX&@>
z{tVxFveo(KR?55}=iK0DSxOxzf@Mc;?xxwBZm$8qJfrEunJR%B+K=pG)}7ZEJv3!u
z_#7JUL=@kGb2qvj*$1&HGWS`-qgDKFB^Be{DN3q7cWh`Ew;-T&0nxNZr1bpyZ6ilf
zx~K`wLTSmAZGQUKuh3G}QxCjBN$hQjxrwGu(nPznWeFVT@zrNIbC8RZQntzSp!_eL
zua@zSpMKx?p-}6k-(yKnH=pm^ATHrANhM|R^(;Yd5^t67Nq4G%Q&YBrjhk|ng>x0x
z_d)$_KOkW6K&WmSbEOrCL<RW3JT-#RpMSnKM~~UDEmty3|9C>wD&sVB-^XBDU$^Y+
zi7|+ej6^~=P;)CE%q^vepTEzWnS{0x6?^YOeDL?=wPpQSXy)F-JrrSEukyZk=y)C5
z+m!9mHi~!G@beR_P`oMc=%s>L8CU;;0_*s7JB8Dl52mEDYN(X`ydPJy;1o72W?uZu
z2SQs=;=Rt3sbZCo3yYk0s~2YptnaU()bO$lh>wC4qgoib>fg)eDP6=2C;_>Q%{*K0
z+){lLfvL$2B0UEE0H($o`<kq9`kck76bOM^p*3mFcN2P&wp$8UhM7}UbiMl9a-0LI
zwb)kZc2wq)UC#)dmNTeb>gr78-TF*rMUU5-&L4ZJtW@FDxGKBYxZsPn!fy%miZAeA
zpC$}@uYrV(IOc==nVP@n1xqa%-IGbPj4<iQ?x07bT@4s3(F(BJ3r%}Qsn#<fJyZ;(
zG9W$sc8P!#c!O%KK>7+y9=d?45RSIqFkiY-ExYy@XD#SoB^Jp?pp;rBn?dtu8j712
zck}y<*WR$>*B+wUPt9uhNgOguM3FtuWceH;zbWD*J@d#{5efWCSi~8tjiyR+uiqTi
zeaxS{FT2cw(vWoo0WzltcDfaqB_7l?PSMON1=J@j(23q1c9t~ao@_0};Flh+Gh$7W
z3IodV?J%avbgqeMg0vy@U61zq1L2pk`6BCkh@iM2jp><?y(Qm*T;KG0JDFnTs?;m*
zMMmgbDZkBt2(Dqz+%!U*`Af8hRl~F%4@b5K^j5BysRS`N7ub(nzFtn@Js?ylAZ|HU
z53fp6kQ6iH?UQqRhW~)ONG8GX>j@jgBern3(`J;1#g~i}n``1)3QS0h0?}!+cVXRu
zFwjwp>+V|H`DTqwjPSUu{xl{nDZ2nmQd92n`K+Q%E&VihOZv9m!W?%0T?PlSjIQ7(
zAgf*jVfb{sRxyAB+w~>iwTjcn4$sAYv5`Zpjr+qB&qmTUiWNt8UE9nzE=pNi&?@F6
z1k)>j-i8T_n&7xV{bSX)%Vss?Ww!eG9|g5QKJ;W`H)O-0YgIiT$8-#;J9BK%k||e;
zh#(eE{`68O(X4uRb|3X2tlNQe`Y>0FICg>kmo+oiqGvdS2m+`%&F9a)N_r8-)rX$D
z^U%GKa$Q!dCpep<q@70T!spi>Z<K}*78SUEm~(U}lmI;@7PP<Rs?fEy-GZ3mmZQ|c
z^UJXtF?@FWO0u4*+q>Z|AV>R6ENG0vOcWMi>sriKhh*Sgq`@yX7sgYEb&RYnA-h;2
z=IF2%lhQZ8QN7^edrHz7d!|}I<uv+Hzaw{8__W3cl2uMdXI2fe>lu0ok?iuM1JBCM
zR!eA$XP#O%QG?Mzu5?{lO0C>k_*!B69N;~?g;|{)twOteBZ^n~{y7hUgVAaj5H-};
z4Ql^2el2n7%F;8iAtD~aNtiU5sNOq9AA>=MjQb*p)tHw)HKSv43XqA_i)@(>NBd}K
zG)b1(#^mT~r=6;`Cr}A5564w*aiZz#4wAAmA}2ReSf8z0X7dG|2o;9clg;4q_L;Xm
zlW!py6XWNmF=_FjSI)r9T}?_gh_z*^{S;ebi$jFvhviB$`XY#?@sV@_lh|q|sDgcy
zu?%n}Oq1b>l)&bVU&h+O+RD<ompszzemkbmJorJt7hUX|F`Wt$XZ`n4Z8ONf{h-tU
zy|@&No||Lav1A*aE|%{-@YM?G(Gh!2Bow!CFwCvbxtf`6{WLqoDbaYX+(gtHL9~v0
z#Anmlc`pL}lYwdK^n$R@eY)gs-^ySYq5Xcv3pEAUu2RB5Vw`KSQOEj)C|9TzOEw=D
zQI3&AykekGsPEb<W$~k-B2jg>wC)v)MU$`gU(V8#tSUUC2*wqKZSe+$qT~xy3y7(E
z<5rcv%3r4OeG0dhRY&wP#cDBRbxFlM)W#1WjJ!|xWB$z5vdKa-N_eC>^u}>4fo+z0
zlfx_FL!GSOHM+Z}Uoybdf(-a3bvwO3C&BbSb<x)te%P}m$j#ltA0Hbb@*Ok+lbsR*
zUe`p~g~e!iBj_O)<LtEKke_v}XRV#VIvHIkNKAcf%)^7`(YQ_>QfmDd0Nv6$+krk{
zims&W-(|*ej3lu4$g$XhMAl_Np|5wZ_Qy~icp92Uwyr>gW!^^P@(#|u70pWcHQD^k
zJYy7VT@D=YY$?EeG?+-~{GM{I5-#%ywbPTYVU_RfQHzd>CxN<I27%aaeXMqmv}E$5
zugM8tl@}`lR&HcjQM1zxER0bWo=jmUSGjWL?&G><!=vY?%t6>q0489)hPQ5_YxprM
zYp5^edT_?Qt9iS;gRYh;kFVTcanUz&H51eX8%t5KuD{RFR}GVd?}IpvLpAx+a|yJG
z5id@){7lRHGZeB-Vzuv{Wug=)M`=?EB{717c>@s*2T)J#!<t>5I~dYgteiJH?UOgU
z6v9gjmj0ZRp6S#lyAduKM)lZ6Y#|K&U|-pzyet@Wk@BYIQ|g=XWtN;i6~WhjFuHxB
z0VN+cIZqHsMkZ4454Dai0I3zbzya*x)qdWcQ!S2vS29FEHZtf8v*68NJ4-z-)2xBg
zNADJEFrp*6YP#ZG96v(JKdQ_ba~5%@d>qy#F@|lKZMqZ!Gg$m`QcjdH^43e(to3w1
z?TTA=?kU25*(e#Ch^&1S*c5DFyVji4ykrIO5ckATCdxK3Fh`bYLYI?+x<r-2Ty;GR
zIq^zRa*$=39HJl<y%*FQm@BQ7-j`#+86%{~BGc^;ZPleeF4&$P5EwYjNy@HZ2sog}
zRv-(tlI=TKi>Gifx~@G~PHJpu6Aj5Xh@F0Dww>SVQO`IqFD-e%nn*dekAC|=x$s9_
zD0;V?q`X^<wmrtAM`bt*l~7Z3b;anQE}<!B2_N+0ZHWty|EZzUf{0A<{F~|=8BTd#
zGU<3oYobdRzB?-!4|$hfD#cNR$CKm52oVKZ4|iw=o?g5*MaoUYJ9ovYV}gmC3C;p;
zq_a8fTW*C`V>P41x>m6@al%c0l=?2zmDK=R|8}7|x_IsR_7|@GAw=;q6oF3`mjEe>
zhlle*tAp9f1tAU*l*=@|QG#S%_~rI>r>Pf{2S%g*MvL1t4U><wVjkPUtI8gpx5$^b
z73R3aTz6{79=Kj>d+dALVTWpY(3dpBYBYGhRby#>S|**Q>f}B_TexA{kDFFgbzTu0
z24D)=i4!nb!H(<^#6QhvmP8utrr%mEPT@#&%QY4&yG4>&X2C1@$g)4)+yi%3{`opV
z46r`T&^S$qTgiJ(ZoEQZjw>v`>8=kvM+>YlL`1!->_9+uWGK%oenlpDZ8&2OsKAPd
zU_PA+_Cz#JP1iawH}4(!oXLxS;X5IDWG9iX^krqTSN8$camqQ%)?1McOmo%mQ=)o9
zddK*;?>_!Wf`n~u$UwwNoE?iC&c+NzJPB{Z_&?YcW^oKoyM@;~OMvE3%M9)g#VMx%
zo?!I)pkWNXeLGv3Fu}Xokn@#bibkJD>X>ehIj6MXINMSKU!2N{HSCEcr-sew;ipKi
z2{XMv0>PE0u0WGAy9(Wd+LEo_3GHKtU#zVee<)$wu{89YhFy=BDV&Sy;Ul1p-_*W%
zkjoy(LeotX>FV-oIToj^*q&j!hC8eL@=IQ5%yCV2hP?Kp)e)=DrVc-^yLr?l;lr}c
zOBNdMtwq+$-vWD^<#%=Nw<0=CrAJ8Kdwd!XycP|*?~F1-S2n#cc1pH|g_0@G-kQEj
z)-kQ=I86=rSXWEoCA&v)3Ejfzbrkcf>iahJqBz2oeY)Q@kZVNe?1}46mP)IH$h#G@
z<gr!81>rXqeG0O_d{ywL_MZ<)B%iOEW~lg?I!?O?r>v|;cy1amLxb>KRr1)N9K=7{
z!V^?!#UiTAx+jV2P8FIwGi$%A>77tqnwp%;q~Kj*Z|5V30Jvb~2^ofbjfnyX*6n;L
zHC87wug(;HszNXQv>XZxm!Y>@NJLi-IOXc}ze?8O?iec2y1g8vcN+6LOMl2`6;k_n
z{2kmBp<`XqqM>c_!F6UiX`%))4)H!|T9vU<4EFk`66S;JC<~OV;oNI<e+Fsf(^#uw
zQjV=?V)OTY?MsNXtjVS=B%%9&c7R;|@{7$i_yOc^oqpY#20eTmUo&#2cF#H!I6Cvv
zCCl9k9wD^pTYH9w*TraSAE5^hNf9kRjUj9W<d}=W+T`4lj={m9vj!RpD>^*k`ajp2
z)BRX*n(w0uNEfm~c3|aityF;yuF5hbRq)vY>!UgD+p`yMci&(1^E{ShS0{1mrwC)g
z+><8A*J`Bf{|FD~LoN>z`p^Z^c?ZuEiZR^HG=|G#VH*-E(BHM=Qlio0UGd}H=Q_aW
zxpT?@4K`oJyDJgg|NL4rMhyMD^19#=#T<1bNOv6Z#hHMWBLz-#|Cdpn`cT*wo{;0X
z+qlq|PEvmaLaoCZ2nxeWiDY=|TW0;eKj^a-Z3;qzd&&j!M6R1qJ%I%^L|PP7+$|qW
zW|cr56CKf1$+U?$@rBYHoNhaHqhg~#j*f}1v7b6o+Ol_3yL+tZR9m58OUC4c>4Wjk
zpJXbO7q^BUk27q=3Ove|z<hYYlN#)E1jK(-y!3QylBqH{C!lUXf*Iu45q;aWF-q!T
zbIg$2FJ1{|@mQx1=GGZCpfX~4mtyU^wJFQxq92^hPb@~EO48Ff(x}L{%h~SPRt7##
zsvid(cYK!!<^>)MCHA40mBH)r;V%n09)6grOplPhA13sj7~O?=uS_JX=bUa{6#Ek)
z7DIe%;D9pGG2{Lc%;yQY)JxSniGx2?lIX(Ou@=Yi3@csY36rB0=352m$y@YeQi6o)
zW(1zE98jz!B^(z(Cm@|ht4CLA?_@SmXHV|;*Xc$(pg4%L$dC&=$P$H*7n<ybdKwu$
z9nus=;Mi-`isR>ZZNO(X5CW~dzUA*z-Pwdrf7>WZ8DnQkuAgoPQH{BeSJo{s;do1~
z(|G+B#{Oo#0f88DSCCOU>F8eecd9uGsKIVmCk{^%cJlR5k5%GSNT7>qv(G6-)XM$j
z0N2`OKgIP$Lutu@`n~GJJ=zZFrEvB6Zrc1}dvteAF+L&j&!`CHgPAQmzDltNGY`lx
ze!j4%<j8hzV_aOI@r7ynfAQ#&mq3TI(|gdcA2MR)bsif#d<UYXlcJ&Qq6h)m6y}L3
zV)zgR5{uj_2Dw-{Q-foidHLtr8+8{zMxD@BgZ1H(%pY>4GUodF``(UEc>8+M_%Sgn
z2K@_X9Sq^L2>BDqAY~rzRDF`&K8kM5N#VY+=U4k8JB+Xbxlh6N<+t13(~6z;MRhn!
zpNa^FTgN%}&69HCh%}bC=9Cw@)=Wlq3Ul8JVRScnsB7Digj{kRGw712n<#!Mv35t<
zwK29j3UU=(w>cK>5XjJn9+X2|_=uQU*>Zlk{pbgdF3?v>!*ZzbemK6z=^!gseigYv
zx<}GkqV@eDS<=xal$~Uw)v)cjGR0lxlx)K^Je2p9ZTbhgl5h`4N*wd_%g7#QUuWfn
zr+0^pEwR0aI@2#%TBT@(v!f+zSPodqxaM<2?tcH^5oVwCV**Eqp+j~wl+Y;$6EuH|
zAv(r{V<}jgvc+b8>EYdK`x)tV6`YZHt|I8RiU>P?4_P;{_S@>H%&q+<T07{uw4t_*
zVC^7Fj2Z+|%va0jI?Q`e5E}42i0CV|Rtd~6FtE%9!k5zSRJ{WQx?e*$v0LK`$_89c
zZyZ9U>y=w}PCvShsZh0)EsL$3bZn(2u9Q+M4BO{y>2bE@TEu;K8t8Ga_!U!_G4p#I
zKLz7hx;aqu1oWjT<E0wagJF~dl9J4W-XP82!+CF4igL%%qo{AY0;}N$lziSP^w5!b
z{5pDtMC}rX4NrCDg84hR!gpU&c7I=7bHbEEVlbE)@)n!Gh6Yp~C%8<%Xgf?$L%I|U
z>2#4~9U`nf)#qHk0u61tt`4mk0=<Jvk`)JeQ;N6`QZUzq=KP7%>(V`7N6D4#INz(e
z4$v)mhN2>jJe&%8SNG|*81-<qK#r;uk7&Ok#zl0R^u+dE1i}rcEu6<sx~eaNFw%9S
z=rVZd)bQguF@`+s-Qt4YI0w`w8uE(7*Yh;vNW(2-iWh!k`Wn0fiU`r1_~&eEC`-H>
zj{bM9@3*qT>`{HTsxs_HbhTf>mfbijMkxmR&L`z<DTY(D-n>Yfr56HBvl!bsZQI&>
zr;`1aifKqoZ(ss2DOg>kMTbYc#QMYv7sR;n@F59Gn~s5wG>X$*jhtjN*f)O?9m;eA
zuC^bcH2`X;$rpYf7YDD(Jh5$GJel0M@C(lp!akpq4qIf%IRv%Zw57DnCh>9wkrdz%
ziEaBfK9@YpXoYeLu|e$mHSMJ|&&Ee;H7X=*4kL=+!)+cdiB-nKDp>BGTHa`{u~=$i
zi`zEmJjJnV7thn&o&oIj_cw`&?T(3c-;ud3pEilVVFvjp6-T`!)KP*Aq@YU~C9a|#
zOb^+w(RvSpzI4gDSn(d%zOik^P2%{>WC;X9uYhL#(N({aeetx9r{yl>`T*cS(3pda
z{t^#p?5Q(!{TE#(P#(OU2?%T#mxl;tg?r!yBeL*OYm$J8oagbHvqguTOMhSMROgjX
z^abXZMQG!FtL#gS&)*Dm1_2_;+Y<O`+FM-1v|QU9_AK|uG~eH>jfRg}Y<_*tnquID
zS2lGmU%l!nO<BtE!n9cQ&5Dc&JD`rV<M`!sPnC09N??3X3Ww%dLg0QYAc+c9pmFc-
zluF&1_9gwd5#eG(QtI#83+jchGnR))!{XqRyD7Iyw<Mod2_aOl<T22(aVc;Q&7IQK
z#<TNjuD5GY%B8-qf6JUqeslu*?rJZ2Sy`$WYLH=cOn>9@-ui7wm{UY+_z4u+?+G8G
zFehtG&?{T}Uc2j2y})ZzfdEaj{ev58hM17L0(4ZP-pz^^d&OjpaX_BUU$*3F;`@S!
zAw{46lqAO-7}OvyNYnSJ2?uxax(^OP8ZfQ^^vj5C%=1tlw&$j&+sUygtZ~NOB@kYW
zB%~%?u(O$X3m3+y{SLZ#uc3O!*}3FN>@^ebC5HwtzvJt9@F^M-qY>4#^?FnYyzs`0
zX&~2KsID@9OWrST&y-x#!9aqX{(wpqAVBT>Gpd>STJ=0c)_Z$RsKbw!o^yaZL%Qy$
zL!29uBa^>QW}-u|I9yZAjP#;z%m+DeYso8Ch^+SZ7vZRBd&x{$VOu>U!Ozi4&{p@3
zzh?|O6y7jWV=nLBIp2M85mdAag#0vgh|gj~UQab(;?4_{ZlA+FP~|s*pnbYa++jl3
zf}+M7Z;#bT%m=0kN4CF>=wC<IoNCDKc?^tcZqkMu`YRvP6>OK)-Y{=IAkfBj1$vje
zmbwHWCq=DU_{e{F^wqu^3C1nxJ0diOD|tf&n&nZWsX!oeuk3a&W)H?m>~&5Nzq^?X
zA2W7P%jrC3>|2yE9#4l0TIyoj2@9YV!$sj^LJgYbD?||QU~(MdAhO=!%D<vvU~BBa
zy3b!EL4WVxht!J*A8%C)m==-RaLKUf?JZPK`ihb32g<q*y|-HLzq!G{J#hWR@XM6@
zMt+#(lMzqb`&rQm+%Mi(eH+%`xn{nAmLFWt%~Vf<Fm8Rv1I>Az*(Yv0deAFjqhwLl
zeCts&$s?tV?7n0lh)4pYHW5NerKZVIwOR54C;ruIfLj%C-W)w9!W*n9acM#r#;~K#
z%B%0u)XI6cSO+~S%eztCt6epX-CNvNk_5AKC#|ctEI0DM3y_K92Pc`|*#X4Z(e3$L
z4m%bCuZMEb2U&Ty3U1!R{*)qEZg?l_g~^RJAVwe6wKUys;mpc+Ow&}NL3Mf|A6)JM
zV&YRoc-&SNc25+yTozpJK3QbT{kqIKD;v0u646?th(o_(rZFJ&M0|<S6G~7C`+U>n
z%?xPKONvLP{gTL-(q!P#WCG9zZK8T8BiS`ht?Lhd0+EM#1nexw4xfyo<yjJ^?2X!H
znS><ZF)JHTWaIx#|I3GQkc4=0P*<r+SJ?*U%o2O~lZ`BWd@Va=ItI~Is9P6bG7{j0
za<TB!^df42*Qb;QXx#Vk9=R<#q`0rM8lR~oJ0)BVUXxj7=*>E)pf%}^86J9=JO@$$
z08ba;QjJnLcSNnJjqOZhT7e?hcVH?hCxUFfz2xNmW^lD1*B>Y!K}%y5`b9xcjMAjz
z8~D?y)^6Wh^Q>DzxK`f_614wgtyQ_gl{em)(d;H?JpZ&hiyB#G%g~%so35s9f7%u}
zLch2foZ>>CCEZp)6l>sh>e^!(GLAZ~4=noC^CaEh3Kv*~;<nx&pq?qrrw{jPY?HXM
zJBX;lo9L*bIc39M=Q$1#J5i{9(G=oDMHQ5^=rN8}b}Lb?&VWGau@_iX+8p-}FDoR+
zK{Kt<pB{LfB9;<mjV+t11R$}TT2UmxIhQ=zEte5!l{NJg=64NgZw91Duc-y>=4Bo1
z%oo1@%Oq>)%mfUXo6q+<zW*~P5Ok8v%t@J!wM45*UFh^!G=VUa*VCDwao=fSz#ePi
zNiX5`p?CX4sXQe0L$ASmzFdwV@uhf(Cfc2f<Hs!gaw2T%RVp!&4(-eKHZ$Q!G(TmS
zr#7DAat-$#ZpmKB(+7AZE{w%PF|U57?;g<IEF*L}d6vDmGM(I&ZNQH#BJmCunxvYk
zkM<0OZ<K;I(Hh%q){YSt^nv$)SL)!CZ!n=$hd6R_knjQhv@~uT4uA0a<RkSxMED3%
z9^K2sFTPfwv0A<uS9XKAVuDHpMI%S^bM3x1hZ=bAjTd;0g}L=XlEExCm1}oVZPM@>
zK(FM1nvQ#4{ZAl!v>11h0_vo4mNL|SsXwxb2%?Br7-7;sq{VO}2v-sAw`I6dZS!Px
zgY8feXY2@N2n`sjvU(^{YX&4ukMczkIGD@naX?q}+oe+j=pt91BQeP5*^!Q~aQ6wX
z)6->Im#-A<uLYtF)pU&kxg140bIO_U^bQN=88=HC2z?8ql##MzCee)P9D8ZxM=9>&
zz#l`Hx95us9Rf>o$L%*g_tLDwvu8XOUETEv;=p2yU{2`}_BTUtjn)(JP)e($NQ=RQ
z^3_f{_=}@F78N8tLJaXHByCkb)8ESEM6DlTmHsCbi9R)wPrBXwLB3ITXhC9~9`E`n
z(f(7s;7B6Q+cL%04xDx*n4Yv?^u98VR++Qj7B21o)ppL<Xon;m6$<>&l@6Uysme++
z$~cYZz>!TH^=J<9@Ytk<9TdcN20^`q{P(uUexmGIGKZ_vuXSh(|Eg!9Vv;IUOybOH
zzNGe-SW*&8<j(BiSaaN%<F5KOY<}{WPYHdal+n0e<E1W3Do9`|;-WcZdrxP7#>I2N
zz;om&(jd+01&z=RhdHg-h|F9dvd>icS9L>-w60=msA5~vWl`ot$q}rOH~I7J)m0{Q
z?sjmuq;qtJ&;|@-l}7rv`HJQbDN=-Xbg>KCltiG%2y3gluEJzcFhpt;INNk5`3;lq
z)A=t7;}Sv4=`Jtkn2MtvBXML6cjx?suvk8w2MSiFF%sMfLEf0B#}|V)L`ck=x4k!o
z$g1Mr(gcMX!o$ORsnlrMP55A#tV7*)?mVI0+K*dE7dt5BxP-(ocRI6Iz}99X+@*x=
z37p>B(~->l8r8ooqKZ>~`-r|RdP87DUROwqvRt|DlrCIfrq-Ex1>)ouiG4z!)wLBW
za6?9DuvNr7BWdxoE1m$o_hg52LG;WX(T#!a4!7*;7jYBbJg}cYmbN4$dK&h)-74*#
z)nb4cDR{nQy}fPqwOSD`u9Z?7UMtl|gHEIjJ`!g^a=gdxNKFc3$$DEEpsBHb+p1bw
z%6w1cIoTf)X4IK03gw!Q(lAmi{AcO-_b&S9>v~It)DMOh8kh~wssYODRq(>Dtq>cI
zm<7)z_dJ=Hy%q3C@`+WIKC1aRXXqsLXI|*<<6Zo9AfEnoz0c|4fBwtgzw*b6)xi2Z
z&QYV+BjDgTdRHmla3ilYTn|;@ndiu(T9)x-SBtj>N8mCb>1&SpYy#1i@${Te8QvIv
zXP9I0pBG+lOOJA@vtIxFN4)%(!r<S>&_7<^z>1fTTC)QU1+`I5Y$K6w0A=S5*LwZs
z^JYmW=Y}g~i!_4I^C&%(vjhkWImG6o$i?P)E|~_!fH|$w8FZW1=~+=y*(AI>3kt%e
z52Npn0f(1aCwplZq5c=)b?tBZ{3}oE{_~R!j~P)*x8`z3@&A1Pap`;{w}FnO2rQ%m
zvJMViJPMf}z;?is)sJdpsvxQSG5Q1;uf~DE+ZdSNyp3>9b<NTOEP*&Y_=`Q6o@2lE
zG!-2M^9)zrMM*bs^YJ05O^4lwbvq9q{WEd@xV~2sC<m~YNdfi$j0ku~h{08J_|&zN
zjoAfd?*zq!x@tGsi&n-d)bk@0d~dejG*r%K>^@0^p1&;H6LOUMN;X9KKR-{N8fYOE
zbfZ`s{>gQKrR9JBvx~amtB8P~|9OA^`zQW?{ib0v5m>~UNfU1U?ON``_eFsVp2+~Y
zw3lG^n%3`q&h9r2LIrl38Ai$95eulRL~&kBTL1Zl{;xOx;~G+3z$G5{D4T}-#Zq{O
zhVgF%0a4_0CLUr}(2I4ei?0`th<O5t?`?M6N2>O%QRSd}xF{Q=z>7vKD$v7$#&y=0
zRN>;+<G<Mim!Mf|2?u}^gw0G9{7o_#6qs^UqY`zt94yn`Q<!2`5Xk_V5wUecMb;8~
z@L3Z84Inm9%0E9SA^N?a$iTJZugm+-$@Kg09i_l>!D+bm<ga<`8>L{lU9mzyKZ}z6
z?gKxh+6C~ep)8w~Swp?R>{<qlhoiu?JvQnEoW*0nw?BN+-2uvtFTwHC09P<VXdV}w
zAXhfif@n*CmwFg@L-VtYkHC~oO+Yh(%9f%i9W+ius@<YD>%8`(Q$~R&+_)EZp7%IO
zgnxV#%ugMm^r~T$;1{(vv_1rVSBg<ebY+m+Xf4xr`fMG`m^6rDDM&xFL;?VWA<9-A
z20*}>rWVcF-p<w;jJ3Nlf3J%F)r-|PpcJlGKaj>ygYY%`iv{r45M_+02PRkRbKt&s
zjq+*A>(48gh1itOVV`fBN?4$H1D0=@Vv9y<kH3K?oT!ORs-cw@_4OuKbzhIjM&8L|
zic2tS%RraK9u)ZwMFo+`^#Ab+IOgtGTYIj`Kwp3QN(xqNU##8zAfMPez}PdJMgKLm
z4-_GM<IG07Se34JO&Z6Pz>0Yd4!o+A*@Yz?_sRe~Vw^Sm)4-^}*MgyL`%H&3WZf^C
zvU5~CWc4R)VFL6%75M)*3TA;Z`okr7(_)v}gCs_-J;wVjxNlI+rOzi<V5Tn9Zl+B)
zXF*QG+SROU4*0zhD=FSEFRz2d8N8hrxbNk|C`lWooG(@tBuc=jQIoF6d;EyVOlFqL
z`}9$8(Qe+`c;}*B(5JQ+25O$i99ZlID*zW0x`l#f1Q2qsc;;v=pD2v)7q`)tf=ac%
zbp4Cxwxz~YG+A<KFaL%c{1>BIG9PSCd8FA4e;9@S*ob`U!Cmq^2X;gw;QO!{jHy*3
zJKgW(Y9px!J~0E#pCEr^9gGg-@G}$C4qg4Dob-m60UENY&KNRG`iV4h@-i(Y$I>ZM
zM2fgVev(Mngt6kV<6y<?|LIEVgMlG?HuxSVlg6Jb$~f`K`OGg`j658uM<1|O)5|y>
zq2LIy+OH^3Bs-&fKT!E7p5fQ(O01uEXPtqCi6Z$h*yhXU*pRj7sN{0+*L5*0LKN^o
z55-NW*>)-{wej50Io@r8On^CI;`&C%^Jp05e)uK`XbkGA_pLUVp^H0U&hqTP#~d7>
z_1JAV{y0*7*M`#Zt*OZVZm7WpLl$W=Afi}-aYo?0h;HMYXj$^G(-oLw>&8$z0qJr;
z=Nviq{)O~D27fmN#4h|3K(%4nFGptOww+hxXHK8+`Vizt%B;Yg4P)n#yK@d0RLr8w
z@y3)BFZNOR22^6CRX`&|x*ine6~<t@^_bX$sExV$+I9_c`WRRwir1?5BlJ5lBLYDB
z8L#lYZp7JRPw<8lTfis$);;ihJIcudWHm|qCO8*aR{8)95v3^`;cC@{IS($%1m%Y5
z2Ak&Nl_$M`)F_5(shn8m4HQ!dpcD9B9Cn|V)PW}tq<qSNMfSE}i~DzT7I|BsZ5Ra{
zj*-!3$8J>q=c0A_$#e`O?HUjbSfKnj3!`3|wR>SU{*M`xKMAg5m^c13EM;L_p8^gT
z3y{;9PmS}Zz<3tV?Rg@0&_#S~9mfB&2z88-bc<np`Fs?$oNnHA+K2u+QONXVElRK_
z(@~tN_d?tN5)bhI8TR$x)oF3IY%Lw7yM`S*fkN2$C2|J3wF?5|>k-zY*5kH<$E9RZ
zs)bgFPuU4aRRpoYnRUQ7+;;iKqy7TeS>7_8vshH;mGm5Trko9BKX>=q%uaHsOQ%&z
z;APsRMyjuiAon32vzR<vd?u5e&#t3;gGRjpLG|-)AWxw`x;I}|IRf=OR6TEico{13
z_m*}fP9NUQs0DbW|HIf<fJK?^Z%Zgc$si)>z<|;zE#0Vew+PY=f}}JI(n^CM-6<k1
zsidTUbc1wve9!FJ-Q#|H_WaLv&9&F;FpSLmKJ~kQcL3zdAiUlpCdVc<?DY}uI{y)*
zlaOY#7^(1E2Syx|w|g;1Wvy<gxb1aI#dBLH>H+JJvT)v4prXffZ{@byPymZ&)fLmH
z35Uamqk7iOJjsY!d&4BJ`U<7=wJiPFn1u8BGER(5T%^|c{P`7FTDg$xZ{o|Z_1NnF
z7JmN6W5gub?@RghA8P+T&jOiIZfG6>B#gMqYkYj_4r&8W-7M}QDC*g7gb`^P_&^CL
z1!}-jdShW84D2{hunpnB;my@Afnhbf=4FuZIow|`xmUGB_DFMoR7yhnu9GU{Hu1U8
zdTEQ{yEQ5|an2*KtxJ$XSJ)62!=D<Q8V~jYU+p!C2$MsZp-@7qL`qzQ$3skrAMmx{
zbqC%BR%LB>J{;983*uv-?E&`1XFqBc7*%2(7S17cU9ym;TQnNl8RVf18dVhlf_SsL
zCjdBEAfI}_vgxk<)Faw=x1D@tmExc($|LI9=2_4O3*E&ZhqasB1~MPHsxW<kM&<?5
z6SZm&^+?{)x+js1$VIGyD-d=%_F~t|YhMKh5yfYw@}I3iihjh`19KgJ7{}rIzzj+J
z-GX3lW)R2x0q@8KU{@3Bh%cz<Wrk5fG7z+VL$v#cY3k&?SV@@>nKvFBiw_PUQIkNI
zK(Y~K2_F^94QLEFJ`QU~gHIrL3MmISIdN*#rzhvj!nW0+P*Q4%=A?PBVHpzXyF4Su
zq8-DMl@~f$fGXUXUon1HHRFB!M&Q9a=8awCDaqY`ltgC_>^gkK7pEhir=W;%G9H$A
zH9Z29SyLcEDvcWILb9huFLvWT<D{{QJy~Tv%SZj`?S5w=64rTJl*(EdJCGi&vyx%o
z(h@=71Dz`+r;9_5J?H&?>18?QIIX%bg0ld(MVw(!8H4ukb)Q=&66>z3c{!EOl^R4o
zQDwXuGdmgMSbKSw3E3!sUZ<vO3mjOgwr#{X^L>x2MB@8w7)g{1++cLmD9U)J^3tq#
z6wI#!l-jVF*fXj*?p3FPN>P%?rQM-Mq$l!5&04XNga@0q$Q{v(IAC%GXXM_*6{t0M
ziGQ`@f&O$LsJ_wJjspH!?Px(iCCrQm@uk|<;6pVLm4hFK*u#M)&+D~F%~D|F8&jfI
z?@v>11v~4&H0Dzq2F?%~`mBd0+p}i>(mhB@nMO3%VW2ys)9jO{r^?c(W(F_ErOxs+
z1R~&tm$P_Y#Eac~_bT^}=OZGfqcOx}H?wq?KbCk}kFsqh{2wz+@78v>`k$R0AAI>4
zFe>CW&6!UOJRbt1o58mVLJ3G&5zP^AEWmvn6nM>BaCu}|U@|OU$Rc%!e%vDpzjyTJ
z?IHNR1>bbu-Ztru0L%`I%(w3Ck69cZZSGh$p0J!gVR`-Jk{EafL{I2wfeTmr2G5t4
z;(d6P$9v(E8JmY-&%gkX>3$O4+>Yqpf>_D|YVon7sCZ^qoA=2B>o_6JRNZrAgUG{e
z{KgX^BMA@S^`^e-a{*idI$dTK^=;m`E6cJl$q31$&ctigKp#}cB`p~GBdrWnzpyh}
z0Xq|iCL@O^WorJRyQ}TAz|dlWDl(VIH7A600Q5nky|`mp$PtdZ$8i%z`VM$@_An>1
zVU~Lcw*zF4$zI+*YkJoXMSKW_sNfz^oMZ~p?3_*mMB>)G%m%omy2K=-zHmwHd!}kO
zrVVh|?Wl)I!OX~p7{`*Zhu?&_4ty0lEL#w3@X&U^f2=!wowKQ_zFSi8MVZDtvF6~n
zyge3+M=TS;PDB}8;m@*0IB!$Kd#bIgb9t+0xT=rp7n;M{lduD*n8Dk?X7iO`BP(37
zJ0rxrS^b&i#9Ko@T4T`^y`ejY;rrD@`JG*F<%v|UvmTY-Sq$!JHHi`?;t67QWR1v+
z+3==47`?9D5g=i2Js^_w($=Cw@TX+LGqM7}H>LU7!R44xn#q)_@XjAt|J6oG5wQn)
zzI>)U#KXJ2?3=wXx0zczA#R5E4t?xZj`_-`Fy}WZB=)j*nasCNE*D`c8CER7D65L1
zdWA&~c!zYfH9b4i<xJKsOw_$M99{o#tlIrT7TH!I*ofav3-1@o#bc2PIh_$3&t8l5
zs#0Vhh}h_Cs!x+jnuWOom8fL>{mnEiV6taFVd*L~r|;|8XtP{bb<DNf>>2D4BVN)g
z^~0uCdU%0CzSS`f`&!Es(z&KXxDw&s5zlVqOMjTOYvTTj#SSl_tXFcoDa!p`ugfN1
z$+MDnj9&K&olW7*BKB&<56NMj(~}iSR4Y09&ptofqA-`xFTKCOhrw&~X>$I#1OD?N
z#}vlDbjJQ(J=V}R5{>+O!ciw*1O6xsZ;>1k|Ef&#i4Z-U<_8Gf`{?`hcH;j2V8@2}
zRkqFO`xwsr5G=WWWJBclqwS(_S=0;t{^kLTeUG7?f*1VT?74c{w1;Y2h^DXTrMh9t
zE+3+t7p6pd!h&&JGxB}`*3nrz_sP8;6D3OFZqh{3ptsS)o}5F834GlSPv~xrrHBNz
zfuDjN3F`M`)g$ld2?Awb+|Wrf?R7C-|7&xMPfs2TRg`s$iTKAso<nhNqord=<ZtGc
zR8^PhYq(N6Q};l$yPC-cUi@agFs95$*(w<=b%{Sf81RbxiTfVCg*e1_6pCWb*VtZ9
zlbxk&psiNIo3eI16xkv8DMlf}^CCNqoW7Z$HSD}8pl)#&9s%)~NYXv2Xz)7YjBg^f
z#~Y#B>pl#B=rfbcgm^A&Px(a>ig2kSm*GJrZ|#D-$VYWTRbm6*mzVNhwg7&KdcZ4d
zIkjY~HL01mZI*~ue7cx5?w#t9iWk5p&)+-5Ve{ULw}CZdX0$&^JB?ONf}waxeh@d{
z&yAy`UsmkEnr^eqEj_hb%GK%zoPD&}Z7K<5(3~jBJZd9TG0UNm?(AZ$V=3-#8=9(P
zo1`0u_Mg2cfxs1-K3x88Mwhr|WMBr|#`U8gkbeGV-J=&H1|1(DKOK7W<eKgMKSlG1
zeKu@A!0KmbqK|WUKj|>ts1KI~f7758Ah#mUCPJ(n@Pa$)$7oW5AQ0K=2<lB|g|{E}
z9hGQ}GmQQ|uKxYXdkQ2xA+E6eJ3P_IM%j3c4V5C*q=-Ftj6uug#g4fE#`+h>IE6<<
zhM6e**kn*UIC0)?c?`xyukh75idZvw+wso4cy6>8Q5?_%5<k`7QimW6@O$|o#U5Bg
zkE5Wl&)e73Z$c`zSDgX$5EPkSCwwIp>D><5A@e)`U?IZvb8t}V5c>2aI-?^e^=5YU
zw{n6M2qm;U78pk;-!b;F5W4dWWs0FI?c1)%aXHd!VO3g9q|ci!lw18Kz!|I{6x)U0
zStu<e$Ok&d?@d*hdAUFeupqdzuj4<FuRa!M*Z22!!Ebw_#mLe&zosoMBQc(5;n!wK
z34(J+KxLskLGH`Yxe8n}$SA3pV2GI;5&ENk`87z7HOX^is)7qQBZZ&K9_YGTo5!7%
zakagO9jV1Xeje2id~o{JJ;Wy^szN^v%8TWg=Y9ZdSaF&_woR-S1hGx7q!SZal$@Go
zwmfPhVd0Y6<?EJe!6TuwxJKl)Z0ZZnu?+SCPQLD_`i#Mjxh?iz?-Fs7lmu>|!ccb%
zapoOXz^H&z>YsL=Ec)TpwcN43vDoEVx-Q8z!szd~A>c3&i*!WTexdzM`umprINkEF
zR>`)>6VJxIr-zXta?7uL<J!|lKQc^?(fn6%tuLjKeEsmI=`XwY<3^0?9*o4E`ljE^
zxVVW}Dmjj}jW(UK?33|?lCUeRkb;<__Tf?ek8`lyZo!%ag|#~)jIce>T}e>*VZK@R
zWkN;6Uc+cIM-_zm3XORv)|D}r>?cK#D*!<umSb9cwRh8mrJ?JQ_p|^hvMIYX2c!PH
zxzVlr++<U*TQ}2~#fl^KTkleWDa=txeeeW`Un-IO#7f}Xc0DBfT^7ZrefI>l{<d9&
zQC9tjFA7z)kgnX9Lf<p1N1Kc5549cpqfVlWm-GX>+wPa?f=IG%!JGu_u*3&jDH+}M
z&wc;~tu@7IUr^%VA_a@hM-cRo=+4rWWT<88TJ^rg<FWjnxo1+b()lbbDP15!V7GnF
zgxYq<Hua9Z$UF0geO^^O_Z&+eEv4jl*Pn51WZ$eM?wUwa&TMJ33cJYc$w}r;eaJgj
zhCM@irkTfx+a_a9BotG2vY&YOW6@moMWKrHgi*r9eva@*DnEy0a&c16(>WQVsjb1B
zBy1&L$NoD@7j273cy}{zNcZ0J&7O)X#?_A@uX=LMxQPq`i|p|HmsF|}L!^3uB}39|
zdRzA1Zp14qx$q2&pT4Bypjz$8!7dt`tk`e|u94NeUAhx@xl7DVe+Ideen8q!jU|M*
zJa$j=at2W^<7SRnCg&wJTFOT^d1tOU+M3%>Gk^M-Pd@zpTea|s>gEzvMQQ&8HfF^q
zFl`2O>#SLB-ZgLxG4Jd}7rOqbnwQVYAG@|k@^JW({%iY#a`}4aGQ0DS7JjX3h60Xh
zMicWPU%&5Iae^lK%(CxHHq&+?Fd%Y{Lp$^~#Ueue!trD(Wi7^?iF3}9NEc!A^thR*
zTQG~%x>OIaxNb@70eVxAJ6CVgdDKlJ-Y*>Yxd-cWms%>2l%IcotoQ&7Pf%o#{`Z=Y
zEg3QaLU5O0exW`lsShjGG6na#IpDgYY|B$)Hw3-FDFK)LnX3Um-suqoI$&=W-I^)_
zbSGe~a(##NWH(QE{*(I)j-whQp|<ksoF=Hjosu=!xK+?y=Tcz3#O*KI=DFx}<6?4N
z6A}F-j7(!n@X5Ip;zcXIw7(`g3jSyOOL*~%2t+0D0!&g!0FD|j5SsWQ(QVilK&`+O
z@yg#DB9q6zd0^J(;AAV)-;nFBX2^IXMkG7w_{O@tgC^`a{NXCy*+5*wbD@nFh+b0u
z%UWeqQiln|I+Vh3l7s|`N+nYRJ{_Z;$$SFLsh*s1EAkLZ0(U=66E@KSWZQCX3Tc~$
z-WMiwSu6I=3C~nY@ZzAv%X(9fNkTH3$MBi|n`kX9&gA)@--j3W3SuWA9_8)r^|Xl!
zY1)K(M&mpUJGkROoG*Smn6gf9g%j7Qvfue<0{+>mNkV}4gk!#ys9MUZ+498MM2p?1
z@um|JF7aUU771~`v+<=K?Ijbsg!gBFff74e@enOKY%zr-7wisfaSKJpBRzL3$DicR
z|HzexQ*w6+MpyPceg)JTZ8;P+?p8V0(K4&K8jG!w6hc@tS_so@q<h4w)(P*$oqrw$
zV1TTIgMl!)GtmFXKv;(e0OLqFTl3pd?s2O@r_*hq=|{yLMG}l!lDZ9^4dp7$b}Z>k
zpQ;b0U0|V{^+>B2VhQgyy5zu6uyaV$nT}_gpb;|XQpEl%SS3<`*x>lqZlei<0w=;5
z;^pzU$xxMf1ix)o&Pp9eYhmw0>IL}PLSa^H;vNBtuWJ?tc><x=q9{|)x?oeM_^88L
zFfasmWDfvH7u1(F=#c$NE%>X4cH1_0Q2QdVYJ0u14O>x>3;ZfsCq-DVy}0&0A&`8z
z`!?!a8eSQ;VXZm}9M7pc=h{s?*Aw2<r2!-q4$VDPopN<t+yz}MtTJj?UndM@57Xsl
z3%3FFvG_#L4580G*jB@F8*89OlZVImavdENLg}gDutwUDk!`vVnF75=NBZ?`^q|K4
z-*0p>OrUNj4s9bhpLy}ZiFwWBM+meu+&SFt&dvl#fNwrywnk4v6|YkMnujr(SWYxd
z)paiGZ7ofy0HPPp_i%MOJzaAJKf{|ncROgULAq?)bEi7}uOF1&iHWSs{1~_5;{dEU
z1q3Jlh{hzw>#r}zr1)~e-PG#1o$-~}og+-|uU;2>p2CQ+H=rFV{To&I4@my+xDO%v
zIkwd=o0@Fnb(@ZQFzJsbJ}m221r;p}zRm`aBZc++wz<+<JQms`9IlxvX$nuylME?`
z2ye8*MO>DK;Jv-AIY3ptJ>Rvq=}9*rdHxPPwUfL0ea2$iCO-YikY`lH6W02QpS;H>
z6-6Bwm2a_W5)Ti~TH1Iza`*}br(O+y-#iB`586?cr@0wO$_iOftqa4+KBA*u?*36<
z?8LmH(Np`joHwDo9u_lmOW6l4H~E>W<IBy-gjbm&0Pf3O5A0AW?84>(ksXDOHqth^
zvRA<Tn)a|7qZZ|?W>Tbn_ovxg?3p(TJmfiN`)@iVA9l16GG`QU;GR&i+<PR{W^>8c
z61%CzULDVQayy>61d}!u#<Ln8a-0~R;b~ORx$)^I<&`KoPy`t7=Dm6w;*;b&2*Z20
z@*arqvo}sKzSa`6omAv@<Y0bFN{YQ0P0U6-7DoC&F6c`=(JAmc9V%(^^{B*VxqMtu
zRH^OItJjeeXZzA@eeJg;BAG7LSE!n;a%Xuiz0Oi5&gsL8GTrDfcl34M$&1Q8q_uZP
zJs!=`&E%?)k61$`3`sfyLc+j>X<&4nYRQbFbt`>PG?fRrw`}@eRBcrem&OgwC%;M!
z)7_7sP}%cd`@Af;^$2-n8+2aBbq-z*0Vn6qoageJP8C%J=I_pYyoRDCo>&eZstZ4~
z&PblNt!K!d1r(=8Ecx;*1M-_%50zNY=Ytngs0XFUtfO^j^aOPOj}fMKs}U{a7nv4K
z;|#lB&!#o=$qP?w$L9A13d}qrg+YqgY0CW1a8Z4bhJNgQ>r)dqbZLLvV+yG$&~7b~
z)exe-I&Pu3X;1pooDdqzW)KmXLoNi`x<Qf6IaJJB7N=qHH)hposUnaBOMeHPl-iMG
zVqV+VR`Hkd02#CsTgNCtvKh1XLFSDtWDM#Izm|&ymXzR<;@uT#2*Sl^r;SKqvyJrv
zSRUthniwT1)FuFfA)hZ>`Gl#jxvP1KpjTA6y5?Ci10Y!{VZu-C`)pa!CMWb~H!yK?
z6V=uJ##6ZpmM1)=O%x1rHj&n^uOxCIO1T+Ds2k+{w|qZ4*hQE>a~uy2q)$TX=~}%P
z_f95Dit=8+wf=1nL3ro`<E)}|+07ni2{k5u%*o8OV(!Gj02gfcmQ?p598r@gp6IR`
ztwRvo_vsyVv#(CA!xCRh4$h_~MAL(&@$#-iJG{TxdCGM;R&4BHC+0J9=K(=9nlZzI
z85l5q%$D0cuxW%}2$6lKM_(YZM#4)V8cqQtb1>&SZrZ6~2y^U2*OK>z8d;5b#>H@f
zhv#~hdu`Y<iT{q4L73c<7A}9q<>^{)DR<IfR@Q%O(04yZ5*%V4tGI$t1hB!xTVnzg
zhaecmKtSW2{TC9s6lZAiEIu}#kV~{Ev5)CzLj%fWQM=Sl_LCRZ9XUf*>r&8{pr?=t
zY^-TdZu|TMZCP@*+)E??ErEbQpp7Rle8Tqju)dN5f8CCpSGXnIYGSpo_IySdZWM-i
zxs=285Ph%5A1SDbL?70S)U&`T8QgD)+r0n{p#X1)I<Q<F@}_I&Z|0Fu27bNHglMo*
zXeed7gzOHmZ?3JXb?^iY2s4Nd6b*;wL1E}~8moPJ<rYV(scSpOaTphl${rTu?L=og
zSen0CF%JS3DB0KJp%FBJ>+sF@vPaWiSJ2z#@_H<Rue$tpsit~(rUiBz-LzE2>J9j(
z9+;a7wb`>{&WiTr^y9z@r_#0~)eiHe^u(b7v>dAjwT%Zr7k{=e_&Df@ZWZ0I-`QvF
zd(@3p4%bDz2m_aoouBv0)y6I28f*@mOo7M;(<eoNn#8w;I-Hlv|DFq7W_{z0w0__c
z%1>!wg(oId$50G<pXLDRmNPyDW_CF{^Ivn2FgJR>+DS(`>nfL11(Q@Oc7nnnNo8w}
z9!2C0ZX^JVW~-gxu1^&?0qFTRta4bEMRXR90W*nmj0n|e6AoJJ4sUrw4*l->aio0N
z_Sd{wpAG*yXiL_!Q`(A6eMpwnOEHDgWnJXrad*wBT!J{ItL%Gy$i?Dm>|VbuKP{_(
z^i0{e9An#2E30q?o+LBViBF|Y6KIc~{Fv%*a0wJ)lOees;GIdpJ_B52Z^LFhI%leS
zW9QpXewub%1UZKiLnGphQSQR<Kx;X0dsI3JF1Og%`=Q?!^iiv9DP<S+)p52bK{q!s
zC1&OZYwG2vpF*+LxakrC4pat9uY!pLQycg;fD^G2jllS`5to|`K|pHv`2edoZ^CQp
zu<}#e*Lw@WyYx3wa;h*KgwOhdgl-=cRXzPb*1dQ5@qR6JEn5Hf%R?vFMtQ!Mr_z(t
z%t2AeLj=^55gbG{@z~1U^+oZ3FN^)WyUk_E@Sfw2no78*$h17);_bbk4|)XK;_(iX
z@R+`yu=ukRq5jtEXDJSL+3OH$wBBrL^LRhvu`OJ@k&f?ZA6m^*$lD#jwS-o?G!+<n
z0J?-pA)mT)czgO=%cni6q)CuI?n!gqK<4kb_fvxG%!5JAQ8f0X;$4_^VO1OpwSrJQ
z?CaT3O65v=h}5NSp)rg91ER!ucUhz-S*Pyk=nElXgUyuAZj6SWHsz)Pj7I7!7FBj4
zB6I~J#{@hNvbwSBM^dIN>K5b=kjzZD$o!F=zWt8*5H^uI>7oh_ODo$^v?<dhwMuFG
z*)v{~Zequ#km~nT;d^;#8(?8&KU%sIq@)2@wPucRr6K!H7eEtWJF1A!2sRd86irGe
zH%A-79193}ygsC&V68i!`BuIhkdUWyDOl*(i3Z}aAEbjIzlsPz#pWA$LiRpv>ETaQ
zh$CEfp(Vh0DTS{mApK;hGt2#PsI%;(z3qEs+cyc@i7}#w>`bl&6osZNv9sdohyg*r
zFI-&*ag~y1FUy`ZyeN7?H0*C}A7Q>EHdg96DmcdM<o^&T490%|ndM#UPpfk*V(die
zfjyZ5rB0)at5l1#sA9-Q@&?jTz=7ebPkq~F^~%UAV9@==snk-Jf`2)EskAyeBjf}1
zr33lJ7-{EVeMMr(*TnQ}U60qhrSolfH=Q_kTey#$%~&|GYj5xz(V1-Gv#mZnym^NO
zE2J)p;z+_|+vF+x{@Xx9fs2ZMLjfQur$ZUvFsu~au-rM<A6_%;_^3BzT)`n$;7)WV
zv>O!>kKL?w7kX;FJgS+hKCUZk$zm2^irE>ozfC@qqT7>`_X$ui*sSh;CES<<<&Zgx
zFV`4xFFFf)1<;BWM#+!e=%uRQ9}B@&oY!vSCB-<ZZmypNt*6kM5YGkeQPVM=qnHV2
znG!E_{IjmO^~E3U!j)`9e1Xi~BuDqHFc+m}U;y7EtTVSARkNZ!jILP|d+dnkx}?9~
zEj&~TRG9ZRgIK1xj<Cpj-71t)Mkc7Oq7&TTg)P4|gsoa+TnsU$@aq-A1YAt0rUkB&
z?V2&8R5$ko;&wd(>h0ML$i>~i++_R@tcG7`JTK${Y3G~PPnq9qW8z_SVyIM~PgMYl
zQ7aS91Qo=-M1#G3RRs}986Z0BU;rg8qeL482HXnb?Y;(>qIl^%tpTkXIEV~nVs4fC
zRJfuI7z!LmtporSmWB91n-*taxum|ubz#QBCkD_HkVC+Y9^e<?K)ScHKp}*<fj<pL
zrPQx%@<s8uv54;vuPDSy$Wo9q1OCW(q=t!1eJ@Ic4ndDoSONoQ4(|@s72a%L639*$
z&f8PZj_W*DIvSxE;uc_L(MG6Y^_U=eG(yt|>AbjVdUHrG)s4>|&sC?o8!;lN8pwN}
zLj3L|*xYw}H4jx_LUJW?2@kU9g<9`GSRF)M0yO>BpwKUrEA}N?Cy|;|&LZbNlu$Ts
zv#39XkeVUFNJH~k05qGbk^$Q(iE5Xkg5RGI$(d71g~~bKI(>v@*oM^VkD1k4D#-)B
z{}Y7A*xiw}22jtv54MXCK5`~vD&;$ex0A~fCumh^u<hmbggIosqaG9GlqR8`Va7r@
zCRR@Yt22wu@2u<hozVGbVS8c%k|Z@m9l3+XJG$W+Pj&p>i9eUD;m2%qTZ1)s*IRLq
z-9)uTe<Gx3idZh?+lF5)pfyob<vXm<;<i-*Z-q9Ob_{1fZ{86>>N_v-I^6V_9woa^
ze*|p7n~CzPcZ~-a7HTF#iqdRqkLpT$5V|K}F5%19&@Y(Eo@$k9R=z~6nI(<dp*EZQ
zKjou@E=;|i>`fz=p3?vNh8t65jL}zWjJ`6#IlR5Q{%%lT-FHE*ksh0ruy%f&@;)ek
z#QRj!Yp6%e=(eMeCE|WeQEzt73AJ`7#|3z^NmI(aq9d520VxhvcPZ{hZ}I6C6mvh;
ztsiFe9|O&OY$Z<AL&78pN(!DS7o%teA@91ulsvM3H#`1b$AMSg5rF=^qxwVT-~0Dc
z=qNXWNbZ@GzC$snE17Y^=zW0m5DX(?&^vE{h=behV#HSY7%J|8h^kV*Nv?4R=N1Lp
z1RvD&`+KLU=_f+OvP*UHVrry(5dt3~gas&@@M+EiDSm!I)1md*GkCjq^U00#lMGqJ
zhqoe$+bI&-yd~>CMQ?dchS?!hP~S?1tvU*JG!KXOOwYbPz;KCKBV5!I2;@Z+FzhSC
zhO|0Mi0M8X>ZEVJ9VUme^-blA6=$?2%-^qX?I<U<{eIGBT|H@kbNMByn;hD5OjqV=
zqS~UWlWxg4puQNAXO;|d$MnvV`};><-w4^8eMDxf7O`7Rj(5b=nfA&!DPi7BBL#?U
ztWq5k$Cx`5Y>XDmGaHm`gCn>Tx)Sz{&c@4DxNz0XH?qQ(TuM1075J?sm17pON<~(z
zyJb!+rZ>9Eaq~Ozes&)1+E@ek#bl+*Len#Jg%o-^+C%*Ob-hZt_2m?`!{nq5p)*W1
z`|~TogY%rCMQ8c3YM4`!Tklg^XTQA%y%XQY1<43WB10y=X!f44j(2(pclBD6>)b{?
z9on&4eK_<($~{kG2)?U}BkA6Hn9r#^{g6J%u<jQ^^50<_pC9i4So6UhSMl=S!a=~P
zG(g8U1KvKOT8hd+-q&E`@eX&=Zq4P*y{Hhk<CCOE<?t7|3v$ZdDC$Z4S9+#_CX&Fe
zdE4(6{I95ZA0}X>@~3g0-EC1V9Jl@K31Y{F3O4G9zrA`9Js<TLNu1Vpc>Igz;_u6F
zH#w3?OTJl?|JUk4BnUlT+W~NPDuF6+I0eL;#pk~Of`rfyaoQI2U{&e5Axk#-Z0MoY
z9r!k|5&QvyjIBWZQUNL{lRF(T<?OY$sz8p*I-p5~(kG_6E|KT}W>d>1yV@pb!dS15
zmVIC(UDWxn-~2r;_=4k&{VadQOnnj*ynpJ0ph{E_uWBKC@%S~g6?zA#g7^VMW(^#q
zL#I!9+;6=PuRGuE;s!i}e3tGMHG|mB17!SH;fHVK8v3qTQ6*M;NIq8!NB*A$@h^MQ
zua_zsQwE|*lm44P_80V~CA<;9D9=Hpl`uvo5HI}3FZOD5wnienK2eXavF5Ti92<%s
z>vn$y%on05l#ua?KcY<j<EjB(1ov9;;D2NV|Ltv3Y)GE6-7NjjfBN|U>vj11)ql7M
zzh034_&n4A45jdyW|e`zOoPAuk^k=L{r^7_6Z!u4>Gi+80=zLnW?WN1PUe4Sg#X9C
z{I?hVz4Ot2y*eDBjX*cgob?<T42M&j&BAF(0HBK+L10M)^lo&>d=S%(Zibi0Ut+_7
z;9%79lMk@rycT=(Ul4w%4-L3V+_Qanzl>?0w0mrB`({XH-v&^k!8*I>HsSZ|1|HG=
z0D<3$#ifAQH#S;klvLT1O9AF%B}mk10&u-41^`4tYYGKHz?!|De^HwqKpK94*{C;>
z2@FVlv?3cpm1_m!X|<?Le@h7bDQ9bO4wGke8sA7Tei<SgBu@JJXa6rsMdYPuyfWI7
zoheQn7QXa<{@PAWA}d-IVmwS)(|R8s%t=}s_?+QB$ET34p7^xLH?EyMqlgE$ZRlF$
zie207jAhrZ&Qm@+t%Fi`*|Yb@+J@)!X;Ck&d#Yez{qRdc%VW*YN3Js7=>LAaXMnzk
zmNHbJc8Q?g``059UTQZhmcV11vcLX=!!eAO5f25DNr|poY4roE=yt0X#@!c@LnTi^
zNbyVK^HD%3r0*5jlg+jOjovyOQcfA9S6cVSa(S^z40M*3K-{+>EHyH~c|m_T$8Yj!
zg;`WEXw1!GbnSso-y6VC*JwDN!v-8oCtuGXl9*HxnOcS`jgy7^#|B?QNgjZ^4Q{%E
zEO{7vn$2X5+AV2>J->EYC+W)qZIbjw`lV#cNkg+8ty`2R|M2iJgeXEZahQdk{P7hP
zXVGuq!I&JEzs39<%+WahbnHW@@Ln?H&MxTkbs!n#%|e%e!`lhic61<1kiezb=&|$+
z7>u!7pRa`8K1$-Ye%|PfLWAhgWwN_fBWP29pic7eMeaq*aasBR7_$r@f2f-2X~Tq&
zD<HCWA@Hf6{~XA=jar;5h47%Q(331*qQ|^<>lj3PECR>&Xkfzol<m+1C`FV@^y^zq
z8P5#I?D_RC!Pm@K%*8TB#OB017CVKcwV6Kp_5S_!`D?fc+<?8}m<coT+HVh$0D_gt
z+D~<dW0t$8YbC#b(-{ze6oHuocKHIr!yUk2R@fFqaMirrYG`h#&G#?^5K7dmi_;h#
z2yu;I!^NQv!2&X<t{k*d)kqHki%R|y(CL&Rnva;FW5o=4q~}&#_mHGlnN%Uq{0b;r
z_{ULu{>xWt#4iiPNal1q&yp-Bci+r8{yxI~9BRKjX}bl+p{Ain${&OCy%?sQV};mG
z@tXt{O4YI+{lfRpjji#pP#A~5b#{6he=I0p?Kz84Fi^XVsxEY}s)t(aRPgewZ2uqs
z5ycoxFb-mLf{dyEHB@}4FzY~0w(oYd?dOZi0R1ABoN!K?3C$X(H5xLjQXjBy@RWo8
z*mBPSAUUQYGmalv4l|tqi}M9w-5v$>6k-5hH2wL#AP!6j;hjm)4NzU`O|?BjcHXoA
zX|;&Fq~SE5NxMRZl+%n7Y~3N`UgtaN;Vv}C!=MGy%0~+dsR)o8N$UVOgI_2<Oat;{
zE~o93z)s3%O|66N>7%dU=qnBwl6<bf>OHg<+(uTpB<>hw=DA`7nQfgCFm{0WL^9CF
zXh)don_E*QYS1r3M)*MZqovN(B{1NI=cK->DFk$&OXWHCBE6>UJFNOlPdyJ}biqfW
zj$3`tSNrw}DFtcD_Rw1N@sxk}@VfwZ^M_eH;d4qLWqAXb$5L*G>-0#tUw+D*Rf2>Z
z1eFX6D+ZEsS+u`8+n`0tOAgiwN*c6Hz9rLtwD|r8G$pzP_5o00ap=kx55i}&al-+Q
zu{P3RRr&$gCW(Sy<EiyO0?C_JUbWput^wZgaU+1q`Eny@_6tDUg~N@0`5Zv*QD7r)
zE(W*g@D9T%ppVCbRal)(AKn{?ONJ$;N%87<wtS;<d}$$SODqr2iNvJO07ZkJ`>?6t
z8IZ+xyBp`ozia(u)UXv;bvdXYY;1)+&n}kYE50^dl;w4GX|lIM9j~&RrMz19yyI5&
z$x>3!6S{^VA*>`XKqu&CWI83UbDbOC82_Yj{z9g<8H1q!w$qvQPWW-2bOjFFiYoCb
z{c^l8G6~jkr8xkb8OL^G?4xyI+r*D*M^iHFdDo!Q0<)C%Ex;hJI5`JdbtP7S%3Yb=
z<njS@Ai7>dQ3EycB-T`??D<dS>E~?_d`#9P=V$%p(dFs~CkEE%Kp4@9)RxcRUcY^D
z-Ez`-W30m)1?|>|Q`$%kox<3w&9zqQKj!?uX8y1LZR`~H7Hf>jVb&t}ZHe?zOKz;P
zpI-o)|7fsQX;WHR{3y=~$ki=1@Hq$Li^jnhvI<rvg?k4}Nw$xt8$9KBuiFzy$U&u^
z4U&;<BLyo=`^!EUu=u^p?b(*=WUZr2LEeyMppZ+zguq24V0%kF9QHjxd`;)HHr&&R
zW(-s;A7l~x;8S&~lejEn4??d{W(NTedBJOmobdy!zZuzMb@)PJ8sB(*P8fngSuA=&
zDJy<KIRFGoh4<LyRPdgLS8&hnJ#zGiILtKqR6Yv#piI%Jd6D;{6)=<a*w!kXoj?p|
zLKW5_p6xRz`~6@ocwM4^tAl9NAlRFjXp`-H1pwrS3ZDlk6*JTs@Nshv!*F>f`o_rM
z4)6i&7^zU=XLGvy%==Ferw0562wQPBe)mJG1P;@1D}ZUx)V0-v^jfXl$I)m<l+72H
zXIs`DE+D$|Irw4MZG2y91eKz61>X8TxLdswtZSKdzHcHoeAut~%A`_mwv4hlEXa6)
z*_lR--UH#knL>y6r2|;DrVg9TyxzgOAedhQ@}~QMt5lfmx|Cr2T4JZnFu)sHja1Ro
z`06)!R0pfFbEjLaTtSmwtdA-I;>wr=ij(~HCw1OiqR(GPuG103zbA)EHc!<d&H)YS
z<8G_)+iV!LNIV`3j(e&)UTGbMfxZIu$mpT0Y{GX$7i0aZLRXu=?k2GoV{$12E^^?%
zE^_PfaGE&A$!6`ARbjvkX(gmq;z|>2(pI`H0{9B>_0HwsYP1V9uc2$T{sx4hKp>(=
zrqUn8uCDsrYzDf9@|41hX{%}@^=nzQJx?onY$eX+MTI3+D&7)F;Ut@sQL?X|!KHsZ
zxuqSXnq7Ym&pw@gL7PWHI3hM|`5%{X5mZKelnV-k?4R%Wp6qQC%};p6H|g<$LDiwg
zenaSVg*pdSAct(uW`Ct$UeZ%u@a+EZK9E1W1lenERnnpcd*pJ+Hsr6n9^xIT)-4aY
z$EvU=A(fpDyNdm(gShfJPwsu%`9j8j5IH{-5}{%rG}z;jqpvujrlGfJF<kf)z|qmk
zM@p{_0*#FYk7xYe*%XfPb;7QW!0NPAx5cXJEC^zV93#sf&Um9u00?nLRd#-;EA1oZ
zsw-dv@)B|qvmP|AJnbSM97ZF_`WP(xOgSaDv^&p$YB0RU$L2u2Ogn)u=Od$?|HXTS
zMsBt37<%cU;=ZS`%sR231Q^5|7#=t>oI8RoH(6yX&SY?f`uGb_>nu7?I%(w0(dSHc
z>@KNv;<Eh~hP)_E(E##p-(e{iMq~&VT<f=bBxtUkpxU^ky1p{RufOEjfV+*u3k=Dq
z9r0)x&bun2+$>%&yy&iZI*dn%$xql7a}ZyK)ek#{j!KV1$vYYRUT=irPpK$Xe5CzS
zusK;<3&PXr2eVg7R`vRkX>9V6>GFavW3_k$-uu=&jB}@a1~qHpMmXP6eD@K%zTre*
zSenr4MfWpVrst{cjB;XpLm>q!Y423*%fmEgLR>}a_4y8#EDoB9Q=<ol#Ouuxm<w{W
za|{_UbbQ0+$B8+VZW2a~pJz1R|1ju6B6hIXSsYZNF?K>NC+I;kFy`M&+oN=Sg4;xR
zx1G9DXnPHFUxBcw1_Ye8ei0tE5(p?|T)-r2b-Q7-3b?x6EM=8qTHAVr*<)i*0=^GR
z_j!ruK-YXSAH=+z2yf8otRgeK`9$Z&)vGbCfL`gg1Nney`K!LBB`!N3Jp<1muDio8
z8K{mTOvBTLYuec)gnMs?(*8L<|NexL5apbUpWu(>T7--CT0aC`$PWY0c{S}`4)>^s
zYKfz~AnoEr%XQrQ7N|nFUsr}KHNSB{vGRzR2}n1A{{+jU)-3(DLY!5(&O8j!Ah~n1
zQ;^#%r<sv*BO}i7+jsuxLAzwWkuJwHh9R`{Pi_aRe7IU>hCcUD@W$qWP>o0bkys=2
zK*o|_aaOOBqwRH;*eKx``o%n0(O9$E<y)+iD4*2pJv@kBrlbp@+2@1A4N<51d5X6a
zKFH&tV-)tkPCCW^_5g&-JR7l>ppX>%F@%=!!(gdtT7mMh+$~h8)Y^qNPZ(q&E1%z1
z3VI0j8<nNU-hDaIuPIW9jW1o7YIyEDBO^GJ(1)>XoMmXGBOaL44xPW-D!#LoaTo3;
z1o$6vlo&qFlJ{|ati*Qi;S)%dLI=b$bKudAuc7%cIy5T+-<{mBc|W?AXnK`(p;N<#
z=r6_jj|CNjIQOL6-9%h;U&}36rYaX375nG1rt}F@Hv9INX-*w)bZ=ESXQf<w=LV|~
ze$m%gm*8RV`yiKAnP6Q^CZzn{PN!R*KLvak?NKVTbGPhU>0CZ0Ig;~Q`0(x3i{2&}
z8+RLl|4}Lb`7`;nXp<tc{8S<aimj+$cig`|SAB{K&ZyIsMjgMO-6_SdQPqO#a6;e|
z6day6$)a}9)0E#VlS8g%oUl?FORE!DMROg15-LFk0&|<|uq_6XsUkix`m6QQXDzx#
z+8@t*TaEIPGSsqQq$xinxS;g6A5RTJ{VDoy=fljcsI*i1mfH|n+Z#~l7%58s2aS8G
z;ItzbA25A#%fA9ixu_OwU+kiNH5o$3scSY|gyb_G-?I@Xz3lU^*TP!^2kk3LR5+yo
z1x;A6m%ySddQMu(BtVZc;s|;}$~HjTvk{=wZcB0u>>s9CE#e>Lz#aG89GUlRT(|VV
zx$>sd(02clF+OUJnJ?6GxKN9v%zR%z*G+^sBkk$kFGZQ?X^j5n+rBZZSwEM-*Dzm>
zP&|+^A~+eN-Ee|P%-&THf0<-=$MJsa%Waor)~qr{gclrD;<L;x>Qh@W@+nZ9W8tF)
zYU3T3OGU)M(PjCAB)hhHLyDvkTme_y;n1)a@p}TOr^0XeK$z35<(U=ttYW7q>D*T^
zX1w12Qbei4P`MNdFDO=i*<|9cW$%F_uJVf(?IHdIGtCG6Y+31P9+Hyx(qV_s!O8S3
z=y>a>=0>Xx0%Ao6*m#-sYF^;u#w9qhUK(}0x&Odsf|rEPAu~58zi^yEz5Iy6eVtv2
zRY&XS91r)zY-{yP-sfn?bx^CBjS4H_`STrN>E$vUD3*kz&rzM_h-K7Cs+WbWER}s!
z+OPiYI=N!soe2yaB*4v~t9i!4P)^7373!Ro=4A|S?yUm|9qEnAZd)k|MU+A_O&K4(
zOQ5?(Om*@}PpgO`%w)8b@ND>wp44A+S5`<$ha;G6Q+t7El2OoYA4(XbO>*J1uVAMh
z;c)rpu4!)yGikFjCugbI%1ojArL^m5uGcUH*Ykq?wGjoN%w1$96v#G@@E{@5xq%zN
z!r)O^CUxE9r2L~PTkPS+xCUqD%c*6%4C3+4(XxzOP0b;9qi^x#SCBC+aqZV`(gcLc
zUu|<Pi%=Pr;8kW2mwv@|ZZET3SM!OgvRqz;hfb@{_heoq?^58Z;f?O3ONK~@bDkAC
zhOavHEJZl>3i2$&STf5B@|7bh_M~&&GRuNnztHcbV<}1sUIqsSai3}6Sq(o`m0;{i
z;)w&+3&DXNM7WbF#{rzqT&}y%l@vOa&E)2V$3A;`k?nn_exEBX?ZtP%KIBH3N7)vh
zV7?3C3d*AQ^WteuM75m7B%ncDDuw1=>l>bIr*M1(@D0hUZ7J1OW(Mtp)E>Urnqf()
zv{P64G#8&gO4+{_=}7R!Ad6B2KJ&Ogv=IIGeE@-}&Ur7dJAp&Gv_BZv{9GOHVY<HY
z^AP4r=z->aP)~F<gB3qhqeQ<l?<T%X8oGT#pg4z8H^`SxI50!>s6E$QT{2V4!#Ez!
zQ~4^PqvhRsLai#e%?Ek+xGdP=&d1EJukoMEqQTc0!bBkFlnAkE1;F}yur>isc|Mxg
zQDYV9qY=Jhjl$aKVj>7`kaON9%(BGi2i*oHPcfiigZC0#c~ujS2a1pxeQ={;il?>7
z57PKRAG(3G6-(%i+x2$idJBjeItJ7f9@{>_PvELjBydV$a6po*Z(T&(5&updX3#al
zBPU$Ka3I6)<)4ocA8SF-AUFWylC97M?P2bSkG9kLiCmU6s~0^I8c@R~ZupxbIymL!
z3JU7-b?iIw&`L$mA(gD$uSPmA18$=R>f*I_DZmZPy()aUd;&vb1F(t4YTQ6H8YjFV
zP~p=-Szsd3!A;#XvBt;?0UM%oxj!(VU$BSY_h`inBtCF49?A7Pl;p#N3Jo(O=Jy#9
zbfIh60`0QbfuKeRBkce?4mw$*P|rJl4a(O(?c(C^PrI&1Z1WNm;RZsjW;3#|PzoLk
zX2mNh{pke|??Y*Pd46)8WT}=UAJ%^!@-cX&Hmx@V+?4Tu-4sux#4Q>a706K@(lK(G
z546^8OoA%eSWtWO>^5p_SiYtb_7?%%`<BgjA(OlJ5kh_0NcS-%#fD|OdT?WwKLLXx
zHBg7@({&ZP6MvqE>3x97oS;pyva1_RH3~z5XE#=nVeyR|KY+$QdES*-RgggkqN%nT
z<$MWI!?S&yHR|uh(}ORz3ZS-xxMv8I*%-npw%M3lUXk7oP2x1CCX$#aWj4N_7OH56
zGu__37<fc;`{o9IDb+Aq!#S8Pm-Q=Ub*}!2E$P#T%Zto@UP2p*sOfg2tsokiXRku9
z*4g^QlhB^+{m?QM@uN<#{pme}p@<(j_6AD-1AYg<8lM<{@5_i6G=IFq$3or!xTtOR
zOH=gdU?0pa15>Y!7X*`lldY%KdjTYg@-6FpgeVUuTf1vCx+4ID((-e+-JH0-t^^r5
zuO3*3w-Sa_8kp0~i$&a<DgbM3Q<h4TugZ>=AzxPwF>)78FMN<)U7}ClBhSfcS6z<z
zOoQRL+}6Tu#=(>0GV=A832^4x>f;W06UfDN)&R^Iumsij1qoR+Pt-gj4>kdWVKl<U
zX*YZT(1>0nf~GaHJrC5lrORp_v5vh@`y!`}50$P8h4bTBa$sT?y$>SH$Ekm!wZ)1G
z?o<P4g}1abjt!PC&1RIH2HO*it=12<M(WIpO?ndL1#g*5m4p)DKK%)>%yrdk@yD)&
zpb|oRh+DM5DQowQBW>0Gcs8iE(i%V`6!q8i8Kgq>J0B_-Y5Z%q{q6Xt6{b9gXV<6L
zCtttAcTk!bkG%vf<Wq&jyHF4fj>ZJ4zhu~SAng>gy{t-u#l3Sg`%s<>EuHz$Hih#M
zCUa_2W@M5mAlCEDymMw+?nCx5xFYY|jGb0N2!q}`)@MRP!m%AM$BW}?hTFQ}r>;%S
zJ2ZbBeE)hzLIRc|A8EWl=bBRe0p#W`p$}Px4Q@Zb9^kQ>*3F#{hlb?fQ%j95Wd}Bd
zPIsP$;kq8+7?VwF@$5?{_Ix+f|E(axW4*q1%TvfrpKx^tFy(l0@t21_yj3ZY?4L7!
zSHMI463C|?9<0A{x+^R;L|r7eaGRkoz_3;M_&KC@K+)6W4B8U=hH~8xk{4-)gX#l@
z=w#lJNDiFw$%*XLOR-=&fiB>cK)s^_y<-D#96}a!*>TvT09N&(xd3l7R$S55p*<t@
zO0!;Iz4KLK!0!gG1GF_J=5?7h?&mcC1I}>XF7%tpRYWZTAYyeN_5EqTsuvdrBdm1y
zzVc>2(~`p>zET7m8NSuYgRDO`HiTQ|N6cxMc{m95PFLw|aa~qKl!k*8R4hVN!}`Gw
zrV5N-LrFMx)iKf~&!etO1g8afADbZ#M3iSk?s5z+;9J~$qH^P^I2Gf`Nd{pB>zG!3
z=IOdaIO&~HYak^z#zGk^1P)$%$-aB2|9poUAOD6VWz%cn|BfUjT$5nlPC;vWJmE-)
z&a&;F4d2&&_t0qq?9&d4U`5xUx#xe)wmlMl)AzNRuSO!^w6r>HjD6?krIn3`f+Y2`
zb&jAO1*=FgS;YAqW^Xd!INE?;dpX+Lz3d6ynoj_2z9CP*N7ee#*fj+Tj@gIK>np%z
zKH3RSIA2VVGdY`#+bTX7g{vYBSV4`HcVeQnfacL8%zgb?I=f3ia8AX6$SZ)0>Srsz
zvP1}2T2_3z!cgEYC2(=u9kKKGYy|*_F9JKgV7N$!-(@2Ni}Tqy6di}0x6f&Ff=?g)
z47Z;KJ!wT4nrc=x+U;065Aerfuun~yBSbLR^*FD({qCpCj5pXeC7BK}rX4{bB@X?z
zVx_sWHXk~Jdt27XB%VT3o4UwubzXt_QJW?nRD(?HtXmKr^5uTCnz8b>-?@da^~J^6
zQExSjxxjP%=ql4X0%|OBHR^SB(L)BSqZQ-Dz5(4svPErx@f{aRWFW&!&|`3|ICf4M
zu7p1r=!;qsylvFff+w6je}ZGgCn3VDT~T_Otb({k#8#Xk+???PSn@|xBU<CJo3NIM
za3kLBwN)H=hcyO*Kn)Q_v~L-ep?HQjTbxGk3s`#s=$mh`5uB0Ke%l;mbnxtCEeYDG
zt$mM4*RFLFDcRqxBWylnAHGJ)GHRvKQkv)~@wGtF`8w;{<8J`8{b>WN_kvz+5iHT-
zVy;u>woLfE@;|xXV6)R3=Y~@TgRq;#CQ7&S6%IXcIW!-+O>9KUW3#SZY4y{jH{~5Z
z!l+xj$}X)34n;96Y<5GLS8v#1=t1H`N3t&W{#YRX-ty5rfeeYn?0ac*{8^8w8lf^6
z;^X2m)I4G)-Bjib%~)Cbn7b{lDoe$%PC)2@aAS`G7)FRq>GF{Jb&B+qGKn_&ba;;h
z4~JyEU$9Mc$m>7}9!E(Y{zc$saGNj(@!FIL&*Kv?O*v#HlzCL|cCeoE`msd4%nyzy
zu}o0_lIz-Sb{))?ljP}J+#givrN`JT@RlF43|Y(|D^{NZsJh~iV{C2?OP2=EqA~E(
z?Q;FhFeJ{iDcM5U33f1AhMNS0n!z@(5Ukhe2YrgZ>NbR<2&}I4S9Mfb!!64M%`_hR
zSui!#`$$8=;WabiiVQ539`y@3+$K7Pz+1S&7I9+ZF|ReQ*Qr&&0RYQIrmW<1fIFGX
zGY(@j@HRWPc$!(5VYAYetCS(p<`Ro>im@xopI6m5HBwrW-psCquRmOC{)$o<cad>L
zXC;DE=sPdG7sz-tK+-Q9bOg$&JA*g@mZ^jicECY6of!55_X6+BwF86u;E+wUU(^;t
zgw2~PqNeCY*cZQ%?m6dqdp_7ruzR$mS4EKWbs!NB%Oz<zD3PK-Pff^;-cjGe-JiiP
z(4Cu1ghEnlXg)4CvgJ!ZSPP^{xvl6{ab<4`?kkc#;kbTvaWBmJtWGhGUL{-8{n<pz
z=+CnNtjg=o!DSMXZs~LKO7KcAu2KAhFULhNV@XK`b<i_0@GiRo^u4mn*S1eVQ^sU(
zCn6^g!wTQ=<|%ZEb)PZxisAU1#}bgwzeRdXEj=K7H>R1dzh;%vDX@R`D9<@y1RUoE
zzEPO3qTGaAp`K(uil;3dRxFDNV<avypdc<`^RmcwZW_HY-WFdxh4$eKm<d^PVy?i-
zQ}TK)_pp&(fD>+~x6td#=H#?yrhaI%Nzy95rfMcLkXKogqHAqsYvDD*2ExaGFE`K(
z0<~2Hz{xosKHsynA7LsXnQlT>s#O>TxZ0{nhn44m#bJV1+LO7%u%XRkE!_Kn?eg#Q
z;m=nz^FR}{N|}q6QT(?PaB2Q6MOM!Av?p?!P5`^blz0NYnk8+$51@rx2WBQzAs$rs
z&#kYaYm+7_BAZVH0U{9Jqq6?sFc(B5rA5n^T7}W%5ycFe^z9MKhwI~?N({uwF_RXc
zH^dZI-n#XbIh;)QQ+G!63~PNbNsz&$%0S{Bt$fyF6=g??H$7-m08}&N7}cFB{4~X5
z_tE%0Dfiqs;)1U%wd+)?qVhve-sO(OyAZCw%sbdl2q?u&)yR_`IsrWgD1yH8v3(Mz
zwZ<st$aX%>Ltk(4`p|?G%w}pt31zg1yNi!yU0s7<qtjA%0*TM<Qw($`=xBOsqq;ba
z#}+t#wLcHG1p*SUvT3LdvU~5`0M2J2Ze<_#svVcLr<?E&+)SGZYtycLG*PxhUSRGJ
zwuY|K_pg2iEiH(I=gLc&9ii9Gs$0Q_p@NEEyoUDPuLIkjV1<}5SX$={fa}vl(SyC9
zqAYqfLtmIf^`P%W+moUPMx8l@N8u7FP;@Z?)LCwcGU!XO$e>Ce-aDw7+8-?>pW-@%
zKQsze#tSwXfeNPC5Pzn0XnOfwCpEa(K7}<#8*l%S2S?(ZGg^`=tKIf<+Xzuj#~~%@
zA1<Vxc*|L1F}*qYL)07RN*02R@Mt-g>?~NBexnPA)2b?2!BA3Ok$CBimA*8;Fg;K+
zbab=!?=WQQiB2qb?E*TcHbq-igpeOQ(V|Z~g|TF_K*RxFYem|dRIk^1sk3MBFWsET
zGyOm`3b|MjeV_2q;0Mfhc8vL#Mi=6Eg=pItth|OysT`&^g)DuWJm%FBKeFt~5BBZ|
z{Rwma{j}HE`3TvS`k<%9`1@HR^cr{mEGG1da=(Aumt@NuswsLNB)7kN&ni|7#}8Ku
z*U!XI45GM=M>c*eB{1u1HE`;0I_FE5n(B4KxH-_Z0sYy{y3O}E&@b;{$3YKfG}ZI5
z4n%6-!*cmM=hw<MYu2zX@0W#+6hymGMnrPsv#PjyiPz24Q^X#~zv=O>WMuPcym5hc
zh?T}<g;}X5>;7KeOl2WRz@Q6C8<Qj0H3$71VJl)bL(s5%y|Uju*Kdoclw=4W^Wy#{
z%I%Y;@#L*(I^orbx{NKF10~_g<K5d2)05jdo-%wC6pMy4pohldaqFZw%UX`_*t#Y8
z>cM^Pj@_1?F7AIRmf}a-aJ^Q9&j4+za>u&+?jpUPsqe$+Pq0$atHKhMp;&udPeLOf
z1!C!EB`3;x&Wn2mj}K%xyrPxydMTsjj?W(pHpA-W49Ug48mks0ooD7%Uex77YMVdx
zNgI?9j~IYllwtP%GG23*J!4sSCMQ9tVK)<31RdFsa1Xh<1&P3sq9xY3eaO=%GZPWt
zHIJI*Gxp51<_Xuaxb;`HLCJok`(O9nYUJ|b@R8Q?U(3s%)wSXc9}ls{2m15Yf7EXO
zym^DgDz_Y+4!y8kuEqDy{rg{3232$v(E9gf*!wS=|DTIpt}GI8rSW`~^j8SRAOI=?
z`3l$%y}Q7TvJ3cJ8qJb})1%q6%}{QXY%(u9ZRQCUJ+ZLZKcfVHelrnN6!4viuTg#Y
z^WXi~liV@%WJd?!sMB$NvUh!1hRb58)Cims76AM0&3AR3o>zOqe{T!?hgTFQk;Iir
z?a;;hf4yjQ!Hqyt!zz5Sukx+dval?0C{KlU`627?tdGCXwO?NmaYC|0oXqKC-~RRL
z{2iwHhgWWaKP&!PK-|A+ME~%{|NMscHAv1;O~vi5NB`4h{`-}|1cOA@yZOoAfAT-Q
zqNs<wu6o+97yp;R^#9#m<p27hgYiWpGJF4j8q<IOfLs>jC}&}Ab9nRXo%!FNgRzH&
zLhye3)<2YLB5hElZ5r=#COD-M?x8F_{%=>hfD(L~Mw66j%&)MxItwbW%8UYy0TTu;
zz>w4Ox;WtY?5ezjj%02iEv}vee%3DF%suW&<jMk^1VCL9M54CjAox`|Ln`bwYdgbX
zKoqrfC!jakezXB31Efo70Wpet9C|*j4@gPJ`XV>&Rba}jY4P7KWv({zdTDPo4g6(<
zi}a&Y3XySHL?ih#<pxctT7X<ORHXB1#X~)Y2)JF~Rs()T51Crp!xC^RcL42v?C^kr
za^tS&eeXpd0Pt4}um?EJ`b0oFF~JF_&3@xzEq29qO;kJnpb4Oib=d-P=r_P@O!V?%
z%c~3cWSLqk8mfDQq079#<^#aS&N|n<OaPHzPhq$MXD5pUD0H6ozwCP89*Yoz(Gxh^
zuKSni@qPeu4sa_P-deyoGTxqPT8g@vvJ(FWvpbRNNgxgh)#uAo%OxM=$=aX(U~QyV
zcig}acLDgK-Xh1&Dl_`p32f1*JWuzds3Qahb~TN{xMUZ6Z2=3HwH%05AjRwf_qRdh
za|e>tQ}2Eh3!dwN=;4G?3IiK3ZKSR_Vx#u}7kxaj%wL#r<oAS4{`+c3DF&x3`s)Iq
zehU?(<OBuxLla}g8qdS9W{HU<RRH&35s4*(`qzX7$uTbnKF#rsOpffc3kd4)$Ps;a
zIV5`sV9n3MXL|X*mNsHei0O|3=8x>?Be;P~es>7^CFTF)?5v}rUfZ?}2r_`cAdDb6
zgouE&APobE(nu@a3J8cO-5r98N+TiN-5pAUG)m{tF?4+Q>}T)&#QVIyYkmK)_S!QJ
ztohw>#c`e&M$z5&D&T~69uaz{7qAnT(G^1gGh&tDaQpb%0}U<*;tdwU#1H?v&m`_K
zgVS&|JW!?n`~+13?P^;p22Or*poXyk%tF}?G{oqFF#Nr|C3vW=f0Kd(x~<Q`DdzXT
z2nXyZwZ{3JKTji-acpSk%emHIEL*jQ01+#7F4$c>0W%YC3;eY6_0MR%=TspCl<em=
z!W=rq(l6?d&G$M73yq`%k(sFrpvGCY9kPAmQ#+f7L)8c{q&}C?%f`LE<-Up}Cr`i=
zL;?O>Pv0G-fnNXU<qTkNazPN;c3mq9R)D$?p<I9g3xFua-i7l|JdEXg#_+4=&1Zml
z>4=_H?9ohvo#G5kb{7E4SyY?{7TRFxGqrJG#n8Pr8ZI!!vOR~hqx@LzX#lXWdj#G{
z8zbbH4SYXwefsleCaWE-R;L>p;C^h;2$3kxuRZgi2vG|4Fhp~t1T;@U$#?u^*WJ9$
z1mQQmYhSGa2#NIo@+`zJG7j=JgmF_bmhU?<V2|Hs;u~;J)Uvv5fS$GT9gR(Rb^?EP
zcK6eRdz0JbKt$?}-q5@UHGpUA0ss`YnL=Fs@e&pA>Q(r(k?PraV}2F%-adk!s(=F;
zy!@Q6S7YvRO1h6zdV7K>F>FR7AMiUjcFVwyZ66k=5=W2K;j!qA?=~BwZ#8)cu%!>6
z!xLTSOA53*pE+VIQg5#m0DIbuao|>ZYA|kH(vxdTUw|%KcQRtTfjlhN6A+S<H{O%*
zk5$^6td8Wxp_B_JLD?oXN$*x~C=6=~Wb`8yn7H%o?OI{>JJ_7QAV-ccP(o3ew1AgB
zuO@xFW)o0Pp*Y_WK|U6&4C(=BwrAk7n_9kT1G=Z2j)AF^f@lO@`(r1pOK?AvtDb%~
z?IP+dA9>8%J1CRHPfP-dXwH=5EE>RW@3xI!pt9n6V`420vFfIy2%L?137-BZ{W=#j
z|AO95u6|;-El~ZB0uH-VwGG|`)<p9bHLo|U<9h|(r0c2qm2W^F77{g8o`REuK-*BU
zc>b;M6IW`xM$Gc|Z3e?`gXj))cYbKC#By9}O!|(A_&V_(X*(qMS>qf~A%@?$>!3fP
zjr*G7L*!@Wl_bC(y2R*fo$uXGp$n%H=p6}xu!WUfK@@BOKfae>R_~%7V}Nc>x?YmZ
zT8qQ)ZUG27Ux?(3A^-Ez%QeJ;-jH<K1!PYbI1g;5GU5B%c*gf3dtiiy1?gSMuz;Rb
zOCJgd)*b79K<JM6Y)6mn`>$Zk1p_s4Pb<d#L{MO0J_@Oi*qdz$jN`h7g#jU?zYdl}
zpY;bfUqXH^di+);P)3Uc2D>WAkvyutW)BmGBLI1qZE+gZ>w@#%tdC(D_<8(jg+hyT
zLH!X2$DdeqKQbN^>X=`EVH3u8hG?AKKtXdbSBJML;0Tc9i;Y)^yIJLfc)rwvYG|5&
zzm&(_r{$$C5+YZ3u0>0TU|~La&w`617l+%Ni10%r$#~L{V@w@yRIl2h39Rnd{jVS4
zF^EDqn+Sab#wf65oXB5eO!`QW&E1aQ0(1!0rhaQ6Kj<Gs1z)AV=I@1l2>!yo?qyI}
zvoQmcpE<i9TmmPd7&|Pi&{8}s^`O1lMq;CvoVD8sQoPnZvBfC0g{+Lk;(G65gnD3k
zF!O#gsnUJPM30kaq&Jcayx8tb6^7Uq7798=<$&$9uOUpwZS40G?7G))PZ0N=feM6z
zag6@@E*dCARRO;F=SlvrqjT}Cy2HeA7&71j&*;uNfur>4m#ti2I*KE8MM?ydh=LQ%
zpq3~G3GVy?$NO+A#mbuq@e8N@A33egMTkc4E4Z<6QoH_w&v&1i8_(a*1qYgrlF(|}
z(V}_-cm2W203%~#do;c`#X5T|TqpoeJ}+i{`o)*%ecX%cJ10S?y2sYSt!E!JBBR9N
z*nanMu`Gh32bKpixIRme5s~m()A=-}sUxF<FCYg{rewM~Vdy^MS^zgDUb(j*rjK*j
z#u1_GBnSZ<+9|_|po%T&aJ~RLDre&}|1VVYOzk5_KR>Diemk66aBxW$f>lJ+=;X@{
zZI%n8fi9VZ`FY(>9L&Fin0Q@l{q-y$A@!~*NA7K|&rRuEv?ik7s2|{yZ%4~=br=)1
zQm%j_U9nDVZ;{Z5^y}O6XacTf6G8}8hnx33^~bDMxh-_J41Y?C%QR159XFNJFR>~6
zWYQv}OL5nUt?>76cl&B<8T~?Avfi37&n$aBfzokCez*9yfaBk<?1x`Wv8tWFN&gim
zswm>EFlYbhQO?kRGNa$t1K_Y6X4bjl(`xRmq4^NCJ~o}h9Vr3Rgcp{=DCR|V=>>Ju
z84lq0pW(xuk!TSA;!FqzI!d|=>XnAaZM*}7IvH_onpa^G_fdWephdZ<uDhbHJ3Evb
zR)fsW(FGA?!i!HG27PBEJx?>JRyU(f-r9rl0q^Sq4;?>@WqCCxfKWwAVEh+w%~^jn
zFR-V-FI{(VA9f*#8k|?hm{(UYLLIH7M#STn{{#-<Ces+4s7f;T;a;gIOy|s%_SK~-
zwWj2cnAA<`b?kyDIkVhGb_2P%z>w;nARkmd1f^Z=I`^ku9pKhm0@J#n8FTm83?AQ9
z-+xz>0BFwXPFIl83BtVY|8jeSB<#t;XwNe2&synQU8?z^c%-1o8fj-qU?+AkW^$Jk
z_R(~6UK$7EwL2)SIU<d|cjCkmy;JcRySM8vnoKO+MSgyD;(`B-e9gQk2kaCwAMoMv
z_|)<Emp_56n?*S!uH7}WAq0}(!tj!{{UvKkczvv3AWiBuP}Iok@1Oy|^T<_}Bz)^`
zFq_E8h8GlNyKC-C8b~)}8#L6bDF>R%7d|UxvsUo;SMqmP@^5UXo({bY^Pk|gl!Eb5
z(k*yj_`jTuW4-?p>BxS8WZj`Mp=*!cWiH&a$xb}_3GT!X#~TInnNvBPab{IXQkU&c
zKIM-w!$Gk#a31d<S-&toL@n>d5sl(A^oAn6D?hP^PO6n~dd=r4KJ;G>$nFr&=Z9b=
zgjD8o-g3_w;{FkF4@uH{k{H6-L!`bPcN~V_k?Yw)t1i8$u4~cP+Cdq1`wuu1^NX5y
zVOdeZmCdZksU_oXdu5RwY39rR+pHcANc%w}k!ElTiU%4`wTgt>i{AMP?yKUBv|^Ym
z;e2t(N?y;9d{K|H$4{B3*+&!Yrm$3<$sAmnINK(;`X$7O^jW({ZD<)OGPLY-h2GiQ
zOE#<AJ%uw@*8BHfXcI);P8QuKhUSW@TgTiZW>AKm2<A!-yO!l~&N=^aLCv>ywNL*p
z!Gwlj)CO_=UARzpz=GD^!oy4*t_CdJb$_pmn?dg*aj{+Po5)%j&|?TrDR)O3-aD!n
z*u6HWn-)YpStyL5Q$*5q`0*frvA762<kQ9}z`ss8Du_^!pGZFRRx}S9glqdfWJrXE
zsB@|)i6UYoSQBv*JJQQto|Y811BSlS%hu}_w#Fl4_Ch}hfnz_HuagM3|F<iS{&4!l
z#_L$br4<41!>4<X2_+$kI8Pxx+b`4$1GgmleX3<h`48#UAKN}ws(UD7Yt~!jOzce~
z{bA>jm-y>_?{83sHcKA~dr(BwNf|+V+cSUi8RJeg+=G6}u8_}29Hu^!A;o?RHOtas
z!95Kq%vbUGasxqmCgtgya8%u5PG7AWA*G{}DZouCC@GRn&rq_NECQvGY`toaHNpD*
z0G{4f*W`;^Q^`VlGlgcjD_X6<Ugx-gRXhZS^`J@uf^WYOqJ;d|lclAUK|8nWaRTIx
zd`-{Z>je~|!=$NZa{DU?32>h5^jiy_u%5uQC82osuj-@ByIZwu090N1fOSr-&nq~o
zk@`y04rnF5b24a&jg4Fd5Y40RD{~hhiui>+j`Y8$x*ip6|BjsgMeF_fj;||g6sO$c
zk}A3^V7;Z4Z6M6pwnWdovHd2Ib^KPnds`;)7r==B#r*SndjT8tLQ8J<$n6^n{aWCG
zWW=tS(WIq^k3=Ewj$QO$@o8M4Y$eCu@V<x;7ixOK4c2fDLmzalB&~Rd@_$L^|43N;
z05W6<3F-FzGxuPPk(dZKC!?c^+jhIsr2p6l1CxQkDO+C7!vTm|Y<o!Lj+X-osmvK5
zm%=W!Gd=`pi#Lm@z?9MXWoT4WiQ4;79>P;gGe4C)mEEafICWS$=s0?y<(YU&eAXq-
z-N6JMaEcfJ>tFfdc1N0apq75q&w}0w>C18z0kdxG5bcwh-dWa%K;BVA-@;=*FWE^Y
zcaUIz7d7)eA0!t`ciC1wCagg4nHYRj3%`za--*lzvIXLgUQ=RBxKksqp@`(nFi_j@
zh}KBfUQzuNa4X#69f}I@M|86q+y<VZ^%(DNi6k^7x{0P7o(oEcP*x8j-!WmgR?Jl4
zBW@9@#j~V&eIcQ}s@0}W<L_k_Y%>cZF3+cbPAa2E|HskPJOyt0is4F?k!3=9!ILA@
zsKFM%)jsi<H(OD+dof9Ji(RFM`I)Wc?D%ot^f7VT@`rL7G#?QZk&2w%xPEWRwP9=x
zYS{FF3DldLZmi=aj^ckUwF)UM*UPkCtO`8ns0<Zkx(Fm3INZ?%3GUw8nrA1nEOC%~
z2U_V+=$cTj<Sa==KPC2ggVkt0|7hE&zGX$JmBs>zi+g}b%<nE_IGH!)w}%4K(SN@w
zn;R3Q$%xES1T5Y<01kui4cIlL7>!8_l#*^8t_(=sxch{O^1Z;07)N#mcb8Pl?4n32
z7VER&?2ndC(gTIHf`|6WDmOv@oTYkz?Fudt8SF7r{T63-r2=1TT;?IO>XzI^aprAa
zRgw?qYVV|%T+XBg0%iOOQ)XeePIQBKW}dQMgTfqw0_@3P?m_tW(k<m70Yfm@Q#qip
zk!}mqnST43Tbh=YpA2<$n{r}YJL}DDlN=xUfo+eJWR!FZZfHPVxTn}!1-=Ek0+TX}
zDdTURqF+|LNq&;7@0wfBsz8b|HDf>9V^ttSJ--Dqa#B>JhI!5DVTj=KBL&q0rIchz
znKsH@5Tb{;;E|1axfGD0qD)bOU{37<fS1Wp?YEJNEPHq`>Y#x6k5hH7UE&uz5#6XX
zQrJyOIzkdQla9BFF0J{`AVvew&=jL=8GmNyq&)#<<wSPYX7;g)vl7ZrL-mAH$ff~<
za>h^R=y5)oUsO~y$8L4{cwo4yD=Qa+^dUGPHZQXL8><5BEE!9&L<W_w^=g&tA|*l`
zc#6w3#2w(k96L#5sp(k`Bk84zi^1*o!}MpR{m3C`Xhu77M%e^Ldluts&JmAsED69C
zn3ujVJSo{gnCMS%tF>A2ryGT}>1H>7RnXyTK+mo55eq3|5__u@$FFz*Mvj0SI<)6&
z3^}RN&|XX#+tMqP+ox^mYWZ&lE%BlFgfci@{Q7X&A>7w#K1Qml^_hhtTY(_>jI_@x
zPpD9l`PrzQHTT$CF*BpX$2R-3M%4*2=hm~Iia*2=<I<T$bbB)(3^Jd8Bne-v&GIyb
zpLAgOc&Lr&Lh4y1%F2>LR0upUt#0I6?Guwr*X^5&#=Xu9=|<%C72~(j{q}~81f>*A
zBc6YBd_cLM9qSo?tb5Fl7ZVx3bp4@t&0K`%OQBJR5c`|bHV#YO??8quu%`?1Q|Y>{
z??B@4u}ZW1Q4r73R#wODfx-%(a<m#xx(*t91a<W5$479XSk+YQJ~9&%3R*Q|qu6hO
zddX*iO9i`#@JixJ)3V@VfAp8;uIJE)WKIPaJDr}I(bPpt)nUa-42nl6o&%?xaX%dB
zBTZ?99{p?I?W@fg5XxbzufVrPFlDw_rtV&hL?O`MF(3Vaj_I`Z+hsG)_!(jvHx<7R
zpy!|Tb7C?;IG^mVGR~+bV$+kiz9+cB&ef!!{AS1`BXhY&S`vTCSM-b0ittJaG=D~5
z9djecC|;)J9{$N-_NVL|C|A<hjd`N=6A=HBgyGtIJF-j?abNlsmX#tJRZN)M9SKD_
zKn1TbIMV$W(A2(gr}eSyzPFP$h5Z$<q0EB;UbQMdh4P|FkAQ4TM#_iWrF6x$kbqEz
zgMgf)bjJ{ok)=hdE#s2{^q&I8Udli)<;ylw*b=s}{B}iM4@!?8wbb7h+3C0y#66s9
zH&+>#?&((=3^nltgwT#G1GKH>Vvb=#{K?#9U%FDl^OPsKIP~2S>EK99(uzgJ^!fgV
z;sehcid7xl_|GJx2QLjzbUEiOmj}dFizO?gG}kVBz{0)q9tuKH7w?$g%tl}l?YMJD
z=Lw90;zJ^-5g$GsCWT;@6hDseQmVUduE(|{8^j~s5NPKE`F;q>8rs@2Lsp>iOSC9$
zsSTTzp;H#&yhZGG#fVA+V`Zr~VeQrYPXtyfJH#Xi8p9IG!O^90m5!3a2w(n@SSur9
zFy-1qR{kQM?E9SDrqAw7KXSq4$BI)i3iTh>*m^@mFHM4c?0<I??PI``W$`G(_-#Bp
zwEkvamL1o!{vZxYY(d@W<101?!enofCndFy>L{K(&B{#f)&jo$(yp2~9ym!dac~8`
zM75@9wRg!ivjJ4Oit1r;6xF4M0c=e@ND9g^lSBGSjqW8*L0FJZrp`VpjjXd=s$2Zg
zBHIDwNtw$~*C{13&bFIC(41kNv)us3Y~JZ(uPpkY8t^@Qw$VC%Vlv0j*QK=%@Dsb0
zp*q($B~*SY4+Ob9LAKKVh<cc+KzKFn1AD7|vTcT*+EJ~1k<I0c*i@JLl7mW6%IfM)
ze_HXvx#Jh#c`Ac7L(iB&OtLq(xCvo);+HOK;?kQZj|ay-zs{pPce=AyvT`bN``?>W
z|JLsC4uQjA@9zAwYo}d=efD4vctPxWP{d_YUKQ9nFO55RCZNSxy}~fLFbvIi?$RD1
zL{%=pS6o27pwoqK$0LYGf+g7qE5U_F?(8c6+j~PPlh1SV;sfxM+?Z`>XTBK4EB=I$
z%byS<u0<-7%_r@2lp-V+=G&-|F@41v5jsjO7D9x-mc`YKIxw$Xy&nXnPg5(fR1(FC
zJyL1e5P}?{OB*K1Qy8ROg@C^mBwSsEGqFbTk!7NafadjUze&C}J=`yjr6NFuJ(|-m
z7Ws_50B-gdff^U+Lg9d()+Y2599Pasp;KP&S<76cum=TN9eYT2?0w^>sN83Wn_=9U
z5iX7FETx1-lk`hT(1Z66AQUgVhYIJFq*{bgXGr)0q8bR#@`$Z6M{70Oj(oUxz$hO3
zZ>c=ju|tt&qgjV)fje*mpKt1QH<oH_Mt|hraa;O2QSR9=lp64!e<wXaL4e-0-VHu<
zmA5<N5!O22Ib)QfALS{V8YHNSAgn7kuKx`jgu_87O%V`yf~s3vSIh5{npkjTCkOM0
z|5=*L#K!V4*#pA(hi4~%aGjsQ%VQ+~m~gJ7?^PwK_pu$-ts!R~B($6Q9><IEvhZrJ
zLw>@VHMDZgVcNh;HU>S^Ol0ra>V5#rl~(iZ1FeU4{>9HS^+~ThXxQrev4mM*2gKKs
z`d6V&)UNAA<yrc)1%C8?Ui#gbvowAt(3Ykg_2k#PB7?N_-M$+@kALwZ!ExlfQ;qL^
zV>CDj5aTivv8^XgbcsPgaa&s}<LzaB`0LFZ&{s(Q!Byj&)Dn&sRZcz&lC$~&O2-R5
zRL0Chyh(7JsX@MO3(0ry4%4vt&2Bk;_8Tn$pkZ_-?DgXWHOCuRJ~iZ%pw2!_RF0uK
z4nA6~@DXC{b<lB!A$^xKZ#a}|KRU~3)`r6ea?>O};FBY|l<R!!u6hE1R--eoO$id*
z3d4iKJ|Vx}Gnj%F8=dtRE*{g`0|vA9FdtCv_>V(umh&V?1U75;l&DULh1UyP#kst2
zO`YO#O5EmOOk9vOGb6&!(U@zGL9$;qWx!OWk-Lg!Gd00Tv@MWz#~<2v8$r$Ih<V<>
zw~csQr<I|Xbn9dA=-)BqfB6qM6KGdR4Vvoz)vkDiJ>yO^#~ZXVgNKhtRR(2j+@nNX
zAIVg0kw!pkWCZMI?XQl`&=5|w+R>6*H1QJUOPOLTOF%JXND(*)YOioR2*rDB4X3L{
z+Vr&{pD~>!$LzoZh?8vqylRW+*>;OyVlH%vFm4XiR_!1Ll*(X4Kna@S#N(PFxte^Q
zf`?TJ<|_&>5I5bZos?D!;sJi@jP020bim(|dKR}9X3R@NdkS0-OIMb8m!Iu<RpQ3a
z&T=7>A&J&-6+lL%=737b$2~Ag@1Zp@M4?~lDA29H974BNaLj0(TKhdza0ZWr3RTz#
zdpG5_r^qHQodW4v-|z%d9p-%BWe&)-{)<O=&8oFp8A>#yDh#<r%HIfRlc51spwCH^
zGU;^LR01t<F^Be}_53l@7eXfw>#_TSEqNh#YytUr2#TJjw5VhudegO^RVBl)DLj%v
z-q?N%!#CU4$%$ivUe|&)pQ5R^3&)dEmXqgmg4TG)P5fN-jaSqj{~|TWUkGK%Sl*1F
zAU!$UN^%tK<$AIXB2lKjbt2()L!)Qz`WHc5MQp6jMR(V+lb;!xn3c&wCY5+E{S1A#
zeBXe6fI?X4Uge=u_ah2h!<=YFF69$@3^PDoK3K!rgvb&}^4D4;9$r(`(ZO5v(<V2b
zrI!jAt0Zy9MKzyc?CgSZ1+!qnDzq-)gqn7!KjP`1VCGI{z==d-I10Nyv<#yiukdXF
zE+Y5N9PCb8`Mh(~F4Z1-z^RpETlJMWM~DsDfd@At@Ms=q?ANz<j6LNkdwG}of30He
zX(JnZFd}ZLn^SvI7r4Ea*GKa`dtczO6eQRc5qc4AYx9neJf#8Q7kFF^LN4@3<1Wf+
zuB24k(HyNl$JN0$gtmp)#VOvdVvVL^jacaC;D&$gVE^Y@hx37qO6A`&DnU%t3?mXp
zE>^&nqdJ%q$t|Pa%E0PDijrfh7UV>pD-OklteBTut{`%_r&KW$SFUEQJjh#NZpaK$
zdL<<wwE&nwAi-i~rM(X83c|;c7BH2;QA{zWSRKmxuwq`3DIxcYR8!rLfz^i;U}erL
zSM$)BTT0x5&mL%hn}2Nbgg9F|X$~klx)M1j*Rt|lhcO4^6G;oS$;iA;nHraFgI)}G
zlN_olqm-CgyRX9}pEzqoG6K_s<dyb%(HvtD)dHv2;(Nu3X?J~Hta)i!m!L4);E<QO
z^~m5}WC3lpv{c=b@q(%)$}DiL@9(`%<R?bOGqNH`0Qwrz(#_moUWhkj(DE}RgeNb)
zvcULb{K0+DcQoyd7u2L(2;?~}54zz5>X<MXt~BSODDKm-kdPr$AizIJ48h+OqeBLS
zJe3wTenP{S`(+<F+*RDpLxGCFZ=aHh(QPd<Euc{e2Q0w#T4M8++m^yGaqVf<<S@%4
zn^hoo%T%Ffqq0*UxO0Kgk)P8cQKGp2!E$iBVZ^rIbN-PGbQid@hJw0JTp`;&XX)*e
ztLqnF@E{3{HzWYcAEz~&n#rZgSNhc~Lq9pUq4`FC2A@dT`0WWC_or<kDa~R{+Jhkc
zxszi9JUKX#tL$o=T;AdITiJ=~5@se3OXDrcQ7IN1YXzx_{d7&jv4+H;^%L<$FH&8b
z+3Ci-B43uF+)zzQsm&z6OI35P1~ns)$8k}`18W@Ad9m21_Hy^Buf5_I=X|m-^-z~X
zp2-JvQU<b<miP6$ZrJ$kP&iPpQ=b#<jiI$KB@QBF#@%6~l2XkfA-x-}tV5pxy*qs3
zepYp48+HdhbF`-`+GL<W<#u;cAur4pznv;<hc&F?Y~DHT_#9+8at$8m*7_=vDVaxO
z?bKY&Z~hep<?l7T*j~eQ6WEAlFc=4oLpNubgC5R4zeQ6b*Ml#FTuPZ-Do-k2+-0pS
z<HqXlf1q8_sy1=)ZUlXwXg3(SXX>N6H@UQo5*sJ*t+2-+OSfP-lRu%0U#2dB)Bk|{
zf!p8&mPytHchCwQp{Zzr=Q=%@;n>jKu5zQh>4<iuDjK(_cM=RXSO5+Z6)3s=MHQ>z
zQqUxfaZ>uLsqioC)ZbYmOcHeULcVm0{JC?#T!$E5IRHKp6+LPYfWL((J}1;-Mdo$#
z7h3ehn%z<>FiN5AU)Pmdk8NamVv1sCp={@Jz>Awkp0hC?HMNUlt3#_K)YIsL6(yDp
zyrMI0r|YrkuU~I{1ifl}<yVH~&9i&d!@Mzg`1BCJF^pzHYkYy{L*K9+GfDXa>ASNu
zay0jcy9N0-=^wiQtVFR9!<)3T8kmAiG8ehnlA;zEkpnDyO8ZQTH{q47Z-%c#bjJ(+
z7@Ip~A(e))RJa(FD(+TRe)~)>O>0PK-wnttMT{%-_|hrJN3uh{@5VcT<eUeQ?gVI8
zVUv}vsI4FugsIxpP=+!cX@+DPHho)&9Q0I@f*_ezTH+6dSwE-YR^h1|9u_0y2HaS>
z#zpeBG#yk*;y!NEHEf0*Y)rO#VoBp<KMyeXxdS4d3VMUo)A>+XvhB>Nr7wD<?!*lQ
zq*VC2HSKx-a$s?e1rI4wQvC`hg*EADByl@7%FVJlwKAGJOJwB~>LyW`OiPP1Hvwkk
z!dC}a-w@j1y<>{LA(u&ReS<v}&rypw#9VD)ZO0+V6pqyvUBz&{l?%F#I14YoBi|hi
z6e~6c8IPCgE<IuyvELMKSo&;VcSuqOb$q`}S>ekcCyk&gVl@AyVdYda%Ot(8gU$#X
zz$!y=3E_2=4<@Hso%T19lQ$Q^J;EzlMILPw)^z{viR3GM)t4)N$Fg;+_^rjz5<s59
zE=}TnztG8c_vo|G>Jfio2P%SNq#B5zx^l+WEak&)YEyj?Ejnnd4DpOs?(e0z?Izr|
zr#vY>I3fKFlsK-orbG*HXBhXkOk^9i1n}b;(R$k$i-j_G-y;Ts2=gA2I&$%d0{Ya$
z^Am+S;W#+Snp(FJi?0h-X%%(zcqN?1hQc_H(%Ah*-f#8958-_1_=?ob<FEKC7`sps
zZ-4|_c8tP58W(XR-h`BFN<Gr<7s1nQc7DxNyttnOin>nf%}W)^&?S0<B)dM>_U3!m
zjqNz9M-S8?u)DtCYb#nMRJL&H_)M6Bh$85)96uA@VZ;{OzV3Ud`X=h;R`Smy@@b~G
z4N2+OX2Ob7L{*60=oTe~0Uf*eQ6muQJTrN53dQ!OsdP1>I65M;MZe6!)@Jjye-BF7
zC`=<F2m+&?#p`!qodwaZ+b+10tm`}z2U8baPl>s9%2y}J*5LucOH_BMtnu8jI}mrW
zrkO+|Ui;cKl^dxQT9n>7;p_BseYkUGM?AVs+)Osz4!Wn(qK0v8Oi?(_$(~KFG31vw
zKFFGhhQ4OcQJgc;E6ts`mTva7vgaepE%p>JSILdOY|^GsYuE?5q7BQ78P)YOQul@-
zhbXFYYQ>$4_<e=PHEB0_Q>PD?Jzhv(B4R=wvG7jGiXS=%?KX<<zuBJ*k)vRv*T>b0
zpO0IaGz_R>>IAQ!MN{6Tnu_RkHp{z{vURe@SH7_av2Tqgu^{%=p{Iw}Q>A!DJts}>
z@CZ9yVY9SqC`X(&bEIxiFA~0R$`HYGGqO6qzKgioFJPGJLDX`lVM(-mKCm3OPcuDz
z#=BESl0P2@`fqoTey=g-jYpisJZXO)mHhki_~Vca#^wOA6M+Figz6Nd*_?0fLo6m0
zK1?Te-~Jnk%kWZ^@m31&908L`8${i4k|P3I7(#Dm|DOTRzqfcb9|P_zKaH^K`QI(7
zNC^#wLSTACpi#F~f7%HI{c`UfYdoj^WcC$~q3pFC8^HAilH4ne#YW(r{bCaQbM{Z|
zF8>DM{VPm`hTBlLL7M_)D>d>jzdW1T8=WnIxwat+D;j=KT_EkT1|H>uUnH;G<g<)K
z$Iu%2KiQ#p>bO`CMZ7*w;7+BHUP<S9RwQ9JnIc0lj$VskZ1{Uc{&6OUq}xm?G!S+2
zNih2JgBhr{1&|p<iJxstjRSLY>$f27nluGyPCikk&V%0(B|uji2C`c=9p7T1zb_;v
z6&R!zn4-M|lDw|n#jd>jy;TZD&xbPUe^onvAfgd6&EN3@_}>|Je-iZK5PWBywF{>d
zC!@g08IHEQ1MO$tfBuPRlOeO+h3^4;6NmlF0Yo&cj$o3>|My>Z1}Mc{Af+%B!D{a+
z1+Ea^fnyM?p&OmUFEVYTSr}#U^QfbJP#|GOj~bp6)o1d~R{A0|#(zTFc~coW*SvT4
z1pa$O5d$yR61`X~=gw$<*OU6qO2B3U>p&v9JLI>h+`lZ6|5&`5=u=aR&)y^A|MLT|
zfTJ82T)|xCchSc`pZfoZP1w+d54PC*4YL2AAAkwu<P=cmlE3w=|5h3MP0{QhFL}ch
zL1Y9|jGz9~uI~T*7yt1FZ!bX2UW0tl`R8^A8?-T7+HhyCZ5MEIzzjh#&3?W;=kc)4
zs0g+tsi#-W-_)?D?!wLg`_ZdI!Bb*`u$TNvsCy$N(5{Uy!R|8;aXDW6+y%xF#UA(k
zVLZknJp+uoJ~o_h*uDhLYG`2f@ybB$;a@y4_WS^##aQoIjYLh9S;+%rUP0Fh5Mqyj
z;%v9RkID!St%~F)DSv?Op8r867t@Hkl;#3rF=iU(h^7`X`v4i=kRsOP_PaCpAIlDo
z7PDT%PyBNgH)hMu00QzIpgjx#;pPw{?pfa*C!h>+L;Vd&Ag+BtB(wt0GHCE9D<EKl
z7X!Mnq$g3-^|~yK(GH+C-+?vD0vZ<vFd4aCwOQAv?_VT+O~KwTezltKQxWl(7r^6Z
z{a?JT!Gx~ejf&|`AOgPA^_lZC5Fg2b6DG7bj+0CoF729D7Ca}fcR&S=c}sau4xDtP
zIsm5vR)UuPW9k4kR)9eA+n9Qd+WO~8ezT1K@lqmK%Zb<#f4;8}|N0T&wiT?e*y;k&
z>1W{MA`d=SHSdudP>iF2zfRx}mj*hP!_^$@XP2ZDB2I<DO#s_bV$ul15D5XN;Lt^K
zLWm;hrrlHa`HMPk{X>ipoZ)ef(gM%kz%<ULoJ7|o+X;()0tl_urSF1+UN&xRv*FUi
zJE3`xNkmEWRkIt~1PAT9sk^CcPsAp`Dh4jQ|FIqa?f*a-!QpZZ;`t{&-<bVy24J%H
zh+Wu#Uj^_n`2c`mIr!1aRNZOzSUj6vg6rYtUBH0H)`E#GY<j=4UHxaOh-qbjFM!Fo
zWomdxelI#IwGnn$QWytA-yE<#dj$h<IAWi#a{5_Dl=7>(taf>ztDevxC9NpvHo#pX
zoXc9{)u&nob8FL%sQ<66A`uUsfIY+C{u6&O_XQ6$(xHWv_cY(h`~qesUqRn|7feNj
z9W&DvK(N?cU2`qz4e!`UgC9djME}=s!2rUBhr)lJGr%+fqZyCk6%tH~XFzv<xPr>8
zMnB!H`1yEdF$<WVI*VG#_<NzvB+*|!aQu*Q2IUyj@nW;VYz;QBqV?EcEdzC!e&iTf
zq|c0;mwOYeW}$%7wjtQ40<IsOMu9ZUXfvVS(@=8GPe3YwU+YOv<Gw$DM!QWD0Tl5k
z<4n*dvtSQ~HCt@Xn>PF5`M3c$O17%qPPWC&)VIYiEkx7A(eR&lVQ#KV>onV%07Y+-
z_%y9l&=25zCyVYhbstXJiVK4Kei5|j-l3f)XU=m0Kh>TJHl&UMD5fx@0mdbV?b4A7
zTLZNDChzFfU&wH6qZ9CJ^*JE=_=!W06+HG-e{-3{nR}++pu%%gz&|_J<G$^W^zbt<
z+)i;H64;+#_!SMvikvhlk-V=4we7)tJz=&(<~Zy3YU{*eM~xF?{FtAT+;)TGU65IV
zeY4Gg#AgL$uX61&XLCTKHE=xgsa53|8v`|Hc5svxN0wfUhlqR{6jw=9*&s=IVd?55
z6q=5<@q{z5DWiu|=MrwSpWHjy1IcgF-gD=Pd&^&WI2sQ?*i#**H@^e!_G#<;CrDtz
zv<OTjZI=dL)NC-*2*2(=JGfn_yzk%y$N2yXLy~~}G&RxhOBA&^L-n1Q1nu-L!{#Vb
zMCHnYZ^?i9hPEZv-ve!E%X$9xL5oD^_I848e*w|3anLK}EUEs(s4@yb@VyhZB|ZEl
zRRrZ|>{?!i0F%4E>Ghdz02X5|3Y;JMOYv&s4iue~)uM>7niVw=c$E4FCV}z@FpnW0
zDF8k$<_?VQ>ULWIoz0H?fQZ7J4@mmK?Fk>^TQ38;<U!!Mvvygi8Wg*C%Sji_Q7?iM
zvAm2V(FVv!KDq(~w-ANdo&u;$5#*i=SU+Z<Ds_<|-C@P$zflKdn@i)cwKgEMG)*KF
zH8<y3!u#zGk?=<=j}V1CYJWzBpc@5UDwGN&`os7a6fSHT$gJ6=qIZGsWhbz#G(p6H
z!~FJ*ZURhtm71kkD}_>x1`l^qImv!6LSqOvJRR+`ZoMaThtZl}Di;N&k&p9&^;w07
z2P$TIlM4@Sast!W_`@PeLi&ERW($C=WzM+1+wB=(5w>76pex^aHidW*c4-5zfP1Bm
zAduN;dCT6iE>0!d@!S3i2AOhH8qs-Xh&0FocS;1WoevaZOb3C%XAq}eUF+i|HoP7n
z;{5<b8{>bvQ(lIH_PwSZ;g;kXZqrZD7GyG9CEdm3gfKKND0`lKz`FzF8M2oTk&LYY
zki2UnP~$H)@Gu`y&_~hn;rEunW#IS&B-ZS$q_AK3_KxTEx;Cl4WG?OwP3IX}F8Wm9
z*`V|$IB2QkyeMGP$Ka_c@L|%<#6-c1ML=`${fqH?Q8$g{O7SMH&S^uaROy1un~1wX
zUfl@HVJcOhUp7`qmB|LVc-?EY6>B0IxN5n!n}!B_RS5c06yf{<7=nqN&qt5;7LUV=
zvScG@Zw>%Q8b?J)$OV-m^iU!))c)!@biNDtD3H0gt{?f4DsjEL-l`Ae+hySmD_a!2
zB*>#M-d<>syA_c3ISaHpeg4kg_$cUrWtbTr;<HVK;MlgBEYAym&!uEoD{&bZL?Ii7
z3#DF`r#Pf<ZXXLdq#tTN^FPL&9|e^NHmkfcP5*$a0Z=*-`b0$YSa1=62_g2R|6{X0
zBbKM%4j|l{4BdQA%~#>b)}uP0G@YQlg2lhj9kv)fTI3O&0X1ukC#pd>xs0iGeY%S?
ze|1EQqM<g3cu)JiHia3mK8V+Kn)T;jo{b=h#9cRT^1<=BM}YJauWYRS93__^>#k2R
z4XilDPJ?$S&sPA90NA`e-s>lHoxb(2!BsmMOE4<+Hs)JI@Nc8#FFtrHm)+)nN2I`E
z<rU-jIS|pj0wZgK)G$NTRQ=VsLw+be&xjk{ME4%6jW&lE6^#{l5;+9pUnmLxnt9C{
z@{%>X&@GaDSDn~RNuc%@AV%la7a~7I;Mc_Cx5S5(#k&N)tbVkKw&LuPYQs^I2~w(?
z6ht8xUR@DcbBD>_hgsc6Rhle<TG)dTcUbDU`$B5s$M}#J#hdei_+P0vbC%&<S46gj
zFNjf(XH|PopK-33SDlVFT$oLdD7wD^*OU+}jY1YBWO3F+?5G}rP!jl=fDcP>S^9ob
z>caO5IncQ)2w2>_DI_7w+Rg1N1;D&{Gz+77g5sJKGi=Q)8klDsYbFBueNUEM-@O+<
z$Fj(|bLc8_oCdPK{mANQiD$@M`JRQ#pMiDLDivVpW9wM_UujRe(^dMM9;@bz(@boE
zn(S*<{LW|c)8$AK)(e*<fle^nSy&Jjc^@Vct(O9LT~}~be3=9RZbFE{Pf}%(cg~#q
zk76&pi~XKQKv7u2eYCP+cfk>CURBawIWw(g`<_4jXnMdT`Ssny4r5~D+}1Hav8-nH
zj44){lNTaJKh?ZT)i9HW2^&%$wecwVhbZ}vZi0zj4oJeGld_7Oj6|2tEOK=!<>5)q
zeGPKs+OAHn#R%wzYPO;wM^kieVt;(detgJFn{{cbEOM2#4B%)amYpbNqac^fk*E>j
z$)EoACNHzqMN@4?^X&{Q_6#fssxB^ZP3c{mwGC|?-GjQL_|N+rorx_f>EIeO{;5Hd
zFAQ8x<3ofJTn6t6d*`L(aR+9P%2Eldhi!tO@|`cA^v%euiya>|?m{*C?7ZZ{e4o0O
z2izW+6{c1uu${#yOmaZMA8@@fcE@s+6;-ExK0mWEkV-Rhjp74vWnNZIqFetp$ux3}
zHS7%#!BmF->Y#QAYdOPCkIpF4myU76zh{1O_B(Yl2#7MZv5s<H|I<L~qZbB$+y`tI
z5P4Vh=t0i;*o}=9>!467q!aU_ejM)!Fu2&5Y7Ug;H8jP>STt>4F{d{NynN51Sckli
z%KXCy7S1j&hxG2^ADTxru9UDyy@&RTU-70!W+w}se}w^u0I35T$Yjd3;m{Pl4w94y
zOr@bNLN8IEM{<Wwnh}a4fsssm-C=VciyuXg@VSqfk`|%>-SxiH39Ncsu?F}9HYBOv
zsS`oQ`1nlbL>Dl*v9PcEzjwLjfHsXtmffBDVK>AqF|3lRROhoO5_sSQDi(Ryki%TE
zX|a0<JaPnz_;QtRV|EYX?B|q2L2tM}3GI?}Do**}vh|W;3+~eU8I7foc1jeIDpSt=
z$U2lKeJEAvcR{TP-9oUWO^GMCeBm2V-WThY3VX97^5cSNn6_;KyB#x(JNjvg$!(;s
zy;*0}o!4|fKfHIbc0eAFNQU|EFvs@uBXN4H$$U86GE)4!2Dav?8J@Wnh#({`Uzqh8
z)n5ZHSFM!<$#*CAfu92xcLHaexUp_qt*<lBBM1f9HaD~-GC!O07Agjv!j0X*@#a2I
z{1%GWauw}#HJ9Qc5_GIqHz0(9#Gv%|WrLiDmT@KTUa(+Z_D`#DiocnUV1mDFb-9~3
z_%>$e?C3v4;Qvu_`0~ee0B+qFwm(A$byiHsof$&hIZ^9%1crb3O6KR!mY8+WZBEZ)
zdC4X$>1R;+rj{psk`w{qFP)587i@^J!hz0Ah|nR9F<#Px-jDm^os?~a1mPl8+dQ*U
zyycl{x#h0@_0-X`s_9pNo4(xSG^3fFm207kBTveuf^D}D^uaYHXz<JIll!D9G5elx
z2Q2t=Qwk0Hx%f}9X+R(~f!YN6O!R5rKC93Lsnc`G7(f!h(>2If0Z&)nm0fcmcCODd
ztHal~8~^!Mx&J5!#LO^BpNX`Ge69`H8r2ln-avTaH_ZYyABFob^y3taNw0-wi-wfp
z1?2%3E2@lSLsBb@CXOWvCyUVAK}~RqA9kz3#!i<a*m0CT6YX<Ms-HbM7K(%jFy%oO
zKF=}2wXKWmok8BWY(+li7f{txO_Op>tOX<+gCf4o_3*}sP>&s4iW2jl&ZkOad?O;;
zayhfkCDQvBR8Yd@y|HzG>443BDdy*O#E9jD>KUCrD;Y4E;O31b|D8W4s1vcil^`_D
zYsUEJc7CoTA&??dcN`p`N`~TMWYs5uefb7>=i8K;8hpA1mX*4yoWg?lhN00ZQ<l<g
ztSNE71P@cmhpr?nrZPzBy!qw0@hM<~6xJOSFh)jrN#!oGjCMk=;K}@=Fr=nAaz0*S
z5$Y08iZ<5d5m-ySOs0P$({eZDMZl@Njx20DD1hY#cn!!Z`>M4D#9uKQ@YNPuYO4%D
zJ7fB?GL_N3Ih2DceAISJLdfJ44pnZ|rC4!QTjxQb+*fT1$?OnPRej%Zi*=U_W{Pg~
zN!ML_B5d(Ovk`B?`VUmrO~#)uW%hB`ISj?lL`kWkZg`s&gA5g9x|N;zhe-~r7Zeta
zzkbI6iFOeUvYL<Fib-1oHbfsloCa*tR~bOjK+c89fpC!;#2e$ETynd0iS4gvdma_1
zJtIz%{g#R^-+g1aMt@*DrNL?p(6$Eu09hBm{Rha}P1}EHm@$ynth&}w*o7X7Jn&R-
zx$w>Y%`jZbSfMqKpdAjj77TGjnQRfi_m2HQip=hyJm+Gbzsp|91zKuqF3OAc-h5Xq
z0xs#4WYdz(HqZU7>FEi~`D~Z_;nPGHRP!|<<uWqrDaLIt7Y$+@qB#v?(P37$(-Tk}
zvalHj4Kst>*3^)zHg%+Wu6``OmlV5N+*Oto&Z_n)J#7>vAgW?#{Z#4<mO;f0*!xqh
z;<rcBchNj=ULF^%AxFW}ihRraE={h;4NXv06(3MpbY9+t?qzb4)w=_6YZoZws=bKY
z-})qEesbvid#`{@Cr~fd6fyDm%k`5r-Yeh@X;ApkT1@F*o|f;vOKXiu1lV0OE-m)+
zqMTC_db6opSPKT7g<7sW^^PRQB4;ABXM>tpKCI%AF$UWGd-j^T?163whzYBjG|6KK
zgx2GS9)MM$n4^*K>RQm2RHVk`M#zBSp=zA+H+L+Y>$B)JWJHC&gFyr`EdrguyXP4e
zD`6@Nl4fvSMl@X`8mc17EeD2-C_o4>Ws|Xsu67QAe0hMY!k_l;iGJqAthnxT*P}xz
zx=@(AOaq;q(~X{z+U4Wca<r)@aNT_XY`iI&#_%N_`I_<vw&1%$fW1zsxZWrHYQ+$7
zE2u2>POB4n4`-`$<^jtH^B|-TKI6~g0zz8mT-H{}x(;-mOu<l*-q7Rmvk7n!thFVq
z@B-3H_QGeWiZtcgK<;o{TBd-+(^Nw@uS|ow$%zSkEHTL1kb_~b_UvBc#I^bMctHng
zQQWxmS%DGrXf1B>?8Aqo;wNv;xQ9TZ=nI9;$>p+zD1Ln#)8-M__iuvl|L9j>;HAMa
zBE0^^>-TdbaACb$nvIX~3%CFzgzT_0aUtsIYiSW$4D0b6<nh3Hfek5jC8s01^>iQ<
z+19&@h&D_b@{P^WDM?pT_PcmIH4hS^7h9<cJ7f&%%!tvHd?X21qPzd)L9(~TLNqE5
z6g4_J)8@Q$*MK@gzp1cNjEoL}e**Pj#V>p_RAOTMZ7`P@`(##XfZ?Yx_ajlBCzkrK
z{1`$HkXIAE_N|`Dq-FXAh9{WP!wD9jKDRXVTkyXNFJ)ZM6S;&c3BST07I&+Xvmw6p
zSTQLwGEAK&hK2CQT5vCmVOQW^@DyZ0ht~O@g*8h-G_s9H;_cF6t{LA(9uH+@a=pH%
zz_K5&pr#M`<E}?ssj!(WH(cfJ&;?1oV4ny;Bu4t?=U?>ty37IrPN~IB%KO2+bfbf%
zU@#B4)y=S9D@k9|pk_gy<n+$mZ;WWAoLs|*wLP;0%m~WjY~5i!xDG-~6k|#F_z59e
zzrgebLx&SkSX+VC3ne0YQpF5lt53c;T-J=4+%YjPnx5JFx60rDs0)52jT|&sjQjgd
z8s3dyBt=4iLMfpcA#S`jleO4qS;!c0g=jrImm#!TMo(<XNDe5GwxoPAw%Y*eE@!F1
z9l-QK*iEYLnV@xc#|LsOr6e&?54Jt36js8f&&M?4Ix&7;nQ_yvMxkk%!@$5mi12V-
zr4RN|jwDf%D>*qe!9~FV`Ha(6mA?A1t-JHq`~xtF1XA9JM5j=Nb)9m_CxM)vcQ$6H
z^MZgCdV%Fl1YTc?^9_g1wsMLOaMOTBT-1Qdn2w3ANMFM1iSIT5`Dextn-0AgQK{h9
z&$>OlVkS(?J+){WyNmdHU%l2U2v?HYzL6{H*G(Dn5z*)qb#ZC_$1#~=Fp{m&MM%N*
zh5BocI}-@O*|69C@R-NrJY;miS)&NM`P8QD;}5Q;;Jz?A@hKysZ=bou-_nXsv<Dwi
zvMz`q8<I33+Q{hDhQ8m-7D2cC{+Qike9ZIaFGSQ1Yum20=_*wvsDa<|Ac*Ux?5xlQ
z-$w=7wlSKR<2j4qWPAsAY@yCCP-q-Ydqi5{v@^Gom+RnkrB$G?C@~6c-odsE4CCLu
znrteW0&P7_6$m6E2r6o9;x3cXPm8JQ(XdKXU`%N|YD7w>zrS1dMrNprZ1$!IZ8FVV
z`SrU=0XVNIuNNuM&!9DxC=Qu!0pnIkiJ+4nYZN>s0rWZ8M%IjyMb7I~&RdqX<Xn&5
zu+3@buCP4`P-!l)Ol(VZEy@efdIQyz5;*}iJhJYq&0<cOzFK{5cpHF3tEoW`4tWOY
zkMBVIOEeYC*UY>B3v>l0MD7Elty=8<a_Bn~-Rw?UOg0Ciz3stL+}(%5o5s1L+pSB_
zi1(zE(`J43M)UQ4fuYi$P}&nba9*bvT~Gufz(3|97qA-XZuQc^IXdqE4V#SvAU*dn
zezkSyW=fMKzF@?O+YfxqcglB6(l5pSD(?S}9rBN@0h32_Hj7<9|Ke<#@gyRPf$pOQ
zjOSvJ|B0c@zm51&WFo7DE^vwK+BM<ac-u>mJ=1e_b`=yF`D>=ZG@A-A4XqlmjUruF
zh!Eq1H+fo{M^URkI%*fB*#gcAjp~6jLk)0ps03dSr+j8P=q(Fak8uHMg6Qa1d2#|g
zD);H6mNg>&Uz6Yx*MDtC;|x~%BvInFN;xQ7Mn_*+JV0%4%?AnHv+Qpyd+}?ciAqbI
z3zt(8N`Q9cOwOk#tqqne#CV5SdouG>?6|)X4H;@W!x_L#$J4t}XX(NsM3N8jXkiKP
zperV4zBbAF5R;XH<9T#jRGX~yN^NeL_4puY+z8KOUf;l!WSk&nkbpRVrmV>^LzLrV
zpCDp>4^UuuqTRm2i&m`HfD&JK?+5AJLcJg^xQ7>^lqspA<!!(AfJp4c7%n5k{_Ds+
z=sBZ}D|S|pxldtq{l+;7qT=HGbXhmzz<cG8gkiFw2AId)hnSt&dMe<*yV=9IP(?m@
zLFami$t?4+-u(zhngS2OVuq@>QiHir+_J9-t(v`z9NX@NhG3GWu;gQ^HpO#UbQ{57
z+$kwRnL*JQGOwBn15T!v-JYHoD3MlMnY);KIO<LO8rB$g+;l*yONqy)dK3c~dH@7(
z?;I0uoUrR{#a>sp(UM$SxR-k}@~SL0zYY{(=^NwM1Lx+y4otUour3$<JYmUD%nZ-$
z1+Ev%Tu+h`m!D)s47i|Rl#-<?;>jeBH9YYZ41m+Yy9Wi~pPleu`jZZC<#7I}lUz<%
zO6d!mv^~?#-5L^vJ*7m_rtdw<5gS~218W}^H8^~SUN~!s+1nZEOB${#0F4@|cRy9h
zc=Us>SMu;`8F@GpTFvvI>QMNXCz<+aW;Xzxb$91^`|d3kOpUUnR{aQHa{?r>!I3N(
zQa)WF=KZ*`@?doYvYx!+QFvpH7>5UqxaKf(iBhLQ)S$%@t?wuV?uCK%-+nkUazzlW
zc~<rK{P}vgdZTPYF}h47Vr@qIiYuvnR3y?+j3>$2>6_b$+#0Yr0jWD%?N%yoYy^u1
zdYaHN3*V9?&YnbwGexVLoc<oy$6wCiTS&&=Q+?6x8N^Sl6CI&og04v(Tbj<+>E@$t
z4QM^j?DteZcl&xSiFXH{yGtYX0PE%}S`M@z6X5FVktOQ>CAv2u8%fZiA}At#qjT1q
z@Q1xyti;|IA3cD{&QjP5@(5pU1qW7jd&UzlhWANzY}2*$NHHx*ey~Gx_=}rQcGuAS
z){!&(o7)fq`~|3W(4WBX*n=k>uRIt5y~gePrr#n9Jpjkq<mnI7gg7pEr}!iloT#2V
zJGF_lTGEE`61T&C^5=a7=1mR2xr2jR$)-DF`q~n0YxhYI7zxH4uRPLuS?4t4jhhP~
zyLA3#Ns+zCx6XFWuSABMfO(D;>20M_FrjQfWSyjZPa?}%g?`QMxw6>T(qqW;CCR*g
z4)qJLdl}+Czs)E|V#1$-o=wTYFoFBh*U$Qt6J>8Vyc<+xpZ^fcOLfj<i;u`>q?2KF
zdoU&0sjIq)diTIRO2@BX`3UkR+29dvmuw9$oeZb0?ROOI3E`;g%Y;BBg!r6MRpFBh
zvAi7TncGz{&MRGRuX62Z%<5ESIYnxz^A(AgCV<ld&B*{#OwMeeY`tk(<#ibwO5nLI
zZ8n}wX9kY|icsi9@~!ZT{C<0@>inN4D%0vpasB6rLn`Oh`dd<ZYqXm^WJbWR={2an
z#06${OJyV%eA~D^l&J5#2-`E%-j3KZsK!NJ%qMfHd2V6y?A`He;RZ&x&u9nB?~Q0P
z?&&Ow+yw0VaprrSK$ej=**$9)Gqany8}sCZ_QXFdGBoIfG^@_@kwdah*OnJ%?e{z%
zLO1EFKh2#6ZOV&<m{oed0KMd;A9!o2yz*7cU;0MtMq3tUT_YzxZjOSS4}UsXB$r}S
zl9ow;m@QyUd~f#Bbxv}5`&@40>7jDcgZplQgJr4wd0V$=$@ow>HCy$rQ+0JePct{?
zL1TA79D~pz?PB6er}Lp;*&|}%fK#g%H}UH@zV@-dq8_3DU$z${jZFf|rKkKh;E(Mk
z;R|6=m}mSo4w~z9STM0{eMs@qirYwuMc?-PJ{XV4PBeX#`EW_?IJVgyKkH7^R~c`Z
z=;6CHM4=s!*(@=R`1MCcjFEX<>myueBFMFY`B90wyn@)Q;#wq${rCi?Y^t$jf<D2l
zc9l`_0ykMadMp>>Az~JaW99a$0m{EK(4x?>pM}xFVK4apbgjz6&=XA9MKns&LlZuG
zG-`d5rkBRiavN%_rE7MGP>UPIUk#@b<}V3F*n$p@yi*?hYiF7n=tx1Da6~;pO37n3
z4%(lHXvH;xyJFW(AbBilKx}o0h)b*PmE|h|HbUyI3UnJRrJ)Ea*3x4TLPR6(^7OAQ
zif(6kErW>W1}zLb>`Ea1HP-EPsP<+`<9jTUY}c}<IDJF;ZhS0i!dkAHlj>g>-l$?A
zyx*ZsX3&xvg~EG8pIBaH&!it{MVIG)dYK==hfMwXGV4-BJk_d@IW%JZX8!R)IeyC#
zL_3-2!<s`9+e+4U$b_^zJM#9B0E>05#4|;oa{HRWGI8^6icl%Z=UUI_@lly{Do^j0
zrwvTQgkLL;e11x|SdNGKfxmt0^Uv3*2TQ4w<~|B-CwRYV=3J9E_3(AFIEG~Ju%~+<
zykC_W-36o8#g1m6NI4IWtUlOEHBdM39*$?fP4$pV%HuLl{z|`TBox87##k@bx%vi0
zOtK$AwjXhA3Y@o4WK?+mo#BjTPkhfcSD$*T1&+l;-%TAf-fSsV-9D_T(I+<d5>m0|
z>R(T_O1)p(pK>9&oW3QB-?yrtZep1*k+G>KR&9CY@tB26chg<{^2lf?Q+1ZkkxZz;
zDPdv(R2jAB*OH^ecrVy#NhB~Zv7!Iv2U5ukZuo2brRhIjMQTVe(yfZ)2$LDBHcE)I
zS}joCrBvnAFV&}qoIl)Dt2M*AA+}*nkszP?d9~x<Nc3DHcUz{YAK|7eii|#C_fLF9
zeSVa84pYkgejya#aWoP!ijTR{@iL*HSFuk+j7Cdd_w6>rvM0X(N{5S0PGQKwu(M?=
zoF*rm%eXiB`TpaijJ>zy?)th@3$7<_34xzXb<~yHnq~3ppAE5w3jzP-Zz+>)NeS(#
zKVNwa-~GNzHlfLA*M|DCJfY8l|1z`o(UtIrY1MbiGxf5<bM!2q3L?eF+tS+(hF#o0
zO_WZU|15tccB0(zutt$g==0-(h*D%~zu>UBo+7wV)J4~XzUjS6Sc%Pip-Pn%AvE9|
z53<(ucgti?9tRchhfaK)0H(8@UwLwDA1CR2*i%kz$muy=dhzsyd(;o+S2Ue>s}>IK
zAR8QiwaY7>;C+15p`n+lJOFZE{bEkdU+S>IF!6%)i0blTqUXg!3nh;E_R~6ahg=D7
z^oC?9&+~OPGM-x%_t!}|W>@E})f*&b@n`joI%kUPIOg|_yot;1m7L73M<w|*?4VLc
zCzRd&->MEj?sexobk#5F%X$>2u=!LeDMo~-?|*&ZgD^NGI{&t;)L#qYG5LMqY}Hcw
zi8Aj|DsB<nOTq<<^6q2FNs!KqVcQUdKa82muxA46<BL5(zXkv1$^80%|MV}<2%y;3
z6DaxBKUdn9cLc!5k4+)G7MSsk0UH7XyNRC<^re;PysyIgA9_*#l2dbfPWkc%dhvPB
zVeQAX**M|;_k{|zw8ZrBBGkL$CiYio6A7&ZrVkrH9PwRYG&hU#wqh)Lc1-l$l5U<6
zFq%|&E{O5^|6}jX!>L@`{_%)tm8cd<GNb`f5<*(0P%_U$rlia>8P+1QsDu;>v9L1F
zb7Uwg^E^)>V`QE}_?>s{=Xu_}pWXf*?{WP8eE%y8Yq{_1zOM89oS(sc9!)}K&KYAS
z#J@NQZcQ(G#(+n`OK$!4-kf#axwmOZ+DHL-mE8F}u@`=G^G))}i;)%$s@8w|ywI4Y
zk&=-dvuLpU{6~S+e3BgDDILRL-)w=wYAjY6ZBh%E!BRIg5O+XcT%R@qzqtLTSXUe9
zfvmt1FFfjtg<Mw`(5B3H@7bS4PN&7Pgi%DU!8oTFq35PWv_0G39x-CZaQC{kECQFL
zf89Q>(;V}_W9f#8_(e9-aXR~8DY|Ts9gb&=^TT^883J`+6)I44e=&^83&0199vc)2
zpIpoV9n)kPn6D82n|qm!f_s_9U*-DupXBfBOG=g@xH@Kg{;u}?{nz+z!JQ3wvnc5F
z_m}){wymEFjFPl<&mSx8UyJ$w@n6#q$j!4Gvzz+;r}?|unTG5&)!%*n_g`y%NKQ#A
z@TGR7{nzz1ogxKfGarE155>+i=5R}}s^`y*{rf58)W1J>m)1b)70bU?_n#L4pDJ9V
zhP*Ev0>54l5I5}3GIDN_qVK6WL$HM5pq#OwN#jAUJqKMN)X@@-g7Bs?;2Na^)hWD7
zM4ZLS3m9*8Ku1RT$FVcI0W+uoUQXP)z)0=&<Q=^M)(H5&nYbSdRRRFeS_~BD=Uer4
zhH2k9It@|3&@*x8ax2zPEi#O49_8Nun};+>4nAh6_W4(|zdmL@9inb4g>JW?3A5Sm
zHYhuQl#L0tHPAEMz5B{eIzIaJ1YYdHz7klQ0>Dh#gbBn9;-!dq-EXeW9Q#os==F(}
z2XKsO!1$d&=lJPNnzM7SSz-@hmX)Owh-(`#q+fs9%w;9bKj;E_*cq_g%m<ZgH%O^$
zz;S;sYddIWKQ9wLALBDsvO3U{RFIx$5i)l-AZ6V)=kM1-^Aj?-gT&f|q@S<L>kffa
zDK)XYWWHjbd_oM^=HMtJ*e&B0kuB563r6cbNNhAzlA9Iz2wPbT1c@k{dL6Puqzooj
zkkXS6mTWc7V0ZQ5vbrcZtL_|=(6%XCPXOk&bpp>6g7>T_l=zG7B?ZQyUpla)jL`3(
zb`;G60-LTcOU&m@;1{Mrdt1=1awF7sxppxAP^NMjLEYo+@c!RC$F9h89KE8D{PXpQ
zn0F);jD~Y1OM#_;R>&tfA`EiMh)Ig=kMss5H6i}qEJ~WB2snQnm^sokSs6Z&pY23p
zgVTjBO!A>4bfTF~B>XD6K}=UDNjr4{kDGJy7*+Vtkd=U^**CEH%fr*@uv=)#D_((3
z@OHX|?0L=wSTx$I=0+jpf4^=bp6wEUm!KB;>-8AzeS9BT40AJ-P(g4p$VhI8S&Y9`
zyH%YOuDSPP(?J*D(OLq_G8?fj0@t6?SN9^fZ)CJ27*vJO{QVoyT)P2vj?Od2U#E-L
z9-8m4G-d#TrED7+fO>2dPNzO--*o5YU^qh!xq-Ux*S>@@s87J2PR%BXLt=9wM_C)_
zfL5XE8rk6KTLtr_LpC~hgJ6>O;%n}6|AfeX$2R%rsf*D~5&U)1fsAIrw@3u)d(G;g
zr>Fa9u*^3gOHmO|rdTqp$a>YlA`!9K>8Is{d}ZiX!;t4h01QBswnH^^YLMEJhblE;
zx-H~ojmd0(ku%tyzD#In=Y`wX`{+#c5vG;s+B3#iy43u@VAZju<-iV*TDl?h!3=Sc
zOU`hrhkT|+fo{Xy(C=-f0dKV?sL?%RsZFgVqSygqr-xsGJo0#LHaz)#8oE~MGwiYv
z_R3Z_Cej?5(Rqt{Ap4#HFn6z&R006v4Vyn)q>9EK2B|h03t3%m5av$Y->W(E18zY+
za3y;XZ6^fR<XM5rVR)n4Wh5Z|w76LQSFlU$fp8j0LTrlwn3|X~P-Or%)(L4j=8Cz8
zGJ@>ulsUR&ipG)(avod8xK{y2G6RoME<nz{7>E20jk-Raw)80mO#=<n>w%PSDo|&;
ztqKS3*pw_tPo8hrG|hVl8%95!_U&Maw|6_$B}8MNhl3V#0Q3{9K#}bM%^~eZ0X#dH
z<rM(_k^l$m1G{FwdRKb4MKI9uvrxljO6OVt7qUirjd#G`DM*RA1Zo1h-YflW7VhWE
zfZSAplcl%`l&A{6rW8MfDEKLenRzxfB?xj<`t;prOu&;|`h8QNrhp?Z$ts1QGMHXA
z1x~JcNY=4y+fEAgH!vD*S?zhbxh?xv`o@+AT_MMngoUURt|HODJ!Ee~Nmai(^v8A@
zAUgzWG#5~|K9GqVB;AMN;z-gkZK0l&IqswuTUb$yfGvR_9H;RBM&7z5q^BPkCYQgE
z!aPi`Yk2|jz(8P&4&SQB9@nToHw$s@I(Qy6^tGSj0OOHN&hJUhz7@MrYm@!_=1)DY
zYT7E0Tsv9{$HcJm9n=u5F5lPyw10F3n$==P)sQOD4eV$p39lR_c4?B73&x*rA0LM#
z+<ji0<m2eDh>Ld>C@qnpVr&MQMwvy~=<F2(u=4x{Ziam1%g_7^M7yCAl$>0mt!Tf?
z`6FIw@X!q|BT3)=*|3C#igd?|rqaQ)c6B63)Ynbzr79~Op3CVQ&|iBH-ZFA{+#&Rz
zK}8{p#i-T{&!GXt6$`L{5%g_5d*CC$##ah}k~=!fr2j%U(3>KCtk%BIS;tAIuC1}}
zCQ%|818CZz-h7EGDKCKRB$$^A5BLC}6xxpFhE}Ooj(Y4osm(8@R9Dj6RGQUR1TDL>
zlnPzX%$L(r;c%~&17+V}TG?f?g;Q!az;wy4Yzu+_FXnUOUH9SrJDLKRt`wVgUphbv
zizD9|a!v8>!5c%sBKIjsF&v1n$T+$oME(WE9onooS>m`2ODY4mv$e)*n}AVV>j1x-
z=F!&}t*(IC(vNg26UCZS465H1*JuW))ui#~oPxq0d0XK(=2SM$4h*6Xc*4G1xsq1g
zQDlIiWgh8GYM$v$n%eD(?5|uw(n-I+SzQA61Q4x$;%KaJbE)E*!+`(u8|pK;=(^GD
zW5sisuPxNh#e0bQ%eQ@3ILPcT^Sffq>jdfcZuGuC>J&Z|5^7f|rdEFuGWr=H^afLW
zu5c)A+?s<ehg(OM`ruWj0)LU$2e;tB3&N5St7Sa>v3#$qJt>30faELi&;=fg@S$BF
zCQaJ4x=2Fva*ujIib*4BwHZ?@H)<4CEo1&zWR%8L>0B}Rxt*XclftOnVuy~Oe*N0@
z;Ghb$t4qqzg}e1H4}#m2LNLXFIH>Bm7?+^L+F$kbQP2lEE3m;0rdU70tj#fZgh_<5
zO5<oFu=lFVy`{j5T5^;vLkg3k<_OlwGEAeW45l}aMbhZkDOI(rg4YeU7hhEGz8~x-
zC2@50P*>G%Y7v^+VUp*QP<7S!Uh)`@4F!Yj7TA9ndmmDqWTO_9VjA)X6gg@#({sz>
zq%QUA4{t}W9AFX=)G_oTKS8bFJCm2$D4qbse4SCY7W|?s`)M14^lL;$a_XD@p=NnA
zd{cRJi8pOk^NpAn&M(Gj@ZI|GP>I6)MT%xeEQKx`a}oDqOP54cQicP4N1~rs9DF36
zMGk>V{)`zB!^0V_dUu(wQQ>N!;E8_g?xMH1ONL=A$!D4^z#g6UnjxcoCE24^218{f
zy(N`17eGig;O=BvR))iYYSVrG{Nlk=Y|Qv`dEEKhxKHqB&^G{Ew?+GIl=^%<XDp%l
z(N~WYi)*-iIW=oN<6_KhJJ@2RnaG}CuXcKmHOn%mUsNeibANcV{cf#zRZHmx7%EeD
z#mw%*^jdtjlby*FaXeD*6wRcze{B(5C8RL5cgJ-!2h*1p=H1@-x!ymw7`%6>F;-=&
z^1*Vrd-7~7Z`&4^^@|1L>{aa)lS8j#Ve>4KHa?sm+pQPG4}0jG!R)=q-vvcZ{JCEK
zPi2m#Z)QKi^+%1@Oh$&~E2&Q0IGcPD17zl1QTDYQ!QL1Y?MZrO4lc4FtXg8VOv1kR
z4@Wnh)^B>1MUat<ILvcUOBxOVnWJG}H^-5Am;n2%sgFzelLljS^c~@u^)tLSD}bWm
zkc*8mjAb#j-%QOY<~;mVS*$Mc(G#@wLwa&WN3JMM`ySYm*zkf_osnu8v|Jf&^yZ{W
zaL^Al=~+yi;VALcm=%#$xQz5!k^hwDM4TaB@l@od^LRcNEtUh#*(P+Bi`X3REs77z
zt5u%t=GS}S+;=Wt!NRTVW5H7JF3GoQldXy0>U-r@zOvo*oxb{BzQe*gisN0%P|(KS
z(ig|RwO&BeX}G96rR*|~jgJNFqc%2&e{?1Iju5It)vA*Noi5-26z+~Yh4lq?G)g2w
z$IljuF14&kJnfe;YxwpYtFm{?r%{H)MF<fu2G6<{`S<z<4biW&TNT;hyP%ADiWzgG
zoE+CT?!8v^1`6fM2{ei{p~WB_=={UqQEZ%VLfx4buE?##4iRU?YvNU=*jyDBaJsCk
z5$DXbPK&0-Zg@mVN*WNa`;9KClo3LgqE4*$h{*h|zhmbTqshwJXYogq<w!b4M#3-L
z4|HuNZ!vY+oA3Dod^Anbu_>!(CrVX3Ojpfp*$8EMLsgtt6Pe@qdF5y0g4fYuDwDE(
z>iMQS;I)>YA|EFx7&HISqq>ReHo8~)y=XwKbH&~>R~Z75y5o%3)m{ZFjxb7J=Pb}G
z(Kl(El)&)c7No`nN}OjHOX=oyoNP9Qgn>s`)G!Z9$y!W}wGx8jlVjBNLW$i)bh-q%
zXYmUDQM&S7XOoDwDO0gZSr;*Z5+xW~v!t$mmLZ{uSBJTCs!zmHedwciQ!}kgbmA7A
z(mr*Q*(`*pR5ZQ0k(6qW=G1kXZbcm(fP6An*2V9+{I}wzLmKor;|a~KU!$`jdfqo!
zOFiti>1B*ffUk}vTAiRG*sqJu$uNH;vlik$`%Kz1o&}!EkBt|%<;aUV$_K%Uc_D>H
zB-4LWaeTKdRz;SEhpKM$tH9MkW={`Eed2-;+AVEnE!VzG6}^G|JqSL!ub!yB&BpWZ
z6T;pL#xn1)d5lZJU(D4HQHO*pUNZ09L45B+H#gX?kBhS{CEGQ~>`%4t%lo`+hE0bO
z1$x1!496F3JU1id!+Aeqr`qEBHj~T*Y=^`yZ!Ndw7A)-YLtOh~Eh&yWDs{RSofi?j
zjadv`uf`SGnCMe8WBNB%rte}--L_q7mKra;rHF)C^g&9W88AuH*U>>A1@=uN?h2Pc
zg9GQuXjoyo-*UaokBfvuJ=l{ewq;DCeAGVSifmR$eZTG)GfpfSn@OeRBvKq|NIW4x
zeJw)qtOdq9WF1{KZggYGOsqlDcs%xb$yOVe%;)zOZ{p7NI&(PbRJL&!_W-xIq4M=A
zm^DWPEvh?SF14#>_48Ss78hG;sp0LjhB(TERUFi$;k{z&bx4XTl1gE9=RG-ny`|32
zJ^bOm_NQGk5n~VQRHulF>E>rM-o&jxyg#m5)49QuS+<d~{xESBj7EErM$*=kTD}jl
zpLtNNJ*O!ldXVsRTk8i{)rV79%dZKq++)(`BTQXdcEEheKzJ7g%@X@)OMJ1f#g^vv
zzGo^Cj(7%Aq%<ThAE=VIexczx)>a$oS6#VALS0(L9J!WcrB?QU8q<xf{W9-yszYYc
z2UI~dhM9AM7+$-@bnm9N966>E){X7&OUD)&0-9o*jjhU&r#Y^9BKN>=dNeT>18CC9
z_FDWxaZE!<<M)sx%16?9>tT8Wj=PQqcZ8KfLCC6<KqHUi%HN7Ddc|VM_(r7%qD-FW
z_|sERVB62?f~LU2u>@&40NJDWc0&!S=d-{0()PiZ#ZQs&DnD5<!14vmPAGjWAxEL0
zRrYLn(C5RX=TI~ollK$eX{w8h9dDKi&ke$CRHIkdp-*g($wjf!&18zzgDlA$Q>&=Y
zb2{DrHf>7S`KdUo7b=PQA~Piow5w`IR-q8s@AbNb>d;y2YuRIvJ2<w7gtl8PYTeZ6
zQTjg7mEh(IevEbhn>1(LpEp%0`PjzBwNysDMV7P0G2Yv1d$+Et+!z*IIcB~I+%xev
zfy+8wfg5&&L!W)Aj?s~h)NC!^WL<=O+g>PEqc$}H^b<P_I4F~LA?}qLGZUPYZ@~4p
z8-h-b1IDrOaGz2iG`?XA6Vus$y&YeV#Cs&NFR!Ql8-@BSd{#{lV6jaF!>>}1Azmz5
zT#R7e4n|2AAPsgzjx}y73|PRv$^^*o7@rc5AKU8PZoa8Nqr^7B!5)9VADDRBtLrZg
zF}lF;NKX6!ej00D$q@BGIz!fX5AE%KXyYm#dZTume~=0Ime7)jW@<OR&cLeVq^(Dz
zbS7qo_COoE_J*QS`1$f5t%y578>I?J)I2^XdBbifjnBMAEH>?nZ#Sb&)FKu}L%Yo*
z`<hgmAs{ye5!s<)qx)H6q|Y7`^(ZNY<4AobdPN^P_*~NHP}~1h@ERte3vsNH(ol(`
z<5WEpw@09m--nCVpB5Gwd2&&%3$LhdQGDqXJwDDv2(XUPDF#u+)biG?%EV_>tkU>=
ze{NLWS0=!y9@6l2I)~~o=Tng;n1tt9JS-F4{^469J>tN=tkpMncj1EUy28P^I<~(@
z`2b1cC*3yw&(=xeg~4rH#W)#VKeaF<me;9cAqC}_L+IwF<J(}rnm8bGWz^2JV%uc0
z{jqs(8JO*%u6&GOBAkVY*n^EabIj`Ez1jZi-3~1<43AjT=QxWE`B+{4SPJt@E#RK!
z@$c=W!G62g2ogD;Hcwqz&XOH%W0t{q-yChDKQmY>qN(j<%|uq8b_25xoiY)6G0fsF
zIeVw^B8uB5sCP?a;!Q`}$nJJlDMBlRO|xfevqtWR&_pH<A-Pm1E(HXFB!-hZ-g%s@
z0q?fLV%S-L5CC60&k=o0(lZ<Uf3;D+4mbfP&5i;lhJ99_68B`7=f=$;^T=;=s=Mh2
z#zRyUV5s|Q%_DyKuy8i);@oZksVp{L&>tpdN`r^9$s3B#@XRkv)e~V1%*@nVo0dBB
zoNwbSbuJXVILA)yyR=&_1G?Cd79|^qbuqj1RyHN5{`&f1YJFdam1M8ysj1j#Uz5e%
z3~_ToqUEfaE#DbYS31WxwTlKhP&mx=UmuWvL(_}oe4P9Led3XLfmU%YIH*xY#YOHt
z2Je%CIsguE)GdGqFs@H}1uH(+vbmMN9;<yu9UhipN@isqb6_nmgT=3gwluRkS1%B_
zh)8VCq;vcpibj#8`s&0(f!N|taaEniQ0watv998k2njhj=+Ke%q|4-MnlWSW4%F`-
z-x3#N#+?&XqkYn{fU6y**~7IgJ{Cr7zF_h8+)RO<c~s|W$@9L{=rzEiloHkP>FBE}
zN}421@T4eCGa|#2U_}E4Z+y+j1ZD>+sW=ervE6f;)MbF3ipcou!6`v;+To{kt)y8C
zRl@AO?{AdIPM&dO0^>-NJfKTyxKghK+dc@Dcu-UaK*UjCAMq{*llZTe_EyOl2Q<O~
zS1p6DzjBqb=Y{+Ezz6q>H8;`S!9js9R)mUQ5I|kZYWq3Up!{0w?5GOIO0Y<xiuZl_
z*tsAoh7CAQ*w(TI2Af*HD+F(@m;5mb`R_yJue;ZM31d)C+Bm|QdiCPn#(K^N((H&!
zeH?c|_X$Lcn)A7BANgUltSOjUQ$Ab|@d~#>2f_ewA`U_7%t4EXz@M5&?W@0ORUD2g
zB;~EyLEjM^eJYvL`4ty%Gvwy7&h$?T5obj7#GYaF!)7;pJYVe)jAx$vQ8t}XjArD{
zZ&zu7rifV&=Gxi>qF9);Vrv;BFFo!pw1E=u!)Tmx&+^d?MWP3XAu)LUi_?O!wIG}8
z6Tp?vk*T1+i7b3L%!c3f%)fR6?zbPy7l;|6AAsdjQz-fJs+Bl_u;0h*Diir;`vGk6
zkgfH^D+dL^!u}QtCv4PPX>hf9Pq9>S(p(iqFdr2sXb4X)jnI#Uv0)U1f_^ohuXs%j
z0x08!r(5H1I6}CzG)Br{{3(<pL0H+LYq3v|+^P1+UXvVVr9MQCTKUi<(lY1zGGOjw
z0d&a)=`>1fvjc4SM=0)}Fv$UOA6Pkg4UR_t@&bTWgr7<baYA`>oGb=5p5HcCk4;LZ
za`!p}x$Lt#yY?MmET>84cFpf0Q`%br1s&P#6Un#@zWm{$jGED3U5J(bR(OS7jq>7h
z^oB*i%SqL*zMG>V9IAQ377HA0MDRTI;fYafU>Xb5PuvALDMU6s&*Dro^k6>_TV%(e
z)H+`AFk_}Se;-K|2-vtm8?5Endi{(PVjx0E%M95+E#QA(ugIDbve9gsx3IMc`zqB|
zMDMMVa~>h?v8Sl&WG6S(3w+C@oeqp)MccN`u!+rfTEE>RQs`dF*lbc?`gb#0O%5i_
zCQO(t5XlbysPfcx>QHZ}Jhuu^{7Xe{a}4U~`lQu3>j{J-v?Mz%n$ich+VzPdeAMW0
ziu(Eo*5h91mtullV=4j9uw0tI?;68qvtzIMJ!pHJ?LfrNe6$lgGp-q%(r>F!bE#m!
zSr?|1E_XtMeD5$hq9&$TlWn&b=Xc${6A2LneSm=5FfH;>-+Jk{RWFl#fsn(S8$j`;
zUPkZYaT@@OJ7Xm&mgd#jbGb&@6{%IVcDA<~B#)1`iQfTqduI#O%XY&~CFy=L^@-;7
zmc%oCmbn40E*CBI7V(%m83&X}>v}RKOyA9H$&j!fQZml!r*sBp5SgwnXWLy%$}yVL
zV1Uyx1jzfP=QV%<Kgd+veY~+0pyFX1^Oa_iBx9hTNnjru#;oCdxRz)lTqzqgw2a6t
zI#SdHDX4wODVS+nM0k$&%HEmy@Z7#OSWr$|vCgW`tiwzQuk;oRKt;V|(r|<tZqluD
z<xA|zrTIWA2EJ+$!#fzft!W=dINa5?3tKxs-%|Vudz<^1plwWYxl&(XUqayU`)*rC
zYR`zE1Li4x{qg?;w60E%wAXik*)7t2Y-9JzrnxTDj|<eSe$KJmgNbs~R$SYRdVK(e
zmAYO#*MC<;?#PB81(VA=0NeAk!e`Hz+n5vVHQjX~ysSZnH-P#u)7e<U{6Uh22LZ8}
zYMwJ~(S9C^E3e&sxkT7aVMdpYuyBXBBh)E9W9>KCQ^(o-A}A&rB*%tzmfz{|Jf5@3
zSyk4q+$^eMFDzf=u#?>S%BYrEOaRJAPV&ww&D(WKVgb~LK8s@>W%`;DxTH)~-PSna
z=JnK>lXHb`nNVI2mt{;8-6#qRZC<M~F|W2z+lIxhdbIrjf(yFOKm6V3S4j)1cO&}F
zUr0<dJNYuP&_i&|%tG>^B#H2rZ3go;xG2ofG%E3k8~OH2?FfjGTw=an57sHhcFi}8
zSn()YtP+btRhqvP2^;fF75YbN%@@^IIxuPCjoeU6U}X>09#*t3Y{|x56$yCAD4@Da
zL3DYO(yL~mUS@D_>;ssD_x~i<<h4={bG~Jy-Bx|6;&qhp>4Jp;^OY9x_I#S3(Y%j2
zpTp-Hsx;Bb4)@*h`6_J2emv0<_w9I;hs%IaaGvZsy1!lzO&2nO5VtCVF$39V%u(gu
zhlSS=Puu?d@7L_Z=L7XOuNkf)9aW?1j5sSFxe-=Pd?^?udH;iq!TtlNcv{_BZkg<n
z&O12lf%t-%ys12aiJPXn`0Ouk$3;D`lxLE!Bn6`7)CV|iU%Pe1IU`89-@GDL1Ydyt
z&_qKCc9ORQrTyu+)4;ZZb@D5*qK7jz?a<VvOSXrZi9eU%!OH^6E6UbFUxYc<l(!-`
zl<>*4s1Db!#TvV`hp*u8qnA~yMfCH66oZcy1ay7<94C-SlR`J3ik4RI3!op6fo1pN
zyNSz_%odf+w`mVTze{XA*l@t_0tT>t+m;$Gy7Q~B>YEmsMyk{5_~8zjDW;fhRWIS=
zl3j7L9}#Qe^~vJ*tujO0E(zIE$y@4-WNLaF>^{SyD--aUsaE2Om1*?FKi;~5JwT<+
zjCp3w!{r>f;MU<kYk&`X!x*olAv>4}5EQ@8?ApaBPn|&yN3Ne_kpDi+e-cE<Nc{We
z5BxC!R}mBaVbRWtcS6rA^GNQPos5wndDkytJ}Nkq$?~w`7)qRdIi~sMY?5=)X4qzL
z;`e>T3mv0^sEa2hT{trvg^Jj#L%4E1>NSp>&>zi(?x5=Y64O`=bEcDiuO=H;t@7r{
zpveC01e5A3g`*dqH322@Zm7J8T6liuc3Y8}`Wa!*twiI=ev^dab`PC)jW1Hu4!eUn
z{V_6o7VN$V#4et_sa4O4vHqc@AoRsJWK+w$PvNaX9k<_<yi=S&>}t$vlCb(#vd3~X
zU4DSXdp*3d8lhLT>`uFYT9;dzS=7`KDpy>H?tv)J{)}Ub8|;OQV+DyWp83b|-Cwp9
ztZrC|&q!Et6=t@$rrNX`m!QoZconRPBPmg|It)1krFExmC?e=Tb(QMZiQ`^xltgd|
zhh?bh8E~p%&xaYa)=5-x<(JFuA$eY<emsdQUz;HZ!&jVKVbj5FcP!oftqS*ufKo^H
zN_0kW#A;**Ftpa7l}}H`HyU&mW-RqfRb89DEH#_$8t}UfVSWcV5Uq#)dLEjRl756@
zF2)p1@^`oSkFTX1WTDieqOtkwWUToJGM4w%`NH`318)Q=OgaWE1gDW&<-dMQk^_l6
z@$G<fUk{w@mS9q3w9x_PGx(+%@GAzxrIaDQVJDAU*x&s{2*%u=3bSgrvGK~k{$1A;
zSstH2%<V0~>?L7B2d31~^WAW0A}s2B)0Pt@<DP#{HTVf#H#;JXk%8j)+kg3{2C;v!
zct%4Vq0P>P=6^i>F8S|TNcfC=SKtE3XDV<U{D*J)4@CaQzr-#c0(Q|8&d9jmg6aSD
zGynHy|8+_JdKsHvlEWl__a@k@qybqMLE8Ao<7{XDa-0A5zppugP^_+S;1{8I7pc>!
zEHJ9-1=RO082@-XLHU}2@f}Iv;<5X3_3dPwKiA{9G`6547b@sL%#!9*lH~4%)OKJY
z@Nz}GxBy~Q=8&&qy>9C{gO@Zi2dfn|Yy2H;;#Yqg(+JO8=TUBof3eQ^$f1=!j-V<?
zjwys0g^kBfk7}WcS7rzLAZkbJOqC56MHH{e<0yay5#9t|bsNsv8GsK8!Y|&{N!3Va
zxN#w(-W4&`iUux)oGZc{pYF;eKuN})n}Lb8#f4-tcEwBI(6cTYv#fu4VT?J1?Y}Pe
zpPp1c9a#1O1yh1{zW}R<7f7K+D^13D**Pses{%!<_Z_(R6`guum=9g595J1LVhpgJ
zT!bqGc*)TtfMeacyi|YD3iuX46^k*#!gxmj2}?Z#$V#vCWUOB=6ydznQyYNa30K(Z
zZI3YTWPNQ*?3>Okk8bo5LjB_-Cq_iWGsd18@ch@cj9K4{-Cmzqg&*#P3uFToho{tT
zB-^<F6ga)WOu372)1?R&I6}X=rk)940(|1{Hhj-7y52Zf^m+yCi}WAdzX>7%;)f2;
z&Avf7Hk(6aTnC>H{B?i&H&y{Jc%420gOf*UCqMr2Rj=BEER$*_=O{_(3e6tggPq97
zt(b8Af3VsAbfy1uANVwoYo0$YX!Glu-yDKYIv4Ko9)C!=JS!);37AJW%)RbH)N3K=
zk=liwrf!2~ybvyDT_;2~ds{~&tiU)tduMyCe0U3Hj$MdDLmTn)Tj<HZA^e*r{?H_8
zM_mwO?qiyUYMM&*3sR}A0?OGVv5)(9YFf@?tGpKJ0vMv}K*fZbrO*5USbN1aw}okm
zAn@UIXvs0%-1#vf>I~RZFR)Z@GhA01(?OONVz>lIRTHvqC$qGtVJdyb9X4;dF&7vK
zKX7<)PR}wvR3it$yVVrr?TP_;GFg2K&!Hvgr&htot*1Fw(hAl|Aq3rXFM0g9og5#Q
z^HYlQX_|9v0Wje_DEe7mAux_H!qiB*nEt#HK9#XD8ZX?jTm_PYY;yXWS)URO%xRQX
zc+A_`BSl<uaG<V;-GYal|H+%~RLxAzzR;;1a8l0&tm`9E=qlMj`mH|#*%W}sn_sdW
z?ZZ^-77Wl_@4caG&wxnS30qt`*p4|P(+1~MH(>EtwMd4l2F6ivj|8YZxE~)vgh8No
z?wuYO7F6<em|tedSKayy*{_CEqWm8-Hmaeh7FlLCw@B%OF1f*RhF+(2(gJn7P+-~~
zU>gkL&%|j{l4m_2oYw-}jE|`mo7sM|hrxy2b@ZS+*jH6L#f?;7vvON%xYYDb^V@Z+
z&22Ci@LT!;n?VJulZ}#R+%)e!FdTCK2487R3!J+bPAek*yOH0|1<T`opO5qpR2p(J
z52Nxqx?jJOnfqL7)uFDL52KL^^pJ|774RdeV)^iI5$OqHqk?Qlt|WQDMdTt&L0hWq
zvIc@INcZw+o+5G1rNH24rHPC2jn2$%6kY{gRz$_A8Yz+wpi}HbxGo9vdQdaZK)+5+
zasZEUTm5DR`qBe)r`B#@G>4R`z5{9cS!$&Nd9foSz6SnyLdjabL@X2e3c4CH5gaJz
z*_)FV<zUkc)4KJDhbpDQ=VKVym%WFCgDe_GUAM3Cm3DOcF2hRa@WTMJXB#H*j3VaQ
ze#*z$MP55}CEZ8)(4vl;m$wjQG>;ysW6X;th`>AynNa#TvsjRGBAGIgAkd>MO;(_#
zQK?=A{B5t*Fbsw=cwdHdNmxC&zmbLzUC<{sVYiXY%7Liq*oM@Mj?)F*xAs9Ig_h}D
zP19onHAoMwOBOz5Sru9jNU*E5$(`le1V6S&2_qB(_gtF-y{=;b!o3M+d7Z%@<;U3G
z>6l(~qpD?n^qg4;+uBRBk_;=FtXKAY%PX-Ksz#T{T{_Ni!DRC%p+_Lza0GQWjmFy)
z2tOBqFLccv7Kr4Ld7yLhu8`E63Mf16TB1Uj*hAtZ53&}f*qzo39R^?fX(u(<MfB0d
z8uJQF;Jc9nUbJq*!m6e^U5{ojpqi$X*GS0U`E$oFZ$&g}ZMiVcPdCRthyoOkl_z>r
z2yxL#wu!i4?8}|?j<f&{0bdL5dzbb9T4w%<P0*kXovlL#t~L<a1e=q{(bdZLC5a(h
zV4O!KUI;sssLu?acc=s{{0Rn>@n*|8keqeqiS<2z8oLoGqPjD=3_3)$S=K8R2vU94
zZ-)0=Mp*ZV$zBfv3X}T6xXF~#Py}E!?e}K`(1{0k&=Tz)3}*6dAHRywo+|+=-#0*G
z55yh|coWJP&gsg_;HAV?W>(%W#Wu47IlxZ@UBF9itn4_cHtL<}ZO%~f_mbo+`AJVu
z3udL#%4$585a!x&+4u8wU!e`9mbuyoSj(pvC0VbL{}_F3818)@RVVjK@PfmB@``1l
zMX=dTmt=Q<EwHJCSyY+wI;+$bjovt#BvKDfLhmDQ&y+U6*t%hA{Nm`A;tUd0aS^|Z
zHbzw`B_(+Ni+!g6B~0SzBewe>A83ZK^W&~*7m76b9J(8EaHvcI(L|AnAVDI5fl7(Z
z#9XZw=pc9&<^FTD%PK;&39anrAL*hVOl^VD)4mzW8O^Rp#`e#S)+3(+asj}oNBDas
zl|aHOd61A5S0Mh*zP6>g%P&KIB|PQqia4q~x%{~oX*ENKvXz$C8(1j3c`I+2)od^I
z7$k9xt@3lI83<mrrcv@z8#s(ihW!fMz?;>lOq!&8bTSH63Fzz#)}h43vqzF?vMO8|
zYSg<PN(@lAXXy~%2Et6j=afU!+x9Zi4C;Zi=z0u{i6q`=#P;?}t&xS^U|k0Lkv=4d
zheob(An^BsOep%qa_Q!oUH*j{f@I=5I$gFw3SwSVN4)~7WrpiGgZ&atP-k3avVDZr
z>75s`c<^B#wK?9{^(oI{v)-t9a`gjjI5(CoJ*zzEP^fm|$TTB%xujY~yRH}Wxmrfe
zKLd{1>PppMzlARy`EJWfvVw&`MNS^_V7mH5WRiBhJuH}_cd=mMAT=Aa_=UxTLno+1
zt}+SN_kEI-2=*(912Lqiinhax?p8_u;Gm4qO=p)Gwww5+<(QlA*N#=z!p<rf@?P>@
zu%9R!fs}tP@5;-=t}(pr$w5cL);wZqG4=+6)~uWTGX`)~t7R0QVB2%5D<8Se^gc)p
z_M4Y*Uwn60@I}AzWIjc7#X10oB~iTvijl{Aua;yq6AKoe8VLsVuQ}Z$NDH?LxeW2F
z&9JB2j~*Aq(_$p<j1ql1C$E`>!Z}7w-6(~T)gL7eNZCayxvLQ-hTmPApg+&VxPPWY
zvVI|D{-w0RbEkK_`asaM>H0!^E6t?!q&Ls(uq(@D5cCYhZ8`0{S#G_-rq@;@g+XF?
zT?Qo!YYHot_%^l?_2*)<b7wYCbvvtf&uBX>Ks7DJBpWF5Dm<V2{6E!OBxKv=_odwo
z|2WmWq|e3BL<eotuTMIvhg~%~Ma}_YkpfUQ!R#iM!-vOS5|4LQ(^6vPU{>ZnpBCD$
zZ<GVZ02daO$4}tjn7>t{B~zPkO5_a0UQMO>!Rc%nD&uw4mE&93O|-Z2e0}2m05n6I
z`2t5E5uvmUou)eO$mybzfAc5>3H6ob=4zSJv&Dt|xYzGYKhkP)70WLM^pcU_&~?=^
z9Q%e7MlaYCHkZp--{pg!!Dw0Nn$G)E__x@Td2%$%ub(TRNLot*4xa^EZ-$KH%vXSX
zroPmh`9vQMFYlRza{#>N=R|j&rhKU(1i-%%O|wj_e>>eXfA3PLf($hVQp%OCHOf%d
zTSma$5Yj>^eqo6T!X72<LGf{3OPr27?Z0YZ5~_ZJ*(i5td}?1ud1_OgUhCv$6PQyr
z->ces5`;doahUlikI&JOCbeBkPQ$9aRjDpl|FL4YM#;mUwG;8-m-1Q6nErd6!S=0)
zab;CLx(EjTbA$pjV2-|=T7W7%Nb~NM1Ah=mz>4xfo0z#y_f#103hPDWXcL0l+f&uK
zh07AgmwBE-B2RpZ{L<%0aP@o1&?!Tf*PragPnD41*RdRh*u;J=tD(gR8@<$biBVih
zN%kI*YWrXwAq&9+bKe&P7KjPS;2c|_mb*~=eF4JVFHnENC!0eBx8~wRbi=}cGwt5o
zXQXQNO%_ZQcBQ5a_j9dSj*YmJpAs*iy4wa}jtC2V%c_)9y{MUwkX~cA*4YEK+X&{&
z9fb+^S8R=6qSRHNnYZL4zDt|SZHf(&stL)7{@f`x@{t?B*njCKRhid~DZcZPu18X>
z6Dm!fj<kf4M!i-}nOVCw7rn@2uQpB@Ndx8x9SzklZnhWwm9{=1YR-g+z`=RZOZbAs
zWN$dYCbNX$*v5D;1Vu1tsj8%2Q3&TfCAS;MVpa4}9#_^bIwaCH;tVqRj~8y#K@D<c
zdIlb>GXyNJTw*#$2|ssk!l^Z7(#pJLAkd}AzvN++Q}lpK;EPdjLtE%r{}vnk2^x7>
zLMzh4R3h<nA;c)7*eR2go_nJ^xKX6DP(h5V>e2P{9U$FE(kuYDkG}u(y?AdO$gn#g
zo;BtDph<KGOZn{z6jKGzTTC#e)OM+plgI=xn7|oGKm;&k`6<FrE3UxZq`bP3mlQ4!
zF^YRHkoh3=llqNbjM-$2s#Q~A^D2<%PGcS$$IA0gf3YPf4j%K42)~39c~rQPtd&Z|
zxz3k9t{N2)<f1)vttS=xc#x(Z+^KGAyhR*3)+vJo%vw(_eSKvGoZ);pYa$d!j`iM`
zkivvI=8KErB6<lydrT3+4-V-ZVQdo_bnZx_T`Zn12MIpGj=n2tzQr(Bea^}))BG@*
zTq+~X;X;m)$q_TJXYOf0kE+r*S5jTtKcj4<#=OFxw-SL>5E6gKx8XtV<YWN(Hruok
zUr=!{gIgJA%jg9^Z%a_!u(G@Xp9?>Wu0+m^>)G9V{sFtAF?9yjMx<V-sv%ONdnK^P
z01KmAmF9-m>s$U3%AjE3CscH8tj_4#=E7hXV`55aX?m5P5M4NAHxdf!lN(2@6p2~!
zLMMiR%3xxzL^F-BX@A(MV>a5KYsbEap8c{<BRT~kRHpV9rd(><FT9SfHM*VleTk1(
z--$T#u`g`t2n7k;)E_0v@9cv%*UERU_<ThdoaH;AZ@&rK9sh}8a21{b8v8Y-xY~)R
z%l8@=l_%9>+dv{vT>x>n%^0UZ-jH-@3boCSYCrV=$|IghinP%Cn2T;DDhu!z_rsvh
zl7T}(pJ9Qw2uhLBg;uP7B+zZft}Sh_e~jyAF=W8Wb4=)2GLx@^yGwvA73a&|L`4G6
zt!Tw~3n<K8vj=#m>-GXga&21pCXD$iJsASj*OCy8yblR~&eVk}#={#dcydLk2Y7gm
z#z$TMW!><%coECIC+Q<IZ`a3vbnA$}gPcG1WLb6MU2YvxYTc_$+x|av#YS06;`&^b
z&RUMJk<tyEUZbK8z~hj}-TuNH{b@!h>IzMv015&&MPb^S`Qk0en;3S$nh%vU0IOue
z6l8{vM97YOb&%D@h7_AZnfv4boGU8MP2Wah>usZ9Om1*TFBx2XhrE-(hdg`-Aqc5Z
ztEX;Z83vU05QHx+w#+@AomO3w2Fa$@s474G!0Gi*;F48gXQDS?4-9H447uqjo1R9Y
z+50t5uxPKA$$x^CZK;;I&-EriV<YKlu;QBTGL9f1xR|WLM!k!!U)rjt5~$i?CkUqn
zZ6fXKg73XUYjS0OI28MXU+DJFyOIRLGwYG}h1D`Y>h+GV^X2_?3dz^UJ=%~XFbL*i
zVE=N+LPsNkWx=x~IwG$a)<dd-DQ>Y*w@YCftn&#mTR7RzoP}*pldN4OR;^Q;9TQWp
zuH+qay>@BwAPF16?aJao9X9GGGN68vOQK+=;+bQ7oZjqo+3w4SXLA%E1MA#=HLibO
zFT|*jr>Q7DS^Bk!?edD)_d#5YJm*~La_c0BY-)P7z1RtNE|863dp~kYy2O)_IQ>9i
z2Qo!443Ajpsi8+@bG(dKQs=mV5ZUJHFE;J3S~Cku6)7P-jd9ZrIp(Z3DZ?5IP488V
zo@Hp2Qun9C`T!>IFE5$F7Wb6QD}9#czd1}~^fl&qGRk%f|4y6g9=AoVAaKl{Dh@Lu
zBY)SbrQi*T`=)?%JjRCPm8|VcPm%}1@CTDH?bAB2ZxBImRM0%kJAHeW7-ynC{V3?^
z(Qnwivb2>N`joo_-Z!-@F%{R%52LQo>SCb^0el5z^se6uCT6qn5b6EtA&SA|_Ln7u
zES)P*w0^4iz8x-u5Lw0ujb1h`A9HU<qK|GLX+p=E0njvqgW8YI0Es{u0LBvp-^;Ij
z73~<;^VVWBJBUHO+LOzRodeL<1S^^xm+*lZtU4datxBLn_zk-C6X)XNhjT?fU7CX$
zf%e35(Ms^AqR792B_F+2(O6ipV(bwPN2h;r2S(e5q$wC}#ZDUr`>3!eurkADQH~bg
z1h&S-=`U;cyL}ffpkq|EY7<WdcEUh(5~|1m^pnRg0ELkPJ@7+e$4MG1C<lW{aq(Oo
z<Rr9*dO%jjMwsOjMz|m3fzSM<7ggTLlVI@Wh!5^)-|#x<tJui<_{@hv<-}x0a8bUm
zMC33h6AB1PH2e(UB2#&ks~U|ANdZ7+S4s$Z%?oHI5hxl?kgKKW@YVK{?&Es8m(<)*
zEJ_l`g>b*4z<L`0G_kQKQ9lqHF<=Vt7PT+0(tyRuoPMe4F9nB1=`0NJy*{;5s;j=r
z4E)}GV7VQl>CxdQq0PX5fr*wIL*4OUN|~*+6c`@iyV9mA6E7U-df4kldJRx0stIBD
z)o(6q^~RR5hAjZCy`;}uNw#JqFbTDH0GZ`Q&v_3O7;KJ&Y$oxj-K1V1R#f^i+F$#;
zar<qM=%sY~*G{+(6Y7yNu#-IDO}sw?TEM9M)ZRymd!uOuK)iy0B18~LUC7139z1S%
z3ZqP^xRRv2WG}P)JX^Y@jXjycOwmfKFGXY_PB?H)v#)dMwn6+%`-DYl#=#Z$qAnR7
z+z#MiG#f>rzLdt0JB^Dt%{7<aSfj%?SoH3EjZWNrxvlyzyV%ynvf{UdzMqJ}OY-rS
znoa7H|7LzXBo}GFg<%L^H}x1F7H|iLyM_k}<Kpy7Eur<27Km21o5}r~5#n=yHrcP(
z61WvcrG7W+WehG7G+MF6!)aE9;9t5VCe>N|{bOeNwmArx-bC4bIyyc~wRB%g@C@Nu
z2dJBFVJe;@ay$X-_j%iIF6F%<Q%9Ii^AQ-jv7Ei$Z{67IuxeMAc7?AyW{!)WN8qOh
z8NZ9vols-eSDq^X&QSzKb^U|q+EojNH|AA^ZYMP@Dd*QeU{5fv!Z7h=gAy&0qP;%x
zaHFuVeiD-i;~mTwolg}Hw8n?8F7e0<QV<Tb?V(}_DYPl6NYp6+GNaE6%uJf=OYS-B
zh*g`%a?4!(ZlP6h`Ht}O`sRDgIt!{u_LnZz>>;TzD;aN-j|}R*%cNC*nVdSV7s46B
z@-eG*RtA{kNv_RJSo49j8Jw{G(pNDn@brn*OU8V`b!6QB8q2lB@_fC$dz`f?ZHpPq
zdTaxkPV-Vf^euietvoXXSrhyOOgn&E(o4-TW|yf2d<zaUU+Yo@hsb)#gz>gsUkg*j
zAE+y=H`Gt2lAqYLI&YdZ=Q8*`yWjpp6~~G)p&TNMBmF#b4=e!47w-2_0(W|F{+;z3
zjpx(LBCtH|aHG+>PMKHoEZ_(=nB1qf2JM3-J8^@bn~5*itSte*tuiBxOW96(U6AY{
z(G?TKVundE>n`grJI(m|$<8|8Br>cC_cLvHkd?4_>IE@SxE5*TmG)3y^25kYD+mtz
zWQA`Ct~~D@lzbbTxt!eh(+Yv{4hXEr+m<6JtQ#bW!wfa0U`Dmsk(fm2ES1@cp1SV;
z!^snD$3k2U`Cr`_A5C%jLS&4fXuYWWgi)wfr23U>!{L8Cpg0gk@#B2i@Sg`*M3fjz
z{gWi$U*4J^Ug-ghvq44+wZrkD1;#?&0ua6a3hMaSA@`b4SQ?a)r@BrN&hjL%Gyo%O
z2GGPy_$OGt1e!O>S=ZiQz+ZvgvBp?!<)#-c`O#DqDW_l@PYu*bW|%8?D6KitXel<Z
zj+d8Oa^>D1!7rn@QSt0olMRX9wy0MeyW-!41i2ik@U+M%x<iRg#^UL$=2Sa;V**xR
z1T^Zsze?VNu3c2#ow|Q@oicD-)!4W4>=Po39F80IJ60j`m@+_rn-kC{pJ#*5$McFX
zjE>g(HkHW*;JseLpH0<8pA`8%^q|ttp)jB?l+;U@8UGC3xES0h=Ox1IdV!2`Eb>@w
zyVFn`gP&o71lBG69U*8Nnxq14m-Y)IENR~NQB0@mUgM(OrZ`kLnJ?aXf+#Cln>h~S
z_pKCRI3ttgCtb7SU)iHC-;wW1J%Bw~Yb)P*n;C!EsgzC!*tHB@GO-S+=1~|`+-*i*
zwPPxVa<~U!0Ki`MA?u){wGHTdR?%elUERxkCJ}*88IwNjBp*)+YKhyHGr4|l>H+Od
zxpNql8KsDM)N03tXE&@%Q>ir#Vsgwm9C4<&{?oNzH)&T@wRE<kXEhhTv;{^geq`z;
zHfEl#gF;!7=d(-!6^SFB3D@2f@vT<i?+eghz^ND!F{&_44Ltlyx-+yx_DuZUj_&5p
z15W<E17=sx1W>9E#2L!P1`j^l<lu>8iSuR^&cGOKg4t?<VfEP}8>c4PS=EP%URfZI
z&!}xsXCh~HVnQvjXC*Fmc`}^JK79_gpMxyHp8Xm>=%Iaa{3o=W4hRbc2q$D|m6SJ%
zZ+>NTe3H`mKJnfDSaI@0m%0rf&`c`RWS52ArcojWgytl1e~3+;T74m<KIR`3ztQfg
zgO(17XG%?>*8T*3G7Fduk&fiv$5Yq*CG-~_+8BM@*-l)`C-gq3eT`Eqw~I=_X*wNP
z$gbs(!9G!8R`-p#IU!so=a0X+;=wS+KC#g(6q1vqcx{iWdY4_6qPTc|P=7(vhKB3N
zG=%lFO<LFoX+Q6p>iS|4+gy~)#i?rFQk`H|&0pkM642|CbG%OW&VtIfjpj{W&cwN(
zEhoLxo1eamI5cr%5+#Dnbs54nfB}h{t<J=+8Xf<qnPYP|LM&Ki82#~!4q@wc3&i%m
z#AKID2=d_3F|hxTl%SyyD$|BF-!Q^kplkcg5~HT_Uz(ATOt1K?^Ze6p@T=86iWt9q
z3RP_R`3@0*Vye`n2!QGgtp8pqdcE852p3&}hew-Dnm;PyNBJ{TR2$5%x)1^qk4dt$
zX=)Rm#Smw{r2S8;>SvKA<_f)K-IDPi^F{_au{MmkTR$?vg(9djFCMtGsEIc4Z9K+r
zt`@4AF3|l*bLl-|X+U*cnD2>CqCEeJ=!Eiiz92AX$wNS$FZLP^Nzz-Z;3qKydu~3A
z6nZsu%;Me7)<HvULVwd%{H5$YzzKnySlC>^r#^9?WVa=dq%sN}h9CTW9segV%KS7;
zTkj}X;(z@GANR7e+=C9h+h^}#-dG`gejcdCK&oa9qf5@&lGMwwh);+wHM2D|TX(zF
z#w@!sqL4A3QA$68J+HUyQnw9thY>aSou;~_S0$?^_Cku68RWvdC!~CF2A^mG50%q6
zRrG%bwZHwBnhy|K9w+TlApM1+w33b2haZ-%s~d3t7uzJm8lB1CI*ho?3`<Al)e@8K
zrrq88rAfjTuhT6CiN*msV&wSlJAcss^Y-l`4P}Lk!8_%Z`#%TL|AZ#|`o5TxP=%k$
z&OZE4YyD4So1dTM|K9Dt*606S$^Yv1LRpnE#ou{sZS=qzKT8~a^-t`ozj=~YPBtgV
zAA1X(&h%*86hCEmzJ(@2_?ddm!U+CT|8)!gv^)IiRWDguE&G~NRzNLqvUgiyAj2-V
z$jpG|DdV}ft<wMIH<6HWsv80_lLy`eT{T}jENJGTb54FP@xT5xdNJT>>}3BKK26&5
zL;u?!kI07o<<8A@^zr}Ag@{=|&XgZc*Y^JD^8CMk#s9t8pRdsWUB<sGpZ{AR|GzE}
ze(ntbS&Rm;O9A<2ZJ%@st@YXl77HTzEyD86J8FS<6TZfD_J7<1)5iZYLUTw+dYGfi
zOsh=witHZ__jU<jVQ^<f^d%r10BG_)0Z4p&v&nPYm|G_=3+!`yVL);IGB7BtVB1Rz
zJNY=y_6hLi+E0V-Y85OjoYEjR5wRQ|pV>i#(l9z9XnO)lp;zSwcjzB|uh%m9o&$`g
z1l+2&;CxDIhI!&Ed)W=<ii1CE+Z6N>u&vc|Y>{#Xj$R+Y^QvP7CE9DGD|*cG!}|4v
zv8gh>#l62TgVatn4~f5nCI=y0&A@nq%Ox6m#Bfl;I+G9jYkmef7h+oU@q8{5#DN=a
zR*7yKA?8augmHhw@QTT8GPa&|XSJ}r2S{__u2Gwa)n^u1rCv<aza1=9vCdsQ88_^d
zox_#aEa42kRWQnG)HeWcb7xR<`OX`Fp28BwzIljO5F);S8<>p*Gy(qa42Amo<LC#{
zhY;bbM3&<FL*zRi--juF61gyye$<7?7~I0M=oOGQ1&lxK2da@b;#-B3#%WeAKY`Sz
z+_qhM_j-V(`l2)^LKE>)PS2_3G)e{Nxam{i@neCpn_bZVr}jh^PTw&6@bF;g+j}<C
z-Py)C+)2WNU8Gbed0gfV5&3koV&^^6FkjlFxs7W}`XXknujeO{;H?w&4@OQA#@^6(
zSzP?`UsUXWSC+Ljrt>UA@BwUO80~xj5(KCgVstOy*ATUx{#jcAsJuD-U8a%~dtp*e
zTY!jMtyaJFbjII!F|Q;3)3ObHH8oYr>k&eKLz}nfz#q{E9fHpM2=MFf>H_LqsrHt^
z$OZtTEVz6^hd*5llcIQ91;TL~D9GO;6UEgR<v-$w?AmUw1OAF<89`DPAMBw*5qFs{
ziw4Xl29wYZy#+KtvXuR{wq*<ZeS{H+WFnxs+Vd(<6y~+Lh=;;B=w?+P@(K;igC6=8
zjHs;Gr&;|NZy}!3U_(W}N!j_6$n-`GX4x~~vsB-|c`F1KnMl7Xc+b;k=D^Q1lzsoQ
z58&lKiJi1dh^fCVbKvH<U{+Mmqcc<|V%%Eq?ZE~000w((l<_-M97&=i0gux~RVT}V
zo>#Vwd8x?Bt~9t3JYK3bGXWPO2T83r?3pq*QT>t2xb~&koSTvU&zsDD%AJ^VWDFvd
zNy_7B5!Grre{Va8;-wxIE42m@S3Yq_j(`~$Z;VQ9n0b6QgOaU>!|=%YI<8!~fQ0OT
z!pxZwh(P%Ujm};+dQRJ<A-so+wpo1W;(>VOWr!(`;i(;fC*-*F9=T|P94;ifWhSTY
zoYC#4{^!fOL4|Dk7PuWtf#Itt{8iA3yTSmV!@HE!y~YHBOA~?_E(0?aa8#GY^t1_v
znK{Uf%32ty(gnfz#W{|4nBQIM4VGa#nm~mCG4I8>kBUHX4CWy{B>qmz2eC_vdGik1
zN2}twEAYDZF)MJW9{!A|j!il(P(z@btmPA)0is8kcIFvivEnFEhsf%0nPt(V8ZEAV
zhC9*Z40HcIm#HyFNXVvS#Mtgys9E9*LHD(=v95%PPZE>tkYQN6MZP%*JD)oY+5E#Y
z&kdT&dWMbY-+aN1H~uY+?KL;D?evp2Tx=3SY!)v=sI0?aO3O4tRQtwEvD+d~6?Kdw
zD}iPRLf{T1%&VI!Ha+*R2m~whY>sP*Py4}>>!@`34eRGv4<N_(mT#|Q^2}sQSGDB@
zMTBX#k#(&sLh@}FotB|j95azQ34MAUCLa5&iAy()1NbE-7l6Db&|zV}4oR<)&9;CJ
zP`6Q?8EjihSRPTf?r}k6B(J$+l6Np$%c|g*)s=IgwBNoCzpDpE8qwOB+uGtz0W5ET
z5`3p~`Z~k#gS6d?aa<%76ebJV7vb0d!CC7YdKh<SP8HE7-mMA#pN9^L6d@N^zB?V!
z@Z^7Z5PHc18~bGvEq)Lk*F5RLa7Darmzf(|KzE{<J@90dbfFSI?EYi04TfI_!=wWk
zXr<fR9(vs1%D#GsEg&b6*_ls%7C{Yw^L&<ptUbvUh-UUL56y#51S(d1D+b2XYkc36
z#3Gp9K(I78e=EKPuf2<7H!!Y<OQ+v)XPzt8u=Cdx@-{(gqM<wR%jMdHCeGEs!2A&w
zvGX1<<uvLs&8t9?HxRNh&qv$Y?VqCOi0AKsFmiPEY((IdBJkNUf9xhd?eXd8U7wcm
zXMJZXDW(WcgRMy!2r2PMjRsq9%3Qj2OVXQSdWC3_9=?rq>4H|AB(eAF`5Gh$dw{jp
z!(eR=!4pHo*;Z2e8ETxtOY;(bn!h%BHWkhw?hWnu5J{t*U^bavge{M;L8d~(&H*_|
zb95UrBn)GUgi;{)(f{}c_7IgYkYq@(WD<5v`g){zZtUC*?tC>th1u%TqCS2#A^Rpt
zqDI+|#X2LtQSc4diWIP_xYo>I;u3Uam6Awi;aGQTFkhBKW>zR<QOEQBW#@0*^%)H5
z@`Fatb_^wU1SNLfu`vzn?3hXp9s?J{C>R(a(c->eqt%&Uzp~QQHlghUdofG61zavT
z)2n9r#@q}ht}x{Gr>1A>pqO-7h?_Irs5;}6KY2MPfe{DS$qF$=+)S(Oh1?F+<L2!x
z{hnlp2>#pIjgX6Fyzx0*YGw`WpY{j2DGs`KtWB;%w)=7E59b7g^(FHkz2S0Gqts>7
zvsmU281DjK^!W*2#LBKcG-LQpTErLk^!!fQS*F@l<CFAyhk=4;X0QI)0e7kM?&%kZ
z!>4iwe_S!$*80`hxlvxAhx&aTeXjax8iy;_JcgumNa(f+O*7>HDa&WIGuyao)QGof
zB}%U=mu^M9_#84}x1D>*Ya;vIA!PO#a^NAAK=@Xx+p{0ots<7$wT^8^#mx!P9W$_?
zF976=>&FNYc;`Gf?*a#y+a*1DD^!KKW-+Aiv}`GTYEbi*<knM0$DJ!+iTOM+%ryg3
z2v<71%S`n8Qj<NEQ&b}E{JHo!#ZLbzbsVv*JSpJk2^-Y1CAq_Lb?f{eehBd>;QuhM
z!r|Ko8?h-;D}KF9EE~KBpx1KHo0W->TZ3~qw&YeXg5=ilyWeU~^RC@cb)jHf)=WvZ
z>>j!yzokcWV~cWf+Yu=7cQM~8_MPY3DrHCB01TV?MiTaPv_3hlv-~#FBljoC4zU!@
z_7R<;9^7D66vck^JGk{dK2VL765oNor9`9XN%HJ14A!?Mv%7Ek<C)hG)->Xd$QwCz
zUVhm)r=TOq8`wVt*YWWAkq($sFeg-ZP`WX^M{;>xMJG)#F-*~>SIGjLlbAhMD!+h*
ztM0t&%I-|>(<okc7iH7nCcp?Qf~t6LoLdNzfmECe0k`9A6Fly#MNP`k;BTM{nuA_5
zxMnLwoeZjSKn@R%eeBB#<@a@fHJkfb8@NjtzGycUprgjkdj-bR1Z~)D#g|Kb1*-eI
z_;-!Kmp>-$E<Z~z+LaBeie;Ls%@|Z>zRe(JxeHY_y<H5>hu{!gIdzie5W#Pj8s)=&
zlIYNCkTfl_Tv;@G14U#%66aprc;+5QA|o?yIhf;Va!klK#5!aswz&D;zK-X!W^UgG
z%RS#Dj(d0jv%V5Hi})K3+@satLRTmYnlDebNL4<e(hIy@v>YWq_^AEdGJOI_D;z2B
zrs!eebQ-=KSjPO-{Uxbd4U9ET2B}Ut<ZA|Kq^qXyy!@9JfMJA$!Un|&&L%jiBZWp&
zGs_sf6*l~sHfEDNDdTv)!Oj}lBdrKVGDS4?;TW9T_t`k(NoGOQ*9!U9$-hH5=Bu^R
zal|*EE2cm>iStsw6FBUKSKQZe#DRSa#I#jR=j|ji?{SLWuukDglwdDTaKt5JXeCr9
zZy1e8_;1JEj5WkN(B1~kN+6gebAL;<knmE*o6sKVT!L)qGa;9brdUo29DH$O1&(_e
z-*VfQl?;rk=X#t6j2+>|?;8y5aBinawow{5iHzwuecZn|nM<NFzc!sx=brNK_>@;F
z;)p$(toX1p&-`%lZ2M`asQ-_#uMCT--TRfH1q1}?7>1T^grSF&MnI%P0RfRxhAsu9
zhEC~_4rxgxhVB|#8bxF%>9ctD^Xzx;59dAS(}KA!Vy*SR@B1IW;B+r(ZU!<wrvaXZ
zK{~VJ!gdbbn@`!rAjRE%eZIS1CEU$8u9X_}vd3k=GtNz~UKek@T)s=P``JD67W{<8
za!g$4o@viZBWH@C<l8oE`g}Ai)uy7p1e{SZ;<DCrTO8QmW#5%bR9-z<D40ACi-}iC
z=QuVid;O_`Y0;cq(BRQ?9XlObJ)<8lRo8&!?EA6EgD=UY*w^1gruqHU%=SLs(W1sx
z)VX<`%T?1;|D@@<iU2OAsHahX<vKfew<TTCj7$%>irX%BMzFlfUXEoc1DacuZOXeB
z!PhN$`|;;yrQk2ughdv9sO7a1iK0#Ptui9z^;qf)jPbYnzR9+7cYu29XzSTk)03M3
zwkXA$4t4)ux;K5gGNAL&PqtaUZ~jQ~U0K7XYyqNG{W)ji9N?Srb-(@H^-@FIaNvll
z7I?3XtyO4EMMFonoqlIO=;&1b+*|9a)y?%qmGFT2yry2fHIsG>=$i89gF4SK<{Qhx
zbvHC$aO}+;qw(t{9i>~l+jaAY%r~)uXh!W^AhQ-kYYL;=Ey6E*%eH|WhR@US27_n3
z=^}<9Hk1Q$6DW*kY-YX*6mr~`c@=tyW_-zfLy&;<=Kz3K(LufA{eLMkwtE6UKQ<_O
zZLP7+6oXXuMwXSvuwOG<ujeVjq-)IAEs;09OQ`{KdQWWICt=}fBOei;WQ8B{f<Jt~
zK`3g6goWvd#?pg0cpL8og;`)>!K}4qDU4u5NQi$uSQ793op3DeJUgp}*BX&JWaax;
zAD@bOT)a7LZTRwHb9k_LD|M%>^@(AtM+wkMY|vDs_RGbC=#Vu=W%-|eS*8~Of7lJ9
z$r^*@SmKCIy#^2n^HQ;5KfRG5ND`f{yBB%t9SFfW0Cz%k_>%5=p7b>qUEoG^fQpg}
zaOP)G;h^B>+U1`jAu>i^9-AYxpqok>P~&A5K7~4$4;?mpg1#xx#%KNos=FKzm4sNf
zm+y4Rby#0M{?1`-<V45H8+VViJ{ONPR$8fxWFR<9Es2=4xGaX$KKI1U7ae#{)}OVq
zQD><s7f)CxGk|#+N31HL`e8rU)5&!;atucnqHQqbzM{q-V+ondRpNQ)iVc6CBBI`R
z!^-`$K%xIY@iO9;UXWlb#YJ4DB9VmiiA`)<XWMvcl$~*^vH6lB@L?S8cqoHPYWSoC
zc_Q23&+}V2tKw)RB<G%c$5A##5{z^qH|!B9JznY^On2IT{g8(fL>ruf_stnRqixqY
zc|UX{F4wrNKkR$20y#i3$I!tGC$ZfnZSI>A`9>b;jTGCiU+7nQ>63GI@0u-&DcT8?
zsqlG8Hf93S<%QNSEtL~azj4$g?R>9Q*q(&Q-czg_6_m63)SFP$5~+lW6Ml0CmnS9s
zS((+bNl4G9ZUF;n2$E<cm=Qh~h(9}K5KRE_d6rQ-Un^2OR*T>Ci-*tOKnZe9Jxs?b
zR8b_`JOexij{BzMq%-Sr!Uw%lv8UXOla{0e6Ar=5aL7#pvo$R|xAu~~$zs73IG&Z^
ziK!(rG(HxC_{;5pdf?9s2;VdBh|7R?$gub9YuIq)XX%|=ppsyfv1v&}vjd#OkTZ1~
zMH!Ldun|EwHu6%}W|TMogo$3H*gW~lYDO(`gQ7Vzc{k#s_KG}=jVdoJU8k$P5m10g
zl~!z3^LVn9o(t_qr5KqJMkMAauj7e>q63)uzu?)8)$Tlc#!oiEXid^{up_3-<3&4`
z_rvua@AdCc8R!Q>sttOSyT`fQ5vMylyq<Af;K%ph4glMcJ<r^BP)}_52SsW72Jq|j
zu~Oo{P3u0iU@RCq=J&_rrwsQ^l_t1QSj_l-Sx`8bj3lnNjO`eJZC7)Pg9_%;utyN=
z7KNk%4nd~mrn;IOxfF<DBQMkXEy8Cpv!p3{SiivndKj#)a>4@C)ddCKSGUrDk6JfH
zDyvodccuKcV6uJ=0}DtSa<X158r3itD~<VmQAjrQ*=+s;oEs>AC$2P_EybKcBGi+q
zGMMb9Z(?=y!b7I^V-9sp%ANHKbuYo3F$$oN1*cCsd$VcKSZSBM7lwDO?*L44BZo#S
zt83qV@2ps<6pN#kzn@p6EF}c&I|JBak~5L7V|-18ZY#pm1VGw;F>Y`|VNikr12yL-
zHEnB)pjg`P#Q6{K*xep%SN7qq=fBL33AIqBxg{R#?pd9IiH=f_wDj+j**5}7`CeQ7
zD4nW8^PSa5@meB_X5oYCC-9>oTsR`RmV+-|iK?_8n_7Olwjgw)P7RAYF(y3F!IQ*_
zAs>P>rO`}=MH@y#qFvg|vd#_16_#g_c=C)R&i~$3Of#<K2QOVfRxT<X{Qatl1r*2I
zPdM%%qCH}y9lVg+9!0?2v1()kGppQNq@!)atmf@?Gq902s~)l-IExT~F>pAq2P(>~
zYnlB(&zEl@LfM%>g(|#F>-Qa!nQTR|;pm7XT#p*Am<^>Bh2;@Sa6uz~eO~EjsxbVG
zhpLHN>xn(+J`h8X#3Sn13pATG!P|9vL{_%~O{m=D(7Z96sKMPuogf9F1D_R6O=n@6
zNKocX&CYzg7onHA9WetybdTJYVtjjDKkJTQRA`x-<nAeJS&(9%VB&QkUo(IU6}Xcr
zb5!W91^a@T|9VLs<4SjY3FFMJd#&!fXI!FupV8@vU6<q@qGsBrDXzvw0AcCO@)N14
z)2p}sSZvf=$2T8(SMa#avD9K>>>P}2wk)qnRjh{!WU+n<O=tM8Uq!@&wVTUR*VC^f
zneTvRgE!;wwrloWH}SYNu`=*xwO>(5|F#LlO%h&Zr7VpyyBx<NXTZR3`DT++k0lHN
z#j8*m#9<Uh;o&_3<znZzLHYZTEP_hc63m2TNYwi=w&K)Jnd`)rpfwO#jrv!?oW2jV
zZS6fY4zLztVeGH~RDab@y{XocBq}VR(zBZj=g*8{PY7FloTuQye}&}|+SOSH9!YWr
zuFC*SDbAit0%il>h9d4{VxmBMqFlC2`a}8w{8}<4d#dI8B`2vQDZkwjVN7rdwGB+r
z$9i=UkUE3E#MV5C)Njv8@^h+?7p#n^wn;SpqcWZqCK4^H^+&`7=-A%5V>$>o0lD<=
zK;lWIgZE-Bo$ibVtNNH{MJ;t3y|R>L(%+nSmjF7YWqx_df~$WVe(o6<6qiH}Hq+TR
z$_~sVNR500+IL%%hk)dt<o%4#TsT?cMq=$z*;uWA;T%d}>>6%j1LYRoQoyGx4DiEO
z!ZOSER$lgnB|N}lcA^%Oe=#@^0F9$;(Xeu_d=JBd9#85y^m0J`0Zo_r+)uk-SpAAz
zQtAg8KHRI^Trqbtj5{6538nQ@2i6WJ+2mn`uS<WAJni=Qy82A@0{7%$$n&hWM%a8z
zvOY8>`tUnmYO?g*kt+r|4CM~sC{AJo#qV-^Aj<&m<7R8?uT1lF#@o;t(o(zRitZSP
zS>*I8KPH=c*QZb8AVK=t^7m<FJ@bl_hNQIL+g~KFAu9`(KLjK}5!8XW2u}^fp!#T+
zFG9HktVI=NZ^h>ti}!NhVZ39BlC<(fnsVcz2vX9VM8|2Hnsg_qega>YEf_CAuJi|x
zdA<4uw8&{{$EEG!*zI!Nwx#@34hSLF*rXW2Y(l^$B1Dk(lmO@<xgbt+UIVooF16v$
zC7@qTC4GGmgG+t1_x3twOT?a1^U+2^b~(|ZvD|uuoCykKx1HHifCVQi(^wu2o+E~z
zR=L|IKhx^e=TZx<zgu5xFZqF#E?hR#F{tGjTSF5^LuIB1e;VjISdWU5F>l~U--3Hb
zVM-*ic_&xyI+<^STC$(E)p@u!|5X0hBa{L`LPBqw*>*ps!mDNhuYlmN-*WMNbt4h7
zlrWBJQA0C_(9E%V1v#=MIWxQ<w%pKP`u<_s&wZ;~e2x4~T9;27!c?bkUOpbk=7|fc
ze@BR_VYZte1G{D(4{Vb{0I+(=6KU9W!?kYyb*ytf!O;QG?s@7t^JOWKmx#XZt2mOt
z8KEr8)ti%U?Kh|<y*)4%`=|7q<a34m;%saxxuww1Os>S3qmR<c?lkVg@HAm5q9!#t
zSY!4(d`qCePK1^g)y2Jq0cK;fvYn!&MOqEKVKaGt;Xe0;!x+6@n9y&my10svTlO+S
z#&O`61nBpapL(~mC%v6GDQfBMN2h!1_F%_u>-xqB?DBkKbS5PgZY05HFc}IWrcHs`
zm3VS|0KGzDKj(-EkjwA9G+1k@e<$$B3My<*(mDB>{D24fZF$rIIZkbxK?+MEIKqb`
zC=bBI@p73zgDA*08kB~q*-43=D7^d_!qQJSKJwcU{JxxCF)_}c|8gWy0PMQ<vAVNB
zl8HMuF*8aLVMkX>|9eFOPJ&<ehECLTyMaAENm(_O%5Y@<{CmPR&=BXeD}CiqI$n1M
zJ)s{Ml9$r*Jl=`8sK0`&rv22-WV~F#1(;4Tv%Gf7#hA~YKD~%EcgOl`(L8O&Q!n|<
zf;9~2D`#V?pB&~fUVYj@dz={lY~Ma*#cU>Qep>T|>=a;yp|iVr6tW%xrq999#d{Tf
z!Je|ppKuCAE1EWGXvM6**e=wgUY5?y`U58?Ew$c38$}DQ`<5_IZukY0sS!afsF8n|
ziGO-5*H1Dl!oTVIEx<xI@>^G3>gic&2c`$OJmmIQ!fQJ~D|7g#;NVC8t!zb^q#IxY
zB#M0OzvuZ_@+4&`MY%#z>qU(JwgThJ<gtdjZ8ra|pI4ForEA7j@kKWsDWueS8PuV6
z0(Q4bcJGhSId}xu+aIv488)q}wD>8zTClx%L<KirA}Ae&w_l)MykY4sc<#N9Nzcnq
zF&?ciCux;3giV`Dr|}fd|K?U{!I4gR5;*`saT(U>YwfN$n8++HQb&Axc5z$_y6X&-
zfSe5Ag@~2jBs`X0?sXfK(RtNEK;i5)k^WMLZf2TEuU55q2z=ksWbjJht+@TlmHj-|
zccpurTp-u*Mdi$In0r`m9x;GsWqDLK{fLwPiTs8+B{!GO(e+$iG<sEdI^WdlQh~N7
zgX6@oM{AbNjt?kwFRGs_NQQr01H9^qm#<G6hBl)XsU(T(6D2O*gkNJeu`q0=)}<`=
z*ABA<^UyYBD)q2&X0()B`=z{zf<JYX9PvB({0$H}A*=l+i(sa77^$AqnOVPl{z<F5
zFvOG_`#kuT&~K+n`-RF^zC?16Yjk*YDS;4h4Dq-%7sJ(gRn>`O32a;q%d@<l89oT5
zbJ$OdDOvw_+_&~3)0@mila#wNfbWXWW3h7skpxGFurIK5BMs~1Wz${U<v@i1D616l
znbb-Erh0T!=upZVgqynzPg%oRI<LK>SCLJ+Q7%-iE70uSuMcz+xZ}yp$*vq0e{_=T
zs=AJ5*h~sr&7Ret`588)iSAvu04c>KA1ERk51CESoe{_r-#x{TPm1WoP$P%ck@ItB
z(+O(7c>N)+BZL&woW`D94##(b1ViVO_KJ9}kKiM0jMz~aR>F*wUVRG7gs43yJeWDY
z9KB@uF%Mc!SW!fluxOL7GJ}dHd|If_LQN>-`BepE9~WInlY+2W3B)Lr2>+Og?GIna
zL-BIkakaY7MXTn7%7*9U^2mOgG_Vl0j7INw-qwm3$}9e)DOwP4mjuo!p?nWE;o*@!
z;po<fj;Pf*>0>~KTVPEHM+WD2K55=JF7q=rq53CL5(Vh$QCNpjA1<aYVfD7GseG}I
z``MdCNr3&BNgcWVV3e?Jhg!u?ol3SP@bxny5*x#^r<UjI;x)8ABuFEDT{4$VGT4-J
z{<50}*5sQR4A+R3k-7C=y-aihn3(0oiriK~v|NW(tWqL>#&}9}I9}K*Nqb{{0>MDi
zc$aU<ZAYv^?+Y)V#H<9yx>SS*AhHC;A>Q0jHNl?m>^%??*RkZ%J$f7#wzU)!8~IN>
z7PZgSLv$LNEe65^Itk|sY^+SAdPx>x{aVkFHl5rg7+<0{UgTu5b6=2P0DIId$EO(t
zYi;*qHQhBA0RAU@br-W5_pj3ksIA|1)WT%XKLK6gKG~qi0I8<o!Vl8z00N%^n~Xm3
zz!g$g)`IRYjL#L?!nlya!t6PTbQ!hZD3I{)>R~yG-KLvu9Al~y`0N=GuIrTQNKgYV
zW2H~Ut5URRctICbo`uarLZpZ2bzpwMzuD>68N#)bQW|g_8XZcM(dV{_g(ML5?8l#%
zB8UKEO=F`Y7&=-o&O(tOYP>JdW{D7KbOe1HO<1x^C(sLZh7t8+2H~vDz&tI+2&~P1
z2Q&ups$C9h@W0UrbQUm^Ed*N&vdf~z>{Hw<7mz_k>oBHPL-Z^zfuqXcTMJ2$|K5?!
zI;o>WWNMYQ9LZ{!H3&%n=ir=&x0sH^y5WiCN1PL-VyDWt7tr;UzVW+zJ3&rl-?%3i
z+?1w!oEM_bU-6=>`_Fq9&S7zc-a4(30tGdhbttm$D@hjzdw)celnVPS-|pAOdR`u%
z!p!@2lwrhkh0^NxKgpP)xV{(u(iOG9;7oPve-OlT#!ip;SQ}wzp(Fff@?FK#>F%lH
zXIT>cViPS)_tc*L(fnw`knHurXY?dqYxAwqrvc9rURk<es=Lw9JX4P{J{uigG%UTu
z)I@2CZ_Ix*D^_zepg3k$<FGq|vvt<qF2XIyp?PKt*}#-AK0T)N`BU;W<^^fK3W@-Q
zi*h(8M%~-JW7BavH?nB|MtW~Vn{vAdes5g-X<>kA=1GY3dkI+?*({;HOSuo$am*4}
zqQ;=*T;CV>GLH)0xz_#=FUZjl-~E(F7z}+y5ryW)5=Hsgd=OS>R5T2*Q<MyDS<kUC
z*2^6loR~gtakyIXo_*({9{3Ap9A2SFtGHwrKUNLSvH}oq?~aigs;+j$qMu>#gs@=j
zIaK)=N*o18`L1X9A(pwPd-3({Ar`80N}}h;2m9W)4WWB@Dv#2Nv$<g&!%_Bext7Mz
zYHWQ|?O>{cjLAhB6YC$3sOfOpAL;bk7+c%R3zi&2zJ%pTyqzD9$s_;D@l81aEplXb
zp3}21&Z;PET2o#<OQG?>bg=SJc4E52QJ}}b8bGYUCt<Jm7>#@Sz@v>GYhVmFPY{FC
zb5_j2#pGO6%3aR*`l%Y6Y~~Rb%4OXarC<7dDsrElbu|CUm9L2dd5tj#!Y=$tp9~8J
zd_FP2%Ig0}HFvFrY4by2N16R>qQjddGmg?myV@^9+BDOTM?;X@sGWPwGh@J~N~9{X
zr+kjO;hjU)9^T_b<z=62%s#~A2Zi||-7~w39^q4)g&L=Fkz#?_6Qdr&30oWSXbr(A
zKQ;J3mFS7vvni+JNQoyz@s89YniVB3nuqD<%vQz;r@<U+76}52^;pv|xrCg_`U;lC
z?J$qy()03K;7`g=8Ar9-%CXsC)k6E@)vP6&a;C;kG6>_!M>fR1y!vtS>uvjd{n7){
z*Ch1T52-zv=i{fbEA$K;i=<W4hlJAjX~I9aP#@TVJl4&cv$19C>t3ImqEmV-AtUz`
zOW!?C;w9}2tq6YL3RO+1g{t*;HF_;gQx-u5hVMTTzREbYUjC~J!FmQ%A<pw!M%J{W
zSW0S|KXl^v4gk>;-ycGL6aU)gE$kuu#lvx~BY&RevtPuEC2_>m1TXJMD8u(uD#;}U
zHI*yL16e=Fl5P0*{`@sX9F}cy;!gS}mm(MDw-s1TY0ckr0!(`i$*cMM-NK`Egz@Re
zh&HA;Kah}Z;B)wkytI-S?`8KDn8|@7q*E>|z%k}-g}sHaoEw2Gre18Ft`8hmVH?2*
z6o0~6NEEBW^jk$+i3=m!M>}D^`O9GS;Sg2D*rM;HX2&HQx;6>8<wyN196@;-@xs|}
ze@nG#WpLWbA-V^kxLz{KZQqW`JA6KROovA%R&VL`<Y&qJ3eO<luz%QJtzffcQ6bOB
z4MW$o3@8*>%C*MXx(yJ_##6?Yb2CCz(<0Z;I9TW58KxuDSz#nJiarsLMHoCv=nK^`
zUeok2xlUJJ*djvkL27?LIicEAjIJP)paFRw#|K#}!AKK53M3W!Aue4KE-A(Pm9N)9
z8FA)3IMI-k*+=cA{%fe_5LK`Js^wv2OXf6|Bm2s*tR``^vXMa2b3eUkT^x^(^;!|4
zL%&$8Y^~#(t>&mKLuq|?IkFyGYmmy9`}Y2lOMmBhsGKTv`Y<>0*a_(_qAdf1)tocw
z7WCb}vMOPOkP<nXPYG()j27KwzXCsoR2!#``?1gU$E-hwUA;tm-%vCS@mpu6N*ll+
zpARgT$#anBP?SvUB$3v7M0GjdsluMv2@{;I4vN%?s0)XT5fu<$Zfl9nI|`*67iodM
zjR#8eS@zPksqB!Zi~2HJ+w^DnSm5Sq<ujJz#6p?hsAc9v+SwuqY}5#XGzyWnA_*@Y
z)ap{Fjx;X5XkD!sBMrxM5VR<s*xB5$1Ra+CFrGx|@XOr8Fn0&h4f~0?6QA&ZZC#j=
zFq~*$j8)gi<{D`dl!KS>;^LcTj3!SFoR(W}nTvJar?JvR<D;klZKgtqLDePnV4ECh
zX1FRl(uYD{PZp*}Kc@qds=$`J_m*tYx@Ib6v4gl(!MJ|jL1UIQ5O5`_Sd^=NQ!z6r
zcT%9r_IvQ~{es3{F@c|F0rw=298fTKzZw_#WqiPRnAN95*1A)j`_#2|5v`~T)tn+X
z2Lg^Hs%R>th(=pO>;CD#s*`@aL3YyLVa(C|!DLt8<Z)D1o#6w<O6Q9SZ0Gvt&k3ce
z6iaiSNT1)|R-p`jasJGS$V6jvcKLH0ailH!y(3Oxb$A6l`-$|+++RNUi?~wi^TN;Q
zP&CXhkINDb_42v#a=3d*d!Nz;{tC9lGAhSmvrThn&s0o?-AJI3%%znSXBu#;fSJ1}
zrB(uFM)*fA$O_N^d2hqEQycV1G?t1}#!1BL=^!pC$)a(oEpEiel$I`}5>3hbSr$TQ
z@J}nOa^sZvOxm~Z5Q|9G=T@Ag|KvVR`xBJOb{DDIw&NM+D2mz#Np&~q>$#JZWw*Zz
zvs^s&$5?}w0o<X|;>%1?rKnAeTB#h3?|DBAFbO`5`d3#gyeQrMyxl?aZvB8;KB+Q7
zfdN;;yz)NAq4Us2P|^|okqpIe<k~9oyFibp1-22UH(bx6rat;Nh}Ry21S_VnBbT9V
zOf)*4U4}V&il=8rH0{NYnP{~-YZiBEAQQ1Bhu?(jyK3k)H=FE7V4u+S)RL4w{hfrn
z$@qGu!`G&Y_b73%S-hwfCp8w%4PPs%i|Mm|HmQl5M2rT77t)+h1B+P>t}U}Wt^$`W
z4C*S`izzjty5w!uzH5cksPV9u6;&$y>-`;e;&f(F=1nFhkBKFbclV!)c1sN%OM3_p
z(OM<vX_QXujCU?DJBgie?Yb~jO<4>CCJ+h~NH`CCo+L(DQ=kNwO(@T&iN6DFz_P~^
zwKK1iFm>)hUZ)u=7P`oPgNp1t#)xAPycvt}1I|+B5h0lfw}9rhz)<kE94d4$y-kQ*
zb(N8%jwuMOd^YhWe}BA4{&!wvY2YLHAB>&3OKDg6tIpwRit}2L0@PsrQPda162JWY
zeC!ou0+cUZI&S6>3WQwrNGTF;X1J(NW$n*%rbFnB9gMZ7jFfy2S;B2f>V|B;E(i~J
zbmD9CrPgF)2ki!=w|2@M;Q?Q974NXV@6^(>Ac+tZ;vuc#Y?FYjh@kHv*Z0JO7|EjZ
z-#wspBTcEYs8iL;c3Yk2_{yQ-f${D`I3uATm1zk^9*ADgL-B<dFXs`7@ci%WK`>Qq
zqUnc;7+`Dwv`fXXHskS-FIjNtQFq^z&D=SU7R5Bpa<!V0>3LVez68}g=P*!Iz$tsL
zd)6U7I>QVW&&irB*DWjja_((_?hL`iF)}a@_SUmli7~W+;%x$p4K03=$<9pKqaRsG
zmGYZrF-eB7##5=wA|x=6(}GUvG;r`4C4$RoOhB-&`*YbF8w{NDDHBqR)Ua=e=>$W<
z7iH0kujbhhx=V3#pTgCXGY4WmL&%D<{F-;ZmT^~Ts#t_gQwY^tk#TPZ<An*}{fuP{
zOY`)#sL6M#Au&;<-G36*^k_4KOorj!kN$_$GfAJ7W%(bnlCg#bcy{MFm1u^=51--v
zs}<c@@>eT5bhfvVRx3#(aDU}Nh$@{y#%-2%+cJFWhsUi1v7j*&iN&qb5IPx6s%0k2
zO5k5tbkY-_z7Eh}{2ErjKu;QZ9~K>}B55Lf^UXyrJ{Z)JC{SU+o~H_0BXtKM`6Cxc
zS9|9uQc`*)S=Nb>BvW_7iJgT!tZBd7-n+9O?y+Cks6wz#vnkUPr8h3=8)I5!u`u*i
z$GxoiPH}8C#nRNSy!l*3gj<@!I<+$Oc8q$P7<T5jSMMC3^wzda6!|42;6rDvw?<EL
zl{iYGT~bP$0hX;X<?s;`|KzF|5Hv9F#Gb{XOs|(~;`@L)Uwu<ev#?t^R+5Ix+g52O
zdJSuhjAHGXG*G^<N-uX=Kjs^?-F?8RZ^3T<I~--bDO+7`_&oZ;cNce`>_s467=1fR
z6KEtG8)iDA_6z-3VzbgW?dmit=X$BOMwB}!zAoH}s7^;6xa`~^L&hK~<1%t$BIiwW
z9gOvAhUk(T>Msg*mCu!GF63<#mePOr?VQE+N;|y>M{|(kdW2qQv<PGWtOHc*gQiA|
zSV2iirEJ`n%|Nt9LaJ6^vI+PG>p57zLj~nR)BFr$)DdX)U!;g(s8nwk=LeH>)3-L-
z>E2lRdcB8tOYNVsCXzEsu(Ej`90lnF$vxzK8cv2Sg3VSb@DNckA;O9D#X)&sbO&Cj
z=L>qW?c)anz@xdgP6Hf|Qu6SaKp8S52YJg#{|y_n+~O&F)B9cwy!(w|?B(zP(_xDg
z)hhtEP%=#tpJ`iz&=;fauX9dAfGa(PyN;pr`eW3a7(dV0DuOx>S^41MaFbQzGc}Cg
zfAusDKn;acZ14rYU$+yf4qat%T~2M@<!bfIqK|KC-27(B-(ZrV41Z0S3B?Y=rw$4C
zc)9-Ar3-M6WdYVmIP{Hup&6}!)?7EBx?-0f6F?O`MimSy<=cX=1e5eQEaVv)b_u24
zkqf%s8FKbJRJnh?iMh}bgTZUR(jtLm)i6=6?3jODp|_5o3uQ7CY?AZO>izZ>ym%`a
zIBmGO?;_20W^w-xT+Z1d@QxRZVbsz9=otuN24Y;X5&Is-_Ufad)J-XL0WifyJA%mv
zDka;mwsU}n!;&^!SFAJqL7eC7;7LQur(=XnELnHtiMyFGGNCDV@}ukf^i5mh&4Lj@
znNbZDKDl?K2+?~=v`#h>4EEx{phIK$@J-Nu_65_;B`NWUyvgI4@#<%9^=NDr*^s#6
zLXW~-QdRj>zLdQ#<R0|^82^q!vy&Q0v5fURT4x%{yPd#bPSmv(#l{XrGapBw=WZ>0
zHFF%c&fTsZC5oJG<$0on-@h_7UH|sh-cTSwh@ph{yPN@lRTQ*bqzbd`V2v6p5sV!l
z-$;r1#*f|<FDUo={Up61OFgW><>~cM+s2q7o7zz~r%VF2=jTM~vwqC0TYFiD<2t8K
z{p;cUEmh(j$)xl|FU_`#qi$A|pAB6c^bH=et?6UHA3%TT<$T4*zPC2g$;YY#Ls|bE
z=a;uT4fs=$%Y$ZBk*X>j(=V@>9$X2<A9z@wzl27rGPzc1ScoKmU7xJ|M1Sd$M*uqV
zo$lCSrav>jfdqeEMC$)#RPi>-xGYh{E8(#SKXsItOApV~>!=J`*+~dvWCZG(1Ggt2
z)ZSsL>8@`zM|IWbQpe<ukn-eCD%XWZS^DpDK0aAXJ9;u|{a7dOZ#9vQEGY~j6Omd*
zh(4-}CSg=a{b*vw1o@UQtV4HCA5wQ@6krG%U{fz_S}KZW`{GRB^d$|)QFONW@R7=!
zQ;Ra_D@nWWL0wp}EyH7~S?96ij|gD0j(?^)ppUxdq5LQL<)4hpGLa*IY}Ia9)P=N@
zdH>TSxOdOlAt@MS#ixbUrrI7Z6>(V|B|x<z8zA%Z!X(e!$(1~zZLF@GbM8f9t^1>-
z62_hRa6IHmKdVz9=XJex*5CPTE-!&%&m4TkTwBIA5%1}<3RM5%{v%+q@q-97vASMs
z*s8JjGe$l5Xn$@hCI50IH3JXX<?0pvwkgYRhCVfNI6OcWJvZVv9ZA5!Rd&DgIalyq
z97+@GI>LTTHHeG6g$kJ|f;)w3?Mr^l1z}@pDbcMGker#09~C(M6ace~MpNxK^<0?P
zf>?>w9S=-@9ZE0XVyD03&^nGbZWvx)=WnTR=FH~1S_qOsweK>iTbTs0%)(4mmMuZ2
zkwU{&wQ-w(h0eml&G$F7HLHM<T`U7V9X!)k@6oMli|8k29u9L)a;t3IwzAz>0EzMa
zZQJ3;&I`Zqe{CwRN_PY$ob1kn?UamD{XVnaE1fIBz?2-R`ikd@wRIMd6)5p*eur$w
zo4;{FOTVG)dBBiJm_zOHQQlF*5z|q5*W;V+J2>bW6RdNhNMSy|g9sj?qjV}4rYP$K
zHe!n!KoDLeuKavG_&V0d3LNT)e6*ehl^tA>@RhzY2DX+FmwVONxztxu!R_>KW;@UA
zS_b6qhlt|@mhl2}8~zHp7P(iW%!mR2e!7GgNu`Ix^(W~qVPQc}u`_x!CNwIg&7KDd
zhNK9*k{uQsi+D;XH<z=OO`s!&czi^&i?iiqmf9KGM;G}FI&gP8go#YrF<haQk=@W#
zK=#x+f(iLD-x@%P3By^gCgHy<x3fRWn!|?2at;Fs-9EO%O!CbN9d>44w4^*pm94iw
z9MxbF*W*pVw8rdDb04ReB*kb7CCtApWVR+^gHG(b+mMbmc3@$$*r3EC7u7Y@@AFFw
zg$FH+)_=YdUSc&uD710R0~q_c7P`$I*LBjf$7^5|^U-kz)qrQXLsyguBCAAQJ+7zF
zmcOQOKYn<j)D8j~_Y=fGfcA7~LRYZ^v`e7pNLX=O%4LA0HSBd!h2VRu1=trw-#n=o
zk%p|chp)1<UAe`$j_IK1f9)%HBqTmN?7H;N<+Dz3&Wf}wzx&Z0G-%D8fB7r769;{c
z>ab!3jW6eAJ?X<d@{N2n8nYXQi5R0gjX8YzQH|(AmVPb=E0)<eZzQ&QkuBYEoC@0w
zv^}l?lRk2A2f4rLNf8%Wp<(zj8eAP-+nOa!sERkvLhufa@2z6B9fPRe?GNsj?;2?&
zq5r%`>`k&3T+frE=haO2*g_`=Z&ttl%X-uKy$E;C7*%&)yi}-h8n6u6o|@pm31h6C
zeb#6-hGT^&R;rCLLwY3U)MwpUxAIR*d+oyl3Q(oe`AN_#-|E4&UDu_CXf_D=q|#*-
z@G=uP>)DL`3d~zxxyCoR?3f)R7etPO-?dI@=iWSh2$QW+UAOHQBcAQ5){(=@s0?tr
zXR9!Qwae)RIwwD*t`*|$VUuwrj?%;UX*By3B|lSN=$<~8)Kl;MS55)}ZQ3ljJ6FW`
zq6Y78=s@oMOdYh0A{czkUo`XLF(YMf<Av$-bkD~FQ;oIfzCeojM+S?bqc|ZIn_%}&
z5DvSX(vRZt?7re`j8Cm@05TfaNkDw+LNFUw14Xnn79e5wntW0mm(wC}Ij!s-O$v@i
z?(6{o(!@u_FD{K?nSzPZ9En^VXkIn%6v~-oyf!DY9<^Cok#Kn_|1b<J-q;h)HWx(R
zOG+r*M3Zkqm|5`YR^cie|9hv1rp+H>nsr<aJ0jElbA|XYIm{q%Qk$<<Pcm3R7IEJu
zCCrmrxQbX18F>|g)0!m4xhc0k1g^0j4N(L;cg3zL*r1=1<_qI^m{2sZ>F$|iLW5dF
z-UC{!LUw|zcDa0Zxk5;O%QMDw693y-I$hxN0NFaC@)t(1cCfq&Lb*D}_(%>dF6JHu
zh&N*3&eZ0LBOm`-5%Y~^%LyhN3U<gCRT?nXK!n!FwioNi%i)}72Syy*L-29nIs}5#
z*rfXJmv|0>+fMqqDhVMo*wpx#5<$p^(=mM{7Rv=UB%I^J>1}f&jN8mULd1~Y*s;qv
z$+c=+NN<`5fuGMR^avTw_wdh2U2?hl?5zF-eT=-6v$@B`9J`Gpp>uo%3`e{|4D5~P
zvU|1Jwzbo2@Y3+`(o*o(6<9)FQv7^7<1?vbunqCRZ`CBA4MOUYb*G`3AdUz*2$*0}
z`89%FHlEBqgYDzJ)Y#J&@Q9B=`{*33>n<cARXW_{b<7VZR}md6T822W(d-A?av?Gk
zl+Q<nK5Ay(3vT)FEFJ8skI<R2k{x+)=0cKASj<KcOhCjwX?fIAh^R98RL@or;U4Dx
z+!}YcG|WG+p2BGQJF{dmhM?V#A4ebPJ8|h^16Aeg=<U6f31WhvGP6!S5J6p0DT&5t
zlkg5vstI5;3DYL2=Cjl7?oXBu6u?~zB=<WmHGc=tR^03enL+M|Wa0tI5qm*4q@;uF
zE8S21u;!%=m1RZ7B+$+SxoN*R;vP*b4Srk=@;P{PEV5J%_9%WvcVkSSP3@%H=E?h?
zVEK|h%yWzyRpY$y6ItH^Gu^@jqhjqTg>zt7Q}8!pZBl2Qe(}AFM0;`GEEf!2+pI?Y
z?@}rgkAM1sDp#0^h{&Bt{W1Jn6~EoLk6XabJ1>k8h8^G-If{Z7Rb05Ypt01ssYo`n
zMKZ-1d)QBdRw;CPwa|(0p0Kbxv9Zt@;9mP<?|O7-`Fd2o4P28DRlWNwxv{YOkfNPp
zYeq@gPo$F;hD8WrCzJ~Q6;sZzsA>F-$dM<qVeGKFdf;2t9P#cF;|Fya<PU%^NDuR&
zwt%4ZNt*aAvTQR9h_+{*T~fnR@KB&SuP-NJYg4K{59|2EOt{~5JD6WQ``rFy&*sc_
zET>Z~Lhf)tVawFUn<FS%rZ(#H93JAW)BYR{!A7?yPt*-M^kOom34w1z-J0RT;g3GC
zUsGQ{1c(nr$^idK<x1tu(&Xy`<7@&RqyY=wM>yy!aP3S0Y<vKdWPbP2n!)Du?$`W1
zeMasdZIU~6_dUqHODOK(@VpnBSD0^5T)F$q-BEPa#(BK~&+J97Wkn5-#eT<0sM8k|
z<#93NPWr5JYSA=Ek9MNq$Dv18MNB!{9Af<Kj@Gu&PN-hCOlKQk44my^$4f$2tnSm~
z&yYzW$9Jwu7WKG1c=gxer(4YYhH_0~qUtp3$c}TVbD47+n_#~LVe9t1A-OkIBi5LX
zcTMNhn(Tzq$7kQPKDkm^9*q{2i=e@Tl01@7N#5F+EgIt1VPO5<F&u4L_e{C;&UmO$
zCFJYNQAq0U9;E$CzbGRgkD;RAdaO8T02{+&g@C~#`18h)u>aXzMvt$0tpjiKQzF-m
z?^Zc|3~697V5qr&cZA)9y(&>w&3f3P_d=;TGrH`&gup~=B%k?U#uCw%-kTTr4SC@K
z?_1vSx>w;@*D;UF1rh&ln#*hIw0uWw$(ok=dxxHtncvx$vM$T?jt~v6XrIqc3@!+Y
zekGfjOMRKQx0S!e9J98Qf}U|&w1#ot)|IhWk^}sc_xt@R><Y3GKoRq*^CT%_;@WVn
z+`D~W<eU(6#k^c#PC3_8FGQhg`CZt(Px+&!?vRkL@z+%A3cbk<dZ}QaQQX|D(d2O%
zy-bbgh+d(`XW4x0R|(3yBU^%ukM6OKa1ys<t&&<kYYjtZ^H$xZCFC$1W}kBK%tC8P
zklOeC5Z#JL>N7(B;4q}dD(_P6^m#0sw8bt%CwFSj4tH9fIitnL;goZl`@`L_N9!1l
zzoCYvq#rX=$qD##oKjiV*WzA!^|TUi7DpG><Irb{mT*?=Qn4j>A4%CrCUW(bsON7r
z(;Z&ke`^+l_HKf@WRAzrEH`n)JbdLO7+b_uz^5hlXHrV<N-3K8V02SAv$(u-;<3om
z<wxnQeAYTSpD$fcqs7A`>Z3*DWrnvN7pZuHwR{@1%omFG*y$5!J#^zmFnt5ij`4vB
zelilbuvMV{<b*|0sw}Ja&IQX1qYk`$?sNVK9W1tx|9CQy;mf@hOUap=-AuC7ftEst
z&Pk*bPZz(l)=%zHau`eXX*#g%mwK63ZmFDKNbNeGGMibHHB~6Z1i=|bJ!b1vY{b71
zws|H@f352gz#mz`H_flXo_efisq>W?yyB6#*8*+M5UL<;5(6MAX{4sroGW{g^c)DH
z&1WS!T2~ybWcgV<is_r}qht0$zthSUd#AVw{Z0>Tgu_@QQ+t@fm5Fwh=bjA^qJk=x
zii#W<y#@ou9BT!Sg+h*Vr=I;%a+r#pV9d;EeFs&9^Qah;?bK+IUw&g!oU$xBp#@Y!
z#DTG2dqlMECjG6J0fV||g5tYI`EX(4cy#Du6B_Sn$&kAJo)@>i%=jr~sdC3(`3TfH
zFMd(1Cyz~n*Y~W<3yYh_P_^xH8%X%qNT2>L5A0<h%lf-}_3z6I+(71NaY+BXaYFL9
zIn{(Bn@y4qWE9i;y*Mdx5O_s!)%9-#;_byPu)~TC6r>CxM&^Coc}#-4j?wVoA)C~O
zsZ`s=;{Ge2=sw>)cZ+$#&s60!_p6o6vPv3$Hb6U+;f`7Mkd<51$cAa9OCU}Dd%&-B
zqK8$(m|!b4l>gfy0s^MIWu)jxp2Lu0POE_RW!>8zry2E2^pfZsIh*;0v|ou|bIC~&
z?r&oE3NJO>uF5_S*PM46jG-k&(IqYYitsDJCLnP>DQ9H03F0xDGa$b&r)S}k+|oiz
zUA049e;h*h<R!*XEvzGM@4@pK^4b0;W`i%ClfTx8qVf5XKIMhusn8zpX!$oE3^+2E
zf3$Oy+__{r$F^Y%5+c}43xd9UmX1SO$9q)H*wvRSwVT!QbIR9hxKi%V3(19-78fUg
zLcRC4pogu=GqN+IC9=FX4{9VD0JwqZiIxH*#~<rw5HQRR{eDmw8~!s2_3k-0X_eOP
zAlXsvCV%GfRLoIpw;C49)n8p;RMdF=WdYv^r69Y{quk|+LY|P=RRG+361z9<nz*$*
zLXV-GO^?A2s$W}s3w-@o-u3q-5Ci0CZROEgKU-}20O~oamBb37QWp5S@(4DcsOY4)
zwB2f91ga$z+7qlMBv}0X^R^_sU2=m=AayJ@zNVe9r6$0Vg~+s*)=#hF%UiG0M<Z~}
z1)F#coc0M1=<2Yc%6g!_ff1|pra+7UNPZ&3R*j4yxK7Gu@=c7A$+vjZyW8fTlHLaq
z=8l<9z(-Tvp~(U&N`S7w;ZP~n{^2jfHuQr-o32J;;$;$tj(6Uyn9J--w{%Rrh>SHk
zHsVMbRbsLhKq|7TKu8N0i#g1T+hCxzdG(3_Jqh%K$U75H>qO>ZXQLD)6zpcyOyMg=
z6@ARcYFs-ew>_LcE*%)SRGqAgYbtHv-;RKuw^HGsb#|H5WnR}MN`i3}YX0}$sEN!M
zTl!6%^9&<kGS5&evEuqs#rpa0=|>Ei^L0+uQ!%%a!SX$UcBOl6%rIpqCOg-+05UAq
zalIzKZLGmB0TB*6&xO_k8`KR{)-@y}yF4wg{_cn~(MJXCf=s&~CiA4+?W%OBw1r(Y
z`f9=v*97>rvTvTIT+&eG=XnLw?Rhg&J{U`zs#!-pji=IliLtb*-je2TM^5h2<&T?h
zvm`zkN~1kc_C?h9_~3ny2Swjb>L?S{X;~=H!pv`3^D&;?Sd=QUs8>?AI)*yNp2Ww*
zKjUPL2YA7r@=gyY+Kw?7kYpYQMk8JZUEWShOQQ^O=L>-K3IxtWKU*G9f6qB_8w(zQ
zhj+k{Ec9C8+RhNMRfs^2cy5Q3sosb%V(30EiSXPDMyB``S?d|;-t+WhSw^AZ7C#E0
zVBzm`B=^C4(h|4_B6U~3sI>b|FU-pxPZFPJg~0WU`>tMkFNy7)U9jHPOJid9&KXh6
zs8RqR+H;<fphaPjy_M5j`Q=x$lJ2_&g8KMh597b5t<NggKSoZm*dN*69=7CxxUW^|
z8~=v%{70C70%s>4$thV<jLgzghXoPJ$;CU#;e5&Niv?rqZSdSQnUKg*08FwkV9dzp
zMm<qrYXiiA_^my?92Yv+K!J}^9GWHH@%(G|f2^F%`xvC3f?8IHeB}vrf?Bq*;W%_A
z{wtV_R24vG3BEtOLc^X0CcLGY$~v>!84q=QhP2M@A8C;-tpYyN_eQ*z4#x`~NqvYU
z6%M2po}<3ilFi<u>-CFip>rD)kTZrpTN2S>Gyt5rz*btLEU!md{6;|^s{kPcqZSX+
zeoSq|%!`Xml{<0_xj`xka+ycB)hq+72M?W3xi0lEO#Q*q`E<jy;>1o`*kd_Z4EX;0
zD6tLS(f&<q#*370W$JJfTlg6uIL-ibtajPH&0>X_97m7?YO{dwbTE?OJw~HFXcnE*
z_a}QVJ{T!y2~hCsFVGC)qKqqrPJSAGerE&b?*S*K;hxqtLHu8USCP-3$s*|our9m=
zT=@OI!FBa@QZBM_F1|-eSuQS`ZtZ*&NDO=BHywy02M6uXxcTlafKxZ|?MNXTT7@45
zy$^d>b6IFx(#u<88kINT94i-$RK$SiMrBt_XogWy!JvYYI8I#v!TaTNmhTftTVVRv
zCMb7~Fx?HTr>r~q*{9{_>UoRWv`^X^{T<AB2@bQMkv_^HHo9W$3*qILF5=$k9$JrX
z=C!gQq3&N&ezUI6p8+@zTLSE0nh)|zH^8%0nDNZqALk9@hBhi2dja@S?AUCanQ_&b
zlMg2tgfJ)p1d7`qUN76F@>u8~zMV*Pwu66_F0y1#pOah9ubJ**QU|_G0C_-t&mxy+
zF*q#Uz;<VgL*}Ok65r|GRYpr})KOr5`kPyMgwOSiE!$$AS#NFtEeHb0S6otuui@;O
zcFz|QX4;N}J$EJ)U(}sEDhQTV>8c0Uq}HV#Db6OTVWZyN_QGE0w)%f1%d~(mI({Dd
zNrq0btEt~DDSc1C+t^Xn$@amA*&FpiGAxA#c0{Xs-}OTankbek7^z?|qr|}Nxq9oU
zjJo!=Xeel%cvHgcOSn_Cq<vEw=rc|+uH2uOJ$GQ&jxgdMrzl^&Z7$pn&+0CqhBGIC
zt*naPvgqfuO5XO1>6lw@3~jcM!umLGBrRG;BH^Wws$n>Kn-*6v@tlcNM?)-Uh}jIS
zCKF-?`CfC3UCW#?;g?A0`}|9#N#bBc?*<+>JHVZ2%ut&@-2LZ(`4=PU0f?}S8d?-y
z1H)svPuNBkztL$~B7U`65PU~TGb~rVh6F)OD2PrYt*t>t9-Vb{jQ2Hd<>O-{hqKxI
zHSGEOxJr<{gARu-%yj`iX6mB@w7LPK$+&o$fU;Lx>>~Qdl$(m6dYRCi6e9*2J3f_!
z0lfD6$^~PQ77C_Yc>RrtDK-e)&p$?N6s`COk(CCDpG-FMDvo^%4iE^utEy&v7Z+|M
z0Ga|cE9@!<xby<RIuCtT3Cyc%w+`f5*5^4*T)H3BkFs37aI6NTKkkucf<5IpS~k-h
zBD>i5LX{j78n|dn<B9-50XTo*M;1l^d+V)orgRT{5=Br91tzVHfL9Vu2wi`$Ye2rr
zHTdIZ{*=oBu%L)30Q~~vMu7GoD$VX3yQ;|NHMT;7MFwA?Cei%Fh+rr=cKjWr*^GR1
z?U&`YE9nu^{SvK7RjXeTyA2+fv$0ua-oxrDmVj6+p92F4@ONOlNnC9j@vWG0>BkKs
zUvoxEM7C~B4I;fW8nDAW5xxkAFoRkGl5DV~j^U?yWMK=zJ;iUQ(id|h_jX4{hx&<^
z76GN;uWDT7Pnr(f>w|cK1nSe?K-8UNNV#9=yKs^W#0`1LaSKb*4y|dl=?{+QY|d`M
zZNx!hSmn3PVWp5Z;Iz^hW@2O}LnxPGokCkD@xv0i49fU}I6RE!htkz$b90u+4AZ5G
zoUg!Av&44ZV5SKJ1g{M>M9>#<g%N>ZT1mX?6TdG<!7hu4FB+Aj4TzRY56F;<&RYb_
zqJkV%FNT7iE|n70a&}=9Gk!lX!NrS5^(>7~pYWjb7EQ~D%A}RA5Mn0GFXY)aPIn<*
zd5>jlHt<OKNf%J4>k20#``!5$28UC}Z?n#sVg^@%{fOqm9_6uXzoT6;#9|7hWiUFV
zxED3n!e3=L?;Bh91j@GG@7PKw?NO?rb1qk7iKAGj@p0t}wXN<oH>@NAkS3N1FYDgf
z--!9Yv+fic48jt^M(V~H?8sgZrKM)uE7Uj)Tt2SZ^vyF;oyFR|`B(2oaKJzpn$uyV
z7)Z?i3iv7#Jbw7Hu(axqWl-C&G4W~Y%6nM>UjAu&X1Pdj|FnyJfXjxtG!2fOg8#6O
zh)^{LgoI*!a$?qu5aH*l<HQWx&Lni>Uo3hbxvX`E$YwZ&RMrbmFAl6}o>D8w8mQ#_
zN^N!pW@`sf7PTJce9Ets7+Qv$>T$%cu_)yiPjyW^l*jMvM50wv;`pvrMp&NfQts!^
zG+W6%Q*~n~+2{nxp_Al|r5=wQ#qN<Shci1VJ@CeR;RKUyGtTupipAe$+>n)si!Ace
zG!6WdHj^q^_6fI3&PeY~OE07A964^|!7vMkPcfep26vcq+R5G2#}RtrUw%mzw-q?g
zOdZBOE!=m16z6|kXkW>J`&z}9(MkBZN+!6zz&nB2>4xQv#CLhNj%d%0#M9Vf)g|_Q
z)?HjQVN;^8_VAHZK4uep)|A&g_TjFieyW-$n$FXTpuC^@InIkCfnw+49s*MQp!;SL
zS^#S%T;*y?yzt62F{<#i^sHD~IaBY^>wDKo439J@b7calrOEtqL3!WAMA6XT+K$^&
zonHXy4kKmB)sM!lq9*gi{pOmi{2}#apvB?4mc6XD_W-R*tJSKg)MsAOkPqjr-s74#
z-FTSzjuA-QEhK<lQk>rGy7r&<{2%q!-_82}ypYS@;WHw^6<zjmW4@QbGXFp&Q+XB7
zUCCnlx%YjD#n#}(U5fp(f3yH1fFG^;Ng@s)lALI*Tj-LGvcLQcI3kq%SHdq2!Um{I
z78oQ_cmJz`_>V;E-<9HDsE3q$7*LGwB2hekb4fG7%vDcFQat9~Kvr~Pg-1e}tA68M
zZY#^wNc(&q5+%6J0Cb3(39=-$0{^+Usz0|>sqb7uTGju2^uOJ1)h7~*0deTf6~I%a
z%LlADuipg7RsqDh5O(M){%VHXFn)~p!Wz!$RTt?~LOgDYIlnd9Wh!9lN(3i3@wXqB
zJHIQEx!PIPiSc?OzE?Z(bt3bmw);*)(HPLDrl1N0JPEnzynj{)@1@N(9|262nmX27
zLd7>ATcW&G3&++h@L#&^Y2QAQr)_xRq_5ACTDKPd`RKnN&wqcHLWyM+S%0+;l)6HH
z7-sN@Jv%VJGBX^6y}tD&$Cdfh8GJjB<Ho_Fh;TIbn9{fG{_p_D-t6>~8!o-Ow&yxC
zV-wZNi7?O*=>ReRuB-Irxn=#WinX6b6|mH?UuMfJ1J(=Q0Fn2`j^d+{f4|fLN}w?0
zU9LIJ`1eQu8A5*_pZ~rbzVEmtb^MMnzsx=HCiYpu{02~Ywo%@GW27k_SQNpIdZ#M3
zBC-peTwssUe~~cPNd5q*ReC>kd=?=TWRT_m=Nlu#t-|l_^P9c@IqnZb0?077@a=9Q
z|9@Wh{}u}Tl()~GSNmA}zpLFB_X7bCLwevISLRVZ`#V4k%$IFqJKXi}H|B-sdGq<G
zf70!5I`sejPjr+qtpH33E>f+S|M_1d!M}l^5Nx^Hb0!AjNNaED)>`mZV>x!~^6R+5
z3gC0+-w4$Gp@Q{*JV*$@IC`^<%9oKw$Hs)@#Ym&p?QhBbM<V{O2tj7by?53dj7d36
zM~r1Q#pnaE<vT92UUe`6&U>r0%YWXtUPjtNnOhD&?*F8S{flw^zrD8T4nbJ}1~!n5
zkh)#wf8AM!mGVma%|-jS`K6Qk-&2)-A`3u-ta?;>oBkOMm9Cxrys?vYClJW*d3~-A
zvlyXpU=S4wh|Te}0H+77N%|(%*D*llj4oRId)RBh;#B=`H%P`C5atU7#I_!m0h)6J
znMp4}Ezt&~2g?%xYc^Bs(8<4^4s@&O*YU$t&-14>BkipTZPl0(?Yl&SXT>|!tlq5-
zH$nqWH^T3xL>P}|%>R6){0eYG3aX}`IlOzoWuSg&<UjrYT2g_}-vPmttncFe;-3%*
z><*WW6j<=Txx%Ql29&YQW&y3gZ^|;~W8JCLTE|cDF8}n(+>-(F7@YkwphC6;Bvgn2
z!^*)Glra14{v^Jg9sF~j2SA1jSMqLuUIvP*uWnfv3$FpqCoCvr_}Om>2?y4i#&Nyo
zOumW*zK@e_t3=Lth`M&5-Sz(Fyz=I*|7EW~6uEF4guj7kkbwf6Swbz}UA?GC4ga<b
z`Q&Tc>G^Bbfl4iDKVXr(2Yeg1ECD3XU1G1%;P(``>gT@x!~f5F-w3+B_s6Q<(*M49
z)}W4fAm)G9B)1NR2RzP)77g{RF`PD50nD>EY@nJ}VVi_r=%&Cb0AD<+%`&grYCox~
zd?#}jFM|){$bPq|019sUc%RuH+bkSsLQZ%`<y%fJ7p+YSE0W7+(WH0Y0&)Jk)coq9
z53tPXd<biP-zIaf0YBJxfN<Rs6l4bY2d~;|d81ZR3;f0lZj98O{Eg2Nk-dOWFT-!?
zbZ`7u7lr@8QvZ8%{Lk+!09Ptl$Hvp$Oxu5sAq!IG6_z|j6Hn~8HWPe#RuyvWl`M|C
zm3d^H>%?{7Dw7{@0+xtf12RhjK`?`vGl1-27QQ*lx`A6rS`rW8aTnINKcZwxKq|+Z
z)cMZ*AGY2)D$2GE`=z^4x=WBox{+=HQKY*Eq!~iGhi;Gt5s`*LLOO<$ZUyP??tRbm
zKJUA~z1O!Ef3RFGaShjf-De!f?|_Rt{+Vz|A$g-(t6oW6he%<J`fa}2l0tlx|L2FM
zAtEi`ZwjgR_dV7E)M4%rD(+kkcqH&9^Ak{kYhC|bzWdpm{896N1sO1-W`nml#nF?m
z*8Z*0J`=#x4!fo9y8&q<q_KLobyENk@1x*L<X!i?Kq5X4U4HnpED&$+EjNUGRR#DT
z&x9|5oAeLV46gs^n#_AVuLjf;xS*5t_WqlI$E|?yc>jHiO~lA$$3LFHTx%}T7x+7+
zoKC;y7g7(9vs-Ka035v)Af^u2<YD*yNFBjqf#lQ8Xyd^faA};SJHUNgutZ*FNe~hz
zlDO`~e_Y1CeTsGVWAxkA$)ori$Q#aS9q@D&5O6&FJ?DN>>OlwImvk*ZJuF86!x{Gb
z_NS|-uH?M?b-=er8T!8cb~^k9n5<CJjZB5X?V-@reL$u`(+1p;*@)4n@E(bb&a_(I
zg67kC5iL1e2q#r1d1j#@uek)gp#V&vntJeDE})~Z*3(uP^_i)2%>Gl4x(>`wDS>z^
zq?k)Yd~fMFAhGX2%n+Umt!AAlFd;MFcIZ)sjSq&&o?6!DdeLLh2f*BEI^Z7YQSMlY
zBOMyY81FwGvmZXdXBc=XxV7j`DJ!bRr%4FL(}9YdIIWJc-T=uc3}|g@JYOH+KON(v
zFS5K_g9SYG2i!-bQeW(wJZ_kL|0Q`6rHRAs-HF`=ga{M7`(TxNf3k4K<xwwKh}tF$
zdQckRoL?#tE)`+GuTLyfFU8XeXM5A<nDe-2^0dIEHM)HB%j?Uc=kVSwoObGZ0kmGf
z#6b^hQjbNw>JK*TzewA!du5LVAGUR*zPWgoZ2?`g1GTzcJ8<s00sdpwCw;M;1+oRn
zT|N-Y1F^?r3Ag{w-}w?xzPReJ#tz`qmc0M7{E*WitA0~@q;iYUtTuso91WLt#-E$@
z=%@(2)E#ZPl&YJ%6iU@uyhyr%ZR{4taPj>-1rk(o+Rwz=Ef&>*jE7%lz~=>0jr+65
z@ZN&)0{uT#&T9-qAZlRd=?llhw8=witnB2Gi(>23YUe|0m7mo$Oi_~W(4~DD8S%Cr
zA9(MFxO4(<jMne1oL7OrOb#LuEWQBE9{2s3)KlF&VvU_R@GiuG7gg#~0l<-4;N2{_
zTL2uk)|X1I*T4(U!qRa`V5`0g;L5st12rDa<#NDN<@mL)TVMx37$5}R6Wp1&1dHDy
zdV@NETFS!X<@|#7Hvs6rUba88BO!Gu+c8u``>F2y@($o7cK8tnffKG;w<J6dpWrjZ
zLqfPa-(uh7e^<^XO3+o?)lY?m+ndn<8eke{#Tfiuucj;V(+zqiQ`j-u4yvX&{Hcp&
z0)Ug*I5c+~+yUf|tK5%W(0z;7sTH?$IsO3hM_{$9i00z!F|_gL2+pmCC9{&J+Hl=%
z-uZD(z{6~S0MfkJ(n%bG42m{p=(AsWXP>u1vAJ76hKBY_|E-_YcqMyD;qaXhC<qlp
z0YBcdlHtAN+3>4w;EgR*E5L(RZ4oEZhm)A_#Pgcyi8%K-REEYG2*;`h$_)d*by6xi
zGHHeEUg_ITDs7tecB$JUDLy~?4H+!oL7iM8vS)cf?E1>+<Nl~ZEdtwyy0eeY_0m<F
zO?gP?D*Rd*HoXHLY9>b@%P1&lZg+mrOz9=T2+!f`O+R=o_xP~<^e56rjV}RDNzwEH
zC?2Lhh8JUyAHy{m_`Y-9UCWOL%i)b}7Zoh%e(0yneaev#f;c5~h3AoA7-=o=7WC`D
z2xrXv+B6nMheA%6<bVa60;-(8KpD@8u~Ko^3-1u-3oIh0T3oc07kUiYKo}KF$rUhy
zvzLcBMnozyL-su^B{2A=0uF^x+fZdl7|5AS(3VqPc|3Pz@CdH0NG^S5Gb@_4l$UCe
z4hRU1zx`A_QTGjX3~wF#^$36QryIH|o0fpF>T9Mf7DN^-7ObG%zq;Yehw`B*kA=G}
zi6gMW+mJV~bzG({05m%vSHF_bxcEBnR*Uqg{-U+Y@v5wd?ID83=c{`3Nwy%`FrN6a
zbh%FbBuaJPHEJ6WJfD)9{729H$+hd9&Wv}sWQ1o0#Wv&8MWt7!+64NlI)(%~61x>{
z;>V&`K|WVRkR6T6jzY;UcorxM79OeA?0%_kB4?WU+C`pmXm?CL^S~lE8%KE{-@vB_
zJQ9iJ&DZx?|5SFW&x3%>v88iTrR!}@po+=6=uX}wt-S=seW2n1jFmbXGV~hTkluJ~
zayp7J=b6}xtO@xWD)%v&FTP3Wed}|o;_&qyf}nt7421K8DMSMCA)1#h;J5Gc-I5S#
zUT&0d%aj(;ryfGZxh|_nGlz<Xw0V$<C5A7}y2)x)9dqh+c>g#q;{NBk)g2QOWD(DI
zx1CP5>Au+h=;oZB{9Wx2i|v`X)OUEE2KZ+&9p6(FXQ2U~#<LEpR_6W!X71Mcvud$r
zVC+_xW2%|MSe-(UP<!~Vu^B#sC!t5;Ym+C;EQ|zeE$++XCKR&BD!MWdDNWh`x!x3N
zpU+K2&t)A|4%w1Zie!+NCrnudlHQ)n&qbd;BiWGQvS$#Fo_h}ehDdTSLCTw3ig2Z;
zjowxHnHH<c`)G!9`3gS>Lsl<9XYHR*^kX=XPb@4{G6z5NG-CSlRXK#!HM2w}#K~8a
zDVbFhz+03Rir>g5l0+;$=jF-;pmmpj;2X;chQf}qSEq*-vw#h&j&<T~ics&@&WEvA
zb*CY%k=LX^Cd$xPP-<bMffg|@p}5@VNZQ-6h(Z{CnhD{7LK0_)&a)G~Xb!TCSk~x|
z)<@Cj#B7Y@m_@DH<hg&7kTzJg6{dD|lb<JY7hwcy$<?qDSwB<ekLDl|Aa~x8)?jt6
zGy#T=B)GNm6q=$tOsaX20a_ugtoT+T?g{v9j1wQ8t-Kgqvq)JHW-hvbU0G$=`G4^t
z{=l-lquDc4wz_RV!*C)GwvPn(w>4(6B9Y#f8_`s4ySi1;#mYUu>?bPllAORZJmOI=
zHW;OXOCxg`k4mwpe!`i3(FkPnDz&u;>2F6jYIq0tiUy2c4xKxRYJ0WJj6dt=plX?k
zPq`#IFp4f8O7x5oe0Y_20~rWqafm2X21%aMYgbN>7NDQYT2hpUa0lEBtpUSC%Zpp?
zr&DfqE?b8LY;~3_wKs}4#yZ;uQoE};UvkqE4ge=lNuFrVE|Wh*sZq<=r@{#i$&H6m
z1FuC&591||iE`D$YoGw;u&<G^XdA<qql!PD^^i%{pXm>x6vh4##O+Tgisc{J8ZQT?
zps>68@6(6gq>yV}a8Siwq+FFch`uaa$&X09fjH}~aQkyA-XB<8LKsjLI8k{0%eegg
z*S~Q|wr|2Kybf)71f%4(D_{wczms*|If(*}Z{*D-sZ={#2$qP+K@kI-X6{t(PMrH$
zFUhyrTVsBrP^x(t^5|MVRx^$TBS6tO5n>shGX3!vd;M!=%pJNO8I!VitT?qeb`5~R
zY359!qk~RO2sAeF3+nb7?bDA2N?-?bUte{5<_1C75dRg&*RP!M^7g|vda24WmBg?(
z(xtvT-J@J>Lz_Tj_UL%2!~zA1b%FS8{e!><eQ0|9*V)}?XNLNbZH_L-5HTd|AcS&H
z;4=7M%g7)vP4mX_%}ZyS5Go$FZ@McAJrsshn4u^TPu<!CC@vJ!14K=pDP5I>N$zd-
zdvZ|MH&5n`!S!DIql0R)c!M0cjBi3#<45v)enx2-X*C&|N=PJwrlo@3(9sDFz{k>2
zJAUo=od^*e9@GRC5}5cvTn170z=%Kn_`H0XF75*cfm4SDjq@-95X?P+5q`>ICHy`r
zkU)QZV+_NHK+|<!=aDQ>W@_@K6XR*g-p(}~^8{l*#SrYk^iw7!J`iEUV8l$uBX~GP
z*ooac9V6SQag6lK$_>H6VvBWu!;&0ohw)ndM4ec1`Bcd+e>h>yKEF){2Z<d9_<G!r
zhi|s>md};;JV2<Rb<K*7m{TU9Q;v0B5T?VrQH3$|Hbv@fVcg~k=M*^VHnH@ls&9Jx
z@)lJR58t@p%8PiM2Dy%Nqg&Kf)6;yFUteOUW<Gl6eJbQi`37+DO3RA;t;GBR`Mdhb
z$DVEc57|=HR!9;TBK14xll9%I0@8C6P}mop{P)saB~lZD=3io}s{=8;AbZn+#>k1<
zOE$QT_Owdb1C|s_?n>X!>8Ibk78+dR_%K=uOq#35Kml>Gd1q}me-~k?t&f?Hul+}m
zT~CDq&V;@&76!bP#ccZ4x6}Sl%c!@Dk^fZiLPHhB@_Xlx?8g9J6n8ulYA(y|MNJJp
zE<M=&O(Urk62Ze*_j+p6OyAoEMy^LH{H=UvYG$ZQOi6gV@FVG(*MdHYf^P;Vir%pL
z{9DqmqwiimaSPOq;Xj(#TW>*n$F^kvWL7NShsaab@yk&(pSnaW$1(UH)gHg)poO<f
zeuE2c|89h;Q3y?Wq9ep%G;Hw6mV#tQ(QnOw+evp3dxIH_&T@)lwVpv^mYGE)6b8_N
zLwGxMYa&#nSQ1TbTKT@&t0Alq3?5HQ^JqMvV2UQz5bi`<@sN|IF(=C42XHbA-q(0E
zBfBv425ycHd60t*+B=vTC@3mb>S0ofnqlKa{Kzanb?^4w{>kr%Eh+I!-B>6L)A*K7
zeFG6c$lG=>uhfqh^7f|zGebT!qKvtt#5<F<i!K_yTe+*ADH`xEMin`Or~GL{IE|Bd
z=7MR-rpz+h7yGZJ@4ZL6#HfrJvykk9=%OR~UnUm5r3T$z`WbO+NuAbv+8v!<-F?>N
zTM@z}d}E-CfAJigMHe}d`&YS~Ygqo_uUPWTI8m)pdKH%ke>^$^U9ns1%I3qz$6;;B
zQgub4PIk_K`Xg`jzGTYKPoN@4Y@E|JrWF?;(N__q>I|}Me&-c>jL(~citxYV@*?%%
z{2|QgPDk{X=5C$QpR!3Gun9Pjk{sjsaQL(Nd=yAYD7XLoFP5?-PWnB=U-ND&5)1<W
zy8-1s)R+oG@t<{J2x4LRHY*jrBH?Ww&xTcik%WbackUKi5O5uBvDiGA8)2+|Xn)3i
z&`1g}Sn3~CTi$*d${|K9?H<bWL{XtTMqo%J`Y+GqpVvMc(rGg(e+`x7RnzDPJV7AS
z`6JM=qJe*5FANGf0^>9p<HcZ>m2SXBL%Dt$kf1{t(DkzXI}yNqEX}V7K%3pus88Y*
zzZuZPgYV^q0bcSRO^riVATB&5a_o}sQhbMuls}PH?l1P~9Fd%;ou6U(OHNt_Bs6e7
zC0JwW2<aw$Hkj_nT4pH&Z8XyR`T7nAoRJ%jyqw)z4FmzBg6hlz7%T-?v{10v-xEs8
zK7NKJt$>_;4dC@OY5jxq+THK`@AaJSV;zDpgZ01jaPE-p-O@*3w~kloayo1y`N|P!
zo@L5A&}X%nJGJFH=xXQwik>PNuh;Mpm>3WYu(ks8B)rq4=W_XWzEg)f;S2A6Tr@$m
z{VXPgZ_^Z2(Deq_OvXIRWpW|Gkrm<D7~Csmn*R1<B)?=9AN9e-UZ+J+X&eUNb%cTg
z6ejAQ9QH*Hf}kvzK;x)dfwO3wkY4-GObaANv+m+>Y+EKO8u>9uLuD#ne8*^^<^Dl(
z3#eUG^B>X`rOVSjT<ls**i{pK_MFfYKZQWuv9iX>evZn6FY1Ni#ciy(&bLo*8FTC7
z(h?5N586EVd%Lx0j)I{TxC?y}w(OJlGqu9AmnSQ5^};T3y@$i~zKs;692yS8K{n8{
zp;LnDwFu0F{;`xPKoJ$h&pqnP;H`Z)%|Q_*u6!{q^WIHyZO%VeoQmSIMV?==GvoVu
zEB%Zl0QI%uAru*?znGquR9=98i~rC3hJ*mI?+So;<y>R}z^U{Q*q97<fTLRfoa)lX
z2+dfU-1#Fg?H}s<dwNSt_CER}_4>4sZD(ezk>?WFhSFf@l^^Ks94eWw5Nh8V5FBtc
z2I@QC%Jt>?g}L91fw2;$>9_iqoKvr^MskfA$wK|WnCFLyM8U%o@Y^M>SA9Je+@2rC
zArnIvLLwFp3~)vO;F4H9-N)mXqh{>%xUJ>#o-05PHu?9{r_<f{?F)SlHx1fb#fakd
z(i~J?zg^7wslC2aR~1MlMz9ztOLBEsyIs!PL;wnwkz)kx^FsI%f#;Hn6dozUPP;Gi
zmqBx6MOo7)DORSYVK{;s)i*Nos|-!Uu!8h_1()v(yfF}@;atFk7h48(!Jutm7J;=M
z!FvtI&0NUFyzSy@**EtH_mTVTfjBSyg)G1!0wV2KUIEn~V|ZCtvvHYx-UhXAI64NV
zHlIAlx>VUC0G6~UObps93$_ii8ng6Hcrkcxe~{Q5C=up@#TTO<*W5Md2S@N57$3xh
zacz1$2j{n0NWG>xlU|gOa7r5G())7V(9yrIxt`M@geUwLLCh6j3~QqYPJIWh;afuA
zvg#@-YgP^!19>oMPU}d%G=lUC1`B_o(w`Fh#Z8Nf{^atz+r4@b+j7tZaBGUy#=Oj*
z3UfMW2K0hOAG=7~cx(X%#sM|x8A#4xjNxXc*|;#*%kId!gx0a!Dap5F{oA2|K7lO=
zi%%%ls8mx*#USE9PV9WH746}f$2;toAtd|goKRW^p%>iDyWl*(sg?Pw1-~@K0cg8w
z+xR$qKiKgbUzF?&ak$siMz(zpAS!=1Y61YSYMS}+3P;m;X0m5uH;t<G0$@L*U*&?f
z=g1<p80)~XdbNE7&KWw9j(=@py@a_)a2C*mtgt<};LjvK?Y%#7X##`=%e188uCDuT
zyG{olLhw|xFcwgUHuL*N#&y?DmgA)1^F}=^<@o;p4vh8aP}VnwaDK98B{Z(nzkoRy
z011@r!vvl{1`pcf`VF^j$q#aVZEfCbi#Jw0s7+3Wb3hLk+wqn;a^t83(+zBo1#jk0
z4Pt?_xW-S61PU5JKtmye?M4Q+mRF$1db3moB06dZQuWhemE;_a*qfTh7Z|WtiZK(O
z_C!BtACA;jtc5Z5nqtX7oMn#H3YZ8J>P-=xvF&(9nY1UFbbJEziR68_r8MB>*WZ=}
z3VvznWFbk)1il*tFex&A$Oolw1p1rN7)n3Q3lY{%g^Prdbs+%S3%+!r?RZxkhsR(P
zm`rrRsOg!0$raE(3YJ`nIBDk^769DK4_{TUiaqK>9nT(!Jorb^0pJtcZNUav#=1El
z1LqqA<WEKEJOgo^INEf{Wz&Ipve{ob56c@BJ`un034(IbjtkvxD@^03e&0$S^i+|h
zMKUS_MhO#Jb1!IjT5i_un$u`A_W@NYaMRrT{_DtnP2Z6b<su^I4W<*EKnfLu(W?>g
zxj^AtAXz+-V&16wAs~z7CnrXbUQaQ_8UjfqZ=4Mx4BqhEs!fu`4k*K<629*utqZbL
zmS62OD$_0}Ml63EU{25Ar0Xp(LUIg_uOarA=xI9&c)HuU980Wuc4G?Byv;j-*vtjv
zR0sB;a3%DSImlpX1`V`&wXRWrQIvi(P0HidHwwId%lv9aVs?Wf#y+DpY+EDK6y3DI
zT7W0q!Z0V9{LE(JhxF2)*s0gK+jnFJb=&CXfAk$HtUT#0zi<W8Zv0vovKnHLcGu@o
zs3dC%E)YM(;r#JErhHUDN#s=2sTOes!dK#V`18uR8<T;|t_UYPSlo57kSOw5!qk{Z
zje1QW3fOO|h{j5e@8eG(7fPW}=CV3ohAFfdHSMp0QGTm8?Hy4JcJc^nP71FO@r0?u
ze3B=S!W!a7A3ueC`uZXG^FT&RoHY6-WFXOE0{<jot9nJ|u{XD4RY^&~?YXuQD*j-g
z^Q7_H13%E|JnE3R6a-rjLs?APSGq+;v?D@P&9`$7Dfs0U$7Anf)SqfO_=SSq%yIit
z4unBHrD+*+n?@qfcn4w3MgkFfwmzf#F`-bC;FdJjF!Gt{5Ne|WXHOu?q}Q+=AYF1D
zq!H<HXi6!y%Sa6sS{(|Y4h9SFk{C}|6X9QKkZjL}5djP+Um^AfQ);ecV|KRp9?NRb
zUFE1i=3A{@zG8=UbV@D|*8yv>nLCS;*)~7Ob-pn-9i~W~>t8iu15R<gYgwl>LawEh
zdvp7V#5Mn_=t+9aavWV$*ayK!>Q)fjo*o%4-(h@kkAi+tmHc+}HYyM%g6%Xmi?RKF
zn=E1J1Aih?vBN1g&3X2M@!YxYF8W0%G*DQ$ka9ilr8d>hT*E<WfVDB95z0e`b5XYJ
z`4F<l^DZt+S$C3iflZ#wfHPo^fq`T{1?av1-=0uGntcu=W=QS@@_onlOiQ7dbu<9_
zGZ#W}(t(U$U*RR-O+CoAC^?8=cNO8pf1hwN4h;wW0Hq7?#1zX6IxSkrUe)Wrs03)L
zva`FNBKK{<paqHyFBqc%>Y$$JT#{CJ*Vj}bEwtLNGBQR}jI)-onhKXmm-Qf#=KN0U
zsg6pw{nJ$v0gNaLy0}<I&{c6%Qi07p#g9sHOITc1J-N?21y%N=FBL)Y;-|2Jt-Vh$
z1RAShviayAhaYhB3n3dGjVx_~>Qw$ivc%XdzOWx?0BMO2%FMGEHppsgKprgRp^%Ic
z3*N4yMqFGWYHwAW^Ze}R{F`=>WUIiEn%NjghBcsiQIQVLYW|%5mG;J~rQYl2&i~ho
zc_hcAiXba#J9V@JWgUq_1F0-YaXV^G`o)r6+g#GbpM@;wXR1THtpKd?97&x^5quN_
z!&^*z-_34(63rU2FemuZxqEjZUb0zBK>hzx<_HbwHpgo$PQk02bI$+RA*vHF^2J^`
z{kUF@DPqAYV(Bf!P|*r4V0~>kh1Bsisy@;`pNr>9T7Nrh*M}5Rygj#u{IS&)YV}-s
z!s<ej-487)ztw|Wpe$SVPjX{1Szl?uy2Q4LpMu)5JPnpAB615Pwj@uY=V;}7D4O%0
z2$UHIeLbd%^EOCv$easaz>4BPY@<SQ-Y}E88os#`T(C=AmfZId<ST(hp>wAJT88>L
zFL`0gb2AU`XePaKp_2Mym!L>_E#tewRzwA`Z*+>lGO$o)99E#?zTZhrWYmlPYJEuq
z4cPD>%;jr?qcU;0TvJzcOS@y`=2nmdaV+r{5i$~-Enf|g3!<w>X8$Z@^GQ@#11N`f
zx5XrDxT^@*r=b;UF?1u9#mbwBwqxuylu_5DS3q3dkk}FZd(#1urc)9U0Zd09P&7gd
z#;De(IX7=w$`mC>XpU1fGP%{g424{N3`1xME1F^(P#NxJsH=t&B-^y*O#s1eq;g=C
zXa*4l&U*6J_qR;bx~J|y!;uM@IDBhWuPoVOp1CE9k|E>Up_hBBrk_B#kPeT1ir%-k
z>V_sVIA`WC#}|!)%S2Tu#;oM$c*tK={7oB7y;NBbXXT5SUi`v;<`*Xgdf>mTsRUW;
z_^|vmt_1|;n1lj_7{yvBe^|Q)-bi7zF-+j^d+Un%<l6K#mT?aZg`G`t#r*!<&L_;R
zHF{U?Ro}>VfIW|ohp_WeTf1^sQtIGcJ#g_5Rx@6rI~u_-JaRuJ)Fwsf?UmLPG={ZZ
zw8}{?Ow3*^{aep`vfJ8bT-T$w=arE&G1EZwS8Bn{bb^4CQQVSo-EnS}+&8}&zbUNP
zNs4Ip`euL}^P8dp)^qF0SsAI%YIZ|t_i5YXQ%FRs%%V68c(%*Sp=JE&>kIvDvu46e
zy*@&Z0?kEDUtd5o{{ftDQ{+2_rYlR)&%>=cU!XIz^7?`oP>KvwB7IK-ZlM3&Juzg6
zDgF>MGK`JqgAj677+KN}E}<K1oW2ATQXU?2IXNv5HftpgYh)FtdnV5;)?lO-Ide1T
zF$i!5psf2>#J%dSgo;78*D3xeC8{F_?F!M`%J^91_tJ(hHSu}AZ;$Q7OhQ5a%q2sU
zfy0PX2ny<qhIaC<u_Y^mS(Q;zyf9xyZk=pxkqPk_+fTcFYC7=)>K7xq$)52>aqqkV
zHnchq%jXjjW`s?-><n}s2Ic4=?P{k4{RpiK{T<Q4K#6JYxCHGj#+0wLV9|dsjbyVA
zi(LW=$v86V_-|u*JVqM`5dn>D^N|Z^LR9}XRCRh>oVz44kgPW&;Yi>4jDKVcOlcN#
zsj&#oJkiL#R7gr?`8IBL8`{>a^>~nU;io;Xv8?D^5HOaevoq1tPjq3i8XpU;)|Awq
z0=f3SS_e)GC-ZwY&QJd+4^0aQO_AC~l@+qa$R}Q7vS95Yp-8}36V|>F>$Gy?>nL!T
z;mEcuD4=|gkusWuS^*O+cd4CPT?W(UQI{lnPJ-^;3D_JqA2~K{X0-#>DGyPDyu;)H
zxM>MMU+C4+CH?rers1yQNTH4$NY^i;f{b1LeL*q-8zT|I*+INGX^4Um0sC#RbehJ>
z&C8f4eYk-*N|FNkeSd(e<OgeP)tu#u>$yZEj}LnRku48$VS}=P6sa@MD|T0ku?x$@
zITT0+`PfWwPaQ!wfeEm6_vfQ2XoJYBj5`A%P$mJQBxq~7k!j2?#?cUk@MLuHx)<w5
zqrE^Lnu!xQ*iqb$HOT6*qBvRBySGi*0R00m>N+w<`DxzamB?j9ksqZ|+NDaan4%>y
zd1oA^9YXC!vDrcm>e^EqXtUVP!Kc#s0k(F)Ks+Tifsja|t;8kE<^+CQ=h{W`x@DHR
znl<n8dlZIVVeS>^NW0CtfL6_J$A^RMJsa*M!Klhb5kbe;$nFBKWvVC+VKe7>*;dKv
zcv@TK70EvX5~H&r-{Dan8wO`D<jjsQ3Q`#0&*vo_U33dn;5^mzeZ9g8^62XR_}^%_
zGAF<qxQz3&>4^wv^GC<sx>idwB;m0P&}ElpTL9BG9jWU_AXMv>(n}wjv$*pEFkQuV
zlrA+-K;I~J&<$KGbQV+_(EZ=mAww4B4eR^ltCkVL#bHRTod9cyBFr%X`NMoHZv&Tf
zW%rM9s3_Hul{c>GaZuqBNfNy^1wfNIuFo;B{5VlecV=`N)g=K3=VCAlV>Lq6qZ0V)
z!xSujPv0}u5-hK38mvRe0?Ut5ZgEM-W<Mm(hoUZb!Ppwt85KSbMnB&sL;HrdAs&kp
zz3%k$9hV9bVj}0emRWKLJrDDm5c?tv<|Kva-s)$TxVDW>6tat|G7jG3>MUXr2Sy~m
z>h;WTTfNQq(?;-c(u&q{w4mKOoX>EYd?IGVgvT=^*Ka1?I;LTJexdNr!R9}mLlabz
z4=0-id{P_i(lQdG;uLD9qDo0v?fhVSyUFiSU*(wotq5$aLV!3by2tTqe>~pqjx{@F
zw4HZoYYoFd?Hixn0E%P7q3qvo;&5mV&0bK1XN{eCdzDklnn7)@CgWnUS;68XRfOc0
z|5;~a|9~#=kEa-3bapIz{9_)iBrqkUw6GHQnSA2KbEXjDkql)DOTQ9jFjHalZ-7$Q
zdG9KD?leKy{0I<FyzyMZ@BapssVp$W3oUtD>=pKZpp!n&vDJ|405<H-C?RhZynQTx
z>yqM0-G4L?7W_4XT#yiD<H+eoNLFs=t5J8CkS-dG$JEN=oIZt$5u0E%j=g|H6F(<X
zL~;DJtm2|?w;<uXo*Go4D22fW>CBqQY{S%gej@>BG$xtL+X)Cy!lTry*xt-$x^Ie2
z67*-SQhG_KSI*9RWG!(9S<!e+4A6&Y+hjR%+=!L@x&v@)L0o`lWG{}OT8Dyual&(!
zGfidmZTw^$r%HY1G`|`gJoxofd5oHAcOXub;%`Cnu6GeaTnoka%i8LSe}Hfgx(D}V
zALGKGsyn%1pLj}2_G*d^OZe4f#j|5RJzFJO=W=8jb|aVrBRSWB3N$71V*Kh$JhFVT
z0Xy-+qur^Vf7%VXqXq0WjUMb6@WFig3M!N+-+f|U$hvRwA5F&13TvPGfDt<<InMn8
zv-Rn`RLZ{^S?_|{U24J0z11H67~yBOTNY?zozD;7nFc7vt0VT4rR1S20=Fn2BzT=3
z7_j``M~F2LA)0xe_t~dsi_hq;R*dl07#694Puj|H{0fjsOkgPAL|a3s;p}TmgEx?8
z1gg6@%s3nwmK*q0?8V6VwmbTwg6e|R-);f)mY`Kug^srnYNRs386B=Ee|NR&r>Huo
zRBs<{FsfxB!b)sZkTSA@>uCqATv%l?ZSp7X06YtBy*Ihu`{fk+yb2Y7gaLXOmJ!v-
zf}cXZIL*WJ0pNf#qMf%6>&>g&<3RPmx<)QxiHO}V<v4$rZy}nqS&sNE>b0S3BrqqI
zJmXPtqb_7K;FM8=Tf`iF7XSgHDBQvn&Y4VD+*rwRqL66$f_mWWKOCijv1K9P*_3%C
z57p*&6){hUF<e8+12uJEg<#)&f5peQ9YNJyjItzh#>adFtm30qu+6@5y|D)N*lmgC
zCw-`FjQBk=nP%H^oAYh2@R|2m6~3>Vh5D^gS35O0JU(b|`R}3cY%eaQ4q_8qIEsX6
zu_l(mXhGEFgeP1wi=*prbSr_4sMc)e@S1o!+k}$uh{sZdX<R`Qzz@V<-7`?U8}tM!
zupA`r$L-RL1daB|nuVozC&K}24Pm#Pw3fpzOiTBF!bD4r7^^7lRk_xa+-_5vrKS4T
z*$gT-{Bc2u221%Z)$$MWq(=M`l01kwh|ktu^mTt3Qxtj$)L(;wH~QPqcFaY)s;dEX
zkWmslNy$g!gEKpH+q<va+rsrxU=w}q$}b}=Qb+$R4-M%$W(ql2%(b3da4Pvay1+8P
zi=f7a%W9`YLJPD7mAEw|1NRe0$#C*3jZXoRk%8MZ#l!|^YwD+?-r@3r9Qj?{h90N<
znv7d@(t+Dz<DVk~K}*k1K1RgplaBp4rW+NsCKy|IBxC1GbErL->Xq6NxOKv?DYrvp
zwIFwwaMS{RnvuBLVvX6Nw4wurSgmyEiDoi-r8qTA!(-9L5;YiA7=jH6p~HaNm>A;=
z^J<0bMBBbeVuVAs;HsQX&eq`sbR`SWMu$Y3IyGu`6W$~dqN-Wu7JFjb9K);Gvs0VR
z-Bw^@WaLlMz`x<3&tM8LMA<l#jgJ7S)ABY0@jg{ZY^DBFxyxDhHF+K1<7dHoLPXW6
zv$g}H+&o~5aCYu4!p-)6=k7mmPl&^~$;Zn;++5D@H{DMC0_Q)!w27CrXi9LW#;2qu
z^@p*>6M+I~ajNaiNXyYIk+Usmwn@h^UD1)?PdugnDE%p~A`k$Iw}#1+X!s&Zj1Y4%
z7o<OI{OzFsT@0s0SbYO;^CA%a=a8Ohz#L##T6_MTmO3a*`Y)p2+t`r*!+5CIgYo%e
z7p4r&Q!pe}CU-g*r!I}MLK=?&4;2?J_}vqFg+kd#35#wwkYG(*pW~Q7&j$NJDL00-
zkShrDCbaCpS-_*;2V_4oD0PX#a=TbSZ(MFPyPHlIqLM%_sf0=6^DbnDj<0*H;anD_
zAz#Lbj+<}YfI=>z7BH2Yp5*xKmdUI)bkr5cyH&I*YvE7}xKP3OC3F*S2@prx*qQyd
zGr!Pg$zf9L&x6$SQ3#P0PbbcuqV_1R(c9RIbO;b%!7ymvnfPDOjpE1@N7Jn}`YWJB
z=VQ{f<ORG}*!c*kJyAhmk&bBbC^<4<0xlV&Sgg_==c<DE9Z)Do<PXrWyOBc=$OmHs
z{p@%wT-|{-31w9-HN}hrW=g}KfUQqX%l{Rdp$U~xEd=@<lkZ~$U~J^HO_GGkS|n)A
zt*b3Ip{V~xAyh!)%bs|BW7idsFzaq)j1|nnw*TNBYmvhNV6uPQrJ!HP&|r3PHvTrZ
zn~^gjDEqa`xbYR#)D5Vq@RD_RbKfPRsvDw`U*4R*ArRqyvnd{NAJQJlB7;&)mz0(|
z4_`Xe^MSU1o`>9WT;0epWz%&hPV-ji)t!>qF+{@bToQ~E9lxthwos`~Yb}tIQ*hc_
zqB$ny-C!B*DXN%6U2^qKUtPlmOuHhk%(4js&-rAG6iMXLYM`M$xkJxE$ga|NapDu3
zG2T>hK$NFUQhhMr*yi9iwOmJopi;3>a!#kLD4doyJuX>mR2Vn$L_cX~t1c$LXq4)(
zFbIgi#s**ZS05UpRpt-;>a5Vw5b3?G?F=IqFR03oFBnxgA^^|dYMi6*ssqRQN4Q#4
zh7)KKm>*#IqBM<JT8tHcX%PFZDT;bBxkT^97mm*9jgfUcS_C9Fj>q24oN3ppTaz>&
z&>}3H6+1}slhexS@irC4jXMuae!QORm$YVkJtODdcnP_T_EF3(kw=$_QAI1`VtyZj
z0XQiDog5%dNAS;;#LIcd#70TXBPm@fYodrhUMM_?*Z}(b=B0c>?MXVx(g!I~_qtPL
zqc`jJl<Z@qnv`P_<jE6Wv#5Ot5OjHC?1kR^!^C`4*)RA!pB%9uiR3m!w4S)<1bxGQ
z0}o#3UWT-;&1E!j;W<+2``N6`G~b=M;2Nd5i~}2asVOh@+XLn5l&h9N$tXNs-0io2
znHl6SW?;^@I)o>~kKI4o-K8sQMDk65gy%mj0rxEakUa86Q=zW#H@Nz?MEoL@>!^IT
ztF1o1`_644e^A9Bo&{ugtr*1NsLMkkB)Vvhny22$CY2I4hEKpo#>e!<Bvb9YCD}5s
zwKMj%JmgLe14sYI^rh{Q?|VZRd3c^pMWwC5dd*2{7vB>B6d<|*FCkMdOYWz47w|hz
z^0<hz+xi_T8SS{W8TNPRACFNt_1x_W_5NHjyX{45joX|rTenm3_MeR`k?wdvsGjxT
zMw%(wmH5q))dg#K+1KLjkc20_CXNmPX{Kyd+Se`r#TIcHV7tD#v*T*4T{&^JgPnm-
zpv4Buc5OEF<rt5ja59hYb<!~DCrbwtieObkX);S52*#_R+@r)Z4e&{0Zd}sVnGsF#
zPr1cqbY#v$%(=aCAGEVq+rI`_x`Za~A(4SBjfU*wGX$Dtbmm2|S{cc&-_LfP(N)1%
z_LBWVuE^i>1WZ56Anr|Ljd(5_pJIQPSw%kr5n`p>g|xmxJ7*GE=gH>wGy%)w>3=1b
zqoPuYc>0L4NeOsRehIBUO8igKjYnX*nyjo12_s9_4(x~+4^l>~c*bX8*_=*G&LO=d
zo0#Sml>Z!hNm+!tPQ;?!AW=YAeR?&~)QW~y0o9bEj3-MYhY6yxT+#FKRrhYk6nKpY
z%<?mGN54||0AnXG!q7ii7Pjl~hj5;nll^18o?bVkpd|Y!n~NgWhD(!Bw<*R`jXnt^
z15BXh;#`x{$^5eD=qbi$!d6Q5KxAZ4K!M0Eyq?0@(tmDsB%5V15jJrC`#4OC){l6c
zU;@iBh5rOOEdho6t<2x#ewI7_^<Q#_llK1EmdbL{u&-+N3A9SeJ+ZB3ut3%jmux>>
zWa2v*KiZQr5;5XB8$R=@(#ZuIK3e!XDkDP=d&O;I^B_Xv05arhA+0C!m7k{Z1H?VE
zEhjD|{@sH<Iqz`^gcr+<E|>|~g(&lDQvAeSS?-X7EnJ-q*;fH;w@o=O;Dev@nzJQG
zc>siwUKlJt6JcD9hGtJvpmer*Dn2D&8_;7_jS><qmZTd~tP6}-Vt*)ncuitUTP8JN
znlkj<)gADE;QKf0M|h|EKz<Y0qSGr*PtR6f6uqSeTx$M%tQSbg28&DrWD}m|bQCr~
zsRm%bCRyk+zm|?fw9Zd0o!xVUI#m6HdJ<7<3qN=Xj7dqO6w{g`rAV@cy;Y<~Ii>Q!
z^G_1*L8%B95=rkjaX{8FNm)|w2d1Auov#8N7~-gZz?_uque|B-))zRRct8HDByjxs
z11G8_R*?wEoJIwvOOu!PkCTaK86$tT#ybR48Wspa-9{fD9g86AuKf{FBSW2ONuiA_
zC)o$KYvlnMZ^d9JHN%7+a6IuJ>5>bUYi!XT<Y*Krq6Gd^I{i+^h^!1h1OxhKb;IFp
z%WZ47=d7BypmT@8lw}yVDDSd@Vkk-~`o84cXSy!?7a{jEa$A1!y(k>712M$7-i&5M
zKJ~ni<UMB%eMXX|?X^oX@g<^WmH-`!u+NUNZ`O(rPE}q9D8KbC#huElcD~j?Da6Kc
z#_vJzzj1UWa%7-*bNQVBQCT%lF!|~(RPv(e_74|dL5{HF2}iN5V-jd(_|+(X3*VZG
zqxEE{AW%p7EH(ZdI@A6}Yl$;#f^|;lHm_NH^btO3Sh9<>0mgog40DQlV+y`Oeuc|H
z^gkv@8QSL#_LLsj{_M;F7%rIN?WdhJzjX8+?(xg3&H`>=eP%HE;$hm0{?shqdI=xk
znSIMF?*gNo8LqviWiTy+(o!ckac?dF2+o(3COrLJ9tXk}e2kibf_N5>u0(f7Km6|e
z74u1v_=vp!C$6Rnl8*~S^MZ0Zc$D+kGOy}dip?tr<7`3?$|0LtG$^3EH-zGSPA&9|
zdH#77W{Z!J0fu88o(X8sfaD1(t8?mN#PpRNO3=x`w)x_6tuDc}8U=rK=!JQJ`XZYq
z#rx$g%k#L?UDFLZ?Kn`x3ya#+QV4=Y(H1#jnB$#EbxOhfuf269`S4aIKatJFL#F-m
z(;nyCkv~nP6+!ZUm9#YdKfkTK7If7iRq07wJ<y>sEuJ=iSH(Zq08(;obsfIZ(D_SK
ztN!7O=wKR4k5rTz0rv&VfNr#`0+RzfOH`32bqQDP2(iaLD&P?40qGFa_*Z(B(fT=S
zXWBdD-4?gpavc;!b|HiFM>H;i_?hMhH=^Mzt#^~Ee>AUzCj{%mDlaUw#g%0T-veqK
zm?wbw;3)Cr>HqHY>s2z^v85RO!l5#hGmC$z$&V|&QT#~V|Ko&_x8RY3HbrNuZ$0>L
z2kq9mgDzP8XubbgfKi^1hEUd9uEhynJF0b;&yJ(77xOF@FIrV4ch|o~`~G>Q{_w%1
zX*bI&VDz6?I}t`Ciq&GReStJ+c|eHFBHtkfL-<tL9ZQ{65&VO^DcvMLFC_vyCgQ{<
zoo8pJgCS2q;VJj3^@TlA>!qY<m8U$9fy^J|bCEfLdR9*R6X`#Uo*)va+<wc2;WDt0
z0S5iBclM{Ivyqd2zsJmLzJ@8q=Dkk09ufmmCpQ&yWiRC+v8ND8SLX?AhGaSLLA=MU
z1x*AAo1Q5DPPPfu4f=JeFqCz4n(=A^^_<XAcS<)AGGTScwa6cXB7AcF>G+f?SFJ^W
zN0}|iD!0V4oKlEuWSPwaKtU$Og)LG1c$+;>IthQE4!HPaC}n?^JIJ11mlZKADRSCU
zw)FZv8jIR+jC$1zOBqp6*_i<K5+GYuc~kc3jj<m{)eGverOC8&9wj1`mQtQFX9q1j
zSd2GcGaY&hE&??q58gTK&2;$hftUT5_BU6B^EMJEObQP)T|7jhuV)+_{t|)I)7^!B
zcF(j;=^_d8aDI%||4rDy%3#my1VqK|`9smTrY|khi$Orf_+$}D$eb|pBCRO@>^OFV
z?1z-PXqo2zoHx#yQVz9dvA)YkVn`s2YW;t0m;Wy4RPhkes!l~oH*}r9wc0tn*7+^8
zU|U3QzW6rry1NF2aLmlm2ZM*l`ZO_$5UgzILw-*6uL0_Rb9T2-gmqEylfB_KnwI^j
z#8v%?BQ+tx>d#g^J*tJIM&48cJbDA(0)h8((%^~e>kXGOLF!Wl+6~Shi5RR^B3ijK
z22mCqSl)owz&zQ#fp)wXn7y$N*8L9FXsh{({++(b1Jge$P|bd@&Mzh6xCZ-;Qh95W
zyYl9oS3RoD6dqB96=Ds_cRJj6$i!dOvKUEE1$BN(oM#VgaYE<JE!755xQbzdM6{u}
zV3xR38nO@7A8?}*y8J;KAnbut8*dW>0ZKw8gFkTz5Il+VH;kb$Mw`b4n8?Kk8t#6s
zIUonVJB&-NgGg^v`QY{K;PR0ha6V2C!dQ^f>|fpeg|+zaDLy^`2jn67y%A$>H4l_+
zY({!G_~Q~Y(2zz|IZJSlFO~SE?xNE_O-P=-7~NOs(_GHjvrukQvC9efo@u0&pG6V^
z^J(PKicD15DQ3I_cndyDLbQxgT;67qLyXKVa*KJZyRQ4HDC^n&i&<PdPClh7>lWj2
ztvlo^`o`~z#P<b$+Rv6B%}PM<UTFShY*cA@fN;`rJ3;5I1&=U9(mPh)9(=Q2m+rGq
zzNEIv@krxrV|ZE^=%*@#FfDNlBT2yB+xW-#y8o@QItVk=QI4I$XL&)L!sc(O)6K<c
zd3IcYItk)Cb<{`)*z8C?rk!wt^6dLKF&fueIXEmb#P|NY_x-Pn6a$ICf#y6PU7qCU
z28x8Z^531{UdIKlxpNRG$ZE;Ge2Mwksq2-Ra1H=jku|oX=dx8@hW8aSEJYb9eP`sx
zx23pi;z+C06kR0iuf(qimG%JFMXL9h4EnrxZm%q4Br~IOlQC9^yxqimC5+LM9kgZ5
z)DE{qYZ?3$WTCBnp0PUcHA{0<_;Styced#_XJFKuNN>1T49A&;T*q{{2k9Hod36sm
z8EGoE{z?uQR9Q^I9Y7r=AidVAT=+8c>qOx*I)>GP{aEV`V5#ys0sK!CC&E#)qAbZL
z0Aj6a-6ci%>1jU){kNg~MLRx5c;Up`LQ^_Zf`o5MsV_~xYL}=Y5F(=P5J_`{Vh}M0
z=VQvClql=&C?zr|MJRO;AgkPDDiI=L|9!~Yy3e^d^gjan&O?8PwzHeod4sKA;%?TP
z)>|!G3ET%^!dyqp@vSFNS70X*;cWwGbADO>`uyP*wli58aqzd-VYMg9<7ulncYUbv
z0ElCM{WaKsi`Y?|-@NY?!SrgB6tg2iyZ}ub*I7a=x%sr8LCN0Nez-R(rXz+Esu$it
zn61$BaXnGd8Ysm)BV8n@Y`j}Eu1YjrJK|jb%K~T{eN7N0eE}q*U>qIoQ02{`T*fC5
zYQe%tjWeW-<^}zqEw03e**oGGsP#|SjO)0)lzz@}@|d36{X>!bI5I1)M03ALNauw=
zzY83nA}g(u8M7QVwh8f{a^9b-;bkG|7w;>5yL=us=MJ=F2d=FdQjXs!UWsk|wtL#1
zV)+}b7k%=vVfx(=AUvyW{-kz}vtSCOq>lnDgR|p_WsRH`Vx)KDqhi!Y+CRT#pl9}z
zx7V0GX6TT+-M#d%fA9T5_2H$zAcdRlqWP{njTOXf>9vwOaCJ@1@MT!!`kfAB$dVm_
zUzmR?gjBVjs??BCU!7^}jh9G+Xts|om3j{?_cg)wcgp+)z=05L-8-#J*PI*B5IcBA
z*f8PJCwRN$(oanaSPf5J_GP8KZ^?^1uKuuF2jB*#tdK2G0=<>xP5TXd8*W4rL^%J9
z`bsVKUU4rW3v1A>NL0;9eWSV~v|aT>S;7ZQLKTwtktm~6_fI9$PeR8U8GY(<kUtL?
z#g{ywNa+}q%~eq$K4wA!#DBlY>QU}hmCv}hZ-Qi}I-iKm9%-%4xO<1&-v~W1%mYHX
zz_i`1c6Y(6V$;Wss+%ES)vN8oC&J^W?TmmGi#9T=$C4&sUgFw0{zPo~Xq5$@-#xYH
z8Msj)KAkSy$GlqvB0)!fx(&NrybqI?Mv?p9SH2NaTY@!^yP18%P}8Kx{aCdTQ?h9s
zJKRgSNw0peT@){ful#)L4`Wyp!Z5IBI9_1%ryo9SyK_2|=W{E_ZerzG8fseZ)!oDn
z7j*h9eAa7vM$om1J-HcUv|0GV0MM>D+_*P(y1_|bu)6Vv!<)X}ccxR_qyj}W_W?mq
z@xI>g&%&hKM#DGhyPK!FB`6dmjCicLoFG0Nz&U|-AYo_3FpZ!RZ5AYq^lEPg^SB+!
zmGpibj{sjQy5Way7IGF*apCZr1jgQ({dzJm<hLvTUaMk9qf6z^o8I*Pi>=V$;48D0
zCEH=rS0kk2A()_Kq5~nHbKA4tQC99{|2I$YhVb{AuHohi_dyS_Cd9LZtJ=_YfO$k>
zpu}-?vR9?gi{22Mn6$H};a-ZSB_C%loVz27+Z}DT8RzAkOXues^G35t@!4)s#TIQr
z(qmL?(kwF1x^6r%IkD%oG5!{mbI+GCweU%QX2pE!S~3eXy6$o~e@-gtd9&0{nshLw
zqbV>qK5FPkc!}&x8YpMrJ!kJmZ?bN{5#rk$ed*?GrO={HGx++CEiEa1mhWag9L}`+
zzBSK3cVh`<1T_n~T^OT4{#(|14vOtGEbpm>*N04M8-9?sF*(57hv66{NBQQHW9hX-
zCiZW~!4$TV_}!EZ8sB|0K(pq&p0j{Q*(Z5AAaRz#ox6pv+xAC;wuFF7PLT8snp9g|
zG4+k)dW6)ySqfpc6;(I1+O3biZAg40MW^dm6UO6V7+>GbkDd0XABrk&CvtOkk3X>I
z_5nEKAT5@drPH?i@b{tm*c^>J#pi-6tJ4{IWG~2$CNhS8cY6yt%y1Y0PP1V8%NY#@
z{61QITu0xdpI&w(IhSWI`^2!`^@_v1cj&{5av^_OOHGITQ$-utN$nSRO7o20|5mVw
zQxxGP9bUv@|Nh3yzx0fCts(*<wDiXigT{?i@U@6Hx-hUE9O;f&0PLq01`Eq0KYqzq
z@G2Guzddo9vXjvMAbLZEagaE;d^neP=)5%<P;3gYeSC*+>LuK`Ekxa@qrAjt488L@
zoV;x;KXnT3c%92!(b*kN^H(0Ze~(fVy|w{V=8gR{z8sq(Pcl(j?S89IIIT81=wADT
zNAGT)+IilcN%n{jh%p>lwYJoeWK3M2tgRV^f>T6NB(?Wxbz>hO&ks+y7ghiZE&r0b
z^10@dRy?oZS-`7>(w22&Jc}GABO&mq<r(c$v5|kYbUz9c%05{E-~VooZD*wWi0~(Z
zlR|ZsU}@-0aKaPS@rEUP)YG#Be3NJF7oQQH#!nIyPMGDJ#r&0r(qH?D6!C8n<vX_l
zu;dzqF4AWz40K5U=T`->G)m~#9*MdB;0)hsZ2BBu!RV(m2HOmNJMDv~J{s$9ELf_V
zW?H-(XUE<65!Wc4!M;<NLC?VS=6^+)t}4we`XEtW*C~o#d545Np;nPEDVRwb;k#MA
zH#RQuwRM5vf2}U2-wVlc(R2woanB$p{skyoWL9l}*ZcMFNZUBt;+nv?j&X_jQ1AEy
zYs?xV0GNe+Lfu6{V)UUS1QolIL+wjgn>Dz>H!jQzzljn%he!6eV$+9xxkx19&`)49
z)GP1E-#-K;()lLdMG7fLE)lp)Y_icq-0%<V8ZQT4eb7@NfSI$zl-=izNYs-ts4k$b
zAaA^k?zGQzukP%s_LYq~`gnapcJMme73Z#)`nr=vEz94e8()^Nf?m%5d|Xj>4w-!+
z^-}IW;w!N)<B|F5)Ax@>;&bnaQnk@S3nK~4ls;LpVQTBuE9mvmUJgEm*pWW!6?;|%
zTQ&CAyhXd>x1bT^?y1BuWJ4a^L{C4|eX&neWA4m!)k^CBy`|NT^w#Es%7txuiLYsW
znNDe%m5~>u20O^W`HSZ*>&D60A0s*ZdTM{s5&-@}-&1IpcQkfQ$SJ6@DyGtB(bw}X
z$ydJt@a>fzPQ2_NZie{&n6!KPI><8N4Ii3?5>P=Q=#U2twv`CmSN?hP-S&O=!+Kz2
zVtUhE+MM$<jsBLmRT^6|q2t4fOrf2JWqekDUgYI0Ja-#;{@zf_*USFh_NsVxyWqU8
z*cq$;d7>f>CZdScM^jo};<I13dsVb~!z0R6u>=HW97ncny3?a#Wr{CVU*sVdNNC?I
zQr}HJ9Qd88Ezw^m;zfD!qFA&dCwNz_<scZyiDxYh5cIraIX3RV$D6o1)P)F~{1tp!
zL04*BOY+Y9!gbRV??S+}sq^?V&v$?^FZzti;)kzdRRw0+1^%HX^v}TD=pC;c-;wz}
z-3#WW2J$mP_SJ+?S`v@+>i{w-s+AmUMA1UUtK0cMAW?24PT~ZDYOA-+Z~psEG|eaK
zFhFd7vn3^H6~%2kNxEXS<?&q<X=EXoyLmD|qujU972rhv?*W8=6C@IWfE{vQGuFdE
z{4^_17S_m1)K!!soNL0Q(SM_c#7xplApKV)RQyM7F9%7AG6p?GRPJhUQUE~~{g>A$
zlN={FkKChO%vSlWDIuWUe*6n1tBb!>Pwh<cJt3}c+%ZLwOgE0qox$R)=-8#d`)S4X
zM~YIn_*=qj87<YMG}JEg<9OJxw67nLT|PYeCe%U;jc*)ajpJ90hW!_FC;?2BB{+VU
zEZDo6H#f3OSm=&{{<$%@V_zy*c7s4nx#!Oph|?;jLTp};3{<ir#|)*EX$@j)pU!qb
z=lyRSCIzkXW9?$56m78|zp84Usop?O2fV~037b``i^WH@TR4(&`G>OI#M@H^54iFU
ze_hqbC((M8(%ih|aGn={G~HZ+Qc*_>swPHbqQt#DpIcb-9=*E~JD{y_W#!{yr}a;&
z$$OM<JMN1kv!RGhw7N;5mi-;^msn;FPH(DJweSD7y_FN^_3X<Za6%1IkRZ}NT<C%#
zA>nWHwbRzENT)@`&4fL^Un6Updt>^SIM&cPO^2W-FbNjltiIRlY{5GF!26$5SfeKK
zM|0F^qixAC{`}sv@6p}4C2#m)U<DOCHlsY({~O}}Z10yaLd$uZjvw$}Rp%XjcL@$q
zK1#8&D&p$BioN#@czir>T_wY3&nTJt8t{=uAqVnM)@pS$x%0Dio-A<?_5vFA{@pF-
zk;e_~{d(0T6r)~M`gMorz>ehq_uTydjuHGMQ6P$AuM}X%ONZ*|5(SAwNi)g}e(sU>
zcpjvHvYGPe0$e?cw!S?s-U!r#nDKd`28xCuij@GI=i6wxSm~T@UdS}^e+oP_0Ypqh
zbdR&L<Nxz=|G#`c){b&~DRW&}<$vJd3Dnbb>Oy~>nd#j%zV&IwIOjyf)V@t?%oY5B
zfphVA0&qs=QArI=0p3qE2rxBy0yLOuuqTya#k9U1rdcy!SdoW;U#EHQs@F7>&X@>H
z9en$L7<&(>rnW6=SP(Ucm_&qtKp-GZDFRZI-c?kZv`g<OARrKW6$wZe6p-EpmEL;?
zA%Y-EM~XyxFVeq#?z`Uaz4zVU_mAOlgbXw%=j^@qT64`cXI<Ga#lOC)?^oC@*x)5L
zJ(mACasJ<nau6kw)(S<74MP$$m&l}@kr#qLnE-i*oh2|>e777UV<luILvoszsHp{c
z<nZ;KwxH+UX2Cv^Szb2u)gO;IszFNhI^YFei@Ij7T#62{OsRK*_tCDPQIr24vreK@
zm_RoCG*Hg6gcL9{?NfAaJ^kr7RBSN<$&!FcvYgOM0=Yep_&@>Pb7CHNJwO*MljP&?
zBU2ZoF7-d2<C}mJez@CFtV0w^qAFdh^eZk3fhS;OHGOxk4jpXt7>Hz(BdI${X5nDM
zDDPh4%~IC0udVVb(>jBB>MD?|j?0xl(4DXcLP*P7hHtjRY52_twJj3OfUJvm+H$`G
zP&5wS4Ogyg#=#lZ$4}%4ZxIwukZ~4BCWkr5e960ny`1=0eDXzK0Pzji;8+O@^!eQ%
z{0EYwqn&!%BqBBGpVzGi3lJxm&OU%vU<0Trsm2?gJ0)2pxmttTl^gxF4u)4D2{1lx
zU>*a>vns`LU&^Ml^z^oZ%jk{Hlp~#?R7*_<W0BnMf8DoQ8L<4vCv$cGqxAT@2Kk>a
zQk0-&!xp=R_#gNEKce!#?XFpR;tRyv=QU@p{P((Zpz{S^tIZrAYt+SA9Kr=WQw}wb
z!7ApG-n6b+Ydh39EvEJBZ214a(7hL6Wq97!>G}7g;Y~;bm{;c!0cO&Rw%+?X--9MB
z@UgrvXC##QUJKg1PJGhIfq9g67sudL@xE$4_hhEa6&$P|DGaHA=+8ll7`MhM>qfPK
z;MfvSeJO+1Di4Fm<s@}F2h~j0d#_EK1IQJt_vji(3JxK)5c8fS`b=GrQZJ@FJ%6cz
zMWgrsR(k(*@wtqQ7(>d7FLA$?hNKw@sHAVZ&!*z6!7$x>P9?MJ*GX%D7AY3M3&Wdp
zpO=6tUMZx7z<NJH`1=(R2Lc`;$~0SXe6+u`-o;bio2~$_bjy{NdD;MST^oQ5Y`?el
z><W`O`uXbWvkq&B@iN(rEB8<;CPTi%1OT;@^4EG<4lE>nu^T{Ltz`)8iS8u8c0j!3
zDrg{Pai0Ao$sA#(4I|3FpZEwWuJJSyVBYieAH3@5->wJW=cKn_TD;0;Y`>oRCmLT+
zmEqAIt5VgVeJj#+%Ng!U45LT1Ptw@cZhb8VfwZ$5=+t90ZArq5%|G(V7A5cXXD<TT
zBlUi#4)q6=moAWG-HkH_IluJ*Z&LY-B+qT_eW<4n9PW~AfIiFn(Y_^tE}r_DS|I1M
z4R*V;`+M#;Y!Ph<K>16;o|tHelr6z9zy7G|HcYGM5#tK41dw7o4iVo1LS6mbbw8(k
zb#-6f`|X-bAqMMN<AjR#-<$e@f+5JTi-3Vh3mW97^L*;}Ws(~Bx;RZBwJ2N$;+nuX
z5bb&#c;hn3UD?eW2vS5Y?c}_1Fqb*x6R0Scz-toV36x*X+s2427RV^>I6C(-9RF01
z>5#9xlZXV$QC@}ydwnsW{y1k+KX$kl?ma3WdWi*NSYPUc=KOVs|Fe|%*Y@{R0NcOL
z<UL!^uMkus3ErV%oO{6y)iZvwXJidOA!f-j6Dc_djO=gxeV(R7zP_j;-N0(+c;|O_
zB7hnkayYH0xi|myQ4WH>4?kzoU=EIeWTdQc3jntUU+KF&1RHwEI`tV<k*2<DD*%{-
z10tYshii|8`kCMkNelt+4%>a8b7Kkuan=%$ZWi?Tx!UIr#wQJi`*YN$N1=jw#p8!a
zIvrf%px|S|jnA*Nj7gyacG2hL`M@U^`6sne`Yx%+`K>-ho>K6#5kS+QrpyZKL18lt
zr0F!>xak-3-7<b;_drJ(7VNMg;0A~A7~e?R83dS(6-16x31<LTUj|^Z!F@1c?5H?q
zQN8u1E8{8s@Ue6Ke&QJ5J4~iVt|t1^WQa)sttv>{d!S&Er*`wjy(G~aqEAApd7BEz
z7QYYFA31%O)0xg8oBRPe%h_-gegIFqBTFgy@i9=yVG3b4V+>)}Z+DY{#g;FCu~jbX
z4KO)c3QRA$<ipW6&?IfTQX)_i!h-qK!2+XZYsiGsb9<<r0OX^&o%|Ru`Gsei4uC9&
zcPY5IF~KL_I|@!V9ldZ)e?UO4g0XEfxbLpJ;IJjL4nVNiar0%QwTW7N0~aJj3vEDv
z%{0sJO8J-w4Z3;(^{21PZ6>3!ye7sqx6VKcM$}IW;C)~dF<?(;8~ZV&n&!YK4JxtE
zC^ZU72lPdTQ{Ry%xf0^uxOaMk)Fp(Md*)s;bIm3gz@x`?Ovjpp-Ojd3_P>>R+^2AB
z#id<j5#XPJO5J(@-U)bLP&@qS?K7cLjoRri=sweF$bzRQXYRsYAJEB4z*;4j@Hj~4
zLPA-z+obNQ>RTywE)!jBZIwzu_;Oc9uZ(8rVH6Nk>1Dkei<l<S=Lw!W$?#J$gwpPD
z^sE6S7-{7+Qw1)s9YW45W8`=2MZXP<#5J$GHcz%o9NCw^laQn;c&*Z{uR0^Did>Xi
zcd^YsOM%`u!8YG|A5_H!K$qyoUb_rj?4nUWjn(Qoh5Gd4hLD}7%c|AV4flT(+;79@
zTGI+9_Cv!lg#bP&4-o#wV~N^3v2n@s#f~p<39P}|sMES)7*}wEryN8!5E~oNEMoHE
zqQw$3!|tsZ1->7jk3N`J`t00lEA>cQIRt4&9=-8-V{X-dKhFN&!*x~%I##)C#ePLA
zx6l)Qv5*wvHaH4s(}F;;$SaFh5?%AJ1xE<RU((RmCxa8iCr#(WezK6*k4P59xmcui
z0G9c{eX*^61pmbqa8OLk{Pg=t)_X23qW)lp2g4}qa|>vjGY3|zcV1nBBscY_qUQiK
zd47%D4^6d-yc*5Yq6ulH4&?}cP2TE2fg6}t#2hZl=WiEyEDmB<fKTW;#x2zgCWH)O
zL~vx?pI5=Z9XkZ+8E+@<#ECVQv`^lL3kCrV6e2PnJ0Kj3wq&(hFbYbVK8rYM-w5e+
z`JGv3<nh;lo!iY=A{7~8_XNLR?$9G9L?S+j5EO9VUPPb^L5lff1RQCau9g!(kOCw7
za8SV0uTHAkC>dz%9ixoD0Fx6oSu_Mu2I}8XkQf3_Y$3OK3zYGx(18)ac(L))mgIF=
z9UgG_VWKm61n%f+o+G~-JF&@0QW5BleCPqB$Ch4@8y7$_p2NNoIp5_BGQ7Uu6JZAM
zE1qBRRJ9)8;v~`fGx8AjtA46`A6_VOYj=?*!2%u{Sclr5>78m9-!2>$ux8AY;=~IV
zmjow%m<Et|5m3au(h^sImhkdGjmSpG9Dx(%kHKpP>uI-Js$q>Tg3?!HvjZ!ilQuFn
z-R^dv-{tLYM7pyww79}Zhx9$SzNF`w`L!93zXyKW%q25JC78o`kk<10d+o{^$Q1<7
z2J<yhq)r1j!3#0!>hLyG|8S08uaGYxn2&vsq97AHfxU*<xZcdqK{mQSd?WM%@d#uZ
zg3%yc8Z@*jA!F(_&-r|Vb0$;|=ujG5Z(`<O<^tzs;YSykdy6WpmL0X!MtwW^-OV)*
zt!r(3JJ-CpDkl7T4Pt{)+NP1Q-zJ|2R#x|iAtLHNEsLC*nFH8lXPL<RGaDTs>!%8w
zL*bnu4Mh`9(<3&ct2J+V8K969#>;;H!KTs$Vy3+eCBCox;=Re#GDs#q2F&Y2FmbND
z^Prpg!d;2&hR3@`u&vH)5T1qn$3tR5RyLP_j;_M4VhPvpjz^|XC=w4T<=el`Il~AF
zJ9`zvR6YKgcAUH02VLbzkZ(D}kumb27)>n_?Bxk*b$GG;mtXr)czy%+Ze7lZ-6luE
zR0jT(>stE_)QZ`1DZ7J|6t<aM$nirn>N!7Z!|Z7JMWtrEH6Xlv9No|+FcsiY8~i&z
zV=Aw4I7HsT0<Bla%H;KZX~*a-sKO#h7Y=4Ewo0H%94jU0XII=Zp6G(PxW-37AfR*f
z<R1j$BD|P`T)c8m5Cxlg-7LBW%3K_EmWn^S_4Cc!HOr~RkxBBQ22G<$;~VncoSexz
z4@B;~=hQ}nl?PLU`-(t7;Bi3S#X{n;HjREYx)B$!GiW3~%I+m$da4lLdNz)XI*o4d
zf|;FCR9tXJ5wOs_;SlZ@N9M_gnusR!E=w)$0iljyY}&E1SCZm!8c$^`D|H2`tH2)h
z#FX2(b97oQMxp8x$38gelMkm*^)Rm-TSHnfGJ2~FWPb`?+~=2`*xsD*D`F209_UNO
z%RPw;=1w~X6xlDm28KGXbg5A@hoe8lW5$3OrHc6;+q8eMXb~`WzT%MjhdhI$ik>zV
zJ-q{wwfr~3goekK4ycU&nK|Md;kqI7cr$L*vx3~Sxz%m0HgUmPd?+(>?wjZ1n^+Ay
zT_)4=sBr~k>s@T?Q<Wt&Ge3b7bCg74WdTojRCus4-wwN`v-nblasf7xOr4nr^|p<A
zv8e@UXzXhbo;^HGi#0MPptV0Q(R)`7vN}j95zv>5!QGWenwje6-p$(cn7L{YWcAQ_
z=dnh0Uap&=!*1d0I$trTn2BajyiUm(&&$b5!&yB}7Ew7ipT$-&J!$08FTvId5vA!V
zP~6F*+AzArKR1oXEz|S1iJ)-iXV5N&{ZT(O5(~%)Xz=s$`;u+(hKjDAbNY{p0bXz&
z^k6a`&#evGz2*E6QRTdTUkaTV7N<UlrQ9+MxhMD`mN3<UuXS#7>CwWw$X8$M;e)T%
zX*fN>IE4h&uvHBV6c8=G@KU6A&37n#(x&UtN~zf|mV}*1jJOf*`c;h7+zi|Gb0JB<
zBq4RBIREnR_tOr^>btbRf9F~UDdUt;<WWKM`30tTsgGbwqYJtO^dEPu<|pr(L=HBW
znJ)|Eb_oRDhB9YFWr3D1>$_V_EIYEXm1i2YT6!$Lqn}HAOeCG&z*SpUgIwC#=|*pp
zudgy(R(|=^m5popg$R9C{Zz5h1*H=0I7Gp7w-TSXBZ>%B#!F9l<n;379G}J#tvxJm
z5g25Z{*;Dz?g7&OJ>iu}#cI>12qY1vEcnw~t@AAbfwQB<x`FUqDq>Z}ozs*vG8M0Q
zjvr<xQnl}9ct;?ND^Yot$HK}#aoB8pYl~-~<iz)J^5AT%{hV+X4C55td{{@b*xXA(
z;$!)Pr?6*7&#q-!#zce3sx(3Wc$l9%GcT5rT`J?dK8qrQuVq^;tNF5azG6J{oz%L(
z7syW+<@`)%(dYShWGS0W=G9&yW}B+nN$>_Dk4Gy4q#F*L0?j@Id##M0O6S^`NgG3s
zU0f1MRuCB};plNv@+ciaZbM?~D#yc8-%=R(wy8$Myzjl-O8HjCL~^&sVl;z0#=y(<
zX8)RnQma*tcc}=jGP@q3ih-Ff;@K9ds<A5tDoQ(9uO#&5%G{0dqs^AEssf-2!8yq6
zSy1lpqyEtdtf`#krD3f4)~y*yOPo)V1VOfK>N@u)p}G1H19i5xyRyw8wIyPL5lVws
zS8Nn_!=*+=i~=mHV@LOG0~N;JTCF{>Veq}mWY{wE-+hmNRS<vweF%XTN0WKk&;M1m
zNM@2ChFmT2S}j`$$IK~e(W+=yTA80X)U_G|%2u*n>7rYq6{q3aE&^AxE~mxvncab<
zJ^wMpAcD0P1b9l3TB*s>S)wy{#6(R2eMNTg3+H4aX-^Q$a6Kp2?fb?%W7^3RXfMQf
z7R(!*_<5p-#Ro`6SL)Gov|NgTv#Vve@+#wX#B+>gW7#DZ*^4f<(z5;W(vmfB0!7kK
z7xTiy7eZABrDh|6XV+T%C;aN<s-q0}ZIyGvF|Yi@jIXxT!a|-mg-!9-uZ7y0uDA_U
zPsO!5?0SN;TM7}fsd2~e1ytK)hNq0J5eLP*Gfp!w;)!Ct_vp$anTD#FS2}GKS_F=K
zA%7p3Mj_7JxZ2qB=1?}nR4Q4c@g0!ZYXJ4nJB0mFz+T{kuf%a*J-ln2b7h`4_L<KT
z-e{ASq$K1xbeKWJ^n7OW&MQQO{i((kP?Kp;u*V3Nfmxg_L}}>#0s57oH$~-y0&E-p
zK@iruT4>%Efr#Ws%WNF-zbq~nk3xKEmK5?J*$!e%UZY$oYVA)9q|Hx+(Z$8Ts3?rZ
zZT}u%a!YQ|6>UN>NB@nA001#-gVwgD7ZRh{P>v@A7jA>o;+yU{WA1rgc=Drkg>XSs
zXe{pJySH;d*U3EIx+thph<P9Q9Q@&&S23ORXc3eL#UPRF9sR5^_HZYGYNxuKZ==B9
z=?5_D(+G9fD~Ueuv^xq=G<<DMoIB&<le{5cUsX8DJdTWex<@XGJa{p9nb1^anSS7;
z{w00#sk@!ucHL?d*~e$e1mcDQuE)-)K9vO>y1I|JeRxsD0{Abj3W5%VyBiT1<oRo#
z%8XE;IMW>y;Fc$Icj~cN!&gOB((l-rqQk!E!YqM64R-nNkNA>IcUUn%Hb00am~+XX
zifNo2q_SOwCx-3!0)p@3YZtzRWrjl4@aS}MW&1S;B~5~H(3@nkO?<OaJg!mz9{Omx
z_v)DHPTXEBjzIu^{BU`zOZSuca_fiD5w)?>*2GZ3;SOC{i{-+CE^6hPS5>aF?d<7f
z&cWFt<Ic9Ogh`H%wMKFh7BzQ1v_=y26>O3{w|RHU1~cDr?_Ec|mfaOS=UVlC^Rt|N
zV_&N)VSIHy&mR7qFuso$BYus0k|z>_{evdI;?)m2kJUy3rud^+ozn;v$MF^MoTtId
zc$p{iSUa<Gjp*5GKsB%m;opS2xCKrGgugFBbGtkjbX+kaPmYn8WMXR0)BFAi0y&Nz
zjEro`pkBTvi%rXSj8U040BaTfhX6rq52^*nR#gUt%4Ve{9eNF$7ly^87$DBC(Q@x+
z$QmMRsG@+oEoFHR8Mo<g_R>Ks+r*ssQ<#mWcKCzy<6WYgRreBv3t8gt?l|qu8ND{S
zZyaJ=XA@mO+UbJlZ6Ab`bGc)7;rFa#^su~pTSx^u73ZYFXFD?EV=fZP|D*6?fC?|T
zvzYo{N#hl^iSYw`b734#O@M?v(0Wk8Wg^K2-+bb=`xa@W<k95v5FQIyIS8gow$y_h
z?h)AevCpqUY53bj7fr@?@5^4?4H>2+J|FGI*kosGj-l%sc>853cfNOi<B}e3sX~)u
z!_L=81gL<s$*dCe^P5hK`lq_G&*@VVGwDmul0=!4N+r6No}28Fq$GBuz9rKlFEM8L
zl-#=ijS6N$x*~Yuh1c^F`IHz&V##&L+9;i;FTa{%5d$O8EYsxDbPD5?;aMRpz~#!Y
ze4yMx>Oypdrlf;yMkyy^7Xmd}8cMu*F{Y5e(z*SG`v<ca2bu4>7wunI*usctEZl=N
zBV~qhV~~0vB7^B`W}*ZTl%r`X0RnCxN2AORY{0k%=8_Gos^XtvDVvl?ti?SfCR}Vg
z_I*j;ogCLHkG=uq_!bCFb(3g%a-uO%#SP5tE}>_G5Cyq6-aY5#|58prl9!9lIp@Jt
zB$M)?VjzYT!?+RPPGY^MHNJ*e!_d`F_>ooVj@)Wl8rLzYF<ML=doCe7_agHA-#Fbr
zzSu+)9T#c--vR)JWb~k&l-Gk%rwSwkVMKhZ3bN_ts@?MKcX?(0!<#~()=mDTG?Pzt
zcA1Zj?-(hL#jApIEgA93pX%|eVRCX&<$0+$8awX>-#x$AqplvF1%Ei1;Xd;`qL(`b
z79T6-j+OtZ*hI5Fhn_Badj2MRld@n}*rx`ocZKrBibbwJ5`{$OGNP(<r;W?;t(<yc
zC^*7==-HXtOP2;)O`XC4NUoUvZFSLO57Odf(dC<@);ow&Nd}IM*Bp>Kn<`otmiubY
zoHp4@B_rAx`eKtUP8g46y1S8zLtmQo0{X?@9yj|u1}-si!M-~4-yWWu{kR&&fzNQ~
zfA_4zaqXeqlkh-t8O18Lz5>+F8?E*%YuvS%YnkG${j{mRq%6Y&q#`Zbr2F@^9?mbI
zuk-J8Q<h0dupGP%_IZS<rG44ks&DU2xkZgn^#x!H+Uw_}p+d)i`G;D3t5YrIrP#Go
zw2TMXmpN>yzHeij0Y{DEDL8DNL%^Jp^ht`F<@ppVULmDpnLxFi0Xg-$GE=3gjM_nW
zRkOOqoBF~EnVE^r5|hn8KFhz~+$1Ghh{~SD{GCwL-f0k%HbY-3n&cl1HVBkOOj15O
z%X#2t*@qXCH$&%@8KK7#Z!u+J!Dg%(`ExVUc^|CBxm$FJ_M6rH!DXwd9{!3!>j{Qf
zV8@!O>85qIit*aF+FOD*>Dxt+#7y#bO5Sles>gdFkQhk|5BgHYoYPW*qf$(yGn=%e
zbyG91dEgV;pan!*=@pV1fHtIG3FL=^NN9O=r*G)NxmhE$j3hu84)f^G)jm<W5bd$z
z{w??bY?49hV9{vX<{zNq&k+uzv^C^}5LLdNE3vNZ>r-=_q(ru0gK_`pBa+39k;YV(
z(MPJd$!8}o2VHe>VIRcH&{?4Mh4z0)9LM!`+ojpMR7~K-tg!oUt}O6^y=Zg<rtKJI
z?7rN(6-56SxHR^1*Rqh9(it5Mf&aJ4W+h4Y6W}ILI%uhznCF^^l<CQd^=*8%@IFgK
z(*<~!bq}KZdO#<NZ!yZbQ=TKJf<^~vTFurMa<n|zOWSQ5y5491`&Pj`)G82Dn_>T}
zvhcib2!^sgh&o!SVGP#8)XN%z*@-4mNhjl3Wbe9Mzd~{I0Sn0iDE}i0wZ_BP?HT(|
z@<^k_h4z6SU=-*i=#bkN`5b|=j=%Y-A+NehxN(x$P<#AkCMLMS;Hn^;j1En3Q-i^}
zpjo#Okm6q|+O#X5oC{n({{S=Ldj=iLRq{!ie8Fe%32=~s6W+U!;Dc?}SY3cZx9ss&
z!EhYzF{cebiHRa-p)4WwZSzT^Ik{u8!1g9k2i6a<yi8E^dlAX3V2<{?xhJO28oxy=
zo9i;FYaVA-1R7phMmMNes6Qj_yvn?;p9oMZvuu9RB~Re4NVh62sl_#uR3`|8B&nSd
zKVFEEIF$iWxd36rh*b;5U>Dn>y1TzIB1TDU?67H6b4lZ+8<YNY+y~b6!?#+m%=K-M
zCR1bm&FKLO;|H}aN3w&9M$_p@(DNKW03~dEEzn*6K%E}X0%u5o8YbGsds+MLokgE-
zyvya~e_5|LvRs0m+np<xc;>UoQ`5({{ECHeNU5QPAvnQ}^su*=v@Y1YSL~nbWtCu!
z6ZdpmsuBD<q5H=dd*Prvci674MD<r2x)V0>3>I$UV#*O<-mKOOltR!=G_o+W@h`Bg
z;D`vhN3wE(eCwn=!oaA3NBX*?OD~w&yyktgsfFpLVJZ!;k@l>i^aH3|{8Qo|ragLV
zw*64o0bm&xE}~s*HH(QrpmYd$)kZ@|^R8N<7-HV{gazn^VLBU`hX6FGs(D}5U_jU5
zo_~#=X`~#VC?l0x_7*!~;IpTw=H${DOfLX^)HdD(%)%O}l12(nik>r43&bO;zEcP`
zuR#pfeIPM#VMNXsRrW#P`%|Bs*mlsuSTPKuj!<rO=~pCGxiUTpjr&>y8_csk{xHLi
z@crCdfdWx_)4pLhGoz(XPP=j$@JOFj({CEh^4Ao!zLf6~$fe5E!spbG1X|1J!yiDf
zU%nsnO=6I>{=us>yED9uSlOQgAbyLr`7kNE=s^ZBTp@N{E|i>ZY^IBnKZ_35yKol8
z54y(>?)qIIVTbKNotW#MbeyZ*uxW4_a{*Fk*Zfar!_;F4yPNb3wa@0n#L99$yLKU5
z<MKLTcMwqcz%id!aL@Y6H?(dr_3P<_-n<acL5rKO#0%HP?+=PinRNQN2(nlsSM2)J
zt!utlU72qbPMnVyrx}51CvJbXsDo!d0fgYE%1)UtHO}Un?TWXG4TARD@$RBp^4^1^
zvSC3_aWQ63!ACKC@e^7g@#{7!Fr|}m`B32qbaZhAk*>Bthc-++43X?LtRpSy!`RW}
zMbpf7KS4|Aaqo}0AJGmifyg1qae`Za>ZaLknwyEuH|hB~WGKtV9*VJ_SyB7IitJ{v
zYal8;bgaR%R+fL3-E}>E_<T1(P~mz416yBgr)Z}L<6(;MIAvF)>?mHxzc%~+uDDK(
z{{G)*#y`IBR50X5&jic=-vah&tdBcM3Dj}&Ao`}eX@o}idH8<ST{NznwXH+ga9_N>
zvue)$h7ZN!l=D7#aS;mfm0BGhg~PhDX3NMf-3n9gXkL_D9M9%3i0kIu#@7rELPcHj
za7iH+m0BNkeJFnXIRz&0=z;F^6x~L_%ShA+=uo|@{K4apDjV~}Zv2!ZVLXnB!JsB=
z9W+pj34Ux2;f9ebz<h{V$Lytg)u;qwelmE4kTiy`Ma1FC1W;zN!F<+$x*;#7n}V3Z
zuHXns6H^|AOiN$WW0k7+(M+F!1S*4rN^qrc!>Cof3hPgkGs|P0S@+`2mrq@tZ*$`Q
zbarGiZkdJM`opR7MlpjFOY_p*e?+#0GuWzS-6f@UnpFNC)$^3B@XU_9*|KYgt|K)z
z$VQFIx<{;84vl`opXX#0T<G~#+u}>o=j8zRL4G*B6#(pwLIu8FLw%CNavSoa4CCBQ
zf_v1I#00{iT#P`R0r15Wa{EmoUejrJ9F!iG0nz_c6g!rvvI5X|xHQlgE)4JTIIe&W
z0Ui6uhtaKaDrI2hf61L`VsJ=R>Hj7tZ=}?^Jz?DT6f$*SXV#s2yJ~{B^DU@!`Bb2X
zdi**VU}ev~B=zdhMhCgMcieGEjj4pab@b?zU^ow&hI)h4TPYuDtaMFjw^EWT->_Kw
z><Qcn!3}aOH44kPSG?g`U;C6xATF70{~l=zUmpQ_9*dx434Qm_pe?I~?^qNyCY!k4
zo;#jpThT7zpcJ$X1W4MndiZCPxliz+wq(b2N&07c6Td#))`4`QP5yB6ai71wbAf6*
zA#82voo!Wv=wgC+#BlVs;Zc(yjk6_cr!%I9uk15*q(tu<J*YW2+s-W>+VwV}c~&BX
z=Ao`hGOwQn?~bbX9{th`+GwNoy><LHbU;#+Ej2B<pr^}&Ev!t0w>x7SE2f(a9U3i{
zb`uNUnN86q;vv{4I$dq#zSCR=It0nxM`(|Uq(Ag?_S-y)PV&9C-rXYnK>bOt#pXTh
zh@JCo8$gG%MK^`k#E;C@2Iy*r?8fYfwuvla!L-DiP|U~9DW=8P{eri^*@L~U*L+|h
z;TQkmSA+WREDm}h`3(U`G`7ME9Di-cPj5*Oy)L(!5Ss<U3ur@W(MDd`GcZ|E5}Uv`
zS|wbHoDZ1K+bQw>+1~{Vhq6g*SwYrE>on#fz~L@&V8l09Ut{?0v?J)lKm|6cs7_}r
z^;+&R$uo{hi)?ny;T~^QFMmsP)iwVcusd1>@*Pw94O8NsjSkNrrt~-0)((Mw^b+7o
zoRcCkrrPm9ZAoT)$LxhXb7`?L%NcWv1{646S)%2SpAbI`$rgE*jOG2$2HxD2wTy02
zhV>xyl@(s#UX_Bhsd~A0+4Cb<|B1uAvdCp*KI;p1LL|2=iGRZfG0Gu87(l0`1KfS6
z1I|k}Bi-=aGVuF&MOw(*5msVOvmPk4T9to>gK?$!LEHz>kEI;}&2#hUr}UjcV=oX7
zF7tAhh5Nu&7^ZIR$fG>Zx{pJHBR{V{?8`t{+S%55Gty;x1?Js?UD-(E_P5`3Ct=_a
zFrGutOWzG-jM$c84}qO3yg;nyZxZ+lp+jhu;`lcHa};#UK8kS$#J1ET%Qgcaz<2U!
z(u}Nw*23(G3n{SyE)<a_>HdR(g0iEx;~B-%k6<GOnDx6habP1<10XBSTc}fn-r4p&
zVO-<$<9v_4fI%vg&Dq{`s~$;V=W*8`&u06@urJSYg2qr9e-imy4ORPd9p9dM@;^3Q
zfT=w+SwtXt0b}?XZ!vosV66{a&8tZJ>b<^8RUW*d_QipECX0v9T$PQ$LsAv8g%7f3
zaQz-x%Sw__w`|1g-v={i0BOTSTxV14)1PpRn4nVqhwS@WgpE1J^9!c<47$W&Y_1~P
zmf><0ps=Nhsd||u4`$$^&yTQ+eCy`z^aM?vSoa&DBQy-u?bJJQ@pD#qlfYG1g;7wp
z$&e(Ck5sv{dPaK%3CA#)M;|t1REs+5|7tD$69-5(#SJ2p^gOun3+POykX-ehcqY-p
z1$+9!S&}Lid5jT##A2U}^CUx_HYS!OhWY)JX;5Z;H|a)#;jky4!g)y3NqKRoWL>_b
z*uI|ZvLk(>RSfp>%^B}#-=5_@$v0<~q{ZHd%-r5uMZ9S<py@KM9nMkU4ae>PO(n~g
zq`|h+2+~hwoO^15-H~r-o$dr-)5mvP4tL9%%?lEvpSxyEd%Q3(zO7xrkfh)`5!mr;
zI%CRV%8>T>lUJatY8ubwNZBLZBUe|}!AJUZzG55g0(5LV^NmiI=VMe0y}$2k@fs|D
z7)U-e=R^i7pC4u`MVj4SinnE6pa1CZLZQMop3AC~q}%?s`iE+`Da9ZLD>+27peo`N
zb7m=6>&$+dXOv!HqHUM4dj@Y=^}$A2?^x+lQv345;~LMbngUv3%Jyl;Isp5>KGTz-
z$8M1>TYtyo{Pf9zzAQUyL37sl$I-S&E<*>=D+ancSaQ!(BDZraDdt>`qMk%*-U%E*
z2u{(I;+Y>;<>e~v-4MJ_+Zbh!17ZWAd#>4>!DiVEjY4bb_Z})8u`4Difefe(ud!70
zMSS?am|@(WVd3N5Lq7MeWLsyE9-kd9Ybp`LniSmxrFLXEzir9uRm&hv120k4Otxl?
z<otzy%#i<nvlFU=y5$u93rKZ~Yq-spcrjIf?fx^(D=ZLs@xJgRNi$*XToo{SB%jI6
zeJm4uVmuvbH|=|o$hnxo>73EI+}FR0tLJWkb$7nQTqX7IPlsL@T2i>oWyUaFTmc4!
z=HMKUj5W0s+M&}CscAwkABn<0i`q}Mz`7LKt!s)of*K4j`>;uh5|4ux9d7`<?dBW&
zKPF9oy?N~c_=3U@bTqd917Y(ek_FHWcVNUDSo$r0ZBLNM+eFGy6|%h2YC13e(@<o>
z=j3VSAoYvL*&9UGZV)x@*%h{nSB>nEbyHF{qZRfL@g>RPorr7-0}LTfUF``*+pLrz
zK0tF4)b`K;b-)s6;KkJfr_HyC0|CFgHGz`HNO0sF4=#N8k1unr9Rav80f~UwA8_q?
zG>NZu4<JI90piXO;Iek)y#SIW#I`kS@-Te^;NkY>Bvu@dm|&TE5L=7Nz+<goM@mPE
z(j$z?=kUUj^Ki+-Qn&2Fczbt<8dYf0J`9W>yN6x>7qY<sNLTyewMqZU`f@nu3p~z8
zAi3h3M22<iP?NzL)bd>nePyt7cZdgfb!=<ZukJ&1nldra5*g;T32ZRLtOBU@nPykf
zdN-(epi!n*iG94kuhzXF-lBx$0P}w>)BjjV{}}B4^=8TyDCEhh?*7L)|Br|NKfWm`
z0c8XXDGQPR@p%4;{PxekpiBvF;ajF0kN>|{{9j)NEHc1|g@4+*_1hcuZxq=7V(+|p
z4oYJ6$Oq#8e0~4(n|y*p%)K+SZW+H_$NzZu{(SR+9B8lQ-nzo@>kIf2-JSp}kYdg*
zYhav#j+j|^D=h~2EP^I&=X3yUVGV-9^4s?L8@0D2%f^E3UY5cF8+@2jZdgkS;LbAt
zwj+E^6592WwDPTg?RsAg(hvrT@g$JYS9pByh=e>iYrp9H2wZrjht+lmQ(a47$SjF@
ze&o#y2rhxBy4Zddg&_=%u5Xr)o<WwxbmZxq!Y6W`z|46Gl!0Eb=o_NgMJ=c*e^D(R
z5ALfR@%wJ0`rVJV(9B)w2$U|3oDpC39rjWZ0!0j!1DgbUM>a$g2gJH{T)?Q{%6;HY
z8}czpOk}<cc;gO9ouR+}eSYZCv*%@Ype!8$w5H+trK<bCJJ0w0Au6(;HXQjE*fGw0
z&8UfFAACcbaEYS|&siCUB9Y~a^c}Py%)vZ0sdW*|M_VCY)m1SuQ_I<H{!Q2AmtQPc
zDjOMIoov>rQ$TH#6AfX%*>y|I0CHq8<OSi~L5Q<JkATQ)1B&6&X`wO4s4+lBEcFxw
z)5*Lds|_b*qCJ4iM?eJuK)*#$qYcmtYpbkE!Shm9O8IhQ5kOsm46^vw0$o{v-n&@{
zGPFGLV)P>5SMrl=La}Ug3mQ6r^5{cn-&0h;Vd&|KYD+kbXc9I6*v2Ce8Lv?@^7)+v
zYa*lXki7*IZ?EOByPN%XRlJ1u@P&|0*}v}QUU&nb?OH>mC)iy;e5|LIrp!D>YJoty
z1bn4SfoblM_n99JT%wmT;!`VtPGA+(P~0PK#Q^xfe|7=E2yRiPyQ9iCWQ5HQ{laDj
z$cEd%;Cud{!SS9ZbOP9b{1VVyA;Gz;?@Lsx@Lm969-*;FK<<Y~@YMZGdtcCW77NG^
zgJ53$QD8LUE98L_aHe=`z@Q$R58+%@7c8C5Lr5ykAq6itv0hCDuSQw?IFP5B&4&gA
zn8^I?*b?mgYQcP;VE;3p2h_c^cU{l?1RcvqfU9U<&3@+4^8*z`^WN{z7B>TiuKCu0
z)pKccOZC)m!chPfM77m$&jli#&}MVE<_p<|2b5m`KzHZ_7_m$*^f{h{dsiH_(3S0!
z*B^4VoO5B33GBb~n%6|$u6f}juv7ii$yJpFC+#TPM{IjDJe%DEZq1SOAVbCi9{AUq
z6WoO@7o3cw4P0j^-NA;R;`T3@ZVM%o>H}0}iYBT1U@3@|0k!U1u%8BHdk9~161SVU
zL-4T9*_I`qB?>m%`$beHP-6^pk0F*Nt-zyxIxdpCoU09??5ffYva-p)iFzl4!2UUz
zBk+LgugEm(1Kyx-`20Q-ggnmHw5JZjzK~la07&_(fe`D0^xo7BG9IHqn{S89&I9b8
zTW<}>^I!AStongjJKfUh3&dZQEoDfquz^FX^cH}P8I(wgj@zYmes>iRWO*|G-UFt;
zuEH`{QUw-k7I*LU^&G;>=Xa6Wh%5Oj!qpJwH@S4mw+ryR+4gtpk8L2!L2K~IB><)y
z3xG3>+YCX3_p&BV{a;!D&~%LT-sW7u>Zuh6m&T^_QBX}<gIntyBieoAM*>WuJxE<s
zPeA(9nD_A^wu`HLuo*64?IKRgYb@)(GXhEto0*rI%P$meN(zra40HKIwbvj}E`o{B
za>$WlPI->4$rns6wpBtp4dcKP_ma^Gj+5h5YkQ?oXd^m9XGq3XqWlm&$7seXItw7t
z3rRm=_kIGWou)DnkaQm9WI}DO#+U`tN`>0D_aQsag03Hmu)bgPTX2h|-9ZRqO5m9B
zKQ(Uq9HQU)0NhT0><QQSK=%;{I7YC7f_$jZUo-=8elpb(Okg0-z9?hsf27@nUY!VV
zmKhE>usqXL3WRI(<n_%+^B+gCDV@)MG04bS5Ct~wz&4<o1|&|}TsTi_T_B}R<<x#N
zM}e&@k%t6bRIng%FbPNUxr4!&GN7_`bI@qOrFaDJ>etI1haP7qJAW{ZKr?b)K8*7!
z0zH#2kE8-`buWPc@aK_(O(=|7E0}#Rz(b-7raVkWje%2b>(^B(Xr5xE8kjBpp%2>X
z55se9s<X`+x?|JFL42)x^d7Xh%YeVe*RbAWSZjjP!=w%dlj^7&A-e>C^Sn2UW4KD=
zU}-yyO#K1W7L37ibzK_1F9gmz(rQ4WWu|ROgZQ%emvqxZc;*nJ5KVup_O<|W+OM01
z*P3Zrmev91dIaKK??8jQWkAUBWq0fDInp!Gu57Cax_1Q331QWjNd#ArA(<iWAq=6a
z*{&rbIF+OtM8uI3WIMg$wNpKdv#s6Iz(2(Q*qmG*Ctu&P8ele5+E+M=J82}sLt?sY
zC?3FOf}W<tHoDI$h_6TXO)<SV0#QWEeZ_YdG{upNHxzXE6^j3)Y7ex^j|YpGeb99S
zC$xeKwo}Pq@-TMi#Gas$_eR>1@v}~UAJDDRa)IFuszF`SDx4&dW%Db;_@``W)&}aN
z3mVPv{A&Ejrm&O9wv<DZeXxP(S+oE;Zw;g_Ls~UtIa^zrSpBY$7{W5N5N4^{fSu~V
zxe2_4S4ts((BHb%eMlk04sw~+@>8yXwT}auKMoOD!vPOY1KYzqG4MPRCQT!&4pReJ
zyN(X9^ZYY;;B2|@{Jec182_>iWC8B!b%=I4r+$^p;2kjV;^PShM1)9Sm;E(4Io06)
zh>v}1?F<Ya?Cp@&=m<2SAPu`s({WPe^Z~+BxxUCh<$JKfK^n3%LJm|R2BKx!qBv%0
zIoZq6!J-JKhP-+ISpR2R(W_e9z}R@`zUW$u<D1PWLCMv1@X?p_iI8s5L6X`T^#Fx`
zeOBrY!>8t_b{Y2yXW(9|PmPiX`D|p|4MzsA%v6G#RQ3HR-}~_D@X1NyH09y%z*09H
zSO*^(^hfQJZdLV&?>lNAWPh||@IN_P4x}{vLzyG=9#_stx5!o&OhXF8Eo6FUSHGQ1
z{F+WNa^FmvdJ|xRpY^?V&15=m8#JGv(<u?6o2w2q?^LZmtl`XWbPM4<Wj_vPi%b$y
z(1RC@?^xTsRM#M#)X4n?TK1@SJfOkg$PTDN5z5UaWXA8W>obbV7u?p@)N@iFv>N+(
zJT{jhmRq!h?UJOY+0PNe#1x9jV!?Vc*i^_3G$c!<BDQHW+x`YDqMuPiEdy8j7l3CE
z*$b-D@@iZ-qO=Xey9ZX?E`fVKJ33mj^%n7Obb4m#@FOe!@!7(gWzQtnr-FIPfT$<S
zd}zlb2!Fgmh2$?Q0}a;^=emu=fHW6E3SjhKZ@ablV&7;7=s(CE!MF+yx?i$rHH(oV
zXxL)kx`?m9ZeP1gdo4dE^UAuV^Z9hq+pvy=YoPXxm56Ppt}fIj2zLfvI|7ojRvl?t
z<w<9WMwS>_e{An~0jhG@4|+f6o3g7l4A)T26x9<7D1)C(sH%%QAgv8!@MmOi?Vj7H
zQR}sh69Fbe-xRzKwES(X6AF1&^h(dZ+GJQf1}RUdJ*7R@Oj{WTnMCwtJGurg_=}4U
zyxK(`PbyrtlTtn&Tk$Ga5xmL$y9~TmrlJ0>`rjotH8IJknm&Ok^N)MFvr)Umo&;M&
z99aS%ipR80IlfXd4|{KT_0%vP<;ssbVb1N+;URV1!CwBrzFW>V&fTz6*Z<`hfD-4*
zs+3TarggEeJ(nGlyGA$E321*l1G|qNsX`7&1wvCT9_3iv<3Zq$BfIcur^@nLo|DQQ
z0@>R*vJyVjQ`2Rsf>&@e`!y*7+1?O%dao-dt~&Jam5Q#@)-@zl4q})kjyxJ~QdbOe
zcx%s<@t9k1`#fLZc<P?jIcjwmaH!vqIjzf$S%akR<M|ukBl<*jV_)q~xe&(fnRtF?
zu1CcLpZCk;!CFA{+33h0+wxPDK_`>BU4pT*%i7U#3wpvRPTgqj7^Tk7a(+%WLwa>f
z3CRLh8ib^;d?;H}Fz@Hq#-9vE2Jp4_0>(z22K0<5RvmrKV-zY3*&NnP244nKG*LZN
z7*tk5O_(lYt~x|2643Qkz(sR1t`4*YU;+*p3y5N^z>s?GyG;CN3Eh;7Wdke}5}zK5
zbre&K5WG<0rrbweqcRpHHpaIZGVk>WPlw=Jwc^}Q%&hnIKbGM6XdQtz#W$VT&3N#Z
zsH|CQ!EX^S$7mEkml0f01jgw&m%fU71D<@F8282QQJqJG@e|YHi)Tj;zLwWfntU9n
z+b$gzYYn|WoPK5Garrs!*CBN?#IGkN^a)Mp_53y8ntOTKskjiPMk)e!8Ivcz%+a-Z
zylfLB0HmYM6#q6P9&fcwBRJfCs&Z(1#FpWZ#US8**;tejzK8oPIo04flkp*oWO@^F
z6SViHHj1@PYZ`6NPuMk1OTbI5KH;${-H&SrJhlYmFeiAHcDie%U~8CO%F9Yd0fIr>
z&2jQkGF?K_Civ}`P*uu5kFYF*gDI4XJc_sN=Xs%@jlL4&INeo-6|Nf1XuvM{^3fZQ
z*cCL^tjQibQo(1}?~GBH4U#*+rYRkG4zGR8tf8%P;j!L$el>4aRd>(bO~@fUr*Rcz
zuA#dfr*eRs-gYs*wUKwbaj?}bF4%ZsJ;yJTY@C1mVFBElvG$t^3RNfnEvv_5C-emw
zW%@yg9!DkH;w=AaCB)dsZ<PO=DD+>Tz@I<vy*^PNp!x4RZCzRtVEgvmElO;Oj;dc1
zKjE86+KBUUA|FQ;6;YR#8(AE}Jqr!Hq;2hq5>6TUQ!L-$B5`m{Wr)KNRzF$M1lI@V
zAZTB)P`yTeW;`$^DvgDcRd9Y~axpC|Wk1c)Rh%8X{m>V+=7S6AaxoXL)GS9NRkpVr
zOb!c<BK<MvO(sQ>JIF8rIF1mvsx!F<bNb)pYy26DzqF)@Elpw=b!s!$7Z7WexTJGN
zMABu(cLaScv+lxXF?KRO&nHguqRV2HmcOvKa4oxXrQbugzGT@$aU%SvT1f-R|Kw5k
zY;>xE);YM4vQWmDn~ZnT$6HAZ;ChLb3LR59gZa71G3MeO=iO#Isk2oJY4&FaLQ=0`
z;okOV?IYb&@zC*|sc4ma8y2|wqAq}bhJuMlHrLhH`N%_ksW-?*Q{^;u<v@!y2P!S*
zqeOq)4;S`Pyjb}RvcF6uvH20rq4QH-IwKM9O;=gJ9M9O8*yPc8-L<l?ohTP_B$|nt
z#ILpGXEWV%M*38Jab7fEIMc8KBG9EGmR-nCX4&FCx8fj4n&G_OT|3Fk-MV2er7F2h
z;l|@)am^PLWv4*Z=@nxiUo-aE4(KaYAnba(-)AI+RI6`>YjpOb1`IXP+(Q0SHSUz(
zLS4R1CI)#eCh<SDNq<tQ*-FiZlyZk{p0l?Z*N6*tk)L<?uEwIc&O_m~?j@;B4d{aM
z9JOq$7)C)*bt%ezb5P39bQS7UZ2ujsmhuR~&h9<f{cD<2QUYn%OWz0^chf2`^W6d{
zD*G07G!Ih~ndp|Qa5;BVDd3GWAmEITjC{qC>md3u`U<~j<@az@IkK!ka<pTbdWQAb
zJV~pA0!d8Phc+(%8OXgard{%bP|_e`))~#6Rv5iISZE>>{bKtw+m;O7WU6L$AoXVf
z&g*RXU@nLC%1km$9}(*o*m;ikQL%qEBYb*j=7Om6WFis+e)b`2yv3{(waEj=#9~Oe
z6r<sj=VsWmu7DY^sKcaTL9_PH)Y9Z|-#^%DZtYI5y;V7Wo12_%K~7L=c>zd8PaY6S
zh2ZEuiqg4UyC>Ln+lA>0$410Uf1eULm6X?mE-u5X=C8IllGRH=QJ|+zvzjG$+$^J|
zEp17v*;3Q`pp{MoPk~_^f4HXl^-T|ICC6AIXh2l-i5f4`37R~mPdZ-jk!xuYhWcUw
z4u}EfYoMddWg^O6TuVy9+CsO0QI)CDuQ`qUlIT1><fxm+6z(xL_*el}0jygbch{uF
z4G1<~6Fv$bG*p$^#&Y)*uFlLaS-n6Ydhnqauzu8N+Lo*O>Cu3@-Gh`ud2gy<4QvZd
zy$J36fzljjJcD^@;gzAPZc5)3E`R}I(?J#oVu?-mvzKXSQm21YqcVNcAt5lC9H*)M
z9g)E7(rQFB!DtUJHZz2e1A<%kK8ktAxvN?IQ4EcE3J#f4I4EHWsjl(*3rYTXF}<N6
zwj|5t+i$S%ag2f}#z83Ia^O{1av5ravTI1;3$2Y86aAt$MYcxJH^Vm<$MIYqNbFj%
z3&lMVfR9;Hu-(Te&lfF`brXy)Iq(FG6-^L;XS%3xMK;Qz$6s%K+6Nave!6Z#$9Y_4
zdI)s-KAsXMB-PA!&p3PZ4vp?7R3bj<UXPXd@^*y#F)y6L2F+BM`NM@;ZUYfT2p%J(
zzSgPA?hG#AfO|h&KJ5IH{HwihTvxfnxDnA^y5OPmO*h31cx}W~DwGYc6_M{mSS<H`
z6qPj<&37R`4ECBa$`8(dw7=YXb#V&#X}qO3Ne5F>X0Jx(a%?t+y#d_$I@mpAud2Lw
zR$*BTubX1UZB4ch(Xtfx?bQ!OHJ?dUe~6uuHf>&ptQR+~O+m67alv6}`4u}?Hk9+)
z*hWVkUh~$?M7f9Y?F5+p2|6Bb4fG^~y0PL5=xteZtLD%v(*9bb)X91wX58%I;;$Hv
z=bX`QBchIWn<gk<Hj*`MuwG+QSFY8&Q18bYRK?2#?_i6Ly9|!{bd^m3fK&*H%iSw}
z(p%^{Xz8%x-!^!EeRYn|lo4yPVPd2$59w4^W~(}2H1G<$Rs~ka_C9)`B|sp?Q(tu{
zUq$2*{46&<74Vu@*hzTnjA1EV`^{!W{)~1g=k{1l{MGuEB+8G?t@08&VB%~eyUKdU
zbo70$w<CwF1fAi*6RBEPVLg*Pt7uE-y>RzQ0l2Nc8%>!rz+n~i@zt&3dRVoWoO&&f
zme+bl9UN+c*Owcrfwnr1K?t5hzb$#pt~e(axx<WNYp-Tm@#HSD&EYn4osaz;j{V<Y
z^{fSP;0?qOAXfdcy;$6HiEI<5E)Fya1i9kejf)_96qSVbs_iSp1wgk(Ot=OP4QB=h
zBFZb<V}}dnY0*bqBszf>&&ifKuQQ*p&kqQFoKJ6&>I!K5B>+&Y#=f&oXA9mQO{Q`g
z0ldy}yA>y(gBc#@o_h7_eL5NCpj$H14FQ2uEdX9JS?2A~D%tsz`jLnG#)6TTOobnL
zMS0natMhQ1ipc<@c+vdWn@B~cKf;Z*`@EEF;`b4~P0x;S@7-s62D6W=*e-7%x(Mie
zX`i^HDDk)RKRD_e4}xN$&m*5i5HoEfk1%<T;mKG?yc2vg8xXu@-v!FO(7O7iS`B0P
zuOm|ZbZo_$O2hf27Hhz<xD~tjb2_{n8Ym%5&dGRah^&1Jy+SB3PQvp9hPysC$hxc#
z_%7j|_ZSlQBs>SjPPCT>u2rJ!4V|ZQ0X{6?MPm2)p3t)sRX?v;e-%rh!R`Sepbz>$
zE7Yaj@SP-Qi1G_TE7qfY5C|)Wk;GMJj36lUWlvVuHbAdTph3exCS;a?Vvze)3GO#V
z-q)|H86->}o>O@Tv9*QADHKAejbJ;YSDofcp|VQN1I}672-j%slaa(3zN)l)N-kKS
zk}e<T{`#D~=!*HE&`Mv(!Wb#hq!fCPTH>zJGUri4ri!Pd?(d*@w_pI`)yiq?{_3nV
zP)Oo^2RPvNmx8y@cX|nl&iWh5!G+30Z-n8X%jlA8EcLDVjbPQgWntr(8cPIR9+Db>
zAa$n!fssasJ1UY5(){B=2tmf{meB}4u3|hNVG`clte3tZJRT+Ip0^?k3uJ4CSnr|<
zrh!*K<a0I#$vq(@>Y6=Rp&oZb#1ygD12E@sMK)LmUDrpYZHnCQ)T}dh>6zy?Qgboi
z_!OYWy(G)|<)K*BMMC`<w?`wp9*FNvXYyaUoH1oELZn^YBLFb^n_Pv4x<pTmNx@^b
zHJUU75?QpoU@WFi`j#Kp<PwF-IVPzZ)0x^)m6iL?e11ip30^QIy()-9eFBz`VKS5N
zt$-jZJUkfozDx9zIxnuZ`N{ESIbt;<>V8q0MX&Cf@dlZ}Pig#4;5`bIC{Z|2dM3$*
zP^IAV{-tRod0{#(e|ZXdO)pSWz3p@x><FQRagk#TVw3M2y2Yt(hUUn}`un&<%VmIJ
zt4V4snsty0EF$xm?b*b%&LgQ*(apyPnZ)bo)jh)R{K!4}et51YBgFQj>TBmU(O_I>
z5ttP&IQuH_pd6fCt9y5cR;fem!jGD@LTjhrzh0{b;|wMu5*euycjNe4sTJd5egNdA
z=+VM&W635t(Q(tiotQA-X7wYHA9YXeo=2nj#;G3sDX};nOLNGNz4;OZvvr8l{Xt-A
zz@y@5^#>OHao%0Jj<J<&_2w~7?qTcI5v&__IP(0=5cIg<hEOK{)H(8eMYd5HMcsvQ
z_rBDuWC$XjcH4msuHHobVOnQ5I#a{rV}Tau9c~mGbG*?*gwi)d2i!=Sly2y=a$87o
zul+C@WzscN&CcIBm+Grim)%MBkl!Gq2b5p+WyMZw(gY(p{>2*Y^@#f3EHJ*Y$+tHE
zM&wn1yhR4Q;u<R`2;aU@;3^)KSF#xq^EY3%=Zuj#%yQqycZ1?`V#4_AUR0f?x&)8;
z;#9mf=t6_WRO`W>@wZxIxo-{n1u8py>hAA?Z}Y*6lPZP2;>AkLmOtcTx$7rWFEhj8
z<OveG<x7(S{3yKvdPR|6tLqJgNi{D0qb^C3tMJmBES%Fv-P!2MSx)upE(^dM=T$=`
zn1STdrD9;m8;p<p->Xn4#n-M1c%VIWzb8~i<S$>^6LK)oKK^I`vqH(5q34AN83#A^
zrj2m)uPRSm^R6iSNJlVkz&EFvqggAkqV+0r=-q=ptP-K1<%OG|VhzuGAe<40^9M)2
zb+O{KE=b0%?4N;4?0%2pN40%D(k&UcaeZ6mv%^2`9(xrKpxNjJJ3KzKS88$RJ={!s
zfHk1x)SFFm)tX*}vZBGS_<>Cfiue^Jb6Dn%?2i7WeQx70@#%4LepDiMzpnS1$uFI!
ze}Xm<zGq>pT^4`K%=%&^Q`iZHg-3bS%g>)3{Yeh5dS}Fi1Zs<OMp#;4=}h$jX?hW5
zi{#`41wowqlg=aHp(gmTS%%7dxw2sM^<@&-37>+~Mtb6)m+9SA2U<q~c58q;xXfhO
zVGv*DUS!!L_RzPS9>cg(y_1TN^ZFB)DaPO)eW>*_12DWL0gW?Gbxk27dU%zp1&P~I
zkbXCX#`5iyfux};OHT6VeCqXL<7!O5bjrY8-RF7`(v5j%Z>x~m!?B%T6nqKv&EF^t
zZGgZ8X7O$Zer8un_b>1l<e{*C@TF&?2iRp}qD?~9R}*Rco;OB7;vf_LgK&MA;(Cx*
zSV|}_`J}D(`~#Sly1S$Ci26&++M04nocrgqYdn6P>KfWCkW5Jx{?y#w=FlA7tM%K-
zhg>A4oI^lBmw$&Z{`rG_$M7P|9eKy)afDU-kRhjxe<%DzL9cv}S9e1AM1HL31G|vQ
zkFw1rp|}U+yoPID+iyG4IlqBG)V~3g7&yAP!zzW!5gY46cjiA-=r|oq4tBdVC%eBt
zCuM%sEN|*k8(-2ws$H?X#Jh>aGxl$Au|1;r$Ib0!%p~vpW}wM{N`-}U`(P?`a;({^
zznBtZc;xxi`5SYXFw@=_efOm=%C~g3b?N_-T>IyhaRs9Ju-u#ZrNee8>GRB+>K14O
zFL0x~5GR>Vv`jOM6Rkxwm&#!xaPDNjh^aGFd`O=e3Uo{bDwUF~2C$8djvQemV%8>{
zS^3~!(&(!`PjXrv$7!9k735@(VF|MZLS>lyTGoZFeEy1%Vp3a^$avgWaE)3mW|2;S
z`m0yePz>MdI-ZncC3RYFKL(9FAVXjR>#jhnT-JRaWECd9H;ZrfcvU2OIcPwPDjs<0
zZLQipiwJXPug9g4MSdYB<3xx)-_<M)Pq|`XP=?fk#zFMLgw|i-4jET<A7;L6B2)HS
zOK{gBX`iMc&Ktk;Ia>+N6YP@rTG-w=F`M4{T(zGNx&$GAv6j*`wjM8YHsI%rQ(o|^
zdRMO;p0JIB(TawcV?oygi-ajF4ffWrF1BHlPR3!hW`xIIu#@@q7|)lW6Nd$7QoSe1
zNs%_a@vKe_8=&><pUZF2&8d-E<dxJDD=y_d4PR4pHvWE4dsw-IQBAL1I~}O^Npt;!
z%!quVuo44D7PzIDYQQ0$x&-$ele-=gYNQ5rX<%oE+Fost?yrC}xT#EpOQufTT5SbV
z>uFPJyJt;YXz2{i?jOYMOyC*xsl`A%m7wQ|7pl6dqtwmDgN4PeDJyulcmb7~2#Clq
z1eA=B4%bZRjYZm?c^KcF7G`<?S4-9d4JfB+%BgpN2xJaA5iYO?UGq_$K6)nUa#BYj
zH2i#UMLLrpoL-Ma9T>nt1IqlS-0%W~YaO*7_*<@zd^^G>3X{c8dg7czYsZS2^v?z7
ze6}y|Jo-P(y>}qheftL<DMzYvw2sK(P&8zQ?5rdqm5gkWJu?p?Gh0eVs1${)tYc+M
z_DHtutYbz9zw6Ue_j7;0-{<N6{r>wsf2p3^8K3vKuGjS%p4f{ubLX^&8Nz-Lgv+Vo
ziz(9zh6A`iBTn~x8F3=w-AD@k;{{wD;l_e!gkxeyC|$`Dn^I$|H|{E#FLf43HA_q4
z9nWTIN-c9*FD``aHb5U{7;F6OWifznBj{_RE(y}?1lPRyqjQcpp7Z0~*{hy6*xY{A
zUH_4Va={7{6?kPcQ1Rafq{QVisr4QpOB<Cwv*O6+#Y%qqRONGgC6RFxweWsTzkqn5
z?EE;Mb~=5j+RcvBb-dz$9gR2fOEY3I6T=x9!+9Y#@M?n$-8l0;JBfs~Vdwr$+{Kpk
z_ggN)3snajt9Oq}?zE%va(=C2ryRbf|9b6yHnE_rPdJwNI+Nt?(AOJGe1f;L;vUzR
zS_+F^8I4ei2{dZJ-fXPa&V!PvCq$TJ%<+2K_C59P+2wYn3}KR_+){P7hI*SETEK)u
z?cA#NK|iM&WS9#M+b)SeI$C`&%B&nOw1H^u&ZS3yjJLCfa-#ia5eiVUXj&is=72X6
zG4o?^;9A8TSt}A?-53{mA02R`o6X)D?Kn<7XYX8?FKnm$nyC7!aOBYh7(1s3M>-f&
zV%@4lImD=raI|n?8>M6Q58?_xi)h*wzG_p_aY75vhq>!fcRePAQVZ7@UEe5t)}U!_
z!XeFv)%o<eyn2_z)SKiOzxp5--QEtLjY)xyp=q7%LSosu#m2z~yUBx^6EDKPVsWNz
zV9%b@V>nV9nvt6uc(zd{a9scCSJ}xIVC??DBAKTklYjldw`V#o#?4=4yQm#5#}>9!
z_3=jPC38ew%7-%K>qKsYYh_7mbl_uCzn^`uYsxHobYBfi%^vJcy4eUNlKMwYX*1Wh
z+M~SJD@cgz6^Bz6h^G0|4P2MPV#WvhhIz~%XFbZIpufY8sCgPivy;q+-$ksv2J@Ix
zQd)*vMI-SC5*c~73I(nbfTq*vQ)u|4lVSSy{vt5gZrmlpt@QAe!HH(}Cc6gu1}YB!
z6oP|_Fvmq6EJrU)!01+n^!e>LA8@Y_%g)wtu7(+dd?J)cIq)^3`cBcMPn!OwIh;#3
z=<>5<#`UqSvDA6`4Ylb351gM6TS~NI{QN?>?-TdakGYk}G;$~N_!XTOUd#|eo2T>(
zHk2^LP?zX)+|j<n;SlP<9kMh~CcK;d4gEKp=D>v}<!{0=`qxypb+43Nlnd$ylNlxK
z#+Bx@srG8e*8M?JK`J?GKzP#M@*+1-Gzc=#)Zg-W6&Ps1vCMe)Zk7Tzi*7UM(fI6g
zNEByA;NHevLu*pGvx@4Ange=zwS=%h3j=M(x#)$`95~Zz2ox*L&f5~Q{lWvFUZR_m
zTIQ~C$CCHlyk(&R7O-B9H0unTK8}`)Cri;CmMwqeM=p~C7llBUl6~kG$EozVHK<Bs
z{d$6Bv%mUhvL=Xhyd|za{DHSa3-XoA33*<{l=p$vnlA2ILEfc!7^Fs}XVuVUWvU-~
z_(}wmmw7&b$r=jLtn)$E<TS@5$yZMv)IWcuFHryN;w8xo-m3mbz8mOuSGGj^1qO+G
z-wc+<qoUK{3{!iiXlT0p;EM^+3c}ewGld9k|5%a|2_v1-u6%Kx8RH#(>T%ZSF?KbP
z<R9$XEM*3IT_ta;dAl7r{YscAZhOY>GOc2Ndz|8gzx>AohP9=}2c?g91g=P3sbv<^
z-&eImru>S!_AE{+$Me}v8*1{SAH$;xKc`~$Iip%JFAFHMI6~#}f-kMW!SI?&J=20h
z6uhpnZ5OMh`zH<7ZjA1_Lm@J`%E$hL#l;$Y*m@;h+bl(evSx3frqGPi9{zKfOAk3~
zbv;r0XbxBe>J}(Z<s?ij$&C8Uf0ml|Jx{f8!0k5Iap~-!X#a~hf!SHeY(V|=&fYHL
z#NfRfSEzcwwbZvhY0DZ`lPV=W=aEjCq%*%8Ymp~cl-umkn|K+S_vBt4<);nor`8SH
zXb|>iy(RZ1A)nU|f8#wuZx7`s9GM-}8#G|Ia?mQ8c!&vGJDBx^a5_+sLRsLO0q*ve
zRGQ8)oyF!Wvq*{w?c9za<z+kRTd4O(=V;b>j~qH^*9k6Z#w3YzHS#s$HB=vm2L|+i
zIHKR^wHpo#E>52dRUGH?o+25(>vY68v8-&tv=U3sRdek7)q7K(U&0-evu{)v$d|qm
z?Qs)eMo*O&NDRyF_Fr%j*0U4nvzGE#c0RHh?ueV=Eg>+!9xVx4^IBDMpYe=45<wWu
z1%AsBv}1QI$vEQTo9mIq?-$&?U!y(6^$n(~i>iQ<OFKscCK@Rvvg3mbU;T=yxqTzP
zEAO~J{|JNP$>$*@cYJkYiu;`WN9;)fAxb_tE&iL1WI*o!QjPY+P&u`yx)@$tJ}`3f
zJafWDW^!p-)ZB-NVj474&q!!-z4y*jI%4YQjj|qXTqmj>8h00?xpwZdGn9*Ti!&sm
zW<q@2`}|h^wZlF-9rR?9mkmE3=a`H!VsGcGUVMiM?%I=KE7Ke_8Lud=9+q_WmRt_?
zY~M-ziwhKz@$*9(J>mRX0MFerjlJl+I(pv0r|kG+mTqA~V<Xkcqw|;Y<l=zzUn}Xx
zvhi^>;}QB>lQ^2SOYyS5&Z+B(9g3ntxfUNK<KD2{bGmq$ze4Pr=dKt2H?N&m65bR^
z5I3wK95B>#om7x8*&?c#^QQBZ^3&~XThrgZ&#+Q)=W>pOjYx&!;%D8AQI*Jlkz?X`
z9)0pLFkx35xso{jF$7ka@``@!qcN>cMZ?dz1a?k}s!A)i2SWIy<j}VF*Tj=U1+o*D
zMbhs)H7Ks$uNPBurN>)j(LlG5bGzFF*EiU7?54SVjegDT#YCkfhom1tn~iG6W1Bvw
zwmLZGg)f@kX}JA#D2@!j213Q7;-r%y*x-+!CKp_V=LbGG)joZqW-G?v$Cg%qO)MdM
zX;RI`U7>a<AujVcqq$Hw-yBWXj&GOenwMiA1&k2rV$U~!lW8WJ^eL@6u8M`FxBq!P
z{2$_M2-Ud*<PJ|4cGP=tJx6AIPA|)EMLqnrsBWC1IMn!DYKK+2_fwf2e!-)at^4CN
zQds{4>@$(8AmqqB0UxD)<6F-TkV~nfK9Gw&S8EoY6N!c5+56pzWwrsMzuZ(LFLYlx
zN><TI0vaz3-P2WHBkjHQq|g7r<ree<kjzf;R%+6(cf`d+g;^xECv4?3BCJrQ1l$k`
z;5r&ZEntMr055Xt0FT{;n=h}vW=n8wa{xy(6vM|;l9MzC!nj_8UrV0NNKN;La29Gu
z6V;#ZURROD8dw-zb!-X8@Lh-tYcI@Kuy*(%quLM#GenU3)K~gjBMlfM<RMjX9%L-`
zyjD)My_0k5Ppk%b!{2}tm^SfT^f$-IQ;gLkxfqyqpMlOwr#I&$GECV9bP!8fMz@?s
zAh>6VM{PL+;*UfDQ|nQVCB=h_b3UT6ns>d_e@&F~ddZEOKYyF2DU}HDpB-wB-swV^
z^Y!N&(Le>b({MERU{BNtYJO`Toc4R60pdD2=r>-l`}+TkcKKh^;b)A{^L2&Je>48;
zy>2Ug0j?aim9q>EUMTcz@q9&frI*n*yqh-%9U8z4hNjI5TtB^%_Md+Jb2(j@J5a{U
zv9#-chQ13J1uzVd1=i;NrcV7~MajU-@W=yWfC>NCyZV3k$K5n=-%~0Zym|eH*Z9*v
zT|Exm%W(gH@CW*@Z%!8jatvz>Uu3}lF_8Y#tNx8e|Nql3U_Gb|aWztp?1rA9?@n`v
zm#-LE_oo&sgHL}YBK)^U^bB!B|0a_2m%B0;tzC)bZ&VZ<dxSYMmdaz&cATEyTx}~X
z59Z0ccX)!yVkgiSEePwA$&?pNZg@MOev|MEZy|6eV!l2V@@-7<Uf1LCgi>=oQTcyb
zFwZW*?Z`|s8~D4&+iOY=U3U(c^nHYI?BLh6uov$riGF(otoq&~YPy~hTKR&x^GafC
zFTtEw3TdmF-k^N%me71<rf>xk@_J5ucIuS1QWRIYJ6L>v<|*DF70m+1g`6v&LhSQB
z)d;^#tQ{Rq<R>;N)6Z?^bewZ@QU2ucDcR?9<$qm9|N17M`61E?JeSsAj47_npBCHv
zkcYDYL(hY|HtUq!^?#GNK_oSkWTsa5cjsz)?Zks#Ps+e$)heTOJw+6XPB#szZWa3i
z(ET8z$j_!n)AM>v^9J2Fy5A(nYY2Cmj{r?3H`Hy$zybJ5N8>3{FKz0n#@tIWCQdDW
z-B*;7e)r>FbKL*(%u^5w^S3LH6Mq*hWjmO`GvY^C>GCU1In;TNo|nlo-C%V!1BoD6
z$dQ+vKeaA#0bz$5pd1mZ?5c^0p=ycWhT9w=VHSp6XQ|1+6kRW88%!815mTlUyQVpd
zK4P_(l;STATCMtOeRDgzEdBe;`j2oH8I5rqti`wQ+8+P49lgRlyum!Ub&pJ>Zt4!{
zB4GP#soc*x7vaf7!G2bak+I`Pdu<tYXapt`{`If_>_7hSzkL4@Q+$M3EHLP5=dJ99
zc%`q*bb|ow5-`mlA*M<QY)bch5$?gFYwWPOkcLM)dOk_oUH)wDBZT9Rth<ZaYxFuU
zlP6+DoNb+!G(h80k=TyD_g#wcW1-dpCcL@Ov9v(hZ3r0X^bA5Oguu97#Jm>))3mfF
zp4ursdR$BlZ3$Z)y!XBzZnYI4VAC5O<<|G6nX&;(@(_`wG2^`cbd1;49IVz|<wt8^
ziJd8f)#`Z7GLQoK@dmMQrxb}qT32;!BPR8;eZT>=1n2KsgM=l-JIq@`0ee+-s_&Kw
z^hXz+Mt}01UF?7Joy~RsWvjh*53=+Mj>~-7YQ8_pzK1&S<q4`wSHl`Om@<!_kZV%E
z%7dc5w4agu;R*k$6LP*XYR4(bXuQP*3p`Ml4|ywH&D@%B%o(4K?&Vz_h}L!>Ep{%k
z_iMrbzNKDQN!%Q;+cHAziP}HIFymfm7Jz&kGl(h!{9zMTax>f-OA&buA{q3eM02?1
z8Pf;#BcDWIz$p<Kpbd4SZ&mK3)lUOY;w)K);%V^c=KF|DromeiY^p^+hG&)UrblAF
zRb4pmgk-pnu1**v+vW7b4|X#htO549c1?75p^&KR8lQL%3%(nu))wIWX47<9R9^~C
zl-qrkNvsfxm}%Ad08!J0;ib&qyo$_5Xd9VxH8*m#zBx9Un=1)_t|oq}D-<^EcpcTA
z<7pqirg;U?;LMy5M6TOBC`iWZBWp|dkVTI`lwCjxs0<v2AF$O3Wwl6Be|t@cR5yB%
z`{J=%j+ttI!S)UAX2OM;J2gOhIA+HAo+?TJG56MIG(r!8nZvLn`LqDx@;jjN3B^sQ
z^c;4h*V!~4T?f`_HzJlnrb7k{Pr<Lm;%m$Eyox<%s>D9bg+f==6?M<>P23<~@VB73
z&BQFx!e90&|78!pMh*v2wAL?PP?^UGkMHJyEM)@wvCIfGix`J1e4lQtfGhbY%&sP`
z%Dwn(i57>slGT;36Hv<OLm?)QqM{jV5a0e$BRl@n``LG74e`^h5ZzCA?qs}dmme)9
z*>3V+IMRfqEm4Bszcf%zL1HL#g%on_;M<0F@=M+a;vnrTfGEw99+F+E6dZ^y!vZMW
zIY83><b-CvNFj=nL5%xjAvMqfnYx1=Y=&wpCU_<TB8!?qc6EsPoHt?FFe%z7AP;XX
zjS_JEZY>OftprpsM$sGw>nnhE@_#QghdFgF;xwI3b8S96o_km{;tJ0LaQzhzSe5Y%
z*tUb0?E)IR9(>Gr{7cViI1~>Z+XgY#F7b6oNZ)=^8okBLB=g$qDoXj($PZ-B4dw}g
z$WZoxLD5Q&*{Tz62yYN-p(~GA{cy=h`x}Sa`0cEPXD41EfIahEVbpFLCbFRp2GoXB
zzKHi-_uCKj37aSRQ`!Z4<iu@-g>oNgDzV9S?&v%{PEBQV+~I+i0(bSHv#Da=zj)08
zPtkwvJIoo+H>r<clCl-AQcg3x7=lxrf7~*w;`D4v9N)XHNQdTBHAiOw1CQ4zZm!`v
zIKR@<<>2f{#_v9kxO-<Q;yY*Lr&2Lt-{QA2aYH5hl+~jPqTI|tTYpyO&TN-T{;YWB
zTmAS|!yt{vdYndc%*TZZhY6G$OfOtlyx5tyKV`Td-q2Imid#->ch-KT72NE8h{z+p
zwy$RHfcvcl;n2ZDIiZ8HGok*=P2$_GW_=ODYg*UMr37Xet5?4MaS|{6%-dr-q2k|;
zDLbEf1fFX{KxfU{j}^t%Mn~@lECcsQkciD4jdy(*DB6ZlwMVsQ9#Y?SS?;XH)K%Z8
z^SK93O%9hyHfQ4P$Zop8^tv6&r$|T`m>iY#7>#!@Z7#yF4vcBfJ`7xWVTs+riXC|g
zj{7Hx&$x(c?BL7KOzdKr5}C=YWFFkbsvLT+Bg7ROasvBGhTiC@P0eWhY8x`&))3Zo
z4bv>6-LzJ|WZNu|>yRU4Q?9+3aG}oU;@f~vBnGfqt&`br0<x`?lELb*R<?qI9o!1M
zjprr_gB#lGfDM~9+eqpdCsUgb$8q`HQ?yP(lXBUKC?2A0`Y3afYf;+>xV438{&>=7
zpk~V%&KIj|uItugl6(UD?F9Df<I*QKj!M3DyFBWAcGc}-q9T0YA7V}&jo%v)NOKKW
zuP?)~7~oi#Up_kAnCS)4Bdk;99B*oO-BR`o^hIP0YOW%!?&iLE6+P8nHcj>;;;VPn
zSIrmnT~}H1U%|385jU1i<`3U!n{4&D$35VD3=P$oShMnYVBp#dk!f$jYk80rI@*?-
z=*RO6Dhf@$9A)m!fl6%E*i^;g6Dg)UQNrT(jV#il#tN#YtmMHfXj+J}@==M-L2C7(
zLQ}7ChB1bjq<1%0A~Vef%IT_wX36J!gqMm0vy=JK^Mmd44OIK{kh7DXm}A!{_H}?i
zA}}i=uwGxM64ynEn8uA(w9;~p>#I-YFvc}4rOcSZn}1jGeK^n{W6Uvr`?*9A<;F?7
zD7VE&*A*WOo^8OdmOf=nWq?H)ab_}dlO<nP_K#(Yr!DyppZCA}bpNs)Y##p2qNG=#
z8f2{!3+}hEl3y62eXbjgX^{d6WuvJ12?}|CmVmmFFFreTL7py0I?N$$Q@ufNk)v!o
z=U69Ne1d}KxVIy3vxuFom9;%HCxx(U+6snPeR}vu?BEZhz7Yd~7hV^KWJ!L6C)DBe
zFNy}P-2+-6E6veDwj#`7zEVaG|0WDCv!xW1Y9xmH2Y5GKA>p|Z_r$uA6=`m<E|%BQ
zWux(r*4Y)=39?qs{zrz2KWh_C)MRikC=ZtGKC$|=!|6<V|2w^$cB!=_guWtsbbVC&
zn3Qn=d~*V`fqu3j4opq9%?9@0bk87Fe_!qH*b_CCQdb0u`gz{pT5p{^_Vxv1T#vMT
z3q#*b^RY2L9FEt#rsUCy)hp3z*nm1wPG8-q+!U3ms&pYO-dnk;8LKH8j>u&28eTVx
z0d=O<2gCJ#C!TBi6I2Ls?}WrwKBw2LZb|v2MolnyE$vaurwL$5x44qlixTMGiTHEW
z^xA02h^ItN#Pz5MEPIK~-!er+CFOP_a|b-$B``Yn*>L;O0h47jeX?^iv17;MKLP_f
zy3p^sds6CXVSiqhQCujdbjjjzy&tCc-x-Nko@oI2Tg%7_2v(kmXj=LAx%9E$X4+Gk
zs+%YvMsZ4>gLoS)vHyjX$n;?!AJ?jj^PaMWHQtz=&DWinOdn&wAZ{K;$tEi7q;Il&
zZ4tvNCp71Uw-5b#_IVocTDgd9fXwy;dITH4o$V;G?zdV5^B=LH5>rjqs;gkTc*|)S
znKND<^m9z+=|pYMyL{EVaB!eD=0wdRT$>@>MW8La9Ed-;28C?Av68yvRsSy(z2=cU
zht2Kfp^ivY`IdMv)|j7PWpwpU$YLowMhIduraB|_Z(R&I77Z5v+4?s_mmFpT0^ku5
z&Lss0rF3e6>T%~i*Ew;O#wGqMZ)#cC;+^zovH)hVb24Tp4IQc(nTE?=bv-sQ#T_;#
zcp5s$i-$m<d>4!8vVEw{X`?TA4EoJ7F)C$UtNrbUY*vQQzrNuArEdT0kC#j^m>&kk
zf7@dxzA*YV%78>;(j1yS%Wi7y@o2&l_v$6pGZXq8^de4~h$r33yM)cONF9ga_%U;+
zipKF)BW}b7WUhvL(WFTm_HOv=bN9wd#TaTzv$Y?~*4)p5L)vGieKz4BTE>h+<Q>#v
zeVWH^I%YUY97SI@2QR%~=S#)IZ*dcdfq!wW=%E;b!i>%PC~hb$ypsJ^g87fk@x>R(
zLK5!1@pz|(U9!RXBkNE(QYBg2lRYCkF&7|4rQ-vSE+@P>eXt2aHo393v?Jp~Azsmb
zi+rnX#w5LFk=yT(n_YArv8agl+XHdSw;E|HZ%zVpFBi;J%S4J(iOE$s8LhP22DKr4
z=%v09?WTFmqlF((^gMbu71G7ZKZHX~FA|ErM!Me>l1-hGW#ez9pH?Ix4>}$sDMB-g
zpq1fS#c8aYqNN!4#|Fsd=OVh(wN$c<_$_XTTi(beW$!Pm^BGVW9Oyk};TrbM5^0#h
zO{jF?=47tlRxT2W(#P`5S?zlp&t~IHhF0+0W^#)2fR3W!*xK7KlGaO@`SE+km<>O_
zn0#RLwQF`+%;?RqaP-YBIH_#D){l48SM6^@ulAU6bcNnCEMD4Cpa#ap*;%#NdX;ZQ
zNw-!*`|dIPiN=0`VhaquFR0)D)ynR*CP%!U(F=6U#fS!duw)G&$U}^)Gzw6TbCigQ
zBcW&qkI39pp}L0AXWHQk+m2vCDOt9UzpG57%DO&#lg5k5+}g@xJpQ>xXPv_8SL`vI
z%U67?Y08_J$UClJv&2x$)e8;A9pq|5amI>W9ok0`5yWXBePc)uHxeUOn9s6wwRL=;
zOY)M2>nxxl(9h(l=U{0rrAoj58CZ1E4vyk!<-4wX$ckRUwbXigqcvS&nYL>W?>eun
zv&CU@{FI3*O07DkZbMi><!#Xc@~u0^;qbdYCNkHN=d=ZRFgc(p7(gV!1V~|R6m9Zp
z3EJFAwg!SXQM+bOQSVXM+yTzOGR(ZSUe4%pMSOGF0eOIYX)0UklQo-(XHU9Z`V%n2
ziX4B6WnQW2|LO%?xltlOD8*}5O7+7~q-#!b<dvoGQKL-2L)%3}%rcrGyQgnxY_5Wu
zGwt!)%VGC1m6bg{y8E6>yz|ERIXneF(;@p5b-~3A7M_ZwCL|vTGo73A&dDl7aQrLY
z`@`zO$s7<P`Wm-LJ&5t+$ps=$wXW*&$Ef<yRF)SqHAgdaoG3@vA=W3y3oX`VybV(Q
z132oX4aAITzRCZCm&59bA)CxA=gy1yc=Wf0Doz+b-uz(DtrPp9IXKYZ-uTJ63`r(w
zQLu1NRpWKOm;5HhURAPeJE-dWz`dGk_-<?F-Rm^UOUuQ*b)B{$`sbeMjFl{d{Uf9o
zNFgZvYt=GS%tX2~6})6bx<!VGQtca?MZT*J?OUy?`%nJ`r2P#y{0ooJp(Q6(%lrk<
z=<WcbeSV!(dlCf0mCkAx>|j0%hEbD9WLLeANU_;snB64y{dXBHK`i0K>P59DwbdZa
zh!Qzxp;hXaA-^Sc1|n(-T{-#=>d>7tzPb*GQyUxwU!2i`N9})rUra-{LMUm0<735}
z{JW6-mL$g-eViy;F%YogvJJP3f?Xq@s%^Qx?c9S&fHy94q{-vr&pUtUA01e8d7BtP
zj>xqO7i#EZZw_hy(BDhj40PI0k1O{lLAKi{a~Z{)VjHN{H*-LFl9NV0&T-O2!wBIH
z<31KlZW;*`{#@ND>#YwI&KYRRxFb)10IUIhMS}5{qzv)(uZae4O*P7_pa<^go{Z^E
z(Q+$M{Zv>4A}KcXx1Qx(j@{IeZ^w;CDs|RAsy-y3EE+6bpa!J9j^5XobLF1Z^3S5+
z`X`Y&&g7a!=+Cslh^FLWU<sX7buRVCh(NLkKU+m$`pHJ8tSiGKu1beD1?SD66UXkE
z@<Ps|zp4Swy*DoYX*-l&qCj0w{&yaQ2NhF?B!*`;r#%S>l_iEsn3o7~@Qv~X2)p9G
zd4_BN2+Rp!MAAJvurtDMcN#%S<^Ch~sxJ)4aL2a!mi^rGeU!(nFZg#L`_h}~d42~Y
zw?C9>QoeiTZ~>ep(9WJsbq4`JJG7?PG<K846ME!v4>_fMh;cayD^4|M`3rx00h}Dd
z3@|0ps+xeHUl9y2T;=fz1lGL^`xw8|x5M=R!0d8MeQ^F%x;S*yqw&ul`q8D=H)B;$
zWRB1-EJM?lYpAohK^NNJ5_)>v`0(b}4pzGEoA&()f**xbF(_pvHiu7>zmKDa*5^`w
zkxii6`j%k=?JwGMrQi^Gc0B(<|NQzl?;ZF`R~gsUt3|AysJb^vwbaZ#xG8B{^&9{Y
zD}S;jyJ~~OKTm&CRd_WdL(K%VjY_1WT;(O|>wwX=L8~3!w^LUuJjMO<-RUjm&+1x#
z_V_tYMP^9xYnO46VSr+eZDW7e@NgUsox2}TC3iTlNWJ!%5Gm$u1Y1iJ00b^Ot(;*M
zV_I5p0vuZj_mSWEYk!Z0^cAQPWi>$1-F~k9N<@0`;>W_!R>Vw11K`+7sZKx)l?fpS
zMaW$WhL_<W>R5b(F0VyyT_-DUe3hC(Fkc0IxL=cktCJXFoFbAB5LXgnDd3Fi8O4jX
z1GiY+^<0zaUA^H%#|6d#Yo|fLPj5X5EqldXJ_S%PdQRBsd@<O59CWUxkgp;`5<#7~
z*as@T4m39lO{d|bilMA~=y#3%h#Xx0!4ovv=>SDe0&Oa_Naxcoc356*-n!YXcFg4R
zB346h!?=?bf*){8UGm>y1!nznsM7&dR&vW|y-7sWc^5cwN38tz@$B!~eg9Guz0BSm
zFW2G&`U!{_bZY7Y)}ICnyy3gxcJF<s`e`!_ryM}ql5}FOPKDqn9FRX;{Il4f&0#Eq
zHRDod!IIc6__0`KzTs&NM}V2U<$jQ1-`mbSIupp3%x<r*8kZJrXfbJ}{dk&V;GlYF
z6O6|+z_QYBZzvyj&VAmJUv|M^&&05JRUd2%EiK%ty1>T+`X<s{;NELJqg0)H{|+Vt
zF>a}oN_`?4FO+-cG{fKq+5+l{>q=}7?Bf=U`bz<7bwMqv+tIXUzn_7x=2Zste^9Y1
z&Zf|>ms|M*Z`POA8m5j4;HKlmU{|~mf)f=g=HZ>2zpHU2ZY#ac-oA1xy!XXlAo=gn
z#Q#``7wpKr3dXi&|K9(xev!l$LHuQd)Q4#5dUJ5L;77}wUBANZvnEThQ@T5jBQ^Nk
z9Vb6(7ar?;ey9*vbg41~-oz4D)_4Wl&gKcxm@9rYbaA|5FCgia4Khy3oamxwEr?Ue
z5?8*>J)qn1Vk5f1Mu^Uy6a1yeUr5@Kz0iUM3c~5{EEKWEFCH}bxQ{>I)hHu6j$3?D
z?AIH_0~M<;F_V2=`ktn)G47hcRQK%&1#?6pQSL=q%Uiu&KzULaIL?9Uxur#48%1WU
zXK@zR&l8-DD>ZOz9OJy$vM#XyF@^PPnc2`>sN1BOX2h|d3o92zxOxs-@hRSK7HkTR
ziD(uane7LnZKDsra*dBw`GAbQa!s{>^6qhf&2PMoKdq=KeY8)L>ENa>U&Ge~>Yc&-
z1!0x$>+K0WFbTEF$>-J+CPlYu_8B9=g_W$DqqG)|R=(r<&uJakzYc$(ZP%pC`(W3#
z-=ZvDvr3nvr(R|TbS}AYx^VW4kngs`rxFEHoK)Y2C2;K*nZ~^sX5iooUMZwKJwvxq
zlLx_|=C{BNA|SNJSztLa00Gx_DElZZd-OI3DUP3<_VCNxM{lI17iYj=lXWdfT%r7?
zrY=4IM<Su{ot9(&Cqy}hJ3HHv-lGmZ(F^E_K5G^1r{`zy<i~DQ70!d-Ml50N95$-g
zrk3Fj%f%qBRiR#kknH$^<gSyqW&#Mod}vM_L5_Q{@<lC9xuJd{ZH<E~@J<&Ys-1%;
zC^x|ViKsC+8DzXoUK0$7>rzQire9&~X`2*j-ZWYLU_~_^O<mq7HC<C$E~1Y%%5YS+
zODmdj+w!Aa5Pm+RHM^2H1!bRBYlh`l%J+g0$aiBTKYUx_VuSpO{I(a$fbn+v!zl5s
z+G952JX;$Hite9BT?+NKKUlOX|LOjdekKUx+t_!l48PzI+N%_OOA4s>$b79A0CYY7
z&hOs?X$^obOif0|e2U@-s)g<#xoFjW|0^JP8|bNZui{5#U;^<RFH4`kX3SFELIoEz
zRbKJE;KxG}<?`N%YC=r#V!pSC+8mvgvG&glow%g#FA+x?W8ESOoZJ8hmj<bkYH<n!
zR}`nbe{VW*oKEWQr&})+pfgkG=vF&Yj+11e@N5=Yx`O&-oanc)PcNMCb`Kf8qVUDD
z1FgeQ;%KP$BYCPg)ANhEP{a%PWinx3XjeTZO>*5)vAuol=$)!iFqwFa*SL>%4m!oG
zuu6WET$Ccw@Kmq2A8QWW>Egf|lgMtl3#NGQz*&%4nli|c+qvsD0Sw;*)XT|mg7_ng
z^AY&{C(yfbz8!s1x*XSp+uW;O;s%7OB)aew*QJxX1U4mdstWj%EGK1E>~i83Jme1Z
zw|>Zb|2YmanO`9@o;0(ylHjGT5*;5d%yUykk_^WTiw-tarAy^rXTxw&uV{u#zoC0U
z?Tsbt+N=S+WK<mTUwJVwYm>qU&&#Tt_1{Y1ezNgUjXKcm!otz@>caB%=B;X>2D<tU
z8UvQ|cQZWg?~nX&Wpj8u{WWqkIpW*t)mITL1QICr_bQ)!ze}xlXakmdG{HZZ+eRq^
zU&wKAbiy&eH#B9!nax2)2g<1VdvI|Zm}A&@SZ5y3c>Tb85-dX2hP2(6FdYoNn+SsD
z3kaIbKOku1FiJIVy6E_8<f?n&z=cP(ey(UX?EUj1B|~+T83aS-H#LfjH1R^;je{sz
zy$w*+7>`4uQ)3gD$-_R5#cb3Y-*<#^`Ar0nOHNIUxLDpwPd!=n!TH#!TPq~}Xibj5
z1N3uz+5&lR7R2Tnj_FEB2Kmp28XT2hznl>GAX{eW#0dF_{(S|Vbye4-e7l+up*6qS
z;vd@SCc*{*ei^05d+N_ul5t};J$G(rzRk|N&S6(S_slh%-a)EFP*0#+_;%A79E~3j
zYlnjR)G(vP$<AA@NlpW2H=m)QyW0w(JeHF4fcf**(+THKPwr%<4i}d522U%fN1G?F
zF?6{^7n~c+s7PPbXnn%R`#wLyfW=R-Q>&-E;0vCnX!NN)wPm#UCd+&Ckn!ACoSP<^
zX!-<C#{0>|_p&;<^3{dE8gy$!^eWMBJp5?xJ5gw|{akc5O;toy=Dv7C7G*(Oq<L=;
zB|%oFQ?sW{QJikJ$h={lx4AbrmAT$>l5+a6Te7G#<88~F7)Bza^X8cIR(MoI5MS+0
zgFR9;p7-Lml0MA-)7kLLeU|Spu6B;xvuI3W<J9tR8P)OwkCg;5y=(YCqNxD1d2+DA
z2nvf1iDCUk+tE2ken8)~^`hqHVdURQOwTrxg=-|<{sj>Iw;zlXV4w_Ph7o@?V23eG
zPf1g$8J1S1Ye^*#&qVXW68y<SV<w^u_9Gl3)1gymoZ?b`{YFoCDOunD)XDwz8<8Kl
z@CdU+l1|~9%&#MKH<Rb1hlrzD4nl7_&$bT+Um<|#zMFl&y~W=ueXu%r>$k6+3Qfz_
z^}}+Xe*gJ@{OZ5IcFR75bKt9dmHyXG`9iJu1uVUC5Jx~tA5OPPd48B~t$tAb&AYuJ
zabZL0@FA7IqQU=T6$5{HK>(qO(O;CX_;26kKm5HXKSE4mJumwEq5aoS@%K;kx1alw
zG-*rPUAYo}yoLS?rvbi@9)ukf`{l0%_dg1?|Mt26_V;&p!u?jHcc^3cKdsQ8Xz>62
zK8zz^wOQP6H~9bh(SQB_|Ig$7@1ONQlT)540AiCzr5O5GM7RG19V*0dCh}VL<C_|9
zGD2@?lVHS>^+emH4(#16V*!&nlc96?uWkE}Us{kjyhe!lip#rSkejO>3g)kDKRGIJ
zJ2JXm#N;E61{T|hhZg}h*K|)lcu@5K?yC9Z^XEWl9eaS1V8r?Uhpi6do8mq&GqT)X
z65l>Jk_n?BRqgku|3wHlR{VmbAtM0=H@^VU?j4Nn@*Lhq*eH3peMd}2VQyKASUh$e
zkGP`WeuX!<5qb%H*dUk+S+qQp7%9cOc>U(M{Ov*i{WAu!ARmV%Gv(&*Rrh5Z^#f!K
z*??hx2ASzD0>TUdNFx)hpvJcM-0lA9bpRf@gtsb)trvJ}S~@JZ%6?v4x8s4D+6Ob=
zt|UI&Z16Zzrtdt^NrV?auprE%^c<3nRyRUqf-fe@BD;=tWL+iYCz-B>imd7mfX&4v
z0JY%HLE2ziBlMpmqgn?gUm>(7EwHZ~Qpi`imlUPHrx(1_gAG!vw(7;iT$YWIvdH{H
z7=8tsc)qO*LjWb^A#PBJ;hpIf4T+AwXpR4I-Fl`Y6aUZXaJt{AIXe?}=-Ge`*V#(m
zFF;1Q1KH@2o4~i6>j{A}j<7Q1$Jp7E5Ao|ioqU)>G>pU*fs;kH0UvS(X4rGJ*FPN*
zg-OCp*-muOJ?KkRb03F}^cZuJ2Lt1SaO}O@Cxq!$R)C)eN>L@|sTO3l*2MVA3W>Cy
zdVH_|S`2d}R&FbdiA}PPC5=(U$+{^U=$uU;dMrHO$s@0bjE*bmNdRG4KpLJ&$HFU4
zw^P#V#UlyZKa!D9XMx4P7_Gmb?El(My3Y>4q5Ul6Y0ueT_sq^mWuSwy08>;KRcbQ&
z)?q$lXhAJ*p@4|7#a+Cw5f1QRm3x+-y$#)=bhH3YXAUPAw<YuvQk`01t(qVQu_%ay
zRBj{-A$wG}n5F<2@{&QUc-?9I_rdMWL5%zn`LiI*%mbc^_cPH6rE{&)KdCRQ&}KhX
zap?oTn*k#czGw>oSHI%FkD!knI#O^bj_1!n_-+gwUz9DnZ{GYmPPp)x@63D1*t~7%
z6WrjFLNr<t{E8IhnA|ugOfv=>a1psLdLjC?!+q!SKbZjEMH+x2SC0;nUkBvZGoCOt
zJS&-ufZF-c@^VOG6k&YL4~aNO<E?M18$P~fB>+z|Jr1_&P!@92<^y!yX9<R|MF4MA
z?xfeYoxklzYX)YiGr5A^OUM;G1LP(yNjwCQ59XYBqS@A|Y1cR}z3Pl21E}bIFG4ZP
z8AFWC!tOckU0=z6Z-L<JT59Fm!lS#^&xq-=a}Y-lcxfqm(<3nhw_MKN--u8WI~s~9
zod0+xBL_xRyfui|Eg}i~um!)M8%V|GPJGPL!wB1dV9!E8M-=En_EggAaU(o0U#1r-
zDQ&E%{O>@_>IFU*W`!aao^_Y+-lGQ`Bt95^=Vz^%1uoJ)Uu$@L0_L#8joFWUcJhyb
zI^*)(SVM{h^XJ~)@T-vxe&*D-%(G;iGCnr9a#Y;D2j59}_4zGWVomuYYSm7MECL0^
zAh)dtz6IlZ-zyo-A#aEqhfvOD!F0RCx9g>{w`tST)2#CP<8}mq2lL@*ulqQYm44lQ
z;Wr;BO}%Rq@syMD+s|kIh+A|z?VbMY>35+iJ@XB`%@lqxJP?y~WH}TA7k~x8Ap7<=
z<nwi^C1y?LnwbkWUpJ5B3$?$C(8J(-<ESXhcDQDxZtB?IWZw^6?MjOjdaPde$;2wi
z?2g#^=h7-Yo7o9G2938f&>|cD%#`^22ofbMD)?>30gpUpW|B8}Dk+M&;^n*iyD_>1
zFevTl7F+Vlea;PG-S4vzPm;(I*ef;h*XqB|QO(|NJpO~jj8S`RJjx&hwN_Zi8e8Sw
z_5DgI$=fj9i)HB>FzNR9Zozox_PZvbRPKD*CkQol9l-gH^nlsuu%?Y7Y$zSyY4@FF
z9`=#;5BSo#zZs%UpuRV&nGfP^@*iX*vYXX`$!8?z$Oo%kGslw+Bh_obYc-zJtQLuq
z+f44<5%>4o;{X1<7X|mnet*4dzfH`a@sUaH1vH=mVdY@jUEwG~(@;zL!^o{C>8b6<
z@81!x>DAXLvc1F96sMM&#oyDopvjI@2v(_}9${>3`tEIS&iaY}&<ho^tpU-x5yGI7
zTFxmWr<SG%W$#<eZR_A2`Y_(4Pk!z>$*F%t%!6fz1WNA11+g5KupbEJ2=5O2a<*?q
zJ9P6fBa?X#{2Iymi(~%Epu<fx-YFSlG#MtfpQ?4A<bfVVMli4#aEj!<h+2gM+;{#C
za1yA<u-|YHTSsG>oizYYluw#lFd^&06(_h1$J5#j9`lzM!4doeBc-d<^$AMc1LQey
zVsTa(_`70s4!6T5KepoAkGL0??>J)kK&<t%k%?xP^gBnsah5hor$x9!=r*hA4@-tV
z#Z=-Oyr>-RhXrApz-P&cYn8@V{}yVW;5J;tbi-Y6?sJ$Q9HL!X24p`s+9>8tmv@h#
z{v#yvu+ty8>GFV2C!;>)!&{*w&5qdZ$x94D(Q^l?2`f|><X%L+nCn;owAQ<yrLEF|
zZo*W>0wU$9XZ%@0{B5G)sUy~)dxeo=Wfn1cPNQE&-(r_vK3dM9muhmbxiZ-P36Wa0
z%CgHbIx{)KUO3;6Zv^nS0Qqr&=_OzRV9dy8E~QN`KOo*kY(q=Jn`yHgGJTo$%6QZk
zGyrnf7gFWowRuf||Ks|D9fv#}AujLqa54LQEDTW%L@)vZNif^(2clnI6n)NNWd_Z_
z2P=q0&k7JeSSw1~tg2pFU)(EqUd=p&Yt9g_E36mYy8s7{k3x$;r;zmXuHA08_vw<Z
z00-)<)J{iJRpQx$!&&bM+#6i_1_$gGp(NRtWy2CXuo70Y=teYh>j&|d>9Bj}?kx<N
z@EKN$N|deoYSt{mTXKM}0UKD=XI7h++I4S=?gjN93t~?a+ZRUGf8Vyd&y!)_zC@f<
z5qHPBUFbR)x4LR{owzF-_RUMkQ{p6|n2DK5vP%7C3bqT9wn}_;yY|%i;3i?C%W)XV
zktWX&QD=x3P7*1TMj3}9sOelfBzyxeKO~e~Cdr*4VmOG7Y5$^mF)<Shj=PP}4_+cE
zbHP8XQO>pr2{qoGQ<=OqxPH@?M$4Aw%ChImQYeALI>5c)>q#UL)XkmY3gUtx_QX*_
z)!LR!bvPk#$a-ItrgeMve(e62h+VL9XLTU={&@q}zF57W=YfF_2~5<b93(m5r7Dsn
z-sWhp$Mu%c(0`Q|PsygD8<X@N)ef&<rQO-%t)-hx%H?iNU~ufqlHJhipffaHg{f@G
zD04+&>7k8p@XD)BVt0bh#D4g;>d=r^7Nxfn5pDLfaJZ*-2M~g;kOZ?yVpx?R%Dl1r
zdB$D@cX?7Ydp{JUxw-e(T-xN=Ui}>7eNB<2x2*xzE4=}dO!W#ONb32e@p0rRkUnwF
z)R$COM-jiWir;j+Yi064umB93PhiWcgsL?nW`_~BlVce*y~!1o9I_4D1sz*X-#Q-y
z?pi1D&9Xd1uv>NP_5h!JPNk1<-LAH*yzBw}!L&HiE?y$G;LE<%?2fv^xM?r9cxuvP
zZX!y^4fbIuv5A&gTT9E*d3%Ly{c!{}$3X+KJL61dMbv^zv*)vo!rud5g>~wVczIIl
zC5}F7QiuGlNk3{f`BdN3K<NmAL~wQ{2P;;P6>E0_yY&j<_fDD9#EuJg?!eOg9HJ0u
z&wyO^yYX!C!N?o;_wi<^MfA7pb&>H`yEhC>GdE@mV#z7(22>-3a#OQ~*d7&ck*epL
zmnuaUULih4xzkUZvp(V4IE2Rq%*f18?+I!L)%@rWt_<OBMFd_q*c>g40$t_ll?44;
zcmE>x|E)>=cW2kKNil!**}unuWS9$G_jO-$6F;Y8^zbXzp{~LCW}p3KPkPb5Rt_rn
z`+Ccgc!ESY-`c~wMv;YO{Mv7)>hF9~=SmJW*#pAUfYAPt%c!sMFAp%Yq0OH5ZCuNu
zvI+S*1r8#pg=V<w22!~=Ra<lxB-}X|Y2+&M1n^8_G)1@d6m5<{hKu)yiA&;_g5DBD
zkR6gv*igE+0W;UpJf@X}`TRGntOcQ$JRO;8Odt0d7^38EJvMQ$((IOgx5`&9J6}`4
z51#8|p>R0>^@^i?(y3sz9F|?WJc)r9f8$gV(9*@dc}OzvqO^n^n>Fye;l$$R+q0P|
zOTkAvdQQ69--iC+3!a%zQjh&Eq-5>?_%4z-QD|(Ih{+7#pWiPX7*vZmT1fqf>@9N{
zA0QU6D+E#Mm-9$qS3=4#Hs6r(=DJ99TzYSH0F5b2+8E>X`Pq+4`<wiS`+7AO!qp-$
zr-`eDqqsBEq%msmYp5BF$oTqV$>|XPz5R(6S}LeR(212{@Z<E`bK9HmyTtpZ&6`pf
zx}RX473SBE#(&bcNb6rdND<<n-4|h6QA*>QcX>AR?p#=1$(-+(NWP4yD0-3#x3{*x
zY*q<R(NOg1nUwD&s>xzlAX#%L37SgxfjUuDR<s;E8KZX9?Jtz#s0GVtt($tiDl_5=
z^cnR3UK#&sxo9DBzjV6&e7{G`DR>W$y_}bvTqSqhKe>NMqLu!}NGyiY>C%ul?Q3?t
znAGz9{1-Le$t)BYRB+4*isNv{+U6K&qR1v$Br(M)crmlcFGG5(lGZ+y)uk^Lt0+4L
zLogW5fa=kq^Ii;+Df)|H8F+X0aX#TWD{-N$2ig2gB9*-~+A-ND+obGQdbS*VJVx4B
zKdyg1ov{ahF?l51#{-lG+UVyuEH^-RS2?Xw%7Fvs-~7369oK7nt~FSR$X*@SQ<pE9
zbUrztCOd!i^J&95)8G|t8G0wq>8E~W`{thGH1a#@87i`PI8%uI2r&%RE^^n>>8R9<
z$`#bJIq3dktn-~RW{TtNi`3A=d=!Mli@~OaA=vtfCH5QZz_*A2_EOrM6QR0tpx}}?
zxUuQaHc2{26F}OggrY+zp*?44c4UX*BQ39>iV-6yZP6n90og~8rgMkyrm|zdPNdVx
z+vutzw&GK#f6VP-mO=Y4Y#Ostr1hkK?BTw;v|i+Qctfz%+hfN5ZB?Gr51Hb6Z8g>$
zFy(;TE`Z(sANG9-#qQQ~+s|J7d+V1`6@00Jyme<<9XrnJ<BX3QccI5C7?v_qkPL_@
zKQuo&L9M4930AKGr6P{>ZdojvRQExo;Vr}6772U9A1sc=^(FSvfxD(jhdDLb^`IzE
z1}d66s3Ndd_O9+g1M7y)1qpbFK@w%pPTOUuIY7-Kt<y(Pug~tgtL$$9wZk~h&E@Q^
zn`KfXF@YuxyU51%!?T%lMX<PNf;lhR&;(bL89ArVU=XCqWQcRgj%;?i`ktVlI>skF
z`E!x<?N%e4RJ3b-McmP%IeIWZGXm5`{hH>69LIREZ^va_jv-V}yU>T@1IGw`W{<f!
zEM9yf)RRP^)*Z)7H2#cu=QyK}dnjz)wkzV*sxSi@d$qc+0BX2~%EidiL=}3duPr&g
z@UBW8T`I}Bzip(Sf!od7$&8-RI5UM?*-Jl>;3ttjCJ<lnx=Hh#DNeE=UGuQT7p)&k
z&p;3mMV(*4s^xEh<L4ikV7|C?b(~+n3bhm#udp#yA!;|2<@~{BX+G~F1J3e%=XtEk
zPnk`}r<4RZTVyPA1}lwY{ce)%<4%c9#0w_A<+u_kcv0NeR)R=e?NHn9qc^#Jgy<?P
zi*C2X#SDntO+gxU2apy-5vb0Yd*rs12*mg)1iOs@g3P~aZj7raTsaMXhcO?77JhLJ
zE7nIOi`8!R8yKpGf0*X1hisErLRijCyEE_2;R*2$?6X=8iGF8R3rEb&ZVHFCLD94>
zw@#h$39tdY-ms<Fq-AanECZV#zFmNlO%1^Cnou)Gvm&rkrJwTg=Ng@DBmNQAy08k!
zMkb;5;@MyJL7rj<=efeh2UZ$Hhq^&2U<|X;4C2I7OaW7sWP>Ny9Zf1Sn$?By{*umI
zbKYU;)t)<`Rka_Jyf9D#O`OkH8fC>)Z`21VfgY%34PZE_llsCY<KP0>VOM1z58QZ*
zh-Q*At;SMIVjdD5t6N+7YWF<~QPOj>r&MIMuN0O|&+Xflg3%AM!ZW?#?|f(l&HcWT
zY3Bz8L2D$vN{f^?Nhycb5L=FtdY_iu$nMaH(>-B2wJr~W+b&8!M7=9W3MDB(&S;4<
zorPs&UuCTGS4{)gegkckJq_12N@huKXVTEy8y8fo3$@I#!+f1LS7%fkNbSx?PA{+z
z*F5c$k47*Ee3fAwBldx4KpxAm)az1Ec1083=Iq>mzDz_jk0|gCgW|~R($8{tidru;
z;xft!wm%Z?Y*kl@@v~sAFppfXKcFY5zW~pLzbc_|>cn8lR19Ij{4~YOD{KzcO9xG-
zZgb)iB+HESZ*}gSae?j1)46lYDKQG1>XD+N=%0#;YKnQ(>wljt)>H_9L8I|$m&atQ
zETaNPdO%p2cj@ttmxKU$?}Ft7mO38~1>wtG%BwXU%VGk3)Ynh@nl|`2uL5rOgfkXL
zwJN7pV?#q1bv1E0JTqW9pK}`(*Y~7D?!4rb72CN?0!5lx4*<ePH@)<@X>Qs9qQ!2+
zC;IcsprXry_D%LfGoW6)H6sW;%saT6U$C&YAP00KtXNBa3tj;vryfoLLoa#Dn=~<5
z?F%2Cd-j{D%TaGfj8&wfVzqASf@nyN<LgX&-^Zm4?v7UkiYuh7cWLWlV=Aq_HCUR$
zh+L1WZ@u5!bF12hzT(c%1O$Jg=QPbCU2IJ$6+vN8{mJB+=Pvb~{VAQZ`bC<@4-d%%
zI=hKB56Q#OCPYI`4~yTm6Fc$Pmwnc(zvOY{#Y`&fK3O%bRq|*gIT^)5dNGs(NgT4V
zeeZ7DqSHlFSj<JoG|zX9Xsg!%w)9xjNyQ;G^0Q7PshX>EqQ<`HrryvfrTBM=N7Vxw
zuN-DlRNM|$lx&gos>MPxTFQv~1S#^Vw`^}Nic9;8KIBUTlR~5nFr2*}{}20;hKW7$
zZ!~6BYrq27(B1UIk1w_=kd82i+MJ^3159PuAkXheQwF=})-d+Hze5l;o{6zRbg!)j
z<ZwLnd&KT`uP8dsAMG@ORYv)9eH1PC`PyE>$uAR%(TA-J8}8if)d+fZe;e>q5jYht
zIXbF$J9dC^P_hhaEuN|}Tt%)4J)x-F#pl^Hkdvm~8c%MX?zvF?o9-t+D8b540A`ra
zN!yOs%Az0fxBC%V9btXSCUmc~ra$>a^qu%oz59emZ%q3+ZuaRcXe_8Vy}Mc`3Yo9~
zka2Z(Ac#(u@z`f*wA>`U{loTW6%Z))E<E?^RY0vG;9k|h(7+D)!A*#&I)l(29|#Vf
zh8E_1Ah#i>>0{Vy<)X7InEAIU3M3OAqH2ATWEs$RX2yPSSZAc+9nKbshb;<q2iE3e
zxB@Nu35E!DOyBL9IEn1ttCqYyz3KYfM=v?N$otc9qYN7jLmQpZR@K3O!K<t&awlwz
zW#MES00|mBj1*@fWozX8Ve(<;#!B6!eFA-%Cc+{&j8LMO<00@^;>1V8q$FK^1u||@
z=~V$#a`;8i&9TK2q#|38IYd4@8<4efHSwrHXrkY){b;)qs-KKIZ?0qp&xwNU2e`P&
z6D))gB+KWr{O|&el5syfMb)hVP#r$>`?yaj@fE|G6>Gd4qHn;-sGXS~7{}YirXLpZ
zzB&%@1uDa(`f`f#&Tnsls#YhZlArSi2{|{XZPhi9%=FEFe^Xzw^ESF2r(aexk*bj&
z_-#eC_@-+lR9lixe%3+j^IhHMRwk~c5oX@bU0dM)7>=M+G1!Y;N{}dG0(LDy2E3f<
zBB8Y!A}{&n{Ue?Lqw!qjUS2TNy7}fkd-rv4gO2D5oP&ERNrh#{tAw;*LmZiKZfXs7
z<ggQ<tR~>-wO=B|Psx&<=UKY&&j*HSQ8y>R)QHX8Qyejb#cG5|9M03K-fGVdFZ`n@
zgSmwd8G^<<zbf@>2=+{r-1Ydylb*?GFEV#DOZwkH5WL-J+_4HTn2m}^>Kxgn;&fF~
z(`({{$YGm-lRv(Ltt|^9*-l&^279O-(!`t3LF^-D=Yjo^Eir=luxAX2Z<+@9R8(yA
zLi+omZws1_slrjiIO$`+o>cOd<LFe<u>$!Z<}qe?8BmzB10i-5go<b1(L*1A%&GdM
zc9UoB`tqCa>=QxzD$oUo=4j~hqcGz5kVJOv4*}mMTFRUZ&V-;Vu+mZ93+zdX>p(#I
zLbI<~7TEo|E^sJRO3}?G&iV(<^ViVoUI+BX&Dta<wU5y|uIoFJ7c-=4CnhUWA7@=b
zr8l9fX(O|6^Qo%dLDAC|EIXTRP_>xWp;IAMBCe2J=~ay|A$RYiJ0y}Bo-%PVFVccK
z-thkEj~L&>oxv8*N*cxhpcN5Wr`huZ2v~DNbpe<s{+w7)*-NYS*DM1zUZvsrZ?c>g
zjdwke>_e`OGzeEnVy_>zcqZ24%cfFn!{<A8y9A-`N{RSbL^=x@GU&x;8)-xKG)-Gx
zm*UI;vNLmCFaQe#(qOy2u&VzTo{Uo{QzWjp$G*u*^UmTs!&>$m6@hr1XjXGGizo~y
z$`2MXy|m5S2C}_Cm3^9xH`?e&{Q)z!W^QfNVBj>~o~G~#W(9_S3i@Zyravc^Dyk0)
zhL;?{k3s1g7qCGOQpW@M-qWdR2i&TK@zZy??M<ObIDcoZdUuveC$@7myHr5z^S;kH
zyuH?9k+apodjl;HGjCJ-d3p*m>+~~>1nNz#w2by8_PlCEMBf*cQ4%Z*0r7QP?IsD|
zivM`zmrcm)VltM{JO3tHeB?n8%tmG|NMX&B398FxO%Gi}BJ4~nxn%udd>D5N91n*}
z3Fvd!<6XePVGwo5I0t`i_gqqBVwl~|GlW=>a&Z(`Z+z%a#vD=H+ZZbjR<2de+Z-Zq
z9@~`r5+{W1lmy20_a@@DyDT{TxwmPpNZ}`^)GTs81!cpCCc?%Yh>4w3uZO)?fiEi8
z%GM(*+}9pMyws-YVYsE-|D%$?Wmo=it@eA5*`@)JSo`|R&HCkot2jdvk?s58rTU(P
zqlfCbFXDuFXbWf)9qOgzVglbZ$OwezgYY88FN6=9B^~=jP5(hbvjX;O6!po*$~!d+
zu=PUk>=#|WYWMv4J_ez4HQh2?ohE>xJ%H<_6G>C59M^nm!VyMA@SrsKf`2s&_e>U`
z<l^Q5w!scEY*|?Iwi3E(Ct^Vh@DCAYuJtRi)nX%#(kz)H>c<NXZMQ#AI5a)E+*iUv
zXC9rn>?t_Ke8g@c*4g-m9a&PG%hVe^HkBsjamFRaqOv?AuonIDl|1ju#rL|!h-zA7
z=osO29)2}WNJJ!6&+$<r$2h;k?0rEy?quSJ7EPu*9M||(M<W^EG7y!-`-P*4T=kk1
zM@Q!FKc&6idG;ao<X*zKJGV;fZY)K0VU>m)IOkfWsdh99O}yZ;BYVtk#9_Iga3-(3
zq;%JHEH0M1Y*;;tC{U#PRlBr;e{%a&^3wy;S_^B6pDm_b0BmpYdH6$5^+u!6uIAyG
znth!|Yv^L9&$iw(ua+^*122iFz>9?Cl6;{a;#|d$+L9WZ7RAqL*bS*celRvb#%T@J
zU0HtTQU`ssY6?Fr&G@8k&hR$eTM_7dig;0|-O*B0Q&j-VT5DruoJ~{{8>ri=E%X-U
z70BEGW4j*B24kG$dGi6E(D4fCjT(Vcn%T*Y^eLe+EFynMd(eGRx1#WRpC4VqWTY8(
zs+R7Wxl*@w5JX%LRC|j8>Ndz-<5RVz)Lrwfs`~4AJX`SQ%F-ny$Z2M%>kBeK+K{lC
zo2|;T>z^fwuQ~sN;Ai)4WX@T0{>^Wfh?a*t7z?ZlCs~mgKdJdI@wLMvi!jwOs`TAQ
z&0Ph;qw5<?B;13Bk~1=KxKiFK>S}DkXC5St)KuNo)v;NUubk@s*{Y+i7dTm^F8~bo
znmEC2hgQOST=;Uq4HwDjjfTK4@$fz=CI8R>`@1e7U&y#0_%Y@FBm)Zh@1SAFOG|lI
zQ{JoE;V%<FRleicq?OJ!5XMPdxC<9(hgyCL<?~jgbMlwC<ZtB;%<4|u^AecJDQdFt
z7s3HB@&mEmoF640fPzhDl%Up#iND(unqp3_7fW#ErC`WOm=@HM4q&2cJu2(%;ctDZ
z5s0KHJmyH7>I|jcxcR$ABgAuFZ=#FWU{8BZ2k0$2Kh^OKbJk!M*eZ@oUq;B&{z}o*
zU0S&M(-SB41<-QGB+L4?F7`@2^;QeYJ)W#jZ;O*%jl$lSr0@)4tPrH&_rF0?*~(V+
z7G;M>0f!TF?N-(3m|sC3?rj|4T<KB~RXB0gZq8y{&tK|>CgHVpn+biC8SXMag)+)6
zle6an&TQApC~jQxO6N(s#G?>7=Gc56u?nc6`?$1B-Fn8&kuw!vVS4D~nC0DU%j=ZW
zXUPSn@4h`?Py%?i&!sFCqN^3?`}W(=F`rnSiF#MN7GBDgPcwx--+U^=W=1ZJ*mx~|
zBIvBa!|JFYHnV;A6+3TqhSc0REMMj8tho)k*Q&lRn_W}{CbPr$RacH4;BUrxcs#n{
zqa`D=y@|GC^gL%4Lim5Ed+VsGw)KBl5p^RfB_SY<ARr+U(h`y?-5@Oz(g<uq5F`a8
zqy%ZDLt0R2=}@{Gq@_jR_biXzINy7{@4xRDXN=>(jce_-=6vQ8pKv;BvkV+^6_)Pe
zxuq#TI{iVWE;fsj^&OPkqmG2)F|xL=BFe#uFAnB5r2Itj?$REbSU7oDWe^OtJCwdV
znx1p`C1Cg;@fFAMNJt=KAt#wiRsEHTisWIg@zn_nxo_xSw7H1!f`$F#qWYcbxsX{u
zDcqh51e5W}iD&d<e2cOOZ*JB$?i7l<RR;3m5d)8>o1zkT!Y05~aSGl1)tm?Cd{K;J
z=VeTx$`B9s2eWuhVo|4R-&aNnDOho=R|}!9GtWlFe}0<Gc(VG0Xz0r~{DGK(I|0M<
z43?L>@>rPWQz=yjF%6&b25ej38CoDXNYAO?E-hsK?yFP%Vq1ZTY=-iHb&%P(Me!ZC
zf~wj(=ODJ^6pL_YclKThA{F3yi*5%=43SxeR0^o^G7t`}y?S}HZ&hNKFhS{W(>65}
zwf-@EL-Xd?Ms%U>THXM8Th4syxyrcjd%hx<JBR^X=_V;Z!nA(&z*uf$P_`<F-eJwv
zirk4+c<7ZwoMV?qL!@Dr_^W_LxvWCeI<D1J6XL&=lf#gcknvS_l+2Xbvi(Y)WW591
z7&E_tEPFlch7)@Nsw!ZM=v7WEz2h^T&QnXr(I)W*V`;aY<JrjKpy|@jGwAh<^TS$L
z+)ji+%tOKZ<mk7ns{5y2PK;=o6_t-RKe9c7#G1X5Wux(o+j_T#s~Ufajw=Pl!H7pd
z$KwAGDA-3(a0iB|M1L<jKJqeHgY~e)u4bi|NnXL!k2Dugl<Q5@_zZ5+vL7C;sA=30
z)f9HOn}gDS!K?fQUm3Rp*X;4RV;XsX2cUUAyu1do$+y(!!1S{Rrk$0sef<v~z8$=b
zal-#t!m)Lf)UKK68|Ht4!~P>CN~8x?Yj*6LuYZ@G_GD$S1|z~DEQXiz*i|y9QD{uH
zn?&oMBD-efhoGQJApSmbW&?Q0{2dH$r-qF&xCbTkabPV!L_LIj@;Kl$1}vbSM4(E=
zj`tY;+%%`@yaJ(~pZ?9<-;pnRkMR(9`6Yld6d?oPu5S^FIj5hTI#&o|-##$FRBid}
z51_^|EcRF6N3M<7i<;rg+-UfLRCX3%A<7L1i{6#R->$*`8BZsI$wv=hnb(;AH_1E8
zjsXaz7(_rl%Nf(c;t+rrmH;wwHY~}7yZtWp4_s!^rZ~ZEBZSHzyL#o%?-nqO$o1Y$
zZ@&CvQu!a>+s_}`A4ecMdbnRO|2VS!FM$eHlxY-#m<{`E_dj0Ee;RTiztVXUsm{D^
z$`kyzNBTd0o~VOh7-jEt{Eub&-=2X8F#=@8K6Uf<-}lFV|I+{VNl!;)3$xx3{3DU`
z506I&0kX1^U~>H{kaavzV*KyG%Ma@DYF5b7)_aU!1|2n)G$XK+?mo8c8T%WN_Mb1q
zPECNojVW}tZvTuGdR`1`0QMUMJsncYw1p)Y-&jK7_iB7U=X}6N5e93ZxD6fs=s4;J
z38njp=#t24`(x811cfPlxE`_$y}f{K*5WUUh@jPZzmlxyaN&`gdO(_Q(EJ9Zw9=78
zK$NuQR(xebLkw7kys!{hbPeeMwt(#~g*uCvW|S=UKKajg{Cm0Fvw*kjUeI{2>MzDW
z$v7sW;PcaCy#E{2GXuarut2aa<5h|q5{dj+2y1v7<X-!rq93+70l=z$+!YGMY8tBK
zN7}EL=Hrj)R^nm+@F9E~9Nb55fhP>2YQKu(=<JV#DD@*A=uog)WyJQWVW{MCxw*Y=
z-g@>C*iS^FAMXku6j|Y1k3ec}OXzEtf*jcA%P~gyaiHHZ8FB-PFTp9w9YnzX>1+Ex
zcP6MX+-PC@j`3zh{Mx?u^qz>a)jG#Tg%<E7Hit2P2!bcm8Bpbt!8~>CxB#ykf^tJR
z{Rqk}_?<1(!ajb))T_wA)W$6{(F$|z)GipsHj=Qy;KvcfYIc^ZSy_D!#xHF{v|pK2
z&3xcSFTbVvxo;;^dkn>r{}E%m`ZQ%fMBUNS(`VAw=*j@9=|lVx#Ppc9BcPfW@P0gO
zXc=qMaM2PZsICa4DgBRNG0zWCWxrj9{#P7MG{+j(hxGZybKHD$NvU&W^O&HWLm!7&
zN5f>K1Tdf<rdBGU<w`v%vAGWRwB=LZm2V!QTo<**3c)Akkryro*vG!uhH?<04s{8m
zm_{NoGj+`?3d2v7DR_XN`Z5qhA1$3eSUxo>g@=&<cZ+}=1}-ScDX5*)l|6lqR5|MP
zpOGoY@#b-mI~S8R{(a}N$mUKvkz#V29h0s-M(^o0J{szAZ>|IaMXRD-IjIzlv8b^h
zvpgjLR6wkc`LzuE>AzZimHFeLEP}}dr?TEjTos>RZ$8KhLxOs|Gf8Y2Bt8njkF8GE
zZA&H2ZF`_rYa9`P2Yigf=X87%&>x^I0c|ZR0a?6+P<96({}v#iN5E5~_W>9;IJE~m
z>qeylp%w1ehw$i1phUb@_{w?qBBCc3AYl<EeYpVii@0mXdq&o<YqWJi`*a&P`28?N
z!lU>MFz=PMx{_`g<^ec}zU)e+^=-zp$9ZeJLtkzQ0ShA=835X>LjYYwTzIc+4Q{~F
z8ibA>*{Ihz@SNg?5P4<<^UPj9grhsNHeL+e&le?ofX6Ky0|aF;+9?Q;ZL<JyO3q9b
z%z!4Eg1$>R5gNFf8unlvB9`_21lnu5l77RWPd8KoSd@zt!EhL-J6&CUeW~a^WCi*j
z!_d#a0ksTwmiZvWB+{vuoW8)t0|dCm`z0HYZ_A$nE5Gg)z(Jb!A(1PZ4C<PE`1Xzy
z3n4Z!?%x*{N8KPZ7=(`*00KEU-f7aGBi%c~QqJ3UdJoVPW$UWT-(dvCUMWpTkW?83
zN{a@nEp`nrg~nq-D-L%&j(BzvMqBA^?g2Ur;IuD=7SQNN^?&|1IA%u_v<IV0K{iZB
zgV4db^+5vIyv@;yUmivn$0&vN4TL^x@r<kY4cN>EI?kCs6jMM+QR2?$A99Svyz8%r
zzM>JdH9Xu?H^DhX?{2o0^Wj6F&LL3X9MvOa@$c_-cs%qzLtij@0O-)%kvZ?;OGByW
zmtvhOKsoSBWD_wsnR^1%&ULRa%+Mys(9%SQABxH)0MwTt;(%cEOxe%#e*N4&D3x6@
zCp5nyhvn187pugR^9Xp(zKC>Y`2*FLaS&D2*U~Aon|^^fErcFCqMX@d>LdNSXy29^
zd`F3YgZ9JQfb~xf4z@-)6Fe#jd0(~wa@~ZGAzv)Ix!<>WZzv3)()Wl^!dJY*N60k3
z*Gm;GmHReNa$B$x#~|uP0qay`e)RkpVl}TxM3L@`L#g|YHDL`s526OBeHwvl`-1x&
z&`-4_dswEwu=+OB1Lwj}irpW>lP5ww>fPv!{I&DWN?~ZAoB+bscv@4aoklMP5O!kI
zPfUNp8aYM*6stAZi1;=O$!}DXbfIb@pp8N2E&5a1ZF?6oQ$mAR1}7!SZN&XF5xnKD
zyb!b(eW)x!0*z~eDmf2;SUg|F<m-J6>qJ3FEEZ|K7k}ODbO&=aA1bd-G%EKEP8Svi
z!C*t<0<>-g6Y>PrCz>DyBo>Q6Vlzqh?#J6&csgCD>Yn+LgJ;+{g_VSSov*9+iM7Tl
zoXPX*H5c@8L}i1a!SLY#s@Nrk({B8SyHm)yd!~R$R8Bks(f`R|nYvA;*G}R#!@Lk{
zXZKqR;GRE+NTQmt=!GZwtKjOLMa`aeH3*v`i$wB!fX?#!{grAIHI+f>Nh0-+j{W+l
z-$9%f!+rWOTLD!8=OwA*O&0*S8xG^CUQcJTz}O!EF0WX=j2qIL?k6-1kfO!z%U`{<
zUi;|7{s?aBA)qt^+-aRabjR9|Q;f>Ljpqzj;7Mru_TjG22sGf_MA_xB`4@}=I#;yl
zDeg(K+FJ!0n3p+Q*~MBpF-x9J&=;sRoFejT=Dn|tXW#81PURO*@(vNM+r4g1K!ZCg
zxQKxv<z$ctHssmWSk%~d#Lw5tq~cq<h7(vYwc$cfgw3%HxEpX-*SxX0%fT3Jrtxd=
za`5SmfW3VPucySki%?d!(D{2Unk{{R62!~7CK-TLl1-gWpS`oj@Wn=>rXjzIiNF;%
zqOg}^)UQ>8^Wq~+(`@2woNVLlnCwfaEW8uJpa_~CYnEwx2X>Dyxvig$Lyt-q>+I#f
z0HAInl@bQ+`g7L8FfQ`CIlQ~-F*qXWT$x#%9st67uN^yv;wkssRDvXPv%+Z|Til<%
z39fr-^uEC4n^t;h#2mlCH@hE~KiN?UhkSZpDo7W#y!gEM>`gl&&kgh$1tE(-EDiyW
z#~=BgNRewz?-&`1zVsiDcENMscsC78tQZ0;$F_?*KcXoo=+2<Jg--mRQjd((<CJIq
z4YzspIMBRoy5_?)#YPzVgxm@TQ+d>z>Cvff5Q2>vK5_4-zDUvQ!venYx#(%is;P^_
zQK`rPrRgN)0;Gv|Sp%e3RPq})4kviakFn;w|NgZ}@c8AO%t&p<7uvYW%!?Gc$rQwg
z68tw5+4kzEDMP0OwJeKOzI8fxohCk@I~q^;A?7$}_2D2au&w(iXp-_(V2G;Yop->3
z`C?ytjz^qKXFkmI33fv_C=n$9|D$T!D{FePnPSyGNn~?Ga0L^acs>o$vMV2FX~0Nl
zEs<H_U@m_I7wbTOnZS5AO|k1mP<dsFCauDj^|GHU4o9@M0PeD_ji^q{9kyX5ZCgAe
zTN_YfGj6JC@Dcp@WZkja#P+7PEwxBd&&l9w>{E0WzP2kW%;o)WCGGU8-44s&T)pJ*
z1%{-dFvXGg4I~T0er+&JB0JhqrkBU693)g$21w0l;i4|V28oLnw59Ky>Sobzn+#q#
z6Y<xi?}uqSMQC?MYVSsAM-y#$-nbKAv<~I8?8@^i<(;(UyG4?btiw{9Ar3WyvGhCd
zqlXubo40T4ATlD^1guL->#ptA1!@R{Q{S3BX)K7#E<%W-REOk+wq7)4#*h9c^0sF-
zT^81fY}qlan$K<|I9Et<U46eh$#FLi6ia@nFZVY&v8%=Otu(PUOhLMk(bM9OeKaW~
zpyNlfk&^rP9{c(1qzHXM`y`tT9osQMCoZAU_2R%-gB&f;j|mUF6OJtvj(x25JS$9b
zO*t3Gpr4T(-`@wYbE3WcwhXOa{=r&49Ja%P+zN)?O^|*Sp1p#`H>Ww!3o)Yn*h{AL
zZl7^Wxwuyuw5PVS5E69Ak6fk(+P<D9j*dPHGS=-N=N6&iC}r2fs1HAk1zJ`EweJaS
z8{;gr><1XDGY`zEhv*cM>|a9PG&G>J&MV-E8VE0YDA)fXFl~y*6T2xzxA*Wz+{gOa
zg7k{gsV|?>GsjLmv+OSZW0@^-7sF{q;h!el<CulGp&?Z$xn0yfxs`HeanY5e+-RLg
z!TBfjb*J85m=xpoLe&Ykg`tZBLO!9=V$(pEo!{HA=rt&V+;pALO%)C8V$8I?Yfo)@
zQ_gF=Q}8+f)&{s};#f?XRb|L<J*m<~m%3T11FoU61jNmBICe(?$6+%jmeo7bLhLGj
zpJl{cp2<++M%@X&8r;&zXBxR#BeVv$8fvR>8g_YlrxIqm?|OXyJb&6vqDbKduHOr+
z<AKI_8SuZXLac@pHkuk_`o5oV;zUJ;SUB?LU;9p5s-}+Zq0+uHVb-bQR#)(R@ZKfT
zV!2$Gj1n3aGyK+>Mg56sbZ%I%O@eax$-Qs5Wex^9D?nl0U*I3xyCcxypdlKH?OU}y
znB^-H`K|TRY&klD30+xOd1sDMR0PwamD4*(V&d8}BSG)yzQuP|PO?&~GpEC=%^P|q
z`f(DC6jq4d-&yl?--)XxZV%^HELps5QJx!ob>Dp$FFkCHdx?8ipPyoz_kMi=S?{}w
zQ+p>xE$dc)7+$)+{OQ2ZN|e2sv_G0}L^2~Uyf{6R>utfkgQ{Eu_X}eKD^3>{-+Ude
zQh$1hCt&2eqQ~g|UW!)-XL;$+G<8Ebdfwu@CeGX?yNe2s3=Q0j+BVnfsr_bn^bHM!
z%809R!#DVISRB5g6SIY8{@jh&L^xG<FE``;)no8HE=h$gBap3+<0-!q<a^_+mHu8u
z&-0M?e%x!5VZx4R)3LSf?xh<-XW0O4M13f7wksOQU;b-SQ`8HfTeJm-R)RpKk&^_|
z)wMA>ZyS^x?wTtF7Byu;EG`E1v32VNY7EAWQ;|bY*3MeFpt73rDL&w)Pn2h9G*hnN
z4Moy7e6Edt>YG`(A3xo7A7}0&1s&N*#d?=3zGe1+56LR@hoXx^{2oOx7iPpW8b0w?
z$k@TjfO$L5Y1E(hwff!lN?}nEflz1l9^I;Ui1IS^E-$2(0eJx%ESLh2S;yuQ&`Ji&
zh2Q4Z2)S=}iu)O{6sB!9kFggnroS--r`;tL%AuGP$iyJ8;<h~R=#}VS?gC_`zQ`;C
zfo~Opoj#9$3M5j~Q(&wotGzVOew9sVY(BTEz54ak_6K5A2d*ZtcQP0FbhouR^5Wg#
zh}1C8)Z`32FrBLm*i&$|BAiml&I{l0WW-#$Z4t;LDxUYFy~V4KhW7B7m%|$5;SUz}
zUI<^VmSseX+pNr0&|$XH+5f3k?`b2chpV!G`r<#$#E#2-tST8m6eCy}IdJMV2g4Dt
z7}8U;yQ#I{wlix$q4-6>C>{Wl!?o%W8zCRK0U{6=hz8k~+pk{dd(a`07Ht9%*@%y?
z!|Hl|I)dBmhcnx9Wg&Nkf*6b=c#irlK<ed5`CKf;PT)mWQVtxe`zwU4+C|EhDoy2q
zHLZe&F}~~19_{+Bn}%Z(wuxEC`qp#qxSDQeu#`u!kO5!R*=ee9==ofdS)16|Sl|Au
zSEn$TH$u~96W+8+M9fkyn9pcOn^Kf~K}M|n6W#<aGb-r(5C%km@JP?fS{=%9O`(~D
zi6yWr3wEp^&ZjUIy#A(~szE!k7Zk-z8<LsgtdpXh{ZtS+<F+dDQ?fW%=0Aa|dYruP
z%DxDQ+dNbZNx`OXt)+=YV-_JzfW`yIYudy3hV#0ynnEiWOdIU2B`_Tx>~P<H`@%S;
z6k(_BEZtw>dYj8)X9W`E*LFr90Nqq|q{?+7DZ7^S;@iTyBm0j++R7KrF&!5s{MrO1
zE}k9XjOUo8cv@)tj-}-F=NsPvK-3~&l<VxIQ+pUoZ{d4b&Twl3cTpibbA7(0Vr0mv
zlWM6(FsPodE1PnP2^sDOA9$@ZUS0~S`gAg**uuiMc*N)Ntmv$!5W|#RRhIcD4>l4z
zoN|lTFKq+L87s1-9_SM8n=+W+RxvkuF!w!Lqvr?X``nRr#$JV<y^Hr>;%zTZR=5wG
zTHVR-FA%zRZv<_wa{H^2AnP#qEb~4hIi<AI3J2@O4;{NX)vHwvbrh>K)w?4zNi$?t
zeL7Ig$E**1I+-&&`e1Pdi8#45|3mQVI?^asx?Vy3o4V>Nau1U}$NMqGnRf0Gd!-He
zXzFe)U=52oO;0Gc306k?-aJdKW+f^j&WZBGRi~EO?2J2(*F{l-i$iSoVYi-5=A<Z?
zTbGr;&eg4^CB6A_2RMEM%1XC;%=Zp20D6?#!Y=P=36%YVP<@-PlwTFU>nC7JD_UoB
z6*IMMueJ((@+`I1%4y1JJO`S~j~#ljlBQs6h`+USS#=Hf@vQEf2GAmuiCeR-n4O4t
zR{jk-MT5|y&|lfQ)<DojnoPMvu5K2$wk-bgvxO@hNJj40+EMb*%!o4S>G=kI0HTc)
zg_kNmxS>((Jf;L2A$)jQwr#`KAu$C{%lcUC?%EV{t2isxlS=@#xJ3}7#Da8^%bNq0
z%?hawqgId8rt!#OGV1to{`bA_uALBZDgiDMnYq<H0!M{<p}{6irJ2})K6joMS!V7R
z7A@}!1azo53(5ODZmTO<@?B{g7=H4)pyb5?^(l`;4wudO-K)_aeO!hv(32TNz}e7*
z&_Z<w&d_19ba(y{pbMn)LmSzDssKJFn3wgF-Z70FCewUyccfM$)+F{p7}{3adkgWh
zz^8V?I?>$a<N5$_;FjNA3%XzZm}bD!EFxLy7VH#Pzfse|C5sG`ZSQ)nyAO6oc^FeK
z#rBLvxz>1Bl*UbCeB6ZY@MU6=TT*(wxb9yB{UqM~xUXSLt@oJ;>zv%DVburFLy`uW
zL**<QycaRJpS<<k^q=Y&;aS>Sd8bkT&{XYjjALaF!cFj&SiXvQxdJ;%FHM3U>!f<|
zZQM+P;C>!`NV*z^PiEqsD(Z=L#Ls<uA?W@#*LC)!RGo-8?be~}%nH{XC~GwquN~0%
zo&q1cXQgI+prdYY+u>KdsLJ3SM`y%sxM~o#3EH*kK;xd%f8{;uux^~yI8>q5L!Rc=
znl-9u@&Rdm1k6_6X-r{TbiCJKh^P&+e#|Xi-4L)@ece>`blYhL(MUwq>&SVZF<xfD
zJ||aB{(f^qcQwGWb<R7PUe3%zQRU}$?lCv<`2MOzl}p&xjeO{0$mE7f0!|$uJdb`M
zJdsjla))wX|LwHbQJodhPR($()YVsG=UO)kzw_-0%A-B+{CS|EiJ$-C(LZN;dLp><
zfr+UxAW$EH?vY)2z*4^RA^l*+(1w#*f1TRf%<l}IUnuU$R*=$s0J^WT6eUIKMJvO#
z4;y@EeJS0`SdTU;c>!)8?u&KqrV{6KYJ<oQS(Mb+hA$cii;VF+@L;B?0|*SB+e8XM
z$6z|iPE6Yyw@9M`gf(j%$Gd=JL>(Nb@?0Ru2-VRn{#>OWb9ra2zNv~jvgEi8Rk8@q
zew=`u^@EGo`0?hhXBFj%iX=S3aeUAvLHum*OSaHAL2`#R`;Iul02OM?%IPrKZp33C
zW#wfxsaGd9#{^=y<Y?TOISc9EASzF_saCqt_4OD+?3{K0Lgp%~DrC&ZUi0YBdedO0
zPE8qBWt})23p_m2kvS}h9P#R$El!&}dB_oe-=i^Q*P>_cga#_QzGOw8Tik8QZAlDp
z^ja;#EP8GCU#(i+mRxF~JS{JbHa8SS^{=dYHdNUEd3+*&2L?2@tj&m+4E*PLTmrJS
z0Nq_u{!Te42m)~`DKHAwpAT+9bChEAJ1~mHMM@D}QjxPb`T&2_a5{lV@R@Bz#*Zr?
z$_q`k{!izs3a{P3x$71p@J;G{_R=o+sZasN%bxlUbJm6D1aA8RK9Bcm1$BW{VCQtJ
zhMyLr!X`%GtQm?E4C-p9lOSmZt2T<28|p>)7?YQ5VDy6JDPmajc^dpL5(K{JWjv~5
zvZ$#gKro$?T<R{UdE67+B^>t*?Ka=W@VwQ|6>dC49mm_Ac#ahk*M@J93jG%DHt=U?
zs@>dXRP_P*aiPmNmvs3O_Iw6tl~bHp$5yA+dmvhVa+*8{zoGg<oea-+GUH3FS07=9
ztpHNXRa9)$!_+DU@qN(sc?_gLrEF|tJC;7K++Y<K%+zO7=2hm?pYKg*ME=}Qgytr+
zfsVOZ(UkQ&+m9fCs=5h<`aI76WH1D?(Hqp+vpCoY-19|rzgGY1UP`c<u5;Q0jb8V|
ze9Uz>jjz)?A{uVZe;+*m``~U~8s1)lT*Pm+oD%_vjmvXH83MO>&9k(NlZ=L;?zmqg
zNM!`@vORrNYmPmtg_?mqgVS4CwnkpVDbPvnUCrDB`uda|1eZ%yXo{BV2=0>WnxM#}
z`<qQD25bdUC`26r#oWr-chAEO2Zzf$mLGt=-Qd#|iu+AD9+$22VI>pcjY*aB)B|f1
zwV%hUpFBbGb_$k&NwP}H8W(Wyxq~!#U-F#UUMrEj=rn0u!1H30!=p?xqE2!f>6L!X
z-FdSL7*#pMA2{enWdWu438$4QL!sn0{o1E#NR!g(f-@mAEf|<_*W8)&K0wT29zcKl
zyp=eruvuMMp8zGF@oHT!;p2yB+#58hH3ZX;W@uEZJZcw6>Oe3*PS2TxnWuu;ZNbAs
zDl5BCX2fpK%epQIy-w*>1ylQ6){y-#ebIWiP=>Qq2Fe0fv)<<><M9;8MtSi<35O(a
zh)1DZ^l$F!ihk$5+`I&(AIE7^$*vMRjoC}tYg9cX-|<R7wB5l^!gQ{`z;MGb@)Cc1
z%4Xbqj^J!(fFz+cp-ra5VR7T&mPo}HMOmMZ=N~>};k$(MIR7%jGNpEJ;3W3nVgI1<
zXua8A>P2z#fYaL9siq1CP@EL+H~(f^^ZqUn9M#_;Mx8Hpn%1GH4lF*uYPBR*kEtbI
zC&8a$)la!7V}rva)VQ~+zUst=A;7zV)5V|kY{Nn5nq;%$KNhFP(f2sD`XBQ~0^f&5
zO=i{&UN#CrQ||Gl4d3%OLJT>F^b+Q5QA^G*4ru0PrV!<(?2D1J_Y1CmFh+o%`5O%f
z$?HDOx2Jig5!nD3hh|m?o=Vz;f$L|$B9M>1iYu6hMQj-NKSPin?!ckPt$%_s&Wd2L
ze#^(By%mjZ-70Z$R+K&tOrF~y7eHJ2Q#dp7*cH4er>Y&!3M`i-MsO5Mu;MefqQ3!$
zC8wN6z^;RTc62VN;>s$epC$D3^I;g*fT*O{b?HUeB?XnMC#y=Xd*6qSnV=nx&n#u>
zGOYRQ#ODbEBNFyc+u3=l7xf9V#Ca$OgH&RJwz3gxY@mDDo0ehDSVY%0ZL26w+5-Zb
zlHJYJ@=|UXO6|4!9Gj)EZuPO5P10!dQJhU$o)oGCh$L`U9#wW%Ek*}kY!misv|~-Y
zjuU>LwJw8e;g?WjhqkyeD<*bAj`}w$ZKv1199+b4G1wlhKElO(jXnO@R8#R_;QpfA
z_l8|TSosc>oql3@XtOpWfnm^c87_N6R>4(P+-t;V8~V%rTD8SZSnbawu-X?wkY=^$
zKdF$$!dHxCf$O+Z-!3qnUcOsBET-Q|Fu0TuiVVEYYw^uQC|Q6Atl5(2@^_dwa1ozt
z=vQj5Ld9bLXg$f?CA<U`Zz&Hk|CfXbZ=tf+@8|8-PsvUA=MaYbc5xHmA$mJGe3I4?
z44%mBV($aety;T&*6D+9STQJ4cWdV>pc6PAA78dUt{)VREz;`qyzBItYh$YF`aUF*
z?fS|^%i1w$%<Eqbb=lbjQ*GJP9$JtR1r?x4ijktpmVenOxI&7l!LvAYD{RXUp>Plz
z#$xLj92i-qQuaT^zGj$<_}+HCZl$6ex)kz5K$5rMVSaC!6KgN2MRQfi)2m%i5&nH|
zb>VvA`G!D;20<axDH*|RoK3u;rx#0HU^+5H6Ub6Sdx@bp8N`}cc=aDy*P7@kUvzkT
zq-V}r6m;OXvmwmS5Nn!jO6xd<A#0~^7gC}p`)bDpsS9F!)q}aasb<py%~EpePj$P`
zo(m#Vf^GKB{G6rN*zT|-XVjfpNy4e;LtRZ>RYx({<BOSkLL7O!SL;cm2}&kBKI*ad
zcWBtyKS#G7XC)!)y*t|i19?@oV7IWbQ?Dd#sb7`zbo4WT$9aLehYOnR!wfQ%YjmlP
z4c`&zWpaOonAOZqE~uZ@Q+mO7ASkc;`rDrqv9B=sRqqf}`vp*7PeiRL4AfdZ#bxIq
z$7@HxnFxHxXuWz+(A!6fG3w@gX*=KIbPH}}8L_aK>tppZ>rF3Ncg2<-^ZL<>iU_rb
z<9>9OQnS0OQp`tsm%?pR0RdLfka<dQcIi#o`H|v<QW=fM``PY(6lx^5^XA1mPD(k5
zXV2v>WtE*3Z(jFsC8rUYWa6q(%ZsH>=~|0+HZ-n6Ma|$X_GOh~+$>oZN6i8-xgsOx
zf?o=*zHPl$EP#1=MJN9<?*g_g1u~m5B;PX)!Cl>uLA}qXZm1W;I%!vz{zL53JVkJ_
zWg7rRJ%;VV${9*Y?(R4mqi*1-)STxa0}auJWP-SlFB*d`8v5ZvgJzV3vd+wp9Iwhi
za;2IBQ}Jb~Oe;?VNdZI6>A`x=&Epd%420{-{0aiFo?qmy@86;48kBQC^&;pb4a@g1
zi6E3oObTi>ddLqu(4S*-?Jc!v)84a1Oqv^*cltf<NDQ?*R}KjZJUw=b1IvKYDNK?d
z_i_1Ljh^BK+S$1*-V%PviAPZ80Gq;*J#T?_@$ANn0^@_)+y!6s`?FeY<_`>qV3ivP
z`UO0MSu`%typ7!SpVht~y63u{A5$6p;+bS_ppA&%Ee&t){$mET-cH#%o2MtPPU%V*
zu(IReN`<SZO!%A&?YC1WaH15Qs8ppjFv+Dg8i5rzS?-iLAq>mf_tS7~zvK{Utl}eS
z5AfW}Qa}3oaQ7sX{K%@l{ZNU)+$1;UvmwJqO2GPj(Sbn6pd*6q-QHJq;Lgn`XW6Gj
z%Uk*1XeK`rz~3iE6V*6i*0>$idReFeS>pTz2b$De*!&!Zqx+&0?*^ZzQGsy3`0Ed+
z1GX0a0Vy0mdi#iI&GEwAk^2*|(8S}S1tQkW@-K0R&s=zfff+@_cKnP)AQ3aZ_7kB0
zS(jo^iv(zS5udqw?P!Pp!%@P@z`MC&TJ+4Yz{cM2%F5Cb;Nw<XUlxv4vSCskJB|`L
zhJpL|*zZ4DVo_mQw0M*M^<$A1+)haq|F`yXVFF)_d_?fgQ(NY1EmenUs6*=%zI{)&
z!{SlC<*-0oP$ypTnyyPSf#G%PIlgAu2yZ-@Oxooe3?{03l*KBRy^{A2u4G)5jtjtl
zt8C^qnj^fW;nw$#-YjhQwlsxxSn3aP+$)~4-J{d*2VXwS6W1qwOV?AQSVMu6M=ltR
zm=SY%6x|(F_@Xcsc1aSem2NPD4)e>qt(O9*CJOt6HMynh1s!HAu?1KIISn^g?>4Ax
zXE-MoJn!ceyw&5hP;Qi9$`a0}{9#u7#a+dyK`L}6y?@BpoCxKR@6=kd->LZ{LJPb1
zjqS{=!jxGjUQK5R4OoYDm&8vmxVfl$;1|#zbQhC+y|}(KpL)n(sL?@wdoXMJ!DwdH
zB+W!5$%1<rTS8&ewj%A+^IRpzD-+@U=EFUg2g2G_N(WOTcqkTJidg;qzbd@rqWJ!~
zLM6-X^`iNzfVq;HU5%_jafs2f-12Ab-e~&82it-=1)~|azlH?&D3%Bh6iTe5Jj>;h
zj*Hdc4Cf2r<-R5yKpJ8)^ol(@EkUM)I<C{zl<h&V8RbHdJ*dsaRT7}3Hnmq@^bAs6
z)-*4=XKE?a>z=be_-;_v<^90eiRG#%`~UK`|Mtg*`xvO>jdNYESAYLS5vC_V@wuPI
zMK)e@KKJ!b1`)O1BQrKj&tsSsJf16+qYhTKMdvY=2}0F0U{ymBHnfR@9p>?=^DsQ(
zpR4NkKlhKd=BY~c7&o8rmOr=XujR)+iXS>h?sG;5w9<bX0XtGI<6#7srb{<c%kz4!
zydNbT)o1b+VIT-)5OKy-j(&7dLUr@s*6zPuMWllUu5kYuLlxZX&vl+ykEIO~?v5b*
z9wDmC)H*lcm0}5G#$xw_?*jm=GDZr>sBK$7a29|Ncmp@19Q3VR^$@9bizyJXwtxZn
zy{M9qD6wN0IX4ptd*Ag^Ni$P-zIZIU)4VYTxw~$~9uRWtJ1Y?ZCpbYZ9Kq>v092*a
zD>|@#Z?S*7&A;xgi2FELTnf4uQT*DAo)owAz;)gc^sxG&<iJN0i85~GtL5pIbloBw
z_~>~;)>+5KcFb*O0ETMKgRe_&Uu&Xs1ZIflD<QXvWQ|G!3|vzVjGX+@ViLtzY)|~-
z2k!E_V_?$|IkO;&-Ub;#Luj);LqEzm#_%6*==Y22?8d~!DA`==a{3)7Tu41%c^nSF
z3^yxG@}tEO@B(Fk2GCSE5btsTRZBnMfre(|@rgd!O<)qpb<R>;ME?CkOilf&mXa8A
zR1yC1<(|Q}8)4gD5e=tRBIiF{KN-3H{Kqbf-`D@YA?v^W`ePIoX5IP!@1Ori47@6L
zj9L#HhvK&HKYsMzuJM090GB-V|IbU>(|P(*xO@^tahrA~2D{!a&D%$eg@D%4kH{ZF
z26DgDitU^??EI(4%uV{N=W_n%Q#q16MosnN*C8{XgbA3I6NvrP62Q0$AunnN=mgW^
z=hq=yi+yYX`=`#5$QREY?C+E!OAIiN@tHGxi?aZhq~xDZe6j@K#o)XM5u3H~>Pv^;
zp^O~518mgY6@cE9){Zq|Qyuf{rRyB&ZS75!%4$4~0U(g5bCg1rUwS|bt_3kC>E`;=
zYKI%Wehc2c;jq)0zcylL4we8wND&PC5Y(}jVANRxWe>B<&OQJeZhKS!V5(7Nxi&TY
z4mgN`nT}GAZM?$<h;|ks{;s308@CZ-6AKWGEaq(IO{u--QG><Mk01a6)K&odrE+%X
z9O%Mv_v@WNRKLvKpL*ev+(T{F?RCKyS+9RFe}56r3Gjf+aE(GlT~OEKzyS?~h$T|0
zR0Ve${Ap{5@X$gSj5~*F`xOSS%Zlj!Kysw{lWR!Xg{WE~TwMw3#FYckG5jh02-Hx;
zEfpk)xk1d^hM1;uwNHbGFx_dm%;s{(W9IC9FL;G6jsO;P++Ad=n5W?f)Yud_5-+QL
z-k*M^Fw9o9si*}nz})4j=+7_o<<}QVK)%rK;r=S(5d0RJG+jk`?m+tOEwvi!T|`}d
z`hWJ+?_dP4$DR{TsCNU8JAWrAZ!vYj=9Li>VM>_*r@ZPmKz@x?WaRs@)ba`|D+7Oe
z$O4nKA4FioAH9>^Km=~JV4B1yfqElf_br=oeHx%lYDym<?<FXgp3Voc)xOQMK2+@Q
zapyBOQuy;?rUEUyI(liW0cJaNo_`4*Xz|l$y}wVc|C2-EJ{IA*3?CxHIe+`fU*G1s
z=r|y(-h!55El>+DY5|LwG$VOP_;ZZt$6I=v%1KJJEKa8|G%#92SnzVG+1>F)P`vKJ
z{8fK=8lJ((tDWjs<otEMeMTX6*4(bUYb1v|h!+%$QtP0od&NEB0d=$L5Ri~9Vfk5n
z=Jr_xbQ-harKUFj%dwvjCywE7(ibm5U)wR9K{O2-2@MF``KRtCoP0${Q2)NQc)x|^
zZ4g50;07VuI@_MZIanIc4ixTP5S7<=1o_|v5)(DuJw8p86EwPFE75765l3U@+%f~c
zthPN+=d>gZ><~U02LFLCpA2BMXQp&VKc*DA6^`X4&iM^NG>ucRBBOKwL6Xkl3`Gd^
zmJafMYC--4;KFcb2zn~oN?8bU7V5x8#K$AB<Qm8kX&oIN+|~wIHvuqAHh!w}!W539
z^gYs5Ga{h$_+hiz&d49ySby`$sR1~gLie@9GD7=deiDyp)f7+`W+BQpARnYZ1ggRY
zelpLb;XDnK93+p+5wk-WBG0QKYcMoz%}9@J26Sw0{|>;9bCOHjA)2vX!}p+Xn}AU;
z6@|lH?yUC<+DX-*f4CZUTDDk=LNxsT)(^01A!u0;kd8fC{cz-N`=I4&>C2qLW~Q`7
zle+;4lxMDHt_>BQ^I?zL`HUd(z$i##2$ttivA@aw=lI@Mgx7KzuS~rXM9MjWDwfP3
zN*04vhoqJgnrF~7rSKt-p_(z_o&=|za>*X_GKT;V^$=2s!i=0kt>sIa`v|W(z+9~2
zQd_%1Pmi2%x8;Lb?F2N}s~Htl^z(*C;JW3vT?O;l2Cfvv1BCnu?@B3cXrW5BB+2<r
zm5T1e;#u%mh)pY$c;yPPe?-oxuK1pLp!<{g(#PA~pu{?AhKE7KLGf%wSkTfv2-G$Y
zxzR9Ros7*mcnB#?(4v=n9Ekg>*;u_XIV}++S^T~H^xrFye>q&*@I+Ak<iZKRHiJDj
z#`k%)=Of70X2k41{m8%AWoJdFSIB=JVE)gU9m7P-q5rPnn&e9-&?#x%6M2qJTQ%bz
zIrx4m_@#y%0!~j#;0PX0in<(Sl!`AV3JG;1VEn)hm`>v?Mjc8(ZaKntlf0SSytoJE
z@Dd!}x%A}Fsv+dNetiuY+Dec}r);`oF+*$r6CLwT83^}^d)7mMHvQ(39<t;tT5dKK
zs$~-&Bk`;eV2eZ&;`lr=-Zc8pF-=MDVhtb^DSn_ESpfPWZ?Fb*;Odbn4T+iNq_F=@
zx;mu;{M$0wbe>+A*#i}HemB7LCD}<LDU&`s;^0&7W+kRLepWmH#|b6Uncg7~cv-fp
z3(IDJA_#Jj(o1<ykKqT$<@@BdoB=)vi3DW4)`>_kzl@iPY3`CYnCLDCW`Z{7&6mD)
z!*(H%2$@VT--A9_gH7jKNPF|8=aV5P7ys(&Pyd45CaP23&IW#+&EKK^E=@aaFmLjb
z3ck1ys2X!zD5<L~oI@UAwj3w1L_3auygMw@#`n^nu7L%|0I(^HK%qqS!dt&Bpo!s1
zu(lPI5565SZd2a1%^AV(dhqZ;uGw;5nfVB4`#qSd?5WuJL>m&}ka6*y3=@+<u$g7g
zk3Lrlc0G~d4@Iw}<qlzq-C8f5ece!@XyzD9)D<FQnZ<DLU=$L8lzb(FAJ7G2%e|;6
zMmPT1$xIN@rnU2_jqQV(ikNHKHmkoEVxrwptQi!kJcsbmK(qAi0T2@Wlee5wJ|1;Z
zym3%FA`$r@Nmof^*rt$kPqU8!lH7nK&p@cYwqGS&7p98^j8e*!Yq#*NPN6qrPK759
zOu9yM54Ki4j?lS;7q}?OBg3)&u&C?t){YWoy?&i`P@I&pGwwESjqB=q1<rdNDWZ_c
zaiyy6w>iq$CxOPc=(t?58J%dn#dVze@|n*uth`u<>%1=>v58d+9$@ac5)Ef5S+9#)
z-?g0pG&45hUEUA!2%ke<#J`JXhAv=PU*<SX-0m>U3rV!4z&g`i9Sm=3U`S1_9ZjOm
zsIYDEoOneiyH1Wp+X?SH-+ZFuy@_Pc-IwUu02aiLsnEsjqOapUMKMpB1~E4c&}DiO
z7~mBs>4n|vRW^TjN4xVP5r`O%*Q8iOcT&wF68$8=7{OwW)4Uz8c{fgR;MLG0adiUf
zArz}$VCC?JUYv_|E8!{}JKNQ@I+Agrx64U5%7~kJSy>&i5wy3&km@B?3z*wtv+@R6
z$_syC*byZf7E^KkU{R)MR&t(L6La9nY})tu)^qFDo(8)>2_lAKx<!`tJX4zuaZAIt
zS?)Wvy79?3)=-JnJ0}-(3D{C=Td&?JV5-t`b6^X$S$We@_}oG<ZaB_pz9&m<I&+am
zNnxY5UcXu*r21rPwIFrF`=*31x@{c9N~dzfG?~pbK9DXOK2qmw48BBmy;R$j@D3z?
ziCnrY*BgsgDu~<jp*A{A?01S-v2?+V(=k$6wT|3@18tobX)_WPZ|Z*8+`IQ=4jvq?
zvfrAA4d2=o`MFPXX9vp>&z>ZXKTVA7i)icS+K@IHjUdwMPtcWTS!d6!46v1&<b*Wo
zPEB@<Q&86ev4W_NI431mI8zCo-5U=%E^}#u$u(n-@0Z3!ciF+yvhYSGS9`sGb-ihI
zRd7SGoMn4DO+(7S!RyzpIK+vPql0Xl`*a!%MIp_qMkiP?gV5rR<Y9|64YV17w0|g9
z?a7ZHWVz!1{)C|4M?~R9I9pcxi(FW=lvj<W&-E1_7Fe5{)m8N;{X@E&s#GOz2B0Hl
z??<)MW7^=@RSVWg9QVV-TbQP%Ownr=lem?nD0$}i=hdF_L6tyUcIP|Eas?at!P2U)
zDdR5T;N+0jD}&&qhxO`h%fZ!z1ux&VAkphPLQ&=nlbw(`j#FEuNHO2_X7cnxrv_9O
zii-G0(;2N>_bSN8x{V5Ap2EQ)qYJX@P*lk8Pftpq#tRZZ?`%LY5aW&KatMch8(LtN
zP{H>*6My^VEwN<sfgp5c>*zzftJM+B0Xg2+T~w2c=oSR=NlfNK(aM2d>^qau7cA`%
zRi<P;DU9dAGHtCmIm{GH0s*r0rZPMH5Ax7Yu~(Yd&x9D0#Ygg?+mt`hk=j{>O@P1Z
z8PXjKe6|5EZ!c%jkM?`FmNMVmV5*MvrM5DBYxb%^ed8`ox>-id4NYa?%;LNkXk@w^
znm-`8c*<X(m^(EbEf<tr0aT9brfim02`+E87?UGH6%AZ-Jew*Kx0Er&+M<%f(HKFF
zS8R$KxZ+{cD3k0cV-*#@Fz%cExd0eOOyWzw{RlPHeYpc!GZk6$<(y03muGaf-j+@}
z;7vW(;)wZ|1NiTMl;?$hg%53-=wGFWNFV;Z8TB|qF1b}1ihGgI-ZUE2L&94MQv@)j
zJ86z<rO_RKQmMP}gj@8ZOvE9ZFdhc?H46Mg>7YUYjh@D`XQIcFrYrll7TNV#r{twC
zbr6OCR!?SH9Xi6BY~iP<uiTB9h^`a*e8T9JB(W#H71{G`^T4isHmQK~>QWV@aXlCQ
zoFGKxo0APP8Gd}-4lP=*1i+?jdcH}%)wr+NX-)?73jRYs4OOZb_q*S<lXaa>Mli~?
z7E0?T(~*43PoXc6FOkFRYeOA3mc@5yZc#6Gmvx2xym$)dqPC47Q!?8EA1NW+x8#=Y
zK&W>izK+UUoybTYHXNL+&jA^D@0iSj)5+-9lPmd~Y4vEtqi#317#4qNC13V)S2j0K
zLx1Zmg``c=(=P2=bCigdE=32NyJq!X-jBbrGKM7KPd&O7wx*0O?;R4Du<JgBbxBa!
zH5WJEZ{5`KNO~}aT+7+rbg7uAeS*SW(850Vv8JvevvcZI0|ESCa_aw3g#12P46!y9
zw+_3WWBxjfL~y9b^))RaLFe8<nXYiXnb6&nw6e2FD95e+k>dVts}i7mh<Nq@Xcg;I
zWL_(Ui58_5u+>t_J5$=+h#6zB$yx<FqOsiuhwom~HqxUw)1!?t1(wLeu#qH5Vx@kB
z%C;h%lfF}y*eqb5Q7$Qhu*l3^(&t6nf)JXGQC4n6CNe;@k;^J>tCAi7xiQklo~pb^
zLWfn+*&46iIfLp#^a3_9*qV8?)}Ga>7zVK(1J)mf)M8J?hopqV1s_>Gnw@yqCP)?&
zc)R7{JH2M@FQydj{o1_7td5*GvKN<4Jx>zlL#iKxjny)$qw0C@0yx;8YZE*ifDt(E
z&_YdbKN~Us;N#;bT5E@6L6CFj+Y2|hfi2JmF$8;xC}%%?JL+f7jAMoSJ*?W828cj^
z(F~;biP`grO)c+>m4z-7XcrcuC3LTI3TvY*SkD9+F4yE|w&61r*iR&t?-XSk2bK*C
zn5WM+APgpoiXkA^_-5*a9yE_;a+ufU2E?XlN(y|*ah|uZ11sph!hXu-lZ6t*S7HLy
z+I-#;?hmw*U(KW>BvY$BM_dx?+g%LxSNkD&cdNRIx?bNAT(ToSlJODr&Ufy&-@sks
zTAqshn!(@!5?ix0GnpSwhJA;e>UF<6@~$VYH<kQ#0DOr)^5|~f5Ibsu*S5kze7~VQ
zO;ElhNl*<v1~Wi>51u=aHI}0-Ar{Ouas3YUe-8vmNq!b$!KK<r&0lL!=T@SM&tv=@
z=V&D|u!B@_#D7~>TEPgS2I?18I7LOCRYOW}sr~_eyNWD6N!bn;;>+#4J6)KJe`Rao
zb0cC3SNKDdjZWS=&v`aUZ^*5#*7sIYxNmgHXcfBj%IfIrB>)%YBXfADtpx4ZycJyE
ze5ZF0Rc{asm<gZ?YmUvRt+kC6w^-5dQq^@fj)Rh?P3(@O!c%d1GvB(PE!6RmK)r|e
z!nbEH)l1h%Vp-AchEs`4%RyVjW(Zmsw=E+XOH&l#X8lF!uYU_oUp_-c>AF_U2)@D6
z>^)r>Z+#y7kIP_uCR9~VT3DeMeeJT8>38Gngk@V5X{>Jq&yMN?q~^U)=Ulc@R$`hx
zazgi=sK{|sdn8viIW0n7gwq^MXAkW(a$fdaXVN1XX#+NYaiE+{%SfE*h+@~S*#m@$
zU_fhT)ki^<SoE3TMFmZidY#ktcSlA;BVRYf&><Fb=u~$~Q=HD?ZHhMMHo+Y;()#*=
z_rx3&1AOIw>>E%+-@vd;k@>H;n%IIW%EZklQh1X1b;<#Z`z)X!uBX)_WH!Hr+WoQ7
zF$84Qi#|3KnT5?-hmT~v3bRe+C;9x}Bb}Ec!t5zIig;bubqTT?wnlAto#$%<Xt<q+
z!*K767Hu-<dsl`?)6uf|0YJ~<-RLf9hL~tPXx=iqs6~*->wcInmLJ8TPgRcpX2c6T
z=tpI^@>B$Qj}eG1)FSTBmw^~(L1A*;1%KIe)fs}bpm-XDC7UkJmzdQ(VU=t$nDkb*
zmqr+hQ{bh1W4ZD?8#5L0e&M>nHDWG4uUtot4D^!KcLCQBq}y~oH<k=PrCICEY-Ym&
z)bANz@z@uPXpt0*Zq;gYUf&)W32-^jMu_BArEVFdMXG|{{=000u3TJ<UfPc%AXGLb
zl(6;tT_kNs>?Og*x!lnjt5wk(eF<vJ^n(pRVO(>*dFM5xQaY4#|6$ks?rC*;V*Z$-
z#rj9vU;Lx+<f)?tT!HJ>N>#QU>;)7!=d_gs^WJ|qV_8k|WPb7J_+4vH$&~}3g6>qG
zq(Xds+N)j}A^53v!4-}xO<&hi&=0X!mUCx!oEW?NZ^7?xa7qwcm<YUwvvz#K>`CCU
zC)hTm4`Z>hy|72Uus@JyPC+vg3#k|o+gmahT^wiLu5p9XaRc9BKJBx3+Gj}wBJ9@K
ztILgJbHvD@GZZ#XQRsy(YIXCSjhW3sW42bkYPYY}U|H0C**(DZYEg$GfWK^~U>u{a
z-(#T6mn^sceJpcck{;*A6bRjixFL$~neU|?7bjc#C7;>cPi@!b<O+_x2PdK(^$3K;
z*!Yxt(C4(qNw~RKjwZ_K$I1D?dOg?a!@!L}3t=Yi>~=ENkLLm9soqJdeN_;2pl#TC
z+|{p^Co(%RE}Kc5`LF|7-AQ7brw~rwnT=?E82+sNvV#?qg<0$8E{fp##HQ)&t_^GW
zruqG})N+KIDiPPLr4>jLBC~p#@?WQ-e|v!e^i*tgbidoKF1Vhmgcw4wdaK2XPVHC_
z4yP!s_&K4V;4x3OPUOcvE*N2Kks_4+NX6_H>J2xgO&y*>;?zV`mgCJD?wxevZ{iO~
zm2g%s4s$BibEwSVr(6Q#^6IMdkyS0VJ*RdKNd)Q)L(5V%@i#!4q2x9m4mKIm%g&@h
z4aE6=eP-~Ho6zajTL<WcQB(_FRyCr+YhZo1gm0P7>HYX?3i^4V!KEE}61wQ;Yf!+C
zwJy3(I770E)kvginlwJzSpx_Ktpdvx`m2L^;)#%}YM~elrT||^u{<O;r@EYuPCE+%
z5D(v6{_;?K{>u`1<S{u6^_^yof`hMdzE^XYFfAN|xragmp5<*k^>)J-J_5n+%-z)4
zkI*8g97gMPCa7O^@WGWw*MJ$)#lA`euxhYTv@%b5%Jj!u(T{&+q(f_}36zi)n5~5l
zsrVdda&i&jpEFT{^Cu_V{P(l|52KZN5@-qYKV>fd)sN+id|Vm7AeM88Aj5V!)VxCI
z*lxHmei?Rjb}HjGo;iMt(r2cg#aXuzQF*w1sX1I-Y=)kE9|E0Si_>lwJJ28+z85CU
zet8O>WVblR!J&+NEKZq5-?MUn1(culLE^yQ3l~f83x%MX5h>3@Z4HmD^H{9NG0VLd
zS;^gWA_mg}LD_}Qr5EKZbeLKohCr%7X{3qM&CK#!?<A4orJ`;^975<uWf-=mpraZ%
zN#k#y3uCp-WhYkj=V+O;jH?zJY?1gMpOJ9TWP?1!bA<k<rRv)-QtNpu?>>_qpd(7w
z)#vo@rI^|P&m4#1a`BS0&YNe2vd#$=(7^NZ6=z|lweq1b;d5Eq2z_b16BcN)7+NOj
zn0F(RsPa=_%#sTb`>`lkc}qTvCC;|vU|kJ&R)6sDt9yxuG-Y999dqI+Kd3GG0$C-e
z_0HRwz{-av?xrK5^Rmek&=wYOik>QT^ngBD*X;N0WimvpVt6al@4dY*m(oF~lN|_2
z_$T0W@J<C?yK5UO9838vod<|8r*R0#`cmwV;)=6O=H6J`Ot?a<ZC$+v9FwPY=#<`8
zP{;N~V2%OD=t*EYbob>65bRZ{z+SKXx%&sNzo|a!pG@zkdXN6R`_aLUELYPurPH`)
zKJ>IDSIgn}(urASI9aU9VD3}NX!ekyuZhEo@@@j#BxBwyO)UYsN4!TrHd?kHR1Jw?
zoOY(}p?I^Uv7fp<N=L{+(Y7LNCVMxAa*%8$QdRd}@#WO;?X%8oFUm9>cN~M&pnjq~
zxYPcpgnJ6Lsn~N!IVSx(Y!tB{CwA$Fy+@87PZrq+c?)Q;n{eWgY4m1!o;W=Sg_H7&
z*VsEo#w!=IS;nf)$Ds8&k9aMrT(+1-HbOmSy;wHbGdb?Al3q6rU8rK!WG5bv#tv+i
zP@C25HMmGT*N=4BS-q~s@2rlEX`}RjA?}f(8$YkxO>{(M$;!@2Ob{JdZ1Cy}QSsPG
zp@M8|t;sz^0Z;XA3h~z`Xw7R!yZ2Om&^p(4qD@$}8C%r&nW}~Wz=!d6H)!w*5vx5r
z@q=h=jaDDlS!%(yTKcG-fhCFOQ`9R6(Y>6cLt7rd(GcS2v#I$B`!)&Y0!moyk>#hW
zc~=Iu!K8Y)CBB@8extKvHksaTBIMa6{$9ig3E#ZHc0#uDRj!!fq%P|+^!$PhS@&f?
zBp&;D+USI+upOJ`tZqklNF3Dk+Y8CJ>}+I;7zV~b?E`OrLNwEK&ZXT)IlZ7c-DI&-
z8=sb#PXbNCuJ^b+v~@T|XX%441{ZNtgJW|81fjUS0vO2m0|1UNP8F8;EQs3bvo^Fj
zQtu2(*w1=t+9|{%&Y$Uy4ytA*?qwel6xJQP5Q8)ak&PCgw99d!mgRc9?Y9SHBvXyz
zW%_g;X!F^&+1scW*%Vk$Xlr}uxDtZxF3*F{kT8&0ww^3D-WSDl>{%l387&3~;;na&
zCJHhRq=$;{xb!=bahXNvKnKlcx$}oLhpBcr{vpzj|7;-~R&pI}Uyy*vZ_ZL{PPrtx
z$uEXQm-(^gmID_f0#UWFdHH$uGwFnv*@5AQ=$tCq#oX;_`>@+)A&$pQ_}=B)4a^`-
z|L>N`j^>noaQRGV18L#1`{VxB0zkjQZfzAqHQG2-mz9@w5os7b?+^Y*XJ`jS;vx26
ztZTI#29gZH{v_|%*$-K5k`l}B-`Vi+(`VJTGQ@_ieVk0wpCBGJ03~7}ze4ysF28z=
zUPjCrUyl%@rEc5d2v{~tzjOq#j2NNO(x}Y>Cm8FdhiuDy&v%7En<SpZH;a-Wp>0`v
zs&)0SFXxLH;^@NA5a-$i&w}Ms6_L*g{8LU&>fzDGcZi=$M_*Qm4?|-G(bQT9KGV)R
zTFL34OL#+{aT_Kq+7A`nuJ6bf7v$gzmG#B%PUs|`_YbE-%#JNFNy{U9o@-1yNy*W0
zyzx~pf<%?&%mUqP1j${4OIVFBK+=P8v?YM<^KHkDN88XX6(@^i6yKhuc%Q7U;&pY$
znovXPM@&@R9U3Jp40=?U^-WPC9D+*{X)DegXO!r<n*#2Q*y+3Gp~VZKgr*Bn+H^0L
zPNo%n3o`0M$$_k-9UG(f#4i=P4fH<25hAG<pvdL@W|!x9*)!K$2*fIx+p3p-Sl`uq
zuCKe_?fmt%ct|nbK$?Yum-5bFB=g~$w1ivD*ZGC<<E+7eK|g=$<yL5BS2|s1Pi2HD
z*Hwdusqv8s3wO-Z(OrQwl>W4!Y%;@Ggn!RJ%J?4sQPX>4LxYGr5(6sND4w+|>=2pZ
z`^scz?WH^D(r{sYa`;kt*E=5{Lnbkei%wdi_2<9{S)g*gIy+&+ec9sK*$!$fbYHr^
zXRWtgUVeqp76-(OQw3dMKpo;WZ#ENK&VYBnaHVQ@GSFs3m9ps81GXqjnW8KQ1?j1b
zF(Uj23>nbjFgaSdvLRNITAsHhk74b9yZGsqyN1bQ1}py80uJcU=0yZRQsxkj^QV3~
zt_Vu~tfbkmIk!Ei#4XePFlAeq@=#?=yH6+Z$)!3=IZBA`{z(*tNfDi1x1F=`=J6X>
z?{qfK7KPFt2t|fJViQbxpJA~&T9vXZ?~s+(tfuJkB`-Ow@|m=pnrs-!HRfz-VFqe%
zaa5)fjacOE1`(Kj^3f0qQcNcCg>xti*XelRTYl-rDUfPV-je>rKr|_j9awj(=hVR!
z{||$ShFoO09+BbdQjWPhS%OVYQW0v)&Q)1{e48uN2j9w49a~BWCIWZquJ8++70YhD
zt$I(l$_q`cWp9Ezexe=0-_E|M1hM3$AECGDxaoxo#kO>&u9RGg!AED$TryLW%!@Je
zwUFgzCDODqGAz8>mM6olkg7v3M;eo&*%4l>IM-B3#Iq%4r)sIvJ(wh99KOQC-=T7!
z^WalSMvVEEh9OfzlS`VzP0>#>=DCBp#hQ&~S2`|dMOw`!%)VJtJm`$gDhSsxJ5u{#
ze>X!hE`S85ia~?<>wt$#^AGu>dNq%4__3R^eVSxOYMMOe(hwuAt|p!RL-XKoaErgX
zSXX655-4UK*g=P?1mr~$lNn2m`K4<AaOD2RkR`yNkT>BR?}*R8fCwH79B-yg6ure%
zbnSPY^@nAIYibGkq!#*zso=lfcP73(iInppk|oN{z~qN^AKaHLAR3i|6UI*kv^dbB
zlHIxSYx(`I@BU+7PBZYB=*M1CHcjk*xe?C_lsB%yUl^?fr3A%1-71UN2k8ZfF)ldS
zD*hQk_WOFBnh0R3*QrbMUsL8Ux5uH&TTJP`OTAP6|2TW=xG49o?OPEPK}8WnMTQWN
zMiG!0TBN%XkPxLqau@+o5d{fBT9A<LZcz|W>F(}EQptCnZuh>g{akx{KhOJ*A5muJ
zob&wETI+ip>o|Sb2S`Gf4q)&824VQ?24;u=5o7-HYzB}&|8V^N%?@xFe*!RuFdt9Z
z|Gbtzuk;`9%wHcMY!1`8=i>NH=KuY%|JOT}z&)5Uqy`4=w^#QMSJHA3xqY=fAc*^i
z|NXBI`rG>pvmJi=s6_Mg|Bw5We-vJtaZMgXUVnSb|C>7=AOp{b?HtE{-lvwl<3CZ5
zdkyC!VaK(En)^a;lx5BkSTdMzg4tUSD2DDMWV=s6^wj@43heRV@^nR_oPPag=@}1b
z7}16RT0L=iWI8RUcqam^I`|^m+VOW`h9G9XeZV$d@|1_9LXI3)rz%4CFx)~=h76z>
z1v2{OHj?6}W*|nlgoaRYNXMf>Kq$C7Z=u*iJ|M1mNQo99E|ikBEEYSulyGnPUL>7X
z#gAPeg<UaE>Pl2Ti@%TGl@HGLXrrOuJ(@vex-D^`@{j&;cU!J{K!ni`5t=@f-Hmk}
zF4;9)+l2hk&DJllA)8&k^)h)LLm(ETjsu0gAXtX<T5AE{;h*o$dfN!W%q9pj`VpH%
z5QF&$Kq!G=1D)E&GQsm!cI&WKXTzk8=z4_UT?v|->$PkUHg3X#5n*xf!=uB4@5_Aq
z9E+AUqfS1AZ8AQ46OuVj4_D#W^8u==uk|IruOfR*JgPvcTnlH=7#&90=ttWO)akzs
z4&e+Kwq4TR5B~TC4=Z_nD5?6;Fku2xueic0Mc-Nl@X!6T@;tGF&`ERyEw6>&Av*$z
zy%*8bh}~1aSz3k#_@NdeJkSK)3PP|UeRT_=!RV**2xiOJ$AgxBdiJg22xVF;RN&j;
zik}WdywV%iSB?PQlqpm*<-b$v&s5Z{zh_PX<^vWaTxn;abQMkk2@m?w>!Kf;D7%(8
zP!0Y~l-MEs|5*C|{#X5nbdR7fF$HG(<j?;Q@O$+1A>d?zh=JV64Ww}4XUOvO(aF*A
z*o*iU;EvXM3;JN3iG%>WAAg@jKJSVb-jbD_YV>(J?FcY`vjji6Gy`15HZ5EvVZ77q
zzJIpbnE9cjlR{&kqto)8_(iZ~+rI|_r5k`_9X_|Kqv3mOEq=t~9p7)3fn-UTc;cJ`
zzt))ia~}7EI4|Q`WPG!#=|YkZT}cX*WpMJ>&iNr%6wcenDatgHh#OI`Xw{dI7}=>-
zm6u39B*r|G`c2<$UGYiW?SFf`fLv0;3>>skH2GC@g`v56(E$jWNOTI+UL7`}LemfA
z!OlP(!atD%DzPMBoap%o##0)?HX)<$R+goU{HRW-*1d4=<|{YAoDKsS`eH;glOwRo
z$qgB-en@s$Adb1PNl-EN8Pp+L#;PLarx^$-i~bXiT7cht85=++o%Wi9?FY;bpV5#z
z&={TkhSeKTK<ID-g4GA|$DfoRn3;Xx20CegiId)?a%cmd!1ZT$qQQ^JaHb!`Q5;>t
zt}8I9iV#!7d-d0c5Woh*s9+gTC()U<Qc}6Uk!6W6lrZZ#Z_D*MX!zHGoiS;hurD86
zwUIjv?xI}oxj2Cxl4Ss4IQ{BCLw*-o;3l%c?7k27JMGp?lE5`k*7z8BZJ<0W1xFqo
z8B^NIge8$jo)3V?n)t0unu>9UBHq;fuc&tDwV=-Nf^%L9deM0sC=5G~pKNzFVQYlA
z*;Y^%nk8%k&ejV&mx@C0Vh9pCj-ypS)||WD&;|L2pG{*uj-j5L2X7}}1YwOpj1cen
zu;fbC@NG0e7^+tK_Fha$qJx@wSUKI-ph-wZYcWL%n%<30dP^phai+5k+HTw_V0S5e
zj@d_C5unt3uUTO0O$B5*xRLIA*`EyKf23e&&jDF#WMTEHP&ea%!Zk~>>}z3x(r7Q-
zH1;xJ)D{p%Z$7Ci@c`+l{3k_Mmp#|(l#o?$!}k&|hu<VnZR2jRPOH?oN*E=#v@c@r
zuo_ZY7E;a%LU)w#@z&0@ap-@N4d*ecXDMCQ*+^*EG>SC(&U_0Po`jCZPya=8?d5fH
zsnM~G+eykyh8ZcUH-7y4sHSQ`!?13={_f}>(@Z$@%y%$rwL<WtJ=jN-US7Ty^wbPv
z3N^MDRk`KZuR*63Dfj742aLp955I?225U9`pNYLBg&jcrAR(f!M?w7Q5-ev*5V8aE
zDN4wC7wKv-Ym_gDmie6hl_{+U*>Q*o98}wO5IIkvi4{UjEe1%=0PO6$93nXjcin&S
zi@BF*;;9Q9pP=2bZ6q`il*4%s)u(>w{&NuzLtxH~B7Vu(V9Cw+ZW_evD9@*7<v5il
z7?c{3rpGnnF8k2qJ|AiID*MtF*$e@9r4)Z+7XOqAX@5vEDLKLkWE3#B+i|5Knuf>u
z0W!7|7lpO{lhS59?LI;hlaLvQtK})YD5SL^b$`+LNo1oO=tm_RZkrKd#H(ej!lD8u
znLJ6X^a&>0X{eT%BTQkwR3pWUx3mEP!fMloE)e~O-MlXTZi!Q(lk$ZOouXVIZl<C8
z#48DUtwj{w=ML=^fT{Wb_%)0$(#F!>-Gh~6e4u4lW7tPpdz|)>@{C>Mx0BpGRPVQd
zyL@RUIUn5>3<xK>&?xr{eIB&^sX<Voo@fhxKb|psgzh#hJkRiOp7P+k1Om7%Ae+n&
z?Sqah6+7(iKYT2~qtLjMldriw*o5V7|5%8e_Vt2OlI67DZ1kASqU)y$=uH=whAZ85
zwm9$SJNl%2m)>RMCZ=a1HSNc^&c7^=TV(2=y7lzFp<qwVls0?8mgMV%y<77kH(e3)
z!2I^RmvWaohGMwB!D8~lQ2uxw*NuGZ<Rxe`UPW6gxVuuF=O2>k(<2eND3NjvFWP>}
z*?wa;AX~va<YsK{Tf|4T?`hY)7e6?gH~f$1Pko^W@RiInp7_WR)PBC*P`K!-WG+|w
z`xM|2{O|59bbrcITH<hb#oSl?D?V08dqS_MtHsmroipbb-eOhL`sxe1w3)xxjf_-9
zj~G^5z1I*n68(N`7_5LObV$zbirLy^Jt^5JKFwU*LuCQG`Y5zMIlgrAN--66u`_kC
zQ_9HWHzQ-IJ@RC>lU+(T6Z@`xp$)<%!{Cu+;*zO+5-i52Nv2%ty!u@sb%b)(?#>Oz
zg`R9s3F2+NCqcla&!ceec)6$kR;S|}(im-|DhHZn^mU9@1&&muj#O1<L-N1&{XB1u
zy^^*~&K(<e$B*KHz!L=vEDr^ZIU^pYIlgSpK8rRAE+UH>{Od_M-~6{e3fOr<QJ=4v
zDlHwQYajYe5qtKwy+8Jq?t>^n^<0H5=?gl^WIE4D1|oG|Jn<nntYEu$e5rsn9l&JZ
zl`MIRHJOX`uZ9TBCCAfr!_vc3?DcLWu4)8DSPiiZMcpmOWt@nSu{7=t&WP5ispM@c
zepx#wC9^k?1)B<ctrw}?magS-3UM~8Eo<G6BGs`c9<bBOQ0r1{!ZWj#YvtGZOgtT0
zJ1F2A(nY=@w>DT)(%PhxOs>=M=5}RQ;?pNzOIDb`?~4khU47RnqG*YNeHZCMwkNI8
z(STOy9<S<XyBM8+g%?2}TlI9E!`<F3lt0I~0S${whOTo^SSAt6Sc@p@yVC?m#^OTr
zL3`T=vwPxwjwx;Of{cBcxsBZsG*%dMmBRGQ0tuFqiNKF|k3Ncr+E&PH8VuSkXMe0P
zsQp$7NcUPQJ{!|Xwt$le5_m}{BRJQ|Hw_!?yUD`ey|2m0Rp$Q_fi$PDyj~ZdHtn@3
zxTkK@obs_ksCJl{?a<50FKlp99XXJsE>7nBE`;;KF>XU1`)+m3BJFuWVtCHD*6)rv
zzK_bk2K*VA%^b!Z$dg%73_H1XQ&jl4Gp%U*u?71;n`VKT8vifEWOsu;*%G%WpCrC?
z*VB#ow&YBLf5TD=Gr^@?2NDO?YyR7@j_@n|#Ox_2QOcZHem$8_S41X|wnxvN5J=vo
z7V-QEMM24xg#bkaVFi^^>~^&L(j>Dt|2bXz){*GwF^M1Ru2DWYpDq(~4guOJg7eWL
zP*YCy75Mn_c%$04%G#aZ_0t`GcT=r^K4<gvXQg!-gAnYx`C2kI)0><5RBo2pmkJT_
z%%w#R7U*26G!YD<y{w$zDEpgpUwO4tgoqzYzPoK6n5<D)0*VH_-Ur+-VzEqwOF8V=
zv~2U<r+ri(GjA&QZqSFv%di%xlCYY|rqLZYqciLp25kIXx8J9uszpQ8{4z;2Q<&2C
z<#t<-F6?MYLCqK)DyhEqQQ+gz<F&agoov1-*qbjcr3eqVTUvD$wmUd<Vp<&&+&Y-i
zYgeb0dLBY;<JgOvd0#3xk8XKi@GM@c$dCJ=Xf%a#%k!_>lN?iTq<ry_DdFB$YuvU{
zimUIGQ>U~2<Zcg;%<&(31ulhdavxXWTHYW+=O3k&GJj+tcJQ4l%3IUCZNv7|-%s6t
z3oa-?x1(Q|lW_0XzMMpCvMH7api)Ta6+RUS+KKjKV~EL@T;=}mw9qjkezJ-S1tsBM
z1E%?0o!R<sYtE65B5>AYaU$3A0&d2h&cqUE1kQTwZ4p>3|M(UW>l>_)uHKL>`n3rk
zbF|FNRO|*#7WWO63&bv;)AXb3a&O`ngXm`3n>WymmZY`#i;*6mL((_NqBK3~7+hvz
zhg0O-EE+MdJ6I!wyDwuD8~mJ<Z`IwzXNZmXM1+Y;vi}j+T=u|EeNLuARG27|g16o%
z+^egR^!`_@F_*PbaG$dId)(Wr+MPrEwA;Z~(!m)z*Q}Em=j)TCa<NqRf_L!bu*;?j
z$TQnI-E8Y;nyDrOMu!lBguZyzHW^v4Xs@g5pqH<kLb-w8oF83pp6TU9v^u5EXgV0E
zf5(nWA?&TI{7wXV9rJv;cxqOluYR&3Puc7q<zlj%s+qaK4_z*lF0T+R?(1p&<L}y~
z4gy4m6xy+g7R=4u$P-$gy7aH{-XUB*KVf#Y#e<yCzo;&svprU2iw)KL;aEbm<SQst
z)2>)fi7@LHhODI*N^4fA2Q35UdRpOX>RL5Al7?U@evPasc(7it_`ZmphfhfzO(fJD
z@`@tgJGMS438+FpjVnoR9;gq+FsQ6u%|;@@o*0oQ4K>fFXuP?Xl___hl$0b1^n;%#
zaA5am1eUlO&}dV}exW!-qG55-2RXJU?3SP1O1uZ6t{X(xFAayX*+Iw~iB-M6_1+i(
zs<zc5?Z{~R!2J1&U6%3G<3#65eqV{wk4v8HexBIdviCL6TP+QaBOb?YVWsSvupjVJ
zTiJ9+?k|~M%X*&Dfao?+OPsk*ufVI$@9P3Ge-(1W;po+h?yygwG5)b#c?v=V|B{xj
zyVVhQP*Z*Ae2X_4OTiuaAto6n)~pO^#eRfDQD$tfz2~C{#hC?DILsow_PB_wZ|A)E
z>TI&1SzF<u&JsGnU>1F4LyfOgLdh;mcM5gbt~;J)s-)`znQCoiIoCDQNFUFJ3eOtb
zCg$#R%PY&XUyWcczc2pY18&&Xc0?9k(Uyp{_D>IaIP9CE_Dh)vwiGWcjzQszbL(ke
zV!F@ZpLE)m_M%#ZKU+`H;Uh=GytJAJ>w~-_dL?`oB-*0m{Qej^>*Q6{4<jvTqV5_&
zhls!*Lbn)<&0RyIiX}PwEFZ)_FQETlFkTioeN)tj_Wep7Mcs<&C29o5hA!RCUu+5$
z+_#wJW$24jniTEw6%-bORzGeLND8k(<kthCl_?x-qf2h2F&HY5(<uhd)Cr&p&M-l>
zx@zFI&aVD@=$qEDH-~hICpbmN-2-fXzpLBG2ai|cn@3z}a8<Db?Yf;V#0F2iME&BZ
zNtdj~U)jgN?0%YHNcGAH8&?5gz*TP(ga>2DlpP8eq!$(5((E)S40XG+S<XCdQpozn
z0cN+r6^|$jUAym)rn%l5oyk@CsSDU>?2va{--LiTimoT{&<c2;TNP=)*afBA*)_#b
z()!ewPb-n;=`|d)<l>EM_EDJWMX*H;n<f_A0tk)MIzkF^1J{??=6d56r5B~UrnR3w
zg6c*Q$lmw}HfRmALTGy)C?qRQ%*9sJd%HNB2R+wbv!4vsD1@w7_rThHk|k;3WDJvt
zTXeFJu2TGtiPD~AQ9|-s=6Rxm3g&5vGXOqxMa0zxxcs@d{zI}dh7TT5=MNjQUzsZo
z+=#uFftRNh^%DL<uFBE*Cgjw5jYQpwih5<Ve6PDX_M}Ltvv`PiJ{z?H6w+Ut7h8Ts
zs)6CvCnOH8zWvqbv-uc^1kByPS5wMHZ}-W-$<}H8vmLnV1(V7>ePlKclUm)89h(DZ
zzgz5n;8nne;a%_>k9F*3#$C60>Ikw&=tRXy6du0l(pyk|N2+1cz_fn@|M}?^_nC`l
zaOf4x-M&msN~zk(b`#%4Aym{4S18N}m#fk@li0r2mx-pN6Mmgtq#UTPVAmifBa!75
z{I!!Am&Ghv!%QJ?I2y=T44cwAbpBWXvneUP{+68Hy(^d$G|?Qp)T(@!QZ!wV6Kx$l
zosF)`XrwSGXBi+lD5h|wG1cX@L{q+rV%+La(u@R1GnCmRMnmrs9jfC!gLQcX89eNb
zv0Iz1G}}rA?Q#QPmn<2f-LWe<mo}~W6h68jF`8TD+;0}(2&xfET**IU+x0^@pkcFI
z$$oQk>uyFt7Xd0Hdu$D4b*?LKFMX^zV)+%xF&cn;^wd&pT4P01!*_sVs)02m(+POJ
zqfN$wM|R{H5>q>ejc^+en5)>d-t2*q)IX5MRqc7n0J2<!V`@J^r;jxkMxl{|g~|eg
zjif^!a&rWAvL8@f^lMw2Q3IR9I)SAA5K4Z%I7x;yNwD#d;Vv61Af&nMg<!Xa7Y1!c
z6>YxBtgJMP80j#U;L~<Z1D7q%$5{!pD#5QLDSHG}!!&3W<PPKcsIN@Igx=T3UcoO5
z2ZxH)QsXNglFc98I(-36>sS+)dz=V!sjj~!mb8BMnGf|7ub512dTicvOmxO|>>_oC
z(T)J8&8B)en_+aB$@NSO+tCJBiHvZVT&vMgr50=e?M;FZnbegczFke&O+PMOxb7Fb
z8L}C;d4gz=yj!x@`m)qGFT^xY5hBu=XJmk>BSacof--TWWR0UNImu3ymsfY$^nuda
z<&>=GpWxC!wQZ!tIbcdgsE6uTjh%8&W{l3Ne$o!g^tb!XQFn$1XZ3{eUy__-r)`>T
zlv>)7AH`gtj`?J^x4l7#y2tSlB$aBWE@9m(U`eRX+CC#kWPRZ$kR?jDWkZ5iH6%H5
zZ}_s~wn-)*Vv4-5m}|aOPJ*_uE{0^_*HyZOAcous$ypE{oq{aILy~Wpu-e>dhgbNA
z)tVvEg8~_40*rwjLrtXFEQIF714mSP_aJp@?7QmY;`wL{dcF|iZRwFBf+q`y1oIc3
z=mvWdj}Q(!`BH8Co?tu9V-gLi8o+U_KvRqJHKD<C9Y}$@RbS%W&1||e>0TlymtwOP
zxN;+rei=%tuy$hANWJDh-g^2-99#@<P3z=e!DTUwmXSq+I2dp1#QA(~b?GH>wo3?`
zCh%nPoM~5KkThNP>&2O70j^u4jt{cp{l)u(`)>QM$WHq6j6E+7!7!NXs>>4pD7oJs
zK+3g@@M#se>X&4WJ2=5g^ZABCM-~}(nSZL^y`5M-J{pHjXIfVVGdJ3ePCFJ+IC*KS
zC`2;;XN~OtJhJ+pg(49TfAp`Ci#;<A@hMrb1rtF9LlRx1iyMPQOr5HbKCeikFBiKS
zo#bDB<k+c~W1La<D|#^|T`RDi-Td|eB-!yimsxpUGmOIK{31GG0y71mvgNq1q+SW_
zseqy@$JzL3=x~s}jO+tYKC#3V9g87aT@9NU#o9%#055mTHmbZn3d(@SEFB6PVyKRZ
z3<XhA%%;qh<M(JoW&va((Q$TeYqPwDh2zXm;T5{WvKS-=5AevPglM@hLfV$3>f{9u
zTIbK7ZD6r@2AmN6E~M?emqSget8+t!VuN59nAJmi{?ek1W5Fk4nVVTbd<q8`d{t7*
zpieS;Yb#j>0Avv%TDJ>X?C-zRd`9j3Vza15Rk@`Yg34v|d}*%+7Nr9icdzDduZwr_
z(pTQpO><qz?Z``XiHP>KZ{mLUqjv`pWf4NeJ;nlyFGhj6KEJ58c>w(86@O9}RdV2c
zq8N7HK!QD9jsm)pD%sODyAi*UJ~27UBSzOBNX2U~5~~(}KYfCacdA#V?G+rUl{FHK
zu(0t>m3Ybe+esP<V>%nW{pc={Uo*V9&Om?r)&0Ekj%3TE`M@IeUUeGW4!uM{M|HEV
z%=}YV3WF{&XYdAF$Y3*`77{0+Y(5+oag30U2<@Y!741gZKsG$dT7oEVn<8e*WoDsG
z0P`d0io)K}6`PN*a=i5X7v+{GU8s89;TVOaen#cq*n7dMf9Dgd`l-|30W^LJDQF={
z{j>_6sEIQ+K&G6$m`3~M9OnUuzO!Sv*rdeMxukUeFT@s<*#BX_cn*_^hZFKHK(?5R
ztS-!bf{P3{%pIS5tL^?0UH!n>=qN~_6hfo#5SRJKJOF-7QNGw9!nHtJrEH1b2On8I
zmT}5wjh@G+s_^Va71I0hsHYxtB7zE;jpCRV7(Q(LtDi5+S=!o$lw7&7b~<|MG1U$%
zBn?75kv@bC6_eqF7o?ldIjn-bzaLKUhS*`3c~-lRPd0!S13yFpJEf;3pV@O;%5*30
zvrU+<nQdrnI6oIB7jUs(r~{_YnC;eeGU8i)W3z^r`JtH52u)u{pVrUWo_YQgQ;Jn;
z6vzxG<Q^+qFXt(pX61*RO@pU&&QPw`UKJksl%XgVh@Qq8HAGSyNqEf&i(a=(w}Vzs
z*NI%smONW^&_NkhrPa>QpIAuA2J6fiot3@nQ9@RoyXo@$uHCZj6zWv{s>7%#9n#Eo
z-1&1R=e=AbG6wOo2eF}u0~X?wauZvxJ`yK|*ew_qPCbH}TfKcRBDmK{f7Qga3kSyJ
zg^v}A5rmiPw<l5Hq4(-mr1Y?jf$`mlruGsn?n`q|AB(Yp&8z*k(J5{jL~J_Vo6g`%
zP`4psuGsoToVE4nKy`e{X{IKPMzsush<UcQpT2HC8=V&}>=+ZVnfS~{Exqs{v>WdW
zVPn4SAY!W{)BfWNCT|U$lGLVDnBmYNb|lj4povxe6S})}VQ5qqT_dvA?%9&KWqu%O
zQ7hgtH8j!JaEB{hsbAhndOo36T*`3+g15p8K}bX!S-`x-eIJt0E?Rj^8N9iyWDSvC
z$?VKogo|gFX4`ahtl5-(cf2-@7Y($<bYn=j0x}BwzUt`Qsg+z-dEaBYY)f5YnprB*
z$>WDA6kNL9cg0~CFRu_JAV`M#j+w-n9h1C02Cl>xK}%OrQfoe8lv26{s69%?9z?XU
zc&m~Zsu+dEiRM`<53I|k4iu>~L!r}<osP-|xng0;a6rV9r3<|S{Z=e@K53t29&`WQ
z*;Whyu?59))nA#2866zr#F;0VUKw0l`=oe++uxZDUV_$+(v5CalCTb}@Q=428@CUn
zl8K9@9fX}v>1l|L%s>&8T=LJfN#!RJ)f40Tf!Lhj$+{8-k+DB6hxSZ71eCNb_1@Eg
zrIEa)MYLt`Wg7(ed!c}B2x4t#68e%Gw01-?N+c6QwsEJGEt6l<76v6t;&8tBMtiqI
z!sdA5Cnb<%Gtx<~tyEa1w$|AnZ~CORD2e-=ctbLV@o>T?&s27HI#J<k9d*tHcU*(0
zBpk}NG8uYaN#3L>l}<^k6T>0dI!$Joqf8#mo)d;EOq?*0@Q!jRoP|J_Zh+XSw-DEy
zSX`Rna#;iY)&~>l_$|#{)>vS1>v6ZAye^jnUG3`;jRi9);xKP>?k|Zodg+5=@sOlC
zU{u&-wS93jow!mO-G7GhjD-{fok;RNrw~-*%6M*oMp->1nU!7YHTvOqn|5?oNWQ!k
zB#@PK!Zb2_G0__L`4r*|GE_p<&)}p!<PeI`aw@1bi8k4LeOk^awLsiXR`m9kp&4p_
zaG>3&^o8-XZ;xXs&hdCUe04B^^=S}LOSE?Sj>&VRyN4J+N`2SK_?Vl4coAS5Lu3}d
zR&)0wB`e|r8QkO_39e#ef{U$v^>kD9^e^zZx_|0v&|8XI-!3fbF8#Q%bQ@va3hcn$
z*k_G9B`Mi8zt@d+(R95>xAwBBdmbF_lDgvE@=SbXpO$a40zVrm&aX_d^opl1*h1+b
znd*ley~x8=qEocVU3OAXdXkzO`}Z32Ash-ZB*kTup#E3sDS+V?rP~Iz2Bpo$D<GL(
zCm3-{rPF1Z&%Pt7n$5)+(;C*F)teo!9Lgf9-Bt;_D&a5}mU&~ZDXyuLtb+MWQ_AA%
zv&*r8L##?SGeD}W>Oc27K>9J4?Q5?$+H);sXU|EI)A|LDu=?v(QL*B#D}I%76syZb
z5q(n`?w1K(V8qDp1z6T6a-zmUUnVW2Dl2RGieVYe*A@Z4a?@yR)z(V3HcaIWlDjCA
z)BU#GpU#1PoO-A0(){Pw=PVM2FOw_r_sK9_X^rvg44g})?UEDfo`FR5D<?MUFx)%D
z-ytl~#)wN}=45iaI7^qzEP{9w1?hIYy7>1xs|xk{<@rWpCE}e`aTbsoVe45f)`g&H
zAD(jt8W*2vE3+0RTnmUYLl?X?7c`4%d%GEH2GQV*f`iF$+rzGb4m;_KNwkbY%naS-
zY-W^}iIh_0<h=BDaUu8lUC@EtoP0uo)AIsf-H83mUDVfj&TL9CBx6Ds%NA}e1P=4}
zI??MGhl11D{~}!)15YOwKRT{ID=UXQMDa-&aarDpsN#57-HZv;Pb25b2}opEFBaXD
zzI27iOMv%F`@5<PnY-^cz=NP`etpT0-LZ34A484%o3-f>P84DUloUH9{%43M9K#J@
znECk4ly5=83S<_H`SYY2ItEBR@5C1@)Y()w_Pfre3scJ$^_em6KnqxMu@LpO)n28&
z#aYKrrGeyMLHxg<DjrM7h}^nBsEzaMnESz5egsI@0tl0a-tVr;U?ClB3H-=4biC-+
zNeaFWXcOY|?ak>y$~(syo+ea{`vF%CmP3s^?1(DyIh5C@vPY}Ex=Ct_%D$CB>T(g1
zvaFtrArffM{#^uMgLI4|NPcd&H7|8-()3@q8sI@HcouM$DuU><=yPgXSEtHis##LW
z@4O7a@x~vZjqJKptQ5Z-Z-iqyf7;PF!Z`+UfKW9nz4=HQg5-=ZMx@7a#xEZ}J^>c|
zhFAzOrQ=HOTCFekcaz+5dVHMTgqIT<P~<>@xej2}&>qJPJGjOoWG#R8H41`cZ21jg
z2c5lrD7{*EGbYY-*e;|vc0M@kar8Hg46sy+_>fKU)PG0(R}2URXiO5w^|Pc!lk9y}
z<haz<ABQxJU&0wb>jAC=+%EO?o0zy4^tpnr+jsMBkMUVIr3n%E_LGj<-p0}X=)3U3
zQSB;C>Yv~FZ%Oq3893rFYAs%Q#QjcE{cocJ4TEkiIpBBpD1Uv>-#^~u3<5IVjWMDB
z{Yd|(ms9xzGKFTE_-mg1pOM!;z37jR*M15wJ$Hd2;`amQAJMkY;AnK!pWt_naQ}9n
z&LN^k@uS}N{#Tmqzrns$cs>pBqNxA5Pr_cOoq=;=d8!XE%v`{B3JH$yX=iY?umo4W
zpP&phlIfKsMIZTJZ;S^UGBaeptv&d))2!jj^NiwR!S}^T9Bht4kuG748qz0?aUsmO
zx&wyf(ilS4`~V72oF>diyKtCwFz`c>G2VuKunru+d%*}SqAB%9x%<KXCb+KjgJ@0#
zBz`1lkk%|hb(k?NHQepBT8B6p^y0~f(rNMfCW3XH;u$$~8>}>CFajI26-#ukNS8dz
zpPdrs`MwI?od?75fJIP)ISBO2jzZ#g7&<Zv^@4eja%v*HBhx4@B*cUd<09k`KS=Im
zDvvxnMPUZ2vO9;lRPg@i_Wjqxwa0~-M<Ei-_h+hEas-y50)UE!A%oAg1N{|^5s&;U
z?_GDDI|2Q$7TR4AlhWtX|8NLVAYmU)Bi>>RFSoZU4=8rF!BV{sj75ycYkl*eupbw@
zFCz?%>tLPfXj$_Ww1;@02WkLHnnD*XC0;cHrC|K?l#56+Ji|d(;~kf49HN!@x!b<J
zG|++FbO2FIAfj(XTq>omD{-8V&W3j?%xD3ErmXqidj#h4h#++namX>7TdYgVdT-XP
zV<#u|P9=pEvf(9*km^Y^QjjO!vdaE=lpAi()`XS%B00B`-~ES0FF>)-1*sPI*J7sA
zW^jhi0edH^kTxMsUc@-<{Q=UN6}meZfwh__KbH;pZ&Ht?-Y~;5X_~RWZYJItB$*ih
z{4QefjfmQOyYn9FUDwmN>@f*Tn-$VdW8x}g&@_kNyJFEnV1(6L>ql{WRh@pc_l(3P
z1Mrd5tp>3(Q{|5-@*U_3!WTRY?(N244OcnF?=vHj#Qnai9<2Xqkgm=o7=(Vx={Rw{
zp=G>AZ_XrliO%GwbQ*RH#HCI;caG0H;S<|-(ABKGK-?VdqO<;wHa5o*4oQ!SOC#vr
z)jHyV#2TAif5i5Pm|yDa=DYavLPXSF2=rFRgqrO`xt>o4w@{G~jnz?%4}{ju*t<Zz
zA>C<t2yv(j%`NbI-V1qQQxL)FH5>r?M(%zL4TYYc2AM!KSU^ccdG><*Z#MZDEaEFw
zQTwFp*OG9NJ{z(-k?3OM>bKlF)I=fplK9172iFfaVoG1hDHWinZDlVogoCe7uRt%7
z`y>L5*PGG0&$6Gp*=vofu**Mv1R$BG95~pg@Q)AAg44};Ja%7#!<8WRU~qiJa1T(#
zc@DT4VncjF!2IO?(O>%uedJ9gqX$N?`8`J&U3p_uMVnSfffzAsYgm<kt1Z^sDn8ZK
zLv1nU9+Q3CXS8jSZWWThJCx^zf;1`-fW+mMoE5Nb{0P+4izv;+-KF09fuQ^y;B^o1
z54S7w{JWcF%Sj{>d-g;$=-1uyMc`}vg1%@G;8@<cct6TaT0cTPaxi6Gfm3=Q&|A^u
zJtJ|(!JeI|px|h$hOvClr|-<CkTwKzIJ*#@mNp2lx^*MC5`qEVB6^H?`DkIDJ~HMh
z{)Z~ZuYIcpF&|zi6celdbr~CHJ`)@k!{@f>xf)cIO^a~Eh0w<{*wdkN4*}6!Oe4Vo
z%MzJQfP@S}vegYqG{|jyu5UwdS19T}P*$rMhZOlavFiwgcWrkbDTRWi>mye1-zfwK
z(g?JQaXSLwVxfbAbg#=WM7Zju^77Cp#Yja{qd>|O5^;WdAKGFU19QbuYl30Zop(58
zJaa36t&@#Kk=80uZ+C?J$<>8=?#uSIzhWUj(?Vx(H0YWNVwX<S572Ff!1CSPYUpi2
z=P2ix63}q*0BO(ZQn>(BF_Of92U^xix|;vWjrL(#dk%_Pzm<in61w49fvJ9?soqzg
zAu!^k5Xd=sku$jwtzmL6LjscLb~lq9^Z|e5@-^Vrb>G{d7m*G`Y?vIyD!6IvO9A+f
z7pBMf`w?JDpiNU^Uf&xsf#!C0lXsM}(++y(6pfqXu=eDG$5~DH6(gk0P9IoTo}Nzr
zB+;ira!P0Z2lzn*TBah6O>HgPePL0&oM}EQ3f@vhU%CQ0fks0*N(I$Bm%~9yZZrV7
zCVt=($5DuhNN{j#4U)p%%GvCSnh;rZ?>k0nh=I-T#^T$I2)=rHu(RBlk&}PpvN2~2
ztonT&!U=;VIO{#%Dc{7|qgOnCKjScgQIz~enGDu1(XkwTTY>Lfzrmegy)A=q=M#=-
z$JcV7`{U5_rgv_GtDIbHc^S6tQX87I&i?C;?%Lkga6dT48-VQCd3P|B?d>_7&E;XC
zXGC2OyRFYQ+F=L>pec$4*aFK6bxoDO8s@TOqCWDQ^_K}$4Vq_8-uN|2GsLNwabKIr
zj#9(I`mN`T=c&cPvLdNn8iQMbLy)EW?x9IpI2~y%sG=$5db3+%3`)A6{1;QjaH2u*
z#q;c$Ty!If?$8ftfNA~KZ94@aOu;(%r*9gGxBxMZ)F)eTZTy6tz3&go6NXzI5}V=o
zA{gIe!Q(?|CY0^U;hrY`WeZ|**ddF!t500NDe@sTEi-~m{X1sp*jjlg$(L0~@aG|+
zxrYxf2OK&yNSZ3Z4S*XOR`ob}iUbSOMezk1L-!fTjx>Q<*$~-&EfQ<L!)f69h8Rt&
znUo=JZyK%)vCE83AszJu7zm#QDe=usme>~-Ng*AQLS`vNdG#HaaamyL=Y_T+nmU(Z
zSdaycNDpw`%GJu?hc+YLh-01cn5-vj=>622Bi%*TcY-pcFTIh2ghkdM@YmN=vp&5g
zo|3wx$=b<eFWpW3stxXu=GQ<?B%;ZIq)qPYP$%UHbaU6Rq|n8T_;5E9p@8#RhP=Le
zN69Vp5xEc&At)W9&oYdNC!UtN^wZq*6*OU{P^CZA)rXj#hHoBw;bINx$K0jeQ6l%N
z$HQO4en@VqKr?3FRz@d@FpzS3hj-RzgO~$DG}U&zmx@AFn;{mn1o@otfEOgg1cFv_
z)KoGYJ73B@_npe9m6$KDRTbStIw%s9(w%F!#>t>2e!=u^gII>e0I4*GORlFRMSG7~
zEq)twucL6f`AVxw&%lcA!~w#i4`2QkVR1Q%uZmvv0I=cnHEFqA$HfrWyxYLsw`VOm
zPWNHC?riL^N5RT!V)04&pjh<rp1t(bN8oPhwp)jygkA%?jmZT%^E=ROiH5Lo6h4k`
z&t$f|b(kxGfaf>DB#+JG-`yAf2Uk3TUtE_kLi1i(VfToOjfkJlv?m;kt`(oZT&zj+
zU1C)ELexy3IFSML|2fXqh0m5r2=$1ED?}R~1zhnCddPLa17Tdn=9-8NwL<UO7f+w{
z$JkXuwnN9+dc2$6!OQj|-m{N2cpp6_KMJuO+hRJegCBxE=f6M7Tl}5A_*m=+>fU2_
z)l+O7Ow93w2P2fj%VPnWn?Jcn-9)*Pxa&IQr*+<RSQUupJcRss72;2#bLc!9AatLV
z$76&B?OJbN@`v&SSFvPbx-P|+wi?~m=-hFQu15`?grZ|%UyiOP*@nn<DSP?sKgROB
zoHXWNI?nPENa()TK9Y~IGz1`Q*0yO5;PO>y%fGe#3f&*&$0qA-q4T`cD$m<+R&;0t
zCo6kzv;SD5kgrE=m1oq)iW^7CT=2*~;}S7n(81L)3-xgv;~O_j|DJB~(Ni>GXZCDw
z3E{2BSOOp8K{3s3u+Hsbttg9rw?*ri67Q2>o*JKwiSF)#56ejE;)|H|>bmb+QCxDK
zvv}o7<}&!?+J0gCq+zrLun6u}^f}H=q%xpoS0IsmYU|T^ofch31+v*|4-AxfHhYZv
zU$;gi^Fe(4o`+KBr6zY~X_C;QooS?niW1W}Kmgv@mt9PIDeeV=k*SIOZycY=CFk||
zZ}3;w7`}?!lqY+Gw>6@%Ef;gGTeCPoU_Vh4N|I09vS=B5uOls$H0$7m^=81kZo=tN
zZHKs*h~q0c3WKvQUT(+exMSnDTT)ZHhPk8;koKiLPGtLdTiX}i(@amcEZ_Vl2#4x1
zz1<|CJx0vCg=7xj{Y=KR=Y)TS<o_MuB})q*(7gQ%bRwb(IEOR$WFE?N7itKJ<t3p?
z)V8*r&1#^g9y304w+b<%TgIAnajTU$y%{msfBe*x7<W$04bpkEwvTS!Mq1i!I@)9@
z=D2OOsFhwh^EuWFb@45;@%D?pIn*kjT$s}Gv9)`esc*p(!`lqaVr-KoIh$E$cbt0Q
ztj3CB`bPn#f-{{CB=t17_8VQJLvQ1Bi;IAoJ^jRjHR?Tb)Z+*g=!lrJe1dCDB(Sd7
z=x(ay(>D=4756Boi;uQEgRb*3uBvNe?rUA(CaCG-zq2#H4CHC3aH##ks+4mxy*I!1
zqB-u?3GEP9#9S2<eeS}+IOV}}(R3u5OSeey(e{)Nj@EFiaNy)`Uh7k~CihN`?{za-
z>>e7`TZ^+{Hz$k8BgU0ZUK+)WV+se6?wOgVuj<NUC8AGXWxTwj#E+aSY12mIH&Ms6
z)!y#k&BKhXKfBZNg{aG1qp|O)SF`WHK>U3hf47pyu6s=_7h`Q5tg9`b(^loB!yBzY
zExVo+hu+vb|9Iyt>aYy=D0N~r1TaD6sgr&i`<YU;)dH)xY%AT!5`#7?V1q4OpnUgk
z`NmEey)N1B9B~xGA-ef$QT!{iHetlswqcL0L3n6!%xWKvhn)Ir)f5#MH*ojDCsv=N
zd(pf;dK>jI_e*tUEf$i<X-rK)MKKQDx;JKy*sH5j6WOAexM!-}0wE}TGvg|YQ*B>{
z{-nR6_*p9Flu*~s$MOpyU`6+cMnPgVPMuN8;<7<PIklAiD|w2VT)o4(rE*^|6%Mpx
zTtrX$Ywrc}d7YxTSW!pUy+|={`UE$9)tNp^y<mS4G;79kD=nd1LTOIUnu`p5%xSdK
zaZ1$s@hs4_P{<1b%U#^Rn#wcEnEQEMAR>NwH>h0v%lk=R`7dI}B^2UTIl|6iqz*A0
z4nZI33Y;V5o9xuGrf6(fE8k+y8seVlRDqv1#YBZO#84MGvv?EGn1!J3ZiHNX9s0)p
z26fC$`Mz#+y5)-ESZlp*TbDZ8u~3dy^h1=>Pzh)ETCr(mdry6&4L4mCPU+*pKfM6*
zmH<IazK!xPIEM)ON)BeI`~8<(beESS?xL-rJ*_wP8-zR*Xw`?2Y-Hcc66F%^{&x!H
zuP{rP%|q|&hQGyM_Bh&a9ubZ=R3}GZbvTGJ%=csapn@zEY3R*pHWwb<ZMU?SZiv}h
zGD7urB4yTG4kUgkNxEd${A7w+r+&$pU1#7vWH$Uul2CT+R>no4x2<^XK0Zd;4__Ou
zZVtyPT_^MJLA<3QnoPhX8`#CZj)E>(Q`C>^qnDEj$?k`HmFFYw>xmki)yjJtr^BaE
z6peT$YQxhsTxIKRGUH*zp*6X=6d&90u1!&HH+nwf>Q~c>-;EpM)4@N-Tdcuo=Wo<I
zH&)=4>vv*cO>=3_;+}w|jYvRAaTgL*Sa6YM4XVVK9II|~CF;I3t(GBFrmJd<>H>O4
zFU7C?EZKT*0JWH`<A{Ofn)mBTu6kw52ezsYr@@0-QQc{C4AGp-e}qDs5o)Z+zF$l$
z(u>+(R<{`G^Yf@ARM^hI=I+#}=lDa2)BzDJ3&-BOzI9&R<$c==FV-7G9#*K-Jmpu8
zYx8g*!sw-3+bE>YiguYdDWpU#wd>?tAnLgiO>~yjH3efG+PhDV5n1JhAV5Uok9j>s
zX2CnA2U4gNo4l#Vxh6NXvm)hkrO2}C1UHrj`TLCX92`5>qAbt$Ro?)UdPGr#SZgQu
zSsv8kP2r64bayt<3P3EI=-L47a%1Cz7IKZS=I*WBv7tLc-p@BPht+exsLR2ckYbG9
z2UTYuA{o6`$uj38mzJA2`0uGizF^@@biiNgVQs%h<$OQ<pJz94)IWu$9iB)i0`AL~
zxwgcb>*MtcR2ezr%l8|=s+P3k)4O^5Q3Z=>pa&zH{HED-PO#xV*TW-~Rvs0f)RlAX
zwW&U76prZ41lgB4qk`u)-n5^PAw$>hpFDq+>RGQ!a)q?}MRU&vhi@w6BorTWQ3@`W
z7uOH2&t+WxhM&9=7_5vPrOs>a>_6(s9ek`z`ye+~X=7jety+m$G*!E+X~$)l`Nly6
z8(Lpb?LU0jN=??^g`KoqeEfArU>%Xt==K727k@*7?7T$Q1A^-Ge8}dDD;Ru>dv=i0
zJ+7q}ywTO#fNsWCuFhGC=?TD37D9j}K$O~KE**)^SY}(b2pS$l;}n|e``ovOsg61<
z@$!5tTghF4eTZ_uex*mvlm<zB$UR#4UA&^w-0XtC*ePDxXa3D`jBxJ@^)i1YkpAfj
z#IPO~JyK*Wdg_m?UJTuF$%awXW|h#vkLDPnX@a-?mF@>uA546X*lvG${!>LO-V5Bp
z9gVW(QqbGQwa6_tBmz6gR-BY`-f%TNugQB7>Do0CePAf5yv1{VU|iNl0WC6W&ClHV
za|I@lU=*2#6YIq~<u`phLB&jwe2<N3dQ6_VRyf!nqt*|xbShc0pdPVy-|0U;RxE&(
z9oBf>Y2Q{L%4!WYL#0b8*~7#hB}vT;yNKS==)BDq&H4OH*oBdW6pGjXy^wx@k1HL#
zngV$xJrXR_f*m(CZUZsnW@(-!%^94;YFCZ9R>Vkaz{f0LgfGhMOmx8&9H-`LTOo(#
zxY<-*C!9aVG!uZM+q%^sev>R0-a&Nb&CiMdxYZsZR8-fza4$$7K+snm?=N^!V0QLM
zS@}vqXvCQ(&gVaT&)I_rE_M4NQNejcI5`sQ+W&C9`w6kPB(BpJx=>sqyz}S=BN9RY
zn8MgcgS-e&q1dwLux&R$=uywW(ose(x8Zix`dNY(#ml(5;k_j%E{tY+N0OhuR@QU&
zEX$2956HBS9ZWhDo?Zd>w-=A^G}he<L_cpD)f1c~Rjs;1WP6GBv<?L4@nJvZoojnY
zYr!8}8w63SOqoS~_ogsIJHZ;(<g0Tj@Y7DLYh88GF{0&&?l)mc8{J|z>TH7*IU}J>
zQ=cB=m-<AnhEttSjm3T5-wgts^b|&PUV3pM^9L*}FHS2SP>Eob5@<cbyts&96A+7}
zd9jk-Cv-1BQ`1q2k<{|6i^7>oMy7(I>ud9Sx<M}`xQKj%0ae?}oRt+UC;Zo7T{sP5
z^`6wr5@o-J`UM=VYhLB^pChAhI<1m0*vb{Wb`{LBNND+h*v+l4(zV8lZmUbL9_=Y}
zT8_ysz(^Sq9e>)TY2-UYu@Lj*ly?mYlNz-R-KjUwjG0OBZN?e^cKApJeBl5j)aL*@
z3uRSqcc`Fjh%`t!ib$VdwL^!3+Xb$~sc1;`Qj@!>5s@}3J9IIRwO{X5wWh#7oQHeh
zG}sGr8W$2r9RqzK*zFFuXr~uoU=TD5?pQQMo$j!Dff&?9%?#Qt^yK6WK<e{C_0cRi
zi+|JfKa{2ytD;vuml5BK(YyJ_QIHnit`upd$uP9<zBqU+j6`jFbCfLayd;fC4gTY>
znwRF9*S)<xV(`Q4G;0psdzsD~^H~4fS*EmPpW|oRFDj9owi6XOd_8Oq`*v+FEoCVs
zdNzFaf=k$NNv3UST3VX?phKz)+s>%K7q{6pkzgHAdiTyia4mPzBaJ9kKk{6UI$aNB
zlCC~ISx<;#*1#y4X_nh*X*bpI!g0`{TQMaQ8W6RD8cuOgj;&>@d$Y&y$)oK^0|<>@
z(#yZmocd0>GT)J+(vhl`HUk+=mNl1$%asU0S)PcO>XPOiH;H%HhnGIT*>F!T){=Eb
zy|>al^(<~j&bjgB2t3Q${WHRc4&$EwuRlCWsf15wv%R6)etrlCpX!k?FAjkgLrlDx
z_O`M>FUe8WLyCt_;4Q3AmUqcLV^vL|bE_1!b4fp~mD?D_e}ko)Y>8l+k@l5EUvw&#
zwPL-k^%=R<d1TGe>JQ^GSLf9K@F*D?vb91^<`QGit1Jot)u?fkg}qYoYUDZ-Idzkn
z6hLh1`fcoiyESmV)aT<Y83PoRve|}f3HGYW&xg!C!BLFjck+w4VU0pgGOwymd`aA!
z(P6q;020=lkmXiboopDN(@af$S1Y<IK{|Su*ZAw}^vNGh_Z^C27X_jCVDOgD$@JsZ
zjQH8RWT2aBNt)<s5}rUOTb#c(y-2r~&mr=`P?O0SNvc1aw}C9W-z|R>1=9~Y5!FRG
z5C6@@{`*X?>F;!pBLM%K_Jj~^rn$ZX^C`D?J5L|KFZDOs)?Z}LU~GxJ6bkpIkJ{^!
z^1YJG;C%HcOOkm$g?SC*BezR6$P8bdd(75l%K*Jrl5mC^Be;AJ|JxbZ|4w`yMY>V8
zxoS=qKvrMr(sqM@qujSt>^dBkIM_{7F>0P2JCku(g}!Np-J`BUjLD_Iy4}jE*wpU2
zAK<qEIH?bEVw8+s@s3CxKs*)<-flN2#RzWM<-}RCBzMag$bJ$L9*l?UCc_EN?S8$5
zIBDZM&gGdySargke<qKnXb%ow2479_{3{`SxstoI3}lMP`8x!s=LTk$iFJ>dZtWZT
z9e0S)bXS=`nOmYm#aH#Hg>-|yaSXbz6#&&KffTNaQ7AB=n}PGh7i}(-9kh0oEp|7L
zEOiUY&~)#nE&5{g4<qsC==8|N!983bk&XMuPkeMSDUNDVQU;X+R57O}pcr$i-FlAD
zPOZ7Mbelua3fR?#@;e`oRb;Mc<?4cJ-JqnfHQ%)T^vai$#*8fAp<Ly4$p5|mkp7dC
zagd;3Z+q=efWrxeee^!OSzsjF+IQGm2<~t*<G+)BWUJyV+5VjAgtA`eM`zc)uKfbH
z-Rs~SKT(e-#KwuQ$s*$*t8#Hp#sOpQeOE3rB~_Ckhv9C5UCQYiklTe7@!fV{QB#vK
zqW*E#%U1QrwUTnn{zB@3U_$LEONQNK{nRQNK6ipwsZ8%gschTNcW_)VrPyHPEgau?
z;ND$CrGJgp9WwSN9mz`6775pip}Uzl<N~dWyO6Y<%_;i~$bC<>f(YQjY_*-I+gQ%)
zH}X0n(a{3fXLc!yceLC09jvANgap)2s~Y4DOqL!@7S374hjnADhMCB5U*B1}mZUXv
zQ7!CBueZY`0vyNXA%Oue<m7wx93=57Z2Y*m(nT9frygv*UGS+*-cDZ06TH3&*=3tO
z=ANWxF1$X{?_almoVAugrSheXl;PZk;voU)y$wv9<*Z<wqKo(<%;l5RdEfKIwqo;R
zw8Z<hPG^Nyg*|`r2oYz$OZnu_H<x_+bSH1ibjVVF1kJ;$r^mJ4u9fJ)Dg2a&tS3LI
zMYpT?TcL|rQaMtZh4ysg%&CSeDm9HnU75RvNlNOp$eR&zB^~RTnFLyG!91j8h(7=!
z^&Gmb!!+CEH{FlTy0q|i6!4k0y=y2<ol3Qf!MZ;bECk!ehWT$#40m<UTKf#cQulMc
z`>#J{aZcv<9C)#v{Iy!nC4Yb_++9Ju`!f~CRvMBt8~BLd>NS_AYIkuE1%)5GT*XnL
z(J#qOq}T(7p5a>SpstC~kZcp8!mmf$7@?%EAJFjT`JtT8x4i#8M!0q;z_ilktp<kU
z0H?Q{SidLH%d4T1p~>aW>I)5$>sbSLQ7F#05!coNHQwYv4Wto$^DbgGj_ILW(|&&p
zFpR<<ul=S-7aES>A~Z?3yHb?(MtQveuHj&*L!NnN3H>skOKez3?&hpUV;W6_!Z(Y3
zyiyD)YK_Y0hidc3x+5Hkc1xqN9d2rMZBKYNf3~W<qg)iFGjmvZzP~67<5@IenAXH?
zAMGkFxt6LJZaKA7$6On(!N`$yE?DadZ{c-9JCup+C8!Ngd0aK`$u?+8bC{lm4&RMw
z7U3EXII^CH7LC^wzRa@A>beKiF}Lq^qS*+qtcGf;saT4&S=9{V`j9C#w8{?Gg9*ts
zCWk4~l-%tlm8TJ2PpT?Jw_*npza{THSz`6fAHaqWsSH|@tnoDq93|q>0z>A8BZ?*L
z34+}MRY{UugOv5=D8D=eNEyuYakzv!#6@&8T*D_HY6pzF&Uo)_TxX+WD1Dp30hG;}
z%%QT~wU7C;=e5bs`i3VTreYP^xiYZXdhM$Z-wO2=FOl)FZA?P_`uJO^i(W33$39!6
zWfXD{afEBIHT08J#F$S3+0LEiVw%f8_WW|XOeovr`y^F87SC(3m1O08_jy85G<m&w
z_akecDi~4j`?Z9%IIrm6?+R<+TKBHo1g}f4wGlPBMzN=(`*CYC)Dt$8EqkJ@e~j=Q
z1{mRe>XtVKtt%q#3}HP52_5`fhAqC(3lVL5nltJ}?`XZ@o?3VEKISrB^mMr9dl|j?
z6ST-b?&~mm87{-le&RUcADe#iZ5)DI3}Nw!Hab?>QU)n^!=&ebbVP=;4+kxGo0}0V
z&-e@#GxEf@s1A^9EAh}|%=+kgigaAbnjsT)?+ek0W;{zGgLNOc<9Z>AoJEB)-!;oE
zY_p4_p2SZ{QhF~yMtsztgZz6Xp6AiSK9(ti%mH!ZAxVYoWKooRGhtr|)zs^hn+PmP
z+B&@Q?UV;CXNrSFKIQLyVzU(rGUC?@*(rOEci3SUhD~$E?nZp#yxO>OWzd%YieZzx
zy+Ygfj$$dl(eD?(Rve=n&Yaik7N?eT{c^H!#<jEe)<E6FzTM7kNCZ6K4?c-AS7BSs
zR1!5%XS%guNFl+2A2yIxE4eyAe@6zJzTr@*t}pBFuw*Y#w-xf8r`?aPL||}>6y46-
zg=zWG?cBlJHmx4(+BO~b*re|*j`^H|u4#@MW>3zJr7KMG=;4aTmYBih6)s<ByghYS
z@PL!v9_66tK-q#(XI>t<lYUJp9S-N{nTwht-Kx7cRQsmd1@9E-_3gBEHB#6?p*u5t
zZdW&`%PEBVUzQ;J1~Mvqp5ez5e=btOsgE8%B0wdN1w^DWZs%8|%;#ov;z4_-GpWp}
zPJ)X`h~Ww~$t&_Vtc^oSTTGew$!0J4PN~g6hGATewAjQ!>_bvL*IO5Xv(BA7oxCjh
zH2RE}jNen~Gp_pzvYhuYJbIX{;Y_y1zirEhapw{biF_!<49##Mr$ksaJK{bdW*<yy
zDR5Pf)sa+V4JtcrjsP){HgGYwypl<iz(}{I#*(_UZ#KS%pcsf8g!*0ToSqrts?!w3
z-(9=ZD{%%_5YG5l<qA?wQ)Y`2Wdy;_!#jJUuNPP3L!WJ?43-*SGqmZPtN6bSbj>ha
zd>l5}`$0y(=7Pt?pAbkjXX3Y7u1Vebu8i@d-OM965f4;VT<LE|a$6Ea({uiKuZ&5b
zb-3U3CcKkEj`enE+NFCGvI5R=DVl(OM9PW_Q{*NpS+~8DlA^3gbQnVFqO~tCa9j4}
zUtP@1yix$BJDs=uE`^7;6}=;R+C-{`w6!B7W#i&ngJQP9VqXEFk(kPtQ|GUKD0N!a
z>)6Y=_U||N^yhO__+g75umAm5JRac2F>uHkp-tOp+l{z9D>}6EO$-pH*%~#bYSDca
zc3jobRY<qyL8bY+UHP|S9%v3J+j-ga&%yIwI~+chq8*&F%$q`3FaK=>5+*!KycW&x
zWEE}%CG(nesDc}Z<mr!+DJFbI#(Gukm&<xqC6l5KA0{|Ze}C=YK26vb4<F}TcTmQ^
zj_`lneHGJ?%L<N+`hR?x&f&+8YBuhg|Lq$7;cNfx%i?#W!id(Mym$Zq?PVkykErx*
z1$r<3ng{>*P5%B#$gdPI!D6gl(tF|m?PWEjRQQ@Y3Wd-9*RS)R*CCuia2hvV&C2m_
zFO%|s2dYIPUI+)_D`acVhmYF`o;!^$``3Ae1dFMRaNQA)$|-&whkkA*schMh`0Rru
zfidi}rkyvc_<iUVBd5pVsMz>s3mk&#5HpoH3JFU~lily%kS>f3OKi{t)4VfPZHkxZ
zG=s4RW^GCftkM>NM8F?Zr78mtql5wD?Y~{}e?A&TKX_gbY@Cn(x)QTH`g0vAn{aT|
z@<$1nxlb>gIgD$js}BjG6CxjAHsp0}HJ+*`=$QI(9-N{PzqCzQvHBpwu?C`+u|HOt
z1)B3DJ1q8fY8P5)y>L>q?z}PNTn(WrI`W#)Gr0&(#)obN=G~bt<g6Y}?D+M>JPR(}
zo<jTItD$A1Dl8)r6%ZylLOHz`Oukf>FY)OAefIu!)AS9_P~jJ6zmf7g^7G$3Ja3L5
zN*2g+w<Df0o)gb`PQaF}5aYVDxe54PALu)Ehs?^=WpD~Gg3zo0y90Wx42TU8byOdK
zRw|S6@Ris>lXnAwWf~nJ9nNH(lxzKl-zgI^<4&6Z74||XYKR=cE_6*6#<0xiYUwVP
zB^tlwwM$Il7=e3dsatBF1NGFLOw_5pTmQ1c&Q-vzcM-xp^hf$2CLfmsKb*%lt^ydV
zgwAU&PY&U5(X1e@YasHqM9OQv7H@sk<@H70_(T1%Q#CxDPvise8VoOEq3g0%k<}oR
zQ;WFf4nS}xJMqb$QH?w<WT~3fKdW{cl1p3`5ShZFkH<gs_+Q5TYj3zBW!|&vf85T)
zxQ`Es2p`3P$)#8cn?Xp|4|;q}?dkYJ7qQdXh3|Ta{$<+zA1k5aJWlFOdjLoJL2iFK
zvDadtI0r;SW`)pr>^4}RSv0-jMQ--wVmIt(ci~^N<if6ggtCX-4v;mLU?7heYd0YH
z4(shgS=THkYGfFdBbE()06(1B{ssk~en21$5tjl>i02%^e(IUqgw@R*gg22@=|DUQ
zxm~yI5zN(2gRmJM<sBG=x^w2=s_+~xecK-&M$85f6X8veYvR6zIA{pm=H(DmHQ72_
zkm_lHKS2geuy40>-843Ma=Dc~uOM`2=p8rSwWfQG5n2F0R%3k0K(Cl&kUsM!1<BnY
zW`=!$t&wE~{>X3u*EwF51U5DDcGVa}DeW-R@(~Cb17JN}?hStbh*&-XQhUNkv>x%i
zx_WNB+c<uf@Brplw#h5;b2(6?6mTn<sdQl71tKvI8lp0)rge0LM-ZzTVG%uHUGy?K
z^*ZGA(vfKt?q+(zerbRg?GE{us_$=Mc74${GK9i@Q+njX?3mh{xkNBik_4c}kvOl{
zUDV{7Bp>y&v8ljj&t;xj=gziTzXiyOWHyv^x<FMXY;XR~0YQ8997Y=pU7N%TS|-d9
zi<;srPJrlGx&#zmWeUN$e*p2c+78sotc!T2GcLPXy3A?W*Fo}3NxTFA*CEc#2mR?<
zd5{M%i1jvrSueYV)=G|uVhT4aPb-ZTZA8pD9}Md)PJh`0XS&cOI&dC)ZwcJ1BN(yu
zyUb_YS3rNeS$xaE0*U~c2$7T0t*zE9u_~O(Pai1q0de@>@e~x|WV5ea24U>?6H-gp
z-O`&+4YG4Ku7IxMOp`>NAH{g+^~&1?|6hAw9uDQ+{~x6^Dta<V);h_MNK~Y;RV4e&
z*s~O6Cm~~}CT%K7*_UE4mQf)z_GC$xtl3KTeczXU@4HUtdA`qi`kwRr{{D7dU6-0M
zbKm!8doQoWIwZ{h_^XUt%#YA?E-@``O(%O0t%s!~jb*M;NIi0UjLsWqwh&*4q&Q8~
zp4=bqs^4&O#Gq-mJMZ%#q~#Gv8Dx(0L+2Tbm=T8r(B{14RWNu&d10kv9(3ATVLJ}h
zPqh)VvmU1Klmp!PAL$NWo0F|jV>orG{j1G3$aKsxo^>nTek>CTkcT?T#ko~v>OQ+p
zt&IaFm<eY!GIMS%aKbAWO*bcX4c>O6a7&Nc(1h}s@F!=DZ@{z=znk6RbPm?hn0PJm
z&u8WDyT0ZLBpfIY%l~n-Y4*@BKU?)@I+TGVSoF*zsju!2-R;KU$Tu+TbA2!T8Bqt~
zB}vMx^V+cr{34`RvpilePPh!?4Z)|>7Ub0V2ZLAE5nn*h-D$peO7xa5ECWz-JDmap
zOV)N$RNHF@2WHvTj5H9d)Q`}NLCVFG32zDpbqm|G?IN<&S9>9Udks7i0dIr;1%H|Q
z=9gK|NCUP{QBOJUVRM>cM1h;`H&(s+DNLp_%wcl4&cN|UYe`i-KyJrI+C*(3Nu#~%
zrVvrDun&lEA-}KW)qd58V-Cg~yeJM^z#xRV%*s&_4E#J8vIb~=M_Hjxc|%E^-?J(H
zZ19)(SRII+#U}`s>c#b73@&wV2sGNOSr6WdP<VnewOT>3Rk=cxI`#haH!m=xzIio4
z?56)&RE70tl<-hEj_;fw3xh*L8>0LNLX@U%X93I|32}L`{zW;pqj|<?dm1!>OQ%Te
z>l3m}PO5L$>JIZdRq*q>J*!Jhw<7hi_BTnL1OwI*z2i{RDRC))24pMZJ}b7~V<*Uw
zGRG>W)v(l{*WjhVALFUEa4hif62>%s<<>GH=cXDu4<5!292&!k=|}7|7GhyZ)F4&R
zYVzDV*gxFdKql{R&ev*hIH_7AU5k1#Xi&EjtR`<m@@I{0kis!)(+LG`4cs0j%W!7S
za6vL^<ib&g6^Z#QpU&q<J#z(4q0+;G2NPBTmlIdA=pW^ITcnG90lRXwf!{t++_7Kn
zSTCwV1$Jr@(cQce4dL`oTNis)a`TnpRbiIU_B_`y&hh=>26`c*tiBpdX~Ki?FP^p7
zJS&qgVmEdxM$CT2Ha9W{G6f!sIu<C(;0tSHnznF<ff&W!pxyIKXkooKS1VTUo1mPJ
z#$XqvmX9Qars4D)5Xnw3q5h3(;ba&>O!a-GFnpy*qerVpqqwxg6>&mG{1~u4v&7_?
zlv;oDl$tAo&Q#Nv*dAwU<K<`LA)E*D3I|_lnwfS6fHQ00QB+JzZk%UlF!VWhzI!&_
zo<|)rrZ^Tve+HCzHo%RahH>i6Z{c#3>*w_t#3uW2Jgdz#Zy4}jr}M)G^u@k#hG}&`
zxBKuf&PHE3JI2MQ|B(1Sq)F<b=b^tXv&v57z)Tkrdd+SP!clN%_5IfhMd6?T)4j~v
zr;I;|&MB|m?Y9h8J9=AY0|!wMGc9^^A9F8>H^09-zapjM6wjr*GM2bYd)7~1={3WK
z+`LPp*UJRU-qRqX=-DeDQc)@&6ma7i06D~`tBitttc?5Dh6@9JQ!rTo$YDjPME8s3
z=#?IVOBGGlI3xDKzVUK^bnH&ea~}G~5_?Zjfbc%*QMQ`f_Zt-V2Kj8|rP1gwu4Y$H
z4D%>m0;|+7bQI<C*cl?=?v!|L-jd-6$&zHYC-uQ%$Mw=hSblwbJ1BtlW%CptzPI>8
z#dV*0-~{X>HdTnRNn|_U<B7gSqu_dr{VT5H^(UILGfk*0yO$4puP&yxoMhc;+_E^S
z;uNI2<Gp)+tA<P{2ayn8p@@r{kluX$qI8UOfOO2usupMNw-@z|dGZ1gtMQ8K59@(#
z&qa(CYqeZ~L|7x|QgFLL9aKE)mu58$kH?5F_&VmibBP*>K4MuHD>m*T`mwcs#OmS!
z?CTc+0S8N$IdOfG%%PhUsd{MQ{-TPMG}4;;hJ!Olsqyd}#)-q%y+^pLh%bq5f{{l)
zd6Z?FQm@8)kl$HB^u3Sy<EcDe;-uhGbHzp*vqEFE@NB8#$9gywxyB>%Y9x=!EOL#9
z8SfY8Ft>{h*N`zq&rc9&dzov`@O(u^czN6`+EFA)g9l)b=_}L#-;(Lfm1~6BfE~}F
zXgkoTc(?m%b?MXY#jmw|QJ0yG&hS`nTc|g}p5McHZh5ppst_RAM_LSI-pv|blvpt8
zw@-me-z4Y!1V7k3%<x@mO-yHAv*dEu#NK;@z%K7S?^qKV9HHqwY06zF7sxx{UL-PI
z=h^!$htJ3)Z=gzhpy3;x!lFlMiv1aM=ZUX&Y|AGK-Z@j7Pb#d;d@&ntmsCIbm|X<G
zk!I(;x(*n+JW%wi_riPCxuhN*_bZrOhz{8I9FT%1roUWRH2Rp;(_JC%)#SunbjFxl
z+Vkd6G~c=WTE+DsU6#P!!Q!Ec(um4&rM$vn2Zgn@q!yQx!wv-<<e9H#t&SQW*KoDJ
z&cd@qV|{k2EB=VM{iF}Yez{Y?1(tWg?>A*TTcX;Q9!CY~TAn?rO?BE;pt+DN0@E6I
ziJO$}S@nLh?3HfB*|m?I!u93~ET|3pOQqVrz}dxuS^W~S`Ms@dZjrD)Ni-q&J<7`n
zRURign(waJ8<JMS-fdO9D~g%A$VU4JIrH)R3LmD^5=}`3iwo8|LAtba=hccgkydF!
zT8x|e(t8{50b(H~qNpoG7iubj^$@<u6sqmaJqrA(9b~?Qcdqs<V!2(2VNZ{|9O*Et
z0Iq2NYm9P&W{iIhNrF_$%GLOQ{v|55ti8!iE90$jfQ21}_1z8s2vj*Dm7rz7r`vOi
z?nIR6+S~HSE_@gbR_*#_(oG^w+MF>$a598C<UP<-s$HJ08_Lv8=(I79@d&hz9A0G2
z)B2*>J(x+pnhsOD|K<(nQ6P7iyPa>5SCgQAMEl1Zdx>2XQS9_#jcmvlv`6(NQL$03
zG{K39s>lZy0=F{Ip>-Be7rJMJK1oIKQZv|Ua!Xv7!td*r58@sU7$%%~S}rd~t!BeK
z3_Br^d6ksvQaF9b%B-W;HHkI->RfW(lcxNIk%=wTLwo$oP#=eQBHEoFG`iKxiq&Nq
z>{64z#xU|ojh`5E>BPvgI|NLni+DNFPj|eyP|t1djc_iYa`_vdX4ZdZL+urq$Va^p
z=1BE;R^n6H(x}T2f$r6<(<%Pi!Ds3zlebwi8}0a^#nWKgQA1_rVC@3WlV28I{;f2@
zbV`o@1j!g*Uz^CQw)WA;PdMgIw(2+5Ra}k|x`5TS3aGGkPsb^=Cv5MaK?8bTqPrIF
zyq1ENt)}Dw@96nA6eP#m5rY-mWPtyaMk805Q6fXPjaaThRh77MXf|u{?l+Lx{+#Tf
z<#L>`TiYcrlU(JB<+if6JeXTRpM>J*0FC>~vhFL%o9KeMaXE^`LL{NDli?a0xl1xD
z=cb`krAenkaQc{Ks3jb7I8H8+a!g3(DHXFC&k|lDl<ODlIgzY7soJhNX#s)cww1GQ
zE@LuNs|Z*Zi{V3XVyspqHf%>{YU?f*l2Av7^`;5h1eOfwr4wu8=o?Y5F<kmE97Q0h
znW;KxfTkyu9o$W8S~DCth5m=Ib%+#i8Rsq(y*;F17DC6~c@e~LYrMefq9wtBakHh$
zHscY~SH3Th3Wep0#@i|MsQ9@Py4A9^b8DIshYfTVx~R1p=TE@*iR@}r3&%vJRfR6E
z9FXb?9>EVx0m5?EI#ydxU9H=NfWhBd0cuILtpS}}pax;HYW4+QC9X~QB#i;VV{aon
zuC40DG*GgW3+tyPJK<nm9(LTe<nN1K>)Al)czhSj#FW1<Fv9%)7kzxF)EuglbC~;P
zrS`|3axnJ6%!`ZVZ9z&rR_mv;mY%-o_v4;P_WW@DV1GkFn8q$Ep8JQHjV!o_a<25)
z0!twSPSQ`(uA`xI2IIS^MHfL~H6jXVDE(-+m&Ws=MWrkG22YnB=jc<*G(4B9Cykl4
z%kEaKjswGk=)yT8rV-XsWvD4s-c4Ztu<>bkxneN!&1f9MjF|p$rOG+zITEiK-Da0@
zw1BufpExNT$sCxHv&0HcJ#2`Zz<HoZSiB3_!kP;GA@U3nF)4RAx0ZGof!T%Y-jkpf
zJjtqo$l!wUQUz^qEFzwaTncKp?o=LOE$zib1A9b(SM^5UZs~?>nbZqVsXffx7QV%^
zTG*SA!*o6Z+n2h;THj9>h-wjq={6I4coLirLUr#~wL8Sjx6<~85;@|lbCf-T0}QK6
zh5F@#{be$@Gc$gI_PPxuyAsrAPqOr5?n`$hI%`bdd7@7pszkU|JasQhcbrHqStoKd
zdvdDIs2KL+d@k9&zwXwLvkBJ_iOiB&dMjK=^5`n^+ZcTJ?sSgF!noN%^-wW>8;`sP
zBL-8Zc9(Lo`<y2h_)IfuXb-1j#RrfBGzQF>Cp8^a9#qTRLSqYQsQ95(vLy3fkejO(
zrZ<2>%152#dJC)wWoTXWnit#C7TO)kke+j)-lZ^$5=w$oRLYU27YE<ve^#{W_j~C$
zCTY{0Dx)&vXvwjQ=iU+Oaij6P5U={gk4IzZJ3OV00Amt-5`vI^0=-(yXDmY|QL^7r
zbav8+<C^h@CYplTrDJ8-9o#A79@r1gi`54wuw2(lu2*-rEvO{HZVD}`cHynz8j`ls
zY*eoB6(_gxZ*^~-KTlD7U4c?$p4Jm`0gz-qZ-$Np{j$)PVTzRVIj+s$`Cj~yt7o8z
zKUs3N?QbXwqZdOt4CT@r8e0i&x<LCFOk_w-^%7QYNYNFA5=oS#U7N&6R>uKUup0vp
zt}YQs!TO@2gM6mRTmhc>D$z@i!}PHvET;JiYhK}<=DUT93l!ge!oDLZ!tsY8!Dng1
zyYaoY8nwz*IE`Y%H+n9Yv2eZLE*YZ0qFMg&)n$zt47JpyG56{Ki_@&t#nQRh$pCuQ
zkgfFdAd_frpbUFhcSwJ=uuS;7Q$#g+kxm<HamK-rEs90@7ghP|kFz|li}rU!nIq)D
zWO-TAa~G4@WTp2f?oEc5WJvGm)a@zKSrzd2y#(756OIK@YCIaliRBSUE|fe92fQQB
zO^x3&3)z024sW8ndu2jb`5KV0$sRVIb}1%SV*svV`W7l5lZy&we}&nQ{w2F)G>?k8
zF+g}Q2K-C0n6IB`7wi}r-ANxO!@e~5T^=iFE#J9?ZtmFm@ijr4^4kaUl!P);rU=Fj
zUNtF-5!i@n2D%y8gQzH5@ky2Veq2Ig*R}WU$Ed%o(X(n571kyG{&JDRZ?Qw|v?|Te
z0vs+|JxqN6w5UsIW&)4rZAb=#Tr$J+YQAOOLJT658@8}dKOY;&cC$96GKQ{b_viXt
z;UTlt98@N1HYNL=>pL#v%SvRy^i`k?@u@w6whm}2T4PI+Rbr2o?a9HN?}E4Nn^vbw
zveG&!$WAnU_BCpe)3_3&dO7@!6GnCWXj56q?8ERU_?GhOup|4_CNRuv^OO3{)rZSZ
z@x*@^^B%scwmo%d7n5CrxB#e>PE>3xMXMJNp%+R{mR;Y^z<t4j{7lVWv%%Fpq@d!u
zZ&;aeGNmLrSJEX#7j<yZnEf+z!~N5^jUFKx)ddYaZY~+Am{SeS$;0Am+Y$2TQ{o$4
zp}Vxb=bN3>A}A}KFA_H&Y7VVyEHVx|no?VNATK8AoAGJU+~XrxToPY>XwySSzlBdD
zG@w3@5nM{A2G@7KMg~4+SdXozYXTjx@B9hl+ei8<SK3n^Y(95BrOZBWe2zFSKdIaR
z7tyDDk$@{5nfV}m;$|&fXF%gXht6ovt;EgE5+6y~KC3;v$qTnT7Bph8C#NZKS_R2|
zJ#+ag)_S!YeWspOd^1yCymxSxtD;HR_mcO=acWrL%rA<=EON2h^g78|`n!LxcB^QA
z0dV9UPlFPg*Nl)i|Aw!)x-$Ury4LKz{Lve<?gBwGFb1^|tM5=vd>MP)H^GVPa+He>
z;{oF71g^M}Xtt5#sCGj$n=+GC4?_IpO(=FBxwo5iy2>WP_T|~k;>~+bNt36;HPSfg
zpBe8@skXktGPyJmBRXS1lQUxq91FKZtq}VXSnk)kJf>n8wQvx|?B0b9xxzRAN}b$G
z9D8F_j!a`J6vYzj9F+*yeNpl53@OTf$0eoGUcBQB(zwdA<lI19IS%QTBNw*wa+eE=
z=*MTuindWbs0xwpzV|kq5!mxKj7<X6AY@DyU@^|E4|lE1E!^{u$@QUHwpWsUQ)v0a
z=NUuxn+P}eAP%0|EM3AoZf=Ow-L#Ev5vf^jYR#uNA578Rc<|MGhz1D$zKDQIj>fV?
zzs2%Xqu5%LFkRM7_r6{6twuMl0(<=#cx>}eTi+&rGm~?YbqBx`yOOb*#LRQC13&R9
ze}eeM9)~j8-JS}aAFt@e(o^Dl8$zD!>AjoMFDMb(UIXn{7FX@QqUmr2Ca4nM{Ya^2
z6=rmMyHn{DC6p|>DBP~neH`C(HpCy2HR>RQUDY0p%k*ndi@SxF7kiEFIqvjYKWbo#
z{TroiprPmH%yX1mA9u`rqNyK6I@!svfE?Qij<0A*Xktze8N5$rdywMU{R#vRWbsy%
zWWHF97u+Q~v4IX@1!YX0246JI(#ih5Z5o(+TM6_8f)?olVt2&r-T#{YFy&05*lq7e
z{E5{n7-kCRrp9?M_a?^)?LJ)0vN(Pt-XeQCTqq$u^P@_TMh`dA2kA@WTxyK3hay~b
z-VXU9L41tJyHLS1ESdu<J?RfA0m%n4!<l;S_@Z`vhFMB{P4>7gRFTk$Q|Xr0El;u0
z@AN3dvg2=S6n8<X`XDu7Jw0dvT%#hdMGY6kUaTK@4|Pm7NvJ5}2>L$fL_iLz9_fsG
zClXW+&0oRqXG|(Q%UaU~(4A@TOCzsz7v>cFeWzhmi-4xVJ;W<J$Q!I0aWu`;n&EU$
z?baA~sT*6FypIU){Z(c3u`u-HjChPseP5BW_KbqJ0Pc70#+`$?xK~!vCrGZdBzG9d
z=b03-WqLz-1%^p=slLZ7^vXQBLTF5HtSC(~p&y|qc}l4UQp6q?v=Tza#+a_C#Zatf
zXUlT^k1zRqyB9WVm2s#QhuczKUISQ~Qn<T>h@}>){IEk8%@UM%__rmgoZfXnU6n^6
zp5g>8wXL{0%qltg;Z?}0_g(WSbK~)u`V0kT78>9|nT0x1OYWnjtvfIo2`Kw34NE?-
z4eV1hOrqGMW)m2g09@T-3E#sUCqm%SZt%$RAqdz)yRZ>N`dAtT9gNCW(ncNBOp>aU
zP&r8TG>n29PEiONzMj67;k0w)aA~B})vD1m#hm$MUWUz*E;g$&>uEN>nZThu1x4qZ
zzV|!OH{SF<I;RkOw#N;Wc6ONUCOJ8aDl^ejI1M~$Qx1f|hGo=D+56cyG7nnMwmcnY
zr<_T!a%ur=obp97jqHuEa{2EiEZpNcnWt*v!b=c5M@Vt@Q9Ihn+$2lbWC)rWr9B-l
zQdZKN>vJ@-Dl!iTbzHLGwXxP7Gx6h88w#`KOhK*vHc^^)$y~<qOHt~O&fP0m$`JwQ
z^xn*|?Dm^-89eWgeRAf?>uU1pG<4xrURkZnqWBI#Q)5qZmF0A#Wkh)PdPeDvyX#A}
zJ_5%~#>(5M%O@|qy;Y}VXVOv*CKKV3Y%W~Oo_5Wt`i+SiF=A<M<2C2{q^ao^E)#3g
zg_NSi4eft`bO)}&+sqDTTGsEyZxHD<4zf8vIP)Fks>`rCM`U|&at`U(#XUfYkUJ;t
zT;W;{VCOVhuDnKVw8<NYmptE-XK$WtHSQ7<7_pp6A?>lt)XD_#L_n<oL9c0kw!l+l
zy_^nOVcUB2V=h;Yh<skjG-{!)8#m3R@~)*ly`<Z9)a{pB7hj>=#{AxI8=w4)6<zd-
zgz3Wr<f{vFB$-ctGO$UrSZ7EhR+PxpUEom=|Ms<sMjy1qUff?CCayHXf8q^HlMfrk
z`{nn~I0k3gcec^?hA*+d?l&2Y@Ts>s=`k<CToD?uyX0u*XXx}`N8evJ?l-9pXyE||
zEgPu%>y;OZy+&k5ZO=VG^i5{&IL%Xi)>2euXY;lURtp!C#@v$j0~JBEzv)f2@aT;(
zL~#nVJ4T)oeK1@B0aBx|V@guHF7h}XVJ3wJR3Zoxa?hmbykcSgavV|a5G~q+k8tvS
z#~ABK?z=}dueR7$mrC^eeQ3P$_>C+xYb&@$eMoeZfF%5$Fm;2uG?v-n#^FBPiUIg?
zez=axbzrWQw9g-Y2z}}r2_BllW1@1S4BiTB_ncU9z|Esoaw12i)6tI!ysONRB#Ss$
zDs#}rc=zF1lZ#p{H#G7sp*@@pGh8zAa@qj@6Gv_<cE~;{L{0T?8xD-tr-+iS>@-ol
zatwTNcQ&{@zo=@HASNKgXFgo`E?XmQ56?)~Gwg~)9n$NH2$23z>d@c4qv7!W2;FZ-
zsCz?<SktY}w@drv`IM<m%sq@IE=m~V`(Lkm8C=~X>rEpvwm|HMn$a{3n{#5flv%+?
z0~dH8d|WEep={+Fw8*Qkdyq$3uY32SMLsCjg{fH{9L6S3rFf)|hvZ!*Dg(4~%WoeN
zS+*~jFt6Uc62!GD44pD5*J28ektJ*R@(DUskZHL!1_Iw^Hasfgop;+atuAuD$z!$|
zt;Ahak74HNfI{h|vg5o>yLTOka(+aV6ZP~?ez08q31WpYpa-MMzsd{EJao@!z8tnJ
zT8p@Gcs}WyCMtJ2n>6L0S=i0}W*oh|tx+oE>G*Ddmurn1rmS|Pd<@%?S#miR%k0H@
zbtcxGW8X5D4zGD{!`RKM<erAP=Pe%%Fh&Ezx-KFq9uqBdtB{D^@F+~LU-TR<pxWSL
z5_yc7E}FF1t4T(k2t3#Md{&BVqoLL_2TeK&>M830E+^%ju7+jJ&Yn8#?Cxje+Q>|g
z1mlJ|YHeSpK;fdnLXmsUWvf-IRIBPLjRoNtos%8R7HQ>q)Lt8&8VTm@H*3eh4^7w;
zJ79P$*XZW5489@1V&i;wo_c!eukqOwCP0?taZmhy_Ly&C-8O~MQ@nM?q-L1Oje$2)
zm5!Z3k@iS!ar&4^tBQ&B2-BsQGc~jloIqymg79KIiw<;c6)e^G@8)0_dm6HDvYeu+
z7RMHNt%U2=ZN1CNs9ki{uOx@q6kkVOBu>9xrfztDW9u%3TcwFnOlIs1hBRqSx0n)_
zHVu2>ke<+!9@)@fuS66J3T#`JCWlSCkFy$2s<`zxrKp-zC1z20l-jcIWL2l=ksdxR
zm|ak@Vz4`k-4R%iG;=(ci^OUq(ph<MVuYS^1IJBpi`~#bVvZ9dQCYPoDJK0mRD`ZL
z%1j~<oi+WAP%x&OX(X>mm0|ZkjG7r-vgR4Ze?vPqIWW-mWZVRdp2MJL3dT1+;}-q?
zKFPoVkZWl5T&`?vy70cr_<8c4ly?eqiu_Ux2IwSQk!qayn&7F_F9VR}MMEItGKpYu
z>5b7@9V^NObKvLo!I4@wf(lJyCP`3Z(ezhQn}%00mgI|57;!~CB%?gVaLH4srCPv?
zVu}d1h}X^)CGO5fJmg-fnw_O)hHUlezkj>>4#AD@D#Okmxfjc$O}k6;jU#X@I6C_!
zOcCe8d4B#t`OZR**}<Lnv-Ad@Q!ai5Ji(6cufJW#UUs%C2PLO9Dsk?^qSa#L2)KUb
zFq^Zjf8O|MaJ53o>~Pe@C|Qyu+V!KW%jXZ<yZh}E#cR<Lvv-pD&65@|jj2U({x4Bd
z`OtHY<3$A%(d8_*jRU#`IzlZ(7m=Qxq?&x$f;0Jd>=(=F^3rFo1(r}zl%eoo8*#bF
z+AQh!F5jAo&A9{?N&BgMrM0?~d?wHobL;Dve0{3GuKazzlNE>I)J(_cWiD+WH2u6#
zK=*n#8n0OR8rXxAVdRlB<;mn)9j}5uo4f*NHH==0PiW9Qv+Bn#VZ|67DO&twoWkRu
z{jDF^mVZfd{Jw5w*>o^^zoEkS|1|a6D(D<5S~B;%?y*qa+a@~nX065AL|u_2ANLrI
z6*hk+RN&ybYYb;+pZjXJv5y<b?jEg|&MiLVwKIB~LbApcO`)`ll;!8Kce~*4eMMR5
z%jh;jWbJJ1%q8tQcL00;xJ1aMFb^Yw)cp=S)}GuFI@I<g@wN=JQq=lK*{;|36$fx^
zPW3q-IW*-cEL%tutme1M<GX~F?3CJ@Lv>Rujmp-NM;%D|=<!DHS-qC){^^_pD7oKO
zVrNnd%GCTg&hJ}3Z+z~J{l{jgeg!oj6PdBvtH^S0;aP{k_2WfPRt&~x6i!M+OJEzF
zv8de|Qw3kwEschi)CMH6g;ookfyHi}ZYK<6bJNEIqzA6N=iW6nO<AwWAF!rb@mixC
zt?u<NSDr6Vnc@sk>**1V=gn;w?mamrz4Y$Xbl&Zj4_T^)iL{y2DHf9yY34ko!4Ate
zPL%4!TPyf3hus3yiAAG9<G8Ck3(43)Z}u&{@7u}7!lV(P07>5;?pW#D<8!|XKcMks
zIfAVD{!Xf}`^f^bQ~7yM(F*QYR_!UnR|Z2Z^t6QInTH*0!AfXVN=Pa9U3sz>Au}J8
zP>S$N_+n>LH<MLcqCzf3AqO;5Nt(}8f8C)e+XdJ=nRD4UeqxXgX41o^VHT>6Ys$u(
zJ-R6u?n^Gvu{fxfe8RpXCzGJ3Q@e>{*WBoht2s>hz7(Yq-p!*BxOMYxe!TEN`Vj|Z
zNAmD>)M?5E>!*j#KN6kc_yZpHFS8E$zgR{jW%I2N_?gW4k@QWGKuxGC%C)UhBE%%l
zWLdOYfXgu3mlADd-3!8ZvZ}W!$&moSSQ!m*)1e|4GVFXcV@_EvWCLNRW-(SR&e*b{
zu@r)yObGptsG1oYqPdyIct?I!$zi$<7%NiuL8om$VR&*o8Nk<29TCqwg8O7GU7U<d
zGzb<@vjum^mmqI@RP~#|+w(7zr3D;8c)AUm?nm;QIqGL)!i}dvZ0?nPXC5he{}JIy
z(>Qih)q}H%p&L^J=)=SHLcRgQZKb*%$pOede08AAW(Y9iG03B``#(fnHm+ST?XM1-
zhcu$?{)ur*pw)bM&H2C7!ucUX;&h{-e?DP74v<hD+=qk$YXH>ho|FhQ>0q60gLodX
z(hLMDKM%^X?XzLUZtD#j_`jV!C65PK&gJ`_xA#BEa2;zSS^1$S5#f&`)Tvl+Mlpfm
za$>*W$WveJw}9<?5SBv?m)q(wzjYp<?*svBx@dN{rdGoW9+L?wh1nH*@8UD1slRO9
z;N<~Timr2G`q_)m5Mx>ez~50oC##+=>bQuA_61_roCw1$R|&TPwG|`2B~7y&5qqKc
z-6SJ^aBi6H;xpqY`yJUlTa(&GI&!b#9_THskkul{1@9Do0ZZKA*9K5?y_J3hJNa|6
z{f(deo*`Q!=l(rEw#I*(zQ6ou>^$Itn+v2oenbzyeBZyl5NnAnhclHo|BEYxn@I8k
zTyT~{4bLxICjasC1jAw4Kht{2|KB_iK%Sbw2#!U(Qu_VZk`(Ddq$vyxB9yX0zoZrT
zusQ+m>big9VutU+6)ZcErPP0X*6{73-pb%HWCaP{d3oP$ur4ke(%4HP!o8I+M|iVa
z3VnPK6HFA5X@pg95~Aj|ilBW*Y)nAXG8-1_<>+4*@BaC$W5wWkk)&R*{XPd9cltqB
z^(|`~f;?437=gkoK&Z2Wh1d*dk_vCx{-;J4Hg4XQA%A`IC!pgJh89FCj;TgTH-V#~
z$VmP3a{tjYU0x4(WxF;r^VkBi?KPmaLTfsKB^-Lq%lJ<x?w@C#GEC3V@e=Ky5A``c
zsEP%A2t8c74jS!_n_$0tgUD#jgLeoX!+0FDy%8<WQmCiK7$xExKmP>SJ{nzs0?Z}A
z$VLTnN>(+t0F~|?V1Kk8c_rX}?r%Rg`hUez|L8$rAnZNZ_e$~e?^7nWjUGHDt|R)G
zgdOzj@L2fuT2u9p*tBI@Wv$>M_6$N#OfB+wJ)$Dg)?t^h-~v=$y%?+)BfXBn{bPUn
zhChAvH<(X+`{;guBy84x!<*?w#G!tb=Q%%k;^#K;s}d2PW)94uk~F<t-#@_b5&I$x
z;uvfjI~s}$5Gg_co*9BZTpDbph#k4u_7MPwTETvJZTQ70EpzbJH43H70TTJBodY!D
zvUYGw>jBQXB>Ry=q@>N0?cg}4TKWc=*+>=+n&-iYhFUfhXIlZ)YX|tZ=#v0o{qBAa
zj7(c#Id5+mrI=iK15_*plR2Y_b|}t`!BxEmBx!S(_lh0P1fJf-vG&)mVT+cvJ>u<f
zx-jJ*n`mH#UV;s7?*e#h>eYdo%F}49Hy~g~0i=?N0Ja6x!p&j)X=wn>BHJOFt;H1R
zuQk(%Gve*b@L6Nfq>cfA>$$jRF*WV5HQFHu7Bqz$cYq05;{sv|?h1DtTGI#bc0eX7
zV%1(fpC#vI`8)C!A(TbF*}6+0G&{{Z?PZk^BD$^*Stin~oT3nYY$7{gtc#)2RFDB+
zz;<AY@yH<R*U$Hw8cY9;$&UI5Y{e|Q02PQ#2F+#q@RE6`^e`14W=LcGVh9brQRtgW
zlISBOtS%rD8lheDAPp#ZRc@yL!>i}_rG++aWjC}vj8vf%9mjxuyOj=`{|%9r1$_{G
zfDM+cphr(k7Hw4Hm$dVMLVOk?ehU3wj*?rbe8hGBD6rKKLLcJ`3~S<d)STisC}vPu
z1RG|vQ1=k1Ja7)V-#{I>)SLB!EgW@Tpu>ICqKefJ?0FD**w^c8D*;|lq#i~x5Vq3u
zfYB2_%Wel1zbcqYT{RGSxM`6J!J$0p%N*310w!Z>Dq&%=DLj&9<8DE=IRK?+L2>+f
z^?asiKKV;ji971^5Eq6+%46OZH#rS@a2Ws>6B)?|!-5<LhpYrzsD<m&s<%_U5+u2U
z?@OBCVsUgIky`D&1Egav(dhmNX%|XbMhU9BgFGkO)sMJgBViEiZohF|r;kEmz>zE9
zyLvFRXG^6ltiR~p`!z%K=WZd_2d9}%*S%}M<5QjJQ4~yUqF5)08(l!e8yR&)URLa#
z2RXPKU!M3|P%;uV!{{w5$r)|n+-(k9mq*g-HNgGdV2P#oLp{SxUW1BgfDjSXW?dLv
zht}NH_~jU?@N*j7@`UF&(g^~aOz%TQgqG8sj5`2o2P3vXcF5eh?@TMq152pDzJLb4
z)Qys_*4Gdq+29CFMtgwS+9h$dK0D2XlXd2zC|`rg9`LXm7LP3uOU<Xnw0U)*MlIxy
zN(M&$2<uCH?2Cu4;7;}lb_;tb=mmkYoqimB7Xn`wIzXv<R^mPodG$VqmaKA{f(mWN
zZ`UO3gp7*`TFYTWYp6Q-iyLlb27%>0($1F6hJ<2`ZjH2*jQE;42oj$Fk~ZF|fo#@3
zHPh=r`i4I|2`ly$PzgjMnMq<z&Ryt3O-_L)8Q5G7U5!wPZ@VcZZ45%q*#I`ac{sKp
zIY#;zK=Dy#?_k6Flvz-7?3eY!e>vNnAcPPeC~NyMc#q?K{XGEia*UylHQN9vO)r<B
zf#h>@*fG3Khd_Vzx^H;+9Ef@`HFdN8hEI5%GaPd?o+Y}(JzQIlE`6$}yuC`88u=-}
z-&O;NAdCwbiNp=h=xrzliF~mNK9n6vh#fj<0TH~6qc^*>Z`~9Ckz4r-51r4ufeFwx
zx>A-(@tL|JXha?ZFtD&$)tBC*PRKBXzS+*O2c3EFH^wBC8x-YSlH8L_6Ua3LbiR3T
zH)sKjLKnAU&nA4Ux-2DLu+|)BAy25o@oBESZvj(6J1EC*qpSBE=`~EpgrfT!XG^3L
zIgeG6t?r;%nnE?p&Zu#x76?R;ckrBa&qQPz%%sf;_wGfM_)_i(HX!U_E@Am1y@0<n
z`=Lv(Q*$@#M?fPeMMnt1eX$1&?$)OHuM|5gZe+t%MCT!p*O5LX^w(?woXND#t%f~#
zEk~fRq4=}SC9w>2l5-<Kv1X+#l<k?Qi@Of>)=P+my@fw1{G|?h9nY`XU22YF(Ijk#
z<NLe~P#2~XQp;;)W>g2iV?W6PHSUK8&f0CE+p?XLmBL1O_8E8)kqr?`8a>uTm?->}
zOVmrU+8eY9>#Q~S3~j#?gcaO=EMfG03v`fHq27^7NYODvT?#}K_Nm%PKSjkN`aB3X
znxyi0%AL}cG(*J-4klVZv{FMefQImU7R%olLj~T7|7DOEdjX;<^qv=+e;nESMPezA
z5%zgjjHF~0BcdG;O30H>$l-wz3YTE|6g2~>5G|^{mWFacr6mu?Dp+2DJ4Y|$(T5OZ
z4u?>2CqSyg$|#!>QP%(;1Y187Vbm#Q(BkM9=FIyP2GFWFcBA?cwGW#<#%BtLmtxsz
z!g>6c!zK;R)s&a5c`hT3bHD&Wll9jTtxa1Cs=!m=G~O6;y<8j3O&RWM|K1Y4F%D;s
zS)NH$>2&yVa%>_wx*1=51&6Lc8kef)Qc(~3B-QxSrcSYB$m}=%9dq>O0Pc?%ZyF+~
zYKKzc9}BQ?tKV_Q1eR!We46bF>{7O%T6Bb=|30V#pgtEMoe(iI@mbw!zF2>I3RzbR
zo&j!?&7JOP0~lp=6}}0p7Gl&yzKUQnzVGgEAd@js=w^JDlA17*IYm(RHse_wdn<y-
zR$lE{HkZ{Rj*2XZv5w+nSgY_MB{>+bq@l-&s$myEjNFxWeC{=62QW1uh^Ope`pY2Q
z?+&u}+rH)eeLu-8g2jp^$9e(J2k|9OaW)yKiL!yx<cnZ+=}M=>ZR+izWPT)s83h7R
zEzPK!?+Z{ZIOxq_>N{Ur$z^7JjFSZcdCrncxzqzOEv_t%Ant)`=uij#(5FC^lR8CO
z@}N8ZrWyoebV1y$PoYVUppD-zPuL;@vTI|2SV#-H<NIA6IlNyuJ*Na|v->Z1fdFc1
z-4~G8VyB>nkK^TE(p&-aa6|6$X*3-nU;I;5N=~TJgYuup+MgoPpDz@Z9Fk*kG?Ojm
z;zj0ik%H(X^i*$dXa+S|b66ASLhirDC90DjHUUKIAC#KD_6m629?_w*qI^ag;I9l0
zL$m#8Gz9GYgHJxmfAtfpv-%vaf)L)aVaGf<W&J52i+dCPGa?1m4*g-?k~<b2`1MB+
zs2o;hPqp&OiUJz$(Yfsi19m$vmUae;t^r3Kx;h{bIIu_3`mv;C9bXo6pU5iY2c@$g
zP)k;vgPvn>RBVk?B*-2VWn2Wc&Q_!w?f<at3FQz)$byJd9%w)C>uSQ!G|`Ww=F9h`
z#Py6653=n2+z-T>8bM4v3du$$4CZmX+53P=n9Xg-;flKKGW2o(t8jV?h&M73)(>%`
z8!8Frw*67!v^fw0AgumCD{KWo0gW{YC;QaMsj}L;aSK6&1~FF}ATwOSc_cuE7>3s7
zX>?`LY?_rgJ%ea)DL60rTZgObuIc7D%?{RwMf(6{&5YBOv;mHz#_H+b>kv{ORV_<l
ziK4%@gy8>nzq^sG|4HmKKoT#4=yn1Ewd+&gorWTb(kA^es(C=-Fwc9|w}r<o775$u
zK^^6$(4#Q+UChqWAei@&<&+3KI#4wFgn%K50}v?L)y2w(VAHENyZtf?`QwFyr{Mf$
ze->OB_dQr@G@(;VTzTNLy6;B6Iv{ruH3jjIHP(t{P~0wQOo6U_OmbU0D4i%c3bWw^
zVbd*t!0P}Lq7btw8&1rCr)UYTWA>2{?-2PnB-NP$5>7m%6;P8YEUXb&uRKg@(@Tr@
zZm*hOTvRdybjc`$<29;#QpT+A7&7D{>{PKayVcEVO6?FB!I6^RiI66c>;>Lvk8tz6
zpvV$OAL)PV{2SNp4b2qm^Dhx8<T#3vQy5uHr`5LVF$T$ltIwf05d2cuz|?ifRB902
zX(*Pbd0XBw+WlL^d|{*Zx&7=G<T1i6^@e3@)Mwb7WM9<)ih?T=U4d_sV-N=(O-=O(
zrMflSuDz}W;<+2v#Bllr2y0Qe9iQPWM8dH!8bM?d?)L>k_mhxxu*dbiL-<#l`gZ^@
z3?Vs*rzC|x^M5XV8Rdug0fx2AB1J>8)qO(p1XRwlfxuN}00c69r5@#mb!%R@TKmv<
zIk4@Awak41eJzBNA3-K!d+r@6k*I-0#!c@s^ec<|VJ$HaEP>|m3$Vug0E;Ip)%)EZ
zL|s5}1HM*Q-y$R<{T=FW#9>AE`1qdx2+RCPa&l&}q{SNPkyMrrZ8;TOzmRW&*b_2=
zP=xq!h^b*x?zmK{c<fyS3k*9TaJj;?2;h5CqJyd$TB8Ui`iA>ELWIyxz&WPqQjPgN
z6I;+cTmat<)z+a3GcLj^)C0Pn!k{YM>*I__dTt2ZR8_i+oJW!Vz6MH!RcJLvJE}tH
zJxgYZ$$mZ@fXMCWe9>Id#wExRKAp>6Y;6HTYU=P6)IQ2TOhD{ZJjEv5LQOU$!}ulU
zMpE5*aplAon1ch3v`UvbJ;E}grsK#321rJXb(k;Y$#&Rz!afzyN1z7xZ*+Apq<2S*
zQ@gRG?JkFUW&{TiiER9(#@ULtbOm2j9{?Ef4z&`bCB!BhlJN!AE|djDA#OQ#if<25
zI4NwK@Vh;AShOV#xTmGrk^%q$<*s>6A4{oAS{K}ql#wtWWXt&!k+=juL)lpqeyuT$
zf*8-~vyO;OnL>q@-)Zi<{voI>>CpVER3j~5CMg1jX(pe90`&LvZvD%;LIw*|{)}#^
z3j}<p=4uu);J0h|NvV}lgsz|rFs<Yt;DQn#ln17RUDx8=9Y~2BLrc8Ehg~4M@6J^q
zp&llqV(G2gj5{*DNN>M>i7auU2y*CJ<6_S><RijW{H<&O%nr1vGdLe~DlS7~0(bYi
zZyTAyI;_;RB*BnQe_<K4ea`Bji*rPmp<-;(4u^|5DwfdLm<WBuON3ddyYg!k7b>(t
z`_>u`Ad6wjtIM_7XM+VkK~Y+fQQZq?Axqj%=+i`V5r#Ay_y&%0&%C|8OOZpn``Afc
zW%U6w@KM-WNZ36E$I^BO0*hwDr1eU!8l46M(*e~`(<7H_whd@_X(nm+D43}+?Cc9_
z|9TkAF&sk4Avz{shdd<N5S=ZdbHi)7{FQEoD3p+dU<;&t?fBc8Jr{wqy`vaECr@`a
zKb?HBj(Aqb0ng!#lt|(f88%5`8-)1Yt@IxI%hAyM3_*DF>mxN&-@!r$XX#OWnB#C9
z(pYZc7><)Dco$KT(GaF-*f|4CW60iwTj%2B2^;W}LTU673htF(qULvPSz%>ue4OoW
zA}*JN6*6aA5E)k#XXFrVa7c^Dp6HdHa=#*Ycp<A}AlK`T$<v5!C@YiznIzP+P^DpZ
zM&CNZsb1<~kayJyfn1Dg()!>Fg~x3%Tz%xRAxTL)-S+n_jazaBJ$En2uhr!UI;#%K
zj2zG0&i!inIj`MwTF<Q&)&S|O?fE8aO&;jg;NW+u25m&GkV>SRKuaG=yP9}-hv+2l
zUx(Q|NMpMEK#Zc&_w6ONaT6d{&!xr8Cb}{JrOz7J=Sg+|Xg><LQx?9|PM(ZofBm7E
z5am6YUy8f-BL->|hFBijD(s#aP^DsS5S9h@C;v9R$vygiT|ZAc07u|R<?Vj4-!mox
zotnoR2<Tex?LJ})Yr`z>6B37yr{@h*j)InGsuk(n-0?x(4_jq`7G!qEg0Q*K`Ry;;
zmH&*r{cPbfKk>-?Bmx+)-Fj95YB6Tesu4e_V*}^*VkaDXz<F!11*Z1a+9i!O*3e@P
z(MWG38`8EGjScrX|M`Kx_$f&aqGbj-+P&XHMoH00#DNLwYwt%bLCfzIqW`F0L;si0
z85yxqQ3x&z7#{o#wtn%Qf8-NC-j{?ZnT&K`o|pcNQhyn~{*M>_U18^cy!gxFY@-xB
zSeNm~^=5zTmj1Zbewdq)4#oTayWbj)gc7dwwW>j}zuX7@h_jGOpMgnR^ZE-Ip8fJh
z{=*9{>M)p5CoG(Q*_HffHRWG^@>_xd7%-e<{Cnf|`y=s_J^&56nkaGUpL6H$U2Bmi
z48V+t+aE32|8OIJ{v^35q?0?KzMK130vG>z(wP>>rER7QKny@2!CK9q`7dwMov$v#
zO!35cPLFl3vW+Ao96d@a5MEmM%C!pqccYUa1m#QNR!{)7iTB9j++^FCm!%ei%|El_
zM&K`TqW|&XXj14uL>zm)`^cq6<ppa=M#R~z4TJ-1{(mm2E=g<O8|WtdxTsagY{x+R
zAU$ZK+4DaS!~eF%9#e|ISm%O>L1t3Wc#}0t#+m!R|Lu)EfO+5C6I>hj%P#AGdLWHR
zm(b-Km(!;I>Wb&|V7Z*cE^$f!Hv=ixiL_f6YWbD^$BO-{Z`S<3Tl=@6`@d29_i6Wk
j6Y}5om;e8@w$7VAJ*XXtx%k@#_;dEO`l)mUQ~&=5as&at

literal 0
HcmV?d00001

diff --git a/_static/IT_sample2.png b/_static/IT_sample2.png
new file mode 100644
index 0000000000000000000000000000000000000000..27b0bda7f9bfefed7dec5d4df4460ac977a67420
GIT binary patch
literal 282807
zcmeFYWn3J=wl4|<LU0J~7A(jh!3PO8!Gb%1;1Jvy7(4`b3&9<N+u#t~A-KD{yS>TY
z`<%VcdtdI&eIIZCx~FTZx~f)Jty*jSSFH|K{0PE8B|(LQgTs*iAfXHghl&XYhp_bO
zB`n8H%1s59QDh-5t|%=oPNnEzYieO-0tfdYI6e_sQALZ;=hy>~`ZWPNB<^=IRVtja
zn0clwGAT6*woH(3h9QVR!;Up5$cJX<XON#Iym3j8p@CBQG)83*<Gx|rxBh1z>bm=p
z=jZeEB(80b?NpDWde<XQM+!JRlTytTc;**TO#RUkHy&JTYq&%=@IG+d83-TXdNxf$
zwN_W#;NqF?UBTYSqU%?AlWwKn-p~4y0f62uc({P<l+KYYlPn0?Bq6Zk4KkcD?;fvi
z3N6L9OPs~)YSFmkaL4{PSmBOih(^C8<Wvz!O-SGnITM+s;e7^0AH#I@P=5p-1XP9x
z0q`<eq-JTkxI!b(aN>r36HS09Mn>fB*cAGg9q9WU9qO4J(sT#DVne>puAv`|Jv$0M
z)f$_A$lqdrAD{9TB}la7ZG+kMSe4%fiwr78vLUrMjhCNFF8rv<;D9NE97p;-?i<Qv
z91>txpO{DlPyv(SO9;|zOSB40Ur5wQJc~h9-goAtG>sn=tk?Odd}j~fLZu3o6y`9L
z)U|_)-i%!$=mH>Gs&}6-=%G1(`!cpni0i)b5|#g~g1=A+{;b9^A6BF}&R?~@QFDqU
z+%nEFdgf*yQ43+Sm#PEmMXds^&9<fbd$ux=9o((zAXwvdzN;$eGgN5C8E@gyvBQLK
z;i70YW4KzdT@u_1CBDb}==hw;EB9+9$Zd*?*oDXC3NCDUx^Mj?LD(LP`fk!<S^G8I
zQC3Aoh01tK#t0nQG_uc|<g27BxQw8*`vDy{W$B|F+ch?WQCpwa8SwHMXz0Gg<p?gn
zQ0m|TD`*JsiGmPLKf|y1szwE?w-B))TZ&RMzD#c8QH1~5Mqvj>_7e1-iw|koUv`D0
z4&KH;Q1s<pn=Cu}lE2a`zY}bvEU_O^6vpUUVlLRYXFib3&x{Bmfv+?1J|IeGZvNIT
zAbRWjHiM)9^A1S>B{i^Abn&-m0U^oPCDC*PMEZ_*n#er9W^Hock&xSzP9S_pdWhR?
zNGDQtWTgREZSp7J8xja<LK|Qe2ayW(75ONr7@NLbsyzzv4RJITb2Lp3dZ8pqlzPsK
z`rzIu>J8kNftiLxM9g8N^l-T0s|ExH>V`a@-?AgK%YSznl$QgT5}jknwtX>Ttti#R
z<iSk#`xM67nP6~jpzs-BXlry|acs{=Ms613*p0NZW^YkvQioBOS|_!@ughDC#p~Cz
z&V8ET<_aN73ms`ATytH)S{+&QIC;E@L`3qVXp2|@eV__@b^DdG{e3&J7?yz#yNoJ5
z3GSz00|{Y!CVO0aKt;S;>Wl&zU15|QY+e|ZNh~l5Qzk?kKDbM`7iDC8&REG}2fZ&S
z*Ur>k4Qe+D&<jv7R;xrxl1~FA#PG&k$BslF$1W=~;^Br6()q@S&weP(*3AHBLR53!
zYBM|F-G&t5G2<uWdEn_Y;lzi<Ys#R-oyGOXMKFERAW^1&o6J}hU*FFh&m6Dtc3F)=
zDf8#d&##5XKdlt`3mhaj>)+~4N=$0RmuY7-fG+*MP~vBjD9@@=II}tfE@^7gzhx(l
z@m0Um<~U?83FRry`?-}tkPVe6Q9i8x5f<@VeuvEcr6=wq<0Ix{RYQb;?5B(tz2^7L
zphy31bPNE7uT->DcDHr-$PaXR{UN6z)*nMZB5T;S3A7!n2`mrC8Vb1I`SWyGzqFMa
z4a?@DFz=Wnn8TSHsJ=FL{Svvevv;!}<m|Ge`<r+yci(+CZ#Q8tZIpZ>dk?&Waw&5;
zyx07>fI~A3U5Y<Dm7}<<gfas#SvGn+M#;_29mLbiYyQs7>I0`S2|Z~736vO<beA}Q
zh?Vyzulr}YccvuyB(}Uq>`$K!zr41r9L|~Ds4x7+)bnnndg&sXxNC6SDq5>#menzZ
ze;l|7$-l)wn`s}VnamnnYoN2*HD;<_u$UFX6&yFoGdDdPY*aU9PEadVYgQ}cUKE%S
zSZ<iTn`Ft>LaBld?tQ-;>K>)<FfAJr*HN1norlb;&-1zvxBwev8g;JlJVfr*u2`?i
z$XUn{F&oK__?5}U$)@?C{9euuE-`%Ed>8_x{0ssb{L3a|-&Y4Q1_#(Q32hVA*z$FS
z>jmn!>L{Ea2Zs7t7Y{S$&u8<hkgBeSj~sdJwi~AF27p8q!4!=`#%XtHx+B^n`Fh{<
zAZ4C<FIXJaN%pb#<7>lfJMX_<BSV=zknU5k*4B?zHpVxL_DBzPR_{AUegV45mTYI1
zN4=At>*1_K6@Ml8G?I~HMym7>V>)5#;>=J;V1}UoeC1AJg~NhkgW)TVAP)ToMCHcB
z58w&(W&b!cRo=6jw;I~+o$8;AE1d5zT=i-`XcLD#dZITV8ku$(1O{1$xD#0uu!h@&
zDupBwst}%2PccB>s1xuq+ej|dxRtmWKx7ewsp$>$+k17F>*u$wMjTT#_>PAm<F~R~
z`dS$^R7P#P0WxUyXpwY=xJI~G^j@pq>@DkQrBiZ#<`_^U*QQ>_9CraWu}~L~==^+*
zuQy^gwwaEYbW~od2&oJx2Y$6O2sm-p7S(p13%n@5c-%E2t{@Ii`a0M-7)zt3)KDNK
z%WEmEUOS_#%b39G@TuB*wEWG7N&snpTKZy)xHxI{$HJrwWDDC;kd4VvPZvjMB&(7Z
zjrLgmgyHU8?hCLNSOn0)vd(I$Q`!m{I$Odo#^+|vjF)DfS0^rgt2x;^=Y8qRox{Cr
z=`~Q>2k7&z<#DmL^!?bj=hs5{;9USvWr@RE`m_x#y#ozKwo^7v&6h1#Jn7)H?tH0`
zx7RvN#Aw3M@UDC;e8{f|?;~yG;kTZvA+FV`z8xkQVMUfvl{u5SgMPEJerpm(5Z^Kn
zwVRLU46NKc%sCu8tUOyeOCXmL<T<`kj@Gts9BZtskZJtlIT@JxkSf&J<if4GuS%%l
zy~MiUS=QZ}S2&dewZgJdI7&0~FuHPi7<{O=NuO$2^VGi%y&AxjATuv>EG-f&@U&AA
z(gl=I0^c3~4(|{S)FCM*`9iWA-Ua%Fvl~s=`sT@D+x+(^)M(HMqbEF4BQ+z9@r&P&
z1Rf0De4YXAHJt{9muXg(&Xz)_MVDpw3cKk)fh8{P<HTc0eq2Uo=fe>xxap(<=?*Pr
z)&}z(X!e9QH6s?s4Ma|-XLPNygaVr6W`IV0Vdt*P{=TBYqM_-OQf_Afm$aLq(^}k5
zA3w=Yr<T<&*wv5D2{qo!MVgJ=CJXZd7S|RUTm&y%u2XjeuaAZsnHzgruOGa+AE&V#
zD8qzxA79-U>{OhsS2D~geDUUeZdTfQBe=LfK7ae}Y_~&LZ+SqZ^7&>=|0L|J*h<tk
z<EVb|T*cqZpI2|tbM#sH!NIBEL?K?JNH3{b;3;uhKefZY<FSmctY^u}t?1U}SuoJm
zns${|Vacq8@=@_Z>3rQ+VNiiQ-ADv0jP1$qS@~)5!E(Pg!#1*;v!}aL7sWLTUsN$5
zHWuEkiV{u+2sc4${><)LnVyDien$D(@0a&lUz&$l3Y@v<keg1YF!C=PReAO7dL?q(
z;H)M%D_g*H;gV~X0wp2YT~Fn_S;~=ZIIXXDuQAs1r^V;$&%mM=H~Ei*j~Xw18m165
zVAgAhiKeuvygb}nSo#$lB0LEk5-bG|yM*CM|2r)O&j|P8ANdGyaDf(Zi2r)#Bkca?
zh=N^z>iqNmBIX+$GVI@L*yWai@Lx}(VrIPfuQb9EtPJkGinz2i?5<+uU}9q9Xm0Cd
zP#`}H%RsUFpydb$M?m-If|pi)a{_CB+Co*+NmE`<z{u8`&A`~!(1gv++U`#~a6mT!
zSkl_W$$-ku+RDaJz)gteA5RFt(tol6G*th1#K}^KMpIsqO5E1Lgo=laosFGF7?p~O
z3g}>LDxfSO^{?u%e?m0oPEK|L0D!BjE1N49o2`QxfP<f(AHdED;N)b5J;CY-wsA6W
zW3_Rl{bwWp)sBRTqmhG!os)&F4b`7^4Ge9aorGv;{&e);$3N*bakKb$Pd1MKdM(%s
z0{)Z$IM~<$|J^pMD)3LPfTD$)iItXwg*9x<V0{Sla=ioo<N5zj$-jI2kD8kQuF1*G
z`#-AwN74Vgs+yyTgSf3VtWPK5f6v#y`u)$se^mqm{=E8s48=dk`5(Ekffhyu0{(l_
zgi#|4kbGeCNM<3SpbERgxa`jXZwvcj{O28(hTq*HK_T#igA;|5mUyq~27i!>+@vcx
z8_0nDiaH=lQ5-*~_BTQ>ws;u=DiI=H4zpT_DvCV(F6yhXWT#y!Y-$<aYQ$HPRMo0t
z^3-v5E!RmSTEl5VjjpaEvmVz=_vsHd!?P+|gTV2nB=B}BTQold0|E`zU!A|1UO7dt
zea`=2CaH`lA^JBb5S|J1Ep3p`tG_u#8EB}w+CIk3g#7%g`aZ8d*einnif-833gEzK
zcggBhSC#zBCPMuCU;ORR2g7JD$EAa2z5T1Hqe4(7Q>DO$$N8Jn_oIYph<_?$1-{x}
z)rZH4TFW^5yE*rH<wOIc{XinPq#)!ko4`YkNB%o@mB_$_(f*l1#!6D{Z`d8yYYtxl
z@b8uz6+&u|4<53hk|FcYzhoP#uJ@aNH|Ov;I}$M3XM$hrDB=HQ6R@R@^*8GrHqHO%
zOZ{EI+KSNi!n*0ukKBpB;d<Ch`+k(n(5bbtX*`5lPV(`>T8yF=UkIf1n)!({-Sf9B
zRf2MHI6p0P+KF325|*b(cCg&qn#^Z!SR|!OiTig$?F*$ZU-3h0U{cAe{=)u4%-(L^
zy4G1G|D*L}(7w@MGSBZR#I1C1uj^kgu>RHP#m?waQCZ8E|HDRqShDec{^g^NA*I5I
z3H=gw$?=zBWvBBWEQ+hIMEF<ZwgdT7|8`85gW&J#|KI-qh=~6?{(sMZKa}lHLIEvy
zTxP??Ox$-@hx-jqyUOE&?sec-z~_haqx$VM$7Qek5>ig%ARUX`_#X^*<cx3L>X0*X
zkP_i<#=dNipZ?!l;J*<O|MYsdU*<Zk>nJbp79n`VS34?2I(ZR6VS|a_y6BuJ_;@kC
z7{{cdizjJw+C}KPo#s)h11m_T^&w}(`~RW$K8AFs>jlS1kKNqF1#v8)jh3goeO1+F
z_e+O)o5m6yoBGWK1@AkR+WxmiIl|NEgp?i!9A$Wc+|9!~Z@auZFnFCJMV{Q?(GqB3
z%_z8T#%mCi?!KW!c&l@3cd*mq)Cn0Qu~08evkX7pC;i_WAVpT+JXJE|<Us|I#|h7i
zQP5G-`7m>ls7pUnew{(bxBA<y<dMtW@gC>>sggOj+C4|3?l60g!<s}L&&$by?p(DO
z6aAD$c+Xl@rnlAiX9Ivxv6^9l{W_&=Dg004kq3yX&`PcYbYZ9Je-p{jic@cUxO}i-
z@(2?~1%AKCN{jXlpEv$jwM+&ruDMKC8{f(OQR9dXewaSmwr$I8o{r0`NykVY@MAL1
z&>?i*X31}jKG|wbDvW3}Zq-UPLp$X8V76*8MwV@tUhcah-dA#$nYKmy9&*WPfjcY~
z%d8K?x{(w@3uKOiI9^*`s<Zv%7BDAN^LFgihq#2j`l(mqXQcWK#5U>|;Im4vP@h+-
z?>{fGoj2Am0M&qnp#Np^A77ihceN4fLq<#Lw|*3Utq9>B0S*aPkV-}iKHaP@-fpKK
zzp^w_AuHy2BGk3>6<qy_yKu2P?kNC(&KMs3uq-`oQpRoZ*MqG1B1800MjIc_1{Pn-
zviv96LVw7P`mzrrTlRreUm`teZ>8r{Rsg0x3Svad*wE^2N<r}o`~S-d$vhqLho1e(
zW}?1JQ3+K0?I)IM=fgSfOFb^#rnL~}g^C`Eec^`_amtk8JU&h9hy0JJdlWEEzMiuz
zt4|a1Xi(GmG>Jj-uAnepj5NBHt?_`f7Pi{)o4hV3H5NmdOEqs&uY>#5mm~!9ybt-)
ztEECzbw+BWG}o44c4^ADVaW_0n0Z_ro|yK(sKQ_r!(?`bQ{(D4(uH7#w4Eid`@?oZ
zf!MXrC8y6@$+in->7H9#FdaSJ=z5G$HS9#w)wZYHIMt+<TIT+f3<iP{XxDYQwk2T+
z1>QxAq+dBGv5Q}?85$%B^(R;xX4+X>f48N<$lHIj^A!bm;r_Vwm_Q$5A6R3-kJIb_
zt3WB6Y{~~2lk9t->zVk!&1wN|_`ZWwfm?hqjyX5NPfKl#8sRtp1ML(d3WTk#B-^&^
z>VY@oCq7*5Pmfrax4k@yRS>qfLcSl91=Pc&!pxA!^TUzH@_pL8K_K};J6d8ok@s=)
z^@7OLDOMR`uZNKPg`F&GP292=acfz_-rHK(nz>)r_qxq(JZVQ;%uUqWd%9nKCI~Qt
z&4O!wii6DRoN2Q4R?FQU^L3f`W04)J*I2WMYb{I=G}^X4P6?g$(%NkFMlZRqBA~IX
zcwJ;u)?ZHRFAKpWxYro>lOEr;@af&d`N(2+q%a9vA{p<Jo6&TUMp?}MD#Eb{jK3pT
zWh>J79Jk~^M0eN6$3Y!1sX<BMT+>PELg{&-w7PiH2V0|BE}>u2?pOS=7IsFnJ%41R
z@OxZYT|XXsFSFJzIN1F1fwvizBn<c19wAKv$q1|W1emImS>ZKGC6vIxe=Q?^&<hjm
zwa#GS<)QB!9Yeg!5Me;9KF^|)N_u{%UL|FCh{uCqO<cu|$0}s-JIB##Af?N3%Y7-p
zKTdv+k;S#Nyd8D5q9J1PE#}tHF6<4j)jy0Q$UjVda{UBqDoifKQWVZx(lkvnS@>fq
z(}c1pUwlb(HetOab2jD#5Fi!57lCmfRe5UXd~#{a(5KVH{l&EV<2{%8{ruE7Q_GSn
zA^yu1<{Pt<8v2K9EjNa)zu=MUny==HqSN=-y^O-|8um-iL+mBKJU?E$$^x350)zD_
zaEO6!<;Q|TFr@`{V-S(Z=y;vs?Y-=IjlbslaN6Us{~Uk@gYSQ=Omkd>FpO@(Ox#!9
ze9*ZsML*j+8WZ)s!0+|Y0jZF)J{y?A#S|$LmN6|OQEP-Ko{2WfK{1KVC%MTvweXq#
z!%+2gf%U_Px#To1e^fpGk(ro?nso}MPhDXOqcQfQKwl=Jf8|`UR+Vn8&0>>tW4wmm
z?BkJZ>v@P<Uv>5iA;^s#ugyZ8pCk|uS?NZ{t_|tYr9p1(6O6NudPJTo8n5Oo{|Msk
z?nuh?i=P>GL^i;S-{P2?xiA$3er~_CGZx)WGE1{JO9SUMK3>f?r|Uyb1H?0<q{X1U
zFpVX+2%BIsJthGP@J94eV(UXL*8LxqT6Usw@$1<u#c`P5W`MPpr9Pl00=7n%ncug-
zD(P8ZOEQz8ua<YdmaVg{U7R{C;onZF%Dc%f;5O_GiaB;LxIe6Q^}gw#lsU?*4vGhi
zdGyCI=`qFFm0NT713m99PG$U&DcmwNN)N_zW!oa9Hly2-In&-%t$n7Q;cl=(+LzR<
z%(2N}QDW<xMq{mx8WBnNx+?@IFuFu1^3Ivlx=tv8@U3GPgX#i{l@*B5jvXXW&L&h8
zoB>TIei~HWBK>gl-r4A3ev88Q2kS2L*1Y<{>&NOIdL?V0mr_dWw@6&4w9Im+cK3^`
zBIaQWcLB!y+ls=e!k!Hpuc?MLStNsRz8giHqOMb?F<Nh{X-sL41U3=1kCF7qg!r#7
zL8im&ymLYi^y#Kc<`--dg<PB#w|3QcD|nh+r&+J-F?c98?pS}<%6E*SQ;x)T-R3`c
zJwDx3EpX(M9Yyf1W|#))6@8<|>UzINZ(QwAYB^onvZ!TUGub4kaa11xf$^!rH7D2M
z3<^VvqO+=P{U$+A79l}JwV8$jQ1gxRK}F9Ul44o&)sIG);q^e{do@UXTt41v^d1Kd
zb>j<zXu9peOoh#UNp0!lCcS%hc4v0<2d(dJSF>jK!oo)6VHWHy)YGRZ6XQ=#khdY3
zuijz7cl)e|wrX__iIH|!7umDy8S93g8j%_0U&xnV47oX!l*0V|Q^z~EP--f0hIO55
z9F%Xwt)SE*xv8M+e>h%5seT*aAy-sOnh_$sFJ9{+CnzXqxtTGF%ze9)dBWqgqbSlg
zUu8DZUZnYxZzD>4p=v~UKVGAJ&sU6eemCplVq4^SR$<{30hNhy`|;4$`&`DWspFU2
zLguTW81e||5?QwT&deF9I$|RCy~6ypNDoO$dKEz33IZ?qY$MC#;WP6_s_WA1qT70S
zp^S?G4M8qRd#E8+Kc93@jsC+Lp=6?xsK9y{7X<QM;xhu~Ko~M5q+DbfhT@yf=m421
zop4hv_eVM^4_HDMie-EsfaO1lT)#39tH%i`DM}<o_ffUqRHZ)-dM92hz5BWevwT+7
zwJqps$3F_*2Ak7w+lTkkp0b+>*_w8>HO<yY&5HS=ORDsrEI&WF+S=nWLo|8C;1eD$
zrdyvTIeKx(2n7CEsAc9eI}jCHVl)F2G~|A4f@{WgMsz)=QAw{x&>muOnsO^k<3Z(7
zr1wL%h85=VltZz)J^L}R9g#kyixJLnm;0!kAoPr3S=YV`{}Cs&vMt(0(=9im`wTn6
zZQ~ytT-$ZEACHIpeBN1gG_gL+nHw?-pOg$Vh}jpo*3I*oF0(6{h6pWg@;=R8rj<2G
z&b4<~+nud$6x!TpHr8>m33Or1-aL=jWB85Yr4J@|aVI_vrJ9ELxAI?!UhmUhS%mOG
zO{vZl$vXRYVT$2s-;91=(F!vZ5QK&cW^ot24G&<Ddr3p;LJtz`G9-P9;9Cp&nH2$i
z=%9T1UAxJM1=Y~71!~%TZKqPNy<YOLy*~DSp0}=B{dR~iF5WKRid7%vz@Q-rUyL&v
zDalkL)EMzPQYsI=lB=Iill<#+j&$tXqxRpu<-NyKw2HR(<L=ZGi7@+;=r}CWvI-Yt
zy(Da3b$!R|tG#Yo>*bWrW0u@Tue*zQ0m!tBKE*wc|J4Ty^GBy0Mr-?}Q^il?cVH!b
zy#K`gzI?uz*?b>zh{)bYACxv8h4hRnbaE1hUJ9|OEqY!rdab;9!j|!5#LgrlT7X$`
zN;8;YeR}l1N}>|PfqjfE{xswXM}BKm(-qx;2u2R=CM7jf4%7PH&DRtrlEsFZ8{fuA
zQ?vD-@1|GBpL$(q#PIjO0b^CuQ2L$G&(;b3MePakkUO}c=2A)YyJL9BDBHH)_d2$t
zVcSY$x3E>StNzJZjWQv3XpOo^&0hxwI3C}ql$<hrWW0J)>x^lS_T(tRe+siQt`jJ4
z!#Oeg6Ewp*IluxFc%n+Z?Kq}*fr>lmY{$UY4@2uSgCYK%00oy4HK^3w)sofz5F%%U
zL;F|3O{~DpG}!2s6&z%AzSoOPj0dhCiC8^P>U6uM10(!soe5B*_=-_0$%laFgRn8L
zl;`m0<6HZXQa-FzRHSZ??R~x_k$QEW{=D!<ebKd<>KGQks>J6yHKr00t{X<1*k%k@
z9$7p4$!uA&-G1#!rM=q<6T9h<heD6T&FK=h=i#-L5$~qfdGm9OMZMO!X6*L*t-ZY|
z624!~HAY+wuFGB{CFhTsM?dG;T;`v0CcfBkeXCb8=uhtYibY@l+;Ww7zf>t(PuE&2
zk?prw{j`<I7bK~(!S+<{2Z8B#ceEMmF-D9mxCnfcxm<rtf!={nt6pY&l|w*#P#gJe
z14+-6B`}iiDE(oBVX|o~Zp1ho8I9e1JC{%MM-zJ{#JaMNK9|=yR}$f(t=#IJ>EL&Q
zbn7i-0U4mSKSx6KZnCs}Ik$Q3BX1>C#G3L&0BrMs627>NGwl;hL&adD$?aNB6!B?B
z4yK{|d-00E_q`}dlp@b%9Tud3nY{q4J=y;OSfgZq;VJ%@oBhiC1EbKJhe`FdOL7V;
zc&b=Il2Fghz?cAzR;4mnPK%LKvMpwNuu`o}M8zlA$}aCDtfW6|5J7!am;s9P#W*D7
ziCBe<rQwgcOSpuwT8mPgWws_VSgGE3-egAbn`VBGjgv0BH;yngt%CI%njTAKFOB52
z+cX|B+BH81QfG{r+b|w-Xc}!A%x=&ft!g}Kur%%u4cHEn4I%{Hta#hibS|^x;@t*#
z^MG-h&Oph(&REu#-<cF-u_qyz8Bbw_K4=?-bJM7tF`Y4&0hKo_FhvIMJ}J#u2iRe@
zBwS}rRR$=4%Pog0O?09NIq1^ozB<5%XpM!*ofhv9pMA$DCP+hJ0Cx)A#Mgu+wObia
zDTTIMtDpE?x9H9gI0oFC(s5TFPM}>|@)vTaZ}bJ*{lNbz1{Ax^|Cm2q<7-QaKZl^=
z#0JKc$^&ZSEduxZwfE|?M625)la{?mJZ}!?)t)aueSX^cMSERxcdPwiGf|cObYGz0
zs6u9?v1)^eRZ$>ZrqSW8{!XY?S;kX-5#QFsZ@W6Mmp5;PVm4uGRL8a%-)2|yn~zq{
z8BpH7d=U3=-C@d2wO`}V-$!h=7%|G)3hmMqLvM0*kMXoXpq-e)n8nAX9}k)9CQD-;
z&A0ea!6P`ooe6<F*oW07+-iTdA-mI!?-D{JQaDfBJd=q6lVZ&qhcGDy3I^@xHnsdu
zj5H*lA+AMM?Tg*&yq*4Z!SE3cliA3s9uFE+^~RoudaWcJpQ1AMlgn(Etw|quv6X>+
zn3=t1*dKiJ546e(xe2{mh5@Qw4X>w7fVcfA>v(w#elglNvkFMtRI&x;K9N;(f|I6V
zCk<5s*9q3ZJI!|wZCG9_a!mcKEjMdV@pK8A?G>ENM==D*!LmO#nIC%IirZN!$pP0f
zWlgPu3CO&vy6o&;yhOJf5xUlGd)8jVss&MczH{4*%;VcBg`uqWnhs3CLEBl~y3Y(&
z<U=ka_O)FEI(2~*?ge%z0}jRgCXK&VW|p`kVX|ysf93$zkr~B-c9m5$sISOA5MM@J
z2f4QGC21W)nj6i-MbGlX7VnaP>VqMLJ=T*6t5F=5`imZNXR-tl|7=J(zN4MyG8f&(
zN!6YaD*^K9JB5!opihWpb=fT)GHQ_I)aQq7k)=w*2mav-RRPG&N&DtMIY$gmdp9Sh
ztF4-<xcv9rk(tNQ#>YD&`^9v~)%i$XeWPvja&0&opdkttRu#Uu9x1$;=3!dxv42_H
zl<>SJd*!P0V2;cCIaL|0R{D8~qsSL);kMl4ITdSRqLA{Bd!SB|P^S}P*0_A%V0%1Y
z-gV!WuY9K_@YthM{s()PeCk>BeaG>zy~jaok&aP`T4>rTs1@nHfwgwN`ZpUL<@d`j
z7T2Dk>jvAFr|#t+FoS=gT*&2uK)RlH%{Th!_BqU@%5>OaprpEZ-hTQJZJtNr2m5Sz
zoo!Cn^|J}$TH|jTGy-7`qi#9Z#9JYctJ)_Lg0HU2W}2&BbA5HZ%g#n=-m}fZSMt|a
z$;aOVejk-SzkXaKu`JvBW&`!Tbxof)F3K!bSiVIq+wMZZ3O<&odPn36@wmi;o|>&~
z_Y#kYoM(~pui5jiJ_eiOT*}r4ZN}yfIjrXCU|m{}dJCs)r<VD(+BQEyJ+6`nb(g))
z8{5REdGpS{>EK6#HObH^3%WVUII%0S=QC)p;~kB9A})1gD&_Ad3RRkQL&@r9O_Oi$
zA|V8><9YJu2^JOF;p@WcZq`aN4;<c4FdQwQ^I4@Xf_s`;{2QK^$oU|bBtdU^7Wo_*
zRLrPIwtX^RfwFc|T?Zc;3~)PXjr{{_B=ou{ABr|kDAB?-Awn=O)Wg)U#qsB86WA^h
zv(8Zm?-vu|r2+RrW{OjYq_)8(8FN1_!$+q)s^b!!KR{9P&{%t>0x152s^gsP5o0Mo
zy&C42yDO<4Um7@{gF*6j7+%9!CL|K^YO2MH3~;<N(0twVh2a(pq#sezZ2ew>`jsYm
z_YcSd^)oFq659Bw$0|@uxZU~g-F|7K&CM>%(B<hNu(3~Mp<LAyy6q>(0?Xb^q-*{N
z6m&+g_j<gX_Vj3+P?o)|4BdQ?;v0Kf@;JWK(}&rWEn(dz`8T|Uwh24~=kdDDw;?xU
z444tDOBF2r!*G1YtvwVTEDLOXSm2)QlS~1H-sEhl@H3NGYlc2hZ)tB9%xjyg<NbK)
z*_^T^1pY;D^vHUK2&Uktco~PwvacdlUW}D569JSW2CoJXLin+cl>=dD?MI8kd$2`y
z*Zbm}Djm4ZuP*+n$Z{4-+xdT7k>|d|M)VdDoMQE#fdQ#l<xV!CFPrhTb=Wr_WjrD+
zwdUDdx-fAkqU+qqJB0R>?r30+bSLJL>X<6Y#DEHhXR(7;_W{r#nyr0{dIGpc<kZfX
ze?<9uJDwtk!5ZCUJ)C#x#KdP&n31+xT=1wadw;&xR_WckGY=<``pGPk&za1;51Co0
zo#!WKn8(3mHfR}!e)Xv%y9ao|gv2b>j?jXs<2RzDlHTc8Ndi$NUkb5eD=B=77${CZ
zn6OgXxBgfqL(`l{G-%Y{`>;o{N1&s*^BHY>#Vu`-*pC*2m^ZgHvlP|(lyiV(-Bs{?
z0_LHz8w0o)MDgJc=XsHU0C_)t7R419X=dF)v~5a=GUn^@f_byIFf4uiCA$sTp6MN1
zT!*QnR=;@~%e)Ll(7@RHZ>A~HF0q_aB(yJLp<%m#)`v4X>7bO5lt9|Wr1W6HZYL(!
zHfk=~_0CUCPs8F{rJ&NjIf8M@?#Pvb$el){z%bt?!_bXKy16jVt67ExHvTE*BGYgX
zkV<RsN1aIT?8M{7i*EFp@Id!@beS7S{FK*LD#O}p;dRr}6)n~*BLGYlI+z9mBHU|l
z`ktgk{M(rMorL0q+Vx%_)QXGD)rph1H?*<NFkB1qoILEt^JO4W258iLJJt0H^{bUi
zSo`qXqT)5!7UWa=dd=IoI2AI9Bo?PL;pYcO5i05JcTPV$kk6IBy{|JXPkx_NXGFzX
zO4K*iZhHFL^@5mPQj1ZzS6}u)N14o6F$sQ_{R4Kp^PEL~{d%N`@WTZN$}HK(arAv9
zgc&!FiDxnvw6<)%n_^v(v{AT0v?FRk#TU0UMt|Mm8G$Z_+v=`cf@%Pv#d0apf>L-t
z*w^mnCEp!(qKNo+$9eSI*~!T~@9RIKiabVJcgzmc9i0UU{?Ri{3N0UO2146KAvB36
z-2(_FuW}=_Kb73R0eqxRu?X>JF0IR={ER+gXPgxQ?;K%IsPSssUC)`$Yf^dS6k^#R
z#e^B*vH!_+M!2VrDK@XpBvHQ`gEO~f{@EpQ-v(6+jdfyVD5AyLHtex%f6+|1_PNcY
zf!uYz#j5(a^_==T@x0XTINW38a?cu9AkXdw=2dHyUT3HWTk9OwZyoI=R9ZU#npgd3
zM6Tv96?up8_FGaeb3HCMc^7CN2`gKbB@1rlwGKj(pAR0{&Q`odx|0p(pBl;c9^WfP
zwf5+_|872?@;GX*0yqQ0NQ`9MJt0Y{@4PBW-x+5J?yrrrSS>Dwjzqv<fXmN&!!1Ia
zIkWV}0h_xj9mhp()b#fThbztA`z_ELh{2QB<)+4wh%Fv-FChgq*8BM!#r4X<+htrN
zef5s*ch6;|bV*rjMTwPs^l`GSX``Bk{wOm?8hDiHC@|Se!+W#KuRP@>S!Iz+0jTqe
zKu7zyV!8BaW%)?IPJQ=lwc!2++Cpz2C2zoph0W?ysGIt@+^o}HT%pZrF2VVe_4zuh
z4873#payl=`f~D$S91@JP1a^Ucx%_Rzc{&J-P_x^?$hbaT!_CMDp!45Ct*LSG^p9T
z4T;cglLqf)VUPRu(P5rK`VoWn%r^=x)PROM1QciC`MJP`<>L4oNlLjDx>POe%UB4@
z$M;L*Fn=YNY)wGV(@CHR9yHu+ECR(Y?m&st&;aelrQBKu<im{O+{b&&<KH~Uf%=RM
z=Z`WG(z^7U(Y_XmHw50&(`EW1BEAYlW4g^O;}1W#we?P4h4^m}(~l|Ai}6q`$;Gjw
zNRZGBuuG?H4YubB-tA;JE90Vqt25=|Vv$e$Su^V+QzjS#{a#iA9R$vVaqmslW;lYy
zJzA#qJiwA-1lSo(@1I&=TpqGv1f4W_|H_k-EEeN|nvlalby$gQ<|?x8j1|vU5n{qI
z*@jGcpie6Ye4!7eG<zLk_nR^-F_Psr#Obp=fsBq^9Ov$k7)L+1MaUWe_rur2C~?eF
zRJzSI)nn&4+7HMrU5IGTX!dfBrdo5;%of*m33U3=iG^s4+kAThnqLDP7<((Jg50q|
z<iCuazOxT;Q9QjMexePFa~+F!kko{2QXJuDbRhEYdcj=9IbNG_Dj!ds*-YAs<?6j3
z)+r;Nc+qGIvzhWIY8mVGAg5h){qk)pX_-rs{2>iKE^u@b?LCNZL-$t}&AZ;?Ih$t_
zol}u*n60%rpON(=Y>417OK<07m-5|i#Cm>e6FkWY`xVF;=D=!=64GUX+W}KRc=$>$
zzp51&nTPfso25D@Xkh2sWUyt7DDayaFd2k3vTV}cr~n3<=F+hvSbB<`z4dn6{KWQ%
zjEVMi#-gBKowA-l(J5x;$$sm41yxMrVen&VC+O>Bj2l_MJ(j@Rw@qSt+vIqYiijKJ
zdmMO~y{%!i%01`SX9QM^iDCN6`KthM$W<65EXI0l5;wArQfmOTPzDtSS!^sgj$B&!
z%5@UGk~8=K&RNQA365;6a9gcgk2rr|(q}1_YB5Pkyy&3p=UE5zqOq*y>3vkO>8HU`
zHJm00@xSap=%g*<K?f7wBGM_&xo@cT(>t&k>Bn&qphvS^!2fZ*_GF#^L;dLeQ9q(C
zQfc{V;t8f~-`vk&yD^t%^dWO`vSZk}e0zr7vjt}B->{-CmA)A2ef8pUc3t4G`H&hc
zE>)Q-OU7xr(6xVD%;~CA(={O*R^J}VgGkqQ<HxY&up-|KN)}80_V9hHviB#r9}lL7
zumNW;O#<qaZB-~0DOEJ0kdt^bt;1`#o084aUKtc#f3JD$73b*}r=U7EiJm<VLbgzh
z1S;Cj?F#+dg}?IwQJ`dvs>t~mHXd2kVa%9g^@i;%^0T3Bs4ow#i*!Zo`}cBliqtZ5
zQr4(q`z#35VwMMFiNCe6zg!~@3l@CIwqA6(<3>F=j$(9@{?T>e4#awDD{UXu_JTfa
zN1}^7c1JFwi=6!Z%gD6|c>0G|v9BiI^3=rrQ`bzE8?3b~RLPBT{K6sI#Wo`;?TJK>
zp(|nNKGqN@Q=>+2J^MWU!z8o~PaTkz^Go95m+y;5QT!BQ&0na!uzf?dtgnfZ_T{;d
z*qSg7eyOyN3CQ$U>$n?o-&LgiX#pJIz8neG(kVy`h$1+!OuOs-Qs&?r{IJ4&VmnN>
zvWC~sUkYLF7oGhX#;!KvakIM7k>-P8GOsj`T}c&V_I+0!+t&K|D+ZmqTPvSoRinkm
zyjN~#qh{I^k5h|Jzc2Nu*@=&nxLWLV(C;dk7UdmVg?8D?sAlpNY*iBOyFIV^wvU3n
zg${>B)&WajR>c(vYbQs9FPqjMrEwV=UtIHzJ#N+MH)4Fll<{hJPAw|Zas92G=fkJb
zC&aeWbN&n-^|Vxnr>wg-aJ>zfZYZ*8?)~J@rj}REJ9_xIuV3fuKYeE1({UZ?6e)C<
z)41DG)HK<qAo4WW40JE{7{Vir6uK~W9ci_BTN%c8c)pvbdlOUT*|B)YyL4#Na&&An
zrm_11inVmE;QH<KKu8~OJ)nQMm7wXsVQXy<t+<(qyMB5cVOUg4)IF}(by-he!0c#I
zx$6%_0_qVz{=DDmh~&St8A<it^r|=>(XJNj>zDv-Kdu_O&Zd{Sh_E9K7??IYv?1LS
zdn6;4s~y|^pxM+vq(F`982Y)nHE;FsOnJNf4BjlN%g0)NE|@<A-Q(@`BN6KTYQAVw
z!Do#MxYkjzkGcxt2}Id|VX&JnQfji*UYgf0+baeGm2oeL`tcdrf7561`DY;!Y<>TZ
zsy<cD_iF>;jMW+umyJ;mnJ#5T1k@t`7_94iNs8B9{tcfjR;X8*ucd?XjPgaLTrpR@
z`25#}uu78UFr90Rr!WekhJv9k$~f!_%ujs=TEREiQbg>*A5vl>pf<lV-g_au1xGGl
zR?3|hrCEZzXc;lH?0#*$Is@#=w0x#P+Z0%HzuV3Cm}MnaAA3zlI`*QA4v*wJk(pn_
zb@aV72#?Ilqp_0UP3+UuS+e{3b?b9;qEV8BaU~%!5;!%O{mz0>Haf5O9U~#lKr{eF
z!%j}jkTs+Q1#{p_+>0|bU_bxS_=p%2Hi2e>9}?;Zmcce0DPu``CasjO{>)$zJ>kj?
z<_+)mHb3e(Dgz(GS7&c5LdA)Z0tDL6PdczXBwu;c=08Va8H7DET5twSrHqlG5?(})
z>rO!=Z>4PVpQtH0!IE=`Fn>k<fj%sd!wUyH#GKJQ;2}x_IJTy~&%I|xP*0s~_z8J4
z%Vx3R)@{_AKC(Tj3#=(FEwHD%hem+SdGzb%1#S_`32cNhOujy`S2{o#d{zi?+gtA@
z%bGU1=90pB=6bh3;a!+}C>mxRWI)>5-`;Q-q}PhSfB3T{g?q=c)Aj@V62m09g@mlY
ztQ;tzTOo|+x*>ap;Hh^v!PuEC;hzP+`CZ!KXtYNn7+?|bs<jeo00}%%8u=>l4SM~@
z|BL#-$dFMi|A-`u7tI%A(*{E#H?aXGdWQ-@3eBJ~F3ZJqkUO^59Deb4&fx%be+K~z
zbhAApya#hab6I~hXC8Ge;*M<LH_Mq}J|^s?wE0R2A1fvCZorGq)3->TkjSe9;74~V
zK_$#y5f-`bx5Od)6W$LM9j`bstZ71u!c00@<pbDZ%IJixr+Ab0NJA$r7GR#d8Zd@%
zh=0gvl*|}dsTo%-u^Uv8u@aT5R7x-ZdDyeG!p^6nznl@-TgeV&fFemAv8s2%916sK
z_#g-kSo_?}DY!b4jUg_ajq!#x0(kbp)`b1STM;vB2}Dd$AM}K{iK6wr<zrPni^)5z
z1ZyTN#qDv*+y|;q#gEb;W2`-mJ=qPCxCPA*K0qI!m=-yYl_oV)v_{mYC~y?@nz|S`
zsD%ZL6nN>er*GU{hj!_zp47-XZ`dL~Yn+mBi#?ZFhopALaw0X=c-l?cTujK~(5+>c
zCaFH8lGm#|jg7S-+ArcjV9j{R<;hK?cuiYoFbGc^bNYN!2i61o*%>=)%|`-_vy0C}
zDL6eSP75?BdVhR#*Z+?Q2t;_?J(!~x4s*`1(0l5Z38m3FMJBDr_gBoe6}NkzBTt{!
zMM}w1q}5r?Mm;tK>6;KMDnKQ^p5*bV(e<J$06fBS77rw3o1ucabRj(g!>iFDAW|wO
z@bXK6(4~rxDt%m#4)qCww5q;}2{S9vniDy02AI#5o<pX^LLCb=okJEvca2hO-J$J9
zpE4t}8e=9cPAd8;QUbgc?U&D*l9Q{@3(|^!_Ll_!5?eo6&bH?_&D5CAW~G%+)<k3c
z2JzI9&i;ZJ_rHwBV3dNQa#W(E#Q=?7Xw$8ng?LPHn#QWrt<qu5b{ZSws+&Apu@@GQ
z#YvP8256s4>!pa<S!GqcY03HYqbhhd#>~1sV`sFwQd8kmaDL5%qol8S03}!J<GZ_)
z`P!nV{l;GpqrRXcoy+t2!=@TD&z(_ZiR@*M%AMw5{broU%?8KE#{t_r$I)i&ynOxp
ze1J#mj9KDNGqI6u8rY=WJBt$ilgOOYez?pLl8DsfwbFBVkxgUW0oUqmLWG%~TjjN7
zKzu7@iNNR0ojE~3_|yF;8j^R<#|MRrNCw-;Af1IWZy-3?dj#eZFYQU^KdPq9=5HKM
za)0h}rHEA5HK{j0yhN6W1iC7`^O~}x8)@cTO@euruTGaay4qsyA<PN{)CV;i!IsB$
zZ?xS;k+K)BDF}qPT_oP!Ci8~2-!5HfWZhFV(rLQ{97|djS7{@ToTT1xShj@RBjHwI
z?$#@TO6!mIM&o8}bG>+<D7LGzcI)GnNmZ{6%7d4B=>;FswUR@uL`51lVq|Hm`U#V=
z6eWa=Ww7X$xsghP+y|l5A#xQAmG}#iBQuDu#xjJ_OuLeK8Q4h(hNGTdiqFYeQlri)
zkf1>tS>sR=)M@MFmR`Ibnz<dBVR`^<GF9ZDn1nH^u&^!<)4o16glCo_DNBV8fO>7!
z4&W?e4OnJP{XY4A<os&h)^>L^<{n(ma#}r%rnT0_Al_10GDo9BhC_ER(JKTk@5}ep
z#+FZF^lN6~l7MDpZr~&@gvwa}D*KvxS>8$TBx#(<O`-K(=M~^2(U>D*-7(pN$Og4P
zy#GjBrm=3-KDv#Q%wDi{+Dkt}a~D$njxKM)wO8=sUQVfG!%OeBKTZl#7wuuS3L16&
zpkH;cdz^nM7)mjG^J4ZT!DSurEM9&E3?{EnXh5qc20djQo5`d`GdbMs_W^OS9249+
zANw3^LZXDd=^m*7$ywr|yV9u&_a-6II7Uc)4(1_qUMz;N5X6sO?aw?#vck(hkj`R+
z;SLJUhu+E?ha^1-@+hV6DNKdt{aT-Am?KD@U_BkCFQvWXoj|tWY+)(@LFdJcQQx$P
zBo?DF?oaq%ypUUE_`*a1gBL14ubNhDLe8V*r9eE<`ay)t`4Jzd2EcKV#VjbsQ&;)e
zqXG(NrT6>tSbhpi)Xgy#T|z26L2|cmI88ozCGb_fB}zJz`3ii!#Jfo*(&DplwGpd8
zuw+2yS7{{!A7IJDe_WQ-uA-D`8xr^=aE5R2OG%IT`u&xRqt;sV1;Fz5!`<|cR(Wve
z0D9g|`dpd{0~e+-nz6EffR&s;5@)^wv&tE4M}>F^?|@bX7;H5vJNnqP;Y*|Ne4VWh
z=HCUCR4Sc$dQFpjE4Kd5%6Pij7gtWJlEaGn`Y@A8PBIU&H9m+Z9D2k%A6abjj(NZ|
z8hMb>PXVB^a`xC|+{)AY0;&cv4$W3Jd3jge>l;sIGDBb+=lP~+a1q_weg{n%v!7pO
zEYfOJt$zN=U0tgNhf|W(8RHP(+`zOC3akC<!bP!b&%<UA%1*1I8E&=u@j-nRXqm?7
zrn)V?bO<7tWt7OI08`l}H7~gt$0H-WApiFVJ{erEe^rXAG(uIi4d7<f3-Di#6HCb4
z8~F&zeL?*GPp?#tze@NYR3P16&9~&fUowf9eIoY{9KbGD{H6n6!wsDhIvH5FT3f=}
zPFv5^iVrMa>tvCZfQD4AQIg-ekqw22b#0%H6kn*GUS5E{u<G2zBr0oKCa}JIJ(%@P
zV>qLj^Av3x7UIQOX}v}BGIwo(-!^CM^AxB0ktcnXk_6MW*_4%Xyz-c1kE@2|5CeEK
z8rx4B4_SKISUZak#}zyof1@QUM2IDBnNs_0ZDnkLg}C{lZM-l<)7D!+;D^yLf7YrQ
zMXdXC4bzCEDLk9DNqBsg&a=JzMcbf;Y?{YOVJpZrlBFyvF~3q*k&Nu2<1)+R+0*ge
z>3|_pTaKaDO{$vx$%LaqP{ZITUBjktk?a~}OWP=vKSEu{adW34@hSRFH;tv;)l!LU
z!J_ss+ay7!ACCx>_YF%#Y3_c1ne7CPQ;`K@Ihaz3Y%r#5%`0;Qe0k^{?B#s3G`iYD
zxq(l(e$AZ9D`t;v@=cz1apIZ7!b08X!I%)25}IldtPCn(_#S5*mE5)=gUKJ*#}(HU
zGt3fy+t-ru{X=8#bs^S}w%V$F+<u|B&K-e&FEm7i6HO2&gjq|M754>%VPHdof~BgH
z`^hH3ZAIZbix>)Zp#*MAM<;(jk942|*@6<0m~>H!5$8y;r<>M9v3s0H@zK61N+<Ks
z56PrTuMA)i!pqxAU0^I=nv4G<u>>Y>D4z83gDpX7#`n+|o3YLKvhU?tt%{{`nS*_X
z1vyz_oIow_=_0#Cvlc_Y#{?alV5?B+R;rhhw0m#Q5Y7ahRq1ly6HSmt7C-biNBX_B
zuq5F0)T0;bo8FjqU}Hp8oPN1^uu;W0Jtp;r9MzpAiqlf4us?Ja9EZFrr>U)6hju*S
z@MejTDj8pt^4j4zCaqd6t|mtuj2NJ_H{=j;JyaP1pk<s@chH(IXAcfudlAYU8-iuA
zJe&>LXp1Q~vGyKY(-^2}+JAB%qh)RCm5PIp>+*{!)w;bnG?_bR#2gzzANgix#SjtX
zamp!6kr~#7PXG&lxaB$WW09f+9%)j^CfW2&l0@cs!(0+$ty2#}Vz1&r9}@or<h-$d
z6{=-*6*6P7iW=h2=%@@zbcFOknIqI;ws#ijECx9Q__!RYy}6k~dD$x^E%F6JJb@?d
z4zo)aZ>fK>qBo$oFs{T@gDkA7I6utPV3y?!4qcH<VyQIefMCv%z%8E@Z$Q?*Tra6n
zYi{i~VR2_ANmrZO(i@wa{-5DdOkWBu8A%6Iznkn;IwRTVSo1P(dU1$qQ)ZRi8-Qym
z!M#$Ky)VE62bbNnML=u-;hz}ce|<*eJ1ixtgFnwGZg*M|(!os%yqMoBryJ{yrbV0o
z9B{wdJ4J1!hND}eqp58+k}QvP9GK07>5luf>RL~`^=N%*Icg16H$cl@VR5`??QK7{
zdxr#UJtS@Acut3MS2`P}XnjQ~t9C1<d$M{yEwJ8y?dow*5$Mk!Va+&$%D+44?#BzF
zY_)muQ&OkztH<|zta;DP=J+o+Qn|FJMx;J6!+7lvE~Kl(RX?K}Qa{&?L_6_z)!yX(
z9I;&VBZxdVo+dFFc!~R%V7XxrZC>FWrk^DPhaHyBbo*Ibu0^oEEA}U8jni4EO5lLX
zhuD%*Ab^MYw{J>NJC2(oYPMp5=dbJwOp4a3-G3K5-LiHeL{^*a6*8ZeCAhOeT5BT0
z=YH~i^*%O0uKyFL5l-anTBquuTRnevRDg@+97yT?xlf3+*s*rzs@eNyu^!55L|NNw
zh}P(ZcHxXRa|M-UYp&CAxjnAG<k9T0<+Ac0pgVS4O|4yW-e;?g<_h(Ps+}bij;MN^
z_J@`^f+*iD2%U6V+mu|s%z?&RcnMu!oZAev3C3Hgk$ESuLPmhQO%o*{-u35Yi=o-w
zd{w`88ivMMpR`B$ou_xqdibio?QB6yATewT1Z~Io1lseDgHSTzJ}5!y#>|uDp$U&u
zvUPKX;C;-phWo_Ocj56ds{(7=X05#{t#h@r^{<B=eJYPvd5;7KI?fF|hDC<Z_uO5=
zloBdbc04$1wJL_mdeAzff!uc>QWF`j7B+Mp?0lgfZWXVydAO<o938%wJySqai+Nj1
z<AlzKt3xXbL`rD0(Yy{+Bj?FfP-sIM8ufO2a4ciFVW=2%YppgQq%q`78g-@9Ms23P
zJ~I{rCH1n})D9i+H(^~ay0C&tFx#1*-=AJsByG(c;cF#XH?Kvw&ML<q+O5wYcBi9k
zZy-TcO?0G<4}j!nBA&R=i#N~lj#k!R)UTS*d(`_``8Rl>eLS00A-8--k@9Re)+KTr
zL$;m-&k^z&x7O{m?WAd^9yie%Eu50LfTGG~O@2^4h0XkD`dU5aYlzTvwDk;U80+#E
zd+;{%o%}`kOxkpu3B)t_Vc@ED!icrreSI{hwPt}A3Dn9>G9UBfb;$BJyrgd>th2B0
zVUerno+3~+Sw}0vZkG)!N!uxlaf(xualDGA<x^zXmkX<M(F;ikdCLggUI4F;#KvU)
zheNx`o6Tw!Od|X$YxdQIR??b%Nm=DwNI3dEf!pT!jEPnvR_g8@n-jG2L!-iqwQBA>
zmZVa;Gul%u{q=vMopL^YK+~nzq-7q!E`^0)oa8SVEjfD&?c%h5tT*1*pQDCigM>Sj
zrsvjBTf>{QOxY|z;BTz&EGP%(=tR&WIAt5jM*I!PBCg`DP(aO})tMi8ilF5=CoQZq
z&eh0Pi^EraZA#bUEft&QV<y3%u`(De-qp8!%JU8;8z3ssg4xKRWP}jfkG(5lOg6BX
zc45!)yTpka*mr!Xl!3;@1u?Wj!@zU7R&M&`_Dwph9M9T+7*|4-N@J1Q9Z|U$Mv!3w
z?OS`ph)uG2;$$_&)~;$TJmzc7Wc?9ssj6}-0<e#S=AW1`o`bxD%~}gUBD>Y3O+tRE
z)}C{pa26}oDW`%%(OR88Yu|vD@sX)TXwfd(56;wM-tU!P$P`Aw|7BVY{fUn^;nb0s
zU3=)8HA_G2QT=X~9Uj@p0;V}qXx@fNPc7qY>_FJ&pH%O^<R{S%<G&;MQ)7HE47qEJ
za!css_9&SzW7q(e8pWusu$aI{RL|jAmqXZx4<`w_n$(HtvCdW@ECv%_f`4kUU{<!X
zuzG1C6?aX(eolb>{RFQr95cAC?jSaUDe`a<vps|G`~RWrEQ8|gwr!o@?%uc)91<KF
z3Bdz|1PSgCoZ#;6F2Nx{aCdhP?li8!-S5ly?S1w>=hVHmZxuhNN|0{WJJ(usj`0jW
zwzcS`;%!`k;XS{@x^)`^nbZ%BUx)B?2Xfqiiz;Pg`QeuzjElFGJ+juSp+S(H=4d?l
zs)P#L<Hc^5piiZJz)h6YnVQu5D-)XDYsTaRa|oehB~MbTyAAWqR{2zbVg2jcEpwV8
z3(cK<A`;hqOv4Pq>sy&|h=c}Nnp5xiIP29i&B9au;(IYJ!asFLT85XnkK9j!+uoJO
zJEn%Ls>E)N3qyu&iFmWcc<^#k<Fy7_Wk0Qc`cnb*=T(DH(?3}N&)?3whc+@ekip;I
z=WDYy|8gqkG#N1|w-6&~#RhLjm#sfdyRIOF`l?#~JX}LR@G@@_cQwFHjz)FrD=N(X
z&Qz$Uo5akoey1-(r<8J{e%vG2F#(e>7{zI=YF~+n3z?~yCb|?D#UZC$=uRN!IF6Pe
zouAPEvRu}a{fgA6;#(B_&@aj_Jz{F>-q+Kr$vUZG2CKQ8uoMvG%11pJXQWU*^Vksa
z#AcE`-N6iI7a9JKixGy&-!xj3?aP;zT3oNO80l5LbcQ%<KgoS}-eP3Z)Jd+6F*7}=
zkp|S-Wm+=s;r%M+awgKGu=AsM1D`?RTy;E7(_d6>`KEDT9v$*kU*#S3_{;<P1d?Ri
z)W}&t&y-(jgM=4A5KF%^s21D70p9Axdlm0mW1V)m=!9ox=G+8GG}`Y4Z8!THo6!zc
zXg=CR!lM#x)=`FQ%cKKCblKN^eT6YSGn?G!^kiyk0c^lA#|L~#dJbHmq|iG*u5?n;
z;Xzdp2gVbt3oDkp4OkgeS3AUA%MG-vr)>xv&2_o}I)F&+GrUg>`lfjB^O{sg<+X!V
z=46D9fY?ayS5HUrX^^My`QOj?^UvjW=}Lyn+htzKd^xW9ZY#!*e3LV94cvsaIyU8>
z(m*SSl^+Dsfa6Xl_j(_2+PP`^5qP3EUZ<sEMo?L;n>AW3j)LnubhR3nXjIFSBtbaR
zS*f=98Vf7bmwBL(qN6Z0#-8E&tdeK$6VKql(#Qm!$|OD$9Ti-!bN@fiIkF8?b?Va>
z*cI<(+@on@Ywf{I$5%O8^Yojj5jBI*;ZgHd-BduEZ71{lKaV(o{o5R-`b&j9(Js%4
zF&;ef-dHWCx$fOd)7Cm%1P=#cel$d+`-D`YN=UIfO+00tcyT=byBaS1NQknc69V>P
zdpF7U+=edN4hl6)AE&NNxpf8=67Lq=TxZj%{$IBv4A11@h`_+_w7fBv8SYmM&MdqP
z$#r^494MDhM@%8UfHYNs$w<A(ptWLpO<ZaUn7yhEsWx2CM_}X)zwN`5M)6B*%DDPi
zKcj^MZ!uj^78tunf@3J^1}IDa>o8JM@x?~Drz3Of5>m+eeyrv4c=t4sER59U?%3mC
zJ?EnUbxNCdhRH^W2Z0S(00iDHWSUhD@l7><H-$r^A|0+RP|BRPGeGH9^tGqMQTDan
zk>z&@iKnBeU+Wcr&SWIOD~5|x#{8A*I<_DRytjZPmeXTMXOpE$Bkrr9dF$=AN0Zu0
z*%fQ6DwO=Koa@by_S1xwTBI!nlD;`+=qJiE9O$Bc7kab0&2J>{OE<NF%g9Zvp)DnP
zLj3xtp%$joRsLyNym%#I>aua=Y<d-obPhADn)hYDj7KVcRAz)r3pbO<shg)|5!U2e
zVmG?Si(@5D{R!fb<D?Owk&`Dhv~GWu&=rd=?SCO2=GX2>E#WPs8g{8J>r%L53ss;f
zz>U1@&B@OHvht#gZSG`_op4;T2_yrv4efZ`*Ih$AA+!2M481!C(nvOHIIwTzB4~5G
zD{%~qF++T-zu&k6at&Ir`n1&``5}(9(lpPYSu;w0OKc5lkuer$<XLpQp-yzoBjVGP
z>u}Ri8J!>k{~#RGM;a@3p^n;RqsxgzB|ImEr{R5#Nk%k4WWqsGm4Lor*Q?v9Eo0$!
zcJKViWj2I)$+9OopY_pzIOs5az;{W$F_XtcxiF?q+9Ne@E(kW*W}ckYIdfCrA?vSA
zv87URDdVqA5&3c%@UY;aiw-zP1r{k1qd0RA%LthL_7@m5!AlsSMEAl6Od3g>c*jx~
z^yK4LN^NwS>WpXs&1z*D*z2)1%xKPF-d>D{9aSkCHdeKFW&-PLs5>=N?~MGHSs#8G
zvG+lquDDraIpAXKESAxSnjQ;OxI(@cejzpSdy&BGN5iiZ+p+xH-WM(mywHhc-E!VT
zR*M<~EoohpvvavP#8_WVgf2U*iV3XFo(78&rGBs}5=}8V653kgYZy3Y>n&i7r>>kx
z_vBbCe@2i&!1h(MF;*{(6X>c!q3lVs*hsk}CSh27VtFb68^?G%nwyn$wtQXB_{nn<
zs{N3?iceSUS7k9A(rLK(23N2v@+}S9qQ}qOhUPQL=zQJ>6N{b>V3$-J_SH0tXI`}J
zP*+9Tq9Crg!X!3y0DNWQw;?-3rZ!Pl#?1Pr$~OvHwI%eTQ%DkUTTY}hpX1@;kyxoe
zegj$$o6$}f6YGP*F&2yjo=lY`BQg`w-e|NY8!TgumB0;wG+>6u%AuzNbkr))cpeS<
z6yNuA*n`o<exhJE2tx7)ccQdqV`6IO>~q4<FCM4sevndE5Y~#7JReg^)xnR#YjnPE
zZ6+(}%Q^TEcgjl-g>|#NB)ZU(l5e>1Qf2YVnnGWVe}5gnVbv-ki0tcqpsHj}&f$Er
z`?UYvnGp@Eb_e3Y<Z=)@UI;<9ks}4^+OU;8iMMvYL7FjF)$kFY&lPJJszog|J^e6&
z?xD|X?2oG)WG*9UrA8Fm+@IO6!xUF>80Aj#b%a{7#JAsc4RQ8cGQnyZQ!KV~1&Z#N
z?JyR4tNMt6B5#k#BFo4QgTbKz8J*0JZ|>J&oTDe$g0=Au^C|<0^yd_*@4nh#JXgT<
zU7O4BkkQg6mCvX#gaHG=+VpayV1=?y&GcKO`OAWwh_etZfRLJBmtwq#C!so!3-nG0
z#jJ-&55gtX(9P<-1tRi@?yj@SGThR71U3%<)ey(tM@R#Wp>-$booKiM@dMEEYxY;a
zgTLh2r=K8}@v(tTd)ub^r#ne@>6iM#cc25{Ld*WwjtGaRwwSRh-_rNjz2Em#z@RMS
z@sBYst`Q-(iBYJpb2n9TgxV0ia&;Zh?CR8hXKVOUus>8hj7yNoxu-RD*XCS>WYbgU
z`7_d^l>#2S6-vWj5)PPz-aSF8oMBBsZ?HaNtr>VwoADzK<!qvw<+QrL8Z`8jRhPx-
zE=(lg!0+i>0Afmiw_qP<+pI4-@jL8g94Wx48`$^ivw8aK4Qb{BD??Bu8?laa#cZOU
z%y=alD?=wt^V5`%Fs+=cwL*P?eK4?<ywZr*@oa$370>kgJmw3CLTTy~Buhh9t8hw%
z>S>el4lb0rR`r*_{LEu{^J1inL5_jY-;b`Al|bbnv=rXMPZHfMiqp7%aXGvrWHa}0
z;eBH;NxU~92W4$qF<dW_z=7|@Fmgx^rxuYuZGaBau%qonQ<)-P?jxIZxPRFR3*PDa
z5kS&lT0$Nxf5e6UY!0rBvrw-d;abF@1!jQlC+0krDJ4WNE~9uiba4vhPneEz1qwbW
zPrZ?<nZf*g9#WwpfGL3}y+;&EIu&80c+{wfM)^nx@C3qIeG3L-zdk*&{+Q2vq)PGr
zt~;Yk@1t9Jpwv&mH{G^NNI*xoem+Qbda`A+0_>v{k$uSl=hc<s#OFHrt&6*rm;F?c
zI9vW0b?;_WyBL3CO3+SD)Ygf2_*fXP`^Eu+`_0&5!?5Xdz?)?t^#C|j$=C_S8gl>W
zv&TzW!)XPek$kP(Qbdr?)0Zr3XZ4UBqI0&T)BqH2vg<i+gtYpKl(Iz<3h5tb9-U|6
zfmu#G0r$FA6(gEu!{g8KZSVq*Xa!?oHLi!Y+aa$|q_*7%Z{b+;#go$0lNx?TiVOHg
zX{166(C4htV|zX>-WKZC-+CA2;3}(?7O$@^Q(19r^*3Z^9CI?ITwHo8J(52he{`kc
zJIGebY4K=r-RE?!L}>o?g?RHec8rf@U&!iu7kwuqiAX5DgJVsZ-yH99f0@(8_K#TM
z*j6DA&rwz41Kyl(07@9wimjyG9_pSc&N>KOo3Ku9NCxY$Z&8Z|s1VO(enHnJYnWvD
zzjPsS<Cg$l!l0$hfFA{Ro@{xRs`on*V&&>vu0v9L##kBVhOh9^jJGO7-LIa?r#THH
ztY`Pn^wbU<C)3BU&>xUv#Z8e4*Ia=zZ;Ef)rZc8&9Zx(~DQ=}f`s>1t`MTei-Y4rf
z*|8W5I7r`3L?H(k{+1;)lODaF;DWt;IQY|-N><Fye)clA=^1QgijS&R`H`t{8J&-|
zea&aMcw%V$f7l>(2GJxOoM_4Dz%&Uh#>%q!wAWVTbQUNh;_^k!15mkIEiSm5Q=4WN
zIf}NKEJbgf)5R6N9zxI1HpowV+y}~kIIcMR_e22|g<4x@IaCPAurWrz*~bDgR=r>M
z5gFD6XJv>tj*}pEce%;&V!x;?!F5=BY$0HpGdNk|w%?cL72t@$T~oH<JZ@}DzYz9D
z>G`@@W*hM^==7D75p5eY(%|7`{~uw~dexLLsDE~^=E1oV?EiI8nirr?%-)`80ykKi
zNSGdxO|J|<mhT}zbNd@GlsC(w(JkcG56O@Q-4l+Q#V2;TuFb6g7OHEq=NpN(b{JT#
zTIvY?O^)KiaS6({np_?pz|5NMzgs;Se<t;yt~Pr9CWl0X)Qc*_cdBJLu>9nDh_#KW
zQ%2nbI2PU6<&CA(>ge>5IEJweMkRO*gEH;<xry?4=9nyO<_LfO5E%9WEwIJWPikmY
zexou7`MS?<pt2nR_@b9XHVB4$)&LjvQEC8HEQjf@z;{~1z&SyJOB$}>-{uZ6Oi}0x
zUqAHfYP}sPkUk@M!9e`8;dmvTX^IQ42&JEm^nF=(mBIPF2yCdEE~V*Gy$xoOSnL@4
ziLaj4t}0~|9qzDG;HX*o3BF2#Zu~u9(bg`fmz=X~xl{%Rb-3Qh=W`zczUuWD>ytl&
zjAeiL8|WF?oxJtWUcn|B)ZdQmKS;-ZSxeP@#dHLpfaY=HOXlP+f&)S{e(>PYc3Sg)
z99-1Vg#MVQVJoGJMIq1^Z9PdL793ffwGr9o$7%Vxc;SXq#A`zL0rZ0v@H5i*hMAvM
z4RZ|Pq;ECl%2=ZWunS{6P%di$-^zNC?0lF?oOI_Trv?Z9fI<B|E9;9NO`$Y&Coqb&
zpL<h1E|8E%(lr2a66pB9Z3KW0?t|O7*HLnn3G6F<DIb>hO@qj=zjsGg%DYeINT6$5
z(t~U2UCyLdejr$-tdrZ&91_S10B+kya&#ni@7EIRr2ExQ&F968MI}4y5UY}Nz;{Mp
z@Yh^+>r@{9>Fy$JjIpeztmRg553ggxxMdr%ye^UY{>35T2`)V0@|Zbz;#0mtyFGlu
zMAERyV;WV_Hj6Ki7<10i|Hvyqbp8}8e(n4`178+ALsb7^s!AntH&|sNez@GSu3hvo
z+Z$Ajoz>0_%y*HWFh-k5TgGe01bG(h$I_5yH`bgyd*?(SBA*|lg`NfOxBEM5+I>#C
zA)y&<k9!1Ukz|&w-gPnkf^ihu1lOu|`weZO%3AK?ZNEtI*yP_xXJgdwJT-l2xxWax
z5<1x#pDc@@zngKL)DBi)c^DY;z+tYK(i?Z+w^XwO`_V8WPw;hf6HlJOT<d^ZK@r^+
zzTt3c1o!vMp)VuuHNy2_yea;p#<a+tJ$w{ya7glYe3_^d&&t!u<#@AT<Go#8x;8K-
zdh^D(7cv6u!Y(P^e_W4SW&>Y0VCFTyHJhh@pvbQ<WVgk<w6dlj%2f54=q}goqKsQV
z<v$`Q-yz)Oz!9Gye)`0%Cl`Lj%+%aY^bUzdL!PBS=w6Hf5$xG-L4V-~Sxc%tR9`}e
zQ|+o+>(AwYzFBkRp{0xQ0n3uJOG$!8{iJrGtuq`Icyt##HnRzic6zYdDjS(jg;&sS
zk&@r=CQC0=-mGQ%qMhBs{Tw(<jf#%KEf*c=X4(Te`b9PeV4>TBeYc$X&)AWP=RTy0
zlhB7!AZr6Ad`#&`RVdx@F4EdB)Xv=LGZaj@3^u5@0aSQzok;zwZY~qf=saQq%O!n@
zCCPlnwVBmehf>OJXl;b&U;FcP5)T>&(T!QB4S?r9PjaIi&$zsg4ilnQ&WVYoG~aNJ
zSM#QmBXwH!#wx4YW?T>Y#Q12(Gz5@V8h1gPLjdApULBd@SnK=9JDS+hHGig41dl0K
z>alhV*d3=^>gDnPe^I1M!P~fCiwUq=S*&(f1k8nm5U|u^R!>lDIiw+`9_6nvot;ls
zJJLbU?|4=OFQ=5;d1kad-O2FoMh4!Zj~Vl3d-ZHmK}68#NPMYQs&!@GPN03dDt}Al
z*o}H1Dl@N&dg-ds5q%v@ckAyERvUE|V+YKm<MPGavGxgyBmB(PlCBOh5{kE=Ju54_
zBbg4-kB#k}rykWNq2W|XvdtC7{3=EGIY)bO6g9FA^_RTvf-b^mn>ITDzso2Xn+)0)
z)m>q;#wnR*H12w>_6y4)yJwc0Qhv`6OlX)hUIX(AUexv8Y~Q)(jFVsW@K+%pFuha{
zzF~GdcBGh7siMRtQLFs&hZZcyI5|896k?v=LoKEoFd{Zl*Yj7zu&;3@Os)$86?Hb;
zd*w|%V?y5mM>>jTwJpO=kru*znux5GU&>sAL;(T|Y-d|qfcMl{X$9!94(|qlEKd$$
zLzzt^#`q4Hz_c?l>pJB+67ydTqQxI(HoYCGBAw&L1ys301%wInG$<q{jyHpv1N^Nf
zV(bh%p!d>hC6lYOS>6jpG7c-|S30GLcZLPVjGCTtT)r}-`fG5_o)A`ey!)(ly>4Qr
zFdE2!Y*BtB389etFf69LmAZNAa=+c2QBz7WI)%K6cxOHQ*t1B&Q^6Bko6H4nkh1~w
zMD+$X&eddUhE$yyUr77KjK-)))waA1zJ40*T*7o_7Q#<<Q*QgWNp=hNt|&~g()8z1
z+UOss!#p4p$Df-!=1}k6xp4*U$np-)lULlYp?mI{XZEQD&LyvEpEcX3{IDlZv;Vkk
z<#7%&CJ~II4Nb2?i{~u*G77hvrsib>emaeObjL<}bglh5H2*56s)ThJWYxS@+_g6R
zErowCH};3U6UiFlDBQ0f470rIgaj<O5wr$;pbKj*#eH*(_GGBsszc#4`zxd|KkNf9
z-m303SxI0><l$7WVcSjCsk3;Nuv9gg%?V3o%sBW1_2+%Pe_&8*$*|*0u&q(=ibd_b
z(Nlt2jZH1((7Zu*q_LBb^QNZLf>#75vWfrVbT?TqSM-srNHsOHd1cRh=8vil=nPsX
za}ow)^*_9+el|uCD{cDFE~CniJO@m7_3Cu<n2;CW?`Rx*8mcs4VfRHn9bk-{+ibq(
zE_;3f^km@sc-HbQ&PxxTa3<;t8Xo)>L$~K5hyw8s{*sHiy?I+{9*u6k3a~9qPs~@B
zll8neB%NjX2Ki0(pa~m#(i&+tQ-Q-YhrG)|du>^4%o=)|#YES6tpyv2y~exJbhj8s
z5@H?_VCFYkEP?}@>M|Y|nkHAkqr`V=ic@xNA>2QYWT*XP!BMmO;$Z!@&%5uu51DZQ
za7GP;M;lXPLUzaCW$n?ni2$XxrpJqooF!3Y*3|^HBUk^>pMSQWT)J*ckD=U=XIUz)
zNvp55U2jpj0gqqZRr_np4d#G)U(Y7bwPZ7v$JD0tg!?6IstgAicPpE^iG9Wf<TAb?
zfF7W-mVC&y&?D#MnSC&0VOh7@iQH(|+7{<->~Xz?uic0)l(gI1a$9<TlM}g2A#r%@
zQSn?lN_&KK+$+z*`~1eS_vHK1U{UEVo~9dSU3bjma(;%}(NEdrecH$jbLS20x?3(#
z6DO+x%~J}P%wzL-ISIe_?mT#duCZ>#sSovO&7~DJ7%X7Y<$Y8~M?%#0nMWdmXDf+w
z85dvk01jOf0OE9VSWet9SPpKYFW+3#)>EXSIB1XFO)R@~X&%9YN{#~_ALw4#9!$++
z0|3QmhlTH+!x55(-gwO1RIP44AifNEU>q_$E(%lMnuetEZBcO>P6S#ml3DF}v>vmJ
z#$kr8po8}fPxGdaT~YbxMt%r!0v1+NXCIJJ$JnET6GP-U;C;c*aLmiQSF*f~Zv`+&
z*=+}tb&;UwCz$J}rj2D7nOVVW$jG7V38`i4Eif!o*1COjw70vXWt)n&)EYI0zt<K8
zI)74vO(uUbKWkP=^w;2?rE+fg3HeF9exL;^zT!w%V?MXdW)SC5H(!tXzK%vXL(BgI
z)z&3>I4&Ql&NR^6dSX5sjgfQt2XnS{t$0c#@+HPuB?GmSGr@cxlLhVyDjnCW19QsW
z@v9_J1(6kU^u-W!<HZIZrDkic6H+Y(;=VA#4MPP3X5vbM(@fF`{(be%q~=l1e$3ju
z)T-dL8^#sZR4@CV0?Kx2Ts^dNp{4$0TsOz&6aCk(gJKzCZ=5J3QYmw-u+7)o=Y3*M
zq}P(*hkT3)<@bXF87!TX7jdxMS}EIpqfvq;`#O?8WmO?k8A~&333=toW#0<mpjR^N
z@Tm73^K@nih~Iyc#!$YNwdFcFzNh_ZvVV`jB?9skO#FUEW7{A%$gkeK6@+i-dood$
ziTZ8}$r!tJ6o*fG-}sEtkY**H3jMeI3rR0@=+%$dYX)%q54GKsN+MZ=3^lXwo8!_f
zGOd(k3QxP|DS8#h&%XoT!uog8A=WV@@+gJb=R!Yz-N(Gk0uBoVkM!&B)3eK9`3Lbm
zFMRi0V;yD3i#E?Mw;9h;Ehx6n-(Au|Xi$UrAFY0<^<UVdJR0p|bxNpCZr&81(g&-e
z;`jU3*7sCVhIRXTu>PUf4kP>x&koxuI$u@N<*uDRc7npRDtwzMFY286LkY50Z>-LI
z*K8T`5wK`!ETN}Y1S=Wy(>d{`oZ=|e+5sG->ap32WWph~q0Ld8$wwZ<vnqy|`%|c!
zoh^hw2C)Xf`rDWMj)DwrAGE-kM4p<;5sD5nOlK$P1iB1jO510K?KtLw==(w|bS%~W
z@Wj?$$IpFz1}9XJTz`Ppz{Mll1F}7eJ>%m@ea?!mdBKmOIyac-Go4`V%W5zMiZgK)
zaI?>aw?}sW-m6}q-yb^#>i--kqQqnq$K2a(Ug1+_Rp>bIJuFhec=GuzjiV!g(&HIK
zk2sTll)5($9d=933#Tk2C$k}$@4>sgJqvg1C!ii%Cm<8k>I4I_#zk-`6*hYS2SArG
z4&`nh?!R=|z+8wF7=&MIXAjVsX@B6g$2;V<|7gZL7bP^sEK>V+%M_eGY`;$kykZq~
z0>z28TxKqR8toYX3#AF!hDi>pki5H?3J+taFy@Gevjcid4})S_qi;ZlsA5_c1OmGd
zwg{OO(GR&t<Z@Eli@kGoHPxYEQc7q;yy}_ji%>^O5~<m`QjeNP%f+)*apTP|$ANMH
z+T_|0*oZO@!Kt%YG73pFEuc!));WT{4QPBjk}R5)0i7u@u{~i$p0SGV4yzutd5=CN
zBIV4>n%qRVptpd#z;@R(<z8*rPj|6qGPK*Km_R5d8H-sp`0KQ~P~~U;*I7Td-*jw-
zmaJ_tth)!}SHH`$S&S|HAKD*cnDrQl&<7di{mKn^jV!>GgCCp+pw$c<ZVbcWJrrmw
z`=M_z9sB*lfSzL#zOCPne^IpU-b2Y1@|ZCCHCM{JVHIGx#YHN($tX>Pw)K&}#n!W1
zV&sL}C(wX=9}|3h&)_>zOXGhHQ6{>V!iKv~g3)P5cpZZnS5Ngvws--L;HU>*`@%O8
zWt82yy}R6EegmR=vf+pQ@g&6R^m=2bJM#4hxGUfrvRX{1=~t#iP+~XRCEd_sE6|UR
z>l&1YdSml_M5$+672gRNfVqL`%S6HEM9E}--3OCfqHJILYb1I46}ZAbJ4=m=(IxV_
zC>Z>d-L5vk?(DU0V0|+r+|h*&{sVtO@SxA;{~8KMMuD0p!VW9Q*o_Usl<(x<wW_jE
zks#2?eP&mZ#iJ@eUwMf9=mkBP60cG7hhR>B9bSCyLjrT>x0#?HYGtQ}J_8@mSfH=s
z5aEcDPu?e^(QwjkCpqbPdB3Kmy-?6H^r0)OSE}nimrLOI@IXMGm986;YAeEPNQqfG
zr5Gm<7e+5O{RVa2O7NEYn2*O=T8TLKsOHOn2Ud&-;y&#@I~@LP74sm8Z_MeLe6!C`
zQEBndbeURI3Y2gm<xj-#6%TuhtgbJO@~np^-ANsPN8dMEp4%}BT$KfqIYy?6>MWrW
zH`gwjr-maX_6j`5H0@l$Be%I;9&MrV{7JMtvV<@^3)(@bJlc=ci<%^sz!PyW;EDmS
zG0jBY#yJBUc_ki=PDbj|ikV+D(nhe6hkn;jx9yVGCy$Yh1K9aan*FzT!#j7$onN16
z@a&5g?-@%Po7KN1ALA@`-8(M1MS}RZFHfrej5ConmefLu|2!g+$=W)uxRtpM<7^wr
zrmneM?5`y}rK5f>YpA^UX26c%h8>U~eW>&FJhupFw1+QKTE`Eu9}{>gehjz8rjxC;
zeP~c|N^R37s~h~r*ibV|wyONq&{^rtAMG>CdIi?ddC0KCtfogSy75WTgiV?V^b+2p
z<<~QWE!`zg$m#8dxDq8b>7*QCh4Gj(fVN~k;+p?fpl-JO?uS}1lQVaj8}7B!b{OlD
zbH4T_splqxgcet5pyhQ-U=JCCYkJ^1xikusk8U>m(eR)@XXsa&OKj*xdjHHh3I2iX
znED(`ZZid~CtXU=Jp66c0O!8z!n>-*!>!ERPVtjKZ<`f;3Mqx|wRDS`?WpJi$1bRt
z6Eq1N7Xo6#nA{ls^?-qtYL`eyAjZt{U4)dK!JoH%xn*kn=h(r1sD0ccSlJmj5=I|L
znexdK;D@TMr(H_YoiH3GFllmES4aI!6`*v^{(Mui6y`!hn}<v0JKCiqm*A|3q!i*1
zdL*Vj+$D=b-(&p^Cijen5UmyP`v}NC<sZ^&pG(vyZ|a_^0BsR>@SV`jc#Mqgn`?ne
z+>k_Bqwz!?UFCPBD{kT$aQ^M}V!AMDv^*2=u;r_gRD(o=n9lsFl?4D@;OOj8*x-u+
zFaznwmzAZY%lYHesWvg4TDESla}xW+;0(GY3Cso|VTuGNn@zdOve9YII8i?qU2>b^
z;k}E=*AZ_)2IHNRbEm=<h_2AI_^x9PD^^mW!%2Yg@p1(t_`F<_pX97<z@6_NG7syY
zpYGkYy@Ty1VmRc8Cgd|UGl|!BS)=M=1|2)vTq3)JJqYc&A8}iQCvc?;r`V?!>2*5j
zn6;zR>G746qaKq_0vMg44Zrp(t1{t|4I$``0Ok>!M8xR?Vt0wS)1d}{46xKz%ye3m
zenm$bATTO5>2mNSFyU*%Qzg7J#J~Ys%UY%YvwvtOz_&ngM#JCNJR>>#s=6e**w3S>
zP%2TFauvrh<RlR_pzt7uw+@PH%g=Z!1^YK+_1=ks<E^sGC0BaKm9e#m2c6>xHCj#)
z4kbvpy5PH?^s&+^@QE$1F(rt{+2%VusZYs}jB?@6pzKL+1^RJwp?|2^p)^(Vu>c+`
z15p<auI~FAv8T+|ERs#qLnU67G6647=l&rz!<=H>{wh>oL;=EM3T6c%x<jAg$UmWB
zqFRjkJd%79wZFI7BD$TNo5{MliS{aiORE}}uK)%Pyc&X!56Cl@(NuO(UQp&X`$zou
z>YNKrLiAj1DQDvj%u1N=uic=1Ute^HwUF#L2^bz3nS`1)nV=6$#)T`4u|LVQkqU(!
zu1^?rGgW>QaNmz-sGNCs%Jw^uMyNaS&PV*!1p^^SqSZ(Ug=K(!UEut^fWT$CvFX1y
zrTUjm3D8@gP#qO~-*sdk6$K|<1;EnT{u?d;6Y*`{I=K%=+QC<1=<lae)zyxBY-~Iu
z!uB+M!Z5x_C}@)vnXv=p{Am=l?{>W=DD-2nMA$8?LC=^8$xC@%o$ZuQY~>wZUv}#!
z_DYPfa2fqgq1OhSexLq4@AsmG<mU`?cm49<gdD?vB`7yo4K1hR|7Zg~^&N%Y6KTZo
zq+LNkPjlwmefapDxrdD9Mr8o=Bb-l6#7Bxf$)uYNb(9r$uWoXMB%P49K7JJG&$cz2
zm~-2EU&U<7(<@=IZ?=2TjAXIiHkzN3bjsak%PPPP6y<^p6PsmPy!CM@v8{AFf4IHp
zHC$~NzYSoUQSU|I_hnvlA7rq{H}vKZZlGhSX?$sC+FR3W1H%-Q2ae3KTv^pD*@YOD
zTd|4;8rZ<*ed&p^JeOAGCuad|z<SE2-xHxv6WG(?!EWX&aNjemX1TC48KQMy8sbq)
zW4=b@TSw-8lU5@}n^i$e;HbKm3Y-m@(ze-T>|&Sfl<EPSK8ji7w%zhb+3j7@`d~J4
zO;W^7ur^sUf5%9<L*!LMQ;nIR%UlM-$q2;C<2mi7uIu(<k5jt%kH70_jzyyRDdU%j
zAn^G-UB<1`@LR^P<(<dp6KvyAMV(Zk!{?G>%(8k)qb2lpcEgF!&wo&mL1)sD>4F;C
zN~C|9bRZ!X_9yw4cp38C$Y|OyR&PjwoosL8%TX5LPo|skK(dB0aH^3h=TN@H@vVz%
zXjHCq$`GSA&13JUjrUb{Jo#H9aT18G?^po%VyoCAj*9gf(V6yw0nN3lVFZ&5N%~JW
ziR8SJ?1-NhofMI9YOHcvXXuZ|ZL8oj1C6U3_ip(IsaVrsXbu$_dpHl}U!I7Zu}stT
zwz`PN9Jk_&;mtA)bW7c2q}8r*8PY6_L((3zwq-L;7?u;1>1C^=$5baM&aCNLCy0mN
zINbWzZwM#g2{Vj!rBaE;7zZ5BpGZ21t-36R46Vd{PI}7DEqE|yziGrM<vFg3sN<b|
z&NJS)M(`{#jN`x@{C<Lbb$TYNilAg&U33vmv=_yEOS`t=+8^&1e8*Ul5sxAxcpK6R
zXKQXf@K&ZM^s97E@hRYa-jQ6P2Dd};CJXUC(5vtmjIS1)Z-mfOpa<M<K5o9}yL3>{
zT5rfDThom2^X2kt{y><YPK`cJ+aYjqeuH)?RaxBgIB1K84H)ye-3jVp<^|JthTMJm
z$hf-G?qVBJOX+_n7F|!C$)!AR*4bJRK(@+g5}BH|(#!JQs2>LhD`&3gZXX3aXTQE+
zR{AyivEo8A4MaUC$)dP7Ai=XuVfxXYzBx=bWcUUp2FZCSSn>}vklsFihFCXziAFzH
z8#VeBWd)-Xx_Xz5j(g`g^oOi8(fM>Kdz?naUdb6Xfmum-{{7c;oKIqc(A<Mum&o?i
zcvVBwRyI8%y&xZ%p^P=H27SRZ;giq@$@(IjQvs~>F~AH?6wtg_?zAL1wo0!O_MGl<
zw+DOhV@-=kZ$5&=)>`$dThG|63GfbDy?6(^ap5;ih10|H%^tKCcrQozpOyY>X%2Fo
z;o7rr@lVyAx_~^_FHKIk=PhU7e6O=t50e#l57ZD@ghJ-A9(}hjqQOhZ9$HL9U|amx
zslB^PxFytF4iwT>4S%6la4iM`d8VI#*bLs}{`RP8b&r{-_Qstvd8exiIJ0+vJC%xt
zN1N0(yvfwIUOxo!Kje&NJ&W8*FX~b~HIM#VTTk#W?DrMQXI*@67sSkhLFI{#3Rq?m
z4PMOMVHkiSs3sRSP1tY-^9JBGd&)Qhz(n$>#Pv)Q0P1N#)!t2pmq?IlH6ZO=%erUr
z*TBi!h$6GGP3=?mowbf4j*JKWvWA6n<UC(XOM`8Q?;$g38bh<{j-k3t@qFdb#}p8k
zc@m@$s;F8(Wvp}It*3p70~9dDG`p6v`rqLnZ+(D76OhOkj}T!JnDVRkTA2=A=F%VU
z`cA?kD1VP5<}#pFj_}P;Yz|@r0BL8gMk*-gmjo6WKRaz0BAHi=-*O}(9bWj~O)r4k
znpv#_2LO@K>+!i2K#v1IubScY^rJ-xc(Tt5z#u_BqSqX<1NMG2tK~Pbffgr~IPi<F
zTu3syiSTFqqS5DL5=tx>w;En3+${irKH|`3ep|it@Fi^LaPDB}Ut7`N5iSa-0RCE8
zOls`!S5}Sqx|Y6qH#;<>bwpXrys+a6WV;7rB%aYkc+_<0Gnp{~1m<|Zg`!7W_+V>V
zXD0{qKJzsUK!Rtw?#7nKaWj(PuI0)u0l!dE0Sfm{-A5JyAaUkp>gVksFKqAGB;P3t
z(;E)jVMevA=)toDy7~SGhifp5cAkI*{*iJ8Aj&uVB`dTVNXYLeQ@5!3;u`cJ5S40C
zjMSzZb0vi)`n^rmp~^fSD6bt`USJ%Rn5W$Qp~7omChz`FRiylECHnpcF$5idJK$b?
zxmX`b%sY|a!FN!+(Kg-Y>FEd{Wz96oC^okW|8-gX`^8Uv1VVvoI4A!B*Dn!XU9@Vm
z7YjgP_(Z!}g|hxKL$L$^Z|43A)noF*4v_l!qk}6#C;zW73rWH%N}wNIBlEpkzjwql
z&oSCRp(N=0bgf?j(1=4g6oOS;3?K^uWV!dZZ1}ebLc!57Iil%!CM2R0yx8t=xKN`h
z{s&+x+n77w1DxX<Yt9+3jiAbTIFR|O=RsR7h1Ue+WUZ^^OdH|#t$D*%!gr8meU>`#
zOD(^!>@DMo6*=BJm&j)b(OlbQox&(wbaQOZRp_@&&}pL>qV#kq$qUvjPZNgc|7882
z>sWXZ16Y%b*0U;qKZtKIoB)-{<BT8olC^KDkh{x3EDFx^t_1kZ(>$2^zw)+z3n8)<
z@|#ZNimvYb`2G>#t6PBw7FTdEkfMJF|Mm6?i$ndoXIvj2QKmcrMz>9XB6^GwObc!R
z*4Uu5CYChg^M}*0iw+`tYhb{B&7n7zF4#pwlA81iuNw^fJ!3E&4rWG}8#ZHTX6;4(
zWi0*&g+J5cWno*?9~Aw)uyGNVUa+Q%F>muaPKb(nJ?|zn8llm6$TcPu|JN7)zeBGw
zkzO$LP%)(cz-((m#nQ<;95<hT3?<WinHxI<Fbfs@f5$K(0kC>i;jF)(iT|31|Hq%~
z-n_u>PbXE`{~PG_KmK|QZz$k_$5Q~Zkp9ykga7_E@d>?zhCRGhM)`ZC@DX}jEt}&v
zSn;bBFIw5{SbuM1v5dF^z4q7mkAFhu@=HZ*lHq_={m0k5!M@>0Su<)FdjrmGk`vQO
zXK~ly=C+#jlX(62pW`1og8%31{-5vFt^rJIkkEI|rM@(#nM3YIK}|5XeB#O>HxyFC
zf3F3?sn`VXzQu)R_rFw^k<JMeU6ZQ*gRcKsJ^!wuztIqZMMy%M;=}Y0S!Ddns>Dj$
zZki><fv1yC|3m6_A_6S^;|A8&o6gJK@vh7L=^^IklTHskTOcpQQm9-GC%dfeX(7#S
zLh5MI>e&<dv{bX{4gwryjejb<WG5W`sTy;5f&M)e13ER!KoW=kO=Hfs2f&Kl1SmU}
z<b(RUoPme73CLIB_bA*xAMLF)5_S=RN%_$ZL>%n{cil$O$8z8X%3mWll4kPDF_qsv
zh1V`T4nF7_K!_b3H66D?`2e4-Tmk_aCryfU0Q1^GUhp!`6{wlZ0DzF>jFQ8#&Es)N
zFO+$k3*z>DHcuuoB!diKKP7Sl#Phw7v3F-iW&tAho&dk~{LUSK;x(Rb^trwoVreQ~
z1Y#KYSW$YK84xvq43CIfwxa+P!u1zs{Y_ru=wg%G44%G9xNKx2Rz<w_H&%Ni*RjQt
zAH{b-S5Fwp`)caiq)5mDsWw3KoBZB58^W0?j4dq?pZCLikpCaL5)(5hfEV2djA=kN
z&yx7_ZPKOcbq&ibu76$}QY91mhfgb-bxS*3b52nQb;1y?P??nw-<KM9znc^JH4HsL
z5|}4H;$?sYU8?!Flk}Hfq8aEY?|}dVI_7);h6RLgZC%ajqrWim>vrXhJyroZnxndT
zg}6wTVX8%~jf%FreQ0bOGrRukd3hJwc`V#ZlnU_P7pYL2@3pubwE^3_<v(HPJC_<O
z6zHV(Q;J&7xaj$^$?NfI5YK1#S-@)`3o(&PNQU~AD4ywyT~JpF6nd4v<2Rme^#6+6
z03?)sFHsvKCextEW@tBnVOtO2|K^y>+o7CY)8^Yom5;ZXGYmg^Is6f!+In3s0Hiu-
zrQZSYkmY)t4PtC8WR`jzv0(S}5$+d|<GDRP<Z!XB>^YV9`_EK%{Qwn8Vs<@070P0q
zD1Z;bi-U1mR-$1RkTt*mRI68d{1f=Km`R1)>Rq>!jh9}gq%*4esE&ZBDA!Y1rHEFz
z#T~E~)ZGYN!fzl0I&%5YF^6sV7t_x*@OI;2G~cX;9(h5dorGc|;~xl^j-9Eghiks$
z5ZssTRdz3FnMyw7wVI8|eC}IuZkLfOws?rAZFz$*krzasO-bn>9;f$OTm%**o;O#n
zBPK4=;e&el6$80ot|x$8^_AjO=$x<%=APUy=~m}$k1C_8dBLWEKl6gF>1&RXT!tLy
z87awh$HGj0?2;QN9oIl60M!wC;&@xneWi_f2u}298vg<Mmtl5GDe_cAn8OYKa3v9!
z>;7(j3)R-b744`As_FVu7p#7q*GTZ3;^|{Fud&F15AU~_lGMZxE_ddHF6WZ(k}u>p
z_li89#X{ogOS`{+q3j#{LeDrB<_Q3F>G63J%=pbz%3?X?cWJMMJQ9f=Kc$9=kNvd9
zkcri8cRE^Xh`teCtuvjZEd-QoQ^MnhiJG)GGMhRS=stDM=i37SK`GC4^5>^kH)m$A
z#Tv83^!{R%vUFm%laAnKhKn7xD3ntgVp<WQ+f8g<U_pz17je<&0AUZ7zjf**H0}pZ
zP?!taV04{nbUf=N-%yX)ymNwLP0H3JjKEYjU7s*w-cNmPc6$?9q-hyUVz6=CvBDA9
z2w!J2N&R#%umemMU+9Irz$3x=fzd?39F71_B?5>sA_N{FJoJT&#Q<&p!j+n3a8*Ye
z`F-^0)2n*(rFunAHs~T>`XSGU)4<xww2wwF!9Fh>cw3-=Aq%m2`->o${D$-f@+JKV
z0GRaMZ!ZsK@h=`vZNz_Scmz-^?mx-0?|gqmgs}+4XVEACzi+>1U10~Nt?SRPNaa%6
zff$`jcv4Hn^W(PX(vsdLuw{WlmQVB9AfpenrW+K7tgwM4@j~EAi|bGjCe?6O`=ZYR
zcxZF(3*RJ_wdnRnIBec1Q-L%8bgmmv`?gN|$Yy7XY>t8O71|pwMCIb}mwIsl&#3wZ
zZl?HPs*nmG??(}P+^lV50n4T6@8$COKVL3O^}4YScN+qRyhqxhzJh$$@z{$bjEPz{
zy43G`Lh)m}zSczPeM>E}+{-qcsO&)G*uET}C>$~+4E-s$Mo#HnHbQ~(_UNz<k^l$|
zRDpb+u7ZINOlv`H?~j1=ok=~B<|wf!)e5~HdPZjgry;X}>i19h3E#`F449zWeL8`Y
zVIG~YMK4)wHHPy~8J@W+WBIf?oq0BOVmL9N_*seb>_#Xxsx@ZxxC1NP1WNaSh-nN_
z#y1%cEg)RtZM;rfxgSY_svAHb&6P=H=@1ymZ$c7lF{|=SkRSM@LmFH9)Q-&BoXM=c
z6=;339?F_9CUh?=1QaqsXXNqBMViH(56jm8S>E(@_Wjet(%5@~%Dkg7`B69J6`0Q2
zp8Pb26c%|LAftt^j8O;ku8+)fijg46>~syZqD_&)RqAxQ?h}t$o+p$3s(4+?c7R6N
z&-W-FFZW4}D_k>=QE<sLHvAs8aFj_B_@h>9HQE(wG3Tn3Y2E3^02|anq<jeOy`vkM
zySnPnw*6m4)p;W3soYj9@o&CJU)nUM!*Ga_o0GAcXChBDNqDqQ25b>sJo&_tifr`X
zOOm%CDDWg{=iU<kRRCUN_0vHG`}ul^m}p#D?|Gc^+V%x?c;tV-nv3z5*$N7v+x}tw
zJ5@)QR*AyFch_QjI9ZC>?2kK_XXw))zHbbW&p0hJDwoQ|%#24gDR?pu=m0GSLOVsa
zzX9*P9TtN6tih0%?wr${4$1);Q2#Tnf5Q#Gy}|tHnGmpwW5PPw!vUsK(;=z;T=)qJ
zna!>cyFZ~3d>M5P%#08g@$jiJW%+o24jp&CDV2sFv#XH3Ch(QRn}6#NUwTnKz)0eO
zo6A>4zI8}$>i>oBq~i>@lNPCf8#W^Nvzxu&9RT~6eAyc)<VEJM)j@n?WYkfIFr8PY
zoxNq`Ev}n8KzS?`IcAmI!+aaj90_w4`pVz;Ry<{l;vePFjWUX5Z8c#;eQOz!Y}fEx
z(U-C}hkR+&+8@&xAfqW%02Tsgj;&a&Dd308-M~b);d#<r>-Ak=rRj!|d$Y*Ko$ZuO
zN!(@E?b-6|bG&mKQ`8s6X%3zHFY$br-v*@eN$D;@(?Al)M7O-*<*#SvmKyEwU$)YH
zvDu{^pN7vpv@m!bOon{(=CkCe+A7Wm?lT6-+RCEs&1Qzm7fGW+@6uTViD#Jhi5?0n
ztqge&!)ia}N=7Gr_khP6Ri~~pc4v(*`W(YWSFJoGOkHR9%Occwr96)>IfTZS%^K!<
zQXXt-=3lq$kUI723*+FyuP;S|d-)mZZ!-!~u^(=LII3zx{_lth4EswB-{Ype;~7;3
z0eXNA%lORwD3JbAkjFhNm9!l=-2-4^)mRq;Xja%Ho%$&PO_1R|@Br@e5BehIk_oM;
z0!1-aLgR-nJQ?N=@W$6#bA}g~_$^@#Z~=1M8s;3E>VrciqJD{`n2H%!#HYJC1?t1M
zqUakJ0$(?|l(*wh-N2vpBkNXNgtt7~D`Xxf$4b2(Btk$;U_~#_yA>zEth(MyzPENS
zY|l2`Z1FGvBO`0Tq<YfUSSS{VwvhcBhyt>UD*v$$i?5N{lSlAd<VHG<KC2BWFrKYV
zAOsW*uc#bNH&ddXe=ahT@($M2W;0r%w`I&4p|M8uLsvJna(wyQd%%#zdcHF@J0zvY
zD={Sx{7395WZ+$i0DWO{yag23-^@7b3rxnd7aDPDdEjMAfCE~MIiyuHvEh6S3}%+e
z_YV&YCLg3vXzD&YEB~XH5EDoaYd;VCBUv!xegx2LNL8m&Pc?kIZc56}bwG6OMX?^Z
zH9elNMhMs(GD?facz1|fE_9l8VCB$LPlNDl4I*;1gw-SRq^$svu?0JU%x7m3+IJP*
z?plw;AZJS17SLP4@@_^Z^Hq-x1;~r$aXWv7GO~R?z$S@@e?5q5c&ysOw^p|h4uOoN
zJsTn6#8)iU=SjWh+q7P@PYpklni26J#y(P+?<Ie=E^w_(*cwVaxWlqXVbaJB@8W3E
z+Bwo*C+_s(FeOeP6m<^^{>bBpo$x8fMnuhWrCIH_yco#mEVLGSMp;4YcbUvJb!bx?
zFA%@9oY4bQHcPXwuPbHS-Ep{Nk$hk82wCRdn<|VwGRqkYGlY!KatVwAka`A{(jV7X
z{l{#&K~C8YIyAUr+%iJz$^9~_tqtb$m5muVR){Vj`!;?ik?RhqN9t<+AozWGCi~E9
zdW1J$>nzg}o|5^QEiO2_dZ90U@AawT7^)uD05mh!UR!61`~B5n9#F#A`+a8~xfb>6
z5VnB_WBQVG5Pncq)3AXI?_9xnbIWHc`D8M!RCmxsYi(%i_5n*JFNXwWg4=fXR=1C!
z*%U9%FzG3o8%4wuUe`2z>qy-|@6VuLR}*RssmOt5>MfE?a89Y%vaiRhVzP}iqRS;9
zekc>tG(r#n;Ylmu!*&;K4y5wT{wU$<CA0xRU6~4ymxoea=Tjl`QFU^v95p>o?Jp;)
zU8=0e!=p(+topUI?a8G9(6(1kt5lnvYD94jYjUkw{r2#z4Vxw461n@OJG2XUWJ>R2
z5qzDwePOl3l(OXg*)#X2iw-KtR8K9Y3ZfA$4a~YCgN(n?Nv5-12kucV{Mkr8fmvQ~
zxkmVU-PDz~=eUsR?%+t=c@!uWS#CH5asX{q>h=z_tgTsLNyY;evk+_Jl3Ez*<s)hf
zW`c%TqnPPJbNPT)?YxhOH!~XRiDXfFh2T0O?h{l4P-FVRqom2t=T@Iu^<ek-SM_-B
zeV&8!ypUtPpjlf)wi&%wx8>9~<Iw0NzthZs?lVp+562EG%NpBMgV!td@6psA@4wC0
zEwtLM_-k%H3^jg`Gl9%`2H_D|E?0gD#m;au{rveG3W<QIPzHx4*<!s%j)t>^RYSFQ
zWP#b;56^fet@8H{hf6eV^lOwbLHE-^X#yp=FkWM=cLi~bWS-f>PR+LS$F^JIv!<u|
z12L|Qrv~Z`Tj?U(cAXZri15Z^=_QFpUN>;88LDTnzSt{>Vkk4ZCxvdeXNVh#`%-SW
z_kd_;SIQ(PpY0Wg<q}WJI}cq^G#OJJa+I~=PkA9>gq$DHzonR!_jq*!7KOYGuV4m^
zz}INaNmTNBlz7<HD~Og%o}sRF=JS*ou&sdb-qOktXg<1B62k-NVb)nq$Fz%sS=N1`
z^9{g7Ff{}}TdJwfDgyY8K9kS_nKR)xl*NXFapP_$AgF#ITud9e$DH*Giby*EG0N_N
zBOt#A3eLvI2#P2CBEu`JhRi+8u`zN{Z`lfBKDz=uyDZsv77L$4pU6$Rgz3-N&ej|u
z?)m4q69mwg7!7D3E?If$MnZCQ{o7MSY+92Ja@rh4v}hrMOyMW5JOzY+=B=4S`QPs9
zhEpE|NP+k~O(FI*7)+ETQ6?A}v!olZR|vCCL-oP@FsvBRqG6%c?PPGeHqQXS<#H4J
z@Ln~VmcYEvc2&oG(mx!&)5n`5P6Yv%t-@!3m`&I{&}AnpkaU8QJ&0TpU$oc-Xf{ZQ
zcR3J-p9yrJr`m@TviV73L3<^wN~{PIdPkuNkShplzM=@pKu#g$+E~LcHHfr7UeQF3
zk1mVOVCM}7mx&2^0bywTmQ?b7C62y7`xa`<U=t9?d0q<@{gVaoCP2)sqnB}5lt0`9
zcHuiyco5>ii_<-H=<b-@LN!WQSR>+!P@U%jV%hfSKq6n}`EpH9XM#h%=ul8j2d=&a
zL%)rdfgCgr(+O(kP0<z6w4f_g!{U!=d=qt6+K=Ob>Nl2mi3Psa%y&_i-n1}diCQj&
zme`8;IV=ETg%ID51||lGiw#eQk<fg_BI${6ifOrt9hVi#(kO{6^XQi_;Kw}v?q^%D
z8QW=KD+BFr3yUHdvW!j_*@+LQ1tAR1t^~F1F4apg6Q3ZPGm2TD4w_7!q~|8IwJ4vi
z4{tgKSe8KchMWMni@_V7uJ<&46m27PMzvr)!|#M<7g#}j_z~U+8H0R|pg7Kq@WvqY
zF^@x4jqcuTNS^9}c>*Fh(?jZ{G|j3qOsPnB+bFF%*4*=e_HJYInpiIIVx;7~e9_PF
z7sypbMVAj8L=HSsE!f|)(W_&OJptiL<CH{Gu)FwrqUh_Hsg{ilXVFRg4@O{<wBUP-
zctz|4^F=sJ71&_kLkBh!NdMF=S1&;`u}?Y5BKSOjqW?IPX{1UP0=312t*=5$ykhZd
z5A)|rO?+89#&OIpWQ_#tcuaKxPh;Fsj0{y&)D0o@=~XYG7SN<9Kvy)MN_IkT1DKLE
zugH$8bb`5X-LWn>Pot4W#rg@x48>L=@M|B0B7}$*R4xvG&vw?+FR&7K)lUK4>W!;o
zjG0V}B!*Po`^R|>EoeNE`(L_)^ed<eS5Zc(Tow~#H5>XUVFS7k=w!E<fF^6#Dm&8W
zTKj@~q|ck{LHHpYIfZWcQTHy0>)wf0K%|aqMX+<~k#=eR<Rhqh%yL~vA4DtI8K-D9
zhZ(`PQOO0AUETHnTiIQ@+Tlt*Ka8}6L<Jfgd3q*<#4}0;axO~cv*b&|zPrh6<&Dh?
zO&mz$Ude}>1iZbmS}~9Du%bM6Zac$IBDLmEN~On+MdpxNG3j^~nwtHYIGH;w^848Z
zdFelzpB#P#chZC=3~)xak()iMGS;CriH!0cL!|r~AkWmGS*w^V?#QUht)W5n*c@0*
z@bgJgEp?D_`r?{igKqz{?IH8HPw6~o+vM1@{`9v!$8qw8?QO=bB<zSzX2nv~Q$5>m
z#Xu<=8k6B6_p^@A4XYPgeChQLkZ<?>OSqh|l}w2iog%VF!^xZy@>+W3@RSX78E&7@
ze>d>z*y3Uj&Mnws6m;m}(Xz2i`3}5OmG`Yd+rapM;?;3SQXH-ewGZ5a6QtwcB+FFy
zFry^soEQMcYL^LY?5^^9=xEH=dN;sK50g+&&SxiI+IfZoM~mqwYb2M<L9@>4pd$k9
zn_yiY<2}IAOmN5RmKM$yLBKAmWAj@u_{g#4nxL45f|y{pYlsQXx`P;641sO}PFqWi
z9GU{-y<9CZtq}WZ*jeanJ(Tqqmq_c1J_o*U{V6fbYmrH+^o&9zSdIZKpI1z`sBn_t
zpjV9!hrim=C7%3fP#!uOcN8T^^5AzVOw6W#ed&fOSo3<@{+^Pl?&8n9l6r;6tylZ&
zTP$Y~UbW5_RA`>yR#I+{pOrOVK3H@^V}o8oDQi(+D~M^?*x0`9Bgp3Fpn22Kf#J>I
zZa#so+x4)Wydi*v?}8dUfp|$g$XxuR6AHREsk?lyvln`uM|E5DE+<_iP3H{eJvoNH
z7|uD69lh)3?S^m=XE|iHN1rpmQK>eY^jb{TPe%`Cova&IK%tkd!@oID;Pvaqe=ECL
zr6Mc6__qaAzlCn)O0!r*-F*k(z{XVaa|J-zh3_V%htT)Y!`~phnsGVZalo+XdV|<)
znUv%m^?OV-=bjXFUs$Uvk`dhO^+a4uMn#s%>RlOWyJo)&A&0bM(G(}>>fzx{go$~+
ziWMwS`t9jMj)~paFXn{yf2e!wxT?49Yghs4*fheXr5luP5TryDVIv4^LZn-|TS^28
zML<G8I;A8w4bn)1bSmBXuI=&sp8K48KhHVOecylH&xdn9Dz*1_tu@!2V~#QAE^e=%
zM9V0(*o)hr*WGVksD$c(@wcp?WT~zyXqUO?XP2HD>={77JT=0=NOSTdvx_M9wzi&q
zkd4%<8Q?0CPDZpazIrt`zG^Jv6e&t;Dn^Zv8Tu|C*NAoj3+4Y6X=6jTF@91cn|HsW
z*{m-cRX=aLb!#1`oOabL-8vwV2u$<2rg~5IJSdv9pFC)GxGx9`^;`}X(?@q-`kc^U
z_&WqR)?WfQpEXe`+C6EL`4y|-3-8O14!9S^0O{7nLEjl59VCVh97>l*Nur+)x)%9n
z*}dz1CN-oW{_68KK6=O{>*Y30u#$`Cu%9>PsWl`x{XE@Uc6#xDLTye0op`&4`DCTF
zb4@yTAd`JvQTkTc?YmOl8wilvc5B@{2<FiB&So#frnGtY(=f3<CU?&|g3nbavujY{
z>MK>SrcVgSY~QSva1i5f34Ythe6%&up`DWlG@3NiA#&gG7L*|v^VAg90;ed)RoAxg
zDQIOca5d;BY%muiadgenr=@+M7kBHLGg3o$gfY|l<J0@11X24u!Nkpdj~X{1CXt&F
zgm(sWorXZ_Gkih_0=Mngg2?rA&gLlO<iE40;0y&PmENrU9(^iC)n<v<f7~3Wk$L@;
zo;AePOh&jL(15VJFeMoj42Ws*;mP<*C>L}Q&W3Cj!c^Xg;zZ2q6jF$NLhHjFRabu-
z>pG9X+%v*WE4OUC2{VGOnXP(WaBW1?G#eS~jkF0@i>Xer_(`K_nyM@Q%59P9tsGkA
z*7#!nbQ$YfL)V$6!zteS`~3qmr=Px$!`|oF)J-LoSt$&A<-O>0Npang@w!aXrim0a
zLkyo6e>^?cX#ec+(d){u%wqIBXG3fAxHaka2G>gCv&iH&H7Zf5!gKa@@pF-#VdMA4
z_eI&4tK85lm5*0&Y22SqS=OIef5LB1993-p()QCJ-Rt;qjVjwu+A*lRyrWr<6i)1}
zOM6O>d5@RHtaOM99&hG3^p0@O)?-u`w(nHS=m)zStaITR`Gvw34CAADOe*ohU6da(
zIAOvTMttj(cLsjsVA(3F?TT7-Dk!RnD!gweFdb;onCDQ$s{!A_vn-C!isKzsI<@T4
z3)KwBj(U6Rwv=&Y7j}?iCsTs3g)if>o<7DfT75=*lHIU;Lx#QybxZfp;n|pscsu*J
z#D0x{nkY(Pht$fNB6M=+w`e=VTJEl?@0DkI(c>+ARTEh5xUJ{#WHSzEpuoKJux>NB
zJq{y>npP-eTQAC1WT1Cb-}$halcf*6^L#6($ZI;A`VQ}p93^u_wR%zIYqP9`Hp6bZ
zQiymB-5u#K!KF)FRG}<gUx*IDCu20rzugMFAA%e<s1Z96MMNfM(J1D?lahf(R^Rth
zkHn6}N_==RDucO#n#s<7I8#}Zb(pR+15GrFQg3AE(fn}E-f>JJKjmK9E&RY-%UOv*
z%%8jyuRsOK&-`VWd_1ajcs(a9r|JXNJDP1_EAe@B?jU02;WQnf@up&%I%rU_Uc}Np
zuo>Hb5NjGm^gwK^%@NF+D$8JIcK$-SMLF62$h_G~oXU;ng7Wht1_4DKH^;a;A3NZt
z9|kmt%cDwp$wi*Mh+9=W`V!awRYR6(dO6W?jG^GcX58~G4dR%3FWN^68a%Cz7P}J~
z#DhEc+o9@jvPvu0SxcfpFQZRb*^YXNcEW)70mDN@othz2)_b(vT|>9Lb2?qJ9anml
z^pcKWYwCcID*DLle2&#GB|xzGXIybN5Itj98<%o_<7M3m-E2SHoSxL6ZEF!%2-9&k
zpzu0#?On{{b@mZ&cq9^4eEOv}&F5pw+1rZJ%A}VGN(Po0spd4#1Zb3{GtSDo*rNjK
zxp$M@T;!y*yl_8$@BUiWBWN{E-~%wz9DxCERvAlGui)Z5f=Z{E$HX6*$24@QoS%ty
zJ=dfYIF<|JqGqc9`WX^vjbXhX3DKc4#?^nGTU`DPnE}4}UJ0v7P0B6daH06>&AtZT
z#M0dg?$jR2R(FX@1TPsX;_4%9(D8HoGE-~3|A6`w_xaGd-WYhd!*jmV;XL)Ltle{b
z(T&ICgF0bZ&<gi^s;Nv=yDwDJ%4V;%-R^Ifrwrgb{oGaK{?(&YC`WLDcSx#iV5@J@
zH8Qf^B9Z@Pxdv_Jns?@@tIE;)I%7ZOz-88$EE+{^S`AU<&<(Z0e($jk;!WurKN%6x
z#a`kr!u~(r)lnZ#GpbIfa&46hp8k*-Z1jqb+{e^aXpZWhpk#4zoRh>tqrcWFLXGw6
z+OH33e8LWbIm7AGiubhIE!d}Yg}UYDQI9g86Pk5+37ufC&_Ow$<mc?om=&F-gwR5!
zB_uWpC3u%uesl}Y(V@lmHS7dBwpJvFjLr;xce@H7op0;mkj}L0Z`g+W)5f-~TXKBI
zq4RFK)lj-j?npz-*>xJ30o-yi^QOo-HfW=sxNG8s*9Bx$asN<ZJa44qKsN{wHjQl5
zh=&}=+ti)$f$@QqBdGVr*`eicX!Ox@RaMPURRlid)%!=kd|B$BHPrVmf!@TA;_J`2
zC1tJ`t-i7h7;vsYBZ>#=2{?}wh6oAbZiI?GD2&{e@zN<`p?wWEL{H~~<mFQ^=Zn4G
z;aR<t+lGpAV>pQ7_C7w%J6=NvMK9;Vy`)q--lwL+;!<y{$F3U+7=~6d&Sa9cut8E-
z79Oe*ESQbCQurXOnlUk~7sH%}Yh?x*c|HhCm}73!2MZX*^G?Nn5qcxYsUt=FG&?_H
z_4I)b>TrTzR;k4XYj;3YDJyyPw{j)#pSbAFNV-+ALBJ)+gh;4$+}+jChG*?@8dSug
zZGD<BBO*NOyi&v*$2PyWtawg*_qT&*a0os-PK^DYR13;+KeJAPvU{U?DC%|zf&4Tr
zm}HfS=OU%96LW8jR`8@tYkJH3MZZ^hA4H!$JfcRzfA+dR&bCrW*+Abuj`o<V;X{}X
zmD8*CM#fN_herz{@JXXO90r4Q+qD%4ZD#Hh{{F3U<BVQ;j|j6d&Ww7$p2jA+`A@t}
zxRdx3t5o#ozlx@Z@T4dJQ;EA>RJIOYB>Qtio!Xi1eW7AD$R)iEHHG(UjJ%#|3Y}2&
zxk<~r%ml@KVm9A=ebLyn>H{Z@qQv9{>Q=35(POO<sOisdUR|BRb=0J3oTq*SF+24E
z_5pwzbb>ALrKw^dIO#Ge(%Xl*-O}ohdm1k=fPzuIWw0(}uGyg>8l(9X&Zu6sr;{jY
zG$xrM29-Hfo2fk6j8UFWE!ZlTIs2az@kk-&1-#vKi0}U9vlcI+d+`8s>U=%#{2Oz6
zAdR^Q8a1z)-!YN2F_j)-%o$z;I~#H0x!-UO7wU8FisSmZe^e5j2+D^*>iH(x_&z&t
z{ez0K)2$m=Cshl|b7?+J{KOZ>?gTuBHlG2kEHMx%bGI(e{r!`AZFN{}L5*7xUjrXj
zKofa`Wpmwa%ByCpZ@b+SyGcol(JSQnmxkAbbb|sa2b-y`G?_n^`DG>b2esb(49qe{
zolfUe9({hKg3^YG@oiUXLYYXu1Yq6<2q{dAojY#&C+!zsH4i}=qr1~+A}-Hu(J7Yw
zDEt0|ENgxs!Y9-hA56)Xz<RA(_f>iy;70S<y{~szI_|*d`K%{yzxVlg|NYzTfsZNM
z#P}ccBA9U>B*(%re0zN1>DQfAp~eoJDLAn-ZGFW%yqAIF!G!n|9mJLEPpQ`6fxX4N
zE6$0WK?fo*wodf~WvXY4dc_bwrCY@6aise`{XV|F{hrbv;=ei%6Xi{~r=3-=+RB*Z
z#J_T+J)90%dW6|@Ua0ml;^WYKTFGf}rBDoNG-$h<O9sgZ9|Q78YwPvyi4Km=8qGaH
z`I$bctGZr;o_9i{BmN%ys~-xen%IR~fKX~~bVUfR%xtjq_=u_jVfC$a1^?9cY3++|
z#CsMutu6=^6S$v3J-_<yevIG4-wUlenpicX?6Yn|jYUA#Kbba09w}K{4`0A1=j&!}
z&^v`Ce5_noWrUXND8wrJdRDQLmb{Cc{e5Lb11pxoPRqxrdJ8XpT^x>5Qs2CEIj8%$
zRuN6b2R4bGL6z8fdWrRE@A5oookiB`TxrdgNQ32_nJ4V`AHi?(6xl~zM@3(?OKVtt
z0Z`fLxz3MXp3oa`rolcjnkDl4y4EoMi!zh5izH}K$$)lGo+lkO(YFi8#T;s(D{c5&
zr<P+8$+7LSRjM`uT{lD^@no8p{1jDXG=>zS%4GKi{}#ZCFRk3OD+bI6V#%sT72E^a
zG6dU9ts`SYOgqn%d_iaJkKPpM-0`8D&_k^Tc4q?-Rds;@Zqv4b&W{fZ7WIa+J-(Ke
zY6w3^B(KLKnvQSutL`HSY8vEt)S{XMfrjQBQXxx1yw?bArH8x-m>Kz8gA>GyaP|gQ
z4>&q(JqIF9t@j)t!A<<|N^A`4tdImbcH*J&#=*A0O+O5poQA^ohq`VTaJSc>Yts@c
zRR?W8$A5w>I1R>(c2T<B<evK$r54q!vttA}UYs5bHPEN1-hw!`Pvgq7l-kYK&y0AF
zxgtB<HTVKXX|8kr{lw1ie+BW|`Y<AD0<mk!Uw!H0l_L^(5FbOrbpIuOV^JLx7VL+z
zI0aw9Hh+Nb?0r^c>{$GrO|>k%@D)Y&c0%o=*F8BL@b@Q3Y}5<W_}~!0u}+3=-!{Hp
zbiPa}uH$0<@P#{x|LqU7^c^wMcrdo~oAX?>)a7%V=-MwpXR}`cHnFh%d`lH{Fo|aQ
z1LPgT0n&ovb-Hhl>_)6C#;T9d_NPULK>8Ln9HTwnsn#oN9@So8*GN74N`bMI4Pd<|
z=a?IntY#40dDM`yUQv~mY?JN>pXQ1O`dS36+flbn0?_f4*+eVXFN>}1^|$<s?IP>n
z7`&bMBoJxW|Lmq38{eR-J-d+*RTHnd{}FkW=A;aj7o#{(PX$>Lg(=$2Y&?XUXZS38
zd4FoKE^(6jOoxMEJzNXlC7tP^x9jUZ*!pfT>bBE1zKboFE;>{<BnXhlK6`vjGy>t-
zYu0Z4)XP>KlbhQnll5z;EXEh3!wKF$-x<{HW_(Z0-JvA@UhMU)t_ve2HfJ+rwaBzP
zo>>^{^Qy;*S0m}KtxLU4=Oypx0|NIB%j!s5tzu%$_qY|+h9L%nF=#@M33A_l#KRHb
zwlq!_XO)^!M6p33!LKp5OGUfqi~Ml#m6n0vYmI*{Ei*df;q#aNyt>)`#z2JWkb5)2
z^+#di1}jumGUFqE()z)jp@5=i>9oYb6q$a_i(Jf$?uS2ay1c#uw(a%+(RpJRMoc_v
zztFbUq~2hvgZLvtFW}X?U9^<mgrJ(_+4X~3W8sflLy+02Bp){(7Ide=<Z7^b|5-o3
z1tU$O-6uf0sQeL-6;zUcEB*mOS~%XKO@8Ap`BM^Yy5%1AB+5{Hs%V|EXKSKA06FwX
zfo=Ws&9VJoaaZ96T_-JwyK3UZI#AVHG@uG#ehXiKpJSEbfnXHFVl`Y_Ifx34>^}j7
zw{XLt11;FI*4?KkKOcg$KM>FQdcLL}-~mh!aRn|x(lV|BmS*T)%NgL1vD?(FeN0rG
zhZY8amkX7jE*@KP?u;97!v03&#XFD3(XtA;p&{(5{)7v^rn;GZI{j|A6IIk9-r60L
z4v4j1HE3?x7mP5)7J)Fd$%{4~QUM92;QeqmqH8O1T!pnGor<Q_qd{w2z2wSJM%TAL
zW1Nwi6y=JRL{j<qD8xE^&z)TLmVe1(vk4M>#9InZ^rVKm{BXCu#Gl`DL#M68#jM1h
zXV?b1F1{(UQ)Ae{?Dcrc&S|~hSJ$V0@R=)qbLE-)k1pzX&B10ZWRCH{t2I$q-x(#v
z?*%77QD7G#-vL3;a&Q5PL(c7F>{3c^t9XO8+Z)NYdRyX>h0|a-VbT10EtP3<t8dsf
zGC{rnSGvtBy&ob;dP)-p4b*crL**&LSqLc*pTmFjN;17#%8It{nAqu^>dgK9O~T5}
zzIc}>x?YEAYlfRRpgv<!y;;GB+?P|^v+-yMNSYlluEdPW?;2m{UK$-HMmP!EJg$%H
zy$HH~OpMX<vi>!poCKQT#LI=u#0@&A_-V@sUwhAm&-uZ-N98aLv+e$kdHuZ}+eZ~w
z#`?6>N0bW&{JMcBwRa7KafEu6I;EaKRrNx@%1hO^1W*B*TFPR($_sIf>v-tY-$7fz
zHDPsuf}+*1`SNUQ>&H~J!19y2+eYK@uK2M8uo^s-s;p)iOjwpODTaywi^D>D^yp5H
z)?8Y_{x<);yG}89oYtzVMs!~Q^$DwB(^Vwv%;gn7r63JOd0=bR=5#F!iBse2d*a;=
z`rJtrp_POL$9~tfVb2h|>fZRoErvnv1M0$wh5YN}@pVx_)YtF@44*SfsGcL*nvaQ}
zQ7H&N|J56@{)iz=`C*Sf1ez|z3H#z+f!clR9!Ffv`HS3gcqIb8J3o?2=mlt3sZ9I0
z9nXb0PYsNkkK`u_c%3>uRW}q-ERsvb8Z~@4Qq;FtPZK|9h;_63tNVMl3}8Q-bfmd>
z^xkDAN$>+D^H4EKN~$nF(Y1<z)AGCFWN$Ho0@g_Jjj%>?A7$6<C?3XaJ<y_uwcg4x
zm}zNcHn>3=sYpX%hrwWi>Gi5;;Tua>{Bzvnv7AY(eSsmuXib(N>yck60yBgd3wYHg
z2?hlqsiH$4<U%??$Rl}EGI-2LYX7FdFwp#0bs6mx!Iz2Px4QbO@MO$r4PQlu6o2BA
zOY>qL@L1CJaeepv9M`MmB`L?cw}xn70ASyYEI^N?DGtma359i$SqAhEaVZPw?3LOj
zazW8i0srXp3<TD&9m-gILC}`%VntjZpk01Qq~j-F0}nHbI7_>lb%vi8zhRkzD_2CX
z)(e3>{8RgtSM2A2dXK<@v$DDOBhOML?on24Au$FH)tfiMi$Nc@y>nW;N*x6bSUnWM
zG+Tf2`Z#YrfCL7VCJ0xo58KgD2@L8~JL<>v+jPcP_x5AhI3MOd8DAT`z%^TnGRWwa
zYZ+bpF_essG2%iuviI_Ix#;YE7{14bQj<KQsk+;2Jz*^-Bc$zk?m(dEJ2ItqKiph~
zI_q()WDZ3+rntLl4thTnPOO3?g?X#aaO>BN<r*~9>-#$$ZBE<UwxIEFu0eqw0$=~*
zf{1Fo;VWvDu#@VtKFLYHT-S!{vYw_duisRa%vKQ&RnmmrZ{hdf$!}|V-3c0fpZ56C
zm4}wx57x7|rcX*bHj+S=eU_{#PyQ<#;g0{1^Q7X_Q0^lZvxT-u{aYFhqxm`ux5}*U
z4y#@FMq|r>F8~sD7Y(z>ya+B7UcGEZhClHV1MOQ52VI=Jdlfaa=r*a;uA2CqVNF+#
z7_@UUB-e5~f)_Oy#OpasDuQFe5gDhb+dlJQEDsO66jCsCw<c5xP#ZDd1N_)+1)SGW
zT86JF5|hcjqt@)!i3PrMNe}~>d$@4g{4$_%GaNU<WH)RQ*8h(6l}<lomN%t2O(fog
z%b>n)4=|xy^w2fvcbO#Jw;dCY6{jN!1YSGm7R&o!Ba+_Uu@ns(yhL_8BOqFFAx^b`
zog(&L%mx1<Q4=-`7rd*)yqz072jCs2`%)K9-!Ch7hE!iPY$I@%9}EksPSY-FG}5l-
zC2(kS7e2QGxbKsy=<~Se2%*`918c$4q_t4RPq7HW4UmEGF+<cv*D|X*U*8iBeFr*X
zg59k7Ul|DI2`-EqL`7{(hU^$XDqYYX@;B#$V;`eLp117{v)bFmwogX`ZD3sUPRuDI
zi1iTyfETGA-3RY%m&YGP4npUn?a5-aZ8+PkeWL1rN~~2YWHt7R^|_8$?|#<n65-I$
zvFFWJhi3w8(DDIoJWvJFKrZ08xE$NFPhWtJw=mS+6~hDcD-Ws&W)99J_pvuZBE<=@
ziYJu*<pf<6Z_&U(PVJgqF>Xfi8hz6P@e$)3`xnX*OwHo5H!F!1NH0BSNa0?W@H;mt
zeI?B>N5{Eyv46&i(S7Y*18AfPOSr!$KZ#tx8S$ZI^wh21)AriR%})Js+seVWp60Up
zt~2ENyf^HO%2mz)!(a5s?T3(5e(uoe7-O=HM6K`v<?z7d2ZEVp@!XadE)f(zk_RdR
z9%U?h_vV_FJ3qS+Wm@(=wx6~}O@5tq_~KGvbFSlrN!Q1T+57UXW83)8_ZJrD>U4xW
zxZ9}yEnc6r?4M8%4W7)0L??gq;R!tss;U=ot`oQ?v2q^X7Y03QYd!Z~h)qZyG*w>n
zyIt2IB#{}mU}{`=U}KGgA!I#~$>Pbk7B1!48r*puM#JCE`+%g65}jH4tMjv)28oSA
z?Oz3B?D}4m@Dg8SCp&<oxs49a5aQR8VBoETV@_8rdpwO0^Xw6~7-n4o?rshf<>na)
zYkPYHBWiSQ_UblQfN72&e^&x1ZIUVPSUj(Ud0zu2<@m7?5L6>0v)M+!UL&UD+xYe^
zjjYU1=s&f*E}($Qp0=pyI@-SCV(y14-H2ze3&?`ekde`saQ*;HQHq5xk^9oguWZ#W
zn@4zT7lh^C>nE7UcL*JKQJU_6+1inmEy#*{e+MvPxY<(~uHaZ|pAD}=dFOEF+Q22d
z;tA$s@)7@p;LK5SI$pDv_pZ|Hehy$22uZxGhWE<XB;wA!#1?6hCSSI|W(Y#_lcv%N
zRRcfvh7_}U(m9$X%5vhkJBFz>Uj->as65Y?{fF64DlB?$nkCOo0fs~keoLbb*q0TL
zD`FTGEeIo0TV<$Bhsk1jo$=zWQ-$r*xz%CpgB<DdV!rN~K3A7Dxb$8cf{|r!XtsG&
z#Z_GqHiWd83eh6V6xLb85}hI95}?MnZpDc;)CX-cz)r9RwNNUdsIddd2zp7JjrV~W
z;;>H@im)2!O{4a2O{}z@RG5Hz23{K=j#_BOaY^mG5RapUv9tt|PO!S11&RkSx%Wz4
zdHy7rpl_%l_TuFXi4eq&#j~p-j~!tZg0T~f+@^!J?Ai7HL>ywwk|VHUZcISRJGX~~
z_vwMl?59197SGwBto$!tw2oehT<e@ey@m_~=0@J{3Jg-)#05BUj;&9b`^5Jm0>syy
ztI$Vo(zB{*yVGrU5YKK_ltzl95y}BZpc74yaPJ0U<U;?-|2)`PZ9kSC;rv-*)g!H8
z1`tT!XTM6_Zi=Dprt7<NG`4OA0`@5rGeNVvXgr@lLyw^TOr_2dhIM$)**20xHTfWL
zTsQSY9jBJp{Y<uXxnZ-JQ+BKO<{1%h#r?NGB-7U&H)GiCth+l#_Y|xJB<;78ZB-rn
z_IYtqI~R!2z$aufZ3QFvd}H3}h{p?W_Ix%Ztn^u4zUlOxOy!Xl;izil(;zhhfZivT
zaAdY4MHaWM#`^tCH-S=RqUr1cxU#Gm=I4PL34dKK;wbIXuHg|KL823;t{L}}@_XSK
zhSXFjxaCYXA&1u&XhA~$ZR-?rQ2kVNSMW5CV&QHbpWrbN-c{2k9h_|*5iH(mK|M&z
zm8&yfi+V67i0d8n^T^D-u5BB0QaeZrqz;>k6m*7nMOw}%D_2+2Hk)AtZC=}OlD2^5
zo_8U-MtU$QtHm&0FL{?h|HGXrZ?x&pJDc!~&AxrHf*wha{_;jI!9mo^K&}jcsWnH%
zGU7YkG7M<c-<Ef#SpB0mLzZWOjre9|R<U^XB}upKUyAdsi`>!N-dZcFKUesqHL18R
zlus^_6X>&>nxLq(<i)NSC6T*%2AjUxh~6`uto-TVnn&g^nA_gzOLeMHc=K#n*zK(1
z$n>-v%zF9qb}-Y8SVQYZmS>PdjOBeBx0+$w2ii8(9<S=epZTBfH}jOf=y4yfaSP80
zlV#JcPfoV$DjBTSWf3Vh&*<>le1Bt)Olfq;T^W$Gx;5G>UmLZCcfM}zHf?^3w|gYz
z)4inqlwJGiCTE4k=r&%!&PJz(@gmT!bjbAJ$F1Kv8XC2osJc6nufvk3T^Nr?C8%5*
z#nR~&J#A{UGu?agqOjcB`B`8d5qB3sx7kZd!`&}4Ud8#EF|R$|xX@zPMccf8i$;Dh
zNU|fRhTeUCvbO{#SQFn_%Z_H$v*48}tXcom;rvVyuRrYciSCtVKw7%INIE+9T{YHs
zD*IYok6F52J{~S{x_A)FoX-5|t{T}(q90+Uf(*)#xNUn26t%V}Y7UF}Crh!f(>Y0%
zr)!)a&rCA|raGx9T*hLwkjujN?r=-7?UURlGL3lV{fpUbX0*F_&pw@(C=6eFmvjBG
zMlAIN?VSOx^*nVAl={Fq<%7((uZWoZUjw>UD5W&%9UY$aJXH-N2@Mn$4dM+^d<!LV
z4W6WQ&RZ`ji%Aa<@76Pj^rKgt*9Pyt3ByZA&&>o%S5c{4CDx|zyGz}7vGC=+Zg48n
zxS2#4z=!12216;U+7#6Z8!mPQ?=z%w_+{MR>u9xL0owh=WY(tK96xFeS#_Bcq|gGN
zX7CUUMki%5gqUV2*>O$1qL-6b<KEcUBrX))S`Z$T#)Eaos!OtUImAVo6+EeMqR}GW
zOlQt_mb~kZOTjBWoKRK8Rlt4kCLR?*W(kqNs(6unAPRa%cBs_4G}EVV=7pT$FNyB2
zm9BAR0FZ^~j$Pq+;<?lLPWwh>2OjRyRg~DK9c^i)Eh7yk?)pbS=`|S6jLyMA(FpjF
z9>|y85||#(N5;fCa?pNfRFBbRnsoAG5LZT}FIs`6T`H0KdR1Q5kGps;!{5FlW18=G
zCPQS)2c@GYI!}j*;+Pkp&DMw~ZBHskr|t006<0iFaGR#dVS08~?bfPzTK=1LchKZp
zdVuxguFh<jTPDa?DgjZ6X0t;A@Ium=q0W<qEwSlnN}EIG<DUnl$_ROfaXK(MPz@)u
z)oFinA+kBmL!Shz#BHnhFSj{%DW`5G_`WAHT$h9Zr<gdRE366ngMk1WpEx_QHY#x(
zXQ5r_DaYz(Iv#ey^up#huJ?y;1#%~>o_+AiU~Xf$99tz|y&jJ2Ql{sRynOp(<bq?L
zunEKd7(g6LrQO`bOoQZi;DwWB`vC*l@>lJ_kBDtWZC2G0A7*5wh%b~rH<th+ob-d>
zFWk5|<*^oFJQ~C`HMZ&b;_Kz|@5Abup}lioP__tJyVsNn7Eu>z_s#pCvCaA0G`tcN
zE|l~JOslxZPoKV+-nv$~?#(l)>BC5n@&2ysfVu)Sj#cLh-7@Y5(;yiW-SA4k+0OD$
z&D6E}87(C10q9pSs)66vT`UbJ$hX1OhhN;CMP_|br({MhocB$?wA;rN)JQ$o&6i}3
z>5OIBI5Hg8S*^|Mbl2$4*$_??rQVWO_RL%GbqJ_Y^I9b%?Mrk1Sio<A5^(vUG&9`?
z&=}?vz9bm*?l(2|3NPY4TJGVin`9>pEJXdv_inQIXIf;vMj{`2d9<soyzo?SG*fRz
zQMik85e}fFosutp#*0s)sV$A65nO+|O%PjTC)I`C#y*VkVHfDM*Id~5Wwr|gMNt>@
zbA<Y5(zg6dbQi2fk}Q-=Bt8amO*|}5SeIkH@Q$u$I91Q_7%-53pD&5V-_)-==|*LY
zozoB@R~vK`>OZTh8+3?<KgIj;r9I1j^=$vu5f%9e%QqjHNnS@O1mxrQRJ?^dp(=^>
zpWuX*5j5bT)u<XqK$b|14=CO#_)srt=y;GbIeqhT+26QG#*;!UVo`a?6Bng<i97oA
z7ZkV<P&xYu)(uUpmzjx5Y@ZjKl6VJFC*X<QkK*l*k%j?GFPNzV;L^V`XS#08TSWNW
zPs>%?q^^Q#R+UJ8yv|ZqzLNQ3YRmLhV6#lUK91)NUH{J~6(V}~*IbPj`w71Hyg8aV
z7T%>|XzVVkmj^^M28XYh7^&XirR}vp(aF0Mw3*`66|{ZxFiRyjVY>8Ttv=VdcGyi&
zO3Hgmn%30uZ9dhhHhS|z16BL%TMjeE?N2=IDPvFYMa~V~Oj;i89vS#p(x(gIwtdIj
z5bBB}$nIpJRekd<Z{)dzY0yj3Oy#AQ4>Po!&Anj)=7ae$!?^~NQ}eac_0}Qcx6kcQ
zc88KsUYMo&1naquyVZ}n6HJid%W{VJ)z?7-Z&|EmC3f?!=I6JyLXT)37%*(eSZ?h&
zU9`yyGK=%V)>u^cJPJ#9Kirk+nolEpDg7YuI;pCiaWO8jw1r4~j*R+esSdM8A5{lp
z$rCYKuCbxtr=M+JWJ7<-3&U%oK{uznbsN>r#s{%!Y$&V45pz@E+4BL$l#nJ}Ty%Px
zSwAdST*@Xk70V~SWQWz8XMD5SoDyAPTXh*MpZSXO^XJyW9;hB0&eO(DC&HTLYg0ip
z$RdebBc3(`NPhYn`Ec#zpi3<CUrAJO1@p3q8DxTAk*1>wS3v!us1E7vEWg)Fyl~dn
z)_Soo$$Ux|$d4Zo(1r5a6A0hj`Dp{1AZT)|=o*;J1q71EhFrI|UwTK<R(_W0dY<Si
z4aJb`8uTUmL>J9)nJ6xuBz(G>apMGio=mub{}ZPg7CH~;Fp$9?NKp4Gh^tmL68js4
z`VT+6Q{&05BIS;_v2@+^ux6uhj}#vf&1*UwNdH6d`O|<Zxmy??g6(H&t_8UEark{E
zT%Q$HeQV6{Rah5jcxafRm6edmG_{Z%5@KsR2<badKldB{w}zFxdqL?A=t{~2E>m>=
zvSNk#{Lk&*>IN>>_byBB%8gwn6>O&^r}aGOX2g;oZTrxWkzchM##gl6su#dD@9)jL
zuB8fv52D6ib-jL5=ra)ff!1GX#ua~FcDP}6GObWnEJ!?5avbZ5hXL@Q3PDSurpWlb
z^c!uXiPXdZcs$bTL}>&!?8zt8_f2`}@$RpBG|GJS_fAQr7!2fk!zF3rcaPw}nIR;a
zet6L9iQWP-p6)Nu-9y7zctF<gBkO?`a!NwHnD|_fn|jUf6w8ZZxaS_hu<mTS6zkbl
zogabQ0|U2{$xml$=LoAEdX&>!5dAdOG^q&wU`hU&>bNXNzbX_{%-(MPuVvPUNFjtZ
z-Yy=q0I^8<0PCaYv30##)3oJW3Mnlf36mde!w9x!XL(MEAr&>$l7`#Z^Z_GDm0dYl
zw1M1R{)v}i1S|ct^3~luE29{D9+jo5Opuqe%?(b~#Xk+le#ON&4pVxq@Yk_(0ny*K
z{X$c5hNSlPcsfdoIRzwMm~u!UMrsXdnGKT4<-CI2@h)hG@E0y9eO%3Cr~L@n>ku;U
zoSEnScnzlr^x{m#FQQ)7W3|1f8*GE!I^A6@JNafTxF>kp{(R^`2)qu5c2?#q3<4b#
zC^{kt4W}0jU-7Sm+gu^`&&uF+==NAc+IKhsFNpE<k&4>}K_Ok{U6H$hymCK|Fe2zf
za*LzjIc6-2)jpxx6eOzHc(|=dvr_Q81YqDLd1qIAGgD{6n;M-^MQR@cap2ej+~H?s
z)^*=oA}0|tzKAA7Om76!AM^}*hF1VTA<qN)EJbF{dAMI|6FawMK$T7V;rmkO?vWhb
zu6Pa(3osp=50|dMpq74@D%R@FQCOwmJQ=<Mx2D)?T5ZD<=hVH)tCduVQqN9@@qGD^
zDsumZ-8XTo>JkJ2ia>xnw6l*BFxl`_%%$^XRztVoohDQnEw%&1(~l9S9NF?3kJ;HQ
zF|WLpHtPFo2?!V2nT$9;&EtN6i!k}BwvChE&$}h&APn^JV{dz-?i%`H)z(4LSWlSl
zA;`5Mn}!VJWS9?=78O(4eX;eQ@$p7t!BlOUcY{)kzf##Qn!{3?d~tVgR>22}0<t^I
z5>;CTUiH~=KE;n!bS0bj3)2gw-4PGOn3H}>|8o{1b)MA>Ps2d%#3Y!HSfRKM=kLRP
zOJiso7~|H8b8wHLNtPLA#EW5W24!wly>v`%HcTbdXGrcSv%6<&TaTf#h=t3Zxbc8_
zE`%^FX(h{psxksS5~B-+GnC=7<BEB+xC={v^R&MxqNb0<2(7SVWz7K&NdDZm@37k-
zytuTI;zr*Fx9ZbOhx>|n+aDHKLTrZBLO&zp2M`4}jR*^L*mxUT3P?PR3=xA?9%{6w
zCDS!zB-8bN#0&B5hcQMS0a6waQGqXkD8eEGD&owQkAtxe&K){OgmzBNDiL9)l&S;w
zYtE*NH-isbrtz8FndbxeLp)x!zbfiKl&(wBPVsM8)rK<=qKHm~yJi>!ctHcwAidqq
z&wQ?3X4s$+moq+pr@;)t_ry$(cO9N<abC6byIENYrI@`L22BYO?0K)Xd|J%;%6oN%
zGQ`Y{g({}LgQ_iyQR0=)E3<9SMXXGTb=;Pl&ZH?}-nFzI@0=ULj1##gC>Gfk;rntA
zn73zqy`j(<rHH-K+wB2X(_Ttbw3VqhnQ1x_I&?QHM|kevGg!24aB4jOSfJ*lkqfm6
zVzYqiV%^F1i^XQE;5}q1JwML6QXG#xha<^}2vM9VjoKJodqrF4#oMrQaxi*Ph~xNG
zL5!l(mgHH)^`E5c9VW>_-`%&cz^Oka-FKmgALspb>W6ECy%If5&v|tph171WUS6D8
z(o@d=kW7gZCH79ozwJ3tNH4YU1#11tprIp>T36SN>L*7V>Hw|mRX+#5i<AJ>(ZvM2
zi}2nj#;+AJ6TFT*l`>}+Zb9G08K)QSz)wFhM~nrtW!Ma)PO9(M?y?gsyHj~nH24Gd
z9Y|}?tsebcsNTXWjjkbdCZ0Ef9@n;~*++iv3vkhk6DlgxIBM=wtsk2BQRZ>7crf__
z*~IjQAH?@*wwc;<P0*AF*d<YJ9fgxeF9`x6y`z}dO|d_8^s@KpV(fUS>{^KNpUI8!
z6QWNdnp9ipr~3l$NT}`_ce26=GZ;b3o*RWTWBd(`_Xj^yHkh5?K{XgS)U!oI!|0dH
z;;WPg5MImhFU7YK@Hv`92x16t{p$6zK0sUUeV@nm!ib+Fo~T4s@E0L0Q}!APdVbWk
zon#NDAkFjH3P6Wx!fGMD$ePc<lX~bzQv{j;q0Ei;<Iic%-@Hd-g?{aIg6I6M@acHL
z8$iwI?uaD8XKsrs%H!>fHmxtdh01b5yf?2y)L{!=>4}>7$;#Y@uBYDxaPHVJ{Z``C
zqig#mt3Zfp3#aIo-xNKs`~LGU;*9xR@ZH#`KRdhc8-eJI8?ZRu5D!1x{(-UVZXw`(
z%P`?uK7Jn{?}IE9esTW<fy1+U+Oqj*l?v;Xv_6_u=%p2yX)ElOf_9JY?X`B(47Y9V
zLt@aJI~xk6|KmZ>`Nd(HWY|FrJ{wGW6k9#$##T=id<6JCTwG6R(ia|oaerLBclquc
zQPUN-Tk^v|oa0s!d<B+SK%wRU&K6;uwn0qvhUUe9q%>IP<|qQD2rYqHG_pRd0X&x7
z$ZOrt>p$E-k%1|!fz%_MQzGm-JLgGg)*8U#nSnOFR{v1KG};Xni$od)#t$quBsIEQ
zu8tR?OI>%2%ts5~>(_X+gc8Ou$>SK3+%)-9VI6yILIOr94n{yAWrlA0eK_E^Ve|L_
zl&}kk{~53Kr?!<Q!|po*ShU7Aa*1dSe11^<nj;%<)<D7&9p&ajnLy|`SZb<rF$?h2
zj*Fx3JTMfYn6f!VhATJl<p1Z9H8BD#yGcbLIPv##5D|4K%+a;rL`S(QFJF~PCrfxZ
z9Urkjl()@eo12~eFscls_C#$Q+UgI5Dc+YZTRy|Jp$Iug{XDAgp#&N@)n7keZ|1@>
z@dZx@J&on01H6Y$P_(B9K79O1GbR|1s;i`(h697z=@rSzRi%&;kK0Ww5*UhyrjO(C
z_8h;@<@r76)J0g=8;));fh1Ei7jZJDjE3#<$4+#3du>7v)2E+>GvAn7<b7iV6bp9i
zT3%+cmaQJpA!Pwl(O~*r94>K%t0d(6^ue5+CZ(;pIN$4wiQeGj#>beO?f)RRXEkbl
z93xe-^ynU)5T>1;pwM6mZ@WLZo|8`_#QwOR@Bexd(nY#u`js}90>N%OU&2*-ChizT
zk;kKgjgo!4{<(gom5#}STc`Q=fY>5SHVAvB`CJVUztz!T3VT>izU_t&C=OT#JA^~i
z*|qX>8tx0VS26h^WDS8yPN``(+4oghalfz!0w?nxgaohupl$vpZ_?9WvpfLp5XD=S
zpsD!=04W<sO4!kiuqru}6NPIdl0vpK(7n1ZM>D!bUx^(NE53MAo^?y~U>nh*Kobu9
z5OOzO0_xih_G${)pL^?_G+^lL;k(V%A>V7#X2hL8$)f-IyZ_UlO<seam6X|M7W@14
zQDXjxr!MUdtN4)+tT>>8gQmRs`2I^3B^=2CoZr6==^65OANQh=^gR3d$A8lE3^9dE
z6)(=R8v{QVW0Z%Iy!n!O58mE*{QLL*r`PtI@A~Jb78yaX+9MRCz56E-{|~k4KmLCb
zGdL6<ANlsN{N1bn&u`Dq^A`94l8EJuKQ`_E_5bOHknz_Ed;8?~Q~CAMb!28i7cr(l
z_p@^AOVTn1aa^>2x_X&NtAaHK6`wy#Z$&I=y*rM}s!0+amHE1#?}O$vGthv>us{CP
z0lJhtz_Km&%i#nE{0DH7-~!y!5>DMp)fYd;K5X#-d9`MqmKtbgFgb6GsJTk1zm^uo
zoI>gqjoWam*FL^m1de&C?To7||Mc~l+yF~f#&DAY^Y@Nnr9iv9JhQC^-M%HT#bYZh
zN0pbl-W+^C{5c5#W9ujK!fv}JTnrK986u9$&*3BfH*idV!P>(N1w5|X;IvDbuJpcl
z7`RK`32cKXwhkj`3`$MH&cSB}K_5RNIDkN%zz^`zIuJoPO90~oAu2d?g?|6d?>P9k
zYgXzD9;wbNOk9-TZ<2%r<zKyCRNAAcD}tJk=M!8YS$BPEUA9KZFjV5fFe4_+8-P)u
znIvSJ0w&=+cq?el4rKD=?e^ug5ZaMEZFPK5A%8y;@Pd%78JZI!lxn6JALMCq>691+
zO}TFr0BimJsc<@D`G2~>uhan*@Y#(&kE35EmY(NM>n>74WEzY#d4b^7P^0;V#j{-k
zCxvnt1f>qNoAp4<qHy-oKdMcA6Lygz;nC%Db!Kof7edRw<0>i8gGOf(bhnd`QK19i
zf2qsa10H|=g1<XA9Z;~|0_`d4%zu2PABy5ExEK>2W+`_n(t44w{s64jH?j9cm7et`
zvw<Ctzo^J;yFQQ+Vwh_RzB37X&EpUV-Ixr#4&ETI%0Y0w?n{J9{?jNS9gGExP3F*M
z7ydjE(`Z^?r+@d{i#*>SQZF^>BpR=>&xqrG`lOh115B&@c*8R+AMJj!u!dH_ldJ%w
zCxNx;_Jn0oB-SBYb}|Dv3Xvk<=rnYL|B9j+aS5c(i>F>pTA!Y5hJ!8^dv>aK4GG)B
z7B687+L77W(ZMVgQ-Ib#{2+THSEuab>^Kr3k2^r*P(76*<Al)xkH`0woESKvpjGq@
zE&_~iDx`|)Pd}C0e%EK$F3jJOooc$a36}7i$~gW_e)+4zOY&&HqnhFdc#=KnR`7!Z
zqEqi#4Gg*DEQWLFk%J<@zLDXf|9OOqDMaW~1&D`A;L=<|n(F}TiN{)ndTjTQeW7l^
zJL3SyZEufZnD1zhuouuBVeP0XG+@q+L(>iN4v={+0j1R|+lC7*q%lT9{U&%jQ{@#g
zEUKBs5r>&du_ouoJA(zfLXL^L|9Dr30Dy2=?=`>d@36Pcb3JsU#$}5e>83=sD5HlS
z?g!Z`fR)I8(k|joc={qqU-0;QZTb;%$R(l6#cnsxE+7xzD|^<H&hxP0^k9Pp)-|^A
ztS^O2`>FGrA2VM7GC0Et*|-CA_6muWz^MmQ-UA%HBX@d8myDe-9D&9~5x7b?dI37p
zqBKZVd^|3&${3;WnN9)X&by$k$*x;zMJP@%U)6AVs&cjiIs)BDlGB1HNN#~%uArWD
zz77l(y?gj$oH3k?Yp^e{s<*Gzc-F<b-*`4OD}~-Jpab9h<#E7IkSC(4+&lU?sV(ZV
z31zIT1#HsAg3mL9qeg7{uqb)=+t2%@!}mvkh`&3YE0!@zShle-lqDeen+NS!!^!fy
z)01a9hT^CQ#TFcY%}~)MPvA*5YSi|cNu>+qPK)_vy)SorwNmUnfJ$B44R9zbgh>$!
zqW^e39wXPoqKjYu_j;sZ0=C(s&H6JNF75zW5jd?AMs#Rq#q!5ME!zbm=?(uH@X*ir
z&%i;r;m=>?hl*h#MW?q{11LG>;Vv@^KoQa3eI{#%NjNv#6tD!;p%K+&BU?y9K*6lr
zg)8W}-$)mN-2gUy*_9B?ys%K(76*t2h@u`2y&@~C!lYbJhBtyFeSp7-B}b69zFXEX
za5K=Q0+wBHv;d&Ll+`uy1a@+Zb$h)--qKE3Dq)aM(}0KdmK?FLGS9bNE7-voc%7jJ
zoHqGltJ{HZ!4*?Pc40kPsRnj_b|1L;R1~71Y(M}S;f|ZX2TxiB{$%DQbew|732%d$
zO2qGZ=l~lX!l0H(>^klQNtr)^<S~4qFAszf%=cVJ6<fF+{HGTH`g8hQyUd&qjm8Z_
z>A}|YG)Jm(EwZqpBgNt8fb_Qkj#@0RPLz!Tjr{*NX9c2m4Iu=}+~bzO!EpzUh2|!1
z`kd<T5&{x&w6NiKxI>!<%V9Kby~!f(N>e81zIHpnb#X)spMJ`6tQa0w1=(pN`ik`W
zLXs)@0W;a!y6>k@{%Vsyl%nx9P5w0erru}NY-Imtv*F6y9tUH3bboa~F608RE$nD}
zKVN`(ql$%9^VnRNb(=LC$R?nMikL01vuVUZ5pE$>;nW6K7`rEgfZ-B0Lct&sx@-8Y
zPZ*XFc#Sw+0+ZPq*su_YuF_rLX5Nf{8(a$<e9D<xf%TpSV6`&m!b;M~ER9a)!9&#3
z4Z004iD8K_*yQ4*VNlIoR6n|y`lrd7pT8`?;rw4e#Qifpfze&Fv^f@#m4zh+e5-D!
z@le(E>E0krg5V<@Xyi=|yic;A?yDc{1FD?X9*WEb5|<Z4x{qt7fe;!^wiXH3-Uqk&
zQEtw~4IDB|v^WTztlo+<Fc3pd&fEoeV>gnJVq7v1NUDWZvM5jp_Ex^M>&goSVa&o}
z!Bw;O_Wfxx6Q>OM6cgazguTyQ*Fc1AIYW|^i-B36kOi9+C-KGbBlR_=Pd%-IyMa61
zU+&De0{=7>aD_}~15V$+W{UrZKa1Q2Z$4T1zC!TtM8^or6DUkjRGcAJO)sJGL0nq-
ziIw-Xv+q+|!|_zU8|9AG=`jZe0Q1CzwV~{dEXNCZn@wKh+&Pevt!ITylih&JYyq0X
zfXYct!AA8~YrW3>YEZGbujy}{sA6N3Wb(($w(T~<)<t3h=*ndR&>co}%v$v<0|Q_a
z)!>@1p8-X+#FPB6I-|*(bF)CD&j1c%q^lG%7*FU`SiJcGis_gI;8q+1%+bV^kpxqD
zn%N1z3K<x6%iuTI$h$wmgul(&9FPt))o)89$bV;TtSDeZ)saQPZ|ZXrgqy<YL;dNj
zK(sCOzHoN~L5AsFPyle0B82%izpil;!K_Cna|v&`p4K=m?t(>9$$gk*vOb&}V=VK6
zpA8zARtVCAC#y*j9c7^MeR94USw@I&bkjH#axn#BGMfJvD?kqIAokt~uG$CgStH##
za(XKNBN_g?8l_qpET5j5PFwh|<xA@)_RW&vti}027w-{NQ3I+(OTKZCsg*Z^O4u1i
z<EbDHMoBqud{OIeZKr7H^3({HnA0j9WCm$p+37y#@sv3zpqQZw+=Lw<QDy~aINSDk
z3WG|>h7<U(CK@Uke(t_8mli2L_sr<f<;NI3WIR2;KjzO`>o29(U;ir2jf7+p9wpcV
zI13a@tix!*1hl}mP%TAVhc0ukEDikw3Fn?^nca5fq)jYv3wr=;NST0B7Eatp$YdmN
zEuwoH@U1?LI~F0+Zl4K!A>G5wVT4m4pf4yafb9^6ycLo0`r_;;x`vw^xa9B3laO5g
z=gREg^HM)q1@J^=<7Xqdf9(kAIr{!011YziR`MI&N2g)vpKZT$gWIPFms<38UQmG4
z#eQruaI$sJHEPK=YJF)>Y!7aX(q}!C=HL<=zJv4$D7PHF4>SjC>f$@{ad%{of<VC%
zORn!Kv;7qWm#<#RTL380odIU3I@JzYKY$4x-^!5JEYQll0`1<XblB9wc4=D+hDi{V
zc6F%>&2TOP@$JT(4F{QDTVLkniWX@rc9IQ_*+YOpG^n^n{L`|<!5<s|k(h{xKWoPp
zBH!(wAdUT&-Uw`Yfj;8#{$1eH_Xe5T0WTVn6OdFPEnXZzl;}n}N`VV3oUjEhvVhEg
zzE1)&ZLVH*DcNXsD&~fZ@aD>k%_%mpeTK6?s=Z@gN%g7s1Y`;aK)h6Z8fEGlPR>K)
zj4OpK?zj$Xk+mYowiD{T&bk0O3qpDy^JK<zPb~@1-^i3?r+~2o8b0Oid%%2!v>H%V
zD@Y|dmn@<3alpSf9z=?V&%P#pC^%3{<ZVihvQmtsWI|f$kdgYc#rp}`oa`>~n|6`N
zKx>^hn6rV;qKK(=xT#Z<?Lhk7U!@_iB+nhfCGfqA0F|2p5@|e{4Io$qe!_$KAgv{I
zla$XaR&WKczgwVBoDt9yqcn3K&^yqGRiVdY?;yp0-^B@?<*=n*;dgkrO2(&10{t{&
zJaH41iluRwHB9TaiW&jP&FK>Ls_fvm3PI!+8rxg$i3KSI)-Fm5QcZL~I_18#VEa##
zGS*Ol4RD6$w(<Un#1Ns;ecYJ&QY+m*k|u*{K34nzndn{xn@Wbxa91x9S-;0$DGub}
z+?9FtYzi1^?U(pmo+#4#uj0&0V%ARKx6^Iz>MFYeC~|i|k#?<{UiqDXSyZ{z_(PuR
zjl+eUjHuU0J6<o=(&#gN_cbQ29|8ADyf02g>xxe??5SEJAwx}+bH3M6%dc@d>|Jfm
zo+I%_kTu6c?HaE?tbJiU?sIiv3Y1;`u84>9fo>pn62i##+X!cnphBvmYhyV`7M{#j
zx*u3H(ZU#ULAg-{bz6*?p@W^^`7y@Z1WT7F#WIW1K&<D-K?OY^&fReC(9^_aCI43L
zcD38yGwU#j*3G7CWE7y7&1epGgh(KEY0$DjK4_$xf=L_(AZC3qNgL<F6Lui?))>oN
zm=W4iI$`$SIezwuFeV1QRS<0dB}ZYuks|bS49jdZxOyXFfrqg46(=BH-?uEv8Px0M
z`lomXi4(}2$@WQ1j{ht#uV07RdVnKT@c>6<EFM&q;Y)@Zc->btPW>|cR%5)@1^6>g
zKgKlZrugy6=IJ;E$WVLix5!53if{9ClDSjoQvx<d$V@X|1A)jQ#M+*Fx#f!1H|HlP
zHWSn(PvpZV=mG=mAo19iBl)>Op2bF(J4V$GOP*hJfFWpQX(+D@3`+{0OE5w~PhvG)
zQ=z|1;V&u`VhIkhB|vUm%9q4p*&L5PMv%c7HfOPY=w703I0*;NAw5<RGNG7+1esj%
zBL%>K7g>Mms^$auXLx~7D^v`AeNY(2ShxJv5=~Y_7*hrAOvrF=q0S8fB`6KL+mR$-
zreG6vjU=8MNI2|4QRse^XdE@L(-@8PNd?UB%xIwK7<U+hhpc&>1Li74A7FcH`#n~{
z<OOfcgiFgw*YftL{hx}P|3hx`$ET{LkyqrsyKR2IV<Hj6mq`M%i4G+Yf~_@d7}m?l
zh0Ed5*7QWb^EFeC`j?>Rc7TVg(OxG-H#-TKxmovO7-k;1CmF*Iy1Qk3$<?>bHiMqy
z+xr<J6N6LUO}}lt9e-Mo46H!RILC3DsO6&}(=y@|Jox%j@DC8g-OwF?r+wTKLXkaw
z1FS_9DH>7i{T~qG(xu<-K6h{S+lV~4mn>J~)|#?y3t*xc-HToMMU$6em@6P=GPr;k
z%wDQldIlXF!sKZh%e=Fjul*G~89Hg$B}F`03{N9ugcXVU|Dz8&Z6wr@$JXuoPjBtN
zT<*RE0Icy!+lKtp-T#*?V38eszvCU#n>_zC#{K2;-UfvGGy)PDhJU}l|H;4j=`jH~
zw(uk8!hZ`0OM;NQ4*>PQ-FyG}3tr_R;oNU=rgi^l!Tx<+8$-aQE&0GT^q;Nxzx{xJ
z%X<H^LVx?$V<3C$$G|4|FL#i1ArVLewI9Xr{Fh4zS{fwyryw``FLzMKUy+Qz%d`Kw
zkh5|CK&53B|Hp{yUw*(}^7emQp}*Z2o<Op-iW(I1w_o_5zu^CaAmEn|!S7AvlLH0N
z60!JyeHf>h9;C}-i2#1r-W`7(phyRe(EO*hf(DGeA^+|6`iP}jsHX!w(q9)fbzsjK
zezuu50qMpGrM=`oPK^KZ-1ybfKz7d$M+AD}DQID0fRxGX1Y{Arcji4Q{>wmU@>Uu&
zPcwn3{Rw3Rc$&+W5huL=<&I`;L8gH%VSPFO6!QMxqdJKsGDE4qZ5Hu6L;1hllGuD8
zsN>kr3#|WTr~kvh21SFA7R1wY)A+x70DiKn$g(Z@I70NlS`6tqWaE2I#j)+r#`j;=
z|8M@91_6a_F@!}j<NiNIUjDZS^WYeqXV~#zGT-kG$UnKr{^e`_|G)gwTG$wXQK3E#
zplQS~D?pu)rPbUWZx)xW87@$P9*B{OXn{}x?40!$U20qet;Xo07FxViLGv>Um}cAe
zWjmnQW1vcDmzpTCfz0(g8_jQ&K^pC~x#IM|I6w)q01;wS?xel}6bUE5chm%Y_by19
zaVF+1zI{>x0DUH?#ZG8a&b?=&sKrcxyEB_arNdHJDIhdtIT7&ma-$$SF`fVmDmCk)
z5!tL-0NoAdaneS}2b~6Quh7~HTfB*@qa!W}MMn{}A-Rb9e{JM1V)y|Wau+D$Z+&q;
zAd0vi;vBR_owBt7cT{PrlS=sh^SD06!<g6Y4S(hdm|U_5h%O`%sI~2Md_v~SEInkG
z)zR9YG&_|i0o$eL1X*)eT8$fdFITq;^S4zFiQK3BNZkrMTI`I?0uRmot5_fA4$A*J
zQypd~I&9E7AaZkJVJ8fqZw>E0-dT7=Mj#0WrtAXi6{g)MF3F(#bKtpw>9RScL$@2c
zf~@;<ALnV+4Y5%-Ae$1v@@o6b3IL4bk@I+tpuf2aBv6#j@{G+f>0~f=z`ThL(()X#
z@!p|}?O}z*<0Dh^$t&<JCcw}A$?ed?Uz$?hXx^Vd$}7qV1|od1NgW~!#q~G`a;8q;
zkNOC^6$}UCMAmG|N89s1=(l^twmQIEFjIgBx${;6WM~(Z=Ure9Ym9^+E(H-T_h-vd
z?%C-b+q_5FQR1NV-r?!kPSU$SgqyMSXJhYIT$6<y10=%f%CpJ;hJVQpDCPwMZ5e`o
zA{XLAsfu5n$p2ahQvR<YlAxu?Z#yHNB<WR?Lfbm^X|no8fOq>Aur3r?&&im|9;}h_
zIJCMc?rJyK$NDfcG^H;{hJwPViQ))3NCTie`82xsCo+V|gRK=oBbp=8r&X~TX~}yV
zi1d16UH>KO@XOJ01O)*5`#5khbOP5N6EHTR-lU%pD)1}<xONU5)qR)2B<$Bomh!3p
zZhmnWTi!F0^3H28B8~og{W%hyK;knfCG(&KXN))mmy|LnUX7zLM79BTgClxUyuRqX
zT5N=P`v**#W=Je8CJKO&ix9ssSA1&*&g25U>W4@vz8@=XJO<rJfb<He(P98roAr&7
z3xTCUhd2R!t_>bJ+qxYEYtv$Y>Wi)xi-!tgw>y6O&==q*ZkKr5{Pk@26RD8v#mK&5
zb1)u~EY-8_$LyKo>!*nOl2jo<`<%};%)ty={}R;V;Wua_NSs2hKZarK*8q=k6@a7N
zSmDF`u_F572P?AUw@KCf|FHMgQBk(x-sr0!AxNr#4$`Fv11Jm)QYs}FfJn!P5;Jro
zf`q^T5`vT>f`qhm#|)h+T@FYy^w50w_`Yu4@$G%~KL4C`)?UlyavAtA&wXF<i|e{v
zI>HQw(ARkB;;+Y#cF$bv<kXQ#_@fMFg`B8zDYhATkhj`j<K;oXZ%%qJ{$Q5Lb4jc7
zy}c1uzEeYGe(JikHK&UqvU|owvL^3XkaVVxQLReYw-`N&^*P#Lsjg_ZDIMbmHA57j
zSZ{+ZBMR}?pT^d`G?daQwKWh2ZkrFvVHbwAFL}?t)-N;<K0=(uradl#S}x75p6~>S
z&ejeV2zuDF2kWxfWr9=&0sD)vfpV$sG?u*2Pj&^-qaB?4Y}rAK?he5M%E0wcup;z2
zzM(WNx2ITh&X|<@fonHU#T<*+Z~#Pj@lk`&kcB6YS{y6bAvX{e<9#H5X8=S@!YQDi
zzxH(5Tju*q81@Q_BV%<}IsyS+{9ZP@<4!fIL0*vdh>%=-pWph+rVTWf#6;ZE?wb(+
zG;CHRUWO0{FdEly?Yz10z^aoW{V@>$Cyd0w61Zs)6uzD(H<Kzr{Pp}0^#79I`%jQU
z-je2lI5@_8r3@Ie2aM2r!IA7@43o_KUjEk_``7?;bm!3;V;`9nfzTc^s+ZFhtzhxq
zkrx2QS83PR6gSH5G0mPk29ETFO|Z>JCwkzjen~XI-MANY<qv#a%U{d-U03p4@@u`#
z&+K=?n9S&KG*R4v*pu~wVzvK~&|Vv2r;OC71^&K$>sqJs4@&w!62F{1u{(Z+xnMNx
z2&Q;f*NQcWVvj0iKvFv*e|kW=e7lCq_LR7EL%Ihz0$V?iK!TsW1?7(k#=}a|Ue+Iy
z1y33)oED4gX+n7#=u$HK30i>1EC(`6O?|a_)!T0f{Lh}BeW=2Z**Jdx=VW5&TLlkE
zDN}dK39>vuW9Deuj8)u+CQ}x0wiO^5_W}I=*>IVJ#gr5&I@}8==JxmDGFcce6W$+L
z*|~iOe*S^@-)+}3r<1o?BfhSHoe<9V(!7qp_@})mk$83b7(n@nKnBq|z=Ayn3xATM
zjpi$EUfr0yD8mIQ-n;qL%6S_s)ry1jO}+g(+avNb?)!Go$&I*x1!ZWkR}3E=<NL9W
zkHQGGcM{AA`CR2E;=Vce@u|(>9iuP_@Lb{U=?~p?Jn~4)Q&Cr*26X{}jg7QK_Mz7E
zu8(C2-lt2RSY_r=NuJE8Np<={dK$R@Rs|LtpGrP!Z8AFZOkyE>HqKFIF{j9yl!xPp
zPspCr_IBD5t7bO!b9NIg`s<M!D+gg-`MwH=pMmO8#fu5|_5hnmcV6Y7r6lQ*X1_JI
zbM^P<@E19!%3s*Ah4*zIkMMQAL{XJP4)=RpcFe~krNY~WQ%)UvF&jJZNo#lVwl+x*
z5R&H+?vYt;X#GwM(dTOg?;jv~ooe^UhO}7B_Y8jUu06cBZ;XBnJ$mleWQ|zxbC`g#
z=R7%8`jItPx1fPQvUO}R>PXB)--S#c-%ILrzV~Sl8z+tfbhS7L1W&T+r?!6AZ4u{B
z=JQ8?)$meJm2_*%aa4m-Gb?;f)7PHTu`Epj(O2?`p0be-z|cSN@J6lkt26M?Pt}+d
zU4<yFCR~4deuMusX@7M7W-K9o3E!Lq1Brq&alPJ`7mbA8RRF%zGf(~X&LUuiP;SW&
z)9(egduVu_g3z9ACYHib5y*X^M;&!ZdzWF^W)uYjfIw&7%%=I_IFn@kBI5P|fWK#W
zpRz<2qg9x}=8djzA3vNGu#gs^%LBYh&rr38%a_AeGSbd7af9<;vtMsh`(+I+*^8?e
z#Y;9S|53%#q&zFMJv*O}fNuj~S#rk}dVkhSGtkGax-nxN75&6j80jVNZA6;)NHzT4
zy4>YoQ58&uQ8j~(+phxcge0S^Jo6`h5{b6LxA<2rb|S}5hDnQTz>>e|H)^<#F?^=O
zLi8H4^3y=k>XoN+o<2%c5sND7bUu4@W9cJ5%K{Wkplb5vP5Fb_6IN$keBsBT9NNRo
z`+3yBoNn0hbipUEA>n<jWP6lZii}sWY`D~VkX?pW{9IRmxT8@-N<+N@l{jYr5m3Kr
znC$&zNq03_lNnc2Lj+!ucleXjn+}r{f6kYPqpBYU8Zgu30YGRl3s64E<a)(?5b#xz
z(*4}|z!MxUMwsejP_$TFC;3c#JNkM^12;sC>*(n~t`mW<fCCPlY7a<!_N*SZ>2V8(
zKD@0wI|X#m-CC;cO~h>HrdVx{a6ZS2A7}cLfEoj{6QfQ_`0Rs48gKmZ*+)Q5c-9g1
z2JI)&?a%2ekVR%utW%}5Nor?JwKi>?+Vx_tvcD>=-Aa_%B>gWX%ZmM9CUXF~XU@bM
z0el`mALk`<ykNsqXeDKLbg&z_<JvgS18HI`dH7@qf9bQQ1ErrLsdef5A9}iDm|{%6
z=ltR1w?~*i(Cujk-;*2=f7id5dF?PXwMc<TeV`;tZbA=4Lr*QwssL>ey<4Q<EP%Bb
zIy}h;k@)Q8lVEra_G<av;>W$k{F*|>b<g5Nhsa1HZ!&f?y_F&f>RA8qf^F^ef+8^q
z$KpU9OHBIEwVt?@DtgW%n&&rkXT^HrS60*eeaL)zC%}5RBAl5v7~}KTXHbS{fC|v-
zust16x9LMoaVGz4T>9A%0R^dy#!H^$I+dAB0fqPlNzVwEv}x1W^6b+K`w~GjrtX{x
z^v|25b$*rFEB32EYbKeW6DY)#Q$xaL!p@)x8wAr36ia6o%K?S27<Rp4)&(1u(X3zJ
zg;Us5j4T*Ky@8wX)3nQT2<Y<J0TIZLv!#Co2HKMzQ<+%~uRhyOw?x`Eo)P;gxqjki
zU<EvJO(?A?o6)iY&_Wy+xLiWKI%~lqK*L2!p2~_yC1sS0<`;j-`gVGfjaB|^>vvjC
zRT^>?I$t7Gl<RG}v&2);oLYCDKS!u&+@_(SEvUt3c0YOIUIBMc_i%>09B=I}YiMc)
zW@S`lFPz!>CT$D$71|E6)IdN1WeXCvQ{{3=g5+8g{n-A*@zLttJu8%77vyXo&g~-T
zNs5c|2UoxfIJ^&N5OPI0(u?x@Y{7tL#~*QVE)N@J96S+()v%(IoA{!j;I0O(%In{B
z8<B^gk*b+;tr^sAMhTzi5~-;GMd!F)&{I4R{#OEFQuo#3o*T9=&OP9j3O~?}h3PfP
z!`oY%D~qrGTq6o?B8NB+nsg<SCfqi8Ne4*h+3QQPpc`b~FKSJ6@?;TNF7CJVxn1kh
z$Dp?f^{EDIOgUlELp90863zl(eDXVxak$|C+7j1BoacPHtf*4_{<59Vh~rh~!NLmL
zB&To466-Kr%6x{6b9n-wug1Wx;fN727wz%Ugp$GCIwI=n*o3=k$Yr$?amhAqB6v_Z
zrcYqcpc$`xDt_fbSV)=Yy>$gnLWKuKv}8!(l65N`oL+bo=!!0;^yMFdv3?a#^Qymv
zyTQVd=dlzax>IG_K+Lnqe<5xEONPwMPfbWv%0ZJfz`D*1=$Zy|jRNUp@`^r~JCY>s
zdp|1#^rr(uX%U-m)=03tZRYJGqSE?n$k0kl@cgbijiR#?_U%_u@EG`1%JD&l&m~4P
ztpKI-dYsb91v)%488yj(IN3aKZDe0?t68gYtxSN|zV5!6Sn4`*w$v8Xb1pw%b16kT
zVkK_p-9*>v%q$!p*zV1bfZ0Fq`6u?imA>&_EFNF9v<E9<VzT-xa0#c@_7?N@XOF;h
zI;Z<PfAQmc%WO|&e|C8hG;-sJhua>F;d|`=tb!%1jD@C=TZ0p3`@Jc++(J&)yZM|>
z8Q2;kh{wCaGlV@egz3z_Szo^5FmQDdz7csS^JUs;8n3ZnHe~VI3A}cA$>5QP-q@`?
z3+L9w8Aplon8%X)28KKnPHkNrw+vNW;JqZ((zK{Ki=OI|+x16pq|p%4>Q14`+7aB?
zTv0<cP$UZ31)iC}Y9`MPSteW^pIOGtN%TaT(>t5CJre!;HT=hz9atXf`1+B3@UazG
z;^!=TyxYA&)i3kUVpXtimyzbALK_kR^n}Rl+=?uPolttT5{Yhtb>Xe55j(TF=9}nA
zx_1ok^{f%pJjM2q=?g!Xnu8YR=Pc;3UdC>raSqoJI@GuojUP7^Ny(@}nRR1^lA3)9
zBuVUs$ej|sB-;Qc%T=^gA_r7a1kst!j;RN2b{8`kbhSe5v*RcB7JKKGA@r;%C8n$z
zlmZntB}3`0e_g@U^ILc^fMvkGQ@7sIW9pfm2kRyMEN)HK`G7h2mM;6au}!7(AL=CM
z^v)XR<%bGNweuIj5;JbXh{9N8We&@4`>3&0HJ{f96OA^XACKgdQLX^F^Qwj7Erl9u
zJqp|%7?{eF5;!^iMcu6k{}f+MuRv@T#St!5Xr+A$UnjR(jq~c3MA@rz--K?*flav8
z+Ghf5K`PrXM@7OqhlKChwng*RprDt9w(IBZlHGo+VHpymfLITb7bFQR_EKOm(eomc
zK|7JhV?ZaQmni_XJ=OfM%y!2d@)=~veNWH?{1~4SJUH(ooM}GhG!W*pYlGIVJ(v-!
zrk@7Od8_-45joFUX7r4GPbDg$H_b%RZ(*a(+TQx8ri*iLY;7e)haYpuKIk}lFf)Yv
zD)Oq+?0|Vzc4sYz*n15(2BTU5`oWrUrbWxU8C!;8QAE^I`S+5(c}>I-;hOc}Ni#`>
z$n(mVOtk5jnKXJ^Bt)^}z2nW~nDMG$)it5v^M9;jW^^ZxBMVRqDjFax3ijSen7DS%
z6vTT+phUFjKUzX*SnOz6q=5=Zw%Q^?*u|jG`;D*YMFn+3xUde-77{Ebb%k@7A6s9}
zmXt@&z332dtFF@E(RW$Hd%qrNeVSCbar4!H+d%I!-KxKT7ZtN!_Lrxa`pbJc5jYKI
z(t@c}YI%lN>d*@!mTJx}D|m_h-l~glpRxc!*)jRp5NKk5@W5x8v%LtH^Pe5g4kx)0
zTTZ9n6<+=RwtE(6m2Z^Qr^`B2s9LiQ-aGH`*Xq0StFP`f2*bG<uP3!{b;Nyh6p-vd
zvI+&(HL-7ST0YyDwf+3{0XbWiUD;GX6Z^Lvuk$bXb~iI*lj0(!*+RWU<KG*biooda
ze!Dh00imOFu-tU(5Wq3Cfm}1xzRUGmY|OF03BrG{)e-b2&cV!7>lK|t{^ov*V0(oD
z*q5w6TLe)m^WG^A@>9oH1^!%EUDv~H+)I)$>@n;W!XI}&^r)Y`kIStcmIDy5V5MR8
z@G!9sHm><}L1C5BCQEH@OUUE4i-(-GYw#}J`Q@x_j3|e+5avx&>YC_S)Q7mv0@;F_
zPU^Mw3n0#K@6MG}S+U)=bu8SB6o+%Alc{ycXY=7gujpPbdu(;NtrI6oRLTYiy{|`M
zhj4ou_7eKs54G32j%Ko5CLL$dRF;&hSn7p#h)n<Cl*<b9LeyGRC$UD>w^N?3q7BmD
zH52Ykwcm($O?T_UPBd<VD)T~H2g5i~G7;J$e@Zu`4eTlD6l5}ulJ0vtG#xq(hO%uA
z7$1DSwR`oUN`qwGpwKjr=djw3bkF-<+a7D7*DMMfna@k7WiI)iFx#rS2d^IJOtU7O
z;JRRV55%`0{u1BC{*0S29Md#X{;nGv6eTOJbL)>a<;FX-`K{XC9btz10B-AK_R)_{
z_eVNce#xYq$P7KZG7;8caKDBUFR(8TVszWrTV8ia?>q50wb{Ni<jfNrX?*D@;j34g
z;&(MK3MXMP8_AiM@E)O{Z<k&C*^Bb|;q`Cy`34y6=KEWEY#Zl&3*S9L8up4<((gs=
z#N~&5iWbea&MIQ*tjzed-WQ>nP?q8nw;lZK*KYE<XFxNlZj=E<oI&}CE<$%#17$~{
zZ%rrRw8pQN<QRSF0_D9^0Ky~)&p`RXP7Yg#Z+i3Ivkoqgj=$;=ci~$(wFU!YFIG{;
z`n+n?6k(~jCTzm;6(>;rJT#?!*0=9p&!3`)k<Se@5f?SqzbB8IZt+5{1|_k8CaioK
zNZ*2rqFa#pCjx7Nz*E)bbKRMSm9<^Y<^;m=`_GADt{%n@zN)ig@yCiE`d?*Q3M1Q;
z9J=>Gu#hRCRe@SMdb<~y5@^O!yB3%VO_SI&OF$;*7MB($PSr#+s6h4K3~F+c?Kkpx
z*~Y8~2uB|-HPO^xx7oHeja+!s)?D@+*$y&U<ly?fR?7sNZ;q5y%8aMJE4nqBlcVKC
zi^P%=*8nc;8dwgsH0fxRS!hoc5HmR5NICX=nBueBZbY~6?A`|fSjrl!S?wK_DG?Zz
zJ!R(2siavL8An27qG6Pf0W>wzDCAn)5&AIp;S)ZeaN9P!b{3xZD&e+emj*f<AUW8f
zynUv%Q+ca-9}SJcci*zSkb|#?kNyM-gLaW8s&FwWRoIcK9G9NIrmP!Z6~|P+<!7E2
zoq|VL%^um`*_6hYO|y=@GlM?WI=-ukU3<2TCG^ZycbAcM=gW;Q931x}%d^{F+ni7&
zrM&Kzl;iS{#c4I?uJe1!R|D;fj+6Jfw*0CzjuxYnRGwF%4ymgxBX)N?G|qN;2!K;2
z6TU0~Azkc&N6r?a{5CnjS`3FMl>ho$1x1DyAq(xX!^h*&S50jdyu8+<QaU8n_7&xR
z-2!7dGD>lbJc)CO)Q!exmnomPjM$WtJO+x7QWhFhH4kuT4(*a)RKgeeoa?YMu&HK0
zRsF0%$&3C{tHDJtUNmZKn6|TsD(x97Giu4*Xeh%__UuTQNG`%|a=xXpf<hbs<`ST~
zdq6t|J+N%2*n6xvf1-p9Qy&*lq6D78c!!SH9@6wK^k#WSN%72bzl>iEkrKTISGdgH
zW41Vny;OD4p|q)*(UI6^HWhG#Lm>w4hln{Atmd8vWX^l9KSflHagxz8?wi(l`tx&3
zb6;rcE!NRA{pw-{_6wOA+>CB_T&Gy08N4DYsDg*;kedzhmT&mtg)MIH4EYEryq@#T
z59U$zTz+cw)KpcmG4+d$f&b!+I%lf{6`hSl?reUN6C^EEPo)NU5GqVMRP-8lJsScR
z1HV}9g7h{Mu!vM==R{|A!E1BjoLxqxBV!lX%Z!o;DOK3%Hq$~CE|@2Ia1(+&gfTmG
z2&rQR?bTZWVUWn$&inhWKO>_QJgF#jgQW_rD)i8c=sP!0BD2>U&JFt-eapgP2k}Y6
zJS|Ju<^HblbotoqotObWkU6zTyDQAaK#}Mqv79_1T#FB>(l7*vB(Smybw*h2q;jod
zPvjs<?;gI|9+Go6<!c*~#$bi8Xf70#K?N3X=((X7h;cy^ohoOEVu8`A4?zmfK6!h+
zJ2IQq(}D*=w=P!ay!sW7z%!EgvVq%^2B;!gF$tE2IbbPEvFChF=a<_^CCNWF5R7G|
zb#Hik)wA%Ck?4z$0aJ{LCy~NrRzOQ-TS3)zVTda%uX3d#9b0Dk4H@zH+LYlGJx?pl
z0(Xlu71oM`roXcx+!@9i;+LUBn?!&zur34@;ogicmbb*ZXo;)+Y*Ct4a%v{3=3`Fp
zyWFF=hunR3bCOt1$Q0rvhXjM_Lz>9Qnm0$AnHkCJk(1xCLo3C&tNYPTW<i4hs7+{l
zarbc?tiC!^1z4UjAk4*EhIwF4i^leIy<+W+#5PTna$Dwy+TBcIWP@dlRFL5t>te2|
zf`=7`hfjHYL#_f-hpog4sQjr)*HeIm-L+9jRiNDbf_x;A#8uhIj3T#o394_qO9NY>
z*W~8cRnI9X{&N^k7fRq%C6l~R-m88r*S=8piWg#X%D%rmgP9BY{TT<<t+M5VZgGE&
zg}(I$`zuf9KHzDamNlOg&di}?$^9A}dhb5w4oFLGws`S-Ge6VPm--@sA@%jWW7!xB
z$A5Pgyn${PG~d9ep%cHL;XVV?&5+5C!A`<uHkk4n33~h}fW8;_o~T%+*MQY)lJGpY
z;K|fEe4I$S9{u&}od;aM0~sUbNrgTnO`|tNOOP`6j$<Mqwvw>`#b11XicX8X`Q}Xv
zW>Vxr$m?thJPfdDjUW2DnO#N;te%yENmKh1v~%mZMWDdVm<<dAy*FNcV5+77%Q_Ik
zapHw1tr8hLaP@U(|8oifMq8}x3~cb3f)OED84nV|#3-mQU)3HT;$$u`dy%2I^fH7K
z(dYH83y68n?xjR5EQGW%GA}HID~yLkPXHO?16WhHTs1gAp5~K47Crj98dW1!EuRop
zK}nmkKRWwnC0fk;B@e@$acBuyiXz`5_xTUiQG8T>Ut~p$^^W|4EZCe({U;M`7uZKb
zx+_9yqT#?iX>-|S>1^wv?hSu(PDFG3D>6#tLh-?J^fxEjeYewi+vLcyjxk$&2wSQx
z_wvHE)vS#gSbl)Zb<fnD@Vm)aR;>pyzvHhQaRAGS%}Ed$RD^+)mN58H?V((3iV`&q
zI&^JTTlMN!8g*%Dz9KE<fN)K|D;!_nnSH!$gx<CNEKPX%_ex}7B~-~yElaNkGQ;rN
zl1W`2D$ccE1=l6GKHhb{w6eYjfcVj#8nY<=DEBFtZYkl_vu5|dt3~Y>=XbWTtC8TY
zZ56VFZJf-A7+Af9VEXrO`_9io{_V^9b5uzFa;LsqJi8OYtp28)X&~S5KKKswj+0=-
zR2d?3(*Y0zPueJYscEgr#gfDj(mRuW{Tq$SOqhBg_GS|^%KlXHun<^b(tRlS^W_wh
zb(2q$Mn+7-7;nzqB_yQ)F<;OITAPbN>UD?N?D+KeiouXbiBtD=jP9Z5?&$bHgpJY;
z&|3MtU+<5nF2eEnKAN?!dETE*JnMG2ogwD8u%OYXoo7TP03d*#^0uO*WoxAMLK9hi
z`PdgF&u)oOnn$>11|v1yZ{nd_4YQyPx;P<VLMh%Hd(E<H8-O6!n7&C|^^yGGG}i5{
z_rjwRo_6B)Yoonp8ZEIXXQvB@uWl3h)~>F$yP@wtIdnU8iD%63T8h^E5h%eW=0JSW
zQrRtFStqZUhmE;n!^+()Hvg(U+RDQ!IFz=xtgZ!HSfojquVk%TB;p>7DBL&9P_tO{
zGh3Y03j`fd(y*&w7^hp3_5hzGwTs&!7!ms<V!wAe=2kTVlA}+E<*NXZ&=4K5E($0l
zJtVubV6MZ>*O>PD@Uf@Q(l@reDZu@9hf?>J{0=UZjJ|A(upOXm=kk9?MU=G~>^BPJ
z>py%N*Kp|xwojhvfk+o5G4*B*K3=!{jEL&7MC{B<T5F_YP&f66e@j=oFv^brQLJo%
z&E_@J6h=EYrY69Wh@}ns6+TS8+>Ml<07w;iv=hYtq}J1o*y=RgLz5ThwBIWjcxf?#
zpAV5AXV_=NBDY^^8_Zm-@OC~nzQEqnPFR`_MUk7uF7dSMuSpSR`Jd6lfwvY+cx#P6
z-x|=Mw_UwLdrtCJQfzHfU&RW%R^x>S&!crz4Ayv~@4H(sOFQF>px8HFDn+QNK#2X#
zexXPD|El75r8OREq*3?6RpYnm|B5GOJOW%?pZ{04Laa)S9oQ6LV-g)Xg$;?0iCl;v
z)MUsSeH-fgZb`y1pQurv{NT*OmD5g7u+{gJl|6BDD<ykMhCFzu1)<@2<CC8xaN-f`
zdw{mTIK1r(&f*OTcTj%q<vc+74F&_M>S*--t+|o>1WTQ><W5QWp+)a&H9TF1Lq|KI
zB3(iM!K)&B2Vjp}E8}LlkPqjBAw`77%kQgPaSnEpU@!Jq2IDCGRX?PFjO?>g0E|QG
zxxk@5I??c|ZC=UxeJIwrcPmcI5x`yyM^kk-Inn?6rfEP(+3E|M^gf)LKz8_S55c8#
zD)v%^l>;mGuy@*(+U(d{&wZ$9(|0^vI;s~{%s*HKevG&dpeiWuzTqmPi@w0+{dcNN
z0IBkO5fumV0AJPwkb<l^%!3MPgzwX$_x9pF@7Dp^2;*8t_65tnTn!==NQ9MD3t8Es
zRoN{2gBZ^*4l9A*y@3?PZYN0(4<rHVEhrg6dHCx`0J$$fp)QA|cg02A!VlF}-6x4y
z3xaA+r$5LHrDRDldZ2vyRjWj=i%yIn*0Xf@Qr5bJ#btdKh(2kzg!y(VTw%RJ)uPJv
z9Qi`-lLZF1B>OLaUn>B5Yo?G7J3a{Ki&}UIYss<(Q%=@O2^Xvj8PVbe8=d*YN=k7u
zR|+7p76{Jj6&0-7v=XGL-sq2WMdDW%;SBKv_2w&pc2=6mK|+T->4bO^+9aq5nv(JX
zX|Ow{=(XBQ2&oa+nL7>8rj}|dqIK)bSlCXW>;W6zos1-jQ=H2AhAQSIq1o!;>dC}T
zo?>_uG1A?W7eHq1w54PbO69bLc}G&I<vnnJJR;CJ#m^>))u$V5R1pTdR?IuE4_GsK
zAr?QCYJ~I$(>{-?$PC%EEYS+2eZE-0iInoD9C0)IymK_PtkFK9VP$|XGRkO>*5B3u
zmE%9z3IVcJ#~q9>gOaPObG#gwY62W3_Jv@^4wpz5%xrH2;?NPfY7GE~MK;O3vwxyM
z+{6f0Q?;^sl8df^7tu*>J}d{5r2T6qj0)?mSrVVLnV4i`YI8#G1rnO-(2cPt9v|Bl
zqcg3&5Z266ixERyd3P`3xRfA{w@Q2W2qkQz-e_{S_-rCh8g{j00Rfqo23r9lC^Bum
z9!O&3S6Vses_;bIb2ovXyp&j_+BtIoguY{+!$qSrPRfQnn#mBB)SvYvC@Be5CG@1i
zFp@@r37rpS>y^l^M}|*#vGXm$THc1m$4<FVIgFMe)ngj_sHgC&t#yeWATX$wix>n=
zN2yytNRQXTebPi!8mo6k0RUa}`suC#%J1shoXXFyElMom1d#$tfj4{C_LwYfWB|UM
ztol3T_1H#FIK_xfzZ;tc;ZFCJpG6L@$Ou7RAnbjxV-ALHcQN~bcrBIr{S!S0F(r6&
zaR$uFWVt%`YI$S+40!O*8fDYW00Z5Ny&NDYOl#kyQ?njZe2CVI;9cU^yasL$7wBL$
zO2(Y^Wj&7UeD?JXKR-_SodkmFXt-Cw!ssDEg$`K6eXC?JjIg10xnx9uSq!~51CAWj
z>iPqIjz$bwd)rMVVLhI`|2Rn*Zd%_o)M=7W?vGf9pbH}zN9K(iz{=w!%IGR@Anp(=
zUQ~QNGrp|NjPiS@ITdve)E=Q&R(vx#GxN>Sycaf0pBW4sX#>xt`_8Fg6KeMsven(Z
z0D}8n;U7-qCs;?MY}^8)BG~N9?*!%7uLqm_#V8Xm|4IxhU=J}@Rucs)+3nQ_q`@$N
zqChUml`^w_Byh+U$fEXVl$1st`_&V2@1j8kU+_hd7rxV^MMvag2nI%!Rz*UOz&mtH
zE7ts3d9e5bG?7|zDL-8%peCS(w3#|H_h+AhI)Z(%7Hh**v|U{9S#f}#p;e4A=gV#v
zsO0bDR!#aYw}WLlhA~}#a<6ei2(UEV1a*Y0Xg65_`ohz+fmU0`y9{gB&+8id22`yk
zu2*4#D=5r9>^L;|lh2zdZAk)jn)8d&5D1v<yq^|5w-+CHzn*#JG72n1Df)t6dcio#
zD3{w6dAaVh&z(g0?q*b%397Dm?vG4poHukeVPSIwWQ95yOMWm7hAydkgB!rZPv2A-
zwgF@ve2;Ai=*_Pmlzb-@Da@q~X9RtQG%f97cEL2N>n^F1^dyn{FSYfUWD@_3r3%R-
zUQ{6a+5gIZh5EfkY;M&qD8EDYybZ`Vz9K?#5)V(d>M80;#>RaUxbFlx*zNXNd4Hzw
zm|zN=UtahEf7c%5<i6(IbIab$x*#Lj{fX@{T0CX%ONQsQrfX60VT1#qPuS~xX6E7J
zcQ3TZJUkPB;sbvt0Ap6N&1&pqq{^jvQcZr+#McN@-KU>EUm(%#NUyI;dJXl)5Vhw^
zH&TjRWA+BSbVbhyW4U%4$U)9{(tka#{KuPB2VmchOP`=kEDDf}m8aSbmW}75&nPg%
zh%yIG9RWYrOX`gg*cp2%S`}}6<-~_tLC_Yf4-5i2t8d~n+uszO6d43qlhl``y(mv(
zty7I20-)9*97D^c;k2pbrX1)rU?5f#Hj$ex`@>YJ4M(G1F#qFtL`%gjBsXb{5T~cV
zo!COdxCOs{P^$)={$X`|eAsgt4SpMUKoW6=#y7xTECUu{;wWE=qX~37v4-D^`!!Tu
zQ4KqUZj`~??M+1Z>TVLwljmSbw&-O)0`js6Mo<r`4}2f&6dmahbfIHu7ezfxRNxGc
z0A*})-<u!=GHYWfZ;>f@V_<n3bYqe27k@0r_IxS=JePL5hRK6juVCA*eF;!Xx}~S=
zH1WIw^32;qqsxt%8~>7m6^+3hn%k_sCz&+CM(k!A)n_%XfEDdvj(RWR4LBtGJgo--
z)I2Wi9QR`gg%ex%vP@uBl}5QlX1IjwH?GQChCaAVetJEe*v0Q{zh(hczpor2<m1u5
z%UtHMQO}Lefu3A?#}5V)AckJ-B6lwWL!^twGRZa42bmki0d9WRT_Xt;Ra7H_RjW7L
zHARn(@U<Syw_5HV1dGgo!Zf*u)Usc0>;0Nnkrq3l<i^N8)&E^gkVMN(NK(;B`_(h6
zPM`arDp-AuInjPeVlgM`LP6226lht_U;=zO{LqBe)l&$Pco@{*Y^RA4(-9NFXioPD
zDq{)eM&R4WUzn``>eJtX{kz#Xz6QU0lXZ%bxKIi%(^dZtssS8(u33~X^MZ{L9YUC$
zAa?(++z)%iYSzxMtRVn-ph}RG`Y=&$^Iv65X%}smx$-w41fU>bdG)&Mp_!1vK#t<I
zF1V{Z$1|rhuk-!hXTk3ejrRIpzgaH<*$Gp+HUcVwN1<eo-aT65IYeaNQ*qNbkUhEz
z!!_?xj1XYZ?+X3PU@cI2<|BTJXs086s}C`6P$5az8>`{KW9F~dIMn2SCRN7&0O|)z
zgIZG5uC?xv!WGa@Iub2HH&-fgdv_~7g&<^8#V~l&i!ZpkFVEBf?=MBniz)`B=}?sQ
zHKkmYvqUJa%hxm#_@fl3MPvOfc6CzAUl1TNU6U8-2iT8{84@Xn70Hn>pE29xCAjR<
z%N`%wt@Wd-Rem95ZVOd{xCaogB$wCZI*O2K<{D5RzA`%a&LFeE8Wl-kb@{b_l77Ex
z4#w}qacU^rgxVZ>C9U3dh&^MJNQo4PAr6{Xt~3TlNk&FywT<4NX^)t7PLa2yT30<7
z>Ry+aSxA#vnACf{#O-!iol%&iF<Z@1IYBssbf6{-Cw61jHy-8w{F3!zas=oNE2k0>
zGFM{HaXI?UK^_3a@7n6=Z4GYGfO`N$`6$hv2Xlba@0*TU@Y1J%^zC2|*4$WXuF30>
zq_TwBNrB`}vR&nI`#1Fj)g|8h1`rg2#d-1dsHt2a?UFye#!a$`#Oq*6Y1@@nAj5e>
zRR~;wUz+IQ<zTo3tJt0qBWwi>6PLN?w8ZEN>YWd6ldx2#BfvMQ=cSzS>Rk68BJ&0I
zWar!ng=7LmvTv~kt)jmUo#r|sj5VrOFotew9o^Ls=Nc#{nyv2!3eQ&H?VMAv6QQvj
z5^j@EwFX`O2O~6p4e7`zDOcFv?QC5SPo4c)nDvxYtPO5g1_N>6RD5Q(%_q_tyebJq
zbLnP!KZ1B`hH6T08xb3@A9R8wq>fK;`leUyV6^S+w26Uv(?$w}lPc4rFDYd{0g><`
z4laBWY|Qx5(ay1d=S_nEYyx={&3D(6C0RO9h@$Ot-7DlB=R%^{$sPuF0kkp=`dX;l
z4Mq8N4C58GfXO`oCK^5@4pkkJT21;O7bF!#KYm7RbTT|on_(&dNQSK|Sd<0Oa*D=g
z)=HN))~|u0j2no(5bHe()VuBVbD-5~Vb+htS;Xq0HM!r}GQyDt$?>{;!?$*kPny=C
z38mpa2iVp|YFWt3IoU<m?L|-J3&I0a49RT6iAzn1agAcNyXvx+s3ZA`N<U^3VrYrR
zVvSug+Y5Ne_5n~gq`9Qe7%0!zf(Xx}s5b~EIYfc^>G0l#%HMeyRAC&?Hqc->-WOK<
z#$n3w0lfXy<fAeW!>_e+m^|ug+MPJ~A(qMC+AI+>C7wagl=-s5Akg!cxBiEB0(o~!
z3b(Wv{)T^#YcD7=C23^C%f{>uIrE|h2vih%9>=UEh?Q0&HDk^r=N?E(^Z-4IjGlR<
zQXf`<rH;p0X_yj{TA*53cmgtSrhAnrNx{0?T-Y`zdm0T0t2vHf(Kv=>(dn44<1p?y
zfpS679-uh+pb<7TxO&UqkXyH}gwkC0_W1Xv(+D<AHGlMw_MaV5h@25X&@WwV<fq;F
zUeD7-t=PrUM)6!N5e55BCfj2AHl#=~k@1UdemzMmL(S?7N2WX|Yly5+y2w)sGY8#Q
zz9(h@0X6qJ%M4A9aiE4!$9-T(?T^s3?49zge<qSZs0YnU$DE%c2C?pnNBcd2GSgwJ
ziU%PZ^d-nQ=G#<6lM!S|UO@T7&F=<5weL6eg!RioYmz}pa%QR#>yQp~Y34EaAKW5~
z{KWZpdUqtBe58?P2r0QSajq0^axqXSmK#mG(8zLx7*6TIS%V?}<B^wFJPH*nmP5RE
z8d>(d!-hVx34+Kg>$L_*^O*6%Cq5lbegw5gy$OGf&S1LW(RYU0ID+zMUb^VwSge>w
zQWZ{-gu|Wa&>7J|;T9sxlnv!Je&k0#4i@k%@7wOJ?v&-9X$LlIR<=}j?Axl@6Cc<I
z2v*`(gUARoEn?vn(~hF4fQ#;=a=D!%4U}VGqmcO3uDhDj_hIO_YM@^$WU2kC#V{JZ
z$NJoa_vt}vnRR-+<?Zb;mkI7!`8K7ac$-Tqlxh3`Z`62*-f>?ORc*Lr-@>LInbjVf
zf~U_wS5=Uq!c7`TXL9zI1`b0bGHmH)&?XV0ffd2==n(?q#S)G!P84JMh;qkhy%DlA
zz^M;OEGro-fi)g4_|0<HM7w3%17?)YEvMT$q5eGelQgUn?o#o|LM#(ZE+RiVxfAJv
z?7L1AD-{Y2$UQLO!xHp?MyVH>lIcPB!Im=CGV0X&nvL2mrBU86<B`Pz(M9F^W4DZT
z1DDR5|1FU5^~y*A=l<?25XcmpSvW$S#_10)SiPNbO7zCg<RDLW4{Nm!fYDrY%Xw;0
z{=UvWhPPs9!#3ik`#J_+9n!*E`}I>ZI?^xi@yE(53;M61Zv80G;25XclSVZUUE>^P
z*>1+5N7VAczzL$m%N#_#;Q^j6G_70IJgMAFg^T_G{wTS9t;)*JXzX1+oaT{5ibDB2
zqRv>3g?n#uz|l3Ej1E+e=}idBP)E@y?y>FwKBYt^*Q7J(3)>7{LS{vvcflr}d3MOg
zS{Q%5%W@Kya39MrRZ~PJ(q-8L3xQ#+Bj&AFmpU?g<L|!19#m}ty_JMWi=+sNq`vUb
z(2@q+&knBa2JbrE6}sPL*$oS!`pz#LUKB6b#ZD#>a(ZUNl*kQ^UfAsWF*;{kK64%(
z$>eo($!(ez4<JEtkunK~LdPH1zJGi8^eQ|+9<FiXNGf4|wnGS7tvnte4&8HD_x?vC
z<a@+(C;d~ns(kHjBQve6gB0=JU741|{v;*+5Ney@6`(=DxWKh26LQbt-U?nn^>XA%
z_NTtF`+x+#H%TQi_cqq-ga``ekm^`ctm59iq%1bmX~==A6yKH9*N!G<c9Rub=EytG
z)-cqe3s=><c6bQ~{qXL+29wd}+W>hi7kQD&qB5k-<t=5>J9QA(Y{|Zlq`*?3ucmuC
zq@UgFRxjgrgFiBDkJ;I?$T2@H1%G)P9(5&ATzL<>m_@PjRAdDRo1H;?j514X*6n7$
zy6^F(pd(%(jpe<Nj2WbPu#M8Z%Thy9KXxSEzNd)XBEJRF6?+XSx-lq}EBrpVDE{k_
zbbjk4@5I)fmvH)Y<YoOMVI16<AE<3g$)TTRI6_@2Z~Y*iTg+T4m%@gIbbcIiy#3X^
z^zn@M!KmZZV-+4;7}MnKu)K1+B-|Az`NQ=UcAVpz2ys_<By;rdsv98Umrm%nrXwI_
zU$j-Px0bX>5{A}+VHDagKDP#SaW_<67WQOf66#U|(wZayCWpIvz*-igGeSGskWO<N
zWmK72OXu4GKQOZB7yxMGq>a%pT$1nG<z>HKe&@#|unz8Hg85>*b=2K4Fgb6T|Aw)`
zT7iRUMvh3l*}D<Yg}f0M`Xoc)At?H=OE9X3t+#e)L3DnKz!4XbH_YRgNGfhP@+s{_
z1L4h5b+6elv5iVfp!)vn+Jg@Rt?=-Emp}L;-jiPQFrX3>bZ>OJLdNLLL>b`!j`2M+
zikdMK>+hG+DHH1Qc?_+`zW7&AFNBR~R_chc*whi-`}~BJ<m_qNHaEg-B9>6o#L^*Q
znc7im%$X3fY5q46$5=7Sk@|T-!T$5kBhvU$2P!UdpA!DUZhE(=H`7-&MElW09q(Or
z`U$&pmT@DSe$Ib4#gybHEr$2XST{3Z*#3;Qz`LHK7&PXGjmt1Y%=~OQwHK?Gwy0A-
zOCa*b_QN5Z+JzPge~a+>;;NyIjT#Y|w=6i$sij6;;u*{ACS<74FyCz^A1I0d3C!`(
ztg7A7;q76R)ZH|dwKd6UW^7DFUmuG-7^F19h_t{0pF#y1;|0}rJ{AeI{Dr-YcH<o>
zla1mp<8hl#20VTdi_RLqkPQ38kdWacnp|4+gD(^oDno5WzWr=b5<AJS+oEeb_Vuov
zU;HsNn<$YI=}M=1$uHGw+PBK_m8&xo_3VesfQqK4%(Flbv~W5NoI)1?y0U3V&Fy&f
zN5PZN)Epm`-0hr#yy<_H)c`4*yL+-lK2&H!J6S5Ms=+_IkbH<dBjM*U<4Z`<Fkh*d
zX}31qk$D6}F@3o*?)614mz0ySibq^|7{&1s<qQ$XwA1Z-867C^sM+-UH%LXm-*j{-
zc^)!i7A_63nFp_b#X9C%9s>oXq0gZcTuZ8P;Gs-!<2ynA+h44BO76GSD~@aQsmQF;
z=Jiv}-Z_FNyK#Hk>?y)qgh2E~i_IphwJ5XrQN1DWb&ImZI{b+jRde^eM)_3>6z2;5
zl6(EqeJLc>ozr<Vg$}zo3mP{KTuvq6t;@iimFwq1=F%9K?x<ntvw~<O2v7m`v%>F|
z@%Mdv>v7VLD^J8^4G_WG*(E|qF#ok+CBKnsBkE<PU|3g9NFDGJj=9uqFyo3{0U}%h
zCP~&Hp4%JXzc~HzWtdNhQ<ZR}P5g;+6aM&ZHyO<g0zl#Tc39-~*m7|oQ+9Z-27RA2
z=TDeK2YU3*13(ov7OwGQ|0!I4eS7~j8F#3C>thFki(MMXdCI^ZsG+>uC4dC20GTy!
z<}_q0`jL*6k}=S0+XD9lK9WXH{3L0^(rNckT?2N%`;?w5C+#6i=kyG8{_V%$t~d$g
z(Jlm_pwHs9?#sEfD2*p>VzKf1i2759h0`g!9q>^Dm$8}GAMM;;dH)F#|Bb|gpLx~a
zvP?TYxL7v*N_7+rEuEP;0>ni-0m(qkB*c)!{?*Qw0+fL$GC%%O!jzZFPbI@`ouI=a
zN}wH6|NTg($$&!XuTbZZUjWxXNS9nK9U$GrMlrHS$UqTRD<DI0R)HY>2}0w9=x)K<
zgThKM<#3FkPCi2#a<wE*5;$H)-#pD7%4@z`^ZQNxq97EFZ_>?yF3}<cAj?Y{lFQfd
z1Q2x$RQI+(tIhmMEzmyw7&K(%XG90<fElq3K&rO3pFV%H2kb89j=-yRfnVnl+5w&#
z5qJONOO9W^_4hl33X`cN!g`rKSL<!T05zE>b#uM%$4*l4Po6mecA#g6EAN2jr122Q
zx;Ll-_;srKZeJ-EGDDZ$f1uZ_4$>J1QS@!TfBiN822pRmHJaW6)E$Pv$gTujv_$|^
zu_ys3YDprV+n(|zz$7ccg=G)e){1@bZ7j>I`cjjH{;u6N&w(Smecy>!<o6-@`>F4s
z>OfAO=O72;yAEaS>nH@PCJFZ%4-~N@oiJ?ZiXY3t+XS~C@yr_jA^e3w`cE=HRGG0A
zjZOg6B&ISLMfXSa`Ij>?7bI8Px~RyK2zvi90E3Nai8~^kPJ%AC!~?*$XD#w2ZX3w|
zItp)9h^JwL9QhSFzaROZ08i`-e9N5#I3E@CT#<g#?u}JCvR_lSI>97>>42wwa`NWK
zzgvKRbCaM@V$M8jO8A)4um6Q`Lp=x=BRQ$-KRnp~^7Q}LU-+`A$f5ica@(R!e&@yi
z>)-wlKbYia5YkF`hRDAS9{+My|MOoMx6nplCnK_jEp-3*zW(!n^RNH3f`=7TzcMth
z_CNd%812E!f=y?sG5skd{%?E;zANP5-g`Jp{1grR2e<WKKkvW)3jq*tnyJb!-wpU5
zeu)2HI0kY?Fe+fs|5gpCld;z~-~{IIfAFJ@h7$_hiUhZvt(?O+1IiB&NY0?nI(T;e
zk=UP=!GGrs)pJ8Z6;=tR5KulA-D^M+>}B=@S^NLXFNzHnAs7KH<%)X2lM+U%slGCo
zqsUG8Q%?8a_!#Pp>EkTlSV6k{h~mLxB3LPMJpQL&>Lxe17m<(lFaL@3{6BvqbwM29
z-_z>MT+aWWy~zLl|Nr;%|KIcX|JVB`XFxpeg*U(hKU$^ukGS`_stXeTwif$06-<u<
zp;!_BKIPyC5XCxafMD5BT<cd8DJQs<=8=FY@EHTuJkHR2*P_-gm@5{cZ}a*7UZv>o
zlxh5t76xX2jDg}CTXYCy5TnZg=JV3?2QGnT10XB48mE$4l@kLrcG^T>BN`y*OWYcX
zbFVnpnt`XyU=+-R4p8U~Fkk{&(^D?M%v26Q%04hbUZ3F1bS4zG37|9`0|duN$vC2m
z(5Z|8v~u}qyR??DJEP0UEVulJJ&+sM>H}TpDB&AeD$A5R)SLvJmji%-6D|U`kG|K|
z*L~T>ws-$U@MQF@ae{aPo_Ms|d%Nh%UvZy4eTaS7_6dk_hsMrK<W;TT#)EAg*On-Z
zEEAE+r8BYm?h5w@b?`hI&SK-PX6{3R-fJPYKnto~3m3C3n<BvzxP`HygL5$!U|PEi
zqY5dzO<^ndPZyg$fZR{p;lW7PN?@KDAvhoBg@FGGtW>J8j;8h-2cS#RQSO<ar4-mp
zcmOO6j`ZAW2tZtDxUT`7*TmH1*$m+89IOVw$$ajM^AB5WihGskU2KBnWa;B1oU&Mq
za)8f;D7h0jyPL+480Hngl_OIQm*AiaGm@Ef<8@F#r^*6F`6!?}r@>UP?})+#&{el3
zN|>YEz#!T&Fr6O9lS!cNoojcR1xHU`$$^@$%W4a|gD%k{Ba6@=Q0*Fn{Jo*!4Mv^>
z><`o)&D4P3t;BiZJYK+#z;l-Yk6hn?VE0C%L)ZNA9FAZE$0-I$y`VRdO$Yz;mrp<v
zVGBO_6$}^!e2(1nJADcwTzWfBa<fN)`OxnVDD^+g;FPCMdM(0&v<R<!Ba-*8V1)~j
zOHi6fXoWHCNels3{DQ=>9Pb2Taqn3UKg!O4<6SEWYMX}=^`9%8<0rhgvW?stPiRD>
z0ViuV(5xcMGfesQftbK^GYo9B_@*+#>I15uZ4Ac3gj6ahU*0?hbVS2mz-iD`ZJQ+7
z%A(=RlZd!6WdSf#K_imXpr(02pWTpISuh^H6m+u^ko*6@ckU|BZmIz}iJwc>9=`T?
zS>|AhT5@>Kx@q8Q$zTKFB5i6g@aL5RqO$@Jb01uo`eT&dDxXwzWnX7RnGz)-XxQGY
zgLATi`B=N#X-6VK@nHoI)Rr*F=O78PJTSdyT*c*z(>jefS%wgV1gW-<7OR;e$QR1N
z6koTXX>W6LdKGzYEjIeqF4oEppcOTYm9KeACA(b^UJf*{RWvOG`KR-klTIlh7$hK5
z!O#}Y%Uvq5sd;A4`zS>^?zo24@A=*g*~}xX>UeENdWEIJ4oAkHUw6__S;bcNdk}}a
z3x|Q;f6+%g;p{0z`Q_CfRe)2!4{6tm)LnrCACf&#pL{;n;0%5cz(vJgVGxw6_~k>0
zKt+oIXw#28WApb$XJuy3Jn&YbK6>$?EGSvtp9jK#t5_6Jxt3uKwuEPo0^RBY6gIh3
zk0)@NR*Fq!{ES(8RN*)Mu$XWc0#+(U4A0ZU;{k1fk-TCHoVn$%(UU4^r$SAYP`-*U
zo0tmc@S_w@Gk!)5ISer(jrBJF@>PvRSbu(~X9@KlI7B#Rp`f#>*9L)Uv_sO>#Zw`P
z^`xKNGa+)UQlb&*l~2IcAX5mhm7pRKAdGhopA&^JlPWM-bq8JJ1-jTYADMYG#Q7$-
zt%Qk#^hPadL+!ZMJ8Sr=L_2VjS4wLSKN&5=)lg>C!(z&nJrCq;Wsi2ShL?USX++0e
zvA$KGSFXzuyVgVw`2gan0tenQk)zI-@_+HcXVK2-137g8l=Hq;gIU^l&%_ZDVOsIE
zyD59oF5h6=gJy?3T4dk-BNtVT_Fje8$#W{ja-rSwd9|ow<thNrUoFB#^^&o}gW16E
zqr{g+QrVsQ(Hs+~5qLjiKq}UAw)b)(sJVVAf1$)df2lU<$J%q=uyM{bO^ulHd=a$|
z0G_WKHRi^*KUIdFJ*g<n-IfR;9T_MCKZ*3wR&yYLU1#^E2MCXq`alL+SwR~W1tKrn
zkzH4ZfhjF2Zl{^}3TW4~Pf^y{ARnULc6nrXgGa!OUeZZa#f$$U+}|TKt8|nT0HncZ
z$R4(;*t$<gsvOYS@xR!5NA1<(^+O~DNia57_}g#W*CSWqvitGjR4u|!K>woYw=o9f
z6Z2=R_}N;kgLdvs!*&2!i<$BS@a%>gDGkcv791k0`{pUMA8?XWT|kUk3%O*yuDt|<
zIW61^T%#!g+lkjK8r}xb&wxz0t}*bp&DvLhtVboEQ;5pHxS!uill?Ld0Ol4G_{rj$
z$>$X34ViiHfDmdNLo}XQ0d^L9LZ@R_id1|AesS<BawrPOH}ocCUmRYv&i6GE7Wc%&
zAHPx+xb>a_o2a2UR@GE(HVaK%@5_jod&-BU*OZ!iuiG2$IYnXAQmJ-zKM3LkG)J(#
zg4c%bttzCJt!lB|iGUM_XhjvZr+qKp1u*&Uet%$5hh?H|_0|KHOx`7ZV36=KpI%M!
zK+`Vk+Pe>1Wo&K9@9G^zNjwBLl0*Mfx;~RcEMyt`nrL=@ebtAKy-H)3vva;*a-BSZ
zf%!ir$TL<rSaySq1fsc=81&3(eXM(HZIgo$SCd&5l9AN&AeE%J7|ulfjMTfHQ>%Ne
zn%;dXi39=2VFALQC%C13dOt#eYHgzEIIgE>QAi1bTt#2B+r*J$rP~^+!J^l^GOk!%
z_Jq#?IJJ{(m$&?=W#&?@ELZ^&bUpH8xeV&Ta%urU8=uOUUXN5ZY0?edU(cnn@L&Zd
zI$jl`II7wvT2eQ(Uc-g-RpWAJ7|qnhlA(KWkLJ&A0(uW$yF|5$fz1es6OW0&EVio6
z-B|<9+9Acw!k!Ygi>p+go9|h+=$!}_%93uyZTBSaJa!^vUL4<++@?PJ1f7YrhE#U3
zUa{5C+0rD!IID2sGH}n})WG;ry|^h>e5J)iiaI(B{+w)k(+{AFF8$YwBz38nhF-Y8
z>r6AW+Hdse@@SWoGoqo}A|>6^=MJZ}iF6T80HxHWQosxU5N^4oza_bwMomGx9Pg<-
zu{PB0{Y{W3Y2aH=joO=Qt$2}5dT1k4bdC6?tfFt)rTN+gjkQBJQpP-j0mg7;dhDGh
zXVxK@JGbNw1Vnple@!4K-=o&GnY$G|4fZ;Xj`s$<WBh>ZQuyfB$bV9RN^?I8J5~9W
zP}X53&lz1ZfQsI8j8k%N{a&;cLPmKBATY7UK5cv%JG}0UdOv`!1IO7y7omu%i?kyQ
z+<-*k*Uj#+e`gI)4TUoLcfqifEPT+>3q0hhn#L^>z~7)RK{XEKzurTZv1i7COSFof
zqn&-KSO}=05v?bVn|1<BSTM4o_HHCR!sj5Q!M09dGF&A=A48?U)OCh`CbZy;8N2dl
zpt`WGYkzm~r}P@Z`HbT{6$N1qIxa6r;N~!3y1dLG?9F|$=eQDQf=5H-AsMhlL(610
z%Rn|3I9cK;paW#U#W9v16G*YfOvtPts#(iA-XivvonKKsmE_W7nT%VSorQRNmEyG~
z!o0v^|00sEzhI+S*mi%)(HF_4NI`b~F7OyK!h)zXPCSWMHDFW3xYaH%gdyuOumy#}
zZ}<fG^#B@s5b-qe0Ga<buQBuF59BbwT&w9>_vbF+hu|!DfDB7Y69*O(QNijQQmMGr
z;wIav%wi8X1<4d)!VMna&Tp!6k<8YDCmkq4YfS^y!4@h6mPf-h4+Iyh8kW;Z4z$*c
zIGWCIuH218w+b)r8@@D&CBQ}<$U)`Q&7u(+`U&XQVK_xGpdi>2u0M~-0wotFS*R-;
zh?U;%gU(_R@saab3FQVI?TQs!wrD5_hr4Wais7-aSj`tKTPN($;$X_Z(?y$NEn86p
zS>yZ*E`O6w2eo_G7uaA9+SH+8FX_6aaYRu<2*YOb1ws{<64W3W{Qg{y8{XsdJtXrC
z?`5N*!+Io5h7_a=F_|XdD*HH_5&r$2CkRSz8f#9A)_E<Y-Mn4Ru#BnIz>c~@q%dLR
zi&sMXz9`U0?+P6P`CYtaAe@v(67qSnZ7@!FHD^Pu=ZNk@qO%uR$D-by?eYF4rJ~`<
zyTcTFe+|1NRu#JaaQoo%HgxFhu*NorPH(Ib@=5}@wZ&H^_Q&y6^n6bBlC71>`t1(B
zL{hbRCBsjK+S4bjT&6mO2dm7AD~scoGMH1xGSnOR&Xo&tjxQ66>Z8N=Vc5eTf(fov
z2>*7o2<ehGm%VqOi5eX^WieSJWha%8-#qkYoBhf_LxgkCwx9P4(RlArc%(~cVgER?
z6EsR=vi<kh<E6!FWS2Q;SL6BRWa6JG`zT-Rs8sT~Ap;}j@lH$^=h1MIqF>t@RCaEb
zXMt;(JNz<wF;WpTvOtl~k<3_aZnnN-k$|@>mUJFDM2u7xD}lm^?hBcQ7&$}6?DOjR
zkfkc&uH)Wt?`cfa^ov$?wZx0L!AMzo2%{{9+>5DDKIU4V?gNt#4gghMCx0tauM8RH
zb=;bJ8@>1X+3?ZPr-XMABGz7^>filqP;W2X?*~z&kA{X=T5PY!_vRcgu!l<6%_?95
zmwW}74mrA>Ee~IJJf7q0?WTUvAc_Rdm4^q_cDEt-=fmnW$@Sc;AhSE6-V^sy1Zh8C
z3nw|<Opc-evicNBP?;XIYn~0Ye#uNKq--+XhzKHL?PY%a#cB@9-QR*e+jIFlD_SHp
zh|%)-J5mm4<T<EvWNrOTS42OfnX$FSAR0-dM6WDK`IPaNP;vc9EZ<B6o1bzLXcfLa
z^wWZj{cTrcJ^_kxSP+(6;|$|4QjX_x7g*(*aM727jQ1j9SaNqdiaD|n(Quk-LUFNv
zH+%w-*f4bkm8K%wDqOg!38R}t(5ifI45lI?wmAY~u<5XUwvz@WC+AoYY3#H@p|bwx
zAiIc3%sXEh8Xl3v=RgH2vZxmQy2TDw_m(mJxIx_4JE`?lEntM0%}zjwkX=FaT9XLv
zdU+a51k9ZkL#s|pdY-Of>|Nlc1?KPrs(p+d?1f1&n<)d{cmhl&!eP$W(*{;eL0>44
zYP|7Wzqm#A)cg2bY0V5T#Ub?h`9o3EEm##>sE}y=9bZ4P8McQOiCI$gQxc;oHv>a6
zaw`@tPt0NwNMH(d!|*0PW7IMBHx2c}a;Ai`v@rCxw_t=Y5NwhHqrwc=0&k_1%Q*7>
z_7sdvA8i54g*w+~zBrhIZjRx+2L?_!B{O`b46GsakdpLM_nOPUoT9o#))pFwu%Gfw
zeQrUI%>uOI*YNw9BOq#x)n2U9k#c}_t{!L&K$n08?5n+b``1qVB&;Rv!niu*ZQKy1
zh^m{=IER<I<szS;?iwtg>w3w!?JA4x;!b782U4RPieF2S5(cl@VIEhuU<sOQ%fMNt
zSpjsXTFKNQ<-lY`)s<pwv7Z!+V(F4n7CTX7N?kNQ(Zzo|oc!L93muQnxQk2?Uvbh>
z2Fujp+r!TERF@=gA`5XZS{r6B8XG9ePl}$fEC(lo;D0~H!7`*MjyP(zikFFMv47-u
zz+SuRZ7*PRA?jnq@r~5Z^lFYaw*N!bdxyi>bz$EjdS^t5(W8$pjNVJ4a}z@JmPGGe
z#OR|%XY?RJL<<t5Gx{Ki7DN|qwCLiy+|Tp8@B97fa40jgueJAH>pXv_Pi!&&sA9_N
z+DR9q2-{}2MXsC2qYPSY$9b4@xwoNdjc?`Pn{iv+Cx2A9)Z~a`YwxH2yo<*ayj__|
z(7&5|o?;&M30Kbmv$Ej%zT*YnIFDx6*-`hf;vtcYU-UM^;oaCN<g=qeZ>duLriPM7
z5PCiy^^iIJu$tXJB;lO>GkVdgG~IYmTIGMNfWmT|WLY{)lV<<Jt#_+!av1QM`5HAU
zPfM@lb~EXae3bz~K+nvI&oia@rj}F+*I15?*E>ta6!W6Ok~A9M`)Z3~?4^(-g?g98
zBToAw+K0o04;9<g-ISEXZK;?q8+TRGb?8YR4e1=0tpc?JFaI8hxRzt{H)XtML^X9Q
zu@s57;mOh#J)WqKVdA&rdGZ@|{cN@8^GSKwFML{Hqd4Mpr9W_Leiwm=3Bw+E7i{%b
za5@Xae<x#&ofYl$aL(+&{?af)Yz0ozwc1c}Fbg3ZRjVVvN?28dz(Z4q8mdP8>yQIx
z8@50}L;1#=R><e-1zT)~Nifxb&?9-^=H7y?e4G|b|MQ#FQW|S4y(!4c!AWWf4vD2d
z2gNr|pD;*k71|R$w}}wDZ2?jpL`gU{wtlsz3W`#2Z1``sn?+=1b@jSA-e%Xz3$JR}
z=wOq|XS2>Obr(Mn-%V=Is#ml@RecZJ7^W-$PBd_hHn2Fs>wqfN2#sOA%q+^s!nhhX
zq#@mhKhC*+jC1^;f8@}t3E?&9#{3sOH71J{|28!%Rlz3eZx8Hq58^cy!+8^u<p!{?
zXJxT#*<dT#0CV5f-|Q}U1~z26v6A%{f9sa-UD+gp+6qK$!%oXjuE+svdJhTt?=xct
z&iChk$1<9<wj^BK5m_w1bcMsEg;_tke0~0+eM`dYn}(L&6JKDf?SXS>M`zn=*pRWt
z^3>&q5=##+KN9ZMJ<qusfOe9IeQnQ9&2t<O-Y`$gvNfIXph<xUM}O-fQT%Glb!mg*
z+yJkbc0zPcM&R_2ip(&z;Q_Y4vq{z!a-c)pk@1PmJ~lC~4V;{;)u%R3t4}CURx-)~
zt2LiP_Fji-r7+15Fid`wbmiC>o;<nk{=vp^xrstNBl7C2Pn9@b2@%BykM^<Et>?Q%
z<@6xzs8CrB%}!7h&TbD}J9flb)7dw_olGVuy&|n7Q6{o6%7b&y;51vfM8_tosRt25
z<VIUP?h<5AHX90YQ=9JrPdwbW#VdpD<bcXQ6MJF1p63Z}^k-y&%xyH=E*tr;LXWhv
zN}k&!j>k%V=WGOfPCVY8(^thC+*k(mb${%|arI8izc2l68pdinIRrcDe2L|m19`D&
zYRyi03miQN5G}9oDc<9l)E7T=?VfXp<D5NBwpGr3iQb8oeB+>dwxqqhKIEpq;T4AY
zqqB&6$>~+4Nk`1m?EqTE@-q1&QZM=5A?$RCzan2Sn&*&8G+7+nEepY}xo!Q>`U+YU
zFSvoo%bIY&?mGz0J#Glw^mlXiv<NgpK~8*5(mg6xvtvHfHu7&*wp*q<%9v~ZYvu(S
z%44KhsCKy5$P<>E`YEh;<0b0lOBW-Ya_QfjzAv$Pnc>(k^_6@s{k%S5oO|eLwm9=K
z7rBE_!-g}a4A9>;PB4q*OV%tUON{p=Kl8qDsM<Hht0|hPpptC~%_c6vuNx5i7}Cp0
zczMX|X*ea845{2O+643*CmHtwrB8NeLZqL}7`uFTu&F#gZ~T~UQJK2q8|2)yzI7U^
zy7@=>RzIyNW%AU}*i5m)S}X1^O$b7gNPJhl@6%-dy+Gs)rkit|3$DGbG2_sD3Z#;*
z%%@t+fs{hW1I!fgZnb9YmT?xOjuN!Tw4LkTWy(4iKZPO>KsPL>-JM$E9&)vW&tfGl
zPh}Wxew>~mJH9FE5>ilFX}s_B2E(ze3k$=0^qAAw=?H~1)=R5aS)xeelYcSDHwuGS
z{GC#Au77_5<MR{2BxG{S{klInQ42NB*TD$$+{CRZ%z)N8C|>5D%Tbxa`b?f{w5)KK
z=k2d%eQzN&5%fiai060784=WCXI*+q>lV}8f8<Qbx^F^<*sM6jL37f0COGrg_e)x#
zomNf?9xu|i@F6yZe)(N5S#m3+R_s3abui;=;1I+MW!-#rT=J0W{QXvLw5zmy_04${
z%^{?=KjN@DYVRBHAAUDcy0u8_=ew!nU!M0S2ocuF`9!Tzr1QJJK2ZQTH+q!4U1dpx
z6TCH5|9C#L>)Q?`JJMQ{^mu2PuaBy2gm{ZzeO>ZkSp#$G3UL<aOD?hHJZ4M+fj~MC
zUc2g5Ybe&HV*KC7R&V9b_TGb83ykEI%_2%Dc8{nM+o5eVrS1vF81<AAQG3)!LPvD*
zW>)ZQm9iymKw^YenoS*u4EBjC)hC(V-IpN-pfjb7Ojtq5Y-W_zx;<zDK62|nn~WXN
za4{G*Eaw*H&vDhfFYQvMI<E<N{%YR^Z{C9<LLhYt^45L1{t4=3pR*}Qo+nYn9+VhC
zbJ0+PDQ)da`8Eh^-<4g(K*_!@;z`N?s2VP=Ynq|rNj?|HT^!RsQOf=78<N3Ep0Z}6
z|B6+@J6WtK&5@A6RLRpwc~1E;^(aiR%^SXlm8grqm(%V0eYskNLps+R&B$7%;I_s}
z5O<0EhtSNx>NhEkfI<y08q=RVi7++j--1i*xP#%J7hY!^w%$qE#=>y2AW!GtP4#7r
zyrge6o_%d3S^Y|~;-|D+6Z2IPG|W2aQNvARg+_PQq;bUIs1(RxrMBXPclqPiE{M?y
zFf}ftfeFHaj~E5uJQpO$2;Zmb>#eFvxnVlWy6F**E?KtX<TL}3bQOD&n}}bQ?lIS3
z@2JVL3RPwBHllk38Jr?2%6i@zJ9Hdq@`&A>w==@}?E^m4{IrCkNmxSp{gmo+gnI=C
zI^4iR-)$iCC&0L5Eoz1%xKi@ZMa`HvAKLPrJrjZ9t2l=CV6f&%WGN$I=is}i5a~>9
zfYDxLne!W9H=Vrx*4OSOE9!p9K{Ap>b-SiTzm_S+qT(}aM@j9KIX);|P%{ky5w7~O
z+L(GKlztob-^er{YGJy9My-DuzBh3!RIM5?7J4^1p-aRuQR)Tv0ZL>Ki9cK(*@T-e
zN|l%r6N@?{k#drD6?RQpha}RW=<m9{L%67ETS>dQU@x6mzfD)O$@}f*5vpNFnjGs)
zlM#8Gi1Si%LP0c@HYCLfJDXj8qhod0qf=0~|NM=8?h)0{oPk<K9{*DWIdMkAFlR=#
zm5e)}0J~T(t!wXF`ijT8W15k29i?J9OqB-{mavBGx?Zl)b#>O02-tEPya1XgqT4B@
zvHXFNK#yqsnyUB^^Qlq$Vl|V+-t>LvQ@}3?Howk_=6GewBqmsE&Z`L%akcw7AlfC`
zlQmfkYX(|*<ANT!vIV7`!yd!6C()88YcJ+7qC8!K47~=eoLg7xGQ{N72g*(BAKZtc
z42K#X!;R=))AA?iZ#-1pdoYw<Px6VU<#}QWy|m~Lhgpw|M~NWK^d-O(8yb-9t4U!d
zrrB<XV;ek|@Xhc6J$EjxSEDbI4<czzHZohF8R8hr=ymCR^1`0VXW)DHiRNAgv{vI&
z=&Yj`(I3qxu$_7pnYssi)I%W+Pa%d;KYKlUXL~$RyAr+amWvWsL7&@mSL@syO^fj_
zz8Kl?P-RFcHKLi~uLljhA+j&hp8v5{-Ar3>)A^1DSLR8|h#|U;*5HCGhxmI;ng;cV
zCe^KzzUf0Pv7gXOCFzl)Eo2hp<Npyt-YC4`=yhJt`6qwpk*;_x&}n%EQjn6=?)#xp
zm{(=G`ia+akTVnN<?Y?i-KufJa^*N=pUQFrodDyT%}Lg6ez{tFow2w7`A(OKm#JWM
zqe$QSLm{o^mE`uIt^6r-fIUO3%N-&~rj6a&GTkGLvH+M$F3lGPzSJc+Q!7AI?QPtf
zHcnzdk8cLy01sA}p!(dyIGm`2)xwd`6BO8tW`N<KK3yNUT`IGoZh@{Qmn<W`cs%#k
zFlcf1T^L$vGms1GhlV<LgOku=v*S3oCuuTJ2I<m;=H{KX9fwQ<R*O}2j?N$eHmg6v
ziwWgO!afVb>dh!WR%3!^1CI`=rlc+GnA?)i@vS?!fGf}Yo#P$W#Gl>i+RQiC(^ji^
z*)822-NAYvcL0z-&Ts48aQ-V99Am}8(I>vzbB?{raQu${{5=PzA<f5RP?{2-^pYI!
z!h=`TiQls)_A{vQcP0O2czsNS_4J#N6_%IQbFC9NL=-42D9fSxlD+1WW}xXhGb_^m
zHGsF|+-=>P5{wV9!?(S^EAA&aJ5scLR5w!+m4&y_ImQ8_6TXFK4Op+u=A?X3A9k{1
zFtlKN>(@iSQSx6F_VL$sTcqrfj{0(5jJMn@I)%^{mN%laIO>cD9d{;@RZTMynX_va
zq$u7H12gJg5@0X7bv}JM_Sqm2T42<8xA?#Cvsv#q=LKbcRnrfO1lIxcTq&?=3Vj^a
z9<ELw@a96PS8D8sys7i@_xluMh2_xxG^IidmVIhQ!0M0gbO)3C!7}u3dfKprWB_{d
zC9Cv=gvyd=#2p8iZnQ-^;*-$7S7u0Ht04PcSfw51QtP<H=V4%z3~t<(2bT1weY1_G
zio^(DD^<OK5xNkEJRe^RFZTLk`QHIfplYGveKt?dRRyZl$rOwSe3X+@P&l9(-P8QT
zSj1_}QdIcfLI5H;mcA(x*7nnQme6uSmy_}*FzZbmFxDzmDM$A}%?SGeOj)1>2ElaQ
zPRhA(H=O^y59m<~6*EvY9sXU#F;}(_rFLfdGOkG^<xmceSfPxTlFeJn>ui<@R|7Id
z3c7M35BG{$TEOY}Kv24HXZcdArU<-86%K;1C6njBfe+C!WcL*tNyj_?4qQIysP7kj
zQdcU$m>QDBlASpymzu_WF=rfm)JVN6yOuYg1eMFMrrc-t7RZw#Bg4}?@y79FIlN{s
zIFw28A9c1Tl<%ZwynkV<)sB|h@ek3X1koL}QRqdm(8=I$iGCpFjHJoibbH<{@6D-o
z(&g%?9~a!xp(1c%vb82YYnK}Jq~Y;TkyaecBtZAd3@zWayxD%nu={dm$2q(Agl^98
zD6RZ#g?{Dct?a0*y>e2Yw3^U6-y`SrGel#GZ$i*JC(E&~M+0ta#Ihe$-OeyKKg`_r
zcBJ@C5*UmIDeDxaJMENrd$+W*D9F{<0xs`)KP#$gu-x*f|A_Ww3)(YY3Mkl$t~nNY
znKuxx^>>u>(w8%xcHp$uBK~OH?K}Q~^PzdjzE^=)<2!9LS%aoO`p+cm*7w(f7H-rf
zY~!3mD;sW%=32hZS2};K2*xxr?``DcG!(udi0x24xtg`BgkE?k%_N0JpDtuzk+obq
zHjS#t{3z2VQY`K2&DPSnefS*!SMDv%9<)w5mam3Z_rCXdYc`&Fv^W+m()fO_=o3L>
zw8boFrB-5Z5(v^kwjn%TA^K@okC47ejd3D9Jo`f{H?w>G!75?#mL#hxsQHuCaQw?L
z=ds$|FI$4YBUWIH0-qU^l46_a`(UA&msZ0gj{?mo(Mbc4xFWaP<0FG-^URy-^Xy3-
zH6Gwcp5#h|q_2#N6uxA5fK@b<niZc=QmN6fU7PlshlQ9Gk4aN}-x@O&{lG;;9`PC3
z&jCDq-rQ@^iE*X{bgjy~KuwcOT8hbf^D86sj+VQ`%_=nIA~W1#In$J*YTW7Z7E43g
zQTgE$_^&%aAXN4YWKW4rsT3D^Gb*bl{U0gkH+F?db};^Gzf8chu9Wyjy1u~6Priiq
zIQN&Enb#5iV^1w}GvmM>M~U_wawdrptn%Ci6jge~ow1?GF##v0(Ce9ZBVUOIO%9o>
z({G&rEncw%4|KizEg~?${JJFx3iqVLwbwZ5rPImkfAbE>&59_rk_JN1!K<wQZt$Zi
zyHE|79F8s+H&%$y&20O}G;N|5JX){Ac6cD(WM;6F%VD^YzhEa4V?l6fG?X&(x@ie0
zRjuA&mF&cD9DIHfY*n5$(NPbI(g(Hy)Mxe}o%w!FpRx)OJ#oQ^WoW+&$4sGn-<A=l
zum@qcHE>ebe|V<}vG(gLh}eS?+@JHLE)0mOXjV8`ri;*fXP>~Wyx>uj=*Z^qz)h5?
zFr<|6+5gV5p{!#cu~N8OJF^M$#K}AZQdeT=+5yw()N?Hq&sM+3JvIO=W9|x&yIy&k
z+acn}9?HrDI>Cbhq*}1Dh0wR6ZUqaXloR^<tlg{0ivOGe_}X5Ex3-&{#k{unI~1dA
z=~Mf3V(AcmGuHUb_Bn0L0iW^`TUc!wLlwhvPBQfW#{%e(rHc?g1bt<i$7#fMW|<T`
z(jJFCFr|uu>kBp56gn6%s68jirDJP9OmJ&Rs7kPA?iLQDKn>NrzYgwOCcDA$)LWp$
z_qbHiQ4{Cfeta3jramAt`e;;~a_OTe_?|z?Dz-D*kr6;r*QM2JZKIa0YfMB^C>(k7
zUvn}ZUDt@<{^`F``kJz1t#cuqQ6i}HEIQ&9!Yax-4Dd==ScZEr%i{g%_Ub)W>|wD>
zmg)8l{=_>!Ga%6X3mZc+6r>p7Z<*W*7-`FG5_!xD^420__P3mVMUy4@5i+n9u#W4D
z=O5&EWNjwgqn{N|zb^LcwZ_I*S$w_tYz{idhse8NWH7g=xXdqMR{7g48GdxoxAxCb
zFU(~S%WOMu6WdXmExs|8@Azo$mF1MDyNfq}13f@|`a|Y<ux9#;v2Qkq`w;qb9A(H1
z9WLgJH(c0L3EuOaZ<yY4|46+3Mm)FOnGI7V(-oAN`4&yc^#qZYyceJj&37bX+Is4g
zq@<9$>r*kQ{$jm<<6|nu=ahLNk!R)F*yXCS!Oy4hnMd~a%zhd3-f0_P0*5mX9fTf|
z3Delw=kNCDde^}pd$5#!D!eHfHS_3wFk$vAfV0blL7otS{ksov2(!qvKf1))m4|x_
zzSrrynTaAmsAj(;uDLJLxN+kcJUnyHVrDteTDMTkdY9A?#SI@HQ+4PgUrX=^uxN>9
zmp7-spR8o5);>3Q+FJ`J2HYE-VQyv{v%=vB87ukHuyV3rFqt=`@tZ~5ogx;jk$;&3
zY};>6`J_s%<=3MnU7$zAt9;IC4YM*yGW933@PT9FA0JEh-1?K-;9A(?zZj?nUU3QI
zw{{_O@Ljuve{`%fd2JO%a{_?lE3=uG*seBeZ3mBQX+>+*SZc#{Y!Ym~$+uVww#o~%
z6ks@E7s1&@?aPh~c3c~sd($tI!h~2^YIXP)(`SB(3$usWU_Z0LHnB;t*e@W(0f|23
z6fY@RzP1RhRpYHC<B$Y+;7=D&;HOlEp42}Kn+aaGoAss57<gJ|{+EIEip}b2c@N$v
zm(LD^vpv+$WBJLmvy#V-{4XsTVz@Z_#Tgx=0+(p#MDn%UBN@vkb+z<}EzLr6DAIO;
za>bwYyEo}%U7VQb?0Zd}VzA9PUWmM-4(HdtY&Ka_Kr-hQX3G^K2TqhZ>i=o$e46I?
zvfdtKc)H2dt%G9bG=Jn^<s=o=Hj|bd%OiQ$;Nwmr)ck9~X&1Th1h~uuD$WaWgtwF1
zB6KFE+fKjW&_rZ^Fhn`2#(FT=PRcBn!F07)bp0iqm7eZ%Xg$&Qmm9se${|X#p3mUx
zNotsLwbSr$B5GCors{d34(6G6u+wX2Pn~gALR%NqnJSNYN$x4pZ20L*payAK<!63y
zv?I<Z^a49e^}zWo9C+Tn%+cBEOq8-KbZ*o-Tp4m6r>vJx|LgiRrN`kSLvE2;Ab%b3
z;q1&;f9F!e&#~bYR@JvjoEcw4OK4wUhTJqC_=S3sI%bJhwzu=gE*`yVIBckNFdy1G
zb9r*oL!Hb2SuxjmGWv^5ZBNq4%d#pLsqe*K#1md~6Jtze1U{~2JAM25Un60R0y9X6
zJvVb_&|w@LP32v8ZL0FM2Tc4OV=Z3*ciGn20@L-eFBJiN37C1VHcc&;Fqf1de)fb}
z*BMk8&kAA#A_JiWwgy~8G;Gv*dy1PMi61TwsilpAc~=ZUqGX&jv!g?<aP~S+Tw6k7
zi(lDU_pM(ts&#8ioxiP$&9h2TrFoJJ;?&JY2<4YS;27sFC7VR=SjiU~77t7(5l+Zq
z7d5BRL4-b)9g{arKmUeSYF32a{>84C<A)xS^uH#eRe>2ZVpcCfT&-w=3ls@5R<(~8
zsUsmjV~2tT9}h<&QKQz#h5c-?K6?2?Shn_uC$4Q54sN;27t2g*3y3zVcoxi_fc=r?
zMSu>ZMkTu&Kb-^fPSxe)!XD|0HfSRYDD&U+C-eUS@mIA;G>ny8;A~)~$*cM3j0~d*
zW*3e&y3IcdJv7fgl&>WSBE;Z%9s<z6Z69-O+7B<IwRMg$xGuP(C8~eVd<Laop?hK_
zqZ_Bvy>lNd(tA}L%cST#Nz{#E?GXl;0+<sjFDuP4o*a;3H-9Z<&T;+p(#<b?`2!a+
zjD&rWR=)_%BaYe+x_^VZ2X3zn{>ZA+s{airaQ;GX$vH=RG~%TwIS{X%af<v+>wM&G
zVa&T%`JBwl_YBjlYX5Xte{hpFV?!eaaBB$*mnZcYgAVq@F_P=+2<V^UyqM2Rt_-NP
zA%s$}>{w{up83p1>tp=vt|8pk3`xQuP9!^WdMmpb|B}>;?F<t%uKI7?<sLN4zVZ3P
zY=oj_E$}%hxjPxoRqEwS$xH94SN_aBn@0>?pQjv>tlrBN8}oiLC$TFpu83kDx{4U#
zdS5ZBLif7FY^a_~2u+L$<YUn9*S(${XfCKNQDxE|{s`?9=*!APj%OYtvu%CeRwsDe
zv(2-Z0rdyb1sOWT`|e}xJ9P~3!g2TVC23K;+D+pNQK9fvWy_4c%n492>8{0jk)E0t
z%}E=ShT;*U=tQ)Yrh<65VWmm|2;f|V$e(_x8$cX^yavx!xlx8cb2r1k2|IeoZ*Ub{
zXJl?WT^(0fhOo!q6IG2;S+S>c>Qj`u53k*bwdjwbO~Eor!KwocJzH;q10l(_d*ka(
z6?J8|qB4+K*T=1+FPJ7%gEmD~1xTv9aiM_6nB75k=5VfTZOlJy5GW__f9x8QnuS0E
z*6&O9>f&LSgYg3gY%Wl4IAR7~z*$PnRg_h@*Tk*%b-JSn*4D7Zh_4^_T;n8zjGMcJ
zo9}5pQsp-KJ2|UpE=3vzCtVP=<O`fWMX)L4q4L|YTy1fN89pZc$>R6JoR>EH;YS5e
z@J;POe`47fm5K4*R82Sm(ilWJZ(>Aa)rV*Mn}=}B_|P8krFg_RgGn3$kJ8;kU^RQt
zOQvU>8jGn^_HBrMo#~WC$GTK2#YtsG3HMUoJ~_7Mj}BVP@;qRWc}f(3iwgn8(@mfT
z98x*y?)x`=h?nV8BoVkyJiOPdUTG)V+ZtXrCA@HL98HqeiUHc8-vwr>%-s&^OnYV_
zkApW&*0QU*xvqo-W*}j1RxVJUV_aMYoeS>zx&bz~b18K~F}m0@CR{qyJ7b+qx@3K^
zeZXS(br!t+?<MuCm2<f&%Tb);rQZK!0$a4$d2|%@0sG$H)lRQn+gH<idi>?$sX#O1
z@;#|PyK3Q%?)Q|9dw3qbrdC7MIkCj5iXMX4j@C*<JaJbq>!jexO*--^kFCRtL~?cE
z<#7I#J&z;ar%+v%XgiR(&?~k~6P(Q5(ZAAYO(_5K_oA5O^mH?i_m8sfr|R^pFC<2I
zlvQXwe6-X*Lf7Z<Xe71%F5K8;lK%58kaW}&(C_NEaxg{?iO3^T51-kv8Ir%+yk{y|
zWBn5@7xOIvBtnkP+(P-^)~N;O^CA8L8X#7#`edsUuNO96Y_#^vW&VVDvh*#p{H%l_
zlO{Mo)yIp%`>#_l!9fkjXz5KU{T1^k(g^~>qgYvfB&IpBoivmH^2~T$Ht7QA28?FY
zlCxhi-Q6W{o9mO=IQ03pP;EelFn15_En?$KwT@QNg|XYE2$VhBQ;~pF=*W<S(2a84
zyl8BL1RC3n+q9B48vU)9#Gn53JkdB{XRV3%Mtw?H^@U5=)M<6%X$5G*K56egGk?&C
zai29;5$P{RLc#@J-{f7Bm`%KppvHpAOD)({V?CF<c{#{Sn*K~Q{LK^Q^Sllg9XJ*Q
zyV=jGJB1YJ^*IFxznJ&gN*vHEAAPV&KNvHTrhsj_^Z;>9Sf#Wy^8}zHHCL<N{#gw@
ziJw$NIQ5OYtn$H>tt~1l_eNV9PZPj*3haJOMP-<EguKP`;*T)QtS<59c2aDKmyiNU
zpDFa2;BYi1pkVlCEIXMEw>s2@hSZ|f`c33Lv0m5{DG*$c4#RK$q5T=;k&s!mS`>0s
zp+h#_Vb!W|)emka){`bH)FJ&)6{r3p?x7=$70q?KhD%1m8JOo@Tqr#9aJKHO^ygYq
zCvV?smDP}#1h+r8V7)%~eKeWBk0V3huFszr|F%>LWw8+P6sIqvFG@%qeC^LICM7)V
z(d<f0n-toD`_WD%9<GjhLs}N#+&Ud6^ZTl_<1bx4_IpRsY!)kY7sb*asdSWlUJVU!
zk;m5R7mQ;`d%YZV-g8);Nf<3XXswBS1P$+r^(?7svX5&WOQ)p|U$u@g)ZbMZ@DDe2
z<#zMXvHtm{OPt9N7KGcUc=QIFSv?4m<i4gp)@>RqQ%F%DWUb_St0f5??HUeJr;_`K
z@1P)(O4mn$V{6vdEv*C#iXN59Cq$7JEyftDXX^_@hb)U?0*Eetr3SOBh~`3#9gY~{
zgqro_;PgIRFpZ;F%E;^)XUF5AmdCv)cv{joM<lKM65XSIdAx#|0r<D}0SAc{aGXK@
zu?Oc!ki0)Wr3(G$;x&R*qroBR&iby$Lf0O7$e3xP*-8DxiU5THogRw8{^9W~u4%G8
z7Ce&TWJAjUeSylxXAMRwCchZh7^ML9Fldb$>>AMYfp}5Wfp<fa%pk0xbS-5-VK@t3
zKtHH-)h*ocF@TZ<TK2Sp_kLthv>(oo2Ej?vdgBYGTl;kP<zTsw)9w(L>F2LMrH7JV
znSHrwd~eznKo{k)J<Qw9%?I~5SFSEe7~!NA?B_%((|OxCzS@quEDVjcNJ`0ZBAM-x
zgu=cXvy}JZInR+INMEuiCk3YTbB(gzkn4P$ypK`QkU3MKRU}k1d*}0%E2%-AjzGAr
zQ%<g6$DQQr_fY(0jeEPR9DUe<j=rvG$-|3{kk~%av}{K*WvA<Epv`P|OfI;c;g}$-
z|C7uyY3hF{3px$#yoXgQNYac|E9q5nPo~R2@csH#%A=%X0?P8|s)e?QH(H5?OqZL^
zL{Uw~pm%rxdw-K`tse<vO+6$#a?ge#1|;FxbV(w3Yr!N{0RtIB$DW#%PP_1+n5cfA
z%PC-JC3MQ*P5#M_96B#j6^a}Z{Vw2RleLeZ(di-dLtP&w)AI1jV@&myhRcrE&In9k
z#&Hq?eYw0VP7%9F5fCm#f){2cGPd(HFdz(!NC^WPY*jz!S&moN!1Ivq@U32_l5ysq
zqo?Px!AHYXk%W%n$brt~=7ZVaK}_gXAT9&Y_EcEt&Tfaqj_&mOa;^U|fvyr5ra8_b
zDMom8X(uXuiy6d?826Q!QC{>rOe;tsQ;Hh^WBDWm_S%?Vy;)~!6xafd`lwr-u3B6$
z6%Egfb_%zybuPz?UwhXv&tqMh6n~F`Veot@5FNJVwy1Yog>@gdMJcN#vJxY<_P3VC
zOAcLWIS}|jw3MPeD&FO-YTvU3`5EAwJI8j!gON%4Ek@e)>zIGYx}&;Pz$aPok~&-c
z?KokE)idN|e4{)@ZcYM8sHJ?URkPD#N68F+{WW?w`*1S5Xe7HMCHL9ugFMNzi!U>u
z!q56@_FG0EC-bN8B>xUK=LX41^NY2LTBA(4_fP0R$UI4iR6N<57SHlDwWF+;p$$~j
z1S*=}4A@(t$0B|X5O?v^LDLha+ktQEWmfJEV?G<y%T=tWJKbhZX6+Wi&4dAVH$H=}
z)Vv%4r0(dQt7jGhb&N1}R1<K#@9UWOAXN4W5MV+P`_4kUkVoI@)w;7;dt?F0A{tq+
z08Gnvk8Qd%1D*32emF^r#z}d}c{m6OAR9<^@_ZB$rEFRQ2dzbaAb1`iNOeN;p1)UV
z7ng>7<a*7LM?aJ&sZV3^O#VNXnlyGRv=|y(^<8etq4Yd-5Ptg?TYSss#P7r_AhvNH
zv7RR>4n+%xcbdL`vE(}8FyWT{b|$-|MRfB|ndVsgcxH~!)|JHNKZeHAASgJbLh>!t
ztHQmk#&6^Omb2*-N3_LnA3#_!4+Oa}-jEGAXkPy929HnJ&IpyZ*XwNvw6rf55?RtI
z&r~@(4oWTEX5ag*b<e#l?c_?Eb!9lbcl<UN6ers2J+a`7N*3)L%TC_bV%_V*Jy|$8
zKFKUw^DILi!s(@9!m7>5{4Oh6PCjMt|J<~Wh#jp`n&hl>SWbh9@+7$>y~qak)H8ux
z7xl#v%8_i_JZ5>1Fp$(Wv^&iO4SpAV>zcyS{fR-YQV!HO*&ukIcHLA+QH*i3N8;@`
zpq6`xc{Q_fVXf5zAX{Ea4x+uR25oNDL$VDkMc2zk%k!hWDno@!BI-F3WaRTbaufE3
zStm@z2*O^T3ahaRIyfDUmP-D*^b0@ajI9&pWGh%J2(&iu^Fw7LbvHED`c$qR>W6a|
z$w5wxwwm$Ddj+aq$w#()hDuk$>5>glv#z1T{9Mr_(Q+0?QQ+m|2Uy6lKe(OJPvkZI
zB`IXzB6>fIc{KMp(X9mdyI$*aMT~Y;q?V|XjSoBXdFP8p^L*94(j#<>#eC;7`m_xX
z;JGfb{Fa+C#18%LG*fJCtxFpLWwP&OOIoe9bQf}uI+~jLS${@2=K4;<%e1FId;_Gf
z)cY=K9F9*58npUfHBv{10l~|`Y)x6qj~CaQM)XJIF%1ci>>hhH{P9d*I&2Ty6)i0w
z7O}c!nBVB%oWCd69b~5y{BnLE#9#M!ynk4V<Z6$!gQn7XP!`qrVWK|sB<U#Cp4{Fo
zcdFsbb{EO@G7(Q%uVJqZ@YeCmr%(Dsty0^sTLYp$_|3Uo4q7}CRg?7MKEud`P>enZ
zXdG~8lUL*{4U<sK7FEZ(MMeS%mzT`(F~5@}@O_$qfPu@u0^n{@Tl7$9)xB<{90WZ7
zo9k==;Llv&j3|!Ft0Scw93Nfso*(W5Ah_T8{1}lsQCsX&Ltih%N#C;&&TG};unoE$
zt$^LLvFPudMYMYV>z8(NTBR~0WjdRyCVJ>dTlsD0!gV%we?cm~GLy3p^~4@Nw3WZ~
z$}MENU88FfCQ~d&n|w2}afXpNRC0+i#-m|JCDZDtA<g~>hnZtf|3!??t}QYW;f7ZF
znyE}1j)-=SHc!qRV>Fo*a${HA*bJN@N+zUFD(2&rVm8CKRdr?vxJ+=Dx6|=f-HJ@^
z3{0s3vFH{uZ;J`*^y-LBojikgn{nh)thpG>aL+Nu=VRrP!9PSBe`GchS!^PdK(lEF
z&T_5!0Zq-tMwmG)f%8$m+hEK!{X5meV+@n8w71|Vs!{AfQq=EH`31()PJb9iMLnCe
zagIeEcuK?u>HGC3%?u<@2c|cR2WnBdP+^QT`u+Qjwx<2@>cyqz=-24ySQ1~SkG0r@
zDZ7eCqsx>lk(-CT$)*0!AkER!MXb7{cH52Mt)FAFHp(0I+>wogVEkNf;nr7NaYx@@
z$p`XXa(Rc;GB!_19=h3I3Y|oVW3{edjk|=vkpKK=Uva%kl0s4Ye7W&y@H3jLtnktE
z4&}OW8h<P`o0Lz`LJ-JyDMuy8v;}{Op%X01R924(g$r|ue@fo==j{42mv9J8{0ncJ
z$>1k1|8fnH-tWr~cfo$aCH}q`M=%$>QzLI3r1W`V$SxJpTAgypB-QI`SU#GrU3Czs
zdXhgGW>I|GtJVws<_sX_mjkz`YDq=*)xr{VDW|QpB7!4a__j*AB@Z2e!+Wr{OOnjp
ziCLq2*ow4Uf$Jq8PS#}*q-G!>$a@4Q6N9l@ziA*)LKo}$D>IUEM71Bhs|I|ii?wcb
zl7`7b)CV@U8g(@@7FGxH*O&rTUQfCqVfRQ_CxJa4Q9P9OAEB)gk(wA4bTNBbw2c;n
zd?=C@_C2pL+!LQe^E2Lgjl$PADW8nk(H?+SKT*Q{oqU=UR8}atX`Q-t<*g1Eof^Mq
zoW#c94ApemzrbYxYXQwdIO^<mUa$pmfC=1(`8<D=F@~LttG9~YBeUgh->To5*do^l
zKgVH-W>?c!8i-WJa*<D@<(_bVUMR2pG2$4v3yY8=%pLES^q}VD6Ii`*Y+Je<onwy0
zBkFh1{f-nlrtn-jM5uTgez^je`UuQD0755m0|NcRVRPb54zRwSI2+PtQjs9GkGe<(
z(D+9R?&=4p;vXr=>T?lzT&uNil>UO!!Q?Kjs>7+U?Iq6EuZ_E`rX1fo7QwriH>u-Y
z5zk24lY)h2aXJCgV4O|9?Rjb#;vf2wUz5{1DVF}JZ<+m>^vY0#Sc$@9vB<;4Wj$&x
zAYW$FnA!s_N=X#(xt~i1xfYV}_{Tw92;9ItDJ?{PCVKUgD%Hs-2l^xQZPz_jc6-OF
zsIEmeV)y1}Cdbv(vgLmZ$?i%tnOim!x<1|(q?5@y&h#sb7L^*XXXSoEv<VyK-7{*&
z3s8D8*rJ+?{~9M0C-|tx>s9@O+_b5~i%A|eM!4}L^B%b;0nz>K4B^LU&sq;Qgyo<5
zyK0<}+N63#Vo5B|9~UR_k3D(Ez35uPs83XQd~gO;7n(f*46EmZ(T^ile?ZkvJ5c-x
zo`6EeC_Y4eY-4Y+%K6c+Ev*oi5)U_Y$wD8`yR*g*pNjIL@Llk49>)IVm7z6{xP0%n
zYh+VQPqO6}ASaCd9S8${6H40+z40^-U6Uj^a_wj6n_!CU@ey{0rA<diE=k?M7)P;D
z9#lRs5LU<o!cvw$aR%v!TK-&mH6p&hT066t;6A|2`!DMDDjJK(UgSm!-3i?|>w-R9
zGW?%)NH0pE9Ix^`C*n6$W2s;h-c3i89sieRvums=I`~+6)#YSLHPL(7M2>#6FiB09
zvRovpDbyXTLS|>9lCsKtv;`yxa~?D%{J9N_8LlTNef@MGd8=6INfg|kn|C0Kgfgz5
z<0<{ScgZ4UY=%l~%KEV)b{fAel!@p|I=!cFuR<-_zuW?0u4nY8b5|s1$b#t|#Fx{D
z`GER+XZGr%3X{2HaE?y0f@3Iy1((n8HX_|obV@B(*_lUIieH;8?EhKV45C}|@CAnP
z!aQKR5FBMncc=VK<DpKu9fj-a?w^^NBX~sJrddqHkQP?33d`V`VQ3{igoSQR8if7E
zcPRxgzGNZbQfle!0zK?z87$#|)(zVTdbH$7d6d3~d?B*{wIIqf9a=lf7dgmdk)c@2
zEVyW!1)slwXKGtNUBo8}4lBAj`fg>^(<SS374Ue$0NU?nxEIP^ZLu_RBel+LdL(UI
zUx>_6N6DBpQIVa#LVPHbpS=0@wxvj>+ey9QMOFim&vps?v#aEw|F*0212SJ*o$-WP
zK&$hjeAQ?{%E|B6J-=jcg>K*DQfh~fU}(HZ2=Dmkri9i;KC*vAbF9kmaePY4IOpJX
zVwt;86~v*05&3=|?WB^e3zs>h5*sb_ZEwk#{;`nQ=uUz4@{b?-Re)6XnctBt2_?H1
z!-kHKqg*#;<vLU|c*~I@wCQs18-s1wm*IU~Eo@t3sD@7s%Y@F_SoOE8^K|)#H8mZ@
z3+~vKVmY|@!58bSU9<_FtObJI-Uc){SsKUcEPI~CqqHR(kpeWrrr=j(T4OosgOFdz
zw+?t<H;Wi6#dm_LQysLWt^x$lRt3&@n2AL>j*tA08wWqw4YQG1vd|w`nK5nR=XWle
zyiQyirAmH-^Jpyp-dlE<O6kFei>IMbMvaqa7NaKU`fNv#yxws)^QdxYFw<c3i0FMq
zoeOdCQM3H)o^IiMdz{2$E$;jB$tkWa*O}A$WceeXIMM5meHzoUJ*ni^S=wWsXk;=^
z_DdAsbbjCb`nThj@Gm#R*&#9t_w1-Nz0P=(+P<x_OA&|Ho_m%gLwx#KWW@T7^T3dm
zeDSl-j_niT`gycseWo|VU|Vi^<D=d@aix$BQY?<M&h#4#*sEOrRdsH|8%3m%<nPxb
zh~EX1(jFpxC6+0E8^Jhj+!;k*!_xX@sqnq{HjE%F!kh132y<1THD7YA*>dldKD@QM
zTHx)o`Yl#aeq(ssuq@KiELH8HeGJujq3`%f30NJK%{f;ze_&9tpKI^$a%89YFX#d`
zo}<jS(-aC=MjKL!P7;C*qi>~d$o}E%^UL*>SIV5ODFv5~RR$tLw!<5x^$Hn9PbJPO
zH-fe-8$c!^#$)PMPv%a(bnbJ>%~1r{TyM6J+@uFR3FH3L=>@~2!$}W{NHf;U+sW2V
z=gg~@L(5*I%gc4X6{F`02pz9}j_aT9QX{h@BX{f^;YR;lwtLVk&-nsv5)%HV_R}@l
zV6%EqDgQrV{qJw}7VGkJCic{FS#*vf3)6f%5g+n!H#bghAXtPN3fNRzkFhqmO@x@;
zXz-T}1YDzEx0>T63WuRq@Z(#MEK}etTVZu8#ZL`5OTpTIye_acF5VgixN;M_<w007
z4#|C_XNr#$x)d$8JCrGx_-g;@CZ-(@xp3`218vL<f3ETVp#z=KF_}e#t#leN{%Gne
z?o~uGEvrU+-L~@box@h+@E3-i3hU7&{<4PD>_)j$c;E{D!#er!RpIW!-~5P!2wKnD
zr-@UK+U_4nv;97hjSrApWoj=XU6=l?H}rj5YDL|-%a5`rF1TqWh=9$pr=0GZyS?cj
z40+~TQ>4~a&zpWd>?P3(!5p8Gf@c$q{L5Ago(D?8NJ;`X08u~1_GebG@6@=f(;@*B
zA|lj7?t|-RFNt55Bc1GC;~J9ITV<9glK*X9Co*Co%h20)7qj6%{oCKDiEM5`9<k}v
z0*wl~24E0C&)EW5o+PBJR)f-{qT~E^F^VKume?V^CbNukZHN`A?%eP<?{hP4t${)j
znU92I+yRg}FL<}(yMOsocUH;7vzKnyOhXw<@3M8r<PNC~17&&ULG#~i!yXHqaS4T=
zvo5gw=M%eoQiaF}VD?hO=m&=pZhc=#D~)@j<xYwcF=uLo8ih>Z?qTUy9wr5qv?+d?
zz-N5-6L&vsbq|}hm~TZj9jn?Yj2Fkg7aw5YozkA1YsHMm)qss)rPXXS!*LWV)U2U=
zckzrgtR%0p0|{2*YJQo+zVxuE`0#(`A&)x$NKm9txi3GYoZvzbcoZxPF_uVtjDZQ&
z5(w)ZFIR_>r5IBhZ_T6V)j*zdw&5*Hk6E1|U2hZ>>E{mPdixvCK-SGn0xRnoY!--J
z+OR(gF?*ge=1Y|dtrJXAgshx{7sQGkZ<;1THMn|ABO006q<+6o^+d87FbG-bB;saR
z_i}DZRK=bRxHa<*_bIcCZ~B|i4gRr|rCX_Bvk2wgk}r9%VxPxmvB(R|MqAh{VqWDY
z+_6wDzBST%AF2=_j{V(%Smreo<pvX&#utK7Iv2xOS{%_KAZoj)aL<S8+xTeZ@2D8A
zCS26d6cV2y#ax1LHOe1O#HcUE;U3B2-EI}E;9b8bpHZ=)grv&~{#yMPB=LWho4?(1
ztcyZa^lx<kE~QY+H0?3_?38FV{CC?43H7$s(f8ZM^!iT^{Qp;xjK$^fP5f6)lLwG4
z4i8GfL7HXO3IZ&DN~WARia!sQ6H1n5ekO?PZ)8eZ#Gr?gZW<5Yz6z~|{h15hoMWWh
zM#{_huGK}oX@2p~1$6Z#@<VC>=?Rii=$$o!L^Db9Q%$1;yyenT>Zh*{a75qq?q0FF
z?f;9>KsJl*2HfiHL_Bheu)G+2MznGI347^v7nQZ^XVx?~pC`b->qh?(p9_q&*dcLO
zGXA@q;P<hK@cR@wvQ-HiSUQq4mp58}MW4GTm?5tgFalkkFXH)RAN}7Y%M!j%L5F^8
z<P}%K4}K)SzT(@g1@XVYV?_g!zB$B{T)|f|YeN(D-(JYe@W^w(H(5tO9L#Sys7xK%
z3{zb<4v!ph-xJ@(?J@Qs^{z^E7lKefux9=C$%&$tVTFEh_Nl){TbHUOrSUi76&tJ;
zqhHcw@jE>en#)$rPW)X;zU`{l#sRP7P&Q|PVcQG|I0^+#Aa1l>E?$ZSghR1=0>8b2
zp~Mrm;r0@+h~*zO1Vev&0r-a0*4b>$4-5B@ru%F*J&cf=CSpkbggPL|R^g6B6F|ze
zbB$p}9c)!t53C=wx*T1r)c`Qk4e@9_`@)Vd)<&NTxT_g%kU9HKg!nD`WIZ|XX9Q{=
zlU^w#hi^qR&I9GK(xW&hsc4`52WNqGr_YC`ldsj){_h$7C`N5(l;{?{3jT0>{Xx3?
zgY@Z{4RXg0SrH!AVHSN1W)vz0KHJx>y#Q|3A{YqW96s)lJ)L1+8u{rCShIP$9yl}c
z4X%#pPrHEZgRL3A5V#reZ$7qm5okZ}uCJ!;Qrz%P6P0lwfqs?)v|UG<;&62${2)H?
zHIO%$KEDnGIGBm2gqm&dCAbCK=G&e~)8lU^-h5g)ndxCKba!8w`pS{<ztV!B1a5VF
zM_qXG-pBR{O-@ZSHu1r5uL#^W6WH&YVBO~8VYe|a`LJM0CLm=fqI5ZcP~+Dv$g1`Y
zbJ1XEBajHXDe;7o)D@r#^#acP>svuz-T$-HfK~UH@F0&o-kP9fP+mnerf_i7xd8C(
zHu6OBO{2b;R9J@H{9z~<f|?A2Kk5m%-r;1>Kk)8+ZRFfNtd>F@Z;6jxA!SCiQj|+V
z<mAs9RGagM-b&IW=iblg=vr{1+*htIRYrSZfc;&sCgvNEQ$c$dcR@q;j%o87eT_J_
zn}vAbbjxD;S?=)#5M)I?u_BG%POhiZgF6*$^@K7$f<U)|?IDgB0ezQ`t;66A<Y^Nx
zemjT-?>*xk^;NR94)I=91^|+pHmY@FNKB5|mi8dvj5FW5Na=M+fIoHhWjUHD@B@?`
z(i(ET$}ITY)Q9WyU504=2ZBzs=105iu=j0uhH6^Asv0FqSB`l1`_qE)X&=9aS*H_+
znoRLt;m;d~*fp5iD3a}@u1WkPIjj9(@*GTZ_CBO}teRrS_=*Bdk;8`WO~obtIpvEF
z?NKsUzE#W`+j}#`Sfym}%!Gq?exC0$^4+{sb19l_-2Ko5R=IC3(QsuEZLkHdP8DT_
zg+@dI={Yp;FUvG73g=NXbq|3l24J>7Gv~D6td4R$@q#=}-)nN7Shi{OsFInM!81D6
z_iJZZf3pBUW&eTAZ@*o#>>6>}KZ^+lD2KT+i;+Qnkhd@GJ!%+MRbG&--%LugsiVSR
zyA69HdgOq0HH7KxBCkpE0NnSACgkRhtt|T2%SH0pU3QVDbZKZVzm{)_oJOUNWOKV&
z5&F1fA5pwNvs8IloqAZo@Tc=<DcEo(!&c{yB+;>_IYuq#(~1?`ib+HM<MYf#z(HD<
zl`Oev>X<jR{iAwVG&ai-Vzare$$O%TV+Y`Sd}vZq<f|eY1EYisKG)f`Paz9dKcU<<
zx{;91`m_MWbjjl^z|kv`l#A3*k*o2Q81swCsEKl~K^(su1qQ2v<-IbWzB=#0&evMd
z*YLWG<B~fxYhi*oes!nRYNf@Wii~6X>d$7(rL`2OkSa^@=~=)WkC-Pp7U>xqDFQAD
zUWm~AEF6d8nepGG|Es>hBVn7Skfo)2r&5e#_V*-PO{hnSw?W>Y8wgvWv-f+Gl6e+|
zHcL}bt-C#DzC3LtW$yK2R^a@!H#ku`ETK*s0Ji0c^ALq=+jk)FE%S#(&>nTX0Z(WQ
zukP3M|321t)n0)go0VFl7P@%erfzqK2P2|X&5y`PIc^oC*)V|6WJDi{(<~S?4{NdN
z^%r#aIuI2Z6Q!6J>_Aa|6YhjRH~l}$1#e0(Jz>pxvTO{EG-1=umkAhq-^FW9bg0)R
zrxMNcbF9=vOS^U}L+LtGX(mmHBJ*@H2t(af9C~<|CrLwKXwvnBby3rrk~7l2l^VN(
zn(Yhq+d=C1i{AjmMVr-N_@`3C55asL@*btYZvg1YAg0(eB<ex^H6z<k_D${q%NTCM
z7+gb$Z&^)psuFkhF}S!inDl~_)l$IAWPZqnh33WariuCRI@$OWA}njUGr-=|W-kep
z_6$=C0Q$_+qb-@|PM9MJYv#U9q_9B*yFnHmjD_NH4|U>~j^vvA;!SyI96;-Y@q9cq
z?xOdhdQGg@8rH)6k;F}*9z8a}M3f3die|3BOAye`5#x#cz<$Rm{tdj!kv&*l4_g3l
zm=_N_(X#a=5VSO19@YspXDS0yX1F#>kr}~}9l|MUkKvp{?xy)l@MI<KWGCHav~(Vj
zY$q8sdB<VSRgy{WqH<h$j=kil{7=r60@6K4sXsZ%5WnH(E)?eu0P+>lz??PSr1T*N
zc!PH$<WO$L3uFhSN#7-uRkAb+#Fge6Fcb}({?QKxu7gYo%nVHgw4H+<D48Vp-4FGP
zn7ZS*DIuXwo&bm0(6!0=qV-K3%n6biAsk(sEsk#PrlSy;XYlYXDPVT*z~WIUkwn*N
z-{oQ7gnll}V+JVK2@Ne@ClWl9ndk1XZTSM<wu-OKv|`wf>k~JvkZ<z(R8WHy_@#IB
zxtlmWKg1TTb;5mC2lOiU=v)m^M&IJW^OSjP<W@8*3_bc$g#J<?hRbU<AS#;svHX>i
zLVlliHQy!?d<mXA`M8~UqI_uV(<Mur&Y=EAiP`+O(WGx)alnCE5Y?j7S1ILVfPO(&
zGNhL%Ikm|>`i|nXR5$E$V%`45QdIApH-L=W4D|p~{>mwA*Tn!IQ-(}5%V*%^WSRfE
z;VIG;xKFK+4}A~NQM$;s6pe}`kQ(g;zPm&%c)RwM(Cctp6fVg<>Vm3z<@6Y{DRYBx
z418RJqTU|^L6nf0oBqS`bQ0NoeVWvF;o&-m#R^1+Hu>x@jvbjK8?qmtIs%PMj?m-6
z)xu{l00|@LrP2N4j2f_+Y}(n*@TgP#@}NvF@uPrH>Wny1H<VWX101yQrbC;$rR!r1
zMlIxk(GkE8YFdju61H?C8#v3P$lUlkV)^$d)0_*w6HIRV*D%U8;l@7Ll}j8eG6K}{
zm%sHtcLW9O)i?L#JB~S67T@e;&bv>ic0E*TUmd|eUm3{E+vp%~4W<+Ziq8UjN0Qcw
zY1?jQxZoAUfwZ(e<E;7rEgKzr1T{1L#Wp<Gul>lq3}9>ohZR^@4~XpvKzKwk#a*oq
zG>^!<Uo@d>^`fo#A!_#4-Fh(SaIdMAEoEZw3z{-hb9<)5EeF8tk0$nt@tVWM(X$Tc
zh%{=Y-rqUG=8X4W&X6GfNMR0BO$!b(p&li!*jstH?*S9M+as@jM$sikdSkDbRCg=s
z7QCi_v#>$#+Z5JVy`xp;k&kbWsWNf4^LUq;O~FIOT=YWn+~Pj|5v)IPiR+v+9M`zV
zVU09}qg|51<V5Aoj}Zl??zd*&0KhirEgemx>6ZF=EO(4ZZA~I~qGuzKbuAb1`wtSp
zxn$Tsn^>P`{THs_cGDEG1#+0)IQ(JN^Il6P+e2n+1HP0wr&<*VwZQHmWCQml4i?Ov
z3o(R}6D<8MPY`1O@ZAbFZ`?m-scKv(U#~Wm*2tw*1Fj-uGI~YYh8$-g20`Ih-SGXX
zsnt*9xg|>9afmo@E#Xw#vNR-F{3iGG9PQU8s_zH~ey?Ueh4sPJGeKn&(UIcEGhe7`
z8lUDaU2F~gR4yRp(~NC=n*VY>hSu2BK3#^a>j8R7uc<&3y#OG`rnt@a+2_cTi1<f+
zR)O1cT0o)8_x6AcVP8?T11he!T9=PjQ2<SG$pESW_CdvpkT1-z?Gh1Jm=eVLr*&YS
z*qc%$vG7(ArSV8BwQghdMppGJzhc|4;yj%Kp2w~C;HhA5+xGqAh^rkm&o=y~^&v)z
z{`2-EKA7^L+q34)SZ2V}&q^joTFY#5NtH-3Lp&TSq(X<xCIF#y;rqL<mTR?3KIC~}
z&$N*?yhM2@vZ;yyP&AeH>=bLI7xYnEf>o>bhi=(t;f*o-6b~iv2Z!oqkf!#m?8i{x
zQF%OEPEmCn?D$vt*E)NC=_iYCFB5}46i`Zw{2-<ACu7TSCiB^F9Oj^0?kpk+9<?IF
z_X6N3f=^j~2`V24e`WX+?nH%-;k?V9+t?HdDY%e6<~~wx`DHphW0^<n?X@d7wB@@(
zr3Zz2>@2?ffA8P__wMxXHL3&3)xpL|9O{fu1i5@{#sK-O?nGH#JD9}Z1o_9>qdL90
zVx9CcxeY5Bu@k`?lXI|+Rc!*UF3qF~wTfPpJB~F7yT21#hJu5&V$DK@p~kkrHpYMe
z>_%flMsudHS?Q5EU2A3XphsW)Zv_^somcA^=l{pvTZcvcZtdb<5oHjC8EMG@QIMgf
zLmCAE1%aVUKtvFvr33~TNl`itO6hKdp_G>HlrBkW&+~D=`~0@<@A2K|{Q0}yzvRNX
zm`^;<de*wvecubca#@7}tbd{!>?{m@oxNUi*2pf>1>JBT5qW4_CtgI@fJE7l!XNA|
znX%zt?`7M6!|_UmjgeL{Zkwd#wlEy_g}5w?l2@1TSWy7=EgJV}6J5#S$o+!V@jANu
zgw38B4v(JAPRcCd+azSh+ZjU*`>|O2oM;$maoUhNTh_h5T{xmUjFi^w=JX>^Y2A}G
z$n?1X+Addj-{ed38IJ}Ew4?PLKU!q_?y$hDx45IMM=jY&yGP|E!$ePm*?S>{uF3A(
z6)MXS?}R_Pr%h;<e=$I)GI(r%xzHMyy|(O-@PJ$^(R5pRSjbV9W}aG-?hYjk7N2bT
zHC{lrytPY=-tI!eP?WcerB(YI-5s9MC)elS=_W5r+{g$Ju@4k6^(t+yteDYAcjWEe
zFn_&&Gu={a-S_Hwo#NvH667d7s@IwndnSLa+5MHLQhDq<o{uIrjWKeD);s6flgWK&
z!?*e@+%lJM5zse$E)3Z46JsoycRRJu{Nn1I;G;@%s*xBOVv3gh-2IA*Exx77gKly&
zs~opEVYC@aCUbWv<baWQDX@aku#IR^HSvyTl40LEo|$Kb_-^z@MPdv-8L>*+)6(95
z$pQcEM{k~6&-@fen@om?*=egwC<Lw`W7q)nqX(~Dy24~J8f?1J#k~tN(moedi7p*}
z@Ghju;B0$1zC8BXOdrf!fEq#tXM&&M>g+U)&9(lu1pEEFxS2_Xfd5aS(yh@q)6#uE
z5hm2d+Dix~3qA9h6SeTI!S^gE9%IWjVB7cE=7ybx$3`QiMXQ17)#y1{JZ6R3jfY>+
z@h`!_<Gmp-IW_5$$+N}&FNndv{TL$!n-s7Cj*`nWDm#EUt=N;MtOV=<rJOn)!3>J)
zdMh1_jU`kUU(^a+{ssosT}}ieHZC~lmYFnmPlAhxg$5iJ0r%Z>Yw$;CqaGbuP?!E3
zE?&QL87DQ`qM9he5831{0^8kI&eUeFTO8i_{_@~LA0$*M{(KlX7T4Pe=cIn_;D5=I
zfBVrp{0g_n!L|cvE2iK0vELbz(rOkIYaYi&boS%XELS>YZ)`VC2R-a}Qj@@L>?WAS
zX(E##Pyb=24>-Q1vyXns06;&OJbMe|>5I(LhG1^&|4!74aOc=l!2h{MJh3RaNaczb
zs>R$nO2z`-L9iAC%UHF?KK_aC{uBHAJ@%ZJoCHQrO2F#wqfXTQC~b2v)LdB+3;S7M
z$iMoFLXF|+!W(~do-h}fXmK+R7~98@9J=De7$;L~GT!kr3MG4;?Dv5DTc7?3yEmZ>
zct*H;#b=-t;7&YPKEtq(-@fy=NAz!R_b0IF31g+Dz2Ru40^Q4h^N@f4KrtaJZB|GZ
z>?hdyfA{4x0pOB1)V1dB{b|Mj{U!bX_jbLhBWiD5O^g5|)WY>DfO+PXRe+uGJ-Efu
z@51s8n$oMBC{nDtI{>QZqKzh+@523`1Hf!*9(~|)8lp=g{Po3k;4vyFF)&sRETVeA
zu=7`eZ$I5#tvOt?1pExgcZuJ=_t2}{INm-xaPbpDfi~O#2wbBXECyFyEVa05_Dn>;
z%(!mg5*fX00-Dt~AxDApx?B{uLFYokdxLlU?xsN}TZQTX6$^I-h>xM1Uh2_*TxdpT
znSe<+Ns8NA`PFJ*l*pj@9^zR8oyXiZ7<eO`UC?q38sJ3Ve{3+CTno*ph6oEMSK~q<
zgW*Lm_sS&>$lBvO_VN`cGYyX)Rsc56r2reC$Jz#7p1gEWp<|l3rUxu!wFkF>YcHR2
zyZ{ICj!V0hL64=w@ftLkr>Bm9At4DsSYQ;*F&mH+nppzlwI%070)K^P5>Aue1Ze>J
z`MBQ%>U%M9;2ym=6~LeZP1^a!L=&JD7VYF`YYR}9cH9Io>EyEp-(LPPMCR*h^L-F+
zQ~?*V5J+dMd&_KIDGS`fZe&gc+&D6Rn;kWE;jxR%Fz^9u3A170P9t;>>CDp2fSzOr
zc#=y6+n$8M{7?wd35j9&^A2p{IEAtv1GWm<G1n8Mk9Kms+TW%@ivM=-q8BNF&9<%0
z_hAZrn~KFW6onfY`b1*^-IRsW1296zc4iRFLXL*N%;PV*`V4)_*pOOa(8#xFIj1ts
zG*mcEtaoE~F+EuY+m$DBli?|dq&;B5#;~1xyWq*;#E%Z2Iv`bNUIhb+7a?O=9cb_o
z*VYy{AtROS*L`3k{Z*8!Y|_U83|znB!WY`UU|dzij<Qx${Yt6i3+DN_9su`s@p_F;
z;5QKtSv{e~6;S)b12qC*qLrs5@KIh??<etCkRd$9A~px2T@iTCX^%a(;yrRUeDRyE
zd8^FFI=1o|G>NthEep%(M~!oN=R@ZhIuo(Q{CjZv<@BsAV*w|O=w(zsQgPw(BOq)Y
z(s_xiQ(4%lCRly~EabJ~Bs`wZ;J=LpqhGmKL8J~b1vc_labm7dOsuFx0OQ>gSc@+N
z+UDKfSt;!7=x3<ppheC60QQ*i92c(9qMkcAA7(%HU|$6&E~&f_x|>FI_2FV_bcg9=
zFe|M)7#GzI&cG<YN$A|6AoK#Zxi`^dS!b7mENY=TapmEm^5duKPnUHkbgCOX2!plp
z9Kr1p2_!XD{_k}{4y_YK<rtFxW}S$8hk2}+5@wo%kDF`kO0NhAC=DO#?f97GQtmgu
zntbz}fAE+L>|PJ0D82&QyhmTkT2E>TBwpC+9|8B!qU1~yR6G+v_m81BJ=PmNIT}jz
zkLwXmN_tnFq~2?`czltH9f|;vaoh~QJnY}Xfj2Qs3^2PFfrITWfLl7_B3akUZ4vw6
zB$P0&wt;kbeC^lO33y1{TfxBdX#JokePF$vIP_^~rGj-MJb@wV3UmL;ya>!Clu*OR
zU`2PX8QWPO)RB1RnRzZnoOX_dtmq3;C3wGm(5k*hgg5wP4J^ppA3<lN06W?K*wcR4
z^KezSB6amT39uHuQ6S`91pQ|kWOQiR62a!aLJ3XWP6em(j;H~Bh7N5NF!bf?xpBOx
zT7G*pps@-#pqE!wtwuT70xQ}3(IL<@ZVGuuDp1gA6coRAvVN<R8*fu#S%Om#`#J^e
z@^3-sxQZ7TuB;54WmQMEo0py&V8?VK$%C9ATw|se%8<$hiH|0TbG)W>{k6?wS%?g5
zleI6DT}t)(>9qi2P{-<1ki~WSP#sKMR6^aYuF4O4;J<ro@<3T&D8(+6Zd7;zgzFGf
z;41%A$FWa?CYfM(5i<)Q@OaP7SIPG?jlO0+Qhn?;GV-LN+Td&MIr=C=&;1p(g#1@k
zll>6|!2P-Zbj6SdSgYNT?W&$Xls^q+^V-~HYXuXXW1v#62slXZ$~3#Uob1nk&wut=
z4lUw#I5II_b^^^qqi?1mFbSy20XwEo#x=P@27hPLm*mQ5(X`E>W3HRNNrJX%2)|Bf
z@iEilw6_Yz7s$289*vaoSyY)FnI=F(!=C8#_Sb+H(y^m2F<}(Un*H!i5zT7g!&-9I
zPWGhOqFck4R-XU|LB-%^4a()!4<T;PeFZjv@h;a!+l=Qh80QMZNey;!+1|RX4<(_`
z^CN4w;8A(dkS<bF`^v7*Wrx5sUp!sEsT_ddvm&S)>dkFhir>e@f}FuniA@L7k!)9G
zW?Y;o&_K6$e&{w6@AOI%g+>uTNmt^$Tl?C?^6uU4EFHt9*=^8~NJ#cA8=s?AqWh8t
zjU~rl*7|tR=b@-Wr`l>qrqs|z_}q6Niu+MfIuG9;1$)lWObpJS%#kF5?sTY#V~fjQ
zs?uYw=CR2trDLEXr#sBtr(YF5<GLs>{K4#rA;?ofXnSx+Q29@rU6K;UNM?*AFHJ(F
zF`bVCar`3Dz^usvaFR}t*#E5Rge<PPOAFG8MG(ssQNz`<9FIRwGdJ*}f|=fexq1?>
zPQC@)_<F{GA{t@ZCULkjR6#Ob`Y=P67;_JM3UTX&gcs`k4{*{bH`BX4qR@OPNa6mO
z0Po!B^$wun4|&x4`Lsz67!q-r$OpaP30B8Q;2|^+vk1(766m%bmlDEO%3)lAHNOr;
z9tz-(!s}QJ+=|YjOUe=q$$L;!yg}{V3LRZ5{)kh<(nK5)>lSvT`FL-cmeL~!gAX}t
z_NMi6fe*0J0kEBpn$ZR)_m<BQ7(*DFH;3t31S4y0*j_xmfB*cN1C!Ho3Ki5-EqD+m
zK<K3&l7#)I7l57_bf1r~S}j6|*B(pi$t}`6S*Bl9?DF>khErH|`FlEZs1Vcf(fp0h
z>GKD|*3mTqF#j^B&w_ekNr9Obh}!*@>vhbIJa+`*&jdC)5jfHaeZRZ=egvc}H@?UR
zT=Z{!=~Ou3d1!{9kS3MAgUwxV17$EFSJtzgVT}z?J;cgU!AuqnL6hRK`7H+5X3_C*
zcx`yI<`gjTH=n}_MZ3M{J3!4`z`Krlrq?t25&mRqI9B9c4nm;kM!tIq$mr%%%95=<
z4S-JBt3&n;W+UJdTr=>A*pzzps}B+9rz3cv`q?J>CePt!kR?PDgFFU7zDKMjgti!6
z)f|~~ZQN7-;d5ih|5NaZocJmb=(*CSsj5@L(iU2KJ*X9t%85>Lhx=egLbM^C<fOun
z>2j5iDeb7jHr{<P=_aRBC)me!Ju1Qv2*gPCgC87$u`R7WYNN%S*qp#jJyyZhAEFcc
zDr;fE>=@t|&6o(%t#T4*L|32g!AcLO2(>d5ZMGLxy+lgshT%+gAGoRVX{-bHJ~NvV
zaK4I{5uD?1w{tq6wmXER-`A~&i^Mt@OD_LtOMhUB-GPq-=X_U>ml-1ph%HY5-gPFx
z4Uz$BuYH}sv`M8&tVtSwZ(AcNzD-wggJFvf&F5?)CMOhjGx5M8OLj$hygb^K&*PGn
zXIuQDTl*Wpb7}&cy;8Unwbip&LNlYEnE$cuI~wp#-qD)5jw%{m2L%qT@<AHMpOT9V
zslUS8!9;?{GrnjwZ6oh91+@Us#OKCp!I_Qn&jg{y&Z8fK{;20GN;KjJqrfom6)zxX
zkmdgB{gDgU=R<jD3)A@q58(c@W*QP8*2h6W;hsld&&1Q;#ww_34$VZdKwNN@Igb%q
z1QGdDRO%p82dhk?7~>P%Lu#TytddbE`i+K>9o?y$mCtNn$k{cRV<8jTW4+u44=`?J
zUd|4~-fPLetd6~i-Zn=hJeACQ_SeIM?M3Z`vy@pVXxzlmIZ~63heqCsPaE+rnn+P0
zcX1ENHOtVC6wufQgkOe?1L!oXJ=)euv{w2py)yz{sfe%<b-@SKw2JD?dGK&A0r7)_
zhzTxt=f*l{+!nm{E_&~*X9<P9PmcZz5xk5}ug6{y&%=`$Us{*2-W-L6Ww6UqXe=u?
zgeh6e-oij)6)2b2mW?Dizxh}QrL8vs==I&uU=>t^d#BrY&x7c%B>AMSiDa7mi}BSI
z#7s`|E<R`kkq6q+tV$X#KLHKHDcnmZ$HO;|IK7NP`@XxuyUX}<01)ZhhYiCGcQ5o(
z_D{mkFl{&RA_tI1dZ519Ozjn<xQ}=W(#@QGs{J=4rs5^%X=T1vLKx>15+~JCD<tpH
z9m-j`Qt(JTd;=aneG@Zvj$k>sa_qs#x0CMZnutTV2Z2&K_Oz>Jmmu{rTy-nhG&p??
zL9RxL-EO^e&Qio$D|n!YaCEfiO5Uivms!<oeI{H0hb3<gchvO&hNDS>@?<_6=0Rzp
zn>7fzx}^wpmCsQG`8x8#77kO4?(m1kb}!~Bj!#X|nEe2OETQJ0FSn!^rDHK_M*EuH
z)zugbZ@7g{&eoZJ_joIKKK6DZ&#v)tp0I<dK@Y9P)#T3vtBEJiUc3}OZ+ewUx1mRH
z)K-Rm8J-rcDctI?w0pOSw_&I4R*cfiv(0cV=3w&`mKTD<{5Y36u}8uR;<&gi*ftx{
z%;iq%1L~P8TbJ+?s<x#jribbWLyB7-X7pSEB+p(;!TDG-Pm!^Hy7il-9fO_6re-R4
zOKBq-8NGR=7|#m({<Gup_u1l1gk%{Py9u(wa82funjYxF$f4xPj5pFI_HAfk;o6}$
z^xP=&B`2`FquX;c#^Ud>ezWdk6rY3gLeZF*H+K4t>AJHpJ2@56(SCB{{Y+T69)f5;
zB*kQ-7%R8HM6pEE+>&VXbqWmfd4RA%wz39PUr@RuhjkA`zvlX6#g8eDXKJ9A!|DXW
z^X{q6dbE()SOw+9sbk`s&K@rLArn}oZSOTqyCQtvF1>H^gx%+@hYL|4&#{`^t+oqI
z%;UX>p(=W|;X;I*WXCam!QK~@eHzQWntdMmloj_x{iDRKO2SjRQgm7}=%?3kAtngv
z4yJ~RGshK|q_KBKfyuHQEFcK3v2DdSTU?Y0*NI@zwUN1h4+AfsE!QLnsDA>^A`a<K
zcN2iT<jnaG-uXzq6AEMM53ps8SZ(`}TmwUk<O-wPDPg#5S6Ud6yd2t`U3#HMfI8;Y
z$2qrh?+s4;7&(~r$atp<-P`(v0Jvju6(`0dRttT5**VD`InIXTu#hsew&5!ZhjURP
z>Di_CbTPKYhP2fpa9n&FrM%1q$L28+Yf(K%O5_O&w`7XIt@xCnFBApj;bNQhhL!Dq
zu;*rv<VfGs?MeoAO9@O_>5*RqN|5f?q-)WbuJ^E@{Ntvt@V_XH&9Y%u7&&b5L9y~0
z$@j+n{njmJS@n=Ii*33jKV%h$#9kc<1@alY^r-eXT3S{I(@~BQSs1~{74o-zOqdMo
z&giR4ZB{KpB4hqS;-x>}yp-8Z><3({q2&sdchbWyT2#SbHj^jaMaFSXH^zA0@=NVw
zG8ko@Hp5QGn1(JoOQ0^y2~DRyXTl#pZfb`=Bj3PNM6Au((-87JP8O>8a^(Y}|9wxz
zR@*}fks+=Tr^I>I#eV9;=9YxFHo_uP>*5y|&YX(qKl`WL>>0lKUwAMzxTBv^3CdOy
z^=$AIj{=u88}0c_VUM>YR$Rhkd1O8FE@cVqxR(aWrr8p`YzpO6t<$#XGs~qyqt~i6
zn?>4X#?I(d3?ev-b*Qc;_i{O2%Go_Iqe0w7zh%csk3E~cvSvGw(5T!lP0hsN=WAS-
z1Iv^??w<O~<WfVx499te70uE|Fiivxt=G{;NCvLr9tDg^1*T*YG6Py3u+rF&&RG?7
z^EIJ>jQ9Gli5GogDxF~3c~o>CxlK`^Lk0?A)U{skqplQgPc1lIXX_n1;+*T}i&Y(d
zUJL)k*iWk{DR$FfQ9ZR*0Oeqg7?k|Z{Fgj8S=1?A0}4&G!Ud93M)s0qG7~bALU&vM
z$`Y8D73pVwT7RJ3J`qNcx#-Nh)^9mtY+MK<@|8n-XPLk2l$e&n#K|Ctow{@2reC7r
z9sXuj5|&yw7~vO`9>}3@vcp-x9wKWuy&%2r<XsX7k9>0b@)YAS?<gsyaW>EQc-!?c
z;=400f=y#;dVR`(f$FOK?pghy)TfqKSep_gLv0e$3SmNZ%|@U9(EQ=h@ql$B0&r^8
zj15LOUkuo4>0zwCdqy+T-fyGJkZ7@Bn;r)ENuDb0*aOl&*m9f3atB}L{=)r_k46%M
zomabQItFh<&5^awQmY8Zt?v}#%fEV-6KLvm|Nj66|D!+#z;PLO!wxFP*HntXvZ_9k
z$!6Hr+Z7DizAqSDmc8o4*q$d6jVmyAdav8WK(sZ~Zi6QMoFmpc0|4sJYZP`o{j?Pt
zp&%0$u2cpof>&Zc+BC#7$2h(x+jC8xUIGl|G86PGd$xLc@`DgQ{jpw4l--mXiI!Vg
zx^+$uoOG@rz=*ZdkYnAlSR9~!1-h+yRlVyvqvSJBNC!fpS~hc}&`ou8_~pO`ng5JI
zsZF`bDelUbjr{x`K5)(H0wp=U7u}DiXx<i*`IN7m-F)p!{P1!R`^{15X=!r7cc;(j
z*-gK7Xyqt<1mr5Kfn|1Hw;9C0zK}Vq_`%L51C?7o7e}Vq-Mv4boIDmQ$n;>?n{utq
zj*`}-ip(T3jeW@rrH>M*B*|`~jZ#xb%eQeRWj42ZG!5mMMCm7~JD2&e)RL4omoa++
z@&E{H#Yo%Foc!+%M-{a6(E&n2rSW-_kz04go2$zrh?QK+e-O|4P&pT+_#NKc99V{%
z#H?JQj7|tvmm&Na)@Yp5iz)%dNuctoO|+rji2v;P%FBv!i87EPP8;Zv^f+g(W$4bb
zWPM=KQ=Kso#_WYC**+$=*J~uc*ID+tX;sWd=c-TU^hG#lkc?2&4_i8hDCJi5<C3vU
z&Oq>IE7KVxI%r?iQI%$RD!`So!GR5BoCZ%c7aT9EC4OV21IGFlv`*wX#6IvU5}ixR
zmQIvVbZ?WBMkjZOZMAKK@orTb7eh?HHC*6d`)DU>aCWp}W>?f8_!DPU>rA_Dj<)+o
zqdNMOGP1iPy|VAh2g|<y9l#hAN9b3C6)O@$k4-IA?xxcw=6$Cp+ztd7yg@j>!JmSd
zWEf!{Ht;np`~seYEISb@PtpKGoJpx&c5$swC$Oy|yx0ctEVV<Q@K7N8B&VOvZ921)
zI8r(g=o*nYKV?H*Ezasu1qBxyDrR=KOL0_<!@f*J#rO=XK41{CWl?FUrm6|;0nq7s
zQ2#-~(MJ*)a<QTc=j&k?G*<ev#ni51>ESPtyfQ_nzuUfwAKyy)o{|TT!JV<B@CCCQ
z9NF|F1YXL!(B#}I+$+_-JY1Gh;Je3~70xnkN$gw9o};k$g=oqMt=(o42Kj;#sl{(;
z9=n5aU1<Ogd~^>4GQs+nYD!ZKju$D;sF(RF8!PkrWj3MSdW+(+^LFFpJZATFIdr#9
zC~Lm?k@d5xqc;?%E|uGfpghbsu1wd0No%ViGqc3ju)6iHpI^1gSeL@i8iLtXD0BJ`
zFLJH#%Lwu4O&NhuBw|eyB_>*5G59R?-48$hMjc$8)9N`XCY<c152MxVXV{>5k1|eG
zjqnYCBPTYdrAn6T(;63Z`yDPp#k30R6r-3?(aFJbdXI`|HeHf?mF~>iodyjVk5DI(
zT<W~n^+KQBuZ`}d0S+C%A9=4>hd5{!x(25xbACM9bR8vDs^Boq?ooK2M@CG1{VE$N
zXY=bhN-a1sFINRHFl|v9u|dfBTP^#;KI&S(jWbFV$rkO*ntS)^_)XEO(~N$sL1Mla
zs-0~q;@$lzO*~Vd+6Qr>*RB7wcAnvjFvB&o1M}`{W&v77BbX4VHwtjL#@_eph3q4F
zIj{8No2Ct%l&Ja_MU?Kps8rNyG`7p@j5O9prPBmudxQosW+Z1$+eE)}_#yD3wOJxI
zkYub)MxXMDZ~1opstJmlJbURns?p3jScdHoa4Di^EN9;qI%v=aZMSD!O0J*Qa5@|i
z>M|!7@GAR+`mrY#XPmL-SQ)9HM><1~h%U4M<2cPjE1TgbWJnW~wEJBQnyZr91B`_^
z1cA(S^kw&B|BJXybp&2YR5WpaPMIYb<iimvvXl|^y;MZ#UQ|T101Z>0=9g{ZU8t%x
z3_6@Jz;riGORin3S5tE`9JzL!CmpE|Q@>g!-CWI^^T-YvYGrhwf^K-|=Ebiv+;Ftj
zIg#FxVZRi9a4y_-XS3<#EX|`A`fTd{Z%&Hsg~zc)N}~048l%~Er79!D%OOMAi$*gS
zExY_j%1d*4a&RIxp)`Uxne0w;%6>3PMicRNM#>7Il~Z;-`DRX_5^Gzu6UZAkO#uwq
zH1?+G`5OrZ(b$0^4x<oKvD_F7-VQ&<)|HBwdCHuHMWuwt<V)Lkr`h$%15&?ge)%p;
z!8E~S@ODwqbNWM!h0UT`O{a1sNm<=&InfrcL#A@XMt-(>I>4y`!=&IwM2UEpebL(B
z4b3?+$JQ@|ZtLNaxiBNop}%kA{qwiALd@`}hXq~p{xmk9-ezeIkx<5IzvxoD-@kM)
zMZXM~og=dg_jyX<T{9|Qiw@!)O5*DB1Y(v&^(KkYk=Xm6EXrtCO3OMz#>j~M55F?$
z<p`YM@aCSmF7JGP6F628NrEwhg@Q>G4EK%Y$}V%_F6krIETQJ`dsu0e9NP4nroLm}
zTkWcb`Mk+{jOQL+Wn$mj2e>9r5J`QArE5lKloL4x@gZ%drzG0HHXrb~F;o%!D7eO}
zmC{a^9WdL5f9p~ZPO5=<>Y!i{41^5y-uw>0DyN$`DcJ|0TVzi`!W(rFzRk39DT9QW
zn`qhljH5(9&KBbfj)vu;)m-;kcI`*v@b$c-nB@#DT%D~);<9}JozsHg+4IGo2t20o
zb6QQO$8kck!v5E2@1Ktou+~9DT_*^37E7T0V4J5qW5bqX657v6YuA8~xj5;}n`;?8
zXB!q!{E;bZu%zmOrtvMopb^AKv0Kf+;=9Xw=LqU?R&`q``bM-D*sB`>p>bsWBE}bN
zzF8&8*~W-@(6MXtz_{L_uIRhcvN!lkE}V{cRAKr99jo5Up>TDjS^=+`>9uX1*8Pw_
zvKP8!@TTyJV?YmmIPh$}wK*b(dkb#?O+oBbNt=jja^fRFztnw{-zY-ytQSGdWUd2%
zz98GjB9qf1L5HvfFjI6OilZnVJe<0?5k?boS*@w@3bQz7N81!X->EAlXmI=m*}9?I
zzym)+u?)j@!3!8aoG-<O%(T<y<za5F08jZx5`BBO^*~e8Ec7GYM0UyW2bZ4s@!kS`
z6P9!N_|r88FaiXNno5?WUuLU(O}sOhnTA1mrevSL+-6ntSSWcJMi2i~3(ME8lstOu
z+N$e+nf5ZV3npm61fVQjld@h?V$aKp^kLDR<?W9^7f!A)vSrSV9;Dtr%L>=dD+mq#
z+{aCd_sd`%xG~!aVA!xjoij>fisnKQGbs)OD1R#^_=O9QGEDK7NZ!}I0IS6Ld7SET
z>SxIhd@kvlfs-1Z1Zzw)3!v@Vqrn-DVgYZ?5wMvGHtA!k^W?LJu3>@>#55TP=}`{6
zxzwnRH%Su!g$@La)U4UUcfm7|ddjR3XOjulNrpgQnsr{VT8>@m^OR4S(d~v0jJTsY
z03vN1LXAtrni)}2dta^nww2+b4!tfSKrm<rs4_n+{(sp;?iMXe_5a(7KvUEgFF<D~
z5N9PlNNF&+!GuSl1xz;HSH%QT!Ak3}#1-VW0YNNu6ir^xAVlr^9Qx!xrbxUNm`!S{
z=k@H{`xMA}L!L0ct;HksLjQ$SEqj6riyWErn|>)LxOu0^Z}0ijy6$mxRdEga4qd&t
zB*fZ7JF};=ky^XL0W4CR&nrJWS1;?keiZ(Q_K2hCPiAF6A$T>GBfyNfABWnKI6RTH
z(GN@)QMw+-%Z6N<0O_BmTX#`tZFH8sZ&sZIig6AOWxEUgGkL)RcPg2@z#DY#;Bu8u
z&}>&__)C0GQs--KmRAXu5^-BIjDJAAw6Fk(F9cCh8riB)<5PPjfIJ1f%S^L^uMA4w
zSh|d@53fTHn)JVdX<ErM3XG$p<}EkNDUI)HGyrFuu{B^Ro!A)Ci&#6O`@G-io>_k=
zOSA5*w>#5t${3HqWJSgKqV%5xc`d7)a&sFlz`W7@c%p7Fu}C49T{e-HQ0AN)8sDH@
z5)EcnvIp_ADfO)iN?*)!Q1gep$nTzQxWBC=dll=1D22xI#%i%g^&Qm*)&vgj0bsw~
zvA5&leOHE+3@iLphJ&tvK#($*bPClvqLXGWjb36NUXf#DXg|)R^;-Y<<3wWBPsE1!
zXPC1-=mG=D?x9LYpEQCQo@$VnXd@Vrtm(e)Hqq)~ywj)EY07IVu$mSLq=1z^L~Di!
z0S=J6G=hm=>Wq6?XNT!`f;07psiZ$VoRmA21>>!Olpf{cQS)rf_IM43Shac3FMP!R
zYFB?XV0LGAH_j^ObEj2H$mBT*W?|3h{RIczU5l2_45q9(qph8%wePMpm9=g~%`<qs
zut0iONM^N&8H1J4LGcX^=}<nrS*uFBq?PC~%ra{fY*X^!^Pb<N)8^OD`*Km;PafJo
z1%;zsBo4T7u>&8fDyn+-K-Y7c2A}IeuO%)ywH>E3{hD@E!Eo*jQAzJ<pFx0TmJCAZ
zZ&~8}4lo#R=J1df9F{FzD(Z1HPsVfHdjl|N!KRaZ?(_wDVUet}k6%d+5Vpw%Z~P(z
z`x}-HcPI?A^O{WaZZ{-lFhN}R`-X&ud+*z$R#M?PDXL&FHXA->U$5-3($qPgMiWad
zMQWxzvbXQZf}GhJU~HrB)1X4K*`w|ZWr{6%w6xJH+(n&f9b5#Px*C>B9Xq3jHc&#M
z6~|bcPZ7|*kWPu|VXUzwiN`74J13<np?N~odTa4pJqgzhfs(OLGSu4G)9UDGC1g>{
z&c&cnHhd>G%<28GgE&B9Qb)(%vdYSrchx;tr+n^>{5h$M@(jE^=&veME!>9n`Ec>C
zLelP0xTk98kh14a89oKW5C^bVF)D1b57=a7EO&=2Q~}k(A?~`R+BxG}PeJvkj!sdU
z;uLI@c0Cua3CLnI2E)o_wMs~rPdk^QdNzxZD(#kb$oWqM!RbI_Al=(vqVLCe)i3o8
z?G4M9k2iVMkg_Ih6oY)7yzLO3#;=U=`}0jSOi&F=pavUhk<lOX_pA8HQQx)w$x-i|
z+pCvWvpL;s4I{s3ovH!LbYPRUTs&CLRW+vL7`O$)JTW9eD4yd{IQQkwFt9GN|M-}k
zU?YIPb@a*t0n$bpsi}y(qlolZM8+s0GurF|UbFbgrIF#iqt(DhJ@HDj){tS3N=RS-
zGLVUs6@#Xm+|GlAKRV}q&xQz@ax(&nx|wP_5-)u|TDmHxIW#XLZp2QMvu}@JB;62l
zUSw;KUIYzKk8_AyK1l2MjaPR{qTzHFB@xV7a%u2w{<kDjwrlpr)jWP!Zl0Wr&R96<
zF%W;28)7s%$?zuI@>&WIcg?P%!cyViX9WmJmf{6O?ijfv4bu~zD4wHyQ=a^a@_=vE
z`{>TcVE&?@ygU4N9Ox2>!k?V;)=xgs2>n>Z(^SoyM<Uwvj@Q9zG`pwdGf|sss2_~&
zc$UlSvo+>KYK+h33l~+TOXwG^51o-V#q661&hRaJZ;np;CBCfiIVW<TDKcED*a?{P
zeXw5+=Q?Yv%gjyBr}FJ6)oa5Ywro}I(Va#;PxU>P0As}c#ov`8AR`l@;27X}k*QJs
z80TA};bDF+m%-}L%!@M>4)yx&z3!i9;qLFcve=w0J!Ti};3^S7-qA)qUgWy;(n&^~
zx1>8vS6pGa!%8Re{VUTV#2iO3-)!M43ziw7W|_+g(o_jRP9QxPCoMF_6<gtPeUv2~
zzCG(ozZH|$Hnc!lWHu)vb`_G<V8>jVH2DfZ4`J4`kv5{HLoaBa!T!>L0~%iix!JZw
zGbaJAK;TQZNb0czZ|5V8v?<!<ewTree6ON;GTUW&VgLFGE0!)L!1-{yHBk0PexCT{
zXJ}4T{{1hf-_ldywDB#xpZ?S%Xmb*ExA<)Vs=#HUCl%o=6tngE+RWGRkuWdV)FnPY
zv%;z{m!FjXLo63gM0s4{j=xL5ol;ukIuyiB7tZlM@so=qv-z%Lpe9ISmXb%T^>Psu
zsO@+xvTU$M_>+PKl`+OP=|{IbBBwT6xl|!BKx>$5P~MBiCwLo7k|yCbm*6r;rY!%w
zCsDEfjdd6!d<24M29^r9XS#VlI?@B5dsNi`uqpOX=BzU^`0%KW&9@@W@N_0kX_&S@
zsF4v;==GLpB%07Ui}{*}l+YQz28q*tc@5F%E(Q0lAUG#`Vb>`hlDCPxjF-HLZ~-zT
z;zg;GK34DD@RdP6$C-y#_99`wL%0DyCh{g>dgp}9(oVIVh&l5TD{kZi$BH3!r0yZ2
zI-VH7?=W7LWtBGamdFw@3%^;pm)VR!FpobVju1o<rm^RgF%hvKth>12YhM`P-HbKt
z&BPd$ufA4GXB&{l)_kQCyqpcjMXPFzLm#Rc9pr>G&td|Cje{$(85uOP*LzPN?*I;;
z#W}5s@|?2tSTkx>M<snGTR)QN=r@6tY@i?mRWeEwL-&?*lho!7-$10M4jV*|B}AB)
zo}Nv~G2%u)FxNMj2!V=0&MhfwDqUl$wdAcqn8L|})7$F!+3mPGXd_7pe>&ZHs|;gS
z)qqx^nI7_nb7!5*N+=~;vP{YhlJ=k3D#Z6DHzk@@z}yKwF1m|(v@MP|b9VYf^Hl!+
z7X^mI<_k@qAkIyQg?r|5^KicAaXIk%IP7$v<KvJs*Bc12i{|m!Cv?1+V{3ooeon(y
zDcRSJge%u?C{*&~ent!&HIHIS<VP%4<S(Ol5l!SRzO0G+dX@PVHVgozuG*_bA4+1|
zKc9AeljK6ttspe8CTwty071tz{xwW!p}wtg!D3WZZ1o=-goPoojc+8<Dx<klc-+x+
zg`r(o?sFE+Eb9zl!hX37q_Ni*<4gY3U3z`x6#ie>GW1u4uYNH9MqODGd<uouXFKnm
zf<Dah%ZwEm1z^+8!-y15sqD}O<)=y#8()%!;q7PYq`g-s#t0r=pg@1+%&so3C{dx0
zo2eHeV<Kw_{wvULRZa4fr~T^_IK@4gNV;a-NYZ?CAD*V`4WiawaYuZ6lxMh2nXGlM
zx~sLLQc|Y0DE{eoqD2(FsddKr>|t^{Mjz=aZx=#4b$hYd<LY7DGZ8ajUf~Ic?ObCl
z0di3`m5j90k7dbajAf4P#g7(E8x)ddE^8LAX@-2PP)&^RA?i*Ur($On-DwlT%hVKm
zAX7B^1~cb0;!d(y>nlVpNTKuKO#%TXaO!|JNVqIO>`Tp1|AmOm`}$3CAKwRZ=oUw7
zI?8)6wJ`mF6|!>eJaZImDtdUsy0PH)#|n-^I@1PQS<kQ3NkxR4GLt8~s;5Jo`?f`g
z(l31Z=zlRzTWfu>sZ2NBT5Ag@ahtHuiE=HokVT71E6ll2e^z#ep%ZWTwB?ui1?JwX
zJ1zEweJ+OMtx;|dNWq%ZSbG!JGx1r}gngL9+d?{X{;mY|4;Oz%&v$gpaAt-XmZGK3
zzABR5E13g*bk2zMx))E)#NN<nknCo59?#7`aHM@dm+tD7=c>H5Q-9o6=s!<n9q@!(
z^-V}!*oL1wUv9>3kfvqt;yaBgMlQpMHyVZ<o&1k8adk_=P5E^P_jaD{idvJ7<@8K-
z@Wr?CjCPLEuUGKws{4P5QNK6jm`U3wG!|WZlN9Jmx*L>#(MJDG|FUfoKd&I*3{wLh
z`Z!Z(IKD2(W&_adsj+{(B}2_K?S9Z1O7sqVaQ8A`fEJMw6JPCF#rW1gueXB2dRqqI
z(RF0-Tb2JeOzK)z#2&zb6CfS$O6Ue)&30%$muF+Y0<rjw2J!PgJZ+|PP5}f<9{neu
zSr#FtqUmrR^Csw1y`})dv*K^Ag%JH;3L-X;g;V}bD)~=X<{wRp^8vL*mOTZjbr3JI
zdWLSDsw3Or=eIQw(z>nWF9OB>!<3+SL%p*z{!gGLG~jWL<c;1GWO||iIK$Xlz?+K&
zLgb?DPt0K{pz>y^{$9P_o>OcWblVH+dJ+Uv1~mh0)(VhC(z}6=W%4?tz}pfBb9P-I
zHgZAh!YSsB6zH-EB>^Q0KjVH7>z@c2pz9+4vUYp1r*OgijUR;~V6U{^yke~by*7X4
zzgjcU6sUSoKF3{Aj*A#S=r1+j*aGYsbkY-WW5IuUgSteN4wUSRKr`9#9!YgMZ_F%O
zzkq1N{dN1tx2)Zu9C@_{xQGoBZ-P%xO#-Ul;!(;4Q4%0MA`JqN`g4tmD9!(j)_x5i
z3=J>eUU(NK5L&gQk*f%CgPaO;_R75*=?H_VyrT`CvO#R*=ah)JdWcNM0TW<&d=lK)
zWs&SrvTqv)xO3frM>w=zMgT3I+p-{CxN4HEYyM|o_HPziOdh0n8T?Q>`8VtP&kKbc
z0oGFjw%8{2k6Fs!{aqs*q}(#j9QXNya_b*!@%L+}m4!k-Q$HZe|NRI4&%gC1R01mr
zll_#4^S@b1|Nfvz!=ZPKUXIfK&))IB{DmthNc>r{#8LLA!|ESO3K8aSVEd+l9vOQ(
zZ!0QMZc_i3=U1BnJ-82h%6fmg3~qNkKpT530^Iu4?vm}#0mZC15JX$<AzT$Y34ii(
zKV;@8l8A5LIRJo1yq_k-Rn>9$BG&`j83fs-S_ZBx$8&-oZ-F^rR!;#;Q*e_gzaQf1
zMWIce7uB7mNr8~m`C0B$g!c%6)W+);Acku_t*04(p(4I>Z^Vc^^zdIVoa!n>8h#bX
z8~?Ar{k>D_{jwHZY9bgyJ-OYhu9^z*a2E?&Ih7!c+jV%A3TX>PI$vTD)-kUi$CiM~
zWC_t^@iOO=y(A#)MhPOb_RLFsFE1(g095!Mh@!RpJ^9Cxg$|IIZ%shz&u#EQ(I6Pq
zi9|6yqWR34B=qBGs|O;_Q5d*P-EH%`_1&iy?DhDc-BjetOxTU2Ie?%UTsBSk<@zN_
zz!q3R>b?8pTHN;e>n#tyFRaH240zmbH9X<X1|UrJ%@!fi%g^rE)?X034N-%K!RGGP
z9p8+)zFxIu&K}NH^-Y@Mgrm4u4UD1Cd>D@dKuQGQREKQ?{-2;V@bDPIO+G4Yk1AuB
zngVe%hJ&~Ra5=4KsIp1XI(nc_*7Nylazdl;yGW9r)eGM1p5H4iqxGFWOQVNpKqMTF
z#{_HqvWym}!AcdEv1<6`CV$&d-s2bd{z-se;h0x3X^vz!se^cw9)Nj#8w@h^VsOqW
zi=}UC{01afF*apW{wqsV(RtXjnZ~SkpfdY>47@GUAfbys=$S400wNR?shK4K-<b0z
zV3b*xT*tont$`4H1ERqjCdQinDaJ`&0l{HK00ZO=oVvq@5dW?pz=ff_fsHjl4aR)D
zVsT@6;p9uzBA|6zL1;5Mai}@VHuq{~PQ^87_LXIRBZU$G0VmvIUf%sRI{jyew<(^+
z)Fdd1k|E}p19-9NUchK%u{hVm8e(eqKuBJ;55L7Mkfd1Fgw8-Go|B($^&B`<6@iXN
zb*5`6GdF3*iuS?q{>a2hlx=p9)p&Z+(`%77E{Kna9}uV1;OSzj`vB0Ny!!Omp0OHC
zs1*BYbpQa>n_KOIhEK<q4WfT?+xY=M;0}>T`-bCYxe7h}PPiA|4&;iJ2C)zq0fE3H
zISW<riOI6CT|fykV@aNI&**Q3%7N_jo^yQ3i_Ubn!I{~+?&A7dMYu)CHX!(0c-dO@
z_i?>kB~-qwiZkf--&VbG?0{q8$%$x{+@JckHxrf)3bO0Ey6z(*<C;k}j|{is22z0K
zOcjf7lbrrAGC5lB$t_NTeO&Pt)cJ%I(km(zk+aQU69^b<kYt8pX9UI^dN5SmbMcQm
ztP^I?-YN8pG-Jdmju5x;0p!_Yk@+e`xt3#JL1Hs}8^DyQ02ma7lme;{eXwxkVU}pg
zHInbf)Zc(EplyIpzKenSs%^CLKG^2QoHS~$^{YT~^XZtXbn#bCkI@kVgxQm&HriiT
zz;FA4*IP{p%Suv5jh6)W=Jj^agwQVY%OJArEUnqehD2H4-;@j(xS1h(p$Zx0egLtE
z=Z_#&QVxhd?kiw}-acIa=moU^4)xSM85sg-pWJ*3C^H3c!nz6YUK8ZZu)x0vbR;ew
z0+rJ%D78mD9R?ib){#OXqHsEC4iH*aS?j^ZONL*D2--tW8KjF`=VH_C3<aIxK7H-=
z!2Bsr0tC}^N<()Zq*%Q^58mfFG3G9ypYV)m9n|wLFc1nwuY;`Z7N~bkw#>>oZOuk$
zEAA3c%SN1iJZjYAWLq386f#qWl&2B_l1r4LT%L|UNcuR5#^H41K*6DXJ<zsDxQn>s
z#*d?@^KrS^Th#eu5XD=MGQ!)zSa#wW2i|Q=EQ~4hRM2y%J@!*;L^HA1gc*#Kkk(BJ
zyX%A>*-+joarjk0zxw0a2}rSG{mP1%^$cG1gTN32>5=Ak<&q#O<W~vBaRxweM`4dP
zqdz9V9R&zlw#qv_IEi!?lVO60obFPD`m|=;78La2A6xHJ8>3b?LD+IT7+d!o*t|rd
z_=|MW6cSVB>))7D`6VF!c0hGSNwU$(qA|}1c@Pu_(20Qhb^zKEF*RF>>z7dj$OoDJ
z5Fs|m9}Z*cK5XRjyepo#2DDQs*6GP6n#T#mH;%rch*#|Ix=Uu2wQ^Q|wIUIkeD<h1
zcpDLlOgL;3tO@6cg%aVR1JIe?+)ERzevcq`l^rbxh*B!BO+yk9v<kubkAa${qn|84
zX%9p=zG{#|J{%p%20X`8N<q$}b_x_JNG8lGTYADvIqg1o$g^L&>vM^4#1765-x8v8
zh)iRd<G5OJ;td8sqCG3L@Y=V4b+w-1#I~eGfd=^zdjs(%(V_d6Hn+d12|gHcc#$q@
zHSd9fJ`<}oi#&5qala0G5TezsvbynrW%U9P7J!}36y-=C+3f&w>EXWAXoI-CKv&@Y
zc7IAgmOwTWy(zkqSKkeKA4zg<u+6*$1w-cBK(ObrTbSG~Ej#=jq_f)N25v;^=$Bq-
z<nS1Tm`6&s_&j@c1EqsQcuzr%5ipf%H1DalxzFZ05I2@vLt2+M3ZnO_7~cl(zYumj
zHu>+h8k&6l0vJ4nqA9IjrPjLSxruT8lI{ib;B9C|O8x0YGkV+IWW>nHp+w219u}1W
z(PfaW_;5NIlrbUJ&_TA=!}BgX^`Li+&H^b?pUm;DOxZnJ$KJcgq4tb=X%xnd5Q4)V
z2uJReU&(q$Zk!JodcMNR$R|toNhC;}Yq+QS2u{PGJQGb!agD?PMx%@g^H2^w&5l#k
zL9D$vOjvIUF!(xUapO>%(;h}B+A$5<8rtrYrx~a=kH@Ylz{vTQ0d`*w`=b4ETZT*r
z6IsqQI!b~3Yt=dF^pv?^^bY_YKXR=ZAXp@tnl6PcKa(p{<&<U}<oXOhAxEM)q*HV-
zjKveK#DM$|&5VhV!_FS)$y@@iYBmxK6qgSZmYV|dcw=UW>mDT+r339|Sb$%xe7W>_
zZA@`9Qx?gt($g^UT~=I+nH?Z~lKlW0io-;bFUC@C7nG9a&_<|*Qr|Oc#w*k+hqNe_
z&R^pwvyx<xuj%bimQMeYC;$78GAws=Xd7BaK_wZ%f#sj51e?oDunA3PMJ@nD^^Y2}
zhNg{nbo6@t(eINRrLrQ8#a|SKfph*5b@o$+L$vb@ZU6VYP=B+!he9B6D&((+OQdON
z=@&o!(0Kxs95EaNzya^cB2c8dCfhsS*k&Tp(s$JG;h=zl40txQ0LxO|NTG|W%%W4o
zA7dqO3_PQ>8Ej=RI$UxtKOr(t5$h<G)yq$ib<re*8{<2qXt=d+;MG*jwMwWIi^erP
zG|YS^r-a(~d(Bz7k!YC~d!SbWVy1bkFAxpo7L3eFT%9?cQ)7n=*64&RPxl?NJJED{
z#(w18B!IernQi7x0NRHk{!J*Ym9WXmUvmNVvbpobY;l=D)5U3UG$v(xO+PHxfFbUS
zob1i;<-$){3`}O{u!l{}#oA4of~O$Dr<Y3e%ft2FPymI{UM8RYfBBRC7U%v84fuqw
z2v}4*t!oOp#)O&o7q@#<J8hopCgw8JR!#ZS6vV6p?tJTr=7#-w=O#$9se6vV&ey$1
zS<uXQ%7_ySLGKpcl$_oO4o7zIY~KMhM|olRJHN;lvi`HAHr52^jKA>u`F%pcAJy0t
z`T4>YY}XF~3E{q-mHo+w<S@{g96hlW4~Caqf)lF%*MO;~75WdZW?4HQ-wBHbRsxJr
zlOSXS4A1nDt+6Gb(`2A&Mtq-svIjMmX-$J`cerh$>V3>+Trv}(s~IBSi>}#D-QMr?
zs_A3#2t{5S1EY_7z^4c2%NTxpMvxL~FXO5`+b_}Mw|aCz!H5+nm4!17{|3(T73UiS
zOE7|ZajO!H6R;dgvyC3*^(21jRHkSRPL4ZRBP-BIv|i^@Wu$`Cp#|h0ZMk&mbE;S6
z=;dXE21A};T8A-zZ9p9*kiyzDKT^;Y+Iz3+aSe+2D$ulAC8dNuALBY+J*QF>AD;N|
zJXpbM)v28)fK(CX5m6$=K+9@GM`i$$Po<J8!d=A)hal&TKT|(-@oAot_-~<53QDSt
zQe#;DlvD$gq2LnRZ_XbN^>u)xuRziYn&KKi%bT37N6tKmYoWtSe{&X<alt&%Hp9+N
zpKFu*5<sLQ-evVDah6}U1{rmI-QfK27Ybr}V0W+xkqS!g40VV<jy2Xh1-#*Qhj`~e
z#%xI6F|xr)djw_e9K<+N-wOf@>LpGLKWk1P=_?f-6FLFiMhmTe2r-$75&b0}tKCC_
z<o~EQQutpTl76dRP7!V+@0{B_`$R*GEXcPk9u#6$hYk9kE}#S44NC7M=!kasgbauX
z*xCgg)lR&3AS)97DIE_Lo?8IX8=4zB41_Tkr?WJ1QlZFi6`-}>10*_w>nGttT|r{O
zfW1UB>oON>(GB&4$H)fkq(p~tb%0LHZ4Ej!P%A|l(ngP%-y@T=NiG|?(h0I}-DIog
zO51OSKMu5X040Vu2xQ|xBinP^qh+4pm-l1z9~$3uU6^8={s19cV<5jgmX#|zd(xs+
z^AeP0We$K2xl2myvknSMWyqusI>DXvE4rkK_%kFx>beN&)1;s6f?U3ojQHg)R37V-
zC);-jTLQ)3<I#z#El73WycJbimAZc~yR-UJ{R0@}zKL;&e^hYGi6CST()m7W-33e|
zB~ZXT90nAYgK~aP@YUVR2Hz{TK|&@P1Rs&_en$!p6ZB8(M+&r&H;P_XY~UIrs)2tU
za?f=Dy78UL`&C2)?^O@3P>8Dnr`qi|Bpw-t)UBuW+?KOT%}+gMN7%a3jX;%Kc}Ukp
zeD-zCTzkV`4tc6d2!t2|0|kFsH@~eA?-KkZm+9w^QXhjrrVJ8hJ4p1du2B?_Dv5D2
zJm`+@2Abh&2T+Av)e97;w3gcr*bB3g04|vG+we2Ub28t)FB<}zYJf+mCLfN@7ueb+
zxy~)Vy4qd)VG43&8>9FL^2Iive5f!K_JV~=?7UYs`vQ`F`k0!%My}m*h9C^K6>AEW
zWZ><&J^fxeF{aRfg{?w*P6!q00qVS$Jxipw+DXP*>LI!RHi*viqk4c}J9Nx}hXYNd
z_o~E{0M57$ObT{|H9cP&7=ohlD7-=+03kiVW==n;LpN*Mu1eS;lNP3y-(3n2(-fcr
z;IjtbdqK$^+xmb6^7iaGL#g)y+PEtpE`9&f_YEW#DJ;=?ywzZWa@$Vjf0?8R8~XSu
z=`EPdUUV)ks<t86d>*MJ>kTPs3td937pE^NxkJ253L+*=77~q9w63y3&~ZLnvlV1`
z#KZwA(2&E=Tmxc&zFM9Dl|-#4Uo6WPe1Y0`y^xXg3xmx&B`d70Lf0O!SU?)D7NST!
z_QBfhj3Ujyr0s9nlV%p=+fWVyLpq)FC{Ohqe=eKq)uljB6_JMLvx*1PlBbs66<KG6
z)4PAG(|#n<wUhMfmp9);55N!8T$Q%)Kf!cHl6cr+$j1PqtKA?639JiVoIf=VK&tq4
zx%<u#j~{`U&H@FIAO{e!Le`0^q~D>r@&co_^{BseED{|HURy8qz8@-Jn63x^A-D=o
zQEghkn}p8kQ0bm5R7MZBt;FUwK&`GKLuINkXrds8$1e3W-GD|rpel4z7FP6W0Ltzg
z??RM_r+}lB-015CwAmL;M7kv>{dTT57>yB+eX560Yv9P7Je!4>0FNbH)6iq57geGM
z0JVbGh2`h59Cpj+o$(xlYT;{rwLJJH*?OYR?`UE0PEf`roYiDT^xkEF4Lnf7&cffS
z08hoz3t2n-(cwp);5$_TSrOfFwV%N9G>kp}ES4&wiYc_=q5&6PPEjYQIkthLEpzGP
zVRBb_$?rfLo(4jUV5t?R5UrV2vB9!lHc<l(xy>CQfmETfF1lOtL3@teDhRs@f6fU|
zuks|Fov_V=sYi&RlO?T>Ez=`VRsCGRe)IFyc~-?KMw51M_VL<roSK@1U_QT6(3()7
zAetlHSckBb2>|kMIss3?)}80@i&N|G!X)Azxb?Pz!!<cnXd0`Rv^lP+JKO*s^SpbY
z$Q=EzQZ?=x$Rx*b$IRxB@b)6uN$qb@sHDhKJk+=<wcda(VuU)I^h|}?g2s5g6Tm15
ztsa70z<|QfP8?4LBem|@>DK))Gdl|2V0?xo$PY%S)N0Y^@~H<z{sGi^nMeLXXcQxp
z5VI;;q~V%wVPhRH@mh1sRG*e%V%`h?*Mvw)JXu)_7*qae06Hh~W_ECTM~Mx+P1Qz|
z131Nq<NP!<wOMq7t!m&>G}E2k4~@6>H~A>$^6=5c*R+yMbxUa*y!W=kypBvcN$_}<
zPPaOwD33^H1+s2JW)yKC!k4ESN*cjPWK&^1Fzk(1Cn?(At3EXbXHY~YawN+xMXItH
zXo&_BW}GO>5mMG%u)>;2JP3f{WqL2$&FGz-ZcMbNKD)#Kby=B?8n2mB8)1qi@l52i
z*e=>OgA1jZKc=xML;h4=AL?2cYvjB;LY{o-M7TY<u;29X@}p0DtFXIQS6^17=4!}A
zn9=reaJJ<;C>qLU3B6ESUh-J`wV*JCoQ3k$DS_kFA7VTAWxO(Dxlevn8RxNY@)$MG
z=C~ehG?R|-GJcb0?WKhiNdVIt?L7GeY$f~zIV>ZIVd0&wQ|4|yOVn_;B6U`0&CujI
zOE<vA#_~e@uMu}o>R=vY<6y8?pX}Kr9yuBchV^|uJ+0XA#P^JB#A-tr_AFxI0$B#B
zUb`l>prK(pZC}{~nXiOAtWhvSV8LVYg>F%>vN~s6Yi*{6df5~G!<WQ54|M|%*WGU8
z>cLf#SNnm!pW9L^MHn2&>7W$JLKa!8P)^++I@@aBt6ki-O><yO?LT~@Wtm_Y7jhJ3
z404nQUyGlH*4#ZOnCU7{bxWV0pyB=3*aAE8jf(_kGfCdV_xL%djoKvhjsU=4o*Gm3
zwE7pQ{zdH|<U766_gwhpF9GtG^ceIn@8H{Jy49W{xW8;)=Enb9UJO+f|Cv`%+Y9`>
zO@FsxC{?T!F?jeswZKX}i~7=UUkCktDt(ZyDnx&L|Le>D?Twqm@X9!#{s5^?Ajg}(
z{$ew@B`bArO_vZxVE0(majCZmROXNp1O-{-`m<*G{d@Tn><lHwZWrRbeo4H3!Hj?b
z8WrO=;0cie#`j$=F+=~IcUZaD?B8?+z3)OzEV8NiqCcKB^2ld?b-=Eg6=L;;Ftvkd
zPB_p2aZP{EegE;If+fHa_Rf%C?zijw(`)?ad*e2Nk|U~zq<?(r|NQi?{o?=m-aoC9
zU!VT}Jm)_}=-=0sHxH6lY%azWGA9!}xZdXdr~UHxtKJBounmCXi$JG%sg6wj5-F~Z
zTuYA!c=($#>)4Y1ngsq$(D|F-AmHssX&^KP(i!Hs5a<fQ;EZm7zu$O!Mv&(}Ukd-c
z+1dxBG2qbc0nIzUw?kCm0@IT{MU;~zJY<P%On!@Q|G^b{f4Ra98oKFcKq)0#$a4kb
z%dBJa|8k?ZjX)H<YCEgz@PB<%F)kpSs8orR{;`+%x5x^&H^eKbjePv&kBQgcJr|>C
zsBg2eG<xxmC&b@9!&-c(b%>``Bll-#yT4gv-p?U8ts&E1>5p*Qzb_;ZO-W-w;zFpm
zPoe*xZ;Ss5!dIYzg*3N=Bme(;Q~b}1{WU26b0_}x+y8m7f49m0eFyw=$KId_l<+wC
zVY)6>n|h^0NP>wXFault?7}@b(Y0&4<8t|LP1fo1-<Yfqx;35+d}~x6nygRergiV$
zy_G<8?Mr9R`RdQD=7ny~g+p%N3oL5Z$Ca;qzCV}DX3=pppWp?!As+L?CDzNsMV9JB
z+67`%c>d3jMzCH&Jb=j+5c$+Xwvvu&ChX>>j7B>jc<z)q|JO(UvwtL0VN`Z`YWU0B
zuAe-CFXr15$Hnf(o#laC)a7yaEoDLXJ=^cVl)Do66b?bi9OP$d53~dM%RT8~@u1~u
z-%TTElK@#y)|Z_eA65bqiKtlVX_;C$H~#rkGOzXUK?Uwe;GMY9CT$7R0|uUJpvT)A
z55Y{Op!FHCYeY9@w(}uWqJ*5E9}hV1+WQiaI02M9Gwn;Eh3Y`1^Lp>%knhj?{*P{I
zmibO$Lmj}$heN1%q&ZHpkY_4{(dTI88r+J;m4PE{zynAtrd+pQYh-HL0L!i?5EOY`
zq-O@mltx0R`L5h1RinLJDLu8(%-{^#ZaD^_^SO6`ZJilK>5r2`N=4ot7m8i^F#1&h
zHl0xz0$hQw;IYuKT!4l;uzXw|w#s#s@jq_|$lmXkN6QO!0qJ85h=Ip|W_cDo0m@Q9
zY%B-I=S1<NA>P~eD}(uvmcbqXvA)nK(m7+`y>~u$#2dV7aLbaMc;NAu)9^pMXLJ8+
zutd}eWaZ|VDB>8X_dZ#ssSAirnva1!7YpT69EMOtlTVX>{zVDx)nxYq)7ub!JO(Cl
z%*%(;4kZ|8DE=?T-ZH4lsP7x4L%O?>+@wg?ra>B&kS;+&K#<&YZ@L?#5kx?`rAu1r
zkS;+=8qVUrpJ(PhGiTm8pZuUQ>}&0{uK)Vw^3-(sXUoEL7B6r|Z$AI^X3TA8{8M_$
zYR6oT@sQS+SOn+4;Y=}3xjrWm7P49^?XYAey)PH~&2=IEb+Pl;v?)~oySo3g$p0UI
z1YFDKSw{pNwO!<hfhdGUI=|abCU4hzA}ZbXe^tV6{&|fCu2H<r(mET?2G}Em^Ls!O
z#B_v^8Yty{?*b-BqmrJw{D_xM$DmW%1(-1+gHV(oO+||Uh)YOY0}o^u#Cjs5PZY?*
zNmbV9{c@<uxUew&(F^emA&<A(O1qiLh+S(6OvFu)-!*mAdGq>>!*d~<ary?pLlZ-E
zZ|^{e*%!%%P9&^GL}au^aM>KGvzt+mpo+r>-ySb_5D?M`T5BdBvp|2Z`-mw$0h*6m
zvAxm)>T`>-qLe59=L!K$`{!=aec&UahVjT<0gObVHRAXj{2*=6p)v)`M62}TKXpv8
zPcD-T4uMbary1(w&f+<b&uHaw5RtsUw0L??_i@&Y15X^Zmp|8UFCKZB;q^1=>PwM3
z6ILa_?K#jH_l0J7+mZ@0E}w-#a~=S>JUzJzKG)_WCh@L)a<Y;Dr2;lR5LRXX|7))V
zO8+24uXelu2odL}D4sfy4b=<~I^8c?PLeZ*AIg%MW3W&wE}$v&iy&0>5HQm203of)
z8RPB7b_^kTtvQ*Y@bWhuoH17XBzyt<PnyG9`i{2Sli^B>{us~q0L#5ue9WExmD~Ps
z@7;d|NCR>Mxu}#7d2d5xbKAdwP<BE_$92U=_)H^tccxQ!78b&>_~i5vnr#G0qV<+*
z^m56cWW(OgRk@%BJfo}<t(j1uDkCW<wKMm_lfxk;jTYV^s5be}S2BU16GuG7@JbAn
zpe%ujwRAnFgKBbB*92As<9A$V*rx+YxA!haw=xRoVL*f1v2pTme@0`I2x9qsThmrV
zaL$sbcQiccs*YZ*ZAha?(jpEyK0O{Wo+r~T&3I@+?^3qDm%oJooeJx?7#3R>Tim;{
z?D)7xA5Z1~+aGcSrTS~80*TbJXFv*ax(E;znvR!~B0Zux{h#G6&@de1_SHpqncacn
zc8}*5aseWo3Ph8X-2fDE`Z<+j1Op!FlO*I9p?Zmds+dp$q(^11!A`vD+8`?m90;UD
zRPhMSM0%jKx8NZ7siKvpbYKc}b}u=j8)akAnQ;3Esz3ba8AUSv1%m3&MY;p3kH*Q-
zA9^iT2WC;`s=sM5s7aqNRRN5;3H0&E_Ahn0|GZCD88++rKL617gD=cJMjxO-84}J;
zUhIk%KN*;C1xY9Cj%n@a>W*n3WdGNP(0LY&fY!SDtpdET)Luid>w%A%Y!B5981d|D
zd}iS|0WJi*Yo?gt(3EcTDPXoNK6GIP2P-Jc*#f*0?aY@qLNg^Ax#J-&;vyQ<)O_Z_
z2IxJXr@#hXd(@@;{RK3{%9uflw4Q~8r%p3(!i6s{R{UCMCA@?U@7D6uysHTx<2pYV
zwewCWh$_iaLsFiHX9G03c7M68C56ZLeUk7%7e6!Ed#!zNx=i$1=PUuO;bPRYQF6b<
zQ5;0@^h(80A_>qX$TcLd3S(ruLve<kiH^RWzIeC=9TL0b|G-yj-y>vGzoOp}P=I?U
z!d_DSXA}5DFg=iZGW|J83W%X#7unz>Pm*%udQ<P(sF4G*EV~-3E&8|O&Z~xoJ(Nr<
ztf+4RFLy>UaQPT^dRZamZxbuOVq;p-$?sIn7j(kEMObwv=Da574sc<`M9f%*4)?uj
zc5w|+r64P9(ak<9{TMc9r;#+in4_R5<d~!m_oHRe6rrU4;x5Jvp7w1(c@(&vg9K<-
zICpdDFFV66#`i#Zre?6>X$?~PT#TAsBP1)SgeRxVJ#Of=1m7Nj6ZE$=KIB917g3fv
zyc{+7(?5Q<m&eSX94PF@YZcqKQ%!u!PPl7+;))&M8+853e0HmNV<i(^<Rmz|Dw-`w
ziwCZ=4F)-m6Z!JD+M^;V&9Gd5)#%KB;J_c4ko5j{P)3#DLVLM_c}KF+i*DHZ4chlU
z12GZU==)55x$c&?qW-BECTM5dM1;`}B>;&)zgima_d7rVHG868Wec|MeJ*D)KL(>k
zBM{lYNQQv7YsW9FJ<LDUQJU?hP#BQxxpID~)edU1IPAWJ(&3oV=;u}k{p)G4g{;w^
zQzwX@uLHc?#bKfTmEBR))9L8n-C_IImH+-Mu%=+%+E6VMif7gj+<-S}i3JtaQUVg`
zjB=*PQwGsPCmj1Uyqcyb_SgUZlsC#v4Kd>68qfdxox3fBlT!Cfl4JMg+cmB2yiM(x
z^(Z_Hd7dQzhY_&V@NVyJfamA)Q{9n~-t7*T7EJt8>j<}h<k9zptb7)LGi?c|gvww`
zI}-(UOD^r;h1Dc8G%Bz;jy>E_QFXRE_w|29gnEoTI~d^C7aN_|U+VraECku$dzj(Q
z0er=A*hvoRdz^LP*%)q{CsD!=JZL}nc3uyxeh+%}kKS*jZX8whYrvI=9p6Y;K4K7w
z?<jdR&vxT-6$G8CWd_V~DzXT7mq1ZR`A^HELRQi4(?^1;_KuGoY99vXz?*tIPAwnN
zh||s@ITUCaU%Tx~W)rCkL(fa^TQfR#(#G9jfIj{F%=|DQ#uw;;U-6sD-cjxcf>G+L
zg;8x8(!Mg#o)lF&AAn5hS3g3}?_Py891;qF!yxw-&`dhIF%Z%TzUnjqx4oBPbtvIJ
z;FGkj=P=7Rv*OBl$i@d|E`g0=z=e9P`|PU!DHZ%3P#}I$m&PJM2OHVbZNUeW{p{r<
zF!xm&wQy+ReyXaLP3eK61%QN^A5g>Zb!U%CD_v?0YLqa*6U&mhziDv%!d*gvL9Thp
z(wN<|X+0@4_+pw7%(I>NX)y$BA;JTf$XCb#q=ZJ|yvsCzg;9|V*g*6|UmFFl$2>VY
z;yFVTz<asSN&rQ=>u<{Z0iek(k|VlyB&EtnM7o5c(-anxh7(171FF!7{h#H!9t$As
zeiO`SW!rHrdYz$8c=1PoMV@-g;K`R#Yd15R&6wiJAfV8_|Fxsi(!0+I3I6jI`kCFj
z?&5JwRgEEkn--;(&R;~&j@vK}extDB+%q1o6lL3Fs#GfxdYS+xOMc5#FvZGSF%$2^
zX_qnqWqH;DAi1%uzp+>UP0xq@bzpb}4KDI!Ql!`5#cbQjhrj$-V5}X+VEGuhaqoPm
z_U5Y%-GJZpTPG%1xGH|VsOwIitt=(8A6tTxl)8jfe$@W4%}puF3rn7PQcFUJH<)O<
zRX&>sp~2~DKg|CGqSOkaWtqay)S5%$0i{R;5$#CusamW1HN+_T#__&HB;iSSe})<X
z7mkZbYX;0uHhLgQH7N@t<`@A=eiiS7@8Jt(RTa?<YLQ3!7f<LiPEinD%N;{MS(d2$
zFWn}q)CkGW*Fn4T*I$zIFiG+MV+(j?+G8RQTcaHNc(|yK>AQ;HI});kWeXlOp4b7e
z1KF{_?G-@CKCL61Q#WM^?6`}BIrYmhVMYD?_)7K7+5$k$^%uxi=osnr#<vEsHaL?+
zK|4smnD6EA^Z6&ivO{V{h3yl%Re89_Relfm5g+hArpcfc8echwdEt%%rG9&Y7@#!J
z^~s*nIc9zz3WH3R*1|4Z3wvu)<`s)zexxf*4pc?7vL8#v6!4_q6y9q#kaL!X;agIi
z+J2a3@59|-BRwhjlPMX16CLN(bOD<FUQG&Gh^^DvYOi>rX){{N<wS(mZvM5k3@FkV
z_Wke6${i)HgaIZ*oYum3#3P8#f@4|uY*bj*%2e=%Z~1TkVz}Cik$sByad3zw9D#}?
zUJ9(1Y_d)pH}gWn%N)uY^izv7W)E4JE&2AZ3Z|nmi`{(@#ZPMawnhe&Pz2lgbQN+U
zhLtwAs`;uRLhnCzfTKu25^1M>+{E+FlwCx+0|vOJ^7fGEuRB#>My=DP{W33MI~4C`
z{m$%`?Pamo{>=ne{#!>jQ}&o2qok2c{;4knQ?ENIQ8SPA$;t&x*^ts>L=c)88@Vek
zUW%>m!k<62<USkQ`S$x)-h$%SA?i@K0rP+++-8F<el-$c1+W)#T9y7O-`^Hzon2yZ
zaPy73plHV9@rsL50or`DmYh2PdB4*|kEP}+3h|gAC}=HtR?HM0#^#JKgW5%#P0}Mo
zETCFJfugS$Mx(B%J+ems3soQ6D*u42Vuwu&yBcAWjC=j6^p|H*5A+15oh@54^e=h}
z5MO%a6yXf5Mg*;f_a<h4oL7!u@I|C>VJ#$Hl=y~#`KfS{#_wm>=;ZVNT#EVdZNh~I
zEO91j;p1#x)gOs*^bjq=7-huno?-Q$HQ{Sj*>tZ7^x!8I9z@0(To>fB#OxQw)&%W4
z+nGUpEq;<s3j5(Z1D!igo$J<VSEHga1_(aNzK7OyT@#`BDse{zK;lj%`(E{p*!X+&
zdXsl!>O}-yj^(2>pUEJeU?T|^VhtEW@}60<qE`Hb7OFl;Hcd}5ZgSR74!V$`?%6}B
z!9YO{7mUN}9N?UaPo+;|W^c-%#2WG^jQ&Bzc7u*ho<}-fimocdfNlgd78w^HHd}R{
z$_;y8^#xZQacmfaLt8$qs?*;MsUms_koe8`VcM~n+ySl7iryibi{G!QUxe$~b)P){
zqz$osqep;<H4qrb%CI>}7Kg{O1$hY_kVO;$cQx4ilm`!~KltToch~1SHI`Z7BR5qC
z=#n9pV)IWoxFG}axGtX`icOVQ0wKzDG7GV9C^zr){A3ZK4`>si_hUqjos^8M%y!po
zEcbD(w~Sh^HwTd3?D1i>vFTMYD%@;%YjY1cCgrwA2~xUijR}Qmjy70Vb|dixGeJ^3
zUsSP0)9X)e*)r*I!O`~w+(NzN{jm8W-iI%0n2cC+?C}SyDK*7yekQowN(yQI;h|_u
zA(fWMmy3+N{AZg$?=A{$p5ssMk>AdzFph|nZ7tJd?DNPbbcYWEaSWdDztax$!WOY~
z?jBoUB!9ie82q6RM8S?kYmOej@Pz?^h{^)GXBo<{fi0c>*`2;mY^?+F9g?0ltF6W~
zo#5i*Sv}czWzm|~A=hV4K$ngDnhtgo8s{JK%90eLqE>@=v}ej)hEQQ{L~j&L-j#S{
zAs)Ga!9m1_Qr~oUs^-6!rahVfpn+Al`ntnNae`Iz!BsspD&SCn!;aakpZXq%6Q_U7
z0qs^(49P&EJ$=E$+iEI!5Do!;;$(f}8U}p{Q%Q|-*cJB4J1(>>qyd(w?GZBGG)=Db
zoGV^YjE#xnC-`XWUuW2>6@sYcui(2$DueGW%=J}#4`{*L)J7C#d@j$#)PiY0uL$j$
zajRwLZ#obE3VduuMU4}FGnv8!cUPP$4LN4Y$lY<4IKN9uu6ug6D4CZl{dII}t28F%
zdN?;^4MR0fjRgP3N1A&DU}le={_O(ohx{YUXmklehj6#2XL3KWcl3X|<%P+kS$~7<
z$4%30(VaOj>%6S?$=fb=tR@|Q`5P#Xwc7EratmTaKYyMpt!wZ298%X)9_ugAXib0;
zAuZZ6?MYJPPEXYMa_mI`R-q2I97XO$t7}d`8VtwM`S}qgK@B<l&$Tp5otF{uVF24D
zGK1)j$Q2=~I=y4hnIS0$vVQQ)`lXcCHG2Ly-eZD%){A^`Q0>~An{~Mpf!!AX(M!H%
z{Uw|qOD3LzdL}LUH*kiknI3K)0BJ-kL{^{N4bx+m;DkGP<sQ8Ic}@RX=ERX`Q-vOD
zIRLVQTH}%GzVzQQrIZ!L(pmpTjo<{C0dtQIR>Ee8%;11xp3H-<twhg!DUI@1+{W?~
zeVlx#G-)()diaLXTYEw*s{jZ;y16quzWlpU3EwD2I?3>H+VI|5X+xoFw!24M#!Wm9
zH9YI`GUC8-@>nxm2XVbP?ixv-dSb!jbvSoGoWWY`W;RxtR<&6c&+|A%hUO$M?@?H?
zd3vH=2y1duknS)ucN`{+o=7n{K7Z~v)qpvao;gig32oxVZ)Yx32DzunO>y$UmIPbT
z$a~dnvq;sVrEYQ<<RUZh?4&ned(4Z?*)3}W2T$58U1%}*1dnncRfRYVVVD8Tk5b{2
zdA=hOB$X<FA8fQq4P$62*+}*fTFebTiO}Cva?UM>$+XhN_hutdXkpj~KoZbu@G~xP
zwIC=p$b$G?M)Xg9X>yZn%KKMvaf=~qExr?XR{oUW!d#Rps2&~%O;E)nmHDSrv3}=t
zW+bC9>1|o^mY#WL=V>;XmV23yvMmOI1ioEJ0KPPM>RUew$i;#S%U~j~5H`FIlVf{w
zkTTSKn<thfF48p0l-r%k1KkFs`f_8Q&rhYQ-{t>_bxvwB0TU&L*{f~Pvl?YGDne%7
z(jHdNTzDak5f%V>5_ZA6ZP}`t;eMStSr{*WQ%d|I_o3L4PX7W}tHSyj9jMMl(HNlH
zwQ3A^ws$guCvM5UDisp%-kct7>MA`1u$dxXzcE?q@yd&Vj|^HBSa`HR;_RAbByraB
z(QStb)PmMks-jA)w*QMWGa-#!#BYXR(B3m#Ol~W#_nhnDrzR)<O^DtTrAz2-kq!^^
zY#>euId!ZJpT^)lvOb)kMmmwLv-BG3060R?uilqf9WLAVMOy{)dMO!rEo0z1k%5Nj
zs=H$_xzjgLZGt8JbuQ`Cq>E+pg#oox!q96xnAG+6QdJJL;M&Rpx~RT_x8F1+kt%*i
zSZ$opK~93nug2d8x7W-=v0t$JDF~jFR|-hZgy)hC;hi#6M;h<A6!a7*Dqe8+byIG_
zS_@xa-z#udwUu&UbWS^vB0|f`!h^K@H9t>5ha4lEv`7*tX<W4bs=zd;<tWDTEA+N7
zg{`5KYoAPdI25Rfhl!gpcwV!)822Kv!Zp@vxRv&V_0x!E+5#HWIIlDyr5mjU<H}$b
zNHTD>#>eX6h~z-%WNFqUfHkn<f+o4<NRu;zGrdI=wsf73TXF?|W*=ekfM(fFtf;j}
zy?)wjy>M!CPlbN^r17{``qSPZy^zUchp%n_o{-x;K_#W>*>j}-g^m;+IDx<*1Hnfj
zGjxKIb5$ND(a(MXvZgN5VCuO~w9C3X2y65kPIGhFN`)L}hT@XZ2|`@Ixv?PYS6?^8
z9Yk)$^gHoqayam=8h3c)o>>ab8IeT)S~uuK;wK5lbS{7CG=c3xa-QMxGErEGihwh8
zo+Xj=Z<+wSPI#%5AoaASYmx}vRwpNz>cOc-jQgA=%fldKl#_v&Ps213AEnG3`qARu
z3i?EoroSrW@Fh~UDp~l`M(K(M6mw^-hzN5BofU!y(urCQc13|~NNm|@?TfiVD$Ry;
zI?aY%bS%{HpqdxoOr>5>;)f@0*_hjE-uNax4?#NR*rMJjFn~v`IX=hdY*2CHNy1y}
z9kQYSwTWE4n)89mAxO=b?MuRZNk@7UL*}Ph)z=k{9_GX!h^v_m1k`dr-U_Kv>et4q
zRG)3y4E@uJZkPIT@mEjIZ`iiLxS#&DM1eBfZu-BT;_o(bG^3uGL=5q~G5tb5(wCbC
zu8_RLL`T^ky=8rMk$le$VNS^U?tIgK?%U^4(oeC%2LwlMAObW<3e>k!|C0B3?{l?3
zD%!fL6t%X|4lot1nqH72okvO+eX#Bz9iQ&`t}-0|^WtmoF;kiS`Rs+R9kp6l1MM0|
z{0-Lcsk9zwBhOCGPB$beR{G1?-RAaUPU%yWVou%OE*aftMU&MVsH$Z{fe{(}KUyw3
zFczP$wIjNVPk=YWyBLxNtu);fd4sl*Hkm9PU1Rfr@5eskrtQW`uy+b36{5RVi=n@N
z=@Fl(4-oCuJW0hv@o6A)c@@><H}V3H3ItlG7vl{|O|ECSVylxP_|)D<=9?Ie{9enO
z_V;dPCUZXC=5%{St)>y46j`J<>MT?GGS%exeSjE}6aI@=dkQ?)KDXqlHjS-4)L+~g
z$U6G}3znrV6a1}=ygAz~fN^eGQ70O~=cNt-#j<CB`fcU7^IM}v&t{aKUF72-#~HeJ
zeFB6<FP|;m=VM6h2@>miG&(bc%k#RaMGTuSit$VGx35Vi9A{Kl>2=C;R#H06`D6X4
zg_4&3gt2V3zHsAYY9ao)iR_p?jHz#`KYByiEClA9=0*<ENeyVKa^qh_oyEH~Q9jCd
zh*%8I%CeYRDBwe4jhE6aU7JkYPA@QR_el6TDqlLc;nYv~iy1N;2v@6r;kP<7c&+yY
zQ~t*T(_2r^=hw6`whN74Hd=H!2OKTk>|Q(szgct%wC{0VmUuJWCa+D|IM$VnT*G(b
zquqIzp{Mi8lPGm@AW;UQbJ5W_!$k@<K(*^|8w<0_AB*%@-rU~{5kDQX_gxQWwf0|W
zQ=K1~RWgb({d}4h3x5dA)f3Sn8=bIgbSZvMBlgiA8r8JYz_*zEvil2#W0|d3V-wq!
zg2q6SfqFF(dGixizs8hK63C%c+64&~Akki5_>U-KEe$uyd(|ipogd6i@!_UW{qJkT
zWVG0x?&r?Uln-94{Jh|v8+IdRDh`9BNb)zE26M!5U|sbw6L(0ej<~AFYuq5^uO=$j
zkyP<>Mhz0<G>o|MhbS8L_KJKs@KAaa;~|=CN})0Q--YKzxrh7PH-A`bDz;3u!V~}N
zAHyUbT}d(1b9Zww7NjHiJNP=J70%A&1Z63_%(_CeNU*ONQjQma+wbA6$E{d?^YJ*u
zBkzcl?_CQgDyNGP`K(9czQfX!U)yZ_@8a6TP;5Qn2f8cS1Ru%O>k5ak%6Gjk^$P%d
zy8V;>ps>AE@j{b!<L#DMww!xRWwH0gq5KNG64g^ru%}cNzAv)1w{xaUXN0$D!`6B-
z#HkKJ68R;a#>uB)ZOn|K<K`XxMwmg=nq!J-N|O~6G)M%=9t9gF603YkOv?2X`e3@{
zF+wmPw$HxNK@7RQnHstCtY@#1yI^3Z{ULh(08E?{*K#O-R4ZJ>j(hU#UTwsB(sJXa
zr;>G{VB7LcV{WJ}=03w)d&4`!ewJm;800td-r19#v?9yCaoRV)jn_4_Mr-a9`WLyk
zdJuRm`|(G@Wtlm^7*et3^mPeaIvhI@<V@uzln6S^2Rq;Uw}<>g{WtABq_84OC9B9}
ziSK8(_WtRgtoC@{ANyWACOvBv|DG8#nXC<}x((Lew8L%$clIRt5}(Sw*2c{hzYF$A
z)1*NVrDL#Qk7>NO!p473XQsc~(d@drMu+3~?SQpQ?uBrk1Rw5%&?hW8jT#^&+>#ia
z9i_BofFU)@98>1+_a?<UC-QZU5luMob-Y(1%lYSR`6{M0FqI(w{4!plIQ(rxLVHKT
zJ@@`ShjFFK=u}LB?J}kx(o!B?g)HA;Kk~Kt1`6GH!DPo65d4Af*wn9x(h9ZQjV1<3
zXV-<Ouae<>_&*hIPR%<uhwchP+V!vnclA_S%&_I_3y(g1_3!RV^FJ`c(e2fm%k)(d
zJlq$>c6lcd$Fp%#=o{M}$uB4C3W4|a#}=FrtaGgkuh4mBIxnsVRC0~#MwIk=43ZP$
zNO+wo9rPDm$i>E6+%^thNN@EeH;@}@rDIFb-}z}9*X-RCTve}M%dScxARs(E95iE?
z{cr2iK{n<sO!n?Uu|+#xkjq9(o1K8`mr`)JoVLjeCHb1VIG!>Hfp-4Ji<C&l&tH|a
zYPCOQpVZYJd0*9yJe>PH{BQuQPnP$Q$hPmV3vt@qF_8<%2W#VJAO*a2`}vVR2S>0z
z{$stC6DS=F1z`}S0(Fl~a9yA#a_?_-EDRzJ%po)$TY$E$x_AKngE)&X$d)hrgyvh&
zshy;oiG7dZCc#3jzj1y8fBnbCG1wtd&h>OjD7w_;_us%Z?@nJSb$=ZD8$>m@T%P7W
zPy-Y3%K7pOR`B<DT|l}``smmQ0ciCkHvpmkm;sv$_M-!`8^mf*R)THxIB?Il3AOF2
z{eA@I9=&He#2;G$z1$5_&(Hn>@imh;XzDs_>5t6?p#;BYI_}*(PBT}imq9fZ?qUb<
zF?2tA#sl_WkqxX!0L}R4g|_`z-?M^lfP+#2a;$K<PLTD{`O#fgl0(TBGzN73W(CoL
zdvIK8le2uXxi+%aRaDqTq|`F_S79@)Th9>>UK`jJtvNjW!4O7@>1>r=eY9eE0Dk}t
zu+SliCD>a20r7!%e;+#=Z^7qjXtw~gFLh^-q#J-ou{FFV6Dy9WJm_2g0{gwtqxBm5
ztJlwRB$z&JT<^dUm~pvofmFZdGVV9vOQQ#`fUK>5;~Ll$9bs#L+)V3`bW)e<F#R6*
zx6-}MZgpAsKLvatGi`@%zj9ZAuZCv#r*7*S39uJBVgTB0U1_W&thtN7cjn)$0Nrj2
zXpKwXC3#frPZhN@iH-s%70qKjd@v9(lpHOqb}|+PxX3k>KiT=X1A7nKf789N{IC9V
z|8=EMneLABPUvHnSvMff3CF!LeI6*rJQVbRyQljxLBJ{2d$Hi?4LhB~0^13~-3eGI
z*IAIha}CapJQHriMkff2zfElKm{saZ*#lU2=p>ILG}3%U_PoW>(O#rL_$or%E`Uj0
zGLLF1Wq<zuQP1yhDD#>i21p;O0Kkt2m#iJ^A0JTkSp{!*fo#gA5xotqSg)6QppdBI
zcCv93)G1rqQmN&7?7lF0mu3upwcW|iV_=m^{Rsz!YlTP4kB-jrsWS<;-w*dk4_(M~
zmz`CB!AEhusO<iha}G(Ak7~$3x&8{iVe7v-mi{=pmq)Jwa~<B9!=twqbbsQg<VtKN
z>2!g$pO;>0Ub#Eey&kE8tF7(i!)kl^H|<0h6cf^ohVEH55lJ29ZEE;@4nO2#J$kwV
zf4NG`a|uQP3(oJh{Q<}8;>Etlip?|}hQWq2Uf6elpT9j}GU~RJyORP=9<}!>Ga~ha
zinMs!YSoaw^$r?D#ZCezl>4wwS=y7hDq#P}P`mZhEgvJN^8=2O2K1fxAfV+TxI}!D
z9hoO@EU&|j0vJnIQ2NS*mRu9%TL70HmfCh;g?ue^O>xa`@zj21c41=b)IVm>5aD{x
z1e9w24ObNJa{c;fld_iTS8oOMN62FkL0dK1kqJme`TZDraS7Im)`N5@fY0E`h|~et
zQvsLN6VHkpAH95kMq{H|uF&pofcXqDGuC6ao0fDZlcHOzeb%IY*=v9D{nSMeg9XyH
zi<4<q49(7Ck_U(gCh0nCIdc5KQADulMj)~F=%zQNJ|9q&oUCGgxb>Z%${~Gp-c<oV
zej4{;k-A?mi7rYX#cU>6ebT8<fR{g?ER4}+EnA4%Xa5G$rvx?oI|;=|-0uG4R~0%D
zR<Ys#m&*+S@`3>;KyMkeDXbLSi|;`3YP?io)>;4QBJ~%!>R0$~NdD0uq(l|t;qqE=
zhgZ|OMO}SD7hgquEji8ECjx4<PP>4Vgz^yh<u2M5Ht}4i9u>C0?d6A2i*+LFZPpsF
zl-MMYLs2K*ue<;Pd#x>O=MgQ;=O+P}^tPWE&*h1yj2NQD-a@`S9L<tr*UI?Bd*vo4
zI@)%oR@ajH6b5$F`oVoDN&RH~pIh*`F+dzx;*VDeboY;ykCZX9MuX_H+K1C{KohU@
z9jf(z!^~Z)c90CmqY~y1x{~gbouw?rsy%+RUuHRVVG1^cN0o>hg9Q!)-Qr()*o}rG
z+6TgJrHz3%TB^P#4?Aw|mIT=r7?)1JERh@?i~T_5ece5J0oGxB9^+4vC@Srk=jaFf
zsI;vj_;D?JY~_iT>XdLt%YjtJ9<is_!qC0oN#zUA<n1!}Da)PB)DbPHye}T+eG8?M
zA-mmCloB<ydjQ3O7OhCvq?V{Bv64q1a>DUw%3)7#vaTA$oBcBQ6?@M33zHgek0-zv
zp%O)Ol+Dbrjjr;^E!jyF#1t20DdXSRB1$u&2OnM-x3bBCn^Ln-TE2Hdfdn{9sK@D)
zsJuj6D#$B=;^y0{PssS%CJG&q+&X0ASUk-;j5mU}Odjlo0auvU?n4M!<anmv@+DuQ
zR*g8=pFKngJq9Oz*1fJw^y5_wedY`8=t_wlDVEfJyZdcouzG@2maL|OyyPLr4+HWv
z<h?pRdU~U^Q-2;hzcF7Fz`3PI3n_`dv71Io?<x+m`l2ueZXLRJw5YX0X@kcx7?2RP
zu?I_YiO>VG&V0c6#J`TC5KsI$JP4HA=2IYP;^V)z`VI5}RZuCJc1Ob6gQen`W)|E8
zT$B}{bFqo^(Q?-4fhw`E{c{Tl;6g^l)^O#ZrG7jR(qb`~L*gaFe7RAh@>hT_C3-p}
zt7jJNJ1^d&P3A|OfAymo^T+#FITmq2ya{0-=LPH$@;RNAN9|X7{9`DdATk=tHy|JA
zV`<@A{S`8&M1AmNSk$nuNhrJLd%AzGQ2Q}X=Bz({?b64ES2C&dV}%hxJE7RzQear1
z@FeX<3>ajHLRR%4A=7c{;Dl@=IpcScit;K1YaxefyI$A$S0k<?TKU`ej1Ms{h%Fz-
zESu-mTr>)1_pr5(1>+^c$;rk;=uXr6)67CbpsLOBLQ&G>6hi}TJy082YLqwaLQz$U
zqP#mtpl6tf0T(oj@YMQ-$eXGm-L15&+&NNdU}ut>VGjmlxuG)Y>hA=S|74Ego>zPW
zA)=j{lVoAI$xKpQ5X`X0o>HT>&y0P^I$4dGvb*GJY0ir$-v#p>8~e}cnudoFB^D4q
znsl%LWQbpP6KAX7h<AkiFH#X{lG`(m@rXXovgzG+n$Yg}Rax;8phw^>NWi~no7#t{
zkxQiZQsYy9NAwo`=yO20xb7WieiEXi`nS3fqYLedXI?99!zz#{n{p{OqQ#kXuS6im
zD_D5oB5hSc=<Rm|QAy8W!!8uo9>|vbko}74zZz5GHOy=Fqx^zr6obsnr`S8zttPeW
zg#rm0EY<1KRy9<ac!@)*#X2TV2%YBQ1MIVVEzG$5%0CKuuB^IpYaja?RF(I|aLW^u
z#F&+p5+RP3NNLJ0LkWX=_Q<m(rO)D8Le&ILkDW2`rXrqRGyOy6&AYXEyYRlF?Wf7#
zJoUd}e}ayA5@(_u^m;DS$9#(M?1ucS`Qdbrg`x93f*-V|g_M*{2dPJ27@rW9TfDM3
zY*(VX?NoOG!Q^2eJE^-bsTS2fyVgPpCGM%SfXlR&;`8vfdy?=9d_LQtTD5j$Pr=;8
z7rUkec=b=VzR~B5<Lcx|G7iUy^#m-DYNPZ(hmbuLCAvwEA4@y!Yr%TiFO0^+2aKyM
z7&u)5_$?&~8QI)IjwP5Adf{YYkEu)_gOg^gY%=lsenYHFaA`q-7E7vjTcHN@kcpK-
zR!hhhYXo)s!(d_@(YmP|Rgt(JhQdNuj*bF(5zdr)`Bp?C7f}MQ;Ky*b4HEHE-MAu~
z{#u-^wqcpflN}B^t6j=-1Km4Cs%*xhI;^=Bg=M#=U$uRk#bXv?>#qd=P>Bo)#jgw?
zTT-0};+^^J2m#_g<&FVw1{Ck)bbqdQW89tJc|CBGuY*R~%-D|sv3_K=)krsND@5rO
zkNRm_K=j-jsYvc0t+JDFa;&`dgx5+l1NX`V=!u7Fbit!o+&d|~&S);NCu@4x!k3I+
zz?lG}I1xBXg&!SfiV#?m-k~>h!>B$8L`jXO>tEtj<DanEKBmVE<yYICZH{~^mNp6c
zMk*{<fO4Yt&?{rI98#$CpplK@=jyeGz6o7o3+1--<5Pc-z$1u@%Oruy6a8%Dm;%IH
zs*Hb>cGrvCc&BJW{Gy9VdG-sf#Qxp4NrWUnc?(`E$vq!hK3|KN?fPeUEZXtZ6=FmQ
zIbZs<m&FDpeM)>&Y3HVu#FuxJ5w)U)TwR)Mpe4f*$z^oy4is4mI6e7jA7DosCt<{(
zrBkIOcf50yH!;A5-Wb!Tr2&ldc#Fe41E5N5Lf-9N{0F=kf2LK85nYtR%Z$aO@7@SV
z6iNkXtiQr0`UfzUFQSsP_r<I#!o?Q^1}yxGpvaqSMS_vkjw@>m{NnNykbLYlt-kj7
z_5!G865*17NZW;gUahMw{zQx5LGZnz!m(K_&f?LN;ec=|W1EcFGm&^rRUYLIxS}h`
z89&8VWW$<jClXDpS1qt&Dyn&47sg}l;O+Qt$o8F1z8&Hg1SvW}qip8d)BcnB$?>D}
z!PvvqpJP6T6tP%9ax)IthZBs5%(x@{wLClER2jE5t^@0<vl9`kRwtj|L3q`Jr%lul
z<S63vSttuwa0^~jD^TF%i&yrT?b;>AI#24KSForQV6%g3*sXaHt!mrXcLnUmYRvNS
zdQ)SQOe^uf%~x|+B<xiSB0>YauTZa1C%>O2Vnj_@UPc<w%(Hn>U~s`Q<SJmfl7vm4
zBc_!WzC7XAH1LVq%?M?2cAb!%H%~ti%l>0QSks~GS?f?R*q5jL6{VbJ)2(YSl_lf4
za=1#w)M6avZJ{4KnkQRyM(aU~>Roa6JHqNCuLYko`C>Y?s*!c+*yNcqwuJJ#LA#GV
zDQn-s<VqcI6@_pO@p4GalV{PQy~=2v-=pDL<hv2<|NIvuAv}Fo^PE2h{Lr5<C-)@J
z5y|CKA~WO;yo0NgTLa_>c2#G%zFp1CX=z*P@JX64Mt?Ec_z}>A>1iFoxSOsf9~_ue
z5y0dHo%q!)yBf$1rTf~ei+tn^O-ww-)bNlYcJAHcixm9WS1>Y|hitKso}{i}Fh9$y
zBy=4>Ei>C15gsTvN0tK7CzKEU#%HCqz3a_;!#6tX$6}i9s-&i%DB`%YY5BV=<TYMe
zs>*EMgBji@`%`+xH^j=%9P56*{&im(=9y<4u}UG&AD!V7%lLMqK}@2AYswdXo)8;{
zcMu!(o0UYanfHcXFtA3j@3GnUb)w>|=g=9r>yI4fq0VSo1_cyO_EpIbcq%685M>fk
zdnl33*Pe>E7sftgiR<qh5EjnuKzYAmL9yj2t~R!?A=_g`bsDZ|cG~PoS#@km_O=o~
zUDtX&Gbw9eif_+zf$zg!ca_Q13du0mO?Yds0>LZ0#OY%3@J0v8cSJ&cKLqa--(=@=
z&aJ#&c4PS3=kK$j*uSf&7*o3i-O7(ctc@p@n}i$iisxZzy<e3gzG(ty7|eZk@yXb$
zIYS-l-gRNGjx0fWg2RH3ol&N7=&lG`r?t}kTO6GgswSf=yK7nZG4CV%XE>Uj_g`g@
z5$I3Y@19Y3`j0Wyd@OuV%^Iac+f$B8ReS;85am^jq<)M0%Ce6w(A1W{IpMA-{-)P6
z2!XTST3%XD^|`Yq=aKz*If+9u`2>EQi-rA~>8oI?9%xg)pqW3yS1fGqjcTUIXu}J=
zl8e5lxW*tb5QaQuZeBb8YzQksP1WwH-0?@%9w-WTXvAmwSP4Pp0p@h{|J<pzo*^ll
zL6~p@v4ZHXMZRc98K}*plLVoBfjnm-jkGmd)&HW&_0)`Ep{`5fot^iyLkn^y-hfz{
z7(`uT6pLo#3z|bUR_C}ubQXqzBjHb+kp3&{rEhAd@oUtAjbE5>M~NZhoOSOwWkSFE
zC!j~QtD<P}OhAXDBCk%bt{Q`;&bX1KU4>n<o@Uhg_k;yVbKXueA@^5SsUmOevf#?_
zBfn?q?8Aw>VdGI^vaKoLz{O@1HzB2WH4`^uz~jf1Weqs_-dXsi+WrKw>&+{h#MsZm
z{Y{n0`PR4Wlwms5<kV$TrbY^fYM}8CcW5fFj%QL^5ZWcg+bT`YN|Tsj87v;5B~OdP
zIR!OfbN3YcCK<cJZc08CImH&sthG!(xf#F|Dm%};Qif&AX#F`N$f!29UXR7FxCQ3Q
z1x71dFj}#|N-~9K1?kGv3oVEsQ!UhW*|Br)<JfQG+nvwr#3iNNQNX-v%1zIE82l<{
zdzTRxH*+!cFG6|*pF?^Mrl)1{tJk}`$DTQpB6rBw;)UX#h%ZvT`^`QQ(`Io1MhbU}
zZ3yXF0G3nF0BMnMigxBc8h8dLKxcS*ENL-YLqEo$m2+WFSa%>fL;k_kk2Obm_spjR
z=Q!w?TabrTf*2E_CmTuFj_NhlQFOj+)U$uNn0hQUHNq*qp2(Y^zY-nVu(OI}M7d3k
zkFk#wETUu(w^YoiiAT|*6>8gMLK|csB{+B_>tr2J)Y_2j8|o~?yv<lIK4LCuDbN$u
ze9YkXI_I~&AzaN|kGvs!`}ir$Fh{7MRpeKF>TR2HR%{|(4(eNbkyj4X+ML4LGsE^i
zi3vn%Rkkl0U%<%EYm{;{cE5iNfO${zs(Tg+FKjXOPW_uLey}OmQI+{8l<e-X@oIs;
z)XOi+x0WNuxl%S*;j0J;t@8eA<{q|<fSJ|wnRxu^BX(7k+h(yK8qY{6*wFPMkrCH?
zrO@v!ZD73-2|eB#0!otzU3k!moUv#<0-Xbe!?gy<Ypa=lny^8gnz>ricTjY%ms(^+
z72|W`E5lvPmVR^YxATh%rhSCywjh`fh1?H=#bJHnCcCo4Vh#bFA@^{qRBLVOsD_D1
zx2Yf6DsLJdccWt=Trwz>#Kq!j_UTzfPp1xdAJ?JbK>pun-(}`^eJ)v$Pis2!LT~T|
z{tC)26QQ|~y|?X`HA{A~4otdGma;YD?J6`)MYO;!S9Lki<R*jYjk|O8=Uu|Y4USnG
zDoJ&HO%8hs3@sMk2@G-m{_!m9)S(GoP@K=0&l!G_i54V|-i~d`kZ%w2f}-2t_^JQh
z<kruS%J2d(<fL4IQar7Kd!LGaV%076RX6#AOVg8xHw*W(?|e4woKXp3MfGL8M*?RN
z_^rc8$XpJwrnaBnt0KzM_R%zj2vOu#lEwN6ol8ZGC;FBYES8d)?BA*1DS4$ZDOgl}
zkBc|?3f~SUQm=wig<kpm<f1c2vH)V`Upy)C3oTtWsb!?BYNya81Ws|>0kw{7S?nLk
zFg`EB*|kj!niAphbWA<X65jDOH~;IhFEtK--4>nu<uP<>(b1yFL4<zfMpL<0^53A<
z;SCZZ9{hKF+S5N3m+jzBk6B^K@*Gyp9AkN5iGpX<=AOl84HAqR^bcAM<5nM#S<j<T
z*`~`LJ0Mqel2Dn2S@?K;=u;zigR6R;&3i5V<x-GYQXo@IwN5s14q@8urb1sw7Ej7c
zCEb^D1~?R-nwsxA1M)f{%19P5JO=kuti0&-`+2ryB)FAIPGcw;rnrFr3uo<=w~eEh
zks%uLce(AM5wfif>)F)laY&moqTzHrXdd~UnW)EFF-B;87KK6g=wHdAgifSI%N~6E
zKu~X_2bi91t+Fr&$+h<Yi1RB9wU1Ki0(47^=tbo=rr|Tbzb*1b=b_#;5KZ}6l$28c
zzY`1p(w_n_k=KjQ*G??<`HLfs5hwIRU7K2eFbX9`2fVFjA3yrkKq8T;nHvAe(?nWK
z+8B~_oN!b!&O*l?Q?FO-A-tvqcRlG3REb*rpPTg&xD1*luRk~HflkJ;Gz^nYd@IMN
z)&9GtgS{iJdQaN6O`5wdbcz;*I-H|M^!^nVucsfl($7wmBi*sx884aBJM7zAu=O=Y
z1<ig!=O@DwtL~p}i*+|YI~SEmuKhh1?_tg1tof#6PEYWnMmL0A5TQ@FKwJ#>w;7l@
z!h;-0bF1E904WEKU{bSGc+gH>_nyWpee9N)_8~pBH>2}<TtCF(6~cQ*&iQDwEtTZn
zinoyT_sr78Db8jO^)(Gph`X)L)!`~0@<}OC$sQ(!2>ua|<sm@tI7$9`HfM42mc?^f
zfNnOj6J1Q!NU7%YNS_e9h*oOOXKx2g_kuGsx0anE*XLZZXTA$Qg&y;?nS|&q<Luzg
zY&6g!c=FkS;$@qZx|9J)zwk`8!?@5!y_7v{0B@^!ax{KcbrmKpI4Vo##@PD^B5XKQ
zQfV6@@O*SZ$mMWUS-M$Y&UB0oD!Uz?jJ~73bIRg96n;b!pJgMUL#V<#V=G66QDqpg
zvL9`Nl2oRATt<WU=B@YMz6@2wbFT_=ouS5X`;Na!;_NUeks%a0X0rC(iDZ>uBMU`P
z+V6i+Ecby)P%$zwp}!>&hm#jgZ1QY2tJJgKBvA>F16fr5!!`&glW&lUhm+$^56&BJ
zX8e5ufN0DUezTj!Zx2oN*!8gSA|F%=@<4GkwD`rqaNPQb=6a!TZEI6PN(C%6^v-B%
zjHoyqkZjPxCqG`_4J|ja*YL)S%p2v5<H3wN^BU*K$omo+h{g&{?1y$e9dLN(abQ7b
zXB*GEFR`!{EdI`=b25Exg1+mdXmHAP@Fa(K$EbB*u8Yft@R^NU!5^)!gTLtu+Rc=&
z{GX=1>k_vy{pl?Mqgu2guaVgNSw>NR@LZ#9d?UP#CzcXwOdq5~@bN%?jrG6RBPNs`
zp`vlLc*+XD#A=~%Y$CbK`gUkxf~0`lB^IRsYCK91OLR9d&P}vDfDm$#0S{u~$N@_j
z!^xY(AJCvs`aorT5r{pA%7(t#(eqpKzN#ehd|vBol#>})Ux>#~JD>g7{Tp6u1A{L*
zp*sZW`25_s2R<}90ljMPdF$^d3<(}9-x!~xQg~JP?Y4ZP<7Y9*NihBnyJl45K>xyq
z*P4m6efDrx(Cd#YO&7HcHCEyoOZ2<9G{j{rB{$wECvWboA%B~=qISx#?i8m;H`Uac
z;8pa6t71_ZeMTTp03z2{U4u{2@5=64AwGfv8P6QAtA(k@5wO^-J+;v!69^!#jdsRQ
zXN&VyD@Ad5lf=3*+VWYK-JC!=uzsrAE5bMo*9*ke6)q(A7wb)iT1h_>322YJFH8<D
zE{}hY#2?IG^d+yvHlN+*<|HqKDS6lSgDi6eeFA*<bSJP-7R$0ieZtz-n|}|5A{9-t
zTf$Xs!@qSzpGv_!m2dgX?o4SYOOT`6d)!in2X}*9kC&Eytkz+JrpBKS0h*)B<(Na@
z*=LDxA{}~b3}O(}B(<s4Vprvs<Y^g72GG}0eRa0ZP#I0o2r#_)C3tiuxh<5XOGvA^
zD}aFyn#dDSt@dP<aZ)=jVA)dAB0K+H12!R3PQT`5S|P1JWKj`aM1IPvFwv9sla*q{
zzO`2``s9dTbBhsW;WEHCNK`#N65sAFHQe%7s#c1D+xzwU+nc01rF<w+4OGs&TsKOh
zGt`;PLS&7ZC&RSiI;F!8gsh<vG#zN~xlcB7`^B^?5&ga@TJ|J!uC=9OXK<)iJyv~t
zLyzNN<+pHRGo>IOgLQImwr+ZD+IM%7Vmm(hs|qW;hw`zCqw^aI!M!SLA<Wi%avtf=
zo5V)wLj0SDJ~5PQ#Jwli^v9Eqy9?E_<GzjR-uUUi&PoMqLXYZt6NI(n6j`PyuGHR4
z2Im!tEnO4<AV3)DlmO9{=S-3diCXODjBgf?oKwwBTEk)yYFfdRpO9Vf?b)bxMRD&&
zvBrFK%e^z*9bG2x^?jA}Lw_UUjWTU5YT9}A>^a3z3Tj{9S@lkJ;**`#9v&WxqFmEk
zrQE0-gA|i9pr@U6s5<3sH-|`PMG1YU$n^esGb;xY($%K6Ul2LX$oEl>L>vH@iC{{F
zcm1wJANR-|FAuL{9GQ9M5hA*4MsuD+ZV=0<LsrnJF<Dh2UrBePG}i1MZ5>~atHoer
zh!=gA|J;#wU?Q&ajad#0>Yk9NC3QB^DR0y%ITJ3{PSQ?h(gLqfXz#054TI{bfr-%Q
z3Ax2*Z*c#lpLDhd9Vnf*UizkX0DBl;-vA@#zx+NH_!k5!7O!`Fyg7ZpkcwwN?IHcs
z+~X0%K9KXYSr|QS9Dhjnw#3dlsK2^X&S@A_>>|99HY$BGAYNsirCEEdrlS6`sOV7i
zAECX-2-Egyr`yrSg~D!ksp7N|FFTgQ5L54rART*7eSh(tBAKW>*_}=5No85<yEkaB
zj$)@`U-$AIy3LCUY8-fuY|;nqUc~?0$(c=-H+^;^e$DVPiMY9xh`rKGly)ql)uwri
z?E|v7)aPX$MP2I|4Z|nPU!E|2NzYD^48KZHM&tS=c=OB)`Quix*`ilb2#GaTB611>
zPYFa|;zp4x_SU-)bqNi-lNuK4N#M%2mRJyxhkA=#|8{Lh=HIsM@#$4WXVg<%#wx9P
z3d1O(jfs5LR$hDFY9aMbO>R7<HeH(}Y;uN|(mUNYF)o-91hOHovB|4rm7fK+vY<Bg
z2o`j&^}-$e&4muH9}T5x&f={0<2w{R1##QZas=eWLwRiakKrUM-Y=K+0_(TxVk+IW
zSjpXYP{MG+Pu=%M30`6|f6wFIASceXkfHsKPLa^6HF{!fDW`%~9H&n1!B3Hj{zbhS
zGg6Kc(ZZ?QsmtFVk+YAGC6Hp;E<}IV9}+wDKM?gW#`s%Xw|jpADt$OX>8y~cUeIi`
zy;Xt-V4DxV2OY@cGBq$r4TvRBti)YllY7uIk@w(ZQ^mPPBAQDTQso5XB_P#M)u4vj
zDXn9NM+QJC?wC=Io26G&9;8h)vCqWjW2|iG`n-FB5DB~8`Nq*}8h<}h1T;auI2M$B
z?p3Cr5Rs`^PKU7zj7;@abso=W#K9POWKU_jgl)zaxmNPjQXbDv4+CZ|aPOs$FdNA~
z9NrI3ehQaJbIQlnY(EJWH+{+MXY{k2nltKB`D;=?+6zVBj)PYAlxY)(F;mf(k*R`8
zcO@{QG?e_iPv43<=|$sJC+jL>$hkJFr!cZ>VMq?5$(V&Cq?pMB-kvEZmi{;~J7M^n
zamqVKs!0q5yv#(38U7n>hG-|52+G)y@fFMymbc87f7JsTdH<2gF8C0yQ(XwvWE^7!
zYxa>3Fb%j2u$@`5MDcntw@LC$wc-{ZX)iY&2Ve4;muMC}$ZGeEhc0%qg;j)E_NT#x
z<LTn08IA}x5Px83urF7}_)sy?`^D9lvTkI1g^Yy$O%4BMwPM*7A#TwlV{BUWi7pu(
zYZ#qqC~7Pc#_7x#!8e4$YC-R*%5&qx>WuF^^)H_FGgnH_^QAvQ0=4mXMoG3)f{_e5
z<>7;YsFt5vD`rI$Ok8OK#Tzm{WqvplvZq8D*gfNm`i?oExZK|k?d*)qQ8ekyk7i~~
z7goTNO_5Pmf)izKbxRie8JQ{@HoPNXJi%GOw6xt&XM~l!znb_j8W%2|h*=&y<w}UI
zESs^Iasg$Ma#71M2^4osSMN`2As_eNEl<<5&|0tZhXo$X?@3La7O1v9PnT9bj`gAz
zEHhN3E{wH?BGx})$=cnYm}7X&m-A^%u0$TMk`ebc0VjNymDXMj()GPkAYwe}5^7NO
zygDzU`Zq5>v0FeBL!a+-Y%}j@&zWjx4HmfKKRNK;aVkJN$>I4$bKcgxopve;49<Zc
zzw#^6&#seqoYM4VMIAAy%HozWD*Wt~8kJZJvkx@J(I>b;{72N+Jfg$*KGk(w+}PEt
z>XR=&p0J8^#}5YvquSHkmMM9_``{hau13p}{~1VmTXrNq_>X^WwQxhu1WDm`BE>27
zL|@VOD-&)#S8URc`YWa5S_#2FQq5rAjk_}%YAOCd=iSB`kT+q^<hkA%DA<KWWe;<r
zOFR1XpvV(=8Eqt*`*ts1z3>co>e$0kyXU(zsWR_OF%5Hc&sNX(V^ZwrG-00#re2r5
zA6athA%#mq?<~1ry|FI&JS|qAeAgC!YiZ2HYY`EY^dv|`$$1?m3R#(;pL{g9Q#b%o
z0VpB4iHZeOSfa>1cn7Q*(-mS@iziMLR~HHcq}QzpMe*VY((-`h8NqowAcAeuq{n~X
zM+i^V^%?yTawtL+vJtA2Y0+#|oQ38&#Hu?OrST~fdLkjVZ=@#^F3Z0(P1VHAvMf)!
zb(G5hBIIXH&NOV@pItBA{tJ`wBMtn>T9q;0zeckZuW7H~*)Rf10_$tpq}Q+Or&t*~
zb~A@Ch*}kuU>F-D%Y(?y38SCCqhENioFiVfT#2ldyys@(CRAiB-Z)CiCc>j7*+)WF
zoJ7Gr^FEV`kWN!la64-FUc<fYd=#CDHyJ$|k3RH_Re8%x%L`Y$e6L^idtj}$#)+%>
zfxM7eAcv=;il@q$yZ+a3qPFl5PLz2$eOP_w(}xUGWr<NYwRT;5_)bBDW!Kw@<anaQ
zC#L>~YDO|r#l92manE-qx>;!ANRnuF6^scygm(YFN|fLEvvb5wYz0k+sO+j7ZS>b)
z4XnR&QC~tq-$T8AzE0MMHeiRuNv1O}Ox%>>^y1<>Pk*$i?2_^)^#@kTNoZR!&+5(2
ziBIBnN^DSEZo$C#*^+Zi61t~Cs8&O%q7PhDekH!)uEwKG*S3!JvhYH`;+2AchI7(j
zk1LN9Dc|Uk8-ZfvSKAw<qYw6QtH=U1v9YPCe?EqDE3Z5SYzs_tuQ5)dxl{H`e0h#C
zPX*O(3huYye3P>omkHc@2Us@U(=?XK2@K*IHVLs057t%n7&+ZCMmMWVAlKqtyp{KN
z6QGyy;FCvUNZ^XV%L6PkW~jn|5tUO6jtYuFqd?ClO`w8b#^3DZ_DY-d=Hp-EC~eG2
zeMSoQCjG*D6*|)>UKk-XaN!Zcjpb3^h4o@0$K;>=lwCJV^6pK!I4*x^$mLiJB0kyB
zU-PJbz4Az5CBGp#(NUB=j4KFXFBTuxiZ!Ei9IA$}xj%j!sYFUL=6P61<unZjkK1|J
zML)yi6yHLRE7KSlJcv|RyRS=;lGqx0IDllW6y-t2HP18Iz<~cPk5Zq1H!ggX`nWa1
zFrXEi9o5A}Uz+tTM0v>)@ld!vm)L~tk9cpdb2GR@pF*?!H6w1rAd!X;v_7fY_XFRp
zdcG?UMwp8#TrGiZe_d7F^SypE{se%R{~2X^zH_@q*4OwEo>WHszegE|i+3#}^jLpe
z!*mc_Wj^+HRTIp_56((EsU{C^aqr2g@QBh1PX4DHMtg;zg|q&myDn2^h88^?Pv&bu
zZeg1)>Y1FY(nq{*+W&{Kw~mVXYuCmlBm|^OY8a62?oKHIDG5o*5m9RB?k*_-0YOD6
zVUTVZKqVCz6a)qdL0~8W!QURg&-0vf);jO+egB$;EMfS}-uK@3bzgD!(~^0o^3S8w
zZ66A2$xsmz=;(Kw{5q9%r(W4MyhmLOsL@Nukjvl_C`5Jr@9^Xw_Rrs$|3mz1L~g;R
zUj($_J^WwxOaC@}GPsXvah_bft#e*89y#-*R4J#U``ASqnDozNn*tXauPqqiLDQ}i
zWNiIKI;f@Q2^uDnv_~fy$+TJ8oN?0*AVbZQ2gS&T#*4;$rPFAeFTf?GJmJ_l`yt-=
zNJZfhf1v{FyWy)a2l}-)s44R^6>WOWlq3wD9b*xfui#VYC7A*Txu!AxD39Z6kD^dQ
z0;6Wh%aBr6jk09uH;oml%{V<g)y0kq1#fEwfd78k-HJlEr55s}-rW=R&neta!zi##
zQlpV&m{8aPv%5H^*z#0AOub}UaLHa)0IZE>lAPxjtl!VGf`JZv!H<&otQPBYVx0W?
ztTxexvROm|(1G%dNp{bE(uTYF=%P<+qmj(U8fk%tliC(qg5F`DwVJ;@#}o*8mxX&s
zNAVcFFqUW2$eb(hz|qW9<5O0*{hTIZzOK>rCM1nXfxj-um?Y|~<VX3_b;_gxCqGr^
z=f5!n<L$_J1^vPfIp7UKID`jv*MiqHw<V{N&s&fJP<L=Q1!LV#WF?!Bpg6jUH-7ui
zNOeLq-pSNe$~Z@}^Nn(aUxolrX+kI>HkDCZG%%MxoM>BW_$LJEt3|eIwl_&mUw%OC
zHD%e7Py}BNRWjuTulcI)<k2+Kz8*>P?PLfC@a{@>z|0F<>RU%QC>IUFtLvr_C!4v?
zhK-NHeC04NpTa$kJTea|biBBXGBUHTZ_FZv1TEc<Ui$pVMOJ{Y?>*yF+cSzG&ea&>
zv1l3KIO9?C1e#?Y7WF8&kGjNOPHSErv~gC&VX#Z~o>xBqOk4r)5j%MO`vSoiL5iJ5
z_TH(PTLpYWnHk8>9@&4s|Na5K&GGn8j2@Z{W+c@vzVh-f^j~%rSD#DucO*mZCFH6~
z)U9sW(RQyj3$$_D$?FbhO}O3Wg`tl~fvg$3!2{(@U85wFhcpN?Ol#I)p1Mkax#Q*}
zn_zUR!F|!G@LUt7$+UJMr}oYXx0M0isxT_<H*@?fQ#TlQ7f{UJWa{35Hz#3Y+^7>m
z(t~XYh$6pzu=E9zv*rUwxwS|cAxhOmVsnhH+<|X#2xbH<0nG$%*_W)iWEuo;dog`@
zsqSWDh)O~<sg5ExEodj0w=^U<5P|aG=g=F$dz3S)s^>{-CEB@wCi;n92Zl>bFZzKV
zA=swL&Df^iUUjovc`67nac{gN>MK$}9a6IM?qV}0Yr&5&F7YczL^A0>;d}^Tm<8+?
zmV))uliLO6+}037#6@11X9p8!UVGE{WLSfF=D0HS-R(NakMA%dW;f}K#{qK=@ksmt
zyWIquMvC!?FbJ0~i%z@3O)D!^A)O<G#e&yqqa5X#8f(1rqpMxHWJh6B?37QjtWPE-
zii@~dY@j66vsutT<`==d7&)3Y<pkv^JT{FrLt09+ZF-DM(Cw<06EyPa^xU*FCOEoT
zxt1=3!SM$%mDV}huX)DHjTchxw#TF20xDTiM>e+(M{f=#VS5I(pu<L|VXOLt=B2uG
z^@C~^SsO5d<2q>`^l1AcKb5W?ri!=+a`HkkrSVy{5TkZ;Bi<#baZ@tpn>7s6du|*1
z&ouJQc4<zBC$z4OhurrGr>P8ALH0?7eF-{<en}u1k1>)}M;@T@%Vgex(V4`P``CmY
z-`3qZgkdFe!^x^RQ`qF0pc*8b`z}XjrjkyVODMxxNZQSHmmuZYF|_oV$bi%#>tPi0
z<3k4cC#lvEHR#knHrz<%&n>w7V|_W{O$Bi4+_v6uAM7xr+}90Yx=88Zy3nyp;_z!?
z2-I3=*WOVueI<MlvVbWICv-($f(Vq;NU`?SG<p4>6;D*-g#OK;P^QeEw5F}e3<R_9
z|53iV;v%@h`+$?v#6qowI_|>g9Riz$AZ@(UdJw5#w0+Ln@XtQ$nc?`>7``Q1il(2x
zl^|H`qVjake?42jd-yL4S%sb`fyaKqU#vtID>!X{4Pvvvd`A$kKMmMSI6Lg#;4bui
z3uB(Yk=k?Y1@)N!obJC|8*U>H21nIiMJ9+DYWE?;4waQ-WkW#+70UmWg7B1srOzO3
zG?!^K*H>J*hy(+~o1eg3lUhSy^sVc<{R`qgNcv4xbSETTL^XNLt4Jzy#*1Ix;Cnol
zu8c6ff3BDLnP9~|NR@gMF(_h<pG^lBB>^b|C9u$2#WvK5Q;10?c^5dc;f=NKTOGY)
ztEsW$UbNNXF|J>P;Qp^SBbs9^I^2<h%&jMYHHaGR&QHIom&cIjWhL+^J2YOfzBdxa
zGml#Uv+7mJZi%A>)-*rG_7vfS*9!BSe<N3KP2kr$K&C3!^f<39C<Q+C`^@;OiEA}l
z=L$kl+`6aX%_gvp@UpgF0$x&x3#Ox(j{2MaU6SmXiGd6sWmZZ1l{BvpVfnH!SA)AI
z-)!);&$ZfyeB>H()+ZriQ7}>cP)Ghphyjk<2+X3_0_7;R>o==DHJG2E$;)OO#DcS)
zEv}2oSLLG_Q<%JKedT2YBTP@@Q-oqHrY!YYsUJ(vlQY8w=XT95BbMcb8E?wFWZv*h
z(!v;WD1zWnGcEfXlrCQ(!&SRKNl--7j`>U$;pp0KnD$aIN#gz~h5NL#xL>iL#Qdct
z=ayYD=xspwjYxuohfkGUC&-qZliKej-;9AR%!@sydX^lLA++_?Rq3DrOL;}jsZ{Fx
z_%Ubq7`T{76}-zrw9?7))PD^TZOWw{{Dt~WS1^y}v5p$!&mXaapGp)E+BK<DK6_Jo
zYv~zTf`U7|n%uR>>r4reQOyS7w8c3!2v@NhJ1=bBq%pEsQw=e|WKK<fYJmNjG_Q8!
zKj&uOEgY?pa>??gclN!e`yF!6METY|sADN?zOdxS8BcoiGDs_V+f0I3rez#c>h&Mo
zpr}03RzAywx!WA?ApD|a67Q8fH|W0k2b^CbnPspvN04_ya`(5BoWMf;dMJY&&xMp2
zKRMXhC+_Qqn{bLmnR-om%lez5a;!m=kEm^}l6kPaV%DbHJqU^WV~+4fDaC6nuWKMv
zAsN-G$=6%sezt$U>p9<o@h)pY{<s%5@^N9qb!PBsz@n@hPU5Ea-3*UF4)INXqA%*}
zfG;fQ=`$74^+gjdazFFQK4f>Us;<KN*^?w2k2W%EFH<h0-;l$GGOzE=bJLKp;w2&(
zpN9aETVKSs+y5#6k?_<j8h8Zb%%&6m3is<>9%<PT5|CDH$|}qK>-aRp9-ki{)!b@E
z(2A|fQP2Eq992goGLNiugkG90_u$iLP$Es?85YbO^eXx<Nc6Wof(<x0KP%3DQsmmo
zf1fkV!y%QTHsH|rgjkAeIsc1`W8sJyoKj-y16EK#{W#<CD_ndYxIHoEl`F|uzJLpr
z(E{d1=C!_9h4Q%{V4xA3S^H|`#);g(le&;K{?Izn%nurJk-v>UV<x6gNMk5<Vt7>u
zvIwb*NN@hpzf@~<Y|I?d9cs&<Nld?;OU}jF8!4gM7!5%>b5Xz6=ggMU$<z0z&OeNa
z`0N5n#eAa^>iXr#(Ja>RK|^`KNJkZ&nCPeDgMKT3bV{eJ<dN19&v=hELncU~qxY59
z#+n@`r2JLwB7-bxch|cMe#htE(K0C#5@08BCuz46dpSmdON26<!wkuC%A*KzOtwH)
z$b)s1$s||JPW+{4rqHgs>H_P6hFtb;h9ZPh{ZUk9<(uGCd;Nt?xw5frsV!+(?pbdO
z`zK>IZ0rcZC@G`L1OkmiHPpEq>V2;GG8}W?qf)Y-gQWr=-u4#pUcSi+Zgp+OY)b!Y
zT*qEhwc-af<XzNbRutkf`b(<xV!pN&@Fhe6tE09<J9>IB&wn1-rh8r`?35Cb(lbQZ
znj@9?#eV-Gqfdqm3YnM_CzM9fF4TEw!l9<N%`Ng-7apmA`!WMjj0=x{L8z+`Tdu%2
zO$#AWy*;XKu{O!c2A6A-@m>hu3j_R})PJ3rs2oBWwQq!D4mjn)L8FyX3UC4#rT3%U
zSaM#KD^55#K_3Bc{jToi&vHY^`I?nbeLmq+stXn>VL4%R6YzN!S9!%DIH*KARobyS
zAI!Vsr>zq}k(wO&d+Ld>UB_HP%7zb8|Ln#gFVT;~-3?`Yy5xx`d4g@cLyBA8$TVb|
zLu435qqAt!<oLu*_T@<RYboLr%RR*r3NFv^@h@qT&K}XP;kZ*3vm1C6!GAn)Ki}Bh
zFs_UbQoWb5J?<AN<Pttw!L}}pYsg&O?XoNT_u7b+JYb0<iyS@nY<PwbES<#2qM51n
z{@D}{@xQa;3Z1Xp6$GXI-C_B=L=!Ed%fk&qBAXV6Qu!tEZqIh!PXT5ewOL>>KZiwb
z*94{<p&&`_MiG(!i<w(wekg{U+C%@E8c(zgZUZkI@iLiQ0diCI720+v!G2hF7V@0p
zogJs+=Rx^xJJx`pJmUUh3hl*TV|W@!$JX22`P-a^T<Rmr)Lntkpx-71V%tJQT5l=n
zJBs;#53T>kvW1VP;A$aGso<8(i&IfA$YR|RThhhvGw#wohm51~q9~3cPRxM>!le9O
zUmWqJMvDIItF#<d7G6I&aC~I1)6~N`9i762>s%|TQ;2c`GMh;9W{6Q1K|%%(xTDO^
z7CF;q)DvEZ$W!wF62c@leTP*V{`AY0;<_l9i1)&wtW21#-<{BUxqv3z=H(d0)9>w`
zLmd19yfWr$g^w$;n>u!_NhfV*U%A*^+{?axe`kD;-G~k3@hOfMy6j5kQoHyxw`wg^
zrkQHZ_?%rCjcANevD_Zkk@ianslJ_~t$_ED*%V^8byw8ajCJ*UMSfasSj&zwPrQ_z
zCe!><bB8B?U18qH3rv+EoBP?$Z^b?Q=*G5E6ve=f!!y{^2!VtE<wFCM;Hj3jdbe7O
z+WkrMv9tOa3sOCflY_3#ZFm<R>wxZ3bO<|H#ttKWh7zv3TKDptnF`{W`gIcBW7<;B
zno{G%({g@{xj^tCbB9>fZFA*z9`cp4*E}Rsv36H3bBIr?|BMdf@+#VnUDDLCt#f17
zSn1l&R{#GL=Kpa)`**ee=O6EL;_BaNgb9@F?DdJddpn3z;Tz>(DX7=*_u59J#w~1o
zV&MwKw239Ii34wGSOfMSG=nMEhqqsX{aGUauz$mBg<JkA_}88855>ckZ-46d#Q$Py
z|MNHfFJI-~$2XK+qggP?+POWi_9^=BPZ6iY>7Z17bbrhX(e0ac@SgXkg6OjQPauJ+
zgq@H0f2Ozf&4eEp%Dz0zk-;^Mj~J_Y*^lZ*|CO$JsrMvCzkM^H(D`5O+`qcI|K+8J
zSv3N-iuD)i+PTvW)F*M^Iu>If{dEShE(d(zgRQqeW_krnfp>0C2Ym?aW@E5$6C&7!
z-&}#L79=@306I815U`kHg;{bWkEPDD-y=$~YC`WHx(U%<{Ncb+@2_fU8R((j3O;uY
zH&i4{>v1D`*Kvc4JACw?3ZP=0rRS4$J?5XM_@74fe|gLQe9tts2ODVh5U|c;vHq7Q
zn-Z{w852E5hrOq}A+DBy(;R$qG4K}&-&FaA70q!!^)*Q^*1GWse$@D+^L<O-)3m5b
zX>5NM7fV8aoH+r!2~X=s`D$W-7*Y5&Xy%q51hg+mJpZWtgGgTySE#Ov_3@wbzwhuZ
zEe`kfkDs#+eE*_t|HrraZ|vquBv!7m3}V3^yBq%WNe_;}ig#3?hkyTPk1sSLeg;d;
z*QOJwS_AoAZh;`r(z0W_yZQ<aQE?8LOAc<#0ao_}(yhccv)EW7;nCn@hnvon82?>p
zcR9Wd^nRuY!vBBy)c^k;Z_l<Wb^%qtR(SqjKT0H;;ula_v0^*+y%*Ywd<vM%^kP-s
zgG#Jg(nhVv<%12M!Fclj>!&A9h<$v;jeB7I{`<G6kl}*3YeKBk;nc|hFq&nPS_(Pn
z@-$NpJON5kwz*qqAX_CmX`AhQWmnS?2B_t6riQXNJ3wIMb!_{(_BhPRANmj|PGhiI
z#?Aq@M@aUrH@f6x$_2OZZ9ePYL|qDMmoei00e?!L1qhma2tb3_rOV@%$Q_zp5aJgL
z?iGu4^j>Y7Rb9<16mGyf8}kqZ2OXE7oU;x3Ku+zK^RMU_SOHe7w)dp__n;lty{}Jd
z!E2h6FMRmFT?GHvdJ`#|7(q?A?%WVm_3u&;*&5#iw9M1e&q6MaW|`b~feSMm?^~?M
zwjY@P(SX<RxdW`;wI7)HKCr_B?xmZVD@ZmuU@zQ<%|w3)GUMW~k<#63>yNKQV)02W
zsbApkO^szcV|5}i@d{}6#}HR>c@p3;7ZwO)q4ccaQIBuOT-UXIa{ACp9N(*ZhbScx
z8|}`7mGqK@dhRsOULOzo{5DSM1e*%q0ff5br9cae8#)xtdI&i`$La*Zv+HI|+yg2m
z5vt`0<oA6%@MN5>aQ9(#x%GkjEA>M+Je$<;SH1x;uq%lXHj1ep*eO~qE%KR!&;Q>a
zSPb@o-F65v|E~v@(gW^Y9@H4#^uXPDr*y?WAR8(Y`ozjks&?7OPq7RVk%@LeV4j!B
zACN}py#s<#jpTPEIIgR)DtwE;8)e=1e9I7t&2PhL<h3K*zjcZYo>$m*ATJt}Ej-oz
z^QkV37V8;(fF_1NM8wq|EH~aqe8x9A(bs?sxg}r6W*Q%6od997<Ztltkn+E-S9=F)
z=VCjE-fbtm1my(e8E%6d@b4_%$$|(G@@Q->UO(tN7%#!v1A~SQffFE*i_?_!-c_%+
zfv@F$u;zexf&OP1*!5mBpz}vVOcpo72GXm0={q+xmqB7)_+>)r|FzhN)KP-tr$Z=g
zyyM?{Q6!kM=X?Qp%*cuP)aOs1fGh8^UtO@R)5<ra@U!VXAOaLo040ee9sb@aY4Utm
zHmV1!Vy(CD#pS+ks0wd<DzmdmUHVI74r|c-EjjyCV-~w+3LarB!F4WBAdV-$4=8^}
z9!}|U$6}3g)4vdArgl3xJhAmBzV~~Dz8_RBI;_C@in|DT#w=E${yVFbp5jV(suYSR
z;uHAm{zS|!0g+(M-ktW+^w4dmeiFMq@aqxLlS}`XLp)M84!cGyBZK_^vqny%dVImZ
zk$gN*ECpAAwi|ok?BrAco2m3@ptm(E{UtyAPZozrCsC8FPDJ$`g;i|Y^3%nX_bYX)
zodvLY;y~{h<cR;EEdlp+|HP`{wSREY|6`qhd=*>Ja<dLY{!`E-iN4{Gq|3Rt-|J|_
zmdhsCkwI|My?=WF<N=3GYzeuEjY;lg2*iqeLv~Nb?2a2BwzPj0v;jB`Vr>EF+Bpuc
z^>c1xwe{dgs(Dx!{I%vJWPSw5(L)!Gj{vLg?VAnOUp<6z+KOAo%V2$lS%RW8bUY-^
zGN9T)xqQJ-jx}%&91<_dymv$94^VGATXU@H#=oHe5x-;D0vA^YG{)=>iBF&pnFsr?
zyW4weEMij{y;=Wmd6ZOLdf)&@AFz*R1eGY<O78Q$@Qb|xZY3Xx!E|WkVS^dE1t|F+
z0x4pFwld;z$c8%GjS@^wbqqEL9V^4-GvJzQNEN{mV5h&`y?^EOZRy4P6-$uk-LH4=
z%;do4&W#aUdfp-sbu}=B6~C092)VR@+7ScPQP^Tgh*0Oct&}qjs`S@Ake=6%VsO#7
z?Jc^=npLtF!~G?9<qz-$-!vke(l7fm4;3fJ>Yed+gsyJU!AI4fSW*cuJPJFix!1hg
zsK|LlZ52H$gh=EZ_gZTQI-irQUjGRI^NsgopTk37!iSP7cY!5IjZ0-C%N6#7d@5Ds
zgMRh1m4p>qOk$U3Cv0^8_t7hh&Q}55fof-1)Ac>DvBsUb-0#aD0F-0W0fYsRVC(0t
zHM4~f^rjo1Cq_nwd#8zP3+Gp(8Q-IGccb9KPc>q~Znk4Jj)g-PALuH+?Sr~Ky(Qlm
zjnVu*d5Jv_Ewo1Hj&ee?;=R<9360afaM+=~0E>Z>y-IFH-faL*M|Sa&BR8>83jv93
ziC?&H5Az0`HDm7zUVvP=EIST2+@1f8e`;iNSOF5y<3mgRlb?K`2P(_YhM4t(10hW2
z3X=t!&hCdbC+Zb@p>Xz&WtWE_!;>w7n1%3k)<s2WSyJ@s);9Q5R+P$FLYDzoW||{U
z>Fg7zd0_xYu1qZ6xchUe3S<y^$HpseJ1k>AQ}@y*Q`UYOS`$_7>4%z4?r8iAeg;{e
z`fIY?KmXZHBGVLsyTMDIt9PejYb~Yy3*6mL4&RgGD*-PFf+1SC%DSV%j1(}$l?+`y
z2O57)`<87!5W9Z9{q{qZZZWU`uvk9}vT%QDuz#D)Q2FAJwU&2B8(7fZ4)zK1mtu3(
z>+Jg(1PIpi3ReQ1(ud9e8MQ?kQ|1d|(thjTt!AJV5B~_h!a_NSoSf?Ee;-J030EpJ
zv1`p?B<tof4tbyo7B|I<!{y2?PytS!vYWu`|3_$~;L6@5t`&~SGBD+(RRVh1SRma0
zh@>3ABVLj2K24<*Dgl7QUn>sZ@+a$i%#Q&(oxy<uL$><|Y$Qo4R(I@i>>9&Plv)wD
z3Ax?{knegN?OPm`JSi!JY#|TZ?gS0s&12>BXc;1VhC6@<WYb@&ivdfZTWdQqun{=t
z$ALPs_I&^0KGw9N=ixPC8VqSRa6vffPEENgyA0UsB9CY_^0Bv@6%<Gn=36bQWx}RD
zI9zeSA%zeFv*o8auaV<e1CAc?jE_mnFz;aNcfi*d7oU5J#jXU<bnhRSYlmKf51jr1
zIHvaAyDEtnKN{uD#n&i#rq3?om=G6%PrMJRz#GbbNTiC=Mo`TP*7(Dr<*WZ&G1qo|
zqVSAN5LYhYNuR+zLg+#yix*<IpJZXlGOGL7tC@^)6I_npQX0NAg=5`5yf|;xMveds
zox<)MHTyk1rna8b#QExgvJTrf-B2zzo8g3%;%Bw$edjaPN}u!y2b>ei?8)F%Vzl9d
zDxpe1>245)KFQF-ZfG*$Eqo2pQKhRyh;ON1)T@lgUkJNIHgYS1rzV_QrX#n<sU^!P
z#tl0KC;w5;57<Gb_S~J%XrJ5CJohlStRmPj433<B6l&Cd9dko%ITrAs!s6^aAe&Cr
zTG9B-e&UWd<C>76U->8;IOz+dvPQ8wt8CG^jI;Tr1TG!b1~979%;O0iCL+IMJw!s$
zuL8gog3J5Y6|r#eaZ0qRmIi{Pcm~<*peTiB-ts^-nmUosozONMGlE6Bb|Tj|&tKG`
z@m6F)iP3dpVy^V35JiK-NCmdXM;*jIA|v-idfjdnMIG9aB5e?2KgA|2I@T3;8_rTv
zVx7~j4~t$!#v(Ph1dh^=yWJ+sBPFDAsclDXQ6GURi^r+eyN3=kKtxJFw8DGOOna1~
zJRDjx-Q~9XJC5fAnjzb<%gl#1D(ve&d&8cJNaVgN4tcW15b<IB;iN!)1)J0Mk7q2n
zDbfd;v}O$r3^Du4B((~6I3B(i9iDTXJH$`E%fKN{=fPC3UEc2CJ&=}I`$49@rUzHm
z>*hwHfYPAUJ_zD&ui03KDKv?0Ki>F;Kh(N$n&Q&%^8JyB^%kv@y4S$mh9xAiGv-)o
z7Weqhcd;p`2_}ArmdS)w2^iemwKbTn|A1A5Y*vHbh%r^gRPimW>V9$uM``wFUX<e|
zU`2?NVbAC$s{O<NoY5@Mh-$75Cso6UrYk;#xWoejMAp~~3(QkSA=4q>5C<f#^C`r4
zg2E<gLN)9E)z5<C^kx*>{)!)AAC(46+NNDnaxOB%Vvp+p#79qcfIQGBq-t@XUH|Qq
ztS21<366nLH*V?^`d|YrL{?pHvAM@-JOb`KReI00b+Q8c9K@KT`7~(FeL(;;!ko1u
zMwJE^ik}CJcOGl)2>@;hwKv{3f56Qx<bE1+6Aekg&sYj9ZBl*MSHnS#5f@Y`Ydzbl
z`GCiYud_uqK$6m8kBzS`T$Y8|=2w@F&~I^B);9~rKoiUZ>H8+g7ximZEC(T^KL@X<
zrtDDY_IuhL6xM^F490mfN2W?nBBDU`n3NHGJPk3T5L^6FhN-JN@!r#hyxLu{4J?lY
zPkX*4iIS&Nhp-8N$$Y#Wl@6Qx&2);qN5an(Vkk4()a=)kwL)~N;TyPmD-~ZB538gZ
zjX{unfW#vL_o#%#-s|Obr#<aVE^+(HS4XKx_ev5{AgL}IA}f7m<i=d&Aw>vccG4BW
zfx$1)d;_=ks=SoJ)yZ=H`%u4v5PAx0O2i?wLi`Bu0Nyr6{3cqyF2)bX@$9fzdj?a&
zyek|nq#<<HMJdM&1KXQnw79dOh{`cX07xuyJTSRdz*EZb<Cfj?CJDtXf26OyYI~Xf
zOH&F$Uh<A?RXRtis5Z0jo9RKCPt_6Mta30MyIB7QN>z5heO6~hlq|u`G-M8q8%Jeq
zgv}bc0Q<7)LjqT4h5_}d=uAcvt7}aG$athHqA6BG>yM@)S7;9H38YEwzCh6Rn<YXR
zRc5ebbJANwywsTWF(5cv9(nEXwWG=|pP?a5DdUOs1+HCuL4N;Rp&dLE<_n!`FNI?R
zBa{e7ay3~i#bc!&mFu<Y+3m%>0|><&zHjFHV(jr+J54^-GI1QJe59B&4eX9Q&Zy4#
zNW>w8!T%XbRN><I;tjE%mcKOa8Vii>qGO4^?3DS)bn|+66&p^+euc0-i5?CM+f#)T
z`;RhLpt?$|I^*a2;Y2Ft5pz%aKw$6<fCQyKLo^(Qe9heF<CIWnpN?y58r-uT(wzIu
z;3}5Kk0Rfi`hN0?vL`H;?#@0<@-X94RDhD<%QOf7r2@gDYv=fD8bJ+uZ)=8anAsc-
z3`h{{dX8^<AH0uRQm}U#^uNaX{)s*K8r@5oPevs1Nqpfu41tu#$hto*7tFit;t*M-
z$SD8z->5R*d2kD9(x4g|jC<usdAwE{WA3Cf>So-qZ>;nhO$jD;<y`g^L=naV1(ECM
zSbnTwiSa--53cP}D5ZQk*M|zk_0+qmG?JFP?ZRT(ImQYe<<g@fOp@>~0o_02Q`(KQ
zzx$=nEK*MnV5)izcOLsm4;%CY|9|f#nqEGSPq2G4P%%w|?S)Xd{*U%2VQzi$qn@{E
z8g3@ZWD6}AYqjx2yV9J#U!hojyL+jBj`LIOiZ;fU^u~-P<=66rh(YVlr5Wz6Dk4T>
z{5hS=0|=E7&4_>p|IY76yEgn@704uG>z5jtKZ0guDO>qQ+tqV2c}G1nL*ttap~&g7
zLE+UqG!Bo-QWKDLnjuLTXWG)=Xl`wT2c(?ptU2W!{?~2Bg(u=P6UKy(J}x-wYKWOz
zGUde&Pf1jTAT8-Nop(azl_${=4!~L^1lp~DX-x<|R@y1!>sV#kO*nZstoZ7|f*-mZ
z8HYCxsnZy3_P*0>Z0Avw9!dVAPo^<WDpsF<QqrUG=2S_Ns>T|RT0jnt@+i;U4p{fX
zXw8()@H-yIx;5(s9ZiE<WF5RKu$-1jo^n{u$p8-H?GRP82*12@6j8YGz{m^Y`9-HD
zpRL;k`nSjv_vkfmC1I}JSlgU9lQd95!Oo;nB6HtW@%wp<9vW+`^WcS6vaaD+y?7&D
zp76?;Gb}s#^kM}*9n$oPFljSiuX|w&59+K1PcAmmk)Wh%CaXIlbHOAHxT(!<mC(ry
z+%YnRwr5e>B5hG*LFFA>Rp``^8BU`dZr5oSO)6%$ZX_}6p4Sdts3qrE=ISxwzf75!
zBN0SM&yJ&6gH=YC;#rBGr^h99Ukbi{1H<N&O>(9jI)}rs$1I&4ZY_o0mXag8GOM4A
zSD&<qFm)S@(upUr%6r_sXjR2Wr8N*Z6|-2}H|^}RQ??k^o`dj`z?Y?Ep;`5*>B-Fr
zuJz;&=bxgQkoi;593s{fZZ)om59?l80snP~UB&4@{Wr!zq-@d^XA&d_YLUy%f8OQ!
zz%^K?C&7#u5tYwv);wr-omu2AGTXogbXw=Ekvc@8S8*L|?H@7qF`?6q)vf!T>9t<_
zOkbhYwi|_3A{ws@VcyB8^q{R0a4-m3_6{?~M)pc1|5HZI`*J=a<I4C1DZLb?)Ht{K
zRZ}H$l%JGxHx`+TkW)aX9-|6px9*)iP0QGY*n7CR^QsCc62fm=2OG^;lKD_Tty$*D
zpd17OI^-^uZN^+VTI|=QdZ}K&D9UI-^9p+WSi@iH$qYf(+|PN^5*cjcruU9%&|dVK
zkBAX7yrMHp7)1}G>Epvx@HKtVw(3|#-ay4ExN5*#YjS8qqK~WIz*Ecg=4sTK;ASbF
zUzRAoj)<de;VQ2~SgaD>XUBi!F%rlf_u-)Y3E3Q-b`z8`qf}+iY3bFmewnp87{DGf
zmR<O1cct@CzS#$@jE`xXkYIxyMYc(NI1UD{;9?$iinYA@p=>#4OgDiy!G!jC_|4zq
zQ8`JgUbk?110D&KfXeofYDu)$@--5oSGaORU>fJq7+h<ZVz&0R?qA4_zA!F`b$Uxp
zSlvS8(XI56p+*YYE&Exm#dx{!kn;~u(Dn=WqjdsGH~FC(DL(!-1qMeRLezw&uvXAW
zL;Wr2M4W2)3E<DuQM5dc(T2wxE8>?6l=&}-ny@9)Q);AL_XiEE;qZhdvLGtL+e~3K
zL?tNdKe)(G3JOMcq_ycp%u87M*fC5jjR3B94Z3v|?oJFEms;336f@NmKW`EbBl)Os
z@E%}5QIY4Tv9YM@yZKIO4T3E5Nl$u?U%2b#J6Kb^m5fRtL)<j-;h*{~zZoKj;4@q@
z9QK*+$&2z5Mfz)+MRr8H_~=<7a_^s;zlmyb`gHEq+LzbqCHjGGQd}}2!o_;LUXsVU
zr}3#^<nOngV%G9wJx<p9d25k7>T8kXh)@(r(HUbUh9=G*cR<Mwzjgyt$tayy%fAoC
z$)VauU&q3A9I(TOY?v7sqPbert}oe>cmpwXVn4$C{qEYj!U5Lu3}GSkAhi?BY_4J~
zTVl=3I#-xr_2^JUFO94$a-j@XFw*?$;rUJrVP>b7-pGo$*T|rGHdGrWwK#a*AC?Tn
zWsI{h$$`9vraH%`$kqEKHi&(fbGj|waVXooVK9)kUgIu)7zGm?qw8@27m?8wX6?_#
z{wViI!N7T?B*Yzh02s?GGy*3bf)W@<-By?EsN!grV9dN+{Epv`sE$RHxmeRlOUIgE
zZn}wb?;8KT(<IoB(Y)bRf-^(-NmE<NS!<ij+_ls<9`j-OJ0?GHtu;*lj6BA=5KNl0
zz0f=N>gE=!g87X3^BW*09ZCG)xlKFK)l?*iS4(acsoU8bcabk>G`yM-{RkgqOnKUD
zZt=#gpH;r;!5Z<;zjOgqr|9xFV3VcJeTEA?7&OcE{@bM=kmC%QVN-Ha1t+e~mn3%?
zB!}QS5ghUfYHcwy4817KXbD_9l><%1$bBI!l7#?Eu7EuJSHILpCoeU&X0UAYs;Ui?
zB=*_REFIOmPVqf?y_vxn4JP#D*P+ssBF5cZnBIZ}8ODxPLqkLW+e2UVH5+1ci4^7a
zFL;%ae!w6-y);>!Nx!J>J+t!R)ok$WOF~Q(Y*h)qK#mfGx~W#AE;kXqX(A1j6}A}q
zayRAveg=UrnXgmV_&fGl^&qyAW$x9K29$H`{>zT=lk$m`fh#`;BNS5}-NLyruzLeL
zfA5>h;eTxl*Pm?1+2QDhb)>%&QIqm^m`h}k?gi(bf>Vw{nUN!tj_Tq>{|(RAnPHz<
zi|j`%t&?-36+*1qOhTzfhP#qG%61G#>9T(`U@jjZZf=X&4<M@0w#hF%hY9Dd)yHQL
z3Mw==IqpJ)QIb$E&=ucL7L1Bl7+@V_5G2Qh(3PCNcRdyg1@4K`vWx9y)KjmnMx5T1
zuil<pWR$<L&;*^WCxp0u20+88zY>M9Hhp1AL*eLGp&wFCfoMStxoov&l6f9d&Z{v;
zdMEwLP?NE<(>2=}{uRd*_3&6&j-0M-h27|n5EWLZ1va|CCo(UnF+Hvb$HnT)4F5A6
z=Sdqjwn?G6to+8ifamD#r5{d|XP-<)F>ga#h_n^F+agu4lJg{#zyGjEOzYkx)TKQp
ztm-M>)Xri9e`giF!x;5<$&H!`@y(WvKbrwkz@f4%7IA-7>tsxQL26X`agnsOq0@8)
z+kn3wn;xTab@Q0c6;Iy}3TTYPh_Rr(#M2C!iP`!e_Wc#MBCh82g)@mX_lm@JOvd#$
z-GjDoElTEIz}u>=t0986XpZr0fQ3&-0^qkA&+kvwf3WUYA{DJ3k?U4uYRL+&jEu>|
zt|@m&^FY{`h<FT5b}rz2?j>mruDl)L$w~40GE70v?gHi<LP|<|*4A2apfrnrdDu_Z
z9M^j2NZs!o^u+0ww`utmpOhj3(yed3&W2_$pn(qhbD*P%My!de8-F^_;U}mp&}Xhp
zbCaxQv?yrmd8k<^iZo`OF;4l6IhM+iv9yq`(QzEPk{Wph<6tX;L2|qa)gQ&4iB8(K
zPjxo?f4xWF&AgLi#Pi%WdH7Y(;NwJ@IefjvS^owtsxA;$7S++}^U@d`h+|}QnjB`g
zZks*TIbqk07G6tjAXdi)jk(4R++h>V=mb4Kv3Rb`ufo|5E_(_N@_wWbm}#~$4vdc0
zrgq#4lv3n^YqN>HhZrhr*@@PBF#Q@e;_71wEk`_i8B+U`0G2Vr4`+TI{5pW#YFI<j
zdHU7aRInv?#?8!VeUm$}Qm1Ye<aeIC3fXwqkn7X?P;9TkL?*to<BqDk*2y$Ub_#x<
z@DJ9yw(<4dk4k>o$=EM~Rj}eEebR|^Y~X2gooM73dQPZ)TWp<s$Rl#E2#o$jswpFQ
z`ZPSIaM%e1I}fK2^}+^m9yMLhNJ0xE2J$!iNQ9d-rBmt{?$sls5{;eruo7g;;(mPL
z?6C|{2Ey_p^)Rr&LR2cFNQrc#b`OT%r8f}BF|P_u4aEp=MyTMpLubi>d`6*>h=L_I
z@xfso-y^o<yBkHXa90()I5*ZX5?(>&{yFtc(W1vhhf7#a58pM8clBj^fJ-HXTd1(R
zaIXJA%{=0c#%WIGhrr$PWTD1KiW_CoUSG5o1pFOB5_ckF9Vnho7;9TunifpBzZ0!h
ztahhLS2}a!_+Y}_TRP~*cCn!yjLxMIj^Xr`NGK6_jX$1kD97s-kbaoZT+NXt0~JZr
zOD1{QuAi?S#r_k($zw!fnpt~4>`vrr$WuJoua*S=Z0!t`&E$6VtSB}`pn7}h?Dm5#
z8R+%4$fVdYq?0kCY-oNI>t>Z<)a)G!!Ui0Bl1K`@^Q_?q{aI3dzfu}--33Unpx?4i
z!M>1-R*~kSib^yQQhf0EYNjVJF>Btv{W&8N%xvcO0G81x{Qw<W!frc<bgLUkSCCyz
z5f2(s%`-Ndi0@6`=Z32sL(X5afsi59(>qil!z@nCneDcpU}yYj=^c{aywXG+O~Rvz
z`*f>*C<zywk|vi?rC13i57W9A%HAQXOfbqTP7Ek|1Pkis4s({si8M`S17CzP!pLg}
z9UY_+j`XIxY)qh0+YEtR&+^_<c5+2=Jc@k?fDn|-4XcFdm{%&DZqR^&_(}5-^MDuX
zbFGgYMVzI#X!f(UoC!m{M17yl(yV%&F$Hg6Ed+}eR6eR8I4?F#XGDGR5&}SSp4}2R
z3So@w5TGI8K2w6HlL&mg;^fTju1croT={0oyS#-H!-k=LJ^3`F`n`{;$#7cK>w(YO
zC^&r|H|*}h6~e>sV(lK&a+xabrd76E1-yy|=~(B4M&;+hTpKi?NUaU0r}Z>JAr4#r
zy>G?A6%)~5tmRmkcK54<*}l%*MYz-`;A#D&S4X|FB#2#k6S)M{)X4NIPYLb}#y}~W
z?~WlHT=+j5Upg0WcHRqs_B+l%>VIaUf^EHuL?7IkJ2NFfp?u9?ym|$??kQ0GPhF*7
z2GSeN8qCdxe%DE@T=)J7W|782bv%8*)nPQR=LUNIbLrg%H@`b^%nL&0TW@Ea_A(+{
zwlXc71qB!U4{DJnxw}<NVwkzNg%NsB<P;lS;!+~iTL$cdT^M;_?VlWRoi~!j@act5
zE^1#{D9jM(EgJcm5^#n|U5B;&GVo)QHKl#Eg+J~o>7Rr18TEn-xdOaQy~Ruk>Uk>C
zV4l#5ginN&gf&W&L1Sx;3B#F^V!ql2RI?BEi@MmIO4hWqDK|}a=d|pOpzeXtx`;d>
z1U<|1rEiRkv!2~t{jGz%753Pc+`UFjXjXA5z=*)Wvjy2^9F<aNjK7c4ZbxcUhn7hh
zzx{pRI@G=m!AIcF$PFXWFn}B*&YpKqRr<^fR2=QQ3{K}3yYUB%O-X%}%#^C&dLP$S
zP8<Vz(hX}`^4L)}fT2WMr9^O<a$s0&RkM?-78+?48gD4tme`kf%#eL5C{!s+oO?lR
zGu_M(unjZ<#w5L79nUo#k8LTdRuZ{O+?)tA#k@RC;o3~UY!*Vwez-@;h5^su<T;)x
zhP!qif-F$-yQm52Nl8so;D+{7PrFFMZ8>3i{Dk^@@so>OSW!NAj1TL(XB=7PLv?rE
z+*fI~alOP7ud66q&`s=1!XJ#gmoE)1Vp8t=Ma!vmCv*~AWGm3)tqy?bx33~JRw7BS
z*~Vqpi-BK>7boTx4Y5Sh9pjMDp+BsyDWAB+bE0*qjbq*+We&D#>e8U1q#C-4Gq~kE
zVlNf>jeZ-Wn-?ELE3^c!CqiX3hCDFkw9PtmTX;JL&W$|fcV|{=Oym=h?dIeSy1Db@
zFRWjdBnh}IZH0y)M`SEb?yZdoC*?aF#WgUyQ{xP5=Lm<Nj#bd@YohUX0!$G3mBUHf
zUOV_54X}jw44LFuOU~4qOlvco1u9r}056{4VQP~od?*RzkA~1ZyP5PweoyYvt@b31
zOml&BjK15P6O&#fhD$MrxUoq2qiz7J2A`DmYodX?>mZhlO@Ut6j#)ELw20dOo$<~a
z_-}*pe-)`2*d876VtqMTl%y1NrOMWXEn{$oe(UB1LuhJ(RwzaD8;iM8^i|=cD;`q7
zC=9{7MdznKAidyk!792Fur$1T7Wp)O9}-F-7B8pTi4GmrTokFGyP-kJw*KK4QIU9P
zelfQOiV;TN8te5L_gIW!=Ppxg^4D0osEZ6ab*t~H&T84TGm<FYM=b)NO63g7|CF3k
z{rc>|)Y62|3yZrM4m&t3cR1U5)+zCfTVNH3u^WO-#`TqK)b65Ub+H0BFK5g(gVHe{
z8Q1(ol6bt`pE0`=+IIq2P(!dbaYGW=Q73+Q3-NS!ZMsLP%i&MpD?k}dh?%Ma5(-p~
z`;(yiw~VHptl`4}3&^m=QeHXDg^&z-^0Xd(l`C|Jf83P*c8vKIL`bJHGW2NUuu6O+
zeGCDSu1jU92J^rixz9^J5#yd7WB%oH*4~J!H<|DEJ_l4pv#m04jVVdjQq~Bh`Akp5
zCE(#)ow^oRTJFI2oQw)G#6~dL5AI+kqFN3qYw06VFGZkP<Zs?L`+jDOAO{w`iIjvd
z#~60TK*WSz+P9Qt7KI;Eh27#mL0&N_zn_K#ijUK3suyh_D)OvyhcEd_a5wL)Tz|Or
zN}2hr_M7(->K?llWQJ|P7E9vQHGMZT##Qt_r`Y7kdDHVom8WIQTcb6gJL{59d+k7#
zDdAC*+xQ}A_Jsww_dTB3I#B8h2r2{jBp(SyN0sNh5x+F#+KFr_egeC>EPv7(2A0Pr
zp;lq}WsXz1i4p%o2pT0O(^clSU!{z8iPFxZw(zEPneo}UNHRT9{v+R}{Ev-mE@0(j
zLOH*G+%n`;_c3j?&T489QVoo_&F;<Dyc>$<U=wZK80fv5{~LRaeUF2mudkn73HOKa
zidrS5@uW+izWMmEftW%$2156>vZqpLmOWqe0i;Y`<bfu+Ac^f7$EHC>2E*(#wy_11
z-RzY8jj>VRKSEOxwW)&7-R+AB_n(UD`zY{vOB<I^!E8t(pEDq-S3{6i6z{2yPagwQ
zoJ0(tTTnpm^}?Yff%@`pVJcGg*0LW_`(7m2w&^<=lyN8GEH6qSUX$qQpp<T>gRziH
zF`<_~h2ok!NQ@f9!LXz0;YCu8q?JjIPj!8oZ3FXHM$%X9i5nRDs4DM>An_V;OvpKJ
z99oIgjA~!nWgk=yPighO;UR?F7!kLgc|edm87;fX-!(?kBunS6j(&X&=}BZ?`E2wf
zAtOOYeH|U%38Y0O<JlX!AEb??`!_($<?g7smuOPr#RIV+5B`9K8JLb=MrysP{WNcy
z&r5QOtooZK!ehwph<Sva>RzSdQ!o#BUir}LChDl)YY)R956Na{P<!a?Lt9zv_!S)r
zV807tc!AL>$C(<OJQX!^Q!FkX|CX{PeOpHROGBE(=+l57bh7K?&6A#(z{27&_3}@D
z6p!aB?$}L+y$sl3HG5{jbn~NxQS6R6E9W`#XtQXJ08Kly?K!8Fgd?v2vy}hNU{%i}
z-SAaaY#>jYh4*8;pD{N^qM3IZ5p=&KjcfhogO*M0(+5oR%BDb}dg>O{-Mvbrj|+xY
zhPY<w{~1l^vlG>?(BPR&e#_)0cf!Zxf;nW|LmWszqG>qkcN*UF)AEl&%2Z$UFccD+
zw>I7=Ja>uP`fWuTr=aY?Ff3w4=Ss~1l>A_O<|Xt;&>J~sZ0(8V5R#@b+~_I7Fl#Y8
zKerPL-0wgPynMrb+JY)pH(zTfU8%vvGa^{0Agxg=>~UKGfF4U6Y0CcVzS?UjyBSil
zZ{oa8k_04s_LB7ou|!m1xv(sSV1a&_O@R$R2U~+&=qED^K|y8ncWQEIra*2U5?DQ_
z5}Jg-$K1=r-w&rBfiUDdI!^E*d|hzGlL}(DuY{l37~zu}WTwenNoJoRe8b$-M*rGG
z?Jq4BpL&?m)Cp!40MTuuqgJp;G5c?$s{?bI9Sj$vYSwTwy7d>cFHBKwAEqo2ebVXi
zuA*CJ22GW&*`jH=od*5y22{JEhwXGSP)e^bZ<`*&rs<nIlMHapm0mFY#Z!2yh%c^G
z2vfVG)M~t`x_)iz|M_apmj7VI9wSG@?wH7>)D-GHV5&EHrR0e^zyRNT`7k|$W)>#g
z&_wRO8YHK<k)NUIKVF^Kz!i8VkL_|=DC|QLNC%P`@d=T*NVT(0-53pKfvuIm2g~ii
zF>@mCvDW3sy4<J0rDduppLuA^+IGqEc*pmR`wf)gM!Bb|&1mPx=Mp}ZA-!~UE`vP`
z<nq^(E4DBOUr@F4NCOxD3JLQ8Qzx@~Bn1Sa0f<<iAxqc+fkUpQTHrH1942NdVkB`2
zp-B%@7UF}xa<KNH{LfGDSCyph(G-UJ*RAMKRMG&nqSYf}5I6&KJ(gR#A9)`jqXV<c
z!;Ao?K}4Qm^h(D*C70bZjjEm_!yI)7I_DvOpK1DYXZi&mljf(N(3E8Mi(L5Ju$In3
z|3?%0xp}<wSc(r-d`&D)l9}zWt>fi*wg*y3vG#yIQVc}HV9Ukqsk>=iz57HS<m2ag
z!Hi997QY->93Iqi^V++S!nRo38MA~JXeXuAbhjtlkltUJ{uMi&rjzEZ?bt2ptIO<F
zGkfR$yYB=gUoBH2?4pK;IM*>2dm1(tC#WXluI%qGEi0q!tZ8-lS)c5W+K|pIXSVvS
zWuBRMZ3gCiz?zq(*QM5?%*WqPPofn*?bMzbPl=>$m+H!CQg-`bj6-I_0my)*Y*tf;
z{z3c)WN^S|AZ;kTZt+voTYw_RvLKBFjuc}!OJ*lS2A0w$Q{Y2Be)|;nsh46durzWN
z05*~;UwRmDdgE|p<5X=9=~E{wlU?UNzw<D=p=C~hcgj{NmJ)a>`7wEbAEO~>*_or9
zjKjh~tTp>PXg=@5z$6!0W#X>-igUB*690{_@VFnOcKbs;S1K4F7LF>o-lQYL%#TxA
zC*t#QM+oug2trAxz#%&l&B9RsN@|Jf`LC}+;p!p%Z}F|w8fdS(sP}9TEaPM`$8dAW
z=2<_G?XOI@>Jq~q)nrPZY^_JhHvY|+B3gEi{s+qfg?$DG%j-R(O`Tnn^<HW<q9upt
z4}Lz4T5lt2GDdyEg?%8MPv(fxM9@7G;I-hQ=J7TU$h+DR=lE-sds9SAOg2vRsu~-d
zr^W*KloX24sl37!Mj~Z<DYxC%XQur=TMd)$zWy#muowv$k#f`*_#S78BD8*-X`yq=
z1*d`%bKNS7Yxb9v)+3QY`<MW2l%>_vo-u+NbI~vB12$W1Xd2}7hz=?7TY>9SpQPAe
z&SXp6<N6aO+z^=4#Ml~{Ov6_yJJlC|H2VWBFiA~K#DMOmR~cHLOG*vZ^rrcyUy4gf
zTc8?lM#`|59bzO_O8mrEkY(ht*|+vhS+3T-D;Z(cPG;Sz#WU0jQMH_pM2@6Of0e(^
z(EN)yJ89mt%*zwP0Q{9_MP0r>A<64mt(k#7-BJ;8_oleXypO&g$1EXFOUFH)DkgQ`
zBs{PA6|M`z;kF>-l)9Z=v|q-Eimwo|$?w^lVjZ8B+;-lQDs7^#`u@+T8$;)^-aFyL
zHT)}rRbM&x(c&ttnOYzWrdM|V%CJSKDeSE^56Olo4nJ)`4#@&%BYV`kZCkEX8L(3M
zP>jdHWdH+ar~U8^9P)I-+~rwKNactng5|?7oy(Gd-47@kjN3QfpuU2`Ja-d2HFjG5
za_%IWzA3$GIHcn;&b0zrtf|3u;#e$&IoxTqrax5^b^z{>2Q&;d#$hSO)LkjG^|30~
zM)BH6{Beeq`f)imytK^5uh&=%jK6#5vFtE^qCF@ESJD)kxw_|QLAp0GY~;6_5-d!~
zx`pKE_wTc}>75Hc`5MsDAZ@L308*=p3S4zYJ07SCxZm#>v7|Jz)s&58iESJpFzFa>
z4<QT|T4mgh_6w&V++VSjTI8&)RR?4>>(+&;2h;aV2xdClB49;9R)9e<TV=N|<g|^=
zHEDLymE8r22zl4(GS+b#X`D(O)>39oFruXi(ckb-so}9{ceFpE=WEeFdHfu0(xVLy
zZ(wG)Zm(PI8QzNt&Ap1A)3s8FP^4YpSUqTK@Gp|Og_>5N-DIGf5TAFbaeLAmhDY4}
z!1$_<6Uorf``^0PY(FK^unkMKc~~oo8);`B7;hcdhvn=_<z9&ERs-i=uDNshw4O=|
z%cO+okpD5pONdBeRUKL?i!{k-;B~X7kx;pQm-hawq%NyO#5I$Z=#(T?f>9+J=MGY+
zz|}W)WB@?6&RDN58^J*OZ71w`6Uyi`ct{KLiNt~k7@wUjuE*<$sK!J=`kSAtmpqI@
zNCN*3e8%1=A~{w3%zEHtM!={9Y-2Z*)zKMcf*H{ShoQ60xjMo|r>SI8a}WWWcOu<1
z&PWp1Vjlt$uAtdwIpqWa`ldFe1e?$8A6w1038xCw79<iXse4RVX`=}yRh`Emc%{oF
zUA@wp_9Rg4tC`tR1k0*;%Vmi_4-;5=Z!PfS!$hf^D@h9l>cyUiG^hGZR0d{Din=-S
zxGg;E3OHc9%g)dg8*4i?ih`6oWc12~J3FqU%__Sf!f;hiwHk7<MN8YVF!wBF0jow%
zB*$Hf#P=dPmJjdYtwI{rD3-a9(KZve4k3+C)`~P98)@C=zqgW!BjVPc2{Ew~HiDX5
zeRG2p6XLd`imrsHzgp%kx1uQEL;hmY=pllBxk~vuinLLc8fLy8Oxdhp1!#wFY)5(6
zG>~LaQv${`LGqk<5NXTUB%q9L7D%|w;2hF$tP`4-xrCzGN7hpVf6N4DPnor&SNsCH
z)PbYF*wCu%ZspJj?j7d&Xi~M05bvJ+CM1PNAl-qwb*2wD@`qZ6=;-L3#aqz{5s^~K
z3aX!xM7ANHP|`<9!APYo`dvufi+j*xe3TgG#cRVoft~onqyhIesCw1~0Y9ZxH0iNw
zoM^lDWzrGf%~$S9sL{g1Q1o-;7xR0W(@61p?Muj*3n${NOt%lxSUs|O_}~eKZ*cyQ
zC(1p8Qt)BrAKb^8dchYNTxD+_+(w}KJ~+Ex3clfLxlH!d{XVTt<HE8-AZtmAL|&XM
zDZ62M9!%~cS2ZTgVWH_$-V?*Lb=C3cd5HuX7{(Uu({D$uW8RV>eHbe?XU(Rg$RCF|
zg(#skm_p0)xzJ4tUB^P%;fDzYtP^+}Y;M!Mm2H_n@HTihl(7W8X{Sz(qclW%$W0{D
z{Sp0XA)&LV7$sb4Ga^N^HnE9Pf}I83dtarRzv@{kD3y_}Osz#L?vE==t3yYeVvZCy
zRG^JTU+9fx5|9)bemRV7{GoU|kzGUosHq+k;|%E?4sUebg*diJ1YK=3$}X)+7w>XZ
z=zrJ#BO+`hKr+!aC?JXL^nGB-r76khDHo=<73&V<NcAVDG9V8ss>829@Y#Fw%y4Pt
zT|TlDCWfWwJ^`1h#vdCPZ--%c%+6$-2nE0K2%5rI$uRc!kDk3ar{kE8GGQ`uOjfb@
zh}U*?_4Z+xAYu;M{YjqQePlGd>0?zySI;lQ<eVzb`5a<@_93SfZ$^5NG}n|!)?pbl
zJ4`f^K{c68|DG;!#f1pnRMfjnl<+ah(8_=*Y5x}OxD(Srtv_(Lq`4=2p~#oF@nX)U
zKGkeqdWx_qfBI0;`NnZ=ZM1^ZRoGh=4QA52@2zeq2GuI?7W@`aQYo+xn1-6Z9dxVF
zdix<#e)Yk&Uy4)qZ^5q8KSQj;?^Fkhx|+)RO$Y`m1|OaB`x~;T(GLyD2-xdU7QA<+
z@~_n;M7JY!qy)51?dTu17<?trI}_QC@r%3!p7u_?V)LjfI|u6fuW#M0|2lisi>8Up
zYhz}brvAd-wQ~0j$y{5#^e!+`yl?a-*uVBG9*39C-4KhS=6{OtOjy%PJJX^`(Fcy|
z6VSnk)Z_-KN7t?K!)J88dPl0P>0qD;hq#x#GUV1>G7x+5B7928tsSMxuXq<eaQR|T
zy6WD+Z3{o>ANwLPSdHuVO}9r%x4zc?&_2Bn8mnyIBrwE?>cloy*RYM%HLLX5e{+8G
zafnnzoG?t$Ls&CF2qAGi4izU4!59NyGip>vbda+ehcWv~#ymu3PYF+rE4djnN*eej
zy4Zvk^2kBoi3+EJj{3NgM4bo?(UD`A&$r>Wh5vtyy#-VhY`6C<C8g9zcL*Zg-7N@;
zO2+_F(k0y>;DCT24Bep8At23=(j^Vj-6`pJ&3)g`^PcxP=RNCNYZeQ^1;fmB?fCD#
ze>)(Csnl7~qq|f)&FpuuPCRDuNBpDCaV4t0{q=RV6SM1HY;U_!=_Ve;I*OdVy+>GT
z66I&QNSWV=4V`-nKklSe1mt#v#54%N>62XfShI=!KY_g|u@Nqj?_!*|ndKHzz6O{C
z{=hDNz>m1+7EvcH@lmNRlei*~R~5dhBOTwK*~6rQL-vi8FP#y3$86-DnGX*1{+cHp
z-qu4qv?FSHv9esrH=!luA3qn4QKPP)_2foQ#&wAecrxJ0_H=A?(XKurH#2ysv7l-B
zs7tu*^~Z(>dSY3Py}g;_w(x)jO0{>~v3#uTKlscUH3ApG(m8<<=E>(>bgxXrWe1AD
zM2FV-*Lz*vB=bsdO=9S!AFH3H2**IWODgW5;F__yC0EC;q-tHxwBxq3MjnP#h{}lN
z8p%HCpi>mJrmQ0M#N(FZy(_V0`-;lIAvwCbqLavNLW=oOH~xi!5jQJrC#+0SAL@*5
z@V;}eWc5{gzSH2B83OfHld&TP4HLmbS&96o(gZ>1Pj9m4MoL!#>$rR5IUpR<JwpCJ
z2fhS%6ZPbA4mr?Kel9c*ymUV97fMlUYr%OQb*Vu6<jTT1CW0>VBeI|$URf;pOpTNm
z)0VDM4j%k@;Psjsemq95UXPhhK%_=J*Ii~*`a;ISZPW_<3fZ*l5^w)m?;`_i7>*U!
zNnTD!zB5B~J;)N5-+1APP<`hb#1$mh@bAh-Wd>d53$5t1H{yY3`Ma$tegnT;{u+qx
zqz@Rt6p41iVS=3Co;>XQxbb3RWh024Ea^vkYsvBOYu@T!{;ISgYQa_`C8{yOc*P=1
zHNi;!A0DI<K^-m|KRq`-3Ww7qkQg1@cg@SbP51X6S<m@6SN9}L{1|hW&ZHW%`%Mf9
zd7;d~_uosf`~K4f4G~j1!nFF_Z(A%7*@`4;_i?SS(`?Bvj*<38{k^b`l4k#Cx5#oK
zs9gQ9Wr`arm?5H}3o;Eh$%+-PPacRijQ5J-?17_*-Fy3L2hA;zhxjdJ5A!E-^8tYs
zk<*+$!Egyu%6rQl&(Z35tW#Z;bEzFaSNzH4cYTP|Z*~LBmQS`zo13&a5ARR!G69bL
zytx_aiV$wN682=F^A*{usv#@m{q@5vi1eDJ376T-#e*_x8LIa5=cIKX>vOu3r$A?=
zhm3rueUurFBVu4sCzVI&Od$MVk_LuUXG-`r@~G`>=X1#XZ}r89LSUSA-L~()U(1)6
zQg*(3CX^T|L2%K_ps*s5RVW%;P#IB@Jnv9wneUNIqx5CQ4RFLN{PQ;+*c}f)@b@1i
zgkzynlzINU&5{bVVg=lFT=OziEPRA8dJxF)<Vy|5&ZM*AO6as5*+*QDYVPh}X4*Hu
zdNZD`H{{49+QEL1cwG~6Sr4_aNoaYELC5A-Z%q7opWJV!4;~mg*KDwcjhZsqjYubJ
zUT^C3Rfe<r4tBQ~tI2LU93@~94%(_RRG$^$95rEW|G4FsF7~p`o)GXfpJ9Z_1AX@y
zt%Br+T|w)aD0U$ON%HUgxn=SrgL}MsClYe{pK%%n$Mvf2S8Fg+Cx?`GiR8N;P@UHH
zwvb@kx@q3)aORljy3M~W>(AFxn;tK)wYjIx+L)W>qeh`RqCjbSr9XnG#;_GBlRqlZ
zS%-NlH?y*V5>PNNy+gaIWEci{ncDcF-TWsd%o)z$b7i^5gI`5k-D5tn?)vDzyPdZj
zy`m~yT3F1#uwu0FifGslIkMCrc-hPfS>fExd90>s^vc&|(yb>-RNhmQlXp$&X^D<+
zi4EI#svo_@9!9T(BE-I>Es0$?Y^lAmar2~?$W198d~x$2*4wJl9<T-FeP6RB<X=kj
z7`H1K%G!@8TD%}59;`cyO8v1dJtg8+P~!JCf}HmW4c9W5X`9$qQi^JhNPW~fr?p42
zY;#08rEv9eczi3b;fd17r|Ep=E-!OM>Vv;$bq!>tAp_OlK6fvF(r^E1XETEVHXugw
ze8bA)s&lG-yeCS_P@<Fk)wQuHWd1Y<T+%sN^<27{vnuv9137IJm)QQhG_f@@^(=_t
zmd0z;8>Y<%7r8=r2ln~Y*C~(o;*9siXx(56yW#{21DiKL&@7C_U);#aH#BizvE2B&
zLdqWS^i_L|F<{Q%JjZpm_15HIt}3>init&UVtsl&j#~FQv?6acK7IKSgYE-zMD>Th
z(gK&QPb6gd8E8tKv$hfv5ykJxQ{J;vC@_)zw)aEv?08$a5UMT_(c%&Ld%OBE$@5Cr
zLK9b&YP#_c7~eK2&0&g1o$$cj4+<pT%Ecz+pYYEZ8B&t8vcHeFG#?P-V-1&n{X1dG
z3bwiPWUu<m%ybk}Pq0l<n|?ZNapY&ASIZ58r#SiqtE?KEXuO%>x<4JNX~q%yl{Q)u
zZEh<B7i4aQrmu&Hre1H^s<{11Q8_0Q(zaVoFQ{bup0b55s`rY3es;zF=Z$M)gi+kQ
zo%aK4E+a58nKv9j{I<MG!tvFg`TVG(MShKk>kx}=_NS4Y*>P`7Y{s&a1C`3!Y*5_3
z?8VbCOW8hGqm;pag1p*G%ai$PGh0cXqRFWRb#uA2tKN3zf7FIqr`K|@&u%|3a<ifz
zBrN`ysx4KZCi+M~kunv#?<Cg%7JAjzz}@PjX{{uhy94+9R-IUQWA{uh2|hF6sT;q)
z?)_(`x$=HsoQC8SbNIKPbGE8=1^<K^nP})M*bjR~-#)Y^SbA5DsN#W=|0oJ(nq=ZI
z>l6Gx=G_YSM=b(}jh{K7iXKn4`lI$k%45EAJ!1H5m^4f2i@azcfyH;Z6b6q$l(k3g
zER)|X(ZN{0@kx(Whdrl}S)TXaw}Z>c26v-=(WL42o`7Kl6=Q0~HXIN;occi$8y@Rl
zzxrJ<Bu{gyy8RILoK97GF9qF+K$GEOY=E))Sy|!%BBJ`W;9No-FC-$E;B^aDh3+d*
zx7eS^Yj3M?*AZsbh<;ulHi_V!#$8eI#NzUvEwS#kEaaPZd)PziwdyhpHDMjpctsW_
zJo#zX?nRC6Pyt5}W4obeIYmIk(U{hr1i2vg5?wBnXOU3Sz1UjL19sda`l&S(Q&Mzg
z?qWTUUR1v$l`D!+VP$H7D3DkNQ{fIP7RtqyB9Tw1m_WJRuzMDIU$>Io3g#6@Oi&#b
z%Iwu&Calz0YACv-rEm9Y+C5izMR(%pE5~cF7S{~Kj`G&&IJ&G3-qjlo`)HS~LDsLc
z8*r4uFFFFKX)cPX)v~J-sdYS5q#knxGwaibu>h<@I)FdEhnf95pqx(-p42)Ue2|uH
zCz%Z0-xo%WlLa&LakAsfZvO&KtldGimew+NOhwa8h3K|GLl(nYnbCgT7n&^_w+cTb
z3Pv}=5(nL&U^><4`eysD1Jm~w{ot}?g)6RIA?M6?Fs0zvKO;B%=yWl^z%Jj&Y5Ot<
z6SBCQ7%aHK&$l*$Nx0-#8s8)7>v23z+T!pa%hSA-rBY95Fb}a(;>Zr3<Z!rNDa#nH
zpb=rni`*XEkxlkHipw}D$@7k>?U5z5Rz8&V%QiTu3L~8rw}Lg)f<C@l(DJEca1zEM
zGxDR>?{BJ}*XEzZ?kmUzT5mR{ZO=J)D+ExnAsf?rmH7VR`e`0Z5l&!57HoRD!u5!n
zNcZssC_XoWkbiJNLR~*&e4K5h22iziyj1vBsv0C$Ce@Zw9iH$1q4jn^;ZkzD{gaX1
z@lY8XBPe94HTVH54b!t+Wf%q<NrM$Ofp{wpI%`K({6a}9s}a9&Ao`;YoIKeq?mz1{
z*G-mxDqSqk=jW0d-5riZN+%lY#hlkt&xezalhz|v2U(UB(Bn4krko#(aey`VB|7Y?
z`Wd2N1l(XJj5rsp24+96$fZ>c6hF@D8x0cex%$D9q1Lxr)fFu*tZ!+Wn+hv?#9^9s
zUkdbW{Qy$w=8|F$mW{V(wg6dP(ofSLXAS%@>hdXLVwbq=*<I4U6p}u*=h|y}bKAJ%
z)4iOe;yh4=IG<#;-miXfRqdA*fGS+86W5T-k;uE=`Gm!@<*{EMnknz}!)%Drn8K_n
z=SrG3O)uO*(nLXD!PeGZvN3WadJCgORhRKmM_dP-$4vBO*JEsf%*A&*yWi_jpg!95
z^^Eq8f4O=_30J(6VEoU0k}oq=S(Ns*^qaq${du7+0qq9nETl#rCP7^!(+IHbc;A=T
zSMZcsLs+NEs1*ZuOPqz@eHlCbHGxk)np?DUrzu#7yMDIW0z(Z)+5FY_s&JyKHbb13
zMfy7Qkz0O}g>zw2yhc&m;pEA>@A=&NvcQF+BSyyFPV=a>^_sWsU^&f5m_3+%{p6)P
z_}-NKh7&AOr?rbIkCE1;(=tyl1B{OE7~<)pE)r*Xa>JfaMN1fdeMwvimQ410NWIZa
zFq@oW>K}^%^Y_eBrhM2`tFAynV&t)i9w@-o$(>=E9`s3j1iH8D%ifP;6A|p=NMn~o
z%Kkf-ilF@i84tr07?U=b9q@<}%CLQuqaOb%zkh(Q((}!lJNb<vxv#%iA8Svy5}$dd
zi+m<Ax}S8e-~wBC0M<Vb>Cga)TD-FWCO8?#WQv{%5`pm6g=_~!rmW^(Tc%9#<E|~$
zb4XbjZtJ`(-dSc3*MRH5d^qk|04w||87j2YY9DXI+&BLj_YBH3nPP2JB7i^kA%P%s
z);5q~vTgrhAsF9_d{R!iIYE~W>+H<546~Wzdo=?w3Lyas;l>>Lj)`X(=5ho27I}F`
z4iqNRIy;A<?}}9orRuA2+#kr*<U(bjLuD8uK-rt=fBXc_Z6aHyFa$+}9rO$R`t6`U
ziOXz&jjU#Kqh#&#o!hk<(BIW(x7!f?h9^j<!#(`GtfL-k8Uo#W+4lK)&h^1?j?z`_
zYt$4d!+A@2h)LT!3<%F%NEb$B1BqJM${_9gwPo;>@t8GFY+5>I+wN8p5v#?l&MZk)
zrat%}vIhTMd8Es%3c<Wv$+>>+8xP}f#)M2_K=6Lku9}yq1UKFxyNgGqqY{j5?_(<@
z^D+9I`<ucZwB+6m+1eduuLEH<ACf*VO>TO5ZWLo30H|d<gb%U8u{hvXkKo(WWi6<<
zA$~WP7?6Yzz&7TZ#LbrUFAF__GI03MLp1W|;kXUZAWSZ8?~9>m0jICVu;oc5QK-26
zY%MgqI{CB*U8I%$zUN7wuCkEzW(ky`9E!X1=>jq#pxWQsjxo<{DymjXs3uOJHc2A%
z>C&2H3ju0Wd%GtW>lygW>8HOv_Y|o!ZJ&N#aC^9qu=vCYf7GAB2zTS>f<Ul5_18sW
zBSeBC-XQ?>VS87S3vWrtCc%is3X|FX)KZ1{*^#9hGyG_s*vbCKF(ZV`^JSk{lBH(N
z2z~U!D{Jph!kSk9VgXd0x^fe0)ZSihqp_<oQL53cy0wNp2o!GsyvWwsvr9fgh%Lq?
zUghVlQ@G$u{vpn=d*n-l?dhPV4Q2RvM{Y}SH=bY{7V#mC<6=QK3j6YLq7XQ;BHn>4
z@Nn*-Qlc8qQ>-dlh^1#ecVT)fI@rJ2UqSSI3#LnhYY{uV%_RQFVauwq#pU6M8$j=d
zfr`NNjEME(1;=UN$w5&}(>t$LmzIj)i1=Xaov+it;(ReX@#Z<8etV+C^S9B-#aN3p
zo3&{KOTvwmtZ3sUz|gA4iD9If_=IXSoB2<+5fjc>UxS6X!_*^wn_Qz(HEt@k+LfC>
zfbY4y9MnkGV*N?lUrnm@x9XLt5BIi>k<^#b6PJ^#J~TCMvd?Xi9pb73xK$7ct5y#w
z%^1g)jssS?MAy|i7|}z_F#0|UTrV9{nRYUIG`;U^*lAgb*i2n~npVL6MJS%pmu*+r
zfr>-4^hfUtzM0RAefk@gOOrozRHO#pLa;F*cQGOR7?A3_5Ow!jwPC?Gdmn?bFMZOx
z#5>ttw5BoKlD^|DheZ4d!Y*esvlXQWK%>+ewzbSEw>7;;|2}K0Q42z0c=zbKP@nIS
z?6Bx_tt3tV<Rr_r)i1Gj=BMuFCpQuj1p)fx7giz7ABY3Bh+-VeKDiRt{9|`cG*Q<+
zF#k0#@I#1R^4>yRn40*B=7tDr_^rq99?N^amnVmm#v~8=-bnvMEAmjgmVezX$4Nu=
z*i{OJhGHtEt#!VoNZbQv7m4Y3ww5c^y?q<M`5tL0=umgn12AXSvP+JR?rcWE=pr9t
zaRRnU@Y}(6BP@RzlV`jr?atx7gc5XHCiJNajwGKa^lUhqmE>bcx*`FeqZ>oSoGDnr
z(?we=0!Ic5K^yeF^Pa_2U-K~)eN#yIZ$#1P@3-jY;qF)p$0j6w2pCt>hSSfhF4)_(
zw_N8j+1BB)h-m|rcPS1{B6JPX@j=h-(X+tz;_qkshXihtcKtO@{(CvP%jz2PoE)}_
zzN8S)&fqIb_fA$Gj%Tsjmz3Pm8URHnB*b0D$|HDvc->;4_142rTF;X?1Wtcf{*DFt
zAnDJkcf!QwM6%svZ+^1Lqc@~JkuF2PVh(A9(`R}#LSAsgxQxf^5_Fev**6#^trWNu
zKXJNB_XS(A+!b!aSZsT)ME!}ac0s3yMrJhl43z<aK~iYF->?Dm2tNyL!Hj)g3>_~D
z`tINw$YhESKcEWYr`MBt*?NpK5)^<+PcKV|I*8T(H4Y~KrXyqiLpa%?qF{`1B@-DS
zCgG7jVVKS|q?^r=pv#r<A-%#2QO^xFdw;i`-okEl+aVIyiOfw5hv|4rRSHtaKm`k2
zcRDdSsSp4Bh*F+E9&HzJcD3(|>*p0==_WMs97k8kFW}I&a~Tr5$FX=M7kZGgAwHZb
zc=d5qxfnYCczB~=CJ#5qJW|nZ&8&<ur;g^z`Ng4GV<y^-=F4`k|EBUh%5f!HR9Z}7
zD47^9;Z><J<^|pT$T&wHRnNetRMGayK!Ty7ThEld{*}`_EbqR$%;<+$MOK61w1l$5
z=P#59D~3U*G-deJF{<e?QVhDff7y-XZG7U9-2eS{CSf9d)H=6W>YxS9MZtnfTd`S{
zw)HxMcQHwPH|9IhvcrQF9(72?vK9F;MRe_g)Ee`cz?P)15JT&vqNyK29$!Y`Pi<1W
zoae&bF#MBm`^l53{4pdw5#=$0h$g*dj|X@RqVf$tXq|@MCBcRVgfl$egl4;C7x6x%
z?(gHXSuHG_>D}4y*|_p{RwqpQ6dO&JD5;Rif~_n(8F%7lhxE1JVf6x4{MiCMnvm-h
z)SmBk9w%;Iioc7Q^2I@><XxJ1I`kE0-K-nz_8Rg^*4039{No4m(&)$sUGyC@DVIBI
zsGfnLOSKzGHO4c4e}~x!b=Hh-$uyQWmUI~fRukTi#r~nQrO~Wd;ire1wv=v}YV%lU
zqcs(5<Vq(p3RNV;8jBt)m&-0GUs;l1{$N5E8Y`yf-78kDOi^)YLXAU+<EeP3L&Ez3
z!P;E&3qLH}5&HfQyPg-X@nWh2hzBhPq^Djea(69}POMI;vln>SUdHhkgucbPTY<aD
zPGe^G=v{-C0#$e>fdie9>}-4#OZu;`8f4G%zpZH;*2UwaW@%xpIq8JMmO@V47W(;W
zOUkmYMQCoTc$RY52_y1~>+_a1{0r2w7lLdw?N5kT(wy^Y17FeLZxOhCXQS75s9blX
z_8?%nxy#l5@oqW{zO72?;)rLR?RTupA5+>b+Nn^l0^<PZ+9>IHdU%Dh{{uZuJRSEO
z4b_@a@r1+6&B4jN4MqyWi&zT}O%`SRz++alk(Z^-s>aPp`(yOe+S;-SZkMUNmvanm
zVre8zZSgTB?rTlc9N!Vbjg5U?tEAujBQeG>OB7OaqMoJ4vn^W~`O^-r#<_fkoZO7P
z8p!e;o>vhwW!vHi<e_&EKT11`=0QKCJSA(T+}z-xFO?`5H7Zrlk7k?RXHXBOUwz^)
zEAouSM<e=b%87q2I(IbgW`3%H+1n?lncRlV__NqbuqTe&J--vK+!RlRI>Qyi9A(O<
zxR{<74DTAMDbCHMMiH$il9%NJ(=RVQnuPROdN@ZF*OR7v_%G{sZc8GYGQ>~XQyKb8
zihOdH$J_}w>`{sE-U>MxYW#Jy{yt#;kB`jGs5(<^sD!2{MYnkzyH8hiwa>u8V`e}N
z*DOngO|sdUyEF7;S1_R3J<UvTEQ^(&Ai@(Q3j2TzF9Rb!1%w1cY<@g8{@XA7Zy$%`
z&;c*1uI#6l{&z7zJ!=>)COUnl>8ktJy@B7+dS&df80rXSkwO7`kJZ5MVs*YifGSoU
zT&qX&(uh{s->wx%=jQqZ{6n2^^Ugna?(duTKYt8V#EPU88sNBJ(tLYq0`|pPue_vY
zL%OIrb!dAy;Ud6%6&jEVHP_zjyuUh>CD#U^DDyFtn@{8~89+?@AP_meV*sAmVlciW
zmWI@gs}4ZK%zYEIkF|~tzI*>GB^ODw`;1-90f_Iy09$2f&=5THo75)Xt9P-CN_D*(
zb+J|j$mgwwiq1rpijJeJ``|BI@jpM~zrPi@$8-QHr1cy58X1x%?^*LV22{88kT35L
zz#Zk7>T<thElA}(*`DBH2Lmpge(9j`Wtr3WJ$QN00=11iA0j>fBA^*&7~gND3mgPH
zJbxO7U%jLF^N|U=Al>T?nb=#~1u^>0Ku5>l2aCxWW6*+r@blwHP_QvsijTDH75*P5
z<{yLgU%sn}-UM(F2g6FVud3A~FIG%C$t_)lwuz*)%Qj$Gu9Sg*4kH*zD<yfrAS>7h
zpt1Hc-1QPyl&~zoMl^rx0{7bb(gx5sGbZ(d$fT(yR~fv+D&=FJEdPJJ#sB@a3{*(~
z$1=`#m~HugJCJ`xGXLX$h+{y$2pyW*>Hp)4Fb!B{w0>mFYH$KSFj*G{_;&S5BV%~h
z8?lkw+5h8D{O60@*pL`UM<PxibHjhGW*w{raLTR@2E_JX8rmbVm&ijkJ>D1X&~(8Z
zNhQMT1HR~Y*%FA>23Bi~tGyIeLYL3P#7YJ5bmW;lVB@M?2r>bqHoJi@H$FxYAoo@v
zv3^D(tKE#z_ke2RyIRu}F3o}3hrwC5ksZUN@{_)c#@iPD=IfWQ)dH6}EbC~$e9sMk
z16UYUy60Nb!Dd0mJJx(SgBpOEXMqFAiEaQkk0bB__b*wu?qoz2ndl*qMRxAj-7(2s
z-<*})3uJu#*;HRR!Tz22{`XI$C$NCK6i^p^^3PxI#SsX0B?v`QiwWA+F20jU2t&db
z#et9iv@Hdijhcl00Pf{wwZIidRtp%?+DEjL8vF1d2OT+?Yy~CLK(skT6=-pl4ARU{
zFo(noP=J&sEdeNtz{1#aqC8CdX)F1pu0d%O&-!y>29H^%hZItg32|EP4AnOF73c$?
z`cw7IZ&%1A5RQN;;+lL5G8NGs0Hs_6#5$9LF7+PFmg5*XJZvQ5k*W<$4p4PQ)9#O-
z0pTK9*2?RR<N*Nia8dbRA1;pTJKV2|0Nr^%|J(x9Sr~YUj5!$^08Yt3UWY9hNedNT
z3TH$dEVdz|!&lp73HJMY{8jqDereXeH~8P|^>jQq)OLE@PZIu#@Z%Nkuu1B)n8?%@
zpfZq=jGDC^4FKQ^iMqt;SnZchjUp0yJrgGy)nFh~;UQ;}Mh`dTfze*W;-8sl0Gk*V
zj|Zjj<-+jp19qZa?VsDL6jf;zOM#T=kDU9%RdN(ttLxxwdQ0DV_fux%5?t?4F!ND4
z&)WF*7a$~(;7MH{1PHU=HK;SCb^t3HVJQeTzWxU#-`6Pq6xog?;-M4V<Fzo$istq@
z+2A&9$D(bs{-YqP(T`~l-s`gUbnkcHCXB>vtA-WrO$g{Wk?7R}FwujF9@Xc)74Za|
zYgAG^1{!fN755%$bve7IE?0kg^E7xg;(xniLoj4a2E)}*A^pdemJlGjPf<y3%Q`1G
zQF>n@wG387t7|%m4m7X~B*?15@ORn!3`0-R-=7Ik+Lp=u222ib17RJ0TJr__Fi=aR
zMUFh`Po=;^gTwWJ)mlYbXYCj|P0Jz~c9ORE7wY6%0It_kAUlDb3Wn-TGJMZi6@=kU
zZ`=nkfh8_D7?Q&et>14F>l6CrgE9;irL!tp&xf1+^8a&lFl|$cG5c*0ocPVr`dx30
zyez}ZTYI!(+=%~vhyMK)X-Yzn@kQatsSy72cWMz#0w87^;6&0bLfIA`tAjTiCotRh
zX??iT`?Va@+?K?*%5vHtrC6~rkA<tAT&KGiNBGXP(@`0iMG2tVC+yd4mqawy>V0f}
zyLqMmQAFYKU&-8GhB6LNK%1|5ktPA@kKpLeh3vrF{3CLKPg>fQfYpvDRr|#FkJZLP
zp8htwnz=Qm+k-Hv+J`CNmacx(1)7Xpv)_&Lc}J8^ujkPZmNp>zks{5hqrha1gK<q{
z<>EWaI{hXOTW!e{VncDjW?KXX^4@;qd3yaHeXtOu?V~Cuxt6i78yG6B4yC8@TT+a@
zMpAe2jt9gXYJtQtr?9R8bm=lMEB@~G<|<5y#*+z9+o;ny{lIP96G&RnOcIYF?ZL*i
z0bspoaas_OIDOjgEh*dC;;B^wb{(t`fUP5my(X~=BoR_Kki>Bve3;2*ULqb6&UvkS
z!JdqNbOYKGLXpg`MSyTWuvr>#0vZ;UDto#7q$!L80mz$r-;K_2PkHPVT&Bzu?<4bk
z+IN&{2A?%GEP%D;wi?Pr@_}FHJX&=h_KCqV$oxRg$BeXrNb~M!^NL&N&6hj`Sk|$b
z+RcUH{u#*>6w?u4tB<<us5>a_Yhvc03B>g5M4ecP)7bNUpfwNG<hlm?NpO+jiU%Mu
zqH>NneSuw#0^80)@aG9zalFk%VV}>Ks&S)lu`=34^JabuVzS%}Y1?*6G&fl=?~yH#
z9E(;4Dbjj?C+s2i(6>sXC*Z6aDDEKxA;J9#nC9Vmvt&nG5_X|}LU)TAh8@bFKevW<
z$i}VeKs3Nc>yrIVuE!?sso)Pc+^BE$hs#~soULRXlkoffW7unuc&Wg0TK9&{05Qug
zoqN4p%?%ULzw76L#L~K#j{o{N#=;qZKRug$!EmMpKJMw7P9H|uQ6gTXSyibQy@XH_
zpiiQ@V0PkP_DvZ3&#klTNamJ2BRQ3=wC?V_0EflHSg!dmOHXbAZ(WzEbx8NS+(dIr
z5>YaGl(Rnws~NWIKhusfQtCo_Ea@rq+Vk%K1kvz^We+)lF?w}XF`D_iPBEaId0@&T
zAJOoM2YB3d*WRCL5{VZeN<?G;65T-`-(U!Mna33L+J_4?XldOi^c@;!z0*qa)RiKn
zUC`%!j-S@h4O|}sfb27BTfCeBqW>DO1kyYNcQtxxqT`=PBU1pK;5W7BO6;CqY{#rW
z0C);u2zl=}y0`Qj6$ik#0W;}7sFg%>T{BQDEb`+{U_JK`?p{JS%^#3|bV$)3&PhN@
zr`c+^0_rKd)(a+L&4jJUDB>QKaM?@k4iJgGodg(1gRyr4SpXK=(P3raKU?e6RQm^z
z7#UKXI>|Zn<#Rhl*gwQ+4X9wAQG0D<alVnoLJRxBoWkyNHf1?LL>juI#IE%gN>9sX
z5q^Tj&PnIfheHkS92cokEVF?<gR<~$)#%rcFBByQfO~ZWO2hXuSSXQvP&j1-B|UYJ
z!iPXNBjLk#NE^5_8}*;j!Ugb!7XkvFodNdEiN}{hENct!RFJ4Q52K@{#}&m^l2*o5
zuck>an425&G_EcI@(zrOkS_(rBWt1L<z=pb(EJiROmA4Rz9+Y@>FyzxTY#V&*kIz-
z^o25<f$-t55g6DnD+4EejXAo5Z;KzHJx#o-K?%5c_=c&$E$_~*w~hbQwOyF6<;blV
z&OUofK|n|i*p`8zRF}-t;_SL*uZIa`AiK?4lsUUY`2@ciu9NJlry4W@_<SYS{QziL
zz>t`f6`v#($vXCI7XA711xC*v7=b(+42L)^0@J*1cE>K@8Y!o6iht;;+Nw<ADN3rM
z^Rw_-i<2g&0yL@bzEci<$CRYe!|rYO%1=5-jSbwUEVj{0LPXAK|NIWa9p0!PT?9}0
zoD_Gm^AulnBeP{Tn=Olc5vtLLS>YL>Hu(V>EEu{VM!D5rZe@%d7hiqQ$-DL<-n_9z
zn0%cXc!x4vc~;1p#Yqy@(@j3>b7Ta7G+;K06Sfp1gbeDEHz%XMM(rNGsw7n^?Txt+
zm*H3D(~dv|Jzwvb+Nfr@d6hFUa)@-o&jeA7;3n)AREv3(M(FyWg5KK{xx}V?hl^mg
zw*er}CAC~i5vj?R0#54vl>vHz-W|))QuDou-rjo|7t86fiI=ruD}F&DPGWbBS`2K8
zzv-Q2zZP5b)Kq~C8heblH$+>8G}IRSN31?nj|?R1dsM&tUF9GM<Rp|9ZbL<7AVgPU
z=S4y)2PG0a4A5;n1CI%<=oeauX;JT@L)cM|9VvPRcjR{XOxwfpw8D>;UhC&%<Fc=o
zlT{4K;$Ecmf!VxohcQ$zH$b&+(%)%ACBR)l22aJRLRD#DzjXu<ddqG2Qah>LEE8RI
z_%r3NJ7pyh{PKo6-bl8k0aikFNw5dD%pKgjSc#KesA8xr?1$pu%A)X~%N76=Qs9c~
z89)3&Lt@nwFGN24UfE=~v1kn5k(wT1F+G!-zSLtjTCCNwvGAX~?3@U^MGRac9STZN
z1$04td|@R0Kt>M>Eab}jLvJtJ60kw=9OIC|Ab`!3FNDTCjNu~FH_Z`&J^!w3L1E7P
zFbsn!Bp8d)$ydd&K*EKCB%~~vRN+hUn;gT{ms-Ycrv%YZiQo>-_TcTJ$`3HNAB?!R
zVN}lH6LDAj%~I~I5g0-wqdFmN(-gACMMJ|)1ikb}xm+tgMSFR2Nf*jpGu`LBqtd_B
zC^MsWKx|_#5Pue!i3M{3$%tVs5bYa{m2A%e3cBp^mvAev4?x2q=LALSYHH9xy!!&H
zjX-hq{EW}$o6Zg6?n&_Gvs`h4hF^e`VYD`eaj6!vxT)}YQhNxW(f)Po8(>3Doiz$_
zh<{O1w<SMCTV}Tly2X_d@g>tTEe}Y!uLJFPys*$3jFu3f>ycj?o~4|YS;XqzY0}9t
z{Y=!+jszQ9$HQD)o?!0N&}YZ8f;F_M26<*%rBo&VHdT_TXAU1ad;&uMkdgHgv}S;O
z)ak{ZWoil?gL%X2*%pRvKyBDFl~G?>y^9`xk^$;QrLc=51{=ZqJe*sPP9u3>*Kx+^
zuyD!^8ya9RTZBKX%C#4GW?B8pAhpbyd4B>Njq1M~%?6C{-bex8pXDY2Us}Dh(+9@O
z=y17GAW+<1hj_Mu#mRdvk2|gRMp@-LpEa<K;q*l%WtjCRdMl-?IF`0uM4E-4l?%hu
zCMMpOiq*JqbspzF&r$sDQ6^SeZ1XMs4wu6oh~sXlf-k#|aWVey3_8$AvBmxw@xOC;
zW>ZvvtS=?<zBBg{?`g;<cwRv}4Jxz6a~2*G?N{Pw0b)EhA`ua;MEVcZ)x<TwHIL(0
zCdb3EK9BEx-DOr*?#os;&5md%*@C#-*Z=m5q!8lrEpamVxE)<P0$c6tS>j<^2wOzN
z!;i=wJZE3XI1{tulN}WQBcRP|bE2`N;4w3g71=JsBvv%KYp`)45S*90%a|A3u{x%*
z?d|M=4=tXSxAB<Ua;w@$eD{0$yNLLM@DkQwX~a)aDorq3CpMO7>**{5p=4r~R8bq7
zT)gz8RBK;eb+CU4f%G!v&dB5XedfqyVlGlvlaF<OvuLMqWU?m#fe0}JcOYkqsBI^S
z|2kqu@{6c)cJ(w_^pl-G)$CV6y-zWmw~t099I^G)j_a(K;_!U2SFM7emL{k@$*jDK
zFHN<EMR&r|A}ky1ZZ|djx5}g%v#caXoCK~Kl_{)VFjJ|jbtl4m?Hx3xT!(1-C`?Rc
z8`{x7P0jrPd29w8F2qCVY8q7b^=9lvr9m>hOqi#TXrM~#mvOH6h8cN%-+O)gq>qnW
zU|VMK5yg+~njV*PTn=}WX(<xf9{!q966JKEPlJD|X)hww5`4kQtf*o8yt*Qa%X|!b
zDJv>DkF06@UiC9LJfLRnb_WmA<4v8-waZ@B=0nF7#hj|zx8*!v@~&DyuN1_$Y?hYe
zu#eaf=9h*rULhX4#LI@0h^YKl^e^*V`v3(V(byOJp{IEiliNQ><0~=Hz->C_y3c?c
zV4N5(N4kNb$Ea(F@he{lUiB8y`x+J0zF|1ylxk8$<Mqg56PGtF+&u~SQl9d}w?v&r
z5S+Kdxol3Z!e3GZCwujcggn!;xep1sZ9W<`&FVk?KPkmu>BWgUYC~Yjza0QKA><K*
zG!L$o3dpbx$6n$R;(zW(jluQP$=!Qm_1a{bo$BG)xR0rp%^Ssj)C-Vgui~@mKcwP!
z@1}0|!J;4}R*a><mR|_zX5;8I;Ykx`tm09X5jbxBZjr>o(~A<J>-yExAeK5U!ik`e
z9wE=3JU}{O`%3bWREUyAL?&W#mPNF(WQBM!Q}=~xAH{1^lDqZJPgFlx^V3><V&hSL
z9-)i%C5fg7&FKxtBD8SYsvrI5)brd@dcAhWD!wflqfFaHsJ-p+TldL#0WmI3zn5@H
z4H*VjeL|j`$`1RC!RiaI-shq$7U_nbhN@h~>!A!3mNwPlx{l&TS>z2V41@U6?P)8G
zQ^lP(oGNM7;FnxZ_s3(RT2@v}$%IocgOv*Y+YFYK!el~jy_Sq-I^@{Ay-xGA$r$+2
zADS#Wh_%)Fj-bv!XMM$KlJL4wywBOzoYMIHcm$cQ1}^4|i6nt;u9UZ?u+{hrn-AV6
zo(+$nB=<&t6kjzC*Y(2)!s9oAt-spqAxm%S1{*4$yW?rU{gu)5w(?iXW@&O`!T<zB
zxPcYGL+g<II9ev>vxpoVew1T@cKo)pZ=TFU<--*>Hm0}Prdzs&Q9z!L=cbO#|1ErE
zx)a3|n${OS|3@MXj7Vw$8Tr0F%<<u*a!Wtep0i_cFx!F`N{weNVdpC&%Pt}fMtd5t
zc_5>1o5o;vk>R^miv+~p-N{-vp>WBA12R<Gt6y0{CUn8V`OXpT(?I1@?716m`{y|}
z2KxN&6Kw}7^&g7f@BHy=06NH?CNXKX#Rb=X3}VJnKkDlFNR(KiYs=3C>w+|rJ_;jD
zZ58viHuU?gMOhg7@=ih(wQ0=HI%^l(4J&j<oX_E#C#$|fs=<3lSQARMwmSpQm}o7&
z_s%IJ?wq%OLoW7!^gT(hckcY=cCXkr=a3@GM{s7M-DaU=1pPVnY-Z<;Md+yE&m|h_
z^LkGd?U~vIf!>Tg(YW7^`;lxrz^`&4<Mcg>y77y4U4B~6%{Q+4vHb9|5yOuu0!;>q
z)tgly<R6(EOvXU;l!JLVWA4I&R(2)yN~;9PhAlDQInbX<3Ur!e1MLnfgDP>yXjsZ?
zR9~W1GBo?{O%10Xis`KDdpJ>%Cg!;jqIM*^KI~ng)V@4s_`U(yST{nx-5{7H?fYs6
zrBbPP<J4m@r4hBsp;Dri?3-1pLA-Jj>%(M#ovedsp2{|BY|f^jM(DC}*eew5I`LW_
zsME5l5~p|?9Py!XV5*l=@^(pZxIHY;X{v|;!@HH?nSE}K7Ugb{kz3w=>8C%6)ug+n
z6yFp6t~h8BK>b?aA%&*LSOoR}%?#zjN3Z-$(bwatW07Qz01Z{T0!rbVwIJ=k7WO;e
zotel$Y4`Qv#is;%w>gA!BcfCb${A#1Sowjp4`I}CpKI?;#2=8upL$BIr_c?IrbNNE
zotnEPvKp^RERwC6GfH_isD&M&b5{kn2K~33AMbjTca08yW5x8IVw_avY8Ait@?jW{
z3&?a2EwNsS?I_h5>|n^bKlS-5b3OhJMq!G<0pL-erB68+-bOjs10^NNiB*v>&-({g
zUIwgP(ig3*>~PFA9PvV3fZJT&>tQ*LZVkp<=lx9ea)flmTI))JHWznd;EX}BERk+D
z&2`>a<or5EDf-OYea#0cTzG(AhC2B9^YSdIQ>lu`h+q$%L2r8^O|h;M-dV*v<COce
zkjdVVMPN#BkW!YElF59B#0Us$j4IRM-}a&DV%Hv*SbHg(EVUDK{(b9p6rV{8iz?pN
zn5Fl!{-;Hj1ofq}Cf*nC|CO}TqxR!G^YG}FvkMVCl^5jU5r4$5>Un!dz`pQwsvE(N
z@019Jg8iTAvq@H`m~$AbO@f;=tu7qCfqO`b=p3MFslaj=uMg6_#$EF(GfGIA*H(>`
zORB_bYx8Rlol>~1IBIq4%Z74KjNWFZhmAUBfxg2T_V$aVc-`yG3dbP>A&g4lN&Yws
zb)d;?wE^XYr0+;}u6Ojd?}-s6P{U@d=FopmPT3kma<<jP5vD)+N7Y`u_}zudWU4cb
zZ$Owiz}w5-ghjMX1d~4v3Cqp1k~H;u??a-3mc4+a?~CMy=6i1P)C@LBpMmbByClxo
z?83w8MnDP-5Uw*Kll1WrzTC)h2@N*?r1U_P(T}gN_|pnpY}T=dwE!J`j{U^V((3g+
zHb%qpV#fDEiDj(@6=8gX;-g&=ptnI-9Hd;W2BP&bA3HR9_^6S_SliMrewl*!YZ{Of
zIV%eGHDJ`s0;3tb+_cICEPneci*(t>J<Z6gl>B9;LHua!g~s=QVFSWTvH|g6sWtv^
z?G2XCIi6XSUO#pJ4nupkN+}vkJ|Jb$KwfP5MLJ)KAU<?)3u3`O;pIrqSzen_&OMVI
z0h`@diEE)k^3VE`2qlw<m%rXvoC+N{V+Y>eWCr6Mk-qwhciBvv8%&Dj5rIKZ6eMNz
zcHuI-&!Z_oIg+&N@?>*4y8y_e%|yfu?DB81z7gCzQVSC2o~KhzwXfT}&f^KO+Dv@%
zm@xhzFOr{bXFU%u{RN%4H<*<1RqM-7x`OUi{xB14cQz3-Z3Lc%(_MTHC?IHqG8I{o
zJ_Ib;{2q)?^7i@wN>p;)OpJ@?0KsY|=^}z&$YTq5=bRp9?CbNW!;k!@q$L*e^2F~>
znZd5}KhT0)-iTx+a2pX`hpjLF1&fbRJ|FFv(x`StGA@yUpngR0Gpvf|=!0;aa2Ou-
zmYb-r-|+sd3w?VYjgx=pHxHAKiEN*ZbCXEvta5VGNQ-J&xF;q#xB84St{tTUGe4dE
zxQ@P3oE9|g!5}plIxhwGb?SwSKLr1{fmlnYbv@)erghGL5loS?lRl^il@-I3B9J~4
z`MxZnClhBLK-0q}#h<m~7+L}d+&qm?O-CB_bWBHU2@a9&OQ;FzbA@_csz?Xuaao#B
zb3KvA0!%sk75#gsK4Pm$uy)~wSBIx%gkCl^FVU#3Ew8r+lOX}z{Ap5658=VKw=}jX
zstTl@^mIHE@0&JS+O`K!wJ5w~k#e^V?n2frC<_>m65a7sdKvf%g@kWJwoI#fcmTe>
z>rh=9;Y3UrUG^;Klp`dF{bhwm2LfXHlEk^MqmWD7G|S$U6mMiO6)0`32H}^cS}Q5%
zD*jPFrAZPt+ps-N5e+U=oeW&K_3YWhAIu(Aif^EmiH4uz7tNLF{8*x4ux$STAfL3N
z;i+b#dZ$y7wO^I>tgf)4%4gpRCt%1Gr42<hCg%<?c3M#4_MNP5KC4o1uSI&EiO<i9
z2^XRNxSd>1sz~-S@%A7=*kq(e1Ns3-HtG+kd*PriYF_ecVPNZ3A-Jb!+OeUMNDG$9
zf(0(aF*>?hc!WpnT^>?Yt|ECpcW8P*CI_`g7EFW`Rc3AI{FlGKmwSshN)Q7AbZsKQ
zf~`0kL%7I8t`?`AR!rKR@!ncVUlD}J+M>8wEl3gPF6m9#M{2tnU@uFQC<xsTj~0Lg
zwSZ_P!#q3+y0^q%#A3Rd;oV&!M{W%@@Bv!YH4Ppiy=_t#eb19HP~IKENb|9fV<)X<
zR9-e>*`K<1E;;2PQc&b`Aj_D|Z}g7e{fCi~GS`xZyj8<Wrz*Nfmh=J{+8};y&H6*K
zo{u6n>10v0a1EYl+gB~Tbf@=m!yDjX-BDLKd&K3k%8iH~p>xti1Lp{-jBslcqhKX;
zqmYGp>R}o!lJX5DeZ~XcuD`S>2F+$5fXsV%Eo#6KA5YqFvK}QYmJ=x(S^c^;87BJB
z*HznPb{qP$%a!hr_fC1(#8Y-@Q_xm3=yG=nYSez6N_zH4d+k?RD9A88{5(+va3aOT
zl&=2q9sjB$Pe%qczRLTrVg7HhIO7vsq2`^775+1bw~cT)W^{zaMydBLSO*mkKVU1z
zJa>zRiP@FYiLAuPEe+OXoN@D&3$;QjIiH;iLt2;w`=1Cm?LdQaUjzAIm!MczKy@d}
z=I7hm*bn@;-i|4grb?#xy^ymSIqs~y30z1T(;1yO552l6E*|!22aWVPb9toCWZOnN
zO%;oUFC9<$^F6moEBv{e+GNb!xA1V5^3X413FH18jI^Nk=xmC=0@2U>oCJkDd00)h
z=+@IN_p0k#MU<A|XP8v-u8rwQHP$Ei)!OgMwt|S~slgn=ZznCs9}wxws~}Z_<kJa~
z>Nn{bAIkonN7@ee9_yAG#d2=pPw>P#zoK=3sYFDa5b4turTBOxw@j9Pvg<WZz>MGy
zbTK7)2GMUn=1&5tBra_P$HK+_-t7d)7`3MRFmz3q2HHdB1`y~eMiWn8Zkh@BVwgqi
z{@ii=k?qEys7$%)^({yja#WQh(5)BdI%$vlG2LGO<z5+AeHJy)O85_F(0@dH|2Q!7
zbEsNz`x6Hliv{u5J%J5UO4ew_Q79s+eRh2{KTz`B_fU=s7Urda;*?aK-2K5u$6(6d
zy;6#X2E4!g&IlH~th?*w-c1tM>-|koN6m2Dx0KCFp4chUSZnt(!#yV6o7)G)*O03`
zS16|g$u?1?=*A|0><h0Nf1%-zzv7a6C)$hPKo=<9BHIg7`<yTdBZ|;{qS}X}|MQmD
zuZrP5s4pe1Plbd$Who!NP}IHP&dRQ&5qR?w;VUb6P}D6S9MRaG1i!;yKjWX0VmR2W
zJ3Tn}Yb&Y>!`qsxoHUIHGV+l50eEFVjrbw6&gr_~rlrUxT9P-<uQkaZ?arHq^Yc3o
zv41&`FcqGCI&3St=QYa=FRnW)GimdwZ#>ZdLF&StUgMaNcptcKON7rp*kq@?qfoiC
z7h)+nrd@fGG^v`7y<1(`BH{=<*X-1{D&zl1fCSJzaG(4;D}pjGLmjSJEJ8oCwoZ4)
zGTyYZqK4`PVxV}>h?dvHL%bn%<0h|WD0Sf$G89gqblp`69<E?iPA4E2JKmO6P=^MK
z*M6%(+@KWedUf~8np1WiEl@vKFO++*Yyg&8Vw*=1bQRH$HGsiPY59Oh*uw_JE!l~>
zJytf0Q1T1zbHfN-!s9v{t$8r$5UeYH;ME`eyT~yHqW4o#j42<N^8JZ`{I4llbBRT|
zo8a#^TV>J|aZ!&VI#h($u<@6jd`SA0L`y81NmbAoBrDHi;lxq19gT2ibc>S?i#!4Q
zBaKU4lJUMv)w$c2*}m^ks(jr=Xws@ocJ>f?){?OP7@tiMHty(D$I92psskmE)9U_^
zhKBtH@n^{82`zeMSqj{ZWZi_s-MP2HDlOa-S$gx4ftS-!_C1MhCXoI!AO0@~QE{>6
zo`ZYg=~Pxc?DZ25(F*pG^2^M|YO|P1>caCgxP>;HUi`F-ZEAj;&FmjPXDpC(FO3;l
z2Mq-wQ>JcGZ==HkYTuQI;4?PYaTXUNj>XLHB4?JU7O{0}R8d#tl^z?*vgwX7iE<u`
zc*FXQE#g=x42b*+Iku4@2%?quNG6GaZU*qPi7J^4AOgpCjINiHl#Dj47wH&YuJtys
zKij_z^!9O)+d1FxbfH``nhiU+X62&GG?T=;s*rntcZ9GNh4;R8&}isO!2_GALSP=F
zGajAoJ}%uc^tVS=UdESVkyZt>-tg1mlakr}S*hd>M0<nu%HD+Hf2e_HETH;yl^Cn}
z=$}GwC>Gt`!s0CmU}E6)sb6c$L>(h^9hXpk0SBP_Et+;sv$Gf%j0O1E^DuMZEy`?x
z@=FpN_d!kgJF>3Xz_;PSHb)j-m-sLjcD>x1Ly3JKbg6I@k*?Were!sLA~b7h<(6!L
zAfh5Xy03I4MWLY{>Ny!Jo(Bu9oQNMyu~8kZNL20?y{NTmV5(4a`3h+psMg29*(__w
zI)Y!jR6cOOU~$-rCg<zqAY=FqHn!-0J*!z#daz7q)Ha>@o#yKmT^j#`e$0!Nes>qo
zIH(hS0tBWUU;JIiZCrd(NL$S)fnDb{+O_FT>sJhifv#VN5PH=@S@Mg_66nJm|3A0(
zH^B;m?Wv&6#3`ub4!cSc8;e86*cWkd5$vki75mQwuuxg~j^32$+;~@WL9^apqF@$j
zmw!fUCM&~{kqQfc??sK~7AA?9edJBO?j!%yQ8j}$dcATA^yWRf1lnLXDXubBi4Z_p
zx6c}AN0jIei5~ayK`yzbv%be~#~56`CHXWN-4QvdE)w%rz~{H)dIS_Fe@(IYjSp+3
zxam^%P!fhINj#yByAqT>@{;I^Sn{~DZ0EP$+i4x?FaZ+f3DAD>B&otN!YW}GOYs=}
z3}}K!ax*zbd8F**>olOYh`!C@dv$ioKjOc7Iy08Tu1ezggA}*Ny3_N<U1Jb)#JtCc
zE<dq6jW?os+0ux{luBVu@g&;u^<LZW{?m%p0Up~fP0Qk=M<3t8@sI57ZyPka_1-bx
z4`1qknm+NIq+v8GvPp{Cxbm0&u(3c239?Iv4pjjeWNF>J2^}b^BPGNGV8&p{e7^G%
zQje*aaw!dAtG#XCKByluUmClS16{ffQ&;c2CiZ%ov~sgpLW0d#+6-P^srt{FxsiY&
z#K!j7=j$}%qX-!n!rh>zqE9Ui$YCACx`t;L0~J5%v}eJq>Yqg8T6k7Rst@pzDObLd
zfsJ~44jO&;M*}^@231A^C4&-u%I;0-Uh$H4)9hojq{e*6UUN5MTJuZ``j0?;pXmVY
zqn*$;4pQ#?>j7a!fndG^cR_p-)rtGLj1FiL3~t1rR`Xc)Ka=K%FVOwf2(r);Nc{+o
zQ3o|Gf#RaHl6*#Xx>2g}6ABN`pWs(Nqf(nB-Dx%D&eHyQ!CZo~wk8BG%Y0M+T^(nQ
znHdiK`sND2;R)3owZBEtGLs$gN76`$!V5d0vYpU9Ycx+jZG~2n&R~g!`)fGDaBg}y
zEEUR-3&mFzYya8(4y5`bNP=vjhJFkd^9s&d6Ehs%nZr})G4PXQAo%;<p)1d~ikjj!
zP@N{5^aK3~7!bYRhigwpM2=PFcXw~N0R;>lLWd4{bk}=w+uRg21j-p06mdbJV*jEm
zS>Zi&FCkdB?1;A-<5qo{H5ngLe%EDxdG$4e?7FmW@<OjJoD9)B%btw^`G^4txm)>u
zBZFyxgc}ivNkgV~Z`nFeGHLamWxBff=`0DhsI3O=ixux-A4fayZ7H}i8$3X0QdX5a
z(4eg29Tb(y!G^i){+gc<+-4t&kSW!l;wQDvnW}w%ZC_nl9@^>%n%KbaInsQ=P3Xu?
z`ey#FGX9L(*p?j{NfpQCovoONZt)+?={U?3XMPcWAI{7Zh^~cpVNKQ`y2WFrE2@&X
z#DzR{?{==d0A;nn`wFAd8LGd`g6th;EXWxrm)>wMzA95_gyL2kHE#$#y(YCq`SSQ1
zvn<Zq6Lz@y5Oj%V-(4i*@=3FUX<Hc>1Vo1%U_c&vW>_r=G}D-=C9M)dEEtNQ_Rhf^
z$rz9?7?nmOYEUZonw?{yvEwqYoOqFOeH41fgJ%f#s<fgf9+qiJnK)?C1%KDVJ&^#K
zYkil-6?dUSt3l^o$Pp$aYR#y|DvfRg^gl?zGh4Rfb<GZ&dfHc7LuTitl75Q}S@LKg
zTSvZL+6}PSF-Mk|su+4MM+u=G4qJA$xF4y^Jq=4w#?KvX(RunuCPR%oacUq)ya3Jj
z0x5GQ=$J;FtUYtJe~mEaa#&-ovhK8lV4_2UUzVVWL2fQ<2MieWX%t2YxY`Y17n!=b
zU;4JK&`v0Ag|ehGcRvcQQEUq8EB>M#eZEpjeDf;yWWxaIe5?{$fM!gn1%|D`dzhiR
z-qCiQbR~7U*3jaCxT(@5@15D5+35XliJXrt{9rO`GBD7X{9UzVI`Ovk8#$Yr@W%Q0
zrmwgIv}nIaf^|@+R%x9kNW@w<i`cr1)YGSkxp?YAR(tg6l=GksSGwC~Jns=0Qv_W$
zbQu~4Mfp=Le3W~`wap7tFWoUKIjN8<Z2i{|mC0VMMqUROd%<VdY$=xMcF(0UKo*<)
zeC!RafX2oW-rldBj#4P1wdFXQ%tKMJ4xW)*zfn6i^Jwt5M<<b-ZO&NKz@DzhID+qS
zWyt>d9M1>6jJ{JiY4Yl&UYJ$=e^?VrWZEBvtEBR;w4WD??lE_-+9D9-&N}ScG0!ZT
zaIFdU=_{5O1x(W#F!xfUGb1nyu;O)~3IR|O3xmnd;MbKg>!`K1UW6G2pd+fuBXutE
zJSA+=f}YJG5r*FHCF5~@*k$+|5p;J6z1<t@f<Pj<SB{jM0L7Y@wCB5%n;RpH;_PiB
zSAYxN5=A@j^P#R_oyv`6+(oP@f&l^@W%_;bNQGw+nISE0+9bn-Qc0(eq$Y}ZF}%fQ
zNM{M<?vru1D2R&d48Hdi|M>9_{iJlHwZmW-TWLeZo9E5rdGFWIoT%_u`oCu5YJ4#(
z3u0TX1$zhXy0EOcec*4wl)*uuqvCfEq9baJq^nvzvCQ-d->)&|M_j+)BB-UuG37y{
zxuzeNoedQ(1byeqlu_1?mg8LcM<-^1_&i@SYMn^~&(ruN-;kK>_&{+J%9-$0KC6m#
z>w&D=<)cf!kS`}rpq(IrtBUR=y=2h$*{|UNqlbbi(XLbq&55yhFG=8K&wiLv!xYCj
zLVlejh4b}yc$biUI$>nzRSjAMyHy-p$(Fd$;U`ZcN!fLtTlQb3S|$RsL4vUYB);T_
z2w@V{O}XQcy-l;;%MaWiE=d(kzbz@y`5m-k)_*P;=TWF(BnuXSWsL2T>C6hjzBTEP
zc2qeTUshVBW$v%*VKEVgwxY-KnPFwCSGExKWtZUMRon<}W*QeXwF|=4_0Pm^&cUwA
zXM~Xj=Zr<Wdl%A|m*J=XAA9c|Pj%nNk4NN4I484gGAdcwTSk;Zg(D*)Bs*KkcI-+S
zDWp&tQ9|}OMs}3FXOg{Vzt^YxTK9Ed_x=6;et-Y|xh{`Ge9rrGKJWMI^?W@?JVX19
zZZaH8i+OVN!pwMBYa~{{;&LoSzSVX<{jGbvt@ZcMM1wwRspH;R-RFGmx&roL<MdAK
zfyuZ_9uL^p$!DArI@1)-df(gLwWuget+*|EM)H?X;0b3toTl>!ztv-I+2+;KioC~k
zCU7&}Y!-;i<nwuZOu>T0%y@e;pm{#{UI9C~3>zaan6RpQJddW+enKvDhEW>Q!*`Xh
zO?{Ro%cP6Xywg?{ex=#5^LZf|Ny}KuFiEu;0QMIxv!~qVCEoEbPcSu?5%7BnO<(&P
z6aJew5aoY(o2OI%MkbcON`%&5N}(QQC8P|zrzPL8(MouZu~kegj*my-QB8mXFYDL4
zn$enm)~=k3_XRad>DxG&c{$m8s-Msk(&M#JHP|AHs^bc5e$p<SHw5ReC*y282t{(}
zf_$+OtV{+B7>gq>F(yRzi7|CH^3DgE8V*&FdRRVqHZFCSOqSusgJJB6nkP)`M1hHs
zf$Uqp46$v-@(B;_eW!{&F{<J{9(};RsNUC|*ymZO#)zs{HnlPBr}NDk+h?avKAQqP
zWqo#i)u}QWZ|31d+ZVm=Vf%LZZ=W;6F0ZQ{*RoDLvyG<2U$-pCX0>~<hv8nsNd{p#
z3ly`s@bVRYX)V{E!&opZqhUSODn<5z<+l2FN^g-p_O1)YosSP_Y)iIgycLSm9b(2T
zn@)0uWxPs!q*fditH=D3VD>DI7L%h@K8P_^l{M|pa%r)V)1ED2?;nqk$=<=bnc1`!
zB$is!WK*H=Ln4_mkEFZ>=WwDbESGxgm-wSe>TBMe_A9s~YaNsn59h?W{ntO&2&L%*
z-tAFOw<kS$#^A_2opEp_Wc8f*BwqZOqLBx+nqmZ<pWxnMk6J<zZ;2U>wwEE>3io2>
z>Bc^qukQ)AvL^%Xrrhz|=@;;(N2qmQ_sbaThFIzS`*id-;^kPJngyA7DTHpEH^yHK
z#h#1O+2~wyxDol)2(RZ+txzyJ=S%1Jj`nWnQvUMY-#e-*8KO3cMFB^cX^KRaWdsG3
zs{7U6F-ni*&zLk1_7GZ&p}VH)<mD*H+}|G*(7YEPSLILlVtL9^@a^{Bv>5UEFt-%9
zdR%ztmrCScr5+6-ipAk-*jE)=`*)Oer%AqFjK;_Z29#JRNnlZ$Xg{^b2WZVP660OR
zGhhB$Ai}@>Dd#5y1e~y3w6yDw9*R_UBV+NciI^Jb;cI{bE`=gdBaQ9IE06x_Z4kFy
zt#$<01bTiOx)ipqRw{}+w!A?r?;2=y4Ya!NNE3RDjT#j+w}%O@20G+Mc(J~$NfqRa
z#SAa~NYSahvIRXGx&S776ns8Q*Q>}fur?}zJ%emRS6AMVjVa)rJp8<iruKls&dgW3
zSWOYK>N?6gsn(A-7u?W6X38Pbb>xiU^%$a35B3pv#^ll?=a`<CS!N5TM|ZmYy>CQV
zi#xHASAI3d|5~IooC3GMymDw)AR!)LX1{#(77HO3Jw7TG%j7`rvNai3A~mzF_WTMv
z;o$iL)}FlA#$307)B2#{<4*hRg@64D_-~o-0iET%jzXaEhrnP<5aHiyr<UCcU5CJ7
z6w;Ns&t?}6TygBXex_GxHDoi|45;S}Z4#-hN+ttwBc}U6&N>pp?D>j}B+osO0=I?G
z{(Mj09iUx#N8&G-bt^_TP!%LRap8wN*d_pIx|lp4yN`*()N>`;97QrIbElqI>x9x{
zo(}>SX%uQRr>0ed%w}J0yikX)_Ry^ahS~k13T64CiH8?;IOFK>>5y6c#&(Hh^0>??
z%>t~}bh!YN`xgIoT@8DF;BsAow=?2T77TG2{R?X2WFfewSCq|d=cAdK5FJ6z8%LIB
z6j-Yyw;_)jNT&H=L#J<(7p82%Yq0FksKe*zc5(2u)cwmwkCdnpVsnXmRNFU+gBQIJ
z=r)`<FIN3{2cdWe&cPKXy&DBi5VBoj{=;2%W|Cgo_`XI`9wWjg)k=*hZ9fF8bUnJ8
z?AtZA&b$YUIBt2(f1DoZvs9<D-Gb0>r>0P=1^V&2NT=`s{70x@1E0C}eL8cq-ua47
zzEh6kQ<Rrg_!^-{nw?qXDsT|p@~q0nKm1K(ma>OX^L)%xssALe%g+$G-){-98z8S;
z1h)thX2!RW>SY3doC1*|&8I3#8zpCRR1yw?u0ZGGfk~R|TSu5AzrZC3=t^L3C>gVJ
zH~<fMbBrn6i-^`JPq^V6<NN{;y8P7t0LehK=0b@3fUrV0JaFN$hOT-Q8H&3FK<krW
zR&oW)(co|#+A=$@?)Y>2;qp-!PG?nbfcZ-ZXby&n<$tj(`XIHMsqjb0CX!W73DYEf
z^UJ{{ognmheTpLLd~--q^j!^u9*!u*(m*U(xHHVGZUA%jpd^YU6y--dBMG7vL&l(;
zaiCh)*9I(9{PdAmuNOhPo5+w0Vtb!Q2s_w&13k{80~Z}v9SceV3e$Iwb&VugzcnAi
z#}>5&N%_4~Hf}HwXX*)-=WYAo1c;okJ9tKHi{u^;O^I5TFH$i)#~XtFaHGP<NG!K2
z&TZIpD#0@NMYL>RXK4H*N#ER~*1k9ewSjY8+`oGPoE}r<9sR&Qa$yHN63HNee4*6m
z>RZ*bgvE#jLMf=XjNX3apAJgPXSmT3wPTx<>>y1mDAr?qu<*^X&ae;4cvqYf5F&l}
zGfLMn(e@Km59T<crN_qlVZtP-Y3WY-2o1^z5wU)ovxV)gu7Lr-Brhei2il_s9of5Q
zl};hxw9_rOv$3!+DFHp@=1K@qQyYTHA2oqipDIdb9oyFSQ|jzbIhQenNB``R;55(Q
z$Ig7n*g0qU%I!Z#2Pr2BK|$dyKJzl>6s2?MiLg8FJW-GRp~c9%Dkgl%(16?J!ht4M
zENz4zF)OVQmfje{+7_lHt&_|ykcYu_iH@H;xpi?vmsk8f_OB$<7)b)&vm7>KwJ+$4
z;^gD$l)X`NXu-5;odIiBLUoe)V|nSWaD12WM2utT*^g}Zw)xN^zSsA9kDPWt+baHo
zjP%9iduQpFGoZwBHK8{>Yk(iCqpmt;Ro%g+)BBE=#EkK2^kYTd@s(5t{%ekHVFxgu
zkNK<eZm`8RQN6Bphtcz;8n~y&=K(Ib93udxD+6?<=vQp}5jten%7H-=yaegljcL(m
zo=}dR+uZ^NgNF|qr^Sp-AMg7@j(>w=&!dn98g@GWjjgCkG&Al3M)@fEUY#dIRC4>5
z>k#vjnJQDxx_{A7dIj8tgkfu^w#c+r^H{xfQ*y!d08bpN0FyYgSgRnKfBvc2t8Upb
zLBl<z&2a5HdgH(k@`R=;StpM!vI|0I^>ZIlw*c5(Uttf0qmjyf&Hm?B%-uUjX6kOA
zuu(m`hSK6*d-yIbJ5BkKVALu%U$aM&g1%A~wk^$NTiW|Kq0YZXpwJ_-z>O}@(Bt9!
z<5*yCzd~3l(t2>TRG^mnYLRC4`;cmn>D&5_guJhoT<;y>tzH~Ss-<pJQ6S>tB;@rn
zli>~{eh)~HiUF!^zSI8C^DF)lO5n!1h9G*LLwL>RIMb2%Xu18O+jc>f;X9Ew^1Pcs
z?JqWt6#vMguo-eI`mXZ)$<XM#)CDfIxPVoB<rcPNI!%=K_1D7&0cM!;+Pm05wU-iM
zl-AGmyVlvQ<PAcxIq!XO?_xJGpW!s~^b<;e0x*-_GMfKkw6nn%whnyR=~z_8liR20
z1|9dZ`e#=?mZWs>dGjG_a+{hYX8Rblidf0grjA#&|BhXj$(6T7Q>&dCEO030JlUW_
zahowkSfw_1c%%=uQ<#0by@RXo-U-z2bHXEx1~qF>7F&+PVx3k~oJx(S7|<CHEtcz+
z@D~{;BAq(sMDN;di#fJivSN*(Wjt;7sI&gt-+R-$NN-x!bpG{kz3K0F<!?ng+RsSi
zKw&W`UUT)6V~A&rZai)L#n=0{r?yf_9uWnlpdbH@i25JD^_CbR)}uyd4&eBSoG4_k
zakC8*QT)e$=9h%jA8pIsU;o?*exD=$L2X68FivIl?e7SnkmMODf9&gk-Y8foUAYX%
z24^+M*_aT?Rw(=1#=w7X&;La`bVYWX^`qCe{||p7g&gIzR7+>llWXGFMi1-<s_l&h
z7VruCkPBmoAVfAX-|<TT{BJ9RO#dkn&kmB&*^WOy=f9(i|MrSN9=fu6WOHe}29Wow
zpE}@9Q#o8hI3v0RGvWVm@i_T5<StG($)EiDZTpXf|K}g@v*UmzE_~h<^luCLkE{HT
zFQ!Z*GQ1V@6zadf?f?9g{NpR~g^0ebUvu`{A8S$epb}2$<2glRVfRmja^H)VSx4<{
zugym4X+7b-LWta3goA&|Mc85m5z1V>B%;0B`(+wdjY=ho@fadZlDh|ptO<BYS)a;D
z5}&cZx1(PKDN!}Z>LPL*fkEfaAkYl4HG|mE1lW=`YWjBO9;-w(DsOuex0}zicnc#4
z#A99rNjKS^$#}clMThJ!-y$mEpHKm?eNfJ(ikR*QsNJnCGgce+hh&pBP{M}S{(Z>z
zmu2H3fDhYc!l0}_C3U-Ugb<Nxiv<j$+U9*rm-&gerOOw@L9S8*4+paml&}s^2?s;*
z@Udxnw$h7G*U&|K^T+$PWw-PK_WA2pX^*eyuZmn8fr-jtgw;<r-m|e(&r+gkR~<-d
z^293r6Z25Jec&PB`rng3vH`sFU|`L)1_hEUcPNrnELM|^=GIV<Ep7o}S#h}By9Uw!
zF1Fp<KkEfz7ZcFRB6aqynb+3l9*>fGXq9?kc@TxI_dW>Gs*4=Q2~+EPEtAWLuX*w{
zBI?djr1}{4*|FusFxcj6$&Wmu=r(YQ++cF-_gZP86NRPUyqe73-bFM7ia-HqnQ}o0
z=z3Q&gq(pk$!q!3BB*);2@W0#gw`Ohzq9=@+4|H{>FKRb!9PAD|M^7=@;Q>Qlf3x*
z@~6aMDwlf@SY0LZ_~cks4%g2cKeyw^4q@2l;0i~<uZbcvy8!C<Zbu2C+@Vbn21%BJ
z?D=sS?jsO(!v)}~gFr=G`i^O8_v9BYUrre!4M3ltu?YW5>WY&egwol{I=LrE<`In}
z6<zEh5=v%9kfoLKJ?(uT!b)v$Z*Q#zNb$_U6imlyZ+I;UKo(-Lb{!&JNlFGU0Xg1}
z6A#n4^B+%Yd7A2@P3*s)MZQ;(OPv{j{kHq<MfAXNbF$o3>>~gwt>}er;=<pcDIgJ2
z?l1|XgM1HAJTp7;L4i{u9>1Ij<*0WlqNB`=`b>$rrAwGUY?<UycVqtByl^>4f%or_
znvt8$18Zw|Q0-gCn&+(hP+Id0>!0W5-?s_BbmaLKmVEa6=#afV5oIh$PIeLSXzw1l
zpA46`p$;33J;cdqLvJ`w3i{SkGUD-8#NFX3*cJSebjt0HQjS}=ah~pOr(rbQz5_i$
zFdUc})Ie+YC*#fa4PWvMa6p{Moby;sk1D-EJ<k=TKS&t1x=>U`lbn4jTIc15X7RgM
zlRc0glejSy4BIu=Ar<aoWh$#T@3|kt`|E`S;r=u79&;en_bUjPT~VghJAv3H%qXrJ
z67|JaLW_uw*EK#?m{EO*!Db{@re9k;eySs}ZpFEH?RTaAKkS1kUhvq)wp|naLkOGC
zM!;Lmo8Nv83K)~#`!ChB4Zenr-J)q4Iw6d3GzZ>&Lu7-@ZZ3mwmTxH!o;-V*l%(mU
zo^%cfBtwvI5p>rgDGVgKwOB(XuOPAoq?Wx<X}J46Oo8b7MPo>mro;vVB5M8gZm0Xu
z2gVtm|8bKnkekFS@vjLSJQ}taMnG$Mr07&_YFG5V-EjffGlQ+)mURsZ?oV*bIc@~*
z?QBp+OE0TJXDn3>V>Y_YGT0t}c^oL_z(+;j?PPduDvU2{3?;_q9sVwM?HesE8ihwA
z>v{uZewx=nknD7cqhA=-PuLh^)1IP$m~ez<=y;Fa-UOQg3S?7rhsO0FAD1JkSex_;
zQC4YIe#eE6w0PW4q?Kp6x9cblz~aC*y;VanY$t<*xtB$IZ}uAQOP%re8;9NPyVVvv
zq2F^@;y}@u3nO9I@5f{B_<j0ARcze{s$x1v#A4v1xjbRVWSG=2bQ$8TQS5vJc}Mwa
zU&1ep$=zLUnx1<8Ow2W|2vmFJ%j>?U^$B&RjYX%QI=NOl!BvvrMFlnI;d`E-{<TS0
ze;xCsg<ka_pORUn;kV@d)r_zmla7zT7p)s~kxMk+g59{d_I%$sJQ7T_hR#}UlNXaL
zVdrgUIu~GpRiU>^U!}t8*1S9Jtt8p*R#l>-ufd{Uw_(xvqIaLK#W$ZWTNh~|;kp@0
zqqiWST+w+UqTg*h8yHYu&e%Af`A%wkBKY05dLLn|>g(jy!U(l|ngR7r%2ptmFFh2I
zn~}5sI7wSr3U1zhX|dN&3DCB?X*AQ4bZUNca!`x#T?8F9=~Z>Xpal!BKu#O*oUlk1
zd)apJo_#?iyH22tU-A3TDhz2a$l<7sFS#1UaIBsu(T2zR@`K32gyH6(SBV-jYmcTQ
zePWgpBxqtIL7{DwtlKP5bL8>2@B*zY?u;{iqj#Q^M@SGmy%A?gsW7t1t1YEOtS3%&
zPdB-A8%{g%l-E1XO~&0=_d=7*C3&?EIc}B1;p22`u=44VrYf`gsI@KH*Z{Q}UxI_u
zQP*jW5)KM>X>Fi@dhjt^ckAYjiSUuff>zM1uPourUTb_gUCpw;NTRU;e4Jgk^<fZ0
zF+h<JH)!yEA-}c?PN>b@AK*N&BJDWWBHG`)^URaZwqT*Hu9srmL-f-2ncoJy|9O(I
zw<iHIZPRu~>&<VwL5dY2wyTDQjgKw^Fam)l-uFcUlv*Qj_8IUj{jR7j=Lt8Cc7lpf
zgigcZvgA|P-$vhL<}52);I(l?j@}Ita5nxNc+6Hm5lRtC(xePe*NgL^i$s27fd7w#
zmjZ7wtv(!gz^^OfY%4uHa`bcpFFO&Nj^dFA(s}Z&OD|)HH;XDSB{`}+A?JNl0oy)b
zpsC4Ir>&-7S?}$llnCL8ARd1Oo|<za;Z=(x?n-pL>+GINCxoGfoL0u5kBvyKK^QP8
zsz<fnx%1L=s{EMzW+AwQ6usgoX-MMJrCq|zX!1SPkVIGK8Ebe@Fi8sUF2_|gUs5cw
zvr#els&0g?g>Lim5SKX1=g&}F4?b_rTM(<*iVQz12j`COqW^Lk9G$(|q%E;CKzybG
zL5E!7e$Xq@QT*Z4S(>4AOZE5J6iZuKlTunn&*g)rHTG$Mdzk12{d6NoLm!S^dWkzX
z>4AxV(U3;0Nk}MEJ(MJU@f>vw?VJOzImN<tTbVlDF3T5_S}!ZYkDz<0vvpTJK=uEU
z>;vXI8*^G)##=?A8hRo3MdP<MtI4YpNZmuMf6SX}Z|Hdl^LKXJoao9#L0?g|<j@_D
zb%ehAc-Epqpg>>>Is1piuI__5iMj9Dy&o{W{x*IvcAZN_x7La|-->7z3_unRj*EJ5
zF?ualjlwj2Q7}j-n<gw@YaoMTd#7G*e{{^X?yWOYzZve<i8gKkllm}huWsAupGI|0
zQx#ucvPtpQa(gu$F^zhZVE<Vb{aG++<<5b<Q~aWM&HPAqRVgqIDZMvU3}MO%Z5g<@
zrDPv+$7vuQj@dc^zmXhjJXPgg*rkTIdaAuffmZH7z)g6)@h2QB8R55&EqqgqHxsO=
ze8S?Cb=}muaSgWk71=o9c9#2xunsaxcXJyPxvNovtf?2jbKh(JdTWP6C1R0ICEo4m
zFtm9iPRZb{&{ofbf4HjEd=Gu?p$}1&hhH(iXSmVV;eOr|iS-~lBl>gz^rp}Ejb8fc
z0<IfD2)jz`nDX|%bKsmnQby;Cg@1Uges6l_E<v-n&c~8=`L|N-I9Fx$$yk-#;$?sv
zOLF~db!lz#H0xbyi}a4DybO6Y1<X1WouTssvOY9A{UEzo7M1B1_!QT_M@OW5gDQ%$
zCWO+hYDkUy*(2HP<!7~5Xp>2?6XTR7OtDt6MxItTBDjob1NDB~RG{%wkjzc-f{|%S
zI|Um$e1*5+-L2MYVL}vbLSA)1P@;3C_DR~zT=f73721RqB`%a246)7WaA|GP!!J-4
z&;oAMdeBzoeZzfK>~I7*dJ+9W3RLwwzaTRrN(6Ev>?+1qT3`~0Jv%})BfNPfuNvuD
ztLGVDK2|*oa?a{D{)!9oB&^oN+Zm8uJdaKMapt*}qxSnPe-T2m@5k1Biwl5C?5<}q
zj-O;(JH$rb#fsrcBW6`@e6${*uS|3lC1aI2QxH-V?97haMC>cZVs1S+nb`VcK}GMV
zTq=`+M@>Mv|GKXO(Zv_P#LaGs0^#CEKCARTM>WOAHayxVb)5*!=is|YvK)+h)HaWb
zj6MHgpHGPA`dhS$+2~P6=2W4qozLA5M2QkswNZ=TEWdSPXg;%a#G3N-Sqj;)zD$3{
zdN{+IHSAoeQt}Qs5NyT!;da5P=+g(CaI|Hc%*Lvq&D*l|!-xjhYxlV!+hxTs%1q}`
zp~OR>hT^KdtY50fSj`T{Ivo?c)!lFN{t}XubNhPBnLd@#BTjvslko*l&KG~Pe{D~c
zMR``vq_j0yXo|X5JaOyN@VWj<e-2w;!GvO5%6(7#(;#WBw+$Lcl>2!i&{aZ(RjH?9
zkM$P7N4VtfI?p36rqy1QFeYu~lWgOI{Reybq;*PtqPD)Sb-VYsAQ=tso_}V!W+1rK
z;-C{<pWK`al=o}ag7q&V*;4SuY&yIN8^)7IF9yz3uI2TpRzzxRPB}ew0w5L=`%wGe
z-eR_@kB**5Mun=;b(w$ago^0Lio8;u2f=J;ZyxIc%;3-z)wH80#;TCSy6Mc9(R5g}
z17wbQZ@1a-bZpcY^_fUVTYdgV;NV~VvOGVVF{R*Nw7lmC3vMO!jCtLG_KcfLe3|hH
zfPlm;cv{UE1de@Z?Q|)w=L=^Zad%_EW-YA?mHyLgJf?8PD!e-M432wBmapvNlp9qe
zC)b4w&6jSPvS;CGy{O(BPu3@LzQ7YXhKADLX%i#RH8p)hv15}bA-EKlF`R{S>0uG;
zNzG<C7;WKyrekM|D@BbNtPCimo8M@g#`YeiDx6TUKlcf5@RO6PPHp5;aVJ45-nzO0
zA9$cux=hYz80?j$nJhkV2@<Fd{78QEUeEEzBjvAZw#zkm(j+@^=kqB~n~BZ7rm5%O
zN2CRzdYoA{VjBaYq>mqh@L9W{mf!r5RhBLnW5j3IP3h4<s-sRP`%La;lSk$Xn_M)d
z<)xz1;^O)-qxtrd!Nd4WRr^A&0abPEmDzVe1#wkE&wQa#Z<>t>ZS6?BJ8_o-cskhp
ziqqHMs;f?!1(7-*rJAX|xs^d0cJzBDY2?$q(6~42^(=zqE$azfn-|p$V&ZdU1ZS-T
z`x?1jV(y4OziLX`a<t<yxdnls;PcC-)th*{wsy&-o=5HFE$=K%X+>rDBV1z63B=Wg
zhL#+qO5)}UzO0VE;q)U?y7Bgl)e`9Vbs|JJI+e;oNf+*Hq>0fcTzxriHC3Yb>0<LU
zJJafqfyIv~B5K!#6if+0Q=C6z_;?}DTeyGYtdXF70F9H_-O$vdZo8(m2~<>HIJvqm
z{&Ix+xNH@BNQQ!HZEH8&11lMS|BRs4jd%E?7u}fz9gRH;qv^QbX#7xq6ewByRDdDA
zGeM*pj}HwEC466!>66&d*J;KVOLaHzgkTj&nVT8~^EHZUC2Y+@ytfdZetY-o%@n(<
z+|BQtLnIh&E;%O)MD>wLh{Wi<T(FN@Dz%u(SNy&I{tNK&dl_|xbkVio56)eslInp`
z)^i0}@q`c1JE_LeS?^Om5jtCj7xQ+#b7&%TiRd<vFD`m<8l^XJKFgSP?dW@fp2t`3
zL@UTQY)x6T5xb*QRpkhuU4PCy%1XAX_xVQY6gXo|ja6<+?S-N<faXa$^NI$|690kz
zhP%pvcsex26Ybh3yT(=3E=j7a{N=8_U5NUmuSMuZxBmOZZ=W>_bpb5IXQgvmK6BCG
zy#2mo!lfxI@wX43){?6F#qIZ)Sc|<TWll<&RZYpTHw(Ff8|kgjwSn`CO{W@!$@5I9
zaOW`1Yh44%&l0I6S@(I;!)4of?|VG^rY_#2o@~{{SLV(B{cMx_(mN(uHI^Cq6my-@
zH|rdC&Or!f#~tPAC1ZSm#j~#~La$LI$0}N`8=2BB8?=i6=zx1MWvOSI58C2mPGL_H
za^BZuohd=t04Vthm}C74(yCIXyT<CRWXgR!OUl4wmdf%Vk5Y8Do)4pP8f|KO1L&xh
zr_7S+(>PZ5rgHbHCKd}32|?>dk)6YQ4No0m*s&3({*JxL*+10*&GtgCIETS<Te`e3
z+~K~5TM{ltSjgTy=eFJRtzN5J`M+*?XT<-qX!I*!2Y^wZCgdo`Z}s*hMxL?3)KS#<
zQ|*A9j;#e-VXOn%kvV*LDvYn!eLQmT3cAa`nKMzp?JQkg=*MrFVQ;%A0|Rj4&4wn~
zinr~=Hkv{|&RjybLQCiw`thmr>8n`vbET@=E#4L_D@{i!y7Z0DN5s|>j?(z&^KO*l
zO$W(lWj;i4{)Ew@!O-i+wuT?Nu1P#6FBfKaJ<ahE+jb7+7sO1+PC-Qdg@Y(iYrxnp
zkjg3Y+08R-vD)d%wHzM2$4%5&{LC|WQkuh~Z9ku~It@J4JoSt+puB{yyD5MF3KmY^
z<|+;PGak5ZGS4bmR##bUt9NMZ2*(wNer(I$v7FbEDZ@^whromzY``wr&o4eJu(R`&
ziBeM*M-gRBn0|h+l%FIXW7cHz!kmmG@!8e;Z`lc&^1rl!!z34KFg{$sj)O;O)1I%4
zig*3%VYt7JDx)JXm%BcD^zmHtGjzAbv=NHL<N5Pgd2tnAUvj_nkA0DeyPX3fJK=}*
zX6v;{_J%2q=uINB-E62SOfe;0WXC-m-)9>J2CQXSvE$}#a<0Qe9I{lZv-Ni-+h37e
z`5gq#_HI%+`8%ua98n{h!Za^b*&~bEH2YAvB7he+{N?SSOI#l=%50`;n&e~#QgGC9
zau{Z-2|0{Z-Ynmma_Vltu5*y14q=4tV7UK`ug*0Du=^$t-Vq#4rDt`Wa15ZHy@vJa
zg|!PP5=V0be$bjnxBl#>Ynp%w`YQPsZeDJfnHt&oEx<x7&wY_nV(Ue>@n1*oTCej`
z31*blt|<@uRZkv^D{0a}ku;emv?|)uQpA2n9?&AJ^1xB$4^}xd3xKY@X3<p=$7Z%g
zSYcsmGPoy|Q)|7)7OUq_MLf7{Ea{CD<QWUl!Ctm+-HSS{y)1o_l!QQG$5MKZ_ffA(
z6D@`4*^F<z6^4V+wK7WnsP6@Uf!>6no0<;@*15uY-{HD-vrJSlo}#VnwvUQZ9TtIg
z(4;ar>c&Xr5RgcLB9Thez>g<XnBTZn`AodOgnhXLK$)^gHuW5uREONNY}P|tWeVa@
zi+Wa9r9vV>#I%R`3)}xUrS=mp&0vT7%A?T6emb)^DoPO-g*{n4csyqdD?ALoX->6w
zJV@<T#BpmhN^HD#!|*_U0-<0sj3=d%4)J!gqP)gJxxe$(J&a9)Pmwa(S`NE6b3y)9
zC~;P)X1W9V^p%;4(pi`80v?-&>+9BE*Brk_rlo<dpf`K&3?OEFtH_wUGz)!H%MPn2
z`6IVR^2CcQp);qLXYLF1x1*Xo*}pmu6qP9L_8lHI*5mD*nai%OdbeG|zV;o;tJVEg
zX!I{~%q_|oy`ZU9;EiI-tx^`HB`aCpfL2pCE${TXG<QP(O*L?U?s-BJpv*PXCe;58
ze}XX(qeGlPRR%_YpTKaM&MhxI9-Y|$QJryQWSzX@8$CYX^(OFPTsqs!=cp;7z)j2E
z?W<_k9n&>&YXbNrtq2j>t`SyiO$nfOI!3{C)7h=t5VMWYc!%of&8&me^u)0$!@~eC
zWWW3B`aj^h^}q9Kf>6{n6ztTMo<s17(C<Ec^~^@8M=0C|h)3b(!nWW2_B_O3B9_ix
zcc5bJr2{%#5pkZYPPKC#l}%*f9W?2k0}QHletFoOl*Ca6H#-AiG3`Imr`XeHU$^6W
zva^Iu6;`<Aw$zpZQ7#rA2fw;y$>5QIWzQdZ^#<R3^~Y^^GSy$W2W}glW(0Mg2<&>(
z!qrFFjFV-(AA%-Qj{i+#f!zL5bZaUN+zzbHQR^F?nGU&G`#Gv*sS6ozkx2#Bc2}wI
zNPHeao+0dzi0i8mfCaxCST2Rzddys~SNsuJGA=!8zb(FuE1AzZO!E3I5K%57XuGp;
z+45ao5NsA`1qQ6<eIOi1n|=frsq!wi$8#;sCr@$HZB}n0K7!rU1A?s_>`#G7GkE>Z
zP#Uz~W8u|RoN)$1&vUvD1Cn$JD)gj42@_FP#-}##K7GHgnEoU0mFmuj`2%NpgXVR_
zPrmG9joVly`v~+ehHezXC&@^fzU}mzNH1t^_Wj<e{=<{izK&o+JM32oe;XL4NDwLq
z1>D=|A2x8V8gtLCtVU)mLs!BBilUyG1RuK{b|_CX#3875*3-@mqvF^M%3a;lQZ9w{
z%KM&Oez!jko-oZKWD8Vys&ksPr)q8)z@K=8tMM`9F()J9DO&1fQqA&IZ(!lKTY2YF
z77v^6xvWr{ON1^&g|#TMymz`6A8Dy1uu})gRhO3f9*7dvY5IO`NFp*9@h2ECrJlTT
zQ9Mi}8|MCXoxO)gSv`0H%%3RRgNOUqVVZ1R;OD&QCJP)bI&qmq=_>YIqFIzg>E--$
z;;J|DfjhztgEW2r{xMHKGsHECg8lsWv(>kFvt>ZiBJu)M<!4LxIw_=_wAF{e=yl^J
zlo1+8QUKBE+hr2Y+TyCrO}AeF-6^PKxrM<V%;L_{y=L3Au^eTkjo(?HON=;nHr-<B
z>d#!}QpoL?V7{6ksWQrHMq~GIg?CE^z^i@#1B6Y0<Wwk1LPO;Vr^iLUvehev<%zO5
ziBEu#&D%b66xta(I7e(<hS)Z?I6mA&4|m2&IfWZ4i5CG@TO3qh>iM|!>gftMi%aj5
zGRK6r_g#;*>3_ohFEiUm8h}+SC!G3SQuPg}2vZcy(mixej)G4p!p{_7-vJD#t9Q}#
zIn8H0V37~M-H9SL&8o?I*Ldm^*sy70<^iF0`y~Fp0B<T3UdwyE4e=nj74eE4JTW%S
zhC?<FH(3<PP2Jt5%#+Xdos{9FV4u1`hApBEf5{t<*o7wf1Rg?6Hl2taCP<z)X@CJr
zAo1O0Obs$^n*dR=vu<@!PRk+sAs7ewkPmQXbXMh|h>GT6rN_tQmuV&Hih80SbJBiu
z)O$+%s*+_R1+1(Mh2ZX$9C#?4cCaGU3^}I|?0Pzm#?$dgx5~|6oZwq&)2E;P(Ubcx
zSoRtQHS4_xmqI_}I5SoDQCL+8-5j{=lV{WQW)VREL`=$CC1<Y%I-w@fneUjilwzIK
zH9iXgKR0uf!UG`$qi2u4qQuc-gwK5Va>Wi4CN$}(r$yw|!<?{`Q2%hwWA4c48B_VJ
zaKIeD^cQfC{p?U+c+qO|fV^-LcOIAd<Ic$d&eVtow&Gg6q3r;zTm?9!#MgGng~y)P
z+cW&^sazogF2y&&ym2*e0-Rdaa*Ds7qKjrTN^fHlGz}!1UxuP>HKFL``JO;+&A^D4
zCz$}(up7M{mxp507Hxmajx}Z6=}g|grfqM907@IbCKN+-#+>tT0=B7K-HIIXES^bU
zF~b-DD?Y{Ir>4Z~$Z~w%sRH6Uj;xlt;k7KD*v~2oRhwch!kyPb(~j|D#%)2o#@ArR
z`${C3=GoOpT8vux^<IPuMy{nv0w(^`F~$BVXZWdYUkH+aF(C?rdE}D{CC5*kaVG%p
z((?c22BYa!Q7@3}xS+cpa`n+o&jFq%+qLwmNb#rIPI@*+skv<E=_wmy>C_u3(s{<}
z_Y#@;KaIowmmsk4l6)$ZA#d~8><z{e2Px*l+BV*Co9n7#PPJrwMJb95%yK?ngmlYp
zAdRumII}AiB?Eq&>1~2)PNHS`*^Z)akdS_eexY;R1M-hbBBq1<P7(f(bIp!(@nDVx
zCI4T*ilc;JCwr#{3@Hhd3mlFiR5%ClXC>CpUt@f-_RhE(XG~US^5vCQRp7x=RdlnV
zp}?$S`5<jVz!@SEeP8maDM8+M4eV7)fxq&Me95_<l5rg&<>EP9fIi;*L_L>y)wnv<
zSjT!dCL95tM=K9f9w`w2-YpuJYXaUk%d?xoKX)v32CD%k95#_ExrC=t&!V<rla0rn
zUR@ZLi(jn27juZ}%ps~BVk-2uY{#wnGXZDN8T_Newggf={<Z>Q`Bk(vGKhWmaKM?7
zZI=<#@Zl=n0_|4L3a%8=f^Q6vrK2}(#-3?>PDO)~<jE_6o>Z6IT04zyWcJ*AT&rlg
z@$~7X#o8<*Vk(0}R3{Hn{UmlWqsN7pCTQ};YvYaCb8=yu_bl(=j#&qe?$@TTv>^ec
zg#LP^hq)+TF{*eZh}VsanP>=n*3<U5Mv>3Y<-W9R(ht@IRN!Fi!s@M_S$=g`J!{N`
z$0*>A4~T#K^_CBRA@L`5@+YM&m8~Cd+!e0!?>_kbz>nsK@#i(RVFrJ7+Sarc*8fS|
znJei^jUR5AeUYGIa@P+CBE49aosc3JFW)%(kW@qJi*1K1z@B%d1s{0u_LAr>ThZo@
zJ>OSjNH>Z?RBEbC{Xu%KmS?YKT*{|cR~qXUHF=uxAipEF@rbhClOo>uGzE>Lw!K7c
zLq8Kx`KJ=5grwtZOW)X<@9|t`hKL*GSsH2eB040?1(>*E?my4p6~q_`2$~3p)eZ5V
z^<i|VB#^CB3N~L)nBtR;r+HV&?Jc)gLRwk2kP;am-?C%h7y5CPV)gEfF?+fLfMO)?
zpE9TqjIWgdchax4efr7$laN`@SCv(Di0aLjxSF@%?1w5mcIMTtl$mcR_361tpCq5e
z@;73E;yPM#C=KbQ!07IX^F-q9=LLqQgHmlDiC^-iZ5GlOIQ@2LII*W`_6aE=r2NzX
zV}hkw=vj1huV7Y$gtAfj@xeN9WV>$dVQ4ISF1gXYwKR0=GZY&$F5w<sAh(+Gwpc$7
zP~{3i-Fd3Y5QLQ(Z{B*LBqceR-=znCPxh@K#!Qsd1on&-S95+tBC0NrWHpKSk+$iB
z7JK6HuLkR_r=mQnl#=}<IBbKhtNlgmq=TFj)^1wX(3=ttQrJ#{Yt6be+KsU!UQv+y
z63Nt|X6@U=+s|OKJ}_O_O!G;~jCn<BKi;yp>A+AjVi&6;e92a<%1l5jZ{>x-rCTiO
z_h-D1V%|kcP|mnR+3A&YarjeD1$Yf^i!^=T={J|QKm79Yzq#A~JJ;wN2BzifDAzxD
zFxSslX7PM{MgD+@K;d9Dj^~wpph=gezusYItsB%|<3CjSHV~SiszvbUaFf6bNHUJK
zHDIh7<K77k59-#|H54g@=ZEjw_;p^P_uuU2Z*0w^QV{v!_4!#DN2F<8RtU+)Ush$E
zC#HRk5o5KHZ*LeY(|b?KJMd7pT7|*(8s<X!ctE6HqNCsue=5Rm;fpvvRg(uSS|-=A
zwS<zXGI#>P`$l*n6elUlhEJsGL}~~<R{Z?#eTtWq3IwSKiRgK$V_4pcPpgxVAD=i!
z6e!s10FAAVc)d4H;gj_dhjS;W6bq!sT#3%+Dq_Rv>$4iaQitK48g~N^o;!$jrbY8;
z)ny-kbcXPn6BV|$GE0^Cy~UJic~Ne1Trj)&@o`1>z>%0c0^RrhMAevtH!6L+i6|{T
zeWGmZvh7R$D3SAJkKS^~vS27ZTEV$$4VjbE-DQ<lKm}j}Jyq_WAHcl|^mT6>xSjEW
zXSa;!V&JFzg;v5;J*UZ`UXx36{4b;%zb}M+$&B~dmczfmS-(z1eK5PH_A`Wj24+)l
zfY{J$$u22&`Xyh1ADLO4-j%hyLj)XS42@J(r%j(r6R<$q>Tgj**;*l=Na{Hj6`WOB
z<fYgJ)&-dz)kRr@C4rVpxOb3U(8+Rs+<Lvpb?=#&PFiQ!B3a(Oq=ty`umY{I6!lR{
zojHT%+M>_{F+xg=k@4gh0dGy$BW449W|tKt!-xVueW5dIcHbNgF$`?;3SD}-pXeMz
zN3`F3a``)FH-Cr93kS3bvy)JrOW(;A#>P+&V%tFPN7hETY6Ybbl;ai$MW4V5?{d`^
z{wjYvmsN7a)m1tiH)emylxUoUU9#?d7X5wkV>{?TS6buHlnNczwyAg013+Nee@HTB
z{9SM^e)a6M0`5X0x4roxqOOQ%GnMh7xiRv6@f;m-X$Se!7OO8GF<;w*`BuI3%+>#a
za_%8)jD7Q$)W7jX<ZFnq`Io}i&^v6wthpnRsHmVHtEK|6n%Ds)%?4BZBaGo+>1l(<
z&&m2@cpp0Ql1=VniVk&$5%?Vd|Cnq#tU6cu_3&AUX$cT!9~dcq=|?;ouR~jECVvEL
z{hSd)y2aKq5kh>EHcs!yc{Ek_U~Q4Alm||Ss`s*I=a_61oA-=Yo$>~~#&(P~6?gc`
z%Mi0qlsf45-O<x|sLRJ*9<;hlCXF^m+aKv&#`Wj%;{8PkGGnG>CFj2vv^kf(cKhsU
zuD4K{Gao+Zx2zt196fYJ7R6o6E_5qhz0*o5%MTa0-LlpoW*fR-qnNMUu$*F~B6vfx
zymX=0Gf^g3QiVbBfMQ98?Sn%(YW+(0&n2?4UKo8#^IR|XWu&kxt?AvLKYOI8iu%*H
zfAWj1NLwTs-cbw6X`mLmnyz`z;4G&Qw~INC(9Ks3n!(oJpIlAHmyw2s-B&FNK6Ik7
zje7m(&VH9<7f?o4o;t0|ZX9d1EHi3z0t;(O>#~hL>)oscx0LbXt+F@MTG}4nV6k~Y
z%Ie}UX8ACoU7+*ScB$(H{zmm7f=1Cs`Q?+TGDDZ(|5nG}URS!qfREdFX6+z5DYM1!
zT#EfR7Sk~Mh@|B^$M@qy+KswI<#@BAz|`VXkvsC{O40@WZo)|#hBQHT6Q>*LzdI$7
z40!0^MQ(&R@@}zaTZ5~YMu(tgBD+j^qU|!KElFDWX~T;Nj%8Z&BMT~0F6RpuZk=LO
zqNY>4*~$B}ZhTW6kmUW{lBE4Eo5Y=EJI#BQ!39TLh;-7ehK0%I3W%5Io>_D&8Z@q?
zJZ1ffCP^aM%G7nyxqiOheX^fId8WBgO1$vRyL+#`Og{+uWq|abfc&plg83i@C1n1+
zR`Z|Srxb1?1icb)98)Sh9ug!<B>YS8)J{rZkUM7aNQkI#26^FXwEWpoV{XA)T#P>^
zj!WwN_lE-hlCe9GRUO>r<Z6xn{5eT3-&s6Dwj4=vNVF}O+4On_&3Ah;QSPZ=<+r~m
zOP|K-R)H1m81A^Iru3TDuZoRzbY#S#+wgxec8y;G9nj&Wj8+=tsQmuM53vH4M^_|O
zg;>8^=Ik88`TNikVCWAKIdI<vJtv7l(@`|VnCMO@cM=M9zdo9Ur7ns|>d2K;p0LUW
zR^OmDFm~ofT%Je4nt&WJyX_t?;u?a>tOi+8htTuVCZNY>lwgyDtsW+qfDm?Dg6vGb
zCZY!Gzmqqx%(boc=jHryU4MRI+=GV8?vXyF{1+DK8#doXlBG%*u<ErCyU>7r7)EUb
za7c>tOfHYeBFzXLZDnCLkPC%J5YS!%VOmz3ym}Ir(TtE%Zr7pm(q_MX#hlwY0o9S1
z+5<?5f6soMa%lnHCjGC-2DErm6L)nL$Z|y0rRMTmE_eI<_c9-jdq0c{w8fZp`NxcZ
z|G8KG2iRX^B>PArw%{L{pb=8%lSE>RtatL+2-yW(P`lG|KbPE=fxuu}gm5T&VrfQU
zP%Vob^W@WQKq@i;b%AxyTc&@2U;cv3aKu4U3dct8Df9pRbN}_b?Fec8(1|{{tiSx)
z|NF21{|oppE70Cegruw^JgGpadJEo4@cgoo|39wseZ;6H30MA`AlMgm?cB(N^UUBV
zrVjaqTWjm^*yX}3^V%I-!<W|yk(R6mDqBV*^kES+UM20fVUj=w?fN|kzWU+YHV42n
zDL_(gMbKbXBk_{#XYkhn_s~FmJZgd&?C6nrpW7VSnW2>jh=AHB7X&<6eqKN%M*&jR
zVP^D=L-vTj^J4$wm2dn=+1tIM^Sf-x_zearW@<3!88JrSzO)GJAVGu=Yy#+5^6ic8
z$UY88|DX&65^K{aFt#b-678_Jo(CBiuK_VJ80cXfNZ*3w0N8uxfxm|oBbXcRdm#a8
z0<a0~48z7lyT9;rzfCz1hGCiEs6?Zi=zH<vj;2^7$q51Nae3|8cNInelB9qqJBG|8
z`QHAP0Ia~qm}qhOKZy$dI_&xN^^uG4)+n0l(^Y@FFO}`wzPFB4u~z59LoBm9MpuD?
zX?9n8v^z^J`uPY{-P8y*AQwD#L%<V}4M|_J<6K7Ay`#fOq>Jok6{Tp}9axAEz{bR~
zkkmGS7@f!h&Nd4uj(7!zz~#s4Ga?4KiAH9*K9wC~h=8nBdPSX%l0rw?d3Xp7zrzO&
zN={0;>W3^DKsO#5P}Th{DBwSC&%fP2KNBQ*%w)5j+Wfa0OMjWjXD#!_$20{v3Wu)N
zJU(q&1+pwj_z1|F0Hl=VG#!>W<7aRrnn8?DJxRLPkZJrV?PufTeIHdkcfSny08U>G
z;w&Zro!;ca&Zp6$KWlW=uM$+f7}=GCwxTqJz~+Aw4gZgQW&zHwMH$+@Kf0`M1aL&(
zJ@6s*;t*&O?5q|*y+pyX(^Fque2R7l{A<%4IxmU2xa=Dzt-|gqff$0pEt^vOHd+Q`
z&#7EQXA8<<0my&2H#7Ro+3NrNfrb^M_W`V%(Wgya@V6(xFA6mhwT7I1$hqSF{dGn)
z#MEdMR(=3IM)9jdRXy@Llaj?I^CQ>9iw5pA1;G(HG``$ju`j|4iOi!MEL*AwWnJd-
z$E$`vRsd|vFSrXB>Y~#aMHcQCs~woiQ48!$MO?PDNJg=*wV}aEIS}*Mibs`ysXT)K
zx?=rxhIE2*Vou&`Wt8&XrXuU*tLf^b{N<ab@qk@dvxoB@mEHQJ5ezwm*%`c{WJ|~a
zU$X$A9gCnD39119?0Je!2=BUwb*!bSLEM_BfEwwQZ#|6BR8F4=g+w(OuQ}@*WD_C6
z4l?f1$3;#J&h^W|?<8!Cc0dSZkE=j()KHP}(-`4bye#=%;Nb%gt=VRaz0Y_zhwTKZ
z0SB`BCZIOWmMZ^7Hh&)@yKDf063o3EB%ot>16hG-w;=+=9(a3n5vac*Z{dSzXX^-g
zK2dVrveSSbAY5l!p;t`TJ|9|`gmLHW469L*s}T|x6e7FI=EiNf8dVO5Cqbqs)<ubt
z<Sae~c8BO`$8|H}htr51hYoe;24BojvYRoStYlQ&myaN7xC$w$`64rV8Bgb5EN~bZ
zjVv4)hLV}z&1!Gx@lgL8b`5;y1O6_o8F@6BUFV;$-k$}_@1@mWUl{^T(a2rie9qs_
zRgS0w)uMPQToc-tEviP6G4-U9GcM8<c4fNDYp0fg$<ISsb&N0xkwYL`1>eBAFF{4q
zcZYY3;-PSuO?8BsLeN19aaJVMn&81v_Nr)~)fXSk&<EyWUdH$IE`UBYuu#VwD)gyw
z^>AX$jH%(RQ2WT!Qn`mId<H`czN6zt$i}}pFfB`0zM97Wl`jZI_JHZ!nIiUl4)V9<
z+i%`Kh55*UC(F6~UW>$RSEH{C!&|Lx)i78Z^A{IqKYC&i+r}XStmBY_^{h#!4?maz
zv~E&;G)P{E#Zu|+B3I;AE>z7Q#UuA{YiK+6>&+qjUZZ3Lx^>Eao6J6&qkQa>daNBB
z(fh#LTlDfHGx50rE{5TkH@p$ihv8-jgUvEUPArs9+m${8XpSSu<iPMzn6b#AAyT|8
zj)KMW8{0j8cd=zZFJE*HPGP@CXc)821a1mRZfXZNI4d5AnS>vpGO^fCsJ_1uw{aKA
zqQV(|Vwg}~K4f)X^OFeTZ`#&rK9(RSRwg&uXyDW`IVL{_^{K`oe0<U2sll<m6U<NF
zBei5WUvWcI>gp5iiO^L~l-Z}aN5?Am*WPv}=l>7x^k2?+Z)A2--bP6Huh#?~lCnlf
zLa4C;eUSJOleQygn@6foDXF2F(X+eoK)cOalo?}!0$_aOPZolfH+g0QW4v{DBqDSa
zW(+P(E;yo2U+qc~X7|ckLsT*v$0ivz3Ys2Wc~pq}cQ`7Y1h3@N(?^d|<*~=2@+Mpx
z*g}ewrRFJF0-X?HxPBoJ5`Ct+C^2T3dP=X6gCCmE=kbPMfPDe+(=`;Qe_Q2(NEHgx
zacfFN<qO^mna}hm>&_GojL{B2R?C5-RV-sd*g;5h8--oOg^efnK=`?vjS{&CYquKW
ziuQap`N?4~enuv_)l8Db#J<R}XwaYw9gg$90NwoL+a?8+F5rIl_X>b69W8ei?|~x0
z?Cq@=R=BX8G|fjgDCX$hTDAH(2<!PJ{bF+Msmfk<Bp(akjPYvuh^liqA5kfIi1)Ke
zFGXgk^hKgHWkThg4=^z-p%+ZjU*jIrU1m@(->l^?ZXm~;@Hl=ta3T{1V0^|jE3m=1
zT?`bY!0F_5iTmuj-&@D2=DjO|18b+G&Jnc<5*YzH_J<XDhq2nY4LQW=WpF|-*||fF
z&jG*8wu9+mH{IRtu9)|50d?6Z6c}U^<Ew@_>>3lDYY>v<?wKd5Cad*tr|o|~i&ata
zB2<4!Km9Q3m3XZL(jUjK+^WuN4pd4IArpGf+r4;KTT1=q<c7e;`(C70Y7X>viGzYK
z<E&HtlLXu=yL;Q5Mm2!}GxNoB_Fm%*y}(b`FXZ<Tob^l4b8KohylqjPU(_$xa`bwg
zOUw6^`p$v%uDw<|s(0oVWqfDj-X`b?HZ@<rF({ZT{QSB_?55EBq__mBiQ?hRT~q(U
zX5MBIhY4mb;U+{qVgvFc``K8C$BW76#V6P@-EhADvA7lND>74)pUn3|t$FtSt@s2S
z?_F^EJsx#-_Te$dv!C@Wy4Hn~v}EHoKkFrfN-H6>by?1m>u`&@aC}0M&{=|8H48rk
zG6RzKGjodHEN&PrOc?e6L;oyQm-&=zZ}N@(B(B{i_1q^3sv>6<Ztq+EEC&w|4@q^^
zgamBPJpH^QAoM#H=a2dQ?gg+==GoTr#@HgY<w(@ho@nON{nn+S0Z?Lm5Kr*!95+2T
z&?hVUrh&?ZE}wsbqp|SvisioKHsAkME5z}@mzRQz{etap09U0dx37rUzJqyKnj~-V
zC>L#$P0`~yWgIO@UQdX=Nmk=z4V=`Ib`g1!I@Dt#%OFfKtJrF>FAh3O#g|qB%qYr<
zA?_siRQ;EkQkCV#gOS6z<(Ab7g}m|10W>1K=?RO1#xrmA;)Tq^^m7<wJoELhS-To&
z?>JtHn5?i$PpBK5X@x2Dd<Si&9ajUxq9{l~oq7;5+43;t;0cc`DU%pKpp5J0isjw@
zXu`s0o#$Byr)qK9`DIhB5s9QbjVD_NiIV3JNiv^juJh+~tP|n)DB{764`v@VHM}pK
zM2Y3ZAAe3dG;ra@{)W-L`bdGT(9$5zX7<@AJVAIY7saY&*p6Y6+i?eyh{-BAS8KAv
z(<|~K%`azROoNs|IN$QqV0#BBDlEBA8}43UjvzRlQPhf5HrBM_9yvuZb*H%b-jc#z
z>)y6XmHA}AI^&t^VenUS{^J4p?++64y!mZ_BgZB)neA_e9F@2ezWure&&J+-<#iwq
zl(I-@3krY^$X$Hx%T>e5pu{gmfv!sTAmX;Dgg<iLer4&_^zoL!fEaEKS3#DXt6*O-
zxSM7@>}Y8dyL~t)z@nCa05(ZJUd9IECGeh{B;+@kVZZ)Ha`8xuq#=^TKscZ7*mQBk
z2rRGevl-je2}QOU10yZgIx8=@<co-Pkd^cCf?wVw@d+ZI79YR$N?YFn<{W+V1#<!G
zfce;u>xxyudih5z*gA;)w~|Tp#oYnMD>2@HsMcl6xab?(dvH#z?}p5pBsm>NQ^-QL
z+`OmcV6t6|oC8yY{?wa;)m0EU+8Xp><^#rUEQ(&YN04<&N&4wwlFSCGrR^;Zr8MB1
zkih}#xzGJ~LxO+awfxGtpeel}=fn9g*vj}NRb?Y>DwC5fLtVtNasL;fAZe-%z_KYC
zxB|0@q8MTdvJUYc6{RY9jG=gPw+qy@Qh~#8z*h}$ig`~A^bbta9wxz1v-cc_u!_!=
zaZ1#oVI)6r%yJ%5Auh<a%sGogXNVN5QiVTMJaH><RHAd=Y=EgsNu{7{zn`5XeRb8Q
zwDE-jePUjPhi)6t1r;*VM~jHn5eQDdREw!bx+OQaU5TFRDv<K7AT}tg!nAs0>Qv4^
zgA9xm?RzEAf+G2Zgk0HsUKm{0q@~b#!2SJbqL&14vJ3=l_@2j<_5IB@>7Q4K0NNEz
zkKZdn6&?a!a(13`Jh;bsP&suB_Oc5p-VZ98gOs}?SZ^sHi&Pt5XZ(@oun2vD3)c3-
zSA&!fpwJTb?$dgT1^><}TAg{c_y>mEH)8i)s$JjGL@VA3?(aqlK(FPP<Rj4DSCFAT
zQpTRZP>$W%g1+DWRCPr6n3F(Y0PMeo`i#5Zbt{gG4mi{ZOI|9ZGjboz{c`2vSD)xM
z>d&(ovTLvV_AiSa(icYsfG)h>?)kffrm6UJu%UtE!}rPL*g8D1!|~t}hukpf2g|C3
zuwfr_Gnaw$I-}&n|F_cT3MrBsJpW`Bl7n5I)6R#kjDj5vCAOA<uUcV%0>l!1TQ@g1
zpmX6u=vRCx7b)GLJ2rGK=(uqDV~QPs6c_HTgUN`G)DJjx2^@7EINbL}l_K-9T*$K&
zeJUs7dY{7JB8MryOgObFRp5UA^y<?p;AN6KA|myEyAs<<_9;N)VpLvd+wz`%xR2;>
zACdJ{(NY+>eAR9x&>e%oIrGl^@u9fg&=T9Rh)pr_8bfica+0dYE&cxqTmF9Um5hxM
zsGwQ({BLdmmANMtgZRDJ)WNfKM6ys-coW#KK~$H;t7=^p4aGNMU;LP866jsT)Jj6+
z0P-UTB-ffDFo5wc<br<09nB0Y;(G=f=G$?bNEl@K%Cm){19Q+R@KaOBg38JSs3lIi
zwF1uonYEst7K3{oROzCFH3{0)NGo@0A*RK$vZ9n3vDCRwaQZh4j*B_%JWUd#hoO;!
zG)a~ilv~5(fR1_}Pb;>KLUJe!vJwu<_Eta#@KNd(-VNJgL+db$Bw~g@v{hbJ_7Yg!
zOwX6JNzU@pv5A(byj=%T*gZ-FRn;d}md2&L<5^^Aa{(sDVGxvFu3`Jv8s73XmCCK$
z0>Ec;C#kRl>s-42r*qPOKrVlpDagwqy7uIxf{uTs^tS5~q0O0`+ZF6CIe->2GjtN&
zxHlzML5fJMYmoYA5%$PnkoEHYfcPf6+H3|ghnC8*!^cj0LyP251P&wB*R8X)!%3zQ
zd}uy>jbV5*xPBTCC^xr8kHTy)!81HFe-ATX25*cK7YX2+RmD6kISZOD3cBEtC)C!?
zR+8wIQHre$xxLpVXZ9hHcivx+$Q4iNc-+o6j~5vnAIhRjCkPM3H5Jv->f2y`L3c1k
z?ixIZ>HDgCg9%c5lZ!NC?rI-%I0GW>izQzb#)K@I=_@GhGB`*AAEVMnLV6IdH*$o0
zW+)6!%8`gILFdM$C&HfwhuK+oM@ZzbZUgWx7}5d$DfV6>YoqCIAWE8pZh;92LEn6T
z&y*&5PuWj8q*fu&aH0gs0|(yFTUQ>dGAM}+3~)`hNn`_^zN$x0dkZL>xX2tGyV<}1
zt9YXt5O!wppg2731iiTDzS`R=Z+Ihlxq!)SuLPa+z%9bVOl<HCAzdY9xZij~fse5a
z=0O+;ZVH^qpKC?7l1d}7=)6;os>BX>jL9MtK{W1eK43qwQ-W!Wo{?9&Ac^T7E09j+
zZhT3$#IZ4LSPU9q!Y!^t`1cd10~4XI3t8Z55s*a+al`NhPZ|&QAu*qaI?L<7B%+X!
zN0(SPj9~qL>AU{h*L#<tAM<G=xKQy6_N&Bk`rao$ZH=0V1wl2Wci*B@Tw-DygqD!T
zb<m|d7xKcl?4b#KRiZcuVX~)8{No|z*_Ev5vGOZ~(^7K($Y+KnHr97FbA0o<q20)t
zLwo(!a|ftJyK4#ZFfd%dk6M>q^rIx|*eI;nYf&~qVz=ed^&ZKac<#fUaDHnP0sEiD
zv3iW1om(jw_rW;m^vFR<>0AYI)tl&Bj__#x+~L#I2e5;%8O;Oez<UBgrVaU#AY~c6
zu{H@2)`4{ob@HR!(~n|p)Q(JqS;)-jOfOqkHIoCiEhZUs^@0OXNi&Od1hyj?wYKP(
zY;Xu6P_UF`Mc3H!r2?Kq>&KRp(pgy!%Fndd*3;nMaMo+#?l76xLoqcnf6b06r%72{
zhkF6i#$h4XwX=x?{}f=9)w`_0%9>#OMfvdmVec&is{FRKZ|UyNMJQbh5RgVnlu`si
zx)z|Mba%I)f~0_e(v5U?gLFujboV=#|FieA&vVY%&pzkN`{jJYrLMKO?>Wak<~6SI
z8=t_`%xrBqat5eKo8y;Z$o3$CdrdCkRo7_`CQBcP`V_x+kLxrD5PDxX5$=Pv1r(qm
zlm_(Ku2*do&sXDF)mq@Jtqh5Ln2{Egq>b<lE}qGjQA@x989>0RRqrQ%b%;GqLlD{$
zvmQT4bTFwbL9k5=>@i;6G6^e;GF9jv80ct=oweTJ;4D1DLM^l3?KMJMpzyJu=;Az%
znb!R`bfKS*3G#oBdip=%=HEx6Yj}yJ2jkxs6ptT<LBS4?u}idnN~iO~-Gjfd7byyz
zONvi{<TYAw=8JGv^!!+jv=`uE3GoFCNjOp4Owf-qVsO6DFl4hlUan?24qI<i0f*jn
zxZlS|6&rB1@}fOf7`d2aID#CZkINkJnPtMzTdPr*6lO^og&37JC0g0E=a&p3kk*$k
zPt7E*60wf)f6%Wy_6ga*mby<m4(I!z((Bd9C8{7tSIxc)mr!$|Ygw8{Z-L`^d>+Vx
ze1wn@t}HD+)4zPtPYAgx=7l4YOQ(1}Hl!h#)lL0q$ZtX<GJm#Q?*UBHUNLeAPZ^5=
zD}mv{N_QO61u>Sd<|N}lvVTaN5Z$fcTegJo3F+$#;I(mc!w)$S#V@wA8IVp0N{GBx
zp6A9y$6oXee9{(%+?EE_x5Y_ib}e}2>(J_R{p?l-JxpDkL*P){=v$lzZ2^6UOAOv}
zS=5zK>NNyuqCLePTh^7~c6^ihcNZ3z6<U+l!`I?xLhQ<bz4}90;^x@;h96U~L?C28
zt}eVwpyovpgG!b~6N$)2<saM{>e@fbl?O*F|0Pxb7zt>~SBtFdk$>G@Po-ZV#yf9Y
zpf}*})5S>hkj-@&&2ge4t7(emG0?q{{sYqrP|^Gl?rR(ZNrTeuKc0O_45lY0Fmh7M
zdk{cM`o-IP0-Q#q>R!;3lCBjmIm1;_te+WIWzp5|I2axy>mvs!v;mBkr&;tZeOntL
z%~^-8Z5?-^mumyM%mpJ|m!Ary5L)wuBm#jcvD<3&>La?Lw<P$n1Ud0`!v~^*hZvhu
z=piSM09~s)6DeizVYl>i>@YeU+~nSyO!nCf_fAjEya3kOU(#M-oAQzgHMf!LFK-kY
zmULyPF|l?-^EQD9Ud&Y9zd)o1#p_u3^)dTV4N`O?*=p#ywqRNawIMCi1t%s#iNF)D
z2i#eRFXka5RB39Wf?gmut|q+zf_Lh>@C3lt@JR6yn17b`UxnI#_xex~kkNlnFc&%e
zy{<Jz&{VKXA56V8zHWQ={a=XPe}6;95AdqzNjIDB-(Nx^Y}%sOwSXs)wxAAC6Rxh4
zWK^R156HxSRA>LwpXJ!VVvyg8Q?L2^UsIz0syqfO9Jku$0UdQ){Q-!i1n{lK<8^%w
zhxP;={j*-bzo&$M`$3Kn{t+XiFJJ%tBVH4_06C=<4DTd4k1$MiUBdvw?r`Cak>q?Z
zMRG99<yzriyTt#dFfst=jSvGQkMM8OGlQzkJuH7Px!nQI6MZ8w2om7k&n_UVDA*Yd
z`j_^}U$HU51SGamD>-@P|5DcW@53nkEhMABeS1WA`BxVA-?`&|-Dk!`_<g_FWcz3B
zK7U^l|H%(AaMBlUNW<SDHvd)7|Ic4<dIFBYzn*y2JWcv<T(*CFuN9zABp_uJ{U_h;
zU%4DWJ_a}UeOlqjKhS9ZTkrJ<+`=`VjlBQi_y6A|@Pvs63>t@Xi$E?eD93B|Rq0<l
z7XB>;{C($BqAP-jnV~<9fcfM?FOBqLq3U_J?LYwjE%~%LR`JiAJ^$go{{E&i23-UR
z;2hQhp4mBX>Vg7nZD*_P7C~#^=kFL3{r~6R{gb!&kAG!QjL`?2)`mduR(nVZzUet=
zp$Y;garf^`I)eWXdg6a2+4(jz@qpH-AfQ27f0r|Uf{vE+Cd`rVU#iOfM-gVj7_98m
zgVt9g|Fa8s84Y5~%~{_EqyNzt@D+zEWDp6D2mb`3^&c<n-(N|;gR_La2X$Eg6{E%f
z{`&vd75Lw~|G$g>|M$IrU8>*>YSpdaP>T;saU6$xzZ^V&0HoD}K9F<40h)W4l)`^=
zKL#=S0xFWAi54l`U`d@>B=q;xm>l8fdJpP9f3Ua4f`HX}@D1l(GTZ|e3<1`WMF6@p
z{*zCF(>@ugrh#uUV44M5RZeAdBG55l6hbXVvi5=C5I|v;3<VvBOn<~)_JZ;k(wKXI
zo&XmIex5>rv}OoQ8+o1+f8~dgdl<mU!np`aH$`*i;E9C}&gEV^wF1J-#T}Gykps>f
zFMk#ZOK6#4zeR*2^%6i|7nu3d1#B`j@bNl^DNM3@$lvMGoxDki9V5Rfr6*lp1BL3O
z!Ip9Y*f~T%hbq%%VUX43An3!=E4U>WwARtI{;^6KHH!QH2*~6iO)(_6peXN+1D|?o
z#Uq*)qpqR#z~g!uGPqZG<pp${KvuQvv7K=JV%sWsB?1VWoV(wzj{pn{6aFaQsX^a<
zjkB@fwp{=>()dHtqV>zl;cPeYVVzoqXhC3qO18bdSP%ZB=f=uW`@7U|Se6V2&X98f
zmKvmwNT3)xY76ArK)Bb>LJ|K79Nz+W&*1`nv53zVR7F|<a&ph82?)RRJO}TCfJv5_
z8i-qqkh^nv`23nyU?iiFi>6JX&9M6OEBhiM#wQBS+DbQbpVa260LCC0*}I9m!+F6i
zr7bqZ47<XsSRfnJBbEPzB2+hm4_XB3`0n{^tZhqMC{Zi;ZPs^u7n4)1M-G60E|@V&
zsAc9QM@Kbi|852DI`L%aOTsg2O^tWU1TL|l{k7+cw0tmFO|{lrNQ|@nA>l^0_}zF_
zgL)O6|Ls1I4u~4SRqpI1^%|)ZIUbU(zfI@*0hGWN&{W;sVqEc0`T_nt=Ux+#m$pdd
z7sJPN!KMGRb2u)|5a=|xKu@mnHhX^oOgL{pRPF=JN(IyG6{wAnoNJrJ%3f0=<K^$m
zUytY6f~4y*U4r2WT-R9j`;l=~4`d;PUM2YX2Ar1SwUg<+o>PCy4m?`)6~gQFgqxLG
z;M6sTsi$XRfLSm8r5<qd1Ep~v-RA&Zn1(>rc5ZG3wJGN+H(R(H4-nh<euFQfv={lI
z8L&-R|L7wCuV#_}Y=eKuxjsXO;8f@YT>{Lnm+PRD2o#^91NDIm&<p!w83}SOIqJTF
z_Sa3kBdi8@MmISgbMzKq<f<Jt+h(BfY-$mdvhX}e1MjXH?~ZCVL22kXtqRmAY(dey
zChsC+xuQr6aqKHF<-AFc+NqM9F$6K?zzJ!mW|6W+32l$quNk;N=->+Xe1{e|F$S>*
zJzbjP?0sr2j>!EE)DY0}&*-1hC`;|!E*_T?Pdovej@oVy9-g=f;FY0bP$vw6LpLjn
z%>AK+m#v_wHq8Tie;EPT(6xrYO#hi?_fNtQCkD3G;xFL%u&I&^CZTO%7C*p%$aH?-
z{Bw|n0CC#0c1j}utUc7F4JW@%K1MD981-hK61G(7p)g6n_fOTYa=7qqdjNOT!ArBP
zoB-Ad2?i?bxU78$?>XGnHNnA<ZE<mwc1I2OI#%(69ePdYvC6KGp@@Ad%p(PF7ImpJ
zh(71Vsa7;9MUpbu(hw30vKk4keTE#61&ktOsQ^;TbzTSBBU|AGXz%Zyya%AgO#4zM
zG{FBM$22^l@Xbap#t-MAdRqkb09V54>=2+QY-U$YFDFZksd`Rbnm|@v&m~tJRiVeR
z1-CgKn0Y!W+>;Vry6|$iP^|SmC9nztOqQBwe*WICryPD|Q$F)w3G^bv{b>5={D+}=
zy8#Gqr?+EpMfVFyiGXpYlXp_O!(<Gds;s9tsDK87inF8SqJbihF){~Tye$vmKjUJ-
z&(t}{LHV{kY4`zN`E{KSR0I(WY1WPBOy8h#Ap8mNaQj|0!xGW9uZ3hl31D;w00nn!
z?qu<!&})aM(cv>$NN2KeE1<4osAMcgGyKUwJ5r{igK2zGqSYizNJ>iZjTnW4tq`zE
zcSl^{o)^zOcl|lT<nWni$bqtj4<Gt!20zn>pDoha5mS^E?TKWz;25q2yk>f?^&`Z?
zLWaz^%a8Z407(RvXvj^?k%WtZw&0G^Q|htzblUG8X+w!Q>tgz$KRPxZVoK#8a9q+3
zs+as%mDde11Z<Ztwarg@3{XTH;0^e@D2`y)(B_+OG%yT7fygo=8T23>)Y5pacbyyU
zoqxdD1TDa+MluhI(80iw3I(KTfrn&07rnBJzW`%}q8K;iS<S-rnjpMo_B+ib;H*xg
z_=6pmu1zBh9GRan(d-`n=wZu4o<%X%@-epCyavw2v4WeQT*By7A-rG!p(`0GVO7RH
zSf{hPyn<z1GKNuKr2u;QZS1Dwi$yvkBo2{%%kx&@2Ui(9Q0}LSMVWsB{e~I*qP2$A
zpfwBcuClrV$ja15bba6+w*{TgE{vW)I>&%_*~3lEf1VU@)GXHye5?wvlHrntsU-3~
z*!P6v=ISjK;X!o~gm<SB@o{oEY^KT}qE}~Y9C${Xwv7m@fPc_k=Lq8lm}Duyn8<1T
z0tCj{PFZ?VXTfQ6#;P~T63kxmzoXR``BnAdkZ2sE%VY9NwQTVJ4dAfjR`8Ia1w~g{
z6+A~4jmRp1en%*f8cA(>bAMd|m@WbD0k$uI<G2AJCEF)#6AykhxPeOGcSV+`Nr3}I
zC!*o3Am*kCW3_pyUCE2L_w46B&s-{O8RQ+z-1oEE&1a+eW*g(|UpWAvxWBQWAdNxL
zSM{w})4&o8;oT_)K<wd}M2@!fo1P2JP^O}+!tNPGqgTO(4`a?4YksvoX5Hi0cnH{4
z<38#Kb*Am#H9OEhbrY>zy)^GMBJGht+;Lla>iQ5ACNEpk9*%1=&R6c*7pI~PidVL7
zx~s0yIF3Dcv<uhQ+%E%`*0!R+w0L%&AHW{w*@T3(68}LH3q0$U++XW`vN~%;phaFv
zc3yP@+uH3BEm0JSE-6pD&qD`}@guHw$@zY5G3L}`Jmd5IU<qpR%i-vbq7PmpXscp5
zcTc9_)jUa0$(zvC@wA)YW_s3#^AE978X#jtV{^!<k4R69Z|+48dhXia>gfX-kKQd@
zFPP_kAkJ~ij_H5k^)e-j_r#Te=+tpqoShJNsUe}kfwC)lrS}VU&sr$k^h0%>%rIvM
z776BdpizJmCeOGV$WMboP#f(04k|O#e<-fZbkpZ9sTLXwXOM!e7uR!pUK3ZFI<ZcI
zslh=;>GLAM7g^8w<oZt?Avipia$}p9Od7(9L!O_Gqr-Ge)ytdjPzVAM6rm%ZXPmG%
z7fk<gZ-S;Rv0LnoF{GDzzlo(yJbMII29i}AKTlfVm+z&Ez}fdT+P8BXKCwr<5Yl*d
z$me#{hEK9QbSy@`ZlXsB{UvEU74dV~GxE`=W=!Di1dcKBy$bMrJmIHgyK)b!+T-F{
zG@y#U7FoE6s@*VsALwTRRwb_DeM6GKW>eI~N7+v{z^E2Q7HCWhGQ4u`Sw&24qqGj8
zr}1k6rSeXZ^nMQ9d0+?IWjhR(xc;It!lMxA&cI}6PK>LOeota8!MFtoCzQ5W^}uLG
zA{RJt0<UYusbG5$2ZG}A04E#6Mpy0oD54PCYCu%ht!t4H?tD(5`ek1+Az2()NHEhB
zhQ<CG8*~{enL{)he}MkLJeN_DLc;6X$B1jI3*>xr?ZkZj{Bc-o1T(6y#KG0(BEIS~
z6DI;FyO)hz4AcWs0Y2wG_TO<GNH<pK%pa=;(nSgT*ZAY2*?mdvsM!@g>RUef{R!B4
zYs~3BrZ*THr3%?oBk}6^E06>Tq0u;~j$W>VHf`r-i5+xA#S3W`ftb}_@vJRM!hZ0<
zO3ta~X{vY(40L*MxJYLs*Wuk&OFA+&$i7KcBnAU8=mJ8sQU<Uex4=CMGolWCx%NNg
zN#IocW2bw5WE{kd8;jf!gYZfI1=1(c>{p^AZyjrN)rB)`V{jilW+qK+`s~11c;9At
z@n-(4VWQk$c^(dj6@$bUxporh{4oN*9a>v*mibbTjgAU`7F~%bUlU-*E#T*jber*_
zlyL2|^S^9={}vSS;32y8;e^iea3*J1LSnmE$UYb7RFU-dxqkZN31y?TMiT)n+mI&>
z4N+p=rUwZ%TF>)DUH@fV2e-BNB9=ZnFh7OsQ;EAeDf&IvpgX@mCo12AdC2tG6g7i@
z^vtNG+OjsSU411OP8;FM&aP90x34VvJvlungd=Yen4n;&7?yR#K}jd=k#4-gih9pE
zzXvF2<K@FvDK(cKNPr<(tef;7EL&6D)nNOgaHL3)HY;P-Cs90&vKPT#zl|a6b!LKl
zo!}j|{oHp|8spr8Zh0KZTmytkihb(jJzK7nJv(Y&^4p1O?#ch$WKl~a6Q`b;vbf5B
zBIE+BwJ3-8eXpN9YEOq+efmQAkGnlroUt9?{|#L5B{%d}>K{PdFWoqFsB-TV^b?`4
zr%lSkbMG!P2Sa`(iQhkwu(}`k27_n!ac!g%3l_2QUDkAMd?V!vvoX`Lh^EI0a0-tH
z2CNp%x}bE}nP2gs@oM@yW7}w!`(qr3Zjg<ja*G5#e5i#>Dlzv8L>N!Rh~PrdghRuL
z2(wJI!|f;~*Uq{Hm;dKXqXk~nZzdXbd1IC_NdL%~jj6=)yNDDVV*KETR1=gNvtE}p
z+@Cn!NBY{aVe|87=n5Fv`nA&^BP8@nZ?`0<7a^tEH-OHKt62i$nFv|FwZ3x!dxc}e
zgk^J*jcpzkd3{(hV(QHPuMSO<c`S>cc9dbJ-ajqIPO@p?*?>X62?c3vv83;i4~7FK
z4W$8bTirQv$8eyrTlY($E)&j3F--OV8|<N>l5E23u^e#{)nK)l@=~ELWOCBv4Ho$l
zM3e_e(uxg4jo(PBnJvkAjOhF&gy2{Pt`3$kG$PW4^A14|#ORh<mLZpO@4LObi<9Ei
z*dQTaba@?-J>Y*6f7BU(?yJDphEwg{u|*VDXZq^;>)FJgLS+OpJOeQPg1220-|Zk5
z7I5gw#lJ?-#`h~S@NvL1<`jqbX}u38qc;$IiBl-HG?4ele8)Il!%Z%>oi9@mhypIL
zf78C}qPGsjS$sr;E8_!qdgFB%6ku#yhtQV9C%pccF>fd_e)z``Ez8Q0I~@Auwz-Dq
z1aBM*PS<G(O#u1J6esy;6rF)*Ogoxl9k$7HE?(Jcg)4srDj5n*$HyE`(iwwoX`zf3
zQq><&Pu1|^^zv;CS;Mx65R6-rFfe4$LU1-5_Bj&pGs^Uw=tMh^!b!5o-f96SL0Ncm
z7_wEHFG3;g#D!##vEy6P;P(;Wb4P#Y5*1?G<-UT}CRP5{oN;6zEAj*0ihP^4L%v1u
z!R`>(W1+4KbT@pO!|P(5;l9Uuw>2qV>xu7Gy5vLp@zLMYPS@8#fL-&H-{5DYLNN#u
zrD38{R8-l>J&3n1g9NgrnjySCrvghX)gwVUp0uq75_qH}2fK%bjJL8e(zmRbqM0x(
zS0TtEnJJ1XC8G2e?!0{J!~6>z0)A8X`Oo^pH^OtTD(9jrTu|>*2(70y9lBB`%w{Rm
z(dB1wJ$9f#vS=-Z7(L#?xMb0Mwa$%GXFTRy7mQv+U}gBNbgOBJZ;Ly>ERLJC*0soW
z-DOwtcO28Dxolo|t*6As;9=DJGuBUw%?wZu=nJwJ1N&zcMV|e=mP@T!P){8W2N1><
zqS;2JKKG2Rv&K2pj-F}(&s!+P<w$(viUld<FII%`7HMdFKaqO*+waw=mc0#`(gznT
z%HYh_8h&p)-r_7P&LqC9eBs6RG6#P~bhBArlTZ^lF>Z&KS!_dbn>4ws?q3coIvLy9
z(jge-xo=b%A(k&DIvYsQ-L!a!RG#O5Yy2b}FF<;O)WM_~;tHyUC-)hSp(P#)5W`|b
zKdBL_PWyz+FWfShXq1B#ECXtBtCDuZaP+nRGoN<fh*pDwFxMd`BvXRrg$TS6Z!^(`
zt%DP%<rcfS?94d#a{J$^xe}k-%hGzhMtZ6r&LbHc6#AEth+1UPB6l}lvz9aYu#g(+
zW0&^=3k#_1cIfJQ^L`xk!{|712_|>W@5z-ZdjjnYD)c{iHr>?sfhYCZ{Fgf^3bOj(
zhqb6z2iS7+Va5#wQ9D0v%+fq7w>^ihi&g7C8RmYtsaMP^6eB^O#1iE2`_h$Pye(WP
zV4kXlR+x9R^R+i$wa@Zh)}jZyC|^=Sw<Y4YMM0-Y1M$jjzU{HgJ*JQ@%a#Xa-ST(J
zb(Z&-`PfNr=i)~aF%UgP?gp6*d8x9x;r^kTG|T;_J;QO#?b96neBJ@erMw-9vcA|<
zy=tqt7l~RTJ(llg_H0m`IB5czCzYw1dh*k4PRsTL-->vNC7r(!8bBt!O$sF!l}Na%
z2r{;`eZbetlc;nEW12NT*A@A5<XXU(%#woTWzBRLFjVx-F2G<9@^|c)d|~QCy`E+H
z+IjN*-wP)k$EcXHqJv=q_qI*zLUrC;eV^v>O<str=|#qM{8Z7r(1)tVTPD15CU_t{
zx99N%$AEo?Z=23^+^{G2?x<PDJrQH2zTVKN>3T45L8)u?EURGHnx{D^%{Z@!S9?m&
zneu=?&;0fZ-j!6daBdW*kQkcEOZ3}x{GvBsTEjV_;NxIEUtD6$P>~_hl4Ikmi@>W<
zIR+aCO1BeH;YhU3PYP)3Npzi$5}qyg&E6ZEC-P$4SM2MNf%$vh&Bb2X#JN)Jm1EfO
zQE_zLd^RnIC}Dq2R2Ae4PoTJ{Cu;^we;Z35hBDa)I8{7f%wdSny`w3;t{0+sIoxBJ
z72B4rY>4Je%7>L=K26EwcaAeX<?E4~zv9RKur8IXC>kGIRD*G>ZuyIguest=|3#?}
zZ}|0_-3mS;e>eSaNmu2<Ip3WcafD3;Cg-;_<|X(Rbr+$>C6@R2-u|2@^`$G??qaOR
zWa>C_*)Lq}wLWabxW}8+H>emfiyv6JC;xl&vq4GW{3BpttGiwTjZqACe7zC3yDN;m
zlY8>&n_W5Kdvjy!7uycg;*5>Yqer}%hLp?NqY`ln-R7QNNQ_BOi9VOD%?^9p^rKHi
zcq}b$2}m<MqUe);p?##=CY7p1r)En{5#p{j(E`*~Q@(zjFVDnzt?-C21*lVPqWP)Q
zX})5Vk4ABN#)WY;okw9Hvemp^HU*kUPT`e6<f}ZG3dv4mM#AxD)4vi{gAIHGud<ng
z7cU4pDJZsS`&0#J)Z0B-#v!s6+rgy^r03%m3Rb>#>nHH}MAqJ3`4u6bPJv|Tp!n?o
zGt#}UHA{k#pSK<tmfX{6W<~D~!!oDr5TDs`AA<1aZPZ6u_ipo<naPRq^3jeTR<n6i
zvzdB-uo$Vx7DtG$X=NBvrQ}#Nzzk6!!+fI6`8>%*HcQS|hj?j4h8UOOBTp~w6c}F<
zQxf?EVe!kiuQ$gUEmCDBD@3I*hRR)y?BoGpOONveI+6{`lSeTaNcM9{p2wLGtE)l9
zobkEL4J@P=#4@l*7UPV_P=%2XVUmaI^wj*@Xcu#i3l{hH>$EY@Bo0DVgc@sz;<yz{
z8Q01`JnGi_w_Xp)^igHx$BbofX6UNl1~bJGhol}FJt*^8CL(_x&{~mDmws*Y%3)&Q
z<`F7XG{e#U%cz4N^79BShodVp479{#`3;rMuRrf&?I6jVj7jY{z-lPW<%G=Hb9l>K
zG(vpbAhFM$*M_2C!M?JUk>m!hm&OUlH)E#C>T!%6XK{J4FcNP6JlbHO_<(^%&?m4Q
zNsdSGN}O(J*;9bsjAEx5<}3Q_R_<k+UmP&&hoirBU4S=W205p?Gni)_&2wN^i2B;b
z1hT+nGd8b2W}C8z;Vr2H+UewUexKW|@@Rs!S_gcfjov(+W}&dmqQR7tw+K-o$L}{x
z)LDaa5@v85+vWW<<sL;+XZljb?ec3o)5fF^?t54a<7MjYIF8tkDO$3}4f?lwroDZW
zIjgQYzE{s*6+hv`L1#S*UkqxwdM+zG;*dQ9*8Jv$rSr6SIJkZ2tE<FT?DL-#;+RM<
zMGw!Ye(!-!kUz&*q>rZa!_8i#HNdsm)b(J<V($I>bp2-{PYw6c!K$XIoBX-(W5O?(
z*E8udLF23->{S1`rCWdLMk-vw<Z(ccr=Vgak(L#G0ameC>=#XM;;#je-dnw&nX(>N
zhWd$4-hAR}YC<8Wfl#>~p1=VN><^`lM+w^zUt|nqey|CU)FiQ1_<f}}cuI1VpYR*l
zp-ls?P-h=xH#7a3qh=SMgBDpL&EE2LE<MGacnP|2?Othax>_1H%*%LM%Xx#@sb^_+
zjDfX}h|~h$EroN$a%Xzy);4S3$?fkFJ_$TSGz8Mvw_1{kWkGgwKc&g>8(3WXt{XlX
z+77CYEy)iYOQ!~<+(&3-<oJr%n}w7|)Q={ZYG$n{2N5QMWt%zf|G+Yu{L01fNT7z}
z3N+HuC>CouxLHB0m-AoP+n1~Vc$KfKLj{*|c_J_zwlIqnU)X&Md8ShKM%9@@F>;v;
zp~@PniPSxpw^SqOcthd6cS=O|ZOj5@`o(lENCa(5VwUssYE#XOp&s>FM1oJwt__K@
z_=Slxfo!cwy0M+()7Rg~W*n`gubMU3e<Y>C4$>nZrzF+Pt!V1#G$jh)?1m-<f-XLK
zRY+1yf0{zsa?kJ1n-M^PB<fpAuhrgNd0p>Do6<&4tIb5C!&t+XO*7FK7{bYMa7u6W
zli!xBFxU<&FzJ^@b3T8qwxy1`F;nA~0RykIpmn&)Y#>`<pJh{vuE~f%n=X8O<fhdQ
z%&O^6?6=+P7V&o{pKZY<4jMV4E3EuJtvmLsp)$_LU4*LG;jGnyLe~LDM`E8g4h>DX
za^qU#>ie5Bo4SLJU5%;_;Z-nGSOjD@)lN#0@@V(Ph8et13gOI$i6<iG<*fx`y8)t$
z^NCdkz{rU@LzKi?k6wdr4nHkYM>4;Qq)(m_Mrq?3y{OtO8nvRkQ@L@rV$oQ8<hkCt
z3mc3bJjC<SY$F7tad*G7#Rv5>M^6(m=;s`<PVOxeR9iQzpRncSkCfGjK-PH|K=aOD
zNgj1iU8bqfmJYnUU@n_JTGO6C=~h4Rkm7B}!viTanlW3fGLzlqMjVOwNhvW{IC0$^
zDYx`puDK<-vjx=SycL35sVVMN;5o8a4+;Q&TGSY@6UIrSj$%+3LD3$9^d<S!>J`#l
zH{$4sEWhVOJ*rck==icgd(}-Xoi<Gx-PX!}r#Mq`yr_V;Oj>=E87)iY@%4ko8&~}`
z-oDZ2#VBavslMBfI%v%W-+5lf7IdI4wI|*cx8KgMHK{)|W-j+sGT5>%vcoca-1$_O
zWVGYH3L%)c2SL6pFXsz?T<iM&ILYxuQUsICsd}1ew=W0(&PlWF%6aCN3j1S3aAHVj
zrt)o7yH{%3WHWfKbo#lj@Bd(1lD|X$6bvL@M>Qs%70)j;PirA(>&Hqiq{m9ah7^fT
zhZd;fv#VdViZNkdyXIJU@#G}tNj?7*M+>|>n9m|Q50w+q{LX)lHh_kC6zv#|nNCd>
zt0Qt2l~6S=z32s2ZI?y^l?G<3<06;N1T_=Ouj(vEscHf*3hh)7SHrcEPgJ}Z_13!k
z)@5Ax`g=}&3U{^JD}LITjt~?*dt``}H~aI!(U;>7iy_<kTRw35<VTGj?xPHRl>$!<
zE#)z;=?>IAq7!^mBICE!j=Dhqr5ebxHY;=iXh@vK5`V98pz7`(wP!__kZk!(+6c0r
zG}mEZ3*pY~9Obgn1ctt~H~V=aFK89!@v=Am#<ZvQdvBgcO^?T3CGE~yYDDcaqF_)v
zsdnG?&AFJ@(&Z<o$Jh~uY7I$JS=GG~wf#gE+DW{uZ={#nDYUb^BAe^#kmskI9_Q?M
z6^7onlUk!~y_l!kQl0)7FE&etgXM^5wWn9jraUtruLhdQY6km<m6ebVGIaG`?mpLu
z4dTyo)T)taOsR~{wfWZRdRt!gt&<V2Lp+?Hd-Eih`<wqr)49&`N2#=N)}v9OCtQR6
z0)mBg6lmE>9n~DG-p_Tuvt5mzo7di<N02Ylp@`$KQvRsxx)_SGjZx4VP`ZgcGGb88
z9Li!;V2leA#_^;)2AIuK6~e^8;L1RMY!)@m{T9A+sb-ULMbRG6)TffD>g$V7xgElJ
z=!+PE`c>-~3x>mbSEj<y*5Tb6jUw_ye|nfzD9ey6E<!Q8c~EBwrOau_7MrLD{r-mP
z8rH1~528vcO5E~=!nBZY*9o9YVFH!;qso4>HE{deIOcl6olw4@0-?m5p)Dy5?E5qK
z=&6?|whmyvZo$G0PlQ4Y+Bojqn7`%vzt<b(IB=tRXfbjhlJL~K*T#8NDm&AUB{s^t
zY)BEgl@5YMiN9Te2`?)87hF=6wlwRBMQbR2zR{5T-SKc#S(k5-Zbv%cnd_oK-;C&l
zFY0@O)8TG2qwwt$208{~$qR7G!?OL7qL_TKe2c9a1(BFXI|$#^4>ILGM7}{pmG4_J
zc<w-Tic_&ZTwoh>i??&ErXu!Gl@yy8pL5tbT}90__xoI$a<xFXei_KpRA{3-sZbA=
z?&0Jo5c6&C-bM)!wRddC-<a&jcx{4qg<pGod*pGHiet~K_%NPGut_MqI=#t#Y%BOD
z$PNFda8gTKFZyY56>;4(c`xn`Tjs?nT(p>vqF2E?QI$>2&CpeAf5@<Hy)f+3@Q2ow
z{su8+oHJsIL77q2rZff{L0P;w*BThwF6O>A@I8NaMS_6uD@zDRIQZMfI^lR4PMbyO
z@^yDIMDg=%cIe#mv~N>Y>|+==d>4OU-w_IL^h-lmo>3b6x+XA)!b|;3_d^(AqBqam
zJ>P9Ux#-CQz!g?=+kRSQ(fJOq;~pscQ`^|{S@fDbbi^I})G!BcoNKunkqe-|k?er}
zYFct*vPhgXVR$bK?fX`aS#F;oBp1+%2-SS=)xcP7d~n}t={uMfF0O^Gp#Q!-?pQDh
z<H1aK+=>-x=wToJy_ime?f^blnr5oUS|gLn8`(F4njHPSrQzSYCdwh*5`DG%)ZK$w
zS5=KS1z1)J681`cfzZJK$%i;R#ztwfGAq!_sI}GBG`x$j>*f~h0F{!3_*DI$0^w2p
zVBUGK3-=SeJ*;KSJ~im44(N{fpz>0COFo#{-Fh*+yQbSRiIxUcReqkjpW2jg?F8%%
zg7M$lK8aKNVYGd!o2V;_w~K!jl5*jxtBtQq+|#g*o=1Sh`%1y<*I?a%kbKAECuo+g
zyDM02PJOi`7<A!F6T1zio4>>(ik`!GAR+sNsVgY+c5UMkc68NEJlKOs%^cxJBdU&%
zV)fk9t$N-fH;7`TBA5zzUr5d}D<^O4>D~&00eNyV??dA8oA#sjQ{U(OY8Jv>eV+Vj
zKtHDs9^@{B91rVzE;DBYWGNR)O}!Hj3wsXPl+;K-E}1zue-=QbykCjEN4AG=)eJ1^
zb<It&u4A^k2*y_NB=O#3u4@M+PTRb$43!nb@pPD3w9k+z@ea9t-R{++SSA07b>(au
z`cL>w5$v(b3taCkAfp04uebyKOLlP7rdi5|6C}E-AR$yc?|Mo+B)jllP2UU9{PetN
z<f8?NIhh_H>3i)-<B5$`aJKo%9**saA_)yc$h(T~ai}e`j@=y#0=hu4#`D}8xl2Q3
z3gx@N)1PqsgsH$qtmv~8rKGwp9~lx;Z%*Fn4SmG%UsR;csj6Rk>&ac&pUgSN;k;?n
zbxf;HD?vD^@x0)Mp*-L-l$x_HOaG*W#8;*;@H<Lmw`;$|IxwJ?y$sbHhN1!s%&xu3
z6FKnK&nH^TR<~NpJ1ui`;?t&Cw+{G?QnX{%DvW?*oib??`P!M=%pW=N;W_4=Exqsd
zV<P@{J?Y4k`}rX@&x?b;-$mJ+!3;q|w4-y!c&?4m{Thz9oiv}d#tz&Ug;oCY+5Zwl
z93L9Ur@x+$qP3p7=M1CC3ld3i;VcWXpYz(1-@@O<p4j2OYDe|Fu3@@<>-x=npteiB
z1V_|VplN=D#<H#&hxGEs(tv6A&E-_2f|m}7c4z74!;5BN*FbAQL4Y3UL!GA5(OKh$
zA>F11!3yn<I-QT|BYML?X#AFfyGwSIQs80f<~|)PpgO#NMaVgZ{5KW=sH?9IcB8KF
z4cM#?qJE_+R6D`lerYC9vb<w5FQG#=dRpO|7$MrVCu^9yia}-Dtr(z2{IkVVddJma
zJE|R(xo$z~E+@suear*$J>>{hIQzSk#A>3FPDPzA2#;c+UI`dTWpFM^#awk!6-HcN
zn@5Fb{d}>M$X@t#TgAdBaW+pzov~?|4>s$|s+D~ap3@$X0Vei3Y+eM@InHj8x;+bW
z%sqLM%_pPr*3N2aIqPL}AXt{5Pb&$aCnlcw^|g^wtdb7Spei1HvbmTerGRbf_xiJ6
zmW2JAIFR+E##MgwXi7N^W#%tdA*ZkMHYpLV>DbwIVzDq-EIaPAgUmap-;q|CZz!~X
zf10|l4-22;-P-j?AIQ{&5q(Y!ov*`Hj5gG1uDda*L}$s|v-}v%t34Y(m1P9OlC0x2
z^J7qfERR{Fr}-uE6Cu0OB<*ER1?a&v@>4=F(Ce)dQuR*+b~LA7mRD^aO)a(ck;b^Z
zCSTxr_Ike2M&Q(nC}*ClIP@ao2hn6AdHYp5OzA;cJ9lEJMX83TOoyOl#z=HUsJfZL
z4B5wWZ$(W&d$?bQG2e$A@_`{eQlYNf9$IK2u2_ZE6BjYaBY?tpWU_K3nuWwP>t8+k
zF<T$03ciaQn@zfFO*E*)^heL2CGir{hs1l-7d~}x!Nc@;QORN3N=-Xq%*XF+kOK?G
zq<L}{e-`dx8h4kdHQDPTXdcH~q`}`fYTL}^%rnsBE~O>qRo>n*FR?`P5Lz<GomgDi
zn^z?s>D(dXI1p#2(i3@|OudR>XE+dhA-jgAB%|tVNVH_i<Tap39RvL&n<%X%`}zsd
zawg2>K3|^7ZT;@bohR#6Y7*VV^5B&T7Mhq+Ja=1?hm>$AmX))M8uuh(WkzoN^eX|H
zKCGjtB=zjRfETp-nxA7Bl6VXs!!qz7(HSa(?#gaQ_TBX?%)-8i#!KzV&>PW9M~g>f
zRjxMLCud7SnV|tG7<=e@6`}sgy`TcPXFOZC*Z(*XKtP2*w-=<`sQ-kRP7^~@)&I&C
zd-4)e*(5{?<a=k6&Su(7=e@GTS~Xt}*P*Q@aYhVS`bI!{uZ)dmrVD!4LMO*z);Z?+
z@168KCRmHiM=Pm>EzU-r_<J)GL&NyAEtgO5>@vBS8^O9WzKwFaw;Id&*g}w!hK6ir
zK5HsJ<*L3<Qaw`2MCymxQcKI*q})P{XgiPm-oul3!`%^iiDascMWHBse3z@s-%I)x
zX<F*AJ&A-WFvlK28$4c|IrE~u&G3L2XVb{Khq=c=%&B;8dy;7f{T9L=g8(}0SW(o7
zUi`vD6Il<nOlE?=5LPp)PfjTz`-NsPy&$5uPse4)fvG<s>+8|XcU6=026Ns|Snb`R
zpc>>QjN+htloH6WNJqng$<9VVdB{H<c8gE^`9d9nzewv(z(Ve-k^P>#PArm`trmua
z@v;ZTFeySOOBO;Dq8Lz#iCxI&Dz`3&v{Ue8QxK$tUiXI-m@_!izR*!;iJJ|*_l0S*
zY9!p>!rXG-Dh40p54(n|#U!Z^=MmSes8W)C3L*$nptW$9X@Gxa%>Y{P>#zO@Qrez-
zo19ycc=6dz-rRpwAalXpO*_+(W=EqOE(z8T^O=dDJit!c0dJ5;3X>Y$J~2wF<U(SX
z^*!^GB5jj_C2|?R|IQL`I2A5B{LR)DXg}vp?9?l+Pz3WrD^!lo4R@%^i;7=K=dXV5
znzelJk}8wb&r4|*(ze@<djnH%zcN{|V4_`|-M$SfkQrcU+(aTp*`m7qc$~R0x+Xn!
z#&~K$asE^qfHc4Cd%cW};=kE+5<GEr6Q2RL9#9aLvOx@|)*5><)qdYXWCwqM(Hm;m
z?Ze}(n04~*{U>)qKx6rH%9WdPK5mq`z49(jm8V3r0?yb7YIrVa(bNB47FIYlqN`>~
z^_*~d#zO#0|Dxv=Z<){%)`!E+p$W-*u#e-q#P9b;c|*;|C3(wWOIU|t9oI9Pne>NW
zaHe`CMMy~%U#=~8udotDptBITKEw)P&ec9d;40rCU+2N%x=6bbS$ZH74%_UT`FtV3
z?fA%^H^Lh~8n%2BNVaEdR^>@$EnR1^%=0KJCLUwW!v)M^;VTnfzS(xV5=be(3qtIj
zB$Z<^AbtA=J*rgJYXzx;hJorbAR_j_a?~f+M?B^S*6|s)M5s!G(Jh)CD$&{+uYg$n
zh}FF5e7Hx$3R9MUJ3V^3JBiU<OoRQfNQWm>^s#F>Aia@P77@I9q0DH>PI&JqDsi2Y
zjP$c?U-nHi@1AQ9S}fERp#&ZAZ3ph?#WNQ*aZ;#C(ddp-q66JKsq3(GKf7Q2L(#Cl
z$RCeKUh(hF<h^5vl&_sJ?dYhM&u_&?QCnC|BkRy7pWrEC=FEJ#=4!iGx@k`F1J{c~
z%{$&o?WOwZC(ga^EJebS^QT`@#xkw+%iDcJey}3B$Q$vnCpV<iEs!!S(7hj=us1!D
zsk2cJPhLJb|D81l@K>asbf5jDD9w?08UlPG+edhMtE5Ml^QN;Kmi0nbdjgb?)Ty=k
zuo|a;!;Wg}^|5F~teT(`p|C5h&P4v5TWwn4lBhqwYE)97V%jSyRZU`WI1Dc=@Vt=*
z{oE^B>%ixMI=pF0_e1NKtjVfe4j4W*0X#9KlmQAFnEM*O1A&RZ2KqQn@>*+dab%sR
zzxC;dsB>vPG8jT;=acm{29sKl_HvUpFN37<B|ft{LC($3hn?CmWH5D5nLeXv3?*9V
z>ZAPpO};}0KjV1;4KlL=I|pp&`?nn8O-6J^-om{W20qbJ-W-JdRFWC?ZQmI+azAUQ
z9Us?N2}keG0gLPQso>R!g?!L<JOM7Rd3O_Q-iL4|GV=*AiAZ<=f%HP*BW!a6Cb4sG
zx7kEK=gX_X%uyFxk3<Jsuis;ijMP~~7Q2|DgTt<tBtoRm?)A5o6%1Id3+ChSxW?>F
zX0P*-5AY3Y&qj6r<MvMA)#|w~n`cHiwJy;=m!R@FQ6%bGrD6}eBWZ>$t-QZztcd2N
zQ6@-cuv&{pRNH(%6y)D#4NeTO=`j^=ZV#FH5+Ir9ckDPCJShLca!69%zZ*Cd9;(YR
zv<~0y%Df*5aB&0AGQ%8JraIt7jDy*8;8Wnhway-Y-}g1FC}RC}x$EaprDd^;h<PfO
zv<p76n83YfZptJxf0K7Je~WxRq|RTD?U0%Q5+!4>WRByr&xg$bqrv|p@)rV>$ZC$^
zi^viIhcdmUIccsxSM6`!v}E?fH>1iBt2aOfch34~J;D!-RhWx(EO0_Jh9tMPcwSzH
zycbGyQhd(i?Aq`oL}3i(8>*NeW0&%UjXJB}WOt1>KX4u8o<+^C2zllX#js1<%@RxJ
zky|QE4rA?vx-?X5$^Gj?)JOOmYXu>El^sj6$N>e2JM1#AL*}yO=pZBOixeGTdNTHL
z$$PzdiBmi<^>wyohE7#_CE2XEIig=xj<K&v^GTUxYC9&;k$;FDDrQQ<`sN9UgLYEW
zRR!~(j_dR+o!{d|1@b*kfZ1jDg^%Zl1^CyqG1nNXY1oo&^BkI%+39u^P;kEybT|}U
zdf1!IsF(BGi~^|g0<zpax7NM2ysPR{W<$TPmA8gr6B{NHkIhLXj6VpVcH|PEGa#zS
zc9uWruW%L3Kh+XYX~V+FwvGA0x#e7bziy@<aY>4lWmFE>iyg@=!5|{1@JrIH867$I
zs~jTKl+)HIu~SV*8cYgpC)Yze-b^+r^3?tLhU}ES17wG43T`}&A@lPKL5z^VWYXyG
z+`Aae0E@vxH0sl5h1B!<jW~tlu_Ze8V~UWbYXM9%HWS3i0%j``O4tqo{?Gxx#In~;
zdCb>lmCn!AY8SSHFj(E@*uF?yqoQQ&Y1h57?`<^fg#DOi8qe*-516?e<ffsm(+Z~@
zOZ47~bfwexaS|@Xe)%AIJfO&D%sWn~j?6`jJJ;kjDE5<lO+lAqtQ?s9s@h&`*A;j}
zg;h|CczT~M3p%0x!1dT{%D^^5OlrRW2P_2)nEXzx5~FT-B_c@3n+*;q7V34mWg7<1
zU#-0wD196}pzxae1(~YZh>-O|^Kqr?+*jlg#(|IQE`d9K6~`(7f_|V0Ra*BMFdn`~
zXqQ>wdybzcnqa9SLdZX9cOY3~E~v`ACY|5=Vi#K{4u+~WaucF;h;Se_NN7iT@}M)v
z?w3oHp~Ma<DPz_Sp%*hv{&{wxDJZT#M8&nO?><QA?)-d@G(6;083I!w7}GNV=3ZS9
zO>=J6KS9GHEmo}p7!QHj?_ul1w!>V`@d{{rD1LaF#!xoyH<5eXc_$Rf>r-nBIHYg$
z0bkl0>Mx+$fFpZto8<r2c3!qkvmPCsRjC}_9feD~q@t^LM$v5NQ6v_#V47-bd|>`X
zwjhi8y!ak|W9MMT8DF+pt|;D;J?H{=OG*lD^Po_nx&TDIGv|79D6T1v#wDp>l}TN&
z1q6AV!_w%%{c#1m)egQwwzkqz5MvH<ek(I~xkM+RH&3JbiOx4?@baZ)3&ko?DAr4P
zXHT`s+|KeYWQM#yeBYfRYD4NAt9%)&n6WOMc*eLhT8fx-#vf#E7ZciAE*@s5aD=mt
zy>|87RM?0%(iAZ*Wo*6pSBm1GS#7Ro$L;menrn+d)WfV6sY0bBze+??a!>o76%Tix
zWg|(^UXdmU76Aoy>gz9-sXnq(PD&&bELTz4b!bPfSqDSFl>=bsLz@BWpSM65A`BM#
zp<5n8Tl84-NpqNrxT-wg#j^#o^rkw5WhB|oj?gNN@Y?K;w#8kMx{6dSj9WI-%l(Lj
z<K{`w-4Z}EaifAjCjcoWg+Bj{qqwz1iSS38@RhxG28Keumx2D(b*=A=+5OR>N)cGl
z)RBRb6m(jUM6ghEl#!c*hKJL$=t}P8HwCx&xFZr2EO#=eDPFv(T|zWmy%ag&_OIGv
z#R=>^`y)IZ+=^hGowxKK=MHzQrhmoDN65YqSVdaJ#+oetJsrl_C|eaCFCzP;&plO-
z8}4ULiW>gfdHk++#(~G{35A0@5jQz>pd_D-@@z-_5y9SEx0zP6u85;1JGUFOa85kh
z#cRr;4QPJ9H2@PSDdT%-W7uJ#qO1BT1-}R|v6W9f6rNI9NUQNJX(G>+V3FW4Cl-_E
zo%h>CDzT#kx!r-AKPz<vX%5BcTasVx)yy7;%kBjO+ox54z|^s>PwY*aG%Ei4zQZ(L
z65!lbVUXwObs}Gv|J=;snP>r4Ld+cF(u-=4iL4jp?cK;4P`-c5eG`KPBY2QN2{Fv<
z?5Ff>W-{}06vAA7gS6M&2L6=l&)RxJPPB;f7W=aDn>~`Z0EM25xJUs#asD!O-v_lF
z@+5OCCQWSESB8WT9u7dvX=a-CQ!j01a`(sS3I;>xc@xOFCS{a)h<-P7sGjFMZ1BV3
zK~B=dVtR)qU<*;?A%k}U7!c)ExZ}$w;h&a&W<4`V!HiyqC0UX7d3Y87<Z5Pu`jZeh
zvlU6!<TvG}*u<$fa|oksCnHn2;AvDN^CeOCO$T(gW0p;qm<r?@V8TdP5+=D5c(^A`
zyo*Co*6(QxHMMF`pl%aCtp9-1hrAttO!QWWjEEZ<OkH}YDbJfg$Cto*QFFBT)*3On
z76%(8Ne55k+av9y>LU-sICa|KU|Khp1d=X;xUaWb&h(xH@FYf@(IBo{kGA77ekCc%
z&j4w~_fQaZE9@C}J-?>Ez%JW~f|jM<VH2}yEyjr74^IrH??>A&l+~P*CRvh!8@IWQ
zRS)mQ;PcMn1sTTXHrlUCv8jqQ3iWS@`kIAjn^C94>WjhnqUPCR7g5Xo+2zlvy_KQ3
zqRIj+5JiKH6OJ81qAmBiXP(x1#eO*aXh|j*0{1kXF{p;HnTGNztj=SFe==?O7rem}
zh|v{EwVI|ITZJa@KsC`i65pv18KO|(<CF^;{h}svLZ;$Baw6J^p5zMcjIo|(Vxq(x
zU$iAHSIJjfIuLwr!2^9<9J+0Q6gYnxvaWHX@Y#~Eb;i1!rVFi-oMF-r(@dJS1lOb3
zHy}v+lP4_>8AnNGlEYAHIfBW$3RS$QjsNugIPdmH+NAT_5%Ka+EE}xDWb?SM&9l$k
zXTntG-%jV1kDBMb?i5cz?@RprbKStYP}C-C{Z)w)Rg%y&+xhE@$F4?Lq8^Pn<fufm
zu3i<&HpK`DEi+E7yn4Bosp*lfLsDocROAQGUT92XT5`uNiJh<9luNA!-5ex**%uAL
za*Ys2+9Po7$scx|b!|Lfc1_$LJc`oNiykiW{QN-7=960L(0sNBWAr;(EALAVfk)()
zgp89dU#DG71;1OLC4~nYAidMGRvUjY{~R*Zm=VE;;Fvzk=qYOo6BnzrWsM2`hGrnG
zHM`y9Y01Ema4tA>mT67bFcy3KjI~#(cy~3VAo;Z6t+?_Wu&1Lil^(mbGDX#=Dby=f
z1M*rO<msICICK5HK_llqbb;J8&(}l?s8sQ$Rl&_}65kc&r!HqvFO#VT@8ihs@Ohb0
z-^+%!l@LxNj;)W8M9{P~x%gms*W2`oJZo6f9jroII;aE<F}DPY5d@YD1v!r1bvD-=
zNzC8PC*tenNcPNEpp(EuC=JJAF#7vw4dGVevri&sMiI45@>&^Q4yuU^*`#wZs;VTY
z4pnVV3|W7@Q%>TLSnJ<A4kRbajuci+$C17ZbPeWDz~Zqv;wiK#q#jLsGsBd8csa!s
z`}XSle0t;HiPhqh#jh4>L$oFl87%c<H^<r=MAHoNkfNNqfrp9mK73Ytu=UY39n%l$
zb^Wi%eeZU%JbRM6s_mBRBxbfIt-blU<;Bm4W;jxGI`dhtJn*Q=?cb1CHYc6Y<jz?5
zS77$p@Z?A*^^smyfN_6&{{8Qv?1#P>sEEc$&^7_FJv@6c=+N!`$YVjHbp0-BRS{3R
z%P$dVyk>&<X)ZTT5~1wI&M@=VD^sS~qXV&dvs5%=uXq}0T9x~pbyOw?Lo}4OlUxP+
zI9Ty`-MF}X@6fkIgvi82-#xuVvxB9pktyW(N6+zV(}kns*T+W*?pbL@8H2|;sWzV!
z<V_SVB=yWmS&j^pyn~@z{KwyupX?3q3K;uDEb>Gr4<J>!dx<!b+t07)CJ+bm4l?Q$
zSqd*X_INpId%qmc&ds1LD6QMBtYf>v{+zO)AU6G*GG%vN*UT%>j;KTK+XOkEiD*zI
zv+*vIfqf+6j&w(@aV{Wjzr`8rSQhFS_!bVjNV{BK_YOME+`u;25{A*!jE$lS)p>?L
z56-EgL4(J84T6L{3d5a{mNn;sl#fWJiG@}=$49Q5M|W+`T(i#Txyr2!?%7ZC^?ev7
zU3LR-G>hWEJ)EN`&=gTB5gA^26$kGdePJ*<nQcF;FzJ1F^EQwSiAu4}iI9c5&rNQJ
zSnsKDM5W&>b0@yuP_;~3mt<f8xve22MZqtz?X;U&m=!c<`a7#$2jAUD^SA)e@_B&z
z?BFvMW^oBSgS>m5OX7n*p=4_9oui&Dy*fv3vae<=sIYfcF>uG|l?N><`VE8wNgvB0
zqlhvx`702mJQYrMpd@&XP0X^w60q4MW=a&TSVupQRca#1B$3!>!iV@?D7F_FGD|M>
zMOrfWLo*?h%*GgBiN+W`6^q{GAM$`4B|;m4!zqD$IVQG%n#6P3R|MoPl9tM1Qj#)e
zc|LFb9#HPXWIKD^1BC4ud<O6z+itx^Z8#!AFkOayL_k6!8>0^yr{I^L99%&}n!}Dr
zbW9CG$a2v~rRgtyk<Qram^+wjmY85R;wu~>T$8)|^_D;Um19e@%iJ|{3n{8ps2Vo*
zZpW$IY^Ru7$vrz`1J<B&UBa^i1^1vD5<?1(5bnN<Oj`NtC(KzBEmDV}S|avR%qcdA
zj;<a|A)A``?=e|!((i8Ak3>hd<L7pnXjWup-YJ&0#dz<^?r{&I#i~)7Lk>hO(qhqE
zstj6>a@3;f;+%%z9Z~+#uFv1P6<AjhtiIDd#I`a>Xsc_#>gO||3o@I2YBG+~%RgXx
zi0m38@W@m(ugN}x^Qj&uru3pKKKHymvK**wq@0DC_o{U?r+G-tB>9rJ8ORbnN{*(B
ze1}Fr|Jdx3z5Heqt7Hg`o{HfooAl2euT(To)V6~V+U`trTUQ7<u?b>yw*!-E(uYKg
zOMG|xLM^KUgm#a{2V#}qt(fL=CGZ#de3kBkz1Rl*&%#~eNH2}z6Z_{6tmpAAEi2zt
zd?cPX-LgjvM+zbHm36Q?ATuTKH22eLJ3E_ioNWACC{5<=;9zm}iqLuO$uKJ^1CjR1
z4^X&tfTHhEO&g5yLHT=Uj@wzW$nAeoVn>u2yn7SGb|^!#^O<)dcZ-GwbI|yWIC$&=
zyZ#w+O=59a38ms9on1)W49s(=J^D9aFLyJC^~DO8)uczdl}y^l1`e!jRLt_tjkLB?
zMbU<o&?;*t{N<CG8*1|xKR(#Tc+m%;nqToPDqf|W-m@ni1A#Katazg&(Y83Tm>1$*
z*gAXf=H)b9J^CD`L<p7~F3-!R^G-=4mP15)X^tLx?{dr(MMXT6kLu`Xo1@0nNMAH+
zeamyOOlQQ3eu7EC!;Qwxyg`y13%se_o*NVNw7|tG6aWp<LoHO%j}Rxf*FFAZQN-um
zB}#JSBx{^>Tw)-7YeHL`A}Jz91Ib$wQ`r;I1ml`F4Q7$%$pgOJO>wl9)&tl^-n$Gf
zo!2|u?A76$mJhsa9#C^!i5gA!{B^>vI*})_3S;ZJW$UfA?#(M5OrN>{npc0}@a8R;
z#ZujsB->HP$Wd&Vibf!|oHKS;9**zDr<@#h129=e9%yft?R1I){8KpI8(h!Ssn&!j
zwKQkeMG~YfN~X=7wwh-i6~12}?_DSQ!oGuX$b96q##_61YTA1ng!M+=o3OZfn4Zd(
z1cqWS?>!F9{o1LQ8#S=1C(x`VB4IuGY-`uevW25{EJumBx)yRuWNFag1ljiK7@ajS
z7rz~kExbv!E|&1-nBX?T-%CohiIc=xpZVqJ_QZ6?wgRL6Hz)=S=F`0s91tj(E17cL
zZZaKr>D#S8(`!Uso3SGN%9ARPzL;&xn0cOTXp_+#Vdme<pQ<#>$5EH-1cxU{Lh!pj
zrIk?kduN}#QrE18xut13*?YddzO<=)m%Z!cu}<x^_y3W0)?rn3-J(Yt=@3CWBy`g$
z-Q9|Sbc1wvhitk#B?Y9U8>B;0X^?J^5Tx(i@B6*qIrlv0-gBSl{)La?hPBpQbB;OY
z7{8wj>xs+~Q)2@rd-L#3GliqRLHNE1{*;%XjG!4pm5XhPQ}G{-qm)yNI-8(K%#_^a
zksM>P56$Z((o@Fs9u);2Iyh$^8@usmtmKft{$3B2U#fGmHlGWBVQ1c%tJOl;=GHJq
z0|J7DtfC<IguZi+%^d#z9ye<8584N20k^*j>LLteGLq__*>IkU3U%<bNHyHZc0ehY
zkgy$C4HG&jxc&DJIq<W%Holl$?j@Su^DS+m5p8-%AMMhQLA@{tkilNiA-^@;gF0EG
z5mcltDgER<jr;6=?%i9busHeSX71bXv;K`UpK|`8+8ajsosL-FYYc9YoxAcsr0Y(0
z#Ur(kINY=XZ;Cz<ueB)OSs+Lf+XjUpj3Ld`OyeE~E3IHJ_j5jE7*xAvoHv}`Tl%O^
zyKTb1zPXcGm-8Gd$hY=Uy7tIsoO!#zsQvAE!8)R3q%r?Xy#}piZux;m_g}&p47bO<
zD7${@9)SYnD~G9xN()>|e|P-}#tbOIBBZYw{?HwEom@A3<SBKw6Vr0i%ba?+p#{v^
zF2&pARY%r(?6cjExAl?)X;5<Q=F>F+d|^{M&Y_X7iudfxTZQ1@3kS2KmyAx~VAfPi
zpkYZF-?HiEyIKEKJ5@G!*OEPIg^5j9LUA53OW&W2tT7}9LF<TXZ}Ngw<UuU_g)5iP
z(lR4_e2-fNoY<abhlE@4dv=E({#PHq@H<rV5@@oA_MS1Xm(KR6r!n!JT^7gvoSqZ!
z7$-b7+u|xH-0yMKseb!peMfIe!0XR*#o8U)^xZF9=(S{qO`xrwr&c#3D`7>y7i?&!
zBm}ahEmD+6L8<Zv8oq^zSrm7^&dbM%K^w%&0n{43Agva@#y&38#~fD7gh8=0-n~MM
zz2S{7`Q-ka68<cY%ckyyM^8qwqI9d$+?cKI6OD-?2JMT>kFEPf|9c=~ej9liR3E|Q
zJ35l_p(VWmssBjm9<=k1DNJuOM-3ua!eW{Ii9U`AkK6fQgS<u1ZyY)s^(6B)mmEsg
zMwcUqJxnZ$9`DgzGM}y&wP6Jvmd+BA@+f{}(%EZVS~)uVS!Cg@#Jd_CS?{A-)hi}+
zFYn8h=RHAHcQ4;y>bELc_vIMX04nyM@OIbPorCl_2V_)V{AXq$JE-aMk&#g9zq#oE
zi(bV#tRrL&P>|}{3DD!LPfWN(>4+nEaSc8ro+ap<p=OZb-qLVb7Q}-7$=xr3`PE!w
z&|OeUGd>useqRy9L`o~<;JDDrU~)b)`Yw}&e5{}A?wdoHgXX~mgcbK;SJ{t`1KqnF
zmDP{*V>j=o%!jCB1?%*gkJ0ov5f$Ix>l3f$YB3Cru)DmDdybBeGZy=edX?&=8HlDh
z&XnRDKAS~cYf_r&J-y{?Bfv_J_%YJ@(&c$o_;21rGP>amB&UVUlW(b;J!T)K5052&
zu@Y#$1qIs?>uFk<V<XQl{QF2Wce+p1Q6zd$$x{vN2yv87qN!OkpHV%|!qpV8L}wg*
zVPN*d-`m0T%MFc)p#!=HR-Rm0XEkJKBx0@9eQfFm3#4ztohg45w2HB&^DGBaB21uo
zq_9{1wNg-TE*x<qzYQzavbVWRY+fG%Fm-owi4tqRJv{X_`LyeN>x_@qb1M<26&a_c
zi1*3Gu4*y$ttu$OHNPfXV@<Y@7Z(CR-)&`%;)4BdVam6WC*>(m9Q^JlVJ9WoTGFE<
zH{ymZp&_}D+q{bX5c^nevRxfOzIaz(rVRQ(^-aV*B#7ts$H%(=0_&e`!uSXT9Ai;z
zmh(iCdDb8$t9ZCC3}axeaA<`8n}KyN?cc?N+)PZ$;Z`vR8nXF+{@j)c>4&2wi-b8f
z_(K4}mGqPB-^@eMy+8d#v&c|HS`<+VK7hfqv%60Q{qOTQdL8N0vz_3eXkeGxiH1v%
z)l8Qpdh&oXN*Nfwa%H`lhm2l#qUp4nVN=zGTzBm7F#QV)2d!uD5Fmt?KVwd6^bUEH
z!685z$2%y1aeGOE{?4ylIpgL0h|XA(LB)V0OLFJPl7-ok_>!E!<+@I}EeEt2fWJ-i
zrJ@{HDKFO>ULgFC`Z@Zi2Ugy@aOG|!cqNO^z?LS@clLIox*!a|jp;vyAwR+tLv~u)
zQFxUjyZh0Zl34;nMra<y_IJK)Sp{kP)<FJa8)X*#YRleAucsv~<g2?aPmVpkAZQM9
z6({dYzrHftwE-RYgW-gMcr)_MO+Av5sLLbI4YP3Kw|!vc1IB%kgCm~{v@gH)$8m11
zu#`J?VVAu-x%>HLhm6=dSu^%)Al(TU&sUd-T%sGpjwbrg>tgL<<r(2fKmErE9F72f
zB8ROjYwRH`gg5jO#Zc_~qcVQYkPR(EOyMn_@<3tB<KU*kZ#dBOi2e~BmzytCE0=r{
zd0(28T!m2YkRNMa6qj=~%pTn#R+~PPc^L^SYEzy9<e&9(^M=%S7lOrYBxiObQ|@RH
zH*4nY7^M+1gDmQa8dbn|zr^lTm<H3ar!^7(z9f$W+bf&x&8#N3^Ly_&F#a7s(uU6O
z3b;u`CqAIgy~d^%_yEfBJ(}#8-&$?=VD34To1G}>gDNMNP2v8Y`flJB+P+HM&S+rO
z>OK3Z2Tr1|)I->ROM}CSxtb1*O<kT%^(SLN_41I{e<Dql-o<naB=4&)e%X1Jh+B)`
zp{%w)2_q~-pjoc#0&2Nrv8Q&fHGcy7>waq2zEeYt)`&<RTY&gxrl#@LL}4oZYB5#}
znPYx``K6=ewn#A2KC3y=RdguRbp+N*yGe%*4MSO0diR^k=o#N#q>bL12YA`8+boJV
zr+$>*DYp*Sj>%YAQvf5Lhs(08(K}w`{Ydv0>v~tN*!@X9<kQFYMT`B{ruS(us6@88
zMG95uw+|o)d>C6!>)hx5{_r4wOY4-#u$U+lYSJjlAjcdM)@b6YnD&8vBb{AO(pgcm
zD1P^uSg1-UYKRJhak{*0Le)dg<>iX!{Fv{>S>~C5SB_J*(>J#^z3qd$1H;2Qq5YPk
z<LmO{1|suY@*LC=iTdY}xbOiYlsKq}e|;f3LzS7ZyE$wCBYMrll&FBrl+HA3urSVY
z(hO*OhEn?jFysu)z$KZ}RHpl<(Gk;`W>~6r7Ad3|)u@a&QqcePVUhHR7WkGZg|pTs
z+Jqbm5voO!tW!owgQWqoY%{QL>a-9SA<We2D|F|?rIpe?&6w2gZ!C)DO~?Db8B_g+
zT-QvH92@*f&v0;5J8OBP<NT$pfs5}0RKidmQZ2EbzL=dxxTO#g5@nTXK{c|${)J5=
zIYJSis#hZ3$}`#u*2OxOlVOW75d(~J3liyUHp{GW&yl(GI{<QCVS~y~uRuIOn}y<w
z<;3VXS{Nv+?3n|&W<zSix&l&+IuVPSU}NgzG)Q+UH)%TR8@l*(sEXygkp>t<y4L-m
z5nB5LPr34GZLBR&pL~H5Y;H2XG^+<rpo~QtufzgnS~+~J@!DC-npk6MInQ*-cc<nu
zm5YwgJd&j!r9A3uxHVf=ZkOa^!$rMY2JSSqbnL!%1K0pcA5gp}YeIA(m0)JyfZpTJ
zO{{%3Ml#!ZyNB_Ud0(U`L~AStQjtn%PljMmsZaeVmHRqXr?JlaUBn3;o<DSatA^H!
ziQBLn5&zRyz%d~H%fRF)rdKOtTXQ-^@FDDjE2uQ({LaFfJS?tl9K|KTw2L(PJTU(x
z%9}_QV%jtyA;pS|gy9j-Z<itD(@Y;zll0vf-=ot6I^~|s>azC+|I_roKZGSM48u2z
zUF2encO3&DeC7ml2C!3A8zkPikdbVPk{M@exxwl0F}Q$Ux)o`uKisEGVr63sCX$o2
zsCLx#&(<N);!h+Xzb-_Q$;irpVq!%?kMx(*&StE3bnW*I$pW@96dNoX4tAv$aT3au
z!zkYUl5H3?0Lr=NK$ixqepKv9>vQ83{RZ54zly9}?n-DV3XqnL0OHF^L&qUx%t0dj
zygk}i#Wu1{vEVbotAS{c=Skksu%UlPU2ZRGVrC~jp+5W~wV{or^;h69yt^v4WblRW
z3!hLv%F}}FEFqnOH^(;;l$Wjv+mo6B7a6VzHOarZDyJB;n}>q_cn(U-N$d8P-bsIF
zll)%q$1vZbw@~)NG`^3yc7m~hn|>u`G9_%*-0QGT!?XM%AM#@!bWCjgUJF1WXN{}A
zB>|IZWek<fKG49LMRE;^si#pYt)-Y+QY6Is_07~8mJe6i)`4%z;C}cs8(i()j%|z~
z*P)n^gl)R+d-Qs2cj`=()+mAB838@+(mS%hG0Zig`V=7hofZr~kSUNLDQrkm&{Qe)
z@RpJ>U;RPxiq=z4M!s{H=jv;RVc3S#3Dya<qHUn{Lw_)|5??w^A>&a{Ske@)D`v>*
zi~b}xw(awp-mG!(BirmEH3!B?b{lWjzWqq5w2m;HJ>v|Z5b7QdcFpo2TF)Wadw=uV
z+ktci716=oJH>B9=-+dS2C4|0;m1V!z!A{>U6Cn{TS>OXG*n&vIMy}$ncOr%YA(B|
z1T?ancYqZwjTad{pwnGKgkR<mX8hVz84?~6hYwenmmzdfU6VjXlzItooh`yR+I?h!
z?GkCV7nc7x+W3<|=}(+v@LxsjP^)GKJ{rx_&M*`9G2=*5u5BV9z91Z;q;q~8`*A~W
z3Jsc-UA9k-?t=5fgl!0l8AP?lYGxb|CCN+ZrUeehpTr9kWPpX}s+)8p)AbR2BBUy{
zeDBIkoKRk?CqY|}UQM(yR+7Na2Y3fAUpcolHn(A&Y*w@!z4ydqCq>^yJeAfO7E_+$
zz>%WFErKJKAyn{YOT7K^2YoQSSBj8+!jCmu(Rk7I%0Lv4R0f}xA2kqSR?``N<X<aV
zV^W>b`H>#6mJh))5s3#TG{z9aKwp5Ii~9!?NMvhjO6Q0Q4F^Ilctyh<gE$ceF&qij
zJx>Fa!cLj+io*7RijPeT>AitEp<i>TEs~KMv8+u&RWO7W8=n!%Jd6X6Y8_2&hbZGZ
zjSXKg4WrwPZK_Sk%G`@TXv(DgrfS(O%J8Pj>UG<a!HK7OCHfZDoR?)FKtQ54u^Y#|
zk<lFP{mW*L=d=+BxIEG(ka0GW7iEcX$Ud+Sq<>2I(OD7M`R>AID{I<xIq%!hP&`#x
z0r6aDYZXA2c=4lF{LI8w-+y(u2+EcdV`|Pb6@QS}HX81>@iD(0m6k)Ok-aNoF}7AD
z30QtAM&}am8icuN+9SGFGO1L+IqeWU`y9PgDL7a_&fr_}DZB@wnqhXX!ZOvKD4%TJ
zEO}DYlW`<*m7x#KL-{G*8&j(N9^dbSF3X}BOKb1Mzc2ru)Kh1~wh!*aPDVfqYma;#
z`eGoycu=ynq0HJ_<`Fy*2`bSxspn3f&ESg50HRF?WcbRAye~0wh3bTTD<rcX96wx}
znNF%5e5Je_<jYR*dp*w&1P-E{%*dzm`Y`t2tGGA^QX4uS%ymZ8{qtkX$WAj{)F^@1
z8vX||+B_<`6u}TJ4D~+q$mdk60pgEGIEcOmyb8ETjKahD^O_+#WuF0CCgCTB<td8r
zpMcBZX`jC1$b1=WM^Sx+g&Yd>>A8zyfj);@YMh>~6YYz*vwv1@gGH*&p6EeOJ!p3~
z+EB`bt!42-w?shS(I$Z=oDm*8p#mb!Un@N8GOh?)Pq(O6Zk9w>w#JV{!KX9etpNy+
z5@Vg$(xhJ#7XHZn6%KqxWt}nYiZL$kBr3A7VHL!l5UZ%39Gj#urht4)iYt@TFLB2m
zrqdJ&g6ruv=LzWLCKS+8F;>M9BksWsQylR_dHnMcBbwUtV|EQ>{%=iE9il%y<HBC4
zEP~|xy3!b%#u)w@O#K0eTcgQ5fN1<~%>?$I{LJ8fyyXI)b=tM|;-o4JDim_r73da=
zHb4cj#UKBSk>ra6k7+9B8Ee{}fEpZhMfW>MAOXAYYT(f^Q|-7|p+rY83;HdTD6OK7
zUky1+<`Yf8^Hb&Q*CN=-VNq9J9<OM!ESV)dp31fuKowLvCY>xD_;G6|FGJVDNX1Mx
z7oeJ5Zk9PRR_)Qw?=Sb{NFwebJL2`7rB=9=0qx>slJ$ewv#+HRQBni_s%^K|%hbls
zcsJJ}*$L4mNoDSp6)Wnes>#TEHsm}TPfj^W5ED#CYc4%)=!e!y9iy6|k*}o&czxhZ
zMYKdHAs5C4B*fG!j06ib*^kVl8F5%@-eR~UNjF?E6y<|Yy1^dG8%q5sx1jIlrBT<m
zSEqoyu<ZOS=KT8b*6Sx4vT04u9!}3;09ZQwAz?1UeHqi*r=vtsCr^xD&O`RzYf&U$
zhX+g!+<C9)T?ni^!d@#l&hw$dj6F<y8Y={EY2^L2x<X{3kH|4McD{YhNxk1Z(am@2
ziO0)i!FAGd_97L!9G65C2bH&pNV`TlaQBI5SH%xQV@PVAd$u>PTOCMq{+jNSy*Bv%
zneP7;N&dHAcH|K?OL}Y|$%j?lY|}rwuknIq-hj~-`53XP8I2f|XI)bua@-HqK<jZt
zi)O<+z^A)m35feBVvvdxcJV&dKYHx%Q9Oi$y7rUdWtEKWzeCsm_6Pp`m#oRSsUVbT
zN$eqe!Tjd$7hyaG)gTHro`d;|gC8;N_C4cs2%Dkh5%6|7H{>*a$!0SS)xOxvc_SqU
zAFwJ;IbXK=m9L!u_IV(7r3?$o|J*~}_Mal>|KpGEBoc*dZ~$$(mX5{O0FAi-I-gAU
z%fCdoZ1b%K<3D3|?U$*iApsx7`WUwnM369>j!mOi?~|9w$WH;z2Uwyutxq=wfoVG<
zsyj34o_77h3CKLG8nc6W!?dqJ)xacZ=p_ct9$yy=&v(WwHz;w4NZ}{+FtLi=U;O=C
zM3@J;?1$jP{+TfTr`r}v5&&vIM`1--UVs-$lA|<C2qQ<~BgeejBFX?RE0W7~+P^+K
zPUL473Pml<<szy?gL!rF9`NH(H5zGDS+BSB?@ym*@VUIlNAm$aKMl{5HM`&H`?zQW
zLAMQnagygv_nIRFtV+XfaKAu1(cd>%5_W^@HmH~W-)^wTPq&VntTw<~EPRy)l(N=<
z8;A^L75nQVX`pa~YQTawhpbhVn#=<{h}>fWX9O($$A%(s@bsqQ#D#FH-ar{#2e_*o
zy5Yu@thTVhwn*S&!ui()4uLUV#U6|+`aj*Y|Gbm>@5TeGCv4d6FqX+!{{McI3UL4>
z*4t24>i_z4u2Ep5-x=L4uv>d^mrx=z7)JdI#uo%Qu~%LT3(ID9KKt6zoJ=Pt|F2^K
zKHy#iJdit8thn%h{*QqU^)HZwT&T62Isj6)jBUTQ9%wxP0Dlkd1rl9z7@?4+9sFr#
z`MijS%eP;a172L0d{p1+9%=3{{AGoO4ov=Q4e|taZmsKNQC7ykHvq!69*pK1MlYDc
z5NM}9?amwcm4k{Gz!9vR1N<_<S&jD|N8lkw0ivAc{UL#2l6SDeG(A2HiDZ3!zFQBh
za<j0kN4WVwjLUFOEIseP1JVC_tN!cb>f{vzm$_b2MN{_gao0fwr)qQDcIi4f4Pq7(
zn5>c7P<OW#G&3Iz(8SaV-|Z@2MCjjk`+{EQc|3WDs>Zo7b9NqcjFLYr%fO5d>kJzO
zswQr*%U@Xi0J`M6-Y*0BmmPptSs`S@<hp2j8yzXE3)Wxe$|~F6!-Tjjm6x|zgoh<#
zFx;|Wbm<jpT8T0p>}}-#^E2qMBtN<a{5rMU>_->9rR)v>5bSXyIPPXi4?|93pU;+b
z_yJ7mEbxw*q5~6n$(Vd4aV|>-@PQ$FXSv#g{aQ5u-u^}q;!meQO(PmDd};~XZvXRi
z|J%`Nuz?Fk??8%Z!N32Rr~%bP%S;&TF*7ZK3%fA6XFFliocUkwph@)*Tx0``%UTZ{
z%~uV{@7{l!U!$<X#dPs-?%oGOviT5@Z(RXlYz~H%i-Ec3YGJalDJ$c(WuODM@dNkt
znFnxRx>y5(U=2XR<vD!LNC5QOMLo~m=QluBwK8gzlye8JWzz+MOIWr=WAafbjThYN
zEM8QP*n0a$u_tM{g56?kpxJUt$}w&qSbkaRLG$_CzTRbX-)~rA@(;=IDr^Gz49t?Y
zukcQPx?_R_M=?zLrr2Yl<M6QZaHo%-7w@~V7QteBPH&7`84In}ER8Gvzuqs{T?%~)
zUQ;ZV<&(Pq_*RP`IFsTSp9AAMJd~+o004bP|6EKBW6eOU@XUc>IR9W6N_zo-nI1Z^
zNW-i$Gxo(WB;aoa(qM+0oQK=~A3$qHuo*@QyVy0q-nUPN_3!pHF8(5c`oVy95Bfb{
zy?@vURlP$~Weovps7a+)n*=R@#GC;cR7_=sUGO8ftw!4p5S812nm_Vkt;1BF>oAy`
zbnrPJ2{w`N0m$D*xmD{#JeP|$<_cDb{2m|Wjr2c^5K2~WSUB*fDw^d#eJcuLXnYA_
zPZw5aLlalxrs~BYCG-wE$#ziv!Rt%Mx6;2ZV^izN@Ix<Wjk)^S2_Q0m7Xo9dIcc7!
zbrq8)BhB#s8kqLvzdqh~r^ia}(o?S(S|i1r5u=k5XOw(~@P8k-lsG$fh=IJbE1USA
z(El0aFKGN<w!kJzm>vG>bC?{=0Ed}S%LG>&1RaN9WOa-JjyluI6&R@R0g3z5Dl=X~
zEZ}GNAA;h_IzZI!hv~gK+5wVu4#pzZwk%9~x6Go~_;7!-BsUGn&c=<E-avO~1E!?%
z&((d;j^ykapzl4_D0|Q$HLLgW9UA&R-v^0+A2btid_5J9Sk62sSFi=oRn{kOmM?|0
z9eSVK@g!pa*6KSNJ8%M32{ZA%)phya_vI>~1K0wi5!?Zxu)|^4-p*rKn-rZ@&maKl
zG^QcbST@@EVS`@SfoB9ZB4CN}df3ueKf*pdOP6771QR!FFTd0y8a!nv3Rzi-?TP&g
zYyT$4Of^(}EN`Ou#tW?1WavXq0A2e9yclH%PEl5EiTx}07d$N+@$%0ID@%l@gJ6Cq
zTsyudP<VO(K&y(vHelXyPP5Ztc(t31C{ES^ONkkEN9}VjSWmspp8D!yl(weOZ+jNE
zEZ-j&kJkPG>>k{E=P)q2UH<Y#eZ!oz@U?0NpwG%@%-OMvl2;9VDD3ji9sPK?M*xD(
zvz5{p_)9n_Wn1fj7khRee_revQP=__(jUc>SnrZl{|u{ZackSxy(+)Ha%(>$v)jI^
z1`x#s;4NJ_$FOyat0UNv*8s-K^Qo9~rpC42-+us`eeWn^cnHUu2lQrSfP%w}WxLBD
z-%MW{Sk!{;vm?&kPwG_bReM~C{kEcX2#~VcFWZi}eu1aYM0yNFk8`HfQBg!=5TI6E
zX89@hih@CYh!g1=s}GKGoZbfzfiJ4jW_I*l7ajVY7h<UJ743Us{+SaDMyTvTw*RU%
zInvOfnU|2ri8vY~iVyLd(2cdRbtd-C#dwH1CdhwEPq|;kiepuSNaf4>+NjHk6<zx-
zq>8DHwQu~#*7$vmzptgXKKw%T_%lD#{-cx2XD^H<x@s+xVTnY(e}Y<LC#G02!)&Jt
zb*G`~Ck54C(M=aM>$`dWQ;uol@Gc|5F(mkWBrs=1C&!07!}2MA&GmwmaH}C9HUp-y
zXq3z_v73^ZOu+}(+zCbP-KjSDD!3ei6M*;YV+6CK<0HgRu+R|=8)x%k(Mfp$l*>e@
za{7v2@B$(_=F=cp{2>?(B)K{FUO-3XR){ndp69CW4QgWMz|?-wBRY@(W3+E0tbGTQ
z*II=xw!kkDWn-F?2f<f>lWGlkOZO4p3Tf=UfbB~`-wIo2l~^C?%NX@|{aO9MKgUji
zo$7n!l;ecA4OSN8STcqb*gCEB*ihB~CYE<T1ccz<habT+Y{JR@0yOppTpka1;2*RS
zK=^jpEt4Xx)!a-4B?GuvJ;GQKEZ(1RH-PMHr7Um~j4z@p6G*>V1wi2o=etu2c#K|5
z*r*=?D)uwJ<RH4q7(o7)7q&Ki5D@)Fqga3`whNxs*U$BaRkW!DkB)SzyF*EFO0;i9
z34H<bGl%@8;Exbft&OFve4>7t0YnK)omH{)qPt)NMIT>~4bBWkX?hP5Z;tR9*@^RZ
z?2nRbl^JBza+i!$5z`I`_N;FrRzCgxK5QU}U4g*o*oU+tp6XWg2R~Q@bqOiozs`9$
ziScvylheLU8}shF84+-Z@kD!d3~se^prDV15Gh#akar_RNUFg)J$KY{obA_hV;?q&
zb40~FIrd8f0|$<Zlco{@X}%v!Q}a&?ENLi7eNQ6vNq6en+k`cnN~!Tp?txgR8n~rr
z7c>apfx#$1fv^bN!0ncB2bSWXQCYarx*SN`m!&sl3k6!h__*@9W|av%S7#IvlDN<u
z&m$|=E%qBm>2smN_mEf<_-kl_!MCG~INLZ#hs(|2ieIFZwP!p4VBd@5DYSwTJPtiy
zMp0Q^n?Y2Q&XRna4wBH`2+Ja8JMkL*-wW4cgqMGlGEVK#!ej<BL4{Ur<it+m)<z0<
z)324UkjeUPT#(LbN4};~e7Q=D>ZAD6iLw{&yVxgACNA7h8rs>%vy}z9Y%=%r`uSM%
zB`Z?RLD9;x(ygh}Eey@x9#wZLs`1r3YxI&acSiI$8@n8rvW(Gy((o%LkAm<z5$ZV?
zd|Na}nc4f=TDJ|Sl70g9>c~rk0AJYl@$M~kT=qZjf36fF<ug*B^9d#PPNPYHy*6hM
z)A|((LFtMNafa*|P|;IR4jCk2i4n+&puRs4TLds?pMXqr)iqHu!@OG{W37>Vs>?|c
zlH<3pz2aC%ntV^GDq`_fj5NW$d@IYLkJ#1*dsrbB8;A?elX*+^WC&5HZmSqNK(xg_
z_!1XFxta%Q1j)wCR#d}JjWmQI!pVWCIoH$0w=ue7n$1DwD1&%qqMEHii-8{lZi_lw
zVqM9-e~D_8Q~<N5(l37#F^^l|q`kPLbTu-}&FbxjWfp;O(zQXZ_mYxD)KPR-_7#kE
zP5uDR&N76SoDV*nTF8z2`8kW@<wIbxbMq$B#$!A>TE=os>XiZJh?h8>DyTGHdsPYr
zs!)-xg4jbhWM8oWi9)Dkvylky2UFRwR}dmmjNT#hVpdH&g>VETd>6D=FrsL)A|Hz5
z_B)hhMzUUGkV<J6US#jOtj>f>!yER-DWk*+q_SZgk|-aa$R0=bXG<nSyQ9M|fYyih
zCKJeT1aHHv8M-e7e|*14k<?SCYo@|_JBZ@*%!Slw2AmK6D3RCvXdsHwl>6CG6><1+
z?Og1Er5kL1;i*w-3&xS*m$1@=u>8XU@JncT+b>>=Eh`)4f`z)RIX!?f&$kibikP+`
zbg0BOURm=~{h%d$>?A|-b$MO9GJKEpbp`F1uLHpyNx6=C9NN#^@~S`bY2E-QY|qXx
zQ@>!#PftM4&gS?v75O7Wfk^Z&xeu2MJ|05`bCv&yJ0+4#(Exb{$4=#(bDaLc&!$*t
z@LSS<he-EQaPr#c`y&+1Xgd}}9IS5*nAJ$dUGX&pZXR$uCwYt=xwVV@mHexnmiG~w
zk(|XzX=<h6YQ%gin!fSv&nodBa{ko$;uHDM66A}8_v~%`C-IRlsg=`DQnf{iOcbKY
zD|yK)_xabE$hZ15^W;1kxs9{k*Pu~N=bHV-wF6K1;D(^*AJ?TdUJ-cyiDS0ohA$q?
zjeS0mlh0#@uP2n-A<BY4P9&uXj(2otQvB$?f%io+cgDf&`)y?MMYK28@6?Ykl23do
z$K6mZa&i7%!v1=X`dHzqNle~_*$(C<zDgVLtwLtkYPD?2VYY7GExI280riyvzvPsu
z@Ow7?gu1qR2aD6J!AT{$cyT&-$-c9z+}_^`N1PmSXdF%J!x94nD;0-25-*h3Azwz~
zp_~bW^G5QEYTwM&s7<I5%)s`~cKGfVDULv3w1zJUikfU$tWJkiD@+|CK?ko-r)GMe
zWDORy(Sq1-8+gwX0_%3kg*(NP&)9L**qg*h3!JSyLebMRP1Q2%4F&}jL)qh$*>QLN
z%;8-Z3cSNBLU*)WGc7=huUTN4=_H-|ZaZ;sv-04Ju1DQvLOo&P_;l`+`RN`5cAOx)
z)}VDbh1ShvN_*lUyzY84MAC#h%&f+Dw(-X<;;RXzL!}t!R(iYy<Dha3I)JQQ;wjxS
zZZs9!SW>Vx=V-$E&8MjQ4P{<o9t~HLgNfD@9Gywt*XLZ8Da1C~4(H1)tiX@>u{dXh
z8VM{7aUW}pet=+l`ozP}EnIxm344=1=;J`AaWdz%e3q?%9@}f;#ZR^7IsE$;3HN~=
z_wFnu{ac;;)-H@$o>6gm8q{baIJKTn_N0jkU-ffJ$@N;6?Ph7ILIE3+DP4_2Q4|pq
zFY7m6pc-J;Dm!tIK6w=PK<dZzhR}}7clfbMUt^&zYMR9d1)YCySFaHsX%&1+1_u?{
z_+NFLa@kDA;$pr$txN0wjeLq6TPVEc47R&k>pU~2)q*dhFZ+q#2%0u5ocu8mqP8*=
z(vy1NA~0lKKgs)Z<PQCk(sZNQX~sGD<hn9%@@tH<*2m&Jh{@r~0&ZL|2X3}@d@lz(
z^3Q8ro5i!bh~y7zVwB^>sBg&3-;VyUcz+z~si68~*j*TB=P6<p`~NmN(cp1Vsoo5V
zX(6K_*E}nDM8J0lkA6@iwR4;#C{9|V_NJM&f^4bX0;{@A*Ixl9^!DTnl{1VZyn>0w
z&F5WanHkd-EqHga0%eNf9B=~jT*vBrpFgyhzvQ!&ZXx5)5*S)}&nGR)6rr95(K4-h
zG45P>fHIpW5mFl)78*;TI-eZk%NSEr`&$k#9|O<w3sL2wuUXBwN^9nKb@H0}UCrU7
z0>7YBP0(V%u0=j#a#(eutCit+6`IPOs-nKT{5g;t7sacq=Cv;vK)j}b-Xbrg^pIcB
zCq~r>Y$L81-QTFw$(T26sumuyti}7fWrUvV@wvWSXw~4Nz(Jw1bD?$@Z5AKIc+?&_
z%aba(vPz3J`Bv1sFsk6eG{Bkj!@tHsNV)qpIIwiJ#*1xPvj*iORvsp-pvQ(5E1E0K
z;*kLJKU+9>8vjlp3V71fh1=;<m2=>V@oLt+f89wuU{i(V6x&Qdh<v!FLg13!t#W^X
zhIq^(zN}h-RDr#-$kv(f6fdFrq=IOJkuu@Ka~|5*^Sxg4H8;xf%D*4x%zt^9-Rjk|
zc>hr;Q5^}#Z12z&@e)4nc5yRH2Iq*|4ix)*0h`p}i=;~LUZ9SNQ6s8FSkPr#Gb+~Y
zY#W~b^jcd70z=O^J3RnDfwJ@^v=Oaq6v9mTp<7bSQ5_<0f0Ul9E7TahpruH+_?*f%
z2psfbHfxJ%#X3FxW;(V}xdifQN1jAHI0#Ljl9H(1A}G84$p9w)%HlSYy~h-6Bp26?
ziX5&Kaxv_aBD^dar4|K?RtE0&Q%(n-e{z_Q(tfAc-CI%a)wDpbW1Eb`+Vi^@6<4ug
z{{{BmcOX`hb#tJR$CGMp%<Q>~Lk+&ymY|-((nu(j&9`>o;!+_hUl?^gabEwH#9L}G
z$Bd~{nI5y6h&CBfHoMXNo-*Zg?dwLf?hwsoF6)jV2^QEu*!D8bD^+lQxloYzUdUcM
zir)j*8q=-w5Pl2-!1i=jMNG`iKXl!6?LP1|rCH6y6dkYDV9gv*)6Msh*kv+bdKYBj
z?Jg2wt1#Du&pYYR+!-}Es;kpYrmm5!{0X%IdsDJ6W3xMd)?;{dRxL$8o<Q!9Dhm0V
zr1-%`!P{E>;Hw%f6+RjicB%|pC1??M7C<&nkD}qH9Kof|h_K1`Wuh3?r>4GYZn1^j
zZcK8mz+0QWWd9X26wAN!)&ktq^ONbT?VRH?^*+HyZc~NV$=NTiI(3T#UT|*tNAB-d
zzH(G1$5#09G0KCJFKjS@KBc+vkPGwlj{!Bc`*PYCD~gGL3gL!jXD!vIgJk{T7^<{~
zAjW3=({8Vwr?sMahc~;(j`l!#f8BZv#1U<S)HRk)B}^bUQ-sM}XxiCw<!JhCR5R6G
zCWyP>nI~@h*z9|dG?L39;p(^Tk8LxHJXr=61V7w;Ko-ztk>u&USNAQR_iKa9==MjK
zX;--+k|`mu4eG{|p3QWRuvRwry9DW&+w~}?4ZJJbU$?M*abrK4lc6&LV>$5d)B}Z8
zK@M3=n*r2Tp2SgMSyn(-He2HK<akm~#ckhJJ(E42B8H`DBfdi0=o^YuBiHled4_4#
z4_|rjiiPh!ZHs2p34ET%IdBmt+_gzi<hD#lD6r#jYXw;Y1(NZ}=fXp$X@k$a*XM!c
ze(TVEJy1zz|Cv>JT*t<CB3ZN8c|fCO?-Dnk*nAokpT7R{HWTi5=e(3AugXK)teR1Q
zw2S167*{2dmZ+M7&j~zaEXUIw0c<XumGmXz6%6n?0h9W9>ymVFTpM^xG>6p#%?QZ^
z$9O30ROsHz+b2S&;`6&GI|!j-*un;Dp(l^_keH#2FFlAH$xw5Og%RbeBvYgp7Maf{
zk3J&&Twok%fu<N<D-v_)$sV;4W5ZIGyRPe+GmS;KhP?@}vzMC}eQD>~=MU%nR+Vk_
zg9Cvvmqa(4;@|m(c+Rdlb(HeIexxvIix5g?7c1KI80h4yEPKv$b-`Je4+>YRV$h&L
zC-Y+HOeWORo^7!xpS{8U$tw;@Ae+Sg*(oz^a%#W2;-X@vUO49{j{J3OGT&+3*y~Ib
zc$xqa929Bz-+#>h+M<Ku&8mP}1-g-m%^Od{q4;w7Ai5u;@2wHq*<dF~K)^-Wbe(xh
zIsd$cY&8Ilq#y$HUe+#^`I7ZMlZyb=dag1AgqCso75?&1YOi|oSm(gsMWV6x4AuCH
z;Ug76YLk|1n?k7#0AJSa`KZVc8NmkzDOJJ&LFd;x3RtG9<4N%hH>!VdD8eBGST(e}
zL(6aK6<PMBI})YYbZ?K6XhTo)N`5JwG|=cyYmq~`sW|u#YZC@->ax5~2lZTmA}4x=
zm>?1|uBr#q<jKA3H)fH$^bTHnVS(>Z^%Qcp!j!Es+HXIEXlcG0TaPulE%K;47ox~#
z$*GK77=4ki5SwE=9E8-Vw{Y+J$K^lb6%-$$$FhX0D38xEA^uaxTjIVGJbr?H%M}!g
zW7CRw@+Z+D#{ZgCKEBRKvsYx7U+5H5e(Zr{rTQs5+j8bZ;mv=ZL;T?+a0%7Y<k?+8
z*Nzk~Va1eDiO%mUv$kejegNH8_z~%oPo2?{sxOjiZkX6~57>xS))poKDZQoxL?baf
z`;;AEba;yt1Y)4d1|FaFceCEnow@Zn7epcHPd*9=yjldVz6lwrB+l4=ppNT76NNYr
zuZw6qu7J(?5npWI2XuN+gX6^N`U;#QdM|hU3`F;w?@DFE;S0j#S*-fC_24UJ|LSyr
z<`7DsZ4W;WdghOhzciw0JyOHnd@aSj-bTz}=H5gOj1r+jA?~vy-LvXPeu{b|nLC`b
zib)Uha|1ap{GkV2y}-MYDH<qF*F_;X1xveARu;HZ<>o~M?x0mad6h(ukbty*9yU`9
z)BV}Do=DM~tBl3z5{^)H;aq|&zyV?!HDiim%>cqRwM@Wi<ak&LzDLO~zM)s9FZVY>
zBeCe5Bc3}3*+vM6bBb0XU4=JGH^<5TdWN+|>G`_vH;&OKeOM;QSe8u7<-2+mXvKn{
z19xEQx=h*$v#C^8VOxMdGH$GS_-}OT8CZGgN?th?f;i3+&=HbkG<)Qj%RFrs+Ix8-
zZXmvr2ya8|7`Xi8c3e6ZWuAQAcSD?cOfh-uL(<QTEzzA9XM72XM{Ct`E!rv8Q~DFW
zuRA((9D_RL?>=U3NcJrl?s1wvG7N84m@AIm@=)J+d?ng}8uvu^)$a)3EKga%540n9
z*wBGlW`E&BtY*+$Njgq(N;3HeoRDRz@R5fFZIE2SbXjdelNC4@D}H{k1Wg;39VlTw
z%spsY>V-of8H^({LDhB6Cc>D*OcOhjp|L^nv1XfQt6=P#+qe4>%R1y=5$lRv^L*D_
zAS9BF+u!|i)Xw+IF24FcgNa_`OMm<xD-5Ar4rJqbrAaYu)HvMMT^`~F+Ea={dQA@3
zkAkLTkK*|@m9?SFF}0u*!fda05sn0W(PN9<*1IuUD^)9UiROs7+PiQgJJe087cOj|
z=p%61pxLC44(JS{&h6v;`&9L>)ihERvGxmV0y|ZL7;8cekJrxt6n4gnrMe0_2HTJN
z2)7CdaSA<M^{785V|e+H82{Ad#|SlV;%ZF|77SQ4>9yuUvs(Dc-)r<f)pIvInPTd2
zz(l1h_@o;<NtY0qw{)1i4|4G-{ZM(CMO~~d<?xTm?k0mT5+)g7N8L3Gb{&FK(PnOy
z<!GLCAB{tYwHQSfN5zJRo5w-ykzB90G%44mO_-%`wY^ncaT-g_rfmd$&d%t>!RXG;
zlxsrmZ=!b4++xX8Yocb+sBOixQ>$ox5}LS`MnKv-9CW|*S)OPEIqeu}SP5GCSEO@F
zRYsrYnu1m8Tf`TMu_4|8=Awgo0Z?7!p>VprR7CZ_ybaqUy%|NZwXrxW%%yFv7afwU
zyi^-+5SX&72-n8r9b$2QsiAI4-Spyz(Xju(VEUJq%9(ESsV_keyOR~z^0=$-q4g~z
zg07x)(f`W6uA@BadqFK=5-C|*lA4E&W5}F8HX&w5C(oKzK5c4xHd4x9NyR+{UvVkF
zgM0zph6+E1*=;f%HQKy2f<Gx_l(XZeu*b1`Tff*3X3D8Z$7X^ve`oN`7VL{7eN)@j
z-kPy%tz6zl2<%uID-GQ0^yDf;3@GlF#3S1f+TULJc4wZ|bg(x-uYy@a5qX%rz20@5
zffIk{B`lBb(E=F>8<(de@^$V<W0rX<E~qs}TQvcZxqgQC`HInn#=2WjOQk*?4U+fl
z&M#!^$LXQ1@leEc%SVDa=c*3NheS0q3aVJtil6|u$rejDS_Xq@g6cz!W3r|AXW?NY
z<&C;Nr+Je-Paa%S?`RtNw`t|(%>pVb5_H27Z-#5BDx7UuPAi6~t_5)ipPdxpo4g$S
zG@w?<*M4q&GK-nKGjB4;g3tDKGqj1z68{lxLfxbSp&W;e_csfQk`8L|r@|V;$DHI(
z=jQg&+^2+~BdHTY2Tj}AX#+FMwskj@yM<({x<fy~`M^+#aZO}Wmh+@+8Y+`k8grQI
z9&ATWHyOUInv6!~<?PL<vRv?iJr8p;#%VGOoBGL(w>YAaG77us0(Le?1?bY%M2?wH
z&=dZCdgZ<+bnJShSGe?9yuM4D*1bQR5A{c_OOIlk1i3{-+|!MC-^g|;iF~@9(xE5U
zx>{Ys5^$rb*lNP#+ceq(sbM8Yt{Q}FcWzVt4>#fkpA~40j)OF@Md)JXmy303yIxI6
z1u67MAEOMd5n0VhKu(1elDBkHl$HvYKH3mA)!Dd~)wWQ3Hb+)%1&*jBUs1B>H9Q%3
z_B+wF0Ro$BA{rK~3J~T*xp(bKo(?2&o4ETJV7!ReEgGF=Qq}mJe;yL4U6OAgD@-?O
zL>KXGIT_Edw2ZpVe7-hqaL<75V{37UV*$B8m#25!k?w~FAdc<n538-4=Gd#0`Q7f@
zO(fs*Gr9E*2y(RkBglF0)$dkTd)K<cVjNz2MacqoTfQQ2gs#E=WJy6$OAxBXrElmY
zxoP6aQOXt$5(<7r9yQj-FsX^px)p^h%1=9ff41}B3r**31X%4t<z(`%-LDk)RWCQ@
z4f`UN@K98dEk4*3Eqx`%AJKzH<UF)+*;JzM-31l7v&p=?Z#KImGLcI3&XCD1Ma5e5
zQIM3M)M%4!#^FKLNP4Hp+!j~m$Pdpiu0u)lR7BaN$&9OQtMW(+y|gkxlu~A(b#dqX
z`Mpr+YdRT@$9{W^=t$~f!b6BxyI_upFB@V#;KuCbZBDXTu1v=c3t|xF|D90l6YKM7
zf4bPnU?PP(PuJj)d2Vl0v#e3U)u-a@_gV~gDSf`}%b#`Blb{l$iQiW+x-DAZpzwvw
z<%o3x@dV|O6Z1%fKNpzDe<=+}@NcYn{uK<aKYc`qKSgaxjwB-oNoIc=*PC(sdH&n(
zWf+AR4#|_(2C6ChI3p%W!z>(F0kxDbWp=BJjCHU4B?cR)s6<QRUv@);S4{L2b9A^b
zD`_xyD2Q0jpAm@<;iV;%h4aB_wG5gtPN13LXod#}h_(i1B6UFwbNr&KDKnN*5k#dJ
z(Jd<mir(}wK_a&hXrFR31fE!+FGW+8ZD3Oq*MzkeQVtsu2Y)lllCa%iqRo#ydz1w)
zq3lsSu<^D!#2fVrsSWWOP3DPYnykb3>Q9WlNzqFNXw^jBcxOJWAbpBF;1p|N(l*3q
zl@fF7)s|U&!s|k?7=8#RL8E1MHkx<1hT=4-;hnPm$ti(O(UID`<<hc}%0DiY`Vmtb
z0x@#yxCQ~jC1rz1Y+ePSX=pu#F4+bHH`Oq^HCs5<&C>8<+xr%6$@+;UQEoB2VF7fl
z5tNNbVt9xKdyd3cDcfa3PY{*&rohRkiri!A9LD`hp|Ql?*<^0wKq^FO?i2C77ejQ%
z!#>rQ9UG=4=f6A-t~WE=Rep$bct8g>i2G!7&x5kW$PR7eq0|nn4@&{um__DgkydJ?
zZRDm@6Px<vZgLAR_gEY?>KSQ$zQ~b1k1mKXY?CLX%>W>9l5)|(7bSM@LBk&Y<Ypf(
z8HJ1cDA$a0jY0L2$rL@UxbrnmNCaOFpUKBKqO3QJQ>E)8-h7^~lxuVgDK=Zw=kI-q
zU%wfZX8F_4AJC~n2`WkRM7Y*4{>iFb!#!%^b=Xu>m5q%L4ctf9{}ZU3%!f-|3aR>X
z6G_%wP`LE%Ex{<mP--n^Qa80CL=~|3ET2}z`{U&)P^I`4Yp+p0rf+h0*i@dgt@B<K
z7)Gos;WTk0lOv<+IE2UNrmMrj817a{M44qj>7OJ|zC1-|GqtpEv*DFBa!HgPLK#wN
z3cm<pfiM|d#dT6gFOvST^&s>L4-Y-}ovQ{f+uRL!*`|!f-5YoZD!j{is#;pC4VhxN
z&`AaK39nUQIVeYUkc0YKn8Hc0zWcV`qvO4h%8SX18e{#V(8Xp8A>G(!tzY7<ns6-d
zlGn8$GB7tYQv~*+@l(D6@pe_SrWA8j2H{<LOYKr*!V=Ki5<keLPxsWFm=I0MiO6<N
z$@uZSdsC&u%P*3*vo6Zo-R=P)YexRL3LZ5MA0bsl5`N22ULtnYP#v9r6&%Q*j<0ac
z^o0{6q71;F3q}@)Qq!&(RNH-ZL|<XD{T4U;YQ=<C<z49p4O3qYZ)8%#+S)IX6{Slr
zMe*-Qx?6KNDW`9@)zW#@+6^MU*BBW@_iKh=7G|9)E=HnNt7*^<;h3ktH2PGXrG&zc
zx$$*^u$fH!?I49G-Z_$Hp1i27L>49LtAJOMk2S^P7e>y%IG6-IyX4a<Ex`B`v_m=X
zu~;IvXhF;1#;ysOh^T7%18lR-jDxQv(L8Xh)CMk9@KLC8YH0#QV*|=xrTQNw)v6Wa
zb-g`G^&wy0_1SWZysD;tM>zgm?E2*wLUhK^W|V-c*CNUxy<%ZzU!QoUwQNuQkcYif
zV8?4eQ)on0tLra-s*2xaP0s<9nTpWKLn&ew&FrPtY7xosJ*e()X0G9#NgQ^tv*NQ+
z5q~@3l@J5cu?_M!GJ3^&Yl2}ekN2V$y|7a=`R12s@m=sx%A&21N8ASyG+m83;MegW
z+(oX;4-Cm+{DXQ)d_!*W%%O@eBCQsXCc}G8%h{hI@CnT?Ds9GHtDQSbVq?1uNC+va
z_m!&mr(HRpJ7$}-q+C=iT{c_Z+*m58jwthhLnM+8gn&2+aCAol_4Vfe=X{b#3~`q2
z8x|x{1!j^a(Kq_}I3Tk3sNk1>dFN4f>qnxQG9BaX7jh=Z>ynaeWA<5SPYDsj_$esq
zKL+~`glc~%cI3bML%2)n>-vz78sseMeK-(()rjJxB(Sjxz^YCmsQi!9IN9?CmF=|@
zL|6NmJr})lF<u~ti4{Z%6J#C0D|Srb!ppSyXIRh`AKY(g_hx;MRkU}i2K6H5;P|{o
zIrPlIfg%E<QLQ^fp&QloNu!v~svsfCR)(FB@`8OL%8_oGOuoI?6{r%*8k3*%;OP;y
zm%{Dh@u=K+7yicsIfa#BB%)^Yu5?qB6p^5|0O2WTBd?50uq3H#h!a_Z((d(x-@f02
zmWa%0LiF@gg6Um1KDCu_szG<pP?`ww0w_|*?Qiz0a)e={^vUBcX7YmVi2l`Nw5ni<
zDEPP#j6LIaMOq&m7kA_)yM_qY0@Fa}w{c~B^S+ZRgw084d#-w+!zLNmS@U0SdG?1;
z4!L`FY&MXyn@3KD_3vvE4mCwWQo|&12rp*?bP4TQly6e?Wz4Dkk*o=_8)fspCI8N6
z6uO2RfHuw<uL^dki0V1$_PbHqV-wh)#6#T2ee?g=%t&Dp%-FBcXC->*F%Qj;u;Y2@
zvd1NUZyfAOM>y@0iUhmRnHyTj!&){<A>Vq;niL$X_~57%(tNxGa&DWUD)yC8d0?S@
zpN6RqkwdV8zt}%fivWD1x`{yNpI{;6h<+T&stSaFkQ1q5#h|OkW<lj1Q3%P_*XJoZ
zYb%Aq!zh!3t<8|>q?lTwh=qyU*-r-<y;d#-d3%9JBIpN^Z1YLGK$yMiwA%W+--vw5
zuCF%o)ljx#PnT6i6P@6)qeM>6O+J$|KhLLl;XV)RRMUeT_zVRG;XZFP{V?+#XG)|^
zgfp@&(wzw8$FVrXAL_kaDVwN$ouOIWTu-K0H6aO`ujfOQ9<|!%VG4;Wk{(Eq?`}*4
zopb1ikhX!khCfAAIwEKNTM&r2+SVcWMbT|OiH<_Xbk(%Fb@8~cO7h4__@#T!aju(D
zwl^pmcDoJY9UzT;HDMpD7Z#+xu=Juv(NC6`z(qu|eEbPjis@$8m%rZr6XvrsW#7{T
zvf=m|ST0k5OtW+tWk79b7UhL`d2)Qri3BMk{rO$3Mr0gc9EV4Jg{vDY`^4woF1<dk
zjZbnSTXPEX`oZn+5vAo!)gJy_3<JUC2uZu!lWt|xs%tXMM3ZbP2AP9W4<DIH$tY-V
zl`>w0WpF)l=C(dp%wtZIIdLdwfBTv5+emSK%30UIg~%Pl%3<B%aWR`@lJ|L_Pb)qu
z#uIoB+-hXh|2PB?F^Ht1T#JW>nXBEW4`3C^##yRubkv;<NicIT<|rfxbkMtb(xqU~
zsRh$G0BmnYVhUZ^yCP_gCGKKDB0J_E5*bVLmch^A&?*CJs&u;OqQML(;;g$=0XP1q
z(b|MDiF6Le9tA>og#?$WXsv|6xq^}lJ1cmznjsUlAE|WiQM{glaaZnB8{Z`|Ih(Ai
zKvE}_Akh`nETj&^()Ff6gAEGeyoZ`&EW4ItuIx^aV3{a2$}7_*Hnu}PZd}Z>m^7L9
zbVxqKzt3ZJ^i%1A`{GXFoNz@WPh;7qYg3J=aZ*t&=$?i|TY9F@xJO`Y6G0_*j)Q2f
zXt3f~1z9$<H}F_K<S}EnGs?R}nZ)}smXLIycu`x$(Djx=4_0gDjs#CXL93z8tD7=u
zmh{a~%Ke&Nuf>G;xmx+9lUVpF+4|Ru>cc_qwVK9i>8X%ukE?<36zMS>ZQ3`L;-J=+
zn=xU^oJ2ZKNP!umI8@rlr4-A;_(50dl;?+zlOn9H{64<l3-mKq5tle*H7hThzUnq*
zUlI|D%Y$XBSiVC|uchu~1dF3^Rk{5<!0w3(e;{X|QZJ^^z_GlfSZz5y@j`D#f{zPV
zOfmdj%A8TC)H}M;A$46FKisFjZZ!A<oWF@)Dds8EGm^wzSc6JDxtObJ5On|)ncmYn
zdo$|Lw6KMuUNSrL+5vM|u1a?Iz_L*A1v)iBIza*5OBy|vIvn!>+!L9r_l}<psnNZ(
z-~Nax;vK~I$uEz0ZxQ#CTM+Mf#63_+NU0dXxtlbZpXf;U(IbBe$6;cV1OlphSH_jM
z7g}q$6RJZXjOm0(&$LSA>4-?EoT$+jU{!`Sw2M3a#QiGYz>kz6(IbDCYgIs%1?V77
zL*Z=L#uP)b)o^!w>u`JQhpH_?k&R+>cpn}974I1=_v=KqOV}Fwly_$4w97{16otGt
z&SqH4jWOPrQlLeAs31IPrI0_#M=|*JRY${?a>_JkhLy*RIRo2e%NP_~(^=^Xt-RQe
zIa?XcnMp!+W{M=6pg+eO?@;lrVNk#3Ge)veVq|*l;lrxM>f~WCGS4{hy*fsHVxM(;
z1O-=2SGb$V8g$Z)t-Q~SsNL9<eJA~T$IMnux}5>PFD{aTxFbKoD7sTK`GkC3NGppw
zj!hxM+9xu#UH;BcZ-_V5V@E`kL@DYDazgM`#Ig5`n#D5>CVJ_ek@v#}9eY~NS;mtc
zNm3Up5%UG>P89Q;=QJ)TT7B$IbKJPC3M2t?`R1(^QT@xlgcP|75(C^#LBCHrbhxe;
z5f&%%!r$yPcTC?Cu8ckWprcLTlEXhKVr1HpV?=P)7+DE_L@bsAZFkM>)KN8Iw@awi
za?I_2xl-(t5@FJB#JW<Jzx0`U!9aF}ao>pSOzF-T3v@V?+kB?h*qNCKh)uo~sE3H|
z6n`SwQH`4-7{AQ>ntOamph+R()^S+B&@5Cp^~Ly*oQQ2j2&8%W%s&4$>#@!H*3LL6
z_D8xp;NR7?($0Z1!EGzPTBk^sD{(rlOIg0M8VlI0C*Pc^WE_9KG>!bEKW7?yucB3=
zKze~(>&wIz35EWd3RvJC48ZpL-?N@;v2UW?+rLAe=M-%Q6vv40;vwLW;lty(FrI$`
zo*e{4&FLSu5PD6vCUGr<)kviGhk-7Is0mZGa52<GNHmW^gu~apf9mOJDcpJ#K1O_u
zio-B;8X##;kUWpOEvH5>6x&9rg|poQ#n(N7hX(qKS?5P_4@pl<H9;!U%I`Q){R%MS
zk;2^)FBxW4>c!#3M6u{<TBlE$=t3dgBJ2kQltI5R*IuG4e3VTn-+Z+yLYUx<i?XN^
zEC#=NqfkrbMJP+EFrd2kx+pf4fhSQmmb*)ZUtW>$(X&l%(#=Q71_)@#-NT?$a`*9J
z&8=HAw`M$10feA5QTs7No$747KKx_W$zXP&kXGwrEA?TTUIYgOM||{0)0G6BQg8Ql
zR$1gJb{*Hl1O`j96Ff+s@@M8whY}=WCpMI)a_!mQ>z$BZAr?y<e8wMMYxwiO7<&t-
zDBExSUmBzZ2M~~ykOl<=X^=(`B_t&Tq`SL&XhuS5P^2WKkr+xOq?@4yBm{>1-}Am-
z-}9dT@2s=VS}a|PE}nVr+V|eq_1R*RUBOlE4T87$+ua$mb1#Yr^JCxJlthm`RLC7u
zg<#R!1MyIKbx1;T<Oio07Bfbeh19V3zIm5^paE^vaP<XHd)G8yep@K!po<j7eMHdp
z*#C$Lj^}%Shsf5UsEc#@&NN1Z4HT(s7Emg$H^j)W5MZplFj+RoV$Wt;-g@QL2+R`(
zuX-PzQR{B+yJ?FS8E}(5N<TLb4CU12+ka3w$*Il3!a-#oe?0f_><+n;@gzHxr-9)>
zuIWhsuPBIwne7}j-I_KJI*_uf;h0+u^oHWjV4XQa!YS!``D5x>le0LJJX!-wD?Aj~
zw!hZj<wv{u=bTJWlld0RhT9t=9>Iro=+w;sm80M2MfYd#EhWO-rY9QDiZ_IqZmWQM
z&Rig5QWTV_!bXG~_|vE&9`yPo4%#^lZ?23sJswk}G0RM*`&wD~srV4??_E692+J5t
zNCmCU+7Llwlg}}AV^!T^e$$`ogHl19M5m!H*<q!hVR?vnJhGTG`J81maOCUb;082v
zJpzp0+e5^+9V^HbYtq$Rtcx<8qJ~;U25X}B2#?)gbYPs(t-0HPl7Y+}s&l4^f$el>
zz8=%vw)AFA<G8Px&?-gqCot#OV;UsemTn~Ra!Kj%d1kKW1=4%NSaC%?Ty`2QJTM~I
zwWjffncm%dK1>s!W!3jegJtTmnH+keKCiTGxK-;^rf_-y5x>NBkxxPj|A>RQcsC{#
zIn@S#@^hd3`=%6)to<ze*pFSJ!cSCOo$VnMvSuzO2WyB70iAN=L7Zn?vach4Rn@+|
zYfl5#JEP*4bgR)q7$>dZGM8w=I)@N_yNF85Wxj#jM90e)tiFWSQq`@LU*9*gj@=_e
z9v$;*sqSghS`>PCi4i@i1@HZ1?d2tZVe#8|!~OUZ^gsqFNwQkp|D>S7Z@gioB%b%l
z6u-^gGd`m}DR}nsp;Kujw6T39-F<;$Rc*z~>j_zYw|8d5`;nuV<Pp`ALy3r|eZ1~F
zA0p*gqAhpTH1Vmf(vu;v^$wq&$<x~<JifE9<hee4s})xY<_0&f5>H?9lb40ozDt%a
zX3q|h-^U$im#}3TC3HZX*VA;DHj-<`5uWaUQbThmGnSwBOsigLVQ$Eww`Vu_s&N-i
zz|GC3Y4&mAJCHf(H{?Hd%8!YbQi?9LIH)r4sa5hMq4*u?Lf)^|`M5RiXC3~t>IVe^
z@5+}wbszhDKGknrfLrj5#hQ!zaoId*aDgvS4<Hr#Fn?3h)Oe&?34OX|Sg*nrd@o9N
zAeac$gdgEA=4>T46?61p@H&ef#M^E2-;HHHi`|&+vR&$>dtK*U3rA_y3DZA*UJ}{G
zmiU(pDfqr9a?23z<#2+X-8ds%F7P{yZO3+C;$rOm(nN6FoZweBSzebE%}=wP7>T$W
zF33Qqpm%@h;<wuRvwER!%j5#i@Ackv^EviPNRFg^@;#5V$Pk&orG1n`xR&ix2GMVV
zU+Pgf4yabwf1fCe+Wr<E!I8b5dp9qsbPN{ufg6GCwAn-X-T>A&{eFRLsaVs^_ouXo
zoOrLStc@qc7h?hH`YzRbv%}bcv+vB8#Yt`DDTG&(oXP2@o87aA2^)@W&pLVC_lpOs
zcW-J?$IObA2;Ohl)}5V!^dw)5ch{(7;3@aBzY`?w<xgOkbv3*45|J;f_GG`bdBk7T
z_WMY^{{Hxy?$E~wOs<7Nqxa1nbN4+`HoXsqj=jVwDWVzQq<*>cIy(yA#M5&#y`{b0
zlKFCg%h8>)$}()mJxC#P*RDyE7Shg1#?F&BaLyGWiCEPx?73f9Q|49|WkUxP@-V@7
z=2^c|SLJz_5LH>?@1yi{dU77G5*8N_vYCwU3YCS|RDS>6xs$ylvO|KTEd6LQ{k|x_
z=gEY~bZ`?+jq7Umke&fZ&Ea)h?gTXvc{a}M`hA)2uV-o2`^`3D=myjwsZTJpHu4j_
z7u_zsL2e;DiTg=LgTPsmTcF!>EiVDv4_2O-wC^~H+-QiC==8}o<tcFr)YDY*PldJq
zm5rP0edyijm%R7F_WayjMdRT=2ks5g<-Z-Oe@<LrTVslnWQ^p?N&e9h(}jaePWprx
z9r|8J|NX_O%u-}-!2}8aFo)CSq2U}q->pa1@-#G&)ty4;7=ihv+_!MfnOy(z?d?Yd
zaORj*h0LWsz;%=<<Ml;*4jBx?WQ17h8mELV6rKWE$E{W(Dseg;{jgamvFmzXTGyn=
zC_P2u6woM)383ZQbR?-tDZOugu=d^2F{{fL*98b5d7`>KvpS_3f?ahb)#wdkk96MD
zM+Vy*C;!-2u_G4H#T*1ZXS92$ZrzcED5N+=Pu9<~g<NODI}(Q~%|pvQKJI7KBbvXi
z*oW@{cS>R}zG8`f6<XY*CQd0LJV%mejL77B{tLK26g-w>xnWrs5v4~=Km$iWDJ5u4
zLH``yuf)Dz@sn0Ctarqeh2v!lL(Cbv=d-f7{<|mEPRO79Ohe?fLNbTE&d%9%W#0)M
zEB=}={ZT$F6zx=<2D+7R0wWKSi{3t=79{VKQc!H%YiX?OC5aTp$~I5pV4m>_n8ZT6
zr7~m7X(q#o%=aS<6OJHIgYn-(5mYi~2})cagty<q={2R@;qEl72rJcvAadttZsN{h
zgn~ONM%Sasa}?>|IP$4aZuOpmBfP3?{ROw#XhETAiinb|B87||yR~;qN$n=aSv$7w
zw5;y!Z}e+_<*DpFeftoDxvwG^#N;LJa|F-VN)z$eeJGY;8L9i%lqy%EDV;t(=}*sn
zBkt8MK97ePb`xbjyIn<U|3@M6*OY2|Pg)No^ZFV=jB4C@0u6Tu5o^ZZ_j-bv;%`K0
zB?~0e`Tx8jmos9G3){HeaOGKU2cwM4#By|CR^QvRXEAqe?3%n%^Ude}dx9EsGCBZ4
z|CcVQ2I_$d#5?rpODL|6eoEs*qXF|gVrFp#99JyPvsa4Yq5HroI-?q08Y#>sWT?uF
z9TDu5Gb4NR0}~yTtwOSc4kD>AijvTE*SPTL{^T)Hm%sBY6Z-CTmg_iyvbPS^H=G7|
z|2NS1!UiZhd&^p{&Mvs>cvVFtj&g|2QBFg#ovtI3-<#~Z{*3;{cnR!Ag_%Y!!*Xi$
zui{uH-m0FOPHiEQo_<RO?m?9@E@g@ZhJ*qqb4_hn0eg14SEl|uw`7hHAs|KZ@=_E)
zRJ-M;vhcn+Yp;WzKfj`3;E_%rFyctnRDEYJ`?(#P`T=L^>ZhkW05GI>{PpID=X%zJ
zhq%A=StU8{R3QHl*%q69lNEfmx`$Y*o=t$Adp=YrnYB_q+#Z;>e`>xvOHQI@XA*Qx
zO&D)ui`Nz9Y%om49>KFBYshO>$d*!|J8&7(koWYyH?hUdqoYvXs9&KLcU$$~5rBdf
zx+S?Vdm?{YNNk&_9wgT3-%P;d$U>lfTw*7N_7<l>)#!Jm?Go@kdp;5<x1^x!XlOG~
z_YW4Ok?v>|wYjrf3)-68y9S;Ot2a0v6vX!)GfmW~P5&e}DnC(>uk<T8rr199^vsEg
z&iQMhXoi=83$IP&y;OLrksnf}tm-ppBf}vI-c%DeN;5XkGWwcH$h@=tgA*3F$CeZh
zgZm0h-#~OpAfBR1{?Q)DWOADhs2&E!qC?TF77!nPtn$sV#>e2%`%5WhS!8?8zF}ES
zG7}cdkeCPlhk2tUdtB{Bfy^CuaV}8cIm<*q1zmvh$R2L4yO7$VDc`8<uY4^}o6p0n
z4%z6ZWaX@zb+Q9|1Nwy(L-FUd6Nr@$PI3B)#M743DL(HngB&=7x>^$M>8cG%eDwso
z_BmK_cjkYWxto)ePEmgKh;qECWvCde-{Wocl6m|V4sua!i=UcP_!$w23Gn@N+5nrX
ztCQ&E!))oO93oipM-^{c3&`PE|DMQnqX;OfP->1P#2fa}qt+dIAS~iJt>2#<;GY7f
zaVH6saS{Ra8t2cFVWsL145H~SqXYEKVQv)}<Z-dGN-(=;eGmQIop13E6TZAe+JOCB
zzhR6qz>0Jj+73KMgZVTMxyVn~&F^)}Mm*&DMxgDqEb1E+0Qs#NEzbh_tr+}d32N*Q
zgFID%P}+>-EBYpmC_MRl^|yJ*1hgn>X`ss60$f3FBZ!ZIf&1K^>nALNqZT+!VYnfF
zPj_cj%1BhB3u+|q-Aa#iF?-o0kU*-k``i{>_3xe(q#3?(x^2NYna-I*HFOIYa^5iV
zy@fln^jouZB%V~hIIINj#|(w2D?*0^Z=KP#2V@NW<pus#nP$LS>TxzDJRVs94`kVP
z5--fxuZG()PQoHEuH-#?AC~k^$EaosS0-SS+NsDAk}=@r5yzE48dS>&E~&~Qj}{#x
zd|p>&Bb>QOPk-joU0O2TCCyS{en4ggM|>HF82LDYZgoMCIwn=S_S2ZJoYJgcvI!+r
z$MC7%gvaVH!f=TV%c?ThhN3d3(-$1$T>|z|90takniLLFCzaH&$6hUpUljWtuo5~|
z@PEcxoUMU!?<+<Mk1T}=*;(EM9F@*mmy+>3&atV#C{9OzQJgHq{JK-&H_SP5gZh=G
z!p&2@$toQ_iYm?XghkLJTn(7X$zWe)?^KB$Pk!C@)urj=u}<YC8^cJ@Q#HGnBqw5e
zm$@T(D3pzr^rZ>}1|Ti*O8NK7hu~V`$*+2Sx9+-%(aA1X0t@+UTM&}bi_T0+_9heY
zMku0oVqFj9I%XK6KXCVq<$MVAY5;i447jx0_2o&10yB0PT<~`?6mYamds2OPiOF-y
z_dQ92VntZDW5ecy-BRl1<^{Ln;kt%AW<N${Z@3X0>IgFSWx)^9dl~cknLX6!;g&^h
znJF$P`w%d1e)be&<r=d<1RklkR6m4=az1yBDl774WY55LQ~k|R(+a2DF}#@>OuM6T
z^RJ6fMs>iW%X9^W<4&@t^)c!aY~EWKBIwiHXM9^qEk-5EWeVZ?jNYwn30Ij1$q%2U
zagc_ocHXnUmN-a=_h@||i`F!bTisJlcGNn?<Kk4+vUU)uGWu<JSAa;;Kb&BcB6{pc
z1e3qC7%6G3M$2NL%_`{idwA)Vz2%%MJqwZ|hICdkcpGj|#VtqO;_2N3gXx>POD>u1
z)UYDX1nRzo+7vJS@pvuTiR9IiG;28S3S=5K=02i566b8d{ECckQg&W|R8O^YGj-_s
zr#z=L8^$A~aJ*v-r?cP|R@Hz`xSupaf{e6rxMz*53Xp-knoS6l3@l3Ju6xqMM7W~c
zSQ3lk?J3WbLqG6qXu}vKnqvg|55@*-zWlXe04-;L9xZYM``Sa0gi;9@l!cQwp3MH&
zed}N0U*1hL^2X^o`{{dg>T#?E!b(m%c5de9#eL72Yx&WgepsMzcRA_I%&<i`It_l%
zpSw4DV1o3Gs!YyG;Z3w>#P~~`L7yeLtQT4Oov<0WxH3U<;7ko(!qm*!*KeT9bRU<2
z#2?p&Gtpc?qNrR&)*DVyrM5D`kSsON@Eo{4(x(7&0O9C*#v1*ArOvw`{<*456nW{Y
z(g^5Da-Wq=N-2;HJ0VHVy${<2V%C3Am<;r=<7U1Rax3x*=A5Zy5ftdRqK}X_sBYeR
zB@XX{ncj~M#(sr)ixhu_q=4R8S|Pc=j|H7zFG;Ig3`+GZ0F%Doqow%ULwE@^g<iVw
zyIL7NX$4yFjaPEJI@G2mN4x5G8Po!)%-*R<rJlxM`NDBzND);dkwj2W`knk6-xksL
zm%hsobKTSb#c58Qvngn67gXP7^6+<7tQs`D*oef3o~j1F>b1Gk!Um=y{H}2aP51+;
zmx0gQ^&_e!-v=n*YJ{{2>u6<VGLLXyUAQWALWQCNFZvSS_x1;n-D9v-=b_)#prap=
zY*Dtwk!2le7kGhzAkm$1L8>}R5y{k&NmZ#B(hEt`D5EVi*i!CrGvDvSzD2ky>1$wH
z^CJd{VJ^pJvPpq|&87^{SQtq23WVJOX13GqJu|rq(>Q4_1Z3JVMIKl0HO!pg*6X=I
zpJ!Ueyy|QwS=r@(<rdj#5rMV035-0Ju<WVZ6C}(6yl&>yDA1IG)rEtkCp=18WK{Ni
zq8CRn>@JC!{6nJr^BC*_paC!9R?3fKHS+`-%lU>~;F3nw(ou!NpUg3Rx0BrE!G{jU
z1Js1QkR=m1R{NjO86Eqy8gP6ZLOvLo-vhC60kc3h$Zi(I8v97Y;J3>rWV)4(S|g$I
z5~^&+d4S6Hcsc5tGK1b4E^aHt-2PnNQX5X3@6zqOGrje}^#Li))TyN2c{O}3Kd@R|
z?I?7Nbhl=^hs=>=gaM&?6%i~lpcCWOecn0B{X8Y|OJB0f5&t)vwDNYt^3Tm5^LFTZ
z`Q$oE<9zb#mnm^$n9-$ZXcUiDX8y;u;~OqH?WnIp8U31VmMKO8Q^)2N8n(GRAG{!)
zqa=D_#r;og%$teJ>lfJB#VEndqrl5KFMW%tLijEEc+oW1he8W%Z?kryTdFrn@Nc29
z`p0)o%0h$Fu}Om@&kj<pJ|Hlb3H>p!APq@1byXH0^K&Ta#&hU%pcc65oHnfJ4C`Hz
zNGVQu_+vMHo_<!d$|kJRBXRuxWHEEhHJyv)#u%e#j|waf<wr7Er==*_xbr`I#MI=F
z_aUN$>7}IR1DY@!N+_Rt`4`6-OTwIflC#m_O3U1)*E3<@P9Qs2K(p_L1-FMCtb6_{
ziX){n$5Z6ApW~wQzslrS48d<vY#7#py7$4{FiZP>NrpjJ+e30^Aqz{Qi&B}@)E=4i
zgaZwTvyon!CYiTT`?wbvNF|-3`eGr^Dsxy<athTd9((i=Ywo?(09T>uY9BGKul&9e
z{wlxY1aqkq>8!Ayngtz^c)a7MSIwGkyg2cGBp+666pSZVdbl?>ysMS4<MUD&M^oBx
z7tu1!)ez)(<(1{MO7}zJ3>^_@ud!atl+3!S)T%o^U*!JifC%X2;!6*Ren+3FO*7~*
z)M1{5R9(UCx9EIv5kd&cAj(ZjuAEZw_0f4nmcyYJw(bl?fefok-qMfM?`E!_Uz$j~
zwU<BGWX~pkP{uC4uw#FA=ZwNE+=o|9s=F=zIMhX;3_cNEW&gY+51<$_c%HF+{J3ej
zpUz6Owl^F{_`K*~$a(&w%9ugn97+khaqK>~V?b*w;Yk_ECwI5;3pkzxfGct736KJO
zjrH*1`PWAMKhD5Ef5{hjL+iLg5>xlaM}JRcZ_~%y3C$ohq?EyxRBCtPHY;yRnd3|D
zf-BK~pTMnVB63UGaO-ThTaEyLB-FqAh+=D$rrIDl{;xlgRC*CG5<aU6&5h0ReM{&2
z`sr4z`m^QFe?5^)><z6Kbil<cx|*!{6L7DpQ4r3V>N2J%3SgFbe+w+DQcR;J8Wc9C
zEdceyB_o#qe3(Bm?Vr{2KRl8hi3hkB&W<ehh#&LmfuwAs=fiA5l+MU_l2mxVlWYeC
zE%2cbYON>eCR_!4A}c`2+U$#UVagT81OG0Yq#iI!c0k9<s`YpN^Qe{qdLS7-S0iYb
zP!bQMU0akqka+iLE7o%WKd}L>fsEEhfN@5O0-8sKYKGvT!4vlwX1l+<KfFXJ_L49g
z0&eoZJoA71SYIRCzzMdNp8@PY3t$UQUAl1b`TOOUk+$*QPsP$qcOBQY1+=N-`IDB3
z-XRoKWAxKs2uoi2Uw{uZsmPiI2KFGzpRj{Ia0RqX-{9`Nl+N(iO`XDH%`}b|KPz@r
zQ*e}|c?_%rCO#wBjrFJ(zJiZpFq9v}&Hlq+_`kk+1j*^l!qsaO_Jdj*Fm5KM08ebk
zZany6HsGEbdf+RWMPP-E7XwoC?k6E8I+i)1vq8rDUi(52nd`HORtOIW_6A=^bAPey
z|6TvrxB?z0nCvT+V@yN;*QfX+Cj+DX@7KKoL4B^7-1r!rKi^rQ{tH+&I9d*XRV&mH
zaNke3A^JaVTF63BGw;9s_y6rrG9jo^74Bq&;D61<GwQ$>+6F&VXBR*~c$xKGTxrg#
zDC_l*#C22A2JN0lILglBoBZWUK?(lCKUpR}x0TcbFez-OtF6ND={rD;$`%x-w)SM@
zxBjn>LTQRxDlxY$%~=0D$c;ci{2ng@4H0XN0f2?G7S0$w309^x8H<YL`K(S0-@n*T
zeeXXbk4B%)0ImERkZqiQzp&YZ<2~2|41_yijZ~ue2jCL!nZalJMB=r*?C4mueynBC
z`OlR!eWyskig;PuhK_%v+xJEJn@+F1d&Av}Q4$;!xZ4|u5O#o+xeA!nL=zl-@pUHU
zQ3V#$<ux2J?)-1B_CMb<#X}8V!AMNgKjU&Ox7Z_bRk;IVg5wcoo~b?mYwrvoN3j9u
zVXMVMz@i-moR4eaN9#%NF7kK4XU1N#dK?Oh^G$(6#)rP0r1FzsXe=i5od=DLHv8AT
z275pTNcpb?4H!E`%}a##6??vW-oCKb1eNDJP7f0|cZ%$<0gvPJ!wLP-!l&<c5$%3T
zdO}uI9LNopp*iskbJ!(FjC@7`a_p=ef(ZpH08FZR#&Q{{5g|Diuc2d^fcQU5E52wj
zp?YkFpLG9c=(`2bL2}CIC(7h@y#vG)#|1pTy;^`IgUt-Z_6IzV8GvCh1FLIj=kvSE
zQd*D2z*r^6)%rgyfb*3l1Pbh##0mfyJ5m<zV0>#)05C2y!m|Q#pNco)C^KXacSP`~
z(3xocnr?paa${d2ib;S|Yp`wF@6LmSR3^qR-M;~^r8}SO0hb2-fcRhTxqdMYXr1f-
zxcvXOPe%jx^?^XrqCW%PrX#Zmd_}dyK8qOGFst7`wC(%m0UM0R8|d(UqV%7yKp;c!
zAZr=lW+8q^<^8K2M_$YkJS$)-89rH71KlBp53=@r){9w*;(%PnFmbM#s<VCf*PWfG
z@|To5_b502F>C+lO}-^PMPZek2l&DV!Fm`JTGsyeF7el+0VX8)c=w~&Z&Cf(2adud
zS(lBY_5py&weI}_Kp)ux)SyCIMqGVT6Y8h{t!C;_e{E;G%VqBw&DHp`>KMRJW)1EN
zHNU4LA!T4Gtk}qn{_>*aNR;*(P)R0Jv<+c?Uvf!LdVyZfeq|aY7e{5sy%)~sEjY^R
za~FE%R#7}@Kn#p&6|fp%MxA5Sdxd1POoqO{L90kQBh(HH?#HGyGgt@b;5GEvqh>-K
zipeGDx-~Y^BXKqyKOKTcrnFZO-E0h-|KwCEK<Bx<2;90R$;7D~?Fkq5)x%F}L8MKX
zKAX_N_>Q;l5D@Ylbh`)-!B*~;X05-*4eYYVvGzV+SF_y5bxaBMoO<samg|)}y(p!V
zJ@dal@jnJ_F{TZ{4t_3FG4P^#D}u2_N5Z3Co3rG)^|XuU0+5^mN$tlr(<&19!DccR
zE?)EW4+_o<7-XLZKfKPnGC{jtw*`7P7B{LD#Zi3Cm_K6qTCg)9Fr?q~<LG=?p3b%!
z4<&M(Q$3%VKNigMQ_B<{e*M;}uyB2pc$%Z{9+rNSDH7_$5YW3*w~%e;-IR=#PP+9?
zenmzsn?3_YyA}hUv}4rf1`g;&?rsWz+sY$Cty7+^Ah~zIkT~YR+63Tip49P~>O<`j
z+hBOpeV57JRitRWv}zk0w*T(`{fUf@Kfr43xW=440OKP3%2$|^h;bc?NQg%Psn;@{
z)iChSZkerscUL>?-?SfV&pAkTI;_1fUZo@!dW1p|crT2rm0cl+{Jmy%&h8=3X?Ft$
z<Nc19&cSZBG^vvPc*?PtH8)Q9nNRb0{E>K%Va3n7GvH?~|M=LeBo8U*-TZs}Gf3#S
zZ?mj+U?+f~9JxBKYn?(Fg-r#g%RyMsGsAE8z^K)Vf-1uR43=Txk$ntCnvMycJ?Nmj
zD^gx}kqz38HUq=LssL9`noqvMRtJ9<MbBSrx0?xlAe^Uu5uGkC;Tm)hm|8bmA&vGM
z67Xg5Jpl=4>!U!DPwCE_^@~BZefK{S=-8FFoPD?iGrAgd{uoMM=M`d(LxnBpR%aJ(
zrXzkmEF(f*q{eok6l7}=Am92LzPQ2Fr{MYDmR*}LDyUtl|HZWZXUnGxYzN+bSIMgY
zdaTGY$_k>Rh9Tc?y{yp3^m~%CNT3r&An^2~ovvl*&eTUgqWpq9`ZDBF39+~KQMA0N
zz!>FKkX#&8<RW3E2d1Pxc7Or;_;49z(Ef|$MD0X@KM8`7cqv39z;s<2C>XLWqWBu3
z7r(w{{TzvqWUZ%s0z-nMWrioOc%Ar}D&w2gcTd+*;C)&k^+B;AsbtGgj5aQIqPp)O
zPipH`vpUI<%0>FyA?KTqqzJk^+ze0zQ^9n%X_-XJ!F_g<{{W2syx^?E*IIU3Mb2lr
z#dOJB`S&9#;3p{Plvx`dqog0$?cH+kI7Jf3z$XDLZgku4=Zmspx_X3Mc6;n81HaVd
zO8{JZY^M;MzXuR&g0t~VMVy>VUXP6=J{SXhl)Tqmkcv|q8qOr(3mUM#c!ebH!26x)
zRr3cQmH6SU<mLg6?Cl7J>f0zgFqvS<P}hwoB>KfG=4dolqHWO72MVtqD`$l=cjit5
zp5rP))`1fnx1-*Ml@@cpCrcD@rvn^>rt-9|$vuPiBOX)cX$?Prfm$?)_7BM`#UW60
z(X%gC+y-#XT};)A0Frtj<rG%@e9DE0^lmyYb8qg)`@DF=kMqq4<k=8|`yR*-8}dXs
z<}aYOux2{RZw$1gI6K^}46~?^x;GOjL-V;c+<00}b0>KE=osmA)Yld9I~B|>nCOrS
zN`5PFv8|iYCJ1U>*?FA{b2N(OC2&^iDV2KdHmnoy(>tY{(IRLvK~`Os)qP}-l+Mfm
zHwEZRzL=)mQkSQBu`REpT&{Dhd09MtJfU&7d0YcQStV+F3<x84%cJ4MRdPmwG}6$5
zp`vdLnn$!TdGx>6w}oswK;2*+A(`z9E*%6Tg1(}cbdHRBO&IK?{ET}DnR29p<5-}1
z6vyALx)^QD^r5?ed?nsul80}q)kGw_!Grlrzl{dwNxWj!y$-!d|H!K*OA%~XU|>Ap
zbu=vRjjTc@UcHl8`(VQuEUR!mBwRy>aVWfIVpTR1Zx!?4ci0V)M@=tzCrpEXP|yDg
zn`exw`+tH)Py+zhkO(O{I-);8RM_1>AQpiJ(eG5^UV?m8LrtE~hYIwtF7)JCaDYYg
z(8zQK3P;FMLy8!}Dxf}q6!%c~>?{B6iTbA(Y!Q{1I@OTw<r5&FB@+dOmJ4FWZy?B!
zs&h#lL01f|CK3N#@)SwO%uy=Q43KrMiTIs75IFK~$U}u7<c<BPJ;?(AP5K3xSg!ZH
zS-V&eR1i~o1Ic3(v`a8$zRAxC_q?Q)?W<D9>aRNe(yXN|e#+e`RZ3IAhSq)~mk@!$
zYr=quey5WLm&F_(gDDov(^$h6>;o=c$KW{LgM5c#Ky2x8d{;S)_Ny0MMpe1(NlNIU
zT)=bDG>_Zdcl$R+2M7acmd3kVz-EZ-AddDS;lpE_V;DDCjazr%h`T4t>1%=qRllvg
z1<~SYL$9JoC@M7Qgfh``nn_KELVJP`px?pT!ly&J1&d1jkLh~$%{58)qGiom!p$Xq
zDMD^UTfH}w-t7pY*_XUfHouu#&Bd8N5~h9<o%#eRKE9w+k>?w1DRbU~j*C{x;3QpD
z7~3Jj$XqU)U&2Ax<@5W1`qArG30R7y5gHQgPY<xBVlPHhHRPkYx7eL=SCb!h>Ohzo
zXkPj9=>~-wWA-MBJf&@aqsvD>3MX)q^CsYn<TrCnsy&Sn_~xJ+R)NP(Q8lZM*_YWw
zKXc@spK=x4rHvb7+*Vh~hTA~<@f}jqpx{dL3WQ(bNtT}Pu#D!V%1{~B@(k3jh)+s8
z)tyW4`*h3QV;vMT-If_coX?&>;wpcI-PVrF7YY?wE`BMYpnRM2yufd4_&-aiR4MQt
zzoM_SpWvK}CM$?B#vHFUcr1CSOr`X@qDZyqG%vfI_#s+UpCye8{7eR++tpdnkf{(U
zqqy52%YApO-2f-_OqjR#UXS5n!g4W2Ke_gQq$lOL-&03g`7Db3-}{FDqRsIqF`k7#
zq3TYtX$riEHw>+OKz+dW)ePBU@L{+F#TBguJ2r~OxaTXh;yqs^@frgSf+K1hWkYeI
zZe(%Q{y0@t1d78H!-YBdJz6{vNpO(Pc@Y0e|B2oIe>w5R1+WzVSGF4{vKilZt*Xet
zRHD0J3PD$__{uN^BbWj;Oh4&0W2jk`agt*DL{Jd<C@jJV28k8ed1K#Dj<u-*>Eq|(
zx2UR5?iEoT6<{<wJyDdu$@t5xDm}HA%j^rH=Tq(UeKJ0dq>jc-ul?}SDf5?$-()lU
zWPkO_ZUJ-VevgN?EYNtQE?aZ8%Dp+NR#aSYDxbp|VH?qEoS`k(?y53H7|fR@5{nne
z#jE6hR57bn3F#~>*TiOAGJAH%f|2UMoU+6j8a`<XKPjC6sSAGs=d+QE2pa2luB0DQ
zX{4mQZ+G3#hCk_Bxwn|I^S_+2Wami`<|j4P%PzrC-Xhrv!DXHT{E*93uK{;C$zD{A
zJ9TfuQz~EcX*egJ8|^2B0T^V4zhUUNV9sP*M0H9eTUumWY9utpYU$Gv=rU?nRhAa%
zkY?5XFzKU_zgn~}tkn9o(-h3AvcKH<6X)|@d0n4oLo!nSq|oNniFRo-iJmW$MC)`A
zs@``A=E*9|6BnOU#?Sc8C>q$KWn_<3#%~IcN(v-kULt)kk7yaKy!6!c7)-pb+-%pK
z;xnyY2`1^S5m2#jueT{8UJXS@1gkEvymP|Y{vGf4T(-UN{NjBp>8poiiVw*cKCk*+
z1x4&!(QYLw?7?uuVesf!dCRApJV?97@5(f;WAPt!3NK<@E{hv?_6i=~f`C(lw8bD!
zHpuQn2ID2twfl*pmQ|72&hYljGj92WD*;jn|MU#AebwEKOGevG1NXTf9{Of-@!E*d
z?wSk5awNQ`-SakQv}4!apMcD2cjz&Tv_HcyK%22()HI(?N0_!QKh8aE<osQwuyLrF
zKrBFtrFWM~g@%iVa1?9p)c~~@Y@JZ^){4Mw66ixk9-}pxe@)i^gkHujjBVfko6znF
zsX{c4lU!!+{mg4A1ay^3TSO)x5P=0Nr!>3Wrx*;3@4ncuLW6+!&ftpndU_{YihKYE
zmvnkR8|6=q?RRor$7~05?ZU_kECvg30k(WRv5UF<zEsS-3aY7;|K`MS->VLjXMQim
zK^rHv>qr|pXDvP=#6Z+nW5Lpow)t$7G?xk=0~IHZy80bqK%gb`D)1c#z1;hV`-rzb
z1os@FB8(hihOE(MLcdHB6vvnObqnE9(NeCjU=hJ%C9)l>Gsp_SlYGQ?-~ZF@;r)<Y
zK-Fs>#gY1;>2bDM)VveCG0QE<@_Jt;b(3E=sDBqI1XLgtS?O%n8ETRosdD0F_rVxw
zu0P^`+MFCZKrj_nA8s>BREtl7V;);!E2>OG64Bm5X=e3(#xJ6H2i@$RuI-Tv!EHzN
zT`d9K2G=0(Mi&PIttX#vYw0i~;V%{T6jN6@Y(_Vk{}vEa(PP_Af^oXmj*T7>^vK2}
zcBQDK%FfhEAl;+Zn>#h+FN>qh`+5|{<8v&}z1ibK<gO-FX;nL=bbk-XFXdZo)o}h0
zG>`x&?ktk@R$ND@UsL@OMhDNyfiGIs(<Tg3)5z8g(1GNur-2S+C3fNO3oQla$gBiF
zKWZ;mvF-S>n~zx=m-asPc@UOBa~=Dn>51OkDU|^o$DB~e9b3itqyH!K*XB=(tKIeX
z2WW!xCnIqmz4ygFisxIEy*5QY1+N~RnR^d(NnHkGODsi{9uxpFRh(Q?xV%%iP-U`F
z#)G;FzJ(N1pFEg%sLQGEXgnHGzWl-&L{-M_s2k%guPf#<0jrPir2SZ`_i&y^v|Wy4
z*Ve%~7=O?(7*agv%VHPkU?GzJ3xyIjkX#yX$WX^go%6AOZ{a;CA4Q?v#dujCJ><gW
zPfps-*EPpeMs}D(S3v^nm3{A87X$9d`hH7Q6lwAGZL}&Ia%Hkxj@lpU=dove+Gxvg
zMOpgYSvbs*dRi2z^&H&>5IcpEc=Nz7u`^12zQyQe&W_f~=OKEtb{+YONh+8U?y-6<
zQu;ACb=6hxx9r}`G>CQ2r9mWTZ3Va+99=n0V2#&ZFI<Y$@6lx~?;K|<B+qv*Pd{gt
z;Azx-4eK(9=dPVJ;np^Hec+2Dgm@~B%x}E+bRtUrS5k_A`%-3WIETgbI4TMC$=WLW
zsVhKPFgU)5(R;H+WVa;JU`qxYIb`PuAk<#Ue~Peh$KvQHX}rJwvXR$z`azn6F1^h;
z=Ov`f+<b75MJ;pV(zx+-+F<WqoXD2^3r<bHCslp~ax)9cp3IzHpYq16&M;;}tTh#)
zzV+yi>FM66u4`K5irpU5IAhTzx6%K4uMXbn2BNA5WscHD0p7~`kwtAfaeUO=-yB5^
z37P>u_UV7$tp*4kG~+)m18><#&+#<o@CF5C|H!+vL0{OrQGXIM&l!WIH9oV<!Au4$
zTa}|q$1{V+xpZwc+@`#?$#3eHx(vAX`=`B34H={}i|QzG7LvLwl75)(eKmVH+oLmm
z1F^z3n@i+coET;VQV!XEBMOf!nG%OMRzKyiPi+W&Pxw9ok!jJdE*b180XI_neLjq&
z{5`9N68skj3CpH_%*%N_T{<C7DXh5H+#*c6qHfJZWyzZFFWLed$PZt3UjQ8M`N&SU
zMuv=PhJrGm<>Tk0c$;SPge>{dkAei1VS>-=MK}2yOt?eEqxllxj^Y+AzsL1_%YtK^
zvyit}73PXXe{m)7h`rR6sY6}VTL-baZL5lH4|0vC^Q})7{9A`w9rOk>zi>)4_2{cY
z7B4-w#1DPC;|E-a@!*HJ@r3v3DK?0#+|w82{#kGT#2>@qDEN{QOfKr5oSPIA6hf@-
zR23l!y-lT^co*PY%Q{oc*@7i?2%IFW0D-H4tO)zroxZcw-mY9!;+d<n6?HSF;8H*B
zzn|1oly%UebGbz9KZ)2K!4n~&R*!IFbV_(azU?%+64GPxwh=jp-~Lg%!@xktEvZ#)
z==g0s`MPAM6y)}6&`^?t>E3S)2s<H(q-a7LVt(fK8n~#1G*tp=!ZfJv*Z|nLt4zbt
ziLOpUsCtq)d+<>TF*_ZU8feEMs`Swnyfq+$wk|O&oTkUc8u6RMk%WNWCh&)3pd5EY
zsp~C=sSYbIqg1U}H+E8MvqwMrAX_7co5G2m#mU$ZQ--VW+%mk0)E-HzN=77!a=bFN
zI=mbo^Vo@S6^w80ml;^f)0piO!qh_c-S+wuVU@T|z~X|Ec8-YJlKLw11YT`}<ghbM
zKg9XavF~9ckKjDu`E=)c=qhxwnOTD#nrz-?ZPx$n&68kn9(z4Lb$s&duCWHdKLJLR
zl6ay@zG=Bm;R%56f-SPmL!N%{(kg>$LS|7HaI#}ngGlnzqh)qq-Q~o;Qgx6bu=t9z
zA{+4()n#SW*99AuQ7|&4VYf!KP~~psTwl@ChSNAjRPLiQ-ZB+)A@QMeqRz;rkF3l|
zAQXXSo8h&P-e%@haEoDLXqKsFB93`S&&(fY!o(n3jUuM+8)mV(=A$mZpPP@T`ziON
zA18%cvufQQ9e5nar!@8Y+jDM2?A5zL?t|Iyi7x~Td9%60L=L1fp(Ab7&+b<VfZLqN
zqu6bm2VQ_5ZnqJVu2)IL70OjR*@0)jfnzo`qnJk`AC2}1+I8}bAZkbp85{GN0a`^*
z>_ZU)RjmjO==C9%HWTOdn*;YFA2s^z0qm(1ks!`zJ#bOyH_ORIH_|pQ&R-10i)^v6
zE;Gsx%pf5dmQxjxJ_)Y{oz&$OMbEHs69$$TTp<`p<m|fmdWhB*Yp3d($e8W)O~ETC
zfBHH7C<+r!r%3S?6t?{f&EM&(f7S_zohZTbadaTRt;PKCPd#J<5?JjK5R{e$XMay_
z&dNW&SkV+ZB0!2iDquCid(*^?gP7%geh6<OF@?TEVDDU|9yL8J=oFH*2Yt<OBRZJV
zQqu6z)#=JyA6nnI+N%e*aJD<D$hhZbsTG%vC5!|4sjZDjf-1VFxbx2h#=rvMK!L)~
zL2+vrv&gJq5mXcpupZ$mLT#dRxH)^!9ZC4BW}VM+^+=1{T6KkOEwV^V7^0Pu(QKJ^
z$T1&q^3;CrOK%ZK>|`5h%O%a1hzR1Gu7dVR=g;<@=R!P9Ng<4r>ZCaBse6R7li*g$
zwpTr>CL>ML$%q4a0DP_Q@|QQa6s}KD<%09xzz9F6H|x{vpH)FMu~q6gd>_(;w~Far
zRH}Hym|apv)wZS`6u*=|4;9vVNT<j$Qc1x%+PI0!caJVA<ou#p_^)KTaVZ0IjhiGo
zT~mg|<Pg`48`qG+$fq(3^tlOJ6~VPdw&~!FkYGx6$AVP$$w3|Ah*$_CI<<9$>^I@h
zl#~LWVNeUO3HmSOcbzcANsXKQ#M2=9*q|d35+ipf4YcZb@;pDmQ8+`r7IZ0LB`Oq-
zNP9$scrf2{$F6bThpFSP+Li)FMhc1o&6+R8h^Z%dEZNr=j><6OWzj@4q~K35<ny)k
zD9iBi&hvEl3%IUgl#@V5CUyGiZ&5KcLnWwYP%T{4D8G6=UP&fTFE^Qr+ZQ&E97s+W
zj);DwEo%LXz2&|<CkCUvj-)Ixb}!*A=6s1#5=QNfqRAUBAeAJj%Mo53bp=pRM4wSb
z6h&{6gNlu`BW;^{kzZ{XSrHLFD;jxaHwujv&)RS1ZU9X7Rz?RL|96JO5o8g!9qNgf
zu~l)FPK@wh&vJL0>O^PulI|m0rg!9R3DzH&O>;6g6jJv9Kk0F+PKFBRO5l9t`JbxE
zKNXhm7Pyu<{QrhxX<+8-4?H-6%T8MN=8)&ZAej6+&%afhN+h`k7Fy)n4CKEb%3mDx
zsGUIFB)sp)et=iQp^AZ93(GT38HrYCOi5UcDrHI7ziD^EzE$H6TRYv^Q^ecaOk`|y
zw0)n1x{4-+ehIv5-C#rka-1g~ITJqlLspe-<}8=ocWVykDt~mS?`w?{>nJc}Gsr))
zv3U8@Qjv+mOqVAq6muw?{8zzHvcXTevIu!Zvr1>a%D9BsORh4{EFGXqs&eChNs(Gb
zvWQGJPpslZJ|tb^C*_@<FJB3p<?$3je5l{vko|RVtb}MUTi|L5xIj_lV33Cby1Q^Q
zTBf9#WfiR%e%)tqY$c-<AM(+WetX?ic=h|_UapV*j=D5(F<%*7TGb(IzWZd64=*{z
z4rq1DZ5xW@id7(w`E^6B8q!)NTO~z2{MdpZd;DCuuZKVvgr~#SXv3pLa-DH|qw*tV
zQIFqgxX)m5b$I%`^szaC2TXWJ=Ea=>_HJ@zLMffhFzQ$UOted^-%p}P@U$fQp2ZtY
zEc=YQayfbX4j7R)2<Q2;W9zaFR;SOpP$Ad687&?d0Gw$d{3s8g1|50hZdZzu_5Qe)
zwS&tt4@lfv|2u@Y0wMf6<Oj~zbuk!J#J#lEEV=4>G}v#~_e~o2=}yeVzMp4;^gegq
z!XQrf<i<TS*YD>kRgY2?l&|BPPj1xZeYk2(dqBu2C!4PziTf<y6l(awOFn8ul#Pp?
zQ6V6pT0v7pshE-0YDDU-0@-ttPE2W1_9zoMv-#h;*9+o%nU9>mHJ=R3&L-~pWt&ZB
zi+TIIWq2REkN4?grZx0W#I$P6zI_QeRuTaN`o*aY_gV(H3e0zj8aBOXaV`^Zu2PM;
zT^U|Y=pM!URe@}H>BPyz3nGg9*wz%M7ThA%p5l8gSB_8C=rGK=Z`d|`iwyL+UO`^i
z4xI<;;SA_2q&28RM^#?|Wibq2YsJ2AS@dZXd_jIJ51(CtqQ79Ya+PXVm8q-|s~0q6
zrO7{$pcoJo^Zq$mxvx@3GRa1<XLpg>zQEE|0jfbg3k_F+_G@&c6rdN^!)ooiBUl4g
zcZ}bR=At`Z{ZWDMadKf<bO3I|zrIv-(S5KA=+PPf{^C`^%kK*2U!s>=3qM6++Yfmf
zKaZ-3eu_kw5Ijp8*w15JZcz)YztGWV>2TWCf;Hy#UReJeXf^wGnv>Wj9przTPGa%P
z?e=b@+R)Op<V~D8FKIzgjQ*{(n9d__z<Fq;HMLhuvk4R9VspJkL<E5VspYdZqhX7P
zSF>BX_g~tO($Od98604EOyz2g5;g?eg$8pzumR<v;RT&)$<ChKSy1fGRBraH<4z5h
z6OMor%%j?WJ4<;xhSbNONk4d2OOpJwgDWM3lGmSmhiSt}+p2Q0SEq}SBJfP=CkmPx
z7C_Rj*Qj1VvorFu-_yN3qXEA<wQD&UTl)F*gVJ+l;f+-6JpY6=w+>qOoZHJf2g61k
zQk_Y5NM7gYFxx{kimj|J^|dA+2kjx4o?7oLqvEEtRn)fKyS%-FOtqX3yiwJSig|HY
z8Y$+>{NFG0b}L&m&WfkyM?3Z<@o_$;GSx&@?}V9uN&BQ%A4&X-Qu9>)DxLMVtR$wa
z@f!x{)v-H=GeZRjhtS^r3$YpQqa*jbCpBb52SVijdxzWx<QE0&3>VJP;kMV^CfRE%
za<cYWL5)&GXp9|&Uu66dujoU9U6X@F<Roi>slp5dJ!*Pm19cY$20m$OLIl+g&{lWk
zV08T5=^E98E$(OmcNrT0y_9AddKwpEYAXMpWGqY<J2Zh{-#Lcn>)Y2vu6qo_@1p3w
zqq?rxAss>%woiq>i3N2oNbDToGM%ye%kK+0JPm9;nKgM;3p6>$<S!=vX=0E1RFat3
z>}Ys6I96S3djFYe>Y+CzoESYMeL9y*8nK9mWnioPvJZA^AGZ8l$2q&8=mNd%2T&05
zfjRE(_;;OyI4#_*;NtUDgs>y(-PKVMvOo4RR%l+>nSwikq7#R`-g{g}e^I_0=xo@s
z?4}O%TkO{9v`R^i@>i-|CzQBO3dkb-D%z5kt5)SrCx0=yUHKHbsEPd^(_woA@X-S9
zV#$f|To9Vs0@Vv0fV->{G+24hoyvvh?wC8{^j*++bvLx?g_iEUW74oS4E&o(`lr}>
zr8N61-uv<=?QKDVUBqiTwC{bteEb<hH`fzS_pXvYpddQwsraLFxV-p?mavX?qtXS*
zY1Q`tS6)uWm^%;eZ;#z=D8H{fQuOqh5s1vBHiJkUD?UihZ^(C@sUXY0m_L@fke@MG
z3J&cex3oY>PO916-Lut%YR9k`g&}uHjrqB6A4VES3h#P}oeJ7rwbD)tMjPy+b2?9`
zaY3ZubmL4Fvu)|{^cObQ%r6RPx*oea$Al6<Tu6=A<SIEY7&e@>SBG0loJ+AF+sXWn
zM9Mrmx*^IM7xFaUOha#NL-MnT2OdVCRa$P_{V-8cm?)t*kU9f(WfvXZc1G+?jd)jV
z%s@mi5~u&cu6Q}S=Yit+I+95~ZlGc@ff$kEh?c;YJBDiJ5{VTZ9I;zDvS!!xa)gx&
zvcX1uJ<=3hkm<s0+duuRPx;4#tqr;tx`kB)F4x7TKg(azy*Ok04ehq+X7VV6auOYI
zr|o?n=+dcPZHBg`dd8pB<`z&*^P5^4(nv}eH&JdwY(4yLzUWIBDemJh_zF?3L?rp^
zh;=O21fh036;;echl)A*;0npvu5;Z+bva>pqlhW*o+ns42C*LIbQVjI0mQ~07R!5|
z)vhb_;TPkZv>!CZrwz8_9}|Q)RKz*eM{kj=o~(UEl}6c}-Ny=aa9#>;aObYrPvxex
zRSg_a-|$~a?pk$6ch50!(_7IOePzoeohGU%th4%+v>Ug{&zP!0b7A>yqhy?4A5Ka$
zUPKtO@u?JQf&AB(vE{8(lBr{ft-m&ezrGVjigsN&uTNsa{iXsXSJKaX#T~|1$88Ms
zPQ2HApG&H_PX#Dc(xWj4DPng}7l)NI&^yM`4y$|e3$*tl+S%DL9%=F!x%|1Ity7=o
zHL6`^12wgC`-_$N>J=@*9T4*g)Xkc;p8Y5v@L!BqM(OApEE9;SQ0|t`p%jL~b0d>y
z@ju+YOX0wH^m`t;H<GV9Sck?6|Ew?Wy})?|Y_V_gC301e6L+tbW_>iKE}{?W(>yRo
z9ACXhrXjBPWOw~3aNg>vJ)j_bq6_EQX9vBjd(52!m44;wcd6q5meCxIyB?8I8(-aq
z>8=U;R*c(fSy}FUFOJXLL*ojES2AOv(T)Rh7Hi}5Z>vNob?P_1&Av!zh~{uk2x$mt
zzrgo#=F4fIHg)DpNbskqf7{6QrcVOLwdP}T1-|6KoTq%F3xk#est7*13YvcCj)c!O
zzlkA>e|!+Z)pEMLR@OO45tcjq7|ItSQc$h}I)kB@D9$IVYvY*HYejun6t7^>`U3{&
zGa$7}jdZ{d8ZO-Ud?ZHU0Mm1`@g$g-6?-0oVMohw#=DZj6PF&1Jt*h$Bx_XOm!Wg^
z;58FrH=ho#;j`Cgf-AJ9f%sTV*?NYt)sZX9FiH$cVxLz932M_N&*GAw)aeNiyHUkc
zI<trC#YUKJWj3^RrLCcNbf;&RYaaN4y3t#`5&??wsg2>hL5jPB6mO|@gQBBT2Ph&;
z2i%oo`0DEW>z~j}_p=vensDn2QM`O&@TE(~^ulo|6B(3=?9i`frKp893JOr%lfM+Y
zRq`ra#9gWar0iY!ScbbxG$}f}FIJ_i()){*>pD!6@6i=oQ$)UvRvQfM94_<85t%$i
zFF`K5A5r%$b0cz_SNBkf+rdzM$NPh#pyA?|24itLuC?5}IFS#V`Xa?lvG*^TR`c%n
zCx4SXCO%Mi{d;j2N3Z+@1x=xDo3+K;+WNz@FJHeeC~v?waVGs}3s9<0^n>uPmb0K&
ztN2~R6dMcN)x7TVcMqvgNYZgEIhu$Qxt-~qMR!t8IKGY2>8n`3=p!XQy#7$7vRUDJ
zlF=>76xf0YN>25IQ2c&}+5reO(NW1-6GMk3<BM1=EtM;7=jf3AyU*RS+{NenXZv)f
zR)yD7XH4waud+HC8T^HKA&rfQ+LJ?~?gjmr(XQ6Yi@cQ6jNp@@Z1lt1&Q$&xXFW1v
za(lmjU%`X@eQxcIB5F>{re<RJ)(^dj{2Py1#x#g19(OMkH60#aP1`Qcy=zqlr1`g}
zVg)lUUz&$Lk<9$iHlp3T=yLyM^0scludyLO@ah|TD(r;ws(Z=3)Gf#0+LEHRZXKBs
zys)J&OxfC8a6WQQ=Rb57PPXZCq~X`IV!ufKjIO};JB}!ppa05<_|@yzvm5Nc>KOOl
zUZI&i)o?ZV*WLnl6JwHFPzMJRt*9#I|3rN+@aPjAw_ok&`HOS;DY!(x<ianBkRicm
z`2H%v1Nas!XS<2Ru0rb>Fl8<l#}f!NlKeGj&^eYr>qPU0&;^rjk_YP%3&;v}O1)1+
zmuByTaH`d(JY10<H(A6{#|*R+EAlAWq~yzQl)I^+K8hvbF!5lG+rVVM(c_bciU@Ak
z0eU)lc$c$#uZ_^GgA8sA$y<gX*@)mtf+c)>G|sg&F2k{Q{?&rM+8HPMDsRxp<L%v8
z$q_D!u16zTiyhCWmBs7LTOm^G#;*buz~MIC(l`|N#pPov1GIopy%m@{t#rq7eZWTW
zm;-{Ylg?0IPz{%-aXBs5ORZi&Ng<y|s04~DoEQ0%kz@{ElXMp@mVJ;;PmJ51t%7bq
zL}k7!F0j%cDZ2()VrjgBRixFAt%GBo3jaRm7Jag*ipeWX21ZkId|3X{xJvy!X+Hv$
zb&|+-lS0_*C{4Rh+?~0<)B>w@=()l^tg;nOOqh1R!CZ=nCTZcj%aCds12iZ#Cc}6P
zyuPW>@=+gZl_9z3vYF%x??PM2EqP*EaRm?266mGfx(Q_gjsy3Egf)R>1&2|+(;Krc
z01{N>bI0}Pp^fbNh0IS`(VU;v9NjjU{))nS$B)+@_YLIotYG0gaEg5C%4Wks-$E6!
z%<b-h^EWZQgmcFw4(mWwaEh5$&~{rP7_)JQ4mt4LT=6BwoCI7ZrLr~dY@;^*%RMV+
zYd&%Ybkijuw7PR1f#x~EP`^5^vtjY>L<rK37cmMmoegrJ5fFFsKide#dOd%bj(=*&
z6q^)5>`83ODd!RUEA6}OqvhAv<wU9ZtK1x>J~EO)I&U3WBU?s4SKme>B?(+oGkzZ$
ze33QKNQ?dnq~vnBc9Vy|O(clRPt@3N!*($%F&?>f^D-ekkc4PlBkp<8gpt4R%*oii
z1D)j(ee1)0DKdZR!g!sT)eLv`I&u4pot`~yB=1;gaFfi4(c+TU>7kr;<Iwr@#@6OT
zq5a`It?%*|NDn5edH&pf{&NR9#ZwR#KHXM$L=K({^=M_d^qp{geFaPn9LBn;Hj49z
z5R!iSF?Q^bhZo<Y0?^w@$6udnK;_Ik$+B8zSqw(xo1R@TBs}P4?P#oOd$`OTsH{u#
zZQrL<HwBMEkqJ+8dhXq~BqGRdQ7lYV1RfD*HWQ~3dIvNAmW%)G)U1)6?V6U<U>~Nk
zew-}<o<!OF&Q_lQi^gL<9UZ9Aw4(UCh}F}a3y|xu*RktM`?%{SHyv?W;ivHaX>!ea
zz0QG=5T{ko?ydTSJ3}1Da+rdOX410VINgy~H+dy9WksbepPc9V^NrIVZCSm7Rt9}x
zS^(;u-GWane3rWNr16X*iaElN<XAa&qs{D#+Bc65tXw}2rQ_gCV=>bAeLEfUP;Cp6
zy@?bt*H@k$aXnq7n6eb8&LXG8nbG}3FLof+_?mkW*|`d<7Lqj^6A0Y@EM#iwWoJ}7
zhGYJkAkTI<&I4i_)xflH(mzsQ;3%;g@bn8^KTm}%MPGMkRlQs{QrI;=VPw13KmUA6
z%N#aRJTi2MEcvo9QQ2B{wow@0AwEc&oZ((Mc6Bgu;OyM;boAP?6&cUB5wD1wK93dc
zjOLqO`&B2g8`{~N;gs*aZQ{8UN^ysdouneU9z@Ya@|GN^G1nvvWQKoy;|M*eLGcNa
zsrLddXdROE+_+xSn5oDccgj=gl)sL<kG)1G&Mcexf`tY_gU%!Fw<Pl}5ckhSB$|J8
zIs+x1M)m0}`)CU0u+OiMa6<7Ls&rzhx5M0Nr$Mth==!p}hKzwQT$J9wu`P&L{QN*q
zhKeVP?C%-&=T}N`OlBGyao3W>f8Wu^6mD4i)N6F2-?eKb|2s|$lR%5TlqruxwLESF
z+PAN7*qsG~jImJjHGLX)jDbd0J><o2P;dVKID6~3sJ6BZSP2;!g&_y&5S1EI=`Ixk
z1qlU|5J6%D>24TFgOmmd1u2Okq(dnMksKOEK)Ty+ZBIPsJ@0wW_r2dA$KQw?nAv-;
zweI`6ugF%u9f&;)q|zu_n{FGy+E3pN3|MVVJfUk<tP%F3_t3SojR7P!?kM^O&(+)z
z`}n>%GJ+GYhN^tM8PIyW(6d=VLO-HFcvF@6l!BPPsyL$8#EB8vfmtc)9Yq;$9RV#w
z83?v0Z+Rz3TUUG>83$LwXiKb~B<hdf!;r4APE~>xV7P$>Ve8INCb-hMg+}>>m$gro
zHS=pXj{8uE9gtuLKlwNTCg#J!^0o&AI4Sc)?70pj|F|m98azosR|8#}8mG*pr+>vS
zOb8C;A~{uXg3zRXvUUy7VSZpOF9*HPZTs!}`fkXX`E<$U*getDf(UbqfC)S|$Lv6A
zp%w*Qem^di37`rf<?#Yjt@O2uj<<OQ!{>+9`ZJ;=m&FcxqP!pQ*Q_e&Pj?$cUwDr>
zM*O6(0B#9S4>?E`sDoQ70m1i@;V_H)Tsj+Lv*{BVR{TMKMXfm)eK=tG?zzpzC<?TW
z>*-S{VM+n;7HGK&<&By}ieK>Y&~H{YRH0|EEq%TOs_gPsdTWuVbQI}EK$8QdI0v;D
zrs!DIbDJ``)l&-4cAbVAKy00|9JZNKO~f48Q3q)FUSBI6<vv{K!1eLz`Tz>yl%sgL
zehbfCLcD^}tlYEEm|Ebg{AGL8<^G6O-aGo(T~{CXrLqCj;|~jO^K8~<K25%*cfs!Y
z>g+)2CbailCylE?2J7W~FSAKXWR-qp^C+Kf;j)}e@sp>Q?7;s?w#GQ};1jc&r#`<5
zFGk04{TZ^Jlbjx~t5i3WtCy>IYHq(i_{48t^jbr_E&QDJuxu3{(rXoDSCh+M$fyIS
z&FfnvcB)@?fs8q@>~r+ReIb@6V{yTiTgCtdGqHar0_sq`5g;pR+;?wha8j22F8=yy
zeO&Dzm$6rLdBo*jK)2*=&Q6~C@#=p-S(y3>4%W$ui9D*=tA8nZj7ce=@a*Z3qnUy@
zW#W938%bh<wFR<(Io&kXLfsz8%<*gg7a)dN0)NkEITIXq6(N$!L?@4Cmpw^oRih}O
zr$mNAE<dQ+xmCVd8N?6Uz&;`B_)5PhJf~%^2D<mlN=lgbx?muf2*%q>K)7M+PRqUs
zS#PkNxC=mFg)gn+%KF>WD_MkOV)f2IJ%do4d=bCq5_{+2qw4{?m_5YK^BgF`HCT*n
zjnb5lw>N@Oyq8RBwVrm8^|Z;*GtXp)ZgIBA#-lZ$mY7G7*Me0a93N@Hy4vVYZ{{<a
z9x1g-G5pzk`~64yd6Vb1Ky*`)G5Dl~Mo)w&xy5&ovZ8W+<@N%t9~*AN&jHrlZYVi$
zi;M`PzcqXAFgZLea$m0Ks(}J;2*L>@^hTBYmDQS_`?Lb7dk$^YWr}64GRWyay)4(9
zHbY~7K6Z7VW8(S&h^H+P?iS-17F`OJ;z;uV0jp4GafEY(9c@Rbn`rF})6NJ;Nq)>O
zLo2v;)mNvBx~Lp8NcMb=0M3ipy4^3Uby#adF*N4$^N6k1$bCL5;)+B69Sdd3YhoCE
z2e1Y%Pf;SsKrVDKTg@mS3@h@|ArI?FSUkUUgCN?$NPwX%Y(PpFwnHyqdN{3keZd@<
zn-l}Bk3{>+WxkQ#3n@OwdsrnVhp1mLz<>7D`Y-I@Y5DS&e^&P|PfMB#0q?`LOo<U{
zyW%GDIsJu?28ve)%Ag7bY(c|BBHQFk`#gANHo`_yrSqf^m?cp5OMF3}!ONSH<r{y-
z<R0Y~Td87mw321BnnIx?`3SkpZhcX<Gq@!om%?{pXDCiV$>@+iAxOtLf1(R~6t?q9
za=K?`PODkKn&>j^DLTgWVcj38V#$izbxXzKdndi~8OQ9~n1X{uh}2<~2v&9pMa0ut
z86`_#xe%=pWBW}CJS&8dozixEP*&_+Zr)ZJ4t7h$!-RZV+QJ+&-cFO#0*`f2MrH<O
zjIaH6H$cR@Q#6Sm-IkxZM2h-|Tz=@KC(o5>i-9BVbA!7C1d6V#+{)C~jK?&V{6%s#
ziM)<zig+<J()rUpG8E|=!d)V4rOK{t<3Bri-`{AWHw8w*2zg@}1dC`l*E7^|lJ;Dm
ztX9$(n&Q|rMv$1!gzuSeJcV^A5l*lWoLa>`U=0ibSp)NdflaY;-z~;npG}A6lxDVZ
z{fNM`ZrMz13$Kl|DmKe%FhW_+$EO+@)p9nKn?sWg?|xtTTKo5v&jzjhH}F}dzuZe>
zQmXn!R$Ym8xGjfDI7a!*9llqMG3md~g?eiCv|=Apzkajd4%GeR6AttWsAf4X*5F2e
zebVLzE7lQegex00_o-%oeY|p!$}3DY;+Sn%Q7({Es5b$GeM}jy9Zv~9dX`F!^#Ss%
zVi1s9Yr@25)bZHuls!c)n-x3j4ffC1JmaIrSq^m8eX*3qhMFfS%q7=2oP6a9wbL{J
zWk2#lYs)S`tu(`~L^r4DvnQDZpHpyb+XzD!ZtP@v;Y7_w$$$8)rk~AT=U}~ocGf;T
z9&9AYbN5tIsGdqV9F+y&`sKj#mhbFIH*5urGP~E!v&X-XYtPyKyt*c7I~n`)%ky!b
z7nj6*B5I!pChkd1x4)_apXtp^LJ38x<F$Yid(GN$+LbPrm#555`OgBuRq3RPIE(7R
z??CNuMKLQuK%#G8Kmc&qlKfQ^|5^~goy>pxlWv{J&J7YZg=~rAK5+kM#r@^jKEn&a
z!LfBIsQ+1e|MAQI{M}72PUw!=X?lRg9g;UqOsmgonAqNOSd@>Zx&D`1{TqH}jG+Ln
zEk-B4<X3t3=MVYcF9M#QK&ye%$qf6Lla<kPXVk@Zs`=?QB*lo7TEEI5?Y?Q;S+)L0
zhy2$cpv?r{dQKI$p5Naf?*AZozrC9@Zvp_lM@BG7_77$Y071q`?yXFlZ#sHqEwl!m
zpgDZyc+KFSuBG3}^WbSd(KwL#|MDA*v55M{36Pz<1ag$EsUJ0CD1xo0zz=?7uOa!@
z_x$H;LQT2G2ERhW+27vc|NYNwI_T5WSxNu(i2c9(m{Td>F(SNA((d=i60S#_Uj>Sj
zAz;d412kiWUBH7^v52U>1cm`|3Ted;FZI1xzeK=(6X~bQ;?-woFYdlMSM}ny#Gjw)
zA79wQu8Iww3fK5^;B^K?094PlkS7k8cHX=4SD@;1SN}fDWVZrJBt%%xo*4M~rCkC|
zXZEx|45tG38*r4|DF@QRAyoBZZ=MJBqd5eSAd)Nst9SH+-PI=YieDEypNS$~i%6vZ
z23N0pBy2Oe9_(vqD3Wrz71AyB18rh!vGcx$aaN3O33M1hVc(la4@cSl_38g|k${qH
z`*a-pe?9`WAC;z5@gw_Vy}ErJ%r@Y@<=!)&2k>kWlxudpRp_I9C_3iabU`zPQI}9P
zoUSKNWc_)h?6K<sP*d4Xw?@1N$ztq2*AAL|>qlM2ZKvC!DnP0F^@94V2cSI7?-Fcd
zY)X3&Pz}?&=avPDUm=GzQ#*^&5Zt#rWD%F8>-)_q<~aO75!5;6eXzv5(l|AWtX&O?
zTmvOOLp8XA9Nj>X_81slgz>%a6k=NGY3U}C0hy~Kpv{o=0CqZ(s*J&U_dVB?J@5$@
zkXbQHaF+J-zptOaJXNPSpzXr!-6hRGZ#5s0{te~kn8vu;P0kUZcwKt^Qsa7#1e)Ru
z`?0CuD?6~xhCW?PwfPBRv~56y-=gc|p%&mLEgBerA%W^rTIKJ-0)BOev;4t>)Oi8=
zS)SmH*aGWo#zs#1!#h8L66*l~Gu`#qm88UOD~`^7(HFjL;J+?ScY1;Vsn!4#orMfT
z2OzTu4J~sJ8De;4sr2iW?iJ92wAMNM?Sw#ju`w`8!jypUoFPccI_eqzW%-=;^xkJr
zys%30s$W|lt?RfeBn*Y}Ifp|F)<gfYP5y1~JcEEP1FVMm&+YAI#EJfKsIR&QCkfMp
z0??L#4%B_1u>st^7H^>H8}f&vEaALGa)NG5<WvL@rv;+QQIcu+?KdZj9nHV8E-3B-
zj{vDuu@Cbt7jcpPO-*U8Z*SsG(1uZvX4`?QGBTf&iUx}%&ufN*<ywv+HAbMw2)j~y
zU;{q2NHYdf&Yn>#0}I%*dQM_ojo_DJ7m^{~o?mwhVR9B2{tjA|QMJ3F7yj*u{|1(<
zenjdr?H`8=9V-DLA0uO$>5pRVc(8A1lkKt`uaX!5iG6U5ogFsXto>U}%6S4pwX_lP
z(?w$8Job$su8V0(5l6#=k*gpoANTg~0Q#l#jEl5s%YW;(|9WoC34=>}5Pn1S_e);?
zg~OMamv`a3OPzje0_dX~Uv=CN?s@<|biuo-g|`v@6h8qNX2tlpda3+Mm&n&aASyh!
z?41hg|DvJ*$x3LCzcOsekqFLn`n3*cP#`4O9qfbFAjqH&V$0%_UG5!BA?qx8@{9bv
z<`h}DtKThE0&L4P0`}wnUJ<AchX8{wQTAKOPf)8}Hsmmif$3qOCUq1%$J@XRLElvT
z$;`)zK?ZB(nMtsPMtJ-@{4N{P(!?37&yK5sjQy^dGzJ#+E2@B8-^&L;#69vN$4|!Z
zGMk~Y%-+n)Z6I6?QLY1h<!70_PnyHuWp>0zOvCsVr&)fwKDuHO!eBzcOJVh;w)n)}
zLSCT?J%kHD!lTnXH?y7|j$d2AgKp%}%yg8`q1p7k1xMiNP<%{D21$8?B_9xMDM}yV
zMeB<1AFs#twB?!&llJpMD{iC_&VT45e*9BZbky<Xa|?p4S}>_T{E75h0Xb+Q`$nXt
zeC^TXP9^6@eSRK{Z-bq-&B6Yx#=(R92CD(UJEAvljadPoc=Y3`<+S^tU$#?o1DdBm
zhMp1`<<xQ4=?F7kE8L<aPK};RfZ1z7JN0^qiOwbz$=jqZZq^0G&SUT4*nFTl*Adtn
zBLrW%Hzw~vpT+?6pG!cdtuXJZQ^vUG=_uJh^VK-D=BubB3HGS@75((KmKq<;(pwwc
zJC7P}{F}lyY(P*OsQ#Fjl8XcO%{w!H`T&WoW#W0%!W)lewuxX`c!o`-O4Vj1)Zvu_
z!SpcTVp?>n@70SXR~##kJiuN2()z{)Uk1^=fEZia;89l0dWGorxu4g<+%|y{GV=S^
z$HgU;fKjR{`$!43*mo;3`oaBil3uSSw~=>t&zx7GcACK>e<mj4s~dW+w&mb6(C=A*
zZNR+KqLZ-3I)eUk`*Ph+@9NoDgA614nl@$+M;n$si%lo6Baf4-J?$j!Oa-ee#Ek4A
znGtZ|G)>7KrJQ2`SjU~t%9ihr!u_vB=o}J4HrU4XHok`)ICYyi6<@*^=a}K`@eYmT
zKD$k>pBW{cZRTxjgoRVLso(agYCif=^~X8?zxTEJRj379eXALx@_Q|J7sg624(wBG
z02FWq>}G8WmkIc2F_7Xrc!J!%7hH)kEYih8zgIrrH+S`sA%FZIUI3sM4Q=TJpNhAc
z;&lLQ3hZfT8d@NtE})!VozD<ovpeYuPK6+2#M;>%ur~qIp<C-k{dcXQn5l>lX9?LD
zp%!Wh_|RECpt#is`KLhIcZTon&P8(NfQ1BlbCZ+o#o*}U-Tf^7J!u;}tbs)&lb^sj
zXaR{kJ6=3>fM!gK;DIHVr<yuPyMAZ!;bd~l41^Ih>%dF^v3UFH4_`tu%fvA`qut}3
zI+io~xZsoQ4baj(#nb*C=m!|=5y>=7;3PH!F?Y#qjcKF5{1+gY&g320$sa(kVhh3?
z%7GiihBIiF%V=0BLnM^zF}}1XDcPIBPR$c&Lc-*D6E?ZvtaAu9;?K@N#hUAYGiTpb
z42QUWDs^&F*Ti!bmF!m=NGWV#Cm)kyhgH6;Rn#5v;mmro*C)Y582+5{yRp#2PP!47
zaj%VA`KZ|Q?fUYvDsqIH!@{!@Ql^}(=Y}QNYZxBa6h9|#7J|uBFtMj7-Ds&ZO_Sy*
zF**E5(wYSxE?L)}3y9ELJ6x>YL88;lX7o$p+gh}_9_t%Gv_@jRE@;|}$qhM=f6WX@
zk7GvwLnhU@!c1M}?C_~AwL5R%yMoNZx?fIFq<v`(^UCjD<4#RQZ-43l)*JNf8m$to
zUi&AO&as`(kSyS^<A%dF5rn<vxgN(otygz|qL0k)k{P*SJk_~GP~j}nC2G5mA1%9>
z)l9|<0<P3nK>c^HE1$K$uoSo!SndWaLW9x-M?c4wy8w;q2-Kn;`R#UkmFl(o*(FsH
z-O^2ne;@heSrmGV1p62pf({Jm+hGsmrL-k|J^)PcheTxJ1Uq*4T4RPDCh|4=ZOV`J
z#FHRugrmOd&R5`?LYOdvl{wvV)iYlH&Yx=3WWaV-Wf|8g8=eR|tqD@d;#I+@8j#FQ
zHQK*tjuc<W^0U5&7jAv1u#ktT*eJ-<K0GWh5YI=^?$vBLvj_@q)gOK|{B$7L*i}-`
zSPRHnF-BHNd3RYN0!x2Gw*FUB1r<>OItb7!z0G|{qxEa;8C!AsW?jfPW7Ci;^|cEL
zvZTdO*nkSrYw{tv+5=vZY~N=m1!-cNYB!Qs0xvyv0LoYg%;AgrD97R#{>G=%$wnOc
zvLNUv#Fe;13gn3@>MtpX1wIQ9DXs~|K(XIv(&=;`x>Erd8FL`k{DSM5h0=M5{3W4N
zWllk7eFp~DXI0Jh-zozuS85?35pkAPPO1J0&}anKzmKP*-71PcRRLHHV!UY1SWG~-
z*mO9e&R*6{aUxKMqxozAAM5S<x0JN22)SM~LYS@7uLq#r3P~|IIs8V{mdNl0UOU^c
zG#DE>Z4*Mgp2398U#Ho@j{g*qDK_>zIV6~eP#HF7vq|P#l<a5o+1Oi~*QpeQh&-K$
z7U+r@jzTQHYQ9lT31U5K&o?V4<%gMoWWiD_v_Kx~!~vg)s?ky*Qgh=;S8BXYZyV&x
z37{Gb)p|M@d8vNz`e86mnmz^e@>KncAE^{>5M9FDzeLhMnk&p(mL+G@cp^_NoMNW<
z22ABHFv>6ubJ9*xA6G&IuhLGMj`?#pUU${5kUd_D%0#~;u8IyRH8+&l9yC+LVu%F-
zrTd7dREb<LcevpB#ke?fosQ?4dA}YV{TIhktD7L(T_yZ8M%<)+@H5t<>8xmAOfO{V
zbS+7Hz%^#QcPX}pBRhcj$X_B$osIFeOwnfW1YD5DHm|dnWhwMZ+n~R)@Iq8cD&QBh
z8L9?0d11EZzGLu1LfNn7SlIZ9T6K;aC@Gi;EX0I^*`&_PCfbx-1FrQ)n=Z{E;h&)@
z8si;bhxqnFb|KUnlZ_Cj<qLVC!8FSn0Tz*#c<@F!!z8P&PjPsdvRwWc4KdcdMr&Nb
z@}pSLYK+A9d%Y7<)3Od{r!Q|n@*7i#DTO&$pT2u%-%DJzT51_Pow3w|-!=Agq1q>K
zpmMmB0kl*^o}EJ&(npGJC(UAtFU%rON^hJyihj1wZ5vxztetB{mKUa+<LO-0dO#fh
ziG*DpPdAp_L9nt`n&_M!xo;U)-u-9W^|vw{{1tEsj;CM$i73_!a8AH63H$^WV2<R0
zpGuC^s=9g6!A}E^rZgz9Z$v&8_91TfoD|zo?(7T_=3B#&>fzzZ2?jx*4CMwKM}wFW
z_@6vnmJr!c?&;*K5;59Qp6Wzi6|N9@S$UIVBWO4KM=}{s!Vuq&mr$N}?FzbZ>&pH3
zWxB$zEwy+P(94EeRMm4kz_rw`RX^i$_@QGz%9nlFNR6bnD7EriQD+ce0^Fi^ez58z
zuLGL`e(L?zCgN1}0g{oEIIm)<OkB%EV|Hf{Istw%ERMdd^x|+ff0_l#x9gIR3>I_I
z;rShWLFq;g!vQ)-(BYN}-aovv#VKNA5sT$^*!h%TV9<`^)|U#Y*;ciKYd^1rhYBHY
zly6cXOUzN8?Km%o^4+O03zV_dr{HR#Zk&}x$KK)T{UZ3(dJg@8&Yc&};4d@DtApBl
zrLv!3!N*{;o*vp`s*s@Lg~$?z^Y>P&?f92-SW4lj=ud`+y~|IMEDRlsP1G1$d62WM
z!oyp)&~NWX#*!R7tqs4t$<^wJWHDb>5zUmP34q54SM<eHUb^TYXtef(QIDzcI@MCs
zHRQHRrIA=nrE&GF22#`<Wz%kG8B@8{94naS8W7f$u}nv0xEfb!C*;uch=blM>B{XZ
z7qYIRb~_AbVk&jMbn-E^*5DNmv7Lu1EWx%i*m?nnis=M+vM{pqVQHqPF|VUqc>-KV
z7`fJljjy|h1C&>tagxwII*qo`8uFV1H6z$EyiBHDo_|7^v2bfu-<V-1^M>Tg4RYzt
zn;h!HXB?PTjGa!on5POdSL;i*^5Q>u+J%KC7`%yaxtem{rlR?sb^a&WxTD&B)o<eY
zWV9_pjvj|)g<gCPqKihOWB0dubRIgs{Fl<}8C$(x*Dw6~*YVI4eZmyp%%R3tb{mwY
z6)~hdxGh^)VLRgdWSW6~JK|9(T&iu9#$x&IJ1$ZMg+MMCOfm>GlY&H#F|zIq`|Pi|
z)ZQ1uznVYVyi5+W$6db@ocoTqevl7U4ik79_Uv@}N&kwB6T(?7jkN|@@aGN2NV06H
z_vXb6J!Fl2=TA7o{n_Z-+d5V*ru!OLx4*Ev=#%WFL|$5FD(4w;;_JJe7}91f&*lJN
zS$%(XQFD@2p<KgMl!yt<#ZSa&o>5g57%*BCNMlR27HjaX9rxsR<3LFkBU(q;f#)&b
zfY>b^JH0qNfmHE!6`w4_P5%8$?DVZs&goX-?Y&DK`ikRr*3JnAmDZBFt7rBTtMO<3
z(6>aIx45TicJxleM!AHij4FEjuW+?0v`STwCREaVEMLPu13RwcXs@N}l_g3AvAO<J
z!u!{&?Q2~5{U29cM=n*n@-gMebniy_)@aZDy8jZhk8<z-z<c>tw!cd%$LxAE5lphK
zm_Mpj+u;RvMCZUTo8C4_FekZQbaZ6koq<RvuF)7D8$L1?>WW}9lbkd*GE)~M^m-S?
zy+SBe{o7qPR2CZqx7+rP<1d!e|K6*huhf5GWz}mr=E454Wu4I@K&ik)T~rzEm9S||
zOetZ4o$ojk46=pqzb=PZu7($;{0r#p*nmeC!w4{2vH|BX?>)2x^AFUv2Jr8;o5xg3
z5-i^ueW8ZcTKYu12If9YXkUBDP<UHPqa2nS1sWqaJ9#;5N{g*kwcL=&9Y?M(gGacE
zvFYVAL*;O+aH}~#Dtk0TzuJ<}dyH5g-~XlY@+Fk9y@guZAy%)oYy<F2sS8{?gu*-&
zD(aEw0zeOHD!DWn+(Ud4-`Je}!)44Zr}FAI-~)Hv!T$EwvO~YR?h&Z7(lfMIepe8>
z@wq-o6s_mgAZw_AFPz<ZDhxom5Rcgm>qVylBhhw!N1dd)I-_cK4VqY7lz{`XA>M%D
zOk7ElHPR4ciV8gK($XS^wP=ng4&!Kwa9MNLY(x2z>9t}{*naP&iYwkMG)?@N69G5G
z?M`VVIA1qxRmBBIg1H?;E4EvC3ZE#sS;lDoXxls=q8Oh=)}TH{*4~c2Cn4&6xJ#Eh
zdi|anj(b|?dff=0(^I&KVbwH!nYLYC#iT!7Mcz0tKpIiW@@i~(S2;w=laD|}ji{g<
z{3^e`^H4F++%QbcwI3!n4z|8!*pLE-gUK1eVNS!c=F8L-BTsI7vCPHqv3d8j#82L<
z5-qU*afuuC-N@An{iuXX952#Z;CUoWKAFNGI)YBF<noRo92R^uox-rwyhKp&eYDXE
zOL2i^@8*F1OU;-6W0R=?uGk?S+mOA?=WCmKUYp_^RevL}{sp*GGQJ8y@>g}@g#N&+
zye>}kzv?tfrZ>Gz#D^B=Y1Rk!{pD=u?RGHEz~JXqwSw(6snaBn0Rs>rQ>yea`6~Rw
zGX5*>$jDH>KA6>fbFQZ{ezqER>L>8QYxaL*@RTPiT5<%*roLR>if(^QlP-6E!xurq
z3|f0j4wjZ=X99lO<vaBjf<vQ&G>@$rwF@zdB=7B>U74aKUIGxIh>JbbEZb-F<<4C4
z;Y<G772Expc17AFHWw+Y!NH{n`_Z4F;5q24S0aqYVdq%T_Fr)B7T3b9au8(ft#f@B
zdZ`e6wzYeB^$f~_Z=8&a!JMC6T#DH!YXPp(^{&$XZQf0={e;*do{MBKEsPiq2ggHn
z2>6<@Rn)S(^G!I(4p7ro?=R$Ko>5NMEH*{g9XUIM#pvp~=(OfdIqbV9wMya<t-%_8
zZsGe+^y=~Gi4hfbe@VQOi~QJ?dpYn;d8JLXeUWx)>*ZF7iq><cU!4`csyrB%=lvvx
zk*hJv)rtf=#?rblWB0(C%oikCfV|k%ano?c+)=JC=uraWH+Dq@#o4=u;5eZqbCwMG
z){vnejE>5KksRd7RPlxV8+1m;P81w$qBQ$k9|3ND8uv+Lv^m7zw=;)es=yzn6*V*{
zgHa*l)-BVk1)-zdI%`@T#r!A>G#Bascx;SfA?H6X$mp=~EKh;H#|HmOM&a|DpCZGq
zl>lV>_007ZjjeAq_wKy)>LM(V1@3)&C9mnfXQz7aFAkbT*)rxZEp*=J7e*3gX=jbh
z6qJb;gxXUi{nUG9ESrP_gX8xAa^rMpu)>~c;S@ko*dNX7_GUy>iCx59D);bwqCThK
z=X^`3LN0dl6qe3z<g6M}jfTIBai#9(F-3)}5l!`pvKDi5K@PAlG3CLzL*q)S3~5}N
zUUa@LpZBkdt=XDY(m~d}S#o)DfGbdcrdjbMfe(0z^{5sCO2i_B)`usQc7XRSZrC+7
zoLuezAn8)RyWgHfbK>HPTi?Yz1S871xhYZUX%He1wH{r`^G@hq)@Jo+l}`P~ID?L*
zZ#DAC)vKk`({qx2Mi&R#n5b66Hot*>2%BRW@dM2zvd?E`^q8Q)%7}GL(28kx4)2v_
zmL0s?A~8m??_aN6IRvl}fXT!6N&qhSe!d27mO0wL^frE{78KP&97h^{dqLg-dE2JB
z*7;o7BLn-n>oWE4e;D85Rjg>!F$5=}m!kpg=*9nz;0lPqCQ?-X+zpK5Dec1i5)4q>
zd`w|2%j{R04PY<Ym1M7p&CxCk1*$Fi8+d%ez13<pb(LY=xBKCjP%YGzT3}wsukfmb
zWZTK4iZ`MbpB{=JIZ2uk16%_o9RRUr=rqLypl>sT9ZH`D>Y}@Nl=u6sf%~$@Y`XLB
zEZ=WVegMD*uEjz{M)K)eE#Mw(C)zzLUo1E$T%pk~IA8~CT(SZza%!A!ybcM2s2q6R
zNeR-Pxjf7Rkn8g7(y2fu@EAT^bDWNMSC+eT;VP$=PEQMjV0QAEKRj5fxIdjRx_Fxg
z$2pl+y@fv51zdxw+c(<JqSJ*Gu^mS+s}2HN(;_$t1@fh;ONF6sV5|5r4LgmC$@uwQ
z4T`tndR!6cie+{Qd?b_eV7jV__T$uBArl3Kjq-HWm&a`YV}sW`>9gd<YXjphcS?+t
zLFC&N!o8YoFFXZNVzk4h5a@<#9XGHtN`c671!LT!B)g8fA)vvfpT3sxYaxFuDBQZ`
zM3xtzTzqDfS&xEj`~S<m1FT1pg=>}nVLjTEe7Z2C!eHBHf9vN9HPlmO6PNMf_3-nI
zO4KvUUrU7YZZCsP+!DqO(r*f~MQ}7W_yVSBiKz^YNubR6h#4RGqStdzaI_Y61G(}Y
zTj-X{-7G%;`lV?)%dJt+G=GccBx2Op4iL>qPYDbFr?~~@JElX{#eSeTSwuR&@uaDa
zk_T^J1sqln@+j84smR!0D+)Xfy5TZ8ieeaglaoTPz`VPeV~1ddBqum^G=8}%aXsM6
zdbGC2Q+H*Xgo?QYc64zI?GONHwC?)o(<(ZC6y8@8!&t(?)B%d_qB8143K^&A_i?}9
z-UTZ}$v_G~0g{|O+w_e*3k$%$(Ebsl%Nxd}x1FS%qF+d4wu__6XwFtKEY0ajWxi}_
z0G!?q&-<<P!wVqKDcTDFyepS>Rd_Y3pjL2*<Ne!doktnqnM@o8U<!v1RGPF^V-BFG
z<#%ZkqquhysL{{m;O;LsnDP_sgE>iZQ?HyXWbE}G3?9TifYe%T`t{4v*vy3kId|{j
zb#GrwHvnXIL35u)fB`yGiOiHc)sSyyeF6JA#_7IGPa-sVOQ`03{7J^FvN|P7Fh;QO
zhM4|<zP6|9{U4cc?TUUY9v4vqO<FFZrnuv`QRt40gMXR91)l;Fk`NVXuDIW4a9?On
z<|ly3$BiL?vj{cw?)l0+ZJFxHR1MKTmFL&;EHD5U1y?2_fb~LNn~Pu$^$}5I|1-&<
zJHZHK2=O(Z&39%ik{2lkHWL=$+PH*apMu=UidIZzJApq@e)H^4lQOafwPVHu<w=_P
z=TQ*)p>AJ{YBd$p?OuNrU1@V6g4>N%Q;LP+B794#&~h|NEJ?M`5irCb!cK;rw4ViI
zTgn<R51n2F=y)giTL~hYb^Q*o7%YKwS<Ps$g}fw|r1Jy=f_6Kk6emHWC~wj`kiRno
z0NZbfM{L*QMILEkW)M<ry2-ZXL;lyE6FUX!h^=c#dIF8}-4YE(UQ!LorP794EG){y
zNJ)EU8X7Wiu$*uHQ0nIDJqf>$`7Q-UEl2F(=~2Cfc3t6vVSN6|h99-!`gjOb7Y+dP
z&zkGvMZ17d>){$>GM9WvMP3d-`rtHjUGt^4WU(_RpP$a)c4brPhBKOI%BsqhHIniK
zULsJ_ZK;=r-PgNg9#f?C3i<YnEP_ExSLwR>Ngk(gmvBzl?bA{Ww_rEyJDxhJ?$aF=
z4_*MkPK$hCn41Y|iM-<-Cg%}Vky7BD*i)bL??!T)sdEcY6W9{uWnl5$QKEJdV4Nw0
z%ji80^IxrM<yqI_gs5|wKAzj}Q@arA!pF0ONeS9^OYQLuo;qjrF*l|?=r@X=#dT-#
znWK@e=ulCxk}R-yHQOGk$yy;TFX|I9%}*H*KfV`ugW&=1Fc`u#mM}P&+_|@KN%h-~
zpXuJ51?<H<d1)Ca{7dP~zc341ysk1!99B8@@^!bia4Wsukn7XxFuVx7KZc^Iy<WIv
z9VV$T_^k~VoAFc*q1~UcJX;3mL?W1Va2r^?$tnSCJ66cUuY~1jPE}A344>AA=&kc9
z94J43j%wrL>&+GFf2Q$m=e-2JNJ9Vo1)YjcolZr_7Oq3G4G=4)j}mtRA@{r4TTXqd
zY8ev$GO-s}Jnf00+$V4NOIQ8Omw$g4A~r0C4jSStYUONDu_ch5t^FACQy0>xi`W-G
z1$QXcr6P8*wXChTMRRt71B%n3LlX}l#ly$(VIS~lnl9oX{L}>pzfiBuOcvX!R6@|H
zz)2UuQhMqM2Fa;{F4i_EF##l10^G5am>s{ox(Fb@;L5U?Y!O37Q*?5hA?%yuF*MnT
z=JGKCmc*j>jX$Rf-5<Pi|IGJ*r>%Ry%;cLPuLVSlsKn;mr+D2jy1bsC@J$MPwas{*
zmZs}nY(f~H?Wohh+2#3jQeTN}@Kcu@>|a!V{DdP&hCGM4uG^S6f%ZI+e>X7<-bu`P
zE2popTJTE6bqmzZVkoZ@c#gzpdu))PsEa`RAfdtJ)OQQ7py+T*N9KCnlC~~#E5YEC
zFp@&WW0iQ#5?;!W)E@@EM4fyY31J`o<xLx$TxKm$-a<%RyupQ1h^}h5NHcrktD(a{
z$E~{8(-}+-aXI*~r3j(QXYBwwqIc*B>0vQUdH{U*eO4XQ6Ds4|q0~Hwp1jJ5ZbM^p
zbVRq|AeC+H^P4TnV8NmGv*10Q#Fco{>kjE!(SG}O@CSH;<V)n{d~nnyk%~i_qvzX;
z83A+JucuAP9po{UAFqhbj_4Tiexi~Bp7hs&F+{BeTDoiEnykg5$r;lmlNvukw;|yE
z?r%{nwMm22OQ?3ajJ%<eTvev!z?RlyaZZ!Y8pp1));~91%2Ob?-BXJVSqt4R>z4RI
z=asOQo3D<DJ67~IYCgBM4hGvTg87CYtx$JGkcigH?RqwPT-wmFp#L)Ev+KO+vSnfT
z*Kms>i$fD$mptblX2T>a)OPrq*r>sG*O=Z#=N|;x4segw*gZHNg|h8bL@kGw%a2wu
zEYL}Ae3k$jF0<n)L;8&bj!-kEqhR|YIyw$=i8)pWc@0C;u>?>R*)$L%Z9>~?8W{DY
zQrRobwH-Rr1#7Ear3R~arH5&!&pJ#3t_zI=>ry`5+t`53IVznol2A9(q}uuHMEV>q
z*%gWa=~d?XL%CE9wz7!BgA*&4YJM()BAe^(EZ@qB*KPg88ThH1_?(3Mu<*5Td(?U<
zl#A9XyRPPuWow_}y`<39?#6a<j3r(*!NVKOhT9~GH$n`eK;W_d61&$1`B_sG*ER#R
zuI+qcz`Fhp*7ePlw|<pY;Mt6Fu9s?-yfO`UVEsAQRKqWfszTJw+Uxwt<djeTD;?l(
zuj#fS?)kP(KEZG7j2m4<i1^~&2-i7~_;z}YjGd9zheDq1g1lX`sbhjZ+i4Pt1lEU&
z6ZJ<JU;R$!ANSbCrs=1molmmG8Q8p+T*h(`necAE;8zOA1n4swnLg!I;8m(!KPmsh
zASSG(_i)eo0Ga{7!qx{yBa$mGmtUAPK6dv`Rd!Hvacj8uP3k6^$Zya0rZHqim!y#N
z^cf#9kr}KOagT;jYSc;>sDt?yy7^3ka)UiNCK!bu-Y}SC-NNLOP7;_hR%Fu>)Pk<^
z2VFy^LA6K)g4X}sFr}@ZbF%_e0NM22F^#oon39<iA=2eyb<OI=sLCw{VS<#msg0AH
z&&@MF`huEgz^ovK8(wT<!MHNyOL+P95g@Co%6(Z<`=WR8#}#k^X4&ngS-ip}R7$TV
zO!y0w8p-v|pc}j1QLD(N*Kr)m)r81JT3}ns5eM5ywRM$IY_+k2+EupUKqlKWJ%ny1
zN%EHzM>1{9{fdJQ&Q&9VC=l0PZ>!T|NH-eCJSAX7h>70pd4@+x4lYS&C!-A}!VKDn
ziH@E*KaD?i^3<LZ?t03=U28svFk^iGCwjAMcj(WL*^0l7yzEAc<`PtLh>lw$cz(r9
z*8lRpzd@A<mbyjGDN1ioYRf6LJYs@+jnZkle=__@Pvkrm<~4=<*V?&?PM$qEgR2Dd
zhkH{prf8R011TokXT^c;^VIR8bcK)d%>4w(OlZyM*%bn<O8N0O_baORuT;n(Gd`9a
z<i!Sv<kgZ>Z?wb{GBhv3UQoQ{SY%xEtCI6n_1dIHOn+VZK^hLn;PzYt<UEwqG#FS}
zPi003=dkIkuPAz+k!N_2GD8T*XH*eL&6$M{%pX1bzWUf3$-se9k@{St#mJ+@T*YLK
zt!qEHle|JWlSM}RTATA3iOUBxK+OWAfw_-h;krhf2Z<T81I1B<OSJQbWz2r=3zyM9
zfSEt3RsVS*AOcCDM1H7O{!Yp9yGJ2M7i1aI@`&g%?2*2u`C44iRV6ZgP3}-Wttq!j
z*030^qM~b-F84#@#-ETYponhqT2bm#tDbn#jNaSb^O}0OGDg(PgUV)?k*DrYJi)+w
zmqsy0>e<e1d^F1Zvw`J<TPZ;Y)QD3>cH`P<HhV3ebR3Qn;N3S}zd3ZIk9-o>W`8<Q
zW!e2b@7=l-MO?_NFxL2kP#K|%3q!_frOSsL)OYGn-{bm17S<>+e23Bk5mi2>fg=^p
zmwJxm6=%~5kj_vKe2n3ONrja6#5uEwI`8uHGj^)-^jja_9g>kk1isK)&Z>c(y+pE?
z6~5VU{P5b?ob^n=+NokVJ6*F?pKWbdTugvNxIyXJh17C+AH2-|8K<WBey&xUR>}My
z_WZmP^)lZ-BGZi$S%0pwqueY!{W;B9S(k658_^p=gMijo(S@a1-JkePje3KaOfR<{
z)rd*7Sd64j>l}-F4dtNbcAlO+iWUsGBL88gt=_Y0&;GnmE%rWR)K{l{%FpT!_ayG)
z);)(^G|dS2*rZ!84wKlvaF>%Ce`BP_N3!>BU*;yS@Vp@YChb0O{RwM1IMrnT*v)xi
zd^DEu<l}o7((3a>5138SubF1Qmx_4j6qk-z>^R=H`o8CU2pELYfw7w(4);v&j)q@;
zEzgp-?yb9XAjhpoL(3JxZ4kH-9kCX_-GJ?q`e@kft(iWuKgVTc^f$52U-wTsZ)o%M
zc6v+syKqO&mT34iiM59><=f;F{kwM7PrqHkvBd=eKBfe&-E=b46;n)oDp*c(a-x1l
zMe{fAJ;Lv&>WPLa?!0`;#xGFz;oDYrT5YGTJisCOFNlOdKJ@hF=jjFgPW<w-0`DZh
z_3`!52=jCKUk!bZwkBWDR)heSg^~Vko({r-d-h#n9ymyS$Q#<N&Ib6G8f;(O!+*_D
z<O*TN?{+}bSps@A=`kKu?XD@Lsd0B&u?ss-+w@LFhxwoAyuXpz{MbSJ6sF|J`^TF_
z=+1$-9w`q{j4gpIfHDZ*FwC6HMgzrd5on1tqfb3MvCIajieK#p!t4+`c>qxpQSwt|
zc6%McEL~+wcJW3$qCm|BjBD;y)>tP#Lpds+sAz=RPOK_~J!X<Y*`imi0o|h<h`&PM
zUkfM))PKKRUq}cprh1l0gHu03(k?DM(a;bEvL3wG#zG`l-mqa>`fhe`Xa4sAk{rb2
zb))}}N()8+CFGoqs{ojpHzNn#`66YT;2#n}ejKada}I(MUGXsPiC^{PtxT06ps})o
zq6pN1j9^}?gZ4Sw6vTZhn|-CRDSI+71<K!(2K=8^8MZCKzyOH13Mz<#pTZdze;aVz
z*di*r8QwBQ4In`GL?H371=^Ex5NhyA26TxzYE9lev`w|4I&%MVM+m?jfy+4pYVmJR
zknuKP*bD&btPPan1djy6M>XArySX=ps4!i*Ecpak0)VK3Am|1??|gK5K>Uf=q+{TL
zRz9=Z#By8&ktq4TJ1P85W!euo!9oRvXhgiuNp`-Q2>5@~$);9R#bL^s=PXr!|E_;%
zxQ)3XF~e=G9`5Y_e)<15nd&cJk&|nbFC+Z#pZpJc<zH9npReOX0Si-%MyEWK`;QCr
z-xk<^`8p-p+q3^XhWg8U{O50|zpVra>ThyObZ-CueEZMrZrc8J0P%l$oy`s{c;%RP
z8UI4U{pUCS`zH$wup<2Xek2fnx(4u?pm12oIl>w|)}IAXZ~mPM`^PFZfGFUx(Nn)~
zo%Od>AkvB72QVzSwE!vmplR)!aNkfKf>|P<lM<}Pzp|!BLS=g*5E;_94O93vs2GDp
zEr=*K1gdj$fP$zMmWbL<DOFGS!+ASOvUmUBy{mr@C4o5d_Od6_rvT}6%aocatiFg7
z(BjAk!Sju~G(H)^f(DCLdm7swiBGNV0oeME&Ef9aHi*1wjexkUkavTZno{w<i5HCN
zsK8i=POsq4tAKz+dP@NCDh6O~ou*uCB$G8{%_05saHG6WOKk{D>;`fULd<5+cgR#}
zoucPKty&8B=^Zj_Pc{3V<EYgFyzM&$+Xf+a98hffO9Q8mD^`8ET-%;Nt+0Q60tlun
zpup@aXq=ej6o$jyNz(QJQe#35ImJ%^Sr&mHr;yo33Jw*=8`A(XDg|ubHb9A#9sKy5
z14>t0q`Qw)HWs760w*HH{B58EfI3u2^Y%{NuwXCeF(2{d#7PoHgaRJ4xWfRauFg%Y
z&B_@3`$O^9r6b@8ol~AabUFVxr&xWRF7fu^0FBj)-G5>lV0`ZZ56%0^bhz*^<gERl
z>!8!I2{@~tfb?*y$UoTC?otr63$x`OfiMm<us%3_3q)h_3Pp#e)o7R#L5yH9v|@hU
zwS9q=PHA-$Py_ih?t=&(XAosI0s(CCR1#T!AL)DgfWzfO=)ypU(-O7OfReF3#V$ZC
zDKnnrldX@f0_xrf03hzDoxA*V^DGCg_PI8?iXss#G%p6%W3)nv+Tg!$a&kn$f)Ji)
zKl|45_iJ;)58&Yq-dlP1&&BFiod$f%f+H{+im@v?6+61hQm`}8N{t-E9*zQDYsdpZ
zifMKQxoEIcKFe=>bKxtL2?tnxbzw3UAg#+$XBntSgbm4dRK(($S7Qo2Ie81NiN#|q
zB`;H8!Bh_lRLIv6i{E{8eC@n@-BG87j2VRSgYliTP|NQS_&@EF@ua`D$J-^LznvQb
zPE><{{<#F|0^lzc{P1Rs`Pn;kSFRcNJORGbFYN3F`*KNzaL^sbS?kZ{K`JUIs&}yf
z@cy<tkN)qwoZsX%uoX_;Dddj%bqv(s)|db^>&#IA49ic3$R4Jf*jfQ2xUAc?fN?qu
znB4c>BVM|N1pp|u0?Gw4sk8#g77+(_N9Dltto(zlkt>i}oT>2Q|FHQ2_^BCO0m{8W
zXh)Lr38K+0A_pXNE~`4*agP=0YVglXadn4T4Rhd3toS8v086RJ-36lK5m!ceEt*A0
z^?iDwWS>aV1Q52b_q_Mao0#T+9Goe(z8^R<Bm&Yr{b$rNVg~|5^?hgPoqR8IlpCs<
zAAFbn`3Tr_<pSNfb3<i3SsS4aOQ|2w*>nM@_AInCZ3;hZmzc*Hm2-xHOe@vC7I5T#
zeZuEM5cdpneTWVbfl`g>rhI=X9pPIR)$InJ3PewTOhN9+YvelxNtust7WUnesXsx>
zWdqWnw%=%aaWe%QC@)A<KFwkTtbK!N2}ynXrdO7vn{S}lCTN)G*bb=cIuJ#gz766+
zLvdI|Ajc>(!og|T_}Yv8s|8l`YNhj_SQ`OJO&2^M<ru)?Z4W-yZj<9CrWE21=rqB$
zAoEL=uRdoXdvFJ$BxL%G1C48g-&)9VEyTQDwAQZj*;Ey$Bd0sqFsYo4<ym^XC$_Zx
z13{*x*@v7}Hvt2}@GIBGnL5+064OuxAGYs;&FB~2|K4QJ9wf)Z;+8s3LV<VTOAu_=
z;~kbBDe`u^$28^C6rJ8qxDT`wnB|)=r!jf$e@%^aiRJg4glDZ#?|HZ4>L!C3plmKG
z2ca#614i*DuLYU;0Xiy88?osPzIzxlI9Z`iJ)F5{)@h*u1?JjZ2MP&^pRfJbP%kP}
z)N9Xi!1R07S8M=4lLtwu%0W_OmVrUJ&OV9my9CL}T}WIIq%t-zu?KW@sMZkh^iK1A
z=1p-uTx*+_#@MIqNR9=}nQdOgtAf(F`9G!cUpL1;UOxm?1<mhgZ@s{I-%Cq?V_gA>
zsSu|R0h#+T&oabPwgt78mD<V42|$fzX_}pJ`<`q8ivkf=-TKBB>2!Vr6}4M@BM)Ax
zU;k>%b_{e%<i#+5dD?C8u;M>bvBQXN05=w0%*S>A553LYNe}#frUj6d+&`Q`vfX26
zdEa#<kxfp{(O3Tj8w>mhL#CDck`q1&9iO9M6L?P2%pRRu!{*LG4(^o&rmmgn@eH{j
zD|`w@ytk7$e%M+x<^nd_kZ1laPVpRcX2pjf5e7~9^sJ*7>x?__sK_`o!*uG9Cm3v?
z(nQ9kG0rD@e(}H|EzMaI!kwEk3)Brv8MhZ|Uedic8vY=j8W;d5W6zB2TWEmiLybbp
zyvNKm#OLq9i2~tuwx<*ziI#&5c?NdP1(xR%VCn6(no$%l0%?g#VtUGaK(sImI2H^g
z!Pzpke@3_y!9EB8YpHTB<|+D=fSXTCjUB-o3#Q{}d1D=Byb303rw2*}fDxncqp(tk
zb&P(fRBRA>pTyP&v{7f%KVc|1GBi2p)4MOKzoKRrZ!LhR?_xa~Sduk%uVjyPleB%+
z(b=>cz+pB`8Wa{0Q$45X?pub921*15zyfQ50~!5Ns&9NqxqXi^GRU<m(Q|7@+Ntz3
zo(cbes0<d!C2B@!)_jP<<#<hPC$0I)kY)myKkbdB)9YJhQCG|@eaT#-jG5hlaOtdx
z_y)-lo#2%fY8b&eDnu4iu{M<tOrr!<?*T@9&ndS;6CD0Su)S&B(RP^orR?XCz(<(T
z>r4Yf#+;_-6cyX>1)t9GPU+5XM@w(9iF+4gCacGf3LvX2`;WW6X|L=9$Td(OHhWz~
z{rSz!Z^(ksQ3acXWeWq<pAfnKaUbD|DfmyY)2$DA^Z@iISEd|ab(~A3`^rVKMyD4q
zDUMh42hCZVD)$qeq(bSXe<7V>z|rw%c`3kqx+pD@lbO~L;*9HiLUoc_N9Z#P0MYV(
zXuSqxZc+$g)_TxYIVmJ54#Von#AgtSt1jR!wKTd3vZaJ!;XGF$qBe|#La#V2Qhymp
ze6A4b8=x!AdCwsE-qU>jvAqjcPB9=E3<BHYAFj0Uub8t4+lC*2b8)sqQs1JfM1$0+
zI1m}O3Ht!mAP8&rDy17+V5M{9-?Wctd7|XYU{fUDu3)AnmCwl@+SE<~pI-&@vabpP
zelY4VISlP35?b~lcCBkBX+NexaT&R8`sw+$Z~ViE%l%h}tZj;4RRAvhsnpg_II>2G
zW;Tp+Ey$k>oFiL1)kxk<Vpj<^GXKj8mB&R7W5M%t|I=rPVx19s-uXNWjPw(UDRDDh
zpXf<JLR2$_Ci~5XyC!3l;}ErK&L#RgG~I=<BaCJVGmLrTdr0OJVdh0+*M462`)(YA
z2+M>SA_DS3%4koBc9KMSsX9@`=EdgXa30KKgjfA9A9A4jk;!*u+vN>LACP2N=HIv_
zLnjp5<AaDaz81cL38ctmbvMbc0tYZO00lCO6md8D++(1#y98hQ0ZQrzq6B8VjcG46
zB)F0nzLt|svMv$wltE4(u-6nq8bBX)1;|}<6J^XTNtFwRFF=~Upt75INjxiWT9!CM
zP+(dnLzs6=OEuhD|FhwP=gRb)`5X~%y-J@aoE)1R_1mMDXV5*jcfyyPzUW~n;-wo!
z!?~67`tcnheV>8W`6C_DjXX^n4fsof_b8AJf>U=}&o2<2bQEMIgNQR{)w2Jj#xp;t
z0lQp<Sxu`ubH+`)@fZhPz`_|Fd}=hg(u>Y?qG9Hr@v0Q@?ERU1?2gKZ>Oa2|J}(3{
z<ox!a?h?!S#y0Ar=Ib1v&64oa7rX9*rao*hLi{zafGV@|poz(B+@av{2ALls&VAx2
z$~%M;tyJlGGyRVP;GZk#RUe{568m4iH&6%bXL4i0`u@RPJUrVd?hCPILfD=VlF+<%
zcQiWE0M_Ao@`^(`_{Tc+0WsI9rx%5g2=nD_eg`%)l|#?fpvg|;uJHZ2E<;>wC5+pr
zvfrx$9L08idibyd^+(mLJPzRHR6yrDiOWVEEVoVO;cE#73hlU28$9ZSaC-_lYT+YJ
zLLaYy=`=cDUew)YK8{=y_DSxb1s(^`&qn5$*M1zss2EjV<<GR?I+ML`*YjxRCy-!X
z8~U*oD^z(f-Wf!fP_5Nz7~lV-Fichmng0;=<&)vn-X|X-sZe9#LfBO<2SHtYf5F`s
zSzgt|Ih`$Z32-`t0&U<K95!~hAd3oAXojuA?f!8644-Prw$7GdbHHqRLTme3*(?FR
zcYXS{^tC!uMkpp$7-?O4{<{JgdS%>t%!fQu`gT0o_^FdP@!S$uncE<CLcEo=qWzmI
z?-4m3<@zYe%tBx@N}|2zb}r5@?5hwm0pGo8y922p?9B2PYzF}vW-7%|8x~ap)R5x)
z@50Q@G%iK9d{B3}-jbK8j=jk>oDTmegsix1KFXJ&JnrImr|78LrS_|%G*jPmm2Wby
z)`jTG!!MdoGi|&LJMzymu(wUr8Rj(eHBjruy?CrG)k==kce&B@PW~YZ-C^jPRLNR8
z?>$^vJ<^3U1x%h5_T;H*?XPEznEM2O&af=82t7byJ8;r=&VUr4^7F<9@d_Fx_c>l~
zBR%q1-aVX4;JTU1@wv{GV?k9sIt1dlFI$N_{CF-{hrL?Fc^1iIv0N_Q-pV*jaM1c;
znBt#0f<KXy)xtUA|B#<tsdCw!1I{$cyKDg4=v1wn7<qkPmx?!h0~Fe9I8*ydZMnff
zyGEl9=TP1GR|ff~O};`H*;O6td{zZJJ-2gVol6Qdu`pSc2>KYdcx&Xxmj>OZ{Wuzf
zG{Agm)ZGLh>wZ@$!F-1-dq0<*Q7LdC_CTm&Vz3xjddFpe(Lcb{D{gK6N9B2(owfEk
z_A=*&Q7IMX+&t4<ZpQP{E^N$=jTs1(G+dA}L!v?~UgKM>Q-?Df%q$-Zf+U29ZngMc
z8sH<R)oio>cI-D1X@N`)OAEBVcA}{X(ID`UhBzxA-3O=;wd%c}um%N`N)ydsasZR2
zIb24NTH)K3h8l%&MN3`k?woZFX9qQ_o(mnI@p;|w9#p8&;|=9O<ST4oFs}hyPiUGS
zhZ&6!2L_-c1eSTi6zM$$<d2X626J$>Tmya-4S)77hGg`v$xN1PPyn>s=#_&b3tQdN
zUv9X*HdHt)8#-?v?uxkAIsQmW)${YN4lS<N7*88C(#^-!1?k-v6dD_ZLM8r<x^|gr
z7ZJ3>(*GCJRW3i!bjuj(R_VkRT*u^@-v4;E!pHpdZX+6N)I7?LGTS&AR|hIkW9+jZ
zpNIturtR{cT35|<UryEi=7oG4A0~Q6^bKkPKG1<BZRI?V*0Em;Rns<PE{BhnG#l}J
zu0Ik$ZAcIsa-o2c(4<VUHV%~19d_!Av3=|-UvCJj=RBAmHhxQPcuh;A0)a%cdvtoD
z{KIZ%E~3OD_B)KMhL~i%JxgUPqWjEnv56jH&kdDRKprT*F*vf=KGhW}pFGfp4{|0<
ztL#1*0<v?#`<9k6_`OUP{O24mL$fLFmvTN$U1p@%_~xKhNJ$^SfHpbczzPT(o~RaU
zvitco?<%gaYU{1tz&wGBfq9Hh#$)8hP1T?E#Ny^odCBT62A*pzO~j71{A=IIkBH;4
z-6|4)doC`Vs3dHatk@N8F`}6kigizWC#*V=l5|A~dN};T&=sZU(C;A{IJ<M06~0m4
zLG6ut&jRou%-ihAK6BQe@FyLcVSV-17rUr#Nz!RjvvCN50ilX#kzZO|UPL<~=jaQ@
zSb#H-<6H9%57;zn=`srlz0%l-Vq65!kO-0$&wqJIxEq@_ZKt=Qfc19JF*HB^WD#HN
z4r~pC#`xSvWb?M8Ha(G6uMEh`0c<XbQ=c}Cmt$Iu43^x7EPiovIUc>SZp5Ovnh_^=
zmm6N4K>?sjn|SmuBQsV44S>F0{sVob!xL4}RCDqIq7fuHWl+ckmWB#2B7;oTD2-2L
zqwI8t0NtTtSFO+*ymBuF4&@?O@>IK?g72ZzKjO}&RF;V;-kx55+-0<xQ2n0A>qZ6d
z=sQykC53P)mCy_e7BpQ!zH30M`Y9DXtl20Z7`#kYJb$-3py4))PGA6Zb{hdTwhGhs
z`FKa<FkV-92@nY`SL_ausq1}zX}1b>s8X5Q>@c%WqNb{Wd|bdh?mS<`6}ov2GDl<r
zmzyN$AiDT!g3`MA)-QMOw2|4a0c5}&W1o9}ADN+Y%y##QDDF4vwhr3elj$gvtYOr6
zj>RcD^E7(dI4v+>^Ba_!q$OP)bi)9Y-45amHPt#Jfz=&lfKt_pSmrNornH4$XOXuw
zxWEXCm(2<Efn}ACfDJU^An<Jc3B>fgf##t5GJxfc?i+`TF7MZDCZC^W0rH6p5YJoU
zTZ!#O#+u)Ool;1)%Z9P76m)6iV`_^ZNNsH3-q1wlv8`ttqU{BIs;0fa9m`LJ!(l6K
zq7EH?J(_cH&t^Z?XKqCDo^MA?gK<w@g9zd$&Y-oY;lIpO-J+QYoHK1ea|uA6r{QG(
z0NeioG-o?QAo3EQ9FUwr=nmixKu>@GxEZi=oN^YM=bWva$iPdd#<6T2T5@}K_OnFP
z`ei0zoQV+P5Ufm3+cj!yHT15ZPc5ZEK`f&gBAALS9bO~Sg<0cl?T;P5dW$wT{C0o!
z%bMCLX}LWp)f*9cntgnRKh6MpEjnwYJYiER|CDKEsMJ?ltz0#+tBI`(tQmSrw98L~
zctuUFQ8k=js<xcxpG{xpV>jl#Y7!A$9<@r?5f4;oO3kb72BmK^efrq23}=oqBcGJ!
zefFay)7t4g3E^x^Am%mr3kk&*fV9U^d3@iC$M7gGhVKa+!>4N)o?8vR8NPt%?Ioem
zgt(Ws4$-ZCLX*Z!kEmI212T7Oe9{&O)dGN%9;r}@>8AlB`H+ui?1eTYSqO~H%NEUo
zZ%1-BHq>Hzil2CbWV?`$Uci*&MpU7>LEGWW&oEk&18&CrVOduh|9#M?;<f>>QZxg*
zNEEnf&wGuX*)i^;id(|>&~#CjyLRMDQ`)Kf(NMNLB^fQ-g8qL;B#jp?Ovvu@{0Z-$
zF(bHtFgzes;n?SN+RH@DVaLgK#l`+6{;_rfe{V&zJ*Wly35Aj5<%WqbF5<M*of}VI
zd@xhSi~ZKz`EK+9q*TJQN0nT%FEEHjS$@Q|-z^87%tNU>ZQmIFR~E}EW!k0xPkYxH
z)>O8xl@?||ilYP}nh-`hp-LB|1PfKiK`?ZrP3Vy(B3;mcpdwPmh9E_nKmsVBln{b+
zX`yQ<N^jE7%G{&pj5D6&{de#4OnxOh``PPTYk#YK@B6*$BftXKXHex+yi!4)1)UIo
zdJq7)Gn1{9US~O7!74oS0_<;8>LXRXX>PcHcY`ty6re#MKBKGdr<*xE8c9J1_JTM|
zdSN;zIPKWu@yN`H#mtly3zuZ#ME<0CZwST<?+3N?voZH6MAKf8DFbQ===|kWyozjD
zoLREa@*D^si{h|hNtl~tDxc{ul`89zY#~M97<Zzm1qT&rRtnf~3Tfm1(Pnk(3cTyb
z559L2bD61T&gm71gxyl@D6@cGVNrxGEUoMMzE73}d*K5Vi>x$aK=(u}YNe4{;*{Uv
zsh6CD33uHTFMVJs+}RNIYkw^mZ&QfYau}f9@MzOM)=|F)M9ISP*dUmPoi|yonM^*a
z)8BJ8kg!WN<;wb`Tz%81Y4XAeRqU$at<hGL1tFP3uS~5$vb5A3JfT|_ns**~Z+O=(
z-tm(<nr9t`*58($fA-!8=yIz2d*Z{^^h!Q<gB78imMR_bT+F;g&3_a`h*Ikd64s@U
z7ILF^n0=3?HiHE5B(~d4D8)Bx$C*w;YiA9tDan8WN_(|BsB;eX?QkvV=tWE}B%1kw
zFHrcz{t$2+nIt`{L5t9q!~f&WY*xs?9V?oNOVJ)hsx<Cybwqp$Cta-~`qLIS5NFtS
z0U^FOGr(I*8^N!QUZ9haN<5NV>X}^t19N?p$iTXB{}OkU3&2>CFvjG2;CLdCiPjq9
z;Q_p7&G9B&GooVdeSy*LkebhQ<Q6W^=WIZ3SdSR-z~{7Bn$fF0_dJ14T&Y1*-j==F
zZfWfcr`cK2Bf(h@BnbRu7l6ZnV=P-G#Uf7d7XBCt=_|stP^f4-v-XmE^vGFJj6z@5
zm=dn>@Le6VfiB0UVla)WF!}z^yw5|}@wEkGzJCbMsH?ANax(}HgFM$Jt~V<}xj;Cj
z&1XWKHR=){Jm9v2`)DT+2AI3SD}&VD)PXlcbf)Q>7s~DUJbk4Tn=0+#%#6jgLs=j<
z;9_qBU|yF#)YQq74!7@ZV1LRE2GmM%gl)zi4KP88HYVg@x}sxuk@SR5n>WS0P&*Aq
zi=eY$pv>N4)Ko_5tABKY@(Lz0_5*HGZby|9f2V<&9{}Hr!N4xmhjs_kRMXLw9zhI{
zYm_Z3Pa~phi!OK@etW<G5C$_H!ayI%=K2iR-P?ulHhoj06;~nba$U)P;oV2AGDF(T
z<oR$TRy8jByuy0EUOI(gV}_+*@`^+u!r0aZhP9Pou**2iEWuFd7u%)jTXW#mi%suQ
z0fgWiuqMkejW$|+OEh5n7(WISXUNyz`vD<A!5vdaZCDTLV$-dqZ6I-UwTpOD@OXXk
zf)+!fbzqallml2xV&_C?H{}bgb<*ZIoAf+<?^ea&CeQc2plX;-g;iSnbDtLV4)uF8
z5a*@Butryb3F`nDS1LGOVy+wn2(=`FMV0}Ch26_eNTGfNM{u!Wai$75#k!b)4?Q7*
z%l=@6i9U!3KJV3dxEPQus~WQpZ!y-ifJjH8K;MI^{fz$Tvp`pShf8D#FZD99ZSaI|
zvh$N4Mi4C$E9thE8mES#KrUm#cZT+yH^b`2K#5Ph9hKlInUJ1k<U5-LXbvhm8emDJ
zuXiYWiF3UjXe7b3&iFD9#4GSX4g@TiJhDD9bO1<>B$V|igM(I{dF=`=RF8o;<L6ko
zT~!SGi{_Cf^ZO!Q`EUh1$r`gW+9`>+VAf=%sFRDYG@k*X{EDf9tJfmVF%6{kDQtzR
zYo*O|=3JX*(t2#=nDxj2aeWAY5VF2CeA7z+4vd~>E0kbR@GVP8yS&gSAd3~IuU4`H
z<vkCQ)~ip6%*wI8Sw!lGTe6aX@X*HaN!6P3)`b>rZ$Xy2IiS2his9+s>*-Ws>VI^p
zRa>qd#X`e|6DU$MS1V4EP6S!Q9$O~ylh}A-@UvER@&N&<>kbDOIj6;z?Q>V?5^phM
z7hmQEWmbY<N+raaoA-v%(dDyYg0qabHwS)CdGHLpdcc81J?i}~&lDbo=4O<ANyiO6
zt;le@qtl`5y2cs+^9lFXChHFNM}VFwS1&xpVKWBwY>w2`r~3G}`~RmAZbFIyj<ng)
zJo~R*E=chyyhTrbqv7AuO!_E!Q@y}XurEKscg*D-Q6_R_){dxACu5JBQEQ`5*KAR6
zzC@|=WXu6s-xNMir7{4#=<v-bw3@^0QP<^sha6Goe|!8IXYsS`cq;A&G35+S+-YUf
zD5&qU%v*DLr<b4H0ShKt?eZA^kTi7W$V?u#0qsPaNobbrOu|^<)d(0{ocNLXO@Lah
z(ne)*HW5JIeW)1CWA=EdZJ<d18qpiM7~vtPfjTd}FpMA2JxDN?@cYC-sHuTAcDTmm
zRd7ESzUPO$F%Wv=cjy@a?I~<ce4ag?;>tci8{|lu9wh96`WZ_==SC?V@f~w<dm-8v
zi+K{QfL<fzurW4UJbr@V!$2=5u?bVqnY3qVj|>O)7Z9VWe<yb1FVu81E7L|}oZMql
z#AgVHwQ;-ko~O~#V$i!)aL`<u?X9eH3nRAsa3f(vv=9z^l7_7A;ei`yJZv%6g-+`B
zgAE{3uCHI^L#`d_S(N&_nH#QjkuB&t&uwQ|1?hFv_^4XHSCeFO-bNB0#n-)ojjZgr
zF*$~cD+1)3BsMGsZR<H$=b$@Cok}Kt7;abd?$8*n022<x5xSX(@B9dKDP^D!bxzwD
zsXla12O@TD_11>zHk47d>8QeBY2&R&Z@gM9sa_A-jxUVRY2}>;+iG@Ec#*|SN(6D`
z5qwLee*2<8jB>OTML@J*_SMov*d!IqYy5$8Kw`#q={3Mi<Snk}VTR<Ptx0U*6tuDo
zM=7=+fPi_fVA1Faw-Z|26X<d(LZy!hM+JdWm-sq#E)6+10bknhYv%%%NPRR=GVgR*
zuFJ0-S&w3G<%i^Q_91e4v3}_`sWyh=J?Pgm?&UpH^JBc-e%+)paT?6G7dFO>i@+x}
zOOopd@rfM?7VNx>&+$8yx_%&)#e{1!yS{g918BL~fz0xH0$oqHP3Y-4n%AH(ojSf4
zD@3V_lsQDqs7=iGgXHX;N)Ae_;GmPFJuNOo&e!Dq0uugP;2>>4vAzjp{{~g3D95d$
zpxpHv0yy~W)y}8*ZCg|Zb^C#4r*@%wBG$jm3;RUZVh%OMv(JuWIaZv4zF%Ll%e96%
zY+hoWaTabr0;H2;SJAW|LovBGR^<jU{|Zt3#Wuv6Ku@?ObY+o0i@z{4NL()OlBoa-
z$>=&`p3;FLONjzOFfb)Vc!q~acx48YMu3vesmMaK=|sW{+w_ZOm1wSbPB*is_94A_
z*-02`A)T_!6Kuxru$#w*J5+{gm!3qS4lBvjJYTh-VEUQ0_Sq$njY~C^K3*A8aCMxB
z9p=IaAdl6)b@L=(5}vA2V^468AdsRuT<>b~oDVQW0DM`BCud-<Re6RpJ~5<nb)QHi
zhJfXFz5hYbo?RfR;?o|@7);3ZjwsWCh0{?GF7MOuD1)X@%%L9zGB7at+TE;sxWWbM
zb(F5Qb`AVg$|~N}KGme$r)FFHGeK!Mn2{Zjn~S5@HEV<Y@I?r+*7NI1*%rr+=?#?I
z|BTafZ4(le^Ea}ElTY*j5;fmLg3l!JJ(+72Q9#Cr^w;H&sf{sju?X~lUEp4%W}-2C
zNrpV;)}oztPU!k!&VeP)zBmk=ZJxt94#K3mB+L-%REFBGh1QXEzN{${Nw_V;`T+In
z);^YpL7iqCWN!e<geQvKdD`Iv()-v$+|)TjHR|=J9%l~o471l@EKMT_(Q8+O{n#`w
zs!OxT!ry6El*Y7nN;>PFt<|8PV4<oz-jRc=x}l)S7yZi;+q&`=BF4MG&hm*1t5tu5
zm~&^_@y70B-H=iLG9*mPm1Z=E;YyuHy*B+|q8q~rKhxh_Cf)~gm4YXVN8M(a6#P*?
zWR<brPp56Inztoj&lp`&&e83K)KvWHfx%3!z{O|U8c6B-IKzs0sdS&rt#TTiX3i86
z4NYU%O1UhTUcv@o-dS6>n+j9$#JEd#<__$R+vkDWdCa^>7p7wUm<tI;4iJZPBDi#%
zqWw};gd~~PJAVq{JgEQ(4DCn(F*+sL&hYrz?XLVQM`(B2Nh+fMiEpNz3eYd@N`^au
z|D2&FBzN;%Gpp5AA~7h~mi&ykP{nJigr>%EO7zg`A|(12rQlpI_Dc?Xrqx;J!iuV+
zhw*Vb0Vn0Tg*PTO?w#D`=6)5jVDlgvZKH%xBi+sWMPZ#>l*mINF-ma3=gfH#4=I=h
zv5I_B1#hGs;mi|508k0Ok69=t84*n<BANX&*`t0_nMdp)hj^>kLONyUom)%|$!y>d
z>#3dv{aT324^gm*U*dPqBazZf7vC}Jq0^Y4q8d@xxC~lRHq}QTH(hxt<#mV+Kg}@B
zKO*w#4%p3Bb{1Jx1mvV<9<qt`i1m{8k$x-9W1RjjMa5Z|+^kWNzr{%NpV!OgY!8R<
zH2R^mCin5;vovyLkn)l&&kQE#v%$Cw9Ue+ChIJDmr(O2;Yz9paw4D0Yf)_IE{py&2
zSQJdjgQimpQ14=zR><Jh#bn5y^3W&9(9tv<=zV$a>j)1n-ihwnYEhX*qqNzqjX<7w
zTH%Y9S^w_A4R*`7<sW&U&i6Wu=gKE-VCwzrpPu{;cPUk|40q5zqBOVkE^R^nO4_rr
z$(@~`APtH#_fSVmBc((wtUY(hPMju=CeFxQ^^6%zxOBm-UShSLl28565vNE<Sk&i@
z?pPG-t{q-%7+x-JX}UmJ^q2N>Z#nwjNN`pVr`*syeod`xVZ&LT;uuZsm-FHzdwbmp
z7*2n4C*AQ7$JpBKwcDx6gJ{7;`J7dYm6tp3*GXH5bVl36mfb|*j3>(OQz?#G+ZW8w
z#&#+$lg2|}a*F~bBp7L<rC7P4g|xU<+46D=8nk&wIo{Fi&E5LdO5w0wyBU~Rxe&*K
z|NV=_zE-84x@@w}=c~qN5aMDP^CJDxju}E>)fPOT=0>!G?vUki_7VLqNBZjKY0W7K
zc@xD9*wQy!DKgP^=01aB%LGczhd)34_bd0s8TcR-XHF`ADf|@Mg^03<h=}ckG=6^m
zD%U(SzT=-%MGHzwy6$NN4JO>Z+rP51a>r!0Qe3lG3uBT$(A(>rZ&d{%hnEJTUn|vn
zSC*T{eo*>k@fS6smkH82W5O(+K8AfUDTB$!*SEjIclL_gk)*s6#qGmOOBYkp)7`OH
z>~tL-nxP*)olreEXc?4I<JfYqq^_<`^`Ki&P*q-j{%V<wQJnmiz22<LYc)m1Zp)XK
zdxso0AI1So&<#}|OHf^7MY)#3e{9(=Mek>l=P=1CivDsT?I*vT?ZAMImFX7U$4vTi
zXLR$su{k+8K`*AAcM#i1kJk(h4Gq(a<1N#G!{_Sa^6~0Wb5+{*hwl98uAFm{RuDm=
zgM&j~d;6J`v^3Y%nF0RFi!<FeurVs5)qJEycEf(bjU(S`8fVZnX{daUFXmY6S5i^=
z4X$xzvsc1TVf@{@3jsivq8GlQxSK%{5%*-!JfG+Wv9q;(Shckf2H|AUD4+WveZzC)
zhsqP5z2@tx$M0=bGR>yG+WVEeeJS*-FUP!rRb7t3n_pf3cUS$xU&g17jafhSV$=JU
z1z$UQ7c+PZ>AM$x`d*7ffkhc)Nr}JeHGeSWpLIVA1yv4YAsKzIMH#>^62!_1{_0_0
zTTj2o*aB2}`2a%WdoB9E_!8XEW5XjOHmd6C&n(&2e)y|;e(g)(p8JW%Ko{YJtHYe9
zFl|bN^76iym)B(5hYfYVx3!tv&JP1_iYpF>d+a`DBF|xHU{Dek7Iu@xSN0oS<<EwA
zbTAmDtGlMF2l7q!w<;-}aT6B&^IhM#QpH48;Ba$yFLNJyxpiyq;P<{pE)j&Zy857r
ziHUCVl1Z-8r>%`uA1|-}ZSC)Oe<;~losC^`rCzVExhdrH{na)$1)euDiZ}e;#$E=l
z{WgbK``xbncLD#+2mgKt|FysT-9-N{n`oO+%E?UO*6OdyyTIop+EC}2mUZYq07Vp)
A{{R30

literal 0
HcmV?d00001

diff --git a/_static/IT_sample3.png b/_static/IT_sample3.png
new file mode 100644
index 0000000000000000000000000000000000000000..b8c7b8dbaa0f51df0beff2e18510530baae14aa8
GIT binary patch
literal 396642
zcmeFZbyOTr(*_EG1PH<12?SqQ5`t?2f#4Q`d$1sjv*-qg;I_DXaCZo90fM``!(xly
z&HH}xJ6F#6zWjI3y?bVMW@oy4rn{=2s;8<USV=(&>pA&zBqSs(>5mdBNJtoWNJuDI
z=+6*WfZzS#NJ!6%EX2i?q{YQ)mFz*L7FH%mNFRga;?a~;HA%i5y8~1Gpt#>-e<#zX
zBB{JL&y+=@c!_~46Xcy?Bt@)l%N`W;jc(^hkdGzumy#eOL*?>mtjZwfeWUh(fyZwz
z>u*OMAJ5VgxwhT6Q{9gmT#h^)sF4UwN;OiD*-#=`2BIY3?p$kY1f+1}Z%Eu3Pd*~`
zZkmK>uCBHt#j)JFxI)mx)-UrW!KDz$qy7hfVBZ!pl7Dte*T|MhmIvh|iBQEWG^8)Q
zd%QX+^witVu@?B%VzI|z4g;@n!W_m>jeknWeL|%$AxA=e6VE1%{Oy<6eW;G!^Q6E7
z|H`l+AYmpza+aQpD<lGj7d!NubV7=HWJK<ULvdi)p0VG-zJbL)O=r*#*TZji?Zwg9
zql4f>-51l3`CFXt;!;>Kg2YN#8_ll9s(dy8GS9J+jb1|NynIx1kw;a34Vp5^y-wdJ
zctw+pM=sRe|6Vj)ND-Uq>-VR#tx>9g{_l|^aR9@ryacwyH1#BE_N)Boe6Ty$Lgfn8
z6t+-|)U|{2zKmVc+67>gWZynX&|ORZ_C-vYFxPF<#dCgG1%IJ3GOWfSA8|`#oWE*)
zqvrIfNbC6PQ5e`Ryv~EePO@G|FLD)lWwtFj(7Tm^X76TI?}0O3@4c$}Vulv;O9m_Q
z3*1nVYothejcBe`TxTe_P$D5ZsnZ~nSMKLZ5O|7<%$dje5-D_fx_|uyDq@H8@@CRv
zSqmTOD668PLUp_~V+4tE8tvPw<jcfMq>P}n+d*xxiu6$qXpO^g6!Z;016e);^MyBA
z`4i`#81=}&70f5^NQ0i78X&KDe~JwL+)4^SvlM&D{4BYhM+x~yJGCtm<uj>wTzpT5
zePvh3>yd4I1I3=*w99h7Sn^dy_c_6Rn)N;@lKRUF&G*i@1h8)&nFh>Hz6auG5`ILL
z&fNU1RY1z>&6+`8fPM2+03$W9RBZ9LX8{Sh-;!9mAu40%TMaZGZ?kr}gr{ij$|oLt
zPxVl@+n=6D)>D@H<Fw14xWdUjD4^}YRXkMM=jc?UQpLE89g-c9sISPPXxXCZa$XdE
zAdmc<gVGS(7x{97;8|d%5h*EKC<P-DLD;GxvEgSU9s^cRG*0;h=U?)2KvU8)EZO$2
z#_ScP8rVG8$v(QF>|Id9D?>#Cpb^OUw&K{1kCMvlyF<^@l{GtydXsvr`qX;K1%4gg
zQXF2N-gWL%DA>h=G%aMLon*~r1!r|+&Hdya9)bGQhq^s{Md~AM5c;*>n~rxKWbbhd
zzj4ZZVk9Tf4K|bzv173#ume`afm3G`DH#eQ<q&dVRw1{*DomLWYy9XY;Z~H9VUV$s
z#VPf!pj<0cXEmt9#9z-}@rzpJ(?t0+DQGlr^i|AA)N#zR3Ns<WcM=BgXz|&Ph1oh8
zLYW?)a#*$4><O>G7ZI`%B@?<6>a*a*g~n;fV8+5?2V%onbk)gK7+I5<tKu34*y7mY
z6j_(msFgE+%>3{x{PM#}iNC=9!)61k-lW8&dR&=SMx)e)&sQ3vOmdakPt;EAPQVMg
zn)HC|#4*0=w_2|cIZHx#it~PKWe{iAN|dM^Rwspq|CZmObbIDWaL;^?eP7iWE+DI$
z(W=+-u0`tJx90^G5X)OKN;130I&38Ag}nZd;}Cn&P*Ow<rxvl6Lp8DG;aFn<_gi0{
zPU~kN$<fekE^716IpR6IxxwlybC<6XJ3D*u{U9gj9i87~W4ZfoyLr3Ny|huPiR?Ys
z9gGW^i{ZT%gM!x@p)Vx)vr}IemzB_D04K{vkH=`ZIk|&)`gqOXf~`Kj`9jV}0VS^`
z!=~6J^CxBJ{lV*IAotdkJf9rId&K!*VDuH=vT`_QcB7#%fTj2CNcGZr6j}GLajPiJ
zl38|#6#j9cMUVVzEX<jXQM$>jv9(49tKBaw)e9E0!UTfjCVA$jhrgOWe_?~Fm8!L<
zm2oc$%m^$u&fZJ{vNbWPYJcq|TnzP$GIpAl4T)>3&5O-@%zvKebtZNeYLaQvz9e)P
zy;ZwpzbvByP@!TsQ6BNDP>EAc^VjlwIoUf$^KtWG2~hAe38?cgn~Wu_{=)h-$e}?3
zidW;v*AZzDXxOT!cDf%N8em^M%$Pr$&8vD^bv1nCz-znRI9)#|L`oe@-6Z@a?IulU
zL~A5pFF?<u%u^2q;P9DzA9p{lF08KW*6#|fmd*X?Z3@oX`myT9_-4@_#i91<ZP&<8
zV0YOPXl8j7lK9wwWF@Bb6Z)-*k_tOQwU-Rr5nBguhFSvq`->mwZsb;Y01O)}Z}BJM
zwE;rUxv`1-c>=vT6=tT&dsp*TLpmU-zR3h4`S!zA=<`9FcvMjneg08s^vhDNQr6$y
zNUe$4!)!v7zbBHYlAOJqVyb=hnV6T&=EFh_xCCtIA^Sw+C8ME!N1x7e!~E9eh(n4x
z-|?`=__geozGg-Zt#SLVzYJyrW(0!~fiVFNqt|MHon-^PbV|;T97F2ly40)a<8I(4
z&hv$*3_jjpt~R1Kwpor?v{j#}3abvP1o~MS`ky#yiD@~_1)i6m-|rfeRgi@x`u*zq
z6+@?{+*lwi%WEn9xo$>9hZ)LluUl<BTK?)|C6HnuEqyUsT%00Xp)m0r%>q;^Wn*&G
z+x<Euf?ZjYPHU`T!f5v<7sb`fRTS6>SZBA?E^YG|f-Mmh6LGU;#!0iye<mws)tGFX
zgIsuX=Wy>@dJWd~1N$L$JkHjZ-U{t|KCLu&kOJT*KrG>s?sjeIjnuD*4w|N^`LgAT
z2W`Bzov#)0cABU0Sj|}KkjneQyZnl<eu^d@e(Sj!vO3M`>tW&%b~Kq!GBBB&+5jtS
zR+CubxYqev+xfURft7oQIfrA1m9P~UluA;N=NPUMrDfMN)>K&`)AZGIGBEWnRk*3y
znfufJClYnY68nN@Sx;MD;Z#no6^@PKQJR^%@ul<Kue%1D^r_Z0PyMTq%Ry`jO7k*@
z(jvhEPg_-C9bgHK(A(qRVVxp@+T_LLU&(jFx}|>N?M9Kby?U_UHvc_ZYy8U?t2Zn{
zJvAeZ`KwP7ln0A9pJz~OO}kO?S(=rllcn%!(M8#<;%@p6p%Q1eak8;QA1-6Fv*GX*
zf^-Ujbo<saYs2|YOgj>rnh}fRMpDO97(<&ZiGT)`8L&xT#HssYpugx>(a>~CDYuh=
za~gc;w2nYmL05h{wXANzwqbNmxan3d!ffO^S%eq3xVF&fEO_pGmAWH%bu`?>*3{c}
zb?4P{KaFEg6Dp!}kA7RQQvqAAWSUd_3gLZhQQmqbxVS$)fBhD=+bN>AJSbZE2p`iw
z3569~iGebX8Wzt~eZ72n^?rGdKC0Z=I~JTM#)%f`CAJ7W#82y|cG`8`mob#}E?I$#
zuALtR16{1?SLqd(%vx#gmClvV*1Z*fDN>~yi{gZGJor4SJWSqM?$>32B6{BR_H^lB
zxMUHDDf!35AcL!DkYt3ACMe7wIbAB#({RmUGzLCDA#43<?(b8O%*BSl+Fc@OKk+`v
zf6i`DrUC_LH6vMpfYXIbE?J5+B$PM3mGfpPN1!lzZ%E%4oJZZo$7%ziA{2Q3J;}ZL
zvu}-4Pt+0K>vt0kX;XQ5Bv!;ZIua@}Inq<a88TuQL8kbxb4g@oB$R($e}aS*Xn}<K
zA7vB}$3LG)#QvwwzmF)<0Z3?wU-*a}oblv8OJm$+p#0}KN*3ZC(mPdgX=%h!)!5#|
z#KyrK<hY^;Z$@0eu>Gj%fP_TM@MlMsR(W-TXn)$`lZK;)yqthB$eP3O3&_ZX18i;k
zryV3AumIxJ+QiY27Hn;0;~)SQru$b30mS*A%RoBXe-&}G6sFUVSE3aM*_+VvaBy;P
z(uq8$rKJ_J|6(ejA|d%7)e*mh>C7D+Z3Tcp7Z(=}7cLHvy&3Q|KR-W^^9}IL8+Jqq
zb_Z7*M?)~XjRXC^8~M+6BupHP?JaB_EkHK3f7&%P0y#Mf)6xBT(0_gY{hlUZi~oL-
zjl+L*3(-N~pF6<U9Gt-aY8z2i=+9LFB@3{Lm8OJ+HDb&V&k*6~;uQK<`TxtE|9;{>
zYHIv<%{ScKJpZivk6Zues%j1<_TnIG#4{a5{##!EQTd-Y|D&Q1@K4wOF%<tE=YL&A
z47A8|A>e-rP2~Aja{3E|94Rd%6h9%32rc{bK^{eXG5`CBI7bnOu(SjnAR&n%NlU!@
z1V%ncMOz^3zV4+d&qQa|pd&tL@uqT*(5+x}9-V1=UE8(9{q+D#_MqZwiO8KqkQEQ_
z6YhtfPj1tXH$;1vmX=yv>fC`>u!n$=!6WO$)N6%VmNw5k@r{)J;BV+bf3>~k_QuA)
z<951Xwxy$e@;BQG#>bQV(7&ro`^5U&H&lo<Z4{vd9xn3VY>_L9u41SXf3?jaA!9tu
ziP5nH{Y}UJbh$kp_E*D>5LPwv6a2dxZwVFL4}bOY5@J50n}0QAh=DdoL45_cL<!EJ
z{r;Elj|v$;`O8uJ{S5u%$@DXP>34*G`TiIWv1orMa)}Hq+{l$OtXI)K%zvfsv<RuO
z;QZZW3lgKCrES-z`1tCn@?Wa#E4V84-%S);Zyqr*pEjPzkI%FI5-boCUFiRx=>J>U
z|9^U-f5E@gb=%G5G5wXmgDd~nY7e95?0msB^Kg5do*t+MkCi@|ZgK%pyf%^`iWp`1
zOB~|d#N8Rq4s1Ca7Azspl=Epz6Ly~~H|*5n{^^$Rms3Q}=1JACfMpTIufbRFdH3TM
z>xly8nB?R%;9sImiA;v&bgA>CntIJ17Z=9EGaPCw0!BIWNu+(lzwLJeh8*|gC!WW<
zv%0UGZ!jLTxO4ux!_~%r%-DLD`o!Ovu{rENHdsN6CFk#YbC`f|AP)IXiJbn%d%*Rs
z6-R(#ZOrHAg?}Y+F+oNIq6;c`e9-+XM9(<<V~!4U;@-3Vm4f^ZN4Q_p;RO21B!6W~
z%?1BB0)iH~k$>k1{J&rPwCe54GZn^RZCW}J)WXH$f!H+*S(eE`KP}5zOv_qte#ocr
zjZW#gTWfoqb|*7Cswyabs?AHOw)!F$t4J=b?iKRC@}PchFhv%v9M4&n)gYrhn;<M&
zj}Tp&(sfx(IvVXBdbbiTcvL>FAUM=|dn6c=)#rA3V6mMJX)S4n4GiJ^WML*`)mBw9
z;^AZ+V8ZfaEBoI~x(p2T4OY1O^=AB`HyU+$-Lls_wcF0i+rzqLmF$^ZS(17CL3Zm^
zf0A0S`y1{Wwfg*Yuen~slA0-*u)$cba+SN=%R@^PG;9FvPnGpy!DFjwT^HTe1ZUXG
zjLJBiUY5mpme=Xkl1XYmwzbx;-2(o1bLx}qA7Q{x*R9WYMtQzYFbf_xpB;Q>EB%Po
zyPKbGGjG#W;<P_C)aZQp6){HEdrmR8v>n~L&a>T6#FsORQ4t|z%q=g`v{K*eI+r;S
zs_*66)&&}Uci$%`X2QnR;k?=&3suyjA7!0%C^@~`;9{IP<{7>J!=EM;3|NzU1m-@~
zEEfp=Po2vQEWAFo6OssC;s>53@H#xR<6))U*Sht3i|x*)CpF#croVRHM$jF;8h9Vw
z86o-r=6Y6TX!w=o0fFJ=dM$3Vzca-{I?5+MYtN>-ENMDg$vk>94BD;t5M}YNhjMB4
zhErj-oh>QP4kYiy?&@ogFqK5HvLjv|FJObqE4XKd@AjpuMjJv6e$aCIUX)u1%ISB<
zHA?(X!Dx=dI{@Qh$!(u~f3eAhm8?rSRlrF`c=`oeHghF?gwPmA;|^V?(NyDpspX;c
ziMsV2uoZ43xL1&E^(*1E$m~{<B}db00EN{M_q0{O7pof11<2#Q^HIY#aawP5&)F|7
z#Z{LjK~`;7g48$nmV?(h+$uS9Zl&BE_>i7@^Xhvm`t`tSgmF}2@aEu2mY>dah*RHx
z^FPIM^JG+VMmWOEURf39LS=G&K2#UxrwLr>a8Wkj!3Ga!j3a7}+a8zD@>?(-euO>t
zMbTTY{PezTN24MAeE1OrJx{f1+)a8(rp^9BUC%Z5eiH~`5yGJk;b=7fe|+x%REQn<
z9EtZFg2t#*0Cm~_XUP4V2LaXbAM5tZ+Lj(pBOo*J`d*rR<7udnnx4n2Wk?+j<Zcw9
zyX5<0xw77x?sHKqWsMW1+#t2>qEc2VgqgDc@+*lSD08ANvgnv+@}x5WyDn2aKx?GI
z9kD<BhGD7B5aHj+8)D31XK^|1C{eum5;P=ywGfbN#gB*Uy+xPtMu0`Q%LpIyFj$Np
zz4?4hc4<CNU9SOeX;ATBL+(wATw}@D{`=PKPnmxLo0#@1&e^AsbPrf?%k5E9;ep%v
z&T4#McQoZ-UMs?bTH4dQI$E@TxZbWSsau+RI0=Li%mrOigIADm&&L(9L<L-qtj>mb
z(=T?%drecFK8qLBAm;P+a@!p&1flw_4>O3F6Vl<k*+`(kmuY(t0lCYzyH`~dR^uaH
zPV~CjwayOXsSGAynlp|NHdPe4&Iu!Z-685Q$X-|7>UoO*)RrwK_}Z2Ky_&l2q-naR
zGlj5QweV>d!K!V)bIZk4RjF<hloAK6{&ChUy@q!Q+$jn<n5htX;V{gfj~DoXSo4+6
z-ny#womrv@F`|pd&#!u1+%EP^5Kqv#QyP7MJl^UX7b=hWwP6r2&b-HD-{^ebcz!Tj
zwd3aX_Dw{rP;ofQ5CCpT*dz%pAu~<LWF`tiG1)=L`*Jg#A;&Dqf-yb5A4p+U?j4mM
zQ>ttiA({_;L)->DXuH3hzx*wZ<Ldo-Fpz6j6ajnWPGX<gq9&g|z_!j;18L?m-ub>K
z45`f_At3Ea7D&B88s)UpGw^<Dg#hPD{;h=Uoiw5S0j>S(FM5FI8opi#uYk7vuxp+I
zZNXGs*RkcW;#i^fOi!>nMnmy?9x5b8!@SHImz2}sn=W|OzjoQ{c2>{*aFXXqiPCp5
z!GCE=MfZPdN^$+#cQ22Fi)nq3-xaLt9qJO~(X$CL_n)_p=NUI^DWJ}wc<V#%Cly5R
z9d8j(?cu1dX+k8bK2E$=>O7S5X$+xIA+?PzySedmX6c@v2|apdco!YVvy5+E{$uuV
zn&2lK__QU1T4xo6PP;xewzyx_W`_$5Agm%Ekd}=5Yb;4##TS5VADza!9;%~s;O{UK
z*NP<XrvgXsn!Q(0+70p*Qkx?~y%S4n=LU9mPE!m!0~lIhSB9?id=lWG@Y-2Z3S-x`
zoX$Bax;TV#O&&F$DI7ByQKtoqPV2dsEP7roIwJ)B3Utx-cr)Fm_4sg)wfb^ItRQ~t
z;d%rz*Fyo)v1k2G2h3}LuZ6TkqF1c{;u<+9yS<zTd3xY3Ss}1L?f8?byqfAA9=XT^
ze4Sxr6UnECYnU&4gVL`4sB!-wFWClW^zIo7F54SEPMZb3{i%{<5`JpO-|v<pu&D$H
z@610@vk1|yD{x^xkhhnPny>|3;f_;5?oNB+1C~9`!lIaPJLuAKInYg#nwy6O_Gd<=
zXsm;*eJv#xe3vg`D{O(<T_GCyAMOT3!r#a&6wKqksggVAXgV;89dXmC@o8)3SI$p!
z-AXKcXWD-(&okRKEPSPNyH`}^u^7Oa#Ap06*4u9Hkb6oa+g8+Qzj^PEVnvc*dRzFX
z%J{Ukr+XMiEL{0TJ+d|^q~=vo-Ga=<8?37u1wpOTQyL)dWRt0*A@Z)znCfYOVbOaP
z!Hb!y!-5#>s?G?(jLEg%0Gt8w6Djx29|&NpJ_cV|A;j{q?gthEHJ?g8+C1-tXXNO&
zd9CKx(kT%kGxmk^Z;=XKZ^U?t=KM%@j6x3*E5a+N3muD)LhZOUwr^<eR04+d6G4R4
zQCac&nmj(aeXiLiU0(6k5D6-p;uaT5dii1!_QO%P<#feZswb=U@xfixW=Lu~C~ZT6
z7WgD|_BD3Is*kBd=q`=hPNvs`&*_lp<MoJFbJ=W1B5|7AJoiu}^~IFB1ts6-k=y(f
z{lio6(ylU2CxYx1^>Jd*;2bm^x0-D0?7IdW;r+uCH`jX-qk9s#gMy4lCDFxU*So>;
zq`?~njhInyeH*W%xOVQV-a>%Et9lZBk5O#C6(nHZJJO2Bhg)~$<?3;T{d9jNY&^=N
zDQ!^AVeR}drVQ?O3v7Vx!{D_&78OmG9iN0d`fdV}7Xo(=V^tL@Nc6z%>6njr9DXty
z2Vw^Cj_8;!3wBc3G2k(2CI&?k#(P9jvp*hS%f};xyg3@Mk*Rz#>gikkFbt`nO{AP%
z_-ii%Dc7zl4g%r0f#C19W5{Ak2vj`9Y^U+n3UXQtKfx`nUw1{nrgq)H1C54o9wA<|
zc0}YBy!!k_xgWxpmJ-bj@5754nNU?cY_X7_4h{tFdm(fgW;#u7g1JiKg9+PgxmvOm
zy*u$8+6Vk>CyQdvS{&qP`n7!;>HNauu=#9Q&wOgfLls1ZX=pNwzV~VX*Mu^2c{5)B
zsO0t~Eye|=U0$bxYeBCX8+=eOg0`~fX4f!~>T`jaIK1LlkiQZspmxUSvYSMoP*;48
zy%oOZJo<rT7SZ7&P^XcBo|6gt1?4U$Ka{2H72JK(fBK*+d!!PhLOR1xyzNx>$H*J)
zh;Orvj)z9-%WhFlhf^to#S!o=?Rm$r!-E4^g<9^^oiapvQd0-gJ>(2~Z8n=Jbqu<t
zYBg&TPYztc!R$e(Lz8mg8*lG|6du+X5<97RJj>knJ%GNVDyT15j2*SKe(m{kn&i0)
z!Z)0@FDfNML}i~oxqvE*9IT@JM5FO2Cwiaw5cDtAKL5R2fG(%^#5F!b$jG<beJ-vu
z5T|Ue$uvlc%P;e*0TD+rFLkJHg&T%ZK7vY2M(_t#gg%cLXTdA@jzl9GlU}~$zf(3#
zxt{$?n6tKD3i}Rks&-}1V*0JD+l01$oNT<?lIZ5$*NEOTg$Os&N*@IeZjKl9<vvJf
z&+Y%o%T-0Z`#In!uYEWfRgMupA>A7KIquA;Sa&nub__JFKEPOG7Zdw6nGdf(nnt+s
zduIg5oYD!Cw*eS}XI`u-;ojL%78z<xH*Kf6`>1;j;@6-n*anT^(p7#q2p~#8D8)QN
zdtW|Fv96tM{u$1Bl54i`RGIQg@9)Cz>55v{KLvh9%W{ylbsgtpJd{%hsdaIF3@SKA
zSe(m2Uwplea9Sq5V&=*XP|2y3U$=S@!lkl1832hdtfArYJ?!g0vL?Wp<Z(Z9G@o=l
zypi~V&THr|j@xwsxfxHt13e&AiBM?RFhB7r#I?$@q_P{oHOPwWZ3lb(ici1^n?{o{
z#GmN%3Ch~TlPemiU`1=gLFHh8I95OJ1@!fk57bfzA@$H3G^xaQCsMk-D*XW0L0bIq
z#y}Va$=h23wLA5a6`V2{hyDvUf-=FQ`s`OZfLqTwi~MVWV7ITlb;WKf2Sd>)6P*)P
z_5Z@$hLgMKk8e-J7}}S2l(P?Ga75WA;!uqP<uA*fc(uka1q`O&7vL&zOqvDzm^a#H
zNjtYv3UxvCZP}Z)E@iF8y&zW6O7dw28$(9$X?m-ZeEk`rvtbj5;c*9;#Md_>+&Jwt
zbZsunp3`-=<&rWKvB)}5oe06!gdY5TDH_iWQOBQqj#}F6UJ;icXr+<Z+T`%*Vfkfr
zOi$>3>!9wjXIJ~;K@gKwFoylTYhb1xaSYZsYQPgB^bK7RR-qqzsvbL`PHkFUMnkv}
zrs(36k!(5%+|<Q#?W8O;)zi@dL_RsHbSv!JuR%X&w5qgY%(}y|co)it&>s-GY0zIn
zHb!sh83)k1Xof_4Rc|l5=1WlnUWfh+EXKdve5dO5-=SDUWPJsd8OzS5lJMg5Ub^<P
zx{bu)HMRa4g=*XeU@4s$7OVYQLZAE($>BM@ekaK7{POb${!Hl&RAYb3g*UBH`;kP$
zRqN20uoT;V{dH8xhwYA2@C<_9E@oM-iKfJrc2uoh)idzZ*TstaJr4g+8|B_kb<t}N
z`IqlS>BZNszbYhX18dkIFLm%~%i0ghCpah4mGwyZU1pnlOIN-l<Z&WeueEmktm69}
zo!n8#)Joj4z-z+|0kRL%4e!dcI8Qz598E{LZN`M`jWX!kU<33Y(zZ@~g~%0eaYW#>
zX!?Wh%9gVo`9d=_rnB~gjl{2=WG^HY;Ge&(0n{b`P?k~d9Pg?hn)Qqa8+cbM>QisF
zioGRM1zk9v*AWn)9eXy|$}SNPH*L1lGcmKLl$qpDc|<cYb?gF&h}Q51(3~0x1SK~Z
zD370*u~csekn)e_w=v$uo9SxG3X`NQYt2^>K!RFR`Re!Dj;el99zweG)b#a3mctq^
zCl%9R7DtFw9_vEb%1h$O?TfGQh;%;2x~#q8bV(aBn0cl@@%M#ZEQSXGW%~_4wd9d_
zR9|qO?C}1-S%%mxgP67hg2W{iBO$DH4AD1g4um4fv213Ot&P4gEQp*?OCKP|VM;jl
z)M0`b-6S?Dp=!~sh`fMZ4<JIAjgFP+W+EMA>Iny{=3kJK7=$aQ)Wg0^P~Y41oEzB0
zR<I%{(0>_b@5^@?EPy-y*FjVsC+$l=C~n&uq(681$Y^{XWe8?(IUn;%tg*$KM+8P%
z4X$yD0E2)kx%Mac#7R6e1}L2a3kc6et3Ip@SOM@aysnwnYqCu-c?YUr4d?(lr>u+s
z+cX_d9w8?e2p_pr)BZV|-n5QG*yDQ=OuuP;NK4&$Z+x0xcd+`MFhKKHbdp_s^@&vT
zI6$6({VvsQuW;$OULWPzixd5GKjVJO@{?YB@%BfpwFp;R=uxO36Pyc>@ziC-8|`M3
zzY|oep)QvMH!C!R-eF<O+GrpP!usetGZ$bV+dEU7lNLK@Mt@~{AXLVvaCgIgTEp{|
z=+i*8z76|1KvQ4zqFubyf-j^W)_JNmkU)XT2|WxG+^ZMyydlI^7Ac35P|6yWSKF18
z5@SN(sZ30F#{I_#U)Fcgr_&8@iY)b8m2Kz^nq`=@6gA(fAHnm$vtl;O!tLkC=DDdd
z#1CUW<$YFzv+2$bgBj^gj&r*m6{Dhe<}S<jZStXvb2M(lHWbzst-`lmn3sDYpcy6_
zL+;;JM|uu(J=&!p&(A4e7^+s@8Eu~_wpHIAd&~rQJm}v#ce&V*t&&-X^KX~eOz97`
zR0+;Uh=y0FO=~smlm;GqE@=3`n?O=?noKDm!Q%#B_DT2CEK7JhtSi~7#&ESIw%K4|
z#?HA{{wj$e^hUKA=zP$YT8A{y&`S*-)1!jth0*XX5$jpKocEe8S%CHE-q-vg+fqmF
z&A%mQz=K#;m;9z|LTlTvjZ>Gez?dzR2Q!)3hmDG-Dy9A*j}WsHCjA-hvKC{S=Bo;<
zhYg41%(E9f(?Skf$3)Odr-%By#odm^hrIonUyEj}3rQrHLLTQ2OjbLPK%zU-{E_Ff
z^KS-$x{k$TICErWSRS2e;0;X*fep(Mn{fU|gERJ-L7oLG-F^P)$2owsC~R209Rkp7
zy;<9z@|-%Vy_xeYISb|TQzfc%)qsbHesk_eIFN_&j5k+pkGkba!A@ggJac{8;dx%p
zH0)g<wpki}`<%J~v6j$o6+oj=193<-1*o~=ipiDWvqFWyw@yg$QqQ8ph_LzuwVp+8
z-1jAhQ_HiNUf97Pu<al{*aT&5qic4FHI?9p_HwJI>Uw)s#m)E4LvZ5Za$6g)D#d}O
zkM#4?o-m5-c=vqvVD6o_Xq^a(Jp#nQk%fNWz84e3FJjf0wW&64NO`>57Tpp#Ik9#x
z7WyrksVk>tcrV!H#9lYSB|Fng1?@SdKLX<jcw7A8<cN^kPKkn1iQtjBXyj%4==M(|
z*M(D$jwnYB<p73b!z;VE<0?ceG&SAC+Tf7+es_2n_ol&W^8P|Eh6tq77u*n$)=Z-T
z!ttzs3xV-HGXpe?Lp$cS5P6{HrzRa@N#*A}#2IJK>7JK~!Z$nFOWH8XL%|!B2{PYj
zSTI*#x435S%@Ovxg|TuO{ozRp9KfV1vE9p5i!#-;a$H5XasdBoMrk{~1%o;$?WbZ|
z@Q0?AzL88%G=jZqL@2EzUK7C2J9mROI2ct|$kkV-F_=dslTZLGh(VXqdML;b#Ru(K
zEn(yc-y7b*&u_;m4S&MC_z`XOalP;>(IY+uxBK3`yEe#0-gDcNS@J!1h$eRIY#I7A
z?Z;`di0^V!7OVKc+}4%eEavWelYDh}vtO2eQ{bjqDS&c}c9gG>aTEY#WB<*&&IdIu
zEYi@V-!11;q@r<|u$VZ@)DNtg>!o%Fcm;!at*eH3wnQSC1=4a$`^WMSG>h=Vk8qlt
zi$#}Z-JSTAr}s&}C0VykxVp0O3=8sGbi$Qude1_o7CLB(5+iGyV_ctYhV#s03mzC8
zW$%A)A5;-xb<^4GZ(iEgTz{7we-rJ3c1U*TyN@pmSnXXY6eQEm4hzNf%xu0@LKK)O
zpwZl1KN}IfDmryMoSU4!pSY-w3l^^|QE>OUyhg;bEknZ$ozhNkbBvu{einLeeA9on
z>Oy@AzL-?^Z2ERdS>zW<TwRAD=vg^cdBV#-2X@VqYPsJN+Anfj7JVt|Z{2d?Tvc<r
z$CB-ugLWLoyX?WWa=sblOmM%yvZMH$I=>{q7+^7PU1x2_kl_Zn+f-Kcv$bUVNBrV<
zBWjY?2?;bgjNZ^7Ex5hj_-AxGAtlww#tOf0l$;MO+&EFA7BJg`ibp_mD&4K+PrHc<
zQP=mZIjrcV7Lgmy%D^F!1&@(M0B7UkgZu+T?K0OXTmWI@ByZom+XTG0dFRq$*Y`4z
z5QEq<7$WN<3Z*y1Wdr~zJN(V}5{$z!MDL*{r?|u2sa291Z_QH}KmK5f=t>_Q6Y+f`
z{mv9d{j7^r3@5z4BG#UvSB!{2_@Z9RUq^@@$GJqaHo=1y;%@>X!~Ivd`iqxAPsSM6
zh!BC-&7s%rp&5dq+KpBSt%V0_4w7D}K4^aCIZ8M%^kLxZ2+^SB+Y5rfSy%<j&TE}X
z>!rM}A47!<vl{n*!6VQItPgF-FLe7aP>vLpc9m)E@t1<+r<vPnp0xPh&{)t1gq2^D
zBrbp6YieF}Jrh<anC`|vGOG$EV4Tai5^?PKL1vxq>r!{<ysOSLceQvLsq1kRr$EXx
z(Ks|QwB<b5`v4m((_@~pxKZE<(^jdUesVT5S*Q1Dqc=^Eawk6d-sui_IhvtpT`T&_
zs7mnEaLnoWbrn|qp%K&Aezedgg#P3F{s>ybyDj}@PU~iZ>lVHpyX~uZ=Z9UAdBL(a
zGvO5T$WzXa`mwKr^VjNWtNvh6>rE|{=ZQ4W(RRh<`cC?S14rf3Ywz`Z8mw9r=f@%t
z{mO9}hhCZIrPWtHKKO+32><EYQ=X}7-u(k%9rvuF`#n@<eNX<0<))eFkeT_iHcK;o
z`|l++?R!rlFeC54UP}Aumh-!!r)?0pSL|WqRk-ts@-1<=+T8WU^zuRV`f!CAwfj;}
zBW%W@G>wBs4wNl*Z{8V*b!F5he7V@qGoSY<0czIz+|X}P-&trz*xgT_cd@#5HE)0J
zlpS5Gv&9wjA)(K}XA?oFDlj?fGk%w5A4&+~RAA91$%OMfk$plRy^lzbzdjj2CGL1N
zmsRZquYbt`Uhxe0X@khm=w+-Gw-XEZXx4Ffgm$=hd$w4kQQC@)a=YIDg!rMOA85Qu
zxulPYtB%n4p5`*2=6<AkdEMM;4(W^$NWY<8MMBy4Sn$@s6a_2Vv-@~0`JzAfU8j@<
zEIZFz>*2P$#{@IX&G@xzB3&H@5GV>&WYTfXF_kkN$t}?t(Nd9}1cp?-J$MRu0X?%>
zN6F~K76eFig8D_#d3<jX%LMLwIi;waKJk63WC(9+z=<<gUFY>dN*jW`Yi_fvZo-qL
zTM}Dn2X)Dz!%Lmkvg@<kRsg3rjw|G-_Hu-83O_FSl~vUqB=#H|)AUqFxIVi+LZl4d
z-jFn}Cw}{h0}~{euRiR*3bQ}AfGdUAzjsq=?KA6&;K`{fH;vbI)|9>;Ag8`j&mp-^
zl-5FR_b7g+$Jm;K<qGY83u89=bc42u>T=Q5so8Yt`s6j0h4%?D-s{uc&~{(@*QUQn
z)Ajo!HgIlqEMht|VW>ii{F{VL`=w1nTCnpa_v6De5uK0pd`csu=@}0zozyocVYWfx
z1x+k$=lun2%=xj|kw9D&7&1A{Ge%uu6GBli$F3aJNRG%szpurx<`hS>>e%pf4ATkI
z2<M7^3!t6^%|#*t{R<0l1Gy_6eZIgFK6D%2Ka^w8l(?7l*v&Zq7>m3_Gx2zC%@Px;
zsti$n=pe;A7TKH38SIF@VmDUh%%?UIdNInjD&`^W>BAXZ{c^~)u4@s&%54y^+(go#
zT8HOcY7#7cX=k&Enf{2uv^#i<^w>YxJncKEHT}gx6~PzZr|d1#MK2-9s3(7W#Ba$p
zVeB^hQnIxu7t%dzV7&3C!&9MkJXDAW?)RP0=HaG<Q`0$=Y9hScQJ$>3oQC2}`k@i6
zh|N=x5ptZla*w<IK$cj{J*lql=4r!zkjQziyf0~_TTny=x_g>uLGkG~;foFa?UeOf
zzG^9R<8k6l4tj<25;&)?h{D2aYRN_UCTK0SwG&KrZ1iYDJO|z^2}=@Md#6vDjnZ7T
zqDY;WD&9mhx_rHz4q_{7w(pET&;6wk&ZDQd8<ne{YUZhSe9M$pS2;01G|^L37rGTi
zeLl}DIN?x@aSacDooDb0b*Zcw_MHk5^=-wHqzHMVs0o<vMYs4=v;vH9qZ88B{wH6V
zV*0&>3c0_zV3T7+)v;@*3DXxCw4h^uMu~?4n|y;}X`$ac$%@Lt1YgIbV${{LcMaA^
z@XI|LF|b}}U~mdq>qnJ<%S5(+(Rlq6SBL!+Bz@BB5JcnKfuJ4}J^YMRG0@qB3_~S>
z7Z2)eAHT9hbajp)jK;VewwnD5f$ow*PFw=BBxj)OWmd@dZcdBoT&g0qK7Sh5a?7F0
z4`%Ve2i)^h6AF}UxxUv4T^Ud4f(*F1hM7ZF%5ptoYC`QqUFkKc%fEN|S=nk77;^IQ
zt&L=uCOwU|m7l9ckWeW^j4)fPo#sl(MR{sH+%3rwo1?4||M6?#cE_2v<vhYx_*B0;
z_At-qGn}3;YuWd6)3qKv<gFRHxgTw8^o)Of*rs59an-4K8l&JM>=iS#33y<jdMK{z
z*y6D{01IAD&=*g8IJob-z8Hw0zRwnZoEUT-?kAb49By352Hu0#MB(iYEb|)Iw}a_8
z6>Zdz@{)>S)>TJC+53iHFj*qcNiDuzXYgecsCAoZ;<#-euj%g2^5BOyN44y5(_Aan
zL&>4%3AMuSncg-*YR?l3_qkx0O)KOZ39eDpQTCN+EzZ(?U--=NvGX!l-ONCdV&#`N
zTlwjS_jRJSe2jC}kFE1IjhCXALuK9XZ*ZZ3M_v;Ox4oh*A3B6_X1ZIB-2{{k%^s;A
z3@31o8h|rIBzl&jjb?hTXKg~qwje#g$uZ&~70UEdPs<KRl8;TA4r;_hKGQUK-|QR^
zJer_zv@A_&xTwwBMI09Q0v3gDPdQ3N?4EqA4|md0;JYQgUEkU#O6ye6yJkJwMKon0
zXkDS6S(;)}dvv+<JHF*7i<j=4*OHHD+B@xMMHtQ7*DI(m38tFi#!rMaYNau8vUnu4
zUL4ThbgU6s1Lut7pqv$*r~(AX=mBe9%7JIU=)?Is6bnuiLIoMUa3@#HBT`owyK&9A
zJ6@0ab$|9FTNO{FeYMu|>x+n8_a4{bK#sVFL{kEo9(~R2S*69*O9OEoTmUMay%_0}
zeRvm92PNZa!pAv8zQ$8yrQ46p@Ee2gMGg_lHu@UOnR9zKhsp@}Y&J2H>4jV$9!;_1
zPh?CE^fd6XafQ&4k6OO%3rux6jWnkz%~hp-;sB2VKXjzgXV86Ia5UqpIMV=E46j`4
z%Y%-W=f>|jY@`!}!Y*jnd^a}`Igw6%U-VEO_AKtXy*tT+r0TAKnzu_M=<mTiF~#VQ
zeH0s*cVf7G8;nL%Io(^)eI|Tj4L@%%Pko)3^<$}Elt$WJ1&AD&b3lIVl1w5adm7}y
z6*Nk!%rMe*^PFypGb(~Kvj1yw>-h1j<#8JMMz^(hZZS-_cz9AUhW7#@tncL{muQ8V
z+a|Q-7+<p}wD_x))(<8zorHMciNHGOSPs7qpb;yZk4N?H&X4vmKu+=W=_Lj{P~iyj
zw-^%Czg$hzPjVh5P0zRu_9_SWpxTU0s!E>v)`}d0DKCf!%;frV`8&FEApGizX+O(^
zI=?|KDp&Mi%mj^yNt7VgBJTNaIpC7ExPZ66bAm)#wiZD)aMJ<Un0%owLWh7&Wf#~g
z%`xU&b2+*RVLL7qCuZurvNA`{3lT8$5JawE%OY4(ja9uoO@ECu$9jI<4{33`V3kO8
zq4ebLN}hPE*4;@SD>d{XC)xY3_L1z>;q^%@4#=08Goq1EHks}iri}Dy8~cLiP<T#?
zZz*+RGDn?(%@z@26zYu3ZP8)6>9BU~p+|o($9Y*Twi1#7y=1lK5?R7HoZaZ9MP-hF
zDF<l;q!u5a6`y8^tkK(vWMxK(H^t-lzW0>&S{24t-{V>WR`cSu0}y$GR=Lg5Kt$Xr
z`dg)(uyjkeltOj#?(!Yz;7j!}*KdWbB=`);6+Z70j-M}1{?2|o{q9P~y<LdGsU<83
z?f#R9d4@>-asX*?V>)>BidMMq>U<SG!_ttYC?uqL$z(<WEIaQTD;utA{c&|qFb1@2
zSp--&15DmVoM#+<XwB169^UO7Eengl95u@<`Kf{9DcU2PKpz|t=jRv<vc&sCLb}pR
zJpXAc`o(@a<x#9HkaQF@t2{2?b2E%M>KDc8>@!NrlsXW&GEDQIej0T;3hD2YKabNP
z=~!np%-V!%bJ%Y^h8(R2CjVq_7{Q{A)gi&=d#BQWA=HaC7a+@Ov-m}>|K3J=Im>+k
z$)KNGs8gPnu+$;%Lbc3JKa!-h&@>+kZ$e&(HfzM_=`db+rP1%&A%F8Kb-zAVN0z)1
z`_t-}(AmfKaboAOF{X`dfcA8h<!b+?;<Q%n2@CKK11VtA$>*|<c^5@+^}z<8+ur9}
zpL~kcu)WNNUj~C~55|Mc&rPDMVv}E3D{+4DLL1M#>6rlRx7z?R;yB-EbSS@)rAU8>
zDtb79cx>0S<|9UaNUe=+`GyuRsg|R>mS%B&Y1g>cUPG2xbw|YiT`;rlx|YIyDv-v-
z;Cx9PNVK@^XB|b5cYUV5;PGy>0XNT^M(Aj8$;SPW|K@lGN2c_8C0RXta{7>>q^*CB
zVcUnM%nKF^>EW)^SGi2GwU(Xd0t$@Y*8J=^`yhLzN!PgPjXY{bzsv!@T|Jr3V$y5k
zCBNM7^*4rJpM=!i-lnXS?SSuGdsFPeZzm^IbJHMaXWLo+#EGI>nV|HY**co<t<;ao
z{Z8e4J0SPl?VsF7E;A!BA5N;ELhs1y8Di<j&f+4{zXv=%SSa&7-kxRbjK&{-C)RR&
zFz+`ZiCB8P*~~gyP1HP7I|)48PZuT_;#@yZqj36?9tRDRl!F5<8cuI#XhiMg#AOP#
zWmvNgcl<P?l9hr}ezV#8RuQFl3;tpsC*;Rd*<ci<<5kkOZ>c)K<v*3*c;-mY<io~4
z;*AL4bK1uUA`JIC-gW|hJ8AfAWC;OGLLpXpt&U(+LSbcVd{9%6gFjtprTFB=GL@Od
z-gh>N0RF*7rPP+}P=tvuWX<!bm!K?ul=fFnd%B`5hlP)%DU;=WSg}$HVtD*)Bu^h<
z3w9kDee#Tw3Y>%u!Si%wi+(yAA|HdmE>WP=namxJTp9&*JpOi_T-oQ_A5m^NZQgU)
z6A=2ZpHhLN^cTAf?raUyF9}CpwWOF=+X@}<MQLqxQK#`+d-5idC=k^upl+;7C;!4Y
zqT_W{<LeE-ReEXveyU)>y=`ahg(y0OYma;i6#hQZA*(VVpgnk22yGyPKwVP7Lkl09
z$u=a#X`nCon2vW=+dd5i5xmf(TaMDRdCzR`lbW?1DRM;5sqZqnmE}q0;<^!B53%uK
zv5#pNNgwUaUI=)bWkn#l*VasRfpm(XC0+7S1nW>CpGH8na#3N`Ne1){IrT^F7?%$%
z9F}P-4Vs;i#Fttce0a_q66B*Iw+96Lr`csi>t6U$W;I>$%2%0RHc(AY$GzG#_mZN&
zO@hC<_z>LQ*x|wq#-h|&vSX90-MDa|LgVbn>R>5`edGZ~6Q9N$ompF-+`iy;db4%_
znSTd}>D9S)diD9QrqaTEdAuicK`^|#QkQ)Z^ozT}mgVs~>j3Z<3~{}~&9^)?)Pwjq
zX&(e5Ve2AKf=F+J>rD8MMPwpK*YiAMw_qGJ79#~08p8j=U)^01+#pged3P5j5radK
zH?U^cUW7CM$3dr+tA>!)y$U@iiVnX*KEoQ8ae#eVts~I61;eNddQHLt3C=&*6zVP0
z<laMnO`prB_^XQ|2!oa)pLS!u<{Vga+!wqxS~jeu&)_-@d+0A(vI*ePqc$OgYL0b!
z60R*%C@4|zzchVgaR)NvUziRV21{ou#<>|D7_eJk7GFplRLmi=vj3GeNRUH?C)K^Q
zvHvcgu)>d6*>V4c1chVsVPSgQi<S}lo1`1;#tjwnwx`Gr;e9XI=QFG_QXdVbfAvwW
z&x6A2UcrsuzwM7+w#zgEmsdT5y0NF;tiXt0J%zAYlSvr)l#Zp}W>qpD9Le^6pI(oz
zvRB@%AR3uOpd;X4=twGIORP+1Haik>m(a}k$Vs{deHOA%iT7cFfBIv){?wb#M_!B@
zWD;ywy?!S4Yqdttt{-D+wqlxLAJMfF<ie(rEeJGpMo!^Q2VAElF0PYt@IQ$i9C6Eo
zx_}i>>r?j+z5y-KtQj{R=Yq<;g#wa~eCwdXZPA;{us-%~3iig0j!||T?4VW|y!)5s
z?mPbTs>f2Xaru4h=ntQ@&Ej^Vyj>(Zy;z7Z8J-%hx_oY5mfof$aN3!9*|>fRi4S8M
zaT^Mrh8|UYIqz_f$kU5w3c;<@LeQFrYRzf{MoNPCkd5m_8$MX^u85M{BEn5I+c|oH
zD?r)S@1g{&yYg4ML{#97rO{;tzOLN;)WVU_Wr9lz2Dq_LkLGSrO3jZxO`YHc^%?fh
zXcO(#XepDm(ZNyhA#^`nXZmmJC=oe>0-AB>z=&n^j)ZE7g20(@Z;L%^#^+l0hW6BW
z*;Rfjlea^fB<Bl@B!@&TE>nZg2~>!H#p|!^B;<Ag98fGSz(<hATbUE7-(`p;Q9-r&
zRmAb+1dK$f-HbpVIlZ5+Sh7CMWyYaCt!@v(wEjGICQZU<j{5`u_-$Yi-fK%s{*KV!
zk)nPQ_2M|f6CpL^L;<vL@fS&qL0BVB&@wtR>DOBJ{S2L%mGf?E-YXG8?WkaWsr$8J
z`|IHCdS;$7Av_4j6Xdx=n`_>5>QjfX1u2bSQjGSm484J_ob)v6>t7Q&F&|74;h0{2
z*21nn?q@H=(Ph~IxRxh2Ay?bJqJ|VO2AZ4PFub?Mx&$41hfo(r8ZbX*fC-cRDJJo-
zkIDu>6$(y8lNI`K(o0Hu@g<4AIHx<&EG^M_4$}N>g>uzZ>oGRiVM%nNrJqNhelN*w
zj_{>X7&A@#v0GgYI!6ar(o1U1j!1QP{*%7SAp{UmphD1%25Z7nF3f;I@&Q$NoAlSA
zZeG(jA|v$F!vf|TfFQ|XAB|bs&S}l4GwUNs^aq6N<eh)t+pAywh*+PT`Dv7fX0I8s
z+{$D<>4O1ON@xdVJWVP{b9hw2P$x!2so|X21ys|UoEsL;AUw9F1!@yOq>W$5vLB~6
z?;V<&uWu?#8YXfJm$u$~U4F1<BYG9n^JqF8iX)Zc|1fc8fmUQ)wXLj}Bdp*1&xOBQ
z@jnUca?17%+yX7Ep#*kerfsi+DR9yAzt;lR*INxZD2tp5d+lX!1lskxP24(A&b0FI
zxo#|E37gur+M3^i^$@|(wGsO#iVY@%YvBxC8X_;Y1v(f86YLf=gU@6YD`+SOQOdhM
zjq!VHw#&IL$%Zqg$6Y3ybZGo2i8h{XfZKNg<~ISOpBlS?hI<kUL}e^4_^L+$`uME{
zG1)!)4%bS{u*wgvPE+%Dl`RSNE(Fg+8UZ2isf}4)K;IknFV|acM7JWMwRu<L&}&l+
zHYj*telkfB_KUtN7s}w7nlJ)X!#Dd`sdu&Dpt!1}=fprgr7}YF<hP7qnAR9CYTM%_
z+DyD};LLj25(Rdg)pml7;qAUCc5P?)V%r9#$?0(NaZOBw$Y7&m>Zs{$XR#N?jHX$7
z_f{f^THAc|y7-u!ZT9WBLW~y4>MBXPrv)3Jx|V<6zP3<Xg)fA{w!iHVghmO9Xz_?=
zZ&^_B!Wj5^=YF5ddl2DS$D%pL>vim!uWMzNX6v;V-P#0dDQmg!alaf3RGg_EzX!2Y
z!CMR=9e%+!!7J}-cknUu{~*fAuMei$!w=Q-JnS61gDUS^wsnVq<L;LX`F``Nb;c0t
z?J+C)dE2UsgY@|H&-U_D9&zW9<(`j+AJYR(k6x=z?#=}<c#6+x1Vjs@FvMsKl#Z!b
zE_A9$0+{**LpPgQ`R8z(mo>&B8aa-@)2B9Ot}XWA3~}M25AXo?-9ZP<o}k>Vg`dq0
z!5B1*&`A3%>PH3V7*sEQrZt`ZifVe<NxmY5EmRb>T=o|W5()Y|P!m%&{<#Fe>7a_9
zZU5kCzAa57={)zePE~OoK(Q<OGr&yaP^H%9%_NXW$3l7;j-ok`=<!kSrH~z?LeOqJ
z@*35ipZACD6O{1bALaT24u~ZxLGId>ifK$O8cA;(blwgk11q}ly&ANjrt%Je-Yv#Q
z4?StY^xoGu7yz<;Ku7{XIXrjdB;I6bMSLEM(eQd-D2?pOchk*qsImkwuGwGF%-~5?
zUrA`XU=oIA61@3@B|Ub~KHO3xU`W1)7|K7)h9DvRW9|TJxd+3B-&6M5pq3i_=%ieu
zRE?Sg<6X1g985UsjT~SlOR`Ooxg;T9lTZ70nr9N!h95CqY5?Gfs#nF@!R_0=Z`(Rv
z^N#MsSL6b8y{<P`%a?g%byW!q$RUwv=Ayqj3<|Gg!WkyP+r9Db!yybKCj8(z^5*fu
z2Iw52g5zqub2B1PYPBB5%K-~`ppg|6Kn`K(6(Coho*2(VM;_La^c8irN&PkobW%t(
zFH9v%n}2cIhab>9H|Q+|wT6OE7Rqyg#$82at!`hH%#Zb*Ce;`Yqa|Y_CXgn;<x^L#
z)TbFOhj}0FG^(md)p@$AZ%L)>p@k~(kwasF1N79WDHQ}zduN)?B!c|}g2(I4wA82K
z%aE)zM;br4h-|~<Jc9afEk_M^r0WN?03u@vToGa5oby9VC?!Xo6=y`=;Rg?MW0le~
z=b^El!Dwo4Y3lpE;tMr&*_o+-`Y-_q@1_ECPh{>}v+P1(z)K4cedlf>^|v<*<t{X)
zgu`lcg|$OzQo_nClmdw5I{)=&o4;jH$#a1zI#apuH%~1wJsA8y5X%ie-Y!3e>zFNT
zi`Z)1tD6}Pk^nuZNGbI<{TvMEi2-+pYEK5SU;Z-OXJinc_%Szy{|E)#RlJq2vssks
za6~LVjQ86t@5E+n2rm;@w}`2GS{W@i;ff^7=FK}Vqn|Z(9h5cym0f!w1+8fcpzYDp
zkFR7`YV-T8k!JNu-~8#>0q<DHza5z3j<+3uAgYi46g*+HTIR4l%wUmNZSxWnvO9P;
zWi*7vX0BnR?R=ffswJ1~mQ<LXttirbz^#9GuIKnzRVT@m?ygha+bZ}$BUszGrFC>o
z7D&!s3HKSLsnY<uZIz%6E~V$IH5sfP%aY_7Y?dThf5wEgsOvClIUUAXx|wR#!@l6%
zmGLMx-%g5wYK}u8sY+@>8m-_sZgrU!o(K=3j&zfGW=u$?jd8<gs9<4_;o`R|jb<v<
zBN6D`Ce@Zm7IQ-m8Q`uVSQ>;ih*8iSrwffXOdO^MSl$Qx=l=qT5PH+0(*If8db9JG
z@mjVcT?n`M^i_P9ydKHpA!9^yE<e+!5Ura|pJ#)=#!UOi{Y|hd8nyck8NO9p<@k%H
z9iyf#t$lejQZDg!aL|bT)sR29{ELz-K$DmVs-Yp*Xb`oE$bu-QvEDiRh%WQ8Q1-b7
zZZh9q>~EI{RC=V{(!R2%QGz$`iWmhgbcN^}B|b=G>y)3U2mc?Oy>(EP?br8BhajnR
z2?8n&N~e^FfCxypbT^w$>5`Ie0YSREyFt3UJ2tR^=is`Y`@Vnoyfg32`_A@{?ToYC
z=Xvg99qU-@`~56YXUCG{D2+|_2xpx?;Ts^AwLI;PpR=~mf>1-d<BW8kmRF`EVV?Dq
zY~9V9GJWevOX?hxq=5L8fb^fA`Q*-26$h2E^HEfOGS`2ou#_F9{#H~|fpbJDRcFU#
z5#vy_^3Bn!CPtPxXs~~C-0qXb7qZ)$WP#HxH)|?|v&LCr!G%!W(x3N<_|Jc-aaxLL
zD##^Ki{*f{0jhb#*+6e<(*9PIzs&DN@uh9!z}l5T5b8Y=hNtW?;Z>C}rJ0GzKkGFU
zb?Q3n-}M?+2Z|e2M0#CG`Nch=Cd|^5Zz7!iSVy-Yz@|cd<sDC1+@V)+`>L&2#ivt|
zJ~ZNuD%-H0JadHe$iv3M=#TbYhRNgp97*Y`YWt?gbH$i1_vYXKb*_lx)S6^}($7DJ
z$ly7!$JYJK=@60EiYJBbUQMIY;*x(iNUbf$>_5#~B%(UO>SOWVigYMz($7l^@yXFi
zhV8!+TR-(=ylrRKY#sf8z10?H(A~$AN>~(PmEzxQTm6jO==7}z<?Y+ewIO#KMcb!S
zo$-w;iJ%A3yw2<QmPtmoD@{onrj7?&IeJr2_yvzH&iF$eaDbGr*8xT4ukEr1WGV4f
zBj~6WlII_#s(C_-d<c&!vzq#Gd}U~C{gV5F`q!Uc#xtsNuUwnHR9P_i2wnLXYI`uL
zV)HV{h>!Mz{_TGkP$pHzxInW4txS-7Coc1T^?Ex8J2f!X?XGG2FxN6TTyUqPNA+9C
zq=x6;?X{!D56=TOUfU*V_%vT|YVoP*C5S7m5keKq2cyNEwJLpt{F26HxGe0UHmF~p
zM_d!`Zgi%s@SIq<;=-6pS`ftFCwx?Dv*{+^q8Ip$TeaZ47S{B1oslNSIl-XpcSkud
zR>RG1K~(wMXKcNzj3l?073FH%@rd8X(k_`-*>{g6(C?wfxkkgO9_Tu8@I{-$s$#!s
zsy3{=5p6=RSjLJRwR>*&NSL7dK?5M$tu;mn$Hq|kjf;yCoo<5V*x}2m^p33d`?+kL
z;m`Dpqn0s=A?k_<j01E(Hb$l1)1jDLpn?!^&=Vj|bMZuNT>=)9g-pVsG6OMnx&%&0
z2VyIw^t*A(!u92LV4UL>TPbTe6>Al(Rwxjr^010?H?m-S4@;;naBO3|pJ&APgsh{t
z{h-A>sp+^XFleohplvPqgMmv(<MW}8e~;uw@0_EZvRS>IC#0%rF-J~0q95#qL!!s1
zT!NF|l6zI%-iY<8&Yj`^-S7(w%QFKO&rduoFHK|lKn(!wbxJ6es|Kp8-rcH5o*;F9
zPNTljtG2ZDxD}NK<Fr*ON}hS>@HZ#7F#fYvUgGj$Xf??Uj%_{a^VbXQZ#fwcs(~9s
z$aXR>x*EEqiew1s$V#RXB32PwLc0TN<@83)lS`VEMcQS3=AV^vJw*-eFIXHZ4~q|#
z6KLLJ%Hn5;I41(SqoZzajCI^Q7c$D=QO|5S>Dx*IWs4+%2$#$_>m9ZKIez1Kci|ZS
zTpmt%b%mHDg^SuePA7DRXU7)V$3gV+0~XG)1|0+NE8e`Hh9ZQqjQV6nrDB$=Co&|5
zGXwc<p6A28|93~8)+PEhw=XrhW8v!GtJfj_w5*vdU;qALhDJkPR?|NAk$oT5b^393
z8J--qh69ZpG|S}hVYM5DK(Fk2cQGYtN&G$2`T*>9;#0@%O!}YgpRRapX9NTn^)o~~
zgCw>`OcKDrp*>KKVqdMm99IY>_y^>=*z?ZpMdhpVPusM`CLn-Qq*jLN;=U#}M0veX
z=ILMKPwv;J@;l>Rm$l@pGALrwfCk)8!<C5EPt`B&VP(cGUhuS^N`gDH?dLaCl|ho;
z**-#YJID;Jzc1>~;-h$Q`YmkU;WwNaA<;l^chDh9Q=M&Pe?T4&PTAmm4jiYcxmL1I
zY4oaS<hWlZn(pR*8K!(ZN}s3jiCgiu>T5v1?s=p#wv`t=hec_r80?pim5nN|3FZ++
z7%h6dvXv&3JjNOhvN|Jx;Z{uvxpKZ&B=MA$z74B7Rt{MnW~yJ`tBRw8niM>`WG=tj
zbVX2Ve;H&|6?ZL27M<>jUydz2iK(wnT!V&OgaVhsi#PO{Ywn%(f8>3HkZ39v7*w0{
z=$KWDGO%0gO@OEA4MUp|4h0)B<L~F7?rMs<wzkr3yRLnIN@yiTYPQMEm^kk7`<n$8
z#<pdGu+ZzMefrKjD-%y!xe-=$pXI149{kV^vV*F((!3i*#TU*vFgCUn^XiT}>ah8-
zMftJcc3LOGz=DFe$F@PX;|`wR4UD*2B2_sxP^kK=w=isx@m59ly<ZI7m%&vl0fU%^
z+KDgHe__)Nr%2dET$r&4noJm{b7Rsh<bNv0KFZJ5Gu>kf%8HQ%Iy3B?7MdNW#gevB
z9_>t8TS5$3DrCY>6mRz%R8AZCkvyTkI{sx{HALZW=^9IW<dUcc-(?m7vme_AKhxbC
z^8PPV>wmS1o;8^cgRO12hZ<@ocAdw$7+X|u;hrkV4-NEjY6DshMNZ>$DI(WjgPO8B
zaXHwrlKf~e5QNAdEpPcJ>6W=q`&FnY_2(g5Dfv?dxIZ^=vAa}Eq|ex7XbA;%7~&~n
z5C3dxlvkXTWlN_1lVkpXFac}O?g9Utb1zi!mM?<6ul)35gw6(w8=U8(O#Kqb2pg3w
zVxCMX%^417rM}{STk7lmybtptCHX(BzQX$GZ$BMN8ohgMj^0d6^@%s`1IAW+Tu{`v
zENcKVV5-IHr)))*{}RuvnF&0AOk~XFrN6ymBabl|M>@$T3EOH*#Gg3>SF$9~V3nsu
zE|`}OK9Q~vfE~BfPT5gv_b9Hlb?7(~s53spB6-`>O+Jbh$Gf`Kes5WS@bx@!I3+gL
z_UpalyiGS+0SzPedtXv`T<xAw8>82?%cemOeGJ7Kc+9K#PHrzMd+Cy%X-X2UMR%!W
z{WSTN5M1PK1I>I5jaQ~Xzu%~|uxQo(V-VuJ*;-WDHyJG_s2W4~-oq$CFs>-;UKp+W
zr6@OH)w&%cR%d)`FU{eL;%`j)pM)=W{}#NQXAJ3B#zxSi(=Day_+lR;#~6x5DaenM
zv1{??`lXyK61&WC0FJl8Wm|j@rnTO?H`&_L>@QLZZ~qn0{zge1_Xq2HPG{U96Nvp8
z-bX6Qx;gYZY~8#yNOF8C-TPzGNZakp?sbK0A9;cD@zp%vgR7Ia$J%wiZq4<5K^8J$
zLUZBrk@uFR9YT#srqDs1!}*|9TK0{|PgPUsIYxMd<D%W|gc(S8iGYn}?I!wRr*K4O
z={4_8vc5ul_q((QsAa_>otdO&-5}9Z*z(E5-sv-#yWkD^u*0>|jab2-@{jht7dzLl
zI=GrF=vpaP4H*eud@Xv3q}P&MmzrZUKL02?^<V1cjWSaejN^^PW*`B@x*w<vrWuQ}
z-=nT9@{U0DZmlu!od(?@^tAEEn!9R<!QMv?+Tj<!GUZzM-n?elPbb5e{1p17`JLLa
z<n*H3nT*g27a)OyI=&;|#_^z?8$NN4jKG0LZ&?EQH8lT+dr^)MzLkQ12;CLZ)16OE
zLcoM98u_s*P;o~AcS@obi10GwuoG9`*ByUi)QNJ!4W(n!-gu5ZMoz()+xtaVios_+
z2>81CL!?c#yN0)7>EFA}OuoShTuL%{0IAlMfsdeVLvg}6K0M3CbBiTYjY@n&o95%B
zOLw&e@@D4RH09KmC71R{Oa+(|6C-fR^Rr9}f(px;*0Y0XE@FmlVD~#(d!Ax11OmnC
z+6Onk^OJ5$ddjKDrFD2gi46gXM#WyXOW)^&H|;j@sM4^0>0`<3Rej`L<-&qRpu)Xh
z^ZJO8A~Esq-|hS4ZGllUL96(2Gf>OWVnQn85hEDq7*9ZN;GgYXRmIQM%sp)j7u(MO
zEzqQ$4Qp5uIMKTIAE4brZg-+d_5Zce;~4JVub*z6)RYWMo_6?JBxX_eLJIRHcR@i;
zTJdb<IVDM~nYwPa*V}>^ipuW-5cmHTKCJ#NxcxuEM?}PTfGycqRui5l<5K)$G5`DD
zdYgX`&r3lKgTu($#J8d^O|EwSle!}8j$}&+d$FjO?#q(9v!8SKD3SE=dS^dx(xF$z
z3t=J<rvTdhHenuoXDMBxE8xQ9_lKXO`nbxChy<*h78$w&Q(e{bwjpVSYy8?z>FDc$
z0kZ6ewe>+82FreYZ=y{XF~(%oPq?i1uIqJ5A3VATHCxPJx+B`fKpi3?9fW|TULb{9
z?g!?|1JpJMR+v2<WR(E3-90c1;09Eern>-Z*al|66X%>&KMUm`3Nr{>_ME2P%o-<f
zQo;EaRBHWnm{!xL-BfS>36-us&sa@AM?}Znpnh{DlIKMX(`r0n{p1_<G`<y6ejz`%
z_HS=ewE9q@F-)ddqkJkqoZ6(?T&31Nl*O?f8WC4kuz#HHz3qs&LBZgUNvc5}!U_L4
z=W;5g1$jwPL<NVwLs*qf?pw^*S`DPm(Bg!4o+>oH@wm!?@}k13)R6An%1Z|Hdz`pN
zuJfF20s5a9o*WKG1P2qI^=($+o~H{*w3&s)UGCdQ3)bZhS^fkmOWJAAem^CDh!P4{
zsz;1~N&6kYVg9A(xg#ho#$lf-$h&%kDsF`h$=K$<TDn<&cb`aHPmSeTxb1cvvS*m$
z9%p>5<RYmOPI`Zv>%WJ#r71U0MD8}xG|I_mDbG!DXu$Vv;MN~X;kt`^zu=S8(DKq8
zA`5GR-fv|sh4Bt^`x3@HCAkc4#=>1A^-T+t<=)#iUg;z}pZuW%f0o^1)ALhA{^8Ck
zyspOiuxtuzG{iS}?PlZWRc3u!1KX_mo{)sykbq=GXBBDgcUlIFG2RX?^YK%pEge#$
zmV4Bt29iIei@)W8a~_9$^=I?LMZdpU;8{D%B85fMV3*du(<Lpvr1ry?fw#qO&#|Wt
zBD5%FgxVh%*%XFZKqut6c<hNQ$cEVRy#tfI-S2VCX?D?@OJ@~^<d>FQ_Wm?AE$R3^
zZTPzieTjaR3Owdo0|Da6ws9Gc6o@6uP8W4+w3j$o@&Q9^-DFLeIv44Qjz-5>vrN%D
zDmcKNh7lrW>BjHV`GheJ`32tvZ$qH#{HPYz0Q6%@T`Egsdud5nNaX4j4d=HXAx=ik
z9(S2Qs9m^D{0WP?Co8PKOeM@3jWJ<A^KP(PB9O7T{;+O4lw;`F1U>+0sFwoDlb0hn
zodT_<Pqn_G4<GZHtCt0LwmfVmSsjYMtSG>x>54aH>h_Aj1`LgBw2;``R2S1;No4*j
z5UQ>+@sSFEwU&?jcm-!C@B&`&P>Q!GgWPLGh)M2caq$Hg1lfjkvK>bdMdXdU^z7Tf
zx$Ucwt@LJuaZpL4g3@4*IMVQUB;6BPJnhI}%?^{~`_8F04`T35$hv!>MJW=Obe7(W
zfL{GRG6!VJ2HlI4vUq#dGo<?$iWm?=w&(cAD^AltSV(sA%t|u9e@G}xe5PZY8WCU|
zP{uzA^rOmn$qjw4?$mc>cZ;00lTh1|%0zi-nQS*>>78X8b(gg?z7nhlT}-ME$18O)
zs%TPiFXD$1zYrGPmt2TV9snywH~$FQtrs&6<>T=ykXMg85k0S;x9t}PHXfO%zoII~
zSbTTq<9E~8@8tz2LhB0|^QZDduJeAdvG;vrzh4&!kG3i>QJ~Vw4yU)keWxEQc;Jo`
zAxYsNGs!8ACN~M1r}<9(t_+J^ega3|7XMVBi#GOFeOchCT@!(o|6AJF!D(iZ=7-{}
z0~E~JPIo|jFSh9WG#?wiF35+ebyN~aOoX%B?!qwWqmW>qA{8|12qaWJr#2QFuBhEH
zW0jR9jl6t%VZZ}AgMx~a&Mj$|#77iJ!BQBIiQ0~6Gvn%PVu=>)qAc>DQiM&hHvT7D
zd2_gd3fMGxQg*6qt^7}}<7$aiXil-hD>8+8_`EKer$(w?cm`BKYB=|*{;7k|M0hwX
z|IBIXySeYpdOX_^{+uWu%nCi7XWWUxbSgQm2L7j;D`aV}>h-d!nes_|la(xNoJ|X&
z+*kCbvh~XUtKVWuKjN)EpZ<k;NHvF@t5)|g<e6QHl3mKMqNgVP$h=p2ZL6l;WWj(t
zu6c3ueO;}{;*UK|`lWfVLLP?Kj*oVzy&ocuzCSim<OZ4IO^VLUEZN&{*GS?xfoJ~O
zs1_dn5r|PMMP-L*d_QC9)iNL7E8*;8hd8>)Td;{_wDK9$H%SI%=n_e-S-HH{sx2XG
z&Du5F4^H#(Z#O%nHtCQCndoErHK7M6;ugRf_>PNPk(<DAe(cjGN#|x^_0L=Ora){P
z-!#&acSiCZ^WVyv2W6;|)T^cglgjak8s5y<9^p!ot*k9Gz9tw_#`omvH5+ottVQQU
z%JO%edBLP139crzws`xFhpWb5sLsEZrE5UO<JL8qb!6VJur@rChaqSaMJN<nC*nz|
zeT5u}E18`1iGwRDPMyLAaba<(;oc6PbCWLi$zp9UZ5p#bBG3MWLe&RtuF{{ReDtdq
z*TLIw%gm#&lf7D!TUcY0Nm)6rE|N<fd{=3wJ|AkJ&h+$$^>BffkwGf|0m_|QiXxu(
zgU~Wvm^(BD2VBv|+Wz41sc5;=B45#KTi<X&*a=*n_eVQw3A<>mO}gc$ah5HG*1e_$
ze+-AzB5<$1C4+RfAb%@wE9F%KPSxZrF=1F$n#H%>HHx||wtDD1!@}E<?D68})3}SD
zIrW2l5B&bV<hi@XQ}Z4{LKIk9ZbPw`BaIO=^pqZkYhB^xIj$@wdzKZy$Xy2RSMW%!
zPnr*g)DO#*wcHjq6OFLcorC-jPDzKVEjb*RG|D??J<E!lNemU_I4T{E1j`4TKeOyD
z>Kon~z*veXC=IneE_3eA*NFJex9`lm@8^=#XREEP+_rW@imIa{1o{cozQNzNCzsFh
zH*S{6(|DHPwl|mKF1eWr-5+OJG;E+a&LVerTb<o@leuL!o^{=NctT57E<2YW63bDH
z6l&{e4sVKTe~{btt(=JZyDw_yk~<Y?I<~^8mmb*)k}bL2g*|=P<Rd+-+Pk-5is$mI
zR4=M;|El4@b30x%R|7Ryu1QNNZXz>{hn;<_zpX;G0M@uf{L2|K0V5}ibjtlCT^jTC
zY8s36t1R<=3$D_XtD9&ZT;0*T70E`%t_V`A`t!N&u5e<~hSRwU{?;H{{zcDQM8b%I
z{d~zZhaH|Y>DyFJyQ_r#X3?gqjKib3m8IEK)O}Oe{pxaHd(YWcg=`em<Q_Ghr2Z*y
z+w^yny>~Y13GwF?fz7Ua5GD=&+8g14)Eq8V$fdk09+}MevrG5QE$Nxuf?rClL2nqR
zVWJ&A{$?E4sfFb3TY9v!gD>IrB1(3X{sTcpFvx6a#sdCTe~_J<YtU^zO^R3VyjV!B
zh~|=|ZU4fp<4X&ktCLVqtp-iftAbSbUFq3Ae!E4;);q3~&AF_Kxq0D6`QkHK$M+wy
zEVjhN0cxo}0iWVIg(M$=BR4FI>5%S^8x(`EHl=b>tv@-C*(0xH__ocxLSNQ-`^Y!v
zcDb8_hVj@dAAc>O+<7bEndOj8fKg2^&zie5OiG%5l@&Tpuw<BI;Z~D8(m#)<DK7U*
zMuM_am5_n=(C<#78Q#N2IY_c*D2UR%+5bK~sR8gtMt-a%l-RG*KWNZ<oSN(~a{k2c
z`kAzUags$TDug)6Jv1(1x5{G|jqNWOhhK9IYkhZ)pUgVGMSP=mj`{=QvscDD>r&zH
zT6CRnRrqEEdn2NPYKeD`wMkQCDd6;{(RKR6rdWKF+&UfCV9*V}(4`diF4|2Qt1oSH
zG$>hr_8d|EwXoNbCTNiy*<s8#%9RDQ(Dh09HL(X-6G`-e5$cA0$r~S*ir2V2<eBq0
zR>NIF8{VssZuB6YPwm)!k&9$`l{SGpA#}bn6M*bMxPz~eYA~T%f3o3KhJ<>v-;u`p
zIs5}<Un4%!>=TAU9G~TCP^{$d15Gp))~{{c-__+?a`_yB@cnMeKHkAmb~(>W#(0?1
zc=gm5Wl-oO!%H?fgSxxhBHFOodE5!kO_glu{u$-n3l-K${M|ppLJ#vCv%m!~sG7`n
zm0==ZvTG1TUtwT8&ML9hLF&=)op%KoonG^|qvH<zG1iqe-)^7$lH^k=*tR2h`IcUT
zo!*{{{uv`w*^*vb(>D!L@%orv`g-}ka7f0DbnU~*NrkP&%U!Hp<Wre3nL}rs)I!~!
ztj6T^FT+PJYa+yslXmX6-RNa7ZdQw@V~h<)@6%R8SelM6gbs8TWf}@rna!>PZ<U5U
zwEodAd1MUPQEuOIq8wSwDxLE|_04}=)!eE~R{N$HH@vKaoM)7E`Bw2?2j3^%wsdcy
zF5Y!SI8hxa&Q45po~O)94$+=0!|Cs6w1e&TyCbrN=IlOfR4y}OMNBgP$|wv~*B~^_
zk@@cVV%9lV&uv4!@sjOg(4E_S&(%;_Jdd}bsq?PDG*FDXjQjEmnl<y56jV2@-t#XC
zsqP*$%sG2Y3CuP08ONP}_?~o6x)A`8x`~wC-`|T|cUW=K`bXr1CO2J&Jp`MVgt`MN
z2;pFg|Jv+-vH;gb0hEmru{Lnf)CSXo<5;D4^h377fR_~QR@3Aaz7^U#BJ*s~gSgDH
zt3}sgC$Akv*4NVX3BANu20EiH>>eD;-?_t3-r=jxw*%K(^4p39!Z-I^(W-|bt(ow3
z6(u}-cv_0_j?82<_n1X>ENgB7#JFV9%KXvZD%E1~wJfe~HMrxWwMLPMt@yqj8W?HI
z#`_%HK6yQ=ee5W_+io=m;dkp8zS5sQQvVKG>c0Eik@ohxJgp=N&eig9d@JVab|K=8
zlqCusjxlw%ST?eHhfD|RsIv7?@CVfsWgyJrUuA6ZpR=8*`o3J8TGqMFr}x{otrq9n
zbVlCsXNigZ47EleUCM@7nl{n17f0+ZG6}-3b{WT^=%4)l4DtF=e$y*U;@@HnOuYR#
zEcbQ&xzlc(b5rk^F5s@9iFskvIx*!r0YCBjlUHj(9r}yZa;P%3+uFTtQL5VyCvr0G
z9Q6ASLLF@|wx3?mM3>OhBgWfnDVkL>mWPv?O(o^83d8b8a#+ez#{n(-lY2)MOP*<;
z*^I}BM|v*>r_l-cWbhaTA0|RB+s2EZEmXf>6092f9=$8-r;R;-VZ74#y~b=9(eX9j
zfxvYShoF0AP?UhnL$u{8%#~$=X0WbyZ%aKro)C|@>0FO?wFKJkNyIao>H*_%U2mc{
zpPp)O`~U~98=-3G*jAfb1&k0gZ6QqC{`Np6+I$oqU_Qb}NsAp1J8WDXt7@g&ivfJC
z?|kx$jSrBKRLCka*T{Xe!_I8yrqhbHVJ)pi>|f=IjEB3~pykx7_~Po5#H#s^jqQK@
znO3`}P?}5-)7(3oD0l}#?s!#DA0a|j9qb`|FZ8U`okx)^u;|qowss9#Z_nb)tg4H@
zV=`<xjGk;LYfa`1s)=_)MIc1%&Vh~FtH)goR&$>TBiMxQ%`MTjlbo6$=I-6rJBuOe
zPV|g2bq%LV^Gi1R-E;m%nwG=-_YDrV5J-JWRmFkR46ldO%v}Sr=chNUR6}C!i5Qdy
z*+I^27oU7MuDiRRR&*OHf6+4b9NiS@ki1haTFjCOWNlT2o>_>ErCfFBg)nWR`oyam
z5<gfP;)S~Ot5(0^Q+?K^xEI%II2+Sol@OD;C?O_F<g6@TO`}Iu$UR$B!RBk%PvDBO
z;#A8N?PzJzZJXPe4i^_VkMnd)(2KN<x=obZIc;P}?g?$5i&{7OSWGXIO383Gv(Fp4
zW#i(Ot>boV7F<}AV_@mew*_4%Cm09bAHy$^+5~24JC;<h^6ViiWG3Qd^QS6JnzBOP
zz3*373ngxvP%Pfi?Qc6FvAr4AWskD3{v=C%NqxvIxcnYX`UWX}0Dr|fj_H67DP#yC
z!{}pj55_r-PB-QIXGTxQg?@o<ED+dNv%`O!#bc&3MZ><1M3?sy*bt=dp`j2t4tN#E
ziy}|DBSnz|2u2C)U!*YAtWx2aGnsBgCq!e(@D~H?Nfx@Og%q@F4Yu7Ve@fYqozQEK
z6aKoBP6(^#?|*glRWMw1y)}Rt%{`ygc#khcOw0H~5~2KalV?CPiw^65+;4DDcFo3o
zs)S5Ln_gmnHq1LCUiuI&wXQiubtq;YEQ%A%%0FRtufqV7OwMD#2^-JNn5Vi)$D$P)
zYtF%zZcG}$M=I8eo0rz?{rPd8t*axTsFlq_Y$;)D`FB3Y+Wjf~x3W*j&1-Xv4#33!
z_gBvs!PS#|r!*VhU32G(^*laY-tSlDu$x|T?6x10L}gX+YoB%(in%BzOwhbmDcOG*
zR)je}td8A03)Mc@6#NlX4_s=8$^y5L#^JL-)SXs7U&p7{zBfM{O^pk_U!h&7BrZp=
zemf8~2<ryy#nsz4&^5-Jo0<{2jEDF9#^qy^xdNZ_Ejk2-ZmNzVPFmsa*_K4=zh9Cc
zVc!^_os4NrIxo{xmLD_c9tl(Wz=m9tn+Oly%w}dI?)b0iqlm3Q>VX)}Kq{i}Znvd;
z@h158%;jN1-D+#03**!7%%+k=dA6il&CdAvVn%~}6GLso>^@`y|HOI>HCI-LIr4rA
zoj&=q5+B=Pz~z7TvVT7!O(VRq?!gXW{Y&*S<W@xTXR+G#@8Tb8yu2xx+3b+uC6y(e
zF@KhXgc5`F6cIucnsJmbCu!eZN(k^!bZY~KI6PL3(Z)wDd)bJS$y#LV17z9o^*;eC
zKgyXi9RY(gU7Pr-WNs=q&ZTX1J2Zk4bCOk8m_8KBu`T9I^h_rcJ_OsOC<i)T+u8gv
z-1?V{7e2vQlv|tNupUBa&=)$QAoVdD@{_!E(_R{g90Jh0f>@7?58dp<x_673-!AL4
zL*v>AVQ<xegT$V;P(tDxLiI}mq-5wuKPaE#Zp5G`OKg$0saO%y?A!OX-(}-QXU8D-
zJ-xuU$^ZNJ|DK0F;=|Yy#P33<<LeUMQd{&Ebi_-I#L+FC{$`K@N{4H;>n$61+Ez+k
zitB|}+JbKcLasIj1n5+YHq;7R!>)I>GUs@Qw-;$;KR+jf*iKvt`K2*i{3=UG&Z(|x
z|Di3p<q2@0$+Qi)hbiwS)iKxIJ6zGDeySH$ciMNzF>XJ#^QkQria5JYyIuHFGzsb2
zxwGqUT!V2e2r%(fe=EK!9&->1zwhnTuZa+)r@Ut#s?+4!d+Sc`WZrPczv(%EBP7eW
z@UZwG^$^&7MbmUO6;94Kq)w9DbcV@{wa}2$V0#jHtH6apDA=YNmQ)&Y`=O<?Aen|-
zf+xgJ#W<U3W4wGXl+$&1rmu?Yv`@L0)qTsEy5}X!@J1m$-leGqFpq(P&hxpqlCxjO
zyVg8*i$*SAXjM4j(4T0T^^R_0zNfydrLREG=YU|JV(u!;YKK`{FY}x@vFk2l6Vf?N
z8@0V*w#>BX*Om^yl2n^x^PoI=N#;#^vL$%GJT`V`(|kHjFR*Hme8p?wRQOlVoq5N&
zxMJcv*3}4MI$71av9f|E{?##-<xZ+~YKgVbk^WfR7IJE~1;6D=>T$y{-ygC_!Sf6s
z>CGT({=K*={3F|%GRjl<1N2$0lZGiEc_)O8cA+y*5HnSpv|XsKfxjaQ&Cs10!8B^`
zTuzaOGdz2$e$W&vj!zb>OIB_<N654Ivv-&0JLS_{OZU$~$%~3NdH%-8w+Y`7@5=VQ
zUP=Fyec{?}SH2mvj6DZ>2J{$xdPIn57MQ!{yZyE%QOcU4cRd06=lZz4G>Lhe3*<&W
zw*;@fn<(9g(=v-Pn{vz*&#(#tXs;t&1P}9A)`orY2Yd^D4P~e1Hg&UgDO#j-UtCPQ
zn<p%uy4D<`T~j`+bLp4ldH5|L>n^30*Cj8Z)>j_QKyh+x^i*{6(w3i5_0lX1b>-K7
z$C7vN+clCgz6xx#v?H_HDy_tVYPalbrT)?eAQk9BbnN9tX;tNEr97b%1bOam7^=g|
zJHb=LwWJAcXs6X!S`{4S8U0#poM8hW@k0mpEZRKf?0xL)JDcX+Fv(+IC#mfKwup8j
zis4OnS+bw5FU+}TxnpN}T8CY#21g#XZxhHDmj%%crhkfSgodxMH-51CH9;!Z$RL#`
zKln}WQjCZ<W-hWN*$2+^ajst^Ze+>L1~Jw0&+V)31>t<{L%U!(klPJ{EJ*tOFNj$f
z4G34oHmxzk)FTZ@B~-nHjjbumO=|CB=;#p+5`+_%u?wVCDk2%=(tj)$!!&tm5UAn2
zK~9kQ2B8wKW$Z;<WGR8yJB!-aNy4DIX4YK?d^?rx8CyZZ9C$74j0Q*|r2V30$*KSL
zM1|+*N5JPCbbn?8Rsv%yq>^8r6Y`Rfy#bZNc>MqH7vOlCJ`!c0(rVHG^3kz@zS|5t
z<8veo`{$N`<n}WYJsbUAWh{Kde7z4JK*RY-Q*gcG9!F{5y-2Y_f@dVvsq{5)8WOxU
zZ>~6eqzT)Y6iN7rdQ=Gl#JXBqZR#uV^`XO__xgZgk@5#rge4Y2X{rC^B6(NIJ%Z5F
zz6X1w1-86Icx3badBlDu93@%S<-ghUh(4T5{m&+^>m&KxRo1=pAM!bgn9jr9Rkbv8
zZFY}S+__GWFZ5JioNznf4lk_KhSNk<^9A4{=B2h^ZpNMLkDTJ={k_d&5KoV30od1r
zbYHB~$xQB$<p=|NoC*Mvohvz?&y`CX1J<PA7uMjDZ#l6jVGW>UTLol3w+8(xL*{^*
zTK>p1Sh49;n<=V4+>OU|1MF(p*`ovhVIG*f$L}wSJ@Xs?9xIL@&LY6a;n&~a>^1&X
zqLbhFSLi)U=N%z<(}M-B_B|;K2!BrcH6OLIRp8&v0ifU{C(W<V0PtNQjo`H45U&Z@
zi1puCR7Ht*u_&tDECYPOdB;VFk?JqYcplLxg8zP)-m`%9tIWr-@Gsr;vv*5J5WYp_
z->>D5>_^Q`F>YVtyW1-k-Xh9qv`-c2oTDCT_QI|}rm2L#Z9J|bc?BB8BvLwnv*5TI
zL{Fkl=F_Zdl=4^oQXME#@2d1AH`*nBwXR>@wqpxbc;26-0sLt5b^1ut3brI^uGLbL
zM|0Fe)Dc=c;KqhJD)VF74X_0erpgLjOoU<jzzDoKMZ8YgY}Y>|fu{E-gG`y12?(zr
zWuo~=5@948bIU?^=g~rxRu6@Zm$Ov>ww<$>>-(r89SKERbidMkh$3(x4!<Y&<nF=;
z@L1P})?lIZhyPU1i=`ugpWU05(bxY-cmOS#9}FDe)M<gM2PDgV05xZ>O)yXf*lhhd
zP%bo2ahQC~Q~X~uWkv>&i2Nvd?TtnD#Z~wmcGu&BI)40_kCB7RpoCC;>gmwupEECq
zn7N+lhs{0!STExzm+mySgNkvIulMW!^F4Gvz7Ve819IN7B~O++;WG7V%Siy0eNFp2
zH~?To`KUhU%SoNP0+w+%(JcHE@&BC5LPY~EfE{7t;9vD@G}=4>O+TWoe+A#QDPeX2
zwL#S*_tH=Fe;t_r2TJ#!{N(@rUQq-rwl`Eec>ms;T{ImFLoivCNo(LV05o`=LZ)^Q
z{zLin-@mK>mp9&0faIUl@&55g^#AA2NX6qMfZ1&MmjAhX|DOgE>EMCeMOQlVA3>r*
z9Rcw4FQjeD=Pxzv2LD>}yjrCmDMV6~NMbOEp%{#%coOzcND;oJIR;vq>kj~l&=<{g
zYKr}d^W4Q611T}XtpZ0Z<#A<s|9X+mTW<2>J$wQJdR`+xGTraYb@W$O3^Z>s!2g^`
zOZyD#-`nXZr{i;HiJ!SND-ckHpz|ARd|Hvyp#bIT?rN#ptf=<a@c;lW4{xS<-nShw
z@&VFX2?k5U$gy2<<7MA!?BOGd6AnIhbT&iwwSytenmpg%Uf1qN(u219MJ#rQ4+gXG
z?SA?2`qRi%=6`NB>kQf>BbigT4e3!MrjFSTlxU^^@Lek|zh3+Z!D0eDQl`bJF=W%U
zO0hhfEtbafFOQ!ALxk!aM2cdW%(u^`4_C%Sa1u5CdTN`sn`lY{14@-?i|U21HLxC<
zh>97*q(8^%eAKKtIDyTG?~^Ch%$Vlv$7R}w6RUo6n6>4Eh^NybK&s=@Ae+R1KYIH_
z_Yuiv-bHN02%8K?t03@hICC`fS07F|1=|MKMozxo%YVN9hy*(9F-%|jx?lDZUiP@l
z{k*ec&NMgwwfFYV!L%pe-h@eA+0-X=`oo%QEp@xbzvgkO^;}B!zp4|=>#$~cqaOz)
z0G+AbC1?-M0mPlfBUJlQw+R5Z1;lZ0Qax@=)Byt!W<xDxJVM0>NLmNCN6lBa*!;&{
zf6Dsa)LQd^hS~ur?Ue$V%?toO*J!|E=m+I=pvNnYK+sP>CxmVT^4~uI3_Mw&T%3tE
z54`{WQkB+v0J&WL$b06s-^Rf*nFCYBDnmh)IOP&NviQ$C6Zz2x%@6mq-Dmu)ITHzC
zp^~W$W*2+l4wZwTx$p?TC*?H%>98}_nWr$RYDE4(kv%j6+K3?y22@voIa`G-1RIG1
zhx#>=+DmD;l_7^I&8<0c?=}k8eO(|su!HVYvosw0AQzCYAWiMptc7zf*ef{l@pe>K
zRy|H4SOB_J@q=J!Irvq$EH$OW`ZDd42V!cMw%I>kiyvQsJ5h~>qdN_nL%KAQ;T4ZW
zPRrU2f;rGZR%<BE!I)LM4fQZ`^NE3u(#@IVrWArWCRhgsF;oC5Q|YVS5)hLY{s67K
zcsKo4#G`m6)n)hf?RvDFIUul>0?DJTElTcAs;2Zc8t^bXmz-qug1R;VR^`qCOEXMQ
zS<@jlj8dpckDo|kXRg|+)9kmvOkUlu*)19<-%fUPs3Ye_iX(^qM}^=l0m=HJAEM_X
zq;jhfA7ki-Yi;!r{S3bo8y@dWR+>%yMj%{-U(xG$xIL^dPH|ohTKOz9=dwGHNHK!N
z{A;KY_%mRxer)M%`duo+sRHgvo`CCqN~tz)-Tt&ol8{8yd=ax1)&U^$rug)-jOgot
z>~7(HbN-rVam>46uD{~pF~b3)fBZ%rueY-z%)t}0cui!n>6Uf-*#f2G0kvYDs2xB_
z_qJH_<Q_M(?(f^ADAm9|M(Q9(qqzvFkJ61@Qs4wJ<JbPV(*FO=1aQ1<MvZT0BxV^{
z8Vu<v+~Z^dWi&8JZ^C3a2H$e_n+rT)+a0S6(RAMvpR81mZ!3LmVbsC3H2cfU+z6nX
z$_Ve?DZmUXC?@ko9T4SvtvS$aXAI)6!#@zgtrVD{WbWksm;_r^7#mTS@CHaNHXDif
z91Bl6xmN=pk-e$zm#XUS7k{G{r!H>6$ctmW_hOy_lpfBS9^1dFB;iIeeKYM-d*eUF
zt=wN$H_ybWTdF?G04V%Px$G$H1Yu#u()AzGF0IRt=yCf$9_w=Hf8s_Xtq`JKh<=Ci
zhPR4}^-D8VDsso(I)2ROASI5N&rsfZ5uE`PuwI14G>%5paNQj(R?T;YA&g~#jUYg3
zK`Dl8?jMWah>-aZ;gulzc~m^xgE+X09zb(4p>#X;b_YcNO8~fcDTn}c{H2he$za;M
zmPVgQoQCdu8?aSg9<iu4zJ4Ej-3@?(x&ag}d0>^xbx9(BnjLyO{t4OB<cb|_DN7Q2
z4j{%Xz#NCmBfmD}*@R5Aw{@YX8P6cQ57T_uHPWWPA5g_yG$X^Qs6P}_H`Y*-i$%p#
zGs*}f6Ko)h;Nm>XA47Ug9k~2O$wevV*BG)G<TVBwbvlML;lq!IYl=Zyr^A8y+UWVM
zA$QK1;c|t;Pjoy|va;73eod218Ak9P@MPap9RMjV=`o|6Gt=~><h%|IeAQ#<dEKu)
z5ruI~z(cX{tTt6FJGgvUP<<FO1n&?;!fE))j$lQe(;n}fAlZ%am-LRnlXG^#dG*ID
ztp$+d0l)^`vwV7S@>f~=9T;a)&dg}@+?xfLiG=m%9+0@?yM|7%Xem^h4>V85sPKQB
zkp7k(;c=d&q-_PW12aNcbCYd9s`F?=HlB;B^Qm_M)OG=%e#h+7tONZW>Oh1h3!T&T
z-tR(FmeV2$t|N#AFS$@_j8xxtS<=G6?`8Aymk&8f?D2u|63vf}<)`WT-yQ(>UGl;U
z3_uWVRXNr;XLiQBdGF{wX`Iu~IvIkT%Q)nq1M$Hc2@f5W^Ei!ul6_RERk~r|j9NuQ
zP|E75kC8qc7B^u<ZaTfK_)TBuo+_{vJjpC!eHbb84VNR@{!>PhL!;n@>RZ*xn5*qy
z!U(Xi!%LY;>1QAQc@a@R;ci)4kEc)SzK4mae+#wXTKF5f;H-J(-`HJg>c?8Y|53W&
z^IRQIpKPkfDRFt9-d=(o3*UnIWkMK<Qaw%!WqMs2Y@Oclq?1sY74-H8b4T&470Y>_
zrc=s?d{((yaMHE|S##Cj6s5t!_Yvq&ePl|V_g^J>eAVY9$2;?~;bgyLa=RV5E-A1g
z71(Tn879C+*?f*K`z6`*aSRAx`sxpW^FtPHmC8cG*R7kF-Qnw!aQP{#^e6l%CG1_I
zRe#$>ZV#+H>xGccY{{AU%PB98d*15h4miLCgEi!6H?13x0JaI`tXtiQ4&7l>+Z$A>
zZ#QyUia!8)^MkiH`Z1T@;A9VW>FFI9@9@={I4XCWWkB<h4yy`6znVeU_8y#wx@B#S
zL=;j?)oL)qFEmf4H=R7KE8)?BXrMt`8o+Zgp)i*z47ZI8mH8$yi^1xnK8=@)^g5nS
zK7C0)=7YW1)t&+TvpYL5b|D-EN(z3_XaQm(Adj$!kwa*rg1@wrkA;&zmTW8SFXtYK
zytcXIWRe$5ZbCnvT4Z<V>Xzj@$PU2(3%&Jn;QRTLdtps7V{cQc0K5!i81!Vex)v3!
z@l8+QRzdc3!>>NBJ0my|^U_X&EUaTmMeDBmwCULk5XwGZN8C=+MIt6Wi<)3Lig7!f
zLxGFOKNu7@%2`msRN-Q$8EqUv-9@<HJsx;n%XC+Vm}$t2V*ZJ=#WiqIZbA&o;%G^I
zJHC0mJc=@OYQjb`l#((n{;}N>b*U3u<ePwbPovM?4J`{u=dMll$t>J|FY!*4T#Qqb
zBJy`#zf?K42FLCEW*>hnC=>VQ48Hzkv$i{*A+ds%0tzzTvQ?k)mSVq{_rI?f>0XGi
zjpq(QwaT1x2Mv`xuxZ~j4~ojGRF_qgv(kt{q2F+x<}3Y<`9uk-{@+jGoyTdbSCf_c
z#j;y*+yP|2Htm92O_<Um_9G!;PRk(IQsG>pt{2Ud_o@ht-dQ-gBoYlVc*wuMEQ7sk
zlId6Ayo4e%wFUg-i!bUz<HjoqSveGiL!By^>Ev*3jt)Xth+gl3)FkDUq~&BL2F_D8
zIC~F_Y*Ecpgg+GRLf}p$bdL-CVg5WN_75KN+L!cDZ7}LYz3w#iKq&zc?xAe?WzMLK
z_pi`V5nt=Xu|DqR<yf(tSJFK<aM?4|75T5>?^i1aK(>{K8z+eY$YbnYa)D|=jRt^s
z+UxqVuTkVIar6na-z?b|i}wRfJFfqmoRthX^&^BK8e5gnXHTFZ^50*xVVh5HA2OOg
z++Q_Yan6}@XnJjoBOK#lphwcHEDO%IsCsBh&hg+X#%tJ0UFhpzEP~Bm$?;{PA^w;#
zGVHp-A{c8U-sJO^I46oT<~?R+C|)CR$x^q_{fPv#BFVFZF(*l($47NuZSz|<Q}~}(
zOXO!Nlx3?^hT=qR4f{&SRFm0eiYG<+NgU}<bx%k@?d{O*fPMttyOE#%1X?ZTr#nvv
zJ{A;>h$LuhxK%<RV+<~yC$vK5lNweIXFZ}3iLM^5X^``X2cuNiO9;o2+u2WlA(%)x
zXlo^P#;H`$L~NzGw$5As9sGmlpX7F<V3>NnX(@2kwP%&s@9E-ixz#zCpPM)vI%jq}
zd6cZ-xiR?NBJ|;Y?4{lQ3Vi~()6yhk{puHsVe3|c?j*~~j08Y>=JV*U7g_J9KWdl?
zy>1~Yzq~Q~nP8a8c2xTDYF7e!JCQ4xQ?OzQY&T)d^5<+-D&qOKWARNBD>^67$-HpH
z5MTayQV=|S6ZuyLh#&q~sw&3_UCs3MiCb9@yYK5$augJn1=knae=}pOx^Fe<57zv;
zjz=5W0A~O@&lOV2ikM^X9)2b|eB}Ay>K^|$T`Wb1mh-55aFWEVaP*xEnL`a266A2f
zE1^}jsw2*&?y@}nu!8NG7nKe@1R}@Wt|!fy$ysD&nOEOQEk}AdxU(QmAOdiJ5iYg5
z^##9!ENsCr8(H60TuHTs=Y2}egljyZg%x}Th6Tl$f$kM7!z<atbC-cV0$#O;jaOR8
zyCn!|uQUNtzPJh<tG2--@(<!s$pV9Wibt6K_a+%n)I5>%Um)54{XEF;YNjgiKtw|g
zcnt^XlFV0`)g=kVqdDo}YYfb|DD9tp!KMxup~pjEqt4I(W@0tYn>Q6I;LRW?mMlGI
zB{mB+)kT3(;&KN5>YS$gxDGb#0(Gk|l2EZ~#ay`>{-h3$q|exr@gn3iztMBy;GjT8
z!ca{lw$l<guxkVT!g=l042ok?*Ia9qUQ+>r=P$^|L(9vau@Cje2ikBm*3owH!+yVN
z^|a|@k3jG-D8$JLKW!Hctea)>UvHhiYC*)V5QL8~x~E$y{KYi=G2u^w_3}BMR6*<x
zS~3;K=%v^pwJDRlae`f0KIt-xB1Im})bfYg?aUBG0^I`LO_r)cedgBIQ{cGKqQPbx
zZsjn+-W2`Tp|u{($aDiaI9tdRY<H77XWbog<ePcME3tQZFnz$5=kjfbXM2xGKjy@L
zaqS!TF4EExIAz+xH$fbR5LZefJH?4?kCtD)&+U`W5)lkRhGImS=?!M7wou+7QuH}m
zF=uFJio%`uNxx`lW-jh!K)rd(4(~>h;$IsNfv*kd#1rQwf&3+>+*a$^99agb`1pY(
zHqUxw!>S3-PqP3XC3u+ssl_(F?^GoAC9b{H4T`&u|Ci}kUsZIG|2l-+_+dlWAerp&
z>tf3Eruh6`{3Q<SOp(iT?LU5lO(Oj4=Z$;dz~L;7-%%ZZqBYsu<znXHfw7`PqOh6b
zG6Rxcvvd5+tMZZ*WF=-bYc1mGL&8XUb!)r?h9npAtWpEVK73w@vu_>d0=0*KhmzyF
zKg^5DJy!b$$erdhM$=25>!h{BDZG8CvE3NWR8B)IBjX#@VQ6jo^i5>{<+^Y3;?rY7
z_B1Yo(5{gTjU>!lw&5<XyHGD;qhy=!vpy=>er^iu446Ck;T~)^oRCrsEKYH1<4nR<
zq0%Gsl2(*m7OJur9)+3D+KTmdS;RM_uO#CTKt51*F01`tNz#XBud;<PTb4Sy(>QMr
z9n&Ev*Mx(a2wk^Ol7gRUGBs@j?0bv|aLIEGw?|HcG}sTSUX~n5ja^52RWmyre@=DI
z=ZxYJ2hu1Z0}XFlMBv=}i%Md6t_Cm$=d7A6Hlwa`pZmkYrj}TTF^;VUH1r3m_}nM=
zE4m#m-k;w&J+zYX-=m#Pl3FDy(>EJ>*!Nxy22<1=pGc;?b?tXTTYI=nvPy8ezYee6
z3Q||!SpIfpam6Mlcpn4VI`B-Cq=d{52pnmg-a%$lSM&yw?LOS+Ca4<iQ5fvc<fd`B
z?z1j7P*o&Y)|m``PYQb%fGqZ@5D}`+r`W;_N^I3H#+0H?MHW<OrR+)I*u~(_qZO=G
z8vSS}TF7<sK=vZ@^p5Dj<_rAqSBOcaVI&{xq`fvSXU$lATPsWvj9cSEUlG4r<fb)N
zhYKO1p80{;nQZYDj_(HD#qF{@O{x7|I)gC>7_Rp_+WQKm=$2s!D2>5*Exc8PRMgMj
z!c}x$Wx!vT#*%%1))yd;AS&*-$DP`p508eQPTBX|Uf)a#^NGfqBI>ir!E}*Yc-s>S
z6vie^C+Vn6-mATPkjxbsi0?2U^l@h~4k38O&U9RkoA>+3f4_vw^HuN{+`U2k=u9jO
zpG6+k%!)E*2}cQs{$Ak*^JuK8kO~E7cg(5zR&#+mk~8~{iKkb(2a-%GGMThcH|s6X
zW}7aa<5N0=z-H{%CM)nbFUanoa?122O&1$hV!W1%R5foacPEhnMt<fq{ZN7pfw8Ng
zWhAtC3$ki3hLHRUoB(4^k%e0g@-5S>A)Nk3oH-QX%)Y#O%cw1W92C|rT5ToB)<o%_
z9Wl@N!>yd3gd<lybVgwhw@S4T88LH1mwg?=d;8N|U#ulxfx&tn4wj$z4k~KHpE_gO
z@pMUS?lvG<Jx&08j)R$oNET=gm*%yFdl7m-K+Grp`Ym?V->yK}_gK@Q#_5piS9EMw
zCE<M}NV|-<X^fc%hnp0~>bKVLOB0+;^batIcFMha*;};p_&+Nby?FCtPJ4870Kex;
z&$|JABJYG~_zw7Q9l|o&84K{9EkY*ozi7LitU&hjKJpcu^37x~eOlp}M-^6veud>y
z_yUFRZ^^h+-==Uj33&SQN=E%yzkvY8ZoQ=6qdsTC++s1}oG(j1Bc?nNd{6mr+=GW=
z#G2_1WL3x!9HXX*P1>u&#!|N(x&7e<4HRA-75AILY^j=;3md?}TM^aZ-iTQTOQKn<
zSd2h;_*HO;mXllhFuo0O0&)P2`gKP8Rt^g|OB#Y?Zt&KGWGz+hzgPgqhyMn|KwRIc
z^!W0gaQ~?7(Cx_z{8NNGf)N;))cLbOov($G>ghVqrv42Uds#oCYNChuZT!xbjMe-b
zujWV_3h8F-*cp?6&rNgLRrXc1=7MZ^YgzG3?t7=7nhRg(zx~>L+RqVax8FLw3P)Q`
zmCwkAHTwE2wSYrHzki>tYx|#6Kd+oOFAL*_8ogeHII>$nXqzS^ue3RH?u8i#J+Anl
z0n7b9vNtR=56jp(xUbl81rf|ey&R&FR06{XxGd*u+VY}aoIw=2j4{dkDtk*P%Q;cb
zh24mIc%aG0*t)JPXA6Gk=#tGNO)`Rvn>vgFz?e1L9Ca7fUt{Qv=T6>?Q75AHWJ8(l
zJifs;#}zT#i?s`+%K!5s5%%U<0t%t?!{}O1{msO7k8987$lIicq48n?^Wyq7_WI46
zf|&%PgkI*75x1Mg3nwkhBe!)r*DXpbP$_dQirsd1jKXI5c9iC<&M-{)aOc;!(Nu7t
z)Lh%+;d~;#kGh%A?{O@OF?+$K(XD0OIdd83T3l8w<d$`}A)u@VKQp5=r7^<{ZM{Aq
zCQP30^DW~S+<|u;w1}TxlN0)JteTjn-4zr~Sq8>N1ue$+xL>-DHxwVZr#4;}F`QqO
zd$|1Av}w^@yGSg%_MEsVesHT;bANC>gtakv8jCMa!CnD_K}4zN?Ua7(q5}llUA>qx
zZ&!@lUGAZs8Li}^jH&(A>nHheu->0*M$o7FD*W+lXzkak4;6R|J;D0zPtpg2ZFM?L
zMox-6MKRgGVmAK~viYt8|LQs-CKW@tg}Hoy{cC7PoE`k732~3Im)}=R_^&axOa+Bc
z_Isq0?Z1YKeifEeLs7M`?9|P@4qO`yE;FRf^nPyS)LlQ1le1J*xBHub_d^A%dxRCn
zhmI#IV^6k6gK_8kf*lxyh3>wK^_%odhc2k0uwkjer>fXEtJnyto_fb+zEy)aRJ93K
zg*UHgOmY%Yhu(d{#QXX+X+vPP#uaa&BbcK%*e6+?^CpYCx3F-LsW967sd?wCOn1DS
zpe#`nDbGlY=nA1gc{(mnCxt5bQUt9MJoo{2Lu0vFwP&sti15S4X?RpAuHq4y`)cxa
zvG;i7BOlj$=1*vJ#2CV{8Af%^RcAl7yt9D#PayIJi^})E!euQi<gd(3SD$LmkAQQI
znOuTxa?Ld*HKLJc&A?C~Ojs@q={gKvfU>MFhiUiKi{UF~V2*gf9KFUbnUw^Y{D7?f
z^l-iO%20iHHdW_FA``oi_mr<cmc@oX56F3r`16zu-jg9sXW&B&Nj*r~)Zk+rLoI~B
zYLyi%(w2JZJITMhsn|!=!k!Y6AbI)&u8#3SClz0(=9Ie+d&$r93Afr_3%_&|<h2CF
zF{85Orn$|!I()6V4NMJQQSBSSra1J?#&a)wEqN4=h*~va8?|sSL+ul!(CeD7kGADA
zp9c&Ye#^G*U1uBS#lYyUl8La_;5X1phPr^fp6ieyR}2cpvtN7M2Ao#yOYxMlvY2IR
zR=OnRmvTb!{N=Ykek*SHkEC&Z)IRd+efKK^!gSwQWF)54PHy31QHei^k*^_=)EpIB
zB67ItY2ubnF6hJ0b4IDsUW`k>T*!P=b%^nKh@iKibUm@|ERP{u?64fqrc2XQ2$#OD
zs+@7U46jk%K>>caGS!Bj&Dm2^Ypt*41QZFY2jm9ox9N^O&-KnxozYyP;F|V9qfaX@
z0Hl_9^4afxjWmyHCu>9sJfRw4x%`~Uq1=u*A`|lE)e_y(XDwg;z(2m%X{Fzp8AH@6
z<VKhgd`ikbcl05AWWYKK|4kM0MNXAJ%UxR7<~L%0n?OG$1Z`yI$t&><dK-49bv8z>
z676&af}9Z>3eo0>7v&1gpPSUhvP9+yBfrnO>$xX2&4&n-GY{KuJSER!yNPA!r-~@U
zkvnLERPFuV_{}Gw-?v}d_3DPbDtsT-TN$lhnqg!Ssbuo*$FOb)VS3HokW<t64WkF@
zepgvdnM3jROEInQf@M9QEdl|!K4w;Hp6nH_*&jHDzl2`%Z5jb#uUY&Dwq%I2*zEK8
zXEE0Vmb$zwh--;F!x=Lx5#Wjxb0P&-1aZcydl46DTf`VJ`2QaHMfmU|2U;3T(e}e$
zVR$QH*)N?(utAkD?>bo&cq#djh2Q7Kyo&tq*9hE`4mcQ(@GL>voY?c!JM5WmLD>G$
ztgwZr>lT~qrJn)_VRXs!|A)G_4vI2<`~7!m=`JY+3F+=mK}rNfy1S)>g;hWVmQrb?
z1*BO@dg(5u8|m(5;oN-V`Tm~Y%$YgQ`R~lmurlN5eShxz`owj;->>UD)jy~%K4D^U
z(kRHw|3+STYUBPY$az@uVj_bK^lJ>R_GWrwQtEWqbvBSCW9Ojp<q0Tr<9Lv+<!%q@
z2@cw?YoKpXJQ2Bo|G_zN8}=4oxan}+h4Bw(US;8yo*t2LFm$^dkReSl4>xF1;U%_O
zF`n4?W{+gl4a(6zwkTj!0Yy9;&Bd)w8RjBB-MR!UM+crEe!>+AL^*EL2`Gp(+V(h&
zn=(@CZ~sDPMa&ASGH>6P44zea78>$Z?bU)WWV+jP?NfMW9FDndzFcwZuZi027>l-$
z!|&4ixBVk6VXnuRn}jRlJWSU_U2Ne6iL=b`(3FoX5KQ|TlM%HV>OL<8bFKJ2hWLc(
zi(AL=o~*^zU*CSvTECOQzYY|tlM06K5<_r#q7-C3YEQASNHe2~nOqu%740O`f-y%h
ztm9OOofi?5otFx&2hYy<l>B>Z7l(hJ!xevi<UqQDUz?|aoSv&afqV=|MewsJ>+Hx*
zD~W-hel1ZY8>s>OodupsHKzK_rk$5~WhettM-=HREL{RnJj5hiz2RPG)^Ag!TZTCf
zeLQQ&cEC50)z5(@>_VkV?tWuJR0gL@;3NmNttRjr$pY#rjEInda&0h=((lsGg2TG_
z6&JJ@ibrEddlyOcN{{BIOkjOZID}E3ff2-qVYQe46k*EY$DmLe$EusqNg=ZJ-H9dc
zX|<%=B`TU%I6Cg(@I(1&-eGL?$T-b#x`Wc%90STY6D9hTJ2YBbdRPo1uIG<iPbWo-
z;YSCY!}{<KKe!X3Tw2C`QsCFI6}05m<nFzx338=etE7asXw@;lqw{iJLcJRaeiq@@
zf7LAvzYg(_vE*1dI4*8?Ov+E%F?ikCxO3UkPl6<!f1C-5_6axJ4}2~;K}A0jeZ-TZ
zLxwQLZQeVT%@@CDbKFFQe@{iZwxakr=SS@KGTUM6FLAiupHVCPLlz<nj+PD1@-Le&
zo(D4yKMf;r+{U<!<Q+2hpBO)|?A^@@sD0rqVTUkHAa6Px01nUxVdq6ov>5K3+CNDH
z$d30$c&X)WVg!|de#W-dH&$EV&zLv#HScAR?Xl(7p#65eePT#P=-FnT-}$2pqlF3z
zKZB>(Mb6*rbh4k;Sce&5{)+4*<RDk3U`4qN9okT*%t4N1&0vWsPz-e;vfr)-ty)|c
z>}rg{4ZVsq>@s@#TN0ZGZ+dQ~*pC#SB-)S&{iv|)o&ic20ykexy8r5PorqmW`U}&h
z%6~w+bks1+{lbL(&(F(c($9F8HzO7}v&!i$>=YwH?#IF2+ZXE21d`W|buFlujK>RJ
zC^PIRjo^fv(I5Fgq3?&DEgtxm+C!zG(b7%ZVcr?27geU>0n+~Mu0a#p>)G||^JTy>
zZcQq}F7Bjdo6rqA+G%SRtmWwk5;E)b;r>a19=iFVSO^8Cl>kCyrDZeusEwerL^=H^
zI(tg|__XaXLH+iVg>uy|2?Y7@3QenwES(}$8aEIAJw!|!zUVedNdig+bHzDR%<auF
zb`ij?DZhum>sGxrSA~z~)f73snALVdyjRdEo(alASE9H}d$P%HdKG~}A;x#%QvRN_
zQ@@5BPVy{5Vd6y?Be}A^i6{t<@JOZYZ>@&}kf#zT1a}E8hVNvP)zEQ>{L~?^C2gzw
zf*w~tTz;LVlGc=K(q+eeb$(+2rLAI9(vV)xWO<+Lq&7V|bm#dHt1179Rr4{FDt~}L
z@iQ1C^MDB{(Xh!*B1$g7y~JE1u=E?{hgL2l=MlN33tnBKULdT!%g27a3(LmPnj}nk
zm;Ht0wG`9~HfAkD*0+YcH_d9td+fI6u8a$TPFP2d%jMMh*VrpB5^79;jzAkm)54}T
zevWe6nSgm|!tH@Kyi#o8E<3RqnI=bS5hRtRae9=eWb`f|u?`<qA_(YoK(Gs>JMy~x
zXx0^$(Fl{iZ)7P7jZ@_W(s^BbOV3ymo*i!S^qh?Frs$uWq@C<TSaP21pip<4hp&7=
zr}o21n)A2rHvcwk0lJf0ONsRxb0xsM3_c*S$668_)wZ5>yRVo#nQCmoFB1g{{p059
zSN+717+Ylu7=>T9VKHczJGoBA&9rTfpDhFc$xb0|&qXA$2hS)+90}a>C+h7o&yeAp
zzsFW-JoF@w+}`L2BTQlL`!ki%f%G~y1ZxfI4l6+7u9GsSN>Qc7LDi6$mP*`I1)DY}
zi_d?6x6dfu*iGrR?sS4Kg}ClGL~fI6#h4rk^yu=ymI5eA(8KZ=aT)*d1Ks$<<ix2#
zbAUABTDfe~g1Q-lzO&f@ymI&7#ZLT4y6mLiv1V|!zuR}h<)G`@KsUp0K$*ORz#-L!
z^V_H4jS@xY&KF$Xa(yw9X~vrHbNck)m~tEiq0Uu(u7m%Iyq~NQJRSe-AcOJpL+!5B
z$*1IRh7r%!O{Se{PbSt6EeVE@lK!5x_tUcy$W{LJ^@g&6yRrIo*ZR&ua9FgkDY7fk
z@M5laN>@nOwejv#9A`jOF_w3&cr@@^p27P&-N9dywZe;H`#a7pi#H8714@og&P;6?
z2F`oBvw@xzEOcHeqU(oWVbG(wlA0AAO$leI^SHApg-iZah)qt-;<nM}Hix^T-%KTc
zXgr5p(8~2#9z$$x`16wk3l03=n`W0&g8x#1tEr~9%kq7YK;Fg%jlr@*%dCTE=YC9#
zA}2picSZ@;P~tzYp@0wua0`({ZnCw?QBJg$3RgVDf~`Fcg>4(1qRt&l=4d@}H)&3&
zP94iO3gJS3DTTA0>~FA^5IuTPN;Na1Es*o+{JqD-R0CKt9d1T|{BAbzgqsN^nkY7f
zsB@Ucy@TjZ^V1ze7npGtjS_5GgG^{54DI`DIe_Jc`20ug#&cHiD&oj^=Ls}AGjJl|
z2+K%?=?SAYEzBv>+6rf7g2LW*I%NV2@w-QFu;gZ%1UyIO&PDks&b4F7LcxAq*(z0l
zarCsRrDuEOcOY9AEh7;J%xT(&NiOHZr&INXe9e&ydtvvFR?@>3%$yCyd_r*5H$jV5
zU)5o(BQyB$v#OdW){HTLj{i4}@6rd_tB?u#8lO=qZbJ_$u&Vd!Fdpy3V_px(jA01F
zh7Gr+z{c5!`HvV?`2YbCj#67fbNCpHLd`oz37Dji>s)fzXnq18Y+ps^O$5X7p0|2l
zk`UGofr(qYR9;K%k-$^N)4XTw>pa6mPQ7-qHYI8|59zL{JypE#+_*!cJ&(<YKDN{`
z%@R|CW0QcQ)La120aXY^%uTwQw$sh_K}xiyd^iDTA~JXv{Zu-<e)7G55?#r_o&Qt(
z$BYOHWV+KqDza{L@~;A~#jYSRuXca<?GpxUXM$Dk@a@8jxE3@T3p@5ii#}AZHR`^n
z+K=Kj5!}cS+|WiA(8PrEy7=lF@GaH!M449$8%y*7pDW{^I;PA3M(CT$2Zd1AFA*MW
zuP92(F-Wb(V)Qs&aA{ahYi7(7%c_Fi0ix5O)zjht?bGP5qgF*P8HwPp>0`0pDm~4g
zOqdSSmI^k_R%@m2|H4tu_qXR@B*&NKb?K9s`V-cJdZy?=#!O$vNdMQ;iZQVz>oAeX
zdB(c#q{c-^CVrW0p1B4E^xdQsD-le0_#3)MU41M&X1CaiWjC$Q>v?0uqd+}*&*<px
zxu5~w?$y4S*9uR09~-<I_o3nqnZ%MYS<??kOpd!SE$IdCDU-h~Y~H2DktCf(KOVX~
zTAAOQTLH}bj#j)s;|iwZ{p?+oAiu(QgVkloZ%Eg<8dNE}l+yRuCq#Vq43t}9*ebi1
z8bU*8mg%iN6wlBlw?}?~KK(X1DCX!h>$qt&hcgW0Rr=|v((*ieT4gb@^R-P;NzHFj
zmK$f^Na%QFZM<}Qenl@ZEB<2^e7@vZt%l8@v4cn7J;X)SG^WIDI6}THMH@oBglG3L
zjn#028f;Y)k7g3Ni$4E-fBsyiRdJj*!U%sjG4x{FpS4G0ivx!Yw!5BvZX3S8cw|>V
zHxa@+LqJ;?tR?T!K22F(`GNtPQGBC7>LbM1+xC?KwO>6)=y`Z!<3Faw0E@PeM;KS3
zSL%qM?TbtT#*@rHJ#u#6@8N;dNw<CR`O$ZO7u37gwHUy6DSya+GKHi3^P?kzZh?B2
zo_tB+`N#MG><wSt!?<h@e{<g4e~@!h>?Abn!035Pwdh`ZI;DUxH3Xw_Z-eeY%u6bm
z1JEy~!lLN;xQF^SrySQV+}IILx{g*m6emTd$U@eJ(hKWIOLQ_&^w;=cr5Zjy%PyF9
zH^CCYJ}Slls*_3w596o{?hqEpLGh7%NO4sk`YkwnDV4IdNTgqz3G5BxqZxcrqoHM%
z;KXmv6cM6FW|pGxmk<7>&Wl?Ob@NgAI*_d65y1qz%KE3>z}Mg(--Eos`@Q^KH&@^>
zG*(qArdRqqv}uH=L?A1WDU(CmeM<VD1Y$3>@4rMKTHpNAk3P1+csBvblWrzhhJXKx
zO-V9MTSc#N+voOnUNn;-{9P_`etyS$*Ba*C@0c5y$Qre+kXRt9l&$$LAVH`tmf(;$
z$5k79i$r}Xxzt3xIE>wpBw8eZG8kySk{;J1fNj&(lcqio3Yt%=<L&aRd3i`qQ%dYN
zq7ot@IA)#UKIwx0HNFSi!TaJQni4gcz-QK3cyLjYgpnXIQQTtlv9L?pvA>V&n`i@h
zk+p;pt#6p9AVrf_XbR?Kj&w&XrrauZ-{bA#r!jd<-9B;4u}4N0aaafr*vx5dHmN^s
zVr!OnCg;%^Yxa@KA^}y7qV-;<=)tiYA2pH?QLT`w%Y%y(d`$lJ*F|4+J8MQK=vMYC
zVI8yDACq<Md*ierAo=Q?4QvPATy~-4!AT=8ZWasL9b?gB6B{@|OX4DIaX&|E>pvLn
zz+%iXZ^6V+A3)mO9x788_k~E_caWmxQn5OCB%^`z75Z+3^A@0@;@v5!H7ODoLzt4y
zl4OcCzmT?CHu_2&DIQmX45gkvj-PI}i1M6F_d30q;`4vmocd!;#Y7)``yeTN4xAQ#
z4<rb_<Ju3k?V`;2CY@_)y(3-V=+q9{1sE0A(ND{Hwt%FC%Y0aOCL<z$^vuAF+!FQ)
z$PIKC6!Ca#zPkj<w(5J-9BJH!JO>MoZA7zN&Q(hvaU8_0`#U7mZYq4~M{MI?d%;xE
zWUF7q)+%Ij4yGu^aU6D+CZXSUj<l!nDBWbq#}ik#N{c;4ZL|jR%g()RRU}}cH@8ru
z*JG|0r{NOEulHu}8m6R<{1A)}J<;0%IAP$WBuFl&C@bT-hIHWov!yVn<<jfL23{73
z=<b#3zr;EZk{wD3bPLz5hPLPZ2`&cWgE#pVnDli)`}|H@;@xSUQ?`q@I{ODMk+WeL
z$P&nX)j@6d_du&|PxaZ~lkuVx6+;QlA-m*=B$o?{-z!E~o;4_}s`E91z5Zp1gsa4Y
zgAqE<xKmXx6tXHTUe*vBx0`ql3f9Mm76**++-5@!c+u=HJ%Ej~NTm4jDkRX-aT3nN
zo^e{&$O%CYvFV(8^=sZ$$W*(A!%!V|-XWCg)Y5nUib-3pL}9Z6?l2kg+u#inz~asK
z^$D<RpS)wYtkI7p+nmXBl6|I)Qc~<xjWWeP(0>2Hd3-A2YrqtrB9KjVtuZE%k0z6e
zz4zTQ%I?`ZC(h<EbU(MhsU%}Iw1tN<uV`hHZhP1N6>x0Y(6AV5cfQf8=|rDB>QR-v
zm{#pJ=_<&0Vkl9e`{8EgeJ=5A5dV-Jw?S*t4a*SQ8s=#={B^A+u$W(;Imk(3OtVZW
zRTvub6O=oP&;D9{R&&ZZ`#YMGH)5UO7&v<(J;H+xYB!0@e1CB6F57<7M2z*#{@Oq9
zrrF8|j_g<NT%T0g<BTe6*-Z)Hrq;Ff?x6{G)%D}zhW%fTYfht{5QOvzev8^G*p#2{
z(DS!IxHRCEY{^?2UgN_h3lHaOu&uc)Vn%^uxsty?7M+wm<}{03SAi+!edAEV4-F;D
z8?PGUSwKyzD@Drlfs7$G1G07k?ER;t(m02KB^U83`TBAthTH#Ttz0Q%<xTRrSj_bS
z{70EcE~gkw;j#v0w+2OK#PPmv*AI6gW3Upx%>1#%dY+31^xi*NW4-USZ9vD|@5wFN
z{ms7dVa6Kpx{~`Pc_0R|%;^r+Sqnt%lQ5cWI1Z%BupYo7RZDHo1bP@=->tGpAGvO%
z<Cw{v$Y^q@2B%0f^;N`oN-K8$z9Amhx$<3*ujs0C92UEVNL!uPU*~Ev3JcAT-a9i5
zjRvFrGP-692XbRsu3P;3-SwUo{w);gRF<&g`Pjg<zP;#R>KqiiKYsezMGe_i(I_Sr
z@OI@LO>ZA9llFkU^cY?xdF%5)nD#DhPZII;9Cp%}-tq=#8}gzfO5JDRZlTO^ReJRc
zAYv7p3=A_Spe1eK^(sm!jMlc2Tek<peFooVt87=IT<-V^bp_CQ!n{e#|3(FFYLQ8Q
zznRMP#S3*8!Z2uD)E3&kn@aDu`a}uIUTOEpaabJk*M|m@8)hhgs{x%`Mmz(_HqiXJ
zq7~!u8!&!t(*!}Fu5!>3j$n_}a-yZGG|R5N>c7|3hi|-0nLnT)_?Ia<_cmcMXaGLX
z#ov~Ua{s;B$me=KQXV?pp7@78on<WVZr6`R4iAyLH+g=h%LhKv-jVCvhS%g$E60T%
zxdRtc?nRJ*->9uu<KSEK5zy3AZL8gkyVLyACj9a-ppHIzl&0f`@#qp&i<Ri^HQN@e
zS?FWRc(C-VmZxEzfI8&oUIyl6`<tiivER}a5k#=3B2XUERq&mbiq(&AVj=aSmrP$W
zpGWJ$s8%Dtf`2$ulE_en^O*CNwgB1?a<^zF<t)TN_H`pr_tn!odz+>}c@{xDG{%RZ
z|Mct{!@5U}vIczLJ3+6P?Y$~#0lpqxCw|tUT7?QpIL;EVjBR5>d84BD;_XbIX2td>
zca9d|xk7sHPF0jhr)4q5QO**_wW9pCO5+<kzH?I;?RI-<tpk%9MN6NqdXHPEX;%Ej
zc>)LAP>M|E1{3x__w^~-qeQuk6dW@CV1=WPj%eq%ejJD2+}7mug>!ej6@7T%IBU03
zagz7AXPwqz^YQDJY&ykHZoUNX@>~u-N+9T+37<a%>e!Pp14X((YAgTO%MF1B$df6{
zZ;d;rdn4y|s*(Y~u|hKhrrm;n=f^$dcai>eG!U@<>~LCnr;Ur~GJVnF99b;ywS8@r
zz5HgHPt_WtoVrV7Op+`%{F7Y1)P1ptKc?PCH2VN-7Lu4FKT-n%joL0LDzaf546hr9
zO=jlzkym2N46ajn7z06xa{DA_DfivAa&4&Ple5#s5|m4Mt{hXf4xf8n$(UXsGV$uE
zt=|O{=49SC$<o`n<02&;huxqpE;!^B$tt9hYnY1OR8qnPS2@^AF`We*OoM9wj7Vt9
zn}?-TynuJRR+|h#zqI!hGBO-_0)%0I{{KG=ZNGaZa9MQ@O>iOsUZLK?dT^x$Y*Z^*
za)Q$9EUfIQPCfdl!7NWrAh}}V57}t#^;ZFM#80Fr1G@uIed&-?3u=$`Ug&nSMyiz0
z*8@Fn{BkyQFxJ&tG>-y`uO!wfISZ9*Z-6wLU1fb>r0fy(=4%^E8BCdIbSt$6Yn@`L
zxWUAoLda|rdjqk-H2Q#X4Xyp-+vmjd!>Xr#XT6sM9qan5#z4&V3;ZkQG@&hQO_;gY
zEdNTi34B6PN4-t{3Hu(@>bLj8mfktT|7KJOI@h634-cdDjHTtDr14Irt=+z)J)CzY
zFey`=Or}U5J1<3DSO0ZQ(=bIGnSgRf_XX=Ct4sHHZv_`RuCnLdK$)>>HtMF>*!nDf
z_ie_T`SifMN3wns;M>t@w=|IN^?<EZ&S-yx*izHQLXp{zMl2<3?CAcoMWQ<wK%t6`
zsIaIQ;KAh~KrB!+AEN6ENCs}HnZ8=V{1n?50%)l!?QS-}cfK3`i#O?|0r!3S&A-8X
z^58F=a<A>P2-d)NJu4j~ny)c&w0k0mRhRQpT?$nh*-QI~7eko=!&SXfMj<@AnWIG(
z$!xgJ%5Vb67iuU+tPPFS<|Dp1r50dNO@FE0BtrN3qDi$Bb0m8z)ol1}(-kgQ9KZL;
zlOCL_Vig1AY+f;j^pba8(?H@_qZCGEsC+JvFD=G%OqH@sqOeL`6krxKhhg%DHDC|v
zvoox+Hkk+6c|ss2;nagLnEi7lLD%NPW>~$bJM$W?%lZ1r;&5d7)=Wl$6I8}}$ML@+
zmyM7Jr=Y(LMXt+i<0;v%HWV3*(-c}y%A+??_|i<z?G64K=pFHKUjx;+z`(RH>T8Vg
zGFJb%CuxgIy#t(wLB31Ck*{t(L2m5>rH6V!Wnqb&2XD`LL7PwIn%5+6cR_!p#v;CR
zOx{@gD0GG4Mq!ikY-f4M-+idT=h&($oP@<;Ro|EIz2ayaNuu((h|@9ES|X=}?gR_9
zw~0mEAfVdKyVgv{CzLN;Zt&XBZW#dG-G<6k(MMN>lD}0s*x=0cJP11uOpFnqSQg|!
zvx=6hNfxmG(*OJd%hdeI+jwB1&8+NC=!yS+p}5A-hduN@KZxOkq4rc<6hZG{Ws;4n
z;3F#?$N5K$F8m+SZ7uJr_({Q=OBEWuWH`-=kT?JZ-I`#$5(f(RUQafF6r$cKRg-kV
z6)J)rs(^6u3s)(L_SrxVHBqielu1@DvPUe@tJ%bK`mkMOb`@w={xfMe@-)}<s%kYr
z%aYWMoR6+XYFdxqu()Oim<0(!qW>A$Ke$%U<_`k(uZYEL7mQG2@np;ls8=bUp_~Hk
zs;A9^dJ-rM2Z17?tHCerXFwjW;>rbZ0XZ-Z7)oq4dp+b`O;HROaex^)qnjqDbgqfc
zaz`ucpYc60*d-HW&hlNN+zyED$w1wk8h;8}$A_jN@ln=}j`2C7*KEu!9#<2irgIH6
zz<%h~<K@-{pwRkp0eyY{Z>4aT76N>CcV5=L+dhmOf*Y6Kq(TOq&QBVg|J1MFSq8PV
zpq8KS8U{`H2+su+BYTu7#y5J0F+0EuE0Cu<d!9>l<P2%|;u}s|4PEc)YsGi3MQ>Ua
z$*VG^T|E6SD==sO1O|LWcKP6!KX$GA@ZIOGZZFRe(zn2<eF;DClypCq_s&bUeeE{S
zIK22{n9%&7nZ<Ryk=)dQ*O*sHw1B(C`XsBEKFO8~yj5ZsbP~3+i*tW)G{HZ1w$ai!
z5-q(M>Agc+<1|0KzFf-Pa1<ck4&yH<haFwz^mDjWR=E%BiB6#3E=z!oK52i^Jt*d2
z`SM+@_!da&^$R=&&r$)FuC09*NqFBzVNvPt_k@0&kJa%ylS&A|{1uWI7#1MMX*9&A
z#arpV5b#_}x1V>lh#y#D3J`daRdwtAoc<hot8od@TlwqM3R+>&yUz@%z=N7%;?_Z#
zKZ}8(D4pLqEq;=yMrBNSd~(f6TIyh_$H!o*$wuGpya7#qPSQfVlwW4NrY!nPAa%JU
zM991-pWG4Nfn7m=yM*~en}_wW64j<g-Ii@XN99X4hTmnHUxxryvw`z0;K?wQ^(^LK
zS37mI>%ray>{q6>(QAGP?qX`I9sH-RgyV54(9<G`jV#ZX#ejn?!I`FKQgc}11Xe>L
zkhd2}yf^Idc4isijX6qo>BR#Uu*&0f2_75R`w{MN=AK%>6uLX75+i|{NhYIXvY3a+
zvK-T;Zl@R(VIv3%sKw3QS-=tH^X9olftsgDA0kklE&GQbcQYeD?8(^xtzo?^t`T`7
zQlq5l)8F39+{m6jnp{65zkU*sTp|R=Fj|+-MqQc<&NnkVBhd)B#u@eCeZDrR4SKXJ
z9)P)^2|->lc8TP_T*YBXks2}@6r%aoec#@yDmrB%mFi0eMKR1<VX%@qCh_E6vaONp
zUdHi#l4`Nq>SLS?ZU^LKCMAwOEl!cb`gyWv?CVsQoA7?FA~?u;NKDW7li%mSZi#Q0
zfHVqIZjE{~%raPNxnFSvP#VSM+-?${+Fz3!j&1b%1v?OQ-+wQD-}@}F>Wsy7-3!9v
z4--_9zxIZm+sQp_WePPrsB<`oEsDjPJBSu~ML+`Ui^D4hrLJOBoH)>A*31^opzA-+
z^HDUO9;>Doiwztb)JPM?hoTFB0s|m1^*L(Uk=dqeKeXyTNo;+6a>Cje<L%S<hnwZ>
z9|~EFF73X66c$flMr@{PbM(z<j-qL$>CBcnePQadOf?hI>Cz)TMF?km7x3#=L+WZZ
z!ARcpf`kvV9*<}RzS12O?Fq$&so%XBYe)GUr?N={B9CJ<_`E&|z9LNjFlY(9N?UJo
zWv?s__b3}bNpLk^TI`W16u0cx@64PEZ@1~qDhw)_hGEvPUbg|Kr~Fe>%P7Iu(TIfQ
z;AC1!asw-(>S>;Yh~Js(1bx9q5hj}7dUp2-Ipn-*Xl9yOL!(b9uOYa}k{F9#0@P!a
zUID3hN>n2`w^!o+(fHpembu??d9QTbo0Rb>&hu{1FetTn_~CiLUSaIcz;t$fIi}A2
z%GeprWOfG!aK24yJ^{9oX*7h{@<4iUY%9%kXw)y}Dy1Y6DYZTXcupY3O9y`#{G~qg
z_FZ<Qs~Hj_Q;yf26?eM`B$8C1n-kMQfXV?)7BQh&W#jF-oXW-PJ@TRAiobg_e+qm2
zz=wDJB*}6tzA4d?gZBB&AYn^qrf<~O61Wh?qW4Ql`wO7WvT`t}e3N|P?<<&;tVtFm
zRboWF{Li15%26pc77zMi9`X|u&5kAgtZSSz2E0Pe+l`&ApbQip_b0f#KmI-CK7509
zmN><y8_k)McS?Wtrg^tb-}A=Y)xNw~ID;m8mh#%fDN0X0UmH6}o$1l5eZM#_v3FTH
z=%v44MR#*TvR)#*$k$);3r5v&RKLz{dJp&SxZSMENXjsN7hktC<Ujb4K5dwz`SkDl
z0FDM<<)D<~PI5(&lMA`A6OrfUs3||mP{W*r@4fuGYY?ROW#APVBS@0}hNe=jc_+D~
zW_p0*d{oi!;ELrlr82F@N70p4`efb7H`4)D;RI@QLHyjedk5?2qS5nqk|)DK&6Zr(
zM{73Yf`Hj1YIbI~d{!us{C(ii#jvn);L00F&;t58FTIhk?8$Mi{y-%6?R?{9ZQ?*t
zS)+r{)oK2KHVgd(AoW_wx}g)wxSS&Atp73@JUnNQ_H9jC?<(M*m75DYy(>jOt#7D|
zZ4E>ll|<-6J?9=uHLv-IlLsUK3y~6(PYGnR{*WOhVn`(-gxPSTG))7R)hOs+Sn;m#
z-?@p!j!ZKv^j}hpMblA~L4x$|4wT^nwoEh0M3%3MK3i-GTJQ*F|Ek@0e~Q3iEcyZy
z37pxF%xe2wnZHLz@lBbAAf=XdMhObJ=G<1M?U>5;B4eqJvjZ#9Oq57m8rNj-YDyf1
z5HVK9eRh5ur2P)YI<AyWvnl>yX-L4Nez7XSs1&A0SBg9GO7CRHKYJG_K<W2C>2hae
zc*<zV`Z#>T9Ia~NytE-!cUeg$3Ah_o<#dXX;0aBTjnX#;Mh!6I(lcj@3zG$H7_L&<
z8Asu0MwX;k;q35nGcWOlLbA|E`W@2})}HHhU>qkVQVZmzgkYb^-y%3A0>9g$cb4_X
z$sEVx(XwYlr0d-?OS`vGhdcaiiL7|hjSf-Mi*mS$;BWujt`d}g@L=?w21dkEE!)V!
zKjLT+OGfJ;a<u1jtIBLx<M9H)&oJmq7*sb(<TD;bHjPI~MolR@-%dK85N7JW&c?ZE
zit=`Qa~9AJEe0>3yj>fUQ3K0tzZriFxz>cjm1<mt0AfW!&Eo(S+z(1Ln<el=dtNJ6
z?vEy?WZ>Yf%MbJs+QXkw845H7!fijD?!7s48<hMkwcLys6qXTLv51AGaf7}R>;E@x
zG7v#Yp;btPs3Bs29y44>1<;sq%E-`mqZA4OCf!mZ%qS_81M7g++O4nwZdj{pz{+E>
zjQRk%uDX7Cnx9H(u`@B&k#1```@%jUF}$Xcyhh#i4F}N!4JZM^D6x}u_|3hynA}9|
z$rGq4PjFPm-hLmRYuKm!f_WR~{%Dvbj>41m-N{$z*Z3GwgZBHu?*xUW#zDxlnY%H?
zr{|R6N|{LP_u>4VWSZ81K*NmbZ8lMU$a&dB%)R8JxQ9IR^=GIfjMB4O`jO0nN(Wjc
zP{gwDMf;W6__~IJ52a@0Jgcl--@XViaGUt$rSACJYwbOW6oM%@jrzsD*Y;-Ve#O=I
zd`U$^O=pu556DZ;TE?FB5c_IDo=#B4Z(_{ce}}0fDSc25^&nwKOT-d5GdCXTU)9mx
z74oHk5?~64-em>*9RVMmlrj`umpA(h+qp8TpWiH!Oczelc=S9eEK;$e`yMv(BX+Wr
z*R}bwllys#l}S$?VO`f6nOIEUUwTfKW9G&~AV2JFu%k`x`O^f}yyvT(E=ARLBkvn!
z`UY%D1!gULDPyW+qs*chrjV%6qC`i!Q9BJ-3<rL>uiQ?mZ(h13KZz6<#ZLLnO{obJ
zv?AHXnZiNKED($L=l1W=I0EQ!JI)4;w#@(wWhX5j!gVfaNt&!m4k{e_JSQiaL?6e^
zMzUg*jY769pA{gEQTBlgrg%vp#_cSVA*1|P3L;B0UFdTn?om)FtTHAY?>^E>)%}-7
z-#<zrOeYmbtO>*dgQn^3%dE?+E4gw$<l$HeZdI~7i)fykX~gejq1G_3GIwFr(Civc
zaYnqFf6yxD8$IC@KLE;ItMdO`-1rc)mm~=M#JWexxD9lWksp+v?zrg&KCo9YNyvy>
z3{j;ksrQZ0yLT>7VP2)=vhuJmBZJGLH0T}nR!1dqp}($EZ_0)!qu45SqDOdD=?-7n
zfm+&xQWDFp&4kI9bQgx-yAlG1mvCB=1~yN8Z&EA7BNFDMt-RDyUA=YYaMBW?$6!GR
zL(4zi7_3WbB$hA>0*kvgiVsbMZ~20F_tS}69(jrhz3+QAI`^76OY_mMCcz9qkXO-d
z7wIdR*6}+Fo`HL0h#N6UI!$S%9X=y_dv&iYeH}OR*xLN-OjxXJMSdJm!#(KUoL#L<
z2YO=HvDI=YXfL843pSD(lTL>PYEB;on{V{x5eAZ}1!cVq-X<}4pX+Ezyu^L>@jnaT
zr-xHVI^<f^^XN*Nq;qTW$Fo%PuC2lJ6(8O|>i6lty>2*tQp5ux`m_G@hL+Dx!to~U
z=WkZ%@2pC%uA(?)eiv5`T6iz|vz*-7FY-r=A{M79zB}CA{gob&Hc7E!Zs9Ws8Jrx%
z<$vjFlk8<viI6)<6_w`6i+5e<#Tjm~D{VY|dy;Cp@Z_MG`C!n39C_Gga}}S+TU6Fm
z=(o=uBv?D=+t7U7t3Dyf9u&|S?PQWHFnxQyK0AP?kNhohKS?phCTTN3gYqs-zQ6qT
zd4Ec0qR{|H)nO3sBU-B4i!MXJprRpp{_-R>)m8Wp&R)&cZT{f>?&)-Pcf$e!_%=>-
zh#5vgY4S`gow}aqw*_HqD-(PC8P-Vevk=I1it_i?eY!sI(0hZkoLZhci$*z7??;w6
zLwJd9@6%Xh(7%0t{zA{5`jKGz4J$UblJj<}=qGF9E?QVO2;xR6A|I2{3pD4ovWkZ{
zYtCDFeWhM-pFq8CGXuHxgNZ<aAkuzj5E*4mPq1g2C$XYN3tJqUfT;nk0oG-^`{&A7
zu}mIqf@73-;j&;`1|_y{WiH{~KL#jW+&DlZ-&pnLO&nJbs@O3wW{AV+qd&pXV(~B|
z6|#OeD>j(ZwDHK(|0QTAA5tVfjc1K>j$j|+3dhLp;+{c$_DxE$;di(FB%tpq<i(Uv
z@pQ84@AT;jYi7}AeJhR1^L0;dtQX9{qiK)-6SJ>Hp-YPqdc(SC2K!C%0_4vtvn1ll
zHd|{Mt&OMw;lB_GaU&)l9gJr={fzp_)uQ<DYw?gUNY5xcM&-4J;8dMI4eyK@pv<89
z&64s@6Wj`L2)S^kfmMk{KVHv#?D9)CMkIE*6Xv;&-{nX589yD(^_b)Y#qzTP9c3o?
zv|b^|7QbV8=T(o0j~|Ez)y?FYd2YLVhB^$q&JFqiCyO4EJ|$b{?E+hWq8l>*F6=FT
zjMzef!o1nkdZ<B*%d?91TF3~|$H`q~sP8QH2T{g@&WcR8XhNL+!kPmiK#~PeH7o4@
z@kV@*{dh5I>s}Shbe-^#!_AH-Lj;owpKv=WpiKf#W#)%$HKyldPa02S6O`N?nN7cz
z2LCKefn3*arx@9Z_R-(8nBy!KQQ9)bpzI~Td0?71EBo5yE5)A5kaGi-A(v|<SZ?LA
zwjMW695!m2Bf0}KY^FGt@*Z+`;oXv4(bBOQ4ivV}NFLD3uTm#!Zc0QjCe?jKGfVG8
z3$$<<6#G-jCp;W>ct>cmk{It<s{-ZANo~_iCa@#e4s*c;foOXMQMqD9j+T!<kQ2`<
zQ$WC(Rz{gQm6Z60DBjpsexPr}aU)F=-nqLHVGD`|V78-um1oz2=VUX0ama(*GEt1d
z)WVk45JA{iYk5em%s$By8w3^x$z*Es9H?1I8&e`0`y>+!xXq}yU&;mMXjGxr3tnaB
zESr5I=ZMa1>U%hW=reawfk?@5Mklj)EkWYJ3g`Ob!7ZDuO&jz((I;=NC9NODP5Vh>
zEI!KsGOQV{ItkCi?|QRU6D<ZhD|~If^QuZ(oU<Q4KBfGD+j9q+FTZFFPHrz>;}u#K
z-mKYq)D!BCLvVgYelQ6Lm;nPGp_o!HDnfK`mQyzDso_y(m)*jp0)}~9jw?ae1Pa#3
zAb)8o2X%Kky(R1mizk^31d)5nMy~KXF_x3*I{BPsU6Eo5WSk^QV2I6TB^4v7R4pTU
zN1+zh7bw)>%FAk^V`X{h++CX9gT=uUlAk$A=!8O*Li(G<S6Xke3m=;xWTPA_DiF72
zl@3X92c^ePZs@_F_=g=2n0Bj_Rrg3$pe^3-bBq2n@qW+-CSF1n79`S|>}5Z8FkG-Z
z?Oxkr&eOH!ti}&neJ_GD-vn59dqMc`9xb1)1vPH;B-v1hPYLHq($gPqEUbLyN!IN{
zE}jgTe!opoIQ5rh#Mj_fS~fs46X{Nx9ympzK??!g#P@96h<5w@9is}lbOn@vyXlR=
zRM*BI(dw{%xMW$f{W?|Eu5Le?FA?F>Dnl*MNB&_x9g{frc^vfHBcd5<RCH|2Al9cM
zd4&7g7M0-}Kxb+3UkdXO2I^r#yw5pmBmF+U?({2?swVLr4T>0$h3=#nkH5#cFGVm4
zP=l=Z<PEgDwDPov!2Vi6`I%qK)L2FD3gvtjrEP9lRdM5TtkmeEq&B2eCA&xQA>c>a
zEZZ;rd|Qg~rvq5|fu0+<wvdGL>5bI)czBzMpa~=YckJx=<a+jhC`7sm3Gm#@V+jPj
zp%^pbWf>=22j(MrlD_GUpqAgvtf$t<I6e!fRY-dO@5QJmP?#cHF5T+E<7T+!H`m|Y
z0X#+~i{2|g3t)H1Y+hASHyi0^>}KJGOtZ<=M>x_kQyT_ivC>fWw8Ox@BHNnb@jVjt
zm1OmTZibr>M;~R(OTk-u<wrqGT^zR`{h~y9Q<G|a7d5Bv4o;{pXs)cUe0xNIGzF)=
z%f<I>OOwTmj2|X2s$)jeO%VMm!Cfk6!+oiE*;h<wsQG9jyHVg>+`%6%Kj${j;9Y&c
zl{7xymP?Sxeq9~7o1}_U_4LV1PV|FqEsOTHlX_%UR4<lnt8Drz#C6pBQ1?1XE!c^m
zZ>evsZ-qXN;Hv10T$i2QV)Qg!aB*1936L5t=2cDoU0ayD`^Np?68@->%&%RuCtB)_
zpr)QRKK6JP#xz|ZWM1@L=r9katoSA4wWcsADs_0LZ#V0N6)q5v@@*}6&T?0qTY9Z`
zZvR58QTp5DFV_af74+%P??>qd8Z_u+@}E0=rCe!`|G;?7fo$8_44_!RL)IU;W#@|=
zt^u&QA2OuU2XSIypQ@yhB-dIlr#;0kN&4x8&FX=?zESKHkdNnX5R_?XJ*X{H(~4yH
zk2lQ=4Z7?g{#wzx7@<%)KR{0|6l(f}QC5G3WFgoXOKPF0ByULMVwPlPb&$78Qy950
zS{*&Gzks*au*%=VyM5xUpYNhPL%pz)KT;>cJiO}#tAF3Itq-^!N=;zsyPsi#cd<X_
zXylmu^A~?@RCb$m+r3PlkAG$bYYW;koqsj4z9ubQz?&j0b<+0&x!K~9?rweIOb2A&
z)_RHRuohyBLX4CAqNMxQPyN&7ZTe>7<knB#Sbl?#zZgm6v%2}3I*F8^rKNPV_Qt=^
zparrRe(DQTeLEQ=QBo_f5=;P%uY~2Pa~E|Zq2I^O;X}zw{K+Yl<ddaD#dBeO;b^Z*
zGDu%<aG=SwXWyUK;$}6)TVjfrY7hqPe}6K0Nr3_xGid?V;baJ0@CPn0C`kQhza3ZD
z>`eT)j>0!pHf=N2Y>>7>_?J)S>i}tNBtA;#6;j7ooGtcRfkplIcQ@?ZKr)T?B+=aS
z&o2f?<Mg*l1|7P`=ms5PUZ?ybWr$#(^(UJDULYaGAi1CT%7sz(Lr0m?$3DiJC;=tq
zgazc?Tk0HS-eJ#1M%QM~9|=f>I686JmPQWSUicVG%J^TlHq*)(#(3s`U9=hY1HuJQ
ze}x5M>Ny)ZZd4pwe|y|N2qXydvu1+;ORZNiwu=kit6pOCzgkOx)>-g@bL3*$Nx4bh
z`eaC7hf6(AW}8Ia<@++H;kN#uL&HBVmiZg=SvYf5o4>Wn-Vq(=@9E7P&@2*O!k8&t
zW2kQbi!JtY+^yEsWa|&*U{2rRf|}(!-`UcFReKn@HuF3^6L8T~OlPTAdk+`Q1t3d`
z&sBcB_?K=a=_4DD9fF=H^L*hdXpm`o|DsfUpO=3~AlSer^hdQKz<#4U`L2V{IVWeM
z>QwU}jxUnFTK#IX2)S@}eyCRx?r2IQs|?&b$p!&WuZ!MW`ycD1BJoVxW+AMHK(nzV
zppBAgj^TqLj`4K-G?`h40l1yop~D2uvm%>nV}}@<YGo4Duh$ABZ<m|+a?gEZ$hVwM
zwU>bIV`C*305)=?boM^Fc_yZr`dOgh^^8E&=;(N!JuJHPJXZ+#Y8Il!6I5-q5va15
z6U&t%N=(nHdxkJ&MgMd2pNf7Q>%(m_{G<;!2wNXjrV~#4nwR46Tyltiac&*a-u-}g
zR4V%u)Wrlmy!WlEy<Z1wo6k7g;}q_Z2u!KWIOGJ-(Ma-v3pvQ<_WVUQ0!w=dkjn^u
zWme|{Ku7+67a0{!@`8Xzl~+6BmcadRZ-)Q7dxM7ha4>%dIt?qCuH_;Aaa+`q|M#~w
zo^*QgV*PA#VY6WLCLhw0?E1k^d|$Iyn0Tp7w1fb0000~K@2@9Y-iD?C#R#vP-^n&9
zKfSoW|J&s7Kl{c7;$X>j%8(~s`rohqm{#T^Xp6S|08xPofH4Ba{bBi!9rb_qry15*
zP>}Pg3dMgOUU1GsU`3yMHK&1syb2N){{8m;6O{CS{u3(;_=iHZxSLY{%gaR=2OtF~
z;*uC+b^o8c@O!>{fqS^C(g3gZxF+^X@o|_9HNUa6blkDV|3{Yge?B}9f79+^tgGV~
zjsNpb;6|fN93ITKk}*K)Sg>DO=LzZbw){7p{{QR=D@8m|9A16XSp2_vj+B(>08mtj
z|IbVQ<aoWeKlP}XHPf9+o39;C>V_dh4>tSiV-@ZwF6>+>-ujtJw(1wiLN?C0zPNup
z^ARKLhY+7-j_3HV#pa)jjv00q;vn!sykrZ&tO|vWiRgN&W+4CvzAb<Tq^cZ|!%0S`
zFmhA<>0M}>E%OYUX~8u7<yO|ecfLKOD#r#AAKy9-0wg#)pr04lN%KWhnRgd}>WoM>
z|3ozaASm_4V@@p|s62w$FH;11oq$f9Hb6a9NSVf6p*oRI^SehM+xnP}rXXO9%D)08
zp4hAE<izlci&|rhQI@;K`<oyPnJJ(wF}>J#CVKGzv|R^?Y2$V83V(j`5&j-n)y;6N
zn~~b^>bzI)Rrt$z`Up<jW?&V<_Qn*E{=!y8_s|i{5BM(!<~)a;Q`#Jt9``)!AqKEa
zl6T01pwtKE6;p3U^SIf=Fa<wmA>;h7VKO5m1spMfX6e3vA0Y*nU;z5gm9SG~k%kW&
zFa)>o0&qm#qA`HQ_bVAdrx{MFi}~Fibq@lxV0*Wr%dLW(=)y+6S$Y<(n1_yXPBqgu
z`|$h~peR}xF#||1X8-taE#GS~c>Ci%Hm-!cububi2QbN2YYw+ZF+l<w=|4Hvng7iK
z@ORm{;)ZOg*^DYuI2(th-EJyNzf#8X3tYxA#p-p42)F>4paQiE?;VTw9{`JW_jfmV
zL~KUX(t(XA&*#0w@Cnd~59rftwryX?Z<O!SEcMMq8m%n6=vMgfoRpHS?tZU!Q7DYY
zL#0TC9=WMJh%dft)_gVSTC@M;fZ-x{>@oV60KJ!G4+s43E&wOm_(=Y^KjZ$^))PQ0
zrvQU6bqkEAFyJGP>|5+<I30if2m+9VkUy{1xP$D|1CC$t1BAcSABiS24<JmNP%^!O
zFQKh30CEID3IO$Np9839g&RX6D=_0lBrpI@bniE_N{8e0sTj5=Tkfi56J;jKwsUu{
zN1wc3%KfUr2NY9_6MP?_LjFK;k}y87kMlWva>4(=Iq)db^uf%~Xe$SLc|FxFmA8~@
zirm(hMD<d3pc~ZVIJ)Cd{A+wM!vcyxC0`>jumQ->TR(Xo7WQNROOeH{z?Lp*$0W(~
zsrJTu<iP=3*&71>3}E?#q|7I_+yVu30!V@5#DKB%Rj=k40eI7OAc@yzEnZ*GWCl;A
z|5pCnU^W_75Foh~RO6^n7-eKuqO)2kJKwA}CDVI{6-4Q6_`>*V7ij)nbhU8NtpBr9
z3c0133Sfka4yP$S@-jpjUOIi7qF*xC(Uhz<aQOLS0BANkuTQUKCkeVV4mj!*Tev??
zY$@2q)ryAaC?_0!esc`hR34(5JsiNK^F=(AZ`OekczsCcu?i?~QXocUTRK2Af4BnR
zbdCU@l=lQkFF!xH?_iNKm%6_i)<3sjvq^WsmwZ_3HdA?Y|GhIc6VP6=mA*+%7IC+5
ziRY>MyOab}?L32GyZ{Ya43O3Hh{TS5D6Qjl6#<iEy*5<J^Fe3P7FPN9p+jo_b6Zks
zXsAKCNlk!=n%0bn4GDDO9=UigC_?li2ymN>iL^V|W<P)Z!X@pOEqEIOJUoHL>xtUA
zlJsZ=4>N1>zFD5!NUr)GnX<-^oQKItdxZ`o1=zqaA5;`vSkx+ywMEI;lxkK!B@E`?
zJNc;zDLhW3_gf|MJZ!~G4Z2;S|8xQvgS(2GueakDx&03t7Llb_!1ll;TH?4zC?7H{
zzEyj{>IOJDDDRA}f!ElTtL4^1KEvnH1{Z!;<7*~;#ADtzef0<K;sapN@x*H#J+0M?
z#zD7Vgm}2Kw7dN1E=fQN;8+i;aW{<#s`(NP6*1?%Sz&*N@98P?*+?g`nKCcvvYkF9
z2)SdLg3bq8xn-okQZY<`TEx~GwSg~_hXcBd@sUJJ1fKb7Pze0f`x(r)%LAu>rYbDi
zQT=Z(F#M!mc}RO4-JT3Deu>%QhYFCtG5avlG#q$dxefq>>6k>XubpM?ftS>0=kXXM
z)wxC+GqoXn*9VyW>ebNGmrvPHA<wRg6-#Tr0*%&Xa{^o2xtdOY*DT)RI5CCMdMEka
zU2WhI&48aFR@_3eF~O>X$Dn)2kUMa;XnTY`l>|)EWSe2~#{lJ%_kkO}Jz=e3l->X&
z%W>gPhSb(yL&HQOY=Hv!?uJ-t?TiDv7hl2$lJMVn2Oc<b;(T$BVK1)#0QeQB9#2&@
zj(53uNti$vQ9tz9@gK6dqWrII$_&rV-^aM01L1GvEDf_p)Hme;z~Js(U_-a<y9l6n
z{UOHArHtm%{p9}&V*M==kVLsHiU%qOz8S`XB!C;~1JSF(n&K6M?k|GwCsEQ-5LbUP
z?s5%ZKD=$EPO~sYFo<1Y5DnB}^nTuJa!Z80pS}UZ`|0VU^;ca_Jt<Ef%h0SR!b1x<
zT*Lr2CQggI!DW*ZuM<q-dYVr6T9+#725Feig@rZfD!h9EjS3%FCqBjt_{D)&RKoOH
zN<NR_ngd5B0)>r1KARp?g}R!WD7W@fTJv)3qU9cfe#0Av;>#kp;8@m_xAGl*RiS=g
z8M_42Y_2Tutf=0N+U)dcBu6%p_K`g(O}-Yz%)rVs;u-f%RUN>dAKc2%D8=zEUpo$y
z#+Hhh7aZ2>zx^=0wIO|9Hvzu-UJ;5%FKH(ENKSTaIWGEFPd4KzQQ}BY-#Z*~J>L5`
zR`zBW-=H|>sEzkW;S_6sity47@QuIV$G!R&JMSL^J$Y7!Oe}WQ(z|duhtAf~dJ|xN
zEukd3F?q$i{njAPR(Q$`wW05M(J9h@3pFPgCHt3QxR|x|nY8&yv53U=tBMDDa)f!N
zAQvS!T(^MCQf^ifEN;;EE%5=Vd=m!1^KL{TNPzACn=kTr%7yQ~q_Bo>hO~R(6e_+5
z!^QM-VYU|5eqB1}4t@)x*Y9;)GBii&4x=8Z`VsBz5zDXa=MUt%U19<gg}v`eHBG&z
zt!EOg<2^MqG7tEft=6)exrf%$7r6c3WZS<r_~G0+^7j1AbqN&3Vrg9HLs>-YOoZ$p
zwpp&1re4WT)5pmYZhampsxpA}U4E?2J3tC#RZY{T^>DTHU?R>J{&j@iwPZQg++_mC
z(^F@ZLJRao+8uU|&rn{?HgaM(?`2p>xHNFfq`%3CO4o!PDQ3xS@CajyB<Ds0aW7(V
zTT47Y6|rnIOF$)qfY44u_B}Inra^Q}_!8`KnPTZw>yL?PJAk$UDc*H!2|8jNfRait
z4aZscV`3>_6`5%P+D0Yo1to53IS=EtcdGOv53yGR)@ANutLs>XQYM;E2_)PoRrDHQ
zhcfOdc0uO1pGn3eJgYgS-P7KHm21;mU1IlwoVj}xzYN1kuZF8R;97oaHN*WuGW&ff
z82yx{z*+V|-s6bwX8x=#l@7fptECkOjY&TVQ+f^%OU~FBI#q;{ZIqL-WW;wmpFf~G
z(<A1XXT8YQ!O~ZUVbV^c{&AXCN6FDR2B6oM(f7g7C{9S{Ag!PBsfz*HcFTl2mkj23
z=vMFp^RN3L2l<I(N|!n!XG%3BK6Cq1Uox-Nc_ub)Z$Gj6Kwq{3J1F$+Go}5uiB=JA
zxTsgZZ|?f2lPK}Q@GGVvuW_jlO&HCCw-tci+i=|%-1{c{uW21ak1Tvjb`c=yo^VJc
ziPHPypMYv8!1}sF+!MjX@qK+YZF*3CF$(hrPPI$nSl6T-roIv~MD~O{g8glvQ}HGN
zoTr}Wu|bssI-#yejCsz!%?93JPR5=#uw)b4p#{oNkU}-pFvtiqngFcROf$@hAuh6n
zBX4A*oFZsLe9_*Y`M~0h^Ruku<lXfwORVo=nnEjdCzSQt({Bjk`HfRYfC5%ib$lnM
z7;`U0bMS5?F4{pHy7T_gX7|W@EESEbXUcY%Wm&spvNGK>ETb!x*YQOv9DdVlEJW91
zbRSK+(_SWzfLXsnYsQ6E)6Zf6aav1FYgQOt;Q`{}Y6baFhS!tv)FaQjEQ9rk))NXb
zY`2>xf5%M6h*=VS7jQ~A^I>&%LN4HMpcKGOV3hkM6@Z3j_upLrSs=OS;?qCU;Tks9
zf;u~I+inxZ&Mfu8onqiS3mvc1jSY}I66JOoVaD;xy-mBV+|WRG-dj_z!ysHpCwvL6
zs!q1%@+lKO!1Fs4bXTDBGFI2X!1DGJ5&le0L9e%tb5(aaOOYSvC;ogyuR|qeMoF5o
zo&W1Qt#-N&aQJAxJ|d&&Q?bj~h2^=TfuOy0e|BlKt^TKXPWey;?cYLK#v7@wQs`nT
zOc4{3xX&n)BfoI#i`;P(DUSDsY8R-snGas@ei~P&^mUIW@Q*N#F<(LQ&T2k=RXs=x
z-+p^Tg-;(8jfp!sMrs+bHqQADKk~`sYCQV^T^9uUoHBEn1SER09*}Ne|0BM*_-~Df
zxc&X;Hzv8#zI<qCX0U?TE|0h6l1=)LprAcR(?1&^2QO4~!DI~=ldZ%#&3RzEXZ(Zg
zog{#e`$n@<T&b}`7yV^4jZzan|23L6oW?c|%yevS81LTkVvPC6BPiqb8{sDAUR!6T
zF`P}-QDxnQ=R1GkgH`b=B;=W{dW#)DIkm-$72?W?n~F{t7ibZwaDVykzpR(fajZd1
zYv2N7@}3)xE@;LplP;8%+~-RKnOGX#*=SR-==Shx_ff|88F!zZ!uPNdE8kL>;JQ2j
zDsl++q%!9G+%llz7n-fdc*U0=kIcxY`NmaGD|R&{O=CJFwke(;1!hnm4Hca~LQwUz
z_5>4;>NwwWSZtyncGJX2;1zuAga$qh!esrnS$wAnk?dYT9U;6ysbN~rc7aNhH$I$5
z9oYLmUMB@v>64eR$XsTJ3=)}{sbK#ro<e~hqQ?0+0L?gIW`8pTlz6LflZ}INsGw)2
z%^vxz{)c&Wd`O=y@smIMwTppz*+LXwu&78NB+xh7*zx?mi8ZKcFeCQS_Ud`~O7-SV
z#eUdCFi`G1Dpq*C$4ZQ(c>C1z#K;6#GrtO(Y4{3=(XNR1x1Tf*ORffh#(Tm~W1Rl_
zgon63v+!M$sPr3a`t-l3`U=0O<7VBZyTc`yl<v+2Bo#qGO1e?Hr9m2(ZfO>z8|f5;
zW$98<T43pJyz6`Jx%Zs^fY0)q`OZA^M9pr+NuXR;Dj9>lX%SswzQLP#z660I`?ddD
zE*gV~Im|r$D~r|)Kyd%$%W}wnV>LP~7zSyfjJ|J>ez}-Rx9d+eWM4;eSn3_-(T6_T
zQ(4Yf7oTytJDLL+z5^-=4c$Il`LoQVF4Dh7e~7d0NEmcM@&4}m9hCG!6?nEHqgSN&
zIt0Z!8(U-=CB1hN9Z+IqiNU>!hg8zg(h`L9Olc=LJIO|`s*o1{Ot_n)GxN2<$?eCK
ze?=?BP5vF0z0=ERndYcf{+U^M8YN3*HRyN88JsK4j0v=UgPxO?F)q2!9e9M~y^P=c
z+8@P*FZUq%Jk5Wk7n=3nHzF;fnvNnA^XVuR$lMf~cV;ALvn25@u*hLdv`jbfE7N}M
z>cO;gB(;BbH%Pp)QTPfEuOM`n#IP>kACrFQQ&|ISQ1Og`6w6&HZ<E}Fx|D9}K9+P%
zp(lqI)l-+4?4cBbCGgwzxtZsbl9H1BiV7A=YacI3P}Vylbrd!wKQ}&IWS_Kx9dxDo
z0P~SfL_vjt-<&;724(J^jUxp4m30>(1JdulaqqSHk?)ph_JSId-uPoY93s(3+%KH5
zKNI*+<S-aIlT=528!$?gt5DzIhM+CUi}kJ|orj^D?5~LQZ8J2p2$;yLhFTQPclG@*
zrfX@(Fnz60`$%7L7n*-qsA+_<D+VbZF$%Vz-FQk1zjQ`^hp>p<pT1{>%p@e4F@kvJ
zJbXVAg@&(@r`{=VH2X-1z%){eYmlj}$c8|x{bYyIJMyZ2%Vtoxfhmd9W@+Jj^e11N
zUQBgi=^_rw5+qb&uz&h*|JY6HUSVTL7q)(q2f;oI!ybl=`!+ZMfdI76+T8gc)oDz$
z46{9hIdPc@{`Yj&&(2Wzb*z4f-oe@{XgfJZXzT$QW;D0+%)lJ!c+lU}#l@utVK=Sz
z-vIKHIzhb-^P|t4td$~3N<)7uTW9iHe9Dscb>9~-)U`(k+M~-&w(sOe7cQz6i#&d<
zG8MKA5bb1M`VPdobw=+rABaBanReq<Xd}VskM6DTBD4CG6s@eR(7WGagVD8s1dbq8
z@y8$a^}2v0MN&t0F3|!e=dj-{(_Yv-tncDUb}IE$^j4?@o`3>hfQC4>1M=2oJ&tOY
z|Du7IaX-HQ_CDA!$;x^=g+wkRNmuQzrYiYow<mH9$#&Zuq=JvQFefP*3U1uYoNDj*
zm*#qZSwPany{d3v%^H+ne+kNg1VkHezFx{Ke3^-!wi4G<E8?csx*A2ID-PBeDa}}$
z5m)3tR{0};EEuy^E>1W$O({pZPY#`tKD^S)qC(+$A3Wk8=|3s2CqWbA37x>j9`5ZQ
zNOVR9!+S6*PZsHW9A7We*TdgLnZ;<$stQB&%o685D>JoyL$Fq6$5o{ZSsqOJpQtbJ
zmRxoOGreq1auSHEV%Mb~ak(Hsp}k4SV){*#pLW;^se<d#zXDP8=bo9~@SaB~&Iqpz
z;;crs+~CkNniK!k8)|cOvK}ZeYg+R^;kru)`o}rm7G2vcsiMW{nRv}1W9>Ex1~aMx
zgdsR?sMIJ^)B+6DhQl@2NN)!==J=u--Dk`hN!4MVPZV;oNjz0epea4bX6Jy%g=Wl0
zD<qb~&z#gDoN&-c`lz(?X2HGnkZp_P%sgPE!hl)%6|<JxVxHTxRC#>xk>E8p*rJO6
z?;BpkfPuwRQ7yJA=XKX)5Je-sz(;Ue)rgQ49Ib7Du+`9qm$JVR?gy)a1;WBcsZb@~
zWc|bDME>cf<%ZjP-q{7DZNIZ4JU_2|^$MIIxih-n@FR14>4ejeqjmx5yB4681<LDv
zFik~V9A=)|YqeGUj#4QM?lWRNi#-!r`1953cPh{F=)W+PfzW~nX}(}j(K9lYsHC*>
zyBtLgy}U8wmDf>B6O+<k7_LXs&8pkNPI=9}nAgOZjm$U~CIvdiK)x4;+osX<&&_4h
zHZv#Gx;rOacH3aRqvxb<FE82uJUSK;71S=fkMR6KW}@hI#n~mwOo8gqu-8Ul=|&Ol
zigdU5^sxkTIl@j#3eOeuJaKO0o(|X$GcHg2!F*e^hA4&<kCZDfbACoF2&}3Wr-=H>
z+@Bz!I7jQL6p~)0E4Hrm!(q=ng<9pBvF9HP6@ThT041tkCPz!yB4wKT*`EPbm7Usm
zkPzV8L2C%bumLaHA9**+iZRT4h&bztrq1Zpug@6KRh!iQcvkfEKjsx}k&ayMwq>Kd
zNhomu>2LO<JiD~8%hCPajM*?*oWE&2d*3EvUx*o{HSZgoQFSJrqJk(3a{pmSi`_FL
z7y;@zW0A6f$XSKTGt)Gs5gaSDhz;o2-xv5ZCE1W<RwvoH8Vvp^ab<px2zx&I{zU6Q
z1~m~WpKeeqN2XEV{`01e{bYiY_qr%S5Q~Ygkl6(ZY(_|9)BHeUdliB2ii8~U2fd$9
zx#`N<(0iSWoIg7^Xqk1$hy96ij2eTqb-g4lNEm;dSFJm=CEHAMW5Nr8D2l+JXdmA-
zlZ%W#Ysma@P}ffgzpuN*i&h?ToWk-|DWa=bf$iZxHSShJ^bs-3*%{~oE-cfwQ4dMi
zj>wPz>AN}lt_!X^$7m<<$ST}mrJX{e$tPGmM1}N9^#Oep1mm2)JJU8XQk$`vn?Iwv
z^7@#fh$(_<4`(hJ&z^sQT+bI%y0M7{iNvPML1*sGgbje+fNd@_t=k@$%9p^Qo*GU*
zJk=!_WutPlsn~AIXs9k(_nQa5lPl>UHQ8~(zK<zRC6%^x+Y(~O{6kg_g9!{q1TDA4
zyDv^WQ64LPj|6bxvAOq{4YB2ov8TuJex?1wV|C#`0@V4Rf_i%f3`CVJaiLbr2-625
zqa0f@u5!hXb6}K}1tbnAa<95CNwc0n@p5CJv2zHN<fpjTth9c)KpKUu5bpS_WIPsS
zC^l&(*0pkJE4se%;+{NQSGbDFG7VTsZ>Yw#28v@~Dj*t({i^iL<ZTPj2a07Wl6CD7
zIKoKZ<*zFJtP-uqZ=*3Of)duim^8DRu^3T?*e^e`b<>Kl?G>?6G}3FdDbc=AY-ZIA
zlB`A7sFSXJrHG20&LI9BW>B~st}5Pz%%HhH{#Czr2W8)1#4J*MdlXWYxD_}k#npwq
zIxf!KPU!S{iTD(nU;dH>P0yKRu9Q6ns?Ge=OSYTgs5F>atN}3*XhOV{(hJjl-PNoD
z5uDxD-*BAl256S7%*QWkqIx=9QrvJ<uE%PdhJyv#-*;~zeSRm%owhk~TWjEC&}y4k
zgaXfZaiu?kA2M8oT0D)C{@GoM=5^F5=2+O6v<w7nEapD>Ba<KY9^|d1{;=GMKhiv*
z)n;jvdK3Urt3r<6=|&PRhxO>IInwpL489SLabhsgGHiZLbB^{H5!BlfExh?LJ(%Uo
zXK}bxth_alDd?12I^_8N_;$x}UpqY-IQT%MNnBTYhTVKsV8(>iD>lH9n5|^r6Kj|N
z>;#9K{Yb`g$D#Y!sPR^^c|f6tyfpMm=T9$^z{i;{cT||rCv}&LfYde!UP?yUoL9UU
zaisUENwSy}i9Xm*c>f3eh=j3#d+u;aU}vh~O!xM++%q_z;WqNlS?YkEE*)6O{T*~h
zXoMHTY~)MNwLVpt*p`Tvm@&X%@uN-Gpcv>YtaS7&s&G1$r#B~ORJ3&T)iQMTLMQ@Z
zp=d1^9}1H2&5sUL{3HL>Qr3=llj_3cs6Nfal+-R3uGGm~ARHL(R7_>&C}ebzH&PiN
zpbVCyEIoiF!#>UaNQh^Rx8=ecx{&n{oFG}($muwmi35{Hyai}Tj+bewlD_%cjPp>(
zj<c(O@4ArwThWWW&Y%X8mf=!M@^p6QdR_N8`E#^V*Q-uPLqchvC0;XQjNB_+MM$q8
z3~49eOJPWn8|xN2hTg`SZ3)XN6qFJaMpAaA2KkcufHS)cC(kbg?@+BkPt17n+>#%(
zVma~!2s7{NR<x9F%vTg;V!V0dx851@Q=-q8yO(l{?tW2$_%_uzZau3ZJicB{aMNfK
z-J%*qJyZ?;lnxUf<vWQJT$)Q#p)Y?-kd1aikwQt@MrSF#ol=2;Oi1WLRO1&0epb#W
z?!BrpgANIr)+dp6!N<G;cEZUtV760kD-QX)Fga)b>uaRz0e6m;l&`7ja#--p!~$u{
zYK)0Yfy^Ptn%*O=Qyk<c&v%fZ(31Ukp~iE3-z2sr)AWVB=y4Tcjz6wU44dGP<LMm?
z&!WI$8Up0xs^WjKSgYXFUvI6u<%wflxYRR^saO_9WJzXuV1>u=%FBuMb%w4L+-`DW
zTc2{6Cp{;IJNor%oL0PZRmL})s#VK5=2~e~>_*Zig>93Ej9t5~uclizj`JtBW<i#t
z{239?tuYR^9aFgQ)D#-`Zv!1cCxIw*?@*#EG+50Kngb^B>hXk3*MB|6pFA5;ex^g{
z#H<S9HONi*UpOKf#~Z8@6>P7V(c(d+oYe(na^vdli+W{yj+chLpz#3wOl(*2-1#Eh
z^>Wo)WqXXA5ZCc*Y^k_6TuHtM$ZJsP(hTh7;B?}eIQPO&119q_X3!9(mt47g;ZTyC
zFQ93Q70a}baUF3OkJKxSk~H(Ru6^qKMdQJjnohd@Dh+5K(GVFL_h!51bpaQEA;*@K
zhq>jqcn80pm1|!srqiCwR;Ugdln>LQnQlX>L6A=@KWN$UlU3(0LH<1ckhZ7R)u4Nh
z6_!uBHl)KZ#YFU=Z|Vb|;igxm|8$F)QrFYDT7mr{IKjdu4C9I)?F_F6$TChrX8VSl
z9f7`OSYDQnl0&Fg-5(GoGL~tDygi?kFECOq1_=YdYl>g2e_PJLc7S(0u0Eb4^{q>;
zC{4-Om^?wF9=^k;>VtxVu6LBZ?`gT3hV*;xbhoosAgfLJ`%AO}lC-Y9D8BiQ9Z^tf
zGyw+d=6mz_H5s@P)$Zmf2)Kd5XUU#PN-F1Y7YWw40PN5gL~+Jopq#kGSQjMB91FD8
zZCfYFM~jv|($1r{4Y=t0Xk=UL!k^1U98vwj`~&jImC?u%7r6J^|8l6lCUSv=4}$?N
z1n$smq>-=&`9VZVazG<Tli807tZm%w$HE8BNa@59m#TeJ`43H8gwJVbAOtxdFYscX
z3gbWAajg*miXJt*2K5cI{IdGb&i2yMeUTgY&_|{5f%TWcVRlxdKlXHN6^W2Ci=C<V
zOYdW(<k*Aqfie(($enUl2lEDv9Xpx=Ds#^7P9jMzDAoY-BN0whHKkNhGDYGEYVP26
z=5sm*vJ8N)dqM=$qeh`0u%RpE7UhGsbF;A}p6MN?ALpfoEL!-d(HyLz(mV*+Ao)1y
z9c5=iH!1^tsej8`6<O%R=s7oL;Q`=trUR>j-o8gFo^R6q=;Z?n(7j;bXBdHO(r(PV
z9%RxDds!;?n-Hvx`71iV1@$Z_fFo7nmbQ+<^c|}Nmg^i(;K-Tc&+n$5f3ob-Wqw}K
z+~Ed>cgwIox7HneE{SWE<rX&nqt(i}K(j6y^9fP<pFvI|%N|T=KNT5Lmskrhl;|L8
z23hkAVoPF}S4)=$JKjQ06|<H>-tp}*N)fvQq;%zal#{C7+*3Sq>;5(zgVNMYG`BXD
z{r8X8FD9lI5{IbJU}Y!I>|R;_4*A<q2T_VtYG4XTw7cye|At)m-^CPHnM-W30x)5(
zxOpwOZJ!zOQ*xkr6NZ?>C?_L);bX%9OccXcid>I`51mGc5>4{$IlMI@ImzZ+B>%mM
zU8*Z#6y3gM6~|nvnf@mH)rst&gIF)9fzU>A2<wC+iT#o;hP1|=puw4vf1`UnKID>x
zvm6l~QbDx|c(Qk}CRxDw8n)B!vm(Hf5cEx@_K|Z|lvYPC;(DrPJ_~sK_Z<8e+APy-
z%ez#RepTkc^}fA|T1PfaV;*b0ROunv4%zu(DX<Dsin*6>^)4~$0f1G8JAj;y{ptic
zls@P-QR`|oIC!pf<rBUq-&UDq`7>WCV$mm(n@Ghs&Hs|t5FUUDR_e;}I4^ieRY>qN
z8cEqMrhB>E&dTn7X`<C3Z<M$xj501iVH`ObX_R}60CN^VaJ{3a!1iRtMb*7A>cd=#
zu^-?6qX`sDkjs@}9O$ump{*yfDSxU60q6KMR7ZVsCx!?#dkVRp5<oJ1L*&9cjL=yk
z`LTf|>?FR}VR)Y`uKclKRkWDJ>2cB9XV4trR>K-4{ys7(uN1dG1Pjh&-P(y6h+o1g
z9P%3&#6N4BiwC%P<L;Wb!<tG|q*8aMrr_Knqp-9F8$eOZ-<CpB-IR-Jo_Xb*vtP5X
z_`+!;Ng)8G!-a2h6d?9Zkqewui92+S@15aOSkMkft;#8dMob1oG+C-2;G}wEbW5r$
zgw-;QG1W|>j!yZYh{fWsKk9>kX$&nkF|LXZJ@~3;8y=E^)s35fN#=nvbr}oDd1^#D
zjvV&+1IE{xfm{t!Wb9<;$}PA3f1m2|M}A)5MqlRYTo@GrZm(kzIN-!;7cZNdzoW>z
zw?heYNL#}IEq>Rn?MgVlyjfKWpnW>Z-UrLz;@pCe%@QkKt+z-iasv3sSxg_4wc=ab
z3C8TfO~W!?umbs)V94RpN{@oZ6L&Lk+G6a3fhG4@3Pbv)3x3;K3(|fiHdl|64*2sh
z2rs9YRX&WwhC_MP+rt1R{Z4O?v*$JhoEJ4w^!X-<q}2@lrvDV5uisD&G%?-}Qwv!!
zWdEcE!crIg0sGfqG|*_7y3b{gP16YlYF}LP>r*Z`8|~<X#B$qPOi_pEW+d*v?{Vzt
zSd}R1{}*dnfriG7+9}u6i~RCT!|-E-=z6u2JQ5m&uIjDujZkSrs?dmwofWQ8_`fSW
zr{vt;v*&dIU^ZZMb&SfO8W*bTE{T18krkaZZ&(4#syTs>!SQx^Dreg4l+Q`JfJ+pg
znW;vCEG_WyERu{enLlfn<xp`pc1YkGts~XM8F;A{g~XqGc0scu*_I=AA?>^R5AtSk
z1Gd7Y@A(G3>EC$t#)wQ%qrMS+<n<FnHjfvga4s{xPVn)g>5)J_iY;JLMfla$AIDD<
z08twO(Q~5~^aQ1r5_)dquM-kMBiDd48{cOI%m8a{ex=A0GCy=d7J$H=;S`hoUpk-P
zuku}wtrNNcPthBSR49w7tN>g9`+ZjCdHptH%H8L`$ebQ_nxvaOPza{sYz!Ul0w#TU
z-1m&6GlW8ZQ69JFdPkE?FWI0}2<acgg$A+Xxu0E$|5@&?Dl?ncfeoD9z=M_$iK(!7
zWIU$ALj4@E1X~xP>2=z_Y3IfY$QP0Y*T$Ea+536&I%E#1_B)6&9chj5K}kF+)#Oo@
z36ShNuIIDNPvr4R%ReK?kGh@kkvdi~Rn}k$!EMJ(vN2>BHo#)|EE`pA(rkWoHQc<K
zWS2>lX*FDGFW|HEDzNngCD>Y%Zhdn+5<eG^&>#nyb?G%jqIKL7Z@GT&{6#VlC(tz(
z{@db(1$VYYP6rUIm|}j(i%Ag<<Gv~SfRvE}J{Qzk0c=J1BXP-Fi_lv!pzOGJ$S80r
zd=GXH@%GqMJape?_mkYe0of045{da6*S-}s2=4{ay_3WOV>U6NNhdY&H`tIl#jn$E
z7vl4Wr&w7mn1*@34(Z{h7^(Fav9|W;ufrL;^?1x>X9CLKXT32%*(7=Fdz>1I-s%_S
z2*}s%@nxWcW2h3A+8Gv$k{k4P0dFduHy1yXNa;~2+@=YA4F^yxi(@fFqDAWFi_@dr
zT0#!eji~_kl&3ho(?BvmwXP<tTg!{y8?6DAwa^Q!=K5?>(#s#fBU-9iiDOd|p6GYI
z#xI-b!0UBb`0dDecAwGOCy&DM7iSgwy~Pl5J3#6)6nq$n9y<iI!;CrHNe(ret5z}f
zZT-b{oX}TGn_tFEL_Wtn_-Oh*^Z}Evh=}ZzljtIf8Fxn$th0Lq=Q6657RpwsZ5r(c
z`Mow`u(DFby2x*-es@J9aMzsOuYf!ow)u5=o8z=>GpunJr4z>uV)>GzJjKi2*_Tk=
z=^I)uN&UI$j6N##71lb!GESOXI6F$4;{NI4b-{n_w4+$E5{<3(t1Wyy0*xZ~bG;5r
zW7{z$o&UB7vZ;0VU7wKx9Pj+HjPtKe$&N_s@%@;DeyJ&bV)`PmhQ)NgJ*F1aE&i-f
z*U-`#s?JPs^WyqDNjuV*s%1FO#%5b=>#enD(Z1z%)-<V~!1$sgI#>mZvSupJ2G2=h
zd-P4#%9uq66n-FkV%6O90(0am61eE3?-eUxl8ase$Ss_;IyB!6U>dM~$o3xslgbdD
z5Kry4pDr7m+v6|5YQv<C)2=(N5wtN|Q7Ir9g1~Ba0cSw6ISm36hJJ$eG377-Y4G{v
zQzz4?`f2y2(mAiwODZa;E=q9M1x}Cj!ntG!xCkwSwTKI4a!<dA$1n#~RZwc@8tKTc
zzUcrch+ZK!2R5L42ugo5n&r76UlX)$bVR*D<+C#7HTp=N9EJAq!!ZE~9DbmQXRH*A
zL@5@?OQflU3oM9ML1#p@^x|-C`mr!gP%6?(kd!AQ`@=yJX;M0`qMIFY`sW>&1^XU7
zW=LrLp`mYoo$7?ckzA2;pfr>S>KK1^?z+<j9|?%c^<2x|{lxuF`FY>ZSE=uqHW-8T
zF%C}yDb4Hrse0|ssNTX+pH5InT*G~HZ*wu?&`Vr{VuH|~EEjhzhYOjfM3KBq(zhC`
z|DQ5iCLhWf0uKM$=_LFHa!xB3Bv7Gv(~sSS^rGFX!0$EhawZCMk9XBD2n?4lQmF1Y
zYNvLxL2|=&SxX#W5#s--3hFDwV?bF?Onh)O1Ds&>&EtFyT6s9HlE=al!(?g>DdSYU
zza>COr>2khX?{E`QXEi8J$3vpAESV?TftrE2*BN1;EZf(==8hgnKd^YicaKNWq^m&
zXS;=*ewp)PkAEOpkL5`hIIBnP%EvJct4~X7JxWN+f%rgB9qCNaNtLr~Bg|XUVp8*U
z=TA`Nz`ELVSJ8%%Z-OY$%GL%Yv|biG^EAgW5i1RczE2Z0k3Rfnrt3tcPZ(mTn8fjF
z<xgFMPifO?v}<G--cy3s@Lwx@G<aW~ajHUY)Rxh&(3nlSI;JHjewDPxwzLeTsn1py
zmF-B|PA9%|P41xW8$~+bf--u6qiY4cZY(n^saHAHia82LHoxV&6@Q|!qo_9N_dC^k
z;UJG!tNUqow&%P5q7tl(g|3-hTslt&OUzBWDJos`v-;bW-~`njXs^V3LAUGVC^<t4
z*jNQf|NFiax0jnI+?zfRgG@9z+=eFX!U<nT9t)k`?mWJc84a9~(26lN(G~SLZf8a=
zQO^M6;t<y}Ym~>M1?M4CUjjM$W=2fxbRICn+d*k9a6^w5mkL>skCtqiN-d6x=-o2_
z#-9?$i?Ys#I=jh88;{v=9J05ySGQ`WrbnjSQGyy$Sd(tP`Z60;1C(ztJ8QoDs*}Mw
z=4tG!XD3=rUFiAm+SR38u&A{zeC=&v2Hp~4`c~kO!Slqi&!_nKy?OWu?FqG_!Ka~z
zgao?-YZGb|E08dl9Crp^<T{=hfg)s>YU|(B;$j2WP_JN)p4|s6nGS##dKkgK&sJA#
zqT7wr0ipcLm8egCdt^4NhZrX+J%G1k&FeNX_FB72k1ywM4L$-wsM-cJkp=Cz{y6{=
zN0d!*<y!<r9k)VR<*ckKdcrOm`uTcZ#(qYytHtQIjqe3i266F!-`QqXCXu2_^OpXE
zl*=(ss8r9dsrT9Jm4nF>1vu`g&A+7+xq3RtUyOSueVux}zP)gpS3hj|j7h@QC&_&q
zTV}9KMT-j9b3<|J!^8}bJ7T@$<U@0eOrPJMq6-L&X#mJ6K7C#Fc_odU6=YnIcE#Gn
zRxLhvxBiqTLoL_}a&JhA<OE$%K6r9quGPZW#>Fg`b``eY=pP4x$ku9C$u(~O#|v95
zhqQ`Y(%MoIN7y<RtFH8tiVwOqp#ARqNRU~&gplm1U>A1O4PQ$rz$lA4qy&5>W{jPS
zT*1%VikiZ(LhY4r_CS@!c1=uU0(|+$2qZ)cGb8+)Gj|2lnYV;cQ1>TWPEJeQ2vm@l
zFtM>?nk~jkx3eyW*&+WDnv%uf<GI98=3IgBP)NOm^RoDe_Yy=v?<)!_i^99w-l{8Y
zubxl8_{RnE+QGEa*UKIWoZ~hx!Ct*L`z}*@S4y<ofXK??3gv@&8|-y`fe>8h95g_P
zo;aUYaie(%6SW=(+U@0HZAEhw*RyjX%-;>OPJ~?kJ}>X$+6bYfk<3M|)_?73K8kvH
zXyJO%C%c$neKpb@r34es8heJTFwhXMaPs}w)%V&@(x*SV{C;nB$~DA>fT1Z}M!WSq
zd)Io5Qqg+Yyd;b&5~7(CtSExv3cw~Vuy0X)&s!@w1l;60?&vxjMM9AFS!2wDBIYbP
zdMnyh-6l4KT*+P?Y4wO>Lhf}mPuFm!k^U$*+mu*9aMc5roYuh3!~KcJM^j1w$rqKG
z?WYPSlKnQIUnbXQk0L#k@HU?Yl^iXY1z_Jby741sZYzx(Q#R2rua9FEq+@8eq=1P0
z`FN5A2&|$%vyOy;=+>&`r3}EuGQ0OSwz}}c(uVktpV~j2VM3|ph@w|_ANVGo%@N5)
zSj9UD`oBI2g!e8>my|e*iWkW$WLp3p(uSjJOmeH_x4t`4HwC2g5NX}IO8!d2nL1tX
zm^XL=``e-VSctGa%?RRFlg5ZDHLD_x3JREBQB7mpa4Suz3YC1()+!1fPM&@qGi{qi
zt~~fwRc*{VOI2b2EsOT|bMUH=hw}uI*>xOv)HTtueEpa=u@jiTBRgW)y(7KzFn;_n
zE`i@b<l)o=bsARoa-%Cfsy#0q7SVICiul%G@uKJ2pgz{6c(MC!{?J;&+})S<<$T&^
zI51Kne05;WWmZsty#@+=VALtfw?p*@M`>{lmhWQN!KZ2?DabC8E=Pj-e1~66%XnTN
zpiWO5A{o}@J$fdLi92d&gUU@xu+h!-&c;b3mMKy5EO#Xa@nA0n9^}xUrEo$-vZd(?
z;6lOUK5%qtJxLW%+PxgqTG83AA~4q?XjIJb8_z91AuJpChQPxKmQ^(g_ZhD-^IEDW
z!<vqiUK#)f8RYCAwIsz$29Nx@i8L@w7?y&zYHi!-t%vC@oK(}&-;Qv$^h3T;*;znR
zn?opYhmwActT(%v+MPrdf-eDs>p+S0dsQdk!zB$_J_nr%O^)p%5y6w~m%67{a;T(<
zA?SlFlLKoB@*v4!T`~0q*m*yMb+;G#u@&h#7K<c^vW-z&@Q1GU(?*7lKB*KyQ^wZ>
z8!z}#t$zN#bSuHbMB5d|*Z%m&m0wRj33jFADL#)HV>7y^$h0P!6%lrZjmA!X_8R|}
z`}{UDyz9*D_P2U&%MbTtl&7z;V!*958j%~CYDGu|%-mbj2ySjZeBIk3&qi@nr5;)o
zZI}wBwvD6QXL4#3uxAaEf0|mbdeaf3KoJa_9n6HrI8+Se#fU?U%)VKaCH8nU^VrFu
zF~4vA`90Gi@zln4k%f@@!RFnJn<CB+A7!cGmjYhFRWP;e70o2c49wtJ7Z@%~F-Pnc
zd2pX-qIvkHYqsrYIca{RPNOszC<AQ{vjH2-g#@T94C9gj3ua`s|JI{Adx`qpuBeSK
z-E<?qyDUb|b&Y&eGq^hU$l0$t=w->bsHtvPDJIC?YcR1pQMu!QUQEo7N2a0QQ9TG0
zd_vJrY3Bv&m{i+qga33agOOQizCG1J2Jth+VPykR$FUSQRHFrEu@*Ry+%Q;wbJEx;
zY{(5kjrj)|k-q&&R76UzrZ|5rHt#iC7Qd?Lk^Z?q?+Tvj$tM;U{0}shr_gGPZ-m{3
z&)Kh1pcTibQng>cU;mt79O`e6DZ~$I5-AwPjaR>;^%9}it9$WB(K(waEa9<reD=~p
zvh}?nGMXkzR+xMS$zIQUeoc7yXXm9Xmr7oHGKwAMqzcc>0thm~zBJqS+L-3h-l###
zBw(oge*fXU`(X$gp{nvmv0Y}r{g*p5lb;M?6@8%F$e!IqthP=;6SBwIG8@Y`t;i}{
zCpsh~Te*|CI0UBAdd+G*(r4XBro>GRHCDS^Cb;2#6bemwGRTFF4H0`z64K9RrRsX-
zZ%X28MB^=i6ZRu-r)+~DIyH+w2{f51nQA9R`BAdwK<`1B^C8_aX|Ovg$;**!xBDii
z;flu%5yWNlmeBJ%b9AXHUgjWVg#kxNe-}_F2~7T>p|LHK;cR>epxJ{35l9Lq#hPd-
zyFuY2v)RBN!Vg4QtrKQo6|=Ck-{mw};UT6LvaOO~V1(}?U=ermy1}6z`BrA$sPFB<
zPxEVcs2|3D<p`Mmt0$JaT^wP7n|?q76_ajMgWKdNIcDTMXW?qI&rK95NQn=Xn(h!u
zu;ilxOlR51IVJX+;o~mE)=FZht<p4?;Lzbc?1{O`50Pww)1yJnq~w)x^|*^ENOJrq
z!<#u2j2Z$D?u_vQfiRSZdTZ~VCW3Q^)O)8omQFRy8-Zp;ivhu%Fj~5V&bS23;4sz8
z-!w`|eFfBE;40xCq<<cl#f@sO&MQ|hEzoEle{V{zGHyrsIc5q_H`Z%4Km$H(eBrEL
zihV(IsVn*x*u+B<o8Io!B9!uwt7EnaT$6JPox+1hSmb!nVmkGU#9;<+32_5G->Juo
zBK9<c{j}@efK&c@1taMs@7_6)y-1T3Ll#<77dxc<7&fG~{Qs&fQLF!7m2GdbA2J*_
zZsQnn&q)|e$u^APR9567Q97@L*PhkY$CbRKub1lDecD7hM}nTz{kqEg%#3WLS}anU
zpebM5^7}wf(|iiUe<Q^@8LK~jUsL^NVwvchtQ(DOrwz7!dBm1qTFv8J2~zG}drqZT
zqG8p0p_SnG_Lp~gQ_97ZflT3Q@7>?X-}k{e?U8-Pn7dTuR@v&t>CBklxaKA;lJ4hd
zv&_Xa;!!?yt6|1`5!b7h?+Gcybi!|s)1qBP6$}JhVd-i2xnMSaX&1#zrOGVjwkQ^=
zu<}v_YEQ&jJ{S;01yMAxc?}A;Sv+na3(P9b`Z?K^N5DPMWwu&Di0o{E#pV+oaf5S%
znIEQ!oggb0XW?HO>t%y0Xq<Nw&q5h(ZGf!CO8)A$_{*R-EjoPGkBSTVXA*mXNRnhW
zqRT?^vlhV<e@$;IM6}6lB^*UEP-#PQbj>A4P)oz%GT?0(Snt{jqbDdbsY+2r^67B(
zshaxjq04*%Xa1@5-Ay~VxtZ+&c@ivXm8Ay#l=x2Dtj*}F;QyUdW3wGe{*wAplAi~S
z>FIy4=lb`i2^G@hDe5T^%IlB8@3$$|4a7gGY3_ctU(Ig+ubLb1$T^07=@#7*da(Yc
zxaFe9O!sFoQD{AyHnxwF7C{?n{JO;n6lFuRSN7TGlA0^Ifx=HmEWYz$t~N+ynE9EB
zr<N1{iA7;p5ZUwFue)L=afLgeZKDc2*PMs4^C$-|%wS_X$3*WUV~}eXL0C__z-iX>
zR2<WFVt1Q6vIvo){WN3=GN9n#GxEpufs{kAE4)x8=+M%yhC)8eLxO4~`Qty4ZqpfE
zSeg$N=z>*t;^&5T3w}wYawD}}E{KoUM(adjWaNObRB~*0ZlAH{uaC?BOc9$ZK#`&N
z_%jw9Qs~x_x&5|3_)~Cg*?DiQSUV)UZF-i(UBXyTGp$2F<BnmYfEk8ORohQHhL6+_
zaq0I^c0p%kO72_A{pR!W!n44JlBEFC^@LXXz>BH(=Khmw)GxW4F;wcOJikS2rOrDI
z6vo}fYVdoFAk{$&IdLW`wxZg#Vgff@M6E^EHb41n)=iH6^v#$*ppFcZ;tj7Qgu3df
z`i^_#>C{oPhE0F|k(3XVDs%*G89YMna(R3av}i%8ubXFpzv{<`!fc?;+0vT+uWoHl
z%o8^akx3U~i8^TxhzIKCyVKUaGQ7_Y(Ah`nRpmgD7->0LTX-CNql7Q;{kzv6-elb&
zm1ntd-l1j_qg8#APUZLgh~J<zn_Q5SvBQTG6$7W#L(#=4wHX=M8qo_BOrAYvfik(S
zcOp0a=8;cCk<9(gr;7AdOL@-70#!K5|JmG%eE)d$HwlZTOKpu5@$3KjLgwjK*q0hE
zBzb#hYAMCz4Y}=q`6q)apBTa=OMRwCy=V`O!?nKp8n1hOi+N*Z39sw^s1rOC$h$*u
z_`pB31}4MVb70K>_4-XGqM{O$nn)Yu_-#(_{k8z+ZRHVC;YRJt)sL)LoD6A;?^|Vr
zCM3s}Z-vU7=(qT`Gh!;MEQiy*guW>p7l=drYmuLs+V-fL&DC<`RIoW)QaewLz%oNM
z@XZ|6d<#$hYwhQgY8II=N@za>cDF4RISUl3E3(u1r6;u*>G<~9j@y3F1GNYksWff2
z?hP6DN+jWk486EUK2@nR0?2prxM7iQX$neIuJJ`l*}{_bNjr)_=!T|LJ$2Je1~j*p
zkk&N$8mdz(dDo{Iut7FsHrHEp$K7M6AUndvYU)1B>Lu@8!`M<cBG*%WIQ6-{9muu&
zQvzRkzANOodYmzZA)i72S}5^De?s$_RgSv|S^GLS5y88Ns(7)$O{Cte)O@ZH5s$Y;
z;u}Hx`P5U#*DYDriKgL$(W(l~8C4+4U^g)PJ(JZpR=c7w=GE@AwAu@6DK~|5n`F73
zWBwjt-VwE#K3*#b#uZIIOHk8s6Bt_cTr$2BAB|E^{(1R~VS^;U)dw#t)L9F4>VbKp
zKvch{3aIg9dOg}KznF8-#d%A9pIOh29emB@xu+)-E`#)}qWL9qlT84RWJJMHMb4@A
zsv#OgFNvQF;&A%oA3Bbrr;D8<{&O|t5Q0Y%+{7syQPh4^qMlbt3Jr-!)c2Mk^kk~w
z+9_m_Lf%%B>k_m<R?^`+AMC;incV&&1)hr~>K4lApG1X11$vnYJwNO}z~IWMUD#dN
zw>DAQYcj{an(uB)S4TgR2VhSRV7$PP%+-wt4iXckl7`XV_IW&500s8czsOsp9e?Xf
zjz^&bZUlYDVY|TRC8eJz;8(~p%m6v|K~NzS^m_rRQm;Dw*Fv$rN+Sg<;*RBFWVn8k
zNdvVx32jz--9PypH%tQa(;(48gM~I=RsG!l<1>$(Uq3?ky<)f3HhWt_v=bN9*#+E9
z%}3z}Ekr>lVdZ2o>$th{8(NR!`_m$Dk#3s@8@Z_J4vC3icCTzn^0%XuEHVKTqmdKw
zei7r;%96nQzzwqeU8=RM1+{Z%oOqL#S(3`=>>xW2&FJ^FyZNM{OiET<TunqHLJaOv
zH?r!xutubO)2GLkiUlbJ8dRm@kgu9Oon_3l^uA=(Rr>gJbw)|I<&#_Focyrn&*;<y
z#xEjO)^$_7-I1h4Eb>nqc>^696=+w*V^2$^w>e~j((}!*ciN=Q%K&Vcxg)E35^=!P
zUNWm5r1F}BD&cLKCclbS{1nR|t4+V}8HB1Lpa*~S%EJ+wvb6ExQGH2$X}~*tuHkd=
zvI>3Q$LPB+F3ktgXlK|<nas}w){MQK(ks=b1>Ub{etBM9yO@XtcVv^WBaUsz_={My
zM1(C5>h>!TD&0aJv_#SiMjP%(C(MoKW(_*2l~<0Ndur7X>Knh5+bUk|i2rhzN;>zf
zX5Y8sPYfoa{+QF`NVclBIOR5N6c}|;A1HZ9@JKgpB7$BdYt=xyoxD?loP*c`DfMps
zk-Rp;$w;j~@c6UwUJ*fHWrDs5O;8ihwWfp7vN9MVDiVx_lmi;TeeRNO-^psb%t^qp
z(ej$6E(((_E92+3fzeR#w2RWc+Wr^!OvleLQ=APbHuTK6TNhsmB{J@^T|&Tj1-ibq
z<z`MYQ5Ixjj`&~owhP>W0W<;qvo5<zH(%dZRs=V_hBbuIx{E!ETu3-lMQ{46T+0L1
zFZO<cwiiA~ebM)_y!>Wd6RrP^aD63+kSb?RslrRYXJ_OW6+0^!j2zN4_!2xddiXUP
zijBv@`<_%uKAiAZt-a2LFmmm{@@!O0PY=IiSuu5owg+S>jw5GJWOpq6=WdY*XSl=c
z)`U>Z=aE=(SUYtGdMbEr7dzcLCJTm|qCZkrA|RW)MCa9HQCj+^j!-(M7or~;kD}Jg
zZsjGwNR#sv<1h>CGvBuEouz`KqVE1T*ty2JVM;Z>_(_MbW&e&3$zQTwPtrhV1|}iH
z$EA6GALI>HR5WCDd>DfX1aGp7)e~zfs??XK_^e<_Gkx_Hl^jBa+zo=f%1_0N*JBoH
zfAhUdRtN4Hz&R}(`8Hh&H7dI)`upGAZvrxQdlG$aO&eas|A0k{lABn8ADIS!{NvmT
zlG;pbwi#Old?!R{+TU_B34Ha@>1Ikf2x3%IcD^+EZy8B8s}dc)ezou-!+>pO8k30)
zIlD+Hlp1^AgpF!H5h5<omO&TQc3>nVaB8=FZ1sFg2y!@8Iz=z$l>*NSe*s1}3|a9l
z>p5=;=gpiexQ#6Dk7{!>+&E_0Grza;>~#$O?8dvzB!ytx8nu(&q*8n8LEe4Ns|$zR
zwr%cS?`|iRHvG=-<xl+D4dvOADr=WMbx4r8VxRxoy?DGPW@IQt1(l{WzGVBG#{zBu
z-A22)(tj>^y+Kor5I<`y&bs$CF03jY-<(<od8y~P$Cx?o^?C$I#B*M!uI3(|vtsjm
zi+cWN!9h+}qiCpFC?u9c!arCD|DZn=q3Et={s_G$PJ7NcE_m$soRBG896(;=&NZ;z
z^5VcVLcE_@(MB^GlbO-^p6s6w3$a~Cq&PeA2I(>*ArlMOoz^mnXJEmb5jWW+Xeiz=
z`g=h1%ZLl!;~#S=3dS1k=f>m>3d`|~5iuYw3%a5=xzl7)wv3$EFLJtQHyF&LUtZQB
ze~jxJl87I~1ZD~c2I1@%5#l=Bimd9<v8y2<EfR87w(HFT{b;&JjhkB*+-XMKAB^6j
zFUZ$sjGXf~{@V+n<JMGmLyQ*cN>}tr)a*#YxvmeBk#l0^(!Lmov?%BqXxQmX=TnCZ
za;hlM(&B1akQb4dM(0KmQ=l?>lj=Mn<%m;+TEod_T-4gn3FoBa;3>pXhK!tp$!1e0
zOxq@J@x$lKReeb?_K7Jt%LA19dx@oj7zv+>`Io+*A$1u7X*~_tH2QS-Zt3o^nuJje
zAUmAW{afmx`7FGtx^OkV+!6kGcocD*Vfz_5c%!Pirhy`Dk#SrVzy8IHZ+=R<xT_53
z6PeI>W418i&5O<}q-I}E_T+IOP3#4&E3VdX`cu)V23!+?*_~w1|LZb$rxoH4iIm~!
zY}I~SfoHkXvq|BlZd}jz2J=hne7VNNP+v7MR;I!-f>Dvx`7HkonAXAQY(mU#Rt!B@
zgb_Vtvk-P72SueU2yO~7#aQ|81l)0Asdh&1=obf_QONACAB>zV3I}r60njOa!2zC}
z7jH$s9!pcdj%ej%(s~u|;`y1U8T+JoJbo~2FLOT#uJJacxIN@9@h*Oc-EIqqH`xD1
zDd5J`{zVZgcx7j`_*Q;NL3~P_t%Q}3J;J^z7qLXEN*h4r=It7-yjWiKcB$^cSXTR3
z$*O)_Of%=b(Wbd&kaZuOhi?I0Y1gQik?<A%R7i7*>7p`1Xb}r8N9(9jW$&`dU>c44
zW4v){o?#o!2^;_YQn?Fz{)1A9?HL#0Tkqc-`D<c__^U$4+TG<=<-zYu*>z?&rzT69
zj|$@BQm_@?PM5UjNfH@;^1%TGt2xm2;<)KKaMDOTh>py$?Ab$)*j{h`)OQ+2%i3pA
zL3G%!C7$u!>-?fT&mEL+o&CMNbCz_K_8>PHr_5VkQ-rUjsr`%bTDv*vSEAPD2a(&k
zn0Tikvy`7%EXCFCA|$DQ{Uut+3<qPwd>E;ly%v+~1L!so)@p>+crMEgxsP9p>BfT8
z-ItMR5vcAYff9>+)1PycJ96pVAYqy1ph4fakv@0kw^WE3d(~8gv^D1)7^WOR)MR$C
zQ`Wdy;1$hP#oBY$aVuU9avFJ@S)|llBw_|{!0$p`Q1taDN@z|&Ueu3O4bYy~Qe!Rh
zUCQ@rz6HlLaUUcT=shRp83~tJs1jv;jAfW6OG`lz2WZV6`c*Jn_%r?6mw^}Js@c=3
zS&kdK`PRPS7sg1(g>t@`t(7dlNa2F)=tx{OO!#SK<TC=GcAjF`54pcshF`}S`Is}#
zyRJMg*b=}`H?BA9QEYA4Su`u??yj`0xKrlL1M*)Nf0cuarfe!+oT0%K48h4Z;)pmC
z&h{@40hg+oTNc8}*pFLn2&r4KKuUy+Cko7QDui{FdQUZ5linH&1V0P#{^!wA0DASP
z**y@7bTf~Q#N}8t9uedhL|+xS%WR!<Og0*MpRG7P`lQa(2t6(|{-G984tg03kjQ@6
zrYSUmf?|Sjva(loe>De)I%gHo6%(oUuu^l(4t}G)M!0nMiL>Y5l}_=O`T4IMBlA|(
ztf5(D_d~-(o{_tB_8nzH<=1;I?sUG-Yfipwu^vHNs;_1?x4y}N(<O0L=*Z%zr(52R
zqdzE5WL-Yn_Rk@sv|*$UDGp*@-z2g@?~;@h-r_mSIc^_Et|TW!bgXyuh=_y9wwVlh
z+fnDtr+63g=NE-jQHg|}U+Q9&b=AZad+akEUdeLpQ#8cqiV9MPVjzkoRc?!qHT_>K
zg=v8IQKqaW$BQ(TGqiZfZd8`12uwb%M{p3S@!8<z${pbe9+9xcJeT!yv+pN*$Q1@Z
z=g@cBjbuyG<(P5%<PIm=`f1;k>N%NuevlYbq~(wtmZSr7>Grx)+>&s9e%FMgnhm_q
z`OizZ@!jY9M_Mb3S;N=MEbg`o!HDJ73gcvna+k8&G8g;&jr5=8;8<tpM4Bj@E0<(X
z27N8rDV>6h=*Vd`OlVg@UV*eb9UYV8MUZ<Sa$i*2vau#x<XDvQtTenzAWKNjQ3{MX
zqNkHvm}&_VJHFU=Ielcb)6VCQ6okD_2I&UB7uw{N-}9GOhG532sYIy{x0-wJs%~(0
zmp?)^3d}|`eT1K01=dN^kDWu~V_M3gBd23SZd-W|M~)!mqT;j>$@X813Q!JU!S0+d
zxEAx<$Z%b+JM1K8LwX#c`jQ}AHZ|7m{a@l1RUA$K#hK(#f|pX{IBt=ub_`9lSeUa~
zl&N%)Zk}J@gkkuQSUB%^-7EOJa9&K~y{b)+TW=x}bFcVjGqkDvE|i!y<4E(RB)SUu
zGvKoVX*r-u2&9oZ#R7opl8egVe3(Hlx&84{%8nnnt&t05Mzc!=hDW#&<%Tb~r_?FN
zC=4Z9dE@&*Y2O9iZqKk{Mgl{W=ErNkfQ;>k>7YZcpA-v(KAmO6)2jnL71}<_%Zmam
z3qm!NMW2raX}T+_MKJ*|5DZagR3#XgMK)0z(ekrZ+}x{`c~*)cpe(ZM+jM@G607-v
z9~7J^g?hIOYMUWu!1tCn-wVjx7-iUR-;k0%55|`q(3R1doICEM&37BuO14FtN8EcR
zQkk>JzNW<`=@#hrAjk29l?z1+VfJ~pQ^{567&Iq=2HcXxPl9MJDV|OUzKhoAr32sm
zD!fsoifl;L;6r=adL~9~ht5`lvdT`l+0{TIj7y4d5@YAREr{YmjyLPJp!HPF7L8VO
zwDdx&=M}BusPptgfx1plIvp+mGqv%>FiFPjG5hey9&92t0l#_r=cKb1UoWQ3rh@m}
zS>jA=sP2Mzh9BAs@>Cw1AqF0JdP8!|RfhX812-C8=c~trhM4AD6p=-r)B~@07knl0
zSW!$7_7oTf`^Wd1@2z5Ja0kTqJBIMScOVTXQru+SNiWg2v#ug;1V6@-;zv;J*kl!q
zxr?YtE|%S+2ssDNeU+X}k!Ln#Ai!r5>J=Mh6e_x?YH#IK<b<XYKys46?4_wjaEwF7
z$mvurL`<`B_Uf-<ujlVSR2qNlE3L_8zyGWrX{>9<(Ce|8e&`NFem^$gK^c{9MOj`c
zgd(N_JYMt~2`yVAX82uo(sX@AiPERN8JUxK;i3%`Uc8U&kn^Ozk=P4&@szu%ugl;u
zr&E?k{<E;qH_vtOJ@6CL*&I%!SJA@?@s({<Bt9-J{r<^n_v>;(z!a9Uuvq5cesddL
zh#q;<#FMJ>98JVa{lTgZZz$whWaYMZM>*IDiLi%@Vr5Cym>SctVOF}E`54^HdI72N
z-J6Vn1&L9>3T8A$JPY-RB($@Sxll)QcKIc-(X@N~u)H6dcWC?!givv7r|H;Jj^NEd
zY=Qdt(&nMY*GS)=mebxV_EV>GL&)n%-!!^*f!58n*m|<Ci-o=@S#m*#h0_C;Gr^$O
zOpSJ7;6=f^^<An-Df688P9^`agM{Dso8yxI)x>)ZQgOm|$@kFW5WAK9p(I(!&8=oe
zQm|FgMJ@P%@%~Hrds;DYgZlStg>2;&CI86yD9MZ%1aQQ?8o_U?oDurumk8}v{41_{
z#72L?k@Q2HEqIE@N-peo!qa<W!?K@iwQ+)>gamCN$L|wqzHY7sUNBvZou;^?9w)c6
z)kvp3#~?ZbuOiztv#gh2MEhwht*8ou%M)!UKRBqu#aw-{{O{gSA=ZIiNPkNxJsPvG
zr}f2)vRar5LhnKqCLacYE}n#Wswaa|(kn$h`;kpIKocUv=wLh4Vw@)~PhEMZdKPmm
z(!we<Rhp})$I7!A07@-y-WJ;U6a?y`JbY&&Gc4rgY%I}u@NAoAA^aE;L~=)b2W?7n
zW6M%bksh=#jPGaJlk<r-R=Y>GS$Yx)$Q70eiPh)Sg?3qA3BubPX?lIrL89qTRpAHG
zjxezhUWzU;CyTtMl>W_VHxII`e7kJeW6B?_b?YdIBKYgUP50D&>b-z#>0gy4BX4zQ
z=r~`3O2#pf<b;!Ywq*N<kDG8A-C*0z7R=yA`33%ea4!-hBETsiRygI4eZgO)W6zLi
zhe5XkbSfA%525d)w7_PhkWY44NZSlCk354OylE&<XP5duk=u0UW&4wa5Dg_fWm8Vt
z)yCr6IFV!o`%hxrEILiQxD%4^T9HC9Gdp17rd+G8yx?jj{W*Rr!YfF~ikC)KP;Q9f
zhhvAD!S|Zw+ws_M?$qo)TgWV-+oa>9(k&dvVLUidFp$(z8YW|WB<JQhqGfB*=a%?_
zBh?N!J%7_^?eTd{JlU}p?;fp)o2Pq)O;emWvFIu)rujgWM>j>c!1Qa=ldhAf%AlRS
z4fMa5fkvBuq})oXv?&Fo!)W#_F#6?j<B)g`#9~=|ZK(KCHy#UD3crHgihs>$>wY80
zAv~-y2)ccoBiY>kg<_R;^<O9KyB@et9XqJl?m1dw)<`{V$Yuc};1t{Kj<gCU%Epac
zrOalP^zR+`(+f;{Z;GgQ5a)R7K5f!l7qj0yyl#~ih7Uc1GMHCc;ey>{2->s!PR0aw
zXju2m0g<mfUCuV@Z#(mdvr)IqO&SIyEOaAJi=7YNFn$)5#jd|u46SYl3sIMLSEPSN
z_{6V5VQ3P(o3t_D;C|Xex!f~~tO%Dc!f)(0o^6M%rsBcLg~par<=u@dY1?Fvu63^o
z{<u~QWu<adSj3%o{A+MJ`oH1V=eOcEQavMX%ic`~L*hHzqLMwUwfT`n`NLd5jSDX5
ziQ=C>*~R0VZN{soCaFt<w5FHyksznx$C+hH&1oWscfXQaynNqX2Yp+fe6D}LGuQ4h
z-Tw7m^G)vlcW%5{8J(_3+M!}{%)P8|nDwW2HJQ0<+fA`U+(^<4Q@{6uC&QR2+jU71
zWmaEWzk1QT<E~QqMn7R>#Y!L9Z6)^Cz^`uoE)5Qa-X3QJr-*I|Z0&!IZ}_nVj&~br
z6UJV?SV4H4C`BpdVnu!9!gQP4+37!SZ|_+&{S`Qr>JL}VrBNB@z89()H5I+F5;;1s
zf93kPSouIZ)V$s321h@&ImvW~grZ-rml3{@_vYz~eor3ur`>hYUS%jPZm|YAU?z-)
z2NXKmwM9zC-$f4Nhn+>slDq2|>L)#NgKqOCBSbh#JEMZPU5NV?JT?Cr2p%FfoKj=j
zuiA4W1AZ$GCD}+X5-yS;{19ncUUW{wc3w8sF##dckeW!p{4*tRd_Xas<3=J^CM53?
z4fZ>g&0zMDasR=wKf&!BiriRYoLg(+DmLG;RVE)*#CP+qRtn;+1{>O~HrKEW1=AXe
zApdv&4o5#5g167LwuHLLy6YunOco~E&)v318_c$ZF8QUC{G*Y>d)X0h&%~`b_ejwH
zVehTuqFlE&U<D+kM3j&i6$PZF8x#Ri5KsxFr6mWXV?<KALrO(K8focn7(yBxy1R4u
z?nn1l_c>?3=bZPS@ArMb_n+-{oO$M+=egIq*SglVt|c29sQ$p(<#lDg*qmj{bPn~m
zx0~>(>0HDI&sDEI1r#aqLYeD^T^~*O@tmDz#H~XSK4VusL~2pGOOmUZs)pdk-kA-$
zjcUfM43lP*<Nh(+=7qi(eTY*Fx~<{MUtG)i=;s)1vbu{`l;w17gs`OIzn7RGU>Yj;
z{%}9-xH&^tm{qunI~F`CzP2$m`hL#ft}ZN0iL+{P_8lq<z)z2@F#&u=LT_3}BVFBC
zhuF80)Sq3k5j|cWZZ3rZx}F>rvv6wWzMO?#9MLtm&jc(Dcs+JAk>Z-E;So`)*NYZd
zNR_Ul&D~#4rxE%d{D8(U9X-MRV+H4ftIlDV8rK$DQyiC1T+F7}$S=D7V{yN&^E@Ld
zKTuGq0iEl?g-Ag_o05J?`C$nu8=vRc;xx&dn^9cNbW$>o)fwWv@F?+YBaNN9eVZ;V
z*U_28=prmK-)h*)L4(kU$u}97HKVR%a7u_&Ev%Jz-aj>|pnSHmcvAv)pfL14q40HY
zdRhmk;`NDDUR>3zp|_u4-&gUrI#K`EjU%{X3A!s&%%0l~>PLOP=q2%Lt?<t~`g76R
z)#tTK`=BbE@%D!RzHj!L$(s#S7P7<1nLx#9lhbvR^`S$nN#RSgl871S`?b%4h?&(u
zZ~JFswf6L^5++C=Qf5`swc>c$f*6u|Bp15cQz?#nbTlhQyy`qF)Wei_4bNSSpxgqe
zXs&?@mk6bEp5&JYKcBBG>1jfA@{J<@^gkD8ef!jzbHhZv?Hh4nxRjs%eEHX!eQ_c2
z2tpw2W;k@!dj4Zn!ux0{B|H{&3`p6qMp2sP;tH5&9)6eLICp&}_S;r+4p4hz8>x%U
zm6ROIDqT5?19P?<^Qj%P=v7jYV?Gqw{eqNhydI@)rFiq^=RdzlRt#F17_pq&KP}<c
zUH)UIULvjP42odA>6%s|Q&^vQFl~X?#&f!bUtff(=zOAnlm@REyV%MO(mI~LHtOfb
z{>u*g`RbqF|5qdU?bS!(;_j6<nyw`={>Y4eS;}v3`kVjl97A&c^Tz$RSBS>IxO60U
zi=I;Qd(r6MzW*<8z(z#=_X`cKx^S*@%;GF(GvmE~y4^p`^6zdM_g)^s-`>B!xiw(T
z$uD1GBOV3INGbl@pa1by{%U>y@y9sMU~%uozLMqi{rhG8?aw7dvta)BA9K5L`E51o
zf?lX%_}|^Xzx_r$qcuayzdYT4ZyeCYWKToJbUDob{eSlU-|wIHRrg9HP|+O+5{@DN
zo6jqfr~<XU+CZTd{&eGnrZI~!IbJ-VTsxd0HBj~UKlS%D{ACMT%V|d_DdR<Hx$b93
z>4Mb(=@y}*oubP4UR<i^|HTi?!egmC+8Ccp3QFiTlGR(=1>@_Oj(`Sj%>Vv^1z#WP
z1n~4lf4?AK`b{2=|K4<~uF`{@a<#)bjP1X+;2sI!rRJNNb(#L_kNQOeGxX!%lq>nK
z@1-m;w9go_{Zs$zkFq=iK8|&wjc@+HHYpAf$cPilIFtYDkLLIQlPdDzLVfsyHTs|L
z$Eq#JLppA>ME}<w^8bqu)=ilZ?Rs*E@&;;3cG?RdKbX(yzQ;`HoV+sQvO7w?QgOIa
zq84vL?AYmEbCx-yb6=!~0SXfK7PU@>_@X0JGzD0IRk*AMS-I!ezU>eJLfJ&bE^`u#
z<_>b-9e%Qpc$jT|TpV;@vjCCc>HB%E=wTF^7gDZ1TL~<(76eHYkwDur2Q=4s2Jj9s
z3h#Ulir?OAvwaByKKU_=3q?R>)Z==j>a%baXWIml;t5~o8X#YmFjWNvPxr@V<Q=<(
z{l@pYz5r?NM#L9n>}1P{OId74E53ldAy#z4Zx6HWtQ(||I9`cftgHcn%caiK@(W~-
zVBgROA3o{}qO_V^gbo)Gg!pneTuaP!NO_=Qmjf7(I9pX;O(YdR^z+YHvrTPoY<r6=
z>CBhwjOW_`iS1QlyHOhz-BpSn4FFRbK9(1h5%SGCG|@VII$)d<j{)zM`P1FG?MuB?
za<2puu*c)TCoNQQU%e3TvNK3WbxZ>oR^w%x9aM9C7^_e@N;w>ciFr^N3+7xdi__I1
zooBdo+Ijj-Cwe;#iG<gx)>co@s|Cnqpp&Bm%BP>oXMY3}(X<$HLtR$SIhY#9s4%Vl
zn>ccZ9Tziefz)u*hAx*ay4}K+i`ro-q`7mI7_leuP>_TH`-%di8Q>l1R37i#mwL%X
z_F7<P?@{8xV}3C9LfW_f77on?K;9+y-wGr~!Z2pztRL_LfgX%|D&e2NICsZAp#MqD
zJE#}7PuhG__qHP#bLn>zwQ@2vWRgt-$y)Csv;$XJri9~l^Rd}~IWUgXZqP0mw>a6c
zs0C^d<e~>%L0&+G+!X_LTh-LX{&11>VpOWL!<F6TrEGR#nzo#vC48;D+x$BNx{s*V
z;#Zv+)?#lq<|e6s{o;J=Ll=k(5%3pjV?s<&hu?N<Kvi*cz+$UOme%kJIhpUR#^Ve_
z5oS}MIygQ~jIh^59Ek863;9+b16g76gx#{)k1xzXCH45##xGwPO^T6u=?E$)G&3%P
z3FCNBDsA>EQr)WAa^f|E7hlDCK*8=JTkb>`kQm-t%S$}YSh0ONI;WKFsAOM5ns5%Z
z`(>~3AvXC6kOeO5KpEt2%??nAr-U@hx9m><u6Zp3<Od@CjbAJqC=ES!%U4~-fkdtj
za&gtMzZ43=d#y+DzdJ(?4J<aGaq;1$;mjCN+8X8CfObZY@3TrU!?6}|)QI?COFWG$
zWgFzsrdu-;YeBm+XqZT;al8u5*3o{q6h-3-)UdsQ?I4ZmXTbJ_^=a6RKA@@u1T-&^
z-4QNJmv<#V2#_TXf!@DS*(7F=+b3X<{{FHf_!vxczeS~E@#6Rf7G}7KUl-y16QC@9
zNg>~$@p%Af>0Yc&O-XpeF1Jh~$(X*bL?abol3AcUCfYJAxSJbQ&H|}y$?zInT*{S@
zO38}OHor1E@2qcL_M4aXB7Iwlyd!AqGR|-M{`fMCVIB;rq$50Q$F|B^hZYo0;Sn2R
z#INnH2tKsk=|3sLV!cKqTWRsC%;4FL$siDhsPRt)OVp?V`SAQ6tmUer%Hw6AT6a7m
zxR??4A~?u(E2$Q;y*{$*^W>WQf<jdw@4X=-e*oeGfqDPd-Aw_(B|^Rt4!IBI<6cx4
zg63p?<HQnTwn4W}$^rFKKHt6V!!n~#cJ&Eq;@%zmjj%-O9Uu$;I$)IxGwA1m_4XnL
zJXzX=?b6IezR?$t;2<@+2q(jSbpQFu#92m3ZXzS<m08i<FIPLZK5hGPhyv$H20atM
z;_iDSMFVC_s~?CPKBhi$oLq1(fcs%1!mzs>#LSBn?zp!MS-!8<niY`$e%G`e@j3$=
zGBJFO6dv8F{W2sdcB$(k>IA*($6s6NlSklZM4)*`UuE`5N7Azo17<~|zQqgQ{HuK$
zMB+J&{rZ6*?bS<%7horQjip(Q5#_!{SbVzc(I!FQ1+nB8(z?l4nNTI&nvw2AVn=IN
zAE+~;XA$TWA9o&Wn&padx>}L#RC(5KYhHX?E(TXn6p^kQLYMk?rmP>|X2WIK{pMt$
zGSsK48*nx2?vuJ^;d+P^b~P)q)233rVe&)x1kg!idEfoThgqH+ZeTk*Xf!8ivCpz4
z9A7jR@LTLxrGNpiah6UFOh07MsOpS55auw4xvIia<t;ZSHf}FwW7@pg)a|}~lVusd
z+`<rX?Do0A^SyiOAw313$k<CvU;g!6AE8ed;OpV9KHpN$>gmLRIt}A@-ettT#<s3k
z0d&EumpU)PIcG8!^}0@@fXh6in>0t(3TIB_O1>wE2AULkK4H)OjP9z`(TtQN1f?He
zWOwZH3)Qm8c%yETL?fvJv&*?Lfi-xQZSCQm$x8ZLhyFjKrs#_pZEvGVyh)xudC8ne
zSM0K=hI)R14ci%R^0~rwoE;n6dVYc>WN5Fz7<6OIeXWm(b*$+gvtYAE<v`F#zWmF=
z?8Jvt*V!FpR9;d9_figKgQHM#WFuv9u+X)x7sqcpQY|Z8#CMqB`YLK8vgn0IQ2GiP
zbcm|mLYg2HK1EE&ZwQc*&XgXGuDYt2=C$+QAcVJ$@6(R<P4-_qX4IS{m``vHR5K}{
zP5iQUE@@KT{3cr<S7X?F-o`VMrWINuGtI(Zwct{A@Rl1X7o1Ooy-#mC+u)1H&obLh
zUjNOjN1wvRPS41_7j;CVB6#WOVPfgz&Uoxr$N`qXDx16$g61v$4KXR!%W~Znosbd3
zIQG;S1*1z=U%vi2r`!3F!y!6XYxW?ZlJkmFg)BF^JF4PquUEwO-MN64Ku<lOcbJke
z+r?Z@bJw@HTXTP1YSyeUi4W*Ck`ysLSXZ-d=ke>27b{n!cE0;F@oeLEqvYC?ppbm&
zG(`s+r27CDF?Z=>G;3SYrpM2-ZNbnHo&9E-wV*;i4+f%?7Qju0$AWTz%(;6)4)<Pq
z>&0y76puE?fK!bm93&KGn@<-M7l;GpQIJX*V%0DJuq0uv5M)6Y`hdLshz+rBohdFi
zT}X^{LN3PBNcl8rgYgbSZx+u9CS<aPDK;YM!-+2JP&KNWIqc>{;q6TUC#uGPCzD^6
zo^hya#OJ-P)*>R%Ujkvun<KyB=Bd`rjpE?V7n&G+4}UrtaSRE(x@BsiT!+Xm*t(DO
z1qvGfg@*aspl<0)Jfha?N8G#p;{}}?y5XE0_lC%=P#x%l6A+aHkGU<h2j-VN(Uz5p
zuk4bd`lxX`4LOB!IN~WH-rpOU73CA-oR#)(w=PSQwv|sWah=S_$Y^}eoK(U)tXN~v
zd&f5CGHa02$Yl3|M^Z!c>`(Ykd>G51Z7j&lmc&_g9w(_RO}x>n4NQ?g?%XPym$SIH
zs$7KIh8{<%m%6^1lchKkSvHnm)Jd&IsYjjqj^%WUy6u7?G9ud|W~dIK)e+JGaDHhV
zE`bm<xX@|NL8k`abUkKx+}U~lzSbLO{rc*pK*2QodS*>U&<Vy?5BYmMiP(hxLEo#J
zpJ>y)9&NHJLD*pb0*PPym__&&3QQ-fhhji7*g>avH!rQJT`N-?esu_M%SsaM_NP~y
zX;k%NL3aD}0|wNd+s?KilAPtTo237>#TDeo0Lu4h{cDP7K?>rvsSB7J@wJkBY=Pfp
zPx0U`=Bot0qU0}?M0K8#u%rwAihrC|<OxK{M>Npjm>UVlONMp}$6Ou!NzSFS!a<eF
z+d-Q$iFQDQ-M0HtfMI+!tc0iVT9#<*DkQn%UHul{tSCt!!&F?}$)7FG@+ZUXt~MK9
zTX8iCLSiRY(JQOZJ@cUe6SJoyw*v%twSFK>5NX%UAl;(f$?+!K(&aUOzQ#utPACD4
zqiW(R2LMVNq;1Tm9^Pl=q4T;6;7DnC`@yjw_1iMrldzLwZK~n=+!EM)Lf>f*Xcmbo
zIMYiuUWnen)e#CakK59%dYxO2>SN-fM8#|!_NiIOO`jSp?+AFl<~<Si1ld-VxN81x
zX5M+${OGrH-Vx6kjKN?ZXvoP*JPA(`)}L}ke{$J_AKWcZjn-00E~q4`_l$<g-EZVK
z*7^hm=J@hBkI+P-8zDg#GW4X&k>-Xr*zoDuL&~G_HBWXHG2WC}(1n0}K76x#fZgTT
z)<1oC-9TYt$Xwc56-r#XtwXTMO1`Ys;BT`2I;gcD;IiEMq{wePTCXHuBL=jGyy=33
zIqQ_>Yn0W@JmESj{1CIU7=%YP05R^&THKeu@U&L+aEWvW%oCcDfBa%fEnb&<4<BAt
zuvxk?1&MW?WjGt{Rtd$nSBONtw)?fj>OOv!;AYHl%3OTPt%{4<Yv8M?JRYv}1YA9l
zI}&wkQw(APcTNkA*iQ!N06b;hF#!hmth^wF3^*x07nM&0=C<4dm@n1sxbpK8+D58i
z9)pCst&s3Ip8O{lvsywvf-)6b@QHIOkC?PJ)PLk(aaGkgFzWYRTJlU2XA$F2MBl&t
z=@cQHr##E#2%O41HS8b=%>iomJ0q>+QpSE+Wg9J1S|bZwb*!>^<nw#}hQf49AReE0
z22(QR%n<#U>y7-&aBf-x_<g@Lm3h0z86MxsbLOk(-(O`=2g1(dg~R6f0L^|r05X>v
z5FQR~eq)?pc5rC&1QS^3O+kEJATQ}@?c-b8Q4Rv;lq*F1piU>R{6s$@r=CpB<I|n9
zUC~^3=+dt(UfzWALBcKy=8VFj=hG@B9U;seVkbw}Hpg(Rlt&)&Gz;eljC+tibhs5L
za4?tDvuYGKIE^#U7*z=-&C!p5K`5lt^%qstNhb6Hvn)4=^%6gQ#aA4Un#y!8cf7OW
z%dDczZs)q`Ix<72pnL-#Ixd-rp%@X<SbV27aLYf`bpt;pyF=2wxr(n78SrSwFxP*}
zkKqW4>9>x{mh{;d9SE@DA43ur)y0(qr5Z$bzpV$_eaNLyIbQj~_q?L0eG0fGlkgNX
zZ?!Wdgwl_`oOA`bUYKA?y<*;bfp#O<8*2f>9tuWuK1L`nIGTawE*&fr|ALp?JSDV2
zquSX-;4(}|JzS5#wj+dhYb+mc)GHpYCGCU+KS|FrPApxnaOI)xMfLP)8MZwOKhie6
z%63md&h}vRu^nfRi|yfJ`wPK>7z*}|U9{l3dERPS^uglX%AVJ_x2`a^xpZvD;c@F^
zcJJX_duK@DbVHZnbWS!<x2U49j876)*>>`WB`p?%5ZFm`KLv4+Qe_W^nn#|Je?x&;
zKHkL`x+NFwit)7eYRDI8H{bjo`{X@aD*}n^N4LdQ#L1z?>t_mU)`^Ge_F!!Ds7*YK
zptJjPy@FfE42Y=fvfMU~YA(Br9aZ6CoWYZF#P!7ufft0VHHx8c;sp_~Rc^+w%3{XC
zDQA|<`qwK@4l6Oe$O(#3U`j<ZzXhny$gS$WZ9K*4^;)SPIkpNm;8`#ZNt^Tb6g7Df
z{R7ONR@Nf*OLYhkfnA!&h)8gIM2JocKV}8qU0dfoMg17DuGmm%niJCL8!4zg<t$8Z
zImH#unOQigaqdlwrfr-_;w0#YGp4pky$;K7%&*RN>`lkOwpMBs$KkxKev_ZjK7827
zpXN2@GClB)%C%HGPT{AqreGVz9U>qXCWiXA_(4pHfh|tp@pewjAVwd<3PZ}IOwF*<
z=kqW~$r1?`Ij&C7VC&e!V(|9QoSC?h5*K=OouF8WX0?h#h+gCfy<3^c$gc2G-&1vm
z52iR$4$juI6ttnJ@EE~HEbpkYsp}#|3TK~D?}Iry!g3$PpZJ=3*1q)J;;5bkiH^fF
zKLW>%W(I}^xt^eRCSuoVv6JCSiwX8}4|fdH{I~ApzHpPdNukgAS$}^zihXn+)dMF$
zM@QeT=%@S?@MhJFYh1`A6x!0*P8|ZX9>tPMoIVYmfL7$jbkD}0@kZm6^hjUw(bKfy
z#f6ty-ba75$3RdwhYePAm1JVDDDT0VU~K@%NO-pqUf0dPiXMG(xi;0eB(-c^)Z337
zCl-33hHyDtE`p^ZFBQcQl&)4B-YFraKW@L1IEr?(j_-JDC5$>;tSfJ?*nGB-os{IX
z8H<Xe=dLAjp(m}t`4Yps7kei$Vp0S&FOk~8wm(VR@arM?6F>ktZj<e#vHuXY$vfOt
zBEE5(XGFqo2aLrMs_V5|q}~2)^EzOBuH~U)W4}S8vUeX11E(X)YX4?qBpj}vkiZ`Y
z`jU+C7K^Rqt8KMS@l<zRsmoRzqVT)kX8SO|z5pDa0SV!Lcd6oK7Pc=_p%W8+>};A_
z@5w)!z<ZOf62+0jEz7jO?56p-o6QgMi+{W5s;-mb!)*S6U%9Bdb=@Bup|t9&1?E^v
z#%(L!`6#!+PBd{5j9i=bW4u82`B6d^399(K{Au?zzx3<2_3k+n;?K8lBEDmvRpF#Q
zzI`B86%=D~7QB&ktS5vrBIw>URgX{iIUCrKHys8aeqT#0R^}shVC>W}(l<>vOko*R
ziP(=q%wzExJ5dwj`xRd@CsDyzl+oq(zd<B06?`@8X}@)ly`*m8g&=s(u*{v>Fe3sb
zD&mQU5jTU`ABY`qO^^iH-$4pBor<-na6qxf-L+Fco%2#|P{VQlqa-!OLX$|-MY#UN
z)k~+zR-I6XXZ4cfw7^6!1CooMZ&WT!MH?{=G$X;Zw{7FaA(_orW(kuX-Vdm))cK)d
za5rvreai7^@Ra06r@I6$JEsM%zO*{0BF(1ljxd$+D<fBV&(gw4t>%6?Y(h#7L4BDC
zQVGKt{7h}vu~<kJlfQ(#Lz$XOwY~dE?O*~CyZ?<hEZJemwm3#;8Zs5<6j{lnimODq
zWDtJz@c0~|vs24cHm9#J!QUOS^yS5`lUUnV9}uNR;DnHdw0nQ~;fbyGCbzyYENji=
z#L}i@Rz4_^Y7XKLF?%*D6YEkG$G+3=)L45k|AI3Q9%MbL#qiB+C!YKfB7VJQq<@w|
z#gOJHft8w~YtSZd;a2?P#0WaEg~*EIhTutavwGH|k<fA)$2L>ImsrF&9O|#=i_n5{
zmBoK>wx)SPbZq!agJY{87Gs@Q@@Kn+dzw3|_@O$$CpF-2>3ptS-kl^nGym-qHiBZU
ze(+-a+C<t_$vA-lJnviUrnq<cDel)roD$j<xIzoY^64U93f<nctJsXcuv*NjS^bvZ
z_3;U4CE6p8a$0drIQ<OENL{@>$470;=IFIr?KZ)T$y$`czC09OzGCP^j#+?QlNrSF
zWqKE0={Uu$qm22C`Z@M|(<`sjQ4PhX#eQJ*9gZRE*K{)b48|nd)wG$xupLGw*|W~|
z;S~n#@OM083j}oP1P&Q?Z1wzc_+o3qVgBc^4@F@<UJOYDm&dLzBjO5Bq!#|J6TWq4
z&e9QV21|0zI#0#0&`r<PFEa_|5?T>kNwQ(WBTBD!P^>gv-fQLANKaF}MxYlQm&iER
zggJ7}q581N-9}0lguaHBjJDQKTN^<qeYd1Q6B6DbekO2_bDs`1*~IsXis9I91TM|l
z$#mH05MQlDrC+(s8G2r{x)P3k4mYda*&<e^N7zhgq`3D<HP(BpCyrnc@@UBMr0Ky`
z^PWDNM8lh(L}#LP1@)a@Eo~|Nwy_+f#sFj~=i*H8Y69pVk8E(rJ!Wt9-!E&%S_2zw
zaw~*>-dcXBlKT1~pcFh9l4=j)wJ#jBqa)nuk6nB*#`8Ls_}FFwWq@+Hxz&Hul}*0$
zW1u1D5=T^{mUV1!5YX2sc!#`-mLo(PNk`{^!+mjQPF$7W6EEmVrGWqa;bkKYd-!a4
zmOba`Cx0-7fzO`tt;(Q0x)Qt$X0U0zQXsoMxr3oGZ>WS%aQcY@aXq&A1$Qxns>dpa
zP-}0`^$WSrwMT5<GK<0or@1f3(77vWX})G}yy)2qDglu&8F#DWZY9uY+my2a<J!)3
zR~!j_N19Tsq~mr=XV7>}C3*Y{2p-%#@4p3&^9@V+K^Yh-JaO-rL9Bf2TdaRui^D8m
zG9w{!xZb&r<Kx26^Pu&phy^RvX*#bBh;7K}uLh>~lq5hAp&xIznw&}{Dv9Xwn7rm_
zQCu>Rs4wvuUcW%lQI~1o<KbD-eCBW|Po9B^0w+Sn@lK%c@TBdE^{ewCy@7c6?v+a$
z1y<D2_Fc$;Sn|Qj61>Yw?qXB<R?Zs^pyQQ$gLUeVT1Nf}iBE2xGmLjHarD!C<6Pkq
zY*r{r`CA=|x7Afm|5}P&=wvY_uE(GO2nRpNB!qhL)*d5HR!_FWlkCbSX|Q_W8H`U@
zP2lM}SIv)ymAr)5?XUjIHaU8+=7zgrBVd^7Bdy@NoHq_i^i#VF>DdVcHE*sJ8^c5o
zdG5iCyG3_D9Jf->>NFZJXNP4eZX^*Vm~D?PW%KpV6EIiGtstY&f^ahs_jMjFqK;SW
zgu?7nhX+5t&*)i*+U3$;>rWOn39g1Q4<vw^gz+mr8|u>32y!F0dM_$-)jG3=o-Dd!
z{>RzK$3jnuaCDcfO4~k|5@^``gd?=saNFKg!(QtaM3$U3Da_Ly++VT<dBZi12poj;
z&AnmWAdwO*Hjiw8lCV0V1e8@6YodeP3YHcM!-lop2^|Zk)7i%3Jj=m(&AdA0wBKaf
zp4)1S+#L-4yu`i7+_MJdHB~j&flmqQZ4MM%`1EYGeD}d{WMmQZblmcl!9rU14Z$;t
zRA!1j)vpSNbglLQ4y-Isbx%#kk>gv<SU@HRs6(~BDm(yZy%W4gj_>2)7rFLE#o=$s
zA5G&<V--d<t+z-QrtJZYtRFAZS?}TPio~F7hjtS|<l6yO=7s@VTQF7e26UKraK&-N
zoObU5b16SryM$L`YF^@Fw|WTy?~dIMBlqnCxo2NQ27&!j$i#-4jk8j)?KK4AIQy4N
z6??!DE>_9VmR7A5t6!PP8QLS%3=ci+!Mj;*MlZK<!Q9O2aDu$DIEA_#8a~>R(DOhj
zbC#0+=KOCdG#S0yNIx1(@9dHp134zDz)x=Fr1M1gu$G?fFcvkf%4f{}eP=wRsr^?w
zBh3YVxHu|wC&EUsw#6dsaPwp31NGr0A@(h_;K8ufXd4l|+=wjYQ2~g*^wt{F?B{a&
zpPFP&fPuJ{<d`qs!VU)-!63NeXL=GPBi{-}10y%FXkG{)`3-x;elgFY@iFqAuQk4O
z9_@R~$T;xkY~yp@42AHb_pJ(VE**^`PL4H(OZ2PL1z&nqntDdB&gyDzym6b^*s<If
zm3pbv$P(vV>C8fPG_K_^9e*?eK^hZK;c9jjRI;q2v`jKSOQ~31I<gsG^M>5UneT4n
zY&8ntEUD#s>5j_JYpYm`n5l2>Jhdp_K6~``xH&7ckdaxwMMoJ2A9xSzk;5CA@`Cdp
zj_z`XN>n?5ke8fq#8Tp(6P1|-!^<lSW`vigSPfgNBUDaxj12nn31Pc!Y&+|D-LS%J
z2qM>VhEG)Yk!3t?JF7Doz-eD4G&v<n<fp6VjuUR^PiO78`PsDXu010~j79~3+Aczq
zbBE|X76${~pmkbi3R_*{-#f?s*a8&<m-E~8R1*vg%)DA<x77_cNjzN!dx0pA;=Nz~
z_*%{yt~Sd{(YQ>SR?^9sj4coh6iVCP@w$`DlCAkI=w6tp^ZsbE%Lm>U1YtYut?RTe
zL^F?Vy1#OZ*%tz*ur&xA1Ql_yjvyEfM}1;ivDGhgE7oP!^P85TI`<+Le{PKdDIX~|
z!fY`QoAT&0tH{Bk)zGFuNd**&%5l69;R;(e61dDiT|6`CNv^5tT4(+96yq?#G61ge
z@$q6M+9Zg>L!mB%5c}Qcj`9n=*7R&{o;`_kEZLS}XR?Bvr|3i!c0nSYP^>*A?H#Tk
zvxwCJCK={5CemXgph3oFWdrZs9^6Ok3VBf*2AVE90U+;1&DXJP_Zm{@3-RBGi@8@W
z+<Rt9xdzgjFRN!uw2mgoh2n;u`@JC`%kQO~ZQ6bG^kfW2tSM-Vdib725D;Hzn>tkk
znj}%zVVCK`3DE-wMGtNut(h}&ZgU>#zQ<Ivemu(Pj#DP|Kr_5H!*7<K`6#3^qeDG1
zmOne0!~x9QJ<_jk36yORzo}xIl#GbZ{j!{kdaNj#;UC^3yt401DQ*<ooFm*`HbhWl
zem&fsSd<?pA}~Yw2`ehdWl6U*VIA72fB5&Rn#<yexPz;N7WNYTbs`U?VHpol?bovD
zrLg4=*7SYlfGXDakTP;oWX}2N;o^r>W7joPD&gU|ykRW{K3Yq^)XE2GKg{+3B#)@x
zRm4^j>{z^-(vtV*Gt>ZpQ9YpS3`Fc?oM_|30wHrF@r6Z;c0J(CNknu5O+;xCN9)tl
z!oLW0c~aVaglLEekPzW>&X>bf2=Fm|8AR7#%ebw>PS#->=YR!&w$UKVZD}NZ!2hmQ
z%=NLSaf0F49>9R^w@OB#kh7*qsy1>2XjM6eLq*Z6Rbs0S$i>6PK94#EAwo1K-gh95
zjMpt08=;`h>;~@bbis04Ucz=phxRQsGsQ5-k59(69*6fq<*`jVk;cOA;YoNRzO}$h
z@$g`q1%k2Mk<Y-XUqsqBaQQ;OKl7O+0@}3+eg{ROoPcV~Bi-wbafKTpy#&(mpN}i~
zz4HY&8|3q~{?apcqtE^1u&q_&^VQ7-3@wY6qKzKAcpa$xZg4rcIfwT(%j1-SDz-kd
z+Z@Nk8z_ypE7MGyEcB2%Su(JSC263%Mdh+0a~L&kPb`?&hZZz`{styIjE&?ySla1t
z4!}_l6P>`653=LcI#{f$kP~U(ppi}PHHgj|)ItHMs`TJE&uq>6K@2Z$e<5@fqd;X=
zNgdQP&IrZz7MQWLZ4_j?uo|v*PRbS~P2=Wh%kp$U(DptF@+Saal|d41<1Y5lt4m$!
z^2F}JVzw3wnDAQ<%F;`EWeUliU&p?ef-_N2Px@f#3pA*Er|b+uc2f_Ay(>7ag3V(T
zBnBQ@wAqs=0|LRUl7+Mi#EY*jYFtKZ{qCsmm1weSZi)oxQ<eid-o1rf6kkyQxqPko
z>xq2<Ps~(*7+DbU%@wVZjodzkF}%fzNmg4S7(=I}TWW}HkV~b%_R_AQ*^ip6kr$hI
zEkz6M3Z}cy+UoK>Kk&X0JIr=y#*|4VF~2$X)28@`2zDdY%x|D~EKUj+)u3(N!vr$K
z+)M7uN26AHkwVj&e|>JwU_APq?6;>-S%?APr6P!}<<?vsQA5dH7<$bc(~Wz+HcG^3
zNP3~4fl3l3VtkTIy?Sf5XGn`dPHxlPaCmG`DzRK^FsG7M5CwPW$KEi1$Oj(aha><y
z++NLAX}qVRW>z32`7ZM!8t@W=^1Er>3Nw+H4AvU&IWQ#HKU+Rqm=CA2D1*DLe0Uu@
zZ<|l;U2+kv+9vgDp);?l;wX-E?AHL3qO9jj^I2c#j99`h@T|c6bmL5*PkOY?wWnKk
z1TtpLOflSUrT)4x;-reWijDOqx}Xe4zBSNfg1K)0=zG?6g#(08>4g!)Q5z1LJVlv>
zpPGR+EyFuH;o*H|ANd+}2H;sZ=X_wpGjU*UeLX<Bdnq}XqCB0^Z~DuU!{QXNiS3Q!
z%I2^@OD%<Pl~|#5$NUXg5XhyiEG3ZAawi=6*2DUha+3QB6cV)2?3lP_T$GV1l=oxD
zIqBjApM#8VF<h5kw0uD+`(9WQ;0s3SjYYNYfO2nK7AQ_vBD>aq<J;zl<CX!llH(i)
zL=2TBFo@RW1gcip=ryu1SdADax?qPIb$hcrb>;_sl}dE#M(YcFbMHtgsC@9|%?u8*
zusSb5wDj1<YV!2#S;RXEJDVeLmNhD5s3$%`^1(sc;{vEo?8;A#KkBJUM}uR{i}Hws
zUiULJZ@H90k9O`Z6sTfgmg<-%4|iGZL4|91=TxV^2Dd<{2~!=&<ERBGcw8=m)U3^P
zV&8!y4T7eg7x6<pn7Hx<YQ?vD<2PX>pGYo)jAdsi<~jhev}TW;yMcgVray^-HUGXU
zttApoC<rPsK7qT?VZcI-k#y-O9&iyv%m~cd!8Ag}x7yVOy<|L;EJo=f9Rh-Z-6fe@
zGyDWX9n+`ocf1w)Qwsni-R<*o4}=6v1R;)ehsY-Ldmb<YDtYYFefQVoPjq-&#!eFe
zGvh3UtSnm#ZT`8QY<kQqA?VKlBv#5mLWjTsPz-kpx6bgmT)nU~Ef}G6l(1LFKxR5w
zYu9pi?qrSlBv^T;6cD1kr-FYc>?a3QL6G?~2r8rn@<%CnS0FZupDV(`u_yNV;qlhz
z@1y~X?$!;1J{eu=E+o$~-S%Qw!28vNe8eC-VecYjZZv;2x0SPq@SdeN6t{_bs`V;|
z`r_M^5#@q~dv9?>80XJgl^oCvbL_R+v{Hd;$_3fsgA5QfV96e<tE_fv>I6g6)RSRe
zFI<4ilNA=n!!|QxQ8Wm;*~KQH?DEIy8_k2kWxm#L5fQ<hU6%W4*jF1VvYT<y2-K85
z4@vfl%nBgqNs5U?_r?}<>tjJgnJ}xWgxPp`FD*T5$QpJ=rdP>xgJtR%RBz^m@8OJp
z4stOKLO_8{PozdX&V2@gfkaSx(%9wMV?Mp{7_v<adSIJCMF>n^8^!P>1d-^cdF@+N
zz&0m-*(Mv#xv|<p`DxZ`D<I|0FX)kBh<}`s*CD9E*#sy^PeuS)>W0jd>Zob*KvSL}
zWI!0XekDY+$SOWZJ6*KYR0<N*eZa*Ta>&p^i*CImX62AISpbL_slGKxILtoUJOZbm
zhRb$u!TxRZ2nfP$vR`BsM6B8m@P^Sl2$GI}19*Bp#G-kYS<rWn*{WiaX7w)f3s(TK
zvY?MqC*|5U%7rK?xUAOQUv3NvOmhTb8(XbFg6AD4Owfpadm61MYPAfNy0!S_4%Y_V
zL0szUlB>8i3PMxsIBy!ric9{q;tI9OTpfhV3;}xPJ|f;~z#!4(-EO}{$$WB_VEh5K
zd#;Z8Ly%DQH$(^8PDSem&SQbBH67*a<`H8Piq)pNwB88Q8m?Gh0+q|X_riTcua{2Z
z+oV8{vk4XER1z}FcZO(2o%RKOJ2FkrMc)LpJV3jWnbR2_5%gZK2GH5)b&8u#o*j1a
zy3=nm>KI|r4DN(RF^JBF>-`-UvxXqfphT!PH~S^t?%S7H(R8!NINVU<#E!5u^^=l{
zB`}HPBw?1qQVg(?T7o`WW6wjWcZ!k-#aio1s=UkdAQ(jQ+A#Z3k_4U*n(hOYzwv#)
zr6p-N&Tu6@IXnVdm*2MSL1H_qeVVyhNkeV<CVJ=5(<I?bQ$u%GFQ2sTW&u9Pknm;H
zfm!Wf<oR$s43AnSLdL$<z#Jr~nB_miLasG3Jp;WdiQT$3?l>7tku<(y%Arjg$YO_(
zV_m8!>LBM4kfCoZztEHPbIx|oAbNHTSTEMg)@|o2Cj*5_j;3(1l{rp8*J?kgY#l6y
zw_DiPQVF8r-3Ege>HaSyEy4Jq#E8u#=Z>>P?&{ycReFbeL*U2;@gTD%h%i!m)yMXV
z8%47AK)9Y7YOio*BjKkts%m4YMzCi;f1FU(%N`?3lpU5H!T|In8FrWLc$aHuEv4t_
z<c5GfAYvwkm-wNKhUcDZesx}!e?X66Q6*YD)igJB^jC!S<A6?z)ouHnp>>7b`wN72
zFUB9CAt_Rw>}2@;E!X|{jXmAigN$2VeNe`sB8&h#{6aRE{3ir&flvC7Ttl5R)9!5t
zg)X99Kwh;~5#BuedfM>!I9XD=fdKsty;H>S&M^3;1E`3+z>}iDpQ@f^dI?GZ$%}(5
zYgd+zK|(+0NIh-|62N80gx>8IX5GuwGn6cY*-_<7`o>PhOO#;Bdi?DQBs}$FB+~%z
z%~Gi8N9Nr_SNcR<Tvb~rDScW^he9^?DywByiH8#H1CKwMFNg%>zJpiGo=<cb@GA-O
z^9od4*<lb1DBOMhM=hA_l$(a_;K!7TlyA~qN;V9M&Q@-xoz2NNLTq^~W@Aw?`C$B*
z|6?nP{qHqjoGyXFUfRiVklYQB((B~aRA26wa>Uu!s;~ndsjnf#ekb<zJt{D`p3b)-
zt?L`8=8YuGwn#++@HF#n^hY%t*?d*tU4&dYboqq*Z(9%S``j9Xm>kY4)`|Bq6`Jlk
z;E%yKuX^t;s8p6P2rDz_;S+VYi_2<I(wRGx>)Rb~_xESLZA&{1NVt#O&IiiSwNXpC
z?T1E{U29g+0`RxsH~%+iIo`n!4CoK3OG7M&W)UG)M-3u|=QyK67Dy;<gH5<l!v@{g
z<Ym-i#}h!ju8|(Uxx50p3sMuS`P%r7J=vYF^tx=xn1kZ$E&KA{+!%)^Ry0<id2Hn^
zwT;NI9YBT)D&XAjzR#`!Ts!QB*pNW>qE_$r_$Gc5fMk~;NVW}5Ya0h*{`pJS#fHWP
zl~L{J1I&+B(GO<b7#hW5*^g}=cK(26)5`)yb_U*^ag!8JbEI_5)4nTa2VgAvcuZMi
zJNtz74uEr(Z{7oY(si3;{QY6NpmHk9B3tXcQ<DVkLx{TDtJ3|9DX%Lq5>s@&j<j^3
zr(*8U6D_z(oy2H#Ea<e7a)^!z6BZ|SwyNeE{lIkhe&N<r7Gxs}Sakj)1Ch%ReOL%N
zJ-fLMLyyCJ!QW0+IOlYyBLM=Ry-75Tg8)r+3zh~2Z&<!g5WljtvAN|>oHzcA$d7KA
z4tS%AGwJMRV3zKg^sR{1*-ZwKd70*5cH+(uG$j5fBIO~BRK}^(nGE22aQ`t~)%lY$
z48jKAwOgl<=vg{0eyFhgU`RLwYMF7Hzy9*s8Hsdhkp)HI__omiREjQqs$Ix1EDGtg
zhzu8Oj4*4i`(o(sKOYv7QN!vr$u2MCHHRv>)l6S{*M^LOZVsfG1j#V?u0rnBh0D2n
zYiGOog}V%v>?!R~3G>)=X}yaB`bh$shOrNZ))w-bShN6_RgfjbJuU2ly&<4h%eqde
z@a|0t=3eBKPq<T^R!u>!Q`hL?FCn%WgIs%R1!Y?D)b@d_T)~n}!@|RMp=C|uE&$qJ
z#tY)6!@POU)3n}VLg84=rdJ7HIS36H#?fVT2*rFQ2BS4vDFl^yu?}bGD4@WG#d%hO
z#6_il0kFVeo>L$y9wB9?kDTJ4eo?5nGt5FKpgBo>L*lCdJIG#~x)V0%Agn92ar0hq
zujNq7Zea{}drx7lCzxg(B)g+?Xr_FqTPT9!w827QVcz-b4(XF#%&h#G6oOtX|5H-r
zTg@%UuC6_(OH<I`^$CznngvnEOv~uW(DOwcPABb*;~~ODyRW&d@E%0D+ruq}0Nn7-
z*fqaCFS8={c>843<=yKC-p;iB?I|F;H3Cqw>cJuefVp+qX0;AFWF$)9_+HU)SmLT|
z41r7fSH{@46<_lzlD+yNK!==r==Fv@ePwIOy&6#5@)QlV3NkODF4_>7EEf*5iSlTr
zU~CASh7WHD$p#xdyJpjBT9{OqxzysdOJ$mK%Z*Lzmxa#wil<{E+V1IzUz8}y?QF{M
zm*X4dCY<O#eWSFNNQec`h2Xi)DOFqCB#@dZ4JDtet<JSn5pCQXA4Iy^R^Q`uXcfk$
zVh1ShwK{{-A=LvQ8sll_cnT;np@_Ooo@@uXr>B!M2%6Wmx5crFbKlXRGjm&*j0;4t
zqwRqI9x5iuo(1BI8eSd{wdj(EVQ@v9*Mh)&bueA_y{4sz4bI`|*}=0hPXr*kcwnLQ
zcYG}RHik6J<pl`8)vigTkJv^m9y)hKikW_X^>7?2o2H0VGy2C2WbE-76=Z_yEJWl=
zEoUuS7kEHzd~$Q3rdXs@W>}Y=KwHn_or*w)b4N>2hNFEQiS*QEr@+bNqQ;~*_!)Ke
z&&wzx_1&R+8&dYpOgc;F)Qq3j(X(nMr~GlC;UvH^D1|9;yU-+ynV1oCee=H7!NaI(
z81m|7FgXyO2He-M(<H8RvCe?<z4}mU6cip(S*CMZrKhMjNF|?AIh?+|;eHXbb$rUY
zg{(!!{~*ouE;)6GL&ZUFwyhOr{UzV~(fWvmzz@fr62z<nY{g4&k`q@+YtvmF)*%9Z
zbHd=NOsng#{<TJvuXOV>R7oAFTp4W{)|7}YKh1DuuMDpqsfwO{qs>y|Hz`~p<NyGT
znS475aR4iNOO%Au%f9n9{^+aVK;Y$E5+H49&4o2kHI2$Hq?ds3EXjVqRV!hu&8D<c
z4*?`GbUY#wEho3e9kPvKuMOd#PCLAyg|;AAIAQld3*ImZg)Oacq^?|hWSX=Zc?bZT
zSWu@uR1De#@LEFApj|Sp$$NKWF4--U;=z?{E`x(KmGZP6`cBcujf$_y9d9>irOvia
zMtmV`l5P-<dE8oe>jk4_9a6CTy4n3+x%Wr_qNJbQx;@7kBOW*TScLOiE0?feh>`7A
zvAWa)XLNhr-9r4tOk3_oW*DV}mvJC*Sn%)JCmzeKF~Cx*mmH_BQ`$H<xEMing{aB9
zI9GNyX5b5PS2Szgt?spw{_-sA_%Oy1PyQTl#b^E7u3~E^YqiA0a1Kt98YlR4=~{q=
zGhN-F_>~SL0yw>%50`Q`SC_)ox<JbTzzM#6!Crz|FKb({!vpX0<hl2^H146=`B2mo
zZBge`D)8VY#Xx1P7Ii#|`bakmx2ZiBKrKPNCp&J*dNzlqa?-T)oQez{d^_Qs;n0Ul
zqa}*>BsxA))K`c)y^Vy=P52_NoL!kO2N<15T?`M6*usNvfC%ShWT{}+YVxDK;SAOU
z%feSt?f|A`Wpy)_JD!LGzXrsW4qI6>1@%$Ww|tZNk__=L9{S|5;~d;qqYgL%=DyVF
zU>>>?y>pYuo*?3k(TNex;TL%KFfw~%+o+6uqNQQ_z65g8f9PD%03lqd!VTFFlXF9B
zT(s$md$4GCB~tXzqxhbxj~b14agFXCC=F2o>RRD7t-9BkSD4yE<46Kff|{XXP1hmF
ztazh^xo2}ky5n`&)zFN()xwQeFGCO$w$(CHern>LpE&O4szzoqN_6e<oSQW=BJ0z1
zT0W;DK@`6D>8&jC;YyxgI<H6ZwzN;n;5ii+B6tqno!W|Ph;@xS`szLnkykF=rBB`$
z36$$WO&w;mJPQjyE>3%vuCKS*&d1JBy6&K^Bj^mV$B%@G%>lY_h3TEsw-|2JB#?-;
z0J>AtrSJR|c5nnX!dE}FuSS%;G=4mX)6Jdc$9&4mh#>BOD5+v`^JmFU90N32jkyy)
za9k~k?mPi7jYf%i7YZ;y_M^@PaFk^29+D7Vtr`BdQ^)}@P8rx|YQWN5;e{~Ha3uL{
zMEn-_eBOpSRSD@)wcL=MZpQUpP+TDHpJxF^<q$}lmRZ52tKT)SkHFVUJi$+`3=|E7
z-+EyWxAV0;T5sH)6QG>Ev)Y@iYkFbyPH#Xu@^+&g7|g6xPPHd=Y3;g;t?L{Pkmm3a
zikeCkX*9Z=IHb@Jw5T4tY}AX`I8{z+rWqF;*I?h(zWA322oxYrrx#<iRpo2@yu2xH
zOR(1P<oT5>)e!WCP2j%f{!orJ!E&0H6F^<<7O3l{voL3I(<x}K`qXEVc8{%dARNeb
zVo!?+rowIS`Ld7egnNs3X!A?C5=b@n8sAoTNXagupn)8v@f!6RoU>FDbxs^<XJ6xH
z!<FT=PKv8fiLL%vkhUNmcj{v6jU<tP^tT^(FzFaC+@9twJTe{K{{+x&+XPD^KclcT
z^SGVjR2K1rJkBqd%+Vsx9XFla1DkM;0w#hl21^>9;0Pp0G*BduO$iX-PNoHp-;j5>
z$3!}%6ep9Y=Ux2NGE^?OAo%7+osZQ(owp27(PEMQGT}gf(=yP6i+zpwi?7TD#%A2a
zW)hFrrjpSVtf9Ou=W{QvXSLYxh0(a`4f^JKjnwho8uVel;|1jw*`|Hhj^!$luDJvJ
z`iXW!EObp-ll&9n>QeA2waPX6W)OnZW$IpP?*B?-Hhbn6&;N#^X%G3O-g?%r66&MQ
zU;CXk0ow6LL$o&SxwkM6|0lBx2R&OV(ZS!ocG=aa8ia~EhlxJq5lsTFctuNjvS!t@
zTZ@6Uc0-rm{z11LU5CE((Q<awYhg>#PbWuJkG2gVUraE(l$V71FhXE8r%iSbK{Ki`
z^7;A_C{nmtlaahuJGPxynOre2cD6bfrnvgfk6Hd~R)u`Rd*%<K({aL-+=VUt45$Of
zK)i~%)m%DxV}?V47hb}(TJa`%G^~g66>q((=YseXAzs(bm<_IkpM9imgY=zg)hJWx
zZ)-od+K=BSOXc=`0EwgFjG$DZ1?cUN81XG-prpWlKWCtC>2H<ir?&3_QJVHse#8NP
z-k2W;*wf1#D#tbi{9>R9C|RQ|daVS&kKV;RPKM+;WTDCAr04bkz(k1#HJ6)|dbEIh
zBE#^A_i<d6iPj$?@}Lx^OA5MI!IY@qP*n)z(&ol!YeDqoB}5c9PskODi$W|}rYNPv
z=v#qZFf}C$&@nY1XE>ng;!eo>Np!{HXC3d)0u?lQg`@v?Th08-^?95DvM7xzh~esS
zyo*AN1(NI1_x4L&{?WHqHY?(G<N#P*RKQ;pZ7-?vy|NSi)v)}O?rpV}3FG#!u8i+J
z;Z@og01VTxADN`>A51U;2pj<f%;Qwl**=8`bVC(X5|yp&mBMC8Tx-`YerxFbv!%Q<
zy2^PjJFi7-xSrN|t=c$UO*vv1fW{cZ)QbjDr-rc+OS-W82*6`@g8CLANs)69-FPs-
z=iRU7Oi|EvyolM*`g7a<csfJ$0M$LZk^vHyafsZmnO)HLbUavs2yfUlPtLxMj<)OK
z2mpU5*-Dpkzjp*eLo<ZuW#rE`(mI(A0359}Ve_|&-ycRHjuA!@&3?R%0=U6ANG%B)
zsjX#;XWp39fYw?{fXEj#pPwD3Bh|L^ggBCNLv?FbkOnCzxKA{p_g?*kO`>mrPV4py
zPmX_WhVMT?LE`1IHY5gX33BO_CkLQ-_rM-h25D&bo1~L7UHsnTLY=F{0NTOGYdQgH
zBG>CT08NWJVKZ*l1ER?^sIdRMJ-^XW@Z9#hwSXkZVJ!gasX+Uqg$KYE>RI&|n!nye
z>DM<x=!p(U&IlH>D_!k7-D$a#<tu)*PZ)TChIFo4C9U7`6Ud~#o-dwFaH;|Q3N%p3
z$r3Q3J)m7T0{W510WXRWY96bFN(cle%Bnh^A0E2`aEww}%Vj&K;gX5R$H=E?ee&%8
zVuNr{y6pp2H0RhxD?1G8+-yES18!|GfbByQ8kCNuq-KA#SAtcvod-Sd26*|^f9|e7
z41tlfswwE=5gPQxc|oN}H_D^?ET8YF0{U(@EU!_KC0(8E(dG}d<#z|ck%zUEjbzbs
z`Bhjw9~}qoKp59ooPS|$zkl|Z4gP$!${#ZH1`?{D+vp$m>(6)o^Bc6SAl;D;1=#=a
zo<CyhfBI&w2cSAYDQ$Y{KmWQvFXh$@n)3MsBwYUL_Wq+qwDChvgZlI7e|Q@I$))HS
zfZW0q!S=0RC()lv`n8+?>4$#*eH@dJ1OdDb`(OX_|M=?vdk>eC@(3FHH4cn8PMGZA
ze_Z9izJbwKF(5(@B)nw6(*UlQ`MKfJIA|kI_+LwBWB^mnGZ*6vsYcb4Yf(ec(U%#6
zz5m5FdUAaYs{NQ&?9b#~nz;j%uP!~p4EyJ8{_h_0|4CE)yY2sD=lq}M$iFs{e|Yu(
zG)I1hBL8q_{_24Ke;_84Q;g>fBoyRML1Gv6gf(C9YbTs9UT>hPWk>9{eh)$#g=cPV
za!2NovM#0%j+-+2g}t<CpJ=mt1T`4_OV1?B4vFmWtzK$8`ua$o@MK%-glB8tTX^aU
z6lKl}?}E&M$9W7Bh6_QBC5501(5t_E#U}nR;m-?ljAHSAv!7f8^r`F4fa#F9;z8kr
z$D)7%)p_=S+R7!UI^|tdHqDC99nl`D@6ziZCNHB+jnT(5sjx}&H!A~|x0Zfu<6aqS
zUUCB>ik#N{EZ8XCZV+pxY?p4m3^k&G-d>kkyM3&r{CL};8yPt@XQx{;6Hb6gfZ^Uj
zf91DzYXFxL$^#(`!OC<7AiSVb%hRtO#wG@e(sJu+kh1fttML;^Q3@Oln%mE%Tp`lL
zIH<5+ipEwZ`T}wxNE@X8HKxO!{K=O5xU|@TA}7TNBx(Z6;$Oe4zW%=d^5w^Xq7@!Q
zxG4s)Y1k|jo!JkzYLe0J7B||Qu?6~l-XNU(Xl)j6(}(xH!dEF0B2J3Y%91Awh6-*b
zo{UVv6G2JBAf{^E8%Fy*cAk6Z?S)Ci{}w(6N?fxwKz*bp2OxR8t|`kn!-4MGO2Yyl
z5Y{qpmsfSW^JT3yxQvHfwEi~WsNy~ksx!os90Y?Lw6!{Sy+}sD9Nth@D4-;U2CfLM
z?pmB2?-D}@eu)dni_W0?00V66&;mVU0S5QPo$CY5)UMg`&d0Ct*>IuGwkmrPNPWu_
zDg_pxLpwmD2<H)~s-3;hTJRWZ6b*aV09C`coCR3bU*Z^7MXwxfB8eePJ+snGQ!i%~
z>Sg8!4G9O1haGkmT=mO<@IB`^1BQA%f%^SwftJnCT7HTMYaCQX+dcl#&9!{K7-;U)
zLi7v0yAMXo0UyTOu6$7=m6pvrAE0s`KpI96>Qp0U&dh)K-0z@3X?j;UN<5~jWHBqY
zr8Dy+v7F1T)ST(?<ldz_-Itlr`b<|pZbaXI{k}=zp+U$M-JtN$n~!jB2y^PUDSX0Z
zPtaq$&i%TK^NxE%(YJ=(%<U7G?5DFDhKc3Ke4sh_Wp(yxBGP3wFQuUWAanjH_|GCx
z_1J7h50hA8(PO!RW@|cX`!P4EgIEzKTdmYr1RdDvfLPsr7vWO=bxk7auKTVEA$h1Y
zA1{_vm*|3lLCFq}&F1o|5Juojx-ne-zx{(lgTyFq4QG7}WQy;e_?-?1DIzn3NwBKq
zAYFIzY9Rroi`kIM#_1Z&)BT(fWx(&0&qunuR`{&>3jO>GBhYBMi{{1S*mNq|ig;0l
zz}Qnl_I=mEK+TBL-Sq&!pQv&QKC`(s?cgR{1nxZ56|0UofpJL2at|aWn*O=#*6IgP
zd#r&&4$B5aRL}u(jzr(vm*?L5wb0^0ME16la8mTW@pZrNzlq!H)HSh@9NqnCpi9F+
z9vIvirTXl3p~cv%ZI@LiEy$8EkA?5-f#FA~%amepEXOO~5oGPr?5OQ&&GkJO=_(Qm
zLJU*!XGGP>@AFTm^+&k^O%$vqIMfK*V6bm<!FUawrqX<PJpZbSaY%Z!OuoM&*}UQy
z`lKZm_36L=BbgZ^fE4X3F5Y<vSnfJmI)1wc0%9P8Uvjx?{D=~8j$ob;V{{yp8F>6W
zin|2?$zmw<oGt;_+-_g0T_&ke6qiq~zkfFDej+LWc6bf2Yjd&~XtK-N-egeh8HSO=
zKu&0r7R`M=r9@__<zK4VbRA-3w9vw9v~DoDc0r4D>$*}!;pX`yW8@i0;avAFR51a}
zM;eStyewL{&0O`Lzr=p68mAUC{0b@8psszWHhD$Ik&4?EzH{@@#?932x}NF(VHbVh
zVd8vP2HfMT2*8-~SNFbnjpgz&U@Z|N4u?@SpnH&3S<^v3uWdJORygHCBA83h*&r|*
zr`s0^T6wa9#eyR6My6>2fn*#UNPb^7GOr%#h!_HSy_>Dn!Q4L9z)|2sb<Y~LZyS-_
z+_Js(fDY4|A)#YD=74n4oHw+_M-Yo_DM0g}3Dmw-k6vT*2OOZHHy^9IagW~rFoLQo
zTy=mP=s;)$L~G<fIofQ0${aH0Y6CT*2QN@Qf_fa48VCjA*L{mEnUVJOo58nyii`w;
zeJ)eEg1QV*+5MOHd#}jLs$4gTO)6A9jzN2`PXApT-Jwsn<JaT3Lr%bS&6J3;{)(Lf
z9Uz`lunBa^_|!pZbsRiG-u@;Fr{Q6y%+b&Lr_^%}@TL5h$6mD_LgEh*x<-q&ghyks
z=eD7)>NOzB#GPY&DrN@KW76KNhSVi20U>WV>heJ)AT{C2JG4A{s?`v1nd-vBL95AT
zc~>iYfMJ<5_PKk)K(_(RRJRK9&xJw|Td?+ddS+jV*Yz=|&YWrOsy*lcn1g1e4b1?Z
zz^o-*O0=KAt3@<lkK+Df1QY{v4m6x!%Mm!cZ_y!M3aJLq0*8vs(!~o96f)_{m{<%*
zbf2ZM8_{vNMPI|T1$?^E2CY@s8<VUjZ-PqjK!Ni9l3HEZT0JMD4bA=GuY_yWEyGe&
zW_o1er7qUd&f;6eK%$@Z>A_2=;k^y0p!D^3W(9w|!O;D3Td`#dVi*<a5hyV{^pGpq
zyr30NcMOJ1=~u(HUsc!wmPxK?JTJx!gwxC$M#B`m9Nr?|uP%^L5YNTt2zE$)gn*qH
zSiM~WXjt)X=N)cqNw;T@Cw|0WUo_86(4+NH_ygS@S4tA{UG_Va9>E9H2Nwl|pHBC|
zn`Z}JA_y={&XfQXpQTHe2i6+$m3`)e|FXaPejMKknVF+P%&Xn`lzmz^*I4*#&UcA#
zv$(_zSzMsT>xBI_!y)T=6k|i8;G{!*MRM~_$Xcp(oJU3SgM(v@KV614(>d`9si+$D
z0$Bv-MdLM#BN&Dz?TNSNiPfKurNOO=uX11Ra0&EFi3TEv`7~%LsJ75L;1bc|-%YKK
zRG!<eN>%z<mvxsUIU!K7(!#{2Id@`N?gT=Hs0V$d<3XM-WY<&)4%!Swd!&r2a?js8
z1Zk{L9Oxw0z=Jv!(pl>!ZP!85P!TJy&aHF!Zc64aQr=gHzpo;^=2y>n)Jj%%SNiIa
z85pi<01uJmATv<qnB-%7c{zCLb0e)zdS>Ap-w%g;-<@IfIxdI7ts(yHua9bhlngca
z5{A}IaF|ea01c~C^WN*JOk=crZ~q_0-a0JGwQc*B0R%+4rG^0s9a6eu=n@cQM9CqP
zk`C$a25FF1kZwVc970+eQ91<%P#EgF#&zG%{k-q?J>U2I#agbl$;=h!c^=2H@4r2m
zSU{H;hcaan`b$AL5NA~EO1P-H?dAp`XEI{)lsbb%Ew+Gx&a_f)5NwoI<q{y(<k@S4
z?g@T)Y%R&5iP??10XBl!zy+W#puh-V4+BYjz{44j`qx5au8-owaC_=75nfiYH1hIH
z35-^wx%O)~5u`DJ30dIjfctE97~QAv^yY!kIept4;?_@+30yWJq;~=wY^E!ta)W^f
z=K{y3XK)k1{hKzOLL3nfT)0oEj5B34DJ+N<2kLNUuQjrVF36H8!p|3b)-n=yz#3(i
z=li>6i-i?4;G(9f#o9E*ja&szn+?`y%JWW7HRsEsP4UDMoqtFemWQ-Oq;5ivuxCov
zD25YFrk~kzCl^v1dzvLHfZxgc6av;MmE|%}mLzyh9-Qfaeg09h^9H!+bw08mx<Ej~
z@z<OUIdYZwm4CucC!IM<&wa$RLGlcmyYDHALccPk5b|`suj}QrlXd_^h=+ivV|Ir$
zgh6GXo=bTC@l6?Xick;#AT>D$bsbled6^!E?sLuhW6GyFtX@|@Of@A&X(8K=;~x=s
zQhdTXz|Gq;hLMDw=uN%&t{vz{s4L{@L1?&B63gfQSZ1rVjHFlge5^`l7e4ynd+o8=
zE)K6q>g|<#jU6jXlwfN{puHW)ALa9iIiy*GSw1gQ*pZ6Hk>h=ps@MBBh&|bod=eA8
zn%;_KTx|4`ph(!*NL!lNEB?bhJ^|dLu{?i(I9D&d8_ax}|IJsW8)&Mx#Y)8#r;xFs
zFvk=FgY65=IGG-}hh){{jNqI!xwKo{>Mz@!qNQcGf4@*AcLFhqyBKO(iG-O#Z1{Kl
z>TLoGJg$6?XTVLKeED(pNm@VWu?}Gff1A&QDo2K2pmv?ttCTFi{sgI;qJQlnxA3b6
zn63oHe;8w3-9wwZcHqPicPf|Bm*_H;Wf}M{|J`+USosPG{0+%FTW=HMK|O1#`#>bo
zLkTB_mu4)#flk+Mrx?d&`%fZAeBh9i%WY%c8;a{h%}F~Y4A7wHD-1g&fuz2NFVFzT
zAThXCgt~)n(H(H^*4YJzhAukmDw`{L23}SGTcDB^f~EO^?3smJR;UN{IxyPsH_Eq`
zqpQ8;$p~2f8x2o@*y#D^I8S@mXBhuEv7nU$medT*&G)(*cnMbwCsqys;4CpAv|9Eh
zBIXgaT7epI*8;feIOGz*i`lLXoQjls#IDa)yV<ksCZh1j$)YqP^b)#115)WGczf^|
z3gbEK6Ht-nbY@A2e=tQis!CxH3jRQb<ggI25G(TSbMBKtk|fN?0LZlrx5DfqMY;fH
z3sD0nlRvF9bnot>D0>Llo7(!tJgWey@b8c8H#)fh!@aDRgpLeCAMyaS9$3tV!X}Ha
zy#fqlR&0WO2iZuxZTuEi_>O|GoRhjVxLhAM0+AxtBa82WDV50PzyqcYF|zkqwqyl^
zO_c(+@Xm?hP+TM%Df=vL1Qf<FWr5F;APoj3G4DvR38PX#+dWY%YibbyP}@#0U*lZ!
z9ynZfD<oTf$qRV)vW*>2@ZawMy4{)+1@mCZ)Cl)A?1BfV@x>OGuY)i2m^tE8aM$A)
z)H8J54UI5(Qpar;#-jM^AXIM<^%d@PU6k3ZkL;a7>FW3Eo!BIT<{Vd)(WVJyoE{*A
ztnk>Fw(0)80s&aF<kWi#uH4pr$zHup{Bvq^zWsRNs9r&YHU7JkPwypX35Fm9r-e|q
z@_=ooX~XSOz|*8%q+YB%kF)%4)I6BX_x9SHIatb&aQLJD#^_xmj)=ndf}}ImS5yb&
zoixr+w^Zps$<lMM)8q`sRj<Tg1s&?2RNb+?2Ec7MH7RF+5*#WEvtsC$Co;^>FJ6+r
zjbO02?2mfTQ!hDoaQC4eDan4a7sq}OKeiXn3bQZA6_w{BCYE?vI{>rSZItKeeEHJ+
zUvJ^FTi?$3e|;`dnz+*cyYrPufrhnt+YaTkI*Nw*@({*)*&Tdwh2bOEt6ay$KC8xl
z%l;zaw~jeau7Qw*5PLrxo{jOo5WCQ~x#qGmoo!AjSW5&hFkzk}`WMvnotgN0aHhgx
zWeUT)^mxat8**EYv-rK)snVe>th=xiF^mXxp}^W9QZk-6Kr=^j)$F@sz&T5prp+o>
z{zDD#LD$%ZIRpXp#VTD)6d%dS7z1cae!N&$*b>VZ4~4tm)cmt%&#ObYos8CvzWAy@
z+q}-dD&Q_*-mM%iTg|QPLE((wQKg>#57q-$nywcAzWI>~vWzz>!2%2oqCzJkf*BFd
zdKh+mxrt$kQw6tk;{G(GldV^B4gz>FDIOd7a0F$8?8`q5Ds&B;)b|DFS}N#a@ZAXK
zW82NzH8M+~E<c)^BW^Nu0TN{2R8MQOJZA2`aC!-gp4X9gTK0d|HfVBip5<fnI?h>h
z>=e#DrOcIEfFnOtN4>s~I|I?23UW+I?chYgrn+<%e1R(H??z3_gw?Vd3mJtaN*CZ0
z>QKKbWJkSVMYK(H`At<WaM}v%Bv&_X(y?}&esr7@<=i#ZQTyC@k&P5ciq(T$x-F{H
z#7Ux%oG2A>#(|m<Ph*XQ3W>RNXssOOOQ;%rE{Ov6EC$l;7842KQ_P(S5md|2j}eF(
zqf~@O*Dg{usDhKemjqY%!oAc2sf+95A$twJ@@b2Z+6Zc&is7jBrQfgJAxWR5Mn%7C
zlMX<mV6>LSq6Y1(prUV25zXd5m_!HK%gsnOyueE&GMBGYzP;ltc1fYCy>#DgN*PUa
zATun$XZW~#3Dz3CBNn&H8BDzcNvFg<PnYocg`5oVQIdY|G4&#mVIc{!k~Uy}LI;_4
zHBDuzOeP`I<R(H6Ks=Qa8TINv79#6)3>Y;xQg8>S-TVagIZ!I$FE`IGy01&5qlRfM
z-$8?3@`Gg_d6*hM7Q*BD+%0w$kJPbjQmcCTry^A~vnfS-YSlbPTy9FU=v{h`9Qq*|
zt3@Mj{)B4XTU>zo;q46>T~Z~k&xQD{y$kNrT)vtwSI})Ld1@DLqmPh_wW9H?_Xdrq
z*&be0up{my4neO0R3gu-pBxi1>vGewg%ynkjj^q(B+IY7C7Bzm=oBxIH4MN1h@vyi
zCU*|RFhxJ#64_W^NHb7lDJ9w!Kn0ZT!)j&VB?rh%q?Nnldg{|g>!;71dO4l_Y?&60
zf=p~EvEap#^KHxx{O`s+l^Nk5MBY2-jKY3L%%7PF{N(J2uawQ2D!`Xxzj2%7BFc<=
z4~^H^uo&|ZIhAxS6E4XxR$?3Gtit^r?ts-EC>a1-J1xk?m7;WkZ%{&R3#T%~TW3N7
zM%3v&?d_6gi0QNHUzW3qoMgQxDL61Q{y<bvj82Msd(|b2pKmmI`?M^!(ST_yoBTJ!
zQ@{bmS^i3==sRb510%XSP<n$zTM|_0P*^)}yGwa>>s{;sJ4J+zgFv7p(eLADqt?KF
zyp=)g|F={R<Y&FL>R)7*9nv9b0Yz6wbecSiSO(g4c39m0JXbvA*M>i=Yl8*CLN%z%
zS?-qLCOd@4zy)e>&j6!@n4+UHdo2Dljmjv(Ka!nYPsk@or_kpX?@E?CMKX{B<T~2-
z$&#g2aG*N|xFPs0*@7$mz<k;0G~LAGU+)_!7Q0f+qxndCHqZT{eY)TZ2}Ss&0P4xN
z?ig&1Mugl~%9`Fy_Rzz(rM6>7lKyX(p(6^ph&FllEh^e_mXo1YgZp1%MOA3%8NlO1
zy1G?1c%S`|Ea8UYU`GR7%36))#CF=`3xZ%$8(kmkY^Si;Q>2I`R?_>5{nTkGcu$>p
z464H@K~a22JrzPn9tSPbPh(-z*lDOB@wm_^{By?Mqb@8JAd|(%Nyx{fFDo7VTHWa5
zo%cDc{rY5_#~yfJffAr*%TUQr!)YigV4wl{*~{jG)|wOX{K&CK_!W_Zolhc_&cllj
zo}47q%LvPlJBe|-tL#7TKI1@ka^h<SpFEuxvdgO5db^!h4muX!@Q8VvqRFpeZlb4Q
zQRRYcv|9{H|IUpBSr%yrra!%_iKL7?#A~U26%#aj2dfS^Qur(J=aMYU#1!~GF%w3f
zI{gs}F8Y@D_iSg{${MgZeA^MO)#QW*x~~HAlJu5Qa@33aOyXD@G}bw81}WcWb?1uD
zD)VNP1H6RAuZ**JcKzSa(dxFF6)Hl$57g#;WuChfrFzGlGm*bdtLzm6VGiE-6uAlS
zpMTv=WbbkxR^bSAG8)N}-S~C}fEvYKoEuFCZ{WFF1Kd?s9G9t(M~*KLzlN;@c8FYF
z?%buPk8YLTkdo|Xb|LDMoY!+Yo@opHZsQzHOm3I)NRKH24LsER(k?(hN6`yQU`50m
zMDmCxP^W1e1J)x}+t3z&SAy@qnt*5J5POsjyv&`h%BGO#=#}qkgi}ch;@FkJuTAdb
z`fh&pH~Dh6rDET$8IQtjWN~2$_`q_}=#7=Ar>{*y=~;)<Y|R?Kp17j%dGZnMea{mj
zXRY7zlSQBVEk49#gZ>0c_LnV{RO6Ks#k>crzxIGqe(UvH<?ls;_~d;-P-yY!f=;xa
zTyS1StNo|(b(|)ZbmIe#1m(|>MpDloxDhtv;3yDm+yAPJIzWntWQ6otQt~8A{j29F
z7<}uO)pnheYQ%1=!>_0OKxW<o)=X7%`aU(Si4%8ZxzT>it2dNQgv~FwT2H<*CpA=p
z1?E*=pr`3ZkLywN9%0S&fVsEx_npkn`Q|&)7(U-jKNstY<?~cf`)*zEcS@Gt=eJi2
z8nYJIpEp7ZBQ(rL|DZ>#rH|=czN$-Y8I+<tU(d-&!P1ht%JwCY`j8ky-=w7X`%F$h
zRwy+RC0@h7aU<0-Th4ly*qn`-UUS7<M!f0y&?zzXwPXl#!?fJ5pOiwq(Mml><ng|_
zuJ?W%zS;&m1x;olKGM4R@y3Nu(}Tm9lqxyI@asjVemB4o!rZ)D?E7N`*CS^M<Sc~}
zv{y6`*c&g|fn6`jEqGo3s_t+S(qxW(vjT{>8*Bx2)!=#Wa}HnRsf#E+5co4QOphjL
z9?Ca)t1q0@baJe9lp*dG)SRiJz>j&6Qn2@ln}i|e)P=7k`rF&7$3AMmGa2eGIuY{9
zynV%8&bZ;t<Z-aH$j<NZgXg(H@O-YVT1n^qdrp(qMdxn(ytVfj5R8qwC1sDUW6b}x
zZ*ZE64WDE7jfw4pzc-FAnN-#~)~F6sFad2#;um#&m$^sJ*%)2E%$=HbY~v&FZIbm9
zPPvm$F)hhX9~kut_Ag&$I`lYKEb2nDS4t1XLg-tvN>MYq_=IMus$g{7YUU8=6n{hC
zBWn={Bd+7VV=txPDc`~wQ-9h0ar@SqFT}>BI)`T5%(#-s?PpQ4)`Rb;<J)9g(<TG?
z0jU(f^y#n->wZ1u{bouo^?^&tRCnHHg5ZmnC!i@6@8vw({HF&ohaZ!3$nn%$uF}E2
z1|wMCmHourwY)c1hkbzQFMj+3S3xA=>E=D}ngeCKC(gps6typVS;uTP6_!I}*t=_M
zX%H-9)DJOLbM=oP!g-N${J*Wm1p38%#wYLWl5ir4yC5&$Qe_&TCrMhY%AX&qjIJqD
zJGFA3ZH(47_XbH|bHlzkV7&`yhI4LH*SmD$`_J!Ft=(=}DD)fsr{QWUiv#XgFbG!p
z#?5zoko+T;4hG8U5I--~GWJM7a_ys}s~&mLrjq7axkPTMYYA)gF|aWBa&2=-B{S<h
z!9efj97uwoSL;pUevfVsK1&%8N>I+c>ysCYPwLr<i(po<DrCDn^2Mh^a&rc;u$FYn
zgF^gwh<iHB!OTddc4TrJ-Z;fM0lZR0DI`Qe66&rWszY&&fOq5PrnD^D!n1;V!cg?*
zs7Q`neCGin=Pu@+0S8ewpRx<PXJS_;lau!)rZM}%X<@Z$2zLu9iF3oNWk+CF$hcRZ
zvw1*#k=0lYP($XFnBZGX!kH6)N-{(AnzRxkbx1RHueQVU+kl4R!evtHbo6s*c}3%x
zw8yW-#_Jd#4S5p<@tqJr&G~hsV)#Wkf#v|^6{p8VB3BsWf3|_UdX2pvC9PFle#@hO
z3)^%WGTBPAM^wIi^j+n`9s~1fQvDF$x!blIVA4P96QdF0NZH}ygqDnQ@MIB@$kOYL
z9AM5fzu|(sRog|mc4yjko>XXU{5~ej%0BDYIBWHLNnBxO2n`MPF{`lpfOpW6yg&D!
zcPt!?-v;n-*j__q;}G34$JP_<gs3oFac|*)8iO%zzNVi!l6>LC6(vTd$G3OswjvNY
zODCsxIIQsrP2QnqJFuunGh8)15JxJxv}Sv3+S5BZSm?ABADQ;NIL(xPwl&su>*vbC
z`tn<!faBf+A|>R!%`DIjv(sa_(&Y#GTN^sGHTkh}XXcZv_L=U}+*2<%d!oM(pp%RP
zuj8*jJQ%wv7&Wb<-}S4f5>`v{Lz7OEzX=t03SdN6Qf`?fRmQYXSRMh=`UNXXVT}Rx
zr{1mwaTWBszwLgRx0CsTEVTGfAea2Y0RzDBl7`9mMCZ}>E(tp;4B^%OYYnZnAJEOi
zG=5>i!uWO9L-X9e%xjRA!e6u&;}RGQLX%VESN|9jy8<hy+|vf2^*Y<iwaRcsx-MtZ
z2nY*LG_-RY-(TkZ`K0d_c9e9Ai{f!)WE?18PzwLJp!v4H88wczmCh?TuJqC*=BH{H
ztNO{sYYEZc6oanvb_lMW4`XJjAGr==+1CR7CU!>uS+4d;X&WNtvCMgXefV6M6SMKa
zfFgK8xoAve#<5uB##e`Wzkog5KHjG3*>fNdp%9A7h{oaiR*2)_aut>(Iy`>>^kj4f
z;~M`=MBZk8<O_W#^FrV$K@W!+r!nyV1GPlwM-b0V6xJlN`HutuTP+Ofosa<{SNDsy
zb{<sII#O&lqf+lDaq60lZV7hwS*F2DHeJ$J)4+_1eB0+0GXI*v@zH`#x@#G}9p5Kg
zu-{8;_?A10y2Z2WH1qUsX8wZV>{r;QilcP(^p#JVIWvr-(c1}K0<RK_JX=uDqsY5&
zd4()rUB-5Oi%-!bds>#hc0i3_?3yrI=czgS1?CYr_9$M9dX^s9{7D0==}GfkuVPHf
z_O|na&frhZUOy92_6i^+B08Aj^FvFU{?p4KYR`fS?v1ziE@F4-5l!$qD@E>z3_^{;
z7k3iMLTfk$%%{w=QhQ6KPv7@_Hj2SNzk5#6w7diosK!swF&Ld^vV|}Y6jwp<8hc;3
z2N7~{=`8Fh>$yR0zUS{t`g_bSW9BavZHg!c;i$nsB@<t*DBOZ#V!AB1!B^if$sF{>
z$T$CH0a(_s2{8S>GrXH#Cm;h2W52kEcI3)7@^42Un?DPanhAO{Rq(C4iau=EdUufW
zs@ZW{xrEmJlCIZN+Ro%r)Ng)QENfmZOOs^ZX@Bv*mCq<ExO{D6!nh}14VvVa&0hqK
zK#(Q>qV}y4M#B)dHqa0W*hcbJ_a|JqLHvTlC;O?A!{#1xb(sC#7zli}ExLsS?O`u3
zNE7HK3E5ka@Cg^R1GWVAgSF8^`akiI=RH80%lC;8t!F)jjr}`o*fB7N`er9fWZRRT
z7*5-@4JF;Q(gGxd4o-gA*FW~l+_`(eZkk4e>SjI|(Dgd1dFe#lHyq2`x&=|MLh1^K
zLnpDQ9tqmzs2s87BtZ?8$jS<Y`R7}sppT8N$W@Wku4DII3EhS)bW!8n@p?|$-%IVj
zci+?3@8m8+(iGdLd%WHrdPeO{OxC8sA)i(1v&{^HMxxYosoGyuO^!s$rrr6NQ|PZ|
zFDNr%GR{2yyBx$5y}cZP-A+83M1qz=9rcNP=8vM(KW%+cMUS@|!?kq6e^OlWF~<ir
zpwlC>`w$+^7JFh3S#Z@HZFWe)OYGC2>1#`_e`R5(>^z1$kv0SphtaXucs;xlKGS25
zv0lp)_W<U&20O6xTw&vTR$E$hH!$f~VAXAHxySw@hOoCYJ@fOIH!&_#%qx)g7}kp2
zBo4A$AN_15-+uE*l!_kL?D5+1__U3({5pvS?L32*{b!NT3ELK}5*k+ml&}uA+V4jT
zq{bXiCzLnOc_s-4q6<!EPUb4!kab9&u2nT$0d}!^#SM<XgX3W^jwvEu5Ln{|?u8N1
z%rEI@#ubfmwx1tFoBJrUM{)6~BFLUe@Tr9kon%++5W&t?rKp`%%iZaVZ1xP5&MX3-
zNo?6cfFXFN@o$go^G)TNAFPBWdQ|o5eGy*Juu}?J8#t$u*d(`2X(^d#LvrbY#eePk
z_&DhMmvRtUx-VzQ=1K0cs=5T;Ayvls`}YH(M6N(%1rh0i2<c$&J5CvKWpH)627qC~
z_L8zC=?|Qh)NK7y9%{2z+0eN7wd{)wW?M_`!r7n3W*#bE&eJ(8C)4ArD<@gkFH~~B
zI41~6#HBU_529gyUO$#KlV0DxZZ-~3K^6=EqHO9R{Ky`z=rIH8LiWo6Z#XPcpQ@}0
zm@VcN7dL+?iS%LXVvZIeB`gP>!amh$l4Z+GKFL9@GW=^}Z^5cXaSFF<+$SiF6vqlH
z1Pc<$pCt>?lpnX_r`MCfy=oNqYmaINK8HEE1A`;mtHeHFmaI3IAOxjr=gKJ1XfYc0
zmkH@G>(A4P+3fx;RA|Gj2n`p@8~kjQesV7H&wYA=Gwa%!oz4FVny(jjc#E$1)vqsM
z-WwZs&#kfX=;snUSl;k6!x4#!-VQzr(@DibS(|dMX4n*~B1K;C%v#OWt#f$x*KUBz
zMfzvS!(<Y#j!um_&3w*>5<+&&0S~URR<%u})a79Q;RzgJGw$Li-tD@JL$;BN;zOL9
zk+tN--ZkyN6u;sh4T}+A42GI0dnrYAoPTwMo+I5+-lnz?JA+;7u#Xg6BSm^ausexi
z0^t?CZ7JSpU>MbTf=#QjMZJ^OU$j<e#8kS%wEK+Io<*jHz2B;+R+jPo9nxz<4+R<3
zJXXiY)7p=fwr+!m-qCY5?n{0a-_Q2cZBMeDHK5nDPZiAfkUhZpQUjHBhcMeI!Df`0
zmC@TG@86=LOuYhbDYaR?ixl0e({0@@^Z)`TIES2sMMCSpiqo^vqxC0@lnfuLFFY|(
zM?gLk&LvAX+I_8wI#l)zJ{wuQQ^-@x@bbJwgOg-1(NWga8v1}RAud9}07^7Dbui>z
z_{~jNS?~LBH>&ft+zwyJYxrz~BVS*jgtstCJ}2y!3Vht3|9ef#V}l!18T*FED`tJb
z3|s3y<Viej`Obl!vC#n-TCNT%)0M2Y88pQvxx40~(r0(-Z1ASu$wgFIinG8o1h2$l
z^HY9al@F#;@jk4a?*6^`8Xh=SY(L|3Ug6s~3q6HPbJEW0h(?5V3?f|%V-CY#_jcty
zUvPi(jeSvo#^IAztMu%TuWVB+C%-93@6dM6ee4Cf!%ZPgfVh6PGQ)xJv*iI==%3kZ
z;CK8J^)uEI3T?@n(iKX1Mu{sO<HB3KxAA|ST)pJ*Jd#%9plovP)oc5P^rBex;%5ow
z87cIf2d4O%#$~$HC60&Cs4hA7Puokz$`$<8Z|*@jqUOwjqbG;JdaubGNd1U$1dnBm
za#<NG3>B>^O{i^XD{H*fD8bw-9ZkgfaNW3&)^b5?-K1eWUzJ>ILHR;bjv)H}_5)lo
zw;$ci*VU=LWd#1nHp3sHzzk>sx|v8d2*MR+Bd&0x<mMGljE9{mY$%Euu~8)vQ(tkq
zP>L#oob=p+4vP3}B(x8{iA`LuTlMdV?AI8{UiVFircs~?f|Fh+bZ_HnNu5sl6*#^r
zi4>O$ahn)R%ilvJT&wY{YQRS~${mi$RXm3A$}2X7=VsP?1=GZ8@cF5*ilM%zfbZn7
zDmj>ub7s40t*e|B4m4;i2bWgD(JYroBYdl)oPEb;7nwRhFE_0_Tiep)kp}vb@$Ut<
zdH@>cN=U6q(G`B%^?`!E_mh2v{L|=5JtcfZBDENmlxI^wugGU?mLR7`Bt`sWVhQ`v
zX1^9&*!?%0`HsalAd0?SfP2NhCt4u>k=t?N8c_sGQE~G->@ErcI4?BnoZdAC`dOyZ
zWiVF=PbJMOpsdRW%UrMNaj6!mBP3UrukbtN=4e7m!VFikTvz@&|NC=Vwoy$f!pOTr
z8VRWC{M(6CFF^Lq^s2nRd7q>FK(?+YVqW>03bV4xEKL>X9~+n57Q_$`VL>WYmU@p3
zH|IHfUlMx1LC`Vd4k;`d7V<T3RZ3;lFgpm}ppkkS#&9l@@pQ~Um8J*bTK`_AbV=6v
zOF%zcShhwFZCHcMiC;qB@ngj^3$3_HmC8O=98`fc#pO|%YzL8<KF{v4jZp`k%V6K*
zl|sMCO`Q!(Ii3hjkvPAHr+QmZE)slyQW8B=Chg}ymrtoxP^W+DAdT$WrVz-ijkJg4
zFVeCO$}b6L_KRsOIed`p%ZX<Qce<`zP#T($j_~7Y&qLI+jn=BJQdN0|dI=Wnr#38~
zIDB(@?3Mq3spA($;RAG;@}0Vv?_h=Sq|B-&64CMG(8Ld<csl%ykCT!TdGp(jmAvS<
zmX)G=5!fO!pg!hv5xoIyMJ$ULmJoj+^H0(t=$;CY6$$DzSI#{11E?8w7wLYygq(Lg
znQ1Kdg4hlSu&IZ^azB#ZsgWwaUh~z^s7nqH-Pa+-je6OS5F3m)X7?Bn*|zwx?v<i#
zC^k40eOa5(!$yWh{H~^q_QAjFz1KM2tO3s9j58<Qm*aQ}WvZL{6fWjU<AJ+)eV9eR
z+Mb~xUJ2_~B33Q4gY;0!eoQQ&SW9TpHEm-({H5UhH0nZ&GQ^O^qJlL?^UsxAzl(bH
zb)aI`-6UMKBG=rwrRQi<DuO>Xq&zIFq$W&{AJp2Cf=GZI+C{qn?oU6R?6H0!8B^%*
zuvav_lkH!}jq;(?R>Y;-R%9C7rSRgAk&M^nHv3XT&%|{wMc3y?>DAn98g4soY3D7>
zsEW<T=5ADHk+D?N>y6!#m{Z@pXXt-*I$7?HV~Yv^XhHGy?DuWF6TG|KcJ2$<@T?YF
z?=aO&DPHbXkEE^`ew5{p2}=Hbr?I(~R`CK%!fD?b>D<4MPe@m<nZnNbW3DVvm#Rs>
zH)ptQ8758s^8horL`Pl{%_S;gQzw+=)f4UzMEfev)j8yVnL}FlFHLHuF5<~)6@?sP
zW)G8j`E*C$FEoP+7pVb8!h)*tC4`XJ%V<p`L~t>SrfTzO^JwgLXI8A}FgNNtzyo|w
zfTw!)*<H;kf|ZKo+uy4eOaet*$kF%Qj}=lzF7EV=uExuJSJ`bod`B{@FzHKV!5OW4
zLU1{*ggRMUc0h}`s6(~ob^8Qt6H|tKsb$l(ElzK(d-G4Fa?$EdTxKeE*0u<Z1`rZP
zw4;a3`j}BuJ)YlTCVEZ2ltb3($3Qc<yJ&9t?{LeMo%J;Uf0<uw9WIbSND0rMN>s_k
z$^ZVrOQxSN0H@riPN$cj0(@lB<jfP<9irL_CjNkVIJE?OF6tbN(!VoImJ!Y8I7H)=
z+|YIM{F_s{Pf;&I_~Dws>uVx>Eo4M+$aZUfjHE^ewa5#W8?*M9bNQNVm*+dcS6A(!
zc#bwm#lzv=Pznx_15WD4{3^Lw-f8EeYw9Pr0XcCF+5GJD!Fdvc;p*YEUD2rJ(8K=A
z&pUL}o#_WBkmM3iV7|15)o!%291)FHTn>$+kb0b|72O(doZv;)Q4ZP;svhopUqDsQ
zzQXDER!Hcis@YHzJ+tyC**iBmQiXN;)TVHY@C$$FgDz%osA~>pJmy`$sv?8Co!tW2
zZD<Fr-VZU~$eAK?b(>OGQL_vb<J;|3^rhOCC&~L`vKHCQ{%ys4$AfNnV>rXYf9n;A
z*cz!3o@p#iyFf^%!<B#6%XHz)@9Ux(OH7JhF|5pD#H*Jtnqt$6+lhaYz-e-X5YMJ7
zXY&K&5b1nnl6=D=|JnsgutD08MDoY?2+28vp0CTZ)r)?E<L!qLJd@<S3Ja+OyUfYq
z)aFN)NU;VkR5}%e@?Ru>OP~1DQWjc~ADRt*S9KZ#oec%7n=&>|(kkio2kKv#Vbygq
zWP4ozrqt%Z#!SFQfVAj20b~i6OD+f}7fkV0u1lMgi!XwKwtkF>&9mvVT&oaQLz^0y
zD02*F;=Q8H8-q`=LF_bg$(vw#SVCw%*jaJcm&ja>uv!je1GXoZ55~upbHEFth6Ph6
z8I1~`D+E1H$ftrW5@y1lC~?b8PvBXRPX)UR;=;oTG6ivoiVP~}y%6gP*A4h|^|SPB
zMR<wK4_8m={~l>XL02r`r?_%=j;GT<EG5bj{+3I2i4OQ|5H$(bHF+6wvg8~~UfFT&
z{U-d#Fp8VrrT&qi&A2#*DJX~IMvoVr$d{%*N3+J>-6z7e^RenJipz#y+78Ipkt>P-
zTdI%6?SXgY<a!Ft1sg_oV>W#Zi{8-t(_P@tQ{96h2_VXIUTBVD9a6dXq4ur0acvxy
zTBKYM8~Inxq4+{_5}DYZ0rxGgG4e`q6T7fdiAR>W+7Y6bV0+48IltFCSv#R7fiXc9
zCV5wP2dD2fB3?2AhZ<II^Nr&d%O{&hFm(gf(O|q;TYmNz52VzN=!y;-4%EHK%L<XS
zOF`7ZLGIt~<reAmh<2+JlCkS0@UB5t5>Gq99tt0~Y)I+}QM~I82?b(y0GUTFYH9ir
z1R5I6`&ksp9p3wJ-A%pER_e(yUh@D^*+A0NeJ-qEe9kn#9V|BmiO<25!3HJ?ipt~C
zFle24^kp%wP&sqG=9=xA3c~O(u@*Y!7yT~M2h6OVNMAqDhsHw~&f_?<xCQy$@p{Il
zV#{nPfEm)}X1>}~^tmlX6ZjArYB$yu(4aj}a#cw!^HGac&-Q3)P@}RTh}v;sgSy^n
z<MzDI)bE#QW`rX;S2ALcZY@ZpV+zf^p4C9eO^nwfSSuZWpeKA8D7O4y1UIvcf{U!X
zB=AqxR#iQzg}n*Rifo>eiGs<18=7{1xTyv5=cW3I9GBqI44<~<vFrrPAMe!T)s!&y
z__en_uf}Iyn=a-Xk1gwNJgcr-KA}2!Q9ox+YfxgEU3B$m)6!2f2j*Y~Exci~!D5fV
zaqxYBQk4<t<pT!8A0Fk3%*{-_WL_)OO8@T?L@Njg`4Y02Zk&Fub!RJWJ#*T15s%sD
zb?c%NtZ`C_A&UgEJ3d1(X8q{bqp)`#R(n-<kO#vZ_3sxD?5=pj1D?!R;%NIdTKMl@
z5uhb3ar-`ZxNYslutNQI*G4fCIZUTQ?_{Hw@{yU6^+Ohk>*?@KdLh{MEpB2**Sm10
zS1zfw7nU=fa`cxPBfOtv-p_Ofpi;hs62FiFyzp4qwqlUh6svbyIbeA&*=p}@@o3?>
z6u)4%+iP78ZgpoXe~)pT8@3Ndw_VN@h?TXh?UfDAE3}XlDza6OS|jD8F)#kn8KwTX
zH_!7A%tm;JewH{w-;_B>aA-})-lT!N)yY)sUc`wB*cw=3NhL0-8~MYQ=ehHZ#>4cp
zT!waxcgNXRY!rVVFmG`w`qUiMNQ>pjPpH4!WX+F<DJy!}Bs-5ODSR5aOcvv}Oa+Nk
z_TktYG_;lOS}JcGZ5h(4!w|ckuw2TZ2$)=tWBKG1*@=0@W&xh*wZ1-YM%9G&n8`-q
zQOIX?oat%HJ_*;1t%!!2N;h9!zaqc4P7)?VO1!=I&H&{nlg%n5P`^?}WL$}Yse5X_
zWYA*qT!l8~!mu>fpg4y=2(A4II(eMd{uWUWbntLbr0e~sD-b>T7DR=6SGBgmgV>f%
zXaCebRv?Tj-4JS;bQ{Yw0jw_6cxzNbV~}j@Ho;34{`PJ}sG`^Y@0u^cp?)ERIE*3p
zu1T-cXUR$3Y4aaHtM+_M`3rGJ64S)8ry#(&jr#B+YP-%o#0Lp$j0mh?$7xxyl~T_H
ztMa`Y#HJuuiSG?XUuKARkg1Wqwj@BNfc2e@817)fLu!=42zJFl_aHhh3OU115Y5>Q
zrszke>;ICX=Szd2V+!7;peNMn6_14VH7mCISzP$pYdT|tm=&2Un0m<A&qzzaNl$Pi
zZq4^P65_QH{v^dEOjb-`B0bK?CVRZ4P~a$lQK8|%Bvl+Zy<2f9Llc<4ltB5c6sF{)
zWtFhh*KScd#VkPW^4Y>;1`e`Cb*Y!DXE^oNZmU?D23OpAAKgn!ZC<aw5OHo9B10Yh
zxbr^BqDe3S&#inWeWZu!K8@N76-U*?8E%JPED|+l*cs8{YErDc$X9E@&4!0Tv`G)k
z67~`IKX7{s8_9}mVCmi=R5MjXTDF~NP$&zRSC{pH2<#Q6KSB4M<1GZJy-dmO|CEaQ
zo%qlm{0$tg$E!+$qT8pE)CqVJ+Ip@6Xc36&{BVnj+$bhkyh<<6q<flbaf4GT0hQ%}
z$hQ%_XFCip2~EGOu=hNf=*tceyIrJCEw58C@$|8hSN>A&!xeVg-xTSivTZzs(C@Oh
zO5$oC%fXSlrncj0u8*JURMIFci6^d<Cx?}w%b!%OwGM6w)>EMG^#Pz~>Nn}Dl$$R!
z{|F^)rHhb*zt^Noe{h2QG?5f=7&l>|ILtz^j7igV81|IL)spH_hoy4aP>`yNmB_v^
zef0dE1z4jlFt7$3YV5mFy-H$lHha`I0BE5vRA`CbWT$+#nmxIA^Y4DoSZraqB|u8t
z!s3q=V?x1O4TfcEnOxv1zA+zTc9G&+gtIDL>D`jlbnP_nqln0cOI5H}1rAWe0+h9E
zttJt?md<y*Pm99CXTzU=IYu(oGFW9TRJCl}n^ee;-+acDiBx{3k@iXSLW{{CO+=uW
z7U#W#KXJ@Zx*8;aB97joWiWMkb~lC08~3gV1E8MlUHEt`?gk;#F2mLb?t!Opj%HQs
zoLYk4T|K7c^ms|_)$3S0O?rg%U@Mt76a3bmiUDsIvx~__i+;}s_IWRU9Xs(5(I~%U
zFTAzN51qF;g{=f=l;Wqk`5a~YyZWqFw1L)*?(^*ko|vzdCx0fn*Sz1SxNiW;vJC9O
z?g0BJZq`d3<}oTc&|bJ?|2!*tu)sKhXei+k7w@!5WHLYD3qA{G3tI2458iW>U!d#$
z*+4y*G~>g4&zE5rueNOdSvaq{^`yOPdg~I*OG93I@Q|^;LEOInLAx8i|C9p$%<H>`
zZ$tszp1Mf#6JLoA=@`j`j)RVW=4o8`f?8+TwCpyViFgF~rUM0x6xh@cfPaMcDthCH
zN8odcoP7;qCCtM=h6?Z{{*3QzQpFC8<O?!oLR|pv8D>Su?dTgQ4;8;k8tC|sDn0cp
zl<d4xdvf??4POr5Xh+`72)`O_GB35$fW*1ROpS5y>0Oe4ot3_VJ6FW-%)*{O=*v_g
z?A`5kK|A+Z^%?Xv!*B|gd6F2S4xC5QzrZq%qCDsv=^mYbZ9v3E?ZyyjhMV=fOaunS
z9Yt9*-6BTfSrXEmzgeS!(%n3M9*Bj)^^aM)`jOH9bkpMljaMd|Er?AGzBzMFO#3(w
zO>i~My`tvKOyT}|OC-TB#i@u&Qq{4G-0Ncu^3H~27T))PJVh-QO|B;cuebejSUr>Q
zCVD+oL{7m2PMlPFKrP0pH-9KK=$+7N$iXc3leY<CXdpvDt=|vK{eYy3{k8|b!O!vZ
z&{}BajXSEJOY;VHZTUxPTjY*1Tt|X<d>`xQcZARO(#pyV*1gY}2=^#wb=!gE=9*28
zA^a8F3%E-@hn?R!22+mK&()40Oce5Pbgs=38{n>?5p%INCTY1Pl{!qOi92*|x1ePt
z?K?2FZax9UdmOq9mKJx{lO5A{OGh=1O1yJc4mL%HJ+<2HYOClPeNU1*j7N*a$9Bbj
z#n)DCZ(dx-{FOT&-vQZ8SSa+5D>K`#k@r2I-aWUVFisswjl~7<aPQ&XfT!;?_Ac&R
z{AMCnB%ZChA@T60s(!P4k9FIs)HUtJc$I|?!3xV-_2Qj3QFx<b&#hN)BVqvR$egfe
zqI<=)6C(%|LX8ZGIrXxEFefVkyZP4{JkI15NvZz)fn_^^FXYne2F0a6nFJmvk+pu+
z0s5_xo&p3RGSxXDJ~4G#vYzI3h^E>S``RON5`+3}^D^}_+k%|Um*x~)#_ET)0YWVT
z>g(snq#9e+EPBar{XkhhED6P9Fb1wlSB5akRG^v}@zyVXYkthq!;Qaf7Y4j^{Gan#
z13m9<mp%<>8{SY<$>pl%q8|!0#m~snoZTYW{*&IpziF<)nxm(9MsMwPGI_whHyEec
zG;X%zP7%_fid2p}5Z#642xo4x+`I@s47Zb~1MD@CR+j|2Li-(R1_K>Nmr~H5j@j!v
z!@+rfZF*z)pq*LCf5~S4^+Ur2rzkzP7ENkd{4#U1CwZ9{R7*)y%(ngKC{dE*K2Hf{
zVa0qbn_v7CS^6)0q?#5bi|8vSkyqn(t=9dM2mi5bvC7NR6Bj^pVJ)OF5O&2D_85Q|
zg&D-4qUY~YHGGkD0Kn%u^obAhC^8^XizZ$s^!f#hrb-i!sbT36kPAh4cpStiMT&f#
zrEdclVh;1_0NUqyNPliI6+~fK=q4w*k`aUEx<ZeTdNA~|p*JrNSzI+>b>3jy(Ew)!
z@F&cBr-uE$+!6g4Oy?=bibemU!sdJ;Wu57fg6APhWUmT@^;52M0GoqBfBGbq0)?CW
zbn!>P!1qvTS^VvKeWhF7sp)Tv1XOyxPfWR1#2&rkT6FD?w=8q#dN|+8G<9@xKZrZ-
z+_5(_E>t^kN|eUA3W(5W92Eq%#~z5cqJ3Gdv|h+|oL{JIZ@#9IPKfgEqv)&0M{^&J
zfnvIkdqNEZla79SA<g`>9XN7%bD^2~*ilL*`I?(L({1x87+u5^(>iw_x~wrPl){T3
z5`v?M%EasAfnP-f&JUxagNKBWs;ATHxyt}4v#<isz3l<$B?frCP?>DC(_CTN_+nJM
z56*OT^2zJj@v>%<&I(hB!wh%4>VR_G{(n6#-0IcS#5S8(VQhtwJ`PpLRc*>S-l!9L
z>(Sh$Iug!{5im;j`dV~|RXhB~9iuPI=fau2`uSf*zQ4{y(@aR`+P&Us0D!?{Iq-!#
ztmLJE<Qz8SadABzU*@xTiJw)}KG!J8fB@*PbVy3CF*U=#6<dEFy~m*ei}&t1$x#Ou
zE-om;$Kn#yKLjnZ-^<W4`166OrUi23%6ch>ObV?)dX$4;ci>NJHLWa1Wpz_<OV0ow
zkJO;JUCOD~8w~@?Rp4sSSZY43<;$wvbP`OBuTPic3X#EQBUJ!g3q}r>$9D01A|WJj
z+W5_E7p)mbfG>1fdlu)>R=C>2z5|o#Nd|N&?A*+&A1LBwuA9Fpd<Nw8H=hgJCUM~)
zA@Z{4CZ+23{o)}VCG%ng-hgo{CLfpJypP^7hLq~~78c!vZOvD?X-otEqf83n>1Ar%
zvgAYCmr9vm9fir1nV37mSrpfuV#<vZ!yJISj_9LC`=dJ{KjZ&cu0I^4my-l?T1B~u
zNSoInqJsi<ZNS8FJ~uH`nf3i(f&Id%YtYD(%I=TXZ^?ALDtW#@eIGLR|GmjvIlUIi
z_JX}q8cUT%Ovy>{Az1xR(~+(Af}_f4U{QEGWP@=mrYPF%CjSA%J%h0v+n;(@t3x`R
z(v<m0{QmZ1WhwAQe^wDtN)i7vzWB$&QG$;JCC`Vd`|nj+J*|5l)03<%EmnYk%;!R8
z`pzD(lr*BOS?rH`X&k@dx+@XaeySx_b6%$hFwEU&0R3^0vJ~K?#1#kIE*@i)2{-aT
z5|x=E0m(fqWtSto=ii=CjDNHeUlPcMP<k%gs5JdHx~_86!pkx8ej4Ad3I;)^5rT*0
zHPlL^S4^I%@e8>xt0H5G(^Vg-T*<QvBs(!jijc%hqirM)JF&kCOasY;pDzs-9Fcy8
zc)>s10nk$YFS8ak9_m$qHAPeN01$qztpd=11~|pq0wy(f^=~+UHDpxwveS75<Mbhp
zHf_5_K%9+|K_)>}HfPRa*V{fE0v=-sPxFtUFx;0bpY*3oKdjK6h{M>RHwpn?x=y72
zELv<@9wB>+*YRqXA<<Csn%vaU`@7QwP%Qyowgba*=(Btr!ogc0PU@M?I*qqFO_J>Q
znNz-q?!X?rK~oQc%)bHg4m8oz2h{;#Le9{biBj3ALF}{(&O(gAKjNX~DKmuSig@{a
zwqM%nyjZJb3?{+B^E!l7lTik6!?7N7PbA;Sak4OCNM2~4UF5OhgBJjQ34#_v%6a9#
zQ1JT>WP8U7dIE~<ji!AM3a7)R6f5LwOk~5Z<creL9jZS4Dq3rs?kXTKT!RLdiP=pW
z*HCv?Nl)XP%3==jhz(hbzw?ibg%`_S1$MH(ZCCmY0H`O2K@P2>Wa=-baG}4}ukL2;
zgM(M7u+2G71HHUz4)A1JsvE@<JG2)K&w_o@;yH>U|Aq!kNZUSNPek8L9n>E2ePfNQ
zKQeyzUiew5gNP1OazZ@AuG6>3l6lVB`}^!w+9x^n*vOgl(^~0=-!!4p$6COtu{0cp
zUI6H;np&X=SAdbV{)xfIha^P;bWgpC4le$2ChLF?*&zBHyZ_h9bq@?)*G%ZDN!8{0
z+66>=Q`Nr_8_pDZr~+SZ-ITFn0!947pw+tp95#KCAb~T*EC018({z}&!o8wB2Cu$`
z^7#pt-gw`K+Tju5W&j_V!Ry{7HZa+c4RHT}tG}C7T41k*ry&{pucIdL<8j<A=Wc*p
zMuo6^$Tvzs@DP(cvrckzu>-fxp@NkG3p-AW#$9!u76d6>?-~h$l&lrgcckU%xI}@Q
zBIT%)f+D2skCnK)ZriifT|i#UUFzaQf(cP_(M@uMt#Z_Ld(tk6fQqY8P!r;g1Ggir
z-J*dB6GEKlT$ayAJc1UfEId0`9=z$$l7<NoK-_Q=>AC@RDAii4V1y+32w(?01PJ0i
z6JV5bg1%3;JWd~Ymk021yW<bP&?SA(g;A}^84v)XyeI}Lgknh{6y|rwnPcv^oj8_b
z;*&i}sA52Ut?D0TCrW;Hq^}FcAOCR?le_><vK#|>Kjij#iohS@@YHz<yG;m@g|`gd
zulvfe3$(QtfoC5(T`yb_zZR`Trze6xbp*8Yr6qx~K@Juq`^=I>005KSchB?xN4{oh
z!{7$-&2rbc0k6CwB^cSk-A}NEG{dFxW1}PJ5HAHI9$Wi9_BL5t1q-|;B^#SRQhWH3
z(ijNn!b_CiBbyiASBK@vN);`Q-<Cm-$KSR$&>RmD-<MQ`H1)`>iI(>}lQ~>HX?+`a
zVlmZj<k<dg-W%v=*1sg{=>ptXA74y54wApf=Ov@-L19@HT><J<!g<pyV|&Kk9=RKp
zPer2~q%#0A<I=!CRT|=4dus81sc84y>G|mnkm%Kb;^3|2ya!O)aLb?mI1LkNdzT7J
z-0A|XZ{!GW{)j{C({i`tWtq<xO9!{%XLaIA#$SxsRLb9;%uJO7e!Pbf!*&*i*Pro&
zegc%`_|yF<33vffttdV0Uaen(^AbkLXJ;u{JW~2OPaHDgB-^3bQAYGpZkoFr`0e(2
zq*&7{2@5~Lkd{;uw;aDz$Ct-$`J);ZjJ&bsk}wBBU+|UqIb^@ZFTFmBAw%IQ`jZ3w
zN4=&jXuQ)#A^u+T4ouJfr0sa54z0bU9q7&x)D*i^A5OVd*D3w>ILQt&@qWMta)}Gy
z)2K>k==78~DjkWkvF(3uU@y<4!p!i0yq-ToI&|`Poc_wG{w-jG;28SInx%m#Y~kwF
z1AG*=B#)S}C`UP~qp|KkjAqj#1EXKklyY4}FDvM9+fw%q{j-hz9~25HQd6@O#b+&F
z2mwp&n=PUK;3v2*Lp9ByY4Tt>rtn*w(Z0X&R5yXMOCzts*@DZcRBV1?5@hJP+;l+9
z$$gqKrRiWOGFyTABB%m83^-4Ys=P_qp54=<J|jLthQ9U$NH<H~IFQgq;oBReq@au@
zN`)MsYEr1}A0<-tBNP$7P;H9MTW~-&5EmJ!xI+mRk_*z1n|=*KKwj^01XMw@-%8*4
zg44x<m9z%fmyIHgUg9V`hiP|;u)JRXOAjD5(9Ms-Bm&ea{&tOJrAH#Zj3c3LJ2)W}
zEf<eKno-Q4hwE~_5ZOB+uv=fCxWWSRJ*w{_QRyK*MVB#!T1AeNzd4VIgl9W<{s3Eu
zwK6g-L6nmQsa1b0K+fn(?yJXSBT>%+a*`bhV8{kt;j9zzKL`b0f?h1Ay>gSQSK|BL
zA!9~L(}hIH4=H4zgSKNW)^^>e;H)uXl}Dbk(yJcuyr&$*_;#eoiQ9u8G{zjjt>cbQ
zL!G`ojOaRf>&4WmO3fyL92Z$~O^%f#ed7YM8KgMHLy@kMtSPUQFtaQtoPhCa2IyAY
z;Aa43f*J93a)<NaU3^dv0RfqS1e+}Wb_yNK=~eBb^*VUB#GUc#fg5#_uo*!1P*U-m
zpUkzoqG^2!f;plzy=hC2V|`a2kn{=niT1VD-cl7w92@#oYU{ROlhe9{@QQiFOfcli
z@HEHfLu_s258&nWDZeW6NoCxw1OH~9i?)ucC7=g6%Y%0rPQp*Va8X)*1Js|Cf0_qb
zeV60^F5@vEERM@qkbeB+e>=SVkAx1|TUbaZ=o1CQp^wL|(V!Kq=@>)7b&(<*d`UwU
z2ov&d<k&5KIi6P%1*<Q<TYS-XnB4PP=^D&i<iAG{2DBB%1O9mNnCz`Hu7ucOCDuHc
z8o)cM180MJVj*HU5EZR33pqN7GE*nkL%}N?lwlabwkiLG8N90dL(*cfdr@)TS-4k8
z1h$jxEP_m$#PS?E-2$}o)^eVF-_Ghq{h$-Bb<%iO@ja#Zk(7y~K`I~$w#asa{7iYB
z&nVv+FqS1u9wc>zYhMzxsFd~NN|N$u<RasQ;YjDVR~ex6!wYg$^I&qEJE(2VlTrR$
zR^FrB_Cr<%$$aI=zkmK|`wySf<J$=zD6qJD^(9fhgT)mZp6=PFA1ZQ^+56sR-|BA=
z6gDLwu`36;MWFJlte{1bv_sW{Gv&}N?XsB^*5ys9LB7~RLq`pkgA1<kY|p#jBW*~q
zV$p+5$44QKI_@XsQ#F)8rBlRG{IHlqMm}iI!@?m)_*s*K2%KqkSh>qD+pXZeuv&(0
z6Plp6G3on&CB-}&N(pf4yVSnQsF)9t<QqXDk9Ys+MA_T}?yXR4S^wsL3flkd-Tzvd
z{@agZ22k_;9<_UfLjbKNuq7*dQ}*y~6Uz^`0wQ9Wm-tv!Pl>SR9t#>X5n+i17aWQ6
z!$Y^%`Qht7-vcA$zzx#ZT_!Gu-Wva3zv4gtjV~VZ8W<pl0`Q`V#X>*8D=et4cg#td
zyYJ8X!>y~li<9)wTj<V{@c79g3(q~Poh)Mx_HsnKBzw7}q!94;0H#x|Gh1Ty_<{e;
zllvNHLk9;CWskoGG;5*20s;okMz&0fU+QvzUk6d{_7@2am?<wwuw!T9jkkK8ty<%m
zlU^}#f=aJI9ZmtoovaI{q7Z{hvH$mq;;(t)Z|V4dvu)Cq$<8)qZs_~~kV}{%HJoi2
z0QDSm-pSo(z^bAUyLbJG3#ZLkC=~JN$kR(OcvT~+cee+TTJp9Zks4mw2Yjhag-!zh
zuV&l)e=Ujs?MD6QD*iA3kQMVcu?A1G{w;j}?@#*w`tkpVKl~p*)3GE76TnIo4A5MC
zbNg0+Sf;-vj3IEu=?22h__Zy5W_aMRQ(+xcu{R(jU;Hn=#1iI_5^=rE{Rbg{*$iWd
z0+Y4GxiGcW?<?R?AiM`84Oys1`_Gc4zrt|<NRfYYK~LutNTdXz^<(lMk5#Yv0r1Kp
zK+9qUe9g-lYlt{WdjJ;fF1ntg*TwW^j|`Jmm*_t#GbRbTInTSH1BUn%<}vn(%x+UU
zKQTa+1tA$aMo?i2UR%e#7cE}H|BKIFh`Hp2-P0ETb;)0`946;ovRwjn;ky`$&PUI_
zD^&G)zz<g-Xe0zMb?~;KQmgJh1^7?Rbt}txX83<?W}1%PdWNp!iX1%FE}J7nm{1a+
zhT#C&Zvok#c}UkZtUmy$>npzh%|*wJk^85U`RXg!ON0=QFq|U}pg$ib#`~}_IQ5YI
z1OOr*5<htT`NlqEK}*RDkR_^t(`+~<y6rcb7kC8^fn-6P3K5>_PEa>YY_#Ke`Q~c*
zClJn3U2_$P$Aqu#J(1icFa)JAMo+vA<8GXVNS+U96Jv;Qvp~^eWEiW<L|fAF>i)3*
z|JjfGAO6ik9=Jvppzkuf|6ZU@<S}r1Vhr~VLq_(N5+DB){<R;7abf`C?v_J=lJt=z
zz_}t30qA$I%}EN8cG7@m0DiC=@(<T5ISqY-uDZDcbO;jQ=Y>Rx-+|~2yNM)tCKFO3
z!5f1jO8d_HNz_h8hkB-ZCbN<PqX-iP>g?DPK&7>3mMk>-dB)Z2E;bvi0|QJ^GF`t?
z=WQ2ca2e+RiQ%RJp`*F{8Y#yo63akc-UT!OUe<oa6vxH@{%Zd08t^stMWg`na?XYT
z3Z(1p=ac(&q^*D<>Yg$10EKN=ZEVY*4FCTU#(v;y-~bC;R)qN>?7uEw4n78!@)Lu-
zgC=NiJ{Dni^cphiy15*^5&h{X{xpa{LxOr8?2U=Q#0M<6=Qt&YCaU7VDF>5(EAtWG
zA7HnMfTpj&Pkz(y{HD0_D_wBWoQcHr46xo-^Lh=acimk0-&8G>UKtaBG~X%@aNPcU
z0wnA8scdPT(bJ08Z7ECmpNuPa171$XX`s?N1oAzGplCVL>t7xn^wPk+3D3JeDa8VT
zL;O_&y=X%v`851fO9%Y_7vESEv?>cc5v#IGfK%>YU-~4l6L`aI34cuH96l9cE;Rgc
zfbHjx2xAQ>9|7L~X##%=-4VBoNcLL(7dtQtuN-104RbHxYBXmaV_tb6<3ji+Fc2mN
z-uTAED<J!HvqAB8B&zb$uRA~y)R1^x5_LF}hW~Gm|KD}b6bIi2NCm4m1lF=svYuZ$
z|5zjHNDrl5tWrxit!ha<fiMGm)(s$>k1h=Ws12xP|6Q|z{lJtHC;L!@((XFx-^&)?
zJ)CO%(1)LB{~yxcI;!ffd;68{ZmCU!fOLxVrb9qt(@0B5cXxMpBO%gVf`ot~2uLek
z0!m2ZS=;-5o^yWV9pin^IOk6oZa1*kcdfauIj_%UE(|;{gCKK_6jyY&XgX0U%;05J
zxzcfY?EBF3oP)ZqPt2}>D`_$iYkWq}AnE{cRt&lG0@;aqKQG|<yBUSFcf0adBvH42
zK~`|2p;iWaBAg>X@$^w00vtuvMD>0%<MJ!Gq5dpmd0KU{(kTRwL5u{c<)a)VgkQlP
zwD}Vr>6Wk@iYz6fAr5koLtt)$CmOKVxp*f5KuXEkQZTf7A!Q4A-2B+zentpo-HZVu
z<OlKiCqwY_+VHV4;Bp1KFg=R|Tuh>y{CjxFQE+p4{o>1-a1gsUyVty-CqxTpm|+VM
z2bW=e&25kL+fz1qxC$=q5g^P-5PtDC0~2SCH-d*)Tf=bDmJ_gM!@=TAwIF%NVbbXY
z{Mg@C3&AoLJ=!AM-$we{15Eef9qqSQlYx$ozEzXJ)mptKq)mGN+7~V*UjkT<O&~ym
zy?PT^6z>2`j8+b@*NuRewX@Ln&9G8!>RVaT*I#aLyzCnbA&%2PJMz?f`d*aEy9n3Z
z==jYme>**p8mqkIo$9{MZgLfqoPN@+neo>D(C$?EX<({Ht-5BPO;fu-WDz_J<`I18
zru^#i?l_QE9Pq*nL_U+kv75P;GeDMHsjblv*P~np8{=RttjVXM_y}p^WvSl<U@^Wr
zZ`<5Be;eozkE~hEVA|(!c#Lu#Y#duZNO%uv@(15-^-%j0iZVgg{T*fQSW^vpVxaUz
zMZ_4ijRbXol6tfT?&nR{h8aRp!SFRP2<%2|2c@aX-_GobPWtQbw3TE;=l1v-^y8p|
zaGu8+I4#<I6ncz+bs5Pqp2D*bK;O&M(`l+?MrFZnT6!*WM$oNH3wN_j4SOw$2RqZi
z_TaG-@`i(F0@TU7h+aSj#hJr{<M62u-~6EN+y3Vv<#xpv29g<Kt^hg*%`v!}^0Bz$
zY$7jEibOVi8=RpFYB?8i-EQ8l+BN;$rlNg0*Vaq&7zjmkJGjE%AAI3!etd~1ZyEtf
zN<?5&)^iJ7T3XFeLlgBW6u~#VuV>)KN3<Uh`}eFi6a5uz{Mh>vTyh2dbvxbRw9ar+
zeb)Tzzv38p`Y=J!@Qpa=N8x`fm}_Xz&=NeOa}B5$)?VNxh>k}fT}(Hv_a0o)Bz}PS
z)bfk=OAx-aG8)3s+=L&yo{vGM;|dXR`m~t<{a*3GQFS{MC4h9mxhtlT6!C!t=_va*
zkZL;!4_S5s()mNSDs1e6CqVI5nNnJ%>#o0q^X+&5?So;>R^2y;GeJh$3m)u93O%%I
z=Y}(79m@E}07a(^O8i-#+C7c!o-Qj{rvRRBJeX}=VC$_)`-+|xA6pzCx{ybo`QfEt
zSpESio72+~D9q@_v><jB2W!*?{}be)hAu=)>RD*N8xXE5%NBkCKx7i<Z`Tu$5xBTM
z4`hI)g0-?}KS}!mCPqxRD1P`+A5SIKv%NIqFlvz$0P)^hGdlLkT*H<63AS~C5R~&O
zayZ2+V*GK}grOSgvGb~|`0GGZ<PQm=irMfwRxS&24~7cADJYkE>?s=*-9j8Rw`XGE
zo;RI2Jh7jW<J6oE$G{=K7lib61K)es6}>6}9(kSqo~FRAv9M^Fj#)YZNk$<69^xIx
zkMqeCo-<3i>h}asbrxgY!kLIlt6yLb#PX1e->%_RDnCt33d9q?G_|Xo8HAD<%?u;&
z3z7HjslLMGAk~;Wl?)un>UZVxT^L6!ql_9(V`DTTrP$-6KvQ5j(v~L~UXY_ZeZhoF
zPZE5s@<-Ws49Pu1JggIm0F&?*7<IEGrhoa8570W;{Qj{pw<EJ2GJ}$w<)*0<r<k-1
zvHdvHpLGIq#OBazhkOhs_{<@Ug4aFtYn^ps{cRD4Z%^Fj4*9CA;kTyqhB%34=V8Wu
z%oALfZ+n%6t5d^T4|maak#=P$*msX1Fqu^u$1>WYKQg^Vc*P%^=2ce42kZS+Y2>f{
zH<RV{3r5kdq&)QCm|O>me<h?e$WOp?ZFdMxlZQC*u_)*E?bZ+lKY_lxbq70`eJ*>2
z?}li=I#D4ys1REk;+)wWAe6d3%!gfR6pg}RcL7MQ9^28QuleX8AXH0IXP0z74wg6Q
z>D#;XRN?gmLQ--TT7$?-5D2pSTdDLS%Z@7Ga!v4CQBSEl4Ad|W?}mjL%ukteQM*t!
zpKV)FJyjt^+?{ybu}SzvAgvsGdIezWv^;7y^dBNk<BWc^ZPM)&9CgkTcTKw73Af&J
zT`-&wPMIUi^5v!d2DXs++lf44dbcljh^L6XTInN>hiR%eD}sx<<-Y(?tF@()Awjrg
zdQ^6{%kpdTACO@${AGl`t0-26<@>*yx9L9oA)hfjG;#S4rOb!^0gNe`SR7F7atCff
zJf-2DuU=&+>oZcl2NC@q`&E^H6))zWNi+}v$>B9hxELOP<^hgzPdz*KSl1*w)veH1
zjs1xw#3jhcBph)K52R_b6{HQ#%)+QGHyg<3Z80<S9AB6rZw(`2ahS&L<3H9z9!Vjc
z`>tndJ=MMU;N39*8Vf=nVv>>GXw+(;CLNB-G>W4ZW2P|c{F#=<+pEgOLyrKY$(xDJ
zm4KT=vig3sU@RB5O{(d^HjVcs*ndD^s11=@Es8UavN=ApMwTls9Rps{4n&9mQdeJ!
zN2ts*^AYWrRA=yy{uksHoTDx?2nyuiqqoErs5jA|Iy6T>&ay+&TL?jx6&;TfOd75q
z)3G9mz657y3IQ^I+UCuVCUr>-q`EXTEJDyVOyES9v?>!fO@9f_%JHV@(O11-5+Bfi
zb6U6-Dc|d!sGCSpKG`cv<;cgEsFELgfRT1BwTLl8y&4>cu~+tpe?h(8$!&sXMDHfz
zhuo5sBKB3dhgvSfA!|tn2+wV06wqdgBDQ>#Rnd<~eTa+VB@xVj1P-oNR%DS((X2%_
z;V&jeaGch5=w~I93`DOD!~zfb0vj?AhLWCKex+K7ah2>vhdmWhWm1a1VPDGHYWoM|
zved-8<9k*S4Gjrgi;&Vo<Td1yc5BdoW0Xo^1fKD+e%_<K0ID!N9PQJhh^*jP?6LXB
z1lQq%0J7(rJ{9B1e&(~d)6x>b9?F4w*zQ{h>y)j1O18EolVyhu!Z9v1gvKv7;u=OV
z@dGGmpXHDt9cDSlUWd3VY|;yg<C7yLTi5HyaD>*%e^U$mo1Y#ZWK{W<ts>f$6d7K9
zv^i?LtvaRoR~x7%Fujji{tdc0Gn+QsRcGHO6akfo$(MkgBWW8$Eb>R>ZG_7F_`}*B
ze&`<s%2P@~0-k~`^8sQevl5BHTcoRqj+cE1-Vd^S?FY5{oK>b>|1wGo1|u?wvn1AM
z5j(hMt+$Uid!qZP2ui4PAI<N^2zYSxscTCLnv2JaMSMcLvSW6LG9_Z=HKF;;sWFEi
zh+)qt+tzS4t-O<Xb&MeyB=X4hyX=7F2FWR*AU45gjfUXYg-OFfXB?t}-L_{nXo|<>
z8{Ga@K3@~xZKv&CPnDNqNoS7AJ#)jKAOO!dv2vjsKv%bHrzWgy(mk`U++AUlcbm0G
zjZl!V5#EU7n{YxzW%-Qu6=zoulD<A0^2_)6^+D&}*~LF2Q&Z>)W<UNP{VZmDgyig!
zQ8bxy6|7#3)C<;eO^vaSuU9r5enkb53j9jc6a+&pC24gI7rA0psd&@BwuSmcUam1H
zO%X%!Un=J)uxnsJG`a|Y5u<HoL)IdO6>6S_HO^{ht`Qd6YTdsDn6e3)y>iKUB~f$^
zpu0mqfF7V;YB(v!`CFdtXDKOEzmPt$?1{P^(G`9NXACjE079bXKx^*^9@}DRny&_y
zS_ISH7v!G9Y&U6d1@BBX<wQ*}*<_9<M*_)s^h{hc7%={(<{er@A%c>UQDcoKyQc;Y
zecqXV(>ka)s+%$~v#d-8X(VN@hZ<TLpV2hR;X|ck=qF&JHDjfRLv(85zZ(*vZ1=U)
zwg}Y>sL7Qj#6cQI@^AWfy^1ymDkbi0m7CI@F?o)2%^tEurJvbply4w4pLV$J=;v6>
z7H<S}0v}hIaEyF%QfX~230G!;mM7FEf&KwBS?tLo8`qRt3#HJ@2BDbudJhu~^$m>K
z3<y3g7OYX4O0FD7K$nzd^5VMUQ#QfPdi)wA*_GWDvhGAI!ml)Cbfpqt{<-rFG(my-
zfgPLG>FXbDq=O}+CW30VQcS#plkbagFg4zfEW&CS@Sgkoedculk5wMUf#p^u11xUR
z#b0~j+gQl!51#bZa%{6th9x<3<dZq|&Am15H$qj<1-#cSQRc5nu1SZjlW|X%{<6H`
z(;CPK7D%w;So-Usc)>;17;MKtePM^E@jo)_82)@fbexGO9u@N<L;^IeT9p^m!#T`j
zSD*m8mb9G*eYnUuG@P?yHXrh<P;a;DNvbbYs+8aLzfLXjQ#Z4XZxR#S=OlE#y-bQt
zO1{o_JtNse7GoUOdgG$JHP%hDqAyx`6#GLIq3WEGm_X8Vz$;}eM)XmQ-y2tg_?Wk6
z{^C(IWgOFH*AtCi+e3ErCUP*Pc={`-cMi9ROfGfl=%fav>;{g|vZ);k{f2%Y|0AP~
z?x;J~6GQvT6LjLOcMLJ_X`r9p%&d4SWBEush(<5sFCR7pAAc|n|5_ESaZ2`gC$T|A
z;CSlw->Q}>7(=!QcMj+Gz<*c(2&j!HVZ5XE+6bBi?^#K$veV+>3cKx#;6MO`j&7XR
z&XFCBVXd&hDBkGh2%s=is<VFrZdT8tY(|{b=?(5b)%OtrqN`zFETEtxR3U3(y_Y*&
zeWN0ZX`oQ?9aer;C39Jmo0#U*fy>Vu>cNmfNb@X8a+UF$b}?>8jidOpI>}x?K=aj7
zdZug}amIp0V>Hh2^vMK)-!o!|#t+OqCVCvhP!oi?#<}O++CJi}!4V2Tg%}D;5{lwV
zvZd=zlGqopt-%DNd)j%SteqMIC2q5iF9io)@i<-&z()qIy~ZlY^=#4whQ|c41<!8a
zfm^?3hd+(1$bE?7Xc*G%JxO;CC(&Vj^Nm%APIrvZl<ikFcAV6{V3;m(J-%mGvI*?A
zY=Vw$9BQ{<U3n)U2pMg(IP{xIMynPVSIABPYxS1m)s=FXI&4j##0HiFck<Ne+Hb$c
zjN=*QKl8Q(^D@^tw>OQ(MCu}2#HoCrfF{)%kA3C(-<3*)uqh~rafWG68sa7X;T_O$
z)78G5mTqg#8LBk0zB-AieJ?;}tEi%O$&O$tn>p`AZx_!Hvr<aUgg8N+l_^MoIu*+Q
z>uKVK2|D+Mc`3CI6U;ew*DDr*ny@nogmd~?bPlfz-E~tSe|}Xm_@aqg4$r13?X(P;
z<yKWm(MuGrK^V*S8(qJAe)eC1ScM*Oaf({@IUiYkXqh^x`NT)Ox>cvEIZjhl8limp
z+c0oNbMtlE^&Q-1@W&?$(8oV^qRTzn@E~q0wmjPK(tRQm%3+^9{{BpDfk0QW{r)`)
zfup<}2`kaE)pvaaGT(NWA0(j8@YHDE=#PhtI0iQ-%O&F}8IAsagtv|=kxVFRcn!ce
zN^PUq+k`v7nu-JB3eq~LU_jeMa;$T{D_*IcL+wm29d*;Wa1LaJ>c(`wHDB@jUZVS+
z#Pwq*(D9xJXr&Ul8u$Tfvv!93*0}$2{~i0|Txa#Tn=g(C6Sk4sj_(0=&E0pb%l!sp
zy?=}3D9pKr{Y64KnQlfqdvQV3yw;-n&&R+ZajA=)|EDUmo$z=h9cL>DOJ(^eCTf8K
z=7Hk*y<KGxN_;Mk2M3}Wp(R#7o0Y36pgp^8U$Rl+a{TG^Rk?7O2Tz|H4}@(rcGSQ%
z#e#3uw@(#TtbR-#*-@J48oAGo33t#zyUPhKPTPOy+OC1<@!}43<;NJ2A5Gr=s>~I8
zcK%Unthq{@NY}c3CEXq6U`bI3p-rsW+E8<271#6W5Tx<#2b{Q_Mn}W$6;ZNcvF}!@
zvEtM&a$5eX2j+k7gV|}9N!DMq=RUbu_x-0?+~1wh4|Eu7SAN4<cdW)75?M^VNZJ6e
zJ@dv1>bl<Rm#Gd(oK9ax$!p_-q-49D4qLuPzzN^W*XhupF0EzW-<WV+y8q4zCW-5>
z;JA{aY_iQ4Cm%<^_MJE7?0Z2w$InkfApWgr=toej;;^EH`5Ea-uc~49h;U9wElvbu
z<QnNQAjx4a65Bq;wbT%*1?R`!RKf<Fl<ulNa@sgycg(HEkjVY*hafIrW-mxq<!SyL
zqf6-+5=Fk<+Tq^tGt@;!?)y`MT~!+4-J(!T^}B;KH|~+_Pr<+_!1`rt^9oTNc8r&X
zaZ2m(gXe2$d;Aa-e$QNJ5o5gpZ0m-P<y?-OuK;l5md^~pp0G{^K6`LVtaa_;DWJKp
z1E_E3tn&6u<1{QAb(yNfb<m&rTrUGhbuzLS4;C*TB>ei@OzIw$g~r5kQs##S@7@T9
zozebwTjruu*?<a_{)~HNi>xA569ib6c$0x7t<nPbE35mM$JzxrKm6B8ZQm>%c)WG&
zTRKJ3x_sdO&<E?R(tQx(gWoG)UkFp#sds4p;G?bb<$1w|xFL<B%7{1MC2J>Klwsua
zZ}|f`4AnvWhr)J|Lz+mP80GxskY3ftal!j9#w{3M+;ab3Z*65)P3@uL1m15nD<B%~
z!|TkOIk=|7UI=~wOMU>U$ulcY*TEbWKGQlud?s>VwwYlT>?ccghicIQlD-D6RVYmA
z1(2={*3EvTq2)r9XVS0j+{WU38-Lr?l}Cq)rv>-|2@rQQi9Rlq*}9^6YI@=d`nTov
z5H8N*1np}4SC1~0G}S}IR(e5PsB#39&B+2q1G_gZDvHBpzQlmrTh@H0|NXdvQLd^#
zL`L(ZQg}Rpl_krjQIdNsd|&p`TqAENAwYV?&k$gw|0)n$@gKa}9%v*&%QW!s_qQ3;
zg_IbGQQ!k}l8Qku@XC2(Jc*GxzDwp+=|@$`998EgmLUgc&BfXvBcQ*~)b`%Y!oI!%
zZT@Ob6-jjel>=j;-GlT=Jn16E<e5X3&E8pbtt1Z(;wdyuyl1S%X(h%*S^KS7&q1l=
zN1O-_oGtMY^w^mlV_8#J2o^k2f|FG0!I+G5u|QvDKw{3%5l<i_Xo=G~$+%R|jF!B#
zn~HRHnX84hnzA3%BfPMYFiYz62T1^7Z(0|}(v}*}z6$spHVO@h#P;KhSnHee<PWJR
zB)Lmddl&6@g{IjO`7^uYz2Is@jZNY>Dk-ymN8S<CfX(~lPG>=yZ3Axu6)SfI@mtWQ
zq1>fi_1nGQ{(b+paz9HTssfYI_z{@qY7N4R4iERRFxwmh_0HGUtWV@_0IxzRQqAEG
zyt77tvI=#fqY2??4=cINUS}UG9yL7?|4jd2Y`@!K*i0PxR~7kYwXmd<c_^{0L{zx_
zs#FC7SDp-R5_{LQH;pTcMF8RkYi^E(<-DMfvC}&gsf_QOxKXGRSJ@ZRKRzHaMiQwB
z({V>`ciDUC1FXmoBf)LSxTKbQClHv?VljT}AHesCZ#9J$(L0&x8^p6kaC>PeC2;hz
zv6dzYXUZHxIf@wgA@!Xg!O>qKu3~)1u`@3XZJ#=KGi#EriEpy>$2^sk_h^dSy!TFi
z*M9F9df=!#nDNW?N1Y~jcZFwA%oI01$UXzTBf$Z>#;2!ZW6?f}B7fx}G%BZ{*|+OA
z0F0>^>ce{RNyl@ZF?n5*4A=s0YtUPUyjXs9(+-&Ad#H7pr_%b<yl)1)_F=&zPB5c3
z*ZLVv5;NRyz^b&`JFC_*a-qqZQX{FCZ~(*AX|+<t_F^}cHMRP)Y@Y$ig>YSB>Q5p2
z!t9ObmGOewZ%@wRHiWL1OB2Nx6<m_`I#CwQ=vPPzDJ$Yj)XkgN4<kD`H<<7E{ad?j
z$&*QY_;YoWacn%BWzf&;T^{eR<&wwdCfUWz$DLBW5NV|9<m9$e5d}DLhb9>=qdu{F
zavtt-T8e0%B=N4oey{O|L_YRo84Z~{6h;D;EugVy(BZ*wn-6-0f~X}2rZSittS%4E
zgLv@0(e4Zv$c42QZ6x$=`2MysvBQHz&HrN|0pH;cM#nxFfGU}};>^G8qUGe4HGGix
z2pg1s1;&pp%EjKgAu+f}&M;0J?yQTq6-pfLs}|CH(AY#&=GvNh1=<LQdabX<I;AJo
zUnUP1%$D_>Ef{aTQ`XFqXB&I6`ZhQ5MVKbKEL&;`u`&T!cp6uoU%7oENEgvd(+*o_
z2j-z8gM)1uEWL;ZkKb+6TIVuYo<`%P`#k}5cJgnB-as1yXpV2Pr3@`96b{aa8X<(`
zNno4+3KNKKw@=|6Gqg~mAZ^hjZ3&$@S^Q}9JYO{VL__T5JN}n+C0-q8hOxOCN?r5n
z--?XpGiYHIyHPc?x-YdIni(2%$cAU-PAy`wAC^biySlr)Cy<w2UjOjI;NE8PTj-3#
zHG}Hs4KOje_;r5pbY>}=+)0Do#b74H;mwcphk>0RkhE!O%2aZ9`C1^<WB`GtKwUOL
z(PyFY{+b=e6y0pgnfAO>I+Y~%^~$YA?4UNR@qixL;ErhR?M4TvDifmaLjLI)^*=+8
zy3|=J=i1079WUj0C=Zk3Ss=^KYLL@8EJ5EEd>lokl`S_6D<Gn%+EB+D`rSLqr4g1j
z2E>ck=sX&mlD@i(J*=@WH8N~UqCBacxm%-a@mn5Hwb;_f+>*9*u#a`B%dV1XJc@Zw
zG-Z-h+1M|fGyN82k5xLaqxZ{tLkCw{x=d`U2-4LMwLgzdXu%%7lP3^$o^q8K&sa6F
z9&|@ED99%B^eA~s_CmkEDrZ__;kEkZ>aYXSdAJOYxiiK?Ob*1h*r~HOR4zUz{ut()
z!Afq!#u^JgksP6$Hb!QFrN&HbRU<kebQqE*=qHY9rtERb>e~!`bhVTUlghP&cL<VO
z<+`N*Nb0lVB||1YmPAEsy#JKIYSA*)`&N}(r%kp#ivG`nVb~SAbCefB_LCTVqHGz}
zub;=uU#)nkQIeU>|19&jaeYNB@^vi4=UumnY)_r2MsiZ^oz^QSGiC4T$!DL=1nBc*
z9qM_qa~rELRnNM#of9`i9gxLx&_j2nE?4XQt(5G5HYwYqWJ|F_V8rA))F@)wxA(9$
zzGP$?70e)JYHVsy)+CFNk3dhF1QltkKMv3BfM^Y)<OTxjENpx?#<$T8?HC;u3})t?
z{p68qTcFEsUoM%Uit`nY+i-ugkLGWfLMt4zx!P7o*rA@wpfVXSqahDXwlgq?_&5--
zm(~g`dZ|cgwiE<mGp!eV&^c>qHkb}KjH)B;P<{VynqwkOP(MecZx`N=kwil8@BRJX
zx?dQPD5}36nVfm2(TQX$f$?SK^RIcuhBLV5=ok*q;fM6u4D5U487mo2#jWzM_#0vD
zwYT6Z*IhK4Au#ef-z^=V+t&J5$r5t@iDa-AiQZX|p_UxD2IBQ9UPn_71;Qjn5>AYC
zI7<wP(w@Hfx}VeqR~zXu>y2f(NyBW$SH?<tL;aT}s~#!8YL%QC+3=U(MGt0%>Q@#P
ziA$QAPnEr#iVr7q%Ci_K<nVw!^d;cl+t?~$ZFGXojgXP)6AfQErvYz}O{urBB@gGz
zwkEb-n4#R&Yol$ACyAvV89Rx;;k9>$fvtg8rK6UWD$07wYz%8GpDI-3#7nO})3>p}
zp9X_2kTqnlFs3*`>{%RpKAvWU`;E$eAxDSozN}DVSnc3*0J9D=*O1Ln)g%>7Pz+Wt
z)NP&r^4vu0mfNAQq2<H1@omIEjw5a&0-KPZBdFrTi;V%~-a8;;xXrt7xkN|09i*V8
zIU=Xc_OTwooWZ5LA^c_x-HjnrhHjR7Tk3l%6DLE9CnW=UVOEp0UrI}60uQz~&B;^S
z;H=TSYJ=}xE|a#5kI*HPf~O4;RP-z|#=p0+La^vq#+XW;nxv6X=voqdP{8EqZd#}<
z<c{hF-cvpaC1;iE2HF8y?;v!T9AUZGc8q<{1XA$ZN5mL`_`?>8quXdnzsA<poHfuu
zl$<}qS%dRi6}fVn57ila1s3mEdlp5R0rrCMsEO?j`TS*QfTS^G9kife-A^4p`2uCc
zW~WMw?oSC4>!;}K<Kml=x2#xQKN@4tv1TY>9s6Ek-c;{xuMyo$EE&~|UzU73yu2jm
z`kXKfy=1rYO%iX4796`cn5%g~Gwl3HPH$A#9gud$m|aFK&U|gLp)5!k!~?XlZ;3Za
z2Ck3ciET!W=33^?L@ufZ1H)d@OBQ#iSF9*MEeD^p>zr%tEv!4^YaRh3OTZHM_&jX+
zeYD{Svz%aF{jacqGC=5hX!)j%DR#^Dg?)Nu!B{}9wY4<m$I~DGPQ($yC*q<jM*f@1
z#UJ%RiM1o0{N)-k6TPfZM@f|mfs!i65I2GR)zj){EVN}+pAaRwR_f2>s94E6QPM_3
z^?mnAO=f${&;x<%Vl!1;aO(jU>rgn(bWBS)fKDti@2oKI{?NDHF<y9qL-YsC%8rcM
z1KEK21R*(?HymG+L~k_t22;!D`O{D+)5(XrDJhv1zog!Hq1AcPBE9BnOwiQHONX_c
zy!L4;JV>L%B>5?;vC3&Gg;1k4lWJ^+zhxg&qb5PuW^Bd7&PJq_w}<Xb$d37aKK`dJ
zC}m%ffKqnqnO}p=yBCjat1wwjx9%Cops+oLD`~quq)uTakW|(@^`bX5U4nEn;puNZ
z0|oQSt0GXls@@Dpvsi{ha~y-M8#t!M_pxHtS>J5^W`1EN?|9V-eZq5*Bm~i$rm5#Q
z5DZAtBR$yEpCmPT+kG;YaP~DBO_cuTB3x>`S*<WS4`Cr!RF7mg7M54Gnub#Q0Hm15
z$f1r@tMdF5$I3hNyiEs7Yv1nVmCSQ1DYx@k-9#{s;l8(Sd*A(!KdYbMQUW;#a%=W0
zoJa0Razm!^R)@-tB9^X_PWd@XZF<9K0owKmJmHSsR(6<uI9Fo)m@$%&EF%K4Y_?FD
zYoEIwBsNHeD{Z9x{U(WwPH3#fY1Z6>dVrQRKO0S#Q=E(eo`GopF6Yb>D{h3No+0s^
z2sK$q(@*VES)rM>YA{K0;Wi|1FapVfhPw_knz26uqn1d+BZ)C7*9Z2+oR2Pd6JqfE
zqtMmkT#<S!Me4_`sj`>gw>a+{Q)njQ7&pUk>WnZ3pZuctBlWWa%f`p^QU$Us`q&d~
z^P89=<x9)4jcUl0f?Vn0%-1W8UL)@D9{ki8fvU|K+;6qs4r^Nt9ZTnobOz)<Ym1k@
z3{5jxE>ry$b-vpDar)D1PpDm&YW$GusDOLD{l1ags6A?=pD|S=WQ~q!i`4$+%CWCg
zX_!fnvvIGTQiNujmhzZwGQi^T$vf9urbc@AOgHWD?9XyFC)@ud+`NtrBD&wF-~GSS
z4;iE%4Dmk+qGKfS%x}CQ%Vbu0KhEsV@u;-_JtA?%%IV56D{aGO1TRhFH`~g5QqU0Z
zOJpURBK|SBiL~{ANI9*7!}zqT-8GZiR8RDUbo8tMb4-UzItJrq@$MiKUQFk$c(xpD
zk{<b+*@;+85PkC5X=F(D*$aO<Bju4tmaarsMYs*@iQ$RB+jRa56!!`6o)z10$0uA*
z1;!r3Q5t$rQ9ZJ;XlAej-UdYnlDd!5#;mS~XcAp1jJYVF{HW)8ftsBxT4<=Roz8MB
z<;F0{VX>vevQP0|a6P#F8SqVo4uLj289zLJGj|$wopCF)(E@m1_{E$(HGZJQ>^<Dh
zTWMiMfsK}mZ#u-aO5gPBBOEnBAR8qNc&5SOA<rX4N-isK>reUnjF&NTMzkvXLS>a{
zLsPiha)#FNo}4<zB4?Cg2<4Qu&Eq9y03!JXTcKFrZA5cu*W2}TD(h9IqzYh=0TrU5
zmejVQjzYV7peN``Qm$n{i~Ov)E}~PrrI^FH-wT~yct?o2P<M|lvQEz6q<)r*^;Xci
zf5vNG73{HMw`^D9$fsj&>y~Yq(ctlJJ`ZgS`-Im*KdQizvF+A-xW3sg2kO6kPCPCg
z7?2X6W%3Sj-~saTvOkY^KXe>)E*<Z9&!4bV2+`pML5VE&D<{kD=rYZMDF)c~=(z78
z<MsW+>*!&IY5UEy>%vMB!M;-ayjTNq-$tGFppsGv0?MOk#~?!T&E)uq;7!*3?4g~e
z*cCx`s;HtwY+V09DB3@Bv%&ZH;Pga8z|8CzJ(zd}dCg$LT^U3kNkTnb><9uUGT9-k
zQrlJmkyy+8j_$EY)JTsPvTAq=q&oLnekEK+-N2=^4~Q%4J2Z0#A6UZj9fLbauT!LI
z>6<zpjQiV5nj#srNo<piTJ|_G#h=oz;ZYOkB8+c8ty;cr`5AdTwCtCvDrmbGY0Q<6
zN~z~u8GAeBV6ID#0S1D{_i@7zx4&AtXet?qmZ$@B1LXn9ywkWCd#1TJ%g01D(Y-Qb
zjrYK>GA_vC&QN+DXCaFRaa||vGT4P*(eo=$WSQ$Qn~law$RXbJO(k9XbG;oS!_xbt
zedE$ZcaE^`jZQgZaHvAg&xo#>h_r_U6bct)DVeC<_qDafE3QR11kPJIKDU|nv|dK`
zau6Km+LYqoe_kU)a{anqp}k!`(?!4bw(4&|nU(0A_vMw8n0#AQ0^gs(auIUPgwU{n
zrCqo);#bo`i9rzu9ZE_DfLfi9^X|dN=nlBZslVgFh{n^B9@YqopB#b>pv!Fc;%|fD
zb!dO87c9U2-wz7z0?6V<E)&C~uR<X}4w*PXBM=TDj+X-9r2`JB_X$N9`bjPfQT#K>
zK8F@qn!>2bB_XwPD9KbI9*wA+jc9BTPGeQDYavLz?@PGI?F5l(WfpDCcG4rL;gA<H
zNt!}YDi(?x`UCRFa)Sz3ds0U}A+;FIDO@QkNi_)&6zME5`V~Ta8WEo6CuIT1B?)FF
zD8cn$`Uw?A_p)bc29K>X6y2DVdaE-0t|**bS^XX-88<X7`l@{HnE6C1pzn9f+i)zd
z?$0N0K6v0$@OJM)xp)oB#L!(Fh(X9+Wf6uCK3(1U-I<K+5adPKEL3$Ol+-08Er68v
z@}<TIH=`PiuJM)a6iW}y{&whD>;%4S7CQaIi1G~mvFXPlQuxIjQO^k#7%rfj*!q%o
z{#}<cO79`;<Wi(qI7o}xq-7k(n}yAeaEqBm^LkCoY4K1q1f?at4V_vltO%lvBo%W9
zvCQ2HZe`2FiC}XuBz+W;_9J680U5_T*lm<oE9u48G5qdb_WsmfL@^qlUAhAq*<rx`
zYIZA|jGiHo6>k;HsnecyzUG4Ivva1J-C*>7xG2+W^iiOQK_`;kZv3s9Uqmw=V*1(f
z#h_knw3N<t2s#hjhc_Gb?$WW^I=amd7Gh#)8^7D!@muCbR=28MeG#wf`VI|*Qtyav
z6%XmxurHG-@_kM$hwz%VgBauxj#kveBpmvraL|Pt-S_g5+Z=^p*-5=uW#4HO%qgJ$
zEzIW7oOKUqH%~r~XtcfR*diT#d#Mtz`h2AVm-~KnC2O{iDbsLzpK(L@4%=MpJk9w4
zKFa3u;Y;$Lxd97&+q&7+UF*HCr#?v%5{i8kO;DGn`w$<cp1Ba&&mldt4$1DSg2^R|
zi<N0TZ~O$9@c1?0rTle2d>(0$$DiPw?!4u!jowbrggLx3-h9;h4C6{zDNx2JX-LWL
zc$6n@=silEz5)h)%Hn)6GyxYD>Z1X=_cglu{5bj(jw?BQ)@t|!rS3-(f1e5`WYEB`
zs~ZCTagTGEP6m2?{EjEwk26p<W!)acwT+I%vBwdNQt3+yh!k|VNVp&!)3Gz~(T$9b
zb$wJG>&-R!A{}UoO@y#A-7h@p6@EL-S+rz|OE3K#Ke%S8)I;3@k@6<#Nll2Lz(V(P
zoi8;sd+oggnC$b*l`4IjepwwQqXB;he(CQ?UbW>jgr;&0e>0w}AZvj*ZdHA>-J5O}
zEs2z_lj#eZ&?l8e<@XO8^+8y94wLkd-U+Xb@>6(PEFPOh*!*sMU+8qf(Rmpdx+`)Y
z_>Sq+y+IOvxR|0qY+2}FWRA-*9;8z^<>`jCm`Bv?JiZyQ?aRh}Oa|kL5;nW|a^AQB
zrkQ9inO&W>TT_WYkMN9?<?EmT5#GB{yo?+`z^cSru0*2FV7GDkcBKxVt+Z=cu8<Z@
zt$h#{MX#qErI1N$JkP@Ec{9WpE&2?ugW0-A=Yx#V^NnGWlMhInBYBO??s2cRlRt6x
z-um?Qq3!X3y|1iBt!H1zq%;8c>1)9R6av>L;zN3mPuvbf{o}f^^(5kH;;yWS9MS?z
zD8CN8X<J>qCYAX7)YjajC{Y=K8sYAd?fA&OM%>Wgm;l?G4KZGEO}Sd{O`qU2=3!V}
zQ$R>UapC7zl+eflQ4HtU<^_S5b9@$Y#cof7JE7u9*-6$dQ&Vo`yE~st>6<jI?<IV9
z5SUdp{O_a*keMjB9|V>_l(^|36n-RWRI(pL*wI5a%Mj_&OXyiC7|oFp(sdbMBV&{l
zDG(2#qfRLKUwBH`)$Sb@Wal5PWG_6u@b>7clyRS}@;jL6IXketL!&LgbZ=$|>SRL>
z3&MXL9E2Vy(Ep`IS`r-{fewPqOc>lO{AoZq)rDD-%j+GMAzS!*UNkADI`hY*)|i1C
z{PBtNgj6-ij3md4U*snkF+6GRNvJHE*Q#ikm{EQmMcqODA^4ADzvFkO6cvSxYCVsW
z#+K1YKqOVsdmvTSPyYaG6F+QXAfKoTqAkF=9rC5`8>iutvvF1Z<_hyP0s;CEu4QmE
zzhxwEuqOIe2khwdgCEHnUO5YvgczD40xyCR8Zj@&^M-b$DPk0{wUTDGM^gu7q2VM;
z?{;J5(j3xISW(ekk2-dKyJgi->2Atw9yAvX`6kgr*oyoh4jeQ^mqmv4+|W*lQMzKj
z8$Jyd6{}TU$99gXV~7#kz(sT$YzB*iuwF7!^I-$^ejS8O>*pns|9z;~z{SUsyi|Dz
zgh1#7r-VlspO%y%)*&9Y<s!{F?@oly#!*|e%%;?LD2q9=-0melZS}0xmi>dw30?@R
zJJptZ49g+<p`yk$C(%qO7?}Kup*+Mr9*{emS`Q+?GFD#Bx-LqN)}$dUvM;_QE(1Vn
zNiwoLNE!**(~r^?^(#|vj_c7ULdd-swB4J}rD<6PpUa=_e$jn~3kfJ`E)(-esTdQq
zOb{g7D(@J^JsUEWeH!soE-<_?^?Jy=QZV*+u8WGva0nCDyH=u3E#8PD8cJM9UI>bX
z^FGxfjjRCaaTVndbgg0-Np;IMUobTwv-9N~RZmZmA2(_6nd;(|Otzzr>(k@veO7mJ
z7hwyHfwHGXwAO1Wt#~d-Sg8QQgsAR6@3oZEC=K<aD@(0R+bK;DeW@az(Nk&n&+@|e
zQ~M+q=N*{N-CN$r-}ciCgwntDnGovv+{SD59WR6;7YXu$&v^UJk#gB<l4JBbi(ndi
zlCSj$3c2N>vA!SA8mS$*F(lw6bzONxS8!$qPpW4s<_XN?p>`#1N-HpNd~zK5bL5h$
z2O%uOINYmVqcSxvvO8tKHs$9?M7Uv=u3EF?Eq`ocU7%T>E_0k2IT6_hnmyQXY?zd>
z^y-V+xZVg==eS~sq2}rW`YhI3;>Um_jR}(4SG}R#$a{%DtC3F-DQn1G-D}X&<eM02
z$Xzw%QUQ=%l-PAP?{Oy0!gn9gkhr?2Mz@VaqS#<6zID%$Dcjkn$qo%^f!S1JV`}p5
zA-^r**Pc88375XSt5&r+uZSNH3;g+*7cbbUG;gohwn8mS%~=u2I}`oWT4%`bl<TsT
zWuLJpCAxZmTs?O|mYRvieN3B&I<&gf-pRsoB(Li=SBY9lq(X>slz677l*I6<>dX`~
z&<-w!+=m4)-b&4C3<vHCOdU2{m8gsw3>zdNE(ov2p6SciME$V*CizX-#IUuXpV{8+
z3O|O2zW;*mW-72DK$0e#-I^9F_NQ({T^$B|cMxTO447E^x#|A1S%mfX69oNuw;7LO
zMcl-3QfgyC(4ISxJj!Qg-3Xw-;jCfQQ5rw4A3b`&%-gV`ik8|_(!voXEwUEX{3{cy
zEUVG2*vU*+Sh#*Po)^oKC-D2!f!fWjK~F*EP}2{i^=VET)r!#in3%O@-?#Q51?zDx
z)F_0QKA+y<L7&Dx3q$)9rs}<=pFN|9wuM;xOh=>j9T^J37~&Wj=0bN~n!kimsaidG
z`oQ<6CqX}>ePJ-AW0%_01Y%sTWYRGv+wWbsQtdL{d1)@D%2$uKVVIRi$sQ46C?;tz
zzd}FD;^Jz<lt0~HV}mS(LVkHIMc;P4N8->j$ObQ!M1O*DXgxci)@O;hY}4kG{K7!T
zFVY3-r{x<?827ti=<&2Rb6GBLnsS4xnHj5YfqQkSxA4YO+6vK}^-lu@9f+=kew8RY
zXMSHSa~lzUZxVPz4P@?r%2OQ9&6Rm+Jchi{Vumne64-oc8uST!Hc>);h|~V8RT1q_
z2G(Q$QtY!^!S63tDJ*BYNnEd4^0ZK2e{F7Q?z+k<&J<Ph4c_!elEt}pT}-;|l*qa}
zsC@Im_*MS2Bk$uPzb}EDFR)Gs!G{U9|6LpeK?YL-@|0(@5y*mupriOw4SgpI%i(Dl
zTU5{lQ_O%LBKvc_eXC%Bwx3#l86uL6pNC2~a@(f$a;(2FvkCg81KN`y%rrWq7hxTN
z)F`X40Udj)LS;7*62~VVB5N#2qi`%`_+E*U^z9Q)`wN*%K47Fi8rUbuWzS9dm83f@
zluLg*B_u;=N)^Vv-F|yA-jzbEC3gMvxAbKx2k&ERB*O(O*3nmb{r9af7mFs+jc+B~
zbyjtYSZjD8u^mlkABDp{qQ)IM46^QMTjHLuT$X#repByb8%(;IPaCaxBi1bo-t!B!
zV)aa)`y6a!<oQf#<Jrkm4w~(<=lck&T+I4Dbhix+Sv<UqC4Zkw{HE7<_C=^%!(;iw
zABHY>`f}|((X{1H&zx{c^U(vFcV@H}O~<5J_R>|JlK>aP1oJ-Yuh&|2F?;0CsfTV*
z4-v`?yG`9vcvoH#xi@1BfZP-36=xiB+{T3I%n-KRS-QJ)jBAqNh&Tj_zO(7NbA5Rj
zrRg0y7fLgpcE~4GbvgSS);gA=iy(g0KS9QIN?a2W&|2vo(eYbf)wwA#!+b3Gvsi(Q
zI2Wx9K16)P{~X?7K@6rbbXiDK%)8RE@JE!J37<RMj34Xe)N;9;fnNckttNe#BnhG%
zdc4&Kk(_~^eF7u0zb_A_lkQ($f-+n~4K_LANzBO@4xmf*gz*-<+Y}7sP}jHQ4m@5H
zJ#OzvyGvnIr}uP5>@@NCq=`?C+HKw++UV$)IPkKH6UHNx$Lk4i`*fPl&6mJ1l_W}&
z8%mO^OJIfP2W`tP$_K{3-UDtC!F>rVmTUFpT`Tm$25UGWyd%@zi0eti&<`7|8Oydu
zZ?k7zN2WO}<d^cET~P^5Jo0G{KTMb`%e8VJ*_oHySI_LE`S?(c|Ai)_$$k`UC%X0W
z_?vaXE1jU1kqgRxanE<Z@sFpMC#F7bk|P1VOk3Cn6GNRW1FPH;MB^c<*F&k<KJ&1?
zY3p{IxzsxfSA->>VGgI83$|mc^iSh253D?-pK>cNPj#Eix@u*w*h&c0WEgvb8NasQ
zl)uNtFfT8zbZ8>iL#?9u7Ui;R=CEVi76Wp%?eP+Cn{gPX(0d#%SoR~m$)dOCgcd9E
zr`59ryhk30)s#j17-tj?9aDhU4QH=ga^1`zoU9LEaZ$NS^hr_(oTzZrw5hgq`MrAP
z&ackClg7euM$FX~kSmWKIE)_n#!j0>Zui%H$$miF!G<kg?(5R`m1-T2Dy*Nmr+F87
z??hPz^E?!(CI?`A?aptV?lVz;lF!J7x`~Hg+$d0&E-zGsw~plQ$71vywMRtZG$!Oc
zxwIj@S6V8j;%^$IivLOWk=HSx<D~nm=-8$6-TBwVR%eFA?X<C<6N8fcC0#q#&QF7X
z5pLi&@fux;Z^(>sI=H6u!{lpvix1$7Fjxcnpoa0^&E-Gq@;mx-Hh<(q_&POxPGE|F
zzg5|hLOsjKx+-17j>sWb!YUNYWW4%u5&G5h?BC^#O@lxx_i-#c^%vwet2{%eY=YjN
zzvj&p$J#JRT^W3HaHU93lzraq1^*crmQ6|ZN?{@iRw$gHl&ZU`ao7RLk{MjPbq&K0
zr)phAIxs@C?%RObbC`n9s<~$bFUm`7q{K<>a)V~CV>3lLZxopT;RRT*R5GIGQuDRB
z&?cJQU*YXZVNhRJ5-K`kX+M6UlPM0eilsLUW@VSsVT=A3D+YaE+?&Mbe0seQ8zwmT
zM1CdRr$GaH^Rc{N%%Ly{XB%|hW)u>k?kRuUK)*uQ423z^b*-{Jt}O7WCd!Dy%hh%9
z{?RZ}0&|J6;;CZ@j`M`roH*R;)e9ZCdOsub2mr|_Ps|8X$YXE4A5Mhym=6r}!h+(g
z!n^rtthUB^g0Np?EE`Np3iDl*Ox4mro4gKfp8(JDNE-V|f{O(y;$1fV!Vv+(UG`9)
z@Hf9x5rUL%&XkuJ8`H{M=e$x2k!Q#V^cc20nSIVx<ry(>pWBXeXXe;I$4<e)H=zR+
zf@!QRGG6<Xut=Cff|cs$afMD@v1wyTOcig;HJ(KEUsue5T7?i(M_?fo-=666J^d)u
zw$J5l!sLD+;?BIq;{F|eojYYKKp<;M60T7B$$cK_5rbnkqq?4k<v@Bk1F|c=%c?7k
zLv5>9n3knxr=Ck_8bGsou?!p>o_CHAgfLEm2Gh*pt;naUC48UQuc!B%W;y$c_dme4
zE9%ZaR2=P+KJd_jnhDp6XZC>29Ptd|y_CC-PLXs=Cow(>yZb#9mOhN&o%_+#_6CBq
z?Vi!ifNM21IYPh|pn5hGrfqGQx<25H@1Ev73yh{)-nK_OJ&&`?yT$Ya3ik0F>b>KC
z;!qBWy)lpn2G>uc%$QAeR3F{_DKQE=W$vfQzKkRnT5WZ_S#EF+Z^(!3pYsz~hxPse
zliwaC4KVNDAGyv(no)*&+dqiz(|(nJpSu0TGLnz|-=L-Zy>I=qf18RjIDu|#i(%dO
z3JjuKFyEBXwZ*sP$E_^&g@UHs2s~K?(hu7Hj@pcq2x<snaUVS#C$$;`CJ}r!Fe3Wu
zn;a`FrW&6-H0W?0uiZ;UdO^rCpdL!>c>vPWo^*D7cQcWPzk3iqxiPYnxi6?vJH|bm
zoYqgd7+fMjp!?_w_EEzK&AZwol;s&x6wk_u6({eQr4HbO;-4R=>GEho(e!MHjnz2X
zK=lNWF;>*IzWDmhx<F@7TiL8>>xoV!?h+Mldc!4u20|Y(XXQA9i;^W%2CB{)LxW`;
zxd*O7@7gt6pg&rjCZzSau>#*k#lEYp?!Xc@T&0v68F7)Da3*`#&cVNKveYHVi)qzS
zv(-r|^x-?y&*h!!qsJTQeJXK&BymN%jwuemmK)xaP(MD2!{7Gj{Pm7{6_^zdHa>|_
zEi<m`OwA?fXU4mc+G{MWtPofcuD=Kin-=p1e)i=<R$FGAG^_#^jwYL}qjcxk$O8pK
zD%}r_<ButUIxXJkP}cs}Au5KeeLUNfP_vTFXB`0#NHAIs5!&m&d%Tq_oZZrHsuLUn
z3*xw=vK``NIuG*vsAOr{EhGM{^<$$Yw{jFh+x#wv)01|8ow_*2hq!>S7E~j(tU|dq
zX@psk`1E?Wlm?Aq$S0d$I))ZxjFkxL&dosUGy+GF_Q}rU4$@Ezt+cG${{WX{!<O1q
z?fq5vU@g?=uD|7NGTO=^mo42VIX@?|!3kX_yN_bwoq9|kw0SS0(ahgHx~(4Ay%4=}
z`XsXBbyu9<-y$ST-Y~Lb5GPaBCx>3^rv9nz=FLu;uc5)ZTIMY>%eth0g&#x?_1s6s
z@Plp>|C|o<E!~a$O%S`=80*!8h%}a7GmC7=p3=CU4eDd3w&`Ys4*uCA!tJ7cn`+vs
zH;wT!tJ^y1EZgMmGZRMFnv)-18G39Z_BN|Sjp$FE8Q$(BcG_)qI#~6&r$zC@>d;>$
zxu+ddR!y&zeG&RBs3E`i_H!sC&4J;`%@jvw1lHqr(}}!>LeUtS1wmFX15;$KSCd(9
z>ws7H-sn1De+G=6?)WphXEe$OzP|!J@#DvndkkP6j>l|<;_J_V6sZ}FVeMvDxz^Lx
zg4P6{k3(c8%iFvKy&pKBXbh%Z#)aJ#fyKvsCzduP350@oWbT>2Gw%~RReMTl%hjz7
zBiMbLkPppM$a;T<zZ*S#b061$>lo~v*_-1Lai4mhEO9!#=Ku6O^4BBt?in&6BYCaf
zeOCBGm8FCp^;0O_Ma!1pZ#jhkf7T>v(X&X3oIu<0CYRMPWjg%BHO~muMog!8$SnKc
z<5E$+-y`}agcs|R;)%yc5A3dzX^#8yB~i@aWZ4JhT`;h|R+go%LvXQ#tk|j@iaXDf
z)c)NuSP6p8kdi}0C@e5ZH|eq(+OgK&8&0623tA48TmsYjwyc;d9|feh!3*<hvep@8
zSR6Q_Te<ohFIC$u5qV<6*Bl3490I+5={LE5N=Ndk9mjgkR#EGXmA4hPq7#S)JB6;?
zN)6lNZz!OJ5h%Rkc--`Je7&*b6z9wpUoi&JDoL)tG4-=aj&=u6*t=Wm`q1tN4at|1
zVxei00ZHPdU+5kLvBWNRaHEEEk7hOJ;*&+?S-o?fC^a>k^AG8G?~ce6_OR&(DMV6}
z6jCXb7bI&|mo(BniL(C5HHj;Vow2;y<@rO&kFlHK9ah+z63B!K%t_*kNS1^pFG+s!
z`sZAm&dnfhGL9jVC!uYxfBwe8HGyM`b=++zB^!#3ih#OcmFuuLnr^C;U5v|ATzd${
z7~dJ<R4NCkSgr%-mb(iQGV&O*<rK$_NR=gMXxTDor5KO$mr)ciODQnWzOGDSjLL<{
zoHl2ns7`VPwy)D7%TI?%a)-I;Yge*|ymrs_K}8tis6-nRCD%f;Arl1FWmBD@yqSeP
zhPye}MV2PNg*mlkb$ov<>JWUN-wW2h`Kr#N<FcJ{cZ`9oxy?q+0%@UO+E6HqNreR>
z)vqEuOFs;%ztGy!q3h)OJ&&*%s6o`T*2$+AhAg@szKJ(!iJ<M$b~shBI1#YVN?_Y)
zNU&OT!0KE?Bl+8$55a-kpL_#x)U8=4?Uz8DpJGpW9GT+xyBQ7i;^xYve*A&3ARHbF
z9;{)t=kk*(`KtabjJ(&Fi<*K$S?rF11%YhpH+!+(eV^JRE|S4v>}PoX$DFWxHM?A!
zrJ&NzN@QL!sH#x$E8V~sb@1q7r1oN#^um(Pcu{zA8sE<k)C&1}mzadjBqSxCc{f*~
z`_m%ZoslIq*oNB5vxYWf-8NK(@y2fYJAZMXkthoM`vkB>00B~qLnfW-pVR4Y(HKXD
zkjDC>nFCWvHEr+5i_E_#r5p+ZDOZdRs^+roq|MiuSi!yC1W-`o7sYAKrPjY57{E|+
zKVQNip$>=Fkfq@dTzN@9W*3%wP%E>3SXb44xl6Q*1SvMPj79Mz{k-t~0Y%)<-6Kss
z%+yyG(wd>BsiaO{x;T(!hNU45QoIwO_^UJ2o3LE#Q;Cw;gQ$$Yd{2EsuSXn>vFE>z
zTU>1L(2I5e*46+CL%W8pO9uQDBR`cltUd`l&iwdBOYD@vnT>&Kd$K-A&5OgUDd>|Y
zT~9FWh*}6VU$z$z1Nri!ruVL*JJBLGCANtCqJ%7YlbbvJGtW}VFpz#$Vpq7J%PGce
z#kizJ;ft9to)EB%@+kI`6Ut;iW4_dRL=@MifK`UF{+zgg<%o)4fXQ!6$WkC(_k>t%
zY+!Vdky>xpZ-(QM8P53fCor0Rb)%YZeu!cCE}=Ej2fH<Y-+H0#rfSDKfY1s?O&>`3
zA=Ns@GQ>`?{$3$##(9j)LA83Ptg5<a?T&eIMe6GmeE~+M+Q{>SFMQu0B=n%W_s)|E
zELbJ9B~%ELIq$X^;wV$A?v*^2Y%Zz$m3cv@*Zf1!n(7sGxoWAug0x9)uM};+UQraM
z1EXb;{SQguO<M_np5uyokz+>7jhwSvaF~eKDLJ#nvdn7)Xi20NSW=-@D6EL)3~VYG
ze+EJa?{L}E?}_bGnJ#!{6YnpXY3-8#=<PCz=T-h9-U_?W64|<YVY2+3n#n%;?=|BQ
zJ!Ih9^9k*8#eapW|9MOPA6I7ltPFH}!1OsXCL9;h+hQZ+sbpeh$g`<cTjwR$oC+s1
z-aSf?7Z!=>q$?5?ZHWoq-<=s#*DMtkJwJKQMhM^TRiMbSm$@+=)%<@|vVV5ye+Bjb
z^@}fr2+MctbH4!d#>unh3t<>r+U_`4pO`)H#~$Hc0dK?pO9CfMCjT*JsJq)s%Qyf3
z-!K>s=2$s!=)3!EwDLzyIY^FU3h;WpgaIvTFph!`^nz-+Uo_~tjpIL~=9IB?ds~%|
z;JB(iWa!?EN%0?vz)V^a<U)?&3_<^2{8jN01@OJLv#po!4#0BPVj%7j<v<q$-r%*W
z$^uN_!+(_p<d{e%{TO*1<md8sK|*0~s}pd|DrVwR_H9%joOk~7DXCy0(%-6LPFzR*
z&rj_?x68jZ?f>bEU>|Ti>*7iDaKaDk|I@$zdwBn^|IAbqAAjwYrs<sg|M&kx5`9Jn
z{1JUF_Z<Ga-<jsn{e|R;fw^n5;&4(-Pz8PGSHAnnq<h8V=Uqbn&!XxF+#+D>E@1Hj
z$P_BQ0Rb9v^0Vvu|Lt#sBnjROA*fCr^S{4SX>9+3bs@DH0MOR&6!=B}7K)`C5pOil
zpQ*$EVo*2c&5^xx1WVEGuFjP=`-&%d?{(-q$m&x#90LS1t`!=;h-MmqSIbd0ySea9
z9D$A-8NIfC0CP>^&y{*R_-wm|x2+{^U~pHiblg<o3Ygp(0JqMS^BzY$m2Iyg14B2H
z|NR=`o`M%tQ|Rn||M@#Th5Q6wOHo5$<$5!uZtVTZwSWkaVN&aV4q&x~|19V7XA?X7
zEaUK%)W~hB{?{;NByfk9A9e=tC-wsf3mLA+W|9-QZ%2Un$=WMCm1j}FLiK+U_SRuh
zw(s7r5;90ji*$ENw{(Y;bc52}Idn@)w+hnIEhP++DvdM)f}}LQ*Z4fYz1BMRUi<xX
z4xJfh=DyGCtncTm;{$jIgHQy#MgrmHWLYiXy9%fpu7Vdo!aw~2ysMr|02?FMePLfF
zC(ix%>##$K7e~Ha`V~ZsST}P4ReBT<Y4O0Gem4r`xBl?)+r!GLdZog#IRH*xke1ke
zTha%pAGZMigabv`5*OI}OG=azBH?;}5&0!rv(CH<nrN1YJ%lNBiDi)CQX3#C3;5Vn
zerS3KM%?~wrIyHlLC>ortThf~dK<l8KpH9DRd_e@!Z8sxBU-Q{taKZ3cn<#DcGi%1
za~#1RA%`>C>JgN8hf%A59)!+o1KYVWX+~pk7oA||Xyg3<zYDVL5aG-YO4#-9#T80{
z8^YHBbdV@SJAaOnf}scu*hw>jFNCm>{}|zdK%|g4wNT#x;0zY7CzUc=1B^$F1VQ0k
z_+5qg3Ba-^eB-G?@b$0m;1kNMLR{@u#oqOd3*T{n6CaQp<%Kl=tRF2mOrT`0)ISZ?
zXj<}Gh(JJP%)jdBIc|ugW>ey4C(8Bg(BeiDgYj}RJ~y$xS?J-2Uf0yK&?i?l!MChz
zA7`}w&aqbjA;<9C=LfJxz*3jEZS2OdKQ&yr(xo*0&Z>=ie?A4690(zp@~Z$Oq=P`{
z!F4HtA=_aSUB1VZq8i60&4B=dc<!Bqy&4FFF3$eC61&uZlz&6@6L6we$Aj<O&9r-g
z9Ep4jyq0d^xadhdexu3F8gZ0&_w|xFKn0zM7oCNul@lQ08_N@$1{_Id7gu?vf-B+I
zDneZCbJ&aa38bh*U5E`5@2-=)C6hhBV4fjlW7-n5|4WF$j|XWJK4xv$Ki%8=)Dsv;
zY_HciPQk9I7f(NN_r<_pCh-S{&NadsJB%<{t>_@r%MrakbjDm<fHGriu0;TJrHJn4
zliy@fA~x#tI+})Q;u@o~@lVQZo~dp|YEZ98b$kL2jb~IDK^LDWoqmVY7n#ptj0UMb
zx_gj9iJy}AsD<I?Q7Ic_04PS?1p$fX6Tn~{0QOg;<W@V7{&xmQN|)bs?CR7!#;@@1
zSSMG*Bv^IE<eT0_`t}JCLl`sn!Dfbu$xeGU*m$a4bQ}Fx8)EhbA_cCdBEuAD+V89H
z-6lkLFPNUXAA_&$b4-W6#sbKKmk7nyp>()@)IpZnDOa6fVJ8wN;0H3iDN!3UG^V;_
z<@UgPscpq)_(p#Wu$4>~NLHP3*F%|gRnq=@s`=lVF#jIda*+FYcXCV(1Ummjo~*0C
zoGy=M8wAg97BDqL7)a{-)q{>xU4SdIh+@V1CooLsK}nRjyKn)Vik40ODmQa`jC$;h
zc?PZWhni0BOOY|maJg2_yUts4c8jmSGUogA#YiKRRGl9T0K9OB00-}GP?Be#v*Rei
z8B;$eViD1;6n1TdU9l0-yAlSDFjuJf!x$cgsO^6RSGT}8<6`9$n&e#sj3av<f~idI
z_~uaHqr2)me5Aj5wG?%aQMT$M;Hs&i59Cy6gjj)phjv_!xMtXKRUmLpGmI(ioJvYs
z`}=J)X#L#=$Z|m3_zwMNV*d%c<hU1r2lQz9v|JzpQ0B^30lL4caW?8HMZRUqT{LYX
z2|i7Rzso03#dr&AZhvMh@X+h!1bDHf`6<A}!!tQ8CQk6J@wq>00x)NuX40{n1dFw+
z$M|bFGFVjt2>93m3s+~{p8BFNgMrU##3#SK51Bzti3l&}J{>({!dl^Q7(l-q%(F9+
ze^o>y8tOq0eG+p$58!uG>%RexRy1S)Gy*7>0VvOlUS-WG(c8@zaMgbLCYzE)oPY?v
z;-revJgFZe3U27dGX{HT$5+)WW_)H?u8%Y+fDCpK)QZc_6&slcz68{J4}k)BUcDv$
zB{wFVNqND_1Tc!kx1n8tO~#C1hLsiE9cOkw0jDN$v!qh{^E7}>OK=XLHQS82!s&el
zWW#5|s$zR|p1VS;7Tk$Vr8yeg_BHRJzt0h9U_>)67e(2)SBd<n0;_lTrNBG<2eYfc
zh0z^F_c4*UBRhK1d7KPZ&Z@niZV`zD*1V&&qB*7*HP-(y@Q^)FO|uQ}ZilRA19Z7h
zWOaw_I~w{QDMJCxzk4r?wK@V3Ttrg#LHiL~UF_^R*xxYukzwFU*#<<gNpQ<Q01F94
z?15g>6M4eBIl3+^eko^|_`z!dsJT7o)s@gZ^>KKU11LCuJ}$7hV%+myKF~EVE$KD_
z{dCd&mUH$4oWqa*VF4&D3C}yvISETVNKd5655HTIAukJ>jH=*%Vqd(3*w($7Ren4Q
z33jk*U5zB0Hwhj!TwYrbZ;JbKJK673$Yx~x&&7J5IvLlDcMCjWYh7$h_BA}B2lR~(
z$gC8eLe;A*l%bLWC==Voh6-+=M;d|X&A_)c$dB%YZ6^g7siPo3<dLA*|4`Agbw<}&
z3yh^(mT7J&#t|ZY2w<iK8&!zqpP&BxBVP!=A$e*PXZ=Qq<to<%6-Q-fB(?CD0m!_k
z#$U*%@4q;NTmN(EzeQwD56wIb{!LSQ3UMXzPV15aVKLY`GT8{dT|w*lvj3;GI8Jef
z_XQXIjD4U~Ai293k`PXLSd4Ae2|W8IVuAtQ%HiUtt;9k~VUM&(MmL~MHQkFrT|=gm
z8V+wEkefr-+zZ3ta9NXy?FVVfHU|>5EzoA@dtGj3iWG5*eZnjL4g%!?T|9!JNx>g`
zn(OnNk=zUgJ=!Ax97veojocGhUz*<T>b29iBM!d%k05ls(dq~S*U8Suypov+g7giL
zfpiYt0&15=8y}B6n;=M2UgXRF0;exh|LFIQt*1@~#~7v;(nB7nh74MTF+P38=o|DS
zv@+ekTiJ|~6osO7wdz40T9MzzlFSw5`>ZvF5H$t53ho=K7a8bIQq<Vb2*T0~2_V7A
zi~<1Ir}yaDTN-suBJCmSLkfTaHJ?@$VDFm0GOozSA4as=t7A0fz=gK>!DSj<hI5gi
z`@pm2#4WRWTYh8>aO8ej$<7N6hmM0h4{0C8kwSMI8yIrbh+O(#`Yv&~R3?acL-}pM
zG3!oT$+TiW{s*LR{_Pl_9Y6}}!J3DQ7o|#tzj~J)0N%^9lTDI|^$~*z-kpm;AZUF@
z?A3IOPh?K4OfdX-`R3+_fYNs$Qg^_##}VBrVRCPDPBXls=KAz<6M<3ODhR#{@tg+j
zqjoIusz+<Dvz*P22;As@9f~ViURgn0aY?C@G<Ao)*{r96juXsNXAZpF$`Uspa{)_l
z?amd>@**el{bmP&Jj!q-dAxC{KTgGo(rU}#wUImti}HCg`~ls~uZajKa6c8ZSw4d%
zJdVe)lV^Hnt>rDMK~(<F?gaz#{lO6sX7iISr_Q|=UF9i7DpsB#b{|&)GEFT5I=|`~
zMu5~_tXp(y+o8ow^fiWz+n`^(CN4i&J){5{6gK8Rz-`mEGJ|(JNFDX$5ki$gS-?+N
z-3L#+ca<wo;ahqtHMlerUVmu{#|Si$5fE_SQlUKIWn^+v+2Pgi<R8E*1VM8xMx0Xi
z*Km18Jyx2hIiiF#ww+{uvQ_Ee!-~IU<iH?r8()}X`9&p1bm)nDS|`uei}#H12f}n)
z4|lAu@va0wMCf~h%DneUtHT!b-45$BJ7EH{(!W`zE6Z$ig!%I*RR><x_bFOGHm34Z
zJhmN-Clw3M_hs=@Id#jmgoFyuK;vJ`Inj(geD=@X$v^>)Yz7L;K&k)Stu44cEC^oY
zwn&s~=L-U&=hWngj1<9{W%^i()EYK$f7k(6k7`xYCzS2J(1Xg5WO5BjozEb{fFoFY
z!qkv3<bE^=Zn9eKTTf6Uv^`~s_Fx-i2qoxk$3?5{2t%S!)=9Q?3srz(NDZ6bm^)&7
zW3~2msR{Ss?Wrv|wSQpMg%Gc;0F#W(IND?qO{(CVb;@&(wY4TCeSJyIWRyKE1Y|0e
zWBQ|}ko0TnB_?*N%Gz0!-iW244K$*$_oXX5DSgWQ_{aCGx24d?H;CFbXv9I|$Vo9S
z?=9QU;2+483N;BrfA7bkEU4-+>Ml1_Y~-FJ7E#woTGFDtq5(q(4s%w$jIlqU9d4mu
zElbs5k%oJPlOOmlefjiWdmNoHoz>{m)Yc|aE50o688HoWoV2V|%2=Z5Zol~TPki9I
z@su%h(q>Yvu<8z2h{GLcZN*QrwMP9^fqGIlp9&#gPh|pKCA<MHN9<UO=(>J1)?nGb
z7u=5F7kR@nOgeS%A8s{vvc2BYlD#PTS<7|r8=?)pjb$ArYpfv4mLb_v!dZO#pt87D
zUn<)sH0$?+Dco3{c~XU3-0oVL&(bPtZy39&r=A`3$5MVnfsfh|`x2FL9x)irGb=>1
z)k_xI;`xCFJExrWk8bVT9hr*~oN{clvQ)1^CeJ;)ZtHds@t+IF`e0vqF9|E%%xUyx
zh1SvcwS@6!iSDIeDCNY_YYfKfFUcUJptT?f2(E6@QeH-XH|4jg<t|d$O+*{%n|X>j
z6nd<vB(FaQ<Q_pb9ztet^Z(Wh;{~8z5Z~UWSy!FcuflC`jl#mzmd)bj^0sYwur3&_
z1M#7AzQ4mj;#_wx$;>gzl_aB#Q44s6s8AwHUs;AHTOH$?spLbv;>9kMt;1XIpNI@z
z;WPXAnepPaVVv57V&hsSNX=Xw4~3~RS!c4LD7Y-omSd$9AR$tzpyhP16y^$;0ktZl
zd9{3jHbY}8>^s#gb*8ab_ITM4Z?*avUBozpC0nty=$?F539q;MRZ37_7R>wo71!E5
zd%Rl#KXv}1;ysydiIgdiAQdYkMw*|`%ja@G4~7PG&5fSxdz(9kt2!Y>{yU}0aQcp$
zY)#e$-$hT3e~QRw@gZ24^#8O05ML55wV7`C+5$L;?1;{anO_a|!;Id_YA?#<)F!rO
za?Eb<y{Q*{V?k~Fl04xaege~@u_@!d7jrQN!6(0F=>+D(WEN=qCx;^^Aq0~ICywVA
z=KYYHQ#o4lri1KahBx($ZwO1jJynaxeB?MuGfyMG;2AWd8pV#{FAAGa8yaA^bzI>p
zGYr@>T$v|~Ys&l?m7-pZryrXcbtmBK+@EeY($&=RxNg9eXG&?|Y%NLH8rXJ!C<z{d
zRGK(3$+u?Ff4Be@$MOkX_ws%~$0c_ad?({eOp9}z&7_f`46U{dt(tEuqF?aKb2eJj
zSVGZI*5B&Dq+{%}X`$KzjS64A8DDeN<%gz|wzEZu*`&f)%2htb*%HAISFmMz+%%L{
zb-DkomASZ6jZBJt9?4?QPk2>~_MxnyQ7m8rls7BGO@#UJ<Q~SA;8}E5Yc)9*vA+3Q
zrR%TO$z7}(07N?{dXZa;3Yn7%YZR1MTwferX&<}RvnSf1^K^mq&$)dNhZbGVuqUx;
zhg?;P%yG47pL8THMWu6zKf;&qRe<C9G|p9Sy?EN8&0TX6&G=})5w1LV2Cd=Xj2fx^
z%xDZdhrHp=t4^CZ+U4pw3zxlluDnvO?`boZ#j@m2!*uv8Ea_HuXh%w&$j3hYtZ)!W
zrwmt<Py9qRzV~giz9o29yWrT*3?tmSS-GZ+`^Dq&$M&>uP<QRfUUVF!e_fdhsgGMH
z$FZu}eU6CMT<;)U-A$?w#N~}aA=HUwo$m#*g2sXHRzdfdcXsPL`0$0elli7hxx=WG
z9<c<QeKDRcWqX^h{C>8t*SQg`XzoV{#<GV|e!laroy@3RPG+$|$4eH4h-q3R*ol^b
zWK`%K^DV=lYf{$J3pjU@BWU(c^4BM}R%1vP;v?C7{<?79!M=BT<MWGju11IPEtwp{
z$*-br1<#0e9}l$K_`SHJY7vHtpv*DTxE0cj6`eqm*=g@=!V|k3G~;u}W!_@EV}_@@
zo+p-Cv<p9mVo%179xN$l@+{X*os{S{X?J)M{8N;z%H4~j_^<64vtis?^6SYeW!Z)-
z9tNA_GY4Y#Dw-Ms(~znVw-^bc_LSS-3**F?vEy6aWA(B4o(_!ujd|}Z6@r=c#&n$R
zMX9i|tl!&x%X}dPXx*)NVQ4a+jJPo-%}NsZUyA8@PN@q|Z7f#_B~T}n{;JhC6Jeho
z&rd9Q&lWm+|1uUe&Pcz|Hgw5|ApZ0AO)Iz&Twd4rS1Iajp`oxhC}jsX?OpnuSn5wt
zJ}tsO3^7n~KQ1%4j<+cqM|ZSw{j5lA1Lt^S_;ohs0&NxNwar)F<T|Q9UbLy@rDxjX
zipVlMGhKK@r5y7d3Whcb)NgbmPh*;Btyik#WgGH|MeuZ@iJ{g+*|3K%*3BDH?IfsV
zv(Lab)@H{($1>&Cicog6PrZvOd4$HpM8)@zU|ctY3;yo|@JViYqR-TM4L?R{lmv>p
zp&Aa~usa;G22eRO*=*9#<KS%k+D&fMCR=xHA8u(>(_GyZ3nYa3w%;w4)$pEA+}UP4
zUk5mKQiB#KF3xHuYC278HyhsV%%o^;(8zC@jUlS3ly$D*lQz#rwu4(X%yRC>mc|O}
z$@XJ?rJQPSDlh8CJv#Z9k0ieX&IO`9X7&iT&^QPboKMCqybvH}!yM7rg6hX20K&1k
z0%M2q%w*9?Hd!wAmE*eCkx{QImJ=fZ81HjFj7((ZMo2Jncu=W?=p?k8M({TNGb2qr
zT$rI|_tS)W5m`GRx*Ls$C%)vdZrhT7<6MqqvpygS*OH3~5MJin_5G(*$$AVKF8hUR
zX!?BSpRSSF6-~&sVuRaSwF{!J(;U*1Uc~+Yii|DTvn1W<M4%S$vh75LsIc4iAd6Hq
zEn>Z>xHj;Fn+dllZJ(v!H6BPaN)y@S(#8+1F{0huK$QtHA+{{SKCT6ah6LZGNJWZs
zU^dZjEPQ;g+Q&h!q<sMNy}EatND`ns=0A4vh{a^z<s@!!U<*8M>1M+>8q7WxoLvi^
z+n;*%BZO=*Xh}QmfiOC8Z^%stXx<LBEYh;K7(b1OY|mzjQ_JI#^66~J=+{X!$lBBA
zk8>q+4#IpT91DK~TWJ)!S@jzn*6bbD?X5cY&F=Y1<CN?g^Mb(WO3T|N-GHG{tcBte
z*aCAKEp-xOR9Ta}&OrJ8T8;|6Wo7W}WqRey1!FQ@18YL(Bs)V9^lA@RUS*$_Or<Fq
z66Pw8ep#|=olA;?8F@uF@1p*Nw1tC5(KkKxFWf4Pwx)=^1x{HJU8#yr?Y4wMIES|@
z{FO`-cM`%L?!aayt;3|8nYN{B5R0>V#euuD-e_$w-z)4l5AkBsd}4t2YEJd^%Ie)5
zB={NJLg-`r=|ZU(`a!X)XUAGBYnzA;Em>OI>DDE~uHBWBgmwG78xjxUM-Rfqzr8Eb
zaji(|84x^k>);4<HE`0EUtIbMYJN)C!`kof-aC7?iGvHWqSizGZ$LGGOmOh>o9DV7
ze`bsVa4c4mxj@+V=%AV<#p5FkYj=!6x0b1os&wMcLt>O|F)UI;A^v7Y<DmQbNqmU=
z)sKUR8NOqn)%?T(-{|XJ5czp<iwrEHHO#<gl+#mG;MlL_YPT{F8zzcG94<GH*%V4g
z;tN~CcLu#2bD+DZ#*E@C95r)_Y`5G(-Er?h4^*k};Hr`Tl?rXCxWx8*bgBYX&e}mI
zS;ua80?bmtDt^^uWwKnY{96eanOiHoJf|bK^p%Y{4;KRm{%)xw)Uox{8rubcr>}2>
z#MFgoon4m<2)1a_%p3eDCiX}#88ze_8cN#F=>lkPiA>2(+SOY7#NInLlHl7MA-NVi
zU^ko^j%VX!rF+L0`UF}n$8dxTbsBs<-`nAK)OItApI4nIC|t2RP|ROz%B*AaOsI!0
zxEh5$S}8Za${6@i!0pSO%g{&iE9*&s1AfJN!J1{xHBxo7&~KXCj2M%?gR|^<akk*h
z->-n?^Qd(kXv7LuUW&*I*yiJO1Qzex9ysM}8tcV5b!soH5KROJ#|udKBnV4Ogqb9_
zK5i46*IWzDckD$aX1jjA@N{u)YHPx4>TY7?XnA4zcavGZppkk5=xH=|hEqfaCH<nE
z889k{s<b_MGUFzNMYKg;%pQMA`vHBTV)_qlWA|sB0Ke+@`|523v^Tyhs-y!E{I?Hp
zg*+Gep<cOvzA-PTHtR8W3jZ?@>iIj3dY4(Dl=4rjUj+~4_xvkPMjHfNq_-;H=RH+<
zHa$myS^2Ah4W1*Yez;pEdZPjJ!={{ZLKKyF%_ou;lHAq@tZ_BXxBPk$)oFCtOf^24
z-Wr01@iN?)M`bE=J$;<1<^;$pE@Slt65}#+Y*Cjew*aShq$nW6TfI3YBRs-zekmXr
z1+~aXEEiqHLB)ubC+KJ6A&3y=VYO51K$v{SHXC;DrNtN-Xnia^EW!z7$|9Sswr(J#
z7J<wAK8=SDPxI}#6Qs;YHNfEgOv5TR_gz#P+nD(bW-WS4fcb)q7^mFBP0D`F22m22
zbUtSxQGsIj*lro?4(x;LP{gx*fj!$^Hqy4K?88((kbSX|Q_I$IN%~8?)S|=33r=fi
z(${=cVq{%c;xPH<*sMAg8LE1f>}0_)m$zf@^m&28y68zg-Q`En!1m2LHDGI;JfDoJ
z<I>jQ$f3#lDEm@ULaSZ*MK29Hib=Eq+m7b)`%<{@ygI02B_gTGbx72Ip9y<5vQ+<b
z<ZLF(Y7UcbZHXWpLCSIz-*@Gi{;~!B-!<-kD-kMOh5dSiG92LMP|fM~Owt^?KzVMK
zxm$wpOPo8#DG>~Z`d*f`#EQ&p4AoG;w<@|T{Uo4bV};w!{<PCY3WPO)A)Pn5ZT%4*
zkNzVMD%^+B%A9>Pw+ItZ7m(@;k>+Eu$IDK>!<C2N#H+&QQ}J=*%%wd90j9YREEf^M
zq&-7IP`JlM>3xJrhO+7Vf$wLYdUma1%n^t5&DiT8F7JXa%}A^esH`Z}Z7AL?=SGEj
zdJrTS<S7+NoFspEzl9WLVz>+Qf2P@pGw82oBa`i;JbBolvP^TYJ|%MODQ*W<KS9J5
z)<HjNF~pK+YXpGV3-|114`1mXiM+Pk&0szCOSYYDG}KL+IOpD$jmmNs<tVqF=-!)g
zhbVxFTFH0WliWvH>ctZGaOU;-E2Va*IqlI#JgAs&8<|1HM6WH8y*+VIXjp>Xpqb-F
z(`PDObrpT*mn8NZf@mh)QlCE9>g`p`N?vMH%;TA4|9P~?`P<z3*A$vmSV<Qk473A=
zcpv(n{kquM>pSFtS(1^+OB0AU^Ucwhl#4{)Yly=j4d(@sXV_h|&tcAQLZqNMzk4eY
zwRR@K0jeQaQWBBkZ=ND~KxYW~>8Hjw-M}=#|4)qC>)%tN7#~xdsedBo^@*s~P%i58
zOvN|OTTls1Ji0KdoJjk2QvXR?B6_$MOgW)Brty2@O2;#?3d$(9Xl#)cqwFYckd^}+
zmg`jOzxTBiio+rEf9<A4;@&sB@Hnp=q73dB`ohpY$?$2?YhJhQa#c#3LPlFxw&4x2
zquXOjAP4sLwM?C~Yo94mO)N2}aL<(Lq+4JZKky52Q90T)UmGEAekC^_EeB!28cfSR
zaGGRzJIU76I{M7Wd`%di_)U3oOnJYHcUyVO6MQ**_|8Cp#DTHuB-hq;9>?h(_I>Y|
z^21lRNB;HwM3{@@%$L@&2U8LY2BfO!(~}>WkM^5YvergMr5L>-<sQFU9T<LIfd-_)
z5Y;G(6M?;+a13*8F+0rwn5=Y`CF!)XDdOa~O`I$a%fYo4>SFxz(d-_6j?Qt>UX<Sr
z7AQ(+WH#aB`Pl4ubMQM#xz6Yfgcc{sU3*y0kc!FV;PRYmG=1MKQEuWfzMdhUo8GUM
z3|SS;#|ewmzjITjI`f3V5xG#UciThG&Jp6yjat`iX_Ep{TS88>4bniW&AH+6Nw!xW
zY?)1k+ty4tneIj_YhI$nwdI(eZg7#|o1L<yv1x5l=eES(+D$pb;Bx?TX{u3~JmEJL
z)Dpjj0`(u4tOk*PUukk#Uw>xU6WeqzA8qBtw|hNQYD<1RW;z_PY+`_N>~#qmS%-DW
z?2So^&U3T&uxGSE)CGpKTfKJuw3NZz<qZ+H)$c41%yTY_*}LSoa!)yJKC-?!`(PJ$
z;|kholWZHZ9t3&)Y<rj2(B1;}gCV}i{Zo@R`EENp;qQSYzSBebr~WW968uz-!K9}9
zJe^ZsjR~ycy8Ox-lWgR&v@d;y6_W5~UI-Ho3fNG$?=PJqcdEyBr?3|T+BF_uG2T;}
zRTE${LBTN_=3i0g@X$S3e(RN)Fojew>H`?Yh>>h6BRZ{zwA{s8%&wg&VsfG@{NqUl
z&Po2Kva~UBv|1UFQeC9>lP9T{v3Qy^oAKURtf}%#Tfv^AI&X+)x8CcvX<u~%D=Qh=
zj@BS-<7F=8Jijji(^Z!PU<^Nf_M{PQ1^~C>UgdK(5;jp!oYlxhvq$vtXUNhHx2etW
z<6U2;3GD8`vUyAJP)qP0c8(Rra}noZ!fjPd;`Q*&<GoeX0(ru`D~%q{P3`Fl1nR{4
z=Yf8l-CEYUx%DckImbLhhn6w&_%z3a$=riK3gQ#CrEapuPs8jRvHT~a_r=a)n6cMC
z<7~;EmcSOq{ETIi)G1~BWfx?xds<M9F>ejBHO3vLgZZX9C4#}&>Ec>0D6VOB<;;1D
zzIXaX0A)NAMnyqTTju?#yf|OFBd^pwz(Z1&4-5ivfFp3M&z7&)lrPNbRaK9>_3L?+
z-Fn{#N58MpBrZam3C+yu{G53woAng)>X)AT>ktb3XNBio_`P;jy8qU#p$X~h^P-k_
z-^X|KHiqU}R%7P&Xb>8w#erRZ(?u|hqkZ_6#hm-ASC-yVVpH6g^t}Ee6)L47kJ`N5
zNuy6`Vrxns1cpt9-_nglm2Y5=Qt?T_Ml+2bTaIjav@~^Q`IIJq>v;F4ORGnBFeEi2
za7`EnH?7x^<e^`V(S#U?xW8t2$V)UedI#$Z$>dJ!l(jEbp|e)hG*1nFRg50|DI-xT
z9lhcB$HN_#KBp1<h1CNT3xZhdDb*rHGHDSF1Q8>@!ly3TR+p`FjPKQB%Xz|TLR4#N
z4Ee_F0k5|g4qtyU^f5&);V`q{SZZv>3U0)G^UbyM@5yeQBuZm{k&4FD<Cy88*EbFt
z{CgN{rB8{8b~Sz!Kd*jq&$_qez0o6Ge3S-JwQyLH$^4RN^`-B$XIkww^8zUM^`}xS
zMTU7aD-yTthsUoCNK`EO8pJoGvaKWSoZ53wZL7)Jo9<tpPha{FHB5;-LCdD}74-m(
zvz}8KD(6gi!>nNMpfEqvh8Dlq$S?H>_bnn^*I6!zK@S_6kY?OiRA;O~1_!+{cTKjv
z%#*<owu>bB!G-QxYSbmH{=T&*;aDLtv!5Bpi<5sP)4?9)q~7?56HnboN0-JVYnY9n
zQzZ$`R_<32hCK_9${G9&#Yy>8#o>uzQ+R$&eNzVU&S^M)auQ{N9&E#1|Fp&&hSXZ0
zH4a~Z58r&MiqHI_P6f?6$K*VLU)EdZxykzUPcLF8&5=YoMXRc&RU`%ev`o3&uT@_1
z!wyYt3Q4fBk!oabok5?ER&$K-wTiKsWPKwmzXHoelQbwCvuNOLd(els$|?W$+E>Co
zuFbmGzp71Q^4*6b70Ib`m8^jl?oiDG*w+vESF^)P?HUC0_6=Tpbi56ZUvwB>`ilmV
z_`Mu<u|eg0BipV}kc#0*+V=FyZg|h;h_ict8+RYxjb7ql9NHl*;ABEI*pb6=T0FE<
zKCXT;2AaZMZ=QZ_f$@yWoG!<=3M<AU`^jDHSp=(?hE(Ta4RtDkg3HPO>#N1n+<(r?
zcn+|2mf*C^y!Fq$!i9xvw#^<Z?}c0g2)N=!2c-d)Q?Sx`qk7$9Wc08YU&cO4C>s&!
zg;RPV7613zu9MoIqzbV=ac#b{#y%skfQZb@qT5wt8?EM6Z=3U_C$4RX|B(~^RZgWa
zzBzQnq<nm?fmHZ&9fvh)OP9Y}%zT@P8Rn>r7+DJkAwMF)xc^gwQwiUihFST6;ssfg
z2j+W~!kO|X$G1GP79+u4KkM}es?T`fr4t*MYJeWTE3_hor}%=3$WpfCZJJdbmaI|T
zm!fz9Z^LB@uLi50hM%<@2RzO+m{ZBh5~Xi%Zl%&$*OED3Y`d=I@@IEN{4U_HwOdWH
z)iBD5%bc~(<cl>(`=oCum7R-`p<L@`_<XS3c38rFXlt`OK6=F66Ws=G2p9Z&!l@py
zQRGyWF#il1@mjhe$?%}V6vVDqnfXxSD?H^ZfkA9-*(eV~V#F|y*E|OvR#T|Or(^Z&
zZxhBxrV3vn4mb&v*nJ+NV!LtvS+nq_BbW#E%_N$cC&e0aslZf%CqEl<g5E@4EmhJ0
z)fSq=odvc`_frZ|=RaYI_d79l9pDRDoEv@mM8ziAPG~qdxYA5$c__cI^g~%1jPwGM
zj)*r(wbjKur=sNt1ZQr=*q{=}5v`bhsj2GXzL*A8nI1gqq4x$&>y19QM7e!a+)ZQc
zvvx|{gq^BZYjQGGz$#}0Vi*%OQ-gMeP#%jSKgT<Q;EO3_I}sW`B>Qj_HW9;x(u;A6
z;;8o3`^{h*qI$$ug>o<RH^2cenIYD7XItQk`*a`S3Al%I`@k<LUC(50Pmt_zvXdx6
z^=!wHdPx_~;}bP@t4}OTY5LPJ*1UWL93{~MP-E+**Ibtz{hqtSVB;cFd|G%a(JThn
zf}DtVR3SKdp{~h*zH^7W4Sq5mfRFrehY!wsq}$&HiNtIJ*F4HtaUypohQXW0-z>K?
zHNyKkHAhqK|6btph<&^H%Zk~*;!>belE1gf*B=2^Fup3fgE^kc5CO0GfGozh6t32S
zG!zw;y4<j}5OSYnh^9u-TNZ7ehH0#_8IYEc%o*tCLK-R$0ZYqEmakp65~ZH<h>^`G
zovv30{h_VOg{4k4lWHk4D)Kj!OMZ?(jbq{nTdJVK(w?uD+4U&0n1Y+ZY|^z<d8X?f
zsTjfp0YA$b#BYw48Y{a<YRZFBDZC7&LwaSg;^?f~IWL#NZ@CE%{TU@fT4lT0U*2Yo
zR7P+RtspXzQH-ioONB!fQr~kw-+E-a?UI`fe(yvo4ZSOFqA8gex6(yGyD?~p=AN!#
z9hXv0STP$hf!79JtXwp!^(G|^^*;8B`L-y0Y)>EmELRYZ18O%Scu?cAq7-l};dS~b
zi6??)-0~ekCY%YA8f{a$Gx8>$)k7hi?KMGJ4A+HIEU%TI*RzTh`0OYWtZ;<z7Sl6}
ze3~r`&$dc`qmTFWS(?aI5l<hbU?H?@adLdak!TaGjP;>9QJKpLQmZI2o2Mp`i*iOY
zFWaH?gMpW7FMNa$>b}O?AI-pR&Odv2Wt*(!G$J75oUuQqJ_4A~@xRizOqu^Pb6^72
zQye#$O?O6*M#ppBmw1kN{at7ob|rtFZTCKrr>V5M!YhW>k@4)_v1d?c6%a7-CQQB_
zh=5Mg%EgEL=oKWC=7&bs*S%ailDt|jJf;o&M?WT+ayZlhQMQD|9Ahz0M;yurhw0c=
zeW|MpZxszhIym*Kla9&}8%)M!m^QKogSqoDnwDpDH}6#`b;?{tt<d|-ss}nAKk89p
z{jl+EAxoLd1p1_xfo_n|8p#}rmAVrIZq_@JX)U$)eM+m(F2eg$lFX+&HHCm2-16s_
ztZt?y{&z<KoJ4mMsfD!b28%t#&)_;e`5h}@lP(vg<hz9vmjN59b6d98=2&^BU9BY=
z=gv5+kcML<&f7&s_zpCJ)YlwKQ=g~DAgmA<@F41WP;Smr$@wRPIjPjQ$uY4<FKObB
znzB{<i=48V28#+-8lUd8)j=}vCVHodi=q_VkLznEt&4KX^S$x;<%>^t%GBwqOisdm
zDJ&Dje$SQ|5(Mx{1RBk~xp$&A$hqS&Q6#CtcnjONbEmpr?5y^R$eL2ws$Gq2(4YQh
zl_9$&44K0>#*}kP`AM6v21`&y!Ie%$CrglN_g^7{wAo7<E{xChQWbv{-C)E9Bz)2+
zDpCYZpx0og+XNKlVJWp?`wDv`(|@~tMI0o)^s_Qz2xr>fVW!@+lsntlZ8H(oaOR3I
zJml~chA=$;FlJ7})p;6+TJ+Wx0i<UL^Ve|Z)K6fa>c%XtxtGWz+yHib-Bc+w`(^f9
z`(+HCNZ9@(pq%6+nPbc1nRi}L>oPlto%8u&p5AH7NG`{NUakUZl6>O8Lz(g7<vX+3
zNV967k;6C?%%Tx=0KOcF5zLBVDH#ZlP)b;y5i7df%fC(lT$RuduX0I*Sh9-ER7|oS
zF=k13?a!SecR0R2km#L)B<Bmj?_{2+3gM|-;=(-RwsY9|QD_YwXxBHc^?5sn_DrLj
zq!DU!q0&nFcI7B<nZ=fBRH)#VGdCe;X-6jDZfquqmOD_Tfwk3}^<J-(2ikmSYcGjt
zK9yyf?4DnKmkGjrCae|qK$Bsr`Q1<>j+_yeT-b@5l-R}_yv2?u8bgk~+*xIlW>8`0
z@Tu#jdxY_rPvRjvXR}@h4-6GB%3V(6-!9B>(q5sAP@g3V+i{eJrSxiX6{(4-hh`bB
znQaJi4wW~ad$BAj%g=L<ECX+ipPpjI>himdc>R5#0O>TDc6vkKCTVCf;;W99g?A2i
zi-fasKAgVX(`Uq-m*1s#6qnZDk}LaaQkIm3G`mpu+<EHkM)0khCoos=yGtcn$ff{G
zT=txLKddnfJ}}Ko8`YxDntgi|j(67)@X_h+yr0b$Mccxkl=j09M>+-0%*O9&l>iya
zNRmBwU3(Kq3wdCuf7F!cX4^g;&hCIpi{&e#&~*Srk;~0f1zKC$3+klYQ?9iG)|~3)
zxj0SuqO%sc!F`Zgib(chRKtMZH$_#`)yM4GyJ+*UneRtag~y+U6MBi{Iw|Rl`^A23
z$5f`!y6-3<=7Yl4?VoQ*tb}8Gu-~ij>Bx@}zRvt(lrGcB8`gZjG!sSFXkb>5bTT#D
zuTUW_$J=Mh6_*Dmq(qx2(D=^W(4??26i_wa??V;n0zzF^O$j1h3prw4CwHfwLkh3M
zcBI{30xmc|*0r|eV2q?W>N@6CV@4B;o$cb`9z0+#b}cSE<tzZ^th>vSs$O7O8q?#G
zdi2;*fmRchmhi$Ttxn7w=ro63r{5=ZBB5p>G3P8?+0lugAtsrgmxD=4#E1_L?}~R*
zscrRr-IlXKuVq&loRACh4pdx7cum>VJ2F(&T3xomU7+{ZMz<edovmH$(f2IvknR*2
zmi1Q$o}JZ&$e$zJWJ%2ejObyeiK52}<Eizw4%10Vd!)}&3*;z{d`bOMouspE4y46N
zvw69OB*FSl!m;X{S$=F6lxIruie1mzoX1(p)HY`x>Fi<W3X=jjqHap_NfoPjKI>SA
z)p4Dn8$;G-)WU)$_$AuZLVb5+Fiu3~m39@y8Og`Jb08%J{Jcr)KI;=Q*~8r{zl0pI
zZgKpM>0k-N7z>!c0;hg9jOROjBu}s+c5rWO=DZAezhCs{4mkUE;n~2lR<A0bs^BcC
z`i<etvs|eTlBSh7g;C^ve0W+VTZB3FM{{U`{V{y5u82}@e7FUzlP1tJyO2|w(*fFT
zCWkvh;cASv&%f`VeFN6zyu=_E4G+wne2vityx*-Xm1YXg+5{DldB19(ZXPsj@~u(x
zPw%Sy`%{DL&=IktI)7O?^RFy|Kr$pklgK{1u;iZEQDbT$9kkIx_iaEO0TOyY9welR
zZ0?VS%977|0tF^ig5mhmig=)IU({oYIQ11lDibyJSYc@``EPS@Fa1X6wQ4JVEj*C)
z3pcTq57Ja3YV1|kxW9X%DD-gfW&vI#3^y}JiJ%{%3qY2}dZ>$n6rH{4ZLQ?@KntI4
zn67D$Q<A}AvX%J$3PPrLstIAh#l}mMxISoirgtj4f9i@Si@yTxNgplGahflGUKBs;
z<Zfy;Ucf8xwAbsEaCVtmc8G#qJBiqe_61YKFwO`*5;N`qb8G<}Q!e!yo&DclF|$}W
z6PSrJ^$)wJLDQTU$`0fS0+I!XiHT_5Zd=m{&(qD98@&~op@wo%*-B!oNsCAYNyjT`
zRpeK2Y4puuubwun<4HS{8YBsa^c6&8fL7@z#hNt2VTKvxHB1U7dLhnNC%e$zVh)sg
z7Fla0g0sj0n)!Ve(&w#VOe+4_758&t76T-1d>G$PrI=haV~Hb<k$t1@(2i-%rxTN%
z6#Upi8n_0e4YIRDa?QSpcE%s)8|Zn4J;ivNJV$-M-b^8%`rW=fx~h0-gbcxcnt{T6
zbnDiUtX<k^qiK`>N=-xZv$hzVTRu8CUHAwN5e<BOXKw`MwG&NI=92pN-T#Uqk22tb
zZ+Rx4lUw88FHL(MtkpJ>*l2+3KRp{~8*7QrkFo)kB<gw>)!3^TYx$L75SjU>)TaCu
zkPRYG6(I9^%avh{Pi84G$et>DVfjQ(BnRd8+1#walJoI!OL85C<$w{%ICr^K)t8on
zi0fjmNLz>S17xIomR^G|;}>=sW;Y_$kUJQ|59MzZFjk-&UnnhA_=~v8tvX&FNaqQr
zRiFkU<xo%sow6hYkD8dCQtVadWNRAp2>`Frg5<!cqPM09R)ISnr~u&Z_*+N;arIW;
zN{+`9glwxqO<+|>58QS3+r);0Hb~teBay9czg3evljW|M^#&BANGwhW=o)#)MH^UM
zKU~}uj>u@Abzq0uh+44@_NL+S<Ha88vB`ez9(or0^1aIQq4IN)&c>%YMN*j7_<Url
zlcpYzhT9M(h|ULc&6BEr^f#08b(4>#_)1%x`kLXv(dy}36(u<G66%ISLS#Rv3ExCm
z7U)TbaVl|IiSXNkjEuK=Q#{{-yib(sLN0hpPqs_N6%PDQZ|EC6{@r+xLJ0AwCciiT
zlgmg#PDAd++7rE}uG;+OE8H17g)7SwSDst46WQ~rKhGtrpHUjQA6ozU>91poRQI$*
zQFVPLhB5MzR5L-ZM+7My<L;LhoWF1xfJY0Cjus<1w1D%u<Ra7SXBvDPR)R!nari+l
zlFXBF$%8y9(P(jVNpH@Cy`VY%6Wri0y9vr?UXvuj<u`zim$Bsy@O*d9mL$zR771*~
zq7)6ch_ifM@1{*Oej0G3pHBHHIZ-Mj2CCf$#AEl7r8$Ekv(qLTwj)FrZ}B?{l<|Ib
z<SCBhnYcE)<@u?#Fkx$VOnbjiw6>NA&hXIGMNe|+4fRTvm|wA2Jn^3P1+Fb<Y!BRi
z(gCy-^C<o@nN=%G<#vC}9X(@@EBhJN&&CG}EJc_pl#O3cBR0?iNp&yVHc|J{H{S$o
zTzAM*ln15dZ7Lo&6$A{&4E@L>(x5KXG`aLzE;wXAPV`FW$*@YklzO7op%rjU8wXh;
zdNg(7xofvnzho7J$wQ7l3O-8tE|ag)+^%e_Cw*sXlIq*kcmX%_p9&%=r4EQA)hzik
zWvcvdlr4-s9LjP-NLUhtlK;#$eyxm<7*vZ7pJmc~>Qp67f#EAOhpB~|(a!ixIEB6z
z_Gu7{P3RdW=fjLa>WDyK$o*S2!2TvfVyGqgXj|sdT@S_Rs(#KhZ!Ll=fvQIHRUBi5
z$-b=CThA#pnQ)d-`2nu>+iK+tGOKI_`@IN2y;o82(f>787OU_k-&|DEgxvaDU7dm{
z18)B;ED{ZUi{2q5JbrYGYXpCV1Ama!LD2HkBdgQ^>=cP{Ieq?DJaR0lZJR7O_(^0A
zP|GjwU?W)2oY9$)Db>1kUNl}R9efqO0MWf6wpHSz3%5E;<<G@(!u6587*$q3KRLp^
zd&7diWcl?rMlDdkxf$*a->#Ex2SO(u@h`bw7MHKL6xg;2J!h>se=A+kI`Fp@dz**b
zRMb*`m{zMT%5#<G@NtonP$)1?<6{)NJ)K^9hE4YQuu3~kriqpUsJdPpHT{Wf^afhA
zw#4$Ep;oJ$<|7QIBNz`J&CY)It>7yc^r+_%j0Js(kTY}-Jw*xK$bx@%fT2AQYO{p)
z|EdHK3n)l<TzcY@M3p7Xnk5EeIX#@rJDO6G))vp*UN)}1b9-uZS$0EGeM1sC`&I?>
z8OIGL_DIqn^CDo>i}D~U#Goa{=5q}nk`1@S8=bdybYXTmxB$+?63EeQhq2%aFxByL
zlCd`TnI4BSbuXPQ8lBc5kN&wRPnmA(nw~*+2L9_McqmHYyv0U*JkWJBqLCp{OK7+q
znP@TI&Cud{m?K?G6W1>|Pna&8SBE&ElyFSg@aullo*d?(XoG5y<M$IvxyD!<Qs&3h
zyxl*y&sT!GyK#t`j?7`VH(>@^iE<1hoay;wbF`0z-kq9o5~CR~Xx+re6DL$<2W3!<
zt#xFktf`J9_?AaE*%$e+&NDoy(c5_mdwHI94G&9kEHS7@P^jcfQsma8G1xc0gN^Bo
zZ=E+#Q5uJatt>zc+%`Um#d9oqzHoLUzXWhHJX02Q5@TV$e8fYO!-?<=$^=P*vDbdz
z{Ui{FjTq;aEg9s`t$9wX(kRNAHv&K;_dDJuq3fqkw+%;QNlHw|7<ZrPrn~Ag*ZAW*
zvHA8O&<+@!w#gW97j?CJuaY)k3HVB9=1YC`STU2LN<O#O1Drvcmn{aQXH9+lD9^>U
zJGk?2xGdcG`b<T^#5K0JF=*kzoE_bk9U0>~@bj2x%U6q^pcs4MXrGgl&yB-rqAQdU
zgZJ>Z%K(S1oxBFvusMI9{HTX+$u!7odOiRBMG^^|6Bze?Nn8c`rEN#wJ)713s<x}^
z#%>)8tNlbr@1xG6wEa`3yx%MlDT#6vNpfECa&o_qzM&EuB|E|YGA?kDXyfncKL4-j
z0b>6NKRTq<*_A%DRX)MygDX;ZX`wvo{qz{k&+jH<D@MM$Y-S!?wj5iMSd{Q^W40c4
z3o=REMMZo8MQm083f24#(9*;Mi9{9tmHrRgUJ2rSZVaui3+&5M;=UCi`&mb!C2Ade
zX;Ae-WxPA}3w2o*s#!t%u{es*FgrdwEJg9p<5yL9V`A?MGV!cnc{M|0RaY0hLhlCH
zq%Rpn8mfuM+RRx;pjdV2TJ&^`Q`SVGPp)~8NKh2lUjLEm52H#pU0>`#ZR|=_gFP+@
zM(*20tksM?Hfr0HAC3hAj@s1kUSJiX$^Ry#*6!Q1==fN@1>bp*r2_ue7#RG~4~-i2
zC!&XA;ErhpP&*v}49$lXxUf~T6VF>7sSc|2(v~*^GUF?!?LQ%!SZ=18u+5V15xKLH
zvo=C3PA#jEYKz52l8YgjDaxQdSCe}<#JkaLP;o4b>`MHa(NmaH4?HA~PsaNN1V2R7
z>20@%pK$zd`Qu)sTseNr5bYcgJp9nM8Ot?13KTr!uBGvy0b$7=fS-?R*|P@x4>!}b
zNUgHp8jgSFJ&wMCS}C^0IX>|n^o*9Vwnox`DOo@jUJ-qvftKdlsdVxv=ee4{a@IBR
zwH}&gdzP3JVEiy>wb)?5Y4llx{?<zPZ1!>5>DJL`_z&Up>4qnxeLwtk{56}$s>q_+
ztZ9ZyR!+bojQNY7-z&SM+FjvgnSKirmV@yM!I_z=$MZ&a!k@<$Gv$w4RCc$}0@2nX
z^Xyme5&ni}q-zj#Y^29l5)D(t)a%3%zmw}w&O+{3h<AzY*GUzwtf5l*736%*xN1tu
z>-W3<MC1LYn(o?qS`%ZVy1Zv!x3np==13IWDZT-f=wrZaDBf|bZnZrZBw8>Y#M`?2
z4ro?!j_VKmsGKhjz*utgR_IbSI7Jy8kI((nhhn%<H+<)Phj&KV8dMC0Mhm|FJ2j8T
zMskC{63hAT@fJTOg!wF-y%9xw9>uUHYgnE6Cmtf&%J97GFGfUL((?Z;+A>hyHws%*
zI{<PXukTyRoShy$gu>OzW74Lq(k)v{X5m;!)tm8$!~p1&_N}uXpOy>N=pLgG5K)KW
zjZTdx${Q8UOLs|FTBW-HF%4>i*|(!S2GBd`GutRJ`{omdSG1V^fnBD3RAnj^<@_zI
z!qHrut$vt>LZG4WSIgX5SNnA!5bECK8%(&LAIkB&=$+aya|6{ahdDy{rZ?2G(O*Vc
zt(yAdnD~2G03&h3+Ge%BYf5e$9>Eg5c@S1+Ukf*1m|pF`pW2<`xCgcU)>U{mi|lMp
zB&|uT{q!07G{H;bu8v7L&y0|@8GI<4Yt5wn;wM&pR?A&i0?)A!OXP+mY0R|TD70c1
zo@+^lrd*?_%|k6IPb;-9H=9ntsE<afb%35$y6ut|Dg6(}b=2N-4-VPPt{;y5!;*e_
z2G;3j8*P$8_vH_t?j<CyBKTJ(AizBsdG6+(wRV_VaMORp+GfTt9p7$NzIe>>i*WR-
zfo<NVwJ+Eo^VX}KVIFBWa5?Qpcv7xE>tH69E47b9ws$Nx$^)w@z0m~+uSxXQtT`%}
z8-DnJ87NxAn%;wkb&+VVu;G~x+j3bCTgaZi*6i7=PLT;uq51@bG*e2f2UeV27i(sv
z@-^NAJiXy^ztQC$9AByX%Pdcclc3KTTWf8(xW~3@j1X6=Cv*_FSibzmV7T($-9mZd
z0O=dV`h!`|P;}{mQdWJ!gwsUgT38Fy4zp{Dn!e&l-w%qjsQTPy1J%ZC&vlJvl~RxI
zLK$b~%%aD=RhyoksbeQ;l=GBV4-U)sD{xkY9UmP0V1Bh&ft}ji%9lb3wbh*v`y%(^
z{4VrvD0-a&ZA)G5p-1Lcn(KUJag}-CIzrO#_yKQA40AD?ER_9Fe4q0y(oddJ6zqBu
zh9C0y0=d*>R+QHYq^(!M^MGO@+N+lh;^%`<A~oyyw!bRvM+j#8$cM1n@$MEs_+3E`
z91vR1Bn-!<6tvELHAe}3=Y}!<-ss$AiupmzZuLv`TY7*(MmNW!T=brfBC+QU9Q=NN
zcc!L%zET(Z6Gb$>WBL6_e|FbL{fuL8#Jm)7(tJt+lIKqa^07tW!M{U`HAtbIe0Cg7
zuNQif&r5<^23(D7J0#f%O&Q6}_tC><!K@Mw;{I?28CzSlxXE5}8(kL7LPFt@_u^J@
zU0_HmtZ=p#gdk7tiM?@SDp>BM&U(sYs=<HWb@_pNTNLdb6xFlg2@#C&J*RY%Z|w#W
zR1jhYfg%lCrPe_LfI=?{QQND;!&nM9`?m$a4s)eMw-r%zn8EU^ft0;ua)X~yDrQ8H
zxdf2xM~^7zBpb43Y5Js2L(S4rrigrzsEdfXS%cJshYKRAY>BJ_rIr@&2^ZlT-;-f!
zupxlO8gh;PQM(Jd(2j7XG0A>P!FVY7GE@-B6{nBHJ`@T{nHaaw`1_x!I1MjvJ+O})
z3j%*L(>s*fKG0#f5Qn}8r9G`|Sp{uB)`%*{>3Oo=SYH9Y?y&6$_9l|sJ_4V$L39|_
zQ4e&C%@8KvCp~$`85sS7$4r?PtzmpT?prW3j*w0VJj!FdHEio>iQ(!Q<5Ci|>Zh93
zR9HFI>|uK>$u3lvX`jJ1&e1#V@y}?(sjZR>JbG0hl#}ydPyOO>wKFJ1F@0ma_Q1j)
z`c~{hj6!nG+f*AM$ny(}&xA<sAiZ;rQJKFoD_~4Mg|!u?_Ozi&U8D8BW5P9{Ce~(X
zN8d*+Jpp*`$}>BBN1G)<+-za%;`FS?j9?32N_sKePC4(Ftk%MK`Mv0pH0t>vh`Y_)
zxZ<cy`dAI&r&lx*kkooq*EDg=KEZXutLM8&{krbkdsYT(&l!N3eBMx!8|qvl-vrKU
z>=6k^N@(Y1hGDhkpJ&}@O*Ud*yVWTlt`gMK>PtQkPmA8z=OCje8P$--8}=8{B-D}!
zq}zY_x+6bT*S&LZW5=W57$+rPkSZC8x>EX16-fwx&HQy6Xm|c*89*=QU8lNYZ#dna
z<vK|FM|0j|2Vlx8cH0FO{}&ts$h)BTFn0AxSRb)d#S;j`WRHrPX4$wlDIBmc%q?yT
zy@O2MGiTST-PH{czl~(-2^0MGl1j3!$K+#H!Lt2sv<}UhA;GXKTm5!e8SNczm2I6u
z0fz3W{!c~~7b{naFe#e3jc>I+nOtt6ke)CfPNXC2M4>H`6DIRy^<?y6%QcK_%($1L
ze`57IicdQ&B-8MwkS^~9P@z5I2ri8~)?i3u9wr>9*9jmxM%B%EuMic^hTl(E|G=Iy
zDJ>m+84V*kH~E*K`evB0^i?lu2Q#mc^6R0{dWCFQL_pvbyy8UJgduJ-cAKg~gI`eH
z7jmcHJTmIZjm>=6w)y^je%axPrNPqrUE!!qtY!CE2PX0{O^tGCieFACi^Npm2)1XQ
zuo8L2&7pi(c~g!pvu)!$oip|bfqY97C*e+cU(SXd1><|8CUhxZV#mKZ^5#S%yY4ot
z5KK>x|EdT1KAdsIK5nN{YXQd|7fGa4>bO32W-Lk6VT9N-)Q797@Xy+{{&1P-bM$=%
z*Dcbj6+G-)mTvU$F5=4e&KcBshrS=TryOT-?$s$KrLzV2&bvI)e@q)`6!OY~7(Va$
z&RtaL-}y%J3~{5oj-c0hj(QoizF#oG(;FeO8nKe)Npe?;``eMaPtElLF@$|_SvLD$
zyq13=-QTg#f5P712X>hE7kw<AMxh=f%tw~i`H5PP9xI|eXOQDrVBni#!N-4CgY+Ld
z#J^8pmjj|^*JtI#e*^0O_R4?XMg<GSC8rtqVYjr;A;_SWghlUCUX~T(BeCG;O=Rbv
zKW8Yn?PC4<KVIZh93n*L2*hUpKOy@6{fGbaDMHa@?|D~j7X$}oaa+pin@12boEkn1
z-~$1iBq87pO#i)#3d8(W#{YbwokZ{kOWafV|Kn}{KOdCSA&rY)JNWFrtLOuT51CD#
z=QrkTIr^6Ykhd!FI5qIY#*u!HVe2x@e>J23s5#OQjGr_&Wmm#~&zS%11Kg+n-~J`k
z9`Q9y^Y5j*{ICD#e}DGB|3X3!+T3Jii&{qi>6}4AnZtX)o}-xhD44PfT2GZi{{|9<
zhxb2b<408W5h76|WEc<x*h$~6D54L@%9#}YZ$Gu!Fdq2C!@LTU{}JPJy@%umUBOp@
z1?lllgHH!QYb3tZVxU`Qt*Vv20OS*VL?wxiZW?i~vFJZ6fY1!u!-j%ewp`aiQtH7S
zoS(pE*Bn#;9Kk+F)ysgKyb99HR}ldqRfxRYF6cfpWRx~zq0*?bWmcuU-I4#~&@h$}
zl(_en-`&UsZcfd`f${}NP+fq7P6QB%^CGg;b-(7HtH3>+JK(I+)Nx4|LDm79^}cV;
z|A(-*4vRC3x;=rQ!QI`1ySoPqp*y%18Z5ZGy9NyqoFq5|*TOAm&`<<-E8OR;bob2t
zX68Qk&+r5&s?K}PK6~x;Te<&iseeAg|6lHwv;^>Wev8b95dHhHPO=~XR$t9!B4xA(
zfDjO{0F1iz0-&mm)w^TQE^<eJf7o~edj)2Zlpd(CB#fmhZ0O*i<W(&<a4y1z8zNv7
z9CRUo`bSg5=eEi;LG$nkWCk9KI0DYl4>vHJ5|D-Xn72y0jq-<@1H5(TYXGFjchIJ)
zVfwWh5P^MsqgE|MTR`=p8~zU$OX^__yBCPA!2>jwse^;>`Jtzt08%6jMl%V%dwM(x
zl!C!jzEU37AO;aO!qVeGZZ!V(Kngispqsocw@`KK{?}6={WC4V59oVc_~ig)t?J1R
zv9^S&`#p?y;|27ce#I;RljX@$OXuEF4ptMvnyDx-dvYY;%@v%?FddI?H#7}K0^0$-
zgPLKcoUpc&PO+p3C$C=eQZ%}vtg){?BVNwc9asBoT0oAVq3~vs92kJdk^?;7nojVL
ziTQ&kfLo1Neb^Q<h5(?H{4f|i<d$Ca|Ft3gf1cc6GGGVPc_H5M3h|$X``H$;Cdvhd
z7Xp;&oi)JU?%wo%!vEN{7XVBx)&L`2H#;?8Xv>8Ej|p!qXdl4RK77{;!;T10mUQ;g
zJ0@krFh!>qGn$dWuu;cJP3d}UELW!&j7<gxpg{+Dsh&UrZy2mrGISjmkm%wY!+ruX
z++PC!-QNrF&=^b=|B^cE!`DDo-SP&S0hq-dXu&~<Q&@ES{{!Enj<}*t0(`{zxeupD
zPJzpn#t;}`=x23+yXHv(SPozSIo|*->P4cTu#Y*oJ>n+TbnWPq>kl+oEb+4I9gi2{
zZXlk>Js?$sesgUtpXXhRpZLP`^eSBSON_UCA%BkoFj`5#)D`F~t2Yn#nrgRfoE^<d
z-}nx2+-`e&0{9V;Mbf5982q6ZaJ;Yk;~X2(KFW*$s}!>rz#&O+IFo&p-?>Z^mV*=r
zOa~hKU;Z>(w9KINgqf1ZJ^)Z6u{|dsmxKEHzMt4ej0PLDmH0$HeKKl%Rkseh`#%EI
z{DF?YNq+a<AItQg4pA`NyTO`}I^Zk`1FEYZ!-0M%btZLR5t9?7EPze<%=&r(gxxgM
z8*);EW@|R36V`S0spl>I+~YeV_n%kflmQMvh3*)<2k|{QI-TA9MfX_Ds0UBK4I9$$
z1<FcN2oMVBySPo(mu9T2D8Bs7bnj)zs0IGh+XosM`<fnPv#&rFniCO6-x}l|pxM+J
z5Yay-#o4Sz4Aa?;Wb6fmVW4Y~2O9>gg%dZT25pkd7Y6)t3Amfd#{xK0o*ZBkDTi&h
ziHWo)SO&Qt%-%W<56tSDz=kF|p4OHq@OaZfwIiy&4Zv*xWxvA03wP*%^|-!i*-@<N
zFtW`@P|pv*CsBEB2HuXEAcJ86=K2h)&M0r8U|=siIRmnp5h;xRpalQ-8`&(3NJ92|
z?_Vqt(0U7p<CLV8Fmo<R{lgyX_caW|`WAIMZ}<b)Oa);eEawzM*8(gZ<)r5t{pgR)
zfaY%vreg-LhRKLA{l@m+m2)&Ygt5soeOtNPQBAin@krj??!U$l>uPn*tUB2cr10!W
zdKf}sk6EV@$$$Y!)&S}9<%!RH8yJ9>v4c&pekA3CL{fQZKrf(JcJTlV&7`~IggXkT
z1Cr2VLjULS{XYsA90YA#0NYS%UiC`-pM8KEipsBWa}VIx){a+zf&LSuiYDn&pDi{M
z(sV<Y;hNj+oVNZ-Ai}iSQi2-=OQ#+Na0~cy0ec+0Y2se7;yu-{ge`0S>R{}dCO!+^
z{hO&h)V&Ng=VLU#miHkTi3nAdy;fc~@bPlx(s|L>aH9j*+1lLi)t!8_y<pIwE?R#x
z??V_i3^2%4r&4}pKbG<7c}33YuJ>k8bcN0MET#sQQGwo<(Z0!8>6Dghy0Wcma`%>o
zL*>JK`;3MdQo`S|o&LaUKD+(_a9ce+VW{nol1~Yq?Y%0LB|=J@=3HT2z`kRm;Rr+J
zt<W*O+hRFLQ(zy|+D;Ms{na7^_o=2T>sMCkH3XX8)O<O1(dO0nY6qqfBF4hDRE@gN
zhMfQ{1^uA1DbJvut+V&!7npit;*-cs^rs}j523HgrAg9GG`5P^56#-<*w(5|*t;$(
zomz6NrsKulvgf8S>$3Hdz1j-#xdRM8LeeS63;?m+48$1Qh_4#iP^4>Z1Tdp|0sEQY
zv*dC7ye=qSV~QlkH<an&a@mTg=b+4RE#eLU%bj^-NT#|dh_nIS$6lZ@{d>MPtPv1O
z0Pv1(!Od{No%$~TYqxz0^tX8`fN+tI&b_Vsz>IoN4PYS@lc_oEIlJl?Nod9eV$8!k
zo(E-D07@WbBgDt92@(_QT2VPo%qO`Bge9@vU$z3^Xa=@nOIx-1)_9I-*HCd@EN|d#
zJR!OzVgCJ!5<R{+`Ak}fwKwDpcca)iyq3Z;wief)$I;i;32g_{$x9B1Pb|uldmUbU
z19VZ_Mr{Yen(+d6#8ydxj_1}$K*Mlj*Zjr4$bUuKBTbMbvYvB4;k~!f$$!t~E;{_h
zJgmYpl}s>xJuQW>D>r#x_<W+HIZBZy!k=gVf49f8VYJNBq`!p!JLJgZKcj)E(T`35
ziX+&eX=R%zTh}HMu|I^JlRRrzjtS=W^^^Hd@(6H2Mu#zgN7#Us76-2%U;rKzGbp?M
z{g&ZDX3l5A7c%2=CI$fKnq!rT!f7Q5k?;@@n_xs6=?i*DbDAG;rIO%yh@hwephzD%
zUv4ZV^xGh!gMeMs;HDcz66b?|WZUasq<F~kp#bi2hGR=QX=tklg&G53^K}79-Z%(T
zu6gI5PJtt3ZUgRYV^R(q6F}0xd(vMPDV8(jN~H>GF#Q`1s?4k^z6wl)fPp61x~?3J
z4-EHNX=U*H`LTB1*f*Tg^H=TR+>jw4>0L)_`hm}}7||>~!72d_E5>}9r!|iYpk%aH
zvV2Fnb^@>+TT>a~(5Fc(U4cWk(k&4+>jD6beMmtVAzPQp{hW|I(fmsipeU+tB%=lb
zVIr|*TVkuX38_8XIZmDLNQ%RG6`r&emHr0eTz>Z8nUf)kEs3M50(c~KFQ9(-l|t$s
z>+>=h5jwmhdEN55UmyRHqVNyvQKwv%uGE2sTB}CZCJ%s;!kA7}oPgb;VB&`i{?nZU
zpR4IThfb29t?HN2X8b^aEkOoAsJOgLP!}oQ9s_|CawoA@E1s)CZI@;Qcj$L)Lsh!h
zMMeh<IBPWJ(^rHg`JI_Qu-tCpgjt1sU?fO<&4QCVg5(M)HN`&pL5G8D;UC#9Ka_lY
z=hD8H>S|k<kfI|5@hW25JFl7XcCqz`c<*3J@DS56`y)DA8uqXE+8Li`bAtvj2DwVU
z>?<$!*MaZ`lyQ)4>wS{}PUD!7`vzWidv4&<ZAW})-=O&rJPO6{+n7<aDjDQ4qza)G
zH|<0r2YLo6=qFy;g|Nt<vcG~eRJ0ue^%h#0V5Jv;f}?TdqgDos=iPciG+<z{Yoz;&
zxP+omIC6OssX8-tUI|>&!&t0Wt8g#zU?uzu(I(^KXl|d|ggjgXH27`x-KrD{_gO7T
z6pgMN(L$6lAwkf`02tLWk~+Za0N<Vj)Z1X;hx!RPmezplf@{Yu@TT*2W|ZB`kPw<W
zgtoXb%QyuBCCx39z;=y2L#cQOv*mYUvjA1<AgXBf?p?3NUyZ=-?(uV`_>>kMs1G{e
zC3x*5!gOY{imlJ-?+`kd)u*$S7hV1C!|*jw0a}gBR<``2?>~Pc4nk@ua8XMQt?#7!
zx7su{MCF&c0Xb9G;S#y&zOm*>NAg5<)XfzI$c8IW%1shwbp<nG(b(g%Pc9=Q=+-!P
zG}Syj5z?V_rLO_KR#{(H0bGd9ATi{Vgi~XMoXka>R=~a+u+5i`2E*R4c`m37p5tuS
z3?XX;@ypbit?Lo6qa_=M-8Ty&q7eQ4rk$_F3YQE@?`Yv{o5{slPmWiDhsa|%W=<RJ
zC4195jq}<UvLSR<QHP`4T5I}(79;pGVq@5pZO9J<$vt@*E?F<rI%B58g)me8k!fJa
z<eMamFwF-Us~$3(W?2MFGYp^y84N7HTD%xu@G39?N%kQJWA7LSyutUUfy#Vlc{#Vy
zxEkQgJRwWdPxm$k#T&EW0Jv^`D8wmVD!d#4tQ76iF~MOW3STUO6GV}FqPSV^am;;<
zP+11tC%b?S@V>bU<)AD9$$Dd&Ai_GeJcnV%+re_D2Hb{dOU}uh@*N;23obb$oO~U@
z`PI?y{uuhzZPInILtY!ONtc5a_`6069hX+I?B;+A>Elkif#cpI%Fcy4i~W~-oHd(N
znxEJ?Z+_G;$oYm5t^|~<b_x$BOH#jm2B5<GbIxC|VVa%6fx@Etv*PM&0~E7+80p<E
zl=X-DtQHsyr34||9TgS_km6B3$_R~i^j|aI3vDp3s|6zKrTB%Isyu_eUa<A$A{J6@
zG&5Yt#6^GDT2a^C{hV)EfeLR4sNok6MO;j36_{_I%8Bh4Sm+<7RyR^IAIdsPJBr66
z%3o@cyGUJT)*)!MvZyya_H$nv{4wRTAnM9*aiKY|kgGriv-(%+FkQw#dri~geBnLG
zS6L$3)gg=lf*iHUU#QJeF(1+WM_5#6_C5!O0^`=EqG+w?1B`K6qG)%dZk!a6S^!Px
zwvMGQLp`Gd)ipp+j{8Z5gq$beG|pg_@UQTBd6K`)Jln8nZ-HT3Cy5BEh+8ektr(wF
z-|xSvM7y*g?F+9@12t)$41wamax}@m<~5L61G@%8gr2tMrw%?}Q!tuQBp73MFx<vE
z)~!AjXt&yuFOKCbDPJ_gTxk=8GXjDy3j*_ELa?HFdzpkS()FyIJ8sQmIp$t-9<gjn
zxP_|#oQ01;Iw=}1znlkenbrZ^S!4WPi+j0yAFy+~fwRTqy2=G;q3bY_a&cGW*N_sd
zyxhva%JJs5LrZ!a>mbd;y^rZ?wOy?XR+@kd&S~NFA#<b6y@PjqOY*|heuznZdt_{o
z^tm8^pUKP8T+V^ng)S$}wr$sX1Q;ON_8DPNBqW*txBp=pg;Nt8^8e^~B)o`f$<z*I
zy7Rh=>s}4RmV2&^ava$q97_zm+0d>d(wuSydOU@k&k8x*jVr2B_OfL0-301p`JA^?
zOhHo>n+o(y3I;(@jeM0y0cVlw#znY>4|RGD(TdV>f-(tKJY3Lt^Iy<whh+Nf@C+Lw
z>q6GpBG%#a<*78fDI&>2b^RjNABF1O%g4llr$z>FiGnQ&iknfT6|tq_(82CLqM;&n
zmt|vmYA?tyF~$0T8IMHj;?*IlDfSt8gEwcw@!sW|;4;Ze)&|HFDlVhvC|7W<dd;Yl
zM)EhsO*g~1IIS!?=n6qXW&0>K|CSE_bZJKPgU{I~7tE*a=}9Nj;l5fUt1QdHWSOF5
z<=E<m0PS~U5_bHIp8Rxjn>h*v6j|mR+0#s9z(mDzSI_nM*W^_ph4D;ldw3c$=kbFA
z*5`p=)F07NDBa%h^S-uq%jr%D(heZg6D1?%pVrtif7xR5Zt2yi0x*?&)kjbdG1f|m
zHGkP}^ZLnVq<g)x;K8Box+poBJla6&$mpzh1Fh(bi^8*lL`nYl{^mxOGt4`rV^<1A
zk?UWs5&@0E+H5}V_vZ9ZO7yPYZ77aW$>4S;fpF)B{@!;FYG}vKFfN+GyfAQX;w9jI
zr{5?l%)ERF%$X}V@~?ZHT}?igu9}I1VmBO|`OHf;tc!;m4>}%9E9CNLVi(puEHGXJ
z9bim{YVNA~@DQ6#87H}N`i34$**4v|RbEJrh^y87-QP_|T(PPDjNu>$cf&MA2P%hB
z2Zr;;<TI(RTadBIoH3rx5?EkPmwrjM;hBd@a9y5%`Z6=1P<ts;wJ5SiVM`;?y?roq
zSP<XC&x*iNFc)h;Y*(N(CsSbECM@mUVsYFOQ$H%zmv$HagT4~ieY2iX_T#30p`TpS
z<~d<LF-+&ZKsSf^={JhhsnjyTj=Qh;-Df87E^8elHtE!eyJK;#6#b>*a`Dtvu>(^=
zqwT^8`S8Tf>aB3s(Hib!k1lx!7zV}Llb4EwFcB}kbqh^%xKRCV$&hbD9&kRWxc*55
z#g%@!d3Wpkdo>dwo-yij%`X!m4Y0t7R;yKx&_RHe4PY0@IEa4x$ctyvpziB%_B}pg
zN|0r6c9085(wY){)g+;P)mzcsBKbqGJAEauao#>PJh4f1FOSqjYj6s8va{fusZnTA
z^5P?v1H9|%s{#sgy`S#XyVklW1>>KkxaD5v2`O$?_^edV<SifzpRkeUpHf{S<i6$#
zp_&%VX}Zgw#RjH0-_Hqr4Rc$F=(UmeG|~L(qcQ)lU4k36ruX%K2f8#W1}TC%&Qs=%
zoTQA%EzlQ>qn)cYHEmmXh33lwcwQaJgX58O!JsuNJWc21CX5038xXHuy1QTSQUx@g
zRx_<*v{3wYoD#Kv!(pQdXd2BsSplcsPX|b)qj{*6b|8v)MVx;Q_og#DT*##1xHu}j
z8}1et!Lk~iVS~21SUCDLBk5)p_fSg|NH)h=bn1FE%z}264dgMV1Rc#KTE&kOQ;uba
zr}mE_9V`&nv2DMk&HgSiaV~E>?AWBCwN!hfNLu=78{rEM5@t*WFP*IqQA)BNiQAT1
ztKj=V!0N|TgP53>pTrj1_dAUEH3&5m$kNK2p^`;V_~LY203zL+U=JC90kTNK3mzm!
zfGY{$L?V84{s~47bGoBS>+fZ;a-Ul!txwne^N<1jPH3!qNaGZHXv0svrN5RGUBzZ&
zT(#ZP;7DWNV7r<XTmiV-qMh%0T%<F`I}=!=EqX3c9xC~M!|oWtsQx+gu2d-<4+z^@
zV!$x_nxZSAk%}a(ACV+YPE!8g?ktHYm(fTB^>YW7AugkBaT#BWHI_{XORN6a&X9wp
z=|pCZWp+z3g}(7Zuv{?lr2lUF(GFoBzY&j!=VD*ttm#GoJodMSjTwTO{4<W{98ph?
zFD=5p)W}rsxp6;P$MLKn@*>!l8J0@9sl+``J0;l59`RRG@jK%1zxA6E(b8Oa=-!qP
zOu3?e+zoxA_O(U`G|`%S9t<RFY)&d|*&w3JMt>bccT9$cnoSo)6Y`skCWa+e`OBAb
z8?^UzQOz_8HZ<8Xg;wu5ZL5l<)EQOl{p;<Y@~_X|AKM-K?R@e*?$~L2%G<fT3pnZ6
z0Sq4JSy|RrdADPHm%am0MG5r>P3^BsV9l2*0~R?!$vMjbMXX%@^&9TQ!b5```Zh={
zX%(oA)@1Ct9A&|fwkHkzX=j7_xp|({o;GY|s-UCzGxRq8Mq>WH;BlDFdjY5{DcDX&
zax&b&Bb#J>-z#4&&}{W{90BOSJQKkoKCV|tV#H<(rwK<{@d!8*R%ot5+i?Os?sKjN
zw~~$exkGCkm7SuMpi=)A<npWft1{iQUl28dQi;bjns>!Mlm_LM3pu`F!nvg(n9E5Y
zMO`faq5r~zb6VBlb1nwsgABhPF!5=uC~`>&70kiYC#N3`vUf>W>XC07`++}le<p#0
ziii*_ZTg6V3SPZ4qjtiA{gVdxswgf==xL-g9K94G0a3u2PiYY(B|*&HoFXDvuaAkT
zrbN0lnBVheEwU_KsgBqFDWM<{jke9eouJw`k(8I!l6M@8EH-b2mZ}Yn7Rr*OJyJ8(
zQ=_6&8{_$H=9V1$2$5nsQENsMk|*N%;phRT7Rz;ylc}bCvB4sIbyQdUz%5*d6Iv!g
z^!)}jUzFf}xiV)&wdJyCkrI^Pe%1P7DtIN{$fEXvB0tH7=5`jjnOE&8V1U&6MppVr
zgctuQ;t)EgHSI{VPoKMeU3poSpyy1lv+8?~8VE8SZgcgn8*Nmz$zdOv<$Whsc5EsS
zA{J#(r$o<o`{d&TfbkEEDQMx}Mtm=*(U?%Yl?XeHEO0~3vd8t;+2Gcs0R3dU<BBr%
zm+cb#>*Tt>p6t4gHnYD;UlN*}$Uylqjp}}3u<C}bBPC&qSnj45Uy-8$UmQ6DAuCw&
zGs8ptPepI+qXvGB`huX&7cSumJ4^gW_?iXmjfLl@yoOJEHgqOlne*CfZ-9<OdKzvH
z$6Z)m7=iLaRX9O==dR681EBhd%H;Wg&8csFuj=z9T*YhRj!#s5f1&^RFe`z)`k>BD
zCxDi6I$w|Tsrvif)H|{)W*#mp$Lve@BY~z%ig{tuplc4fo8`mvpK;^whLwIyDBe+s
zLKc*5IcHKjZa|u$7jSw-klD8pCwGz)zpMl{c4OMybDjCm(T5Z{3h9Lp0Z+ch%dpJ4
zkBrN%B%UDOkk2c47hFE><(PGZgNn_KL}5~gV=WV*Z(>Y)f6PZXRzXm`II@wK#)0nS
zCm5q!D!4Qua{T;LV`x)vw)nfq_|+9u)06HT-oBBlaRTx9yxu*T3{314Dsu9BadOav
zg0+WtX#a79`WMKMPHEovEZ*G37Z*w23D;$wFZO7b$nB`>HLAK3LdGe%0+;1{*+*$Q
zQDDotW(v0%3vKo@E5kp_9k#v&aS&Xlob-bPNh5u<OQJi6bF%il!mWf!Y1JY2YO$S6
zbsuecZI+VXi(>t>NaIb0Q^(hx>8rZ+^FE_I@rsOY1}%A8L6fYPFm)iu4^`m-GlexP
z8&ub){5J4leh~r6^W|F%A$b=5mQ*3!Lf{8>n6-RSz?orKY&fsttMI5;1T?(&OaAku
zrR}uXqNOW4OrY4|853{2eoysI^-bcLGoOT+7`UHEM}Aj+(Qp=`g7vM-Pplz18D$tq
zM{wj1U*k}j5YqN`EmG&Bt=?O5^LU%_;XbKFBomsI6B|AQXmsn@KvEsGSY)Qq(CC{z
zcS9wVEM;Zmc^&!FKw_ZN!?jq|cD%oE;6<3K1CLKSwVcv^?JluBuOKVnjeK#(POs&p
zODdzLM6d`rjV+;9LTGF$5z)U*^`p-3Q_17GmR9WBU6&gj8iCfTfNaGoBQBpI^Q1<b
z{2*ag0^IxbK@wdk#4vB^_B~w;e=413f(Ib@X2-c}Uk}>vG+19$Dzld=Xs2q8Um}R=
z4DiS`6;-<*4*o?ZxSk07#PQ9R*&*$WA?f!*01i(bd8djdz#n)3m4yF&0GBAY78!TZ
z_X+*?6!DYQ)MFi)u<@CuAOC{7ja5%ES%~KuDsfHCM|bb~Qwt63#!lTs*-dx1-%j?3
z>JLmh>v3y-)={|0xp)LMPCZabl44I;$3E<y%xH-*f&54<<O0excuS{CF;g@82yF>k
zer!6=E|;_?Ry41=%jD3h#lD9ew}C$x4c^@8zP1p<x2@+YpJ${LvnWP0(u!$+n>o}{
zML!iCcT18}ShUwWm6KuUqO#)^)SN!`XsN&M3tNN;2k~&Nn(~hLW&`qq^+2{v_q~n7
zr*k`fQvqB~D<<vO?;e2YlQRTmO01ULv8s|cnHSh0cJ9Bo-L8wXg}KwHSLeAvoE^_k
z_r6j_5Kx6Q#l;$w4zxsI360T4&hpC`e9rPf9LS4P*&H64=_k?e^E<h3z@69$NJWk1
zqU$2R!VdeefGls^080j7I<bhL40+PFaLZRBO1zurIhW}wo57<<eV}q6Id?|MmqWV8
z^frA`>l*r`lOqR!=18i)sb%z=+uRYM3Awdv4UYec5fR>?aj>c-hnz-~#=k4Y3^L9G
zqOSy2tMaKggxqm`NkByV4?vEo9jtKUK`C*Zpg$jTXkDkIPpE2cs~QF6Q#$wIR%w$I
zWe8%~3y(c2lsZdHTeHcSa4Ypt(Shp^MC|2fohXS9Ee*<oZxLH*xvbxBq#Q(0VEN*B
zsN&*3;vI!T^)QF8C`ODC?dI)7p|0T$*kVzFOl?MR{t&z+p)*_hlqOg3o|i$j_EP)G
z;tyBD*!{5w@%*?BQqLmKT1q}gB@=VDHbCc)AboBcxUdu-^_yoENbL;0;^aND;FM_}
zGl^U(AXZ1Y?G;qcI+b**X|MOr8T`Q@8({6d+lN~nxCr@c-|yJgHEZ7c&f9)$p>*$B
zwWmr=qL6yW2$r3~F;AZS`jPa`V_R#9#za!|GS^3?OXsEKqIINFBvf+x&C9K7`UvI>
z5^K|)SK$w7t)JVQraFsyyF+|9C!U!?EUoeOoOV4)L*1eUqp~&dcyX_?Wg0|Qc8;JJ
z`x_~uI3R-US-M~V^gY*<)5RKi!F{Qncb#Ll-4Yd_%fsZJ#{<eYU7p*vnxDA{rTyp7
zM<7Rp$4rv@w*-I{2BEawSUl~y7ih*GEu@vJ11`VahqMq&iGOpJ$8JLsIgTvy6Y2a*
z@e@Ol-^uweizZ@Ylr#1qp^x5(*eYJ_>zPFDNFUmh;*g%&xTxB%YSA^&`~7>EuH}of
z_5!G|0|Ca>!)BJAN%w2H#=ZoxPV8H9A2l$YSnD8^zx#(YX#rE<gE4vmB)BXNHMe?W
zh!~s($sOtDea}F8Jp&6O*plhkj(SPDjW;;<;TKk0P^8ZdsEJ8QoXTUtEMSvR%<?_~
z(0?cnx)T)lciCG2q@{#Z@izie#fMg+^O4c<6EEx+sh#u<qtMNJ{b*FN&ffYHe*mcv
z2U>o~hn(J*(;+@wv*;6VvXOb79@+KNX6G<Zf%wBFQXaj6{FUe^?wVK0q1b?kwQ0u)
ze^+2sq?EXw3&T9Sp8V=h=O6fkx|cCes)}c#tlC|`OEbXk$mu}ZqE8+P)9#wg2m0UR
zn~9xd)oHvGnjt{3z;&?ZB(PLYG+=nL!KfKq5_)lqnD%^MXY4>QQ5%;?(q*evdRvB+
z!4T5yOISIxa_L(>NI4L8p~8FRsUXmj5UCBGR>}ymmYR;tFcN1g?48OA!(XGbb?#3z
z-d>bU>>AyD;+6Qf>WdG8GxneK1tO348AQgMTLzl#qkdZ|GQ>v@@@!8pseNqF`+Dwt
zXUnnM7Wp0$9w{9#9~ft5Fz7nX#cfO{uGGmV_%(mLOC+R5^E8{OE^(P^m%b+YcRYz$
zALa1}l<Uv&I?L%n$*7wu7VP<|A>mPHR4CqkMO;ebu?-8KbmFtH+Yep9GCzbM*<RP;
zdB24P334!Ri9~uN%A)_;-ct4R!Zd$cPO+@D=@*&%Ua%1l_4gt~*3<~<mg6_xq`In!
zWL!eA#QG7ptYdJRr2a_sWy~a;*kRkVK0(I5#4fKz{lL2_9#=`{m;e0PU;XXD9$7n>
z3uLyw4U3YRZ~I17=wU-29h@_YXK@p*OQ@R;`Ob&1NZqZem>tR6VND;2-wPig;qkQu
z+mMzO=R;N{kE(PF-vq(@8#J7_aRb2$N$kKeY>}5~SMePmuYRy{cwsU;?z%XfYS5Xx
zTY*S(v*`ObM3})r#7eQyfOdzCy}U*Re-m`zyX<yDyLk^r>@_!*D)8UC1Q0!LULdr4
zgmfsoS6Q=Q?-M|NvnIYH-MQw;=#k^fw9Mry%=h)^wF>Dh9LI^c>iwRsp4@J?QL;=S
zLW?w!ore3!{@8=Vbu6B7Pc69@c&f3)+#cGfv}9{Ci;~Bo$!PWzi{e?8N(8PApoN;7
z%#!azU?zLWWBJaeH7RTsJ>S>xFeXv}N26>Hul<@wBgeYZ+!#C1=ww^;q^x33C$a+b
zjrEtVP2b%+rPccl<3HzDaJMjO4L#HUT33-Eus8FqnXmt31SXBYlgsHluF}4c#T!vf
z5yiq_B-zZon4sgci7X=*q`5LR{l*+*cz};lO#<hYaA7lFu8GC--d4$JrzmXD3ti*`
z5j=V*Ju(f^(Y=K+m^wnhdsP+ZXy5O;!+NMgA+8>T`_zs0!u!c!d!dxpd)zDn=c!zi
z<hJ3PfRy0v=Fm%`<ZgJ7VTSY)`Rkh^NxCGQt-WE)D~f|jot^Gj{%z~FX_?#}k_evS
z1TOL@YYr^apcG<>c@&ELNxXel^balKdVTvZmwf`V?Bq}bNfLBx?1o8n@lJS)6Kj`R
z+P!XSvn&1H5|`IQk25xLIfum**cLXIe9~<-`f$M1cY4NCQ2N~b*xfeKI`4IoZ^rQ|
zupMBOBqSiurBGuWZ-5dS!b4~u*XgAPUe$Hrx5ns*$(z_b^U=b?^PWfWNdR>$v9#uf
z3lkyIa!fZhJkg}qvxaqAm17tMS>s2Gm)^)anEWbEbf$Q;RKEemjf98{3V2@XuqU0_
z<2>EHN$`agzaMdeJ_JL2fs3HzFwurhm5fg&BgOEjR-nDhFx>=?q~`X`1!r(ryaR12
z2gwU)+f!)JT%D%0&xqO9;;$W!ThGT9^{`{kWNT^W?4kOPsgyJiqoZ#_-pG+HcEc^A
z(;=VDfsxs~3*6gDW=x%e`1`gwqokpm8-wR&LN*SsDA=fr(anp+$u&Qiy&%|9UKN%f
z;yY(XC5#L1K>(@=+7k+2Y@SqS;YUl3QFE<9q3sjkS%1(NLc;k#lRha(BKC%hrq&$2
zsMCcXH1EeZ$Tv{d9(7;3*zO(c3@xB*HT5;^1&i}e{3>OG@T>#3loL@zhA8_6MmKbj
zR#eNZieTF%T4wI6w|tEvsG1b$Xu;9ECAr!INzpV)GGnaaJYq-t74<)$SAQMA59}oI
zws2~ejE@Ic`i=TcTv52OU&2q*<>)QE`TI5Xnr2>AZl6y#JPL#NjVs0E-wo%mW`Rl`
z{EB3M>HM6Rw_dJ_Xsy-|{nvRr??6v-hdF_yEa6UkUeQ)c%6Ex-qwSpcl*@c4pAYLT
z?rHiae-a)3oK^R>ybw*h$C_rGxJ`h6#P@elojlPyk(*`Uf`;t5I`1R*i9qEf=51eU
zJlQ4hM(4Qnq(?ax<kl-nKPTjdUJ0q%Xf}jC3^EBRH~P!o?d?p6>`-Wmxq9Vs3E9sC
z*lrk><!@n2s6Dspg*~O#B@0cLq+l$jJ0m~xA2l*AP-dRzxs0$@E~nbfdXM;)$En<a
z{CWm#66~m7E#&nidEnyxX%f)X@W8qR11yos3jaL((lp%Ra0B*C;cDyp2xf^$@JvUQ
z3VuwhUMIcEvy3Eh+}YeBM|k4TH7{_%aAJFQy$BIlM0r`e9xCdY-BAcU$Z}{Fau|(9
zg+<bSW#kb1Rzy6KHqD+wyOiAg+>Fymo);(lBzhy^85AbBD!NP~a`88@B-T9o=JH!c
zqo|f6o1dgSimyMe@}7Lj9}nwV`jFF`%oUe~Q|M-))hT{!fs33HoWSzQ@LSiLd|e!p
z*EehuV22q5f|DIR3^$`Nz3B*<h*oT#VpZ+|+<7e}%I|c_f;g^SYI2k~MX?Du&b=Ui
zuIB+|EY23SFZH8y$rRTWngH$`kCwCvk=}No-aQb5J=&8LOsyL~^<!$-3vaJil1mo4
z$b6tzDq@tAbt?3(<LiBOocgVb38uGgdPN4rQaqWs(fY{L3HNDezRv%wF089|iN%|i
zvz`2N#;M=rzMU>cFwO$=!B53S78z=90NQ-TI6+KxCLgR5Q^4y6pLa$N$%yK5nn9sb
zE%&`pzg+*EAn6LW)XBAA@2A6<KwfjQK~ro7dN>P*s5P{R2pCj8_1yX8)my%o%v8aX
zZDVWopRMaNnu&PWBR5i8Bz!VFIbV(1zX4ULniOv!<Und_eN9t?Zo->7-vvNF1+|c!
zgMDTQA8|pv+p+$2YLfbAKlrZ8#umo<+o`KVIb>1Y!_F{_gGtK1Td=e1J9D`@?0V66
zy&%VSH4-do*xp${rJ**(*H<r~Ai1wmsUZ+=nQ_#o`ab&M^4fb}{?YzPLXkqG%3F$d
zl}Be<Yt>QCtmqI(BwS=D{peg|;uQgv*ZC=^xFoQly2L^rxoGnieZ!1<@OY8Y2=dql
zp751ecBOfsoqcgAIGZ?4HK}kDN)yf*%A!J9J=pg<WKduTjMGt5@{>2>$1HB)@fY-a
zdH+Jh+n{dLvm_|exsRGjnQhi;lCOw&(M92WYvY)hl|-y$&wwHNmh(r{BvWZ>0o8>=
zC0+Yt9Noo9Lq}c^)@#c0(T+IcbwLNXWfAw-@gljSaZ7Q(e5I3=z;-d&niC&t+reIP
z!J=`dq+Zq>z4N_JzbQ@<gW7>0|2Bo8E$=x;uX)&%K<Dl!VPOk;hXe~XFm)M-+k<^D
z?^ePne{5@5Xa1(>prx7W-$%~}6>-h!vL-(B#lMw$a5!#vREnpY+zl({Pp{k_me8rP
zSc{~1O$v#i<YvjcplAv)XPza~c3U*t?jks`js@?SZyrPHTh#K5Ox}kG6-e~(W_kow
z9jGbPaSS9J+yT;R7D{fmB=nMT9BPGUNs2zIAZjz73p0zb!$Hnf&cqiY<4Mch?Axoi
z^@8>vy=hmApMk-tB41S+^^zpw=?R`XoNgMvxe(mQhpp?gXn4FsZkP1=`{@vPfea<c
z-+8MFqKnZngUyl%q+2hK+jFFdX4~TMMUHDF7sDsONDMs6zLr$npF`0YFcO%*Vs;e=
z`=}Xp8xXd2*E`469Ew^dqlc0pQi#Kc*r`XAK;9(QnRO1l&ruE}G4iIuBRQ%hL=kh>
zGc_yYYa>&|7)DoesK?I++3qU{?IS4hgh&l~n<^JQrK6LOVtXKfZfLoT5O^3fm}7B_
z#1h)_OtwXj(75!x2raZK`+`@-TOK_q=@q@%V=Og<9GEhg9X|6Hr<;`@c(uRZ%jQ`U
zX^m^_KVDf9BXQ60ktI3JR?sl6<{kIRxl<&Y6e<b$+EsM^zIT)4&T&yVy;5h_eIM<S
zI97eLtzvnWZ6xi$)D1{`teC`aIX-L2^T_};vd{~{|1N+x_nFHTiBA)2#d%cCsj|%K
z%xvFrmOjcNTtOkNv9C{~yo>XYZ>@`)QjCr*LYzva(COnbb<f=vVnA^7&987LANMC7
zaNnm8{Z4`Z(Y9o47{#KC92f)&i2JKUSMJ$hCl?xdVQj)R`1w|X9a_^Jtn(t__SFG_
zXcn|BS1@*RP%F<Xgyr6ksAmWT-N&*cNE|Ec1s9R{L_MG_`)E5Q7TmiAtn>ZpqX8EP
zSFCN!t;{0#%8&2zdcM7quYJpmPeg|T`t1YJ9<0v`le%Nw-d|wam;9#2>^<72ICTBp
zB85D5vhtQ;jrNK!@gVh*;bF}P-9_rQVzf?;?~!?f6qPX!|2-*stSH1Xa6(Re&FJEK
z0q68^$nWbXzMTecPTR#|=p&D0R=uovh33NN&Xh)AvioV=W!LAL#G(VJ0A8;q9nAjp
zYoI4z9jP^*KVSG(UZ1qs8g@6v^JAAIA*Iri9`A{S3p8kc_!x}sXPUCf30?)p2VXV~
z*2&aq*09y6kYk6G#1-(7o`bHz^|~S_uU30Kht$0dfBbS3Kx(6M=uDgY4t;v>U-sM|
zOXvFc9sV}!n7WN;Jp=v|i_rEehrwlqunmeA;a2Wgj_{(DR+i4mTPJU3pK8VhjPe8j
zqvp$Bip>#5h-&wrurEqdG0zuX(r|A0xBO(@+NmWytAgro`c*&l4aApC6z~$QC5jH<
z?k!Y~*6vieB~e3ve$V4SxNWBXXK&z^M3|?2&AOBHr*Hq~?!I7(I+SFyWNXQLnu&_S
zM4vcZ4&rUhOz`4Se^7_ms-=z*vJ&6+Tj+ADlc`6Lr_NKQ1`kvLgO53J%%$OosWPi4
zc4b*Bdm5hlxRIzTr7G>VA@7$~b|&V6qYM;L@M{Mfv;+1VnF8`Oj$ZR8r>hMTsjUzP
zPsqpcl4__Sa;MgIH{{GT(0jc4gkj2r9mup^Xj|54OI%^g)7^06M8}yOh04^MZHxO$
z{vpTb9r`-cbh<6x!Tfjag)=&d65zFGbNfDQT|VUQG_W`<P3>)crc7s|54P1Q9sa##
z3pcL>1PhiHk&4IP>PdbdGN+@foicaPL%(F|{W#2RFB)RylG_lKG|7%nnxm!&7f9{>
z*%&9USveN<6_Z>-^k-ZBu>3x~faJAV`IHOQSu{#-^G$0G`3RiBHv{U`foaUQIA^ng
z2McJ+fPkZB94tD6C<w!GM+ez9wQq4qP`Jw=qCDgG^em&^7C&h-TZEn6DNtEb2vXkK
zEr~eN(PR>JszKnG7BhNAW~5gNHDZ$1^DQY5OF3NPujELB7?0u(ij@>;H0iHxrSvII
zJ#CcQhK1K{ODGR^iBfUL8^l&DZm#Z1!YP6FYECqP6&<&aS27B-ZAI6RG{`hE3P4}^
zOvRS&|3w)v2IT!5#a&(%g_ETAtU6C3q?MT{20C@oA!bKqj_Yv6FeYoAC}8lAf;z-|
zqWLHDz8FMd)I%JV*|uV;L@NsH(j9j6KkzNF+!8nsT%sLm4Fm>qoiOL7Vf$#Hd*^k|
zbzQ!isc2Hc_s0&TJ8VeVou7Q>^&mUq7zH-BIN|g!Iv46oUyj^KjA+e^qTx;~GtOD*
zJNI_QX4VT2{O)S?H8*PxzIdv<xI<O@ZfZqmd%Ifc=+0JM)`0Cs<W6j)xLTo*|LJ0Q
zht8kpQjVN}LTLbRmyeRr5*8UNVa#20LLk#~ue`+m$ly=6dg7eun>l0wQ9A9eb4U}O
z#Qh!ISx9XgO*Et~ABdm)*?qaX^3Ns+dc{2oLKzTl(PTFJ9oMNazv;rrb)jkcC7c`>
z5{eVIQrp7zo6?IDQk7}44OQ_TZlL-7NIdUNth?>}>X8McHed-QGNpTWlPPaZqqgb6
z^<scymhXbjQvJCsoO@nmGm$5~Jw3Bs-@E<dx&%U{eXVK99`Hv!t5Lx<6{iEYM*ni6
zn^oHiIcr9}lC+vDZTsT8h3&@c0l&>QCPJJnhb7WS!pDuAsiZ8m5$b=v(&S2&vjO@s
z>X#w83ryuYcOK@CRg;)-52joneeu1G6wJzM$0K#E;?(1(Hag>RoaG6T)Uh~TX1@5i
zAhbJ3T78M;s_r1}+$Ia)=Fww2uen?jYs&>yvr*$isRq+~GYxW@)X2;HFqA0b2uW5=
z@x3QQsWx5N21LFw&N3ewi{6^?3XY%9aV?yXRP#V6>CR`Pt8M^Yl<k;yo;3%6dH!6c
zYvdh>^9SGHwnZl-T?+mpPH&;OL(ToHKB+CE@EKamEl7a-cj@sK^_YmVfr;3J8S$<B
zV3MHs&AF&DMO^Tf!}LM9a^S!Mu^5f3eewX4AS>8hO#XyNgYzi(*SB!tnra-SMp|I&
zt*w>Z*Gh!8v<pW4tj>L~3Sh!(m|(04oW5%Ox_e`s+GGTh9HNN43pFlapAw?G&uUL>
zOE~Tgl8o`@K4&`&t{-V8B)+mg%xNIMlNbqVoV3&dY`@jx-#kf$xFd^A%=`<~q&!|Y
zSX|z5ng5>cc<79OU<G$A@7B1RiG5O%ce!#Wxzm|rZP5GGu-C-*awOrFyH%NdU9@*9
zamw=|erhDLHrB2vZ~3Fb5{J|HCFNtkFR5lhY{Up)5nA7npUZ8~bJ_*ra{ePZnG)gx
zcc)h&<XV3eDcBcBX@bmxRW^ah1}n8!w2d-kA-^M?QF}4a=Qxr*x!V92zh6iKE`}U@
z7z?ToiAolZc~l6&z`7?B{YPgr`5CNz=-VC7AvT<|*FA6%wTL~+-4+c5aV3}LqG2zU
z^-D|*Lr9b)xzkMzyH634WBV<K^pkjWi>r{vxD&>Cp=R7Y<;M8Pu6`uEw5G{cP<gU$
z64^{~q+4@=)J-+hZ)X<4dC|U1RNGz#`Nc?dW#YnB&6q?``yF`#)HL=vkYNy<ic!!f
zWakhCtT+-5Gk8i~0x&{w0;+9#;K3qKl@<{RVRuSD&4=&{6{_-75{&LzRhlwWwRC0$
zX02dW2ZXv{as4~$0TGO2;3l{JS}@L>h4DG528YP?HJB~)tu@OkyF10L4^z7_Fy^h{
z7!(vl8`m1OmZiXUbd}E=l)MU)OU~8AXT_jlxbYq@RYDqMNM4{59ka}H7C3yMPC;?V
z@@l=1)u9+QF3TPM7D8U}<-jyfCXKfzQ?q4Ee?cEnbsE!IA&={kZzr@a0x@+Nham+3
z5)DSzim0j1voHVP+1GyRdy$-W7X;ZNYef*>yzO4L$*7|B2kv(zkRQ6^cg0UKtz*0q
zo#gab)AcqTvV)W9uY+%tyK;~o!wc{`;@tMojw_oWjF=7Dci3)ReL735J}*{OAB(B`
z%M&H#pJ(foO~wid?$VIN!W!XhW5XrGkkE;^xI<>FrpQr|t?ypCvF4CN*N=eBu9@~z
zk&qSI5#hAF>ReI$oWG6+ajfgby<Rt}IWu}f3HLj*+GO6(6B3#^qU9)Zh?EJ(U&A|J
z*0m&*Lj!V=r1!D>abu;i=L)%eaNdf^icIflt>O$cLY0qVH{uRRCb*4%0c)}SITUdS
zLDXCJb%#Xh@%7M#$)CG^4h8%Ln>->>b$uR3eOTwV=sAP(EQg*ZnA#tS!RZcefC*5;
zrMsdq5$X(Bvo_y3?Z4{9phc);UQ4XS4#OWhyIA%Qkq93b;;8$nt9#KAnCyFwr|E?P
zQW>aSkj~`d8z{!&{t@&h_)&V$NFzR)NX1~DZ)|Nb3cQeYros^yyin6UAe(FPX4UbD
zDzrjKkz%}W+&NiYA(G0ceAvLS%yD>%=XlU@2wB;lzdkJ9XJK&Pdr^_`fyV@V++9jc
z+2R0m!!ZMVE2WP8rE&xFP=}b-$PSj}zXv9>S`K%K@9Z^-!*3Udnez|H^O|0*VdD|X
zOSqw*v+cifCEqStIGhw3d)}dau*`AL%aiK_DS-z<#)DnVJNqAw80o%gS8sn*xZpgI
z&9CL`Zs$Zlw;i%;klXglQO|TTpWzyAm4iw@y8`~cy9j}yfCbU}p@6<<4O6Ngtw@*O
z7{|s*^La{Tgs_rpQ3ET`q%T12-4p@k7AwJ+Cd^cr{_J3+e?*xIIs};;1BDOt`~OvY
zrWlCa6y}p*%|SgGNX~H=i<2{#SQzMG{lb06;&H_!uTG&Tq;V?N#H?|`dTLv_a)=0K
z9ivI%rbHibkAJrPP)5Ifg#<=Kf99Zl+zb`w@ksU<iZz|GM78FDlT$HfckCY($g?&p
zVA}t=c6I@56G(o_FG$V@5j(gj(>8Eu6NtQP{fOl7&ytaeCcmJQkabs2{0^dp=S7i+
z!<bVWv+nrjcapivsx*>+r*fO*xNpH^d)X1&+UuycFTC25-zrTidsSF<8X2UiZd`!T
zgHZ5`mvoYAm)bUjA-#l9sWTEj5=tXAi!N@FOspJL{Dlat?cqmP9Dk$2<t{L<GeRJ)
z@0{jrC5Jt~V1!<4BM}x(^Z`k1B-{g|(f_DkW2aqrdk2tr^uJ+hNoH-TWJ(&<+lVm-
zjO1U}XUb&Z5qC`yw&QQyuuIhES?cdQd?2Gs6Yw6D^)-IC3aAGf6Lgnfe?Avjj|31?
z%C3w~moN*aK*o>K1YbYO2fh^=CT_J!@1(5;^pMD_{#|}Z4w8hrn9S}AWw{lrFH*C?
za|^*cL=HvbIUC!)`IB6qd-VZ1r(w>$xKiM$(66UFO;oCq+wuk~_Q-o#tg1^OcRho`
zJ5bDdHYm2I+7kKsp2RgxGzt3qRy?l&E5vso?@2(lvdK<-->1*@eD~=g?*`+~)JHvS
zP*tcncsq=fOMTa1=n`NZ6g5~#pasJ6ul-VS5MKxiVc~ASzAWvnRdzg3xe5I%w8?g)
z!pRX^(d$Iw=67iPIIjS%13P;}TEDm%5pCY)4XO5<Rp7XsTN7g@l1>To0yC)&(16g3
zQQI5p^-4yK`y~$9lgdU!<aG20Xbxan%Uy)=;0EmxQpkQqeB>B?Cc1Q7*k=*d@)>*O
z!n}++EwsSmWYL%2q<~u4C^f}YmmVo4IP)8g(B<R}e5`mPnVj_%Zo~n9Eh}l^^1I8^
z!2!qiQV`HfS=N9`|9Hs1`q556&VR1oepxBFmUum7;+lfoijr7v4*~I!m0fZZa|qGR
zS^m81h=TNwvgaUBKb=Ol$-%E%Vhsx@&)I{k9KKX~W;DVbgM{=nX{EeM4wgyYlqF(r
z;cU<d&a6uRCNJEHwg#MXo^25cI8{ZLs~6<Ld_nT41s|d;d=cI2jO|FFepsQNhTEsY
zI_2|jE}3)84!~Ah)%(Ip`Cmx{Sy~#M=08RC-*(M1Ji-!XwuPaJzP+ZXkBivQ+1x~7
z!cCA!=4LSK@Iw71UpwU{_lTPvq08<{<==UmOiR0SY%@bz|Lne`!Po6O5-P<zZh$eF
z$pK>X6djhbS<7O(CK9bSIaVcJcH1tibSHR%-BM1ElA6##jpz0%F3Lb9b|-qLZCdD3
zdJsFgQXE@<=ma%6CE?;y#Mv5dMHi?eonvhF_VCueunldB-tcv$GJhHVK1cik@a2Z4
zw2S&=&6^E?3llsSSLvRaahZ{o=-;Z~i4>7iVKg|?M0+L4i>uIS3iNWgYFT__2vNmT
zVbBUNa!@mVNFCBufyAP8IOd5?5w4LqyY)+4h~AcV#+(YB<Bp(96l+UlyEn#3TQoD&
z{f+mfw*7F`-L0-mo8c4TH8dQU|8TJSQ#}@c8g7d6XQLvnWwz~A!;@i)M%`a)n6iEK
zDXK$-{5_s)zk&ruu1b~}`KO;$U#`>0Zwc3&tXx>Fz2!CC^Q<MC6W}*{xT8<gSj3k6
zRONX+FywFSR`OaEJCL|=H%qIqZwv5MQj`r<HXr0ya}^jS7*g5}=<g1^EsQ%<e=Rj<
z%D+=za{IW+xL}fTX=p-iP#WuV7o{XlH;n)3`KaspDfcFQqH^1l^82ABeaOn^`JrC3
z=?L!=yFE}j%K6Y*Bz2xVl{($F5H)_!>I{9zhx32~I`^=<?{S#Akf&5Z!s1j7(Jz-V
z5Id;k-bvYsk>aKZpH{{c>3<7ols3pS`G2|F!!G?taPXAphp8?=r>;UyPzpiRryp~X
zJuJ)~6R}XONl|3mFbp!TY@a(}Q)E!aJ}iCB{~rs0p83y$v`0dOnW7l512q))M-vn3
z)S-RZo5SU#YI$~kRps=UIoFYfX-#JgTqdqukh%r7L`4<IVv2agWoJwc^z`T0i7rEv
z!uiz4Q|~A}sAJBi#IOowno5Q=24Zzlg6LOLpcR~*%;oHCygm$~I?rTqvgvStJ&5kJ
zvgK+YY>1ro2GZzLl)kgZP@USzQ!MoRdF4}?hFd>yt{XXHcSvHdB?`mfc%oC4iT3n~
z+A35HpKLBZR{L2hy!~VE@*X{3jqx>X2G|2B31x;t;QaO;M#{GA%@)yueA?~VO`)g|
z=u`gY7Y8O=CCCaM;?P^*oX1%PT(;EJP0-W76{Rj8|E~7=n93E2^F-rC3Cs)ZJzaMN
z`s~V4XO~mw$k5Dd8AByk)i{d>iGg2foc>MaMsD@j(^ZouFz9Wa64~*k;f*o0ben-A
z%C#Z2fVUblezwoz&UFt;5E+ODc+H<S)!k3qwiKpSN@gjac6a7;etjIyyg0Al1Wari
zfRoU!b=2s1-A4gpsUu!k)To^=*|z`n8^w!#O^#W%0jGzLn?FPiAAK=^KMv6+%Pk)5
zz%#w;#CyW`u71-&%S^jhjtYvi+RmxjA1%rMuLf)Cj2Lp-B<yi!@t;NmyKCgUgvU^X
zb@0rsZHDHggOGSI9%BzFAY`ca9ftB*j+MNa9c|x7Qh&*lulQJ!iM;%Cr0~&on@h!>
zB@yIZHIy+9nN^CwBSyeD1pXIA)b5%=qS>{3C1@!x5ecHml2lKRQ<18zyl}(;KR!Ns
zy^+gymbK_G^0aA(*J$Q0hm*Ie&va^Dnm#1BxtgAB2AHB#x(1^Rmk>+`p-@md?OHVZ
z^rj|stBXnkD1UEF^%`1EaLsN4vmo~#9Tj_OVw})c^7H8{)J2&=9*rvJ#EV(kYzV+!
z5M$j>PV}<?x4qeOe)$%TDwu{}<|XY9{eiy11mC>XV)9RXSbnl^#n+?gORIjmm-wM<
z;<e-C0+QDr%<1CIre)m|$=wi2Bx1iGEE4br%S$>GwT~4oNwl?nVa+nSeGTajKx@}F
zQAbherB~wj@d0{;#Av`<6xS@-fI`PWz>S_~$mXu}$hjl{!8=onmSp8Yaz)h?)&Dsh
zV4ds!?hfEh{H2sk^Kp#_T~_Z(c(5L5>!>EpEPM)`*B+^?&aIB}`i%q)1e}E4ooAA0
zL;|>|{-yNM9mXM=m5D5x7YxQvSi--7P`?=QG+F=o<f^?M?0_0y6}FQze?W1hNXtp^
zRR@NZ{P7WAPk%&`kZo&QIdaUP@kvgc@Z08NbMwWTV7~1(U^6?T@@i^Zad;j0(M)LC
zQ`vMjL#TZX^H_ktEIxiZjfZNyJJ`=m+u<kMlAj|n&LLXtb)6qaU}4jHcu(=}f|)b^
zDCU-T!s5(@QeT7`^5+yCFKh#Jp6vIGj;E@x(2lUDUx8lii{~KQWXNp+3pH;8jtHl4
z=Tc4hLgG9u)IRq54Q0{tFX5z(LWju+R!>UelcRW->Sr5z9E3K+#31`tYra0fcP~Y1
zP@uGH!g8X=F#UGtt>YpaO&+l;Y16)iPIfcaV#YYq_v3xv2w&J9(H_6yfM?WRK8gLA
zzMOpa+fSN9uIxsye=gqRQQhv!4QcQHJ)8e6n!AaVq><%Qd6q1~1WD@~y&Rho4{fz3
zEZzuF*4O%5c!>>JrtQtW+MH}N_$EHA-Y`|EQ<_}#=bHuT5k%il$aivcyv`viR!1aj
zTO4<^g2GwokL6kH0f>tRz!SVI-^uInz?&?l7Q1j2Lfqcn@LLRRkl2Btj<{o2fr&S3
zW&d4MU9QK{@J=G#Pm(}=H$iVAUflK4yiuvn^W2lnAvUJBiBjJbKH#_fkl*L=uHxfR
z&JoAVisEDnQ3<ezvT|uNfy0OG;%s>!j;=Y&nNi43>?h2|$!`vUwbrmVq%jjxFQxj6
z&LWd2pAvr>n+h7Tu}6~>g`3w118bBP9Uj7Cdj$>Se~e}q@eX_fbiD=}2Re#%3)=sS
zvbPS4a_`>0B_uWtQc^<=A`Q~rASLLQk{(LBySs*z4hcawN_Xebjev9v9Yc3K7yI7c
z_x(P{^S;mT{Zl!PiEF;G*1690S@N4#r0ShAg|7(W!PD(>`mTGbZVeOg=LRyAT>UB#
znx5F>Pt#1T><O`!?F%x|r#C?Ja)#xY_wntFmNK3G@&I5g`A7#=Hj*TNiw<P*%b?g0
zUr~dB96Dk0by|(!h$>^c7w!TCe>PDvtMzAjDrxsq82CDWQzIB@iU|*1b74=(5_y+~
zVVPJb+keC-`TPaN_{-J5Qc^J9*}9SEt40&_FpMdA=ju<uT#_n!I+)Jl&oj&A<-@lo
zu1Kg%td7G?X>OO%n?)>p`0M*rD8*;N`?rE49?%Hvi7KZ!G%9J!Qm=(wWtmsfBjKm3
zLVr-v$HvLs;REyCv)sV;uoxzh6-+Q*`uP_QpFZdL%82ES{rC%0+L&1e$YZuS?FX>*
za8nQ`5n{ygy)M+A0j{_tRp#+`%3W+ds99`yR@9nTB1@z{(*&unLmqc!A-zpARj#H*
z#@1&$U>r3#FgXRqClY#p=1B5uT;}@Nm#!yKd^_>$gZ9H%Y+Q<l{CXM1cB0s8`2vb-
z2-BI@X*^w{$LzM#Rv4UM7+2&_qGvN1(4iymr6Hb(*gn#uYG<*FQOjk$Kd#jve<E7T
z5c>e_v@pN48M<ko6b6kruVW_Nd3v4WDR(0&v*_s4<z1bWytwYm!nW?@O&T!)j~I7J
z50A)}Ulf%&@V#ivP_U$aC!;t$VM-&w04?u~L+@AbaikDxnA}`%toqhg{%mF5xuE)@
zd;RV<%aP#Q$edNnd%urPp<`~JbKqQ~Js&6W2>Mh|RS{NOq`FeNgxlkzwA!QblSn%X
zHTx4VTwopweDj!z?`8jCpPH}ZraODs4{-wkk(7b)qj1ezMLu!57X4unXvt~9K1LC`
zpC_9R{aahb$6yVWz`rNyWXt~hj{5IQ?ssMMcV_(fw<$^_h37(%HapGu_fWCxt#2Wg
z9{x^*AcjB<t-~oU$8|=gi@vCAj*be>p*&6o7LrifB}w=)c{nqhkx_bxO}Ku}^<+^d
z!-%mU1l6OK%VY64#EGP7J=a3^S(BQ6gYqJVs9jtUsJqNaWicQ#JU1Vn$$-jCpQr#5
zl#qh+V8Q80X6)>PC@vX$DS(M#&_qVZRKPhuM>=?dxUE+f1BFYj(sERy5>hb4XTrFf
zL<G`B*k*|}KdfNg0Bs$wCH{ybec`o?LSu29%_|FnUzL}?n0SS7!?4Hy2_inj3}&{E
zOUWT_7ds5VbEj^9n@yW_+Xt2_`>ZeVCUE&Ft;wK@J^n9Hv=k-FI-s3rwz3$|>)|(p
ze1MB2q)9@k=PQv(mwKVy7@`VCfjjQD0RpgtxILD38}=cbz3h)3E}|)mO?)u3c(0kz
z@N|T;9x54K=$o_bi>97YO%WEnSk=ssC$cyG5VTzU?T?$jP62*Zb{?1+o`hNMK!1pu
z9=+b<s*kB*wX+_&&H^|EFF9!qCdr)zg-BP)3z!uK@%KM!zcWuhdecK|s^@#Pp3Tww
zbIWIu8Ig8wNH94nek`&RE6Eucr+8ERbajHo;OaF~#8?TQARM3R+s_rgRXZx;eIFW#
zK|i!53J}2Z#om&@*>=wx^d)Dd%|{4(y(f<}wl17+%k4IXrpFMB<T!*h5Xb8-)>En^
zS@6}w$N*7JVy;ta%gD_F!nZ{HrbFUBI9zQB8SVIZj?~*R_F6dyk2$@80k266Hx*T@
zhhvMn|Ct^wqC2r%CXE53xqdihT6gK&E>otx)NZmt{;-5|C`3Uws0qEEw1b+00~F_q
zDU_l@Sk<nrG2K^k*wu%I$7v|S_YOqMgGMhNr$Bqk{_-(b%@V-wOBOIVPDo)TUdr1J
z-~3lO@8x6lzc>j3r+*CQ{!i=tKmTbsK>tJ$MMH`?ZvG=78dfFictx9TKUDZ&OMjx@
z%G=z58W}h0*1aPjwa1tbLZkqSQ#ZrJnsAp51U(T9-lSEN*jjN)hS&+l8HESEI`qBo
zHDJMkmoxAZ7J{(@G!>WWMAKf3y`G@EdSZ7hCZl7?nv*}MfU=E-C9)Dg0|8;E>Qy|~
za&9h02yLTBH5~jLi7Va9{i{XZ!-3{IC|85y8_jK@{+m$urGEIx^_8jbv;d)I@q$Db
zt;&}q8;58iE;&lm>*Om{^22GyxHz5QMIjLff37o%=1!hTt6q*I1Z%q$lMC{mlb(e9
zQI^%OfnVQU#dbx{?zj)U@kBekJ_cm|u|hoVtV#97M}l)`Khs`&DFFT2DLD>YmML#O
z<&c2Z3+Y0-B4NGCXO~VLmf;x>v)%8<6T(`h+b5xeM2&uE^~%4MGYF7AXl%<7W4b7Y
zvnlD7npH>LQkv?fSSo35`0|VQFqukdLoBz4lg0X1c1Ln;cy~MAr|$@y*fmYBHsc0d
zMuv68(4Dg2?*m-2SV2MJ*w*ULiKAT7koN|vl=WHC=lQ<seiri5v+{ZedX|6fLNo<R
z68T|7zoZQfX1ww2#@zM02WY788=NI@mo6|{+?{2YKAX>U{o9Tu@F)dRHo9~F-xtOI
zeoW3B5VQ%ebdwrbyAyPIF+jFs7~G@RFCC!O0MN~eUt^pPq5~A9*VER5j}F2v27c$C
zF+srL%crU>^ZGvv?tg#n|JN0&681#C+wqjY02rM-XV>|KMlGC@KB`Bwi6Hcp-oIxP
zD1l869DmJ+>hw1nJY=le{_B<7=pI);!Q^V%e;@4s)AzSWWp6qJ=DA;%b>g1noAA@I
zKJwb`wh&oviyrh~JLxYG^bxI>-5j$&?0k0n2&O4s2E?cS{Z2JbkEv9arhl*R|M)xp
z`y-ZNBfF2@0F$k(`5VS!k|V&`IRDwXNRQv>i2}iY`DXv8-{=4NbhYdrKjI=IJ_7GQ
zum11P@c(OFpn>p%TaiGrzklk#U*$FG{ocs^px+|G_kx9^=i;B2{!<Ecdhim7Vyr|M
zS-p0;8G{b#|NBq^o}abQ<44_&WF@5h&v)33X#<Gcw&M3kvHKG__otprjEk+f7?}X+
z<@~JFZ(+$B0J_Wq#M4w41=%J?9@@ivn!8^8`{R_k*4u$r_X+^ncQXgHUDe@agj7H`
zWyCJq@oIWRHn^_L5t{~gKMj6mES`dum{mf&ehuiGiC-m(3%b~B#f~;g$Lo;W^7fmo
z>++p{dJeQ2?pxLXw0msTQ(zL6tT3<*m{xMG2F+&yGkaJc!n8$ZK_9Kzo$vqpQ1^)f
zK({tI>(urCG*iy_V@N~`Jlvk_BOdM%Oy7ioQPq#th|JrK0RURaHLz%SjMG8seb`<D
zaA@F48TTi82+tIsi^ip_{vNg4F0~n;$J9Lt67m5kg~8~&lXro?yN*ht{{rAx6W*48
zinWt>EIr(gJbZtBOJGS60OaUXkE!t-@O}?`C=%&20H-PA0@ifduh}wQ(y;TgaU<Po
zuco&7hn~m!Clfgu2OCLL&@GkMUio7SaVJ3G8L7_}0`TVV8H?<^=H~VuJvM`Nv{VNG
zOK`}(9=e$kmyvT6Z9ZdJl$FeJ<|`F0dVjgLFaB^ME;|2+!n~QPZCMw(EArhb^32(9
zsQ-0pTZ}j0(Kzf<V)EG0wSQjQ+MHWk0)*lly2Z!!&)3UeV(N7(@Sr^s`1S#KWT3iO
ztT&t<z4Zv7cqL@HuSST;t=&A<ORI9ZAeBG&xsE9v{s+sRH4Hekk~=xF|J~dwC;fzQ
zf<T_q$N*SLIF6{O@xWlWSyizs_%}BR!g^p3+1>Ntx}5v%oSM1$4L(<ZJ~umL>stA(
zEaM7Ye?sgiLTzU0ZedB#64jU(nHqrG)cX0gW9cs!v??cb&EY~w6`mC{yEUy*?HfFt
z8SL9VoZD%h@DjL%JYk437v0q+A{=}C!`#U2;K+=Mts@cMRa}@z*CD88_V|%c=zADM
zGlR*LY4~u6_h9%Kg1*1`MDPHXe<H<?I_7s%>sJYod3qiW^Z08HvjN4eA5%rpw&@QI
zH9MDLUGjo$R(FB?5lh9AwW~#Gn(H^h03-*8ZW%vd>!P~UXDDJUxD}``?-7@}hivAE
zSrPoQ_N*<&qesMbKV0QJ`~rzD0(`@&HK_2g?{<o>C4m;2Xyx;DuB~p07jIzxuD!4T
z@!L-Es>cmRmD%C<aN)<_geo;7dO9k-zf)XF9b?$?Yv<f>l+N(~esZ?`MGKrfe+Kc<
z{rmUXkbaT}Gq_(jxQ>ri^VxZu#p)2HJKuq6!$|OigE!3rz~J1F%V3MhFYM}H%xX~I
zgT$mX0x*sqpXs@9>xnS#3#>JFIjKX}%hMw9g_i4|rJi+n7fVZhL{Xk`#|r4TH^6YS
zgW#CzLoh!`a83nZA=%xztC8;5A6=xa87c~Nf1EG5H)L*x%lTPE_^G*f5LpA74RGHg
zckUmlhxf&LPUc(lV)!j9iX*474j}Fp5x)RO=kuF2@q3d-0LGWYA*@bWafyEgJWr2&
zgXga=i`QiQ050Z;e&k^-uo*OI(D+U-@VRJXn(y}QY9Gab#UrSja?|wGWz3yd^f8&R
ziWM<U*O8vvf{Sg3fhrAk^3fhYc*PD_`O&i8XuofaITqxHQ?&=UNrb^owLxXhZC}m@
zjL^rO)8erYFyt<zMtY=m@dCaQ>6R+<OF{GWZGttzU;X#SDf>U!OJ@KM+B^{vT&yYC
zA=+{~jafdr2OK^4m^&YTVY|56k{6%72QaI1eJ@Qn4i8}Q@qb6KhnGo3BHgwg1G>qQ
zt2ZR_{w7?sAkl}b5rdRWz?`gSxOB>)XVYPTGfD<8iIKQ0KB?CBbvrY1KkE+K4--43
zWny1ze#Q}QMQbt{QLV7~vB?Cm>%FJXAu#{XTKZ0n{3v7W1L(M*yX~C&Z(OR`n!W(Y
z@fz*3!PO^&@<)={iCBy6%CAfRdEoNWt$k(vleAPb>$1o1R{lb$1vqJ~KUxT$p{N}f
zH($9`OFnNDnJU)*!KYfey7`LI8fI+|6q)QqS={Gy0-i8ElW-m>DjviTJK;N0=ie4D
z-!1vXnt^96033@K`{sV<=2>N}@3*ths52SMuKbDYLoq1cl*=xh7V_Ja@Zv<!cl`6b
ztyv$h!Sa534Wi%Ipa3%dy+ASv$18U;PvE-y7#H>N7cltS-NOsmh30-oq<%&m9(_u;
z?k!@SDDn(WBAnT!Ek`kSbw&XZoIR*c1xwUkdt+9R<ty>aAaVOr<$7$8p(BG?RCrqF
zWRCmGCO{;|mcFW<*58$%kkYvAHn^2#vVF8wgr;;nwrQT_CoaKPd=OqaJ^*O(6!RLm
z0!;XgC11AhzDX$$Uc)#7J@p`Do){4w4@t>=xQu<U=kTD^cF(6+usB|fxilsC8(6h7
zlSnf=0w@o{pLkuxgB+3{cW@1RDHiVvUVb0IgHz=XUf=v?9_OeVYTh3xbPJGNXPe*M
z4=vrscvaqp7+gLTUrfC8JJ|GPm;ZUy$CcO$V0;maXh-*ockN%VfT~M}Es#pr)z&_8
z3Wqi^M=C3WU#au+SFp|wbN$r{T#Lu&DqdYqFBcEB+HQcmZgu*o?R=(3Y|0gwagl~^
z-^J4tGA<`5_e>O;^1k2y{CRMzk^D6@*NyZC5f7FJ^od?@E9KK35gGvc+X4#EY-4W+
zL2KnsiD24Iyo-Pz(y_FL(rM|-jqVZK>r3gLF*U%XMss*NZtz)N5ncXIJZ)ud8H1Q_
zwYtILO{fOI-TL;8%l7?vCyfukNf3wk3f5xjXYnnLM-j)sJ~OiHvRrvVj^E(}UtmeY
z;#tY&Sztq_`ux_8*Zdmw;|*Vb&!JCY@Q>-6j|_Rxas-Ib75@U573S9h4vw@f)(WVH
z1wKnRi_@h2<l?MSNgWgXi>2i9oT2IRzJN~u3&5&6UXTnq%{&3Zq@E$7?hd0gtQDJ_
zE34i*jKQ}|pda1<825sR<FV&$76_LL7t77FG+?JQp=6rIG$#7(nP+j*xJUJ}RK9G!
z4*=jM)!MfKP${kDsaz%pr==Y00XpK|U&SSk6iY6pUV4w(^JdLkk)4a7SkM7LV`COG
zKzI_IxSl$uL=&Jpl$v=AzAbGyt|u?e0sJ<q{e`zcD|7Eq>eHMPlf>27L}dX7qwq`s
zkS8XdMNAU7$J%jK+j7P>%`xb<eUpkX5#KDJ+Q>b@7|D10J5c$7E&Pga^YIGbKR)T=
zwVN?L30OD>wAfeHnD{h(!Y8(SoHI6!B6r3Y<|^jlhIWrXngGMsuy*z%;DDSiD>bCK
z_Ip4)`~u=C&FOr6Lft^0L~p~lQk%vcg|=gI2aYL?Rn!qy{{*=isE-j|LSQ81pF`N%
z07K%kv-fOvx@BtSECva~gtD>p^)G5nzeo7qeh7zmZn+7OG(XXvSRiGh7WT5TKi3nf
z<Ib|9EQEJY>gm#+8Gj``RG<UrfU#!X4V!Sp9ma#bCWPFFw1%3vejGk{oOUf-koQPO
z43#7aL=RQy{SeorKp#pJjYP)E^?6c?N&pDf_VS|GkMp6lAW7;~ErsKDDU=@J9M@UP
zAz5G7TCr_KHOC+`RdH0LXcP`09G<tpBWVgNBYW<9;pXRdh}$mIY<1*C&_RPH-kU>a
zsr5DBvt~aKge+6n*2&c`i^g@QCQk;7V|W{4uVA&ihvcr?*m-3T8Se;KEKzO(6YV3*
z8i@mEc&9(Cb+tAEX_hNcZM#UY06l32F5q+Sd@-PtD_J8n4*S~)Wspj<Wgoza)@G_O
z`uwq)UO|qm;<2>#ydWdy<v^9wY#UHIJ+N^AW|^Gl=W)pdq6;O<vSDW?t#kEInz$3~
zCjiqGCpgPA3b&qTue|u^iX?sANx?$R_5MU*NysKDp|Rh6s<U&c!qY8HNPWZS7orlw
zuS<;v^(gs#icrCoRiPY&UY~XRS#R8LltUB<IcrIqmY+5^<VCy13$=Tx3T-po0Be2A
z2joL?mJz}?-fnlg3&i1<!`;=gXFpRt8)YxO9&VNpSU85kN7M^LGrRo0Yy7WgZoR;_
z<>A4qG6wmr9%ytlcLI?1N7pcPBof2fD6K(-MorI1S@OXd0IX3ge&!nGp_ZI}qXwc*
z2yGd{|54Rl{ZE!vE$uJ14G<|`iM&0^!qJMWB_ZD6jm9KnVtr}SRle28Kj35y3y$s*
z!A*E*3S{#yM|R@H2x6@pp^xr^ik@pB5ae$qj(b$gNC2Z!Jwb3~+z~#R4%0o5C(Pe(
z#oqVslUtt}&i7=79WmAT({@k>ULz~wcl;qEzNj`2Ft_uY0i1bylruLxOS*$|y8T{;
zs`K2wk1^#`4CbBTn%eg!?xd{8Fe=xV+q|b{f-%?>+Yh)u=J`i!I*d4W=w)IEdH@Q<
zsOQiA8y&qy04vLM&a)VC_f(hy{)KhMI5o&j(){tzIbf_eX7L@1(ue#FuroTTq_`u}
zprmLL2b2>PzU#5!UvZ*I0x&tf>rgRd(sL>6*g^WnU=@ef``y-qj3MqmsyT932p=zx
zrmc;&ZFkScIX`egs=hO;1o?^FsMsW!&UaX%7c&lQ1UUn}OV*%Cqs`8ZKVZ>IWfz8<
zo##4OHRgMNY-w9Fu}PwTt`*ovqFZOp6}vsRd#Fjgvsg>Qsqvmdn;9LFOYqze7a+IG
z^xXMo+t?27(;~QedX96t4A{n+cDe?%J)&;@IV6-opJf;RE13p7`4+i+qCo0!`cBXL
zs9RC4v>@c1_*+uIOH1UI)$8{SAY*01_@T-IszXIsYk{Kcc%TB|wyoKFSzD#_Nidr?
z6M15S>jopiSps&tV$|*SW7Tt<3r{gLYPhbdEL-N)>1Kp<K~BMn<~Y3JB<(sB3|Ext
zEWq?8^x(17^iI$WE9i3}v}`+(&D3csr8!&T=4R;E&HkK#EukpGHiCXN79U4lN^+hU
zPOpeIOzaRhm5I)V?|$v_&UxHenotoi62!~MDRaMKAoyb(h5FE6riH)4KLvW4#S+0y
z;wjiK++Lvni2#^6!sW+3!$4l%KlgUY6#g2>C4IotZA$v{l)CZy3D_?csZkQKITAV7
zN&9iWGppH8OBsx-D*4Iv$S{)cCqjR))1+Oq(`hFVaP%!FRV^n$o}F%*y1y3M5o(4r
zawu{qfsbby=?FbMnIAF{x8D}>6w#&`q%(V#9=S6u%EIXKY|j{|a<U;Ju09|-6=Vu>
zp5_mu#IslFGTF-9ulXujjRAN9b6;iNZqZ|XRG+%4_%S8jwsmXNCA^wh%GDbj5_rEx
z*I=&i^JkGXS|=(vs-MnM>pg<~EfdL*`EGp$p49@W5qkvoj!>-&O|t~W6y1Ad`D^S`
zvWA?Rmr{=7##z%pXnp7qskI`I;B()1aw`QP`voC#h@er;E%sBqH4>Y5hxgw<#hRW1
zX9)$BnpYg@up;;R3QxpV3gksjG%?0>!gn57Tva+)<9*?ee3rJ=x;3a%*0+$!sWo4l
zw1yu~qk>Cl@s3fmAj@iW&T25eXf}PIgmP1-E_XBhsw?K7W8_TU|BC#so-e9SzsSJg
zolJ|Gj>LtSr|6p2=|aZ3AS~s$VArHpcC&wWdGkb>t)C;w24)(KPUgp{sXO5DbWI(m
zM7u`ES*6NpU@8`g6XlT2_od-rv@o|FvjZ(aPmzhLP9?hj2VdDG8Fgb^I4Hv5M&}RD
zH^D@u19YNH1pa0%`Vg@kO%puhlFuq-mOvV1OGIl|Ti4Ate=tQHk0A`rQpG*0k@5QL
zp(>(q*P&HRj<Q&rT6*_SJFw;z*j2|`XHxMVzHS|orMb9HbVe8UNS$tOx=8>_S&gIZ
z77Ug9wDZl2?mbi;0lA6(_M_5Q&s;-hLrs%86(+i8gc-ea{o&q>I*O`XY0JD!@8|%4
zdym{u&XU;>H>1?hwY!LlxlrjM@p<>hI?_^YQQ34{2v+rmy<O?KEw>hv30Io^o^w8H
z#ZmjyB!jlOAg5mZ+{fltsj52J5YKFpz0@;ny}dG+Jg}(O7gLwVm3(h(3lD~37;&HP
zJ*re~I5!&VSgQG5aI5QyU#QyN*nB6L5!vF-;`X;d#DKBf&G=tpAQ0#=uHaSR!^N=;
zS>eeY!5&_qTceptTLXU7zTym`>W#GUk14u|&R-U`6?^<H;ptBwa=AIQ1KroJ*tW^^
zm`IGBU&smvy#>2|L^t#pXnlJ?X2R(!`JTa>8IBT!A|r#L^my6!&#|-&G$w9p_Yepv
z24RX9B(ffdZn^yMRcmQg7vqW?RJt6n%#rP&r9_R(4Hm~{oi*JHC5?uW>FqcUBZhP0
zoZ1>>M<GbtT(KU}(^r%&_XX^u)LL+(qi-ZTqA)QkUA?9BTX2?;>r7u!V$^lsLW`pC
z29uogz8RaHT`yvw#tGYOl9@yeF6sO-w(6-f>V<9tasnfkVO%1u<WC!PQ79Kww*G_J
zTMn&m8ekBRkl?Q5W${Frf*4@LaCuI0wNPZW)dRAWMzbS#|0a(Q6jJ(%B@=^V+O?^%
zGa=y8+@J&jj^yLP&8Wt$=On1xPOl|XB&|)$(KN&gf&dJVrjORf#nAX1kfIz@qS<29
z$7MER%nIHsUz(KX+&9?no<3zAl}M*~d6FzYjDo5ZzPPokp_RIPmn8q&?gx_xU9b({
z-;=%NRQ)od{|**zGUjHNyb1cnGd0SUhQ*i`;@*30&5#~e@biqj_rVJ2uTGrgKC5D(
znL9c2*g>UuKjV(l9&<zuC!f6&aly<;5cws{RnV2?ECDs1pob?_6K)U>cK#CR0e|R?
z2Ev$VpER02q7MmSKfV<h2O-hy6~#F;1m6ZyQP7Zn;#$@EahzJD&zAncm7Q4BUob&D
z50*_Ux2&m&l?3d{N*+dbM)y2cC%osBeC<*ibIMYl<v2X-{`URzNi|$}D|ms_42|po
zqgW0U6fP<LT&_|cvj$w^;Fz=|tr6+nOHdO*a>bX2Oy22|9E0<VimSHa{zB|<=d-4q
zDx)xSRaIeI%<~s}r^j8h4g1sZKY!W;TZ8t9oEaJ3c$~S-@j#kq*(c6Ka2d)3@A?3>
zBGS9p%02;IE;gkq_;-@5`j{jO4zOuZ{VPc>>n3=C=?D4L#~6&|`d4z|;65;l_(0N?
zLfYGqPNjwGRH7nH6U1H)!fTJ|D~BRGbP`}4NSiKpd#v68>7`0e*RB@kl80tuJEynA
zq!mFie0GUxb23P*T`O5p-e-k-1rPg(q#M<b=MS`2K$OV;;2a%<pqu8RVCglMF8iWq
zQnkt9wcSBPzu%(7`P^tL^lKn`1IadC#S1JRy$&At;Xwog<$j=Nzv`c^!x!X54Xb*H
z-KZ)jlIMrfQsh;^0<`=0Q+9V#k#ozzOfuM!dQ#+TF7cj?$yiR0%Nj4^)1l^uNxYD{
zSnHOX!bkj4t+)X{{1edb?xm~Unh?~j+n5+(QvSNfrZJo4=3a=6HE>z)8KK)k?WEC_
zs9lyJmwz3Xg&QWq)FF^BP`xRc3%FK-#i^n)oHLs*(pX=PyK0kVFs3CTNmdTD;c7o$
z5P7%37OQ-jPwnyrDK>Mf;CHAJ6Dx6`n@p=AnPe1QkG*0SEx;!>8&wonP8RIBffY3z
zo9f0@qZ<l&w_*4L({H_T?MC1+i!7T<GV$T-9zO^JdMgJiC6i(ze{eCJ7{bcp4yIoB
zwn7_D`6L$sK*x+>54eGyBMlzofs*0b0Ku+UtZ?}LuauBer+9ZJZ^mL^{%PZD?H)xG
zS5;MTeU`|Nl|OA*uslllD^Gwq5-uqe4OUGv3JN!tHjp8soRiGWx}L}3M<-X~lT5B`
zy@<G%jlp7xuHpPb8;f6t%6HSsOJn0>wYOo({^*^_pQE(=EVkz;sqLcNr!!2WmXqjD
zFCK?QdP%86Yj5GGPpq5t_!z@8tmh|>I@eA9O-Ur8wWk{YD?;+W%tY28RA7J-{{%bN
z1nT_TFMYWYr=~iYd{j7=kF;xpl-d_Wiro;-MTu`T{LFM3X;6Dka>l<CLd%|V%aqR%
zAS1vCSZvAvbz5O^F=YqnYzk<vQn01-Nd?za$6`l;>Fa?Rs?NcG^^uKIsI@S6WNwqm
zPLya&u-ALnU%VAoRuR5Y?%BB(cDXI-J6y4xhRRP65Q9;YEs+dmMGQeeV<pm(v&?Sf
z&H7WS2qQYbRBVx<I*j7AfPL)|&*VP(MD=O2Zf#g=e9M4>Dc!9gyWaMAi53y2W48_G
z7wRqdb-<7Nw10K-tr{=6{PCA>YWpwv_sjU>J(RnIIF<ll`IsvJk^W|{Cl2o(fvfk!
zuZAEW_~Q{GU$M>gIb<w&$-|Tzr!v%g$t)hRSIv4BDJ^{h2$$E5=FQOf5_HimNqsyE
zem!K|CLXM7R6W2xb(4x+TDWF^`~&B%-deaw<QqrV1f!mdRFt2-@9`yXh$&U9$V*Qj
zf|U_Cc|l9_%U04cO&6)rKOd#y=H{Olp|{lZI8j~Y)>&i;wc19bymX1=o?=)HO7gB0
zHh;>mlby-SoO^p6D^leejI(>@w<3t*V^pvh^ux^0B2EmM@vs*89}%64Zl_-Y#MGmf
z3x_blj%)8mflWo`soQgSIV0J_kkLf{{AwaEtj}cWbIThB4@9YKKPT0vSD$)rgB(YZ
zt&0wZSB^JD*4udm@q~PP;Fx^!JZIlj&<aDy5SxXzU9tbfXE<RJziI(dg6?Cg!h`F6
z_4M_nL<Z7+8w#HFuyC3Z=LqD!r)eF)<G%7p*Nd0;I4K8|?ABk?Y!p2}l5I=0(!0bw
z;!>!Bk?l)B?sLyp5xN;vJ_-=`ZG;g`Cy}S0NUOGNbY}SMkx<xVZcYq?_NB8eL^uZh
z%pydvZ$#3|zpLVKZahT2_T{)^@wsd2&{F7nLQ~9zSEOtf(aQ!dwFMi;t@+<ccRlGz
zEoV6SK=9IYlQ=0R{4+QqsjsoAPdF}ix~dMb-d<Owl$x(WGq6r?b5QnJP=jsi^ygjS
zYj@hn4SK|L-_7TjsB5N0z4gNrJ$*9)t0mSp8SI^>xj}MyP6_aKCgolealv+<<7w~-
zxt6=>E7ZyEppKY&Jc^>nBnSI;KzmVR5Z~A=&|+sitB{eO=B+V2yS3&aQ}n&YFk>PD
z6-u<GV-1-!w@2D#&(-H%5(MDWD%8E{?PNtJ#huABgz<k_14F~?b^qcst_~b*s<Eal
z{*(G-ie)yGue?D8%PXX*n!=OphR}slID*f}ZB_E7-SyKtAiKo&C4PH=EII;n(;9H|
z^*G%uhjK?%X0(8rUN1SOgK&X4p-fa;6sv~5vW+|AqFb+L&6o><PERA6veTN1Ri3sV
zNI#d6I^Ct~D*c)`GovsPSHn+yl_~yp7oF1He~o)He`tAzv`TX1hR$ZsI+Or&X!nWh
z*7*BWl%Ny^TU(``nt4L!JSYY~6_WL!%2=8en+ZIhW9T?}>ke&=9j0;e$2rC|m52Q+
zkMcdAOcxN0wu&DnebQTFGe`L#Hpgy-5*;tIn#r_-LGjO(D)~rrL1PYAko&ExzoC0W
zbYI7$d03b)xPbr71UK_V8Q`1ZJT*xg48{ci#Lwcykj!i!on|DHq^&b5>(cs?@35d1
zqOP=fH@~GaU#xOT@8ejH2(23NJ+HKt>9oK}!&)OjdlhXEH}ElE80`so>zBpUC(ROR
zy3QA&?LVTTQ!)LPznVP1&YK&(FeA9vT5cD>_(f!_!fb*s&*|Nt|A|PJGtwBp8$)p5
zvx3mq39{K2O#9=fA;{4-WbH3VchKN(b}UF&V%<SA0OJL_0|Rp`u|YM<SRHxw@NxD}
z=}mABBff=pTZf@+V<3JbIV9Ru)nk0PtwN~*!;iH_{XnR&U{}2rx@hE}A@6X?K6oa_
zy*cF3C-Y;ApB}jVgvL6I7p=CQ&oTrzvBrHL8<~81%|Mc&69_<;LVVL?2hG%#skyx4
zgG0{6(u!0+7c0=q7+%k7BLgsMBoft++V2fwa%*|Cac{TZ(r#DuhGiG{j3gu^jp=+}
z_26TYXhBRYtb1)R<7E$c7^+7(7%72lq;Gqp=6-NGu>GsnY-j?CyCLz2b<ya2)Yee>
z73A@Cx1!b;dT)}`%g(!)#KtzC#I0`8q{de95l9ro^Iw7EEz<;MQhV!?-FWc67z93X
z0i-wz$jk8#fv_tainLUO#Mg24jG=NIhq6N)TBrjG#cC=O4@rWiXkrn4&l)hEkkjM!
z0h<J$vc$}?_IR5am@kTr4nG&fB4ub;GTPz(zG=oj6~Ut_u%7hR?CC{4SpoJ&P;~bT
zFZ)4P9GF=g^EE)qL2)O|NIwph?v*e^*2YApqhM~YW*q#JJeH6o;F)|q_lUmIF^v1l
z{|!<^3}3<oo_0t*&l)4&fn-ZTG$xj0u%M466YaxtmwSeN5_$ps#$Hn2qbg|PwTfci
zAL&86U#_+R<N+UmsGyg$z>sQbVg8xoExr?(G&j_SfM)j<>xr&B#!X|{x8BMmd0qhH
zyJ*zOdXB5W=W^HG5|*8N0PqKoh8WB$s2eco*o3gDP};FHxHGcgv}IS;Ep@+72M-)z
z+wC@>Ex_h;&ZP{3Z3rkww|R%j(fJ|*1*MwfYn=eOY|R^m>}cl|k>7a+9Xk)Y$>P(N
z>d%@WmlIXKR%usx^BiDsV@o~Bf_BZowFjS8rJx9wWxM{MlMnJ(q>zI9EdSjz8?roS
zaF^t8QrN5MkeT^Wzrpo(+^q6+ZtzNMJ?$9<1&>qT?0M#iq_#A~cW!PmnM1{g5bqkP
zA_p3f)sW(M$a-l~Iw<3}LyXv~@PnK8TvO81=vf*?2SAOiWkti11WO@Q!vb**`Z<P9
z1<GYS{<JQiYIq?h(gscQovxi^n8?+tuhNJ`4*uy<ww6Ji<oX?z1HpM4n)GL}aegMX
z#^;IHSP}WN_~a(HT5aBb)ndG9ZqtY>r|QLtNKLfXf`v;KTZd<*(K3o?KMS4)5MR`T
zOqyt(vZ{$;UR+QH6)D!!{Q4G>$l$29+WR(UECIuBNI-{lQ`EOKS{hc&fZ~-9|1;KV
zvTHH+c%AG?{P?)XW|k(}1%gh{U7N^tD<?XKWwnn~hW$`vKp=}_=<Zuc;5VUmL^VnO
z4}^jN`TFhGKgdrt8Dc!U<6%B@<B>FM<HVu#hNcp)>7p->wCR?w)#-1u8D}GMAgPMn
z`krYTH1Mp>U9fzzg}F<De07(@lyoq?@Ygx+gu+LSND+-9HsicqeTG|qI65?6R3I-h
z|1S%clYY;i5#_Uj>&)Z&b~TMVsPdHuK%uR9#PKbbvoHJ`b`DP`l-Q8m=hFm8`l4W|
zZ+hd3k{;)zi|c(>DpRbUQP@^`iMbjeHX}~Lwss!p#hwW@m?RGdF(Q&B#j=0_M3s4G
zH+1Ao&?g6pfbNXX+Bj>WswRe;LT@7g)nj-{BX9q`f&}DPHmt@bAUy2T!yfrcRLkuo
z&!a!+P!Lb2V8#4xl~xAaCC`I{{dc{B71ePm@97c4szgE&3(W{mn2|F5YA}t9JAlys
zt{`x+omfRae8c3CbF%b-<pKW1yqhaPu)=iJ@V2zu+b>0A_|iMs#B94#w(H1y)Q)R&
zDz9#YM<)4ow62a+&I>68LQJN<#B}kVlfi)U5PLYKQ`-PXYtz}~B(MEsUR~HGd3=M$
z5~at^%1)R9gQ83#gWUlOG~;`q>wOYBL@h|cG|T81hk(&LyH5mH80|zD%GG_4OGGoi
zaNt_Yz%l0AWH8+%g<2R)CBuM3W96SKB7c1u){q-R>FdpM6yB9$AbQ4(Y&H(AQq&{N
zg&jR3cX~Ph@m`CMrhs>IWyU1->~w(T%gCJDtR<89YmyI^=tWts>C@`RSs_B~8uG?s
z1w+BU!Hhj0j+r<X7sb+~Cw#w|4|aOw^o9sV0>dnKWLO9ZV7FOx)HSxHb)mkaXeo<a
z2oa}9Vvf|cz+Q-=3|L&h?z^8f^mLuK3TrQS#u=_Xzx%bxfN?`DDMVFB!N&%tk#ao*
z2YQ_Uqfa-(?_<8hIP5dSxS1U^ImV>A-(=A<WFfjCcDR6QbYql%OLVW`crZmKbFVA#
z%KX3%g9X`|(4p5QRpCv9V{Xpp!L7$SF8`~wydTZLI4HH_)Xnvmgr8E%!xe0$ZN%Pv
zU>Fjl@{Ww->kXCEV-Rr(_23W~tzlkbopESGxuH5kYjO|WEoPgclntg?=RHLkB(UVs
z3tY$%<n=67OTCDCw?rN+l&^AO%b#4NlH)d^Z)rvMhiB{J)gcBytY1Fu8H|^%orxea
zQK4rsj2mI*V@xmxeuBtXqMC_Br;0r=qx>?opa?%Mz#&nj5KZJ`XR={ei8>!}R9?P@
zmLs3Te`lXV5l@s4fByprkO9@)nczOPZQEk5fGQ;-B)|j1Pg$Zm%EK0=8v5%HvmQFD
zaJ&8Da|q6QRNH)QTWopYQ>5OvY-6f#`QI6S-{wBxDTf(WrU#3cF|p~%8DJ}tEv>hI
zRul-d$jyS;X4<VgEn1o2--pvrV~SPZxGX|gsj?k+>2Hl#8L(%jq?$hxC^`x3V1;t_
z|4;&jmv03dB~<(@G2{PeOj+pT8sTcxK#*87Zw?CfAD6$JIC|ZJKjn#MNoTvC?b;dF
z6kHvsQnI}@Q9#d@=vsRtML3c&F8dc-=XcF4vf?Zb01p3IMO$)eFR{sgyqa_RVlVGv
zgOR9uvGNxHs{a{5LUN+NyW45j9ilHAmaBUmd>&bz)$LDPPz}gN4kllxp(kk$m4|(F
zm<^vFGk!8p;ly?B!ejTfsM@&v9H3`(XwU>~TGlqfm=S*{FBtR`GWh%9R?7miRdn|+
zfi5~1=HuuTxDPM@e*NrN_{1BAJuujp9D_oEW@!k0dL*@2+RxxggXN&In3oyXX;!2&
z^9nR#snYngt`{pYmN(S8EYBb@1*}H2ZOkg|R~yc$X82t#GYVsyku8qj{EMXiS>}@`
zT*#EW*KHpN193BkK(A^xh(%DoNl|YQW;DD;)>RdUyzr=T6b@>7w?5FEa}rJKMgQF0
zY0n!Xi~)bQ!_qb8xQ7yt?D#Q4kb$N*FD?}bYisguEpnpoxa;37ZlbOKtIxCvC{Gq!
z3+)jpXjhTKfX4bNq~65k*Hj#)FIMn`se?JK`DVsaOAQAM7b9CBCTa${`I(@731m|u
zoZwdqQzBves6O5)jq~07JIUKDtsE3RACr3M&i-KicEgsz+<>A%sD({n{Zoi<azL^u
zMIhisk5Xe^(@l8lpy`<WATzDL)Z&LK_F=0|e*uotYsE!Z;Yv?nta5|CDba;aA0WtJ
zlpD@zqNVlz5%FEE%u9+4LAY@cX93joKiFm9TWu?q>%J+enh;?^M4F`Snylw5qrm?R
z&pmD3QsfgbRML}yt4}I^!^bdJq-N?Lunv?N(7}g@pT6)Bb&-+32T$2;5b$NMdHcH#
z6h_X!-vn$YzfXMZM@5aRpH_q5x9vok7GUMk<4676lj`23z?u|_wB7kNg$EuZTXca(
zsUe5C-}E1G<_^M?z*T#!dpw!zo3QIWdZn^({{SRP9W*Qj0kT(0K{jdZ7)e5l?MypB
zgJckyM4C}5f)G4KmJLwju|nmu(WA5*Yh6xHl!LVRobisH|5?QN0K=j@3lKjv*Id@Y
zLg*(Q$~U+7PX0AOrf;D+d7~ZNI!;enAg;i+_ij^2W+M5GT-T3$LY>;X=Of1N_JOVG
z{&S3fNCX~!go~u=?TRU!&a8&L?mqsNlv(SxlZ?MOdb(IO`sZ*8?d=$hc6~LbU_b$8
zjE_Z-ByVL@QtX}FEC%bCZ-N~Q4Ve@eWedRmB61M}q~z(ap~%<}QrF>$d!4A&D-V*3
zF=8PxW`mu(OmrnW<FSzIg=5wvgRWJc#=)b4(OGh_(3a=YcIF)h+f=dLx_r3)v{Al#
zaPQ;vn0?)d!=8>O*|<j<PwFLl7&eS!M*d0nc_n~-7l*y$;O;BxBLe@yuATf{zG(m!
zfP-?3bG1#lp&xgiina2skVItN9#5_5bmZ(q%n{lMj`k!BO~Mb49LrAGn&g8F2-)au
z8@;~f)nMesqdkLb{QK3jqd0w-o}QCpM;t#eUUi6N2DlVJ8~u7U_^U=iWCcij_`Fl|
zrKXEqkh|Na)}CScnjp@_T(QZ5$@kZ)qG#`F-8jt;2w~eZ4liPJT_8a3_SvE=hM&9P
zZeo+Aqk6)h2J=_cu3X~*KfYuO;{gLk>OA*MXm(|0i$vC$BOC(_(XO7}JcCSW>sNvr
zIps^|Fl1;&5>|dDp3g81bRn(S?Hy~?=m&&A6E0e#X-??(kfc#p@!=bfwfr@*V`VBd
z<@W4zR7`J+jK9{>rbChHk4oH{t!4o$T6#l(Ap81LYUxwctZZ4oSYYiA;U54t-4Wvb
z5@lU*|Mtw~@o^*~Zi_a@1@m0Fi9|Y5dehjd08b`GDzlSM&Qrk0oEP%lJIOK#gxikj
zq`5Dfb5eZJ_u60A*Bh{xZsAGJg-ZA5Fbb{Cc0O7>o%3G#2*V*TMen=^`_NIKad=5m
z`?DRbekE7^2%3gCMeA>04mIG!pK~`QWd5foX$DHr7S|3*QKSzR!wzG;Opzi3JCFr9
zREh^5_CVRNH<iJHgmtmEZ(E`e?oBI#_cP!6q}p^a^pASQ(JY{OE5e~gIu(+HYG%RA
z3o|C8regv0^sm)EcEywj^l>vV9mB@34Gk!4Io%zy>P~2|2t<N%=?vD0^0bufZ`9l|
zQj|BiM3RuT2}}&cTD~^*xQ!nI9js#mltF(m3<_nrlbE1`-HqVipMG%Ead=Skd@%qt
zc}`gtudVQ}$Klv_p;%V%P1a=Vo<H}1m1RuyH*&RYES)h4w_r{jABIT=*SsO5(30q|
z?^GX~Jfrk${wyP~*gZzaxDtC?PET%veOa~@!qr%!X&60*i$ad2@T*OE4rYbAPcafu
zWe4S1(Q|q3PI`1g4W0ceI&g$Li)N(t`6GdScu>VW;H7QOdeb97QU{9<;VxT#QO2%a
zDkSL%P|A|XO=UcclNcUce9+WN^3}dNxp)|2Z)y=@x5Zs?YMAFP=GlRKCFkj|L#<W^
zOmjM+M-jOVtewd%bgg=bP<(x{5x4L+_<1|KhMHJM?8Q7OF($u+)Hf~Tmwt$`u4!KE
z>K(mx6L5^ar#4}=Y51Oz1kAF(`^dy$QfJ2|!yd(G8B{PRfZQx=@Tcc1Pz>vybZW?+
z+myCBFR21hNgCMNpZBvLe183%;Woa<hC8{^B7E4?1VyOYb~Bn@U^Qa%=vE!d0({Bb
zgM6&55oWvVD2-$`JQ%qko*iJ9ENmKwU*L!OlH(~Bgf(4Yn*1@Y)I!GM2n@~{xz3Jx
zWH~?T0R!}Ln0`KxW9{tL6vo~7DATpirC|bc2GdNZG5AtC@-k8_?ezF?x$rG(GIwAg
z_RdMx$%yz;AF}jD@zouCq+a+p3!r|<bg%k_B^;0zPfr~O*XaV~Szr&*U&2mH0UvnS
z2)SAOSnAT4TP+_USE&i*lkCc19I8PF9f}vb=F{ni{H|g^*#vPWz1yiz=Tzq5I^`aD
zMYYbwu;{TXqlbHS0bIs=X!w#hvle31nJPuaw1XM#&+t6iZWaML_Vy;6017d4;&HOu
zVCh1ynK>`p9He7CTaWc%KcVS0i0c0?0Ar7p-6duKpRzzCts~;9F7p?q3xiXC))G{A
zp?A>O<ep|~FXo(Oxx&U)Cf|k|yK8jH)rJE_%QdROA0GVK>{dp;-I<ovvvPMWw_dCO
zt4IqRxLK#0Q^g@dC6$CA;(KHA7C|WTWN{gd-`-0Mbyb7^7?)2ek^1`6&$#mAu}^$^
zW;0hHFO#VDd!AVPRV<UN*^SXidN2K_YS-Pf5h&Mq3w;|W$uT9+#N*FCYg$hluegVw
z?9{Ef{Yz78y9U`)=@<7Y{CCOV32N2AkjyYA)tYho<t#8))i~WE13X*UH^UI31F3mz
zgQ;TqC3DF`UID()F_1Cjjvofs0=_){G!b<b-wPjtt5Jx+F^D@e&2`Qz{W_a`Eqi_}
zqgNhmCj##D@9Tx=g!8dslU?=YMSP3Ew@7Em!1gV~EU|P;FH$c+op3+`qkLYPKo0J+
zx7F$8XA92&*Md=6wzbeE7#83pOYn6VoJ3sSj!r(E<mWF2u6ns9I<qtq+QPCp^@YXj
zsoHet%hcfYD3VPT;pZx-923Z4efFk6Tlv0nFppZ8!t=tXB&{6di;;jGr3;rReo#jL
z)b02dFpKH-1m9Y8Al&=f%(&yypiC>}5pUF-MPeQfUW@a~lD*om$`+ZvrYpq;JBBUM
zQ3-9p9PC1<PSQy_m80|$E8?b6di=TVBBG<g`N<K7=5(COPm%1%d1AN|xh7;vPCSvJ
zRrQ<V4_DA!?C8RLqb60$T-e&@eSCyw>-C?dnvXvE!c`Jl805vN=Rl>1<}}L=ri*q@
zl6x~f-lUs{{i!&cvxjF}x+gu|(7WgEw5DHVDb*UW%fM=e;QYeAHWQt)qfJZFD(#(D
zLSK>}O$1d65FG&?^2%n_DWk8I+!((AGNpPVY3p^(gMO({`#=hCY<+^3a<ZkTk${v7
zM2IlQ9R|{VuEtp&b;@|23TaMbb_CRm<o*&~Ji?u3s=}G2Uw_FR-zTtlVfy5PJFjVU
zck5?4swg9eMWO7FV$zRuD7$Q|2D;K%iT=cDM6RaNSYGNh+bM*Z&4(*=_yZr*bXO*=
zN4B;0x{Q)kNG_@vOz++@dBt2&Q|oh$c1nB}DHgmt$+kb}D%#;Q%W10HU!k>8SLHN6
z!8+cs`}~^5W90DFn+sD&&8EnvAQ%x@QhMxk<%I219lbbwhi>zs2QwYR4;&jT=y5(|
z=Q~k#xwTK0#fHEwX(U%Wr1)NqVa5$_6oXiIxi|~jkW0pXRtr~>5vHLZ4GXUl^cdk@
zmu#HMi;Q*a4I<H2#*!%P)<;CSV{T8to=iU^VF@)J_<<v?S~<EHkFN{t7>A-+b{gJ^
zlm5PHVUqrWq9?C`IY&{jAqCl2hl0#bpOQJ3OmSb9wl3mTbqqEDntrr2+1jL2{*>3c
zI&8uNkUmT6hGs$?!uaH_1ahtjt_o=&!}Nro`x2+qPu~`VR2U6{@S?%S^O2{+4Q-d<
zN672ZDkjOCi-iU)K%0bTP~@QD?9BRI6W<5rke>trKsg9XV>-@6cb^fC?E3RhP4cyB
zF(Xw5wJ<SUQ`^c}ZdWmxBu<;b!+I#+FwSc+K0#PW)eo`UOxX_(6m6`7n?H#35MT5o
z6n4&0Y;?B&;Hy^^rG1FDYP5bK6w)@zU)4zKph`z`8RDvvK{(dEt9M7{s<a_8Xz7A_
z4vE$cl@_<)`O&;eT*|qlV6)ja)$<|xJnDx{yz9UXyl<%I*NimDr<T7@&mCeXlI%`2
zHtu-KwQnOg<$oFzeWo@<H-?V=d`O!4JXbE;0(ksha^T&wujZx*^I`5)b;;^|;;$MO
znj;phAtQjPP<(6*{9Xs8l7zd|Vn#rCx6Qx4or#d?J+HsYbNSefh5iBbFo=NjI!J6C
zfM&4DoQww`)V~;Kl4MfPeDSCz2(3>CL&|wvf5;n76wr(p%1&%d6OTdWDC&%NsdWMI
zz@R~9)r~3_Cn$U5(X8Xs&UDc$Tw~Y~WzS!~coxD-zK7?Cy`p7AQxZ~P>=VTM?D_}0
zQFN%j@nD$BAC1eS_hl@@|K?=hKX%sL4S+QM-KM3C@85)iguZ4&Lh_-ycpQ`A>UoZ)
zzQ-5KqI`;h)YCJ}6;3c$Zl{0#nnCo_1#RSSIBQ@C2pbuHnse|+h~a3TOy{u=t7~GR
z28cp(M@s7!wv|TuS3PtruG_a)O&P+K%83*jjE;$q66cz5eS+@;(lQO03^}4*8kyG3
ziGzWrj9#!jt)NEp!oFaI8!C6&hpyn$i-x`fU&o2J2^KX2MuRl>5cTYecCrvI^8Du>
zbR1+z;nMm<h15L~IF?q)64|VZ_&*NG2tZ^YvKSDV>zDeo+q`K|!?8Yw%?XB_4d=^{
z<KV9^@ce8KVTQ5`x%v_v<4`Al-%%2-!m_e>oWA2jh!CtjAH{7)mzhk4IbUPPGo(F9
zmU#c8OCaVpZ48+Yr~XM;ZfH%^q~&rtH0|hkQ@fXiHBi0uD-qdJxRU(+b%(|PNQRPC
z#K@n5j_d?<RRw9@kx?Ds-p_QDsQna3Tl?yD#a9pliAYMhSj1hQy16OlEjs8bS6I6$
zU$3I9UFt=H0MW6<39Am2k&Ks?vZ9VlY19ut^K~xOD7xgc=X8^AceI#Iwgv(JN}+1f
zmG-HZCXJjotFdI&k28(w#)<>;w@(WAckuQp139{$C=A4<>qbSH7%Db=)jbtHHNux5
zFC8r!C;7T|N<p@(_*Us(5n027f0JgyiS}sVyqaIu7wb*<C)p3P&Lr2~(Kn}S7*R#|
z*POTJ0hu1De1s6~LJ<(WVzkqo<eG6ju7}F{qkml6B}i%E0ls^c_!J!`d8*607=*Jn
z{BIFO(vXGVehZ77UV>zs5CeP~dukaSrX%GCYYG*M6z3C?kTYMSGZ9fxJVUE}FEXGX
zMA;rH3_BlO-z45ZnIR!_<3T|7IF&%VL?zidlOvK6?R=dg(O0s}OgY{ac2c0h?;+kB
zxN-Sp`G#s+iPhlSn5<YJAshVpb$UkQ+M0Th>K9q3_g}o=kvet_rsK-9&8Fa1Kx>`v
z*(*pKie(D`b%RNER1L&*MsVxVphB`;yO+4adu6Kxv(HWRHd3vl6VdO9jT}W~QY4f=
zV#<)r+R@aPpMCI%sR^ON(X0A;+RbI=!KzcLKvVO&EKgM;LE4YUpqnK0K153~OUtH%
zhL2Cdw*aX*_(zP|HKvF0g`ldzl19oHU9k8@lXYZ4Y5NA$=b2;=_hy1TzZU?Ueb@T&
zPrsl}vS9J?7`=e{+wU#m-bA4)Q*1^-6nsx+$>HV6{%aXaw`Mshu4mF|5V7J`t8Q`l
zEZJ4+88u8KiJ<0)opndRnJ{{e;ujePmt(SFl;Ohiqd&m|%?1r!+WMCd+-Gl~%4IJZ
zpSOJF`%{nBwB<Uof5{%4FM(8zf_ONb_4fgxg^Lu8MW(Jb$sUrmcZMq!z09i2BUs$0
zw_fRO9`GC)J<)4*Rx$f9RDlmfm>eMl`&>;J?Ojky0<N!i2z>hF9zQyiw;t1kG3u@h
z6UsU)9lna!2d8K`v9@$CV0HHmhak(butJdgBe?UU6*<seVZRqD8N>PHJxsWR;bGK)
zCq|YPks^NllUJLAYsU-y$ho`SiL8uP(QrB7Sy*mf)biLxwk#rm$}&}6XFqI{S$Fm+
zT)1V^Q_|i3srg~yuss`Au=Jt)*Qm2$w|-(o;Frm>NUva%mB@zniw~jhLxElD$=X?H
z*-TVq#^2n?8Zl!flb0qL7b8dqT0$a9T;*=7t7fT<rW#e@&g`?px@434{vyd}2k$-L
z&iDql`CFFu`S)*HUtsN^t;ZN%?s5BWTyEU4?y4+R78IFs7Kv%ZhGH<}>)6HQ^7MoU
zopR_fg~EN_r@n9QA9;f}n&D%^6E>8yMR)zz_Z!QbNh*b)B-51W5L0N*wX80PDw^$u
z-(qaoe2Z*5*D0|jNIYuGCY$XB@A^S`b%R>!K{;bcqvmAl^83LFrgYXC57s4wn|eB#
zEtVUvOS34zDoI5#+U`s|m?fm)sbZXG2x%YpkZIlJKZ6qhIQpX$1LJF2aAotHke51d
zXQ_e8X40LE`bn0o`C3K4rWq<E!O+4N40R`j`wZ%m$8Od?j1oc!a)t!0XN^z{MQzCl
zM--W|;02x}tH0j7l9~AqyP2`?3a$EI+`V;FRPFjc{t74v(jkJ>fFRP{ATV?Zf^<ns
zcY{cWbf=DlASK<YG$<h*(lOF8G{0xg8|Qq_=bZQa|66MoYh>Z<J$v8#dG7nVuj{&5
zd!Zsd?0`C0Chs9@(n$bKW4%R1!L&gxIH|5SEw1y;6h^$<E}9_@*Df8@r8HSL0pg16
zK5DJkd<NY*9nu|04d*uI0Qp7Nm&md3aiRL;ZCAXJR~b<F16ku(L0qv*2L|pR#^sUa
z;g$?jSiPfo@8>uX!Ixa7c9L^66#7nMsxoJKX8;3ned+KiTXkXo$x!BRFeQb_fj11#
znqpIGGG9!N^YHF{zh><BQIlrUTd3A{jMAoz9AjF$ve^ha3RK>yfM+>$+-=#fsS0&j
z&8P=vTi2<Qju9ANOk|%;&9LyQgymG}3#-I`dtZUE7*i*QO3p(=Rk4^V!1je>P2J|@
zr)?_V&9qwIswFZoF?KF)aPVkQQaRPAk`4N7%P$fGXBbz4>f|SDnHyf@W?6f&@aM<t
zao4X0xA2y9=lz`PE(Ppw+MP{SXj@z1368krRGYN?gC=25Da}4mD?cc|86vR3uo*YA
zK`Uj^@4uL;Lhr3#{TVR$p^9dDA8vT#g&^Tq(Jvk4m!2zKj#~I9`#zfUUl_G00rV)-
zR$WD|@%IbpXkT(E>dm&`eeaMYxC`lRi**@WJV`q1u_Xi|NhPhNmwyr>{#YZ}32%AA
z5CXML8*U-Qi;(cD5l_Q;9bud}h+yYH!)Kd^f_H}3!ytu^;^^XK37(e`Inss)B|x9Z
z9nd!}iJncE=h30Gk2$a6;#<b|pv$em4xA1fNj{7khf*X%Clw$>bYN0&2)yC38k7KE
zG`x6%I2UmagqzK|QV2SgMmgx^eEF)9!*_^}7t(92$6f)@@C7L+`}KlaiN&ESD(=8E
zDnS+_+-aM@lkL#!QM`s`Y$5!>?bj}mmG#znELCJ;_v9I#@$e`cY&4e6n-t${1UdWD
zqo3<KVtwTnQ!mqp?nqF+wGta21WUo8et^lFwLkQ}MsPYe+FltV@#sJ!5+g_stu}YC
z<+x<c9$!8oOxTVsF{OtNBF8l=d|z$>0BU>lG8IEUn+k^0_ZXI5##;>1v0V)E(y_gf
z`;tPpyE?dD|M(!PR`Ten0wBp2A81t>gx}i~)TMl`t9@iWX78!}Yumpsv``oFCMm_-
zZ^FDMh;r1thE8QwZe^#?;=8fn5BlB@R<I##;&U=jE^;y-0-R=Vppx^<mg<o|uuH8P
zo`8R$`~I5Eq#E~J9skRnrq4ncw&eOWY!Pz-5kdz9>#l)rYFOunE8;h!A@%1cD|~{6
z4EH>#CbzhtdJ2#<+z!{c>%btO?DTX|gpjBlsQbUiey&*_XYDNR+jy37K13=_w@(K#
z<WhvgfPoV)<l8$YJ3@wr=LIx8uGz{`2YyjC#UF|icneCyB)@#6gRDouOR!9SY`>WO
zi_XD;HmObg$4sT)BaHACeWaqE>o3bgKfT#&yzi_kd_-3FEjnV!2r!^oVS<0|RewVF
z12M0KP4bY|QlhQ#e3f5z)u<_Z65Qh_YUxxomFM5^YXYKX>?(EbX3qfQ2)-#7g;J_A
z@a02eEKmS_osOL7b%leE2`e<ry`xI;?p1Vbyqjih-rCc^a4B9Gg~3x!;efst@?aNe
zLMhOB$tBRJMe@owlquV8an3Ris~%I5#G_{XSg9|r$D5A`b&BoQWuWXmbY<!pu#P3C
zx!2b?JWy`-+q{ftKkfSjBkD}kfQP@`67TL7Op9Uvd5<RK_6>7TLJBhjvpCA@k1yiU
zX{>q(P2%12=u~tz)2&N;W0Xb1V;W>;&vyB4#Jufq@yfVf?@}@6dAwdye!R0&+Wx-I
z+R+_@@%uY*U1~I0htQviOXEw=sHJq?Rn^_!Vp+eL{kAK1k(HG-<?iN&SXDX@m`K{$
z`2|LPBjTnw5f{Q@H}weo3x*kw>J7c><%oIp%@}1pbeP=8i!Pewfc+F31%t|EmU^7&
zwue&OtF1A^d?dz0dHj*d#4PXtth^}iEj#*44GKaz-HqDZ=BAO*%)=j=h9$NI-z=R~
z7j<G&m{Clj5A({!oTIHx6%#jAq9q4R?1t||3a>h6ZI>h;Y3aKD(C}3>pXhMO!pI=E
zZ~9!uKtjx&ZwBdF1A1P3mY1!O-J8qthyfbY*`yaby)&7~$y#jql*R#LM_`09UFeTy
zHyF;7Ur>K_3e6}_Q_hv1`-VQBYeLvV;%Ex*8%OTH#HaI(qA@VTS~B8fWzk71)9qbf
zskT(qJ_!ty9nh&c9@kx$Pc06YRhHfDsZEZrF^;0mpkJ3an9BJa>!WPYk1-h*UA!>s
ztYx2ezp6S(&CHJuahY4vq%*x`88@?4N?S&iUX5LU@2;@QQJ)ab+9jVMT|ZN49jvC5
zLzB57Kjv;EL5jQOTumIU%=x5o2~<%Dv!hr_$)&VdTO{cT0(Z*>A;^qfQASu?J>{m8
z_|n373J2n!7}tW%*O(2&VADC3)Hbzhr2YdMrv5vdy@KEDB_e$`ttqz$NFqlz;161@
zn7tgTFkbty=&79NNL9Gj^LhKsuRk#nDx<U|yoHZ<3xDI`t-Pb^8yOe`1w#FgSFB1K
zq|P%!=%W8Lx*%&n>_g3oD@v{s&GBmy{=F&^-{$65hKReCiY_^yhkpMbzyD-+!>8U(
zmufiVen-D~J5qYp%0h0`3U}@5B}+wYLC+(W5|Py7ql(;Dw5!H7soeNT5uQq1w`lGv
z-H!qhQ-uG?yZPtw*}!;aO42%KrZA8wtm>j_R2l0eR;&q(CY?9;mZkU=gCVl|QK}J3
z*S1<i|H;D2N^bL`>Ku{pIJu6QKrSmXD9rdv7GVidq26X*$N?#Yaog)#z3=PGGW`ql
zk`K5me*H!`LbjII(f{IUgOD0kM^0i;35k}o-4HlcU1k6Nn^wqIdd73vq*u6GB5yfN
zy3pv#Vs_<#!&rnfgV{?#jSq6q0&cQVnFuwTa8fI>ykuj9L7(LUaipJ&b(7<Ny0^&V
z$9}rki5nRHR1{~_%DOPeai&<M=tQP*!VxUTKh!Ens&lNlw0V%?-zM9~^u4DZ^v&z=
zveq3&e5E%BE*0#LuvrROSkko~%Wux?q?UIFwA0nL`N(5#zDf{ifB4~h%q2odO?!=a
zqz;Fiml>(I6U9xd9N;VMM<Y-<w&awVe~&$*xP(o=HD2?)thcWEo7n>UklbiLPYY1$
z_@&CRw4B7f;43RPk5@}{(;a(a$VSSP+p|KoO)cVKP-Jr60n<GfSd{BiAF)W7kFRgn
zkvxAvIFG_SVj^Otz_9hM)a~NKoa5J~U|Q>dyHi6bcV>l7Ewl|Y3#2d(tM9WGGs#7`
zjHJb9Cw`XXc)%7ro^qkmP%h2xAMZ^L2ARnwD4jcv3r8E5m<)0TG`XFIGh~x0o*oFu
zzH(2%YxbA3P&W2c!(|=~u8GZ0<>A&(A3Qb{8`l{3>!Lke4kA>>$mO-mQ%|Fc>IY({
zTD!SNz9LqIn;&vbFKDK+5n#m8eCpBm;LgpIi4w|j%W7?Cc=CK%@y$_vc;X@e^qCb=
zv&O!j<}PeLA3@N=-}X9XH)Je4q(565`kaen@M&I*za|U*vzElxRGFRk!3@2)7nyU-
ze#XngK|aLMBE7j&10|NP5u~t;>r&b>k3ApB*lorA&Jcxl9wY!QItj!T<)5VTAJee9
z_0Y#p3vIBip-*ps>2io$d}P<&bD^MC<H5}4^05IL^eGzj-60y?rE7NKE24iql6V21
zLw>ty54yqV^jCa_e+y%eD)JQ9{}R1rMeIrBK%LT4lb5qQ{`$=rVXyyuJ%yt&OT>uD
z);w0I4vY(nWng|8WxpK2xs)VzRk$5!6!blT<<g%!mlo|I<@ZDKtk8Dh4H--Er<GwE
z*WU%bZ4T<Q1PK<uL3+>>zHZ@u@tpkT3K?-wstP<=AwH?@S)5l^u!gJM+7_?*sr7{-
zlAj3gH6D7!8=V?uN<G?N)Dc?hPg9s!<kj42L%v<|;SIIlEUaop5{;LusF363Fk{9W
z#z?8btq`-jEEUgR2TDme^~N~9se42ED()d_Q(Uf*()&Se0erVGRbVfk6KVV_vMC77
zB<|{s6h9Wp8M|hU$MRb~6yIL{twMG?d^mlM&q0U&rxSGI8fOX&ZuX5rt*SzHho5a7
zbFssCkXRL#(Eu(JMI9&?)db1lNbozzV%Z!=`x%Rwdr2!6`71-}aYz>s3=>11m|zJ+
zpM?v*w0iDqGn`}A2ZZulCb5FC<#%u?f5ls;nT%dgPa<7Lt`{g1wg>E?eUR`F$8c`b
zxV!orP`}){9{!^bdD<rkIo~PGHGz>&QchSH8YHUUzkS;;PPhzUaxO7!^+PUcLaOLV
z)uU9~(M(+xc_P@d;<D}zQm-UfD=j}=<8Hof$=_QDK`i*+klnANy`$^Gb<?upzmb88
z!~brlm4z#xMX9h%sn@Ys<>FUmOfAh+7rz;c^8NEr(R}uG>2y=&i+-X~>#O65T}RD2
zYS?o;S>2?{q5<5<@lDQGDsa0Rs;0QxH>ei5JV2@9IVC}k*L$iY-WY3QkzvbgsLTBF
z9Tv=N>!puw?u=1koxW2gcZ+B@vlDC~u<z1+)oq7rTMwO+5&mUk_=9gY_1jUB9oJ$d
zJygU?znMgq0fsJ;R{R6gCG%qk7L2H3@3N+Yh%(*GjEkH647>D3gOQUaF@jzp+}5PN
zuo8_fxB0UwrQc)U85#T0d9Mh5kNyAlGaw($1H;}510GZa?UKhar9z`MNO+U^obClp
zuy(p~%u>kf_7>qbDILjcrbCDNR}@x@Y*tKV!tYp}4Nj!qy)WtUFQ<8Ve?mM_tTY(?
zS5q<zhMn*`704{Hqrya}OWmB?%xwh|Qs&!E=bzMeH<?x#=<ZN%F&vl|2673}kW)iw
zlnACU+oSw)m4hCmMM0}*yl`(Sf=}841o765+Wllj2B)Q4IZpZ#f7P)F=F9ujxoR}y
zbm852J&6)plTV8u=&o{kRp_Tb6zuFJbnN;{k)4ecU4+<_Iyc??(F5Hm8q;DZ4r|19
zqUL?;+wsK95zFD<U*xfmr|#aJ$i?|GmOYM>%C4*}WGQqNmij|+Okr!Cc=7FWoCjXF
z7HzCAzCeGn@|bF#m>^H7mSmXv9m)C5*y>w%92CFNjPsleG9FP#+%Fr%ypr{us*fub
z5};%hZl<z*nVFBZz1$K@G_6dP%y{IOc7dUKo%$`&CZjL2pzpG$alFad?1IKV=ZLrK
zG2HJ#J*)3SlPGVz%LgBdSqAz!5hE@l5Br5bP)@fs-^(wlDww;F5t?Y%qSoOwi9$q)
ztle#u>%$zmi+lY#a6cW|t5}obo2QcMT%1-FDr#&;JooKE=#hV#+KGXInN5m7>k0lp
z0^p<a(+)#dH$k<3#0Xd-|I4e(3Pq}8A@xqMtQs_!md7y({XjU1f%5@=t}Is}=Lz9c
zG@5DBLZTe7L55(Br&>ORVN+1QbNq2DtVrUYAKHb2LU7&nMtSVfKNr-$R@Q%g(}oZ?
zvcbO&+kg_mrYv%rIlDl%(oMUuDmD8_uMp>#Gg?^b(TD<D@O#P75a5H(py$0^9UNcK
z1<irCe1?gWX_|m}i;#bQn|CbW$kQDKfl~kF3;yw^MzfHom;=C+Z|Zr!$k|FlHfO9$
z4bjM}Y+wVDH<}x*yW&!90tujz3<*L`VjFiEfyckV7e4}GUo4~nae-&?_B*~m61=~X
zw(n-(=neO@enkH3m;Ar)+a{r}6XPv%eBDSzyj)I!CtU|vj@NJM03SA!;~u3jzasx)
zkjdyHxK5Wb3b-T|q&%<G-?l*sm5A0rR#pxy!$QrHwe!_5H+0OrX#dPK|2z?54B699
zzbS?Oud83D_!H=#uoip{l~CXKU7dR-wuZ1(iKBHrLjjgowx>s<D%K@P&bkO}LH)TB
zF@37hn`yV^-$M%Ix_`U^<{{LJWwo+!;eWp%f1c3)`2F`gx9<AewvE33AAk4XKJL3N
zSo4latg>YO8G~wQeqaXmdBJM3E^D=h{U7o0t`W`UX7k|!smbX|f=UO-hIIv!W`L{T
z!rt+OVW?GpTD~*(bEUsO4-+16x+boOt)<_ismm0_eb!ON1B}~wmEr&hcp;!N;t&P~
zb=9`$6fVzFfxn=2&awFbP5A(4e-L2I`Y!_4aVlrBl4Q5UHl&XgyUk?zYxa0n$&h<@
zug;GZ?4S7f4@=z-Mp_lRE|k?wzcTK`<}04b$Ax^qivmobT0091vdLP`*2v53fU^g#
zuFYpDh9bL<%Q-if7X0AQ)-&rF_ATYW`!bYJng4x_?e*C@ZsTr?_jcucBJH~i6^;VK
zt1ZpNNVXP?8M{W?edUuukTFx1)~}`S0aLXQAdF*H1Riw(`*rKt-gmlQr5zNAUG?tX
z0$Y*%DS*Pr%+t)0%R4Zcf|^gJnH0H)p=L!6Q+wdNFfW%ckpl+`mrLuGBHBE?pCl-9
z%JFaJCOElEstaEojdpgC+dm{R@K~Jnty2BxP5bAA{MY0AP6)Z5g?@TP{Cm6w)ZE>}
zOY)cVJl`znWT~Ac7y?!_rq4Zq{eQ>nTd1#c68NZm_QdNJL&!Va7kpD}ikeGX&bN%i
zA0>bEwL4h@N;@3jIjl2j`QdH)_V`hw^4RbZTJzDW>xBOOaUJUtr03&DT8m5IdH=J6
z3U{D8?_gZlJ{YLmD#e}pIa7f#p#I8YHCwW-eKSuJ=*nX&eQ3Vet~KmDX~GWk3Z?S%
zcxnUlIU1I^b!!D^=K!p_h<k`*HQD902a1@98rs+$@YF3L-DqreYv-xy#6SX4y=9SV
z98x~WJ2NI#|MEJ9jHhUPfl!fMeA20(Mc<~7g8xpd_&7B4?DFuNwdQAV$AYh~E+#||
z*MJwuJvsI26|56+BNG<V=a?M}KwV0(F_B{6#b{GI=VrX|{k^`GsP(cU*C0_mj}yqb
zaWP5(2hQ1+ykc>0v9r<Xdbe?;6~yHtSZlF?d76I{BS;SoCS69AI`y0fHLS^&cMtm6
za1DKrHEd_(nM#O~_1&hX*PHm3x2_msq`YRpD*136WOeM{gb#x<>>3!aH0CaVfBEA~
zaJ?2gamDUIH&@+n!TFqhz>RZfbp6>D;M#2}*0kS69w0XD0ZytD9DExeC9VzmIELn2
zt}rYnK38UJ!lg!cqB7GL=cnuV35s+wH!!%s0d(0dRFW<{D+hTG1!fhMeRi7>m)D6&
zeF>NkGQ#|#dXX_5rPCj8{r%IQIsN}#h+ktw6@u`3TPRZC_WPP0b&Tc_mIGxaL#}#r
z<OE2wv&D&M{!Yp3UiysUfUh83(-)M7LQE4|fsr01#~mx|@OO$2xGkS~fa%#Cz%J{#
zIbzmVP*uk6Yja&3`WaJMQ$;`h9^grU4XhLBJjwCyb>+Qz=KvyPB8Z}tqW!I<^r#Y_
zKK56l=Zg4t;EJ#VmD705aULIXh}hF}S<IJc6R%CZK_;IP;r8VcdO|%y8w?lOt{U|z
zCD15TJUh&}`8Avu?>nJof*r%2L=u>=D|po>cmT|}M$8De44ir&p}qtyI%OT9AHWWr
z9C%Nf-hZIoZ4L^02XNXUdC59OHQ!P%L$1x0q(_Xio`zijyEL8q!O?KZh{m(Ej4QPV
zyGLxp+TnP~lnCFf+ixC>up$Vd4%ak`t{)s&yRb|$oxb49)S-#Z*i;6$UaqpWY=gqE
zm~a{Jln%t8Z)bp$rf>yY$`=oTU*(;8cNAFh7b6z#hM6DTZAFi50pP`?Y0B(`<MDL@
z7HxH$>tU)%^ghqlYg8Or-$Q@<GvDAz;l&`r5U{?jU(4Jdus~Cy6P={BfVAbU4%~dA
zsr&8i7-Hi?Oi1OBAQK)Vp?2ni3lgBZX}>cQ2|M2JrUL$9>OV#60mUK&7@iN$ynXEr
zA}LQr<1~ALvsiQRgrR~P!3#ZV>UcQke{-cT*bF>`8cDXfc4w#Q|9y-7*BxS65bOub
z=NR9Qe~+OfZIl?{16gV^Dl@R5yD&!BVu%nysAxp?j0AM?dcy`%!l1&W^kffklfv(G
zxkKM?7XAV|5#_Bl_tp!2l|~Z3g>XSkrrkWYF!NyQc1bswqATojjjz>l6Fbg9?}?$`
zSzdtzlt{Mk4!I59xS{`5ED3WgV;Hp8+K@TS3qybi#z`}UHx)zFJirIzs@92Jk7dAf
z@5A(;a>ld5l*{D{5*w#NyI-2c6B?#fti35pdq4yOAEg(DQ`?4ub)+4~)n8fgEy@I9
z6SqXN6NNPnzu0MHW3K3dRCk8QCiOID*i?j<EthKnXQA%Nk6Jy7P)T_59=h<zb6P1?
zy!R(S4_pq0%Z!iAaV^OD3-XefGOLwB2Il=PwplMY%)=Z~iGoIvyCjGO9KV=W07vO7
zGf3!Ha+c2fr*jyKF#O<v@G3_7(Z;EzKlZw$L4)agHA}WZ(HID5ucXm{9_M@xY^&W9
z$`FBq8wb?(=mpOnpQGmnXwOb)8lIR0xK->V45chE45GET=!PlE&xCWPYu4d%bWeZ<
z%>n41*S42t>d^64V~?Y)*JD)G|HIz%*CzB7tw9AO(1-`{dw<;cXW_r8z6~>lY8wX=
z3%iozdWL&I7sU^OKz(sH;|dqwBp$szj2?q5Mb9Z*=9abF!-Z|&r?a@4Vu-Vi;h**K
zLxG#yg|{h%8iTvqDcmiiJNMlCxf=Y=eb_;k%ZQw)X4r`|8XsNxi&6X-s}3q={OGVF
z8M?<S$;BIC!jjV(<M?!S!6Z5#$@RZ8Y|(YlRA9kxFVW&_AFkY|qgauFP(=6Uf!hq5
zz~4@Gy17Dx0-2alET`v}F#MM7iPr1t0v!ad$>vRYa_pT9j5OJ5FyGQ>E-4#8e%S`e
z>hnF8R-VOnGQO7bug_%Bq;mVfe!+!^Vf^^>bU6lr;hU?Z((fm?dUHO}s`py*k|iQ~
zz^iwcJy%3+$sEg-P-HIvof6|yafSSewU}68Y50>2BFNJk$mJuecLnLMigF_wG}lO4
z?}tw}09bg!H|4sGDw&-j{Fy@fqOz2@a*pKg-)Hb-cV$nlO<^p5+>Y-I7<h~!F8MdW
zXjRzjn}(+T>gI$8e?Ko|9shk^ffFd>z0vNg6+&iW<(;Rk3b13uu03{8w{b;JsezvW
z#r0Fqm7X|CsaW(`PqXIZaV&5hEIKu_KGNzl#+iG>DtZ~4KB06uWY1D}la10-G4R}L
zM;BQlS@=wa+Z}fz+Zz5%pz|J=2Ssjs1`yGa7rjPZzX0j8{zs%2vIs=eO-9Yhu4c-d
z)zq|AK{R2<dF-R=5YJS|y`b;$47VR$m>Ra8E@8Fap~Hg}&?M2&verz#43PEo#HaEs
z|He;E_8^YN_!c?rUi0-iyu}w-RPTL+LixA0)uc`(i>O^$3z;9-hneZk#Q0yDq!h;X
zZxF)hTr1g0!bd_f`#r>=kUTiporzPaN)kC#PnXY04FT59C~F5bQ|JS^=q_VsTzsmS
ztb$Zt-=w);1uq>wZFuq5=~yBNE<x6laJRn8cm55kc-LCE_!VKlxlNg+1eerI7cHur
z0p<nY#4)I0&V?=JWHyWy@e6tEvkO$zT?Z8>0;0a~=VK4Bl}0``IAYu0#9qYRWURlz
zlS`2OB%2pzO9U%hn+qJ77#{N8>Fg=`CU(;#Qwa>KaZ#;t5ja}kTfrD$*F{Fz!#-dg
zBJVZ0E;P;#t8-M58DP*c=Y1=n#<D~jHyL|_Urrzw8TRU7XLHtm7g?%i(Y*gP2+i{|
zOUVX2UsrItCALe~5=Y}Mi(aHmCKATjx?Y=ks8<Sx-#}s)JV>v@iI_VQOqffn{%xN#
zTsiH^FE8oEf#VkuewOV=7yIPFvLBJQ?ZSeuDw?OPKXn*ZHk?L>HC}d^MiXCk$!}3R
zajV{y+aL6OCiVB}>;*EN{THM^>FxmEf)DEhWS(Ta_p{vqS7H=oma92a^UuXmB*P@4
zmS~ROi3fIQTV$=q-|En-!TYg_)D-sU3r|X?_C`MKbj;nPt5B!`di`AG(xi_<!oSHB
zKdpL7Ceg9HC=8nh;jO9Qn)$GsN!7`&`S`{C_-Otmj1>B?mfM<I&(Luz<L-yxQc>oo
znWrBuQvSHpi#BC;v@z_rBmDl&a#&L$G0#X+GpND?FcfpGa;ZJ2eW+)LlR0m51YLg^
z%p_FB!AHdzQ!ve<jS=Pq^$oM@4@c}H!V4NsJPLxlw;$k~VLXT!i^IS~T9owXx)Qsb
zNq*{}a=L>bpY6H!K*e8SOrh(}gp7`KX%0sWT_pABphTh)6#YvWkYS9kGDERPcz6{8
zQz!B{G)|XwNX3eIj&8$rZmzFH`WD*)Q1zvQRUnM3`rbc2zfJ7Ydg^d-4lnbi%+J#p
zQQE2KXB8DoweQgx$gv}$qXG*6Up$QLF8*g;rhab1YkqT(8&k4i1w<TU+vVQ72Bz>U
zn<5YFi)}sQIVXM$rKdU2usrs?^W5~s_??~G#c5%7P)XXi7PZgMA@Lk#vetb~7A2+z
zG_9YpX^^O-T&2*pyrF^b@sFW^>YVC`i(63-6lTfgcrx{!6GHHkQcj!|E7qLZOq3`&
zO~d-lFqV&mE+B94Ih;r27?g_kG^1OAxv9?;LkWij<A`9~4S_1jX5&Zo<{S34RI%I-
z=Fv3FoV90xfk+1kmC&Oom!9S*(IqZ8M=RUuhX9WE)P*!*gyICfJZw&eQ*e)!I2k>k
z8BhkHZh9SJE(d%1fPs!{zT0mew=$f*m8xpz%uDE?dh?`}IF@%&RhoUxo%rfz+Pdrq
z?0Je+?H}0DKMVc8D*oRwC26oF^4JN={}J$B!M_cwc}zw)y#D!A{2k5oH((z?7^XN;
zOkIob`wOIIZs9S$o{wWl9X8&FHaNtJWPfy&<s^bI((qFItq@bP)8?8YTOrCYcl=wp
zQ_+@Oxiyl);-`|L)C9QrUrv31FM3Flra^rvqkYSnxd*5t^p&YGiI%XJsBkOdq&}k3
z;N2onjjk%*lz@uUVNVBC1-SG|p4*7yP&>Ue$jvr2{PF<i_9zZ`bdB7zO*M^|i=V?Z
zjDzB_;Z1J{4TTa(_>L(FHiNQx%1jhmi>%A58~D+!((q~P>Cs@eHB()qH~W-RXrBYU
z@X_NE<Hv@p<r_}(H5PF>qOYd`ze2BXg$yqs7KZm6&lq1Xl_5-Xj#-xC=#dFg@9kuy
z^Nxc3k$dIiXH#x(dpoF>V=+tyo_FE<wvkG5=Tm8gM#u}92;nQ@;o!fb0-A^h*Vi24
zLA>X|o&b`eo7PeFF0xYz5~JgRgjbkX&S!#+##Mut@S3$@17heNw@E4!CO%-c!f7hB
zCg4iTK$C9vA+V;=eB>EFjW6v5^>17uaa(xwX`03L#qV<`SIMlLYcZ{bttlXn-~zXM
z2@46&MCvtN;g`k1&~7PA8D5ofJ<|b@qz&H|hUt>XX|tP%t#yNka*gw`Dm(s()~BD5
ziBZqK3I{N%-W;qrj{DIG!BU*cATauF>L2)171~DwMF`;0e7We3I>L{S4X}8BsuVd>
zsWp;i6dI!a7>SHJ%}!gz$HltP(dF6^fE?)5_g)&jRQ3U2#XR3Yf{l3CO1A$1M|WLa
z#+B&Vz$yI}b?b3oqB5rcrJaQ6YIfCxfp?NX^G*A$!&RgQjX;BKA)zJ1pUBkz8?b-E
zRbD73R2v$9<$LFkZxM#W<hX_*H)A@M6*8SQ(<y0u7MYIH+@D#dB{!z|=o_JtX!ltt
zOsYc9s<%Du6+cFhv_IbcG%_<?*s1nfI^hU*mwTA<9be`sOi0SR?I8<_h8|c1oXfgl
zgV2j(LM760dOK)jqPk{Go?sNoKC6yPdc*c6F3H<o_-o&Sh~C=<1-%|3FXVExSKeE8
z$Z^xqrLT-R+%x$C4YqW8R`pRGd!cfXLF_C%XLQ6^nlNzdWbvUx2yt1n_J#@J2l<?l
zm3~Xu^ISv4+(MyfIQ#x^ce5Op>w@b!>Xi|vIyYl=>8r@Sm4`|_MKklGp)b0is`@1r
z%tZx#ZVB&vqO`wMSQ4mn7e3O-t+RZiVC<SjHbX-5JuPET(}Pt3mt1Lx<+5dtP0=m_
z>`c{~Cb0LdNPCG7a@MXIy5PSoy_n+_EvK?h>(w1b`-G9<@>w=>cmyt;@C7D?h$PYA
zzN&7ucOT?-?BB(v=anIRK|bB)DzevxI-L8}GG3?Zs0Z72W93dl4=j9W7&h~xXmIAm
zMMbB5olJULwopjg4Y^&N@u;^<g8;qC87WU36Xy%X@yLO+TnEaor!@-+yPiDl9O>m$
zN;?tsRMLbuFVnnb8*t`p8^*uy(pO1G8t8b4PTR#Ba&w4!H;CR$p{ebX6TV@p{c+va
z;jf$4MhS>0C$k)qtMt3)^@`^<UaSdMOC-8B!!9{TDfH+vk#6bp>SxV)!zbRB=d}QL
zNW0vb&^WsIP_TUbuT8`M3otyjMhkO${}&bDyUbncDjIlRa9WeC=ioZ2Lh}Yp>jiZ=
zw-c+P_EB4tr1i{n+;O~o(RjcYBP;6<QYnl?_xRB#lEdC6%kvA^Zu8(@dtkg}Ptc4V
z#9NK%L$F+GA5~c5*IWu#`li`cO8pl3siZ~+`nWg3;aBJ$pHbWEu$5FzQr$3@<Pi2<
zc&V+mhMMH81aHM-)8P?iK0PKUOirSS(ZQE-4NVj=qUV-#R-}zsELG3+z+f!$=@+#o
zYm%P`l-C??CZF^-2~JOSX-Nr-Uw+<*sE64#Tr0obZlS+)d`t3RFnIMX-+Z0&(i4Jf
zWua~E6gq)Y?a%6^JdAA3h798a$e|Gv5-S_94zi%J;D@uei!r=fm}{29+gs@gwIz0;
ztUf;*w1v)6dJaM#I)xz|(?%b4P;ur-wM3C9RSy`uUX+*W(kWQPG`)(`g}1O>yzkR-
z2IjzZ?EMWXnxt4Hwm<U5nszS^c)#d%IafL^6kn0}a_W#F!hYG{pTVT=u+EC$Uo&^u
zFN*qn1`hRIgM153o1&X+LSSfn*Qq3KWU3xwhRe6K389C3?`$ytuY2|%LFezjDDs=u
zHyS#BW_w-IC@MhFTK>^JdnnnZYEz;?K<!YQyXfh8FRTJgw(t@Z8)<*Rq5*$*m7B<=
z-8DMRBXk2YqMwYua^i~9(=IeWwtx-J2IC7BvghZYro9P!-rQvlE)<n}YXi$H7^y<v
zt6rY=9F;$Rp=KWM6kFApi)#Bqme^`UnVp6{c8N`Y!?`A5y!K;k!y#0mJAGBF%AcX8
zHhm<eDlfOgXDOTk#p;pLJ?yde;~{`iwL^te<OZr;D%rCwQ5kD1t<sbvu<23m5Dp6m
zyQR5RaneU@StpsPifGhkFfT2K)+6~B=!DNQUi>Khth8lxE|M|A=X(J9BOWt~v-0*k
zlNT=XhEI6f`BT5F7mg|=7v-6;$!|8!d%<syi;Vonr|?aFt}_%DLDZU?tkmVnFs>Ns
zwq-RSw>iZuB~a+X^5hkJ@25|UzV+j~m9d=XV1O6~$q)9_jddGKTpKNWS+k3j*2}NY
z>rTd$M}~F*inf?$k$pERDf^tiOmwIgp!|16I-C}`-izKN=PkA}T>bIftOV>kn@RO{
zkuP;7{77S4QZ9H*n)|9(t1D}3;&^9ZH+K@OMD!h*)*of@7bc&TaJXzdQz=v2T8Bw2
zXs?(VM&7|DQ?JVS#um9YdBmr4Zgd{Jz(AtG(_Fub$kkeJDqbv>GQ(p0B+z=dBw(!N
z$-iPk|M&geBhVyi^wDVi1CI2uN5lKV91$bdRP#{?d%N7_ZNN6kEjFz$g(16B&ossr
zG+$KXL_&y|vm8@FS4w|Fej?2`#?*9p`%T^p*D01OU&y64cPFw*f23KB5D9;}P`hV=
zpY<ls9f!*A)HntbEG2M{#OkqTjl^kXn8zY$u$Ab0pA4o_Uk;l`eq7<)0@(kCUJIx@
zQ|y>o%u_*A50#U01SW+tVDM5#w9^rEAh1bF7`32ZLj>ajnV)F$gTXF=Qi?XGqV|%+
zcNuiQfdy!oGLrfqnC$CR?W!8O?VYj7oyZQViBkZ%T{M{Bw{}S4iV!i!rz$j(*m7w>
zi_rpGE5<oWnUgH$PMUDTETcrw%Avg5U(&4#=k>Q503dv*tUD9JEdcOPG=*BBaZG;<
z*I5~38^%QaHx(G^#vqt+l3(zbuTfw6fx@2UuK&XsMT}44Lrsf5vgahXM%GjF&~RvF
zqFuDFzDwc`{+hcq`B8m`Bm4Ll$Pr2sJn?noCQ?Jg{7rPklO=VaZjz=fx^*c!iQR4X
zXr`@7yi>W|E!C}?_mVN^CxXG0CT4_sPsgR`GM|!LR*y&<ubVsoLGIvPoo;gj&pVQ=
zQe*FmoiflkVeR4l^$z?0-ur+34Fxg|KJaU#{SgO8*eJ<;0Q4t*p+#f!`725NTM21F
zus7;s&Vt0<8BzEIp2ze_2OU2@S=UY=<w2N_M*be2FTfH$&A8g&T`PlDp#zWw*FPOk
zc<cm#EiXC_h=Z9_)n~cW4+$)f@)h`!4Y>Bcsik`b(L)UIUGH+--Y+J<&Df!a>FvS-
z<5?x1-UsD-gL;Onw&%s8nId`q-u<SN+lw(_p#$D%KOUH;(#$o5AV5;!3o5WT;dr4R
zgN3D6x?vdp$tVJNO2qZGHZ)7=d5b4=LZ8hE7GpMCp$^tB>ioag0?2z^i|NwQHY}z`
z1<Z-6`Nm=@=0Pi5u}0l62f#r6``5A4^@cH|9FX#pUk&h5)`T4Zg)WxV&SQp|z)KtL
zDp@~Wly2!rCQgC*al*yb<2pkj^7y1vyvRm_>Idl`5asbK_QQb3?wc<+!$+ySuS6y!
z)f>$HNn>q^@bw;$hYiYlUW$vSxn_GMYnwmFu}GHS*Ddj1YYFg)goZuS%6E}J)iq1f
z=m2H-a@^w#sk4Ea@SZk?ESLD3_g$bXlyaq$^Y>tspwhbw?LgN%f2N%Q^Yjh?(U(#v
zyw3ttui|#CzhRb8*E5Zif{A!U(-W@n4v*aK3chc^k7%qQ|1@qq^0F#)J#IO2S~DM;
zq@8_wx4;tj0L#5Lv1K?Yuw(aTFI3EfqIT9vVNtZi&v^KLvvR_imvG{t6LvCZ*=CEw
zv|Z!&5_9Fil5h6h6f8oDFsLE}E;@sHZCQA^RpWXQC^Z;Cube3OR%-O%MEgtC>=nzF
zHun_Ukm$uP5d&+x+mCqDiwTq*!s)}I=Vg}B`hYAtyz<VnLo-~*#KugKNMjPSA0{|G
zH{{A?pmYlVT%`QgRw3)xI``dzy-ehoq4c4{4pap{<G4XsxG%Ohm$X*5VpmkvD2!5C
z;ns9{PH)VxTBLO3&+$F|H0&%bE~V>e5*jX@SEKm_O<*`!W^LY)ZG+*#qr+s^s;p|^
z!cTk#N*)nc7YKx3<;LgWf+~#cp-&E>WB2K&1h@!S)O7osk(flY$Pq4LCpUHFfszyA
z7^B5Z;YW2B<l9$dM^af}HE|e*cCi`nS~?}dJIC5pcsFqv(tg$ox@^o9`F=TH)mXe2
zQjIi=9*``u>jY`GPNcshSGs0))vFi<SCoVSz}%KE{VED~exUi=$K=JA&dmQpMffI2
zWNeQtyMvaKL6z8(Fwpz*iMcR$H`+Ft<J)Ar(AfM|j0;jqw<`73yqMwWe|`G@{XKgr
z0mkxHFudW92s5&wF9$22i~{Z^r&=YvO1)+nLh=<fU|HSyA{~fWBA)BUOaBn`*z51K
zULlMmO`d5nrM=JzDw~CG2Z80^=TiIx1XgNh#rDLDAS{LF<SbtC3KUg&AE{N^t-%=&
z&6$Bk19Y`MTUx1_aN(nPeu0roN5&@hzn^<6tDAhzA5F>uv(m`%zxJ&9nBdsYQfmWl
zj2V*&M1cN6F#oPUSO;XYd+13PIU~P_fk8KO!z0iA*gB}NjVNBu>Qs&J4|0#LVEP00
z4C1q_?eIwLnt@Eku$nuY9>Jep0Jmi#@1*)1O(++q4Y1`(=9WV4_N(PIFyriUc2L0%
zkTbnlURuwB1%mbvR(AhXM{v(lRRSyH)$kN7&wsmNuiJ|YjJNFqXn=rDP;<k{cNr3;
z!AC4#1fX9z^RlYDMn1|!z9IwYt`QY2i5x$qF<GQGFU!3qit5@i$Ksb?&Z9Vvv*wZK
z^`N!2jLf@yzcs`&<mi%ZOe#|PfH|^R?ZYxua}I)N#(u5)Y%K1)zfz~Z-vxv|DX$!&
zEBb-Vik;<`kt}Q7PPQ7DkdJv7F+c|Aqi6D#2LdUK_`8l{*i#PBF8XS7^!AUxigHBr
z_&L;h%9x)D*#Yy6@CBIJ(qr>zqmX@UWacJJKs}rfrU`!tfs26sMlV*}rC**b3?1ZC
zF~~OO=!WO^Si3-&EFa*_%E0z$jB~N#e&Dq-H^1+9NLFUQa;>UL7*>0JE@?TPexE1A
z*JDGOjyH8B93A>ra(q#oJ04p#I$nZ2EY|~BY7fNQsI<~IZN{unO$7cj+9b7}IgLIs
z#j<SxJ$1b=0Ecvi3<r;J;;8+GAApx0EAf#_-;#EsW>}iMKDW(tq`Q)BqQ?zFLFqcC
z8W*}d_c$}9hVG$<l-jIGjL24vS-0*832e0;+jHakVi)1Yu&^&51UY}R5i6^Z#e9US
z3J<He&5#=nd+dgxt4CgeWS*=p(^Cw7282V`_Q!*t!ye5%K%HZbYR_d^&+%_T3<Jjr
z!<po4u1&Wpp_+sS(lY3P1d-f0=awIs=duJ6#RvDbjBXjiH)>&?*<(Mpm?osNppsG>
zfv6CZhVq!a!(#zOqQNF$b$)kTC#VsW2Eq_&N9;|B6%mC^GEH4uo|yRL>uAJwX&Uqi
z?eqf^lC)hU(7sJn^>rj$ItU8UtH6beO(@JnV#cw@t*6#vtO3!l*>I_^xe%NN<p+Qv
z;fS%7^9>nD>~IQ(^3kKX-;85x%{kW-JOhkyQ6c+}X5ux6Qdlw*FdT|EGec#aJTHbR
z$5pzpt*M5IMuWX7tnHW(IWGE5wdz}OVo!c%^E<7`?v6&XHau?BU0etA#Rdf0P-q&G
zdV+$JYR+y~un^G3dw69Hz35L5Q50>IibZw~d|TRiFYJMg!LMwPGRMa}8-FoY>?7c4
zFuiJJ)RL)h{k~BALF&=2D5iqKz8+uAL6@m&md!dJioaeg0|&3r4!AT&-mJs)x*`^8
zY;X#MZP#u1R|zbZu|1(jQ<`P}zPkOtwG&4)I@(1{(LeKKlUuh4Gr{1bCcIzoUCPl%
z+Jl1G=Uhf_NB&0i%h+uj6Cf^ifa(+|!i|$mz^+x)YlQnbFpuHhNn34Aj+7H7eKs25
zcE>~K0et3kRdx>6)D<E{%eyDBLR!$zM)C4m0OGgq&ReL`?FQW|Le{#UZ;%WObZ=5B
z^DNhKBUnJr!WT;W3^}KEXoeA^P@ssz$>^x40>}Mdi<Mj)4~9IoUO`()+&DQp%D&|$
z8^N@+z!bAUsnoml3i)j93%bx;2f+Ecvz6w#me!1u(12q5Ge_hjE{QQ?>d#4s;=8Qv
zL6X<sVJ8&#l*TaV@$V?K>doL7Mph8sxkvhJri(lQ59P|dA_j5m%33wYr%_I{ow5#{
zkoM{VI1Y18q8?7~)1!vW*td!_fqT~u4K%7yo-_)?zVwREV(y5GNhrwn>U*GrMk|&h
zC(_GG;h9HK#W46r)coOat6V^!6!Xz`FRy{p&wKQ)w+S8J+$|mYjf9@yc2_`ebUPZs
z4dn~%=*<QlU;DTkpXk(}<#qY<A&C0z$ZV}oL%9#Bu%blHLj;VLifVsAzTZw5l4vyP
z2)ZX0dbhdy@PIz%5g22Lm0XloOTl*Z2p*Itdfeb?CR{<UMF(Bznq?nmK+JAOZRbi%
z8(BT>tY6P2XI@Gz)1ZHTHfx%t<9~G=C>PVdlAeDkTv(Pgf$T1)J|$NC?HFF&qbFLR
z9m8CqU-G0b^)WFsnHBruw;7#_gB$-&!IzM@GS6bHDBE%FwaQ+<b~+WR=fzVb#w@=8
z6KCp)&pmZ*u7n6JtZxGAnQdBT$|HA%y0v%%I8zgs7czdWD=Og~z=Z+dFX73S0pFBq
zf)UKlcBb6869rM!VxSd@lt3D4vS3#(cpJ4fd=pFXL`C34p02^)=GEJ3acx@cbSJ`i
za&`~a7~na*MSeG&!|WjK;`F2J@=%)xTcJ^W%EVxUjM(REUG-U`I6B{2w~6!>y-m%8
z$~WOTE%|BBHYR%N`-uom0`pK$^dE4|YK|h>I9*-5s7j49Y-{!%`|4E6);FD}sEAJ+
ztgY-jA(6~Wh@~S$ulISdH_Ku5$4M#+v>l3Km0A0<k}i9caePAlD}zp7jxu&$QuNJ>
z4ZtlTSIkE2r$0>KqfgkaD5F0*KmB~PT;FO^aZw;f1*mH6LE}c7K7o1kD*hVhBCY!M
zD)bXycA3YUj_lJ$lB{I@%wzu(d`gNA$c5zSg4!vM|G|l1-@&T`B&fi=m!BGqC?UGJ
z`3>l!vZ}JLWzjY}Q0+~Va`%2Fg^tMT2rEjnE{XF&0;KZqlJN9Xc^#Hd!>}F!T~UoL
zI_~>8MP{=1hU5oCYWE=NQdN%oLlBnRxd{S_@`{u=v{0Ib{dA%gTFP=2KuK%qg%AfL
zW#?w$rc_XV(+V5$y>iPptY~|$v)%R`D0*+*jYI!YE*afRg2q8d8NFGkWlEJ25YL;B
zka4KQ2)xyfQIA@PuZqIh274L|)J}|A>X~;uQM3tES@i*W&l`UL*iPsQ&D!3R(T~e-
zt}iBnN=#eb4DrqJp2aF<5a&WnA3$4N-tXEs?MtuQZp!FKzq2<oiDL<rn!zLy=r!Y7
z888*jg?~J!G{!O}a%9#A?Od!J^j_Zc2cPe4((03zg()9tH=4Yuw;rz6Q&-CQu+c09
zb=fT0xNxiC>W(vwy^QDsGw!xv=EDe$Znp-}pd!Gk>Ro57c(i<Lk?>Z;S%C0MRhp|<
zu%nfh{qi2CJ4dm4%orsEEw1{~=T3snEesL{)^iBsdQ#ABHDX{e@io<$#iB%fNID&m
z($zado)nd)ZAQdCN;52OEN+!8*27wtatSV!ay{e;TaFdtn|0`PA+AQ~v0XsAKfl1G
z6rRl6oW}7z!j|B-W{AH}K&Q8f`sLLc;SIA!C4wOZY-bSj&*ljA*t0u054CE^XIV<D
z=6utSMOZ#Ksz}pY=qq9(qLJ)-`m@@`ak}Gt!#I~X`TAB)N)`?0b9e_L%EFJ2J)5+3
zHszFM_&(+`MOtR{8_#t2T_j7k+_Et*y<cPiwo0<o@3GVpc`x!eEBrA1omB+7zej+n
zj{0QIXDNz7`T11l=^WQgF`L)7S~vSha;4XJY{mIy#y+Xx(Xlw08_fmIA;;5OTP8KQ
zMEzZ&7KMeRhG8`qTmLR2A+^8&to4Rmtax2sd2(EHPuYIdc^7d|PWF?0;e-|JJLGJb
z3P(v@E*o=}naMZ0(hWPs8e5{j0pyIEI?<q1n?<rwHQEbP+O;+&76(VAP0O%N!8oE=
z?^udhZ%b5WOVk5H19E=K`8}rR2h*vd(V~HmwnJ{m?QNh!iO>p~P}okC?oTVuB%l<y
za6``1thZd=UzPVnH~;diE2WdrI1!?|KOM6ok&$ry6em{&4<3OwW}{!?Hx$hn{Vjv9
zDvl$Ab+djloFmRX`si$AH|}-A*Pz4i=$A}S%!UljY%IQo<_5=Nd6%Y~4e+mt%os^q
z@qVjxJvmBq<6_+yby;j0_~0VWX+BF2mp*Si^4@~7E3F9};G2LJ&Ee6k5><+yq1m7!
z)whPYssJ~xp?dlu6>q#mg5Efuj~fPU4PbmI*_VCi)KrmtqEcMdM90pHATevgLjjSU
z!|{P5#9(QfSEX+sX-{dA=W@tC^oH(Hqnq2y61gqqR(fk=gN%`jyahs@xDD4t1=f1H
zUtm|+ENh)WT;e>f@;>Ng&=;g#vaexJ);8%&%-&fYiyLrWBRG8P8ouz@v%~*9GfE1q
z1S@ZQ&AT0J2bO;|{P;x*lhP$ydYUK(xm*p8AB<M<@3tpLv*=>?7ak@IJ^Jj$_P4w*
zG9>RyCkF2IAK-y^DR=R*Vu^a7GEJa*(}QfU(sPT<0S?A0qqm&3(917j`4?OLKjXZ@
z?JfaQ=ItT|i7onj!uAmkI9_w7w#B~_u)#VV;F$bd;Y+!zw@0;rr+6fQ19f_6!&nfW
zEyFF5w1yVOT?eX4Z^IqH{<D=rSo_iYGX33GgTw-qicJCfpMeiS`JX^GyPr7T_^U#}
z7;_^!yIjI-xAmr_zyT>YWO^uKM<F4<1bTBUNJ4;dQDHt=Evtd%-yU*-3w*l;XY`Gu
z|Bywp*AP;_nMd-Tk=)4W?UW+;?{drk)||!*C?$y9o6u+~()(LJ#=lU!{%Q#R{zgy|
zw6`tzi%9-%H~g=M2SU&%;QQ!*{nYm#g5$p*^`9s4jtj{Y3Hk8qKiuvAv@!qtsr>)&
z0+4p!MY8IR0a*fDf%!Aki~tbnKoi_3hwdH6FhiqRlzn2?57LF;hTs0zO$%^LeI3;5
zSa;~U`B%7Q(d+??F$h3@e8sKT=Zk<ZTcmEqu=UYpi5K@A3`&RdFar?Ic09MU60N0-
zV;@McgOMzOMKB;(yhhGsY#k7ogcUh^>HJts)H}J}ZoByV7%o3?)^5JuM^eF&h!Ii^
z&<T>p5%@aQjPfa%=*;H>P;BFLtNhKLJXI=_L!_x2Dab4Y)hyEFr<<d9TP|3UtZCzl
zUV7)v)&YUWW)<ROj;=jG{SLA!$l_IGZ%3Q6gKgda+qkey0N<GR)|SH`*Jgt*0QrPQ
z0l4SID!?7c0S;F>ZPC1$fd&`_Fl5>j3INvNx4hFI)o5o5m%*vJ^B<2r??fV{>jt1h
z*a@1Xs)ZCKlvV#KY_LCmw!MHF@S+P1)&=-mt2b)^OezOp?#VOntBD>;a8z3lz@O^`
zr40prw#QOAAiac8xX*GhgEZ+_2r<GieF)0J)2S2i7wiD&Iq_IyYrW*u1F+;G7|Gz~
zY1wg?;U36zy(eFND$(n+r6-`-xktwR_!2OonT2uT9B?O9Pr!cZ01K|hQFSCY5bfxu
zg0%??u^4H0BEz&?WJTLu&IUg9z@h#2eUF=Mu$!1@!EX$mKMforX@QwQ>zr5W4scv-
zrjjbPQQsRd+&$R1Gb9+4_Kz-)%lutEoq~y9U;s&yW~wDDJe@M*Dt<X_ovryu%>!_?
zJHhorAeXC<oa>_kV?cPHniD#&pM!e<!3J`+2BCBe<@OTogeYJR24V9~woPOJ5zcB&
z?dVlpX6}>z2}A$6vSkot*np<3oeJn4s0h5Z9-qU(6!kLAzm4z!Q6xEs3XaF~yClYF
zmcA`;*(6$k9xp8d0s0E=Pv?ttfGI~qInX?Ke#v~|N_|aD^CD@oDAQR{XwBaLLRSm0
z{6r!qLHA|_w_mq8#0~!RRA?nuQg2Kma}vO#9pB$e>e@=@U4^5B?#V!Dl7{dfbJ61o
z?|@Ql52$PIm|~lPd6?VapLw^bbfPx8Jj9p8wZxB>FOd$^8w;)e{9s^?cj9Y7_tl#+
z`V}ged9>K|>Ed#X7!G3IyX;$~`8v(?B<R`bCs4*6g0ayzK|nM$-n|Yw%tX<GTOT`~
zOu7T!8s?~qShGw(|BNG;mt#lv)ww5@X0f<(=!p%}L?;4+M=CGlO4VGCxcXCVjo8Vg
z$-G7HnR&LS`TdVZDMY4~7rU)uXB+Sx;8gV!pTe4)%RUws1Ni6;Ag2fgCnI`jyKptR
zBkT{JL6LLj!}P~4p1~a5!UBMRuQ%I26Kabr1Nb-I1$F@L(1TPgXy65~C&Bake{>8K
zs)yhq!@6(aVSGBb>vtbu06%ZSnEPhe-Lc~BYvmg!#2fKtH#2UNXi*RF&e$v(-khgI
zMqc^>&c$&DvVz8ouFOlgiW}2Udeo|oG}zb!&%zzp5rO_b5ZPivcg!e|6!`{L1Nw<~
z|06!eK^15HHsyrm_ay^t7Ss*cxZmPxV%_klUKn_;%CE1yXXz#w3=dt`z#J}(^#h%p
z>?ffygxyGH3#F1R(B&Ja=(`oV3~rB$bC=i7d-XqJSy8+m-fdQ0%l6*c14^5Q#2@7A
z<eEdO=8CgG$HH2!#POBL7}-`t+BRTyV&njT&cI3?dMp{D`>HTcxPyWK_keta!wHw*
zCn~P+_fMNo5TI{~?4|^RE;{Q&uZ{PMK-|6uw{}yNq4KT5g33P0oe05#5aL%Wi(R?g
zPygA_MP|M7B9K3_&1=#)G#H$QsC!OfetMBru6Ph9tmm`aIx*zlHDC(cJ;`^i6zhxn
z)Z~U(t5INU9=ok>`UF!I3L^SmGLdO(dmuVGtV&fYw2t@7Cm2vsq?}&9i`iJ_$g2|i
zmL_ybLQSS0XBjc6!mUDA1KW5vYY}@+>3gJ_^5V<{4H6LVbiNeHh65_8KzlYp(ZP%w
z>TW0Ijt2)w|FG>h593~eF4Ks_Ca=Cq5@Oo`9fdRg2B2AsaJI6r#9|(diA$2#U)lWj
zBVPxlvt(_><jxl{YcJGJnUh-}ABNdr1+{mFLPRw2o*iDUGq$yiBQ7Yqe+$BWPr;b<
z-+S>_O0i*r<_Cr-Lob@~l*kHaE0B6KgV0`-BslvmYF6M9krYEUgIg6QIt++N1X7pa
z>pao7W7v|-nFPS6LTKvz*bhXqbWMrl^pCE$?~_s}2<e5sf~P>B!a$3A$P%WQ33Q4}
zfDAr4lB7%a&C*G)iYUpodl)BD`aX-i26tiZPNSV4`}@b~mxM|uho8UdSjH=kKGz|e
zw|x1sYOHw}V)J9$8lqek<}eI-A*CRsgU~-mo#?M;=G63ZR@h~cu;ykg&oikIa{naO
z7yZe`+v@^cn#*;|H-_As{TBnUSgA;Aor!ydpI}nI4zD=_xO5WxEr+q-C%t|ifEZ6}
z66wiBpX9b)fA`h?&p_Ph<kITB3YG{9?s_Jj+?EbsBgGsod22)K5up+31lDD61t`-Z
z2m8Rfuyesl1)lt_kcM=Mnebh(Q~O^oQ|(7EmF^vneAx)0*AD972W?8|nxny{y8knG
zA!e37Gwyaa9kPQ|fEVU`kiw3lcaiQ2`5z5~1zNmwgM=~2pG$s#GKN11W4Z0(H`pk6
zj5#05yd@*PIEGIoQ;&rUjwt=IU1amMYu>#VP+@`+hs{B&uS0wt`y46q@6oFr8`vzm
zY?i`wjk5zNsMTpHIeBa_GsLQjI}KkP>wQ(!o(1DueJwJMmr%RPI$j0*Sr-vFR)-<q
z#fM*YeYYq3$yxfes*cKsVdqi`oev+tH{TFhs*W%Hs{BUDLTr2f>GO_n<NwFrTgFwn
zt$)ALC5?20gruZ&$Rq?rkPhjRmhSEj>5>MK?v#*j5Tr#K6p&_8XUw(NTF?J^p1t?~
zoPA!Mc(XpN>1W1$k8zK2jqCdTzSl`+^*vB*Ja+YAtWwvuENmK}jg;CFR~Gp0hmZ&P
z9F-iv?@AQeM9i3OL7Vc#(X}c%%yi_rJOvq8>}F-fE?jndolj>DOfZLA8(yBtw6fxs
zu|?<y{!FP$t9Juc@<>89-uZ(6W|0$>5@+`Qg_if%?LFm%hncnzc8_H-<e)P6yaAAs
z<C!GxQJ&YDM&XZzACc|O)ojAFCk#VwZv|8~VL?Gx-hqiL2F#%K&AN898&v-FFDQ07
zN_%`Ng8*m-86PPC5>UFxcvS_&E}KP-^jTSuwv%O9nx=)fAKkANO&%6t@o3T6Qxq63
zf#hvB*bXyP)2`p*Nj8DBCxWg;j!4+~y5aK(iAbsSIZ5>K>>-#BQ-KzpfjkIo{>!g_
zpE)V4Tre2Am6=4A6T(E(+i8vZwKi&doXwAyK)Bk~OT7(?p^));GoS8CZ-Y3_KoVm3
zLLd6x=kjZ%&m*vfwi`@FbdPY4|J=QiMY8VJtYtt^fpAlnGBW?+G6cCxgvxfcg=Q=j
z!UWoxAT&GMmRqlW%|Kjh;?9D(Y(Q2aV;{|n2?o2;oIgQ%l;xDp6cK}*+kz&(u~)eR
zwRbH{gM)8Gx+iB>!o@JvKQ6dmZU}6B99dH-s_n4af3Zznm@1fiBALLIsM9GurRLHa
zr!o_iDGpVe+MB{=DX)@@W%edX(G@C<VuRPR<(9J=N+0|-&La0hKk{h1H0tDh*OJVV
zQ?B?H7-`PytPU=g_hQxi9|^<chz7K#2GPbMjH6GO9Z|5zcf2KA#@nV!ug+j$GKjv#
zgn_YOrLv%`$X?8#y%{)T5q5pkG$2;kR*xOI^<=5!p?i6fPu4)7I`+D*=whuy{>fZ{
zF3445-exWxCV01;E$j)~&0-xO)Gg8mB1ficHhq-yAIrWrUlBB$gbNIB$sTe6O2Xk!
zp}Y>wo5>gjKN+G~t<3CFUh2x!@+JPD%)UAJ(Wv^d(Au|d_!zj)9qzcto*5hB1DBB8
zeY6y*`J=xc@qZf0`*%P<RJXc^T`&FTp*B;T_J*dDnzs$OXxN)*C}@nR+i55tQi--n
zXfAr}e+q<LIFeWcHGkKXF0@*!Zd!p`v1zv`gH(zEk)hxbA}%HB$<e|T)C~J#TzF;`
zh<o=sYIk^H8#`L&^~cJNcLIoTd$3I`j5^GR@bSPtgdx*q!BVfJMl}QTnC8)I6k2!{
zKstR$ILWNpGX6(IBCos?Kq_1TS{0_e+9OT8GO$EJm_qHJLPl%^mrGZx(Hxs>#$^pG
zx@$lcjunWE`NZ{=S*7~ExLJ*Az#5qSF2BuSXPuZZ%`mgD7Xt*4ZYoLq<1H|Vhg@Xu
zsGO-8yMo3CR8V;!T;pKd-l#!#Rt7EUG^B=B68IO|v4YU)`FbY{ZU>%>WWYT9piH=e
z^wNa?9D*akD(UtD-3F#)(z`SklVU2M>t)({cmm%X1x>ZGZ#aD~N|h*h{6>cYe_l}g
zk$)3=fq^{SL5)R&s3_eQHMUOKQpg#t-F?}j=3<<EygF_Lj@vN0Z@G}{lx(UjpO(W}
z!=mkpb4arXs3LFk#fzHi8L&wfm|_}7S2H=XKehuG7HnmxLSwZJ;%9FgNR*LHjp}fR
zZ8a9x!7~}<U3K3&u8O8DwuW!C)Js}|=2VMOrn6SFuS|Y}khI6Nt!;m&_vXbOFvRXj
zA3v{p-eNooQsRk$g{5a_{30BQ>j@g?)pf;#GE_au$4jvvZARMGNSi~o_S8DJIV-O(
z4ynNS6{n*PJ$sH=<h_)CIpS>zA~I)X(buN^Vhw`h8NA;?z4GL4!&OGxVuZ3Mv*vVg
zZHT3E#Ah=Hx8y3i`w(6pA)^($zxy5>m&?QN%6pdV(nlkghMw5|>~$u{O0VuA?Xmj^
zX3xUrex7$^L&~pM+uAzS&O5U3RYYLn^-y-K*Za3=neUzeDRAfsUpXD|4^gc+H+=Y!
z%!Ux{q}Ge)vE3nV0SZA6pA#_4yAxxHoKCcqfrwKeB{l3gRM*Jz{D#JR`Q<rWd?1s|
zysU^UCD^4y4qx65gvgq?BkdXh`&7v_`qeI2L|g`um=OgqwSp|X!X5})Z+D4eUxkwg
z7f2&A&|*Bs&7694cbwAcR6V2;gKQ#ATS${kGf54OH8dZVfw8MgWxNpE#!acN7$$xt
z%))X&qA6w)RDF(Aia8*5ir=>Os<2K4MUsvSH$E^<Gi=CsH%PBx9SYI}UPMfBS7A|o
zm$F3AO_&$F33woTz{<UgTNB5&>s!DNy#YI_O9d#mCp*5N%4sVW3x@{dyr`~=S_vIU
zJ#}o9XRJVwz+DTn5=RsQhh%tV=4PKh0<Z8Vp+VhE0YH?Wz~o!!?~Wu7#lc0T3F~fZ
z15327ps{Y(BZq)1>AHgKf&m}}r`%VEQ1>CK&?=8c1d-*@a1}KXZaFlf?=OMEn6MG!
zk0*uYNBteAkB)zUqvy62l4Ed(50EA3eoh0cu3k5DdL>|<`Xq0C%TJlwj&h%U5>!DK
z^Q;?*XDme{H6PaNY#buRlV^?xdGCz;G{vGA)oKJV>^|LDnCe#ZIbxQ7T?#b#{O$Wn
z4=<-aj!Af%Owz2irK?ZjzHtR^e7q+D?lg`iu1%V}l)sdPoq8BkhE#^B7`<K4m{}hG
zuWibZk^U<{NMK1JTQg*yzICdLbS7d516(&kT%u|s=IX`5ukI`5BAgPorh41_(r+Xz
zOuWUmQjKp!6B@A>HL1dK-9~xmHV(RP3`HPi(vq8v4+EI29`}PLi!q9YJxecizI`j*
zm(4i_9K46HgJ~r%FrXCOe?U}z5&lw(Vb8vWeI;cV`+T24-shrvi_X2a;+!Bj?Y!i_
z$>(yxd20R_ad<y!>V{;;jKyR?UZGkPp+Ip(Frg)L`VEeTuEUo<>;j)g_BpSZ+M5e6
z80I072Xj+Tce50|j`%cZDlE7pcs?-PI`dT1cYvAXgXNh^uBcLeosG@FCY=aJJ7*lA
zzI<U16TIXPGL4lU3}<39{F1`wYEwL_w~)qk-Xfo^3K#uAnaIXMp%naS#?VNvuEcyY
zk42#WryJ)WtM-!#i+9RAa%|d&g&}<0SPJ&1hd&K{>cT9XuTwYh*dmO0w4P|w12y{n
zUIz#CC{q8X+dchAqfTC?(G@DFz;t$=VU0&p$-pla>85E_i&izck<=n!f!mW=I4C}u
zXTTQ)w^cl7wL%nC%et?<na2MqDtzQaeJsOcNMxAS_0lvMzFVFgv!W~a{3MAM`e4y@
z{mJ7fk{kw|*-!X{kH^PtrryMU<qlyhQ|D>s(x#uVQ0B;o>_sxnU3pmIX*}02WGT#1
z002qycd|SJ&(aA<6|E!5&5b@*_5-R@NUCW0LHnS>WZpU?QG(9Ke3?6KXj{k_ewG1g
zmLl<PBKod`t3tsmx}0SmJ|9pXzFqgy9+-C8Uh10qn6j*i8YX<GJYgV~K6s?oPix#I
zRr-_potWb-1EDk!VMWgJ_px2T<NG=8@67h8Z@ViCv8EgIeBjoOSkBY)YK3T?it;nJ
z?D{ZVUtMx64CakXM^cl?L=mq!Is)F_B<(Y4e8G18D?MX8m&r97yTcjh0<w2NeDy_5
zUF*wXXZZy~u=MqFzf<<Y)s$b@Pks{0&o%<BB+CL-hEyacg<=Bhvd^!7RBZoeQO4id
zPudAj#4!8(zjf~u{iyWaIfbWeNx?8Xx5Z>My3UG5kk|%UMR5uBZ<=2;Wm=?dA<L{;
zCBC8biHsUFruLZ-CaY(Bt#Ha%@hHFhXe}X;6DiAuGv<+!8wo*>6Xtr~m6xApBs|%U
zRX{e2@uck7Mwe^~ty?1dDD+wQ%sG&u63)h-A_WveoLQGRZXtFUzF8eHQnHdoCc-9#
zGpF?ohe{yvVPWY(_rpY4;tgcV8ggj`1d3UWYPhIfBY9?oYTJu!9`E&K)CS8eH_sm%
z!!mfz$_zrHI=FXns--A!!+Gq|&|UCsN5X>AixdcMw5<OqdJRUmnR*$7phnhlGVlBf
zs;r8#9}kfn(Timr(ywDGWarb~8((_TG6@5Y&8nwTcLV)(?{h#1xy^p@{_4F3<1M`B
z7QU6ZRyd!I21_?pzczsckt`d$MZ=T(5$)ld!I;IRETN|AukLT0-X3IeLR4+k@s!7C
zkNFlyA46e&1beC=E&2EZ$1L89E)=S$B^fPw-0e80D>{U&7yK`ZE9jQ_wvYl=L9&OG
z?d(&dFX#mkZA!oCB+|yQNcD3OGx_uj2)fs;k|35*gtsZLOaGhoM*`rApa#Vya(_;O
zI8<Md_tK<!C~rl~_QhoF!W+1YxrT?h=cVZesdveJF4yKOb$MuLao(+i0y!0|MBD_z
z<vxi(<mRM4gtwGGQrPN{()<Q&)CzEEU1;1iTYe11+i8F5nHu&8s>wFtk(by#mR7%9
zPd5koZp!>;T@xy*f>!dLTIiN`=)TKMV616d4CjCQjY!N4S@#aIa`qO~bnAaKruD4Y
zaQsM%b26avoIfi+*=4#i(86_@iGRf{e^?i|)Ep3|Yr3h7jR6rPGTCT}E{G-!x9|Ym
zt79flfVY2F$Mdb4(Xe#>4A(MdKtsy=qc%<^F@7>KPh-T(3z@GDXJEIL-gqwWPsxT^
zEt6=vgiZd8Q@c#16Iq@v+Y@s)T5U@M>bAl?9(FVE?6C)Fyh8lPvLy9jG1`i!MX-!n
z$XS=lq$fe$vg>O5TI&-VtIbr{gT5^Fdvj$Wk#qc}W@d(ci0JDVmpl&IdcsGv542l;
zHi5*sTT<U?`mlg>H0kE8t~3LipiJ$rtd^KOM@@9y6yo@H67GTOPf0bm5?NVh;S;0Y
zqblj11*$$6|K0YY2OD17)e_Tx4DULpKTuaHul9kS6#(7&<ZEI<EZtkfH#A0sDmd^%
zpfk`9JoMeD?KE@BSf-$n=|r#h8g}n(2H2$*=|;`JPQ9g&WiJr<SxSuv%ROQOY!{!*
zOb6fw<Q6}<@kD&Sk<C@eAKYoHD?>|PPs@*^RjNhsF-V72y09>LqxGj1Yz)1Qo7Hf`
zW@@G8V8E7yC!}wbfLVyB4VVxjLk<P2{i~x9L%+3xis|g_B24Ne)LRQu?vJu#>gJ-n
zgF1xaE<vvMY(Q;5M^;P?;|IyRcV82v`i1t(Yu|m`aq?9q(E!++(&EN!P-W3OqFZK(
z$BQ4)%Ex8dtyB-{RNvAd92tnGyL}U?c>xec3!A1>)%;|&^MVQEpU{2DUGpFjCf4yo
z&-B8G1tz8N-G>fO&sW@5_k_o-@GT?aZyNv?tF3as!NC`TNEhv{OmoKKxSkt`^<?48
z^li-w&L~Ia1jC(dZW>RjtM0rF7)7&fV{aI9#LHS3i26reK3!M5zxyt(66zZ9xLc^R
zlo(s;+{(OTRxe5P-#@KR(V%=&@j>m&pAYM+tk2&->n~_sah-jbu9V#m8h2%hA2g<n
z@vUGoK2qv@0TrUPrJ<z`c4}<{hh54^G?EX+7M>8BQWY63IP6liY7Vls*w72H%P#C6
zXpEV!X3ATBg(XQ9v%L0reYxM~tO|zsUBiSV#u-yu0%<kibw4rAE3pHIA|O+>wKJ|w
z%9d>a67556f?AjWar^16TkXB<J=*hL!`_v=gs2{~Qt|ko9tmWTdrA%X^Z0~n7^M!)
z*+4GD0Uo+@nfDG56R`qW+N#N9SS2dKCsj7Q>hUKi&~qNnO^|g{{em%b)<Nc&sF@P9
z6OX3TtCc=;vKy4FpuLH&ND)<%pOY%2;4~12ozfkVrhmE%-8O%9&vy0Mkjp!FpkjL+
z#?;?lQo_lXst?1hO|r)~zYOX$lO(Be*IH{kl2Yv0>2>BQUFoES(w(%CL|RjPI+rQd
z6a}Y4O13zr&o_ohKR*1GKKke5BfiQA3>}9T@!Ee1qOkCYGlIl|0?e@6!<!rjXhJW0
zep?T{$}`LYNt9p}6g=wVRj}3^NWMz@(0zp?({Xb=%+)&g>v<9jtso3W=e($M^{Oxi
z)|rb=mj;)IMQTfeYXIF>_XH*eo5YcmUnq~8#TLN9tx%AY`dF4A&T=~$y<b9mljYDy
zLiu|i@T%CPqy!I`i!-NyNfFgotxsw)T<4L0Dh6PI8M3}uy8($M$evUii%x1|0YP!+
z5W&J{;jwGiv#%OLJ-Bqzy{-yOPo7Bi8UN&ugt~yb$8~*zOYI8@KEN6KA`I1gzb>$~
zm0@@8vI6J=&1z0HVr5Y*vx_FtbVfRFuC7<_59TIp*0BB4oVp8)UKTep-u}duAz!o1
z&5+A=x2dGPdFq=+HDazp|N4(K!N)^j{b-x!S)Sdy8TeQZMo^<44u$;toYcY(uHYRX
zsUx}{wr?a#OtJVP-(G+=c|-kkpe~i95nM@vx(%81sYOvUoDbh0XrQ~k0*6v3LRgLQ
z{A2ex7*0k_So#ZQW;)}|z9w;O%~B~8TM)>J`I9L2&r>Z9X|Vcm^M&)Te>`vA$>NNF
zr?U0?hyu6hEloQE^U6?!(EJH&buS*}ZxjK4p7?py3@JIuY{~9#;HHpu$Q)C9=yUr&
zU=Akma~$Q>I6uYA!G(mWH-Pc(=bhE;DU1%uLg&L-dkffQF$KiG<kDG6y7NucN)fQ6
z>Ldl36YRstLuOJKZzf?_mi~QWK_?646WUveQ^DS3P#kQx?>r>}G#NE${`8zO*s9xf
zK`k?jzgwz-Iw*lMzT>+-_gW?X`V;RvaKhW7k4Kd&&0<V?1L}H|NIw(g+2wMIj&4ED
zhfqV`+c|bXx@@BCo!+~EKV+q%zkBfqi<ah0MulPAwoXvs!~6+~vYA4mCHF8hKgqL3
z_{Ffc9%0sm5!qUwW#_rgS*5HI4%&k1qDOM-pj=V5j^6AR!Cb_o;2HG|gR#$<KUi$z
zMeio)Zovbq9%JIFHFpB1kB5)1(a3u?wIH!Q>F<(3429fQoVNqF(V{kmPH%nc1jlP_
zf;cWlV0Gxz+pMHMsiXY++F*yL_ctfkV)Scy-RZGB!`?hfc#Km_nZsf4)U+BrDPY?F
z*VU;4pcBVpv#x^E09WA)kafemO1&m#-V5JKBVtQb-5vGeE~xzU*0*4P!t>v!-j0D9
z<Gp#I$e(JCTn$v5uRGvC==32VjK%9EkqLqd%K`8uIwZ(3gok0fPr+el;mXG?hAku~
z$KcF{RXGA}y;MPFp5%gDo-Yli*%yi_VJs@qrRgVx5CO?BYRPF3{T^7-%5}k5c6|UF
z#)U<~MQ;o8Me#iEfHpKu264C{kV!M_ae<>0;`j8aV2d}p2;>p8Nre}Trj*wzj6dKl
zwIUH)P95506zKKXm7U9y%banXL+`TRmX+!T(0p=GiF=BR1LL&8bk*0Q_GY@&xb&N#
zmT_KJ8b-@WC-P3E1i>&O>HUCDf%=%dmV$Sb$llAl+na4M*KClQSGOTseT1esDB%8S
ziYluKODeJ8HtKmF0`6{t&7hS_+hdzhB1eKrCx0ZuqYi%(74pd^;i*BdhMI~GPM75O
zWZ!&@L5oqE4FC*#JWLyq8YpWe?2R)EwxTIp1X==ITp4CvI>4~c1f_cUGsgaSqm1S+
z&ajYQbZ&m-?G?yG-jF8n8UT8ZB1A_e2u4fm$c9aOXD(dhMAbGa#^=IgEC6D!OG{EC
zT3GmaRvt=7D64o#AVmtIRZNjlmdU{|_0%^yp!iPV`ppm}B@I-Q>xvT87@_%lO_F}5
zqoIb-l|W83x~a5rTU@nW{a%?_guWLobwZtC9TBo49{pEg9;s~Q;lOPXD<cZg_5kBW
zy3i%UBEs(`2J{ivngh0nazD&OOuQcNu9GZ|s!W(KW_VR_0ha7|BHs}dfPp;4h(<$C
zI|e;d|L_fJC8!H%&84!0<(1Qz(ocZ8f=7hO?2F!+0_x?27wKmq4qsoFb|(lfw#UE@
zd~;~Dq-Kk9Nk630ArD@z24;W);$3b>8LQmzs7Ie}2at=4*znIK-11vYP!qvt)QenI
zy!EB;-&gy$S+`tX`1R^mkt*f?Q6EG=RLH|1++|PI#Z{QXAdG&JdWyq3q+Z42y3!3(
zrhg0EGKe=05Xr=3I7Px&>&E0?ny^P=Pz(ZI558{*POIi1*@nt6h3GkOaQ{=nn{%E$
z_1m01E1AL_8YmOGa}GYSFNXx6F|;Eke(45r`j73#FuAUv&sD(P#XnQMks$BUs%VdP
zZXgI_*84=EQFIj~OmzC*8^rr`tLxZoI-9e=I41$&ue)d+krGLJ03em*C!90qJZ<VL
zLWK=9hPPK2$$ZZq5_3}PK$gwNI%ibmf1_Y`P}mZFWlhahZPY(U$b1%a%a*6ZY023t
zADgrIsb{0l02pl{qUeNag94hW4WCn=!VNC=UNBrEC2C~0LbNd$HfI*jVKU*!S$u_!
z<3oUbZr}vgCpU^2B{3cCE`x4K&fk?Ty2pJ|B?bxR(Nv2h7z`Vi&+1M`2P;Hh5;u-Q
z3*P$HfsPjd5z4vNwWvMg)4`*~s~G{j;f1T3JoHrK3?qn<A;b#o=nQQ^z<2Ytzf?}Y
zi}p3Lya=g66q+TUoXD@3a2?y`$C)5ba=!fy!V2WX^ls3hCo8k>dTL7KT3cl?)x)?(
zmzd956Q<qqrmUIeTy0`cYz~kU?lOs)!Tn;V2)OpUUtMzFBsy5-47WX<oSJ)#|EQg!
zJeaVZtGw-a$e;I9N!2P48yx&H`()J4dBvJay?<}f4~QpQ`!5@?pzzrY4oI_%(8-N&
z2DrCgK2@`vj0y#ggzmS4ZhVI9Fe7V+dKF*In`-r^ofiFI*(6s$M2FSX8qn&;f1S$+
ztw&rLOb)hvD!*btLP}ir%GwZAa08@rUQl!6>%dB@qF{admb1Bj_ReLC$T8Rx&yn1$
z4lbnpkRfhv%mHL0k?}*kv8F^LNw?QW^5CJOBT9SzrY)(&#5+W@6Hb<VbBVXla>Bog
zR)DeM4dQKrwU=;un_av(D6-mX*BSawpGAjt%J3No(r-m=^9o~)XMF+mL!k^)hSI&@
z#73QJJ<&N)bhA;e1}Kb)z7LV>12vN4+=x@VnkGYkqa(%5Y+m$|CrQSLX=D(&Lfo0n
zP0@GP#KIXY1_Q=oYp8a;SXm8^|H5`CWa6~`{i?Hj@M;=^EcBM`@cOjD1xlwEyYlb3
z!AW!=&?IO4_4rTaX1nBru3(fwR1P@_+8=POg<7YI!Y0yfH6>z$?CjLkdL|g}sHSk?
zvEET1V{zdmz)ND_WkAZ0LucoWucC<i10Cp}OVMtiQR2{r|MpwV%ed+>qP(H&Gm!_0
z4MY+mM0e^jgtrxmx6KR-%Jhm=haktj^Q*3|f&G!a;(#o9B&P*#s4kxr`HtMudKYgu
z)(_+<%6Em3=fixhYur<Kc;q(KOct1{Df1KIf~-%}{wt+E;u6g+=&^|cI$jZk;v;YD
zozz$^RxeS|ODKGNxCd*NT@{mma5sR&w&Cu`_bDp{J49Y`gV31h0Nh4U;6WtasLGCj
zWV~a5Tu)*Jp-M5)>(O)e(Vql5*i5qO9RLc&0tJ3$2eeH7t;iY5Z7WR72E<j5^dcfZ
z50po-_3~^no-9liLBHjR4JQF5$xJ`BGqoWmP`Ph5a`iJ@+lvrNjLb8DKHKfpVRpNU
z&W8e6mZ%>deZcpB?^3J&w1I83rn6Fr>CFG^7s%Z_Oiq}e6Pjg##;j{GvYl0r*VD!f
zx``(hfb4b@#?}L1kM=*~OTsPJ+=ex-SRcy6hl=#%`9Ptc)5qOH;Nbc2sP`?n4&oZ(
zYZ9l%ZUWXog#L@xRfX#0El_`zCeo7>m#^<i-cR%sKy3=-wqyfMBOvjZpjhTJtrsuD
zc5Eol5abW(w`uWRNm1L9Vcd3~DuSJ=G~eoUR&EY{9Q2vjfRd9Wt`gK?2oO4C-Zh?n
zhJ&k9F!a)?TLJ{4qIv+R@4$J&HdJ&kA+B_$YM3@NKju7fUye^$KTSeA_>RLko9@y9
zb15>78~AMRa_^q7T6(Tfg>TfStiF&P5_#_e%`a~G?}#XLKo%RVQ1=%&qk|uQ080v8
zUJ9MGyT^?d?{Tm<PL!J>`n#9$n!yAi--64<Psz_W;VKFnovtcJdVCPM#s&jzX-^!4
z@la`TD<O~BZVE5fphKA{#&Px*EXP#V3H#JTRCC=3E^mej0h|65u;m4na2pFmpV6^9
zETiIuyQO*QZ`I>F({F#k#)-!{3$qS32<eq;#i7LI+%ZS@PMj0t40`O$Df+$t_W<cA
zG}y<OJc)VULB~nq<86FJ@t(6i`5zqmK94^+^p>fAap;W{CdN-)m4d5>iFZrsjUxqd
zH}M3Tn4quUct<hPvlQsY$T)peNxhKe*-a?d*|l*$)M6nno{?DGm7#B?Rp!t^Nuca^
z^yz&%i%R>6_FXbF`Evp_;?D$W;a6V-$6C}HV6B6qSQh%vYR4v!`>=n7!re24tB!ac
zAm|#aEM(~eetK~6_N$d|#nF}>?#4`(#^>)v*&V42mkWG{W4`=!FB^$%vl^Wp-mW>x
z^u7v&yQIGzY^Vu-+V{;M%fM^mKI&KYLE$`9AlX4$q%O;o<aXqms}OzIsU0Yk;+@jA
z%kkUHe^$>PWfYWCXVn#F-Q~upYjwvsS12%ha@aMB4b?%e|M^k*FFwu<6_^9}{gQ`&
ztDOZEa%C^~k1SiXDO4WpYK7lPMGQPC#CKK|sEoO~L!}PDv56umfe}`oKj&c}jk9=;
zf@-1b{2;IFf0aG|@25F`Q5gK5>X4dO>Cfl~0K1I?k5KMGkl8;iw*?tySrh|G;-AfR
zs~0kU7D_M$C+x5V17CLoaHbAZ#9IX~-&-EO9If@Y)(3ie8UM~W@{fxTqFz(V|6Ayt
zn~x~#3NuQ14t)br&eZFn1iTbU?NN~G=yXb-iQ!XtA1+RW`Vd6?e}Kc&<!Ticl%`ZE
zUA`F_0=dKqPi#HK>0f34zwg_>`eT6_Y!#Spf~tSTd%5usm>&bA_fJlb)14!bq6`Lp
zfbsk4?I`XsV$hj`oFUV`yAPorg4ieiH*x;cCcYw%aw<4{_P}IM0H2V?`BnAx*N$nj
ze|uQ^FJoYl0*=uxXnSsdy}=(D9;Rl!1la@9hKv2V@%$TJuMe2CoFRCZK<?WE%EQ`L
zw?Jich<l53`!9&||K$t*r}6iHi33FcXc1-4|GMtK2jKtQA1w-Cs-(bu^!V%N{;O~L
z7q9Oif|;F_TO#<c2IjwcJ$D!;`X{L>`}b1r|LLOt>reTA$PM78X;Odi0KQUAO5*Ps
z`o9m*|3_}dzn-e#1}xy+fo$&kBOqxk9#&L6`!o6buWESz>h+06R9H(1!*~ec?pNvh
z$2XsV^m`2m==Z-O@0<LGAA01C4-;{_fsO|r^&~cEz6dB-Nk?-ZmkSfa|61z*{YLP=
z{Oc(aJgP5H&g)n56|$`zf(S_M#p;`1>{$x``X2r7)8_A2ayf9Q3lBlgYU!h*rTpdo
z*W|tb^inyo@GtKlJa}+_ziUtK@fXJ8XBOI$&-1JMex*`XZ5vSgluY1YyL>W;$DBZ~
z!z{(Dq8~J%q7r1XdM#~gS^d5cqJ{q6gjuJ~VpU_yx4km`+V{XQ$C1O_J^R|HBgefx
zds}orGi`iCm)mjFyuzx(f`$ndj)odB$PWhxUQ!(O|MLGBM}PLBy@__dVsiLT-u~Bn
z$l-ABRB`{~UjaTw;Q>7DjiNXWbNFAM`Ja56)a7f6|LjKm<6Al5P{WS+;ere3(f{w<
z)-@&P|6nZszVOe8)YRRphy)1<O#k`FN{R=Gef__45kQOJhw6*}O7=zGe>l1RxF9ew
zQ2vvl{Odjc|HSwo5BC4-CdLz*BTy*_fk}qf<di&A+<xHaXfk89)Bip16*%Bwg$zam
zz%9#O3EJ5QlZt@$2vFOCQ7GN0S(`T!V4Yi-yv*J-Gr99tCe<AaP@u}rDxmB^i~jw*
z%@qVoj#|Gh?JA7t)6>{_Yjdw-S@^yihNn~!?EvjgLSdCi7-eV;MssodX-q{1lF-#v
zz(E@X>JoyiRiEMYelLUGOlvSkkS>g*3<EMu3C#Kb{pFnAs3YmTEW)5<xaPTfy>j|k
zQlAiBGMB&~RH{YD0dBz%Y-!jqtL(1hNjK);NB0j&@B4Qp#cRx3;^tg~|NY|r6~`j;
z=mJ<8gfjsJ5Dvp0;vbw2ird9I>e~x|^h6A^^p3fo8vqyELpxyR=m9vEr1qtq-`GA!
zfk#07U5~#Fdd88e0}<Jn``ve7Pe26+iVXd}1&UqI<HVew+otaBz_>uQay90<mH&40
zM#&fY`f@6zg8&%B1eF*I#L7>8fj$ju>s0rb6RS5f#iEZiQK>t&#qO`x1jgN8UV^_!
z09t(ZXmRz9&$@0>dJS~%K#O%BF&1=?{KGWX4f|>Kb9%19faOX9vr%SU3;ts*_qRp7
zJ8$d^23A4d8&EntCj+W+eVEwZ!-R2TUmplCvr<yP+2aTh-8&pB5NJ3b!pC|pt%Lgl
z>vojde-;Hs<8{<$P-4V|Sq5Q%=@38~FxaUa&l2LduIfTI1n_7|XjT!GwIF7KhGI(G
zkBpay>9$nPCtQOyMYbzv9e)q#KizK)@NmI6gl_VOS;Kk^kYnq<k_hnZ<IY}#^%?bf
z2BhWJM)j~P6HNA+(-lGP8#j!qnIr`Kf1pKD`S)#`8!!ciyMYykn3@9*7#$-)A9O|n
zp9rt9Y!RWGpXOp1;4tV5gVtn~RD&Af^~rpjJ)H0WLeM~^*CvbMdo^|DsK&ua{mX8m
zZg8J`809K}>~Rg&N;#F={N*>7MVp@LeoCR4VxO~*`oTc{HGCqYyMgc8F<8%G3O{)Q
zPF2{zCZxwW21agt4M=9uV0F@OKnK+go`Cg^jlr?Ve~ijVKB{<JAM}aTvG47^{+iaf
zhq~Li=gK;l5C)pMT6GH)X6_I^D^)t{C@HTT>L@m*iGzH1ANS0jqk+03fuN-i+dolB
z6m*=AyMyV+KH#B}WMZl_X1nDj-i3ISu7xhh25zYJ1Q+ZSVfyne$SjX2l$Doe*tK)~
zy*m8UcqdgAk(BX9Lilnr)@NhZ=mFBZd0<AIrSTW}WvPVGl<<qF3K3<^-w)CZYi0WU
zLGKhUy}*E(4O024_b)4b$9G7xJ3be!tHFohzc1en`<{Tat&2v3JX?K|&S)TOs-iw0
zhuGDi(oz{^&_Lo#!9h7Un927<na>An@CZCg=;J=;wLN)}?0d<gObDzJnDUo_-@<@!
zvEwM%cA3NKODrPjI(fUMR_URp34-??2$9}QA`0t6ThOnI_U4Q_9|N{`f_YDUbqmmb
z4>f%R0IXv<n2!GRak6IKzJzX72k@z2gZzZ~GazlrM7<9nYRlax$wKv)K=s-dT1C^m
zAT#lHe<j=k?v}U^Y870#0AhyC6mp4#zELga#xuTKg_O-XzA0*V-jWSxUlkmSeD9gj
z>w=;5maEz8zhjYqy7u_XL2@0s^(j$NJ*aMXWO}YwE;?>LeHarv{eXRS4Vw?49v!8j
zRZ^jg3!pLUH#R&YHJAuyFG94Bm=9Fhj;3?&c-nqEszi+g1pDg!uKO%<;Kn^zA&gY8
z^ITkn5kWTyGwlx-&8V|jw~Vzog8}jA9ep>|;Vo#6hOOk*jjuu=(h1s4F!>#&VFNk?
z?qoCU0nHVIKgHEd{W8QxXn1lk0c*(sGY(H)<Liq2*n1`MXNdg3F%h&Ia?u8OhB)M}
zh%}rPs3S?|PhTHDtvHj~#rtE+_5UsYVEJLj?quOMzfp_~`t~Q)65*{+$Lh9VMJ@^T
zGcgmnIc@lJRq&S+H;~4^y92+jwe@*9wISRtQRgM(kA?4#Edw;WX))~p4bpalH<M6(
zHTytUGVY~?DUcc_uNgNsH44P!Lj``H%EYyoB4;xy-BdxXFmn%^qxcl{7%0f{dR-0c
zW0Bd^DI1b5v}pz(&2flmy7a!KcBl8fihF!h{lFcFl!K^lzN5Sm1?`2x@U^Nao94|j
znyof&t_6u;*EAA|3KscChJe}UH19ShupMNg_vqpy$f7&=sm6|c39!=gJ5}_B=rWbk
z*qsR15U1%BR7sQgEp6$+_x1qYMB53P$(_0ANm>lA{**+{ta{0GP9npx@BXJomxPNS
zM4ivd0p-*O?3j(AFa!szuya6q#arpFSXwW{N9Bj1?%NgCL7@u@bM;3Ha~*ubCu!q9
zz4AXHbX<Y0i1!Z%H%#qrF#W)o6#9yfYN8R=r|*1xea{`BtM_Pc)}`Yx(Mzq}?O_e;
z?7a7NuFpGtq0>1_dmBK^xA4B6sC@@w_^Ez&Pxl%unq)ASt?QV8ffuLe`OgZg%<m1P
z!d-VTm-r1y4kx?Ep!}O3XF~T=uxI!)zpNN&=T5;oZ~|t{+FW&zbXFFxmHORT-+0$a
z2$}6sn?W_&#|b5`ANbyFjld`qlG1Tn7<^Td5eF9M)lP^onm%iw%G|u0#PS2G@-`b#
zowtRnTjl?_fsoe^aD=;)v<Sce(t6~3f8on+WuXyO$jmt9ccwgK0&3zrDN!$+X(8WK
zg_M4Oy@HIr0c~*K5GJ9V!E3bwAz+h2akQU6hG_)+3BnL3S)A5Dg^&N;gZZHPcYIHN
z!FpNnjd#`pCtc87`+@VDH&>p!Z7;IEFBeU}g1jjdF3>+Re;GO=F-y(Aix8hthrMSf
z8kJE#SX{Yx1u}lWyoogp!YMyqX~q=z(0J}H?xNH>uKaAYn9h%3cB2v)y+9qX;mcEI
z7(m+-JaL3h0c8lEFjxrBx2HP!pgDrK?i5<nCfBl;GglNSTCW*zJv9D&-ae~R`(1p1
z8FYUFkF#h;c7)waD1RzK^?5w}H<63xamC_#pW+(fg%KGxoYQwUh*f#9C>7<lQL=!z
zb>#06xj;pgkJ|A&+Ro|IrVX%ARTxY4yMQg*MFk4_UNygJoNyOMF^>GKIy6#^B!Pwj
z+eXOPha~+gC~wC(CAb`klTH;9QN6(0bm5#&z}VNkaC~zG?3^Jm&%5`~T+7N)2>-KU
zRBUVJ+>})_uvC`%1b^0~;mk+#&AIGlT%zEdjn_QYbz2KYsIb)9@w*J&Uj}i|bg)UN
z)TJ>^IuaUTAPlw3E{Tx*{?)=N5T&gcp3~ic3A%b5?6$<-v8!lz{DpxP)rcl8_<bps
zBHo?-c4X_|xc7l$jBETpr+sfBM`TWug&9|u5f*zk4K_U&g9%bOGa>G)B2qeUCD^^>
zTt=NRmsq$2A=W1QH`w~GH1)>w*CUuj;lXa`T`B8#%1h7_&0Ylyk-&rn!FYgSKyOUB
zO%i<=kJ29DzQVPgrHc9`RfN}gLNWE^Gz#AR>fJg2;cFV9`bT@G+XcJ8m~P*{d;<*r
zi$^i8ZFFrH?T5~rl-^UYP^usP$2=_3opqz3H$gBC1)=5QAQ}!0HKVv~fF?q6+{iiT
z;c&QLd&uklaK)O-?<iCSF9F#^WPmk}vy`fU3Tcx&`}aeReV_{cJe@uT_uB!MyRKO$
zJqzFpVx1fR!RakxiZ!3)(dD|^qm0~fy?l7SLqm4}qDx956Oyqmbi97RYup__iml_;
zqP3ZUR0U?r3LH_Ct8r0&9&*gS^bhd&3b+Ch70#RUnv|<&Z$AVRzN2Ny0P(2uxFsV{
z^Z2Qmqc-D&QLn2|ejlfg_YyDb&N~6X(@6`w0XkE-QmjlGLP=dC1$iy5Eu`mrh?f3g
zFPgu@X9{LutDQIi1zae$*LQ?Bq0}4VFQg5}q{;TE&H`&fo=MM3{xFG>Y2JLEB^dwX
zZC+M!)80+O+y!WyxR$YcW0w0SKuk);g!vtP^JXwH2;))+eIi<hubHVEsnr5qJCFMK
zJ0_R!r(9o&#l4;h^~&{%+gEZQdVSn@)Aly&`)1LUxUSmfqr;*H0+H_M-UL@>qN!Vy
z12c8*EMgbLj9Q7uDi<joflraRykLH<W(?$kk39S#6EU=@0&hi+tltB!GxfL@r=*$T
zRZa`GrQ4TB8mAqh9Z6gPQa(e>#co=Khy3&RIOfZDZ87DRHQw}Idp9o|%S#ho+F4*R
zSE>_Yss?Z+y6c(X_wDVdrf9z{t@v3D7vRH9_D!}MIp(LZsHcAaD2+Yn^mi$YJdPd}
z%<!1=yajk-JQIf7?z0ZP%*W3dI5>&YKs;(d+4J-bODZ-WZqXl4KE9@c2D#(**qJ+l
z<3~Bnff)eU=$}I$MDW1aqOi{0-(GvGa-x%L>*Nm$cl9cJxPhplczijP`$LNHfvh-{
ze?>JD&tt2-S(dHDIH&m~(ZLV9UJ8XH<orvZvSB^9i2Z`-7N4x4|LJ$-*t47)WUr|f
zpoD3~GEH?_55}bvX3lLj^?sl?%4sNQ%FtrlL_{1!&3Rs!wcT#`v_d+$__Jnn*vGg7
zr3H@T^xvxJJJXu@1IVXc2Ej2LWN|0u6bdT29WbhC$3cLuiAc>S2b5lmK19SlB0)!}
z_^jYm3(?$eal~n1V$jDaK}X!w$sjRSG;vPjH9sT{!r2{SE!~iU{vX&F)u_>419*2k
z&!#^Uv9K#d?S7{gn=q_u8iY3!1AzfLa;E};){U8=BM?Zv3*jOptVvEhwzOPtV_(CD
zUj>oP7RIoqmAL=sAU}u=A^!;t?i~PtQ}o9mxe0F8#f|^K?(HA<y*r8F4Xl<>H#lCM
z_#=2SV4(G@L4$~T69sx8H)&4&h?Vn9qmA$hks~vafO4`nNeKJ*Vhu)09QTXeIWYe4
z0taG86_;-8w5SBc9|VbFvI_%;1k`okv$O3uk4(2+(wz#;({Db37y%2)OsI+)H@@4t
zJ52LyI^}s#z|CiRXkyMy!J+MNLLC#rQ9gEYr4HEdvR?HCJRIU+7;Nhoa5Jwyj^#VT
zE}tKix{1Txm7!6^Wx{5C98VSImIP7)Z}#cyfKoC)asXvF%t1zlfRoP<1^>#gT%9mg
z_w|+nJI#^}ICidVKTj}(SEP={aiUI_;G`2j4FuL@R-^%X9EMqNz@fsxMLA0yWGcW8
z^V<mZzz&ZVE)*B6HbQO2GeVb=a8Xw-B2Pq4s>2baDSqX8jA6!H(bB1q2W#TU7%6EK
zDEvOm1?=xn*5w#VG@GrS#K{Z=yfY`4Pg8LL6v8%+B|pYpYVQGzAtDUz0gQ6<&F5de
z#L@Fbt4f2vkcVXrFqaQW<h^J;phn=rl6q-?Mha2$Dc3B1S#UFBv(20>*coR>gFJm_
zk*}#>T#i9Q(3~-}4w_iGVmQvk=K0+`t{@=hrYn9mafVTc))6>j6o|`(fCLeZqjY&Z
zMMV%?<{Z$J{sO^4?Y%~u`~&@g>KQc?c=wkl5<jG8IWPR4Dd;EvQtFqRz2WCSyL(P>
z`5i@UN6G~HbSuCB&Og-!gCx!;>W+$~3}?CquKBgIf`o9L($-`h2J~Sc9nO7R)&R><
zk(iM%eD!VIbF5vjj={U)@IRLjrw3&YoXa&u(o<(;uu$3H2d)A6sICz1;dBuWUyK2X
z5LxPV)qIF(yaVQmIIBf%RazE{>FT6|BPy(Yk?5+##ki)giewo-bGw_D!A&gtgJrxB
zoCWk5HNJ^lDsld}Ip(id2dc?38CNBNT}~YsY4BM@j&57GiKM9~TovY`QU<nRKIU<@
z)_<%Wr)E21^G*<P_FOy3{_E_EwtnY0-B8W$r4JVPmiRSA;T~VKIZi&+M8DrK#*>`B
zN_~+g@bWUR$Dl{dii|l|26FGeo&WO^ElUjg)!|_|M0`rd3>S1;E#4WA@3O}WLE8x?
zd0ymKfphzdTEy7H2dpEzA@W@w$Xt=pqXWflF&Xvf8$|nf<+vl+B8$<_3WXJ*9E68)
zjWfoahOfEbkWG;X+|bB(AuV%0%hf6BFfQc9ont^kIc+$4srg!*J%NArZ~?WP=$HbY
z{zaJg&dJkLzlwwhli0KdK8ihddpS77D2Tw}!>dV|?@ao>(-1>#lJF@@yIo?9g}O$b
zR_d<%Dj|1X+Q(=<&0o#^Nc*yMmW|Q}7dG)++Ki+&a0}x$GWtA{+C2Q7KDOv-7VxvN
zC-;}hxX?UI@)p~9qg>w}!rImfb0uqKp@qEf6TS?HwUsH9i$?Z5`-+7ZD9cSCF_Di;
zndeNHjiS>O&X>#XJt~WOm&wK{t@9vOx@^aNHc5Aq&?k>g6hc;Vx-<t%1q6L=i|}h+
zkCx63mxhK(Ylap1g%6B?&~}J53bGhXVq<Q`^H_au{WKot=D7%L=wVF%Qde=EnQBU(
zvRst;t2u?892CNF=U2?YiDY})RLH+giLXYYPEMsRIC<<@7lHO9L+m8MnqK8R;<Mwn
zc<MaG^40?fqo5)S(Ax;V+#&;!cGpHQwVs1RiEa^+@l)#cl&g0)Ycr&ewqbE#m{os8
z58M1Dzz@0kt|TF)ecTM?D~*z-jWR2ImfUWPC|g+>wRY_9`Q0Q^Ge#uU9AUQuP`UE7
zC?fc(BlzNnRu0HBw49Y1aW(UBp9#(hP`@0Ca3JrrL8eQP<{U&hb^vF$x@=@R6)1m+
z>x|xOFQ>;}2)IPmA!fwYMJOLn?X$cJo>H!x(78#|2^Cogbw*u}lo|6FDb?w%W$lym
zs4>>Wt6Of*Uwou|O=a_gN>Pili-#&|W^c2rE!P?=!8g=#1KW7RFp#any<PSs*>H7O
zRHOJMjuO7*KvC$r^2VcgpLH2z7Ed3Io;jI_CSwnH&~;1GL9cG;eRED6;&~yKWap$@
z@6YXoX6YQ|_Y6f}t(ZyFsG9LUEviDIja_vylvzw7idUL97;3fH7@lfSZoFU9QM}kz
zGgTz}MHgxMb$vIES8clN;*ZRJV$6eEQs^6<%qZ#HFjx=Ni&zL$WpQHjBK7cT|5K~`
zionrvG)9U<2Su?HMI950l0cHi(_)XFreR?$_vNQke*1P4!3q#!>0+L@f<^hl3j_Tn
zSClE+qo^Tsg@TTw4ukT16!{~(wTGiY#wJK7W>BJ8p|nQa)Y^i2u~QN00_G9hKc4rq
z<i@ryXPd#{nYodQ+s1+>TTP#BOvI?fCb5sr$t2pd^Lv!$E=Ma7qU51+GNc?uryZbq
z@r8@!N_u9jA7$YSV&{XL`##kn1t(2=Z0qUZ6e<r2I?k=cjqzN@P<?|8Pr6O9M0hKe
zEn5+v4LT|Za!mxT=dbm&c=B>8x0Us$+Weg%CO->8ctSpD&!KFR=|)rUrFKK$jp02^
zxK}pkCdn>zs3=Y`mx<tC`#(s2rH%QJpg8g<Wt2El6czDqO0uo`vo{>uwXWb8{+-1s
z>q}ymB_b}Z@(|8cLIo)9-AF<>$LD<V0c64<8yPd6#4Ztx3*k_4e^v9@S0Q)CAIwr;
zr5u?H(X=5vR&wHGrLP2n`^oY)0v}N}<c#;Fk~LBu;c!g!@QcjJ?7M#)=s3seJ{x5D
z^p#Hupslc;J?!l*7lPDYZf=y;glbV6z?-x#=)z_C+7cITsp8z#?2^`tjdu(`jXa5C
zgxLqF@HY;($9F%{oMt$GQVAeCjAQQ;!bu-dQ4vjF*am9XxMGB3Gp$@~a4^c*$B%D7
z_4QhG35C+at!Lyt7)%PsqfhjzqL_XYI!q9}>&|en@x_I_NM5fq&K}lve2Ob5NUq&1
zic8JkPO{A;3h~am+kiX@ui%p(EkZY&O+9OYOAKvplz5uA>kN3Rvb7Ld=*5bfY%V^;
zT9I6vJ%8;?OSWC|OZeLMtFira65Cez_MgK)J_wHp%JQ&SnZ};#QmNt2@uXSI2udKQ
zzeI3|$$KSL-EV~Sj^s+0kh+AcGryBnn}<q9>veKU&3>E>4PNfd0nrG&sZ>z`QufbU
z#VBUb1Z7m;)SIAJoxsxc!^`_=Y2P<$vvrfA;8?#2bsA1ro+>l{BlFLV1h*9vH9c=3
z%EUPq)I0u7K>{@@Kzw1th;Y>Fr5WK4okvka52s*Z!}NsI4fkUzMT4gm>)%@3$xJJ>
zE_Bw<MIE_K5TPu0BYQ23@%!Y(`@-Bzs_%IfXIM^xxam-R!CM0NhMbmiPF-R;NZzs2
zbI9t}DLT=sb(c8a-|!b9$U7Yncl#B0xiL_E6E;?_^6qi+3ADVMmlIX;xYX!`etjv+
z&GJ2X{)6o+nG*g)@&$&QYJZwDM>FsCo-4R*6U`o0zjJHtY!ij|=V{@eHJeVqw7+ND
z)6Uq#Y{IGuCgp#<=y}np!Koi{J=eypNH@~#*;)xwv9<^J+(n2Z%Rl;1YmaB;nn;%J
z`%drg&Gn$9)#G;G{CHuouLC};#O%~iz7KmBQ5G6|Qz;h?t}XyYcHT5hXP0VJA8s6r
z?W~jeF0ouI<;}fV@E4gT2v@T2@6nO3E$rUh%FSAuRlSIuTZT{RGB_|sJ{S=`YJ3-{
z02*X<`Hef(j(rQ%IByjM^%{^plSln-JIngJSzb~X=iIULZL-aBG;@W%1Kc9VH~6O3
zWGU(OJW(=xA<Kf(jsuMmZ4FK??T_r<`={IUH}=-**vlItO$IC8ud4J!Rw@I=0n1_j
z!eJ=P%|XLTXzN;3XisruVqFk{sGe`0$(y3~SCq2CQ;nUsboVg#-S-2b5|rCL)pL<W
z#I>`yYQzfpT6Pg)4HCoNYp)DVmt{IE-OK>y3i8@R({-i(CW0>B3PlC)%#@0g$TXk6
z*rt6Zj3~^~48;`4xu+M^^ntVqTEo%SuLrKBO|@Ta3Yo>16dBlg7Ezn5`=rEHw_8p9
z%}2^~Il0W9?Um$vP<~8|<>{+MYiiy987}PT?;x^nxe)Oy;Y3LBYxsz{wC(o1>qpA<
zx1=P|R&+}trx?;>XDu~AqkrL9?l>+BIp_#*@8Rrnv2D@iFAN^n32|q<SI82S&{Zs;
z$teQ+&!%{5k0Rwsq7U4~xpbLpT7d3w2-tV#d7?d{Fwk|!Gd7W2b9Bx{Z;=m_U~bxq
z?R3)D-y=er;z=xhvXmliXi?258k`iAi|;M-#i*PW;*S@luVo$PuClN|E~@<>xhPJ2
z90N+T$P_q|D4!<liqL(z>_uSwTl%CXdbks}+y6Spj_@YXMg_ZplU!ZTSuIDb@VD#T
z@=t-o?+1_c9vHjmHmwPc^2s1BqT-IM%eyV_znLq%Fnt+DDH2?mto_6q<okB(P9n2Y
zniJWoFL}2(%s%w#5fq2+Z3H%RJ;R*{&b>DmR#>n%MIi*~%=A{l9z~5#Kidq~8aVF<
zrks}k)tMyt_9=D^ZX{8<dnb|r$YuSiM1@f$Q!5M4d4R97#S^kec!s{*1rUJTuZSk4
z+@qe0;;~QPM(RlQZ8XSD&*@J@KgG@y%|&2Mcs@h1RlQ>|QCo#2)jovM*FO*@0nNVg
z%+JnHjs<QP{-#J7r|0E4X=;p>+fD1-`&+?ROxI&fj-xft7_J15=UF#UPD#Oxz~uKo
zE_#!iQu_xEa`jd6_!CeSYnM#MLAi}^&pfJsb26fyDs}9iZTqTZN_0XeBNSH&r6yW5
zOf+1e#69`<{*Ld+qFgG-r7Gd2kICS3;VWp>@)>D;ek^RYzoTw_;>?LH)%9@bbR!T$
z@%&3LRf$9OTOoeQvRM6*^_QH7sG`WU_jM$F_&b4<i;WZ|#!Y6t`!-#Cudh>k!#um%
z;G^jeWw*hK<+V~~loo3~<G4pB$iDX-aQ>$!gI}|Q9C!7)pTbTtO1XAp^s60X<0*xk
z@<*l!n|?R4aA||jig)8qkLn!hd|Pw_KPKqU8R~s`sCn7)VxCm<k$cLx((*I=WlxY-
zxgy0rAzC`;dSkwX=>zkO0)`-kw&Sk+k{81KsP01&na`t*T51=k_7n%v0XW3wh(%dP
z{eJzr%O0Ux8@`^z-8pIR&ZHYbc<54P7S~IQEyW|KNBDi`W<*e>-&KHKI0y{`<?0nk
zDysBI^|2yCje?&Rd((Ecf==4VaHxz7?-i_1{0Gs3AXB#L7RQatj3PrZhJjjydrt3e
zS1c2b`dx`9GRRgQXdQx#FoxM+<^D-2e%|xGIrmNj_L#t@rORU1RTv*`e@u(~5qm30
zJa~75OlFNkt#XEO>Tec4D>n<;|HyR8+-5gVIGR3-fLaecF;~7M2yQ_0b-q6FN>>);
zisPUa+05BoD*x6D?*V#Yqnqtk?!=Gr)?#fAarW`X$)rE@T4ntX4S;mIkz9*&StTCz
zU(lFv+I|=xf5JPWoxvfl>>9yPzY$nHNqtPeDsI?!8D6dHGlh;8>So)r%RtHY&<6i5
z^;F_W3T}v&)FuYP(`OUKp`6NPc)P(Uwrf1qEj8PQx`qw$5F?*rEV-@(l_!Mta_f=|
zuu?sr^fPMLmqn*xIAU$WO)tZrKa5!zt36Z~DPpUt3dj4*>ZB1}bkwf=$iaC~%6T6#
ztt2CamRjYOBqVo4Hvq|WZz1;!Lv8u`3BF^`^5b#Yo@WE;w5-z2AqnwcIJPR(Bwrgx
zNL}tTvV@SlR{iC>urvN~C>@k3-`9+Ga(}I1xcAP&fCKk;OAT>n@!PkqZb89~ytvc#
zxThn6I^%R^5_A0N+}MU)=xJD5<E*{cnM1<0m%_Q${@JK$z8~h=f&A8cU8GlJi$AJ(
zLhpoq9iC8=X1d^W2qoXWII{S~Sla+=<i$?$D10scDD_sZ4&wpHuUA;Ddt`a};jW7v
zcOi2VitRcAHVi}`+h7IxX_)(y$lTkl3&5jI;m9o(u5tOi#e907NT_+wIehg-+N^$l
zyL5+eV%&F&J7+f|s|-{k(w$yd|1|%Q36;I|8hV}m1EX<<v9Bx%C1Yzya`~7z#Vcvk
zuB_tuJI?)8m;}Ne{=q(b+=CI*0=pz5z3>)EouNdd(L7(j(*RvV-Ept%nh}(RN568o
zY+0ydsl9R?{hvjRNSOB`tMYk0l{Ww2L+owwVjV0$r(+7&rX4|1599}?OTFSh;&hZ!
zS<?>NCCEQhjpudVf8^+Xl8ZCZpGRj>%W|uy%$c%f2vDKyEt_+Db=V8Y4%7OQ=n~^G
zUSF8dd#Ro3PW=Z2Hek5m)%P(BClbBRrdyqNXO#~ejsdvpv?J>-r>>SJL--xmB<hq;
zHuaK;{t}W30;Af302#{2R3X62rVJ=LeT}0Y74jdZ?lT?u+JGP^1672oyI}z#N^bF|
z)yryD#r(0`li#XcTvRge9ih~++dJ!{p0A4)UvE#E()LI2Wfj<cEnS>(58H~W*P8qM
zU6QYgFz7ck{4Ig;q%!n_weD?9B4u_*YQ6A@$?h5T{U@KGPQhdL!cXDPc53DLLmpb_
zpRS(nADM;aM87Prt%F?Vi*l~6d6qv)U$mQ}>2dX^9&3UZd2A{q+n8xTS&5U%hgd#Y
zN6KpKtp`ni;8dF`UB#jh$}F;Urn5GBuh(nW{tJ2E0}(%iFE_$0o<kdDM|8-+*&K83
zwsrA453JLk?Qz}ym;0>LDktu224SH_hZ89;>4q>EmUUYYW%m2>*Ee$zp{`)OgInuR
zxjL-eqjAEad`tpp3;IaE&czb4>15I~=lop{steDgMy<h}ma*{az1wW&V2^`M4sd?=
zofZ6Y!}jO1Vu@(qYLuLhY-E6rb?7Myzynxb$teE^V{aW5)!Vmy)188ZG$`H8&`3&y
zAV>|;QX<{mElB5(DpG?;cMP4<A>9qq`0Vi)_w`=yv)29m$672lGkfpzJim2(4(`yn
z#ulGS%z^>4`PuZ=A0m$PMUFs`I%)Jb)l7)soH?C;{dpD-iZ8_)VO;n%wC{zDkB%c*
zBI_#FH`s=`pqloax0h;qa<ieA;SZG70~B|@>$*mj_T;WZ%W1bs*TLKp?;YR=dF>~7
zp=KR(jQTql!(7YVG5)P!5-NJLD8V4YeV=_xQRf2vws)jy$)am2+AXr${PNK96$?`U
zn*HIY)7+bn@sjBEW^Zs}ytScTQy)_pPZN7Q1bGgl8n?TM%f^yBjV=m4ffDbR{?087
z$SzI=KUn=>ez8pP<A*$zyRhmcCa)F~u7y+#eT9k$l`7(s9ei}uO;?!bX884JVVSgF
zCgk&(kYdi(gx1AEa`qSuS;xd3Cw*j!O8aSQi~JRv>HxP{)aQG3ihEBT{R#P7W>nV@
z3T+l!dBqg!$A5xY4-~TC(P@YQCsyrj`xan$$V@U6>w$jXz)`JmJ`=MwYMa>|Yds}T
ze@18^OCgMJS!N*{Z#^Xi(ceC~#<>^rWp)_1<7<vjk5QcdsKVeRXJGAmFX7B!fD<cs
zrfEPMYx$9J4o$>ArRE7{1zS1j8bJX%x?)o!*I*r8HAeEYKwl!p@?G1R+0=6Q_{RV$
zuY_}}^=Iyc(;sIrMcT#F$@L@TU!N26p?wS0HX$-Soa1`=m561yP`l%?z@AWYy*^fS
z8M3Cjy{W(x(~fENaZ$Ne*8wxbQ_bY>pPDRDFsFNG<fXWx0nuakJeqw-t-`!ehp{Yv
zYN|jo=|SMd*Fpy61BP2$k|793-uDcSa~k=p2zkq6{vtsK-Q5EFuITOVQBm$qYQY%Z
zX-dIiC2u_U6U+LstvLPJnDP;f5x+&4anbB&dU%@wy89z>B3ZOr(trlqyS;N03OYb;
zMs@{Q6VvhXDHk_Z=D86#vP)x3+Cui7wzA(xvIHqI2GLKZ64TEKYq&NW%IYb|dFa_9
zl1_QK+z5@H+G57wp62(p2-CPPo@vgDY`o1xgCs_YFUOd+wF?$2%<u2nqRJyhZHLX{
z!T5GbKuN@<1~D{SgXu|}xc8#F3z9S^rFxso70wP%*=g3&p9xN{jn!^94nn~-xX+3-
zr@zk2&$(X9cv6V8Y>CFCCk^t9zqYk~q=c`2%~#97aja~Yr2H!5rvCgZ!_m0HZjZF2
z*_4KF%-cdxYAOJ_b)cv&AS_^5>#dpU&QP8}r20x&Q=rcnR~wl}ByN@mKqk2|o++8^
z?JSH25|Ejpfct8nq#?|3c8@#K#(0K>84`!l-OZ{2UM*gr3O<bPSGT}S(sY6FwwO7(
zrH&(fz=Qs!;+F%Q0eKz8tLMyVa#BnfkSv~Go+~s@>#nzQmzA815^Fui(N^LEX)J6S
zUwLWefb!{Olzy3_um>f#d-}ae_f_ltGd|-<7-)*=uL2%i)hqNbJlY{0lnGGBv$xV6
zF;fy5cv{O|S=#orICfOlKSzDKH2RFWN$(u2f>=KBm`qvPxGMWwkrG&|PX@);X$(nu
zELw+YyDOeI?I?erw(vwnk{3OMkoZPrX*YyXNmyqbF_A!qu0N;x1nUJ_Kg7O+roRX>
zM8aY+X)u4VWg|_VfUCOB5PLHCLN-KCkEE2T7Ai7_cUZ0fk4zL5?Ehp>d_n;)ZA6K_
z$ZOs#l}6Qp%pr`sDCD1#!gy)r6JOP0&+pyQA~7Cx8s;l(%FEqbBvps`O_k-jKPaxf
zM1dh&u5jr-D-aSWwvIY}^$nf{Rj}I5S<|Ra7JW(<P%V7r^k-hm5({Zt#^iqYl>(HT
zFdca-xL(?|*{Ri&#2KIDbJQkdjJC;H#`G1$RpM}81Y~eNajtz`&t-E5Nc@i)kTHC(
z+W<#3noR{3DKMBibDAhq6vXtLI5B8<lOa}!mOdfn?h3D20^6JYXZ$1i0|tf2#vv<~
zTVQx(K`CYSC{2!xiwpXY)4bqp2-F|-$62bbL%g1g_oI;<=!@p7$YU0y^ww>fC5uh_
zrr<^<=+fdn4;V^CHM3JxOmL_BnD()yLp^RlnB-dPDOX>w!0#l5&4+4kd^u8ubX$^U
zN$v7Q+d|=pr0W~$#h+e@3yepu7OF%SSj}JcFTcDbR{e&5j(#4xU4MGR!(F8IzSU&0
zEGfQvFrJ=oOgrr(2a~7Fc8RJ5@;yI=@s7eQhK4MX{THs`;)4E@(gR9;8xkC<Xh`f7
zjLt>OZro9nJ11yCA-zy4yF)n9Gh|oUMR~-^U*Qwc?BzJH4vQ}{ZUoeW#>&({blAZN
zHTTk>3dUW*9jifO*{gxguc{0vO8B`hcQHGo*H!t9qJCTqcLH=r84<_@2+4GruBSb@
zFmx38sT{r3=MJ^fS8JjK9yWWkZLU-0k9uakfvUt>R9no`yRorx!ptxAS_|pte9SzT
zh@m6I`d4f|$Nr?z>~}n$ELM&H5!d$RXP~*_lZm<BG9uvV{IY{aRCGwi$>z~uOchs(
z^bPDtW2I;kGtgZU^#=gpSkIo2u8o%cdtj=8kSyi6*I*ON>Zh7bvYE;YI{OXu+AhTQ
zY(!UM=@D_H#>-io36hnH$cMXDmiyno6^yd10^;QiZ@AoeeqoD6Umz<!sG^r`mjh~X
zwRLl-tOR$tw33?VN$M-udanAGE8{l7y{2Fufu(8R0_d7^gucl3hMMSE88)|w;48JA
z_bXf*_;lZx`IS0YqWnh%_|}e63eR4Qo>RNFEr@&S*Qj2Z4Gz>D!nx_ySqiV2v6E$(
z-!Gk}d_P}sq^Gfo=}k*rY?VtgC2?nVP7d=#assM+rXxpxAH{Q$*q<hk_GPRzeVF^J
z_#HTPqa7{{DRc<K-;j&6ISgn%CvpvPoKkwOpP`WWiroV`I)L=Xcq}KDCY0v1dC<!t
zshW|kVAj{p;X}2O&B2TzTYVXve1olou+zD3tYB>VgN`PHO#5k(e{20uZjuK@gO3j-
zu@6n3%WMjK*9Xe_uV40Ou<lHV{@J<GV-QHfXJOhk?a@4pXAZEvPLxe%_NbHI4zWO7
zDcH4NpkG>jXuU$r<&&M@7CwqJ11Tid7wG2rf~D#6jl*mnjqMGhH<$aAgwgB?acH5e
z*NSVnJ)9QKZ^}5UFzu7M-_nS@Sg@C{Ek9^dIdkUT{Nh320n9;-YY)$!PktimO2Y8p
z9RFS;Mi!pwJ+p6xP96W6dMzV9klg+_kRe`doOB4?zD4#D#j`MGa1@2-OS_%tOW0Fb
z!KTCH9{H4PIzFF2<=+RW=T3bCysxGc&OH6{R?1(B=9Rokez{E9(RT+UCK?M@2jX;0
zX)nO?g3EAC^@L9XOoZSm0t_{dZwxyc_&RMNKgO50-%3c>&fVr3>n2Tnpp`IPtQra!
zrs~A2@cQw^3s#p<#2oJDl>(`W)nbZTqIo<35#BW{VE<{({VRl^XERMYKeAy+s3-Qg
zcQXYrx1adx`8-4=KjwHkr0EM_RmT(KR)r`ks-S42-kR_*Lm-!H%=l^odh;0R41^R*
z^sv2edNBb`qNixj96oDh&zaDJk~~=d+^L6;zeJ+!G;ODbW|f6X!mQYR-podH2h@-a
z$o>YVXHP=a-DujX<x2^yboJZcD9r8j1cctIkVPTMgxyvq?&!h1I|5Mo6bZOPqTN6Q
zG|`Jq__O$zSnPH}s>EQafJncVqz78EH;)9e&oaAdbEg58_q6I!L0w<wtyR7t2cFgx
zrsL|*v@Ok3jo}2uN`zqL`h5h<Zh|C7l@YC!LWMuem%D-?A$1w&9o^*9+1Pz{?1ZA3
z4LSb^n@c-MS;^4V$8sp<lIHKnI8|>@-RAPPm4n|dHBTr1t~-*=fig>*gqLj@C*?c%
z=ZqtfVDc%F)Ik)tc7$4!n41T<M7w4<&{jUQL|s(}Mm|1Nr#<FoB)9yK?(3gsR@r@z
z`-1eH)q-?AkEhv57}Z%aQDgnKV>F47R&LhmAlLPZryGW4#bqZ$Q7Q-ua7RVkBmHue
z+?Fityaew*peZV{SN@bon<lvS3JxNVc7<AvxLL@1rJ7V0mz!#3hcb+-i1wlF;!b}`
z1$r~Fj@3u8?_-KMS0w>1YeSh*fu3a&itfI31=t%H_pTk6P~>vpZ;O0yg%f%jR2$iW
z?tGQjW0i<Ian7u9mLKT@1RqJhyG(12us8yODl+N&LL;e=V6iN!--$U5uJjw99Dt#`
z$?NWB%)%aE=71r`l)|r?M6Pse@&Hb=9l$LzP1GWA$C<5#@r3X%ux?aXt&J3C3L-4A
z|K#}&>qlM+v{q~laQ*0G4N)Vo%H}`zGx9`TVh7RMzWW{UUXYD#TQYsR(xYV<wC8g0
zVe?jX02qLDTc9VpDu^|o=oFgA8sw-Fs|R>%*M~D=EUCA8?O?pHy)S?gp@E6TwU_vu
z#l|mo?<=HUU?c#$e^=TpiuE2p{Vhb&QD$n=o<<z?<w^(8TEJ48V*9hrkjEmtPj?bM
z`(g(+qn@g*EL-Lp{SN80Ov1<Rs8ip-zYN;;*Jrj=ndOp7la?G3{^;~n_W5QVJm2jd
zvP8Y9gtW~>*JC<25eqJPn9*G3@hLj|9#yW?EREUM5Tc1RZ(FEm2ajdcFEHmHc3d$p
zzdfC;X(tj(E3y2oS80hiGN+_ps?8L^NBD3ouR(I-31;=Snmi?&2C51bv3tov)qtA}
zo|8~;O>niO3wfC{;5Nt@p|83gUoYhR+ME$YDER8Dm2pAo*`RNlb8PkX`-VO?RWNV1
z(<HpV@dYGXUp$BI<#sN8-$6TI;(Dt2Ytht#(Tl8yzRrr8Sns)2QvhnYk2P?>`&&)I
z<Lw6iE9|nbdC>|-aV?~s2zKmJBOwx~3v)Pjtd{#B#@zGq&OU&SqONJcTbn#N$8=9t
zU##GbK8bqv-nQawNS}u@K?&qjgQ{0=?6E?|;kR!f*v&E5qrmci_L?r)8=HvEd!#dL
z#XzylDaKpz3U3*(u{>CU*D<!J3QSZFCK8AUp)PTiZyyXrHxOgK3sYSW)}O$ZzryZG
zzO=GS_&F_H70UCQPg>_ng4CXlZAM0odu7yf6G+yvZ}*|<mEG6iVB+5`ELXvV`*$P6
zWbHHU4{;nSXbfT~^Qz=AE~Iv*yfd==3jnOS9drsVYy%@#fs@w2UE#w*ir)`p&YCiG
z8o@83r9k$CbGglc1}2!p5Jmd~inap^1qKSmx$Dw*CGASo88G%(V9o40A6Y(an&><j
z8`7SPpKY~AnYI!24uDy?^oP{hX%VZ8tz8w?LCFejccw`OgIrg0)H2Ue4F+)p^RPFq
z7ArK>iYGAviyeP%UQ;iQLdRfHk3c%7zzPm5^u3TE*V=YaA^1fZSXM>-3rCCVDv(&f
z=SYm@+LBk>F)*pF?HhZ7IA+kHuh$dOkT_@5cECg<G7)xL4#GFv1)cDb+oMq$ij|>_
zT77%mT5BphbPewD+RNZ;#q9xn;hzYp1LuzNl~{u5Sb}YMwPZAQGLK`33GGbj(%y5V
zUF3ng3Vis#4FgRapBcr=rQdTX?GvUB=uP;1?~`*&{N=eZ<TZA|ndK!WOMLHNachNK
z8pYVgXb=rXC9ff-oL|Eywkx~IXl@JLGj9YS$?>X+Ouoe&Sl3q}HQa7sl{a0TU=r*g
zmS;Zu06sFYEr^f<bP?$tV`RrCYgP6k2OYoid-kUtt(6MGUqPy;cem$TlbV@!q!O8&
zPsSM2Zn1ikP6*eqc8m3sj(KRV-x$h@S&DP-0|S#gI*#}?*r$$DtU-i*9gF~6c@Rr(
zi=2TywBC6OjX~S9$&$Q^uV+u}<r%_)UB`}@((WU(Ck%NWH=ocaaiJd`$K>r@gjQKo
z9CQ*#8e7HMk;b^}Tm;wlU1c~uHPbuhL5lyr>{FWwLpKxEu0}Utet+FYMnWD~mZwHr
zLZ+U1+Bc8Bb*%wJ`LDDO#B$3uM`GI^4~E)te~DU>u#ud!H&8qS4SA&TaPck>m$s8=
zHg1yCIBlX83m@F@pY|p@8K+hY*_krT(5wE^+Dda0JqY6YR<DIOsokPJdyKy6vo~WF
zf6&w^O#~Dw^#pL6mP49&Lvz3TG7zAfKdDC3?{hVe;dp&(7;#(k%v=NOyjivX$>t{;
zYDG}X^^jxAaJ=@DjJs#8>q(wu%O%Ts#<Y$s&YLcj3~UKTHM2gYpyIsT!h60g3qRFt
z^930b`R1o&`g2<!21m$qarlZ$$A2Gx1!kU;s*BZbX=z+LfC4dSXS(sP2reqXeiy5@
za)0gZc~J-;f)Jhwqn^$zvAw?Id$2NAm$V3>p8m?~W{%Te_kkx*|L)=;<)Os{>?4~y
zGLyrDbaaY$FXCn1Q?cy_T`MNeO|%QL_d9FJI$^v)&QU<Oqv$i5_5~(L(HO!|pu$_z
z<BfNMiTExfgTDvIWBr*Z{6V#`nk#_j;<WxngWz^*{ij44-z>@PS*QAurcOaLv?-dj
zicMsTu7!`TlnmfaNB7~qNzAi=pABfc0ru59UFqVLxHRP@WG{uFM=x38ody*I)$M62
zYQ|qkkQxEBLUg~Vdio2r^bfhMe8yHe0~z?X(i4QEI<xXxxt(O`vs()=rgQv3wuD!M
z1F^?j)lO;r5Cxeqo>zlgo@tj`8&(+!@D?;fllSlL*C(a<pMJIbJQ36L=D?O;IF*QK
zsbQY%Q_%D=q|mh-$1mu7!;$`ro0;3uYH>mE(p7v;*JR-$5er_Q<qhN^w0PU~a{4<&
zQhGQKS+><mb~k|iEgN^wbN>Kp)CU=f1)#{_w-}K!912_J0@2i}5jPKfW7SLP*4YA?
znellL#w<(;WfboU(wH8EKlEUv7A;dw4lA2buF8*h&RZ_UmFEhGO=0M-L1Q;pnvZ-d
z;=lf!tCUapDx?Z)Y;ejuvuF+n85H^tb_6WVFBCR%W?qoM$C{$xk}w6U@Kb7}HfF>k
zw5PmCNCSpOv8P^EbUe*nKXc@+3$4NVxmR=u<;}n59>Te$+ob~bduY+52UiiJ66~Z*
zPY>N&=h3-$KuI*GZ)0{``uyo=coT=rz`8^YAir@GC8TPwpj6s(0w~6kY+k3;qZAmP
zg_9)@OtKS#KxBqy`!i}hJk;$ST{HVHs74Bt8*iI9U<HunA(e%r#!P`0sPcsTX&_M&
z?>rr$-PVp2Vy=4T@LzKWj}b|>$r1Ph*|C_9cQX5vHF18?Igc?_wc%D}p}Mb&6VVeN
z5ESd-Fk{2Ko#)qNd6z=pMZCvPkMAJq5DJk&>Y`UXR}BN*i|hnMp!dZ_>a_+01xaAp
zMIrez`#xDi?qLS^2?t;j-P0DLD(<PAi*UY1V2N^{v`uDNLnA%|cj_T(k;<ywfJuS}
zKO51l#K&R~9H7ObD-wBVeBa)>MJ1*opd)9hi~y5#-iR;N_s(EGqcr2(Wxj}M;?`xL
z>H5~s`bC)MY2$1V4~AKUT|i8~@KaQhQjNo)$N9@wjLq5`sY9%mi=36Sp|z${jbaBX
zUgQ%A=n*EEM?fEk{!X);Ig>10n@cHPU-0Y~Ru1!fpy%OZ{umeb7zUbDCB|CMK3%(I
z#vMX+(dYFMGb&OeFx5V2;&)<L$m7U!smgA_4;3T*hXtUf^2|QQHQlUoyKj>k>bF`Q
z#M_S+(O~t&i3legbI^v?m=u-7iVi_rLDNo?B<4E{_Abg!Ub)Oee>f0*cE4i6Rqgxo
z)RwdL?tQNNs9J)+x%auL<#mrx_QsYG5GpFV3{qec9MST9|Dpc$?yU9P0LIpXLa8l*
zvWs$0nmJioK!6x6^MeP~9_bBBE6XVL3{L6|8O8)N(+;*4*;BqRza(oyKmWa*DU#{j
zQ<3Ux3*3ACB|Fe21P>2i#f5@4UAB$%r8eV5L@0nXW(@~=yh_nJ*pPd9u#M%B<H>ze
zDIGC{)0yyjJ9VeTBVO{xaPuB8G5bJ-PZw?>WKqOlm6)>^vz7);fA%EzDADbF+%P<#
zGwY*-q~xUoJ&k=L4jGpY-5rk~WLY(kH~ljYt(lr2f~JGT8TVnn2{^V;r9yV@%T%Nj
zfGtp#DE^|x?P%U&@6f63N_jPq#lz*EL2HS7_bwNGw6`h|Ymu<a=-hFevUDxxjE<GQ
zb@XGz@iaON%~62PHly^FYs7oiZ}Cb1Ky|smBn$&D(q2RAC^8GY<|tV6gVd^3bE6JV
ziDT#Wm@T)U%u&C*8Z2@9YIFeNYzsQ3a&g~oSs^7xox&KckirT%7Emb#l`A<-xAS(7
zb#d7OlTU-4D$Cm$pXYtdDKY&Eqd`$Ch%V#X!Q|1Tgbx`B?yhTKaSUqQhUr<TRnpyf
z+YRx(2!)R96<es1P>2+;b)XkB<5wrmn!YRMQz1FfBQ9W|c*b2i!iJ^sHZK7z#L$cL
zK*!2$ZJLNkH$oHWSb~=xiHQAiHlaDUc^HLWy$gJj+g&7v)~5yj<vZSkKu4iykm8ix
z{Aweu3AznpPTt_TrM3tLKO3SfkV1L|!%ILu4OC)B#`DAcl+{kPe`UrSrWGj6O!H?%
zGl1ZV3(1o_qxV@N$?5dj7R>O*@<#WjP8OwOS1)Hh!9UgcMizqs=C@STDwpFt_{}^v
zYR@8S9pVw*tou!p$?qm@a{Bz1T9{jctk8$46|5vV@{|T=8~wWAh#GCgZ-;Edc-H`G
zV~LFVRI?xCo$b*cB;lFHZMR5Y&VN|Ig&uB-yc3=$|GGWtoROUzpO`~9KxD)c`N%xi
z;!Z9@i_Fp|DTnHURICqqXH~Q?(<`CJDIr=5kTsfP`tOL|X-ySOc7v{gv3gqUB8``l
z6V27jtnk5Thb@a8iVaX2y=Y?m;N$TtYLTR`71zSin`qZo<09rP<{J7^OV9Tq_)>AF
zc5h2y$9}GbYT2;jlDAJXl3z`-&As^`!2Vi#)FQFe?r;of(lF+SnDablV`E;rAv}Oy
z3HYjb2FQK3nnbM3lfd~*#KNJ|L;NXM(TVtiCmH{2fH*8Mon`4|#YAK0o$k&^W{QB$
zd@uJk#ktC7<lvk&zszdOw4*)pKdt(vAcT9`6hJgz$r$_*6v1KlGXTScIAP1K%#<&;
zK9!BiOp4b0Oxi6cIX2!eE!Gg=8h_vOf?f)T9e2yWWB*m9XlUNdOem;#WkrOIA}N?}
zhMI%Bq{n4m-Y=YV2|fE!d^$$bm$%y?tP86~VIx^B>0o`brp{i^a~ZIm&SYP78X8ky
z;Q1EZNl6`DogWF}!2=-31HCRkFr9}EsEL@9Kq;kKqPd*d{^3+^Y2H5qU~OS*kcJ_{
zxa;<pL=UC5+u<`9=^q_u-2^7$G7|*A<Fd`4pI_O2_EiC`;4flkB92j+GMD7W<;FtJ
z`3NM=o<eJHRvQkiUVeCT&*W?>=(>fh$5RXf;KPe&Xd#kK*8Eh*B!Rss=ZXmrQ8G^?
zt99dJXyhMWQcV{g#Yb6qRqQMSO#sy&_T%^~@lJ3@UZXUsb$W}ykB7X<IiXy5o>eON
zBYPgA^uCjF?C(}_Sshm@Fg;%h;R!RADX$h0nPFN4iV$Wd%~o>bti->Uzr-SC`Z|={
z{dG?U+(QS(mHJ0h`bR1Y?jYo5!yF@y9%Gat${~8@uZLnsYQx_>QW^jKG>b#3`U}Gx
zUWe}Lt3ny@I3f3!f-#KtvE=eG_GJ|>w`4!~L;iD}JUNs>EtMMFA7dnL1&MbM`_f2x
zdX{^Le(nWNnEoQ0?<N4mKJ$oKFu94*M~Jf&=xjCkm0<rS(sHPGD@$%<X4Q&cx&7o@
zg|B@L>Zx_+*c}hUnH)5P#vYM(t~h-7!j4jPqm%u|^+Ffy!|ZlrTrBsx8MHhaSxxa0
zF6Tv5zgVj6)>X$pd&aOV$DsO4VzT2gbeGb;Q&S8OuHN)Gn1@<Z9IpX7&L3xnKylP1
z!F`n#TV6i)e%Er$yCAu@Zo{l$7v;=#uFg{-qL!r=V`^Anu7IQ5eHe<4FMgho`iYPf
znp(0y6CpPyXfHQEs(Yn6pG7>RT2nHH;SZFsjra%j9`KG{VajL3%G<?R771i*MJvu{
zkjfBMqh$fRx8*ZIPCzC<CTJ6j<EwqpsdzKSzI1Uf;oDSS%I;gNCecq%C1*Le8E75S
z>Ty1%kSa%N&)`-Qi#8^>(=B>fSFlwNyjwZ=u^y(AZVs805PjY~gAzN2-~l|pc?>vw
zmLJ>SDZvf|X!i>e2O;{>RIKC$laUY%h3;zRIL7jHg<k8z+I+}ch{G8jw1(`Jo@r=@
zmuco}h<@Q&m1nD)#LFOM>fMs?@2W*(3)fzfN{gmAmH?zb?XTTm&6Gmb2j?xxaN~HY
zIM0`#V-dYM_zZUzju(`CCe^C!Eu5B57p-%cymQ7->9~^!+fFlr{?;4ACrd)wLg?EH
zbLA+Z#6Od~p|srjBAV3-a#A6!nuHT)GT>c8Zc^{^EUyhV6Y`<mwMix^6JmG^_)iyz
z5e*SKydUiX6k6kwHV2E0musMv-@6_u%R*fq53z#7QOu^~R)JzE(a>9Yep_)hdVX6{
zwa?7-1jXIYVXT0~vBlFRJ2={DHA}=9qCXV1NzLy4Wsj_y#+;IbZ0Lv4R*wF0gk`#T
z#>eFj)H=q~cKR>+_ot4z0OVdAJL_qf6)keQPNlUm4i>2`-L8~G--EOwA8EJIbe*HF
zE68(@rjpwxC0ShjXS=)pPdh#-7gGk>LT16N1;Dn>&-h$b(PSO5I#*xXDd<&-hQ@cb
zv?8druC>_jZD!#Sw8}gj+AysOMeL^^d`b<d*FNB%3Q)PyuV6~!eV~@5c2U?=t0;NM
zEGe^|YJk*FM{iP#HP2lr8-xnO<h2z8VHw2M_?9Oy3)I5zf<gKpPaEt>lp1%5&&s}(
zH}nh1a(<iCGCqPEi*>l+kxVw*QG{cth&BWI?Z$EjzA(&9t#M51F2tQ;j?)otoAfBF
zjAVhr?~z`v`&GSOi+MR)n5mb5P_od(V<K~jI2vQcbspU{7lay!3XyPS<2veNm`)rb
z$d0~wQ(2mXzf{4-M|Y={OKvcFG*MD0RsZ}nWCmk)q-<Ue_Aop8(j+Sux?5E*eDHQe
z?jY+u8v{Xsa}kY@ttH`g8gB9ypa&!I40BQ^JmI;$EA2fGkJ*d3{p#rZWh)q$_&Q;I
z_It|(2aTDu-~ds<?o{M?ETe^`1Wjj7N!9Zg*r!ViGj9Fqyj<9>du~zF3w@1QBaix-
zxu8)ynZ`4U(k7!7fX-3={+Yj}-27U}2r@eK?fhfh(>Pp^jaUxT-H(s%K4p=@gGe6y
z%1LAFq4H3?r5|{S3^k(t)9eObet=QdCiQI2knq}*o+R){*dA%qPgL>u!naLvp=~P9
z^B$+ROyv>(wJHAn>b{n0b2~sa5ZmdBA@|H|YK(L<<c6C_Gt^HV$QA}iN+C=mi0)Sg
zr`8jSFH$ja3Sn4kb}ml+Y$-knh~`-!Oh4tCk<qG@Q&Pt+XLz?2*vJJBaAYb`-@BW9
zUkRfK(}!AOK7uc>zv&0b^x){7b`s7-+%Niw?@4v*!uLtxkQnz!3thDZ@hO0mPi|Mm
zMV3GCJ7#p_@R1XCIJSyGA-<(+-N28>hdfUDy}kxDI0NiZl}b55cO}bchjvBo6(-5+
z$mL#N8qUcmB!~KUG<GESK5+?~&+0C^FLF7`JPhTIru%w3@g?^;k7A?<?WDQWHa>#Q
zQQgg&fVo<4tyqHfLl-BizAKzpR&~&Cq33G4?@*&C_ziKag7s;f-21Mpx!)u=o(@Q4
z#<f$V4x&_=<EhWqkjF$2{sI_k?j<VqD`-QCDJw=tUay!qe3m1ya@3C@<z!?0X*#T2
z?uNV+bkE@3ph$03?q1w^)-PqN7qPs>4g~K$URA~W4jNUEuXIfG`~12)8@ar_vxSKD
z(9y6wpU`9)|9&k1k!m}5df51hbW+(ch%a1-H+qkZS=3Pc5RmPayTe*?<A$oWXB=Da
zo3b@6Yk1aU6IJfwk4EES>XMCZ_DSzM0h#<BZ|u>NPwhr*AN+Qk5($I#6`e(1cz0WX
zTe8<)FThMV<Fz^>rWMDt1(V9xc+<QLo<hg@?w-ufE>@J-0%?a9_j?ymwo|KWe+G)$
zqE{$or$s}eSCiV4lVKwX?b>rf7&-LqrWgB7fluj#M1lp3pTQT^P@}x%X1BfU+)Gqx
zxgB+Pq276MgYm#&<KwHtooLTmHHnOb^uqQgVif2xsU32_fZcd>w8uK~Jrmgx;0IfN
zYo?MYqwVVyRWZ!9h70Z5w8&j$dAkLZoZ~8&I}4P^IR%SZcKP=PfkLp2-hsuWC8Na5
z*-cYi&op03DEM*bnBIkyS*pNw{2o;O4X-W5E_c-ZL1cuSa6mn0YhrW^YD~bKqU2oI
zU>TOm6H$#8lw+PZ23B_Mb!PxN-n}WRMf013)!G{|yF!mtE-wG;fo4AqQoYQ4M}g?k
zCI4~Bly7HtJ_VF(f(z`{XbYAMRxqx$-_T*u7b4slI4V2`E$)od*RwLNNtxy|1-dKA
zWtS!9*gWBN-g4~_lY}6O)f8#z5jbk0cIf$`e5RA1#HcaZKyd@&LkzC<8WvEVl>&|-
zILAog-*p7;p=~qY23RVG>*(Fc*t{y|v`I6Ecf;aTgJLLYb=9z*^&C`b#&Vm}w~ZK`
zfWOr~HoyHAr5xi<rIGK$Cp(aa%Pd|~ZBpJ<Xqzmpl&A$F&g|gor~f<=Z|D#bR{`Yu
zrQAH=>M%q~d@GLsX&0EYKM|qF9*q&#Beq-_WdC(d6CG^n*;Qcb<Ok?cwqK`Njt!SR
zx|K<<mvXJ9UL%G1vlnK#wt=Uje5+#RYWaV|N}`Vl@MW=~g_HHT?SqD&t``txz+(`P
zY#*uj0ft?USU$**%zDya%67?SglVY1iD}KH-m3%-=~(4>7IUqa*xviKPcPg*mO$Fs
z23ph?-h-L68Xs45$)rQ0E#*d9G0EvDXvNDKc0Xq;PZE-ZA2lAe3=80;smt(!+Z)uf
z0}v2B!{4hTHluDx`_fBM2FkVtIbv-K&{`FZ{uuz*@+&S-D|9R|nd5(|0u5P15t>8Z
zh#d?_z|uqsVm!zSy?_MRaerS^dlxJ#=Np8BteGIJ_fCj3Sxt+@tM%|39Dea&Z~n}6
zi2dEQ`!!>i#VNvdnhW9S!5!SgV%4l+nM-`q0GjqAk-=hw-ehPlohV=?A<`JFx?Keq
zSY9ET!15{|WQT~H90CZO_p5@g$dHWs8(g@~*a%OijYm0k2Wr%+Gmq}ey}oKE#bn7@
zBbrM4Bz<oSkG!aqYjXv3Z7pM2%r6ySA%?I*HBMn4%l{n4<K9vp9Cnw2t9sErEBu8z
zm;vP%fTpws6T@PGaifZ@dcY~ZV}qeONOgqZ$)^&S%DHF^lni;a)iN?(oKUAO;Ba~2
z%&j>xYYAwcHui+t!2%Rt%vvF5DkMazPDTovTMB+Hpd^i@sy@mL)k&ST=~Vhf-8&i0
z>MBZse-WrkX}L1e%L~S#BWfD@=B;P{?MD`ObW2lNw$&QN2HAK^{ltM6stj`vO?BWe
zFTk_zR(Ofs^#(0x-#6|{f1{lkkmWad0Z^q-xPAOtcMoC@QH=DUmt~rLp~0Nd98Qqh
z<7%g|IZ|`m8^0c*WJwBjxQCS8dslW8OMZY*HnhE}v-!_Sgb|n?hA@SuQC<`PL6l=p
zqH?2Q8}vb$sUI!`(dbQPuom$~VLMZeYU)}}1U<s){`{Y-<IhD3=n#~!PASZ8A*f#=
zGh~@&-9h#AfWIM~uCxm+b(zJn0=O!syT<3I(NZvN=*LGnOq?X@bP7^vYz#_;b+kWD
zBu1V=j2aZO*gwc%v0}TPO5e?*I&SuP@*-j9>GM)zSLrK@6k=2A@l!G;^wO&+qLZ}K
zPF5Xbi?HUVn66fws!*e^22^4?u~0;vrdK%Y?4e-qerlQ!I@@UH8??Pv%`aUdF5(l<
z{msD^?*!|0nTixs1IMXF{t7Tgq-V94?2F}v$O=-g6F<#+%ET<X#DKSsy|HMScbI^V
zArIqUr#j{BQ^7!DS)UhqA8{FudF^HJ3(rHU4o&^EwML&NYbRexKdJI9#8~rJvg2op
zN#%3x$zap-mpXIk%l&)X*4VSPAvg$7;{Kaem)5uNt)2z9#&JBREYHq*t44s`D&^aW
zPc!q|=bh)>EV!<IN|OG;NS<%XlJH|lN1;jRu%WK*s{rT9JT+6D8eCq<Wwa7I*yY&5
z^PE7>TCNSR+T{uZra9mMsd_=1dtLQhFb8sz(wrOjW%A&hvlSdPXw@|Av>Tl8dlLPI
zX&h7vyl0e~X*UX<mdI;*yKf|<)GV}7*AA;eujQ6DS|?;Fe|Fy?io7sDQFtglvvFv5
z??YePslKBXnuSR<hGL}WMz2qlI|Jzh0l|tR<HzDt`hVk^|0_NVXs!V`PHZGS4Fj`S
zfvkWC0|CPX*1>s<QnkVPjf#aoL+$_bJ^}!wl?o2?#})sXnEV|#{rB~`M`(}hM|?T$
z07?EK+BE=Pbs69~Q`!7DCWG>l3Cna$?%%`W?@;e8J#B}HD#As>|HIX3QQn=o{hGB3
z2P{Uo0pgx=00yG_eF-Q(>VG1+zdn)F1c3RR=7&W6scruEZ~yDVr)?3ylgCat4eS`e
zt_X;2^2DcaomSvn(&vH3q9e=y`)g{@065fLvr@8u?-JsFf3N@jH#Knpee>({GW`F(
z`v3BA|2)2b{-!h@xX1<1OPqhQzW@FI{`05*bDh@;z)?O$H}zi_&$g!Ewu@<#qjV!c
zPq3A(Lb`sQA5M*hFLo@Y_rC90=Obly&EpVGOhh=dM$uT+9drAC!HOZ$X`ljk?sHzT
z`@f$vaEeeEaN_L%I%wO=`hYQ2Cx9HByAiqoF3b4`=Ywx@GoA|S6S?{s+?kYfUO0R$
zO!NQYEebP1FKV&@C*l>r(i0mvSiNU}jOg|69EnjeBsj;d0~mKBeiLJqNCWs+n!+he
z4aG^n2&|pOXgz-z*RxpuZ%@w}ANbgFow_{#tc}vpHhAXjcV0jfIwUw>rYkma{~ka-
zylVi`iw=M+MF{6#y*ue9Rd?4t2P7xJpoDWLT!KSl?6FccFe@-^NE8R!Ej>Q&#7CbS
z1(+)$DLd>)hfeX{Spa1Wpgj^VayIGes1eKSuDwZeU5#Lf!vLULCc@{V&pQEUGz|uw
zv%@c{DGHB?;cToh0HwkNARK=mPXocANf^*>2Q2PSFx5R|J~D4zyTd)KL!vnwV7I@3
z8zed?wCuA2u=<Sk02i*#6axqH&0PP7cR&Cf2H1{6f^{<20$EQ#Cvj8#|6iAXAKl+C
zeHtT*>{TY?&kOy3ZeuV3+(f0*z4dnS9niwW^KvOWdNtGw*x8Z7SzY0PjX8tZ!$~~;
zD<jt*dRy9&nd~UEhh;!`y3$3Y?lQhySM&hqan=D?nZ<q@`-pSShHc1N%3#O2)KcK&
z8DRhR`Pgr)`}2DUbNT1#jB+Bly(d6Gj7$1M8vNXsVYO`AAodvnbL<Fso!%p{9hm?Q
zQ7bTpcpNyrt`ktQ=soNAyvc^s)K-92`dQVGcJ9sOaha9)!0jbKO!}@?2uGE5!mS5e
z0LWZJ=qQuKNo3rsKX>MB4yUYz@!#yL|JqOfXJPdq0k2IDge)23pF7ixVF$3SRsefG
zT=oR@_kfTI@HAtFGcQS296zwNlND1%zb6m}FoIkm(|~@(EUn`=;|^fzg9qfhow9@v
zEoa$BC}p?OrIdn`ABj5sN|v2fry~J!(-vZCB@lqPcnmw|OdL%#yb6$-*w8$VOytT2
zD3fY3q$`f*+Y1nygFApfiE-pL-Po}|3+>d-IVJwSjV^r!$4Lr5T&;D&(S`_!1Rauv
zZFbgXVgOgnVxrLOpD*z5KGUWm{}awU0`pBp8=phcvs5=!|KolT0sL=&B7M6THGggF
zgV-2N8yt=&sK-Dv=Esl%1@_&U%)KV{EknSUVPLfa)|?OAs<Vp??;(*PYA>>_=_`^5
zdWBQMG5mkL9R?;6AN^sfeXQ_#mIbVIwO`(VglYCM`O|)&j3?3rbo<@LmjS!VhDZ5C
z|9;jy9u#V?Z91kKuD>a=*6d44F%d(!G2t`&8UlAK`2HAh(qP9L^<~{>&Fy9$nV$f4
z<#MtNXx5q~e;!Nfke6&C5?do5A-mt;@mhluvaSHMA#qv(d<B^$SC4%dm-oB_dbrpX
z0CKYv@T;N36Io9fd&J+a{srH*_Bi9PvvhBA2;j7OyqLFJ^x>wKx~u{cKNXvX+mDuy
z*)z$e&y|x7fnAgqsP|a|=a$eDW3ShL-%6=37+^xIS8hB#?*NOjFmxIpZpwFUh@s0N
z7~cV09PyGeX(ald03ZPiWt;!~aM>CTp!H_eT;>3F>TB&Lq4aq6l_U6a+1dh#K2-*&
z^uK|9Ne3R}YNaUqJ2oB&y2nS^w!z)3Inb-Fr0%#p`;=d2TmL?A^vBM<9<*A}qQiu9
zXFCm+Z*+feP(Q`!k37`}D7p&}gV{Gyi#`dL(OfMv915$vMA$C?&$?=}IyiFawjYi`
zbuvdk0$%n5PEz<Ikg9)F`W$#(^)w~^`&ZBGS;_!HsjvI*sUdVM{ddnb)f(P4M|#6H
zQcWb9R&X6LQce6k;R0MzlGoPx$xnY5`R3Q%UR3x~;eodEDhD_lLx5_3^B%dHM$7TT
z@B8K77YhT%rJ4B2YotaliH&u4-CTYv4&I+eY|HD5*MxLP9!k9d0Iks<aHv1?uwGh0
ztbP4Pii3Uh&FSY?&8(S#D?G>*%QH!)HV&BNSZ3TnaY~8<c(v_&bW<xZWLL61uS|yq
z8;>1jlh-rDnPo(O*tBKILu?k;a5$p3%}*Yi-kYZv(Sp$i&KRG$d{(eKfx#TDG<@Rt
z-1~j2W=d;pxb%IIcNfkh{$4JB4Yq$5%~6X1wH$hq7S@2T@`_@k5$NEnDJ*@6hwClG
z08;Bw|0CBF!BnL8;L#uhoemnTMK0eR<)z_7N#>Vu=i}7IWbK^0lyQD-zlx^Q7e`;*
zfrR8ZAKw?Y=d8J;hXJRNb^_%nrPQ&$B1f;<$P+0Y&yV8?etR&LG7__~V}M(13^ylf
zrq*&K1qjFi24**laG2$ec4+G)U5_lVRev0yG)Fn2XD~jAdgVc*KF>5d!l&y6Afo1p
zUOm>IAc{0Xh=+eS&Tta|%yv=p9pdit_nu%Wh?H;_50vC9fePwvmaXMht&scq55Qfe
zeB=xe${xZo(KmNdei^X@*zA>3N6PhVG0S5Nqsx4)`K?xm!<H;Jcc#c^m_4X>`HnL-
zCD~{#{e}l{9C6#0N5FOW0PrC4bEv+Dw*aEQTVe5{JplcOr|vbLG;U+tI@n8g%S)XV
zr%|M4KH>TM{xTe(y0Y8><W(04Wdri&JRF9H-J3P*J`XDXS{WE|JnY`$oY#`PH5$n6
zZVn}1bL#nNTUI4M1EiSQid}pbfr7q3WxRp&^M3l4#AQ7v^1z`95GEl~$>>W}R0B6M
z{IRZH{yt&rlBH*s$7-a*M$R>#*ShDk!4QeE7ov_@zl45-7lF`OQr+T&p1Ry_PmE<x
z7>`{8jJ{1|-%&saLt}m9!4LM4Y8cb35*)4Y71F2H_xTK-?W$wUM5FpE5Rr0*$GMa#
zQA+>p7xpN6O^ZOu&5#^!uZ-rV^0m63HS`41CUc@q0^AyZ5&eWvH<CA+ZTWRwZ}Fa@
z4{NqSF&578rWnuKktCyp7Rn(SAqWWRp_oGb-6a4~*K{~bMC}D@N6=QB#LYHvijsOP
z(KS_G-2=#Hkvw9vaIqKeiWusCr%BO}QeRxsFh{S6OhZCN{k-IWnux{`puD;8v58<%
zk!=LR4mU>>uh>~rT}KXThre?s>_5G2E>D5YlTDoh8=s&FXiZ&;kR)u-W=Zoy%B6~f
z^3Vcp4j1ZKY#wfb{cV}_Rs`7W{nng-LOZq5$h&b_30lFGN2kEfd!C&LNBFW?ewy~l
zY>3rD>E?K`W#_R*vAkXPoB-~CoDG)t?MHVjg}W}Wed`#Nl_>QGqx8jrl+xirzf%Xw
z$q`_Qq4xUid=Zy@;AwD;Qo1&mUI3&_KkLPzwT(AmEp0>1`l?oXU3!y1i>eh6<i3a$
z7n?53%3Gjggp8s*A8+~@gblKJlGXmp-qAaAZRk5R{}V_$3$DpsOIz&~VCnCL(*<|N
z+e~1s?39(I3gYIyywRAGj@(Ot-*pGD81^p2D9BGkBm%cI$YdI%$zP<)j9W-!rYlKG
zft)Jlbkl|fvC>J|oCrQ)Oj@mT9k5W8LSY2Ec(TkSTtJMXZcei!QZ92<fO`PUuXCR1
zO9f4$CZ@S)>7FYkENq8czRgn?0jShjZBaglq}#~)lk$ffnA5Jk$UdpvK-kBcx!J&*
z<b{TaK(74zIYo&&q?~)nbOBzYY$%B>;ipAUK!j=BV69U{2PRyS3DKh`_{VT1(A9YM
zNCBOaQdZk)r@}S3AM}~51PtFS{P8P8p2exZ6UZ1Iq~)s~p$=7{#YrW)HF?<9tX~?>
zQ?F3sG$B6s!d4b+1xlxoJwbTyWn4AJ;z;erm1KO3tFkJ(irSanSMW@o<ko5==fE6x
zw;5*FLjJp;R|;K*25$_1ybTCBjiv)B0Yaazl3QUQE)zk<I3s;eZcJ4e-6_vyFNAcN
zgP|=^es?%gfqF-_!tRD?;|eGrVuJg`K3tPJWAyOaqn~RGj?ikT=#&~TX@T0e7mxhD
zJCeu=b@@csGlMt<P5iDpY20j+cdkEOh28}}2Cb9ad<708_DaleCg5PIQ_Ooxiq-_%
z?L=!c0w>xA=@~X`yH!*6R=@)sO^}Eb^?1kXas0_muf(sa@tO}bhtDTDu<cXOdC@7W
zIQx5k*18m1cv84zS3h@up-(Vs|6FP~g`%4Dlo}L8JX(+UnI>s2{D=LPYTrYHuO91{
z^MT?+b>Cg>mMUB-0sl$mISCN&l;RwQD^Q4jB!0*7ePQ*?y9Z@(Y!;*6CyR&F_r0-u
z_UZUb&%&##uR#_>YZ-P=T-Z_mMD;`PY#i#rD)>(b^^HKPFHI4Pxliv-9B3qBt;yzh
z$yPDvjY_o-nqL9YSkV#Lb4>J4h`e~qV3|IMKz~8gw_1p>LjBZBi7Q6zy98d^7g8GS
z9b}?F@QtM(!M5*WWhLl2h@R{Mjzo&j-cP{Qz^W{b>&TRQN9?QTY!;X;cUFx2f|>D>
z8B5~^Ix})8(T`y9)KhV)m-b~<TF^y9*~iv~YpLsg2is(<BOr${tnWrbwvjoJ^STo3
z_aSlNY&w>C<zglhV9WkUiX@c3QzRtD4t#?2o(ORjOLUE{YW?}54*^Xf1q_d!S9&Ul
zU4bf3BA6-Vmew_ahJ>XWaf8Difsz(6ahM*0TgX#Z{a9a?zNO(hz#Ih;7dw=BpsF0B
zJQ=+(mTy?N6Eh`&{1d)jkxb%(%TI%Z#2K3xw~*%iu4li0%2J2|iLF>R^JJ;QHG)m(
zjr1JmB-N6%$butbrIH9@Uq~q9oRGQD99}eYf(cN15*Z&38a@M%b>-{;?kj@$lrgOs
zAu`7865yQk8|~gIKLM-+6qxe!i3{<H-LQsXU*M%zwawtsD0XGUCMQJoMh5koX#3jP
zp1yHP2EL?f!VUuaw3{FDz=jcJS90IFMs;-Dg15@)`UZ%4_WD`h<3&G7C}3tF?y^9+
zJzp&ikzB_Stxb#v8U$8)0I9J#8~X0K=ZSCGtdp95Gay?iR39oeIUS<M!8$NSIw0{(
z!v}Vau2(w7MK%0%d^X<fEv?&fT&JjalRZ0T<ZMz`MgHIu5gO~>bAw6Jvgh>QXRoC4
z@Xf`7e`xja##|&Ez?|7Xj;a@HC`=U!SGVeY>iZy9Iz9OZ9}m2dWd10$%2MoRJgEUH
zRuhVDS$#es58+i$15fzZFKT#TT$>Ns07S5bCmuI;P-wHLEsOkAG!ut5x3W$h+@fNl
zDv8NBXfS}ZhI-?<i5!DF>dKw-#<1^>@Dc5l>wsdgPq?a)-9vkjSZ3f1bNxZ>d>CL_
zTe}a<-dnzz`tD=rwdA?=@k4CdE!dubcZgdT8JHj%Lo|CXI4={2vQ@jec+EZ$Ms@~i
zP5TV=rxVnAAPG04t@+Ab254p~E-$txM%kQvKkp&i$9mE@n{JLxn(X@MgR@lT&2t8O
z9&dJU`=u4+7trCk#OsY(mFp6xX%{%>aX?qfm<c2^-wq(15-OJxR74}82sY^@@94~N
z;;)NY^lfsS6R(kseo5rq)Id33Peym;M_p%he&7G`)FKD-gna)tROz5|xAv^q+<#Gm
z-9YgCd4gFY-+3m1jXnc7XMi%zuQJ9ZGq$APY0(FGM6rBA@{)?VF@uGi5QPm8Z_tp>
ztkbI(w%Qwst6!qvJ47<t{a=1LQrH+XdH+Jzfd}*)C1#K~)<=WILa;sFa*Q>@;*)h@
zyR~R|<0=o*YyL+@HQAeE7)E1dIIfG1r-f<~is1+2v9<50?BCPG(oTNl*RGV)7L(Hs
zl5;j#aE&Ko(3Uj_i719eTiUZ{KuY$=-&HLJjA6KsDXNVThZtSX3-fef_^B7L96&gJ
zDKHet8LS`qM2USkw0oTi3s0PfBqBjrstZ_@K+9`w5wFL}Smm_iKq-YI#EV8Fg+%?G
z81Hv#j;D!>#uNdeSOyF|@dN2@$N~Ljjg=&dgM36(u}5qz5ic<3kmlaUWN$GE#0W8l
z#HwRt_H@h{AFA0flf&ziBjshHif1ldQa&n7M?Y4BV+`cAg*`>9yTfwGV#FQ~aj5ZM
z7I1QTQ8U-QyF2Oi&QJ~1dH4NF%Sc4JtZ+HmSDYp<+XNW0&SBJ^xm-{u-JOgiR(O~u
z*@Wl|4VZ9>l^o7NWNNw_t)oASA0($*m()&(Wzcg!iLBzL)^_kVW-(yMaux3sY71ys
z#*Z#fWzAA?r+C%}&|)(NFd4Q|HrpaP$6)zmF!iDlbH|w1--j$b3h@HIaU5F~oSGTv
z#TQHc^6OV@O>FUGMAL3jFfjj={SCAhSREh#x>yya(x2Y4#NKfJH2UE$w0o0)a+yR6
zALMOK_fC$)G0?k<sjXYy88};$l$c2_>tkFhLufRVt{s)6*x~jz_#@8-3V1yjY2EeK
zD6Q7mn`44xa?sVu358j-Cp5wITj|(OoW${-?D63VXld|&_(`_aa$U&d$QNY7PoFiS
zerocw#gprp$A4_$nRq;2uxEz;j(X}{>UBEdX9yt`gk!;daUdr^^z6<2J3@P+)R81R
z=a>A-*R#DZ;?oWCnfgX_pj<PXC^XJEqA9~Iq7J>aV^h7n_Ct+l*8-RQV~WPt17CBd
zzY;x-tBo%M@26I@b~(!S)ZOfUQJ7DdersHv$Y2$7w|2wMEvH7V#CWSXQf03Zv-zXV
z;-z1co$R3XI>`vr&1AKd!MZ`0jsO{^&=Z!*-Mz0V=+}I2bolpz#stN6)XXxuKFH7S
zli4#Qr-mlp)1T$r<@tD`-3xiioTC66?5kgi9c%x&SRZ4ZnXF^25zQmaGq{gC+usAZ
zvFnK7!)2LOg7GfbJ@dWD$m7uB{@}VVZ)PX!ikbZz82>)IY1_&v{~qK7N>E@N!qVpf
zfp&=bIJ^e>0WU+4(n9r<Gwnk`*R-na#|7NxiTTV@y0U$ijgj6CncZAWDjOfyuhI3_
zxl{3xBkis)KW8y@&|p8S?F49{l*B+<qJ>#VLj3EVxMH)VOR4y$e(`qs^Q$kJ@gYeE
zZNEZM=MxPSJ)}->Jd=D<LqY51z#;8~aXntfcCMo<eFM$jG;ST}FJSlYG(@LyrXd{a
zu;L#;E2^MnYnyY+Ns&lu=3xyCbKmQa+`klIbY%9Tn}Ga$=09mWu1f+ek7{8By1Vpl
z5fc|~+W@1GWuFCD2QXIK7*k?*irA5>Z^mxl2@YQP4O5W#;Qy`aY{#L%JVXu?YPpI{
zEO$X6C%pawiZTN&HA3rRb_IeB5^uW0Uj_u>K62^MD{y1OfQ0zi^lSiFm-p>f{;)^h
z{#rq@4Fzw~sY1(HDn_nl=yW7R&PHN8FSyD0!oa66-c%)caDWT^$V$7KF;gM}#j?Jw
zmmKH@TTaV|?K}Skl*0%-kqwZDz%{UA4HV9yG2=B6ybPBR;)yO_OONDHdTgcwom+T5
zIlwqlmggxME0M>X*K7Km-nkIclI4WFHUl&|Dz{x*<;gc)aSI3@t~k#=L5@n|GYkgn
z0J{m04V_Wu<nkxRJ5_Ww0B~d@&(zk`%|1YC+0Gs)FBl-xkr@@>H?ZI-@7%tZxk#(U
zi$*Pb#rJ(f!fN<?V)7=jdMhquP3p=PD#qDz^=4Xql8&EIZ2rN*&rJ8;_s~&y!b`$x
zNyo6bdWB2<<$<S12NysqR)sw<-f%|+*7Jf8oT1c1Ns3l2Vmfk@vJx}v;bnkUfYqGD
zyaxhwtqB~d;N>mi4b|ztYQ}jYDC^_XUfTb@rE3ug<(e_2o19M+lyX%W7pYLom)>1Z
zagkxt{VXO*ZGjJXv?{PIf60hHmo~hb?W*YwwQGo<*IY0lu8&_c(w;<xR_g4AF?0t<
zV;TCdYK^$v6c5Dy)?%!S>r&<6{MEk@Y%XaiL{n$bZoyuTadkkdNCA*aLS#s*gMl^F
zKB9;z4)FUn8H(m%IEQHD(`a=PZE_!20u}%h0H=3sX0r7y8`kEairczIZftCp-F&~j
zP2^O);+pr<WdMQ1Vn@gxcH|K4^hrmTx`+$k3Hh=DfT@Y|9ID@<a+gr0)Dw=hh<={e
zUi!+$6w&F{)gWqWs7%h&I}sg?Z&<6L9y6oO%;b>@XsGUl!j^smlQchZ*2Yucw`GU3
zOau|~cry$bwo?gnHpG}utTl9zaNU7Cl0&NnU;nx*oi1sNEk0j0#&K_k@%usB*}v6`
z@)_RFAZh;6-8li?@|Io_>1O}#4V6}~xve_G;nvXj+g1gp(<NxmMOh`_ZPCD13L+<)
z{QZ9@dkdhbyS|TG5EMiO1Q(D7=>?WXTImLnSm|0pV1bM7?lO>+lvD&{k?vHwq>=9K
zhWD)3b>Gi>-|sWe%saEgEX?@dzyCew{QC1fAjgmOa*5@(Q9#Uv<CIrd%u1HMM`9vd
z?4$z~U+WsVl>3}Il8=qF-#b{YIAO)BSWgyqK^9~Cr`pSa<$rCXH#eJfpQPCw*}k{E
zqW5<);ws|Cr5~Kh<ro>cZATjuA*8o75CXI`IN3j(%M3e^)I|QmE&J9yMY7jfi{`yI
z7Sy=;n8qbde#o%7b-o?KbKQMr4F<qO{o!p#To1f(8zG-X>(HVjGgmGEt+!`uX0S=&
zjTTLjxxewzp+OL*ywu#!Uf3({EpZhUm(@w5TyTmNf@^GfEd|e6!h40horM&GCE<{n
z+K`)s<xWM6^lj5&V!Y+$NJ2ZN*IIs)n-*gCm%_?^4lW<Bz=H2Z@+6$!p9i@9j|jKP
zU4aO{M(7)SchB){9u*ALCAl)oAt%ZTp&ak$BX*6E$-UnD$Z>|Vo0G5s1G~;pv8X49
z=*-`^%ukWsM}8gpbBwSO8$NyVqxfCZM~ZV*&ZDPFVpr<xSo`E=m=2U}g{MZtRHE!%
zwf*ehDFGxmr0tRw>Gam=hoYnOrQX?Iw2iN`o92)TnHBllzWp%od}dO3A3$Z&_Dorm
zi1MsFLYk8$e+nshFwTm<z;&7{{>;UayyA?}w(1-<VD($0=Dyn10ZWAjBh4$z2~O8W
zJb(BxUYBxblhS50n7=navGXDwn|?vvb^&q~Bljyd+79`-U05YGPFLa)3n4ZqcuVqa
z=b7HUkSX#U5!a&oamtZa;RA(Mf7b;1C_pa*Ur4se-{LwLw0+E&nz%56+d8C11EEdc
zY7)4?nOoylMU)2?+2@-3Z^ig6{L;d5;dIQ8_lj@Wfpk^X1A$iUqfShba7D=YgkGLX
zpg5!z-a^JW)E%-(eAm`<o?O(H`TIF`sCx*-Uvl}G9%^m`E@X*$NU@St1R?gb4Ro}7
z$0K%oiF@$j_EHZ+f}#{tPb?wb+M8-XFJWhtP$JJt2)Tj9l|S7i35i#a=B(b}Hcap(
zsM@MpkKTJd{YsX$FDrDd1DCkGlpRvj)e+b!v4(+{2Hh&AX~t9~<PyxzP)U)qU+@9u
zD5IrRztbS?Hlbhl13g#-RUJ*T;X>d}$Ry7?+v$R1Pv^MBvfr)yIjE?d1(Y4;&qVhd
z-+H0hC+ufGK?frCLLxlmo^*r=OZ3Jzv(p5$JP;kakvyYiIL3pjskwA-ekFQxE-D;$
z+|?09)T={UKy>R+{Icz)`7~v=Qu3R`AyEh^nUxwfoQ7X9z0S3<!nZCGTVVm>^-`?w
zy3e+Fbdq!D9^_>Nj*s%sB=6~wKQut#v{PX?Bt9C&YH|I*h3nw9^yy9g7!GFt9LQJc
zu1==58vHOk2zcX5bNaN3YyH<Os8AkDo6Y$?ZbK1B(QG~Jmfi++R-@reFN3>{_vwhq
zwLz*zZQAb858pF=r#4m>#Z290t|xR@XAP3=_iD5uKR}vni?qu|-$;_~G4qdkAkgL&
zBGsHDgF~Jb<0iZud6+nn>b8_<t{$`KB=;<=K=uiH-N~HjYL+J`2$M{|G1Kz!*mA(w
zyXoN4h9`~HlNq1%*b8=@X~5mq^MEI!PIzi0&vd0N!ABA^Q?ze5!E}(L%j-)o34kNh
zyrTp?%qqSL<oC&%C8@HMXk`-*i;c91@lc(kF<hj*51#90t7y}HQ@cWEqsu*(KCX~<
znOn}5pX*b6S9v48uQfZ(79Q_La!6Q`R{0s~p7qTtmfAsd90!Zj!B3QT7Jq<!L7RU-
zU7*CBUhYS!I*#BEb$$X!5a2tSjsf@_1rn6}d_`o%Er*vdlZ&<C<{vV(g-2*ED7_B)
z8%16SR=%n9ho|q~E9fGU3xU7b5J?la_##5{GebktS!4S=$bPkX<!1L>MQ;4Pud(b{
zj&f^Z1$a4^|LogK*drNqGMRRSR5JZITbYpg)wor?b3f<hr%5YTYUydS)Se~enSIAH
z=$msHQ$Mh-w`z{Py<Hx*2%AsxcI*3RRao4>NFHMTpK`BhsDvg-e3{V(2Y=*a-WP?S
zNL*>5kd=}8bx-Y=XO``~zi6$(p9}cdN~InsERX!a=y$fQTGyLgyt^M#dC4E}7Eh!$
zcCa`DDP~Pl?K2pgep5Wmiiq4KuIeV`3o6eWBpeliVDjzsM|+4amYxu~dZ#||^SX+?
z*6dCotoF!em+4wzxhDF2xat&!=M()6-n){OyU_q;JYbJ0IS#Hw3V#<x(l_1PKk4g@
zoTKBBekrRmqtyk)m!Y?lMj1V@p=BD+j2<!8NRXwkm)02eFo;-?Quw3X^5Z&sZr~V*
z=!BMy5{10I2UvHh%gdk*uFHpeZeXM!??A8DfZ6KP*sCPL)I$Cw<!Tm*PwXqwsI$IG
z>xkig2vR{`-HED;GV*!(TiG820o`AJ!MyN$)ofpxG)S`}SFO47w2uDStJG9~_G%f7
z%7%;k`E3@5Gfusoaewm`eL9=*0)H8LE@H;90ugDHO8@+8u4@>_Ri019jgF8UGi5P@
zzo~hF5`0)I_1297Nv)?RQMR7V`i+lpnYlxFN-C>uzPHZ&(D}|xD&I4uy~p9ju}2;M
ztzyPJGE!LP4A#~~ebyI}X>$@Glf)L@aZXZa8CY$Rm1+@4m!@i4Hii(}X7yxoAk+@e
zqQ4~DWlH*f*+o<vk~PhbErl9dw0tGjab6Lj+8C<23*_35))AG$dm)`_Z71)Ef85?T
zYm;8F>Ttx5q_zoA;Wd%8JUZ*X-C^*kHQN?Wv0C-kFO%?V&Sa~}*;R3USw7>V-HyYX
zGQLqVlMYjUHdK*a?NK^wg%X3KVO(%A(zl(<Nel7c=uHLnLLc0A`h{b<1yC#1*{*#{
zc2f!C!GsYtbsE=Suuw`X+k?MP*9CevNv@flntpf|u5)4P)HBly$)E_vj}sY@bEyx>
zBolLdp&0nZTO`c*1EJR6qqzH_)u+Pf!kW}?vV5MT#aeImie#0#psk-!Q5jSRS(M&V
z{gNBqa~X4_kVcEcQdgLaLAcCt<Pyf?q{ZHfR*0^vmt+qYXZn4TApXt(M^Y~SNAfi*
zgZd+F`S>^4IzChq+x+4etFCF~bLg+hW21NG);KR#A9tgAe{l*csm~Yd`Z2o2pIYS&
zWO<|}C}vEF#*^NL>qC|YM;6>PPUS40jg*6q2U&Vtk9w)Jri92{_%E0euyDFM_k|1d
zlf-M8{toq;X&X2g|LsKo3H20wjZ4fSg?f$Cs!x^Ty0fa^-=xj>{_r#KY;$`s#T6VU
z^w+~jiKiW}wP->8{k8&jtP}HtuGeZTSj-CQ)|sUjJ{u{wD;`V#3dVc!7x#^KpRnZL
zkzmEAE50kndOFAI_Vei*8b!4dtfy}%;vbWRVX`&RB{C?yiNwWzq4k1zO?>+_KO;An
z>3Ax!`ux}aq_8OE=P$p*2M-UQNaI&kzS%1-&s%S6%*38cqH8oPKb^_RN~>_o(;+4x
z-QAheJL4x7b)w$_1KOd)Zg~?Ob0jvE>Tbkb7EaXom2f!|zs#<_XkW&u@?>xSGhbhu
zjnQvTB;8#D@`FKdkZV#MvR5d|Mr!;FmURTC(wQrj$$np1Ud;CQnIh_Oi!zE0n=6Z9
zL%ql5!w+g~c{-o;LtUk)D9VY(^Cu8wbV*LBa#J~79d}w<F)QGlP__<?n0DxM8PxD#
zX)itPLa*L3@g2gdMEs#nGhy>_n>6`LNr|T^dF6JpLhzabrV}1EMUK1tR_TW&3wS@F
z-UDgG?A9NCjVusJOyp9tqw1EoGq%wR>#%PRVC&9r?dhmj+6fu@bCX+ZFJ_arSId`x
zVk>dCkRdNgh^E6W6x|A4rHlj3QaosDDG_IqC^gY?8uyl`V<<?egq_Onn8aC|iV_lI
ziW`g@5qcM6x5=Ue7n@*9do>9!IB78Fu>EBGNK*1hL<TMSVj0e68164ULVPRyrF*;~
zB1JnyhC;sB-@`R;>Ldx}Oek;Wxuys#^RHNlFuX8XXcb)wIQ3=jn0$1OVj5Rr%Sd@*
zq3%g98+~x_4zV-_<T*_HEtQYIYcRPEH(lnnM`1n_Xo<N=wu0x{7GyV2ps0!b^?4=+
z&%kq6np1bJ1m1r%9eLm;q6=f&(e!$?cq{SV*(+}ix>6aKh1QQu_~yZKOxLf|d*N?l
zgtR!-a*fP?-_bjXoVn|lU+EfyCea+<L|_;1?D?_cZz(5_0lh)p)RE&s(l)#oh@ZW!
z9}z$KlWxDSGF6}*?dl0LziXA7@~XrTHTHViaG11h+efODM9=OiFIy5v!J-?9DYL`-
z?0AE!)qPZwQt7u(1)f>DdlVA|$1<pk)%w*{)k@hie8uImy|LGB1S1l~)~^f(6}%mz
zanM5b63bYRPu4%C6cS?)5`dQ<217%P1Qn}uiCu@JOdxXUQ6}<{TY7XhDHYQAsZshf
z0g!8%g6`}WN0r-z-RzhB)Nu2&GgX{G_?P}~r8tA(4~vLZ$xa19QPiuyZIYI_F58Mj
zIJ45_vj}gG_=<|<tnp)ft?)GleDb4lD^3#BL)@#5AGd8=G;(mBtzk=rK>yk;GpqHL
zW}qtEG5S8$XCo#@Z5x$*&RAZizJ0ZX+1?rHILsE~K2lba#=T&p=1Lc5MO+MPnphlq
zkH~o68i72+g$U~qd?~m4>7P-Vxx0rnJ55g(Qjd`r#t()2wm!ndXO~y3_0_q}^`<hP
zMI?K~zuw#8lN;=3M_Q#>SUaVQ-a?}|9`|Oj^BVsq9$rf9!(sbJ*gdV!<!W0u9ogxg
z{eEo-nnf<UouO)XgBv^pT4Cci;}8*T_<Md0zSv~YQGEc@<b$sx#MFq!+J(p22a#t=
zOJt$7`2mNRUR-zqhgd#37-rvYatT5b_Waffk9m+g+K*4zz7+lZO`gHcdGjtA_haPu
zUt0Z>cnEweT`VaHEUEN+n32azJtM7*2lF_eR|m;ZNkIp?MC+bi>_YldGU-j8Vdr!A
z(k(<R;~>%HJS0n=Q11ObQplYM%zE=~1}je>0vB=rcvEaiOe8rGC<y9J2K42rOWcqU
zBgbCCh*cHtg>^jeGSW=*px0VkZm?xFD0?z3A#vk|#0`0=r!OQZo<Gi0w#=>DZ=!MQ
zxZYYae`fG*K0Cs_ULY(a{t0Lf<ny_&Zd;hV)&5G@p>M;<GMU7;%PSx6_l|YnYNy2J
zP}pNw^J|MwzE0{6@e?bth0*HwbsC#74eB^8WOr^0r2eE)TL>ce4(V&pdKLOrXDV~6
z&v1GDRzb0xjCi<RkwLlvo!Y1A-rG`*I(d?#I%c@u=^duYHf6iZEu5f0CMM7BwS7n`
zb1oUF<be>e%#G=KWoJUxJZ`z{c57_=>(Qlol<@_}(i1+L6t`}mPWliqxH6V7M|_Cr
zQ>`AS_vX}3ZI<l4-Rjf}KT;8I-M<fg^j<dg)AIQnN99kwTAcxc9f+~5njCd8N2R~N
z5)aG=7f!uuT<DzyUJbqk66vkYxLES1K^jy?GkYN++cok@_)ymQ;aKX!3yIcDojVuI
z&F{!F30k}>@>d*bX<EltP=lU3OXT<#ZWbMLd=qX~S~dFJ0V-4rMov*7sIibQ7e>L<
zbFx%|Cg#i@A01JyJ>(seAp+&M>1gI?E-z+ff)7lvnKl*EM2gw&m@fnfwM=aC^olr8
z2sv7`+{3e#-%C%t6nXqqYM?7<gV{kKXrjNMFEAvw;zg6@eSD>#)cAb+`0Slv6BZJQ
zXJSG~_qtrGQ-ylT&ua1wIqfL%|N6;;e*+bvGsF{U(-L4uOO4*JSsD-#y*eQN{F!Qt
z=3|zpjw4zXShT*Pxa0WdsiV_gO30{)L|f9WL)wWp3${}Hf`>-5SQy;+IVBk2Un)8!
z^rCfmQxQ~6=Ev`g+%pEf{ZDj8vx>|b>8<!8TGgp=_TEs$4Utb;y!-7bE+H~Cgj<;F
zS#oMl-hZD%Wd9uXu`*5~NoO*vzA2kH!WAxHc;JFo(8NQ6z8+1VAE(#aSJyssk8MNc
z-4fbGQ6mlzH#6~?@No<NI+gPCjHso-)Pf6IthOrZ&8o2qZJ8-VL0c_%9GwM8WcQdv
zBCUapQbZq0ZXr%IjSC8=ekvFY<1lO++dsEVzPbUasD9>~tY1opFLHt~4)#twOL3ys
zZ1cPUF^{zr@M7GIaU+XJ$5D~T_uCDryo>i!=TpOHoQT2FUl)Q3;`W2e0;eY2kT<0D
ziS4v+8me&Oq;&3%0SVgL-p*(nZ=UlJ8^iZ=KfSq^vs4Cpxn&!|qy7Iruh+3i4No@x
z$XgH2<EO1_(nI)=XCpG4g;~N3jno{T5pBMX2A@>ha0?K$f0?|ceEX>WPzZN3oZ+O_
zm)v9t^Ymc9I&fMw(*D#mCClcSezR-L8=UCo5d7mOswj_rU4kpiLN!M*PIWa-SF*3a
z)Z$idTD4yAclBz#RBjIx!f$*Si5@fyF3;+rzq2q|GZV8cuhS<v^b1O}w&0@WOr^V>
zgv9n~>D-na__dLp;cjW<BkbgVCV0B=zS?R>#(vMAxxVo>>kK0aEv428bG5aF=1#lA
z3j>11M8e*eCzRcaTo*F40wndiu1m4kmvOb)C>!r}F4-lxVy@pZ*^uJW^(l;C(J$y}
zOPffu#j*=R9PH|zx#MGzqr0Kgnis>!qY3d|ey=p0wqoaTw+@f~S6{q*;HIR#in(dt
z?k@HzK1jvO<x^Z!b2d&X;>Fytx{^J0N9utXSHfP-(kECyHiN-lckMKjAD>7WDNPng
z;lxC{I(<}rKLw{4qwb)8`SbFUX`q_KCd$Y5XdvFmaDidS7T=C)Z1HHMlEU_&dAKnv
zS$z9Vqj$xxtgA4lRGgy*t^{V*gkO0^bRkAtF9Spy?i}41n4kKS4{X1~c=4sOu7EOS
zso7h^eLT%=iOfI^htV3RY=uyXcUIoiaLyD{=9L*fR}*ar#RDJ#xT&zWNoxT+mThd2
zXrxlP<*_mkwKf(S&3p=b8_n~rHNXguys>~lvyk1Y`S&m$<Idg+v$*akJ|pk`f!Vfk
zgm5Et-rPmA3~c?=OCV(YP$q_I^R%qIAs{TmpIuCl#3PSn1H<qHw1SdvJavK1N-thn
zHV1|`2EH{Q*;MW{!te91{k5^%I6|c*C^^T=CgsiP;rf;Xe`+b<AUT0H!H6!Q4>n2{
zn>&Sds}U0FAEtn5xln=oBp^(Y^|;4&R5HV8OWj}ejq|Ni9->=rzkiBbewR)(GAX^t
z)?aR>O1<S_vA$@7$IwV6cgtgjV#C<0xO_+yzqFer_v1j~KwI}$1(<P2ZaL#q$}Jj7
zL?TN<UrQ%}!NJ3wtRmlb(6~A06&hym4=>+y%$i8dqfa;)ov;|!E{Y!XNzDG<#JyUY
zHj_Hp3$du7dLGQUcIQZ9WJ18n-`yI5)eHo%XCr}B#*vZMfImAV=dre!@K+*6xxFfr
zIAJ#XZ^jW{f2(Sgjj0uV)wn8EUE`=(DaxDQ$Q9Y}OCNa{J<;?uom#Q6#TPWsBXJPl
z_!98ujWA1!K~Bz{QgwNh&aSw%(Fl*$phwz<H)Ue=Refo{=UK1#^@Rkr>{K0ynNkd|
zK5iVBtnoYf?!M|u;4-*gO0wc8dLKD|z!SMUq|x^E55bWkIeMdvT?Sj%^K+@d8T{<f
zTJ%TiwCY6USlP+kCH7&gyA9^9QADdjxpKj}1CcJp!I33ogY}fvN=!{A(Rt$4!0oO2
z>#N_<s}Xtl`9<0FeDTu1<HwpxV@$hxSBOgHnJ@1-HA1!AJzh~Y>nQ2j@vWyd24;yl
zC(vXf3ZK@l<P@=d)@D4xXFU2!Ap{y~{MpmB>Z~OHHjrW%ZMVD!c_7?ZTqNc3GXCs&
zw><n>ea?9d1sV-!jy*m}h%ny5?~+$07+e{uYDf3P8x>M`Av-S;r`sdB5b|x}>jpY@
zzSE<OWjoW4utV@%603h34ci6OUOY(P$+ZJ5bGLLo@e!4<(M~x_7ek+<(dlKEE=jkz
z#*e|%soj|Mr_8cTIK`ZIU?uU)7spAp%xaDb-C}cdWD5P;6G`vaH)dRJ)$5PwIdmIo
z6|8^!MO~eEuI}TOW5%h#t3=#uyumP#v`lkxi~qRbD*62~Wt2Z|tAB09;{KKO6nj<h
zv^nF_;giJLJ`KkXtg#W56eF?Ti>bgLSJidQeqEx8<$JmlPz*@Qz@Zo_eDRWSf5~&I
zoCM)t$&JsLdgM5q95dcI*oY<N%~4;8@)JqW;T?@|Y?KPRg3@}_q}X`Swl8mpRIfeW
zDF~guc`-Z<8X-Stl7{XXMH(w*rkRVpW;g(W!#0bz>Ym8QBUODEt)>x>t*{bMf|}|G
zZ`M$Kaz48%X@iXmUUKeZppFdRmmW+;yfPFy<`j<o5F@i{ViWS}X~Fhr#B!9%SD|4Q
zH#=iER5)ei1%rVh+V7OWQedKKG+dQ?nDNH~`QpS%(y&LPU`Sg<Evj+)aZ|sKx8p>7
zpBDRBuu)e~87Ee1JND!HCmj^y?Be3Do~U=lboku`$6D9^?Lf~l-%vaZJU>@Bmh8+2
z%*@+?BB9<Tej!#OUGo@FB?6q?;$dj==}kwhXNcJz>N@(*tilO+D)l}nS3s)3=8Sh=
z{_qx`>JH6JCP%Ta)YL2n9liowkGs?BO06b!ruTQc^S8Up-)=NEs|Ch^;SdX$i+5uI
zUS(e_HtXCXZCR~7{6jW!&kv{f_lk2ro|DI9g?-(RryrMwf9*HU-NbwX(RhSw>`EHS
zjQJK5AO3?p!b)Ul-jB5D(cTB28y~p`mBm7vf1fz}G2qj7a|dv8*!c`}(?KGzLjAke
zlBWj>uutd=wNBm+40JQi5pMNL^0t`Zypbbc4DKVnpLKiIuC9pdi{6#dJ?Cyrf8v&+
zqhMrre##LM3!TIFg+;lvr&9>O_uN?*1fc_MgnT)U12KJR3zE4`{>}{nW-UWsu}}~K
zAd8#Qq@hupi7!>7K$%{jS;TL}gpt9HxpLu8<j>vWJZ1!;`$84DH<OHQ>Pocsz2BN*
z6M3vb-kp>3&ELJDi4XreKPKuK+$mW|HqB$&>loW1ijp+KSV|bt7`fX~P~~=RQmvjQ
zeXT5tfSMZjO|l08+4M4X{<AGo8>~haC_%*F(#kQx=S03~i3y1x3HQuFl8hA{=CtXL
ze}>cVT=}0h>IDWl?DbGb3?Srx5)N5?Dm8AI4V4aif8}fc@d=$*T4f+UqIIx+#nbTZ
zsPp(p);cOJD;}G$BlfH-^JN_cYWbhQXXY9gMwPVaMU5vD^=NS)(>rEw0N)e~?@Q!|
zniVcuET629$_Mo!U3V&DSUi)GJB>JaZ#MB6q+0Mf(#6dPF!r)^amQV6%^Tp5j<s(S
zw)I_XcV-^Y4D{aeR#g7cP!knTNZ87?EvqDRri*(aw%~S;({`^!UtL1(CyH(f^RVsg
z!pS^T?Bp_X$A3Bx2j}IzTILvMEY}N_?5M@G05xRj*O^HrDMei_#W}wxO!dN#f*P4m
z=*2})b1p-p-%RmP?H94jQ8k{Cqc${{oV`2MWh-Npi!Iw6sm43cI=}$h%lBAI{URFK
zs`CsVG8^^@#_^VS{7rNHNbudL<ivl=OEmo0=a;Hliu%MD&fOU>T_8d~dy3n8o^)xz
z$aTpZ%!O$BA)<YP7kltlWMb}^yBCvy5faw+IvwGB@YmhW$Dh`_ai@;s`GWh0yWiD{
zHy4du?@h+O@qsJ}S6^sri6t=vkZ8Eg#w<&H4R&sy<JNKJDYE1<6jJ^B5a&ThIc>_?
zVE*M#l!Ha@I~ow81%>UuNs{*VUMa!R+rDkfZ?9wDc!ozu3b{+LK@-_jqVxqn7`5wE
zHH8#Ap-n803sPfd6Qr}fFu$3Zkm<RrhW`cE&gd<<IyOGeX+XO~ou+<lk9E=@zolhd
zH23~fJvV!KcRaD{bm8o`s#ICD;}T{K{1*g8G~+a4f<yv`AH1Y9l0DTpjar)t*7!XM
zNlQ+oE+wSie`S(BRz0^M!m)c!p(jg5Aa2R|S#kBHSoKfY@Y@CpLGJQY6`Vd$Gvj;?
z(E~sBLt#S%ODa-)^%b+mNeNB!C5txJ<HI$E#a^ecSUUVfq~#N;X4{HFMu}$~FF4;z
zysOYizcvF@cgL4ZGcxtR6n-3=pIYT<X9*l?>zse8mnD^|3&%Xoxo<96r}{KOE?GW9
zom_d@ac@l3hBy0q;l1kpCDvZ)9cUykp81H)+N}*1`i}6r;Jw)xq49zu`PFO#nwP2&
znh31Y)k*J5GmI!{&NK1~wep|0FD2Ehp}=fACpr(jdd?WBt%Jx)DORRT*B=}i*hVbh
z{aUJihWE_bx`CBo{)v~IbA4LE>4?1H`^uUU`TCzVzGvMhe70My5glRU=v)s5VIH0D
zp*!^As>5Jh!lLmc6~Bu-Ap*a2CdU5WF0Bbei(ZYzvvgZ=prto?hZeFJj>UV*o4cJ7
z%0OE5?X%1yf`_~Q^qEf#HY8_)2@2sf-O1J^@{8uwZEz?tWY-3w&D5Saj~?#vR)wx{
z-Iku5el6h|V05?hT-yKf+%h&zn=Vwl{80yi7fqgVwBL`kN=3E|Z7@@PFV%zAaV9&_
z-LI07iqut}_nHE0{rM1sLDI7eO-@)Y_i^}@Za{ol_)_VZRVTp(SMU5WCuX(Ge(i?W
z{>0`mF7$X8laUG<E)|7yITJ-1M8c(uySWqqc%7BZG+0>_1YWdQtV!>mYjr~(?ui6G
zVfPo^8<MBFZy>ekVz778j}!^TCLG>-#|pW<pV76KZkw9|quQM_V#`QtgJF5OQZ%~H
z_sddm_ICZe2{|U4SScdCJ-!;@!57SMG^8n2S~L0~)hgR`)3a=%Ct?nnyUl5JSMne_
zo)C)M^LG<#2qDQK%r}Wb!B?`1$U?sfrm3|2?5cIesQ%Tb;zsbQ99o?MnoPb-?J<W-
zp8suy#TbnOGXD+&|K5kbID!}svw?rnj0YeI3DLQ`Fn?d%%}mkJm@=IQ$_c#c3-XBs
ze(qRG1dnWH1;(44YBfp?|K8H7IFS@jxRXfPlWx?&q#0j_F_!F4zhXRL6Je?zl#=o;
z4C_EuCnFtvP;(^8fz;jMhNbb*p9{IzH2U>=q*37yl}}zc74(x3?aLAw@d{<{jLyf1
zQz^q90>^dRA~Vrq?!5)1{S$8co!;JU-cDpcmRgVhbQ!1^Da$2@+ep+U+i@pJ=UZS#
zS#Ui5giHN;OY)`C3YjvmwnHRkmv%RwHSah1(s-=bJzU0$yoBn!ow9uGk2Yb7#yy$+
zPokpLRTE4TcJkO$YgzHv?pyNNYxDN2P8B4Yn-$qg_deY)Pqv67`vgyjEDdmP421V%
znb?DhSfPu;ud6{-@u_tto0|6$!n_ztyz71a1$Sh;w;dmhe<ecXk551@pNqXAK^1TN
z91WOU+07n<ln|YO;#|+K0uYu2r5(ItycK!Wq%dob8eXyKYKaPJ52L=d^fYTnP&lmF
zU@x38fkULdz`h-*s860(WD>1xubd-NoQCD<pA&bm;-_9GI&E2`5@gnX)O7sP<aNm%
zoR}M80h#}V`|A@t?5hIX<&BSI)4}-EzxM>6J+6J!Tu~mDk%>EXDUgCYo(b1CbCJ!M
za!#$6ul^I(-tBU)8R^Wd8<4X8?J_u=6r^&NgxohyHJbw|E=`wc*(7D7b<+^|sa0H1
z$F6f{0A95*ysakZ-ARmq5KD|F#Tmz+aCQe5qu?F$Kbfq5Rv6Kn8P$hRv80Sl=o>IA
z;k}X28!Vv>EegKU1gJpb?ssoSTu^d6-w>Lo5B{vb;2W5tsHI1$!keWQaaF9R+={Gq
zay5XG-wsOQ?buKtlP%JNH?nd~NVoe+E9ycI1=AXfkN)McIy>bu9TghW<*v5iEsY65
z`9iwCr*JAHw~CUeE)M$5FavxXov#U@u$O9DHx>iVeSuM<n<Q0wuCqa6g^BAYchV5#
zV3s(|L82jO6pxCC_3`Y&R`oU&7>_pQ<mq5Pe+5a-gl{34XDUFGP-f*&BtoAfN!ndi
ztcy@Ep*6!oD{3v=SCH&2-my~GX&|;q%Z8ZLjj!nO#J=@)K**U`WAHZM>}-a9sk9q_
zeyzNLmmzg0loSlH!fZ`(zT|PDRf0db-d{r43<T+h!fXLAfYlGu#P4vrzUVjlLoXy}
zg)ME8#}I-b>)viZ9ox%e+W0JWv>RMmQyBX4G{sblWIuPvlk<~GA4$TZ7|HeC#t-O^
zbFUfSlL(=SaKR@Gh3`GHtWaqP(a9%C3?j>~3+u`6jC7gk@Ul&a+2%xNNx&1O`zyd8
zrWEQ9HK8X2TS^3HK?`pF@KxGF!GU_(w(Jx-U(alb1I3SsN+?Igj6cT%ueJQ#&uhQH
zgb;c6^NfhZXfVi}<Sm|^U$6)3^zMhI%T$ySxr+v?8-Uvm@*WcPz9x=x0<dst99Z|<
zYqQAi5bsCLa^%k}QDDZ045J$E^z}3S=|S$cdFei}_Sr_(_O|m~#bd;MU*F$)f!ehR
z(JOKM9XfK?<50)(5y$uVDCX70GB9v1)lxYY&$7a_QXa4G3?~^#b{asy^C-5nL@FvY
z;>^3PSjmc`R66O&L&~K8ge79MH1TyuiYj1z{`npJWHiSFW1p_Yw{Jx^P{mfTk?YaT
zG%bME4hIsh_uSNr`d}&V5H#F70&rz*`yk~G`zZfv^+}<zCM7{IJyhdEE>ihiK{`F|
zM&aiY<VD_jnc7383cyvLnz}K^q4W1ArTgF}Xbu-I#_3jhywFpf6N<i~7};jLz(tJx
z)K@@$YlU1T-#yN%C~J)}y#xA=v|IHb?<N*EPQ9O#;U-VGeF<jty0-2x4&x$rnP+aj
zuY6z$^D`D)&3f+-ANul9`1Mep(OiI$3A1V%d=eLICD2c9B!0WRxh0_b%(~%2CS30G
zsVF=)Zo|pip&pEzvL*b=G0ME(OTYC{7grJbP0ICbHTUwxBq6ouPUDsq>Vs9a=xzR;
zmTsfkgQh9yb3VtxyNZ*iR1RHSjV-c`1wgyleCa#mvwbU@#v0^b_ShXCN8<cgmlJOE
zFE_yB<G@geDXFNtqv95ws_5Yh7gcLIzBCPOI%G+I=N@cV5!M(y;zcA-C9JA9lcYAR
zZ$9Te68L({pTs4-!pQu}3{!(svZSpNmZ?7jN#cK2YgP{@!UGw5|MpIMBHN*gkyh+I
zAbF~x2{{~R-JMSw3N*TxKZp51pom3H+u}VsP-%kN-I|o&+rWvRU^F=Neg&eoMk&p$
zo+RgKg_XzDr>@E~i<kI|5$6_qJm|69t+>gSuG7IwVXtAbS04%U21_+u=l4GBwS3Qk
z2kZ;^ep#k-`N_D0x096$J8m=o3|GfvWZoCFpf6?DTNqH;zvK!_7jON(55k_MO<cqf
z`<eHd=_F#mdKii!G~X0C+Jo#2sDYMG{*`)*sI&_2)cv?)8(n-LO9)bVC13v<v$%LJ
zoXveYejr>?k4WhDfPzPEjb&~{xGM7Jxy@br%^~~_hQnErb|a+P3u+C6Q|K1}P5IYu
z$@Nb9F?nv-_KOz+#`fzED=}<5SNbx7Y@n@;uX4RW@)-85Zlo7X@zN&nEgNaQt4&K!
z%Z<{Y9<zn`Gy~S)^1LE{_hFeZ3;x_%tT}u?$=X|pJ;s`19P-ESI|=l~@<}W1i~A>{
z2dBQ$JFE@%QM>(a0gvBq2`$w4kGrFu3-0}(_j-$XhxJJ*!WZYR5GK?&>BBw4Jo&KR
zBda6l3eJ;6hDb-9G&&eFZn3M)up+Caz!yl}i8aTv=eJHMpJue>KZzi)XUv&MwrFlh
z7W?;M?S{gG@6-ea;SluWI8gXk>N!4@>>!Ld$I(l|4kUmwzq;x0SPCC1l-sb~0Ib+|
z1tXI<M?K&8p4q{Ilpyq2dL3Ry5IlMpdt>9tVOjS*OPyCKPHU7$m!bJxd&Ux>7&jes
z$9FCq2^#0)>n|34ad7N?D?9h4j~qSI#R+#3G>hYq?>Rd!q!HhtG!1^@bJQ(TI0apl
z`4K{7=j+&5F+))=Y}Zmkn12=!2D9WOWd3IPaM>%euLNV0=6J*Q0H&$ZWqnX3A{P#S
z{uf3bJpXQ}6uQ<$ollLUOPKx~0qVFKNG*Y8Dvu3y27jez+%51@lAaI@I(F0sN!Nfb
z3{Nj8lL*TUt#mqAGP4xRLic+$H|_V}^X+MX;g~n!S;IEDYcMB8FvJ|I$&{xN`jU|1
z0h50@+7z8LIAl-|FqS_O`Y9`h955ZerQeV!DkE;BrHHEjTxMw)YUBn9KDlL#zlo80
zpxTvBxHN?KF=liSYV9F{$o?fg6zaZGze-P7J#3|WP#@Y)u$#1~LUJc*@d>9#%Lu}0
z;CvLZIBw%c1Cc&>y%iAvC;l2gRdw$1BslS~#i$UO;-NsZX@~0OS2V2EBb=mjR`5HL
zCSOiu0`75ftPJY+imCXV3+jNROW`9P$>6SgnmK^CU8Edb>P>f(#?^}NCLGF>bkOHy
zPg<DP5KBxercFT|2hrw0%I9EkGOZ+oif-I&Ln6T%qT#h+v3&X>cM;>0icZ{<BKgH3
zWw$KB1D1bzEx1==#u@B?s@H!Ocfw=bot>oe(0#uU0w_P9`;WwWnQ64iNFTw{u{OrY
zVeU)oFXHpysatuuN#y5ECqbE8nA?!a1F<AHvz%0U@%N<z0&di|t_%{Ef3u5fB{|!F
z{JZDQ4}z@6Z!s$+N!txk4Wxb+G}PgR?U>J<3j>2hr9{Uh9-;7tA{B9nU*d+c6BO~`
z=@d$bKCROI{xZUew;+<R>uZj#oV9ZrcMK{aTC-B?85>0&(a^mU`diI+#1{vx3piO(
znY`-Lp{NfMb<E2zICU#?Mi$>H(^C-Y?py>^Db}4zz6mn#evG4F5@}6NLxuISwAD;L
znp+}0B1m{#veuX8%Zi&~|Eau`jw7@h|0x?C-`+DSf>1s@^I=?_EO&XzY{O>_14?-b
zs-{e!O|9`b^D6g)cYsjTqJZE#$QeUHU4Prld!Yx+hG*huY@~fM?+z;vyLR{JUJTq&
zJabti5<z)ce{vOK$1jwekrVe#C|4InI$j$@8E2(=>?b}`uN*A1oM-8C-e1&s5i)Rk
z+N*VjEL=6a+xa`<D)>y^{oB>3!Z|A{CAZXZ;5gSxn6i1`aP_?_vsK%(H8#$_22{q7
z42~0EA;i2;l^5P6AQk>0Sw~g&E*{wta57;IKdK4lb=0M5$@sel1Dw|X{`&upZ2$YG
z-&468ohsUfUQP|+G>|DBHSHBjj1RKgqw76%eUUewWdU5x82h8SIQ{>ZI{};p!Z$Ov
zJ0B9CL;orH{JZw^-*4W~`gZD0TGO3P0Nx>wZ0Q&tCTom=+K8xhx)WwaYpVhSq@DQR
zjtY2liyGYeEyF~X*8jfLzsgJh{yQ1wC)pV1kpg;|$Tz+W&Dq+?;L>kLy#Drs{(VtK
zY>=o?zLQ-b68^6U@ju=9f1UOXQc6Y_C%dy?j{Pn$?(<R~I~(8geQ2LyVatB;Ki(Aj
z!v8;>!2jpA(f1F-eE^=|1E`}p{68=JFPH7#cmKb?`CmT2PZG2K`f|H@y{MiFltc<)
zpaJd(#!$tF^@8d%D2o62ruuX-0oBQu$>_%$T2aN&rJJIXMdZMS{qGON7w5niSVU3Z
zLfwBqy1NFL%V|E|f}5p-O+Vh!H-mm(L02Ab42`QUp58z_j*9fZWJkL}!REvEl(o{2
zJ<=bZj?L*LBdLHDJlicWeW1$;JZxj&%&tmvJF5QOT_Z+Q+;oR65MTTPCIXaxlr{FN
zr!P4^It{~V+g6s*U9|uxMq=Zxk0N4R@JxrlbkzI59^0)XaMw4Lb$0*xpGn|hT>;6<
zkp54y+7R(dGl2HP2O{$J31Eolvk4kv(hPt)FTSGzws@fEi~nAkAO6eRizTgrtDS=A
zzSzl(-H^I$kR-ln=>_>+wa5_z4;xp2P<m~W_WE;v6S8_!3sv7m*(-K#EePg@nn2q#
z<tVsVa<c;4*B2e^jAm|k&BPS^`i&p|ff50j31wRpImy*YyLd3LZ(@D91>f2p8Gq2!
z!aaBbdN<&j8)%Q+70ll?JM@Z@#JzmYBA&a3<&~I)Lo-<k=F4lMawC0?tEcawStEnm
z*A&9e)^c$T0~kGHufKe4iq%j_bY4YaVHVXMy{rK1;q%hNF^k+(_t$BLect9@`4tg!
zlT<7G3g^;0$6(zQu}CZxqm`RnK|-KOLsj%)VZ~fH%zCsqXM|7K_*u{Gq-^Q-|Jqsq
zu{Mp;apiUd`T`%;|FfD(C2sV00VdNUo%CxkzD%@~>Vd3ZYq*Jj7p<eDc^BQ-UjZN?
z&ER9H3D#)hbUb%k^<gLZs$QSh^w1A~>{Z(49G{dAnuc{MK$=dfQ7hj+A9uo&cB*s&
zFd}-sXv%e#WVX+9?Qbgq>DJYoWq4>U@aE573SV}3{rj;$eT>8DJniK|V$uV?P-8%r
zJJulx0g#F7d%)wi1Jn+tCbaw1!KLB~`hl7OIBR``?osV}VYN_2HAkWjPG1aKhG`Z}
za`}S=0!!j%`PoST^+h^sgg78v>Faf<OxmjObn|XeQ8u1KQC-iS3hNwYX8^q7kCvA!
z7<<6veAoqRzH6M5i1;ie<bB~vV&Me{VEAZ2`Ka4+ikDK}$MmSM@CJ$Uer$*ijaqJ=
zRd*A5TkinEULfGVa9LNa=FF1RpGfO~P&!F}_~0b)BR;Qkq{@FE?SG8^KY#SYV5DiB
zzHw9Q;9uv!9Sn%PJ&F5(JX0m<9N^F68J;weOm++dnvm8*?1(I!OJvP%o86>+dk{dJ
z^s-yCqq?|DcC@8m8`Ys{N_M;B0Mo2Xh`8UC(h<e?i`+!oGi2v}C#dR6>YO!GUlmW1
zLt28(IicqtSWecZ-R??>-LboO!XX962}dJQVIZD^Ij*RcZ-v0HxTewCUWYdcxeu2W
z4bPq{CxJS?4aqHq9qj<byK%q6V$h>7-fBL5+T&7F2Xh8QSR3P?4DkiU5=Kc_Pl3aq
zq5D=UwhXY8JDUuHo<AYK20KgdM+Elw6zB1u`u#krau`oJn{1&x1Lz|1b}=B)<TRgs
z{DE`Vk^B-(85zDDC)_vXa{+NGoAS;`#A$<+mU1lR|M*Y;{e1nNN=f+FKYgL2@cGYz
zZ)j1SLGxA15$N-gl*>e_`7V8&>(%LUGn)HMmlDjho{nRNL*cRU4UH#r?6lEYY#M-=
zP(r`oGfU73{EJ9YtoTOo4Y1#j9R+^59e)AmQ4U@Zvi4AG+X@XvN?N0BG`#mt-n+e|
zlY_jNKcd1GtX7v<i?@exsKOlU8G6?*B;I@MD|7h^>;q0trpYhdw`i`Uy9#0UL@V+9
zFUw|Z=Q#KlRsOHz^E24U+TD`V|7`E{T&%0pUY}qP*>iM>?03xqHOzdgZ}?UsAKU3J
zjYBB4-XoyLGr{{H^Xrt+13*2pE)UT_rv9(&UhBHkchb%$K~|=7yP;$R7yt@_yGBp2
z1*|JBq)r0VZmuhX1J^~{Z}laZ>Ko<?0vR!;0zidbwOZT<YKV8xM{UE}-C4lmDy??z
z43Dn*c^lF`BPBr*c0LYhPi=!o00@Jn0|gj&<4*Q+Yu1F8DT)!ZW-tS>ZT@CTfVAuy
z?DI-K_DLC2hCP6Tvl~fXAl%hL^~r<j9z+n+PzErWVn$x!RnoOhXG7=?2c*~_pLCje
ztgQILCuUoszotPPKJQ-#Bsc^L(3G9>IyC$bn7$Q=+-m$DE!)ZYBBu{o@t)MV><r(u
z=e}EVP72?hsZxizRuc*@d6YzQZvAJH6(3u$dw0N>Ux5=>=Kn!xK$Y76*DpjyvY=7O
z#5(cH{OT(l98+dIJ(l^nLo{>E68UZLvg1rjNovfr67S1*oecA?pCJf;@wQtDlb|mO
z8Ft_P(Fhc6QGrl7zC+Iw(8vFIy8_}3EQkJ&Bk&MHTz<PTDS><mGV$SY2I&6ytDE#3
zoMY^I_N@=aIJyC!&~JAgO<MkD28`@IUEA94)s={kpQtlMRyt*(DaLq3HN%p94NESO
zqU%<X@rKuDBcj_C$K4^w35#5(>W#e>O|d-@kIULfU`|Z#syYbnN!0_JJ#qeu%?<kC
zVwbKHq%rlKPQ%#>SM~!wr%?N_n2M<p7ajdmI{RP!4~cq^W-Xdllv^eGJAPTfp#V_F
z*^~JgmkdeKj4nc1irHlm&_IReJmwiY0;?VZ%sB7G<+hIt?Zao3E$*CkCz*KHaXKbp
zJE*kFf(!}2AZMr#z|BZv5aKo6mhun1ZO;-pikUIY5U)4a6c!r_8bI6hvFLgvXab5i
z0Qaa8?`+_SC*8p~AlK;y?p~d5drX4a`<EYf@{&Cpa0m%bPXNb=1X;NVDK{$xNyZ9>
zY7kj-Vi4lc@dT=(aeb(tU&h;h(Prim|Mot&Q9VbS+fu7OPm@N=D8Atx?lX|epd>Ir
zk2gAZzC&mEnms`k?Rl2Gt3S44i{g-Mpf@_#Bq1W<Wn!GRmw>EV9Wduhn~IaIU--mm
zO61zU(*H@nAut^`!(6?Aow><4@LA|z`xuSXb3da5eveLF)8%C<5&HGD5$;eq_B}NQ
z?#)lbm#4p#g_+*9Q~ONn6RC2w5^K)QJKhX5s9?G3(5{<1<k;->eok^cVJl9;Hqc9Z
z|0BR}+`Po{qSERG+&cnW;j#6SiK#X}U7N{Dq7_r;-)^s5zJ*WNob7`Hw)b3L3Nfg}
z6Q#Q<6*_8*!9}>0#xer@x%PET|B@P`lD9MXE`uRd?iL^mDU4}~JAcm$voMDRzmC3S
z!f{a9O3Gu$C^ECBzh2#zEW{}IR|hx;*Z~I*im{#ptcS-#v4RJ4?-D$Wc8Tp7M(jij
z`^erCw$PeMiOA?@On!@%zy2wD#-8O}{+7+Dpr+{nP3pE6Py1h%Y`CBsdU&l=6LT}R
zauOWABPHKw=o+aZ=85DFT-5_kF)neO@|YckshPty{wX^BhvV~am)!_P0GxVQ?7IQx
zKh8Exp0I&+3au!h)6#PITC6xde8mWWLh;e@RtCevmWQV8(N0|CO%J66aSR9)X`&lh
zX-q<hR0L_My%x_^0qqkM32#@>;P~Bax{XB;s#2WC_j-uATqfvEbn7TFfwUca69%-1
zX0PG-!?{Fj3AsSLr!9sT#x_Xc-fL#QLGEG576{t_uLx-q?nrZ?3(5k`z(%an$zb`R
zn?jd~NXieJKv(R1Q;Q?T`g3qfypP}*tZUp0lMD7l^+oGo8$&Qjpvme~I=o*<guW87
z)IAo-xJ!^h1xD?V->g8I{Qi7<jEmySKXmjBv}Io{N$N#*elpxsfBy})Ccrt;C`L%a
zFd(RiPMx})3DiQHhU`T%jn*F6Y?4o(1mp2m+<MA>NqTt;qPXv0L~Jau`BCtBZ4)09
z^S8CcWEeAcKlQVLuVv$`9;4-i!p0xM`GHI25%`Bc$MLE6*+%5dE?R%#*!a~)dzlRL
zTJE{@-b^JNa_eHh2xep1=?V~AxTf)dCafx0+=0;|C#TMSgvSMESXgmvh^A)3;4u}z
zNNKWsVLJJ=MKc<F2%@J|bIGZv7Xf)Vct$u|q!BCi#vkx;Ci4eFM^|hK?)7gE_R`Er
zodhv-aGAEf2wP&Y4Lk@H{x%oF_q2`sOu9$9MxiQ<1mgi+3mTb-eV|iA3EGauSVXT=
z$7SgosZC)vH`r1RZl3npkb{#sUZ1MALpJqQQN7EAjJ!B!EbNsLQr^hmDYa7KcUDjK
z2N(LFi`W<h2KL`hV!RLNZR=~^G)CY`XH1+*sDB)1`oF^P0DoWhJm!B#cn5?Ss7J&j
zFRi{>Kn^o;AH9B#;26rRRE}WG`SK*rIe{6LWd6y5gJrO=BNx{w6IVJDUiD>zZ5+04
z>rG9xBdy!_RVauzX%v>^OP7>axJGKIMCzCJ+aj~MgNvv{3Z+NdedoD3H96z0uq7d`
zqz?(7FFjZJsmlm9R$*2esSqN+7gNmz9j$18TN(aB-^@_S^MP{1yXmRpX%-KnLMbBW
z7uml8*?WE#h-q#1`mE-t5X4tH9h@f^{aU|w$V}e#PD{^yHM>@+<UHA?T$k!HDF$&$
z5jxNKJbdXhZ&CtXl3mC8AxH?W+l&UPE8EH&O8Ynh?6XJO`&wUq8@g%>lST_992N4p
ztlDOo@kRvW>*&&B>-5^>4>f|92ZJ8drgZZgvmIrX&?9~u%<mR)hpgv=w?-1xn3Fz~
zqq!j-^|8^}z1(wh_Pdt;I<{YiY>3jNQAwJsy_N!urC~+&SMEpe>c;r{&3WHe<c4rX
z+7PkI5UFTo_d}ny0pRGD*VKkek4d{R4SCrN`wNy^LEu}wMzIzVrc^u4|0SxwU{{k;
z)sWO7yiIM6e1DfzI*|n1;VFgkV^U>nG7WL_O!XqB{a$e&&y(2@w)n(^?(^@UR8?J4
zl*a$O-pao|$*f;S+32D2C@G6TTWBU&WZi9J@>3jNMBGzipdkiGu*@O`09`fLLm1X*
z*gxCkWf>i}gknsMLih=d<FWrf3Lm#MmP$<7?77Vy|61e!au;ts#DKd0zt$HShQoKV
z?y48O5iEQxk_E3I5hkL$3RjpdcW@?7Our=;gD|<mj@AcxR1Md%&*%Y&845$NydU*b
z<9%EDTn?iQz177Hz$%I#m_a$DDYW5chOsjtNk*d#xy_3wMZaa3PURwgBD(Wsc;9UW
z*!4oYvWwWg?hj*|2U3O>b&bE&6oaR~@|&7`J{;?C)Ph{Y+I>7527muqmkOt}i`D$7
zOu2l5J;u9Rb};gO11^HAyEPuM$&(Z$Mp6ox$R-?BFh!UC$<J*vP2yNSp}gxU0i<}l
zwme*x9WR+SKZ?)<s$7!W)2?N1KPxqld;F==Rk=LxENd(IFwX)aY)YSTXZ2Ze<d3m}
z^bb_Bnt?~y>ap;c2!wvm`PRj*_f_gcqPIQZ^lfS<$I7--%1E8Tf4cQ&EX>JX==%~8
zp-`qnOAg?gs(0^R+;3CFYDFv#uLr}K_T9te@|kI1l1g$@9k&L|w~D@;aqdY`3dA3b
zX`Ca3533`cmZ-+ByWYK~;CuaI62x+J{;pW{IGFo$SD8wV12U*x$LHm@Gez9r{?97-
z9W@hE$&qQ_Qi?DhFzkbA?@E_jRuxU10<^l*e>PV9`waQ}%>fHW<pCwJGwBDpztcGK
zQ8%`jx;w(En`(i7%sSToVHWt6)JH%b$)!{G$(spSV@kCjh#Q3ia4#qC(dEuO=r35U
zWHTTV$3h`KRPVhns>o}hTn6~)We$7$d<5R1_a!DClt76B(sjo;!6ckTta4?ZJ^9#m
z9Hc1>TvmT<0ZgX=T1ILMvJb*o2Z6|}C_%f!(DND36@KcU`V?EQ?S0_Hc4RJKtK8PT
z;1trUz`tM7$cLdeyX!rLd$VJIK#}vSeWk2{UG?j^H%8Cmo1l(B$h`RB$h(SDK0Jqe
zU|`996Zl$7^l3%7rzv6~;*-2df~*OntO=T8MMUdg%TudY+ZbMp%?!ks1$^ns^u^Pf
z6rS^kxKx7k*;(admfsaSww_{=@8cP7@{uoxdYWiRD>~V;dlg#uT^&FMuV#>lF#Px;
zmRLgBa5k@`AYqrU;kvq$`S>|+oJbU^ls#9soLKNt{@V6mKqYRZVBjpF7(8R4b(vFs
zzntpH*6bqOvDYeI3vXp7yYq1CYgII4?ch=5zd;g;GeIIBS=$mNdWm&u&YwU!tS^-8
z`@JN2_x^D&<H@=rNoW4F`6S~u%M8aO?RLrFySSzHM1Bve4wf#s$1SRCnVhN%_Ewz4
z9t0TpI1kFKXI}zbZ2pMPxOp{)mpvQ_QM|c%4T2#Y1or5}i&_u9@#!tkbag!;#{bJK
zT`Ga_*_(qhTlvr7V)Tx&cmi2n30vQ7r^ZJdw!mqWcog+!^K(-_Z*)pNi}f!?pv#+j
z@G93%Xet)uxIKn?JCjQ4E{?||+^f(Ly@$6@J;ZVW@U=aw%C<@DkY2AX2aZh16BDiZ
zy`MWlhN!B=hzgwZu9CDgTdhEU_<g*kae+xd{pR46uxP8ZW~Qf>-+8q)cOf=*GRRLy
zjB@1d$GhrM_{5o+{OhBU*5~I(Y-zD?;ZGfA7wy|Awa3z?S8`My&*C~6{nFI#mff62
zCu(jx+?*&yA~?6q{OYOq-*#&hl@r?Z-RkDf#rs6SlZf4rWX4Byk*GUm1mRu3^Q8k6
zaisdoX(;$Q0~(<Vl@)A0DrZP7&H%VI-qNrj&1)oa*BKn97~H65;&HJ$i8eafs#59`
zHsvYD$ClGZr(F?Ya^;mHhhn4{D<I6G>}s`TesG~q-R1O`L~$j^uL*RQRGPVD<W@2i
z2_1vY=g>&A_<iA>qWi~4c<2@BIcYON>IhM090Tl&!i$)?EbsW5COT}a8XQzt=Iivj
zUwG!_{VHqDWQ+ayr+MwJ&$hR7qJ|Z%1pLL<X#LmGpz4I=Z9A~=AazR5w||dargYsd
z8TqG6_216%f4;eO&rj^TNS@ihdM6*NHbBItF4ur(w+^)vu!OkJ?aVfRf<Mjgd<E1e
zLwUMjlDuEs`XOiYAjBi?e*U&P$a6kQAQO%oiAdo{I>|s%jNPozY-VTLP3Y^JZ~4H9
z98oGIR3?ZOtu3+9T9}-gL#sYmKqe*^X+bh1B@9nl&zj14)Obr}q+shsp(45Y0)6EH
zU!!4o*%HxlA>uF};q!?yr7E&nKNDAl)#q!4s&#C%bMz(1EIGSm#f|`YwZTdG`0Fpy
zsE*OT?<~g~r7DlXp%Gm`Vpl)@dP>RO@g>#en;3&(+)3tvBHlZU@4^@<qR}j&S<8w8
zeIUtkC*~VbeBMzrbJA$u&PVQXH+~b;{fG8M$Y;fJ=eCd)dik`l{94<24^%yff4;Jg
zbI;|4_thm>FeEXITx`YFAD0*`P6g#iC4YbEau{s|fA&_<kSN<j{17B2Y4nG8h+Il5
zZuA6$T=M8S7wQq@|6%W~qpI54w_!mV!A*lGEg_ARfOIG!f|Amm(%lWxg0KMv+<*!q
z-6@-tZlt8UOB&y~&N<IH&o3UocYJ?+e|%#&9H7A7tTor1_kG=007S{G5Hm3)B9;DV
zyrpjsKsL6ToDQ@KcRDe3X69W*C=HbESS=1#CNB-hH>UL{`_^vqPk_oHOXQ4}TC1Wy
zpJ=*lz=jHRepQF9mqG=Na#7@S2U4Xcx0Tq^qZ^jCo~&H2XNyg`|B>i-U@?spjY&oM
zpR-CZndp~8gLT4N6D&B;7Qljn5v+KxKg)1Lk*YU=;JQ7~yxTyVe3lk}o}0D_cX|A$
zPBt1~B)Deu4)F*kPkVLWt<C{BU9t+@r8|^DGaRj$FEVVt&>1vOg6nh$dLFKZYj&{p
zNRw)pzR&kQOq=avn{5HeN9n#B06<Ji9~w+zNrNDyNL~W>;yArSHvlfW->KbeqqH7W
zm3?O+<Id9$cB?MA!u4RS^(Y85)@^#3cWpUl1ib~KcTT7sjYrQG=DVg7nsDf3RR<o5
z9smk+-kQeu<eSS4f~hVnJNFMRpI%LQ&UA&-Q@b%CKH1!^&(i=l8}RUku;Pfi0WxCj
zE9UE^l4M}|-TEOg18BkrTLTvPD9LNP>zyEgU)SLRihGbi_XF2h{cM`8anp6&HnB=L
zcXAr%>g?`+y+c@6-4h?=j=&KtSDLwwfDoBNOX|3KzlrVVyJKDZk1WSN{)N^5x0C(P
zPs^mh4()2Ort=pF3jd1M{pCF_**4fjs(@lg)+v&bAbA*fnSY?wdHHqu4;GF<nsnJf
zF}5PZ`q=YP-0+TWz*ux^4*5u<%9|YR(2OVGjBB`z_}<qXNTi<9O=4kU$iLMmtEw(~
zrnDQ@$}AmsVYt*k)oJ#}4H5kSPJDM@L(N-<Nmn#(mc|Z{j}Lq+Z&aAH3DR$r0d#18
z;ImM`i1Iq^F=t6p%Drpq<FH$48yjVx*Gnmz^tHC$qSc?wN2ZfWcR&_=(x2dXSaAgP
z%agZJO97o==v*rQH9yYVNWR88wHBWFj+sg#mFN61dLyiZPn8v0xX(=5NYDJv#GPio
zZRgnofT?y#Bmea%E>;z=x~~@m>8WjymCSlsJiT1aBNFR8nSlNKV{=4*TwaTu8TR>>
z2A)Za8aZ%;zG%{KT%_loElqYoO!BAp$W3g^?k1NbFJ5WQw%nq38V0!@AGaf!)*4Rx
zp8Roxy{QfXGwZ)@od5o*HZ2gKuqOT`SqH<q@ULKyu!am#)6gA}agw9*FlkOS@UZKB
zu;RS&C|2yMsy1c`wS`mj$paAou%OYfPeMjugbf=|j`xS*+Jy&%-k>0Kh<{u+z6-Ju
z+Jo?gI@pcp5S4o3n_*st^EKc}Wc@go=Ll{>WP-A&M@ycLXkz?L9SmcoIIj$VJ????
zr)Q>Z8i)Quyg&G}T(Iz|KNkLzaWPVaR$TA4acw-V(9~lPw^)`=HQ3@qS#8$)F)Lwb
z5!0<3%fSAH$jd3gUwXu2!aH>frxKe206&t+H#ZGYng-0}CqDozl<t?Ww#@z%8OS4k
z?NuwEJKg%Xus@&%i3U#<CPL*i()Brj^xUZg0|I!ZZx}jnzP*uQqUhZnDl**cM1q%@
z1oFfqe!;<q!c&-JUls)>+rR-FK_6(begYAq9A@M_OZ>3pAGKc|&t$-!i$0)M$f;6n
zZxxu&w8_+^_iZB)LNq~eF~B0a!h_B~@Y07)0%-8w%#iO{x!HS-qgM{0{B>X|Rh1db
z+Xzhk)xIs=ZlhfXqRv-ne)eBYMUG+|LPgoo*KeXi^uF#^UkEvQ@l~n)v_+N71%<43
z+!tdbJzN5$PB>KxD)<Njnz_}P@5z_Ot1~^Ok})Cy4j+OH({`dkX`QN1oa?2YYE-$u
zlrOizT~|}fuvqAk)Uqv0tI?J9IFu4C$@fxY>a@xVx3t{I)t^sjNO1K(5vpzc)F<z9
zzh|H%P)xAi2+;_#NjlioLE<+!Q*l=Erh73}Qx5Y)1}~5P-lPAwIrcjkJSuXtDgW;_
zL7bYfNM@i2IH;hNmXIjkCQu9}=|<=^C#C}wx(UstsfD#khcUEzIh7SoeixqE^-Hmz
zL=SZH_F#~K>6Av^Evd?7sLU0waaZp}Fdi`C52Y4Bk%EJBD^RN!=a~|pT2bqy0d7qb
zo7w3eWbF@O-*2o-DH%6`oRA$Z(`aB+Y9PrWb^+#t7)~19k(g8(14q+~&C-lb(s!C#
zqUNB1+;R;s+?@f6M>GV+zel|A6M7pk@9IUOMKwt)7`B8HG);X|D=|QP;PO7%HBiT9
zJ^#+-@SKa4htJe-{-eP*!{=Tl{aIT&lIw-<RsZD$@C!$QNZfNgd*S+?B@%A1oz@JR
z^qr-OL3vhmA}x#;;)U5E=c(k#18mOqR}oEt*qL=xJ$IeB_7p)O30vWxv>`Hha))0c
zZcLyPUL%nilL01^b6{AE$H;({J;Ej(3gG*}*n`5T{naq<>>EdjC)r*@F>BO{Rksw;
zR*nK)J4A`9JRW97r@`*Sa|R3og?54lj~GSq88Pq~(GFefz-AcsUMdWO$mfKm%yuFf
z+XwY1&Q*?5WHC-EP|HnWAo%5k$E!s!Y`no3bKq&ifvnGvZy#j3BzJ+b8xAVce|}V+
zD8p_+!nuv>bcgIx@79Rx8yUg+FuNa}14NzxOnJAREtg8u1XrmolMheP=2K0RwC-K1
z)kH)j)Ke>LF#=rqXPzE3!qdB2Y-M{DDnq6_yg^c@uJ3asPpBkK#{b<Q$Y7K(2>Yte
zR}J5{4g`cNgM7{vm#8}Aq;#Af=UQQ|2AHDBadDT%8jf!$(U4b6IV$W-=DL|h@WWfj
z4G^?Dm@H!;oN1t4z3th$_xgD*J);o58DIZK0`_8%R^^S)`6aJrkZ46zVbyo=85xfE
z-xmxOI&wO(+*!Rpq}eV52=qh(p-`ehyTrcCMCpPEVAGqB>dv`V{_${`c^vXkdlc2@
z5t)_`Kp5YK-%l07tzNVrqOL<!s+oO(KCbD>%Z!XdM-Ni5uL89;6_Zay{P%(@p1k9G
z|HaR2CC{`Qa}du<jcV=c>@Sy*e;?WY{c*q(tVg?J*nGP5e|1wtao2FO*alN6>x#g9
z8iQ1s5?xv?T)i9T4QCP^jf7y}wSFiUlkOc@-R)TtyOmIsu{eFJbrLWiyY4uU45BfM
zsRAXW@0%5K!d<==0d_!*#ZB&MlLVzCjxT46qAkW8!OdVqwAIOtX0}|x4Sz#UhDL`E
z_S*d>$z9Nqt)mtba~_L7y$|5Ba}G2bXyR!QFN7|w6i)Mu%UTfN{gJ(2&`t_K&{N1q
zIxY$ON8$|C;QSa{0y&wO!UUuC%4j;{Mc)|H(9l>}p2`7*E1kZH^7#qxaYGZEU=2Jc
zO(-C)f9kT@pRqO8|IL1;C;<$ej!xVl303l0upmpK3RGGrX6Ow^INME7nTpzc=PI13
zobj#oqpOf^k*KC|MZ6=6SY1t6O)CkgkfNl6Vb0%ZX=3SDewvPzrMmjz-A6%K%rqz}
z*Q}eB(duQ49!~gcy>XP71rbE^2#s|dBccfu%TUbu`xTLAbuZ_|zlQopIa(512w+XG
zR&IoVA|V)gm^?1>^&8ux(eNzre7vL%g8ei@&ty>^rf!jn5(}q}Nb?Q$rXQ5LZ8A!#
z>;A6bo3ALafN9G2@Vv7mR9di(YAqceK2Cr__vm9cR_=~KmyaI4+4S%+y$+(6ZjcdT
zz;HZ*s5AAycofPTsNV~m%3}2jMn4h?(R+hwOlf@E7%1|x)1+5r+J_pzJWjx+wtIio
zRFtsmaQw57iaO=e-9V74i5T=%ZFnWRRto$1p_*)EafPuC7Ds}%{~G<0*2f;_wAzZ|
zAb{Adu#<jTKjUwL)&KHAKg28(djG-tuR_ECl~DY4yCh#=dA>`!>|=3ve?EFrrE%Fr
ze}7Ee$1Oo{TI^deMtI~z>S|TehNeAVrKPW^gW*!&6n70)XrC;~dz3LJJ=fx7>u{oC
zbjdbPAYOxs;UZaRro^M-@vtOHj08y&2^>wX-!76IlBbE40Wp-3<Ate5LizVny!uni
z$_V|KRu&(?FawoH=Str*XPgR+S-ummOXw7yzOPU9HVB<h^ci!WfXtfW%8J;=8)&AR
z>_X=tf7>DqiwkzRFX#X865aO0qbb?6*)yAtgWjL_*F@&KwEc|yz$jva;N*kzLA?rd
z-ud{!H!_zuC*frHhd~zuzkpFjJ-@9RA*QMSCn4;9Rh?7iZRKj>H`psbOY}&}od#P@
zUXy$+=8VSEL<ZT`i{AnB)Xx;SEH~SD^7r{zanFd>`Ib5^t7CW+7nnyWpLP7aOq5c0
zzp=wmCg(5fv-D$pFW5`P_3-m>F|nr>kox7Y<}jT{V7+csmlUuqo2=3OV-t#S;@ptd
zvQ6a4d09@lZ+lla;ayrkV~U;0b*J`x&2ozZ?Kbbs<fZtY!oLIp|6W4=^KqswV9Av$
zD6ag&1`|(1vsVaz!%LdPS%0YNCNOSBEaXE+pDDy0Phyc_B6pOhj5EJr@=4_EGw2Ta
z@`x^<;$&n?EbU?A+Wuy;dxHbK5?;Z+s@2~eA_i>$G%1|}$x}(SasVN>o*o>g=?;CZ
zLx(aFL1=g{4+mxBGtR4q?@v7rWw;O3=$<p3<p~CMW;J$7<w7E_|M17+9bkfeDIHAt
zXDRm+1dQqXJ=6hYBg5}ShK3;|>Z{E7y1WhqPv60-I!pB(vph#Q@9zCt1Fxk4YA7r4
zTJMQP9|v_V5Rpzmik|=Qhg=7kMp5c4<qZGBmXk!K_BmAfgH)w>zlT`#a!vUr@;37~
zQ=+HunA1MZ2l=n7`HftEBj*PeX11&gf?9qqf%lWhf93A}_g4&Pk^!w*lbW9Vm)|vT
z|7%wV{o_{#dGP&@%Vyu${>R_uzrW_c|I+{e>;Hf5`hS&{1Tj@qyKjL>mI>Xr_!3&E
zT(~YL?z8;0Sgo6G7sURF)&Z)cVL|mW4rtas`v1q}@dqE{@B!HA&+7aWMgCeM<oRg5
zfc~=!@N$oREfK(?s>ife$6%Od%`cKX0p79AS+Sx*G<&RrMyMbUat8_Y_#{SAzQ~`G
zcGRT0&I9uam<Bt5LrGi!!m;JMF>l+8!=c)6rkoa)=z4&+u!4sWV1G6fKU0ivs5r_M
zEoIjc(n6b{vj-%qu%Rcy?2tNh6EFsBbzeB|g9e@m0t7nWgECD-QDEBn&CXu}-v6|A
zS7X3-FugZ4^4E6o^j-i+>w9k%kl67t)he$D^lM?D4;<Km$Ty5r0Hs$ueg{t9W;c+=
z$OR~`>RUTY$w1R78bxuR5vryQx2sq1fm59uXTp%%@rQ00#6J0D{lJ#9@xsv^B>5ph
zD?PK%cj-M35?nIu!Iy{nfn-d)glf<ggaF!G_1uFmRMNpUzpl>zpz4T#0$~m?Ww4q<
z@vM*b3;LvX-2k{63h^2PeqMpFvrs4zPPcuMlfWPZ@<+r=2wF7;HrcOFvS@RD+^z%L
zXESINRNW6OKcuvE*@jY>F^i;6r%Y&BO+pExNbACS(jq#3@IvOwwjQ+!qNPh6Oz}j(
z)n(is0L$oYg!$DfNk9iBr)g8%iUj<s+Ynooc?H}pGeTM(!BVZlcRc!2Kn5h#Is88;
zvRQ&9WZb`m=Yoll03}&E+5#D8HzDh-pTxHRoGyPXvj4UOoh87w#4)?3`qz?UPP;~`
z0;MeOg7**{D|*&3od*JtGH^ijVILC8pRa7j>4L~vlW)V^jexVT!)HT?Zr1rkwS6z1
zUs?&+3L3Z{PrE@JyeS7AqJ<DBbdFri*ZTw{uTDAsgyJ?KMpiJuWpkFS&rVzXK)5Q@
z0nS0eq<2$>&p{_X)M}6`S_rDyfVyM?7!7TiR5)09X`%WT^JIoV(#T4`o-_|JJasO1
zK?q~#7d=Rzgcr$H@<?F}KQw`(@nU~BA>$(ZUM|R=n?W^=-c9xosRVRsnr^_AR$Ez9
z*umzyVsQ951F{{z>ffvE^IpC$`jZXJKb445CnykRsOs!MOZj~lP-%zIO_k`G1>b`k
zN0a*45sDAx<g5kk=GyNRm0zHfv>DX$%f>KCMQOiq9)T(QX8RTbImOp+F%&+OZYA>4
z|H$^?@%`a_&|?^JH_1Yn6*&61Ld<2ILr_AWm;3l!p`PKlG;5w3U&tN`1LnlC@{rqa
zkBR_%z#1H*71aIL(C50uwa><y5~rh;4J1iAj*rQHIZ(Iwe#;}26uQEKIAoVoK*Kb`
zzF#yuX%|pew949~@AAzcmte#9b3u*OQ2@qF#kA{-%CCpO9DrZYdiOTnuf@AsDD+D}
zzbB#Tb2QQnyq=a^(tNO<X%kR5y!?!!pn;{sST@U<bCaM2O5kPW2NbBDx)8xBuNfde
zc&Qb2LC=Hm+hB^!&edT&Qf?{2D#s4*0@cmnt7~zNQjdlsUi4;M9ey+fD`rQ2ckFS-
zgWP2E&R<~ML-s`M(%J8U0eDETff_8MQcNd}Ub4-RP2FQH7I;@Di>Ilr>Z6XkzB=Co
z@j|`q=W_+=$!t)9SjD>&F~FG(s9FYUidP+#TaR!!s=(gy(1-AX<m&*(@2=}=AlR{V
zzYgX6wPSHTMYG?FSjCQHjYb|0DN}3KwaZvEVgR>D))e3=&L{$!WscXZoTFE-u=l9d
zS&IoP0a@mT<~*oOxXSnlpuNX!!wA^rxb|>wfj6w_A%zyI*6>n_BCzr05Ld2V?4<qF
z5#4^)?IPw^UW+#FAVaYkgv(JjfH1z$JFed)0vp-hiq5n-GCGMbV<g@gy_Ero1pZz#
zzF348M^Y@zc>mN|{{u53q6dcxW1Fg4`k$32#>0RTG(Qk~={HqBdHtpctmsD|+zRls
zZ{o6-J^c<Q*i-YPU=qYSdcAbko)5CCH*+I->9-fDf1^c(a}nK7lNh<hwEaW%>mb>h
z#Ht!`6~@LBs12+wXnr_%+o_v?RL8!l2%LMh8w-ey@9i!NZPA}^>jA<Z5&+{nGXy?{
z9-9!v0+`zgk$2X$rlbx7)%bPROEP0zY9P=VVwP^B?g56h<Pwz(rDkO{_-WTY9;3MR
zK=$UdrFxlDfGjQBxDBPCyH!H0wz)O1$5uoN;oQW3{q76a)fG5>YpUiThHNBU^G#O4
zGqU2_a_<zj{WnZ&AMJ*@^ep!B>Q(}6WX)0c;lp1MYP}G*6`d52yptl?J|*j)c4{EO
zfO}(fbAN32q4iwsL3<R%g5(ov$|{MRn>6_AZdqq4#BPmS)@r+BEM0&>+cnLbU^f+S
zw3f)zq)jpUG?%Pw2)O3+(7b+0Tw}Y^68(nl><QbQ!C0erE&w9F0cG`FTvZ9^fymff
zce6NSj<k2jEHhHV#va!r@s2>W-`pjKG2_uIPgf{c69C`6$Hv&US<(DBPz7+6<a6|{
zYzCZ#bARy3&nL##Kj3ZT&7A)@M)tp#(xuUKHmWstNTPokypm#0Mg6PzX(bRcgaVfD
z#FxtX`P<f?qguO_0pnE5O=qVf+A+iAZZ&hPrG9C(j8I{6Mz>TF$YiYSXcs}zSkts%
zi0#Hq4WT3Lj)gOo--36kRKh8_Nf#=TuaexsF_qG<xGN_tm-y<@9vIZQCJ%LEJKSyf
z;koh0nY%5hZ8kGw<?E9EyZZSM7b<~r7!vXSdPF;5%u>7UlIH;9qJZ*-EBjj{h3S|{
zO&rF<uc32~Lo8C<yp+lFWl$q*8~q`Xd<qK>p9htZqY5FS+LM;^L@NIXMDEtk&%3oz
zm^+ijv*FNtcVbFoVNGp=(#F3()_Fm7A0X*Rb-Q;QzJ5r1sa?e!Wi3pag{?_G2}A<T
z0EnZs_X(?&pQ0edlnjCndG;lmxQ<XrAFv4)KL<@9Eh^$0Q-%m&qWbOe(p(4vd=Gic
zIzPTX-->d8wjlZ2&44-sHSE1d^a>&nU$q&qa5nP^VA0fKKc=nJWv=#;0Yb;_4HPEq
zLcA6~%jr&p0oAsC6LyQHEd`LYb-U#s&|WletwJf%HGu>MV0Z;F)yWj>Zx4{WI00ve
z+zl!M?lnLVU#3H!!cV}TK(mc<Y+L#XWtzidz=kC3n!mCHPOtLFCQ)#ptIyML4zqM-
z2nlIQv4Y)5lTJ~}KrSQ9U=29ZV}o7#^$VUX+kqzpuW3ZVneQ|u<4n-SGx0#OnMhe=
zAmm$k_!=d1J815v(R-j>ycwd7K(!`<;c%ou5BExF>wYOr_=`q3|C)kjR~6B(b&bYb
z93U5h8kZoAz7rDtd$Ub|y7v0^=s7o+9+w`iYikYlrFND-3_swWkG`lg59ttoBS_{J
zl$)Tlod<MwXKu16f>2u^_L<90c!JZ7N3L%f#`>k~^?{jUZz#_Mv#AT<^9Os<T`mru
z0lWKkFGCK=Cv8}Ropm?^r;rL4h-|o4l1(+rl%`utjB&c-VK)ZND+$5j#^*K?tzgmV
z3YIhWp+ooyEDZ9b(<`XDZANYM)y5x69hqNS7LxW&eBcsX6>;-(NY!;t9Mhvq83*f(
z->z(7Xl2(2CjwXvG<E@xUyD-&ijf`@P#Q{vp}UY8YuU@qhK}T_WXV=93Ab>Er|zSC
z_Ghcf0L3sLX)kP)Mo;Q6%-;-Dot^r6r|D!Dw1dcdgrq*OHQG_BV1K^*z|!n%AUAFe
z#L3hI`DfF&s3H&F#v0&9+F+q~5YeUK@qefBs<tzJ9DYG47-KCZXn=>=1@Y7qI3?+Q
zTcX)G@Jb`QG!L3_s8P9fx{YWfoas)Q2Vv^VRWUo+rRgE!`_br>XohHL@<n%J>OM6o
zpPW$Ro-8K#yAw2Bk35$_!Ij5~!f>*X*$q>p4tB(<f|VNJVHJdn6Xek$r5Hd~I@7R>
z3d2e5ekbeomk9*Zz;{*5@%>FnQ8eOH|D!4bdf6&>QWO@_cvT^0YcyTJ+AYTofsHJG
z(O_(%K0;TyK}XCce;3!ej2lh1MPgbG7J1^z6>0`nuGU~%F$`U5Drr!T=0fUI$cE|!
z`Bt=OplERix+lff+w-S4Vf$1d6e688_?hM$;)y|Pyq6RH+SYNub(8)_AG7%hrFsNg
zhWGwsGDqX;s68i;8@;<1WZTAw_PER^*#7Ww`1(2f(>^e<o4+JbcNWHb7t^X06LlJX
z26+)w#wd6}<=m7uXP{Y|K%4rR5ZWXdilx1vRJ7#CiY6pW+JO^sy<CyHl}s?eO*Kq}
zzFfF5;<l`uow9eMm|-dS>+8LcQw;TRO7vjQ?gdqY8%9rpS$E7(im4XZoLT`w!<rBL
zPW61FT3aH@CDy680?GKnfGg^_is8e>THzY!S}C~9!?IJ#!XFA~5_`jV&aS-Q{^hCi
zKgxnXD~6y@|A)ltfA?;y_fadDsX!2B`-j|;*1bQpR=XmlKM8wCEO=Rk21pG)_i`%;
z4FpRMkg8{~cc|Y{&{s4eRr*S7D<QbGulyo2+Uzc8Di_NH!k+gyQLX&|{5MQ#cD}a9
zN`V)}h*!a4fNfc3J6`$4PtVdMk{+Z2Z{&S20Rd5hrZ?q?Mx{#)$jP)!;w=K3EJopZ
zF)NMs`Y*1<R)kac(CPDTcn-7(`rcBs2@pxWTaeeqjM*!295awV^Zd>N?x{er9B&uz
z0BPMrB|avSO^8>zw?(N0Q?V~|C)@1y)ZJ@}Fe$`-@Gdffdwn_FQGnJYXS0MU^(jZH
z#n6VdtDkS0C!U{^<!YFs$-E8N*!L`q`Yq@NFYdnC876V+yQO49uG9gmHWz&w8}ncp
zZXwxdfe5)UY{Wbk_$=!?h4zte1)j;AYTq+C2dJ<woV<%R1$Vz*&x%mNNqB*wK=)+H
z(r69*%tvBx@EG|c%yPcGtY^-8H<Y2sKW_3vUnTaSNm_QGC2e5p`SQ)9rDLeaP9?8S
zEOQy0S+bjYa@%x-wT*LvWg#ECH}7@JLvk{~UURWfz>Mf0GAHGzmDj5@F2??q6;WL`
z6GYi0Z75pN3p$;>xiIR2>^^Co^yNPotFh5wi`;`h73LA^#4J5qxeTiAh-SRqM-5zR
zqJ+E6H$#H+%2J93Pg5*Pto&H@;Tv4bWrW=bdlUT5@0Rn8c?$5cl(>hf?)~Sm2`nFb
zl}vrMTXrcs(G>STSe~y}>~P8_{8nc2)qv*fi`joSug$#Cl%d7!v=>o*J$f0ZLRPaK
zbpwiDtqFXH#tKdspEv%|=$iA!b5sP%zVVe<#)KCSLDfCLwsC@5uC?@;K$%##3vuxW
zqjUP@fZ<F#l4o9~CEDz?VB6!i_-;G+D)PMAlef{=xYM=V4`eoEks@d}!l?<KrsDpv
zA$F;pBq+>&#Bid(AbJ}7P!jbDhDr!ZVrsJ2-j3>9>V(*Emu!pC?{o+vkoIQ8?bM|0
zC1Dm@y&q*}+t(!$p}`Zznr8N%xu9agN-yjHsV*H7P6P8N!G8RVAv~%ZX~3)CZ#kBE
zZUd~<gYJr}de9Yg;|39N1+!ss9lXzXec~66MLh7SIxI#8Ay$fYYX_V;-h^lS=LVz~
zZ`2CO8_8ogBHqBf@Bh3F9!?*vP&1d&bBfZW?`mf?cOnTcXgm8VvD@5A-)fGStvy?#
zglv!GQ@gCPq&swiUEz3UJ>1~TEzuiY)~h#bNPJkSU(tj5zHbk9&+K~sYEZYN)8r}S
zeq3deWS>?2ii3C~=aA;?KCYi$`q9Kwp=9L=BF<2Fxn3dB@QZcSgB&n2XlT@XJMpI-
z@qgM|0{nH(CWuVX|ElCx1yD^P2GcZvvF?DiChEo>VXaD3hd$LXe+$p8z5+Jcuk?w%
zuJ(MY1(U3-Dz9aWa{D;7I&wis=PVVA15P69=BJQqzYsP#EQ=xl&$ffeEgIF)z2)dH
z-AKZ_egLrG^LDNoI~V*@*FrP9aj60FZBggF&sD6mpdf9STdP8A%i+}N_B3dAdJx`a
zs)L#EQ<lT0lwj`VXbaA^6;G{Sk<DaaW9`m;hp5^uR(Us8z_xUYR9FW5h@ok;)bN%p
z<jDOa$f~nR`g&8avrOf#&$k@Tm+yLRWz_1nZq2GHurF<=Z0K2F0^m;3Z6`{h4{Q}#
z1G(Fd6KHl#Y&ISfgWte+*y6Z_M<b!47J5Yey;jsW<g-pXMM5`}rdA7pm5xi6E#QRR
zNSYFS!=&Q?;(P3eZ0IV)8<X%%%bkc^LV>53NZbsC8N6v)X=nwc2j};Oi^7s;tIYVj
zV#XN`mt3LuzVX<%!!E1;v0rj*;XA?3JdG&*GTo`OQQSqp;ZG{;N9RZL<ex6Vo(-4L
zb2IEZNyh8Ovrmz)?}0|vXBPpNrctesQr6%J@?ohV5NhI9@qMIz+(d<Nr?v?cqoyC%
zSEszovETJj@zXaAx1K0Uozi{^z=kI0xA~rFk#Xp>TL%=P)@|rI`>8Oli)DdEY{s%f
z?~47{D;&}uoiiTw!IcF^Um%alrR00NC4=s!+;^g^55HN)K1G;j&k3LHJ2yX*n-acX
z0sr1N&A9`p1g8p@UIIRThQSlRWl97=)-planTd*j6<m=}!4;ljEkYKa;#Fg?{W7S!
z--&=2N$XU+h9qyXTn;T^nT4{(X>@qp+yFK7RMKWBauW@W2=oitba-YcU1TFYJ$B7q
z1pO&#Q~oqKhB~DN+d{f;Ew(q;(_hURLSks9xAv0LP9>P#fls(Lq#zMJ9u^~*7o(XC
zu*1C`t_-j0#FW$PJ$ArzZK9?_38r2asxEk1kTM(V*@P5Bh&r)u8DW&l6Mq2cC>PKx
z7cPX;bn`q_2oyte4u8;7UXHsHBeLHSN~hUr1XNTm0jNhFfF`Bf`s^g)I|)ieHRBDl
ztwP5qgcs95t>|0lm}zjG`^GV^!aIm}!7iBPRtv)zVCsLe5h5Gd$fM~v!*Iq$*3+c3
zOv;E}%!ACYn=R~ijqy*u8r4a}aQaBqt@`E|_plo4<LYFtF-QnBF}!tdY0erb=G3!I
zmwp{zFw<~Qcx`8&pn^zF>*&>9q^q3ADDEZO$$f<c!)cc>m$|VxTB*De|G0SmJ<#(!
z|2f3TJxZ%%zDb|RP3p#a!de=w&SOm(j-)6BaNZC9ufPOGYrQ9ZFnMzL@$z%CXny#c
zmQ-H40;wp4T4{qBE3j3Wzx4Lu2zk`U*3Tc6USq;htot^D>;iy29#;r7C#Ey$W$$jK
zwJeI}HGx9Vm?xq60m~OWqJp5`ASHDDJ7^d|6p}eAe^I~~`B1AG?pwRyl}3NQuY;Wq
zs3CZL%@1?26X*&??43@4u)c~_e=2!m-Wq_0xRiQ*5=2(jpM{A=syDhnkeNLBdUyV<
zQ!j<mQ<6f!zp)@40BA$O>dgI*BQ~k-UAl`Ck`6t^h&zng5mk3=h;6uqp=KlcNv;qE
zYg(5~DbDiL$&byx{06mFl^q36-Lq6d?ukIprPv)Jx>?Cv>5RO&Uc#k6R#r@=ADAis
z0Bs*aizrg=j!R-P`q8c9z$#pbusaVJG1NVhD!c>YKTA@#7BOICBUSFVxs8Hr?f3(v
zc?S%8PRH*+5DIUAVRj5ZF0lCcc@bA&^N7O-a<JOmM%of7y5El0u`usS*up01meiO6
zR1(S`z4@L(lZb>)gU5BbB1(rU&XhA@X-=iL)ttO(CLVCGtl@TBX?g==Yy@_EeeXQo
z1rVRfN(uC`1k~Rfa`roWNYwJ7FeJkaWkk1xl&X7kGrO@%XFF}HMMb?{o@4F6@|&+f
z7b2d2g6$A7!?%(1!wqgx#pp52dDaKYv*P`BP22f83r@`Ey`08u{BtHIS8?Em2;c^*
z)y}mD#^1S)fry?zOY-#tU4IlKqiwEOq5LPGH8sl1y;#4^1Nvn<Gqtkan{NAzu-L`Y
zk^Vdd{pH}qc<k{sSzA)8OP#h=8!mz`AN%q*+rTC`er|uJ*n2<T(i_-NOdatYpMF#J
zeO+H>RSOQodvrd(nTf&zBxJ9-@&9|t<B#s<WJ=nz|71Gx@@%Z}r_6JwNp&ZDBXKmW
z<@K^<a$y(>-PQf%>H)SblJo@E)LI~lGND7JgZsq1+-kUKO$<|{F<2A=o=e&~*DTaW
z5Y=g)y;xS#TFJ;T`H2i30_dROqzB=X)n72%H`P@=+p04-vRFIdqy1ns#{F)qoDd|^
zJuF{P7W@Jh=NqwLI1ybL-AR^c@mD{f>K<lT0}o^5GL8_ki4Aq+wz4p6tNb4LZ$cPt
zY7P_Ghm?_$P=uI!e4Ddvjonp(P#sXKd;0W^%`u!<GpfowHBvvU6XU{lIDK*Aw$4IV
z^85pkoyQ*avPWe`LI7PiQd4?Zg&0~r)8V3>Rx41UR|ohtuS(V~F<nOU&t*!n$a+Fd
z=R7RUIlFjs-je`q*|zjm5=ZX&ICP}t|8b<5;#_$S?C^vcy#4t0)3YAbgBV@c@tpNC
zp0;_A{{Xl`iQH#uXQvEQa+zs(;T;5V0yy<nS%+I@YJB^H7)GLPNeHo)#X4f7<`A^A
zv@0-}Q3(0&c+FUo^Q{?EwY#T~U2x0EunD_vey>>=0c_x%F#uG<O7Z`mIsdnr3iwnJ
z0S41)(ztr#fBKxQ+M#ML7}6dBwJW<Kb6~xl??Kwjln)jJ)Z8?q=McRYxiS6?QCuAu
z38~Dcor*4k)Tq#|V`+(A>^A2)^_b>xillAK=3S0+0#+YFESMW~_{ouYTAN?Uo`{|<
zLODlr`QyX5;a|H$K9+t>Rflf;&G*+wEn%^`fbJKKaFw{J9<E*fVF=v*X?GAW6tzrP
zZM{cp+KDRSr+M4@x$oK!Q*Xwm8IX+pl>AOz@oX4Pr_Slf+XJA=9+$ybMieb;PDuD@
zo5ag#lU|szBkPSwL#o7<>Vdp$n56qPKbt&VjX1d~CyfwYK^;14uX3PRjQLPpESP5W
zz5>%kTSz2z^>nGW8qbv@S24G==q(IrrURv=+hkBSA|lKO_INZ2?F0fFOC}!p?1Q3a
zD6YO5dFa}1p4H#QqqKIb8~oS0n(u|YO-aTV?UN0)#x8H<)Ua&+5}mU(F=&V1&TMH;
zeF3y^etEs{4Y+a4p06b^WVv_S8dgG!Wkc*aBI}_N$PKv?SR{&6(g{u4l(s4fOqb1h
zj0b(exq^!o0M=*x31$7E1Ar^sr&$RvqHjMKL{C0iXK-S1!sKkyukbHp|DDc;0<bMi
z4J`g8;X4btmh`kJJcRsalk%t#xw<*nJS(C0bj*bw-XaauKMrnnSZJjv0)Cixoa0^;
zPFj9^R}yw_G8?5!r|PQ+;5<(~ygUD4lUnjFsUy@Ibh=nxr=?mcVSZal_sAF6#^8|g
zKyD!%iY*~8^s<A$;<EBrQ83hNzIhzrq;(=27ogAprh%pXbaW2cy!0=dJyp)qMJcQ0
zTU&3h=^F%~({V^da8h>QBITz#=&nh1!*8-#Bs4mUiALbMhTkhar_5ObB;PNI?Erq)
zb)g*e0O)nqiv?L!o=y@JEyA$kOwsrGP0^GBz37qQEwC-q1Aovs(%?^mL=mf&9FrtZ
zMcuF3BLJe+BPgViZk3z?lB1nb27*}wY*>MZ1lv)lXK<t%`J&dpsIi4|f^uH0uNFYY
zr!)6DP!`Ms&|14Su5Q-KYc7N=1`Mq7k{#~lKPR(q?YPv?^H>Uhp|EZ@N@qSFpdick
zDL&Z~#Zdm9G8#FS1Jm=&hlY1?KL>g!D1kAPe$J)pPe2=AO*DNK%b(Dhz#msR+UoIE
zO1FJpO-28itnf$9EL$*lG2|O<`cyki)El62-@&o4llN$HSYS~6SH@Px{^X&D2d-ak
zjCr4JW#a~a%*zduOQECSJzNxSk8j5bi7e-a&VU^-dPtP(yFcFLOLknBNWh5#<Y<OM
z-lYMKA33vs^}IM)sC(IFw4c6=;?b3V=o5}oYMWG&C6;w3zJR#XBCEv(lx)>c@scZV
zik=u4Uh07n9KDZM7T5bpi(9p@PL-9_XQfUE8*LV+bw;Dc6th$oBlHKqcu*JYDxyo2
zPU$`ScKUiHhI}C4t3!cU)B2B`q;=v;_9cPXdNw4voX7lxi14!U-fQpKng=g9VX+5+
z*hvM|>DIAA>z|OmwNVXXOLD*Er*VD9Qj`tcqRsn<lAJ-nrBOfEFLe`-Q4ohw5R35&
zx59uDA|3B;miC2$AtWE%C6%j-t#mkhs)f@Km@;57M&VQ#P<2}=7ljOQjXZyb{(=UN
zkq4I%XMX`oD%Z4cLbsmhl29!Q`DK6zEdqH@XXU2};gh7VgRs2<wK(&HY$PR>SD<1?
zv5c4s_*UdCFwWvKz5@QHIE>eDE12)dMv(*OJF5{7Gg>xyNTWU20?|tINFn~L0kYm8
z@>f%EWK%F!6Pi&IX;TpPue^eC^gRQ_7X#d-h8ea2LC(PhJVsO;Go^urzHQ}69rMUR
zLnsdg-|QV1+?@_il51dup?qBj2#2<y>96~(p>pq3zCr^@2EDRRn-w8v26W{`5D5Vc
z3hGKS%HTBbixi=ma1oHboL_caKMFjEh%AGP--U;y8=&?WMDD&3ETx3Z3%KI5cMwme
z08$=sp%`uu$Zzfr3M<dcERk>RyhDl$Gtfv7xcl@Qs0XZnz~+aHrMOVc;DBIoDY71j
zefy#h4gZI12h&l85Vz-{MWO!UfHa~>=L*A#gj~@drnpGhEopN1ys+Uc?_)rX%sXC*
zSAJ#-vnCUC=N4j#C~TJIn@kVQ_y&eL942C`#SXpTBaQZ*EFayqE!zPR<QZ4f3$D*^
zo7i4@Uh;51;hHn#Qqkmc`|jHifyrvwZ3j<Vxh5QlO^*Z}fR5V7{gvd?cUK*3N*?Fj
z8wpKlDZNc-x6^%z+}+RlBn-G{&E~!cn?5!f_96;JBQo{$Kt}NPr{}&vBZ_mcrTvn~
z>N64^*zjS}^(87oobql4<J-Nf;GY&o6pft6Oz#qJUcS{@+dm|^+)zvT;G4$F{d3nm
z@EwBTO!Ul;NEUtK1&-;Vn`z5{QoWH4k3zOy*cz+icXGvRJEqC|mVr6;r%TD0M0H*n
zeS%3lOoP?+n{JP<UlJ&CLZL8p?7z~0Wl3H7py+Dpr`ki)sGj9lB6sWe$G>SRUKWuN
zzomD=J}ZTYIgoJ;6=yfOaL?A~*!VK7w#RMK70KFGV43hx(^q9q^2^^oB-qVtG`h`#
zW5&%3?-zrUG$NWC>?A7(cwM<a25O}WbG1J<7Qzm?f64BnbWdX+d4i@uG>+b8)Jp*v
z6>3+r3K^~5_bFE=yV31jHMmgnWA!&z^okS@p0}wi!@Ik>YfQKzBSe`}0B<KSVzT}Q
zyz*(^DjAD)o0yZtmJj}^ag3>Y^dBKZdptkP&$oEaukxbSB7Zl^xIqZf310`Q^886B
zWPFGk?h8Yk1`8Pyk2=PaA!yNjvARaJt=aX9IOvQ=n~*>%v4lH{6#;9Y4VTO`AI%BW
z62k=p*AJXuOHSUkJQ}a+#(ByW`Fvd0r4k$wEnOJFioe=_X!&=km>AZQ9;0dTR@Jhj
zpt3auqG7>wg3$1`3nGPUtEA!#u<>!x^rxoz-22g!bU*#7Tlr;umEQ2sX{EPYslfY4
z#L!ba@AFJvL3)_)&I4eb8dWICT>6Q7yoUpOS^oq5a~exiT+Uwv3sv}fvh;U%9Vy}?
zle;>!JIL#@d&x}1rG+)sS+Rs31$h{9-3j6f_Ye8r!8d?GP9Ll#&F7~T-4TaYA)>6;
z|3bR?wL5&B{Vbb3{4HryajTi8rin0_3XIu9%Z{K`$6s1c{E5_CaYWJc4)z#xbk6?x
z_sks@kT0|0uXFsZUFx&>qcMHABrFDdnKS8m6Ggf_WU;Em6N$RAN3S?k1_Y*vIxt8F
z27l37({!t@4$(@5CNAES<}IMcjw~5bfAQwq8!prC{>I$JjYk^^yPn{vtDW-q_y2^{
zEbH*)C}%6{mo(kKGv)=fv!{H|ZZ5=FOAaoH%Dm3B#WF4`di3l^;o4iksKq<G^DW3L
z#giP@xWqm~tBi>(gM5Qab~95M;|BSxPvEfDm*M8uXD|)6{T#m!X9R0;27?cSr5)FD
zl55D*ezK%m->UE1sa~Plv0j&8Fcsn2;bDZc&Kk1#T~dAF9qVr>9{)J*AzW;=m9Q7)
zoFadJxz=o0!fXAGMmP7A!g%0Ghm!!wav{VHVsO+*W%}-C(^p#-aGtn=*~i_YvHffr
zZ&|O6sZXVa>*|{SDU<%X2kDDN(G2ByPL?U$M=!7ji6ad*5kjiu$7}}}=n+xo=uRwt
zQ~ueqmWNb1_`y!g(W5xWpJk-krfhAdvQEu4tl%<shY#-hgb^4q4eD0xzCeGUEIOOw
zReh;~xsaKJEPYeWs_(8&n>GW6!B;vfH)+U)MIzZ!Emy4(GzT=Q!XdtkDedlAyOX1I
zs<q*LJ3_vFyIu}}LwCGFO_(XO&3@_H**n?u_Pn0^EcSPgmNO43;gyh9SZ?*7K*e<I
zwfHCQmN8XdBenIyiPp0D0x%pRd<c@mOImT=^fHN-Zc;=gH@?Z{Vm{z`<20%Y!#WtK
zbtmY?)xnwk&h)|pckB&27E#>^=e@Fwjm#$u=o2w@Sh!WhBzoTTQ@dJU>_rBp^oF~r
zrLvhOu92PeV~`)jl6@bUkPsVH{gm}cubZk1*Lgh)uTI>YZ5MId*IWHw-J;yGI2w)P
z>m0h5M>=oTt(+fTKKgZEbd7qV2d?;9z5I+qSenW>s`Q3m8>&zEW~=7oKy2^*eH<Sw
zT%60K*fQC)xE58GJbGKH!1rQRRBh-4B~_#h(i4fCo60kLkKfx4D|<allyvX^xXYYJ
z|AL9&W`ZJA^wnr4PK$m>iuj49@qK`Es=erotKahEsCv@ciH%=L!73HP@Acw0(UIj;
z&mE`rkG~#>4e+l$Tb~zpLSJF9QnNnj{=PZ1S%>#g8^@=oJO_DV>)c>1dnfndQPgL|
z3iW~F&+4w*$nOd{gCYa*%hS#oj<GzVDPk09I%G~N)(3cFoenx~U;2!e`FhiE4tS5T
zL(H$)cFdGUQMZW?*kQ(eKBijw{#>AI1g9^A;lt7~<%?qr--0Y%na0LdeR5NUE6-Oy
zzinseE?0L8=01-jNICbw*r!Y=ZV~JKR+?x_ye#TFW3nQC7PK(5L(xVvMdX!Un7g>L
zB@%lfIA2xKeYqy`d)-Q}xHwYZeD!YmzYr`y`AHS`KYW#)EF)eGY*cWE`1$8}^Nel~
z(!8NxlZ^N(otRUnDU6YZc*Ymue^&OGM)(n5s+;g5Io@K58o4yAXKKHpm4iNaSO&bJ
z?sw<7e-^c~C@P@8Nqu);h<@m@4d(Sa&v-xfy1kkH+?s9Y3%nb2&ggv#<X4z7MYk?|
z;Pt3IwOY4yjkS&~##ZHYL;m?e|Ff%hHiMoHr=8rbe_*_Rj6e9^RbZqn)#!UEx$~ob
z(mvaD>l0ci{+;zTLQK3{&-vat$=NnejJyWF;@yPaI@yDl=G*o26&qCS%ZT!Se%Ajs
z$uN%b4{qdc*Eau$h8|EXKMS;-SO6%)f9Un|1_0T9Xxc_T`EeUn?D8<-qs2BL20Gn)
z&Tjn9|I9hxP2w44rblJ2ps=)QjhwKwz$b#=y=4Cv>_;aXqtk_vWdBY8|G)ltl?{#I
z!X+tPRrqhR|DVjS|I5E?_)xI@fXBAW-tXS!|8}YV{Q7qx;DYrIKC}9*()oYh+`s-s
zypoY-uJDChBZuDal(_%#m&q)J#qEx1dCG77PU7{Sujc9~bon;6$69|+oc&L~fHNty
zMirEE-~GS4eAB7m^1YnCta1NqZU%yLupo}1?|0zE_N;GCNcNBGk{ON0yHh~`cirJd
z<E;=@(#U=PyBF*?!~&~|OZR=~-)|he(|WICe1Io-xr1S+utLH}FC(GgPTc_#a%1KD
zxjw?WSw5zxl&`DiB7+LV>@Nf%hA~+BtI3t10SgVdj_LYcp(kc<xTC@6JQRGbu+NvX
zQ{c%7*%H&VkaztQ2D;~FG$O^P6$PW5sczPHM;se}uEr*ruYI@Pzfl3GuUx+_4#uWd
zL6gI$16Z9omq0f@#?^k?DX~GIB(r!L>){>N-=Qtj3E+ZlkMGo2{W<5se{_u-!r8Ha
zDWp=a3&2`*;ebP<Frx)^Op>oK68MYyfaN3s<6C=VS>BMJ9(chy5XQq^8v|IUz$SAp
z6y@71M^HXt>LEY!fL|aC>|!E7Ou}oaR_l$Vm;-$CbN$-kLitGSs9SvZbU@1X9JngX
zQz({3K-|JSHQN0KSP3|iKX5d02Ufbxz?+s3@^3R!FRZDdK?!aI(5i49Q0IYDO>fs@
zy=>@-buw!`*u$E^br`XleiY_&cE~ct8NF9VQ0)O63bEy=eVnz`j)-Vo!HvKL7}e*U
zW5Dtb14dT;D~BM}wQ9Zi4ZmxJN98I-Qo$CmvU>*E79v5zO$Dz6WbV4MTwT%0EU}54
z_gk8lMqI$5VHfZ*G0#5gH~R8r?18(73!ET|e#si2Le?@<Qa#M?Y;6P<Qq{tKw<(8<
z8-pi?uT5F*U*69{STx*QkasUAXK!`@TShQGbrJ+PW>;?jgCp22&xC@j9y6veK;EW?
zAWp(?!tx{gWaWK7TB`{lbEjyy*jGrFg?(n;)wKRn?tSWi9p8S9^YLrxNeNd8ghd|@
zprVC>hL;5>ipaJ+ZjHR7_ctj9UN&zS5P-7ItSGdWZ3w7kKObsnTjq$P760U*_PcO6
z0m6W0px7X90KN>nfXiEX%%bMU#N^)}O>+kBsW5;kg|1?mJ$pFbngh#8$M-BFO3x+Q
z)*u~G+=pkhNQ%C<h}UiI6BXc2g*63XVoHD2#zD+LoO>hQ^P80XY;VDk$oEpDbe-j$
z?T;^qHx?%1Tm66$lK(~3ccaJ<(2g^)m2}V@8RV|P!E=!MWGcW_{zlc!)8+txtvx*4
znQ2@my}CH)3t$ewss-Srk~{eBG)A~vMX~OE-mO1<4*%2c+y4^WigHSAwST$9+tT}g
z+5;GxduKb44vd;cKojGTihL~b#8ie<JaR4?8Y9>PD{vWiR0*zN)bY<@v_LfRD;T2W
z<0q0WNXsH%TM$xm1G2{5sf1{~A}V%hEQMvEzb-cPe>oZ&D@rGPe*a+FjEq5nPKZ!9
zz64nOcnlW<=IdZ1(OL;R%dN{yjH^<BafBVu_~|3{uivO$cpv`lb76z0#m%{!=^}D*
z-&CssyB`Uhxv8Kb7}tO_QO*d*EBeSWRmeyjGhLw^$L~}qr7`RV#UR+=$Csz&`?$XE
z*2%R<->L<rO*3FFlLDbls_Nl{NTZpbova^kt&(w`Xv-ctnNCp;Qs0W2ASq&VOmVKT
zG<vH=^0%NC4XSZ0*j9?$_00bS@L>V3&>&8IY)kNz+%V4|+>HKy&h?1zWdpL5Apl#w
z|7G1V1BmBkPryRjD3Sj;AHFTy*#X8i9U<)ix)kaL-WB7zWmB!#ail~jmVnPo&8l&4
z&Mx0z=a9+PZjSTjCkjUmIoQ?y*oanxhG!_jNHSIh*n~17yQDeq;>^RDd&r555%0Df
zVq%yt%G+J?p8V^C_+L-6fC42@n~iPj>%RNr@fByf>%S|ft&Ql1rK|yqOzaMs(?Q)3
zF+|-K*zV_A$@FhPqr;CY3YNMDQ72DcnG9YhI$Z}l)gKeV=SA7rj5+!lqCTkmnmNcl
zFw8mrq-aTvrl@K}8UeS=xL}IM*#UG~@!(9rVf3)lW@rVZ)%MmN@Y981IrgtuK6FrP
zpY`%V!^63to1*qw+2(!eRJpjh!Jnkw6a+a3OUFs3XPZp+@lVJkzSs9XZOvMYFDa6X
zrOG*QZ3*!VZUh8DL9c?ab8~P+u-BH!QQfjN82XnNfJaOFW9@1{Hwe3%q)F$)e%_7i
zC%shzX>*`uHxZlI{HR(+g=r19Z5J@1HdOA3op0WcHG2R#5PB-FL&oy*E^68|J&Rz!
z-yFb*xcYXoHu70R6VS7fKoYS%@Z;>~)(dyRb0-_?5PoicVaMEK(O-bef&`vuE5`!I
z^zqLkLXUx8ZkWIi_vSI809&9`!3C%6OA-{8bPzQ1Q2v(6vIC-Q5Ob=K@XAVIZ8FNE
z@(1^T6;q{Qm^@NHFOatnxM)g<8!_E8$U!<qT11OnhoMJ|3W8i<kLNE2r!-Q}BiBD`
zoEXO}Ja7O!S9dQDJ+9b6d{wpkJ7wzYDTS#I?8vbFq_KSg{M?zUXSv-cd$KUzV;7XH
z?>nBs+D$FO$v^Ep^pxGjh2A)=I>LFJcsUQi@iXn{#$nH=x<{a}+V23%@wwuxW&Z70
z6NO-v2db7*3Wh`eEGwMvqGI_0M^`t<sRaqSdLpqoJw5_tx}<9}1~~&@v0L~;tC_pt
zK@Wv&S80Tv6D>A?U4_ejR-B9hI;Z>>+aphuMNb(ZuSMu+zypGzppR*6L20M^fbj2*
z_5qX)J!5naXui>2H+^PehHqIM{=5vbC<S)=okQ9X9fre+)?+|c8k^xO7!`D)w|Jw6
zlR`GafSCgSC?qQ`NkzxbFvblE8c)v^@V07p?~3t!GfT5wMpd;E3%s`93XLcRDcA&W
zo<mxqIiP)V<>r}5irKHHnzMtb7pFulcDXr8juM<+4}t0_Q*O%W&ZXp`xN1D<o4Vbw
zub9TcoH)ouAPgccsN0t(Gun5t_e<zasGxZTlfjeN+^U>$-LnHF06wJVSBt<vS%pk<
zMU6$cIA98EQVddUB9=uI0y0bBpm&S64a8<-u8WO}3YP|!iIM2pX))Tjs{5)jE5OP3
zwH_+mJo79THJ3@2HB}$cAND6h*b+^uMUEo~>J0&!bmVzrb*S+pnM<I)s6LeA9R=#L
z54z9JmM7H<r`~)qw~;=vcK%^&qG(cBKd>|!EVjNv5$xV|ss21}Ge&S8i)=K0a_#1}
zLK|CO4gmD{phQizl~*?PM^V_lPTvi~LI6HJ)@q-w6}1EOE~Ar8Zp&j}nplMJ*4*(P
z>Gr67bW#f@yhu<XZid5F;kBn*E`V)nT72qAc*OxE&lyB1g(-~>FydoZ1WRUtO~s<O
z_HzT)TNINb3lyvuA7)RTfej%BLZS~y#IC%8t3$<X#6`04My|*apKP^%>&i;k{{X6i
zZ(nAs|2esJzK{C&gnpfvv!n965w2e`Ft}(>ieG(+q(%}DT*7WXd9XyR?50EffE^b*
z;1zE~2UdqjU7}o!0wMk&$^0b9;?Rxiq)JO7C<)ICHjtPz6`seBPsx+nd)$5yVGs~k
z3?|~CO8nz3rVmXdf*&IC?|3Z6bD-?*Qw`AQgG(Tid^0WtQd~7s72K(gB)h=UjUqu5
z(MBxM-=t%vb92B0v0jl~+1alc0~uE2OyuM7Ew8I$s{EQAQBin^q`nN#loYZsA(5W2
zk-z}1gLd8{;HckR*aDN6o^4g_UR=K@BE;8+C$m8u^)$IOu{2!{1>SR0K-310q>whH
zyG3&;b-v1qMlDU#=BZ+MaS5~w4yVMH37;|#$ABtMjI5V@2vp)xJ5s@2h}VlGZBC-E
zX(hh_GT}TS;XKyCQ#tnNsTF%~0R5p}UVu1~llsTO9Ebv16}l$6k!`TciXQjdVbz`n
z=6iL}hPekTZhr(72rZ4O;|}e~TTl&&VmjD0z4r1{O!QnV@UCqO-5p!4o9uvh==;&n
zydb)&>Qe{7RAl7@69*wHLzxmmI^t$G9Q!_2qsn$)f=xCnGB2uWG(BQ?W|X_>p}&PI
zB==NP*J{n0+{eV~w!>PK?pV-59^62AK>gs26F-kWpVJQx!Yg;um*peshG2kk1Cj?`
zh{`0s0E18qPp!vJjEziHoN+Cu>=<so8n7k4hlOi}h{2BUj5X8;59Ya@L@)hM<h)L7
zubca3_20Z$o~6j7`(1}Z*N=|<{a2mF-wcse12o0D2*wbP&ztBqoXTK)GN9_0R6dO8
zYylG6QX5^3^Ab?fw7*f;X3;LIwji)UJk}eK&bJ`2wQ6`yG(mU!pq0)|CINxCY>8yl
zSQgN@fRZ()V~fc)NRGeR6?NDRFw|KLv>+bqnx?Vy=<piSPjsDGaPD_i?1lJNS{M$P
zlvt>s)1RB;jv1~=Wj&T1P_1f%Qrv2*TT8xC{qR+gBd4%tJdsK0mOt03c_e0jC6aST
zr~~!V6_CxwMsBr{EPT@DPPwZc{P9(-X5J5r{b;+&@c03fUJKMmWzItbXmHZYw}l<j
zMOfBEhx=)W$Knn^O7@d`#%ST!izdxk-=dD^1KDY$wcOiEy%9+jHLVq8fv%|{#OzA0
z+DjRX3awC%L_1LwR6Un$-Pu#fI(~5Z4K4)wh%6Zxd#&^~`f|9uQwX-tfa=j1oaaGq
z>%-0qClDeoLlh+MqwX5$dZF6q{l+flvCKd=;$l6uQ6LNNXn0^1{^=C;^QNY4!Q2Bg
zRXbh>5$%>L(sHhCHa<6ATESH2^#e&?rpRRq_6b|wIkwcV=`%Vf^<UVO7-LCMMpte{
zT9A1xo8UU7$P^x(cA~9V#ABAG`BYOJsPQGO1%K(8>~@)IPja2SS&WrGfXHnNA=PFE
zDyT0ZjU=km-r8j{Vm}6#&klrsy<OBQ0;0rrDCWUSb7ZOFr(6w2B-$ONTe00^<3{lq
zmlqdis(-JVziIma{=sT0TDa!>zn5Qs6P8JcSn5x3Sjh(CMy5VveC<p&XP1}z;Z;)J
za4TlEKmnx|?zazL&>bSk_OV_*!X>qA*Z*;VtREd5O+Fa5Rsn+4Vl0C_;KpS~8$^pn
zyG|0JM*6-#1lA@|z7LMX<m@b~^Sg65LrK3dYqzvm)w+GRy(4pyrO)<Cqb&U8079oc
zuGI=9oHsDLNo<j#6R>@^j(u_#*iCaI#l!C<y)JorCPnIVvDeP3HrIZGTC~pjiR)s6
z!CURJuNDKJ=~mWka?;Ol&)I=meqWTiMsHnu;vtBn)V2vFLXzw&<w#P$2v*Ioi{ov7
zo@?IFS^1?D$8UV&lewZvY@r*)W4yg}!0lkZl<Jlmt^Abd8~cKx!XjSg#apE6W4TDe
zH6tfIwytd(6+Ppb1|NDtH$CRQ?c}6MAx79}E@g3@>Res;#uR<hXie+O9(Z9Y%Y_zk
zokhILWr`(B{N<1G@M`vnclz<cVrV^eC-9##NwS0a<z?k!uXo!8^M*0dbe9=O{vXQn
z$KotLTS~>R{tsto9T(O5w)-uDgi6W)0s;eqbc%wsGcc4$Bi)TO5&}}vEiK`|C@D3h
zgn-f^NQ)wfbV^8bp0&4ozwhUF_Un5-=RZKjnOSQ+&;4BY_qy(1hiW$=oQY>m*gnHW
zZ(ZFX4?~a`j~Pz?d>yj^)`D*ZP9&gQ;``$HEd->?y;C3I2I*Tfps*r@?A;?gEXr)`
z>#Ntj_?DPcZ_n(;StdR6+^hKf(P15U)-+e-X+NbC$A9B8O?}mO-GX_6#ei|?51S*1
zN=yc04^$`=h(0`Kx}GPmmsRcf+_}BWk>}{%vZmT`0<1cp$Agvo8<?~N<ASRj7guRn
zO%RHHNYPb0gc2fK;H6d8?shi9B7gYF=@x#sW%q?lr56&t0>j<EG($?ls}OtV)<DX<
zAJ{Op7I-rr#fjKW);IC2iO$kjbrRJT*`ID}#+|0@#7|MFEAb;4x%tex(E*3lxdzb4
zrW4UxM-Iw${QfkeCEqkVP{PowFe}?m_TLBXCHcW#&lgz=n4#1%Yr`-8d6T^ry<>Nm
z)eqkXOJa8V+HAi2h;Kj4HA5${j$qU3o(_NR-7esJ6wj0H<Noz+Kvv8cVh>xlBX6sd
zrI`0@-pi;YPAz4pB5DR~B@QO?5l(74B-DIwv*|}Br9>YjT8e|P-x!NG!e&jj?G2e~
zhkNUv@94N8R08g$J4B9af2`!g?D&F(stp)3?|VcqT|6;X4d?9tY}33mk$n0!)$KCy
z0_l~Z$_mmeRMaRkj8~f<bqg>4w#^zQmBI+5P(`;~9+|%+zap~Ei~joA7}iV4=^rr(
zpyqsw3|NOBM*|TGiq*Xig3_}6Tn25xFan-_D9VU3^eyC&Ou!+Zf=6<ZbKQJc78Y<y
zK7~HQQ!H3^hw=w_v^0uuRK6(TUHJF{cwe>3tes|{_<-aed{NiLCqw?2l*;gQj@5s9
zDS;be)5`Lnn1HIc`!$cgEWGjtk|**(dS|O|!Y00);(~6iij~ep`lb~-OB)EG*r;GL
zRjwAgOD9cQ8K%V(@gbYw5$am<{4JltR*QBaIEh*|#)LTef}qZoD(J|McNkJ0K-W!O
z)oB#sUPj*XbZZC5H28{CYiKsJO#YC3Gj@hHkbdyZtl9K>+GPT4Cs<GfoQ#-Wc|0A8
zI(mLOClO?UQIp@l%Ym5=y-cgMbl)geKm56f^#?{me@*$Ssqs;-8X7aS1?Obb)4HcV
z<V?JjEOzr&?KlH0iVIo7Nxth~(2s;ae>0ZVl9}P>$`~({fT4qJuJHb0G})R^7a<JC
z+3s~HVQhvsf{|1~+#aBd1l<(oBKzA)$e}9hNC_3Kgt<y8waCrT;ONIj4(t7+&`nV5
zb~!nD+@H#FMWF}viD0EygyVOLEhCOWoYdGQ{rsX|ori#aWyb8GTLyh$WXxzn>39u?
zXZ%S+V}{f|95Xv<lY*=g@-sDnE~&arS>VY%dmoGP8ibEjjrS8#JQAZUL;eT{<`FrT
z%UMW;AI~yb`4NdeaCVh7HbbGA(y|zIl7n({N=eYO09MC$z*2lG;dT78`n(_4F*))!
zTR<orSYPj974bBxbSybj+Ot(v(KT@iFscl<u+hwrl%Gy*KcD@W;bVv&WLo9S8ig--
zUHuM_&c^o=eW=UwMB|2y-bf1BS-{#|QGOc@GKI1Jj?S%95dgGeN?)abrdDsLw+A5u
zcFN^#EhBNMph!uS=!z~8dm+syRuL%olPBH&o4c!jG4R=hEXi!|>OSLCv9W87ndUrL
zEAeH}GebWBMS{(c;Yg2AJUirM<KAPsxdPPSaZ_KvkZ_}{9~$hw+vd7wUy{9Q9+;tb
zwW^xrxLuVoy3qV=|M@WiZ4!FX;RW~km{A2i9hZj`<llgQ>Wgkw(`S}%D59^nm@fx~
zu;irm!QrfD#3|VCd1ki`M*ZdDv<|?VReqy+ti<li-D>pS7qw-uGNpXoVToi<Wz4vh
z!RTTnaz7F_QRS5J>-_PL68|4xxcT9{*ft7${dW#&{j~sv)@)=)I22@z#b(B*^g95t
zZjK8_C*wVrV!%r42acj*@&ibm>#!AgQ0bF9dC_m~R^$*x{Hq8eQ)1pT?o8A7!a!&r
zSmQZv$*MjE*Av+ng?iHIpDbK{|F#S^73h`*_ihyDL~~{^9jnS+rFaZ)h+Y~>-+Tm)
zx?a<*!ba5ENVv^pEPo7_D$65`FX=ljf$VAZof6~=w{suRvIi$=dp-U@Esu2uRrQV!
z&ADPVX{iF%Xt2kNeF4EPA)(CEsx`=;k+3QbU4@m_lOY8~->cjh3#`xt&xd)ueg!oL
z(X4^;o2=>X*|*@wYrxe0MY}V7L?NMq<ybRUqoIUbO{20M;jB1WT|FsST$P@xowksw
zS4g=Ijk%gy$kxrPsxHfW7#Xu7;A-dI474B~d;zLCMZx(Z%id4&$%;zccrI8}8P5r%
zd6g*m8))w5W}5^=75q~{{j0S4VGHWX?fd4zew`fe;v@x2-ruZ9M>!#UnY~jmXuBj*
z-s<=a$Om5w!Z?SotZ{1E+TZ}eXXxD7-ndz;uwqcbYj267E^?@oLU-FK#}%5Wr>hYg
zeD}&t_y<Z#Tm{9(vGrDl_$P~k@P17W0?X}QK^5$94z~xB41Gsit6F3Ac$R@lK{oRc
zvTEqkOzaIO8znwsZUg5k_39qsej{+>-JFSa76eiNEap3yiZQvtsYo+bxS0}75V{jO
zEC7Dto7!+|U5!QNWeHGKMF<~>#$*C%w~)v`8AlY-PeaDyIGTXI-I%>!tO4dHt)cwk
z&*7vXP+`O34V*|XVp{9S(>^-f^7BG8f&bJuRNbBRS~FzbK2YwRog<t%XG-nEMs_gV
z3A*4~DPOj;cx5|KrbQf`oPLL#KlUh=ASa2)l+k4qq|$2Gq*UyJxpD!(v-~u610cdB
z#aXLVi22i#qmFCh-ky1btHQ|{{6;s;zT9gy5*Gb(1Y1jyq<b<0%1Q&({sqx`^kxTx
z2kDX%$ODOnMr}1L7Z%lO{4X^au@aC*@z?2Y{*{EEb3`VEWg1xp;?k-N1Mj}qOg8sw
z&|I*-%uc;-u;kEtI5ksx9C>Z}aiTa?528WdbnKE~?^jiv?IUc*BlN8EiSN-~bl#lu
z(!K2k1vRSp-f{eJUdngEXly3PeY&>BRr2cLyO)p@aGmXS>~LWjTeosG<Hd@s`j3)o
z-<>a^HFP`J&^oGhAP&jbKUi_-CHWw)UK}Naea32q@jc_k&H!|LS7O>Y=rWw#vMCb*
zy8~p)Oe%K$Z9JZW5-?M$p$0#mvqUlbp!)mD%n?wOt^saxXsA{yosL44q0VxM=~eX*
ze%?K$>FsRLWU8tuztK7!FL!O}r;L`pl#bWpi%>C5Av5uj*h}%;s~vTq;L3!N(S$#7
z2V$!T!#e|(f#cdh{(>j+XGdTmNeR|icYt#_`%{R-9c7@NJm;LgFnRZ6rwxiE-H!i$
z8~^8pBgV#r<(m8tDWHp!Jk1uxKZLU&0&jmi3(gU|g{@=JCHyU2w2cExf`?(Jh>E$x
z?Qb+$#a0p~z2A?5P%nU;N^f<d@>S2{JnMTQ&OMKV_CsplOXP`RniK?FM60s~Kmz3J
zVv@*t&et_Vb<IEmsrY8Y2EQV6rDwt)Nj9C%ZzZfN2dcnp>j!2LStj8Ir*&D9!oF6k
zP|Hr@C)nNxHkMY$^GYELcS?%3jtQPWE5_`jRW4TnzI@s=S+Aa0QEI6Ai)fS+L1m{O
zGy((BK9FMBs~z3@Y3V?^-gUmRcvDVq`qmPtokq7sQY7t-@AXYDAe0+HCe5)Y`sF+`
zcIeiWUN0Vk;bpUzp}Au2Ny|o*+hu6nhzUYsZ>CNRq)=3Qo|=D3NIp=O_L*sQd=Jiy
z`S0mc_eOr-M3?Gd++<5PIWyBn{@pCkIZq0W_zdczfqIJs$xj8LxFZu#+yB`@EF<v6
zz$O*v%Rrj=1w|WLrTvQ_4cXF0Z?N^rx$}YYRv2X}eJ5$Y^5UJw$-u7A8IYH?WuO#s
z5zGz^t0|b|?;E#_CkX12Ye8?}QoIPnT)DId7lk4)0#@X|gX)ur759P<KLZd_tBnUo
z0x^9%!s*XQx%y1_4%%xT;5Hh=Wy+`yS`5SRBh#rakX)mK#AxIA-9QYA81|i2$Gcg>
zmrq1sT+#g$g~rBsEB6w?F29NjY1Ai40sWU$zzixJRT%Rw1R?tfSv~Q54j3#!8EZT^
z9N3leg_r_AsOwSF=e$N(hk9r@eG8((_y;;W@take1<a~-H#P5D1J`HQagY_$3c`Qe
zOfapty+HH(#0RriIf&b1yu_#2LCHM@;FXp+;%UU4YhH6?3x-%n!Lc+Aa;`py0$whn
zEWw<C$Ouq{Lz7<lW_VNI-mkvcpd58+X9Dc23`(!JopI#W>%b9unx-GXxh+FeK%Y?U
zS^4_q#TO{rs$&GSn+>#T2>DxB=oo6~dYliO-`_Uj8UPs^66>`W+TS3ZcKo3kn>F~O
zQM42`-IH$#mpCJ@E;!s`;?!X&<LeXph?`gzD4*vi_7m?Xmsz^Vbt6cnHXfW0B>Zzw
z=MI0`bSS@_Dl?Q3ca<x`$gk2L6i9)Ix4g4IYJ&mJ1gyvl6h8yyy;@jM3&}jAxD`%7
zpt+EE&A0b^2kueJ6#PX9l!&uHv3_}7?r_8$Z7dV*cxz(dDD2eK(8|z3%OtDfhILDD
zg<2Bw=N(t}?Hgu%dRz(?W!SX<ubX;Yn95l0`7V3*bl$(8L%2`$E)SJ>o0>r{Ka87K
z=WqjCIQyz=(r;ti-nhwBf9sNoxLFUn=uoWHW8LIXSnsMu+OjujPB!+jmK^D~ZFivG
zc+*lJvEn@iX)WS9ZpiG?A?Kw^AiXeedwBaf#Y+C1a%~B2HYOuf-)@EG8bIn{b8K2A
zm5%Yb>^BGEEst~@f$-OYxR*_G;PB3S-eui_Xq#Z0&^>K`p_$C{X)5-Cs0#B*i?4%!
z{@3P5wHG4QXq8$oL54HQ$lHW#m6De>53>DC`uipQ6`~LPFAiG_+A0ocP4;d?9P4y5
zyxRDEQU8DS6=Aq>k4T-kaFlFldG%|)c-v8-6gVAbK%bfb7Ts%fL(hA9^(Cxl)k^iz
zF<u#x_#)`iX?zWNXopu2o${qqszFCqKLfzudUuN1KRE(;A;CJeBrfGMmR&C{I<Edk
z!*MlA14A`ZEax$2A^9T_iEtIUoJLMaLihn*h!=VKt%LcHQ3n$^9AceP{c0DPjWP>!
zhr;G3EX^BD+MXU<bXPoCl3_EYf+r!3F2de(LtHbD|J%AD#XT0o2SV4Vh~srh<z9m3
zN;B(gJi-7`5Ry_+A*|CamvW_B3-(w#t5<;(?@cXBxUrP9fHYt6@I3n4ib3dNw-?#u
zq#D;`f?)TxyZ&$vu-8(F#3|6+iMq&A3}sY*hpi*c-$0;iut?<Q6_7SZQ%bBFyOqox
z=Ded?J=a00)(&L&Aj}^gEx~&qVhNtW2mv4a#$`k8hlBBFNI*FmDDUL#WQ^i$$AuHQ
zKj=0}u@Gs>XS}S(@)oVkJ<4<$esG2I-9$M*tC}9@rYzylSKBXre5Fzr_86ntc`x~D
zBg?xhyZCqX5T$R$`>CaVn5`oczBM68WjjrI?#WzhO~HtAXTBQ53ziJN%ThK<w9OEx
zUhTxNc}eC<A%;(j#%y9JN~Ds;T}XhnVoZI*1!G0M%e2lk=`t{;HZ(QBfXCR#?^ioO
z_L#dOTd@b#SKO0gS(bON9o9@pJx!2vY&etF-T0Pzp)P<+F(U%dvPH7<YscH?y?8kn
znCsv_w;S$V2eSZ5l8GiZmav&tO~~^*3%6I|aL5k1hKwfPe^&~Aqj7bKBzOwY4V;^4
zMDSY<sOCHSxlh3vY#6zSRXIIK9DSw2g7U-Pnvzj^db#<I=KqLS{dp!4<pGEoeU0O=
zz+cGCL(~(hWd!v%j7K$TV~?7=rM6d|sG-ufi!kq_q%~`!xO=3x<1Xae2}cHND>At#
zC#3N-Y|2nI+KRlUIog!T*qguTLdQE%8G@2K&Q$`C5jXlGSzm0*$uOZD%bnmGr>Y5$
z+XQ&`3z#Vn1Haj^%fG!bNtO1*4?VDGHW{fT`%F@CfQ~Dq*dIopVbDQ6qZ>!^HON}7
z@%URkH3{;1PrkK}uAF$%vN1z2BMTuY=H5T(o{R{TE38b*F3Y{_z>fS$(($ooimLMw
z*DL3iQ>PHY1{%R)MZ6i((<|+8GXcFtZS+M;ML+%W=#wz?yK!TYT_)AaxZ4%J$&Tkc
zJ`%rUYNsE}8)tA1+K6KBm%q|uD`1H^iaOlJEPgS|E*EE2yR*_b5J^3^asG9Dd5VIn
zf1bl|%DU<tZ7shawb2fqS<vZAM#k;ZwCi^s7LF+Tr278|+3LJqh+(ARAFaU9xLOJ>
z=P2*Sgs<)nztI<Zt!5B){#~u@Ps_<$z5-NR#R{&HKPAlYZQ`1WHPYV;j12708s^X+
zSLXl-mdqtGxXn#paGrB{)_XavMnw4N`0GJ-tatZznvZbOs;}YB5!YF-6+=5+cEMY=
zF3(~+tYpQQl<$FXSxi$`#1?XZVXp_)4)aIEp;4Bfo_|*9`TZJ49|YKmm_ZWTziCc4
z6i%MQb?<Ac&mpTL%v3$p9(#4<H7;sR;h6w#J^(=`)mdv+2sOPz*>d;?XWpPL|G&_C
zi!xC6`9=RQ=GO^Ys*(_zPLOq#e97_clXHEq4R_sTaPkVE{ahSx3xZK^R%oy_h~2Cs
z#<@!2F7=V=_wVG_pWbH!H@th`N0VRQf1oG01Y`pB55!l6@|SGb-kfiK?HQe?*6`mb
z)Hk!&m(wRvQfBDrx7ato#NPin>Ho*KLJ+@ue$y@g*WY$?{|hL6xCVey?Pd)Fqi9^{
zfDN%*A1bW|AX#|C^O?s=H?9zCz01b(<nN!`zy4J8D)jW0$PHintx)~1P|2S^yaCCQ
ze?(~g9u5AF_x4}kO35*Uf1kS>Hu-nH{{Pc8_-_YuJaE}5r3N|vKVSc%HFVoxuc`X{
zzkFQ(@^^BcfcqQghVJ#>cK!a_3;vHcNFNP7AEV5+zx4wA*C+OW1-D)SjH~t^$fAFF
zZGVeKzx$({Trz-A4SFNi{%I~P!f-*p0bz4Lgaqj6m-49nB{`dse%OBHbwX-(3Gg@;
zsg~(Bt$^{-(|!+9Ezep3SHti!%X_&#W=#4&-s69K`x(Ni*2fm=P5;_5?#X-w1AQ(7
zdeNz%3YskO=h${I2tZA@1_*sm)2y=Ub__0NhDrgTyA7CLuVn$Py8okf6}A8uxFR!w
zACL++INlx_tOEh2aR!>s2Vilz2u%f;*!#t!zOm123X>8fI0j-gKt1jeNS^NJYX1$d
z3g<4v+>ZejI0I-!&XR1PgS0|qD6!=6=1@Hl!6og1R)g8l&5wSPC7C4h8IUa|JBx!R
zm3_iR4T%n3fMtY&7NP84@;@OKpnc-EfBV`Jbl$8g`3F!A4+Wv@8R!?@T?8o6BB(6J
z0cF|-o?;_KN*-DX@bUxvpZfz$wPdLlFdA^DFQeToiAv;9##HNPQc4v7GeW1er{FkO
z>%>s<`-Rb(0pQ1zp+&Zoe?7kePYE~4ll%*ysogUm*^mh0Qgtc8H7no)NKsPLs}MkI
zg+K##KueCnMsvv(lcYtY$W@M5@~H{|jUL_x(26#J!Ue1KMjZO`bC(I*KeU4blyET!
z`kMhcO|I(w5Exv)*>Q1u{ZGjV$~3NuoBDByHQgFc<+Z;W_8s5}#NaaeMZO&T0)*X?
z_lIc8FmC|I8cErs9|Nn*xpQ;iRCE~x3px)nagXvd0O7p?do952Z~E27z=xZW1rYWh
zG0Whfc~}O3!DM9J>?wo9W6y%-@F!JZJY)U1nrImXFQXm2A0SM(Fh%2_fz=8oFh4l#
zZv}2-n3_fvHo%v3JoTceqV)bJ0IZxpIRwH_FvR#i_n5R<8+tqs0wbRD=ZUxkT^1%F
zcn!c`bsPGM8K@%VSz!2-1o4>=4@3;65BfK<|C#U%8oX5Vz;CYhk=j26q-O2|EKFe~
z89X&*K!!|dqT+TNTTmg3T&)Kq(&12)+n+(!kX6D%ec}g{#0urQY))8Z<Dqwcym@kv
z;qw?I9=ozv`JbIybj&q30DU7sa?%r6C|aSMAu^o!w?;t6D1#70Fw^_>tm_{laSJ!L
ziCUS+E;NIyRtCl$1%nS&+`q7O16<;!>9&!%cYu)X%vSL#5Yy}&N9>p>eF2@LOj{ns
zu^g4(EvSpd(5g9_rT*qGUqQXtIm6jkok#$B#k}&lJNpUbj~q{uS$TFzPM9ZF8E;M6
zmDagEL%_|SK@qsu_%GO84$nz;0mj&1AUUv3frq$i)v#7YaQQ~D=cU(k$Wzcf`&vPG
zoUS?#8kC#hh2!A{t!t28frzn$vbHMbTCzTCh!uVT4JK^`u5g*yYvKwkqC>4<^4rxo
z2{q}cuUF&jAq+7cP4Dav_aR-1Z{3BGTl`ANH4BoV4z`b50d@ldWyoGum^w|4uByz(
zL%1tX#{Hkr=x|cAKZUab@XWn-0!UDOXQ!Ajh%k5xVV=hTkzrr^m`1heIB(8EKTjB3
z3sPt5xQOV-La#i6*k+8tmn9=~B{>t<o%K7S7jXsi46@k5z8>FgwTo@7qvnlr?gmoo
z2Kg}rg@fb<3ZqA0<oL7jt)_~|<IsHPpbM9ObT;Xef3pB6gF6kE0lby6nS!)$k;0FJ
z(U;z(lgTEfbzuGQm_0cDL05y<W()oiT@{dgtg4UCkbXu?!m)shnQ`<#$~YUu`zaX6
zO3ttvFG%*a!e#!fl&SWQ=;5CL5DHOO5x+^wPVwORv>Ol$o6cG;tZ*V%fJL8JZ1e3~
zR~IO`2;_h~uc1Bn6M)g(S%tF#%@(Ppg=;G2Hdzi~i^vckH>lB=3E9%uU~83&Q@;nW
zK4(IEfCP@)$zPiTeezp?fl`OH%cYq-rO}TtrI#TI>6Al}H0bnGAq3#t*W<&e+7Qg&
zFam>b$7Qdik9LDSV~xxha*~gFl+-rTxN8(YzU07|P8o>HI}t8067{3h|D*?V3(>l+
z0Vr=hm5{I-aou3@Hsz?kYhN5CR~&m7qRUE-E6kff<O*7VX@Ybj_DbWFBf`qoPCDyu
zmef>%=6+9h5K0{dH<Q^nJC$~tnM`vH3w!@OX-yEW!c^aKp(SL|3Cc7Nr4-o~(>v<F
z&DsiMe>l@$Yup5y?r+*ar&Ue#Il%+@b$UMICj;NgbO5J*$3cV3oE80|DrFO2Y)G9M
zsY?0+tTwO8fL9@T=(B5c{3(_bX3nMd8r~a2&EIMv*?x(PSY7(IFIXGfKpgARz&_Z1
z#=g73hc!7r?;fZ}6K;zX#E_Mn=q}wNXUBx2I~ZgbCoLB4;3^@U;QT|52q$iiyy>;E
z((HjugiHVF%!NSu=IywH#aC#8oYlK-n(a7i=3m${d7D)V2_H{XlcqhZrm(=~Z_>?P
zGqOn1Yz5;Q)v=MK-)N+s-JQSlI#W_s^`C;_ze6bhd~NLy?LOJ%{eRa!?qrOGWbsiq
zfSX}!*r8$W`Yc%pRUcS^@_Dt7NXX2FB8?FgxY1CCVeS*nJ24J=4po|-K=HZD+-fAa
zY+aJM)=2*#bhxYU%IV~rzRZGCe)fVYd@A%0dJvZHS4z4T4Ima{2W7dbiC{?`I33wi
zEb4wIx7Q$<*Hdg!FEh#sinaD}lMc;KBna)~>7@#bMVNCZ7{$58F@(E~r}d<go`xC9
zVfDHJ=}fBy{(q_NIgkR@J9Yv%_#KUvQ%vM(XkD<-IHD6#XaLHNp4U53g8iF26?s2o
zWqZ2KEbc5TYT``ABJLv!#lAmNCpt?FB!b$Bu^;3ncZye2IL^Q(`3;9|))BMV*_(1q
ze0P%h`opICw+7E|fHu@7$m6})JTahBXov28DTCWtP2kUd*=-<8T%FP)drI%vki~Wd
zNGfX8+(#jjTr*iLGTb&#Gbr&ye2BS!Z&JdtfJ*y1Mai$L7C<ch1jU~xcmhfiG4BD$
zlf5+NDbgbfUm$ydsd52AzZWi*S^@l1oPuk*%}c^qu^*_Ji)&DY06Y#OMmNBA1`@gB
zsHTt>P-p`H{Cpyn`|_Ff*yAld<CVNk%sC9*wgoaZJWx^kDyns0#De%N?9+H`$~Tlx
z-ZUcrIZ&z^iHE)_gO!iF@X7gwqvfJf{d{UM&oa5W?hTYeKQN6<heRD$?_|1V1IYZ*
zYa4n2QW?egE6?`2{9^!-P<oCqc+m!B(HRCJT6GxJ3{~IB8LkQvhsnyCqjcUry>ZvE
zVe4sWHV9`I;~*o8X?H7-+rL}L=+0BXMdKJ|q5lckPQD3Hy3qK_Jj>hTibA5XObY^r
zr17XFs35^#kF9>zf8*HMH;NY_H-ClY7P<oC0z_gTBC9->b%s2Q0w$}x@1$!g4lYan
zAhM2$5>7&%$$`A{{WQmaGMDj~p8;-8c5W4d>;cd09H~A6KX&$aH2^XC0n-b;M+D+g
z?C>hs=Y<K1kvmqFD^jwibRRFQ^S+L}VYtJ)MnlocFr@B%N-Cd-Dnenx9Il{CVd6m*
zV}qRn119DpV06G}?gKyYiUR6Ft?;=Prwd-D7tXSyw!8OpuPJ!+ThY2wt4?{oWS*zs
z7R+v^D}Xsiq3k=)2z2C!;OMIzoAy=>{P?gW?`%m>m<DCX3Y%XX)CN_{(hx@$FTxQR
zDMHJKXoy_LL7hRlXmsceD(;eTd-V0~slI-}gUqyxq`c~G-N$?2M*7l0Ht@9=o3i)q
z7E%0tK#ey0xD8{QCav^Bnr+?7qr=`Dx6fTG+M(qfk)E?glPAn(EadzRKmGd|N9u?I
zoKH@p+%NnpNb)n@f@*Fdp#}7-XktW{Z<OM{SOWO;-bS%w31a{gf=PBO;V_Ijh5!KI
zg^CvlTn}V4QtlFqykmZ}WlC%E@&?Z3j`){}ojeCaQKD*@`kbndJkQ1G1Xu?M1qXNZ
zcp<95loaf<<=gZfY^p8~!OUK&1ZaqQ3Ts#sp0>_?eBR1<Dljq)%?FL}0L}+t6l*6B
z(Ero3-=Olp-iGsjM@HLQ@B&$j5r@L&Dm%e70tSLdSWI#twanR|Wq<gM(6Dp(%hxOV
zK^8Zu_wKT$7X6r%WH|%;{2){O{kWO9*ti!c`tY1GkRDQ}q^l);$KR%CGxWK1aNeI#
z`!xA(LV3$>d~*DXRWEZFyc281)Xj=MZ~6{m)?Hxe3rkmvFw$JM>Q<DewXk*)*UP>E
z`NCj5@_yC^+Np1%h_YGODGhuY&1l)9%b$GR1O163-_#^6`CAYk(=1s#9amKg-M<HM
z3Y2y6S4L1_VO3raUZ!t$@B?x_U9Lwa$J8%~mOThM&8Td2d73@sOn6rJT6VtoT?jU3
zhej#>K^m<mjL!;QPj)$4C~GZqh$=qnC<6O(>Su4#ly;|M@m*AgaUh{uWyYzz5mYzy
z2h?TPMJ$|)qEQs1mB0Bs+g1H;ShwPl2y5SgSWjvqu+@2{^pLhMJ=#O!GMmnW#P9UN
zO_dl)U;%r4%2vWJ!#*S$f0l3~a>t-x_~_WA*ZC|iCf(<fNxV;n<n3{=WuCa922i15
zJ&Vjuf(+Y7fs$7jFX72wmVQo<fO?ILKeBC%0iA*m?`xo3nz+|B`Fw`8v86!DLHF^5
z{qPB#)<Mt~V}#WA@817jgEPb`W=j@zV1PV>Ds0T+a1~bX=-+SSzmM(1azNJcopO8j
zOV)`Z!$By{Sa7zU3p|e(@YGuJZYc3(#LFVBxKCjc9v`If>)v%V0nq%?IC#?aZ+9}N
z4Buv<A&P%SN=qskLByL%ruG9v<9?rQ@J-m0P<uHnjoK%sEXHFd2Ga0Q0@KG(6XZ`2
zZS)!FCvZYXKOQzxJff`&Td`CVo$%oR(y%|mZ^-G6+B2aU6E3VtO-j|zUH5Hb#3vOg
zzoDiwN#*=B-+*0>d;yvM;&P&rfc#|1EL#G>ES?OlZ^uELOnz*={41#1t?vz86b9zh
zl;v5vJPX;VxpdMn(9ly?IqMLpAT-6_g%iWsF~frxg-PcZC9MpmbSG5+cddWcb(0Z?
z3CZ-5u@zqHl+9>ZWhK3w!%LHk;wgjtK?dHr%|*z7?wTTyLK}4#9_5@Z5VLr`of^%5
z@>WcvKuDlE(Kyg4?EVp$f|SyCK8g^I1(5($2xFzeQI4utUv>clrpBCb!&P{@4`hbw
zX<6Ky0aQf)wii`ejB_ycyT&N}r}iDb_da*Zhglvh=0x|W30=8|n(mY!cpGJsCmJlj
z&N;6?e_w6YB^q1TNaQN!N)NxjErQoYQ9yAZE?zXba@-q&=s*(539GpB2+RXEZUg7q
zAUCJ{13oj!^;iJwytlXy0%KPBwOgpzX<kGe;l4zia3*x59mTJP8Fp7jgr(C3X~>!P
z*bWM=_?~|7TW9dQ#TZR7=mg4Y2_AKG5FF8tCv#(#Aw+nq_^F=Njb1&($8R-5gly06
zKT14kd?yY`nx4~6*>e|VKzklk)8p?yAMfpCZYd8-`>l+q0e*Gs4dKBvt07?N)ah9|
zAtA0cf(|yqazJmuos|E(ZTzpYQO@%xCnVnguoMBr3U3i;>mem<rz8fJzB*`XU}N@h
z1~{W)SgVmfga(tQ6^dAt)8G)ltWTPet`XK)MiWaQi_EY!97EA3;|QhaQ7Pwy@8LAQ
z7vZi~$5QnU!tcKalzYLZH`k;R)<|+NINo|z7K@U%*wra)vnJa>BVV!WWuKT1rZy0d
z-zrMg|LUF|HD;T%lZLz&!TKCQA%m8`#~ou5m6}hgOdJ7ADtzyuWe;3=e)kkZX_PN<
zd<u1aOQZETrb~t)gIaT&UAZZWRTO#n9+6%F5??47e5W$7_p3`Lj27_|U!dg+EHw3k
z*%qv2v?dNmo3Iv28ihqAT%0571Kx=_kE0&EtFR{55GuH?n^$WcsQ7+4{bn_{Opf5K
znbO%hE^&Jyk(${(*;}boS8h$1Q<SzN?iytQ7cN@f4*Xv5PiD0yM4QSOT+_Pr262lB
zGVIGy&j*KAWE0<^Auvi)=*>h5ZKj~|)DNqm@V>#|5tzf*dY~jUSr3^C$#d&^&%qq5
z=ZO+K(c%4r=ub(7+`8J6<8ydf@vLQ3F(&J0z{UG!xsv%4UE-_9QR$tOG@Ps}$n`Xh
z+azX;(;GH-vRQA-Qtah+84S*~S<f03f2T<(3Hn~$e{<&zVaXsmMN5pq;Z_%KI(4?B
zLFm~-oPwXxge~qhBd2d${q@q`UIWADRc~A7Pt6%frpRY&4p>we0hGyj#J7T0nK%Zv
zCiJFo-9j;UZQ#SI<G?6~;KAq8qM?o0(U>uh--!#P0o7!TlOg|<1ycCepL}>)oI+_C
zF-mze+l*@$9ck(n@OG%Pj?ZH7d?Yb<i5+58F=<#2pF9Xt5)CtWi8g*KQd`X;tjjUu
zHaTPkEFOf>eCir;eWc$xlFGS{AM8|ElT{y!Zu_P=R3iU~gpeJA>t19rC#0uf9_#C{
zIUNmZy)d`rr)AczZ%XWn@}Aa)*qf_%qA{5dol82|JMZhO_`o8Ezpr@$R;K%2+qi-C
zi;!8si+(~oSlpR!!fLj5(XShQ>aNI-ksamNB2yvt=c+e-k)Ki0{J`u@E}_4EE>hOj
z#0zpSCf0Hg`m<$uq7@V9v`ch2#uA>Td@iMNod=EIMOP*lH%u#C@Nfv-)R$HUhq5cZ
z(AB>A%GzK8Bn5@6Sgy|*C42|!oa4qf-Fqp)lI9ALk7>BmZMj2}*4i2##&czg|0qO0
z<o;y1lHxs|4jr&U9?UnQTXo8;i+C_FhmFn^QdgRRk#^szJ-wYSYnv(Nib~{MB#fqZ
zc3pnIlAxLLr%4-_JV$>E3<oYeGQTCPxNjD<6?zTEU$<IgwWp8l(I>fd(D(9xLFZ0!
zpAC(jmSX?Dwr)rqXj3oCf0Wgq?1r6?sTWb(ejrpRb=aj~0j690aP!LxDyacic8v}5
zyh|W!^!4O{T=doN?7s5SN|TfJOd}BCzsRn|eO28<-DU!}!Wc*Ky+(GPtgvdoy3ORp
z{UPUQF6?z*PIf_g5!9r?P=FR^UWAYhTPi69YAc0%M`T5QMXw;PejBCOk-nf~b7ooW
zXs2K_l$=1c4F(2X&Px*F7rKu>b_Wz07bw=muJex#Wrb`ZXi;^UDn0h(wtQrKsbTZ%
zSUVdzy><lVaI(8XO?>zDtq1QqNPf;j9cjXR%_7^;aaMP}lPr~`4DkaXJZ(4#I&;3M
zx9I`AzlLp~Ghg_CI<PfrD)bn#V>ZC77_!u#sPp>V;+*bAZiC3I*_7c5o9DO8`mxrp
zvf7_N%NkL8mZ&g>Mqfy5n7oPM9TOG4$es&vN5-m`gTnp#pYFYVdhk@E8ceYWf92#*
zl9`t|4Q$`7K<c5S<}BHDR`q=jrKVrX3;AAWCF2WPWfu$gG*!iSdv6{;WNZ`|b5K?4
znIt(s^z%o9W$~D+19FW{eFc@_Fz6rDh0S_h%e`<t^L9b<L5VYvF1)-g)Fibzb~2HM
zdv%MT`p(B@OdG-a%9kCnt1UYhn^m|9Z9UvGQsm3~z0%s!B5k?zisz46u@}^i3V&ZR
zVR=E1YOP1R>eu=9tq$`OMoZPl$KR`vpKS(-700M`9LY8Vk<V?*Gbwkwy(hYtOWM!`
zc#2kZ^T13=RbSygpAsL%Rc}x>T*&jrqK0i|08R506v9@wl9Kf(0@|xLt;jM1F^`A=
zJ@lXm9)c9P1hdJ%UoEp=8Rj_y{HZ(!K7KC*3H-4WaE`jDL!kNINr5K5xtZDUw`r&=
zQ^sw@!MMqdPsuZW(8oZo*)nV87C1+BK7~jpIw1r^V>9Z!dk6azsWr?zGq*zXsg+@^
zoB(YC&?v&Nw50ju;dRUCHDKm;jt0{vLI&=@wm5<IcD@B3UWc_C6b?eZQ=mRhtc!8B
zC=lxzrSg?Vn5{BBxjuoVaeayLepFxwrl}%dahKdpe-8A}K+LDLidF_0`T<pHkZ_lh
z!xd<mYK;yL#849ViV!qU`=9w{1WU%0um&eFF2HUxBs3wP?D+Jm;e>)T`6-KCD4d|?
z(QfH(|IM(iUMmobp`F$nIUdo_?i;p0HbGQDHw3Ssh1->cSQQ|>E*;BC?a}fXOBMOr
z`G2J0qHGwr2oYK{1#uw<zU_|b$6~6UPfF~Q0ib!A@o?*t4RnZftB0+8Y-om3`58Z%
z;+Om*9sTPU)*RqAP<~&~{i~amV@6u0@sg)|cj+VBCA!;XWvZPh&^!uN3;-eDb~6&c
zZiFBLb&xF?)pH5aREXIm_XSWc`==^S0{>73KTjwr*MQMZ`3hiN?ZgtO$!dSVb+yS)
z7tIyJmxiJ-W0#IApS`^HB0NkRVIZR<EoK)Rc*TT0M7$n0m>xm22AF>m{X%6mtwBE<
zgGTau09lS<JSN+UJGv(Aez^2%K)Ce1OJ%!<95{v<!{0I4GR2Y|l&b|xDhAMny^NUB
zvtyb~$IR?~&fvEqpNU;`M>rK}dHQ{*3Bo5?2iI@x>3-NTcG9-!pfO|>-!WKjo(@q1
zjPKYto1w$~+lOGkx!5ZXEWaf^c|eqLhXO{<&yE8b#jvEOyL#l#7^#H^P|!%Ah}6m#
zYPhnaeE@_p+25SOge6+A_4+b(8&vCf&#sQ~H}&q4CcU0yxG+bH#MN%UU@E9~&Hw)W
z9x)YhU|Qch>Qy6HDB9f+yA|gcPe4Xw$KFSNdJ5vIpMp4zL1tC0$RBBr6!8JEBJ<Q|
zccY=6Q1gT7-^UK6E9ab4URb&GPk+!X{VlKvlX7DWflX*|XqfV}l0elt9tjKIn-xh{
zx<8Ikq@o5hLuXnYs$Lh@5GTTN8^vPjj0$|Kuq|+E*`t#yYe>KMpScTz$$S3J0>C&5
z&E(APfjsz*m0Hjz;#mJUdKu<BxFRH9GICdhcg&c5yKw#DE2!xMa&CH##F{ISl##3i
z-?dYGYp9L~)ee$M>^68|H#yU9tA4cMN>8pCfWM?%cT^ofclkL3IM!iT_abM$38sr<
z5I4m-0hM>g@!3q<z+F>F<VO0Y=G_whYN&cF;|A5^;G*~Ij4j(%qaw|lcJ_M9=8~`E
zZ&MC<?F`CkHx6)K)A@EG(jKO~YFB>CEZD2pycczcd)WkgC5L-Y@PJ>}cF>w^O4q)~
z0x3AJl-Qh$?2JmD#4fXkf`qFEZj9r|$&l>2-u$5K{11(>)P12Dc3@q8_Qeq<UOmgs
z9r$Xyu=cKHki18KFK%7$<RlDqpiCV}PTDH(P$r7y%CAc>wA75MZRnfDFU*2-0=3W1
zA{n<rSvF6f>qu7lga`jU<(w!OVgFeP$eoE7akqylR;~J)1sW*#l@Wvv!W#}NM{z$}
z|FBZ-z+^&h^xL2VgrwO1SXiM#aoi}LtWa<7Tg5B(Q<<owTCwPp1eiH%?XN*f0*w7y
zKl?V`PIWXm4jK8W_f4lU0J6K$x}e)mLiWyY0=C1X9$hD`UkO>M|8=|Na8l@i6)P7L
z>PAyA>8{K1$(k&37PM^CYcr6yBVg|fnB;T6h<a}I^w`Z&L{e*wH(?WiQ$u@ZSNJX{
zcO*{XM*y)_er3O3$=w=)DTs!NcVUPp)e0C@TI_)S!7+saD+tuE#$xl!(+L8^Ob0LL
zB2gT39~KqzxD|crc)^U7LQP-Lb{s_CYlnRA%YHUypg<61J)4*aZnlheU{Nurh;&Mm
zo1KPwRMt!E9TA&y=Tk$eYhWcp_r!OHLt3*j6uZ7$YI@k6l8~`@E#1Pe)`Ne1)IPIs
zf8KQx;ubcnDc@Q#{E8>C?G0>(MAZwlYa?LgHy?uO9Pvl!T&jIt(M@Yr8BXC5bR|;p
zY!Vm>w)>zIAV4>VSd;B_Pu^vCAapYYOS(WxX~!Pjdox*9Tvy{`EfAg#sLN5UWD`}R
zu6mPRRz3SO3<F;6>*HMV?Q=gB?7o9^hA0s1l=D-3TOt1OLz^74n_f<Ho$>u&bApI%
zokej?{F=9TIUiMz^VgthfMq3ON1|XSgYQvyMJdo~QyF@yM$}2Axsjew(i4>JOh+dF
zzp!Q;Knm1Oi?04vFG#&&ehY#R+g14AUvUO2Uu>rdaxj>S-UNyke+-@{adO7NmlDA0
z0vq!IPiY-v&|-3m5Ge^H-qnKPlA6(3;4Jz)D9jCz`Bp4M4CJ!aqz^7{^$gM0U$R=l
zt<1>9WJV>Y4;jgP97X5zueK>Zf|s}Cp~M(%=_-hM(ed_SF3*S@7!zip0e%9Cy4fd6
z4NI{*u`ALZWo}Uzcl%uN1DOmEBkoP*e7L(p70&Z4$vF-uj3D?!@D-WN2^reDYGBRl
zxQ^GKD4yk8Cp${91a^IrsD{Umv8`S<inDLo5@taI;eIub<QFx)uw?vOb1$G|g-@B)
zTk^s1!@$Kjlwu#(&Uw|9unIi3ych-5RnqEOVe{rJai7E5E?HN5Hzg+nzHXP|cL}>u
z0YKw!8BPXb{imZv&|l^*>G>i!;UOeit!@yd4Yy>9$g7L02j#(uCvfh(!-MI(T@XDi
z8S>3qwq0Qxlg&ijMJ@)NhnrQG#bUO1eNu0vBgsywDSE329zkIxh^X1C6(OBI$wN2y
z);^8p=2_H~Pc1%|&l5mr@>|9qeDVGHd5GOQ5+fc>5RRxdXPEbYol>pMfoMf!eQQGU
zH3Mpjb8XUORFDCFpBs#ZJylX3lanM5h-yMOp?Sk`34+ifi9QmasR;J?7|D%vQzS%;
znw;*z-}1k5y$0{1T$A%jm87}V>x)qDg)Mgt)H!&any;8ns1?Ya=0CBT1uFHrFQj3A
ze^j*Z3$;~=%cQ~{Dw?-OBnj7^n}CIr$&1}lStzo&cb7GL1l|cng84PK2%l|$s&>=L
z;dFl|^C*fS^jP{m7|N-7sN5pfy&ZCDd96evA}hBHdu71FtjJnPt1xl*SoUk|#R!&m
z+YV5d>z(#qpETDSNL#jO`c~Hm8iYFF$f$dKPyT56?|Ars2Xz?RCC{~-zAF6t)D$K~
z#b_L;?g`=z0vn$Z@5zX!mXQMFcG8&aNnm3ykHHI!*(E4Oq%5VqMGfbI;!T>>>mFr0
z!7YaS$$OQkdoAx0jnM|FpeYiF6eMD8$4h)3NXXq{=z<@41E!=#sK{NVu~fS%lk5KH
zIBd{>CuEk3xwI!B0i|?-aYvHP;o@I}6XycMg<I4-F8Iu>zIx@GY|1m<V`XJ3@O-;$
zILwcG)!a1LC|jPp1yNoG9I1Gbd=yH_y@r9jdWbwHkN|<ZoCbzxHV`M)<2ZcHZp<fv
znXspR9iP>q&<`XGJ$`|-p}nL$Ou$NZL9tn1m&7Qtf&x054~fE*WC(ULx+K2f@022a
zf9$dz;QGv5Q;OPIgGSPnQO#q%DPOn*VeO##Bg;3)m|9E$<qtP!VH1(Je>ndDz|Ebz
zk;Zcif(2}AL(fzurF|;67R}npPBMt-xJ+M^s;uk_1qh}NQ!2e9WUUoFuak<pLXFRQ
z=K=-l@txuybd~N^?h|l_194OGm~777Ok_S0OuSqpQ(i;@26oKLKc9pbebI`PqEZsQ
z7#%}MA#hB`x|7`BBSWS*n!yO;(iDtT=H0oz9gLCKtv@gq6JXO}K^S#3nxIu?>S$>`
z7065)+uUda;1dnkZ0dj}tNKA>msk~e`+&(0>bj}9v8`dGCjs!Lbz>Y&zH`H+V{q34
z@xI6A@oNg2E$<Np#)GIJLjNzLL+xiZZ^caE@dIAH7o!~Jh1RVSPXgt$BwnV1txb0v
zGz@-d?o;Pm^(#^GQ)M}~(*%@vi_6pAW-yz;?KW16fhhm9bjG<cFMvP1<mi^bzeG){
zG|XM{1{@f11`1;gQ$v4#a+J*0=Du7h>0|u1VL&p`>-+$5<o^72G^Wmh-)d+;ES5cK
z{&6+s=j>j8)}q7h9z@%RH;)`8NzBBOfUk)$b?&>n;Kraz?-`8pmJW;jbje^HYbNKB
zjf)oF*2e`Q3Zj#%B9FM=1KM`!iLZA1rGb%E!Lnr0^hS~~2>m&{Z82rFu)SKh8cX!}
zOBJ@p1k{XMR?Z$ix-X49M_rn=IQBYd0<7@2DF*cXincPIc)i2D=Rdls&&3gSxm|is
z<Ph=}CocFlyflkwH=j&C6M8Y$5sP8qJ*H2bmOMHEl*DClSXEr0<+|zRq2AYX?D1VB
z?REI{gI=$RX@Q&CFX?|*>*mlxNtBmGAEf*h+V;Vh0Ft&_)XI|#_;};F)zl{<ZSrog
zNU;Rk&xG8nP~d3r{EiVyfbk~AM|uuMqm1Ki0+D?`Qob}4S1-yel~oMLu6(^Nln{<&
z-m$xzB@-cys0<V4`SKtS^TbUcLq2@zJD<#zAN1J;j0{_Wz96XMrzWi6hCZMrcog$|
zNoV4^e&DsVp$Y3r-(K{deBa~Fg3p35)Mz6XVKpLs7w38VWtC~Xul@xEhu-zrua;yN
z&IgjY;HxpCan+t61Bk#W@o`!HZSFZjb9qFay>x`kIO;o;8<y-{xl=)}Y7AuS>n5CQ
zV~cF9mvD$GY^nk^g1Le^f#PxkSe@rJ7zTHdz@~s`Wc4MO1E<a$OG;my;78BSQPmBB
zR#@weS)FQds{61JgpKi&<+^*=32Wy>Q}~W9rhffWwW!NeXgQ#2GXaO!?`YZhFt5~g
z#GA(*b0mQ!c@WvB@M(qq)5IZZ;ltXQ3(YF=+%zNv*#!5e7&@6@ltUy1ZrN5VMeb=L
z;Z0Ik6+)at!;+G8CM*Y~+Cb-T4$Yh3qAJf7S4jG#mZ$xatBC7m7vZ5R-#($QR_CL?
zhzRPDRj~Y|+UD8jCFdM#^&&Gcu~q&2n0WT~Tt5_hsfV;lTd^>jyFI_08zb6{SsEgy
zcRR=QypE#lA|d(fopYu)e^Q)rX;%JZOSvy(1-In4s?mx&IQ+R`R82d@U1BrgntmY8
znY*PV%W7q)6k^o@L|?;%W7<qRwZ;h*7K1#stkNhs#gXTq^_|nUdN(e#Q+!KVAaA}d
zDr3^XoG7U6lnoQV0|pON84r5c+42nVMJDSx8?{pm7?|teq9UbMU3i-%_uDSAd{hWI
zq0t*wEx8iirEnw|yB&I&@9Qmt61;-=qx5RV=giBXMq}Z2u{8b?^cnCF-(9W6-Iu6;
zr`|3SLugp2;poj@_E=h+FjN@NMBrR<>`;5;1H+*ofw%PHFr@uI@XDhWZ5pQ7L}_@<
z$j&^|dabxIgQ99!I&<pD0OPRLR>{{SOsqz=?Zv~L5(i--o83hE`s%`0O{+n6_1rac
z{JCqBe_Wh=a;!akAtv<Ju`N5&q4%0xc1wasTOi7F=Nz0UP+c{E>d5k<!^^dTx9qgh
zVSe-J=?2!4TtCcs{ja^|MCNxp-gS9#g5tr7?BcsH`UdJ~%S-i*jf`#_@W0<e{=Dp9
z-fOA2XK3G^dP7uh@9~5MOU|BYtJmWBQ%#Pl=gnE=nUV|%Fd3fJI>U9VcK)8?AR5f1
zZ&qn#)dn`hSvuSD;=*@{ZWxPMyF>b|icb|i!{Ns1>#lRm?P6!O1C>7??$xfkh`U<u
zdCo@<UzqH>y^wxea!vZaLIShveT>z)ZPr#gSJI9FOH&Ukc6|o)jc-rCzss%*KN7jK
z9y#B;x@SA4)Oa9aacOR$8|~cYiS}3P8p!NRwNv{%F(5<*t73m+<dK#ijwsF52XP`M
zn_pWd`pn$Zrq6|WK;ch{R*2(1s&)ZaC2mZWm56)Z(Q}{hqiC=2=^Mya|Gno?$QT>w
z_OR`6a+vBk@G6?*M?_W&y$iVEkJ#rVzg|S2{kh$d7T2ns&tGitYUpaY?yr=@e}%>a
z)@4Bh^wEm5LFvC>@KGi>2sRcxfk86RpSzcW>Tsec%vuM$?$2k9c~y-dXSQ&?D7uAt
zm;Cd7ByZE^q91}$^Y<7MwIH}3qiOm@|1Dp`O%Eq8;`;8bp=n@v6s5z5i(Io{e-)l3
zCnS@!j2k9Y3)*AeM(ZGw^)m8jksqtWFOv3O-`Y8{PH;X+VMG7B>r44M7|C&8#YWuv
zLrw&HPKI}@D8~`lb;LuqK`&CTWAO6|sTw)dNPeMRYWM=gA%0$&b^mQ5HD5BYqSN-!
z=Y0GZ^hZjYO$xUWRPy40QqC<hDtxW-+XkN$?qm{R@;u%bb^mTB``6~+Uq8$d0U!0v
z&|KT!Uy*-R_n;pJcw7gFlEGG?%kLJL|M8!nBeNz35@Fv=zV5%*(Z6}=zutfW_&n{j
z8w&ORmrv?{`u*v5!OiV@yAbyO`v$ZjW{$Fy<Da<iMIKyWJB{Z9{`{vP$`ZY<_)wPK
zJzF2&0|iVg^c8xdzPUoLMSZE3asH<l_0P}i`N7BYxpv!q=^w$Zj|uXRl!2Qr6aapL
zbO5At*PM(JT^4l?!iVPS8n%I?>5s<BqGBq>z~~PuP97zYKy3o%VfEfBQ1?9pEO#qF
z)Mv+r3a)0^*;gK!*+&;mSwwG@0xa7GN;PN&S#WfKa=F<)Fg<c_zxnKh_c4e(g+h5B
zc4Jf-%(fwJeE(zi>F_gAyMnyI6;Pi`wgO)|B_Q1dQcmqetb8`qNqVTw_JOCi4Y+72
z+UXxn?L9;<|MNNh>s?_T2X5#K=a!v*T@Gv30PtJ?ik9X8$GTt9(snP*{))N=vng9Q
zXWH68WTd(_Gi|{)U}gSui03}&*qlrO640I+G&l{`tXHJ}dtCM>@BBT-5GlQOTy#5l
z>FZ~p?QmKKJm1x7J_nZOvA{aZIt7l@Gf;pkg|d(;G{LA1?5C8#VNm`fv*W|9$!#Dw
z9ZW#u!1#_<D{H*xL1?0|zWa!#2@q9yNI<N-rwPz7E(3_HT(NEh0e_dt9CI5EITk1f
z+8*~d=og3qIeG>xmZjJUWIFiYd3r(EGU(jh_1OchDa{dCkBVprN52hPSq*2vZ#~te
zd*+{^H{dH_JDl$X^y)FUMnIGs>HxB@X#EIS7M?=c-3=fj1bpr1<W-@PEnxx$0E2hw
zH)6I;E0T3|3|Y$o%_hAMOcHkigAFA*(t&Cwd%Ce;nzk_p*k$>FLrSOtq_6TwxU2z|
z+xO2b=cY7({}6rpw0eqSD@gec-r39;6i^%t!YV|S02CNtMewr|z+NeZWB-fy9Urn0
z1-XZ%|79ck&`7vB3yJf@h?k~+CRlgs?I!>2Q@F<eDod2qKaO-IX?$74%IePi^x5en
z9Z(vJM~nmb{wr_KC-2MAqXC;WpsV^OKjd~f!0kh!M2eEWd^>;iTc%?&RTRbx*OrEY
z;2M<X1jUKtR9p}sjel7Ugr#&m2HmK|c2DV9X&vdEZ^vNLj8ZBH9&clR^?axhcesJ`
ztobI>8R<<|$cqmeKFC@Sg?TJd<Z%rhy#}_M>q}%Jz#(zs=n1Zh@8gF+KbQfsP_Ygk
zK#addnE*4u_6;ZG9iFOTHO*YlYyM!gH#C)PT<dk{8o|vZ1#ro^zZI&zX&*iWS8n`>
zE7S31nx#tskw*bz#%CZPEfo#CfyS!Hxp9^Bax)?s08BmuqDd}?(tGGUwCgyTT=YE!
zIEVMngLOMJNF-^e``HDe#TRD{2PKMqfUwyHW@XVpVRmmDrG#pMyw27DLT5~JqlJmU
zEu!B?b`v_EIzb_f;I3ou1q;k+Y$tl^==h=Z4~NpBMRyQUt#)V`fCnQ01M3v`Au@Ir
z-aI|3V;1c+pJsI-_}gO}OaHw36^ky%WZ(x9cnoolHyRPz$DjjuKHet{A84?f9#p&^
zZ!Z2`L&<Rfds9_(%HH1*&K}pL+-ME6m>rnb7$u6!T8Z2<6H}_`6ouFfv&PiJJVJI{
zLBhkRt7i?U<BLl7tZku?25KSe`vzlE*7S(;K0t9a7?SCqV>ppu?S@j+0M5#mQqlhI
zpuR^0+!-LsG4Y~3lxC&_I=F4XS+UT{*SY|rG&BV{(1%QIP{igI2*DETbVPM@ild%@
zgy(5$W$ik!&;&yPjC9_MFHAs7k`RKVjf|=yf%hRupHW7QsZm&;G%YOOeMp|I?gQ=^
z2xh1&yZY_{?kJGgZrKC*H{@ss#AVp#Ne2s{JNX|`QNl0v0|8v~J$7$3R<~VK^-afI
zJt(2Z@ifdih*S~bIH8x4i7KX5<>)EHhGb2y%}$b*awI{lzy`@-n8Z34W)M_{q0=~s
z+#;5A_#Q<cMR<vt+x;v-lrM?+PMl)VF+Wbcl)MPUsD}cB$(J#0QHh?6YQ_#1u2z~K
zUqP4jdmvrBEo7*O7;>o2@=m60B*ryoIRg(&--3|@8GD)?>FWdTKO(`6zj2Q6rcwF`
zL%|pC2LjEG;C}BvhHMNMQ}U!19^H^(`AN7v0#e1%sE>rz62B{0a+t}=&SpyfArn9>
zB9v7gvFF2-E~NIT2Pl#TrETT`^(%bm7=ahL)tzB_gajKVF^CAUCP&hVd%GE=e2NmS
zV5`Qy8XRJ>)UH~oz{ERPKd-j?CTjD&DuysIa~7=6)M>n2SFMCpS)pMG>i{%!sCQBP
zMxz`PnT)Pjy*+6%xQIp;M6f13bFdrWWxaw-^at%DiR#pg$_WMD3W{yC%1Tkm^|y3`
zODcSo--W}-cy!+tTHDC=R9iGVH9qfwyIN@1fHN;vHWBBOo08wc-g2yNhh~$w+2Iig
zeW6p{jAZro%6@PKz%7WLbQ4oeY?$vCBUUHbNEaGzZ<7g$172uTHn(@4c6&Lr==(5G
zcM2?uXpCKn=LGD%VEPm#w;G76GFHmgQLxLltGF{Ho5=GGxSOukQ57XYp@Toa`_T#8
z>qOQU(tnC_h&JrtZ9XP11yd)pd6lCc4zfhvrv{e2^-f=X4RG~_KNBc|{~Nrp<fmfu
zrEPNy)~e@o2{Yr%|0|~QZ+pN$H3{^^hb@BOm@$?r)Zc|z4%9**ve((KTofWdExgn*
z(-zu6$eNnNB%M#=Zk}=bWk~aCUDXsF4!9l&;dfE?e-g6-fOnKzcUSCtUecYyt(G6@
zc5=9DS6(8n#dz@nV#wM?fH@E*W5W$PEkYh9OQ4_1X%8E-WDhLrx1yKcw;MRkch{%H
zp=;yh?$*KXlntT}k&10}G%FK<SFBY>XMq{!DmSHrhkA<M!yO5$RevC~g#3IISOvDs
z4kYXWiwlVc2SB!bumq`tYMZciQ!=V=puE{ovzQ^SCNGaV`k>a)4kikV3=Npo3w;yS
zwR@D_7V)~|!O7qa#bDdW*)f?tMF&eBL-0`lANJldEXsCm8&*OFP#8c;Is~K<q`Q$4
zQDLM(N<d0FMoM}R0Vx5c6{JO^8M-B;r8}g17`}7vb+3E9-}|n0Ki~WO`TjirW*au(
znz^p?Jda~P_Q+IH*DN^r+n{si9{~`KoZ>pUM}OAf0SM(;uxmrIyIFWx*BAhvO7XEv
z7l`!0ArgN3-OeNXY?JOpFgw1N#If*cNi{*lb7$66$78Q11SF3N7lI)l#xLg1r1F{i
zddUpe9H_vE*$8kVada!S&&t|v+1<?j#+sI<RJ;F6m1*poqlsffcvQe}0Gl07QIJp+
zaQk_Z1kMKCgpdq2reI(X#nKpZG#+UqLob|7RiP^6&y&4l|D_fL0yME^nj+=&ZF3SN
z$xZCqZ#szJH|yRuc=mM%oNQwKnHQhG6yy=b86W}Kn_lC(3lXSIO!;-1spDR}1OiD2
z+oMMsYw8Vy#DMaujVU|>mAD9iUN0_hpM|^x-{NNqttxr1XLkA{O!Zg-av$)d+aK93
zimSp@BOZ{k)qWp8DvvI`#xln5uU7mEB>GS!TIsYSWq_BZEWwjDp*5a0!Z=n*hr)0H
z=@HR4_2W*>H-hj5c-A|M{;(X)03P?v*J$h?xZXE;cvu9C>C!6)m>*ge>{*IBg0jFr
zI=jGrTmHTdnG4pMo?3yr%mHwh_!(axrCyZwsEs0^up~MAo@L*&P09qolVV1BaV})~
zJvC6dr7>s=2gl8qt*2FWb!S;$76{UTDf%=1*t`7g0vU0@==1x<`SGzd(=Tfs+mVYy
zFywd*oHBxASya3m-hNI&?!P*<Y6&9=IB1yCOIW5~zW=+|p2?4Ut$>UCA-)MfiqbRh
z)e!d^J^Y@#2OpClc7jFRy^%=T!;#5BfE59HvskXew~Dc>kSeM<(wF3&hO&WJfyl>!
zII-NSHx=b{8f6~@kpCd$EW_gcP=tS->YdT(gY$-lk+Gp6sUh3GWY>PZe%I~fwY8!m
zZ!SH*!_hqH!%qP(wYH3^tqtKD@Al2+KF~2vSy2lm{cIUD(7d_2X)6=CE>KQHrIEj(
zGX5J(l0(oB*o*3pCIVP$#FXEOyM8vJ+@+$hLR5vaU+|tibTj&SVAH(^^mHC*!1&Ox
zg#t~h<~ECSRQdaV12z1MbJX~Z<jt&O*`UJ5fA+2HZ*bsYCeVABh)rH6g_prl>;W2N
z5XNOyw&W;Rr0!0Vn+#<TXdsN55Z@21_iZpqjyr4GaBta;;8*wq1Xv^#!LZs&3%~uC
za}SC`umD!-XL`IfF?wI@Z4gKrYD61&1(Wqc2$6%HB_}W*Qwq!tr5_1VJ{s3a#hzD?
zY$FbagspFOw3$+*1=1Z;`-3bFJc-O7xX*V)!AR36c#tw_ic-RzpYhfl4Ed^Yi-qA-
z^>@*8ldYuA65u))3@TZEOa<|izU$ZHQQ{U2_juEjU)%mG$W(Ue-S9pVCJK4M;QL{B
zx^Uo-6sV*BX}5{H0L{EKO7=YtSv|`;d0Aby<?ldEYFHRxBC6<|%_ZOUiPW(o8ZVGQ
z(H^3rdqPt1;02Ii<nL@b!Y6UxhyALe41ge+akM*0UIWTgBQAU@6iP5ag5>DBzJ%pW
zjn)~NwQtw+_R{dbdGRHceb0ZG9E*fcvB@2$DpF}5y-R0CG;b!s>G7eSqq<@3-ny`2
zRoD{}L#Ej;AZ~nGjx`!W7OQDo<qc9feWVm%zd(~j5Tu&&knaH{Q+5tf7D9l_mG}Pi
z^MWtHRB(~WZ3TSaD{fp!j&(MP=i?hZ>vDT^+Z?V7_hZCQBSam6tJu}1G`f{duhd!%
zKTWy_eH7icXK3VZ#QcEL>3#IWV#aml#n48g0XnxMjNW83>1i>%eCe)KBHAT}&L^eb
zo4%5z|Kj^}>lEI+Dvgy&v3IJ%cVm~%+s$BT`-3d-Z9f!?OnQQ&eu(!A(E0Ml_j`?>
zj#e-E>R;)TKDXD_KBI%lS)i%A;^{JILE}`5(nix=z2|rH4A=d-{hX8};U&m|sbfFL
zqtC6Qe!@#2W#?HM=^0;_Qma6xFoR(~I_<iR*IAAh<hEDACUhF6RWh}A=7`}{+*Q@Q
ze``v>uZamJXWQ}q^5^=v&H^5M{rYCoG&<YWCu7ibMmG8DLF9KN_V_o!A?9c7w_z~j
zLPGWg4NV*q@Tpgw@rd)2>j+55s?DKSqz6u+-)0b(VK+#U$;A}$xC%?;XSBmI>i5&O
zt2%IXAoD^k%=!xokVbO}!gGMYiV?)#qDU&?j^`G&l)|Xx4D~Hl5Z79w^zSR^zYe=5
zcTJvE(FQ}KDal$Xiy`uDn-~J($MJHlP$yhDtz@_StHgIHp_gy8c3vxx>9rH%M2Cf!
z&<2O$7D&3kFNxd=yHsOOTf`!0@zqymfj^^L+Y2D}W%LF5USb-DidcgVd3&`p7H^hb
z3<-6`8}xAEd7Jxdf@|SF`EWjpO1v8wf(o#$Nxm3Wl7w<Fi!u>GnxxMLVy+cFB!%zY
z^eyEKg@pLx>K3AJbHRfHTyWI~p-j*~M%vRo!Hm<W$i5CeCnZriujZ$N96Ha^Bz0P<
zi=bhW0URU+R+fP@g-mbr!U&f<4H4eI0TAdBnFi5wehk)(`co!j-+@TuMK@f_kcxV6
zHX6O!kFu6*RuZ*bx`z*p@DXC}8#&oxUg^~Ya-$dtF_zG@-XcfV{!0IGSs3)jMzX$6
zPP$xD<X2rByZAb0k)nhAxZ?cOaBR1G4$$izG!;@U0egY66x}}hw!Jl#OI#ZamNz1!
zDsJd&D5ysMcvluou8OnOOiN)I_Z9YAVKMrHYtoO@ouSuE%L{%0q)3aTnKM(#J>P&(
z9`b5H(Q|$GBC8-iB)1WhKY3P~k=SG{Uxqgh1R`hymDi7hh{QW4V>5lKlrLW4wt}Xj
zWQU@Q`0!5VbH)4vOK8|by}eBUT*y3nslRF)9ck3cat?5S+Pj886=XaLkP)eR9W1UK
z{k%eRH_;8sjDl>u<LJ2C7t-s#iqpddS%v&$^kRRDxdq%|GmRv7`K!nzQ*Kn`eV=%%
z_vwQQ3-T;j-x}dZ*wY9$|FB};C9d-#>xQJScaNJ}RK~l%N~d=4rnogc?SJMdJQoz&
zROqfmMy5zM{0xp@;|?dJ;NI~V3v>~RHJ{n`6&NrV5|Cw;X8W;zPR9%E3SdG45w7d!
zrL&RY?wR;^B{g>L*(qcDQ%DLsg$p$?kQoSj(-Uu;m?7WVJn7Gb>A4_IQKro0HYx$i
zWRyx|pOnf!2tlp(0`O@sL=1XjtVq6W)02HyEYGCU<R<SWg$YH%{TdSePRH7%mnAiK
z^xW0+Mr#nE`1hqyY{KD-dSoJs<(^}N<@lyq;ksvCA7IyA-?3ebu)=j^Bz_>RS-Wg2
zC*58C+bjM20lW4h{g&T7orgk^Np;&<(n+t_Q#dSF9oXDoD!T_S{y3jNK`i$2!g<cu
z@yp+?Yzq%et8RBTTRrAVsk)nT`-jaFzAV^!BwNw`*`T5!wr#>;{L8z$Po{u!DKk7M
zD_lMcf0DUBx#Z`DMOP?Hp+|b;c)Ueppl|`0^eZE^>a5K<`L-AF^;0=Ywv#s#owHTq
zd@Yfqs&#KY_^Sc5Af?$eZ;W0E*w*aQn<ksUC!_Y3a>Qy@H5d@$drPUokpNek^O$!;
zB5m4-+se+Oebo(2v1a9E+T2QCi187I!d#=+R-@coUX1hj|KjHMPD$nPUiD);+DDnV
z$H6k9t>lE$0TIYp7?TucLqVH`FD_-YHkt&f+1wo&t%*<(X2g^fhJa;Vdy!KwqGMAL
zeP!H$6JC@;?<varZrHKdtJf~mK{(vXOgu%k%;m%_>+dU%VUk8<^k3CUV=#$opp3-e
zw@2IuMILorDQ;PVZ?q$3iFa@8-f;~(fo?H9D!16V_M&`2@j^uLuF^cGBDJJ7Mc1|T
zUK3f4&^LH<(Xhb?tB-te^_*?1kPanMtKPi93nh`CyFoiCOF5oHnj++0OQT^lnPfP4
zN`W%LabXjfEV_p>51%rY2o)f5jxFIy&znp}cU1#fTjCc(YdP-Gj2@f<Pn1Y^4M#;$
z*fO-3d9aY>q#%yHYuZ2$65QBRX!Xu(Wwh5jp*`zU$MgJ^BNngY8;+hqgqC}!<%n70
zSWYx!<K2SD*q7pqZxrV3OwONb^OZB-BEcnc4Ksq;FDre8g!Bv=3KjSn!{blo{GPpo
z5^k`R?@{*N|3pZ$wc0Pw*A`3fuutyBr2$1Up9n^VrZaGyXBo!rJzivA!}93@_DzpT
z9HxuK17*_haRu@+-Xr|rRg98w<5^<Qa3pm6y<`5%#rWv?gjN#sBf90c5Bj+Gh-!M3
z4lt=|Yf=!at+1DT8^hlL@692bE{9i1G*mTF^BN#)RTa(0-k@2+3YZzKKl*B;GUI&$
z%$+avm9pO-hW0-m3i4@hdRu`>IITn6<%%9CyximbFSa6lR>r0o9JYu+)<2JDwVmU`
zVtC!n*K~GNTVb7pGwez`p&i_O>797DVy`o&fmVg@GLBHQ1QJ~34BS`30cxx?1^Tmj
zV0E#cDMXPEYbjob5Sn9gf4fH#ZCXkg5K5;vHMXZ|B(v!W`}RQiew3BUgKAF}z7v+d
ze(#ZT<)_YpAL(?~3b)3(@2loL+z~-|vdWQ;+$dPaYoXQu(qr(x2U{WO(c>LG5$(m9
zg!)l(EFHwA&FV+XVZ^TkPt&azw7-EM#7$rn(<U}ne4dS|Ed?Ye6$$exo?s?u^Zj3Z
z{=0<V$~+chRg@*JNgznMZ$vV@TQ2bp`S~rED1gihiMYyhkhyVSlTr+PSt`R$ejsFH
z+^?4!#ND0&?mNo^2+l*%F@gCI=MH20hxKLs`^he^msIxdj0}}Li>Ue>UchzIJb%mU
zIjF<w@)<TcyuSrj0_f~IYJ}C^4J5?(b)(uM_#8Gknk1LH5xMRkeUnWL+s~LYNA8B=
z%1ZD`^d$?ne-X0UFr0P<TrN*>J(9S0brxfXMe{mGL=sq|N>O2Toq}N&8&#jz0y2!s
z6@O^fSgEAc<UBgV`qxqN*O~IazKBzo4Q%MtqzTdZ1Hmw*Aj@YyX<nbyfTHSvZ%A?c
zTUv(DpGfA=R&GT($OCua#!mUs>HRM78Swqa5u9WlfAYCwhHVsDW9krNA)Q)X?a$~k
zH`c5}BsXipa6WdG5U!uv{CuX9JwvjK@RBZoX+E9v1t%_iAxwpo^(TplDa<yByr6{#
zKdUVB-GKqx@t0&QT-vXg3uk^6R%nMlX_wJ5zK%+hf!7n(`tPUCfUm6;cTbpEG`Mct
zcSYhH4d`K3Xc}ZW!0YP{+-Q9?r&WIFv0m{9f~P4L@_8Ob8BIE*U`wZ_Q5nJm)J_!M
z18&|zoqE=3ZtoZz>j35Jdt=OkacO*5$9wTxX1CWKtL*6I`7SlR_xb&OdW5O)yi2cs
ze%=yc!SK#{-#>fX4Hquykpzoib##t*9x)Se0XXJ9OGEi+-n=VVy*Jqqca%4eE5Fr!
zSzW2Ot0;yviMG}DI3I9XS^)JmXH9F&c}{&(ib*VshCL}2+bwSIqV=yu^8%ZG!+aPm
z_)2uK|6xJ%i>>o7THZU1Dxn^&q)ifjoO{^sE+*nf>s9|g>OQP-45GRtB3!Yj|0gPy
zLZ+XD1E=*9U)L86Hq>`CkFD)tnxE^?;`Z>{_L<pV&=W%wMH)azpdnF{0NeF@F@P_C
z<goLxG_##od3=>VlP%cKixAE|Ewd~T6OM@z?GF6uLC17^WbJd+r<<r~7J5xkk&F3s
zsvF;Z_^fLn>@hd$UMHEQt*##Apq0;mpjmWNRa>zV5Z@NZ3X^(MMf$r1+IN>Pol)$-
zGK6K5$Mp^akp#ib*k2sZ7Ri@u;_sx)28Nx5cz4CXbdRuZ^{!OPA{X~LQ$!0EjoB$s
z>@$e@Q>llSEa@KtymgX3^b-_ClD;Cf?S;wgLA2lrkhe#(ROcdY{j6MM5+Ol38}iwv
zlN(tixMSGIsV}N+KQ;<oP8T(d@~kLrt4N0bl%5#iB#LyYP_jhS%AJY>m#gj<;qh+V
zs2gPQOv1&SqI`Y>Zy2h>+M(MQ?1eH9`$9`Q&UHkjs?EKCK&k4hY~hviQmna;e;t(n
z6jT4JhX?O8E|P&Zes4&X=^w6GNg+5CY)uS-fo74@`cr`F>gVe$34g{hQX+lQMEov7
z9D0=*Z2%@+M$kvM3wES~SsE_391EtVJ&2zam8`x5XXT+$@qSJ?!?sty&lHlJIna!#
zTQcnWUZBi;Ezo3qzo&?}O}8vMLhQ^aU&YCb|0N{?XFa=Jbbdf_=QuXTw}r>w@2EV3
zN2$emO51-gPD)A{c4GIIz0258526UeOa&~)N%}A}@)qr2kH>UN?^s&Zia8Pd4-k=P
zjJ-(T4wx&1AE$oE?=ho}sg>zjpA3~jnZ<CmzYul~>oq<bOsqbvVe8-Tu}m~to<o6j
zOp{fm#I*rd6P=k{aiT~ErQe`!-4%^8*a>}=ybvrFA?8yn)!vWE)%3r9z)xbMGnNWL
zb2P*gcL}PnF7(M<`cmq_=Lk(WynBDxxpk|=TDx_7#WG-&HSOURkp-fK=VndBku^$0
zay*t!Nsnvz{MOXd8;XGFhMBLnPu9uUYD?tu0ogmF78V8s3Qz+Ze!&dEM?KBB-!Q5f
z!>f;x`a2)~)H@bWKZ+{*)pk!UMtFOu{D0j&|M%DYsrX>!rRthE|6|~rq>p6;L6@Yo
z&r8TB<sugDiOqlc?8JVVo>;eO`C=E(8LM4!IcB2eKI!}&MJjL!7meFtk|on38nUlA
zLRhY3oKLMRo5<|YeS`oXAuy3SS=VcrP@=&WNib|g{w&Y9*z6$5%2IT_ZCsU5$mZX8
zHS$TAR+plvK9Rz(9_tB<B*Vq3%C8uV(-ya9>_MY99r7ImY8DSuKiRN=n@n%io)tbz
zJhueW80Kq9G#2RsA0Sj%c}_$$c1D%zyXqT2CwMpfwmf2(moy&ZXZHek6IjV~)aM@T
z1Ne>}V+QuEqfO_dW~<v8E^vx27sP<1LpjKXPkoZ}VJrl2l`y102~hStH={;g0<`Bx
zYhwugm{N;GeDmE{>h?k~S8NSIFR(8W650;iS~3?D4QAQa%Ny3VkgCVi`$UycYAtR!
z6QH4b%!d9*vUI4-Zn*Y!{YN=|lOU(C){nXWu^ZTaPNH8CG6|`O>*B>BDjc(I=|&%)
z<{R|wn+5YZEQ3kPS=W|?$}Nkgn1g<ay;jh-ZC*n06U9JlR$h^%#O%x+jr<<*py&?B
z8><2%;823dJI#&&NiZ^ua_itMQuD5`3_nS7k78LbU4}j5iOU)cJ)|ykwo9Rh$sEA`
zIRN;#<c$yFSMkxG%Lo6Nug!lOr{wv|r3Ag|4_J|0zWw4Y*J!^~0TD`Ty9|1lWKJP=
z^_7x!@a?e4l8JZZbWHlq$UZqdx|<|R@nVRymNUxCrdp>^3wq4FR^~}*PZvobgj;9}
zj0OBPSXKDx0X<?DkaZ$4w#!ohWMXdbz@&P<AmL^pDl}!{o)c9GzS44Mqhwi$cJ_hQ
zM{XY(m1So>e`dp3{gJy@jgG{mSu_@_3=7wV+%zv5*NN6H;`1b1mH27x-qV_E;6RFf
z0O<#)Kj)4clYIli7{8mK_Y(c>Ips-`oF!dcP6A9LShz^eohw`hVp~sCrkl49wiOu8
z-t1!nq<uzDzG(c2pYEJtTN?m~MN0Ng9C0VHezyAP;Eb=ox>+ct<;{<4ktQf}ZJH#e
zh~}raDdM2<Z?|_P436G1tT_%zHtm1nRh?Vn8~Nc5A{^_h7_U<=hY>#c1(FFT!<=ra
z(B5%v*$v_IHOo0z6hn;V+LLUzOCOaQk9;>Cy;v4$*cHtdp9K*sRGbw{VL!lpZg3+G
z_(y-&)=->Y<S;IbW0K@`iDzo5(!ZPS-xcvLpHIy|pYeHhL(AYS;6lu`*P5qYw+k7s
zUIH6R1m*xT>JfsYQ6HHLc>j2Kuc)UHH``AJnUX5gTG!XY9^LjMk$%YVMz5^M;S_ns
zDQ~_mF&+A|xi(7+lm0NZcPBFkql{yut$>Y}HLYN9=XA=wQI+tjPsE)jYbTz2M<pz)
z`*WHzsqryIU0wv(aL4aU6xu6gdvEGQ{9d0b(>N?Lt0Nw39p19U{m9{_N=~x+eI<QE
zg;iINSwoX(fJa45cjis8@^?1w-Es#>U`r-d%wA%XM9lsqBF={5s#E0+dWsCTwE(y#
z%!OOzklON8H{&dxVN%m@?~+LB-~lBL5uwl77hCV_l|#M1f8OZ>RGe#g#ybA8Th+L7
zm23dF{3WG^A&EvdWzVh0`~wR7Tm}G_-iP}CcFiEH{6X8}qEXU1ZW>r`Dxcugc{}lX
zOQRy$9EKiI`ReyEX87F{EYw6Xev!d5j7K(!OLlZavKyD|G8mMAY)KuQJB2WiH|kBu
zMnMSJ+8-KbxY=;Ya9)1}r257fFxDSxcECzBB^)wq3dBW6;fn``-$w$&BkG<(ZH(pO
z)OIZ%W;^4bd&J*u*rXV32CD7rm9lgi_9Pk)C^S0AH1d}kCQF;lKn2m|GjN=C+yH81
z>*gSn&{WtxLbeC%HGG%2T9WM-Rs%dqHvI7Jj%|kverF5*4fiH)2Hf(VMcN)t1IGTO
zE1k_P+iXXKY{t&Nt!#OcIPxefG1Y8S;UNy0%jtH%SD))2hKc^OGHdqj1lP#8amdhz
z>FNv`5OR&h@r8D`y+C{yNrs8t-nA$F?2D6Twu3@a6K$9ZlYq^0r{rBo27_kdt<A6@
z<M(9@J-6PjSd;bOl)t9ZFubX;LkjRovHm=^GXksRyJG`cO~C?8-E5~r!rH#ef*HW8
z!%5!6`w!-~qUIbZD~!FP^K$5pH{H~>&6k71m;MO7Mo5Er#?aw=@gnn<=0&E=>89q0
z9<n6{%@!Js?9rO5aLuU6bK|8Te4aYvyB{+{gQ|ZN>b(ZT2$2VQ31Yu7$DQQV@Z(`h
zHfch(H;3?Gi|JY~<L$)@MxI$)<|5e(Me!6-;Upy)J_hhl@W}?nA~O$IWl~LffiK=Y
zW4#9=4^4FRbm}4PQ{P>K@B@&b(Muzi-}ae2u!$??Aht^DX}+K4nEAofT%?#iJQROA
z5axfw5i}8X^mm)w`3pWZwS1}D`F6)7o_vC}OSA`qef`snrtGv&JMe|1(BR*+e6y4M
zQ?%dOiu}j+N8fp3?()9ifPj}DiF&KeETw+6pKIuEopuEq8ixYqy|ZSQir$hWjn;%O
znZFOs{hB8~lXIF&z?j)v;)T3=fBW#I6+UlLzAz?&w$CF8zNo#fIT&BB_SsQ`n{xI(
zHO3f|aHFTNK~q0>WwIrxp2949>$c;=cfF)8`~ig~S=w?9orqx~wnnXD+Ou!A-a?kP
zSkL(wtmK9^o}9TUA^dakS?m0`&ZK<4XnH@p7_k0(sPQ>~)eG9xJpNpyjqy;~S)}~k
zz{Eu;NonJYcJ8y=6v-sU$v-n*MiN4yMvHXw(uyh&Cnz5CwA}Sz)vI!Z%);gROOYd|
z)~5)5M2NHEYe+a$yF@lhH467d`HSVqR&EqaRA&wAi6G%!Ug$Hq_qfmJJ1<(oo#MmU
zs*b_^I&_*St>ooo)C!CBq-gp8qjf1|LI&s{r6`I>HW{a6D?+OV3qLLeN$t=LkTQ!R
z&7b_Z=3ngckx-_G^YN=7n7m6Y!zs=j0rEj`#*T7HFcn!SWT+5v7^vdS+}B7P%m)!y
zBMjC!v1mWNR!*wVIfe-u!M0VCzO~fx4FXR;$YxKNp1`ae?5j8COLY1QHO1n}&LMb#
zkl+VlFG&Jo+3zqc+Lc?a%@^HmGMSWg25g&ZRfk93Z|X7ga*R!4cjmmjX-nZEPq1yg
zcGp+B)OD}OgTi?L!xVebw-77p17=fI<jiNu)Za|fR&qV#xpt_WdVxsCadp28FQNfx
z;1irQh(5lE8@bmc0(R8zM{HUDpaJ9D+k|*_bcQ&Ks95k?SCcp!>wZez9?t;1PkL?r
z`5vz2mAMo;><zPg=90{GP=`3|C&1VbgqsWfy)R_bwEbUS#j;m?HJo*4hG{}MKFu(7
z54RZ%qZ>n}I%T?kf|F~4{Vi+8!x(G2HHBLXs^45SpCN2YA_Te0%s^K`=gl>*Q|?Oz
zB||nAE)dl<thTc)Wp%rxBM*RutjyRz<#Z8wxZaT8X+yS{xyISeU_xhZr1Rhm0*?qH
zTOhjk4c2{QQJ|&5J*s$x4=2IOa!3RxYA+AJd(k@8^9_2K%<Ah9HS^mXo~-vQJ1j&M
zt?|j9i)_7T;biZE>ulmwbv)O&m>$wEO=P!oG#viD9eE=osIaajul`eera^PHg-D-3
zU7i37X)Z|-{&s;LM;WC_<gArkLZCy1<ZU6)@1HZKaN6=zr@VX&+f<QllYp(Z-DT)!
z!ID#LnYo)tj<Vcl@_#(tEg0qm;hL8j%l0%~im57T?3?+Fs49_-QKBi_k9Sr_H&M@#
z?kY3J1}Cl^bHZzOSn5KBWW$0V5O+U;^gx|Sd2wPO{HoTpwz~1s{K2>V-^k3i?}y!c
zokO^$ES*2UcUrlbT)h^1m}e{LiGXL(o+4_ETXXi3$kQeSdU*RnT#v-jZn5mD)`LFr
zEZc8Yl)uM++addCNQ##TgLCf1pn^LHba?w&UfRub9EaJI@%}Uzw^Pda#MO~^A~&X{
zqUdfZUsuZzcHo(Q%tD)mXdL8}@8d*Rmt+!PCox-P`s&2!I)V~<Fe}nm=_E|%@VXh}
zCDfm4hd5EO)i;-p5z+aT$3&^0g`of4$IM!7s&e1Ry$1vqXg*uR9J-IwtbPT{ttg+A
z_?XP9<%=bspDulM_EfL5?`s!5c3TwKyLO0k8g#(!81r0xAHC0gdL0gtaG+l6S>Y_g
z?=Ea`SO~an)FpO^lhNrjeD6T*GeNel=R`K1|F9*##QJy6qUQ$F!cw1um5%>kef0m)
zT%+P%;lMASeboM|E2AUhAVayKQcAr;IbJQmO3^JOv6_aI{z6{G>B-Vq5V;cJHj;eM
zv+T>t#<O30vnr2=6JIR;ii7VjF7khTVT#1GdUN>Mv;W*aHP}k{A?R7apFKWzeycRH
z_c>Ra(*KcP;_p;xEPbD9(1anun3u8$SWr6wr`iMpr3Cg|zBGzu5iS#+o=>RoD;)py
z0w}2Zt8)5}A6gbdin>Gj*N<P`f3-lg(wDZO(l!l10sm}nmHBzmC;PyC?z6AP^SrBZ
z*3A6ZurMvbNt;Ohz<#pDJF%lU3(l3^R7&@0FKe0QQjZ|%zy98T@j{jmixl<sRi4UU
zoVIKDPT7lqfXsmL>e~6%aKR(Ft`oBdpZeZ&w>!9*SVoc?=$V6v*CQZ3vRHb?UY1qB
zN7kFk1W<;Mm|7ger{uTpd%pvEjhy4u|L>P#fVq_0`78gr6bK45$^`I}W`=7;z&T+4
z#ggMJo)I*F`u=PT9A$wqL-PIzKgeR(qPT#b@sxEi*KM5rTZY%ad(#N|X116xVo%|3
zi{O9T3I6eBn+5<~1+d>8|MjH)-@W*O3pTsz4;(W8@#*}}{~Plwraj=~rx7&u;``gV
z`@j3>|G3cq`2zmu39z|+A_4e4&4BMbs3f`=`nMCr|6{ZMYj^3P!iw>kT0mmVhjZo}
zZl9zIT5EX%gYZ0|M(^bR<{#A%3BWYam_ASUR@9Tg2H-L%ipiVL7`(D!`G2?A{#ph6
zzi`q|TG(Ho?yqRu@;}8#`vcH4qyc?ctb5C}?jQ4$zh8I%?x(q_jadMs+u5|1@wp!~
zaDPW6hv{ece|KS1V6bHh_WI1fXVm=lnf|vI{H~bEf$QLXroR`E{aY0acr&RL!_;XF
zFV6orLa+b)4E*b1pana5@B4~}s()?e|5x9H6;KYLt(@n%*8h9=STGcn5Q6MqZa)0m
z&gnnM>YxEI%*GCTe-css-+FKxdziq)wMUIB<k5fYWC7E=|GEADbNm1QUHu{9mw(iL
z2R{_%!nJ+PJpp3>BdHP8f3)<+RqY||lFI_o7e9jjTA@QKv#B$v+y-ATw@a^=UF`)w
z$-(hI8_IaiaCy>h_I>FcAx6@+=2t$4_fxN&DS=$&zgnIFoEU6F-j?b6ip0#RR@DxY
zuo+k-llvdn%!B~aDYxW)a{qN;Y6trxYXv9=r0?bmv|-Bf5RCk43JgK#N@M)&^)Z<O
zfbzwFnPCx0!h!4zlR;PEaq$Y92m)Tc+^+dop{l_UAm+Lpj5!nLCTGw`2e|T%K*+;!
zdk1_I%<p~oI_JQEQri<;6wG_Z{zUHZr@g9WtImJ5+CLBk^x?u~&d2+aw-XIXOPKW*
z_48ps85rrmZgdWg4jRm$Z2EXJ6f-}Vb8z%A&^HgiU6v?ufdWig5kM6)Sv&oXQHN31
ztt!=RpAwsZ&w~UpAZv-GNh&889OU%4tco!naOQsP8AorOWByWO;2dVyyE9MOAS$^c
z){p;3Yz$zfllWh3n1q6GVAl6eNPIK!wej4*w)LKJZVv%t_+KG`kPGyA!=;2TAYu|=
z+)hM6&J41^`0#W63JHqNxrJm-ui@-*6#&PO0DUJ-zeKGZ|H|U~$3t<+y3@fD6O76J
zL!I`+k0N8dhqI8F5l%og+0|n}*od4C0=Yxx7+VnUnL|08&(OyEe1Q!ZIQH?2Q7Q0e
zbTwuHwVnvLWIKuOb+9y_PaoHJIX0ZkM>g9T@2Bh><jK_S2^!6exluhbaZ8ima}>va
zX(XL1$m{*>Z#VD8VX{@jozO$kVEaq(5jRTbAe@R$z=|dwJ_CsM%|MxSPbe2e&#Gcn
z*(iGMfnDN&TSNsfF`2cVU}CvC)4n_NQ_qABVCp|!d6FEz7O7==4^(#{#mwm#Z=#b<
zZGTb<*t1vwcym=fSit4;44BD?CZwE|qlNbS?M9kh$4t{cHW(RWqFtzscO&#X$ELw2
zJTpP;YpR?h$zqo=<Bpx9Zw++^KlKkWIXB2|6ST|9enp;CFYcY}Q((KW#N3D>8u+#2
zTml}hllm(R_&o-=<jrtYJN!Di_PcsR8O+(-$j$Uh)*JYVxiFnxUd&vESW9id{t$d&
z=40m&(`54QFPJ!*X5I9?n~srP6QAqx4wZ23;-nwfu6WUB7BjMAG4p%jVB{Q57_;zI
zM*#IEV1-e=zwrb^d5A=FN)7L7E<l3T3lL;#)0#8^&has}kx|Cnf7_z)^Wcb6qL?I4
zKy9}@VjecYWpdh80yK(8!O!#q4ZM~TLX|1hognm1`Z*XQ7;h|DZaSv73MIn4m_Qla
zV7oA`66IsC|FnG>V_NR60Z$b&rr8LL+p+tZD0*J=(IhgUcS6m6PkrYNYq?K9UqkG_
z6XeAwaf#rFwQv)2q1^EI;_{aPY+EFTFea0743sB)ewUckm$jW}M_-mgAh4)Ahh{qs
zj{%=tD(TZsZKAM;RTu^7b&RqR2QKyeAH8QQPQGr#aosj9cZ2EU1QuG7j~IJ4KVm1G
zjfRWKh7bTaPHq?n@9=qbs-OKHpd_JEnw<eR<X80!$3}M+Y5=3?Utnsy#@YsrQW(6J
zuj5z^w2fHvRs8_gAF?`iqcwZrcaL-F@LYq5AF(H(J747VcZPMZATJd@xHg$c>q8Et
z_hMs$f&aeDrEbnP!$~6^$_s%=yB~U6yu}#<a2l{Ka(n@7@<Kj{KwAkI0f%0q`Y*3z
zo-pkB2G&QLHM<NZ%l#SK?UUxQYk<D4asq@1>lE)rr#+^A!_g~n<N`P#21glBEaXT^
ztzs2!hP`nH=DIFsotO%_ezN3x@CPUpqYCXx8V=lx`RtDxw;BJ3vZ)yb*Ejv(UWQOK
z*e;|uflb|Xy3;U=3SHNqFv=B+IAGar`xF-*2fen0Lz46ei<2QP<t{<upKJX(sp*T?
zWIkXl@;x9rDad1^s|ue(s!dZL4Juh~4V*{Tyc(ZARsC71$_fO8Jl#27k+~d?k!T?x
zrFer;J%q|!tZDlna3h5f3cl;Q8BbN&@X;kv&V6DlRbO*(2`LVK#UzA})-K<h!yF!v
z^*$ire%LgFnVpsZg;JFzk>ahB)88Hq624%c08ZOwYGzcq>&#E6ORH5+AO^9`^Gunt
z+*g=oJ0)6T9KE^?u-eh)Yy8_~fie_~K#hF119tn-p%oA)R6q!WUatoS(Q#Jne*eZF
z=xL8K79<}d11_fN2LapZAx{)XFhm&HN7{84;1sF3Nqi`{MeZ-4rs>mqG2A-*-K#sm
zv*p>9L*f4U{NqHJ74>Bgyn+U!Fzrv_oBN<ilV~@TPR2G52TIz5!1jRyOlmrOMgOed
z@3$)^4|pdl_S={_p$tK}hwCzSD)H>z7e~Kt>Ve8;OF*sgkdp-Lgvs4L44HX)8okIa
zgKvN9gSG-q*E06R2+*622Q}&RJNhorG@D2Aj1W?^L(1fsa(HC80c2waXv@?SXf#;+
z9tCk=20P>JzdL8i=L7R><uV?qIr^mS$B6=a;FjMCwqMi@;js_fw@g5qr`8JoIVL{g
zv^-3{!3Gc&cTs8ZRUjQFstww4ow`Ruvg!j}!SE#f>C|l&$HAGt$gqu)ki<+}V+biy
zO9(E4*(ICKeZUoZog~6f6Oo=3H#)%CJim?1Xrb)>u;?c2_(o(lDB&B(GvhYB`ZCFm
zlo*E#bE|(DtiJ*k(pJC?CNS3GIvdO;GiyBKuq^{;6EUH3mSn8t=I8h9mlLR*^g3V~
zCCQ^oyoc+y1t79jH*rgx<%2Jgv<ujS{2Te@`8BV_Ca~8X&*tqPA?3L*&T9>TJRJL7
z%|4B1*)d>XNzS-Yr3~whd+t!9qt^0W^X>9QY_UIyfT6}rZE8g|8$*5W%y!+ev17Ao
zz-{lsWec&lD5YQOlUfq`!pgZCqg%SrC+oJ;p;^<pM$DS|>>V4Jy`+w(6^V{ZKVWU(
zyAVMf<!Sm%=N7G%`Lar%as6|{<~PQAFWu59jG-mf!lkCa@zB-_UNhff5gU^_{@$-?
zVsMexbwFh;`Bd&l^A~(mr3|vFGFJ;v?_nL4==a}jF81*EUkx>ZbNl#E{xFlWj-z(V
z!I2PCkob%yd6-W>c(F{9)Ftk^JdiBKBn>Qbe>v$0xApuKywq*|)K@64cKlLp7HrBX
z#Wc5N1p4k3#mOWWoC7P^bYt=(K-PBxQZ+oS8jTlVX1doavsQzSXv`+j&&P#hSTh%*
zriCn}tEZq)kopl|`4M}Lfnq~TVyi;FRDl##TGAhg6DgPX-sOJAhpXN9AK~2ez13qD
z*Z|xC1}Nvn_9)C~BQ^D*E%J<nTtiDGQ|&sj8R&z&w4WftU6<<Sd|M_{WQ`^E0kPkz
zLVeW<C~m|h!O8G@pB*2qk6H|~;?$;a6`iNvl_j-ce$tp?9(yIHU%?L}{R)7|z7i%J
zXv0)!PV?dHuR$3%ZUH<Sd_sX=q6FjFM?X$iTH>DU8=15?9K&B;Mv<gG*xpME*Hbiu
zH8q0*Z<)@3qQzFhmZc2vjt(jUPA);c<(V2WW@}ia>7$i|VV+f+ruA3oY<rK^Ab~u7
zBhIHn4TGn?$3>Q7{_%@wN;&zF0gogR$Mi@ts63;VEO$`JwRd4XL}H6j$BpuSw^QnA
z`@6C`?b)o%H(I>IU{JHSG7uTEcucL7cbJoCyl$pBCwqdi6!4yKj`o$i#r%%O$~g&*
zWs4Y#_iggAZT0}{j>cDbp;XS`0RByBPF>$jAh(F2H5~6Sn5|SjM=Bc!)6FXm1{|Ay
zd5gaCBz+%A%tjT)-a}Gax0g_S;6F^N&OImpM3{upmnB123o=th`y`EIaWKySaJ{6Y
zW-m=H4^8Na6(8%r#CSeR+`w{X<$?ZeqBdhf-af0yzU^mB^_Cq%DF0|MgrzYO$5?a0
z84!6xFg&JxXdta(r$$owz-yEWn(+$>BDOM2bZvn=oOp6E9Oql4OFHWLeqvQUIMWZ+
zuI}ANk%&J`d>E5$;YIS=7<guYif`MBojm^=^l`IKlFv};^kVzvp!*w{{JaigFO{<g
z#Vh9<YxTKafh#ox*#7sv6nhGbwZNj>Y_h^c7Mg^PK4@vCe%rH%QTuH7&~O5sv-Lq+
zjt-5H;A|LCJ=o4I6$o^1`x>RQkMK!7xU&;_h60?i!cmw$ntfZa=$SxxWoTu*ki4Ql
zurcs{nse+ij-IuiJgU_+8nQf9^*#6p&BN4+)YQ;zE`;(ZTywQRkgo=)hL4pZZa^zi
zVeQJK!cNcHcgJTAze`}LF!)u(Sct-yxFXhm-&oiUuOG#f46|Z>#ekXPFzcO2Nfk-^
z@`s%CCDK~!i>c&TNa`PZK((oWk-s4rrb;8s?eFQ*xvl%3Z>`vcv;=q&w$a&tpmD!H
zq7l7a;w0$NUV9bpT%iz!KhB+AByij+`lZ9|@n}V_UKBgYTG@9$wwj^^fdseZY-btZ
z&O{JBwAie6-tu}&?dAKf_A_mNY-zJ!piuWC%DG=97DB=z*(u=BeRjq5-s7p4Bgs8(
zS=x<^IEZ?_B}w=#ii_v#dyA`FiEDI&Qi5xIb|MFPZ!M&w?*$ZENK-xO*|q$I;oaEb
zjoZbclTqG10_`W&gon%<u?y=SzF@f!q11^Mi0mKcPO*A`-tlaC_rM6QWI71p9uAb@
z#%JWWn0tlGM=@eU1v6C6Zl5P6EPZs#3(Ot|oE|cjacId|n`&T(o&;PAYSpCR55-4A
zAN@GB{LnEkLYNBnqBySj(vmk{-^UZHkIs<H7y-&+Jx5dy?zKw{5t2`}oFBGmb3y_w
z`C#B~&x%xXzkIM;Ka0z^x(`Z|5hD*H>vk;3XqTrQ)=eoZpxC@?t!Zr^Po7jyx%!6>
z6?ZmHlD#=|MrEXy6rCZr?QV>m&$!~vt?O<g=5iDK?&i#raxGdu%0_-5iM0m0f-&X*
zC6r|hqubp$R!ANdifJ$xtwkh9*gmvRq$s^Pr4>~4QaceoS+NovREfYPr`^ro%H>h1
z&^ACo33joAR^zn9J3YQxDm~4<in3|8<r@V&CCd0x#vrFaxx4+lbkndW=s_ZM&^?Ld
z9R>mBEsj7=+O;Oe;1k9c76(V*qeuDi3$RJ|lLuUd1BpS!tA250TB1!00YEJvP{r0u
zZOB~JeE)6~w3Gi;8;LTGIHVVvGzUFns?XwkM+@i!Y$HHiu}wIKcbe41kx3l<swLDn
zCM;U?HlAZ46?2hdQs9G~UPu4(ygM2A*l`sQagw|@9(#mVoyC=KYQ_ExnHg$Ea|OwS
zVD}d6kDM23*YfqhZ*-BU3<kc;V^B|d>wZwDUlqVU$LKpunaxL0k<V9>p5URm5Wo?I
zYRsjeaD}=*)Gn37A$dIErZN-j&v=M4vYeFPbu0olm75urOOmHU%Y5?&!A<cpT^tlc
zs~gOaoN3LqU2suD?D!oFN{BPchPPS$8d3>qHmVJlmk-%IiQOn@qVffy{mXQ7O;1fn
z&a98pbG@)+s{h<pDDX^LN*5om<IKNr3p&v>Y97O$<9C2nN~L92FK?IH9?QsI<rZR8
z%%y8i;pB!wbzhKR4<ZK87E0u1Pk|Sr(3b8j@lu?#&{Cjq5E_D%S=U<U@*JWcV2~U0
z2l1%~#xs`ZLrQ;^2?tl!g8nGH8>3idWl~DrNvHEPne1shGYXMop}oQt;Rkk9c;<4c
zA-$XWw#tD?R|BL$W~*;)^OTI=Me_W^dbPQw7QC~|9$fx!!A!x<Vtbr4Hc1SoFBoO=
zhL`Jflj#?tRA{aZopPHKEZ0&s3;xddgb?7n@NQEOZF}9b&Cg3MEF!GCAi+Hqb(kcg
zo_C#JOW{PeriJYxtwja6%^N0^Zkm**5nqkl3@mB5%FWUj`;FQyI5&FK=I_^AHOtu1
zbPB|Dq43N{>@UN`;LCdK1EmyAhJ0=mh#O7Ax)m$KO*W*TMY}}&Iu?_-o}t}`B`vDq
z;u{*s=y!pK*O$*oIJUuQjqN&B_?X|MuNub=2lV~{G|u;(lh)v+o+^Io(y#GTQ^c6q
zqM5=N-0rwu_S|R=_-?r;Q62rP#4Ojzbedv|z>Z)zvL*Vk##xm6BO6y`;r8x}5x$3Y
zvVKjzwof;xRHzp7AQfz;H)izHaQUBR4V=(bJ|O=R_iNRQCSIw$n$&^4t8xE{XjXDT
zbZ#L1Ni^Ze8)f8+gz(E3k?{k!25bm_pntg^D3N6=Ni2^xOa^OI5RAyITrL=~twugE
zJGgij8Gt4-==oifeqkOtF+hRKby*(4aqc;{f-#dy(Pw(9w7vaJDbXfv4(T(bWY7^^
zDVkbOf&SWwbsWRLk`@j{i5QqhRbHXjd=%SodxBN3;;K?bYn}~VLcolh90nM(;<_rq
zEci6^Zptc4iYe1XOW|m8v*!=vaiqUc!GnRqamCb@^_n!{LHArw37e6oOqj;Tc^aP;
zb?3&5)c15JJVyK*MX}X>??GvkPMo-fvnM>>a&6$6+5p8<?+Y9<9%ULy<y6SFPtUc+
z*mF%B5(m>?$22AyKW=L3Q#iaLcZic^tu>M;KU!L|HLtM=O0$VO@GCA+t5P*-asTd>
zvNTk)z1))+<F{$y-=InHtmqxQLe5f%#7u~*wRDr3OxD{!hd9v#b&cJi#pQ^<UmSS>
zERqz<ohURp$a@K3M=R4S@QYFSp;PLs+bOTIJkp3y)SeE+ENqLW;j6J|>>)f20W-{k
zI0*mF1>K^=UXiBn0d6bXv)bbOX=ccjt54RJg?Cl`B+ok8cB0taHKvcygWG}yh=E|K
z0pT|XMf)EMBx4UN)WrIwXAT5UW6@<>zA^m05@cV!z4>`LXo40q<|Gb@{T4s>QH)BO
zWIL<>Tq!&L+DUdYq(oIVH|Lsm7UA89&_GnKait_oc6f4c=Wl32Xf#(>7UEFbz8Ue=
zxg7S7i*@oIq8-}3pS@R>?vWUr<W_w*0y8?eqtUUBp5)!MN!?>U)EyaLG!ojiv>msZ
zMDG+8P44Hq^@?_>yvvdP$HYiO`s&2JWgiFUyA#NVsHqj{l`V?_-%k#!doxtq0&N0z
zm(nq*3}jmVh1OOOYR3uYjSjISwMe5bk%OAe7N&j)wp%-=G|Te2$j8aLpJ@<lsxi(j
zSubkYm;40-=e$1k>bpLie|tkEt<>u1d&ts0uqE4GyH680!U>aEl(_7fRkM9mlW$ku
zp@)&P_a`P^d-)pm!%Z?btMC2IVW7c8lWj3x+Oy=ZSTFAk-9s;rUS+$D3T8Q^)b%tu
z{H%9cU7t3x=#8_ilTuZ^A9d=XVVTxROL2xf`EqWpy%vE_wJ*2llf<rKVw&p1vev*a
zHj)jZ8<%b7?N%CZy`fH7kIcls@Rf=|y=}Wu$n~(j?@{7eiRQj{f@j`6+6jr}O=^pA
zE3TEEdt}K(g{_Zw^^U_DACWeacInfFm{sGz$HPij6+#*tDxi*>PA0P5(}~h6xbR20
z+*(PTQ<QgMh06R%6dJGb(Hi)hggY8U&*%u3Dp>?mI%knxfgyLqN33q-?8#6(R~DOW
zosDLg`{-;+lB87!v$^AUvnk`#ygGaZul&|3`7D1^1}vHLraEiP+AC(}k~H(Rw`dHO
z)L6(aYn|raRyg<_GR=J$^~`)EEfa=v*?lzL`_h>D$~Z51#+AMqJ<Er#oZxbuemeA*
z_wOaQdbPy4E0!JO?=`G63_W3471?;2gC&yAWO(2K6-v&@t`+&xzyR3Z^h7PGmI&d4
z@BMy7-&z+)QM)fD>DD804jdzBnRU;uhiUBfxoh~)SJHW$W}MXvI<fOsE|hG0p9Mm!
z0dJR6Q$aCl(e6x4cgd3-seT6Au_n8AVag`gwv^N%>!c3jasBjOkmLc|haPDI5qh;9
zYl<d<)|gJmNx#*Lr)TxB(Z}f>XRvSYtuq|Xs9)IU3UFnnA6cT`&|A^5rX+kVGq_eX
znijV4RO69|HZ-89`BjD++s))L3s;K6yN9mH3Lm;o5Jzt-I&0Jl?e@%Q59vyI<gwVk
zhKiNTQh(BgE-x8T`bDswSnTV;XhKTJuUaMB4WXm5q#U1)MrHZN3k<paG5iSVWNRFY
z^%tk<RS`Q&4m(*(?F^Tt!?KNwgbD2nB*Q+$^aw^zf<<u4gQ|K5ZWgXW$xSihZx0@P
zN?P__Vqr=_c&m(X&s<ODlkKNJz4~%7=UtZ;62^f8CsdbY<B1_lPd4sLcC^yvV&+pc
zjMrOM{pT}{r@lNT5FDTtgw~iY^=d0LJ%7E1x;BKW6Fl`v35EaKFs9rRQlrczZw;-L
zqi34wc_u<0D5Kc$>Bh5iAtLO@YqwJ4mK03~@ikgv>uDV=BdeddU!zmO8<&qx{X~a4
z+}7ha>DnzmU1-waUumE8J1j79bInK+Xf^x&&LWDR%P~4QwdtZp`w}JNlU<B|j>1F8
zO8}JrVyp_|71pa@kh9*9{G;eV)g*nTsvKj$)5t37+<dd(1}$FSgKw>OL^@NX1~)R8
zxX~@8s=0f7=+aqE|K_XvbxH26IQHo*G08tZSdYB7XpEq3mvQ!`+7y+M-WUWI%O%qF
zaueey$N%y92sNIk<;CzHG}|Oi9MKme+`Ff-!L)Lsn_>9n_8aX2P6u2njYO`MSAMIg
zyA1QF3|rH+Yo&1bt?qKs9&6FarQYKm&6Vm84PUq};6+T?QxoG->Do2<_@oe!))gGt
zqjwS)o;$Roe)w#l!>GMAB3u=}@IVrz5##P>9BjEe=jx8Ac4B}+L@@NKOH;keD?>Z^
zUQw|aOq^eRh)kfp;3SJW8R!%)**uT-5uLx8OH1<Bk5?+=!tTi+D$Il?$y<i+oW|;G
zO~%DgB{|Dcb$U~rr)+!Sf>xoEq_A>ldFhouDciJI<sh3ZDoF$JgEn-b!%E{|A6<r~
z-d>`g{79h@MGahyk^(+dgsZGCj9;0G*`VE#bPe#Ay0Gnv(7b=bXS5XFV}}|W6kn1&
zw;$@SoNL{%K?u)}y)o-+`JS%#1Y~@eeLM6Q)QM#<X{!1dQJ$7+T%7t7m*LwZFE+S#
zs$_tPQ7{ujuJ|GOn(X(5B*`|>X4>}2)com<o9kZ-Q{$`IXmYDFFu<maclyG#nDn=v
zffr7wPM~QT^f1x$BS2&}Ij%<u)%S(A3X|ZJ6B$h8+5EAniYN0_cj;74RsA>q;q%+r
z!`vtuRZ4y#wH(^r=1nWr=?74v3<kS|nA;1^6VB!7bT<3Ff$$Q|^(Knf%|Kg2u{xBE
z_CSwsdZf=)=`CZemb(VisN{_=hndna1*Y91s|4PI_1`Vpf_re;+XvoYhK^?w=iQQg
z7VHMxyLM3~J@U~EI`3aRCnIj_azYAFB2h9IX2>=j_Y!^3&t2pTma}CT!%h-UozpwT
zm$6lYl9yMzlLfBI?DbNF2ZgKq8LHA!M@GA7?kVmGO5`chm-QC-*D`xW`nFCrfzift
ztSgC8b~|eei*xet2HHr~j45H;vok_hi>ROu;-0G%KNS|~+Da?Ux>jKN>dYofbQc!u
zVLJO#rz`9JB3blv02SY5p--kUkM1rM_mT#D$=)EIG0L{AO)0mw6ife}#f_-uO+~?&
z!0_$B@Ib#8Ls|kmcKlWdJ7DjO>CA_3pUBmivum9^b&ODvE?nH!ah^?PTMF{B8&Sr`
zbE9#9I*?jCkbt!n3tn2}r3GUE*P!Jshs530%sy7$LRe}EcM?@w48-z9tFQ^4L;wdj
z7>r==&dMh`IL{G2HMD=)cZcl2<oCA)rLC7Z+t?&<iO0-8GEAkDyw05^_NY|KQw}Rl
zQ$N*dn<MK?^RRw?-iBEJ@@n&)N-66(*FBavjTPhFn>}60m`NpMH^RihuE@e}2@{VA
zIi`O2(amYr1|NwiL%jTeR{c$0>=|<GI(S@!&h_%_`W`M*XS~3%kE$lx;IwK)#+33J
zVB0;MSy~tvAx8Aq0j$I?I&_z?{b<l6Z$?uFBg`UnUFoCSh^>#3Nk=WR`$$=5+VvaI
zGTab(sRZJM6ik09XyLFWH&zg$S9hF*g>P65oK8N*fm?O`%#fOWh3#FqwprQ{loYK@
z8H*F00aIB|GeSUOx`IQ3YPm>J0{Lc)78F>1rAQ6$Guc4vZXEbato-*>kM5TAN?c=y
zP~g|VK6YMiS`Utr=#+#B+BbR3;<^X)RndK<fleeR)SEszx%R?hpr;3Yo(c13;{K&^
zGS_&eJN{s;>{2rtdSefS>P#IQpd*uS5Cc7Bfyal>yur&CKRhrelRgzbBSNJm@Xhn2
ztQ$h=6<1>qfiJxD=99}__fIF!M9B_eiD~cebr+0hcxSZpoY+mKA7h=x5Nk-^RplU+
zTH0Q!ftBv#uL7?~v80@aOat%H9`Sf>|G{E8$-$h33#+T6l-#dR)o1Wx$3=1BwG49C
zwxlm!Z)U%S+3n+U>=6Z()w0kZL@S50dm-=z&$JJg_JW>J6z;45Z?5i!%wxYPU|w;*
z8#i6Uvph^W*@7h6pXNfHqJpX<{pFzAH2W5AVKXrVoutap##D!>Z2$m1&KV3gDop<=
zrd<)6Y@^vQ1!{h!Kai#f)YX2Dzg*J&D%Y&Lun1GRhf_j+=T6_0*7IO4`jV9Ikj_Zz
z(EI?0c`1Vv%d(9$im%fyiz2n>X+VtjCM2}}EQ()4-blA4v+o`oH4J6GInm<wGfy?J
z!OTr5L0W0($eMP`YpB|ab}373Vx9WWgP&1$vaOAXVkB9C*TvTVy(Sw(V6TsR^-S5d
zjVbKLstogpAmOfbZrPBLFYSX$*%)|KLlR9rFg@^FkU~ne0SGIpW*K&d%U!~ptA@BC
z#T{76*fu4xz+BEP%`f9Xkz{1kRx9ATaq>qTgM*<X88OwsB@4Yhzr}8xaH8f~As+t=
zg$kr-g3=GOg>wKrPg4HSrG-hD?%&*>7VIrWo4wnYiXO}F7ftw%yS%~&Xulr^Y*X0Z
z=Gkh6M_+Kq$1!wfaDOgillRXozVBUPTh#MaGe#|31%}F97iY#RN7M1v9XzDGQt}&V
zUP(v2W+z?96|2Xg?xj-*J0K1-&evNV6-;O6*+;l2ow7%_FXhKTecvR)@Ur^M`oeDH
z+n;2{pb;LmS%bbd1Qwv=@wPHMOTF&(29JYpIC+>jEVAnLhj6h<vm!O4Az>4Nci+fv
zV(pjqkd%!l92IqxR@%P19_hbl=g@^`iQQn!C(wbEltA$p2J(lAJ)MwU7h4x>VS$ZW
zO747fNDKdgJC_afIjBrK$|xykN)_Fs`$&>`Q7c>3T`Ag~AK9<jB}RjXkUG+*lzM3L
z(DANbG-ObVeMdaZpZajUip~{Lky`O3egI@T3^xlGUg}fig3iGUo>J0tKCmd^08Kp?
zs4~qHIqVYhO8>60>=z-NmhJJib8^uuc=_NfY_xG!ex&sOA?&T=qVBr3Z)p%ILAt>K
zlm<t707)rDQ4ndQyOC~?2I+1@qy?l)x<OLu?igyQch7lV=XKrB{XWlq|LNx=6TjJe
z?X}iE*7qRwKV+lcCGi+}v~8yWD{@ypKB)RxygDr(pH6VJPYF3Ue>I=S?X-wx30>@r
zLRk$xO>s&SW4x&x32}_K#k~n8UdDhbMK_bDEDpnmomYh?L4uZzqn<_awX&N)bae8O
zIIOM(-6>$P%y&h$`N7S3WN50Ub$u(ss$AKl3SW&s)P00f9F~)svsavbaQx@nLY;<L
z%KHsBPLJ~5(E-PIt4W*}JSykL`k{dDPj*^z?2|((c*M0YAlW1IbHpi8%KXmV+i6MH
zs))AX9Mt0*r|YzwW4XtbHZM26@9x3pW99d=Q;-0Pr%C2CQ$86=YWqX8xx+2i7r3C-
z-~TSda~s=)tRh>AAuZ*PjReh@(NY!;s=J)@BN`H@xB~`lJzR~^WJ)8E`kZyIEf2c0
z=qAhAV{P%<!&c0y5lq~RzP$O(XTxHnbc^<JUWr$GGf7d8>`2T;qGP9>r*j*FY6>|9
zK}GfSN$agQ&17hjdqlnBASC2EQET=kndK3a*50MLuvw1M-M34L>a(5xVzBczw}HC0
zmXjTf$_>KJS7Gf*ErXX<)qGD!WtR${kEVI%D>X@_n1mT=wU4!-do5%8UYdrg&D~Nv
zQWJaVrlf{f`{Z!3gCX^kLq)0Cb>|VX6y5M@+gvU$hd7ps{jZvXl|#^L3*4OaImnU4
zmh`5*W8_R(j}bwrl43kU<MbJsD=O&QWU*T~Zaq#(jew$1ni9wgb~=t9hY=i?^wk9-
zW+6o%Z)JY1UzG7*Tg9f0bf78GIR}P;djPgVmPHHS=Fzg(o3HpSy*fwIpMEp%d@1%k
zYYXe|U=+e4PUl}8|6`TWIEBVN^+UXm{>W6YaiUI*^l7&U^jdK3`@oV6zd2L%VK7H_
zI5pPTB=3kKzd8K~JN(=w@H7Sb*<jRyud!wolDmzY&*3kvX>tOrIAyBIDMLGRs>LY>
z9w}Af*?Ds5Jt5N0(6L-Xk*!2s0=*UG5mJL3juu%kp)|1I-PKjx{5JWq56v<CW=2TD
zRd8)n%U~2&H4!8^I>01bnt;dSg{HuX!2)?bzP-{hNR$THY)PJMKh3@H#lU6U7GD_y
zw?9vlbr~x>dH;?^U1b8wKdm8A2N&eC*c+xu=Qg+L%&-@yte3^sJOl})=5k!UXxo~m
z5$ti1J35snq)IP%CN{Wl^>Ft`I_}ZtzwR7(?ok!)679;fMzdl6))6F2YS*dqXRQ$q
zA3Ozq@OSC?2055+%tIWNX5Bx$Y5aJ?zp4BuV<2*RsDBL49z>sG{FQ59%Ez6oK5?88
z;=$-he?nu)>pBK)kXbvW7&FpJIBc1RIzlYbS3}D=&G)F+6}5aaaud?@$&x~uONXC!
z*K^J;HOyNgDHpk4L%g#?K8eo3^%iug#&RrAPrjqh>QW|A(bnXMF<J8%$56Y5-YkPj
z^Us`WH=54rJo~m<@fYi}?&~Hx2p1#E(e*=C%_=RY3f(Hahk4pH1GA8s6GQ+UYJ1h<
zw!v2*)FowZpcyN7e2@$C5N`8r>VW=$2bfNQZMIbpkE!<L#^g3HoRj@N(5$V~mI(a=
zHIyZmkwrwBehYn)U@IgX(lINsRlAa?l4se!?K8R$Zj)oB?!@hY=0>X%TYQsWZ=xG(
zw{*#;g&`+j>DganZmAf?q*(*-wU!QveaNNmn-h1Nl*V{5GMt<9@aF+#Gs+1LUvD`;
zQpC+0>J3H^?K;QT18xahkux9BpFY-6gT`U2Q#uJ|tCP44B^#!vG(K&0WjofsbaR1B
zq1=m@#tiAQlS!R<;q?oj52s=tNHDk0v|J`23892(P={7UR*AN!YG5C<X6^P<)^aEf
z@&<F*vp*UZ3u|{ANndYQbGDsxDo12vdEBj4PRQX3aQVqj`We$l;vp_k)w(n;yEFlm
zCUY6)fjZY^h{btg92g=<7{;A7wD2u_Th)JznHn31h=h0g)r5{-j87GL%|c{PUIiXi
zN4UFr{j~_YBBmiAjDr4O<$WR|(s;q6kX?(2YQOsZ_>v>3iLtuBt1fr|xK9}D$k+b-
zS8@`bSQsiIAvNQlbx$e{akrdl)%x?D*dsmJtV@CXLf~Lhl9l#EVXM(g2|ek9MWB{(
zsu}6c`_CdfO%?NXTdy1*o}177m##!!M#;bXcWaV5XLXa(m4##!XzOV-ep|<#?VTE7
zqWv>!i&Ordk3!Yi-x1jNPxHc*gyZNg?{@-mXa71rA$*kv*B;(TPWLd#+sD?zvt*iG
z<_CGU9LYGyyqbLoeacoYo%^>_R`a6SRjEhk>#h&sza4HL?jm(5WEw{8`lNayjr|C>
z)1#Jks4MEwXzKkGmF}l2ImHGe>EDS*b0fZ{(8f|m3fWq@aZ%kT`bJMJb0<Xw7cmta
zMYzxSi083KKxyh{PxcdqmmcyWCAYF3hHtj-F28UnA5^d0xJdHE(<AN_>EANrdL%vh
z5w<2)^$po6?4JeHz4zTk9=k>E++O0E-6C&2J;Qiq(^t;)Tq~7Cdi6C61`X5OiY=ro
z&8UdruktG*oG5u%J?=tL6Mr`E4BfcCw>)eKHPZ&TiZ71`e6uQ9XH1ktJE#d#;9a{L
zDhzu}N-<0-da2#Y{*;d#6wZuqCwhpd*GJdu*vIU(uD{2k=V+vbIaQTXp=a(y^0o|1
z&q1!>)1jchRvSKsEs-3gVt)nSOguJtrc`~a4LRS=ZrW}m2t$nuw3z6j@?xaLCf9J&
z=evKhR*T_P{DZ*`x}DaT%RIDFdg3{bMU=>mikr6(85kcp6_Nb-=RjtLvOEp0S6=!w
z0id&CerLZ*13<cVTE_=aKWjc85qMKRtl}}Bru^!4>s<m|Eo|sfy~1eYZ@-viX}n^T
zwX~$ZaCU*QV3Hr>*sX3;=)X1#GWNp9^r}Opl>gGgW@d=1wbH3fzZ76{e$FS@nMpE2
z>NnN!Y(Y6l2quxSXK*GaY}xq`d=F6-JtePg&OyGGGK27j=BPi5u~&UJtJ=kM^hf(M
zBc~-5y^lCo3@fGPMoRT!QsGNFo{|tRFT|XY_n;99ges9x25!_Pbip~4Ek6kt!^}}9
z)3vdX70l}k@RIjJ{~}$YK%^Xfn^TfR;ROKF8O$m}(O5gREGnTOY1%lh*hI3mw<=R*
z)#Y#@0g2ty;NBX`bDt3u3$hsr5;n;;a_r}QOZcXPxq8~-zd`e`sT7Rfi_OCMRx<)#
zA2~Z^Rd&OihV}ttcO*ic1L3;x?oWsSB>oH5l%>WbnEP}wPT~F>@I%?-r|(nTMt1r&
zmYK>MKG*j1-2RVN364tD5k_Z+z%4lzuXjppcI3JC+e+_OkQZyK>>Dq18UfUl8e0Lh
z6!|w{7VPRj+vt4ENB^^P?=dHH{w-?jXGK+y(AEBK;z#D8*!d@T@~c26EBmLz$N*Pr
zD({U~`1V!pVH`Uy9WKoplG(fcLY<_+-k;ihru^{CoOZPKiInSG>x7T@r+FAjH3(@y
zJ7OssK?T?9?d(w3(kFgL^R@=2-L-E!XD&PIn!MJf8}?RELKldyDCE{HTxM2quH^rM
z3i69?1kE0nO;1K#y?-hSc|yy?G_#0@1cpi<zf9pua+4HMg&1u}CjQc^`u%u|xg*Ev
zci}o;aR@K1(UtnIUlitVPVDpl1hqysY-De|d)Y8E@kDFoD^&7F2{=!&s%$^BksLj0
zL+EW!ST4Q%41NFgl`&&fXOY3Q5YNpBOqckM?FG=G_2h{?Evc^Ee~8yK&awU(YldY;
zVQnc?$_qQAY)?SBO9VfVM*DJR5<mDmkwP?n`<lk5wl^#)B%(SAAEN0jOQK0lcsJY_
zu?_6|Stq6E)@WiO=LYtDv~D|#6AeU?khgzrLi24bdwud<$%GghrrffE=HQztLg^Ik
zH>LS*vp0F3GA~lno^sSpoQi(bM(>mcQcSvV(IU^g7RlZDOG@*{UK%0=WkwuVFI&bu
zj@Jh486*(e@PvTW;s98ODZ9J=bns8_D)K(@4CP0-89QR)AC&TZGe~}~sPD)L|Jf)z
z5Jn4RlzrH~Ul?tw9_*&y;!DAZt$0Pc>(Bxtt#_p@r|WuLR`%!n3ta$@vF~$JtXok!
zsq9WBJ(VlyC}J&1{1Ze`Od+)x{J86C6vyz04aV96o9kgF*bF1TgUe0Lmat3D!_%=q
zHCA~}K>QbNFmV@sLsvCOmE|ZBpVK!Po>o`c&k(PlK(@<}|EF@7(D5m8siGA9X^t}I
zjF!r5w=xw3p<0k^ykFDf6|tWyX8h8(I>P2iGoi%NExipjwBuehMm@>c8T{$IXP)sz
zA3je+q_mB^W`LJQSIehm6~cpImM^^B{bD++d={obrq+wf&3+CB2c1iJ!fWy@PWB^n
zq#RbcP{E%`%^g9P6vKw41k2HF3hC3p2SaN?N_dD~*vrtO?&<Zzd%;oax{Kr#5rN(7
zRFK4S^_o^ps&hk`kMYfAkkWF$$N;9cRGMI0f{VcI6$J!zT8y^Yd9iFf;C9DzkoMWn
z4qhsbGJRRsfL(yP5{y5R5D}0;Q_dCkV1;n8rYN54;?s?tvsWNNJpo<Q4n&dQ5}-`w
zCigaciU@^~l&L%8Gp@f)a8TtnY-t#(vVc`$DR^7YwR^MfLM#;yu3L>nx}aBv9G{|w
zdS;PAoC2!YK<QXt>aHk`P<NTA&hqCM+Zwl(3<dXEQZPh1Nv9Fs#)OokoVr>M_v^Ze
z&!-{xMCPlTe4mEmk3X?^r*8+<d^_yP3)R1>07~}L-I%p&t;r`<Sv$3P$_sQKn5k&z
z1Ub8=?)Swiu$TMqM+}78UC=xx4{H@bm;q2C754L4F$~M+lsfkCHzQNYZVPROCxv3t
zv8O?DATe@&Ia$R)hlZ1j`?u(yFa3^(IAS-fv?3zhReiRt;)YvxwwD^3;;v4(cK-Ut
zwt-$&zx>7f>P^v$zV}?mWAIm-vW)zOVQyfUw{G6zO(x!BJCHu4aCa(LA)#oV1ooJl
zQ?Rv%Q>?EcP9{{5%)Gig@7Kyq2GmO;%<H+e*R{0fQlKtvezs<(bhr~?iyFdg_LfKJ
zo&d*IK=k>C9t(Gz=&QeC(1$~+F2~W7x)r3it^p<+)`#1~KHaphg$>RO{acKlVT}}H
zs>Zx5ZEHm*yW^}!A_jDw?;W8!$~7)|xR9|;+1a}MQ1@Rltbj?So1S)&sb}?hwG?{D
z>Sp~*DAtu6!ALzM(pw&=?3J}4RcrhGXHFac)>++Z>JNT6Zt*v6vY`gy<3@5!T?3S3
z#r}d8%hU*^pYXIqvi|C2_a^wLLSs{3Mn3N&5w&BfB$%T{JdlV8+2>lc<u(gGe>I{u
z2WbHDhe`2l_8pDB2b@e^{NM4XO=`x}zrCG$8CS~Zcgh4L=uhdqlrEImg(SMBZ8WLU
z=Y5$ErKIH97-f#`TALpSElE>5=22_J5FX?lH~;gQTXF2d<hees!OhU4qPF!PCb$}{
zq;|e)_>FG;OYf%u<i(d`5NkX4{L<1Fme?|QJB=>G?N|mgz>RH1Cuntgg-QGeLo@Ln
zln@Ufim@l6u!u1*w}UaLsIiE};BykSmSo9YznnzSQ>w82$4IlT=YR}ec&bghOg=_8
zma0_<wN=QYaouv5!~z96CFwwK0X@G5VZ}zyhm>KA=*AK>y3M+m{VdVND{F4Yv}h$|
z^mFX=<#=)VgdFG@1OV#QSA&w>BhGQjS7QOU`UQnWv}oe5Fx_z3vYrCn2i9d0ueojI
z=Z}q^Fd>Is2C22VnMDr1OE2n%zbNa3s{5_nBodD8*Ziyz^MXtNrC4A>Ng*MC$ZdT8
zQf<yY<=Z&fd{Bl{t1F!6hbMx2I5qgk229Fe5F1_cRdqpJj_c;xn=v)3@T<{S6Axx*
zTN_v&od@6lhXqiz3K)%-4chr0IeRs&ugmbA%6^{fq7~_82kYYEwn8R-H3;rPM0(xE
zw%k%qQ^`^`=4N$f*ivbP7j`p4a<1GEL+F^T=+8x~H$Ni!>>+!1A)gMaLp{TB%zQPX
z?P$w#2j&4s&6lIoWu8TJ_LmOvbJux)o6*ym?5HypY+Mg41T#eBo>4jdRIdM-XvIF1
zNS!-NnhSPF<v})kZ)E#x?&Tti+iX%Y=YyLkt=eT0>CG9ZsUl{6yGY3g@eI0fQ=983
zEZbUgj^F(L+&YssJoKQe7%T9H{d&G^qsR-+;AV}l*bru1<lPAOW0FQU#WE63Ft)F&
z|LAOzo#_<e(wOqj(>n!Eppi>nJ6pv7m4pCx;oYG3rotD;L9}_84BX+Ag-8umLaK2g
zb?_^R>0B=*Sd19EyJE}o>QlO1&LxTYLCCZp2Z`&oI9<GuZ=J7*gs8OF3?!q${Xldd
znSX2*rGjehtRAP<mxh7L!ah1drI)~UmUL7^g%3Y#NP%t!^1_cJ`!t+`bvX=g!`_(Y
zBvPr<Rl8TaCLmlllz&>GaR`3RMh5?~%epEqwtAh5cdI*CuMB+X=J?z<7*&`FQUZO}
z>kCUQ+J^d?aDvesDIM_~A5MNTuo#)?`gEnNk&XX6qPhk&Wo*qTvF=A8h3|RG%OruE
zxA%3HdIZAG#;l^a3a#bsI|JPFOEUMe``5I_#4vW5S2K9Hmd%ftuSsWkj~)(ojEd2I
za96&~eV&mmf7xC?Q?;%&@dPv!4x@<wnqE~u$DYbB(Z%hXD>Z5&aRx>YShK^+$7`GK
z#Ma=q6q94hNtrC%HP7dx#aOSOBfO0cCucQhF7+aAh%CiC+z{9zeSB5x5OVm<<KwPr
z&sRr7QqWdtv#X}ESL_8EqP`zx*2p1ay9uz(d~v`AOFcH<)VOgXB3MhOxJ}BV@=}?&
zg-=6e0>H#bf7heUnZo8UC8ev4>#LV%%Dwwx!WIU)y4OC3W^>UpVOknpJSavt>vLyN
z*EVJGgCCej-j)@c7fIK;RATPw`mg8u#m1OLAw0Cx0^yg(NVK292$K4>@|FB?jCJ9w
zqh)8rc9^}!v7V@@5VgHpb@X@_H^k_a=jjRYF+tel!cy{t$~V$~)PM3sO3`$x^U$q7
z8X8)2tNTj{0&zJz4gJ1USMv3!)rFE4BFBN}i&N8pm(9Up2U5rty2+x6q^kCQz4CMg
zSMpWzno29HluLva&0EdiXEB&M$)D}F&=FF(UvOS|G{s#IZ&9q>Nt)+laLnYSOvs@%
zxbb!+f#bS51F{gpk@#I5DdR1Gixs(MZ)lV}(bHIao3afLA7M_cwJdz9wQ8x|Hbbx*
zj>D-iFIbeCAf~<;#7|(I;_7C9UA$N>(@jZ-2nVgK$;&q;mg-F1Ybl&t^`P;x-uIEf
zwc6x{C)CH)S{U+Yy+Xq}8mt9Y$ncMa$j*o!H}pG>)UlM<BQ~$qoyOJDy4-L>HW`G@
z*n0~?Fi8=IbSMim=B2y0g^G!~Eez*5#LuMrmr#fqG`jqIki%p&i4JM41h&%fc*kG@
zitaUv#9#hRff?KkG2-lrj=_6{!5ZlMZRIZAneh>%iiu@<(WwH5oTOVCE02gj7`3-Z
zMee@05p(DmFbh=49qwkZ$n{Hbbg%(~D(+;oHfe)6rq5WSB$(2e1$bC1Y3_rbO+@iR
z<;`{_Qb@?&kFK6sfYyJfqMOG@W%f0cLfV*)hs^Rm?r-cq4V!`i1Mv<iVP7^<v|eh<
zK$_WQN#1m6n{L37$0V)}?j#~vA|z$V-XHyLI)sBIDI#TcuLBHLaLuqzkqFLM%<v<y
z;4gLMVcsajmTcJ=q;JSKj(?b-<aW>3P0GDO;3@XE2R5N*bNZ)NEwf`Lsntui(oqo+
z&OAI^E^}`8YmnB_OEg$QV_^2cBPHtBpDLSmYaVe1rtD&5uAkjy-fw&*BDi`71mRlb
zF@G#1V1Z&%6AUtWT(Q)+uNC*&?k4Om=egCL_ZdB?v?80@k(}n<)}nR7Y=oNzqqUWG
zxKXb%;oQi*g`{i*#d6^gc<}c=7D?&`$;@(enn|X%MDepZLrI_7SkHzG`DenYb?oGG
z8CEeY2W4OFv?*&TVJ}D%9kQCkxZ}!3CO?F=X0xQY^&>(iALPEAU>Fyj^iSa~ULqf(
zZ{W0pPZQmVl}(yr<k+x|C0)U3lhBsFigUIE9o14L^17l@RV~0~n>?@evy;?%Rxe9s
z6mRW5rs4Vfm!p<naD*)r{vwV3n(cSu5U(4qUDA^W9;jlmVgg6#yJQA}N34=9jFVL=
z5=r9fz4I%q1Qvk`miK&3AuOi2F;r^CP)jdXT&3A#&h2FADihDB&fh)gfY3VzP6%iA
zah`mGy?6}68y<);+Oz1K3_Vh9_vHkQv*YTtAE<lA)L}3SLVdnx4_-dl<=rJ2I*f!S
z*uqj;Hdz?)`>cAO>*+IzPHuMb81X6ai_J0Z1pC!S+zH}&Z0YH_^G8~w^C53$^SIQ3
z>}`mcr(JH(>l2hs`wEm`hsRVwhg8S7#nGP!9uuAByXwcFd&S<5-0UR|qoS~$Enj34
zj=1Ve<k8u8VCbj3SYmj}P1hy!Xf1+pkN37S70or(Q%q8Ni)-yUpt7J=v}gPsb}aqv
zwBjO@IL~t$jXfzC1Q@GdM!2eM;XrSU{?yJa?Fc8kTQR9B*njIcAAYJo7Z;uy%#^fj
zpdBw*Q$n+^w-@ol8jc|y@dS14Eg=TudC^kxsd>!2r8)!#RuU!9;|S{+6*?mW3U7!n
ztWBEs=fNHO(SEEkGOqhqjxXoGpZJ~7JMi~6T9vh8qBoD1i%Gi;$~?fidCs$mMTDJ(
zd!yUCBH_fX8Ne5=S7FkXO^&BQBU>N_d)bnUNqm3DXoWA$i&Sj0s;pc5;Sh{^Um1@V
z4UP0sCFiBvsSsLl1EJqfFVcjgdog(}JIDUv_>vPu)x}FfmzpU<HhHQm9_~g*SP<+7
zN0N<t`&qBreR*+zh$N5_9HWU}=@Kdg`8C&m9<&Lfxg`(d`64bLNB1|*-lv(kdos(F
zQ@$#yIsH~SO<`7X_BmC8zqmS0Unup_&`}zcs1<t`YiM*%HRCwRyod6^WFI6Z^y=Hk
zZ>DY$w^ck}zEPZt2@J!?pIj)r>&XkNxM8EssNRe?y<HS!r*`yIur!DETV#qeKYwRj
zy3ECH68p8b$Q8F)wK=ky8effOu>dBaz9{cPtbjn1#>MAepDSKY+G&J0W52Td;86>A
zOKCn}#`h0BBeXh_*y>{7d~O9c@Z)oZ{Y5TQEf%io5ZN~eLF%kOvaganA79Qgt1P^o
zTy0=?cPwfDTo;RoE|=Vze>G%yY(6)XT`qBx4yP$6aeR@K(<D13O(ibc+xR6_HHe^(
z$?`}FDx%ar{J}%IDVsgNE|3QoRf{yFx04nm1LN|Fk`;-c;vPWe0&#XLuj+0TKejNg
z@*Uk_?<UAOlDex9D<*Bm5<@3#d2h3w{K=0;jz&jMSGVC7am`9I$0ggz`Kvk4uN)%(
z7IOD9i3<lm6q#Wy?$OP|*Tba}AZ#NOoN<6olqC&CBs|_Vpr<!HmT*e_d=lxD82%o#
zD~>OgcC6iH4t^Thv7eU9y*F%R#eu4y8x!NA!iooVG1Bqo)VU0cpTlC;E0=!(Kssl%
z#8`lBanuKlH|!lM0c!k+40St$9iQP!lfiTc>98L;Y2z({g39c@b(}{X%-A;&4%)Fe
zmK_r-7>^IXsX@r2xZ}-=gR?9Gk8YJ;;@vHQ%h?SB>Jf*nuW@g>&HD+;iFh@eKlJ@-
zJN*$_hY3NpjeMI+db<?Ku6Uf=jKtR>xIgn*L}cLIsp7*HJ7%l|m6~OridHn`F3Yi%
z#nzy(`bdEgm_)l?!9({-%^vx_T!pQys7s48q7$ufEjw}fW%=tkOzHB;=Cz|n{(9Yw
zrrN<IsLrmiZ#Kwuz(9Y%NR~D;N7<;BQBUPT>nF0lwMEy|u?TEBS0dchRLr;<KxOur
zoFH&`RIFh4oTc-AFSP+9l%-NH>(fG{gVEjt4&`?#Y{#0D(z;npx)|brw^RQH&*%mc
zb3euzfsp3l`5U<?l3W#`TacMng=7@fM6QZydK<o#t)lTaIMhl1)S}3u`$jeg$2Ui*
zPNveGZ!9P0;O_=MIBdhBy){0CRR0i~X|v*w)~UjKG9tyCXU)G|IoHMg)h+MrH{g%$
zMTT4|?DOEZj1(K4yY*RBD6x+eUz{gaZqRC8g&DC4SbhKXW`wjm$5Q5aR<WOK;J#-S
zMz=`{?Z?SVPlpEtm9N(gh=~=98uu?Pd>QsrPPC{G18jiT><!Z%<<&H)rl?3*f}%T_
z62G9!7<oye+PVo<#FF58D&M#>@nfl3Z@PJaIm7=wDV<JF-9gjbn;P<RxvW70{yq(!
zSmmPlHeI(tnWAQlDBJQAN83!jjBC};6C^<aIi;7L4|(6OonA2nVgFQZ`gkU=y3KiJ
zE!U(=%8eX&*SM&RNLPlf4js4?8)r|Ow5^%Ed6$#sPQ<2J+3NUmvgJ!CxXsgF2Hr~j
zCMi{BIr~NKiE^;ypY>V$1>rSJj}-Pzi6ePcw#SpIUFnk-xM-b!+kTjno(-n4%#o;4
z`Z%IoO<rsLpF07|*k}9-M$e8vTN)nJ`86-wAV+NYalLF`|8W4rlG4OU(v`mN*vK7^
zq@sfyHApW;em)T%;Ylw17V}nX#Zucao^%;k=e(`WT`%Qbwnpx8(?yw`He1WXr+6tr
zHA;CS5yvM^F%8e?`q8~S#QqNbJOP&chhl{q0{gf+r=c1ckr0tnm&66G%dr;YPKD}_
zpOZfobE$S$dVJg*B+yBXP}Q|#|1K^5h65#Lp_9WagAL+Q-EFK%(Tz+LNp|XxP!0vf
zvrxcLC}TQxw=)AKyq}CtY`4%dpZ<Fe9T1_6(x%Bak>g2Z1JltTpS2IR%>lMG)xfA4
z|5tBOqA(4z5a2<p!~d0Hr1+#Zc(_=rwlG^xJdu=u=&SohpH9SiiX)g&j6_~#k_?!=
zsakHWfSq{A>;#bqVaNz?m7Jw$90VIBi6+O*w_((BF10d#)wn)#Mmi&+;ijSox)Gu!
zsg;=&kdtsuFpf*O$6N=|4@9u;)pzb2sTGiQU15<z24X6LuxJ|+Fa~Xf{5dVqrAU4T
z#scHW_GYq1;t?`i@g3tgFWxW2Js_zXAU4`$Pkuld39}~wv~Tq&OzdYpZ~ZZd<mdhp
zzF-|DM~FnbFh)Jg4T=i-!6d9~O@BKpuQ5{WC@rkbHFLPIInvbpS<T<if?t|MPAisr
zwUwRLRm*3?l6tWCx$7J>G}puRlkLR!N$qVa#snQt*CA(MgKDTJ7=+=<rP)b@`brs0
z`OY<Mk5j0n(y(o!5?qNp@Npih#FZ)j6a~kDe@&>nBN!(AI7gqcj*)-_8Zhp*Y0e$R
zt21L;Yl?MWuCSrAW1sYBULev1?9%Eevg}`K-V22uE?iVQLVR_O_&jiDJZ3(x$T>@;
zV^a}bCOOszPr<h9%t9+YfkHtx^)zlh+fKPI4ktLtK9A<9xluHNXIpOQYiW~^9l>=v
z<D$XoO<jg5!RF-R8E5^@(xEPz^jscd2dK^9Vx;<&4ctYR>2muU6@&d<Ia_Dr&T@KL
zO*IA<Zkvk92|ub`g)hvzdfY@a9(<10HhM7jw6QGOw^~<SP{s4twZc{Zr|L_U6fCSt
zv)OgL6B;T9RWA{)b92r4!wSwTZrp_tb|slg63PJ~7ahS|+N|HvO-9qB%k{dgzCLxL
zRJMpGa2CCx!0m*Fm4|32Hm9B*q&@-RE$R{Q@2C#}-QwRQEx=f;Hs_jJk{>ee@19-w
zqf}*&J_a1~-fx@yNtZ=%Ghy5Cw*X+V=z!A3W|ytq3g#fY*?IXBY8^hSE+PYJS;JVb
zR?-+>j^`iM1%!N;MFjQ~KoqA(*Q7sN-iFS(ECWU&HN!t1kl->}6}38%)?#rqQt!!Q
zaZmpTUC;3CfsI5pUC4(w%2zJMtMQtgQ~EmG5(&3_5(#t(-Q_maKBl4Y)?)lK8pof9
ziPRch`_(2CZ{FMf)D5-pWKQC!!ys<;M}jYIguBjc{oh3|t8+#WKtnynWaO6bN1Uu~
zz%<oun?I#n6GtK^=_4EXP^8N+T6N>!`i4BXB42*+M!nU~%$FVpZoHW`Gi=8xx~69&
zKXN^y#^~iSq2@_-H1g<(R2VD>z{$B{ZQg0z<ZBR&p}<n;JjfjYb1yt?C|J}BU2L&d
z)P6>ceMpM(X<OT+QsKg(Kbh-K|NPcsRo6Ou=1Lq@Ngxo;pf1I+kI>TuC6s7$bms~l
zwd_N1)l%*WPsZUCDGoE{`mu8VHtn`fE>b>N_S3M#N(kKQCiw-u%?@*yb1YYP0yWMc
z`CJFPSeN*W?fCMX+y3q`{`C5oapevPzSy}gmvJt|hcHB^6yzrI>_voP?PuD}9j;>n
zBmkvRdASi{c98P`2j)805!}k2>ni%(b+x<2U2xsRme|YH<1=btGbCm&F3BN{$*-qI
z+VM?YP&35oba`^@JK*JrCJG37Xl>j$X-qN;0Q;yDa@qWp!mi_7oc*hA<I#Ai@5j#K
z<O>nf-IsIU+33bUjGF>-H?_nM0F6Me9Nk`JZVY020XUcXFuPd0n%+lW?^XT$Jqw}L
z8c*_XRc6CBTI8C-yi;0<!zJ7C3O4N7<`OA-dhi2=An9@ZMfw+Npjj_|N@O<{CcT=9
zh&<{W$##{cQu#Pc$G6yANl#$w&q=+ZG8~~wH=4lRSLez_=~Zx(343-MQkN_?065OO
zuB>k3yw^W~`8-D=Atx}^;n4Pa#WhOCjbg^KyCJaEpRcPr?*sqY+U0NA*o1LCsN%jv
zJot5kM(NmTf(y|O`E-U@RV)1%w@z?`YR*^n&4_4bR%NEE!Y=)!bkH~g?<sDwk3r=Z
z<6y4q+rL~&taa^MG00rV&r@Ne2~@M=&YRMf*YrbnBIO#Aa62gxlN!t8kTbl+egx_T
zP5$_7wQG0eNPb3!^Jg+d1)FtCMia`1_sZx-Xh>$<vqz63{=J8B@lY_Iy?t*h%_XKN
zQbyk{PQVkgJGg(HPk>8@f?2{|!BcB25#?h}*)YS|W&?U>m#71^27^+N#WF3An<5K=
z#El9Xmt}n4D>5Gn32(bzRB}hY(1QPB#l#(UNAeT6vgstV1WDgpofp(ju<RnF8+lhS
zyZG3d?#sjgzRIcfTXkywKBhgq&|@TQdcKW6)X6ArimZ1B`!(72h?m4wmt>I!);*(d
zQ)IL1h6>A(?zJuiH}qB93eOueh!Uq?nu$4y6RKCS&OvB{sR_vrp9;?T3=c$*ohn#-
zHXusGxLqMMq_<Oy7*GQ{tb}W&z`XsYZ8GBWaRbK42rwL3ZRW9(6j$r~HtdH<Ht;FE
znV3g%g5jgO<@aA76S|q$@nHuoXpQFXx*{Rn-Wc0Jc7cMv1ZJfL*O1f-V*EvHMpRPw
zSOaY|+c$Q9i_m`>3mr*)bVev7$@rVb+q_?paiwqk!bh0PNFtFbFKDdS*!}Y9RQWP+
zw3U+f1je2(2x00rHB78pxvFsYS@U@SI=cEZ9=GzOvIVGx)36gl(3T9^0{Rk=4{0_a
za`Q}YkzqdhXR)@ZsuS1AR}Eo%6hbQAqy!0zva&kRJm@I2Y|=03XV3m|fHds2Uxd1%
zWhK*`P-$N+k-3Qm-|1r4%AY*FwtOdM;zvYgOk=O_<nOe9Uy?}}@q<N%%`EiC4<(Vh
zLcUkzge{|V;s+5M;T1B8L@QIaO=4K#=N$hWdiP|}cg_H5LXNBO=>)*L`&>4`^SE5E
z<+NJ!jnTtbUTrC^Id2$`U}bs?+@-@3F|+ac;U%12voF83>D+wz$KzTOYDmxgaJMpI
zfN4BLYbC^XJZZXY;^O_dbGMi^>6^wuWh&*0=EEfS5oT-SYX(?ZNIUzLs}5!-Ht(XP
zY<&>mO#K1A0|^On$HiWbVoWnN%MO>c5`>=)gNkz+(rr5_<|3aeq&4}lTncBTOgL+J
z%^)PaPQZC}4;Y%pb%lnb!C5x^5>NWed_t_)oK5+#?TdoLLnoRDHOEfJnlHRg=g@w~
zWV_s2`zjEkOH0MX_<1u+_@YZk(F=weMpZGB_2{AC#9)v}FtbFLVKeHhN3(1AqU)ic
zpb)+0_8GlJjPK8`!<+~E<@t=QKW@A3iLNVdY<kw$=hq{4A8loXvj8NS7PKD0i7vyB
z%6JvNv8A@??@dUEDk+YQ?Gqh%XAYCQ<YDWuH$!|6etqOaCOQ<S6232kxS{W2`wGJr
z+a$60!{VkJbp6}P9ufWZS11hh=zqUNrZ76ayo9HLg-%tb+Gk1w`v!IHw=2TO&w|wm
zY2>i1DXtjU<I5A{$@W&ouvJ~y7C(;aAClOi8<z3LNIl0@r8%?LVw}md&+TDk%gyBd
zqKZE{#OcDPG8jvwerF(NQ9XFzV=sfwLQA$DizG!THv^puy5EbgHm;RIcWoj0<^=O2
zfhbIWrC<dOwnxTI11B>O^%p89Y3WI&1SNXfea11_POm->Ie6Su7K;T+3Sw3&t=OX4
zTRewj?c61Nxp^sX{h8=t#b#y!J<3B-i9fb<rkOg0`<{0#gLuWcB@43lhZE$DB?;wX
z_SZKU_V%TYcB^-254>&g(m#9A9*YT#<|%8%mKEh_y*wMYHlwOi*m+gur@ZsJ5l)aE
zS2Hp~R^fp|bH_-&ckbt=d+BQ7Gvj=b;pQH`I^VjijIqgsRE>$Q<K65+0>>jxW!J(Z
z`I`%Kmd_#O%G%^yy2b_fqEh5T4^f(1eW3wwM?9l0T%iIV9^OW*Q6uj4SMIAu7|&m-
z%}T)>Ag(*otj3-@QA!1cjUR>z_8Y5q9k?=deFu#j2G8L5lDA{DF0klHr)!ZEBbIqn
zuW{oy<Xxkr$EbpFN)N*7`({kI5S+KELZf%M)(AFUA{YgZWK&e^anlt9$yE~Kuo`=0
zFcBP5-B<6KGn}tUQJHX1h;C$5EsNNZ>*zl#;{W=OmP~qYtna|QY62(NUyTX<i=hKk
zCX}acTPKg~|6EbmlH&f&o(x|d0{&T||L>RfdFj^@=?Tw4+F-kouK*6pplEYga@*I;
z0PI2~lD2HW|NhZlOMzGabZ5)y-wXKv`f>mD)`V-jz}^H0>20qLh-MhUKn>w#Mr?-t
z-(Pk-+#B1=k*Ns&?|t<@edwT;%l&4$IWPspecaC%^aTLi#^ZD_tYpM3vnhya{(oCc
z2<%$5t`}_o+e7}J9@vI3c##ttebN8v`~GilAnu4Rb^bLXOZD&Afx<BCjIIqH%mLs(
z1_G#za6d4K@$gEg{VO_<?_whzeeqlGJ}%}i${M~jehrx7T<HJ%Gr9-n*3YSY<FeUt
z{`DNYnSe_g*<VD%9q>kalUmLL858>qVJ#HKsYKczHy;iyqT;d>I?a2WF{EmKP{JM$
zZ2_PjQY&DlC6m3}6-mI?&;4xq?8Qw57$^$51r7{IycT)y^p4vc{AN?Q!^5iunXudJ
z{a6nqU-mYp@wz(H>rdoCi`WeMSFrql|EMSo+;AeoMeB{pXgk(_eM5TWZTBLe<l%ME
zx>*QZ9kdI}tRSIvI^dLS2dcN3Ls3VabMTA;y6C(_iUEGikwsKvZ2N%%wlaVbS-Twd
zu(Tunbs&m~+#1K<g2FA(13%s-3)t!lT&%>lgLH?sd;67&=#xT3Vaf#e0RNc4BIxj3
zTX)sq#UVD-bRImYnmki^`7Y-~J?&%!H*g48kqdwUs5az(YQ6@C8y;7~fn<x@GUzx>
zD#X;u{;xaPn~+=z`Rhm>oS6KxiP9Ktfbo66Gkn1gd9UTp@a+vKDg3R*hd3L6ZWGun
z%}GJnj38~*k(5oF<AJAV3@1IYh7Z^?ylyTI0O>EF+@LEEX)t<ql+c_ik~Hy31_Sk3
zqZcsh^TLgVDA{p&t;99)0~1%g(7Gh>NA18tofzDjeUCU6CfUmQ)b!|)3D<r`wu8)e
zdEIsn88rh!^cp^<th;)HERR!NmO>T*3{^JK6S3(guX^^st#ss8@&m0pFF))iT8@AH
zpqa1GwE+=PT&@YUIaomYow2b2`oj|HKY-m?J0!(obVT&J)&xj*aFJer=SY@Z<O+*I
zvZF5Ye*}TwRK>r!N7W!_ouNI?OAENi>j6<b-5dXAq$FeE55bZ{0WbJE_#SZ_*5K>x
zT<2W7{W*Dwl|Fwl=$z`$*``Bd@(Ayr$Nt}=(0{&EEFFl=5ZKzs?0=5=9^X3*SETB?
zJ`yZfmf#aEB7cqv|7M{wl11wZ{>ztVQH!I^flP~~37!0ZeS}{M(~c@UsL#s9q|P6g
zJf?Yo`hUKk?m3fqt`ET9u~~rmd*n_LeXs#YuLMZ$TRU$b``FlZo#_h03)y-~*TO<6
zHKM4JFP}2^{uS?uZpJbHi3ImvqH`->A4YmnN7W;XZ2(Bips-=TT?U{mTbGN~g!T-;
zH?;w+x&Q#7eZt-dAHC>nLlQdw;v}*aq*iRc|Apq_&q((#25o}&M>cBD^TmK3N&gh*
zIal1FJilDOns31H&v@}Cna;gifH7+Wri+{Yoj{xpLk6zR(RZ-z4qk=xO8`OKTfG1w
zf-XPV>=D;3prn$<zqjkDzVHJ6ry?ScvV4(VLw`kYWY~+kLE10mC+dS{C;Ve8?Dk^F
z23TL!7D$fL?}L?E>1DUa0=o!F2t{ZBYSy2eSw+5q*XOlGtC>5*5JekCHU_>{J>Y_B
ze-HV$Zs)x^*IRFIEW)aQ0dPRj>!!s(D%4%#EIdnalXN3`-99Wi7vm1->6WGo-`+)l
zX9J6v1q#;X3r6@FwS7Mk={pmKr^@r7h@4Y_!LtbR*e@I%Pr2@wyu-hj0c?#MgZ6L6
zGHzD%+)Rfy*zc{xh91by)*lZFw=YM+_Pl^AZ~iuxLw|mX)z;#7Q!b=74s8e?uw2Ck
zSBkNqQt3L%aNHI?vIQ-%k7)}u5xm2JhkJEkPB#EhBEOVnOnxc-=y7XUzNuz?3mlK}
zkg4z_nui{+`E(d6L{@Cb*mf;2XZ3W(B&Q5SWMjJNg#AYNOJ3FQVlI1tZ8kGpllud>
zZw9D%Y`t=~#I}qVOYlFt24_i>C#mP@^YeD5h480y8`r6}n6%uWUE?%aKRzZ4`fhFO
zo0<O(*xU$0-Qt$tX+O-G*S6a+eu1^Xas?20xF777m3v#sUF$aUamN(-lvonBDH*tn
zhy=iq+ztx22Z_&+cu6gTR_ZGS9fYbnu)A&J>7pJxx%P8Jb9$Kj$K9sv=;`ka&F@+#
zsyETum$x8x`2!*L+`K{K0{82^FZE)9UrWB^h~<`Ui67qP`B>uDU$IyAt14vY%FeD_
zPY*ffyBmFtUj~G)vy)*8qr}!-fXl<$J_AN}t0@F!yUvVLV;W~|oxOGbQrHf@ARMJ-
z=9>R|*Px<;=%M57{@K6h5R8b@f<DWFL*apaA4d)RpqVfz9YnnYu&aLKNL_>`p<ivt
z!io$X1`CorpP2{)yfhlxu6!&6<KAx|Ka>&@U9d-bp8+V=n76?+9pIjp#9{yo^fxSs
za`l#$v{qhbpvwl!9i{F4pV%m&a{nkSAHRcMWZkPGLSM-p>;)f@Z3n)ki#FX1K>}=S
znm6#YZDrAuaT|cDbem>q5Vvy!pST^Dt6?GVmJ2XC7m<izzCi(lx^ELIq!!3YZ}C}Q
zb&MW@QO3pmNZ8FesE!E)qCYp-?XHWR{jj;YSSLs{{PhxFL<UJvCWy+SFX2VW-qUg0
zeKItQL?W}i^jxNzw_&I6lMrZMerSoo`+!)l+XUF9tlkO37uv`?0KSNAk&OVb>cig|
z0YqMup5rg^?@3?oRH~eV^M;Ck)cq(n;~w2bw`_yfvBeD3kXH^C)etj2-Mqz$wM!62
zKRp8Z5)0C{nf9}o8)Tm?H_tN&6s9k{Hk7tfom#Ctw-8Cj>8AeAjD0=8amOM+mEli2
zZxAVpDug9(NgXo{#OkxPoVR&{h|D<#2yZ#oGdWLkX$B=fzNtOXsSmKMQ1<w>eaWKt
zO4>R{#f2qvo=n?ivFxb#vG7QuP(W^<KATYN*_{jKFuUg5NwMdiMH5)L*GL;b5pa8^
zEi%K0NVoy6Z0th*)yF;r?)2imFmN7LCh&Xg2OP-KQ9iXw-FAnQ7~n$hMX79kavnuG
zwz&|<|FgKjS%S>C1Qr7?$WjPpQ_4(Sr#~cI1PXM%oRUeL=Xs;wG!rt(v!-fMd%4Qx
zwP~-j#%y8N{`Fi{f#VlXZa)XT*kM;^13+Du?>6>7Qvu7WdeRn!$~N&bl3&l$4NNdX
zKS^t*gHS`+eDcAB@!+ia_K&4#f5sI046ghs2+fCrPgA}o>Nq|IV1g9u&(F`o>0oo~
zrJ%4En<ngXr<hI6)kd!~eY<fxf4%42L=pEx2&%gyl=fV!u6b<T(T?UckXS`zgbQh~
z+YamrCr+U|r&D|^%BdSG3>NRFM5m+hvZ~YYZYP^>zPoL)yAeu}TgXm@fAT*=n(1yj
zEoC25KP$+Y)s%;A=KyD|xcw|otJ>|a+LssL2-$%SIGy+nn#c-tDTjHs%6y8@Pwhdp
z0{8eQH3ZdeGN&SEd}|%hh*+En_o>3Rcg!h6muzZ%twJNf_H}PNvlvB}>Zj2UpbV~S
zGJUdEQYltJE@AS4NdRiqxj3aV-4agwGhK_h81@1`l{eE~W8UERpom$tG$~u1beF$4
z8aUx8Uu=Z_bHHZ>-075d8e?wl?(n__o<+}t?!rJW=*d!4mP2cIConG?!v!+qf@?_l
zzJ@zWi8%o2<ELD$cSnnP=RAn@1+!uiLzxG_yW^P`mj}AVwwuVNGb|t63mYc%P!?SM
z46(M9Hg&QuR%U0o(%bN1MUlWDFi=coW$Jhm{FwZJ=bmb3gD(0V1*a|jOym1q_+E1U
zgq-xK$Q*4CZx0ACwRN2C@Ju7c#_r{))>cu^JptbDBaL`<&b^*Xa{oti$u8G@A9)Mk
z;bRBF(PqwWF8ApUTCjX@pW&}iAGKE|iSR5of~Vy!ONq-!Z7}LTf-`!rSwp1fF0)4W
zckJuH8?=yzTxxw1AET`BMGnMu4t%K5H6ozmZkQ?}4!k4zB0KCVs&8)@8;e$TQSYLJ
zhwK0@C)Gg-cF8yvA{663c0Bos34u!~5K&I6zBIdB(TIxX#vX0nuvQEkvER6hm;$xN
z;)VaYyl^kB49X0atE`FeZ#0U^INRz$gtubfO2$q2NK8{ZDn_m;LeL|k!GJpr<f>RC
zS$}QjpnGb+W*5S3Q$1yqHpnzBzsJSqT__{S{phbH*>efTnh*?)r!SAd0XV!1_qw;)
zuuZ=)2H`F-#t`lD#SgS{U}8q_`g+urYPQiKNUcjYy1<2$a4~Shnj7L#f2kuvXN`5W
z=Fte+*SWHrul}9MlP_Uk0n1h#<c%Lwf<>7wPx{%%`<+x_tFhdJfmIV`VeV}>Ea#9G
z5TB~YXT6?V=Xz9_e4_Ca<g3Of-6U`4jc5GSzV)?q&#?PVx>#at*<xhCdr}St1kav+
zSdURF()jgZPlU)q62<;{zinbAcZM$U8x|?CHJg43SI*;lMux~Lnl-MlXR^sJbS!#1
z#9Ev!sHD_3)TS)Xfw_4(OQii~l4EZ4DCL(77(l1G<6|twyBgL=!rBM@8}Zt7xv0#N
zO2=e3ou?QRhfbDE%Vrx4guM^2h@@4yP}8a&!}+KrM=~}(I&ebyu*au(g`$h<nq!UO
zc6c3vP=~S3veF|@^M_NDWgnQc1v~0MB0e-vQKhouWHDv=(1N{dLdA>yekWUM5aBCk
zk;w4VDS8bz*85Q38b-B~O0lisC(<8^j@6qxgI8>8x4sr#r<!O^A>I+y+GKlsoCC{o
zj&I6ykEa$hKy`uHLt{q)HE5fz4D4_#CT6H3tQN5-pX$F@o?8z;@2bTRHfYW@6YPGP
z5=MR|S!l5AUQxj7Y2a~ZR%{zhTJ^0*Zq>*o^7bkAdSiKm9J*idopwV0)5JBobs2s~
zpxML?IbT^|+eciX9iXM(5I*FMlgMZ%K}oTA{_oiBqv!37hneF_wJlHp?PlTpz<*!8
z>kPgusfisdBeub2dSvCdWKtiJz56NBsFpG+f@nB9dLvsYCYypf8}AcVYU#Etjrz>q
zPm_9r918Yqifc^60lCMAm-DxtRe!=q+DHimx(JwFNu`uY%?i?4)~N93{PZ?&`3cb*
z!FC=AiyjGPKpQ+;4kD6i@X*+pHet}ri5|@++N6xk4qFjUnH1{PW1cX<r5+*W$|ky^
z99j;g(9c%-ltUCituWK96oL6Z*7F6O%!UFC7Z$EgY3cjhp(MgCiuFU5(r{C^+nbi%
zWb={`qdOr+?EUGK2aV5fi*l%;n~~LbNCy1$`Jb(MH@t-KDNAWyg;bMo?B*4VVg8<K
zCrCM$Ili-g-Q69d#5QFJ1L>0ANLyrGlSy6Xg16GJ=nzzbOF+}T@VQ4WNl!Bc>gZ4>
z%eDJbM@TvguIsEo`mJhEQ*Xiaz-0kvqb24euE&n%{*@>i*<Ri#9t{sj(LlB=OXE>x
z<}uagyN$!N9v|Nl+&<EsHFD1_Z_-f3VYb@xsc9~r_jii^3i%6XzG7)6ZG>B|TywN@
zrPP1+D=juR!3QzR(50KMRK8nI;|MdUg3c#0dy&;BIUi8M3rQ_2HLHG}$4`cL#8wlr
z%}8qT`}z+_w0wfCG4-n@RnB%+#7jRdJ&riTq_QBiw4GkRoRYGtT?{l~ZZiq>8G&NN
z<+r{cg<}5%dA87ASv$U%ioS?3?V7@VPdaz$rq~2jg(=<oDE*diNXavtqsqvdhG#wr
zD|s@}B;{>ms*L-_4~wZH3lvdG3X>OO8yAc1D>Ab?WC@@|LNe*=dbGz)iA*MC?gdw~
z&Tn!kP3nEE7^oNtmYQ=ye8B2~P06WQJsY5#+&9B^m?=rb&2gT5XZ|giroG|N?RpmT
zR|MY+wKM0dOioTI0?Ne>%i}PNush?YR!qki^@FLTQnUOE`{JQ@NW%STT}$;++5?B5
zhuQ3&c3#!9T<1rAub9-s>M-gM7W@-?%P(p5mA%7rLYAg3|HGQOz;r1OqXtz8qR0&P
zG>A0X)}e{R$SY;iu%<Lljp5WO6o1*NE|jy5mC|(ZlHB7#WBYmiSb0+Pat_g|dw<VV
zntNHNVYJcj&Tb(E<TiGxnZZ1$1!c-b-Dk_e14o7<YbsTDQB+Mfk8g@wg^{1Dqp!22
z>5m2PKrP`|j%e{4uq~{{&t$U>`{o5<(;7N__Y3CRS+;eGzw(yog-bJ{_Y&$lC;gYj
zfC(%FYBL4Y<S`v3R0HmyO`}Rupq=pQ*?IF&Es+FHdCJkKOL7}53@$Xwao;fr=H7Q#
zDzNNoAkI%DueuMW!HL(Lq^jSTI(+0Lx^ao$g?@@6ot6J(C;mt;8?$`GsV+`z&}j%^
zSg7OQT(l}j0o$uxPE9mfEcIBD)f~K=%hjFoxn8jdURt#3?T>MB1PmgcGe~f(CgHFV
zAG|q<F4<ircJC(+p~AByvo*cHfIG34RSZww39*uY`CBtjtYS!co{cIHrMrsohLuhr
zriS;+Go973y|wFC33HtSZQ1J6OW0GHXU!+ytXC^FXZM8=CchB6w%*c@XTR^AWN=Z3
zcCDQ}oY}iD5neqm!1Zz-tLDiZ0{sOmR`U;vLH;2;bTM{u0sU?Fw*OQ14L?OseD)5z
z)MReWU1DBDkw2U*u|2s<j4p`x)}XcNSFLwH?_2Ku6Uz`EkH=H=5BsO$%3}lmgiBLJ
z3pDpYQqIr6nZ&B}LOT?o{HvCC*hwU_{iDz+N)^eO^jUrY3tT3D%=8c19m0^SXy>&*
z7fM>w>q%IA8ADEWo6K(eHVFK{{?;PGn?&MRnK-;3j=4OkHv11tNiP^HY%6~meR;Dv
z+T2eSIn-FkU>cW~)Sdmoex~B`qo~!~_@nP*+8M&5``cAoe?V7Gj4W_LChb{z*1Ypr
ze9_iT-8QN)Nj1eB-4v9bubSyfuk@#k-kfg=^qzhK8XdO31-{io!kd0R{jUn|(iQw+
z$9uny>fW$is?j-aC%^enXjh_RAa%t5{zUDUC%yzhj8;-v652Z1)SEtkTY+UKT-iT{
zk;YdEQ1_~L!VH#BXB&=dBGE~+_wV8p9(8?<N8^R=YjpN%1QD~T6tlehOyZmw9kB}A
zE+iP0dRx)j$@uhQ=8mCggd-!;)1vfJV>~<Gu58YTmRwY;^Ur;T7lGcAh?!so#((wH
zbOq?W9~(>lsK(t)>DBKPcW$-BlqLvBj8NHA;d8*Ea;eumuu2aP^<u&a03e6qKc8yq
zCP}FpimMfdpUqX(RVUax>W1LjrnvJ5L@8Y`?65LKV~5IhcNE)j=@Is~FMAI^pEe(L
zp7mJQW5U)HHtJ7ws6ADApT9hs#{Cm%i!|#V?NhW$|C@)w0B5ns_%Va^%1D?+0E6M|
zFDz5P`Il?TqI9D*W`e3sII;5ND?p&c)DL<(fL`|dgK!qaqHTB4cLVL)R{7+OG;LA5
zUzxUyAzpP9iG{J=qc>v(N>U;#2^`Gv6S8AUTGaF_cNnCxRTR6U-k`3*pGlg3kF{z~
zkI{}p$XzgqpfCF4d^OL74sm5+VdqH;^HjXaN|!sE%qmrVw^=!MrjqtC<%OL;3^kIM
zZ9C_~q_bN?gxEV-55ja`Zqtn__geloA$MToaLP^KU%DYxXFA*9X5nnRaQ&S1(;LY`
z#v{3nsD?QU$Be;$`s$#|8G6V5Wn06HQIIa}0_vj6z@<CWt7?>oiCuT;`uI+7`;EoM
z1PZv&i0Fac6Y3F9z3Vc$%h9dpbhg<P=r4I(y&)+YCN=kSJ+{#d<}ic~d=C#>Kc*Xf
zx*r-aq>JA<^}Aao9Q0JeEjA3gI$LA7WO@(E^w$EuY+h`V9=}*Ve$a-fPX7NGJL`a^
z|NY;SGNgv+7}5+0sS%FuRsjJK5b2f_snIb?YJf;dBVf=7NHdVG0ZM~(3)1cV@ckao
zIrrRi@BRJlnA>OXe7&B}$Fm}TBW#0xZUnUSz(&4%DsqIMy$}{{v@7-OXLbV>Z`!47
zx~cKnliuL`Frkj0_Y2!Tn8kr5qTEvqKRiFJZTbyz9WE}tXIiJat%1SW25!(TAb*Be
z4Tg>zgBM#F*e&EK!M4!yaTG-<Y}b#;^+k$F&pdr7eJcpAmh|E`khi*#9aFZF7~wGH
zCeC=cWz-ffz<_w;oYAnq0_+{rwfWROVhzgXZGkE}TPw@B->JAs_S9xE-~E;nbx-wP
z{22Z{%Xd1DTuEPHE?=`lik`0K`aX1GSjK$#Sez=o)r57pO0Eb-?(cmTI(08#4j`f0
zoHL<c%JQx1J_61%Yqi%J9<#cQ5Ur(JH<o^TVn7HZK6`p&LoEFj<)w2!q)>Q<8Oa}1
z8g^p&u>U2iTowqp{=IsH<^NL)mInu_6H{|;$bW^7ap7K@4+tp(TooNL^l7hjPQL(k
zgF606_tw`&!y6G|r55ueuXe2KihY8#6pjT4)J=!-R2{k?uVW*wQw0L^c8nO+@aD_3
zZ^VwZ_(0IsT|192nI%tSD&cbDy$d=$)^Z~iJDl$bD%=lb=PfOAJ}h-wO}YDsGJluT
zJPx`ZDnpV&!_Ddfr#_bg3I++Jq2t~sGj_k{XMp{s-HxqqD_jXm#LRZkI(PvTAxEM^
zz;ZpLG_R`-)<SP44i#%L<^{?oH(DA!#v}1um6^b;ZLL2;*412|ZiI2qm8dt6%~3RT
z;F+boKERf5c#9f+j{W}YvbmY8E5*HeJ~1S4FDKwCL}**dV~^ax2734IOJRpIpe~h7
zTQo?$1?huN4Z8z%z!ei_pH~b;=Km~f%i!Wg#a*fVb^r|fNhr}WL2nt9q_}#<%k&C-
znogm2Jl))WyPY^3DRbz-pZWIeDrl-#Uf)DE>gePu<gC}uOPe>@AHJO-@*;-{(;HcB
z+&UHSAs#urlIh<_^;@M*rFTOW?gYnIFEh=$C5tBG<$f*@)7cGdXH?}DH9#!Y-9_pP
zP6~p|Y}Vcl8fgLPyH6`&C6J6BVeMq$ub60i?!j`IuA0n{*7{Non+#(g7@c{E`a4>T
zah(t|BFLX5ysFMmBy-@#B{!a8GTc-SPNCBnV4q!z?~tfygm7M7x~i$0P*EPBqRoIV
z8X4q-*<T;-BxSk~EzHZR`$R4HmxLIhs&D9dE_HIxur()9<S^P90#u*!&n+pks0O)y
zV0=)BfeRD!jOHL@&ad+1u<lKVkXh`5N)J~iM1#9~2u9&@+r+Clr01tCZxFe_fMs#;
z_}KmQN~-16eQm^h{Soq6`&UKR8<gu*5D8D*!Rw*X6RZCFrAVS(vOfFgc~af>vWwBK
zwy~=y)^QIRF@&g;H;=O6HoTJyY&pjCFzuSM{oNln8MeMuVrk>xa0Gg+0<|g-^`Y{k
zBeDO*(1oR?oyXgwijX?jx1LU>6anE1#<q&bxrV!qrsg?FCN3Y+MO#-Z8r$@=YiT>9
zC8w)TM&XUEaTp+!0?aK@xN#f!ESmd>m@U#e)0C`@vRro4hi?!spMoAU&Ze*nZBv7r
zUxjbEMNxWx)BUuU%s%(p_jenlQ{*0HL+_b|cA5pLl5};a5dP1iGv(@w`suwFcr8{G
z2pNOQL`={4>j$(l1-I7tcjkS&LtoIb+$6S`1oVE)dLR>aq8vf?>JiPDCrGQlS1%|Q
zc-yraRM-z>Yk}TZbgciXQ&d<jrTy#=itLhy&Q*ahobOV=Y5>NiCH>#7e=@?%w>Z_m
z!GqjnhwQ>1JWHm!J*Xl_Z%If!T{M;M@mi%z(al6vW%fz7d*9$%Hbd0T=xY*KfudlR
z*XlFD5AOT6?Zx@7>k=j?$^pKEk@t^z!e}%vQmU9n4&zWUr7cdjB7^P+E%+cpbA`|O
z);c`OQnG1z%5q|~{GIX!k2+AhcOG7IJb1xp%+_$oH}vr-`QFY7bTY?8|L}H)19PFJ
zuMqF%TqMncq=R7Pk+a7M&ooL>?fv8T6tF7AJqc)Z^DPJ5I}`JfVKRcbdX*usZLS5_
zT$AFni_1_!7RnpHs~oZ}*v=F?K$x2Mf>X*L9qWHrCgCe~F1^oK^iired^LO0p4usw
z^{VwM6s#d6Z@8exZVc>*t1#F1Ja66hf6jTyg{%DQ%G?izvplL)<9Y#9*dhQ3uBT9t
zawTE6O{6mx;+GSrJP-5)bW)+)hA;vpSdu67{vMbiT4rtmAI6O2;DQYu-$N}f=#A6G
z41QV+{BmzWebHd{^i8ZVNLa_+xce<~@Rtvo`m&QGJHUW=t}3c>1t}b4E)@IS&j~@0
z$>XkUFm|v2l!}cP#b7TKV0=~gy;L&HpQ)-C!wgTMO&kf!pJzLadntNQ$Pbj>IS$8M
zb$M%^!py#Dj{WrT-GY_4vRM^sgZhK=)anm5lpHxcIcLwN>!*}YAr5c+bZ%a!s3Y(V
z^oD9vEejmPZWd5VJlH6|e|sz>Id_-)G=}Gy^71XE!|+I+D!s)oN;^AJHO+V=H8ad~
zj`~nB*tA#8a~PplxJEb7q~18IkN)zBXXflBXm~VYFb3;8^&rEGjKYB+HPp8i^d?=R
zcI`Iol_D%wVXt)9wRjb$mSnTgA4a8M|4?Z#rix?Rlj?4RA@I%YJPmRjIiyb+#mT&8
zRjwNuqei%?CPmiW-b~y$%JL}VQ&)el6(crdx~elHy)OvQ;%$yU;@B42Gof_|e%mBo
z$ORMnLRL(#q%)~Al%N!9f5ag@%QroXZ6)Dl@yUzZ%|X6?{&tHskIyUYko3Fa)+ozU
zg~>MPWJm#PUI}Ggj)O>eRra781+)A64WKIluOiXCUnjGDLu{-w49gOvu#XU<G`KG=
ze!pow*T7Kz>|HFCS7KEUH0ZETFZvhQs{pj}*!K0kHPC)55Zt~!7D#-6B5%L!6>nB4
z$EvigeaFqz=?HIF=h}U6*(JWTB)(wPKA;*Eux=^q;Sx<*eBPkYIpOyJSR!=%K(pQA
zA*YJJUw2)#B`>l!CDwMN#DQjgH}9__s!<7`;~#D3=~20%8y#tPtfeg4UEU(@1+;Qs
z(nUQscGvG&NdHb0@yMx4_JgL=NT$Pen-sqSc72bQL+>AfMDAXZ#k3)y(lhx#CvO=L
zDL#SNoo9JDBZ>klYX_&}T)3fu{8Va_-)7y*TUhB`m{=>XZkJWC<j*YUvR)Q9ti(@G
zHS05mHbZuFKhtr;$9e69KUAj*nWC%H3mnG+xpi1a!jQdp#g;nYkH#IF<uJzITy}7F
zV_($uf+J14-KXnhb8oG^|NEqjAD^>@h=n{)UA+HQ&e7Z0QL&l3;Y%<Tw~Hx(P%JC`
zv+0vN)94(n4?63?hd6=JU*iNYyph#EL>?3SyRC1vu$_PVro7?lKiH_p+U(gCL>L*d
ziXctlcC7%-M7s`GIp{Xgp4d(nd5;Y+{9yr*#!O3O+Nt)qb1Os0?)=&Ll?3s6iF)^R
zElkJdLQmhueicTDr?78y&etNlf06kritvALT!;}rp8+~`vmLLNu{|qTv<*{#nQUm^
z9k4#r>g%{RPwBmfd!h4b-vH&kz^v!B?65L_aefM&42td+#=I)^<?CogCT;%ykg2wY
zD5`igABpI4hyJ650f}cqDiLe7*_W!K&^>uq6W%9?W$RqRy|gEh#efmv%Yx;MpBS%u
zI!ZcR9sN2vGs;yJC9UEJp6DG;tXA%%*#8Km)onPIhysnDw5%TzTPbRhl}_sjV3xD{
z281lJypO-;Kpsyx5QNhbDtAh82;OsFSTf0_AuvYJ=P)Ll4>rC+AqfbWKtkvzHL+*R
zesbM`;SWSB2(oirhxj|$o@HJ>?Rz;ashEit>*acmVp2%)jswErnzz)J8_Kt~z-mnX
zLF^VcCqRd^_U_!h@&Z{6SxuY<afaOzb|+TSr<658ruTzSF^#ygH-ohFrR9^@F9sxm
zPGF{V>=9-${=-t?DS2*Vti*0-cpqIk;r2EAwsTX1Y=1e69H~N*J;-u$;`n@djgEuv
z^o_vNcV`ur>-%Sd0}>176q`8Xhoi)EvDT##iP0POKGJ|()Q~8px~+2=cO4J&q^?oP
zzkM0ROvb|FdlSQ$W`Q*aZg2}YL%NOj5j)D#@5p4Ft>g(a(JO<}5LGsRg9Q&@uG1RI
zTlEfS#m;&j?MP(}@XzG<<_PwgTRt$OUO??P&P}AdPjIHLD6%`qGvpb0YCN*8r|m|h
z432*I>qt<kxT}$7x<y2{IyvHW!&Sgj#xXBbaAqlkn-gwmaAXb4_iuYhic*MdCMVUV
z|5yyxyNE!JBh9j#K!9n(G0!>!ksC65czORfyd>aK{cc@#%z|4YERDMT-g4RCRvv)n
z$Z3*~1-?i|gh%=a(Lje$PfhxZ*T~lW7q|+~5JLPVft8aR-sAP5b;V|%(08n@dWyE?
zZ~)zow)h|I7W^T2_03L4@e?*t2!qR`2R6(Pd;+E;v0MAY8hrQO=F7?Yn1K7t<eIKE
zx|!&Q?1VAX;1k+bPSTtLHulfc?~+SD?v9hpmBhG|;>QtR0u5**8@lpR6-QLT7BH3c
z-c3Q(_yul#?h&5xZ1spT(=(H0<8ynB--_?{YA*kw9u9_MAg}oE$`X@5_vIWCrPwB5
zI?C(z{ex8Z&#grfx6U9c&#FDvW(=p2?1ZZLU)O^VT8z*s?oCZjB*{W^3lZa{NYzsi
zP*Qrgn>XM%Bki{KXJPmhySPLV%rfJyH-dpr76A*wnz83<C73@>@%IzVXK335O*s>#
z*~2!R4!q_o!NFbqbgF-7T;TJRCgi${^rU<SXl1Cffy!i+NS^5JmmrBOncA-S4SiHl
z*6{NR99F^hmr=9&VC(*oeiC*4_F5WwEkDTquffNNH%!}}H71mPHN*?7kaMP;Qk+Pl
zZGphW=K8F}Qn*IlYQEz-%Q?Ss`z0o4-3F}y1QtYx%s8$~^EK?pdzS}s>$17o<9$fI
zCKiZjZCCDOK?X7VGMa5&#j#yvARiiqe+aZz-IYu0zwWgJ+#p{}<Ep2FUYhK-F|L*C
z%|^mKEE?nQT=z&IKT)ifA@520$U)x;e&)pN4ZC3Gnu_&MSBXQzI1Ofa37??v4iYJ-
zbFi4ohhpUKruk#L(p7}_>eYwrJk+L_R+s3nb%jN@pp)6uQb!*9?`zbzn$G6(cGv4d
z5<fq$TYm!@;hQ#oT=k^GrWMWJa?3C-QJvTyO;E-tV|1E_v<lWk>1wtUdR_+8O%|76
zg-fT)i*Y$q>t)<?o3sk**?Bgd5Z?HW+N1Vp3&A!u;pxRHs?973TE5^_*G`n>jO&>u
z<MjIN>jZD2W?vzQv&2iov|`jPzUy1N(efTR0}*#8V)ySSU6QsJg(thkP}Y6Z1+QEs
z3(hiWJ$X{Um7^;@o~QLjsk)<XYo%x?4BT^Er$f)m!e!q51TQ=BQk(4)zidFmV7=v#
zF!R@(J!r1LVL|41dR6wcq}uK{HLBgaZI2{~a<-YTRge#UbOkP`>Fp7osVD`Kahh^Q
zuh$-u3r-{_v}GaguvNBtry-~D$8o36g>6?Su)%<6Be$SfN9*s4I#daC6*(LM9WD2V
zmbe2gb=(e-0gI0+P!ah6Q(1bL^Y_9QNw86Jy}*yn0+~UxpaFt(7t4@AGEaeKo6bdX
zK5id)sinz7qY5enDL!doN|cE`+osBlNs30v=&Vku#VE%?n+zkqvt~NKYzMr{0+oIV
z#jgmZV4MCCmK_^`PZhq2><&5eWAl~>>r#+Lg(QuzjemXU$G@L%zD8C5Dlf)mFce-a
zFs=_+6nvCv6SPSQ={iKz(wwN}4mtL6_yK(gIa6wBXPcvm`*V4-TpwuME>*c$cOY}x
zJ~4j~C08s>g7<{agrqTwy#uHli*>i;(7FXO2cqxgr4Y0Vi<9&Jq_m3KEe}AC;;)_F
z&D?bP3hiQH-eUepy~edq5w*naVshT1mwaotFUR0{t4}!zTB0IN$Xmi`a3Ej%ofO<L
zmup-tZ93#tfoMIuJrHjma|dZQdiSJn19w+T(fEb`w<8Um2DNETL}wn9r$x$XBrA%J
zs3b?~zUYVhhFAR=i=Mdp^Rn>R6|$Mxtr>5(N{5+mp0_QrnSQeL)*)iS^`t3~ToZjz
zvMa@QY31=(42coRp4;2vZv}ta?buM9mlUAu;Dh8=L5Rh9hfhH|YdHbROXYa0o!D}%
zthucnFQey~q}>CKx`I_tVU+aGt28|ypRh0ti|H`+j4f#R-dhyE_{^+cfW@!G6Gbjl
z1{F~QVk7s$%oe8exfYbQU@SdZJjCUfoc?!e(5)_siiaQK4^*n3-|h`bk=D!*$S#tb
zFppA~TnYJtv%)m~@Gg_CsrFoBJ=vX1_0wqjs1v@7HC})nj!UrQs#d3gkdZxaff?yv
zR6Z~VECFtHEE%p7#CwnGv*ccJuskm~!{%lR8GwB6XdJgDFwO+1i>>A1B3`>Xowo4?
zM8}>OW2dAIlcehI4CbAAU3{veYp8E6-J+b!O*DGfKc@1)&~-v--R2)#bX}40@?m}O
z7*2X-fd?Sg@{1Lg3NQ1l&&yy+HZ93-EFG|vB!Vo(U50qnviJt|foN&mWj1eF%1K*1
zO#H0l!sC9sK9&D_Oa5cli^5n(S2fKH4x&?36N(X~)NYUtIy6`(n6xzFb90d%9l58p
z`NX))9g`2<Y*bt@cq=pL*pf@-J8p}dG#qu1=6et;QIRh@w(Qf?Gb2B#m^9Z?mk8#M
zyq71?GV^NGOyxx{ZK*gAua_kEHaPEr->TgO8kqNev@h8~exJ5VRa4JPtn8FMF<Vc>
ziYSIGwIOctt~a*b!@ou)94q`b#b5F)kTcA3!oKh{{Mp9kP++@J8yC@h^JdTOoPX=&
zuz<#Jc<li_atkJjyfY3tcpVW)VwDGVo{ExgVVlE`dB=!AUbhM-sS;#Qej%!-l4*hp
zG)WNqUC|XU-NWX!_GUub!HnKMh%$=l$3Qi)VrhC^3)}A&_d>!n5zjACgODhpk$oRp
zD3M?cksuq<OJ#<rMR2bO*j|n?>gLkgPoT2Z;(m<?c6o2wcr?(cOajQ1m)M3ZqoN@#
z;bbjrv@N9XGVaCP@#)=Jhtjn0Co{8tYB-#T-qhAb1cIzc$*OXBlII4?;O2I40d07E
zjJa9DkxBcINgb5-==*^spuHwj+Sy$!loss21{+2Cl@Hi_W82^$TKWv9YhjZxa!eZ}
zgvoo+5>dn2uT3UG?ZZl{U&0!#DpqFu^cZ>v7ag!{H7!$fM1oH}y?)qb7tqO{MsK<9
zS*Ofp#i@UDLHnmYB1#J&5;P-BGx1KpP0SC?<{l#uM=ST!#tDw8*R@wv7=JhfT2*kE
zLaFoh;^1Bobrg6}th+P21u|bxy012wy_!7K(N}XbDr>`w)8NhB*{-lH2~q^|P>g`j
zmC!SXYtS2kb*DGW;0W3ZIPl4{p6EX}Sv23yUDMO2RrXIQ_>2*Iuf-@a03oZV&-vI_
z$k-dcbdiRT6-Q76wYYy0Yp*hgCnr^!h+orwLz~&K{;_sOT4(<d5UJj<UonH58^ZCe
z;CT9afC}=$@Xd7_65JN|qiZlnFG=bHtgAywrrUvL#;l$1&82C`ZNm<1RbQGc+}s*#
z8Ex+N*`b}O-oLj8T`f(*v+59kBw~Jj=2xZmi~&UI>&Cmdh$(8=<uw=>aZ2k2Y3M(D
z_~)``BE(`6sohfFN3a|f_1-~N7|P$#@v)Z*`XD)MrMbR35nfQG_9*ZOVNS4oZ~FS4
z_MXcvl}Hbv1<lDZs$O%YYE{DJ{3_<=j@9(3-nVb~@fJwi&Bo6K`w<we(np>x;eq)5
zZt;gEhOH7`fuV${AS9880^lwM#+B;sSRT?-x8C<QM2bNU&7H|D^|uJ^xzj;4X5NL*
zfAeGH;BqRZUpr0aBE*sVrH$>K6^=)6iXXraM$%)3iZhA4l%nRT#WBEtIbv7R@O>j|
zHPyZ2Lw=xobANaSr=9L%pq)Lw1w6<B?sS1)ZrR3JEj&K|WfET6haIpr7{e>2FiqeW
zm<%|3dI8s}kRDcXZGp5I7(VsvmsmTQ0yk8z=0rk_hS!-qO{fGkl;;4gQpRNXQ^Qz?
zAqf~EOz9eI=Z(=T-p|g3#nLH!mRnk5@`Ed5cLytq2)w~tBxkfD)i;*Alzd26V~k4e
z6Wy_2YJ0Ff;-qJ=?d!0Qm2Wl7>{>!iP)#v2j5?FIJWYi$QW27`4yV<Do6Mx7us1{6
zARsiQD!p}^*-@r$No=B#(lp3nzGnvULwsA3#GEEdh^ANE5<ZWPwO*wfBQ2c<$d7iL
zy$!EK1wdC~tx49hUv6AVA}*6m2+MX;Bp^Jk{dSL2dL+a_<=}|779}QEYMx8Q0O+Bc
zp%MI##rbz1!8@+U_<wJo{@Li-*w1gqclzpp`TG|e`7z=OGXgb6t_nx&S`CM{f5|Lx
zX0jYGm8c%kdho?J{a%DXw!f))f$s#{S$Pi~{*ZaS9&40pS>5i8#d(n7z34h*C^G)g
zuU9<$i{iAFu4wme$jmz@6Xg|2yBMV)2Xc<O`ZdpjgO=!JO-a3denu)<EFAiaFL~jO
zQtFr*jd%jvhUADb7r#KmhSHwRx?PS^2Dy^Dqg!M@aUc^W^LF5$9hf;3r=4hWeBGi-
zz1)fc<0D|?$`forTK15X`8z+Sb#n@-3v`%f%pARu1gEixrwDAVcmBd0D|<Iob`qD~
z-s6qNWs086L^sKi`530=6>h95cO(<~j7eG`D-?`JQ=U1*^`?sJuFPW3b=G1b{&zm$
zQRohP;66?q(U8UyrqFS6KjBXS#01Q2<`2hG7UjX^(+|}1f*$Kso4j2<*+Qwp4z$>S
z`wk13fOi7#n%M-o10L+Q0v`$G8mgh*^xjfLBaVLP2E2S|o9XEgBX^rx=^(k(d$Y1=
zSN_!H=<hn+fi25`>gte?X9AJ%XaZ_0fcuUPTFlfbJH`XUJl-<?q4|z(%4w$Itw|R_
zQWsZpu63tyqzlv}Wh%9RocHJ}Vbo*;G)W>b-XGfF*@&!j&Lt#Z-BoHZ-ISe;V+?Oa
z8k*OIPJ`s?Y}ldS;P(NJt4<LG+rg@x>oo^jL)F=QLpkufSo!A<PZVOaG<JtRHh$Q+
z00gP>9Oi+-T+7TuWIbz8lpLv#1QF1VTst+A+E~w(s2e@y=oa3-K7re&(dkp!@kC2^
zGQThTrT$baw7(7-wc5cA+HJ%ntNfn#PU&YUS#_G=k0jt$Fg-3e{9W`a`{h5YGTeCF
zH_LY%GT!<a^*#$N6WRkWi$>+r9J$IL2bT_SU<R*Lm?&ukv%riAjAlV|Z(=EP`msVa
zN_wcuxgqeYK>L0#EMDfdxhd7|E%VbkCN%LFs3nvPaZyHt7S$3DRA`;QyZ5-1?*?29
zkOGOb#iOs;>fAZ!j|m?8`s^F#9>mh2jeY~Qqik3j_$kU+Gw~Bqq0X=htI4Lz76C1l
z10misBPK5a#@7BYwO3!rmy~p~>zHOr1A&7xQ>Il(Qx^%V{HDf!#S?}0a%+<`84B$I
zaKZ%fOZYy^BK>Ac5vp$-@-^2Mb~g>ojATzU`Y4nvn3BXe>zVj1Lh)!{E{pSi!LYeo
z5{&Y)78z#BmCaD9TXkcdX4i)uaBfW8LazOF^_sn&bS`>&&-c2w<Q6aA*vByflJUeo
z+}?}5AU~6$pd*9>tJ8y!eQWIXw?Z+#xk-LKg2NLyX-XZ@v`GU>%1wQ=Zxhl>BbVR(
zylRA?8Na*KP^9rARP1>_`Iq9p_6xfFs?_RPwt11jLtRUMk;wkiy2SRN;jd-mdU_WW
z?ZyPMU6Yhgn}=4(f7kPJ&@SpWaxb3rGkQbb2Bfs0>z_v7DhoCXUTEm3h$N5<D~chT
zUnVYHy}|xvkHM^-EKZou79UabChpqak?USvu}h^FTb^8J6ojz5r)%u9_FAQs3Pl0c
zr;ak*+n-7`S3+uod^*PhkDjE!mE`&OmxF&i*DUnP7*3Rw?@(sD^0@YT>Ae5Dwr`RQ
zYtD-fbYsV$_%w{VX2Rl9LXhsB<u=Wqxl83it<WU*ZHw)-4v#4Rhs{h&@Lwzx+;VWx
zPYeE_b3vq~oJI-<Q5#47GqqxY{h&JsD!x{CjICcsH+_N!^4+DTVu1uV;|3EX7_7%j
z@2sxy&gA%5?Y(VGNOC_Ztf0$FWIN%c=mHri9WRK1xIH7<;24h`bGe62Op?xQ4ij7*
zuk!AT+EAMi+J;t<HSTMY3RgO@?G&B5BcyfCbQ4|jd;+K~vkkSb!51hN7&2r<!UA}<
z#Eq>v(?4Xc(yZU~NtxR7S-%zkLccEW@=FO-ws92<s{5W|7x#^-Zoa!J4R}aB=#qPq
zW-PLiWm;@jx^m6&0I}3hYp^FqxS;brdHh2x+b6G+PZhC;PZ>{qD&#ZLd%z*gy44A_
zlZb7h(t?;mF4vm}t&9$ps|`A@l_&VypF&8+TpuZ9@ziQ`{OO-Wu@W$00WJ4(V<3qJ
z{oWWxbo+6;M|~@ebZIh)rnL+BvXg>UI^4_iAegD0J)?1l7dJcfg<ruC$s9#IZDgcp
zmTB-!OmdXDJ@-}9k_wr{A>E5#E+9(nkK{FHQ_(=r@4O*c3p@8<>Cy&ewC&tCX7yCJ
z(7y9BOufU{4SYHFs2c-M`?3>`xNX5odAVvlMr3ec4r23!e5GRbZBj{|%}ujeIp^PQ
zH5bm{2J&l}KQn5w?;-=~0;yd<eD*sVPsRq@u8AQ=%V;X6t?mwOhX!;g$+~*X@P<eQ
z<SG1iu%=KXl8-TiQPVa(Rplc{eX{mhT-QU?i}_p@{?4#M^h;tiaYKmu77D118{V0!
zy@Pk~;Hco-b(c5>y|PN8SjBdF+Z|`d?y4h|&Hj0W9D8T%UL*kviAXg|74>wIWN3#b
z!ZG(F)BZjS*qJFZXJA*$+IK(4u9M=#oSWcD6cJQ6yQwVKL|W;$+$ReO^yj#xo_If9
z=JRioN)NYCN6INhh$V4fvpOlDX>e{B9M)k$75r9}wt3i^kKw)cX@igZ<Umh`Y-3r)
z+-A;nUAW83zI2Zn{p6{=0qJdZfmLgC$14K5U&+yi#ha3^RhCcG#ue8NTt3CPRC@E8
zGqg8>kGo4b?gMT<lW6h|vZ8w4AZBv5vBeDMKIu_8N^fK3YS|Ig$hhrnXTiJKTrjRS
z`o<r9!rzNJJO^BA7*=!n%1V!3b4kk-7(UPjP%^G?s&b-D<VP$BCzl$O`0O82$Yzji
zbG@{pL1|=_33YVV6TQ0zy$8tXIHl_Y))W)-?~UyUzHH9LG){&l*pNKJLRNAexNWA)
zBlWP5)LfoxN(NI-blZ!O4)MRd#lIj7%`KvAvTx=(*heh$w^GnjvQNyrS2{miD}R%@
zI1XvgMegoM?>*6b<<x`nlCY&suP3cKrtYh|f8NS?IjD1gxlj7GEOwZqHf}$95zIqs
zi(MNspE4(zf43qoYMK#R?-*4$(zS$N%zwi!&mJQ6qWM?$WovzjcT@Laq)!HqkY#0`
z<4J4uY$fg{l@hlt-NxaAZ$iCXT0$;1m5SR#%G%0ROh1wLwomYjE7aU5(=9DBaXW@7
zmf4jKBdb!_Z=dkeJD#c5zK*u`fGNSFmP5FQn!M9#X*jF1gT;NT;$S(HJrnB)See2e
zbX9M8ORExpA{go9aH0n<dsI0y%i|^6V<&3Muqk1fG5yh1o6#*PH_0Oz8564%@M`(}
zhsw@<!x>JKe)0Uo5p1#6>9T%zrmaP}YvGc$w5zm3A#Fp?aiHI`Lu-bS97UF2M=$lo
z1?TgVCJ}l`Z$?#GE;Cx&u^;vH9Q<RD^qZW-uwKQzWNK`*;@_h<>iiasvYyROC$#7x
zAkS}p1K1mVm7q~d``rxdtKQP>S8Xy9`3-Zec3n^{rC}W)V^?VJomLS(E#{tfJVR=V
z?cZu22h=H^*tL9fz)zxGNhXI4|Dm=bVuEtnrma%V-H{;Tv8ZnP9VMdgNmkp6_|P`(
zK3TNmb7@jt8+x|Ep)AFCGQ0eFqzu06bNP5}+i@fK=(@Sf<fq4CZ}<liq>}jY%;sM9
z4VzWkXkRw2_<naV=m%m8d>4lYx937|j2^K5Ih*`rZ&6@o!pw$$P`{-F3|aYr5I=|@
z;85V5WB!M_O$HQ&LvRQ<d~Igl!Xo<>F=n+P*AGHI_}#b!4rIH#UzPUesr$S~12&Tj
zB5X`QynQwtlNH}kMvY-Sst~cx2$=`cFjf>KLD#rDD~_ill8JcP7iE6_0M<+r;{<so
z3}A6~Z&Y!bBVWi<O-N;UV8lJ*ROxI8JQEybmmH@OWSKd0@oRmT$cr9{ibajj=Z1+p
z%Hk=&N<9xGH_zV7K*K$N`l(GeOD<A75-k!IIU;+UOM!(jWA^Xr3};cuKF+0=dDNM(
zxN3|pSgoFPtaBJfs%TgAAq?ylnF4C;jaKW^UA5HrBarnGkW0Z->bfzq6CykT1(Uth
z;jY}&;y{$}%G;OiZli{y#4t7L*K}D?`?#Ih<%su-AD}!I`5XM}k{ZW?PAlqp@PT)6
zUOZGFH?XSsooD;^vs4GRtEAR>mHAN8u8U@oJj*CX5EU};Q4=6&^8TI6J3)E*q3Yx2
zUfmtMZ$)Qk*&aw?+2^)*RIQ&v&oDhq`YG0ZHq7t`UB~FclKS25p}keE_G_zoEO@m@
zmi=E;c?|-AI2M#7_k{&VbQl}fQ%8-w``i$VodVgud_caD@164lyQr~`jdDctC?#Qm
z7=i6j^=BKQgM=B=>viQ~t5uoQ?^<ifFD#61Myd<b3-Y~~K_tT#L%sxuL>)*Y!e0}g
z;$pbAn^|^?kqQSMdh9iaAM8H7X_XLbf6``l3U>%m6IwnfC4H5@F#Vl~I9b^#Q#r+Y
zFn@nlepQ`B+PH9-iEob~m(QSZ9TR;E@?AW&qdmVel6?OuqoVEF*p4hfGnmuX?SB7H
zzatX?N`{0WMr@tKINXf??pB@^KdzIWWh5t0{4nR$V%(g?9AMflw|>v{wpSnK{A8Zj
zD_XI}5z=<8h{>z;pYQYN70%){NH#a?63#XccW6cUXTX?STg?r@WoFj%tY^!)73m6A
zIu^e?Uhr)JJL#|Kom$vyvU(brW8IchtPfXi@x&}x1`%>YboJ3ei}CD({Jg3I_cNb0
z0h&u8dZ(55*D-T|JpBjB3ALeXoRep%vxDL|pM=#Mh<1@&Mf3;%7gkDkK)*)24TN8o
z2(I&-b>!rRe6v&asspThzPR2%L_u^HEhxDHha~$sU?<~c70q)gXX+Q!HryeGpeZXl
z^lolG28v!rP9@xq9$aWR4|mEaNnQnTYp8d|R!q8{`AQl^7m}PZG3f3Zy}8j3Se25J
zDC{G{-|tI@bAlOoES7PGC@*j+#KY)Z!V_k?yC2^p1NnRYYU8@G_o}FRW<u2#(7Z=%
z0C1%#ct2nc{mEZaIPgP}gpO$iOTH%-qUg~rBf0Rdf+|o%!JEktfrv{|M|mCRA#W^R
z8a4coaYVtxlX4xG01|<-9B)QHlYD7%Cc*BNSAOjz7_BhQ;Iq?$`%?I{Jmjl;IVCGj
z`18tltZ1Q~K?Ug!yKg`XG(r8I6w<0OmJvW8UP=@X9PY8mJ%(lXCOK~8#<)E84T~c>
z<?x~uAoPtpB1GBL>&){<oCt1Z)Y>Q>f1D^TtmgjN)dQJ%KKx=mN+L;BXF5}@nztd%
z!R}*GRcdutJ#ViRHD6$TE0#UVsoa8Pf^7k>KH;-hl3wBRl|n>h{FYPj_~oxQDvg>^
zal|z}kx4lT8hP2(uNx4rI}d#1R)(Z(1i`s%x4?VaNI(#8<VGe22|_HYAmmsj<!P-x
zDFFtVotx3@G81}Oar$|a=l&JF0+CKTDow-IDKUk1QOOSSYbiSWw-l83g}3QQMz?Cq
z4Ypf=`8UG0zPiF$v`-vC=j3I{;5zKKmT<&i^y!yNWNUEVXeVu2gbh`?{Y|qdwMkiB
ztB5tQ70*=h!mIx0?7&q2uvA)FH`9UreBv|zPQ>9;kj&dtjk<E@#>vwk3=@pLi3c1r
zYpMf+Y;7-LVzEBUA-@P}!}<GkbPaFu^qvHW;xEm|z92+3tLKfLuCh893c6Az2Ynis
zQT|hM`v+(v^d6p707Vs5cTY1oNt#5Tsuu)EgKbOG*IR3Sy6b@nmrH~zNKPntjgSP3
z<#2sclA_70r&44>VjjJB`mZ-5<s;x0vVadzwIJ9DVLu2Ac8HbSNtVkv#^&wdL663O
z^)=W5545b$Gd7Qe^Vzn9lSBsrP1(j4NaVTj+Bl0dTq|xM_%`)8HSF|wjM=*v0tyak
zB?Y?fD)91{l@`d`ppbJ?CGzLbp3{B|;$a|!6?tCT2Z}<o*`Z(hO3DcDYji+PXpeTj
zh6hR!!M61|yB{1%%px;*=0gGDtMC|0dJ<M*W{HpM+`%X~h?4vKE#;lo>@bE2l=58k
zJvsu~9ibd`jvl_6G@?9{$g&o>MTpcb?g7BJ4sbHq2@HVj2r!*J7Ouxp*DMm}?NC;`
zV?>^6UFzIUV}1Nq#L$AE&4jfV{`-8dK)wa@>>7z6Bx+||<jLXE<m@Go>2P3ziZ)Sr
zRo-(wb{LQ$*ukWyG{l9=b>|M`=YY$~wMh%Ar~uliJZc-=9~K$L^-`U=NC+TkjuQ#y
z>GYb-wO3kxem98Hu7PI=Q4=cVlr0s^Q+}4oGvY0cw<7R9z8#PiZ=4{49+oQlZ4UrP
z#`nRLvG{u=d;3&XV`d~yFHa<r*&9sS-kXq2%s|%CR!Np&t6UQyl!O-&Vr`vbC#<P{
zv|Vs$T<MFqh*(Ek_DX*khsdQ8>2nX{FtRjXQj4Mr&KQ+9=BoA)UrFk^5`jzla{}6J
z?Nw12f^Y&)1d45T(1hHOg!043%>3ajsT<dZmY<$7=+V6RT1KE9GklogcWSGVv^Qed
zSOSgU^!a@@>$J}~#O+OlmfET^OyAdH)alV&qE||xoRM9H!q8u0gz&uJm<P$;Kfhei
zn=#x7M&^@_5S~(to5nr1)$RQjy|{eA`OZP%qf%y8Ec1x@rDR6&O2vnt1Nk~hGxxMU
zRvUI7A1}n7!J<ayYh$YnSyvyD&~F^Hh;sv_YVmXIK=j*fuYy&-UB@&Q3{Umc{KZXg
z-qUM_t)3x5RkMC)CJGrQ1gH6|s|)o+YcE$tCW7Ktf&i8DWk}=vg>{~&f6|-J5g!gT
zbw&rBTH=86v`lxP-;e2*Ob2HVKxvY^F>@|HT^~fa=d!tQO~&lmHKTg}@}QWk-|+7$
zTBFdo{q^s<N)x23eiJbBTv&_<W>T)(n?18(wXXIQiSUnFgYrFHT^?M&$OMoGN_Ypg
zOOlRhFY1j_@E@%xeISFZf7ucPZ6$cENVbl#M&FC(UNLuSr&HSH`If?|q2sES=mP)F
z61z{lkMA(4f^GqI`9-~021p<`xbxgg)%ZB;c%;3HC~my-Pw{XX6&)cqcU(sy14H=8
zm%mJ{xQ~+pV3(E-Kgf`1mhM`lB4!K&iVr+85C<9zTSMPZvH@hUzkIdhYr%iNMg{VM
zP2yvJvGYMk;ICe*rS6VyRA24eCLBxwad(@XGj>Cy<V%*Q{f3af*Kz;+Wh*}J-m?W1
zHOK_4thX~q?_^~C_8C4gn$Q-K1qO+N9pg*0`Y>MN{2KJ8K!OsjGYv@+B&eI1c#R-(
z1|UIp1Q5FJ?avX^yOL!tC*IXdR*-JXffKMm=EN!Yz!0DrpZ+FEj)i|fMtu<npX6wK
z_tZ5$<lf%y&0fJ6J|j6TvtLdw!~}#SFUPS9yKX5>5np{eRen%>!Og)V312=LkFW$k
z?Iys$9AKZ*$~TyZl7%sFz^zgK;C6?;UiFThy2V|XvJh{-uW4MoJ82x1%Pj@5Jw$Td
zbDN7i>2mZ#gZp^PXm;+K)r@9wG%WZ8rL15Bcet6OQK5#4GI$DzH*nM=waU(vqQH&z
z8_01>d8`W5)p~=Ysx_Byb02D}Pe3eL7%dfaiW^SXKT2joW=2&B7m374kK~|2vGVGU
z!6JiQ5=&RSJ2~<#+VaigUkj7*sws@J0zp2i-a#s(MKatdv@)_c(yF@p>;SC3YQ20R
zBoh-*qw1^W4jKlhVF@G7iRVayoZdVmabI;(Q9~1a$!H|}(j{?5xTg;3rL|%n4~jM&
zv8<C)x3aAUZ&<y3(f-DA>z$NJDbTyFZ!Hg*^S#E5m&U|u=$xmPsx0U<3NafembC=D
zTU)PI3_5?t?E6c*`9l(F_jwmZNJ=&=x?W3N(6U(4M1;DlHo={;)W^ckRm2y4d-YJi
z<Ue}6{Gg%oGcU0Cosn%T`dhDeegUp+^}|B#@QP(JUKvm270urg)P_AG*H}F2j*JsM
zUg#Lgry#fGuA4;>7Nb*{a0K4W$9f3*Qg<~Cx6qrxY1y9=y9N8rU$VMIQ>vCQo?d?j
zU0?>Q7<?vd;y5ay_SLnP));42YrhR%r&(~jTS*vh5yk0AEo*<##x6DYl3?%A3ISGq
zD91Ogz+g69&bY%C?;==DxIXxX{uiC6e@!`=!~EE^QH^bO^h_6t7v8i24fG6Nwf5Kx
zXwDXJgP~ac-t9qAi8THmC0d8ag?wG!x7jOuU<=kN{ablJqU-~Ml`vkQN=P<ctJ(jj
zg7q(W&VStT2w*s}fr`wKVm-ZJ34BggO8IgE0f@PN)B`;Jm?n2T&iu|u9qa|zuiSzj
z@=9N&_d2rI>$a6?I$^##DntaNv_-%N;qFh^R#%nuY;CC^He)`Uk~gbc89`=*40E4+
z)R{>CA(NKn5Fkym8^QG5QP4m${nk>*p;Br(XelQ{yKZwaWAk36ImEn6JP*+={W?U^
zc!G(fM)d{_*TwSfQ@%`9n@LN{U<n)M)P?~q<!5OV!Uu!Iapj{z<*26Dx>B6R>;p?r
z*E1VIClvOBjIZ$xW%Tt-w)21h3ORkIG8t&@IT$%X%ql67`xM$56VW^_^q#9Tq|z4z
zTP?pU_nUyC8;AULu(XPTr~MLLka|dG(zhZycIkbx@T5H1-vP&D?+bKh3afL-O7c>Y
zeCql#HJZHN$GSR|EOB@dy%Va)wA?P;ytn4P9;^R}XQU+IWs9h{#I*Kj!#Hv`we;$Q
zbMN}qQzJKjsUop`UWe$*3}>~0EXnF&dJ{QuMUTO%_sZe=TPp5rG*P%r!$ZQ&s9zKp
zLDtqCOlzk(8Y(Xr1NpD7WoIH|fh7CKUhgS(KFE(|ng4oxE!2ld9?tapNAi~ZiHDe5
zM-6J=iJWA-QKOT^u=>%F0Kl_u402WTSK3A1M>-sge{B3%ky&t#Rl=T)7z@kMJ}_69
z;HV90)I?n4e30h728+7ag>PXC-(r&n2I4oDLC7H0{)@(3+Ew-_Rj9ZJSHpLrfo$%5
zQCGH;Z1q(G%k7sF%mW(=V!lZUPXHhnD_^bE>Q%{L@FUfd%0CUie-wGZ?d(4gZmW2K
zeZsR6;0f{)Vb2*etQK${lyT-0Ms2imO*upAPv>sM47N3(HT0u81Zj+vh;4akPCzmQ
zExA+n6@3t(7Ez4B({<1diCF(ps3cQs_4p5gE%_5Dd`DP+?tANh{-6Kl(*)L1GSS_h
zYhb!?09?dw>?QlC=2FG1-qu-XE-*v-j@-!I{TFotYN<k9lB;u?ssb*Q&2>Yq6w-76
z{pB%xfbom3x8AW1h)sO!RMfh&aSL}X)hKS*_wR`|NCr$rx8OI${|T7-Z=||VI=m}Q
zY?#gzz$=lEm_YYr39u5d!SGNeQ$VRivJx;i%&hqaffGQc_XZ^PdNGe?6Zj*&UV{tC
z0s*#>)YYq^_dz8O^b|ok0IG}xC?q9>A?=@K@%`JphvHs69lX1mivQ(l{BukF`w!r$
zl%7EH5FTqmidCwuP-|8>39fPgI=&#?{V4$8VaHs-{s1=A_1_DNn-v2L?XirE`59i%
zv+ocn5#RyhNb{?TcP=;MaWfI7ey^YQ4S=(|LLwdA#H)=<Zv3}}$t0Ww7(C!2sO5wI
z!QlDpqwfbb(>m2t&H!IK-=HIkX2Iw(RecpGyXJ#rjxYBze@}H@yw+A<W4N%&KrX|-
zJ=VS#v+)Nia`jD>Zd~<K@+KbZ{|^Sre>|E$1KOnR>T`z|J-zktAD>&M0aOos#lKKo
z03*;N89+>}t37)U_!1w{;2vPR!QWE@)^neTBqsl}B=P6n{@b`fHU9?YRc30?KVBV3
z8TwB$Am6t?WWcBQJ^zy;Sae72&ef#y1r^2~hAydKQ!ef*I~A@Mi&2UIWh+d@fRs2>
z7XZqWyyD_BI6svBA3TOYY~YpbFE>NyKJWyAKLDOK@XLzN0W~)Y4XG0gPoPACxvBk-
zav31tl@7|0gaE2Xp@;XrJ)Q$-cx1E=m72MzD`+7=swkPQSd4!VN=(Hj)UR{m0$i0)
z@)VLnbikQxZl{%^B>weJdMnUi=&nV=6%yJMD9!C5QeN!=D&3GCz9xw_=rFTW?lblh
zG?~ijp=He>7T{#^K1>yQPn>jx+}i}0_i7y*QgAI-tEO&E;h6-%27u|rD*=>tOfL>b
zGvCG*UFfp|MZRPJJP4a6V(?@Fj)M$W>?TLLVd}rX;s0}<nCvqFzyIocyTr$T|Nfyg
zcmSPu8}p)rp8ARj_!mh{bU7I&697Er%Nu~(EMx~D35apm{0N%+%dz(BcQE65R?^PN
zN+!h>l?p2E{>k_^@Z$0}@tJ^|Zw@U_{$gmh=EcHuRqY5;3IZn2OCW}pxTNXQ3o2;!
z@c6hxnqe8RR$P(Uj<+Abn*Duwq%H@b-Y%((d=I|u07%4v67luITnzB!(s_riGr`be
z-+h3NK+<&jt>ktgSnM31@mR+5_r+1eG*4OEjp&yb#{z#nH7rFhyKX8F7?9T_f7J!+
zI_PYaSc&yNH}?c=b+>K(=SYm5HR`~)|Fj(ZKSojWsytBE!|JZwE&F>k13rzNRiMy2
zNf^#|MX6$Podx)Di!(zzr$?QzSk%=RmB9<Z0ku$QnLJ_A19V>A*}Rxpu9Es20MSw!
zb%iSgK>7k+V9(RFg6US)3|7GUH|;zqOC*YQO0j%HtgS9;i4U4!o2j(%@XSJ?uQL;t
zyuK>(E{OTAQQ)<HjvZ>ekwQ#gad9|VXWax$+0n@f;ii+mOPn_k-11g?YkT5?ZS^h|
zB~O7dNd(%5sagTi#HE01qSlXIbxG@gvU&vegTZqpf^S_D|IFg9P#hy{Bk=TVtv{0R
z)--qdF<z`H6G{BOVmW&5v=Fr-ChZsUn!03<Y0<bBP&bgwya9k~#Q-5B?CfvV1^a~F
zL8pK;H(ZLx))~lr@9et`kJwQ%<bePgiXR~5@z1nhiEwuSyJ#5DnCopwT;?shSW#i_
zj(+VjqS*q}yNjytP+qRfXxskVLi}HspsT%Go)GZmhTd1>VEnV2Tv7xAIJbNDRpV9+
zw-10h!+WPzhLrUtTi%B&)?YNbk+_q4ZDB?FH!=XF*!k^!e(K#TfB9w9oup?1hDOIo
zpwKCr4fl2dgsvn&B8qi+*o5B|v+CB*`aSid%TS^k;aMNSD}rGLg+z<29xK5npa&m+
z4N(x3xP7DQ7yy;IU}yk$JJ$of1)i!g7N12Jqirpp_m3+IC=Px-{rs)C!Op2VK;a|G
zPbSa_42`AgADoou#GI*#&|LH5ip+~%Nj@z$%mi;~Gv3vraI|Yt;&hPi?f)y%*fZq{
zuD0`D!}r6tCHZAKf)QY-_H`Aae|K8|%pnSZa0rT(I<;x0(lILB@ZN41jd%uBeco0R
z$!RT!x&W6=60V|3Er6cLxm!(ktVYY!y8{%3sCMj+08rm#s}W!2pqwe-jiacZa(OTY
z%&68dsr~GBUw@#!(VD1hw3clq+9_C*-+9KwEJK{0Zh=*#cH4D0?XxafI^Hn;eY+T7
zB*~<)JU$AgbzF5wK!kf}#4M;%#l8a8N1wwuwOK&7qDD-C!988OLh)|}JX!*9Yo=K|
zno9fn>_K78|5!0O>j8|X(eKsUu02tL{ZJeHk7*qvEHwAUUY7P~q6SRpf~Wj|`R=7>
z)R|cs2XKKTcYfx~E_UrKQu@lAWiuKA@W(m8YH%HOhwsMv+apuycQGD%QvSO;Ny-XM
z|9{|d6z>&z)$pKb=<j=dGhF5e?d-AA@#Rs|wgpG{-69ma2|Wf1Od%`kmITu<DLRyQ
zu>}q$gFi4)%LWC6AIwqsi_p1rGBbF6-lVyr2i5=+nzn$Lkra33M=*sWo!!yWR`u**
z>DxT~lj}k4$^7P+*~>#=3aT^=euA{h6Qn4Bk!uA&&dZ|-{^8{4I^fhm!#4}K5-1yw
z8nzQ}{G0)%n={}z6Pt@I-;+>@MUxZD)&a2wApjatitY-k`$pf2_=?6`1KhLIA0_X~
z-3BW$nah38HL4-cVPV)lzS@7pG)w6<qlEyIiwGapQkDczXjj#<h*KO|2;F3F)9Q>s
zU+^}a?_zvh`T+Xy&#j+zvlot3FJcFa&=P;aIRQJAw{L)S9v#S?@ocOEFJl^@{TU=n
z5i$2tuAf#nT{r}Q_W=X<kZR@#;hHuh+K;zkUyGe0AElBB&IDo;>lt%+VgTeEaoSqy
zKC&5?2o*ua?&Vdf8=is=i+{{XWWhT@e_{x1RgznPBL{Ov#H!U<zpps~^+f}vbN~dz
zz18J=9wYwK8Sf|1bf5$9hIUS{_YHv>;(rli0ZHgQ>REUlnhhJc+O?LS);~!*_ik-g
zQQ)!b@*@sqN~wf1sF>eW%?tI%VIyQ+X-9Nop^fqva$X&2Hcyy57`;}r)i*<qBQjPb
zHCeB^7e`-nUb@kLJSFGq4g>is7OwvPPfvmWhed%6bj1|Ijg|3S+)!W=lp*aVS-b_(
z1m9sIott}}pJL^LVKn;2bbW5^3jWWvAFtrmMQePYR>R#G$aBZ2vZuc$|9p;mOk}!n
zMb=vhl$f(5$)Ro&-34g*)-_28iKV$fI3M`-FY=mRfgX`Le6t_G8F|~HE{ws8gATbE
zrw$ee+>U$IUi!?gvJ$PdH$`Mz6_{A$KDAKYB#@~mir3h!+AQxuNN~yBXd7m+{Z!U|
z+Y~s1yVvjFsL<4G19Zi`eJ0u~TB_PfCu6hS<W-=jNK{fgur=2G@-FnlCFzbmlEQqF
zYwqp2qBbllOKjgl2B8*J#-6Ak(bF~wD589N`NcpwdGP>yBGHd3jeF_aA`uCLqgUBc
z0aJ<w`Rc#c-LH_ee*qKnHa=CT!+=o6z099HF4JzKZt-B>0Qazt(t04imMQ~q;kr`<
zh9i;zy$=)Tyv~jCA;3(yeKPy}W6@hyRsCifmwDi69pH?=mcn^ort;wsn<GTD0nO8|
z-wTxVBB67idZB!@qvXx?q2RSH{9S`)2ioJ;gwiAiMz?4-L6sf!<=S?<qQFgH<Zxb%
zoIU8w`QW`uLRZ+f;>=6%Wiu$YWL3>CH|>hctaQjkm75gz7XpwxOyMaju%TB|5k>ZY
zbNkdS{)wA{<vfu&cj3<-_J2RDC=ugTq1paLEC-a6N;2U}u7z)N`Fz(NV8iXPoqSjj
z7=K9chNci-5VjtxNTMnJ<Jm{XWlEk@BFvK{QW1|fN@az;^_2kumhS;C&*9q!zW`%b
zl+t3v@?7zn1W{D;!in&_`Y>{DAJPQaXOd>@j0kg)QIX4PJgEf*HVQ4cw#m9*Jg<&3
z<cR&qWqpTcq#u-Xa2P%iij`2mXDELerMHGG1-D=&H155NlQwC~@7z1ijfrwiZnEgS
zyzF!GTkr(VEBb0k0%@3F`+)DymsrpXwUuF9pd!6@#Q{6gcq@W^QHHLLUO3@fzUX@)
zsDV^-du}oA3B+|Oz$leq?n_Pwvq0PIsSsZXe!n3A2*!qg!q)%37h~W#ZG+0OVPK{0
zC^rV$zj74b`fKb;&fRSEoJ7z;rKhI+KiJN&4vc-X(t1s6J*$EBKLu|Tn?HI&#Mv}*
z>om&mxx$od%`a6$8dKq?ovZC-=SkUFf?>}^cl8d=j*9Dkt~;buIb~R;F<57(<WjEX
zAnl{CkjWR4Sl2juxR$ffwx41F+hp0@?b7mx<fDE@PZ;R`1HAuVf9zU$oB^7DiFiyv
z%s8@r65cxfl#7X1@UYa;N=zkFzAW2w<-D$$DY7mcxzG^y=;dF$sbNL-0{D&c7{P5@
zY`qAXh|xa&jS&FGz3z|ZKwlYYBbXIO(+a9>Ouf<zbzL*e!9~kyPEs8?;X6cM?6X54
zw`B-=3KVIWQe<ZYHMCBK)tdS$Oi^%vQ<f7F5e>OrvzD6eg);?g3|%%EL#dsykoa#A
z?r2r<TNA`>58ycd6W6zPfg3lk8b*#q{7a;};QN41;uOI2d8)~}#ZrbI!Af+B&b{8N
z<l38|?lU(Pa*B>N;C@}X^0l525tHb<_Kpr9DYa%*ujII@CLjPnmG8$@*SRvsQyWvf
zU{~N|eV#q~9!W=!#@}0p@F=(f`A@gfI@8BVc-mYl(ItLc8vdulazmNOkjN|M@nIG2
zERlhAd`kAy7+jfy6C$^m8H7`L+_t4$Q%D}kmr=|y#0YT{<Ubzec1GSalXfBX><n2&
z3M6;@-)G>Tf7y?@0xP{L|6@35ek2zFf4>}Z7i#-Ks|uuHyNE|Y^z=1&&ZZ#K&x;~*
z%vNOF5d*qWJ|Zy_^0Hr)eG^#B74>cCU_LNpI&fIN-7!SHcfMrmadpNcfSn<shW(UU
zNzPFy$9ot(INzhSH|;A|R7hq<;`cT)O}9Jtj2m|oOHjbILjXG0TcWW-Omv?^2ZKw6
zC{F}i#qX4&_JMQVO9R986_DV4sDT`*HbpDohF>;)NgP;@xy~h2u3lp;$1455xO>mI
zrnYT;SP>+GAfY2Q3DN|qf^>uckt#)zF493ndIxC|s-X0$7`haZE=3UOf`EWDk=}c6
z^3HYk+4r1#@7eqQKfmAnVF<}ubIm!%(*{V-@vqQi3wnRe`TC?)<o7ujbq;e`SMsod
zHea66YpcOeYZdr*mLnNvdfrNsUHs8zVq|aH62aMp=P_pWHnMir`=~#IQ~{58Znx)i
z1Nm0m-5+HMK6}IL24Gmh&w3DRU6WmwA8!iM4mL^#aawG#jo1RF-<2!LE1Tfv?$tN`
zev&R$ClzTn{GAzwP+M?lN`0#p7Q{EPV}f$JxXraR+0gh1RP@`zkC@Ps+}!ScgJEVR
zE^HoG^Jh7l3XEG9Ud&zl?y?91l7_bUf%GJ7*+O=o94XEVH54J6b3Q^3G-ny_$InlW
z^mKOozwER@GHw8q8e%l6(fboanSehqkGq%ZesS?l7zl}Z&^B-`bg@gY=kh7bNm4>#
z34s5^G%N@1>8$2)M2oi3<LW_xX>YsK+%Sg<fW6e>y+Bc&#)|Jzgx!El5YM5g?Z&39
z3U8^NXJ+y2C!5~HbadUfrX%=VSAN7>KXn9YN3KXDx+;d7dQX@eu%0HZjBd_j90kED
zB-CPpk3pEW+9yubgWtahr8l<cJOpGaE*NiD%;Th&dYjY<iHk=o45dH;sW#&Qa$o~I
zz??0Dk#qa6nD7N)GV+JdZix*UIC)61a56ppbiPsI#zkU6dN4|-_TignlJ?Z;Kki_m
zRy;ZXLzJV;f#8+kNs!=bt0c`xOA;i_$5$NFDC<qQot+f+^8i=hwQYC{q(@&c>lg5M
zR`xolPQ1_1rRc~f^k)`;t=a3eUN)BX8xY2jXd1Oy!=q$c;$qR!f0tXaV|3;~$o_>1
z*w2X;?AvM!@dH?09=~}V7;I@@<6Lsd@7{!oD#201h5zT1k`ThtqCD?&{!!;Yq(wZO
z^!^OsjCJ&DSF@1&U#>phBEr$bw4y6FvIoOg0CqbG9xqa83}+J*5oI*Sa;%%E#g5f=
zAASmeMbOMty@}o+TwD&OF@m|faUp%PBy>4?o-P+w3d4W1{Jcv@7Rf?XPSK|1j{P_l
zc74Q9>|Kvs#F*JD)iH{npaY&XBH|G$dxZq4-b}5pbDz1Fe&DNU1g7~RF$uA$fgV}M
z8cRW#G1-2+q3du3kk$Kve(uzB<h<YaA*ci2FTp0<ZZjFRZ}8E&-zJT|D{)?S1Gs(N
zWNK|>^(1DZ<(dE<Gp0NS1?_SNZ9EfN;%dE%I=LOzsg%TCyK>XTyOOV?u}RRYu9rFt
zzc`6Zb(3hg8^y*nu)z>YqcuV>0j5;AvabP-_9<A$qMSl!sf~k4hbNbV9WN`SJbM~G
z|7~-+m2e41kp!OaG>Oy(*FWyf;TCQJd2)?)hx7m3J0Hd(zj+Bg9RJrVcP{utr)MwB
zt{3yf?v=>8La*$&0_wStp)G60K`^iPn?-lAhh~XZpEInOe)qkp$e1%Me(N1W<{_|d
z)+;exu3EVeP&wf%u2(gBFi`A!Gnp_aQ~4w@W;4^n`MQMt8=1Fo{I|g`cqDonCqOo@
z15{LVN5F)CiEZ5Eb>0O1_G6(J*bCjWezbXfS9Ljk=fx-5Gq_Idey(JCAi^a~ZxJP4
z4HW=DaaJ$pb}u%nx9Mp`b9>Vh1>!FZz-Fq}kdr&6@XcoUMq{M2RfGH^T%(m)gVmcv
zNp<<5r(IZ?URbixnZo1nxyhdGGB5=Xjk=(K!g@$?pV(`WyTk^fBTs$^yfC<6BvGuj
znEk?{Ur5tHNYf5?a$*Mzj9D(Rb;V0=Bi*1|W^4cvz6z+vp2u&BXV0uG7rta3u|PdP
zmb&I6pZE^HGd9N6kUefO5ZpMqLWaNrHus_!F~fbGffu#X5zYEPZXEfPfHqi{$!m3<
zvu^VD?>n>K{`A@Y1)vP@7ysTc8QMYuo$5c^=F9i+SuQ2&l(q%@2AF`Z5-gDHg1`CZ
z7aF%x<$8%gRU`jg6Q;c>z>Cb`7kmFMZTM=cSGG~oE1M%ss*ioZ9-EX<0SYS`tZ6uR
zDTZ^hr?yU>SaKOBNz9+xFurYW;-2<{;E-SaVEQ>Ua!_w!+I=Z6M|eDg^sXg%;?gMy
zGrYIjovU{GYnfUAs<_=-```<d$gamR(q|j8bP1Lgy~GkH8x^=>#k8eo;RY~}2<D@2
zAyWVn(Q#)CDh|9NRdsFhN&uT3QKoH@3^K3_06W`phU5@dDt6deyW(+Le=rs9Tj+7(
z^fE1&(j_1%nD~hj5PC%~+yxOWLeD?lk^W_$;j_m%AgTIH`_@zNyn0C*fEJZU>&=Tq
zn#!tSixeuOkRrg=yXegwg5fforJK5}frWzY{dc)qAC_PG$yg?h9ZN8D)z!FnGs>aR
zgUY4UYWm?^xW|uh@XWXBSA=)Xmirb?gVXB`S5QOD#&6Eg@~wZ~g}(;LBB&wog6C$T
z<3GWR64_rMB;TROn3%id8O3XEZadUjbW*zKQeG8+;%;qP$KAUh8v3qT-d)NI%mA4N
zn;{e8!=~hV8fobuXjtZF5`dxuSr1SI)gTy5cr4{K@Y#DHFT8x~eynfDyWrhO>iU~{
z*g@DE_LM8wt7Kbhw<d$EOoY>79RU29rH@J<nM=04v0t<ghV<OTJc{0G6aRS}uV=%_
zZaH}=+WQwB*L8_$Fg?|Ii^gaaKtILqV4e*cV+gucKTvgEODm}WsJ=M0JP2f=cV|ZN
zQZL=KxbzTr#pKcywGrN((WWko4|fmJcF3!Vt^}Fm<C;=JzTGC4W`shnTR_U+*U>33
z#OJ4go4d5i&z({u`c2_rw+f(>>AMp|)*U6e_u~Kz3=%3=mZ)`LD#Znx?76=5;qdYI
zY?Au{g67Lt^3G3KYvC@OcPK0$w279?n}AOEWr{wQQywS^UHbK7@c#7__pLu3&VTzY
zN@_&N=zLdelug}br&n8;*kr~1*`;Q&TJ=4x1@VU@9xnx?z?h;7NB8%Jspp2@KU)z2
z$&HG&*0Q8uiNGWQ_n8>Q8l(2~3(Bwt1391hJJoV`EJ=eMSy$NHQZPNUuiTi7<UrEh
zZ!1JO%JbIs9R#L<?8_!uXh<vaaIdmqbronI(jR;kWZ?u5{@^Ak#JFne`rd<tC$rbx
zzG|MR*;6^){~@yePgtwMbv&)RPxXJi`qy%kgPk|(QNR!>D8NL<p2{TxN+2-iX@Cb8
zk%A^nA)ZL#F_A(vk@!l9TuI4?Mwpoq-u4kq1dNKw@5BLn+9+}EBFc;arF!5I1z~25
zFv`X@8xsI-T~e?h4q<eC@BrW|r&m#*@QD}cKz3l-`P>umjY#HnITGi~&bLMy&~mRV
zJTpDvBiG`ymu+nGXv7dV!XhRqG7ivuFH;geV<n)NMH4_kT~>H_d2XDmi|_O0NN3-T
zK<1zeq_<dT7g1e6LTUo$PdyoHty4Hh1WjwI0xL8ic$`F`h8Pf^9(<a2Q6wE(0IGWq
zOdHa)y{+3OCYq8wPsf20JOHj35PX+Z;eZ%|^QV9A%mN+hC9uUiTq}-+P`Az6yH_4o
z19pHA#p}V;=VP7$N(d9aDZS~P>iKl~h}6LYNrU=l;W3U3f+*FCsHU4ECK_}Rzo?+P
z-aV~%E#fXs1Pfin0Uem48~HH(IKX3uXwI-_l-5P<m~E96;1|V#9#lI=!aZR>ML>W~
zC1)RB>0`dq<P2#L8&Uf;%j-i%8S#kZD%RHmeP3Gj?DcB@IyZw=$=%Z7(G#x$$5h7-
z&UC5OURhc~6n8MtG>R|)kSo&CKg;9bvO>V_o<%wrVcBZUk(KA`U`Y}%g*EX^s5EOn
zE0-6yT;QAO#m(Txw^cfH3I_5lthPyfBtU%=eksxyI}{S*QP%(l2PyxYTlfggCgm%A
zZ5&QI#o#lFZ7<ULc2D85lVNB&3khGgq;Az_i{SH@-3%*^LNf1pUwu?H%lPW{Q-s<^
zTm_&2yXTLX8BLm}9<sU&ecbW*sjtZFQ?~!g`)XO}PQBcGqT-K-Z{A$+Npy~IUtSkN
z;Xj9F8u@w6PJo)xW#5TNNb{$3+;zoE|608NvG)d@9c>H5Lq}X!*Zree_m6pT&P)I%
zO}#aEM>F~r%Lf@({&k<m+SFh$xZUX-834$`DGo~s5ua~!$g&9C4hy7>!!KoN=7Pat
z&6Ob6x-C?|_w`rlEt5$h;^Kgk4Up(Al^DXqK6X1QFzCYM#Vvv${%xCCd`99laTb#8
zlR2|1+dpSK_=_JlN?a5>A8ols_<j>KirdF5Cy|WeuIG|MC}Iym5$pB{ZY^sjQ2b17
zBoOTghWik1XWmyjnD*V~qUBC~_v$W7j11G1M>I1}YVoTI7FrkN%uwxl&$Q%(CO+Sj
zgKW<g*0jd@i={uokwC}t(+=t+B<vD$z@`1(9z|CF>0527=%FKo>?4_3Wnh`#Z_0km
zI%VTukA<q_Jy$uOM9W+3t1*PzKj(I9KuL4U!xE^%pWKm=0GwHx#77X?EdEAVGVtjA
zM&J3ZxL_l-5JK<!oP_vn*l=B4>#Vqd;8;6`0F_I+@f9I6qpoQ_u@!bqd>3ZK_)zxI
z;m4fO@rcPx;?kG)ie-TyAn?br{hk=?TeKMk=cc9fagyJ>=l0;7*^QlR^N4eDR}~a*
zM_ufEcllbH=#N=U6{uD*J!%=>uLW&DBhDxM7YvW2cki5T*US3JHqCh<>vE0{ac45=
zA)joSEYq5!y8TO3>2t>W*;u~MO-K(rd!8EYGTsN`DRWxR^$cyMf0Oq8<?4xqR?k&i
zLvw08=r#T(!RLZ#vDe86H!fb(ZIq3w(aE&VPRh>XHh8t))w-)I011Lm%n3^MkUdXR
zg)jva?K#HISJD$XE-X$RUTI=(Cc)BB`<5?IOYi;kTCXO&wKgrzbPnMPre>OT22}FG
zVhUN|xur8?#Jlv|?}{kCm9}m#Zg5Lb$!Z5O5M<8~s5|SfZ0*nbIq{mQVQgk-N;{rX
zjj-)r@+V)nNrTy)Z=x}~yL9=oHSr0;{ihP1==F(iYvQsCZPMQfcV$OEIkCFn(Q%Ca
z)Y3cb0kN!|>z$Q8M+`1wMTYx&q#df`mcd(Q$LG<^9r3tayP*NaP1$$}2E8iV{nEtr
z(H`{<MbQtM-=o4hl6Dkp62v|`(VvnP+m&6~p=khr1y%c*YIpq&PB27P^pkVR2s*!K
zg?(2}|4q*~w_NojV!wq5HD_M2;NQv4q&|^uoT~vnzDfgb4>XQ>JMG2OzWAUPPqm&0
zMiKvY(fZH-p<RS-OrPz8**^oPSn&0bGFF({YqA%cQb1@(gfKqrjq-!X5*E;yNVT5~
zcC@**7vOhOXt5!_{|ayZt2+<a_=g;j(CBVH$*lZ;X+;tr!+zZids+epPUgQ#EvlLd
zgXX8HSIjTpmVKDZB6F)BU_&E!{J^O&dAB`&iysTdXkJDhc$WQ>kp25_HUe0FWBO;#
ze*zR>|DuCIjL#&<#<i!M8+V)7<%<b&VH9~{@B>H>B+zElcs_8vxA^{mh#UKiS_fpi
z&lP(yB^JiLFWM=zd-rnj{_jna3=c&MSXYIov-wv4bwNIzz4!#igVC~`_^sAms`VK-
zscH}k72?ziRJw+~kW7q-R5EY-#{CAPbTyYU^dfmAN3zwo98pszZOu1AuY(kj+qGxw
zmZ!guz4s;E)}4CXrT1Mkc0GN>j#o@p%QMRNM&<{6*N!D^@7sO>{;o7|x+aw^<ric_
zQV||NSQgsdz&vDw9{`2I%imy|%_g8Rc*<s2F*M@;%rx@?gUB5jJo4yg_^d4FFJb<}
zKmCVQWG3x}-Jazn{PRT*9bqgb5&#=3eDxI+3Zj|JnG-<IFs!#8@R#rM-{1RTKKb2)
zlzd6kkAL-7|NHCz+sC59Bl4r7y`l8qe)xCa;J@B0o9ldVL*9sBh4SBi<Ntd7zkj&9
zh}RoB{*#&ikKgd$KHUEwzHA~l9@Hvq<qaA*eQg)ZnveNoEwE-tf2PNm*T4bE2`dkX
zc(MkK$x6W#GzoAKB$28^Ev#{?SM=Xpwk`>9Lj{X_>K^@L{Ux&D->BVaat5Q?D)o*5
zp)5TP(91;tJXsfxg2_OApjR9|T5H-tlz@gS2Z(1H76D$LE)`jkC3kHDD9I$!fZXkp
z2~Lj`RM|p6e5b8*^9D0`{1<5eMv%B#>^V@-2SzChTcjDY@?FM*Mlkk)v54(EK>Q%8
zQ#q(-n*~f&p0xK~6ih?#(ovhDRG;2i08K4R9hd~F4rxF1myHgZXtIIIuX=Ypcnx-N
zVm-G5=o>-ME150=gx+>r?fopk3{@;s9Tbj$hA7z=47JJvC5kjZlN-=_G(T9)K3URf
z=;(QtaWcaWUU8(Q_bBvQAL#GWe(n~zWr14wV+vR4V|O5QVE^s#8yaF(Fb&nftli=o
z9!pvo$)ut)(IewzRMNY7>RFIPvu@@ZZ{B!o+}}J9d{@B*e>-VY_{VGf{V#2h$N8C8
zgW-;SprJ^RI$cyxDCh&bR$@FL83*?rp$WAZV{o1Nq08-fZ`lcaB6WyZ)1MIMfo7$u
z0F=qg6HLOJu&#_6eXgO{N}<x%@}XPodfi@mv1*SysAC;lpo^?~FdclT!Ugbmb+BH-
zTkbqry*vjlBlp&ZC{?FO{br;yh(3?9y}Amu5-o+r`u@3<pjBXRX6pcYvsf}2Uocr9
zJ4a%c(<LNAzttPkQI!p%qU+}+*Y7|KyWX-w`Lk$mAXh2R_XEQz(*Qu8!mm+Ob!PcB
zd0@#N`LJ#?(!T4F^71F^+Gp=VZ*7BZvEGFWK+#%y0HV;J(5(WHwVuX62gyp{rxpV~
zL!zdMbg#9fg&{~nQ@3houdK-E@*R`hGgkHEQFxYP)EJ;o$pe;w0^p}!$YXa<cfOc8
z`S7u?0j!rrFzD~m5g03%ma>Iomjuufi9M#r>*uzfs_BUj(3z?EolG<9FAoB}3xt7{
z)erG>eV<on3}31NobD}1>uL>v2X#<dtGV<i&aV5cdo8{WBvra1b?^|W#M?le*btg}
z@Xk{5x^f=i6c!UrM;gXJ(-JB@QU%J*e(>SqQTRyzvlUBeWqZ&5ls+J3+iaOX=g88d
zF-H|7hrBEW_H%0>#83x?uJdu8O@lp<`5ZE>v9ka4J3Bo*{C*=Yfl5FaeoFfaYOKqW
zJ3DMQwEl7*{p5V@^@fO{SGTJ&v=x4$NI%EB+1Vx!lun9j`_G&AyWQ#qjm4)f^B!jG
zmE}a50tKF%*&;Xsqcwhm=W!0pkFmwzkH>-}biPx-aH)Vvs@5VOht>om#B8poH^<es
z8H3)v^f6w8Ni2|o2%myD!#l8xf_W?)E5L2WV-jMk_!O1gD+r`Y)L{hF4_MLwGHVUm
z-p7S0O2mW=Pi16}IEWwN4?Q#UYQL7?Otp$VrWg*yyr3-F2imcoeJ+i-8cwK{baKvf
zoB%T&2Sl5+?_Iu8Npk$kV}{uHKsH`A_t>1*OIJ1`4XEw(a%zAWpbt3BlI0cV-e<+@
zaruDdr^@Sh*r|W4GEZ=nB1(5B!~+kp1}5_ECvrF&Fl&y1hdhNhZd2`yJS`4pGT<OR
zpv#hoSlyDijX%#!{*!}f{!uMo3nv87mExa`%}a@#f+1?{%X>i9!~e+J=n((#ipI1n
zFi$fKCn!Ivg=Cbv7q9ceC&MN#VJ&tpH4sME?@jofIVOzsBpsdP1#9TG*H#11)2!$c
zSV+d^oo75-$3*WvimK<7G7g&Ax0nr|N~5fhn~hAO6z7Y$EPor5a1wvR_<-rD_r$i^
zP!ur&F<%Z<-GBsiM*7D?AFVPepwR^B1W%_#p%D)`fWT(2q(8q8$c()7e21h7fi9i2
zJ@!(e9%1xOO_O8|Sd^BFJ0RedS+}gG#_6NlUoRvTcCe886`u|9{^MNMivI*q)g)|-
zqDQw=`l~FCLB!5otas4Pr|{Xm0E~iJy(a6_7rx#$_dFFLz6yA-MP@0Gfz87Bz>0st
zK2q**%(rCq&ikb>*;k_50SPPUq^k;tU{}k?5<xd5X+t{28Eq^U&wLor6MX=5OU(};
z5Fm)*I7|&A(71vKk=15nrAHpeWF{O(E~wN|&xp=kYEF9++_2|FW=(X*Odv>ljv1e(
zC`~j3`=CpAKnk@*nF;1!8DL5GGimCqI>J&{CD@O;Z?QE|b3U2uu6kHs4<yL_O4kd&
zi>1PJv&KGxG4eAW4Afu>!%|sfSa26ws7&LM-zSvhO-)JyxD5aeRg1l6uj?<aRZn32
zj?mm~4{!$ZiFLmMh)Y*m6A?`&x@2n51H2RER*RCJh7x+Yb12>vX(i?LWT>em_S|ua
zz>@T5PhfDdY++1(Fn=qZv~@B*cw=ZYrUBkqUKH)S?^5h#m@XbVV{rPJ1Yg{D9<kl4
z?gEz?4q3>LAQ)+{RTne%&2T0^$hSiXaN4nVSQ$==HcS3a9;#(a14>Vqk25Y37qE_P
z-mT@^tsK<<vAX=D)}oaJg;}rQ-sOLuNcAwX0W3x#7A$4)oF~)Qc;PQNYOU%4^11Dp
zb4JFW?G2Bx3V7|%cZYqJ<6&Bq=_r5#Qov}<3Lyxl5M#5M&#go)M~itRHcjM?N~A4i
zsNAumG{APafr+p)UcpxO=%;dc%QqP}ohg1l41J2mNh5C~D$By7+5~Zi=1jK(ya3rO
z5q2Yy+=YCSY&iH3Z!-q_&Si`ECFvF_hBU`1>GI@@1sC+kEbnFg2_^c+CAX30TR=$l
za>xf9Pu+Qg+X?Ioeu>JMK-C}>+z1swdb;9#!93~qMe{qekrDUfT3c2Y5~A>{m_uJ<
zwLO6J<oj=a0xC!o*$zL#q!`7Wc60z?rk}-m(#=XvF@f!C4WFa;FrlMHxhV_&&>-%<
z>qeVldtVPva5ZfCk9D!<hQl{>vi*&$-UN4WwRk()Z+gY#7TSAs>bE>D@SVJ(T*V%G
zS2=moME-VB`e*aMyCLeJ=9i!t`e#Fsjej2bCAxbD$RNF<emad<=gjWd)~1>>#%yE1
z31}RS4AsY@s-l&shM#LE19Zo#2B3vD##YjR_griu=I!E*RR`mD8RVmB#$3$bMhYu+
zg89Q$s-JV;>T$+cxW}k?nwW7JId2Q3X}e+~4Acz|*W4OXP*2`Rp1O-hl_?&n26X^z
zW4M~M@7Gu9vy<ukXBXzZz?k0;cOjsC25Yb?{_-U>aJWlW;kaT;d{GLV^>Nkl%hCR1
zdP}(1jv0GB0L<q$zCaG^?T_w=hC}irG_JM=`=a)TB?DPxL|b&pGZKyGK+M`*>${kh
z(5(@QeJV7Z#XNZ>FW`GMuGFj+5$|9JF)ANaf(;Zh$gWw(2EWI|3eOm{5F^Ps4S4}6
z(tO&hoD|i>Jk)w1>ujcW+Y!a3uL~vy_smjuWA2PeW0Y01GTwhZZ6CX<5W!kX?Ifp*
zmB2XX3yg5?6P^l&1#Gl>W25EqOQ7m%5fqf$Ra!2n69+#hLuhVheNWOp+>=%dZBzKM
zr+|IQt^3`ee?;@GY?3i!0~(Fm1>+Gxx+dm6t-}+<OSYK%F-q`J4zzKh*Ts9qrqSl<
zpWE^Qm*W&=aR{JWwmGJB#*){5K%|@_k{%#b93j2*Mtk17f<FZQCDKytkyNk1Xn!JV
z02u`jB{^h1Ul3b{&28bA7^SoC@^{IanD2pp)mv)hE)Uw+&@}R<UV(c(SzWna6;n%x
zGe;N1TI*%{EiQwpB5&Rgqm}2X_0udz+$?U$)N;Sos<1Z(a+K{`Q*;J#dmX0-k+Q7c
zq01{OW>a0?Z&}=AoKy2dq1n*a)bYj5(-y8}7DpkCOLsQEI(<X_$0qHM7WOT4VtBTs
z-~QuJ`6l8|Q~QfRSK!vd>%cxB=t^C_ZB+x5om*~$>e|tt$TG>^y5I2uO3`f83kd<`
zi!i6v;M^M!w-Pn>CEPA3sRky*pqQ+N{8*M?@?B%Fy`0=7gyt8P3QLWV!?@R`Dr!UH
z5%lt+I;Rg<mAauCmYdaQL5{hj55Kqa7ZLH&FLU9orgrv>RBBzbn59Y4e3F-!N;^S4
zAJb#VuKkoDCSgQA(#9M#9kxh33hS$!FeVr?%&jadC4g4;tk<GDlp_Rto^a_}X}8qz
z%B%PXk3#te97uL3ju8T6jO3RH=tvJ?bYb6PU__98r&L8OI{@ytc<(^2n??2?m3;pM
z`R>d2jk_`NG0T=+sCHyqL@Aj$Melk3DX3YA|3G#VGxUR#<dI=YcELt#IiGK$?E*2n
z98mhV6h+6t0JskkL>Zcz`5*UuTZJ{Rnx1c51r1)$zR5kkYvVJZ)rwxfvL<52vzDPj
z$;zymhZK34dG!+feT8BCBI}7mmxdx^+*@<RUH*o~$k;|mKH=!8-a_xBpN8vlArY{Q
zH^2}pWi(PI#Z~)To_Au)r0@NPnHF7&t;d>;>ELe8)>{_M)XVysGRlqsKb4lS*Fwh5
zEYR|%i}LLY>TF6p<2MbpMCCmnogXVjpl7pA8ML|&B{~ydT>)(%cLE}!e|;q!rqJAV
zIG0lV4(aE30?Z>R^L~9-QL)HEio;?`^O*V+rG_5n7KZ&Ri@DXUWYpD_hk=2hqeEqo
zF48SVThjS7yUm&U8gD3G=zqhKhrm+YCT}ml;b>er<N!1&7MrVe>pu2KLwiG$+NaEZ
zy9LIDqp1WMzUmwF;Z}j;u_i=*u>6zzG9ZdTB!Iu6VBrjC7V4<dYEoQ3zWAFSc}G!Z
zvRL{F)3raAeIg|uT6xm^deixU3k3ept)!tI6DeSr6ELQ*^B!nSMU=f)I+Z<|pIrg8
zj_xfh>GWI&i~uE)6Cupv`<=>&4~2u)^6fPAIuehml(E+WV;ABcuo2=d!^(ujZk|?6
zM=1j``DbD_<XTt+5|1`VJdM|6_GD-7F_GM|j4*##G}aMkUPz53LRx<##gX8YjruBS
zX3inm2j!PI_vKR&Hac{iq1(u>-=614?dHusiOj$tu{1Ed9maV%qBl}YY-vBC>epNf
zT?w^wg*CSbt&`xWYBu~#h_B8$Go>5Ge;?=xCya^h_}*Knq)_#(U?YT9(XVcralK8G
zZRi;+h%z`a02cnM$tS@5W=FvH+r|P4F$NE-f#JhqG4l5sB|(0jVZZKyDFeAl#Br<x
znwhVKEOL=>?oGBp+m9Or6hU%9BR~!4*(LEQkwupNw8`f-ev80}o|e~<p)rvk#s*Dp
z-fdAY3g*<eY7gJ}N7Br8ZMaJ7E!oRcMhZ9NDyF4!p3JbKKkR}1r8`N8QVlcFCI$Nx
zVS*D2y0dnD&S37Y=89uvnTK|;XnHki%*(^nfSJ(mp(jRC1qv>{bF!J!$Y4}Tg3}Fw
zIY7C-B#_9|7rvv*XDYq+0wUKsU!xn*KTL}Pcr^uNsrE0agdXXzzYyF<4a4hVEveu)
zMK9(>wZ*4pTJ9a6yg40{K6xT@l|6G3gl?p_n$zjVc!fLolZ729zEPDZm=tCt+NSQ^
zIVglQ=cVgx+g_nwQ=mEZS)*TjwHH9vF6h^Ay4N}N<alpOd^#mJ@MLf~`?A)-v>U}1
zXAjP;ZrN}=RqCwq{C+2By9S+{M~l>csr?<?5HE9b#WVcR)vXmPgGI|@1m)4ct;_k$
zUN|i4(T`F?m;()>f@+hzru>>&0utFs6%+Lyn{ZgT>q3msrh+%-KG7OlzL<y7d`68}
zq(Xe7+{E)46Ghc%!FdzIu`>7Gfr&unLlnIhVF$ZfL42A)RT^j<j(-=O{~3w=Nd-@{
z=sH?Tz_!FLKRYc^Y9pxbj^1Q@_fjgnO->%hE)PS<W7MXjOd^7E?*=hwqqXxLMs+u{
zMd~;+1p7>=5f>weL!3U*-S<|rc=JLoV(BOrVN8t>quNo%3YkVti^+O-O-tc|`A46h
z2K&P3@?o4VDg*FIPN&NV+peR~`W*qb<fWxJb1li!cULZPy?Bp(a6btDUeI?<G_DGX
zpU|@O{ALA8_8~}+HJ+|UN<M}=KAGOTEfO@QT|3KT-#!pzm6(vR`_7UIQ5cyghWtW>
zcoM1GO0#ErgV>`0)|mfHwH(dbjqF>dMubwq$BVJKnFC)vJidtvHW!{ehP|(bnYb9x
zR~4;$&x^}r!sRhG@&L6PdTvq#3w;Oct;R067TPQimJ5@5_*<I6R?1BikNtr2YWZL#
zSD1Y!PeMU%hKu$(Vwq}8(0o!N!hg?~w{n{<z!cu(3wP9kZ<)ZSjC;Qd;`9hR>$-35
zc)<$g(Y$sN%yw7Wi(&lrr_Nfxy<oU%ds*1p$HGlQ>ch%=u$3<WjlL^xkt;DX?p+iu
zC7yoWtl<&Zc-E@aT};~=<<a7WMN3|yeOB|9R=bt+=(ymM^x}fW{Z@4AN5j8t8lT~N
zqGZa5|832klcg^;x|f)zuQTKw75Y6~m(rF7hKw|z&PIAZ9Dh^j64>Txe(iM7znRS1
zHXM7SpM&-(X;C31Ju2?0$H<^1=gB7dRg>%@mJ~$C&A4Hn6ZxG|B=Yg>^5gYRn-mn@
zgUx)Mx_lW9YnsU)S0<s>>v}Z9GLb29%#A*>K|7}58l!7Bn|aLMVrB26M;0B1!cqe+
zf7y*FMdOObqClf((`8Njk=J)KJiHv>j|QVHD?j$KbS~wrZk&|s#Jjj$E{%F0<{ofB
z(Mx43KW=mS`Rb^a>cR9?lz<*}JS5NIC)atk;dU??VtMD~v{)QwcI4$&{xOMZ_X+4|
zQ|D!87P6Zdg1!ecNZ3ri5JnTtAu~rKt<}bB_{AfB5f6PGYj4nup8#?eCuTRN8qEW_
z?1;7md9>E*TnYa%U$G)<XBU&lO=M-%@Xyr*-?`*hYGRzkAG*9;g5Q(%-v1D*SLwr5
zEZ1c&;dXUDZG^d(JsdM_#3HVUu16v-h^!eA`N1bE9hik7kHnYz`gQ{j8A~lDm;565
ze}6+E*<Kv3>PqU23k+S5^ZlUv@5uZA;3XdJ;HxaspXrSJvjlWUl0y?2wa|5)i;wRp
z>hs`IP&bXyZ<TNvkGY_jIJJ32u^+z54KCk&H9La2U-gf;h@AmgDO&63Wk6zC`G9Qx
z0BFzna~?!msSKzV&XToh!nh;zr0zsAWE!Yt(eeJ$ZzU=HrhIf?)8eILblFGTg;i|~
z6<wbi5rQpds#tHqY4FuxWnOg6_G?V*NcVf|Gn04gH?w%HmG1<5td<Y4^(j54P0kHm
z2I@s~Ev&2PHU1%ijwbg^65QAm^-czI8zMMgFg9Mx%B?@js0FR%Q&;*pKWAbW5k9l|
zxy~H`oH(+(;`eX9d#}3m@Tg8Q%YJtT(t<qE$LbFZ$mthl?_~9nMQCBXMO{4GGz-gC
zR5^V_My%d8*D7B9rM9S>cRce-y_?hb4v&7+wn<kKd{roipFPVmhjJFz7nz&O$yywf
zy)&z4X^Al30VQ7|GT(%#m-oXP$r6)#MBFN5&`51^$#i|`>I)2@%8*<9(L0sztuDI`
z^-#SVT2>+frpT+ar@KFDt<NvW$*6ulL4wnnt3ib=ChmQHwo@ljOt*Y$pl>M7Zp&!>
z0d-_glq8&UeQmZDN5MWI&~OuE#pR|NC4KKv4XzJw>-4xG2D---*Uo(%wA0D~eW54c
zY18^EY`&i1__hG&peV1?Ehml2;Mjpw>7YThjq!gy?f>B{iim&KG+}k9_wkQY=wSt1
zhAf;z>TuQtJx(_ig&9hGzzpBeM5kV@OrtvO@EqtkDnw<%vM`pZ5X^7!ra;f~Sm31w
zx>7kt?`G$dfz_-Hn%>L(Kscl~{#`E}gP%O!WQ=xJiuoxj2_l>+!YJXheDb(Adc#DM
zyvV66fXe2%0NGJLNHzQl*Wm2Tg*=3FTVTMa`6652qwE~%)5S}4SFHJw=PQh-?$&>X
z>Pa<tA*Ve0i-QZ2P+q&V4Cruh*u+~;yvD<PSyT5sEh;w@y^sM`BfwNo@yMcoOtZr4
zfvFc4k)Odd2t7!hPs1L>L2kG#ENgt$IJP1QiM$_*@Dmu~FG9x6C+@l!ll9Vi!5(Hk
z7K;!S)U312TLwObB5^xg#=aR~R*y!GuTk7TkjtGGC7Q;(NxhbfeVqh>^_14Z6i%c~
z>mbSC=c|p@_h~^T)H4+X-61$;ILo#qHK%_+aC;`3??UybotNA1k#&PW+8%Gn;~E*`
z0oRWis=5L{%htagFQ<ogl`~td*x-)jco8}QYxC8A{QgUj54~T15|YY0ou+;#i+3QM
zNH=l9@+8|jgWcu|HtsX|?}NAJ(bB%+g(XcysyXzP^{M;D+sQ*_w_V3bi~f$zd;_RQ
zdM>r=ul``LR7mh*I)uU<L88Cwi05*lEq!pr4#YYs`$I6h?f#<9V8Q)Cdq#Akp3~Qx
zAG%~_3Z4-0_GTQ9%|{LIhO(Ua?7U@G1^ky8bsHPGd^hM#816*C3|5gN6n75ZzCvBI
zh<}2tvzRi*34MhpW8LMyUu&jp{-8mL!(up7-`c_p^NPMj0YsEnxjZZOUZrt$k@2M*
ziS(X4mJ|49^fA-z-QgZz*K&{P_>h|wIr$r{1QZBT4PuXowil}K-$(*SWHOhdePuA#
zHs-nbx>SVdE0`a;7!{zO)3hB}KMtz2^b!C_JU_smlr>(YpYb9dzVZ|HqDvI}1W>_^
zZNbtw*YqK1PSK;aVSvMbP^kfkX%>j$Vn;+l<U?A`<E{$d`Ky^-kb!lvyx7OiLM{sy
z@lGGu82aDoe$WF6iK!*4#2Rb)*ZL@Q8q|&ihUBzs)ck0_{p3Ja2?&#=#B3b|20WSb
zSkeU^_iZ0)&Y;nPuM-|mSSFBgB&Dm?^J^z{Q&hzm(EevQv396wNR5u7&3?&|in5-)
zX#q0Pf%@U~|L|o0Nm=tf#(Udyb%pNFlOypl9z>VGId{Xb9QV-4H6wm-ME(OHuL*PC
zGze0;JKBt)c~JWgwrp^<!*>~G4a&`Vf>x38;uK*t{%54GG}R`u!C@b>s2}w7Th%G#
zj4ql6=#cGL&JYRQhkuCf<w#J7qfrq;bIxUqM$O)9bTsT*`~&2^$8k6O-aFXsT%X;<
z_050=thTS|VC0c9`YV@g>cWx`T=jY-FEV>zjaj27Ec!GIPz5$kwuObS!{ZPIl!z)y
zL@4EsalfQ1K-#Dfgj5J`7Y$ntsD$KN2{zYa3@n<jp({q5s6H4dX>rA=b<bbEl(U%%
zkzZmz=OYPSs}aZ)D#TvoXbOLm1hs(0ZE@6+37k`=b(Lnz6rLv^j{PP4L|^3kbAHR`
zCx+;tE~I9PtHdl2ZN_*gDosQk$I(~YCn*vpb#~UUO$!F)JOUD??o;YsQ+jnexP)MF
z3nLf3Y-wkgY;?i9)h7TeJ$dxx%IW71AGr}8)QIZ&Ls<QIyzhK6RzjzC*>IhU8Z0zq
zCjNYX+_piE<>PY0h<-ATa4depnH{K=2%MHyxMOTY_f=u8hBr}225R`^O=(-XZ#=TU
zSC*<AL4?Ujp8H6D`u(6p<D#pe>hyb}j(f@Up}wCc{TLDKI%3{rw?E<M2lvyy3n8dU
zPbc;m6OR-3<&JD8zVEutg+ODk<#Al=N3?FSe<^n0)4GI9@|-lP{NtGT6{_+OI8%6$
zl4kKhDQ<}?<v~8u*aSpJiEAbHsWzH2$DcLQclp;BUf$vub>yQ5cE=rYEA;Ssn_@Rw
za2{;5^Oi!``&R4)zdfV@I(Y<{$!?w(SRq6;!|A>XoE7CCCE!5~p#QAFmgUN`!|^;=
zGwG?}h`jjEudl~cAa@-cIyiHj-z`BM8eD@t96D?n_pgiW39xmTj!5UEY>2#jA2}2H
zm>hm%G-hTUvM0kAIW5BhM3ERmTL~{LgWSqC!3`Mp<o!^LojlIIurA8c97l8mdG+Dp
zO+Ql8?wQ}RV(2BCmnFzu>-aB|<B=O7xwW9EqJh6JcL#8`(NklUUAzelp%FUvTT8Zr
za|N=mA<E#uv>vEU%)@pE^^37EFVhh@W_h7<Zfit0g3DFwVv;Q~(i5(s8h*w5*9DIV
zH0vbZaQUc{;>clF{N`ur^zDf8VO=G<R}<GM;vnDW1K?V7nh-4h-aa35v1WnwiV^rh
zQX}weh_5-rnofMDjqj~7RGN;4nZW$uoC5oI`t6%4w4x4cKt%J5X?2A!5z-0=;|29|
z;S_K?xcFmi{;5;=bC>w{TGHPF(BNMmkNvZbOmxL}pX}~D08HWQAAklYo++uZl8uQf
zaR<m?7&sn;74cX{zC)2GYo0GaMhJHsOHH}*RV*QUhPUJ9U}b%|=}ztKrC$taoIW_`
zh%+}3CfdX?17r=zI+OTGWZFX)C%82zi&c}1W%~zfa`{(2E4v}zH{#SJSZo+dNqIA4
z9%CJ~sYhg3zW_HpZzuQc(UN6*%5RP{2gx>4&<~eK*_GwgeanUhZR$3Aj4vDMkQwH3
zO|uY?!C`MMaR!MM1PeRbLI3>$_sJu_^#kuvn<tXd_uw=04fjK-uqF>{<#vB{)-P_r
zMr=A*{-Uy!2i+M+zxA!Mez$OJ&5=F>&5_BAB@HbT>XMZvE05yPe=56V74|5k<V!i?
z7!nkDjA_3WB^6_<x<PwSRQSD_r0K)hVEzXcvrZYu25WCq?CYz`i#&U@NBI|i+*p28
zOU_TMflKYoJqAiWo~T<>*eQ8lPmT2Dh05&X+IIy*>P3-aq0hs<5fY+XY)mpR^C{~l
znO1}H2c{}~L&z~O{p~)%Bg!EFG5-vYE&5Vmw42Xxt*|C?_V&CAG7i2P!5`kW4>q7o
z7orhzYjxrvG}55-%C_{7@@NWC3}4KAl|8%*!W)HG#mol<nUQx3K?`!2q-Vr@ioq8+
zB5w?>XC-I3yfmk>cDPk9QxbAM=Agr6mVMrqH}w_lY1d+E2*%OnxR+^8y@kYyxtFpa
z*mGol%qU%Ypyol8_7Z`g_9wy3?^^0#eLPP#TPxQG4}6O2_Bt3I`jd?-n>Dp45sp23
z9Nvs&HA>4FEihM;V>9MLw)w8w8h7r~KOQ|}1WW8tchuMWFzq-37IJd?iw+QZ4B8XT
zX{7V+h-_PF<#|Ol31~{Zpwo9TEzMIK+@|-JzLEQiR-DldumbazH+tE;BrSD$w;6lv
zS_rnkfXgxE(4f`fbNbht9gBaaJ2}~uU%uX-XQn$I8LW{Lh+CwUHZ{e*5HC_T$$A3H
zDhY{qH!NYe@Op%E#9|PI8fLTamb}s@$~-o$+Ge$wbiOad6NA3+cFth;m$GHb%Di>X
zVO9#9%~g~ywTRwa`=Y!?3IC#r;j#v|(w0%jkhUxeWI^#Wg5;n@ppV9E3P|x(&%ESB
z2YUY|^Ku-OeW{EX!C{7zrh=WMBv->Joj8r8EZGpB&6#4Z@*5M22h+wB^Wx$uGau*y
z|A;0!@eV@yJ}-h-5sfQ5`+dy5r0c!;#fcH&a&6>OgS$osvPqMu(P8X8q6$YwdMFn@
zC`|i2H%1nz660H4&QuRKX9u{OW1$OoP*C(jWwehY^S1{0E@$zTcJ?pC`;?E&e*P{^
zsN?=Zygj&yxR4DJU|7b%76CpmWvSW-;EA_(W5wlz-b$zy0Til>4@o<_9sFHuO(0f%
z<hEBn_kw8){8XR%7P`5qw%(xC5|+w|E22cI5jEZvd?t@YhdPjzniBc5+K1c|Z?!S3
zyP=3H^nMW(^s49Ho*_C{2`>5mo*_omgPU(}fqu=OOvgzdg=a{#$W0=~Cfn9F4|`MT
zVnn_4pi~O~J!M|9E`#SA-XwOgnHQy~x1gM4(Xw63RU)!Gd6N4qrXSx$3!|@g?0}+D
zSibqagif$Z+`8&q9?<*T3f@VSEn0NXNZ8%o@&J;P9I{+Z*y}@2lz>isPSOixsbK+E
zRS-uU@~SoHYbwjjH_wb8GF^({=<(<xj<A~(TAPnP)x?0Yp!}AKLFHv&8>`fbX_b+7
zve1P)O<GEACpABqO~;X>zIe}oXHLY)ZhvK%SCHg3hw(hnP3Ya{pK}6cSPWvRVh~bs
zg<1}1^LBj(#w53N9TbU^@7||p9vOa!ZMJkq_$E(}NX(q;Sn3_=ZHa85uZNEdK1wH@
z$=?6kQ%pwE57gM{J)dcLI{|>5AiLG!BDmhG-gO425|L+ZuN{5ur0P+$WZ9<6+1w+M
z6MQ$ZIL66MOEy9^hV<5YN8)6)p3g;Bt=5wL$*YilNJ5Qnf8emzv*_S$fW*4O$Y@!g
z()C$O2N$num4=OXv%Yp*o2_)AmW10BJ^H={n<>vcSn<10-c43N(4nP|2wZa70nMTk
zqGx_nk4Z~}t<!Eb1F<u=NwL^!>(^SgZhYzsl09%_dRn+`)tKbkn!IIm-;Z0V+{K|2
zx?LUv69R@hb5Va?09%{`6gu6vcB4NOid_tNAkJJPO#89vjl_gDc%F4vxHXQ0RfaVH
zWR#1PD(Sj!b|_o3@))t9Xf`wHV_1;=IBas9vFCQFSSE2?NaP_y8L5#hDhVyr<Zaic
zh2h6q!F$=l-Qd=oh%|0S%}0lhv<RZZsGd?5=FI?~--q)(1xQVeBOnG#H4k-UM&B|k
z-3#~y@5TQ>_KDbY<V&XuD<Lb|Nc%IXXm%Y77P0xX3w;fFY(|&+*wn=jzY@!rNEQ(-
z9(@&lEgBC=q^$?IK`rzbI(BsyWn{-aVok~>1zm{ar_2-4=N*L;xPm!<@g?`xL>Aba
zF5?lETIU342XP0s>G15F0&z^5hvHE$x8gUElF+tz&P=;3Y-+5z;7pgJg!b}+8=VQU
z38lk3J&=X6?9CjzykOZZfO}zMQi~MpwG529v^I_{ijyB$5>1XrAs@y6Os=?6n(oDL
z=VE(tUM*1Ib!A%g118M1Tq-CZa;;C~w8cjd2C3t7tX-Lj!4f51v99SU*me2Xm5MfI
zA&Coi2*KWqdoWDhE7K3MyW`}IS!ZiBOJ(w9&rQ|P3U<wQ-E14iXhp1JZEh8X`ObIQ
z$hW;5Gw%A>)qr_tKs+z+9Bhqt>2xA%MYufy3vz%<uhW3{7}?6N!wj<7b(5d0^2yVX
zH6@A1)JC&RDlxO%caq-Z)x0_XdxjhwZOz<sE1^UJ6JQaFS%_3A>h<&YSkLX}N3;mi
zC8c`2=avxvxnI9u&U$uxKDqSg68p;=nw&j4Ev7h|q9;!bEF`u+`foNL%vLwshvwa-
zU2mJ|q8iI_=>e%SRVewg@uK8TLGH|1;k8cpsp{~{;pgb!*bif!>sNkn;I)_Kl}MY0
zi;2b6@DG<sZ{PC!aqfa|zl(nK)#_qv;wI5dBWZ%$0!T}?ImsV(;0}9`uIAi$@z=E*
z9|er#<ZF=8-qmYRMfb3UOa?9MoJY-CxyNBwn>c%ohu6J;K$mFpT^&n2Trx7@7|lLy
z&OOD<`|5Bd`?8)Q9S<Shb5|Bx<j0K{%W_z2UIWv6Hf9-nSkaAqA_D`hn^zx|tFaQy
zFqsj~FpZoe9(i`6GjuU3MC7D~)5Y!`lVV2WOV`f<826p?NTL8OLw}m%`^0wHtLFaT
zSK+5!H%UHz^L<eEqHb^dsiApL{+=D?Mj3N^uyDeuKLaL%W$@;u+*Oy|qdFfI2k&to
zIeNMe1DbU-A^dDh-5Qd{E_9cyv{i0h@%hw{-S_qIm~M1+Uan8JEm(4>^R(-hMUo5O
zTj!dd(dE<y@l^|hZ#e@^9@Q`M3EC4K^t<T?>;}x2(0n}WPBW+$iL<Jng8e7ERZEUw
z6W(>7i!3E6OL6GF>EJsO_*z>js_V9D58j+=kmwvsl(AUz%A#hK6k4N&r1P5J9NXIb
zQdMz^V8N)kq2O-CKBwtxpQ3m4A`$FIIozvG^L%s3rE5!;qBMi$c{Ro^Nx7lu;<u^J
zePA4cp=;ig^uo>)Qkk-kKulTctk68M*Yv<5YB5`7hdU=~OKRR|%GcCrN<=Ab!&s8t
zG{%&T3lul~{08*<4#%|vi?o^YidU!5vC_zsfzX#}QN>TDD@g(bO%tWl6R~o;4-)=y
z@Bd9`{jYy^2|<$L!qz&Y|D|#H#7e;;hTx12jdBMX4~tOemnq*iooS4{9n^-E^+thV
zEo-{|iHfKuZ*L?s7*xnG{sM;kJMB^;1Nd97J`aER-#vx}_j%1PFOTRK?ZKIr=*WPb
zaes<SW%Fd}h4G|aBC~!9n#&%a>K&?0g>&b`SMQ*2YlDC<Qp)k-?W8fTzfm}>;#2+}
z2|f4^1c?6|auAas=%}>J==$RgKf%sj^&g7nkP6j_XdY5wym)Ilh`_PdgOj-*gk{B0
z>sh@?a@jO0{F9&iUq{UUHURwVLQC{sBD=i~#kc7_Z6wH8NLbqaTRVY*63`Xl+zSo<
zhPwH)8vP$29v^g_KTlfm{I3uAHy7%U*ZRjmm>;Ln^*<d9|M~*|`=<SmKYuv{AjeN{
z>HQtL{1<QX5C?9PLfBBpUoX_Z_)QtS&%kaV!Voq1cRH89xDd3Qpc06gDtG)}YyU5C
zk3ZYuPfx+VN)qz6{LjzD_xTbK{NJ<&QIg7#AnBNmKSJzCs+v&k4qj9!=P*xyy+%}l
z^+d4I|MA{}iQqC`Ui0StbItlslDUKQ&OV?lNE8MxPG=y@+qt*Mx3sX_e1T~PWNfkw
zl>ikP>4R4xl@y7Frc+RgrDum$`pa~nmai1C*7G?#UH8jKKl+|CXPb1N#NKwL<k-|T
zvY?_fd8Zg!^8Fy)CnFvhq|2Jgl~s-oGpyK8fWGJzu-a~xwIRbV&-UfMcz+mSDS?t)
zyrX_qDQ#R^V8P`C$lR2npg>Q6IK^k3J?REQ*&`EON_kJ<JMPO2zmW{(cWQ|A9{tUo
zeJBIg>jk`>r+?grPkfi4$lpbfach0=5FE84lItElwmxnqHsrT}dbS^8>n(t0%?QZs
z_XpZ;FCBmwt~jZ~cJ*v1pYd=E#7W+V5b9$a#bUxj;5J``p1T^LX7Q5ze#*}Btus;P
zJ4A(TH_+5?1X}|^b?WQA_bD-w5GK6?B*!ABkWwEw#qIR~veFM}Chr3k+<S5LK9P~_
zx6G#;89;LHstk$qf-HXnoGV%lh~k|A|77N21<8YzdCIjtw^Jrkv&Phr@TqR#3rGaA
zT0f!mki$fe>O2FF*|<d@f`2lSDxVBMJ5C6mk323hi9PM@tXlffW?iufj7W}Rz$D>x
zwh9JW^uOCZE3Q5ro4+ElAtMGv=pSo-iquY$ScB-Y)vEPk-`xUdkTH_|UlxgYb}$@i
zb*IR!?)q1d;=}mDZlY~-Y#x$M)vVVa_kona$}%8?&2|H61=hd>U^n5DJ3&|%1-L+-
zG?4tZ2uhZ=Db&z!Achx*Jk}od!0YK@Gb+CFn?t>iw-e;N2pYFSkx4rs%kKuHHw?9e
z?Y9vdZ3L8hO8?zG4}$|=(<8=5PXE|N9$HZR`mzb`dIDDE%NB^yd>LJN!~;6k`<|fp
zPeIM?Xc10bA%E0Bt}i{G`lR(_x7be=4ANDILhb<_5pM(c`tH?>g%>96K}17WCiPVF
z=+C_hXOQ)_I}%a9$4<o1#0(GS{*3zeHBn}x3b-lyfapIV`ew90R;V28@NPT@-yWew
zT8PQ7G!bPo&!2(V#YAclYU>Q-rYaogN!B~gj8#Ek%GXrWC{ZXVJN3rLl~V5v7qKh7
zEw11e#P48Jq4Z1fai9S6@O}c8$08*3Hl<qm2ss}tC6{$Yi44Z0Af;a`M>@nxLK1QM
z(T?2`#y8n{5v9^KbULc5$?tY`jD1|zK5Tgd&#F6h?n51Fvk&$@%lflJ66^Jlo(ImT
ze|?lXIuii7XU-s&C;J6}a_#hU)`o7cZv!hJLNQ~uo{h-R(4@;Dg8`h)x?=^SAudp&
z)d*w^0=B-iDjMv?eaHiJ`roV$MJO91z*Q_lvL;9e-VZEA31@KKSg(u;9!&jPu0SKg
zRRxwTd!wC9nyh)C*IWd_E0)wWho=?23iMe$zIz`oO;*)i{vh2^b8OyjA0J36#sM*X
zp&WH+9dGby$z!~d2&IpDUql(yau+wM|IO~Z5c#ah*uHj5|Bq9UPZJMaIXNrZN8Lv{
z^%*?jLqnn-)}8OsPE)fU(!o|b7-`7-ARW?!HC2;7-C<B&1#GSw6d_<dBw@vfzVH#K
z89^#TYP<B2k3a`_s!{;j-C)LPE6^>X)!q9w@fOHpDg9Id1|{KudS0{;7;|)K6%L@m
z>op4U+R|oK2_Vo;w26`Y$`*YezvC71Ul2lgM^)!AeZQ7H4Rkn*&n}Q+m`{EY_5)RR
zN^x<&{3@_n=(5OaBI<+<$da(Rz=Gt%MXdVN2gu$G^+~3<@s`1p8ut@CkV9)9K)AF^
zgb?<w0;TBOT8(T_Bj8gG8NZZi>?w*&=NFB?s$PRw1nRpnc^qi?H6jLq2fERxq*4dJ
z(6Pcq+hjw}-3PSQ#1?Vhe%};#Hk~;Gf6kAIpHQ=0HGE|i_CsArowXu+;kATi^by2T
zaTjj2!q9lvW_K>Jr$*NU7svdrYO9V%KlRLmDEV7*Oi(meI)Ny1pc|+sBKz#h(|zL!
z>qzm<SkIt9xt<1Zq^7_Si@~=3Icg|`hl1S3i|)Q(?CxIY%R`?OyNdQM>CI4ev&ol~
z3n!%V*cmC4dND8)C{oC<Y42~0G1^G7v(KY{p6j0|Q5)bCYrswr%VSVJl`FuaCSQay
z^RHiqoqz4#4m5T9o9a6kP?_rROj$U&{3c+GcU`K1a@unW<E4idYhKEaxuJM<*uP~F
zLocFwece?hXhk~}uy+L8^5|)9*;r0UE;<GNil4d;leI*6^mG_TLv-%e2)uV846AuB
zNL#>Sl%pZ&u6EFgR?td9E47meMu1VB&e}bkj3OK}u8$5X<Uf9n)k34I0E0{w<Lca1
zu|!jp;3RKWw6v5SkCL^8Yd%y5Fos0_?*&(b@=D9n2FO--Lw4i3>`{jgwMJVQmPIU^
z4so9f{+@SV2bL<2uI5I1PA2+{G(Z6sJhb{_BK>)nJu^aH-v@jFGY8VpcG_H#vIjX+
zghcM%T)SX;#Hy(`U!7fpvb(%k`oLx29`V{2wOQZ{Y(I9CP6zIu>7^y-6}}1*Pev$`
zHyrXvc+_9!_cVBglr?z|Ocpd_tuE#dPH&F~mPBwQ$isZ)aR}a#V${q$GV>xe!kIdP
z<nIbS6-hknrJL>4f7ZBsmjSs?K+1h0?|e15f%LP(;5*$f;II{pW)X@I%U4Z(JX!4d
zBq2oZ^(eo8+w(*Xo;o7GHclYw^J}bQ%1bJ^H#>`^OqQ=1s#|lJu84d@P<j{5%7aw3
z@P8eAcO-IyhBNX*i8Ek;z6*!jhEp~&&VK-`a3t7M8Fk9fv3r=K>8?XsF^$tO%x43>
z!2$wKzRY)2&@m0qvnohQf7%HilZaC&+^Kxu?UF6n(s}OdS!V@!z0Ruz1ow3$o&}uK
z-PSD>8~zXfa8}0%<D+Yj==M!;Q9OWavZvhHHcK!uXhpBym@lcZrN8fYxr6+TSmr_G
ztJ4l)L7L*y4la4LGLOz7=ie0KTG(ftKGP(b^8eiO24G+(k;dRUQ9P`R%yQo`@HbpJ
z*#?05?^8BD>Proa+_7Dq1=ax-M90i`M_k7!ps_X4<JJ>AlG{l=Dp2@oGB~qZ9-~gL
zxiqiGIfD$OY%&5(bO7B4^+U*GkgsZgN9-fkJ;4nX{0m|Y#(tC0QRH2a8^jn8TGNUE
ze@_|6`is3c==(e<`UU`?@o7eAcr&**OAuckqBGlvWQlr0JAm3DNfxPuV8HDT8B6C(
zr7R@YtRhTiwA<mjJmVmKB*}8zH*IiNOR<dyDt*)t7tKjQ4DTZ%4U{0;_`){WKq+PA
zO_hv~;(QN)bt?%9ny2gy>jN)N?>~$ox)t<ZO;rsSW+9=et_xL`WuQc3mtrrt?8u1Y
zW|g;sQB+qC?0txQe-ZVA&-YF-<nxL8ScVmguLo5wyJm?%_8t2>#zCwx{VmUFU$tH(
z>=Uu~o~GA=@lP-16)jl)f9-vFJd}Iie@94&8QY<3GumufBf^9kMp;57vWD!0#**z;
zXe=>PcHzVzdm^Do60#f0&IvQNNX8PP=X-UZbMEKdzf-@{{m=7yUa!ZWW3HLGzT5Tr
zeBR4P_gtJ6unFpgXjiz?Rz|Dtz>!wT%@$26xR8+a?lS<dj$4V$ifLP5MfYpyBe%tz
zaI?3-IUmpvUl((xB>h|yJhZ4Xg>>A$)9!tLf^3cl`cqqe%=}u4qkxmNPTn@KptQ-d
zaCs6}fVwVVvh)mSNsaLFkxH=}_v8yn?NIdKok>_MFJYbE`5Hw-;VvZnA1Ngj<SwaL
z*Hwd?;+PGoE={EhQco{-I3t22W((7p^TWE7=Zo%TJ+5KzvM{Pnq__Zg->TqanZ6XY
z?9KP9Y9IR8GIm#-MBdEBrw&xPU_(<>ySi%pC9-zs!f@VV!+^_BTq#Ix1is{Q#7AlJ
zkJ?l2T10pXqz~yv1^6vLjUG3uO?2O~J(~zrrXJKU_oh|y$|9u#2HTcB4w6q?`nQDe
z_tfZX;mH|__sq6!Tz-2~$|=#$i08t`ff8TU^s1(^M&qgLnkpxNiE#5?7f6U>gkEox
z9e{}c$PRn8N`VDrw8R~tnZD<V<fP*yGv8p63iJ0Da0(h`?;7{de-8YrZ-bhl3mGEV
zpIC>CuOw~>pGpUv>h{9vd!(}HDc6a6L8-S#;vA-+!*dzIZM+9bXJkQg)rI$nVrsfp
zfY2isYnaS%-0vRX6PRl7^(I$J5q+c)x35aLow~ZUl7||tH1V|(>989DretJ*x28dj
zaTnQ~hYS0a2`R_~YY5jW_oG$hsY-gI8#UxL&uQd3a_qoT35w5w9jSfFzUnw-!d>eI
zUprCy)Wa#$pnF8eBSxGdVPeJv-TL~DB4aIE81q4e#-2ypEn*{zP@D**F6H%+@0;qF
zka!G}f&Kk`P*(w-odI?@^DL80R6xWQU3JW6RsXO2v*NVvi*F6r2j`O=OS4Vj_<0TH
zSgv19dE83lKH!yFUJ}K8M*<C>ggl>59wuBRUVxXt{GM{}B>NIb00VR#di4gPQg_LX
z6hdHv8|SPsm4S*xn^C2y$#~hMoGD;PpXGT>Lfmdq1@vB69*N?8HNusjNPG55rKM_;
zU(fy=l)_A;f<$9J?CQ_6_h<uGB;>5*ZXu(e154(1S<6uz>*S8uyuGFRoYu1DL(Ll7
zPBp+9HzwtG&^%V_B;v$TgGb02iB>e;*!by<s<*9`3?)Q|izLgAF8)QxJn;mSeo4aP
zJtLTOxBRp|rcaZB?bbcaMy1%ToyYxK&y06NY$(dkl&|)AM^{utfEf8m9+V?3q%Hf>
zI|pYkJryR`zg>J;w)7mB{g`E;XwqG2`s)}s`=Ap3CY*eFYS~NvmmY<Xb6i`$-j+h~
zhf0)Ihpf>ybsYQo@Ta~!wKsj{{a;19f1T3TyCOkpBCS5a?4J?tsUvhbOWIQ>kv*8x
zFZsjweWXue;)^(HL|m2*0+v$z#&6g|KE0HXfm7J450c+gl2&YeqNMQrXeH;kzME5!
z0~%#CUuY2ax~N20Kms@;*(xAVVAHypdZ{z3>ud5L_6s3*Ne%m#b)5-DY9sRDp^+^b
z_Vjtq;Z}f~_vj+Gs5Ymp2$2kHBRIljVMIbVFi<CR<xdyfioJArs28{cj790+s=7hw
z<q~i<XGq{~L9K8;d)I9+o5L2SWrsxWL6pG~j*#~{493nw65AvOW;-=xwv1Xqc<{zR
zvsxll<z9h=joK|5m837ku^C!+oSL$?DoAzY^5#BLWVdztq{7agCfT<K;bU}AaW$-y
z$}ldteibA7VOJnkF7iRVIm(u~4xON>VPpOx{y~U+JZh*uh~dm8!0r+FhAI^(M!(in
zbwhlDe*Hktox^vK0eqaeq5$qtUQTQz`nU^+J3gW%dBUtJPWycFv_A0j<}c&)OduC8
zHP@|hFHb-Ts*1_g((Zn6a=Y<d%>Z<9`@MOy!II&qwP^WfrR$<WoyiU*8Z$sltQOB8
z-)Uk}rOl4|yc8&f%z~@lJzCH%hp$GT#9W*lD-_S3i?eLv_Ll3fH}8;RVvO0Bp=!Q+
zo#loEaUD+{vSh4%Ua;y}3MUdBcEfvNPH?yCBs^VY_gFs6urfxWEf_n-l%9_E80wN^
zO^Kt`ZghBbdWonY?XRl|EL<K7-#Z~h>BHlr2cqLo%#r7V#`0CA)!pqarxmp<)E(`D
z$ARVMvNy?H=&cM2rv4E$I(RS|zIAJZD}VpqgmGD>to8(cfE5h;#Iw1GFW68XmSkUV
z1>(b_5X%;bnE|7YCUhE>6F(C_iWVq&`^rmk`$cfJ-Rlo()WCxt)Hjh?(!GM;4fkY2
z{Y-6shas$^;7qT>1M@Lx7BOV;WjN8hsX`hl$&Em90nhc$<kdG)@M4L6__84mXV8Wi
zgh=>#wgo3|eIg9Q44+03_C~>+3(}%tXQChji4Fdb8r4iXp6xz$VHFp{SkGQnML6Cb
zWJ`jmFid=hy!V%vQO{scVwm`<^_FfsYE8}+DY=e_ZH8UfsC+-~C-y!k7JL-``UUj0
zoc6LHl`y*tXt8eOR2K5!V<htEZ5O#oVfyB0l9tq;;c&NhHAFQR98<<*T7bveVbGrs
zl-JI-K{;5vT-ftmi-?-wfquA#lY2<ZVTQ@$QxQm{VC4`$9nc}PCI?6(BkU6qySd>E
z-Xpxf?y&kxt}{#)rQ-^hE5iX{dH8;DDdG?}9A3T}loby1iy{~)FY-LHy}~7W@Z_Rm
zix#h(nr0=xoB=xiS>PgJ?Ql_s48P&yD1tbkfSIBD+CYHMjp(Or$Q~^53_4#NAI$7u
zH~<DBns76kpb-rlW{zbkJKohiA_U)8c6}h3LP&|i$C`w(nsxh7YYF+w+nve(z+ppy
z-aMEO6S5*#q7f*WY%;_W^EJje0{1KpqE&s^9<f?TSh~mbTDpsd;N?*f-xP?B35Vqj
z85nM*&$E|31Z5LU@8_l}iYbils+&Nz*&tY#eX)4z#jtKubku0B#quWpVe*y=X(o0n
z9HdGiz54K=N<7n}SCIDiLkBo?OYlsw7^d`Us|^cI4xiq4yGw=M;?5oL0&x1zrGBr(
zmuX2GYNT%D4(ykjGyQF&YkRn3qYlxAp|}RQGeE*K^<uDhES`)De6P~9%mjPa^xv|`
zv&<nx@)JXzAy5H|#=G|oEaxLXBgU_)>M=yJJFy!|l7GEDEJ==}x?yE;Uff2E&6u!9
zp3Rk|r{XoI)SpFS%`oU0m+_p!5Ag7#XE@*;d4=2aK65#2LkEaSQ2>@GDJggs>!PWw
ze&PY=4(teV2t;soP0Fo+Fcb42_V(6!Mcv9J#rN|^!w=e>Zqj@jZ&v%(YXi7ch#~bn
zLQf2yBbX571K>N?`8f5a7tYgbNJl?N(R)K#h1ZH%X+OVnzPZ&_#CCn3ho2^o{ZO-m
zRD;kRu~lJfOy8>;!OB2i8MRMgKp2q0;gSTpliYSchjsYm^B5jQc5uTVOLYn9;iQEy
z2OEbrj%I!4hUdsf-zILBS4J7GH2*RgRn2`E{}7PIK1#>!S=^31jaoMWU1(X}S7#>g
zu1Rnlo}a^fnjF9*lNrePT|<`VcRJ16wtLmwU3>*6S}i}&a*b*4!$h;3^Ngar;KV1r
z?xtB#!-W0o>dCjcW;ZMyI=6vf`O*@<{V~ivO)@2pwaP@xFajs=rqa_&_>%?5RWYQx
zj)f44xRvQO5mA$Qo9f{m>v6+}Bq_`FntqCsSe85wd^^ebyV3_{$TsOl`sKU7g_}7`
zzQ86&u>oHZfgvTkj;NgvLme#l@z1U8!OCKlp-5*dS0WWWV5%WwPBGA=sQfah3rj%p
zokY&Z8>va3MtO#A=$@LY3?{*mJ_tW_%|xBZ_==e}DX9bAdiYKclEFSvaVQ*B;$QZP
zCrvtvlT`#`e?`Da>ZyiuO2hckl5DO!c=v$?zf)9UXZwK0*FM&Gol)kH+Stf7?}Tyi
z^s0(O9>3NcAWwFu&I=EX9&LrddHWJh{S!9)z3$<trY2xB19sO(egvy}%uyV>>dM*J
zN4}t@#zSO$nvA_TcP$8=DzeONIKC;#q#qNB-EDU1)x)jqaZenm?{c@?BGT33<r|e!
zB_@Nx3zEq5PIVDj5zBymZxX<ke=})H1ttt4lkqc7Xjx3jT;7V<XSnBWna3cMyPYP4
z<PM6x0U4;q4Zk%QzOIxD0<C=atWV3X>Dc7yov6M((|C`9=vA~8+SyGEy({M0;46q+
zk}ZE>+PqW7<!#YO*Juk#Z+fO8_+o$0wODLu+954Fk0w?xSE-!gonJa_%F#2_Y97%J
zZ_>G7mW$uZDCVFkt_iYj`}!tpM$Q?nPa<=}*I+xApr+^h<}N8t#M#CIahb;mM%M~u
z1&6rEd{UqTjZzj&whO&{-lBH9NYe)~B<Q5JC0YA6bV_6BWhAr4uTbN+V+}$*RQ!-H
zKwk9EVa8&u@gu!@DZ1}GxcUv5b!uDop)bsFy(*Zx5M_8iKWF6a4is;r>OPMZGNI>8
zvs@rA7<=wXH7I6~@Pn+5CM);ZGT0v2%U0$KFke2J$6Y;z*Tb+bC^~8CC9b8G#dVt;
z*#%kYD|vA*^N<amGkj!KsvbAeYCZ{~CznEAALzWU1Mbb#mA7Y(Ron4=#;%_xp$~df
z9UD(?7XJLMG-<0)pdAtvWa98RsK8(TOjmhL-?p+gb~&K5L8w`Pi-_dAM3ODxL27jZ
zR6vR1i`%x7Sihdx|2R_qRZFl($ky<@LY!$cjT;^#Pt^<WD;+ri_-ER^$CO5;X#09T
z&oR4A#L+A3+{88Gp1Om1oDFQ3Y8DC!6g&WVEhsN2U%4c;{UVgaCd6|l%s1FrMiBpA
zv1MDCn)UhjbNH_!W)Tq(F3VP2koaeHWG1s7$O1%GXVFgYVAc4s#xu*gHySlt^^v~Y
zpiSKCSreTQ?M6v7J&9nL1d*1NvN&ZM&^>FZLu3PiQD#4^9zJLRzT@Zr1-!lb6Cmpi
z^%cp?!$l+Y`+u9tp139L_;SeXit}_XN$dUL1}l_pR=&q|?og=AG18<js!DVPou%Fp
zF*3t|+PWO1t30+ALi!Shn2^5gVJzBglXvbHtPkhw+j^@>^O5iOMhCOxdsb(Fic4-~
zl+BtpYBH)N7)wN3XG=oPPwJaC$U_Z@6?34L>RlJ{8n)xu1<KHuimIup4E;BCF*Tkf
z7wUt^49Nw;7JLWWe<sr;$0dt=4AAag0gn}P@#M-(-+}Rf+OGM|I2D~fwT$dhE@bOR
z31_jGeT$_a*sgI(x)G4h6p|h$Ye#brr!{fn2Bc%=XU+zVp;xpQY5EdF^D~c85wCy&
zO=)`%T_jz*&;?<^Kgf;$1}A?_-{&~NX4?$1WBK}|wPxh-L+%Y>Vo0~V&gtm@VJ=e5
z!^ROD`{scSm|u2ZC_Vo36VRN@9`PXUe!~V#8#0TzR=$-Sg!jUr^T0#Wy?xWNZpFAc
zV0>$g{~dGWFUAO2_B-lNmzs~40>YTmexws<3qsyR>_Il*j8FFzSn2$dw8Q%fw<7X6
z2o$6{;m@I(n>-RJuzrK(Sn`%_7E{P}Ok^(Sx)`0O6x3S{W6V4aK^P1;<QVQ!m4}X~
zehALkare6C=yNEmlT)F@8%G6b9(5K&H`%`TjoMs;>hf0NPKsNiEmsPQ9-WR2?(h1@
z?+pxosXkDL!^h_VuCd8y(*A_PuiaQ#vkaP2Nm{*p5RRh1Lf(5fsycNnW2GE&AoAKj
zqtD@JF%0IgDx?}!m8nD2RRWH>kIaLB!Fhjr?qrSShq#*7>tG(|{J<b2l}1*)8s*B`
zg1?BvKgBWJPYifyeC{cvC-V;k`kL{=LSz>o)3jbs@G$$_EsAtWcCEm|I3%EKp8zc+
zT*v!9)%Nf3GPVJmkD`kJUZnAN(P9Du2xNm0Rxh=<gumKzUy*U?yL$miXliVI=%2IO
z%E244q}``tStru4atT|i6ZORXRr(t38OW$R;2{J~EazZ`^WCF5fb7N#z`}*91*{<Y
z!~8**a1McP$-gVWa%;9ZNSg~`HV7Ez=br&eaVNQ{Dt-S2*gI!%`%+jLWukolSxoGm
zHwFS%g|oQHI(Ne~r=&@Bpa<C+NCGoo<M?rXhI$?1GwfOYh{_Q_B84I+R^;c^UZIzD
zvB6=e#x_WdNYhGolw!J1(k4lL>eu;npsA`L<2fjM1l+-fk@`wFPd|t107tE%c9^U3
zx7&axS8y4?ey8!SB<2N^5^;m8z@p6|0!jp{3)af-$+1IDaSQmPq|hLMDi{T@8-9T*
zJ@&s2F<(Bk%(iZG*G?;S6#zqpmBDjUJ4lWF-e!00ju-9B_?jbrmID3uZac-n1Ku@O
z3Fa)_GLNiM0isiG=!9y<oTF?&Fd1c<)3#)o3{Ia>CUbl@>Dm>xTp|Cup1j(4n^B?K
zvFQc3FsZI~sJ)a0YFx)jmmanGyjJ=cB(bB0U>o$pm4uuUrybJ|>XQsl_=>cywt02W
ze*%%!u%QH1FnD-)ERk|KOVh`Ab2d)h6e&dG7s06ZsLW^C-w~_r^@x~zygDrOM%g^^
zn~Op?bP=oyTvw~=Z^89ljek7Qf~b-_V)&|tb!T-2;l~qT@-L_WI7<>HG)LNPsVtHx
z`~?2ESWEy>1h;Vj@_2EAS1Pm9k;t8J`GI4L1nNgfUg^Z0mc@@i!|pqQ7GRxDNSzmc
z0A}T~+MY9se#v&S2;ilX%;XkG@S=pg7;qAQBmAO<vNn2GNxm7L`s#T7eNhclT!<01
zObUlQMo4_D^Gl&%v+QN$tXE_~ugB~YwyeeOaTdvnvx(MN7jqsvUHY(niX#spB!s4R
z^@_PLJf8CImQbgYF&ty9b$#m5**y(hqT<ADY_WOLqvTEAfIf?zEzsVa=syOMt74q1
z3rA|J_wpdvN}yVy*SaRh#>b8(iT5XDE=(4?o_toN4cD^*kSL`tl+TF2M=UZq>}+dQ
z-r`)X$y>wVwUdWW)ZJrgU0c?XuK=02d;4SIh-+fzEQX^Ht&+|&zTvo6m*;KqqK6tL
z+pj%I1JtpJ8ZbCxfu-ej{gC*P;&OEQAExD3a9nOWaHt?Sy)gw|77*c-QP6pH=9+Bd
zB=8Fhi#c;plbTmh(HN*R-awZ>I~IYx2_QIAecI`3vRT|0rV~MdCfl$1<QUOD7r@xh
zvh9w?d|kQqvePdgZoXGf4uj9OCq>~4R-`6GXg}HJn1T8QzAg&e9@xBrzdm4@%qaz`
zM*JIGPP5>F#NW@BVfrTECI!-E`b$)lW@0C<sM!T3%uezRryF%YdGh4*Sl6~p(bhB9
z-I<Wl`{H!&*8LifYF1kRGwN~EM+b|&luA8Ig`;F|T@wk-0Nn(XFtXl<E<2kN+;t9h
z1iq|FeX#QKjxt~-m^_={_RkQ`vI#gDX=Hx7oT>he!_~a`**w&PQmaG~<%$~UmJE3;
z_4;^AE`0eKywm$Q2UHi2?;MR=|2EhN>EM?R`o#UL@CSn}cULH3#t2<=24O-XtQY~E
zG(wexf0TgiuyB6|#Q1~>jCWLvjHqR2E^N~XFE$JMR&G{_GdUp_0Bj*@`O9<=)EAly
zsTY;>e#7e70|Ufx4s5GQpe>3JLIwasg2toM{0E5EMpzF~yRsF~7j}xK-ALlnur>oy
zj;1#sAtwEklk`bkQn|fx!7JH;x=aG1q&pQ$4J>*1mAhjAs+z7D7Ow)&abvu(DAGHJ
zu)kNfP2%nZ1CDtR8GbfVV@e0%Oj#f*oPcQi3+H1tIyaz$h_;8_Uv&goV9KoS)E>NG
z^=qT&dm;F!5%%W7Q{7KN-O3v7y719sn5k?oVv6i7d}|`F9733)P^N14oNM_sRgn1B
z2&+_F2^AkjvsOv4>g6?v3*bb=@Vv&IMQ!stF&Hzkh&Dcd>q*hTpA|jW#e;e)$D`5m
zdQzIKnDmtD*-*96cVv65`ImlRE)zh433^>h8|rEzoHr(&aI-MRA+ivZkNgd~QI2@g
zsVuIBGiCsnp`F+_D~_>i<;vGK#>?cl<`y}0%Xk8F|F66Q+s(<Z%<<zn00OI2(OYc8
zuHT!!^R~5Bqs42sk<&mI;>OT|TFhT=H{rtD6Er4t5Z>ov9W7ZS6~jru=l8m1lUvY-
zD+qL(id6W6<lz_QyglN*fM}4E+?&e0wE&tCR*$xyZ~qyQnFF(<p@z!9V}s(KSIf3Z
zmx1s+L_M-8Gc~{k-(KE+Vg6DTj<mR9*q<3Q4Wp7NNmd!t%v3F^I8|&>YN!m5PV!)(
zWXN=J?QlgM`;gGh81cnhc>ToT)_X!mXPE~jyY8B6GkJ(2pXWj*FI}1l*F1b`s)oDU
z;}F7DbU?TR$NGtO$K9+c<C?-A>64JpC4zN2fUt(?t5WGiidm1{+PZt)+Efz3cz82h
zslrTobe<A?PpWZFr9@+@)tkxts#H$qmhHuL|HtVnG|kVQU1u{~MicSZlE;93>GZT8
zKGC0F?u+)pDlQ|pA&1N$*>U|<8MPF}S>xcec$!{|*WQwmNeDc40W|E|?S~R^LQ~40
ziZKWL*JKr{lSDYYnAH_ZClwi<ry+BEn16bzV&B{qyp?U0h&VgFC4zLMRws<(aN*-A
z2rPB_(&*Y;jrwM`SzOyU7<*7?+ARQBdy6NklYUdE8!-u2Eg|{d<w#)B&#dF*qy4$1
zQsAY?IfLpmJTStsAr>zoGMtXhOL&=07IueDgN5S~x~NsEcsP!7@-n|v6Ok1jNjT!C
z9&@mXWfAY5XvIUD5R%EQjR(3!!s{EtQjPxV9k@rRe)P#y(%v;y0KIsh6e)H~WUe}F
zcQ6*i8-v$0RY3D6V@D=m9dBTpU?dBA*`qaGJl3i{V9#K$RvgOu<@WGcN0dz;+%TVm
zAIV{q_?h<zUL1cPpSh%UQKN|@Y@^jzc7ZlyFvU7|Do}lfBn3-(aD=>(i{}_P>iX58
zw9bxS?;P(FJPHEECriF=ozPP^`4tQ!9gS0TdV?!=Bk17FIzR7?k$JuSqAJ!g(R9(7
zV~0_I6P(Nj<s=2eknZQ86cvyPNL;DM4n^0u6vaq3$}bu$&STS3n|Zbe^{Rm(_A)Pt
z?O+jTZF8c{W0e2W;>%SueOY^@dGisP9y+kFWPf?PCsvcGm?@p9s3NU8s!YHsk+z(C
zQ_L!Pd=W%ST%swoz~(^AxTgxuj-=MPZpPW{r6PQq%t&=^Vp!ibVMYq=aGTM3csXPK
z35rp`1fDL9l0(ipWC$hG_`0M|)oan&2}(8#kQI!4Oc|@6TSGmt9f(hh;<n}>E+DVY
z9nV%3OUV`XAl>`p`ukq<J3UO}FtqChkE0P!lc^qjU@~E^4XnrW<f7G+GO)sv*Zl``
zfq|4N#(DU&dEs4?w(TL*thb$RuyOaXsf&%Cn<(<`BlO+w%#$axD2JPPW!*?0J0@A~
z%VxP!QB27XGJ-EA;@G?dR!}rFdZC?_Xy>2knJMzgbDv6dwL^{M_S&A+^_O=l`?rZ$
zi-3*dFp`QP9|?81V0hs>OIbPta6SDFm;6>ZZ&#peVM8;EW0?0hv;nt{gzz9(%qIyJ
zFa5h0!?_{`!i!F=@UKxA+(}Q`m#q|~w%{!%j!4x+wl{yo3ggj%3JC>ECZo6J=c&*5
zZ-!}JB7U(pw%lX5Q^jNXkT&S{&dg8U0DT~fChsXee^YLPqyM(BVN?=j&0X!5p|-!^
z<}UbN!?>0UOjoe?Dp=mZf4Ui*+pUP(-mQ7jXwT&CVE(wGo_wnr#U^#doS+c}fnJx8
zbMv$7$|qt&IJuhyY-`hu#^TP;%9Eu5&%N`_YM)|^j5?}!9Q^Z`tzDK;tmGUzUYzo4
z?u-F$RVIT_MDS7)>?zJ{2#P}+OH!Vlw=j9qZvNr~sHEznkb}n~uP@+)XFXM>wfSvh
zg~?(_2gPt(d=~TWWTc#Dh)kTE2ksIGZ*5}4hniRAy1E&g*cWd7^yO&IxpvvEpGwLk
zZjM>unnl7*KHRS<GdNHfINx-3uxZU;vnp+xiFSB<c;TWe$^f;lJ^Hw^zir+9gZCPR
zo$*)WwIlO2k})z(j5PYmn>!A!1}hS7rJ1&T4t<)#{)s(Z628WO<67+7JuiIc*t!yp
zzAZ+Tx<B~tHF0dx0+{p%*(Qk(Z8PR`uWd;PRiJ8|qPPA#6#j3N>iRDL8Q((j9Q%VR
zV@IdgFitR#BhL}&uwAEDju@fWjDb-!YI-$!P<t0V%+Z!~surX$&Bq&MWY{9bBTm(O
z9;e4&KzxT*0<FT_6X4_&9aC2@_>EhVz87e!QJ=;^`BN#~&Z6R(`<4SQ?pnaH^|T;*
z{Mt2OFO-`dt1M;#x|Tt03SGG{0-WWgnHen(KIW>#fPR<Xydl1Mf1;wFm6(68;_-_8
zL>6!dNr966dx(B%+<J#2NyDrI>fLIQrz7C}vPhRsMEELj=*@8{sm?tw=ai_krIXrP
zyEY^aa%X*|4qKN!KmY!5e^ICS*VwY20{r=1euBS?1M2<gNb!&re=kHS0Jarha*hBh
z^M+7>+O$%a&K{w_tAHL=-6S-#f3L3QS4mjvahD5)6cGOVngQ)LxuCEl^@3H<DiHRv
zJg%(%oeBQ!1PaWC^;tk|zR9uT_dCb23H}Mti<5SM>wDZK6mVRhIvO&mP^CfF8($9v
zW-FlGe5DzxncsT~AB+zMY(fW@T9)q|A^-Xr5$fQMdAW);{{zLqW$-ga7+qKY)))P0
z{6lSEKrz6Gng8x={Pp^XfvE2IpDxvYQ&I8F1#Qg>6ax|JJ<5OjG2eXqzt#o#B_fAG
zo$+%0g@4*4|IjV-GJ+jsbGCx*_xb$$Utf<0>GW;hhi|^a82Hv&`rkOU^`{!fflYe?
za4Go`Qx*B2jCQgaJ<v6jK-@E3=WIrf9R<KdgP7*-MhD)%{`*7!$FP>C)dQm3y*dD}
z>pRPV599$3IuPwg-}$C$%73s5eK>(#e+E$ybU@c}YXWz?+$NPN2>M^?xIH}b|2rH1
zV&%2F(|t7shFts~^X=Xn(r*<K1lXZF$Mg5C+5N=@f4#B(cr|oT&l(gj^&!}uq+l!j
z&co=^+<!81|Bou=?`<{c7w58rJu};DQhDx&52h>9;8MPk4a=Gzo*P{uaHRF-mAdBt
z@TKk6;8M9NTJ5zT-sQ<^;8@Bvuo^V_@k?KTORrp=4w(JnX_gKGVkEuHx<@%be(6o9
z^!sb)I{7;g`kzkEKfc0x6P)~MYC-mAe*98K2u^sB{vqdw55xKuNbeQ2*_!ynZEfmV
zp@aYC$?-EktZoaP)dUaM&6NbnZ&%U(^(6k&$rQ>5bp>MdtxG!g_xH?ipU?|Q>Q-jO
zFZ~|H{Be@~F$;e^H>c?SZ^Qm^2!3za|8K*7W7+<B&;YvrKeJ(%zWijO#8x5Lej#)h
O_|w7Yqf0cbZ~Pb1%Fw$2

literal 0
HcmV?d00001

diff --git a/_static/IT_sample4.png b/_static/IT_sample4.png
new file mode 100644
index 0000000000000000000000000000000000000000..19caded9fef743eb110b8fc738839eea401389f1
GIT binary patch
literal 334618
zcmeFZWmFu&wlF#bf@OdZf@>g1(81jTfk3d}?(RA;I3YlAcb7m2?(PnQ2X}(IJHwlt
zbM8Ioy|><4_vZcmdUey))m6Kych&Yi1S`l%VxSYD0{{RFsgL4{0KlIjVh$=Y{MJkm
zaRjd@vJewfkP;K4P;j&}wXik;06qrCC!#1QY2f=Fd$6Q`O?Vj+H<n6~4p4k=p8W}h
z_zl`i>2H2nhLU({_DtWt`BLuu{Px8P!MNm`p~2_!8H~zrbo+)KfrC%JZ|d(ypPtS$
zli9aDw$nY18eEUOoX7yLO-j|%5Eu}n-VH{J-*~XEuj3HhAov0}vz~nf^lh4iYOJkw
z0OH@>yFz?WM1Ei8Pq~%)_&n)-2w>^oLI4Ehrge{QndEqqOyLVuyhQ;RbMJ9$r%{n@
zgX1h-Rg1(OhdT|v#R_+tKr(6<mr+I{HX#Bau_ZD{A^1W?9>cVC(NlsB0xHA5vAoV^
zl$fJpXAg}$!;TvsBlsmrHaaSE$1Fd%;z%>#<k;}eF++Rk>r2nCbL+sPi6<w1Se>!y
z$AT@8XnYzy+BcCB`bM+siK;IfjMC^BsfKTSD80WZeMcBqf)1I|%CKhc<GdwL#U>Ky
z8F()oAs~-QYZ>x<t~FYTaUdjWG@j9*DnE%KIYTXljOn@no#*TUQuw(-DUBfvEq(pq
zqCabwpmvcZTB3g+|Jy@L!S-crnIQXp(<M6ZSp{$5XN0pFrvmsZ^-12U-y1cj&xKki
zS;x=Z93tvGnH?nR1$3j<Sgy^sB?kMpvQQk|t?NCpChPsylz_7oFO0M35r8klgl+**
zRO&J8tuMg|ZiV7WF)3a8+1xVitKZzF*$Kg1;447b%FMv;lLR3LtT%U47Au;s07p3$
z6%|U8ty!Z0k{J}=x2adjSAeW<8TUh4Zi-Szd3Ni}2IF?Vud)zivt9uG2+N;=+tKO~
zSXN&=6D9ce>{K6N)lWGpShbaa5yeX64IOf72bTiE&kizs012|BC_B&d5&uuCMD+-^
z{y`$hcO9QVz-9l>s9#QAKF@id5=CYV)OZhmiF4-bnXOOvEF|bv_Un&GQrVkhnm-8W
z{phoZeqi1`=R->mDiv88^ZJ2L^mSPz(*TL4i$fiS%g?MsCh0j!$LAAIp69wq+a1qO
zB<e{@1F$+|Paro$p2P_qENj?E6zHg=<C4WMX*wl3qmbSbMpH0EQ|18+KM+N!<{>r&
z_eZ_iz(EemHY6Zm2qUHe;DoOk;2EeIa_Q59P(ZRtV5qDNiz&f5#-|QTBc_T{bxbbI
z)Gs<=Ox+0v*9P+XEQWSQ_Z7zuJS3!MAx^!|SJxdZ>P_k~>eK5b7J0R~OR>1W^!?^M
zO>lGdB*+LI?Z97mUBy}(UH3S7yop45{)Ma~VpZ}Z#W&R3uWX&7orLeP417V-$}~hc
zI>83wLJsd7a2!}F;@#3`<w>Xuqh#RgLZ?V%fl-+DOQi9mySRH%R+fI&Y7R(J^hdd7
zw)Wb$PLlxL0C{7T%IC?l8IlPx+%ea&qtVB)D~fclaYFE^{bI!CJ{IO`X9;9`D(BH_
zGC01z4JmrffSdZ-<F(#9?D(*Fb?FyzXK{mZ5$|-=h!kn)Q|YSW8wMHT8RF&XS5(M8
zXaAi2`L)par?mp_562Ii4fMKG;!|qzWtv%yl9yjB$#JuZ6z7!5T$o%~E-7m=19Oun
zc&a%xSr0)ap<KoJKew{*a%;s)6c4LY!Xn0GcSziky>K4s9x)%Q8YB2V>14I)wurV!
zKKl0pF<3DCB%&p9du_r;Q-HF1!_LD@DZ?p|H6TqqO{Z!+tHX)LADkTiTwONEb`s-Z
zx$I=-UGsSJ*z-fx*XFL4kvltkH~ZgQz&qMwgcINQ-FNeM6ZSI3Nq^<;L3YqCr7uVJ
zTJ(Rgs)qq3cyrTPi_1#Lvsk9e#*Zh+IYFG?xca%xIozy2vKbT65GN4T5@HhX5(W@3
zasT9Y*O%ciB`P4Y<30kx^bIXvSyhhY&22Ok2EOa#7_DBuh$ifTPFhE6l*}<XrSVP*
zEO{2(V!W8`9H*ShnOJY6w%#><SG{O4Cy2v8X_9YldI)V&HD*XqDOG7vDdSw?o8?<+
zoV%N1%+)}vs)g<)T@Lq-({!1Z4U1{1EQl<4E~qYWgYm!uP0~$TSFb&U?^UjtuF6Om
zNs%y{NRD_FNySKJcx!pRT^zwNJe)ije8jx8d}_QaCKE|(Pz>l0vpT+Aq6%|?won6K
z!&W_+%j3}SAk)%e*24K*e%14;>yaZTZu{-Vnff6C0<vJTCPCwjyA177&CvqgKwZx=
zFI_}NCsm^Tm;3Q`;dR~jU$0SW89bifr(v!CK33Y8+$`E7KGa&f?;dSu=_y;bn_U_A
zNq%YoSc@pMC-^pzkYYwE^$}t^V`^j1l8Ix60Dq#o6Io+3qS<2ji9Hjm4HQ7<#KaBY
z3i1QV%}$s1t>v$UcKW3Ir{V||IF3}IE_~a>CXN2p9}tZ~wIT_Tv<Y!1u)$*rw+;Oq
zl8mo}fBt5gw)U+m9yf#Shs7GV5;p_SPtSzj&=}}-_G_;+ENoqkI;E-c9FKTT-hSHB
z)5xl!FzVP1kbcqdB9ht=#|Q_D#(OQ$!K#5uDlPA4o&i~EUHWy*aSzKT7W(3I>MwrA
z*Bdb#+wYFvX(=Hq2`UXK27R?Q2sm-k6w!2<54tG7c-%E2tRM_e{tE4e#!{+$Zu}wm
ziQ7s_wQg2Xn=XOLQK#Bwy!`FQN*3b5jLfANF)`v?xx(ZN6brjjNn4Ynz8==lNT$yk
zl$sL_zYKTpz9T}sA;K(OjK7(zv`X7NhtHOAi*Y#_vg4%~7E}pK>D8y&=6x>xIP*An
zt-Oco23Q7s>bSr*R(^6FdtX|~AAEkWC^N>rUe?*JExnV3Ms|`nO)r$KRKT>b+jcA~
zWF0h46ET`G)O;!*3m*z9!Uu?(xOi>mYY6K!s&7Z|Mww8gm8H+5?`i|BZRkzn@Zwt+
zYV8-|*@7zf4)YEt4lB=A&k{%___>a66r(jAnkJemE2Nt&y{3ZFAJPSzn!%jP`^xxg
zKFdstUS+*)`GwPYwbod+@<$nF9!6K-2k1kCZRT|Ax|iN{=+zLWIEi_gQ)v<Z4=;Nq
zL2Z^2asiIxvG6XTAT6R|B1@v(@E*x_?A>Vmwzn|HZS%45S|g|tMqhZOT6$Ino#mI5
z1TGBj0<Iy=b*)Bu<P2*o7c0TjqRX;-`Q6N)0wrMgNy3TbFYHEU=OYnmIGMzJnU1Yx
zHU<k_FC6e~Yep@O8ws3G&#2oz;q$4Jnz1zL3AyxK4h|GSi-u>?N;zHlz!^8gr*$|w
zayqgz>1B0`_6_6nf=%}_k!GW}sY2W=OY4h`VEzm6b@~qf_0dQZLsMVd^@Dfs;|!J~
zd6<y)BkKK+or<&Hm9+EnmOk81EuXjE@-OXAF5GgQ?RE+2t_%rRKHW^{orIkgTZ`Cb
z9W^YSEBSl-bL&FA#-9`)9G!oh$j1v8=_a@E!4hZm(z_hG9?Pi9`j)NTif+MA{6VfZ
zRBKf7%Vw?Qj|vx`&wu;LL*+>`jfAnnm|<U@6k$^jR{M2Xc9FeoeZAe<Xs$W9A_@Vq
zu?TKe<N#>_z%OF+Cy;ApX2whNGjjbe?LO-R86NM`0OlgYZd%<!DDBwFvZ}capGob4
zbD9Cxb}Tc6%dR={<oG0aeU%GlX-9V9RDM4F##m1}OHbAM0!4^71&{cTYRJBg)6dl4
z+G~i3x|FG`EPx(fMg<@t5CNXUO9=2w2!Z&&%Mu850K|XPKLY@QEC5LVY9j}~|2d-I
z*PlNByd%a00#M*jui%$k*0cX=jSk2{{I4=%4*VTJR7p%q3Vv5Iax^iqbuzbeUM=`C
z0<S=`|ES>v0N_#oxe%lj-=4tX&sZp{JFCme@EO_JFdG=#8JaM=+1USq0}ycIgBNW~
zoDC@4Y^-ga_}m03|IvaEUj9?fLP_zDCeBuZl<Kky6k>LcCKO!EAZ8Gy5IO|~g@B{6
zDW9Ua#J{@3p9Cq*ot^FZSXf+LU720kne80SSXg;^d09YgENpB{@D@x?5L;&hHzr#r
zs(%9cFF4{RPDYLv_Rbb|wiJKh8W`HSI15rz{u$`MkAKe7#LeR0BiTCr>$Bh=$nxh6
z3oA2-<-cLWy9)fN<x{Y5GqKhXx3GcZ3_gYsFFQ!!AMO8NZ~i^vfAm!UcTYA>uK!!t
z|9JKP?5g5q;wWZk10T~_=-=!0uW|qL<-a-#u>AS-|3Jk*@%)ckIHHBn1z7%j(S*<y
zjHO<}*OA0RTwWP|hwrjK2f{e~58XfS@ba@5fc=&`{8<DbB`&J$hH#LM(oFi{wvCoT
zg(B|d&x>WWx6a?tKL$H1d_$$rf+0Oe5)YwN$)n1lY<ck!H;Rl83AF|JJ(k~?2Sxc4
zB=Cu3>tId^>^jo0y?mcOy1l$qWdhzxPIfh2hM1X(7^X+#AYi}zt8+&M3ArCouJ%3g
z8N)M5ioZD{e1U*?Y;h5!zd5t;UqvD4v&e8f*Z8acB1p!79|C_hZ}``?M8W6oVUY8#
z|L;DXk^)g=^RMs@U#*=0_}ph-;K+*N{Ur!O06o?>U(~-jMOfZaAhzo!nWI<!RexX9
zet-hx@77!d$^8uiA>gW0Tq)|qUxGmP9Yp-gH5+?@2H)}cG$~&g|7N?xM`Z&jll>jJ
zz4X)lD1z+U0vdm9{Ff9Pg^+djcWds8dhs4U_k)1oe2UP&1OX>?l)s^K_%g#u-8>Xw
z)oE1Fj7C0V@r483-%_1IT6>(5oCThSar8RbhFcXAvaj&tuKY0nadRvcfxGfwauRo8
zzhu*tfi#zC4btF`T0p>dhKKF>yj7XY+h=#*{*p|bU--?LWiB<J_7an#$-#l*>v=k`
z8+3k+5EmiG{yW0=vwZ_HAEP*5bQ*Pe%VWa|XkPvD0@_FJvlxxKL-n_0!Gx5b4%TFM
z+~|)$DuAPu6s<vs@-HdmF$vCC?;X||_Wo`*Es1{^v)%l=jlb*vf5!hm`2YX6|3oB#
z09x4;W1+{3N$oh(6bl-5{dQmN<DIpmHz&W8<dbPL+&bS%83qCg4e_+?JC4w_$<c!c
zRmr8HToNB{^TO>nYyNjE<ATC3f&ifFGNrP0HpDWDJFNGxj&Hg%mff&C-nAsE|Le!+
zt()JG+$^m(IqU#!&(oe^a@Yw<)y2doS4qIV0-bX0oq}O4EBzp?%~yu9|D9ybLP2xx
z2AnkRm$KvQI7bumgNqyPkD7)z24dNB1>iVHt(nnXfO4;Rniiyk_bd_HS&e$@dCD7g
ze4NAAg$=t=oi3-;)XK|aejha*+eQdp%?@&=cfZN1#0@giyZ+9y6)TsgLB%^0e7sNm
zzoY4~us>@d{BR-$XPZdpgA8R!IPwDH+Tb~p<h6O5x+Q3TG}Z8GhyO5%6Vw2NdLR+>
z=C_^qIdRoaPG|~_98Wf}|Lin~GvvuE)63K}-9`0$T#{TRz+krAU~AdqxTG<kd=u@5
zb-dXux#y16Z61!r^v4mG_4~WP|J?#szt5sV_B9ZJoo7BJmDJADH!Jh{98~l*dZ*4t
z3d3ggf4JT47Nj0~UpH;_65ADjZ4*x82z}bl^s(JaHX~zqSnsAMA-1XhJoVF8h5LR7
zOSLhL!$Of7JSoeao@^NMj=s$(^M`d0%{=$x{vmvxg%dI(E|kDsJ5O67&P?j5z^Prc
z3oRsZiqv)vL{qjEkdn#DFO~nPp0Tru#AgfWPY7WssaB-nQVrjqR(V3Tk9lj`(}sv8
zFjk(U>cbAO=7gOXH0gzz+ZXS+O-Ruum((rKKOV6>*|t4`okn=Km2*5!I#5!1x6`W5
z_^;p;VV&bO*|wY}&lgQuGT0#ZmHMHVcEDqc>KyDQwGH12Os|q0+V$DPZ%sqx_gVkP
zv(@>>Z0UW5n47F=s8u8Hd%0VqA!WqOl`f}b9tQ@xo~Ix1(_p56S<8pHy=0Z%46iE<
zv#|y!+fdRl_sSv^)CF9M3L|~d)dss{se*%1ZoS@!l~18gP|x}Qnt31)AV$Q8@8=0u
z-IF2{GyaE~`;N{ZRo~EE$7z^z%NG$7ge7#NzZnjL1L1~cdR-+$eur~IIVNR{OyN)H
zRWN5t#P(@Le!FGMVKqUONH{a!k7mtJJ<ov`3Q~_aXiuiQ>xR2$Ryh2TN4H=pZR2$r
za64*6TE0i(kM`r9(%v16GrV8ccBI;9R^!2oy@U4*z6a|U8oo#+IZ}=#qJBIAt65&f
zU+f?!&#9?{)i71e#aFGKEh&t&VTO;0Q;;ES9=b6a-Q-<-6aDP{PaC6~a;yf~m!<8Y
z5^~hA&GwtuepRGuJ%l0Udc|jD5we|bdN5mY<e8sr%61NAk0ccGtWUP7onu?sT>pxO
zv*=N*97O()&1_ZLu()T<J}OYv0PhZkQ-d`gTrcu}vk-u<YYw}rYfB!_l~lXsHST}u
z1ZC}0E-h)?`x&SLBzIf&8>~4&GOC|a)f)D^-AvqSy+2}ykKYC#-W9l_?QvK$>^Q_^
zmf^nl4!`w8&}ZaP!z_KOT!rm;#z`y)lXK<4^s(f~ja;m^Ak%x^>uFM6*u|h-A(96w
z{PciZcHW)o)(mHT@vdDswXHVQcch4Pc~lmiD{#flp!KrEuAw{%(;pcFy~Sz!5K6^K
z&S0ONiUC;YDNQ%Anztw@k$%1|qTz8q%&pUU58qGFF!qt1Gqzikcoo$F4j~)0#jf_v
zcSdU-u(V41>m^ry4H3dTYs%aNJ@4C}XS<oBPS#TZ5EregoEPkNo84dfAwk#Of)+7q
zw;$0M;2@D-W#g%t-xn7T*&tzKOUGxY!o7uW3o67zvt=E~Dd*P8wzg$0Swr&N0SeWf
z#zFWl2C}bR%cbgvqe<pm8)>ZwFuDH?Tnl$?`^5c$ha6#kG*QpHVc+bl9qcTzjYCP6
zQ^l=tjMpn@!LF_4>`0FPa6|R3RGSX+-Y0ZGjOPweOtu_*H1UbW#VZB`td^WsHwr(A
z8M%Mw7s}E)o91;jw<Rfj7k%4-<vk|<j&6SGBVaDL6)j!0sC4O4vS@A0Y3Y-%z~xC7
zR_Y%ax~rR?<|sX*`T^ym9+}{_F9PB`kzXsF>o~bmVwP}ch1N)2+v8!aW7sA{Pb!j<
zjU7+ySYFRy_xEI>QQajeAMP>iAw7q!ig*y>Bgfs-;)P_meF)z7oR`eu_l!s<Tf#0{
z#t8vmS%gjtmrUeFy1+?Im2AcP{&1MF3aWTp5(|^fN{2N0#2&)qbDD@EwFUyPo8L<Z
zo6c-mujjn3mxiTi(~XZ>u64P2{5!w;Dcaqf7wJ6?68`$W;<b|<1a;^m-Hua(R(73|
zX<?1RMWk)2CQt!(8tIcH%f0kV8?qVyhpZiX9WB}vzyS+{dd}k?wz{x{IfS2XN7LWv
z-$8%?4TC-SMgcjWu(sc?`eQ9}vmWqX%^D4>sOhMvY25GYJ<VI!OsU$wSWkCdo&$-b
zZDI;uD3~cEwI;p|X7|jCg@z#kJ@Yf%Eej|D$xk&b3%^4HiEYijAAXakzRORdRC}1w
zc9M1KBX!La-uJdE??5(fd%P-YkR;Ln^5yP(qAr_|><?wBSl+FqF2?Gy_kq$_R09fy
zv!w!xIQ?Jn8|MJ!qzCglwIIr#&%JVjm&$5a@5EwhWF1XV#YDmYbCyI@24`&k)%*wL
z{#wBkv_|~vM~}>ky$?+!15X>oK;n0~-6*%+*ogHX@O6HLk?>AfSC31SdIlcXx-b7k
z(6^TBPj7pAJWggK^|uWxg=5AiMUvd>7a<LMzerm2E`jjVI({37p}71#bqayr4DswB
z1H<_sBom*(Z?|2F3b9>H&wF^$qwg`<x+_aoa9DZEW96wd_I~ZsmXF-!@zAs<dnnbf
z3i?a7V=ubA+t}d&+lM>o&?j#0dR1`P4r{hXa(yzEJlq2CaUr-C^Q<*Gcli48*r)CL
z*%g$1G$}WNZ%`r9`{A@t?uxZsTI<`fzWx{N#57b6N)aKrmDr-)^lf|D``+e{$;kWL
z9Vk82NNii*A?s;)Hc?d8x_4$Zjg+3VlAS4XAB2W@5ekC?$G@Z>^tdQ^?=MM1c2GIQ
zZ2I(g9l72nu)Bm5!LyD!Qzy6)CAP?nP||DqaC2(9Snm+%U6qDf?GHG_qEVF15NI$t
zUeNKWCW?ad`ejhEia%+|t9_``MuCzXZ_h4$VQw-|l87}p!ZWaSJ?0a$178w2@W+ou
z8?&UrI}9=BnVt9Vf`lK|urN*!FiOkvB(amF?z-dDyaG(dJ+<OY;1m|pE4=05)sn8E
zY>LOU%{nRJeX0A5VDlz8q@?$YRCAa~^+M0c!r}C7b*iNNz1~~FyT-*M<CN+ud97-g
z1t)AaH_PAixaC^A?T-p;8vOy!;?hNMW)o>J;T*hm$Shf~8vpi>Tp=hdKYJ$O3*-oh
zz+F@K^SjrOurGG3=)u!Ug7fGWOA~b9C-0~Gl~g#crwLnZ8a5M5mt1CC{HuzOTsxlQ
z*W)7G;`z%hSXYI!Ps4oy=V`2%MX3Q1$NF)BzaJjXM_pkV@T8QZg-C(3k2m!OApEZP
zeAjfa$s-n@+fXI9Gnhp|dP1ws0{~F~l|0ai0GPi77ag0xo6db8g^`;ZT)W|+5<!kn
z)(CNUeGSsgqwF#;7AIKi;xFladV(E4L0x8aw!=I@%V?y?0bd!MJtGn9;sT&FG$aJ*
zwoQk%5mdeJ^Ij~1HrDw9BOiKes3O0vF?cKwBl}j&x#U=98|OEZgJ%s&JwZQ)(H*RZ
zhXEWCIB^DXlFLl>t6z|f%9p?kC5N~9=7i@4JvWhTauB1jj45C+F~e#_4%Q|TqPp=)
zv?Q5D>v{U|jhYjw7AGJqT@qw1YMN=^iH6I&k0{5-T;Bf76q>|(^l`qDugH~vzG{=t
z5iJ)$g0ZGv$rhO=Hgesar3@#h`y=o6hug@$rVgReFeW|L988Q|(=-cYr!$Q{qKKuP
zWyn4$yJKwnu&B@63{cBx(-vsh*ZDfNXO}k93Ule0#dP%1m9y!n<x}7KI1_n@1)zVg
z*ypB;{O2I?S?U?yAhPOr(+^k?;UtIDG`x#&nQe>vZoZyC*eQL0Q0BF$!>pDnqQz^q
zB`X$$GxXlV<(=PqMP=Z?%Z=RKg3Q*7zFm|G_ggrJ-MBM*FwK!;3_2y3j`qjo>z9>Y
z=NH`FNd-|HlLppbB<_?A1h)^+W*%IiS5IBTRsICOtOb2)pZR;ZCWQH{I7_-MDLz2U
z<*-y(6Afj}WF6`5O-c1Udc}yKm>2YC+rdu7w5C;qio1mxNGf_DS*CwODA4!N`NO6y
zf0y#K-Tb}ll1t(wBzNC<li(t&rQ<UGf{v2iXsD>Qv+`prQ&;7}9p;C1^857G<q<}D
z_RzTVSqrKscNT;@_kQADBf=r;d_)7kki*)Ak%Ok=R+~uYgXSgMbom9pixZTPJW0L8
zg>;#if61_3@&<=d!Rz|xb|56rvjJM-qx{*QZSJM~sBuhijbvd-^kcR;zF#RP;BBX|
zwQBPV4EOqo%3!}~Y$}O*l~!B0D$oTC7^|3g-s?aR?iTdsrPeKFxdwHZ3MqlI$@eCF
zUAGRxzBl?7Pza)kX?<lkO&~8p9h@e6lh8t`1Q^PiE7{o$0~w2y9*<nxUXZ)?@~hjj
zD}+_N#9PYlA~=TgN@p$ds!9s<nlU&8pKwv3*Y*~PCV+kliXs}J`-7?@huS!!02Uwt
z8cPv5+s@V!-+u9Lp!HcVxv=^;1qrBYlo3%Rjs$qtff&>oRjuoFIkjZh8H7z)Y)^n{
zAadJs>^1thrFS5pc*@f$#Z3(8HJvG}3=pMbP&T9p(z34TS>)nari){%fZgq@#gh`P
z^rM8)H(N=zzSHbQ#etAO{3`LO`D|llxg7lc`{~2BL8@1Xym1JFlF_(cnsCGcD|jnD
z@GRtgH{n|`RB6$1U7E6a9tQ1cr?Gi>jLm%LE`fc6_EyMx0yUD=Ms6Rro{eYwgu^`s
z&3BCr=hGh6OIMd6#vV_5BJ9?e@rwy%Lf;4!ZR-m;-7`%seF}>rSRy<IiktiGoqDNh
zQ?a$I$5agVo%Ic?2<|tWOfPh>l5kZudl_3g71JZ+B7Io$5hxfk$B*)uc1dw4%$xKn
zHb^(a<MWac=KJ6H{Ky0;ufP?$bVi^W{}#XQ7U!AdwK#>WuMpOyB^;B-!ht0h8ivFG
z_orlFxA^xAF2Kdwsbnor9w{ZFh<u(8`R9Ja>W=O-`+z-D<sF>KQhE8;)>j!vp&~^{
zNyT49@y>q!uu^**rxTEg(s5f081{jklXEB6jaN0!Xgn+Kj<k&*@Z6D~<ohO9qY^r}
zG%`UC;{*$dj!we;o%ho2ucMPIHdXWtG~QnK>CcsT1yKqNwoYcwCqs-)qaAg*m@d3m
z-q^!6<|bC93qW-RmcU7SV6<3SpPMP|-Z^;lk2**6?4XA`G*8keW+7c}9(-wW{!1A0
zTP8D<)nGnQrZr}t7|Ag4Puo^et%@I%`WW(jdk4}6>U74ixsHsK@>$WT6tj}k4~BJm
z(g%3S!KA;Wi^t^+)mDojBScZk4^<37)aN7ocAJD&%CWPei-TTr#cwnX>!zJY)e+l&
zdVm$`&PuV`IGKyxX&wwPv@J8X1)F|;GOS~kv(9C$VDMBCGz!OJ(4C!ot8(o&Lj+Jg
za(bWUsot5(2`F}taGNht{+UtZKM}0*OC_T!@T<x-npP4SWZllpTJ843?16~ErdHZ)
zi58BQhN_@h9fE8ztKlXuzmp)LfSdgkdsh&aeDSH;CZU|+W;DG#)kAr2wu(U=t}Bw>
zzDIPOH9W}=oX)5PzHFn<2b+BF9_jokQc<-07QC9ZM|hU0^SSCQOgI?Z^;m6=%@hRz
z$0|+>_Xq@yQ)yb)+iYC|(^mLX!Oj~ZAm>sta!AgRf>%GfyM~yI8+8al6%ymj%(Y_a
zvxFy{qjOU=ubS=@R6LCa2L&8U>0grGoSr5g)8Mp`%$%(h`<qI6*UXGC31{{lg9SZS
zX0^0awBUgz+#~SG%h2g_cZLUHSs`nycQr+0G+<TxY`fVh8J*TPIO{`iDN7X3B15<%
z{#A7mZ^(B?cw>bDFml&A9m4c(;*#HdpO_Z+Zd>Qo^XIuQP{*dLW7Qfd-lBg&{Y5)g
zdY22skL{LIc5(SME@Vs9=2Aw2;B3D`xiSB#{YW=j-Dv2_Tf9~B4Fv@EdG7gJ<RRNb
z{_fmgjq;C$>F<)~+*`+D9Z;+3t?u40Ti1=y%7t8D=-tIG9yy296v;1)uz7FBt2GKa
zG##SyrQ<z5dQ=QSn$EyJ@&&B~Po=sDW0=gzD-T=TTirPF8`bMunLgfiE|?6gtJa<M
z@$C<LU|2^qm1H9m<?XQ0+$ZJ+W0M{{ReQr!bgRX6(~3LSS+bkNwukQqm$jGETBCa4
zg{q1r;~jOK6S!pzWKkJ86e%^pZK*cmBv8dS?(K(L9A?D|t_%p3j({R8_aR$VCL`S9
zvF~btx62wuoq1{i<L@nck9%s_YPQtk>qn$oRDbm8E8!(ej_LNwN*nrx$r>V`S{f6E
zrm@>8j7Lt@Yf7PPqXfPMuAm5}<AYVV%~uRee;ipHe<CXkGw@v(5)>`h==EGAhtOY#
zc4!)IKyBChVZ?qTRDnugN2nyt=CC?5&SdCTP)jU6at|g{6qlDDoYNN-uM4l$wh3MS
zQetS+kUH*s!BYHSd---0-S;LLx0aG^Hi(7Es(u8oT^?>!OsOVc8q+z_YwK|47elGG
zJw5qctgD~ts}kUmH^<C@`$`VLjcuciIt^FjDCEhXxtG`W#^;cSEmgHVk{~-UNoZ`O
zfQ_#Esmk>ti4!JUQAt&ZhG*dihT|!I!lyy2t&z{yNoc=UK*z0*rOB|=$y8B)Gj;|H
ztz{YRN9V2IkwUCPm>P|Hmmaolqs_dhOzaMw;#H%iJ{;5b<HgN;Au6LMMq3)lDN{B)
zm3x9mKBoIqsbx(qm5-Zo;P3s%HkG9w86MZm4sDRLii%QLx}6_nxRDrxdnIkHsO0A?
zTeAIITkkIpE6Fak8Yj~Y@<O-2!Wxgb&U>EpF2ix>xt1UHr_3Ln)=sAq_1KLvJ*Uf4
zRv4~4=Xdit%eAcsK0u|pyz6GmODnge!JoDW8OnML{ep6jn^&4o?^1Ewc;8^)>lm*6
zv<mkg8*$sKFoQl`3+z1HSx=amjvZ8`G;F0$273KSUdp#g2u&}0;MT2$oCQl}RxbFD
z7$<01Xf!<=sD`O_89{LYRYGyDCLmQIQ(Q|D2GMac!%cHlK1|Xg&MHNx76Edhg+1OP
z?Wbd(r=mEV<tiLdn2NqK#1Clt;@PqKcl+J%MMI@+eF=xv_v#&fxb7O4V{(lSHs^o*
zHYL7?nU{TIJ~!XhssmrZAboXR@t%*DNs{wh#e+SrnLaScjF6cJ{vO?YZgI$vbaPiA
z#?S-fW&mHp-Tmc^o@-@beaDxnio?8wl<UsscG){jYlK76BEPC<yhZz~QG(W3Ga(Fo
zR8jHYG&W=0lLbVCwBa_lj&KNv`L*?lJua52ci_8S(G?rG8w+e9>XyPgy9D<i*?u=)
z?+Dp0H{k>iPQmq<MQG^H`;s^(Rn!)^+Q@K>jX%gT^9_+<Bv$^9$F^<)cda~jmVUI4
z=S}8>@FxF;o~pX(<|f>qGB-fb**$n{)EnlYCDC>@Zxsk$!qc*jPL@PdvQ8-JjRkc=
z`PbLgLat|cDh-`yuiuu9k=NkNFB-+!3`VbzrE(aka!MXmeBrqbbJ{3Rf4KvGf!fx=
zdmCIy$_GX|dAvSxcOy%1;$Q3OK@${7(1M2zbc9A5x63Y--EWSa#PM<%M08`02VB8l
zVuqZoR|e6S)+}pbOCTT3Tun8+IZcwep^$o>2|^9j1#lB3BU;(YAz*GqI%<QQ<nrBZ
z5R+SS%P7=qkjLc>ge_F+b3z8jF;$hAB%wb4QqEye&ZQ(lpz5dv)@KQI6})uZNARHN
z{9Ld;vmQsNQgkEFV@Y$H(I(npTsocBtpGnll--yg<QmS+c#=1Ls@6H&2QGek!o0_g
zX6169BA#|mCQ37)IWk$aXP7#SK0JG36X|zBJc<!;uNdUXe5&HdB8K2Q#Sw_GjM{u*
z9%7@^9`)X{`PbEwvwmN6EK_(l5H2xJ^9AJ=Qm><lf2j?t;uyFXyR|2PF<kI$aQ2+f
z&<^E|&G_pKnN!G2One#z0anHg2alHfJEIodfy_%}tGa~Agi5-XuZN<rwl_>v>nQcA
zK4CxN%?}pa%fJkwopqV;P|y0jl!b7}L&N}^>WUv%c%tlq?ibITr@*u!%bBliloB3U
zsFKUUOCtxt?i>t9CdQ`8j%4{(pXbQCutxLDNqg@r_6*4r*!psA>1l@8*0z@vra8`D
zKsrKN;jRwnDI?a2T$yD|L?zX2F`kf%=S_tr#AkHwx*URrLx+@WW(6<7GA~@htEtAH
zzLP%c+*vMXvuDdjfj>BRy?ODlkDP9vd`CO{_W0QTxyT)wwQ$7n&}@Z~jwUPSpvzu?
z>En9yVM?=s&E_rJA(VhE7L-sZ3CRS5EPNwK;7bU)Gr4#^*Ws&ke?a*`Ky$ORnN$pm
zz&%4!Z*peCwqJC680|tDQRMYoI>38LD#n%%EWO}Au0s#trjq;WOKQB3n6<7Tek0-9
zs;sl7trV*?HloccR|tmK9%0$TWtor5g2Nh1=a)dnl{|k^`W8dNnz8oUqNj$z&b|OD
zT!GF3em64jo3&D>cKz{DADB}W+}yogXaG_e5Mg$%N|^)t;fYrrSLxaOH>>_u3FEzE
z$8et+D!G%Euo9trqDhwW+eo)3P2QXvvD^9n3YX*+%9L$_9fQ@%Xfa6Q_Y>QTh&IXZ
zA8DzEKEDeHF?y2uht5R6q)_@fX<fm#QuZY&!!h0HuIS4!ok9XJe5#`xwc4;#A;dBu
zG%#Y2kh<K}x+?5*p=&>o$Oodc?D;A*7RxLFfda9Bqf{*D44cqeEeXiNukrQN&$Yss
z8nzjKE(A>HsT~5GeBajN@;J7v)ABwitvWG$k+Qa5V#3V(gnoW&mYHgipWf@jOcUm_
z1Z^AhH0{%=b&_zpT=(c>a!qxTYZ()`8^b~V-IQ#WK1cY=>*NS!L?5{MaNq=FlYDv{
zqi4CAh3K&#&EeblKPa8A?gR}#dogOZmEl_Y)9i-T{Av!0RzPlnTcJmI_??_0tqnGJ
z49%0LokNHog>GLRr|`3|{?<#^9}#{ubh@_5)p2XRS^soCAW7a-fAAQzRAR8qu{F*b
znd*P_Smkk@izUoOn*q+#0QZ!Ir|XtWczQgZ0s5Y|(P|~T|MpVL_gA<TG`$ep%A`#*
z(9Jx_CrwLwRYdwV0J2h<bd)-ooLz*;TDRm}z<aavnD!>!hZXO|1OLsnWD3d=q~F6v
z$n6Qy>UCqy9HfqveugMr2Ig7Gd!D(Snp)O2+<bTD#t-83Pqy4Ev|gl5mABDks)4SX
z7DcC~aJE&s5Bq|lo||1aft)V|PwIkF9E|kr15=4@8Z+iLxrI;bh{>9U0a0A|fzV&W
z1kky7jU~JoL=?>85EMjAnK1p6Hwj{*JxBz|5PMIj1)DmTLm}6-rzfL@SDnF(`H$uB
zG>%lbi-FFwJ6)+jp=~1L!|f`H$gaMU<atK=uBwsx#li|~x<=Llankg~!Y8+|&v`4)
z5<vQ|Md}weLDTx3*oSJb^>w4)r{bE*{K`J;xP-^d3TO!OsBx5cLKQOEqlU5sW(-8{
z!-A!%qs0(kX5#<YSLlpcJ~_b*z=$Qe;X-V=LD^T;CJXSr#`;bcctv;fjel3l2Q2^v
z5+J20Al8&Y<rO#^Aw=r%{ZOI_#hU=H!v^j;EXC?bnCr6fD9fP2Fn2X_lXSo9uz&bM
z9p*Y`f?stu$S`xBv*pf!?0fM<N9*a|EPJyX%9iwT`3+|E{YN|G@5$D+za#h?!TQ4}
zFLbl4yCS@4ECNX&@K9emi2(@5NOgV!-1~!lF~`$Lg!@l8_}6Di737J>?S+ns`=tXn
zcpNk(l)15g6J`BN1yZbmK7QjL@Bb`rJ_fqk>=`2>vyj&S_r<#g$>sG(frvFClaXuU
z{_{Ly#}{IW)uI?Sv-UCAD@M<VZ!4nDZ@)Bu^hmt!loDENiesg@{pBQ>OeQ3=LA8gv
zfm4YCkxYUAdS|A3KN?%huBFE3yJPnW3PcX34fhPBo%X^7{q`VmlSCp9-;3e`n_#kL
z2-gI%5iObW1iNmNGmVW{Uj6DBPz=x8@zQRQo1>NM#uani;~do687#kmjRj`OeK=Z0
zMcLkPup{`LX>+Nb2Q!~KzHkChxy}@P5bAnG&4)~9F(aM&!(1l~rn<aRafWpE?dqk)
zH*1=eSsjEl><BCze-U_)sf@v@8^7x#jd;FGI_x$_az*a9*WSyG+{D7{%u8duGvcl{
zBy5q55%8Sy12(4t>O`Jf(ffomKjyDsR*6#!HRU6BpO6HnWvoeDvVNZ%c2Gz$<*4Co
zAi@>`3d!mo+0)WRwA3oTw!Pn^A0_*{)I3%gg(S$cXBE9gbvnV%Gsrz0Opp9EkN?4(
ze?memQAXz`(&r)fCuuUK<lM@K$=u?D>)YUFasCi@GRUXYMJ(mq+z$mnEVpsYAnaX&
zttB=#_Wb6(cL@e$zRz#>)`G~vm$)u?f$_iI$h21Sv;2wv^-f)a%))%HoPhYTCUdQb
zkPsW<!aBlL>e3Bzp0QGcb<E$i-AmUl1}_J{RUZ1W$r#nLd;BDm?Q1KBGAfO9IWy7U
z8{Htpvbhi(d&~_$k{gq*hw9mG>G9wa^!nJ85THn_)E5x@%Um+|Nvh7uFi$eM`SAB=
z)orExrwLgbozYxqg}Hlo+`@;hNKet3k1ND|NE-81)mD=cHf7@$kVg;>EJ7G|^Rq8x
zFR)3LR<jmX_(gFwh#MfDD7BC<&{Az<QFH@Cai@1yy(rtED$p4R%g$$3whh&?8}&mg
zFX@2hR5^*_KByk^LpyjC7;D_UNpb1Un$=Zl+v77AoDEuBdRKMV_hASWMR-{<=sM<r
zZ;mOV{*1Xy=(Okvrq+L4*>Z%WcU=(Ou9xUMc>5MX6F+bf%hV;9fIA=_Nh5w|AX$$7
zSwskuKNV34T`a2656}qClPR)q9zvj?vP5?7ns3GD9RjEnH=VwJO%1t-&6G%?1{qr0
z!a<q})x#>J#f<P*f#FGnXhc+D8yOk*qP<vwIBHpqd&W%p3m>EMedD>DY^m=7>+yax
z$QzHoWZ~#%pdz}@4MC-02>Ml?O0{9p@&wBH*!X-vn)kU_#^};I$L7|cP)e&nKOdU?
zB<{@aekubV1?Rw`a7UIQK-7Le+ByLrA0agDnmM7kyG_7x#HUTE4@?MXy}$C<>b0_E
zeG)j8IX??6tDm|{&@p&d1(R4|_20gu*NgBv<2|o4l4JERK3(QbOgBUxy`9=?GFTLT
zG>(}w?IML%$w#vD!wmB@?ALCG_3s6G9dYzB&DEkj`qNx{KbQu@C$i71DCwv@T?;?4
zAGclmCExU>ZjC)guq_;4AT~YIKa{B~EIFOKuA3htqzPQP+P-l(#N|$6q46=WZQEpa
z8i1%Jg_iLTrJa}gq@o=A%vv|^-kI*~G4N7_r9Qb9K@Oq|pwu&&Hl{>8xsb;iZs?Ew
zW6)?gsaXX7gR#?834V&o%$<8Yiz@2ZNqFqo!nXd-)ag)#I|VJ%gm`t?YoP@mX@i+A
zdu<C=-EvVehTaylZip0#z!sZr)g+qlug16bhRW&~-nN>-TsneHtH{GnK7%UMSJz6w
zCVgsp?j`Hj-c3&f-({;*-a!2BZncLFXQ|{W1G8TwV}onNR3F8TnxB^Mw|4vbQa(L-
z#gp16_mJ4%=&BUE39GhtvK)UK<sv>%kB&cQ;)!gqJ1D^wV|vn_u2=v9KEJA@#P=`k
zHwMOoxC{Us&Om`b(Z>i6w-zXGFB9Esh1K#e*#1J$Me*7|O#gd0D;s@)2Rxk_Mp!@k
zcd|QI`-7gXHNf+ef=ZcWiDTT}G?lg9S-M8z?nm7l5TlCm&VO{?u_$mi5<V7gd$d#l
z);{dY`#Ab5z`UVUgZ7=G1sM_i$V9(}HeBaPf3?v0@fgO%aH$tG>+kf46V1C<6N7Y5
zjwiVcdUTQLB6-%o{hY$|(rwRrD(}K%|8@DbgWnX$#)_7IVhj>o%D*1XkA%E_)*MQi
z=;>$Z&y!VdDnYeEQ?g^$;qQrF1SSF~GMgjn{|L&?Jm$7f7Z?gt>x``lA%_x_e<g-Q
zHV;c15v-1zQnkU8N_uS1aEru(aLOluT`MI7fEY8vv_AD`{)%CH=9!A%2N~0|7Y4>`
zc(){ow8p;NioiCFjWnJZwHu*BlX)qi?l{(#$A${B%YdW|^BX<GIA2j511yw*g`UbO
zw{g2(h3FG(@(Dn45*HxPE?DM`Jjm_SN_6Y|Gq5RNgr`Rjt{P9zyP`#X`#@me^IL<e
zVJ;_st_^Hsa>TVfEX4L~vD-a0@85hj#z4)9qOhQ2`;k2E!P-RJn`m9Ml{PVN;QbDa
zuP;LmiyRLqx04%77WZM07IB>bFo8nV0t(9pvYC_d-3e-&A#-d#7X&`?zMduEA18m`
zt2k0f{89cP@^G*`kNgW88APf@BI5$*7YGpe91`RG{Eix!2Dfur%np)tTUvmhM!()A
zJVXkiY@hR7dYS)N&9vK0s)L#LjJl74k|WvB+3$j@xhpSBhUdjwJ#h;|_XA3%_Thrk
z1?3FQ3pr3<W-N>B#r-{S;!S2$uuQkHei|P?nD_LMgP{F&3zCb({8X$4@YUTw2s~iY
z6#Bx3Ze4QWi(}V@+6I6>JnMcI9*Ub>ZW1k{bi|JwjQb-Z;~TNNWPn-U*j>0Tnb<<O
zp6^C2yF3@R|HVLX^O-Ce_})u~I2w#^A@?ZG?-x8u;A=(*pj0cgGtaHuBgF}o!ddSh
zbp2(vfwIBC*8^_IxzyK~llSm#^y5OX5gkPJf`^O~oi08_4evBY8xTZSS#sZ5tdRMB
zf*Y2fbg?5c3<v5Qc3>9dY$p-pLbpU3VoL~CdkklSv33UE-cYzVjrDIbWQIIO4#b=(
z<sZWxD2a`JTJxmUXsqIW3ix<&EQ7c<mUnp<tmZ81?+HKG3Mz!HtJ(Efnq3(S67>oP
z{)e>uT#M;%QQXKQli9p&a;^eQfOKgeLyR%Jb`)HDHaPB<*OE96HRsU??t+*l+~Woo
zzETH{6v#Y%Pn~Zxn13zEl0sIo0UozMC%b)3`ttD81=fX5;qSuz$p{#=L4HYSYYbY{
zuf*l}aSYd3ISz@(-^s59-4R9Z{A6~-BKx}yoKtu%%69@-Uv3aOLw9vPgw$Gv1oG~;
z5FA1)+1Monr5AVh$Di`8v#_pzlkE(QUmF&+mj+y1zn>X#{4aY`eVgfJIYvbUpB7#D
z-QYu2_>Yuwb!w&IDlq8B7H<&&+;LBDSd>xa9~6A}V@hDV%Lw-~8cf|pZOY?8KhyRp
zO;|JZCT08Kc~gwY@OiV@i&me!wQ)?6>1I0=^q<QW=w~TrH-Ssf>d(@a-ZPyOYWXQ=
zSZuzPe<)G>xs|G#t#BJwl4W^>W835?y-Da}7~jr0Ixo#sf&TXC<16+^WdR$<n0O%}
zy*S?akfwpe{gn}|+LL#9O>zlLMQ*YO6rP5-pkSMmx3$^72McWSXjHLN)Vsz~1dK2M
zR_#3@9&krIt}W*pHP)PHO*j489!$eOyc^76xjuF9{-360qE1B?O=~T}rkxt;VV2==
zl`5gyrHrZuBPGKrM`T(Nx;Tor=fh^xKsQTY%>(y-md2~5c-x^-yUI(6h|B&=+t|g+
z8w{1>@TZI2jVL25=j+ji^|+rNLNeFF{Dbi=Omz!yA+Q!n%ZxP`x>(|;Kp(r36pv+i
zD&6vWs{=eTHRIJQyl~*QA?4%fUD7x`W;{-fCk=C)=om06Th=;FJV|re6^>C8_WJ3F
zA^hYzLwzJ@1yFYpd|37lI*OOM4S&aeEGz_D?zUA*V{ABBA83RBLj%ZeF0Ew-RmQ?m
z`S<$>cyy#;w9c^d#9%a*mbq!S*|_7wmVEJ=p5rhnJ}r(Axfbq-!?8eGn)m5SjNP!X
zJe%iZO^g|hkb=vKPnC7+8O*-<aLwvflV<Mkz%XsY-4C%dw8E*kODBRO|B5t_o_2#T
z(7n3o3AEFSJEXF&LtBke=yB$irprr8@uz=KSNMlYsiuw-5xoU(7JaQ>5bf)xe`ERZ
zMh*<(uJD%%COs)>7=6y)o<1x39qIw2Tobq$6MG!pk>pR>+0(ZAnmpUFNAz?k9@I^5
z@ahgt<1$_)kQ%JA+>QC32FzcE#FbqoJ>6Xjj#d94YKUTkV+DM*4BiIoPi`Vwy;XJ^
z)hkE-60AabpD7#J9a?FKQkEh#gm`0zu}5N6)L!J?M}$CAy;2f0D(K&)VWE4e*g?L5
zc4wv9pgw6=De(c#Mc{hBqEe!WS#N&o;FVRFG7fqXlJtncIkX$|jC$Ihvx1%5Dr2BA
zsvvDrC;)r+gImd2QSYZFFL6T*a~yu2s98V%lzS7FSwGF~Iu#&wb-Kezxr9wH=b&_P
zhv+8GH0nse3=z3<slx<wf(p6CSr=wwWBIu5RfKmICqdzE)q=`<oyHjXJzlpcE>(xa
zLH4#1QQ1w0yFD<VptIPK>i1jujFYm->1u}eH<=9)OHpJ(?cjj4ZlZ_V%y_D^Q6g;q
z-f2+2qo9|e8jADi!-1Kw(_`a_(CrjaPvd>ucW`|r<8Ek7<Uxpt%359T>!Zf>EY;}f
ze)%TSvLeLFb;Sge`O)S|;YXo1qnqeWyId=K-$G7wiwpH757xc)#*8|9D%FhgUYb%_
zKie+Ma#L@@<PVv%M4{9ge7~TH(%RL?$#U|Vzg}DrRcG$hEyUAnKia5GJ1&1lSwM)-
zn*!UDHE7kQ;fU}DcWr*Mk~yA-f=wFefzvx9yH;bv#=J)h@g1+M#ADRE&Qlr#xPo;y
z$1AdJjMNMV&S*q42fg;JF5r^cR54M5N@E5S<GAKtK;HdjLS<#9+eB?nM(MV>@PuiQ
z7enc{W(J0`%FZc^1n?^{Llh7Wzq!VDJzDDxr#(ljy#2E_PPb8X$2coi^@LsiF|s-w
z>=^j3Rr&_<o!flD7>#Ckz<xHXdJWuT^~$!v$ydUI)$AUPWYqH&#->CD%Z|FLSLctF
z-rYaC-ea=vuaB*mZnK`LXe11G&hB{`V;mCR9l;`nzNdHp`iUREZTMYsxta9SKg_6b
zn<eSaIDvswGSAYmMP@?n>&A4(=sX*`gRkfXb``ylc=;@<Y8+M_;TFg73<de#F^ubK
zxoVwD`Tn2rPGz6+bRup2z|!;C#8t?trKE||4RJOzd2K2et_NUK>}0|svJVB38DrOa
zVl;8s{E+1u>aP9W^Vv;-vf|w#NA6+GwjwkTPD^enZMZyL6c9N3omYnz@43<Rie*vB
z7SY}t0(6VJIZ9+-4MdSO*9Qn=E?%6>uKoFf=9hc<8fWM~Y$dvz%Fb?m?Rky;U8q@?
z-@q)g^e)Kwv9=1F{v~{{ewjQx2b2NSpYfR1Gkcl$V@Rv^Mvn1MSe1kE5QxhS3Ns|x
zBuhI)X9?DE>39bfJ4&mY1e9A77F`QT;DGTomfGN%Nn9OtZ_0*J<qbEn$wvf`eg8!J
zR0t`V4u5hkGq{FPGn_L1i*-A90_=|<+cHO9cUv<4KMR)YS1|Y<c7Hsw9SdLW2@Wt;
z4W|{?UV+TS6S<1JQ1_dQ<Yb8muX{M$&yjtH%Brm$=ReR=mw7AN>?Cr%JZ~E!0+9La
zWyT^X0YAbVjvSKH9$X~^-CQdB*I!U`N3;r@HzYAWaGw(Z=2Zw{A={}84Y#l&wyvN|
zJI_;X$8IzqnO4>*?lji#PjYVEfO45u^vBMZm*u}PWomE7O$V;{AIlAN(N}(JBm368
z|EaN!^_#gVC7t-k(B4MQNv>cxf|7gI+XtdUUE~izn53HQ`U6gaHh^yqB+}Zi>%znK
z2(4?1gx(6nGb6ip8l0Lmqlqbu?0;nZ_G;3MWN(BLRUcTGB<es02m|Ic4wEGWg>-=e
z3o3!yON(-dB2DoI@0SN0wF?BJOgJQUyX%KbcQR(<A(gtWGj5o<I$_89D<MPdp#69-
zIN!E&)mzX^O~^5XGK|5H9|%a)w0y_?jrcCSf*yT&%pB=ALJ|z_kuDLKmrIC`thRqI
zl$+U|)#iF(WI?59D~6~B+>d$}P6B(W#y=!j!44rFER%rQv%vY7H?GQq%P036x37{D
z0764GC0P)-O9ZFaASGbbDFPeS=EfbFD2_Nx=Rn$bSMAeNM8^cyjS&wvUlZx~Q2t6b
zXBo@9(RnDYLGl%Q85_^Pd6r4Ih~clE@Xnb7U8ukst%bgC?B>CqSt-nLLX=h;2*-eQ
zDslHuwJoi1$s$K?30_*6<sCqavASL0T1`<_JsSHOr_YG7N)wy9*dv1r=#GBFR1i_b
z7?tf5o`wvT*teh(p-G&bswJ-F^-MA((~`w{8m_D)Z=q^b4=byh@CV}t5)h>16W)UI
zB0p^CXh7mI5!?6&Ui~&~4KlE3k;2*0v0#X0@$g5`A8$pp(QBeK3pdc?3m?m6TT4UJ
z%MN{aFXj`Pfp24W){;>wi6+LF9ITf;q%|@$=-sN*8>#N1-VgX4tnT8lV^?)faQhQD
zWV$xwl<7?qt3`-)<+CFZLImjj7%2d0(&5Xbd`z^p+TR*L3C|dgC0ov9b6mLpy!9;`
z?=p4fUhye*X1wmJmky1^IL*iXQN!Ip4^7(n(mPtac{d%(*q}-p4v(j((CXmB2ufHC
ze}c`7GDP)6k{_OPvR@AVmAjhd3BpbgZuK)dNIT(lgopF7H73<IZK;kL&FU!jIKF0Z
z31ut3&tiXp^Bf&lZJN9aSxoYctndsmJOjU^wy*e3fLZ<_l@Kuh{nkpoGa<RmBC+Nf
z#b{trK{uCzbkLi9?jpt<Ppb>FI+Vf?T81GG_Y9A^MaNIm+nS9D@O%vm88AFF2Bc4S
z9DFwt<(u5rD#Nz#r(psVC^0g#z1xS)$|1H_npj`#jKg)rMeu!P!i3<l7(71%D)3$%
zPQ!PEeL|$X^Gr5Ba~KiZ4^$A&2|Nra0<x#i7?_l1W<$+D7k7_u+J-#?asMyo-a0DE
zukRZjKsptXp}_*_kZzGs5JY7FDQS@IE=g&m8wBYF=?00RySux)&K`gF^So!R^PcCv
z-}Bd5i#7Z)*UVh8_x{HI#3G)6R<cfTm7d(_?e1o$oGTmqgw>r|xLELnux2Db&X)S+
z{y`kk*(<U97zwjk$i8ZgmZsxL^?v-uQ3;y6it&|8`}`k#%C6<*hO<fiKAyBHVqyD#
zEhj2Hg!?5^kNCy-?@a5K#w#eTeUkb2%Ej~q;H$aEc5IAwSKrA9VdmZ81?-T#2l0_q
z^+&$<WhUj(HOX}=RNu^5HpG^R@&jf}a&Ik_x&YYPNko#1YEqITgG@3Ns0wU9YSawn
z=re4{2;y76F2%y%3Dv_(gW`x}DGn86Myk249GVjySVKV@h&~tcflB|E{wTjgA!iq%
zjO+XrXjvuBOg8?aHQDFb{%RX^>rCYtiL7GbfTmYB*lW5pHoOH8>+O-8HK@`<`bPrp
zNYvrlgLr8oHB9Owx`RX+sYqY=m0>Qmo%qcXF+1hoBWaaS{P)=>{7;30>uDr9WzfDZ
z*CJ6HzP?~uI$XxQOZ@5IVEP813SWozf@}$5M&4um{S3$h{G$c%S~{@##m5tPmK)U*
zf<RWXD%<>9QqNyh`}$hi>(`sU&}=U}K@mal=owq~%++6$Oyda;x^i8qwWU0VsZBEm
z2JN<Y6T1FDV;6X7xq2l7YwSlAvlD(lMyJZQ`-c^hj&D}+hMpn;wqhD+Df^{v0Et0#
z%qs7LnwhtrbsA2}C`ZYmEBp6*Rmy!30t!3^y`j}~Asumo`T*0wH)*!<%W<`fH>PXF
zBggxoD7|89ORw+SOe=EaZByN7GQ7LV)4m_8ETOLJ?O~Ltvtbh>LvOl#ZAx3$1SuNu
zKMcCW+eX%{xUylG1R8DTA&2c>H&VBQUR;s=63;#JFXyO9rF<3)l94#Ye<dUAuB|yB
z8G+Ka<^+hjktMdn6?K)kury|!X6gLZUS7|EEo0Sj2&;tSa_1DAjgamY%gpF#PdxFa
zU^1;>EtU96xm&~b0RmJ4L}lo?30V!JKU7>|60PU6y1(1X|GzB^m-cE}3^nsbTaa8{
zC$X3H3gv3y+_lPm6={jF6?g&4AzEgQy-(LkW6%G*esa5BQOxj|0n!VgiOO{JV*p$B
zxFn+xK7*x`sPkZ*8_gZ#RgGEYNB{Q*@}nON<n?|jeogzPeS#4fRGSuRDR{jFSQko9
z^#6pI7#vm#9GK}r{2qiet-Oj#(8pcg0c1MPzJYNJeWy0A_wbfq1Eq9J%{+rVrkPzv
z<KjwMh<2<NwrB)1jbeEx?u6~+ztW5@{fd^A9$L0TDQF1DEh0`>^LgbM{^&xWeGqfd
z^Xz%KcqpL+fQSyOvE1!X-2`v4bfPu(Vgko4J#}V+8WQs^b`-)>bMY-CJnkyi`hE=b
zO@BBeV+)$CMf$=aCP2mQ#%KiLvHnfdNpDo3${l>m-xJKHd~WvpE989H>uQ^NG@{pj
zs9|covKij>%&4$JLN9?0!Wdsj>QXb}#16svwLi92e^dc>e_^<!>>M#oZy%!Pz5TJ0
zMYcX)t|iivQ`fPV7x^`g#OYwimc(QJ3;~&Lj+*WL?v&4iwn%bp^2eRvF6K{Xj=AC3
z5S+Q;D-nuI54jIxy1pC<gx#<Atn?BWT51^*IsYEUb(vPz5#^dsH~Qs={+CgPosO#>
zatS*xUK7^g?K-_PkP(%K>y{Q}?xl_Eth~Bh?VyctGtTT8Amh-&z>q9gFauNfOI<!E
zSzUa6G_BmZT>$-U^$Wpqp*V%7%Fm({pvt)DsB{)f&aHm;ZMw47-0x!?7KIr_Z-+Fo
zkGi|zL>c;sM>Os1KiSbQozunnsx|Tl<kIV|c>)U!O;2&r=9sINAGmc=WOFF7t(H%j
z=2p-4QnBGc2Dbi~=fh{9D+&9KX`T`%yi*$^9FW&`e_Frm;h{Vm<j>#2m=jBOIbZOt
zQJCsq`+z?fMdLD_6ru3>n%d-Coxcx6Zz>;XNphzo^piFQNn7+zRSYX6+VO^%DtKS%
zwO&-1;zC8sk!(q;xi{C)lMsOiXsN3SBlZpj6=V>uP5z=542iU$OPpKE+_oV*4BS%f
zuG_fR9S*4m{H0`@@;^yG>iMi6Mje3~#Vq>Z35?<qd&=CyTxPr*6KhHajY!Q2I%kJH
z-Hn16kv9$eg#nS24u`gkiE??j(%ipojxsdr){fGB&(gA{Pxqn)OIpyvYx;z)jMs?J
z-W|7ejs?zzQ1FkcwRe2y(x3b@_+Q0BoMP#mgZ20I*M^zi45%P#zr@yHIGY;i?PlQ6
z9!V?nj<x@zrPak)8mM9shPk%h5ECOK9D-`>=m}QfG2#_d28R-1;UO~|$X>(Ut<L%W
z0_K#7qwhC&qy`i`o<=5{fW~7tXpalnw!9ta&ZpV9H)`by^FHgN%KoHcVNHL-e=CJb
z&KE&(t8R~@xHlimh(eOrBLBH;(;(1mZoXC>5%Ldvh!j<e+Zytacwi5u+@jI!*0sp!
zyj`Pt_gK)}6X{i;Bswc(dMRFb&kXqb%B)a99M7A*Vq7BcoyS=6dqqg{anJSK__DSF
z3MqLd5_Mo%$%ktZ04f89kc6cR*=kv1B=WBJTLX%_Z^jo-b5vMldaJgRJ1sKF!?ugM
z%1wV&QtcLrv+vIMas>U+BCR`#K$_><%AXjX&nHUz4RZW5fMIq%-9Zn}&PbxaxM_0I
z%LoV87_FpNth3ZJ+UE>IbMN_euZ@0Kxw~W=bB*84cx|47;Q>dbz}I0qx9ru99J#!x
zemLwHM|~_nJL&=FO);3P`xJo>Fjdk#W~}X-=ZrbJ+;?bv$sec}I0ZLDTHs|{hPuQI
zQ;iD+HM?)o^H(d96NZA&+?SOXo8In%El5#mHvosq{;oahEr3$0Zs?0#8SA+QIJ@$k
zG2`2bu-#}*h%s#!teHv(Y`NTYy-1Zez9bqARfY=m<E6{(RTyigWQT3#qq)!gpp>cB
zS`i9lRC@794XZSFVl>?87`tx3E-I`*E*>gDJzH4Y4rSEFId9jB4R#Cj$h18VJ1w2q
zV@KmG-dxnvG`1F&NW!=|3xC<1->bY4AxeQ`egx(5zJ0{I2hN!d&YK4(rJ!bElQLTE
zXeqNJBY`ZU`$$3?X8_LH^Z|Bm`Im=0*6W}lCQfXVi^O#W>%MZBlyK(u*_L!`OCdus
zNe{6ryJBsvy*})^`4tr`uA@idpuqAzQRh3GA;S&Trd`rj7n*w@Augpr<u4)6J(c4>
zL=X@je;Kz%#n#uZFs<q83ePbuN#^l&w<l|>FMZ#i`MFkns6Y}W-jZ~p9<}TL#(8@?
zV(iabkTCQ6^wR3xJ%g!4+ahWE+=k2vt{kTO;%kE(s-&YAP+#^YQ-;kL%f2z{oe^3m
z%GhSyd}nF`G}#4DHQu)Wo_?sG0%+Bv{bqAxwO0U_aY1LXWG7B<&UnQYkvQw><vB)R
z?k~F-e_FRIuBNehaLLj({(TFkVlrZc1JUlBfgO4zZQvnOa6@!LQJW%8qB;GbU6osa
z*JL%9{^1?>JtdC-CSyPkjeA>}HQ#JPT7n~{q}I&O7$1AFq4+9WA2OaZK;fE6vI9t+
zw^q9y5Xnmt*UXA|IQ!2f-oT%OU8jp|{^Im@Ua@u#i^;l>^Chagai^8wb74Tv&)H9F
zeZvoX^QQd(TydczBaAZp@MUv|EE!~#N#!*gT8(TZQ^Y3S@!*9fpjj2BKMES95%U=q
z7y{*`ka~IG{d((KR#)vtjvdxZ?M7U&_G6gWhJLB*+CLFDIFATIa;|68+J7<en1+~C
z6#-a1k5O5nb@n;>iQvxs?Z{f!Ak9anG0T{eyEtUrA1Bighz=yNQ#lOR2$so+cb!~b
z-0FopGiMB_D5EqoNUDXBNCeKho@EODc(H@!FbiQ8xScL=<r(VccplPVYqol`KOwOO
zV(?zsSBR7&Q|&l%=j2v_W`(h9wWfeXNZm68jP;qDfaP5G8UXn}0L6*N2KJs{+=U?3
z*s(kVWR&=W=30m?kBQb~O1yE&o0cYB*s}QSto6{l6Tu2Q11pZ0&c?T+{wj3xhTNl8
zdFSFY$((P}XZalpO5>m?1;*$qt@njtO`ieFh3v4|>=I&}8~Ee-PHjoRt(agt41NNO
z>2#zGIXnR5UGtqoJ)gibQJ98{P4wxE0Yr+*M?ruUB6D_D0$r(!`WvQ|)WS*9PJn!g
z;2LbIyN2CK@UD4F7m#Q>*>Ut4)?5#<T<{`{09iVEtFYt}a`Og_BgP>>{&T7I-apg-
zSTt0OF1-b5pOqq8I*Y0rFlN!Zpl#~KN<Q1XeNV22_AckxuJ>JkJ2Gm&C(LCw<;+v=
z`_n;a=yvloNSdW?c;;0@>pk3t6Nk$S+ne2gro2EGDMisYCL0*XlC>}$7+q3Vu{)Am
zBrfc9UH+razNZGXYDwky7yLCLe!@_Z5?of#UY_wHc3eWAQtRp6RK7d^jvDsE99dS&
z{Fgygx=KHt#404>b7s@E+PDnlJ|>^(qKG_S#3AP?DruIlJI=PNqbO;Zt@vCOEXzTj
z`*npPX~V}c!n`#!_B`Cyq02#eqVKHh{C+MTHt#ZL62=^p<KM`u{Bm1)PnQIuCpZTG
zt+WwFYu4U>(Yo8vYOuw=5OWE?v#Cxb&#+56-rpmNoVvY%N=~9cZ`t(x-reKEJ3jHn
zC0m8vX#b$vL2BCs9bMDPOq-uOIc|fsaW|*cuN#KM>(&f2i%~DY^HKX1{bH&h`>al2
z>|a56IPJK2UBd)WH#;jWmNp-MvG_S#f=nIeM`O3u+#gi%q~8KXMEFU<+b1{e)KZf)
z+k<^ffN{nNcgc$%>NBZa@FCi(4SUf1*eIR_>!k$UEIxf*f2!ZX*>#w%xmnf9qlz4A
z3dX<@P*O4AUamDiA<|<AR<2ucO8B@TWTr@vj?@zvpaGc5oPGPMljGmv@o?%LDvH~i
z)2TjX2JLM>t?&AVtOo*bpY9#D&f(+u%xS+F?hm387E;y@;>>AoHC{EJ(v9HJjGL!I
zF~pJeWrk5=-UM3A8lp7T>IU$%eF0RWm5NiWBciqC3Uo?)>eMG<l3rNcH#lfnpRQ9a
z86>7%gkGq$DP0%>K42auw*)GkQ>=sZHEY?IRixa+A`@6H+8?0=@}t%8n1pjz01s><
z285YphQn~>Z)C<rqPX~jY!Ua*i9f<t!>pWZ!V9KwAr+KWh9y0x@w8BbUqOY}dkx7}
zdD27Sy*%_svA!CYU!O6<aE|vL3O?e5)SjdosY2x%&dp*qD_4U9!ZnTZK8->4zvaqR
z<gU3qnYcN-i>?*An2-_n8|tDGj7>P*3;%(FH^+Yz8^b8j0>gzkxb~sgzSAS;5qlMP
zaW>Q*E#xdP;hISWm@XUP@w*M|pi>BV{*!w<MnNnU)289e?fho(s`fo3v2!xM;dAT|
zW5#^vwUEnepoC)M+0}z{Wf1a#m?oF)XYQH-ZRvv(kn6@@VB{_o(^HyEm1!%mJP$0Q
z5U;_3WPemr^Hum5XW}V76>pwCb%^87W^N5&*coZ>FWw1j0}{V2ntPe@V6ZL3B>EWU
zcFPN+PoJ4)28=MLvf_zJEL11K_HycDHw+_gQpk^x4x{BNJ(Z@zZS4?2*a%l1v01?y
z$@1x2{^!wbJnzVucjg0sgxMNZBI{pP{Y3OgK}9gjqxoKf%GH;`)_kJ~j2wQJ_l%Ey
z*Y$s_?i;T<PD1weg|A|<jiS1Z9-e|){`g*pyQ2@?`ort;(V4FhsHbiq(w!3H&_zZ^
z<468$sjG=?ZU*NM6*}+SjZT@*w1@xXJ7>;vupPdhG!uS<(()YEE-BghXew3q$&D<(
z>zM>fBXos4N;e3P^^?9<Gu7RBJE@?L9K+<m9H7*x-S)9VbOvd6J_XKh@=lflR=xeV
zg4%FqxhX&MNVd<|OX4*|F2W`dI9jfABX6C`>vP$<qsQ!}d9`pX=}cvIrOvOr$G_LD
zyn|h_OYD;zRxK_{5<ya<#Ki$OEJJ0@O&)rbiF<K-wdV`{+G}W+VN<jddc#T8YXOq~
zaJge^!M9MY4w>sWHMJ|XdE0AOy4QXRzTV`Aq;g(+QM@5xh43Oq!qelL9-ViW7ZMHH
zza*0YRCtZB=I%jSzI64ug~+jGUjN8PW;w@m)~K+l*^rWU?O@)vcX^GY5Q;}OWV=~u
z!InF@2wL5S?f^<A#hBLme>L!Wawp6RrC&FAQ&uKve}nEsBRH>{VIUEq?|XagH}kg?
z0ISeIjCr7_v0yt-Ghzo|@aHUtYKU5IKnE#RJ=AjS>X-Hz;d%@b^;<18<?)<$Mfk>(
zHUpNr;}7C&uOyRheh>8x6idd)hgw)SZj$uPQ}vR)SRlqK13Ya$&EKr#SY-(6wvB65
z2pg8q>s0>T%B8Q(O{F-bTJEhfs%EHiYd<Q}TYAg_?bc<ut&AI@T$AlNVp=v~&^p#7
zwi}!V;?D*&6<8@1GCqVTQ43hUw}OEX`NuyEw7xuehILyKU8OInq!J|TnWiM+baxZh
zq(((aWrq`%)qTy;8e%kCJV@nPe|!5lfboNNEeS`+`Cc2JEgVuLHSsvOc-$t(xcE@T
zaYX76?Scs99vtoB#TLoK4B?ATIeL_>;>F#gw-fAbS9IQ`a5~LQ)d#|8Yx+qJdc<<1
zUJj~Z&JW@V7kbzsPboGS&G(Ln<m}j)ndU-`loKmsN=g)XqkWz7^g+g}WpH}jivA;W
zuCn2f`zQ}3oWiYMZH4=+t?o)HsA<?JJjr&?v}mKcb5WXQGf9rDppI*U#4b@`kJY$^
zfWbuP_jKs5aF0?Pnc$nXg-$}k-Q4QFtGk*l;Z8t^vCI84V|_=Ya8{=Qjkv#i!gXug
z47b$e!B~5DrWQtiQsEh1cefwkeWr1<-1*k|ZhvjAs(LAvcsEP+$QdK#PVe%(J6?O5
z-aFp3PX4ev9>02RI>)MTRy$}RSbFrbchl79t5TzY$^Q8?KR#`|X{kK>6?rM|QaxMW
zocc+=qLAw#N!L#GSa0WtE~c<NAUyDC=4|0b*SE6MgOm0=fhNK6%e~7b7j4OB40vkH
z%7dnRPQ#tPPy5CzFP&O1P9<-nx699%O;z*{TfrhLU)5(-d>DE6yxnN8mWPDrNQPMr
zU51TX*(6aR2e8eJF9iUH6U7)BQJNEdmp1f{AAYwwZg7IRT<kV#rvE*xreQUtK9N*9
zc=)K@?smP5y|R0oYqBJWoIEtP%#H1#e);pUTI~;*9HNG_OSEUD(?CXuUth)&w31+^
z+S{HB{5vzK#H#(jcy~5olYK;TmI?i|DT3WOzGY%>sP?jwm4C0~mG)Yim`##YD5bUa
z58!KVaZXm}HNU&v4?a__!&zxJF6~*pL~mu;iJ3;0-gFcm@V1L(_0(Ljt3RTkbz+6I
zy6t)gV@ho}f7hfv*#hg5Qe?%Tr}HepE8#(P*)cohzAg&w^%tm_R;^4^-3}HSMfBA*
z<Wss-eG!*B#9e_rvso7c{avqX_}SfYrZ6we@DaY?aVUpQ>z4&Ux5-o1jM-z2k)894
zJS)-jBV^IQyJ5>s(Dx48cl<?xW%=CdON!3|$y@Qso@0(qGZ||w<kzK0?V+L!Ys)m3
z{xX!>AH>&{o`txopTC1PM<_gRxhC<{J>OwpGfrKn^^w=BOS^-o@G|I!OO}$mxFiRk
zz{Tszj&CD-2*EYPp9;0rZyb8cGdbLI5sfOTARF`)g*3~}5Wyel2(9>!d}4e0T)3FY
z>a)6hMwP^^&LP!wOErqDK#R??T8)kyNmRuB1{Z-1=qO3OtmLlm!Q*#q2~T%tBXz$E
zv+20u2s<ah$+MTjO7nC?>PQct$bE5^YX}jtiVBb?*1wnnshq>(Efv=vZFr-QxLY>I
zE?J&UJQnVW2LVkAcT-eD?2I_XKJ2%=lfuKkw-v%Z>`$G?lA|MQbm6?3)XsL(7QW&7
zWhgC<L%w19<6+<U=$Q6usmgZx!yexa6xvJ91Wcz>)Ghyz=5u+O+t^5J?0Lb3$}sa2
z|4l_u7QcZu>v~Xqomzp6W0~0Lke;N5^YMUWO(zv#<t7_(bNjn7Iw-=_ZBV42XTB%y
zGpo2sb7ns{k25NG6SG|kgc7Uy;-1*J^5a6fiUJ4h8bc86mdJ)VAmwAq&}TjOt)6=7
zZntbm*7V9Sxqb4*)eZL{Wf$0;E}=X*c%2ay8LQjnz&N2%eoLp~+w@XGZF?wVz1X<k
zKa5hmWp^$uS!^aS2{|#=tliwpBDcWUF`{S6t2D2dR^skiPp*s#aUD7z<7Lt{O(y>@
z$)GUGBDyCPLyQ5pZN12@>na^fNeh4olbG~ltE}|;WS{?d;-7#+ns$p>wJdmcfqBDX
z=&?&A$m-*oT91@S=O}xr&*<mvVZ9D7GkWC)Pii;NN1WuA{AAX)UMsW{C{n0xB!{hS
z428kQNBUtBc=ELSvQZ2ldtGKRwvrqj-yK>ofUR<qKW>6Ug7?q}9x{BI+}x&r+*V-x
zW!pWdnBypr=u_-m>D&9IpveqXDGF4ZfsaR$bp!gCsc~k`ssq2+;COFrvBZ<r6i+;M
z7*wlU=_Em|WE7y-AGO))1u3*a`BPSQRi1eq&5`Z`{1OR&rQY^e@-~=l8@*nTnrPp+
zsHB*EZK?OPl5To&?!!yz$TX&9W=5E(i7^egtq9zEhO|~8>44T&BBK}u+J*<VvZoKC
zje=o*(A-+K@|O2jee}_RoF876f7CIi9Ug~Q90qc_T3QrEdSRm{nA@`0TLHx*5+haZ
zB)8iO_|XPo4QDKN^lwUT*b9xSnYY(KOF&aQ@?)E+hNiJ^)gd!2YYB>4n{6JhX3GRt
zr_!&gV)#pKMP2I+eOw_mzObm)p3m}RgIzHDHoSdk&C}wmo-5lBn<IcpfAX2KxJ|*H
ztY?BGyv-C><)DvLl9ZqN!v)1sy3_<IuVuP6RY0QJ&7jh+%m7*bu1Bs@SYw@`jJ7bE
z-NP6P5pI}+!jT^*HK(ma5lu+DKYz)ewfGWUoC^a*bxX}stc<w*E9?mAnz+DX&k9+*
z^R1v@@xwR9B-^|b9-`W0f(lK)Jq6oH6`)ha=8`(>CByti0$O<%DHA}q!f28FBdq-K
zTIkVos%(wikF##c)t&CXzYyQ+A(nz_yN!lBqen{S>Kr=bSdm+{)Fb59$F6Op@(KlN
z&bzv+<z#y=?UaQ96xgM7AY&_b#JBKk`HV#_oI_F7IjHJjGrp%`r?iW)Awt=vC$i_(
zcUQNbcWI1VBCjsUVdEmY$GEglFY2=%TWSAK?CL$4xZ~Krh~%XZ0o^3WYm(9#xpFRB
z8(xkX>mk@Rz8v1voq^S4r+&Q6sM=6>!+b?|f>ocUko}(6(P+F`tVLh_dOe%Y_~Ghu
zX$Y={)tVdsrN(M@dy?&1pMYR*VAUB-=}5=B9&Y1Al4_dVJcDi9)`eeiUYkBj{<~S#
zi)5qj)u}*MY?ks+2dhPu)!&M=l{}2K9mslv`pL0mg%rB3`)NHrZ7QZ;xXy2@BhPpp
zwo_@l4<r+ecWayiNvbai;_;4R5zO4ZDSH`bm!A$Be-yh3quRe3;d21|wwQJhR_^SX
zuz$&UH@mQTpIA>jl|n^g$>F^@(h@jUl%$E%a>iW{`yJl_<Iy%pU+l=}=Re*2a{_u9
z9vpFFw6&UTLMI6hI^EiOX+5Ld4T4{|IqZ8UT#9DcVD(snb$bQq-4aG7H<6#)XQ2XX
z)JN+KXP+kTEtGB?k)Y#n4aF258C|+2%~j0U#v4g{@2LV&gf`L~l1bKC`Ppsvh+>JU
z+Yc*1Le|Z^gLP7*j4l7ZG@$Mq`_A@LZh4hX4U#T%nCMSh+A3^iB>zsdAeQ7uYWtIX
z>aO;pS!qdhY&8=%z^yDGKQ%HnD9+Y)erUXjRvSH8l>(pJs7P=7=|j7++^SelWB%mw
zCnowvWxCs6i44u&-jsX(Xe;=%G8pL!*)ULSwmk`=_{3>DnoN^m`_tdZX{fx^b3$P*
zz0a~4lQsrqgl*p8=i?qqEmviC+i#l+T%`HOWk2UQMpJcAPF8{6FODge&~CHcrI4I#
zwq<uWs!{gNFV0P_x+KV$RraK_I`}hk;-1K(xh-Px&dI~|p5d!HGCrPhBpvObtUv{I
zsL}9xQgPeIHhMGFWUo`6o0mp(dfW8OyS$s?iz=f7rsWcxZKT|lu3kXiebl|@q-&o{
z!j>O9ni4jE5Y}8g^rqn99vv^-JUYfc2&ruqybLJg%oFuo`ilNIP$43g2&U9T4j6-x
z^_WTh!o?^_3OfV^`jJ)z;TqdAhO!L7^#nIbKuE5pRxGyUE!kM+uemm=l^NN6R@>(W
z83|eTQAb{t6C^nWbuudc&W8w*t^TOyf#9d_HbOq9Jsq3#X4)~<!E>>Q>9=c%)Z7s8
zI}Q;DV5DZsw}@oWus6ZA#Jy1iYHy-=zbHz^N%N|^>Q(!xUd?(jGcZ+AJ&`-#3*B39
zc|x}E?#_$dkZ86*NJKqUF)u6&p^WF2Z@Im3oRTj?<;6Ovm~toQn!%h_8@87BV0eT!
zLc?hI!?QxVnqddq8Y@N0e9?e_wP5~TrSDHlaAIvi=b_Shig`jK;+Gmt(*{jmaYin~
z3yyjgnejIo){o4KBvhEyM-6%+rH_TLCrtR>DM#6pi=DRbr~R;hCO0o7c#n$zqHE;O
z==-*f!*$x9?LQMfqeOPIMCJ!@2%0Qt<q3qHynsI{ad(6=kNQpjmMlpMvI#mxcQNCy
zT&OLR5S7(Cxap^9-+qlpi{89d=Doh>lt{hvS_hT=iFJn%%UH`+Li>*z@h;vSRDOq+
z-z0Qf5mw8wh1{5FoEYYgUnc!*-8;u;uyU3u`sOS3(Un#C;j$8T4)M20Ca*V{g3Epx
zECGcF2BDhrfJD{hS;=UznuKyz6V^>$>yhCh+tBvkf`-BaXqkd$=%xVVcNh6C#J0or
ze2Dpk{oCoo1G8l<c+JnloOq={;bNbyf}ZuNR5{&R4_C;JiuJhDcCoL~)s?;$RX|DL
zh@HrtlyGo~Y;v3IrsWOiWx=}Mg(r~-)nv#h^&pKw<~9qw(wM(Ab8~>TURM{v=tbPf
zmLs~)z6Y*>)#JWt+r<X;{+%T<<e~Cq?foD7_cV>+{FnWK1N{;P%m2`kT41m>!nC`^
zmc<a=$?9b$YkHnZ3ERg;8A6S246AGsjs!_1mi!V4+fRuz^%taOe1vc$Rg$=}5?^jj
zR<1LYs^ZL;bFaVd8WmPH{#=KqeOYO%tJ6-#k7tctCR(H^>LxvhlZajY@HqaMWnCrx
zzZptpfT1MQtdT*vawCSv+p2`3I7THY>>)%&2Dvor5WCdvU~RK6TG1q@Hk$q^!yEF<
z$S=fBce&t9tbJA#A!iCb(Ci+`NelOWHQjq?a*pNs4MUj!DrLdS_JtiSH3{pI8EEHQ
zMlxW%CN+Kd8BKG2ePV(f4b|GwJoqg}YS%ozcf16hr`PD^3~oxw^q&|4F|iuP0x~ay
zhjjPOfJ$q;L`bD9kN<;*Z!w7I)fT%N(f!TQqM+nU6*Id{-myq;&?GR)P}H%NJjM3j
z2llgS?T5mVJ)fX)Zp?G^`!>Fs@M>6as-1g#Eib{2G1M{OE@HSvVNs%@T-l_DCIH>n
zeeaJIB#b8hUZB5nRWQP9_CCJR0g!E=JX*f5)MlN?3Ss<E4yE(9V6>%^gr<MBn|N#@
zp^TyxcovF_6A@<T5-T$BhV&<fu6tsWD5(J5D-s@B7FLY6PhP;L!<aPdUwmxyqr%mL
z-wzUk)A_qTjDJb_{`7$)>8$Oa7E&>u*}^sd<?wO#aVt1GBt0SAH?-I`)?#e@PfEB4
z@~ss2B6-vuYr%2=w4{WH*EeAgbF2fhf8d+^GBepZI6O9}Sz_Y6C9Hu1$VV87WihbS
zPfk0}DEdg<&T#q^>XqLP=-p;kowV_HuMppk$!1DV|0dCo8`!oRS67g5CQfkpelZB&
z?tAWHx?8zodh|V#E!mZ?lf-CZ<k0x$Weo@8^)A|Ihp;Xq>>^aqVJ8xEwYijq({4^^
zk`&O6GGcANa^&}552<)IP2}I7c_7WAxBaMAuZpJ2tFY>5=+?H};kc>Ys34OCN5w@K
zWEuH1$`rm9Qxlo0UntAbOl^ah;`cIGt-5)~8OxU+kB;xv=zZffxZZOsurUi6w-hVg
z^=UD3c03%&?7<zI)}pMxlb=$zA9L25A`KT#HeO=SR~{$5&&;i(T<Xu6jzD)C&7KhZ
zm&6!1*bFzh{q~7I!To;2%~Mc4`u?beqMqyg_$(`#s(SCf)yzmmjDFt1Dxinx3ngEB
zb3P#XGR0pS&zYA>w{ek=niJ%Qpg|pKE}Y*wQN;-F+Qg7(x4i6m`7s`sOSx~m>aMU^
zax<y@C8TWX0kj7Bo%BjjFV9YKnYK9+N~#_nZvq$N*HhVns}6M>&(M)!v5t8dxo7K`
zhFhJc`42wgC1ITxC^1lSHdfzMdBKZ!D)>+E>)%Yp*%`JEKG$y~o9Lr{xnncA2j%)r
zS`KU8OBNjvpONN|EvE|)fi5r9l$FY_58AGU@bT`@v6PK%#0(YFGS`N@iV^Y-k*;>3
zxb)O~&}Jv|_8UsK{aC^H>Za%=gG^SlCcz$2;<es<bUE%GrxjE%wY5lvZdl-LyytOx
z&<^jHW7!#Wlt98aR?XjA3O)1WVbyb=mMpM|JJ8DpUP(5z7R+xm-{-~|8YGNm{c+M}
za;tVoWFj8;sAH<E@4c;dp2+PhUL3_sDD<gUTXs4xTJXN<p8?t<387mDGl<ljPi^4i
z`*-J03R(9<4lz(#$HTGSFONr#sj4LJ6L6!MerLfG<BRYwqNcLfhFzkYr^(4&SA{UN
z3_1{OP~C_>hZEozhPD|UGNRE%a8LeFK1A|-T$5uk6#Bm2a4J(RP+;NaDU~(>oa(~!
zn5Mf!MhClXnW3QaC&zK;4^99q?|5_h2&(Xaax!|^aGC7pVMKp=ShQqn?a%zaMw#`{
zkcCz$oO=UR_G5_H`HZk96o_m0J;h%P<v&)i7<Pk7FJ_Ttt(mt*@+`rNmT#aX!ah`h
z8zUzCrx6{?dQ%>by5cFkNi+Fk7#r||s-QL&Q4Ku>f4T{k=bE!3%AZ)aoMo3L1=s}f
zg;#$omwzJ3?=5O!{WDd}S5)vdiP5rw*nVQ(tqo(9N?8JjVd5!m+vnL|Tkj@oy{IUP
z>{4rw#&`y~UYmX(J`gCOs`}ndE5||NbNbLxLsf}t!1l{etjigB*j`h*_2f%0TrnoI
zJa5&e_iV;PJbjO^8R98-e%wH#wL(3#J>2!splmB@@(v>=jD}+$?bsRvr^WaD@Q56w
zH9TzTjSb1l0qdEhbZ`EuEYR@%>N@sdNE+oglpE1W{Rp->Thoedmgbs&;-0l?S4TJd
zih$2{$b#>hd>P)-4b-L>Uvffwz?mpqH<o8Se=t#&rVD>lHSj#UkE{y1A^+UWSCb+4
z=*5&9X3Jt<K)KC;RSFOdpY^IQqS&FP>fMJj2*TkztTMX23zBM5SJ4)Hd;Ow6sa|t;
zJ*{x;9ZM6pX*r+5<r`o>krU*~S8Z3q6GT3I;9G!~$ao|CyqOKJ*Iwx6ej8Rl`jvNb
z$@SSrSf*O|irj>eSDF+*hfnv@ZQV`z*-SG%J8sJ*)K=6PViU1PxFYyHF`SNU$AXoP
z)w$r3qVg<|aBhhrYdnbgc2Mu}X+iwLND?~*@Sag&MM?yJ^BKg@9PeQ35L*lK5JaQd
zXk@p4O-s1)$e%DTLtcO_EA*FfyW2fhlm?u*evA`;DLxPA;{^hGoR^$d7$|67al8@y
zpIo762QGX|)<T%HbFD=`&Id@I;hn3Pg+`ju7-zkd!XNY|NyA`9Yk~S{{kiX|U!`~8
zi+#&!Nrzr~{#fLVWOFOkf=)mAAC$JBR;C^Mc7~3@v+F$pho^Oi!Dngf#e`_?;_yc@
zoteesO}<T=w$V)WV)ZFBKKGe*#~8Hfh?mviPw|$7Ve{!0udB75c2F~5CD^XTv+D*H
zr4bzKM!bB5_qddG@Ygh9bZvo@(03+JmkSWAbYN_rDc!sFO?XfO^<@HwNcO-y<53R)
zQ;F}as@>KWoa=yu3Y;Bc7k=WyJgM)TZxlG{QM0*fITN;KnD!>tr_kB@;h|Aj<h?M-
zz!Fpu;;gImN?&`|tcR#xOXza{i6WIY<5_ojU##7%>samC?iJcngU4v$M1Irws{rQy
zmlOxbKxx93;bNgZT~3&T123VeTcAsXqbrXi9>^GVxvm+B?Xwc>#*Z2l#m4KFE9byO
zyB%+XFbR8x<N|FZ^f`K8dU8OX_9kfiou0;P<hS{;EgjEY=iJI&i3zBapZ!p)+c#;~
zqMuSPo_ZQzWkS9hYgXW6^ZP89E}&x0x!tGqASLmx&6ulBA7^k%<4h8VO~c_p;b@i2
ztg4Jg5N+yMRp-0+R)%;U5S*9~VG#h|=(4ZO-LG4b$hABIQ<Xsxfs<KcDGzkW&kmye
zymgpNEEA3AMMJxRh?VMh&FVMKm_{$4#fY_(z^=Vf*`X6XDbG!;HW?!%hWh~KjlZlg
z`^X@<pRzp#$F;+Ca@!eYLd>ZmF}Um*d8=oW!sKLGX|}Iah+eP<2hO?X0`B>D{1bB3
zINtgl+);y}77rdP*_w8Ib}PY`<ep@6&)m4IjtL(X`S1mio6w;(7;p?o?uRqYizfR!
z@eHEt+LiKE4^&*&vAW)pn)nvYK{PI3r#@jmxfD8<Ue_SNziwxNKXfZ&*{nm?Pcr-S
ztEGcK6~M0c66KPiwYAC0l9xKm8su}2@oB<=PUBCW^mj6o4r^T`CPtGQhKdoolgU3D
z4Dq#fWJoW$ew_u>h6G|PE$vx6?F+Sw<2@9w14@8$=?y(^gS^O4$aKiq!_A(U@tCbT
z@(G4UE|b79O>9{zUMTYD`1S3j);_L}Xz%~kMc|V9zG298B02h5iQupa%@Y?j;#uq^
zA~-knLWqIfzhrslUu$@jC(T=HI8%u~yIiaY)oZokw`!lunS05iS!Wz7jFT`L<YJrV
z?@>@oxNFX0CI`6e;(j5$kiL0!MW+`1Ba^PBXA)jJ26Bs!yQ@rVtz;zc$VO_c>pE>F
zoz{KouvZd6|33YhCE<l=q7P$=tAy)3w+o7yxs)sZ7P1ocM=9cFamsf5LsH#)i7(o1
ztJyaOm?x;B1ScWp#>-k=h+2hAxfxeT(?J67V?*hW@7&gAtrgvMj~QF4N8I@HoT{QS
z_uRXG23`s=u?Ls<B}WWdld*NXIWuHzw>PjCfQ5SK!~mum0PimOXQ7ttNSC5`1OR(2
z@~5>@_5M92QsgB%L1*Z^#qJ}LWgqS$Ovoa8ri8k0l_U@ppT&l~5b>$%4r7?4UIM|)
z6GFg}LAkEbqSVfmRIXCvGF6v*?TR;rTxOL`Jal3-dL|XlUafh(4Mc|qE@=ba<c@uj
ztUBa-c!{wtY7j6WIUnMRd4tl{Xqe1t<uJwBbIstC%ucwPFR{}-Zle_DI1F~3iObPu
zF@`FzNlH`9D9btHHwyfLG-`>DJy?*Fz6aO&BAg$b@bSDq-CMk02}tI-<taQH>EZ0?
z!;3ERLf=Ks2s_vzuQh{GOI2Nv0qk|F;TJ!!6d>?3gp3n^A3V@rrS>GnVZM69<ib|q
zAC6)D4S}tebyk`C$G>Tv|DMu{D`L-?0phFDdy)vHe-1uB^VWT5Erz;nGcxbP3)+8q
z;1pv%G}$V#N2~e4<`AerYB%R{HZB_O1`aoQnqsR*w!n2z4TO~!^G)p#_Exr<Kk1}g
zJRuC}MIRFFj3t|LNP*gHGNukdvDWy-mRp`6Y^a3DX6}3FA>09C`Y+a!&0J4QldDX|
zB?F-Y=bAG_NcRoi0>7wt_c#7=3i5b<Pszd)lQfTByzVC^={xt0atuEaL*zVx7A$XL
zl6JUT-eJJCrj0c3zkC$@-lTx$_|h@L@gT4E5WWf|w#}p&e*VF!xqg6jze7%Br~A5F
zt3h63w+%1g+DI6jd3kbi{XdRs`*)bvmtzfFZZ}&Rx%NfBX;c<EM9Q(GNAO7?u!Vm7
z`yJxo>@=3NY_|at@WXGl31G*uT0Z{=BXp+&sF&kctZa{&ZO8!gCn99KS%*xOU*9Lz
z=~C^+HPI`?(N$3XYB&dRjvD+S6MXXbKKJ8ePCIVpn?BQPI3Xl_D<^zojc_=LH*>q1
z(rdxxehGB0U4@^f(E$gmvVg|(gQXex8)Ft9vD>1^2~T{X_43GB@MwAfZhX3+uS?_a
zeeiI7O|7SLR@_c$wIn?(j{rdl|IEtr7ull$Pp&UhL8fZzs8}B35Mf5AuH~ry*Fh?U
zzynwv4B~#C21qG|Jvp{N4^11L0pmWmx`0Z9@O)Z^J(-ru%|%on;bK7;qt}R|_m)GU
ztM6<Don(5#X|JxQv`<Lo`*Re*YnlLbTv8obfrHl^EjS#XcDG)Gd%}sFCK<vdNp9ga
z<>SLm_mKw&VmD>NUAkMnRXun4U*paY85&0Aay-yRwrHMw%MpC312o)U0+n0ut9-=h
zdraVoFngF;E=0H>EY37~p!7H&_2`ePm<g5w-Qw66nF%G9a<qs;!QcA`k?wNlot0>H
z$4;>4CZM;Ooyko4v=01SzXbAE{qp;!0qX4$HB7D_LTv6M2C}>s9UbD6{fd|Z*ZVD|
zTcJP+sJ9m2CO8Mkxr}2b^x{pr5fZ*pW`zzZ$r?qgyipK0OYji~OdM8}ZHDVk(Jw%`
zd=*%Qz`Z;cje**U6`FJ`{`yZac-4@Bh&+4u>oYX61y5|2+M-y%Xl#+@BmUQj$MOIg
ziLD4pe$x*UD_Z94ch^VpO`hoJiq`aq3tGWo#l)%dygmpvX%FN4TH*MZ&;PHvSu1r)
zxhanSq1Zh~e5BtTPD1%l?VN;g9og>z541Ozmtf`qM>IlYMO6!L-ouV3Gq#UUehRdc
zTg3pW&N!g5Up4!rn)zRoA94Nv$sYj$7(o!R%zqBvq1gvW#CS<chxF_<Y{nvYW!H<9
zc%jVyV+*xyR|8uT*_{_z3IEAS{O^vb3q&|-DSvxW`5#@!|MM##U?|X2{eSTq1W)7t
zcXor-;f+fn!lpl-vH>u;j2+thRInQUhn{}x9}(+y=8unI|Lxzl{suYTtiI|37&jrn
z;RE4K>(yq$V5@uebjqyI%LsTTv4b5$mz4_H|J*&#|2`N?Y^ILV|I5$rD~1e5&&yS?
z_0AejD_2l8){DK=9>TrU7Q_K8RrL6UKaTGqfGYr1&9(}LFdI-^$#FoKRHTS{VqIkt
z|L3ky^?%JR@BOcv;`e%9W$rTS0(emM09h@f&i2t-UN9|tT+U|ndx75msQUW;N5bba
ztcZ1}Y9k{aM|0&2oXNn#`*o}Tbky~q0cXjgy<gc{PycJi^^4u>6~1#mx*M;-H>dTl
z3t4)CjD_LnM=dVi>W}(qn?k-=@smj5$>-vQ6Z&zdKeI&)r{7Sqzo10qj=wJ&GEOm<
zN5W3o<!`9kYwgjq`jKGV+{ruVKq*x+!G1I$IFUGU*UP9>)JP7b_jP-kKqOmq9JvDi
zBwf`ruefbNoR1a57mR>3zFLT(Lb^vF0{Nd_Iv$HazC40Hf8q{7#-Ii-k&pWeHG~n~
zRbr4YzqZ*9zyIrvzwiEUU!Fbz-@zW7hx+%2JW4PP>?huYKaYL!F(byT;bycup1Y5+
z?s(8tYJ{nLyg9FCw_5M}vnZ*qkX5Ix{E81lwzCfU9}Cq5xlJ~Qv-Kv+44-2l4;85K
z4`qL(Ed>(8@o2cD>8oS$@@W#WRaR@&hPinq{u`tD@v#C9B^T>yA%UX@H)nJ4u>8+e
zHpb->dU`<rSXM>oT63}W`|44ju-v=PM)|)3Z(kUy0Z>01hh8@;f|Y^T4ehzyDol>!
zzFw`EbLB9rJ@ENEVzwtrdw{O1;pcK=LjV}(RP+H<UoHWsZKsp*%tl^^uT#K%r?gbn
z;tx@e?Uc!yv+iu2(^=pO=MB(%m5Z+jz=gOOySbT*441H#7Am)E&HC#@=_9}zjN2H>
ztg&SLT-FgOcJ)&-gfc_9RF6cuFOIW|cOzh{pr&`H1;h*e?o7;h)9P%)sdAHTJ6Ej-
zQAcZic9ir1r|saR_4({}w@-ISjq)_tk@}5}{Qr-i5ymw@QmrEJkWF73ZZC5UC(E+s
z8)K0N9#)nEHdHkY3`~o=vg<%f!N~-W3+XL1d#?idOg1qe95$eDYlWnPb?;MS3Npm)
z?#8a7p2BW=0C>0DNueGfZo%rfb_geMk%^TLRqRU=>{*RBGX<zgO74WM57JC4-}cJz
zTpBJnbIDdT3+qn(alH@%oQU|obeNU$%0zn04aE6wu;%T;Gv|d-!(N2$1MN(6B$tI(
zF5vqmkj$`*Ki$T?G+eBQQOCn4ej6!T0R%{V5Z5GKt<t0uiMrd($?AxrhRs%}_YY)D
z-RdSZvPv7nOE655%kzZj&JO^(!|G_g^c&{y9bkep)bTVqUDFd}F7m|P6It?4h`Y0I
zz@!p}y37K^9%4c2#$w0|E$3^HQ#eJ$cvOAUlFsLbN9Xt@J2N#EcQFTXUMt<P96ey9
ztO6!jTM${zHA7>u!_lhg-B(H<NbkHp@;`Ize>;m1lMjRXy$KRBB==qWw||a_&=6^|
z$xN{H-!haF%>8XR)Q0-`h0;4wRRj=89x@Ha_MYS75$~ryODs!(%sW$SoC4MztKY_k
z3Fo<{jF%kaKkg?vMv?)<fdjCgdRM4kr|O=Fy!B9r%lX{=>;{0azL0VmuYarF94p*I
z?t-b=C=k16&SDTf%}U`y?Ri0M45Go|gj8LQZv*i1YjzR3SPt(IMW@gy3kH3{>e_8E
zarAyg!Yl<B1@{|KGR)@cSG@>jrI-qx@yL1lBdvg$3fJa^Irw$EDPTuK)dRBN9sqw`
zbwsWOc3N6a1fS+pk?%(;qE2L22%mh(ntezlldX<Q<*<DJET{HG2Xox;)&hXR>pWb}
zI0lA#mF#pKACSy1p00dc#hU$8q#f{xNfq+b(d3i+0sv|10)Psm6=3LQ=F8Tx&||#6
zdWV##;c^mLwcm_Aw0b}_?70ZMhSr5M>Yi(6Hx7fWcM6q#7yM&Z`Y6aoPw;-_{g+G`
z1tA9pf+Cx(QNxRQlz;o&+(j0U??E1!wmtag!7;*+P+eb0$+YD9gn!FVaz1=x)rf>%
zIeT*oaO(SGyEowHj1V<PT4Nm`7L1{e8uHAUe}K&GRLq%zC&<NJwtZic{Qm7E>JNmP
z^R5DtD^?QrnQi7gH!2x@kzC6<y&P6vSJZH6;<rB!r?VDJ@rAF=xpne=s9_<FSdWGU
zz|yMzm1whPKO^4MQV@j_c#&neT%Bg`9ZLwJ(|?f%FGX^KL_SAZ4rSM%n%i^C*cvNL
z#(p$xo%~QLTSV$aHtV__6>{ltlRKI%^5+#Ygb11m>BXbq)4{396X89X&io>dYNjU*
zzeD(?y>FXX?c@@~>;f7Rt$%H#8r_k0W3ce=^5zG?64I{fxbbTha7j6^9zBJyPR)Md
z`?t6CeGCN2-P?hU`=28@--?Kh(~ry4jb>*Gp@zR;eeYoa$_rHqbr(pA&$)t}@g^q=
zb2c|GD#6z}kY-57iwkzX7?%+{`RX@UnXbNJT7RK8)ZpfpW|G6*Wc;Sdh2GOA`({n}
z&TzO?Q{j!MM<Sn%rRX;VJWUxmHfrmSId4YFUO+j&M#Rtc(Q!wLdAQI^THC7$C&Mj(
z(_3X~7#QB_!n|8XHpq|`hK~f9r>Q?qGrv8O<J)*TR+{E>uQw`(3=-q@%aLH%ybG-F
z1zR_)q%oUEdDM-F2!VQcAF{4sFaAG9X+#Y7C!~fH@d!sr{5>*d8#&qaGJETd2|VHj
zv(h$Lt?k4Pxr|1;PbxE0wF(<<9jz9>)9NkvphIRs#9%IB()Q`UcaoIpUcOg<)oU;#
z^uk<L5ZuQ`gF;7F%DUcIj`owUc3|1nqlkkYy*M{b3*utOS~Z#~&ws8E3xJR)#I>18
zcMm(16fQxTpw?$qZ^tKGh%=_FPhmBppjbr{d2?Jj1_FPTW@Kc$YBnE|J1Lg~u-ID#
zW7@XhGB15z`XnVw9fMXjVZFYq{ALpk0w$oQA7@}0@>6`?<xpc~YcV1t`+3b4p8=FF
zuxIuEbMAt87YlJ$SXoy8xhr3av4D;c$133TS+)$6o6bz0e94@z1!|csS&B8@*u-%F
z`^T&gNm-o$MBZG+t|0E&^=}yZ-7%UV>j;liL~=jb9N_>-s@%J-QwhQC9k3s_3QE%+
zt^jNZLPu$aphQ_8%O#ca>U7S{ZEQ(ZK@8jI{z+D*$eR;J9`pIe;m;C*P)*xOL)tNp
zc|88@&^Yz&_*W<p&QHZY@-Rq>qo}IkuSnGpA6zmnYf<0LRJj%=BYPcGH9hGOT!>8b
zp$9r8>ty{mOt3X~18&F==k_sJq(O!i`PTX0=VHq&;K*B8_diEAY7xCJv;mXe#zPZG
zc9(4o{t7UhDEUQ35A!g^q8{qc0X9!6y$})9GYYrs1BEC_LCjeY(3#x}ESB1`9Z<+?
zsQ*g~fCzF9f@jSJ)~BwhhtnDE<)HfQPXHG>p~t`j*#|+N8-$+xuDd7|(^#9j1s346
zhs!OKpL_~}e;gw5LzP~|@;dHXKKg6`NkLvTtz1AO?ym%ed3-Phf1GZVa@$LT{zkg<
zJ>Yc6KBi`u2(0K7-#$Q0o^NkB%L-+<fms%PthG$`n79w5;Kz5C`~Pl$c+{d|K)U1o
zN&Y!fsTLXQ9vSNwcZ3pfkGj*&%D2{^yq92n27WuS#6`+$tjbJsgfli=9}Xx~_HA;?
zd3fdKm&eH$dM>0x8Zil<SJG%)9d8ic(w_)nAX@=_!x%T<@r5W}EW}PT>H}*B2pYYQ
zAz4ZLlNDxW2vam-fr{1#(yvEju9otYOJ!<ocM{f~MXUaB@(@^~dkT3k$OW%38Y_@M
zip9DOkS6t4Zg~>@6TR^%W2Rj4?qNaF4M=_;D&FIMhZKZm`6=a#5RyBe&+wJbP{zBN
zhCGH<?0=Wyi=^OT6F`)F?S#Z75o<>lqX)2=Z9<)SVZczZRtM6&8C3s7`^E3PD1U`#
z5bNVuz$^%~31|l(UdIEL!6XhDS2qp`d+RocTZVHC5FqJ4XD3L2P?*#M4M|ySiBDgq
z!sUcEzBPd+4$~2E7O6~WEsOLLdPoY%YGg&F3?|`di&PZ5(36f+T#-*ngl`Xl5x(Mh
z{}XlGVFhR+T^6#dj{sR`d0vInoDVce?pUveS$h%wzPwv`rCN$tzz4T#AwU>dGSQ<z
z5UyxPz}7Jq)Qs?l4+vttR(^^HsD4RybdDGTmUa{>k4~ODOk1>9Ed*MjLN=j;Y;;a1
zew84W^52aUaoqo(s-nNcDys!V^wEVxxq-ifhhADlEHlTK6ZON%_QcRjv)O4P5wSa0
zU}?p?E?al`!eA4e4W}f9O@uN4ln){|BOF{y+bNwY8ygUADzj_lvyMh%X-VG}hvVaY
z;$1*|?gq`SMdb2f6cwHeYMzsnOv=fJP=p#m^@5+<`2zYF|2&vw^FF6_fSB9!Jl^<$
z`gGkLcqTzUOt8Tc!1qew1<rBP`gh4bO((j?L)W>E)Wht+9X1abBWkP0el9auXyC#o
zW`0TKdY+2yf22DBl8X;>TRn)((dQE)w+#gUfJMtkk4UV*kM6UeT3=TMRixTlL8`CI
zhg;6qg{vnUU`1u$ES3vi&<tOK|D@(VxvUtFAd3Z+O1x|$Uo05Vg9gHMBD)~v<^+jI
z(SYk<wb0$Q)fuUJ;aouiXTpAes+byY9H-H0gJlb6mx%8BqGhl!Bwfmy{V0kUYC@;l
z{zm^>7!<YWijVRUHp5&YLYG-vEANwkYdn`tFyOOZ|I&OhY$p^%#@&0oIns48nyc7n
zK($dh?DXK@CKB$AbKq&-nl$nMGhM!xxjTA%3>^Wb*}}KBhpZX1WA!(uD#Yd)A|kj`
zPQWiCZg-}J?4-T|L`y^|C#b+o1vv-(k@$n9_L>Ek-$JNqhH96YnjQ#Aa_-gw<n5nt
zklYjb?QC1tW<gIFQO7dR%Zl24R!1_%`ie-c<jpl-z6jkx5R8vDYCxZ@6}7=I*8Br&
z!Z&0hx<}E&T`7)qpNeKc!nV2ku`%r%Hc{;v09eN%s%c!V@kWD~IZzI>m~g@iecr<Q
zlN@6|rf7i*QO(Q;tMMaiS|GxF^>3m{-4w;7k}=8#lKdl3i~SPX3i`%EWSn`Eo+6z5
zDGEIOIn>XWhK_Y0Csx}}M#feIi9c9U$cR5>?-;q|>c85VLqHMTB~f4Jv^N^O&wytO
z<mgN7P@ULkndDS}^d2I=vxe|CMBVT{N;!L@lylYsJXmMw2&{l-sjaR;VFaI0TQJ3z
zBjzWw5K?-3P@UKvWv{gu;77nid@x5@M}#UrFw>(WxewCN{8#(o|Ftx718dlM1?t{E
z{fw*w<W0kOVn;dXoF^ignkGc)6>$38%A~o6=7<y~13B{{B<qV%vNH%Y%@6HQ!0@Vc
z=Qvr;Y!u9w_AtPTGa5|)uEDr7f)2^CtL_Y^jYl*q5Q!0};M!3Ke}5f;qfQwLTD(Eh
z+)V91gFv}MGDZqD=1_YKW2P(9SLPhnD8|B*eJ&K_tgtpe5+oom!&EKAxykdt4LnqZ
za=+Z2uCh|=v$P!9ufp>Bl#+l<LS&ys{q`g#(`{fDW6pjF7o`1X>3K_s>zUb%Sf}Cd
zcwY3&)tzSB@uM<*b)%nu@8JhF59+aJhwEWQsXrXMhl;v_NI9_z+g280^^+`+5+^I@
zb5{qdBg{@-j)MYt2(;$|Ee*uqDD@@_&jmu+nzPO%&9b$nuH=G2bHY~lQ)tQ$nch;Q
zB&^+l=Q~|UC}>dB9JmfRjhDbvAh5Sh|KEp_UPLHa!$*SreGc^RM3AgMX3E-tdOY}p
zd*rvQhc7eI-D}PJ3>QS|%|T8+d(zoQMf`Ma?VxZ0A|m{*5bN>BLaVC(i@ooRiZa`_
zEf7#azyKm3plv`vQHc@-QIL#8$taSOO3q0^C5WI%&XS?Xp~xa286*@L5s5_-Bo^VV
z&pz7S=iYPsocH66@!ohpdJIRC>f3vNYp*reoO1<FDtUwvQe;SJy!E28o{+3vGIb@5
zQ~Df6IMS^YyW4m+R<KdI#v$LR-A?ogouL}RlL0ra%g>sMo8m|IgYF~m!fj4=$Q-OD
z6JH8Hu2@ZL5L~TJjagRJFgZu_A%&5gS}QH-bG^CHZ?n#;PN{+{Id@jld4(r*4GvPi
z%5)r8JQ~WsOQn#$6ED)THa}P<*sM$tx+N!xSMhw_zlQdb?nz8=Qk(;b$@+OSXV>qC
z?te%&JTVs`@6sN@zh*%0mAJNp56Dm#iGI^W>7B;u>HO`wBKPtI9S>CX7jpG~6r0ot
zs#Z70%JKK$&M6NOIe-nUk6!)9&%U{gjN%GIMi#&YO*2_K6~pPv=#)x-RJ9fj6Ynbx
z{Vo-y<^|DZuX~ZPKNOpTny&lOV20Y|Vt=0up^4W@EpVA>Pw;OX{o892S(RYwQ>ZKI
z^1VJka43-gDhK?0aVt#eH9dYw<Lj~-V0-K7sOn?#f}P+Hme}M12?R1f7(s|SoiC-h
zOU(}!^I~<JvrJ&j{Ttk8>3?g*ZmYtIJrGi-|3`JGs8{1fYzUYiyTj#(8suJ@2-Rfg
z=Nn%sFzIT6?299r*R=bi&CI8B`uxZKc2j@sWC2_op}lVle*vWbGD7|N3kpYJKcwF`
zN&Rhx`j_>+kOk%LyH$hvKmV@34UGq0--LAt&dF2!w^r@fF_vP36K3M7F7Wra=b!P#
zb4Q@$I-8kE^_P$S>$m=kW%c|-3XwxYw>R+5zw2+GQ$&*J0P5mFHnzVnkpKLL<NlxC
zHickRHIjL4u1vk<nacbTa*zn}cYE;TTRiE`ca%KRN2E}1&Z(U{h|&V*idMMxniWwz
zUjNzV$R4>1Ey7m7h1EmnUY|uRfx_<U%492>G(q2i|Ljz8--?4Q#}FVRgGM<?LKOG5
z)nOt778Jeczx`y$<HaQ(xS0FlsA0H4D2ucm6pwd9qw<gZn@jVLC7jKHx6syMO3CIw
z`8m%f2!c6}*BSVJR9XMlUcevxrz+#$+d~2(RyBC#n}+Sbp8Us${<nVcpDMwB`!i}m
zG-I<LeP#V0EeJnyCo+tHC;NZ&86QK*o;l<y(e@w3&i~V!W^fR-4VXtO6rJf0venQr
z&_OYIv+}_;ZPj!~<=xgQF7mu=gfh2u!f0!>0G3EJ$Tpfar77}edQ%bLTglMNk~H_R
z{ek*l6!%Pb!;R<l#EI_f0-ur+bQ6!oNVy|YLQLWpt9Y+(FL`(JnE(@k(=0RB#{`?L
z-M^e$!c(AEL%0IpKvS@3*cuUjF7$3foIzNr;XxOGzhOM&bvvx*FvgaXLBK9ko%Z5y
zywJ6>J|f+4Sz{Qtw^8`U(9~ii<=qketUu!L#9;@h=Ud>FHdz~Sw#x5Tl%}p|#3=iw
zRO);mQ+&mi+G;B>i}Osgu((pX`>~*1*I|sf3o-}nWIE5sC80*G#U7HhqJhWn5Y!FI
z*2|7oa_crdMcPevpYFFM<{P$hSXHcs_QUjY|A8^dpMP0I_Q(Zk)Mw};-Gk1FnnAd2
zt|P!JR40kRxk_OL&<N#HOKz`2jDpu*)d#Wie&5qwTO6%w^6Q6!wGZz1^RX%uQBwdu
zUBxH_NDDcyaY51>h<{=7<1bL34QhAf(mKf<d@;X;Lp*@5HIkPXfDKGPuw}GSIu4bJ
z74GZ9_N@T6n<dLdn@+Zb>$`2uFeeJKTxa07xdPL*qWJ&-Y#!)PPR}#cTtonT0P#ZS
zaeXsD!|?BS7!S~|(!bwI;{Yo67Qng9k^#Y?)25zt0s&MQ@I+E3jf5pAdLFvK^rp$M
zj3qRbZf5M|OYh&)T<khp=olONJgD{eJ`i&0zZUrnIPPPM%2=1>=cO<lx)l;$-EP%c
z4ThL4r#rsVx7f8JadzEUGhKR_?CW_@uSG(Z_i-EW?5kiN5lqWz=<m3@IkiHH`i>I@
zCYlT!oSlZ?D!Hg@ox_Q(_F@3lCy}9)8`x(RZtWq!4b61vTwe2JSw!XvwKHQhmq{?+
zTu!YP%CO39M<zusI?vMY^3PD?`Iz+eIn)vYSe_C0zk>AA0cE&nzG)-64sG|yinRf@
zHGIkQs-3mSegtTsWBC#sbNEhB4UAd6pIx~%n@5ZSl%#)QaxwKzVv9d5Cr4g#N~?|1
zwdXd-eC0gZaE^2OKl)##0E%!Va}pqtBrcE0hS3%B+ToC=PVcX4Y9JR_dPerF(@+<m
zT19nw{7q4N(i;O0LS$PIqSvR9d!uf63!3lWVIq43puiYr?mZa)QlhiRj+y@Lt?k64
z&()rj(Cuqief=Ufo1_G=(rzSiUO0U~Xol{sO;#S#g?^)|(MWu^=PRljK%{nRQF#??
z-Rq2PfKVerA?NTyT%HAXs<H5k#$5aLJsZ*TwD&uEkx;*n2z4j;D6P7jcv}RXx+|l5
zyI2=Uki{z=_=dRU+jQk6Cl8d%o2Bz@-uQLdeY~^;mM3LAk5-eDe5+tx$A-uZ!V|;z
zU9$NVLuJ|58VhqP*}J8NnIumhE^X5JaS-<GiF%TG!pz-|nL2ka?f|a2uu#Zo+kIye
z5PM#rhvhj!bv#soHIII39PM;K$s#CAqmTnYs!X!f=Ax{m6!*1$qe)oH$MZ@04T0SJ
zmgB*Uj9BkEQuCfv4pVu>!L;2hAm%NULF^kvT7TJsk8mH2G;*iztfEr!Xwvez&P3lI
zL<?v?hO{S#@BbP6FQkd6L%VqUn}mHXkK^(K#5Ru&NB$nL)BfRWt@aLD#9}aajZxi~
zrdR2r>(=+HY?(P;Ww|;{3kkK(l41<$4T;F)9;A&lh91J7-G?%2i<)8QKnCjg&mTsG
ziM}U7j>m7rHjF)s6@=zCmP)5&xeARt-wy%n!w|xWjnIM+^e?kTKT>YSS;x^ybzifA
z6^VdSg&*_crZ8Kde8t{occrT{Rtas%`h9*7E6;;8ZSLs%F``lfOm??7En{8baAr%8
zaDS+K2PMr_+pJF*TQh7NHSg$`m}C)g24_HlcmTj2{?aDbUmTmUO={P37~44}gRx&+
zAhd6e8GFK;y8P6ax+^!G{({rt6ocl_1zt&^yY=U|Z<<}Z$~_c#lJApHbauOxX}8Lm
zhq5w-uCcv0t(TwtO3#8~J!Zet&5KK}20`~ae}FPlt(YP<Bk;AZ<neZ`!mLX6I;jz+
zI!0nR8iPoT*c0|Lm8NeQJ-3CA_S*X~9Q1^3>?+b#7j>aCX(=vE4LLk~%->l=skVFM
zVJ*2hrAhjJA!341>jQcWDW+$7t_Kj&xOuECD$*g%)iMwMLL;^fd4fSwr|xCpLNjdJ
zv`(J*aL2x$9!be4i3t|~@9Bg}sx~Jh%j~+~El=U^zeoEsfjGrS;Cx!8_kk4!3D0`&
z-rU2do~{~f=<PnTk?OVMC(0PB+?-EGVKGwNAq)}Mc-(nVS2|?}#QmyC_zcaPb>^5c
z45~k>%^U|}n}vM&Ylhn+xbk8c%$;n|#<N9aXZ*z9(c-h5p6~P9k-TY?MV;I^LSFAI
zx`fqdRowv=ekWTTAmcgT$eSjMonGx@4D4!JKr4MW^{r0rbX1!(SJ2dRHnTD4yXVVb
z>f#pCrYAg6O#aH#La?>3y~1f#ieJFbz?EaEA&Z=c0wC3*>NW9*CcpVHfz^uG_NsxA
z%fl{fGYi#gGIV_@UHb&imPC$Hyb9KkkaboXTsM(jIe-1<&4(h~JlMCteINZ&kll`j
zL}H`W*hj6NtlB9hi5KSS^=_Hpw=jBUUAq^)i!`Rp6;@bExb7@E2Nb>|KfkajQl$cy
zyG_*&bJW8dan_<*O?G=KQ{)V6+^5sD2a0KL>B5!2QgCrrCw<M#Mi?Jt3cTloZor`^
z*j(8R`s&y%+)dXfXRNi$)2HWMO>%KEgn05-$}M$1F8g(J1Y!$ofAvAETuX^@*DbBR
z-JO-Dc8HNwfPxF=i22rKOVD^=I!mr~A<kia<>wessT>3G44g%`P@lgH;mzqG23!cJ
z2G}b6n=^XF2rd&29%JX>fM|PoO<KcvpC2rm8uXvvtmhpR5FM}wSkW#{yblvJ0`F}>
zo-citsnob;(N0{P(!79BQoIYc5co|^*Nr>kswpIc2e3}pYAN<b2NWeF!_&5>J)Ys6
zhK{IwvpaN4+&ug9jrOlSH!c?D*1>*Nj)dj6W48KZ`TL2@U*glutaXdKy}2=M>|81-
z6uP$=(mJ*;25z#IR<ff9imsFLk&v=DTVxO3(cRNU=~<Y1Ki_4TwwJ0g*lN`UuQsnV
zr4(@^NmrOf>c`o3M8mR6mRGhlzht!U$i4e3sV=h)fyv-6aoWyJNXyb!Ft9CV#X5NE
zj)e@RJ-*sy@R;7fzI;Al+y4AmBJT+eC{YE&0%d-M;7)Q%At=TwbqQUBpZpB?O|!bp
zAa#M}T7n;Y(_P<QgR<PV(m=&_g*H?*1m^3to~UZY#e2Okkt{;@bNNbJ^F{^Maj`>;
zvh|}(5i@Jgg(AR8{5SogAJ!1lJows5acovqz+#}Fn&LEls)PvfDPbTd)LCZV55<a+
z?ea5kVT?Ny=(H&WUfbyT2FguTWfP;0EAB#m!e0V&wDt)-&-$!6ER$cG(>=WtX*D^Q
z!?tiDr*?;g)rgDwC@R&U!Jqb6-@QideI_>A+%wjYcXBs|m;aIMd}WT?N}TQPbb6Pa
zrLuS7@Ad%gYH?+(${W~V$pKI7;5hqMwK*&<R*APC0}$G^zTD5u%c<3GvfVw94Ap(W
z*zUl1p8=9`c)fV_-LczhXf&mFa|5Vj6sAtj=yqQ8Kh|9VPjp-py0nCFjY-Vi8U!Y~
z*1PXe9S^q}lXIrVlWy1D+fWfY#J+G7j`qktbA)1->Pq;O2qnBeWSu+p5n3U6XKNQ(
zt1r8-_c}A7QiZ<3Jxl^LCygb@->_*uK~fd*S6nuh;-Kwkf-hhvMyaN3lV<h=JT(`~
zRi3>Z+>}@aIrs=)(SfHVDASEaK<9`6AbY8;D;sfF6OJ#(`O81g2@*ZZAv+OG6c6|`
zYg|cD37FUM?Ks_d+l|@oSIcbuUm`OUak!%y0r)ecXX5z{&{;?VUx+3@ziod7Twz6o
zhG&X(660L~pdUkXju^p$RrEy!17oeZZzZhh1Pv<H1?Q>kUP)HP_~^qCVrsUTmDJG<
zxOc_}N?#X4<w0y!xtYB8LhWaCkIN*KHa1&JNHU*+#GR{1jS681&<<Y8n*pkn*Ws?B
zsxcpWx<Srox#V;vT|2&DY<dDu^%OV&VlD9GCUad(yBXaX<S4GSnUwCS65h4@{e7q#
z`j^{=t_wWK$>lv0Fh5h@mt7D*e#Kw&qD4e#?ywBfnTZ5eL}Opf{s4xy7W49N_4Rgj
z7UtLilNFpd17Tm*^>M$w*f>F?h14e^q3<ifh2%eoBr~^SyR^LSl2I9o7ZTaq_@*_K
zaW-U~1tu;(%H&jYPo!zmcdFX|Mjs=pRcL)Egl7l(3S5w>NJb`8W@#4l@SJGx6RGxJ
z+<tPS?bceOET?|`A!{H$MZ!eIPW1Z<OW-EflkAkO7roW(g0ow&xEjwtr9YS-kt?S6
zAXEC7Ic2T*U}0@d$x41(7Zgy50uqBGM_Zts^fp)P^hE33YYjvKR8Q2w+f;ulPyR^+
z)R`%j`8E26%U>JQ+Q)d2vWDkWdtZusE4@|uSoXc|BRm&rk)~f!J<qeJG?LC}9O<PL
z93%X9QCqi<<fhRIFahZY1ro~w9l^l6RhD=7fq9PM06)eSXE@hYm}Y`$yF>e;p9aIO
zIPxWU?M?Q~oI<0eZhdQ_31-3cQ<7v6dU)CH*^xwjl?E?{_q}h_v3FU$vv*uR6{?+{
z!0tlNFH~1>zEC!jr;XR*{<N5F;Ep2WnfT0i6>q|`%OFm>uScxC+tIU%sN2;xsf5Qq
zvsOHbj7C&Smbtnm>z90E-TBjbAl^5b#h$MiPW29S-f>Hli+A#ITSq!xiKB!9Es?om
zX_3jYk;#_~eQ|R5mi<2uBg%8&%tIia(c?e6n0wkpZveDS+v8~c?gnn@Y45AoZ;hLZ
z+oEl>uSdL`o|4Ky<6>@|#_LDh_FhdN{qW9ZP)<tQNUg$m%V~ipy2Lfrgu9|LTkyQP
zeYW(fZQ<~I$#`H|5#R8!3dlWfMsRQ9(VevBs5n}_#}8~C7hYFO5o)%QnKwpg1bbm?
zru$IZ0!9uA3rran3iH2pc0;q>de|0wi`i%=s@svvavumWifKWofS0n+fo)bBGM5qi
zSrtgJ=G@(fLWo}ZNLuAl@{onP!^rs(2b8;q<n0!1YwaQ}iXOdrucqTzH~g7$^x8U^
zGBqmIRuHa5ab65Ux$Gtvc0Z=$c_bqnlRg7YNi(k8VIF)>a)n&1@GvAs4E1Z@r<(N2
z5_m8`ik`uAq9Z$@KrkQR*b2IYQs8|_J(L<bo&K?J<jzp5dBv<xrPq0?kDY)30EM@r
zU={cp^)<-i$Mup@9Yq1JfH?w=;?>$Ov3uK~{_oY~Og6h=+jh&Fgq#CfR(4ax#;0Dl
zHJCnBMJhIbx&#;2Nf-0j<J_#iZNr^O;^xs)(3rE-_ocHF)x|uIvo7$Op+1Q`N?BdH
z(#Z7MV^{XCKJS&}g$(Ad<k3<}7}H~lu0fVREL<p`i4hUze$BQdOL^&|8oBx-f<Pvh
zB~MyGhaBg-)r)5?rI=k13OY?synnW&!}cX(Z-;EW@h5MJF}-@E2AtkX33L4hoCl0q
z-gMD-b#g~2+hRMpv3DdGCJ5KROy-MOd>ENlJQLsa&O=yRGp(~<m~Y-o%phg?!ZXvi
z+IEjTEijc%&`l;~t)ly$;E5J~S=1nC-K%c2`7#ZAA6^88Gifb#<(;G9j9ho(lDy6*
zBD^Uez+m!wZSA?JF{2?ocRC&m<GE($eHe6HS5?;W@5WULdP$SIzoqYwqADLQve3+j
z=4TkdCcZSBb_ZtrMm-9`H~wfFC_Ex`GByP|*JE2f?fv3Ry^5Y&Yx8Cc;5i><z*az^
zYs?@?d1F)AI9ebLQwb#<7T$tI=ou0BsISH)ui0O<V0Jp7uzu%@V!Unf5`QY%eVkFX
zB}JBb?zpRN7Y^5C6>K|&QPzDsh|F;w*!MM+N1(7u<&6{r>aL7)XuEwpfrs^E%a3~Q
zj?u-=Bzzh)zv1<mn>yQ`_=vvbvlm$l{5@Qy;$Zj!Jx-YO;IE=F!z0W14fiH{ss*Ko
zW!aif7G|$`!^molq-ZL}zdQiCoVJb?b;mVJ#>nPdbNmeoh3Ov#<z}8w&9JxVB55IB
zBPrHYtqmWB*UFQe>Ht!KC7U?h#D943u=!Fk(SSHtKeHAgef5bx`biULLX8vRzf0+P
zpWyUqU708#@aXI*jsIdE=hIv~nP1^ZROH1d2rO66i(Imsh2%mt&6lg9a?8n$Z#Vl+
z77L*aTN;oO6y@1;Jf~vM0BRE6(KS>UYgk(>&ooO8FEPhtPmTyr5Hq*cMt~1_T3hKf
zu$~Wbab;Tg!?i_TZ#_EWJa{E)Kr44@97$Vw6xefWENKV3`G$B<#6_>pjSHyHXxvQ0
zOvqCI*3SF%5KEGyMUYVkz>xmsgUmbLAS%0-Q^E4UCFSwVK<!%Q$-e5Mf+!OQ)xmsS
zJ@3Jx$529}%PMm#F6=KF>mZknpIBXdCBb;mTqJVPXf0W`-mZXrUV5i%HN@fx*^6Ns
zo(`6dAO>N|F)YIOdUI>Ale-@}h>QZU+2aY4?)zC2=V?3pjVk~j)7EotvcbBz9iWyi
zrc@_-d40D;^H%&svAn{5o+XY6`nIZh?#mHb3FuqWkNh7z$g4j1l!73$hq)Bac%wZ;
zcI|e{ePP~NMFXrIR^frnJHyd1Zr+pJ*lzU%{dXF?eVikc)AZ>oI-Ng|VbfRD2+EO7
zpwf^-nbe>|Ik3N$kUT4ljatL^azWtNV%=zd``lIOs(BY(TUn;;+NYy~)#(L2ufJ?n
zb!=fclt1p~c=FsTjy|hyzY^Dt<y`bNFzfne*yE}_v}!vc&{!1>2V0F^Z-cinkS=uP
z%F?=SZS#!lQ85(FU?`4LclvO(>D04OiP(8#OV4@jUNp~?S@2Y6NkG5eW`WDBFxMYP
zx!A)gjKyr3CQcZo@J3;<OnczzEA|m{N}8GmCikdNUX??xSE4MZS*htz#=iMvb6qrP
z^Zm@3T0u`nJAhghPI^dLm`d=TGwPdenc9Rtv`Rb~_%9bSwwgI_llf3E%vzqxIDO7V
zUwsts-iZ>)YG3mTJ}y%N!JDY4__t!AL1MRvPC0>js%_9bm^l81x)R_`33;SQ7WK#D
zs8pXQpd1S2RHPBFne^#Fpp%cbrxLMrE+;dmG-aO_pNckSoz!~ZoJxwYIta72OWUdj
zPMZzeZwa4t2>Y;k7mFedT_0RVv!V$xHka=CHnnW%)CAYP<~4nbvp&sBRlqLrK|;@2
zwC!WDm)qRKD)s{k0k`c=1o=DKSa6tjrxsjfN$uLdH&a8bt=5-fI4lUv0`j?I>1k%i
ziBWQqJgxJb+Exxqd#29K()058vu3lWpWk15%8^iCHEsK*Xl5VE)wAB8Hslp^D-8|G
z*g3?G(qN+;H+Fa@-~vmPlX0zkirQdp0Y005qF6|3e-U$C!Kz90BLWY}lXyv{GufYS
zZL3AolMZMF<AqE@T-HrK4XTV6&bM;T-<<{GBw4|maE9f2hKw<NwZ&DvV8&hA>u%+S
zXa3->bpJ*`tZfe0X10!z#D|0Cm&J2eDYa?DaSD$PJ?$OKP?BIxZqqj3;mB_3Pgc=z
z)t<Ou<9>E5{K2rSL0n~wdt_kw3rX(y%Is}sNH54%ZHM=uM9|h4^k58ZeY3rTHI5WL
z3Lm$?h-Gh+1T`zTJmS3gj^E<oTS<#&n1)QpWl70ciSeQ|eE*oB9j^D1B~S9NPj`;`
z?i_&!Mx?8?bLr_TcPn2RX|{6qI2+DUitEQy_Y0p>(c9S`!<I0j<`wK{D+pMw)Ql2g
zw`}e)%-7r&YIt%vyXVT``EUu|UJJDiji=I18sEU^@kOg97o6rR6TDIGOrKNVG+sFQ
zF+RpU2Cn@*i|M^R=6^b5pY4dRq^0y3>s7Qx&$`M_q0L#ITfcG&C5vtT?gj>R&3z9P
zr}VS#3hk39PHX4Lsbwk~+3n7-bYLWIKX+%u-^(CeVVaycYZB5f>E`@p^hh-c#mN)X
zh5=5SN(%2{4pY9BUF=Z`leaStj6>M&j#Q`(h!9)o;S7dgmg%01w~-PA9>V=&iHgme
z7O7KJ=4lxGdvrSmab#}AbSvANhN*MgbmS;y+EqPq2Sx#9MUC1xr@LhC=<%~dwPNuT
zYKz@tt23@8MsjaB8}+ZJ>3_h}tY{3wxiS3x+^nTF1%0)f4-%$rc;!`GD_QYG{3Z8J
zc7MDVC*x7l%MH$dB&JPN9o`DF+PXrY@r<LZOA_BB6k=DGMj>*Bsfn&+)#Be!D@w2y
zA!p{(6d%^?T`yK8`^K8Y(HO~|;Z*e0XmSW{qjd@&p{T2-6Ot)_UEA8)&*T!+d*?Yn
zeIs*UmNJ}{A|n+@0evH=r!9T2Q_MZY)UcP$1y6o%w|3glu4tKzl(J<cx6)nDter9t
z`j}NM_KQuYAm>>JsOS4x5<4v`M>>qWr;Tu$nvk|gw9{;wo1!Rzhf%Dn-2ib|uPxwY
z?|LUjFMPi260ch2=F0mdYEP$P?`aj*nH*1F5$0#NIrsKC35xYXt^1U2Plntz2FI_j
zOjeAaRXv12lv#tUO57SL&}Nj+>`5|5Q+=Q5Y9P0!6GpR6OV`Mm8Y?AQX?fo{I+fsJ
z#dy|MxK~H@rRDArt(!{DTdz7d=yHe{I?d)m$nad=Xk?Bp@%Xs3?W8+h$~&5v$<Ism
zWzCn-;{dOv$@ST-Ga3{O`-Iax+tzneHmPo=-gB?5xN53H&X>N5+gM5qiae;142Qz%
z^2yZcW-2%f;XC$YSRgIONKv?>*CUS0?d`5tbxn32Zg$G~kaD9LJ@zzl&>!gRj&sbm
z{u|53S2SA4pPogkmd!b((sQNSql1nccV7Tv<Uk3(L4|ngc)peT@t;|ah~77{lTcbO
zY%M}@B5NRS%TG*m6HTVF*Y)uEW|9kT$xZud_p2}2ugKD@X*8?8JGQNAq^;V%$q3qc
z^~6ZgmUkAnC&0R5($#3YJ-Om4n}%jXKdK_@F~NPpm^!=e0IV*JcUAK}yP2d&j5^;R
zqoRU}TWC!u*JLO*rh2`^C%5h=FEC#ql30=^(Rf2Ktu1j(@Pf$oDBX%!VOJ-I<0HTL
zFXw1dk3bv|t%6M6Nb97@fhq))KRxYJ>@(gIRFPh=@!E>gBSM~cH<3n<ZjC*Hit0pk
z`?_Az3-yLT2F^-EXPTJt$djmZb=vwGN&$xhW@!CZcUe1V1w*$|Vy?f(eInocFQ=lm
zA|^rAsuZahYEcsH&ZN5opd0rFuk$mP4qQpIz9Y;_8O+aw7mCU{f(L!iQa;h_*aFP>
zIKT)!2NC&>^C~0VB#+A_XMMcrsuEZ%!*qzRZvm@&Y1MngEX*I^dhU|tCN?ETATW@1
z<71r;ul#-#_~9r>qa}xBI(CNv80Tyfdi?d{(xev%_p!uJSeN9nAcUfj#aDJLPno*N
zZS}o0yM}JN8mu~P>XypFOBKx1SST-~7ChJ^?7!QZPH;U6(c<_!awT#9N)F)cA9Mj2
zOEi+lLhIBMuk$X7JMY)OGFG@Lqi&VhTUW_^=Z@ufaR_(8-InYw^~=)rS<&vnOJ$)m
zWHuKlTO*$s9ksQa#SPHAr)XL5ySJ<F*mJ+D%svWvP4@h%ZGh>~T??$^saWL~sOqG_
z{B%+jlE-#2KA@n_ghpLWcbd}5z1xu~uctbfd~8O@r_>gEB2WMNd>2wg0uG>*J&{3G
z{(fh)Pa$3iHSeF2d4rm>*%d7!sQqv&X}!~r)in^3x`nsY3MZo1frq`CAyonLr$*Wp
zmfP~w%H;8Js29^tTxWH;X42}XUA9hApuk{lkKS1{gLM09!^x1X6u-eG4A-F?CvMQ5
zid?Q?10sn+8F3ylO5B-FNdya~4aWs;W)kkBK0B?X2=TxYi3e+r?Y+;0EQhXLm##Nc
z1f=9q=)rg8@j1|{OVMVU);N6<m6{^2cK6;>SoqjvoX;ztR)sh{F6zgeqRpyCx=)J3
zKeQ5MS+zy+<u>eqKKR&3RN3id4+V0zWq6EI#xu2?R2sWl0qv4Z_)VUuDSq+y2H6Yv
zh{s~5O+vKr-OlGqhf8*-=Lwb|J#O2G)B%}?i`}1uJ*l?{ol07<E4zSD7X*X+<6=>T
zCZ%fyRPrV(^{yXND42BENBI?x+AiG3y&ZE=Tdf4#j#a4=FSc@iEc|A4^cH5Vev~D9
zm&USd)46T7DuV?=z^O)zwFMd-3#qcn>aN}Vz=ZAL*AP-?)-uxxQ9X9WV1JPOW_XmQ
z=KwD<(o=_GEL_SFBEH0waje>2I)32n&kG4ys@&&M3Kc6iomoHK3ML#D51^w)^t9Qc
z;~-s?TWRD)Eg`lDB^#f%s81e|b%W6PDj5q3g%bKEI-S#k!&OzY=@|(bjT!rqCh286
zSHRausk-0J$NF$|KP{FgLOC5o(s$2+OF>qsI4Y_Om90F3A7YHCHr`d+y-RAlwK*hs
zd)8%+!*Z_~HWd1Vf7(#L6(6rW8;JD{!segR_duOXo7%Yv%{wVJ(bK%g3iNE(ZaH{M
z4#{+EQFTv&t8(iU$=z$$pN3vn8@K+{;1*0iHTGC-%2l>BKBz!4+jTZ8cxV9t^Vmn*
zjBp*fDJ`rbh<cbr)y%I&fscAF?vAy~;w0VL!<MkCCVMf?3#B3iN5-69x&ukyck08B
zt&^;@O*%U}RM&gCb~Ag{6}hUDUDkakW#4l?=NDMDoe!uwM&MCT4$S$Fdm|I$ii0?u
zQPd;x0W0UyeJI=A&I1NglEpvn>LPQYq(Ok+3YPKKnp17j+HSk{NndxXx~DoHQgpX1
ztjy?U9LqVO!@T&j5c0f1Y$}6_;}C|9h(Fm`?F%M%i-4R!qK&bU#I4FTf2BvI-W%P%
zrM9sOpT5)d(<gPKXlj8);7xcPIV1kTjfT;*1AQ@;5tW~{2<puPK9o!`tFUJX$7Ls3
zD^9OO%gMcoB_*kHiX0(!iOe6q0OFN%2<WZfu1T7swp5|`>t(c~E#5OQ`;cvV`&m*G
zZY2n3`)E|5Bd3z|l;Y*Xon}v$jD*auntV!9eneUt9``go|FR-|zEyBZc-*SNU=^rq
zA2?X^jne#l-6Tn(Cq8_<x{ttPhjX{+i;sTW)cJaVR9|5laeKmu01Zo4YUfS&8Jt>t
z3+B^es2JY!H?WVJ{XEzxPtwC39iC1Xw6RZ-nKp6pu8DhjJuBVg_}45-=@$)zZ8w%k
zm$S|oWQ0(E8bMO@sKxXVHB(n@gC=`-<B_w1A*uOB0z9ecNM(kXjHrF<LFs5pJT_H2
zo96_>x|C#5^y~ae#`#7oXfki>dOqG}(r&}APIsPZ2%gc^1(KX6ztkNmj1Zd|+`F5m
zXP=ZD8CH%#bOj3X`EV)6O4K?&!x1cv`7%+ozyFmY#(gE^W<@E~wqGin%S&qtth%;}
z1}hX3lG~L89><vF{IwSWE)UObgS7pPF9H~Vgk5~jDSdF?XXt}4lvS3*Kj-A2ix8po
zIDKg3HvokS@5XhYssRq+T@y@Pi6HEG1NM&-)+?vdR&qz{dQ3jZrI!m+mg>-~lM_$K
z2bf_35m+vLj(iC?9rDyENOqLUkN>UkD*pU)__uBmbI5*5#*-sjZR>x$XaRkES0A?k
z|LZY<htqvTUMlZgB0=Cb`)wlLmyy(l!Wg(sv-aAH$W$_{l4IcT2kaj$4eIlsac7;b
z?LvXkw1ozNd>T${G%y>B2wR$tPAU0PqN}Hv%gKeJj>`Ui8;Y;Ip0`HcxtM=?=ahsg
zpYI6Sua5M*-oiYNeIOtXJBEJqm8V%<S8eC|wq{1B0F+HcC?NZ!coI_Vcn82+I6;b1
z=3Bk80u7s3!_Pw9;t8E{R1e>>ZH_i7MoZKV>gMQ)in~#J-2O$_Xkt-FBt+TP^p@E#
z`um;L?z;pcvRwH~lk}(!0Es?OJde9qlwblyJFTOFSQFqiLr;x{LYR>Z+J@T^v}epJ
zp<0g{{pT4#14E0;FeN1LP?=d`KyE*0B2scxIiR2%ZCClxvs%d{C;t>G*NtuIazj|r
z?PA&3I!7^4F>T|p3#IMAIuk0)&^LxgWI)AkN6>GT1g*Tmn<P9H^}|EoCeUq_OU<*~
z$QH^M?PM@NA(Q;`_B}26(5k6qdEu-@uvx5=Us7>KRSkG>pFd!w>*+RJW={!J`*(35
zZeOs4vW?6VdM%$pudt>*M*S<RjSD@Hs)h$w24d+;g|BfFTC3~kdq6OW?Q_~gBwC0E
zf%>iGv|3-9w&K9oS#qkkodK+Aeps2|q>H4tETcQ|jv-4j32ia+eQ}j?lpg$Yu1(BW
zIJ9)v=2a0!yY8*oLbxS?t0Ifn#@=Qp6=``Bd;rz`@jizTB$~sM!Ite!%qRtV)ukw>
z{8c>fb~@@-K(ML`Yl9<KxhuKu$+pCij4t}W)I0!&kad<M`C*{DVx)C};qk}QxilsH
zgec7d<5#&(Lb*pPP-W8zejJkmThoc0lY_uz8*(8f^vAu+k&hLo#Wflo(9<qxP>B(+
zn>#U`?5g!C<ZOC@ZxvL-az6-ZBJ;;CaG{ohrP&*=rOarhid??_+64)pP}@8bUUM$u
z59pJ8gPl5!#eFG{Ujo09lG6!XErIeF0xm#G^g7~1MS(ol6$nlrSD*>cYc|wMB;*=-
zNd+<)Y!Bq41?AJ%dX80$CArd6gmxLN+RP}yE}MyQb4*5(UUC<PJcbc3$$Nhaca(P&
z<lFbx{ni4qqcfx==j|=d0Q|9L!cvr>)37?2HU3Vg!l?$>NKC11j{z4dbDyozs{QB>
zQ_Aj^JiMn;?&AB4m!c3)NauD3+eNp`i{%~#YTgL}<%H~Wz7z}*DyK&I%XLNJnl!h~
zyD?A(Iqa6;lv7ss7N+UV;%-N|q{I0;_RV(Y3c8oIQF>6r;I7H!wE=wv*OekKZ)Vi#
zJycmC7uv6wOW<L8S@~gf+jV}*&DRf{Lwyu`x5jr_p0zO-GN&;JW{VxaR0fzZ(c`Aj
zrbm$K;GtzM%@%T!T2S)OFYFqs8X*sA6U`)BYbjoN56)1bcJbZ7!o)hu7Hy8RQ_+L)
z`zoFrs}=GncLSO0ppQ6RT(!V!>;cdk7GVW#-KGt<!X8yJD*qF5l!8kl^)4E3S+&P(
zpp|0Kl0v@(h8YLjCr8+(<!kp@fgRUEa%}9qiq+ocyS_2qzYCwb*$Gy!(c6!JMod>+
z)K*9FrNyHawjAN8;A=E>!-9R_DDgh>(P9WclbJz3U9L5;(?6h)CBT2`7WU24Fg>Bw
z=1LRqMAWMmn!F_49gK2L%e%mad?X~bp?|a1cpE~QWl>4wL0~gkno=Y3$v7Op`;OAE
z>zG=qE#74V3m9X+`@9Zh_;|>X(M?C>9#L8w3(v1V)KAbACs5;OXO@ll%3L=hla{{y
zD4@NwD*x=uout55%AS4>i6UH8?Fr~oI)RX(Majso`SNw4U+tz^!p8%ZR^8a8S2t2H
zV(bmS*)jPcx6tpgzWWM+P{F$qyp50p$oRjuM0?!;)o?(c+4800Ar7+&PB<~`BByOX
z+{Yn_m1~T3g(}?16RZyq*i3miU~6ks!a!JP_|Fl3xU>k1wPjMMsP-AWo3Ktj9(aI!
zhf$}#3T=9mvWjt4wYW!B^IfR>%)7VSYT2z9>7J=YJ7;Lb4~)1}CZK_zDrKQ$^+ODJ
zfC<bZslU3x`~jlbE9WAb5TJy2tJn^#9G+B73jO0LPAMGs%-fHOvH2vq@FhFla=0hw
zNM+n0RHsV3iHWTJF~>XWMuFV;$q;+&gcIx8@&(|6GIMuF8}`=S$Mt4)P2LfEZW)YE
z(y`&%zTqH5s^|hia_6FF45M<8jsxSK@YdIN>_dZ~P`|ne6h=mb%~k`nUnZGIo=XnU
zU-pLNpSh#Uj^3i+PC5<!u*1u>&xsa0E3tbr6TZAZ*1Ayal%|*oc(HJjrk+0gR^2e%
zf5p$oUFCA&nDSidZQK9|;yH^78_!2R;T<1g27sl>Ms95iyOiNnnZXL@Exng%7c5+2
z{8|&21ss>ZE@h^z^ZWWXC$57VkCgb~SSRmiZ8Q{^@9=VRbt{#3CdDNz*}in5TAK;e
z4{W?!9GIe}>)KPA@lBUfb^}#CyFgY|0bO)XeP_C5n{<XPsTx0^Pf9)n4xnL6;xcYh
z@!HIJxuL>WPSn>x06b(HLC7%G9?N7r0>tK#Rwm1#iv`;loqz>D@ZRywr#|}wd`U9M
ziC~I>%iKTD+0VS8WK{VVD);KEoi8O5qp>m5VoKF87*Cz<SbR21h<TLazD2$Vcab;H
zENhgO{}5a$NE2CBtg6!V#p(Hh&X^UgXkf;sK)(hCOH^fn2rM&pWA=7kFH(>Hie)Mv
zcW=MvOvBKxu2#Kl`Y**ez`BJjn?(c_Be3Fpluy8DOUQz@%B9$HIFCKUIbrKk_w+vD
zkMiqkJw6jk!Sm{-gG-w)DuDp%>lTuR1t~Gf?&-$qZMrOE?5uPvZuB)yaJD!F3kTwQ
zmeyFO)ccihj6d(-3-3hW-m>*x1*d}pF7mdR@OO;M-67Hi5)t8?MkH+L^2f9HC2O^A
zg;?@r*vCQ>m781l`xh%9#AcIC19yzGu!Q|Vt=Ph>>E8ajpI!hYk;$7$F}pU8aM*V-
zI|;lkwzYR@YTX$3(ogw<YY|IF-2_bQJ(5aueNTJC7KF}rp02w-(M|4ffDDPKO~c!w
z1@>cJp;K;AJxe9in9vnvkYbJWkyS<Acc6DdM^28qtOQ)9`<;|SeesI@f)<U#oB1?1
z;@}Jk9-&*CP-0_r9`KGhjQd-Aw~3lwdmiKL$4B#Ah4&o4?d<E^3Q)h2fB{HzpANvu
z#MZak-M2D5Hq)J2Sc}RxPfD4JL|%Jz|KVrP{q>V~HkEe`Xew1ZyRtCJ&qAfK7lF-U
z*rbRR?Pol-#QPjHq`mK4(Vu_2el{cKd~a??fr(P_C2i@{xA$U6C>f((?az{s-<J_z
z2$l&8VO`r8p7e#vrNJ_&>p6>3IB;~gg$QO1y`3BaG1DtQXvzdp90o5mqu0<rG6BLl
z@s89)C19OU8hH*;kC(N5a-sygje<zCr>}Kup1VDJ{csm!Gjnu9aP#r`M{~@M{u?eT
zQz{KXZq^k1({X(}zmd5ME$q9MI4Ega#cUD0C@kw#_DJZr=>grUsdTbwf-^ffAT!13
zH1;%qeWRvB`<2uw5VJ!2^N5JVO&(08NE>s(f)c2EbronXOw*XM^rZb*U>YoSbBysn
zA7NV>hr(Z~=Uibn&?YpJgZLG%+Af@(ecNQElJ*H;3e8IV8J6~>Sxcar^74xbTkxwe
z)RtCeNJ1Bl8z2hpG<3$!1(%n1M63FP2SirmJ_AN~!kN9pwGQdD+j1r7tWkBO_>$VT
z-Eoc?IQf3i6j~p%p1M`7_XqR48PmphKc0E%-<sH_X1doUSUIma*y|kdY~1@;;?y<U
z<%Ip`RtFW^R#NT1p^>UOUCZ<g_Nx3JxBVrBP_VgJnjyg<IlbMHQee9i)8#3M^@!7R
zY=^Ldn^lnBwV5y9!MWfNmyBou{oGPrtmO+F3+*cZv~ORR9hE~WquDlbqGeu&*&Nsr
zaZ$&mo)Q_pt90eoMsj((Z;Wym`Od?fxZad|m~OF`mUj_}KpI))!%CY(?c-zjCM$Qh
z%xH*CEIef{_jgD8^#hE+3<*?x^NXg0NRM6oQif#<&L=ab<m`{8jdHd}Ot0GbGeY!s
zDciR5;wT(_;y94|x3QmUl3$-<ApTMg)I0Ul>CWt*93`hdK(^Wt^OCFO3Ee{kaqBMf
z6Jo**h*VUvEH%I7+)lU7brR|5dfTsP+Z<`(IyU3vAX3qj)}YU%N7EMTj*h$9&oV8+
z`#}69<2+%f4aNG(6J3KedXZU^yI9dj4c<QAdiW6*-xEM9S`&Qt6<oGyO*CWK`%vJt
zhR0UzzFD(D(-Dq)@xNgr?oIPlpmaE!d-*|%?{TWnRP1PXwD@gpO_n~Etq$))4>EOP
zYAG$IoxZ&%yL^ZrlH@YEss+%?l%#Y}XVS`ThVDTjY*+Jq%<M*$CErZ^tEIxbxeacg
z8g7@gJS|r%?b*m2jhbwYG~!m3_s1||za@Kg8B5)-(gnuM^WEXEW%AOwm%)7{Co^=n
z<V9t6&U?}t=zPAm+m?vl^#7TDoKjdJrurPmh_A&u18*k0Y=M#E!EJGZAKe6tX*!?W
zbf=fItfpk^lrF#tMF&i2N-Wc6$pKGd>s`CpkeNj)Izwx7j#4N{)qL<duS~+iByO%7
zy8<2LwWpQ6NdVY62HP6#1kMn2YFS*7v{afmlbXmzo1M@s+_z=zHQ<`Ke&-Duc%e0g
zp0jR-MMdgaV$u22rw8GN;hOGBdFh<0Fcno|CtY~;rqb1Ub?~GsrO#Hg2zR<W^p<5S
z9bXY4A%^$aH~k%dj4|dKCTii8>Gvm(qa_k~NQJwFbYm|UmIeRPYt)_2dBS6QSxPD*
z8#Bww>9)O*Ihez18!%rv&_`dE=4k#fsuZXg{;ta(4xPAiid);toYJ#?q+E6B{=j#0
zCUb-8zEd~#t&+>s=jy_8<!et2`o-kRlhNWoB<9kf>q)js-HWi-i-)0Hprt=){JN?n
zNpQe2Dxf0GY;MW0Uf9B9F2F_K8x>IDX<;reP*3uib`Ukb+wXBx|FT|y>p(ph_!sq&
zJ=QPFmdkPiB({DemA6RZI;ZQFjlS;=_gXzBmscgABG<^=`(<tbcRh*S3Fno;OZLFz
zI6>d?N<(6Zeo#7rO=8IYcg|GmCSj40y1uV=t3V>psiwYqdi%;wK8dNr^fp%Jhf8xS
zVpU3GW~2Oo=trmTjZN1=x7G14GC(ILZSw^m6J$$+ef4osSVEra*1r7&ii7@@X>yK<
zxMz4l^S&$k`*_DIUqmOn8v5RNggvunC^IMMmXLkP{H0#~+SUYfUF87dsUh0{*S`1D
zvDrd4^BlIb%KEV}+~7a>5PN#<9$7w2i=}n$`Z%po$E<B61eDu)o|BR7!uXbL09)!w
zNHMsR0$&Sq$$jdDOZ0*VXa=&gP3CIJ7ch6Q&ivxay}%4TG1y^bo)exM5VVj#Pu?|w
z<vEEN*LKV&e_K!DzV+J)XR(&Pl<)wHL$+AeqMD9A#;{@T_}$~?<7v5`-j)?(69rx_
z&9UvHM&C6u!}N6xB$rRTGO`WM9-a?sB)*hB-mz90mMfEsA6{cOM#MV)t3J$B%l3&?
zue7R42U50nZaeAriI^l-X)}~PJ<rfw^j2QY{(_DA<R3D~yW526;{!3__c;?Vfv(Sl
z=LHJs2lu!8yDA7;I2BQ*?;WD!`lt;01uF)Y@|_#FAxQ*8Jy3-v2MmBYb|)TBCs`2f
zP<0<?IR;Sb>W~G0A#}-Ivz;GG_C)uHCly7fD_W$Z2gk|+Tt8w98)tUY=5sz2-h0}t
zJ8h-m&~Luj<6|aZ+qi+j%&YUDI{G>_9v6*ok*7z02Z0)|Y5sS~<CylyOKHQJ7H$*Q
zUW%{8;YTr68P+ij<R1#fX78T+)kdE<4l=bk!hQ%oY6kr_kCH#5VyXPs!IPzozL`|(
z0Q2xyrFVcXLu580TAsO~P~(od(0yGc!gi(4&dS0@$q^bS&7}t~fv+LHJEH7PN@NFy
zHhaE5#loMxlA0>jT_cw8s(HJhIyQrTS=Z&;(_DE0jEKbGt*Tt1t`ZOOEKa9R<Jcsw
zHOaV4To(=LZRNwiw`t-yh?h6BDBsEioF#R_^2zSaEhUS0Ez3TQ@muC=Hyb1^c>(ib
zDu8e59vwH4Aw}v$#(TW3RnypUy!S|SGHzdVGnpi2SD}<@LscbDD81C|kan%|8=Lui
z&LVb{$1mu~V~>%VD~hGj4b119SQz-ATl_(rE%lx}CI9q^?!u_RbFHHw1L^jrmRulv
zBY{;ZV%~=}PM*i{D^My-^WSbLBxF<JAts(3kiuUZtfkY4&0gFte6!}49ckX~!fxEL
zcFIIm*VT4ieMMYJc%pr3k#n%{u5F;}rI4>zkGh>Wr?HzZ%tqNHeOj`?p$Lb&zXt79
z%x>G%mIkjz$Ow%cw8;SLDviv;Y^odQ)sB?-29at9ZvMo`35bqYdCZEY$JU=KWXYhQ
zl1s-;pT1l%8RD{20<tP$+azNh(ll#k?JEDtTd`)+fc-WWUY8nEml>9Iq+p)K-3$1>
zZn~a|97@(k{<4MBvn6QOH1g!4ZfuX8edD>eVF;Q~NIN@rSAYs_Mz%sS8+LP@IYh?P
zPYJJEl;o=Mu|mbRI-#iaMliqW9f2fy{<qVD;c)7d@D(Qk+h_bPFjx$ocvRs)`P{>q
z`oJegVxF%5ghfUAiJCyqT=XgD^e<c+$>HK|5O=S91}J&niB;`N9K*mzp(v_LftG|9
zL>_{VwMk9AU7s~}>11?l`_EtX#|7x>S*N0l_4Wm5FPT0-COQzYG);Ej+cjZGSFTdN
zsGjK|Zrhd0>U2P};v`a5nIsNVm$`-hdbvpvH-N*JCjY9>mpT;}v!DAx3z$6&4*))>
zi9M#Lg{E{$&=ztu4Ws4WW;>-IN#IF$|M?%1rk|)l<AI9{TU&sOPC5~|fKS)(GhMni
zZBe7KJ3YZHEM_b>^}c?5H95MsTO5AvQ35@OI*{~^%#*4!t^5KO>JYDi%u~(3n$|xa
z{1&MUP+V>1`mS<683c;KgKTJgvSQQk*>jb@^68rCP7B%m>`gh;v8clS;q8A}tDs^L
z*;KUI#vqW@!=UY4OWvMws@iR53(?*Fw)pspJ_jM4&la@WIe|!3f%ww40G-w-P-yLL
zJ~Wn@^9g!?>Fbr1K)-GJKKH~=4B^Mq2feP2;<J{4mQR4r8;srjY|N6{OSelJ1-A$N
zefGzI8q*Qzv@M@&yju|oC40$dLV(6%3pW46_WZewbqBj&Z*2kWfY_*7LdTgP(atkh
zzaaujqvLAVLfP+rhU0n#+_YY*o2I>!8Lf0hY(2kvagga8A^RoFI0%L>VOK6>{Nf=0
z$R^UyJVd<D5W89zP|Yr|?1EcTBxqg-GfTd_NQO!%+rAp?2_m5-5oywu91iN32zG5u
z7?kRf;M>y9W!{&=A8GJQ5B8Jcd^q(+Tyi&-o3J|Us(uSl4)wapK7)Uu+O6!(bSgqb
z35$E4Yt$ZN2vsEy_~EeD8Qx|7xrhpga-TRP_t*LUeO1e=!I=L6w6rYG%Y^B5$7Sk}
zKu3m#W&@ifd8`Y{IhwC2DJ|UJt)K>P0jFRC#EiZREMLEJexUm!eEcJis_rh>+5$z0
zY=&x93xeQ|u;hCTm>}6G^fZ(vFXi+7d_Ut%K{%hWHm3RWfBbk%P)8stXxKS>v90MD
zw;iR#N+$y4Q*pb6mctSClX(l&Jij}s{CW-rEO9CtyL!fB`PY%=R)Z51QD*g#_UF_8
z-1+}<a{uxLb;J=*s+Q(|{;q%iYByp5V75?s+UTFZ^6RAhw_kXGIK!}!>;BK*_0L~b
zutq!pO7T8XzwX`t_P77@&jqQ#nq+dhc>Y?rzdZWC_#Gmxh;zf_0{!EETJ^vAVG&hW
zll)Ih#Q)g_{GZ-9&tM`HBKQW+>D0;4i2ra&6$m|{y&nee?rXi)@QjO+C_N-nK$zkv
zuBZCTx&HMI|0`b!(_9hY*0e%xqa1oP^epNjFyya-#}oVimR-m#zvG4<o*nt!E(ELu
zZSz6WmPh3g@Xx2>f8Is~B_c3DHZFoxJnuouSppA(M!x^lbmXrFi~s8Jf7W3?ftS?(
zZ>#awBlG{gQ2x3eB60tBc%l5+Ci9;U`M-J@{;xE!38oH%9PFT1HDV#wPo2+0Wa^1>
zXDZZVR{SCR`1xl4mxuPGV+Jlnz~H%w<_^GDtq^Tx6g}c(&<d`m(GiOvq~!o5T-7Ep
zC58*YMRRx#fWEXvM{ndRQ=lb8hGrnX#0GXy<8VU*Ufy+5c8m~Fi*-6K13Qb|<z8s2
zu_xkVI9BUhhEkw@0LB}f7U17J&2cW;aaTC_xPj}xxB`FPmnW2pP}NAz2KF?ikzXDy
zffBo@U(cKTOQn>oGvaFkq{p{gC*2ZH6-L_)G;*1IIxEQ6xlS7D42@wW+|Yf$OD%|V
z?AA%w709B(!CJAbu_@N;D;=-poUjpFGuZzdtU<9doi4Rp3eECK_;*&`Ax@(<F8AHd
z6iDgsvr>|pBXhG_zyx%m37c;uZw-bEDL<J<GeiiG%4s*(ryO*r=+cZ!9uQC?f$d}%
zS$7igx&Y4S!%<$a5<Ud3cIK6vlj>7benW(hv5qUito|rueJM&GE)GMmspNs&I(7P!
zO3F#Z9~JSL9olykIK<rYg5RF>1Mh`(;Pgxj78*H8Lh4L^bI%{Wpo#hn(k)Xtz6mP0
zj0l-QQ9DXcT*nP&v92TFB+vrWj)w8(d(i;>8UmMW0SDBex)!i&EC8=+*Ix2E=$<rY
zVrJ9(k6(7$LOkyg?R}!2_&fMbqfy|ZzTo=33b8O*q~^K{wt8J)RY7wWo2xIuYgTeG
z;?zsTcnIo|_rq5ffYNF6=DB<5CRkCn@=ZmB*bXCZx|;xigRI#lgy*GGK(P1Pj5xi;
z14igC6)bN4xGELs2zGX7?KMRWg<D}PMYPy^Q8oAnCnWl@on~5yT{2Rl3|o~~svX_|
zCx62T9t%2Nfv#n$>5h2sO`wdmBCg1SZ&QWx(^^RRE1>!bU<^*}RvvzwYV;M_HnqlK
zgUFaJL$Cuh1ZiZtvQO~WGWI>!*F>-1{3_%A`@^b7+yn%T$h0#OfikPlx?oxf&8cF<
zXP2Q9W`-b;no6k>)iy7HhP>^FDi9cNeAXbSP7b|8=?q*Pms#siR^5?)UT4~_jUP&+
z4<nsZI#vIIEjuhX4QWw=0S<ezN`~qr>?H?KCjIb3c?QzFRwWl_TP20+FT^bYznlwl
zmvGuEdLs4wAu$hEd(`MxH$cp}*QdgE?WPbX_~l|i_80R!+6cyg*$|v1Le(!{K}?5l
zYl(>V!|U#(@e%_~?<ia;8b&6h&8JOOK{vGPf`~ACnuI3#COwfSE?QNrcMV8nma56r
zr<R~L#rQHHs5W6}Izztc0WG2GxR^zRQ1jxFa_+=NyZf#_xUO&#qVxkjFx%@#YHR+W
zG5+_l>F5zSw0c~{voi~@09W9NoxW$33j@x)Y3aC1bhqt0!ljS9M(&w$&x-n{J(Qox
z2BI}oi1jdi!lNIpy_c2P>H^xmJ#VjlJs~k(gfICfPtf?sM>eRC51uHFxshUW&0yZQ
z1SOs&T@#>bS^BONq_Qm_I-@Off{&^m_6)+$YZZ~2Nz!Xwj@+Y3j5+{;n)Bk3DH99O
z<jkz`+RP5woGq&U+8ELbPC}~R2t4XSy;0Se9ea_Pzf@?RE=1kV<qKz!2p!uWzTfl3
zk<WeEVcIH1{-)pQ0=$aeld_e#8w_5d|Ir&tH73YrVRc%6wOMyofq5TlZ)2P?G2W5;
zVy+XF^1~5CR_8RO)XqIiz%m&0To|<k;hLLw=%F`m*b1!eO;+R~vnJn6)}f&1wKQJ<
zRZ4eGP3KPL>66rFOOPorh!R%|B4;+Q$m*$*65n<KFsN^ZSHwzr5)6e{B5c?BfST;_
z9(rvZ{pr4}tM#c9%7H}|gZ1p~H<A3>^9ynMF;r^2r0wpf(#1L&U7OlcwKpL^y$VO7
zbUh5h(FE4`Ko}4?PpsZA11;!@)%O{r+lWh1mOUt<ka0M6LN#fRQj{eFB>3GCEtT<s
zUc~tH6jyZY*91m$+Yz-{a64@_<GO4;)yB=gm_Aj^XFXXWay!ci4!6BfDa&19O-0Iv
z*;iMiY~-*CC9^Cg=~O&Dd`j4T(31)jc)P{;a-tl_yd}Pr=^AgdnY%N_kx3u$T%Vib
z%;?^T9>**opvVwwsJbC-KA-?)_YyNSj4DTG2uQfULP^ne5}5i6V-&0Gw6b&w8U~FM
ziY>d(Ec1ifCEuJmi7J7?9wjj-YUK-O=BI#@)1__34r7H>4CRHcn%y3Q*xsitVakMC
zR1pp3R}Iy+>t}r{Kif=9cDzJi(Vc0QT+`iXi@ams+1U!shTL5W(1OO~Hs0SA{$9KU
zH*Trz?0P!DImn0#Yj@mM+1sZ+R~P&9Q_4o#`Wzm3Eu&pCN6)XeWmWCl;hDLsVukIj
zT3K3K6={KCS|&CD=x)S>7j)3PL+jA_!)3qe$tZGi>mF6L(bf<@ZknSYY!iKvmtkLC
zcj03##saNRlDsEWC>T;6cklLCy>nYpZZogK-g|j=P2$tU{XgH-$`RE-^qCj8?|62o
z@LUD(bFxP!niJ)SZQPTN2G5#-vD0JL4Aa5RX%)JI=|#)|QmYW)lc<#}Bzr2Z7nTHC
z3{Q34R)LW?YUvm2UMI3M9ph(2yvqz(x*>e`2C#+_wygw+s{&05?=Ox9lLq@3Y3l2m
zpBip{<7;v!H@!|L>{4RphgF1Io(TZjDSbJjCf1-2R)(LZzMLhX5=$I?ILh*NAlEuz
zHoSX{2Fra*annfXz(opAep*!OSOZmcN~9IQHII2m-C$sleXUlkt&$ng*m8NRp}}ss
zHbb%sUS0PKDiO1j!aZo#adAHQc$fwizX)1|Z3LIg34jiD0B%Q~#+Fp!8nFAcDZSI)
zgl-0SowKo4;>FqKa}TyUsl?f=tOAAk3Yb%smEFmpFVN80vsuyLt)f0u3S9WmQYv!_
z2!^Tjgy62hiW2v{hm2OL=%Zj1>?2N4d-o0{K+c`%69RxvnRODK9revbJv!|rvyWpS
zWABXCwPoJ`w7(TB6%4b%3M=?Kd#XnfsOg(z!miXrH&e?bo{B<lXzTXsJ^e+&&lYUy
z`0JZ&$tR?*h`K&I#d@3C-^QQoj!14CK_VgPGxFEJ2{Sz-JaY5q6>tzre5QBv&_RZG
z&r;(&M2``FuQ))We_vXLEBC=})uPnoHKW*yb-uDF=k^M#)+y^n;j-2k7}lQOYt?(w
zwIHDVI=;XCVyk}^+I8N{H}Bw)Nin9apd*VXaKZN@zTw6w-5tOTTu6pb<<^VS<o<Y<
z5LCF&XwcL2B3-$I9x}xDC}dbDpUEIuLWR7J!rdF$NJF(2TrSL?B!0Jadq7CCaN+5x
z@Z4tvm){?Dt=_j7l;n4d<Tb-Om`H1#Nu2mru;bbxy8kI6395Iw{p>L1>0{LUPAqHB
z%KR_FzB(+*zT0{jx>H(aXplxyx<MKR=|(_7q#5ZDkQ}<ZMN&Ye5rLsoNdaN#9*_pV
zo9B7Xd){-tbH4AdxpZbO;J)wu+k3CQ*IN3U#tiyLT5fGMek@?9;4(ntcwVeA5L$1U
zsBx43p(3xM{oW*YU8-0A)fD<RQ}CPqD}=+;n!WUKM$a=}sWM>ZfDyvIrHX>uD&jcb
zjcZFHc7Vz0FT|&ohIn9iP5chJ{-HhId#K?LB-XIIB6<Q$3E#Z47fYJ2n70e|k~PZz
z9F8;So{TLf|NKt)G!~AV3xgq-mhSZFEY+_YSO-?ii{b?Jc#bB_9sA%++G4-LNHKBz
zk-$Lu6;g)Z{~I%dl|L_aB9>|a5DXSRvxDXasc#pn$Db|FUIAOJ!YGj~5Po$i`#{#S
z)|GoDEWPycyldko&n1~uI!Rw+(4+nJLdgjpns!O#r4(<XhPN28^jb?$F3ahlbYf-i
zr|P>D40etouAS%5a?so>8ysBcma?MH@2ZA*0bGBQCMqWgcbEyzua#)NCd}APdyc@l
zdElXhG|gB?zlC3GIohU?P!5#@trDywHhJd4F5b^`Gu`=@I0vXVx2)UfSCa$T!9@%(
zlGMZor*fy4YAgMTAZ&E}tL{F+eMIh5<9@IC2kwbGcRn9o(y^q@h7mzrTuDvP#}q)V
zEMpPpKB`RmcSTUZKq&Jv&JY9G|0KgdzQ?~XjN!fnrT+t%9hJuA?=LvVtWSoN7O&~z
zIvYp~S#g6th*$|m)e*kMWf5hY*LW=O)H++a_Q8Cv7(3`rq+~t#ts1vEoMb-|U7HEV
zI`8}&yHny*70|aqhHv5seCk0`77J8j8Z~J!uHd5PI&%=Z^4guObR?m^V-P5^i)4*F
zTsty=1T3&`GxiN0FGqdA2B$ZZt&*RpkpRa9F^@3>mtu6@g=?fUH34<U4*)r;gQ&pp
zy9fkHSm+I86~M5Gq+DCBV#aob^KjLWdcs7aC_x-ac*G^!gM#zrD<AqWvE!36nV*r)
zBp;|NjgAIIjWWVicq1=ku3~<2%DxCo4xN@{<B`m>1xwl_SnL;E2aftKK6$*kDvl6R
zvx*m$SM@7qW+yLkSyxH6Pwv5%K14fKC6FVMzgMgfsYp_dz-LkE)s0k+`2>@F1@dK!
zgnu<ZDWsZh@qvaCs9pn<i+Ivhopls<l#?wXLdGA1LMP_Jhe5GASoPUD8j1M3@9*Iy
z1M-kHUCM*v>B|8XCeTU=aXnSo1J!JjQFzKZkthBHo7)osomU*QoD$m%T>A;a0@7nd
z6-kJCAF%Db%l54h7~c&36FOQahLy<YPq?N6Tg@ObN@Kn#s-{#(I}Hx`%|;_V`#!q~
z_lSs$@8WF`He~~DZBTYNSs{1r8?rS)=g3nuxs+yYIy+A`vB@&lRV*=b)IpI${=U4M
z@~@+6@3z4?QO{h#Z|)3s(Qfo^>^IsBP!kBS{2mT*6WmL`MFT?zY%d=p9VftyuC$E5
z?;%d98#d9?eTz}wKB8@v9w-MZWNa|@6{*9Ju9q(V9$fUnx)_EIpFFnoj(@~BW%}Me
zZyy%6H9=apRvyhOh=Xy}*X@Qn61sNQyAa4B@<UYoE6X{wHatfg^IAi&1p2ZRz4jY!
zd8A*#)OdWrNzsqev901a;ES5_2qSL0lnwOxr>M?R&(?-Il6yG{S^O5BS*L>wDKLqG
zm(m7nkGewCYc(#lP7Jv@vK5z%4D9dvj7gfpJkF0Tscn;ejn$qZJ>_?*i?i`h-i*!5
zQI7WEQkoQnR?ivnw#_3W_hWP9#-?j=e!J&gEkstAs`=f10HRETI9U2}sJ(AR+C4f4
zFM?`wweH69qZm+7UR^8NnF68qlnvn`2F6E2qX}JS#mOqiRornIJfswMbot%h`dZPs
zhf<i78fH;xSV+*Mi*=shYYI<DT1(RfNZ+>=Wx2F=V$qpXb6;8;6={B8Z@Bo*&ljrt
zdDWb2_jYnHTPDl7Zl=a$oD$NYM^taHqb$jks<Y1HU4(I&9vMp84psSUF5(p#WEo30
z3|lA5N}Wg=zMP=9<Bu~&Be&CFXg**E0&p2e4E&RnM4aO>Owd)4>_<_Pb*x^m*0t6f
z9mQr2lxFgDD~EiX8D%hQ)B=T)>ASgRHYr-Ar}alp7pS$+m;N_C$5~7CB@XYH_Eb1g
z5ryBwxCSO&D7wbm`M~32{#wPIQK%TIF0JCv93>}+hnXHqu+Oo2i!+8IDA|6@gy!9>
zo}*gico!?|rrw)KY#EgF5KHs~Jtlz%E=6r0Ef<Yhh@&ae`?#M$n1=2RhLIJ}#K$vX
z7}Fac@_TfQlG~C`aE6$yI+W=%#F7_MH6ue0)WjXcBJ=GHi@g#j<TE0J?yb>?o6$aM
zKh8O5c*8lGk<g@1peNLOJR-q25J{$F8jG?v^D9ns-xfq=QRL`ywW_Qn7Yrb+kfsSQ
zRrb@&9Ok5pkwV<U<MB*%%5z=}4J-TuhcpQ)gCPc2SsL1b>`GfW22&CVe+X1i@0~{?
z3f0{H0Gw{YDG?><CXcpt;ToEg`wcv{W<0#KQOr3yHMTSAWueTD`5lFO{MhLRB;w+F
z>`K?sdO*CT2jXYk0Hou_ljv{MN)AIV^9R*y0%Qdi|H@Q<B`j*Wpf44aztVS0-%jh|
zS7HYa1vbiZzaY)-cSaN$9}1I2Se#p6kfjdQ!9OYGQ(I=NW(Gtb+K#4dVV4GTK5`%M
z5C{i=uRyK3U33L;*IFSePYwTXZDT7A@9nwQT@1Fpru}?!&Ncf*zHeNw5C6G&8-OXP
zG{n1@IeA>_DUYwB$)4Sd8^?sxw;_G@MPU{9QKjS7Mq(~$TPijX`G(Xu9(xb_?Iv$u
zRKpV#;OEWLZLg}O?fnm}Q_?mF6#tAphyXEcth4fO<??|obFE@tQLmn>gEJ+mGP)iH
z^4@5LcPfIu@ki^r0UP~M)FUFmYAHuN4QUZaK^(|uSa@2oYK~rDGg|Kcni{M00c?~`
z07=enD-BB$U}g^QRl7x+MLuK4gJUBZIR}z-rUoYFy!k)NKN#E28`$VHWC$*7EmMR+
z-usOJ@%dL2G$`b6*#@~Pa?PptbFAut&;V+&C_UqfEO&zUhXPmh74p>^Y?<>e*Kv_%
z?aTg$%EVP;eEw;5*m^8We9R#E-tp9XK-^-l1{x`%gdC7eRhPW?lcEnAG8W`O$?VlQ
z85P{J=kn&AV^0WBe$SEWupF*M+HB?)*%6FrBUX@Br6d+q=zN?CNEm5v3@OW7l3@bS
zXgvc&3L;cLc%ys??y8G%dLLRVzpYT3Le(*_nLx89&Ua@Ykur)D&1{NS6ic$-FKB0t
zKZYw1y+cXnSkPXV9#|D<CCrHWEdL2W9DAD9+rCR!Mfs}j^F5%^IiJXdif^F7KE3#*
zeM^5n;G54AQ=n;W$s`^5eG<<&$FbM}Ctg1om7Mm>1mk2&JMay%?tgk3%M~qw0`MKm
zQpzn4a}R!-4jGWJR2<<lK}w@m9{5@5p}xtpR1LqV<5lF&+5Me0H}!P%JWJGm=h@#+
z8SL+;Z1(bdqZT|%dc`4RpxDMRN$kc5c$U(6?`j$MpVJyq2vc$pyJniq25@F|$XS+t
z0zH@b`QEuUPGVooU5*z$bJuPSjjI2%_4BM+NtS0H?pls7GaK2Rl!i`UBBI|kRr9!a
zjGjJclExqfbpz}8e!j$M-<>MX!bMxVCc-N1n>d(j?Lyh`nt0O95Vv)#*{v5dr*;=!
zM3P@A-#9Tw4Ei-~4WQ4HB4iy*sF2e;0Ut}C^AIhvnUHvA`q^HB>>aQPXcZtgxXsU1
zZ_nQ2GBA{$@gng4&hSr}d8rWOEWS2e{rd;Krd~Zz^UZsmIiV+D_YyY6NCI#tRvPQ8
znNvlkUl_0nI@5#8(4#D#HDJ|!8(QyY=kZz7{@MnA99SR-*c9h13T<VbX)5V)_1cun
z$$tG6x2cDB_2=iRm(+RvT&VufJVBG*)Oj@QG{=?T`Hg{>!OnzPG9Q6f0Gmg))>(W8
z@BM@2)SNv%C6Ri;iZ|7Rp0H<4&W^Jyse4LUVU8&(NkYZ;Eyz%@3Cd1S6L*K;$o>c=
zqbCeN!+g&~Z&1PVvEG($nC|?0{amo-b1wh#H6_@{LnHWJ*=%zYZzRR@`RRI|-OpnU
z{*giUTZpy&7170;eQ#_S)~CI`=(F+&ra!X=y+gR$;CY893{E?PBC}yEXF&P)Th1JW
zUGs5|f6pMdJEAGHFg8)837r><-%j0T?vXo1f*AT<=AKo0Ab*}^v?nQTbd8sE+0T#P
zB~0=kYFFSpt#yj@3_Y32l_WMBWR4#U{HxBqOLKq4`4n@|ZtwLI{D+Srwj=KdI9-C}
zS`c9XN?COxeu8jfjO32Uj&dqj;u7(m<P}peAxKloSJEs!Q_8&=@ou+qt-UAUilA%j
z@AoueD73NPPyC8S=R|VO?9gR4X6>3(*;vl*1OWX_jjnH+0F4rNq4pkA${cC8TYxDd
zeR3g<>hcTS100%iJVPkYL^I1CmZ^R&AO^Jv|1e~B0(eMPNtz<K#vk$mtq<vP6z10N
z-4MF7taHqO8x98tlD`B=hQxojcd#m5d;3PN4a~P#;*{abe=zp-(;QFGn$ckUAD}hQ
z%t5Ckx|jj>H1@8hT*V>O#es7&6J9k=R5>ntMyEfe>zpQ)yTW1bDP4_F9PY4oNX4fG
z+d*kqBfagN(jHCp%O_GcY0a_DaWIKTiV63@H)Q8`+OI!0;IEsk*aBzPGQ&zQIC_5;
zYn|l8m+tw^c@|bkV{_B^ZknKEX<l?W1~?+UmkiedrayaRWxN<lDE2BR-fDAcT(q#o
zmdkpMz^Cxz*LEphhlr-v0PvsrDym&e%Or%7BWuq9pIEt<*y#k=Q<{ewpB!%Hed(nh
zIhUL7iHp~ez0!Qt*rom!qPgSs>-UecO}@q;>qBXq4FI9AwXtfB1b_xu{-=4^Mowrp
zPO%DBU#TIA$u7+!e2TeuMKYfXE02Q1rFY5^^YYc}Us=h*&hc-;uD<|3>iYnN+cnAE
z4pNuMuc_!e@OK;gtdn&1-BOIi41Z@UcLNRf;EJPfRpkeh)1M;#&joV7?7y5i8;^Jr
zfao^1rBFP7NTaP)6e5|AC0>h%3NdcMe4TZg(7vQpy7@%fW+d#s9ua8BTsBZmCW43b
zim;Vh{9dhX+HZDFnE(%rb@bp2R#2gi7{Fm1CNao<jXuSKAsds*!3<R@?p%Jq=UjCx
z&++3lsC3g3Y7e*KZw3k&^d!Sa8P=-doeL!W!dMDmEJ?Oy&S8BFS+*uP**THCt^hO=
zDk+y9N?=G(oYJ9&PVVriOzkT$eIPggb2)9&r_63Z^?9^Y3)_%tlcP&{u^@)q=<LIi
z``>-BLwKjD7S{7%Q2Cx9`j#wBq+81a{`F8z8Z;dfoQQb;T^qL6qSrO}0IFJ&aI0%-
z6;^@o%;rt<PA79JwgmV!DC{|A1fEV|s=h0ZJ^CESVMx{Y+1D)6eeyxPt0QmA5-a3e
z8rOPMRQ8D!Qh_p_!cNn<ixx>bu`TTRiB~6;!=XX*JB~oZew*4pAPC|qlAwYcUszcK
z94ahdXYV~TB?WeL<0ovPTneq-I7S4hA?L^Au6wCx@Zk(P%g)EhpDbtItf2a36?2SO
zok+6S`X1G&5H<5|%?x>lG9{2=@H;^RbIv}pM_uv^+QZ9AQV4}<6k-*%^iO5{&&zL|
z7-kUa6o>#SuLW0?q*<e<`UG=d*zWu`pO;5)BvbYmJ?YJlVAN)4Q}}7kfD=B{B|u;U
z>f;}>rnJ=<$vCEW9}-Yrco2|xeAe;jrU{h0M2&tYwK*xF;ADH-osCsQb9@FMof5T%
z?=_x{TsbRzrzh+dF76J)Z}vfYp}OfMiT#o2%)bB^VyhW)Q7@bUw2&VxZ+ulLld+?(
zCanD?)abbd<<}!H<TFaeHFL4|pd{(umhsU0s0a6uj0?NeQ^ebty@BgNmWSmkQV#<C
z59tJHPVY?&w<f#jQl&2)5_w_fRsfZZtd$nQEZTjg`Vv_{wHwR;Zyy0%T)Vu>Zi;Yc
zEBUzRn{*c^Pklw=F{!fEb)MTDdwW>VR3?5T*fqSHSMX$O{*E!*7P+hyZ9j_rfo=IT
zBIrF&Se5zYl8$Ibje?l$AGCM26JtV<9-pkKckQV0rX1}SEurrlL9AYyml+h)=(XSD
zOwMPhQny+z>R#qRtEE>ElS2F6wB4J37#9m<N)I9&A{e3tg1kbd+CCg;A1U+}(la8)
z4odB}u4u5wTiQ@5HT1Njj34SYQ(X>uE}eU!mlP{Oei}JSys{l5*3Wxky~oxLCZD2@
z8l>#~`c8*u@X;82g`Kle8o>gjIXpM4Z<G<sF$zj#=5*W_ST=l1Aaz^*i6rc7lI8qJ
zS=voHfLhvlov;}2aPfsbOH%ikru=Rp!MOghb7V=(>So0t_(pK#9)~ag8u1qpjP%1~
z2sRP}rGE1WcwSWy$<BRU^+JtpKlFd_T)oPY{zyG)dM4~=F`qI2m1&oG3KuCbhi@Z5
zByQ-3!B-pynrZSczV=Swyi6Q*_=8%ZuEm#cQjXW|P?u3&8Z=bSQOT7*U6cOJpx1=0
zVo6DoZUlxV$+B@%{~C}y){qC;>XVQwvk`DwLcfudhEnbZqz`h?IMfOmU`e%_EdX^c
zv$D|&aE4@qRQt|SKehG{wBxAG$&2|{PoP6XFd5KwUp%dfl!X^21zSn>Z^w42NZbOw
z=0@B`vFlPvk$sSTimaKetSu&l!P)IV$MfJ^>hJD9xSu3_u)mX=1Y{H(mQWB!f#JrH
zn4ytI3OZ6A^!fI9s$Snu90lJoznn;$`}`TuH(mdv9PlY)!IKeGxS$do4laL@BJ(aR
zh#1$=05=FYG!np>%U}J-K)P{&4F*x^au<oPnk=r8)WO62+Aa?M>R92M=aZ-7DX+r?
zlEtEOcef|1_R_HHVTo5QB%q8zei_+<C`KYZCj4ZbYBpBtUIEEy?Pw6<8f%5opHaEM
z70p_BQ3B}Q>I+m_Qu}ml16KZ{ZuqNw(24E@&Sa*}9nbMf!f62mIjqlsz6zM(SHgk7
zW2@S}orf4;NfVhhPVKt4Oxl5$lQ3tNLjn|kxFx|Qb$5b|z@c@1_e`VHtl&eOxYiH&
za17`Jva!65>oq=6CgRICNbB`E?y)`^Q1l5r=W0PN=ztBSxL<PLU*s55GRlTo6|Gxz
z>p{3QD)^T1EH(57jh;V-ok{tq<w6`Kj*_fzYl;xzW1+T(L0sM>3x5Nz-%pmZ%fa^G
z6zyMy9D`N!X5<fjRA(Q`6qsAa=MM?w%S&=#ty8uVb0#-q8<Dr_<pj;(x&}3e$mM(?
z+8Rc8XKq#f1wyNbE<W`MKM5jTQ)*+1{jyOZ8RvNYO8#^keUR6gA||6I6mpvJtZiGr
z$+B_iBkgYjDeSd(yo9KGzsq=qzCZM;nD=2U+81;wgLH<6tr8%ZNIYuW1VsC0v44j>
z!V~vMygLlU{`+AN{Wxh_cf{ygJ;%<~+H4blu-?U^iu~KG3ca0UWpipT|D{Q!@$L+W
zk#YhT2|M`#e+sP}qe{la*@w4LU)7CIbd(-tpAKFvD3}K==csNqs5$)xa!A_NhcPpO
z)5!>%z_LFPBnpAyA3!j|E5T@DvM<slWzIjf4!orjv3nPQS@m~>I)WJ~dS*>f5$l(q
z*X=7{!)Rl+xY5v)c*2Yhm8mNay5&x|(OVS5b_d&tOXBn%J7Q3x*Y4o|rnwJ=DN@Q2
zg(OFKJNCZ3+^El?-E!vzMnlYh*kR#szrhtN)MFkZ>JdzLW0YC_P7w>!F+2KQ@}%ne
zS4PHjCmB6H>&-O<+0b)c;!XSI(N0pi)1n)l#fF=;!L%&*xOyd8vpJQ4K&p&F;)@vR
z*f&G3<Yr9E2NBFOi{&31aBU4Q#n!PU@kb-t!Cw|anfb|a%9%#hb=He8+6G)+nsgu@
zlCeF`8-E{Ku2p_w-pnE5mQ2c*qpTTTMC+IgB5Q1Z{krX3L3y)W-@a$ii?=(iL>;5E
z;ED{Z5bJ;=tz2PR=`flsB)r;Do}QIn>C0ODwpfwDkCfYY<<vbN;xf~T+hm5xH3y0?
zIykKS3P%xy9Vwv(_kT$9DY!k|pq@Qjyo0KNxRUL_QGEwt%|4@ZRB~dfRH7kL-X9=0
zn|i!Zu%kCIo8Is9!ye9r=zTzgSD1WSqWf|Umh3j!&bM}lCmqg9X?{@IU!y{V8tAKV
zLW|gXFRCV%V%!mY2>K)}p2v}Tty9v~YKl|EHb@#Cr^rmhsR2ADd1cKN9y0`V5)T1J
znnjiM=A|3JgtQ{fHxr55%f1hXhsLO<<f%w=CkdB)js|4jf0pbIKZA?LtL99;*fUQ8
zDi8$R(Qu$vH-m^~SHqmwclnMFsmHBELqGT`zAY6Qz*!!in^HG8p`U!!nCPjvKmAx%
z9dh6xH>4xErHE%5t&P{~Q8CTzsfPc)5T{#LxpD4u#UDM(<dMndRF%Dmo6^6p&fgGS
zTqg;5E{bnpS;(Vm7Xy|quISo6U`lG%1y~l7*&*)3^szqpEW$RFX!45?pcMC)9@9+3
zDFgM0ZYCDR1W)A^z80v$a|$`m7hM6#5&4L0&VIzuUP4PFv~Bmh*7O8DVa^{FX%y)d
z`7FZr5SW*?xGwmw74=`M;ztwMJK-%GOufxVC%qU225j$qo@CarOx861FxE9UA7V?8
ziHubU>x>-f;MQ3}I**)?3!;dmLfYlRq$)p2h6aF46a;kgYRM!0s2lU^eCD^aS`$@Z
zX*6F%0bw`7iI@?`#IbvW(sSia>T$-0;>o5?YD|+vvlSd%Fs~@4H3BIU&k6m}nc`Pv
zLiu#)f9@udf>fjE$A%Ll{yeXSUB2J$XU3O$hJ5qrSAp#L8G9CLW>DgnIl?@;>d*H)
z7KI46+OB6*=wW`V0c8RjVg8d}D1SdfHQ7Caz?QgTqmJYj-)pdu=$15CJm0-{Ast0P
zH9k&>z87~I(PnJEy)ux(6&<g*U85Ra`lN6VSrFfx(>Hzk34tP|SJI$M$kNMZLQ^K4
zEXsx|if^i=?ag4PDt<dJ?ZH-XrP6-ygx0gsoLz)Y@ymq;50&AO$7B}2z-LY}JG>pC
zy3V%YN)*S*zyC70Gg7NQ-j(0KXs}M(NVLQTfR<18L{yM6V!|`!ib#4+u607Do-6`B
zy|UTCsHLq+e~MM-D7$AD1SYipo%rU5Asq1)bq+-Zqjn6T-0B@-n!YBa_%hxYY)T-n
z+g1$P9soTNcs>EmC^x`#W~bJzO{TGfvD1K~F*@I$58wpBX&L#fNj~6n*h7g2dR%_m
zv@P1KXwNPHMq1=O(Sku6j4BlOh&jPn+7t#zr4O4Pv+@6w$5M>^X;BeIP!O0d$~V%*
zSis))lKwHmn?tDK(Qt3MAQ&2jN3m$!NRdGN9=yufi(7^F;f?L)n@L<;K5s@4(-Vj9
z_Q}(>%(;ZD%IqAkp^pU(jzx!YaozBjKv$(!#%tq3_a%W|p*4@f-5DoHgQMc*lz3G$
zbMrvwM-8z`WX%8-`)4^RIme=V`J}U-c^z2WX#&$p;_)@-RWfTo0Sy2MGsE0OFPSzK
zbz+T_J4%TF`T|CfykhF7BhntV%Q%BcWz~s>97j^n1?3IY7MT9%2e%&sdPjnGJiXo4
zz^t_A{CeR($Ati#lTRD0h8u)ReqL`YU1Jy5>6iZY2(@i+L7HnLz=$uK-WP*UeacU9
zoVLz5hs!Q6mc?>N|H#!jm`XrsL|@fiDAIW-=fb8vlwdIxG@~M&ITWA%Q1;{l(<v+W
zI4QSLJ<9+<{}S@h5NvA|i6fz#2t|?DyQh2P#!U%q-+PlR_*M3eR&F3xG|3>&tqvvg
ze`o;&!NIlUvJ53oQMQyTRzx$;U9BD~WZ68ae@4Z9^`c13HNTZSAyovg&Nh@#B=Uri
zGl=!Kf1=ljgxT*H&26v8%*_HGjQalQNOV<f6h^u^jqmPX!G+=Z@4;36zXTV7?<40{
zZih;Pmoj!|W1fy@bj%wibuGK%9v^3X87--B`QCwW`r%E6_mFKPh;6Zseg%l9M10Ml
z3+(qtrBRZFfe=ZIOj8eFktuFrA6=a(=N<qL%Xx&v9FrW6;}@~pEd&jT^0I*QB`1_&
zlXQqk&x7B>L_hqChTDL&dAaU)lCd@K9LJvHp_jZd%AuW;Qo~&1Xs=#L^;$lvpypIJ
z<`ZYC8S-JfCM1~1Z1GvN8cnv%L)E4}$GLed$O$A)Rsmx@1{B#D^bBBD-%^Vqo4sD=
zR+p^;J|fnmNCps1(1bt1AO7V@*et14$tD^Jv9u2;EHX8Wm9}hQKT}SFH>N778=t`y
zHB50h(v`10XN1{xQ?LI*a+D?g%JHklA~}{ut&wYwZTgN~un#N{fH@x4y!FVY#wokc
z|41RSeBjw-z~j6o`G$=<DYb{(Uy<6u*Yc8<6Tvz|$RUZZVR4ZAcSCE32+)m#bgrO*
z>}$Erb??h0x>GMmIV`#^Hl^nS8xJk%A;{3QZCI=`9R}DfsPv9WqgSS>4FWT9=f#Os
zJ`0Am010!zZ>Dr?#-S*77xAqm280v$fkD|ZenUQ)^Md;A8Ak7$7%(;NKyopHw5Q%O
zL^fB~r?`F@ho+rcXAC8M0d12^D+}I$h3|Z*RSZC^k?n^HVBz;##Zdcr%hnBUOH$H0
z3{sf1^+d5w=*Q%VUE&qTHX}q6Oi(Tf#5mS>$^f>W76OM((<qMK+0#~aX$LBjv(C&O
zrXDx;ykv1@K~<|{=Q`)w0fHK8j=@8O?um4#EgcP6JUqFw$;ct=``lT>O$2Oi0dVgg
zK2M~xEBAKmg8+7Kew7^@TQ1z(3|n*3om#K12-Vx4uw|hLj&YGMUbC9Ppu=-agNU!x
zG=R$KX@Wf>qN%YKQ}n~4I%C=uAF##5SoG8wb5N4CX2S<mY^tV?*PwRExY)mg?oY55
znWi+5^(eHygd4(H&8srU^4Qc_j=oW;TZgtQww1r*D~XE5lMiYB$tr0lS1tdJBz$;*
zqwyGY2lQ&B4&fTjl@3M<4vlAUP2VuaGet9fO2tIJUZ30T4b7#?$dK_MnadneHPy%8
ziawG9DNe^S1n?>_Y4cOr9Z3WZH6pxCz?P5;2kWsZk+&i>R$XRhFgM~FR-a|{;#&G0
zTZ$_8SDmKh_Z0;e0PUM_n&Xeno!4M{(6Wg#$hNwihi!O0m_flLWUe(9Gj}G8gKvH9
zg1%wj*Z=3s$EVvNb{5VQt<WCUGW;29hSS;znp2;&$L_Z4Z+uQOpFW#OS)KwCGAeWz
z6IRSC^gHsqFaXJ6I36U%OV3?2Y%697@}$~LqBcY9j^?*o9cd?gFX{js@Vxg%%f|4h
zyv>}?cf%o{v@>$gC4~rWh~91sIU0%o#AjJ!LqqtPsQ=WS>r&2101AvCQc;qUc=Q$6
z(OdGtw6k%57$%MwF@oyLvjpt~BAe*r07$!LOXUa|X^Ie&?zdi31QqA9EcSdxZN8+Z
zKg&_uKjYjfJs4KW%V__3CsZvlw-W3sNkJL&gbzr2{*Ce0ms!@cO46(Ii1VRtx12x$
z&1^vNd%@lA;8Iw{at`O35~5?1^9CM~2nEk9OJs6jITMrEK82VL20D-s?RxXsJcSnw
zysBGjig|Ww(mM`~XOTnKyPX?zq2cZRn0K=aE&93Kyg-gklim&Ni;^G(`c50Cy)fm#
zv3s*^fW{97Cu2xhMgAVKlgyj;!S<ilxAOsuyYAKfj_|sNs?>Obwez`~Tg+U~stL%F
zq0g{9!{tI9eAc)uXd3nm#vVvThx(7rDwu_&t;vzspS*#a{K`iYad>Iz3)xk)C2-TE
zN4D`J5aK>fqmWDztb{)yVe$F*tW$K4vt!ETl=q=_a&UohRBylmlNlB)8ds13>(r47
z{}JMMAPLt3isgrgU28!1-`USn5*(x5BW*m8b6CtS#fi{7gMv9<IxnP5c)zY1B<{87
zg*vm635gRiF<>U7W37J1agyWu^c7gZrBw{U1vI->o#}0}lfa|Bt9NF7kvNmAGztem
z<(eZ`OsoE+nNi*XE<e`G#+eWrDf|m)@;|zYOC`lI8PF)}-O>dhK*nFlg~o9aj7*SO
zF}|YkwP)sGla*eIBV+HI3SQ1lboNTo7<1p>VzUuFf=U!awN2TmRNuX_dCx}3mWN1B
zWT-p87y(w&9So!4)QP5v3==Hhdu<ZZLl6qVuCx~z+;9uz%}}omr)3Bm3!87T_my;1
z%srtk(Jvrzb*}MH3h7i{=WbefLU|d}i#WgLm#`ScnP2i*eo%E3g9;5Iq?O6=){f)h
zMHEu!haiHWKVmLd9C0p?Y+TiZFSv_Asumg<blOe<GtgF~6@nQi86}lCJ~M_3oCHAF
z^*k^+VIMk3D~1PRQN{iVDph!dOtqM>0vr?pZ83-71)qP5`ju>NaeLx3<SQf$#!r31
z<EMb}C`m;P{TRHCffzi|gu%Vv<?pp|GmwTmo1?8a*xTygZmJjO3(lH`S@}w(T{V84
zoshjCb2vj9EeeNH5@S|1H0=m-7|Xm{$g>Tku?F9u9lfM`3$@AazZ??fu(*7(0{<l(
zY3h8zJs9DizbcdPgb*M06z7C>IoL@HCTM|%&1n~^nk1s~IB(~pq+T|FX8$wq`s)QS
zFq@ymbRMN^uruBZ6`Y%+*T5HXVnB3nepSlG8FE!>uJq3nUx~WM6++oN>&F(De!=Q~
zZ@m{$rR}#uj{XR1J!>dSX-trYY}ORx+x7Q}CF69L`j0UKC1+)xx(t#b1hCQr6(Tmc
zm&b48F!CnAT~s-5p9?aS)3{b>HF({|YY7?udL`RNO_h$WYre@IbyQfX639&KD;ItC
zxX3!Le*ue5iDAsDbz8rt{^D7RCcD(=)r?B|OP%=W;ERq(FeTL`)rp#MnC%*qw!X-8
zitFqyV~=;4XLZ$Hh?g~9vE~S4Z$K%2s(+)dlg5yMoyVtoG<Of>p`9`HNJ}ef!LsLZ
zd`#mren+62@FoI6l$gl)aZg0%)ZPiNyA$xaqe`$pu_xE?aAmIgbjx~?W6pAm8ERZb
zBF`Yl;Yn!){Y2ee!6oSb(Z?w>eB~<pj8w&_ReyS*w|3>DdK#~=l=ZA3H#Sf3EuGY&
z*t}q~q{maYu!lm6me@bQzvk+gOVn(0@-flTS|lz-<$;Q;XQ?lto$zL<Pv+KjlC@M!
zt;Op%{^n?6D}3K|7hQct&VGm*AG`2@(B8rOm5SyO>X=wuf9R^vZ(~K*Nee-ShtHv0
z`E;d{J0X<U*n2_N&Bh5bI}G<1c7PLCaR#NbRR8L6??1KJp?11xOqMtF&qJ!tMGSTp
zV&KG;)Cv&osgL&SJwtAOJ2`f%?{c)04QF?rZ2F|n)(Onyg>i;x&`5se^h<>*Ck55f
z+eikFV91*#fwn+kSv-wLYvb$)lPH>dj7*lFzvcITyH~^*CV_DujLr~-&6cZCtmA_W
zRuj4yz$L^H(MaCHIqobrHi?Y23>xRBV&f+f)U`onVIwH06Z&u+FC;Vr6buy|S+o@u
zwm5sy>VmEXGN7`c<c&_yQIJ|m(+EDC)Q8A4!*U<6BeuDxMVayEp8-j^l58amJ?R0)
zU~}hZ78eh+6)gx_JzkV+ay<s(A;}I&ByP;~$sY|ZKtcwC-BJ-+WqjXpz+(T528z@F
zF+0HrT9F^A@=CT~_*;q^d$DJ7zFqKHP0RkHtk*>>_Pfs|6-8uugYt<h>YR3!XhNL$
z`;GB%GM42t^@0>V0RB(u6N!3Rv=tpZZu+3*40+zP;+OO^?g<IhSxNHP^w85JVoIwa
zOf-^NOgD(4{s20CLczI0hU^!Iw<SE(h{?PFZ_aoM6-b}Rg{)N&ibFI}XMaj+n-#5)
z7Ui@Uu+}z?^9<JfMY~PeYP9$s8LObs9=0C4Afq!e4S}9fLUv}~_cc)zz2L*B8bI!B
zi9irhflO$|<RnI?2<%Z}Fq^}Uw$w4VPF63N`7~y*<=rWT8R>3ZlBR{-_p}CCGDZRC
zsE4vYk#S5y+5wy*(G20=W+cM>3V7VF<#1K=osf<_0h21Shs^wes!?Qf0(%*t0Mvk^
zJHbHeEsY(n9q|_(6N_2-jH$GzYu!vchw%Ym8@#PB*~fGxl3m~u?ky!;wPw_qH?Wa*
z_1`wq+%VXH&NHpZTglvp)E|#qQogC1T7vV7CWDIxo0jYAFoz=Ns~C;Qg~6{19t7NY
z@&*we2y41^0=wc+N#1}nrEIQViT&0nN!KJwF~PhS??tz`Aesc3s3B{2g~MDS{+j#`
zsIeXnJW|knKUMo7CiBzvBg#M<8(CsqC6LBwQG``{?*jSC{f-P5If7*qB$bq^8De8v
z@>yv|g|E!NYOv%BdS&1BF1XMbX?}!dd(=KAnD31^Wl{aiQ*+8EUN7T^l5#-Q@{^Cy
z%FR??`ZpXqet*Yyqp!dA!nmf{a`qYRz=~sLe7#tvX(eqpN^Bb^Xft$++z=EU(8Me2
zV^zm>8sS&x$^?>!D>RkkpZwvrwRe?cDmc^qw9`I&!r(=OWI3ta+*M_FBM`jj|9}X4
zz=_)5Qq6cH3xoipouqCW7>yCik^6IaQS?5gB;^lKz{;UStF3XZb)2QjJ23g7c{KWW
z<rOKZai3%EeUvNX6*D&Ebtf7#MrZq+%2AeP-7Zt!kuxO@W7^8X`;m2U8Wt0eGuF>$
zV49IilKv13&FCG(1Un=J$?+04a6IIHA&N_2=$In+CSH!P1PA?*B9<zyhv5fnMbKC3
z-ms+3Pip9lgm1i1fk2<Ow&*t@MbP(UJaGs9OEr)by_yP>Z$eom6FZCqBhe8H%TscQ
zWvH?pQ+ioZRxLyqQbHa=jzPk<V~Xt?)2(Eupx+ws^Y79pp8*Yr3Em2Io5H*_>ur=q
zzwu>4F<%Qxm)_>w<_X-hqAHTb;|4*iNSw(EWPa{kMb_qUy`ib3&VznuU(uQsRfBMC
z(G)R~7jWA(ijoH)L*I9v!JWw*%eKX<ld732!rx)6+A+Y@+?ej!aDA(!aQnhy#>MdD
z4GI@8_ibZiLe;my8mgteRr&*B?ov?THpT^-HWJ^K?XU{pd9a9gMC2@UJ=9}T*-gC0
zaNll^$_8toyFQ3p_2P*(6fJ}`$@LZvIleSM4X~Krbcd5VLQc@s*m^=uQ_{w(_lm?^
zI8J$xu_A-lP)y{=`;^jT;v3x;8x82&wwc%z$8r}%^qOcZcH6`#Us#srA}_3rV=0tz
zX@h2+_(ZcfZ3Blx`N@3MV1(4d5kye(JOn6P%o7+SOAQ)F+XstmD6c*Lo(lc5a4iK+
z0_H-tAu1x0ulNr%(b@#v`C{Bj&gX{EjD)(g)87hGUmwE)E{Rmu=jPbl<;XzCenJ<+
ztwESKW>PdF0uJ)!o$8WEv4IljO=}7}Em<kc1UmEXO0WniP7w*_AWjfBwXN2>x-Zn}
z<VW{wrL^&RD^|HP-{Va^iR2qP!x_RhYMtvPnNQyRjtPK|4qq``@^(LWGKqMR1P#gh
z2Vdh8HCWU>Kf}k-PYSEXC}Z=KD6yyPHwU^L*j8eE``m@lU6-S2mldbEqYpU7i$?dA
zI&!p+!XU26Jtm%2TxcYG-n+^XpTsZM-^>0P)Oxl1I3(`DbPX#QdV_~=G5$6Mb1_t?
zR}52t?QQ$ES5!^B4bbj<^=&N`>}_A;aS^ETCZQaZhZvIaVaT3Ta_4j4$2e{hEwM((
zCcTm5i7wlwMP4$;OHN2a>zQ@y50InXMV23Of#@4`<=6I3TAGA=PM&QPjcP2ta<?71
zW5s11hZO(ejb`V;>QZ+fN4K=tD?0yKYUq_PdM>gThe>|o3lA9tvpCKPxK)#b21!sa
zkK=-{Jc<9kiRFHTVH*}n)=)^%8_nyYDabNmkWAJ2dG>jhFl;!_^S_jS|7<O$vtu;a
zIVbBjp>~*bd+x+!?+AIi;nf8jwqkK}Rjg4M)N8ZPH_ea2>>Go)UL5@x<Kdxv$3&#O
z7xU0VO6CjD;xTVcea*afTXRWZD<l|60ZKoG`P|G`prK0~|22hg-0ErBL@}DF2#@fO
zO{USt8irS`l&d{1{yQcmqX_cP{!g$LCaz42hZ^}EknIROU1WUuTZIIl_aR=Gk!}Q*
zO9@sUNCDXWD5h%VZFBuGuawhv)gb-&8PC@RxrJB1v=?WK2<jRSp8nK$#B!L$0$|JW
zBAF9DfL*z=cv|EfCRY@{;3lzRVT^GMU89m21u1{E=@-E^!BY-5Z3-*<U3YzAp^AA<
zqJ$rXkx^R>Cme~7DQ3T%n9uf9UpN@Et16`VoqQiD(`BDacC6FLOjV)cj$YE*(42%X
z4#xX`OXw5bKVz{;N1UhF2sm3rpBEt>B-FJ%aF`}YqJu-C<ZPvPY_@DR=^7A?=aI4a
zF}es06pNTgM@_R7g<~JN_|ZG22g-W`q-!&s9ve)2aR`MA`^(P6b49TjRP(b(L@xw<
zC@v6}du%smxg)L}Ydzz=j-;pRkb4(=FVNnle8CAS8AVbljn&&XBE&jOuUbZ3Yflw~
zMavj>N=N@`uJ6Qevxf4OFjCwoeD(mww80N1P@G@s9q#t|Z`_Rs+ghZ9sph1k2a2*t
zE_>RIHv6$^!6;^W^nS23Y3F3v-TvsmmR%_^!5R!1F%EX_3jq6919ska;P0Y~ta4=I
zU$%sg8w{5c{O}^yEIQoy3ljU+_x@==)=7Z=hkGRi=Xha5T6BQ!d^!G0l_Z)H%sYOr
znRg)CJ-IJauY2-Y$E7~=i+*OqFhzY9VXKoHR9c6VEv<7S?b_St9BJ?gBpS;d@$m?t
zuD(NgygCzbv*L)ze_KE=7$n(lLOU4g<R`i&mpjC)ya52T)GcMCWT|kdz*zsw@v%?Q
z#eD#dJB)p<Tg^F2f9Iy4o+i0SXvWtaib|Yr;}}IAMKZ@l0?Ue5IIgk`L#qI%kp6DJ
zpnT6<rCy;G@af#x%L2&P0h|eNd!#eRz<g^76~Qs|fZt#dPSv2fkZ*oSaF7d#4(rY<
z&q2}^cvGlOB%{t_>?ONbmW?K|u~oyVO)9)m()SJk_~ZH%;X?4;FG6OU%Cp(o;jb}n
zwX@I~xi#TMnM@G8GKXb1wH2%kbE)5%W3+uO)s_~QtZmMBK^Tm^8gTSz?FoU|j)?1o
zNJEPG!X0v$rDXGH6iyjgF__8&xUg#qm}Psjz7Vb;_eU1Rdug?|{CHPw_n5YMs#3a^
zh0w0RF(;pw@@l{SRCjk{0DxpNy<YP($vB%XmVd|_URB$Y0EjWMV@eVU&puw0CXwE8
z&vu9#SU>-+gVBXHc#bPk_1H`87%7KvF}DKTC7W}=kCzJAr98md!k-|q*#X%13;-!Z
zd&i~W_c`Kx;@MAz0YhT;nC>}e_I=<omr8WpUhcfK!TeNhh(g23B3Yw(@+5y@zj>#c
zgD*^X(W}bBWn!tshfQ3bL?vL-pijXws<~6_^7K>3Pg~XPoYw8s0d$0pV@!FX{bz08
zYCQkgQ2nF$mc_IW5=;K;pZ&DlAYYWtpA~r(#57ph-cZ0XE2vH&b4TO0x|HrZ?MSJ&
z(sL@$xUb-&2tIe7y>+`Vl{x(4=h-o1(|dqkl`Fu&jvC5yV1@-|)*coy^Dp5HZBcf<
z!1<3Y(!ZYWP2hGw{Kib()W@2{SPf?>e`oT~H>EmZG8eRdIv93S)LYKVI!#vQeCDKX
zoB%?f4^(xuG|4T!`fuI&e;m;~7`B%9PM)5CGxY4K3rbo*6?$yml-Uf>8=djxX+DHX
z{>Gq5z#c`)%=E_d?9J$<RLQNOA=bl~C@^UDLR!CJ!uoMgln^b+qi@g^gWER%Q;R%=
z;HR8zH&i?HmutMQDGBcQ1XRy*tgr=&Ze^)H4agVvtuH$%aVKW>xF58gVbPFc&ws4P
z+)B#3MkZ8QD2aUBYf`nyW3*B;?@Fq&P2)BaJP(UWZ+d&QEZ-9^jdBH;wBgzTs3Zl8
z^L5b<nmfu8v<65F2f=-JT3_>G8+`eHEa#<3dB5H{93Z6-17SS5O2r>L0862V2%i`8
zQthkv52#zUG9Z9ffx&+3XzFw1A7H7BQ1EolUGN$^0U!NyKL_MaBcXQM+!vEZ4EqQH
zclrGU*#=#)2l!n)+&N5O3|JT*6ZITrk`tuQplg%KhsYnpkuB1e%%IxT(KG$g$yAxN
zu;k@<Jkch9xrI}ua$7Px=ll!9#XYPgzFU%CJhf76av0Q;#K9I>{99f%{#I(VEls*B
za23^;^Ym*MJoJDdfN7JLN{)TB+Q88rL15=XhJOQ=07VT3^9aoD8)5Ai>|_s4E{_3p
z78Bc*!;Dcswmaarbks)T@ntq3SP*!4e~xJB%i&l2P_NImkCM-XL$UE?nqPf?1mF!H
zfoZS0g!b%#9(!cb%M7ooXZ}_hhlqHf+*zFmbjGt(bmycMwFnP<m##kW(Y^WHjk_u}
z-{EFUVAG=3%imO{vC6s^zSp6YO0;y|D;lCIF2c)yx-Q4!r4Cu;cg||}2DbJIoQ@x8
z`~2{Z#<8~{_Lp9L%m6GGGv&=wcb&G!2s;%#n5esXXG-G*xR-Vs6rODLjH&$O?VbHi
zPZNF{e)JOH_}A^lu8B8_%vBqbiLZW89Tnk?I%Fu9LA*eC(<kh4*dqmdhtw`x<`k)3
zF?BQg>V!Ly#ycsChhQ(+YglsGlcI&TYG$?SBx9E!FkP4ospVQ#I-(LI>Rp*Tmiolf
z`=(F4pBbec414Fz3XSTIGGDzLa=Zzaw8`AnHYp|k+Ww6c7PnXA$PWC5Im3K2L}G7&
z${7HD2JB;sPJE3G$rZt(8Bx-S&UG0Z(4zY9{Pn`e4`IeyB<JI#>9`o#ATjcu<#+#?
zr-DUr@)2nf5>I?e&mh0cRWPewiL6N{A)5x4D3;{K3lGQN56Zdqi$*afKjOHTs26w`
z^yfdn{dXz-*Kv@}iNRku&=N*tH%NAhP?4S>_ya%XZ?%efk4IhH4lKk`yMmZgBq=a)
zOj|>bqQqAKy=Mm&UV=`gP?>OP7^WbY)ZGOZEf5Mh)=(9LWANWve_)1-I78^jtnfpd
zCHvB?K0Jsu?K33!rO7a~&G6FCg+7(~2V-n3zCUsm&=w&?eM7v`Se3T-v4n@0>k%Y(
z1A^*Q4f;h1YfuYIk8dP{$8dR(B|mrO^M2;V<==U?CuPByi6riN`3#3SC-jOW1855N
zc+}U79!Ru8s3UxXiJ1$I1>5<E3Vq5ks-r9v#-f_<x5UbXj-dw@(-Bp}zgTV|!RZRO
zuSg>igtioW@AiC^L9n0{?G!ORWKj#Y8*lsAq;r4*pFP!V-eG{Ag>W0o?gJ0c<Qg<y
z*g(s&Vs9I3dXg9iLL$s)u0=LhCnq_gFx^b}MSGFun4{{NQLqFgN1|tq*VE2s#>bGA
z*cNr5ZOradGwRpiyaBtM|AD)*X<TZyqn~HEo_^Tkm=gr|>!q=ex!*?(VH|l%ii+M!
zM`pKEGl}|?%0kq#-r8yg)w%;h6htN3=9Mv$Ig4rl5F!n*gt0EQFU^6jlDPK-5OU$%
z1#(fzn-fJ)m@Q65*1?3!A-ju_^0`d`s2+D!PG8m0;ZeAL&vc(BOzNSvRIxJ5JeOl2
zIbH<FXO{su!<$EpJi0^(JIkzInM?P9fQj2zg<<x@{St0)m4`{Is>Ghu^;EyHQDPYU
zI4f$eI0w+V4Z6^}Ki2|E5VS_`x4*RacD+CEGdp;&L-xn-EMpZlgOtYSFQ-c8@h5Rd
z-^ToAwSIpl9lq5Wt7Fg(Y#{-&<>XNWoj)D&#X8!Y>(Hfg<Rk)EVS-!cfN=B}o|B@n
zgGQ_?d@}zj%zzh5CXDnjK><0(1Tgt?3dXraMl)m9IGO8T#N7ZAj0IET3}b?<T>wO2
zk<a@w|M|Uts}7%2p%G|OD-MFgJ=dE(zjdtm&Jhhp8dD>oX!Q<pUzh^n7u%G>0@Js1
z)9sY$F7gv!yNQpTaR8MRe$&;xP}$7HXG|mg=!;JL`QI_y<HuW6Wk2}ZPxyKou4KF-
zYm<XqK#5wnSrp!wU3Q_P`{{$UB{!{NG<uCt2?M1=$7HTY{;fB~AXk#mR}7Kl=r8a{
z4&Ni_HzCkKotm+3$zQnIwWvM`yy8y92SuhAS`RF6!Zx`O?6zdv#Gc77r!*t;eK0jf
z%Ni)+ed@m+v!ETQ`V4H7Wy0God+XKdSVw)F3;p$6P^TJ3dZkl@wLJJ!jF)fK<c%bO
z`pev_Tmi-rfK5i=ulDcHrzlW%Z4^06Xu~%3*qc*19XIuAE_(&7zSaR{*XVP48qv4{
zZ#&c7hlj&}XozL`5^o%_-Bu{Svh*O>HQ;J~i6I7Z(h~EQB!ax6cDF4Xbly-V<$S10
zJK6htI+W_9=;hCm)$Ply=5Q6$!><yrj63<&TAk}OTpuzh@V`}6%CgzD87g52W?z+D
zN^j^;KN1nw`*cBM1H$M`#<EoaulhY$EtKq>i6s!@g}yZCXT4}K>+6!NF70SRH2b}5
zzZxpNxPK<O7xE_9+p)Gl8oD&80+m=17UJQMyDcW>3~$;i;!5=&wCA*_xR3O|k749|
zj~@Q=cxUUr%p>seyC8DV2D$&ArvI`$_0JB~|M=4R4X5?;dyR-o$emqBXA&vKH(ths
zwne~Glx~Z(kkI7e<tschot$}Z#-iolBr9maJXN#xe>EbNVLmwynDZKKQ{Fl06+^>T
z*6(qG+|$D<{G#?s0U2KCj_Brr)L)yU{J3E5m^om|6?-lUNQSDdoY>O?{myH!WuC(u
z(|>QT|N8j9=qmsHUr!W*s@P^jGWYlxUwaTb{fQK%YF${qhtDVKl+e1CYT^|;x0J_>
zad_>HV2mK}!u@p}gieHW#~d{->D*D|nt8rm(!TRe!|oLk`A2HpARpDYouSw!sE9M;
zznlCi#-NSvJ3BINi+0Iuz!~LN!=x5IP-(vY$rUh4nof0>!}w~iRQ|_?>wgT>Os3bU
ze)jnvQ^2Gm$AN3Z_2{GBc3S&MNkL!J;+vh?^~{$<X2(ghN+9`BP0^ne0Y}jm1Y}=;
zx87=s%-X(QL`F=%V{;f@|L8ycZgyzxMDNbSBGbDSKn3{jR0gbo8UDvT>Dm1|wjdsm
zntz^1FS-u_qDmi~kF%Okww0H_et_t*k{2LEKmgQ;CNn^8^1Xepa>pdYrwInAkTQP1
zD=qHZ0xV~fk4wO!|1Uc}Un9dNCnG+=&1Ll4ZqiL$;0!y(C{|hhFDt?S>zxAbDZd#8
zz*dI5wYWv4{i*}#5*{+hy1(d1wNXs30l9*RB&|pNm&VY}hYx@syeQ&5Yeywvf%6CY
zu>HXQ0Ken*cMW#G=KR~F8lZfWX+^T~(JUN8{(4%Lw+<$O6bg!rkVG<C^(V46)h@r~
z^Rx1~UWLef-jJaWHj?ke4f^<$Hkb0>JL>=O4*g&MX_-sXIe*WjioMV@H-=s2q91?Y
z_&cCybbNRg_~(of1{ym4dT>Wq_)ynN+cOxj4_uQ<e(=9eZyd>!0%F2=*pN-k`KP%T
zm4ww-#446QNM3`NKcqyK_5Eju|9}1He>>m*_xI94z|GJCCCdEQ{p|nO&G=v58=?hk
zfwCLPj!ym+YWM`ffj85Egn*{;*|27N%YcRD+fRt3tR}A+U>A-GrX}TjcXRV_d@LU3
zZC-t{zmMhu`DiYmEB*hu(1rv!VrZ`gk_$Jw{^J6qt6)f9^lOLX^L~c50#zCT5PIGl
z$=3h^(QtlyEc0h_<<9T#D?o9a5np^uh8B(*4LJQ6ij+n&65|58W%h<Is~WMUDR<%l
zH$HB^vzDd%1B;M9N0100m%rMaD`w{d;;H-3_2{3lJF%^aXK!Hp0FhPhP7Y~zp7Sg3
zg{mu1oNl5A-+l1cAn>q4-vj0%kDLHy^icOd<MhE7+b=WArj|D8f&oTj7+_K(_>G$9
zuL!<-@caXx%k^#S*~n4lu`m9|-}OJ%sIM_7ae%9B3dLOO|7S#O*hqT5c-Bt!3b1Jp
z4u+8lF<g#CV8v7eE$|#m{HK8R;v_9R09U|g&wed0ww`Wd+?Qm2-NP3^x(IX`;ddrG
z3&16_%JT;ddu{Y_0UJtJpB<AW+B?l)u9iP%e{L<rZqHY{DNUT}HW|->D|-AI@FyTZ
zT@1SfJZwE}tRVypPfnJNlOP>8^BwmC02#Chi<|9dn&vy#+@5D+3dDpme=bIOs!K2K
zmi@opwVQQYLG*)5)<l3YjAfAJopuu6x>MZnJUX!?i8S!x`1|PP$9z6r_+Pt$)tLb?
zAQ$A{!LRfFy%Ev8%^Xg^8I&@56?=p;^vvR+w4M9$H{d|Bx0!-mcR_}RvCB?UL@qu9
z$D1bB<)^@_Pql7q1BVsb{tui0YqKYhe|?Sd(szO2f1Hy1{{KlGwPSZNC(k9IUGsNF
z?0oR96QEMvOXW4A1s<flw~(7nm0xE82WMZI8IB?!b3++^$sMd`xe!f?Ao5&$X@2dw
z1hypnfmVAG`Z!&ErD`V$0FD^!$OmxhQGgdzJ<gkcCIfV~;lLk1rYc?$!}v#P-!<NZ
zALfuX)^XL+@o>fd4WuUVE`|62-D4K7@S_#s1d-!z$ZhSouNkCtPwq#^&|`;0;#So=
zZ@*uV+g}i@ekifb&6ajI<YpX#iBrJ23-Fxpxc}N^(I_9l8#3$DBh?p`Zk;ce0L0{F
z*pD#u`5J|jk*jlHi#cIkF2%;k(#dE0NgnP0A?&TAqHNo?@1a4uL277_?vA0myyz|g
zDW!+*7KV_JZUt#XI+PefS|lW-Luvq#{?57X`+2_iUF%!xTWkDN98b<;Kla%6-}bOy
z^f2Nfchx?LAMjc`M0-&x)Lk0&{QHk&-NlQd;;<N>K)^_~<<Gq2{XZ#>3d_Orcb$Nj
zHri@Xs^r1@)Aq>_Ug<;A$)}I~hI<Ky?I9l(27jCWe|EY5-*-L&sshB`a|_S|^nXHv
z4GDVB<NZk-(&q}PR{I5X;)IyJ3wHV7sz}qxzq?@%yQd4ovVQ5hEulL!Yky;b<+cqU
zz@pBTPIoa0=@S=T-1LkD#2C0UP$(EoaVvuDSs!{?V@?W^dpJvtdwe)~Y!%-jt*N~3
z!~NcSH`a@nj`Xmm`EUq4>Q{dY)~*ZsFM$hI`1ErQ8p|;9@1pRT*La0DV3p8B`aJOr
z_)0VG&44~jc6xUO{G$2SU)ON|481*y;0a@Vr@-PA^c#S9Rs7cPod-pj?z=u^Y5%Nx
zHg&zlF1IMX)(zZF`#h`dDeqlnfxDeA#=UiM2q&Bs<MXr6sU+fe>HohO*>ec+(4I$p
zXa3)(TN%|?j=6)3jlY4uowougUq|Or%!nPt`}>HjbN)@`KzrR<-u$iQ`B@2J2uIhx
zwx?iyA*|J?;<Vbh``J3RmmxbS>n$`da8PvlBw$#IyCWM|wy_1!6>lB-n>+z|S8;sb
z=id9H)66Pq$9_@}0OK*Z{up0W^nY&+`zV4Y_qg9#+P@L`U2pDoe)WfY%bsF}Ka6$V
z+7F#JLY=q|O%_)ZKq+pjTc7RHMqosQr$*tH=18X@uL?}ywn{;Pd;aeaZH`9%$8<vu
zp8uMB_&xdMs^_Y^;pU)*uqj-u1%P~t-W?eCHo3a7ckC7hEwz^Z{OY|9c>4IBziS1k
zYtcY6O6Csubw8}%Uql@p>ox4y0<;1=-Jqpegr<GirVPw8U8hwu1WC?We(nJ<H(Oa9
z=}=(O!9ZwyvUay4_SGN2VVp)W+<}o-4Ic8Q9;YF;svT|Z-;zQ)Pl3=#My~$)(F4ss
z@#iMMR`yqbOPQ%t<)>v1(%nEg<+8pWwHdQ$rFD>+LVAqdZOT2Ay(KhhSlI<;J1<vz
z4oJSu7iV&rPo|$(1FNR7tOF@>*BNp+NPvr~l~av**DPzjL+_<|FL~kpZzsa<?H44=
zlACFQ7zaOu3eS>#%mUpsUfJQQc9n1cVkCsdyX)q97UZqz_Ac682^C!@b2dDj5<X5L
z@Op?Q-&><2_mbx>*-Bhw6JLulg8XVpZBBtFbdnYQj|!1}4-n{M-f|h6bu-Di*0onc
zJIw$JVA6;Pp~;7%Q@7VrTGBVa+_^B8$d9-%GJy$|S5iUfIvPX6P4JRSRmHNRYS$u<
zX_LdA*$R%My4Cy10G|7mwfmKhVW%6byg+5ssofc|DMEmbHTm(I{VMM%n<h=4;x_U^
zc+UXfhR(l`0`R0)0QQpL#_IsRaGk$Tn~SBs?ZVfs|0bGUx?jyXle6va+uGf?;?$}l
z#@a!7nclM@np)nOH6U6BAg{BU<qIc6Tcew&H!fY0q1R*9qV*+PD4xMvDBM)cA5PGR
z5J(2>P(pRENV^VAMeR&|@7WtqJmd)MTc?8nf%M{xXUbtzXC`vyK5;v$%3CyD#i%&t
zA4zQg1VQYjemwYzyTE;DqbZ=nN|DXsuB^WDq^e*VOBx6-gn-3DnY*U06mqRxRrY&D
zh!5S@#019sOaZSQ;7%lU#k$Z=rVE(iZ9Tuc1<a1Sp=~O>Y+x=X%E#YMa=5@P7DfZJ
zH8yt_IplLTA8ZT2;^t3DBam*NsWQOo^8#)}E^V|E=)XbK#)#<ev41bk!%h8TApe@?
zejsowdxqFZ)=OFWEdsJ>M0mb}++II2YeC+}I|2-od&ta*L4QJ%wi%Xcs`j-j{Dk3%
zE}||ZjA2vvB-bpOxKwpG0d!-D3hdJdZGK6zzO&1~UoiX<#oUlXv=>C%1%y~kNsm{O
zr+sL{OU@Leg@u$r^v+>GRy=ZTx5ocd@kAJ5Z(;7_|AY2{P;na>T9YUZ27neRO^U=$
zPG1&`QyPHb{Sv8%YX7Z&yy0WMwq1|OiP2BrC51jQK*@LnjD?tKgMK$Mv*aQ}&oAUR
zNu0-4bo(BNEZXYc+@2-q1~XX%pAEhMhamxHTe#S6u#wiKmt4%GHdODIP;+>;9}emt
zMrQ~Tu%ez_+a3Nh(I0n5sm?)VpX(Mu{7XMI@Si@?c_b-rcYh7ZN-L*GUM<JvUcIdI
zuULLACCUgwnv(qjD5KawviL}xBEEsmm4Wopzz$f)P6OEk+vO*2Ca@n(pj@oT??^n{
z->SfYTOu}rz+T`m(J|tltsYukb;x%!$ycL#_ZZ^^(0#=&IvrT|2CipH1{z6EjaQL-
zs08J?^7lJpzFjXh=Pn@F1TYhD9i%Wq7z)gB7H{WxPFadz1|PU!AU1sMb3mStgVLZ`
z08XVkUwD}>iGc|?@wtlwM9btYwAc&$kBDtq@$TP_e7Jmp(@C*#nM{;Rm~3<hF#FoY
zbKFN-r2}$Xr=zErC_h~?IOX$8YSZYkDTt!>NJmlk3+CWx`U!p^Kz&Dz6YbLrq=39h
zeh>>+vwCo!-iu7S5D|C!NH@#kB8M8=C5(Hld(KfVxC6{Z53hOq)cRrD+RR$Z`g)V$
zQCu3x@(8Q)y+#-OjIWNWgIdn7fYIa*b$aUv0}Jhv<UmbXs;uN4kb8Vc^HCFW2)L=2
zmftI}g#3xf$&z-R*`0ozUbC0yPGR6oG2L_3n~S;yyZ~KTgs@MNuUd3ZZL}&0%Zo6u
zj)f)^RSubHeWoDuS$%jUOaqG8nS-F%;LmOXkib?~XnDXAEpU|6Ct=!Dnv9MH^qCkT
zjF%a*Z*OuB%mupLhd1n`Ehib<?gy>!Vu&5~<4B<y4-C{VfxM#m?$teD_(pqek^N`d
z5pRrCL8_P=Tq_@4fv-oOiMC0=;)k7kfsf?zIsGj0Aa&Nfz(-mlroBwa?uqLmHelH_
zo7Pid7|B7qCe%vv-S-A$O7c$z&BZp~|7VK(f;<A*N}z>qrOgwjt59fzpEK#|cEB(R
z=f=^V+n_Mq^Vx4ylOgTTGKz+`sSy}q^lYl@|IPWE&!|EM6!6{8`%!CoS{IlaPAgrV
z^J44PF0UvVP%UpCWQU4m^^^(YrBf$D1Tc40_qL9EA0I$FfQgE3F#{E4dMLjH71dSO
znm@W1x=C=#eWiNe@>XS@$I{kYqsc!q1RLVN=AP?{SaSXWte37TvR&Uo!bX4+#t7M|
zZ_^-acAEK5L1WVhhT)A9Igp?b&CsX}`}f|i?5wbFlcO<P6(Yj>)4oHsraaqzVKR%C
zaNnKA$+-=^#1<e5<v-Y<ORN>keQzWF4UJSJ6OB{8G$Q^SD0svyt7S;URwMa4dN#=G
z`A>qV3<R<icp2XWWZ4C2z-0gfpdGc~k8mb;D!nuDm?qLq;SA2?Ng@J0)bj#qoA_aG
z+-PA0C=DG@lM$kq6QeyytzA#_jjE`jmb)yK;GhucS|OKyq5T<@m@nbH$Ytbjdicax
z%pxYXb1EO6;{~vP`5Os|430!2)jt58M4_B);=mBB$7JtyiHJCZ;P2diXxww2qpOE}
zr6F+vFmdvwJ>O$0(b>Bqx9NGG(^ZXoqlGPs3A~6vq2E0v!E(V!M^geP-zFrhg!Fq8
zwr$|M|97vK2-G=;gUEp^4KsvwrLj8S!aAJeU<{1%^3P&eu-hX85rtP>!|@%QE14)q
zr}rmey|!=Rq5HDCBQ;Dzy?|cd0}fJ?DM^j$gRCh@#aQT5yic9$R!6<3h_*wh_Z_ki
z5Fr|f6lVjK?<-|D8&gB&t~yNKhTVV;&>g^P1wehAmFCu;;#D3DUV*6mJ6#UOYs?Ep
zISJg9#xn}}d?2Z0iCu{V2`XMIx|O|Ar07N#$9W%ZhM6gq<Zqa7D#*Yh3Zo^SitXA@
zXa7PN6hhlo!_-!InF%NTtuvM|Rmjy^FDfpc2Y}0Tt3JAvwHV_aXOb!jOA9F|95AoD
z<OqN&c#=@5n9r(>@L+Ywg?FIO9*}x2J6guQs$PHjHkGC5mjGoGWfOh~TL=e>K#$@|
zWXp(&DW2ufhMk4w1f#{Q|7@?25wrEo4{(|OY5>j)6f9MQYA07L?@`$pRaO;RDl-v?
zs@{cx=${X_7vuq?C2yvxSWz{eVj)gu|9t3#<qIT3v{bW-C!WVrp>UeV3&5oI!w{gw
z8$e|*bay(GJl=N;--XHSb1_b(Q(!0jXWX2lj1Ifqr^$#YYW)4Xd9wAv=jYwCZL9)?
zIw_kp4BTqA?{wt>6bRzQH`jcdPEX*t>to`1LyX5q|5LD~UcW{~M<Vw!RH<*1Zi;ZM
zigHz#w$R!)F!%x@czT{4v@AJ%)PA}YJiKOl2Y|zFZ>hp=<a056_pnUBT~uf{0u<V1
zx3)T|^yOMu*5gtgj6(u9V8Rsn=jc>5xv9@ja$JQx>v6|B8~wR0J6et#t&T5c2%4x0
z28Z|b_I1S>ve&ExH`0LqPG9w#aJJ(AV~lPS0S4Cx>8EREVx|90OAQ`)P%>$9H5d>Z
z1zOpE3h}!1(G%e9Mxq28L^$lBH?u`4OmFe)4DX3-xnClTe1rBnjL1}=h6zlTN@;p3
z8`;FtiRhHu>Kq!x^xL)`d-py}O*SlV-f>iFMwNAuC{h0^!?uWW9b{k{P$p(7#akRS
zkMgEwE$yOE=2XP|QF$NbRgtW$mbdqO$!wx98!?DrVqKgHTIWgEA{%+V7br>GFJKUg
zm?w9H>11Q6j&1@Ho1}O7Drt_T*i}(M`*E$HWI<TfbE(7}D!D(6smh~F7{oHAU295E
zdd=*YsoDpIrb<j^RHA}^vsZY#*>t7<OrT)V5;+!A2gbI_FL5%s3rPCZA3M~YPXj0}
z$^+bMvUc7?XDNeYCQ&6~mohbSeU^9*g#m1Ire)={29>&9O6Mphf$i+r&@p+cZ4~=G
zvt?3uC+h7F0|@0+C5t2`{LgPZR@Z<Vc5k?a+%@ydhu2HW<w;B#i{fcR=r_1&d8P4O
z`sGKEe1jC4A*NS^b(fcl4(A+AR%HD_<*DT-8O947k!2y)1AA$cqXBf=)RQ;TX&a?v
zp3IYQJ>s>CN-<?(+_KC&=BqK@abN2)@~`uM9DXMmu&8hz#ig^R2I|Ldawrtr3VaLX
zC8-TjME_hOtmP$Gi^$np9&`e5fo<u*Y)!%FNpm_g{v4#x_2fzDg!fEEi9Lh;nO-!H
z?)xz&CM9TC3gTFpBq>>X{>n8XVI+l_Fw+9y0(PX>BDft94EF`_E#^(iOPwLi3>SC}
z{FmZ~7r&j7VR3zo67pO2oxHkBQ9V(u%2IO*nI+;gIF}kKjXsj%>6kZMJ!KdeQHMZ-
z8OVGiOjV0rimbAVDh1t4X>rfqQK$n@ExNmeJ*Q2Opi3y7>71evI#uU%nX`aPGo{n-
zeL%R)9xFYqLcuo{nkSq&xcP%Ng6Y3^quo0;i638zR(%XIf<0x6rWGbN=z;|-_cW%)
z0$!mnfx&|LBeHuz(-~V{dW)VrM}~h&MAzzroXj@Y0ox1DP2XJlCSDn!>5u$oqS))E
zU;n$XkDQ}$1M1lbYK81p(@%2P!5nu>Kx5)B{OfXjL<<g!HkpK11^u=m>!_6GHama5
z>$3XT4;E5sybn|+*P7@e&@w@0zwfGBC~S7`vz^`pY&@zQm;g=;k&%iZx9r*vs5xqJ
z7o7_9*E8+dE_|xA#sJcf;FI4#!&$%PIPo*qt~6FZDJqd~0~L51Se11MC*4N5b_Et%
zsSL17GArw{2KAFO&e<-jf8UEfQ5dB`a1M+#ffz~bs-;n~fEQo%$F<s4f~YT(BM24j
zpsm(th%!S=6&2SNkt{9hiY|Xc4)^4E^{J^{jCn+DXm?KmxLf3^+*grkz;wKS(SXUy
zkpdwcTWeh>C{9b){|6bcS^7~K`t4Z=GDS)}0az!DoARh{vHJQ=^eNJ$@t@)tnY+A!
z9gY?j9rj?NO8pB0)J(BdbWwjKplH?r<r_i&i8Syo<3k1GV!e-ZjOZBqYk(K>W)Q&T
z6e?ypL9o3ly8asW=&bQgLWwuNu#~vm?n%Z@U%~BvVK>$T7=t9yMdd<)IFD&RXEEg{
zidgJAMX!k{X&V*nv`y5%3R&6N%B~sQ7hqY6b>KY!n(Q_u;Hef}zA@wH)UDhpgRg1{
zxhc^I3gbMmecsx3hODglp|hzLyROMuhrB3CGGJGA4J~v7nlVLWH^!7Ss2O{2M$YZt
z=K-Bvw@*LLH?hY>QF|~NgzNq{&_={W$}icaxVtD1H8?bfP?--jkQ18cYCnDDGU4|T
z;XvUBq~zeH$@2OX>ETupM78j=sI1#qe5ZHoD+kT?>b{WHKV&EkKDQAqbOV6jg@-x8
zD;=;-tHQvi4(a1L57oX_MK{u~M*eQ+Sy-)+j;<E1LPz#oZ>5hr*0fteZ1mLC?zlRw
zBB2lGMb=M6mKB?Q>6U$VAXepg_O1Py#-(@ja8KfmNO)fBC%omA?T91L>%mE*a3*a`
zs2MKPL=H{ZXkM-H;yZ-9XOOu#u)pbcDAjsIOZjfDrm5|nTS4Bc)7TWy9M1`Q@Tm=O
z(9~Q4U)rbh|F{N*um+*dL2ldqqNR@J#g)4YlDV~V3kILswE4_mJl}OrGPI@5a%+aq
z3(S3xMK#yd>j^fd&y&0i3MsGBqy?k}8J^7;!poA)7#8#k583VtE<ZS(Z(Nsi5~4}B
z)cGurgUjo)uNIKq<sTlyIw7{V9<3@)T!+$XzbLl4U8jt<uM&GdOu99WW%+Kl?&ci!
z#0r>Oy{eT)@N6bRZEopXxhPRyIlH!Rcw0^^Hs$ebD4(|S#O4i%)FPbIV=vsYgNcCZ
zVxcErRG)$hO&P(Erg`%;X{={)>z(v&u5`cAm7AqGrP56Ekt_S+PUpWbW8-%bFD!HF
z=9m!6;x|I`jkSEs4x}(gSmXA+D3Bnm`|@Npx({wIF0`I&i@~tq%*!O?XgQ<IO>d;!
zO1PV<-pIk_;ni*-0~CjDxcmPZmJrzxaM$mn@h`|);Lc<2?qiu(t^nhh{W%A-9ZNTV
zD?E4)(6!MlepCpB#oaaxJ#~#nPBGx;iQi@uPyQ(oJDMHa$kn87{e9c)1-#+QTQRRw
zU*`u#1gFwt&^<x$2BtPi_Rsse&^Q50*+)#%pNNoZm09*iau!LPA^o8_prOg8$3Bgw
z#rzXc971p@VoP7#&D+J6xDJvclJfiVm7~jx3z*%GCxg#{Y&=Cqr!#~7SHuKQs$z6#
z2JwQZ5aD5cZ_kMS{?j$0eYt|A*8pn|$IMyGr`BE8dDbl7xkKb45Kh*zZwI^b8Pi<J
zFL8hR*T_l?7LnEs3)vgyVBkr|Mm7z<BG)O0+-WAxk=_^GNk7YDghVIh#UQ;ATr?KG
z|HA?}Ho$e~PFlw<3_<rnM>h4q9vkIWNv{OSWV}LoPsqolWwcgGL-^&ux@KUfeRjDP
zhMP6JVs6e^!1vZB8#U0{C*#BFbY0G&HdlBvyaL5E%Ma2seFS4WLbVk{ut;`GhX7?*
z`){FhguIwm$h-#_>WA6u;%`PIBhGg_U8$2fID7f@UPzK|3*|t@H~)`%h-#<_0O5(k
zHm#5h7$V@|pwmG0LkC2Aht>fluDzPHA%8?#*l%QZ3g+A|Sb;xJlmG1)UOTuHyO%NO
zLU7m6!q9nG=-}pXvlf4eS8Az$Lfy>kZw8uNE7_)L&hZ^c57;^qr7sNWEr<Fu>3Pgr
zMHPD0yUju_zt@K7%Te(GHl2~h$|Y~jL7$PbcR{+2Te9~-aO-wn^bpm$No984VI~M}
z`K&8aC>fy~3n&UGO(O#R{yUAN5JPpue37Y6#lh$#(4H?0C{RhspniE5pJsO7ad_qE
z3BAs6YMg1CQD)oH>R@!uhwkL=H__#)b??fYG>+gtF>$|ko)nDuCBxqzps<=FZ-_y*
z3Yf@5QG4lEvc|BC(vx{`r1AkiLqhXLgVR@+t>}GRb!|+)fjLLTh{V|d>6->47LXNS
zRt2Fy`6$itwIf^sow*<A^++oOD^kaj=$T_+VVlD|_^ZocUBJjHvUNj1`AJTxLpoMp
zhc5oH$Oq%7>?Q_xvS=cZ{u2DFF2(HW<m>WJzpA1*gV<T&6l>dEF0v%hQtbk@HPw+6
zl6r!;3+`<ezH~(v{$uS@ICY%k8X)J`8GQA7N%ox$KQ6kwdp5Ni){w*yth(<nNYoKs
z9?FHk^zJ8Ic2n&H@zSOIC0T@H?A6_^_A<wb-@YrqN2yY@k1-|KCcX7P@Y@6a4;^BZ
z=zSHM*_iI<Vocc=A{55k?lXr!Ed2{OzYCa@;T>}cyE9l0H@p#B71t}=T;jP70{m#v
zyjV1DSkV3`g9g-i_##!J!~cQwMwgI9J!8SyVjOY-hW3WD()6k}ufZTFb#C1C$FBQ#
zujvs1`W_|q(I4^qvAqN)?a00*EK<dq?hV<2;%TL$errFXwW}0HYNz%weVM~bQEUX#
zGKw{e(FVu0qPMOgHjBc1MDnI>gI$G<*k7=GaSL#KQSY&aTzt@>uqzju=BfGya9oT8
z`T=K0d0C3&Gk9~|cM*6re3NL!RIeGRX?OH5|CrMU#9cgT!Pxn^sdhPj$vB<{<Z%jG
zvqo9-ylPVDTw7@I{Clz{0U2?CR0rGzr~$;rorA`6`_H!d!rEmfo8hb64&8uBh!*>x
zgdm8823kH@hU<RLnEHo8?uKs^fEA!O+l9NuN~1CsKWouOJKOf%1hnQ2f=K6$<iDV-
zazX@Q6Kj&^e%vVPWt?pCphYHH7?JvZy@}7zhUSj$ocV?g%F7IlDaSRfAMYQ5uTh-V
z)uwU>g;wr5OYKp+a1n7$^>%9l|7iHco596B$`4#~_&1hyQ5fe7NWTT<Et3t$Mk3!t
zKb2g4YlGr%2337geyYK+fJ$;IDXOHYTm_;MX*C)ak_pBjgF;7#p=x2@qubr4$>`te
ziw1GIG*i7(S8p+H&@RXfOWX_IC)OYNQg{GH+=HUJrB+Q-*@0U0BQz{_7vTVIa*4w|
zUsA1@9&w9n{a4|r)VA<w>4ZepIt2fSL~s|`!-VSoAzEU;E?JrWyoDfJ(?yYZ1Er4$
zZ0w~K)Wd3c&0k*yz@r3*8uoJh@IN%1OjTH2#yB!SpUu)J>Qzi^okN%8?wyFodPd8}
z#IXs)JNk5#gKWlC>niIVjwf>z=zirsL$xu#PWBYC$_JW~^K34&aO;JvJ@4l>7Ag#b
zf0b8kUxQjj<NqzLZC*32qxTJ())jL7XG|EOoG4F?0{ggHPTs1ebC?;ZthSQ`Qj;__
zjZ^r@%YO~Y_31;wN+trGwMP1LlFfd6Y4@E8stSoGiS#$E7&-VQi3(mD#sw=8lQ9Oo
zVdTg8GDf#0qUP#Ny^{`tA82F4IQrOGQw1f)$<u7UMpM}BB5hM*-MStH_6NAAtk84N
z%cNBq-*`pi+_2cnl0{ppNf<)k9R~Zb>cR-g!af?QLSw#kJS(1y3{QHE<!D1cN+xj~
z5-Nq#5D(wlwLJX3W&Nabs?d`#{G-v}l%A>^t8T#aveakHAiMl>>r#(Y^Zh671d|O^
zC88@nvu15eRLmV0BCrR2(Xy7h^Xk&0Tx|)N8uUH@KkRO6_%mM@)yCTXwemb$Pjb^c
zfglT-Jn=7&K@I;V84zJG!B(F-8p|J98IS_UoS*HBy}@vFPU1yMG}0sD8AeQyw3rmi
zs~H0XFpx7uL!r=c$y3bBn~?*bFs72>gv0oAbnRcou48?02K-WJ5%Io@j@i@@G!jIT
z)ddqXw(Ug~;SDf`aP_vti@bOaa~*=7mY~1;v+r0Tnl0$Ic;&Qh#$Y1)4R?SkzAO{P
zs_4`wF70UnPC2#@6AU~RB~PS{B0*a~iV!I$K49@;b4c({U{R!<Q>XAOF}6vGm&Nd+
z!+lmoY|;ShA|a8()#Ps|Tl;_;(7+xB#|aQs+>O3^1-GF3M*0g&@Tq&;9}FE%8)>H!
zBKl$^@S1sAfg&m=oIYBf7XEzw6>vSg`wXE_XTqJLIG++_7&lbIZBw)eCo<VVfnAGk
zV?@;aHv*tpGTey6p#j|r$NlD<R2%BlGdLYpcHw&wyeK5JTiB~H{OZ;i*NKu!KLd0X
z9*3o$G`-Xz;2c`q%0F4}PQTGEDE)oLZ}JldoiBnLro=E}XUwm<n}%BMN^#o&qV(dt
zQTDO1(KY>5k<hhHo`~_Y{l_b8KJK9i;ZXv5<@yb7gUVGVRf9(`7_yOLRNQqMi$4L&
z0Vkch`=Ao_gY&t~TgTPxGb&4ff+IVG&0BF4{Fd{wv~5#)RP2)c-qcj><nsqo$Fon}
zDl0MTs@C`@<aCmkBr6iQtw`sknz6BLKCu<u*Tk}G%5jlDBObEDd*~3tM{xm%3O_=#
z+iv$<omJsdEv!rY?zFlMaJdV^7S97&fD)Q6coGtDP+mZ?ipcs*WlHr$++=EacCD1z
zG1&Hu)BB68{}8a3e`p90NK#}Ka#wR=_%6KLH?eQ@hREFTp?fvb9m!EG%tM4~vlq88
zvJbc)CRrHDi+aesL;^uuNzyHEC*z|;weF=Z=bpdmOVU!fpXhFBYKV|6XvaX{ZVI@3
z+s>1gc~}MtZVD9Cr`*#2D@jV}Lxsp=MsS=lYY<OM3>p-gSiE!aNzo+T;c)z-bOtZ8
z&a@NJfa=|Lk-f^2t@WY{)EUJHvgT?mgR3iEAxMdFOUYcXD}c}zs-%5f1`uCs63Q~y
z^hcLD&@?g;O&0~#NjB=i27irY1pjofb!2&Rstw7Gp&2kOog45Q%kp8^&9OkYddOz<
zHHtV`-KvC18#8J3N7Fbe+%wZ95Zct}8(HQa)D`h5W(98hbJeqHKL1!c8`O_(z|O{d
z)C+Qj<NtYDVNn-i!4kni821Sv*6&I(SRN>8d?s7fj~(KOG3*fbDlbg>=|c6qXR!F~
zSAC`AjqU#SOhzNV!-mYJ@uki6iWE7u=FEBhtq&;z-yRQ<#StwWPEtWj2URZID}{F}
z{^-wnt-*3sa%ic$N9q{R(4(@7rBRk4xO5}b!-w$`E@Y5sy41pR3Gco`h1&d1gYfGJ
zs!aG=GnYNjM3EaM%3}}O`y3C`AC(^Cb9Xm`(8v6y5mXR;7-$(Sd#!MxkR5#5t-d8x
z`@r{B^M*EQc}u378O_TfID=Ti@?d+R1vwN|5|1}?WN^%Jc^jIBRjZ>3k-g|4yYaKc
zTq-#f{H@CQH?A%r-hPwj1dRI@!$Iff8RxGtk6RJo1~7>shS@INVC@)kg0UF0$Nm=b
zF6vv=lqSdPm)DjYGBK&9y(^_?&OI*XkYW@{xTCxRyaqW8R5E{g@7-5tRE7kaN_%sD
z%?b%Ks=JJE%<BT57oT%r9w<FMR9bkpn&J{Xr$BaFKu`2~?0_qfp8C^~U{h6VOtPR~
z)o^8~n^}F%l@Nhz|Gr+R8b=@)8sBuwWcId_?pk~6JqL*ycWeCx>Jd>vW*14NrhTgR
z`LtmykEY(NFw3HQ?Wg&~y`Zf#PY$y*d@bYXloIavE|+uZv{5^KLKTZSP2!gyrAK!H
z9ACG#vnfQ6eDRg!N{cL|U{K}Br@oHH)~_@t`@l4s6kM*o4Pi;+Ahez5RZ4j?S7Mjr
zprc>0#x?YPJHI6n-T2us6K>MFoxN6ur*N(DVB2oZ<nPjqgEAem*Uo`Dw;6GAv)+jg
zlBK&8oV+g?1IbK960w~EVd_28mC_x5IxTOXUU1BPeXb+Hns_0buH7&-A<AaV$Y6MR
zw6DCyK7`URYv-TwdFE%Tf3EIJh)ZDCXM2XrYCv%8h}kVbWrwRCw&2u2BZ422rJC`O
zC~GQZl3O4z#e&aD169NIFM~h+!Shm0e;Z3brw(KR0qO3zx#b#hoFq1xpVv83EA-J3
z8@u;NkQ@ayYQx}vw88%>6M+M6L*h79Zko%7T5Lq|bU!)4pPUngxuq3b0K+;B1ebwI
z3r3n6T76P>`1vBwT?7Bf@uYY!s-@D<6yTK~42q|K4U(~#FqNokTN1alYdRBCwR~i=
z(dE^6%PAdWjH2f&Q=O2CQsZaxEi>GsWkz$=w;c5#G*-z(a<y2qGw0EAKp-(6%HF%^
z@TH(PJM^=mngj~JO6GZs5WIm(CY^-K7WV@Jm>3Fge=>3#5=^f!qE&&)JY&r+)yra6
zt0DD8m&X&F(><{MA^&t6$~ejl$*jr<06G)?Z@jG5PJmj0C6od?yqwZbY(VUQz?XwC
zvtF843Ci_Bm|*(RCZ}{lFzcTfq^;Wl-TfwuZQbots3mZiJT_UpAOgh>v6*kY(t4{K
zTmM?+d6Vr1qy5f{1u}(MpVMIAU_P)QOv!=W=f<3t7W_HV%p(LgSz6gaWLINOWM}f7
zWpKUAt=dq$C@#%kQB2e1BO>CVv~B^xGUrJ}CUGACDAj<v`Rcm@NxUSi+(k+UG(LTW
z#DIfPsW5Z%1g&);t9Kg!>)+<yhu5FE#$K8S^M-22W-~qnPNtc~oj_#7hNFFZ(ImcW
z5X*D>@X_KBZLAJj!plf1WW|KV&nqap?aLR!{3aFFg6PYK`EN;Z_PXIdNkqRc!3vah
z0Y=cJ{FBkN&ndu>^08@RZB5I5Zw6nrd}k66>jo3Li2A!IfB=$%DHAAt?U?}zi<(23
zUHE(X!tHC%Xn1~MjM{Ld{lTIzPWUD^*~K@+Z^{o+%J<m7<B?~6rchV+1bz<><?ety
zn&4%ObM8cfa;k9e81c`uvGwEV1K+0&pYIOcdzb-8>~F?Ji!lI_uz+J@>cV1`O<OgI
zR-Zhlx$76cXM?EGT)%Pm?7xS}rEsCLzT~)+D}DEh!!yiqu}M-VjM>v$bhtiPPYj%I
zD;JJ!Es{0Yz4pfnI7GNv?z5JE=mkGYf6w*6l_Cn`c5v4O{jQN_JF173yf@yX8lx~S
zmu6<J!00eU@4Mh0W<!_;@}(^>e#^%AEjs6leSmBz+MF>XlV<CoLdE|{n!`aFOh$1E
zGYb1LrPwhJH<}tbuT8hxF9=A{Hqy;+`ak*9YvsP(h+W7k)KtZ*<-nPNN@fkBxu6N+
zWJr)z25f3X1PgLTs>Ma1m12!jY-rXv<)Z(3i$fR=tdUNxOFnVk_#zb9`i~F>)@o^h
zm{@wb=k{><UX*Y=rPKy*pgz2a@AeH7+845%bdUHHcLh>Zu*s-zZ^-%G`6Qo+&v|mI
zIKQXI93%e(npLTZdtA1xktBt&MHzOf7TBdRvFL5ABTAGMbQldooNbVgJ|HLDyTEXH
zKTsg0?lSQP@?fyCX<|5>G{88cV7VRQ?noxK<jkB@rS|R@^~PWx)03*#m(R9e8w^4M
z3k{Dq5fHQId1@!Gf0yYH5u(dKaCl><4byrjf8tT|p8NEbQ10gvhm~@Rh(vudI-Hl|
z+PjpFKD9*#GG|1szno%H@ZoE^EUE%0*LG65SteVHP#2$eX<u<l)3c1gM1i;#%CP_z
zQX#JhABsbN(_Zs|_o-WD1ErSh-<scZ45ERfNAn#N8PJAcJe(&0aL@$+4kFl4`(FT(
zJ7PYGC4$id7WeE9^#V>If(YhS>AegpC;o;lUl1CZ)jk*^K#`et_@bm9WfMTqO4T(_
zNLGSloZn!{c70lhYMJLOJ_}pob4=T9oJ4?mZy*OuYXhL%uq0X1$OUSreA|J)8=r4E
zQtk5HCjsXYzqBo(E=tN_Dunv&pq@&R+M|^l4bVt;KQCIHHImwiz~F-uoqD0tToRmo
zOGar?Ox_Vo<j#u&Hu_Qpl7)p{Fece1w*%r{RR;z{i8}#vVe{!c40ZcUfI_G3&yn+@
zE+$CU$f1Ni(Y6M4m3&?~!$v`u?9?#$`7cM}#iSFP%*~b3;H(Wq889#T7>GT^0Je9@
z>w)Z?%|nVJ;*NzG%1&%hvRZBEq_aaCh~h$tB4rT<D-5ghu(YwAW6mBQH~UlFlS=#9
zcKtxlM!MU+CVV#7-%s6T#MLHbpMbi5DIgy2+h1iFF6JtBR2$03;Jgc+8=cv}m<N?*
zzj+n8iF3IQ;R`dka=V*+Is7U70OmRd%y}3*{*5C{SjHF^>W-ZT>(mL7*@t+e8}ddw
z?M)I^i|6hDq#~vjg)SiyO0k9G^J-9@dN%td&F_ucG;bddn0+n{+1HN%QGHEl@<nh&
zS3=h~L!h;Lh&w$f7flog@a7OWzCI9ICIQ`6lPc4`D{SV%y19HU=@DPk(&z%HEH63}
zTIMV3A%qF^ME8aTS_3jWFKzmBOen#iV?eg9==Y{$_qOc^z!Z;kWB>%}@}Tn24XDU{
ziTTU_2^oN@@kI^Sp4tIwlZ+MUgE0^?S`nr<9!>6zL7LN65n4^ak~a;C9|=@$a9<Uf
zU9|xqKXv1(&HokOze0aOg+h%RWn_y*{U$1t|ChuhRA)-#9+D_x`h;m2Z=xpD`O*tU
zDb`tgq!yX!4vCEcEB~D0TA~NTGnS`BbR4kp4e+qESz?bZy+&9e4uyR|(D#X|qn+SE
zGuQ^V^I<$#R|TRHSFElF`#~+0NJiH3#vHaASv;ai1mRI5ww*2le0v})DZm6?YN69i
znY;x*rkmZs2F4x@7POf%8i9D+Mzu!b8i-?3Ll9f|_wz>dkc_)x@}Cg;NxaPxtf72o
z+*GjR-*wJWO)d_s-HeB+d0YtQKDfe$D54P|LQD|ZWoueMg!3$Zoa{Z8pc0mL=8+8b
zyQt)~h#RyWgJY~(KRl)wyuq5c#gL&AFh=VyZbDPYXjh-udvAkhSe#gqt!Yo#3j^#H
zd}g)dR(F-2eCZ3K{Zu`Efb&CSM1<S%eUs-{@x*q?lxjDL3!|F46q#rBrokyX=Ly!8
zgE+2%8D<(K<_{%Wd<2YG82UNb)z0HmQ)8lEXkgNRYa1L-OQuBf)Vw~xRHNh+5WPz3
z4G&=`c^`DE8AqwsAWDw1BQw*Yc5gnC8jNB<Bbr)}J4uj%8Zc|beZqjt_A@&hSmg53
zDVTVX@Sq@>5#mq_HmJQ6nrlQ7sSe(C)nIUD*2)^%s&|vxE-bOS%=oQDAkY|=+Q?WS
zPa43EU2l#5T&EQ=fntHdQXz((UutNScsV*P)Nh`5DtlykNUv%1#`4DN26{6qi|TGo
zj~OYQZ`P~3#;6$cJ#2&W7jL`?{obR&8d{Hy26UQl+6iD9s(Nt~xW`c{RW}M(s44F?
zi}nmsm@h1_+G&de#s!gT@%w50X+7l0<rXcklg<>l{G!M<AWwv7vZ=84nBn#DD^dQ}
z0$d^ZA+=;{5SB_XYWOGh8GMRlUqmQ{!~0LK6);CW%zLMeYpay#+DWU*5>tPlVI>JL
zY4@ra!D4zzU^}Az13{J0@HGY)5PSTRNr5|=UAtdbdD?5S@Ylg%0KKwNHYqgkK^1+s
zWbTgu1)=i7Ms%o^O4y;N(#vMAGHGhXkusyUJgo>fl{M}Sc^HcKX$2rOnH11kUXCUV
zUPnW@ju%i{b;kEzYR+8yqiX%-Zv-v3eG@Q<1=i3!8n0lV_P=siqY8qcS^nROW7r^t
zFMoG*$(qZlw<A=g<SCT@fJ28gsH4&B+3g87!Cu0z!TyOB?g;bT;1>+1VL;~9puq&@
zTM*5zkb@j6?<@hk>`{0Nkm%CzU|tfx1(&LVNVPo(<{V@G-c?J8{gjc7;*T+%J)!;V
z9B;7d9L7!?;<v|SYEbL5LN@QRYH^CMm$D6RiD-cYJ7nV}E%K8)9}r#kn{U-lMSskq
zf-*3aZm1Ir{gt#t3b5=gpST1oF+>Gtpl&ilm2?V{rdq57Dw)d!oob)J=eL+gP#q@)
z#AXyU^75QfR@T#iULU)2fBGj~sPIz@hO06cS?AsHv`<v<o7mT9x?a65PjkGdRQ3Ip
zesbKSdn+_3(T}2?OA_a66T-VbKGRE8qZBk>q^&>buKbbq-49ln*1<iQ5+zIAP36US
zuC?=P?u|5<S*S{B_p7vzh0SVWkFI$WflT$_TL|vU+DBDY1B&Vf?x5ChA`LGJBYUzM
zA-GWOUirdYp|W1Z6qU7{!`h*c8kg>8Lt+deY%cP<tbBz=s!1-@QYSS!b0PNWK^KHA
zJQ5`DL-L!HQ={o2EN4qj?Ti1ZnQor?Of!YjP|OW>gg-b8CXz#vR9tVvB-k4yO5!7Z
zr!NER>jRMxC86kt=+}@=e*98p&uA-@0?UjkC#cqvwXJ4xYTr&Ifc@ad6bvmfhG32<
zQeJbX#tFo~MgVHcwndv68RGZfJmC2yFWuB9Z>Et25*Bh*h<1Cg{offi5Wlyui<9Hl
zBN-dH<>3lv;=KHjVP}3`8xETpYdo#doR7sn3t0F+sA0}*oD&6Y)xu|@B5}&NM=|50
zi5|cNw(?GPO7PhM5va=(*r?tBgtwpSP1-vg%5wbp6fG(oj!%q@aK*@i?0d}&>tuEg
zm4iZ_AcjFP4J&MxB?=aJSrR9FtgyCIloLhIYU9gHg{Jm()`lV@mV_YnrvAFT-N@Tx
zs+><uyxn%HnKXzQhb@d(;cX~5;eWyD@*GNwsp%w&erZG*OkZIDqLVjMi?cX+KGB0&
z#YUVjR`%en^q-s!nY7NtHTWc>dc^a(i;_FN(;bvSPo0J+aD0<kjQRojq)xRs0Cdq^
zfh-F1>ZN2lJqejlPy(^^Qjx0>4ng3OFa<d2J%K|}wS0FZw|Y?I0%Y~QE=-`0LYdGv
z22uYd-Cr;IG;TVfLv^{lTk&^AZy!}6apE~3;*GW7^wgVoni`&?_CDBncNoNxzpcy@
z?qXTY7j2<UeM!t8uE)T5P29&2g?Wz>@t|y8WoDgrCBvbiUSsraDMw!zwlGHbFg-ct
z1lcO_DB1hE9{DSwgL4v3CA3bE;7oRiA~Si;^+x01>1Cn-y1it%3cPZVUo!+B-Br!;
zkt32f(PjphfM^nqrKB3akvS-wP?sFU;lfyLK)<dgwM{t4w*958kVe_t0uZp%>;71C
zNA&+qjC>o)jWI-ZK!2GYE=a|!u^I|MvkSm{zWd}-UcIS}0UR?F>@dx=dOZTgg;=8S
zJ9YXG=G>z4Y@789Cw?vrAt|;FQs23-1>v<o!YF@VUIgY#W}x!U_=&EcAzcn4et2y5
zr#ST@cVeLb*mkaOk<=*`0Uor}e3f0w3V7%cft|54O9u;zm7hS;K{2KMjA~N~5ln;T
zTfryit<5q)$H%KHGEOHed3U>4o&J;bxkBg3`OQL)(6zR{*e1v;#o|=?b=7rCA)WVU
z6Mg4zKys_zP=*S({UYdO#cyKDCQz)h$c0R_#)(s?{G*aQ`j95-3-VRUj57iciV}<P
zanEO=%~QI$A_iftr$M<rlCs`R+|7SCaZ!jAf6(m3%_LPmv-~k09VS|+5=!G(PnX?8
z+22G=pUiodztK&JBFCg(7}h0rp0BX?RU8-9mjX=<M0F>o<Af?t(a;1>zKZ-?*e8WG
z30;*iVOfGN7?d5ARpEk&a|Jw1G^N#Kc0=A>wHo1q+KtKd8m6A~i$wTChe%9LzjoS-
z<_tPLt|;C#o%~(&zD@xrU1?E(Dio!#K_AX{R8|KKy3Pf6zWha<#2N_Odh3MSOUVM#
z%C@@p)Aucvo*(KJh}E<%qB|KMQVN7-+n{`D;{ZA6)A!C2y_j|SK{{^t?n>%Afb#XG
zx8wwIS6dBAz|2BCYX*m2)v|!LETdnm?0}!*C}*j5Pul$zz#5G79$5Yq*`D=BKoXHn
z0on<X(_9luT*w`^m5v;gyqJ~#oKpqyvA+20cJr`AfkN!efJy>t)ivc{WdfrjKYAG;
z(!2eU&5Mzbw1al$PAF#t8RrEUm{`@OFuNuiMr|3CGgUt?@esO<@sAxU1O4*CsIqJh
z*V98RJQ-(r{KRnUB_{e^p|gHM>H*H5Nrxr{qadzpmR|JGpG2Ou;Q=UPK_*EfPqtoK
zQl<*nq^QN=IB}0L7QbYh;{vgGe1w-Pm+;%a;-w>s;r~F7#+6IGRZm8xW`mj_tv8fO
z=Nln6TA5D8Mn@<o%!$kPd3M(;Y(~FEcB|I71wUXQ;lih*X}SE3a3z{=T5y%3Xm>or
zI;o<}Wuo^Xop}6N<)Oz<<NXYjsE<=M?<LbHK$R!>yFQDRsfJ6Hp)<Tx>+5DST6E<F
zOl)QM%ytLKtXX?-n(cx13yceWyqD#-F&ZvvC-#xi{p6VS#LW6bGK_llP|Ut+Bd~P?
zk0AePEI3C#cRG7$<kteQ$gRt5%+^g*{z=f`kmX+X60fFSzwP<{Lqoajl-Tqmc&U_E
zfMck++|2n$58I!ROntogq;bOrb0@Wv)ZwJN0aSDa_P^(}>i__;a7CK*mpPZPIsK;%
zw;hwRYw$`-+C{VYZSse?P-2;LxA$ZS*(9#P#c|mi+8d80tPoJt@I`xkTTM6fjyr?0
z%vZZmPKKMZg$Wm|xsksUIp46hC#78Q;^*zdV6s`Ea_b>DD(I_kXEN^iw;uANK>V_3
z5~<-0?5!(iz?p1%NC?MHHbxJqR7sNI4o)hSlWqKGVuLDF3~=qNdAR!~yPNAU@I`84
z`{K*TE}|r&T9KP&j=UG4{^uXd`M&K)1$jt?@kKf^Un@35h2zxl&z1a@ElP<JpeVc|
z0X@V$>6MZq1tE^39D^vnfI&79v`J_-$)A+D(L_j|x--aFg5ZGPNP1&>JS*#Jb`#1F
zuu<Z<vSDJ34Vp!LqV&84H(D|I6XTakD+ja_ZxgJo&*IzG`}qO*jze}l2NWT`l1eCG
zJ65Igx2U)$CT^Kl;s5M+z4*guwVfNWK$4`tcs5AI-rG!EtqGz^@VVoz*6~ShZ<N8l
zF@e2_7k#MvK&D70_zUV}K?X7~rp^w>LjA$C%CNrp9ks9Yz)%Vo#PxkL@DI5u{iSRY
zb#HM|xo#cn!wU~wh{y?!DT>YNRWo-Uhn*AT5m-4r?!_Vis0oxy&GKpxbR`>UMjJvr
z3}Pab|9+_IaTW&L>}r@uo0+l}q7^<jH9&;5`dJiX@zk~2(oR>n|M7Kw=W{-UodaTP
z#1Z3@8=X=LOu+vwN}(=R<i621U~%T-{mdtb*g0SRx8gU1(I#c8qHzwGvoWD!z~XbN
zV!MG65Mq%Zn>(-#Q}O^C18JU#j+Re;PCQ&_;snn;`Pa$5Xj(ze_lHDzw@3kj{$_^H
z7{cU}2~r8O6w3qkB68Y98@-0-cQ4rFzCDOOmGE+~%c>+LwP0l{RR2r2_FR#en`Wfb
zR0gqDA&O&pCP*s5%9ePF5J9nt^VZ^CA098YToh<#+r0g@<gxgGf`-Y`3Ttq)=IRwT
zs~;yXPeu8zk&(bcYW~kZhh%auvZhETL{Q}I>^=?rTxWYB@$6Sf*YuRLtQ%wExzk<A
z`zhzA<DGF;{pd=D88-A1B<4i_oVA&9>$*+Hq(<bnef~{*^AzzSY4nbs4a6<ATFv*u
zP)>l@+RmsjL~A0tF~Kyo9x65SBB_{!RM1E!_3k&Y=aj%Rsy&__6nE}+R=zv?^R1uw
z5CLff`wP?VXJd(!A`P<?-&oj^H8|>D_J6cud$B)SB=CLWLrd>;;M+H4HMipdVWi2d
z(w^?N`acb?oQQo(kMj&ArYYLlTlr^;axZk>qHcRIS=0xyfj;_Z#D#pzR9y`f)3sK-
zK0B5-edpLhaIJo_Zebpuw(w=>;+SB$t17pl=j(&f2}h~bwymkd!#M79yDx1DrQA)@
z2S@d8B@T8JtfZvj?EJM9<9pG)Q37lBsC`Ydh@PUi$}g<Pb5!gUjt7*s5^$D6ET%q6
z`I{9rpY3&G$IDN$^}V$JzhA%{LktEYWK*cK04++W6;-m4e27;~SS)kIK2>J^Xx{+!
zfZc{7jKmvdeLL>`oJA)Kdr`30J#kBaE^wPb7&&l!No~TjAcueyOjf{ELN+(MAo`<J
z6MZOC)Etn-*6a*$r?KuNevI2y8=?Hqn;&tO0qRBj=Jj3|o~QhQ1bu*6OY7a>;ug?~
z=RVT)1S14BS$L}@wqT0s$Q6sIfPU^mww(<@0j|FzfSFx-&(F`_I%;uq#x(*-9*G9|
zp)M4<8n@zE<VOQ=qP;=aP4+KJ=$L0gcMc?<n!^J?ub;Tc^AC9L7LtrV3pFj|YE<IK
zyfQ-f*3ZfD{e<3c)zLNn#T_v~s>`BE4=vZV`RaGN-f#6s|1=Uz{)>2+h^<07S|8-R
zN^NmRF`EK{#k~3!zivnSUG{vA>wwy23Fstb0QK<r7z<b8FWLjv$WMSorg~Pn^i<#~
zQ1({}c1rLMX4>(LR5&Jt+WpBiqr$3>6+F%qMSDW#%7XH7z;&hwIxBA}DAr>o23V>g
z%G1|#yi;UvoKKo<7!D1!ur}E<<4kjPh!iXUnN&jik~Wc6j<Po&yX=QvOIA=a>XT>>
z?&>9*D0~v7rcbDM-y9w7jz5t$7aJK}jR?7;H!^=QIUv7eqZ!vQja+yI&_;9?*k17@
zMG{jmWtN;&c^KK{zNw!+nPYD*-M*$1dZG0}uGET=g%KZHoddV|&+$>TYZIeq_p^=b
z4DIL5Yt&=Ry%!-{Ty#W;Q`*wrvefZdrUNC>36<X2hxfWKMplMdIftp<n_oO)A5xk+
z5Qo*hs|n!yana1-2_>68p_u-WL#`Ayf#gzZA>9e1F^Km2n(^r`){^&`jwdkd5>9wp
z9B)4Ul?hFsH;V1Ltj5lQo)33ob&WF>;3hIl$u@5r99k&@C+T67+buB6dHxSc^HcK!
zF?F$zP-APLd3B$9#W7H^-@`I_YbqkDKhV?6!(wN_-x*N0<h#U&%EW(61jDsCd`<q}
z_y{DT45)1*eU2(iJiNzdubwAzfh=%xUoov0)=QxttGv2RBM=pz_7Z!yf9oQ}fO3#!
zTE)?>&Dx4VX)(1Du$|QY$#{9asFSB5B0juE;HUqR{O2_?<0mb+79G#x<^4$;a9}n@
z1}n9~E{#RI$A3aD_NI>mTf-h7`q}zBq;f62!t$KG{bkQyh&mtfx@e<Qxur`lY^8Tf
zzZt&d{5M|%Qno1}Y+C=HuWi@+BKo->Fi2DWN)d1}2-rn;j&0q??aI_ea$&4yzv6na
zf$qQNQJNVJ*q*KaI?nF}oJ^Ni7LevJ{m~+h)9XVd{@MGdqg?gq=J^yrhq6<UvWZO%
zr=0RCP~cR@uC%~r1J*Vs_gX!+v*oRk*#|edHRN!%DP)n4EfG&N!F1^j%?f~(`FzoB
zDaOI|^Q3ac(a7Ef7aobj|70>T0owIb3r;+K8_{X$=oWp3WwD^I!}c484IK~A{@L*h
zs3F&GqY}VhfHpTVec&Ery(X|G?#JI~&JW41N|UtPNN@OVWPW`qy7>F{n}R(*mt7$x
zLa`?i8+u6Y@!|wn#tBa{{5iQ%O?g$v7Rke!Dy{!m--Xcd|GreETVxXnvnidOkyykx
zkKd{bw{?mXWthzcme882-uCQ%Q{`k@F6MT|<sZTjus6G_6pJtk0O<2y@;rO?kG`S%
z?vE*~3inN-yZ58hSc$!-oTDLC#2}@)?$+8O){3Up`e(D(6tK#zGU(P|kL`gEl*^`&
zTy3U2HU7?4ecppY*_o`q{<4t^)q~81u<d;CNYb~r*(PhSYi&?Q>T*cw&8q(<;%qzH
z`j0GSc5+){x!eTi%@a~67HJ;FIedLn?NkI(dUe&>HaCW!UDuYK;x0_1TX>S34J^^+
zuJXRuEuW?M(Y~y4LtH3vL6cnO0W3cj_WIny&Fj^IR(Ev^@0{vIYOGd=^$CI2m4tuV
zxKQc?EkP#N{lQd;Vu9(On=W4_8f&4o*N?7|#Wr@9{yf_4I$|q6R+U|<dUDia>k@-G
zMW(rU$%>l<Ik3;th(Ddme?*Y{p-!dkb|6R&3le{lBXo9UtHaG~hZ<lb{i*nW8R1AA
zEuoi0G(JbYm*pLYR#u{2GWYaWTv4{KxOG2EcPzX+{aJc{sx#cFMjOO7fy16ZwNiF2
z{ee7@BbDifi?r=VXdyaOHf2I6>B`nT|L_E@Me66HVBN%d(rd|~WfyO0W<*z!;U?Ph
z_oJ2AE8<oB@prXD=QaH^p~O_GIBFpA6pD{ctp9Ub^m4;%EiM_-@7||IKtvz@rn$_M
zMYeb0w7!(0pRdABrbMS2klcu84nB==XT!Q?<`3u(XNxL)!ZfvlPL<<`80OZ>Ot(AE
z`T|Iew@HPxTKV@RWzm~qzoMW5^;>!Ba%?Zg(^Xs$JoG=Ru;au3{m4&(`$`*-01I1S
zA=sDEQcEQADZpA@@8b>{P^6QEIz5#QZcD->ew8!G`S@)Ny|-QCJmJJK%XA3C^2JEy
z@4)64WaA1ck2<_kdcTrP%Mi+b1Jr1w9a=tKYk{aIp8C>8N5RpiS`AFGwh?bltG#8v
zPk3hECyT)58|{bv-jySbg_LCq3Ks@&urWOV3wE1%r(ni6rk}isu@yuHw0C$X4xjF`
zf_a%_x;zaS?Bt`B=vZ=M+HEMRLh^7t21?M;nYF}8i9mwHmP)`Y^<MEsNfkF|IS)^V
z5;$H@1d}3;>BYzI^F)T=0|hPl6(uTTuB~sXk{(NF8?xx8<xzjM&8~&GO8dF#I*Nxs
z-c|8=Pb(LSj($<4_<uNi3#h8Qt$X;;AV`RmbO<O73P^*1(x8HLm(tzcEz*rBg3=+~
z974K5y1N?=@!#jE_ul*5_ulvW#&9@}sPK!u*IsL`Ip@-jULz)E6A>YV;g=<^E$Tvp
zIiRitjU5V?Oq=+NhKD|?Et<Ekmm3;A7KLFLwbBq%%cASq)vTc{+k(R=7hN)x6510=
zJiLp{BIA5KYH|k)aC?~L)$b8(*soKo(NAv@=A1OU%OIkx9-F`!WI5rjff~!M>nSa&
zfoQIpu$>LVHL^qKT%z&a!C?_fF$#=dr%K{8eXy%<3>hcs&%J^nvBFT#@Mt9>-4<eG
zU&r*%!RF`bcO^zb%u{>abA*O&%S=C>z7)bm_~?o36;4j2{`m@{-t7z<$o*5S)>xwI
zt}UZI?%E+9a~T^w8ED6op`Fj#6I+gOPO1nD#?;dx!$(2Zf1JJoRbtwghi0K)&Zz_#
zZz)GJA1z0q)bAJF*Nx<jtWjgSuS6kn{Ps(<lih43o=GDRuirHEa)t4Ef4~#v>F2$4
zM07CB{87n>Y;}6wLz{OP)keaMn6<a~3}EXt?4EcZAl6rptyWZi%l$$6--OP-pRVnZ
z5!e#tsdAp}wVV@Wrh(jUgou^N&KP|c%Ub`bxG3_Z55#XDOPA}@eI&mhviC|{-Q4d8
zbPvV+#G>O|Xm5H-$aVj(<I<HLVb-pAF(}hp;{AsA$^&{Y{uHv62Ys!G6TpDzP4e-Y
zM+mQrlNdzt5qop5M;3AZ&@&rK#5kA7=<2v6?A6)|RDy9({0*~`S3xVDo<+qX7-q79
z52=tj-#VnxC;VJgVU=7yqd4)g6`g3kPV|<Xem-B%8UGBH)MTtrs-%K-P@l55VX(m&
zPPon+RJDbtU}Gsrtx&Izpmcg>$u~1iDGI?q=U|ugkIU^t9Jgk%czkF6y;ckZ<uB()
zXyR%k9QoLT@x|=o8aV~~q@P}Orl4M;BU^G%*>|7H918}^>j$4a_AEM<x{c8ukZ8!w
zQ%w$A9F`s@sK!ysu72r+fFd>_iMT_~y_{AG7{9_QSb1?0e4gT}N~!)$*v2EG6~h*J
z6!9C*TQjQ{ek*F8X0I|9k9<*_%2({YRK9rt7(+t9YqQs^eyt=pl`|WMy1$%llCQD$
z*DmvN0e)JWPQT5N{H=nQJjpQ6>mp~O%pDr~NMG&C$i$+$6(#XT4SOAfLbdvgH5ZYT
ztcW42BeVqjTl`-wKA!1k-_?E0O18%cj_&l1<YlRi=_~E1)=+k*Bk{~W`WEv;Olc*$
zvQauG+KEwoa)fC%&iQ7jz1XyyCbfz&P~b3Mc0Ag6wtwhM-fDQYxGT(=0#7%3Pj>@m
zs(~z?=>g*GM1#@iSO66K5%m7UhbCvK`Nc3cM-B?Yp5ixSo8Hd)z&O|5hiDN4N_yuH
zBx(R3ecqAO=}<LHuJ-LN#=Ex<Bt(!ysi_cjImigY{trI|T!08RqG}eDb2u1}cKVQ#
zkO`jaeC&pjDLJ`rJRR4>hfA`&w}v9PCv%679DM)yrsYTo%)2zY9f;KZvKSLnpWeix
z;;p*^pv+OwZbs?P-Sc}1{lEO)?uJ+;K*?$NvZ;U;0fB0l<wX?5&h9hV*&Zb~m`2|6
zY@xzsAXC)HGL+JNU(|^w5*^nm^ZN3{?yNRoP7^Qmk8=YX@v~%vEl;BW7e>{C2B6n&
z0bb06W2y!b-k31ZBV_0xxBmuiQ-KaV76s;~*=ql%t8?Y`i4wk<2ym>Li_ilp%U!$P
zr*Bn(SIqMaA&(Yue7XBL1g9bM`=DcsS=4yr`bggY{*@f?y4L%eWPAti<dpdS`#0&1
zAJl)Gsyia{Y+e^Kyw}IWGK(%l$HNN3E<?_NJEh&MdSNUTra@mG-vzH#>!k={6RUrL
z#{X%Rc%xqhL40e0+Zl&D@DY4SSH%w3nY3bY^(an`hem*tEf6@(f6*Ti2IjIYA7xlc
z=o58)Lpy%F{^KdsghBdc)6)hKb-Brb9aW|9i!#`W7Qj8G>8TLrtF#)r(*AGj=wH@y
z6R$M55u0+|IF>(w8=rRcdCO%$JULcY%4b*k3(ooPAAB4cz{{%z4*U#%<|qRXkpu6N
z5*Ai1hQgcX{tX_m)rPh8!OyMFaRlFlnnVenqQ92qkl3i@{wwtU_YYKT2pEu~dm|@*
zY%nTB2Yg`Sg45C&P5b;Kor4Y+(A@;}^2`aya{dH=9PSD6L&L!aZWwQfpO`Abjp>p@
zvbK}qZWpO=(=<R|0hTvYb1hu`0{F(s@4lAW9ZUS57U1s}dx-l1!j2=T#q`&G<09lQ
zp6tyDgElD70$4;2Va{Mexx*-L0{|oTK@t<B^jR_hc;7c{x=4`m0_IX{^b3lO#YXUy
zn4Mz1Zeeq+F$zK?aElCf`}KjQGx%VE>Lq35zwY_Jf4~K?j96;@&qXAjxq>WowciRH
zb8!4Q4BGvHtb`SARjwEEFh19HE{M{TR}+jF)uKsv1EV2Y!LX8hAj7UMJ|2>JEBE>{
znl%$pu=6`fQ9O`K>H7oi_!I5>56ffy7cw1W``1d&NZ23IAc}4u_+vdi-W+9uyYGBb
z$PiLKVrLCrYYz~3!$L*=AHLbY{MEV&Jeo0i&S?Jex{_)4!9)gsJX%SAu&rvCW~3O>
zp7d~Zkm)s}5ZUX^AmhspP%i@NIp?&~oCmekNA_V^t^nn?TBBMcLf|zl0A9iAkx7zK
zd@*ps43HqwDsMfJ*v)+laUPcA2>>>YtnNU^q!&nH%=mBtcxAn<0;XRrIKo%+F#1&M
zmSXsTNTFweD@{b&&drr;pe!qHMkV$C_BHvty~p%9aBRF+ES-qG?6D9)B)|8)`g-)O
zPIcJgTWyzzz^U(c-=4L&2~jUvf)E3frCf1c7V?rp8yI>uox~#9xC2W)Jz!al1uRg#
z;8v0?PljcOsqX^o%f?n<Gp42Agwf4hCY$Iq>(bpPVJMZ#@!>1LZkRns|4QNiiU3W{
zBJ<!;9Upju-`xc|3rcx~@XpVdqCCt^;HKACkA6S@!ePiGXHhVte6rtz`4f2%GTM_i
z?)(ukz$-wXx&bySRn9ogQ}<uXE^J(!PFc0azLqC}9})14Z%=#--wCb2o^DN_-QO2z
za^<T4;*KO%3Z5796^OG7z%&&H{2QGKur0NojDHVT!H3XnY8b^S3s6qOfI9R}f~Kt$
zeDk)zO*LQHzJ%*x2*IQeexeKN%T|xQxtmiSVXu)~CC^=ODiAl^T$-71>Na_^Uo6KY
zEb1<SneSLp0C##v+0q4E6`zuUu~yk0Ke%te;|l`2?blv!JtO9>Pp%dGC;`4SQMES(
zD~RXGW`!7V>1Ixv*(1!ulff61E}E(@>280CZiBE*2iN-b1wXNE1d^q(Zq~F_KO7-m
z0dRsWfIVxH=)4$`*4E^MG@JuZF#hG3)g@ynzF19x?cD&BqG08jQORPO)9i-tbIS?<
z5CniRJq0Ugqj%dC5L*N99<?Qe2b}CoRkEUk`3u@r!1%CU?hBhPnaM{fda1qdH6Gyk
zp)E=U0Zs>Ne%)c$hE{-8BiOwSO+1oyV7CW?fFtzBS^)JS0DC$x*}Dv*s58%2C|<vj
z&P6|F1pDbUckz`z$Wg$>wmTuBFXGw@f`F=*Vo`wsOiAl-4ylD&KawMDlNsnz>OF`@
zFgOSAuNCf73GD8eE1v{qnr~z$IEqf>m-TJ>rNea|@B_6KSkl^2qPNlLGKoDzpvtOz
zI;KjS3Leh_e5A`p1^P^JU;=jVuf%c@(=!mGsy^79?=LK5g|Qx9SylWp{^YbbYl(-;
z<kBw$Gp_<8M6~D<BGJNvf!~z;yKyTom<G{Kn(OAbY>!cHq1&q?xXPx+*$N!n)?uMv
zVoF7Voq?BV)w<<w^%{2Y6)tAWQ&Qr5DhpNeWldBI12Qm(xDN9H+(*)SLN6#zE426k
ztmCsMAJDoifYxY-eL(i+8$K`-+N;xsBkOvgJYx!z(fQYR3Z9w;VmQve2ABGfm67Ga
zlcqgY&)PC#ysaCig7R0xmCI*0cbM|owZl?pJ<hbw9KGKLsr%RhH^PW+LZz?$CZZ4V
z`<NTZ-nPs)`_(txeek4{NO3Ht=&g>&Q!>}}PYt=TM<iUWFet#3#`F$Mf1JyMjAkEr
zpT=P?{nCTGDuWuu{+$jJ_r7s+{X3m-v<R|ioL7rPjizk{_t%_dMXq<Nl+QO|w{O#z
zq-QAJ9v(D)+-NA+)mJ-)*`DEzq=A<|8RCuXz4PXQ^q(2(2vw7a;3A8-4wfZnW(q;<
z!Dj?KGE0+w{yhPwjFbplfV*M`_rViUYha94l{e{(Jq7_%t?7XxoyPqf;p?A!cu@ow
zBq-kCpsA{5p|?b~K;H-YUPMavf-}!AMBRa^U9HE$V=!D1n)h_<#dnFOAvp!;DxtS<
zv&?BkT|vYqV8Q$-6@XJhx;>BWpl%$xatA*dGLCKasR&DSpCV8pRRoi*0WBYPUQc!4
z=$0xynV_qEc*Ax~xjyy&_q@V8u@>1Pd>}fUH%0x@u6I@DlYsr3_os(=wR#15Tvul%
zd1XKMG{vQ>$XA_i!Ny*1{FPG8)|fzYJMUzX$CCFnrlGI~bq~dvTi2WswOyq6IHTf*
zyZ6cPc_MB`Tnj*?Jff*6sN3V3s&FYNh)I94rsv0ZfGshE5~qaQ?W^AYnK&gho%_2M
z=~>ytllt7)5^90>BV1qaFPW|$QQn@v(A9*GnX#j$3Ug6=^PLJ&0#C6~<NnXnkQW}e
zUS0Z{-=>zNg(g1**fVs~%q2|uWDmiiR0?qBnn&3xKK|qGGG8NF)g4Hx>W4g>79(FQ
zD|cu-Zd1jW20&#i$di}-eDG!2MK6oPpv1iGwzG(s%!-cTY*=(O%!W<(e87&q4`KZ=
z@(P|eA8Jqu8LrruA&N}rAMef0F`#}g*<fZnxdS0F)G<ynsuLi7k^v6L$C44~A42Z<
zyN@D!Z(x#it_4vdIQ1<fDJ=tQkS$i9;k9;){JVvq^SVnz>_a)8jTBG)R?A@&Vm72+
z#FS!B2Jh?t>v{hRxmfnL$m_bBsNO+Ye$nby08k%JdWCPVEG2NaU`uB6mAAP{pSv{2
zb^(gIdktAo!(w)y22G5-NRl1kj=lg+<b!~Z2g&I-A&^~=bYX;JE^K5afdlM{dK;u$
zcQGJkXWyJ;AGv9j&8bn<!4Gr{NXo<9t?%$BA0gX6Q`B%PY{XLtrd~g!+!@H4^fAKQ
z>)vU4^$uLEDGr>7@ZuRG!5<0Qz#|_JH3hZ-4lPt%Iq!=(CO}!qP=%!T&~{9kN@ej}
z&yTl2UAIX$!>kzs5rIgFi@qUFC@HYjxaT!u9R)Iwk(dLJl?1}=T;Cy35n`4c2JM&p
zQry#Ji6H{k%j%i3#3J7at$SOM+rTlYt_&f5n+8_+l4Xd`o<F$!o&w@ouiI;oaA@bg
z%cfRGFcdQZ|L?I3HfzSC4!A9KT}8Yis>HvEWa{!%H>dry5JL{ST{#<C0U^*LCQzw2
zAl@(oNPUO*3ZKA3#;Yh0DsBxvv4H2pzdMR>a((+~*e2UDis`~MD1Ho^Im}td5-ncZ
z4-|NAAXP3F>=OO<$BXrm$I!*V<|zWM&Pq8nJempS<%5#DwmBfy8`;o8;90}Wip(#c
zk%Yo~kgp5~#p1<d>Sonb)5<Dp595+2QwCA?L0}whzSP=b;-@HZ+E2CBo%dHhQXR`9
zz<svpS@{t~B%cLtpSbIY%EJXO5EFra^|UgF@5@B^DeY6I7uld5tO5Z^9~d1h;q+AX
zv3PDUQzCGfEPPGWIl4IR`LD|m*Qb!WNxyzwr5vQLyNeFXozVzpx2n!?3+9>RTfVLQ
z*Tc$>8mUW&6R?>RTcxjo*J0wUmR*Y5qNq2_pu`iHPg;vojE$Y&?bWYamxOCs?;iDV
zDLiU8_Fo01@oJ%WlIcq2<5|dgK^HFsbG{S-2O+h?Y6FxH{aUraqQ0>weRbFBVrJ@P
zI-mH^bEgshK!g|0_f+X#s=<MD0=v6Q@a$o+M1Y@UYQBMk-7}iJXmJA6yo8eB#CQrv
z2Ssq>Y&|d%k9EfYiD-DF*Iw7>U0I>ETMYykhf+cC<c%{hByI(#%xXN14XoGmGP-XP
z4Kv^yExgn}2XcwlH?QSZ(&=+?oN0f$5=q&y>8A+!mp2}EF~QSl$340^uiN@FCsjyx
zM(*lZO)#v3_wd*$e4*r=`!tBXLyR_q%<X(3$is(i>Ntj3KhQ4wLyo{<v3aFuLcl0<
z=&G97uck=Ps5M-J4p5odp}C*UIO&`Nz{OoS)rIw7+U4eA%|js7IoROzmkCb2>&YEZ
z96QM@zI-t$E3|mBd<$~x2%T1%Fy=+CZC!Wa3XRQsvO`NF635r!GhCi>-lW%3eC*uv
z9uYn!*J79~_Zu%wl=hyAkD1S9R9Y@ouY(xE`on6-*C`VWgGlIo_3}ucfm!#}o6eLd
zb3RihxPDJT_<E!OZMcD5NEBx3Fh@Cj1pZUaAm&i>>2iz$g|k9t!x34QRkr6_SM2=;
z7`S*>ibUGxD#X1w;x<*6mAE{i)=*<9?d0AU>1QjQiW%^*@h_FzWNH&|1*PZ{zICVE
zpMRlP(A4G3++6bA?dQV;X8s>qDBg4*r6NGfrC}4IswQ~@%jQS+2KzHeA$eXukh1kh
zi(U{pp}g?*aVVk))Ta?xTepH3y4B>(?T3mrG5<9|lm%6bB-NxeJ09`@W`ME}W@<S&
zH4i>+0ryRWEqFX&4V)dSaf?;>9Ay8N#Ad=UfVbu&pSe_KWPZ@sZUJtZn=R=i#)-`m
zg3N>lwqVnZV*yo*Y$7<indl#sAdPdqlbh$EpSp@nvmcUT(vF#tBD+dAGWYrU=H|R!
zT0qv+z#^R^(5m60eYj-CaVntxbTaS0T<QQ+{Tj6HS26L_tY4Bb`_fYQhePp+O54If
z?WRD)8Lh*-N{gufijNf6TN5-&??%uXF0Fr?+aiF_yd>te&K#$lZsS~N9Fk^I?l~2L
zZR1~_?r0pSrl$c~%5O>j`14KXB}y+)i}5pCdw@JvtkLsYAOfT{Y2!?rR0a|o*2~c^
zT&IFzox^Zr7i`{n@2b2{{prk*FBTl>f%_N-MEajIJTD&&I1&qk`seP<lH|jZ`)fl&
z2bw;RsM1dnL)>^*G#|y>U)1(h3<g4+Y&qX$dQ#0&^&UTs6%Nb1h+Gf8DSSo27gDp^
z>H@0;v%uyAD?eCXYTT?_{lxcpbKme5lxefKN~}nzSo>FP;)F^RE~*u;8^a*SzL>1m
z3piO1#G6LsPG48zeYFHgp-}n*3F#5H%VJRVpL4~oVTvt*`2#&2MlPq#lzjpQc+WWK
zYeBS#6EBs3;&sRL8y`kDq&MJy67=!3BsxlePGJSc3s^5j@YJk_lo*P;29rebE(R9^
zyOPTbu$k&!P_eEiMa=I2j>B=)l5idFLVch}hL5fn0C<9fuFG=I9?+Oo?DcC@k|Z$b
zD0Rx~RD()75De(9WiA3<^TlkdL2ZDduzk(eKbIvo>;7ffrz+Nj5>d{-t0XGNL+;s1
zZ2Bp=<HV=zYu1ega;x=$PqUIP!e0c3C6kD#$$RSe-T9Zoxipaz`;gX}$QG#fvX1eY
z9Y57EA`zl&1K47=Z+N^!i%w>-J0?nCuX|bpP?kiHhd`DjJ3LVFCtk{dfND)VG5_MJ
ze#v_u%9p^?->M3fXV)D#!m&f2(onoN?8j<n$Pq<UXt`7)jD%6C@=#D=j7+FkrG$_R
zrBG5@R0Uven~KeZYLJRG&CWQR1YuW4fU6)ptivg{@`@U2I{3{eiUad+IdWTZOjWKs
zM9UK?JpRB*|0>SC(aR$tDp|Nk#Z8!l_>D|>V|t@Ob7!hApXayk3DA!wnHSwJwvDZj
zY-=e6-*l4UV2REuduvgM8I$86YK5C#p;|cp?RkB0Fc#C%mpxz{ygyeT{*-+%xnh%$
z;Rzn?K4~IH1h}<+3CUmr)w*CbBu{(;BVlsh_GK1CD*CIW+yX-cTv2pPDm<(Rt54bp
zRAWWBzU<oepzl{5ASF%;ct7C|9a%_XDqR;~THjr8uvBP#%$iQjRY2=XaSf{qaE)Yy
zFuhkjltzO>24B&`4w8Ks1KXl|{lQid_iw4fk7oxn`Yb97F_79n;bK$cE~hs8q1B#q
z_Xd?bZ~svo>yLit6AR*<$qRr8tYJ#y-BLlsbHiSKA@I4`1_LBdgpl!uLFKt8#?oGc
zyGotOznS}a`kpD%u8z3Z3oi#STBaJE>$I{p2<MFZ46H6*oy!-*`chbWPhI50s=^qf
z1em9?=OR2t0Y<b;BS^?c3(3g{;u>pkMx_qV(SDI<AAIKgRrey2Ln9q$Um`f5IB5;1
zUrGEp7@5Q@M_|9N6ilIX9eqp_srn+N_eEX0LVm2?XpHrMD4L4~J*7Z@3A(Is=nIne
z?+2=Wx@G{pIXenlhiyc?3qr@+V91*2x6-{m@2kf+#s{NrQqKq)mKtwEU5lt|jA1u*
z4k<#v)t^pW>4NCPC9EL<)%KYd!?S9~jc~-Ewj0=5B?vtncbs*HrqCLS^S)58{AEn)
zl@WH$x))VW8ySr@`Jx{4h5prv=b@!-3kJh^EsI603GU7VJie^=o5FiYstjLxO^6t&
zmW>7>d(Tzt+TK&`RrP$PB3n)Kdv-6JmdaEiC`<L8$*lXbi;n!7)D-+|nn&J70o}Ae
z>k2g7;^tiFv=iOwXO0qt!S9d27|4*jp>*6;Zx6^cekD#&d9)eW<P=m$Z+E=CQa{hR
zdqsar^Kl~T^R&$rRY8Z5@VRY=5o_?hvWf9`0;jZ3ILeMf<Sgm9*WaADr=Uxj6MW6-
zN7e4``-~;0nE;0ij9|k$V76;DRHgdiX)8~C?Pi7C4fuI+6F5syaler3c{y{cubUTf
z*c0Nf_4A18Zzgml^pVWs<cEd+p2!j6%{$RICXpaj0R<X9K2z+>&uIjD`VndK!)kM{
z0yg@FLtcw}wPWwH4d`>F%{h5F7UcBti?XlHKTwp7#U-vcEj=mG|8;Qw@aD9#%_>B$
zoSxn|Jm_jTzn{+7=Ml%?t6w+AL}k*eXY;NBwumQ@dS8k-ZC>ifmkrg{m`~Bq*E7q#
zQLkZe=oV)R36YhCO|Z+zgoMc5OQ~X;(%F&WVji-2Iw>?X-o8`!F8E*s_HGe)?ecyI
zalp6Bm|A$Hno>13Rf_SrJJRk3Oqxtfv>xbK>#^1PWm86<n0U^{0e;qEDAb?F2vBkL
z!xX_SJ`Ark@~eI*FDt(Nu9t#o(+%%hi;u-7f5`uxkYlG4ep#$Mby4_CHVm58^%*^W
zW@azhn{%OBucZ;H*+4(iW=|Dme9g*HVydrla}Pf@2M5O8|04Y}y{WvcX_Da<&r&D-
zy0onIvO!&3`{Yr(lDN<i$!Bs-!rrPzHN;~_(kZM9+6;PTIh|LNeN$&CBAg`>yV>WD
zEk=?ROxxV<zB4VK>jfu;8$(82iZUj>_}K*n-}o*6!?PQKi%h|!Hcb<gFvO4BsVE_z
zDUv=sDG8ti&&w{S@)h#&j)0dl1IL1)KBfd;s-)1jt#TuLh$C+wb{|KxPLRQJjFWO<
zuh8uf!<wFyfW-3W;IkF`^l}z>R~cV-G~v(2S+EEWl`iX@k09$Dme4{kKi6w&ASSMw
zM<el*kwvx^AScGv6)^;Wuw0L_f_@%#cj205pI&1@xU9L6#j8E2$feMa3JZel93M)D
zt8@M0Fj!mTwyx}tDqc42RvfFv($T<dj{{XQ<Zt(JS}L5iuf$EyQs%C{w%$k|TwU~g
zog%!^z*s{X-DqY+&FXn5roWqT9d*o65=}VVJf@jIM`~Uys-Gt><EIw;tswsxVIMU9
z7)uBoJt>DK)LbSu%_*s#KY2+`wd;!3O-qZMg^n*(R&XhAeqt-|Md9yzfRzfDq3c4+
zau<ndc2G0z64|cJkHC?e##-OqN9qIslGU45(w<0V-jQ>WX&8_sEQMQRd4?!AD&t(j
zL-`(+FelB4x!ippfi(i3<%PgbNXY~JT;iTZwbD+Suya)v*B&HMh;NMRbNFKB83Kzr
zyYlEOHThe?DJ4iJjZT^Uo0EH&_(%E+2I3nro}l@`V60T|S}#LwUZ&@{e^|-Tc;)+)
z21E`tFPl_&`@$YipJ`X^jaP!a#(fBl5LcLcw)>a<pO%$b0?9_dX08j>+#P(LA&$KF
z;V>?v-_&PrPbfn;NLM31DOuYtIZxv?g&Cu;bA*}LIBjrQk%5H}OFSh4>D0GEZe&I_
zrCZ*Ch24V?UT8nRA>DC08r$JLwHy<`zl=1Wf7a20+CI+I%V5}Vb?TPxZe38VqR@k=
z+uML)aKGG#Kt+}MQ^9?5)`oL^E4MTZtPD_$8GZ^Nl<uL?%aNKFMC2%{?ml&4BP7{R
zPxjpFl>8X<(1sv7iV;ldYDn!(XaG%;@J8fi#0gi=Kgq1?U|gtEWQdD~(Q2WVg~&*6
zf<YJAI9cLIM?J#oDSVCz=UG7&LX+$bfIg=4`3eP9c!)|WLchm)OpEI-UIQka{h3ge
zViOksu5c)NVq_(y=58p$@KzOlbFVoT_d2Py%aoe8560%z6AkLEU0Lgg{xU)Ks_(Mn
zrcq>FDW|FwTYBoR8JYS0AGv74Y_<5_gm!pB-Lz%b*9xV!JBn*R7dQX(Ch_P(JZq#s
zw5!BFZJq0HH3}(EJXyD?Hpr4^L!0K&{W_}>QJq4lsifFYrPdTKuzGHzF{tJ;zl(bI
zas)}UzNso2Onsd}k*W(kY{@SFYcu|p6^V@FI)jYdmu1?{$*zJue^IVC*&Sr{y6ir>
zZA-t>t}0qts2ctF?GUa&f(+Z!*ElD9UtQQ5kgbMmm*8Bb(G$2n6h2ozC}D6))(+|M
zt*<rfrdZ-<#p=BmTZ`F#_e;wy`ut&E{`r&TQ^L31u_-QO=Q`qA<wnESdIxPpIH0)_
zUi*EyBHV$!J%CAf`com8sHRQSbD155vmjZLK7bWwUl&mxb?mSKqV>xT<$ms6Mw_S}
zUc6lwvjEVcnVDhaT)nULY$%<W0s6zlQGkV%=&UI_3xkQT3a>AHs>eyQloLn*TRqO@
zn^r)Zq;Jzwk_2s8b?{T}4QpB<sBgz-%D7lf!tD5~YUDFWE5jZ~(Wp9_G@4<4_6zZt
zP7`2VH46X;dC!a|C%yZann_$*D3jeL=2BisUqagc99#I0;W>8sRELOkhRCeLXn8J!
z9{TeW6Ju8aK@a&KJ9Qqydx&*bhFXxEy$;nU9*)^JxdeMpr)vj2D}!r#C-@78ZKxwK
zSM4(?K6+^~n9LyM^mo^$F;FJxc)Co07Dx*~yS7;?%(4TxMQ@{~3^YZENZ=eycH8Bb
zPBsx{yYHL*dlrBx*8a0d=`{`>aRGylNC%>Yo?b<wUe`~;!saE?>?PL40bB%D(LoRK
z2&1EDW)ql<<@o_B4fIs^Adv}eg|ff*5|0fLjWJlUD21lItW#jY-se%GZ&2HMHdDg-
zaBVo8ld*$g=LbpH=nxAO-FHMtUVzy=A9OF~PQ_#9-L8|LIHjMS2cddaDc>>XV4L_p
zD;?Bz%E8X>dc0oh5!W#BUf@<VMX=o)B?ceMeGES_mpu=eT>eL#r#7K%=>GEw*-<pl
z;6p!R4uiP{53TLsmmiwEALBnkzecZk%s%nBQkRwbN&xi>;|o+9Mj7!`ENhxBx{=ou
zHX>~*#+UP!pN(D!QrDwPTeRsD3piV=g53K0=JK!RImoqs|25!>c;G~ipAM!`>9vuj
zgMwLTaP|&DX0`0mTv;$wHBp5nF##dNod?tVhLqn;TM$3_8g_=W>Rb4|m|+iU&Mmhz
zbS4>qvn*JIPRAzfZ;CF4w?fq?x*kgPfrA2B5k9S+7|?8^ERxQGn<gX$Kamh_IJ=cX
zidBU1j!ef^-0OOoQ7#viPifn~cyKt5sp($f9k4ICreOs0hcUUOdM*Q~!x4e%jD37j
z56p(MLqIb~XI^~zw%v!LN(CJfwz)TS0icH=O{wCKLOhm4fd5sp-RDUZy#Y+>cUx9H
zuP-G8l!etkSnO_kxO02lTNEtYCQ=QbRm;4Eb7|{ur7}`}U#>WaBJDW}{5$2xyEPR@
zr`Hk~5VMoF5e?%X4kTeWG@IOECrPjM+%u}<Cm1+Q3|KAvs^6Vn5Qv)=+;KV7W9`@S
zS0L~0qcj@sPDRF4>LGB_>JQyM`O$E+y>Npgc&u13{GQ^up$9-#suH6wh&Pdr7<vRb
z6#Nj-HZ_@0ynoLbk~8fU0sEqk2Cy%jm8ZM%e=8uHs>GUl3Sm^_!Xw1fqN?q>XnBhy
zSh@Fmj;q6ZhB|s=8#Lm=V2QjUorFljZ`4rK44e2U@~D_l4s|V^naCecb!^J2xHu9T
zr`kwTKFPPInC^LKmHnLj)aaR!f^GnbzL@8_0n6Yt(K8-Y$ax7#K;WwSI1APx36fvZ
zs(c$7huWw1ddBCd0(g7T3WqYeIRdKWwH*|0$aXWNV>({f&M7s*B<5`oATHFDSdM?R
zcmI*(WKZt1O9;^q2@A0PLBIJ)&+?aM-9rT3@zIFmv<C!2))v}$FE4L;%?)<&*pR|M
zCwA8sgC1n9!wW)M8g^axWl(1ywZ()hD3dOA&DMfN#OqoL!9^g|W_XW+{2bo)iBf2j
z_@=7yrRb-+d>#P~%{M9fq%UKS?ozhu6C+9-dNbS=u#<f{`o@zI>XqSjQ=JptR#t!u
zAsn5{=9&$M6bBzNh8j`WC?!AVnu_HSqey2`Roh~wdZnZ5gW_9_&&D5j&;xq$1F&Ru
zG3NFl62G>gr0Ovi=HDXS2W0Pe4{m>1mhZ8Bw=_Wcz~>5R(KXI#Bm3KY#{<uiw4-nl
zD>#>bqV*#}Ps^kFu+ZRr1Iv{U)reR;_et1wo9kW((#%MoNqmUZ!(3H4oit-3`oTry
zzn)ypu*%Ec;O$)!huBniL1lFhqVia82OqQplAF%Au2rA?4_(5?Cd?b-0lpndbpDSq
z^^ogM@=DH}>3@6;@Xt-oV({*~ta+0#2ZHt}S;~(aC?S%HV7j0+02rK8Pp(wTB}_wD
z24s4EXz62)mMUF1l@)%3PhO+Qk@I~k(1EP&InCUlEYK%!hmblU*7UstECGk8R{BqQ
z>BF4}8j4AYC+Sc^DQ;UejdU|ibbLFDam7jVQUQ);+2`Z;T}m;o>I4ApLs+D<I{TCZ
zL0&(M7+-@qDN$p}5KyvZdhkWX)J92@^`awdqfet7SnJbx?rv0V{^IU_(Ps-9)P*j5
z7Ng7|^s5#$9A+=G?c{Pl%7T6S69CF8ljEbL$1g?@>|#<98Ot|!<ceRFW#qc*k)wrz
zSW(F-`Yj<s4I;Z&Se;fW5NfDq->AX0l;l_LcJy-5AcY>%OU805%_p}VlSrI*<?W#X
z=1k$2vBqmUj4G2ywvkVF=PS6X34*>qJi${SlURW#cz}N|CispXBzL_J%=BqSzgKcM
z$2G<z)=q@3l`qx@l0Kc>BUG;JwUgHmB)+G?96hmk3x)o?F@IXQEoE|2L6eE<wE8p1
zh4zroYB$wvHjeKKB<=JZ3-5+r>ZouHjfj?sCno+Fhl;y?t3R<yPIy#gNAJt#R*`Lx
zY3`iCjiBr5^vjVVUObkE5~>7+aUL+jUo`~Tnk4tHU=Kd9W8==F)m7u3A{^OJ@pt_#
z4+>8}R_9Qf&27SHj!>&o2*e`!nE3E>sjY~qzNH(!-S9eoUcr4xa`&r7w6ku~muCkZ
zlo{rgGmar@VNy3&Yr<<0r5Z(6Z_Y*Zdr_?P$8JdPjT0nE(J$eZB$CPyStd$|MN#0z
zyJWru*yz><d_wPo8ViuR^O%-<d?|{Uav8>5`gX>32l7avEM4`4vn+gdu^I`RH*-4`
zu;-y?tu<Vt%qxB`v(|Of<uuPjGON|u-!4lPY1gaMLzWvPj-57+d1xU$8Wa+x%%yX3
zcHf**Oztd>s4og$9)X7OFp4+4`w*0iw@3Y0a$i<MDop2D(iRXvbQyx)y?jr-MO(m#
z%Mv3E<wp1l9%OX~2H5*Y(CIgC-5vKMn`(WlHRUI@AgYElLMySa@0GD8QJL~=RfM=8
zO7r25W_NS-JEptW0Tkt?E_-&^>J28XWcdEt0mxD`c49SQy|1#4;8@Z<<S(O$dv}Ac
zP;n?U9q)uG_NbdHH)pG%%J$Tl=O8u-FiR=Uy5j~Rdn+n}b{1nOXoE>azY+N3*amNM
zAOM-TN}p;Y(+~Ucbz9swSwk!`UiN01%)h-;AYY?y60fQLa*WXpCs|~%2mPjzs<iY!
zd6HlL1&Oe6;ugHfgqy_N_B&l*WSiDx$3cD-Tl*21Hun$=_4^h^n`l@M*d87qj{x5H
z5?kne-ms*N@TecaTO(d6?OTgE)ebVd4KT|_29=r8GriJ;Clt=}K0-b+>WRy}sNxS)
z(V@O1h+mHBI2yxBCJlGF@$Ed4>dRZ5khxV;e2~OY>;~Tw>I7j<n!Z?$V|`tvo5oOI
zRAvLo*HSVDRfs_l2U7mNZirQrH^NgCBbSoI+B&GFyX&)Q&_&2z$33FTYJEQrx{#`q
z88cK-8v<ooAt<o2EXu{lSsu%B&$$&M+@m&LkV$B9SB8gF_sQX2X0s@G04~~3qXez^
zM5Yl#U{9xjOPGVSmK>%`beX(0ybPYn-atZPSSM6dlOD`zdjY7LwO`oS3{y@c93C-4
zB1yQ~y90y2?Wu0kMXfdrZZWG!J$haEh;Mmk?<LOID}(y{u{o8z7|o2|0*zs#%<O}^
z$0tnr#lDz*Jq6!|ne@p7_KN93O^2sF=!1hE3-S}wu2=yq+|laMh{Gm+i&c7oKnEdQ
z2!@>q>m~O1Wirbf+KIL}rCQX<*1rT~Ud9<kMoH&;2_Ux~bH&rh4mh%=6m%_*l1bUJ
z+>2t2GADiL;WB5|DTxON_742&2;!M_Bd=s!gag;}nZ(lF2O8lmJyunyFY{vAaV^2v
z4Lhf|!}G(o0~O@WHt_y;_K<>M>p{Z^@xg8-_C0oy6{*;(ld~CjgOP?UypK~f7M20|
zNtRICS+6Jos<YvdbbeU8d|7wFxapBp{znv|pU=2{317&6{aP1A-VSmprfh0PjgkJg
z`Xf>}r4yaol%l83qqsR~$j0JdPb~mjvmvA)W6{4y=l(`18%!XZNhq6%EBk?<NA4@b
zG(ERph=Z@3WgyTh(QkuhydHeGFE@vJS7ONbmSVkD>A7FdD{*gIK#Pa4=Ta-t@S~J;
zhpagPs1~YpRiM(a#dbiZBsbT88V(pOyLXHi+KYH>B;j36U%A3SkVc}E<-g_Y@HcEm
z4)swe3%}RXCCh&HxtZ`gO4;G+TEv5=k3kkcv25uKR+V2h16ElNrwn_v*rW6+QB7kd
zCis!82PSYHRmq~`nsf3b!iIFm%v+?H3hv9M60?0fe}~vq-1Ru`e9u^Z-jkm5aBhHr
z&Gd8Hs^l|xyUFNL4S<AwLs$j&4!o-auz@hmmHm^!wDRdYnqpo2-R68k6#PIGJ3ZoD
zq}7H2HAO;tcFa|EwnWqX?a8bL#W{T~c4`H786Aof!1Qhb#IubR1C=>nwcSl?8GI=e
z{9WXfVm5x25EKSQw%nYJ50xJGqXX?-AIx~p`=Heori^Aaor1JsduImf3g=7B#*=Vy
zmRMz`CGm15QnBRMY<?%VJh_4G1p&9VZWw8qKVeTxLM3KdDq(Px@Fl_~6i}Wh&8go9
z@gDBN$+GMmq2oN0JUfSg`?7!2D(<6=W13U$10NT=6{bdlY5eKQUO{)ERYhZ1^uVsR
z^8(841DHnQApP*CS83rLNAWMBnZq`n3}zo(KS8J-?}d|?5^*0X3U4qu7Y-;WX&c9}
zCKmT^$#Q#MR-zE9QYCgzTZ6VP8a@#UJ~C_$G`}ciUlPabcwXqLykx4lyU0WnDR4ox
zCx9=-v~amqgaf#cjX6_JS7($S&fBktVZ&1{Fwfa%=1SfPei)AR+QS`i*6QqdVjQo6
zAI6WmI(<?W;M~*cS=m`*v76(E(R-EjKAyf6G%?cJXNPtKF4*<HH8LsYI7iL9n5m@s
zh3xubkosY83n~Z{xil^(4TLti<%b+0?wJTQI-@2s3hYOA90fUc!yx@^8g#g_+uG8o
z{b;lAT!aSW-;tJEq!W$E&Em_7?5>n8hB95k>`y(ONyY=z=SfOb)?9*t=P^(~0{juK
zYGL(S(f%K39iN!rF&}5>Z5N{71y)V@CYo!u+xv{)YK^}z5%59e2)Rf9?fP7!mu?n6
zs^=8-CVRJVL)cF>dPBU%wkH2o@R%qYQSHzk^c=pGT36vy)I+klbC6=zeCLH<7+!Bz
zx&o9lx*y6M12sk%C`tmoT&)j@gpiNi{$jGw_9hSBpe+p9_++CTZ^cJ}AQE-7k-_!H
zoDYEHg=&cXmtjl*U|~()_IymuF#`J++qmG7_N#*!3?nF&@dk4ufdCN~`^FT03m{sP
z_6+Hf6iIK;>jc+-XF)X_q@qE&!5w5bhR7t+6pEr3_i%PsO>o{hdOX9>FH}}R(4(af
zuZ*>Nh?c~r->ZC;Mid{4yF2n|6A!V(6|sfdjz`n+{VY&AJ?OeOGEd~MfS#Pi7<?Lr
z9>ghu8<senKNu&TetZ|5DGvp?1N4G$(AgK>tJG`Wmr2Xi5h<5R6Zt>}`}E%CX&EJg
zzgmhd2kpC0*mhRpt;QZ*Z4BsJxR&<-=_y&4)vtb*+pbZ4-1=4$R!dAtV^kJM*r;+X
zzN~MVkVnz~8lKI3pDRcnl*d+V#7k4!4IW>I;awK5lSDHj8bMbCcpe8Z&ORr*YA3Zq
zZdb%prE$;l`eMmt89JY<&In)CB3J&rkFQhnV^z|N>WKCQ@(B-5ufH^F(xcfqZ5fHW
z2V3JAx0lXe2i$PXNQSu<Ju01JDYb+%Ja4e!y(qpE>HdQEMjxHLKu@TleVWj1YX178
zZT4;Uo`ksxp_@mv$ZYvoO?m)?nefwIw#TcntLA%nY<rH(v-9;_q_`8apr^#z@HO)A
zN*4k78c|=?-;enLm5%>5%r=~jO=j^OJ*7k=TgMfgQ?7dj*B+c}-E`IeZAJdm>P*3H
zdXGU9M2yQ%)g>JS2)K-JLlk{<e`gMM3~2$I7Ovf*<B*li_lTj}7trnt4tAxmd5DZK
zdZoU|zPCfEy-?{HwM8&#R&a<J-rYC&G&I=Hk)eWkJk;`B37M8NSX%*VOd1?4zZCqb
zRM%SW(8P49_U~bQy7${EgA;zTpBusU=|#vd0Q*9d5mko1x8%uBM<yg1B!Avmy?Ehw
zS@MQk?tFr>iL|Tv47gq@ovPxn&XhpcA?+Qq^8rVFm}F*<Vq%3)u#fO0#R?cnf;n~{
zrOgeOA^ypRMki^DA~RYG&Y<Pfu#X3qbr~ZVP<*92;yoEHNo3#iuq(VyvvRuRvxXmq
zMU{<VHXnqBeyNFw|03xZSW%+u&bplGOB~_Detn6uNx$()@MyaE(!9V^m}OR`BZFo9
zxMbTA&j!ky@)PZy#I%e@E`f6c4K&XL&pEvz!xM#rPwe%O<g@)7QT~<LW9lzYrz9UW
z9>YYq_MR>`f*6-u)4Pal7rm?UW*C#>ecapVj>A>I!^L{W)9dm_>Ex<tk(k_}vQ|H;
zI?-sfx+XL?U8m^D>*kaihxXEc@j6qyI&)?dfe>#!q_IUjXkRk|v5Wkzc06>{RA^oY
zutoV4;4D32VYociGrraqv{L@Dl=<6dIsB#*M*KNdo#v;%C&A%)$Rl$n`5z1DXU*t{
zB%p(x*Y=kEG~V76Ti?-Lt$_59z3u+%<9sZ}AHRR+B|4z~==iCx{Ppj|&42nzrYfSU
zQnMK5u2{)c(2qEfGS3u12;3Oj=;7BmBO`LHb#~ZqcSZwslYmAu;M7#qV*oAeQ$4^k
z!U1$^Oj3(-B{v3@Z=j3^DG{+1I=IK#`#(4z|Cf6_!EaLD`au1BdWP~Xb#Qas0ua2+
z5E$IvT(<_2TB>^a^Zh2tn2hzltmS4&<ckU5?_G)#p4IZv26JzAyX?ViwEa9AlSx3F
zg$LKs@Pc68_p>5sZgf6-jkiKr51uf(gZw4?SQw}$cJGnkQSBOk(*Z&Txs9qng*2J`
z;BqI(M;!j%@A-!e58Rij9U`E+lqkGk+GGs`K7HIhKv1R!kP&Q;%VP|DxhG10^o-U^
z;BRUo>&TyWmLCX<TM{`85T^ib)aTa58W7shp9f*zg)5le8<+ldz2IAOkU381Eu4(&
zuw`eU;QdGW)r=_67VKsgy8WFG@lUCyND98LEes?&SaY5*i$~dTZB0}1&~JjOoV)P1
zwY~wr*yDRx3vlHC`m-{<!=DT|VQdY~j!tcR5>O=${hjSW<n~B%QXv2ngt~~R)&N;r
zJaP_!!jSwQDeyWXTwQ`-I3xDwnfx#G{r~pIP!J+@2gM2gerEq;g#6z>+arLrUJEI;
z{8OLypEuXv|5gHI6~(4mp$h*$Z=l8-)U;X<waUM0g8y$He6m1bP<^7*_IFhKZ=vDu
zR}a3TB?49it9QP}KfV9|{fU*wHvv}_GPur(9S|yEI%a!cCqUz%#?O1w#Yu5o@ONFq
zWEz~^t}Q10U*C+#C*+3P8<=$)!CQYIA)H*&0u<_&2WI(7>w<v)RkIJ5gTSZ9m`nh&
zDrSUEbcjeEy1fL<RCzM!F6pX;TM)S;x(DpkV$|Y(5xPBw5Ab1lRtMw@MrZ1l)hjKa
zIj31sq}BLk`1UB{wyv1_o8wL6y;~sF_$^(6kB-_06CQ&?=#v>`!CHj_?3lM_UbjT9
zCz}Ltq2R4nW&tZLJLp>cCIZ;41emY?qpyis$6BQDhpPAltfDab@8|y%9{=6iolI_m
zlLL#vt_lDeB8EDk(fLjH0rYdMPkP}BI%hx_Z2_^AJYN0%H)}u;C3?_EgYTn)w;p;L
zEf3pYXl(`J2HdZJ0e8Inwc=ezdA+C~YJfrHv#;NH?Ujy@f#zXmf4>zch)(LgB<IGL
zm_Yc5s4FmbKF?+L+;bmLWf}2t>j8=MFDXNm2>!4_eIlqDTY$<$`DM*$->n}g{JpS%
z$V#sYaAL2u41kj8$TL>x3c$?hU}S(%)j>OHE2uFw(48cPh<_Vx{qMzo?KIzxwrMgx
zJIuby0ea4IYVfqCiWOt&{-+TB_lw?V46jMQ56I*eV@d%ZB}Ze8A81kT)Kf1ug0e*z
zKs^I22h)Ig9nfCXDW0i}T$AMiE%yR}h!q(Cp~H7d-a7(3D-e8#p*AnDOjPS4gldDE
z0DbZSDi$#=E`6JK?N|sydX?<_GaxCb%9a!&_>SLHN-B0jTx?zq)=&_Ld{1}ZZ}lhM
zz{ez9ZRy<JQ4lx~r`}NJzr~T9Wh#DLx!-~rB;;|4oPXCqZx_6|yE+~xaa?P>Vj=;B
z4Etw|w}Jv70{YlK;IlWK=Z|L(!<G<!5hkCp2+qBAK6nHqH=B|AZviB;7rh6daC$Qi
z<98)6AqL#z8dmk*GRhbbT0D<aI>Z7?O3>%K3s<gHZG25>A=EIBChdjy3KzAfzpw(}
zs(R@)as*ye*C=RWBhm~JpMrkrK&{iBw=)YW^Q{F~_6RcGcfjGA4Vc-xlLpA%946pr
zI5-22I3_;id)jT#ww31*Y2{5&-pa!S%^|H@$yN6uYz839J7%?TLh|`-G8?D{S@0o*
z*JL}W;x*ZzdJ+XBKxXtWS$O|^F8=+Uue%_tvsizG6t9h9WS@H>tIr6*MLYXpI#*fj
z2=lkE3JV;m2VP5<?X<9^MW%O^nj+{X|2(g&2jdN^niMdxCTj`H)gifSrfvz1uNV-u
zq(E#d{8HfdUsS{*D2P=r-Pb<_Nxx9hu=v3ah>r0MOMN?l1ocQwIa${iW7@((03lBX
zuq{=u9^j_wodkHO&bR<Yj@U;!J;052GSkQ_ta<+%p0ZY{kCb4@2KWh0oERIh!)~@#
z!x*{M-F_8qmz84ZJsIS~=)0*v9C_yAY67&dBI-7uPJy$20B9{tDP#aI55#anE&IXk
z74imyN9%x?jQtxQ94sXARlpSi@<mL4@UQ4)1y4pkjUwihL-xU5#bK-4_f7Z-*JOi%
z7!^&(LL1)*l8r6^fzRz%&eGRuxE1Kfl>)G+**+B7Q&}`#juxdVxC)=_!xh=^zJCia
z9>H-sk`k0?l6VFTFzh#mHA-ig!)WmR++HUx^c{kYV;iqpq5(v#NbzfM+2lT8sl!-u
zlHWfz3Beq(&lG$`baOf07zpnXX+5RcO#*SL<#AsAJ!62d9d8g+h6R7FokY7PzouZt
zI|G}=yLMIwu<gkYp8=_Wn?{jQTPPn!t9_<b6N~W=k&Js<c`c*~^RZ|?4SFDJLkBQ)
zgvPip-p<#zMcUf&ahEKGW<jT-e3LBpL8b*so;26n&2kL=Neit(LEsVd^Z<s@jb@rz
zVKw6Qd-tzB0R6VCE=@mG7q5QOeF)MO^xvUr03>aC5CV&7!jl^Hw8(dTQ{#px-cTpd
zW~oV;bjF#*83U>7P}!s1sEponyF$4HvstH5n#-HS3}a#V>+!m=72NHEyPG-HrbR^E
z#5#q)|MZK`hsk92h1dB4T2yj_ON<=%fs=gXAose=udg!$35R5K?jXQx?+%JFo<urT
z!wKC{Ik$rhNAAHsRq;TNay}}Y@@|d8SKuPp2C6X4VsLRB_g4F$lv=oMoCT17^Z@b2
z)@iMv7^1pt7hCs4gMkJO&ULjHhUYbnSh0xzF~<5Fq(XdsX(M@Q;LLv^YEO7q7X6{V
zsCZBw53;$e`r<1>1Q{$@bQ}<Bqqv$~0Xa1QgoIIv0NFmypI!d^2eiBDzz(1l_;B6B
zWh<v~2hh?2kcvB&AkaMUcbrc6Iq3z)%(C|AaUKInvO~}mt*SV#e-oAsqW%v!vGyoX
z#5-gOPQeav12oi11y-lv<wpL(#+yrPkRzk!fBt|vEk^^Ti3CPcKP6tESE%nGCUUmk
zrC;%V8SJh|zwGyO0rW<w@`28Ype#CQ1r!D?88^oa86(+OIcbiQ?4U&v$W&BU`eZ@K
zcDd8~>YxK=C?p0-5l%I)u17ika9)2by*N8K{=SAuFT4;yK73v;{sf4q+~Txe*HmRj
z2r%D~_N}ak+gih+c2?aKAYxA?amNc;NRK(B;Iq#?TIJm#S&+gipTlIWYLd8boo(5Y
zcD*l7RtpZ#@%t$RcrTnPqQL6ossmlkbRB3cIwKyP5mL+Ja|r~gQ{FkeW`WD~&)Ls|
zi5bx*m4Ap_Hvm0pU1K&|FpX+gWQ_-qanuIKc*dSedp!ZPl|XRjSDW&XTJe`hnI})K
zxRqu7Fdm*W7xqz1_e?JRTJ&W%f<ka|i^7IxtlffRhf;~Z|07_nFvI0P3*gkwd#q{3
zO&M2c7Why<U-)VP=cca-)&y4<D-Dk;jsrRFbEkeyIjE|*P7L7oQ%r%ZLTf@n-*oOO
z&Fkhuc}^zz-5OMDgdc?rj-fDE@xp}(;Iv#BR20U0_%&mc<HqCk7sK(+6ptqhtmid@
zv(+!q>ALdxc=J&h>ROI*`QG0ed8iG>cwz?n8LV%<WGrjN^P)39fJ=}y&o}sA-A00V
z>lL(;7G2Taf|s7h@nouR>BQv_Z)qMsdJ!OE+gP1kJD5uJ4ym60w)3fQCWL$B>Nos%
z6!4;XT^oM3vU$w_BZq&NjQ#b+X+HA!KwK!C`-&i{1Dq|S4U1zPfFR?-|8WbgxR$JB
zhqsYHK8&5j9mM)tDg;bJ)r_z)WIRSZ92F;Pq9A2LWG$smf~Cic_-2g2oC7nMpxD><
z;X`@}JP0R3I3vyhf=kO|JwDyE;A@jKhj@*v=dp0rKa7qOkqm00y37ZH-%K(UBKcx}
zAduBfp?FRcaS+7c;vTZU!^#2Vp56<%cb|}y=d8<`(PzeGexoMn<9Fh8K?6}1U4GZg
zNRsezzY>_PE~+%4z64n)yAD9dj?Dbs0(7Vz;-O~|iF>0<<AqRt*R*LP&~guKBS$=0
zUH=?)YWoYM)gCkyqNMuwkl9sRNV&}uH4p{BP`1`rGAt|O1lMdhVKhTxNz{u13;y_Z
z%sf-(TtorXtKk;*J&7E%UyFNMtAi02B08gRmv}+CsX5=p;cqn}=Z~%0TL(Zh&CF-c
z`}3uqRI8tYd24Dt;649z9cD(FZP8A}pNLc~odp$>T^G<R6X6o#2L%^4+m(&i1!1hY
zVWWsD-*eA`WpPpMaG+YjYG}!c$)0}YJ5@`OmeF`h6Q6=8Hhm4d7|18HJ-!T&$WQeK
zj~#w#{N?G}5WdFLuZXMyFB?S6+sgb*=yzPelQ=Tj^FgCn2P|z8u~+GhapI~jhe{ih
zl4?VZwCy=jv*6?uh!8-+eXRF1=E^qQlN=A<f|@*Z-7J`m2_kMDIm$apHD-+WeQ@kI
zQs7p{M){NBzUqcxr_fGng}IDAnRJW%J?aYDeKJvADUTl1^j2o;&cHec-?jUD3h2|I
z9f+Wl*$iXa@vDK_m4F7qjc1eHD)7cTIg^K4Ys$EZ-hVN<uJf$&;}4ZZ;{#MoF=tB2
z0lDdaSo$6xO*-%LOwaebFjE^4B%Ml-9B6lVsG7q?KvlT7TUinTS?Ql$w#1b#l;EvW
z6x?F8_`ml<|E0teA;o2HR?&Zc8-fAJ``V7|8>067oAb24kEdWq?<-~fcq_hC6V|SK
z_{enhS28!}9m0;h+9TL@e#wdRMbGWI-=$6%e#;q#x{*hAkQ<W^^m8zYhE%6U)LsuW
z_tDX7A1^IGBZb2+(0td2^qnuw_4yXA?Ae>;Z2|9(fxb>!78;)@k$k)$2HZI_=UVAY
z@KbtN!=}VPXv=^3q>{8|^sYsyh>M#<%klP@XvhSA9Amq5!Ko3ZrSIDa%?F0E^F0*u
z^DaQ>_mI0JEwPi1Lnem2@6r1g+Jr7c2OtliGycZP5iPCn`x6SyH5yFn<mBj14#7aI
zw@gOQ+~(Dh9r=*hy$tpD3aKuaHO1M`_4Qo_NV2Rf`-jAnZm7S>Mr%VFPo(2}%2)=E
zbPN){<+5MoBg%G<rZ-lOnWo*q)VIf52C~O1u--24yix=xx<1<GXqJ9Tn|#4-v9n)4
zd&tOH;Btw9pLcN-?a3v?zkVj$6q~3k$Z_!c(KmYM!y0|D!=VNh%O@=#8`_xlT<0tn
z^IvJn=k!C>eu9=#Mq>65lzHbX&V}P_1HEMDQR*5;v;U6}H4mAdG}}_3;(ACQzdMl7
zzV+UNyT#Y}u<rH!!L#u$nX(SqCuxb!WxL&Bx%e9?gOcmd<uVfzgV`2p)yA08TKIAH
z$V<rEkSaQ<IGP=<HZ$l^<qAP3YZ>&U7<YK|)b+<cUwaglbNz_UNz2JPu;+=vneFHe
z)zfbo@Gt+0)E^ECVrGuPtc>SZ2QD9lih_3p`UkVN#PKea$iNEh#n;V_{=ck1M&Q}T
z>Zuj0%pzmV!OoHAiV@SHh1YX8RpBzltj}LcsjPD;rd^isEvD=k?LZ#g%cTe6Cg^KD
zzS=>IJ+|ul@!=mq@n79qQssy7&hBh|!}JGaVERhszQIRA@nl)7ai(_jz=!yo#%)~z
z^2X0NX&60TQTEhig^ibHaYNmMwJU&L8_H}LSNPM?nj(wIKYL`U5WTm!GltcUbc-ev
zJ}^%CGhjfnb(MdF3^31B@3%3BD@XWovNIWmsNvWjh@V31Y|3B5I%zdKZ9{-84pjp1
z2n)g!E_13oPY<6@$6*_P8;~g7@YtCa8rs<I+i~swto+WrJVOLhwVoA5Z=7AoEQRib
z$C3?GOs6eBnFHP<Mr#$DGPKO^G*(0ZA7^I)Rb{^Je?*Xth=8bwNNp6PK`ChoQ4tXd
zK|(q=rF4p*f`A|$5)vXM-67o#(%ndRiTt0}IrpA(?>%Q+>%Z2lHO>s9d%y25p7?%0
zPY>>Kp2UElI}#k~zP_L|PgB!8)F*Kk0JBQ()h*g{jz9`=zh%-$Fma;&jxw=*^zSJ3
z%}wQcT;J1J$otoYPo%Ob-E!JuWYRUKE24#G2;-%qwmnb$DN9)!z$o{+jyS6<2A>G6
zp7+h)EZrAvXv!H)6MUk!bZ%<ymbOxi5cTqPESBPjHBO6AXpMD5HQ;6nasLzU_TLxl
z|M*2LB`PfQIxuNc-3i961}*mWFD5mqVbxDT-cwTYy*eu0YjASqU5wNfOOFyKZWGqZ
zv||N9hW3X!T=OeOP)J;4>v-^Tob6X3*;gkSmf5ctEf<v-$8Y$XSjwofj^7}@SePV(
zr*?W!!o==iA<3lp><pjTyyo}&WEF)bb+2_3Uoz_s`%c<qU-m9#DPs4r*!LlrhiJNp
zg;9Ti{mdnfiV<b~k<5r0Is=cxyQCo@>Y+#4+O02#o=#S)8#9ZMutdm^ndMmSzdckQ
zrhY<D2-7q2@UfcPrb@hSkIYP5c%mk^{Cvz$;IW$cRQYAwe$MkGQUZBIYJw4&Oyf<&
zwhfXV4kHPZR6`ybqx3DOH5Ymdb3Pw)PGG_vEe|e}mf{SbsP8aaru&>4CA`S$hwlEE
z6jMRTx*}jSxA5}tnc9bF2)q(Uc*<5yhjCt3vMdDe6d1>ogdoCT4(5=RQ!aquNejQ4
zb7|>ot7RgkdJv#;Ya+X%Exg}*E949=u6SA|nP8{b-LH=E&7|27Ts?1p`0LHhCv&)^
z>>I**-jqLz(dJPNqtlR$T9cIpG5WMTeYO+yZVc*j2b#Uq($<@R<shIJwI*8<jIr3o
zgjk(iP0r=bsZd!L;+7SVWk?Hm3j02bkbQ8&Rv^k;m~YBoptH+ge^}vYgn6dNF!M8D
z?$%SQsD|(h>TK9(|4`B0`Lotf{x(a2vVj|mTO144sAWCnz!w3hdhHLboIypZix!Uq
z1uQ;zCSE0_EHvR8j+(q|4OJeCz9i@@d>Er(Hux@<dr4uL?=Z{%&0-#=LWbF_kDmZ?
zNC!$7&ifA*fQ&Y}Dx?G}AAdRJUL_&N%p+e(5|*NUD83IngyT0gr3p?vzi9HLFtW9H
zVTwQPz`y)jh%%a%I`0?ehF~~E7|bE~mNSHsS%y56$_#OAwihju&fFyJ2#Pv@+`X82
zmR9^t&~nBOZfLmj=7YEBNyDU_&a)S>WX?z4QF|^IZ6n!;Pam+Xc_I@o3KjandFl8N
zm=Y&sE}|utRU^?w8E{(!DMf`Y*FR-2Eu&<`LPSN4P>b)VY_(H4^$SXgr#9SFH!eY(
zqOB&_Zt44;Tdk{Rvks?l-9kRbw>R~#H~smBa=oPQQ_zJmWZ!#t5$hWiAD2EwjAikA
z;Tn)u9oi!%xo_>AJD($rQ*fj|;E?(#<taoJL>&_TI?M=nl90|Se7nN3wiN4}7mr&3
z=eq%m6>y<q^T16}nCEl>AmRTjs`#J(*dr5r{lqxfH@g>vTE$egL++j%{(ONPY!;MP
zw7oznPOsm1jU0qKB^xtgDqGTV{jm}AzFX7bX>-3&XZQ&Ub0sDEjmaun;ULq-$MazF
z<RE#LFv6w*87SNRn~Pi*ECIyQ#qgCR4``0A`YKu4Y9RM*j2o~@E1HeCFruqJH3YSo
zk19%7cb&y8jC=F5Llu6OgX5h+UnX-&nFKky=NGUmU>1sAa6)T27gJAD;7{~{m!%|@
zr}R#8QZ)bV`B4?5^jQWV{{aK(BxH{@*hRMsZnm&pm%?f-!84obZjZAwEBoh@ci8Ni
zw}<0*v)2oA5JA#Lf@hWsR4X&r-73O0B+6d(Eo`14zehPSVg*N+?7{Y8Xj!_ql(${G
z4+&hhIfcunvw+g@>Qp$Ib}j)wx@4bM3b7SrHL=o28|qq^HR&Jr(ELZy^H}dMbEo`h
zM6l=t`AkuoX8le9mcfcj_k@9FaX)EK_yt*rW~V-Auf94w%)ui*k`Y|5F_ZN68o7Vs
zHg@2r^mAAdL)A6Ilh>3vQZ7v5n3W)Jf}A|edqEefuCifk*EH9O2m}yD)&|gsZbmEZ
z7_B@EL>N@Ju{=F9OT;=^A{yToy|lhf7?)vu!bBAGLEe3<X;{QA%C%g!ITzAdKTk<q
z;)R@@J}KJ4*r6cw_9)+Avh$YJ3C+|&JnP!w4tkh(FR8M>Tf^>qI&Wh-%hN?$jRXsc
zEk}vM5BDSPCYuPJB$-cSExOS8td@eVTosf<Wyf(3<sWVZqCSG`dEf4-`0uv^Vq#}k
zadMOK={vYDA!Oz;YDLXyUx6ZEFu}1h$IJOuu(dw0GeDB2kss9Fl6JCOww3DvT?Qn}
zrRT#<xdQzPMje{dpW4vR9|M;lV6@E9QDYD|f}PgrRz<|msV<8YAq~C?{Hf(8iLE@Q
zRYhtd7>vELZz1DZ@%6DMb_Jp>l+<|eNdM$TTVgCN{FTp<W=&LWN0gr~u?;g71zvlu
zHhO2b8s5z0<+Gia19wzNlv>7tPZXs29dC4*35(^OUdzou1h8LTvLLDE;ZOK5Z}<rs
zd%$MMH1O7J7+n##uyqtJh}_x*<>1}0H8%e9t#RR7v<l%s2Phq->1{nV$R2m<$9dP_
zAEQ9ZiTo`x2BKSVS?LyKvlLR9C%}pUFjUvC!VkX#WEjTyHK!QD8SS0tuD#4>AaxD>
zIcl$lRS@U==ohN205V!#;3(byt(MjAiX!O&t!lzQ-ekFvnUZB=IcO!W39hER2GpG0
z@k5aqLi@vrOV5;?cA;KFk7Uu_SWu$uzG`aEq0v2H5Wme7`R4w*4#^%gIOZ+Bc$R+?
zPrzL=RKQY}h`P&ALO!8Z`pf5L+ZrxDkjG@J1elLFkwVDkZUB(c#zt@gXEE2~qLQXC
z^wg!YS^fHebNOo2c#gNelyvi92h$17=sCmxpSK$0v~lkSwAinri2LbjIPAfutj^)V
zJOKTs2Hcm*(UZ!^(2>A+W&FjVf(%uj+iNcnjMjI`;lN3kRWN>-d08{kn}@fQO}wE9
zKp<K{f-D3RqsyQ;+`^Zq7sn%_`l!2EnZR^NjLVYWuWi=I!JII8h2;v`Ll9K9{Dqve
z<s3DqY4#<SE}m?;TMyRmWhG+E4}}1ZvvU&pxh5{iy32rlx2q89=vSdBy>8kSb=H#x
zeoiRnlJH95RB21MUPP$gfrcf;ZZqjaKCS0)N|j`qe=VBbIJn^V<e(;nGH}?F5TWr;
zgMg4#z_+yO`~;5DHEOu3vZo}SjBvg0bW08(_`^mxDW}k)T^^nncN<}f5)d2$k=h=)
z0x6dEx)}tucP&;Oq0V8i?zceW;z17$C%YR^lXn>TI$aKK!sQNO?P!t`+B<J3_wN^2
z4-zQF{+@)eSqGC4l)olKB_4Yc$(e(#Zea)S3C{N*(pQVQn&IW%M?sSOLV$Fzpv60q
z=RZFpY{7U7R*?N=!&C(Gqr%^{YgS-tSIN*?{@7k2hcALx#QqRJx<t@1Q|t}FuvJC6
zP+Uu{;+;VjYurlx#Qccz^3G-bMVnG0Jxd_^9SGpI3rI9C9JJgdUwj4uQsvWoJs%4)
zJ~_*iX%_$7Xs%D#XJ{9R7BAgg8y+VQaHjSoo7TEr(th|*PGXFDGhjE+fqaO}3_*d6
zbnO;?dwR$e_mF1IZx`qc@}<83G%<v@T8qD>2N@i|^w`ySK(ah9sQ9=G+B*9vO_5cG
zLV~t8rOO}p5Pd9W_`FWz1~f75b*=gf8bE(QtiZ`@TY94j+sjxluvLd{Y*SDQQUc!u
z#W^^gptXx5VNUt}&zTx8sb<3{iz_Z^0uJdaF6laM<O4$ba|8e*Q&UEaBKV50Wcc4u
z+75GKfNN1n-zJ^R6PjO%<1&FXDtN-F9uQx`h}{U|pzXh!avc8sqo&e$6&#;b4|E?F
z=YR19j0O(ry*+2sDv=f-Pq{oJoxqK}dX`6yoM+H5UQ?_>6YZ_})w6j+E2Ogu3X^0l
zgwHd*Rx$yx@fwgcLQ9}gPP}@Sf`<hu*G@nO_zb-oie3{%FHAr`%N(T>_fa#jdje^#
z%+B(J&G58msAyV-I7QfTAON_^v~iHRUV1Qo@d2O0JG%BaDqIBg9H;fwS;FjS`mdJF
zA<dg@Lq?Mo(~xj%dkrv0xXYZwBQ>_;=VN4VzQ4fRDaAo+lzDGnGY*x3H>;oDm8^FN
zBU)!gqpyqeh3h42R#CCj{rG|}`AQ}9j5K?h6;e18WY;5^{@HaJ2F5hjh`8C<K>eFv
zq^Wn9>=bF1-O<g(h)Ndn-%neah<^Ns1w~H6rIrIlWTbf63haR4$~5d#Kc=DOcctZ;
zHplJTNFnuAQR9+@kibS==g2Z$R5EmM_``5NYkvJu7S#rcU*dt0w*MNC*iMluoytiF
zDsMTgf~>#)FlfsSx`1pg&~bH<aS*W#%VQkV>rJWxuI%{-S15ZS=F&u$i1MSh*y(A`
z>bG_tra*6MU^h~-Q`3ye9sdpZsp}r;9&%_s;Ann)$#Q)@`Q%ni@Q=r(t|d;OoOho9
z3@@Yf!^91absLWLCll?lf=4*cmu5^t<jNkdV}!tld_*q(=7B%e1FrL6vy9Tz?$E@3
z!A{py>!GaicK+}I*0Ydf{qnVs=J{qxs|MPpa~Ly)G8vYvnsZ1?j!56DZ;aVaxXU#h
zYh>&i(#PMLDUV9Zm?%2)BzmsYaZMxOEI?`2ZBLBK8_IIDmiJBT$@K_Gk~q8aueM?~
z)0W4+eG^eF`~}U>jScBt$)%(RKxIL>9#1;*17kZ^l{DKv2$>pT2Oz{MMAvKBz2LuM
z-hDX5BWUk89{4Nl;Nf`q{TK`Zekt3U@JB-XL3^K6Pp7bk&KPz2EdrQlLBB$z&pL33
zpHs2C8dGW>jy%-~O{MEcf}=wAD`eym&hG=Dd>Gf@be#>-ASY<0w$#f$`3LCre}>op
z48B2A5%x&$MyHx<abX3^uOd`K8qq$^5rANbTYiL%1$sTNapHta2aTpDBiXkQc72e7
z-V;Z?y=wyf5TDQszq~Xn(&F~7a-y%}kPHv8B<<I*y%MB%jIX#BMO4G9e=m_m28Bg3
zQu9)jSnl2#5JveD@Y2IY2SYzd5<a}|jF&1=Ca^rYAbQ1PyglvQYqyhl#cS^f9ClYI
zi7hUW5Hp}1?&7T!!3sNI+jy^xV=2#%z(TK8;va5iKB$izy>sRj$A{mXL)1;@(k`AE
z!`^;1i<cFICujr5P=v%9=;7G+H%`1?%L(vLcwFkcWC9n=3UzxTYo~DdxKzbG3P}a`
zuoJhLs$M<C7X2!Pa`z18f7fy0#2GPkjYf~{i=l<(Yoj+0RV1#A3<!7a-yZiVBwD}3
zLOk6S0^<iBv6aRx(*(%iNwj)0&5!-SN4<^@TLt`swVPS#SxatkacuvD*5|Hi4FxI!
zKPXh?@wVv!T@&LQ^bnpYfhC8S*}=nu<m(T8c1`+ZU(8%?(bl;atH2sHP2iSaRF`IR
zzxyz9nSod(Lh&its>O?gOpw;wf2*^Vcs^^~eOL<J9o(F*Ml&)~`Cq8vFMeBI_TtgS
zQDwYwSx9FShWUzwachwsumAiM)%i@Wo`k#0Pw}+<!R4IiKb)_vZw$Bg>eQ$QJ>gFn
z!|RU;mHhBK)wdFk9QbS=ONEX|y>MMxITE+*UD8Z4sb`*j(hruNP2pNG|Lev-YuR1o
z-UZmIT`XR^BH^jd2u&9qHOU!w7{|=sb=W4by7!7=G{fE|tKxO7;Pi)G2W+u>qXMa3
zhdKwBk1#v8F0TeAJ*?qFMAschuwIvb-FS5Re^i40Z!;#=2P>o^uHm#L1?g)~@Ba58
zLxnGojA)-dg?gZ>wB?;o>dX=<iIKWI#8R|obY-4GVP~$DDet0vC<XF?Ff`}xWlE3k
zKB0wsu}fQ&xI)6@42o@Bo{^Y(VGoor>U$MOx5&I%P+!BiYO9D(c-tsjGNKcj8S6q@
z@w6FOG7?Ur8`Fs6Re-nOJ@i1i+WqDAc#>xA+1|DR`8(4ep5I1QQ)6<BWXB3Tm*Z5X
zW#kj{Ul)E?!^Pw4Gky^ya`6HQ$#XUqBZl0OH10&>g%%1a+dMyvOM{3~q9Kv8iaW09
z2OclUL}S5B?R!FL3<*QeiU}=P(Mkt6Vh73x5mu3(zZ&4ytEnm+g4}@@TYC*$n>DZf
z65<(f22Yxv|D=#ETAtz+XsrHZHB0{xAIAk1iJz+;eIX(ErKo@LJ=c5R$WOMDZFt9?
zrL~azG*GO*@GHTOU-FrFX>QW7!Kz@zEK`bzlh~Ou_uKAoAgn7REvD&`yD#ysi}v;j
zcPlZu<j1eLe67ymEXce=;h_CR=P46T-+g%~vbygWT5`DTGPW{W^6aQ}(`lfjpe4%Q
zV76sCd@Q<pPs}QK=g_!aubo@RM6kuYG|(w!-Hkylhp~j}Zu^{;+*SmdRi0`nX1241
z;gvxte?jPc$xl>5*?!@Gu$%8^cB>#|Qmbh_eoFb^#pv8I-DUH<U2cK4#?%G}e`=Gb
zgKzRkHt{zv?wnk$$bG`3kYt)k?Q}6JB1)?JIoA+F*Bmn2sAC&hWCuKt@c*>Ph_Ul<
zUcT1-`T%~A|LPh%hyl=JlSLkG+ui%YAwKe#Vy<HuzhTGrkpA%rM9LQ1hba8|c(a(U
z|E(&u;|Y?7KY99ed&vxE4Sh?i3WqM2WkLt=bz3OH!I7rqjla(4^HQSu-}pkB{u2Dq
zP(!1BbG^G8(UJYlUmY_zfGT$>fs&2sWQ~^%piu@8I`hO4igU-E`ey|qTw!C6KTU&O
z=2Za#pCyR2*cI_fP!r<yz7z&vGK?aGc&Mr`#H_>Zcb!JmkAY@rY%2261w@JCrXdY^
zs>V=mO4#34aYUcdBI2?kn7*SK$RPWkNtUzr{JIyG1*A0vph5kX5&K9;GFBcnzOV{W
zDQj7(>+vrh!wwo%ko938vB_O^*)2&l;r~niGf#=5d*~=7ka7|E_8k<+bO;JnMeB2a
zs}%xy{F?AikG$+a-!s>d&!ZqviV^&f_bHAP?eA9s>)RlZUWQ3;Bi-D&84<$}|02A^
z5o@J^_tzsXd;1dOx@si-T>@zgJ5azYa8&h95&pXu@Sop}^DB@#I>&`8-upuZ=Kl(w
zen0rpJMiE~d*f34{`SHD$Gbj;1~jlShWd(+|Ip_C_uC`30LHy!7UX^TLHD;S@IO6(
z4hCf5GeoEapa0u;_V2$vbm>iz?Q6=Ka`4ai<v)GbfA{aV@S)4U<E5{3?Jwo&zyFrV
zPF_SIMx`CF^9=wld@5t{ejG0ZQd|1i+1zDC){&`^2NExaw}MbxSFtb82_@$#D8ar^
zLLQHzZg9q*-i|2oB`960L#-np{+J<%2icj>9b9gpJ*q>BQ`vK&Fis1$>G!6#?*~;m
zkmW@=gZxLY02mFzJit-Uei5~;_j7OE28LLKYV>_`bsYj!>C)lYvCD#*xzP1uq2jJy
zxbk`tI#>4RrtS<F{BYV4q&w34gj9n4`A~Q7+8uR>74B43&0p<J&6d>}gF53%IY9z0
zH+$7i4jV`u7?yylAL<N{{W>*4P(AiTZxjP)_X1O>1ZF`Tk{7$x^m{G+V}?rbe-Wzx
z;|mfc#0C-ID>z_UfPJck$XI<#te)cV(^~+<b^tJhs>SQT1<rzB4ZR_v*5L0Ot@OYr
zUGf({n}xvjEr&oBYwie_9|lghtj@8~ob+NFa5@b_kKY(1HJYBpNnzTj54SU1O~E=U
zmZ~vey21L_1A&q+=zK#oRVBWO^&CK>G}I&3o2~dul*_@w$miv!ip1OaIiMtNMg(gq
zo=917$S~d)1A@m?hq4O{=D-=xltPS(%-O5k_Me>(U^+)GKs!e7_+Hink)K9JhOL1T
zN5=B;kJOAM2>S{s%u72#IDAd#W9Qhny-5N@<)>46G{f$i1y-iyo2iMmh~`jK9M(1l
z6>9c>LCgPZI~WF4fC#`8?A^Ma$He@MUZOcTYi*nG+lErrr%nUrdKoG~Y*8kw<KSe-
zhmsvT&sLE+2rz7*QZfJ<E9yvKe-1h3>h-ZfqCW&6n`QXF3qVswJL<l-DloSHuz7Sq
z{RZ3bhlc1oc7P^{Gl)L8Kt&U(tc_poO@lwt2$o3l`m%j$cw~?(7GJfh<8g-4E)UQ}
zC30ke9I*q-aTL)ZK3q5j>bPiANNa5%ZRlAx%9B6c>#h;7vv4nvegH-h7(jh4kdXXF
z0^ax7Lo)7+<$414O;F3SV@%Sr?w~SMND(~^=dZm^tUdvGyP>1xR1Y|VM||Nx9Z8o|
zp!&msI<JQl?Q&q`2-ZxWN9jl!&7K;S(jAP0YTkuW4~1qYH1$+Sx@a2a&ti*~)@@r8
zsoD?4m-I)JZpX%;v4-JlSc1i$wv_Mu!%-tdC`)#H<~+M@0ss_vaCA_wyK|w=KLooo
z+Qk%bf_Vte5y5HC`>{H!u(yBg$OiU(7|u&p!@1I_pFm(YvO6xBjWEH0j1DtJ61G-h
zH>Zf)WNs@D;EZ1x=4Xoh(<y?i8W}jzdL_E)*rzWzT8G{zlOw{%I{wD(tvNt>Aepnx
zx6`j9KJIh3i1{3?3Bs1a$0B&ienRsEv1(8)ppi+jMFdYU{a$~v*WhlDHN6%5e<vgS
zYY-mWcMho{EJLQ7johefr|&dmLC!V^KKDyn_ero>ps>fiDBF+?na5Z^#s!X?Wl-;B
zL&1fWXC0sR7FYrLoOsIod5K1k;ebsXsGWnk0+%PgH$7BxqE#3L<(K}w0FSqHO^1H2
zKdb9}DV}V2N*MNl9k{t6qfDZV+GS765x}oY05;|kW0-6-gl6U$a}TH~eMC4DA8J9!
zYK|bIoflQdjM<xuGkmE<Mx#Mv)^+rKBk6%hyi7NY#4tkkD@FTFwPO%BEkU-zRm$1P
z@TGaTe7EdNGbmBDL7yd^!{}TNhjb!Gz()`ABO009b=HDmZM%Hh5B2crqcqrl20jX~
z_;eO()DfM%i2KDOUky%f^srH$L+%s*a+nYN+^HfszhbB#&H^n^fVy?cW^see@cH+h
zik+X!=B~_(H@8xuZjmQ*b7C7Jlrfg57&y0w@>GZWeeR|R^hZ3tc5K`6K!|A*rk^NV
z#TeG@AP$9Gm==<yfQ#7zoc_~;XsS|hM`+b|ts=;wlE^pZ58DkyBDDrEG2>+|yhdS`
zVs1AWIH=`CX1Or)fR)!oC`RI1)se|~GmK(b{quZtW&OCb#B~5iUNLz~<zU#4`&Exr
zLptk(q@z^xj!6}DcM=TxQ*wmk_KfN)vKoMdQaJnBpQqsIQ+x3`R1Bk>*YjZPYlv*l
zTnawcg2`49rb>DoaLz$O7dG|eOz3WyS;=%74?R;F;QuyFK36&kSB@9=je6uXODl8P
zPJOnX$HGZ{_A=L(CACp@Xy3d(f`dE{DE-+%-)}Ut=pbnB1p&N)8yitBw~xO(NDIdo
zKL#1;n<%s{=OSIN7c4f1mc`(9s{(Fq&@_XhFV+%v2~XPt+5dPe{S_^UHlt`ZiosEc
zM8tlC+X~9=d~baPJ)U7aVxW1e*jzEY2VM{}giYjWMNAaXzo%dQfv^s_st{LnO8bj!
zPr?2NM{w?GIN(LLdKiy@<v@Bag|ZaRGFl$dLeED;A8Mg1L8T|f@<y3Z3*q#bfLSXK
zG}O6Id3fwMbYZB4i-As^4?amX^dJmTOZ`_VIW$FsUK4iRY~!m2IVt_r0ifs=l$9kn
z_+hw)uPqJD{)+846AueraUL0Pb$+3=tRdNfn&lmPlE)oigDH+NKR~v3^<{%Lj8pMt
zF_a&)ZSbc>f{%iMamh=TAQ>Hk<G~DRK=k<S$dbkpSB-YEUuS4xdW~8rvlm(ThDT}j
z2-PNhv?VXT=Dvp~N3;FRZsIj;R+}8)2Y>zaWE&=N%T2jkBKDkqkVG1ww8&5T5FKDI
zbHW&iLB#l+k6t$CI+O1vTz%;(Exj;w9K;GQGu$}-2@Lf!Fl~|Z6tVq{(Lw1}&xGQ^
zUWHWIo7-Pnk~HA=b$+$WYjP;{{)!Qz5w3pU5lWl|N>o=*&ggV9>@O#Y1UYJ?`me2k
z&Ru>e$}a+0^s#Bkv``PjT74>*6U6DKl^X6H?wVQR*-dQHy_+m9lpOBhZ%%bWGiU1K
z?h0slc`PSuzOaiq)NH2RW+K%Vm=&n0mZZlCMT~S}vA8;;o513(bqteT>(7U|E_H%_
zu~#VEE35#)lY>}3dhZWxa19mY*iQ?6rN@%!d0QADK^fMq(ucV2^fH*Af8@>iY?!yq
zo{EWjw&C+aqjFCl6F9H>(=cIh&@RS#CkswA;T7f|)V;#6{S8pwje)}D(0ZvU<~cmu
zE0Z?UWgoOX3KNt9%dYjB%|+R2{ET(C(RQwNYKu7Qm<NV|{)Ym~=d~A?{8);+FK4BR
z9iSmqLOe&2care_6!j`&V|pgOpK7<n@MTZDpQETzv7@TW!J}BlJVLLfKCs)o1QSJl
zDFt{0i)ahO;W!oNs;lFRUc2K*MH!6+^LKPrjgu3Za!^@nqk;`VGRr;Q=f;rtB^707
z?ufr?6LbuB=ovp(%mF)?OlY5rOT5LJf*p0X6NMq>p!IB_mmu=I?gxYkVnMUK?@F{4
zI#Z-~ripZT2G^h0lzCzURX4JquXXnO5@%#1iVWCeTu6MTtwXlzfFkVn@H}m9jP(0t
zpkLs@5Jam=EkTUI>b4O%==6(QCOMrgClez&-oU*f3{!NPC5Y@Ks&vkw-;e#A<(;^9
z%1_DYzKTEDK0*r5()y^uzCp@KSVfKFXxAS19@X5BV^Dnn4FVru<H1wwDrLzTVCA|*
zommF=xL9e;C2c276Jo;g8GNFI+>=$PnyP;Ho)R*&2LNFh>ox$LW&l^Cw4H-^g-SSd
z%Ffa`T_3HUxAk<jf8v--{z2i%V8l~vqt8^q_TQPYSW;xC!$pW4=?e9)=(Jx7g*K=u
z#R(b>+4bM3tT)Sdh>KyIl!sJ7(?QbMUx$2@U0_MG&@Wm|y{~$|MnD0hQ?A>l$)-1|
z#U{ye_xLLVw>;5CD;*Zb=Huol9FzPKQn}<}%s6g#VyO>MzJ?|^2a@!BH`zO5G9`WR
za8x~|@w=N!eHpkU9^zfG5Jhn<xb`!+-2HIj`OLzQWN&f=mcNU{an-NSi^78BXpYv;
zadq}NChwGneqq!Xh1zbm+2iG-)~_!B0Fua((K-1>mXdxs^9{~E#HyoZpv+9?QJ2HJ
zxC3_tjY%oeZ|Jr65&sqA|JkZcdVjz3jLBV(Vjro&Ga1b=j6~7L5BiC;vy)o`E4g<b
zG$Zs>w~si(*RTJ<vjZN7lh|w__tX+X_Vg_#sWu0LJQtD7VDeQqPaod|g@+fxuXDRD
zfrVdET7}B-qY-?Fm$%C~vcPLAZpp1elGB~^@*=m3hMQ&bq!rtX7z%3&L<59&`svnX
zHnSbo{67?iRu`_s<z<^VS+06{5kv$a6YnhjY`nZrHLqB%@-`wJt2Fz!1mjI#$+l}>
zD>OoC5My_}r3TkRF~^!jGw6F+^Hnz8>m~K#)z?<IIs1$HY1be5==JUY9v2T;-m}T7
zsv6w4re9xF)!wRX5^q#2Yr98=qF-_TGP{jtD@-b?!F4swj*DIqVriPMis0!R66iGi
zmeSHJDZQ>S=95v{V>C7*|1_u)m&*Fv&GtPz=tG_(`%X`lfAf>9!r+O)YFu6|&>{vG
zHpIyk30|?E94%;SD=}m$f*#p9vI{|NCH+?|W&Jocq#1^8U5^vGyi97Z+?#TrWZQy7
zYEnDsX@=noVRNC4K^HRLJ0dn2j}x;ylNLzXav7thDukXv1vqXud-Hqw`O550;fc8L
zFW%m50U}2WvR;lqnVBO9@_{pQjjf+iU~Dv0e{o{8{2ZBdL8$e26G@*JH|or8$`;dP
z{&0NiyaD53v?i;XYjA_`Ux;{Z`#L_5R%3})gOn{1UHFKp$pLi2p>3A_1?Hi8Ayd+W
z8EOfaiiSbEo)xVVH7-`~!TnT<v7L`yTc(p~XiTnbv1oLjuYGo5vbV<d?gydojP-2j
zhd=SIYZ(}qb-3~c7QStr1}(VZ3vYsDou~z1FN>oYaxHm^3R<QSL)2tkSB;P{!-~3q
z<E{54&6l%eDy_h#9IJ(0Sp4Mg%Ttis&7Z;{c5j6S+wX)bF5N0eK$VN<eX3~<-K5sd
zpLimBw<n`tWcvkCzMvKp<mnmMYW|GgXqXKZGPBYNkuk&JK(VG=<7paUR+-e0A5_y9
zzhL>&@d4%qtB~6lVhzBLZG5$QLFT5KH%)~d&)Q#@{s6^}U^Hrku<}Wa^j5b!UeZe3
zzVejOX1aW(u*iG7O-TAR^grmuXa})BNYN#Cx+sR*=EQw$a@ue*SHkLF+S`MmHL}n)
zHl?Z@BZLfe(S^aI_*Ji(rm2LcBHo^PFYS4jn4TSMM_!JMtRwH!<J6f=WftDNb4L%x
zcM3!3tZ@F@R-UH=7o0}<vgfr9Nv@ah7G`rUQO(E67UDHrWzOFA58{5kbj|Zv;QA}Q
zcxUtlV<;M=`Em%b#yIAlk$ZSh7T>PF*qH_-Ld%Cy#qL-ZU^z0StCHY)i@L@9(V*3%
z?9+-7Q=EB{{mjvW!(hjX2au!(r{#9tUynaaR3GgG$^>h@n0k#1S)UDqmori~D2IDs
zA?WuhV#BPH>#x+U+g<RNquUAwo|i~f=&p@d(hjyqw^QY?Noc<7@GpN7UOaU<E8FJ)
z<cW`{PrSz3QG6Njhpf#C58Ffg+L=oe=FZguRrW<R9AYrcqkh=^Eo;$qz-;uOPp0F*
zw8ko0f?P|Rw_nqJDUqk%Bl6n$Fvrum;%9FqSe1DS-Vay2!gV?2JK)q4?gh38?I})g
zwLZC2Xb^*vJ7`tOfnc&{c=4Xc^9SOg`WZ>}9<A4c=!S<%*vr0!%T_g~!vqhKrf`c=
zk1`q3PZ!jej1V)vN88miu%<;TTs8t@BpP(YO7cRH@ier4_}8SA`!Hy$Ey#(w=y5D|
zsH*BK11#R9-(UIAG%T%_%lv6UEsW$SEVIGG)<(<N!||#DrtJ5!G0)Mu#k|?Lc?79E
z*RH3&huziE7eb8fdo&U>pGYB|BhODL+a<XZ$<ey?9dm4zVMvl6U@^SaYB^OGPVlG|
z_jEq^Lhk(X7)EU5%b+&(<!B#R>&qKek#OW1WPk}ZdO|-4Pkaoal@)rIc~s6#nU8^q
zk(|2+WVPDDaT$eV4Jg}k7=|a|8fUgeh<PZL%zPkmIVG4l3aY&--_hrkMr!S?hg$R(
z<z!KsON@)yV&{{RVM%FttlMO%d*INww>?~;<gLm*7jY{Rbu4XYxg;Gi@7^-QhQ+n(
zLy_b?KV<aFUZmSa^12mf4lHkaD@A>3f+O^5HtDuA<a3f~j8F9rO}-a=*z_Izg&9rz
zp!}AYrz9uPaW5zSJ%5uL>heTx&g$IpNg(3*S=0tkxT%z|<#ve8YJ6UXK4PnI?`)X=
zAOTC}OJr!)x~tC;;fS7Z7;E5r2<Rf{AZ`C$iv0WJCf5TCO+#d2f_YkE;7@$7$l%h8
zcAnYM@fLC0L0Q*>?tcIFixuKd;f^TI-85aytNhS@E(tYWn0I<(U5&)P$H6?FA_w!l
z#6V!v-z6{m%(^m+c$>VpD{=AF>j<JlTpVoA8&4UrrjIZy$V4QcR`uNuWu^c}kJBVA
zP(<t_z7fRsbG1Tf*~9<T`UUKWoi0$#ihY@k#?iUS=V)%*Kb$CAVo*pdMDr*?ga-AZ
z-SW=y9lqI7)@xW8QRK)GwRrahuTxa2u`D3M(d=eTq{@2A+Z}GA98;>t!yv05LkfPB
z*a{MPRHu8XfIgQgN2Sj<*ei6#dMl-HH0>@^q`kpl#FAr9y<|^3IYj-=ao2}dX@|rO
ze?r~nO$X73i?2ivPEK;xJO#hKfu}^4BPxdfzGNmQS2jV671IT7!IX<xa-CK=+h3%x
z8i?@rvR=ORVGquO#iB0+j6;PnHn&zw{~cxwV^z8rDZ?Q6#SNjwU>jahT=ryIh2pP+
z#yq2KTf@Zaqrv9#{S%r8;vZI><^*WBr01UQTA8+dB497ggwbn?rY?KcgB?OD^_MpR
zv)-bV^n(-a?4n!YCzqWpkVcX-)0qVYpiQJ~7#$V;hl;ve2#Kg&_3Rj^{mw^M^3_pL
zMGk=f!QO7I!J(%ZYJ}C@!${JPB+u$I+9xkt-idhEK-ZeOk<;lrAQ(9eqCgi5M=(C0
zVb{$IcuRC9OD`2>u*|ZJ>JgV^7I{D6=CN5%o1SnR!Yn<WMMmf1+_b|N`#GSo@hElT
z)+{^<6#It(#}bDZWb6G<Y{5Xjqk^I&{!F7(sd{0FB+NdngqXtt5G;5Vm_<dAwwuCh
zRd0xK+r_r&JyKPOUZQvxN+L1Y_dIz>E{?mMaU;`sN1R8)vE!eM)HvT3R62})N8YmS
zKi~3AP`?hi5E+>M&fqlP&j-4^GrbI@ciM+LnO!zOZ}pfHEDb&jAV}}@AC@F66w9Q^
zk5ZFRSmDx_OzZ-s<w>SKmO2z;Wqv7-dfvJjzQgPh@l(^i{0M0%RpL=!5m{;&?0>r2
zsB`G8ET7x7)0bjSH%CD^H&9yg80+q>F6wt`FjtB;WcEt_Irb6HALLWP__BRJV_#8+
zjo;3UuOaeS!&Y9H$xQ9@qzqQ=gAIL|=FzhR&5>(=>acVeW0%WC71o!eWb8w3O}e`h
zR9J|%JfSj4CD^orRQdxE4YK74&8o821oXz2?NKNj3#}7p<gpdpT8*4)gqrWztkaH%
zc?bhl9G4^0`w?c%8!!p`DwSD&Yu=WJqw<5{q5G4nOkUm{UEQpQf*?FPN2Ky`ug(N&
zn49f>Hpn1T<MQs>D{|%F&n}x4Qj0AcVb-10j(`fZu+fVQ0!uXY!ISi9s^N2F7T=3M
zF8(Ap_()=3(!psd>#Df<oc*QT9WpGXGL1#u_@k5?)Vo!eR;@|2XA+Cct4DF8y$Gxq
zi@y!ZutmJPQMlV~9GV*4d{bEMk$(#e+x2pcNr|v6V}J_8kZy8f$o*e1t>Mntj{ld=
za-a)nX9Ko^flc|ngMCAT1Bi_C@z{-XWv8`NQne*}s&UmDPvBj7w0Oa^Z7x|#E`JQy
zrE?Z+yryEu7M(A|MxnOpJgL1@BmRS)>KB(Kd%AVI{O_Z{hgwAV#{_l-&&O`kt=&5q
z{haMJj%v;)39IQlMKH_i8e11@i9!#bE8rwvy(YNTys?_b&SodArZ7Bfy+un#9vW3o
zbu-BkHQu!!cVNR<eFe|x+kF+}dh<Gtr6ULWc$MHnbhJQ=5QM14GM%S~#<Xe|3qN&Q
zMz;psY8L9(m$m+p<@ocVnnH}7@}kGK|MA_cmX#XWkPh}Nk!|hblBsd5Q<wFWGNj{7
z^}AYqDD2=6Qao7t!L6N4gEUhcWGbG|#h&?(?XUl;_nxL5*XTM&lr->_V-jR?-h^gE
zWhy%ld9ZC^C@W5<r(-yCiPHXS$C1Sc8-PU#RcAu`WuLkE?D9xhYfhstN)P%CGEgKf
z^YCqLn%3A5`}pcnYz|DIoLxbB2d_9i<dHi#lBXeL8<)(fm@HEl+By#fdsKHG4`jXV
z@v%udO1<_&T4=;kH?P^tarU&EB%O9h3i)?Fuj|$jy=4^LUXDH?o_U+xF#wCWlcuT>
z5fZKwzi)*6OU+&MFiAiaYY$G(tAJQEfw1Ohg7^eFPQv5l_b$LIg%lbz>B2E|4{<Tn
zl&RL5CVhN<-}8e|jm>Wr?$(rpd=5Gp^DD~8Bzv`$icHS0?L5{+y+e|(U@X#wg-1Y3
zze(=p;=wuuJ&bCl&fOm}|F|#GVX_j7@ke9;AEzs#W!>gc`LsK80r=Sk-d_~*Tx$J<
ziS0Rs@4UaF@*?`dSd4+GLQh28g?y3pXVL9<>g{*b)mgC;SQdyRi67;#jn@lP9%0p#
z9lwr^Ii~2jpQZ@j@C;%sVQ_+*nk~Ysjl^E?fcu~uxeqQ!QT^#Y(8fm!^^Gh!Edk?{
zZPn!=dhh4BYA{f%Flv<%EFENpisgheftzNc6lZC3=73M1U}>AertlL9Dy05CN$cJG
zi}=b!;@1tQL@B=VsjT|yO?v2YC=Wkd_~hr<fLA#P5Yd49tEQN}7`RUxA#|fei{jj~
z8!m!~uBy2WuG4dZ^p(hh;L<NH^5#1=(S!6Bvw6P$6e~l-5dAoH-6YE~dTs?Gf(PTA
zr*3O;XVKYD-h1P#>v8@{pyssJXHp(tL;{-t%7IGK9W~|~mDFyRsIY`|)fr8iWL*U}
z8tCH3x-LS?Q7qRr@5c1PlhPh>vOA*#!gnLrMng3mlGDufn^l-BYENKCLfi3HwAB18
z?LpIE(&SFWGYS%{i=4ulC#l#tN_8FP8*$hbf=jbRsNQi5K+}wrXhJn_sfXHz>J`WO
znpfj=W8jpqAVJp{pPxbMmXjX6WUf@SupCZv6My2EH*xy!st%Q8zsrwrAkV%t)XiJx
zLmr66KXDqH5<~bxEwAUh^Q#-2JF8&po*#<8<D0@1FcS?TE-CdNdEvfW__Mh^5AFMV
zam))Ac8z1wD4B$?E^@aq5^VBr(rb?|U6B~8?w*4~K5E>)V7|tw_9<3I96-ukRCesJ
zB%GqQ%imGPkI_QkHR3azJQ2_0V}iV00(MgGYcZY)7nuce(lp#_zT3}cia&c=YV<^)
zGnUPRp5+~G_W$A%IZ)t_MrP3K#{2RxIm@<hMz`IE)fpzM(w#Lrhj(uITHQ3@hWuH;
zB;y4spi9W8?6)aA&VGNS_0!utVEJLR@4HD&?v(>roDH(E)GB)tU#|Y~o~y-CYtTc8
zB}I7GHani>1Jfjoz^8?GV;NvY9J^gwE4S;09RrG9)~iYMey*X??j4r*D~fF#j`u2h
zLcZDRxas?FlTc=}wW>h}%mw)f-Yo=5JYt)%X;9Sm-=DQsDTn$}a4P^nW~)lFH=S&W
zv=d*j-OYibTdSa&Y0yBm<REM}Q#LWCk(&-%hA*_dj@JELxyG(?7Z!c#6kIZ2pd!N!
zdmG6`#u;v28foqk@phb>jK0>H!^%%@mLh&Oa&i{?YgbM>|JtK0xlm}^bPDQ57KM^+
zX0|Suw@H4SeTB8jky7dBGJ!s^DWkR7s{P{j?Vd%bH*=G5MZ6ur1X8QsUOB2AJ{ent
z$F}Kj*-2*SGF)ic*>1?Yt7)*3)vR{+oPA#7mqt=RH>4Fks#Y7;q3Rs~xvuaFxv)xi
zm;28w%{^zuL|=S*i{L;^oBRu&$yptoab}1F-U>Mq+Qm7sWizTj{}52wBdMJDA{dYD
zRVHYpDGewr8^!NSMO|K&2FTN=qK-C06EdtP<)4=v<)*4@V{D+36!BI+4~$dIYJ!?U
zF~6oJzwK>yly~X_TPd(8^i%fpwbz_)oF}@4#A*xY1a(XS(857>NlKt=H6^ih`9p#1
z`P%HKBgnX+NJmkha^F@X`brzP9K39dSNGA2ay36I3)eIAI(gNcw5iPa->pX9Z!OEN
zv#VUuY-W;wvWq`^Zx(0ve%%}I>?$glrTXm#k-?Q(>xt|li(a)-3>!-hoDzAyWhrT7
zV-yRdEe0*$ua2HzAC+gW^vza!pNFoywQ7~66BJ|LT(&qEd)vL`wH#+b@i=cUYs(BA
z6GrK19;fI7$1uGTXe?@9;rwQQcPVVMvBi}}bEOvarp88xn%-I8*k(^trCF*uQLVDY
zne(c79Se<9h3Zh*P8e=6lVL>=Fi8(&l_SP7ZC^aqI{X*6ZPev^kV@2tzMW8}(dpY7
zL+vKyQ($bBNITvWwpr>|mhP<2@v@$RZVuIA$kM(NrK9JlGQs*NH1B9<yWWq^XnfO&
z4&%@gOM4sMomB!(oX<kQh^SLrned%DCT!UZPKM|*c&5`9>ZP-YYD!Q*2*hyzNu~4q
z-@iU*k>Q+0-Z)FIjW3~rvB<eKH@YJB^5vyA@v1k<NrTNKIYqqT{VfvXb@XM9@4b9w
zaGrZ!8h-J<sNSDgN0HQLrtzd!sN)!ZrFPziBKR2QsMhZ%kr<9Z89Z_I$|D0JKQ;ch
zqQtkgB4ETu9r0;}(`XX3ec?@Ui#b{=LOHU@TOQmfJd$4P&shE&#bo8gSeaRiqxxM;
zxrdE+wgDt)Om(`A*s6nlp41aB`K{v`0SV4_YA|teQD6FaWzKN(9S&y@d1B-223+!^
z!;<iMq^s>dD8zAD40S*jccvzNC6SY-P>Aa}s^L0Ca`qYf!wdEfUvzYQ5?9C7(rfG}
zIJG4_ElozM`}u)Z(jqi3={zj2h|)72dA8_8hh2fLmm18-<j(!(c<S`}nWNJrH>Px<
z?toA0>qwOH)JY;T*<Tv~H|!(g&%_?ROCJOamGoQ=p8XJ=dCc_yy?l);yy%@2C!QmX
z46Mushv}>%qap+j^dcifOO@C?34h{>*kz9j1DHRec*60?t``hS=aabIQfVaS>CSq$
z9&3l}16|bym*OB?H9I*UgeRg)tCS34R_8BuswzE*8GTm6C&lu<a1|P?_*%j|ul7Sk
zcDKm2OTR0rwhohO9Sjb~plU|wFllZLK6y=0-n?}5ZZ+5KEyMdfnSAGARkIe(71JBN
z=v?Y#&vHs~?sX!OB@5ZYtcz3Y&tQH%ZSljL=R}y!`b%_1DQbP^Z?$y`e=v~wP`{I8
zjoG=Gu{>ASGPXrq<FgEHF_h*=q^V=U1sh*$fk~y_w?FSDJXcp;mQ1rIxk`sKD<y^s
z&&u)CQklzoqtB*arpRo)eK||1$O`z&v7AWnwocmiPdfHyROY_-nkS)Dsm*vYt{>;e
z>PYBZ+2N+q52mu5Z!ZH=4Q-oNeBc}j$J*Ps;X5_jMjEZNZXB;{djtCuWQ8OwMIy5>
z4-MpJgfPDz7jR`&`YC@d3wl?cy+oX3_eIm)amUGVf4b<-$Xb_2;px@K+nA-#t2*wx
z5cv|((m5@V+M&d(xGlyjw(@O6)=ve#6C)wZ<UGUs-Y7$%qiP>OYhOo7BZ@qUNojFb
z#jx(W&Z=)obwT~^{$#1KUPE_A1N8|M6^t*gD0=F(-Z`poe{jXLs_xe4Ns@0JPbg%3
z0a-&uy~R9;<Hq7dj%Rl!k!p)sB4YWP1ixpHR&X98D!C!&`I7!7Q=HJomRFpL{$j6Z
zKl7p5P9bzqnSXSqEhqg%O%pTe^fA|;obnTX%iC|9oO+0(7w6+?03NH)*`-^i2X+xD
z1F2h24+ieW+^v^AzzD_kW18exXDH{U8wI^yMq1<RMDOUGt<o5048D~-?@{4L&EcOZ
z<KxvMZ({7oG=%4P$C-o?y_fO8{=i00!`G>Ohk3TNX<y{RZj!yySCe;Ca~w@MpYQJT
z>Ku0S&+fOt-*!J&?QS{d$sSNrZD(()<{iG{f7@}<uOf|v;)wQCmB<14j@NqACB3LT
zXruRW3zs<fDxVjhp~H3#M+B8c+lp|u{jH>MJ|1VSP&`bC$CesOLk~ZRqC~{jCn-%q
z9J?H3`{UY3j3^V<m0QnsQS2m_J4Cwt<43-oVxq-a`}{igjaBGd)@wicCek-Ou;_=G
zXv6N=#a4L-K5nOdWEC@ii~GqvByYd=vLVo#vC0wWWosLyZ~2p5L@-E&C=R3Z6=eRH
zZ1K~G8rI{J()R;O6$47VuZSk?PD6zfgCKuamh`1r5Rb@Fl%$^;b>CpP!Zmsw{BLBe
zf=@}ho<qZZ>2}^DIiwv**TaoZi-(>9yjZqxQij95bWkBp=&^0D-78#y04Ha(-uf9Y
z4f4THDLy8FkM&ey9JD)`(9Wtx?Uq8UmN(DZqy#&U7m+jb7JU8j8`AqX_l5Jv|CSNU
ziDKPVu7u((;u$pna<4b>n?#D!{Dnw~$zA%g*$#OA@<4Le0MJ2BRp`~CbB7QTZ^uxO
zj4ZwA^;%>EaoXwM>X@})J>CWB0`yA;xhiFYZqO}%x0-Lr(y710H^L<%W^JBXd1~Q_
z^9~la<gORhBfSf9@Tt#YgVy-$Cs&@JJ@7f{AOuEFEPPUDP<~yudG=dM{R5FYl_gmH
z893@k_cG}lXuR?j%&JnevtB=%E5rH+3!oV3wnIETkj;?_OtNoL;A-80XeoFUgqqLl
zIjKf%#++@3Ksa{4&+1?A+Kxl~OddfPa<mMD1ZKq$YM-_G%a)*SD?sJ|Ah6CZYf7Y;
zD=LM^T|NE3{^96@Wu}bnG11~MgRc33$!Wb<_3gyTQ@LpxklKcW@=2KIMQWKY7%;QB
zat-np`Tv#w+93cRM$h*tD>BwOAE>)eVZ?Tddhab{KYh6mQnVChtN%Z>Fo0fp5wx7`
z;TNmFSv-IDpB@Fn_EyZAPxwPq<v*3at;}Kj*{tr!E&Or&{rew?WrA;zs@?hjr`Lww
z-YD`4N0;Nf{@A1b_uoNO5x#*^ZCv2rJ*IzMUH|LX9c4(sS2iS-_!l|*?@mcKG5r6j
zdhTcm4o6MXaJc-mlMbl42x6=&Me}mMPg=KV3GgN`<b^Z=fxAN;`9suI@67zU*?9vS
zlw!VRU`|;71(x#aeBj%%F-v!{Q~iq~dN;-4LThs8PkPztD{6?)ttMWYCCeinHi)l&
zchc$@crlitBAt&|)y>WhWJL()0*+_EI5j=Q#dJ5&(go0`wT?NwEB}y7FC~Xi&8168
z6FiZc9oVqtjoe2Z9RSR+bp*NHGLTRUj6R3gfnlCM5(b2eWB3g_V)>1~r~)-QQiZEC
zI`Y}g{BLu;O4pF#VQ#&|a58eM1*iULU5XO@x2)L&?0@fC;Yi|vpq@%Caw9Irq6(h}
z)6W5+rU#{UkK7E|(7p|55(goR)<iVM0=C;D(uf3Z87cxhP_7+9gwRqV2OlbnFL?dh
zg95Zw7&lttm&1K1U?ndhGD}1#w}MX6;dUPX7Zr7MwDE3WJ`%|;A?oSj-Bs7)Ay8ou
zM)Fx|qiP483oE?s7v(hNBAS?3U2Sv1b5|LYK59eES?COPHfT=gr~osaaGiXVo<xjp
zN8yJPc9hZ0<fz%qHuEWmv3{TB3|{k*-zSUnYvfK&OjP_6?VQ-H7$2Y!q_orgP+1nc
zv;{8po+FIyw{_^p<O7NKMvW^XLR7D<b?=2lmP#zbK`8ZmuqAsje!Df1;M&VF&^zWs
z(UzRb7a8?lW#VaQgqTZQj~1ULmXh!|1Hf?yTGd0J!?fRZ?L!Lq9S-H)uH6-vwWC#6
z5s9Z;1rGB3O5KMfR)m!NJr9BDyrobQV*$_m@P1r1n2Y<t6>MPl{3D{6cFu1=o|3Lg
zL?bgirR;E@Mqn$|uE4HYvnlQnBY2JFznRN!`VG#}CV7LKgc=n$qx5bvrp#tyUru|i
z!liSls(eN7H<w>10G^Ui_JD%nA0iiGw>Fx!YU*Dd$Uv7x8SDz+kMesMd)!0p1B+FB
zMqO8-k+Z)9dh>i31q1f5gAExF_GOp?zLAv#tLlQRlKHnI8~k;oxP}sO0<mmAlzG(o
z{R+LlMF)T{;2rTWLTNKAZt$eBFNa`p#v94Rk(y>75Kfw&Pp3=ixkPQ32@JGAF5F0b
zL>neFAM3sZO7ZvGi+vd$*e9MUai|u@x<6#-Fpf;VmDU}Z`-$wL)**$%r{9WpH;ub+
zmuU2VN%>QQ+*6bjr#LVyG3+vTXftd~b7ZIt&P(e_lf1CR>`Nn(xk@TuzwSjp;p<aZ
zo=ul}k~tSO(x{K1tn;BJI)q&Mh-Ul=ZKvZR&0nA9{hbchzYb8L2;{LB?<?5-?PGTv
zy*myYCBv73w{1<?eYXgK)PhC&>>=nZDL$RXSG~(u0u-k#neA^sH>5$VH@5`ajkA=x
z3<4ccaZaNgJ@F8e(-@N8^0ZsNa)S8@Jajelu(r*>=kfUnWapmQ{$%(>_f69CLRf);
zv0m2TYS)LDvQ-f6_(b9gba39SB9m?r<T44>7C`2>O8dc~RH3AzcR|&5#JahZKT+>h
z=id61`xFE!7q_bgF0<h4lt6T%;LPS6-F}~w0JE0A0|?9@GrD$|5!sGZySs{(Dq&KS
zK143A9+*TS_~TLyiEykrkm*)A1qIV#yWikv_3XRfKtIxAps0x*xI!6au&{PuD-I&P
zLXR^*DGp{f&r#72zjh#*T4uyO^jUJ(^nhARXAI7kLm2Uaf!{9+W(ifyC*5q}aL;?%
z@^}C%>e^E}4AT*HyV2FsWGy27DC^!PbnRZ^Sb`Be$b1+u#&dKzCYJ%7j`;zTbU;9t
z&pvaZ&H7?9CMuPZSDt8zlxGtTKcDqBH8J6_!_nO{k^GQ&8D}sFg!1~KG5RWQmw&!0
zt1|mQ<@(bSIbQ+pYINtgnf`!fr=ZMb2SD+Xdz!P8uCUCtrnHJc^IJC@xJ@6{O0j~_
zF&WdiR{OzFGpVjtZPW4aiagtFKIjyMv-jpij*SptGo!%lFn~?+wuFRediXy1cGnbY
zvDN(2mCxY`6x`^cQ$6**q)Gpx&|B^y1}6LI{kDkrx1xk=B3{{Oe<Y$3<U7EcqAn2x
zq{Me8Lg&joEr>4s_Lj)f$xTNgllaob2*G@j5+#ZLL&TT{1@o_$c+*}W2DLCh0IpV_
ztuB0V49yOb_UvALryQPY@f#1!3n$)m@#+6c$))2yVQb-LU5zx}ACC^V9k}I5@X%^}
z_U-nN>V`HE%hA&VOr4p5={53B1I<wS!Ak-mQLBKbS=`DNhFqo?;jWix#ecc`_hE-~
zC~_Jhic2NTl0h{o`m#=SkbgSn3i|pUEbsnP=e2qeZVSaL<a(vAa&v)_2nk!zDkg*q
zE<DB8G@gVc_E`g-6>hEFu<iv%-!}Mel>Rzcq%}hj=$;n`&(qX<1wQ0&G=mL5+?NjG
zogMHLN||BIay~di8)^V)G(rnNy~q1G`klmh7DzpU%EAdv#e}mP3FRe7P**h@=-I1~
zE@3M&`#5`!>6^kk6=LNT0MyXFo7Rw^_!$Ko7*Vh*3Y#IfG?IDkbchq-ezOvf_EjAG
zKsfea5^tXNi}sUnPvyetu$st1!bwaY+!3~%2#T6!`iCD4gF<&iC7Mpz>xuCpa-?7h
zmf$|`z33zv@1-G7S;Q=Xew0h_erE=;3tCjusl%M)9-7Kf4>f*28g@mz$njQ^2%&%|
zIdO)zEOCX>X9Rdn%b`_TiPyOdty4K0im81NyT8w>z^&t-nuM~=7sfBF&9P0P`W`O^
zF#O(nJ5k@?Q>5d?#llu)$Dzn2b$$Vsn=>c)rpV;lK^?3h&mlt&t^JJFw7R$tFHrMm
z0qgRGAWrDu(&p@go2?2VGEzJQmKU5)WRYwlf)k4N>T?3Zx(&oba22MuCAI{L>^I8#
z6om@6QaeZ3lM95u(+KY)`5w@h0s8&097NRfg0KCNKF)2|y=rRhUrkEmun<Ejf$#E$
zbxvs55ylpTv|h8v(62Vh`(ac{y7VvJdONc~szPP|o+PKc%~J%XO_m!FR{5P%j$z5`
zdAzQ3viE8lI}@oGXRlXU{9YF7=#+rlNUc&k-1_U{B)!*+VlrCdJ*{QHK&;Mu!!<MC
zeydYT+_O!v<<e`?&2yXbGySrQ;^dn90BNd;2(V@oVR6Gnmq2r+-cM9nJfI%N+Fe@A
z+u9e5Nt4`#UKKG_0MY(xX7$;+LH<bPdhPBm`}3-nEu8hb=0<~NN%`jvZy#6k=pCL}
zG?;a?)AS3Y&U||xcFRa8TreXArV2Oy{(=#{^U~Le=Xl_w<d?T!2uV<p0hnJ|z+vQ%
zvAa|q|4<K93RZUWV&iXUd|S07PN9hUmvr6D`!ls@AE!87%&O_n(pg5t)T?f&tHcj8
zCp<ge`znq$zG2kZ>KCmOkdwpDxNB#5o6qBmF&KLDz_W6Lhk9U5_vQg;%<~~^xmKz|
zizNsVq1YF@?}|@?FSxfps3BRazDY4NZO@}VP<p4LmY#E4uewAj<$O+NvJV48lUG3^
zqEP2)M`nE9XlI;$YkPsm-yTmxo>+83g@hA-G?L1Gnk2q#jpqe8!RqUr)1OZZ`5Dyr
zl%Va8;2<bKIM1U)8ly0K&U#CzkPa>JK(6&94L0_mcllsS5bTp5Lv~BI$EBW$J$TLc
zmY*)y9*7xBAaE#vZcjfO^p@r+z)tWm)gowSNF|PYi5RVZa2739haPQAu@9(8J%_$0
zQQLpB<}=tQ6;+DX%Y(Lq>TY_zCz1SqE+$Ct(TsF_RZ$ZQ?35Wt=5M@@i6n%`)GXX6
zU4O3Ajv`cFu`iKrly+jZcP6)2fDiTM^4S7U!7t)ekt&U`;j5l;V1Vfj+a<=(R@_m0
z!q{Qfu3t7!wMRUv8SlCkPyPA}Ta@%d4)`~`O!~J;Wd*Z*m4hu`v5NW!?-(^nP<&vR
zdTCj$_EP5dOAiC63VDWu$9b&wO0D0CdAxoloTUK|8T}<Sz|?V)(S6l3YjU{WtaH=;
zNwh<P-m-C|O3#&$0@(z?4VP@2|Hs^0M`g9HYvX{3FCy{MNP~h3(xJ43sDOZhbeGcI
zCEY3w0xGt2w=~kyos!bs^4)Xo_|CV_*~_!n@1NfoYpj9s;$+VG+|T`7_Z8K@a>b@)
z&->!wpT%TY3`gSp4hG<apg)=eE^7HMqYxfM^5$Bf^_G<MagcvbiAaa*X>M99i9XJ)
zPV*~Iw4xe!O!`BPoUv@qPl?n;2HC0U<cJx0_^Up<{*h-r4WrICXkw2W)iFg&O^|Mv
z6F2Pv^RotZ!m<XDM{U$5Mq}IVbE3|9#uY6?VGjIsIBKlMBtOn1*}&3k6neo=X%@ye
zlQaA&V_KH(>wcTnkl&pT84H9*^jJxOg3{97e=L`HZ|c(*|8%J!2QEvAD#2rt@g+sg
z^k;hknn}bZ6ff!o&yf&z#_ZiROj>+t&3JF>O;VkxV}8j>l1Ff$h`eS`|E&6spIg10
zyBV{MqEm1veqJ2$xT%7s61v}mH+xAPt>a9;x}#KYw7L^MbZu2ThM17;eAb3SQ=Pw4
zFrM(|%Atzk3mki9c5O0L!EU$1j@+Y)+qRc*Rrgq|U)om7OpO^56XwP38QV=wZW2?-
z7+Y})@8Ayu4Z8^rPcMq+vzN~6xZCjquTzeAJV;T~@I^62tiW#mPwW|6Mvc+I+82&C
z+%lsb`BM$$7KtCM%04pL5PEgeYqpXYFlxvSZ31ahend)t>P>K+=$s*E(H7<Kjl<%P
zuimZR=G|TU5gcmQowdI3W?&lF8FJaYk8L~Ur)u71Q!SM2h=qQXP4g(r6_JU4&C_9A
z2`@TD!bOe?cQyCooK3%DQ^zT^;(6`2=+vl`nqIP5x8Jd)p?H3VY9=Sa;%v;OU#N+`
zW^q;K%1hn3G38pvGTkh`aYJ4P#8`^crdnz0X)i)d<+v2iQ;ZUrwWusJsp#@(?dO<c
z%VXsz4K~qy`Jl8zrj@%=a|=R1ecaHY*XvF%8_jidx>y(|g0N@)t^jWbjf#EhtptJ1
z1^t3}9Zk|c<KyasFh^S(vjLyxqf}@`RZm<#_SZ?Z6sp-NDN>x&yZRM;Jh6Vo6Z-by
zV&cu^%+N%GFn4I;6psSAgHNudrKfL+Qfy!3_{=7p>bXvT;{wM|g(F#RTm!bGZ=w85
zW}>qXDHe=4DO{JZ>gy^9Zqhz6S`sI^J<?(@r7b-?a?fwnvD2m}N=%upM1spkv9f?;
zu^f;%K}Oxtw=`D*eskX6dW?l~BHCsow(hvZuf}lAu7kY^gp%r;DO*@cC8&u#xq2f+
zNfC6e+wHRl(}YA_9nO;izwfv{QEu#?9IyA6Jiz$&-dw<?_{YF2!*1cE5jHIp6KLsZ
zYbh{O-PGiX2y!P|8aR?>mft;Seab*`gF@4DJyhp{xmZ!$#B$2%DEgy`Gr?|YpIq2>
z8SCt__;Hi6lVi@KDU{!fOa-g-eo7E%>n<9#Q=dw?OtSmxLEaq{m*-Ais?7M$HY|xc
zk`<TF7x)_boWK1r1b1KD8!~56)t7#xk5!I-L|yPu+{x>#O!kQV!PKmjV8T024RbgD
zvpHv{lGTGtYDJs<2e7NvjIJ;dZkYEOtHgc?T5UsN<TxN<Xj60;a+}5lxy5I4h$u3$
zYtCBusSY0ApM2@oDnG)kR_5})gV9#DE+#oOTR4of52f}xE6_0_Cr)g5%!F5fw=FBd
z)t(gy8U$>mif`qU_6^%=4;JAlMZoxYCX$F~xAal-p3e;<Fg{zhBH2Bw{%|0!qQ62W
z&_h<~r9y^gKi?Cl`<SoX^6$bOEsTF>uq#AeQ_>ZDLR)OoO0>uHKK<fLqq_zceNW&o
zX`NdybR(C)^vI}&oE-uN^&WjW&i;TZyNaoM(ZPelCq+(z(R=g>+ixH}iAN;X9f-+Z
zs0<~SWPf2CYVOvadQH?r7RerBx@DN>MpzfQv!c(w;HpG?jnU-Km5K|9U$?U;BpRN;
zdHl&Jv7Xf^iM(IX7nl1a6sU?qut~gmO?nqgwb@6>GCK5`(O`CMbhSV3&1)0Q*!Ly3
z?@>rn9n3mJCvjP2Geom7lH)3<v#sCV-7;CZ&Rvs=CfEQ!d^16FE*DX7@kBGl93l*J
z^}$9L3iX?+cXvrC8N-JwN?wJA`%%oq_}N~3z&%Nd%Q^J$vQ^UT!mywFPTL-e>YMZD
z7}Vuk&YBb$>Q;D|FZ&Xdru{Sx%S*3SgIL#u!!N{o)S7A99+2cE=+%pWDkE59$-5$S
zC(LL!{XJ0?!RImVJI)RM<m#6s?-OU{R!E!5OTHngOjKHZ3=}fHEMp7f1NY~te%~%?
zvM@Tv0a>FomyZ8QbjgRod*ha?ufgZB&RqhfDEV1Mj-w}p=&fG|5*0A)w~?_ECP!!v
z`i~-ry=fTvVmwl#SyRkR$~E~E7qdlVh3>olOKu7kIiv`g{7VsPI%qpRumg<iL1@xx
z78z^qLzW3w-$f1-|1Pl>zn_quV4=(Ox978CBSRjdi!=umuiuYKX^jjwJ1C?Tj(f2a
zn;F;4TNVx)KU|bl9Y$id6b97$8duu-5kt3J>(ZD6UBM9pN52I`)2iiClD)bz$yvjr
z1D2ua72zaq$3&&pf!yY`Dw)A$n1g(WLw<|^qj%m6L%E2+P59!0Jv-KDB+_AKDLi2M
z({0m7>{gXjNTS>KkhVcM$!A^03Yr+%ltW>Rc1ODG@gzZUmD&o>{az8}@7z=oW`7C8
ze)Z<;*v^yK+6t0-V+QIdnfGt7y@Rffgz5%eBQj=qf5UrPTdP&-a-b5-GA%#LVa&hg
zm71PTA(XOj3P^V|GcOtv`H_+x*5*7ms;y5#@q7xr2=FK#^toD9Oud2DdMcPNvOHBL
z|MJvZ<3j<gTSMIy9||<c8qWV9Kr363VYr&=AW&ATl<F|pL+S@H6NQ;Wvo$*@w+v@B
zyy}mEiD5ajbOA#~vD)SjuwRJ99h>J{nWWAX2k|_Z9c<loUG$M?tO!5P^dh3}NA#)J
zEKeC{U-FSXmrbNFNT*6l5Yw49;e|T)Fla^h=|E-$pK)tE5xvX643Vx%nIKs^_iE{f
z{wGBbME{6wV?el6;rog7C<c_FIpegXsvAfy&oc?{+EXqDHiv~|Cp=DR>q!GtOvFyL
z(}T%qXo*#PV<HfD3xC&`#!R@+=?DRQXZ0#AfR0OucHE&2x-&U3Hk<q53!Pj1Pa=J0
z1_YS+kpgqk^95y}FrGP#AT0B~*RiH;wTDB9QCy9~?3Q{kKfmLE6h<S>qs7YdIUToU
zehuglrCNjA?zrU58WI>)xIhnIuP%0&`DUpQ+gt6k5V;TC-B)u$oDZ7dK>t5aH~#kf
zuXc&7F>m;t-)w7r&jteQ<103LvVN$Ywwh^h)Cf^3@Y}ffHpz;T6_FtFK%h+FVFaI_
zftz57Y<%5m@#WF3o&rAXY`1Gy+{OgWeIwm4*o767M~4*#uT!8|{50$ppEe<^-^mqC
zJ~U%^$!rQf$+bgrMZY?Bx8f;0_L021ewOxG*MmcGnbWOgp|o!5PJaaemgSMQi;<Ss
z%Js~t<&49jTh7XNp0m2^qYEy5R@j6|bu=EEP&xr)DiwXjb!;w1ia8*aeq`3eje|5i
zN;|)K7W>Nv{?=&&%oOh&zOh%l?&Ny5U^m!t^U!feEh6-ilu=5XdsY$MmSB9DU}>rL
z52jaj{)tAr<nX%a<q7b5RDSC)mc$_0u?wIwzNL9-Oix*ph|F<J)W5g9-(SH-v0yEQ
zpMC=mP7gN*p2M|Xyger_`J{}ss;=VKMFh@y5isC^;Q59Rqub<pE#K1tPC1)U8?r)j
zxmzT4;1LA|3&+?UO}^OdS-WZt$y-l$Hxi1k6)5a*@Puy5SC|p6v;-tu02xmi#QYeR
zR&7gk7$m#)0|BSZ^!mJ(MO&oOsY^Oeds}Q=I;aEpUsA`H+$k&+qbwXa@)O!u$&8A~
zu!_%XC6oB1+^%vp&UyIcPb&tDtF*93qGZi?=v}6tD?AKq(<L-Zat^Alt7z(N{Q+m;
z#ceDJX;AMnC4RHiHhImk0&WEJubHE=zWx&02NMNJrFq66DGA??xhl3D+`)mt3Pczi
z=1y|kA)#Hib7;l40EDr_!-FD}rW2O)L_f)VROEu*R>l}R#@ouEVDv9WO&RZifv4+*
z5YNp*m&oH+7{*PgkyzioVw1-7>`4OuP!Zear8nndTm&{N#$0uEM9$ewF?N&eV)U(i
zp=@07=diKzG1^6zJmY^^@>K}9)UXPN*D*_F-3CfohOw05EUHNwPGe8KqvFO+O0(A0
zR<=J#-;)I-ORX?#CX?KKVkfUG!0Qjj3#CzQt-OI`qpbJdhkJH)Am`A;A1@sk^d%0?
z?!4lmwf)A}1mH=u{L5=O8~3!vHA*rc_pHpKF-rUEh(7QJdI0eCaB*eqjlprB8deo@
zR<nsrXoBoY^ppYN*O$dkHHN^R{Wv%VrT+njGt`SmM%A}>H#8o~t8HRx!Q5`17#U_l
zjQWE=UCt8N&A85#0|R@tb}hv=KbkpJ+-vg)$9C5Y<5;l$g0P4td+tn^HDCF%<%-Gh
zrVLQc%Gvjy0!<Y>4~NJ5u3~6<?$5~G5{LevP6~rH2m|2ImbbBUtOCXe>wDU5{80NT
z>Rm-$g`(*8gSPPbofL~5b*S%rUgb(T=!i$3GjGWhJf_2BsG*il2DyoSjdMmUWguOR
zu<EuW<^)2!h8xA#BzCwIB@ZSsxMRY%5!$u&5KJjvKvX7^3)m@7xr?6gUBUVkYXXND
zp2&YToM-4h6YV(94<%+v(&WSB#~%X;{j{*tLAyMAJVW`V-?;Fy5+-XR8ly{RJ-Y}@
zSy;p7)SGHo13Vhri{ijg{W(381EF~05@3P29@SD7jb=og%3_3i=T~J7AHka8?JiIH
zBn!hGm)Yz$htkdYor)sqxi{+;bu#KkK$V4ek?w49+f^7pNu;05^B%o=T_G(l#h38+
zE)(AEPbA_2)!nRPPuAZjC=A@`6QWZWo!w|(!}xZ1pVK?Y^JFbh+;8`5gx@!r0gJwz
z^wl*}^{bPw-9hjkg^_)p2G*-3V)SL%$8aU$Pfr{Y;}<2q2}l}Gt^0mARI|EkUOwew
z-t$qJQAA2L0E1D%Ty-iKMd<D4U+ab&+%^IC>#83kV907H2j@nxYb9Y6>Ym%LbEBZs
zCMvuNt*VXUyYInn&g6t{K;1Blk`?%RGZHQG_Vu&Dpw>Qde4n64Z_N`>m|+8vwjXdY
ziJx9bX|i0J5)H(+FJ}iYp_-h@+bEe7e=C9NDAyV$HjHUH?5l{0l6gDQ0X;V<5CK4b
z9WwI1bHJ(C4}aVIjc4>3Zq|M9exe<0@u!9SyFg`%pHaWiIOl!T4@8gCXSmUHBD{v*
zo)Mik9QHOLN+ZrBu+e0@HzsiPja7FXh@|KPoZbP@!rkyt{m}uT#dXlE{c(EHs|J}7
zjeB=a{Fp`^s|S|_uNayj+vfy2yqNZGR*m(Q{G*`T&a8SlXmS$rct5AhHRQIQpz?gT
zxnQ@zYLpg6+v~eHX!BC6!(^;dOPVTo=JC!oD^a4KL<do8l?V!zAo|uyUl3hR#6er8
zb+z1LOaz*Owv86CKrMyC2SQf58r^maCPO-T9kyW>!obQ*Bjin4kBBL7YGroztG!P_
z$5o=8uy!T3^25-8d{%F6(PK|P8WS4s_FP6Sv@O%wC76HMLhbla6z4HpkYE^F^p8zs
zp2R4;dpZtbfgWI`(9?q%u>f5J;Bsx)uUYhUC1oD5G&Cs7(vod<!>Ht?2<S^J`oz+C
zagWP@UBm(tPB$a++vvr_a>qDN2bhYU#_Ox=tqU9Bb>CZ~WbW-VLB<^Et0<#L)gAra
z^lLD-o`zeeKp-^#O7+xclX@;N_diic%-%Q0TXgZv?$T~HD{7*)@@*YVx4Oo{ZH3_Y
z>t2Ci1tNGYcO|QVhwi&Lf{;Fy5Wh1Sit~Hahl6E8A43>AjwLV%;-M%Qyu+4Z2yj;}
z&|BiImJ;2Vc_^q7+!%SsKjt9-z2<l2E89V6=2m~aKSqIwx5~=9cEoMKxlhJtcCxOu
z-rEsug5s?RlB|f6LOWV3UP;rfh&F>iqQEAX?hZJDC0X&vYFFZFx;8>oc$+_Jr&0V@
zr@L>LCbn{{>_P(z!AM&CD2V!IwSCQj#hi`U61krew4JKx8I&qaz@uubV5)bac$wfP
z%15D5pBh-f`?C~uQi)bPNuj&uunc%9+@wdf7>qcugG4k@rgWJ-n%PF;d)0)kGS2Vr
z)z%BD`1<2red@OGhQ7F;@`uY<wXo>4c9KZ(2{bs|yB7?g{qUYmNow`CukM#&SSQ#u
zGyq4dd<4Rrea)Mf)#R~6{cAR9KzYvG_#edf&c@ui1-q5bt8Sm|<u!_n8Gg#0ys;{^
zZgQ6z(A+}!^^((b)bsC5LWI7-6mQ#tLOdSNwYC0CA-G#KTtkiv24N+EZF}}UF)knL
zDL<_6vkr~c$I85}Dv_ODrQUug0(cFT-HlsYZc8$W^GRXhDi<iUq`hXNa^%FBORO0h
zvO+WU<%{EselF%Phb@|aen?0mGzg8o8HYQ^sJ8TYyGFo0JzE-5WkT)Mv6a?}s38C|
zYdFfXFPe+k1-gav6ZV>en5dux<I-%v6PflXS!IDd^^)_~8;_c3Wc4cz?%GY}>NVy)
z28XviFzOeFA9YFJ!tQyI%ihoIsMh2a{8r5{1CFlX^;NyBG({e(;@#1%Jesw1H-~V$
z$EjnmH?^MTGukaMu_}2k<^^DX-8DP3Nj~eG){)i9V?J|C$H0w;@5=?EjzT`lI%OwZ
zPJ+gNOze3qW5)+KU%cUFp=z%+^m6-9t<00zr$v;9znDYjztGJNYu@Ucb>R4UR=8RS
zk9zXWWcHaYu;i_$FML{n@6lIpWwG(CEVIeLNiC@~eo$3h`3#IXCXX5!atPA0IlA~M
zt~Hc~|0)9Jxr$b*vLXfUFi;sDeon)i)NfmU`coBeKT>81WKMj9J{qMTV$S4Ko5?MO
z;)$hI2)pXgl_`v+U)FqJ-rBpO)+6%PeXYJvPiJKDEw|z*REyb^TVHaQK8BToi2eI@
zgsw!>?)p0$zaAM3a7LjL*wXggH?mVy$U%pC<F8XvCnM3eM~o9KEPiw=371k#?4lL)
z>Ykzsh!z<iNJLl7vOR8~GC8;YGH^X?HOL0>;=6YD<c54a84=@q36a?M%_3c!ZSJLK
z$PG4y6wHO_JohHodqrCZfhtGcxDtNB@2>l64mysj?f36C%RL>3+Cs7Iuqs-O^*T1N
zJ{139odzoQo;WTFLvYYA$~o3+BsA6~zI87`iGz&ky<oO-vvBSw9>UCZ%jT67qgJW6
z0lEaQS(D08oP*^$_MI}}ZN_yWF_JKgX`|`38YsSrQdgAxo)E&uwravlX+uACUSM)T
zV~h6Jh+Irb|1GJtLh!~=5mW!RD`KH;IDRwx_P8s|D<qF@C3P0fv-XTv+zc=cT)(AX
z?)t4>lJXg@N1y1UA>jN9wML04d^LY4`O!P$`cx3JIn;R0idi6W5P+yy8NbtVL2}kR
z2T56tZqS{*mn7wPAN5l7!jlDcw1==2)gAvP8zpH`XVDrt?5P3igG&aa=*U1l0w2P{
z=Q~w6eRo_oE;SqXUD2qRL!Tan0!TX+fsq|MkPx`g4|CoxjHQx%{(F%*fujauMs&7>
z9s$=1co?G_#O$Mma`EYY_KJ|2wxT~(#fqS>aSL`;B0Thgi6^Vg2E0vKA6@MW+T;hP
zlWT!zoPW+mETM7#IZB*GB7X;Dq121Gb!w=u`vzC5Rb?q3nTbB{JH>d}CAYoq?)Rdc
zoOJ#;;0t_XK4gx{UvlZ^cUj7G`x?F<AoL-4n{-o3U8Uj5l$z(uXYZdZ@@7^)<v%|q
zYtB-0r>-*HM3)=p<^J*6u<xXkxa{7fwjebdw)aTsOB*twD@&dYR2r-ZZYZS?4%GS~
zE1j;m|Mk?dpG)$CP)MZsW2_wk$ca{?w$*Pu7JU-2MMTt`NeoU!;cJ(_U?8tGz|&7t
z4-QupIXy|+TD$At&n&<vN@@~>Jup!hg&(YNU%<87Z|_;#ck>#vdn%?us;Bkz6R&LE
z>JZ!v{^+AHz{mJ3ZRzb&fqB9`Li-Wl@7!f-XQj%!YzXylNk1_+wWY_!e0T+8T7<gd
zv3?X)WatHf`az<0mVqg)D@j*2ziHPB6OK&Cx#%_<^Ls^qNDR7g=$>M5K(w09YBZ=)
zmg*=WeTV)j=JwjNuE}1aRr46S%+E8==QQ&y)a}Pv#1F`iW4L9hd1B|YP?w1k-(sB>
z5v%$5Et<A*c}6a^G0F4`v4OXz_eb{O!0z>KZ7)gbrzWlT=L#R)+{k3nY#A<s!?N2<
z=C1mftKqoRxHsEQD6p~L(N8pe9$O!jr+b`qKNd(Ag?GbgcvmRBXfMuoKHz$=sO`re
zl&*W5x;io^Y=A#G>9c64ho7Eo@U9bIDzn&Id@8v?Y3lo8evQlTj4fDP$R84J{Ctvh
zo9dR!T(<FH-N{$cldoG+2Ut(p4=|3yt_N)s9~(l5aY#qRn6zkoZC7Qu>a&uN@`d5$
z?H0aOV|bO8Iks;eC9=$pG}}m2WZ!=L#L6RfOsXVzlk%8;IM}IstMSr8MaIFjEIFO$
z^>S)sujqRnhlx4EgYNsCk_S>`o+GDt`}k>0`iHi&TW7x*z~reN@mHA3=}Y=E2<3kQ
zSW3BlfH}ie=j?n`pJ(-jgh@@(jk&Te>azHe@*hTzHxV;j)YVb8-$-%(m4fm83u@vy
zIcKP!Y{Mq~SQ_r-i2gE_51@Fi0kcO!YM@a94a_*hGkF8Jiu3+RMfV3O;cw9Dzd7Lj
zm;XOtl?YmX&MbR=((5(~1NYq_XwhUY%|HTlO~5&-d&iSBCaj7n=8%bf3Gco;G*tu8
zgQJ<Mz>j_qQIkwM*_GdngshM)M8UC>gY(GYH$c?C+{IU%B+h`hNt(*;jq-u{th|8p
zn(nZ!FTQy=e_!^m<HHYeXAh;Vok6%Z1Q7G1W|DPAulw@c{%vrybc7g)L9l_;Da|~E
zfFAQfu=HusT@JrNK>n2<>!U4&Q9TOms%2o3vh_X&C@33oyG+k(R~PCJ@kQ)Qp&5L{
zzX&70E|3-820ENh@1oEj0;^tl#H!&T?f>b$|2r30LXZ_+K0zx6GR5B<`TyO^`5AE}
zK2MhXd-Cu9sGjh#yo;Pe$0V+#{OLLLzq~g8LSue_EFb<W{}1Q<Unt6d`peOm2-scj
z^FZW3d~pBxc|Ia1=_e2D@BT0D;42oN^&gEIFD?{9%v*oCEX4si=5}yp@7U)I)RrMu
z?jzBX?FWv3T!_yTSXb#cW#<0fo&WpqzY0Vb|NQ&~u_}$Yhp^3xK)O~M*!ya`z5*bw
zrBSExs}-hj<}3tVR!rsYtnWWPps#QM^n&;_$0UE*#XCk~x&GigkI9}biQGxc+FaqA
z`E-Wq88|P9wlyE@u3BlB=YY!6)$LyDxAB^a`W1a@7(SME73-fv!tD1kF9%BEni4GT
z+YKP>5k!5kNC5#?k%;5@l4ga@!3K>AC-Z3kb?gxBREN8tw#gAh5bB54@~5GHd9eTT
z%Pngx00B!C68V}`bUd~Lp6~#%*$s#Q%kSc}bKk|PH~#`=6aSXGPBB%^Fj&Lt5IZ51
zM@L{6@A`iphU32p5Dcw8WkNb$*fN~v1mGCjRdSO6*os#U5i@0g@XV)kMs_Pf$>msu
zd1e6ki{vgd8I2Ik`51V@AFsl!j<E`uzL%FEKfVB&GM2Oqjq=VfdQ5@2TrOzaKfN|i
zgu#$`l!Cxz(Qb{Um47%(Bm7Rt%yglKDHA*)*LSWzU%3%|r|H|A%Z4I21pFCd+Mhre
zFRB1c35+78kzDu0pN8A*dXN%VG76S~@2;2WeD%y}PbV$AUx=_JAH*WX5QM{82brfr
z9d_VMq9f^}>IzW(Rgbv|FkM28xf^!_v73G|e5qv2R&(nhy~tWwRQeS!!l+Nk--aZY
zf4s^6;}`SZRe-JQMT%?$X#sFYJEG=EwN5`(4a27fW8hx$WEsym?TlKP+jBzLO+XQC
z7l3bu@QxX$Vj#|Plu3Dyhk(~*G5`IwBT#led^msLHv^<Tbrw?7OTH{tzaX2>0MJMs
z^UNrjF;`9S{450gpusswFM#UJgP@bj&E!=+wVISoRh?B5R0hHfA{uMG_9N_FfRtv)
zwYA~#Uxkb)1(K;^L>|2MGXDLTz7qm3Qa<7u1jtX4@;ZP!`H2ThEZBU3+eg=zfb^OR
zI%7uH9<T^nf<rv-Yylk23O{M9tkML0B<vK&8gGhXMckGxzwXs@4mpP+?B8%6{WA$9
z=;Rcbs3^o_6p7VEUWQco0o^UYKL$YD{@Cm22fana_1Q}g_ra8!TeXH;_`?z~+z<+R
z%*vv|8JVmOR}$)88Em0YC_BkA>Z|BKz57HyLZ~I+X6}H);V^)$Ka0zjuYyj?#(c%N
zHnFhci&e@EksR=L8vt_fpdhB<lknmNz!Pk`u_Ppw!OZ+5T!d%P@j!3P_3ad{2>0Y@
z2Iq%zb~p7@B$gJR^A#+VuNgqle)s+>9R9*QZGay;1e4B3LS7&k3Y;f9)J$*<9X*Ek
z-RN%y(=qrX3YBIAhwm9(eJc)t<WB!|m<hJNHoc(lfWaA;0Y`cn=puRazt#dBdz5AC
zBbh2+Ke0*P3?f|*C<Cg>5XiCOE8ca`lkgmaZ6QI{u=Deqm)aw(Jb^6|I{;>Htd3yQ
z5%6Ie5X5c{Sj6%7V?iOU@%UZLo!#wO-*N-Uc~u^je)SM)hdG(r%<noJ1)>4q{GRL3
z1f?ftmapuElsz0JtHJm-4REtQv`l4i-(m`7@4C69@<nmN!WL}rq0@=<IJN90ON>!?
z;|<AGhF8~@s6V<`oCscb7*#eoi_xvFUt0*2&o$CxVC)Xg-yigIDx-BAsO|$_mp>eK
zL5JqClN-T{O|*vu-!sm`K&>Bnn0a9B%3;@dLaz<39g`mGuxbfcEG*~WvAWh$F27Sq
zd$O&MbTm^94}1T-`n%#_O8Vz4+^G>Bg|%+#O!g^O1x8xbe5OhTayA@n(k_5@GDlOV
zXhQX-??B`&h`T|LyI6&HQ9L~R@io*oJc0m<D3Il~%|f9;CCSB4K3SkY*^|B8vjpP<
zz43#-qH~ezJqu8_cs3@EoueNMOluYF_St2k*QK9tn+D8?pX_(&#`c9i%SVpq2L)SA
z0X$<yUDCo?usxtlIYl2WgUgdz&6%lnmLz8i#s_(N_wxF?e=w=NcX<!tm&_g<$LSY$
z?qr#Who;N&dSG!Xd%>_^33j9iec=Ch^r@R&MRYTH^<h8G&^cF)0j|)s4;#!d5<cCv
z`{Yb`ZfiFy>be_n5!GH?K!?0g8d;OsI|4jWW_^CMhmg}F{H^O^^}yh$6ki0m^aIDd
zvWxA>Hm><*T`-9h>HX4uuQM_OTFy-EjORG@i76Gp=@+f?g6pLp=Z7}7`_I|GxL{nc
zzh)7o;Ebzj`r1GDgDw62UDzTDqX}!6&8h5eC60lf0rl{uXR2-UWeb=qhjO(qlx#(D
zwGZ5M1}W(0Pw#VAJyzWcfebzbPIvcWj2t;w=nv@aR1ztR5LWmg<p>Cmx?8TE^w8|!
zPq{X_97xl-J$$>Y*iviq2!-LYdNEE@9|Qnd#|Vv=v#NFGLPHb6EH#Tn>{iQt8vL)E
z=$7G<9RY{8VTd5IDj(Ot9PHGKpr=%EX<qun4f`TRNveasz@%ObchY<H!T={laI>wr
zS~r3!y`Ao&cK<8vK#aO?MdYp)JXjceLvRms0!eX!`5c~`4KVZqGMXyWm@##E3&n?7
zKhfKX$=~{igkUl!mS<|&vC11$Z#Vl>?<_jS*R47Hd&7Y4(DM_go<uL_OqD10VSUZA
zbI`2+MB2~ebo&p#Y5u@lj{1!sq0~U_P1+iNb;cT~P4tL*#4Pp{<5}t)#C<DFK^7Y?
z=+?lR8^|N2@^5@=0mJ1oV>;kl-$4-}==G)h=W(LSk%Zj=%@OFSE%s;PtfCr`xh7o!
z%F7W_D28DJ-bF<P+zLK;e&#8eG__ws5)jLWraM2Aga@&Pe3s2)M_-&w4lMNRgy={E
zgnumgEn&u2cMj08FtJV0YByFO1Bu?kCb9nch9zK+Q!U><jJib8XfDK@{+N@jK6gm?
zJm_8M3hK0DcGoLDx<Y)AGeI@Zk9a>+9Cw}SXL&y+L*^3*emI9ohLsDR+-VzyxvPJ?
z;Y6kk5;ru%81QbaAtv7*XWjfWUZnsNI}7ARe0ArH{4M6>n75swF$#Yc_6Z}v+-w`y
zFL)pmM8D;2!<$1|b>x2-ZF7Oqb~n-5Jr+O4{7EDRp=`M@2{GHY7}L7Y@&ro)97aGI
zRZhW8w~0ALjLj*-B58<(e(PyCcQ}1L95uxo;#J7+N6XTn<q;X3UKzrb7SNaW;|lsh
z)Wo24w|=%&JlYFvzyo9(?>r~2S%jZxB>P>fUNkPZLL$*OE}R^X0f}a%%MtRc7WE%y
zv#mY~@U**JiLAKNyyJEM;6(iA6-)*XH=qQD7M%B>;dH$4(*@=@^_qa`x*d5jemRo+
zD!y92DS57?oDY1(DD^elc^imA5mGthL<VV64u>HPv#E(W_iWzdx*#-?Wzu-roPM(o
zmO(<r#RxTKq^+`PO{C%$Tr`lRIf2j&T#XR#@ax8)W(~0k|6cwr`YZ;T^JGEME4TZu
z@Ktk_ne!E*;O&RJ`uitX;(LsGaoIKlCI=$gxC!?sy%qd9#j%@N>Rvw5w2>J(1fqAn
z(2JeA_D&oycsFtOGaOAF>RQ??h@hrTE7<6C0R-RxS}|2|w$a~NW?W@l1N1BK5vjoB
zz@@q$#7XwO#M<(o{VC;EWrCjWVXmr~XEJU5xVGK*^x8Q>ck>r2n^S>j-U0=E7}|0C
zj+>}JV^Hi8DNk@OLV$A)B;y6LR>2Ee68<O#ib}_dZKN0K84)U%Dp;1Px$XT0mu$w{
zbmqFbCBtf;pw^?ezts{?UYOj?gf)eXO=|8F=pE)tUM!wO<3&EmX9XX}oP@~6qMdu-
z<>+iREef9XMuu*X<?W{%s&Et`;1{Z^_Jt$7FO4TvT|cY&a8FwrtORfcrmPV4#dY0N
zD2j+jx9ap_f_JJjxhmJP{pOt@4)t?+#*$ka`{ov|h=QKM+ULM*0!DI45kIvc(tF%|
zJ5QkH3aSjJ{ldzx9@#7c_A3i{0M$ID@hgv}J(Sm?g3^f|&^<%cS9Om2rT|$}ZXyPe
zMmWCgv~LxTRWRzwXI#a+nI~KE-<yLZ(4FR>YeNS_UpECLncc|$5-Df=fONSRcOXcg
z1*+L97G!n>LgQbQ`~;z-a;p>~q;$@C*AYR*%aZwg7kmg+S#%o=24K59pF-!gnKWKk
z{&HKXVecJ2B7*vM29x;BhxR*(wUvycU<PZD1+Jpw(ML&XnZy|Pu%yjUAyhg?1Kv6Q
z*7#PoK~M62tRxX@q+e9~fJ`<S$Bd6Aw+Ww9*^%yRgNNf}!VXZ#O@B;AI$+b?MX%`%
z-_OxGAgKJHt&4t13r++%EDBj4Qc)x8-|iehnjU%wtNNM%c2;tF91bBJjSK9aMvQwh
zTdIb4PJ16&pUGjgV>h#CHxYFxV#P4;+c7!3AYSvT5?q;#f_C<9pf}GfO7-F9U&g8e
ztJn*5JNb$NT6S&x>jqduzCyY`a(2^Kk$j<Smmq=1gulGLKkuL8TfNz6dhim#M{S9$
zncx22kVyem{OC<x#RopE^AvY4R9=rOUSqYvR8kY;V7RTRO4&cUs5@A&)dF+*<l+fV
zB#@SvBIWG|xl|7CTO#h7Q`G*!p=`vv(GN;v>Gp7}r!**nL&=DDLtYg3r*GYJFSTa2
zxuHA7c)O*&=pfYU9xK9_9nEUeh%9H<G9{)NWqbIkm1F0&JFoD%!dBOHf)S~^`M4IE
zif_V4zdRBxQmJUH4^|c!XUKG|*f7+NjoE)UzS&n>apk7ygt}rgd>e4tLR62zR$Fmo
z{W3+JoSIb+gjOX#)}2lZ4kY1(<QxXhmYZOVVS{;R&B^M1aA-$!$lTQACE>z4e-)$t
zp~DwNbfuF#H!=S11rWZ2+x((`<wc;I_8rZ&Mm;s#PKzn5<cVpf@m#Y(PMK`^!y-Hd
z=0#{~uK%2UJx)R~dU4=Hg@q#NLV1gS#!2ItOV#MTXeVbB<_%KRZH($$+*_IX)K;W#
zQQ9Ham#_GZ`7_*+aOIUCB9b&cFM*A9%k&BZc8yGxCedlmSETPAb9EA-ag~Gaj?L9y
zcj`zHHEP)(+V5E1^)OjnYR}8#&Eq|B88@wouaQeox|}F+F;Vq$;$G%5r7D5~Xa^?D
zvm{b46&|O%ueyzgvS5d>SckBX{~YMRA<-3HwcNhY7%#z<_)e0%k^Lh<QfTRln|5LO
zoz<IB%Za(WT!Pznai60fEtn74vXF+dEQPaF?(8;euDu`i|H{I#>(t}twGM;T$a1?u
zU2+IToCswhjxZkNPh4X!D=Uysc)Vh&u_%K#TPTnyQ}ag4Vp!gkTmePzBHl`@J6*xQ
z=-_$I;sFW%q1(9Qy;E#Gay^PH7+V9EJ^=Bm+LWoplxe4rthM<yy$Ok~`fx7Z!xqn<
z@|j<OqgdFRYHZY5v<jU@2G4it78@XA?aaNX*g=p=1$X+k(Gt6sq6L>qmS3AYVlUg6
z%y+g@S{%kZQ=Xea&m+Nz`QY2xe}!NF8(c-=CqaT$1+mnZ6E(iokme*~O5ucW`uK?F
zagdp`zoaU|e_%_{$VkT<>d6uPtuYFx84^0?34Cq}^|8N@$%}uX(l}l8+4;>Kp+wza
z=lYRsnLOg2DT7$`DQH58+paup4Ot}$=XoV&`carmL#3E2T|`-s)~KGup{r9mWn!NW
zdZ*5GyaH^3bKf_YnkZ)5b7D?se*NX@)sN=IC;pksp2S+rDs5IC7xXj~>nSgGn=x*-
zsncX(>B1%W6WqN=SDiwMtII6<lHQaQhM@$BKHjr5X?c3&qL3}WKg=G|fRLvW&6<iq
zA1Gr-NY%z~^Xr9_n&1~t^dzBv;u~FZtcsNz?$Ly($3zqHr>i7wT!m6xi7<s8x14q8
z`t^wg;{k1K%F@Vq?#x41!Oz(*-FkQ4<?ep>+(|Nf*!+c*Fp8I`HL*a@>qF~?&QR|*
z5}6!stkxl;H%c&-EXmdCCa-pnG?5&G$xwQ)FN5Q9X<+H_1~AcD<iohx3T<|l1~cXl
zaWOQ`3`r*BisNGBZHw7x0jE}5C!Vawyl4GmT11F3ZH@&ZV54YrHPF#ZEg{))JX6zg
zLu$0t`TOGhHQ47%;I#U(MN;lJ51eZSJCO)uC-?Kypi_*vzIpq)9F$t-1}k(w;0)Wh
z8R;WEReeua?5&Tjop0=TuX^L>7Q>vB2N!7-vJHtm0Qw0>QXoR9Rk9``j0D-&iL$eO
z=j%r*3nCVal$>-6)2;PXUEWv<HHWbNe47?5Qf<e->+vfEl<5!>oMRfG6zyJJmWr(a
z7GZuG*LjP=j&privs!Y^QzqU7^|dD3bR1ImoP7}T7QV$SZsxFJYuA?%h23zn%h=a~
zzO?n~k3btalFO<u&^%~4<0$kMH}<ku67?;xnM>L;<**Y@cYkL_@Iz1ylW%v^g2@Cl
zNSeo$M4wpkbOzn3-cUX5+_E&>ory!@KD?|#(y{OIW4`*CebgP9W9||&7Jh|YIP2GJ
zZ0@cSsbb8=6>7o-9*V)|yBYm;gzY!XBeoUn*pkBESW|WK=;v9g#2v;YNDG+`MjDkO
z617CL$r?ARzJ_v)XN$Ty3Bi2PuOGj-DUPYzi#*pH>v;Soee*shu0wDp*dxGvRU+wx
zd5Qe~VZouxP7kh(k=4$@;}T1)ms6dP!B;f_Pn70mOvLZJFkh2U7Ge-l<c#`VK$7-W
zYZGZC-)MUk<JFRBKX)tSQ%V`RMSdemeud=A+qBEE#n!bKSyZaBQ4JD;mfL$s#1HLN
zNr{vN!OHb;rXK&{p9HP3WE|(%6nhk8)z+nx?fGI@v$X@hEwt$T$`v0*<gEV0h%=r>
zFXBP(aR_-qzd*c8O)hrwCfq?SwvA-u)SChrZWqlvst@Ck`Mf3h`92MvaOrU!dquNW
z`&Y;p8_n1R-=6tC(VSdAHo2dtVv7JP&YTgLBT3Xv*bwZ?hY<*EhoVUTjnnj2W4Dh_
zR57J{LfAUZtJKMM``Y+9_0vEKY|c>4G$S-){QP3|-xg|{j1@=R)vT^EWq+f&-Y=t4
z$|A5x-9=hB%F;u_H-|SDQPv5?h+oZ|6Itf{LMs?H`5zv{B)sG^9aMCi2gW~Vc{`UN
zIW7izqMPIW%R3JCkT{l{`xO^+0t`B4$uZmtcDuz!*t=I0YZ^Sw7yIgur}{l%1>VKl
z%x*d__`fK*-ri~|xW=7qtqHszx<<_bznKJiAPV&Gk0$qL1Nh<#ign2`d`l)Rgl7>0
zG-``{<-IyqG4y-I)Um>>;gFJwb1bBfhb??xRO68O;ztKk@fznZ%dsJ6!)ud%BuCoa
zg?cALRYqLg3fDjndZzRNJjVJioZZ?RXJ|<gmAvkoXkDf(-Apat;HIazkDito;TiL@
z$8Uc{V(a|iK<EA{f6&{7tKVx9a-1mV_l|DNhVaoq6j3$KQ(t`hIVY#|e!W4onz7WV
zJrX&h2i-reamJ;1vbXZB^`TJSAEa=x$yZCiu4bp(cQdz?_YLN$f#fIB){y(x;C$#C
zEah9$htI-<$I^G{uDqq0R;<w`b{wmC(kt%xFEt~hty=ty@o{7j$k#d&E*kr{cSz-D
zbQ$7`#PwQ->$xshi<ce%5QN6p+2TXDb!&9I8e8vZaiMkfN|KfMs(EItnsG=-aZ(lx
zB1Oqt3yiN#c%OQTYuhywa9|c&Ue3YuD6!hFO}n9dt4prhZ!TrGt>z(v&a|&u#nx_>
z-yQ9anO&3q%d)^7J(Fbx66?05Ox+dX+Wg1ePjT^mOEj#8<5X}mYAHhmm6e39#ybmw
zs8u9mn+NZ6?QGZQ!<0W>#ZJSr_e-L!3dbYwuFUj~<c({-5gP;;^mw#swIvfA7+Um#
zFUF$aAft&S-sP2@jkDmnt*YlwR)a4%@;&FPPC^r|Z|rft8)E;V8+%VUleZP$7@y~l
zBLn;1X_CeQ4x`xJ@{Gr^Iu*3?R5u4L$;o}yzMIiLDq5N=HNGeme_F!J^RoU>V9%M6
zM#!U25O5Q{lfwI@Gs6M_d>O0=hd5@X^QwfDw3K#fEk@sx12%}m`u!7b5BYB}d?KH9
zCax5644aE(fc{C7U=)r7sZ4U6le?Z(dGMr;-_X`deWp^LsJE9^OW7S>3sO0fy^@&E
z@b7Xbs$NWFXwdql&)*<@0qx{hoLKwNno7XoYMwmy;%$*U*yGJaA2Utkb}4A9g3rXr
zC_`*l_Q&9uLd2UQ<CiG-P^DO9R_v$~&|m5?QWH?LzOSH~(ZZVfx9wxRBk>_8o;PqE
zBQ$zzEDQwY+<c3PET0-RkF!MBr4x$n=^~iekFv4t%x@mQW(B`C{h%9i&BZdEV*HAu
zlArxKMl+Q{YYzhIAUgzW@6Bk2MRW{{X&4JlNYB;%XY3e>rh;F;6hz=n-WVA$sDCEJ
zf4p6KqB5;T{vfdJ^!;5?P7UpMP6k)hG3pA;axaO*S<ex#;;Y<IU>_)A--84&N`*Qa
zix)RcqznCdy6a5*4+nyJ4aTyV*t++l<V^+i*fHL_rM2BWP>7~M2#uU^8yvBVvBxYU
zo6d{H7!9^l04)<o%}m&zC>-^9$R-t^sb|DWyD7opExR1dR`KWE07+sM7zxo*O>;8O
zHnV(;Ao4f!Tr!~FT->!|ZOJjFx{@eUy)@%dvf#6!mS?;_!MAIya_R?U#<WD35#;Kz
z*OMsZNbl876si^3<RqyDZM{vb7jmjcg4qr?YaUuBKu6%n6vTSVYSccO%&oU+SYZaY
z%L6LAOx`pVaFDC^VvEWzGGQ~spT0n`p6GDXiR37~1dopIZ_iRkA*gR-;Y(6PuFKQK
z{>x72fQn@ctHmoHZ-@jfYVe*Ax7nJBm1JzQsFKfE<+W3`HOIsW7XPAQLo+<J8cF@p
zJj9MwIzjg6n~cNVrxKoxGVjzV?BW(4yZzb)TzbbgO6VPwD{UCl-*_2OZ;`DwOmhq4
z*4V`0df@k2U;*6fjQ;hIR)9inbhq5x<>8iPYekY^9FaaD(^-YNn2JDq?VX}KE-|91
zGdE6q)0p+0h|ueRr${<Ej%FSF1C<=!hwV<4_nlk8O^%{%=&_zgKRbSDlAh+Mq6#L1
zh2W5xF*o-LP`Q8onmg_7(DI7~+ZzGIM=XRazJVm;-J06F@O)vkBf<Hfm1(}(vW+l`
zZw33e2~BeK-<VbT*swyS()#YZ#=vRMEBFlEV%13R<u2|03_pm+6H?8}WXSB)=389M
z?tMMz%^C@zlk|hF{c@sQeMEIX1R7pQWaBMj+Ih$6ahYc#4qx-k4U})LYCy9Ful~di
z7^rMkw?k)LZY}ZLpf%a#C75K7Ceqe?)W+1q>su*aEdniL`>X!-(VZByBH#0fQug~o
z0^{y>^wKQH2pT=KbZY5bb`5$_brfqqsQmdcjDPD#uv8qWTlPhyUe4O~rQY9BO}DcV
z?+`z6*C@Ac31m^7{UN{Ta3;Pbh&Az+?0uJ~H{AA^Ik+CZ>m2bECQjeey%@f%Y9TjF
zlC>o~=JM1%Af^?%MS;XJ>kcVZ_hy_3#tz1h*THj~`OS+Pk})DaYm76iFd^J*-+`&!
z6N^QaoBYZvOheLj!KbU14t<S;num~#lNQ62T^Ql@DH{t^zt~zAxi^?xWE&vUMCd+o
z3W2O6Oj2Bo`6$XxiAUZ@Q7bZdK_-Fy9i_b)o4GOfJkZ&*{FEwR?F$_NooZ4xpPuKc
z-wnN#Us+xtuuEHj!)et#?6zU3R59Stx|Ya|`Dqj#-4K~oAK4`7=okri@?Z^JVg9P%
zFfih3ex*n6$3=|kWQOtgJXV87aes8<X32&6Mcx>#Bkg6*tFU`-Id|ohpHaQq@H4rD
zd)JnlQB(OEIEicRxg%Z*j&@4`mC_(C7!5EhY9CLsG!nPywm;Ft<JMAF{5rUi@9;c;
z{F2-yJm)OPz19Bes7(|8EZ&t;E{!IR0;BKpp<cZc#`;FZvkCzR++i2;!Hr0(Gj+2f
zhosj!Mqo%Xq*)=GUOC&s+k6OfgQ1ozBx@{rXt(H&)`;LeS~!A!kE#+`D!`Js@jF_B
zE*23N8*jDrnd%YO03_M8nw{Y^iSua^V(h%9q@HWO^w=bZr|oFT1%^!1`e?mCO?*y(
zl2V=lZNA14-}_p3MFQz~rW?6vbTcb!IzTO~>l-^My6U2+3doXg@P!(f7Y>93p0+EY
z4C<)Us_bCDa+=zlP{5hWT$g_U!+Jst#;eQLVEW-I^0LvBdT-l6{v)M47cru4)b)+1
z*6g>vm3^aI*Eg)%?`n0OQZ;*;UjnZQ4hhM@uQy&S-kbLCJdo^KHm%3Q3}4?G08C5>
zBWTNx+9Y||hp~gyd^yI~z(?%VuGOj1>hxRtmU;SQ0)vJmVWNK5?PmS>`&60T5;=z_
zEcbdoh>^*Q^cJoHlNwi`icSFJuZ7F$Ne@d3Em(xkAwDw0g{&#gOz{`LtbxO~NGr^<
zGd!jruJrp|sm!<?n-OYA-YTR&0E8po5~@~nTK9Em25?!LaamYUEIkXZ`cVs}Op5Wk
zZJVdQ8smRY<reHH`wG*$(<HM8ZbmiU5q#lJDH8#gd=b|$T@ICs5iY?#?UaP#R}NpN
zuWyWjsT5j1olHKTOx~GB9(zRSz3FT76fPgXxZ5n`>+|x`RPwr1t)<3Xar2eSU<;NV
z*RsC|YwnNpA)Rt2GSaI~w>eE>>FT0L)mmm2y9&zJ3i%K6ZLtXo*?eZ1)qWnBIMJgl
z80UD6TTL(SB4NY>wk=nEan)<?$z7e9Ay{o;=tv{G?SJ~G$WOqB+&JQgJ9RDjH9FpE
z*q67wtsl{GqmnH#t*O_ICG}M!JI7+Xbr544;@Vg@L#wR(KJj#3)uEIaSJ4OUEE_SS
zWw5g&a*mWQCv9zhG`#ukrg5x&GWGDsQ!xDBsj(GuyPIXyt7o?$?=KWq7^Y^n?KFb@
z!H(pS{^RdL?9?-YU>#wN__hluvnJfSDxWGgT4ckSe|<oO^jm*^n*4?3nGMK={jp4(
z^mfXkXBF%;s@acDlf?SNFfRAfS*uft7$VjjGhc!qSG4WJ;@-<ro7-}Gm7K_VSdys_
zyM7Sy&87dc;AQbVpB>b~M8#vFcyI1_uAxQsX7tcEmbY7r@641Vl*nz^Tk@I7wd_0d
zg?la^e%e)Vo}b}&5L>oLQvG3b-)Vz3r<Oq2c~lSU$+#tXzOUM+bMPp>oG8C1mLFK^
zh{{c!(~G4#E>RXp-0L`Te&3_aLW*yEx~pSxH_KdWgi~x8q3bCETj^JrJtR5&drlSn
zQ5Y2_4lc6m9MJ2F<4m+}JP)NK@?;!R2P-DcbI<*BM`)O+>i7nAInHX))C)|>r0)V-
z!k7>@kj6GB)GU`8UB{K}0&p0P9XENax+)jjeU3y4ci(4v08qeHHJs6m@~%m5@au4&
z<>d&wwa#cL0yyqdHPdy4ATA>9_|q@!gF;z};_UIx87KM+Q&=(7aj~g-9$Pj#4U2tr
ze$!<Cz$ihXc4;urSMBpAAM>mnCilftEw_p8h&7$@lLmKMTZt;(Te`1A@yS2@K*uKX
z-|CQhtWl(cv+go>?ua<iP|d!r^i->F2y3FeEQt#iIjV+edZV3@^zI~>!8qx?D7oQ)
zDKV7d*(j`NO<&#}^g0*wHARgdDG`ck<F+11$Q;)tmBkbr*E9qa84{&QeFic$Bwjsr
zzDA#T#R=Vm=*2M^>Go@(!BN7#fjdQD=ex+H*ldG^RrLVbrCh7im0hkY9vS<z$S|;z
z5WU*ezCZhxYku=q<D+AlZ%?`WUD0}n3_iP4FyKW_o^hcXADwgqUy*QsQd`4AwV+WB
zctI~eIkozKN^QvrO89t4bfoltBNLt5!Aw1_7f3s+?@_p3`J&^VM}ny3cUS|*w-w$C
znU`aGqg*(PC$KopLV`9@7*^UsdUG3<jd_Ak`3OYaAS&!6;Et>K6`>S+|3t~t0WaDv
zIyK1kdg^Y}&PA+zdoc&<?T3}=H;CkgpIvi3J;a<Z*v4d;ReNRGpbJy7&YYnlh%OHk
zlXkU2Z3%C-Ms(Ug<zdS&INi6h9=|nXyz^f4)~jPr@UGhDF`^~mM~QGclYBIrd?|T_
zH$hc$u;F+^qzSL*Q?#hN8v~#G^1i)g(Ri+dO_6gxBrWS&12cl4{f;c~KL6|uu8K!j
zHPbyMWp`(MQ#(ZLn-jDe!XCre)e`cvSlRAtZ!USGkZ-!GY%>lHds#3;HjY0(ISx)R
zeKq4LuOq$FpXWtZ5_~2|DJ7)y<MCA)rIli^-h?220vV1+`k0}ZW-F&!Y*lZj6XTF6
z5;=7*e5~Plhh9HRu0GMh+lI3eRlO(jV~KkjFJ+v8G1cT+;0?#js5;q-DMrQVGsj8B
z2^F5Bn6uE5YUVB2_cW1=5cYL?f~TbL>QdRavU*{aE%xh?T+krdhVNdd-q;OY15f7D
z%{Rs-Jzw8Rd8PT*eTh8V@1dbfc9SiIGM=Q!lr%<=sekVC+}JfZ-{y~1n<osCwwV8#
zO>}vE(TK=M`K+&4q`iBLPFKK$pGgRc%S*9h>aFHcXQb!!D%L(NH<1IY{-9-S+s3+H
zCwuS&nz1@r<eM#kFsZU3zKaGG=X^-}rJoIad3kqk``{sMUub*M;l>O~gY^lmfHs!I
zPuphd4O8VypBqCX(puk+ysErGBTwybYX(8yJG&Y4FP8*{Lh_fLJDhKRYUM2|;fQ`G
zGEKK>gx`2i(7x=%9>2@yRd&R@Kz+=~yI(gclksP@i<HcRW|Y_iHhy)pv<F@>5-&$n
zyq7qRZZ24=TsnMqH7}&u*}3NB-0cKhL3V25Gd4~ne(&ABov3^Y-Lx@9Qn_isWUghs
ztMku~Sn#J&(HK*~Fi#)TihquclaO;SjmrhaaINa6%K5#<WIm42#ZnC5{N`!-h3I0K
zSv47LPc!nCIt|&UP-?v(;hZTJY#`%Y6<M7zJX`#tA%;a#rI{lAN+15`+MmQUC!0BU
zg1KV2%=*%;0SFPC!cuh@*b^UC{(7pnJ}u=`^sitED9XWWr1pfry;3PwE{_BIB9Zv?
zw-e3lNaW3dgM-g3O4kBa)Z0W%wTPs_wQOq*vQ4Ux7B!>HvWWs5=rof7u0JqJ>stRI
z{c1sk`)<>BBRlMD!xFG@DxRs&@>n*b&E&NUR4jg|DP#pC06nm`#YfEj@UHsCR6Oq0
zGYVEH|Is0|nC?CN2#cH}hiMno*s-~m2?{Ykv)S>lMzUh|*8jMz(@*Xr2;=OJFaO&V
z<6nO?E@EuN=qlDa4Esle4NO+4gYy+*q?yjvJm~`7&)A(V03Rv)R1L!WToX=@%(thA
zmMupE+QP*S(asjb`8+nbe95dVmUlkvB#%P7v@$DLa^4F#RyDu~JrDZof^6Wi1R|&}
zB8rQ`sX-7XP;8cSfq7OokmmSo$3NGeufCS~91$<Bv<thD(C>efmFx?5bJY2Y)JxRk
z6IHa5mWaw*2U_$z!c|@|&zJpOOZ-osx$ZaYm9HnitRcQ605M0+V7vHeTy=H?E~Eym
z7siJKJ1K580*ik_e+rl*UT?-TSN_Sw)_no?=+-a9B}Oa%wAB5VLMU&HRlMy9p0rZA
z81afN(6x}fINUiB%lN+r-&G%x@7iPM-NpGI$^!oOM7b`*<Gps7KFL!3e|;PNc|YI_
z-5%ugKd(zHAV$ak`G4`z$P^_yQbF(!m)n1RtA795{X{VPZJJkG|M#B!mpk}pd&(bw
zP5usSKzExP-zWV><MVHRG)f2fe1t23(!qZwR`|y~WVr}MjdjMLTj8%=f0z(_URLzX
zrQ`qj*8OpBqYzu+sEwS3_kZpBZzjX%k*WuH{J*(HfA;%6w{OEDyuCkQ-~HFFKkcsW
zC9JB!Jd;4gi=k)m%3r(>%r|wRsNi+KSD=o1Kq!N8&2F}J4cLa^x3TQef9XfVM8{vk
zM%sUC$czYYT0SR(0cJ|VcQCH8L@HAJh9~@QZe{!}Y$A*?;5ZrB%1NOy#MY`#*<y<P
zfBJq1NfcrcsKc7B;=_;4;*=Ts+ePy~|Ai_PQi}ZOD_?*z^tTK5zyAd;7o?Q9q$aOu
z{<Tu%hkU+<Q0MkvEJfu=DJt){>+~0|zXmBqlYv=p|JqWP*G5Xwn8^p>zjpnUdPpf+
zt9{h<*OoF1CsK-5^SYD&+VzKVA*E<<l5OFyJbmce5GT~A-NA+SzjpmM<KgqnM{6<m
ze`&q?G|53JYC2r9pZjaq&nyq0M|<wwDgEE>@Bhy_s+vY%=4lVdQoFx){b&~WJovR=
zgkJd{+sl9Oh3*D1EZ+?=;n9-6cK!c9S^i#>w@*e_Jt;tGK9dd9?1OQx%-7fc`AP-I
zO{_Q46LL3X19^{dS2Dyy9(4pxvxF!6sV7vBSz&WR9TiG2$#nlBt&~LCB^ieKV<nb9
z+ryw%`*b4DZeHxLF#2SQ;wuUWJErlxCU;eVj#rDs&=h_hZw3JYttCKMYk;f-ocvln
zA+U03(WoJoV@JPNNFoOSGKvetFa<K}NYuaxa8EKUytW^Ck^P#F{PnRQ^Yq_e3rajJ
zaP*eiYf|*=a{Poqw3EO>*BLIn!Un1&>taw6X?#F}EGEGU+{~<^$^U(K4oDkLOCJL`
zX*WLXHx7C4(~Wj(5YzC(BDj+Er#CmVH5?KAkwKFlfSUNwRd2>?6)oT?#Mtw-$*e1g
zVm<sxSV4cElm3#-Z<YtwB_D1Qj9>j;$8mU-l=2*B5fOnZs3k%j3kus3`Ps^WvYoNL
zRD*@bP9(PVn$RCPh?kXg&L9$#Nf6(d*9TwIg>baT8FLhL-~zptMt}TG(9R167BXfN
zUp3T;+S^`U=cVse>N}b9`o%f|3?Pn!!)iiLA{ew&ebvDU099D4PnHXC#{<0%K6sr)
z*e844;>Y(-+(3&`1Luo5$tuK~*d0u^31EC?#nui23UP==OU3_sG{ad7Z9LJ$I(+-N
zjAQk`J^xGphqAYhszUABcNY!PE+s@7K|<OE(v2t~($Wn|N=dhLiP9w<(nxoAcXxL;
z=UIEd?{9qPjB~~}&VM+@)~#zjF`qf_`?_fl;gs?oi$A4jk4EVLTf!~|NT{7jVaGSw
zK_ygm@J{e9U^6bo=!fR?t!v+}X*a7GCqBiSdbnwLaQ?O$A=#XK+m$>TTJh2=b42+5
z3WQCNfTeI>xcSw5CJ96WSpkcObE0g`25@;f9J$X3N1Cm{8%Ch_Hvcm=!D`wx%NLl_
zcoKjqVF|ce5)I_^_xF@#&=zpZY_jB%1-RH>$$Ums*dy%|3}!f!9J`=)H~$|GhX0S^
zSXuy4pB7>ubO(0@z~Q?Ik1aa!+Th8L^`XBi-MshUk*Ht7$&RMMgkf8G-QUA=8ze0V
zGJ$D2*#hgm?$SO8iNN+J#^Bwr)Q7j%UEe$$fT8iloj>ZsBJMp!Eiisz!Qp4B0ON2w
zX!dCm5+-=tB8YX;%ONN9;Q}Nj@RA#@>|im$!6f(6Q2@(U1Yjc9{QF?|wFG}0Z#t>o
ztV~%ifW@SuYW{bunf&31yvD@Ca~$M9vG*qN^J;*Q3*DYq6=rQ1qUF>-#L7G%)}Uu$
zu{YuM&#5NF=IMS@;eJv=Vy*nB@-7Xh>(fU~;8B|c+xF2M23*f*L+QkWh-nH|gFA4g
zaSUg6H-P`k13p|gX9YHgFHnBd6@}W(>iT4k92vWx8!rLDASO%Q>BohcyRkyIwN%CJ
zJ8xV8ZWqu3!Uaw9O=n<0_u!;Xx|`h)nDBqm84xZzogztU_cQ3>e%<okh-&s@hI4it
z&U}L<Fv=`};didBW+G=Xx^$ZZr(ET`JNdJ_PHpv_|7GWaJ2b2@1_MHzEo>FV)U|m0
z6oJ%Y86B`*Y$pNI!TE8k$N>1x4mxJRuLK%h04$=czo~h?h-t?Pe|Bd1(-t7rV_^kT
z?~3{#g*w!x+2NWlfw8K66A8f){5-V+ZPII=Wo{AJD^;6j224JyB?ll#s(F$4WTMCo
z9}_;>-_OL|arVcV`1B>;qpcqF#IUaQUxAX93k%Mg)%1_8)1Fl^2lj*|aLhFXA(Y;z
z-cL`|uRDU<laKhHl`6!Bz|;lr0lad^du}%<jON>L8Ps5Pk+j9c^dNsv6|O>*069Ac
zLY*mp2&?APi9G)uKYEDt#JdK&5l=&HTJuJ|L3m)x0v)2XJpzS@#I%@nXNnBxQ^dVL
zmaaSM2p8jbOcrvhr^+%N0Ohkzz`X&KK@pKz&LrWDiY}5hFY2V>pZTV}x%vNEjK%OF
zw@X7}B)!_W|LiFK75O?WMR#?jiuk;zvGLJ<ATIyuhl(h_FAYFpoxVq4)pH4&bffu8
z6fUBdF@I3YNiT1nwtA{ciaq%TOsdHJ9X#_zbFi<8@OF(Z$X+p<BWP}CM--B64Y4d)
zoeX?g>QMxJ?}?6cA`Ct-{C__%M?||et+hgKy8$B99(Kgg%uXC?(z|{0Ty%)Fes8OP
z4lw0j>}FoO(UJ|}{DN40=G&5YlI%#kk9k%lmh=C(-H2Ea;XyuO*6=8}6+6u8ePn8n
zi)XbduYlR+o~y}1u0p}HQPFXLPK1ZHnP0IJkw^TWl0(PqF_9Dkg)+KzUy!;U(<7>W
z_+dM1_liIp21p!(eTI6EA#Q7o+896pkfeSAPtBPQbq4^HX82Kwib?{X#ea9d?Qq{^
zfVBEjQ#FBGkn47*;Xbbhz4aLd0rCR4??N2QbEqJ@c9HNybtlRQNJ$*3@YvnWlp&SP
zj~IOpTsC;`&63=%O)D(g?utm0R5cX;{U>hHuqx{n!bd(!U{3XAt`W@QwxZC~N!%I^
zgS2m(cZ$mrd2kqu;k$4obNm3DD$5;uLm0)}1egm~{SM#3w2e(V9@eMXTY+<b*9@OE
zDc8N5&Y><@0u~I6)@a22EtOBE8D-Na6_0uVwQnq+eYhY+rWjCt9q0Z6ir%GNhiIcb
zj8ye=dx-(;aMsF~*W^x}2I2y?)-3i{*u2|-w{b_he=_ht&-Xjc%*?MRdaIIX>4&_2
zbRq`3&+XBO#F1ip?Y0a}K9pk6;2^j9d*_MklEM<r;K=Kq&$9y$mvbmETG4sj1Guya
zbBdM0JD-zCJuff`S4xfgjxUmN)R{hw2+RP}U0r^Bh)>rWX}2p-GDXkr9v}_~_=3xy
zqwysP+=_RYbWuxJL$4@<Ht@svU1P~>!wF!Iyeh98c{JddlM(eU&$pq>w2JWf&A}V<
zZ`>3F8Lt`ubeJGX#bE_>{m6j8(X#AMHgo~B8lmto=<@8+H7eehCndEbP12BMb>A)t
zzV}S7AGJR0%O(rd6;TWH{P@=S(3SgUr>F|B!_@~80<^#oowVi|Hn`JJWaqBawr@P2
zI?M|nBN3c-wbWkt8$`CbZhuR9y(S@VLKSLO3k2QAtXC|K&r<rEdms(FE9!}S1MG&q
zeuB44LUrtaK4k1##;p|%XMX!XRiT&;160Vv8l3Kbf4ogb-+R|im+92VI40+I99eF#
zES$1^`DDTS6k`^AX6Dv7&c2ps@JyaF^X{|lbp4+v$X81Mar;>xV*m-4j}<-yW{P_B
zO*zfo_CM8>O^(gZ^I6-ZY#WwukvHcR4Re8B_1WNhZ1^iMMbN&(Zy7ft_O?T;(l4`$
zp}v@IP^@as)>&tlelrjp`grhdWwD2MxFmtU<ABLj<e}aSJaTU9J6N`4z!l<)q$36K
zw+iWZ)d<920HGbD-SY4S)I!*BDNb&tqs@XqL4`64O9F8U7I$HuxTpex_eWthzD>&M
z`elRZkZ=UsKO}DZHp!Uo4;Sv_@Hs)KA`QYK_}45V^<asyeQVKqLLPwi@;DeEaPi@G
z+GgWn{2Unk3DfA0|B~@MH_)#-@H$f1mZ^bC=yzzE0w)Pxr_NNA)8qH0&x2|%fjX7T
zIR>{A^c$|2d2~|9hWI>7+yS5ExFk80(#X#U>5R$C+D8IixBk$V#lXQ&LH13?m}$_j
zPpul_b{}S>8no$qepK%RNQc8kF1puSq%#6L;;l^mV13$Z%tb(s<ZOIejPGgO%Wk2M
zW=E|_14-AC8&ATIIMTq^es0~0-#{mD;a8n(4<!wj9`zs}53j;aS^+(ty+}^by=_p|
zigmWX%K*-BE>;B3+Y4a1tR~SBlfM#jgNqfNAHBtkSL~X(`ZA^_xSgGbm>5KVD6l;I
z#<GfMZ{YrZZsAU|@fXK3%`wW+VLzpX%pM0aW4u{OKt-&hGW{*!bDz6}W&gZLb|5)A
z=nyzBE&^d>H@dZqs&(JFCUg?JxM~OBh`_*Z_q<Fb;aOlj+PrvwlXwbGtu(jJdOAO%
zMLL*{aYV7FG%P1T8?MrsJj->(uHBw-xpUhbag2#TjWDhjRX=p^x~-7>$iK`{<=cb_
zZ{)q1PV)#5yzJ`H54YMZk?QlB3hUH>0bZpJ3BLBDR%cL^UZ1QwMm$K_JaTC`<kJ<N
z5_a+Lio*dL&OG%#``V2U3`?}Q*onLr$=D7+iqYCYxTXH_6)-6iMQOV^x?9p$ocs>p
z_Aj-V8gShQ)liaPm;0@}!%@DRmdvt!H}N!fEeQ}q)lQ6EagJ#gemduBE7`x>arx&-
z;~608VgFT`CY*NcsP0`#{uwAF|60UcI|GM0QMbd#9jZ@wp7gSd0P1jlk2&3eo^Tn(
z59IAFpxWSSQy^<L28PjbWS+-8$SU>LVIU>&ooN<%z_~+Qx&r-=M|5lvM}-g9g?Bem
zBRiS!&1%;RVCwq|Dn$mvFsNNlatT@6B-dMR;4?)0F1J}~xNG-EWR6dG0VA^e(qivv
z=ZSRc{@PfGKnOhFJ?(e!5=lbwQrEM8cCE@U+d*7mZ6c^?Ji5~U?|>bK&>ZoUfUh-M
z=R!)&4zt#N6%DMdDWKi;A*u20BS@^dm*Hck6eeIZ^F;K)eT<4h1TjwWXg7IQ)UVg~
z2x6L``wAfs<B0_4zQ1=Y2^K5R!u-9D7<~`vOu=qOnf<}mDJ}Y#u-0~!0;l)WMCW`*
zQPd9jZCD&{dp%fwVMk%XlZ!x7Ae}8b_gm4&5ccpcA#7z@@Ydrn?sh|MuZODuJQzC;
ze}6Fij&Du>&kk06@j<yv`4`gK4ZI>p=4r&Ux0wU>_urxr{}i$>d|dP2(t@h+PFcS^
z>*1xRFcL0mPHBhx<p}%JuZi*$fxay0Kp*uvBCmjt0)@r4I&uIAU=(}58-329&#T6o
z+H#y8Df_H)qy(!cTHEc-bMV112d3|5$_UjZPCrQ}XTvea_Awtdojp+XVc%MG(@FZ9
z;-jR;lZyAAM%Sp}hWRmMcwO%!k_fe!vg)VrXwAhgI^T&CJy@?hB?Xo{zLH%uVwHmj
z*_gOV4m7U+k`{^C-BD{*d)nBx_o*5ry-u>;I6VL5XxEJRk<#4vOLny*>QGT!{?X(n
zz{42gdN*xT+Ph4CMIuk)^(<)x6t8Q`Tz_1sMpv(>2*eEH5wRTIP0^0`b=Bfw&zDO$
zt-LDHJt>!|eYnfqiB*^GmZpqF$7~mA3GgYbIoQw$Y&i*0hJBShQa#RiQTaRFe@dzf
zrR>Swdm|YVJugB=S6}|j9uAVhPMF>z>3ycfTzkc)^CH~&cVD0D4|if@SxReeunja1
z#betl(5_;PzZ)Vky`VTwuS@*<=N+Bk0*~$K!`{MuzD?i0AHK`Wn{Vf3bNC1_(#&Dr
zX+?PHs&n|=_`99-_IsZ`5c5_Zv@#DW`ZAx=RjzJ>c?Tv$BT=X|p7vZx@@<eIJ}RNY
zoVeRs-9}=H^ioZtL)3-OfyFgqN%NXnbuRM+{Y|`{+S$C_zfhmXR_ewI;yu`Ok7}HR
znQ%?4r7U73&!&UXM;z#z%>hHi$IGsuR{@AOfZ6&u*+pC%-4!3XUiV~g@pG}<Zm9{i
z@eOWFa1V05@HKAWM~T=|Ks-UAEekPRyotTXqRqnJ{%4n-VKPPfM_RgWC$Zo8Oc}d_
zTKJnb$(@_{O2L$VyhcGh-6m$%eMSQ{%C%6;x7eqdtT|6H{pR?3<W*LryzkAL{RB=a
z58l5slnHcZzYCpkp65(s*wXpqeUG{wL@V;k*<`soQ2#T&N$RXcHU9@+R&zAuF{0@1
z3wM<Oxefouw~vdOQmiphd8g6D*1p7x2bmxZ|LU(mh&J)|8WKR}0Y4U)B+?I|3h|s5
zH=$O<ZsVI%^j{<cuE7v$e`)}>%iJGtF=+PR@xD(IWlhox(GSK1oT;y`Q?aQk_DT>4
z8opyNO_v+1^s>uxVj>|uSmz6SVIxb#zj)pG+qc8&0JcN<$-8Ytr@%c8<K|JAVH!5x
zwKXh;tgl(ROihinBW^6hWdCQ{>ul;H#(!IhWl&bOcV+#N=OLn>IGQ}7XMXn4Pd3`s
z9(OaQVSh}Qx&NY^mx_UwYWKCwyrMtNB&w6@!W`*z$j93JMy)t3<QwqJOYU$7oU_dp
zCu5JRJ?{H!wNzgg!82vVoH174V5l00M?0+(++m6IU~aA6@qzP;?vM`}@Z0&N+LkAd
zF20MkeWE$;lXRbupY{iM#$Zsqt2eZ}$9J&2vE&Ghv{Vd<G`sMpNgWc~?$cT@e2O>|
z#%VFHa{Vq2BTwcyRnE9dwz$m<bN5Kbvf^)Av@ixza|OqmQ(rW0plW+N5D81fW|r6y
zGpPy0xJ?b1OAQ#BF3);O@+TNJMH_dRx#F0q|81X^JILr4tNHegI*e4zB=LR-+23`A
znWquiPD@!WOSCL@fz&U&u;Da=#ZgW2zIS6T6v|-1`a8p2|B$H~r9Qv+uc_|LjE{VT
z!;p{r&}_TJ@n_hf>*>J5?Z6On_0XRt@g!{l)sBh=lG?O6#I;mAABUN_&hwnZP@KH`
zmp5*^V9;|f4aXBb^!Eu+K<_2$&FOKkjCKgNTZe3?ESu39w4vEfS1r|jy_R!4t`OS(
zu=M<{LB%O;Kf}l^d5wSH`73gbf4aq{-A7waqqTrNEduu85$Nz0S<V1kOsZPe`I^>(
z`WD|E(9YP>=}0`kO-u{>GAcTC5Xi@Jebn12;dT;JI3eSA+&U|%oRMs`KV=!3*Bp8M
zg->*fPZ#>5PCL$t`!3dG;D&dEr{Q$$bQ?D{Y?ANQH{zDca+!A|a#$f6&Z)H<c2Lj@
z4OUUvGiIt@U;sX?u<`(*D51Hp@s-_Cu~R!LnwbX$+mq5`U$*)1V?HAn1+p=FMmqpq
z(iHNDn&oG`hxgc=$W~a)gucM1<tc<J0k=64bE^YqF*Onntffot;01&e6M2XU-xe!7
z=5>S#2P{o(&)<kLV+DqH66;)G@@UiZp|@I8l24pu`w@g#hm8>@qP<ql+nQ{>dcya)
z_{4iep;BfA?(DdJUA$=x83`M~U$8Bef1ag_7s%};l%f9ELon}c`VaesCEW~K4~8iE
zF6h+5J!q)Ry<Wncm>&GRlD#jIf`ZK2?Q~talfa+%n6U3f{=`@rbu~&nHj#OGrF&U$
zAOz344bE!ld-ip8D)y{Jxh9(=3p5Z1=oC+(=dl^<m7^cuSNZ3tYuHsx;1j<7Rk4%#
zd-JwrP3$owPHER<d^RxEc8ScZ)S~H+gOg(fAzIc-KzyhIa)bBxk}Zn)TtuOI>~906
z8zTY-1l&0IFz-VD)$jP5wi1ngDJ3R)xNAXGxq;~ZfwXU8#Tz?YpZk=}Auf>m+u_Te
zkZrUm;z4A69BgCGB&ROCZ<{a3z2;Lr!aWX@N|9B|W-7)4(zf8%xV}_l&YlGR86r)n
zqOPRERW44?ql$cmf9D0X+@7Oka?n447SV3Tw4*P<;WA1`Q9PmX-LQdiJ?Tv`a0ZDw
z>g}JHiM(a!DnC-s1-{S|2J0BfM+YQE8nd+q!BZ96mk~+RDh1O6+<$Xk+Cf8vplkNO
zjtiLmc=S2!3iD6Ib#n1y2AMs858vBjoOT<l@<|Kw8`^{#lvxDbz$w&~X&fn|5Aaw7
zF42A-1VYtHzGk}=wFyQ}`6N38vr?uzK569@JrFC4bL`wa%3(2#o_y77?B)8hvIlh`
z{K}Y0z#+o-{fpT*21o@PMgvKE1s>+CEHyovypxL(jY@+l2+`oBRHm@d!uC4&J?pIS
z7&dJvR%J%B7|bHtDnG%UpbLf7ZOyuHoH54mkpo=-Ny#6^POS}-e`4T+5j{)xwRTvG
zQ9rO2JF5a6rI*^>Nw(Ee(9Lhqw3wLNHm^Nu*Yf+AHR<6~lisT}(yFUlSa<$tH0O$k
z+4*@-<!*Xt<5Q`5SLXd!)*MoGbFUT)o<_&qZO7Q&r!ZN+`aHgkXgUSd^ut@E4Tq^(
z83kOPvXakuWKB$3Qnohj-%Jx5c^3ct61Iv+<PgNCKKp0}OTyuc@ciae9kF<XivF@W
zCzNBP%th<5FCrOQe#+B+JF2r)%N9Rx7@%?8Glm6kHj52U#GI1K#y-Y#xxboiic{Jz
z*w=Ejv`&MgxNG9&G^{ckPdFm7YfA6BRqBVA-tmz}KfH>^>92^b+CR7HPk3{+=~N&H
zr0)CAV%RN!>SxE1i8;(VG052qNZxBVij^FW_q=PKYXq)*=OHnvQRrV{W<DT8P<hRm
zU-qM-<4ShxsnwTY^=I`8#|ua<@u0Vo--2QV6VMowU&jGIRredXOZus~|8y0+6yU@Z
zz3+>?Tz9`2OS4-xsow3#f5O~lEou(!KxI}rHf64v|K|vuavneW_Xve13w;iYhp_g>
zu%#V--*U*5i#tfOK&bFGLg2w8XZAHEiK}?_u-=qmRWHZWrjOWqxEd*!Ii>@3_flK5
zywOUN2@C3cV!~xs(nY&p!fpO#yW$;BuZ=M4mpk){My_<VT8pz+Gh*?LM8dGcyLX?0
zG&qLstHH^SbT?vz)mX&@@0iW$M^bpJee0GO`vk*momb}4|1sE%2Z7)#N{I<8BG-pl
zrU3#c?O~4qH#=F=F@SCDm=ni9t&Bf^NB&g6JtdJGEE8p1{M5wwI9OqkL<eyH-{+Y(
zz)3ijfb&uJs*?!Gbq~hYdSha8LYU#r_N0DhYoiE3*FowIo=RF@3MbWtkvmkK{`ynB
zoohb9_$jBm<YfVE`3r}_!_Mk+roebxy67q+tmcIv;5%}3OJnhI2tHsqj#Il~2u%H(
z|8Cw$x|hixCBAj-=&*Lt!fR2v*0iaX!_i&w6P<hPi-)pLm+I#=UhJEoDQg-Obb(Ij
zHs$6~j^x%bY+K*0m1f3MvQeTzE614_T5`})bRzc4O=O#F-p?$>jMWl<N`|yYrnFZp
z#UhEKfWi8M?1z2@*oUNe;h($ac1*^^gXFSPPQa~)+^*s1w*{?`O7e6H{gxBio92C?
z|B()xdem`X2gudOH<U(np)_^(ey{Ef%UF89h){mM7BNS3UJ`&Q*P-?It%p?Gv6+iZ
z3srJt%1Ej&g05FZ1&W!TAuFYFQP-tLc@gO*FT?$p>O{2+%}8^uBdwDpBc!|LpKbJm
znW^Tl{|ME4U`~jUx0Fg$h!6{DTNc_I4e=|1=kJjs{lT#{R<-O++(j@O=k@vE{2}ui
zk~!xz-`zB!(y`e6r%!~QUGH?7A3?564DfpXe0btkiA4qBe2+4p!p`NLztJLDq(*%t
zfghsxOv=`FP`IgsfP9LRwzot=cS9-y5^ipqf{j4wt9U)RO$&aTBZ55H#P8A<<eNIb
z5t=BT={346YF&|VQq+=&aZZ4qj&qu~7iY{&piV)D%OI=sfD~jhkAsoal7Pxf&G=M;
zne_30oU@jXiF&?stB8g7p*^;HN>1x?J9uSg@w_%|32^j}Lk)O(w?<jtyPAh>gXLs*
zL_r>P=c5aM@&$!$a*9YNVXx3b#BTkw{c(kDyyNQfLk(M5fnxi(6CJWOyRLr;GDFT`
z6Yz9!6%bzfL|*Qm*rBluyX?0~OYBZ(J?WPC#>I@Xq{gV?{Juk&(s}eNI#gn-=VH6!
zGkqGXsHmbP-dNg~&4jo5N$g(2&BCHnws@9Gz|Gx?Z4}YNt6FJMD3nD^$_Yj`4~<?c
zT1QX!(oT%iLSV<C+CFZk6aoGCJOZ0y4PcmbTSMzCVE8IB7Y~ojeo}Wz)3?W{Y6Puz
zQ=|WT4h19u{a}-IC#>WE&5NV7d?&GP$~aMzObnv>IyAHAo(8f?!Yg>h6rY(H9p9}Y
z!xxr0l6m>>o!K7NNU?Y120%-QShDdu!i0}9t|J5tug$JPH=Og|`ZtPMOz)ol;*bgw
zuxw?WvM*MPKJPk^ZZ<|lNBRorg}N)=X-Xqa$Q-n&tU&u~4&wBVU|v;U%+BUc5}Pl0
zn)y&anch&;rm^t$_@2c2x2y`Fwgv?0FTK4swS?n#`Z@lNlvep2sM3nWomm+_Qt?qM
zJJL;TRd7)`tP9$Vlsa4IZ@SThqqfKyDP#-G3DR$8sGYG-N)}h1b-hHiX4PdR$9u;9
zGMK`<X~ec+74sN<Bi1f0oG@Z=U0*2b=qarYm4*H=ITlfhM6d);Uunbw36wohL^Dd<
z8djViJQ;M`^bK`S)@FpX_mk72tBy1}P7#gZQGO2%bCQ~Tc3j1q0}mt|tNGTROM(gd
z>vDnp$`zMdhBeIg;(oxIN}Vw^XS~t$6hg*XMM1oZH$48<z?PGFthBSvfoV7r@n(#$
zXEh*m7(8x??31#oXmm9q9jH;d_1Zt<zF7?~%Q>u|j}niV=!3SZ!Ejm~f5!eKOiy&G
z(st+8Zy`$fU7K;6!k4*_c#U!NCh`E@CD9?MuO<7PWF$fd{T3TDeNe}!H)5%tcu{K+
zJ7wUi^_n!;_e0jl_B0F%L5||8m(Rf{S{F<%n5Wlfa_5ogxX#=LEvO$@&qW%Rflp+2
z?^!nm?fb)mk5LNR;b_3yjxwYR{oQ46G-i(tdX-#z%+sx9&ZP(~Pr43~?OvKwOzXFx
z*G`nJY^uX>BLy&gzj_sO704tXfRWq&-Ke)#!@H`)uo^SRXu*cwYRAH(Bbw<C=nD90
z1g63-ak~UfQ>BMq=dK5&R+h;i10f+}Z^sLwKS|6UGE4;K>mNC-g_rKs#7Xv6%B-}f
zO`5AsqArJXAEy0m%KH6H5=n+oJ5%jE#55Qd&pR3BOqT0R2wqoL%N6w#a*dfa|5dI%
z*WG6xw6Zboei*kWsIG#vyHwge7Q2B2jdy*xH1<x%zX(vC{M+`X_o;@^5@J=C4kE<(
zd)1W7TV{{G<M(bV9X!lwU0mZj`;%GN@aveh8@YY@Nrf7b6=B49)f4&#`uM(2pEw!S
zhJ03s1XNSinghqD@XRxfJJOnzXe|Wy^fUFf-MlfZ`0l9)f3IvI#@Iu)%KLk7Upf4V
zf6Xdruc!eS9>TRCtu77wnOMB*HnVvCN`?1(d+&g_RM*-zFsXK0Y;dzST45>G=;(C;
zFUj6J94E7>uU_YDj<wM}Y^4yQ{K(>sSBLw@_3^9)b_sTW#58QD8(W0dgm%Q#UxHF6
ze><)P6)Dlv?hq@1#hgn1+@3_cTvp-XdrGfNlbGDSX6qdR$}Hi@oM%t0uvb9IGO(Ub
zSZ^v?A@JeP59yAR8!q5#^iMb1ca-*9G`=WP)9SPMtX=?>95HOLv$*E{g2ZTk!M*zn
zEOgSS7TL7re+Q04&WPS+9nMvdaw&=-)ZtW&kyJH@i}OEfSV6FG;;a634d`casI_?f
z<amV8^j{e3wjJZbJS>N98e%YGO{;aXcd9!>p^?sPQN2(u%X6v4g&<cZta;GFDvC7J
z>K<_r@x+XRUkeCSB4C0Gf-V(SI5WV|bOC3`m^Mu6d`p*+yqQ7mUjd!t=Q95ITPj(}
z$;DnSDaoyWP@QLu%x4M>p%zqU4Qe)l$+xR?b?&~#`rbVe)~NlIgXK%=Q%qydHA%KH
zZCST3mO&M}K4C1~#`NmO>7HU(Ts_%2qtNRJKVD;X;m}G)!{^}Jen)GddDaX?*-6@+
z2~zVAt+&LmUUR3(I5NNG(1V<l^pm02ETe`~_6pbqKdmpii9UYa!oY2}+C)oZX>Cmy
z@=3K0<3IPO%?=jVr<%1-td*HB3J|}~mr0@-h@;2|M?I(wcy>!#6i^LR<d#f)%r(Wk
z4k&k;^zJq+^kK#ef6Z>E*1&fF>o3!Qk#5zVs)J(XS7<K0SV6vuVgDL#9Cs<d5l%i#
zeulgtdncm66jY$VC5LA`#hp#3W87|%Dr3Z4ogZp$6foMOzKsa|>|7;1B{wH}B|W7#
z536-c+XV{K_&RN}cB8(0>Uvt(|0r67-HQW4`2^#YN=yjuKRNQQ9i4tbyh-ZJn@5U(
z9l6E$56v(Kuf)98lOr@ok5Avb!r>Y#+v}xD+R$wGGC_;mEH<w0-ZjfjRLjy7b#gQ+
zGtMSIY5GG}kyIRT<a8dgHt?#(A`-Rs&bZmsove(`#P}oK;C{O1UvMQ#lxpx>jVg8T
zxvU!Y)ZAVfj+-K#@R`WT8y%BI&gbm1m|tmC$MQI)_TFu);>UC%$K1id`<_r(r9#NC
zh0Xssu$cl;=us=MV3x~fS${5ln${-}l$mN_KEMJWyJR4HB!qc{hpoBLNrEay7E(Hl
zZ%Q-Bnx(^X%@O1vkUb5aW$p_ye{e-1F=CONVkeXU$>ZI8rLS6{7Q;-FBLDgi!b9rL
z=Xkh+LmCxPe+thJ&5py4Gx57EjzYzBu-JiFN?L3`x(yOZU%u$7H#@o6_sV2BI)lQ-
zDp=Z-udcZF6_&_ka7TgshXpFx?H3xI$$T&0#|LadpI&2p-el#<vrp0z*zaJ#iVL;4
zP*_7^!~Srj`>rVV)6i_)H#9=>T-cYQnKhTiCL^c>Eq1U&HZ{bI!4lV7c#1C~$ygIu
z%ehhi_G$`q?gS=rF0amUI+SHZeu9x|q~t=vzxFYyUFzt_t0`ZvlW=>6lA+#j`iP3j
zbdY<?*iPCxcT_sWo0IiXNloYv=SMtjy=A_Z1?>#fvT77mzMBA%nnd2s^RD|)oKkpm
ziIb8$4~pj9FB5eKjLBxXyM--Z2QU6Hs>44XGQ=GIelc^gm*Vbl%G-p1S*h3o6-p`c
zli1p5T%)}<G#k?_j#w6!^X1FY^VRtB&G-cU-eq-8Owu8yIk8gOmP`pz=vGRI5Xs-@
zLkv)48J!B;yCCA3#n6!-3~*rzk^Z?{XTxOg$Jc$1)Ltj@xpXj-D~Opn!AJj*3zv1_
z)j5BpXH0@MaBgIZ^17LJP($eB{&c#s>xRLoy|2VwpGYr?uzrqmqcTrO6^i#2pgh6B
z-?|t;-JklSe&u$3ck<gb>gsamPc*)>b>e9#Rmx&4YkeekFkx&5I-$HlMb;YaF$|HS
zHEXI-_3?7!F89*~3V*_HFmIodg{N2Cd8hPaT`E^QS!e(9sw@D9<;V@dbrc;lx89^L
zlnkWd^w{>@T2484CJ#7GalZ@`)LCgd4T-`wdNF@vlFjuoQ7EzBffLc)$DN+yHqEU5
zkIr)#>By?(6rX%W@9O-{_=flvOFYhU_Vj?Nt!uebl^@Rm8dT7vX2>#3m2F|ebbLgk
zwVxZ;G1o*-w4&_ilvYLJlwyq$kpykG)!UBY1t~#Gf$)9<A8!m)@9zv(4CD?|$IR|-
zzs`8>>Sfjf7%uW{#B`|eiRck1N^`S|ij;?60euRcEO;G%*^qi%N@+nv^91m6WOeF_
zw7tW=KPScQy(Y=~pbUAtl9F9YF~_3EZ<K!M8G8T#TcB#KUY}=696xdfDx6CWHs}=T
zP1!$2aEZi5RjyY2#K9bMqmH_fmf0`Y)PNsBnrwTkP!R7N-z?d}^gNV<uwG!L>X2fl
z6YJ};?zoz~&Nxq0?lF1J^PtB0S)@y*E?p2}w*XK%`-x|^I<}bT37ak(`c*2fQ+v0$
zGdt@64CMYf)O3tQX(t%Ej|bv6B6#wSw^BKJHP*fX3Q~NyOS|7<5#k<85+M?mp(wbf
zckR&L)SD%}bV|C~!O}wrw<+4~6H+Sq?rHbd2D;b0<dSeV-jnHlNfeXHM9;wJq#3qC
z@6H{q_PY+TV^~<3(`tIBmf;3a0Z>th!xw8(lT)7Mt=Qsm;EldwRSTT@g$^NZ>_tOG
za^#AMP<vw1qd#Un3B<qu-D88ge1NYK&wMkY<HY4DB42N=r><b=4jo7fIDLuppP{8N
z#orEmyjb;tW<k`4tI%$=>GWNFJ-+dj8kEqMXV5)CTIVOeG(0=&b!#B~;rPV)Y=@wY
zx3fHR*c0zF2r@dsU=ycm12&;~I+$K@{${+V)U`C$Hr;O8KyufP>cY;`kMSsZn?OJ?
z*SLUphuXbmqqB@_9+--8>LMBce#wM+-AOC596QRrh)=22`|?UcWKsDEUbJ5EHt-DY
zPR?=mgkp|Peo)^1H9rI-gVlMLzlyQ0eU5+U)VhKY45y^`1M!}gK3J-Gtsj*hNsq3o
z7tA;(g}!e9UA|$*t%`Ta9}``E6~sT}|Az-6B?$%B;j!2+<4+fkqq?yT_MZyl>HM?I
zp1UNoN0hvsjNH$gVL=q4o!mfam&Ovmjn5z1v{~b)BM%V==(z2lvGn|QJ10-%&MVA#
zxE)f_A$j9%WqX^{3H;9X1mZL7D_3PNx)fR-L%ufg?Db|&zlMrKVq77oL#d{^3CQ=s
zeuqpZqBzfTSNB|I|GBG_#&^&m@!e3ReRVLg5u-YMfIuJ*4~~9I82@E0?j~Z?Iw+hy
zM6>;<lE%mH9;J@ucrUfy9pfi0|KusIROXM)L4$06_MhxODSuMx8}vRQcFLw>=)0v!
zd~hmszu5s<B@0Ih&IcDavja|b`*%s!$#sIxd!~B>S~DeVKLBC4VG1s`*bQ^P)xV8Z
zI|p#tYU`wetOZQwN{W?O?YnK25Tg4X?T62N?m*0D2q%oL@YO{>xB_)g6kG}?bHGy6
zh&bguf06``BCtD8ez>LEuLGQ-DY#f^w`Sqa;(Y$1p38YGPwQ775a{5;7Nm~>KGqm6
z>22Ks;>eI2cwEHDUgiQE-)-#<l*T<;Cvg*E(@<16`WFUb)eXVeRZ_#!>BybV=lv(R
zCd&-Sj{4mI)YJf;=r)z+s!ct2#kb@f#8mPYU{Y$@Dxgu0Sb^lu;LlEmbKr@;;VEk+
z0jN8C$kEf7V65eG)YfiZwBZPz{BY@$88~Em;gt8x4InXbH-H=EN|lt=UZ}g@os|OG
zcMAS@93KxnJVYqIXGyg)D2jRnZkKSm#^&IVaEYf$GO#Hcb-+?j!-12UNEYB}Eu|Cn
z=YNO#6Rq2cLw2r^hOL0e&alw!EDs*!$6EuQ02~D`u7Ea)c&{61jUwyL@-0JdwzTfy
zGLBH#IZ#5GX=LUDA!$u1_?}j#NER;J{mJ(%)YMTWa`M5NSN(bbsA>)?BwHQ_!`hQT
zTTlvw%eg~^i}!FTUhwHi{gI@)!^7Q$aash6(n!Sgst-a#*+T0T0I!%sCDQX==D6Dc
zuTL9qw5Gk<2sK<qTnheTEY3g%nw$tm@>2j+#X%*wN&Ot&dH@Lnv93VH)d?4c2V0#q
zJHw$B<$#5306+j=?v?X9HXk_^0TzV#17l(QCc@U$^+W@Z@RTN#9?ezTI*T>(Yh~v`
zS8cinfZCgvap&L|%hmC^>~t&e>Q@;1+GEmhBk}V;f0Lhvz(o<a)EOP%uZ?)uOjwV1
zcLP^rQg<)HHP^d91k;Bjczd|Hk>xOa0J1+K>wpBfAMV!Nj1&Mkvn}P@)SA;RsR6)Q
zPtpX61epuTl~vH!!C`_ky0WQ>XKbI*u|}x^hHdFnWmQUEFujhy-diQrXKq&knp1a)
zM5&~wzDL++=V$_lp27M)4)EnFnVeM5cg+eEzz(AUdXseta=3<Nq33Duu2PuOX5n#f
zQ}DElGne`TYQb;c&pmFLzDVw%r-fEOz!mh*_7wBV^OSh!gXaG_mw%qqMvK~8&T<wJ
zLPtlSWKBTVK{Awv8Fv5Ng{RPNk=vWKGmGm=^8r<_5nQDV5B>Rj$GaLXy37kl^{~Zo
z%ef8!`HFKZcPQI5SnVX|4%4O88U40{$+18?rJtY>ya6;-ZDs&s_r=s%r*}a046Gyx
zb*3gKAcyM)&d{WfBqG=Q@a5Wg3s*=)l2XoDUc7hrdlLa9<yNiwaFziKAl80nM)(yM
zrxErB+5P~`8wNNKpQHV~xUK+1-IbRz)KI^^y4Z}+PwT)H?9X5q`p~OSzLDLV@a3x<
zhKzce74$LMg2OLrU<9~?xsF4sy^qwMWpRwh18JH>uwT&8yN$^+fwLUBjFBXFIQ#)M
z%CNjr*;8pKPgz%4MzWAcm@hXyE%cb}6VH%b62VEv#BH{h;nB=i?F4uF#c_k3PXun^
z9NRhJ^g=2V==bR=3}|Q7Dx5bQ|0-N%37fVq<dM(l@1;w9=bVH~M|uGS%y498DO@Fq
zB7aqp%&kaIeuz^<Pp~J{=3@r>*1r`JlNjqKdK`bb8EM|dIn?cF4jGjPIpn#?)HpcR
zU4Z_gZ(r281AI`Tg>pgxjtd<L!NlV`K=Q8I(YYDQ(_7S6Tgb4~Z=uZxaz5OVi~{$a
z;eZ0cY&73S4=Prqqhl}5?d8b%^~m~=gG+-{c*2Z}UW4^3_r+iJce}=d9Ld4#$=QMG
z0w$a+m)5(H)^NmIEVSpi@HHtfWs^(EkV{FS^&Rj-!{|t3BD>7sDrw+Q=maJWBY~LL
z<hOBL)9VI8H&+Ah7C<&u2ajxTcW@22)V>~$-UW|K&SaXh5ZH7lF#pM)cotnPPP?%G
zISK*hfM%C)Ga+~=e4NVkIw|4|kYVq^aVkDbtK2hychrk!5`jt{{(9aW08DxrZEcR#
zYjN#m4LFlI{{mr^%{uf;h;iQYOXQ41AVQZBt~{a`UJG$waU%0>Z+jrE9BOi5pRYyI
z5jf<cV~mq~KCOL!9Nn3xeY;G!9vOnVFW@-+jo<YKpqJP~8^8%kykOrFkq;6{@zI)X
zmp$S1^!0Y86?KpuVnK~X3N{f1GA$KinH(Iqq%b^JsV{3^I;NssXFZ1<JhWfERArst
z?@b-XHyp~nj1K2XrRHbm3&B%_OsGwfOtVZ?dRnn_@dcAR{VPurpsl@9OcLacU6MoX
z$E0*)lUN+RtW-pFb1{*GW=Cy=Ty%{@rkQgwJa(UN_kj@sUy^PyHWGtm?+wD6EsE(+
zToN!}_3tlmJ@$Zmf8_L!oK+0TeqVswx>%#h!6peeag)i06w+P|bSX`H5$WVxKlW4n
zjm*Ytz8q@OQy|+oIR>yv67xzqqF7ujl8R05aJqC3sqK|FluXxeWe?fu1~O)b2EMj<
zRZ&j6{1)Cs`<8L9<TNNHX5USpB}TMLe<yfty2^itbiZG{!)(*6clul4Bz7QO<;#Pe
zp8D!5SFY=z)|oEYx(_it$|$6sj7NS~5<h~+w&tBz?l!V3ck**qx?VA@!DzQ0^?5zn
z@E|ubYnvYhZ{$Z_dY^V}g^j3W{3(Cj&13i#Y?h#7YoL6c0tTU>?W0fKCXZ%Ag4ZQR
zvK2N9hqEN26I%)H+^)AbpSn7=ql+J*Fsw(q$`^9uvu^WWXX>EH2FqQd>c_)jTM~&J
z)60=R*CYQ%&?a-2Ie-WrQo*ap<Xf8VN$}-8wxe1Lu8z0HBl=M9(%h5~Ok!9_L~&Lf
zS*vgM7qm&I;KuW<tp3b;ZQ!n;UPzcVf?poI+hT0DTK#@i!9V(Q_aw)@?gtg=X>@aA
z^4q6bY}QBSkxU8I2;Cee9=Oo8@3z^Og|c3%j#ZPBAnoD(m}YLDI>k8tI3Bujpu%79
z@nXW)aEx=)6)ZSV8xS!elYzmnO=|tq4wWlao;R%5Ij3Jvyj~?7(XH%G)a~0{sZ)m*
zHwb`Op!J}WU;*XW!3%?1yLrbAadH;^6=={p=1H&{t*h@pXE$2=`YqgDa?RD%-pI`|
zm73W3^lnFwhbUeEC|Qp*j<k*v>ap!nuU}kUdEMI7r}PROM})Ojc4IhSsgs7qR_OFL
zt4<M8!RnhTeTu=Iu=}i6?a7)x`Ay8w!8g!R=j~@1#I^Oxsp9a#-!d|5i-|eXVO+YY
zH?i=M9SbqcJGih(yuwH}31j1j1<4M}5F5In<MdvvW-!mSpRcAXoM0Rfd9%^^=KEUD
zfx}Z*mNWQ>rMiizm2h!AZRj}0sU=a2$gqLDN%<U@2ROS$^1gn+stDA2+VT!eyoT~*
z59b)RveR6GqhwDbv*^i(Eeo&YQF>1MmMsgdU)B#elIs5SRkz2U!8{Xyt=A3;rp)mJ
zwd1qs;wjdJUF61>?ja;;wu5=Hia8u7QrW{v8!b{Db(!85MUJ`~?F7nxr%*qJwHAzY
z-5onGrI}5IFeuk`;O4lARKs0w_4VshY}2fz^s*nycX7%C7fz%z%r<V5#&cT|qa3oc
zxXyoBhC^WqY@Sd^AE}((Qx`!eC9H2uJ2j`~*RzxRLj&;Xr{?~41#{x&w!YI_bSzdx
zeF=!chfS=t>~yClZz&^m1b^o|bP)}f1CFlf<RW9Qp1LuVtzge$=Nk^~B|>OTIDTxU
z0ivfcvj__2R=)~B?dDDn=|M!R(djUJL{J^ViH@5ME?-Y*a3FevEGRBwsTdD~ly-wp
zS`-tcO9nyJ=pI;$#LtJ>%;-S5JD^L00HwvE<o&wOD5KYmGK_)xgq9`<`kPN~<EJq`
zx)&bIWHm;$!wSXIOX3%w4>x4%BNcq%#%!XqR<^44Hs&P`G7Bfz%c<{??c9*9l4@go
z!6Kr}r}HFgMHZqO8FVXuJiP?^(>$muL|vv_rjaJmrje8bRas3@dt^jH(v`p?=iS&?
zgLh8(R|n7du-yk8sZvk9{WVrv+0-1m#M9}W)Nrj~$da#pFaj~E3+m_aOxDqG*|%Qk
zKFHHwgU{^Vl5TC53x+St#rG|;NEL@|%oCCXaeey^OLS;tkYW=YK1+w<Np+DNwl;dm
z9?GC+0kgwbX|tk5oiZN*qE!L(ig5i)f$Cmz$DrSRDvNQ>@ysD2@6f(`yV=%o4l=VS
zk6{p`e6XM(dhuIOf{eD8KJRphQjFFU^42~7mj`t#d$TV$YFAOBqKu?~**5BR%&qQQ
z-IB0c8g0r(;S`BN3Kv>-8z5DFc5zDteM!W$b-?b3$UOSybSMW??erK(<m-MGCEyl@
zt{8BiDdozemFImKx!82zkm}$@Yo>A((Uc)hlYMt)#H&XOjU6>Uy=nSojdAe9tzG`-
zjDgT@I+R}P?iR;}SY|7K*iG>jdH78XYd_}IK<Zx_CJp^#ZmHt=0_xx=_PHE4IX7%3
z?8Q@3MA2@WqU77Bsu?%*q;=zWc6w^`2u^D7YLWsO^^xc3!j2yVy(_t#^v4u05L6Zn
zuO;24%4wWiQ6&x?`0o1gCs>oEQ_*hLKinNeKkn~t-d_8ZxRrA7PTPYG0rPgNf$G#D
zrqeIH4Y7rN3Pe)|-|94dsu0oUZir!i(0YnUv6LS0m0&F^Kz?0D0L}}H@pYkxw(Cr;
zsA)`?-U=n(!>yip&jao67}C9~i)j+YZ&Vv>=<0sdRnw>CqkGtG!B*aMQ|Md2MF^HE
z=8YJbWSS*$3wKs)?b1sI82e)B*Ag0eJWC6>TPcBBA#WQWn86oR$kSvDP}ToiC0LB^
zK|Pc+oD_!D=e0P_sOvb1;V~)bS`n>!n@k|~Q&+UX-wnFAYw{R!vu8X$R!D4#NKvl5
zYdtwLccR^!r{irLim|Jvidf<6OH2icGTmmJ*d7j05gCoZBa%gK4~C%geOvpDG#$h`
z)U*1m*@uR4)Cr%VbD4(s*UKanXjiXh%y)+*uG%GRwo60y;HS6f3$!?Ia}8Ej-5)4a
zSL+tjou%Kc(aBbYUJxy7w|eZf33FMuM9ONh7U-Ddyr-CCqnGi^NnB+!IeQLO{Zk9!
z4W&#N6w>8m$P8hP@o*Vd$<DCMazS+{TER_FHSeHdi`;ws?~P=&Frj6v;7*HZ$0*Ip
zIZ_Z3H5fJgT`1}|X&>eu|B<)ih}5@GCH}Io@BmiDj~Dultu|V8zw<qwbW!*t;P0XA
z378r~FyDk&{`_m`f=eaGCGxDtUwBKCk5<S<_n2G+&lPDJHf{0NEWqS#LSx3<B9HIX
zp5>UW2Z>|U^s4<rTey}4PWQ!8_x0-l<mhJRo-o_6TdIX0vK~+8$3ME8m(_mCJjl|%
z=3sjC+fpPs;8v0=0P{-yl=;{E`vxUDUf+=cOQpol4Z4MnP6eicRb1B-nx8eY^NbW-
zA{tb<&{)<w?Q`|lq;0H6qmt9CUw?bA(pmp@@3W@+Z7KWg<;E!KCi2!IZry=HYR-s5
z&%Wq~HeQsZt{Hrmp81Rh9Gq23Ct03s-XnrxsX~mf>!>bfR*xqV^9l>ry|0h4w<PG6
zn*<)PZHW8msuuov-eIM5IMSEf{8T#S|IbE*HQbHgjSg+v()~^d6~7sho73isN-P@u
zZ%u%MTun*%_ZZ@@8OzoKMx<m(9@|9jw8O+cZ)$^<DC66-@t#oT2Y(+;#HpQys?c6b
z;RZRWR`yq}0{~puy_p)13uF0PmLRVpCB<L4$c?vRZ-RJ=281pRdfS^3ugwj;&?Na@
zl&VrTkjNjYnIXY*xt1^z>pxp3sQCU=;EI}5e~EV5gtrV&>^>J)-2AmWCPQp44*8={
zadaLain8D}eW)7NA#C;iwnvprE|E#Gg;CIPB1G{8_n9a`UaCs|u;Oddx^3j6>>r`k
z-5n;IVuFsJdZV>Y(jG;%S!SJRt%T3_JdEQ@2-MnlfDMv1V(F|!@JX<me-~LO0V;&v
zable(j?`Nr6#U+XIZX;eqkX*v-)NMHrx;WDMbex4jVK5e%JU^X^E>CyFr`&ng`FmU
zr<oyOxbm;Zib2f6K5Oy}-7{7*B7cV<kPk}z(}(%@x%^Mzg@-7jgN|qQLvtO<9a}ir
z=i)hx*QzT21o40TH=9opE+%^e_j)nxUuoMN+{ChccxJso$50CV0+cq7b=JvlBxVP+
z7uB>ce_o|#-LBzx)T{D>r%XAhnvG<IdS<`4gAjS1dM>IIGG_2?CC~P`97(SSsZe3d
zUiZv0ep{b;Q64e|kI>|<L2K#_xG_&YYr}|T;ro0O9EnY3=_>qwW@CEpl$4yE?%J@P
zT4cS3!$5#O;Z1Sj9Ahj?Sy0)Hn8SLI=O=+3tyocq!QbrGU*&@(yZ1A`2skEQ;Ye<k
z9jS9g`D`FhG)5km04j@7eP{7eofb}G1Z!uDMS48@Ht&1G`p(Z{R(9Lx(e<Ypy`-!L
z^_a@*$iBS+0Z04`e;H0^FbZR#TvQK|V(D!t+RuN!A<l1`>yJ*fB~m?S*UAT9x8DDc
zuUpoucQ$^*Uvf6c9G(TrrJ1Gm8)s)U^L3r#Qo2fF_D5rav?;J0u>4N6Mw)o*8>yzH
zSPMn314MrvFX8+s0q{!p6U__ug=cPJj}vaAs|Xh*W~MD`c&HX(g#WIRBi~oCmQ6{&
z*0NYo2NIZw_e|6pGG%^L*RPS)a5&`ttWwAsc1XA-T#Gy2Lv9>4?q)tuQ)#nmFR6dO
z;0AE4A{Q>j^3|hfmS7E?p`E3G_X6_aITil@TpVBlf#!=RrO+b0B^mHsJBJs!Sm}u`
zyx;fVHRL4j(bW#ZRq1IP`oTl4x6b!93WC}|-IWm_wQ5Tfebq?dtOXcR_KU-WD#@bV
z8PGyR{|Fse0HLqK!_&Zm$8q5_d>T>`DaI4e=lB&b6r_8_m2`ZrQ+D8?P!##j1AzQq
zG+Q-j2K&YJ<_S&P;4RrRJL7&HAm>6=16+lY(vNS-t79#@;;W+fdn-PX;%jwt96x#<
z+iF)h8xf}Q-aY(Tbp)q;Js6Lo5}&}~we85(yt;@Fz{T>YgrXIn_U<I|8t*aj7tq@0
z!8eL{A~ZXTX}7M9Xao?=F0F4K1QynbhY`2-Ku3eLSSJUE>BQ95(gEAc%h?4dv47!U
zt&*@rr8qbWU(~UkAH{H=Ucid+G3NHL`CI+XNN2l&L9~li(k}Z#kBA8k$$^nGDq1Gi
z_t(*GlmzgQIx_1zX+c-x(6`!7(F@8vIJDw9qJ@FFnJ2lezFM}ap|+z+0=e5_V|*wN
z3{FaDQBr7u4dCFhdw!;+)zH(8ww14La+71J0;p=eb@N;(xCnU7Hud{b$c^w$l+8i5
z%zz4iPA=4LG`5ZNbR<bd24Dj%3(0(T;4i=5J6k*r`)n*Hb37$L=E+wB=GrF}%axy>
zZ@(np>;`SYq$_4YVMHI+dqA9Jg@=TefLyXSGXS4o0@?@sY3^@tKAP=F(&T1h)}3<%
zsQhx134~#OrdZSjH1DmCKYs&jFt3f}M69za&U0|*g%C14_Yg$BKLAZbwfe+#Z3*qW
za8aB90e;6Nt<d*3Pn6CAH&t>f7-`&F^O(p|lEiV<6eA}DPZ#*V8t=>Srs1mothP-h
zD(LrZETUx1REIdP(fq$HTxL8Sl5zO3#I$A&G~qXj&)%q47oQy>pgkY_T2sO-@LTf#
zc4nfYBb1j50ugd=7wS5+v-mZMdJV{2O)3Wwm*Uc{!{Z4Fh5*!^1?T3&DeW`hEv+xW
zHudvyxs_p_7F}H1c*_1&zl%uUCF;^mdR&zxSv;rdk7A{*cZ~U?r*20y=KME6lN#ro
z2D=b2-WM*sRNdRK%WR=j(7#bnUge~-YlU^0@lHr3(3qGFNhHA)r|U_2X24j<Uan$)
ztbGf>9^g1HD(>4i2HlEGibN}ueN(mqe!$3UKle53*b%GB+;C~0n~hW-`KwIVIz#8H
z-?Kn|w<%eiXq{&~Vs~L--0HihGn5COrbAkw7v2GDmSy0iOOp5osO9Nbt?ub8b?}VT
z2{!=Km<CKgJf2pVof-eGPsiz7Ghr7H%s2y@4x~0L5$gsbHlP{3+SSgXRBVR4iYEK5
z6pR^OG4BDeUU9~6=>Bn-t?TTbeW9V@=_^aepf^)31X|ygF3FtM4dhP(oNZG78AFus
zmCvANUZIcuo{@#Hcf%U?uljJ{czYixp9tOLaJVI}{ELSzZu@36zJtg;nTyG7k~<;t
zlaHCnt6b`G8!xshp+Dow*=Z~XUzNX<`xvd2ad_OIRPlLFneNg^6aJW19Ksm>_*yP+
zTS-&(M^x%*E_716A7Di?b-KIJ?B3YzfS*TB%^K*+2KI$H%d@gvK9VbR$QM${HpA$m
zW@zKhhO?a$67Xi+;D&=ChZIfYnPHyvK-p0><MokYIcKdGq*s<B4thbKFIvAW?vecn
zxenQz+hQEzN)uc(s5YTHIZzG})m-y-@_OardL(k{XTtFoccypPsAfW6KhOC(T{id8
zTz8f}XXwqAvGUeLI_ID=?-cCXp+1G`+UPVfOE&SDtI37`^O=WXIp&>-bgS#TyR$-2
zEk<t!A6)8zK+vDMxTv_x!la8@6=?$-K00Xw^1BL(V|lq}_6??%XJqQHO8L1~urHK}
z&ksf&E>LjRlkQW)z5=~oFnTIZA`yNzGdS%<g^(L5dsxL){(UL^pB>^E;0cj{EjZEB
z8Y={{ZPT_C6l!rT+&8&oe6G&d{<pIJ?~Ng`2tsQ8WMbLdV*Go8oL2&+gbRdMkMdnl
zD!d_`H0D~H@2jJ+YCs6O-s9*1j*A3*Xsz_Vi1Jq!k#dljE-@KC#e)Hz4_~B60A1yS
z?bC0sE_P(mR=F;;h^)QS+r&O|A(|co4{15*ZuY~Qx%;~Gw5^oFA`G5Be}O{^?WnnK
z=ru|gO{Vh?(`sdV#I8uG3k#nB$F~%#NCwgisO9k!Ubj*_E<HNlW~)g^`Yyse6wziN
zhew!!{_z%tqb3-B$Sin3&=Fo^APaZ0NrBs=fTH-xiZV*i6PUH%Wp`_(b7c_UTCMh0
z)UJ}U1f@)o72_IjjEJqbL7Y#Q(}EXIV!NegH8%be=^S2)R2u6r*>$pz-U?UPLO+|P
z#jOadm8I_Qfsi0`90~#HB)+OGj0-EneM3&$>@A5d3>tc?)LNPxg~E?9TNNrLDo!kS
z_IXe~ZDSsAF5hh<o94u2DV^oYGD*p4IgMc~JQ|xFQN&9*UV0KnIbC&8GHu3z=RP~4
z5`KG0u*VT3Xj+QDO}IeIF@54Pw=%CuT>$74GovzgZ=WV+m-aJdW=x#dzT-JwXs4AF
z^cU=OTgBabzPAYnwIF}KMhcJS9(X^-J*p-^aaSDU<4ZaqaAOuur1nI}{rf#6eY}_k
z;-w&X*Q;oJK3ynvC8FNjjB@=EPq|C`x{K*DC0Uf@-OanX@m%f5=Ix7Uw>oE?;5MpO
zKOPj3BRDh3;2!QMN55&sra!08-BucNs#&nDa0(OuzZiS#s3_OAeV7hGDJ2Dl92Jo6
zZV(tmC6q1!C8WENZcw@rq|-rCB!-mkAta<5q`~i+y`TNOy?)=Hv)9_ooBO`vJdg8;
zhq0Fs;{<3TM=z!C?zGUww-)gRREh=L?c8EmB2ICHj8P~tG5Whu9u<vy|1L`YUb99f
z4ErFPyX(<X^$XK}bL=;Q317_jwo?C7^%B(jlfkc!sL5~Tea{qMT+AeWH@drVu7f`|
zdAH%4`<E^zi=9C|TuOeb3bHSjh+;<MbG4T{+tBAtCV)N_vL}VX9^Y^)X42=Cw#Ytp
z(4+(wr$r0PyGeWV>iZQ;s)@3@x%AQ6^4;V=7kV$r=9Q9Lx0zRzOSr=Kc(^xp@8B}j
zD6z9i=!!vU0pCrBiI%u#6@-&RV!n(p9n#bG0`x0^CRqw$3L|wQpW{gIsZuqTzM2B$
z>*c$gfo>5+j37lI-25)BOwwUMw3L@}Kvy>y4l-p3Tx*TnY+`O=hhm%`_Me6+a%UaV
zE5;O2G{jZEgFNz6WMru&2>D9V)ihS7eO~Fw;`wU<6fqf}MQJtr6v2u}Ei}@Wuh>My
zpD}M}LWkaVYpf`W0vD&30im<IDF%De`+>DFLlaMv4r!Tx*;Vc4cr%vAnnN7MSu~w<
zcSU$Vm4M~+Mr6cwlrxiBnMOp#Q&IE@%>V{LbYK3P`wAsq`66O=h8=nA28G{+eVf_!
zo%NV(u1|P}FAXmF&xJ<!g7>#{g>b{j<AFGpfKbWvk?zg~|M~ufa6$Q^Xl)NM+VlJC
zB}`Kx&n52tRWR(Aj3Jv0>#|+l1`CMP=tFK;e}(%9W(GwV)J7~+6q9xCJ3Yn<#z!YX
z3wc;~SNlI7SLpxi57S6A<UrLuZiCJ#Ml~KEJTB#yVFeoOVKg3A-4iuFl$fXO=)M8*
zL3+Z17Mz})bb70UOr0>7&W!(9%w`Lzp6p>^P2y*!a-G=y%I(ZdWycnN{y3t@DVOlW
zVuMy4j8J>?o-V_XB!;zxswC#2t{`3H<)0BYyW*_#TjV07sKxw38l1(C);IM3*C+g^
z1pjNbk8)$Uk(x46st2OXAfU{P*0I&6AZ4It&UHZ?P79Ex1yLmpnwqbwseBK`+S7qM
zm9%Qvg)!Wr14&{4-+vTn!2zGp13pZ}IAy0@c<5sGz){Pms*FEtCEdXRXh~1+VQ$D7
zR^A>{ao*~XVu8hxVC4C=qO#3n#12C<z`)Z5_*qSZ$YwKZ4)K{>pH)ky(tXe@BrD;C
zq6x)0=vw{v9s1X37lQTgD+Zlm2*;ns_A3MYqnd7%_@ApI7chjSo}U0w4Vb2V(|}Bn
zuZGv4Nr7QqLO{atEn^!s5t%B!<3Og{PW!GET>uQQ$A9<a;H?2d1FzU!a09Wx152o_
z;A2go$Vd%%39MJ%%qi47e3vm6jb^?A&;&c{v|ikgFJDvX0$Jwe#Yd-H&MBPDPYuo{
z*`$Ttc>dJHLclR$2c?Ax!a|STDcoz^lR!}@>`+V$4jb!b%s-+}C2K-M2mb1eeltKY
zjJ$OoG5^0mnru|-55Wa%;Kmhz;)g^z<=gNdM56Ar86fc={aP3}EbAwR3a&6m*?3XY
z`zQC%yB`3DrHK|mNI<>?oDjry*t>mjOjE6NV*^7jMrLiRe>(i9q8M%$jR1V6SsdUS
zRjsN-TRH~5Emo+}^|C<*?xw{=SV7xgD~*o{JTxzz2-$x-L;vR&;xZ-y7F!hCAm9ft
z{v{P2SO8x+0O7^F&dKI3u^TWrZb$0B90HwGNm=d469JI@oH#eZ|DRq)BBLKv|AJEa
zHJC%D0Rxf1GeHnhp5OQ7PjS{u(fnIaQH95l6j=H=cSZiA)BI1X<i9ML|N94*7;52(
zxqnK&^}p}&|FVPruOF5W@M<t`{XIba>mjZo`(2VZ!>Am%sT)OIwBLc*^HdyIBOf){
zN-Ntey=4~fP+=5T1hco_{r|CZQU9)gG<eb}5=Qa=_+x5ie*zssEf@`Vxy-v;fVz5T
zMto+xImy7|VcR7j7Tp`H06DgBVE5|ud-r&sIR8r)7>P&a_k&Lj2jPCerHEtVt3WfI
zp3|`W32<5wWU>38sZ|9i5se$K>QIh%U#?HT{rLhc#cGk-Z`^m)eJ>L90axc>=dRHQ
zCnf0SXjsEmN#eH*pRI{7*uv<)z`Os~C-Dyp^uzNbgI0Wa_V-t(G{HiZXu#CB+5F{%
z^CQ%U4@Bb!AKmF1CP{=cy>S5?wiyr_Y!gE3q9ZBI*L(!JLnLZ)H#JuDd<Xb3Re|GZ
z{tO`(2M8z&C?dpPlT3F(!2g)hP^2{n{GLrGqVL>Y_f3Jpg(*`8N?i?js|J5lxXO|O
zwVO9JPXGf23C@fx1A~$)_J&4a+k!-y0~%|pg*fF(WPEM}et0j7H#?rIi<=A1IX&+Z
zK+6C|jz4Swy<|t=jqLd>=y_?Km%z~~ns`vbG%^zQ`tB#PA8zZa|E*sBA4`sv9D@-J
zR(hrN=5GYWNQnd``LpRwe0T!bUr4}=vOuw*Y(PP7okAU|=r+^{av`m2@kvpnr5(^C
z;~&y@d?^m-f&)5mC!hm-n*zrY6*B@n?Uy%|vPU;{EkMEuEX=pTiD0wBz48r^9V{(J
zhV0}(7STM!r{MJ9Q9Xd9kSOAdQ&FK|6YI0rH^4IU^w?lL<e_Km7f21G<|so(@=SMp
zvCf3PAwT29Vs<eM<p$>h><n>p_`Sn-A{ji^w$1=h!zT9mP_+5hFRJ)u;KaF5=Q^~3
zPP5U$R{tQw4V!=#|9`xM|GJA?JAmM-H7sTC$(O&YGD0d0;Px1aVkMmblHE+S*rAdr
z=*T0>>~-;1@__B;9N;)fP6ZG=p4zN>RW1oyap@`3h*9MP)^XC;x6Duk4-^wcfCS%<
zSRACu2ldEZc+YJaSO99C?gCaTO5Kg3x#5|L*_Wh7%b_L0k~KRW__j(6oJuF1P_tzM
z89}TKW)x(MqEkq7sZH)*)q$`l>P)mkZFg|O*#LS_HHvHn2Tt{>6+PR@kHOzetPC~y
zE~EH=og@Ej8!m3ZIellc+W3qE>#w&V8{P_viPtC(s%!8i>cCA}X#NN+g8KjvJ4_}<
z(^bd~rU6T5c85bF%Ru=V|L57ySQ`AajA$+K9GX>LEcplG#Z0u$LEJ45j!%CB(pmrf
zae<(k8-@{;$4FS1+AWSzeEolp_E^QA;Owqs@3Ve?PpjK-qKcT(18YBepluqPTnBO!
zEFmqMi8jtJo9Q3k@?T2jA)x#DIrKV%@$ui!4Jd!M_<n8*GXTo7nW?n``D91@*=;m4
zz*NW_&=!4{W<d6(C18@gKj;$|!<`3NsrF4Bq$*Q@8%gK+<E_D{n<OpDh09n^ZSCqY
zASIoE3_)SvqGj&+uV<_WW{qR0R8=@IY(fIg-+kl)oCz&Hc;RYQ@C1-QkigkZ_i{Ar
z#thgs#{ZD~bG1nO26)uLt8KW>2T*QbI5=|gdj~sBQq|1;%%}`g4nRT#p&mY-44&6u
z_gGt}5HKX}NL7HqWY5DMD3fn1hL|AmZk-Zn(9fS1NcHjtg%8sl=eH{dBmCjl0Dbua
z4t<M`E$g=)Klmqg4*$B}Q2fHle8&wIxHE#uob-6h#?Dh$gyVYWLw}n<thHxGQp)Sp
zhj0S<E@Tg=&WJ7nsg#<`1QqST^y8-|pe()jIe3iZN|11Brq}>;o60T~Xee&x7h9k@
ze~{-3D2I6xM$gQn_$Tl<^H<yraMYloy5D|}L$<IsHnl2>3j!Suz)7B={_0U(>^?X`
z45v{k*Uq=?-NFA$kM$eqe#A0={OOx46w_s{TmUCc&f5*z%~zYVP1zx;+Zr)mD@^p<
z@N`9@h=8>hy;g-hS@U_<5RNf>x9SYQl^ZR~xuj*wXubT&>2nZ)BY`5aV%nAGkkv}Z
z5)9!t@s9-e128OVO&G%2Gs?U|Wurwp!+gPZeqs*1PGNP+uC-19H$$|pvn0r_3Q!g>
zUEr$44Tb${efir1zc$W@n{`j7?K>y~bEimrn};@AAS`0$;3c>egYo^rDb3Z}0U{!-
z0gp!i*Jtq7=G45$K{#4op4z8F#lk!^M0NT1*Y*IQ>N>;3LzxRsgLp?C+MJ^Nzh=H*
zv#m$Af+NHxNzXyKJ^JJPltq@e!qm!$`1Ktq#&tAy2=AA0XGzpmO!US>@n7e8s}RFb
zH~=lq!I1b)nrXB!<7?HHEU*j!?=yr-SBCW*m{Z!(Ukrt8qiBo)D_zWj9qg<}?h<i@
z5MS;T6zOvscvMLQ5ZiY%giHbN*VKU94tTm>c0qw1^)}-UaQ)qz1{o7}#LBCHNIF$b
zia|q0iW0oVib70lXE2g?^uK{2wBjs-bw4N!<I2Hv3oH?Q4?33B>Lp<2j;u2Z>Icj<
zI0&JdLS<5}Z7hI2Zixyex(2Sa1SYgPDD}?9$>-}6nG+G9)!YVz$Qsl|_!o|TxB^}*
zAqJh0Ot#-u4XV!}&;OzdENmZ&ZsaEf41M=hj&Y-JW5()fKAGv82P|O0<^fOC33$6%
z1~TL)yuj9r0wlSzp<;k>X(I|8y&_`xGh9_(D|1ROq2n$^W-r{&Th4&Xy|2Y6!W&sb
z(U9_cz(N$9SFCG&?<Dg?UK`To!q!}nN4e{^j<rJDOBswA3SYnDjPDj7LhK2d&ZR)z
zjx-u{d>Igv>j~|^QnLIK-R9!>JI8dgB*?+%gHo)ok&7t5N9m3{fNo}U&K7PrffrX-
z2A?So^HZ$Px6Rehkq`zogOZGFG%$8X8Sjy=ty7rCd5Ca&?$#HIgM34ciaF3?OS9{O
z!s#fcgB#?0y`*w)KI1Zz+4N}p<-T-X72{ocHZ_zU^B#tV69k))-NZaRMsY{S?hzkC
z3}grupzCj?+I-eT7pwq(uH;uwPHhD)z1KUjz};Ajbp=w^LrcRm9A{w!htjk=)$I(3
z*XAs0Qw`Vox1wrj!A@~{Tv~#EU7wUhSeEn~oN5^&<8<yTBuMadJHWMXhw{Uis|o>z
zXiGM3kklP?avWp34f%b6o`CRS^5Z{XTEUOFH^p3hi+~p8k28%T5LOCT8HK$Mh~Fiq
z>}v>?Pg@aRRzkF}haw85D)fMd3krUtCg@{LHM%q|&7eRyf_BV!ihkrbpruakA0|_f
z8<}!;BeRcAo!PNAWzdl9Cm5^J%*!#Ya^<{T^w+DDT&!m+P)RBQzFjduHm=I1Pg~>S
zbg_kf8j#9fsLCEn6GBzx_J^+?{@o7xZ=m_}U_i*WOHcj)OI-?ultk4~TX#DVy)Fpi
zjW6L1SWsos?@Y20E8_FE5J{<BBMf>@cS>h2FY+)`N}m*KB@APX`0l?enjKE#<!1b)
zA>ambcitKDa?v>Q!$b#3j0`aK&?=pm090nhpsrhXOTm-UO4yWZkX)(PL3lVRXiSiW
z9s;xX0JrSIVTj^O2NdtcusbrZ3__WbQAX1ITAZ?c5Lg}ufL!)7L%SqLFocQXu`=mV
zHm5jYM5g>$k1GS%yO}jK_S=rZ2|rT~%5x>CkkOO^t{Xj}8@;;U3a2qtItv1cKaJ&=
z1bLOgH;L{6atjboQrzy8Soz^Pml2K>miOO3O1cvzDR=c8i79i4XZ<mMN{CaK?(oFz
zDo~EkY9t@>{c+QNn>1R7Da%DQa7D){cf;CN@K96a?ys7_Z}A2BU7a2HRxDYN?^0vf
zDq42~?gc%@^IKOxgbP@TNWZ!$tH5~)Rno5W&}5+{Mms2w61N}IP<+pwb%ItckjW3}
zkf98=Bvd9IdC&mt8@(91ksV|?dd~#2+?K*Ld`PGF{w#cf{8<bmk}d=VTK7<iQ<fpC
z!w@GcEratjXp%dcSsQu<QSHMFtT{h3m<Wm3cb<*7<NDFdKo?(jrvSbLZ6Jivubo6K
z$erdrZ@b#Y+PIzsdMv5qV3*XZV7}YKVkG?sS1NxnWtp!jl-|1!#x|*l%Qi;exmj0~
zyhUn6cQibN-_~VnJTJ;{jZZGD^iN$$cnjlLUuUJmElPr9dAyq2k-G&r4<@6qgvDQQ
zH}y4{Y%+`~ElSK>S5+|{Eo&%ydAkHRpq0DCj63XslG7uS1oG#ejnuW5-+LK$i5&XA
zHu(Y3Pjy*)sug>Y=KAfFBo%*%NA8HMXOk?48*m|_8^FY@3nM!>sxs+a@b8z19rj}r
zT(s?;$jkFQ!WGvxqpG{6%s~R!o@fHH*4m7{ci!$2sW%NPT95ZzDIqBQ!wf8<IWL&L
z!Sws}G+wO-a_oR+jY!w0&^<0LXQLOwG_<GlE_O8k>v*Ie%__^k>&w<y$WMb)p1oUU
zwqw0503SePk$qfBbzS#-(utz1yA48Td7;+k>6ech6ewX6y^p~-sbASV>TTJb+17;m
zxG&>71zRoD50A#brjSkWWL?pdw5VTL*z)zx);wQjj_eml<8RJ_rL!lKC<=-dszez=
z$;~AEvxNOi?91GK27T?XWA2|NGd|E-p2fFI@qFLw<?zYgrk1E%`*2EGRNuQdD*d2|
zMu&+*V(P62QBy{3(z#2c+T>^$*?r!_LGk<MyUiQ`36H1W-C=VybzRM5;en<L?0yOy
z*V5%cf>_E?t1A(1ZpS<$1p{XLiing7x`A}8_{U$acjPO2F)lW>zfn&KS(R<OZ1A5N
zkne2+Nl1!a$m}tpx&PM$Ntl9F!{l!M8xh1SsF&`}15wyYPkJIR<KSFu?0JX!dLUUe
z4^$&c=_Y#<i)4pi5=R4yV4@vk0DM<(UZ``_5W(rvN8x%A$7;(}wklreCipK<r1`G6
zQOEBq?9E@G7GU))+L*%kAR&VK^6T%dbu`PHwcHYDW3p-a{cbg=#wUO})D`int_b)0
z59oWIQ+RxRR}_iR)2y#ao&90H!xj%EVlVi|_{%!*0}i6JPPfMJTNVk(s70hDJHpD!
zB+9n_`Pv9tbi>vmhkm`tboGxGu$c!26ZORFx6q;O<|s2`o&^UHM=(_HcQ27dgqLj#
z>w8y*gito}+MZy45^owmA4aK4F2HrQeN~xh3E`r1RufzfSHRM}Z)9rb^}I``rW1;D
zQSGVO=hYR_+A40OFpVe=(dXg>4#d8BvzIbI`^8%=S~$G5Y86)(hO3A_6QDaqyD0J8
z;He&qI9@qsdlO!yP*~CD?%rX<XpY3j<-Q3YU5B(zi2aLo1^OQ&C1NEU4m|yqA8)Rt
zf3{1FbD|UTgm}8qNUyGA#~Uc&w(zU0BK^&FjZ1$)A+L78;5e?_mJmT4?=}ANiX~v?
z=5jB~_?<1wZYTTEz2X#4v3^MTU2IHF1)FY3bLRG?Bj=1$l@kZMkQBQ(9qQYsa0)rZ
zl3L|qnO+TOu>_7ZuR4)Yj+uc!@%=Gr*~2~k`pI9K#o@qKaOJY<ntpMpk*QK@43AfH
zg&RA}M;Lr)UKo^0x(fEXWnZnz<t2k>ukL4Fr&6J*)}^zFg?|Qol;Z18YoD1IO8D<u
z@g91o*}!`YqnD+}@fj!I`Vi*AA{9j-Xfi+FQ_er}!ZuN;&x<^JGD;G2n;Jb9j?j*=
zpp2dK92inRj{IT^#mKfh5H)%BAnSaWn@g#4w~E_WIQsdw*RgnhEjya4NG;r+Y4L3V
zj%dm_H2r^flLL5`BNpkrMi@RVX%ga<szePBF@Ekpa06c#R#fC^tWmgkd&!&pw{e{#
zQL@^wqk2TjUPD}c*SCT#_D7jO3lQwVp#MC;iDw^g8-N;&Xm^wz7Y*ExjkF|*@>!LC
zh4rZqGqUw~Xo1eiEkxkW491ja&A-2Bdw>+EPHn74Pe6ITAF$J~RlYcqRz}u%53Xb^
zss$f6eJcTVak|_LU2B+giKUpkd~TiP01p2zbGdPs)@)mVnz9oauj`ogIff%rNO9^U
zjzf34xKsNRd~QCJ!)BT>*oJMIhlkS|yKkbv+MoNQuJ3X69NWR~54)=dB__J2aAS5d
zxV9M$5fo2gt_2WHFicOkEnq(_8??pjWzvP+dEqkWpdOiaLN-4@Oh{PAZRIOFR=h&h
zIS9DaiXhVZBHT#y>}Me0P{Vjn`7A-OvG#LHb7=Fr=Je(6_=v>CaLmcd<_#wOncsQw
z+SG0#yJ~q==L2)%>ve_V0#kP9fMcN?&cn!^SJS(nGQJkVADCMAZDt~=W9}5gh`$>g
znG!pjh>b{l_n**N|K4KWRWP6+nsP|y$eP_stJnTofV3u25tqB3eIoLK)6n=nlA!9E
zt>KmCR7)h!pzAcO<%zBi|6|Ps`nPi@Jq$<y>*Yojl-WYZ{*3kPT750LQNK=IB5ysf
zlmJb)Vur4D`v|Y|7S>|*Co$xNu=2c@44e0pZ~s0>?xQ1gtk-^Jj+=V@N)Y>}xwZyC
zV<AWVO8jb2=?FKkcO9tvavlyGJ5Mc*P+0Tnb7pe2yuQCyLs#`F&vPn<o77YsdfK&2
zz)>4pKe-=llS_%C`x%xY!vFZPt&>vsYoVLt;QsJLbtkhG=%>|6M<>jt#HUsh4mfRU
zJ#C^n9!<N%fBo{dqKbekQ*DW0JW{eloDvx&I8S-jaO&%*KCw#hcr@?Jp8)7V4w0O1
zJ4JrSSqDW<XB~x(k^hL@M40&ouE*CYwXDpav$D_E3pg|1-fAwtlA>?r;+*R73W>P-
z1n+90<m(OfEW6X3YS(fu109YrPtu>5ElE)W)7D?)t%=<5OmndVAjo_Sm+55fboo9N
z*NUdAq~R7~uSx1egjDvq`azjiofUKt&WR;(YLxj{^QKItqIit|TrM^X)#!|oSnsIR
zXGijqo0kS&k5Hr|*jl81eAF=sLu{SfV+-&gC?d(cNG!&T*E@llNbgNZ{+*w~a1R8m
zS74;Yt%$@$BpFaj@&>kE_!g{%93+WkiVRWMbe$YNyxqd}Xwzr&E!5~!-$@RU_XD`;
zZ|vNtx96C>3dN@0Pzu?}qP5mK&GYVWhF#ph**6#;kDafRo8Ltq;LunT|01V7$zXpX
zIcX<ckw?oB^)!j%XaU+>JKaL2K4tb$bnKU%TRRQmrRS!F5!v^Fkz$VG5%NOI(>$_7
z^<%v62I`V+u9nKu0@t~^ElSNBuV|s?Ld0~}^cUR9t--=r@Kw3bmv!?VTTY@gp(ssK
zl;eKk<3(x@<<xa1qrm+P{#?6+x*7q4(t~WnIl4WOn#Xf=LGa`j0c1H-q6*;0KHhwB
z9N-dnLe#7Swp_dIblI(AUe1vU*xlt9U*^m5${!+%CXaloy0;2B$D*rMeM^$KUr3Z|
zA_C6eIf+%FroJY>>YlOv3q*^Sn$LcZPMaeeMMB^}j@YziAwlDIYp|%ZHDvQHrx-y;
zh(?O-$`|vMI>kjxD0oFG7iN0AX}JH5^6DEXa)bQa>RfLJVh@@S{}Si=?k7F7URp;=
zSPP8reJG_V^9z-6aefo7a1$E<fSkm(gln^9o^yH&)kOmgu#``vteXr~()Tx~$@|2H
z>;&PHNAk7McH=G4DBOx9YtgK%hjNGlt8qdT%HU|W@~>@PG?`p80T0b+3{&QIpGSFu
z^Je+eth6MiezyT(<sKMd=_&mP$-sN^L;Vyc#MUDDQRI7_wvO?8E`hJE{U$FI(q&h3
z2ziX-dx?l{gM+5d;P;!q4w~=z1c)9N^FTKSqwP%j!2QcV6<2@$%&C&QJTlCaKI!eU
zPmbw&RY!(isehTUfBd0qb>7m{AtNk=+59@yiSru%rhUYFLrzBKUuOnaGec``DuCT3
zFIHo(b%M~PLwkOqm|sedj)z~7prwd8RlvD%Fli7ykzFX6C!9dU?ISmd93^6&{V?Eo
z5$S=rBu)>-G)2b?jU;*bLMIWhGzZq9T#uo2k(eNz2o`A8c^NsE!W=l((bfbQSB;c%
z={0_lVLq%q?wr${Zdu1t!DiemDDvRQ>B}iBdpzU#57>!dE~h@lOi9K&@IF!I6@z5u
zJM?G9b8RL#3-t_m82MphpZ5BkM*=d;3Nu`8G3vzeX;FN1&MzueCz8IG(G$Z!6_ZfF
z_|f@S7emPX5i?zgoh!hW)*W3EVBH2&A07}*iBb;Alm{9GKh9+m8x2RpO9_)#Uyt#W
zkGKjlinx6%faHXEzH{~2-Gy-E0LYhlkdSB<suX#-F$m2mRRy!hHW@`omN89WvQE<2
zCvDBm)I{<&KXwm*#p5W)c;FlonBSM7C*Yz=3*<kq0WU?NUVX379-F|ch>f!DSqmO!
z8usGT7!ugL%lFIO+w*kZ$HNao*m)2w*yoiGVspDiHzPJQb%t<_{Y55ST|&(+q8X$k
z<?9%VS1FEJo;`B*Hh}QNQ2@9rtUjr-oA)r#E89!z_4kizbNc-;UP78N!%3)^C>H>1
z%3sEHrzN_N@PY<2lx)&l^oS&@<k`TMrJ|zT=MaLaUtX>mmdFRZ>zpSqkCU9FeP%&@
z9#tiBV!z3FnA-UZXl}1UePqVrU&CQi>*8$`@KIbpLRoKh$<L~v8|%RQs#^C$IrpH{
zpH|nNFrVAb=+`mKwb<;Da$OsC8g{Z5%JCzu-kbS$UduEZ5|Ggx-`YzfxwBdp_OZfm
zGT+Sj^22Ro>OUUgp`l5kp}ZDdGhQay#<=Q6O5|0r<7HQ<(#T7OpWnNnKfWv~86wyn
zr7u*k0*!>TUU%TYu?XigdFY|YD{!*Bnp{~P*|WdLzFB%%5{-SS_HCZV2TKm^{*Bm+
zTYvYFMi>mSgs<e)g(oi`c|^`NQ|2Hd(3HepGT<(jXABJ^gsQ-FAO?cA<o%=WVU#ox
z0@?(Lbs_X5taeYj=U;l?RuOIG(hc2=g^i4@C+)6+X49;+L13-GN1l&Bn=f6uho!{o
z(!dg9Vm&}3)QW>C+%W)Ry!Z&;y8UT?7gQ&y)ol90;u#+1BLo5A&74w&*+-M)<~vUA
zzQjESrbT2%ek8<$)QLH^*wxxT=9{2>+1Svm_`7{keOz;RH;*;=Qmd`wVM+JFO_!XV
zK`$%8(JAn5cA*oHKY7G>(tED6<-&-@V=N^)0^8619(3y>A?v5rctuYL3x*Nh?SCOp
zt#Z;op_1py#8;Mr4DY#D-xIg%B;j9yfgxYhX*!rwa1ZP2@!`Ub=O|sFF)t%y7!J4(
z((PjUxZY#6xg~R>)BT}LlT9=lrZW4)hfidTJ17$?cF648){#dHzlA_fp;MN~&U|mY
z%PbL!)8q=q%G7fC)i+oXN04n|3i4#`mZ0=DGvPhi!$ogIUgS<+(<yiToqF7J5Wrw7
zx?L$tOz|B9iPeZc<jaz=aJ}o=-gqx_W$P}hUCMT$VLR4iI4+`V<)Im6(@^CZu>*62
z=xD&;-4KyVRb$O-j;&E&pvA5Q9N#9Z9OwpPR@Lp5PvuvwZ()aA%4e}3l9d#RMDTy%
zHD3W&2zqyPWfFa(Tv|p#0<AE~4gw`r)sDP1cVnI#;DZ->z>>*6?Av=!-~nqz6}R-Z
zoGXLqulGEWH{Kg6hPN;{MZ$d6T%Y(Ed4j1yP!u-N$3&9j0-b@w#S;&`Bm>Q&x1qcV
zzanWFrleBy?+0Q#-!r4{#c{@?qpCvZQ3x?gfu<MOW4q&@`+1NB@N_^CpWR{=6{#94
z-|=(KGJQK}z^l6Jppt?sg(x1QNJMUEKENE41!Gmkb?8RN+xv2k`vv6zLyF>&*_Xs0
zD!JFNi2{fl9}325LNsMOSk<$9E;b%e#Og7iJ?na2ILQ&)Q7QQ>xafJ6GxuxQ@5e5j
z-S@6?+16ia6ek{u%nX}&c?cIq$*nO+1@MdVKOI<Cms-kK3(zCbumO$=F<yo90sZ*K
z@jOZS(WukF2-oB>^OsL!*PXP`XL?1Fm4@RS#B&b`j-gXlg+hXu3S*~<4DT!!(Wh8*
zV$jRS{;c53se}&J7WJe-Hkj^j`W%LK-ix&#Z40@pMf*+D$+KyzA2=p7Ax$nu)^E8-
z^J33X&ahvo#p)C#b$gQcNs;D1&MP;|d@oX;@b|udrdK@C=sd>%@Uf;(iIIl(oMD$v
z<t-c!*KU%19QnV*et-2-g!mXOCK5;GimQSBz1%0#QbE6HXL^n67okTz*X3>YF+M5k
zlxDUN>&qrG0!E(}kMa#+LGsQ^3A}-Hr6DIyUk8*Wh;41J;UjUE_UEe!*eNuDJc6XJ
zZA*W6eehUCQoojv*0_L`L>MW%N%+0}t<<$;D-`2&X_T{EBlZ(f=Qe+vk0Z;qU2Zk7
zi{x|;FSUM=?I`Ty6MI%8!GPDcW<9e;Y3F|Y%V{me+(#UIJc8-3TQ<3uHV|J(4lJE>
zD4HJ;wBK$#P;)BKJ^0|1LQ_8@u`5^CFMc2H&FWhNfRJa<(7FUeUO^!gW(3V!Tb3Tf
zCmsH$RrR6?@7HFqka#%rjN<d<Eb+Zx(L9EC1)rMRI&m7<rZ?vBZ5F%y>VwiSdI<^%
z%$GPwrF+M7Ix=Z~Fv3}ytD3M8v{W<@@)vVt(pMpz)yX;+c#~o$D72I^R$(u55$d@S
zKqI4qtaSQjn*o^>nK`;EdHoMF0|-%|mu9=utOyo~X)zPMCyiSXK|Ih{a|x2<dtLoy
zEt7j^cr-zr{e9`o!*}_X&9joDZ_us-c$Aklp>$@E*xsZy8{8wkLh&uHQ_k9A*S|Fi
z=JqRVGSQ(EoF>K#_b{VSbHFs^T}rvahWeVp?%&$>vVR4jz&ggJt%ThaPe|XlXMI%(
zob9_mvC8}F%73IMPTwsj8{s)sNETjXcO>GXGX0(VqO+RP;M>~q#`jqJHy7=h0U3qb
zMZA(88w92m`})i#dV85cskuLrJH)%qVXwk1hJ2xD0^$j<^$-);N4L(mR@xrWz#KT{
zqmRrY9G(HiR3DuK(Ajhp^@pLQkSX`l6?xh49T!hFAbQeLJcasaX`XA<zSpqu`;_2>
zsSdUMSRfq5R=LNm#SWO9TO0j7btwtsPZrjeEF;-o`B?keak(7HXGT$q*o!zWS-Ai~
zvmzbyC6jYuf&nj-LooAKt7~X?I#;G}(ZkItlV;WPcQUWWrpM4>t2Q3r%E^vn_W^NY
z|JfzEfRXhe>?TU2Y}~aZ(Xp+g;Ke6en|l?u6{5xhj-(n0F{wuswS`mH>FMd`0Rak6
zVZ||DFO~uj{CA4soz9y<cG9#imj!syi^|1sXMX&st!^j#VW#*Uw#SUv26;G?B%b>Y
zL^W}CwKaKwG>F>NB2J!9XG|`>kCDL)qFOAg<~tPRz2&peXwu0hM2qDsY`D?2#>0w#
z+m1IBFg*|#dwMV8k@FwRC~Qxq7CKr7ETevve_&w1W0uZNlYU*NOvQ--3C32ka9Zas
z{fHk(xuC#^BhlGIxLdD#1xfzU-1h}s<(#^p1!-hNJ8Y`RzcZ3rad-z)0EgYss;jKL
zQ=G<Z_t)d256CzHL-E<)jpw*s@!Qx3@6UMbS4zSUhbis*YbaGCFo%w_EA$4v8q@o=
zXDxPfTeyWg<Zhqu(A_n0DDJO}YkP2WV$#al(&sLTIVa8R8TAR~muBaG*BrOn7W(Wi
z!X2l>oF^2xcmP`<>9Mc_C0xXVO#{3UO7>AXt6tAg`0gSHD~C6>qn1LGa|L;d9k>1y
zq1Pj$+G~p9!_jD=In;T{6<WZy>oQ45=CPwXUy3JPcR5IlXlbVoq5$=N!eXjOmONT#
z&vPzgnYacl86PUE1bmmn1pGZ7<A7Q5>sl7Hi6XL&obr<<cJT(*zxyhgY#J=c^P!f8
zy7Z4RykUE~mmi{2beqTa&O#3u$u!I{(1}QMJu%oD=X5GjBpx_37m7ES;(1JeAYrQB
zjZdsGi|h-%+`t$drttsi_ai?;3dgu<&mi_yLDAgh^}J(@iL_C`@yHWO$42e!NNkHK
z*6{eySJy@_2}^o~v7}B1$OUM&1+Jv;Y$l=UX!|%JDG&ug2HdY7GVQTgL)u~=Ulaqa
z?8ZwUDqUg$Q%h_8$PexmNL{w?<nlhU@7Tx?W9$6hOT+QQLnQ{A)JnHk=2m(Vcu$jc
zQdRLx6Ycl6<<OjnVwv#Z51+P3drv$;r1g(HGBEX~C9U4uCdtCS1k8`e<k!`)$mX;6
z-Z5#!FM7yMpN!v?U#Mc)_R9$M-o0kmHH(wy2_1GO_(hV-v$EnsFjVAPg^ilA>dWV3
zIYg@!JrSMJ?3fY5QVthmC*5c0yC82>v*Gn7I7}2%eWs|ZR|r$(#FFan-XVY1(A5eV
zss8N4o=^W*<6LYD)x*u<fJKz|UHCrC9Mi6?IBuMCG-{YVtV@D_Wsfy*&VAFYtEm6{
zW3)*5?{Mse(r2eJ6FQZ5*zY3V0ruL@cPj}d5C28lxQ{M!R*Dvq-zk;;9WAiqSP{3L
zX0iVGP9RO2iBwAYrv=6;CVGJm__5xFxe0x)fr$Dy^F^scBfnsKN_7o4gC{Z~+|@Gq
zrkxLiS#VB*LWc2k+INsd=5A<Cr)yDLlsfnXdoOksBw1B;r28A2MqwjwIF$n3=pox$
z!ChL5Lq#F?F$+3GzXeo}i<(aYB8LAbyJ^w|GZtH_mk}0TqEY*^@9+5A8=!<Y0ET_q
z>L_FXg&#qGh$VApK5bv`DQhf3Ni>t9mAeeq6>%TOyS&CkRG%|gL$hLtXe7xg=5Gm=
zw6;6_a*~}t$>Lr7VlmmTr6Xn|pArl{**S@l#AR~cM1O(BuC-{YtaFo@g7tcd*ZkdY
zQ#a78g+n!XEs}@IE&~G7Jbn}FgrU#A>H4x+S3#_tjg~U+ERU{Lg;hkxo4_jiB4oU0
z47qjw$ale~NR_1;6%)$pdL$>RHFpW8p)$bsVoUw3IP-zCm)7cZXDz@hm9sGa-e6$M
z^;6<16{fhx(CDS}zy<<LYWu;EBu}TGcW|Ylr0B$))o!D$y!d44wZ6yCVG`#n`vof3
z>Yj!Z3fvn{Xr#YE)LKVi+$ljm=j4XNJ{o%;zZjdTW{60;Ie%FTV);i2HAIBrrdnV)
zZl!uFCDKRLeps)>?u3^_sWDVA-eqJ!)`O8^@k{;(t_k)d9d-IR3F-*`0pPX;_ieGU
zF#3*B>w<xfO1rnImaX+{iS~@tJeXQSM!OJ39|+BV_7X*q3~GIMkTbgXSPD%}@?bsr
zW<Bjx1+A1#vVqoh**(>vow6Wf2--;i8H6*UU3|<*vHM25QY*)364^h6r@x}Flu^b!
zovX9|td^5&73EzUh}I`$QELV+zoZ5nyNI8O^h_T=-9z3RbY?_@AN7t~6~-R0)+B!N
z6f4hZk6a`ci>vCxou?vMFcG#69eW#j*+aG+?{2(E@1jAzlQEvNTVQLZYET#bT);|F
z$81VjXFGDP=EDn)HBIEO*1_h9h^ftHaxO4Cy~(856hGusM=Q<aD1Qzupum9GC_-}v
z0<a8TZ*3jh3Z9m1p14h>0O_?cAD#{O$wcQIfm5Myi<xXYD^l6RD&o10rqkbhFkwWa
z%kCcgna_s?-&#a~^UqwaU`&6i&;r5rf1~CypnrQO8+4xEAbj2pro^|7G)-1Zm}$6O
zmP>@Ia_#3gTeW|s(FiQTY3%!N+rUR*qA4`UG=qbk;C}IXe*NW+FzMA`RmXO-(Pm&v
z*P8y-xG5<~Yw_#i08pseGV=)1O@lB@E8Bf;!QxcUy$D07U$UM2n_cmN*{8;^r@UEH
zp+D#3Th6s);UF8LtFh$~5)d@W_k<q~&B}fdrW%^vw`jQ?dk)80O}Mp+^V#b-HQM__
z<$1n;y>Vaq7%`=w_|Rg9c6X{B#h6?8l|u|+)R8Tb7aP>kG_8JdA-#3#kpI-r$QBPT
z<6ZT8Vf)S@%!Qe@*?qdb^Cxc{F_Lm5jds~y{b+_Jlewj~IS-Pq1`9j#SpM8b@}`hk
zN7`jON@NQP-b%AuG)C$6Y(9O^dgGBr26ds_JJ>22lU4t^i^yvo2@5>1IX&$7#R~M2
zUcb^%(Z6bothA^2Oh?K^2=N8P7>Yi4bnk1_TNod=VTQE?74e3PtnYeK0+M==SYpAR
z(_&Rl<LE;N?eFj;@Aw|z_ADsdbrE+kgb!VqdRUllvi^|<(%%fH`qi&ozDh=Nu!a;*
zqw&tWW6yLInjcjM1L`X25&TUa!LNeGdRlwWlAaGf&tffNqm?*#i;C(U>jThJYJHuc
zBa`=)aFt+h0~zKIs6d{9`oiF!XB5Ic0#3bQ6T>G@_rM0`-=sfPdZT#XUT%zgr#C4f
zHBSu{?wfZzTihw|v#@l3x|6lK*nwQFA}U+(=7=de1I@C<87E?CiB1HDTqMJ0%A~C&
z2+Ssvh_ge{?u3<vW~gRhn^;h1YLV|*>K&=-o2hm*Q~62w>XoiBLJG*ASTc@`#<ilh
zeo;4QB8o3h@V^h0RSQH`ZjRu$6_<!Z+5gSa@*#d#A;w+cEaQ!w%z#ktlz1Zfvy^Nv
z*#)Wc8Z1sI2Sw4pEgy;fOeh(y%cY-uLMhuA^2ahJV(<+9eU+fJzb=vw8g|dFOW0I(
zzHl#=;sRH~om_{j;0?4zsEy9|%HEMj)=F%ou_bwkRXx)hHt-}#;d-*UEyXiE;@~Gr
zvw7gT>&|A4-g50YTBBKMee3g5nCC!5)1f{YHN)`U>(cF@R>T+EJDdG6wq0|PF_bk%
zOvZW(upyOGN;)_zRE}T#$wCr$syZ>v=~;`Xg7t9-6<f|J42|#9Vfxti21^uu+3GZx
zTp+<|5W3qwutwnDeiB<gc`ZuqF|MM&JWphL-;aY>*MQBq6c=G6pLw^l$8t47qy0{|
zPPJ-*Y6>Zw6Z$l94(0mS|JirWvA$lrVZ?&9O@6;~cQeA`2gR)reG*+ZIK$>!#P)9U
z;B1=b3hlD{w!d;EJ-7w!G9G`q>O2XGqPo<t+j-X4>LB=8GNbgqi^S=()W%VyG3~FW
zH=;ATk%|&OglI%@1<sm=1fFYE?lBvCMT>Xf`wbiZ^q{q|2RFfW)UJj_WQf!A$=c0H
z;srBBt4Yo$8@aUE*G2=Fo|+lYLI84>@lvzy$SoQmXg@wwPnLhnSW?jqwP81O8Ew9@
z<Fo<ckbf21w9aZ0*P78KCg1By#;)`D$x8#m=4l_J0JHL|XV(AXG6yjFF?5J;1Tj<*
zV&J`Ok!jl28s!(cX-O^Ecr}f$K+Q@L9w1F8ON=>aizSPJ_|4=0)CF*L<-E97q}c;%
zG2)2Z=$T0(pTuIsd8L1K0k=POJ`diT7V&4q*bM$gt?z&043l8;0mRiY>oWbE{ENN8
z2`7~}H!=6{$nJFYDMJX#P7=oo9V4)DQ@~K|<%uQo?Oc4b*A`B@8$LH9J{c$F#;mE7
zo{0=zF3lOMu^%246Q}COual*aqyS7g>d5gRTq{x;1}kGmi~u5cO;+LuJw!{LQY`q0
z=iM}_cboRS_f(Q-zB#XdunQv#?vY!u!ZL6`Q@)i>$pvNj1K8)qPSoGPxS-{{cLc$Y
z7Z6(>@06!4*5`@{K+`FN5!)Tcf4>AWHwtsQXG91*rTTn)%smvXb!9Z)5uD!#V`0S}
z{7NGJ_M^=+ROVkYo>It^pKy#($BcgfH069kqZQMDd#2)M<}D$xA#&+;Tb2%b3=TlU
zGVNt^CM3g86x_tyBswhDKhMoHk0&9$znGmcHhsl?9?X#D(xE*B4}YRpSRQD3$N%=W
zr7I-<b|7H@jJcgZA#;Tx;qx)kS>TmgTC{`|-Kpjlhw-NvSzx&tDFcOAna&o><%mPY
zW=Qf&Z!df3i%9GI_evwOe%+9w3))!;v5?9Ru#^-hMSGbpPe0szyDK-!<f3n2&h%BJ
zM(?hD8LaJPiQV0b1odMLhR5|?AD4DW)?%}Du)fPQH!-Kli;e(DQd*+)QI|QElr^8a
z`0p1}gdX^f16S)Vv$uVYxnB}u0b<%2yr!x({P-8X-?3rUJtIf^LEkE)XdacUByOT0
z*7<|Eha#kC46E5|Q~Hog>41Q21f%-)BjT0HACVs;gAIPmi1;N2?NnwMB)A}`r|7|2
z=1_BE(NprBs?^b2%e*i}e?ZV(h0FcAT_DsOgXV5TY|%2MI@EqpKG>k@bFHp%dXnK+
z<c?=qW}cTR*?PWq+5DP^{;#H6`1$%*4b9pgm^J^SsP{g$o58ms7U0{qb0#(08rj(h
zLlc^};TF&)pG}zDneke6^9uQtJlVP-E4uNCqg)pHH(a%|zkXsstFvGBcbD7PDK)5u
zi+7eEA@^^o0VM|3Y@UX-YpCb%CBqqkisw9T<!83j^<<A;;rz<aE^mYA31WO9459g2
z>B?lVN=B2CfNol%k+p7T{e2y01yL8SJ6dhSto#0}Vn_e&<rFe$lef3ZnZ&Y2R!cON
zUp4w#|7n|$51NemTG!mKY#Db4p1rN@F}?e{2byJ_A6I(;vG1^$7GPa$<vg%vS!qmo
z!UJtEhrvS)XDb5fk)BSZP@|5z@WJXGrUlzVBB1QIYgo*ASa$BT7V}U=UYg=)FtIa<
zL-}wJ>=my<WLapq{gbujA~Feg5~~wwuho(T9{8@z;IvPpzIfb{M5c)J{o?g;$Oaa&
z@@dP_g|<uo$6a}F-&!j!1}~fJ>yO_)PfDMqKpr<RNWjC*k&sQEk<ff6F#3|1j-3NU
z?G&<Rn&0=3KQ&s;73wgeJKnqCc~g;7VP#~)-X$m@;FRtRwhtwbCXuwGCyf*!P|qrG
zDj0uG-cmHarS5Gz@HlFiIda(;7ynz^LW}{DoIBmBG9H=MyT~l0!`*7@&`y*;y9|fg
z^C8bZD$f`s-e|EMbb5T!MZ09SSS1kW$%5?F+YN8(!btnP(3MJ%c@7yqah#<6v!2eY
z1ZOioff7W>>w0J^$)M)n6?P@|elU%#4nvT(K^ZTqzUDpImyb%}yyeBh$f9W7zDA&t
z7b%=lv4L-xO(U|k()3X`$s{e;3y#DP!rqMdYLcdDtsBZ2-IKl@cb+?W-2M4F8f=2<
z9SB@>;hOQNGnb=IqGMiG94!-u#@-oF#c*168&0|)%yKmM_|;rAH?FY}QHctrt`WtC
zv)7Bs(@Xr214^N<f(29YvlKo20n!y2UZhCVM#brViU9uIPnerJ3t=pd7{(2JtKW9r
z`Da++A$!`JkJhX)Uw)$95xm(DhvvMP!T1UQVps-cB8Oi_X(EK~U~7Qv$tAy7^7exk
z8hKrsy8^Zg8&Jo#i_rZNgU>2;8#QD47ANuW;s*Vz2{(A>l%6-Y^{PR5#PFN+(P*zl
z0z<t1#uU%Z|Fnm0S3bHw=8v6K@pns%!S3PQO>FeO`9HQ0@GhoCSHht*<?6nCvSZ)L
z7}pFFJLuR3B)mZ(S5Dn!YN&UUAQqR=<;UJQ-<2Fs#Xhg`*LMCZnIB;*t&}6}Kq)9@
zWGRn!Kj!OtTA;raj~)fSUm%HSr_4VL@_oj4zylQL;71+34SO_OL9nvG!UPv9dZ9Mp
zy=I^W8E_D^{(6_H3{QGgc(B?JZA447DB<12u4t!c>_<pPZQ4{Pc@<!{;GoC8=a8qm
zovU>~GnY;z3VwPMbIt=rgoV+}li6jr$X^VCck%sVhbh<U7@Zm{TKnsUA`ho;qZ0=t
zdtUZU00Z*=W@7V}8u`_EOwmoplCP%@-k{|I19E$DAhh8*CCoC$WqW-8lb(^`flV#u
z;4m{<HD1cqs;6@|h&azWhp%YFXk-;)!<>V}Q;jQslHG@c{`t(AGfFKh(C^p0DDSBr
zl?c*lR9#2v)-WjsCP4sF`Bl^0Zi>%tRr0vF%IjS<{-ZJzB+hp?a#GK@r_Z&8D`YUN
zXbv9RX|X7e50g(8k?sB#y1?#uD;D(rD3J)S+Pp2->QrJn4sG3lw_S_yVe^BrzC`+7
zjf|j`<BSHUMxZ$EtUxZH*2TbI>*8hlPlN10<|l?MoVD+FZ}TWk?uONuoR=8PEQ$Qn
zsdbeuEQ?hBL~3r{3{w-FM1;-LNT8~k7P04%e^)gqWnL=D{!Wl-05;LBWe%tPq*rf>
zxbmfyW&x+AZFJGCOJxBh;6)0S4l(~ZYya7OtJz0hWSG+-@W^0Jgq!Xm8Tvc1r}W!)
z;b=lr=Xq0q>!Dir8PU*#mDLtV`4A+;0r6bIE~q@TBCJ8K*w+x#pFKi*VHu`kR`0|?
z6mgL9+a`s=w^!u&f?g3}it`I#>{<gdD<v{5OgK5k+Ao0*wId37i%yPY)(d6wB8#~+
zmm|&uQ#AwiGXr~>ELu#+o3bhdb&8z|Rp!l+y`LX8U92Ew;0JfrK%@APEo5ylw8|;d
zr;n@Kw{Vl&z`+aI|Cct%t8rE7?X3j9KY(X>i}daRW?yZMv`4`m_UAg+H%mQ)O<rtW
z{dEI2M!o!8yIPywM4=!zZG=+}zq=A7nUX_?!`41%8&Dan!2wTXA1jkn)F=wK<%8C%
zhc`nDO|Ea|C4uE9Z#Rt_TL0D^%A6QCT3I{yQ$TFz7`JKbQ8^Ujb3a_M1_<^b$`j~n
zM?<(=wxW1~pE>e?$NBdBO0t{eR%wli>m8+Jl(+Dh^i_a66?*I(J_J8d(52ZclUTx;
z;&&e)4bpe0Ipygf-0lQmd(;9w#(Tk5uZ$2nY2u?YJKk=uIlP2F#3G>-&iwEUV`8zA
z=g{Z$9M#>7RBP`3K{BbW;U|S)!L`h}=TFytj+`?HcNV3y9Rz|SYarY2MorI6h`*%Q
z=%UTD)B)M${zVpao0NV@3K;~<M7{y1tkf3WFX@xR(E{?HW5`@E7`;n?A|X^?45)GI
zt--gqev|h7eN$(keoN#Fn+U)H-7Xo!Cf*Ovf7Lyc$h(j#;=;LsSsh*B_mJ+@M&T#g
zl-8)u%h;=fSe5e}yB8A`$}V>|MVc^cMuY9fsSg0nH-fh*Ca0!l$Lkoq?*c~4G`;9w
z+>jAa(Kw*G=yIf+WJ~dlpU#HczIC;i-uq?xi>D#E>c`rj3oS7>PAdSdhf1{Q?x#1p
zdP%SsWv(R?{PKDi!WH52w9b;($;GDMeMZV%3+S<4Lmtj&{|Dxd&cMY$ykExfqt1g{
z(1?FzOE)OF;wP`6yru1IHRc_mR+NVZ;<`#~zlZk7UvCuZUES{W#c`BBgYYrN`p1f6
z0s4J@#tjA9BTcWUBZE8nWX;+9$%7MRWTqdU2%4nQ`22_}UX`<QT&{0k6W<FPybS;I
znax`lUKJkQ!VAGW3wJrvTwm1{NwHhAkWi*-SWM7GdCR0$xX{NYKcfDJ0vAkyo`uGn
z6~qS}O7X<50d89(K-5Kd)MMjlr*N<($9-h;JMxBdhjUp0DkY!mjWK}Y+q0t4`J5SF
zc5?a{ODx*`<@=wh{sIZ4sa7yQOT8my#UmTYkGikC(8(z+T^sY=ou2g1!*0R*=g|q$
z{o)@}Jau`XODSafOSGq(=CE4stVqGEO1`YGd|8QtPyxWu{>TFjNhT|ZJl+Ms@o>_9
z&mcljXx~QThqyxu83j`mipVZ}z}zBz1}o=1{b-uNs#c8>z^`_cAudrOeBqa^>ywNN
zm-DCn6tgMn<pP<DXoM&?XQu68Qs{$e6r$0AUz)ycwtJ83a76s+G`J4TVM@GN$6tU@
z^fSrr{R)CUg>3KpLK>r#X(HS|DogV1(WQR?Zp*5{9#aewTVK&fD6YecaW>s;=(q@l
zJ^@j#;0&;0nRm)K0#dQU4Qf+os9rMJbuRtZO8=1WDZ~n<5aF=Ke8;rE-K{%J@8G_9
zw3j21l}5Bzo46Bx1_s%_w0c;Y=RPkKBbCe!gW{q`er`Z9-1VWU*8&A;Dw%Yy^;%ZZ
z(bvb##jk)bPR<|LXF}a44Xey)pG&&w4~LXLaO1G3$3KPUrd^@3l~*(4+RBqS2ggi-
zv(ag(RU$vMOr9#w_Q5UV9GVsM-;V`k5bdy)u!BLAN={^TrDLgtX~0|BE-}d|@ZBkJ
zOmGsZ{H97Jdj4S3?i9lkHug5W#IuLTT5oq_<N;Yrl;@~7ES>P)Z`$9I=7^p_e_>Mv
z=zJ%=JB$iAAY}B1WiYjU_V9Pur4!l#=^cEhQ&i^)J|j0doV&_`GxRf0p|7S4Ja^1t
z9@bv)H^LP&(E^MkYv?vi?t$L%B4HO@%9X<gWcdk~Q6fa*U`pVw?%mS+s6?(&Ep5}R
zF;w~kJQ)q(Gc%rtZ-FW~%B|sa%(6X$nB1Am{C=|oqyAVC$vO2O|2WSrekde0;WU?h
zh$2UCNb*Sjr-JCO4vx#8xROG-hRe+6jsaOHsZ)+G3@`FbWRYivWQKHMWF8eJkzAO)
zr(O4Ae8<acvy}H}5iVM;TX(cTy#*KYPU8vL9jcH7jM-dHDOv<QpsRmuVH4^6n*Bf%
zxdgnq-ZZN;fP-4z+Cv<i8v4$@<5fl#8xu#P6etO_;d#;6KZ-nJ3|iv-drij4uPc*-
zzI*aGaCIZ3g6x(4cHr&CGcpnZ++5Ru2-C>lR}8AercXQ9AEoHv8GUw9XEhFD{-sI`
zSx11sfg-`&Ui!~}%>SO?arJlugkcJNPTCc+nOCd<48qc9jujHpuL8UMN`bO>RTt68
z9dwNnjByV{5!l?+C#C8zuOXKOG6QB^$2GP22ZVJ%&_$q7!ebfYAg2#O@=me0XFXrC
ze@){tf1(;tczI0oGBOE|JXR4xCsV47p&YKf*Tfa<I;bnX>kg!&fsyT4OpDS2K~L8+
zBMIY)6Y`3_J)w*(TCXEk5Jksa!q_mc4Z7hoQ)IR3(Z$n=y}Pa=eTHe+t)jwSCHMTU
zj_M1Fk80?rZcE4->a!&LWLYd3L8LP6`FA-XtV*lVf-jdZmg&q?$V=~HV{D>?>C90@
z&kLznB2X?cPH5GSI_@aZ#$u5;t>8h!SRGh{V7-N&xP;dM%(Caqf98R30D&bgCvUB$
zlw#WzH7lvaL0kX{I@9W_7yV5?0T^prsb0OUY1Mb6BM;Hh6FMu2KYY&WrdMir$AJ9C
zUEw}v#CPumR}EL#Zl++`Ci)?MwT%YkKx)@Chs)mYzCs3q{x_z5&W*FWTBo*R`tz7v
z$uKTOeK~!8LJWag4&WiDMmC-F6)K8;mR&>K-5Ye3A6akF{GcL{dsCE)MUp`e1mpHf
z^Qr8}ycE(buRrB;9IRAuSNoOjLE*i4ndm$NyMdTde0qY~MC8jOTf%azPIuzzWyF63
zoaP7Cy0^T4Tl?u8K3!K9T^-<^?vQ9Lp*gd3_f*6wK;(l+Jn%Wnk2nl7hzqHeUBAaX
zUoEC)Uv$_PaI04!-(;zs5n~3Y^1Y?<Wk;@ZzRoEW17aU?U*)>v=t=`_sR*I3`Rp%Z
zR||+~Ys)VzGE#`@=IrNCt%(l2a|0QMru6Nis*A1$Fyq9OYIgaBS3iH|efcxMHc=U%
zbowG1f3sZMKPSD_to&TFTfsxV&C=}Qa)EN-+Tc|G!1zy#DU*$fL71yUrNc@oK>73+
z^mcFN=wFT+{~u{@9Tnx;w+#=aqo9CDi8LZeNJvVjD2SlI2m{g~Fn~jdlF|)IBcPN@
zH$!&_gLDrdAYGCY0^fP<`@Z+{J?nYiz4yD`wfJKc*34Yjb;j>Jen&JY$atxUhn6Zf
zHgw+d_}U`xHMFMcX#Ae{;*|Q&4%Yv)f|v+bWW7}GNIaAc^yUlr#d43BOpi&EZ|nX<
zXz#hXB3h$u5_Qucx2x28zfESZ38)$zpo&Q0v`o?om%Z`j6tA@PyOO?gV@y-LGONcH
zIOBbRMe_=Kub$e=V`s#c<gA}E?<~HRue3cgy&{l$B8Kh0N_SPOwnX^R!}>`^Bc1PA
zW)K*YPjxf##;8mpV%^gR%6<?5k&3;1K*yZ3rjCcq5sTF1Jq^2%O+&^vFTD+9Im|v%
zybAp#27iE)zsGu%G(z5XKU#ejA1~5PA)rs$x9~Bo&8G>82MNuO+28$+gwfiCl@ES|
z#;dj!8_q4kpAfI^C9PSM?v+kqwk$XO$GHO?1lzZsk;)*>tqOcZ3QWx;MSkK)AIuvp
zT2jt=p-VzsV7(^%2DQaN7-1AOSnlOIhHnKIH6-v~cYcJ2?u75r&&O^PgahhwQQuFL
zwm<(N>BlHh8KzHe@(iD*Hm{zaTK@!<c(Iw8an{k7`a`@Vo%FV0@3#v_h2Lp!Uy{=w
zL)|$kZ9Nnn$)w<3`*fMW;)pR?WBaX}n)}D!E~`>Xsr_k$oiHbfgBS1ZZihH|roir1
zr?Qn%9tyKh(&bHiI91c6R;^%-G3(!y<tq-SUY54HiKxSVNT?L?sE&lxJC(IMEfa3B
z=6bB^PZj8nkSS3vQ-@TodNidQY&dZvZmhp@Ie@$e4t5@K{zpH;^zSzR|1^5}<2N*N
zgrfrCwl}t}#Y`Kx!ruwp(l{Nis567OWiRzxj=%hIj^+!6mjdC4iEdWAXoQX)4xDx*
zjdSmS(+=MeB=$U_Stb5Y@%85ljrZs1{%@cC;R0TbA*!@bX~vuE1$^o&8`TuW)@U15
z@ypZZE|!Du8V~;18fcWDqZ$rL*HexlY-k)JZdHYou!7y?vPnMYw}1Fy@_r4F!QsU3
zEFQ+O=LcY~E(Is6DadUIoL|*!->C`HX5SeiADafjP|mYKH`}Nde9ueSBmnKXOS^E*
z<1E!8i2WU|&Pf7YO+!43{Z6+?K){GHqx&Fl^_l==X=<Z|jjad#*StI+o;JD0=#I4k
z+_my%DZmCd?plI0(7Vj8F!PXIP$M&y&uSa|aZG%r2$lf@@`0UycrkJxlnI&MK)TgS
zX><khgfiyu9h4>j;A{(2NkKb7yPHs^6&b@RU^9xm1!{5h1l`onbFU$dGB`$scLU#;
zu-E@G+#w;x2Tk#6dhnR{L7tFe^p{m~xEFAAO|=D<1P=M0CIJmED_TpBfu_0e1SA=S
zOTnrd7>falF%BVv!2M@+6ff;IGy*0e27=?enESw}4gVipOT!12pTPNMlz>!%jA5g5
z7IM$0oAP8&{JuqsY`FOqDCB7oSP+%9{MO@-W4e66A1+joP|as8S-nB?0gqYs+;(7S
z_ZCxWtGd)-hOEc!&b5*r6tL%!+baH-O%rhD!8g?5TW0>>hv9!4lmEwm`1k>*8=n})
z;IMyoZU2Ag+&?FRcRtWE6mrGyM*hFNNuwMXE{&a}txA7_!~f`Z|F^FsqXU^AP^yC5
zhW{$i{=qZ*Z*S6I2yX3n!}$PX*1uj{{_^{M-~tnT?IOD*+dqAia}>by;AKNGHxycB
z#K-y>vFF<W^xVp?!RgKhSo#Ilrx@gK87*st8Q$UmkNw;=ydf8YWXa(7;Ku8NVQNUt
zv`qf@<DuEV0hl;Ehw|50$E|aV{J~-VdDxEgd3EWNQvDZSZ|sBhqGGuSoQB%XfoWcr
zb>2e|qQ=7dz9tvAhnDY;TQA;T#$3X8B>|3IvVT}RN#yDIgf{>OFGI*#Q2|}lN>u_r
zq^g{Qf@*#b%)vY!tP$xFaKtEvETK+;^QWSwHqt?H0wfP8-FOPjeGdbn4mkNQ;2`LV
z?5f?pmC7Yfg+CV{QxrZ1L}NBVH)3h(;c5ZT^3&_m>Fa<C(0b!}#?pH5UQnLg7&Jpb
zcvl2SWGN^C=^6v(@kaj3WRf>j01PH81BRDu2T`l+-;WH|pM#Ip#{i-IYD&#G-gdT5
zZ71-CK|`5XdQ#;8#T<uVp<m4F$-r>|GS@A0lk%K{Jf0xami^g~+immUum14$i~uGL
zs-_F6woYzX{`OHI5F5puK;|#d0SZd$qZ@JTW$^Z8e$}XO26@_^5>3)m5PATAwu!k2
zY1HP+_Pei+ralHRJcq1%Pe2q!88Au8+Xu4qf=|CDBvq4t{>Kx_IYRC8YbXM6ot5!0
z5OQ1iJOHrG*LOj%j53jr_zKpoCGh6a3lIl1VH7|gTQ8K=-k-zoS3Mk|4~_{+`daKj
z43q>pmKK8oJblf1edH)l&^n@n?0o!6d_Az@0PhqrHfE7(OJ`f!NA}6`XPQA-UHT;{
z;7G9sJjEPJccM!`>eh&5h-DI|9V95S0j*Gmniei4Fb!3yq1{y9H2n(m-%W}E0{oZj
zcY!UiPsh9dX6;A_B~)d}vIaayotFW#6{^68mq-sp0Ij>!AM4}H(-<Q(!N9T@n`6yC
znPWVIvlz2b=zAQ_J0C573b>i412V^0Q*Kv#oxd6=qVm(U>`xOZm3n%zdo&DSHpsSN
zUpub-%tl49;6uS^FGqZzOoYgs`oPgG<jTGxA0UIW+M22TrxF*XfFOS#MEq5*qbZiz
zQ{sVci0ZiO*{LyE+SZ06FenedNchK?b_WhN74nzG*g8B@j|M+&SG4USUI}H{j~o^C
zc<h&4x*}!32bm3BOED!6eE95md;|*l%H2!K!(XlVW<xopP?NGk)3FG7AVZmlkX*}#
zMUMZuRle?YU)Cn5&>`2MF>sQmXO%d@?tbO;<j$TsMUdMmU*aI*2D^F#WPH?M9>($?
zPUQb_KcKjT2Y)SsvzH$jt_HMHyO$e2GI4}Gf28-S&VI5!oCYCdlcU=rUB&Tkrh2_#
zIE$X$Q*S3DinsSy=;gsHJisfL0wSbI1sFH}K<gpM*TZ`a3XjBc$BkJbNrC%th*ap@
zBl&9-)F2}x)Kft9KsPZktOAn&LaH}N8QOQD3a~T^g{3|g7*#!@?3)|FhTqh5lrICg
z=+-%D^F=UZeX90JXS}KH(mS9;aNi<3_f@^H%}1oQ`(*{T?;{puhs4o$=7epGO3K#2
z?6{5pox>D4bhH5+!0lh2RfBEb25@q&&G6G{9s!Rh(Tn^#k<d2beHH;W#D2oRUqcB(
zWu#lg%EM2v7E%;n#=&%21Mch3RjAh3m$klfyQQZ(I#N^w{S1Zn*DlDY{L*Hq4=DM#
zeb3z%ZdQU_mms7}H&XiE!5TrbEy;Ge6`Ad{@Uq4AQ9aN<e^>||gG7*o_2OSoe3EoQ
zaR`<$$ZJd%OW6w6GrYh4h7D(AO97(o!zRE2?j7VTwGRTeS|HyWohEer1uvV*a~dJG
zJ3rHYfxJkba#<u;cuk69ZTCinoj}-^bx<GJ0bGFd%PLDKSjl)hqC<GAKJx(3upD5Z
z;~*-ncUvFzJ!lJH(~vZ(I|cX&{Tx)PyV3@^4Cxtq6OGqBV}o6p0_!k^US}q?MUt)X
zgk^fzEpo}nh&+!iT^CSjD(iv;P+F*e%U`6XQ5a|6d~X`EqH=y+ZBWGK2^n~w^S60c
z*(*mp2EYMZ&`gi!DQdamPt8^52}Ia_-_(<YyQ!YJIZT)*+3L=AI=f{Mun;Z31Qtr^
zxi!mbcEyk_Z}}pRaf3GQ<)GX{vH7czPu4nc&c7u^wfJ%x35c)vHy!vl@yP6wm(Ep*
zfsF*Ttl*<f)yW@9UJ`c`ZvyG<0=*>3zx=j()btyv=71BRd&Y&6jH&O(NjAkQ$`Pp&
zp1`WFtvZ^=@!It*WI*^Cu=yx1jX9W@+Ar@YaeKci-e$t!#Bg{9*#bg{qtZ2P1K0V{
zqZ%M%WK6cC{^hZv>Uc$Brx?i6PNet0;bY&<u1ue&isQa)bYX*6miieRh8=<b0P#KI
z+(_$%4HYjLrUTwaGwNStIA%<>wU^<E^yXn*)wAi8*A@#?6M}cQa~2cUH&!-QvMU@@
zk0#Mg$)-~I>FA?#5;Qxasi4eGp9Tj82_xG5tRz&h!>3r{L+x=bsb^vP=&4#T`O~Fe
zJe>GwyFj2V!uj|jLKaGL>^DoA1HOxt>(?qKLMJejrnzRh?QeOEMaNA&;dvKfNz068
z;R)Tt8XmbO`K>oZ=Xkt}jDTc_WMmm~OxaBUR%;>)jteEPcEoKXE4_*qfUQG`3VQ<E
zIOW_aJ~8*BU$y34kK77kVL!pz!q=HdT6UzIqMT#mDY$y==(;sE_YAoqHqj`UGpoZw
z^4pgju#6X6>2Z7J?8#$N(-onIv9+j?1c)2g;lrEsdBpwbu<O9<iuU#C;UZT-FO^j_
zcA3)+AY1$*S=QWvCE9GMt-V-G^5$ZOvU{=rJ?|nPv765p>1+yRt0r%>Ymc0i5|8$e
z6k)%KM+*sC_cr>RA6Z{K+`Bcb^;Ero@j&`saa-p#DkbdtMV4MlLN-wGW#4~H2xjVE
z{M={h{S1p#v+}ii=DrsLE$Ymgax{bYEtRmEzYWsoIqIgdpjGK<3%GZ>fxou%N66fY
z8Uw7=TI4XM)!=Yp?SWqeo425dSbR%tDl9}E4WWVluNPjiNGYHpyEPMntNh=<Ab)Pt
za7FzZ{9HNc>Wr4=-4x}x2qS-D3-%C7Vn>^!Z5L2k`XA((%j$-jn(X*eJcP%V7J&nh
zM4ifAR%sxPe*|CYAZQBK{YV=NU$^%Odt~>$!MpJgY&s!!taUQn0e9D+gx?~KuC`yF
zQazqq7u1&1d;Z?CC5#%xYpMnHER%4?7kPXqSv{sr$oPz|k2~o%WIS8rg9#4X2EB`7
zZGOl#vk)@JU65oUhZW1JRYF-D2f|5RgVM$~6x4R1Ai5&dHE01?2fXSi3kDX+?FdU_
z;51cm%pejDT-59tRVJg_z@}mWItljz`Ic@LeyAhAK*A^`ff>0UDbdSb=zs6@P+mq}
ziY)c|%kSPw6Pb*fE=hKEu7!Cq+FTC6nDk2+4S$mXC>oZYMW;H;JpG&w1^N;O?IB?z
z6YS$%Xh9S9S@IPYAlTwfwjxfx1i!B?HXlv)?Gbe;r?2=f<+%78Gr}?uu4x<y3t2Aj
z&)3xL-bGJk^7LOOztBh81Nh0&rq?4xX9I&LY+ADJu9v>ML?@iPlHllVm{jLy8R>BM
zLg;s|dojXq$kMBka-ZD7zL;>~H<%wQCzzO*;I>WZP4G^98eXb+Lf&to!k&9WI_(Fo
z63r0=meyFFLmBBNMed$D;5f*i#UCd))yHuD(lyhZ+~WJ84pzyLp7psqy7)3jcjm5p
z7ey2^qcYqMBNzKRN2k}S{L6ilkPBi>pfgN`CPAz58@68ic*BO1qKT`mTtJ?7PXAg(
z2emcDa9o*(<3L~4hlCeU{mcKbKgCF4qP1IOb^R*M<6oda_<pap$k_zP75piQzT$AJ
z6}Z}1gBH8(?|L7J`4O+crO)W23x{l^ibc;mt!bf2h9!ppnZUDqAwqZ00%p%hN?XBW
zo&d)Zu48agm}Ia19qHp5<luW2u*Dh-EPF*eza0e%U9e4MWU~Q@#h{>NOM{bJMtKo>
z@!$`xf`aF-etKgFuTg#rV~`YAutF#{Yw9|@D@dpjW;N&Z_|*5}>a^YNwJj_#g1$<(
zkEk*dOqD3S(9X_~v;+?jEv60loI^vFT10*T{K>|^(s?bZ$|j|_F(flzBVDOgGB@Em
z8q06%s{=nr_VK#YphgjGpq6;cO-@n-4XxM<)4<2Tw{1z7szB2t7+<F4DKJY$Au5{2
ziH!w);Rw$zJp+}^I0ps8)^ozw$AFxiK21Ki-q0%KMID&<O2>M@0D9c5_-Sn5;rGzU
zcVTOgcLDR&rY1Vn{sFJCpl}Q@%^NXPF#w*BM`T<JBC}t)((Y{EOJ?S1xKE7FTqiI2
z3~<$r8i9viII)JwE%>YeA<vhGMHt_E8QL-=K}0pL@bH+D*5Q$(9Lh`WVpbFppDz=;
z>F7IYsagAGJUDL7eps9zKq)1hNUrY0Ce^4pm5kaV^(Jx~jaGiITt^<>fFDK7O%_`C
zQJ=)dk=4y`C$-ytsx|HU^rzc>`6|c2`V{WKWg8#11MJwd;-5SiL5M0I$R&{X_wz12
zvzcXW!+;t?&OV6F@kmAE=g|Em%iD9^p$UO6Ep<3S%C2Xil(OEM>Ie7P)8wP{T(Fku
z?Vk6g4!le@(3@50uY59i<qMd39VR6w?_I2V5PHVAwtmgNF!qtJ!B<%H6>N7Gjb$3{
zGd2rQnNVH77vcQyu&^cgp5J~D8AjUIoX9J0H$tEtYre`MJ<NOr5t!xq9;S*85>1BE
zA?`IR@>AiHZ>#}hUjFtz<?R{3fFvopW@~k_Hxl9OCmP5fAr|TCCl*|;Qf+Fm9Q^13
ztOL<a^Uy;xsr?E{0};2_w>Fx(&|p3+lv^90^pzFOhD)G`rwjHL3lnbn-F)pb^Br6l
zNnW_ij~iin@0)#{$gQJ{mn`)kxz!1dU>&f>@G5eQ9f*7`D*vkIVz$=4!eY%aY>8dY
zReXxh=<URGlgb^`w?V$<Ae))dBsscbs*^@;+Hex9dpV~ecch9Xx@coDfE<_;z5ea#
z?L_mQ<|Y}8#n8<C>vj1AvXcb@G}jg)$UJ{GK=yZUAmc^mZwjyuFxwdB;H$ZAl0OFe
zE8%z48qQl00WkJ2f4mv)HxS)VAV!KO->rbfR=)m%<;iqBN)BgmTm{NX1=G5|$-Z=_
zO%dJjstZor=Ukl|pDpB%3QWh_?8NHrci`5aExf73ilk29T#U39hARZN*I7!i`h8`t
zZ9J)FAbK9Xkd&8T*Hm7)NjI8ME#K`{B82{xLTk0)6B1nvl+hA9VMka+r-TsF?Fz0%
zloi`f_YuaqNNz!UOn~~OdEHEE<>#9_YOS3m`cAVf1z4g>2_Zkm-Fp0l5{Od-sF{_-
zLR>j~GKo6`s6Vy3956JBm<U#I#~rfUEXWmfT6|?)A5wf4Bl_^N*pJB1>J=n3o7$7R
z#M7U@0YAzbtcbZFwVOGrFYDkQvbwb$%Of&eGm>y``ZH-PuZ`v9=bNy9djWJHzdr)h
zLT(FQ?$Bf2rR=JaDaCa6!PZZW9Yiq+#72VDG1@ma<-D;(#0k~N++cNT-iKXClBCMW
zTRUo-SLv1ekHG1h+&21^{YRsrlVaYEbiKXcx*2?5?y`_*slDalNCE1}bLtgx(I~BO
zv$CN3Z_Zx#Te@GJUaX4ZA5f&>-x%Q4znZ{bN9m)-)8&7e{@UKgY}V|0;z(F@dZ@=k
z_9aGMoBV{-tINZe1gY<4b!D@<ck(c{9W1?x%zllu_XLT_5u2rzW6G?^wb;tQ%$fd}
zA!%KDLAq}*H;AaQib*c3ZnTS>7%#C|C~IA3!G)UBj5i}fOiZ$U3#|2B$PNnkP=`uU
zk=}j%-_$$jRxeZUEIhltClNNun-St@2ck_ZyK;q)KPYz)uNhT0Megp$73DjK;1cvU
z@;b9Z`b)qOg~NpVL8G5lg5I+(+;h9&m3|Y}IsLp)Owu&g6ZvRWJFd0z;LdOYZw3}w
zqrG#N7axm+Cs!`rMQ`a337UmXplXGW9=~3tqK2EJz7(z;n;AUv%*#D~Y!J_-|1*x4
zOBqpAZlpgqTYKMa%U9@b4<>UXn?GO?@u@eZa?`kDHm8cpJNCj)YnFd9H9nA&!xaq)
zeu9MR>`K=m8*;tLtMk=ivFT$q+YEAw3hpRBHZ<{Vr+0Fqi^H#qK?GfKx1bH|q3Dcn
zf*#DKWGvNll;@-0y^FFc8i5snC*lvkXe3kHaMxy>)zV=VDa7~yU&c2yFrMrV_$7*#
z?r@S4CutS<`Jwd5ZfN%p<EP}et0-F%XMF?^EU@<gUYdXpkx4SKb=W;GY3y8Bf#9i&
z41{!WB9U>+bSzJX)@;a4iGc4^Ma*>LWpX*jRD=&v34-bCgBM%gok8$W8EPcdx-^2F
zyaVVR9<}nOV)T%mxI_)9Ti<9+mU;r^$hn6v+C(=;y>N_y_2!wIqlUQ9UC6RxGKW$k
zm$K`Jpg!%gp=c$!qs!R(s}g03Ragp?sfZ;LOD(T%2`dk{S1$N5>P2*4&)3H}_%dxb
zeJ7!av)!3AzGyq9lqwL&Bf3ugkrvM|;Us;Cza-}VeXz%N9@+zIhBnNjHDat&X{wtc
zuR~WF{l#Nkrp>~zgZyr)XFYc3j&q5DC;)Zj3A>M)Cn-j*o>J8(PuZET-OLjh7J_%o
z{D`GteT}8wJqAbJn+o%sJ-ZqUk+0AWukv&M7-e?h7x0)4?Q=Ye{1x77b7A}%5CcEu
zqm4PNaSQ|a4Fba_io(cuywWwYs^K#6jG!Pr?0Zn#keKWwF0J{Q^1uOqr%Vq&*GbQT
zrS}S6C2XG|`>(4QCVa^l7|MF`3dpXwWU3RLPU{le>gH%6hSOEY__cOZ<=2c8*U>U;
z)Yu2ak=7NVP_!I4r*5&u_lJ76-yDcSR@HDWv%?`?)q=R&b5oZ^KCcrYooAnuP-kH-
zbDhI|nynik;+c`U<kI3+u;h_@V306^$!fj9iRyNHcNZ56QUa(gel+t$_#GP|Gb!bB
zxK&79GBM{{{fwl(pHjP^p`i8~<1XKtPSQezu{?K_YKT*t=7EqBgcEg^lmWhSzytG&
zt27%-AwA#Ad-J5nK``6XnxQY&xOcPo1xvGcYC~U(iOv?2Rs<wp2;1-bq12yeI6iJ&
zpua<b*=bfY@6(?KqD4!TYY%Sw2}U>XuvobX{>ue(O$s1~tCGWx5*r1R0u~tNwspH&
zuOwrfS{1CYlgK{7Htk?tX*K;oY9uT)0dv-pCy1hPm_y#i^s*!E`_GTR=B=H+C)T;N
z=vI+_+9-&Ta9MQ}(uWD|3+xNVR#}ST3g)i4gF`5MWL~h+dJy2=y|^n{8S<~(B^}#{
z_%@yzy&YHl_ZU2CoM&H$iF`&86~4WSLMP}s7$v$it5l~=&RrHILT1gT(fu0adw!cw
zWRCahNE7dx-DY3bMrIM63nu!KNIAya2`{;_tmg6nSo8$<<I3_Ka!O{3<A0ab8S^@k
zr_UrNIM5HHS_k$4QzU-Ep*GRQdcue~$?ZebR{rqKSc6x~_IttAOB*D)VE}M%MR?AJ
zi_~9Er!RJX57H}|-EE81Vuet1y2;yf!BU&}qCI%jUv~>;6IsqwOBT&?j|VDLLJoHU
z<jvovNG;!vl%;ElDLBJ?oZO?GySNG6Y0j_0?fVOb!lYbsgx8pR-?}}z9pVKK{L3Fl
zPH&}L87D0z-Lhc0x~h`vP+9?E$+Gp*G*}Q5D-AQkCZ=>li`eI13u<lH6-fl<2vU;_
zzRFa)&g-EcJ;0s<&kg$Ce<jD^#|4+hItPQDvHZ}M$wIsD{_`R@wzm<KxXC-(cJ>J&
zewk@zGTQ^U9`rZhfB)X-BbIj&L4*U8o1EDWg1x8;*phKXV6)<_4aJG$bXAOU?xJz2
zmbEthQ_<#ZhQoFNQw#ZY_d8F$@1xkP-9M#NOCis_lcu&Ee3f@4Acihw^%k)g=8hw|
z>ip}yhxP#>uUFED8U!X3wkE=cmABmU>1YM@Uq<m%uN9ZQwc^bG*Xm`XeBxs<j>f~f
z<3vs;^}S52NODrG^UAc#`|op^VZL4$Fumu0JffRF<63e19QC&I5<bnyk}H?1*?xp}
zJ63yz?J3L=&rnO?YmCLzJU8iefN9ur!>W2ExMAfYb{Ve`jOM2eFAzGucbw<?U85AK
zE^y`T4w$MHq}yfvN}@gP(j76Jry{mIoeS*9Cw`KyF4$w@B)q4V`?9!hDZjVlr4E<6
zNhdy&YvV<x9USN?8^j0=%a+Ex%6IL$R?%JxE^+w8s~S}5P1&WykDG%rA)X~4<paz^
zjou%VvWp{AZ{$pBQEFh8EH86O<Ub%^x`X!=wtMEv@z=v5P)7YmZy@F8{9E!l+8P}?
zF0peUJR+_V!Og8-`7@o98}iITq%>BASL35ZF1b9mptGxSKu9zfU!4J160tQmLYhv>
z?Pc{Qnu7Qh0x#E0?3^{r#;{oK6{8F7BConzrHr&Lbtz5KB2~@oYWHG|%E-Dnt`|AZ
zZot#Az@<1i4{&w0o<5Jc5mfdZkIVznF$OqsZ}hd!ts)H@`-rG)H_D%PXR{JpAd)_I
zI|UHQP<tb4r8IkVNwsN;ehy4k%QU~C`CSF>o%Y?LwFSB}TM&3A`>aKC9su``&F~7b
z!ck`oi<6S(RWu?;Y1M)}=9!tRN}U=?(l<-!_efKgbNAR>&fD-R2)RsKZdCDqE9(5g
zLW?FxBQC66?lS)2s_Op@0dt%wshsF6WLGxmT4T%$Psl6eYR3N%qI{{#wTydrxGRZR
zy?PLzDfXkj<yDhQRIl_0uW3IFp_&Rld_V=8psQaEJ2A_%E>sW)H+15fbuRhN_2IL^
z+`t>phEeU;(V>-v(&OnN1x?$=Mc4HmCc%JMJ^o`rsA$8A9=U*=0)I*wZjJr|qxCl<
ziB#(UH7XFN=#S43MAIPgrRh0~HA!xo@kzg8t&tN=&e!$49z;oX$Iy6E#NE3AL(eUO
zDZHI8IO^W|d>+sR{sJ=tzp6@>Vb%Unbqk%RtH}Ip^66z4SE^yt{b|p$tSHQb+QRJ_
zl<tHYfi)UIh4$9i2Z*s&$_)l$mW>yD#{hQK3o3Bh28~Fwy2kj*q($8daJz_!o2ycY
zp9)ZqnQavFL9NJQ;HsCsaMw&plQ-9Di0gc8cwyZvU601pY=xeg%BwtGn4?R$QR7!4
zFQL4vR6DpSr<~zj@nZYo0UlUY<Sq2Co`etsL29mHg?ao4lHod)F*7RIq@qiySgvLn
zHEzi4N4h061T=noxI`o%ssMicRlp?(iM{{e-=((sG9zL&hZ-43KCHF+bjfi>ccCdH
zu|SZwer34b8RyVm7(BlrU@fH+`a_a>)HJldmfx`r|3mfOW$U0XZvR>zqCc63n;~F0
z0LK-36P(xP9uRsLQ9PwW_4{|g`KXL))1JbPyP#*+HvVBsBDpSg%B~IHFk$uDLIj}E
za-gZ#jVdpPlpy$##z8llR#J(bOmJ<pHZFcjZPmK2*WOXVzN#1FFG;fntwFh2{NErF
zW?3<v11O=t;xo<J)!KuomG)8nO%JXuu!B6+GpVFo{jkUV+*l*tN$oXNBA5m-iG$U7
zR|>T$dXN7KtNEh_(pUa)*AKmSzi%arW7*k4j4Tl2SB&LZaazi{s#6758MXO#UV#O(
zg|>b|OBb@&1cDl(Du*&a#H+CN4m@9-{1Ox*WV?|GZ$X20g)JP$==vk-s=gC9iXFgx
zaS3`X7s<ba-b0+^RttqzPn^4V{JB`vOeU(IuU&Bq;N-Z41OY_(EtmpEhF(JlMgK{S
zVTW8|l`3W>cmnqpnbC~y_MPH|WnfW497&G-D$T8BdVbkWGD!f)EGd!oIp_2RMyBIV
zp~@%g3{n(KOY3#L>q5G8gx5kqZ3yiaU_l6n5UD%ouiMM$Q~e;{!k_hr(qH4}|2&0}
z2}7)XA-}3pAzb#HRA~x~tSCY6ea@lBEjAMYTPW4%fre8=sf8~W1Z*y7e=LT4P&Y~>
z_ml*|=Ntnh-JA>AMFy*KDzbq<bTWDuHhe{<u+Q~5ay)p9=v`oeOJ_}Wr2LwtL3fjS
zN4MwXZcj{Ez8JVKf)$QQ>ma^(45SjKYN!(;H3aiP)9zbe53NoW*BybWq0WJLD)80|
z*t`=n%tNKZ(1t5<WqZCy(I0ailAJh9p7#+s%sK-5Ky%b&Q5T^X7#oLu_`IK2_ZU6i
zdp|I{0DGLbGD>mb9&D7tTG92&U&0GZafG7{O)KHm+_}GtmP74tFfQinRELtrRtLK_
zWW(to2<(-XRB%&jH0O3)&+WysPFXieE;`OU^N4h%u!~$~Ho+##?DZc53h{D`7E?Vv
z{Jkz~jMDS|=sLeL3vzBRuX1h#^EOwh7?}EMNpOE+x?}n-=D5`h-k?9>7e5aghw%Jm
z+HJ2rDc>g>#@zB4AJLnPWGM}S5>?d$zw=1QzhAidp6sysh!h{T66V--`47j{GmzPq
zD*3a4z;=Zs$7DnNc1t+OpSDWssp#yV93>*)CaCP1$=;nLdI3jor;*U~8j!F5NT_-+
z4t3Hmugbwa$)5ipuq6zKhZySiJP)tFNz#Z<H`!VTC-$|1G8)?aU8EwV_JW)Q$JO%o
z)oY7`)K4dEEtJMWoW`FsW=>aupoeqrOg+5waTZWatxba$p3uY)SSzTU1VI1GFH@RZ
zfkmzwIP(_+V4W3=q{G?3LW0J7I)vMSd{?{P4PJN?aMe&rzu#ps!6Jo>i-p8T?6)RI
zx>~6&<B!ZCe!-%6kfkS&Q$G1Y&_qPo2C^4i{Dm~eUHD6N!?llHJ9_{mOve&h(;M#b
z8!!sYpll;ykf)_EVTLsg<@Q&k*1yx{k^77?PhMb}PrEv4+gWks@D91Bk$#j^&`*1i
z{PQ`Hc)MpmW9N=%FKA-mgltMZ^Q})ByC;XWDL;mu%D(9iPtZ!RntkKkEUG%mZ?HIm
z2R90`VEERv9y%H56|esId+khfK)dK{zO!KN24%b1_;?^WYq3$Yzu88)aINlm)MnJD
zJI>TR2V1P?#@U;!htK`^G%DRmbF8O7!p5}Abs%V7xr$pm-mH(%yIpr+oNFD}o^2GY
zz8T|W<5ctD&`|B|(EuQ>P*JH4IEhn@Tv(pW7iH)20y)=OaBe$bH0mt<co@^IYnB$o
z9xRrC#<v}wENqW3ggS!=bwi_qoZ;-e{A2ReVsj&#4+R+wuTu92wOgOc_oK-BsJj|}
z%U!p29+nyO^sr%;OkrlYyth!et52`^ffnCn>#d09--8hRs^2%NH_HKqui_Rpt8(}q
zYKi#j#hjMowy48%1(h5hujmX1*?kxTzQfuvAd*Ycg7>#Tg_kKW%V!+gA+qw`SEP(i
z&`#J3G+yLk8~R50JKcGvloypmG!GwKs{3lbk;e7%$lSxt0+28weplb?!U@D5`lnQ8
zf^sIR((ERh*+7c{$wc*jl1P8B5%m8I>nc&aS^U&i33e?L0D7e+K|V48khk{GF%nZI
zhPHO!2i&^&kUA{U;{|QJFbT|_Ec%zr0#yOFm1qIwo-^gboOoznUHL!=O&lCG{hI)_
z-0mjd@5VZT8!D_M>qIsf{m8_AM@8B~{CMii3Dx4lqCa`2IfSO!!ip<9SVl2)COh4~
zg;VWfZ_K+`D>yGEyC@~4*+4NUl}iG3GfBA%!?re{OD;%_Qm7j}dhs#*(7dLSsVQ81
z<#PwuZH%=9dew2kSQ~9ExuDI^-TGxA>0lwLxN(AqZ47+*>zf)2#^0rE{i_z~P!8=&
z#C6NIymBG0G9TiS-{*nMjhK-`-uG_uQ*I_W+?5ZNpZjq1Rub}clS8JZY88*nc{?sF
z@EV8IWclhYr$-O8%4ld)-#W)g-JvzQb!mD{n7UHoHFzAhV0;NpP}+`&bf%*;r;so*
zcdTDQUx9lG&UxD11<h~%Dr1oK5Zx%ht61^M6fhpGLqm280PjWMD8a7yyv|XT9@E8G
z-t7Dz=|jSO_uV6dirZ8X7>-6H!URlJ)xr<;RR=9d5>wubB<%Jta?%&xgkB_gsLONS
z>dr)us|Zn>8TF$U#Il^ENnLt<-Buf&gX4pg1*-KlJH9M|hgCwIyRSk^3hIus)g<=@
z`L!$e`YkE!JpePcn@5Y*eKfDK6|<_{acgr}+<Ou9NJBJ|G2NnS(iWl1UzxiZe?*nb
zdVaV)LOduvmq|RI&0tmL8Hp${G1pQ`#MpD8A&v=+2|b!-9&LoqXb}%^!0GoT>Zu4)
zt18y$T@4d!Cu(kEL~}Lzsq)9YquRmfeF+sya!oyLWi(+iIiu3EG2(R#|J8Zap4Z>@
z%2zB-0t4#>T(7+Q+okJ5!wA-K5o`Jix91z#wf)ka*N^p{j%1Kj|CCsG7CuO^s)Ubb
zb(+Wj9q$sn&R;vdQWUxsF4PRFBn#f{;57_fYMAItcZNi<w!P6GgdWc9uJBN`+`BZJ
zU)w?a809K%UkcHN=|1SFk#n~#{>m;4EK}OVeaZw(Nay%W3W=+`c}j<ok4xj)zu~N%
zCIN8O(|(3tYCcU*`h1)EL)*)q>n&&2G+YRm5@(b3=x68>RQ2Alo=NvZ&mG2Y1$_(?
z=B;}h{rHq0wE@lQs4f~YZ<HJ;LjhXYHu#AK<bS#Il6mJddGXBvg>ium`|$C(CN72B
z@QKdL@MxU4uHss+!4X^#$hHE>bTbx5x#icI%^fZ9lb%`4)e82AY?IJ+*|3y64SU&=
zhu69w;i_=Zx<D^SdrCGxbkXML6v$KahhN5n%}NW4lT$Fu$t(CAkl-Qm8q9^?PznDg
zlP%=zSYVkY_hqvgtV%LMk5t+kxa4>*Xjii^kXv5i%{V`1nQ3#_V9v(kB|zX56n~`=
zm(1Jp8kBiqVcNoIa^LNM?;*uZg`aQuvPpLrnhMfsP_RhjTd{JWyF4AjV0$3bG*r!H
z-1v$EOQOkRh8ec?s7o*L9Z;NKM?bs7`7np4apRN~d<&yWIZ5dn$~xsvs<m1U<XgT9
z1_oMEl(Mco2j}ge?q6=z#?93H7E(<2CO~P)8th2U#bVGIpzl1E$?}q|E=FpF>YRL`
z%w(p_*v^{eoW0Wn;pM2&ANfQ1pC7+u@%U)VDk{$w!bgEO43c@T8#m=&fxVSI8iJ@W
zC}!SmRd=TFEQaqP@g$8jX)^dH^jkU?^U8y?S9~O*#muXON-QS^FQi@WUx?Tbf)QwN
z8Vj#{qS571zF&~wkuo)Z&%IZNEWbBviF#K}LC2~_!BM^v5Ykfp5L~xsO-@-Zjbrqj
z!}yc7d>rFTmgYE<)0V_jBOJrMt{CO(xf3(<uYVDx`egCCpI_S3QTQ1=G33o}?c?*)
zC}-@NP2H;GVyucujq0yu0(cR8HwrLK1ivM;A9@pbS0<J#%fRmGsW6w>j#KWiSYKRt
z(dN;1cbGEt1nXnPdw;Y1m8pR}u^Iobx3PN7B;>~Tsqw_;MDaJSZl9!Mi|S*X)H>;u
zeV_1UO0#CD&jfSj!#~nkvl1rilXuc=($K|oJ+&^UMlcGPV-%bB8MxLNnzR+tavvXj
zl(mq1KGB`;ba+mh>P1&h5h}CiF{%wQ{N(pbA1V3CD^<pM_p88``}A>1ii~~NX!}_O
zLLIwi?dbajzC@m*DJ55TJ@6P+wVV~Ge#Ym)7oOoI6UnHM8M&Fx`B0hM(hu~X^LhuW
z|K>pRhy_nLx^TR4;rVO4MlBs%w&EGT%#hN`Ap@yVoXO){8nVkp{No8;GF-Rcu3+zm
zNfA75k^VtkM6KOP=FP;h97?IULeNL2O`t+Rej(Orv{32iV1id9QGQ0_IY~2TY8QXM
zJN-G7qlJsE`1l0AT8CbriM)|gg=4`>*=sVDU$}(rInU3_a`^KF6k`EHD>=K}9KrH4
zeF`2}P4<-SL3A|Z<v>0%w@fd@8@r*|HbI1j+||x35!)C6rqS=Dt15eh$(-)th?~Ad
zbl(D(z?i=|PPN$D;PO45v~{0U$5-!AhOuk$(~#8t6!MQ*k4%*%*2DMX<SfQ14?7p%
zz36Q#QixR$;jAr=7{qf5CyR1cq3?gVlp0OSrJ%AK*xIhUH(--wSteyDTg_pOHFa~H
zkI`XK>x^#W;w}@g`?1RLjMOZeHFHE&T7C0HhC5531&;k$DeE}Xp#{mklOzvi$wh+s
zPuOS*Z{Br2hkJ9pa(w;EHio~IaZMPXz_%U~1M(iHOEv}xcfTXK?o3E;TM3DqFHGAm
zyf1HvkHp{6N-Wv=wLNF=@!tIU*N_J)0p)7P4_5KU$xSX=MLJsyU*(hzW7siq$*aj^
z<jimkIbl7M>M0f97b(eCah;1_*qpr{H;$?^*xrnBFc^Lr=lfxOK1jIfRms%o#8h?I
zlnJkbS1=PLxna>sd7MGcoUoOk`6oslCX|VUQRsribcJ|HmgVgN`blfE$)4(6hwTs;
zU$^wG82*m+k+f#r^+F4^1kzUFQ2XREi;JyNiL*{uz0ILh6Sr9H&A+5R{Mk}?^sWJP
z0!*6yFA-adBh()M*~(oK$H#nZrMw-xZ~rkL{CRIZ9K3P;ie4t=UvCQk`mHxFvzONO
z+_4!JiY`w?PbJ_GuIN&(zE$PW_;;^($%3%(y{KtNWFZz?6H;ak%2e-DMH&2tQ?2P+
zd4B^N{|h~gmV?E;yP52$iLtsNOX!91b6YEFZ_3+-HdqKn<($@J`Ew4Q$$tlk@<oyQ
zuk`B>s|pYr^T87D20^W30QoWkV)jJ)ATRhXEB+POetM_2#@}L55K{Gl8=ukR;Nv<t
z%j^Rh>XWzEMeRR8Qe!BaWgOTXSP+~e`HUL~`p+fFtP*b2{xY~b<4bW1-%a!P@~2dm
z89=3)0R7hIP)^+{qvl2MwQK#sGE$qB9sQp>_UE(tUtcVV5f)190*qo5T-4*QgAnfv
z9OJ^&_*<{Wh$jDuw?->_xyBueJcV&*NkBA5Br<fQ3?3*meS-<la5_&fM)hw1xc?oV
z`#?$&2`ePk$q#1#XCv>=A^5Ksy)Mweby?NELGe$<*Z=sNrDlkB*Bq0R1|H#`7yHj&
z?CL{LeX>gH>;Haze+Nq*@!?&79ox@mTmCuX{;yuR$APEqUV+jQ_)9Lu|G4gdj|Fcj
zRWNm=kd?O*|L*DmEBF9N57tk9O;=cDC9t2iD1uRp{3Eg0?5R0O$ndPZz$Ns@*YK87
zli5*D&7aD;_({0$s&MUK5A>LN=)8h&|IK3#Y5){bGo5GdWhbCQBrgLiz&IG4?@_;2
z=D<PH9B4+o|Ijf&j6a>Y%ui;gkjxZHuY8*aj^ASn6lzZ@H!j=+2hvhfsRO3s?T}}`
zn}Dj#CnYjZD&~B=fHM;40Nk^Yw?d(79%QkHgwjB>_M}n>pqk?VVR=@$k#n=P%=p&g
z<-2Q!AesVIWwRujwGh#lt~>#p4WGn}Xm@&BLXGj93=zWym<dmJReC`HS3ym`CEZ~m
zAWV*dUeS(5vJ~TE+c#3(iTo$e0!b<~!yi&3NP}fAfI-KNm-hC5oss|bpWbyc-$8|4
z2*r*9$p9LvLf1j9UcoHyzw4Ja21SMgyw8gsq?-!69sGP4w+!SjMLGtqt+Aj^d{{p)
z%}@gjeAhq)SPJAhgHwP<_Zo_D&i<0MqCxZdSee1K#lZ8#Nzd}5-x`3o>C+FZL7<;4
zAY@bkCZ}8IeEM3sBrqzOh>$+D1lHKiG+ghbxS?k_sR%Qo*fF@`(hbNe&VU!~GP2k{
z>e8>f1=3pWe$0SwY>~^f`v{1eR!jnf;iS%y_(bW4;>&E+%UMbHx!M6<Vgtaua<GvI
zwnT#1?s9sMgBkv55uz;WTE`zz%GxJ1KLEL-1l;Ct5Y&`)pFeB71$1nTpyRYVpnykh
zns<5&{QzIIAFPd2g;Ec7fn8`G^A!US<|)r12I5<_+rqP8%QwhB%YuP^?a|d6e>470
zq8j3!c~}ACO=<a`<sdJ60+4C#plu;l#sL5k6Of?_I9*uWrTc>r4)g_UK(n*4#&xn$
z3HZc4z$as-@D;)B{IR7q6}&CRHx=@vl7s93s=q}nIhou8ZsaH=x3~-LYEpUVW2KZX
z)I^|R-RY5ax*N2Bv-FN!;9KA+Y{2(#6>VzpCa?vMeRH+7yi{A&coWbB$3Ra@=NNcj
z43zFo^9y-OfSIxnJSdcsV>H4ceN^EaY$CZgHD1vf&~Z{Px*w(jQfU)ePo`JvIus-y
zes=?Nb*R@twd}$;{~7fWKyg^=!(s`x822J*b^$r91WfQYGC^aF$USmtjyDwTKrmUt
z%>D{ND9!?wdu3n8XMeNWAJKY4OdTi{dyf<EJ7}(1gB17rrmztJV&ImqO@bY+IgLJ6
z;_CQKuaJ45A<I_22z&g^6N+yJVn%5tA}>Vrd)xA{HBg}!0--9OS>8m1S%odm$e8>&
z$W`9hFJAKb<91CFl5z0+r@mVGGI_Ghuvc6Ed;WG?gc1nn+4nt64}HOg$D~c(p5Zs0
z(kqh%Vto;$cer3{7i{sv!(!(E)v6Q#{`@V4@&T7GSw~W+-O)hkD6_dvxi0#}>;WCu
z6KUR+3zy0w=4HW3sr~U*_}{FQrT0JF24r@#Z4j;XZM5(mtkC4m6__(N@TPr%B*60{
z9PmDjJ1?$0P=~f086L2KoRY=}_@c`WoL!K#^O3NhXV3FAW^iY<V<q3N@tl`_g8`zs
zb#UWKwX~l?Nm@1cz3YxvxF9Kt98(~SEGqpCJX!?|Z~?1ZJb$whSVqF~2T`?q+RbKH
zQI8!coWVSr04S{$IAoO=71^}WT9kst@D5TF;_*Y;IKLrU+<i&{-Z4N1xB(X0#WgS`
zDu6bPgF8hHFqn;;Lc2dh>ZaC_?0oABFb5sKq!G7<``^Jj8xxT&-rhe3#*fL5!~f9r
zkw!dagPJqKcDY4u*{|x>LZEnEC?GxC6yRkBN#HYbW`n6gePr|_=_U8^_oh_<sPjru
zjYOM)DOliIkb4j5FA3y&Gl_>_Kwtj#8fuM{#sPwTMb}&L2fSmzTS$z~;=KdODqHjg
z;Zn#Fq8a$olwH>URpj6b<OGTnY5Gu&v$Y|ygQZyikVEKI>b{+X?`{e-iz;z3a}XgN
z5;@Ml0tk2cb|;q(?xzE*9Uh?nGnW21jj;IA4w*T1K2gbNQz_Y#efGh2hQ$?-?gfw<
z10KP6IkG@-`z8!v;H`Q*&*H|)XJlS?oEM#)!qnC*PO$O*EX1lvl7R$9a>#zaIp!x&
zn^+(=u7l6C1%7GaRk<SOp<jypJr2Q&uS3Sy*0-=^@h|m(^C#;Tn7ltp5$F)`DQbm3
z150eFQKt_PC+^%gzGf05S@^lKPESatEvn7}r(9Y*{0X`)?W0k%;@2t$(0SL32!kvW
zUwk*%0o!)O)(ViKx$g2WNRIbhdBEMbAee_WIxHVrc{##vqNc<bHvoE;Vte0c+Au0-
z^<eRCqBN~_t?U!l*iU?8><PwCn_K6&SJ4^pdI+r~U>4-HbI6%ul(!)AIuuWMp8e$w
zt!<1{lX#d+k|E9_tY5=2ORDp|a>${zlIY0V@5YMi>{OA*;VdCapr;d?{;BK&=FYA6
zt3kGD2`>z1Tzfyr^2e31-#RzkpX+{bHVQ8KQH_2juZjL+rE4)H%NaUkpoG+U@=k#B
zFW2^7oeNAhO<GrzRtU~tCh^9B9f!(TOaTplJFI6XdAyT)W)oOX7R%(Q`+B`>9o#Qc
z?|UhihJm;$!n36Weu*sD9caM);eOT%<jn!35t~1zyHZE4z*saD3Kf`0ft!v4+zI$~
zrT)%w2#=3)g94?Ql`@cl2B5~TCVNxJ-_|w)OqNGJF6FAtUQ$@AKRX3r>MyM+nS0)5
z;SjKMMQ4>zwi}Xb0+QWhxX4SpKh%7mp}tx9SN5dA8XYAzTSMTx1>fobrxAK%Vd1M|
z9EwkT;>jzjAuX%5#|CJLT+;bwBMK&JN+TpJy|BeFZ!$pRfd7VVoXAOwi%IwHLcEDw
zuHONHi+SnI7T}$AK4tkLbpv!JDz#EjbhD0qYF-3U^0aY?%*hS#{gvGN7F=VyhoI16
zV9fzzqd&rdOQoOx6v34H3d~x$j(f)lU_ZpQ`9odzO9T+{4TGJ(^?-d=UO0%{g_59Y
z3UmNO+Pr3KwF|mzCLnLqQqChGy!`j9OY#`)+dn4QPA*arPFWz0RGDqtNLT!e$Y(~$
zm9LY}8A(m@OAbF$hf=J6Bs)a8vA|>xBJodb0eL^<H2BzxFa=C?CAzIUEcXaG0_Dga
z^|!-!H87B{Wgo<*m3NP-)i2OtrWIsVJht&r6_d_-3jF3wQuM+$h|?&|9@<YJF}svZ
zny-zGpJVRa92M`V)y)<Y<}Wn53C-KKIWb%L@t@2KtR|o!>d-LwQ(6nll2Fz`&a{UM
z!{EX56Ee4M-Wg!;G<eLUs{lJ|ARujdf3NOB>4TLh;dW#SqeaO4QqD*+C5+`w<}`mS
zM5ANI)B|n552MNT6W>D%P!W`Yl}u__doNx{kJ0Q~dTWgk>w9bG)9(}dLp@2H%2A4}
z_zlIKeH0Uf$ZM3a{TAUqRk4FDs#Q^ifzabJd}(``-D>cT6_xsJRav2T#!XepYu`X;
z!L-C^mA~8+yzB16_@_T>q|5Pn2l`wx^Cb>^PiD)c*<?)76a`GOgnac_l|tbf)WhF`
zlQ!b?ofDw#o0+6plgXF@kmt{jA>|8SG~k_o(&AQz83sV(#zK-E1OA;BgTT0Zj5B}=
z05BM4_U5!1P+T+ifFdx1`H?Hw#BFw$`j`>2ckI7=pVua2X?4`CRYFANPDc$hp8gCV
z%LY5M!#aCdh_<h$&4<R+J7j%>7opTm9-v7oL&aPPuqV}^qx*tbF=-|<B9pr&>~2L2
z=Qj*4ET%aHvRMDtdY{YU9OD;RxK^UvSbuytVJ=Pc7z<U(ctjGY^^(zO890m542Kk6
zj~@pe9ev7K_Hlw)dS9<W_osN|H}Mgj!G&-th$BRY$HbzOYZb-V>HE_hXe5M$O&-g@
z0zaP%hwG3XIXu*-h%hAkmZ8y=t;MwmrhyLO&FvsjN?Ai5Vy{3mPR6FlV)+DHI^wrK
zl?d-+PWC4*X%NemvKi#WoVg__aNK)JF2>3(_p+;+uSt<#`vqAyq!D0Fgft&oZ%tn1
z%I2yiklW{hT_2G5lOTELBQrBfBT~?}{{@Z+fgVcIX_J{=mc$=7LAhy)Au>g(n~1zm
zV8tkp!yR_;AFOz7(R-7g@CKb|jw5e!nS@pX*%TtQdf6jExuxQVj{~eo4Bz>P%a(SF
z8Qlex_SckK9wOSf{IwSf-6q}`3C`*BOLrRx23!Z+2mUseD;D2cp5I!3FYK-OsEw9e
zvOhmWZrgl9A;J10x4(e0V14g?p=b!v-IgS`0AUq(qYII)F|L`?e%~&_zL?MeCP!VM
zD4dx)%|M}fx&0hgde>;7;P*jfe7Pnx_&KGQM?>xN#}0k$oVBFwvaScoiVL&AxYMso
zmZ>$QYv38G+fZ~z`-sK5MsB4#)W(+W*N94$B|CEeu`6&&i|-5IpE6$iM@>wT`a`va
zH@R35CP8W*1r+joOeJLr)=z&7fSZrMezdEysH2ZMIob3qXvw6Ywl?G4{)#k>OuMD*
z(wbCRU56D(;c_T@h|4p}06O@Q_H_WpFrytFEF}F-2tf!;+vo1-wp6ap+?&=PGvoE@
zz+KTBD&eK-L@vcwcHG$MZp13S()v}s-Q-7{pvP~sM1gYPf;6z#UVT~%n9M{kyr~!d
z+SIS;s-eS6V~Lhl%s6rc8id&-EYE4Co`a)?=0a1L4T!zZ1I&I%wIN8|Y*v?$P|0ya
zvAs{xD?!g6d=_m_FRwlfv8PH)&6^VVqKdBh{tMph4&cm?h9!CgL<qR5rwc5>SidN(
zEo5)Fe$Cs;Uj8GMQ@j+Hj~2L3>%+u(!{;}fzY=-xN%iq;fkK?xN~sbF8#OxBhg0=b
z-RKS7xr+RK9XkDOa;*mnFQEJcspGTSXGb5mU85FiA6-Iy5u^^(p6A!ceCI;rAr_3i
zu&p>w@$g-f3e`k@smkvd&5hskkjFoC?y}fm7e+T4oodg&FJIpX;2Ht9_?pJ<obIpA
z!1POe`HX<Hn4ni}^Daa0JQlaDv4B5|&5J9qR9DDBBXZ{SyH$l~z9PG>ci<cV-ON0?
znug{)Bu!YKB=XX#=(40R%jJZG7nX4-TTv59O_E$9NOnPEmoN~r<GyuR=e$mZ6g)Fh
z`V=xB`zpPz0?fN?hX_K%ugc6;45Zi{RxKGi;;?4mG0%@Gsy{$gq3(p`FD-^GFwaYR
zI<B^81mvoVQ-Gd{x-CU?ZGw<X%G30{zFS?tD|QycVttDGi3W;ze=-WRyk8wg37++7
z=ia5AUiAZnakC7K+n&LjhlB@;@80~)aU-yShP}c31TbkPl-~O-eakb^3Q`*7P?7=S
z=#wv#$da$V_M!Eq{5|GE_+k5U;|l0ig|WlM)YQtPZ`jEvW~iYrS&_`CRIV?q)gfB@
zH6vkgp0q$+;5vI1mm*RXq-@?vrBbFV)1rl<=RrTvqiy04P5{fS@~YpHRMcTnHvN36
zOGU*P+dI{_5y9v>KiCk@;?Py*ENZgMs@AlQ2Rd+=$^g%tR?Gm;ds9%ft1ra171u10
z^j$Lktm}OM7?WJ+IFJFZEFm%0vZ)HHLpB&z?l-<g#XViW1hp5lR_ce}sVL;{#73;D
zn1t!|V97?9JMt)Ta)$7l$s2}A6Q$=}J|$NYHFSzjE-*)Y=w|6R<91$^wPm0k0D49I
zG^wL>y%*2l)nApJn!I9E4j_@N&sFW_3x(Qk)m5}sKmGOah?yNEi_GtDY_&5+Y<(V`
z0s|y3pA#X-w{HjRm=g4kA3Q~&e~9_+{q}0i!<JP}$FU}o!?rJ{9qZ!B-oSUNARJwA
zrrpyRme9o_)8+DJkC7sYN;<ncN;{8ou6Kh2;nUQ;AHsM^=EFGGEI{tnDGf63YK+iA
zP&0)Z*p>~=w(A5oP0l(f>5u3V|Ab@%W*Jch=aYZ&;=jOKI}3UAU%1f!ispIqtHDGx
zFb*v@5D<Z0C-`3k`U;jNqEx%>c)%xXLC+3PvV;Ma4<D}Og<1m<YyN`$`JkM~$c|y+
zuYh;{+MzE{j~9omY5?@<A>(V9@|Hc`jRJ5Yd_TLMZH+UDBE`8^z06uM&_Ui=2@jR{
z?mw7kHwKWNG0LYaPZCkxthj?Za8#O68W3^6BvORB#Jw`uN<CvRsOTm%@L+E_w!#KX
zoczO3y+h((Ihe$6=W&RmtjtQXL1f{(dM&4iNf{ds`b#W6n)<LE=KfTOVs2ZYcM@VZ
zcZ<>`4M5^>>KqJRrsPbl11Oz^DncK}DD&FmsnUj^GzWYzkK2fNHgs3FY|5nUn?t%(
zk@(Fvs(!^2&a4|0v^vP2*Y_5*VO((gn=4H>(8SNRc=^6O=x?XDzbI*yYqZvud}re>
zq}abj-4SYs0?ysHQMn-|^zL0e3~h+JpoyL}J0vk1{N910;)fd-qxka)i7pJ@16)-s
z{Sq}+2gBcL>-xIU@7hdN%~Df}i*9nWF6|~z{<8kEFYy3Ha-yCriO%ZYl#_RQWv0HK
z(~wq93Kt+gRb|;b7-K3d-rw*9DhfSDkaW8ci{Gs8^15d~^L4Tl-uq7MXq&_-z+XRo
z7H)s#UK;wZg{SI&E<83ZM0}fNn&Bxpb`Cv3Yt+<(k9|j$er8AJc_t<EhLT@jMeLER
z8lT>7mH2wLgGAb%Hhb9<rJR}5t|kifcPy$b)`mb;V@_SP-d|UZQujX$b-yZP3I~U4
z6Jzw(JpM7jY6GNWAq7Soh01vubk5ZU)UCTWU<&a0;eNRmnJQG-m#~+a2>QU%(Vs6|
z;#>7>^Cx+-qSS?34zzUw%!RkWQX{0R<U(YLy1~;T3-{n4Xp;72-O%pG4WZ%p#wl_$
z)iiwBAvms^8`8%zliUKL^od5VFmo3Q<j&;Ppu#0^mo~59F-_Z71o<;<fu$~DKV;Mr
zbkzY+{F>X>G0=3fHL;TJ3A!SQ8KLI-pgmkcGmvZYnw~zMA?+SB-Ur3kC@t9NKwBa}
ze!pE7Ut&T(__Clqp!rU)!uaZ&C*y3!8^#B~4a9I)(DzN?(+OgT5J+C2NBUf_&3Uj~
zr6L9Mxz^5kpP@-vVHVyHKtkI;Hj<sfWpdfttR|;Zm6#gqIXDk6k%9k@y|<34vfJ8+
zH|a(|N=h0OX$hsJq(cM*X$dLm?yilLNJvPdsC0KoDk0t7B@NQ>t=s3k=RN1=d7kI@
zz29Ho81ESV!PtQ8z3#Q<oa>s`b<L+tMx`NYBR2u$SXI@jX&?HTW4rVjZiOUSQ#rT}
z)XTNJ-}H2spEj3-t;Pv<vkDD&JJktks(Hq6L=th$e0-7viBLZD5nj~x2@Wzzl>oK(
zu*t(Jg~kt1zA*#&hR`x2%r1+ws9v%*5@vI}Ge=XnEQNDIg%(qk&qx;@V?yCh1$q$f
z&yyM3Avo>N|9QR^$w@k-7WD~K72)2*6Yb{gWLdV-F|^0nodJ4sV~G3=ae0&RVE|tm
zjv`93QOAaOUz=AK4Oh>~G*ol1=N1-Wv^(>Y;dU<aH1eP6X7)TNA?2e-^MTV%q1mc;
zh<SYbFdd`vm^3Z}SDQMe<uhFEqFNUA^}_*BzZ*gjq6Gj)tcdyI;6;~1gha5D!kXa{
z2*3-V-+IZ((yA{JiTxB+X-jP!nn_*w;w$|Iuyz}uZ=eXt8Yh6osZ!y?;^<&M%nTKZ
zkrD4zTf7j7MbaT?3M@j^?z6bFgk(#AbW1>_TXZAfxU^2h-I>@=Ki1(38gZ^$X=ZtM
z^I?mGD#ioXFC-i;JF~dAM+j~?$2gGMt3L}7;qP$2tgrF7@0>J|-^+aEv0EF)-M5}N
z4>b^T;vl^~B;;si8LnNvhJq+QlUsd^q^EHRrb%k%&PWDoLZY>Qlo07^eAOzj(dY?O
zi$o3kl>}~0DxE%h00!>!5?v7i`yvme&)@No4fGSs{SKdA7Q~7DI=<2;eMQ9hHYx?b
zfaE`al;9~F#{1L+b!rRt(Gy$UVYbiVSKWm`hh;7NgwPA7T{iSiZI?HYoFSnmKBt9P
zQCs~rgqO9hel~aRAz7;#y*)pbY_m*T*u{F?7`hiP8#H}h*(V0z$un-7*~Ts%s^v`g
zmXoI{3b9qvdFi#4uzU8H(%5w>!_9?GJRJ#`!JHKb&$u)ykVyd9sWI^=ME-EipuW1N
z5_9%gZPyp@U0KP1!5F9WL(N)M$IYgyt~^}{U2T}UV?iz^@v+1yBr9L0{b;&v#lZrJ
zNoXUt=6UTahDbewgp`4XuDGkUsLV>@omNL~XG!j<3)ngHWS_i;y5)&y7-}qQ-q$b?
z8!AbC;aOi81776s*OkWk4U??Xp0)leE<>N^%)8C2+k@_~F<WsbG*0sChK5OUY!xoM
zeV?Q1^gi*H?!%5BR-g9!EWeu#ur+?y9!;zw9N|7Po2cKNp%;><H03?5tigS(^qz9c
zPk&i#%L2>@wq}|t{#AN>;M85XQ!)Ft>ToJ|tf=tUX7MX>^uwV43O=Q?V4l*!jZD*a
zwzzJ{)ci*3<{C~HfcM70ei~o1IF01ubatQG+CNCw^eoLiWiX^<B})4}AvpLNg{`D|
z?8yjKg9%bgtKTgBka<d$zx%%|71GpCxU$-<jj}tAd}2VMY2BUD@8rDJ_c+C4@7qnn
z_CHF|SVo-#CgkB9yPU39f88V%p}Nz%Zv1c_Ge8qwwT{M}IQdE|8_mb{OMFf4AJZ3D
zr(ZxQlMY^*mhY<rs72sh`K1S)(ct6O^>B;%_kvY5wP6aEZhplF+Rw=GbSu%?t8?*k
zXpkk?hg6Y8xH+ayRk(qBWAN)jUA}vjvwmNhUCq=El<#fcj2!J_?RWw4V=Y)}9NQ`-
z1Tu80TXQ}#kL%X;zJe8k4FI6Z%<qsph}+nYxMQd+NIEov!`$c9_W=fIWVLxu3lRDr
zEiPJ>a8>js(7etghB`^64OU<78Il(ePu4k>ZbP1z%XE;*%d(o7d4#If#e=p@kXmX!
zRDLw}!<M9FcOY&b13nH>gc8`TSt-B>?`wNAj<9x>>M;4RD(!1oNda3<w}Vw&KT4ZM
zBvS2lw6^==1psq*7^3?l9dx;tK3Ep6Fx7}{m`CfWj-7Ry;M=C&bKuz>1Hej3mFmO~
z09-K859Q?7sKC_!NC(WHet-5qK!8!SB49T&?oKutGxc<kNHQ%Y|0QCWvj#*YD2lB4
zpQ|sK2%l*ztNG<@#s10)pXJvwp5Z~Ts_I)#=D$h`w7tX_g-E4f#(@!J+q`UrlJg4j
zfQ!3xf7ovH;E`#oBtMx9D)(>^Rv=m<pveoJ%>`4@Qtl!RJ3)6(+>9_V2Vo4r`zfA4
z^|AX{@Sx%9H5-ux1m8%Wx*}0WQV-iW$Bm_l`g1xTCRXL-aGU8-KzVEoP!$rZI%lyq
ze6x~pYIzJ=5Ei3_`eVK6kM=8M1t3ymW>^f@h5iXHrsf(y#YD-VEpX9$!CrH|*U8o2
zbuGY3J++xXp(d?+hG~f$lg5M07ml2Av+y#Inm|qqzHf>Nm;P8oYd$S^=n5Qwo|R?B
zQ1(FPFJnGPC@(+wO8ayBHzE|e2g_Yjor~<2x1<*$YpGEgi)CkC9<Js;C;`$h?he71
zgCAw>EUJsg<Vn9Oa`du&vS!JmQJ$|=-+u0OVie>P%1Q1ts$ayBw4x#D`TE`|lV#sj
zRuGK)9xkooLCJuk@I!m&kF5qVZd0-oAFz9;m8s?thXSK@V0fYG?n${El694u>wkFx
zkd&<66O?P{(7LHW1S?XGT@j;tgw@HTvdU<v(4oi=2;^|psrS}urFQWta#9_xpl=)e
zUL?1!+I{15@xV+GT&hOm$=3@_eNbIvOS&o%=28K36$^lvc20Kdu}j5BlKprklKLW7
zgys8rRDsqhzM=2}?mor*CjsrjcrlO%fz?ZgBD2(<FD$a?RS~J&=Mx{7GYEwaEe%Un
zAMFlSn-C{8zrME*btK4329@QEpr(&n=}XBAPGbF{2+mTIwKoU2AM{A0>LW=U(Vi<V
zp7He@eCX9cqF&%PB;ND_6?*7yk-AliK6&f9^{;y9zCffhPhQx?Co3(5%$!r#Ouv$+
zDd$CaprTt^#jy0-tEgVy`ef->{DAhw;?;vN>vNlB&@#0uS#A?MQ<9<}T$lwdBbhzb
z^e{&SQnRN<vcz?-8r2tv$<qHXF5uL&L_1x1KgU4#-HgqLeMX$enbCyp05N%NTvT<)
zu#c@}j~7;z^xSI_=xVH>Nd_nA)hEVps-ARBD&$nXZh^ckA%{pya3;B&-<z>L<swE)
ztSNp#j4}>{xt2iV@u@0DcR$?SzDU=jGf-=}NoZd|#$G?E%BNX7>uI`}CS=74>5rh>
z`^0OSM_sLys%ASo0Mal7h1tt24$#R<(0c!>wlYx69!dKA;puATGMQI-E0NZy=rxL`
zCz7Oh>rZ}p8}D>w^Vls)gZIo@wfN>_wQ?Fb)P0`s57fHJL2MUabBHx<HPxy8!kDcP
z9%uE9sbg+S-{0M>AdJp&dA>(_W(s{Ah^LH!w$9UIL_$qTsDjOy4fQ^8<T&WH83OZ1
zrBa&uZ3AWdw>`@lejhQ%ulHm*wQ#YqaT32w4CU?OdDG2vJEwn^l$d@LG#rTS^>qf1
zK$G2y%>&`I)&vB4=~Y@f_DG+r7=|Xh^#xaG(K9bQvDmswpjP|jd}@_6%Q*X5&1;5n
zI?v|3NC~YT?~%OCgwXF@t?J8Sx9#P8zCeRV!rEPWramNS(QpxWj%S3jy=3d+Abyb$
z`g}e?ZG$eAidY+c0CZgMoHAW}<LT{|pk3Z7*#rdIaHq96!u^6+w`ti4r*wdaj~S_+
zxI&Fl;K7RE-mqd}K3csuNM;#i1KbP8L7`7zX$SD-a1Ja=)c&l7s!)?m%NhjRybc!W
zLRJkc_vNf4$$+84i;Nu{Ll#4JPk)c<&X@MjJ_lxhzNe#3Xt)jGv4%Gd%P{l00}Yll
z23HNkL~UL~=&`V(pB7?<W%N|xPHIcP2qUnr-IXMtO4K@_7%vOO05)r_2Z>e2xi;om
z&{V-hOzuqVfpRmPu15#=_AqeP$MPBzYMk;tKF9Lq=I-XdNgz@rQPep%VRPehj-<SJ
z$MMGNyP))$Uap;a)H3oi$)o*Li*kjSst?q{Ti?e?gw5o_0>kRq^de3Is{*$<;#k0_
zkRqaV&mqi08-4j)eV3pI*DnGO<*l&_M{Sz14eJJUFB4Rt?sE;WF7iv#x_ZUY<PB4D
z4KIqcJp>)@u*2uzJWG0l^ZaH^t?lyZs9OJOhu3$06A&aK%3^D*6vr%AUB2_J#D_M9
zU3Q*U>N1p&8&)gC6#Y8ir}mu#<$$U_0lea#74S3fZe31k9}|1V^F>}|L-aaH_Y8#{
zZ}j1e+%=7R*;gB?_&d>mU@bk--`#Pbn~<UoWA<NMgTS{uU&E4S5U98(c(_hKKzD38
z?rJ_UT+^BK7U+54$*}mce`Ea`uk`_;aJz-Zg1hQHgFUk#&8gT;L@zQDdXhX`3iqU_
zPtxW{h32y@v#M?Q>esw?fwQ6Tid%&9b(L#Z)NXV*h&E-04^i`Y=#zldPSYjEyBhb=
z05~xgCbYGk(1x<$&!;~9aA1`qC2nVtyF2wAm3ut8_v5i(Qy~yzabJ#BW_ry^@Zmn8
zWym6jdseauRBp76%4bb~v|b-%Wf%jbDhot5ZX~tS?Nn4$AH;nOtG@3)C#;1H?YYs$
zW$Gzt*fYXu1m$Q_C0}KJ|D;&KM&w7kZVxszcZCgMhM02$ndOZrYNXx1bhg1qvr5L5
zSC?nqtUtuB(k2;S3s?B4;TeA5PTr#XEaDM`<9`h&LP2MKHu!a%&}$rU3z8c`w`4wy
zQY2O)o)>DVgd>?ME+z=a)1Ke)hU3*RmG+m9H3|WJAk$`konSwasr`X_6R?#$8igrt
zS9Xqm6v&@CO-}}H8n&29jL7Q9hD2ig$jBn7PK;sSfb2irMSL;{-0V<_%uK|X+Ce*#
z2pM)t?<hW@cOG<2%my{K40!=%ll*Zh<Fv$NIGk8_%Ye5HrLEm<!70yyEZp9_jN@A0
zGh8Z8HR_+sHDPCn*LW?v&&av75}G^6w{Nl9*R9|&)(Fg+jv$2f8mTVcIVoOA;!1fJ
z5GI-t809m5CD?rBCBrxA=nN;pbK`v5TYs#Cyg?W+htPZ(qM|Z}D@<GJ)Y-`K7C;d5
zGY9h4{7z~xL}gFjiW*_K&^PrsWMomVtk+`yJPj4lX(<178m5y0)1#B)!vw0KW{&w*
z$~e7gNjMyq(C{2MagyNRA=^FH;Y#J=F17-UjrWE^lH}_5vWilvD(;_of8^}{_*S^}
zNE_lUPF5>@|7+gjDFza%m&HedQECGfK8;=jE-G7UB0Joq_dZWB<ctxI3U9_DC&>f*
ztc_htWYOFk`o740NKLo_a8&M{6Ez{drE5j#NORQNizsS$jwP>`^-CXpr8;R@CqU-v
z2wF;uxf@`iLwHPS;Qt+UPvvID&6(y7JPTYPRAQ-l?cIX2so~1_VW8_lT*j=lEAW}Z
zeWWgRfjjwLT>>E?NEXJpxDj1xAV3>&s_)MkXmE=$g>J@x>rJ>s3zfQKCbs?*YPur6
zj`B$9vj`*faZ1bAry0|GzQz5^dl$E-f^~Ekr9FzK_dV(AcuFu?L;Z5p*6H4N;`Pxs
zJwB!LjOFZ$Tcxv+;2ohzPbd$cg-s*6_^<Ey8(xng!!%{=jUdT{pSYi#ej9n7XKN-O
z@!%8r+E)_aG6iuI;NNFx>XttoWmgMo%~+0TZjq$MJyV0H8+A8<jdP${fWx5DEoPFd
zwNSHZQ|rmcW@T<!sjqQPDfY<>Q5Wv`wft<m1dX@i_8K+P9U;p^TY?+vBMOePRN1<(
z7UXu5x=TYiM*xp}kwWyYq3}i2?)6<E?}Kj<9Z~)1o`$S34{jiTo##zS*}!fG4L4GP
z**Kh&-EbH2r#lMinL^{TzvAmZcGSS?XD_g$MATz+A~=TV8RY)hQVd#j$Rf-9!}f3&
zjR_8-><8Rxwh29WU+kxq7=AU`D3(-?+=-xospiSR^Vb3MpWm&Efo%!duDO8@DM`Ky
zA=rGU^hl^aKtPsb30%LGL{O8}$Dgh~L@gW*Vu-Z?K_Ajp%Gx4%^vm3BjpV>><HVu%
zQd^}!J?4P#^UU)#rMPdIK|=(ywP8mQ_)Kp-6Cg%xm6Iy>U!s@)#L#`MXngK;14d<l
zV;9ItY`N2poxz8AfL3mG*XPXm)T8)&2=k(f;={6u{d62|#aTe6ygwOBP&kM<w8%B(
zF{pZQ<v3XnSt;w2lA5Qg>j=#w=peuGufiYpDdl0EVNkeh9p@VX6JNqNfpgkJ-jR+(
z_&JnphVcLT{QdWL(?pShIp4F&6NUWOw*H<Vn~#IMV8WbN`UaYB2iy*G9>jm9Bu2tp
z#YUBkq}0K~h<ZNh1&!vdquJ<KxWadVXzmiulwUy;e#Y$W@W~OcH(<1E0!CxOE8Ul$
z-dg~wjjC<+Y85K(p#e6&b3J1hAS#qMB4cLB0byu6KDZFJtA%s^UsAPuRWK$2v&*Ro
z<L)S6J*s~*up>h(Wdom&acH1tRW`xDJ)za>AS5}VslNYf8qYsLt=?=5umC(iK_XKx
zhN-jskmu@tYKx;;n{EDY|1SdsXB$O51=PjAyvY99cYDJjin!$HY0Tes;{F62iGkz^
z#nMa^|Ms^2rk(T8{Ry6@B(yMe7M`+yL%RI$&GPTG^A3o#sc}2@@0g2!MtgUl13R?_
zuUzQw$g_X8!bT93!16{`>VNnFsv&F6DY#bZUu*iICG}Oq@Bp@^+(0HV4zB5%3bNb3
zy9piGLD_Hb`qQ6(tJS!SYy_-s1Ay=svvkthfmU)kP+nYKQLZkTfMPp{0?;A2o3Xf{
zJJbtEYkcosRPpa5_ulJZ3ANI*o&WX^R);kLeTfCIV60RSI$2D<Yy<r@Gl=Hx2|x&l
z=Q|GkZ1v^Xs$OKzY_H+yg4@T>q?z+JKr!ZY2b@1kfFG`|X{cTPz8$JxgLTw_Ka=@=
zN&b&#e_v`A3=k*>SJL(f;Y$}XK^L2B0?}h<K-eA!L{USC$2tZ4a?63^wBzYT%H=Fz
zgIPizz3P`|po)YhrEL6ve-p50^^irU`pyKnJ&-+KgdD@cU8uST19q?OK>6348W3Je
zfmuVfKAW59Z~H6F8rpi+L)o%mJkFn;@L#Wc`ZWSKI19j0twHq(9Iu@VBC`PI)1?5!
z8J-4VtVu`bcl7gs2iiy#x1PEF^>cM2_uZ~P`)&npXAu6y<f;X%!z}>I?vu8?pTGp*
z(tgC}8PT>M0}<_6)#JMwabp1Zu~h%!4)EelK&dEzWX`|#84xp@gR8uPTj_edV(Ln0
zzCD%qxotD&^Xnc-f-$5ZFHoL=4%nl~azKS21Ml2Yj<xJ*lA=7=Oyy12nMVOXaB7(n
z&lS^WF}j$B>GtNQ4~ZSiDehKUMzN|EVDo%D@e+E&8C?Le_W@SQmGt6`j33R3Cg7wr
zQCMHPes4J6lY{iY;nlP5V9jbD0$spQ0O5^;0q~TOiolCYDG1DQs5o|;GM^+<2?mz7
z!F9lZ)-iPe`?nbAdnpRS<E2s$)BC@SuOc|~JD@o~QK;Vxaic=DZ;<o$0#hWe3D9!N
zv8w~Uy-{Fz@)GontHOaRKQvln3{WXG54!-L+!`{r-UNob#X#mXKo1~j!0Or!I|kie
zdQ@{Qi=x0h0PkA{m_tI6hkIY?ex_cXx8qW$fD@o<Mg;^GfVo*ALGCJ&CRN{w1rXG{
zI)KSXjr{7S!|1eG(8VFMdn?55tE0-t$c^B9wotmvBDq5Y2%3RiF4@u}hboYy#{foP
z3bD-dZj05npFv8OI^ei0S=ac+neht1oMPgYQ%)0QG7A7{z!ri-+U@Vh!_cULZCZXv
z_Voi2d4N8|gLTpzuVjp0#`XXShitk7(+at^S4n?g<a#-<$n{zkObLIz$oFV`W`X1;
zzXb5+zD~O=U_*K{h+I?#xYf^MVa&epF~FL^h1`y?iY5!Z4s$xV)RP22JlOo08F9UD
z3}V!&UE#i?QU5i0X7Otjck4}+QJ__sYHOm(_yR^NO+W-d8;+zFuQe3yyBm&(X&|#H
z#O-~5W<*{9JPf{WfK}B4))?L~MR8~%0WJ^&PY2UW28LLw14n%v_aSZhqn2QX$nCIO
z4>~S^R%2q@>!>DCqXCc)ECDwB#Q-p|a5!lFWY+=9aJm@ic7pZriQpLf!*Jn~#pp_J
zFk}Or3pEt;Uabv+8;lW64TkN25EfqDmVA%)5eu@@%~@f<Rhf`w&U*pgvb$8|M*VL~
zCqi`>WLByIJwnCbBG133;It!LUSOFlX5$1_a?ILmKPizE8vnZq4r&G9-j6{(d!J3f
zkKA|^78J;vdjbJ(+<J{Z9b5>Qtz73r?IVLJs}9EO+)2f|i0efp>NjB@&dBj*X~e~k
zhY69K>p<Lchaci{a^HFbdwm~xnyYJpP8?kiBSp*WECCQO&^m9XMgRk#S_osLA!*s0
z16*%?5ahk8Ex~)qgs?+N&qlKr1Rgh#S9V-Dh&_aIOF8IwpdLZCl~dQveE=)I8qzE}
z0UF0RXvwNnJ=;blZmAU31$Jm+_tm%gT7S<i2RKlajC50D{F^9=gp|7tnOFps@uGS`
zglg4%2t;hom=eQeeNh?RnrtaVecAkK2I6WGq3$2`^Qj>hxdp>_z@(#K{jq$K(>E|&
zr91i2`(j_8PJ>D_$OY%ww^kTr#XgY_Osx}}>JQGL(3(?FU#<ThH^;Bl2etz~5T>~&
z2vbtKPU>_m$kZoUkNK|yxzi&!UXiH_X#P4JRBru;4`vH}u-)rwyT9k!e|^P`*!Kk*
z?uJKYW%sqoZw9tFQP2AxvJ<Tpu2*z%b~-R%odK;IHpeYm&VH7l`rEKtKp3B><bS*b
zuwTzjFrqWb6R5TrKijqNQxD_Y1`C4T_S1RuWQdCyg>DrZnjw$s1BOMuf+BI5Zg{0+
z9^-rv$?iO4!dh*%9`yUpy`ly?_v)ibtny#)+_$*WYOrj8KjWS%-0u^Q+Q$(-?*$X4
z&>)qDB{<!xTNbXcU8q0P^oE?xNZ>lCTwE&OwD#+991tVTLZVtg4scRHngtCR>q;8Q
zCu2%3;9dt=|3?C7)IM&Vxy~Q91EO6Su=Zz=DC6U1g0&>Gt6h`P`VR@+8xo9qk2J{s
zUS$1ie(y!W&_mlv^S@DyM<+=Vw+aUg@gT5ib-AI=8!}PJ1vu@b`7d0pLAYq0u*@e`
z*Yhnzb^%)P@~XQ?X=BjAhsGh9Yf4+7{P3pZ1}?UVM6gaG7-Y)DeK@Ex0b1Lt_n2RS
zE7$)~%&*>st_*hx^9+738UD4<I-o@Uc6}C_VqA!P4643Spmj0@b5cSfNZB&?N(t1Q
zP5iSEs|FOHpHJjzeT-^=$=cUnAs7@TDh^8s*bcZ-mH_AcI{df7Cm=sqfV~2V&hBgD
zuK+ut$CS|e*8q20Cfcu({q+E*g9Ik!Keb^MCP=t!>epHv@12-(JAekA@qWh9zi!+=
zzCNV{#p{`F@@MhCDWQBsijko%dr;^&pp6fKJ~h7<V194JZO!`QK=nvBkPkGB0(%st
z{VFJsgQUt+v-n0%jJyJ9naW*n<oE+Kq>P6kL~{?T8>;@YCfNiW5^VE;U@9+{w-_)m
z<b@TQLybpU3*?@vOX*z1V0OrM&9)E}@GL>xuTXi+*9ua6e|0LPA>D+Oh=9DA*Ue>w
z3B&V;tndLrK_m&pv<VMIyx%|51$1M)$f;M={68P;zddJPT`_76q?RG(j5=V^7WkkR
zgFZgZ3ZpO)qKcBz79~54fAa!m#{|S@=`p0)0d`kKAmQ`^^qL&1TAEdNz|@RU`iCqF
zG_)<_)Wa~0EJ*D$0XR$6pc)LH8~*8oas!Su3q-Ah^x<lB3|z<}xyQ*s>ZX5T1FrKC
zFMx>T4EL@O7+UfZ6vx4Ln$C}Gg3o~fL3|odfvsZFjwW8NNAC-6A0@J6OIVB%z;DY1
zF<Gc7K}oZ_2}vut$6gSeE#33hxPv$ckEuy^9-yVhAo%%1oM**U_Pt|Je2jup%MjGu
zZ>B(H2W511CKQ}KEI(P6hg%@`BpYCaLWzt%;RXH#(CJzL*QY!WmAn68aU-F{E%>f#
z@OO&~WA=e+fXaQKe>I03mp0i~c&1$p5yOS}!8?{85{$3!PJlyT0*1!rLhY>bFX&_e
z$53B8Ct(_*!GqkdTMIEm#sTtg2wf8DASWU57l^YN;JV9RZwq9m<>oX6%ToM!u0Vqj
zglikCss<HTLmyJTPI0>c6HiJ%9d~Z}jU1a9G_UT?gN|?R<WiB86tp7KbPe$97!WFh
zIHUxjSZ@!AO@{TLF7wm=xkXT>VZ^~H=OAO~c6)URahI=Df_PW@w!0uKeXjL}Y5`E3
z+?o-yK->|X<i(-jFWtjJmYaM)*U@r^>u?CK6OBl{F~#e2O&Y~_erK+kd2LPnw{V?y
z1zmBjhN@(L>2iNx@qfJ1D+fP6o4G0n`raP3=bnx=z-Ztz%is_*^6)0^4h>85Jk>lG
z>Pz*MM~gvl>>RJns>lmaa6gJ^<8h_h@U3i=Kz%#iE=x`H12TFU0~xpd@lu{{%0}w~
zI*IKJNL3Ukji)qfr^iXazs<iJ2%uTF+emDw#RT^Nnk|`op}^IBAMX(KeJ22;R6m9b
z&1DdO0BZZ|+in9*Y9UB=Jh>cAZ))|D@s#F3i^Bk#S=l;HYeu^TiH4Q;Yy&6^ns4I5
zBgpwUb(~%;0q~FX<jeXz=ufH@^{UH=|ECT23fgd=t5wzidc#eN!*-$DMbpw2@O9{W
zGK{ySUxvd|2FC2Dsw0Wt=YeP^5)BdfkM17>3^3r%!vMS5P#+sWkHzAtRG)cd!3zpz
zRhA9M7-O!ZZvsTy;KoG&U4}EsK?riK97xs=A{~zj6s!Y}fW}RJhKyZLBz6{0bOCbD
zJ5lVRej(objQKVFQs5orM=Y@U@6JG{eO+k|3{@FkelU4OOWAcTg^Jfk@6PooS~WtI
zRV=yngHPEh#o#dsPhKrg5oJ3#-26Qo2}9XP*fH<T-((|NWcb%0EK&nRr;TW6E6<;Z
zC%%FN%&z!o<xe--A8EMv5nLNggM`Y*0Om1}cx!?eWyk(4fd1DzU6p~}piEhM?Y}0+
z-@iVPL$36D2+-9Z;MSQNa|LwEH;8KyFi@1bO#m<go2TqgcyF(P3rW26D#PDk0EUQ0
zNaqe>N(FQ$izR=dVSElrIJt+Osz6S@#?kVeGE4Q3f6tM>UO^Tq`>)URA3rE~aTN&_
z81xaIfm2|<`~sjPC6N7h8`yD)M}+?o`Ro;jyc2{sblwyF-GBZc|LE!xw7N@#AAkH_
z!2Z6*zrS*z2o<?KZqGjbb@l!yYl7;=2dW2mS2U)-cWV8QTk0YjI=>~;sxSZY{QlY6
z`073eS6^LS-;Uq!Z+`;V{@WUeT-}B$<y_}j&EG=rzuxiw?CZ3*P{I24S=xU<1^=~R
z{j;zC_wkFog~KBGmPyV41?2Ru(*N{1%;3kn%!h1((&$Iro^uGy2GU$Y6X3M2r$zqv
zBX7Ji*_NNpOIp8bxK5}X^OcT<GZKTe;~iEMqxjc&zNi<k&<%qa(%*2<s?Zpwh)KvW
zCUnS~;R-7!3}2mHUTU=qoo#XbC>g2isp`4N9od*JS*aU#FL3o7cK&~5F@uHa3t>Eh
z0E;r5PgcT%rPy^a&H+s?Su||*&VM{9fBzyo3o10%wh(}2VIj_v60l>4a}ly~i2px2
zTS#~$+{gv)QyXn9a+$O+a)Q*2|M5HX1yPRISGeAf?QeqgfBuy0UeND*H6(ld4}brE
z|Nj53|Nrj4|KFcK9j8<X1fByp$PIw=)m>qq`T_idyLJxS8a@E*T+$+YJ^v9x*x4{9
zf$IiM8wo%pifq@sUH_TlsZnEBUZMTxQ1f4zdLaZ9CJSrk;cuRAgKM}A=oSS)?Epc5
z4b6eUgP!q#`rUk;ITsf|8R7K-9tetjaFyN3d49npc+mTZjr!LJm0+QRbYS*_8Mbo*
ziQa7>1S0~{@!|j^NY4B@*&pvBQ*jou={N>A^)a`;v+1MyLeP6F`&p)x+9*k8c@uXK
z(2180t%oTV5<!j=#07c<Sz;gj*WfeXmruNJrX4JkLhMV9`gFTTRH=u=9N9l+<*l}z
z3`7eUudN0^WB@QU`LwC-8tl$i^MkWM*v85qn;Q-(HcvP=|E72pkwP}4hd~$2Ex&u{
ztvQnrLoH78<)_aod4|z<6>mVgUh{@Z-_|N{7ZB(I7xLC-jS6sy6b^#ADTng>d!T=^
zHSFcO0J!+r-e7+3Is4@!EFREYJ@(2jq^_R;_n?l*TV)r}2uF{jni*`6y{%m)CZ?CY
zA=D`Jc&8JX<B&t6Y}mW+4>%m4%*2>@00ibpbQ@^Y=s=&~YWGVd(983ZJM@a`19J_F
znk&Hj#dv<&|F#MAH3^_=JPZfb-|mk^pTZCXxWXJD-CKo(@#`p1UV;&VytQCP{Dy<f
zD7X+bMS3032}ceM1I6;aTmW5D6pVySqt0G}!O_9cP|%}Rl{rFrbaepewby<0er+=e
z@RK?4fDWp$qnSP$vVvxs%G|Sj;w(}1VfW^iszfJTu>mkI$5k6m^YGI4uxfGa3Xdd9
zFx%4|Y~L~x$VNjS;HN)-+HB~xLNiN=A=chK;Z<SYjocCsTHV(7{I~xmhsL6|$-|^q
zjsTQNa0#ASwG>c%$ULY2*=_ArvTbN~BVC>62r%&Lhte+#yf2sq!QI#DQ-u3!nv~Da
zeCN8H3A$U!&9c*wJqIq=JivxZY)18iPn=F>0Khf7f`?rXhG}C%^XHowv)|z_1BQ1)
zDGAYyM#FKlZ0rKpZgY=D7r^6Y7Unz?8or$UE&!S=XN3R+41xyKvuW$u>_V_#!*T1q
zCV&>;fnqH=V!PdK5O}VU83QAbt2T9J5!lyV(DUQhQ+j1mRBB_#zBq=sP0COO;PYnC
zF^GA*@$6U4o`!;y`;)?LH9h57wj7va-QY^9T$*+Ifv1k~9XO-DGfYQd#)m-d+0^z8
zZ~?4Ll@y22nZN|&#82yX6dTcvz{-d)1*lPXBE8QudjvkDbi%nZ1|ZY2?6OS#8;5Cg
zddetbcjLk*fsIkKBG|n<06X@M_vc^tB8yr2=x*_@9%2q1oIWtv++BRSO{Mr;&PxFm
zow3LJY+~kgvHTzmU_t!Qfb+R{!w8vRCht_wpBkx?qGTSB-^(+=5LX917N&wB(Z8wH
zKGL!qBa5KJ3V=;ooCJBChlEJ`9(N_=vz7~JLaJUC6o%XvK#}s}4BDa2VbD9UN^WnW
zq%G~UQ@0l--yQ4m;%2AzDa4^cC))1V0;kHMk6i5#f}(c9tv#BL@J8oe>3>NiATpf&
zQY<P(;TWR@V-V&#5<<a#e$w$wJFVh$J*Ngyd<=Pw(OH1`dvhpM$3B?HuU1}7fR_wG
z!x9sD=0#rF@XKT|m@NQ-uAwt)^na8Uhm@7Q3m~+zT8u*%THkzB+#eT<cZJPZ3EEqM
z*?*GOmSGwnAInu#Fx~3{JM~hDv~K(@^=zYW<G`_@WL=peM-&!XDerXLa#u)cA^+2Y
zasZ~hUgI7GQ9yf+`K=rDyt#!<*YRPRsHPouwQcErdq-dg6M(kd3~H`IgjBZMmCN}{
zf{`A+4*OH&k2nYTF!~Obx4?2614e*%@i!=noL(2;8{z0XMnvrNcr0@*sd#SEu`GD8
zAVs`?^s&vZ6@nME4!|NTK;*TtWwe%?Py@PWUP>@{heHF~qGrci0D)2zfneAHeBZ24
z4idqwatp&Ddvv~_vcwmGMeAPT3F_Bd5=CPn+oB(xXT>cP02n4AqL3r8z2OVy)#<7W
zE=}BmllzKagYM;N+`iQ)UA3@{oM3W0M1y%JYX#QK$#-Y|S7_$$(me9g$T26t3hJto
z7{-kq40=hDS07lOd%h{BZbW5TU&Mzo2_YJfsh?DM(Id?sALsA=Wv?EK&KH9b<^dW3
z$a4F|EF45=cq>r1y#Aau&b-iP(Ck~wZJ2l$WExoG*Z^STs?Ae~+gra#=I@}cdOBQD
zKcoaZ1|hU2unzK?JO^3A$=p>PrNwU+^kwc#EMRp$VuL)ylD9pwLrHQ3oA;ccImNu7
zSuWPM_q3Xb<8~*KfdV@#SHkyQt(&mon@d1XyEU^6Kyu&~s;;@nE4Qj|tMvF+e9*fM
zEsvnDN9-PEXDMjOJGyMU&j{>&38YgGhlDmdSy0m_7|grhoC6c&u+@u<)|_*9ly%R6
zBK7AAEp964hE3s50@;I_i}G|g<Ve`r1GTfLUwXF-OwZAlf_b&{jZ4>rk3~CAcigA<
zDUZy@u3a5@U#%>$4R_$y89s<wVF7i3)wggN%w%$+h@!9R4l!#no^>60v??qlTxPS+
z8ct(F-QTnt@~V92JV{H+6!KMVha8#aK#a}__T|Uc_dz>MnYq|<zC`fU?}oE=-3$F$
z^RQupkDk2{>k#_ULJ%9z7H{Or#ZkyTC5Y42PV|V}6WI8%a$QOZhPZQgT^2s4*W#=T
z3?XatPl9-A({qkK1{Y<q`{AyC_Au;crj?uT0F=AQ!%OYEkT!_jb#HM8L^hYxc-uBo
z6f-w4I$+<8E!tLEOX6*)k(+TeuV~#l=&;H~UI)3M@X8jpF?}3+{>q^U)M;leal@^F
z;f#k55j@9uM?(c(4_=QX1G`AH)VgY!>ZJQ-{W;f|%0DJG2e`;yhO{d0$?gd@2`8R7
z4!8-MF)Dm+>RGx|UKAlMltR8{4H~ubHPG(g*^4OGvja|@4)I_z<v{)w(M&T^%9j$!
z=05a0$@>a$ty9XFZ27c#fhg-YTl`hkYyNi2K%}(WzT52$(Cw2mK3F6u4M*M<uekU9
zOa`XiwJa^d$sNs%Q9^dXnKG`^YZHTcsj3*T6SaV{)ehSlm&X(TEEgAo8$thT$&&?0
zeMXOWOFfv7J#n8*ZlUy})&$gJpHi>dlXQCpV+5&wNtT5(^t_W5%JSo2LX8oUM9ZP%
zk0k&$(jsyDqK(1SmcZ~12dnWCDqWE5f0D}cL-XmGGh|?2ec=AAEUm5kLZ^L^<ry7d
ziOK3ux~;2We)dzSK3GqoPWl|c8J#!0u{1dlr$v#TU94E;l>BXESo4MnJCIQrB*xTB
zm|=1T`85#g23-p#>CLwT<d>CZupXq43@?JDRu;v#V%jYShmDR+|4GD`Z=gK6!yezl
zqX9C~`NNmwPw|F1K%@?NoVEqyG9V7n2lEr&X%$Jx`<dn?J#uJLZie41B$x9Ft0e2g
zlAAK*h`STq04fC6lUJ2t9*}L$QE}FQa?8~8*ao>DcU*BhEs+tQH=GdHm7UkCSM-|e
zN!Oc$<Ah>IQ2AK7vt)#V9ls$|8NrS{Cls-F?F-rx4)W3CB^<A`$(3Z@7BUOEkE?{n
zEmH7hy^f>f*=xciz}!W!A+_+y4ccnAk(y~jP>>N58y(BZUg5rOya0ok9im91NeXG@
zHLl~tQMEh=mYEs$v*)1pm@_Et&qN=e0_7fZlV?1Qn79if{^Jk5ukImS$=U<keQQ-0
z<UNEDw_#Z~q%Y8w=x6+`ebpn5zf{nBeK=P8N!6s;f9Hq*=gsl8w#i88;L1t4N-Zy|
zDNp*WC)w}oWlvo?wWjAIsk`^XLlzKEKXDQ1u*C_Mn7vDj2*)Lnno4#9yyUI*xE#n@
zr$#S~SB77_^c<w<VYE1}=BrDEZ`-~kkVV|Si-<q~yPi%>2Oi{6UArDbMOVPPSvt@J
z^5Vk$CMajVEzq}zXhALd(U&=S+AhN~t3$hUp6^UGx9=7h72~cU?%RMJS5bEeTyqUT
z+9o?E@w3?*F(5<tB|eKW;q5?UnSIkhGzH9M%!o_&u{t<Akm^OPg5mEt!yF{5!lV-`
zrN`vmO+jt#fV2ae6YFz%ZdcT63#*xtKkpd3;@F0GkV)~0>Xv}UWsTlBsCV}5VoN!|
z)Hu{*<>9weF|t%u&3##d_qb%v95+mW=y$x`YzTFpZU!DjKO^%;zy>Kz&x2spG&8(5
z#T(e8MsEAu^iUL5N^ES=U!O@HuM3<AGKfZ69!zO`bP#P>_yrva1x=*}YBeKAU%iU>
zmgRrzW4y9$Qh6tU-)GmD0^iH<(%ZeG7V4$XQvAHS>YU~cw6&ldZ2~lbP0!q^Q`U8%
zijl*r1@VHTCBYtb4$r!S?RoAghk=x6A^ZC0JumOFC(ankE+uTz45q-tZXlf+LbRfq
zU3e+4snF_#!8>0W_B#3!9Iewds^mSEyeG2WZmQmEtG$CN5^2K71j;aJqi7>nyj+gn
zc0EirRb15m-JaR|YRAe_?s!HZINRW0vZ&jCxsLB&BZ#s=hV{V%B&q!`Q&z~0J)T?k
zLN0+WzP?YH>;YS<2osOQqP4$MosUH{3L_fHkTLKJLkl_lm>+|x9@wK^i!;o}Kf2?6
zH;Q`L8kggy&Mo+h9qb}hA~tA9NgM+b4XuAUmXAHQSd-{2l2+%JT2P++z>Vy9Exw?7
zor(Cqo<3qP$CTz10ok4_KLu_eU1J)`O%C+9=v8Wo+X1T)*LQT?H<t*HK~*s$fgW?W
zDo_@&;v!6R4Km*&mZi4HoLA6L2&UT3;*1!g4LFTCi<$aHgDkr0D`F0!=FWynvX926
zfVcMdIib-wvZ7QYG#_M$-WM*1B5=G9))(8xsb`RLbk)x@b6fY=3<z3?9E$OXP-)e5
zvKFw5RQr^yeICm3(;-g2N(s#$(N60-vGz283uB#<#9hG=<p|{wk@M(g9pF)D<%!qn
zKWL8Yt*QabN)b09oNr@#yWsxxVa1GLRZ|qI&mbOND(%ONOaD<sgw*t-NwLz9ven{+
z8k!r66L?3fHre?hut*2#cUf$qcNz^N_&#K|R`1zwY|+7+i|_6R-_Q+6Zh0gw*5ckI
zlbqln`7m$QuBZ#|F5(FO)Hd~WFuPXh+YxoLau5CEH^gtyLFm-p4x0vd3(YSH{WJcj
zT(pfb!$s&pxH%m(B!)Jb9q-Q~dFSE8<k8o9e^NiBbuiTP_lW*Zygp&H;ilS?>z->?
zms%TFPL%O@-oV!8Mnhq0ZZ_@f_u*oivJoDn!sK>eX{?YUYhy#6Es3(`&cFF?iV}tX
zZ9p2cio)o&S?FV0@h=W8z~k{d$EcW>LQ)v$$2ty^?w*PcT(bmonT7~%AY5cnN>6Z!
z=^V0t&<fx6ZW5zD83v4o6|dV7H>YCcsrZ^j;}c_6B<Nc=s}D<<z-;4(1SAY4D!TGi
zP_IkLBewjHZqf%@WO=N#+z*#(PNf*81j<BDf;S&T6?j~UPi(gp6JVkq4I;H?TqRVK
z7~+2tPLED)8?bHD2Db`c>l$Te6A%BNKU{1Edq{B%{OD&=4}J#T^Pph?f{TSOXW{44
zsme-~-#Ro(KfT)g#D5Pu@IN^{-yE3CCI%VOwquNEl%lMmH1z>wME2_Xe0X()7kZoa
zT2JF~hkDfqTR#lz(uX-6iv}c!{ao}%GbU`#ApP)jom#A3%Pz<p2<X+`e#g+GFBnjp
zZ0qASW$G2l7;EyA%t_}ivOj(r%a`7phoWsVtD)H!O(e7@OhG)W8|%wKte(Bz9a;7r
z#7*s$v_|jV!LbOD<n&ZpNCQm$<4_tV(C=noKgM!s&|nSWjwfP`im04nD(XYG$i!}p
z<ZVaVh{-;>Gd>%YT7hm^>RV!p^0?l5STo}RuGX2c#BI%><&}=Bu641}CCg0g>eA$5
zBrosa2khe!#LgCs4W>j?r<V&lb>!P+qA8{sH@q7hZ%Ys?Io5TfpJrl9I@F=7v6o!$
zM{LMXuB_x5kSlz}%TwZ>zeV6I4$RxtNJEeRQ8K3``#64(+{m&J9yN-rokAVyV`*}%
z`g|ILv6j)@8)E$6p61=x&tD|d^tCA10>Voew|$*sCI*9G<kffQ`T_npb3?!-|7rIH
zRl1KqeVo!<EjB~rG(5&IMws5tRG=rqu~4ccj9A}<Hq_K_A8#a{S2z)uA+eA>QQfrV
z<TDEW`kr%QZ`QiQQW5AdcWsXW3P5d&QZ2k~kQIK-$aqX+fiaPtS@W(+=^f<E5cCPQ
z^n5*{o(v4+3sUaNUMHcI8;%6$W)#Y0R05handMSHg6QXFJ-HK__~J!+G_oAtoHy|l
zAqAQeQ1+*nKPRAYtu1Y{y=@trjA1noh&-eUFv2`vQp<VClQ%49gRvnz^nxMAj#e#P
zeAdZEwt#6yY0E}o+GKZsJ|M(41H(5HV@6OXzA%4?(kByJO`_kk54-cIQE`|N-%?G#
z&)w5HiI)%muEH^~<u{EUlu(`lA<j@lA9{3VNMA-s{j2N}i<Qqu1S|5RQ$vcS?H0H>
zd$&r+S2iCc<*K@-_?N0EgpHK24r2F<^1@&uFd1>tr@Kg=@7Q<UN<~|xH8j50eB)s-
zF{9h9>&Q1dmA3qKRnD7Ce<&$b59}x_IwU%<J4ogw#}w8MqL|5-Y>I@XU3I-4Pn{-U
z*d|{A$-!JCs58A5U0KXV==H?+y!n2?5Wb&MVHatG?G|=Yr82sG^q8fWs}=8O61i_X
zCVzJNF{t?Q3YR|42I9MCDCoW{VzE=0;Ew9vm{<}Dx|~tpmL0#tjg&pD@W|_dA#)ho
z=pu8$NLV)SOG)y{o`b7ir@6=-c~q$x<?CDx2mV8ACM;Zu;1qMPC-n|pmm9E=U*r`P
zeyt#BUUa|nBt>Zbz?o<PZBwgWDw<<%e*bjJ>ek)WN7+;FIxxJt4Pqmyh$F|cs!he}
zyy}KV91jLYQ4LHa1(JnWs+Lh#w>9+x2{SMR%pY3ZZ)lI{;4(+*ZKZt@4Z8>lV~$V9
zEFd1gE+%Ab5*@k#E`pX+Nm&imZ^_CJL@O=&1Kvi4^?2+>r~V*BO(yo`xa-s8^oo~>
zMS;m(0oA7!#^qwLEOl`YJ8hgL{VPmAR?Z&_RPc=~epO(O!EI%vv^^6dt_;S6i=S4=
zBDd(UPJ2Hx@t17jcXP_zmm{D<_pgkQ&C!V7^(8?!JU!vs5Nmz*)M|p$zptxK>K6A0
z-JgZjw-|5&h^!@Hya2Ez#t9!ap7q<FQh-TclOLjXb|dkpF*afvx070X?e$0Gy_ABA
z^`)<hUV8-h&ep}mVbbk<v^5!qHi_SFB-IV@hxz!PUjbUb6-8%}?-d5uJVw%x)`@h7
zGe5JdcPtE7|4p}gxG_>;j!%cNx?~CNl2~@JNalL0KOCK;Hl4<hE9f)j8p`D$ahK*0
zVjr%7`#kYMIo?V?(mET0OP?bbxv}q*GDI|ahCbuLbTmk__94Nl5tW@h_XO56Jeb#G
z2d@e6<<oi5wJmvO&V3~23<*S%oK_;@+<Pp}G=-X~pdKOlF3qb4l~(+5a=%v861ykP
z`H(owRPBfMO-mx}P*IBs`p8Ew0$jGhXah@uw}!#oI44gA+`Ltg(lkcwIkGeY4xLo_
zEU=Vm`|nuiC$g*6d-ZybD=lt9enz+JVmnFr-*=x{ykE_Mj*O=GfrK*%Wyd^XG3pM-
zWrCk9N&2NCY;HDdOj3e^b-KrC>Qn=`-(GgT^18oqL-(ri&XU#z=907NA9neA>vSSI
z0kx^z<r>ZrF+FRM#Ys*yn;0!8h@agCM}yaBR1o-b^sG;CrT~bj#X>MJZ;f9W7Hf)~
zu2Cfv$GEC;&M|1DC#1$=kIEZT*Cq=~Z0^ZC468mIEfx;C0|r_L&$r|}FE=|wl#>Zs
zChR<;;zuiAZ^JmnLG-j1Bl|NP$lZ+O`0|>2&$M!|hXF~CvKqK8daMtpj%_Mr?mi5<
z)ANJKDw)8y8r5Ryjp^`NhDLPWkE0*X?_S0aT&pI;fNiv4=?D{{25G{imp(^tNWl=)
zI8h>F@5h)04ME-FfJqIXlFllB8iV8Dl<4^uofMGKO`=^nj=wh|64)$D@COxOh&|Fy
zd|+N(r?vUTPVuyv35)RC=f~tNf$D*Co)?#nB{oXS&GiXvhd1Ep*?8_xZ5R2)3Dre@
z+;8lDpst@GhE`K7hvJ*e;&8JdTLqKZCsQ?wtzPce<(oYPKDb$HiqGAFs>Np8LWhgi
zuc7OJApLq}j|sD^Qu1a)-nWV%?W%2%g(1)$1Q4q(S)3%4FNSLAuV*i~y}rRAPLK7&
z!9F=xW32eKka^KF|8H^VPDFM28_U7ED~X6Y$qn~|J^P^(kSkm-98fuSo~n*y(Z#sP
zQ0xjA^Oa0cIf~5lI5zd!3~u65ame+=C59PC2#<5J_QgF9?;50g*L9wck4}GeU!U)m
zEiw|dSzBC&tnQCf0r?q}$Qw+sMJ)ryo1x=%<o-JFx_I2!iKWj+JcKa({1<4wanC65
z<Z~Bj^YaAAJX-wS$d9MCy;kC|PPpZ_#NuU>Rsq@dNaM1!onPabX`+yF5>d`uiMQjH
zMsjxn6156wr0iLd<Z}!`4V4c;Y_Cm)N;7b2?b5aRg65e)K}1zuR@S*a_y<p>aqtmr
z<9-Z=Q9CNyco7^aA*NeAvhQ8#6R#_m$?fcL$;!4&Q%m)Rc9$27MI$*ypR6AK91;jN
z&oSe@c~|!$g>PJ{K*+Mg@<BMkv0yvbI__!(R?dD`h=*l#y=>*0yZqv4X5k18fRA2K
z)x;;ZcH+Z^L-4AdSTJgX-j2=A#S3UDv&0+4=ncy;&0GMXL@;FK98dU44<(B(pCdhG
z#@e5j?)!U`VZ8Z`kO_S6wI}Ry*T0~1`D3r5Zj~(|`8&p;YUwh@6Q=9=$A5{t2B&wy
zG$t){d_G<YFfjiRaz6}MGmye=2lE*lz-P)u`}IBuYk3LH`k?qdxOrZ;MtYtY8?4uf
zb>`kxEV~<HttIuC-O&L|CZzMj-ihFie15VqGDfYYmwdqgoj>yYf?I^B3(LGPA{(z_
z-?J;LA$SF%?j!ce3*Te<Yp7p`BuM_)wdFb}?bzI-)bVHug}^qcESY&b-skqH-V1g)
z-!B0nO9SH&q1+ck&ZjoEIKs}cbP6)lmKR)DGTqe)Wve!2$E=(iR8h2~`0z^K;DgAe
zrQBM11kV(^uUI3Mu?F|3T|5PqR00ke0SujBw@k-G4jv48MkAf{1ltwhkB!dOygez$
zGGq@`S{krCE6)lHj^z*1xMeOdV%9Dm`-KG?y`*w*4Zk2eeh9gMjPPE@<6=OWzLr#`
zE+}X~c6r>5>T{_esr`D+r#lVH_<hrbL1);o+q|h~q-M(TOshiE7rL%nnIj3gwi{*{
z9XNwHmTPNxjtW|S58|l@7aXg;KlQpXvj=2QaQa#byn7uZH>#}-3iUGb^bE%J2<-h7
zn&8}cts9-jq5~Vn+`erM?n45Jh2-J99H7mFRuYFVe_vy13a>qA(>>njUc8Z2&zYo4
zc_-5Ptg%LaCjvBdvO8n1u3}g>$u0U7pxGJjx>_rK+J{~xZz#LQp=Jabe|<{1j%rGR
zRT?>D!-=x_);z!JL+tYj{2YD20LCbP>cF0!vr@N~(s%q`L3SdloN6t+-1IaFxT^ZG
zUJ+YlXr^syJ17iDy&On{v5(qG=1C<{b3UO@fc1`L5fo-&rE4o#{Ynp6J2BqZ7<)Q<
zK57Ybi5jK}NnCtE^_c@e^ih2%;0W`Liu>#>l>H%RV2>q8+cfKxMJvJ5B2^xLcaW1*
z)@TESpVyLBm0c+=+DnZXO&B`Oq(%d93YrJ*Dm1w8n1~(d`7h0ZUINC!$UfG(fnYN|
zr;AQgoa2vU1ca-9$+fQdKH#;4BP9@Djq8Y$beqcJ)3lx(RYiVxS)U?VxN~2A`8V~g
z)IAsR2pjfcy&R5Gw9req)>O8dK(dqGyfHfNI|@q*t)|h5H%rxqZt0|yaPDs{&T9<`
z5Y-p^`zc8apt!Eh9Ds;aF(T@IXu|F!rj8(P!Emv?HxuX91)(UG_mZU7c6m?IkLAds
zkGLL->BR41HWh&`wyP(C*NU0*rZwr1rB(#n4<MJ@A|8J<rTsY0KTB!!$=iyMiY<kD
zzRh^M+NCU3{aeR3zV&@yZ7s8USS~YJPAKU|`nB<i47<K6^>e>_o?eT5yyKPukwS(1
z`BuCgJ^2K80cLB(PqoR*A$^7@4z7K_dK^?*QByU^61x5J)?F912$P!^gXT5rL4CSe
z4N;fer}w0}cCvJ=gEZY9GR51zvjM_iH*eY#Hp8~CvT-K(^ji%!;Z=d6kmJ0SeyT^$
zh1(gLsfsx)`q?}m4`N6+)Q#smE;wy0Ve>HX=u3K;GS9kFys|6#@%8GPHQXe&bN6B5
zVuFOQ5*1ytl2kg&2jP(8kS%>6Ts{(d*(7$7*h$zaylxr8QA<?;h^W`JT6@kS)=;u~
zf{uO4VOKE=&-MnWm-MQ6F1Yhl?i-Dj*30vL%$xyo3L#1H=WJ`y4icyL7w~kcD!#Q)
zzF6;Of6w;Jflr+%^_g?heM!w`j+7lBq2Pdqjrdke^(j54Y4|8>qR=nJ;l-V|G5>s?
z-Ls-pwYMvNwr8WJ#gS{C(BeUl9>I`5tF<<ku6d1F?aPPKWS}-~>;-=A0r6#c*5Yrf
zVs(U(MXqx*-7}^vbrkR?hU1ZF-?6`4^hQY;uEg7KTE1kbvm;5b;$-$ZodK4W9te-{
zn_;o9lxCe8k%37s=mTsB)0n2vFmM%u1#b{?o_jG_uO;(&ug#|KT5);QDnIr5x%Boq
zr3LqAp5Zy2B@&VkT(7PhRj~tr8o_r=KSl0eX3SyI;`6k^-QgEhNRRAW^uvV&9f>;D
zifN~=W#K;O$FOG@Fh)gO=6<}u*PoIMarHoB-EpoPntigVOe{@Zp1F|sEZEiU2>}u{
zg`mT9?{5Dmyl!_3S7Ty3>>QX)X}y$z_DV&FY537i0Y5~xYJN2AHPTIS83*yckp4(R
zn?M(##0VtPTRSZe*6l3@(j;LmdW4QEpG6z!S)}M54-&#|OdOFsg-M?UeKFh<E14hV
z`b-Tl5O08*9@Lh9N2^-D;6wE}ChHTFCt+92Fygo9s&=mgg#Fs)ipR(3bwo<66tU$$
zoL#{5^7FShY|+n7oQ$J-1)hZVSVNh=9MBv8s7<SgDz1^vGW}O=%K6E^oWm=PZYuW?
z`uJcSx^l<%$UpM^F@3KbOnMbJv;Z`Tr-QACMkG<=Wd`NDKGaEHIo^U9+q@QK8OMWn
z&j9jCL?sVJn&Mk8Epb?B@>-^`0WM1PkAda%2pNDrC!pC{huTH_M2#lvs;(SyG*x`E
zoo=1tYpyjsde%XT+qK4D))x;Kt6{bk{i-Lo=jIasMA?C_=}Q{x68{g|Fp0wMSyy3#
zBRq|rha_BSJ=1Ep_b`|R{oQUOW1Yf}Kd4^d$?s)Tq3u#$vNKE00s4b;AIYx2i|<Yl
z2WdTHPk|lcE_<O*!<yl&P74}(Lsjs0=L7lr`M2fLV;n!5vv^M}NzN6T9;`Qntu&(a
zJ<CQK+S|ALF@op5WWZ;7Ye{H5xOB)C?s{VEQB$1R-K-~Q8&E`wrP15Ay1YJYnABn^
zE`H$q`WpL<ueTZ6_7NkCJS_@M->3x$sPWz&Y^(=mo=@`I(mvxwS6IoUxQlK^r9T?r
zLC@C#8|N=)(&B%5k`!t*)ST3Yk&>`Ls_M1jz!Iy!conlEPn{i>;VFH^Wfth*FPi+h
zR`}uD@;ft8mx6m!4y>-iOVsF^Br|MrD+p{HfgkMN*S2YAK48%*&XFES5SV--rZ%mJ
z-K>3=EqJd+biwI_&wnea^+f>>o^IPDo|-CYBF*_A+Q+1)Rpc|7GARk?;;!L%w>wl?
zWZ=ZRTt<RR90W|OhI;38CJnYl=a3U-@*S^?$BzUv8eZ+-DSmDKHo@^O05^=^d<<Lj
z22tuj>EkGWqk&USF%Ab3l8m&OWG|<FzO|sz$I}>w9oFlb@9w71B0pxk)_IUC>djFY
zxD~j=9wegONTOEtc#Jjqfecma@mN;A+_9yR!Zw$&2TBo@qsqF{pd8{%zC|p4Oa>ND
zr;B|`;gAY{?JAFFvI+n#2M?akQ`_)62OZ|Q{djqP%C2k_rKlu4V=tmhH{=+K!V(_f
zXE?HzM!e<8_4%wF7rv3(=-@pDQ=72lj^@zJ7A^HF-*g$mZyaWippOgXuA$b<<kqBr
za?<`LBrc)ln=!qkY}HNOxU?KQ?$!BD%I4e-Jnp0u;yk)f>kjgAzOiq^S)N-E#+efK
z@|yMEW;Is|%JhWFHAOv^x0ED7cYN}8XXTob7z!4OcVftfeOHaj9i1W&bi)a&D_wqr
z2>VtPwe-m0Pq*@v@bi##1XZgZetkp2{dK13o&3F`pp<*q2|ou46f6+Kur<wRg*OgO
zDOM<T*Qqm?5?RvOHD?T87jk1<+ukr?gs%s4H*z$&0|+V4zGZ>U)>@XI;7EMqY3>2C
zmphnKm8oF7ds?aSDyKFdqq7e|Ff2o+FsZsW^4Y)n;Ni|pnjCDu&Vu^RxSrm~og~+5
zGhO^Ys>=r|t#JPz_TD-w>Tuor9zwcD38j$~Y3Y`f77>sdKuQ{=n?Y$1k?ux7TDntO
zLb^jrY5-|??s4z4pR=BKpL5=4z5l*zxt6*_)-ZGbuKSAb_4&pZEewNq$ZBc?ckOLe
zrjBY&GVh21Us3J|#@FnSgDjPsP$r*n^VCM>e7oV@4a#-w-A6l(MU8X96HT^?v$6Xx
zU67g4>Jwd5x4&WN^K{;|RNv7%`-`u75=u&t*-kXyK>TMiP+j@tCK}zOdQC6I{YST-
z{-_ia+h%<AdiKv<fmJ`rd`7>1!$stzr$k%a4pq#EdP>H4gz1ZH@-Y$Xr4?H80mmp4
zw-i`TmCPBO>P<WwPB)zpxf0*Yzfm*je(7~+7Tj=hSou3X#PTYtx+r8UE2O7!!7L`7
z-i+f!=Avz@e?y?A=uLJar1kPwzc=c*W1)VeLjw6wJcX(#h=h;q^EFR*dNkxS(y8pm
z@yXL+bFCO*t6&&NYP({@$NjYG9+8RG>>~Pu5sanl)Z?F;r$~N4wu(<k)vZmLc#`DU
zESgt5n?=k^QSlLZX`>K)H_xV6iv+ID9s^|<>ENfVmLRTpDrbAByBrilt{Vxrqm0rF
z3}VzC#4+MC1`Qy;bv=3KmLaE7Asv3j>~PlULed2q882E@ZdJ|oV5C)<M~+c1o{3Dd
z<qK!s*>cbkU~)`$A$Rl>R7{>{PmvwPaNGeRo@3FlGs_9MXcS9S%Qw)_)D)o1ofQ-3
zkyDg-%^ClQ5}tQ75($M@MX|hn?$c8m5|f3^<dJz?eO}g$ExO2Oc!I88PYXg-at;*w
zN1k_J<y>O@9O&mLRPgQ&BHnBbX}fN9WH(DvNVOz<b^FoRz&gBTDllTm_f){vdzUHv
z6tVEq{77{EW!KRT`LN3QEm->9wQ@_gtqvI%Qnu8d4idMV^LcW2uo@;Snlr~>usUpV
zr99gywvCmlAGtsBa9H^!pe3X*i()ebv}i)(m(5jv?s%Fj_w0NKj6n$HKZ^5Z3PxNd
z)zh1ic-xqZw)M~{>~w0@CJAyUX@n3jSUxAn3tBJGT!T&LG%GLSpVR`tb^ql<l_|23
z?>SYn_XYZ!7JPdJa+{sO&02kOvIa{`;6H!T)_fVUZ&&og?Ue`+Zp!C;v)d|GykO}b
zmfEQgq`5&7Jtt)c5BzJiEN>4|RpzJ%lfKr#gJ;Ksiu5Dhv`+M?p#DBs$}A|f6d8jG
z!AK#7NCo%UN;E-Sp_H!OGdP|Z3M?P<B{qck3B6%X=E7j&E0Gx|oQ2I@m-&nhn>_jA
zH-Y&Z(*wk%2h#)igs#GdSGf+d-@glQn9F`~>#G@b3dTj>n)>2oMf)){Kas7ZKIeq8
zIb@Jv+bu#Bx!lky!^jm!Mc6?LSrcIpwK!-sUv5ZfKHJ?A&8L&eM~lb;Ga4!yzQTb|
zFLHc-^o~p`P{=+lq7hIC(g9hJGY*EWO<WD`aGy`rx^xp1li*4VANBJ-#k+J5r8j{S
z-rb=3tG`*g0DY00apF#h1~W3FgQwTZrDR7h4my4vT(I3q*m3cme;YhFu%f>vmP(l_
z;`LUTs!$&qmrcyPjROhV@@#blu0#j5BfH5jyfGuA2}ilV3+$|Y7ukZyLM-DXg$?)*
z4nZ9!{B<kv%Svko<~l0chbDAC^iJ~+a>*Jya2Vd%h}<JWR6@!n4gva-Cy&BL%p+YJ
zL$7Aoy)GoR&$IHj7`|C0V?l!Gyw|Yg6K!W`azxcm^A`VhLiqYcw}?49<vu@4eP26Y
ziuLh$U`YbLkM(%7@$*0Ig~^T3FYZA`hX)^KAp)N;g=vs-V<|T^{vN0go?In1Mk36p
zX77!c4j^RL18o8Yu}B6g<j)UdFUNaqzetyk(eq^w%G~k}6=T*(M)TX)#m`tCGpKK@
zMzx8~@Zi->IM=mn<To2D?Fw6j<Fw)%%Jw`|*go)5x(dR|=*zvpaK^2)S&_pIoitu+
zEyg&Pe0%6N!_=JkO{mnG0gqL9yVN|juM0~bZLzQ%a|LS3H?CR8#OB}wmT@DLuPBtX
zLBp;ksbU~nj#!cYid|SHv+6QbTB!M+sVG*^g`@LM`MSkK$n2KkMRzc7mFJfYGAd`-
z)sXv4>~($Ka=`=U1ML`xHHD4rd(0T;dF%TxPCfaIb$wiCN_T?Z8667r@P-b9hDIM(
zr&jQ0pRM!-M&hy6#3>t%>^W?rqfu4T*IlbwdAk2KN?@R}AbZ1jddF*}z_nw=&s2x;
zAt=YJ(JrAJ@^c*GrM)gOy?u=}Tc-txOWLje;@=*NWY<dTH{2`1x}L{3C3dTOT0UDF
zI@@8!?QghS_`|Hz)a-#B+*dcbdB2c|x5=$lzny?T)b?EvGQV*068Eq_7iNC<V0~Zl
ze*S{;e27!o8!u+ee#EI&{qPjVjqR(N(q!?ZiQxiVURG=rx~vwbG^GbP{tT-E#}ck2
zkYB;tDlPtd>9v1_vmd(fgh?=bAA0aA&dsRxjB)Z3)#h2Ed{q7YE)7GZGH%r8Cuitp
z#T`lWd}EXYz~G5^;pOl;ZXi70xDyZK#ZG~XJJY^0LHUFCj5k>%V4Nk#Hp#mhbc8jL
zFfBa=g#-Eaj|cj92U5-NA@ui*fbSiB98JOJcgfBluXHxzari2Eu#~20_;4tQ;AE4Z
zd23fZZ9=eXD%yU3e3(UB%y(v6hf|}YzDMLIwy%uVGk^obxF68A(ojbGJ)<H~<77JG
z_1s^DGgmaWx_PwkbGaM|^wAf5W8y8<Rq`&%pKaf<k0y2&?`N4#DAgUsRrvA5pjh_(
zNQrd1)AGZub9B+wrkb92CSg~>0*&{>d9HFP38X!G=xHbQz^j}?LmN?UH|g_WGZx_%
zAtRSB?2nxCe?J@m<Oqk^(+pxA7rb`|kFhat)`esx9K<FaQV2-53A;S7H;f1T?(k+3
zqwv%|wIJd>O|<ko(ZA2mS9Pd;elE%o-(lY=yP(*Vd&bc&g%(E<!l%$5X`Sp7h2X`U
zw<jSOm3=eEqnmRzkIlT8l2|!TcwW6;C6%2n#m%Inj>$I6m!Bp>MdAouJn50RIn0pS
z^`JSi-0sl9VNpXYd46Il&~$}{Juyyvq}s}vtHCt!(J1HoSnozvma79Hb^wLG`97BL
zO@C;0tJTc>aV1-k@To!!UKU+SrCF6Ib)Lhgs2{(wyzej|lAIcqCkYVAh&~Xi(5rIN
zvof%67XXk}+&v^jam`iB-F*7C7l#(@_u5J4)|_n#LS$Mkbz@*WZ%`(QO}Lpk^;<AM
zMd#xNrY*FWUsROz*{YiodT~Tn(DoG~-C}2|w#ZvFdA!%~<U_Mr)s|{h-3f3<UmyBZ
z7j0Wma{Ahj(wcPu`OojGFjUDjmGin#{6D<qZE<IZqILQ-OZ8cmjOShsY5R&n4^PQj
ztPWq<{zKZ2$L#L|On0&ATGYHN0x=xP*D<bF)~h|atMr53lWret*)FVnv(Ypk<!m(~
z$r!1Ya+ru(6`onbBjT;)e;HZU7JNayYkp(7^VUWk1s3TrPVu9YhhBPEwY|CXlLzIa
zO&hT7x}`pAfB$Z7iA{gyw^St&D^Z5se4`ce!_bh=o&Ktkf8M-$pYhBxz?po8Dp>xU
zzw5%%gXx6sg1>w8=RHs4XA^-mdbWMU1VWgq3G1guIw2jB#Y@Ev;H9$kg4vE)g6Cxy
z%p<?i--GYu)pp1KW!x&>{nTAFSIvaMaQ#B)0&bJ&vd^q$=K9l_FsUG_S@I&%GMm(3
zyF>D|!mHnFn}in6A>K){@eFok>65P?P_G5S<G58?<YN~x6^(T07}Rn~Q=PtD4Q`rd
z(3H=f79x}+dOktQbcP&gUe3dm#4<Jy9<BR~tg~Uh9cOi_EY>s1yA8Q%v5N^bP762q
zeCgs;&hC>=IcpXE`WZF^^UDs2U2)*(ZKPptO!i)1pF-y;08Pyu7RDI8(?>7FYWx5v
zm6D%Nl%Af<R^#LuNhrOv#ovEdusexi8lY=*aMEU*(KkFTX;IcbU1M8Dm+-9_wo+An
z87fj7YezNThQa?TNY`QkBLSW$M`+H?tYgYIWm5o)7=dNA!t(avAt>;{99;gXm7PyV
zh7+gNxdZ`;g!ufmq{Z2^I`8`?VKpd!U=cSHBE?yEm4Exy@=Oghz<y1_bv3Z_J`heQ
zoLVF}&?T@Q!R}NAwRPHmz0`d$FN5oU(pE>~GGPq0`5<4W>(XX!{jJ2d!uWk?sgd3L
z_obS4(jBn(4rP&a8#Zu_Ha1TbbAF=vL#>IWQ)?U65EvDttymZR!AER<_2pq9ppAsH
z-puaRv~9!aBVfGhXtQ~71Y`#9G|EtZ!Lp<*!di;k$!)L$U~p!X+gCl3vu~^Zg5jVM
zkoA}+B{KoGS?92TAeeONgd;TEz0SI_M6V~wM*ZeF#<cPZ3_jErDBmW0@szDcrIoa}
zoY>|(119Y+nJ6u!AY?&r#~B5bIl-{l4W<RlhSSNG8)vTafJ@fY4AE+AR70O#K-Mz!
zNvw0^j30ZlH5n*1^xeEkL}D8<@ezmV25bDO4edhzIZ)=$by!yxkAA&T5`$M1ZQ970
zW?b=c=Ug(RX8-;rG*0!2mpgnCmt}*+Qw3|~1e!b&%W=4IL~Si8V*MZo>x<zhjtWJz
zwoY@yJiS$M9r)rkKJGT>=e7w2Jf=PD&tJV_Mx6NT7!oYIc0|W3ZI<W~Sm>8Z0c%4-
zU7kMeIVMX(-eUGuBhw{35MKZA^w8SmMJbfnM{krR!THW9!QxD3q3y}j?W_s9v0nz(
zFpT8bofFy}zN5=8xWsD3tKB=={8LNLR->^hIhG9)Jp_&s`)2TWh7x%`@;)Z``lTXz
zt71!-D`d`MJIcjgyR__M@T6|+2Z9dOIoOm|^X4xwlH1}9ms@g%S6ttSSc!BTfbiQm
zh*OkrkI<UII7_V;+$Ec#oanu4nOqF`rgNjeidPvEa$_K5&864{O<$j;tQ#U;RF-Tk
z#<(T+ua}(a$(wL3S6wcmd-@|H`8yb*L|0nOWye~Wx3buV*@OCpN8rwVg4#Rd1QVGC
z%?~fdcl8(GG`8)r`P_XfeG1T=3jv<!w0Om{VRQ(?6_@1&OSRP@+OF9sYa~yEpn3l-
zdNMhg4N+TEvcv2Sq{>ROVM1-Hg*8^(=|i@ayK1w-W$J~-h}O?xK{X)-D#p5T?e?1d
z(l(4nc*xe6)$m(e6BzAcu6`zGJNgX6Hrm$*HXZ1G#(+iir=Gk=acm8mt&7RhZ1his
z@gVjOi8mAH0gorAPmGyhxcmoe&eAh@f)|5dI7e&di|n4x|G=mZP+`KYW?u?+9g$F(
zor=plbuEkH<xZ-({4%K?r@_3tUT~^TCZG`!Z`D*X?7V>SqDfpfwVR3a!qn<??9YdE
z-Ttp9h5E}GyEdfpjRYE$dIgAZ?2E?RR(4d+>zeJPcj4k_%|Y0IW-b^7)fOSw0>^af
zi$zYy*FVHf!m~Cvqq6qC*Q(ZVQ(Ocbl&UBn|KUEY-Q0}lB95<7IiohO>+5+K+%oQL
znXj&vmu5dvA665&&A$q*zH-P&Q!yxtw{GXrzMmVRbPL@T667)Lc8gRI#10qWCY3dc
zQ^~?Xx=lrOCA=>+lsfW(RFacghT*MjckUj0&Cn5w9{5GE=9;b>t)OiQelzOZ=)XL+
z;KP=m7i_ys4rGuQruixeN(<Wyv7|Eg7Li?n2&gD?5AP@6i*pq-a(6CoO07sf)#+DS
zq%js_uJMSF!pkx4?f}-On?|g^=08)ICyGx6J}-Tnvs2`Ah<n%-s7tplAhhE0vjIcM
zoOHEO0?fz^U;gx#8{5}7X_gu#bLNCzRbNC}9cj8~$QygX9SOQGJ%4f*UXAULJBA=;
zU&bDA`@Qqt)|7s$@mu@hB2#!c;lubM$^ni1?`;)|-iggB7w`gm*PY7D;kbG&cQ0nL
z{$pFGYeV22E-o%@^fIPAtbQw7@jARr(>JR$#m8wzB$aZ*%6^1^!JfT=^04qqc{fb5
z1)08nhtN}D^NDW@V)U~XJ3ORCcXXx<DV?fQ&3Y0h&LM0HiIpNH;fhd_!Na<*@EH$v
zsUYLWrkVO!ahV&2zL&&%=4CdV`z&+h;kgW%o12~alI*-NSgwwv=~Cn~(iM1?sd{j?
zD+dZK8saX?>*o$;=_t?dPvRYnj(o<*3%x0r=`KB3)_uX_jzN#PX94+=qUb~wF~mo!
z)b;5ZBis%P-3>#U<i4}lQw&k^&5y{s3+T%Ia4H9XBSL)yNIxrK<qw*u)%0J8_hfLx
zQP&0qg^7qKa);jbS<B3?&8}_naXFeB3}3&gWH1PfZvfdVWt$MHQ>Vt4Xr&;2DjxG*
z+5~6+XDGBct3JHzyD<ct2^qihDwCYTJW%fW$V6zFajxL?iI?oJvyd$Zq<_lRP<~pl
zNwiB_q7XUs3LU-o-VP(kz&_`HUMHMJM-x6j<UC;xv<QR`os<(np3_Lq$>-l)M2j5H
zJj-4U-me&a_UKNs#;ao=>xSM_OR<L?-L%UY(2io2YWz<J^NDw*8#_m6^Tm~(iV0ON
zYz7on4RdaUOk!bMRXvx_zEQZ48dpsHSS8;hj%I1*_~uX-ZU?P7#B7fPGT)IgjIH!D
z)!GBJ!`NP$F4a`TC5ExCmmFBR@St5}LvV<T*g~lhWhNWu!ae(3UW8XDp4)v_Bh%fk
z%(l@(HhBfOiC_iHYTwJW<@a}HqT!JtDP@KGx${yRB#T_$YAdf*;~N%+PbKdT!UT=|
zaJK^d+wXorBM@vIGszdNByS$)ejUeimB>Y9EAM#-{CIG%yqJ^A(g-X^1-I~tJxMB{
z)EK%Vxrb7tnDnpvO4o=^N2Q_|7BeG2i`<lpSUyE@1+9<eDs~R2zJ)qQszfauQX13t
zXQHxq(>r7JjrJaK_g~Pd1gweLnr;)BrzRM<Hm&flW6%?H-XF&%9{eMsp0to@^baK(
zt5Sv$5cV7#NoipaXnH319+(UDV)};J{wO5Uh-0YF+%@EQWVy<v8**bBxVP`MA<{IQ
zaQed~Z%^7Fil^||J9QtLUHU%Bi_xA@E&Pd9SEijYLrFfPo^>4gR1}qMjB$z(vL;IJ
z2VR^^EOD4On)5ZH%Nb|N-A~AiIdLJ0ooB!)es1=g6_4d%^9ui|ri2N5USaC{?~yL}
zwy$TxtDbPPw7O=B+mYsYNyaLwnXYe#HYJq}#L8J6@$^Rg5n06e75<2hBtxRDIH%#q
zd;UAS_?7xU$%gV!^MxHn+_B2ZPE2Ns0rhX#jR-<;Kaz6Gt7@f_cCqS;YNxx%I2;=I
z1&li>xA(MTjn^%HV7AMw6jqQ&lb^e8b8;-|0T<sW!4!U{@<v6}Ry#9!k4W80Tyw>V
z*`6;gT=H|O!}b><q~Ko`v-@u0d7Rr8<d55HiPzc8EHAR{qqy;zMBZ&{{!L<?upn<^
z;b#K!sae?lc_(HUXM;}mlAmV=xvc_grRSSi?hf=;%wF!KC69vdcdnIR&taUi7vjf%
z0(Zd?u%)Zw!zwO9_nVu1Xz1)w8W|=jfs9MR7G(piw3<{st<Kwxl2R&+xsj$z5?VZ?
z-VdMg1}%~#(ml&qR%OP(SH}3n!sXB?qHCMEtepLAb5oh~)sf<cW$p7Hq*U}zuZ+xF
z4-t~kdrySzLL)Im;9u*gQuzk{Je;#(AeE{u2b+hH2ii5Ryx%2*h*R<KWZLiSNZ`R@
zsn9<1$$he<_qM3*h?EfIHuayj)}p$zZt>=?e2WPc1)+i%eweX$*BoKk_br9M--x7q
z%;Dj_s=Fwvl(3DJ7c7EHNVo|k_0lBA`ETodt*-9A_xQ~EndY4jljGB$B8M~)zI~SW
zZG!z-d<%XvFM}JRF{)mLt-r8DvFAWr_XT`iyz_yqd}<EMTxJKAzn4>&hobPy8<iVY
z{=IVLKr^(y^OZwHlh7Y`%ox@swG>*>gV+2L&S+;YZ{a>scEglw`tuG;_M-~rY1%h?
zT3ckq^7&Fi+W}s_E7&LOI?v@-qb^Ljn+Zv}?M{h&n(ysrm<H{K_zHfIP|yP%pb#}@
z-MH%~mMOFm?NOHvLhZqB#o9V~Yo{z1G-1VSVrz73*g2fSl$Ok8^Zgnd8$Jk8a4)6l
zT<yh*dx|w|i(b8*yi~NnTCqw#Obv;uI%hdOi~RC?i+Nb4Dm<5vy_9>)vY2um-(5V5
zyBTxwy$hR(Tr;*Uw~r@s1Is?rQe0>uzOR0z&ct2j2JY*qWvd+-<$S?%5qm0tEkA}x
zIqkqyy{@I7eJAEf8%v`VKKAhaJ8ucc<a&Ix+wq@73rB|kRHYt1&7eV{-IWa#J5C&%
zOr?zLA^r^r%zNHpvjv`9i8-4@GgB4b989mBAhzBY5v+Qfr+HUB?R=!^MKYLO7Uii*
zg4#ZIrPrz}A-(*>aBMASX4PNFFmU_xrQRTOP+NmhPpI^lVW#;pnW}C{wr*${pT1e9
za1-1p@Qbp{?}rfO$>%+pO-i#L_?c&wpBZx{yAm7D49rtH*JVAZN)?SmZPmB>3~VN}
zt4tGMoe!mE-YIVzBN$N)JDZNXY%J)B*cJY)CgG%jlx8j6_^#*A2xuT!QgO;T8VQBM
zy~d7h$L#0beg>L(Se)0C(Da>Frh{4H@b0@}O^TO|&B;sQ*vo6!$7r8>GQ);p4^T)j
z4EjhL03Owm#yQL`@+w_1Is6_ou}5U>{BElAkctYbWP5~L+Tpv4&zmtzWPgI=3-ZgK
zVUhlFs;306bsfg7ud;I&USadf$Xpudg!X0Z_JqJ-rz#T|?y}-Q<l$z~I~lX9S`dX<
zzw-Es<wzdPQ8L>ADwnCuY{*AXj0o-ZQ!69%H^gQB3jg?!z|D?AoUn~GHHw+=E6n`9
zcOkA`DkPDGZszxuCGY0WT%I*jsR}D(hNWhY2)krCSIZ{<e)2O&8|7b+Gja)ZW!OCl
zM0C`<lQR50IcTYD<=mtTm91oQAerq)@tN2D)o-u&)CDTAnMN=ovhg3fHVy{f8Da8t
z7Bvzz&I++4eG?Yc{d%@I&3(~Nw{RHItiQ}f_x`dd-GgP$#z~<n_U--mZS4M+A}Rxu
zheFl&pYst<2Osf&VV-Gy$JgJ<Fm#V_N~YBk^`&nVHY6<HYZjw0Ti5-8V+va5B$5WP
z2Nu>x0vSeL+7vW$Pz9#yCLwB7y_m&kTHS;Vb_7o75{b%7FTphsEf>SAQoQOdZXUrq
z^Dz4RX=QJNBUq*kGj|kC${1~V{}?~nZtL6{DxVeN=3zLw6#y;;v5w+6k6HFz?dznB
zCo>Nh3pHbVl0IbHrL;m-+M^$(M#-g+uO-JxJe%Ji2{b#Ow`wV&8TIa3q#R}uFP(Ie
zFiTMJ+SehS4RuSyoCAz?10&fRYb9MdXq8N|!=%gCkjN_vb8dko^@d%2$DU5-5%xq`
zr~?_H=X82Dp0($up9ue#i4GH!9{;o-$DS{ms_ZUSE~g5yMwm=rhz(uVC^xR12wpTe
zi6kp8;n{X|RLdhhYJBFH>Z_?Fz$hgfyKp8FG(eIkSV3P}!-yZnxJvQs2vc4aH``u9
z*P10XJ=MN0EhfzD&|9dwwE5!AX(@K-1F;gj3(Myc9ogn1!()*lc?}n{fg{x7h`|c$
z`=x!uW6fCOjDA1w|L1eYhV?_&_~lo6s$}!RSrNkih@mg^8Q$mts;9srjd$kx;E1`K
zT=MR8k3s!k#v&5|if}sxW&W7lzgEs9(LaPpa-z_h@S}@YUp#wFTfJ#XW)D81kzyZN
zjKzAPk|(%((SJ-wMoQ#<zfUNL$!-dfV1)dZDY!foMoThx9v0$dv3hAzFHVl~05ynh
z^0oS14T=@QN=j;XRb~s)YJ-HnzyUUtB>=#nvaOpbS$UyXGK#4+3~bX>OAl8a=+Y1z
zH{yTFqatuYhm9|(6)E6Z*r){ifpU6Z3u9@fp3g2!lry4b>*9nGCY4nfP5!ehhMzrU
zU3|x{ZuH!;GeoM{>Y{q9#;);xx1f~H_6LUdzCRZ80(eK8zpg+fjx!0})7_bU=iF8!
z4DB;+)Lp<NhZ-(OF6B0@!J#gv!ORl2<J7f_5W$?T`Ki~~SjHHGStrJSO|8ZGHMyB+
zWVVofr1<Gev7snB65gGmk(Up*-3Bw5|B*C+7^0&4tAI24o_*<XDKIXfZs}XRgF@B5
z_hOP27xK28y`^UB#_b+t28Y|wS2{r#Bz*_-=Z0P}Y$&iXx7dYjBVzK;fDQovH1;1H
z7g|)!6u-w}l)(I8@$8G5ci-(F&~vW^dm*)FQIT7Al{5LZ8@!%=k%yiBM#8>8dwgxF
zWFpL<_lfoB!HFO;41qX8UM9Ns@#zdgQIGV=42iKLSoMhn(<q%z`izwS%}-WVM8#V%
z?nQgl1)z=9dNXY^1is34re&v9W2#}*o%Tqc1OwWX7F<&YN2=ic9j=*Bvm$ey2xA8v
z>sLD4<z`;90THs$V#q%S1J%Vj=^F-B9mJL+3o>|I{p^*sw}tG}hYM@I#VxSh$Dlo`
zCtb8VU*|&7&j$(V>O5BU>Cfu1+qb<>a0c|C`w{?9;=ObLTm|ez47=d_&S3E0@Sgla
z(hi^!h*5>vW08~@(0z6dub$fpUOJdVa#|PN!NRW=4t5<z*r)O6Lr%P~J$Z?XXy5$^
z{uve%1C}n_lH~=Y_asMoC&^7u2FuE*6IHUXpyLPY{t2yuN`?~J1*DzYArRkiHv4q)
z!z-;MQ_f^lsK<W-84YZjFCx!@zStk-93?<s@BW;c@ne!lp_KVXEL3Wzxkk-u+=dkq
zenBX^2*+l%=_e7hqA|7RN)Nha$qdF9b%(}kF<M}W2qdd**27aX^JS`FGy|p3sL*1*
zRf^?>%Hh)(w^%r%Pl1#3D!OBNimP^Rl#{sj$(~f?v{MPk8gm4%V{~Kw=BHbr{^TSx
zj8V0z>s?_qlotdzdr^rp#x}92$9dfBTHHbi0J&|^gEA6wGyp9$o)K_4SV0PIZFM(;
z)jw4Hj7y9FUa$Te0@cT7!3aj}IgU%~kKWsCULC*zVYmw+OWG+`oQH(7MI&;5J9$M>
zZawlQWS5<-D7QQLwyE8D>P1t`VFxCIU78U!Aw~en8x}8ulx5Bq6EWfXVa<f<mqUWP
z-i$sUS!elIAj~U~ou@8m%Hl!43bxe!8CoT)ZvYfIP0gx?>aUaCe!&vcey+t=rEL?i
zivQ?V=oz)o7giiQj?VV!=C#50s6kcmk!xTEqdjL8#Pi}*lR9Ezw&nKy-X{^-TS6E2
z{H1`*`oZIlzxDOsi<rpYQNERJvW<Q2PBO|OqTz`h+}ZBKw~1GebT(wmB6lY0zPNWN
z-(v2(v$_-Qpq9kPn!wlzqx#YoXa6R3>@Q(lnGbbBiqt00gUg!bYsT2!U&HJs=nt^4
zkW1!es=X8Lw*LvUrvojM%7wwJmkWzZaF|MP`H7|ws3#N;#_^*>5ej8ymK{ja%fNHB
zo{N6QczqIOT5_P4P{tx3Kvgf3+J~V{`B(Fu&{s}Dk(42aPGWg<sNJ`Pt`Cn-?kn>V
zDcg__jKa!tdjdj;<_Vc_j;r(VL+@0|k=ytQ{U8;PGA?W5@*P6bb8ME_r(lWu$+<UN
zi(W@}paK^@7xJ?|nM`YMN6RT&LKuzTGkc>~MaNY7H^BOt^5!F3gxVcJZ|(ySV;AgF
z0W^+Ung1p1<xaq3+ZkcZDA2zZO}NuD!OforfP)7}ra{6I{QzK{8-r7DWN$)x5P-dP
zH5})U&kvyh5|biWjM1ypqd{AD{oB2{mOi#3^mcnV;%hT1)E^y6hEY2$=~UJmYaFA(
zkIh76E7sFVOf8|(NT%FL7H8n1dzv4@3fB$9VdBCARC^DuL(8*vvSy-zfOsQ|tjDLk
zgjwmo+AwL31(ki@>vH1&%+cF2Dbpw;^R4dHT-{GNOdfdESyDUK>-f;dJJ7BRV`yt2
z(idP!JOk+0XWKes+B(zrHV_S?K434s)aw2^#c@z{_4$*2`{l9p8@?d1dEyP*rTLa@
z>?c=lhcp9q3+Q-V9m@f@99RI0WY5(mHS74tw!&(*JkN2qvvV1U?g^{Ivo}<jEDuF3
zK$x;_3aaYx1iF`$GO(fTLCN;($yZ#Z2Wv6u0O8UOcEdG=#}@5c90T$}DIW7KP5S3U
z)r0W7#VV`DzAm0amYazP=x2M26i2y3;@GW@&M(W4jLt(h%xuX7rZO7^q^jfVJ{z02
z4fmIsxGUCYN=$W*E+gTwOfrW~{uMQw-zN`f*14>b8<Fchcu4Ad2Ux%T<zDlcZqJO+
zU3~r7Na7h+LMvy&*<3#sw8l->6bRO(sc*B@%)n|aHqw8LF^JN!VKbj;)*-jVtbf(x
z`a`{ch#(w$JRl^$DV)BAH>~>W<k|(b$`~1S#d=DU>eHqOMBKh&i^Pc>l=IH^W2Vs~
zd*vH>-Nb&4p}BWsDu!g4;hEwnYLxEZu=4|Si*WfBJ9oZN3UHfog^f~6gk(FYl~@T)
z{f7vT{d<j~1%GtMDxI^%Xb$#ylc7rLcXmz>gUCOS>jbmO&t~WD@ZT{;pX^sG8PuDT
z{YEbJ!-n<D6U~qm8eP$~Ss?!%x}Z3LB^Ru^uVRWHpbRq?6n!F0xbyBd*RBEJneOQ6
zG8I>iK3kZm=!H^8QXalRr=%kCzYqAb#VAVg|Kj4{k)Z@Wz|;9*98C$w=kuYMN>CVL
zzj}{J?Ur%xwQ#x(Mx!g_E++I9ndMXDS~(Lr3PJ(V$0px%KQ(_na<9<ZS@|!P1tkBH
z=yG9@EUwBw%V;}V`)*^&3u`7cGM;$+8bMoY<g|QB{2V@}KR&kEqU+B+ZAffektNmb
zl+_PL^qtUR)D5tG#0@4I;Vv7XB|}>eb|At|crQQ}wuE`r4Y4P&p~DgZN-eaT&0m%c
z^lrRLX7e0H-X&y8+djeF9(9z<Fxmxn?u)e;!+kg=ycbKWFLK`jEOuh(6^?W{1#^<4
z|7vg$lP&{>L^GIcTZ9UGK;@2)bZQFwJ<7d;vadKfQrhk+35liz#*|dDLFz;j2~Q-+
zvR5!GCpH!dH8OQSIAzk&OHnQh4WOrCDL6XNs?w6sQ#O*J&|Qn@D>dr>l*tb^7!~s|
z-J&resgzk?dBWsx2!P|wM)k$}^FCdMLc07a*{lxfmvb8_LW^!-Q9vySu1kd2M^{3p
zFbi_P%w%d`@S&8(4eXJf<g;TS=eGXTC4X@$|L_wfDymFKZ|qUhyT&U$?cZ9>EsKI}
z-a;MRmcpDDzCQ%pc$0j_@($UU`{|+8JKfmmz2^8O7D968!8#8&0;LfqfpZ7Sk)ogY
zlv)G`qWh4%8UEYEd)#Dd`JSWpq1srQ-z9oIdtF&2SjLF$eGlL?#dkHBQE2I(qHi|F
zd%-0FpJACMut?4aA~Ir>d<yhFID0H)^bm{D?>#LDw8X*~H#p1YFqu$jiZq*8e^vVZ
zL@X(fQxx5msf?%26tUL33@}?Ullv^%6lhC?{^Dk=s7+4SCHD$Y_=!?TKCBTP=J(zm
z7XtP;FC=8WOajmdG_-kMNiIaTe6oH;0Ce*}1Xk6YqVqA1=;--=6{ES>WBT!2JNf&v
zo&B!~611=($rZ{smWcj`P(+v^0Se4r8s(Hvkw$38;_N-Qd}TX1(yyiTK}_WL9U_e}
zjE*IY6TM~$J78^yhxNo!XeCxc;Z_DF3FyfLSgJMu1N>D|6&3G_J6L<2T*sn9Phi>T
zT)emq_T;0#tBjw{_S54o8k1(bmq)}{Q_t<hLr|khNqvZ@X=_a~QvBbt<1QaS;h%en
zF&2{9A$hCoMhj6Hv>Om-4|#z&ow6N^6)=uJz+>m8uiji=^7qS*fisgHlT*rlgb1`p
zH{-)&v?5PUsknqNq9{7qh0}T=Z=A3WghPZ9^gAHThd5HKq`2odIz>`<zY9^`tcq;`
zb*4<Gv><T|k#BG|C4(J(v!wdR5;Y&zTa#X#{usgHHbz{T^_`3PQieZ$gB6PR8NGC-
z>AXqoL$?y`dj%NnTC4krgT9kpIBE@B-LEP4kfCI&M|cuRM#n$Y9<m-iMA#dYYoFf$
z5w$B)M9qq)vLZ8PA4-$dSv1EEh5CTwh_LC4He4qUc+}UG66q=qvS8bDmpZQk1P!Wg
zyY9LhlS%mtpO5>`N$Dw7v)T(ULs46Sb4G!)BHNh9`XrdJWm^HEQBg0y8F36b3mlwP
z2F)fv1RlR^{Ep|Eg3(zRpPZqhu2Q0mcewC_b#7x_uq7w*>$1z?=lq8UFN0!~K9Emw
zrak(#+FMUQslKlam7P+v`{BeTpGuF}&vdfhnQ^8UTU|C~t#n6Z!_LZ1J+AANus#Co
zp1&=pyulL{Nmj#IKJT@XbPZ~a4W=Oge*x35Ud}J_(3kDu*&M3WpMZ2QHZ9kPxBzM1
zRNwio>3^3Y|2vWTKN!}ncj;P3OLT79o{s@>rI<<YT7B9x6V5ndx#Qp;=JU>z&S8Ph
z>A5@kcqr&xI$|;fa8pHVcDyGBs5*`&eO|+q_JdT+PG?r4;@L3}2hk6*KBWGK3VuJ8
z0mBkMozh=c_-|3YDF<GX-MfG6Sb!+~-~EDFVwH_QoJiKWamr71dGouJ=4uHOWk*=Y
zr40d%Gti-T2PL%im5mt81`Bp>f!rf~B-R*xjcgr#6E!n1Vo@Du<oBA~h=g{KwU-|P
zB8O>*u063o4@z-ADEYkobK`Xl&w-T&7Sd_+pR7vB2LL-h>tOl+#&Zy+gG~vX9-8@1
zP%tTfOg<lKYH$1W1<4+q;qV0O4rIT82UoS}422;oySpe7cm=S@Ribr(+8GGEF&e)a
zSv+;jBKW%8$nTm$KBOL*R%%zbPFhpw4(!{49+lM6+kh;mxs$O!%C4f9$v8Lm<^87k
zonCq_6zmk>Ft#kva<1}d4GW@+a|Jj>ct8-VD|pa~Rypqeit`7^ty^Cxg7^CiqLNMh
zUy!N)@()}m#BRXes{NstaPEJw008la+#CP}zi>L>L?e$89*RG}Zazjj2&GM~P_bvP
z5J60A!8CF5*!vm`-M&ACDET61Lp#R6JW3^j&*~HIv;b%kaRE|K)sWTSn_C#^JOQPk
z`hPm$Ai;hA_Lo1l3}nZY{o4&7E`?*A0KYqCrPlz~<cGz_fP^bDh=QN{Yml<YZq@Qv
zd+#FLTVUE&KjHYdbdW^a?YU%Uzx^+N_WxkB{`)WB$^oeg<tQP8{r?b+|Br|2Uw{4o
zU%>yTcR(GE<f7d(2}c$}MUL?t_x#&{GlL+x!2JJcEy}fq&H+r(kkqno=T3=caoE86
z_W5kNorqA{$nN>-Iwd71c>E8*xvFw^pEPjA2U}eRv*oHHCetC|k!pkJ;JS`<Kr&PV
zz#l|DP?#rVek+Y|K@~zI{b7D%FA8Oedw{f2n*xWj=dDBcw?$4ssXPEGo}V3Dz>eoV
z=#5zF2U#SD-U6P$d!!|>R!@?7k`?Mn(InuVcB~B&`ygFp+5x_G3spcBP~-B9M{^Vz
zN|L{0Dlm_yast?ENVt?S4FfpGk;Ilwa{q1ML^i)|u)VgFjNU*O01fpo_mu#X>Ye%`
zb?;$#vk`wM*#d}v#z6Rr3)o=3^YO@R=L|3*Dfe1#Z@7T{(dqiz@~hhfqz&)^azFfa
zH2^VU-mKL_L2b&6N*@SbYu)90&CTBccSnY#bshjeDqd0GPWLw|C4lfG-z6|<=xK{&
z*mkf3C^6eGO<2&c@77LdZ>LtZBDo@iEOMMyfx{m@k`+vQ``KC!_M_Gt>8As1sl-Hq
zDH)Qri-%ksOLEUyo`-+PUbddK8`es@1w36W&`EPqQaJS|^rB4x9L1-ckVa59q8%bR
zE7*kMXRuoEnmYgva<SX8ad$0J8wr2w03INeMDBedt(L%G!J!I!=ltJ4^#3U&Xa=?-
zE82ta3{h1JO`v}h1dO&ADS`3ZSJNhbC(-Olk?xLjBrbXmoPn<>%IwW4V8M0S==MT;
zablp8=WOW)6iXE3zt#4DM-$<JH-HaM@K3cMO-=wkkHa8ukpNv)^sw`w-{^)X1+WLH
zN)jr9Zh`AnpwacVHrPD2!#a!gW=~vqL86b_gCwq&kw#7v6;4UBG%D%w45e1vn0qvT
z++rERSC=Wi%n5?b#=uYYRs7#KHR6oWaU7s-NX^sS&4<U5&U2=b@8-0gQB$&+!Q0pk
zK%R7(0&nQmGFCT7{U-hJu#xwygB{&f!+M+^YX2PaTr4BuZEm-hixMemyJOmQtM?{&
zzHDg7_5a_zmrtc2{y&+;4?|u7bK{=3qS;pihqbo=do2cj9O`fAe(MgZFXD8pl=@lP
z(QzG-9%N^~1X?H-4n7D}aXkiYMSqF9J#?aHCR&6_G9ACqy1{+4TYwEI4-f=>n7TW8
z5y0#zL1P@D7ulFMfqUzE=oX?45406sI077+@yFqckg2O3&^n~K_z~2OWI}h{@l)v-
zNqLP`*LQR>X|xzW+NJiuuGSc7z;sz%lEOf#l70xdcJxU6tLA$Mmuut2o@>j_irEZ}
z6Ft`q@^Y|k_xK-uLR!^{zS9ZzABJc8jr@QL$q1R8-%3am#NkV9)HYwQ(%ibWKAVyu
zgWPPX`QiWoS_`^gqRA0>^@#x@$5GJ`u0`fI_iKV=Aya&m=fTYvIbbyk?C{JS*|B#v
zZK>Eh(Jhs6Slzx+n61fU0@Q}?BA1yNsQ(aneZL3fEdIM(3mGz$u^K^svJ9Mp?H-r*
zDS=CTi)+8}sJbaV_uCEVZPY|Q264?80FG^U^YP;tFG1v>A^+$JaIEtPbdw_2##JhR
zsjg}B<i`dxfAqY>C*$J}+xtjg=6zb7KQrc=16<9!H%M?dcvzW{PdBw4`%vS_S9|A2
z+Q>C}!p{@#t^{qmkrv0R8{pZQTn0X@ZwxK~cv%d|$o_g=b%+cy$O(g$0I==mLvB73
zg5NcFbY_-j0aQ<A8GON8SnD~U4QKbfKW*CgHa_ekBSey;M{~t|F)7wKB;FP4#4U>0
zTng#@d61WKSg3g4EO8Qe&@RNwAbndqkUb0|CNg50Eh_mGBj9e^4xr>b^lnJru_(ve
z6#!x{%#Vet3L1AhQ_}i?p5_;XzVI19LC?l>zqT?*OvM}dOds7xcRq_)=kNsH8=f3j
z07QH!8fMYB{@5NHX*${gQo-NnYnHl@hbkTL-1i_JpB^L*dv$l{|Jeu8az)hWa%M74
z?b|NlE1xQ{hpR&wTc#hcI9wa=qd(odfcTuX++Ip>GQ$vn>2N2GJXahEgRlKr0vy5;
zw$TDJ*&z?pOvGZ}#Z!{Lm;^~Up^@+eU#a4-VvDox&5qHnlW#Kssor4e#FR!$e4Zcn
z`NON$vvme$M$m<$w9}k<xfY>S7Y&FTFtBy^K0<Z^FEHJA0sirqyO97PcvV12E^v$Q
zVfXPWw8G)*w+V#y9-v7sKUh4IO8|FRy@$WqI*t1pamxn}F~=YeiO4@4K*+k&KLl{~
z5_sSTu8|<b^&TfI8f{pdS~Ai=j)1SwG=vw!cziP7lj4lehP{j|GK|l@atE=YeXTdf
zyY3C50iZQUcIp9&SRYuDP|P_w3mL43Yhww0!M#i6d_?Gb){&`wu6I>Z)s~m)@`@05
z7iH#cm4AV@7ic9Mm)%{|SjeM(deQlY?PY%NX*r2<a|>YDyFJ)(ZMjTcl=JkfZM$6S
z764U1&*^E*{y(5c|JAdThl+|vx(o6G4ih>-q(N&3a;*(qc=6Rgya9@y0Lv}S{TBsI
z4w87<4g&Cm3qE}va5Q`j$ly<w!QJOfH??DUDr{7OAB`(r`n7td(22+aoCSOUHg6z|
z<o0tzay7*+kz7sdUV$sWDRkec52n4v0E)|nyl$KE?!!Oa0^X_@ku~^ISH~6bQO#)#
z=&a{c22@fG7a;V`r-0Mf{_-t2p*x2t+?xQE{+I^FX&45x0=d1TKc-i(2ok^Djx1Ux
z%<u}@kC2P>C=w-fs<eE|BFGM_I9CI>Tu&+0l?ka{e-xQd_UQ>V^H?mq`30M{pK|Ih
zVzRnhj&+VHErgLxF(3{Chp6hO3F3ZmsX><w_C4Y^;K3B31db@~fXHYk`ML56(ND%{
zWxEfKgz?d!fc<W~z;|Y1LVz~;dL!D2Du_@T#d4>!%1+4Z8<zBP9oLYr!-7k>%3!Q%
z2J0)>pZQzMnPM*ivqV-TJAKax<C0=Yeoo5;t7TYl6_4yDZaDXvX<gn!Zf_B(aP*1i
zHN@|uv+#yU(BcHpwUcx*cdAexCfBNpnE<~(YK=GoO>f9;+#SE1$DPl65zMHhH>$uq
zn+x>m--b2hxvO{(w7|<A`ksR)dFz~FSW;Pe{!XUJG!?)ia@eRkHv$&~))bH(t@mTh
z3{%t)L!Jvv14~4&EsqtFbN+T7ge}vD-yj$mBP|IxC#CwMQkFM2>oAzNe`z<qLFQs$
z*p?Eup!+<*!-LZmI0zz3!1g<3pS{<v1)FiVJU~@Q7~D{BdT7gHy#Z+Nl48{U=63Pu
zeqP#r&t)ved5L-G19I7OQzSMs;IG`(rwN7nXqY;^0H7RCS%<zwNm`GoNu7f}z|;wa
zzCWv<qI601Mlr$B3tCNVg}AXRIWutYMg~y(cw&NWk(?(ogKIQv4C9=Il<Y|CZru~b
z&Ip~jHBWVM1fXb<`$o4M;0<X?NEWjSG$Va<np@Jn{(FB_u6sl<2fRmAcsC{^XeeY8
ztmIdZg^SU73jTl-EN)XZB8Re!iB{grUDnY$32XwEvy#AlK84vxeq}0o4tTaWz)jJg
zFqjGypbYF+boXT(f)p(OOk3E8T!Th9V6?<JbD(nr)6ZJMU%5v@zMLSLsY!zWB*Hi<
zH+)})am!uXE%_5rWHu*CF1+pkdEFRcT$4A7fcw5fkva9Q+de2Bg{2|ET7=EBpiR4t
z)`a)#FAcU|s@DLx(}}Bp7q~_OLQ(11(Mk5pK0ATW^c=#l__=M`ko1|`9xZ&XR|25E
zb@EQ}ckBt`u0K4sXngZYuDtJQV}}O~=`B(IW56Njkdw~&80fwBzEe*I4{wG!$czr-
zw`Q7rb4XccAJWU&ZZ~h65^c59{YpCRSz13t&<e7Xv{n-cRq_M%wDOX!Sx}i@<AXLH
zcMejt^3fQ?qfw@F&U5k=tU$j&*7ydH0AHW4-9h$w8}(baji<TK)TpK1)z?Y-!TBq+
z_$ByKz?-jk->IvJX?aMCrj?K|@h1xt0Z~-VlnE|bd0xd`>*s*loSV+D#{hdAs`=rI
zxTH=Ws-z6pXXlu~QL?xqqH~g-@(S4tr4lQ{m^5A%rBQL|$1<%t^Cs>Z#plm!SR0U*
zk9JS+9r-3&h(<KbHyngk`A<|kKd$v+<q%(#t5CN)B5aF4N1r-ZqzPM|NApq&)yAc3
z2(+yDu#ZK{5ehCrN?T4%Gxx!K^-rzZ^wWAq`$eSa#PWk6btb17>LEU^KNf`eK7D&O
z)yG*z6x#JgkUs50BZv+dI9BIBYpw~5TOVppue10P^$-gHfTW?Ceeg1?HF16P;<Ee<
zLGU>-E2+VdTUWz(31Ze&g-K-xT6;Cr-zyDavt9e45{H*P;(xeHJ!yaE7)pBwAH#J%
zHfH@M>F4y`Qc4r=eOjPRTv~9XzUWUp+!4UFD<|VWZd6zt6rWytO~(u)mi$jrkbi%Z
zz_75UV5(J831=W<Db2Q54kQ`uQ{hu0VJIn=%CnS?wS)^Ef~zHj|7_eP0$Xe46hNIV
zr9WH3{YEg>XNK}qk?trnejYPlue#FtYhSo?^hjO3L30`@HV=18d2IZ^_9J%^%(Ao2
z4pVswYU@5=7k?}oze*uyglNIF`N%QDrO8(m^;J)KfCx0w<}Bf1UrlWi7`rXeQb#Z(
zXCr4x`i6SrD8rc&Ld_Y=c10aKi^b)jK{C)aS-sGbU9Wtq2c6C>(|+dpN*>fSN^xRh
zymvT(1U>TfZRewfc!mhO>S=T8sDV$~eAaQcZDs!ZmT~7-SA;$X@anf+RF#ILgC@@p
z)fZ=(@*+EtQ#Agh+!Er`X_z}mNO;9;qviUxf*v3l#PZNOHy=y-%xa4SEUu&<v@?)K
zqo*GHD6rWlbuKo&?<~xu!~Cfei>}Bw*A46nW5k_)xvH#a#V;Lu7g~AyHO?1{x?l18
z%bPV`0?yWhQ_vW&oJh~Ye!+Ilk9UG1DraaMKz^Eyot&lH5{gv&W`~$&h3I63e0X%7
zPQc%5?Nr7Wt<C5E5)pZu<Tld(cAKv^I%vu%Gwn3jw9YrV+ZK>JbFxFC8q;1L(>N1B
zO)uTIDZw5(BHKz(`21_{q42quN0jd6pIT$3cll4fub-Pg&zl;pw_VTIjXYWfrLXs+
zc}>zp!HJz)Z?VQ-vix<ei;52dd4drQo66+(2WRZ7awEfn^kzYHpTT2tIudOwMy*Lp
z<ri!LJ|J={_5$%DhCuhnuiLLX1Q6Z})O_V{lU;$=akFI%ahA$mYmqdUfeT~4!)#ys
z$}3@d`!ziqz|a{AJ$pL$G1&tvqciRwd&~d67x%Y8M~;eW9x-^w%bbcy99)qMUV#pQ
zD0~<dMS+gy#6K`QiQhRp;eAS$6(YNz<Q&)6|G}(P!acIhRgZM!_7lXSv<Wa=U)ATO
zFaAwZHhW6-J5Kq-<VM*0s_eLTUOECCf227T3u%`tRz@Ml6E$FpYDwDSfpa-fXp#cD
zied74oI4d7u9B788m7G!+t(dV?3n$+#=XR)eftmiB0j}WS9nFE?6L(x2s1-;KNkw`
z_FzyIGjo=&m##7;+mH+vs`hNWrD`{>jDr#j-KCO#m}dMkNYs}t?U7xwbG{{Mg_pXo
z4$waJMVZvTwV4DHoFjlo0ey?3@VpZg^vr5JXFgoN*!G`mYI^jO?;m2B7RtQqcuu2$
zDyg?8JoBkO&E9KUGd(oe?7~q6kwnRkv>oC!)trwIg;sywFpn+1hu*0f?9}J3vJ-&!
zKxLKxrrOMh!7;zjF@?)(M{luESZrk~V_kmCA^bq$1DZ3U=vX+=Ykqros~%xb0wG-`
zoCN)4p`moA{LKMr8@`={uBX%0fGMQ-%xjpf1~e8t>c0wu^tpUJU(8FHs9nO0n126>
zbqa@iG2*Fk+}}f@2)LI+*^k$g8-M<>o{;k9(V1<#-KC`#=norVr+%6G$F!4bn*+2@
zW%%sWEU{r>KmX}*^j~&OfLWl9`6&DB+4E@jEv<L#@x4y5?2ok!sT0<jY$Ye+<3xZb
zMpz9XjbgLLS<Hfh!V7x{afdzA46e!$2jMX6dG5(B8}2LDQQk<SWIPY(RBfHU{@#@R
z+7+Mg^%GZ)Q*K;13&o`j9)|l|TW72Nl}9i(4s53{cJoA<(~ZWXHt*H%v?EIuyn*Km
zXTL2Zb{2EayyA{|ihP=jkAUV%PxOgN@_PGHZ-pZ}=ENX=#el)NJ2*#Id&V=vZ566+
zPX7p76~+0wVmRixd8R}j>rdF3E#n5-0eo%xkV1=j;8`SRuVNBUk&IS+Rb-+hcbQGS
zF3z#p-7)MqI{uxc<>JaA9KfDa5f)t``;{Xbx5tYDn_%=C>(^=jm>a75CffC<qfmd8
z@c!<`jPMqNg^z$`<2JMMwIQ#ix8DYoc@S0SFLO|Za`WO@+%VIv1?w&XG2Wq0(|WXa
zk6pW$B-@;LJf5$6n8{@8e)YQwMZ*|4<2t{Rd)~g3k6D(IQAI3cIBULcR*cgXC=>C`
zqxlldy4S9YTU1tEN(I}V7qdIRiBa*QSY;dDV37a{&^{xTsGq&1NzGxkRp05u(6<B3
z@06Vg#sx7pCz~l6B?+Zw3`O4Rd=jVuI2KK&3NPna2D@aIcQQ?4+UFf_b$pVlN%X&;
z>JVH#ws%)5!{#vFtJ@$@VlS7;;81SLis`2+1!Pg^#(SXH;=4lDfuQjA$rTtxIKaMG
z%)BwH0iPU3W!aW3OR}Vf`238;mh1LIhaDG#9SqxWTx9<<Z1|ztzcd5?cA%wv7Zuns
zx#<8xpZ5=fOd5EBU}fOCbPO2evFL+w90Nuvku5h-Bn$@1pRJHE{|#O9L`Hq!1J6*h
z`=quXzez@jf%?@d1RjI5C?YgWj0F;&t2=roV~f456}l%_)$Dd$k3twQNE(`$f|l5O
zc<ay0y@ulC@bL<Ogs_SFCB_|O|HJ`g^tInHGh0CpOHM?j?|t&Kjs2)f3>Ag0owkx~
z+NJSLPjNoF+bmDgS2t~Es-A5E%a>vn#kw!gjmaa|VY&j8Dyv#0e4z?$#j<xQUyO5O
zH6fCuFn4T)j$tp`0(6J*I=k#5QJexpt6!KVZ&Da@?8_@#Dj}Rb1H<079K7em%Y|@R
zoR8dVhcNHtM5Ub1XB@uqqjhOC`*8sw<O_T2f7V;`jnPb`84g9JdAKTI02N1U*7doB
z@L^Pp_D6H#6Lp%3bw7j6r>O-rlMhRK+eYL5;8(L-lreUR4byT;wgzMM#F?G)P@!-J
zm{gLG4+Q8+wB6rK%5!YcPT?y0(pIeIau@Q84V*4hS~oFpBO9H#y%-)_Re`v*qF|!y
zMS6k5V{v{24rrZZ0?%}D0mjM-GJo5}SXq0n^!VODKtb5L2+9!h=!86k@L9h%22I$a
z<6p@7z&R8UNLJ)4*|%d**z9i>rSxv*cM!;N4<EsBgx>6XMD71D{#4FI*J<vR;k504
zkqYE^Bb|PZKqU~HX^yS5$E_qDyU~xS!a{7jROdBynt%Ms7eG-h`5aA<hHTj(RI9*#
zYrFrdT$x4bXVSe4%L(Fa-IIc2-+cH#B9!V)k+2;+w^_eZ7cG>`&@Gd5vWvvu{2P1H
z&lby<;#s;(e?3W70S8fNg_wycqH>U5PZZgCyiroR8|kLAEIH3Oh*ML|-;-pv)Ah6(
zu(RfAfgG=H)biCtW%BLdeKWpH<sKyrgOaGm@eeMmK%unYI>WW>m@F|M9VM}9A#tfY
zdsvtXaNWbDRNahzZ{ujRjP0F@RWH6LF5=%WO$z+By&$!XCK_Z`WRYZ*XSotz*i%i#
zYn<AnQkgUDl9I}5K%+oSUz)+ryJLO`tYM<ezx;^bFR4T?3&9v$W%CtTe<Gw)@i||Y
z_Gmqf)l`LxG9Elp-+8`{&tt`4A2jWm#t5=q$M3+#YUdxRPRZa$sR01=Xg|1H0UzX8
zzXJ{~w-2SER@ksg$kd;L7uw)vwe49Jp8?FLz}RY6Qr7`gm|2rIUmp!rO&*FOcV+r=
zkKUOYdf<6h0~)|*z*^{T(?oUZ-#ZO_o47}NYv8}|DSe2|2i>oTDJ#X(6?~*3GhY)5
zAtki-Oo3vp&fOyZr>?ECGy1?ASw1DMk+=~=o2Opf$!)nS_0sdZ(~IK&i@i4wr?PGP
zzZWS>l(i^BvP@CRT;_QeAw<YLgpe`wn1v;?gv^l;naND%c_^_AnKRGx?01~kec#vf
zyubT?uIu^h{pWqR?b^0%t4mJjc^v1l@8A8ie|HS!b%wilpAVtm<Pq3*BOMwNBf0am
zzM+HfJy&sl)os+nZ}tsCpv6Fc`IEZTQ{7#gmG$af>)szfoS(0CIrb`^EV&wdQhW9D
zS!RRh67Rg!TE<<Y;+C3;ce(3eG9;?PEU~T8=UvThFWZxYO_f+SUHW}8ij=!YCJsZA
zon%Q0r~<8~7Bxl{rL8mV+G3nFIxhxd!|N5TUy8y}pg6(EOVTMaNxp?h{2eKq!SyMh
z&PD2@BjA%S^)pzFxhpvW@L3kc#v$QT0%{tvYnu!Wm%bD9s|@Aorh{08-IMu;-509(
z4853ljJu|9_d#UJ`p&b8KNi1J`(8qG-quR`7iQQ;8dteZS{Tbu=g)>qdNmTFR!T0b
zdKmis2FfFQFhXmN-}_19ml*dKi!`ob&aS~?FR~rB#@z-l2to6E@ZH@{He25ga~~>=
zR>|hl<~ih7D&~EibEDOwXt_*0K{?0Yax&EzC<`Pa{R6rSGg=PJotkgS6}-xQHi+ps
z(?q3&FgQ=2S(loJat~sJuDO}tCng`$dBvsM`(jiJ&&RcgMr%2e!03ITC1rkxiOTL`
z7(J}Z^2`-zWqzq;*Tlv*fnFvBATxsl&Ufyj?602j#eUWO`GQpIq7Q0`A5+(PY7in%
zA7*kgyG}>|>x`GEss@w2tE_mg|2g*aKY|RY>7<yNGuo3%pma3`s5?*{g4sYu2zOU(
zU6BjcgPOxY3;}~QSjP3n`?M(BVGf1cLOoO2pfdQ$0t4%l^t~v4g=*&acWj|?Itb$(
zJ|ke$HSLXWWq^KehTw2BFY;+_OuR_!b9%fVP6iG(b8wt72NCv~xX91CIY{ttN!`>|
z{QOF?_(?Lj6WGGQ$03k3Imx^Qer1;3H%@kJ#4)8B4}x`n56c!~?573X`DXg?<wKut
z3TKE^c!Q;Kwb$)nMfK~o9=mnWIEtj3RbXlo^gP}R6(X|OP>rvSIq?Fue<^6>sN_W-
z7Xs7!G|IBOAFJoW-Ntnn-)kSqCWEmNPU_igyLvfh!amrG?1iX2?zG+*1(bxy4$hH8
z_CWlyuZin-<7S`;lP3Y_T)e3WbW2HRRFiACiMX<!>GfPN&|~QZB4k@*gbV=X#5P8P
zeS#eUihhx^C~%g@^4|DRTlCMeVmNF&QM70<9T@^1c2R9zIvh3?(R$)%8_sQ?Q3~^>
z;Z%0JvsQGlU6!M5O`iipi0>E$vyx2_^hpNsX5ThuG&13ngJJQMt3>I*Bes!H|Bp}h
z&ks`>NT5gT*l0|I+Jxt5Fw>3GuaHjHKvgA|H^J*>k|Bz8Li0#mtBuKTrji#(J`onm
zt3L5JSODpI@KwNnnuBu6l?0htU&Os$DKGS4dM#)-d$|LGqJWjn$G%^AdDeMAS*yTR
zDQ>d;?d@2W3sW`aF(-kDtID$N4j$~}TReb!HYo0wq4Kv4v>Jdc;Qd*U_Vj7XV-*gL
zV3r<P&+WQN&MgRY85au!bHUAb7Wx%=WC^GrVJ5+uz)t@_TcH1vi-Ll^WQ2q(lTxCl
zBg)r2JDOJkEdtfG?_4k%0GC_!%Y{?C{QFG>m)0`<H)g2$rP3UsqdT5q0`y~AAZox3
zFCo9N=1V#*b1#$|vX-MDh8xFUT@L&zTy08nv;iPT>(^J<`7rw%wf?j)q!m}cH@rXS
z>klyNjNT0tm7`)Com(EZ(4$o$#~+odewu{@7q%H1&r5WMm5GX6UuN)D8z_eI_3LP7
z6}Qlg<IQl!aOYH2=KRC-C*wbpI|T-3Imeq#yN#aOp}^eqVpQsBFKE))n8A_Ir^kCZ
zf;WD)VHKEZI7pF!rjM@U9=cFXQ}z6&L9SfjhAE3M$Vw!s7gKiBTi$(y-3c#bVvp&1
z7J`Fje{L<SX+VN<10K$Ewzab1DF5v#qYGy_LuHHlB^Oa?g;ceTyzk9PSKhfnJU<eZ
z`*t<5^#oa$KknAlJHI)sz0GfFKlgy|{(e7?-5W_*@tpFW>zw%D-cF<tDV+zhUyZC<
zHFL-}+SeHvu78+1#a{WX{7Gmo1r*u&yPCTyaxLEy2%3h>*boY(#vbEb>oqCN$#tMu
z0T=YX-o6QgWb`g?6mF&Q_Ru`eY!&QK_J&4Qr2(k*76yi-01WOb-8HUcH7$nR+M^{y
zMacw4YxMlP5nf-VYZFu&5C0A)NDW19AZmEb`)Icx+c^P&2n!0*SStcg(w?z<VWCvK
zJsy68WO^=!+lYcW2@3b))vw)>z3&OtyQB^0rc&hR<`Wj?5>}_QCRpy1PsS0}SOL6Y
z>WA>f;Rrg(u8XWw)NEmeU{Y(TiQaDFj&1y0!^LO7y{fB56K(GW#N7{9*)BgI4zyys
zR1(P`JEA1GdTl9|#m@#k0zyE;+J2p#BS8DK>r|W9ZmK&b8}GNNIkJTdz2WMzolIEp
z{p#+(HPLrvmr-zk%w%oqf*J9g9`Zu0YW^WW8~hZ@5qC?n38gNm5z|@KKD=>eNM37a
z0fwOFRNRh9v(j-i2Q%y}`zt_IzE#g-Q`&*pV@~hw<sqFA^XQ=(60#od*BOM&UBdbP
zmD}`IlnJ(+22(ZGn}pWC4duI~k^a-~j2EY>yM;%tt#uwNkX3bz$m+IPg5j)ju5BU?
zn?9}-%Y75NKH6DKJb|maAPJen%_uSsb?+>sb{^vlc?$+qzHvQpY`kh1qR6pkI9j_&
zJ+9*N?Ywpjbmz?JZ8u=&B*fBFDk9jTJN+-uHQ>!q1H>f?R7NZe#}wwK8gP$+ag`yP
z00%j4d~ku^Y1LP;ZN#^^gu^mWY<W+Z&|RYu@6JWIUy;SR(`urpRabR~Hu~bI;mXQ*
zO;rW9Pv@2a8yD9;3t5ZkzMSTkx@#8=<NdrXlflA(ArtEL>9psiIeaPABFkFD<*7e)
zXw6WnJYAk$NrWr~K8N81UDDd$c(n96$jW_s;z++|mkwR%THO(ExlCS7_4^^yzTC~h
zhdXvFUI0RdGz+$Yz~ikOmRkmkdM=Fz4Vms2h?cK)EWA!CsCDqQyHiD2$@i<^Nxt=c
zrOJga6u|EuKi|hQN7Z75a$nrB6soI;t*ifaQOmH}5$7{!Sqt;>336V|MtFL@*sEpN
zI`r`7F(vnl=)$}in+xYiH0RmdIKlU%zd92RiJx16l}$9Sdt$w&E0}vv^I4yM5z|@O
zWQ~UifXn7kvW?Y^6Y#E#TSsf^PjyY%Dhana*PnOKNmQi`nI(B2Bn0yJ)I1N|-H1|P
zvpADP8ei#{K&&0Ib^TsL3Z7h<9p?)kr=bB`m&Jj?lI}b*rf@A)*Ww45@Y$kMPT9iU
zG2dXlO|y-*>+`NR03$MG`(4O~d^ua1<K{%y2)0FyiHvIqLmMY7Ic$leraakTR@;{I
z<$6a}$)K;T`~G+8|51|p2g}lT7Z00gx31a3loN<sW}l$m-cZ=yc!@vZv6WHbho%<3
zhC4iCgLW#^RE}$u^oN@L&OCmIUmHrvW^YM-X^a{U4gvqvGv*!&uoWvq$-?-e>op%6
zaHxXT-nU_i%owZy8&v?Rvz%6U>G348CiO}rsN0!zFLn6$1zf3nL6Q)#dC9`V8|B*M
zYI^Dc1ir3Du~ATHe1;&3xJ~7|mXuBV^7FtYTaVdD+1KU>wlz;iES}CVi)nh<(|kR{
z#(rTbvg_}!c<+a4AV#n2KBA^2Gj&K2ni+t{VB!(qD`XoDXy6(%-RlQn`@OAwV1?yK
z;4qXZZXN|-Kx50xV%|Yl$v^fV@=p2J>a?i)Ww$Pd!|{--EzG}qXSO+{X}d+6yfASK
zgjk+wGpRZ(j}d;@kD$$-nwg86xW;T{{#oJO_ASDA(mTRv&o$rNk8*En<<9MGe8lf4
zp4+oa&L!72T&fLRx@t^GFHUl5f+4)s5mfNQG<T0eg?w71sXycwhtI|HJA~U0Q$#$8
z43)B@TTfK(PQ2(frM8^jH=$gg7AhK1PgK@R+0DgwaqgN1g<IljxhC&{E29P?(uqss
zVt(djPqd!73r~)R_}OBL`eT302pRFCCEL-kh|Rar^O|qpG5(&CrJilPak<bwaZK_l
zg{Wo<-PjA0&qHNe<n(j$Of7GL8Jb`sy}s0UOhMFQgydC%GP0CT0q>^6foI_dhq)=$
zb)|C-c&|cY?Ic!O+#;Gh_8{=WBh!0Z@(PJJc|qQ7&LOV(=Zcul!+!X3H6LeP;U(D}
zC5bE<(dR5dq~3lTk{?Yv1kMYTepAEG4t$A>sU<u{g19(i)y$4GGb(t4bxq$ljwP5-
zkl*o6)C?^66gKpGfZ!X?*i7`z^=Su>2GV{;Q7N(LPGSaxPw}|fitNp>NyKyO)7Ja#
zZj0`wKnT*9OV9bc75@ScVeg_jI7vW1-j!Fzff3ypY5hbzeGSbH7ZVyY#lkJwXr%mH
zIt|iey6KWl8@n2RT*M)ws<vV)<iGBIXwKHFombrM|5Z>Zr@%LGo`#3iRn&-VDNq)r
zn$1<f9}tllFEyTl7I&Rh={`=V!Y6YnO$&+Ues69(f@3J<5Oi$LYxTO4F1q1|Ru`Ev
za|X$3uRHddHCy4gww9Ly+^>HF_beNi-o^`buRlFVaHJgRu>0<G?_H3{{M6CiUr~$N
z8dV+NWjB@l;7sqopCZI@^kiqAd`@W{I%ra$aFQP%3+5HEidXcm<&71})pMcwZM2eF
zSGd?CDzs=g+Rcbgf?Ks;*rZM7prqNX^t;1B#`t4@q1a<CiDDfocVC3dNtJj>SwB+<
zP8YABH6K2=#qKN&BA=eB@l7@_k+ZHBPwhmsM2mCr4ZgWhhM!_#Jiahivc)HN^)2)8
z5zUWZlB7xFv9zN@XBO|j*UX`<XZ$TZXkz+Mi~0VxiT$IX<@YH3x4Rw3?qABf%iDC4
zA1!Xh#7}&G_4|-zTXyTzBfk{2IA44-_Ong)$yRbN<^1O4&D(Pc<?Off8Aopqdv#gl
zW)w3WHOkU>Uh~3`%)isaFLIJXx7654^rf79kn>3Hb(^PI>c+MlA9BJRXtxFC<M1sm
zprd~FRPAEb2l+#|iDs-55px+s`yqiL<aY95&YKlGX4PKGgVf54sJnb8p>aYfUK*{d
z!Qdsjnp5*c`+p#i|A8g@W}{&VgJ@^zdWk!C@NUWfxd7G3alYVx%G=YhhKP~WNW25P
z>)yunC*V9Ex^Z~-_pb(kVObeYx+VRM9Q8zU5(HNTpv90Eoj(eA!W>tr{Xb<+M-E(V
zV#k6@>aU<eV(N7`LeFKa+)f5=(IAxqA`~9m$$E`DL3+b?zz*Sgle=~OKjPGXe&~Cb
z0I_7h(Dy-_E=eZrCZ)s%J?^b1Lr_~Vx=WShU;p1i99{`(A#U}vHbg*+DUecd8lgie
z?|<6R#`8g2p#+_Bp-HklE*l03pd>*!w3~4X@XCVIuOD9gH<0qryY*jR^z{&&fM#GP
zfIEyHY|apJ2c@+g@OD4=fd%BMHB{{H=IUheUp}J8{pUCJ6~fJh`wB<YN|E65d<W@$
zGV4H?U%YYDlOz~O^KN%#_y=Ubod8lv!l9iEV+6Pm<C63jbpQGn{^J#a3wadB@$J*2
z_o_u?poO{S3!wc5v(+-AhF+=uD?m;)M)bL*{`rr6dFWGj@4fQ$128Oi1Q0#f(cTa}
z?4g}PjRDJvI*U{N`d==mf8F2zyU^yF9SFW(P5D{Qr~kTD|M9Z_hY$b%pa1{T{Rcr6
zlX9|EiuYN?tK~t~6U%oJ3mNo%Fx;ZmOJeFVOjS|HKQ}M<Vn6~kWN4c&>ix~Ki0dGz
z20iFJF9x!3{UPNRfV*=IcG>LEJZ)XTj)5-cDy{RgQ-xn?4)lF&1hdwoyF2+#1WNvS
za<n#8pAWcnbA##uIZiTH48%kcp`l_k<!7J=(6U34_p6XED3D-5sfF$MbblgP&kHE=
zSP&&Wb<7*|izx|`8B8B^b1pe93<a(Szd)M6dJF;p<0L`nGH4n!XbEwGMjBZG?WaU3
zN#Iq<iyRSt(zf()$gfsLq`d?*BWr>XCn|GC7cU;#fCMB-|Mw#Qx5a-)b`lVArQik9
zT6=b6c?^aKT6M*7@__nsY?W{#4Ubja6_gM*;41Zt+ljUk&F-&M@V?`T9IxDW2aPK%
zZ_;)$idq3&_r)5B=<ofhP4zeG2wg{=gimuSxR%_G86v*wmWtw9e{B*yKPjem>=Wk!
zvlys&UnYY6t0d$8I|=lIF}MQ@_jv04=W&LYMu9qH1%$%hDme9B(UGOB^su57uOsGx
zWzaFrH5UV^q?qcWCuJE{ueo@NP+Alv`)1z9tGrcU`pzn-1eFGn-X00L-(wEs*ueg>
zy0fF4SBu}90kO=pJW^`up8vUH!T{HGryuGHSZLJ%*j(5uK&_pCD9T&D(@|Rky+v!j
zrgr=_Rz-`;DmOWCe)hEnh>d$dp&#9tY7CtLC5o>`1Q5?n^}zfw;+z_!OV`(bN+ab+
z>+eC2%1rcR!;fyHer6WbD&uRv*q(q6?i=G3$zDf`1-|;_eLn2}xPLf5Rb#$90t@TK
z?<L2qp&3SqWFw?Mz`1Y<?*kxg^%LEx!5(lGT>SW!$WS#AekHV=68!lb^dwD2Mkr@&
z;Q<?^D%vcmF6l}z!h03uNCXbj4tgP-^GE0TCDDo-Wnj@&xH-)(=UQx<*MJ*13T8^1
z60(*&*0d~0hVWV{bh?f#JnI(jJNRtB?50#un~;*rV;Vl&)FdE;Y%+Vk5qk)Zy_lx7
zw}1QAd<RE-cLw=ht~~4joIm*A@TPJ!v@J5^hnM2v59l@kr$0)nEMy4exICyV<kZF4
zN@2*`$yoiIZY4`5QJXFNMK5nhndPGX&lvUnQX1UQqMSX+RL^Kf(4<lfL2EbYU`>r|
z^m8U{Bakz<+y0t0MtmAQUk519jt#u5C&IqoCo?ooz<Fby9`>lVG?FLML;0jeWf!xZ
zqAhZMGB1k1?v2*d1@lGYUHm`NC_$(*;nkn_Wx<kBk$4P1rg7jgj%B0fzDEEPe9`$?
z?{9wZKG0TrhZ~Vs#P)E~pIom1$jU#UV_#?|y1HH232$@-{_4(<hyXh-wZE7`aLp~;
z_Fnrf#P8_4iVCAr=}Rh~Us4~A0c4X3&&vh(35l-|0eBMJd-KEe)Pn5b8k~+lJN@-F
ztcKJ_Gpnrtr|A*`QYH$N51f&ECggqqN0sN%4ncCSz^kQ!4Q=EW(YhqG8$7|7cAoh}
zhY={L1~w-lr=a!2T*G8{rMUM-@@SL=5VWsM{fwIbK9Ri+xS{7N*M_0U!5h%G5^Yn3
z#;qewL8OhVpnj0@rDktL#w&w_?#)b}7TTV{WEyNcQyQ-W!#h8q(|f51l2H*h8C!D6
z>8vuLH%x(`{{D1c^t=$1vzjYkEkC!+EtF~AJ6Oz2SphLp8Q(1JxN#X6MOM(>#Vz!+
zX0wHgu4D}0;zm#}UYp?Inou2hAAH>dud+*{^wDgnbRg7=KYNj#@wy-{TK9K9%Jt_5
zZJf!~uM;n!E<|(o6&@wmQrVg^>R7+t+*87Y8=_tw3n@E<M+@L(ZiL;N4rYG|)v?}P
zCO86B(osob;U@(TJHF>UWq;<_FMX&jhhb^Iue5Ur8aLQPx^~|pf;l_^;o2Zb;3_X1
zqiGb>cMiHL_L$b)0X;A2t)vIR+94v;?h?n~#bFK9;r3^_4)Z;bI>b#tr>jia#@BnM
zlCS$)>(bIS_c4>|69D2hP>4U9{B|AL0BA9_0!XPGrl{I_v^&CM+UCB}O3E<vz*y8X
z(5pd6E`^TB6|5y~c{~aDE{NnyCOso6eEE6#lUuBn_)Mz5v=F%0vwhD_J`9X#?W(a$
zBMP}<mo~xG+M-9oJP))6QvJLGt^7NgUxuY>cC$)_zw;m)6??DbPP7TW?*O08yyRME
znRvTBmMZS|dPub4WhSX*&DyW3Ol9&gt^A@G<IskKc5T1fj-$^u6Samln&tev{_`!N
zR8RRjZzOns1ieD31$!pH9_j($)@W)xSN}#KD(h!nmShq2WJ{&+`<4(&(!npA{DbsZ
z-E`G)@4%m<vTt`SvkOHJl{{@ypvXRD^^@v94AKAlasS`Hb`-<KbcO++=YsxS;Prv$
zi}U(APddVg1-$F%R@Oi@ps#PF_`NqcZvqQyFm}+1U%>f>Bi(W+&k3B*=qu;)bUwq6
zJb_R~_}fG+VnaRfroG&hPBy!h3kn=&;&3#_74crs!a~zT4yPMlIEUoT4Wr3=U-1XA
zJ@eokDgf_bc1$i*26!9>3jT~3a+hR9GpYNa;82@n+d%L^X;Fypx!OLIf&dBh9_CAy
z;1t>LgA8FZO#@%rk`Yi}P`FIMJP%G?<MVy*b8;T^@HH70--2z>K}UqEFn+<9XXto$
z+006$_rZ@xi9Otf?OjZ_a4FH45@rH2-AYG-b*9`~4J{@e>B3QmriXe;FboVGYPTdV
zQI4lS60`S4O>b2t-@5g8Te`PlEbzTiC}3Riuk}sx1n0~*FRWPR+2WoF%P^SRaCw^F
z$_-Shc#c%LIIq$bHTEF;j1nXRh#rndb?%3Y9JPVASRr@teUcMlqll`HDRd=i-lxEC
zlf`WU9AH#v^N@QxYD?dWp++_FkLZPc&?9Dv$myRbyeM@ogt+}=iS@$8j$-ct%crg}
zP(xR25^!qT16CyX2@QD&jqhX_wlKh&jDo{7^pIc<Zf%hwq*ha=NV#?6xP2LHlrqS0
zbsXF_4sr&T40et{NP>sb9efp&bP1SRfa42hB;50LL{`ZATT<mAx4YgiI)ZY@rASpF
zjSWeEQ2ZqSrLlg}NZ|;&AOZ@n^goypFqTL}%&RG_ygj^*eBhwNosCjadR7wTwxpu$
zZa|uQ8YdZNsVPhQTCs;F*VH^5ic@my_+&6j2^rS(o#oFIk2M^Kmn8=SK7wDIHMcp+
zK{644cWt5$R1kK8x*^6{93!7((4f=kI#-x1>^=g}>l^6#3@;O8uc=9gM;Z9Oi#;IW
z%!K%?sdbz=R=h*dl3=?)e8RyL9=-3DimFOO*$Eb0ZP8!@D?DxK3bnw1QEy1d{gcpo
zfltV6B-lnjWPiMZw$Z5`^;c1FbeGlfs0ICX%Z0!l9G-GeA=;Sa$77dtsg%rWW}>DF
z#vrkeJwWT=>w{b*1@^>9lE$tdasm>vow5rI)j?ZY$rs_Y%G{E;()0*&mwB5ud68+a
zp|!aVFr|iS05a{hvbWj7OJ2;H0B1jq&>5Wp2nXK1hFO69Mm$X@v_Y#nRfE}{jDF-|
zC2=AtjO~)9@OgZ(ml=Zw(dzQX7$8Az;4xS+#mEDDHCcr4Ml$$Bz(vE@Ay!eoI#@-C
zCmdN%8)7Ia^nMbmz2oD22N4Dm9RezN7|>B@x(D@#3c2sEQ71Rvqe`QyW~0k(zfXse
z>}wWu9<j)*HwohhE9}kJc*t~Jr_q6=p%?3S!?Z13c<ZwQGY+i;p2Yqspjrv3RyL~^
z?mtvkN$kxLUMjWjh<JM5;8b{B>f26px$t5-rX0~hgc9PzM7!7|7W@Lp67SBHquZx&
z8(oZ36oGq_o)Q2qCsi!#eQ?U!nG$Z<w4YF~>+Mh-A<JYbH;x{PriE{LEd}lKw}gJ2
zXJ~2;x!?^N1O4VRK=6K>!=UR;^B1h}4D2;6&m@vR52WMYls4p~J5(<v_x@uuy~ClS
zlOU3Mim93kb-_>Y-BslMb*<%UgW8NVl8$z-l_JIr=#|Wpdd8xA;q}DNB&z-9MVlga
z^wAO_%R7NNd6jqJer&Y|(?ZDT?wkxM+K_jc1*FK-p&huh#H)|oGf5q&=pW9hpj$B~
z@7ZUfP0zk<(>pM|=4v@`Ja9f@pTS=yo#U)lqp$wV^z&|qMGlXOrSOt9u@!EMOOD|0
zXtjoL8pJuB?H_Yq?VAd*SB$v5x_tPkkUS1{-<gWSs(zN(aPNca;@qR}&{!|LX^0GZ
zD@#eZh`%Uxs@c2ABC2V4_HCKQ_+s_OTz7fUe3xNk0QaVqtd~o9Q|&{@B@XdqS}U5S
zw%!^9313mdZs^?&&`@i3@_1&_kNE1o*(rPPu2_weH1)KT7*V^MAlY&A?hM}^tjCS&
z*m$)MGk~dUe6gCIvGh~b*gM;KSx1mz^1|DwztNa5VpMdKhQbWw?A*+u?i0Z}(d36b
zQnP7anTECnrV>^n-H(Hy=9e~E%W&+!ya0H2-W2UB?zIWOlCF)UTKmAl24ktr9P=e~
zrzd|BT{Sja?aoj13Nlt;AikpQ`0Ju)r27@ZA<xFT1#OQ>s{E-V2}0Iu6+_8w!amKh
z4VgAZvbEV%6$jZ19v+h)!Wqa>GAMfv`+o6RiQ)s*gO9>GC*&2WY+IofZOwwxLqG6q
zaVR=waD&>UuOVC4bhB*}2=CC7Kb09^?UmyHwQYpwJ(0yWtj8u4;vvazU#L8r(8H*o
zQYtL}Tzlprx#z_IsAIwrgp+~ZY}c;Lk6ln!toS{QD#beKop$=mA$ew!^Wz6brSC_&
z!!PfR&Q6~zm1mZ;30L{os;r^Ne@U;o-m%Yr_xokL<9M_GH@04Z<=nQ5J&IfQk#de8
zl#*qG!%ta$;k)`cIp$k^Sbxwy>3v%#?pj3zeOhm(8p@QmS7s+%d*yC8SGk;^J)Onu
zbbp()zjPLb=zDJ+49QVKa87XEylVsi8NyDnLGA~^Izp(5gWUGG5M4dH=IfD!p7i87
zGHaVGy-QEKax@aM;<T=l>vB+8m;Tg{cCO6gv8PjQ;v>C1czNv$UsvZQzsPpH_jF-O
zMY)cVl_^UWPlXE~wkS9?-t!~ts1)licyH*8kq>gPomdyKy%8_+vhRO3Yt(n}q~A<$
z=B3IfN@Afdn^}pScze4|%URsrHh=HPKwILy<Sw7Wu9^dBw(sF%39s;~<bz6YElakL
z?c-4lCU4&CJ(*8C$Lwm4GcUW*5?7@uttoQb+qpI`6~3HWe=GB;`uu86MtDrlQ-$!;
z-yI*ety5qfmW+QRh#-7<<a|Och!-hcH4EUI)M-Rg&KSKLkuI@;rJ-ZRDi4uHCrbNW
z*DHo<9>>UBh7IGfGLC$>iQz!#m*W?gYf7gS!kcPRrPq*IU^&*$J;2RWT?9RtA!ES`
z#SJ3+Wn{0`$Azt6u^qQX$O(JKf*;MI@;Y87w(igGw#$49#DtDrFlctJ+|Q>Ca8wqF
zbl<eta>G;78cBZ`@G21_O$+B0zGiw5m>MObk`Lnu-^+3`E>k%T?sk@HJPDSzUu(Vt
zSL-x%B90L+_rWv@k3Xmg^;$)RzAj)TqkT}p+(^*Tu2~EYQ|YzUd@H4~KwGS^vrLJ%
z0G)*Wx>f%rDyzpuDOE$MZNh<<IHgDOw(njreIa$8HJE3_q}AbbWkX~iY?s(!zRon|
zm5{R3I&M{1oQH*QS<<T&XJy?;cX99wSQXuoo<P2!ahde>m6j!g(O)+83T}msT!`!e
zA}c96jWZhRyf2X3@2{Rqaaplp#W8qR7Np%I?OGb>emP92%X&6>)QI%Xk;$*$<53E4
zes0}#4%Um`Ohw_Oy0=_OGxs7=W#e3jAX{<U-f*G2jCek4uO8Sd=b8@HzDv&E*B;yz
z({uW%f#>ru%=Ls?LxD!IR`9R}e7dFOnNkV4I=m|V`r_VSOSLToE?LcOjRcoI_y<<L
z*@Id*p#C{rHX&2GT7&zA$VHH#T|Eq^;7l{xvofE)7*X^<6^7x<QT;qiaEr{>rTj;c
z)}-{u#8z4%(2Ei}R1MCP&(n24#}IOn3Xvws(4!!PZE9jLX&EG_0faoHU3Axem)!MN
zJXN)RQ+`Lm{lW3p-Iw6GyAU_zFq~)Uli?VkR#jG8tddU&RSE)YLjs5>Nvc_<STWw1
zNFgD;4HyRhlRWd@ksmF0QUEaxO3E3f%I(P<YfPMeW97S~!_$>#J2m2dm;8C2pWJ(g
zNT9W6RoAGdpr2JFeA{-rm+L{A_Y=#foNrl7>x=K5E&zlpr1;(;|MRXmTYZL$K3dj-
z%h+|gX21F`i3qMsehc`|skLAWQwqUz#MR++Cu!p@OK+r0T}+$(as!1!H_^TS_K0O!
za$5v%i7Gw()GK-}w1rrnl0{|=3b=-W2?Wni^>Sblm4$oT8`$pJc$)P|e=<X_3^WX~
z#t0Q+vC9~}Vb9v>M$29Wo?iVw_qgSl_RxjjTm6S&%0TxFn`rs`;V+LSF+n<8aDE5c
zy1R@w6jK^DBX(Cn6gPDY0%;V_I29udQbi@tFhcPPkUeDi*J#b$f#TAadUVg}@QgE_
zgu*t!v$~fcoe|*+wFo1w(%GA@(a?>;xpkPM4On0-zpcMm;W(7n2i%t^P^H>jTru=7
z@O$jE7!C5u!=Da1@_4U#Kh%02%cRSMerE0z8co+8?aA?;Iwc_%M>9ian(3ha%_1S!
zs@Y%XeNNlM_{7oyw;sDpd=ga_oGBsXLHD{_@Ybs_In33<QV=@xjMl*$70>UI?-ASG
zn|=eFOiF3xriIb^Iw0heTxeF^Z7U`UU7-0nqTcmDffDiKJIc5<EEGK3pjA8lbbv5o
zomgPG&MUG5e)TmKJ5cV(sRMTJG$k3%Ay9BgZE4w_kN*N0h~NGg6zV75p5?>+?AT#>
z#>~B!IiYSLqCC-Q$C{Q{W<^>$z>vS^LxYMwjFcKVrXF`WG;9X4NsG0jYVyp>w_W!W
zv(#ajzDPG<SW%%~1F9(?71Z3+!}9?s{F7$)8?QM3@sJ5|>t+7fz^p^HrFr-rztgv=
zQy|g3E>vCeFJI=qPGYcdQWs6hg4g@3Jj$EV5^$Y7NVrsC>3b0kL)f@t1fie)JgD$t
zdkwWZth`5HS?X6M@E-@Gz&@X=&)^hfT=YS&n7l<5$~!EA1wv^3CN|TtgPu)(UVYt;
zAK3<QYD!Q-bLKW3y#ZF^5&85e|6CkHgXA4l<;HlxLQvuBhye^8XtF+je$fZkxJ-<y
zflpCp0&mVcXgUudCr_(Yct`Qhw2I?XKY4ZT1`c?Im<#MC7#gHd1Dz6(wQm41{i(<2
zh5LVO&4J3Plnf#)(W=VB2lzsBx7IA9VI0JxJ$GRysBtic_Kj7>0M3UTWaBFnoJmNQ
zX9aExAHuq9(E5SBkTXKdl_q6OAZ7SfklHZCZN+<cgr2sO^QgkGG1h5hb}D+SIL?Cv
zxvZD2-qbX|WQC&v<TB=}xi*I})@r^$_R`C+#q+Kt$-t{$mR2#l$e&N;AvjlOHxGZS
zC})ri#<R3upnYV{f1LG7_%Rzz()t193vpnbd*iR0DM~KjV$KjbK;vRn;isddXKNx4
zB?0yWQ-La(N=fm*`kI!t*T-ulwFExH5WdC`JdABr(ID+mv(-@0$^(*jpAyuK(cFqq
zp)(QR(UkufC#~~jvzc2bR|Zej+R3#{wv?*ab%z~~4Cq<j8%Ml`4d|kujJM1jP%u|M
z@WO0WX^CZ4`-GS8@$6-VX=C;I<~X>6TrOBC7)2oRw@1w)&@$nMK1lkAjf23aMS3-E
z+yn|Y>KOj@(zNp|iuOgk16fZPS8c4k3o@hT1|zJiv#j_sY#)G$eXScC>deir$HQ1)
zAJ4FO4WfAE@MOeDp^O%Yt$?KDmFS~5Db1Ax9t()4IF^2s(5+}mswcca{w^_1$jU&u
zmg0v{5>q(Lf##qEP*NerA`w3>6}GmiT1U>8Kf8PG*83AIh4F@G``{?=ThZb7IyuNv
zh-E|4E#Aujk7uw7)z+DZfRVhc(xD%Fu&3TbWm_@-S$HA~1o(^TzdN*--BNW#1SU2?
zqcwJORFnHpGyDHoS24%OLudeF0~S+W5^qpVSOBpO5!EVqbVb?9e2*nYn1n4dHVZ2h
zw1#)qvv9?#Y3WmqX3oL!ugGvmKt`zXl1)Z*=49@7_wPgSDY=6IMV{eX^d*)RX!4qi
z<T@N9pthhe9tEBxSEGPZ3w;Ct^9DB^HyQqXoFj-xHu^;SAdWm>?(tk5S8NNgvQXDV
z|L<6RRP{OcZ3@=?wHLh%$@PN;6(D!Ynl>Ig2IGdKuYcO)u;K-F9tO)P9$OF5vDpw+
z6!QKzSMR^xlsa{}reY7?$g)>gB1QP$Bj%y}xgVF1rH=#0K(8zb%oBjiS<M64H4Kp$
zEg$^~u;#=p0+p~0&@x9bJpzW&(8dG%h%``7yGP5voLh{W=|?QWhbDNQ_b{`1j;6Z1
zaJ*i@I$H|sR^7ub|6jWh`Slx=wO1R^*M7|3Ssr3PYNR<s;EitzFZK$x{&F1G3bVWp
z+hP;?uq7D@9M<zzawrAyi6{_wK8v>Pi-FqDU~;ZYTEw%aPhI4i)aZGll)4^vDWY%L
zM8gVHfF|-?_Q!5mxTP*}Q}3d~5(6#H<iOC}=fb}}H>qybKNm<?c7Z*5!@kiXQw{|Z
z7Y$JflZ^pa2&r;2c-o6JRzb)>l$?sixP#5>7aPl`-0=zk!ZP=A3nUI?wGeXvHsIZ4
zoq9V1NMrI-$wz^Y={-?XN?dj!fpMT_aaS8)z73}q$V>YX#0jT<_{3M~;BQ?Y_kc<u
z6a11)2!)Djed`|>>Oh6Vd=$$jty2TRZ9MU=NLdtRC`!1p{nIT}8vK4Gc<cHjKDKjh
zP%CTOc@97E9|9UZ;rz6v#id1$qMMCtz{rU(jZI>gE=^&psdi)hH9lXpke%@RE%R^_
zuU(JUo<ah&5r<S3S@BG9Vr$V(75EF5GDX*-J@zI%7mG2g;_fuaV5CpfLJe&LM?wrW
zN6cjZ6%ipd=4L%_#1*3pv!Rpa{T86UX!uCd#STi~;1+L#MB1>e&(F!8St03ovZFUd
z{P`w9rQzyy14z!LK<&>WLpCkfv@ORQaD3GrK7sNrzx;(~?@3@FMZygdpm_5c!uaSz
zT0OXwx#{t2QDsT3SY;G&$THr*h0Htxy{=OsjgdU<nw4OGq7My0>n-6aQMkD))6s(&
zJ^S2y+~J?*Cxkl1^e|JD&YL?e6yktC9=H30KTfVCOCcsd-`5{Jyxl|C`>{IMX-Y7t
zEY<t|98jCw4B5;SVZ}Qmn6ih(EWEdWoGlJ$SkYSwZ@mbQBi1tps@<x25MvB{s`1cJ
zj(a*ccjo2ES#An?GxV7JL-3{;IZbQIyCZVZ6R3I4&wLeB8?N$@#S&1R06(+qbh*i3
zFE?mk*IR6=Ca()Vrf{E`IVg@+{IuP^T%HVt_OBq3l7_rc=?!TYT-Y0H7$^<SRC9`8
z`7g&a#{fOrbk-ackv{}XQPF^YuM-h&QdWTS$bW$}|5Nqoi(GxPP+2EiRh$iK7+4C(
zKJ(aY?&clnmKL5%<8-h?R}{4Bz|B!^y*eHSz_TW)f`e&DwJNS@OfbHpuUALRq_)Ch
zEu&LM$dnDiNj~`ArYs>iqN}Dh&OPb!{$W63g>92gi7^`q!EZ&f6<`3tHBns*d11N=
z+^jPD(7->@^|ck{-X%1UayXZ34LZ=uKv;qWH&nfR9!y>+1)AJMUd^7i!?&mkmu`<A
z!(;C1E-kz|-MmWOHG6S3O86QeWQ}rN7>EcA9`BO&oRhe^9cn%bqA&R(M>}7r^?{CN
zX)gJpRqf&2{r&pqY-fb$LM%q+#UN(<sdG*34uif9#)K_`Xnx#vo*I(1LyvVnK6a#e
z6h-ugZO(R*s_O#_Dvh&5yf<06TiEq|aCH~2yJUi!xhy1Ty%tuf<1@7m=SyKpDIw)~
zm{3XE&NF5!xjbIuPN#76+U6#CSf^+E2ujO3!QOOLf0om6;jx2?*#dbmYnf(2d#}iG
z*BDhMXINit2-O;MKLCtxI}l6z17R&ADv*iDn9#`<R8;X=G63j22H<7GH7&LfJ2t!r
z`x+K(_LooN_MCkED`a_}0bO+kRNU{A)5Nmkp2iop^Y(T(i<E&{>y6zPY^KbTs8fa<
z-QOw!ambnX!F}DrcUm&L{GalGBf-DI$F4)TX#`Z?P%Y;Tiu|7KMx$NiuA!%0glq^x
z<MhBNo}*!~Wl=|Q?pUIfE?3mu@v&#A75n^+Ub?f%qy&Vg`=C_8(vd$P)Zx*9l{vf-
zId<XYLaaw6Z#_rdHj13j->$*G!tl&--2(_nre{I-rFtgy211;7)+Z)}c3#90$JLpK
z19N$<TwxtP6VO)g{`N3&7k4?p<juUvpf1)q@kAEKkj>=R+eg;ln6|QHQLB-&4V4Ww
z6#=i9Ev1NCUk0-j$!*0%O%#VzG2YMJytUm47310T2$hS<49>PAYNh$@!geoOZmG6=
z#~}uTW?DiCEma2JbX9b<g1Sc;=cgm$7hT?uN#>SJP(m~GGkvjq#0uk$6R>u);#Eqn
zTRV-jhcce|X2KDHx`l%dNVkJ$4c-{UV8_YneT(lO3uNkhChG-gE7p30B~jjirmN{%
zLl;~NAB(KUf*5}F>*^ExzmDOr%J_sf{ZS!m!A3hOD;TP!bU%?&h9eO}&2_7?umwYQ
zhfJS{Iq>S|<oPk|;y`?JFhhadfmlrZQR^k4giH+HLpTB(EqOE09BD}V`Xk_jU7*IM
zDGi>*U)u&#aqMu|uTIseo4Dd=gk8LK*>IO++6Zv{G6C*Ly4V3zQdT9aNqQE$^}AEo
z8hD#tV$jo1!Weha%~0&Wgh#l~MJ7dq%44-^+gtj14}N9lWJU=2rD8tnpkT!eM6BJC
zS7K|DTm8#XsF9xto(7npMD=S7$<%-hm9>f0gWGC@dvbH?>(1yKje*3(5}m591(Y5O
z9|B`4pE$Bvx|@`)mUCzwk-O5e*ye`@Yy>!&CZ<v&rN75uUL%u>kc#qq;kIc_-l)MJ
zV>6#B@NYQ~2Y+CHSI)Yx7UBTHS9^q$>ZK-k%&o;@xEpen4OD-{Jp%5~Gb^#GQ{Y_#
z#Vhl2bQ`&yxdXET<<<g?)+!JVZ&4bml@QRINxj0<Ac>4=LR`Bp7MLdX3Yr$KR=0Yg
zGrL+trn;_UQS^f775EwAGsn8hbJg*g^1zx>XzT#(<GZc#XW^RtHTmzV7`I)$TzqtC
z<Lqs0{7+s21*Je~1F>BOa3k`b%(?-=^IO}7D1+`P!zYbn%{tBTew)6W;oSJY=`!XW
zh>=!L^*vu1u10t9o+j-QEnl<}PSPG_+WHJuJF?g?L~N`jp3snqe(Z&6jNRAxlZ2{3
zu>5Ho2IUC^9c=pN^w<=}0}y0WH?~>#aEQIy8U>KYCa3b}cNpJDaV(;05>{uhw@gj1
zjFV)K?}}+uH833*5?dnpieyZweX_Xb+9;Y5wYn=mp)|D4uXc{e7U$tFWfvzeYC>s^
z=4$rHf3Lm%``0B7xK?m4fL@5h8(T!EysYS3P`yl#;vQq0=7fwwWI%h-CxHYvtA;5g
zKC!L~S^`cUe1edneS>M{$1q}bzA$xz%J;Wt@*eC7h1KluzwZ}_2XJsIVf^nrP28Yp
zxMY=x#bkK2?*L5+uh76Fk3@mT#ey_B$e-4QIb8>1+~KO%z;HPs4p6#&zqicShPCt5
z!uKb<>mC#HyxZKur24r<%e$7yW_wr)%hnk<ls`$kS|aPf9KP3PG`hI;bF{7vA15=A
z9Q4IdRy(wmVdQY=Se#@%A1<CDl|G$S&Ba@GNowKS#D*fSXX*{j_DS9h5K){kTdDUW
zIeqON;%Q;{;w{mcGp-$D12>6ha;?7g=S(8k-EbDCcTXwh?b3P~hmF;`9=hm!X<2@r
zEtDV_+g!GdIyUUYhAXPDSw&U^Y?6#NT+n6VF(Wc!>9|MdWjJO$mKVLD_f#qI$zp|e
z-jvEU_IgZQ)Sk~mlq;vOY-~qW^5N|yVxvyYJ0UQPYr@{7;LN2{OmuQ@p|KUi*`}1h
z#4P7-dh%~o%Mqz$NS`*6$!*n)Eev~jj-ap295WN1d0A2YLb&%vOsQ2yQzaRW_V+b~
zU%iRJq5U**fpLu#_cr4?v~!-M2^$XS(9HjerqeTy?YP(Lpfz|wYxPa3;I!G1fX8&5
zFU3rj=nUt2)^lAocG8w#ZeR_ihd<XM9(ZOZJY)xSNUYt6DmO%!{_TIh@o-vmEzr&C
z=i*l|4EJ~^m3qw=kOcH_>CKQpe4&3x+rD9vV#~f^KVm+&5B%ZET>MMj@_+HK;~)!w
z8RfUVU(Wgr(JMd)W22{kECtKAaJfZKB#?Z*@d#A!_AjwpO7kxLqcJg6mlQy%y1Xfu
z*>Rb0QbYc1SNzK_fR!EE;--M3xyr#Z$mIee+%Yf$J2#Lj1V+TG;apWO#Pd}aJss!)
z%U+_*8eqR)JpkJ*q1g#_g&gfJKb?M<^bcUzoE^}@=IpE)_HV>?!FJ3|ez*45+3G)U
zvagfEB#>l^)5AqZTEc1cvo7GWpD6<B2|@rzLE32mHJVOU);`jOsLPvDsp!`sYCv-M
z0=n|bfJ^EKgp2`7F^qjt?GI)*m6`<H3u>PDH!Qf_5Qj)9R>|_0;_|<MT5A7#Gy$nq
zM>I2I&VAsBJ9_zEt^q(D@4D*V78Jp0O0}sy{B#P48UGFt2j&ibdB*z(`aHiQ{g0pQ
z2|WeRuPO5ExE7$a{V8Zc^IsV8Up|O`cZeFnfjr5@8L~0|Q*it@pGPo(j^%hv*k4!H
z|NRI5^OIC+=+03;h<bD9e|78r{c8R{a|6Jt0fdtYNH;?vgEt<qaK37m{yz!fLh@zl
zf4@ca5Ss_%0c;-G-@S3}Fxdh-#9<J4TE^<w2Z6#>&4JN{@Rv2a8qjMW@^UIZJ3Z#z
zQ0(}lMO{0r9#q#&^g;8Y%M>WWnn0#Boi#c(FC@fP`|*O5!14Hm7BVZrf*w*G$JV<!
zxkiw@D3IxQL;!?Y_9+`52qHH5{!fT_x%aX18!;+Iu&k)8DXuI0<-#^ty0ro7F(Y6*
z<pTbSnnWyLuLc5p#qluY^)d25qA336hd<i)L2?zeDVt{xEeTuu)PrsS2u71;TpX(?
zhH_gTFwpK=4-gcW_rPnAEbNdA{2s#~S{q0kzZ-lx)z3+C5`>$DpgHg79I!_uW4%8!
zaxcHyi1fBn$@Ql?{VzdYMdn$jr}0Q^Ri#T(min!jWvDZC<Xm=U@(UpM-+i(Q<hw#3
z-%S9f=b!=P7BZ9bDc)s%R|VRb6yr@m7+wteD%%gL#2(@a+yJ{$P&@du$bWd4|9Xpn
zzgms22Qcm`@DVJ6C@bBebe*XoZA(@vK_@FjE(>rp{{6D^^aL2<nD||~a(TKAy!{0B
zPu>}hGHz@%5|4rw*%ctml~&BfcLGP#k&W=$GT`$?cG(MzWf%tF9c7sJ!4y8IP^}YD
z)l9#+X9X;G*v2y1=_8mpOgpYY5c3Madtcb)D`L(OVa79{tGh}VScfi7FP2{<$P(5@
zl<EDvi>K{SZTapp??J}N0nAW4apQ@w?WC9>AY*3Kj=JO~L2U9BSlRYKXf_ovFW88N
zGElQF_3&V^Q6N(pj9QYull$wHaF?k7<j{D;u9#T?BY=@sg>=sV^l`k>X?4U%-2*@n
zjdL`HKOqp8%spRk;0RCTVDf?>;6o*_ix3wMF__>bTm<;qaye|^ejUUwU87i+f#b@C
zP#}5Um;sCT!F-H--*bY=(pTUDzk+-Bs|*xYHJImt@0SspALwlA8@#UBBk2H2U!cVs
zX>b`5e;kv0DLDvKPizQa3g6qk>;c6~xGYKlD=vO=)jAv`Mb`O>Pw4;h!1m3Ks|Qcv
zrOSoIEHHMV+(La<;}&G*dXd-~MEpf`xS^zaIOpr<XT8spUTGv)e0o3q<6FWA#2Ynz
zhUO(qRo7y~m~p{g^V%R|#xY<PP@Cr`fMxbLrQCS&i>#$8tV;{GzNIV=&nNVE&e$_3
zS^LtdZ}WGP6cU77i$ZU3-;EG++yR+0B;G!a>bhZ>Yxt|mncbCDgi<;Zf>-x|N{j_7
z{-x;Y&fCHd!0(w*d!p9U55MGKT*i|mDr*0!Jj5(R=dGztyLLEadOjd}h$~SGL0;`@
zjCBmxf+dTPWpM?Ckg2>mRUC=D6L4`a0WSD$5YiasI^Z7FR}A$(tDJ-D7y^4)q><ql
z^x6bzF@}!CcNEa|FTiFG0jQl>88D^B?=^P$dM|ck#4pM2-1}^srZI<l-&jr9eUC_Q
zvqN}vyq6(M7Z4Jwz=$4>I-Nd(d{}wxrknc0(7%Fz09KJy!tNZH9=}n(?pby{28%b&
z%1k_g<TPRjNbw`{rwuee=FI@Io3`cPb6cqzwY35a(Mcpwz)Cjc+uYI4qS_!+)ZRn$
zS<%x)Wf^Z@PL_w!h9LT3yIgpa?ZJX*3*Wf1?#96ZpqkNcU7)+54D?n6RH{q>Ba1H&
z3A|2spHhi>Ro4LW?8Aw^*P-dgM_HQIoC&C`A4NmipF#or9M`=^L9xjMO`a*5@dLQa
zBV8WcX6jZA|2Sy{-h+l$eWd4mk{T0n@LO_aKF;2hIV%x9ncjhb{W3WU84?uVSKz&U
zU|&#2uJ2K3`$*TInSZH1vCX9%nm5ApZ4R1<(Qs<LsQ~KKHZ-O$eYCi~1s}cuUAyp1
z@VYZ@REi>a1{oE<rWJ5a-ZK_~u;GxiO|XId_ZX!Fti=}=Wis8gX0_Ost`q}*t9F22
z1=mM$8f9n1TyC&$iFD~P@bYK0#3%7g5UF>OY^VkFX6Hsb_niHpwTD^q1dwGLZc@Kg
zI*lzAUNU{Z?-Ki2qd2g=;&y(iibjmqT)6%9m-)E$ZmV~P^IUV8HA$N-XFylMFqy7n
zKga$qNaI7S0pDLNl(Y_q(meQ=EP;Pd?f&~GC6>4hvSCE1R)lPUo&VD!An4VYBwYoT
zqRRkwY`09jg*OY03FL`K;!O=hy&lB#!wXUirQqEZ5NaVs5h{<ls3frXk@lKgo!^xo
zU}uh>9Szl8%C}lvzB#WF4s`ZmP`1)<L*=2>1;HfH%kv$cbS5l%GiS#W=Od8vz2yp~
z{#ijI;^@gVi!3U@0X9l-J2AqN&lpVw8P6cJ5lobks1Fv%YqVpMZN*^G41;ElecD8E
zgAIW5j)E=n68Ir#h)*Jiu1s}Z;T<zT>&v8yC1dfZ(%~71pS1Q6nCn1(zs1yiwHM60
zy^B7U);x}_J7K(r{~Vr%Es@EJBubL6qW{!*DJq+_PzuB4aj-!vA~`1M{mBC>2h4?_
zW*#C5)dmy;?jxo({%H)uYlAglqzj+Ho2Jn;AjOz~AY6GAm{>=J`>(@`5LI+|5(lrd
z7NEI&duX2sJ0cv8gp~{USBWHInBLF2yx!6W90-KI2ROIV3pNQ11V&6?1?GYG;|P<@
zCmX1w_<_BDZU84g0`9>l1A9cn&X@;$aeP2ugL{jMm^<wy3G*Z+A$M^9Vy^GldJnqM
z@<|FcVeJ6)@6V+|zVdqS0II&U)B@l3+q%_Vt?-^O;XQ19DCgTs`Z5e@1K5j)y&tIW
zFw6+1_4745(5^O(i5*hrBX0J;EVTZ?hFt{ZZFA71`4l12>RM9To4XpB5e`+KA)5<y
zDV{N|v4}D5UhhpAiN2njT^?OJEs6Cv8^;3qM3r(GD0q>^-@mp8qPu_SmE3ob-Fkm8
zTrE;u(3V4lWpP*okpzuY9p<y)CH)rtK{70lF;{hklBigpR2nfoOKU(yxHtGIIuy?3
z>kzQA&-t<+A7oPGgO5YR+X0YJRM6#6tuUB9vhJE6T2%CL4B8^QFT-)`L6SZn4-Fjs
z125iCZJ!W@Cu=d>ciMEmPVZSQW&<kapAhxGSsu@i-iNXWhvn>*zm8L=U&E->;j~ye
znFF8siVu>3F}YNFT~Tz8-tsp0rU0>5Jgyk{GhE`lOplNfb~y%9N)lef46p@;;@Bt;
zk>2%zsi51*a4>AzWB88L6K7CS!H-lqa?{(b57t|04tJ(Kkq`rYI%c2rhi56)>EO3E
z5P=20-AB+c4VOqsn6C?wZwCnUZs3u^)3)SiL)%~~q-l+iYu`pgFabTcl?gDXBczW9
zEe{V$Zc`;=ej>y({3%>&B<s<wiQqd0Fws~FX%ipvEU>L;tMQSopnskQkJt5Q#+O;Y
zEZnJ1;DiF^%*(l`n~;_}uaAGQp$!XZz5&00qy(=RX=!!ydn}af5c+5k2nRBriYOjG
zToIx43mjAd8<8R@WLFrz3xLrDerAuC-n5DcY0<|?Uh@^E+xW(Vz%Y@z+M{eTTuvm%
z0@|*AyNud5qT2@LWooN1FNw2IIgA@pkgh>2E0B+DQ%Y!Vs*p$=mI5b1YG#sp9zb(Y
z(qx4!roRxof4fINUX7TC2wgb-_g<2%*uv|Dw6BfvxH5!b^w&nfZy{On-PJZxRv3{u
zB~CPmDMB1=CT&J<m*886>dvkfUdd0}x!n{2jVx&Jka2G!Aoj%)=}?X$GU>A3kUBMC
z{rleA5p&*O$yQrYjBtw}wr=Lb1U^7p6w<Qx07uT}*5NFk*QNx^!VE&o%(@!$v`Cv!
zmPj?uR=TDWU?+lXs&r|SBC5f-r!ss?;P)+SG-gN^r({5yaw$&{grE7{7MoKayd5g1
zr8%pT?4sBZr!ZlJGR>8YyD_>45*I3~$ii6u8!EKb?=2sOkb+vBGeW0`N8*8E%9I{o
z)NKc(6^+{hojWMo`h8(w3Q+?Ry1%L3jIK_Kw&=DF)$eanTW0S?JAT#)UxpQ9)%7ed
zA6>x(=djm8Q~Y~i_2j_uweOrFHXjVIDPW3%6(n9<@~C2(B#rssv9WRt``@2FweV2m
z-wHiw<4t1NuP3iCOG)$aAZ4ipyMUg&LTat_o;|Ui^w6`VddUTu+6s~iSi$T)Ds^IV
zHmO>E&8?&irAh;?=pPqdB7=b+TGSKA(Q>;PdU7Etkd)Z>A68#=0v0Z9&8(5Z_vy)L
z1FUotuUF90+N)^HCcO85TH!+*{gao6aQ<N0P?3S$Oqwk%Naq1r#r}u1Wh^F`C>`>?
z-%^}!lYT2bB>3GousIZvFjW%oAYZAri>mz?Bm=G>`QrdfwzOc63;gzxR#SSlCD&uv
z$}X?A^G#8WN4kgIyL%g-?Rm(Uq;^&;AuK^BeBJJw{##2d+P#Uex6&oQefI$Ej!(3|
z0-jkU;mY)nDN^@Wb{getM)_%$uBh^QX6zSts!$z|!CRKtF`z_62BwFO<ruzQPBhSQ
zL^?@TWs>F!H(4x5E?Tx5yWIa&A;#odxwP&&(Ab<C85~GY&JcKXOMlI%r%n3zl)#Kw
zdK6(VZlL_St*^l>>b*P}&NE+zA?AGdfEV^Or+qa~0}3}mK5AoM^Cc0;g9SM(p3xN-
zV%UyYf-ky%@a9O#_|bkl<tC?1rP(m1;&zgb)b%+#^4LKBTf%i8*L&>~-$}<fhg05X
zvX6XK(oFVDiY+cgXI3XVDVU|vW%+fzK^>bTYh|)Xs&0V;d3%6ekTLm7@|VU!Y2DMV
zqO}oE-Azp2zp4&iR$DhiS*BNO&T)Ew<R7TiP!yfeZUqv3XZ-8UPT>>jvkP%;{-ugX
zSB-yy%oAA=IL<*_DLXRQ@Nxf*m?XhQ4)Om2c$J_HhV|7bOAEX0z+dOt|1_K#5oVEI
zOZt>{Cv|MwHPJ>1hRt#tsLZ}W2>W&Xbawo+yvDCa0$CJK;cRK{?crHH07WFgiIJ{u
z@X0hTx@%&OZGr6ySxt=bzR&mEg5DZyX%PPtPxq%03H0Hl4S~#=djK9uW%cXcGqiAJ
z?572YLLWjH)qk()xc;;DOMwX~)rVO3h`V?7k6`_80zE?!fjqHeR3@`}EA;zP<aBy=
z;X=MR4bk>N-EKTT6bJ%c0<JiBEGEx)m_{xNnaT_VVgh(rVkx%As!U5jDi~drVT-B~
zvu&=^?3&f%@oOuvRSRO*!sB%PdEp!7>Zv$E|4r#RA*Ww*{W^NMV*9m}|A(`;4y$Ty
z*S@ESG7*(IB}FABQWBz4A}x~?5D-BD2|>D~J0zt<5kX23q`Nz%L>eRnM3hcRDZgvD
zzP<OmpS_>8-s4?|e<-d)7;}tq-}iN1zw^8nQwis-+l7xGSAOrR$>AuUnb=h+JU{mi
ztuaP&_dCOQ@N_Ws>GNYRv@y9Uc#9!X30uj&WoJ&0<TjRVhAinU#l2F=i~xMpV`q);
z^i`hld9*)zf~I+}k#Bm;Yx&dl$E1(_^52_Jd%RU%QYQV#F{Pa%c-NJ4VI)nNV<BF&
zcQeO6`?W_6=lRbM)$a|G@9Q=;9!M>xl$&qfV?n^$x8ZUq4ayf>&|o-mWU(;spVpfq
z3Ie2qh(6$RSR8u&RMHqv<>`xg40(SFAB>3KNwLaE7+{+VQ^+Q-5=(1dBD7{)P&U)x
zZSX&1`dK*8?^$Gu0CxFt>oW%YB!~Q%de`-ak~991)=h1aG3Z<vs&H6nJ+hrCF)K!u
z9z!yGVun~;hSi`Uxq>CF;g^aI+G@5T71J5H>UV(?oA9z^AA;pAe9g4La!SyMq{R51
z{lG>1N>=7pLs>)ovsX5z>>R{+g;~Xh?7JsC=k<Ltl&^0?!`D{HlpzuPoW84(Tu-?r
z20FpE$56%eb_3lqkNZq5-jf>49bz=8#ubt1djEu;<i|1973X#(9r_~9$q;s?hD5Sn
zfeW;;Mqdhzj2C*|NfYhOL`%+!HMpIzq^x{n6)Ep)!*cQx!`1v3_f;Nj(ibu4*|S-t
zOCAu1YNRX33#)&;P9-z^+AL@IBBAwFQi=YDWNnzbHT&He2+cR~ePrh?<yP^11}03M
zllXk*Crzw{8PZ4@k=Bzy>qsz>30}XRVDm|MFUGa&<m9Wqx0%{v=3KY0ZJ$2Z@YdF|
z{5ww9_K0HZ3i!V-IcmONcg{;IMft`uhp^fbFt~hVYJgd<arU=T_RyRw2~;;(&3`UK
z`d&E)$;`TWcovve9sN!i?mV}*h-CRFG#$?K^#13Jxf#s`flO_~d#ad<xZ*J=d1<DX
zAD=AiENk9Rf|DEToSQ2m+qzl`ztsL)Rpm`1HccHB=+su-H3Xy2hO?-T8!r=o0txK&
z`t5t1oq4eRMQo<otMXE!T#PjqAVN=8@nsVS6J6S*!JvZWgJO{Q&c$w4yKY-<eL|aj
zF$>NQ3cj88E~WNxGk01K|K!=xwK=4c@%5rTGiogD8@Eh+1#PV-3hw#!)vjYFE&P})
zf^-e~)0F)R26jK(%pL(*$^;tK4w$K4e&GISHuB1+C#|12<vrI3`Q);F$!J+GZ*k!q
z8_Hnu@ntjduN6=fz2dx?LPC3dOcBHf`bFq?JjwV3rMoyfNVik?_%ZQ%#A4)6j9O14
zvNu<>_OG*jAM?nRil9#Jq2-qZTE6Rns$(DHomp3Q>|%d)jL?EdT9>}_VaFRG5Q(y#
zqetpyW36h4XU2k!Fj9T$FA#9_+IFiUt?3tPIv~}p0=<b&pC7{HTRq^oH1M{Ucbn1-
zLMqX2>|+5t<JIr)a-s~h*qA!GHp8k$7)hg+v{BJiA0g_}&h_)39e}3Ac}Oja_ZWd&
z^Ci-F?fl)2g^o3d6$~fzqL7`_cSYn<SBvW}HUIv0-aLccPDh%m`FM~q5QK{7qkrht
zDwIoE)}xhAd;W5O-a<fw!ax{3N((5|P?!F1b3Hdj?n`f=0%u9BJe@k>2moP5%_GPF
zPa)oFNEsC;4>n>J+WksFqPg+x@v(I)b`Igl{K*f6&sTG9&z{l2IJu;)eZ_X&oH`yc
z`zOnU743{^n3OdFDlC_k@?*h0og@w3L>EvOeWVLboAiHNZ6Ukx;2KtNQoOY><#oWu
z>8<b`(sOTCkeNTH$WYVDr1P)TtnY>?bIz>)L<N?SlA63cLwgBZ#301ZqS2i2BArIb
ztPP-`u3lQ_P387*M~x;fv>oOY<0D2uJrtT<j$Vljh$NkP{#zJ?bcDC@b(kG3wfd<t
zyyC>LJdmkAys)X&`T{S!GIN&ntvM~Na0gYsSHYbJ*1opIv?wzyY8o8Y`Q|GGbwV1w
zht8kqOf)m>vuQ+?bhdAgujy%w-YNLp5PG&2XgpZVowf}xtb^vNo^{IBrPOlF9JITa
zs5~5-ws|Sxh%@>jh0fpJIx#{)@*`B!l3bmc<%%_PR7eXT!s?DRmsa`?MEFiIZVQr?
zm)1Sq=ABO>$;C}*1Zh1R_)Za;)dH6F3+2YiZ8tu~ujajd8^^_c$x;40gBK%dw&s>g
z6iLll8t2w#=6hvcjE|M%@iUXhRjIB~#BW=NAaT8nRMiymT`pf6{ANK83Ujh@m8Cs#
zo%?aU{P%MoA_AH@$j7S3@yiFCBQRZwar+y2xq4kg@s9jEjpkqKd%qBw;4vv*MDRw2
zd-7}zI8KRTtHC~w+ul&O;}r2C+WJn5no?B|(dME0_1^WxdK($n#ZaLlyPtO~C$weG
zNBjok^?ZAuUsy!IPN)!}(RW-yfip56u?J|&ds~83)T)y|7`)iW)#vpGU7DOL_?ues
zA!~Xsl(yaa&13nT)0)Y^a^U6^Kw{o5GM6CmF)W5F;Y3&ESB}?0&X0$Mw7kg6rAv%~
zLt=F$U0RsTWTZOd?^*PYq_BNFaodk$DMFntbm4nLM{Bhx<+SJ^8nOT5pwF^Q%w3Z`
z+gBZnr#*fn&+L<ph!rLNW!BF+zXv^_iPJ!(r)rqEnpY6vnsjn&G%mjURh>26x^XeG
zJp>y@t>ujD-mSR2Fv3b21yRCt6@wLbLlccG$if}nDy)TT`n$WT7U9&_fbY~3nOO;z
zYisNXsF{Il9eZzv>$W6*v#b7}DN6sYm2AvN_=O$+o#EH~nY-oNwe(vRpdPjl`YU`y
z;x)ioBh2oq!+8158Pv~lse%9h#kX707q4oXPIKkX7|POd-&Dg~RjX>djPnumDg&8O
z8qL^FEc5zfYD#V`v<CF19ASv^i9}FUZz*Fj6BS0$t<?{X+1%>Eiy9Mt+_;f(-2>o7
z9O>t?5V2C0^oAHqhpPp<B86cvcU|?DmlP)$$#2Sx3?B0eG#`X6YpZaK6xb9vGal~N
z-Mm1=@D1i_YL)q}c2+YU!&*%m%#-}efsVze6DU!bL|L~0Ie+p$y7%?q_FJ{$1cKLF
zcYh%+>eFatxfDc24t)JF=>0>u*zuG4#o+JFljRHf7-AH7m7SPtF1P5iLq{zL1u6QT
zCO?xxSPF`yAGq`65J*fwXN%|Ov*{j0ux++(ayO}WauZDqJfxnw7cR8XeX~top~m)4
zc(HTz9P{Ubgw|NR@Q=0b)gC1J*Ewws9bqP@hlE?Z*AGHuM+cZ`^<xaH<O?%r1Tul)
zNAA1H@!}k{shP93-`2ulv{b;<j>=8Z^@2<ef;u?7-9Sg(dl9wZS#;JNbXXK$|HVjW
z_A}5jrI=h8_<PX>dNkG`I!Xu9VQwe;sX6WIzxW6V&uF$@;#>|uN==JkZ<Yo@*T(nN
z3xWK}HfFvD&sIJBozUL5&620V9MXT(!ReXY;Z5#__c}k%PslrquXF2R39j&+fQE>&
zicpKNQ^b1Sb8p^|bvf1dw}OP?bc;qzGx4;`#jKj3COncovFpTd_@VM{f%e+QcZRo5
zwv!mcg|5;FyV#N0ur1k4h4M>5V@9V4c2SDz%>E;>MfrKhb7h7MTxDoS_=_mECg9`{
zIA!qDI?0>C!H1eVo~HilL6~V+{tZG5?av#5d%ZX|LbpKIHHY$s6YM-|&mADT+3(76
z8+zvt#8!7`?+Q5eYr4#(93^Rb62d4u`o7ls^qv;L#uA=#R%N!N?j&!i9-eBLl1fH=
z&#xbbT~FdH{c^QI)u?xm&9J#-t~Qc_p%1XHd#uL|C02uo#G2kho6xLGN*g1~ki5m+
z>P@loM&!qQz4PLGomvlZ$=L5lNPGim9|)o!ksOn8Yl9fPTf;(<_87@1n!Y@HA&09{
zaY&T!lzf2#8NO}frJL}1t30Ke7fxLeZA|0z?d`L*GnF!<6x+=NVuVaiMN6P2`<C)F
zM?(_XwT3=&i_oQO>dY--2};bq^!=|qH=NBYnN&c;C=QRxU^#dQ8Q(#jaJZpgY@A|S
zQem4M2+P-n1ama4D>jIQ%-)avfyi3(5IX!}sBT2oP2P$8NqMgFi}wv8@IGFFR~baA
zL7?_g=v@H_YqJjJD!=txz_Tf0L*X#qrul3=7YoEV3sC8>7t;o<x57~<ze8~R!)E%-
zwLeNb?1T3{XI!~i#;x?>A?-prbR?K+zI1nGIlOFnje9^fIq(VsI)oHAbeh>b%6w)t
z2D7u)W^LE1=q56iu*NHdg33qe(2@TZ82AsqVFYz#+tPq*`O5aHArfPYjBlOxAzuF^
zync$)R_X5axx3R6Xj>&i1-Zopm&N_{u2P$+<`~zOtE7ge%fwAuLevQ`Geg}1S~FLu
z{9*z!&3?Mc$5~%KKTU9Y9aC2V{m3Xc!zh<B)_O?RK(C(~Nvg7{m}5qgP(Ja`2Y1@K
zL2?FpXZhDoEmo77>&9F{`H*Ej>ekn-<z!M*PQ7fKA6fpU(@nL<jd2)^gruOLEOp>9
zes1#z-w0&ZyBnqxwj?gJu6N%gErv?|ioskGyX_Tk3z$^rcQt=(>RXuZwaD#1Fza*w
zJ?-yU$a~Z!<-(k@Siy<ddf(Xz(AHJ|6DCa5{1iV^jc|RFv{bW0FoI4)z%o-u*LSQZ
zXZcvS)`rHNJ5(|;!C@W8?52~7yN~1fcf`jz`hKaqhI}d)==$<8W^;LL^vjKx<<XzC
zZed3H@ZQto3muJauU+?fs2)%hIONBcd+`*N2CNnXb@oMz71K;bl}s3GI~nV2tZjS5
zH18nUqLRB)D6YUcTK!$F)6jQr%5tZo)i6z~>H)3(M_d^v9X@J&$1$vjZZ`-!yV*>7
zIY;YP!s+o-HI0AG(l^ER?mE4WhgomKD@qHwbC9XP)B1T0nE?>V1vwlKSVd2sB@=}s
z?<z-}(0SLdS08_G)x8#y@?(BdU-#W*C+e+k+Q=tc)^B5-(UyB4E=c^@B^vgoF5jD<
zd_eT+*yLNKithR^BzKHU%aruDX=_Gw#YeNNezs3H+_t>}kT*M*1Xp4eiL33A%5ta6
z8!Wr$+(}&RtU{Z2_Tk?ff&S~bLC3~w$TWed?c2x3{8WU;Z`do_<|!}sS3it#ExfiJ
zWk!3V(!6}k_MP+WsCE3>Ivl=E*_sR!J+62r?P>sS`izFS4+5Cci~F50eN3g4GQJWR
zrx3v_e7cvH@!eXBG`k`eAKgAw7^b|udt&``$x9<s!eE#78GY48kr@8jz75_%6E3Gy
zQ?Ff9iV%ZI)&(sT_4L$tS9s`_nXbeknI#?7g?_&VKjgKIHc&@sg_CP<Jb_~j3nDmF
zC=SrhRP(hrJilrbFIteJ_Tr1iwQF3@HBv=Dde%FX8}Cb70kJ8U_t*lC>htU0PApmO
z(S0$DJ9`1Y0+a?FIs&qYp=n9h%=26C_PcIY9jvhplCB}}xcc@#9AMkY#KgR3y*&dD
zFeZ?~cuTMF#2PU{30SIXLT(*Fp~O(KlHzxcMG&{lHNEHYp=#ncZ)fZlAMz78Y4cH#
zi{^{L3xy8h;bc0C{D+c8ktxqn-15ShpfZ*!`tRxQRRUi;myq0!G6dE{3zN~*RoK(@
z$clqj;^K5dkq?HxhA84r)sb_FnKNW>S-x>edR#RB3g>Zt@HU4z=LUQw`S!#h2D3pg
zBu0PpfXy}zV+u(&jMKE;1XP5)Lxiz!fL2h#*z2kX5={v<cfA+II%>nlBHsT*H>{x9
zP<~=`FbrE$KwJe9?rT>M_Di&Uj)-AF<O!PU?Zb2^&sE@c>lAi*9dMLyo6Bf9CX@b|
zewCQ9Uw0;;EB?AX@feb_?F@RE`bU`H0_D!KJfP0}KiNtag-9nSTl5_Uk~|Nx`5-(`
zuE>Cqv_6-Lj6+6WV$&Qa)(dl`y*AGSHFIOh*>!3n3_HU({$gA0MsyIT7SMx}&Lb43
zH;9{D(U#Z=+=!kKlUfE5j=#x0@Yq?$wt(HAq?2q*USGW&*j~<O6XE2WqT8BWw;*KB
z`YE3zCIodQ=JfUN&wSDZoJAHKn-oGwm;J5|krFsGyEdbpaCvqsCX>9fmER;ePzK94
zvqIiH+508^yGaUbPtUi|n<77)gzGmnmmpkDSoRTkJ{2VcDO6D+_j3~XC){kst3tiv
zA!l9RO*cd@OZ=<97pDHwI*HGNCaI!qS2{kz4;|{xsZx1qsUbu34xx24;+!SbNZZ-Q
ze~lo5PL!6F2_<=cf5a@8ckg~PTa)h(l|kfuJ-C*mb#N)Z%he+CrQzT&#)bR5ZBKb|
zq9<_-<1^A1LuZ5A9GPp4@-u}#lkbl;j`-L3-=;d=ac+O9scLrI%K~ICy$dqGVi&C1
z1m`Q!MMTI`9&lIX?BA1a?*$2*1nZ|xCt>(Tgp1UNn)D0fS3WTjvI=!E^Coy=YUm}_
ztUaQ(1p2fdIO|ADdQtl;2=Yx`N^x1JZ>E#*Gd<%|Uc`-d2n_t@s}Xjp$-swGzg~dh
zar#BMpq0#NNXy2NoH-Ybj#)L8(kDA_qg!zc?S^5c7m|p4rl>VV+kU|+<Xc<N?v!wt
zLa*tWq}PzLDD2d`o|jR-pv&7;?B|N%{AwRi<JD4T_jDNdRB};`>&J%zt;2zYR4b)u
zZib@NfI1TQ*fnnl@f}~W6fKD+AK`8#OiexEIjMaqC@fLYMHbB$n;y4CF5g6-!%Te$
zx^nWIr9qMi`f8rV?&*Yb>dV;YY6TWeRe@}41#-1d_m1vkR<0ap$h*j^PC9b5o__lU
z)zpb4anW}Q8E0M1vZWV$INSKfXxjvvwi&jv4CvO}H{E(q=WB#`_#Sj0x+z8Ro2XZ@
zfS{A}RMk9S;KY4Q#pI6{zAaK!FN^QU4PN#zsz$Jb(23Z?CK2q{7tvkOH!~DME^nCZ
zir6=mV1w7gue?((A%At*iqfWL(<7=tR<aKQdb7%VOWLEwM9sPC3wxa?L=#hx^LYzI
zqo<B^No9pXk5tgpKs&P`bQ7JxkFA~LjiYl-{+v1R^V4IUYKNEGuYLA4M!5%jL_(B0
zM6+zqu79(qw>PxRnYPN{(AvmC7hQz@sOIyMai)KQ8!=UpUzq~(^s@71)0zJ&=8Rnl
z+lf(ToNVKyYuv%z=ezC%UIzX*f)Inrg(IkVo=#|LtyC1=SxK7^eWex-^F+C&zJj74
z&%EbCWn@E`t~W3|$H@OH-}e7N$NpRG&@DsYRObxN<i+F{iYLZDCjcGe?|RDAz&*0`
zk2z4xgAfK^oxxEng&+jTDQa%|T7pX~SQ%{HyG7_cEoNlq6aYPw=m7Si;X3c5s*ZEF
z32CHJRwD=yh^}`>P$jFHP}kVUz#tlb1jNGpWA?iOB=h%geu>>+^6FlJo;jN@CWXfL
z-nna!ZlU>IU@rV9ASWY=W~usrC8K@Eqps{ANq0^K5_JV|cJt2*@l_s!6IYf}q6lyG
zJS3E%8REmiD~I`(|31LJ5U@mU{ieXYupldOl#Ip*66Xi)EnV;)Mt%BVy)ld<#N5zv
zn)DmY)e#*9o<Q?^yh{*#sG}DVAkCcWrQAR4cO$&s;Hzr}I^#SPITI%~XM0nEH)neq
z5oZ(}i#E;SsLrO<NA-vN@=kwIH{0{h0-D|ozD$JQ%8wlR$`Ei~9|HNr6?9km_Mi%G
zCA+K8Z)eL(xVqs8`#h!P67Y+_XpXsTEt<ffqL-h55ArjE8_p^2Wxtgy`@-D&7-b#+
z05wbx(jDQGE>cF8u^xZr{C_xMzY!+efe!Z)U`HH3Q39~}X(Fl>P*x^gYO2G3ErO3P
zdQtIKX&fBxttdR!2GiUR<!oY#cw<fiC<pUFvnPKb`rq@55en_E`wdCYP{yVNEx7-;
zNAcf1qW|;j&yK>&W$Ts3@efz>9~HyD{%>zm5W914)27JZl6U`}m~BN~uj^uWDE}WX
zpt~1&3rxPbB>wF?@b6!b?_!XAai0DX$^YX87$fX-36D*+(my@PY~Bbf1+j)k+LaE)
zfK8;u4@!LaPny{H6u}D?r4rEDz5qo@c~~rUcB8XI`YSv<MXXQFAsETZ)lW4DxdU(F
z5#T$XC-N!VKdgz_3SR-KDu&v7_;4dtbQmaxv5^lwd$A1MdJTk?ko4O4Ua%DL77c<(
z@F-zE`V~^2?SZdU;s++z5`ua!BqhPmRD~qI`c-7*H`72`SSCf-CSdgRbOiigte+qm
zpne@L8?$=wLfTK3z4Q15-2D<n(iKd2{;p1Y3NxX+KnJqJ&tzOLgIQ)v&<V;1<nz<+
z_KMCY#`3ZCx}&5+AixIy%q*ofz{R$<FzdF(g@lelM#T9NjrLLIdA1$IvA9uaaNqEy
z8nPa>sqS(iSN{o7a_&i1i2uXqFNq>B@4z_?0V>kiT$)y7K>+$#F#<Nj3F;x8+p6~%
za2qAN^O>bf68x$ah?xSzx3x4{8gnD~a^oblO!ASq;HT_4#>km#?bOnSG32jpL0RM9
z%h-%?W7)v0EEVxxDnS=%69n1-io4*mJ!<$$dk}&St4cJ##>x4Ao_H4~*bV`{sn4cB
z9r%-BiU3u70GJN!M}em2=WGIr??_{zf8Q;N2pk<+h$Y-MohYW2`DZ?{MkjY8h@n=s
z!?1_7>jkH{SbT%l8T)=3z4j2qJr9GbYXZ<)EvPV$2f^yS!HI84O>oHo#5BWt`Cl07
zKPl~|;Hc3!#WcqF56=@$*GW!%zqIl6jUh-0;KEtNp&DvI9vnnxJ3<2QP-_IEMRxI^
z=~PONRuv*thdZLjOl5&<CI#DL(<2C`Mh426Iaeu+?|~NnCpH!d5n!EwWC3%h&#!?L
z8#QoNW5pVG5w4ye0ieC{;rVpP*sMQi>6)X;Yxo;V$VKR*v7rlBFS!D`It1ji)wM1m
zTi$!*W417`YZ%EH-N{W6Cawi$R{W%rcfsHjAp^4CI$5u%K0=->k|IwqLMgvFLRj*#
zf<So{9&mFcb?|LR{xcgS#E;h5aoCVbQ$L95#OFSp2c8G^u{<U!lRV>Bpao?NJJO@c
zxIc@q7%yU=@qfPjmlbKpb1|N%>>_@^pX6EG6B3U0GbihC%HMD%dTl1+;*X`0p4&c;
zt4FcR92dX(oG|G56qoGJGkA<-FR4t%?<#4t1o*F<Y<D@ndOLIGUPWp}WyO59YHDn1
zt!vodEO63FjPq?sk0MIP!x?*42`vN@ar6NDoQL)sEFeN5fK*_{^5P5TGcx@z*X$(e
zpiuzuycKJopngYU6WCqeE5}~^1pgT}mJKmf{%5$6jgXo**cfxLw`v8PQXW{li#Snk
z0EjifTS+BjQ_(F!vUxU3u|>SaknDGS%~g%(iZQG&U<(~^(jx<(P~~ssBRJt*cX08k
z??UnqlDqLvI{Y*eODX;I;{92;PnTwRMt)86*Up`XWn5CW1(vs`$g?PkxBurIp<4;P
z!E)p)gjiDsMll}IMJJ!Sm2?zmi|~O7Qu&F|+5QX!+ITVj+~11QXZ3R}qld0GGsHeq
z=-+!sAnA3vdPAiQfP4azTqLj(LEm`{cR?WblVyQ36*;oVqMl~_>z?;7d*PeQ#32`M
z@O^;~WfFYv0n*qQ(~NYX$LO9K8F7JnLb86la2|Kb8Ah5RaL0a6C#G}$j-`^ThK0uC
zkfX5U479+@xxtY#pp}?QUqOt$HBX`Pdu_5%`4^Pdjj-xrsk?yOnfaD%<-VXf={m8O
zAf3!kLo{ca2-hiB7~=kcLjB~#PBZ@{dnA-vm`N`q+Tgit1K1iGf$mJmkQ*zxMwfmR
z0h^rwxo-r6!2*^XPLtZ<V*CK_yIjQMN)R8{+(v~au7lp?-DSi9OJ9ANZ3amAL723_
zPB@*|?+ooqCxvtOj(i0#i0~PDy7v=cbT54g)L;u^3nY*j@P3Zc?g>-qRvJ9&H>;Uj
z^;3@iwW6|KD4=p>v(G*oiVl*AUrzZQYn&9cFR1bSZfR<7jskZ$vYs-<PlLOJdJr7@
zbx{wKN0OBmyRNjH*8_c#G|M|lLeQwQCF&5aFM#fFH2|N=o8=19w~z#OF5=!hRmJNm
zdo+uIyo#^8V60lQ-Z{hbZUCnpji~?HE@F1plNZ9OT_SHrbAm(7{<B=BgP+wcHY$@>
zfd<y891!)>8E+T)S=%vw+YGW@9TNR6%XtL%e}v>=UBjIr%5|9wX5=?M&~*VeQ8V}o
z8XaOcHcu>2{U$m(@<^eO`Q8keWERvWCPF<*X;W)5tnMzVZ55VX&{>D<v+@I`yICpV
zFA+S*GMCk9za~hrPtP~b8tQbT;krr>?7gFNd^JCWeT|fqm@4nCIaiTj)IvA$u{)w}
zTL6jOmy`X{De%*Mi9MG7;Kr@rTcx$f#@R5j_1-*%J7;^p%*dAx6CN(ZcAGKJyj2R_
zpK`iv!Do9gMKHoWwbj1Q^QRUvlAsaRY0Bs)kK}F;)H#^rMqfE3iVpJ)t!%_6_1S#i
zdpCjO?nVOMx*usx{ry;y+4}^6c$p^-KqQwLn?v&HslZXlL${@JW4I{JjTn?ayUj{j
z7r>Ib*P=`4;C8YWNc4mIk{5M64~kRSW08~!s1v>tymFzRC9+tvDNm!f2KuS-6hHZe
zEx{%pH{F?=oJrJ8lH?-G3|H?z4CQTrTIK$WGfPN546MKJN#5*qhu1A3;U8`og&zo>
z2?avpHHP>Fh*hTR1ZTi1Ap4^6D(xjNe#u<W6}eowc$I2{-c+A!2NMz)F@dHzEjasz
z=JcFP2|B`KGgGJh_7i7c7QgUAw<4xc9G}Di#1_rA6#SMY5Zm<;m;+;mL#0HBZ(Xoi
z{wqj~&%3HLyzp_v>N7J!ocqI`*kM5(L9L(56s3`rr->wr48awVge$kTxHQ?{SSnH#
zfFPSMv@v9o+JRs5-a>2yFNY4x2m4gMdSQt3HY+`{TQAyVgs#auxEQEw8FOy14XPIv
zMoawp4#D4WrmC(*<FHCaDwxAMugLbH&Xd<VzP@7I$EP8={=0%k7MqX61CbikA7sss
z#~ybai<={=@!1#%j8zhxyGV6pdapN38<rHCNoejldJLU$f5T1p#TN3$iIqftCGY7k
z9*sCvGaJYlW?j!Y!(3@_z^$Iw<j8o}&xnR-A-?<`u=0(tV4a2ZyoMiYe)(rEoB+Sv
zg(H-o^J=hrxX2Oc+BJlO2PI=+N7;X6g}GMqS{+xqF}x7XZrtA$-97>Tx7c-IF=jgu
zCWc<yKisa(UBe1ZA&BXA`ViYp0zeg~J~S<%XSU$<X#M+OTm8`)I>iiiHw)zTKp=Ok
z07uk5Mc2)7HYb+JaFkQjReyM@r&>#I<XHH?7>VYIbq=Hb9rhFFSMlg*XXy@-hRRRO
ztGOL4?(cOo7uSWp5BH_ny-?V=h%z`2wRNcIxz(9Zk#~=IEsaRd%d>PP_8%WroM02Y
zAYF2Kgv#4n4r&B%!F8Iy3i*FmOo)BPmcjnIf0PfKIH>KY9ZutiA+#Y@=0$SR>H@4p
zmT`k}65PinBC%m)455LZ;3s>mM@@k9Wp`th68g~MJ|dmw6PTL@`stSdne4PUh#a96
zCV6{Pnc*s_KBe!Ya~4Gro`F)y0ck;j!bP9*;VgamJdC1KqjLqv&r$=vNd)>ha`eS)
zGluWJGbHVP)H(c?R0fVAk<}|s-&-T>TU7g79?c8Jg!do90dCnHU-x8{#YSxP1Cg~a
zKQg;@2)3F&CN)jpGj3(lRRJ2$(DrWUr6PT~(MMX4SeFUhvO(B76EQ02q;Hw=3k5#n
zxZv9adVO#_p0oWKJF(CFOuv+0j)hOWNTTWP5EG6aSIW?I=@JQ7nXU%3eS%LrU!YsS
zhUo2~nJ~NgN~8USB7w~Iu*y?ODbM#aM;DsqRu3|w&KQzKEW-iz+myaS-f~d?hY{us
zy1B6Y&8jrN;?5C7T{+mj`yn?FuPa`pAo;?!OSp9Jm?U+Uvib2*Ny+D-kJYKnV~L{G
zRnC$Hot$Jo5F2`e@wE?&4lA##MMP)bZ=DWHPg{c|v5bCK^<SLg8=?o|K+r>{={-7X
z%`=M#M_8Z;X@-}n+IvY)cu>|K6(RBFQclq5vu5hOgoREAPUC6-csXN9a<MBHdlVb|
zpOG<8nw{HgWS%?fgO$W0=1&s0Xh+|#!zzDXh1d(4{7d}oTbCY+P&=~J9M8XSc)qfp
zFQ469=<=E7=AfK5nyGjc0~b~{NaTKQ5o6rVx>%fL?=nz=w%a$5`JvULcjgPtC_H_`
zN1UcYkJcpE`>p`cW9CftHJs`-+foSQ#Bx~)u^aJ<$Ig-Q;J}k|O-2b+)?d#O5aEtA
zP)eoS#VV}EwMjuf|IKm7KDv<Pnwiou12SP^6gKto>8>y1fvBM%g^M}yLY^w;;%ihR
zuY~neS>ScW$nvO^{&856w+OQ?oBlie<6!3VLCjBsC__^-q@(2cUhiCf{f29#W#5>t
zKQk<qj*{WhNI<wTRyrxDQKi^wsnMBdEr{-SU*h2%`>dCA%+ybrF|1X-9`_9rjp<U}
z1r;aZH#4qxNh;XD-W$X`AI$8PaotkC#h5Pd9r4>F_vozaihe;)l7jq_g8Y;4M>vF1
zU$hrYH{voS$@1`IJOg$uNo9DlAHS+TpWoN<3leav;XI**%s4bvye6(w#FNs~B-JD7
z&)broNBUc2hMPBEr|ERT2Vbk6q&9r;?Wk-ni`tX1GV%Fo5k0-SsfnCc3U_D}IIs#A
zzpX#Fs1wnZCZpW78Pu(bG$c#cCVR6}6`j{N6E-nqe`zR`*w%D9c#4v#&E1dKp?8kc
z!ekWJk0<MuB#Rl;`nU#|Ge<*a>4~k{xL-G>`<lYAnw?bgi5ECc6Kt<eY7a~*E}i%t
zm|K*Vp;ptIq~ctSH_h6)Om<mXUg4}lljZ=C$5(li7F<-tK{R*8^m+3Z$(|lU`5s*U
z!1+|4{MTP|r>UokSd%<R$pi<^D^bMn9$KzensAJj6{khp!1HTXVHV#ZeYU1qQs43#
zvtY)?a8uA)QZUz3=BkYTe8f{Cwb$qsNV=H5daGevA#I>9-9CKXLpN!=qGg?k17AGR
zN9fo?xY8r3rj|NHM<7X7iI-P@pf)ipeRhia6<@u^Y2l6Ao~445zGO+nPt5#{a2ZY@
zasJZ!nG@|9zW6BC*j}XK1GAIV{gK7j<!>4eyBTRAz_91=8OsPA6?uZ{o#FP~*?#8U
zgB$LLcDstog95!yMHH|sd({dl|6>Nk>_Y+slzfM`&tzYeA1g$w+Q7Ky9^80JMh0in
zQ4zUy@v5{{Z|Wm=*y$M8mv{i<!S<+(Rtz!yUb{Ye8h2J?knw84tncv68mkDmq?9!K
zN%>FWnR#e$uJ`3oxCfth)KVXGGq~aNjL4hx_nC8qOlZO269goZ?&v0t`-(km4~RsA
zPn?BA?s|jmb5u8u0Ka?t4o7v`s6tP^;zD}?@vgEzsypciela)gc6eEeOtM;u!o?$R
z@EjqO#p(KF_=EP(3?apK^?pA2`{nprEDeqZ@AcIEJN~WKL$=4--1qI=&)P?gjJhjF
zpan*2Qdq+}V(_KycPn*s$aG#1dQr~hWoXO4TjG0Go0xLN{*fBXX!u2xq*$<LwmOIY
zn_zcXs@hyq@66xZqm2LbAtsO%YpM5Y=vFndZkg3;TA7dCo(LPt95b>xTXTHv;Npv0
zDu&l6)wmH^<M3A7i5<)&AN!VtTACV<V>UXjOeCn9<G!VKeQgR^n5qJ=TF=sng?8$e
zvLfQ~vmCodNI&fzevkb;*~_?qvNQQtYJ@c_hMDijhg;?o51v{evT@ioir_y8x8N8l
z=!(HV=Oj0O1jmty#a-4$=7Md7y(J+}LFd%jY@6gYV|Ie0L0&4T(b0mZe^VnPf3?AT
zL}lnXbGThDrEG?tqfCsfsZ8&7j7-k_m&E<=EP9JW`I7+Bre|~=ax07F8J|m#<I<d1
zJVYiM!xWbHy87v_<z-0CBMOBmvhgo7uP0>Fhq6Cp;H5VSD)b{SPc}+PAWt8Tv<`{2
zD?t>PRDq6bttEZS9{1M8clmn<gn@VxG1`Cstl?FENAfYIO^)@L%aq41orK{FeZ9|V
zi<I-jClmC;$b)`8S3$k7e96;*H-tj%6tfHaqIA!%?EaUAWM5uHW+i`cheGB4oWzeW
z3B8*U&8IaPE#4E$UHY7_8l{jd1>%*RlDY}ORi#t0N?kFX3!$8IWwMU9+E;%~$6n66
zD|=6VSDTF5wPOAlrAD1Q_ArV6go%)Bqge!}naa0ED@I8Jy82DJ1eF=<g2n8;hSwZ=
z*YA_pwR6h84WcVX`>C0HSc_q<Tequw`NE<FXYDf5f2~u1aevWB^doJ?uu*)Ee-Kkf
z?`vD!!UftLD;0b!E3PhKV`k?>uP@cGVTJ73n%dk0OB#88%UFf&*Sq6wZjqB_w^qM)
z$*9C@_KU7W;r-Sl*LG+Y6C!^Kxl*4i2@2i5lcy8>_1vwCcQ&<_=PDL$RL*E7>nf?n
zX6KeFWw<8zW$~(5P`w0o?b5TFQ5*r)g`9Nuhv%nynQx(&=9Xe^y6L^s_e7lmeJ5Hs
zf<V~bS+F$LB5B~)EM@H1_-{!1qR?g4e9`t*yV0uH({v(V`aZ_4@?BWA8U6FvsTc{u
z^7B(sdHO$=pZ~j4p*mKq;}KhtPfGPtzD}U@&Q2#$WfpfB3HAbgN8zlR{pg&+k*~-W
zPj!yIcR{w>4yvG?7PsJU>H?O*8dRsjV!s+7y3e4|AKUK4p%M~I!C_&nEw1_FJD+2F
zbT}NG`H?dq7s{Xp^tCN+Ky%B}h~CEFwMn89u4wR-N=#m8kPhaWYPc<eE7g3BMJ<Ck
zQ?J5Dwe@an@BKo1QP4PKjaL+;J@%8xylRfSq8Cf()lrK#Yp11Q*`rkBq8Ir-Rhi;x
zUmue)X<zB<opRApJT@h4`Q_o)g;gwaa@SmPG*vewL1Ec)Kp{TW;<F;5@tTV2Dk)`9
zw_jgi0@jOu|K@&coW=<oD)Ft0D2~SM?UV@%bZ$6l3FDTjqe%NtUZlw-wsQb7R=Jh5
zWB|Q2xT?8~X?acWlmXu1lfr#UNXdD8<Ia5wQ;OoxS#@mMu}u#`2Zx~aTOw$-q(YUL
zk*RCh;PY4$>yti}d&-zUEeXuWoLPa%m%N^{J=uIc8QS*23iPO=*_?WnD)sk1qTeR9
zt-9ZcAD*|^{51}D`6F|?(;@ec!vtG9jkF;_5hKCTXlPIzb^6I`_qPoK?_~zDi?kxf
zXCg9(eOS5Z4svpjc>!pnBz8xa%}iTnz?cK(`4|rAJ~<_nq{-(ZSHX;bhufw2M%WZ4
z$&Ms<5+o&q#Mbe_Zf^f-sbL3EPbA07cb-Mkx>sPl)^T~|SzG^|6_C)jf7L*NP2%SW
z@yR;WFmsQ@)8z~$FO91X3F>1}Pz^aMN{?*VdEV@Z?mUT)zP$E1pi*xB_UaPBO8#Q4
zMdX1{oD_Ry2=x`5#-)$``ah($7yEPC;SpFK1#$eG{qVb;dr$SKW-hMrNm$BTEMbK>
zUzSQ+Fw!t4VSR-H132wdgHu}kP<fXREN?owyIE_UKz;nlD{0T9e_5Yfo@l|7zRva)
z>nrpMD7$T9-{o*K*dVHU6{Qrlw|A9YKb?r{Egr*d$sNd!+L(|<{TzFTn|c@pBfR?k
z{59p{mwmbxWyeg(vPaZPha{Dh{R$Uj$E+OOBlXA(g*^)%nr+4>h1eLZ<ho<2y9tl0
z)I1DT`v>j6;$<DkN)&+Dju_8z-<VIr2L?qQFVQ>4OcN62QaX7(^z<@3WQg?3D)a-I
zA7NXQO!rMk8>Z5GRg&@Yug)r1V>k)!XXFgiLh;;#qihEFvppP{YuDp@tw++9cP8py
z&zxJdG0SV~D>*PisHcg_zVZ?7y|IZpNw_BYhgL0QaG+mHY&OqiKvxjnC&M);#WQO(
zyn>3?`OU3Z?zeY}&@u|+<y$<#))WdM_guev@>0h1+>7bXn2uAlz)id%rrm!Kx+v<Y
zaIbdhiE9z-Tfa~Fd?4!mFm7PbF{2DBSglfI=k(DjVwLS_vasMCNY{?>YWk&lsXt>c
zHg9NwQ%*<i00f0I3eoCnEu_H--)CZ3SzbF8Ogm-dp>unB#iYp8-d~zXaG6i<!Aaql
z%_0sn!cyock9Dso8@!r}yXx+F&x1<ys<<AHjmiA9Q{};6X$^%Vq<OyIn}8MkhuN*K
zGY+;;%Sn@s?p<rc^)*8=$EV0$PuQa}-wcye`0rm@Y}Mjby)`>;@iWA%v;z>P9+mt2
z?&T?E+phNs>nc8LoY2*gwpVJs!1`zHjxL4yHXN@}d2{9Mr-nTpG8Kvh>BjWQgsWA_
zKZz^Zf658JYMzueqr|d8ZW!{8<-H&m!BIakYSyCz0)J1LNWO_B)+pkq(nZ2Tg`YBU
ze7XzWS1OW#zM$DRdG$W+QUqeYHz_Gt?%#<F^vZsaU-Tp*d<IFFScw0&8TFdyeo<{1
zB)wETGpIjC6@sKF7v0G|mAkN^o4Q^<gw>#zOH)VGjIw?e`2mXYdmFhs@8}L#J+><+
zLgPl$G<dF2+28+O^15&LK1$=gg2t{yY3!p9mz=4E-LtG9KB1rar~8~dj17_tem`_u
z<}tK3%4eq`d-tPbb>sy9xYN5}S5kqmUV+)4*#jpqEf;|3S4*|~vk)Wc`&567C29xX
ziS#MlZs_X-*WI;0FCLCR9!g13lp%6Z$ISUWt?NS>L^e!1VgnC;zN!s@X;?Nzt^RAE
zw#nV3xm-G>vZxl7tGA@VlV67TF(vU?&L>tdLaEUBySbe@^r-O9OI`UdFSXFRlg_sU
zO6+y}x7Huak))AHSC!Hc$xMd<PQISL8~91W&D+C&w$*l*KbvZ^fQwj0BaT}aksx*7
z-#_{f667O-Kv6`$699)m3uK;M6M>@@@3-^vnddH-e}7&EWlJ;22L+}Al3}=~%@6@b
z1{CCcMB2|Jnob4@ik|nFv7&SJsS&7Wt_i>r(_S}+fj@zPT?Gg9BWEKjF(nMsQ%*`u
z_;(Ub+`qoRXmM-GNyxF9aY|i3qM3B=%#l?XP!<etUy9>VO35Y(Ex%{jo^FsHSJqW{
z=#aK3oety{F!qEO;21<+>yv>9Q`2>$!Ry8?jTp`-{ecezl7<&YUZh;;q}?j7w+uA*
zLlf_b(cgZ=*2fGX2^vO^=1iuD=-h4(DB#+#!%t-_qQ(p6slQOirVQGRJ`|<1Sb`*?
zr|3joduR?y%kyg&AHPvdEa+?f(siA*a{2+pOqix{UHG`2ZOYBGuK{TCmz=<~pi_n<
z+mE#ME2I4RV4Ui^;l`Auhrv+=LI_LAwfvD}22*2Wh4Z`jboF(sbNmnmUsz}zoo0JZ
zCi#(PGWYIH_-W@$X<QzDjzi&+cgDACW|A&7#Sh^kJ_qJC?Ha}nS&7h$Lwv?pzEAFj
z3uHR7ky1FXPH4}Kt4lunl31L0VtwFTO|5Eea9}7poI3%A+D!o%qW<C9rl;F56cHef
zw6F38Wu}Zcj6S4wN=Pn%PH{P_brJ{)JKjkRwj`r+9_bMi?IV^YocnfLE?d}!Ge-n;
zbiQvJZ$szD=Ek>suXIGe1lyj5E?BF^&jrTOSBoB*K81_LCZ1X&gxlw@Jqp?LK>}iN
z9a60^oI|I8FYJ<}$);i}&$UQWt(-p|TPOOkwn}w1;QT$!&MtX}u%CPJIpK;HNqsX)
zG)k!}<ndvQn>RJF)U$-lC@<kvwSQ<5i4pEb*MYn3EL0Ljx(yNB2CdJyfA6mcz^=!n
zaj{$%cj5Oo^a_scInh8H<%KoUGjlQbDX@l&m*et{6t^}gII?v#V6lHfe2jfY3w~2i
zh?+!F<C$wnm>Wp8-U>v|<~>%tb|Z?0M}sGI6>Qj436^xCjgb04eYn#i>W}EH8tsGJ
zaL_tNwn0dyacnP|jw9QOJ3=ilykDfI!T(U*#OFF)FtKlg9D6qTtvh5K6c66X%UvIA
zbL;j-VzZ;Ed-p`Z`}#}k{{0?ZiDKO(v6+S;4qmGvn4#d4;*}FaN+wAdyh{PVPMq$y
zcX>@`<b9#XP9%S7aY-Y&UYBk^J|HOF2S0Km>V1Dpm-{X9$(ega#k$I84@N*i=g`<-
z5hZfH4l3^x;%py0_9W$p4|^sOOrj44J)uwz-X)`E*(#h(3O*iu`s}R@?5!*+-8O0s
zI_8SU%jv(75Ekxwb?!A|?zKC4X~KH2o|6t;_3JW>a=2FaqS%10@JUdoPmsowAWoHQ
zZ&hP;wKWWwtyUle;uJDvGS`so$KB=n)9dCEqZt)fJ@dBi)ly#QSn5u?b)3xX7}+yI
z^9=xcNb^csb?bfG>U~QM$7_0{+&4&>O-Yz3NnlWfoo;-zHD5Gq!@&i!@g$w&!6=$r
zdE^}Kwm<GWj1*_SP!wcCT7}DIa2))h`;AA3y(lQLshPvjlsm$bd-BS_B%ti4D0})o
zqZbHuB(LaMaC_@;_mt$~$p|Y>y>8`3_Nc=yoS+}X&~R!H%FirP9j=J(5ClJ;`K+<K
z#SI_b9`8_L(9I{rPpavd3GHHwc^dSf7v?x2OIH13@5W|jM|ui=s~9ip9M$ta=*X|@
z;}mYw(C|GtJC?R?_(?johIun$#{B0mX4B1?lP$M6Xki}C@_Hl9)b;qwRhRQ<3z&X!
znn%stzO<vaY?xGb=f|M|<V9JZ%egbs34N~lmE$|*=52ez<Ax9rm8kpel1^rp>f%IG
znI0HRHsuupDHg+--IgYmklgo-%?Jz8N-NwL(qVR*cpt6*bn)%`?UrJN?tCP)eO0KD
zf7LvRlsRo<TSVZVo$>Y<GIUq~)~<u=QFOO;MhM;KnsN{Usk9++P0>4%zuxWGHIYu=
z1<TBr?LHS0O5?M_`wdw5dgj(yHs6LgK2>av*nOLgK4}N&&S^W6?w!9sC{#Q}(OFxS
z|DgOW5RM_j64Q}ks~83+AFjaLx9Ico;)XuTEx8s7>&)hYDiBgLl6}OUk*=geyO)Q{
z*hKBi6Ye&Z?g0u^uLn(wMd{dd5?hmJU=}sex0}8~lzz=0G}A>xOVE!oQn(!PD&8D9
z8Y2CmyVFk&8R`m+1L~7j1xSOpW~QX;h>kXe2iEUUE2i_)QLeR)di{dq)ij@XYiM}q
z&`}?Blj{CW$>^teiN_JFGbujZ9leh$TAvCMk{QL>Nae{G%|&v<khQYeOcmu`Mg8F*
z1>KuNsNa}9hOiGF`RvVMUn4iY1=v3l^swu<n4je{*BCO*%&vOsfR+D_&qhoVU20nu
zJYf#z1U+ql*%f}GSibfuaK$B4V}mu(mJl8CWYeUF*gaIAG0<$4E>H?<3H_)@U$KE+
z^~yJwgQla^eS-q8V4vSJ=xeY{{$>0j2UCqhEBNgp@fC+JJVI!A+qc(L(ebQVs!c8A
z`<L8UWl@2`xG#?sCfRoahRF4p8l(uXQ!4l5w*3w$c1QY;qPm>l*vrC-1I4gME=ph*
zBYjX9afTFrcTW}%DlALVecy<XmZZMBl2PMpYm1_vo>sTVg&T}6?Y#Eg29F0C3HEA)
z$qF-kk@MJ2y>(e`?G4&GntJt%mGYk{E!NHjjyo&K`W1Jq`xkfL5|@l5ERC~LzSjt>
z(2PWbo$xoS<rm|gtJ>%hr}32M+H~n$wHICy)Lpe8`0W+mV053xmP&p-fWUe({QBxx
z$0lNG!;r6sbiAr5BHrx{LJHD2sFFHGyP_tzOIb$`MO5sB2J9<DRBptNNFuGMfUDz3
zCu%;p1jqdha65r#Mmi7wK2dH~jv!D9gTsb~gpau0y4XOXoE#YD$9+);6U0E-O4-@<
z^v%vtpce;)2sNXraWVY~XOO~O$gf@rgV?b2tEUMGqLv5Wm0SsXMDmdFCusw~5_k!}
zN28`QRSwgPV?;It1T46IhCDf20jP2J1Ogx3B)=x7v@XuWBk4CpNVg9&$yz)lrSW{?
zwb1glr*cX@#H5&<CZmA4%1<1w^HcZ+-or{~Tz;q-rGWVnh?W^T{w0r{y6(w=n|Meh
zMWD4%tw@1wL{0!j$Ile)Slyg*C8c9F@;z}|cG`t4*xz}yXU{e8?i!j<RG&`Nn={pR
zQ@==SvPGL6C+*oBc45Iba<^XSj6z7U3J=lnmS!>gvh<ySJDARkiIKul#SvMj7``cZ
zu(G3y?o_9@JvGj_=bv?RUu0Vx-Vgek?+?g#E;3MV%HA5%tf@U&8~mF^|9r$Z%;N@I
zH%@R&eoP2=a98-&aV^l;Cu>2-(~WhBF^=mJ`mq4Zuy)V&bgtj#vYR{F<`pr|wXNpd
zmcwPPP#n_ZCUnMMV*LM)XQaTwC;uypbK+DrX-&uFu2YAT*Jyrq)RY7%E@hlE$L$HV
zj#7#wItDySig&0GDcD|aVjginOx}C-OR=X-K*jTq1pa@eK>@QW?iKz?lI7+z={s+2
zlx5OocDo?urr^@2k+QPVWd;QM3Ptymq4+8uxd}kiR#+a53i@6Kwb#xO-*m~=J7+!7
z7$$P%OFuyJJqdELYzv#0A=y;dKGkNCc`R#1z&vsU($&H~Ls@G@47&frfk;<~S*+>0
zNE`@O!;^4o-tI4flg@+O@W3<?$aCNBcR*VNx}@TeA8btGjpzx~4@a}?V4i+!*$A9$
zp$k9<0|{xbbC|4qWm%5(#;XLjOB+1EZM-*5$ac`;>$#M_e52MpcD4qaAr1co2*jY`
zn3V9|{aY0WsA?Znx=%PvfzF6@L8Bn#tu{cK*13t1ehC;o-Z{cp<BAqlu>W!?t;edI
z_5g@_5+M>`uG1-GPKyS>PYu9qxR^_`AdIe<)tNGA5eQ#mo9jk^^E{}boOVrmMUlZv
zdVnccJMXECMolc8oq@<qYMJ=gC(?V7lfj1S=$aS8G7G<SD9Hpfp=G08`LON!UO|vh
z#M+AEFyEC!?Co&axOT|^;oa*qH|6*TVCpu3H<CQJ2nM^Ccbs{yLp#PFIevD4hOY;j
z<Ho@mOSL}2q}MP%=Sp|L9iSJ@CQ?X|nNHN#9~%fJ0}T%nI<Ys*0%Aqn<45Xwj{TMt
zN0lLAd@t|bZ}SW%{T(rJf$ov4`+wz0d+#EIT$hyeKRpq3XQ*#DWTm#qqcEi%fNyZ^
z$C?+kso2kH;y93e0R}v(V?q}!B`?WCem&gfL>`T)r*#N1zuK>=<ahCQ=eQ%JADq6{
z!F{tNBXj$QW2K8r|Mma6<FgU)6bR#biT~k~`k$QiUtj5Vhf7(kxvm`k57+W9h}eJr
z@r^W`IJN?1{yZE14kP%RuZwNN>y@`REh+quruF~b%<fH#@Y;;H_<wha|Mc}SX;^TU
zW63uEK1udJulRrb#swk&!vpPn{-ITaygmQhFFQJT+$*gV`x*c1zW(pm3edSXEgIyK
z8<XH1I1RAHPfy{voJM3hc;>a=go?8}<s5egnfY@w<O-4sI<-+#5%cHUit)mC$#kB5
z^}qiR|M`<`uz4eARNy<#sktL~Lt~y^W5)c(yGmm203;)52RNJRcW_K|fW+hL6_tHc
z_g1*TX9CHKRKAEe-n?wm-(oZ2mos3ojT47krK?~B2nb?+&3X$)JnKQziDY7bn!iy+
zo|A$(TGyv!WDt+zuEET99cG*qy;9Thkbip}Yz-%od5A(}YsCn66b)5%W%;O@`?-k;
z5Hu1&rX2OKyepnh2a9Y%*8(g7v7*Tnf4v4%&)}_?!jxG4lfgz9D*$f3Mi?QieSHZj
znGjRH0Fb8a_EY*uk`FO5h$b4)=9UjjyJNx$fUo3%E8P!DEBY$n$0xx-ObV;}isRJc
z$0LYbHR*NqG+O$+jikWrn<w{|+X0BCK!gZu$WHcyKk(jyYS?lGz<Fn}tgA32#~%sq
zzXe4!XK6Nxq^=vTJMloR(&HOIt|x)#J_UKP9XmWPQ?@sOC$_V(Q#$qgE0HLAKT{oZ
zFSnrm5-nSpZ4~f()*z;ZsuR>%I*Vqul%?+W{|1lO*l&98a?W?RAWzhLhOJBCFW$ks
z21!qz$p0Mw_ho|Q8-G+!E9lI{a>6~I9ALDPHQZD6c^B6i7)-COVU}GCNO#<u)+ZJX
zhqLd(8$h+1B6yC(D<b_+J+PM>7xHS~mHZ0(At_kt;oyPz2a{B320;R`&;y<I6k?#<
z=?V8BuYp100c`K1ac=PZP6A4O8j>_U_XQ-xD6q3#gu8{crk|DKK*^m1ZtXOZP~e%c
z>;I538y4&3d8|rb5Q1ITJqUdEI&{1gikASyvmbxA@L)67^8$Q3{#<EpDFdnTYt5;>
zr&f&OSLF^wK_uV*<?sGV*~vkls_I6@U+=-j1!M;bt6u)6<bL5ZUm5rAYFh9r9`lkE
zehu_$BVa>krNYh+Kcusla3_!yrV+GDBlOzD7Q253#+w-FanR$pAQZj-@eR?v;Vz+F
zIuk6FhVBuigIEjSjk<f@@>|hc_&1_wX2TCNv_-w&SU_?)l5{y6t~$069e8{n$=G^)
zsB`#}N~zid3Q*_%emfLhd##HhlBoJese1szK?4XvDpR%#+N0T<?rR=*j|U?y;YpyR
z9K6jS>THtK(^{NiCF9axZ;d;$5Y{}*z44cK>OWV6fBlVEIN|QMqu1EXYS-geVC|h+
z+;EfdB=6r4nqQj=p}4D2f6POY;0{m>XxcdipC7;PH~it5qxrgzdHu3q8=W}h!whWw
zK3w@Nq;GoY)UyJ(`@{rWAci0Ozh#&_(d4_!!P|wnn(V=soGA0nL&3j7aNanAk~(xl
zYartv*04y4H}Rw$$=rwuLAzrO@OppboPcb>hetdh%JmeCkk!_kScb{1t8n`BMJ&^c
zu&#O+WkBeG!Kd;*B~f<`Xkp2Phi$bohTrL>V?0e?F{-L0{9_w&e>Ki$<}RB|!-H2L
zb|)3?jQ|HE@8}`eGTAL|aRt$I`qvCoyGU*&hT#_Pz#Tp4$%*99?SCb*&t*5y3T?+7
zp=V1tYgR&E@e7V*iP=AGRRoPy`aP!3O`riMH*$gq+6GvTmTY&&Ry_Ejx0S@k&<yFg
z)bPljHu?##(Xn1`j|I(%6BB#m+i?#Q7zjv{gt5fo4K~y65V4##&mtPlzR3I%8AHlZ
z(2AQgGSOUoop1CJsSQos3hg(yej~xcgM@w?_28#*7Jx8eQ^l5LB2ro{)d|@T@F844
zz;ijWAzaDFs`nDfoGQ|(SKBVW@C73Tl~g-Um}OxKN=|F$WR7MX?$j*0MIjxHiL8J-
z(MoPNFYSbZK3Hzr`58<!c4ePS;J8?kYcj~b*MY2QXJB&W1D(u`peqct?NpNZydNV-
z<(iOSSJ1{x0`h8MDtB&ar|d&W5Gp+`&<TB&BNMF^lB=8W+3#A+9u3LMFOc6l+<57V
zwFq4+^th{SKfn#eTL3aN&4PfA&!80(-hQ8_^uJN!2Q8g|?vf<x;3{&8dgc7~3#R6q
zbPPx5%xXt-5?gb^`xI|{a$jOgjeCCKzF{rDn`Fuw*SLu^s#Q=EP*L|gNK+Rcq-*=X
z#*Cv4bHk;!O}8~}tQalrEV=y_v|zl9N)}Gor1RXj`EhsWc7kvf%XboyZG>gcNB~G&
zD8*Ju*|u<fBz@w?*w)(s+MhE1LO15vYr-69a7U*|ku_|^sAs*z6t(-Oh&k|r&Fnz;
z3@_o|9uK?^<=+gDb_7aeID~e-KH>Tn11*3HryCN$2_i(FJPOr)BdO<3BaJaEy4vsv
zQxVMkRDMaC9vJlV*jgKin^XD~`}v9<bkGh=xb!N#Tb?5y1cfaO-<2>ZX<9_59{_Tl
z;DYh~h~yYS`bD4^PpSFj6F!qUfmh*m@VQxt-8u#{EqCZUfGXz-^!`AudMl#3sQ&;b
zeo7zR^pd<uC*+h3<J89}j&P*G@dYBJC!yJGe}pzkXgVl-&MjFKc1^dC;f2CI^aqS3
zz4f=F?ljjdtba!^QGfl9ECAj?4MuY~MyMrSF%RM?N5-P?`&mcsS=YK<hfCMiIIh{u
zV?LB;4cPFs(!vgZlCVIsPg+FL6L?+CGEB?-7*Dc+ZNtIFyyyN9%t32}Namru?7O4|
zk6^35Rcb`7R|#l&N${46=7t5}?6&9dRenR1`L`aNZ_l4NSa#$Ze3<Y)!Ns)HYCc9)
z(h83CvY`T$SB7ChIR_yt*Qi{(S==Npp-ZD}2Z<NLwtngl*je7wbURhz<3+A~SweO2
zvX8boS;S?tY4FxQ#_Dd>#1+DGTlW8P_SSJ(rtA9m1HuqWJdz^$pfm^ysGyW|N=f$s
z4-L{Cf`IhUD4=wANjD18-Q6JF^<KBL_WG^8_u4aizyI)I)FJNsiZhP$dz|ioooE9X
zMfh2|LD1DW_8;=&EyH>_Y-!(~Q#cwerU{sM(Wl~FAQ>U8;0IBeABDqM(<oAgmO`mk
zUuXuhjMxeDY%bJ~#8f6F;R{j9ZaC|<(i{uuz66kP+%$!kAXhT=eSTU4b?AFV1iq8T
z8T)cviVFWYXT6J;p=Y;Hh)mVEw5;(!_1*;p?iJgxZ{#=4rCC{r@530X<tm5EX<<_!
z3Us@oi(tOa&Qv%)Ut9OsCbv4e%F}8{+}5~fu>{af{&1rVWb7&RHJH@u6zLehC=JUB
zHD>I^!tTXBK6~YbX!@Ld*l-iI>^Sdb^&p?KS*vxvc5(ilQ|PX?K!a1ooX+!}1Kj>E
zzjRQCDp0a98b4h1*V)QBy7wDC_5=7K7{a>)x{3zWrX&a!J_p39(6fCog5nRN%PCqE
zN?=o~2cp(IeLHa2RwI9F!bHVJ^A>@_&c0PQ{dT{qF)ha%*AxG2OE$>bKQ1`z+te+T
zJ5(`nLu&O@t2CnrWTed$zUY{akP|^pjSeudWOJY?838sEoBE={3c}^kJjW-vJ~NsZ
z`zE8QjPfGuCBxkzb_~#yEDnmskDxAlP@<x(8oTT*9uhXL*-=XqtS<|HHbAK_*eWf8
zPJ@nMT=7f|;eNYE-~2db6<;O{Z;iag2qKPmVz{9xfe6q2bC~d;_r-wJBP~*k9WX<O
zmlm3VhZ%yu)Ks_SgU=!&lD3@;hG=H^x4y;;J$RndDjy1nFeJ|*n{%$%<~WX6HGD@J
z_E+C3I+=2Plfh7B&x`xRxJ~yW;-C=%KhWiyS+xgBYf1{XBXsW;HRHtPE3v9}8{cZo
z`nV@4^Tm__Lr!2+?$j;KED#|U_1w1}a_$z`ZFNic!E5r=y+76lzAQZnO2(q0)z=#!
zM_aE5p8KO*ph4}Rh+2If%yUu%sD6*zPmykI26fY#AcnyAfsL5_-Y|SU2$sL#Bi`=2
z_EZBI>4K<(@8Ri9f$(WbU$r$J69)H&n1z%w;pL(ihD0Neg5%BH`YRi*EJ%ibfJ4o@
zZ0H9^gdn_HH`QA2MV<u0dBrdH6vZ5)pZ5AezReaz6YfGR@aFxD;A0~bqrd)eT-~-)
zu}QzHH}<=_!6_Nqiby)R1Nz;o^9v|Y=75mmXD?}M_vy4-*?0;^Oda2cpZz9U%sa;4
zGe|viJ$R2wgSnB+k6=r9SYYnQ!t(~f#Oe`qQd>Tvp35X;zJalW^Fj)atKf8fU~Wdo
zYIo1$sm<KDw^yPgjvKc_bdsL%<nqfvFfc(VPu(jybL&?d?(^(jc8u&_m7~AP^ZU}#
zq#H?C)io9CctPu!c2&bJ3%L-RV7AqQd;z9~QUVuJu$odt{BxfBV+X~NY`2JW3hK`9
z&E#>Se3+daP!>H_EJi-jZU7>Du^_^gVLF8m$KE^%V_=X;L(2P}%r^p7dt*)xvF>Nv
zsrH;rAy10%PWxLC%A>JoEv>3ERs`ElrC=ylk#kSwvd6KWK<1PnK2r3dmes;%(5<`X
znD9T(8I*ZIEE8~-b=p$d{k%z)^={~-pF3J%kyOVxA8?5IQw}X0bPVTr{dOW#3zrE{
zQ??T7b^_Ur2tkzch76LF^<2FSGb*}zX2!(Oed$7fVFa1d;bvcW&Apf1TZhG%8aRr*
zW7u4Y6skfZf{_b_3AWPa^A>S&H0n8n-^h)FM8blAtje$OHbI9dL8lSkMZJi;UCr<A
zL(Fve!Nm3jGxMRF#S|l1BDaE|T1=|QZmtP|P8k3cy6DKjRBb<D_i!}nbaPax3cqL=
zX?5l5Q1lbQIJ9L!eG}a~UlnLMowVmq_g*?Ss+x;Kmh+1wr(X1ISo_{AaXC{0k^@Sp
zqA%iKk6u7%%QqVXxCfS~7089}O3LHS6AI)tI(bUOTMs-3LmM{e8a`5~^LjKRQmQRh
zCdt+sST_<TKL^$&i0Z*PukPu?ML*WlC8~~`i*HANwP-8o+?_NmYdYO?rlWQsJQgGB
zN4yi2dP=#CcP9mFB_%Lfqj@M)#%{r}BlzB8ulURji;#tkC+k8n|5IuFwTjM~lwSw!
z9W>NS^ml)!Jq-|uy^B1)^+cbq6#<NfJj%X3w-;-1#>9qyqDEz6ng+5oq_VPB4e(RB
zdc}qsXI?Ds2UuaoKi;%~^l`L5<;xMnR%w<hBwmKoC|gB8$SP(q+5;(dSd8prwu&+R
z{sDVp?4<oG@llPX0wE64RgJPb<ofx3?77v-z<{S{5KNWU_lLHw$(g3G`0PH_4QpqT
ztz<L8S2=)m6-))kCDNk-yNZo*<zo-S?WZ+hw~ms4*(bezvQ!b;w>IkX4J<;R3*}nf
zMcNRy)vCAPW_8Wv2=-;ZG3|Q0cd)(zDzTtiH3pvUW0GS3bb-*b^5mEF+q9BGefrrN
zv?+K&+{@g7RR!9D0s3%4TK$O>AMscZ^#_e9Efk|P)(IdqF>S32@P*G9iG0Sqt7dRo
zCF;lZf#DBNi8Tv(G+Lik(T8Ur9WIrho=snpPVN!hg{?4NpE&aBKWe=Uxm4auyw;e!
zznR*9rT%>6gI4r>Wf$_@-b0|H_6Y$hv#Z%h9(zIG&aE=b3<wfWwUWMIra?sL>gl!>
zC;E0Bz=bA>#0B}=;S*Fz6mxBzKaJx4TLtT%w`p~>Kq=#j^n*o4DT%#mK;e2yj+-qa
zfny59rGA#zzYkC+oIVJ@8y97=SQ7qru$4Wjb07uw#!|*biU1W&A<k&fT79TRKvHM`
zB~vI_0>v3rlQDvHV{&d+riIgLF=YEmz*QWI!}^liBMI$;Lx!BwI%(1KM7lQj9M;c4
z9?zBs5RQJUk)6?m-_x+WNeC!@3rbi!q1)C?70Quv3P1j1=zF&6lAzp6=^K+p$o+wi
z@Q^b<?Y1vh_6cw-oMks1Vvws#kTLr>u^wVkx|zB*BgI-Cm5KAfV!etb?RPC8Gf3oE
zcal*1i0y!km}my8ZD(EkfZIgqy{`T<xi&Wt5TtcsSEltnVEqZ~n-)wgOX=gS5$qlV
ziAIVr<6au&hnu<I3QAVXYvg#$f*!1|2S^O77RJSO56&!%e&{LvW;+uqLUKU(jn3)d
zFnWyIghpm}%fvxKN=s>%S<@^c?E|hW#X7s5VD)av?RD;AECmna+v}cSHcpA@IB{Yy
z`r{az>QIl;w#I<dKs&IWSmFzrB~*CT;$7BEe6K6)%-iSYKOGJ}QLDrh!&h4Mb>5dX
ziG+=byG&;8b_4z{qugdHhNs()C&U*US0l(`jd4t7oUN9|ljS0ldUi}Ef1Lyn&rcUF
zH$5Bwtj6+$rP+$wo_C8ZQKb@v0;)oK?*&7}v~l0;n!mbxKm{^Lpp+_(b=kNmfT?M~
z<hKG7rBl)mb8Q3rMA+)>jwsX%jP1H6Z|>{u8pK#~sSXu+1u!tTZtOMF$%*@rO^8+*
z!X29-UUmgAYtKOQUw4uEes4t8Af&m5`>v_lVZCm)i`Lv&<lHWZ)~zGHh$Q9e8)v(}
zOHadz;p3tHaJKx&b6p>WuV5C;5$QonThem70ygZjwYPp(TOplBqhkGa7m_e>9mr}8
zlA~Y*q1|IKHBNUWmiX=$6?3=pb)3Hde#CF1z!v&v^ybDk_0T6!`nx66V!r+x3iT$z
zN8x}0OTk)wc?PYy4@e?cfm45Ep8|Hx%Mu3`GJWI&SQ%zq6XMRuwrC{~4ht-*;KLSr
zSrU!ofYM$Q^9%r2ehVTer3BJuU6ag@0J5=ISp9uxg}#J>p9V!YX~EO8lJ=O%GW+Yq
zH8l_vTgCQ^B0^6?9YxWui03&~d2~qvdqHBn9u-ruJ%;9v9NpO$LlAq9-H4~li@_xZ
zpJKu8%MWmXmv$`-YB|@OAIXCLvQ#3_K--sRJr9!8{VaXZG_h4RiZ@JH@S?T=`C-9x
zM>i5!0T`7=nV$la^KYJ1xDW*t?e~vtBu+}Y-`xeBp*L)OO9q_sRmVW%tebt}QfpjY
zuZ1!PQlLe6YrqR+w%2`!#DFkBtML}AQ`hxIXIe#MAP`HS3%?@7By1@B=E37}P+-<{
z@{JoB$a)Vdb_2AJ-mv{#P6Y2u{^_{_EG`9hHy)>$%7+;tk4W)70_=&Eg?zV$Ko{RW
zoV__<U=_u$d2nBlT^@EIk6R-Mdj|MD0#VJP%{2)qhD`X|>IoxWjW~j=sGL$5Kjl&3
zZq&Z6proKVG%*W`cJvRJs1epFZy|Y4(|}Knr{Jak^c)^Qi63VGU+U9b%l(>rPxhuA
zE`0&}^ux{7G?*HxN?$R(cn2%G)1W1KKk(TC@65gUE)a~rRF%~7v=W{Qt11{C0J{qT
zd%#nc?u-X1am1(!Ms=jV{Dk8^gyo5eAA#L;6)M0lWs*1lUn^|?rAr{tfM5qG4G}55
z@x?e6RHZVduzobJJ#gjz+`AN|uLYCH{d@t!1ioB6`6{9DbKN^v0I-~%<4z##T$c_j
z$l(Wtos>Y2oDDGiSNIH|cSM6KRFr2)_>;i$W{QbPdQt%5(|&|q8z^vbB|OX2s~k9@
zgXOs<rNct$zA?YBTz`Lc9dli}>y*=KBpptQfkQ^fDyf=NU@k*SN}_4qgp_Yh5qT9f
zbpS?0!l0RYFn#kdHC^uLLbZCXlWS2wbNonth~(Tt{!S2XNF)z|)-x%B)w3(8GL95#
zL?ZI5PmGgTOcgm%gu^OYcq*d4_NSnalo`vDxz|ghDcG_o$C0KLXr<fojH}yqBIP$y
z0uLul-ygl3Uc6t_mZC2YbN@sKW8vAdr%MPTM#=qYgB$nknfiwx?^nxQ`+V8^eapCF
zKjPCT6jkITopn6@I98}&CDefw8bF$Gq>4D?53M>%-yt`S(|hk`KfJSK^~a7?n~r_t
zY(81k6|zGvu8K3ki+fgSvSBY5SRILW^sZL$z8>%+P9L(suenc>K6ups(cOIH&j2h!
zb?!iVX{Q1y#N$Z<`~lT2olgO`AY(9LrsCI4P4J=vSU~cqW1m&K1M!~p@6xAI1hA2p
zav0EE{Fxt)d1b=>H=7q+>rYYh-E^<l<vht)u{Fv9^l><LX~t9f+Isme@^)VYVJ=e1
z^a7n|^J1Nv+I!e8IM!^xbd6|OQJhsxXVrr0DEx~O6n24eCzb^@O%RtZg4Ai-30&Jx
z3EvWFCGVKgqRfG8dFw<CWVfm5_&;>i&)fk^v1(iQmB=!vU3?%GWdEf!2VLus9|X&Q
zcwjxBojwsnAp1;j$z!pE6GUC-L!+nytpdc`k096aegtZD<6BUw&Z}V?BxN~IzOBhx
zZkv-cj_4T&Xf93IDd;7Pkb~^J!#zZU>|26?k}4UJBz>45hFE(@KRrFn0c8mVgS_wc
zXYT;Kh*`i?P_)Nr?IB(+_<3IZ&5JGS6>K0G-79Ysi!QEF<g2pRV}3}rjD$b>Nc@jO
z*XaGvn(fm-VDjGXj_B=@y)E>x_0y?Js}#WI2xiJ0j~{IZ)bXsFd*Z_$Uq8NdxM$H9
zTYLHw(x(#B458Z+b3NL3L#Bhc>x_?Zx-e;%EMfs-*!$yAQNl}CZ(ps35$|4saBh7K
zj93eJLdFW7Nm?N?f2r%&twne-%z?l)`&_J*4pF+i8HCV=@BI$=T6+QJAfnY4o{C`=
z{`ksG^Z|29-*6D;d#8X4Xh4N6v`bWOzFyds7xb02zJ*&f47&UBVTPUR2C*~Vr+6d+
zTWYbY;1`AoqY0z6DpmCiq`LT$N|oOa)Q={yQI(u>0^FHaAm!p#pO~;SG12%bmJE^;
zN-DWGuUlkvtAKS{8$_0bo@2!I5-pQjXG7-^s1100iHrE9OMTWaOmKF!zvw#*YNrbk
z%lMyni(hnwWhn_`U4r(`5^7mW*!m%f4c>wNy0MlxWK;fnTb?*jUF@Yij$$1Z%Nc)S
z0baEMqcO9Y3DnDi=DwF5)eROEK1HU$EP*{HThR}(dKRGo-QfXtF)8Q!kXa(rr17;H
zt*|TYW4Z;N5#@T-YQ0LVR<13@s$<YW3&y<78j?3jh$^u4%u40#c;YN%`>m%rhRrO{
ziYN@<LFH==5*B!<PQ=HofuD>JZF}1HhG~q`!x9YXJ9-984|sTxpFZBeuZ70hwyA5l
zkzuw)x1TjWZV(-OIXT^&bU`3NF!_TR0fQ#ul{b@|bzcY@OCLxk2TVzzH3H%sn|i0E
z)%#*<7DGOGOtUFyZJq2bbqLlZp67z0)jPF!_yu`E;V<vWqFU4mmT-P8IHPHMsgl%Q
z61B<(aK`f@Vr{lh74ZEZ&VD0KZhNK;@WhfYD<ZcFe%M48^L1{>WjZLopZ!^eGxE|l
zp!d7jOT#2*l6mLQ!Z!aEcRH4M;|<pmtWe{tw+a4_FWIg)92+jQK7r1yCfc`BudH-X
zJPZ+gvNUZU_%LSIMmN{FA?ZBjPhG@QR04KtTep9-hyEh^a~E%HCC~$cNz7yKiX0N4
zxH5?Y-l^~0f8^z?R*D$p<}CWB6W|bkeKgi_rVH=!5sAM|w{n|~4eMBtdkkYqSAMw}
zRCEYIVQNjcIU#{Ni(Ovrblb}FKFct&NSs*@B#2GG72sm`K1;c!6M%Eo$$?u(aQ<jp
zNj>VTHR!2f5QzDL&fD`ifu0LNjaUE;VlRGOl*QK6%F=|U=y5wJgOCMPyT~r&I_z3O
z`WqojN=PLakSv4wN8R-~pFnn>!0pTA^irH-4ql+qpld>Gp4uB;bs$)NyNILQx>a@3
zgyz-6C#jN9sno6q0-xK&+3n-%4?gNH09Ei0T~v7TBLbVVPffRoA?y-+lSGkS@=PdJ
zq7N4NJNGO+y6Zcj8RD!8+M@5=YYV6RDF}RAbXZ;%PX3WzeuwVG9l9cHVn4PxaSm1W
zS!*DFzbI$#Bne7et7wgJZhpgf4Dh6#t!W-s1YJvySeG5|ZQi<=1wo2WAWE$qv3E~L
zqB)&CAu#~&Tg8EvjiPd)h0JFn08cv8U3jWv27-M1BBhy^GrcKp)$TwTaM0BH5#@em
z*=Uom#5~lO9IHrwic9BfU(@^a+p1s?K51MbaD8M~RoAQJ+G<K^^V<Pa9w;kNd1nl`
zYGQ9Q%^LAG6l)Nyq7%fu2O8N4_mIHxsnH_Yc_B|0$&`~(-@xd}xG`4nMn;PQ)!gc(
zy~uPw+_c0lFiT^SO6#s}(%0}wlac@v=@DI%aQ8Y$kI6?uhgmRntHBnicYJR@s%2Nj
zHaAWjK`y6pJMOlfvIXI;x*Jaa1f@$4ssv7X{UU~Zz19GDw0&=_T3Wa6g7OL0sWri|
z-!El%-{?hHL;L8wK@rBY;+6UR(S~a-bAZvdWFEJ%+bF#2Q-54zcpAt_oD})Ob{Au}
z0-4z4JJKX<(({LX`$3C>JnxB}R)5TMP8+Lb{N52zRL|b8(Qqt!Dz&|3AnWc1q`)np
z-LYoYizrJsJ<)mL-^FAIH1!usF<SN;mG{KxEs9p=KlGMn#a;glH#C3;PV4B8z94ZP
z{dO+h(?+#GQktG5Q>5)wSoCcuda!H1rZxz2S82KqZvj|+621Hcum0IDEU0!x;%(k1
z3`u9Fe>U*fb|BTx&Xc^ozXoLnlc3=js4CHuSa1&N3G+*!#lB3&OC;7}qqbaFq)&Mn
zRJA}Q3Xj<n97RZ>2T>E_cM}HIA@OMN7x>U<12EtZ)v#4>jNvt!pB{sDO3^n&(C=2W
z{nubj<UUJ+LOPDHIA$?JunBPCsSm@%oX7JE>kr6a2*8VZ%Aw4JSr7WBF}mv(iMb*_
z_QM*3mjOy6<Msrxwv!&ofNEKpof@MC+8@Nx?c;-UppJ2ygeIdM9zi`J*c#gjnth8t
z)Xy|HDM@l^UWogJO=?D-fA9(m+YETDIB)Q>x!FyE1;h&6Bw}w_7jRXn&3zaVM0{hT
z%$V7K1mJQ=GZ8y?`02+OjiD!G*Z;7%xB0Fd_Q>~E2=*QPMs?v9(adOvI}vb1FEsqr
zgt*H+%%+dwn>@xZ^l*eB1xFY+pg|Smhrft;&Lm|@2*|mfs7Af4L~rjdKgDQ8Vt$L$
zkbGH_GwF02gGK+#1@p)=edUz|&%9f1Fg}n$#n>cAS`l*ui@0NBj_tY!%!An@9;&AZ
zecyKh%Ek7Q4&(u<kTrM?-FyrY2f#M_CQ3SlWK=s;V(#PfBlc70f}j!9)m|&};eq0s
zot^^Mprv&Kk;axMw;H_qdJ)dQ`Bvr0KP#$$Tb9<x%aQ6min@HpbEGn}c5@x+CZlsY
z{rFOt>^65x!V;A}{|d%OUuJy`-?JC|_>y}y@=YVEW-nCWlaF`uN-ZC;e=Y<vlWjme
zc2w0djD9>Smg&ya1;w$6WqNRf>DiWB7t7CMSuvaUvd8~8iXS=%HG*d_^KrRaddwm+
z&v|d}Xk<sNJL)wZY!VTC&|tFj^qdd7f7(KFN$Kee@>)b8xK0`csHsur1mX-s<o_lD
zqV@{E9ii|9eNT+GA0=WH6ic~C-u2O~BFGo1Qo==r3hg{wO@J17nOflsPRbBmK0JWU
zZHl$jnm@%O-A6Y9sT;Lbyft=D<8C7d8yeYoGlqUEs~h>s3t{rZWNzgB;@4J0Dd9Ao
zDTGHMNHxqo1>hy*zX@D<6#c7$!j8wAtMG4U_;{IhGG<Ei=T;Ost*6H_vBdGO5OC-$
zXxa<@#PGSZrHe}tZMf}!F&MU0yTk^H9^gEdzc17$ihiFaMcW}bq-SS|R9`rS7ZEOi
z7FRi&G$DHGpV*kp)l4Sx*(|&9qk1xC#~-gr<w=j<;glq6X@qR$Wb$#_eyXQ^tRK3k
z^__d~2`0z=ZuAAbk!M#Tr|qZfkIX`aY9wHB6c}6ZTtlgu_wegEF~#qN{YtYxiM^>%
zEwCLhE(n!b^D8f_jq~yheUSlL3Qe)p8N!-Bcx3o+J;sPCd>&;IhR|+WP3);;Q<+~3
z?8MB|ml)9KlbIKhRLwtewxrH5ZipA+luW79w;JPwqj$vqQT|cX-Dp|{Xc$6dP0m!~
zH&d5CuCg%<7VsQR+T|y550oZs9$dfUx%|z8UZaFB(#3Zq&woj=bamxqZ?g8$?^0H0
z(?e93NTz@WR^=l8`=v+gRTWS$(KidpIcANN7XKD_0SZy1k1)e=0jphjLIXrV9YK7#
zk^hKD`nu8Zh48>|yWcQq>i%3Q$gzU{<~146@JPMNe*e$v=3jmC`8levfzW4xv@WO$
zDb3MF*sYaS`p^Hv6cb?DZm>Wv-R&6|M`_8DPi+Qa?p^VZ1;5%dw;<^2Exi~n(Z6ic
zgL)cNQsaAQ4ajWH3A0f-;YzJw7+RK;#}26qEby8^VT4#Tn6y9BeW?2%;w)V^8tdz`
zReuhHcBSedSol!C+>1To)4ecHa$qa%1<09CltRtKplx*yIqzN^Y?+$><JtVB_)$x6
z4yZ_Q=#!lEPl2X_=?e14arXq0t~>uBkzI~+_~$d&AfO(Y1-&}0=Xzjr4I2=VarsZ1
z@S34PM2M-^$*~O;G6X#7p<E|fFXZp|vCvok+>7-~?geUNxQwW}^{-yNzk>yYjh!l+
z|3p&>pm=*;ZR$WQi)dPi3fK&=&1=mKp!{2<P)GqiQ&PyCn;f$=kK@CCDW3fA{_~I*
z>MYkV4txFc*Zn_z@*666@EFZ3X2JiP2mjyy%~dEQ^01WTqW)j3pFda8zkc-ghM2$J
z>|JE3e_8MQpB@IN1A=$f@KFKh|DOm}ml}{It*L*Mv@`r2ZuQ@;$i5xuDA>tRy7j+N
z0h5xT0oYFBE@*c912^=u1{)i>=~rm&MBYGQ#B(`&bEOjpWP%;TtoF$~XiH`rUiapP
zA$~{z$NMvovU)}K?f$v99>Ah-{;v2&LU0NKM<U31WEYf9{6KAm<(H5ZTo8a($NtW9
zm*Rz!HOWc+jl(u5ePM#k#5>4<bQ#ZM_Z`HUJe)-_``r%>Xk_13zWWb>tlw&u0$Ku%
z^{}#vA2j#8F(g8w#d+sGcV14x3|>LqN`e_`N}c&Qq)uriko%Nx>f#CFEJ!W$OQfo5
zUfh;Y&*r)Sp(A&J!tG5CgvuxaLjE@%C!>iatErFNyCDU>HW;tV$=+x!_?=)mzaO-A
zdq*#UT2_OLzwRm(%oA*1=eYl0woq^^8YF`SdR%&6a3uKx%H9J=S@-Ku2<Qzs^;ruD
z+htlzY6Op(5+c9vf~q6>iV4s$V1r>T=|a1uf6=x|7nT)@PSQ7iqOxv8nFp9YObiOU
z*B3bJohWj1KzerviiG@r0X~~+AV#6WC=!DovWD`5(4Za@fbK?e4Gu$J5^bj@zr~P(
ze8=D{ePPm;9>`+L1yD7`8c%mTZ+P8+HTCfUa4@g82Ct5@HiMs>#g*@TU=*5Gng@S=
zY{J;`=TnlPPvmq1L;v%j^hKW8l|k`QK=>R?F)ANLI!-{QJZ=XcXY+<YA7SyQBNI^t
z7>gV9h0}`qO3=kW(8&MwJn_BnfHUC<dl~cJY>8<hF9uoQH>3tZL0*)M0Yg-G5?Dr5
zFqQkEyeg=f6WSMq(FHqKfwqL^7&JSKp0fK1G*asZK0@`{7r>%r9ug)*9}goh&>%C*
zB7niJIknoUJF#q_(Rx0IMeQQ^`V%(l4v-1aK!vsJocS1INUp00N1X;p$9Yf{NEBmC
z)c?Rprme9v2QoZjd}`XaFE_GMHxfsRDCeo*3X!Vr5E&CFhRj>c&Rd%PTw~xzAkqdL
z*5sJSTaw&+I>p**EN8diDI#HApaf2hcnO>bnt^U{ZKo7EyHRhPIs{2Sb%Nzi2r=+)
zzK^L&&7So*HUs@Ng%M!i+a>%852#U}gEM)>Yc2C{jx@Vl-jJ@E`g{`L7+OGg1rCMa
zfY3moRQ(=i0oG~W<%sMoJPu-XGUetCnEfIMPo)a*0ozFTr{dFJ!y7?}zbN!tp`5L(
zPw3a)oS+d9&daZu$T30Dy`}_3QL7eEyBvssS%VdgBaC<fyqj7;LG@#Uh++fWHH7^6
z<4?e{UM6u?M{dIo(7#6QcyL!#zyo1^^zPw$RN#4F;acv%)-(JF2)Mx}%?ALN&43qt
zSr>DXnO78+l?tMl<9SeMV?`011LxF^Cz4}WN%3fy-ChUUpldGRc{p4VX21LI_s75f
z$*qV+z~TIr`|m<2@8y~pKUn}~`$vwG3Rt*!xG+W}%(|8X%U%hMhprC?wGq%gn-v}I
zP9GZcU#Mvq{4uuKFqS|hQI76RuC3Jm?014&qjb=8PZ1!k_<`A>2h^YB{Ia~jaia#E
zv$+Z}1RVH#>~P#8u}6V1zh2&JX>cFuv*r9X@&Mo4Mi#1eVu3wgIqL|?oS_3iSRx+O
zr*h68gUH4SyUm5_l|iugp*h}*T>&RRM19F~NB<899IOXk^IjISGLDf4zQlV@3sCHs
zCA6P1*e)ME#w#SJZF^u*D%bD^^f$v1CD8kVi09d5qC?G~Tv5HxyA5gKj)&UUPlO3%
z2;TtE&du(}jj9`Ph_Cf$2E0Fogu_5+*F4zX4YrCN<h=y?Ffu%KMp9Hjz<(zBG9vF(
zj0PL$(+xO)oM!b(=6r6y>FM6VcMi0%y@t-e6);P+3ki53B=5}X3_i@r!aHo%`Xx~Q
z?`s{*LULnLaDB`dn?Ix)lVE)5ur;;|qB1{l1Pxg%05r0;s%dt54no?fklFSI9O#fb
zI5U$d9S<S|PcDLuPbsc=?s;?KsTUzaY~mz!p1ksls`LE8ANb_rTnQmjYD0aHiQG@1
zhu#JC{qAiL?bW}nL9|h4OJshutWN1-ve)dY?j#-{fw%_aISK9qq%%Nw>wP3qM*Gtb
z6IG9%TUM`|1<==(*YVBB&UMh;s{pfSnj4^;ZY4{#du90?U1#!NWtW8NC7jS6O>Ac;
ztx5L+&igXw2>)iRQfT(&LGsNZSEPp=5D@E~?#?SWhuY;)ct&ecmE9`6XZ(qS>t39}
zC252L;~1JrwQDY@(0n)_2Fnw?v~i1=!$k<>A^+4h&kRBlpM#!E)Y+Y`cP3)bMhgn(
zXHH|H6cYFmy6y|E&M|Yms*hfDfH-=-{^{(jOT*+nvHE-?3!lwSaxvcehDmc&a0*iY
z!CL!@2254S#`Um98k$C0FmpE@gr4Ms`Inqdde8`Hx+ZU>l<-XIgjyyLz;737;^cG^
z6Tb*k-Gt16Un=hc`zc~0P@rcAC6f0$w-ReNjh^pcI|0PmnJWjyVqN%&z5<C3DC>Bi
zbbyF;Y_`}*_W004;ohfTq4j4Ru&XOjWMcm=4za?Yd-cSG7+y^iy6}$PL<B<r<V?~y
z@PPb*skvrLu;2=;v(<%I(1)BxHx01LT3vjKnhhA%t?0ZWR0uGb-rb0yf+*hgj`DPy
zT3Tc}k)}$5%g~Jcf)N9X&O&n%5T8ER1Rm}lIPRhhCt!Q<?b`5Y;A?M#la5BjZGH-x
zq`L(0uDnm8>bw&R^^k6SE|tw)NVU}p4F|fqwK!<q1z{rdd{7SYOI-`;`wFTSuy1sO
zt+Djx!)ac9;`PaMJ6oi>A!P$(lU*QQd@dzyi^+&j45u&yx0XYb@;IYc;H6WXNc}vY
zn}E@sslqJw0#J}FT6>+2EBJtF{meQ<99B$${!zfU<HdeI#pFi5&*?5uChyy|AEMAO
zqr<5dl*SNVv=gfGf(-AqKy!wR4Y$p*$O71uLx6Q|lJ}5HMD*VBq@E0Q8hG0niL3^h
zjouEYPQ6EJ8b2zF!7b_!O|2+&PlbC#*iDCEHxGj2YTN?1|Dh0uuksYhg;9Z;C#Iz1
zm>vcRe8>!MwlRU`@q^KQya*Hv910#za-hbUr{_Nl48ouDbqcbHmfyVvv7QTUN17oO
zHE?7!C!xqV23LK55zsxH)y}zQBzw3ds|+L<OVwO6ok_Yi15D~xec)FqbU{q`=A1X)
zBHx&YRkZAtvV21|m{kOp3^lJYXu5y}h1yUz<z@nrYjimu&mMsxU(3F(I$R>%Z1RXp
zQ!Yn2CBK*tbn4|!)bgY*+3J5w_FucJ$JVQGDf}co(PnAVVuY@$AkU_3^8$hkU;Uu3
z5ApFdNIv0A`Vu0k#&U3igpdRxuwL6!kPUx*+D7F^Eh^IIFAf!CbEgnk9mUJ$^L0=s
zVAt_^tAjA^l2#<Q6O}-C1DF!^1+d+Y@h+x|Ut1)DA)E7u=B{iph^J34cz2jye01(z
zdj6g2C=Ue9(R|2X32i6THyaloPWqb|piIIG(-xdgQUaR(x<Pnp0h{<)hPP8P0Ry*@
zpT%es&PQFu1NaKJP*Wpl#j1-w;Dx*4KUC<1*bb;(;I_$NFK1uX0u4~g=fP|Gm{Ds4
z+jsp8!OYfCmIr1P1Zl5<tv~WS3kt%(7r6p>f19SMLXRV6614u_S{Oc@s{bqg;-2>*
z7~4EzcsGt|2(=(SYw;~S_COQ17R3LCVJ$2UxrW@{Ex9wm2Vc;3nn%Am3(~nnTml;t
zdp)8>ntk|H0#Mq&T=)TkqosvPXpTzjr$)PfBVI9iiUe9n(NfC+7@$oY1ingSkckG#
z+%Q3bGS|cs9-4muZ1B_F?}1?*B7R7oaV3{C+kn$n|F<hvkbc-kV?p9U)_u#WrZ70o
zHo&rM%L|8FHN?*b$mohJJwpo;ws|0lNA+X<RzFj`qOW7X!3S*kJNz7L3J3A~Ziiw4
zlD{{B^Ou9#T5#<>jExE>Fw17hmeSy%5WWyv&nqt&)mA8j3&iuC&I%XalZq%`qeuMK
z-pU2G48~6xVKKxtO@;7W(CJ+pfxE7VErZ}Z9JGitpV!zZn22X+xIg&&+2_4fqquN8
z09LuNjSv$>O-!W!*@4B=gZq=l6!1jCd1kZQN#g;Z0+>}BPEimxh`2^7!;VPv8gPfW
zJv6(_N)64Ci0;vBAxrjlr~OS7ZA>Fw(NlS-&=MbpFn<a3ubQV?55n0t9P10zidED$
zQwjLxSDvu+ej$obs;<z|X*m+kuNy>M!_{y+y$VGzgBt|We0y6ubZheDTEBNkIkZPu
zRjPfQtA5<xAGou_{c+AYK4jq2%;4^g#8~obqDO4g<MQa~3Vnjp--WY)t!F;}E2O`p
zbkf5YsQ|ll{-oCkyIBAc0f{cId9N_^m_5aPl9$MnSc9ohsP*Npt>2yCCf&NU5vaG@
zFZj3|u=dBvcVHgU;TLx~|J&8vOGn^cbC5NO2|Ixp&Q}L+Z=k}Nw>GWrqtU-_(rJk{
zwi@^I;R*-s4%_u^y|U)qQ{em}Q#Rz?KcMhN%dEK^*}s#51xRvKLC5kX8&22$?n6g>
zsz7mxn)1zx;8;oaNrXBlT!&LnFujMb;+%_4HfJclh&JIH+l1JrBn9s=%Cm$%#tE^B
z92bv1m5Iu4+*}>fmjp7wcV%8E+va68^WnP!g{EeixL3cgeY69sk1FRE1sr1nR*|5s
zyE4z<+lzL?q+k89uv0`{29=qhmLU7vKUiu>CW@>C5q~YtuP8zx*b@f*IGNUavmiW4
za46y+;1?yZ79nsLCV!L)Jn{k}`_UNs5i7%--+y6<6EHuokP<<9h~CO{m!mU-nG!!6
z2;SVt@pzW*(|5|_E1<9&dj1LIv;AnOTpjU>1iW8m24t@(xkxSxQ7h;O^Bt=)W)OGJ
z$Uy(wB1NPnkEcid@q8iG(|+~g;=>j+|9-Sj{b-R9<<YFf{mkl1KadM=tXf+6Ar95Y
zxdHbOEbbI64(!P`l{oPyka%UGCc(CISB0);&!!{;)XXtOgOpTvtQn6C&vS>qyxC^b
zimjk)ci^VGBdN@4Rm_$2itgO}Y{CwO?=65EOYyr-Ig|jR6krFo?B8NF;8QQhB2YZL
zjGtpu=bQ}i{6?H)ONSX8gLt7aZQT%TQ)Wt#7@Dd$JKN3Ozu{)rS@P|f`gRMLcDv77
zjZ9~}pbbHy);oa;V@;iU89#sfMRMSJ%2ku(^3+x+7J4eyb%}C)?**(^&@DDl(=2qA
zdBS;gd0gg#IloL{*mkP*Oxr4=jCn_c&Lub|Wqh{o;AF%K%2ya7VV=)3{$90ih&BO-
zS%Tk;?NSg^xllE#7%BfXCH=Pjo_5_qmK>D`Yqe$>gP;sn3VxiLuTTSbxdl8PLFER&
zYt+UNFtLGEKTvLd7S|=Fef?0L*ypaVb~j<qBJE=;BlrtW#nU<S2rYzrp$fhk5I)IX
z7guVad%00pMt$!`z}}1zg><BJe#JzCQ*tXf2;MP@AknG{L^3mz$($^t-hj|E-s6o0
zRDBc+s#j&$>PkZP(=5(4z0*SbnBQT^f*l@&%d)S0{Cb#gm1d<BQ!T=STTz{8nc7+i
zrr1jiM<AHW#l5&H#mqadSoU?kSjXH}eH4shBV>gI`x+*KW6ST;my)IGE3g_s`ZX(R
zzOY&ZRQPo=ubIk+bF|Vo2P%~1Wvkh0+B}MA(W<v&-n3~0^pETnn!#3xG3j;;2UY%r
z@LgB6vMC`r+f~WXvQ18QhJIzXLi;oI@<7>k$gtoSjhaVKtye&O`<ACybIot)p3gxv
zC=!%H!dk6YmQ{a^fknPphjL@UIC3^#v}<W1e{zX;2Y-qtbX9%Ez?Escc9RZ3`!3DD
zLQ?K}V)<tp6s(Dr4Y!n)H~gWSIZ9f9(X0C)_wQoZN;q}feMUuA9u26j;j%?(GF%U4
z{#>!S6g;_^(ynO@$xJLH#MBGJ--eWVrS)opQootP8L^h}EC;02r2@^vc>z-`Y!hSF
z^KczG^*(6abg(SqH){wfOq`gR!5E~plzl_#vPQ!n3skeaf3O{a;N*bwdz$NJcDQLY
zVE{B`_%5);g=adf5SZmO@twOSDF0L|41XVeSVIT9W0(Q!nF-^3JJtk(-H3ck(^<p@
zxF8p^^(1bC+fDxzRHY6NE7Cr*mg**_RkY1)88l>O(Gfk#hQ&wH#H`%otrWNLCLBEz
zI#(UgO9I6u57}#{<EI9VY~oGELMr^6Mn6cC>CGeTd_q289iTFnuqu4c+%~gg>yXwZ
z6OW!y#GV_%WpwTcrMOS&ycD6NT<V9i=@Zp9-t1EJEedKTzP$;588h9XE&fRQycf&?
z$Bw^B-;NA*1<HsMyS6mz2itl~Mf8R&^<GNBCX@qvC6`3msj=#&*$IS=>wM-C9Cx~A
zq){|88YJDVh6@JlP0Uha1k?|V6y9Z+?Q0CYh^+pDFeY!*L6Ui0+oU1#s6RV-?iw67
zNM;Sd=Y3<*EG(%9487*Tq|<GP5?R3;?XG}~tSAF2zkjpxI1;90bKTkZDkmT8sDh;M
z6aR%+2-$M4e9-=8?@!?4)tMXt65+Il+qn0+SsX|4Sr#%gzpgk-O6m}}o?qNrUvrk@
zq^ZsP@-)J%yL|kE!;C!u0NnG8olhJA^791RbZ)+2&ZF_DlNayid><+Q<63-O^7(c5
zUw6o%wJY#>7}HBwmZm!$D=nd#H`_u6b9h(tLUlbtdL#Wb&ygd$S6oDzkQh;d?;RHK
z$Q%mx8-fp$kN0RF_qfrZlJ%d9H-Ty`a^aO6h19wBKp-?0Ou63IYH%iX+tXHupqP*{
zS4Cn&;pd<t;4aF&SkS8!g*{L}d<rNTRHQ+rvQ>DbuOT;jV)qF%6i9|FALKKv-$2r5
zq<)_`**F-sz4+7sww_;iTtFKwEP?&W@nM+Va9ZsT(T@*&9vaYY{4O^dX%x39Bd$6Y
z>t>iM|24q6{oG6`pij-tbHM2_F2OtQ>M2DfTWu>*SlTUNlm`p+*=?nM;^xBBd9ewV
zES|$Sa9G!|N}xfmLv(@E<eSvD>6;~!fV|L>YC@c@-<vdwY#2sb{W^Sv$;2{yi-u$!
zG}_(3<-$0jA(@@9EjMAsCE{>z5jjl9b5BgIzCq&`iQ3n&B>u8un2#K!&)4;xH4jJc
zTAYD%b<e0+EXgzNK(@CA)N)xpyPQDTrX7SkfRNAaWM@KfBRT_z)FvltbWTUc58;Ks
z&Y=uUFCa`c<x8jJM$!GneoevjQWidZ1zJf@F&R)lU(JU7-Q0@ZeeXB4uzHZ)Q3eJD
z=s-zpSPKVDlo(1GVWoYqjJ*67twm8(mVIv*74<+;dixQBF<svtk|Ab-3;$kQt6|eX
znM>J=6x?zDX#wAQa~g#YSOGp*f~XOvP`O8!5&Pwbxb#q*;Aru;bcjQ7t8ufSW0P5h
z3>V?bltmGH;X4l&hz|O)`bZ{bK}b>k7!?vs69nWL_~hyW#Wc=CcCncEh3i?7IQPx0
z`v%U@31s|$pBKAo0^ck>SyJG)W_7dtWm_q<WOEwR2N6MJB6&ZOdg}x4Fh{IHB%iQ1
zV!~ZObxP-HI&XzAz5Bf+bHG%IE#;P4=dvy%>L$0=Bh)!i(IQ}pc3>BjV@ENZ-VAgR
znRR`FO8rz$9u`>fyS{c9s`#Nu%CmUPXK|z{>?sCn*E#nQu$b+_fwT%!mG$Z}3oa%V
za5X(F`I3Hfg#2Ut{LQgO$nnT`U~C|lW@CSTM{4_bfPEIr<&H?H<Ju?57yZU!Ko0IJ
z`bCDW6USfJ+r*+D6SfWw!+f7WRf}@F7wu-Aq~X-BAsL!s)`L2ckbQm8VWann++`Tn
z>SbcDuurtc-Lqa_k9jbk4)sFn#1fX>0ZTol>+Vt1nvZZDCyFkT8&#Ki2Vm}0<8h&`
zq#c3x%xj_?nZN6<dG}zJk)iAYY4x!<L(B)31o49TZXM7;K~hPkrn6W$C4rb2lPO;m
z#_64`<M@jU93?`4AtL3v0@r|OCdhTnm(@w>J|Ty^8UYjSXTn#dWtQWwlrQMa5574l
zN85Xp5uk)4`48WT8-`#T6z}jAYtX`=Cd~}LRQwc@ln*bssWrHG_{?%__XalslQv;4
ze^<$T4qzKX>Bx?<Y+__9R$d^ULEOz^ZQsU(+i!`+lGERYf_+i^bt;?n#vHsDdU%gC
z`#n(gbJ0<W`B_v{DD3+Cr!hk-F|hp^u|Ed}*6~Aih+bNGzRl>8OXUx+0QX_$7}W7p
znPJ`;33sd0+@a(@L(zQEWy#U#It#7|4f@SgMgEBPneR_n$|Gj~j6R+UyLQ>mIp%T$
zf>-DM)t{JadX}J-KdRqcK(NpAgTQaabW=ew8}$R6^3h<h+QCz-jBv#U0v{Qetb&x;
zK3TS`&ZS?%#NYTc0!C*#0lK(g25tUkvIfuTc}_{#qFfd#owt{l;5wR+5S|F8%$)&W
zUplRD7{UilxcWVU;Xwq$K#%|-{`%cJZfha}4-mpa{+Pl6l8Q=*9~YO&v8Bo5b)_di
zXQpK=&$?F$9335wf5h*6il08I^Gq7NkNby#Fbo9^U*O+;_*{cEbRRaf$GDXmJ_Ouz
zb0`Z?nW0ky4OTZGIlO%+YJAJ<yd_u2QQQ8JqOKkYMMAC^YSXfnxk3t2Vc6r!YzdE5
zX#{0%saW(MX6l?alc<xyh7R9i2Rqvr#&uR@J=)04D6p`H_2u4vuVScLk|h_tNthRT
z2FB<K202vfleNuq<PsyHnfZLvjns_g!Dq-FQcg?5^<nRctqZ_I2|dW?Sd6|kS(~i$
z#U>-%%%k7K-khgB#Uzut(jACSOHEN~E-PHLtr#*RydsY?cg<$&-3_X(3ez%`(?v#4
zfExD`<Rqu}oF3Jyu<lr0`m7G*JBHeDb`5}ec`VZ~7NPUxrwZuy{Z4MYOxTJgJg8;`
zgp#bcVHFD=(dXJbRg%e4VMp)Z4y^8nCB)cn3}qV*nyk(iE<4xh6^a&wiS7`rJCU8?
zN?9oJaWA?J@zPZbjpYsq7H(1ygpiAs{=Cw%mmcCRAS-6!iB(szqJkWes;qXvOwit1
zv_)6K$PW5d2d0L3EDbs<493eYAuOm=ERwMk-6?|NSct84#P6`z=@p;xg&wBL*kOi-
zzqXEw26H{Xw{rULUo3dp)gfL?Cuc)hdveh-haFPN{gRsli(BKRx5Fg9&?ob--Uq@A
z2ap?y4cC=@xyCHO=Ela$Ug!kWoA7nKiRd69UK9SDt#rOxV1Afxzg}mmtfJrwnZ{NC
zM0VS%Rn2xq8Beo6HV4Eci)ZpvX2WJ*$7{%eW%1w5O^7E>9${@3!4729TpDZ=dB}Mr
zmYl6gh$`hiytjeZzHaZ5cSmk{vd@)|$WRI85&&CBFJyPR6o$8OpD%@*Lvxeu15|m#
zmK#<R75Bm>8bVcQGNEQlo<#Br8x!+6r*nL>Y;@3+$K@gwU>6>#5Dk{GvF03SGI{gm
zo1;}8dN9X85c8hc9%n;kk&uVmDyy2&j6pux)!c5Tn?(>$8R8P;$;7?hhNQ>i5G4~Z
z`y0CTPRpvsfm|iV0!S+1kB3OVKbr9G*#tt4q%);fo9*S<QQNc3y+}%oJEZtT)PtLc
zs?H&6<vH?D8a57!z$uBBUteELh$bh4G06ooR<cQK!)zR7(SfRZ;96e#J(AuSG8k|e
zk{1)1bQy|^CqNN6;#EdbS{uzw==oWDAe(RLXwrVwnR!37INwaUw|wpV&?JKvb6YW>
zqfRubWYo1=Gr_*W^mom5C%f7Hz4YSD4%3sJ<Dah1_d}k11@gg)y}JH!SB-qVLL*J8
z?kc$l?|MA^xi;N6dFbqti1G#;PeHR+p7hfQcTRoA&Z2r@m`jWH_a`^1n{j={2{z>2
zRVTTRY+a=}T@nG$3O8D+&Y(H-ID{i}*60=n?vF#B22C(F!C`qcoiqC;x(n)X&RU%(
z<{3x^dGxOFUIGKWupO6>E6{W)wKjjcuc6YAc;qUzz3CYLlh;I|Vfx6G_vWf#7KzQe
z!DAo@)~ou%J~5n&?m+ba>97L-W~HPN3`rRQs*dMjGqXv&T%-|nO*-oq<QYS>ED8{c
z8Ey{TF0?cvpqDKJ5ogbfAw$Z$j|;Jpl!T?=UT6n0f2{jlrcusuoiW-j>xabyi@?wQ
zM{aW4=K5gfBp8D*1%3BQXRnO1kR%BB<*@d}KPJDc;Th;_nFFED%JAxs8+<bNz6))W
zo?6o+Y{EE|{ZfeBaZ~K%0ous%{v$tf8^(b5FzMB&=e9=V@BC30pX^6OtAqzuyeu=+
zCSv6)G&t$|<UB^N4ROaSfS)TgD%N?t{Ka2^s%z$S$<P0{AFnu9Xs{2MIQwHE0mb>3
z`mUjOQ7eP82`aZa3ptG_*`6dzR%U&bEj;+}az&6_1LU33s~&*&z<o4*OO6gBQ3da}
zIJqDaqiEj!&mY%oJAn%84JLmfzOnYhj#?KD^5lt;=Ld5iozdGn>Bv0ti%$WD`lHJG
z;428eMJILB@^yX6<>Avo8yCL@(vOq~vr;)c`?j&;jbM&~muqAOuda@EbuI@_WBl6l
z{AVZ1<_CuoRK7jID{ucgaC#W-T?hJTD_<B7UaKEMoy!W~I@2-&x)lodH*zsHf4tI3
zw|9YtWqflB0xuPuL!6wr3QUp<xR0N{UHbM+hw=@J`xeN<p5XmRz@YxYyxzw3iE7g1
z6-6cn&K77S%H*1ZS-uL;Po8~&Mn0P1PXXbiRut!ADoln6LSAVVQj$GMem~`QzEefV
zGoh1u!aOZjo09E}*-~4Qg+s>4(39F;ULYUxB(V~#qUl=H^zD)^HjZ}Q4V9NWPhA>J
z?KHCs$)1q$o;{py0F}u?Qj+zT)4X>1u9(dEOm0!xDvLoUcKJPV9I19;td|1O(besc
zDx}IG*ZgkoM|*e9MZPb%T>BIwYiCG1={1iPy_+gxyM`&yuj+2uSfP1vkiFe9*`lzb
zXL`^WwO#{tZF1MocRL$b6ieIPU!Q#HVJQy{G@=xvJ-N0EwzZYc!V3w`HBzSwJ5=3?
z>vNyHxfMa0e2%l)bTy>(V&gjB^{Yqf+S3>nUp7~rhFkVJgN}6`qw6P$H~t)ZuHv)X
z=b^_puUlVWBEj6EVDMF^t0!%PeMcCT6=*?wtxu3~=nHG=P^N|OZSdvInn6&R!=_hz
z#WT4Q9X3v%`w!)2BUv!1y#z|pF=5;qQOEz|?SsFT=5-zC1U99^sfk}>j^7TU%OKU^
zk-t8A&Y_sjRfHXAECv=u3Gwhz<go(Cp5H}3)m!(@j^>knW*{3{#57wTT~6i)5IYCR
zHyZSHDZgYNCJAAHGuxc8=ndhaLQzldK)ecnFh9QQ!%|EF=kz#C5yrHP@RU$_sRoa$
z1-#sBR=@DtjUeZv=(5RXF+Z{+4EHOV>Ip;(Xbm$;*S?kVT_k)jA1>?7U)V}aBXp!D
zCB`;<JUHWWReT8;d4ZnT7mT|Z2Fh!%=Vgx!ff6u~pJCI!KgD_y_;Tuj&Sxwmbt->e
zu2%V%&+#gfPYNvwo0?O4pH}x|9)%oy(G3#%Jkg*s7s$8vf+6aUb|ZOqu5Eq_|B2Vu
zeaw5UV<R>UnpwJmEQC&b8$#;_>0%*)pwTc@tDkfQ_;B2XC5GD0EX~o_;W<=frP%$9
zs^NIAm~=n0Cmw5brSut`FE&)Du<Le4F&a`0b1>ftC-(CoVvv?v9+TZXVz^Ba)Erjb
zQ};H5acq4Uy`?#qQMA!m77aE`<wt3UH>e7ENaQI~u!YPNmfb?1t}RJ(Uj9ZZhh1fI
zXZ^*7_Ll>D03sLDAw=E#;6QPln}kDFr^CflFr-o_YFOQFC*xSQ=Awv{AC+<qsuv?f
z4DK<GZ7Ca?VcWl{P%|6LjUN?DJ;9=QMLVk(k`yM|;j5xl3S>ZR&4uf33`_Y3Nk4<C
zF*YycX3b2pm(R1FZbeV@@Q?7&B{-9ZOkF6>uCBzrrXsti(p@G^xnBIlK<Ebk;U_CL
zhU?KRwANh(8wif-GIofrxItV&IcoLfig@Vj@Q&~m_n?)A-~J($1w&Et&%kJdnuytI
zZTqJ^A{&uDD;tHx?({l<=(wlZkV~jElro!(-EP+JSYLmk1{Slo(QSN~c?H8ZjTa0R
zEcAc#p-%$EFIDL*Emg#GN!qiuup<B3I!BVhs_rO;(JoUr&c`wb`;A)nquKX4ZR@2J
ztP>6A@)q({0)kJrD@%O7oTu~eGF82?=wDTS{#yAbA)(@Z>}RZCb`b8NNHwg&ip|aN
zI-K~5Lv=W(EbN#V-A{>zgKD2?;u=UCf;Q~2)w6Z55vHCSVLs73#ypPdN;1&=7_(`{
zynT)=J@1iNN8Z&MIYZf-F23bGAfafFd%gB*Q6O1SM36#?nIuGTTW$rLV`J)U$i0{?
zMCJ-!mumZkZ{wMg%xeQ)zsJi-?c%PBk10}*^EpyJhYNXaylXGJUcbyq7702Ar98DE
zs`8J-iBGp5H7Hr8&V`*v$8;7aj;QdN{#d<W(b3w;p^WpvicU6JF{Ah>6pJ0rh!)pe
zRsJ)z^dxtv-sb^hUv^&_s#mq>k>)D1-={L|rf)`%LZ%9bVny4d`3esTU0-3>AhpUx
z7KLfzLy6H1lu_Oom;fzd$WCY|pW>kh6w&BVh0M@{c-yQUI#-b|7%r2E-Bpx5RyQbf
zbXC<&zw+@A97}FzcBpswR#+X|*d5Ol*8kY%=wGWctw@(Hvon%To0U0F+RwJRY4|dA
zlarI@ak0)Ta<Z2qa#QT@aeIxj5cYXufON#MDh8r#+sWmTPgWP&A6FKD=;lOM{g4VT
zZT9uq$^Pul*$&N7;mpprvPYC~uz-%yFkuy$Ew2B5@%(c^**)<F1_VQ*VU!d^e<-Dw
z3<qUc4J*~=l&Pq7XAfUwTotf>E{e@~J6blSPPp)fNIm8!?dQ&7>@c&hqnQ+=QTYi*
z<q?ct_Xws}A*%DZK?;&8?-i8eWyyKLbltVhL;lU${pYs_M|dp3>!YX2ID?1ZaxODJ
zh$kp$Mu(|?&GH!o&_}}@pR_A`Zo;8XzJcQ@0#ZIVC#85Pq7qK7gVsQTRc-$<+iBFr
zj<BaPUvosF$&}+&h+KFz+JZUcPZ-zUQX!Ay*4|=>`$HSbl0g7gbU%r+I`}Vtp1vA%
z&bDJ;ECT&=6$u7d79&okY$BN=9>i>eOsbz!tOKD&kcr2jT`AD7c!LhIA%wx31fntv
zz<n#vubin~zesQT$;j@!zZA#Ab)-G$91481>XwS{W3DBotScRK0pkQ9dF*W%?YC_g
z78BCCxiSh1EbweS#0L*<?HuG(HP_Em{Qu_NMPdC&;Oc4*zDGdxb>RNh#Yq_B8=|5K
zyFnRViav>LqL+}p4Y%b~mBQ}pNjEU-Tf7d2!3`k2#WrY8iBK6*{Oj3LCIDX8?bjxE
z|03{(p5z<sEY#UDlfllMToW>FnQ|W8Mz5Id^Vyqg$E~Nzu)L{$@GMB&vow1_g82W^
zePjp>rE#Qcfazbq=f6GK{$+K*P+5tnVPgu~WMa^o`hQ>TZ$A6X1)ibCqGayhJi`C-
z{mYcVBam`7A^*R8FPQfTCOE7^nDY_;=_miYZ{4SWg6hS(oNn`f_`RU~gcdAhu7a`;
zf4z>|&@*H{8mYyM_kDR)<0Q9E`JW#9YZmZD#{A0v;%Dr8g>KZFWC0SybkLl;`9P})
z9M#RcSSz=<#V&wH^Z?)vH$<O<8>4FitY@i0*ZoLH{s6I;N?vJ#d&nF(Ju_;kYzE1v
zijIST!W_7n+U8T4bUH-R%T)j|@t3@Os5QI-lzt`l<@K+__ds2r$$}3k2_`~$e<C5O
zn$)I~Pva;2zPfdwz6so9+AfSR<1BBY{pYVEU=JSSu!R%GU%WGy&y7GIiv(s(aouqo
zEdaUse9|E7<<^fUz2LXWDVhXvw-&ZzKJZ{W2AZj=$-{<-P)J2^4ET0Y32}Vx)f7O!
z<N)k5V@8G#An&d+P&WMrxS}`12gUCz6WR5AWa1w`oHzp%F1;tlJ!L?eI76(C{32Tp
za(b0XFhOqshAuwf&qG@fG-Pt}f!@<_`wGZ>ih+(R-LBUC{ID1_B4XPWCRE)ofHEve
zd!o_%mIVMLGt{jFjsr!ZDIRde4b%rss7jzytU6Sevn!0Kq&(;XS=)l<^5{|2YIPx{
zO)LBul4Ij$_nVbvR#}t($6FR)2H|prLTm-NApWPP^|=cxYhkhlkDr{T`W-IMHzo9+
z5Wwf{Ro)1Vvc2N4kaF<gK!c|zfOK}}Sj?o3@Pmd6d`cRl97q7hfW}Q9GSpF59!+K8
z2NmX$A`LAZySIy=PC5ZJ%*&^TU;^F<yn8*RG)=(B<Q-lB3L#S<%&pM51hB<Qt6Ai$
z`36Wnn)lN0oG=cuu91mmk5$#j;srtY;{Vs)cZW53wte5As7-6pM~0%JwHAbkY*a=-
z6letm3}Np*RRobOL!`>ksvtWckU+weFl7}L6bLKKAWO?0f&r1CyyxBHQJ>dpzo&nG
z-|_jE2!?xH*L9BHIQgEQF9lp7VI_0hJtltKW9{wyvZl{FcMct`<0E&)d_cnQ0<2YZ
z3`|1|;BmL<*@lH0F&8_k!Hl5=$BJv|ftt;=d}-maN;CnbAf6pBfafI0=ce)TIsKo@
z3Z<03E!lu=z5!31vOW$3HIz*_=4Msmc8;LNO_Ci7+#tIb>F{U(C=~n!^%e`#Jg>Zg
z9%)I{Vh(8|LeRVb=@PYEJ+u|7EzgHw2JYIP{!1~)i~E3>$`#bU7Pe5;g-y>q9Sry%
z37t3h=Y49b_(S)z7ZoLB)?`t_L0MP?6u~kZ7I+pjP^oXGgf}37J!pdh6-wk(zF`&*
zU)vp1JoQYAuTs_b!Pfd~AI=UbbvW<}4?6zc<|$8tgZtd&h6dT^J8D(jC6L%T^k{*i
zj4Dz-e*X>C{2hhPRYN&f0(}nVy{d(u<UqhR1B1CeRu0|dX2-mp0p9)gabHN|X8>A#
z01AmbD{LW<fR9l`omS43EyZLAnxs{Zn7AeW`xUhM!%FoWbXRNn>QeQI4J1AH0|$lE
zYSLbF77RrZ%Hs^}=B7BwA0s6F-d;ePs;B!#iJ1-EaLCx8HglkNUp3Yx8o`=!h5kfT
zmX-52#0v1S_5^)<ANUnmfdC-af7YEC-h%D9uT&3aIaa_yUAHO$%~~#69xfkwe3ydR
z1$LC`B<&^$rRRw&9b_vX4NXCckjv`A1pZCjf--k6Nh9v8lX-LzSAJXC0Au0ygLe0s
z%g_#^@@e~w#@So>midDOy%@wRJwTYZn#wLbqv2|#j%Y_$0g2cQwYAjX?&2}D*=`Iz
zFRme5WyCalDMF&QVq_joV`FEI8aPlZq-<y5fjY71N6q(yH&dGn>O$pp`*YGJ?ByhL
zQ6q_+C!eX}+fnC)3EK&s)I)@y)f*x7JEMC|ozY8LjD~)4xD-b&rFevQW(sS*d^#d;
zZKYw1PHQSrILaQ#T!&@lE(Z#O>D_~GRdu!RhVVqr!;-=__bqmtp9eA7^+lESJGO=<
z;Wy=9{~RG`dPi2U+<crrV=CRD40h_vf%GL>Wu0BXpKR2nDB2l+18CVX8r!W4XX7$9
z$;}_+K!7HIS)ILH?Tq>vn}Q2)G4Q-&in??VZR^4TU%LQ}24j3e<EijjxALP-H=GxB
zmTJ+4OVwPc=;I>7%=^B$sfuYOtXYR|q@G{my$?GXVV8#3oXvJ!^Xs`o><;!F0vf+s
zB)wULlc5)^^v}bz3I~nuWe9t+?BeLM8ibU_9<BH{uDtd~qu;+6I%wzi4NUe@Gthj6
z?MOuum4ATId-+=dU|%nIFa`q*??&!E+Kkc@w#*2Xl81i9<?U|<g7Qv$n6Lqyq3c_|
z&7#xbb;$kMfxS|s35bE_@Bt1JE&&s;ewubRUyX;Egy$xHw7rR8`a>8lK@~aMU*j`>
z{+);88|hFx%mv`Js%@|81NY$>;Ib!+fV{$X1~r=TMB@%@vYr0g`_<Tvl{u`?frcp{
ze5vRtN4~NnX7<pnbmT=K+XkzBvZ|qp9VBH}2wDNXRmRKrlaZmD0SiB_($5E2$KO~0
zx0qtolvms~E9&I;e7qR@b70UQK0<A_Z)@*$%Mj1OxXXq)8M8G3$>3n!$pA$RHENfn
zKwV|?!{W}JF6R>`S?D?|9!V+4R(aR%9+}rrqvvvKm1t);8XrM+$0C{HsaavtI1f`H
z8>&jZHvlSq;1^}RncvGI8@od6Ewt|o<u%)BX&T9CI`-}1{6NRbx@hGp=#gHw?9mug
z!jB2dQ+?Z^MX6gHjyg;OOq**!_PrK-1A@zuK2_0pw-y(MLlX8ZqCKq8?_LMpCbXvV
z2Mbma=d4P^!v0m<vUrv+6EXvy)s9j<ZNs2bk=}%#I{Zxr_Y&Vt+iO;#jBCU7h5Low
z-U-((VkawA-S>?Meab~5A?<X&Y9;ygl+H*b1#<;exlWG)%OyhS7$J2Q09h4XC&t|*
z5dLIW0?r}h*nY4NV7vo0cUGjGyRx^8X(r3}sz&nJQfMVG9`DozP42D$IGIpE4sY~Y
zGipJE38qM8UWgJd5lSZuhrC-^3{Qy?{@D~*+1ne=hGGd0eK14r$JTp!lO&7ZYE3W!
zndOP>7b!=`ND8|EW<+RzdcYCDr?OTv)HniA@q8wdDSTD-Jx~;W4EB<}2DDyP?&BP<
z13=}asyz>`?h&m9PN$~0=NJLp%4g>^*miQblAv0eIS({SPe@YKw8@8R$T+72xD}b;
zSd`ADE2dkj+R4&g)c9Q}w1cZO5wEVmrbJNJU93f!@he$X)bfO8D!Vc)vdxYrhcwbt
z2qW!JOV18z<DX6@uKyvZcwIczEs;Xc@P}Auy!{{=I@1&}M4?+G%*`1QSY{pruRDKY
zfTK%J8s=Ae{$bgeGHyiIoXZHmO~1P|I36=7TuO4fSS~=Ta?lx3<X|=*d->gNDF<!c
zblv@W-LbjmMO5F=U=p<H?Gi9oWH)0LglkFwZHlf9a5TrPPvFQZ!!JM_^3$yA#Jh!_
zaRtdD##3tyGWk9;uy|JMb=ykNhRreY=R}d7Ui9n|uAw+PgPP!m_Eb-SwYA0+lL-*g
z%S2)1ZPF`0LE(VOwLk8u=Cb_cqFbgaRk4cyQ8@+iZCu-Y=-ZD=>_g=KRJnMFvvF$L
zm<frpv^7Vg5lNziB<r+on|j`@81Eqw7RKMU@;xM~MDHO{IWTXR1W>J#%*#Lb)Q{QH
zwBcwk=(&E&N@^H^sR5qg2pVq7@6zZe-1?xT@~p||52;Nje~nxJ<^W6Euy=KtSF~bI
zc{Hecjp8-S*rc|5rDAgnv%<Wr_K{A3DDi-l`~vgj=?r>Bx%p^nK~sop#wA?j#nH}u
z3C>Yleh>p{?SA$<)+010Y>On3y=b9#ZAZ}z!^Vz3wz1-X_4^#J+$GRfylcX4DfZ5b
z#!3mG9_aCxhb^<k7KSWFxf>*s>emj4a&EN4MY;uCn($70*$|RkaD_8EhS|~cdYkj;
ziTrk1g?Zug;y49A!{RFDw~?1Qee`@EK3e2I6H(=j<UevPI*}HmB9~Rdsp9HIcycVN
z!MWI>ui;X;px0Y*mQz}sKBGx~Jl^|BV*|ax%=dx>OX!It@rEpGqE6DZd-+L7ah$#g
zz2dCnC_~+zU%s*0)F;=uKICymw~AK9io+=PH{k=q#@%tgc7_|0{Dv8+VU?SaG>-ls
zEBPEpqt8m_Jur0Yr0%Kn<S!S=X;hq;R$x%=RUewKwBM6Z&<fvNW?iX|r@x1}0d-D!
z@$t(Z-EL;>o)ZOaQ}RXl<(_eAqh9{JPgxNYGhVG^i2_$&iEu$j1X9Gpm5*cG4yAOq
z-4`S3Ga8|3aWC7b=9H_090#dl#dx%{w;_H}Sa<vWg@!AJMMz9AZpBZHt;X*WM&Fy8
zvZYNZ@cymaxI8g)HFXS<3&R}Z^s}(viRcv%bJt7cU5iE)j?w3%&9Gu`A!JCd=!hVh
zDX3&J8T`NOETcJhk*@izbWDT%T}yQAsWrg)f{7QDM;V1YZ|1=-yrsg&FkvZPJeLdi
zI?J-Lxuls66i`VHIo7ZoQKmRo$u@2HzR#n}LP*c9MwwD#b|$>O8L59Ey89Pk9L!Tc
z%wVag*GTn!)4MrxU~qq6%QWCYIW5)NZ|PK=9cE4_-IyjNhX6OOFjFzfP=6QF;jhc)
zOmsTR5>Cw!ki#G`G7Ex>&n5AQcIwR9Md<WBc}J&`M)mDc&i(W5*VqKNq_Mt6#q$e%
z^pYp;A0Ky+6&ne8v(CBUZ}zCErUifp5kkfaJ$RW&o1y@pCr4Pf620dhgMv!Owb0;I
zP;G1s1Itn8am(=~<QYvB7nA!Tq>C@Q8ksMTM{jlo;3~?|7LtmNhLV?y3lj|M$zBLw
zLZpq0f!Dpa942usT4-prP=zcNy!75K(_TYf5$@=yU}KF=BS^@yW_T{EX=+ZV7RJ^Y
z;BV_0OVJCO<dL0>B)xNEmMXiCo=enBJl2M4`C=zKHUx?1WRc!1OxhKLPMQfo5pTWT
z^0HoKu06(L@%IhN(DS+#d)osl>O<3Apu^t4xSPtq?XYk4u9pq#r)*)iSrN)KBjt8|
zFLkDp&oWU<bFc#Jo$I*TWQ!rdc2)2h8Y)sM9Xxz=IAiOinEa)a7vQ5)w>LV~YoFhD
zgE@aAX0&q${ZY{<w|uya<K?6%b;f(AMBmI#9#fAtO5apDnnqglozvtP7`QJ)lHOWM
z8t%HJTjp@XHS)OTvaxHM|H+rXV<!oH2fMm`TU<&CWCxnG*ro}_FYm<LJ;%<(>2HY?
zYLc$>@WmrjCzC4eohB`lj+&~OapZTnYfj||G3}cZcQNwGM{Tmi$rFK$_6XR6gfHzu
z=>Is^@_k$rf7y(q<$<+_&91j_wNq{9CuE=o_LR&^vBQ|qN)FYeSqwnke3pkYpM{-4
zt0{2IxpKpkQj|LM>ziRbVhmZZr2?QG7}8shM+L5cDK$|yGa=cuz(u=VjhOrf@gaz?
zD50I@y2S%zfynp1odpjTqeTH~-I(Neqp222Db`OpU7r@%1oH&!FcOnHEBI2DrHI?F
z>IQ1BLJWrei1A$J$_wq6hrt^MYzVs$m%7sS?@>i4%P||B+!UEYWWSEtD12yy@s!EI
zYhnRK-c6y3We6Kx6%r<vidcOaz9W>Qe)re|lM1q@eSJ*N??ag?k(-dtE(|t%8=C}2
zC)2nhp6p$uBL@|a3&5l^u6(bAHj=gPYt{GL_1SGkrpoM#Co}~0xi}DRpHAtFD|~&Z
z<<>*C8r9i2I-q@1vZ4oE7WhP*?p*qm<&Qw#)u`YV`q$|A?>MP+Z(HV31l0`o@))~p
z?|iNMj2fdtsNkwS-5A1&bHM%Ne3eDvctXpmy|7{e51fz<=2QFdrL#jJoKX$uDe>gF
z<lrz0jWp>;In_-;hM?u5v6?7hds>!OTMy-b{U6yE;<bE@N;Qop7;Jseb(lt>677^^
zRK!jiiV*hCDvwi+>g3@F_B-9t-UqWMZY~1Wzfhhm2vv1Ygj<F+{wH<f#pA(b+@goB
z04wW`CZPc(TKmGBjTd_|F~bqNBwti?ShYKpW)94mYJHCgKc-+VVzP@$#>;i`^}E+m
zQR@I{K3Vx0g?*fXE9osJZ$d7mP%6k5sO*n05yCwb-}{sUqu9eTJ}oY)NWLLtvOA`+
z&=?43+I;E<D}^e}H`QdT{7qab6B<grJ}o!6e1>j1NAZi(x|{&wcgwiVbf@<w0Vnf3
z?|EnBx_}J}y`tUpeNMZ3-V5H<ImM$k{D^RlmKJIU+eCg$uZXu8jeg%g`9O{Hu?^%C
zNMjdo<y@ei%M>>fK5)cjjQgDg(e9Wp6uI9HYGErc>eFVp*V-xY^G|G0rg>vm2CoFP
zmhJ)KZ;Pw6AMK>K8;9QH=lbqw(!O^OVfB1lLbV+k!!4CfgZMeqmeFI!q&d1J@+(tT
zb`jZ+H5-p?qVLpqVr&R#+nm?1^QHQjnky>ijNAN!WKuZfy>_^0N7I*=!NBin6b$=m
zIM=<l0)D)a;DJ{}YLT)sbltr8CN*;}r$L?DIN$8_17K1Bzk>GYrt>r>0Q{_cxJR+R
z{`XMAJe6Nd@Gh#UovCQI9%U%<R*|Fm`J&v=ibua+pi<|Rf~0?Ln_Aug?;HNd&WyWt
z52Mx}4!%3*2_yqZ8j!<%xoy6ru(N5VME25b)Bcc4#Krwqy(_Q$U-`yuPH*gzh8^O(
zaECY2EKSAr!S^4}q7;*w2J_dLk6-`w)s_b<Ra)|dUPQfyHHO9Vq&jDRaOi{9UAfpF
z<doA;ofdwxL%_ytf||WIG~LgqwTZkQVet#>mf>QpD=Q2_$O*GuVymcID)Br40;xpk
z59k|G3(TC}kMka8#M&9V77-TGjF49BE<4;M`-efaU+xH~>{f(9WZ|B8BXY`}+GCLV
zTv?w69}ba9{a1I6!%@^#dUhOd%_KX}xwhP<@^Exj+_xOXYt{D<yb6;u+Aio+4$H9z
zu3z!=m~r0YyrysLQH>6VJAf>ODXyhnD1-(y8j?gR>O9ul{GU>7?fRvWOt-v(S?(Q$
zfa-|mt~VeSX-iFZ8<B)dNxVsasOGiZmoW_cc%En<frk0>5dBX_&gXw?l4c7{^>gZd
zOU7touBCTR6iD2?_bOY9o&X*PPQ9wTUGECr(lgwu0jG4^8B0E5adAGTADT|L|D<c7
zn8_+MnEp2{%NFgg)Wd0;pU*x7hj7w_0)g1DMBrYdy@L5|xRuQP&%s-87h#AC_e_+6
zi;Y@RQ(V!xYKJn(+RGQriZCI}SD>=zCm=AIXrCSUxA<gv&#1rC&M1T#>uNnZClrga
zCa*1{O3cU`zw(-T=p;A=GfT8=6VNC%vnjD-0HB>E2SE-*?IwQ`Jy}7nnp&RRy#u)$
zsx`Cc8q9f68O84`>AhEc9NZoX!8fL4&QHqP#F31$^fspcaRD-r=#|u4vC7KK>TIeA
zG}+E2h>HPiq`}rKCD3l&*3-I7$S9;m@>3)8)_OGGIbO4M!mb;f-y*)=yHFv?jSx&l
zUlqW^u}ACuXJbH*)1sGn>)KXB4@$At7w7LuyCc&sO}cRtvttV93AAP#Of9#_pB<ZP
zsMB<NAJY@~lD#k*B`yc#ui~Dasyfkis;}ZQl008upLG`6xCXuctERSu$AUyZ31di+
zC5h)M#E_n+3V1zSs3(mooVM&w#stdp%62>Ia43b+F3qyk3!!XF@HvCzpS{}sVM!>l
zI94)$Sy!YPd?hDG_n^fQj68(UQA=sooW{Znp!G--Z!@lK-L2P!<@}`=3PngiqjR)G
z&y@~uVhQ_?1|4VZw{RKzL5M}cj|K^{hN8T=C5@rcGXrK#`O@7clK!-Xw_s%x8>WWx
zl0<S?j^|u~S$UN6T@Q@Ga^jbwft^eb*diM-RkeWoY1jwpK+_|=w#$1%!up#|dbJ61
z&i8yTH*LYck%kI;<*(WUz7y`fzo4TTWv5}e>QHuCf_|eUr$C&RhzWd-%<~J6#?^Zp
zw9LF>pU6MA<qv@Bzpn{3doYIKmNR)*-ckQdXw*2l<FpE0%tTsYKI5Ia*J6fQWnk9R
zfm)le8ZH&r@j0Wt(0{plJ+1RI(qvg^8TfR{%7KZ8OKK>Ou)oauW2^)d+F`q?ev`%t
z5P50$21&_dvP(exEQI8>p#SDXB5<ObAL^^nzyL?SLIJUv4CCPx>xIL_mH&oRoOkpC
znX!|7($*$PFuW_HU@G+>3v<zli5l^;oFascwvVZHMTa*j1Um64E{_1@B+hyLiNVu6
zQNG>$vNY7WtU4}ZK0};34=#`PlX+nx)&xlr2Ch9#%iVQwrK{vck4MrayE5nx(?c>w
zgO0HrE1aEPKvS^<ZSK%=nhU}@T)>>6;EaD;rRvlTo=E4&6a5c8a=aXlHD~3D6Tk2F
z*f)>4;^ul7n#?Mik4srKZhl>lXl{{H1Jc%Y5vAr{1jcyCES?z_c8`v|t7=Ryx99Q9
zu8Zckph2zj-0$?O(y$4$sS^$3?oP@tjh3WbsbhTV<-8dgzxj^7oqtIW2_sGUf~lul
zq;Kre_%4H`{j}w~vu-VmlL-X_*tB@9WmkT`s4KW=_sP@vT-I4@qI}ys7Svf{#7*U|
zMa+Kuh;vH)0{Fq5{f?Wy8dZW+H&(n(_EwQ+oz|R@N!ykf{0lC_X4rG3f?de0sKw9e
zXBNeAw6oPEh=b?Xp@?8}fVq4&E^{53W{|Zk<uKn;k^nZ8WXEFImOH9Pxse~Ym4Hp=
zsSDW>NM;?^cQB<a;lN|^doPumEft4$rro;1?u~KJ;N!p`q`*X6p3+@?gu(E<2Y6Wz
zR~5DR(0FD*rbwPAcRgh&lHD4!Cgdg6wc!lruw#}>3-wUlvi4hzAEb#@Jr%N+3&vw7
zVvJo1LS11=>xm4ddvf2E2Q!ETso<NKL0$Kok+N@zMjHaney0Omc?`eKxbJoWWnSvr
zg%e~2A_bSzGr+^zU$U<A$D8t2W?4t-h|O1wxRiB|oley`&dnjjeW23-KldO;#j=3Q
z>vW%Qh7{wt=F1~FqCOqV(RXUqjp7{*%c{}#Rp25NbX1=f9WlM2hkgbkgO@|2x%Uvj
ztSLU&RW=M~d)|a|g4ue>+_kN=e3C+)hFPW&EmO}A4UxqxS1RiAo3B9KaG*4%4tU^k
zohy2cKA}vmQ%$0>W*ot+@Y%HT*-9(oa>tn7u7zpnNnmO5b(f5|&UO{0t1P_&$@u={
zEA!#z5f;T}<9^SK!<7Rso>7;~Wp(KJ97JLddVM7K8wo<Byi`p4I?qCh&W<t5oFq^X
zgP9PD2I+!^8k+<$SBi<hrS*dn!5BZTzP{=EJq}LfE*1aA^LfJ7nEk?g?$!!Ky9y<y
zO?op+0GBZP;rNVkBXC!AF;<S1?4gc2tC$^^XyXQy&^zt+{tbQOVeH2+4;|MrtwODt
zDeZ95>0zE``RG^q#0NWKf+u<iZ5r=xQ@ftw>(gsAeaDo?@STlZw~PRMn1Q$XD`GWi
zZqM($6`{Bi@c2Vm5m`Hq{j5faY20ijwA|Nz>5D!tHYU&hbc~TB@<<K3Ea3H`;eyj|
z-LJ!ogfq3cH!<Gu)xXaC3CVOf+hyWhY&+Tg9D_8R=yojfJ@ea%N%!|U-4qw8C=YtN
z^GSm8r5q2MgP(9j=KkDfopSxRv@$o>3EX|0a0R7+7-KgmiGR2#Kj||kRMm)k!I-KO
zzninbJ7Jb4o;#I)aiU29KTuwhp)&kTw?KT|?7dH`^j-Gq@_i&FTc?vZBTdGZJr@!Y
zi;OKFyGKEzD9fx$NaX9ph1N8Vc8@elE?L;;a3=@Yl6j+ieO<8x0KE5tbh7hk?)=`+
z6se2?l1kV8l)IE|4?Ok-0XEo(3U0>pNld`hsR<Gs`E{|iG|Vf<)BRGN?n}R=lKV}S
zf?7caRK<0;lj~)?OtMhm!_~Nq^?wJ^)@T#~94S0G`1Ke%et@0AL!#)!WAvlfbvs{y
ziHz&z$-X1)i)GBMp$9<$X&ZlT`DpdBrloDZkK8f8w{w|2_k{f&Pmf*SA=uUTU0;hz
z(EQ8#dfnd1V2Uo~#&fXRA{a#G#GD&|d`vv2xHgqiPR6*$*Lnd0mpSs%B}9p_QFkjk
z;_l`Rs*`RW6v#D{POF4sYd+-<3AIlF0H$VsuaA3ty&*vP#(|<o*ujuWFycaVZOK`9
zavc{DZ2uF{X^5S;<`u1tKtOQ(Gnv*y=QHC?1hkS(Fx{EI2fSP9g?aJu?J|)~yVy+C
z;y1}vN9;NNz=bli=DSw`46wUrAkPu%a{*(D2cS|Tx$G6)=1^3F!j%-KWe~H(cuG68
z?l%TT$LT!p;E!2us|j!^=!@ynU%zmlt2I#CqNTe|(&4?@>+j!Lmm7xCW%@lBXTqY-
z`c@9>ur=(Gv^%xU8BX-^tzk{#pRKN@6Yxviw<VbV4Y$#Bqwf59DAH`BM_1m=ILCW>
z)&#QkkBE&6e%=UIFoFnl6IZ!s&*$Iykd1x3yN^)|SZ7)&tq9+8xc=)cgD?C)*bQ5d
zH{*=P-@ktC|M6Qse8Pe*+_A!9(?yfNyQU96{jszE|8MQHNAiEkTTA(gtp?R_fXK;L
zYR~3RMdTNj#>bcaVPz;~)IjIK9=u<3P5N(li+qF3mJ9~G0VmFq^hZ*EPx|`&^Of|u
z#|NvcHRGimqdfB9ssNT^0zqQzDO>)IuW#`$9k}H;YV2{WUFcY!(Vs6H{R`OxL$@n|
zI^q5Y((}zf*?I`Osfq=3Qu0xFEy{xomj!8cGuWJk9KVQ6Z~ST|`K50J<sQgyRsOaY
zDqNgeIpa1mv}fYD+~1%fU-*y|UWkIzPXvi4{}IP<yb7ghzA8%_`|#>MTdiOIveE#~
zD^c9?9@&5L(3F<Y^UA+vufjjO#wT!%zVpw<T0a3sUwXTr{abk+1h84(+41B5=o&?N
z;2OiVaA8qjzV$CZr~iWL^V#M7`Q?8TWdHjAf@ZOhovtR!`TVnM3`fuX7k64xKb1az
z*|~qcX%BNa_xqQgw|zp3zOep2qzxagHpmPmXYvyLzTf*_{Pf2kNm<3fGOiknO8!UJ
zxL^g>So!u3_wg^kp3l#ef8S-lJ}l!Uh06>7h${R4_EuZpd~r228>6$@hrpj-RL-2r
IRkjHJ9|;1VB>(^b

literal 0
HcmV?d00001

diff --git a/_static/IT_sample5.png b/_static/IT_sample5.png
new file mode 100644
index 0000000000000000000000000000000000000000..b27c3eb11eb560f49d90f17baaeae9baf320949e
GIT binary patch
literal 632293
zcmeFYWmJ^W+Aut{N-8yUsUSHp<dBM#lz?=%ICKsT(hVZrqM*{<-7s`_58d7Gc+Ppw
zbJq9$c^<!C-&*gz?tAS$d+%%K)z_Yo_X<+DPspDD003MW>9@)Nz@IE8(IZS`LSLpt
zhb+jm5Ep+hBQ8$+-rmN{!qOA~kPeAUz<RHuN$PXp&XMLHPZ0WbD1|l+p#0`zmMj(p
z-D84x!M>SBQY7lOY{9`k^qaqfKSNPW3WALcm5L{EOM{tqj9LPFAAIPlZw4P8PBN0f
z8}1uv?)x<^`<@Qe03y>ujZ_pCwCGp8F>f#3!7D3-z)KV#08b{WG@xtUG)!}OxdjmS
z>c+*@8%u2UJZBtM=<WTWFA>1ejX(heWT&<dB22$|P>z!dl|07+nDA}$>7+7HZ#aLo
zcvdd<^&rBb_c?xq!w9-@^IN$O=oF^p0CcVd78w*DxY%8|j^2~xpxuDdh+qz)ELO>B
z1~51*>iFr`{vqI)6!qYs+%>ym@1p(79tZoHSN7>Tef|U<{?jWs`y&qyg7;M>X41I`
z(CfHV=EuQe1<bYP7b9h#*I3^@!A&uu^QQOutnve8SOwl^#wf>`u|xQrCgmx)P)E-j
z(MTahJjPF<7}Je0Dy%)B(SvcUhGjX4EJ^9=$<%BYxlj0yZ(V;Wm8hh$gg;JO**)#f
z+yYk4al}Y=?~n%HHso%c#TE&JZ|csT2ppFP{8B<Wu5idjR%wh1l&!8+9AStwj&cqk
z!|WofJlO3dtA+HUmpLxXHza$z5Sdu^ZkE*^_@mXn%PKfiwAdz@%qTbn;UZUnXa<ch
z;6?)Hc-XJEiC>c246^v-nwNrM6JRoDUgvW__~K;G>S4Ty9X{RlxW%H@Gr<10l9CdY
z(Z<X{0Ocf>&-0Y?q;o)KaQaQ3HcVM&|GUi!yWy~n&$CPv`AlpaU$SCU=jO-NC>%@J
zsIP&+s7D4UOTHhXL)039tXNPnIws7N7T)(Lzgwtn0hE|hufhBn1Aekg<kcwFenDcG
z*DbOjoCQCnN1qP~FuuJ>j;1!j(R|}fKzQuqk!8Sy8XEL0i%1$>CTo32>nD)empPOC
zC*Cy%<Z)V1q1gP8=TA~{{{^uOL-dzz+!|QCzUD1*i5OTdN{1f&7<%X%Ef|NA)s%$+
z_$~5>u9xH<6!9$_%TLj1pFE-(mdYo1*(%u@jsBc0hL$CU{yWYu3G!&Q?`Sn4-O+Sw
zgqT5DMnE7-IK@i<VZ^c_iJ_ViuK_a%3nZWD440SVFaw_8%C>wmW-BSwz~jYB`K%kx
z)*f$oVW?=pVPs=`Q*vO(Pf2AS>d=X?v|?vbZCZ_6omMS5C!oVuh|l-AYnA6H9_Hc!
zOb;7uAzg7<!e1U-aX-Afj6%owOx+T>BqdE7{OHP`tMzp&*&BRAAJDrGFUbjYLk!=F
z*uAnNwBsm=gQZO=Qoi^VEr&c8CS`I9++V3<Vztt4Z{6}TGYvAAzJa7(|18$Z(pe5}
zH4V@UP&833#YmD*mx}+w_vIpXFy<h3QJIN|FqHI#?-%iD>0jA8nL=3}AHFkdvDgz`
zh2{~l5T_8i6Y0Nt8W$d?@ecdz@z>t3k*{>s$(3I+r!bYp)%3E&vBW7dFRD^2W&NJ|
z?f=W<x8-|*pY{^#HOzYBZ^zZ+inKCorOrNoq9M*ASDyYr?ZoE9aYkQ}5tyAc!e7p<
z#kmJ62;<Gq`HjdV$*z1`puAU}93D9&ze(wa=}CCUbcc6WRvQVC)y-_wYk1utb?4WK
zgUf;ID;Xo1-Dwpun2aN@-|yJZmfW8lRRPi>(Q+s!f$oje{^a5I<88CTw2>SR&jwR}
zY?~pOc{<Zqe(}-eQ`F|>_T^5nlk=v|5ZTC&9k;EVt@!QqVXCq0ZP(4mXYbAiwi^t7
za%zO*ND5@9apo5l&}4Fq7Y!eb(C~nGf_c07K61k>rMXPVUsA-ASCZjTY>@>3+4z3*
zxf#fDn~~>|+wkp!?hTASJ%g4Ge4k#c`4#x8i+iwq;WUP<13qdQqggP`=8!5dDm3qr
zdxeWV)jCW+{%vHX_J!q^$*b}?i)mp(!BNwkk7j%DIyDoPc-2DH2Gt^-dB_xGv3B};
zoHbkXaakpNJMpZ)bNFSOSy8{Zw(6|dtjDa{ET1!pvryf;I_+~JchMWwbGGv$Dpo3V
zygJH#0c9$2%1MDr0WT+e=P&#`{J0Pb0Y-?rz@q6$;xZf;-p8&%YLlSKo~t8L1F1n&
zQ#;-D_4l&P?`6)OOy`tglwAz$JMh_V)J|6S2?41?sOyAH(y!BX2DJur^#b)giahnu
zSRK^JcL;Xksv@e|Z~QN?Dp}kyZc_1ARu5FxM%VMUDfYCNZ`uc&IXa3KY^D~6y^|hl
z0G49!o8x`zD5>zGRJzFU9PxCXPEo(b3&r{U$c^0cDeGfvTwifi@ybA<Cp>t>0lY!J
zAcd)k;;!YK<*-)oG`|!=kzD(MvPZMQ>rbg-#<~MyuoxDlT&1i+-GEjkY!TLBN})-l
zDx@cL6O5J5)kyeQtR?0uU<EKk4_Q<Zx|fFft=&3{HM5BGL5Eaz{(}LJ(JNVmzGh|x
zt#Qj%z&q?3?5Gz;gvNyUFTIun?VvRbGO6Exe>bE~sY<)}a?ruCj{js1<HcuRlZ&-4
zYa6fjUummgstBv}DF^vm8U`FXX^Ck$%><nmpWba5la-J~B>BVJ;j#3pO0_?QW%-~o
zYE@IpI!y6w_PXU(!^O{~OF1Zd(=+D3h>KHXEBs12#j>y|l(IJ6@9N+Ti(*sKq}Lj$
z88h0t{(<J|<tob2#=6P|)h=xE=s#W{&L`$!$%>O<nN=ezWY!pOn(;pK<@wID1@-Ey
z>fz||uHtpJg8C}7Y=3T~x%K|Z@qzU#(Sq(qW#P3HJgSwZZeq4*vE*L+Y18JX5_vn#
zqXgV~Ty^i#yI;4tB@sOob-V&rGZkc2n&npmB!g^N?>@Xces^6NXlcc4`jsTEakkQS
zHjXQ(bbIgn-pF3*@zQZTm82l=!KHGHmR;RQU1`a?x=)_tL20*X!gck|JRf#Gkg9tx
zu+4cEbvEVvn)qI6iEpjApKk7MeC~V;zpb&(m}p$_)V~Nj@56gb`LW2MFi-HOr>%;x
z4o3lv5ck1QM4L#EHhDhzC-SX`4yop+TQQ_f&+qLwJ`N368pDloyCS00(=yYUK7CG(
z=f&mA<?Yj2(XLg*Ot*wOL4}X<&Wdgnw=#YU6*#+%l8q#N1{<573`C|9W>7#f>>G=$
z3}@T0?MSUF1}zS1fsRMVFPdaYAsSTX9Ci94P90~xJ$dlF{>jur9w&%%`epx76``(z
zuKZ+LQPrGn&G3wH-Hlw7`QTNG2p`A%%3Q6p;HmRP+NR*e{y-f|U02h^tykyWB)&aO
zxQNc(qnn?bCC96!j5CU#y!jp)ln~Da=XXYDuegu5+C=mg`$S70E=Tka!;kYV#cVS7
zYvxZ>{Ji}5^x&Su56ZXpjz13-<3#iHk{TfQ36uJ1ZFX&UMK6lF7A#?TSI!TDK`vGd
z%M6MO=8ZIW?@yIZR(%!Wic}fKqWIzL_n#k>@5gVUJ5`xBQJq{}o$Wf0UA_^Ey$^_u
zMS+#k0Nx1!#wb2MfLuy5(g{8u(-?ei_Fm~pcYl)#_$by7({2~RYJU1bUM;&uiOMGA
zTRp(ihGX*Ag3C8W8dA#ZuF_fa)P0)>24C-P6Z{9=`G;}?p**z9+&j`cbxfby2~>5Y
z_8MxcA!8;l4`4>-9|6!&$N?D0918LiL816}UJ``~fcB4aQ~)5z0)YOnHVVk}&l8RO
z{=ob*Mf(y6z(W3dhWucesQ=aa2_O^gzw($w$U4Al6>%9EWU69pZ)$4o@X^L`PreL>
zEO=}yt?2*&ki7Wwp~xsdKScIFW${77QA1u1Vr*l@ZfIg-WXcY+vi;K!KnMmwX01#e
z4QXLkmevjsm@xf6T0oHbKgAsMwEt-02o<K+kbh4rZewps%gYX82hoc>p{1o2vNth<
zD8H5b7dY~lF#ShIM_UL7hl`5~y9=1z#@?KRQ$Rp~1H{F_#l?nf!RFv<?Pv&Nvvy$k
zXD9#F&s$RmV|xo*M++Nk+CTjo8re8G3e(g78R*~7Kj&!*v-t0ktR4QfEaU<?{?u@A
zvV%DO-8T|c=uau+y#>tFQuD2a6>`szV~B9_@bC)#1K|Hv_1{DO2dKt>gL3ikg8mcq
zKdSzppsEh0_Tn~H$Uz-N{`<iG1^l0t{{j@^__Ox^*o%K|^gl|GJ1z1=h~wX<Ch{aw
z@bv=nASo^0Dt<twNHqKNp#8D4nEsg}^XNBqcOQw6Xe9=adHebU3}rVBtDdrAyvnp0
z2<kLwK3(ASU!at;$r9ge@c$wPdzBTKP5_}3dqkkcMCVWK_Ay7@7&MhL>%%$k2+D!y
zN_otD+!)xOPa590^1gCuC@TY#Zz%~vAdpd(k`|_5pGSZ9I=(_}Fab>c=#>p2MtMr`
zH?Nc!rqAz3^t7md^A2G=?gkWNlO%s({VRA{R4%{=s=u1|QvzRhY2@5>i-*3N{N2)@
zMx$hI|Mk|tM$Ucrb3o4Xu)pjA2cU}eRt)`b-c0Og(I^I_tlTd&{|a6V-30Ix@^@!W
zi>e##g9E5D$a>ZDH#qZWR5Y>izroc%$GrI&Irk|HY<U@yzwCkx@Eh;%$Sqim^f@hB
zGrPveccp(xvE6|8PJegiV(7KBC}aTVmNzP2B>u7sB&nnQ<(d8eoYX)5Mp=qg6tQ?I
zpZvV{Z?_=((dqvDY<~T5uYA9Q3ND~(<`WCpu+=}uOf2Rvxqz5C!%&mAx5KDB9}kbF
zJP#n%^L)C$?x1mY42Q}7FZtT0CAwL<tJz5l77h15I@NLL=SfNT%^6Wz`^&Yz_i6S)
zL4&v9i|Q;5blO-Tx3iyb(X#Ohk<zJ>&GMHBnMpygIhq?aZQsY1k{p{u4&dck57qfg
zmgWXRlE?W{tNO&>&8C+94|&a(s`UOH{Qn*Q|L*_)H~jOBCkH4bbN3u3Fv<z;mAkS}
zR-}bDr`R-FT*`iFXRSYMHLIL9+c9~yOWmV1yG^}5wX5?pojav?h4g<V3w<Y^ef)*8
zB#tj^5{O5Y%n~$fSr(z`I4t2XpB*NM<-6P#loG~X+Z|3VJVORxuiGs}_@Lo|PWt)u
zH>3^t-UZC4@iD_+nn|2}53wb#_}_^_N5G$D>UrOtAxit$!K<ASU<Rp3ZqeIAakKc`
z<WJ0HW6H8gnr2BIl=j_pW)<UVef5RlRFCyIbserl>q$cKhTDVkQ3b?yUQwaJ)_7im
z!2RWFe@Pd$wm_~QKl)hZitX-xiDQ@BNGMm{<Y2|hm0RHdZi-{{L&|8)ilTR8;Z%YX
z_2)BCME%JC7+>Vlz^t}%-dVeBr?4)$W+Q#}?rdk0<D@oHB!q62vy0Lgk?Npu5+r(?
zl^r{JyfNAL;IimxZkBBE^5kkGL;VL7xd`4U(R<jqfafbkF{?bGwsZn<mpsM-rZTrX
z9hUSAFG(XW$A^)PGhOtznVzVzNj>r$H;S+Sr!29S@XzK3E)~{pGnbtV2nOXccMS@i
zglk&YtTQhPuC}y?5cWT2Of^stKKoEMp#@EHKWL}|zbvz`_{8xL`%>PXnv#?MNgHG*
z-*_Lnz{3w!qMf5JFk?hcw??~`a7#2_Tjgn*)T!6&Hl&w-<DS{s(25B>1vi2T`@gT{
zomzMzG4Z31SccvmoXD&fvvBOi|5V@%IX!YrL|G~}Y=a;7ab(DB_i;3Fd)}P)-xh#V
z<-WB4kzr;{cjsVWbwR`VjBqcYc=+}JDJc=i=6s7@Hw7$}WM+visUF8YW*MI6TR0t4
zrQAXna}J0B!GnAun}dc6Uhn|F@w)piT1COs+d6X(-5TT~-jeSj7ZHC*w)i|FsR-_W
z_%#wK0LY&Y6-|nbb~)-@m9|)Ja{W(5@lX08Lrfoz#>;P-_B}81o~<@5u~~(_$~n{*
zoZt9EaUg37VYhNwZACD^lKY{k@%neo(r4^BXvRNq|8gZP<?do}@$_J^X*A9POn7?H
z^e}I5x`B3PHY`DUrl0No+e$)ehK*sj_v||{#>hv}$<T8DL8Lh#E#5^R9fefQPmV^$
z|A)J->xXRb5I;(&C}bE3*=9OvM#KNGzbk|3<9ag4!j0TjD{w#eXyUYaM#Hb7#sVT~
zMnU9I#u#6!ta(WC3V%x=O|Xh?;Aqm$Hyi*oA;rrgquD&e%UUGu@K=&o^J&h*>1c(w
zIA7g2Mj#-P4BQzv<8jhI+5b*)dP^Ghw>p)K=zZCdfv90CI@q8Y0KPjN?eg|9BKi-0
z)-54+tjVrZtAV75Fv94_tT0l8v^DR;7OX(N+kT!&og9~g1*gI>MNRbQ_$Erj0#7eP
zd}e|^eSXBPkU2^Lpa&s|S1?H+*?GpI{|z>qN#p%B6l~P~$R*9&^AHXmw2ap@BLmMn
zR*+U}Uc#0GNZ~;=ULPB;w^>GW6=^O$P5hyBf(wQyAtN<SQ{uLZgQkatIV5Ex2EFgg
zYHs%H@<z&5LRlb+F86Dhul?rEIGS!tMODuP*1xKd7cLXW!u;H~ek8z2^*s!)&vwk?
zwLZNnYevOR<XiL@WrKg1zkFyq6ui1dq}a?gXinN%mvoSTVGDW0)ifNQz|A8My>VG`
zjYssJWpFOp^$DS?Agc9fV%Qm*5Hvth<f{Eg5ZCfS7mYVJ8Wx_T#d0I-fp334L&Zqa
zhKD&Xz>=iStdea;5!j7pwO2MM;&2w}7T)seEc*KMMo)f6zuCI5oOU`d$-+idR|JK*
zWF)19$o-}9WNtM641s5Je%Zh_2g~X7PZR@71tblnyI#z@fJf!-L0WW+r~Ri`NY3ro
z2ll*f#pamH!4IBRM{I3uGK@NLf>mM^djTM>A!?HO1{XXYmpNF~w84o}zZ*rtc$>YA
zoFWKtdsgJVXOfG}QD58@SHv|Zgi57Ul!zsCd)Q{(NQAIN${aqV-tG$fEky%PFs9yZ
z>!h&RMU>yWPi&;QELd3AA4#J1Mexor?L12(^p!*1<QNLb6r_U-Y^0j~h*RsWS7X!`
z2T&2evqe4mF|MXvz=OY!P;8+PT73^p44ZHZlYvonr8(i;+@1{Lx&cmk{xHE+42bw?
z7`)9q*&@eL?NpfFw@|BS>_RFZ-RavG4>hbvuV=p9=%Mf|dd93SsmZ61Kx+I334ns&
z9(NVuwoCNB<dbKOMAX4?RE*Hjdf0oikMt-NQ7>Cn$L1r&a`^!DtC5264pRNp$LW(l
z?-6`WoDbKNO{sPZ(4q!JxX*jNexxuln|GSznYC$Z0LRNn`DA)uttSXFh~nTh98qA}
zhOw3ggMrsb1T;o+H=(vSPePN$@dVn1PT=6JpQ+!R##Pk|Qa389m^$zUcR}X$%2&$|
zG?@l<NIb=HW!P&A#B~T{2R3CE3-4F2!mX<2&HnKF(}I1$qS$!@Kit$9_<`4jU~Y9^
zErDVHk{5eSC|YMdTE81r6x|EY1(l1V1RG&>9QCpkSx#u0Q<dFc@1*p+Okq{OcQqqn
zd{&A`h9)oGpI2Fpw3i*#o-IDyPa^TOB{+KWbsNU2C{7LUtR=YYbVOEtelZ`Sd=~vo
zA5w;50G{Hhvoa)j*IXUxd_1P2Si|ejm4dMxC34+8ZoG~A#FFF(Xyiw~e_dp9%tdi>
zF)$@JBH*Gl*kPxjs<wWZ9puljSKcvgmeOD0%T_fvQlT<H{5XOf(1~zF;!@o`2TQ-`
z&Gg{h8}7T+>?n~>Adz;aD525ZQ4XuRy~=^Z*~N#draLw*{iLdl3#Z}SAH*4FH~`V>
zC1bFCyDTr6%>4J$ec=F2yTfC%CpXiYZ4Qc}HY7TZufbMLjcGg~Y_jTdB<2~nd=4WA
zihNS3uwsF=%{%S*pn>DeQcRz2=~H<aA0C>qJW~p7U2zCH8UjSCUy4?l<lE0Gl+ai<
z+*~c-bqf=x$)`otL=PRy%ZDAO8pnR%AACK~W`J__Y5Q010HJDCBgr}`#n=>FLfh;1
zsB4yIiT`RXHcjeA^mZ8^v6&sUKj8s^B5FT1g$)4i`gV$%_B(M;*~~LMoy|MuY-@)?
zGPU23jwxk*OyfTt67P2y6gFS*e(+=r14St_DFV{wjH86rLN*Jk7P#DZOM(QD1j(|o
zP373h86MG;pzE9zt3;-FbH2C3TYk1Z-baRNT{Z7))<xx#pTKsf_C-v+Kz+XAfI~BN
z|2%OuXsrzR;v4_lOiok@#QTVj{5<j(LrAKT9L?G3ecJpA|Adgzx!5<SEa=-wvDv!p
zu~=C*^TNKE*j~2a*Kfx|%&2J#0D>R0FA{W&3hVYERmD#~x1bSXeu4=c)eW^R9u4@H
z%#oR=Dy=_Wj^JAWF{U|w0R!H~lw5D;3AQ~$SM(=zR=T+?Y&h2{>*t*XhRaD~?f+7W
z4=_YZX4~`|{w~fyLDu@xSzlwKHN$4<$3@sp4|fOi`?H69k%OWS9_Ic*zAJlGi|O~%
zmmTEc#sfkpa#g_o&yJAWRhkD7QbU%#E)R>X=Co};#LKb_W2>~Ce9DJCi9V3wG@)r0
z!86ev_kh$|`pEbe97pA?>YbyhM~E@gU65p2a{{~0lmpGon8eCcO7#LTm@WC{9aRz-
z!?_wRqRhr(K%S;if=JS^bT0U)4g9HW!_jc6JlXj%7S<t77nKFXZZD@GvO_1DIVCRV
zn<?c>3`@HxPCvY@w}LN59npA_ht;sIE%is&8v=q}FArsj>r%tc_)!OuR`-TY|Jgq|
zJ5ssL)P_|fshH(1Uy)xA3v}8+0K^Z1iL|%n;R6-RK(95~tCv3Sb00NBB>zxHE_)?p
z3=E(=Z?|<oWZY|+I-b^euCsWWE;dIOE|z=tPm^XCMeO8QIUQ5DEs$n-8=g-e(K<}o
zVI}@ZIGDziDbc*WOZ83^y}#!^sWZj{H3?g01ombN>VB925V)Embv*~_cgtLDIADGu
zU=oe@?a_YX!=ZGawb?;TkYadsOcYY0=k_}F^JgBki=}gOlW&S=cpS?x?(n!p74H0l
zdP@DLZf*P6sy^^S=pO#4kP6OXyvmm$%~K`D^OBF&cxJfL<Hb4lHB$!!HYm*ju-%fb
zT3o2+<fe|pW5;nXQgIkBR*5FQ=Mx)Q194+@JUErhlpz*h<m8Odbm%Q{0pVuh5Z`d2
z#f^tLlcG<AVg0W*EM7E**{oIQbH_@~@X9cyyEt{dty6b+-Edm5g+H%Pes2x}NmsQo
zyIaYh(Q7|7Y)5)}x6tR;vYlVjUFAP1Y77=rc}$cf&cpLs7<3#t#Ca+Pr~0zq;)hii
zk&<6_bC!Z9BkhOJ2Em7duICbDsJkQN_)ck2KhX&PERGu=L5DP;&<ebD4O}K&_|Uqf
ze-KtOh17wdf%r5FohOt<C=C}2?s<ChZAMs6pg8@9v@L}hbmD>h*~v=Tk^)8EKr6%c
z4O{@HSX&vHqyVc}fZ_eshRcRP7wUoAPQeBay9SfUvOnQmS?dm`*otaXdk}pG1;sVg
zdC}9we9k`zn<VZ?0f=b>=oZ*#ywfbQ>E!~nv!K_VKW%=rYrg@(0rYT4N&-o)#p+?2
zeJS2whk$COkB9?V(V8i&t6L~xOTGiybd$<2fgaaF%P3zY9B4oUt(Pnw^Y3>dD;dPA
zWnc=xm!`yBt?^2jFzX(ClB1s4a40Z_$?7vC>=<GxxXa&KY?<zg_(cST*WXBoRdRZZ
zN>(wZ!;Cs$XhlgtoYi510M5V?Ci2I?Hcpuqn1Id4F+eoT*kL73`cyzOun!Fr$JR86
zky-B`Oo>|0s<jyvzLlG?A1)|BlUD?v9du<_16lQp6q*D0&(HY!y$hR0Z!g=T;!ILa
zaZ$0@Dt>BuXIHRnT+0c^O?xg<!9Sdrfdly#VSTJkOXP^sok2bdo0`|=az=e4L@6DI
zx#@f^i_T4&R(maGhugvjah{wwfTR32ROYapg+%-S5=?{$?IfMHE1H;|yRVSTT8va}
z1j~J`nMb<Y_OAf|h+Eq7K%m>KJK*%k6Z&Y0OzDQ+&|G*;)BRbI*~8sM0*V1{@Tq*=
zel1iS(+4&vsb&U#e8LkjLAZ%`$DR^RqV+M09eH4wl^7%%&6%;yZEjKlrB0|e7n>PK
ze|zdPlT}i;T1i~M=&#L9PMn6vXOuSAU6c;tqZy@2BUp~v9{7V-i>K2<sJbP3P6>^o
zlD|5sMW1JMFcEjqp<X>&OA=oX(^RjL$^>r$XJoUzI3#^4Qjt3J<n7|aU53Mq5^Ma#
zys9%biQtbfSVn}Z9`i^gUdn}8e*MQ5dVYyq5IVHEP4Ef$A%CS*KKXp!Ws!GMYc;H$
zdHLB*D7>A#{rN<;t`V-!jE*^}9?OyDpkava&jKCeh)0CbaOJBsw7P8JtE)lp{lXGV
zAJwMmj)mR^kvr$3@7u|TVP{k02)F&!n)5Td5MuvlJ00J)$#1q7-Vv$MeK@1I&_ZkO
zJ)qQ3<ghAShSXi^I8j}dORA_&esQkbsj^Sw`JHz?Pfqn>0#tnLo_?zx1yHq%dCBr#
zb!DZw@!bneqmUQB3(GLQR0qq8P!;kM(tbnpkD^&lK&2W^&%?o;M-rsEAL5*$CPWM~
z$#!d2+HZ;cbXjKOdA{TseG&$|bPn<gc<eOM$XmNjzP&iEuCH6uNz>k47nNs8QC7sM
zn7<MD+i`X>`eu_bFHv!q{de!dDV1|fE^;z{^=HptGLFbBN+tPwpKeJ}+PyW>Evm~O
zs<J{lR#Yc#g~Lbon9TH(ax?ncggPtqA1*D9m!oXBZ}|*z$iCcpRjId4ZiLinPKw-Z
zqZPsnPz(~@xqxB8601%PPwxq1IgGj6+O1clc^5rSepdrs@=F|zBJO_=Xp1{CZfLHX
zjb9C%28h`7yhzxUlTp@sn~96EMrQ(m8q6a2(xm6=@kJ)O#rbOizWYN5e;iFAJTJFV
z%Xb)Ku1KxA$N>{g#XU{&1%&7Qz#c7V!T8#T^SRAGT6ue;bq&ZTlDcg-g;z(I<F_DF
zs@=Rsf|z=~`uvZBB5e@v{gh^B6lQE*A2Ptk^{Ci6N8ke{FzuNy_1nx{p?s_?Wwn30
zd25k@#ATGBCo9%nfPLXMp9^AEuMMOd$0b=eufDgHJT4$Xz~&LGhkj{=LXoY0`?C)g
zV}1Ny^=hQGRKB3aL)!^W9T9TsTqj}rx2+kxf(P|w?z&d%{=&`fsCt-Xs6};a#I;++
z1CX8SNv)$~UOn?Gq?HYXZX%qASEoM(5Ubf%!I5Sx@gPhjatMuf!_qUaeqv0&jd8(z
zs%6QK$LlvN?`Q)SfV6mNcAF}f!>hRS#%$BatC<w<3s7g)^PbfRH-iH>um?^&gtq*B
z)eoa5KuZwv!A~NGZsvMuB>N+|^+sb)AHp><(&o`<ufkgj<E;9~2P>yNqjI~2FXqaW
zEm?}{OWl7MBBx#4X}Vugvs?poy*P@qGDfO&sVDavtE+?@u4CRb-3-(q&ncw6pd-58
z#CM(4g+^=Fx%SrMD@NU2s{!#6Vm(2h7lv}`J4wAdOiy*#8M^VF#^^lG^;+QJR?Tzb
z1dtdPD1pgtUb$$+;&5vrWp<|{aM3?Wh*Uwtix&yOx)1jQ&?P8GgLS8tjR#HrMOK}4
zAj)XUi+H^`IM0EivGqD4+%HPxcB&ra@jdBqI2X~!(a5bLKi+kkVF)QZG{R02QIWT3
zyhCV%A6idT=LJ5s%Oc7YMZpJ+CykyMhrWnlLVR97^&1OT9FDc|j_=v=L7@jPJ<S6}
z?>tOX55}QfWh;XT4y(_bBD|JgoFEeI&MUqwB7Kr6)2vm;ToWg|iVFmyWkqr?scdOk
zOU71m^dG#Cy2(pV!5$%K!{&>Si_)pS{L-1wxWzI&kw3P>e!m@J{)h$`nVVW#m1Nkt
z)vt2`w;&-ak8WXEG?HaE%GtVUQrGG<dhI6JbXa<I8mpN1g?CcOAxU5(sbO2kUH?$>
za->_mKbKIXg4(Q+?=ClE0M2deBY*SF?3(P_e1W=7?<(j$ql}{d4P*y{G-*+8+Wg&v
z%KTs{HNbNX_5*Pigf#k2a&;CvpyP+fat{c7B?!{e*w9T<f4<`M0tXfW<g?aUC*n%D
zQ0}vy-pKG?Xn2xVo$srM$O9if+G&8?1WTi>270eIh|+kiqz$>>wu#fFr^L9Hp{6Hn
zWz1r)#qqm(3Z`oxZOVBM+`;+^$IWKLs)iR0InCfeaY#O0%?VLZkdEFsj;PRLIsK9n
zg<{8di~)}Wn;TWV{@}^4F$_}4KN9q2&ciAiioNd6CJ{)-las6O3+hNLJj##29ZX#E
zF#=g3w^bQZYDxydSzN-x$fsBiOx34*vsJk9-_DrBnVenAd)=SyX1F5A5{S3);074t
zdMf@{_PxhQJ<M)h&U54DQ(l3ZD;uoa*H}ndg^>w=8{I)oXc`>U!6HPxGm8x;v)E~G
zVQIX(J0)`-4YL?@#~AF&b^x}DFbZWj!RM0hLtS~(zccA?^KLc!2{{Sov72_pT~MyS
zt|4GTt9Kj48Vwpz|H3f?O#iyu$)!%cC+z2L5k~k@&Q@>quw0%ZTmEvgZZZtbW}@EH
zr!kQCojanjgG3unrK0y8jMKmA(sp@5onV!nMaskvihR#r{MeP3%WI7-D*DA$4wVv?
zM`r)9^>FN^7zHlcD)CXY*(v^}#t4#|+sI`OT6Y3T5ZSvLlx~3bJOMe2yl&J}<B5;C
z-}tZvzDF6Ks&Da!d2oOOd9s_fL(^wkuO-}zF~bpGWZ5kk6nT}qPAZl35p(Kl6K8fE
zm#iv9FfW`dR=AYZSYF}Wg}XE7wrY|WJ>pVHO$0k>Ee`r7DS`8A$yfD^n<KyKq&`l}
z0qA1wSm^ujMO$)LOe#f2Cmk%HQZk3vYo-gzOB=l1=pMPEV$QgJ^K&-oUe{Y<854aP
zjARPwOMIe(>xoVRjtB6YvNtgzu>vH=VfbUO*KQoe?j=bsP+WsNT!q6$^_}q~ge2Cy
z9y}g2{V}d-5uWHK(1ovrm=ilp`ZOPN;n<^ek{Mg8tl&uH#n8<{7_TSc;LB#;Go9Xy
z+E$&R!3a&of?V@uuR^JWu7+~7P9sEFz3!8N{I;R^$MB6r@6Jahh;q(ZrYq)=12<|o
ziff8q13>-sMa7%^8mWa^vfduk4a17#_ze#-S6Z{1?pQ=Vw68Zw(~j3Z(koe@7=Lpw
zu}MXg4DE_xXyK2=%{V(0HHLj!B;hnyay?Ify<Th`+eBacZ16?Aj>>Zy5nK-v3mQU3
z1;=B#pM>=6r`uBmR|Mz%_o<VD*9Z4tHShIR=o*-XrW+CNzSZm|3Q?dbJyt50d)3C6
z;Ze#Za*VU^=rb!Y@V)c0N3sj7ATmWUQ<eh2j&v7V+p)I$8|q$HFs4{lCe<LrwezNq
z*Q{;<z7nJ)rZp}|r;^JPPISw!#=yh?!uk3@y82S^&iFgvm1`T{*8w2Gu^hS0Pgtbs
z*c+*!*Lv4y9WL`u>f<|`AdU7%Ari7L1AzpAn4evb7t>K_!}Qk@@Pu%LMDMmCI9Qrf
zPue)VuVfvNbgZsmO2$O|iy`h21;0UQOQsgboY&3Xz|gS1ABxF@Km1{U@iQkjLgI~H
z8+-cGKiUt;dgI;sY+ZoV$lw~@YTXd6BQ6vs>xqX})MSPtfWvvL;9)uoya1P&pweJ|
zg#p=aY@1QA5|NI?A#bNMf9P=_QuQHW%uvt^aP57Sds(y}D?=9>4=OH>G#EhV!?{=8
zy4T5X{Bn7G>TrKmaBUfnvQ%BwnPQB0oqYvoHX+fl)jJ96pWvC)_Y6}l_E-yF*eq(g
zGW(|KR`a4CZz7W*yG9G8t%;)WD#~@Yg|WXqWD(+5)svez*N_Y=d5xX&D>rH8bZ>H!
z(!F<CRf*DbDMan{RnFCQUQxfXc!r~-odbM+LfUGnISJmgL2X0qKK-r}fAOK4@;t>D
z{#M%@ba}IULj$jH(fK4&*%|Dz=vGv7P>9vf2rQXibi3HMs+@8*E<4ei6<{ch8Ei69
z5V&Ve**Ji|O<9b6uus*zH4jr8d7PV#C)7gLTvJ`3yqV*khdaSSI_sw{*<^O_I_+e(
z3$;lq%0EJNO+iLY@{gp9%+jnfHLVI-5a%fm(8%l67PDXL_qp(}j?d+(Znyeyhc?ju
z?W^d(0+*?|-P1g)yI-{#aya|-SJYir-+9t^O^fzT<MM!93Wj}79T^-pymO9JCtz*9
z5jwMmV>=aR^M@@D@NvL!PE*>h<!v7>oMhH&#k=r^tPf!OC%7}I>yqY*v{zU#QM|;_
zo?wz_#xOpfkgX`8j_F5wq!1!)=*L<(Cg`Z4QaA*gv-t@m#tmA649`0;zjk!Xp0>%&
zwKY#cg{9Ox#joG>?nVmk|GaU7jW~_SFxHe{zhEuJSv6CeLK5ZN&PJKn$p+HtNqzB7
z>E!^37T06MC9}9HJw%6c@0GEqhpXuHTZc|Eb6_Vgk+-d}{S(0vpW~o@5olqJ*}_i;
z-NBQ4q&)$-XVo*P4th$zWyp!zi3AUa_PR2KP6az6S0UI&|IEiI^ggtw?(Fj(>|1If
z@p})$p%A&(^D9yO1}%x9;a<p_A2kr4r-~4w0(Fsev#(NlH~>yt!Ocv+xq3JUCaCJW
zLRoQp{g-~jZX+f%C4f?+^JKNaTC{j-){l*5#<Lj6Yf2`g>UhzWZwesAIi!I;|5+TS
z4}*Jt-SB?6(Kk2fyJK<(%`^u`d--t{T|6;BsFf?y{3I+7(ZAbPzyC0=m@c2@dN!`k
z;wt&H;~RM@@61PrvPnJn8j0B2r|Y*d5Lwq^ADJ7YRjR#2-Zxq>k+X<Kj2zZ$gYUv7
z-O=wJC`G>5zvXDAOgv_=rNJyMpHx-X9iPt_XZH|-KTHL8){|r2>#gK8W$V03DNm)S
zTRhu(bxf8W;f5yGNgF=5#TQJPhIvQqNwZ3G0v3IlGpbcC^q%<}IP*>kYYH969y)>y
z9BJO3uNrAdwM)`3N-4(!z0LZElM*tJ|4}Q{8QtQz5jIO}21M5%yQ!aQu-d;D{j~`C
z#(F+wjPK|5T`hQ6e>_xh@T$~U1ANVLEpR<kc>2U(Q6V}=gGgM&)K7Y*n{4_s?^oXB
zOn4|#`Ymvp=+dZ950;)W2YSR0i0AF7svq14Hv!Ln%ZWa9xMgC=h_lGfg7=|PbFsUr
zBC5lSx(&i<!k6o&Bf5ZYJS>h09$$Z{b53*JKin)n@WO{<`atpGl488ZavE&|GBJ6t
z%uP~3@9#GUt|Hv4?~xvS#}qy5hBGx6_?-V@^hM>YwIClB?LLTT4_wftvjIXQi#&ty
z9&x%TP<$$OO^rx^TS#M<US!;#GBOTcIkqQ?6UW_nl3%j+q6X20ua}Fj>F~A&p->Uw
zNjy3|fbS&6=7@)j#Pat=W}7<Vtfie}U{AaX!S;OY3-Q?x5lHe>339+wf*O;X`l?`<
zcyIHgpnWXVx9cG8UvNLzn|R}jcNSH_h=7}l!v(DJd9`vD&xAmjuZyJbM|>Oo8iUVh
zPJK1$Cg!rG=fef?`+Bt(98A}IHTD>6l1C@IqQRqfyw_B1x~ClBRAYjjU!<sA+avtK
zeQ&d*vNH_W{Rcj-M(`C{q-v>kQDrZ({nkHs5WD8TF8w1Tg#&d|xc?<1fwtcW-{8Gb
zID}~2R0=RoKI8BtwHG!<+gcM`MWJ{yMN1y2ai<D2XXsi(a|hrQKJFIm7PTCZ#tKi`
zn*A8oEf2pNy<*bJ0S=Z?BlxI6lJsaf>v*Srn_8aaLmBRZKONo98m`c5(@<!iAcGjK
z1mu8`>FlX~u9;RJ6J$8XDeMVr=b<pwF?rw^(xnA@#eFI(+T|T(I`;WYSP!?}c4t-U
z@n!?AedOzBita!nySqRNYmJIn7RCEnFu;U9u$6`Fd5*Aq3_E0{>2{riQVlh>c}U~8
z655aRZRvFsTuKAl9)J=DZDP=t3E)?S;6RtSu!&N(bT@RW18h_bHJxo`+HEIPbc|KR
zEJL~t##~2CXmf=?2BzQOtp>neAfQs+rni!F-`t#muowhVdO~_-Q-9RSJQ&Ed{Z#Ce
zymec3>nc(*N>V+}$S?&+3vgM;VqMddu(kTM_Od|uk<P8>g+h0flT}>YeSkwWD7Q=h
zkIAH+!p0tt`)N|M2?4W;bihp#&jgtfBHaaVQZWj6|E=acG$sF{p#zSgo@RCUC=n0V
zJ?nmh2G7P9w%{$7^gm$(6`eWq1*O@_aoaRqMY<kcDIwBlX+VAX)n99Fjwtf1n=a>A
z@?#dzcMSIl9gg&ixOZ%7D>dzZee(0PZcbU!!`jta=i~?v=4WqO%0Rg9{Mes1Pv^T>
zyvp>Pz@1<kW@fRuD)KxZG3!u`FZ+CGg79^x=3Oo5YSG?*$n85wF<#)E)ORX#b>HQ8
zsNk8aJ8SVyiVM0OY+I~5s5dXHOa(D}tPJP&ui^`vE~I>zAG|CN_wXtgI0z4epS0<L
z^0Xtd*&t0xuJK9mstKlN`2KC_l-3XFWqQx7ydInir!7+=ZVO(LVAIzxXH+Fl>QSqN
z+|+|PxZLl}Hk=#Y*rUa+*4;Z42^<Bhz{{OKm&`!?`s+{nt#Xo_$-&EQj|Wm#{4PX4
zk?7lVI<!B=5*&<)!<zQfKw2~>cG^|E`tDNZ?|ZK#Nc4w<xL4TuVz7()Ad#kkVC7|G
zw8y#hIUzM_iu9pvE3o~tANaOeI^gls6k$;P6EsOCrmv5IMSEtXY8shMf9@Kz6KVp`
zY&*GdFud<}^#9y95Djm2MQb~uJ*M!Tzh>H0`fSFwcAb(GPRGIEcNCHZFg8&W)9cvG
zQ4`A%-Yss^A>T|nqHl(ERtMhtvBLL1)n6ESG;2@^(oQnOX4OlGq`EA)&VNtzFywDX
zC&iaOCyr7$55M!Eq1lze!eNpf@wE@B<#gWFFrT_QpS7vu>OHo`&LvdfZ;5+s&yC3o
z%*Yh#IH3srwEe8tGyI!|qYTzpVM}f!bwQ`#k~seNpXq3_aKx0tj(x6cknx=F*-Sb`
z9&h`ZqSVL$^i^d{w#cxGz+if@_z}JoucaY>Cn}Q_l>}B(87Ian<ZI+ICN)2(sAlt9
zSn5gu#McF71~6zyEt#Y|0sD^6=K+gnl7ByyhP;T$j`vjv2z}4Uq(~GC&rMux4Tii;
z9zkJvOVaC+Ir!Rj^d~1ab-(6mwnD5z`fKu!`mD`Eox!oZ)Zig0L-<<?dfYXtD0`-~
z32hr;bN&#{=e<lLz8fEbtES{_iy%;rp^m)1|48_G%Ygp1-I^Tl55wnybikAF@)jOD
z&V(G1p@Wj<x88n?5=43BN;to_iZsQ(6$H?kHBoeGuklM#{P6z9-WFXFE)sD)SRZg7
z&=`E&?2~r<Xcp6gU4?qgbB5`Tc`wcmGh2~nlTkpCO5$knz$^81vUo`zGAvUp)=Uqu
zkjDy_RjNj$L4U0FTvL}RG-sR#D}kJn8$X{vF*iXo!En?`nC%bNbACKZl-=}XmBo$H
z7zZlpJ^o3&;M;BMjfdcoCr_YoH@=1ISIB^lC>vkAMYaglKSXHKzeH%<z1n16;lhP`
z;J>X$NfDM=N%5qF4y83lgmSl3I1N&S9@OgwAH|6H3+U|*=^mpbaGf!esGkD95<6hk
zA2Q(6QVex5T+rqj@dwQH9&`^HE|H%;?UvW8@{#L9kj@~5F?R3fO3Xt{%}~FxiwD!F
zIla_qXEfa~CuqYU{2c{=E4Nb#y;N|}+>T7#lf$MnQ9Lf@#<*C_Sv=#7E7zj5jhP`@
z<YZRa=HP9|PS)+oRrs)ZJe8CTmQ!KyeyF=m%{qxXBj|Tl6g|7+hcPL{=ln3xps~n+
zq_CsymXGx#$-`z^-!lQnx8V_Rv|mz|aAPmH5WzG#+NM0Og=L%Fwd?XlqQeZ(s@$?^
z=OoYzB~96lEo^ZZYXqP3!XVoS(w^*2RFJ~CVYpRKe5EHz9Xu=&`dz?N5k~+_WuCpb
zh0W>FRr1J_!GYS*wA{4w^}&zu60~`MF%1M39ydZUZoFA-d|D@Q*5`$%A*sGn(-jR3
zMYQGm4$|-WH9pg&UuCkvYsU<1%((Vpj}cq!u-POf)8C}fbboySER1R2E}9c}xR(-<
zW(f4)A2gI}h4GYc9Vv~_lp(y2dl>5?GP)%UMXC!wCXM7~&vr@%380@tywEDVMKa(N
z8V)`i<jxVn&hb2${ui6oA?TJr_M;6j^4iT<P6idYX1@0@>(1&xT{x<02HZM6I;8Oc
z@@og!ou8{|&Lbch)Pl6nb&N?(fA6n`>26v$WDxVu>TD=Zp=9@;33V}>-@R}W-Fsfp
ze^*(eK5rb*%{zIBGdsmRIUShc8etV42PGHij`iM2%jySP85aq4HqPYtAAFxx;AWY;
zM1zjcA+(?+8B>v6`d`}f8RDd81-hAO^rLz`W&(aVr|aBo)LVrm8kHbLY$>+Y<FiTj
z?dQ%%+$CXRAw0~oyR?ZKFC@(<s?hW(Y)g*yDkH|3`Ran>rvfIDy(gdsMsX}w$Kwkj
zT7%gOymk05tu(D>*bB{H&axL)T0E=?;%HixNbAd_x?b0Sa0~eFue%gu@x48=mzz+)
z{n5J@bq(Z=UZVwkj`@?bCVOj5igO>F76li>74U?P`9hpf5_DWHM1OmoSh(%?OMTIc
zJ$uJH<(RfM9u-(sgUH%AkR#Eu^fl?8@teIxtirCxv2@7U>+WxpCoiG1H$rFiggQD+
zQyvW_g|23b_XolHjBR?}Kc8`I3GCGq#@gve6{a!_)KS+h%$wr^@&<wi1-GM`Wb@72
z;EVcKo73%XIkE^{gTl4?zCr0uulqOSPBgcNZpTXwF3)$Y>f_x97K;ROw_P4m6Ine{
zUANos2IrD-v_6%X9B#9`P=DAs=&6g<`}J<4j|NXRKcU@ZDIfeo+qy~SxHET9{cb-)
zvr_L0c3Ho3MDgsVSHCOb{!{wZhTOu#eFc0Y@WM`r^NQ;oxZ;V%g-4!-?U80%OgnQu
z;f@ny7EKwZL)?bOx*;;?K$s^~nA!&^#`XzOWv@NzQ;;Tv?&{Wb?Z1)THQ_6H7n-k~
zH=xOYnZTQOSp?BP!1he%f&Frk>0iuW;VHqy2+<|7w|H4#XdqM)y4_AL{ql6w`6=D4
zz;=o1?F+SEvo96uMhAQ1$4H^|<A{qy#{4fmJv0;VSzn9bquGb?>?2oJ;(zrTqR}C4
z6sH)Xt0B9)-`UO#{mo7%vhAb#j!{i1Es3=5HGK(nDPgBP0>#-uGKGK1Bo3o6U86%|
z{Gg(d=$Cp)U$!*)r}f;JlL^pBg)VeAMPpUQC5)$2K7U!x2zTgN9`~=WCAzG(LVNTd
zRL3T0IkK#@IWR)b9%-IY=;teG8xez%aT+_DH@fmzd7F0Y5jfDeud1Rd=08VI1%!ei
z3qR$gsdsm1^H%aM0=)31hsz!n{V3EjE^a5649v}@nUx?<kjgu?_cW@_C<~h`)Y(O?
zOu^U4ZvM@!kv-MU?4z4u8bLPOU8hzOuVLKR&V2I>Vbq?0H=!K~U7gTi&1(4Jrz<U;
zk7BE=DB3J+dOd#%4)fxem3)O?2i&=Kt{YP0;b4>>BZG#gf)YyJH9gnW<Lqs-)^T+D
zws85eLAlziii9zbz}Zd%?MN%3e9dDTVkiu)uI&Me3FDyc?yc0Y6tPl~PIhh<wVIO%
zMDqOcWcJRQ=bS8VbF$ufbUx{PG~<~~CrN>uK`c9gyK1>sZP^e?mofg%9O=Z|AG!z<
z3Y*9S-n|8PMhQ95t5KL*zkcvx=V_4ind`c3cbj~dU!oo?GGph`gC%A(mRD+!w5Tb5
z(C)@1n_r@zvYAN374_O%Nfv9z<_&MA^^`nE2#=Tp#HH?W@5AmIZlPiN4${~n)P3Z{
zo_bS!FHR-b$C)oah3jIq1_wI)EVcK>A+eR{ETiZYT7fIpae_)DeUh327{gAJOuL2;
zua@ofrn@05c4xzo$I|(Rnnt)>t((s#$TUSuI;m7Q?np1+A6&@|CMEH4;hRv)q)Amx
z=D&AVpvcLSrdYq%xU)`-KP}-Quwz9=>#i9uggfUL)aF?Q@%!KgJ+ZE9DAcNzo?qpx
z*UwAs&0c)_j5R?pH7<Bcv^li&1#k05fw|%&%@ppDswkg$m+tq)7fPn5eRDM8>d(Xr
zU!~w5f5CH|2|{A~Zs1lsxs&>Y=E#xK{wB&^;QguMzI*sRGkArPv4C-@y|JTw8?R2O
zt49(U0gTo+L3Qlc%(6_H!5x3>Ea03isdl<gY5x*x*stgJD+zG-p3X-V8ss0zEmER3
zWNieUkF7&NX<@(XdfV36%d|WLq9yoVRFJz=WC}=(<nxe(QTF`|2<UdKy&A|&WL)d<
zcE|LYcp?gxKPOhV`r_Y8$8zU5AkxmYc6`KhY~dnNi-AGn0KKEL^|#n?FM!^;Leu8E
zSHVR+vz5WmN$S)p*<OyE$Av!bBxd>znhO9+;!NB~n~so_+T@@ZO!KPc3DZW;%s=6I
zNR*yVi%4?HTIfOAQEF627QM!R?l+$_A}Mo;RZFhljU?n|oLpIqo3r2m%u`*(>>rA?
z!MMNmfEU%^!<#luDf4>~4|tnQ(^iiBL=~n_RK2as@qX1frZLbk&7sGlhiv14^Tf+5
z%HkA(3n0}o5i$RHK9P|9NraD^garzK9aU^V0@h``MRN8}ti3N6dr}<+)922IDv~BU
zOqNd)x&XEZXwI@`>871)IX8s^ax`{E9xjIm7m99*({(w9Oqf2V+i{A58NhG*?nBf)
z$vgNmpbs5B;M;};J61F?9Nf*ITa&K+U5BuJXLFCwPw_F|Fx5R``d;Aj-MQ`&5o-4j
zU|ME1EBP_AlB)ulDd{R9p=e!SzV(+q1xs`8fTz+ds9^+5cARd_wvmt7O`VaqNBGW;
zfEvX;!4lZvAtSya{H2cq^|uUMzSY+HwbEHhlj((_Vl!OkO91Lt$IpbB+IKK`6*B=P
zMJa_Yn(I~cXTR;sgoBl+O-BiGa$}5f0DC(_I>Jwi%wr#ifwD@dud=-~ZHVJ!Ij=hi
z2v^m2VlwSTJjFB*kXIT*&UA)4KL$bI^q7H8nw<qA>?YDUQZnLhoNfBC_wa0x>|z*B
zq#Y%;T{@a4RTyVK4r}~S<8I+-a#U7>Hb?%zP>&nmuiD6Si-zFmTPMpBYq;5p*A@8z
z*5OpIkViayyLv*?X(&AS!nq0x``vGs<W(ha@>GGyBHQaIRHD4RrA+ae$8KI$+VP-U
z&uQ+wdH8W~))+FTDWxh&ajRUZFvKxjWX@(Ejn2z9BxSOtWpt{JYgV>z+7E0qI#V~j
zmgAa7GB(`WU=)LJUh*7<_<V}a>2AAeH)3&-ytskLLUrAW1AN!cpXw4fA!gw;uG??(
zc|%cXDWUzZ3xfC<y)}y_=&nC@&A*_rpq1EBg`^8DT8vuu8n$6~LMu@1DIBO$KaOE9
zR0sF-0i}NMIhf2-WDh>yW7W%s9-`FDxw$)V!K&#(<iA=<<A~=Ae9){B#l$`)o&|u4
z7JlvWlgG=$Qq_5nF%>2RRS=Zn)tnw3-5gXODS(|^L3yfCNma$^{`GT~Xw6&>Gaws@
z8K#S_ZI`z4?@>EsfX-z1_TMSUfXY^V*JZjC^^&XUj8yokS%=cw0eC3mpaAkx3e5Lb
znFYirxfpc!2AG_+%FdiOq7b&7(4&H*E7P6xQ%8lD#*qBhZsEKIQhs*2nRv<Hs(8~G
z+V1vFL8DGO2<a1oEYvSvOq<hGHJUWs)M;Ot;D*$CAw4>(-;lRHXXp6Oedd2I=rJkh
zScaQZQEw2JA2~JypLSUxEjM|;zEk1r1tC=75dcA|Mt0eILtVeRNTOx-yP&5AT1(=z
z()yFcH~`H}+1!+aN3*VP(r>HG0UCX1v|%r|qD@)fUf-b#PImFWf0s+QBZy5s(eE)Y
zx}#hfDjXYfzW-QoFMne~Fk!MdtlFp)DluzhTe@FBIQ5O_&?!ZRQ}PiGHxuC2^<i2L
zt+~_+2c<0+d<&nI+!>EJtU@v%Gb%bNO)H4i*v4pBmQ}aEG09jP@;|DUMLw?K4WQK4
z_dDi&>`J4VPX?V}4T8t`K_+^BEi{tGVBA#BX`{M`*i}VA`l}NOM#gjEs-B1oz4dU^
z#p_QWDE<$7?->+T*KG?oK|o0&3QbN52uKu=j3A(b1eK)Zj3mi4&;*g(B*~J4q7qti
z&Ka7F<RCfcobPJi=M8nv{i@D+{@g!bl~q_24SVmk*PLU`ImZ}bZfSjL3K@bnkmJ~3
zrDs!To4-&G{d*2ujREm-+_)3_5b618gQQ1dDJhVPCF|ng$(oSO6O%E6g6$s5B1OHC
z(&bd&L>B3e$?22>q1DN(#2M2EHj&0_s;BWTmQ_pN6C(ZPCZf%aKG#fv3K}GUG0r1^
z<^?vh{vrRo^{kKjeUC;6!Yhk^ZFhoZ_>9=s%Q0!{r~mRB+XdD1TS^qh!$czM;Unti
z6+9=P%>Q)tB2BAavQ0_NVj*oA)^W{H0dao0pCxe*jAvOp3(9IvSuo*~cM%{Kpv^xi
zylq2w>qzOIvqo{X<!b6QKPBD8??~cMB3yY$Dj4K;w%YL{#R+t}OwHn_2n!k{U;R@H
zK-V$i;sYj@VsZbcnwqE8LE5x3{o&nm$y*#j#xP>N&8G>(SixpRXT6Y{zKjd)*2AtF
zAEv({CsOiKbo7Ulc_{37c)4vsZ7tmQR=>|qucm2I^oTl{h_r~n?EBCfBBmx2oG~&g
zk5!7cU3bJibGkGZGbkbD!!}r3j&aW70FgA}yg>AB0ryAD$Qo45n+{l+AxG-P?8fa3
z$^O_cczZj0nqQ|0=kD8Um4v*cbvbIRtI*No`9SzSP2VnOciZX-EyVK;wIY|^#|0^>
z?p4z=W?9cW;s7b;@Z?Zjs7}fFJ@iov(_dLy;)fu9$u>GPeR6DkL?y)e7HQCCTHHTM
za_~cB|AL_hG}~n*`XX?=K&uqT3$X$s32DNwKF?Mn5NML+5v~h!Y`f2ro-8(;*}8cX
z9#}H|Ag{6|!66Cho-y2yAH)~X&VSZ{8yaEQmU4vIrWMl_iwbG{z0g?j7e<<AFTv2$
zseUzT`;a4H`w`M|o-lD$h)@4g&NG@QM>W4Y;qWlJCiF%1VBx;&`_l^E=}wgSK+l7I
zS;NE1mHm&?{|YZj`qs1sbB~Y+jj!;`6zxSfO`$?mVfio>M4-UjyI65wJk2oLWVtN3
z;|BmcKer9_DYTyvlz+7pZ<80Ex1u||)3@2hW?81RhSisyv1FVpw3h3Vf12$;yX(<<
z{U+~CD@5H}6Yh6LsD`ph+!ihbg*4LZQc&D5>b?%7uiT_PBFFpp#TF=a5z}SQ5i$18
zfpD=cwJd#KY0l*FO`n`MeqJtC*F}6qao$Ino9+NSfkT5$BXt~mq+0lo-YO}-un)Hw
zH6O-PTr$1>;R7)5@UB(K66=ctWspa!yiA-EPNIolm7jLzJsdQcak`v26qpb4kJ|qA
zAy)vxG2RjfH7-_hD&c3E&t>6ReHs@gisQw>>1W)<gTJZ|NIjxhbzfq23-a=ZZ&zo1
zGlvAE{c77$V)m!=)GJs~t9_3Q)o!j4k5O;StJ&;HQlUp8Xm$<_`BO*+mb<7RH7Cu%
z{3%NzjSW3u%$+^ALaJcmx%`r=$+5d9b);`tp$b-=`i2E!XVrP)yWGR5u*dRT$vGFn
zzxa^Nke`;-yDMCKpb2=B;?|b3!jBPnsD6SoX&3`GJVx-QLC$WiFI?>u>I+H0>hP<2
z{yJG*cGNeGs(||W%Hbt6i|4a>k4qLF7GfK4Q*>FVQk55?zpcv}W<Yko35GtW(zpXt
zb34-e`1EAY)f-H@mKFWf4A@t0QTWhU?i&_1tt_73+pF`N0^z$d?2v}iHUAPlRw;J0
z1zS_)>tM2>v@_*(l-2~Jmn_dIc3jsF$8ds}$8eAVpc{S399pzgv`y*hjvmRhNYIA_
zJidS9_3A^Ncy&M)DouzK4q<uj<g{CnsRo72{>qG~vnYT|pbl7TmrHpxEbl5?&vUWs
z={?oV*Xwp~ysF2@kUWj&;7ZZg0!JtA<n&gaF;@74#4&sh=-xYlQ!_e)XgPu}PJ9H9
z+{O3Y!ED?<pJuW?oHAT{S*f6<9(ob&sTkh_HPk3P&8`Y2Fel2TZS3#qv4B9cL;O|(
z?ydQOIR&i}o#ybKcV7yB9sH<7Z)U~=HZ_Sp#7;(^FiBwuW6y7yTphFhgY<*hw$t3o
zd7rB0AL^QklwwQM3ig0_Bj74OxmIPXVn(fUgmY$?BTmXhAh5-4(~?0aJbgz^qkiQL
zk<k3jSONqz!YYo+pJ9>u2V#OQSjLT)7G_fvLbk#OT2cC=A6sF@Og6tiI%mAWm>UMz
zliEsIU4t}_{5*l_vlT?cK`}wmEkb~@aQa0S%ehVugnQ_(!wmi&&03iXtWE-{aJFq5
zb%9_ojd|fC2r;tvaPx-oTu2?4grIlyex>*ZsE1AHs;S?4m3KdqxQ7v>Pw_Hma2HO(
z@&u<1+epug&Je1p2<c<0R`jN>g4Mc7KM_s2g`V;^m9P7ZyQZvz+b;a{ftYp<?|}{k
z=JT%XgaSsWNm%`QIakO=#}N+DK{q74{(W|Ci5-&Y01A^KkHo6A@Cu#2VvsbvvxbG~
zebwbzGKQt+Y8%iVt|WgPSe#3|zNx2w|2)aC`=q;G_w5$npEl9F5LD;uz(;a~gI+y7
zoJL9qYCy&6Ya1)Hc^}6s_uX3##{AUeJmM%%s0Kn_cR+#I%XnNP_spCLs%~-(fxUhE
z_1aO6+j)7O0i<m{q1~;5qvz2aLCT$@cw>l0Ch~59=!KRl<D2a-caV@!{cpalWu`h*
zpU+K~`P&x&p{{y2cDTt}PnS^i!t1i#hIm}$yKxlxz;bQ|0R$JB|FWQgt<!|Eae+rK
zxV=H56m09*DbJRY>h?%OC84S`caCyYs!>wN5UQ%Ak^<5{$m8_bI+Lrpa~O;`zFNTu
z$A`(VN75r`U6PF4(hySg2odD1Xtz$*N^b0i<)!@rC7WYpTil&R9yn@KFmCR_VX6<2
z9q$F8ElgVEY!lT<;Z1#^kSidd>)&ZU?q-hFZ}hvzQKJQ0h`w$}*7DS)!kDq!Kt}kt
zD&m<NQ>0N9ePqh}Ijo%-t53XvG`oY@Ced=q51ZON(c76sks^QeN8jO&<mw$l*3^Oy
zA$CtW8po*05(*F-IrNlU^Iop<`rUPrt4k26gsCj*a}3|Wk`!1fHoY7C65obCEsIj8
z5h2>1Ek=30ssPhiEw}Fc>|j{eIpKnpMHJCH+=xfC9*|qA0Xn9+(O+eITHj?rA-h54
zhnSV$zDvw3CTT3FsC|DimY`qPM;n<l-8mF^Nal}DT+7IU-KZli1W%6Ug=&P1ANLj5
z$PK3UiNg?XH7ddZlK)cLU|&;pUOkgemRVKJ@kpnblgn^qG@;sfidDDyiFnm0$Mw$8
zR#hv?J6?o4uN7}T@OZVoXl1Esm2W2})R_ViupOr;tg+Ov_Naoa|D(Fy(1yz|Yp6j_
zT2RMDzD5ztT*Q$HB?J`18w;+7?RCf+6C_sITE>*7bHB)t4Y6~~9mVmISt`<hBgFaN
z^Dlt+`Zb0$f^5XcOVQ?Z;-S1L`SU1V^WQo);5ge%UP^W6l*-{S#q$!jFoxc>Z9k&1
zM|~JijW@Zn*#jR8-|muYYoJ@QqMPvFyw&5fyy`A_r6}=X&kqo78&dL45x;a*W#0V8
zx42)}Eo@=;+Z$IHiSofEcA_fkW6D@-!tp)V+j`ZRri?=GUkY8go)rP|V~!@P0}&>T
z3FMLc@p~3!hPtdPsce<7`!TlUdC4#xHKw|;gkcJXhi4D@4Ilx#<Njw`wG!okLC2QY
zX90seQR#hJ*ZDyy_{s=Z1-fmt?1?YI1xWhayh{=E=aY1F&J(t(vZ0sE!p0NPVSxLZ
z((w@SJ^D(P07CK&E(~^DZoNd@_2jEbPo4Mm*Dg(uD6&Ie%TUh9XA?s9N*)wIeYCU3
z$PP$|X=`YYdwNIknI7kKUD#X}#^ZRsZ(8&BR68CuqZ+qur5T?HQ?J{%mQ8M+Wf8gU
zanh;FjCTT)o^rjF5&F>_wF{Ec3swjFYR@E-w#eZZ1{x%OaqxSjwc9Z2;nk&vhefkx
zg;WqC8V|YM<7GF#kUH)g?(@x+Nv3r*7pKh5qPt**z<{!ve*m%_KuTnXls@!xoLSmD
z+l)g!+VjHZa-BJ;YUR;9HWhpaX=C?V`fPe=<kL0uVk&9>`J_9-<!mQY_sTr7@d4>o
z;x=Ci2EF6T0<qsc8@LeL9&DztaZZC4V!no+LEhu==OOT8-m2PD{DNQCcpEMlDX+sf
zt9wVQoPI7fN2*RPUOlh_ZXBgud$D1Yq%W7;byqw$Up&Hjri=0MgE^rpqW)OH0Z8}>
zelTe<A`&{yjnj0~1W<r1BjJeK-?_|@9#))1EOGdH9057BIE#XV_naJC31F+Lg&j;v
zlm_$FIT*i>cjl6+K6~6#O)jcF-|fs(EIK;9b^EakNoHuP03`E8zGVGcIIP+#_x*^z
ze{mD!X0w``&I{$kh7UB$;y?R<`pV)nz;7%>4Z9w%9cN0cP1kzb%?7ggWli!>+4PI}
zUC3|{F-8`KhvsxZmD!_KoGqG1mrpdTZjVBdUp5Y|T*m5@#V4vjO`B5X-K)^Sgg0pv
zIMl_-s#~BO+BY@(UJR=adZAif#ee#ClqV<tA0cA;e-<JDtU>+P5D^x4z?L%&3+sEv
z&9%t?jYo=PE?oo8Of?z46mjM}ZY3z_FRL^2LI)MExjp4HZWRf5BJ^prz6?XuKXU%c
z>?eOpW&cF$uUqN@g?RT00qrgOE9tdIr;Z&aR1mVH0l)J{|H~NBW}+zIH*^CLn?sW!
zhcC_NFaa}G?SG|Kt3Yk+a4&!UGxUU>YbCrC{$7IW)lR6tPA_t1+tp~DKtt)<9Z*9l
zp6#)xrJ|TR^-=uv;#Splzk)Y_l#ytN?NHM8G|iBk&J*PI9T=BaW?-4-ULC6~n1({`
z-dIK=kDiIuKhkCp_a@?l%->1I+tu{60dqPDiM**<$OTn{mEYtN4Hk`)XH!qF+o)(F
zUl}lj5H(Hr<fU`dvz|_$v*5pXk<_zmn8icVG(rM@JI5VcT(GnVhG19SKXyB3@pFW-
z7qo2y=C|+`&SmNLK;@I89(}vpi(R@Rog9T_!<)0sUf1qYQg7XCC8fg`XT^tuCYR{`
ziRrNtL1x1?U_;+68#<`n3#3|qzuO|U3OH`6j)ZEXX>K~R2vxa;{pLs3#Ta^1C3k{B
zFd2d$O{%MsyHDyvOh5Hx5;dx>76V+eS#!3jc#L07+?*q5JU#+ce%8s9;I0=e_?FN+
zy`0~VptUMUGEF0rqv2%yvmDpN(&VQ3C$0q`kd<+k&wqM;)!rS<k@1|&rcwOE@js^h
zlP5cRM*)e<eSEF7a)`g9Vtchvdt7v>e}EV1X1&g|d1TMkc`5dA`Q298H;ms{jn<Cl
zWxI>`d;`bZ0hON|b4x4RbN#Dw9%{L7z&1-GFWnVBFlZ0)C%P>q#C}THtQ0;Ok2W>d
zA&vA^lFq>K(z?pNaDHYIXL~xY>#P+MUz4)SEiAPBZTjx37n3aI7c-U_-ElZxb2YR{
ziEg7S?HK+Wk>{70(vg*PwfQDv?#UOX3Rse4&y@w{U__%_1Z^&?Q#(#iB&OPk&KYlb
zZ^vIg0zl;qO0|=$ewUuf03ww_>x$|Q3T{tLF*3c-*{8noxdwhV^zhOUb%b;`|B69B
zMZZJYVwzhw;!7@hn`njGb>FP?Rah6KtP6Q3&VGAr<%Pg#X4k!{8xZ4NFz7yMl1RX1
zV%PokOLkrbo2d5QBc<whYOwid(p`x8<Fp(M`n`u_7OdHa2;uS=Ar;n?Pp@nEmLBs6
zF^_~tcjF+VTl7LMnL%((j#Huk_4cU|(KDU5GZ^8b{YwTDOH1R)xCfsvV_q<@Huony
zui`CD#$d2LzdWjSMhXxU;$85C`;Xp!#dSO!s_ist%+JL>ur&lG5$FtxRTf#IaPf_b
z2~r{!DDc0h>}%_TT5QuZ%6A*wUnOM8U-v+6!f)7Wd7Lch@4oTSDp4#deS`@0cQ*;~
z;W^EQ)^&vHcda9XICwuQyb<&B?OA+;#GpTFGRtV1ljhJIZXI4W{i~Weq&AejIS4v5
zdqdfA|L)KL7$ezHATrNnQhj61dCG9{vrB@hVvxHIm$IiOeO__CuUjybycvL*?N-<E
z%d@SmshXc)qP}ch@AXG#s(JZ+o+`FSGzML19cVc7TR)Mv>&E*AAxMIi)Yg*bupOsO
zO1qeGtIBNK4VZBTFOFhcMcbImnL%F6;r_=&#LP`s9*Nra=k^GD4<c0u&}&Kicdx}3
z#9|HmFb&7Tby1F_Bk1&A8<r?ImIOphY1o;J7P(;6ea}iBg9W}AC%!3B!9SJi&6(iX
z#{KY5f8{~8F0;}<qZnRiD;nioiYKn8Qc1CVKz$_ep1dyW0Y=X+f2f_qy7xZ2B`Pa>
z@|+(XzIa4upp89Ay(<}GNuI6HxIcUCW|(IVmllxL*(@8%j%qw>FdXVt6hE1t@0BKY
z+{mvk{<5t@Q>v7@faAsLulc)<lpIrah>o+0<d&HhHk)aTpb#6sTJEJV*`1|iw@m@t
zg7;GH*fyC>_CTn@-t_tg80Ddob?4C87D>VgVIjy$w0oX#4A;`AG=RB*B7f!Exc7q<
z$)b!}CFCwIR6djBJ%=?@(A<e6d6bEq5s9gFcvpj!@v8B{WqGD|+CTqxBuldY3|(b*
z02nGC&Rykxw}2a?e;NPOHEmSFM~UpGyQ}+D{CnP;r)w_O%RJt`)WV!sB03Fm6CW8&
zXs*wN*iNuv)iGKHpEnx}|0v-%XbO0)z3uox38z+jV-1Sd>nR=SFfcwAX=Rl%8-S9u
zpeln=m+B=UwpU>MJa$wzetm$`{0v%`ww*eec}+DX=1q#s>a!f-TVL;k3LS65w}|0y
z`EgzBd#uAhsln9n;oVMSPgG4op9d#ODLWM4bOW=)7M30sB(5jaW@9HI=|^|<$wHd|
z&Z{j&*I$#BGmC5Yf*+IzREk@b9ZoqTEB0RPCo|4C)<6wljcivZg~qt|_<DE^?S4J*
zRMNR{arw115r$i5<vhx`!l#-tU*eyadpg<5vsg&}$vi7TVDw#U*DcY`nT@V1S>Y5Y
zT-UBSg#7N3mfgFBs;i`)9a7!A`J;~mVGajN&Cbc+68GGnDuH2O(cMXw4Ht3ag3hV&
zdxqjvD&?Ee#3?WGD$jz9_N)Y)a6F)p2GOe6xma-*W=*T4dUQ`ICegDwz8%zn)U@}y
zn_@q)=u~hCog~ghMyqKvXu>B`K<j>vz4JrC;mJXYrsIccy|t7z2|D-9P@^=NB_%cL
zDlNO_7vy(s+}|V(>mw!N5)Jma2kJ1|i7IWP5f`(~KOPzUVo#wJr*k<JI1)X*X5>RC
z@HQ(i!q~+Igkk|r`Y`>$aJ3@FD+a8Y&q)p@FRGULxE1x~7(EqqRN{`eE(@pDoWgZ4
z94{k1B25`BPVC)>f|rix!kk+;t9V((`WhmsWw{ajntoQ$&UHX~R*VRW<&B!q;XJxv
zO#U%?(}(;k?^Lh5`-;-6OnKu+zSr}Nrj4}ai92+hh&$V6vfDw-h&GrbB1QMuwB9?A
z+c*T3NYn;dpUTkDw``{Nw-^82d~*&?CNYBD+~Vb+S7pL?dyk=t@N2a8;0poypOq0I
z!fhuj%Anf98dhNbbm{(<v%trC3Z{aeiX=U=!CcAffS0Xm<twaoVL&94u#KU#y{BQ=
zLnt`W3qwAo%FM@#7HHH8aEkKy!ndQB8CZ(B=w8(?oTFZ3u}Tw+%i$G)(;+w+dO~}M
zOFb8OE1nOY{LmMJjH4zEY8aV}jGT`@l}$1t3hej4c4yYYD3nS(qtB2`ek4)O<NMM&
z@6t(CC^SQyKv!MMR%Upra158qguBjIoy~r}2=|4<?IJ>>b+N+{ro!pFgs6lW#RMWO
zW1m?#kbkfh9T|D`FWg#SH8{+obA_J`RIa9@Wyb2LdY&Sl?ZZ$#4Hfcl45M&ykjuI*
zK9VCf;-||@>Vn$@K8pSlt0ro{#iOpEx^rFJ)Qx7u2qu*M&TKzQxk!<_l*z{OsU8zs
z>xvWV7=!&3FRF!sJtYfNx5BbPvf$wWq?olw92dG7aAw%nG7oC+$k9i1;(a4w-uaH*
z@4b160jq&&x{Q#Q!{_!vZ584VjKw=a(|?y@Woy~>SJLSvm@QM2cpM)?BhfJwSYQ>E
z|KTuDJSF6W+D$0zoW$L;MHmy=Vc~}urr9Ma)VX+4b9_}59aN5GAKXgomh+rhvgNTy
zbDekjHT#x*6Pd-}n&}odz5?ktrFC1yEujE9V~1Ou3)4|uz(+u=A$4f<Ael}fmM}#9
zBLy9+U*}ue@fKoH#AIdKUygc}3#O`j=3(d@v<(kLzI9*h)8^T_cQsGS6~3RcQpXRK
z82oe=@0@|9FCH+Ja>7VkU`D)Lpi!7<-8aoYj-?kbWHl;qqqN|{`ZIj!$Q2+WI>Swp
z4C_xZ>3iIpAtW6gSUh8LqwZ1Be4XOc(_04xRgS(jBx2b1u>CQV4>zTZvsX=W4S()H
z;Fmeh@vT$$+rPLPGZ1-qW3)(_=#0nxkyP&eYS8gMV|~vLd8Cbnhf;{kAJwK0cSY7{
z5k76QBceRhEWKG7kup~b0L*~Kn=xg}S~VnnkCxXcVBnWYNY9Me${Zk^4p<4Z=#Cgi
z$oaZpyOKovj(^gtR*T1_b*a<9_@!|zDxb!3(LQ%r$WYh9JSSG%(3jUaOuLHD3FB+m
z!H`H)eo}FMkc(mTWy>ZlkCW6Z$E4izmQ5(HQ+|s^b4@<J@TPO+d&~8P*!<)SgsLnB
zdmOa#*nFLxB}DtFgZWRNSOziiaNNqhmXY=_3DuLLsceKQH5L0q0}b#)(}*M&ib^ka
zK(ZcbW{Ra9pMkEGqf)H*%<n95(hMzs@3I-HM9kRa77y_b(YK5Q%KYMY!sz3uJ>PA)
z!P7#iTru5(pfJn{U)wb!zm}axTpR8=GUEHDu!Al8wa2@=S3^vJ@?qLh=C8<_EO0f|
zX`RwKnl0|&wNUa=THLb?pmU<K{OLW{$g?wY?Y1Fd1>%5c89FXw7wgRE6*5_EXz+_<
z9HzEQHRt83$1(-#+#ef%t#t(A;mUwoK{uOwdVZl_d<$h(`NYFQtI@#3e5|1Sp_%yk
zb;j^~d>=n28y0)3yfoc>QX|7^pC_Fn@L(hW{ZOkZN>-@FUFE{CyRY`;o$7ySFx{3;
zrn(OD;kx*UeSs`d9@mWltLWD)oOdf~Mkoc#F2`=5qU9f%M!(LxO06Z726-{UBh+Jx
z^77*|v-TaqcGE@nNHM3H;|W&#@et>6hjSZ^t2IsXgl3!vmP2nmi=UsmzI8gIEvsA+
zv6}7xLRysXl07WTs;?CHW^n1K*4lf}bX${2LQhJOvhB1pTD~x`<BunMDcyqEIW_5;
zHjTxo&DyDU1+lKz(Jlg7j_i9a*@j(J_$VvC9H9}8vTgT6p@ET_MYp{je^S(dlBfl3
z)saf!rL0A~M3TLhDkWRQh>y>qc>fpUH5vgy{09}O6x@3Ri)+Xlo)pDn6oQ(kqiTD?
zjBmKgJ}%q((#<77q~>xpq054+b|%zcV(&xw6<}mx_-|Rb9wn!*j@^r)H4>PzyZB*$
ziEr1GS9@Lz+l|QT62B_-*VIdcb(yBPTTi*#C&Jd#vb40O$3+R<c7^PFRn9bPq)|Z<
zx>M7e;`)xKflG0vEunl=Rwc=r@B6MPT&-jko*a+48);VYuB!6wPsyRAf*+VUo&0h#
z*|QSNtC;q@-!bbpUYp`MbZ-+T&M98FioF!8;cDAyh((m$P>U?<(>g89Mycz17<*p%
z<|D5O4D2VhxbN;{l~BumcH4BR<EZT@SZo>yD=_9Y{Z*a6stR3UA{MnS(Qs?+xPSJ;
zm=CmY3FZ{t(sYsDU&Bsxt8KPUU2CrvxIB5lzD;$@JJ?*9&)s&l5H-j-VqwYFSooUX
zf|}`B1hP2-L*MckS(0Fj@hysyA^h*YS9G({B~3h;jw4b_v&ktaSS{kGw_950<RCqa
zY%1KUIis|Aq~j>2`&44LD2cUViLa(^K#?ef|CncJxs9HW*NKrY%jbYo4>W!sk%z(g
z3uby0#}X$@#6~GBlxK`!My4Y79d_Rchw@IZ6z1A~O2DrR+h-z;G%@15%uxvk7aKkw
zdh$dqfFis(aG1?r8~(G!FjR0o_fvhBXHG1eI?MW|Kr7>gdv+>M6wThfepA9|Eo!IT
z%|+S?n-Ja*_~egBjPQ_wdNyVhT28Sgl*VHHLvo~0?#srJXJz>}UK)mwCMAYa8$~~B
zp=^6Cqw*I7_|~D^%{3IzKGG#(t@D8ZlWCQl7f-)8^>H47P^--lYg3oD8Q&13m77>f
zE$xCQOy9#GghxB6D7g)xItZPoS;aOszr)|c#z<&a1c#U92j5b?|8+fAPB@mfvb?ja
zXPG!By|q|rac~f}sJhRzFJEMz{lwM^Q*+u{)P8|Z7F9g&E2Iw1$<N8}=--agpK=5W
zdSxD%<>9bG{-1VM<D8$Nm9S2@X}5*0(#AX!l$$(+A`bUWSWw<3-&U4qnvJiRE2abf
z`=27m<MZ`dKIKBjn4bHD`)-*j1wWA|yiQm_31|IaUe4O{?LwrJd9>E15USpMD-)x;
zh|5L8$J8*yeezsj^5`s}<FHaZnK8w9jCGe}onkm{E`306J>7<}XYl^KVaJCh4PWG%
ze~xWVw&2Qs%md<nX?XDsxcbp|x<UL(ioRa*{;QV)+h1vZt)9ha55)8?D7TtR@5j&k
z2t1!m6%X$*@lAJZpe&@Uteoz$pg4<*eaSp%gL!e2HyHjUNPj(TiZ`}|Hbo0|a^AFz
zbk_p-AXMM@EE@_pkqgylaNGaf=8J8_?_h!;95Q39LTNP8snh*Xyu9Kv9y24*6l8xP
z(;c0wZoE*~t&FUY2UtTVdV{4wSi|8wK!xS0T)aSs=%>q|oiHBE(GVE*9zWgsqPbx5
zn>pJ;ph`t$G-}f3GRZ*4vlhLiJ3m0l1zn?Pp{JuKXE@__mW&Op>P(!lPPMg%4_yln
zGXdn?j*J}J=zOZ631{S6JuAF;HUdq1oGZ6c@qyO4gO<!`>w@oXpAa@qhE<Way{a0G
z4?N`;3fHNF0u0(Ncr<#6;tMl$f9kAGYDEcUPiD@Jaa(w_EofRjlufxFTrg)wF&Y4)
zE#QFF0D?#&qSG}iB{))ih%A2ZR6ILo!iEq(F`4LuHtEh<R9l^>4=jm4{o$b1P^ku~
zwJfVhR!E<0yE+(rBH=`^=M3wDLO_e~x|zK^AWV-|JZlK*ZX7fEQ5RUSD~UQ(p+AX!
z+tC&u{gjcKTU@X5af**Mu2}Bm`cjLeMh3ei7|D*~VU{1(`9lT^Nz>&7EdYk$s4ltD
z!wA(cy+@3lj97h#efQFae~^RX-a&F-o6T}oPOFrmpo?(z6;Og*vz8`qI>q26DL4g6
z6)_ar2;LWR2t>f;ZIOGGC+S~x7~m;aq3a(MCvew4KFyM%#(cmJF`D@e5V_L_Ri<{i
z_#Vcm8W;te(t*tlxKeMbz0&CKL|mg{mSjES8LHn73najgBEZ9A!TZIb_mlM|4yA}x
zRx!16FegC)UhsrZT4Jrgj9e7bp{N+shwz5R$9`9g#z!CTVY@L~WMX`I*wf^7lOf1E
zq!}s<Axm$=?QkYLkE`-eg5nEsJmHyP+QKtp-Qi(0c-*l$>%3FoLNg^tN!4Bp?|Sof
zMm0{3jG4j=o*G$@OocdDUn1jpMrzyugBG&3NG+(nwOSxci7z{q)YPycytG&}O%`sr
z%WGohi=h`$BBO2m;dU^M4*`us;W(MG&xfF9kumJxLg-l^^UX1hZ9M~p%BDv|8M<SH
z8(4D^WP#j?hPIq)Q!8&9Twi?+iI5n(=ZD0jXT?9x$XPDTalTWk;S(-3ZZmis<`O&@
zIRgK=G=E1ZvlExhN=PZie+Kgca*}aTsQg&MwE9Ppis;Fi;X!oBzWqvmeOB5iT39c|
z<@mn+6on^JWHY?xzSZ<;th|zai<x+}x{gCg`QDJ*h5AwXVnud*HPP|(-HRB{Y3sb|
zUFNfKB^<wtuWP2=72BC?Of~jOmA&p~8+Z3Y?mY1qJE>aXA606(FLAk@5pUF^WJ>ru
zrqs`Oe51ga$}oR>#!!dG;?i|-I@f!Ym1<b<>lMM!NPlR2lOM^XJAU^Zg492U|4~&J
zO^WOJ=;Db!SCLSPIFCzlH0sj-vgeV=#o2aNN~w9-2m)r66AuLWa`Ky<oh&y@7xQJU
z-)<De^G7A><yFt(IFApBSX-BjSV#U~{U)a*afm3N`TVx4Ged`uv^L%>=jcf+9bUKi
zQqIsLdy>M-77HP)iXRC+Tr!$<XT7>~R)n|2i43ygWvi6z3DeXC>IwKzfB(xP68pVc
z8$O#**!Y7?rj!?n$GP!&o)U%YH{|23oTeH$3XY|$HbYdA4FM9DW*<&AX#L~W<!V<?
zKl~cHEEigsAZ?I;N89=kLM`m@E#Wu!WvZ9_QL{pcvf{1?{FN}bhyL=lX;+N$=&lEC
z&MRKLKBmgUWB5&my2P_B_4~zbKNP&4@xi%p*ysISp4Co@EfS1{r%=cjV)8zu9{K65
z@Qmhh_dU<Hn6?~NbDpPo<j{iI4!2N;^s1)!wB^YUBXx~rl6|!rMtY8CG&g{db9dbR
zt-@99p2B$dLS-+5(H<_nI=UcyBTWQrglQq-D1fe6Akt(}LSeHV_vob}A`)?xDN&6#
zHy}%^;sgpK${)__idpgS+N7YL^m9n9cXy6rbJck|GAt?Gb%NXr(jZv!1nYP}AR=#2
z7c-&5jHQTVjM)((YH0aPXh1f6#yzD@Xi*Cl_Khj(-a+K<d?bu&A!vuqvna_S^+jKe
z#<%7+#2M==qe)n<$}MxL!y=vA6S;#Fi`i{#93E*CQeWTQ{pt1boEiu5of`_gNE(F}
z<A;KMBA<O9D0;BQFgP+fxGCU~rS#qSoUA2jX@o0=Th|B++ViBF#b%#x&}uPuaO=%R
zXDqJyf=<U=7LQeWf`}mpidtL@_i&}@De2cDmNnn>mpV<Iu*In&_BdsMVkv}2soKOw
z64!w2v(f|V6~qXPs48APpT}=kGOe2*+Ef4Gn&w%@JSyM+tTP7JR+&rXQNNQpX_59S
z{VBug^Hu+Yg7z<)A~;(sUMeh(iG(Im5B47D`nTcYFE79DoqWiCJD0S`@%Ea&WZ*mJ
zg34D>f^h7OM+B&HD6-*qXNH33ofh@^xcwN4#*RaEo>NpPb)(dx<-^D7LppW8vtl@Z
z4F@v9O$angtBV9r-9}xT?K|2`lhXbn4OY!IuJrLBZ~2%cLoqy**Xp(=q4WhecaPr0
zp?67-Xs5qP`#)2o_C6`iBjeTy4579juHF>BBS~3p!w~+7mU}h+GoaEVco=Or^Bpy(
zJk(Z^ssL0!8sqRONJnW+$;5xEL=Z}#<hZSo{7bBS{<E6-^Opr2amaA8p{};VSe3qf
zWg?VFcxyT5q%4?6`^hR28WRx`r~5N+X=y3bzsqU4w8}bAh|OLn?;tIv8+Y=Kg;bH|
z*o`$p%R^~e1Mz3@0-eUs*61M>s_c-M6=$(I*OYy+!M;ulD8vq_Bpv@J`3ZyMS5|7t
z%1+#5O((PWO17Kh6qR#Ug+WhMZ{7n9NcWfamxD{|wl1=qcDsg(b^M`l<?nZbnlR7&
z=YkW~aq)sxJUA)QE$FM9*r!?YyL0b`4lcD2mmRIDi=+uOyLHTOx6J&He^wgo;db{x
z`3j{#zH1GBEatbj{J2kEzc!rZXP)mArnaCJSv7Tu<_%lCqCv>bHB-^R`RjPvc(w<a
zI2L#w4tWQd&-Iu^hOEk7UG}#Za<_a;k|8B))0l7K!;;c<y){4gHNUW~J(Lwkv%gJD
zA^B#}b-a0vo2BF<Rt*i$+`MIoge?iYZ$398;J_Jxf^<^vf1dx=SM;Wh+1=RClt&}&
ziW#Po8Qy8!a-&`J0s0Kng;H)Ot&L1OzwjOMG=tj|f>`BNZ+PMnvrcM44QB}Jd1%V~
zA6MzLy^`w}NndLoZ?nnehH&(k=snkLF~G%VR#y1=d9=iZ=pa@`FzS}V&^Gt%ICij)
zM{0ha`=r5bobo94U_m?=jzodVsVH;9c@m7)QkAj$7xY>@cA1{L$!29Geae~1O`AUO
zGqgzI91f<i4#&3g6W100Z`tR4>Lo1U&8zI=YfF^Iv}YBds4`fto|8tAa?lV3|N0Q0
zlh^atJ#bPSM9>}S{3_Qpp))f{G?!eU{;*@gbuq6Q7J0~DV}h|Z!$pt+KP_JoGq&OL
zwKD7xKd3tST}@1g^A;h8W%9Wm7S_0ztS;$y^fgT4YTbTsFr())&;sd&UCmv_<8+(h
zbPE^mwF&z2d$ipU-OC^z6Ld{57D%X#zlyERDVr*pI@{hAoXJo^z#a2mTu}@3071>&
zgUjO-0=C7XB_6-d8}o*P0kK_fbJNbrSKM`p=42Mz`QZm6Mko5WM|4;_KA3(wlx@hd
zM;YLt9!Yn}OSoS#?XAiq<9BRoO<7Gl47<xs1v_)L`4ars3j$P;{ehJHv-f<wnQ;WN
z^U8x5HbOpv=2}}MA|(?eo|$I3kcYGWDT{o!wBnP{OyqZRGm+v`@!8a#Cu;AwZYn+S
zA=|u1bgDGlxQuY;nc8l}d>nHg?ZFHurmb79q%d4Ts<icQF%=qon?}`@)$c{DMBs;a
zJrm)jKbaHkxzYH|wIexwr6~jRM5Y}>o_S&L+PuZQj!_iHiQt8(#!FgMLO`VJ6N;bh
z44u%2aEWP6C<D=^9OscOJwA`_Z#Ok?107LvIS-8%kP-GpFXp*151=|zsbQ0KP(cRy
z*$ou#VpNf-k+M;YLn`M@E-N454$;U8DJp#;^=wgh(Hf#xr2#&|GMc-s=aGc3&Smuw
z_pEeG(uuZFvX}{+1&ej_@get{niJz4UXe#khTq|FSqSqY;Y1nu_d>W4RGU$9$&zm_
z5+j=*kKKdg;8+}Ve6Z*f*%oyg<`wjR0OzpnEloXt_oO?pZ)>sL{&Xze>LG{4sfvx+
z^KnUHFKh1;k)=M>y6kG@B2o7$MC-DS`}pkW;Nw-^rEjI!-NUPAD}3=HpU7547TTc!
z*1k?^NS$Mcs0)slEDh%;O9Q1J4PR)OTv8eZzk#uh>z<fKop1>I?Vt)SR7b@NGg`TO
zl4>ZmP2Dg~P+Y0Q6(vupe5ExtOVwP1lDU8$lnKR2P9ac(nmTIzp>BK@_>+k2PZ5=y
z5RCTQi?^(EXSNbMNY1>GFEkQO@ykmFvjrV|_xGN;$;qbY%}|6NFfG!wZM8kp#26(L
zO0BLd>zg{SQAlaut5&=Q3l(|}{T92A3&kD6H(pfcSpY$sd(+T|e&Jb7bsm=X?L8s~
z8c`+?vk&5vZToDa9B%*8R`F{iO~atK=h=VjQ6qoaEA5NpbujOdlCm#)ON@?pe`n|>
znbPMjJmaN5!cOe{TLKN&D56fFRO!{GyKjz~<IucY@fR|S`j=A$DL}izq$G^-gpJLV
z45l_D27a|+67hc6N7?;{Gh#9hK*mXXRfgb#cTQI2uR-CCB4XW`6(u9QmMGc9`W3f7
zI$L}0N5bX~HEFa_J7Kom4?45vLRG%%Ct5P6hnY6^f^lHF%fVhAxZp0KtM$IMWd&jD
zp&w(P4F7cyo`0IK?fc+47-Ft#G|r_FyEy8nf2GG1i<bo5#$?o&>Z)O2Ah4x#!HFrB
zBnh|y$=xqc&^936YNX$aHTqW_mSO+h1Yj~4Vv{&)Md-3LxAC$IPs~^mElmFzv_{5-
z7Td;dPZE>_y~2J+0SQ9j1xUJMR*4x#z7Vjb5_nJ0EKBO!jh6dQ$`+0KZ2c4xMxA!%
zF3RvbUiLAzE3Y>g-gHWz<&jSl05&i~CK<X<QpJEpnq^qH$DhmUmQT&%h;f*@rQGpf
zEzeTjNb1Jtwr3jAvgdPE{?lMeq1)bt&u(;bsmMiTK12-&SAKMp$FPC@Sa7VOaKB^P
z!wg6c$=a74aEt2`U;wK7jLxQRJ~#YE650VlL^qyQ27ULv8`r5FdLJ*Y7xv_JupkQ?
zqv1@*8`%~4xbXNmck3G7km~+oTzwTanN$YMfy2+E`#ijLHh3|dx6n^gQPNAj7%lMp
z9=R`iF!*Y6tphD_8_TQ1R|#YXo?5SCFJ#{RuU=x3rYzfia&+-P%YA5j^)=?+3!Rlm
z4$I1>KU-Xztt&ScJ%SH(os$L(9-aW@1~u;b+cjwM<&*73;x2uE8WTV}81C2-zUgg3
zf;J=2s;1&G_*f5J0Y5AN6upAoCV=yab&Qr>yNTGFogm3Ddhkh!I3UqJ0jw1SqtDR_
z(~C3ybOLDU#(2X}ArxS-Se+hjygkvzrBMP)NTZAM+#BtQvd@aCfM-h#j!vcO3vPho
zEm(%H1Mt<4?zPS<Ja-%B_-F0vS0ZUl;A!|>-(TU+ji1=JK6ce+{p0u7n~$x*F$%Q9
zyTHwwC_h3gfh!1%8gZ(C@0EhV5P8&VAk3H_I5_a%J(YyTyn#wYYZq7T=_<>lu*GdC
zhaN}K%GZR?;MsCiv{q~iXH^5E6;wM*$~uDezKaG3jtgk{f1Ql=rOl)o16FVC&vih(
z!bb7Y4E7$7ge`Yh-s+{<ab8aIuW)SROnQ7z2U|D}Vr3B~cF1km8k(E=F+ap3h5y1E
z3zt&oVyf2j3Owm$7qpe%F|YCs^oyC=dQ%T0c2XaBK$M}CsK)2rfPk5xi<747%9k$2
zV2=Jp=cNC%EL;}CmJOjaZV90pW^No%6v?F!b$Oq#C(Q7Fu|zb-gmQ@PG?4<y%G#Xs
zCaS;@twk|ERgVCdUYeyMs0uJ%n7e}|`5xHH^yD{DMSZ|43aT5{_1suxOpqPW|NbTB
z-Meu`38iV7)XH1v$;mD-sm536zjGAMAf4D}wRlZ<lmTs!$1rlB{KqaY9vfJ{kl6$M
zD`Uzl7$dL;2(u>%YbC$-KQ%P@>sGzTz^~Ov2=O5KW6cqupf}$Zv6$vh`%ZA^u@6o>
z+FD_9&Ed~m^<Q_y|F=Qnz7hJ-w|_xa{XPFQ2Tk<n3uLlB{=Lut|9}7g&GWCjGTh>*
zyLUgYRvY`RQAG134Vm-xTcS_#HaAPkEW)v!oe4QMZ%xo(%6o;&2nMfn5}imdZ&gJj
zO)OkptCptX%WGYe-72#xD=JoVH}hU)DJG#KHack6D1ZP5FF#au`d_}`e|<jm9>^@A
zxx&VO&xbCRixlmx^}F6@l33!rH}>Sm`hQQu|C8^)!zF7yE+#;{o9O>kItY3HyUu$P
z*tb}iV?>qx@BFO(wI7S9yn#grijaag^ULTvaG7Yg8g*UxuYOEc+rRa=fZLNG=Ed2O
zGg{|TzVH|D%>o+o<+`31#b2%|&;Czt^#AfRz1G+*P(uPiy*jTLlk6`zUJts;Z~gB)
z&i=8Q=odo#&kIqutLpH#?$Em%{X%r-iY)xi7lN1+kgMqO-0uKOufO5jnQD*r*tK6%
z6#9D)V7U>U5B%A6W&Y;pwa_7AqT&?*e}Cr&PXE|4&UC2v{k_i%$D#pU>1LGu-&?-_
z_shQwI5Jn~OA?zvQaLZ+OtAJOgbJ8hu(z}ep8y<Cmf~@Gf9>_mh=yUv0<ectKb~Ao
zLR(DTd-n2BAS!Zqk%U!*prB#IFl?iR#(5&=l`!)^(AQ(^cjj_)vqg<ZR{*WNZ}ca2
z23C3wy6}VOj0eeP>46gnzV2ma;ED05QoH=h^XkYmwAQuZBdMZ>l0&nVG$x(LaoQZQ
z(0ky0;V_`U_kwD%AfE$(Z|4wgjhP>^3s*OuEqeS&SbONK@m%>o7C*HW?++{nSN{F2
zzV97sVJ{PD-POkI+Nt${{`g^M`)Nfmgg8+D+O6~39$HEILG^yW&Qgc-0}BOSak4Ew
z*?75el?N#^=V*z_qq#5%zZBI{F`lpE3L-&yZf)TV)<81e0bo~?U%RXy{vGa$K1+sC
zLZ-k(%_cB<`KIyJwuPc~1DszRsW8O1>#%fsbE6)$c>3TFuJU*vM!iz}&WhRu9U<$2
zcIND;h0i^Nw4x*UpEb!jiU(g>qs;|{zZm`_0?|B4jd6O&No4iz>hc1f5Tr6nz=!$`
zCbPvoIYNhjT%f_~By1^Z69CNvSufD!`eNgQO91^1oQ^n4C+s@Q0mEs4{o9*p|0%_z
zK$hh90Cj3;kB+#^g-lM^_h=9E2Th1$>%b3GodI&IES>fkXV)eNQ`KDK-MpO~8<FkI
zc6Q(}!*A<&_~n%XkMB5WUKJ>n)c08<2LvQ@IQ|vK!G9h~{sy?v5%}L#4ws@$W~{;Q
zw4smgi`_34uR2}JSgS#<JG!CQp~*3U=xcxF$>|!h64?C>gX@QHg8Om;lk72Wz)$D~
zj#qw~Zutdpbghze3<vLC4KRvWW4n{ROx+Sls=)0{1w;g^Jg9t{<4r!dE0uH^VmcOX
zu(o(2gfcp1TE##apKL)_;@{ydHS79580rQlTyuXER)J5|0K>BJHIrwe$G?57fem7}
za1x9l>}fW#ros&v81ePYT@cmN&8&@CAUAlpyhPT2`lNWD^viRVlQ79f&4<L_gG$bm
zD2nbp0qd|-4p?8&cl{x*rz=^p)Wd(J?rug&5m%rf3u{aVwrH7_iz**~+gc8OQ?Xv|
zO(XJc1Hy^x(M%{jZE&uFSB^g4`U#at0>ZbOUlXj^T$Z&ha*L~mYr!RrEy2M!t8<a=
zT27(jXo17&b%K3l+3XEDcv^trR|_@0PYoaNIhVfzx60D1%QLqvLd~7A87NK?w&?gl
zvvThFTkCS;2wr;xPs@9>NruSkw+vxg@}RyPiY|BExWR@KaN?scOho&&p;Kk!O(@r>
zZJcQ`tV7W#aPDBpLvYyKhUXXjKSSnc7RhPsj9QzTla>uIHMp4`%=zw7pULU=uhXZ5
zrazj=oeVu`jFR5a!CpHT7yM%jUWq-P%=Ec`q^LX`Cb}C?x!r(22}}s*4}PrkU8kC>
zK<od7=C|5;-_h)StjODbcN3f?k6dT%Z1==N`({bxsAmo`9|fmzXkuj40{@e#6sg(}
z=Qqsdg+B-)zfx=g3pUtf5PROp;0?a6k_?yMli+|$y)b53*fjw}OxMu1A~}pJasS@H
z`*_~8{E&0twl*n82cP#%lekz>GRDP=e_-24C^%Rpy2;lgI;Zx={7ZrS`n7ue9y#d5
zLh-tFG36;J_fG<tvB(KxiNZC7WE?Mo610z~j-JwzJdT$+<LLu*TugfACE)wzxyBf}
zKj5tEz9aMOv9cdf+w`<`p~{%7UWj=+0m7(CfIlOFVFDJhXfIDEDVKLH9x(DlHex--
zd!;D!rSUD7T^-QUB_|T88WabvgknPi<kl0?BKi~9bP$Ln<k^$QuQ{$grs|saN1t`q
zjeNl{f0Grkm+~+-ijn(DJI<rm<^njuuJ@>W^mOrmbzJTBNVqjKVZFEgE#!GPYC0!P
zQ{MBEH-Bb3Gg5IC?Lk;OcwM&D-Cmj2)EDjB;&J~ybho<}teK9tycXiwjzE8C+W1-#
zO+35Oe}blB620_3bl01SFMb`k%Izm#SyWa$Qp<oJzk_GaRu&B-t=Ev!rIBkHqgfem
zMk}jrvMx`TS-!n>;LebL=t}G(#O%+vath?}cPg41i9}Fcz(!7I>BH0khzR=?BKr|5
zb(xC$0v0*>eun%wEeeKWO2QuC^nAL)0-S`lPHMriw~MDF;ZAQvUO@-jRKg4>=2X`+
z$%a+6V(*UwPG0u7LGXhr7Ok;5St;5LR#(?C1M2bAJ+wUfn5%&^Q=&u2Ki?5aToUf#
zp6XkcYEKN}c&{uMpBeag@h@~@Pc?n4M>$BHf|N0B@jP2Kg4W}B26Y~fp_f<s5Xp9Z
zU8=ls2-sOxhutFm6A_lRF^k}!GU3#Il)P1QzdVT7^m$5Cl;l0zE_poz9GHyl-l_ic
z;ijrhvPt_<<Jg0C;BPo3)7rJCy$e#(qT^bL;?ie}y!&i!W?7@3OMjODAIXelz-$h0
zin0v6KQks?3p4#0vzws4s-U^<TzjTHHM$bLkFwXDp%7sD)l@~vEYMjVPM%P${fJIt
zdi@Gmm5S58efy4m)Nxpw^p)t@F+lWg3?WYAD~G~Xf>Rw~|9lGI%q(ezjbeIfpc5W6
zzx7H<thgljRX*iL<8Z&+^T{7po=&cwG(RU_(;bMj7DrW&pE<1}!)Ptm$!HXY+_pin
zh`A&^=<3j>s89C#^j40t_4zMm8)gA>{<TlE5`f?ld#6}GpZ5C2OKMb|*yB>_R{iN1
z?8A!Zd>n|3;i42%kyCK7EP&7axo6tO^K86{xq2uHSLUx{78+PrUFY^mSN{Nm>^Fp$
zTXK?&-JC#GD!e!;`0|nTtsO^wIKzYUopDjBJ2=!9&wpNvK+VaUOzjFl?yfB!R?O>w
z+Ti0aY|qfBZrK<Um&QREER&NB_={}qqqJ`85fsiKB5(vIv|{Y-tJ+Dty&Pt-(wPhD
zj?ZeqzGtb62R$)vXQKwYc593hLpG}f@wC1y7YT}wuPy<D<zD{ifNzBW7cB5WbN|r6
zTZ_dw%Uj7aG)^N7`S&z}ICGmJ+*;h0BR`vB-tDaH(3matb~%o9Pecp;2hQ}~#RdUQ
zKthPUZOM{cL(dyKzIv+nDP2vn_{-{(m5}4430;5n(qD37J(6O^TZ=VGr)yVykacDn
zg;TLQ^BXP3?ps!1)(qHvJ(s(GPBe$KrK-QYCC{M}Ffc9>T~n!PH41QyM(?xc1u5KC
z>|<By<PFv%6Y3FBl2g%3K+fT^NYu0uzoYwa^tH;+Yk}icBAQ12aY)Plg-s?ozUBJd
z^?7tj=}ioM+t^FL_=aha>$Lbm;nW4>U*5f;8(4MM$hCg{dFqaY?C?KN9fn!*<S}^Y
zhos1|mG>LP!zL|Bz>$IUE#qC9>Xqp<tI%<$+jJj+V8UG}ZNCm0!?lByKCpiwd=EqP
zR;$7+{B1*eZtxBWv@9hK4h&cDF38X1-CD3F?P!99<<EMYzz2m%+u|erFRip|i%qOS
z)##E!vgT{zigFkR{qPhBo*uoQ0!25dE6Nqe<12Unfo(zU7W@l;0YQ#wW@G-&1zrF)
zJvHAN`eaYH5B4M2CMv8H9ql}60w&ey4n&~`QPm0P4d9$Y#Oy)5y<G+Z(w%*~WnXq*
z?n}xw+EKKm1Dp0x#uB%g!V=ecH!tK#DVf?Qyq$(akcyD(Td>lM32H;57K3)ujUABp
z`qsM5S3jh;{VULFKTaJcgs5<HfU;)%2ijy4wrj%t9FBGLcRRzb>2pmHLAg_3-V(D@
z#PJgQpecwhlX;8Fq|##50W8H{7;@fa<MC>?o}D-VD$#7kb5;)t-@gI^^2~syd?<hO
z=+2gBROKt&+i~?lAGq@dYQUlG>4~OCfcLc+gU1*n4`x96IO@UR>n@b42K1eu@iZYk
zc9s6txdTp1dFO%uQDcj`RX5OwO{$<gD2IQ+U$r$XQlB#=?#=|y?0;D_9HF>@l+LGG
z%NWYah$wKj_zS5wZDEkIHohZONIwU$V|L))*#ZG%+#F`K-<Y=X+@_EKa{n$Z?SSn`
zH89hS<^JYCzvc9BVhzMg`LBQmlp?2=C(mU}J!{VXXXXB^UoImz&h-SZ!MPa##BEF5
z(?mcKJ&9slD53i0C)~GWs7cSOi*D@kPxjPhpzb>XVAv{JD%Tn9)Cm$=<=tcoNCg`I
zQHK#DDET|!04Z(;bJc8n3Mj8BJhc~~UqMBe^ypGcSfp5>GE%P2z-s~wgf1o@4k*l_
z-i8*&b6A6-DO)+>7<m7wx9_~K!wE?$-#I)frMPeX=^3*v^TP_XS?CL~gF%KjeNJh<
zWOiU|!2xBGR0f<*tCxW&(Vs`_Am9!B2ZrKF^qw)Guu>~`JJ^Fm$L+*xsyT$sB+Qsz
zJYKduG&+4f<1A)HV)Y%i<%q7n^C7W(AC72(dALa+^s^B5f~Wa*H-Se`_g+TC;YbYq
z!G`5C11G&i+ZJ1J&vcPA6Tt9Qdr&q3S86)>D>6H)YQ-Y^LSW%j;k|g?j8RAk70AIi
zmcGRTbMm5X5kQV=xIRg3kD6hvy#L8v^sUs$C?wox16c8{Gk-QqKmL>sPmr9keBan9
zYT~u9z%h?+MBDEHAb%pv)B#<qJ{zCDT<iw=5J$k1>h>3XdA~<*hf3ZPbhoj!iE5_B
zR2&09NCWG<(Iy_%e*2(?ws7`6L3y@rt;d4m+1d+=r3LYMem=<qUMgNPW648&Z{XSL
zvD#_E<lh-w;%(Dczbwbp<H4R1!D=B>IoJby#plqzucd%rMwT>(+w-f=8%sXa9+Ip%
z{TT$p5jZfpLL*zY^1Lb6o`iNF`(W{7plT}tX9q0D$<ih8ku9CmUYcA=noUZvu2`7>
z2F~j(z#KBY6O_uT(x*@Vv+mw2hD{{4uj~vu8>VD<<>BHq7Y*;yz-Fk1-!&R?_*UGL
z<g6=^6t=&mxP~0?L}rPb#2b^zT`{u{_UkTHs2z>2Y-QK#XlF&L`cqpRT?)fcg<BJ`
zr)$fbL%N>uBk*v1`ixgRE&`Wzp`IZ`C)#`v2V(FSPu~M;kUmr2glcd;=vMoHFA-c~
z0?|kxe^MJf#b;6F4BS%f7q9(f4S6oH`%bQKQ{V3x&BEQw0@ilANcFCutiO(>=m6Av
z1Jg<$QU+@5LO7pRp0s#(FQLsaaMR}I*2h2j5<TwFs;U&(ENjdKuGKPFOAH?Ouq6Ar
zx7;^@W!OH}5WxiRw-$O)ta^t`AbX-?qZgpBUA_${r{%z{Y(VUEIb##pF70CZYALmT
z#&Gzk(Gf=kwm*fafX$0dUvi!86X}4KvOZCMSccTDL1zc}659M2C_fQypzH3EN4LJU
zLbk)DiEKSD_fq8bEWv4e6D$XZP`9`27-N*Yz}Z+eur??9;z7-=mG$B-N6qmpk&HZE
zS5YhLd7Sb#SXIHh>LUhtPJns+=a3q^2aLoHW&Mi!*mHh<5CNAsi9U|c3BH_a2zfX0
z-Bi+BR2Ot);_oLM8J2>CbH=NmYQkrtt&GqoC|doI?TYmPk=2i3{D0Vc>!_-`ZVhx3
zQqrJ;v`9A_$xRDL2q;QQx0Hmyrje5Fl9Uhw1nJmxOC#MKD&2k8*7y5f&v(xG?z#8=
zamO9^jNu;$4EEl?wbop7KF>3sDU6YYi$!;YjwZ;+$<2bo-@ym=-e--{jryxgy>j_H
zzb2ci<peg$HZsBh6X;8OfPN%Cm6}k}t7yAzb-eoB2MvGQ!C-!6pcjyHD>rqlzpQnv
z>4q`zm@J*Ns77~Ez}zK$J52CM-*&`xXsEz;{6FQL3rxN6OH`Dbwck&Nt2()Zt91?S
z%DXVxy#8t519`@5G#J;5#(v(e*A+x)5_Q))7Wn-{tG``2|0p;@BpZ|}>C>~dynJ)y
zxp-C!#KP$W7pQeExc<CCAxGhMlv_d<H)iGqw>|q#F<R_Px&98gclzPz+`E#I5h&=f
zeaSx)@R6f?x4z{@=6R2T_MM8z7W6a1?;DP*!=nU8h2<Nim(>Nusu~&bb-ZqG&77ES
zJ$oV9b^7yr;Z@(*hd0W3gN{?17cY0y&1b8)4n_wq3aVD#cuf&UAi>^u5y6ghr4x`M
za!>URV{j#b34q})QLlC!g^}*>dKz!U(ho%XocY*1bxudqK%lxLtu8@gK`O<)A)o%(
z-BBVa6#GBp37@H^U`lvjAuF;PnR3*dHUJ?x>pq{Kp-LQGX=M6>Qq<yKm-RzRcsT*W
zk8-aH7&DM0a2sRaUnn8vQ6w(O&%(1lHC_p4;xVpifsvoM=Rl3y;GKX8O&{}t4_ts@
zNxLn=*1_aF3G6AicFiz+)|ZF}OUP5bg!mT5^Ltug`FFwO;KxSPL>HnRg?O7xX4aYx
zqA;mbLn~^C;==mxJ66J+B(nf#_QchyL?RTYjhEky@-gu!!ifQ34knMLwzCJyEkGTi
zd{ftHAcjBse1O^^qQ26`u3VIw-&)hCgoOm^#KMvdr%<$xk>I^L-%CsdbGM*Fx2gtq
zG4#r8KZI6(UbxsNeu#|mt0uCd45LYT)&2}`zH*$PTbW?)-Bypgr<Q-h=Qrht>jdH@
z6R@m#QVal_o4MwC^rAH1uWPaPvYmCvy%meju)gUsXsw$huxbJ}A!pUCBN0F`dTLn@
z${5K8b{2PWrjT!a{PdZX)$!P{Eua(3t60mB9)#MF1;an?d-Hc@Mcr(-kqW90Tg^`2
z1z5#_o*AWa7!S_+HCr9K+Ls*)hLb-_W<$(#qN*@lG$-!H3DwG|_u;@t+-H=Y7l6o3
zAviGmMe7oKme;iyu&PuspuUG9%b)9AS2<dHDK46lf9JOMGtaQb?uv0y++j)w!+n!?
zE8vbQ){pgun||+KQr`Xt88zdVrj;%QlOSZPeRliaj7ym^lD0JN68b~t;$UH?$U!(Q
zNsIohc3PjQ4}O?-_SoAK+{NstBPrkFK1NEAc|ZJdeOD8;QVVQck6J}Y39dsm?A)wV
zSLX)DU7eRH<|1!iLhIt~#ri<9n{V+K+WIS_;h^&wjn%`Jgy3QLI2GHl%-v~Z#<DAC
z*OaFwKVLv_*!<`lwt@({ZqK!t>631ncbPCa@k>_%A{=M6lAw)bg@ZfwP-l-aHtK30
zf&eBV*0Q=rP<!IN^Ht1FV_&YH=H!s9%3z@+7a}{-4!j;Td84k`8@r}mG>N#IU67*p
zA^XzEr&XOGAq0$C-|M+=+$(%i0}x2!J=CWb0v6ugi@HXNu8hp|)+;WI8M2di!k%(h
z>VR7<WV6%Ll0v9m-`X`v&OK}a+xWb{E|4S8mdt6qxSYY}o;`BvjQ*u;R|<AvQ==OO
z>4P@29L5#X!klLvsOLQ726<`D3gFMnQ+`cN(20QUb~NQ+E>V*P&Xc=>;p@aYoY}M`
z1~5L84nlo>LsY4`ug+jFJjYO8&m5;xj72%8>KIQWecdgP`4Vz==r6~4X<_*E0gEjo
z>X_@6Mo-}2kHkEAHO~;zbnJzY_Lu!?mwt_8&si&u+7djLm7Slf7h%V*1a=CeQLfo0
zn%A!O<ctXSoYmouy&3JwGVer$xJ&CKc>)|Xd8-coK*d_ibk>eG{SWjQ+*E{_S5o3p
zb)247IuG->gJGNBiX!!@wcl3|%s!wAhVf0T+Ln836`b#!fbA7~<o1J}ART@<&Fpvj
zqwI&%+^~{G(53K6AIJi3o&HeTe5m4PMr>O~*rxx@Mw%MK2fkL);az~?vxb5cv9zfF
zWFCxG<#&3mWQ%{=^ZePaEx76vZB_bX+5QG|B~*wkD9|MQ1<g3$U3=_htaK~rt#r_7
z<HUz|Z&;BUk@F93xR5Hw4LmHg`Y~uQ$F)=?;q@Gulapqe%c!K3bt(1j9BJwWShi2q
z&*Cl|`=o-(iRYhgv|{gk!|=gEtnXW6*GXw}hlI0k<6CUjUtb-MMlm45zxc(Mer8bZ
zg%EC)cmqWLBGZQh^U2Vq!Ep<45=Sj#PlQd}*`%4ME-G8LRM<?`a93(7FSY6vtr7@y
zV15NaESjyS0!u*!9KMV34|~dM8LqK8(iT8M_F@hpWHAk08kCWq2p>blYhN0NV2*Je
z?y1nlg<(N-(hD+zfV~%uKFs*#cV9G?==~J61{OW+&u)FzH;sR~pcr~cQ4kvSnuF_^
zL;V;S%q)I+Y%EW?n*hvr!YqZt5bYt7FtR|0FRh;(*-1upymY=-Gm8!?Q9xEsey$#b
zp<nh`Q&6eSrT>mYAJo1^t$Ny=?0)Qv=E!(P9Mt(}KYp!?`R_IG!E(Cg3z*P*hP<~t
zYYOWxCvJgF8+pd}6H2}ba`FOfi2~2#C{0n%9J80WJ%?P<>7Q&w%IcM?zJG{>xvmVm
zsCm~qrzw}MRwV3g4C*r#g2+=-Ti;7&q}l0bC-O;?TUnE%^ecp0#Zw(5XFB`>z_Cn&
zSjTN9vI0O1<??Q=c#Kt`^KJ7U9D#?>9*zjyAIGN=Lp;#%6Qi!p|6>>Lqx+WgKyr#)
zqYUM%N}EuPYcq#vsTG%qG3f&dzw<y^tn^Q*Q|!PI_*nGNh5ALk7kIR!y->0=Wz51?
zxC-rb8@oc!V)-E^*mW;Bt?fgpF%ur*u8!R~Zu<{89kV$wy)w=Y2hP4e%;$^C4|%Mc
z#;BW59b|prr7~ss)JwzEYF(}+p_A=JecdB~HV``jm{BYed`Wif6(UKHHaeNSrRUnn
z<xQ~=Xj5xrSD>OrkA}nENxFNZoXpt^lyd=-M&102VbFw@&_usBT4r{71>v7GAa9w0
z6h}pE6yaiEM0c3b4MCF{?uC!YRO<hHD-<VW19@ExO%1KEL~8)}Uin?>8K#9OGZ^5o
z;C()H%C*bw?vANUhl|y;)3A&Sr^_mR`o0IucF;*P4AR|vkYf!!&U?|dSLaHkA})t0
zv9AQCUJ45)<T~x67Joo3_W}8e3%pmmyp_<geJ`y5A+N9`g)6(As;BtAjDTkFV_xJ?
ze#BdFA$2&G+2kFLvS^4NL}T_LPHjOFwe=xiCQohSq+r>jW`DHPD&2F{cB^m}xF^ZX
zx;WL8`GavaD<Aqknm{V&11a37_zy$?<9!EX?WGnK-YN^dY}os2a&*d{CI*`DtT7kQ
zO2vZYM{)j%NZjcsL7(^vi9&G;yD`{A1QoHZzi{p|27HH?ke->y!W}VxZ9K@jDo0TR
zD#MW=NTH=42XM@255gtsqzvZB3w@V+ZxMf?g6-!l?uY<fu67Kn^13g^0|`suc$0DY
zJ6d;2Q<^}?zV+83aY-jGUk3~y5T^BB1!<cpWZgB=(jcX`pY;7ADtiZ5cxVu1A||r5
zF$i{_M@pkm1CBF`H*h!I;VTPWNCuG`*zb(yX)hD>siB3tH*Hl-nF>*^ndCY(e9~oh
z9NT60_kj^b1aWksFLEq^JmdjDmPKg}03Ea21vTU*lM?mAe)jEwuD|=29xHC&(;n@*
zW8m}>X}5}ZlLxEyHL(z@4Xvi#$tRYKJ+H$O($#Pny^QMZ^8rpY960oYdLD{K`Vq*<
zWE5w#NRRC+LBgc4uub_#AO@e11|`A6q<9dZCsDL6K|UI90bB6-K4;Es1v0;a3lNOv
zRFv({FIqj{pkE$9-WBAlHG^Krl~*e`kemXcV8C3%J8goD>f^eTY`e8bq3-FAINH8C
z$BLs-uFAPrSzBjAYAU+lr27FThd9uR$A!R|2C1)Ji8%s=hnM`^8;SeT+T#(j6+1JR
zU>@cnkiUgx|G6u*Witx^9K%q;|4NnAZdBWfV=Md`fd`fssFb9lCzrsYCle9ZbmJch
zf`g?%Tk_06kft0k6Dg7HG45ZO(q4$A`;rsqw?YUBf|duH)zpp^xzyTu8+M5Gc_Fw%
zWjB!6)g~jj7q~Ypahu_YJX83XjNvZe=59zV58oW-b&b++r{vXF4({1;pex)>P!#t>
z#G~POA2-t|m?Vrr2&)3Lnk4mtD_p}{nzsQj8e0@6CyYGAW#L0I9(^zjk_VBefQ7MF
z{2I2}F34>lo8BjWb#Agf*VqYkGlUtRt+)fO|6WR~r>Pq5vLQIrbCJ&EsvRlvWx{xs
zyNIpZR{4*1f2M=VVqV=VRMPRBv!suA`;YTMi#ja_v0LqM2gY&^uZUzRKZEj^xGJhr
z=`<w#+M?CMh4$&T{kitpYxsUu%7%X1$>-A94;L@LKC$v76ae@B{`jzd9=~i*-H?9X
z&|d*v0b*ZYKO*K0tl#%W%%6g$0iAWwH>mgIumR<Q@wCQGnr(7GkgIfnuFz0Xq~KTD
z`1{C;N@i~0LhJ*=2r1CBi(31`Q6@n}54>}EA5aB>Mbig2LQH8@J1Drxm{yMn0L>6n
z$&S`ddkVyT3cb<zfo|)np(qAI3}{X4=_p+nU*eEF%k_#l6a$n1=U?K*PuN4xAz0|7
ze4*^c56o%;vIPDAK9VzCEMIm4H941_&N!^+MJCqJ>^6g^o*J>~z50Q48bAuXOp@qY
zp{B;cvHVe=%(;D)1RmeApWh=>qOAoMK+wJ9b`dktw==sMYGSJwKUYt}hMF1JKI1|Z
z6My#ZwrO_F;d_Q`+Xij`OkU4lRPk+}PthM2*J}wEB7>JnbJORbsD6N??y~B^D8Rdh
zWvR3RZ8`}e`8S`hH@903O~|p1Vt$sJowmN>?RCk$aDWGxiX8ptBldiqJ|bdzOb;1d
zJ=5L7o(m^Qu=Tkgns{H#`4MoQmhW8-6BqAa^XT(Bs+k>jfYaf}npyf`!GDyn0_&c;
z9w3%6#wrV2K01*;3;Z0wGoda039KHCt03M9?j%4uxgxTDVdcnF322?jeaoScoE%Wl
zYXtlFXO=c%P~5#ZJy0}DdpUvx;Mre2{)93I7}mICFoxpUqd01D;8I@HT?v#HhPwbQ
zGEWx)!{rqQKLFF?XLaRcMxjff24g!5S|xx(*d;%s{epWA(7`p1jzRt0Z;)=r$RILN
zLQ2o(5r-e<^5#=J_n%PPtFJ6y7IUqor;0cdgWM)(qV&pa><E|}jKKTo+z+{xj*Vb1
z1jc*v-3955=wy74Z}HS1&s<?@zB538+A2O(>;H<BU7{}GVrOvHFQ1y;6Tj63x1-tP
z0%f!W*j88OCell8e_GG_OKXf>{l&;E{_;J=>HwF-4Io8Hd+>qsxwOvTT*(?WDlp?T
zdP{QzAVvOh;KZVMsd-y40k2t%dNv4;LY+vK!}D0sSNrV2@~POtzddl60QSup=yaT(
z(H945+0eQTrI`YTC14Uye^b4iHty`5z{kFiW{}rAbjGdA38*t5gF63CQ?_mKQsnk3
zT^HIMs4s8H4g9*bekXaQVG=sGA-u8v3^=c{?$q{Zn^mZzH8_+!;RL=N#2s4Hkj^du
zYb4CQ5>Dgt0>YMchV>yM$T2XaHGkJ0({X?fhvWw(EQBEOu2tgqqYy&l-F)XzW3Kx1
zV$5g_ohMy$l6o8V?;B;~#^CV+OK&I2D=<krXk=wvPf`FN(&dJB|BEvqgl~qZ^%<bv
zy7%~2K+3o>wI6oQ5x5Xc#r6)qsvjiOzu6VqG8b3LE~%uxJ)VeS;6)8d=IOmpgYXHm
z6HwItHfm7x6U^$C$G|016%67z0=?lYXDs1EX`g>9bN||@`lzn2WEt;og&w_g@b2)J
zbPlUPqC2fk!-GMQphF-sPVQGBow}uPmNucEr&KnAX}>p@ESI5s@$$K`6VR-Iq6?fP
zDm28gJMa*Tjr4lK`g0lhThqsBnI$u58Tej1Pum3{5I7^gQ&QCfH)|*vI<;V^p{c<4
zY;ELH%+fTsMzt5Ux<3A+#&>)5YbB(Ofb#%5Q4C=MW4fzReU8p;Wl%R&M)YbE(^+vJ
z12~$!6Rm`WXbxBgWyd%!ZGl2a*_;C$9K!4-&?iuGw`baUvfys50wxe92S@ZGDG*TK
zft86$4wPJvn#pRQTl(i^^}Ljb!C7w^5B|P<qX*?0BEaw*DLY~Wf+;0e>T-ZI6$kn-
zAHPEVj|ax?czZs%_=fD`1An(J{T>MSmbpHL4`&qs7(+ht*)a}(=bM6UoeeTI^%JOI
z&DVqv(45RunKlgp-FrMckn49!Vp7E+0838CHi6tb|M=&G?Th5GJAqa|5va5TI|2Hg
z1Uvg#VtU1mi|b&F(Dm49y`G*^QZxh>tnkzSaKRosiVo|8i4KkZB5OZn#U*py;bahM
zaL?8~d>c-?RM7Xa6%uuZPYn|^TtBYVy?(YPOXgib>Xt<nzSZ{?kIvih3@|XQkN!jz
zIq3U;fI2cuYaU6$QI_h0zzw}N1yVs|=#<}et;K$&F}9!}LWOKyh;aqgMrvdrkpbXg
z6nht)g{hu{YNeTbIZiDhM^2Btv=3)#zHC&FO_&R<f&KAscx75HQxB~Hsl3Pue~XuD
z6vBdoh5|49Rb>oN2feYmADpHEm!|$`<L)SdB16)3Uld5$1FQkFlGq6hmotFkgdZFF
z+E1w^A@@FjcHL<zK;+}J$l=6EE>6pD^ar;af5d`mqhP(EI6pd@Mtz44zyswUxy5@8
z>0D}uweETx?(qwMYRQ_YLWn~6A8#%3*fEw3S+Z@0I$VI3i$e_W`jZ~Jb7uf1?JHK(
zIFA{2K9Zw~PX$0#ryqe>(m99~!yn~V{}k;AM?q%7*S6R}qz9hV`B_t7y^|Fgc-POt
z71>+eLh}if5yap)KKZPo9$ANkw*Fa3{4<~AI<kIWzll?wd@!OD$=bzHs_31Dx;$>a
zb$EFSEQ$4987;1BmeH^MU=^k8xeck?A5`>OpgV#Ar0;sgSGA0%pHP5zuIn~i1?l#d
z_m#Uic4buOkg`f?_6`S+Ag-H2IcW0H>QTJWzW!&PdHWa93=q+nTd#^?D~|}x<ma@7
zZ^o&r12vmy^0G-kTLi`7&R8NoyVb4UpbgHNpD_^$w1R<qhM`4dQX~0aPk<d*-zB#F
z*o;FL2g7<4b=qe^aGTs@*kOjYMXmevcjc>)^6mU2B8R+_U$6L8V=$rMic3wB+EoAu
zBL>y@ZsY=Y#I3_-_TCqdsBIpN%%P-vexoZqOmYIr%p=C9IPkcN$EQ10$lARG|NV8?
zo@oFgE(_2rDfKQ1=7ywdFcq?kfYq!wD085B<8j9(eI#rlJkGTj3zU-&UI3?FmL>Y7
zRgjTj<z3w53Y>q2s0EWjLl$hR*WV@YM2VM>z9V~faZu=epcv-wUyPx~8ee~N*)hG(
zOYc`rsYBXg@7Z!De!eZ;nzQ#;aY@SAcjg5sJ#3P=?L*Xd;QuHL2P9N3-{)h}^_)M#
zz%7*hMb|B9Y8Xh>Ki;$j#akx`2T3fkg=Hh!rcB{O6F~&z??IVq{VQDmb94M%W=49k
zx;LR2fE<YB<?jkJyGFsi-W~Mx$ce*D50M6t-Q_+!J<olI`;0_)fo4mq@~(>K2*@gV
z_a3CQ=PU~vV38L_5dKI>;tsBmYE%05Hg96!xK7X}c&}Zs<c|z)pVU_m3^E2TCCG`*
zoI2_-?Uq-AfWw1gTFHZyON=>$;xq8k=P|lJAPOW*-tlC&IIGvPWCGo>LZRK<(vRLu
zWk~K4COz%ky*{sJeR2V=z5r(Iqp#l^h>k$y;b<`qk`zri2>E85A-Mw7iUM#z$Xr|?
zEk%|<nyri!TVA!tc%fs=n7^@n7xN>TD*Jbi-LH!d27X7zY>C>mfl5X9vFb!3P+?Zy
zy##GhiBmxID6^Ad6Z*g!<VY-4qaf|laiZ}87)I;}=%0G#pyHa@Qb@YXY?r&g=;*qx
zP_Z*F@U<-8r%~@UNjXxcBR`ZckG)euhzm*n5IHD7#qAMV3jxyE)Bt9YjZ!svYE`jf
zr+q+AC#hB-NvlNZ%J&1+SOP7Yd~x@zAFbW-`_kPBgt|%R>nh%?!O~O6!Mk@CD<Z{C
zEgn@#xv{6X0kH-nGvQ8Y7z(4qO$-fo8VAsUbZhWK#kp%J{t>ng2LD3TYDvZ@lz6pp
zz;kI>b!9)n?bAOYjW~|Jv*K<7#WVB>M2!+_Qz(-s3hCE#9-`M<n0~eYu`-f@OhaRv
zqp@efLw<wPp3^`p-8Q+?Mp5`agM>-libm|2nhMQ_aWR*Yi>|0u&~>!&3vEZjp$F4L
z`ZO0$)W5BVMT{RKvu_qju=DEr_2uesdG~4cWNQoV_e+O<#vbMR(n4&?bpZlFZ7&Mi
z!?veYfgHw#bW0!l+<+Z6?_P&zNf>2vQA+MCqZr%Nc37UYN?G=fXQ%mG&moXlzE-jN
z1CG&F$ZW@hkQ=n-UIM-BVgM~h$(q4pXMp6iAh+u4icng1q<m$^vOaQPGB3mH4SM7&
zI$MM&5OE@|TRiIuY9O}93ZSD-M@6Q0FNER1CLTwM(|)_ox4l--1MD+-^;PTedm|6X
zEb12{L}I{6sp`m9uB6dPfXXuCp*b%R_(hDN0gU8l0+8nNlk-_bWqA6Dc<g!uJ=F=j
zJqOyUrQiL}$ieBcAQ`^XTYdQZG7ye~K*f0wO@gpeqMe)4FJQ%ZRKk2fOzaYaREKEU
zlnpB^Eb!b6cvQV%^Ns^`9hr(6Xp*f4Us%3NW`u)CqPUp|V0#pes^o$W{AI+C=vlKa
z#oNQEu{f$xhEW+S#9px~W3300lWF&2$&(q(f+^#cmwkh;=g4DW_)HO;ECRZhgY0Lt
zIPOW)UH>m_x7mV4`4Wv4g}aTAb$Be0m(0%QC0B1_kJXC<1c)J(+g}Pp7-S^1S!5W#
zx`<X1XnHJ2<X2*|y}&4~45e`#dWi(JWiPTzI`imFKq9?Cio&tRS<b9Y<7viUt|Xgt
zc2ekxNTo|P!C_}`F?u@vaBFzr@yZjowjE7gkK^quY=NEH-H%8>PCE}idxWgo;1MfH
z#`ssF=8R%`+Q+~|y$5zgA0c?B3SeX*>C;Jte?8ORYGdIbh%6Z>Vm%tqXu7C*iTp3s
zbA4{Dvp2u#H_!48ZqF4QnhaW<a65h7vtPBhrr&1TGqJUQ=?3EGt-oM8JNEpWY04ud
zIzPSu2~q4S?5x)_w_aIaoXl@eRIcwSXC|6y7OnBC`AHf(a*g|TVO}dpP%8a!?m<fP
zck{KVf$M=*WPUjhL!?Vi+1s4NL9)CArdF^{F->t<Cm$=O7ENFstx~ip*wQJx)qvai
zEI4LO-MyG9e7uOpT^^Z0Tr4d*QC{5qRb~Ey<<*Hm+kiyEmZ{H6a5=LO-${Py0KDc;
zBz1-F+Bf*%OGZdel)6w(NkEf2!s3cx>e_3I0%%`c{Tm6Ra$jMyO9vJkftyxQHH(m-
z8QDh3S#>yJ6$?o98Ja~(`$`pd>p<8Y_Cg@nlxchQ40<%>$OP)$LV=EEHf8lX#?%*0
zCV-3QdW_&siEUit#>#j})xNFl*o4hQ`qi(|Xv6`pWn;F!{?@xs9DZb9a|(XlX%g<l
z2JiTB2yr8ul)CesQ6sIH@0SC8&7H^!vPXu3l^}s(0(vJ-Uj9(UDGMEc0ZBl-kr&6R
zATkXbT4SJee3R71_kpRwIYZ+%Tj(_#o>+7Xrmg_lE*;L=W5Ko)U^yI;5v~8hCQmnL
zI@vsZeHj&TW(2uUk^YoZ)2JoL&=3X{h?p$Y7tdg{Y{l?_iFD9RT~Ef0>}p&@)1%WA
zd+j8hiOZZ1nkK|Z2#}0dAzybkXC>2t?~YNW971NkQ#5+qXxaRAGc;l<z~d&WsbqQ!
z*1eXliN}Yas2nB853m{^(Y{1ZAMh&T_%VHV0}wp|!KPo#GA<Aq(Usi?Kh<n!B<qBb
zIh9r$5_U^$=W%YG<D=Sr@50!&bTR|>dcpfphjg9d^ywG4Tgx7^aI+hKi{~ZKg%crc
z^+hIjQF~(xC^*e*j3>Ehu|vsmt@l6YW5emFPFdMIL*?#D(R;dn29?46XW*+%#A)JH
zm*k=M*MaqmmkR2No^y?#w5H`VljLSOkU&=9!#t9)(Hfh)OW~J*(NEqY!{{4!;)c(}
zmSTxc>zh-?@qFFpFpsfxqm1_e9-WR}=kdX>v?8RM4qP8j0RRB;c7iU5azSWt1xY^z
zxcgU<5x<j$p5{l2UkXZcm5%DUKOOouFBmA;(rUqnUMH^foZv*!j-1}~z(4)DoOIoS
z&6F37>zwZn%P-l;?@~yj3p=clO6kG(XafZ-bK9$~nJ%*g!M!)D)={N|?PlqMjv5=?
z5BifE_2vad7}<p>oSgZyv)5d!X!uosT0^Fi=hGsMO1e{7gj3;%jw9I4V2Vf;H2YdR
zaIUx*Nq+P;1<2~Bki2$xSuMS{RwKq-8VCKV(y<f2ao4!DqB&Y-Z`jRgsp}j|-HJg_
zdR%j~zRNybNGeC#s#2!&&lHKxts8^o$%#l=l0$i1QAx7kWuMum;GV)_A&(4@m<ftK
z+^b&Xf|>?!DL$aL0^CKrFx0yH@5T~|IgsVNToEc?Nqr+(4fSK5-5fX^XIGg`p9p(U
zug~%pXk;eHp_`xL^0PVk>DkN`Ql$`P6UI6>N0Ju`ve+R(K$<;H1=P*~?(bY~v)BoT
z=g;yhj_Lyai&ABWu&zTlX<{!f3f0CX$VSv0#Kek#idzs$>9lgfCibk@4(ca*1f<;H
zS!E<Lb?c;LcXx>LXOBGO8@R7$tL4ziaB<NSYeGl0Ya7dqz%`BKL4^6|2c|Zc-1iF3
z(9nHh3{`!pp$E@}s(9G8ye3M|=n+0V;-gpRLffGnFK+D(OGiw}kXy45av3KV_ayI_
zzr}sW{&Mu0V>q<%TisNL2miNmI=STN+!WBLLbVQfibI*vXo-U5!sQF4E~6gUSw|_j
z9y&gSqo?C79ciO&Ws{o-rw~4IkS{stJ&NFC4RD*GFgbfZ`|Nw%D%)0dCD49QZGFaD
zQD;?N5t_l<+x|X#(2XVhyDj~%d~gJPt>!IBFAo2FCzWG2k4#?O<*8MnetGbgOncWw
zQM_5|Lv+XR!+EO>`qrbuioP^+fHuk@X7K<5tfuIh#G}et9otYiUZ*n!ApP2E<vfbh
zA463IPhxat;eh*e;qT;y#UxJxs%NsexHsYKP*})`Cd_J(x}`BSMKFbooaVV(hMV+O
zB7XSy66=?O=4>8Mh8Ek7q*R`_dFa{QirCwz1JZz|z8DT0N}C6u9$QbzZs*IEFLb|e
z9!ec!{`61S1a$h}hD;En^_p!xWuJVk$gzz1LHaJbd`i`VwWWx714Ts1J%L`NT<bN#
z(`j=95u1*n0qR%eVD`;`Wt&7FEM720YY{W$6GQh)(9RR+4pPetiF<8cOUY+EmA0ED
z(?}^D=64${<96C|P!^@+SCu#e6!8u8X!0_$Q5>U;XU6A|wgoa%O6&1wNug#!X5-?j
zdOR`8zp;%{4qxrM1N#-8(#FbQ{&0SO&xqmM-)%ylTdr0K*d%k9YQ==hP8jd?{zzh1
z<uG^=`N%t0)eg><d!v_*#(+;lMdDK8($GOF*Jt3if`>_x0!jc$EQ#vPH!k8=>|+kJ
zUJIxp&S7}b52JLVc#>M8@Rx$*DVeRJm${i4#1uGix3J(N*VgGYlJf)$i;RnddM*!&
zX^U_4@y60HPo_HduJ5T(<(a|;v3>boSqBWzJCFp1DRlr{_ca6<fQQZSti3#8cXZb)
zA-454{IH6#&%6ZzYm@m5&wxcGx}o+7_B~U#csH|W_Uc?W)T5_@eyf^H$RWpgmo)Gb
z?%D{&8hx^cj~kbo$KSXpp)WtJgm?!+#B>Ecjn^|%uzyt2J3_f8f$+4wXMf&moj!N(
zu}X%>)}^U1*ZF3`rLgG-b((^?ai!UCSi9uKL6Yo_m|LXnTA|uUfDe%c?o2(wSu?24
zt&iBmZx+IxWv(b#nx{%Jz$HVKwOe@#rh}S>a&o5MO;wH)E-<$qMqpMWi{mt<)>%QV
zMj-{Pp$~8|JaNVHH5Q)NTfxqJbwaa-I=QXMa(aaor8lNR7#rH{%l=T672N3u^EC=s
z{Rff6(4YT&ZRvcZC<||QJGEbb$1xBq6di>9hzzUlWF<w0q`(?(XWmuw@;ZL}FU`uI
z7aaI{;+2Hm!DzB;C9u<W-SQTsGF*nJ-)bg%{PqaDh_x4k!g}r{U(L_r(2Pn-3X4N8
z{CKo~ytr?j1hU8bZNgL8e|+oryYh{oZLoHZI|jnBnxh5c1dFcS(}pl-7+efghMS<x
zW|w*pRQmt`KsFdfFc<Q&{2`wE2UU|Mky&x}1BvU3oNqV(>X%vps>wo3O@RCV(B8=<
z5p>a}-g)2(v{z@tY9qQXx-M<x=5hHJqn4(Rz5WUR2*)BjS4(E%tp4R)a^7tq7H0oK
zl>jxuPDVDz7|HXXF5OPJ_67)ADZ4yM_G|vAx&H|nzLpR}pLko}%ew=Lo9Y^n)1Aj`
zGo`y`>3ALto1!PF+!&3=aid@~G1_jw-z|Q*6S;|;`v7n<72S2_;=e(MC8UiDJo<_7
z-&FEnA@i>9d~=XH!zFltKh(5A8FP_;2^O>OU~9IX`2g9J`URJ;<cdIW4k9r8M6C+Q
z^81iNN|~MCx}<+u<NKLHs6Y%s1B?c1fUoEj=setDOfo=4>SnEo4IH0N_74N;?w3h`
zicP;JqNhwJOKuPXEdIT?-xe%?Eh`MG1Hw%ZqVS*gk;5Hsi+)-cAZ<Hc_G||&<Kija
zpmB76<^pueKC9>31{%60v8!`O0ia4c0!n|81S}-)G!4dEtb<<LL^b6&;B@m3Q1<UH
zBxw6NFAe}}(+EIjo6TO4EhnUGGz9>YQa8&s=unXnv>9zt+p1uhf6M?)VBVkPus`i3
z#otTafwZWx<`icA*DY2n1!=Bh`&`*ndwKd}6JQTcNa<wroL&Bxwq2a%opzep)$5BA
zh)WJY7vrNfB6Id|01?;5m`^|XLo4@>cI@gjiJF5^ts4fQ8S?1#m96=#?Xt86^!X4o
z{KBdt<dN53tCIQOb)sjAV&9ct09|fsM*!iiiKdqmHajX=+DbBxSO+rlJM{>HNpkbl
z0c2a)h8EVc!hiK9!hBAD=&HYEjrjX<I$;AVAZK}*k7WP4(f)DN{{8noL%{($Q}_H-
z`tN`5zx%1K6xipwkE2-w|9($@FY$j~^ndw|#Oxigf^|=4S$`Kd|MKJi)Bj2&5&{+_
z@kvg|pOW<d*+)lyr2Y!9?ThS&D-{0dsrc_c>woe6eg<UcKf{d=M*kN+^dFyUp99(X
zuX9i^_$P7kKmS|*?=OEQu#TYd*!DdGN2nS-{{QAW(4v+Rk%0tq8t^%fb(S!U%YYjP
zNS<6kqy1Aqt#79P#k2pPta5)c`svZu7ML>HonTf6-YSwfIRhZ4wl;);^IxOJ|L{dN
zWPyYTWfR~tHkReu&mpeg9tu<;W9<JkI}Q*lSAnrp?El(v{I4Cy|Hth(KurE$H<thZ
zy0QGr^6%c=9@fyW0y$4L3K~YGJY(|RzMOb<8ph1kk7umjz&N5#70~#KM>mui{_cJL
zotL&*=G^{=7r>uo(LcZTrAFp})EF}n74CrAnF4vc_B|`W^hM|ZAtcUD%w2l|*tX9Z
zkTG6-LE0;DfacPGy@}sSjx`;0KoX0&0FhEPptxrXfS||o(>AM37uNoW;{oXOuLLbE
zgLGRhzN_xtJndlw2jdr4pl5g!{5==(i~U%0Zoqw114N*bH4@~pJr}?N1PNn(ysbfW
zpOz8SFEo{j`d(EpEPxY=Bd~GeIv-K)BRNJk1#%BDDuHfA+khMDYbQ!h<5YVV-s!Fj
z*mA=5NkH5?je4D(kmmb<XB}Xmuj&>>FG1?)sUy5uxmXP*V)Uy}CVO9=xbuluC;z+T
z%ZbF7S^#nm-q*AuSM3wtq}!Ru`_LA^D+J=uio&|nk4Q3cMJQ}2LDL9(v{1jkdk;?8
zqPYpiig4Zy$Z`YCjGX7nTe#%BEqE0C-89Oph0nCwsNJo9>5$T%D0*#GE=C|>kr!$Y
z?>!yg-&x}W=2FMi*P~VEka}yqV3iYf-Iq^Y0ACUK?JtqWk`hEZ2U%trSd+v6oj@lN
zASCsw2H@g2Xcbk%UrZJ32D(PuKytI!1k2Ulz+kTmX`Mg?RIi(_=_B3?ZGZ}Lup$<C
zeB__pmiZ}ye%tf&8wUS;mj)Ef%B!>bt37#7!0XI3FN40td1UjVS?pHGkE^|mt4dv$
z>0wX67hl=|`LzQO96TE?Hr#GJS_iB30W6-j^c+}zCqToyw3GK44mkl>z7l+xUZih{
z7cX&OUK;hpTd~_c&Ayn_YsfAUBx3{DS^_$T&He9d2j{B;mz19S1DgOtXNg<RjudkM
zD)(9u;LdWr-OT}Wx)G2<^P6A_Z=s^(yENb><Z+YxW4rr4c-f!fDDJ%a)rtRhU9Th8
z)dLV*h(RITcM+g@*N{}85uL>K9a%KqZNB$+^l`7Jdmd`hthUp5?*|eBDanVc@cCI#
zXUv-cVmHK+&v6&n3p#+VlQ+#t(M6y^xZ3k@oCzu)#trNXZWS)I`PhPA+&M`B9zdSw
zmF$#0?RH*!6apif##I!ym5mPpE1SH&0)G+zRDEX#<q)VAiaYgy_lqqM^<0ydy^~d1
z4}>|jM-mtd33_gpE`ZWbcqq;EyIguhFOgCVu!+k0tBWnMxBZ8P$6nx&krL;5HZ`kn
zcQoH@-c56!=tOo82DrkvQo);>fo#9ej&MMF8T51_<(!*<_;f+)u6jn9jb`ck*Bpo0
z=}dUTZoKL~1za_J9Jj9>sF1u$I$c^1$abayl`yl7xu}l|v4Te?3;N;Gn5=*H8ITNU
zNO}qO%F#v}u+bQi6mqGX_0>?t?FDryY_`2A>PxJ&$dDWq-hf7{H+`%QeH%85(?ise
zebK0SNON^8M@J7%X?-Hz)ox-vX~e-2eyl{rxEY0=69Saw!uxb&5TowU)DR{u+ztRJ
zETnfQY%=`21XNcV$;5&O<BfzNK659z9nS~V%wBA|*yhHo@2c)xRsbql1088BDuVb<
zL=9geIsz^MH9&<ka@E#>-|E1Iw>smo_s|6hc-*DySn^T_fq{yNSFm9Dx=$8`iWiw6
zuR)kx4&YfwC{T$7IIk=2&q>+=4f&y|k?QruMs>_xe?u%i)MC^bUCmy+cR)4#LY=e%
zNH?m5R+6MKe3G4<etA~y()oaqs4uq_0wv+xD~c6qr2-n@xHm$!RlHAQk0RWWZlT}~
z5Y6Z|$$2H973Ry$Uf3SG0n`W^;X>Q_z;)LWHC#+%-RaFm__H<>Y;uTMY|C|u*+gwz
z2-4Tf9j#&Pg^<WoDw}J*NJAZ4z<4<VL#WV613x@%`_P0uI4~uu0_;%*5Dibh72EA-
z2HNF$QNk_=_f@-i<YCZqas{BGMs+ZJly(t-r>mu`0SI@Z>uGZ)rHapgAnyNj9S}~m
zW{=k3#LqJ(jomXS8IF#7Rhiy6mX6Eb1qE}o-Hhw3K3BEr0yM|=m$^n0*fl?qwiXqj
zTj+F|Olzg+nFL)Oh+eLRZmzCU@{zvuZSVqISL=o!bD-vlfu{hgscDMi`Gca=JZbRp
zv%X0paSjV?l!UY8l!s@;sAzdWL06#vS@bad)e$f3n$ePOq()tot%|j!xTO^+L%}qG
zNLlP9L!W`QT9Nr`(9k6E7B-zKXX1_9ZVDNrE}OzXh?b)UzgfM?5_y<c0YvQk!Ac?K
zQoo%_eK%>jfr+kul*8%LCbOw|9I9P4((DWk1NTL&60f|uiFTE>kgqLVt|pg~qwvKv
zVA0}yCW`P2D)Bq`KlO!y_UM_*w`j$m*ndelZP8ClQE1zUL2M5`L-tthma}{A0}1bO
zt_AQR5VqUDnZT$_LV)4J{w?$tNx~~*%Q;thK^RLBJEtEV`%9#QN)?d6)nA@|(gEHA
z7mpErqDQICIwXPN?7<zxLbc=_*A>r;hk!>%PkeGbXky~!eZ3DZ`P1D@*9Nr(aDUDr
zt=YPNy!>sgmg$V@D!M>%23lxtak_!XNr|ye=0xMrK^JgnF;-+4!5_{rApmMHDGv-o
zq^*&h&VvS88snFrS1G%2vrUX??EDlt%Hoh-T{xpbR1068=u(1E%Gj=-E3qm@c0SPC
zZ)>2Gaf7?EaKvk_AOYb44(-ey-3F^-*F%sl)KMY3bW37-H#DKVGPcRbSJ?B9j={+|
z;Kh`E3uMlFvYAN#*x9^2q=;D^!NJ=NdO(*kiMC<n70ZU}>mk@<oq&s<8pZ6RJwv8B
zM{l}IcmTR31UsXnFzM6=$B0g);C&<vms{$R)kdqBNY5X5@t&@}Ih~_wo#ZWMczFz1
z2-iASr8gjbm<Ecig+-Ixhg92j;bc5b6gYdFP{Ed(fO%?5jq3;<I=pmjnDm}kG}F?o
z0<fPS>xj0W(*fyP*HFGc9bTP$+%>J(PH|>?S88~~w#8^j!<qjSO%ssMuEXcUDf?E=
ziMT)eVq;7UBQ!4nCb@~UKr)@Dq5>`ph<sQxF7((XJK@w~Ni^^B_;Apx2z?u%b6e_~
zh5@RjP37KKmvxk7OW7|TrHS;^Wq-kxNN)r(qtT-u%oMZUr_`_vgfz@5UW)#}cfd`v
z>CG1XQE)TRR;cLL&yOL=mP8m{{dl>w=TdSI*dH!|TO&ZvFX|0j-V4uC{ultO(-_`-
z)OBJuQG}@1fR~Eb8bZkGnLxK#ivFcQ7n<Qcp~7evz9F7II^hgLBl7qy0eFs04plZs
zqoDS0>L`;R)nhWc2C2cC;8e{|AVOvdQ^1bqJJP?P0)(21pJED?C0I^fA`=|TYx#M3
z?hak(clVjl_vP~I-t`Nkf?j4l$=N<#FCbTSpVrM4Ss4x-ATU1=!(AZnlQ4g+oVtU;
z$o7OnN8+8K#{$rRZC9%~x#u({p}NsH72>2x&jIn&9zS`Qm5tRW-tx!o<3dm3)If(x
zom8d@1w;&_95e33QrLlL@-av+#7=^bS+sf02w+cP+M9YExO|BreLrl$`#9{PZc}7K
z%2{Uw37Z>`<^@O&)D(z<N{v*}2_CvioAjrl$ih&J490YP8-*~;$9YxtZ6O>$1Pk<<
zGe8JEfJ~DJ!sZ(ZC4*6{639VAf2-Rk!RkU}I^%Q=Wg-x?&W#U5W2fChlos=G*PrMG
z*>T7&dwBEQG@?KqMY=T9U1Y0;p}?qYajh&S#|Ws*>K0RMt3TM0oTRt=GQhMp#n8hX
zfaLRf1&|R1N+;{=2G|SPa0N<|<>rG{vt)Z0kib;*3jZ)kU!?Wh2hp|GcwBN<>CO3o
z8Bb=HLkopX=_4)r!(!}F`z-S0tcb}c;UV)hu_VA2iHGZ(fm4t@(S~3Mj;q0xBhuR9
zoeTnGZ`0GjPUTtv9;LJ}5=qi`j&-hoETP<#NkUJSj(w?`nyJ+I04IQgnjMZ&JotRS
z3J3mN#uJ2OqFoJ3caZ&7BL{GHWwYm9fPm<>h#t1dO=!%rWw3o+1Gj_w{r$lE2}+nt
zstGGAqRVFzR3*8bu8dK-@Ossd`0K7YD=8#1FVnGcvET`=^_p!d*leIrUzYQ}?2?G|
zT#B^VCh&^+7$UJiua*vycD?BKv);3Z+GGK$_$Fp`B5Vbv!~>3a$ewp^{W%r!d62Cg
z=+(~SxL<Y<Y+-#h!7Y9CLiyhCljM^IvgJbWi%&<pJ!R-Kcy%R8WT&7Jc1|ek8T+l<
zo_&CwoNW^66hJjj1j*!=CH|!g!2IpZhQ`DSQCS>Pf!1xZroCEho0~S+eK!w$iD*3u
zPMwxr`L$_)IImKGvWe{0Vm-jyvtQ%d@NRMXt|s!^g9pkU@(o;3=G5GfrD<@5)7Qq_
zSY(|cue<d^3S%6&9Zdt*m1iw<m(lG<tjdK07vR*Ydlc?kc)K-2xDiZ4XtR%Jdf2lU
z6&&=kw=CYfC*;-xO5t2(LMbdm3+Dv1dF>0|p14;xU&=)GqGUnRPNe}CKB|ty+KaJh
z(e@t7@E4FKr8eucp;e>QZAiA+E!W05o1!8Gx?6TUjjPD2gHp@<4EL{FFdXK(W|2%A
zpZ;Rxfwt$tZ70*U79P|VEOxzzd4;@|rd0zynKY8Kt$RS@)u(<B(Upy797Zj=54^^f
z+cA=?BlNH?m}Uvv?)6h7Twl8h$|(F<x%)H{G(^pksKO#nnYZ*kz6fBjCk94T02yl_
zs8iE3io!{l)BT+BV0IlxYyAs81FUW#gi%-Uw&?mrpGC$tTvB9uO-r~2T9Xr>IRpzh
zB#py(xxb|}a>m`%Jb|iqHjw&-<37njlY&5pE$`pd_plewY@}sXwB+##k_cvE7&Lnk
z+v~!B)<B{H8=yGlc`c(yC<x}B-P;<!mg|H_H5TS7IPYl?T{h}J1lc&NI~&Fp`IO{q
zGu0Rr+GFmqz|8h;H;fX?d~S=S)ZaXtgfTkzOqOnAKb>p2vE@BjDz|EggP)`ZY9OPN
z!%m3*{8KEC6fVTJl=urDCJEs+vSmY9Aj%l<@tVhLSYH~Xb`WjDT|dNX772@%%A9VI
z;OS~seBFGrP?}lSVr$HZ*+<ey-Jpm}`h7kGjr6)x6OU$l10*u2RGv)82silcbqw=T
z<AU9vgJDY1Pe02>@FW`o-H;>D)l_DjM9T5OrVU(ukN9yp$kFp`Km2gYEa~(Q0Fic#
z!nvUO`^P%nRdVUijchcs31oOYoo}ZgLN=Emqh1hw64M6G(S|9z@E6w?i!-i*E96Tj
zQ^1h)o7N6SS~yID?ydTe95n_Rop*dWb81vFw|0X83ETn9w%u*3Cd;6+@-!%U1d2|$
z-Gjy&M?f&S6J*4R;KQ!7UxdFhkbxGpUcalrX4ZOVk%#0qf$osdF6Y)mVO0%~_h%gK
z%ZqJO2j2-dN9A(v_&_GGqk9Sjf@^ze%6_W^1i6NpydI9+ZJA-is9Qp^fgawjpQ8(T
z#BzRiY<OgEa|cpJL@TDFBwvAa%)!;HC6?8+^+E7zMztX$3yLW{Y+c&g3aS{L_Dwdv
z2+8kW2t?Wu?>eq|H@vpHF{{1F1DL_tya$Q7$w!pgVlyJRx$p#T^{^ZA^Lk4uoe#BU
zgrrl=D!{!BoVd(mw{G@bYkU7T&5vo8EzSd=HF?eb;fL(_Jwup%(5-%uysy@mOtIm^
z*kB^)h^gHk+zp4<qrU>e+SQKO!g8cD{eFO@j_azSER`-O{jQBO6d9!aBNj~l$5_w`
z2#L-5V6wSfX(6G)WEZ|EwB(x?=qTU{hZf)A<(BpCVCe`{!jX|}YQ~+#{gnb+XrPk4
z*(+#rx5O$fD;39e!aNDbCfx7RAx}a`=|f#*Y=6fCq&bVEi^`RXBsy(B$CgD!Ty}I`
zOdW|Z`MxwxIFW>kpBQE>Pn*I1!yDRi5KjhSXc#BlYyDe%3N3}hfHg%RrN0{bX~vq^
zKuXr9<_9<zOuWP-6ZZJ3oSBnTpxX%XnaGnS)|3Zwe{;djX=3f=;NqR-!EZY=Og=&1
zk-$a$(f-9Z;fa~P`8CQ+E;(ECp8<sG`bM~zOxSd^j^7C8lJ6F^QeTr(Uzcy6zik8f
z@>6CMD@yl!WQQ$b=(l_QJkN<K1g+QHC@PhH`n}Oy)OD!EW(~&s%22*!*eQfRHm4^w
zW3%0vA)taOMq_3%=biFO<cs9HmayCBb`+WqtqlgZ@~*TNW0O>)$1>%O0to}YF&5or
zjAL9W%YDIKYurX1Ozu8(ZXS~CM0~@0GE>zk-H#Zmluy<N>pk9`@KG_svod=1^pS^V
z3Xg;lhHo&-GzrBvhj1ay@8YYBgxe2o79$z8S-(YKL^u=MCYoG_fj>(7K^Xr?!kJRH
zMGo?1(7MD#bvPM${Jyln!cLqSTA;?Y{(@ztbo@{j9+mR>!8he2lJec}^WRmjKUnI*
zC?kQQ%6F&Oo;pd1SVFN$pl%A*5&}G1_ZjSW1^Yp0Xupte#?>W;&M9=i2X;7bV@y8C
zK_QCWNyST)m=-^x<+>%By97o~*h9<%rGr^aiPh>D%`B2M^GeoW;+2Kuz6@qzvCm!(
zKMn3VW)Ja8HEz2R41ciKmrCAAn#OgMJ+u1p#(^P*npmGGT{F7SHyE7p-OmF}19QC?
z59)4XeIpFz%Sx6~Vko0xce5OkSg%^5*<AdXgDUo}8RdI@sJ244tDvM8l)=JdVk_lQ
zN}`GlqoD&J=My~8x-~g?xJZI~c=sJLS=1E<JB0hO6w@i7$87QuE)%#yuv>f2v>e0=
z(acO%wtPM8+P((Ee{jm*4&=`s^Ny6Jp}n7#-p$<a9lBQEKl&Q&z$#?Dg*2wFPxUi<
zPu=R&H{4MVt!@L}Ua81UfwPFf(;{+2I{fq`AdkF&36#HPM(Gd|C?^}MzTSbM9@oX0
zMx5*SAeKd9D5??_)kO)_cc2Psc*Xyk-{+7K-$S6M<js|`|D}V6Smf5zRK4soc$p#G
zyYTjj**o%~$%uD_c83CPH}+m=JMzM7vv=^KZ&_+sa-Hz^^JAPO@eeYV9dEtx2>+#2
z&YZ^Dt`!D8KO}oOddDT$jC$qLg_IFT{;B39KEx>=W45nPO}y5iSp>z{GM>fftiWpS
zcNsfg_T*#@+-$=|b)^npJ-67aZ=Ry!h)P4!xEEJ}o7h2MPr2b5&KDwXl`StAnx`t|
zn$Ek(*46O}`D|=YGshLRtgaP=5WZ1%za;wF>>h5GK)(7_tuDLM<ch@CNfmBqk|nrf
zs%qRl=z{}er+JvM*V<**dQ>Rq2*EG+LWo9r?QOSj5N=LEDG1s)+2?DEkdn$nG+BE@
z5e<f!9}2B3CN5|45h)jK=A2o>JzNhBUI<@effqNrUxeTov3W53%Ql+t&Q4L}pgCYn
zRh)r$^P8*g617UiI}X1lvRL$1*Q!u%NDe3Y6zgbG(=hu$EnP^@d*;Nrh~E=Ep3wP3
zW6eI>BVI9n<3b|7W}7~Yh)#je$akNmiOcDjL1$Utk>Nskb{iW$k)!3wVkA-AgXGvW
z^@o**@RvNxtepa$L&Pan8b)!dn{Uh_M%mQ(91D-NaYLBtF?@iTvYBLh@FKTlqZxjC
z5Y8Ey`B;8;OM*KGrQ((<|DITKl1p$4;&ZMF(+W$Eie8bawPJ58W|XHjzp0Vpc56`A
zroisAfR^2lze?Klk2Zd(oi*e^nG%Z>ghA{=d!Wl??;uY7X3OXoJ4IYqhFc_jgTB1R
z2oq@Tw>+GB9^4s|*oWr=AL2250w^Z%k0|dwkGoQ*b1_nC6Nql?R}V0FVOGLLf=knN
z4Yy<!2r^m{=s81P@i5Gq?U`sAg^|m(6!jz66YjBE9g42%!O7fP;kM58$f^+YB947=
zD~d_SmhnsSY%u&bUeGyz;b|1goJD@oTur(ur6T(b-i(?~g89D8l{XjI=<->3lom1+
z3E<WO5KOXjDM}ne;*b?2X;kF4IXap|+?A3MH$nMd`dtY*V1)HHJeRBRH`@)&_HGX+
zIurPwE8IbAIM*DBdEBo*Vg%7tbc<ie)RDzEigC>L?c6ONRLD;JQO$qX$Cx4g`aUF$
z_(iu5*e?biA^1~>i`Jmp9*n@k@&%&-`(G9{C|F^Q;%8Wg+6^1LUFdHrGK=8`A6V5i
z&C|`vIX(**2XyaIwmd9^e_^xVd5(eNg)Q|=ViI$xMv+0BJ<%ZimA)*lM(a*7&OGA|
zal=Y$8%<jeYZuwXPr3Rhw?Bv4w<`!PMM;0<L<ysqKze_3>9=_<`b=sh<QtCH?cd&v
z8%OY+AlgTtOy~optd-tipy{w?d8EUaXUe$^#AfUwor6Nl-$Q=bNKE2U3ibm`*s>QF
zkCowr<L}$T6sxFozPlWqQIWa$$=waCuAUSke|G@Aq`@&ASB6`X^KR6uh*+dYea%@(
zCGT++SmcLnO;6_~%Coy0wqZ~3q}Xe`#GX5ToHFn|F~nSJ&J9(ur8ozsm}m>^Rc~<R
zQklvtPpklP=#$ZutTI2n8lOXbzSWCck+4?YUPJh!J;aYh_DmbI&TD&ai@LLU(=#bV
z@pc)ON66hz!}Noi(%8~8!S|k>6T?CGpd0&Ja}3!Vi84j<C=shYh}@I&_!<pCy)DP#
zkW+4B7?(7i_i(YbO@}-EB&~U|HAa_vki>M62`}&Mt~r07pYoSc$<jU?T{&-7>$-DI
zq+#V9ChYV`qungi$34<9SD%4Fz;esNkES|uqSIx*<D0|9d=Kf_Pd0a`W91SAGpViL
zWqbE(jwJ1tf4_K_YS^ijKxr;Cu7Hxx{!&U0O(22DEZ8>Tp_x#$0?JAHnQEWmco1QI
z@)0bjk+*;y-Wk(5T$eWg27yfN-im3eCo(v*U8_tnEYW|LoV$YsAI9vOdCauV;InxC
zLS<%PpSXce8Gz^1aGz_l&YI@ezNa41{SfV0bdP`8=N7p<9JqW@`+O-SnbW;?!{YSR
z%)8_#XBRhJ#t8AQo7|%Nxc@`*Slv`SSvZ_oEbq6mt_do;V$ieH;kI{O0h=RX2h$zV
z^tHIs-qBmx@FlKjk#D2LVbr^>l-*~ANNX<2Nd=VYnuim6kNU4?J4(#<4En4sX_@sQ
zVrl^0g1GOk&oZsuH{kvH0Iy?lhu4e0sy{i=h?0Rc#w7Sup^=g9gts3PiU|>$?buwY
zYTZgPe77Z_p^%*nypb?#1jj<~BkG^Ot@uHoCcbq4RDM6hpi>aPkHMG@Pzw6*yzUeV
z%Xpc}tER*)m9&oagOK$j!?L##4!BRfUYUe1@_3h?iivNdXV=~4^bpxgjg0X^OHAqX
zZFx~#aJ!WZZ!hx*GTvmqy@<9GyTfsQ;{a5F@4>SOyR{rL=>-at2<X?kAMp6>=+zJ_
zO`JH1^b>Vem7(jiY^bN0yODb3H|V^|*MAIZl*m6xu8?BVwq)mgHq>1qp7W@Wf2G4^
zKwWlkR&xTyNh|EpE-bTsrONy~e%8a~n=s*K*O?_7LQsdOaTKi&tM7(QI}j4CW_-$Y
zXGv6XvP5(tw&;rbL~f#bx}-l!7-GM5tUQn&WnQOIS`Q-`lIgEXsNs)Xj7xKfBAB3Y
zGUo5>bj;?c2;PNHy?cK9g*4;bq`w$X>0+rQdq$V*5r26O=N29qrqEMrm~*3^sdRR&
zc>tRIBSac>I$xZ9b-72??O5L@(oWva#95B9uDT&=wV&Kkav>Sv%W?ob4$t2zA4iP)
zFFAcV-*OTe%J*m!QKmL%65Ar(67RdVA@s7<QG3(Vrj$}E0NSj2^>9ATAT^8RmC(x<
z{OeQWtPcZz@?N&?j&w|nCHwPWlkHuU49BNF?B;4A5P=utT8`A%thmRoo(-&tXxqVo
zW%NX_#3Tje4sFsca&ZGGv{d{uGDy~(%!{Q(a#ShoS0!+jT|8nOg4>e9NgiaDYuxJ6
zQ6GjOguVl2IL8FsuQK39tXKlT2b5^=rk<<eEVJ?ZXnc>g&LED~6wAt>N0{Da7{Z)O
z1qlY!*i^FQVA@5J9W$G*pFx<a@9w+D5}M}Qz8D`8P!<#{*He+`+Wu2i=Hrv}lu@o(
zu|az&o(Jx7HLtRww(>bb*W;fuzZ2Bp1eeT2tOHe%!|1z60p5nG_b9Y}9pyurTOH;s
zBnyav%goHiqDxTLhp61Eoy}JoaKgw!k4gbO4NwA~0F;}`LHGiGP_pxW$uKlD3FeYV
zoa%4lh(1d?(}@M<*%yt+c+p~$Gxw?XISzk3?P!K%t26!|?7d}JlyBGeJ#;DUfOJVW
zN=eMnT}pR@lpsiV_YB<)f*>s|AUTAzfPjE>cbD(U|GKX`p0}Pa@3;4p8_sMnbDrmM
z9Bb`st=|p|$n#c*1r^S9KJ%2wSNgQjHuP*Dl2pNDFS16nIG~F_cV5I{nqleZ5Fv`>
zljqN`AX>6koBF-pvB%Fw1$ccqf^(7w)wd0}1<lDB|0YGpFSG+yMUv<5PK=HnPhu;J
zlo)?JvRQw{s9F!@37!h-xMa#Kox;?p`iyA5$TWDi0o&9tN+lqrpR-z_m@AxS96W=4
zfBzq6_1RCIx4bHdRqt2AOtFXS<6&P{yfqI6_@W`A`!>_(Kd8!QK-^XCFGmEzP@hO-
z1F)rI(|=J}`-w`Y`og+s1?UBS5t^b&3ZfIj3*l9%`i??X0<s<YoO!uc8k905sOTGV
zsbaK!7|F;dX`Zl4uZzdF-rbh&BBUF_C&P&(7A7~uKXzVBcmi{VSkJfa#?L36AP<;G
zg@la9K$GJ7=a=76?gfYrsra&m?oebsO!9h{*;h#$Ty2}A59R?AmlBgwsg?H@8fb}9
z<)Mp}@q+%OjOjC8;*qH4WXP+$XX&ut&$)P1C*x`|1=ln`DZO6+3ZeIP(BG#%OMsx>
zz9f?y?0fbC;AB3^tnMjl`|0tgdXBj~TsJi$i7_lZ&Z*AKIg&Bd>7-!vhce?U6Ug_v
z7y=VvU{4CG&+u*FxvRA#!AVTIZtuii)mw}P*ev7m>5tQ*izciK0*f2!aWc7V?P!2+
z%!v$iOKaPXSElbn*JbcA*UnH%g4_HYs?HOMQ{1gddr#TXCn#(&p35oE($a<)j$CNt
zo&1~Y*(l?MCh(vb1AqbJ*hObD)+Z7}SGZH6$q=6z6`t=>w$nQb8}avTL8oy1R3|Ps
zC@_lht-+sbaw$?U7N|rW-aNJ>KNX4WPh(kfSX9(c=4h|49%V2*W!>obR<EBF&M}I%
z<kzk_CF?ceMjOcJJ~qEv`eMt+R7S|-5P=ewT#wQ~C!Og2*+*zes&BfCq+NFn#hmZT
zbU5A|bd~{NrZR225r#G>%;GXbq$&BQD&(+<xUmh4!m=g{r3nT~<a>`DKE-goZBe_8
zbVT|6_p!J`&O=f+N#CRXR+(mpNF6}?ytim-y4%}i!Bg85A=xUht-C4|-Tbwd-KC3k
z$g|YQ?pc@aQZ&@-w3EWa)5o1=F@_cB%*Ocl)E>1^GmC}teJ9d`z^KqZUSI>VC{Bw@
zgEJ!2GV9})+aG>aCtl(ip0aiv<tvep*Mv6S03>f^7>8Y_fZp|!BnNuHOLk1EVPQ-p
zbBrgTv~lUexPY*Li1(4p5n}<M!A^Yu9`e$HlxHk$?0FF%SOUOfS~buo9m|F@vze5Q
zuqOa*1`C^9Y$|MVw8Ue1WACl8<c>vcs|jEee9%?FJ$hK^$m^W)f7D3`Dj7YbHO!h@
z7&N>lK`T8kx!lB&z42PuwR%g00%-8Cbebvyn0RDrmCL_b6tB}YO@!swiLG^y?S!T)
zl+=ercOoU3a)0Fw$4Q6fclYe^JfU&eHmo9k4ammP=$bHZ53Q{bYQUIO(+R_ArSK{W
zMS(HN*;VZH$vG)!{b#Wg?#|L=c>pKL+zyyxj^#}UjO$g{tj~9KO19)oZjYB)%P=hp
z+NCe6<0)Fj@nS}dlLd==p#AwJ*&JaI`G;=M^58cSpsyNvm5r)qK<_dyqj?i9xdXAP
z0!hux+q5035h;!5*v^b3XuT>~Vz;qvo3ycu{Gj30M|GST?>TcNvDz|UM7IDBsraTQ
z_0O0^1j*rB<71G{0J+(V)exXEm7Ei(H~MNm?CXxV=BPlOf$pxY>`?Le@UhoZ&?yiY
zi2%jzl6)W@3Us7XD~m18Lc}6cm1v~+4FLh-v6Xg(>E{?;kNK0}J|%V$plxW-tGpz;
z8wG1$6iN5lDl@uA`F3pja!@{zFWP~s9;pb=FSO>IK!p<oNaW@!Z~jPN@DKI9IU_O8
zpKYLI>K-Ftaz%}NZy7h)kvBSgtu}(sWWRk_5DZa`r--tCE;7%Z18E+MU452YyDrMZ
zxfaL~e1CQ+(aDVZTQ%jf0C=}|tC)n?7CBEtuoATvkBJM@IcTVVxHF;UdntjZSdLt>
zb~Z0GdZ2E~-J_&T;E4!NMlls!lgA8#A??RKWv<~^M_Y-pNZVI74ItU7-UtCBl=t3)
zd_#4ESI?+cc(Tm9oTi%;_zsyZ2UdS=WL5*iu->X^2!FoJRFNkxThZ#^#Sdr6J%wD!
z&oVe<`9`1J@U^4)aHTcL{Drq{v`Xue{_U1aXcrB8v1RtWafmOkbv1k0$iB=y=4~A^
zusKXrz-Ai7+4kS{o5lHj@~FCY1bpM!W=vw@jv2x-NbjBf0J_>@b=*f#yq)v1rz{49
za`_L357U@~3HM;@fB^nCwPqMUNGd+Gn{?6IHA^J^>^A@|vKId%l#RROb5`ubHpaHP
zpWGtL6#AWLAAOX{f#bUKiuEvnhxw3Xl)(W{#Ae1*#smq&^+Ux4<Zsn8B*(4AvAQFN
zwbot_t|f=|A=3HxPHmux_)x6#<bQ=xbD#soU!EkkY-1#SNajjU?&{WvEi#s$$6e+$
zJu)nPnwPjPb6(vlt&Kk%DVg;2hnVjKA95~n<cKdy_FdX4+M3sd%X!Pi%;#IJ@x|<?
z*@R2dPy8|bv<O!(W1G{o=V0cO`P&xnY|}vDtY?~gyOK<VB1h^fVX}gFv^RR%5j-Q)
zcWoniZPeEyEl9^jpV~9(%cjHMRr%{slj=~m?$CyQefkyM7(c2K{@za$8p7SQlpn!v
z=ttjj(|GCl;(idzRUD6qkl@Y@hxdq}hUP}T5{_gMCL~fJ%;ZoP%vt0K<m92{^qu|7
z|9xU}2z9jJ)aouZZ5gJss^PkT+R+nIRbXz#otlOL?;&Q7WHGoSJGfA1|21dgW7cJD
zcF-SP>9mfQ$pT9Y65+`&E16G&Wu<j7)`XRs`4FKJ5{VHKA=vsR|1na*4;(~QQcswL
zeEplF7f!=t+VLJ~|1%2{(VGkgetl*hY56|}({XMH-%P9QtBOfCpoYgpcsi#s^QvU;
z*unJCjqJnH{=6(-ZPn$(!pbJ?o3-TU<Ti5lT~byqC4x%!>B&7Yw?uDnudV6hB9zR6
z!I3|@l$Fw-<+A?z^hLn(jajhdi%t81kRD>u$mdikS#}$cFQb}lyWPY`527_ll~z$2
zlV71DFMD;>eKG}CPE7{C6|9!@u)z5o%Iy3rGVHaI)ONy^)#R1((`F|$%J_Q4e|U$F
zrg9(|_!JCRdFjl7^8V(eYm0JqTfeZa%f7!?#!7b+PBrIO`*Uc@H}tK-h*Q12*QzKj
zEvSS!iiPO@a+cUgy1&jZ&D7nDD3Or-J+sj7VQ(rzKD19<(JZ0e`Bd{8=YU|yJ?i)D
z)NydRhd3H*3IBjn0uTyiwjx<HBx_&jw}0VR1Eipz2_ODIoTg=SJWKn!(0hm4KY998
zN$6($;{UIG@Y>EncF1sP&v{LYZ>3Twy=`&)Dh$EoWBkvAciuea+jr(HyMX_x!E!_9
z;9W<l-1s%b=+nEmZCuGw0@h~uJZ5V0>2yCC!nq$@-g(f6PrX!um!`?%h89VY5l7j`
zY-%(2$7w`W{5|wm-pz44^gg_;zHHCk^vk)_%N;%4^Uj?%+HF*sX_<C1Oa+@4@?S`C
z1jwAt_Fm_JXX#w(!AFIdWW~HT&#emMREK59j|-i^&UvyWGM_3#Z6{|^eXC;a9xKwt
zP^s9#t+8&qM89=k2Ql8?4H3Yoq7mG^Ac9uCC~Hu#&)Iyc#;;`vMMxL?!eKx3oD$yM
zVge0^3gvg3C`*UrSKe+H#qugUzxbUNkzGPPZq-tXcIo<Bqd`P6(~hmExr5)Y_oX(&
zVXdI0o91*+0jBeX5V!dOtwgt4F7?;g2K7c=MYDZbhAKI+6ObdFkKXM>fsbGL;i3A;
zxYoCV#c|_6f7a9_xsBv0!?Q1R2E{|>LeoD}6VjZ|?+Y!Gnw!{nr3v(ePn~s-qql0d
z>&CP{dAa=5)PBCfyE1W#HLapcJ+gVWO?!j;oGEfx;0A3l>7xv!|AmBCQa{e*?})v5
zd0AbGdT5^-T&%EdT*_6=)5PEzd~szW?4Fs|*{}8IQbiTmEyxF=rBH|%4l~vC|0N?)
z&oX^u0oS0v*QT=Z_t6g#ntVDLCIuR%F1r)mAx)9(5x54=PCYFg(Vnj)a_gsFG|yq%
z9aQK1>RlVJ;_*EQjemM@rTAy!ilp*7(llt<I3kHmIxenOYyKQ-PS6sCB*-OUXs+@$
zs~#S2R#e1xzS`@Z=*3A_UM%$GaO|=My)bq|y-2}MpJl(|?|4U4TQ27>X9}H<z9u1w
z4L9Y&`-S`b!V(Sl>ar$u_$DOF_2|OOWsIN+Te`d<G7ZbcuXZ%G5j?C&)vR5-$TZQS
zu#KDI;49Lm?X|CyUs!LZtA-|`pT49B$HE7(D5BGa{HX2h9A$vD<xL$gV+tb21<<Qe
zBN3A*Rj9E_+KdK&EF?1x6H9H^eZ7A=FWQlaFKUPiX-4NxP*?3jH!wjS(BZ)OE{ff}
zQZ5p`(jdsfDkiYwNzU6_x>VsswGzbMr1`sHch1z-O@i^-J4X@q3BU1Mv?K=?xCxPx
zuFRI_lOVF_7wlVkS7q2eD7H$ipUbJXTf{2pY_isQq9&Tzj$f%v&_RfhX|TE1lrU?9
z<)Y_m7NqG^<gMOBSNxPdG1AOoR!Ma6p+JS~6dMF2vwF!hIp`e_R2ZqEA>pq_Rd6P;
z8j!q0!N);b_5;Q|8A$h;%i|hmvC3+x^9i4)i^Yb9)sEF?6EP{^_k}%9bM6tDf1B1K
zS4M?Yf1YEOk*OaNcsJub#d%%n^Hr0I{cbQ@PJ3ZHgy&f=BaX>EX27H_70VZgqI@i*
z>evU?_2Pk=ZBqWqhtILc*0av_Sbq15ly1=CFZ>yw2CxEz$R6ei-lJL(AqQgy)cTm8
z;Gb4~4yeF`20z{ECe)Lm%$-2yJF!nb&`C*EaxzFBDVdq4iH7n+7iDpKsv){YG_{hP
z>4V%mI^w_AY1!V(AWNulCD^(w$T2Zwe^wT7NV2^#nuz_j_JWH?V=vFs)vIcN=g0l^
zM=Bb1Zfmv#%^^UTyF6JcY#8U?;lF=+N-d`Wi;fPAZNm?j`Bf^YAU5||drJ^TSw3Ao
zdnGGo=~<rEt2LU559<GU8J_^f3>f|uEsQT$F-(2>z!z(2|FB0wS~bfaNyYY!Yxh%$
z)Apqi9Z2I+Cq6l}h1o6xE@q(y>FB;<j@I|3wxk#Htyz$~^sW4oX@cHxrn&WPym8D4
zu`8jUyA>7g1f8;W)RvJ@!YdrIFwCwW)<QO!cPkjV3KmQHO3P0(l;9H7v56pM8)ao|
zg;;Qfm7IXQh_ApEXlFPT^kWBW$JM(v4JYHz%2Xm!UGu}USps=-KPFC}&@k~$BZACA
zI<Y+UQYblIpLvtE40bnF4|k|vx+gw-R=Krljh+6{*s+PJM1nji0g{3!rW|YyTQt;v
z$f^8Ps3fi*p31dXZIg9SS31yTA<Xr<wvN6Lgt)BvO#7i7o~>*P<-?L^7N$`~tT`n`
zx<qG=u%NY%j!**`TE9R!V+dx!LMje9xk=&+yg1&?ciD{#Ynlg@3`VGI>B$Wv6iCS`
zJ_~_&b@W7v0r$w;^!|Gc>z*O*?9^*1r5@C^g-W1I(6A$Fg|o+dR0<^7PJ+^SkXC$k
z08Uwp9J@(Y=R1<uSI^5QD)aB`L=yXyDr!$CqMBSeC~wbaY@f)Dx>#8A6T$AtI-<4E
zPwK1Z$u*uz#jbb^I&v^ZVD2}w!xpV)judJQm;_h|R>)QzsjfJq5zMN(N;cjkR&Ubj
zRP830V8^sAkQkP@r6hVHucTdsK<@zml9e{NbzqH`hWEP?uc878ssgcd%%zJp!4PI4
zbEcXIcYe5t)>};V!?lO*6<t6dDOf-*+Irp3c>ejz38>19QYMbYj>%HcmG{l6ET^>w
zuIIcw7-Tys@3&5*1F|4s<>Fh(AE6&ANGhSE+1&{A>KgH>5K7UVkTO;EEKm!mkiRa0
z7!;xOvRrP*19QFt-5iGk8D{cxThuz$!|rgp%u`?D%)PD>`#7Nu(px_T^UQuJ!FizB
zhaN`)QRmWceour=Hi*vl^sJS6F4_Th(iMa%82CIo)bn`(%p5l;l%4o<s^Ldn+mp<O
z4|K)3MeTX0=8o_goE`F}1OYJTJg)LQbJ{pR?X>y^^K11vJ}|?&xGlPBAFA95{#lg5
zeZydIt(8$EFDKd@6?qe>ei>U)BHFo+SbEO7VD>tEVw~aWvPycutJn8KeLjOIz2Z71
znx6!<8pn-`+g=LsQ+Cg7Rb^bHuXPLv&Y^a!w4VL4W$)$PNzr%wC8Ro2-Z$1T$Lz-8
z8{I1?m$5BESw5jQ9H0C8aTDjJ#H4vmYrXLy$V?c<Lux+6-#wQ52v9B|8MSOTqsZdZ
zqNm5T8}}#Hz&-wiX0^8SIUs9c^v39?cejyKFnicu@DoIvPUQ^1F?hx{jadx}&5GZ9
z^Wyet=Cpclxgsd5@&v>M{#x^1jJfDMcZDMQa*i3WmaFfKd0@$%EV*43WJESe{Nb;;
zP4-VLKnvlq?p=MhE<-lfuRhAm7J<{eSirh}eg0205X3Gc_8Iw}#aC0ETL0YV_pTdO
z!0O(u>;MjG=Sqp&VrsPUWbT@CAd9)~b0paH)AP+YbHl@7O0FgqEap04b~@m4-nE2w
zXnKKo$zK^Rl*r6iL#D{P2#+`L*0+;C^Qy{jf{zea+?8Ue1)MDsuBL!{X}!Zl3007O
zEtM7{-jl>p=xEWyNBr;EHyGa&PqMzCb~fxvWglbwrd(yVyVgACG5l2db|6Z5ozL|<
zf@Zkt7Ht}}w6T{b?~6~8Z)OATwoeJ_oWOcgn`geY5n!sURMHbAa7i;uv~{^0yJniD
zv$6F2!d?GfGy7uvBRM7i=tCAgHwz}ajKq|!l5VGJbBJF=_=U3a-VcmcFpk0`4J~^X
z(8mIeelR^)XugPHG=*wm?+J*teT{JN8&{+sljbl&+gHG(Bl<`MUzu6yl@)&2dId@U
z>&X#>hQ$j_k=Zjyyr2=_fucs|QAR;gWMDMUq+~F8%k0IbUreFF@%M$|SZu~$M?4{6
zXhUZKbv+eWz_)nDAh-22;|Ur^1)B0n5Thitk}P=P08N>FVizi5HWuUnTUl5nQRjv^
zaxkFj>eRRk<27Cd*N&U%!4kx5Cf+{hyCy7Ox*NBHU=uC`M+88F-j6L46`AlZycELY
z6{31Mj>DTz&0tTum`rf{6xhmu9W5{rGs|^C9=djt7iomgA{w!`cpitT2UWl)XM2f+
zGe<k+mRqAF#oo%OVE{X)B#&SFT)itqU>AhRWjG@^jw}V6cp-;1>2?T=C^SC!M{_i6
zwHQ6a!-M<vnX{M$A!MKybVI3(Ay6q_a`ESHq)RaML=K`-c@z`-X=-(0L9_@_Yy4|=
zIUb_=JaQ?TCv**H1g>3xDnVUDflZg=);oSy6sa_RHh+DKAaN-+HE;UZkFlT7=<=yT
z>K*sBj0G2YVeyllGNPUr+T>wOv$m%p$ymGuf(ks;k7l>r{trHEIB=!YQFlN3%z-b8
zV+^BxzH96R78R8+TnlQJ#^43IQ1|xJZ0K6{NUOtCIsYI^dNe;gd`hK$Rm^77X%p;#
z-mKKx{dy+j4zaF8P+y^Jb6G9TqUa9Ao}vl0gz<ivq$;|4d5tafCS_IlK9F4x<6yiG
z7J8ki@zMsvVhlYoV8I&*ozdX_!E*Il7rpgwauB<SBaJllHKREaH;E!9yWF+hi%1cU
z7oP{}5I!L#I&_lnCa9zw_s{U^-EL6fWc!D&syt(hDV45%z%RzEU<JJ=GJZA~Q>Gj+
zbe8)s;;5?cAJMYG5q?nYrJPa4<L9N~CMq{l4PU+JU!)}?@xE-L0t7@Twn%gpsXY09
zmY_85Qx_zPQuBKC9j<#Jm)E|G``P>|8FRSGbIj*~FC$H?Pl*XWIK(j}l%D^6?1_y<
zn{qf|GSh|eibUxAN{r<n3>KpDods}_4i1Sz+58e{EWaxbfCl=$K+g$)8D+DrcsgKa
zR#GRTp|?{4jmCqf@C30BlU)GRzVaVf&m}QqCK2p&aH=&gMsBfjl`YjYu97hs|Kyc-
zF7~`(U&oDK?eoo(%4RO7E=pfWO+!9GAxx65lkEo1A}Ggmx%+4fpGi*gv*K_g{U-3j
zLFQ$>!M6Az!X2krV>IWqIeP8%&jVh9g0$vwqq0FaXRn=sx<+}Ub7QzxAwn}xOHGRx
zIh(q~djBPs=%NiAqj<PU@|BdtBDTju?rc36`4rOU=b6dVT)OA!+Ba-WI@(8&*1?@C
zEgm4E0S=Q#F`u&ilnvGqfb_`+deW_jWW`ez+xsYvHuy^Aq`IR`dOjh#bWO!LCwTI8
zo1OSvt@n)nFJDKIDEHEkq`sv>)>cY={!dr=EoccH_a{r`&}&Q1!lGvJQzHl7U53yL
z=sZdz)O3np!N+0BVNfx%#{ND|i1E~7Fzo_L-ARG<n`$N;t8lIGi&yBP<)5z5=}WC*
z_?;TnNrL*h-0JS|m$clWcQbO%v2};kFKCC8)2ZuHGB9?Ke7SeaS-72~`Z?$_V>z~k
zTTIX9Ya=Y(u<z-F9bI|0G}3}6b0b|HZj#Q7j^)w|=<(D0&P*$eCzcAMsfd9AvDi*w
zx1VGmAH+`e93*(Nnaw|B2EMX%7<pxga*cs;!|G}glQ|{-o;49;P7Y)DY-<xBvy{*@
z4*3>Ji4WZ6ilO-~=5YhrT)%9NZg}1oH7z&fTBMT%gr?-T#gEygd=D_KGtP?UCtYd6
z65?H?Ovnz}RGEV)^_Dx-tIpYDT4ICkv3mhI24MUlN7aQ_7G6!vTXE0zrAh>vpQ$`u
z6%+Qn(Y#S{B~KE))F5R_*(UF%n%}G1sgG}%->Sa8*rW)Qhnfsx1t`U=Gk#o)6bbSp
z?8C26uZ$<uVK?XDM8gbW4p7glM*Kjg8no4wuR*ibMet1If7G30Td>xMQbY3uNkgeD
zl>5OyK7?W<q9TGZ3p`E6O#f7u4&xL31{+V-Eo_nzeG_S0M&rVqv*h6i5-68_PF_=7
zXcOxZ1u{yM8#0P$bTSM~YhQ~$$ftCMKTJzJ_{ubm&ss$@$?x~$zKkSmB8LjZ4DaX{
zxE~;-gIHTqfg0HZpvw=)K2b(mm-bybBjs^0@E{nQbe@KHZSdf;U{hR>suqqSwP}S(
z%!?neK2o~@yg}-T_*$YMdIfY$nchwYt&}=sY>hsnK2>5!&kzBcNYBTlvb%h>cYH75
zEOSa2Cs^xrb}0PYD%Ysgm3-jBF-TwF>>Ds)3PK#ey)(gM0+rbEHMSh_lxCnyaA$y-
zTq~wBR(NfyD*+5;b}BFi>UA<E^X*-s2yD_ue9clVFOK$8!;2+^glnoxQY<y10yU{*
zfhljn?0YPYolObqng9cO5;+Sbq#)j;`$@=?K_cnbmQUe$F9`N=iU|fQYdbeciLqfB
zYM&>>1_T%c*2LT8_0;wVPy7m5Qj(B*-X@XOJKdnR$JQpENnDsa$9iQC;s}8?oHJ<&
zK)QMP6|`7zsvGv|A}0wR$wl?)w`klFrb7nE#XfZS6UO7J4ghe8F<hG?=<E21Jh_GG
zPO?Ft{=5Zl64+#mO^p?b9a{18qs@d2tt#&fhzof?$rI!Lsay6n#XY7ki35fNs~he5
zm$cN9_3!S?MBt{Yk2kpcQ7w>f7JfY+u6=6L;<nl2lTuo20~Y#&EU8Y!0zS(Bg*0yl
zHhaQL=Yh@6zi`B9Lf`#~U17(&IqL?Vxasd-u5br!UD0}D>3TP}MrluKZS&b*+7`y>
zGLLK*-l9#=4DmLQ&K?Q@851BGbh^@X8c<D$eBk3BOh!AKL!3)bppMoe;fU6<AI&i)
z)?S|4*(+^&Zfpd1F&9y+yld$cVHWLPa7twFYZ~<dU56@*)Y@B}(t>Bx8*6^WzJis}
zCB!P^q#c}?NV0uULeNV4YSpmwum=1=opz4OG@fUl6vi3ut&j5(vqgnWEXn*11EHB>
zXVbdS?gw9^pW1AknHWxCXqpNr7*A?_%h*QkbiTOPTz~uV+f!s3H?%Io53-mZTSVi!
zkdKn$(ndU8$k4R*CRVi)`zWh!-pn7g&kEQ~?i&<jGggA8elXtn_zL|l7p}!sAe+Y?
z{zS_I<+cc>xBC(^V{4enFWb(I$<2vOgYGcak$m3U>my+lWFn-G72q%vDZGkjN4q8U
zi#dTD%$g$YH$vJnE<?VaN;YI&ah(?qiPvDNPmZs3GEc&y<aDqsL7f=7dFD&RAntl2
z-~YqyiWFuzKqO4UzyFC_g?&;WOsL_v@lRW7hsJT!ll@Ua<iwE9BqGp&NR&=GF*Ozj
zahDP<CjJoNw*9P}PJ|+A*BS1h4da@QC{)LS08UAx^F6?)vnNCxc}YcKyFQq{w;TSO
zf<5UO3K99rNbtg1hD%-qQna6RF=$OZWL6|^s!Zc%ZS>Xk4>QJ)=nzj$10fkKb98T1
zb@mT=s8AXVZ+7!~mNTv5fvSI!j%sEaSRWm6F^nQAP#OfrqJf*(ej!Z@z>{1wx0iw&
zfJw!}f4L2}?)g;J>U8=N$1o47`$<I68pEQmBnkLzn8=-B)Wp2(;6%>$ev1eZZbvPG
zcOH$lr9ju;z2k$}8fTv_xHf%Rqi-4X9n*!-oac2nArH5I<5lJ{rRP)6Xb(^7p(y6G
zuwJ-fOA^aDs&3dNv+UXni_f1u=1R4dhfwJ7+~8h9%mGf~5^H&v(U;#jm<A!k5@HgY
z0muJl0lZkdES-<UR}=~+M-n0u;>Uv(&xS~PPKeR4jB`c=H<izMPyykepd9(t3FRQ%
z3z{v_oGX`ufDIrU>Fd-&dM^jq$Q`dZeNndavYJu{qgxi^LsB>GW|xGJV9gTPkrDh#
zt%HC+cV1TUg2p_gCUH<$R9~dw*GQS_9-`>mo02J}!Pt|+Y^|^3W*IwKGY*Bz%7$`H
zd=ETEFpLeV;cqnaaJ-L2Z&cCIRzB-f_58S`ab3*6COQ}WsXQU`Ee~j0@DbBZDS2G?
zIl{Jx+gxJRMR8}2T)q-(nDXEbsZOA7;sdL1;Q#^qPK>EfoxJ}~ncQD}K#tjZ##w~f
z_WTSdMm8oN!Ks<=NRFns)&)g{^f~XC(;aCQcmL~F1Nr3>Nstt}#DU2he2Zw2v2Z#r
zK0F4_o3_uLB1q**_6xyH{LC>DgS3^~j9Kt7Q-m3N5imy#XIwp2;i=w=GSDswHx=s=
zIeG8U)Kehh&kj+MP6JjmLI9@AoLHdQL1$#IaKFYRZu>zU!RcvJsPx9i4V*UF>K30U
z1+v#`{3SRT#({}#=_2MrJB>uUU1N%AXWt#7ev{F#^HYfl6bA$yCBqYZ%2)vvd~~1q
zIzra^Nj1T<XOh%B5uk`1A}5UD^I3!FWEZ)MGoJftecDsg;~ak8<P^ixwpwQ@a~=<i
zU=t&mEGtSf-(O9b0SNM~?jua8*Ok~6-mCc-UF-HQKn;dWJ2B$#F~*_+)<e6ML(OLq
zbVDcBTy-uukFBvs_VZW3@>cJ7T{bEi7w#lME|DYk2}&(za>^iO;R^F;Dlj&gbjD5$
zvYh{q#88>^cAOxM2h+xE=lj3T;D1u2N73h>6FCnVNjE#-DcM)Zo8u7UkN=W`{(0r`
z4Rg{EtpjEB<X4A-5z=N6bQ-c8!>MBQjh*V5s|D=f)Wyw$h}Y`VUIk!^LjECbvD78l
z`CG+I{*P;^Z2m*SZK7PeLCK>C4O3$IniIcC^2d>Ij;D0A$ucj7Ce(2%eJ5#DG1a4^
ziVr_U(?!j?OUWpt_W4a&<uDTtjV+}E`?XQA^6O*3EB!-KP57Ye8>_i2V$1dRvvmH!
zm(uJ7>B*V=QzeS#;!!|DQj+w3rPBagu;GlE-;A5CMBT-LTg{5a6#rbk<D}o3jEGAG
z_FNi)G)>f8z4p1>=)aT#LKgNx4J}o@6C64rnP%-#uCEXJj^7F~#lEgkYA=Z^m`N4I
z)Go}NM)8BBJrlNCI425zdg|8g8_Za(lqw6W_w0>)UBP=!7g-wwmKI@M+!5wVVY!cs
z2%gT7i{}6OQ8YI<Vwh4|Ebb8lK=E?Jw%d<5U(Eko@>0^LnEqK0LvTJKfoI?e9VkU~
z;)-kurvIb##~<~~Z>+0{nzyCO&00BhAT8j^lDa(^S}OJ#n~BUzH>TRY1kAIEEOni?
z+~<J0K_B(p6SrI*4KVj@Nw&2mGo40Q@7P-^lQxNuL_)77`nAz4egIQ)yXr>jEmXE_
z)Q0e#kVD(V_v~?ny_<~tS?vma1uMX0&D^B_V084r9o@ZdcIxeJ<yt(>P-XlR=ZTO0
znxA~@f6Ig;8pKyHm57mb29N$j-~XTgs09WPf!G3$hXB#&zrA8E2|3`6bp^>sNw;UE
zG3?GC1C52zwS$zF&TNAxwzk4L_~GyA6<v=39!XTK&(t{uP8U8Wjro9^$&A26a1V##
zl%apEgjsZat;oEasH(YFu3hbv)#y;NE=ypLwT0&x7=Ar1h%AMwtIWHC#SlwuF)ldh
zpqKQO^;Qh$m!>T&^BeQf1^ha&qUaOz9;OO=QOEfgr`TFT(pv(7RwIgIRAhx35gO0I
zX9lBgy5$FvzJ!A)2>}(eQo7~5F7fPuSrta<QW9233`xgTuhxt=Zvl?-5HYV**9ia=
zVf!+<Q1#B((w6LkYz<Y89ADce(^K~aavnRY+aF{kd^(L9N&yET7qb1W^1(`%DH@+V
zI&IvD6F~Z5j#~qxa-5-MwqhC6Y=Z$OkM>0?ZJwul?(ngM2~4C%C+*1E7Ojj_e$K(0
zM1nd1gR+z3_h$MarrP(BG?~CTzjd3r7AUN6o2*^Uyc=(*4h1a$GfJ%|sLr7cG>tWs
z3FOX*Rvn+o%thjLX#OPF1MEWMiz-EWqwhIdeZrc8vCp1t3!fTr)BX1KvaAtbu376z
zPYuAkm=#bFIhOK{d$|^P!x8PS4z`G8PFu2D6*!id2!|xE%KL3IS<tc#&Q+Ou06dJO
zjlx)0$YMfx3#O^gqIJuhkU7phs7Mu4Dd|&1s#`Ljt&MOPL3qUzJt`J1U81i}1FU+R
zWHZ-n7m{t6$bu!CG_M>1f=2eKYqm&oDF2QW`6`~viUQqRCjd#)FixN9ASd8Uc78LN
z`%wSWwAtxkP=)*i<Uk<wc9DAppdC+2Hqqx!p;##Eeu>U>XwcOr6SU2U&%tzi93YK_
zl|FJ$WY)%NoBBOt#jt*15<0*wa8qDi>&?A8A`Ts-eSwz8<iR8JfK1vD&p)@7=X2vL
zYmz{a1}5nwyAu7Up7r0J>PKO8924OMg`M;m1m*w@WKisfIK`8v>kV?z$VAzGs#*Vi
z5yW*4OM~3QfZih?m}V}GizMt>eIA<DhpI;+1R6Y~3!uF@j@ko#+{w>HfiX;!gSz@L
z^2Mt@0dx*9hAttmdt{Sj($}1TAs&Lf;cNq*0syx$vhZtT=4*ug%=>RmJ%oy`j(=`W
z4G+DvJ6~`hHRe7#Sa<dw0povWOC(s`2BcB-R-?-e_^CmU1m}UpS{YJ{pNtVwCYSR!
zOkFawfS4UzL6LDQq5A`?7TL1W5`Hb;3LXlU)6HjnS|ks)3_?6L1z6@-bS7bdN&MX6
zB%MbS0IvM8pO2}s5eD~mK%c<SqEmB~Ruug%`e=|o?nZezx+WAV&x&Utsp0>g9H+8L
zv1HL^Tp&W{-GP}c3I@gdfaa#e7u89zBQ>_^QLuLh=MmgLZFR4AoK5w6Y7zs`#jp6?
z;2&uds+vc9=jRu#cqTTSC0l{MN_~BNp5Dqw^BpK^*?WLsr0);Wrvp+8aNxtMFY~o#
zHnzpeFiwJw3K3iFGtnlzUbJfVl#Esd0<DQpulxB=#*2&w%wVRG-YWnSMLYuNHhIpz
zW~&SgOQdA5nJ>INp{UfJ$32q<hRf^De>G|KDRcO?M>s=6nrM)<I@jra-9OWNl4V@n
zdm0qpm2~u@F*rh{<!T>{7}{xgtE453rA3#?4(y-f)JHgaCjMkApEnttrBl>^N>Ozq
zUDnaT(BTc;z=*QLJ%{=8O*(?`M;DNYmy?=QbaHyss_7as^`|0P04JN@YP>Olikp+%
z-@=xA?17nR2i3pnb+NG-1+D0r<(c6WN#6p0H=z1aD&}HtO>4tc;BUU%NgwfEo;JPN
zY%R)Nvb?s*V7#&JlO6mBD8NvQ%DktjNR1tUuEFRK1CLQ`!Y(oHUw>`uBpg<J=D)3I
zyp#08GC10*9d1xuvN`ta&qgz;)KMLe9xAjo^tH`G<pHpT8^*7^OZRIYKL0Q&Jl=OD
z*_uBr-Bx!4@ej#jPAEBVjmdAS?6YL|6a+>xf4H%y+rOcLgbA-i(02huAp)s9e(iC(
z!;Fd``;bi2K1>6z$qi?`oN&f##Oi=QAMUs#CJvY#GiINUmWFTAacg(mX+kh}qVk6@
z6kPHJub7Htec0I~y<K##CKjlWUX0hr;$|BAj}W9up=N;uBa2sTUH+Glh9b~b7SdX-
z92uIB5P*!1#^vu)V;Qw=t6IeLKO-ZPX*$q2PC+g^<@3lm>5salb-f=W_}L32p6rp(
zU>%7aBUW-bI_YNyg~Q-4Xs{7vZ(y1w6E-xM5*lXe&T0WuwAzNIBUwTh%srUkb$NGe
zn6QTxu(#DI!h&hMkL<CXJcsmSj^HEn!V(Vw?DX#O4YR5(7V1ARaeRm-;@Q_{HwPy=
zSA<Y8bxUz--`Sj^P@$Vw)-69x*U$oW53*I+AJ<>UI|V%3reL%OBuePCg>bxzo+pGj
zL7Ii02@#a?(_zC1|CR=34+sK*E&)<n&oYwA*egOGyJlH;M%?)QdH&k*A$%JNKyR|R
zJO>nGxVwdVAO}P=Zta)f4!<q|)RCM&y^#wkALslM`QH)j8c()-d;h141}9X8!Gg*c
zFtyfg9_PO*G@$i|yqyO|tL#;>+q2^w)`FXMrrG4|YE4oS%w>1-7W~{R$Kxd`6aycE
zFQXV@TssEfA0Z1a*^WJA?MSqd?K)bAK^2haLZ~ZbJxyjdbC)Em<QG{F&sH3FF2{2r
zIf_!?9X=`A6FQPnelUzFLb52mEeqfVeEhp9{<rxn4sgz(c}UI$5v%$ZR9UVvu1hed
zU@R#!(DW%`PI(Ff796ptxyEd}x6OD1ctE#ID|&>0ytq^btpW5seLVjD<W)-l{F=a(
z+M$Is-`F&>f1myuQajo&yjlT|qM8;~YUPRVt!tDNG{X`Hbd+b~_*<lx%*Jx*wZ!(h
zbg{6ES)Vc2CyKV=vz~uKNlGRM>7|tUPrxYKi9Z*pczutBPGrPNUt8wOXWQ7_$?sFF
z2)YHgr?^h~zYj4P^1zNv6W`&;SN04L#t0;@tz@Ug;Qy$`1WU$5gy75kq6mz63*Ol%
zWaq4;1xyw8zfNnBX41SF4WWsyV*E7GXKQ_y=xHJXVcZmxTBsf4PqAFua{mIXhc6f7
zMX?t{g-ah$87YyNKq_8Xo;Uy0d;c@Qf7Imk@FBXP&+w9%daD6|`^O3-@Zojy|2Vb+
z=*Jk5b?$=S!olveWC%;cGR=(7hY|>QYc<4xeOa=$Q{&0;8NCj)p|eYk?%CVkQ8)0-
z2RhnA#JPSrW)~m3-071d7_r<N6%ZgqzWC&f8o{Ka83^j^5?nHK&RsI-4kF5Df4Ynw
z2p2K&#j!JQlWhN<=w1y@1<&@w{U-Va!m)JjXpLEv%se3)kG3H{yofyuIW;=coxW6e
zVY6)e#=8wAosfqqVSQ)R{koS4BXBh9ed_WC8$UzC@6=nTUSX}W&9D6^lxq(*l2Jc?
zb`P{Gtjl_Bf7;^f1W$lH>GABoeAQy&+HSqBiS;SE+L>aLD}|w*z*a}y<Q&clSs-1d
zbQs0f85Fj4^>{}NzA$KYQy;Yj+uUHDTH_%#u~WzG*)8lF=GnPPQUtq>Wzd3(c+{&-
zVjA3h@9&?fLyCz#z#8o5IA#R<@P;61h#Y(^UR<I^gjAc<7!w$KlvyZZ!n5@CZ&Cf$
zHAmwt@$!XJQMi{x#@qqZg!?3c;ya(*c2IO!%%3Tg8<ZXFOF_b$b^2cI%d~cBeGH5_
zK`poI;E+f+p#uU^JeJLu8*k#=`m2+Sc95n$Fwh2rPgrn@;$1Irik^Fhw&*?EnT~Ae
zr@qmjqg+R7kGY>xJc-xrE69?4-oTN|^Ksl*y$>ci>7nv150AMd|H-8KsX7>AW{YoI
zz=mnwvDLx4t-{Msqtg7<v-Gfsgjl5B&a^=*s_GEb-(2;fmK{!KtusX(@9KP~918DG
zHZB2qTDB7_Akj!7N9^B;=l|Sn|M!2&Lj|7Xlz^4g|I>r~zyCY_dFTKCng07O@PB{t
z{=dIYb5IkY*aIGCEw7kR389MqXYV133b=G9(&Oe}*d3Rpe){Mqt^>e*x=(X1+rVaL
z%>O&LKkx$oJ+g_nj+fHElhIDP@J@i!NJItr&Q$tT0kjO$#WqjFnVal-0@;qMp@Ux1
zZteDy<Rbw8G`}zOVIu^vrfl6)@0i991V$>^9!(Vg0QM~oq=_8|fL48vf3<7;m+a9s
z_6R{AKWonA^9QG!&FDDFuj=Ys3c=Asr6q)kxrO>}8RHL7PR6V9{9)Jidg*!uOCg=9
z_y0<f`Csn@=69?BBb@*Gx&Xr|obsd2qhY!5;SyjPW(pu~@Bmhi$>v-hwL=fj7J2V2
z{=_06wS?zBoPGKO*e6-e0=`*Yvf?-6E9TK_*MPYwsnN|uYM5n;Zb|YT7l0W>GMjjW
z@hSLFMFN*UiT(&USB2Vk4G*z_iJQjzeIUpE1AIYt|JkZIG^E4BKIB~~DX-&RlJNb8
zc}{>F$NA?&?(Kq9QJFCvmCq40(dn_Q<`;i0`v1P*19gJ{&A?h7>c#)>uNk-v0qy$g
z^vat8<}f%q5MdyEoGcFP=U*Ycf_I_lfZBjR*~R5|9Q}SuT{iUkbTj<XuIgy2!2RZS
z>91JZwz3m8!0++!_*riB6B{(d*=m0UB#p$6&d*;7b7fb6o41Y$ssHxqxB`q9k>CR4
zDG<`pTB0iF|M-py^K-y8{$l96^CmYV;^nzwkFIt0vv$Dm$(wo%kKtbxi~k{Y{QFt^
zkN@msmjL#@x>9aYwtv6U=}LdJuqh_c2K*N4NRW=w2XbfP)jjuiU{oNq<F>Wq19BeZ
z=GUkn*yyZ)B=qr_G6fJShZyr?myfMS^dJe=v&S^_O)%&Hz>XdMT^ta9gW|r)wz%Y$
zm*yruyBDx7&c3q`82=FEb~Lu1fB6IS_+T|IdhPan`bo6)SY`Qt`pm$APXTPR{QYC*
z|Fua3cF>NYdsf8?%CO)wfgY&!nxug1Z@j>YboAHk(ir%K+?tqzNbf(xE)Ic8LS$!U
z>jBwgASs0Oo;SMx&HBIo>KGw14hd&{nwG9*a*ElR%|ExC6c2gPI2J=Y{wJ^`81aAy
zjIl~-PGKC#hNvHo!o@?LY3b04Y|E>@W7CW82i~J616))Nl$~**e}9uVk(~gyl#IsV
z9-vx20j4cfJ%Fj|Zob^>;iElk@z-r9$yN0q7Fyc(EV;^+z*<WW0Ce^B-OWq~ZqH?m
zwFte)q30;h>Bf(`CRSti=5cgSrzyer-j9Zp;XwAsw*WXJ2HriQW&YTzdh6)`7L+Kc
z9iPO`%U+b-FDVKkJ@Rv3ia0;L1I&tDO^>rSMLzMyxfXJ!_+JkhcRj8-1db_S!iEDR
z?T$KtpD)TKp!76!uB9&l%?JiM00X!S@bO6&n-@xO&nXmxk?eE>Y&zR@IZw0SRqQ6T
zTa6v}Uz(-*(2l~z&hi13fc#9P1)$r704sbwr9CIpzZF7OEu4gLay&QFK7EpCe0X^q
z?|)~0b>QA%=jjQMExU+vucXRTmAC;9nDdF3M<dDtzb(ea?zdI(04zpHb7_e|DS3#s
z=&o+cKQ2=v{r67%YlV*A-jUBU-cz~-_|b(XYw58Z$>tq~FrSti-?NIWv#8dz<I&k`
zl77wpyCo~WGbO%KggcV`x{=9i=laoTs#&EqpPEUz_XvR5E<}hHJvT5q3L*&sw%4>*
zZ*r)mKt=IN_t*}(ALfREclhk`Jtwl+RrNbbjLtpCnkk!K&RPiMn>JDAz!Se7P_6j!
z$*z79nAzjW{ut+Qo-Q?8Wb|tLaBi=P3K+(E4=faY|I5s!yS}9NQ&=}csqJUQT7ag7
z+K0RIlPKp1^|jMUymP`~Ni~4G(_gw1|9dJvU1_6_+UGVQnf>|Zz_npK>A}@v51Oz0
zj8uETuA@_McvgYD*dh3}gFhfc)m;n-iZkcw6n`Ni=|}}I#EFMnt)v)EO@GjLs}f=)
z^;N)sAI^>c{#;9Eb?+@QCvCP;BfD?^S47=e5DZ#<5CE)$@y+)J&wwsNZJ~B|(bB|6
zDCfqn>vK(Q!=HV1cXQPpmp~s%0EH_+G{(IhtnkHcJ_B~P_H8#<GF%(BJ{}E$ojLor
zNiWqqHjcONtidkslB;}rU1i%wvs_?V`kSvGR4^BfJvrW^mY^fK?ml9hJOIP-z2L2W
z;C#|n`q6*S76u%$b&YfsfJ@n*e(?uq$Yu!cuVCP#DlsCffU<^Nj^F<00z)A%p}Wof
zd!73Sxb2?_SanQPqKps}k8rvGlYh<&AHebPFftzNSenGpB5|eO<jJ29G%JJ_pl=#^
zbdxWdSio{p**Ieq7TKb$rUWYi_}zOTUz%GBZY$_S%@hY7Lm6QYfSF3Ycs=4ymR_~z
z9SM5$)yX6`iNImMlhiO}B8oZ%cKb}l1?_J;nwILM&Q(|`z!Ej~05*09OX=qte#Q?c
zsj)qm#azAvev)HrRJM;BIEQKx2ik<sf^&cR1a$#O5FgJiK$;+_2|xsAb_2BKETe6T
zC4iQV5g~w&4hXF8^aEJve{<IqawZcbXjm*`r1)e)GRo|L_RVzD^qgJy5X=+mzsJ;n
zfaAY~fXM?qf$3=sP(ZuI@4Cg)yx(9-c?O`ijEddr{Q5KPIq%T;H1~eb8s$nupfk`5
zD%8I3;KWB~`QSI-VJ#_^0kmix+s`d!17ztx6N+448PrE<TUIKwq=z4mI2nrKA7e%|
zEc*a5Plo2lsaoUcU8DbP<9p!vB_FrnE>7jAJ9wmQ*Qq_C@3<*|QNn3lTLFAWz)N;3
z2Tgsh7dXwYaepQIk(&{W0uO2!(o8PQxZD7gC~J^D!AQ$Au{4<q^Pl54?|(cjAY`;y
zK)<!y?ce*3zt$FtbF^z60A8(z?FOI|i#;NPKE}3O#<{oAx2<zHZF_|!^4}<}*y>ef
zoVh!kYflY5x`+0C803px#TWkd1Y+DDwDPDxT6pkdjm3YkC#wWvsiQx5{<Ay$IvO*7
zcRL=2)(9;KA4;_m&wTLG%^n4DO+poG@*mDL-+%?3`!)8?>qhz3G_eP`m#LiR6!0O=
z<~2gQLi~GX`nVV<%-=Q>@CH>j4&vxhKzQ5Qmn@l4FyJDRGVN<42gGD4-QLQo6#jNm
zp`uxl#4Hj`4FxO>CnwZCj#mBW^KFUA&Hw0L#NV+h7Q(ktx<xq1|0%iG8SvQ*eZ(4n
z!11*CuTYCw9A^65dUD}IlDqX}9T1+T4m?GeCShfwz&c>!otXDXEg-yA$voTNj)lHI
z2PkV_w3OD@cmN<b9P0Zm*<GnuyNjlYoS9{5){O*h9q&-2<GTq+V?Z}c?%#F|8we6&
zJApBAb3tH66r*EHwf=FXet`-(6F^I(mqKtel|>S%xMAN*=2p<p<W(C#jS|>I0Z~*Y
z-PYkhwe!dDU@`!N^Bq4^ZMY`!sh#uQRA^hENi+?vJn9h7LU<7j0~}8KEzd?aJv1TS
z@DJ<_4YtYi9Y9!mxq_8wfnWe?p99{O&T(Y@0H&6TX5b6EqPot0P*EQ;2wUmoDHPv~
z0td$u37IAY>m>gC3rE9jmmC*tsFN^Sb1x!{*cB)ToZ*|zbtW7_Mmh<fDaauF+1GM`
z9#okfj~{kM=HN4(J0@`*<`4#y@APQk8`_8WPZj-_p}))mvDiKEco8J+M`-jiC6|H{
z=(JDS!I0&EBB)<95)SgA_tNc+-rJ&)Yydr<Lb9lOoj-Cxk6YqoT6PFv54p`yP%#FX
zl3AQ3)foQb2h-fHZ*0(_WqWR<{{g~*7&xc~ghe&wbOdQue(n#{5*qHt!)R1u>4#H%
zUY2P){F$ropr^!lO>`5zL6K9(_qKJ%_bR17quw(MVMaLqyaa-pY@mN-vqzS^kXmRt
z0Hs1qlAB#m!u-zxz)oyvPrdko%*N206>_=n<o|jPsN^ECIq5rmO=i*8SuR}ad-oj7
zZA%WT{YTlC)9i)(l`syp)Zxx^fQ(4+cmaU-*;$C5N8Cd-X~vYp4LYWBy=X-m&&V2*
zhGs2uzoSMUwOnxT7*Jye`jRG=fWAU?F4|PTWfbaaujCtHu=h3-9EB!Pqg})uK^s16
z#_iS$^r=|(%7i@TAu<t#2T(srD+uRzGmgNU5KZDfi#aNW;{vyVZTmvcd*n%gcG*3V
zWvto(&)y<QRLpc0rR+G4b<%llEfVe{``SD!kOBG|5FPMONL;W^WO-=Zeb(ryv4E_`
zy~4@t58N3);l+})SRPj%B_!o<zyF<VG%_+eIUVV)An!w6o<6wxq!j?Tr~E##`*#=$
zVL?XZB8U)R)D6pF?eOf)W877sK!OcBfVb8#YA{s`#l*fC=yJ^#w7PkX;2b80)xkx_
z>}Z0aF;Wi@N7D~XsF?qKxt)i{nhsOW%5~jUFUh0|FQr8uj(wbc`4|$;qfhXkEZ=LA
zGka0Qb()}({+)$|kD&cVSv5{~w12PkAZ3W!)=?h4u+|YC&g80EA$oC|!^a{d1h^+C
z`EU4k79~Y|G9Sc;!4CghN!^MRSpy-h8xRQ`rLsw#zZ1Q5e@D{wNyBpsz;WIOcIra!
z0>uXP-V%FqzRPdfea^w!AMd+ou1?g@1SBU?EK{uq+S!ONO2h>DI~Qs*vaymqsmxtw
zQ(YdacXV}91c(BIjf(|Ek~M6es0bqIycics;^HmE#yz<^y@+@^R2c%20<d7qL1XMX
zlh`qq536*uMid>Q$NxIjKeD#r=XD9AF5J3)Y`i(eFXAQs*X3N5*RmWFQUuxTM$OPI
z1Fi>rq3<~a^3@*Y7tNTVU~7WLTFm02NR1IG_=^JKFYptf2=r7v>YRC9nK+$CYd!SV
zgc(Osogq>LyPU~K5;dJ!!(!%JgNl`+7$KGNu4dGygBn|Z?Ym03K6NdcLd5tZH8tju
zem{k4>aAqa$-L{KHF|QH(eR-inoWfCnTW&pshgUk$m^M_V<5?Aipu&qITIoEOd1jW
zC{7p1Rf_rsfE@}J++~}6zltfp&S_myW}>r|hR3bUv|oIc13D`2qcEG)wU7!Ul&(2$
ze{20tL91sDa#Fp!MuU47$^~2(Pb|6J4ma#RizF}ZP)xU-FpjJy`x^b~o%+mR;>kca
zV5~uH={!1^DOTel?%y!r1YfBbwPb(xhkY>ARn?U)SXMWOr&(&PKIis2blaxz`ttT3
zEKeKtQvW6bGoMjA65@1^eJ@W1`zqxXTjl+GS7`&3Xa*ZvW<GHwhKom4P{B7(AeqIg
zB?c^ScARx#>%)Rhi*2>~en0-L1=yek3cIKtcNdP_M<3cvDkiOeU0Ijj$-3^~pObc=
z;Qa<wZ<l>wGtw^tWt0r_f0=ivuTrmzbF>Gn5y;NJ>Rz|;k7rc;E^!>bE!Wl66nR4w
z7a(_BqaeWlM?QCa;Pk;!hQxTO=h1S)8`urRNp)XmVVMF2NVng!O3civ=DV27RDF*z
z;D02E!!)QgHjsD0olL1h$t<!1Y~|eQ2O8cv@bfZeJ6`qaK(B*h9OfuX4(36(;W4SQ
z5(Sm^-we1DPjm^)o)!rpTaY#H)-757RdcD0z#T7A;88Jfp{H1%V;=0%(thb+o{nIL
z`Oui+N@ndUj4i2@GR;Xl9wu@{#H=Mo>S?AqBGaHU>E_4Os(_<=B1L|{6aheR0t7W5
z9XXP{V7<!CPckV8^y0EG+s@%3jpLfRyGZ35Yqa&Oox_=XuammKT3i?6ho)Q6mCEEt
z72~J*$|=3V>XHwMZllc6nTDuVh)7snbf7fNir{|zWK^T@=5lYrQ-^N~1M)_te_@0}
z#QZL?CJdaqsCzw`e)XD-{ZJ4^@hFAtH{?QL95>|~wmwYVfC>i5L`J!>nDVJeXChG9
z=vkE}6zpHHP$%15Q03KAsVYmppP9F3C5}w9Po6HXbzLjq<v-|Fn~ar-9R&qHi)ClW
z8ST|K+;ll6(Kc}IWfs^loSBRy0ud4Vx&Et!MM{F~ww7b2moa=F?ac9c6M>BIwC6qI
z7Q%e_uZLj#J@WEjT7KN$vwwRJI)J>|<TIduEx)DRUvfreBtn{rRkmvO_}mvC&<dJ<
z@=uum&n@Oo#P7^5kHFm^BY1OhA5`%-O_54Bb6bRwR|nRc7ogZY>i=+xq#dY<^x>LL
z?=f=vz}zUJS1*v7>}Gfci&J0i^kpjPcYYBvPi(LNV_GmHB;j<b0-EA!%7u%6_4HP-
z@1^gmve+u-{-LYH`M&=sgwc)7epv0)K>bEDQP51Y02n)XEre(>stLT8jl}v$;PC`w
zfhV%w#__!~lzCSX8UjBW{d+fBJg(Nry@YE8@U4Bf`3jAQ^T!ST56<2?D(bcU|D{2Y
z5*R>gXpjL3rMp2wa7!rNNSA`r4MPkKA|a`Q#4sQT0)pT$bO=gIBOua^=N_NEpYQXW
z^Ihk6*7<8M*V@}HAMd!X_x*mquGguVN<yJt-2Cvg6^t)p%Gp%?D#x!JcxREoooV(|
z`f1)w!nm(1L^Mx??0>5D&N41fy7p*bd`1C>?zImnyEB6BMOtmBco8PT_my5IKGtLA
zkSC-r=-^d2(K_S$Y3Ox3Ucl<~hjH8Kmz6TIi7$;#<~3GI#v^ubh0j61i@Z#_e0w9-
z%+9@46$j?$qa32i778DkXuCXEk!OAK9mRR7Bm7ZrV>JH|m9GyLke6WJ5nL#Ak1}49
z&TiP=is|jBV^Xn2%Avt&bNU3&1)Y07J%$@rRnTkJQ|58^iJiz7vyTVQgya-{7-UBF
z`^g6Ff==+VwOTBAAw0tC%^TJ~t7~nON661Bnfzy0I|CV{@#WymD1E^qGqa~L2}P5P
zry19xIP;T_jWyMzjg}?9=X34d?l-@{uZm1bmi=d^QxG8X5Y`Mbo_ho5q$j$W!x6DE
zkCUdfrw8dH#BmASev(Y4+f=Ou`~=pJIuU|zX;NNa8qE>}@+MNAjWn-I{+KTC6`SkF
zN0R?U6ZJ^0k^OA@9sDdKKyGmx=m4(%g&J0}8OAtiJt<LVgm!1x88gKZR?}HA(a<~7
zWsbNmzE=rW7Sr|RXOG~|vY)NVjxgMGSx~ZI&=KN$Z==$kWxvoW&9=`!No#+gof#7W
z3nR7@*+riIs4#AY!#G^Jc`Q>Me1);qdItW{Qt8Jf?%>5#6AQ+xaO+*;)f}EJ&Y)z`
zQHRc7plzF6xv`69ZGwJ-tgqjn7XECo@mOR)daJ@&rku@c;0D>K5>r=Y+sE5A;%U*$
z@ZhI)>B;X+8tzxG2_PocZ}yN~l#1g%kTT3iHsqyF`jwaux1uG4LhY>;um%qEK~GVg
zsYfH0%WRdS>H#rQ9AQ#~r?*cPG0U^hw(B?6NY>emECTFh*pAx1A(G!8ym52Q-G4HM
z+x?1mDS+pfhGK&iMcgIxg`iuhWL3$LWYXlouw&`z=d-boO+Kr;bppZ=5&81=;{nMB
zW%txAwVJv6=Q~XLO!O?f_txH_aO?k)P(Md2JGT^V`RL`Q3=^a!GSE2jUMz)6`!T&3
zCZ<}mOu4fgDj#|&KsfnY9H(q_p!ThN8cFV01<!?f?!_119>;_<bQFIj=+$`WYco}w
z7m`C=6<JF4?)OBEepA19Dkb_j^)fi@Kx<OFO;JwIExjSpOlQi{t^P}8(BZ=@ogI$0
z(eXI=+!{RR_s|pe3J$VS%hDztA>Q8km>#i$)3I_xUvrHSkp#2RgVX{^JTV-gISzy>
z?{Z21BkD=vQSm_cYjMcqY{o)GX$WC>!to1A-|4FTqvS%s%nkCVzIMA7rjf;b+Bq=J
z_K;-(%i1`cGzU|JSZTf4(@DGdp7=_+18y8qrM?U@`*r0U?MXZHiegCy;>@_d-57=Q
zmttQ0cl39zm#}FPBc9!zjKHU18|mlD@1d5|6~#rI@@+VX_b{vGHy$s*Bc{2CP2Z6s
zrN0moM_y0=0Mq;{%DPAG{`OI`{KH)()~Q#<7d6*?t3lZVQ6PJOa9N&g-h4b*-!Lhq
z%90+)N59^DS8s^EH__#|Jmj?86Q~i4)Cj^pxX^n(S0DwNJ1f=Vu``7mv>_l7aeG)T
zIw<|HlOY^0YSo}|XvK-1Pak?i`MSUjo5?p*i%^`T5$wT%>zfz#9vmAO?;nvM&zkA;
zp5*n5+Nd3?p#}e^j_I}>INs?lldaXkK%Sw*_D(vu?}2=tDKBI0hPovZ1qZ2fzAb~X
zEWvA#A`wC^ZSSV{+Yhc<g7lSJ-{ttb*873>B|g9bOOU0oiLDCSKXGDTI|#&W0uL%I
zZ%d^T-hu?NqiRe(05duv<ZceW*f=BtrXNFgvt?*WFW*{))=O01A*r{F%CzO3+S)?%
z{Z;H$W>(yH5s=j+mkaF?!@Y{|7`h*-*4D00=m{M>``KLG#hpxYsGS;<5|v}ip;4~(
z$>v4iHkjHg5x^pfkEGjHVz3gqCaOU8vHXR1d>5`(nkc{f1A%TCjFb<lI3^T-&gOfX
zl<E!>`a$3CUQL?7WHkp<A&Pm$JdPS@e@p0-W!;Wgnl;n-u>K;n8t_Pmp@iG4J~HGW
z=K6{oH3G4gpD(pa%YU5@2(Ri+5!++V;pAd*COH_o8O+m-w~o2yP&7G{`rD~4{|!`i
z*_JFdj<YQS$JwK=HoMs{I(_E+WY4YcRpmH+n(L}=tt&X<9ahG%DpMk#Vu;2?NwVq@
zu}DOdl+Ub<l^D&gY)}Unuj1x&C^|wONj&Lye~M$4=i<J$@r*d>w`T=GhqcHEk|_7@
zbp;o$IF^$z$su~fq<vE=-_MK={}PdE-6rQ=a@uq<hCQVizAv;CQP~syp9)f~*V{&=
zW_Ot^`_WJ#QkL7&F0qZgFFchgsegd2U@f^;7y%1PVE-c=Ik_ZDWhR(2h-=yzpSs1W
z$=<BKt~6^8+qzGnxO?0*7&Kr<((QBl>{M!3e@eZ<#zOCJ)hM@rEnyj@vQL8M3ur%X
z2ZE-3)~jc(&`j>SbLz7h^#Uzb&>o~+R=vG^Bn{%PtKQFC{g_~{H+Cnjyk&OjLdFD^
z)a-5tgp16WG6uLyPdQn9H1Ux7@v7Lh1zo+DE0Pgt^CIl<K3(=ZGk{<nS0BE4Po0@D
zV#sm#r}&iXXqFXOZEAhnE%8wg_KIu7b{H^8>eGs3$?YCz)Wd;~FA`O#@SF$jaf^9N
zK1Of~Pho2$R{?{;<i8c#N{xA@x#Q?7ZxK!6U|3Dc9z2`@mr?0B6rY0OSiX7FpEWG#
zuG5<(u1kZ3&>>Hts!!_qK&EqK3tu=WcX2&$!49-gzqa9si@2gWoT?~Nt>d10?R<Y!
z|J~GLJCXvI>KY}S)+AC<E16pab5($-geeu5<_9o=1r=Qh<$G9*=}sC_BE#n^yKAGo
zD{!MAm+Ddcvc+m(12!3VpSW{hZ}d{AD~(Nqo4)?bfB_Cxv>J!h>NR~adT1J?VI7C0
z?q#B~0<#1aA12sNc(dMpZxd&G*D}Ey(WNo@j1Cv!zfOO*u;y86lr(J&g8Ji&nf`q>
z*srCRZ{N$!fNaZ{fajaKyWj~Q#HbZb=WZ&R#oTH)k<i0~jtervPopS<e?>&oT#&_|
zAcesZ0u$_V0Y?a|rOqVmn(tQ}#%np;gb)tX^oYJWE=k;)D^D-0{(RfCC>%0#Dv2f5
zS(v?*J-Jj$m$RMZN-Zp0tifhITJ5Wi)g>vvi@ho0<8w-~q&}J)g<aWC>?C67<j!c&
z6N&sZSh;Mu!#XIn%KXr3`9%{{RZRGY#;5P5n#5~&5y=u$u3~T(FyXeo->xq=B`(Jn
z84IVt?RX$kgV2<9jmK44z`X|%t7TA^QZ%yi%37t6TB2fFt^H}vwMw_hy8NQw-SzU0
zz2-$Ib0`Ir61|U*z|>d|1bw2`_*%pdFNZxZdx|??3_n-RkEQ3JCa96UdjyqIUea7{
z58r57&3Cbr7GGy)s@1%6!6AgcdI4iz-Jv{dc0McA;eWDXg)U2fZKi%*Dk=b)`|hjk
zr<Jz=GLDQPxtB43Ej{x8ReDee8$*A)BJ2l<)(KgdmdQ<{#0mJ<u(P+>KcT|xnGWKk
zf@2lNZj-2^3@{Kfo&A6@jg?;i*1{LNIIoj$DO%WHQfNh-L-5_mq*~+o`u0GBGV3v-
zo-<Vd^sjM!#raPo0{$iPWD5lL373iI!!e34u4{n4$;q&KvZLNPL_-=XQ%?RX;fzXg
z_-v>^CAK&T%3H`3&esUcM>#R^=`p}Be(Yqs&snJ<ydtblC-`y(c#XgROZoD`m%ENt
zzKEr+h$=t=hu_wUq#kOvCAyOvLKo>vV6wdF>)i;pHG!HJSBduN626#dwx}>2FfeJ7
z4Q4P}IT^l7u^;+8d{|@CEty8F`L(BrnTVMo1S9(uHSE}6@JC_ZHy#x%Ffdx1u0d&k
zGdO~_z5J}^?zkrrcIw-Idm?*NcPhfu5+x*+=XR?)Vy)pTS2$}e!w|37O{qJxa$l`1
zAw$h011U7@bHf7+j;TV|nVnS7mVF8s-wZU1>#RnM-dv=F*eN+W2Hk<q+b2ZfkHiOO
zv^!9U%ZrSA*_~@-8AD9;?d0b_QfK>-yt4iC68pD~B}RhFuUE|T*`SK=QR##5XJbpE
z@v?N>kw=LtX|sIX24kvq+dk`MIu08Aoukfg7i+vUza|^%sOy##Jw7xFzLS>t>Wcbu
z5H)i#?PW^KA=z_ZRbBNWTCTj$-C?D)9}MsuY2(U4uv-D8XziyQpH3oVuN2P=H-_L~
z+^25yvu!f+D+?4b9z>^?LCR@K@hvmLMIY^OJ=zujWy&8~L_2JN>C1Nn6O%sk8Cne<
zoQFe%i_-Z{O2>l-GXqigtDoacq1pt_8q&{GPQ_fED)L;c`zFKFV+`W+U#Z3L9T6Ik
zDpVX9x~fyx7eV+r=3qpo{IQ0b>im-Mckc)4e>u&ikj+<ZyohCBv1jLgWQ|;9r7ww%
zGgN&rxkS{mn$z&3;c<Yv;H|??1wx45)>BS-@;=N5@u=jHxzZN#^=bx>(!wOe`SP`=
zNrux?*@KOSM%7O(UU@0SdsX-#S!wHq4H@wp;f1XBU5&ep?mEVxtA5osGlmyi--&cw
zelRsT01YVGobgRO?r9SjEV)ykgruOVziYl>9@6tKg~bX<HKc4qTPuKVVhrrs@;~*w
z(WKq^yt4IxVWVk~P%u`5I!Ajo{=rfb$EB)i=#8LvKWtc<${8kVe10!HTMrz&h0ZAG
zflazn)q8&9d<VodwLoDOl~eU#)vDCed_K76j*I2lgJf5~i-QRbQx)+KA0_RpWp$Ac
z90QasMO+nm#)|6JwI~$Me`&IyBOp4(5C=~08$OWM;hS6mX5%!vKgS$m`6@K?j8Rlf
z0ntsGQ!ky*hKjt$@Pg9oU-6~7CMW5@G1M&a;t}2EkH3}v_4{sS`Rl5wl9$dh8WPb}
z1_X<-LZMC8tg)74JJ?jz!?7JyqJ7#^wP!=&A^Cx4{jw&iye?uFKMs61x+fnMqpVLk
zOeKB{*S(Y}VN)dLHMFBNax(*Gf~Lu>x5oVW??0DanC~?A$qeT`+EYv?j^!J~Gsqp@
z6!AqPlN@c8E-gkM4@kW7nG76?%WRF7=5uMtb2&%w;*UO|J-eRrs)!`Oz(TYL{Y%c4
z>?!fM4k~Q?6eT*tumdl#U%b62wP?Z0P_3Gpj3i}Hv%`3N+iv)A2%TYIW4bRP{=;PY
z@du9nVvJ+UXi3|1+7sA%H^Ut@NjN&ME7YCgLTJN{qnys;NglQb)>iC_)65lbh{Z8E
zjr^m>S-Jnw<4lri*T48ipqZ@<udaPkoxRP{g#M9oqQ~k~x8H-FN$?q4M9#mj75(dc
zvM3aXNBBRS{_R(lL$!GDP-c7TTNYcn-URNEc3X-+o(c&Xs;Isu<%>Wy!-VR`3=JH5
zD_A$)&;He;W!Kpqwo6^Aq`<*-ES$rtiepy<tx(M<Vc=opBg%eJ%ZG?*sMXOPFdzXh
z8Zs<*=zTQ>NSRF;>#i68edvG?_J+qZpf%<~4pU+&v%&!v(TO}udBKO@BqGA?U@HBb
z>Jesv=>V%@JUn>x*ZJ=1)Y|s$@v+>Err0NwPOB(q%?c#WwR(I7%mZEn4O-r@XQvG4
z5buvR6Y$^Al^)SD{*m@l*pPebNTp6sPsdGL;?Hjsr(vwss1a?sWS~&kYkeFU-orJ=
z;iZ+Js)}J>A72;yY*3Bv<-0J)JWIROHd;uaDm1SujaP$ILDPq;_l)m*044%!S}IL|
z(oZjK`YDvm1SA_eU4Wg&s&E@r2_8k4kL%cQ5o)Ttyie5R$cnleO_!{!ix^NY(-1}|
z_rAbQyIbg1m_L}(r}>KOMyA~M#2;@6Foe+HLJHYemF}jh9br_FB+;^nrCtoNU&Kz7
z?q3ON&i6>*o9N==NNkO8X}x;cbu@DF%2^YH4ENHHXpN<nQFgPIknB5R+dnm!ZN!*P
zu9t6SlqaHoy#;>*2;QY&5zMXuObD~$6_HhW<6x$-b5DykCIMGn4XMuRnA<kx#D<io
zv)Aojdh-S~>|oa~f_bzDluoecx=<6&t5i6z@AQ3)x67tg*i!%5p{vhF_HZ4^GS!WI
z`&3QNI$!4lm*v{Eo^sgilTrnR9Ctjchi>UjJU$ggtYXU<CnS6BO>DYZ599_i1JlC{
z%Plgx#5U{2Pn#Fsy<%F+hDsQJNm-E5osvRcjchRE#W~S8vgLeQ6nIZ&=@YryfDlo4
z@B)H3apCQNlz;8V%m#}xrbVS)<%^hfl5fonUu($XDqJHhANoCGCTKV4Uci=j%=hIN
z^LQ@4l%O58B-De1MvoBscpJWzE@=Euzb#%JE3mksIQY%8h(Sv;ht@%2CJroiRE^&0
zIyl@@5-DV|ouQmDX={JJ94}DCu*_JxF9J-N9SijIxWIScNJ5Kr9RB;XJ!FU8J2nu4
z^^xKbgifcjkqS0d2$$Z9cZ;;jnb^j|T^1&wY_f|J6`ZYu=gdr+q}fs{IyBySJN#0%
zb4360^G{C*<*v*5z-Sh4i54^Lfnr%0ECP>RFr7Dwm6U+wwrHul43QQw%@pm(J5wF@
zf=z$BEFZ7%=!InS0Y1-V3n->vFTNwm&FYHMP3!0=C0bg(>m&8PuwWT(0<uJ^n^62E
znIiPfG;P{kgnFED)dSe<zI(78HOVd>fh;X^(@2&mm(b&KYr;wxqBVWom!KJWo=(lo
zuTn{fR!gClsol)RWNg<;Dm5W%qU?!aIm=<UcFiBl*qp@9ot(+$j>IBKxPvWtcN?Fd
z=YuM%!s7PF)F-)zPd+Kl==r5KTzz>nuiVx&Z^>L4FiiZH9<-m@O{!LrD1DE&r+-Xu
zoZ@SsX(L76TaE}eY19mB&_0W#>JOS60@fjifWEWDk_4OKXqJf+xdod|#gi-Qn?<nb
zGC!HFn4*l!hd-~6aNpa<&>>At)vT|m-*-bf7xZZ4j#rFjm7I2ZrC(MhUdDr&%7#x~
zsdzUzG>;ktDxGG^SkxItl$!m59mcTyV!G5kAgxGAXR+nBC5)G>{9M>L1qMy`dz0Gh
z%LznlITv^N(hZ&gGC^hNt<d}GkYQ-w$|9dv$`jbtyY5tk@UX|aG8&H~_8+3mlxMO8
zAZf!j@>~*pU3{C&F7jwX@41OOl0m&k7@|^F1yN7uPr?7J9BpbVp?JEK;w0%u^G}A!
z&@(3MK7B#8dPlGIgD~jR7}lpwvYR!X3*Z`enLwPKx;+<wqXrh-Me<+sXmR5~k$3JZ
z=BxAEBtuKPmN9bOv(9mPfi%u<kS0)EVMnlHVA?BQdrSd@;hH(;nfbb^f_CtQa}GJN
zB-_zLB`GZ1%E7vfua0h1K*Fy~fN^}W>#qs7kttN33XWY;=Aal{(o;{S$Pz=p{5cF<
zu2J{z_~xy3B*~vTO-)~|rIzD`96J}>-*exU1U(f1Yeini)Ze(fy}lj&&EJ?#qiv&(
zn~gvD&r2D|J%tz_>Y~;9_=y?mHMbjv2jcKAWZuWq9_!|oSZW#J44<g{bu(8?toxdV
z0t>5`*edtz4Ls-)=Opo|ZEJ*C1T~=@fgO*gp1p-*Y%^%%DDdX_Y<DDo&>fGZQ0ML&
zr*RZ;&wLdwdDN65(Ntt9m!BlAeL_wOsDYJVYTW~XlVoa=d_&@X2o<|&mH;Ms4V6+l
zRDVdehGy8Pn0!NnbM|ue$UzJbo+3pa&iePc4E|w7#9pYj0yQRAB7F;0l5*-`;DQnJ
zyN5#_N1$z$R9{2COJL(CN}1;!-o1ho87e7k;RH*Caq0-IEvSpWQuDX1ApY=95GA|5
zGFZT-s!8T8^X;H2mQ2YgIGi@Yd?H!l@)%0^XvnT*<2~9@DXq53#qAUOHGQh@?3*on
z14@zIg>gz$T0F5WkXxqj&OL=VOcA|<g?Z;n@FS&?cBsAm5Elp9+|UHblHq_KMUz(8
zMcIo#sNKV|Gz)dYqgTB_X2?7#@m{njjmP9#7A-=F>=g30FQ1ryCnB9viR*TA=4Hlr
zwA{$6p8G=s2hasYDqKFz*Vj~;NjRWi3mzAk2HZ0~?l%Ad2ftOmU1bE>JoH+j^_!lu
z3^9{f0=ietw&#KWumG6Gh-+>T2}+wz?rrcZU`@M2KdT1>cX_yVpqH|mwXFV!Zz`&g
z`*9_Y2Az~051wxueqNrGik)XfF;(tKH>MDcfT^BG4^4YB?tgt$@m6{e{V5P}7G0=@
zD$FZU5)#;wqArsjMD5rD*IxhAF9f4y7kQ6Yi1^#X)2A}8p7Kx1xN%dCX{DrB-t4a~
zp97c5PVD@3`Sb6TQxYi8k#+qg_Zw(IncTW?+I%H<!%@fPq%m_rZK4!ZM7r(x?>nz;
z+?4I4pJhE(Yts#R#WJQus=Lt^s|oKIKeX@RbdvBh0Rfe_K`BqV@u+`Qy4~5i4uP2r
zeR*uC{CsSSmU!!KbM4j{(un3bA3dSOrS8%>N$3*fbDzm=(lIru?BI=a$b?ad+FYNc
z`kW?9R_gTW1JhNKggV3L-x@XBFpY&>vwvNE%ksKy?6$Dd<HGzn9e2?y&WJAZjS2v%
z4RXOp3j}1B05)dD<~tY;fvG+P@S4J!Sy2AhR9}%GFKHM;j%i+Y>{3#oc@sR)&I!6`
z>n=r^tmc=6ggh?IXo&|2^|dBsNTnt7&SU$V&rGRuCMFF<C|Z+UBVDtW@Gz`RcXN3v
zs}<D?s}arR3md^$7dY1X%1#0m8%`EGq+R{m$&=D*`Khf?H%tjK=cc1%q4=bonM^97
zR*k7xr;J(LBE=^h_HgR>uO~uXW_OKTB4T4ZgBv!Q3`P~@hmEnCmN?~R{H6zQ-w9)H
zigi2|*oG_hfXN{q+hg&MbRT>5Se$(3GyXy3<JZJH;Bo&F5+{iEPVI?R%Tw9mw$t%t
zM?==AF?hRI103YT^qk))j~<3GC+6!cmE-V4grf@p9IBRB?vYRB>q_u`>|@GQfAR_I
z)RYJrbD0#{Yf96X|CY}}oawibp(&@twBqWnte_F1@qudw<+5*xVxq>um+nG_>Qw8N
zmp#1kL$o)aZDo@0;y$*n5#NLCd*8>EUE3RCddC9n0PUrTU?t9h9^%`(oUG!TL+ahg
zDY=QsN2_bOjCCl+{*`i$dDFQY5>-lein+x_qZ({|?}^_dWt~P}CU&i{pAwG4P*Q^n
zIzjh$0=gybU`w3&8lU?bHJcSJ6eYR>r6%NPG#>ip?&Jo??Y<@LY#5dfS^-Dj2?KcJ
zKXD(i)Qu0!^tEL+wfsS^s!7R#J2}}84C@b1*L>@TM?TrnoBCLj|ER1RK+1D9+bg7K
zs0ooz5?Ho;@fq;1ApcnQdxU3vAZR}ZMeaMJ+W}n7k1qPb%;ZU~>md`=*<M)Bb2nWM
z@7I4`@Q*DCph`V&lBlbwT1JcHmFr+Tj9M<miQ=TdqL!FP@1=4=sD+S(ra}LrAub_7
zd+<G>PG$d0hfhXGwSN2)fcf`840qzdiOEY?DW)$%l+0c%5`T*Uc}1RV(c9UUlyFyb
zUOwI(!gu$siZoRCb*N4m%Y#E%_&ybX_IceJ1!IiXw--`An<oox;7FVJ9>d1EBwxzl
zHYjD6E!uUfgdME~Q!Sj1m`5v9VDxAw5f^tKglkNmy=1woPQe#U5Ji?<w4c!>*D1;`
z-HzA6UlIi*)RN;j)Czi1%jp9WPg&7I2zgpO`)p+bWov<61N9CrwlRH-s*qvgnD!|g
zPM0nSxMhl=rJII@LZSCbU+Wdpy36$K2ck@?pW958v_<@tiPpauQOX+=`T_!60_pPO
zfSJpWxqZr*evb<Gp4G=_LXOkw!JwZ$4Vo!G>5}=B>jl1{y<uY*j+pu<v?}}d2nN%H
z(o5>y;D}2lh-HGA0Q{v=v-n$<6p2eG$NfA$$tS?L$I9)feP`x<-t4LjY`@5fd+>F#
zQkd$~^@_E$fU3$w@h!1?($B{PV%>F5o5jxrFiG#bio_SM2k``P$9w%C2$C*w&cB0H
z?kG&OHS|`NEZ2`|S2vGn@0MUME*dL$O}QCwojU0x;?}hXkS6&tOT=3DGa1~9akKAz
z{)pEBe2RENxi{l;#rz9e=|#Djn4zSj1B}}YX;-Xt*us@_60h5b?W9T*eGK!qN+_;|
zy4w-Z_LYS7<EJMR|0_2U)}tCKk|8B$N4Ij@509)v>(Z7zNhC6Wg$$D*g6OrsGDl*8
zRtNGZyt%8<@j5P5Q+zuEP;3)yW~GF_3H*M{f;SHyi-U9ZYify43N5XI$xra|$dD?g
zuT-S@;ErQ1;AO+lG85(I14Zr@ws+si*J!;?TlOB9iU-(n3?0Aif4eChqLP{JNHi7u
zsSXI(dLNWENWw^zao1I{G;+Aj*-s6{JJF*(sx&xLkBW6kZeRRpuQ>wy9-x0a7mN<g
zMlY*i0|q^6ZuLI<?ye){g2){P7BOj|0FfHq*WMbgKBoKrmBzFO-L7kD9ulu^I^=~O
zP-As!Dzi0us6cIZ7&Y#`-CFm@7foV8241tk*OTv#HxG7r_V|AkH`wQTCU|?%jlf?K
zY`0#lSq{-;M6Ew6ws-ESCg;FWp6wbU%$<)RubdH<&<d+p9j6#hynRIv$2B~^O(u;R
zyaMP5$>}%$V|;-UnEZ0|qIVHR9A)3K!leLRI4uz&PXnw~<$5~@E7?bHImdV;$HyDV
z@NE$hRvsA00F`om5?=@{;jO#O41oxn=x9(YUYh>m2|1_EZmhEfmfILaXa+M6gpaf~
zybDjBjMMD1wvqgmS`zKcFJ0HV&V=o>ccVF)Zrr7>d92x_=v~5WoLMj5q+X*I@)K0F
zCpS24s3h#T9=}Uv`&FuUH|y@MGEU_idcJV>&=QZoO4Bx|{=%S3*ICeHn*+-1s&e0R
z5j?3yymDD=-x9T(mT~Pp^xWHTKn~=hTJ%a>+*pD482lx<%|9q_E4C=|w-PC~EYwpD
zzGH`a<G9h_v!H?-heyNBbz#H$Dyt+z?VD2Ad%b7s2C7V@iKFj^SJ;_ew&>WgPrAuN
zhVq^0sZdVU`*J1lykQg_BzRZ+UPh+0pG-2DHyf8__s!kBQ;-N=3OnCPM2b%u737TE
zdx!PycYm=AWAu}({>(R^FTKw7aA~l*{9Hnghnv@5_^?dz0ZD{rOxl?Gy~&Cs69FsN
zjt5F@inFgf(MD7bAM-?BG{hu~c);{`L?L^ElQ6iTGqW9Y*$3mquJxd9o7r-+S01om
zqx8RGT-&q%A>ddcsjaRms!N+Ob%PfGL<&vsIpidBhd7mO+!cd|#Diz(;aqh%Xv5$^
zxkNl|ILXIPdhxs^jhIqKc{6Zg)igB?P42%&{>a+xCQ{wi35FVEL`kc@Mf$6M7)J#`
zt3{HFV1o+314j-1D#Z;*I_&H=f@~2-z2?Gc_%e_~Y#5TmM7R1TAUNe_Z8SQ%3N`$M
z*<}2+v1_PxCk3lQO}e?dr<zA>t81zt_z%><>(8!e_1Uw3z5dW^%(`G5E4cyR`5<7w
zt-6NByy%U-+cZMHMw4x6%U{qxn^5c8uq3s=rxt`aSXk$<DE;*8jbfmy#pgTfDLClU
zD@Tmgy(?cR<@ON)rlD(})zxMD|2+M~;*hMYeTuFfF)jaNZz=~Ic)dSYr$3X;J*O}4
z$r^3VbI$ySUXhM|QLh+DtD*$kYQ@$^CR6^bhlalu$!RNvWPI=4@@R})52q6TqD^9v
z8e#bLKg?u*J5ByEoQM*ra=^WhSnH1!8ea5>>ZIFn#XE^Nc<3m9LQ*iR4J9^%hiWda
z&!d04(P~JbP&`vwDgJm@ntFq`9!Y9L@yJ4&`n%%+dIY<C*Pfma#mJ8}8J#P&AQ{-|
z8;x4N9GO7^UC$>R(pv`)nqTYk8W7Y4C+~ldcy!&)bYcWnp_H-1>NjR!z!agq-Ak%P
zn<E<k7foCRgKB>KZo1W+qIl7HHwy^(d7tG&Ok*x-MROq;#p2?sOSx~L0y+jvltnGt
zf63WQ)DL+$50w?Q4%P?NIiEr0ym6W8r3B(*4Rd6v)rzKz)tk!TlYtLT%z|xiY&?Fr
zWd@;m%dF;^Fe#biSibkLB%ANfmO)9B<^@`13cDOilzs+t+8aSsGZT=a4kxs@HJjAU
z>LJ=pna?^gj_Myov_rlaG0k{=W+{d}In;fF);DmkIg`J4wDLz6d?S?9GNBIqm#wiF
zgUsXjth0P@qB9DpPQ!Q2sMRRqk>k#DV-63Qa@TB^hM|}TLCy~b06~S%+oMGHeqvpM
z+gQkCElH@{POPUBk`5~nPR=vbhqbPz0CkpW_7~9|jX`2x+!q7jotzJE>8=|}uRtp4
zshWlQ`nKWhKP@N2ZU;yvP;G<fvGflE2D!D1ASK0I|C%Eo2!nYW_6lVciJ7I`ZR!uV
zB-A-_Q`;gar1>bG^P=$_$9UkfPtOGJO!6edhb+Gh1R!=SO1|^y8_IckD$PDF5^v9(
z-Ge8!yQ<@m3I0b*`riRct4wJ70)*_8?F6miTASLn=@nJ7q<Tf3DWdYey6Kw)V73NB
zCBq~BzHk4S<|F}orMU9^v)tiVPGoyeV|qBOQ`Iye5r$HdO{;I`%c@y_Y{!WM--H$Q
zBA+_qCLTe_&Aqzl)K^mca`;?^3|MiCDsBB{^%d;iPW<#zxL~)fIp!7M$#a>m;0UY5
z$)45Kh&2Xrvu97MRs&oG$V=`(wqgB9jNyCWgg_KPaZ<Z&aNdFE!4b_+R}x8JxGSDs
zRcj&;H%zJg(RF*=q*L57RF_WVCBycsRYcvjXr#7XK&t)LK)r9Ww*Xnm9rtfVm4%E_
ze;L7>M{e9%Fp_}Uw;XrfOv2K~^>3k{ugjn{?ft^e?mMr=)vQGEY(?lRzlfY?f8vJ)
zFd@a$c-6+I7O{yS&Qf%uo%2Qh&-XK^yz>3tsqp1{M%2Zy=tCG<#j|uq#eEt)TtCA!
zki&Av1txhvk)d%AboUA^BbWurOx53%wX*v3iIi@=cmm~8v<T%AxJi@JP=I$q<7Mvu
zJ!+><Y)~hwIZ5}dugop|*2txu^^JjCiNRp{v*L7hFw#Ky;pEX1`#0S^|60^Uk>=I6
zwzgvv-shdVPl`si`sOFyUzFSiOFPanf?QlUC#r!@Ak+-m{*`n%->WgcM4(#2AoHxK
zg^h8YIDm(6^2SzLBIUH>i?j@tUR!xqZkA|Rl)9<h*7_jgEx<A1ZEQDylq7A?>Q|I|
zdF<irGmo@UguY?u5La(ZSqLKLDwAc!q|>N;nuqvX<JUK}4q_JhL~WS<NU5(a+(2A4
z1!-GpS826}<9)rP0hlGlF!%11Y`%8v66SY9WmlaCQsE+``&y!!y5e~AopPi0ywV%d
zXTZUHJFSA96oS9?{PDmv5sO)hj@C@pJGQjpaWD(8@tZ>~U2%<%sewe({Ev$2pD7pi
zC({@r4u^k4vusa-X~G(=tCas(>i;|7*{-9eNcMXLDNxErKcsNElN<p#jDX<~ZOMyE
z{9F%a(wN?N8lIZ&!$OQ!o(nf+%-nVNMtvRYb}YL^!M_W9A=%AK`_}w7BhcyCjMD8R
zg$J-(#_{W|Y%G7%<ifl-+9kQ}uq!Q<gfjV?Ziqek`G8p>>q8^8ZOdoB^Oal=R=d}I
z(*xzVoV^rcUqP_rpP@zfz?;<2LIBlkZZlvwQW{5YGEW>J3G9BNRe5gF=kQZdQ{n>|
z-o0f<y*c`9o;%Z|$cN1MY2#Gy%#4fIgdH_?qV(X-BJh+-ynEBw-it3}TJQhdl2msj
zRG=+QPnv(eo9<n>N8lpZ@isZ|y8u)!(@3ah@qQO0pxcV8QKO!OiGaZHy$8W`Rc2H+
zW2VJZ=iWBWfUv-LCoBJOGH<nEKWkRwnYqSn{17}*QulGsb$G?9EEL{Tq!;{p?+h6F
z$V+x^G%0c=bF`c0I4gZI(@j+{Q(iREAI}blX+54@Ca&5-{dzQEXg-AJo-ELXL~h2v
z`mQy4_;FP&n50pUcD?IIbfuhWGxYbnX>WK>=-%oeC5JSTY0A3xj_<@)diMJ*=84qC
zUOMZ}9|u79fv6IoP6bEc_4;PChTCm>dD{djOnq+OK~&x?Nb59MTKr?GB*2M{@OAiD
z>4m%KQ-@Zv#7k14OnWXppXmmSww{=e4`{*BWGjdFxD+>J)<@Sm+m*TaDu6;YsV#DA
zz1q+WV6!i7JGZ;>S6yEpodB`t8elLfU6TpQ6OaGKOq`X`)!Me_8NpV^&gM2=L3B5|
z>|)0J$qb#UZ*48ySD0vY)>$CHikV)k#||6c0_O)!vgX=%5K*fMQi1r%TwGb4YM63|
ze=qXvB$vppnBSi{k7ZTnYJ;XWB>JE=CSlf#ke^!t^rn%)Oidb@hvAhk9{g65Z8R_4
z!bbObD6!r|aiUZpHknb5HWd7d<c%aGgZS3xr6=(K4Lq!`c2rM>ur;~Yy)~wi<63e3
zcM!w<dg`z*V2_o;wWObqR{&Y!;h)`$m6SqM<!Ewbp>>=$(W0e2_cp5ATI+Gm&R4Kq
zUVhcWHZU3ou%7ILTs7I;<>5QGbL*=e88>&iB!wxJi0oNAzkX^oyNcT5Op@eqQmLvp
zUlD1%aDYLSgkxJ=7<w1udiGJJHuGnFVTi3%vryN87OmXkf2m1dfjkPYK5OC>w#Bw0
z34+r4jsLVF|F+`)DVwNVKR%Fw%iJyeOJz8bBSGyUq-~M{S!s=tq|G5ncM8l>Xxbz5
zrk^<E{mjZ1km5&wPk4n*#h|AM4_?7X@jow{_u_C=xlGALI#}TDaewg$u;b!QVK`p8
zqK4JQ4|$XyY8RN9PI4%42}0^kW7F$1p%u(Lye$BlM~*m+D`>v}pSC~Njh_M{!ah(7
zRp6`f7nJbD5|8pU*fW%1>|31#kd`Ch@vYLxEeVTKC-Fw?8-}7ZrLiMNn+e2AQXKF3
z_dcApfk%T`6YPhzPJ<ns@Os!L#@rfFpVeL@R6-i-vj|GedGvB?deuTomys#?4U3;I
zR74TIC}_nk^+(!dV^mB%l#zzQp;QJ}k*}^{!Q2bW`97Lt=%~f3x|77~MK7SkBSb)K
zp&aRkswoCXa5VsbFcI3TOL)dlrnDTm;ruG00xyh`<KU;g(@KzMm6nBaciBu_9fvQz
zR0sR{&z;Rqd}+ORT@q}nBXG%`@}aem({vCaGX|*S2(Q6A=MAc0DE;Vm>1E}cyt<tD
zhCnb%nHZq097PA&LkWHGHOa6!MDyO?x7Vw8{eLtVoO+_42NcTPd(3}#?&FOpWUIoA
z%hMO@Cr))1x4V#9A;hm?DWhkl5#pTmRbcpo5noWHvkb3c_ubPR;QqZ%EgVE*m}K_&
z`3t-zmeip8Ry3XE3K!0ty-EvN53O9b5pULdp(2?68tNahoNUQ!Y!q_)cRAaOClG34
zOdDvXkOcfSLnr~350KYjse@HW7QV{>+@C9g>1nUe4_l~6n#JjjQ@eZtmtt@?gYEQ3
znso?K`=(nSNEAi~Qhz>6RwS;#h~!#A&cc!_#`kauZUeN}2RaXf<*(u*R@qOK3^Phy
zr<@SJ(N?(DSx*O|nZ;MQ0%U(v1l_dL78g~bEz9Ff6Xl_4&z6rQCvzR+BSQ<1)!q?m
z`h*FSk7^BXOrf?*93E+3!Nq(APQ8bJoO=9PgE9Zx#%Sk=!;vvEZPCXqWiYaW5)@8f
z*HB!K9${HF7qQZUn&QJiQR?GKFi2~Hqr`Y66PEh)-)w_2GF%eJCr;GVE{upY6WNHR
zWVM7Mx)j+&#FM4qzFFIy1h2%Xpjlh?geZ?o5?Hh~H7_L<k!hb!WX`$8j)ZEiXc4V*
z_wKx~^~rLUgpKW`YuYV@d*zp9@C5y2BnL8e0<mj-fkU$1<{BohPfrH_l=2z>BOW}$
zD$jCo&Sf{(pPiSS)C4B*@NWMc=LO(2RG2EgOhB-wd`B$5L9?|aFBe;r<OM3cBgzSB
z-kS59($FS$3+;OH0~fMmLXx>5#rYv41v>g{)bt%#r{iv`0cYi{_qN%^Z9WhSm;gkR
z|Dv?f8*g~Q$<f}NU#)9}g#mrdBkh8Sb3G}J-rf(j2e#99E==1VQ!uWB*m?aWb>M57
z{dy`QD2~w=I543makg!s3}Ds<?g3Wr*=bB`l&`n>(+@7zEtoz7|9ZoZ`%M1h>)Wrg
z5~A$)j%0Nc77=qqiI#p-Qgh+Hl;5B0k|mr36k4Pj#mYtC`Q;s6z4BaKn@J=3WxtAe
z>c%dGY3$?NUhVxE6dB0_v}A&F2^sRHoBiAR@gI-G?mNG}_inEddZhG8qm^^+b(_sS
z+yH1uU~y$)!xzW0{js;kx^|zu4vekDzwDao;iz`&g`ueU27|Gub{hz7jSLKccARtj
z;8|wNy_l4@&^|*RbTmqCsn6O@wcEz-{v<5(cWNa@Qb+nt_d>c*_p{mZe8ah+lo70O
zAbig+i$4UUa4=Y!4C9UeiE90O{rt~+*nc)3VeHiHH#m(xnpoMf&-`}}+`i8tOt))F
zhwJ&&XZCVpP<F4c220!Iu|s!YTN$fe>{?&Sh$~0`qbmQu|IZw7@JvQK79dp`UIMWj
zGZB>X{zN6yY5z(3Xx~(|y0KjM4M4b513(rp65DFY?hmPaZw6dm#f@imbw&YI%wau$
zSWTAd`O5!ibPKlf%f=zVw#Pn#*{QprV;G*^wE(@4H82H|18;G7(-qyLi~sy{fX5OG
zf-)6(V0fSEkZ9c@_Yr$`x?VoTgIF?p3<`m(fWg7!QTCbY47^A%W254FeT~CG00SUc
z(D+>r#BDpk<M(*ewgUD;A+aCIE+eMA`~TaA;-SSQ38n`F%z6F5U>6hr0iS#t&!|Ie
zn7tB-xsmtmvh%-wBdc=~otVJGgX$1DX0K1w*b1xus!;3k0&WwcjWTc^-x_d6838Zu
z5>FKaijPR_WMP|VoMV8RmQr+`21fBtJ((?9bUC)e%+|qj33HYgUjIM1g#Sh||JM?_
zqB0LAL0>Tm)cUTa|J2a`Bbfyi$c-%50B%iL??2J7zt=`K2ll&mlI~+;jF*6Ha=7FT
z9Fez1%mx#$v|*V>5A(0#B9?^e`b4d}=!~!S-fTOMET2uoV)ylM_ND*3(D?uHt3E~S
z&xcRepY|sxfNeo&sgFSlnkg7y9b&Lxk?)Ly{^~moV5}7U)*lkgc{5X)+WYJNE1x4n
zlND_7#*6(3E~7(}tu5&O_kHzWUy6u*o}gWc<Rs{y#cD+a{k42EM9R$bBcU}|V`NzF
zv7=(VTwGH77<&ycu@dTx7XXX)y(5F)BOu8D?)cm1mq%^eC3~PFw=e-+h|Fr;!DoC*
zS*q5PevfDdTVn{LLO!T2ea)Fx?4&;zcm!=c{0=z4YQ~})E5!n-xV(U-Y9D&6>HpKG
zfJywTRRpd_U**((78YsIi6vdKK3!w4pl?Jz1Pm*;w2#wGKA%%x0_F$LU4^}xA8#t6
z1NK&yvu=dEGk+(baH~rq@M^A2`-RW{-pl<*XZOE;a1#(ulRTb}=)`hjJV6AU)&mrl
zHJY3_Ud!NXxcG0@eug+R%kIsPwEXm|CZm);E~IX3#DBg1iAH}j*a?z910JBP!9z18
zmR$pTf`HaSp8+UCbZO>;M@!85tH*mO$$?xTr%1=Sx(I4`G5|)H(+n{8>Ax-Gw~Wtr
ztk}4m`4E<bCDUM?!L0mMTEPsvbJ!NE6$lYTKS))F%jLG5Gwn!<Os>m=<YmIuuaD&^
zWHxI&!O7uQ^a1dkIspWol|uGc`#^oL!erIdMeFU1Bk*OkQ5O&Y&tK-xRuk*(zqj>I
zlGT^}S8Z}oA4w?IJ0b}VlJ`GP;y7zwan=BpF&g`K_5DfP*-Q~d)T)At6)3w%{@}a-
zOU9x36!3g_T|s;QjizpI6<|Y{IQ_Zt=0RYCWdJn!*v%DF4p=yxj!&gOv!QLUF%QB1
z;wAhdS)=p(lRn4C%k&k&cVM-u^c08~b}g(b!g4311s)~4dwIU9Ft_j{-ifGS+ILjf
z`rXW9vR+u7PUt=$qzGKc{Y4pa9APRK2<R;xDH^Pd`2g$nF8>Y;=Mv<1*FH7R1)mXC
z&ar^xd7jHfFv|ShS&++(?Gpy^45zbq|HnO)or8VBbjZvtpZw?KTT`9~<BmwO%)GU}
zf<I6v_{CShL4D>hH5S}h87dq~;9XT?Asq~f#vxE73Bf{!ANviToy6$}a<l;A(_3Af
z7wJ2W>4BT5Q9ay_dC(!x+@Ig4tEg^$`W2jrG7}*nI21*rC&_LGR3O}=j)1Q=aKR1$
zaFVVd2?g?-{&An9ysZ(rS6t%MoUCGO<>C$LWZSQEjmE)eQ*-jSaBCR0`TUBGe{$uW
zF^FrGes}izzk^=?>zKe^6{6V2$}SULj``>3{P&&O9ShhQzv-vHDW_iNW~0IhIjk#)
zV)AIIt%%O}636!Y__H+_cV%uz6580^EzLE4TqXm)+50=~ZQK7WuK!lH|MOSE_po=A
z4nMzs`9Fp0UB!9qu4!!U?bH@{OMARQe|h{_fynHf+Vgu1T*2(8QgkE08*iP~<sVf3
z0zfO-%9dTsWBGxQMkBLEPwQiF!ZBhwYjw)?;3jzP(hbg!DEbd;YPKh`455U$M^C9_
zv2-p53LqXPTa(FCQLPN*xwx~QaW8>5uTP;)@a0}XZtf@Sbv4a1kR$e4Sm=QPW-;6q
zmzuCwJF*M%5LO8yyw<=PB*8(n6HYxS2($_7JiqZPcLeMIVpUr`!k&=sttSqhyDwu{
zNwL5t^4_cuoA!rHcl?)bQ-Z$fNcyK&!A^o>B+x85+gQNAQDAaX0v{urR?Z~c_57Qy
zHTr{Q`RuE0G9v_BNP#AR#MyBV!~>DW%~fA|G~z=&W<hhjK9*#4g5_EXZ@o{+o1aYe
zjumk*Rtcp=VQy6TRxSJSn(O%S<&|?DoC3j&+=RMVmQP}G0*ECJKOOlu8ak7M_lxx&
z^0>EZum};2sF-WvCRm(cj2|~CEL+ToYv+9cZK2_Q=97iNw=v(n=>iS?7ibEc{37M8
zre7v(-aqVbIO~)4E7}t6-;*Ns68az!w>Zdm0*tg5LBprO!Pofv@(e%oGB)J${NxIr
z%nAg@0~{9vlG+cSg4N`<6(?eK*GV+^YW&&nx+ko^8tA@UuqA49+0>{KF0#R(q_MAP
zGv0)gQ{fU)mvTZm3`_SEoec@q;$wZ&t4K5hpcs{On~&=I)FL8$?AM1FlvaPARoUqs
zetWX*GcfyXpm}wsd3@Dp`<vfkpU3I*xysv^3mdl3Ui0bSMolxNGadI2q1G@j!0e6E
zU=8?G$vIN~hj_NXYM<2`(eZuO_Zm)a8K)zQ+f+;#<DwJ>rfjMA?2gRfagxX)mZGMZ
zV`iUm-&Mzoa{SzuS*3|bXBG>^yPgS7i47WTQ@^qC@ybgNfp^=tKc3qE`#x4@Zx4+2
zqVy5!mJ|B>)J{umFF{SM^&yH{y^twP`kOr53f*b%zcIea+WMBWo&N}1DlubA38{eZ
zBLCjm*V#j5Mpc&<yA*(5>HEcYNC<f1XCaEl7a^k$Y5RS#-6`4&qnSOZdJ;t@)Ck4a
zwzz;c*NXULarAy7%B|sBH`)Oof75dhcm~OU@^5wwAVf}N<5bmPC~hM=YsK4!=cUfz
z{@MDK@w)n&IKw$OncsPQm(KCP+H!DR^h*jFmGUFdAC*tkncQeSV@yD5srI6S?B*>A
z1mABh;*gf+wIcZQhbu=dt8(ALE<G7TuaSP;e!o!_t-z%q+z6m!08WJ^btPTNdIoFv
z07N)z@JG37pGpfqr#JUaX=;e<J4BVs2`at{PbbDrXjI&kJY6e3Tkx6>BPJq&mN0d+
z1LgSZli%1Jz0c}Ez015Z;}lrvr}<jEmcLIY`VAqHhWYJbja1@xN!USvkRv5Hr`u@C
z!mnbs3qEErHZPiUqMiIIpk`F-euRDKy!i=c7H%lv&$~S5kn85)7v{((1Fn-8>L5d@
zs3DS_?*OYapOQZfgu0dIxlgs5*eTAh5uNOSaOiDUz>`wZV)%fgAuD+>cCQl~_*-I%
zEk`(cSwG;Kr5@7<jGbRQ$syB|6eD-OVTneGuhQhdlqT^=6h-#GW3LXzGI+XbY2{V>
zRVd}g78IMx&F)@8BJoi1n;8rD&-=eQGzij2?V+;8lX9t6N1D^V{=TI}9H4y>1*T=a
z5mgHM03zU>U?PT?y#YNQbop#Pf|8=(_xURRlGminB*mmGNd#bUnVSSGXTS^AWqwYE
z>=Tafavk8dT)y=P>YaR3573L+PHd+`F_rz#y;bO)=BmJfi!DVCG(XHEPG4dlT}}-&
z=Zi7ER5(-r>=S`4-V5Pnk3qSktC4;{^kG^`pjnSV(@dLU_kMk3pl9R$$Q?m_c8RhU
zJ6nP!qwW~01BIieK~n~Vj6N{abaM?F^oA?^nR^AOixgo1&fOzkz9W9|=>G3nA-fO(
z_V$(wFgx;HYev&gK4rG2;OEf&=CA(SEWhqcR7|}?2R|v(#aCYu_Z*b>?L3nNCKt*!
zM!YU$`|-Z6$J2XtIOpr8VT!qrpE0-BoKOUH0+^G0(mefhVP0+p$n|zec<j?}79^rQ
z#n>KA>^e4ry|#Jq>2EAnpB(}8h0?qhUdm6P+j^FM$=|=)hUg!VKikBSA}*pa`erg@
z4o2~Ij376ocX9E?u~l&$9*GoUW^<A1@ktQyVxYEF3>VH22&jSLfGu@y)CvIJ^g&?W
zU17JB$J-5OavbxWAvm><2yk6NVn(8f)yTWXzjNUPbTJ;1ETB$O*M@aR+^YFAa63Cn
ziiFJyV<TmPWuK~}Rd)PAfg6dvcDg3E7<yx1MZht-=nzJ+40hGDwVgZ(T}9RVH&OjI
zIN4TBq{^%(nFPoga5c4hf{J!>0pH(-{Mt1fqbuwWu;nWRVokpcjHO}%AxpQn3x{rE
zK+WoE6xw)*EpC{DkH6z~Xl8`+mZ<HfqU3*{3`%gYo?OZNRDdMqFVlWs!WpIUREx%+
z!{h+}9-0yX-G~e<G_%_I1?**_3#(IeAOz73tQ}A_&t#<cq;vVVBJy~07>@FSj<P(K
z6a^uURWY$DVYU7i_sc-XHpLmNBQY4?&8En}=vbvDZi*$`Pu;sV*Tzpk*=g46ff)tG
z6-&d+<A-&6*$X43N!NJ%4XbJ%oPdtDbMlqVv+x7HpQHIH9^m(W_7!}xx92622^;U?
zH}4<3HFI;w5={})?{#)nesO18&*t|i4-3EC`jauyl>8VR&8cX$+vo-p_tLG1W1VId
z85B6ZVgPgN^Uw055s#!x&ZsZT5(l`EACGZ%&j4Xmid$0?pK7I7`6;{BNd9n)Sbpdy
zG9H+Ytw?Hc6ruAeC5$qegX}rZS8f)m4W}vjt2*o<qlcpkDU^|F`65YHX87OJTf*0R
zlxdhqF?bY_by$Gj@09Ks1TkZXB46Y&rs&2X3&7i{-@p<>Uf`GzjSl{KenLm$?9r~U
z+27gf3@;aSWJUW!$bKhWl;4{yvz1<y-JAjlxO6vpTQP>91-|eo^A(HzauOd^@pEg&
z4+A5hTNF3I_B<bYT|KRn`u$y}MV*%)N!Kz+jI#!d5g3|UQFH_@fv|%UBve68fXZpq
z9bf0Ce1h)G8pF4%GP$L0!F>*W7-x;&pNNe2Mbe-a4ZQ61vJ~LH6Rw~}cu#cwPss_S
z8V)dQ)kZ@v4mR?A^%e%t^!)H$+ssiuj6u!xF8sk}B8Th2Om7fJua7}w8~;abMpm`>
zI!R2?Q2JSu5ZXI8NM-z_NK?>&tG2aJEj59oUM=PHsz3_f*-yCdH8sOnB#OqKyL_9K
zmxEEAU#v(|-vGXt;!A$5gVPdqrVyV%e>4~Jn=kdRZ8||x)qLvO>L-c#{O84*b_SWG
z#nb(VGmDW&s>*N<Je{U2AvA|Mt$|3M(Ip1Yn)XV~E-QVhQzFMRRm@#eEJc}bL*Kxz
zX7}xEK$7o1-q{?!se`CaZ@|M~p*$BaK}?qyZ4+-iB3tO_BK|LGwNp4p&7t{f<TH*7
zRY9bhX+$jY*7#9EY@C5z_HpLB7VdaS+R|6+%>|2*+lC=(?>7m{qqr*uIZ6+!yr2u7
zv?;dQ-}EW{UKb+?oSkV?ytQ+5zu$?wUZ2g#AL8d<?&xzF{(&ThllWU{f1d%5_lmQ~
zu<nM&&Wx|@l?(kO-&=w->86u20D3w$`8ATb4JE-bt4{4<+cZ$R>V3q<KP}XT8r%kq
zU?xH@KtuwcwQ`0AS9YYEE=pQxqhQ=2{uS%Gu|`VS@*8G{QPJVsENO-s%i-TnWt%LL
zuw4-CFpBay*l9BHg=}gjTg~};PVb!4?CV8l{TCAr;D+m&7vOpOraK^n7#%WJr*9w0
zZSwwI6q!;TchI*|)oYb`nV=Rq0zWSUWapY}wb#6$g;Q=4q~L*7^$S3VG=O|B7AH$O
z*G;+LFhNM1i^v!0v-q`QVo*)5j^wSK7S%X|hR`W*{XIu8Uwmy0aFZ>ctAi!1xXo3~
zGs-~od!LJQsBDHuuIWtJZ9H)j1mo`tA{y)alL@aiMa^@r8QX>V!`?49^c*2)AW|WN
z*rgjLOJbGR{@<%ummb4jr)Xx&-l%Q2+y!}JKU4MN(4<P~QG`{@!b79e1f6J(a{w#Y
zG-=f+q{3nMM2A~$NyNe*Ut>y9$@F+SsAQB{D#z*F^G0v`=?M;Aw%JyW4Kc?oR0}kf
zaa26&<^-W;F$*Mkg=RO5uaVn-<Hy{%yS3^=g#!;$AyuTt-GNcFvF9pjze;0X=Zl$1
zC+IlyPGqtt6j1hkbq(Tgrz?z!A4#7=KBZr_5y6!1r$6f=Yq|3wL!fJ68TOfqpWWqJ
z78fGGo)Cl4Oeu`BVeVMq8!1b|kgARS9DE=uvTT?sRAhE5`C63g;4SWFe$HY(>2Y4m
z&bg-s1Cd&5A>Jj$&N+qsk+7@Z-)Lido3A!PWQ6vN8ov!L)pPg62^3?K`267)wkK2j
z#K?GCx5ZaXWY04zb|rE%<EqxP)1{vSi-X?O#<Sk9*iPBND6%(e>qRZGYP3@B&99Qm
zmmocVO}XuldaC%CK>k-4rK;(O8_Q3=y_WGX3IAFb>Fa@y!sPxh#@;%v>TYZM1tb=r
zbcZa`MVEAUqo8z`NK1p#-7P6dBZ>%%?nY86X#r_aQaaCE-uJ!te$IK$`|iK+gG<20
zZ_YX97}xc^l;0vKsh<rT8;Xtvvf}jKi%+=EKP6tTuuIk~A2@h}!#3c&|9eX=prMHn
zsYC#m>HMd|+|Io3G-6)dxUSS|E7kqc!H+GXp8f7(f(}+z+Tg3PZrQ29lWxmUg>LWw
zJYz=LMbV|OBmGxd&jcNr&{_aA5o1>)ShT?F{;BCc!$xWl#Z$9u%xYrqg@5W+&)8Xn
zzl$g5#aky(U)Q6xiQpFH_H6Eh?opvxUx8S8c86G6SMdDKQaSNdT>Y_9x$ldPs{x(X
zF~c2uct6t}2}KGkE$8R@vK|iwO6%3XgXTd=y~D>#$gZPvkL#!$DcbzK<Z^IC&l}z-
zAv-!=86PI4p@O=8p;GyUwUVF}X?u^goIN(3%w#HNBK5F)Vg%>9bcX0!79C+arS-hf
z%6mm^tmuxDIM>1g(&^j#5l_iqs431+tfJJP-9Jb>F32@XS#M2#$k4&}@Dy>?ab>?s
zDG_D-4mTNKb4+lqVka<OY&pa%_m)elJ~Z(dHSnwL7ji65(|eXt0T}I>gXRn0!7XKo
zk#iBha@3G&#Sts*+fwjcEX$U@g0r`(=nFmh8`43VHJMbrVTV-&Whx*m0*z7adyB`P
z2kwoxes{gvHLUgP0mG4m(Aa3T>d3R@#(Qb+w4`4t&gpy$k1e=;7bP=pS6^vj9?2L~
zHWPcqvX;~s+mr$BAA^$%Q!Npf9ad>Z{63p9G+2l+-*=0apK60zdp);e&o!A`3N19Z
z-?6o!>%HXBX@egchV^{0>K7OSO@WK6!#nWj<Opw0>*e>oFsa=ws~NrshUbnS(8Y9c
z7XLbZU!Z2vL&u~Nv>dan`TRvfG}gTxIvp<5r`#T1FLWOIBy+16Wj4lTZFf?;`DzYm
z3B`>E$_e54iMK$wJgr5e4OB@Lprj}Z;cL=kUHy>(kN_!?8?{AYdk4A=PcgQ?p1Ht&
zIYfv=mcV`9e;&eId=$0UZwg@DVTmmiauv&FdXl_JL?K*?PhAWwEHU-4;Bq#tpr1RX
zlk|Wk$j%)zpK6tZq%j;*wsDw$j;)WcQVdHfDAc9lC123@+*76GvRm?oJ&?Cn>iZBM
ziRRTW?ymfkjws3mBCCF`J`~dgxy}W`r87h;u=&S5CkX5g)JLlyhIJ)ng+iCbW~7%R
zMG+jJn3Gfo_pq{*kHCb}>8Kzv7#odis2j|o5jULPn=*8b#V4^!tRyXCyA1tUJImq&
zqL#2+Ue7e;;6S@1UmAe+$5_#<2@zkyzUBEy#HSkE+Ir-zm{92kOa7$1vM{nrg<a}k
zHqEe=IFT<A0V&a}pPG3PsQ%2u*M{y-Nr;0DH7IgiWWHa1K_fLZ?Us#@abccmYK7ut
zCbcwHXEY4R#)-09%+>l4b!CLyJWvqMOpP>iC}wleZz-okiYvH<tt-u~6r)NuR8)DO
zNs}`inZbcrX6`rqOz78WZ}*r>fZBj;xtppqwHm6V;<tq9532wN^pHoQ@Dg4I<oxmz
zRPrdV<rdr1Ll0sJkp^KGf91)1$2Tpw`&%KzFV*{w>qi!|Cls#Vgqh>?5MZKkldEM?
z{*C-=5dQe#Oe!_ur*YlS>Qk^9J_WHY%r*^~<#;V)&b{cpO7iBTq!~!!rCzI05W5AG
z{AbSUU#mtFfXW>3j-S_W<h**B)`4L^kkyxNQpsEfL$s@wKE*yHubK#5bQ3iuRZspY
zG(fy)!3(p>Rts81rsVCXA8?=To@OvEFg*Y@I{#zRq_Wys+hS64s$&Ey?518p*ZxXo
zBiXgf_=4lN`w<LsLAV|(C-n(QTq$+3_^BTBQOOg((&Us!eNGc%UuFzh?Y?U(STWCz
zY0jqzFeZeNPK#z7J?ehDH{P3TZSjQeH@2JP=Jg0+ApLQCqt^WM4Dq?zk4k&L4fs66
z?<$HW<<8}Qe?M0pi4$FoXI)`g72oWrYtS&J%$iZo8p>C0GOKr(#PcVnqF9?>P5h_c
zSFBCX)kI6SxAzsUO<muJj_B&Y+t|chqh{d3u>P&7r3Ygwx0`XCbKLmV)%S=$oHW!A
zJ$pX){AmO3itl!!sL#=Wlh9YM;&R*-<6Vz5t}Ml@tK7Sc-^JWPm}9o*g3ZoXY=CW-
z6j3+8Fb;Y2wo|B4FGpcO(B0wE^_#}*<5`Ns(3Qf^)`n8o(CmrxWx;5_<U?j$bu%z(
zysJRvEL(K{CFpfjdHCy6+a>w0S<j_9bTY5AUK8)IUMo|4IJ|3ta$$DSjz;T~bwePA
zFHCr0J<ECseO=@t>T<l{lXc*cw0xFyfVr4w&&$s!o*-fLr?m~E2Lm7;x_^vVOAxj~
zNKF1J8`DQ+TJ7eq2T-`q`AqSKsy!`jWdd@f?Q1P<?j9B_J|Yp-`+A1cV3gPkM#kRp
ze8s{&G&VBKgZU8IV$j(fNn#ZUlX7(G1Xo4zC&qC>f%6F=Ac(ExWRwF{5~at59!r`j
zS`|`RAHsvw(~dy8SAZ3X-T>QvLQ&X9M0~cTfA0A>1G<|4a6-KTnPbRB9LbgoQlnA%
zYXAm>>#}F+%o=Vax|bAZE=W~p^jJ!&1xODwj+GiXD@-*|f(ez8`ke8o27?ac4y!?P
zG0o{)&R*-~ZrGEZJM4~qub)nXNP>^|&9;71?|CVlEt9rA*KT|qk>?Z2#X{Ortjo-K
zu~t^lj|IibQsHcwV7u1%Tq?mPZU`KDzNyFt40Q*rYW>DWvqOef`F1(lffI-wA@kom
z!a5X~EH2{BbE?H0m5~aKtTm33-(Fcx1HEUsUuvj$PWT&sRZz~jc>S{->?k7{f;;rR
z0T2@O@vhC{Inew$%6dIOI_X&ctIV~OS&9?ct+;vZjUR$c;{9_yQFrQN)rE2hS(Rx`
zx;Hd+T~E^T4T?~JH^sHYZ^wuW_s(gA$sA(OWUemT3<*Bns-n&2owB^qNqMsc3!7U_
zzd1)(Cglv*of4j@3;qaU4&hjTF?3o<g=@@~DU)d(SOW|1SCOWpeUZH}gBs9t%s*y1
zq)F;G<^05`KDc}XrXw)_A~+VigU|(K1<Q}tAn%cBtKy8Kv@RtY_WI$_f!=fp)f4%V
zG(4I<r7MlzQLBpSHUK)C3FGTy5JLM-5Pd#D-{<-u$IIMgAGFhb(nUPWjVzL3Rx!g7
zS}2jrpgVZizS^u6w*w1KwiN)Ayp-}S{3TO9#AttHvBF_vT15kh`DnF6pGKk|*8x4X
z=m%C)>^!uu1Vi@*#%ez@hds$@DzNDOiCjp4n#6pZRnIHHjEzcF(Jg}BNW--n&riq(
z#+o}&fv&uIjcO|SDc8d+wJe)p&cz4RtGvf|mI-!n7D!xFb}x|gLSw6SzdX3bhuCkL
z(mLi57MG`8#;251Sv5H{Wmn-^<b+w19pdM`%^NQGf^!%qbs4heenBySd`-57cWp~Y
zKkA3*$dKIk3S|FU5tus<ijVJV-hA6mSy!g9OBOVj>028%8#6wV)M~#q`1ve~?i-E-
zmQ-SaEV@@aRy*Eb;MpJIq~)!HF<w471i5Dz$8SzBIVED+Mj9mb)#0ExIa&ZIc?N86
zKR}j`3=n9y*WpKwyNk6}SzbiHiN^%b`<^*Z@&fU3uD_X>6_~zbBF0L)o2Q)xY?V?u
z7xLuN&riv*M@p9OMI1<S(~J!^T>M6e)%d>>a#$g|Y=FrN<;Yh<-{WumH-*%>IRzG<
ztMCey{wi&wz%Z;LB8=u=5Or*BlmeJgbU!O=chu<^Vmo<Ryu;aD0F;j3h#c97o-&K1
zFXdIdgXc_~G_9ScS96zkS+QdM5;n_2wPIB!hJXyi2fxG-rg^*XC6l2#o-ORictzaG
z2P|j|gfc`}Fx~X-Lk!wdwyr)l)y8xA*Vv{^YoEM&SXG4r;b@_buO<0K0D9UMGeBre
z<Hiw`5h*TgmY@*08?hXUDZpW;(On=}PSEEAi&j#}uIYJy>SAJnk7mxkDxC5$#o41H
zKS2RG?2y<lRrE6ovT|(xYIy-$F(PZAxKA=L^qWK(+j8y8l~@4|zn4qgYeG97g}%@K
z$If8(?f<+pn6&e#(47KEG-2Q7(LLBAE%a_O;)-mX%#+AUo%~p#i2#Y#W$e+oLc%Al
z&Sg|zTG@l@CC%eECa!-C?ViUYxjrTF*(x}*`qh5bpj_ejq#ADI7sicExi`Ik$2s7f
z?YD04H{G6~^gDTLZ(P5W&M(P&S?72c4?pn2wWDW)@tV*XUffp{Nzj54kT6;cNB!Uz
z>U}Q~y_k@)xe+0PUCdTNx(2KVcR_Vfj*pejTbU6hH^rkD)o|ano1u_$upUtsH%@HU
z^`{qcES0YmH@?6y+S4z0Y&q5*#Emfj8Sy4e9GO+Zp2(sMu+V+F*aY%rat&W^d~kOL
z`vge8I||410K>AZatzJ8(VG+TK&jjrRHBN9J0RG7r%yz!FT&kfb*l7LAu;@pwjZ>L
zD72nYB^1va0RmbGn^<<G;@z|MyB8O;Rr(<zc$zBeGjhb8!tjm0CPS2l)J8O$z;mIq
z%5yww{O%$|2Jwk6f5wRRXsVR7B`@ZVvfgYhU(bsHSI-FH1paqZ2{=j;7hkeHW=qqi
zwtV|0@wi9Ba|Q0G90{jLK$vxxgY1LVgYQ+Wz*}WS`yR2GOX)L7(6R^7>~J`6kc@CN
z{o<iFD>>UZ8`hK9-{-deLiO`=`(0}LZHaW10LI_4M;{ZK-t#}(fX#u6<QreIt)egE
z>$FF0h4+%aLcf#vlC8y^An`B?0^@(5g1EU=P{s3tk;+8R&0DCJ^i@F)@nwZ`O;dMX
zWOYC8_995-b;t?X3}#la?=t}bz=*V*=Z%hiW~L!uk}vHgpT*%E;8MAy?ohcCwHzQa
zciM{`_;Zs<mSZ_0hp+J-&&bB3JwAM0%Z|B1Q7)rd{IMwZLuT8!bzU-?KCY_S>V<<r
z?YSSf33sdHE0rdU`_->vf-Aoag?WrA`1kA5)^RZvzqK5aG8jDFY9kY<G9&U8_I-F9
z)mpFGIQZ?YjHK%1J8~1AlUy2Uf84EjuhG2<+VEO4;HD@wdfOM97>?-QW8<UPP8j}s
z+zx~&&@>5+-BWq>NRJQlJ2@0*%>x-BZ~tKd5OHcK_Xd8gZef;_n^@(w)ifXvNeR^$
zh<<0L1>8P|8C0Z2S5;1mO5}38(fZw+*r{hq8RSK>v9o5BMKQbIrx+gErrA#5)V#y=
z-h25C`KvM}zS0yl@o;gO9&Ii6bMgFm2&b*I#M;Y)AR;!BS`QKu&eOOewg~vehzRF>
zJR0V#{FI#6R?^8o-eq5Ca9~7zI;K`pvhlLXU+`@4N}{Lk*%wTu>mFk-(m$UZaaC(B
zfUs+$%eh3%dXvf#M7az;zAT5NYNWCiK%AVX1(+XcdQ66WTSNC|Dl!EvL*+n%BBbi2
zLc>TGBfKKhSWO0l%=fLyO>xRX_%)kkRvomO#?`Jow)mzn`JVbnBB~lL@}h)DcrSj@
zUDgFhH&w&1ZyDwR|I{ZW<H>PN6Y)g7|GEU<*x}Q6aIK{s>Q5oA!07~mB&3C0do{!b
zt?CAOJtgmYqGxqBRch#k2WG>LYie*w93QglgzD^6*ktr2#wookaM7|@Q`aULOgQ>t
zCDwjXyAawL2-(vJIepg?s;Fx<O6Okqso4Iv!r~?(FAUMlHAbyYpsrb(zMzlMOaJTu
z)W#Z?VXZIHKSkUnl?#ieBRl%en#1KcwQRk1Wwe?P(^Rme;#O=dws<_<wcB6QGOQz)
zv?(@VsLGJZyhW11870VfC~74YulEuYM&%YOL;Effw<(1+5iMV_L?ik<v3WG#W6@JC
zAyWT&{Ok);S4DiQi(On{o1jpVO9xSzbJC7lmtS3VRyOQW33zh7)JI#Br(wga5U+H<
zxO1^%L)r8`!vfflifkwR;4MV5$3QQxR3`>|l$+^0nR9}>W>XwafT#kQ2v7JJQ~rR{
z#&f$US5NP&2)YNopkXjMKq#EUBi<8{n@kw_(6+#7LcV{QU4$f$2Vt4RZy{+U(R0o^
zR4vD*!;r(jDMtM)AcxLvKgt)Q(m08Gs9&hprA(Pb?<=R#hk<WTf8OH*=U`uf&8f{s
z5^c=;>W-hYxC+i9&-Mx<I~^s`;i-9BzS-GLiRWt`VUS2K$=5*;NnctJ9L3Vv%5~yL
z*6V!U04j!K+0~m?GI#bl<_Zq?r<7R^hGE9qtF^G7>CC#9MSgfU=#=)RHj)oa7>B;v
z_{*V}H;S=OWu`jNDTzo5Z*po%B+<o-N5Y32@2u21M<}Li6pB-3J-_2?Nu^1Y$nM}E
zsjX5FQKDhlv%It59Css%CT}9!rtHuBGnL<1`KNMBmhML0J@-Fj>D(A3k?a3l^t3Wa
zj4d9S$JcpmQW{3B#yJ%)zVO29t)d}RX{1iH9nRD}D1^2tp$&3PMfRN>*u1{UhOI?L
z(;BJ!s^mkfj~wr^i8riBdf_2yph_djHn9D4=g$byMa_`nkXi<XuB~&TK^U`!q<7<l
zri$Foi`tE7RH^&uPCH!H{mmYvpILhNlIv2OX1`kz8bh9K>0huX6GkV}QqMAyX_!Ah
zH91I4q|oc0+68OgUqk!X#cl~9i61nESV>D2uJMw9Q;)Q|Uh6d`Q)@G}l=}MRD=Vx}
zG%Q1=P1w$vcr<fi>3c(WH3gO?OkTg5EI7fF2#T%{>j-dD*&(ZWh(kXcMDDF7T;n=_
z*j8rB6#sbW2`&^XPNlj_d;l$^$fQZ`)mte&<PQ2o%2xtd7%T4z@J?ZOA|Zv}>q9wl
zBxEBoy%RkGd+s%aZN9Gg;dCm&%`(HPkd;xT>efF_b!?$|`Ia9`{-V{~Kq@;yZg!#O
zP}-73imiS>hiympFnGnqWUvYDVQNJz=iA@}Eu*?th<cJ{LNtttp?8n3$6WYPTkxz(
z%Ghk_eJfgec&@f~h_v6-uG@zu_yWO)&TY+d#aD6gY<lW$E8(B8>9O!}d0T5AM(03l
z%fQiD|HC@ABc1Bo35*LYJ)0W%z;s|NIj}eQrbG)<3LIn~lEJBRfUNOPJKs$qmvt+k
zM{ylhtJ~51^Hmmqi(`SV5L)&1(P@km!8w`{>dKMS7$0?T=Y5ZOA(o4KdMA5g15~dY
z#3_G}eR^P4N$fKFZY-q!G4Zo3XM3SH<<78hfdTI{7di1h6a(O=|Ipe`^`^QXMg7~(
zfSp=}A_Lw)M{cb51U|ptz<1j<fhOZ2DKg2oT~y@KjIyEdT#KMpNFaM11E2;q<V)J-
zIWQ9vjbP_?Q878pWy&Hke6-grx=v=SM&wZhzLgcgAjkNgJC8IaK4=)v%`N+Zr?l~;
zC{AGHt<*Eig_I+E)976{!?7B_zU9J1_=w!(@W)6IU&5C&jl=GLt?&@1UGXxTISHVL
zf&DC$r}!|2Ti~D2!!tLc!>2%L-IT>;dpXA-0Yw_S|AZzAMJcNGVXR8MLQ7swZahqR
z3T||lm0I9q0bJTHgnyD69kUnfdDphvId3Eq<i`$B29}2ei)V`Ch+PVXc!aqVPnK?n
zyzI$@tfUFknS({^OC!3<J$S_9USDW^J@sdXL5JvJx3v8gFbaN66<oB?q`-#3gs~FN
z*?NYa8snMQ80(_~GnUB<RkIp4yQ<EFw4tdPSf6FZIfRo;$EfO%=-AU)q4T-snm!3C
zK}Wmq+S2Ptfe~PCMINVWJ5w=31Y7m+_X9@zcE)n1M;=3*yRt4=yAKKlxjYFo-7uK(
zhMJq0UP~21c#Gnj?it<^IjcI4v=1ytlzJ}XYG<ZrHF&YZvp#4hVhuUmefG;4n_<v;
z%l)#)$Jd261RHV)@hsKP-F#L4CaouHDd^mDIm+N3^H(<N6jJ>CAfoJb0x1!U;UXTU
zUxU)EEBB>SafUx4aU{Oz2{fQGTvDEK;LxuoGv8uM!_<4k<5_NCUm4oKo`CXs9)HOt
zY;VO8=@b>Atr^ZHV7eYOb5p}OKyAxsP~~2U&q`xhY66-}T=%q~pQOtTHy1y%exl7w
zk{UfP+>dIjw-f?un;q|MQvZ6`h@%{<g=v2FJ1J$a53f>wJNMmO?$H1;)@E)`J+#Oc
z`Ur0xIa<`(vF!z8(1#^>OhMf8LWE05G)60Ps1m&Zk*WN-I-ONSLuh_0|7(r!E$g<t
zuCF6Gyk~3tR9$Yzd)+C@O$*O<h^Y;qj?)z0BzPq8OJj?)SPQl@hNn2KZ)bVj0u5?F
z!)<~V^{Y6)z=}-^E9%YxHjYo#c;7beiz(cWl2(#DpbOA@n~R3}iZV@jR?Vs0EviFl
z>v_4t=I8gQp5E0KN3WKk=FQ(6=G_-Tqh&o4wT`i#t0PuLqcN4Wb6dami4S7j&u!$M
z!i~Ka=SO+UwHge$D^;lNiYtFr4KuUod-#Qre6Ck_t&-R=<V9PUiU2#;tXI&@lW693
zxGhj0)ZV=Ddo4Utd{5o${L^3B!1foYWOiI7-Na^;QVNokaj#BU@VUD#9P}0h(t_N-
z?)$EX4P}taaMhW`eNLykM!Dwp_0M)wQ!uJ8Ff5U<c=Op1Rp&H<{~a1W-4HR?Z>{wQ
zbbOhYLV71~=Xsb}vfmL})hODFITyz0m;4XJZm)iqfN~_=PD4E6g{|ubd~W@)>Wx4@
z?s`wXTKC$|`>%d!5F74M93=3)j8<^eiJH*p?T#7U4yhy)M(=gflD?LCAS1Y0RQ+T4
zNV6UpoiNd2$k*B1@d>f^xZUR7NaB(_`L!{DVbU_3psog<nZjWpoZgm4!-Bn|F;yGp
zl)o|3YfU*$@L(RNfRmD8b4?M0qvXSpe~)Lz$`ZlL$4z=Zt;5`oeCIeX>1Pd(+K0Z4
z6y=%Z-;e!$Y4j7rvDSHXV7}oOM^y}lD{PzADwM|{#W`h%X-76n1#CXCICngM59y7n
zq;ku7Dp5jJ_Scyn_J~}^Ree`me2&<BMhtj2(1P<Tn1U8U8(t8&aSUlWr>V5%6xz+G
z|8OfUsQ2gzWqql+?Up_Al}DTQ_Wi7rTWo!Z6%~Kk_xG3byFVQU_;>hiq)3ezR~fH!
z%OWRFg0=QyU1GWO`Gscn(XUBX=7go*4rfNQ&1=V9x*Tn6&21i9#dOTmjE6V<RZ_A5
z={*%Ov;P;w9`(mifrna6|E*6E4_h9eXdZOyboMleP~<xA2tfJ!iJ$F}r#yraC2@7E
zwy=!pzAWD2Bot`P``$0rM5cxx$)iKkALQ|HHs~Rt&JGb@62T9<I;#M~)F%YUkBg7g
z`eAkQQj<|J^_4(yn=wX&2H-nok42Cz4=CvflV!((I$u|F1+MCdWT}ltj3u@mN(&|H
zM8LXGm$S>hYKixn-?6%5Ra$m?)G9S!&+*d-`4G)3wV#w(a#>9f?n4F*F2wxg(N2z^
z;;F~*qv1VI;PM?hmJks|sHlt}Y;RS1ruANH3TGipO!=0icMv@MkWaObGKB!vi-GBW
zj{r}ZlxYD^mw-QbA$S#|;Yrg&H9eKq-}k+xXHazd$P*sIey`w>;62XQ^rJ;#GD8lP
zV>wjdxWn@Eq<;IIA%2#0QqO%NPqh#<l;vTRJE-d0gi+1>JS@mu)>#<^h#I2&*dpmq
z4MEaD>o=~ggIVY~OGu&<h0-6@^D&=j&UzbXz-qQTzv7(+lfL_oMAVNup<?yQh{p?0
zALB%yXih#=8`W$5?5&KkBjO9YMs>(3+e+u5l$O9|(pudrUyIuFWS<KObW{3#8-i#5
zqt>Bmj*&Wn+cL8I;yn>5(dtC(#ODzKndcib4DJ@J6}c^ma%s??)coQWqbYs~&iL2Y
zpuhVQwH>wa_D==|mpp1?5>6Pq{FsRhODTO4muG@|`wKA8nmY??U~EKQ-#uE1A>8DI
zW<Fv;PWWYZKt;#ssyW6`H!!IHPmB{!@xCuBsmE7)kH$WoLM{`J^nIrtZybC0I3Ld1
zzP!+$X1VV3@FVD8zl;PLEs}Zhe7L?4^@%hg8v$mB3BL*s!Fu-Q@PK8l)LzU>C$iKK
z-Wa4lUi$@5R)~fih+SZ;4>w~Ap<n0DL*fd*LYp-hNs+m94DKYN^4xXx>up)x0djSl
ztQ7Pj%xi`Va{k5qPvi_SuW4C$(u7wNWYEQCw{yx&(A}647N6VMxYFhv8m;TEGCQ~!
zHGSVMC`tpDh@-U4C+V)K0)lDcR#v2)#fJW~==<+4g$6y6J?MIh>`dNtDz!Px`g!+2
zK$9$P_@k8+7n+r@F8`pW_O#$2dMiAp>N~sthb`~qt3D;9Ut=ep9}9*`O9waioBCAg
zkl@9gnbO%YW4MzoBAddL(9sMZAp3o1<jLIH_GjpINTk(l|GDA96(ty{Z(YUB`7aUc
zLn$zfR2w<8{+Ih5Q!3#5TP+c*5BcFW_aV=f(kCRPnWN^LkJ*Bqsq+a%E|dGLtatML
z5$!{4G~Ez-u>|&t91N-0Y_$YfeOjNPE)hjFYq=U#6l(kOCG4AS-*Y;=o@wqOb=k2y
zyQYH-1lZy3Sey+uT%A8EobYj!6IsjzxCqUbht#b*8E|0!6&L6pFz&(CE*D)G&1z0h
z%Fl!@O}~nzuT85MmKlvafk_!g=_wi{0TH~)It}3uLm5TZ&Sc(qZf4>v#~cd1&TzSi
z+Aw`0C(kUN99N1vfxZ7gtpH0mkbD($01e9vz;pssjQrv#ZYnmQuOxf(*izX&6Pfmd
zs00;VCN%~WG{XSSlkwoiW5fFJB<fH+EHx6Yfh~#?zeWkL1U<86YW|~)A8<VkIvjR-
zJ@QP(m;>8^73hBE<4NWkkEqPP9p7=dj3@F)Gf64aji+R&HmXkRx{w=Fo0HB`c%^<T
zBM~H*-B|tmJI7koscVWur_vZdb+ZIGhyCKYA--A!htWRtw@C>CatH_}>o&xUsH@|G
z#!~m<V)^)p`qurgZMH95QfEvGFx>P=Ddmy62RKnl`FSL#&2Nw}6V4x8Bav)8lDKxG
zcU0v;#)EewRCcR4(`-Yg$9f#f4)lv({Z4qRG|}`z#p8q~Dq_u<!?W;6$3%J-`B$cw
zFR46ZWrUhFzJR;DVrq=RG?e$fWNX3DJ;0b`2qM&`r47VM0w%p_er?He3mFr2@-W6l
ziq~gIdIT<vzp>ZZW9(=fme=l3-#yM4&J+53Ss~;3b^)$iS*wicGhXgvf154JpCPd=
zUiUj}zuOuA&TRaAQC?L_D6OrY7o%~X_M{>(se$PCaCLvGXfbt{=8qsKQtd|?-EW1l
zauRxSimc|n^6rXLn=jj|l|gUq&#5`pDj1sVH&=Z<GzhXZc<DdL=ssE_cVU=Rtt0}7
zbQW^!CWGS^rm4(T1$7kDIe2zKT>e)}LNta&j%Cz#j?L?@O)(ZW76&0cOrOo9Z`*(C
z&atU{xtYfduyVVe2Pd4oyri=Ezf>d$uiG<rl+0>5X542E7h*r++Jj^zvDq}DaWqo#
zNms2|+ltUy-pRGW<f^W1rX1E>{hVF!6V@h-Z~Cv%pGg!b7AI4h#b?lS$>1fpr@{yn
ziAvqW3j30OAo?}!mh@1&j~bKP`U;vMsjxxGn!V_5>lGNjQn4t#i|NEOe!J6}N}GXe
zH!1954ejJ{n>D8?x2Nc{OZgFusp|NQ%iujAtWjdIB@7x?lYJHs&$Mhr)@zQP#S?Vt
zo88B=+Ol2U7{rZApZ2Dj16!(Ktjj`?VZU0@k59v;1wQXbdDSXRIr^xiIa~|)KTY^4
zWC7itt_I1b-P<h}3$oM{1&88Dxcbv`^6d&Xd^rcy(q_)(r=ok!zwf$w&%X019+Y!<
z&qHO!*uqJshf((vo;mGl6k;B&a$x?Yc3o`GJH94ksra66;z^_nt;)2<>cX=1Wt<JJ
zY;_F6vvOI^xk-rszJUKwRxdkdNt5yL#%x-<W=`vCUBhIJ-y7ePnZ#Nx?dvc<<JBRU
zxwmDGMvC+6KM%qqP(C@#JrX0#<B2M>)EJlFg|Vh-I4Tt=o-sE*@%ucptA4KAg{^id
zgK9#VoFb7I>xAF(SVK0w`lYlH3%I?32V?Ah<e4DX*C^+55t1KsW3Ue;?(uB<e@dp>
zzP39n!tZFP&XyChj)-T;$mOKtCQ76_>*acxmjSTQ_*{4DrkM=37F;{}M|Boh3^Aqd
z`40gaRiIupFzr}7Oup(Q<wC)$TDV#l$dWE8n=oqEsQlW0QIe>$iXwJc2Ri=h3J<Tc
zsPPgg?+%I_#$Qd~%=bx0Ag+Ni7|43C2sjsB=_?oF83Q+puT!E=|CXDCv?B&ZR%;Kk
z`ERIe$%G=1U>{)eOM=T0^tNp(i|qno_()B~1^g)V3kk1{>C=CKmuqNu7qHpQWWf_i
ztgIM!fxjOAa1kMlHIKOctxLEpt<~;8GBEf9bo&0<9X<fl3CBQY2L8t|ho!(~bY!Qe
zHbHW9fn4Q6KSBu2Gv4(ISh0)T2a0_X6TK-4K8o0!<h62GTF~pg$Bvx9h6;<nTLY9D
zHMOVb9|<8iheM_e;f>a_#$Tn7hGr&HA{pR*)K)X&is5)6(4nW8mb3%Mz$C$&Qr%mX
zfJsiXK@y?wa%!M*vyz3#%_b#7u5h?cuu`-$So0H0!f<y179GlUO2_&iDGI$>wOQ4K
z%$g0C6}1i$g^&3y`T^NyVIHJ^Xxw+VoeL+rb*^pNqIB@j`uFZ2oseps={2VG;NHBO
zq#wxB8P?$FVEyr$PudO}!J8Vm9Cxl%o8hRN(>l0Mik#rY7{rz(yB*RjxZi$9jg)Rr
z;dQ+G&e!Nb;eW<*l7zq*GL6daKlx9=4E7LI6qxu-<N%IjS%4JEYPy7nT)a%60DZu3
zSwWP0cz-k@$5D+vyOt?@E|kxv*4{TtJwK<hI*{LGedX^yP<>@`g6B#bKv10?PBo?1
z(s!(e`x!r74U&7MzC!O8FSJfZuh93=HN~!aK<;oRB);Su;+qGeg|-82*R$wYV2v6i
zviq!iX@;Lv>5y^NRS)^_%)Zm??ic47IA+#^)tph1c=As&wJcbao)rVA%<3EuUPVtN
zm_YCMSrs4O(>?c_tje)k40x=|=rFE9tUYJ9Bq<U0tq#x-3Q6x?Q)DUW#?V-5V8bZ{
zNp33`5XP1IJ$>RypI^WWl72|l{c<_@xYE2HukC%I9o73zdlz<v26KF0kO~Utab1pl
zARi%}Qkrktpyp^eJfNN}S6i+NdO_L;(D4f9Utq#sA$$zRTJ+K*0bD0Rq|H#^mZIUa
zbLkOW7iO~G)*EFJBkSjL7u~I2VRKALM0m%wzkN!H0~&j7>E41j#As~5Ozct>6kb&E
zMq0FTU;YZ=plIc1(nGrHf8;l2CvwJ5f<zlYB7naB+%5v!{W%yp=9nu*{RB6hG;ao^
z2Ln`719Uq5&dhp2od!r2KU}gDDF>Qd9PF&f2J;)x1~u`L)Av9Vp`c}{|5ZiGht)HX
z$vDep?PL5pZb5|_N3+4!Kw_k2V7M-E9GYk$mud$znwv6Ei?&}@oV={5TgDe^&YPOG
zO|^u~97Y!vR1W@87;RqP0*F)>+dymMGa@DZ&PRho^>UB2ibln>tK96|W`qvN*FE^m
z6#21Z=4G$a5o@&?Fi=_2<nh1$P-`WkKy7t5R7<ayp-e6XbF)nC>ZnAb=KPGpt_*()
zfSZ)3Z3>=0qMbF*E}u>tGah|5@~IMs2WLyaHqyejuF_-JxiTC2Ws7n%<>y65xedaX
zV9ns&Fi_iCHZ6hAAL8UjXHS-Rd8{iS|8w?j@>|n(nr_{Q$Su+9N#7<`^qW?&iw#ks
z=E!PJAUFOG*P$KWh@~LHu`1Q%7Yuq%9LzyP4^sWqnuCd8QrKiw=(vyG(R1OTRyPS!
zfo^`_MHttP;%66Y!s-NqRcR}#w<lbG*S0)RsEVLeG+Ijn8K)8^_3DK{pR#6xO5cK*
z#NSs|G8z=RsSpqWP8X3vP_1jf!}(i>T^&8~Fo=H=m7NL$KeQ9bs!oh$WqUE<Wkej!
z`_fsbwq~MbAaTMBk=SHS_SWu>+}W^`pT&wt111izMn6j+POZ2)96aqDUACwPt|~Sh
zIP{rsM;ERHRrLtf`fwJhWhu)G4i|-l4{+$zbE)#3AU1;I&ZS%j;R}>4FFGD&gl{UC
z<VBb7>33@z+P+UdYphb`i}qhoa$HV@*HrWialmAO^XR(>aE#phRM)cO;5#KZf=Neu
zWYlzhGI!^9VSa?9<&<R}Ti^Bby;PFM;0S6d64RGcF5P*gx`V-nYs#tgUBLWkyJmj;
zO$Ya?KdNxc+LC=}U<x?AgCi29EX$x{PfTK5O{H1~^&de^${W@*p;MVsvQbdF=TM*J
zj<fKWQZsq;F)|2pO$89lU?l1Xl^LL|1F{^M?0iN2M_|JI0WfKI?LM`5Bs48EA4hu)
zx)>#3k3dPINO0D|co^~}@e;WV-M?v(aJahd4>8iwp(Q4CZHS*XKvAM<G+ZhjItG=q
zixfg2(I&QU`#kYog4kHm^?a)Ye@jo*q8sIunV~gwR-c47Y7y!6!G+}(q9*JenCts9
z_4l}E<pwe#LX(a`0xvs(DRiiUXiGQ(Mj(rK56(al2M~Es7xlesDh&AMvjf7?4^!&}
zO@-N{=OZ`TM($eC4-Pi!-^5WJ^l3=Qa^Y4xOjx}jY3QAOn3z@O_=Kb5sbi*tpOi1_
zpMoJ<@nCB#9O%iOP>p?xkPIt|$e+|6@$YAMXf(Q<tNH%%o>w=e(P1$EK@9H~NE*O&
z;VH315ZQupkjAC(3dyYcHvg&Ympbw61JDoDHRzfS@vF6)`J$yDCy3*^K7@%#M8k6y
z;_4+JpYJ87gP`L#el#(^QOn$X@Lo3LrQPZCV9el?x0O+Si<8FRLC?=@t0=A3FL(Qz
zW2!OLp;tCUs%h9OMbupM@vo9i`VM7EE!)cu4jaWiz-VZgEFW{?(S9~ROqRxvb(j1>
z^Ud+6x&wYMZKbrED<D0n<|{1}vG(b&&#v^>*qj?Jjki<Pyh=qVP1-vnK`BG}#pk!~
zzjP;Yk_xElvQ?%Usn((^<i)bgCa12TGI<_`G{8iHn!KEU#gA2naQv<f{-b>Z5b}yX
zn4M<Jbnf$1cLoDdUCTtdms8h&6D3TWAzV4nMlU~@T)$^0;I%HI4tv+x%cu8%7v)~`
z0$xqB<ZiI21`$LS_^;!QmVFz#_&QY&$RpI{r0<4MkkHbOSHQ5OlqYqA*lx&~uYzf=
za*s*jCsrLV-kuSB%?H~ecTD+bwxNM#LtOwZih-ne2CFik#-(yVlGo~gQ-u(d6zk>Q
zmxDU=FY7KAu-%tQLAf+TN&ErarK$;2K??;!%OlujYf|MdJsYF_l28O81SNzdf&wo9
zDk4=MgoKtC8+wKyz;X5pAtoM*T1$63Oo884`)kCVV<nQvm={Qu?uIDL=yVxuU-ZaT
z8DL|NvWwy>&x1DJZ<9KwYKeWQgHw>MPR{oE7Y0f!Slrx0cB&fY=Z$eKLX&1kZD)WN
zKHovxFx(JFkxD&5PgGffVqbaOjYKAZj@5h&)P-<k1m49otAl}aH4|SjYHB(K($NSr
zWg2utMxmb$qB&0Fk1>gmb0Kvsyc<DoBjd~+!|}Nq7f-{oX0-K@$X8F=VohA=r4&c8
zVPS0{3$@X5dsD+UMmPH*<rqUqp=<@(qZbGeo`WmQH5Ts1J|_fbr6nA`x>L0hJ$Yx?
z8pj}9D$^**z4Lxz?V0K~=QNKxmS(1@##M~0Guo&DtdG7)9!bKFXaYxCMWJ-mEA$Py
zHI35O0lI0ku8K+Xd-}CNuGtxQ)yss<&)7FmQu7@iXJeCZV%|(Q_)%W8Dw_Bfk=kqL
zxfaQ|&#_W95ibc#F>s}ys2ruQ13ym;DG)Dn>Q`eT3}t&1D=M?y3|iNwjZ<p^HP=3m
zNN&q^N_^9fRp;hWW-^B#!uj<sGdQk=3jHDD-4+QWG>?0YO2i?IZ5xmwM})(->G$H3
z|Hee$0jW^D0-oPRHVLU*Q)J3Pnk#m8RtCJ#fgmmMvJzr6I?qG~h(<(Pz9bO~A@}tB
zos95(-Q2K{&QAK&XV5|7uD~*s^J3-M!UC2zwNp?o_&e-CjGsnwW<Q9{>7#etr7t?%
ziD-RVpAdD1+oR)jp(d$*qjQ}SynDG@0s2kEi$PC=j}_S#C5eKnCC~HAG<%~+@CKcK
zM#&|jO6#dhze$+yE+_K)##x9A(Zr2tmlo>SrgEp0ranNTU)IuTYeyM^s%rE?-)k<=
z20h*|xtyIk`F8zYu(|==n`GITpfybqv*%S$ZHaCe!o+Zd@7Pq$r)ytGMXT!&WT;Kl
z*^8pDQL+D2VHZ`VVX4|h=}qa^!r;S5#$PTXdUpjrb5o}Ed$z{PeS6B(9>X8<s+65P
zC9c-FUrz$obY!Ocv8XJF`kG|&qo#LhwDlDHdrda1_k#mXYQYX?Z?J{p_)_GE6Q9ke
z#_Fb%4xC%?JRRc`&`O(h=msC)Nq;YnlZj)|=MK}w$a_D{w8}EqunluyJicRJag_!h
zU$HUZbC)4oyqK)B7e&XZzKp@2VUQLq2r-l7znZGC=PhQZd=5JymBko)U{THB4j!1r
z<1}ve#R7@EoIu7M&dEH5j!FlAoYa&9GeFG>kNHN}UuF058o_l*_eKyZ*4Ik^TN)B*
zqvxUr3KJ*}ulUX2r)xpHuG6(<#6A%Gw0fm&Rrlb}%^^@P9C3d%N_`ddPY>_^jg0tD
z6wjZ|d<#Sn)AD4!rNl=N#aJTDV)YT*8;IA8vEc7vi@z}=Toj0Y{=EnO|6j_)zqLXC
z^9!5O$^($cnN<Dpq`~moGVsHa;|Oo&=1}g`Oc{%m0SE}T^S0=AoTLD1v!7en*Toz|
z)MoofjtgExe}BnC1O>-O_od?BUb6q<17TD<g0HkbltFmk|H?V~2hjGe?iW4~>%Tbx
zdaQ~cVU+66hs5;b+RyO!FSz|;uCd311JZW3I=;7U|D}ehFhvkGf_moK*8lp%e?bi5
zx(B%eUT_5D2^#^IDfaO#dab>!cjt}EIHw>=6%z6%nks0rwGBaw{nxRWQ_DXGBG(`S
zp}{uz;?ZA8qW{JJ?|*+u`;a8)+w#)Wh!Ow&f-0&&fWPwm0Xo%~pr=>WwH(bUJ~+C+
z`qwumLw4fDSgGP#$d>SLmAmW#ak`ykzVTImP3!)H#PgrE_kX+(AHiE>YPWeM_m5}A
z)CCU#0q6qmzWe;Z7P$Y#v(?pwZk=229o`>O5>TYq;#vHP7FgSMx><L+C?Bp(23hi-
zet_$VH*j2HATTMmUTp9_Is{88EYVR19Je#*{nFVBTeiGyK%I!Xt*4O+q5&%aK8A#I
zczfk?uT8?|_HuF03^)%dQMYHgKzOzh%JC5pqSKtbTl7RNhQ0uGI9fzUpTt#{L<qtU
zKfvqP@$W14|KGz4mgiH%K3*!ca_}#x6$SLggoZ-s9UMXEA8J-^FplkhvBWmyiUtJ%
zpY1}lQxVvkxt`mrueT2elM<W}RGZty;HqB`%~8M*AVwrGA<5%00Q7`41r{ECWwUoL
zxiuO2i&Uwq5hl;Z*&zT6csLJSyJGWx;7JMt)dI(_x<$VhyQ*s>mvsbPS9UI(#!qkv
zST03@>7E)3o1p^`i8z0NzazUSli<3K3g`%XAwX};!B#tMzH;6U$*Ya+0`7Lg-ImjZ
z{Sg`5P=q7v$Gq!T>K}G1uL>Q6dr9eWXa4W+i)j*KaWHJeeIEYj#bo$_2o{1ok|-E{
zbb?8b$g;srI{ip|yZ_Z^f3i__d?Y8JeSQIKZJ~%Uyczhzj=vX4w(n`v;yrMgtVz_@
z*0(9zzr9(!<waSth!=ZCQUKmIiBNDU61}=!yuD_Vw)&yI?N*ba``5Mmj?<@>6D-|w
zk&|hQS5wa}k+u1q>VKfqE9U?o_m9P!oyE@IfRvC|(Y%Dl3AsHsxh3`k4+^gDBu!sD
zQpf>-zVJ{1Q{NBwPp^z_aAWYy(*#YC@mN1q-%AnrQlipWXwkU*Yw_=%`rp^V|Mf$<
zG2$_4IlG+tXS)q#gUka*`VUZ$+Is!YTBMo;8gVwWm(aL&Yw_P16~IL>{1Rp9v-nBn
zi<|R}EV*}lz&;gKBvBSO>_OU(ZQziR!cVAEfN&cSBC%rOkxWAu;*^m@*V8RxkRZsz
zRvq;F(I!`P%|}I?_YmysY0G3a5+Br~9x7+Axw!=ru=5?*Fe0rhey+9*@SphS7xlmX
z{C_@aI*Ztds-^a5JpO|>1QG?nXKz=(8Y~==xJJhu^8-KI18@$MDO52dgS@l0Dkxt2
z%Y{MX6FB1AMRYOXA-NV>NUb*b{|9a^NH4@6!FWP=W-CD6ya>a;H!&`3{Il<SexyF4
zT-dw|f8o{IQR%?8k2ypXpY+dm#D#?-an^y0!eI~%2Lds)8~NA3m_Ud31!R${IS&gt
zsyh3h0*Velg4+E9fj2#yYrCG)2)4hXD&IZ3X}kH^76#VLblEUFVuR89`chjLV7`X)
zAFoInH18wKQD7~Ik%TMI_)d(eQW+ma=}dE%GlF+_W+0&w=@k4jS*`%F=%JlfAtoDQ
z@rIGM?uUl0Ena_LRP!d)$puasr(YKzjE`9cvDt1#x0D`XA!t`#fDq?I^3f4=EcU;G
zUE!IPBLczo$$QxD7rAk_*bZoED`4LtP47N1xmhzAA{axk%vkAHnPur0sC!-)qZRu*
zU<TZRRp4Ml`voS6AmCAu+@8hXlD){gjIkD__cbqTGgTs&M-G;IKw_B#7%h4YbWRxk
zr>&<gIjkCn)f`|Gc&$B)-~xpq?$5zNw!WVn8f>pSkU2d{X_%)u&ZWp!X}ig0u?G0d
zOZz`>h;9gJKM*Y0)v*3hXqP1vccuFnwVldAjr_^s4@B_F*{T2SsbtnUSft-c#7hY|
zz57*OIy^mZaZ}HFxz}>4KNmJ<JgYJbx!{_&$E8O~nrb#l0Z}IXz+dPkv*byXQGOGC
zakN5nTTj{p_{+ol7o+ihmR4DZv5&?7pm)N<^~j58>MwVjKKdolJn|lRQr7ZnG}@Rb
z)x_^S?|jrlRFXV5yJWEgZ69BO3Lk9jAB|_nkR4zvm=|^PNxV36C{S<Nrch$#fv2{1
zu#(_HhlYWSIH`l*um1h`CEx~v{sX_E-}#yu=-c7!`D}>%M*Br8gn|u0UUvtl`A^?_
zpXgS`oR7qpF~GkYfgtlU6eviJ1h;IdIoTeyO3dF41jo?gk<r%2d;HVT3T<vhml%wZ
zw*{v@;80Wj8g@ToFJD2SB-YqJeHCIcEpXLJKZ1sQ^b<Jojn2m3-dJ|z$^i7?>n2p8
z#J#a+iym_hw7t&{Uae3rlIb?NI-`=Hf+c5&2fGyb=dcRoh6ZGTbY~@KTQlGT;R3vz
zOyk3SQh0kPsWBM>e-jGUnzcNMsb+t!8`y5|fg!SbJ@9e;i7kM-#mE2&Oyvs@m>f!T
zvmY)WvND%|7ePs=t%zTU00mKh-gspueT9Jjbb_tYJUJdgwF|=_Bw(6aLiYZ#SrEV5
z@<j9uJc{?rK_M3qgmg%`WlC_T0}TZO8(9RjQ;DLH5a@t<ps3mrniu$*Ei%>j5HB0*
z@C)R+|3Gkk1;Icl<`-(|9lFnmUQ@^eu1+!b0c0-vyA_Rq{>8@OLlZ;$LfTjt@R>UQ
zaEUBLurLemA>&EYuWc7+Ir1!1>+XL(?ZCG>e6{d92LTjohL}xS9PSK&jl7>B1S@2(
z#K*`YmhUcxV!68u0yLx;&{x-b{82Jpt1uPUcT>g;pgGI1i^q>^*WQmSuo~~`OSphL
z0x+*$XaWKdWlXw~(HS^O?4!75W85c<H89@!1NV-y9E<ek?6Wv}2@)V`w=;Nxp5oKK
zBkGIifayKN=|JGHZm@Wv{ejed)mgE@1P~BV(xst=u>nYmBG?UBb$tn+&>N7#*gV^j
zF?8QbUBPg>gT)=v%~Ll@s&4bt@+9&;me?a`jz0jUulTjo0KdPfbXKU#TMx=mSL#5@
zB#NCIQ0Ch2C5aRDJOaEbwT7cz&$hY4!Y?dpX$S4#w~K$X_4cO!0(S8@;u24AP<p;J
z>*!7cqcs&2Ni!8k{A_9e{Hy;0Bx4od7vjs(eFe&la&4iRBEvOR1;-dM(SHiNcKmnp
zOYaKe;M~kvP(=N{9nMnjw;{5K2^^4bV*ObiFN^T;OoOKoWlLxB?J0zkAR~DAsm+t}
z^gZ5B(j|TJ3`atnaR@arC=w-cht^*}7(u-mt%e~cRGH<?9XR9(8u^)iXclM*h2i)G
zb8#31FE_`_0So;FXg_9&vu*X)$NOQZbXmV~?rm2)7V1<yI51H#HH73gA#gdv>FDTu
zx(|e}_PahGH`rj4wt#?xJ%<%mel<ioe61h11O6fIR~E!U{Q*8l`15bodorFRGlmcw
zuuTIJVb^Qzfid156Q(Mft0jzKeG=?hU@xr3zena(H6DYx0mj;7UU6s#&JgY3h&v?P
z#usC0`c%+>T(^)3bRpDe5yVk3ol>nT40_$Cp`fh!=CMPF%l5c(BpywKO%*}mLD2U2
zq%DbY?k=?*zW$_p<s1kZMb;#+2`H2n?xRt>BWGrgi@Xn*3A~!9B7qABah_z6n9>ks
z>O+*y`>VV!0uvfQFLkakFj5Q^?J!2LA{r8kDpHRK38bHqdf%bRaIj>c%kZaCFw~1t
z`R4ot<SRu6p@yPr*TZytaF<b|&Ox#XuOyyX$u?~`mH95+ht|0@3M_(wT~KmbM>7Nc
zUe1p`M<pUFHUdfYkiyQ@B(wmQKsU@%G$GJIi<OOZ{{T<m#N+%->oBPGrQYo$jnH9m
zXaoXHCTVV=13gAfB9o;fa-PtLcr>B2`(*MzPLOg3HT*}j1E6s|)n|vtSy>-JNASUV
zg50Cnf;(1NF-zf^6AQHzQB9`4ttPa=U~piXG^HgC!e-!^g?7HDYfW67c%tYAi8eaY
zNlDXCd^SFFn~NGW5@ppJ&DlB~aaovk0ufEsHOEvT^yTu>EUue*Mrn}Fn#|H+4G`|N
zNRo3{l|si-rm|Vl`H{oOc)he}sYQccNSZXqc$yEcNi5~qu3v|GA^fO$0~C%3i>=T~
zyH|K?7Jryu@Rg~Ltbsgv!H|@t#Ks#}7UDgl0j)1?4|_)5dj6|!3iN<*m4}M<T%~d!
z!aD0S8W<^45TLf$!zhU-fk+`L>nBz}Z(=@AD?WKcSfK%#9}Ru#@aqG_FRy8$){r|@
z;cciNzE8S;Zm<i`;s^Fi1-y*BLGoqI;>^fujKM{AJJ@^<HR3t_G_FEyG4f3GvCrFk
zwe~?)MyTKMgt^@=Hb*w6rqd(tXYr*7_v6bOP5XVv?+Zn;;NgE=rQJ<NfdaL&;8oL&
zm!Wy|Drh0~VN=b1M(QBpoBPc0zpTf9RfNFiv#c|3{ni)weEN_yx?=$9ML-_E|6+mg
zoZZ%xVojPW>er|z=V@eFnH>IHF)Ayz=z;6oODcYN!BEhnhPOY1=1(Ya=-6fT@-qp_
z*!tw~4f7R^pX?T4)1}Jc59TwiY-#T41Yg0V<q@XES5r|KbRN1CMQ8Dd41yh!bJ{@{
z52MMod_^h`Bf>&hkOEPEDOm{hAbW)v=io~X*%BCm>LRa8CApkw6y2&^=LAFHp5M2~
zJ}Q!DLS^r*8<^ngM9vR(K}L^;He-n>Q@;5`TptYn7{OSg^L}FhFc#%7IVB<~dgVxJ
zIh^gwkd!$M<qR0ylIVapXpoC`&r-^6+G1wUlUb*MZoI~CwGS7+Fn>O`7x>SZqkai6
z+Q3tvSs5r-z*s4s>`oay|8pQ@_U3W07Z`pWog5+2sFU4~a&afI5-Q&tpAca0?5S&B
zW7j}^4=Zw8^i}@?*lsX$PsFI1-kAl<foi9PWUY)n?>a5N=1g(NdHf_Qpt$K4UvFks
zcW0enbQSKSe9`_m^#%Q!LF%*1-QctotI1+pWTCM|qR>`ZY&4C5hh5F(oWtDr@y^G`
z=sNwlv9ON>6{sv*#(Gf!gUzP!mJ`eOhl0e#sG?uwh;j2b&-og|HAMxJkShca<?jNc
zHw?+uOAjY*){Xm6pnAr1_Olwjf=3XI&)%y=%D14t>_Jd5ef{ZFWgX;yHHF0mHOP63
zigoncQPq^MDU+1nrC-NI2+FuJAY63{{=eX+cz=xCJIk<_rTi;_<}9oD(0uw7o#=?1
zDz@2&2CWa*fM4ec$}x6aI~jh$K{U)Rz~ZX&2qU&Ujr$!WV4Q|LzN1VkAp6%putmgI
zzk~N&nZXR75+s`8C~8<G5)08YNY3g%MmfMUr-=TAM+%+ilDzq4Eg@11lT4SCx{Lg{
zJc3op2988pLugYdDA^bAwD|i}ONC{|DABkUGJ_@#a1k&t%IKZ5e9)I)g^3Sxqe>n=
zX_6YOd%0oCQim^ZoX}IfaNHQo6hfgF6%EgM-c8ehT#h{kkpZ5G7Bgceor%_oCG<7@
z$xq9<M~Mn>5<Dwn=z#Fb$Vf>vl5ULdxQmug-JtW-#VDV|b3*8d^ma&$?t>Hul7LJ+
zzO34e9;c0gcn)V+uXV}_l|w9BoPJAU(@fyolxTw|h6A=ni>}ps@Fd%iO_pWPm%1V!
zV<-4Q^7_U7w(-$h29#X3A$h$mR{4@WU6Jb*1UGVRxn-}B4OL#RG#Xwg5*nlDgJ(4k
zazgD;50{VK-dKooVHVa<IR_xU%RZ@$iorBqMO)M4C?oW_<3E@3fohA4jJa*=-+i$0
zsDRUB?<MOS9k~!Q`XshTr@kTgqGEnL;{jz*gA0nON#X&af+^jf7TR!XVHQm&(lp;Q
zvik|g-_$!+Aq7uRudEbuX+cSA4@sBUg~UVzG$Bn6ef^gPCLI&<(48@o{4(BWm400u
z|Mz%(Q`eB_?h^>MPG<eDuK6oLG}6%e{cZ<#rX9H(Y{Y3XTqx$qXm#E<FGn!t4B4f(
z65=)lb?<s6$!R1w80w1*R;gTzASCs=uqI>GgGAyoC{I>{MSHWr){?g&br2&*Oossp
z`Z#NgiJ??0IGP_$p};f0HQgYP+lqM(j{eR&S+c0&uF;WMsrh(&tHR`0+o*MDDuday
zyDn#x3~Gi*D;!~NONXw>YW(CxWorUV3(oNgye4ue*q>g0`w&#tj<!#ayN$~Go30yG
zPoHX=pHio{ew9eCmZQGX5GYn%<ld=V_a)NQFQ;g!xVlPZkTd94RGbv(#-ocZRu5At
z{6CbvbzD?!*Y^!b3=RzrAT@-Pba$7efRxgWNOyO)w30)rh;(-jp_GaOQql-WDAMmS
z=XG9jzt8nP_vigfVr2H7z4toSvDW&1f58Gt{Y58u?b#)7izSC7Kb~zCnr*8#QQ)MF
zh9f6GH;q}o_|^te{Bc2%YMO@r`aix)yf!#KTP>Uw?_ay4FPw&JzGyJm{<ggz(nFHn
z+M+qDxkvVP)a>WLFT5Zd_mTuE3*#?<qx!9Je_Yob>}k3I4*~ImnRD1Tw2j@p;+IeV
zc_>-MLbvE@DAe4jLa=#N@34dh?MC*>VAeg8kw(w$)ba1F6;v1_acrBB0y<g9?^3Rn
zZL?nV-rPGG`q1dzJB=+=@&|t&;0uZZ>V!QV2C$EPU{XWGCQt_6LiKTj3!37WRkeuI
zXI@bD9#^QE^y{~cwv&*=Gkrq~tOUY|Nte}Rf(LGycTAw^qWdTk?sti_dX8rsV~H28
zQ;(_NA=|2x?&HRYkY*!d|5&$_I*$9e<r`$UKaSl6OlvDEjg6mMl9O>LFn;6lH5skQ
zPfcw=0Dsl91edi7t(KvZ9{+1oslA;H>rCWu*2<k2J=aQ&DlU4j(m9bzIIfqVGmx;$
z?LU<e4l_(h8IS!AYF3|XB+k~-FQ(=FDN|tYbdxDJdF{-D&F%qG8!z4l_jog_T}Bd0
z<SVQH13%z|bk5MxE5>JX_MzPjHCt4PMy5VtrCJ7QpZYq9PwL#CblSa(y~ou$<_^`-
zZ?s40{KZDuJ;aPPbt+tqk^<Ic!Ek8ghzRv;`K!d4<PK0B#hd%_^t3Vncq}zrjdlt-
z-Ytk&t%o11AkzOhIyIdd)l^Ntd3vj^ad2C9PQmk$P4Rn{6bCw+JLA&{)nr#sruDLx
zD2GM8V(T6Rij-4N-c0Uu0`JeTru;6Z_j(C#MJ^0WzT_s%SF;QW9F}_$2lo7Khlh)9
zwS>iZI)s&0AN5W!A%>y6d%@0Z!QE<fS*wbz<(!in`U%o4lhwLXYj5ur&A#p9?wsKE
zx2!46ySI|lC4%cGZ`dI3#rJLfwNq}Tn%4a8veFafujOymoVeCs2pxc$j9_zzm;bA#
z?_p$1mLZ*hvH7J9FX)L;<w2G9qX)@6kY=Y`<X7kBJ!$vk!ybfM!0?6BOmDkrsoIlt
zDzzQ`o&KCL9wEz)_kX;=J{SmXsNCvX(LQ{0PNy@Ozwd->4zAn6$nj$PJRRR>n9utx
zzu|3(*UH(i+$2;N#rxWS=kGPqmKeh+mdOBg|50^rCqx1UHb|Q>GO$=(t({mquJ&9C
z5xyU=-c%`2rePk1E#SYDBelUHy~B((a0>%j6!n|h&4Ph@%nu!lI=zGGLGiC>;f`K1
zB&+pkc(_bfC`W<%Q^QmlUN!%mpT!?-ic)ke7QoWO%~_N2-AF>&p$0gmVYM1^nHIP&
zL_WMFs>wyFaVV5nC4bF#mTohE!r`l*InP4@LILx7fJMrp^n=9;imv^;W8pqtpr;rl
z)MV^a52j$p*CtdV)0&P?bZ%c^STn+=F*okEuADC7sS!Yc0Z(|#QB%h%u5MVQ^nl_1
z^G-#|*v}{eTca*(DGqL=Rx2JI6}l_Qd7%Q@nh)@%qrg-aS<lc7=g2)t>!W`+<5L^U
zr_!p`b8?CTjdR`|TV2j@@uDGA$8g|Orna}E2$}r-2sgo>WA1a3nmcNRsf=&`Co#d$
zm$>LA><{<-ktP{@+(;|~2w5u5XCpfr#(CZw>+y%M4KtJ_K?_LTp3x2kK?KKra+|Ab
zhwe796-!8oZ<nS&D;P7@*SF~dgDOs%YGxxCcc}0TJH$|>hDq6uNyZun{BTmHn3SL&
zRxrDv&hd(Et2n}?klFS{oAc-yE$MHoikI{so;7~bw*VF9x<ACT0Gh*7)1x6lky7V}
zU<D+Uj=?0y6v^L)C`vO&;wpa!<L&`0dZBlY+C=~RURo&>op4C4&QLlv(ha=i*ZX$A
z!X&;xpVXR_h+2V(B>??a%3>qE<5zelaE9!&*M?{D#@M10$sW{zK@{}|_rKe3%$tWv
zVdMk;^??cAQzdHMF2k)TbAt?a47jzJ(Na@e_=j><r(NaS5vSapCpvGeQjK)Kh1X!p
zK1Fduj+`yd!Ke>W9CC?a<c{P=P!qEM72zlFD^$7<L5(sOlg;XW=3#v9e2xk*B5`O)
zZ2h5<k9s@QPWTwSdazTc85Sq4OwR{6E4A=0Fv|IYE9qdms~BqD)t?t<^>c;xm~Qk2
z0B`J_Ec60lCxSc79!(7v<l|DH2V$rZSp82H099NqHTj9YdBF~wzp_fNV)6SJp3nlq
zAUc~Qvr&F)M-@`k2<F3SciwTK;VqKMfb!o7LWxaLzA_uWGcOm(tWEflG|j{l@k{U^
z6xUj7Nwt;P>{WXQA#vVs(kl<6_#lWi@p&THd-jatUSS>RDuFqK33%j2#_B2P4zp>V
z$S!*V!-pr7IM8#SX_-IIK2%6#&sUyEb=zRT4fnxpt5V{cl!x9Z&6Tj9?!6h4>b8uy
z*C!2~N>Aj#TTRZU18jaTVfZw7rQ^Up@JDac=l<j^k-C~UB%D{GR+Mb8J^!Uc)8yTA
zY70*`($={>t}Epayr$W|lP|ZQC%?XQ@^}kE5vkhHR|<O-N4-P`xRdTM#Vqn=9C5n`
z37SVU7=?HPF099EQI8G7-_wQD-u1)<ybm*hPD^j)tT8tQ6^Mdq18)D_U$Di+aEfCh
z5_q`EBP;u0Z~)(C036m+Ri@UZx8LZ$VLe8=vwqg*Kqy#ONRHw%&$IHxl1F%RjDF&5
z@Hq=UGhsgefv^^4CJk{lZ|09b2A&Tu%OgdBuSNw2W+)+(koP_Q2g<lHb<u;jGDKMY
zjV8@2u3K<JIUqG?{IL*>G+ae&Rh0}(Ac;ax9l1v;RHa?}g+IQi^oeqUUMpr5n>;HG
z$~L38mqe>TxmZl4JOb@YqcOg!*2@5=9p&W9A{s=Z45ld<2ANmr{GHG2w9#@@@__m&
zDc@=mHKq*gp1qaf4;no~v|>qxhu&665i~m=Z)D{zJVCnfnf>wDfsn&H=CYdym_79$
z#KzA1-Cx9So=BIbwUltk8~#XFyW3ysOw1M48_oh{j;J}{qs_sbew-}gNsqR{ZXi0H
zcJwXe?@tmIj0t}%1N@=s9JR*9b0=hPTV5xx--UhIifNf((AB0G<|p_9h}X~XYEc9$
zK@p@?;%u$yttp)Q>Y$n6Tc~Ft1mPhV_UB*vKF^^k+kpawb5aWhAZ_yBN7btPf@*6d
zDAP&gfhv?x1!%w3JI&Bpsl+;slhZ%@1~$}6Dv>o|f-EW_3o$B#a`ey@!wy+E1)?4e
zr|Kvp?ws5#u$FWBn4|RWw06NRYz6~+8S7PA67d#%5;XajY_17vD>dOy(+ZeUpTIZt
zjP@Q({Df{NNJMy|1sjGBx$7DIl<1*odxjfdd-tVu&9j)U@LF>Q7g)b}VC;2RpcNMu
zFE&R1vov-LbCbFDkA!_|L8CA#57zFCw`DeQc+X+yOgqfhg2-Tr7;@|lB_eMUbRx$P
zj+XT1@#aZsnJk;)yuhKlIZlQ|xjMPi$WMC+Pfi22Dg`v_E~TK7VDqFx{8FO%N;*T@
zVFs8&SshT`u4qetsk}Xs;|GnDUL928QyKqgz++iUO+!XeA4~Anoh0%Pe{%#9$D7Fe
z%T`oZJqK{>h)Y}jax%2>2&ps}72-+1)z}m6pXg-}OX~0;x=cyPak{BQxU3^l9&Q!_
z$o;o{zIkP5y3DzZiVX;QR<055;XANs5Pgi7YYknK1zfR|^WH!qKOTBI`cXLvMQI(f
z(U2YaNgEw*2+JpaNT{KE+4~aCvJ|uuW1Y+3Y$=Yxm|$WWG>V5Qy~J0-$I|D{HAT|&
z(z~VVXpwnxW^$4;!y|!NLbj;UDNC9x9+qNbFE&!I7J*Iz=PE&-CWMAXV8k8GqX>8W
zuhB=42XvC^`B*)FB{TwED|vOAHxuPw>uMHdUK-ys>EC+H?A|bMa*<+e)j_>yl6~F5
z9x`vdJN<dB?|E<-s8#Fn`t<a1et*0{n|P06x87|7)Z&FAND_@aRQA2al>_t?!KdGE
z9C3E%JJ*Kr$+6yXO0|~&PBrb2Tx=*a+yC4?r|NL2Zj0!sDyS#a`NPU$s{&1LmsR3g
zXOJ*IC3<|OKAWKPTA6JRk&&q#UoyJfz)03jshxf_dLhd7w3lhL!_VfS4zpsbO8lGy
zO<7U=W4x_^WZy9fp9}<i)Zcumb2To*TU)hqwQRAL89hshjog1$BsB`BovX}n$%`$J
z)9jpUM<Kp|br)az_Vm%)E)O<FPUEz@7u5#&_S$^~pAfpA_PQK(Aj)jHh}fv0S@N|w
zZ&f9Y9E1v&w(4|PA#93Y)s3Q2MBt~W^HXj1_)@WxR%607R)bc+3`eCoY$tE?Q!ykB
zp8KyPg6a&vv{I5h_cRTGdxN0+SEoI~UlmCz4z!IFurea5KW)GQn?w5X1*76_kKN3B
z=Dkm)C(@4<>II&nbE>||c+R(;?O&wG*k9_ad)u;KH@<OoebmY*D2|~*nKSR?+k_zP
zBtvnTB2+kNg&Ubp?M9R)`P8jZXl_|gm}^9@8gO-m(1ZVOjG7nA%d9n3YN(-bD7Ah@
z)?VCCizhaO7;H0Kl+%O|3(QJJbFmDx>bWcqE1x&!AOA^a*wmoBU&>o5<T4<zVP3%D
zDDasjewjqEy#9dix`OTwaV<A31BZf9Q!In@uZ`<DPYOl3#Kf35X$uAhUc(j!%3o2v
zBQ?P<>>4P!;xUmcWERd^(p=5M+6mXXLuQXt6B%s2Tn3K<t62e>8K1pf#Xq&J&B)xw
zS`1yD>`%jUcD6hC#MUK{490u!9IuL~$JlMii236`-hFjTF<02dlnk9CKF`MHU0|V|
zB3B&43kA*$Wg{YvF6ZI#P6ZPF{O7R|js6`#P1>R(-IF51yU%(1RIxgdf$9zb_;{2t
zVyXBJDC>)AOq6KYYg?~)ia)J)wvh-RRR4}{Bd4&{MFGrp6@SLhQ7p74(<TWeTyoxa
zVs~m%anfECVrZK}6djNd&uzr4Wh0*t&VNmgdm|B~Kzxc&@lLxgXKk2g7&oM3dAIfM
zb)?;Bn?`{=3uj-&`f*Xy?R<`vnWfTT)*zgqq^g3s4vN5G`qpzEVl#xwX=iW^g+bbO
zN!8C6*R{T#fq1Z!;n-L~?b2LU%cM-KH0yvMhcn3!EM>;T+0{1TJZ+FRx6cv^ZApn9
z@p9rCeq`Gp=4?k7EB#-Ue-B&r@z^SO8J*Uy-r&^iaYAP@>?=Cg4j|82aZSS}^Dn9m
zL0cw5+j@<aAy5LPVNYb4bVF>~srSw;1Ag9u(5uFS^{35eGPA$gSJ-Zm!OVy^OJWbR
zf<NQ+mT0=qPPHBU`IQ@k)}Ha7=oW1Fr1kT-K3Q>fjl+l8p3EH_aGA=xF)~b39|~0m
zah4*`82DGp*C<)gT|o<j@-_A6WT}K0<69@Py^**1#^CPHz1`>J27$^BkrA2*Y1L|n
zU986IBw}IgF~_^t64Uz;q=RZ0SqKLb$vI3GWqc+sk+7O2@qsh0NFWx13#)@zKp9Gi
z=wGrwY%6C|Zr!;Xa}u0=E=yr@9N`Tq6W+Vy<&et5ro?8Vyv5lq{`95jhf4||J8Nkc
z)k<hKK@PFX+RzWicJ;^hHt^cq!}cl|Btkv|``mXLLtt{kD&dO;*;|ddHxIT$fDPQB
zk<r*IV={CWm_W}JXxs(T+1BmYL%+D^T94jzy^JD-)jPa2;$c&(K1BPKaw!Jt_@jk}
z=9{9cJc))pv1mDOKs|IA<JLFnwZGd$`j}fja7>xybg#=f7v65m;x|ucrcotS*8B3>
zS;(hLcH{V>DflKf7>al0>ne@NhQ6I#)ye5~P}5h`R+^Hn4jE$B$!WuHPoKhsnajJf
zlEQ5$KWiOdvSP|?hzz>Gx^GQAPqNe~-OlN{uMM51ETwDxonKd}+F>+cd(Cv;$ST)E
zxYt&!z)246$7AV9z2e!s3Z~&T3Y5iN4`<{1E+OuBUKUzM)%m_TpuotHu>X>Ww)!L!
zEX5$RFaOYnpIN6)YSZ~_bkn1p5vw4Ui3rl;_vMnmQy90KsPrpG-iiCATum01{ESDr
zh#Yp2@p!(tut}ZSMo`0WUFED#XJ8+*&1hCtV3v!GsGSv`m=<Iv0<-f4N7*14QE{5I
z@=g&d=nB0oJ&;?%z&>hwc)?XFtWr1b_y)7{7Sp^175<w=n)S!kQdNIA=P-mei-jLt
zrE!eRU3UF3J`*8EqsJ~u!1h6we_XQQV*`A>2_YK<wNkprD4lhPXQ$r4M>1F&zvo%T
zM6N2@xC!%I!+h&`ibCB}@uA(h#dTF>_K;Q}?G!YrRgfO@HO@+MOqcDf*h2&-C7|V4
z?Q3hwyhJ@NyYaRyx7?W=tLy%VMH!l1o9>G}2B2p<E>MZgYxUb@^Y};gJEs>?#M=A#
z;UVEY@zL;A!Q?k&K5?EX2I`z<b7_Vu$a$0Y_!QFyoY&nTS+j7-8Y1ll#Zj5N7=)vA
zvC`_8$)X!MZ0*M$gScaU9I>Ie+g!I>cvS2RlE;5)Y1<Z8^&*~+AJ2!Y>ufOJgC(%=
z**PDKb|#dCo!kqHjpoY{sMD>_F&F{6lTuedLTD(F?5KX83QoHeJsLt{h=h2_&Vblj
zgv^_js+Vjx#fJXvdL5%M9ZcB!9{ItDp4^$2c=m&%mJCP<-=6alM>^gEIxwJ~U}#7=
z%w^+k!yES;iqqeaNID=7ORqVHH{Y%0;>;MyutlnJRF@u~JUMl~CxXxA&1rxyC3rH$
z^(&&sV-4rCbWP8L+*Y6+x!A$=O7b5IhpN(M6zXO1puo#VC2G{swjWb*+#3ZW<)tPI
zij6Cz(s2~&rE7v^smyGMGrkqBv7FuW(eG3w?km!qg3Tn?5-H8cev+vH5{chWb<PVv
zl3O`sM{!mKBQ@&MT8z{y=*$lx2@)KdSr@G5*c=u}n@F~sT7@S+xD8oL=|b^B!`ol)
zX}N8Q(H3=9xP&JnEfs2wCx6dmrwLWTlGWM<!W7Bpyjwrj=3i5jUs$i=6|i2jWxESX
zN2wIPG4^^e+*0$)_3v#!ISx8R(r~C~iIO=neT+(yf}8^szHED6eGxW@QlwKhGyTn;
z3JG1WnF(y{@HMrmr9Hy@-d=|se3#@EKO4sqv?qcani#T!DTkHy>m@Rr8I43ZHY)*4
zGG<6@(jGf`az1;Vn}p&u0E96$;nzA5?00>6gQlZiXQD$$qU=d3UVXX~Ksi=^OyNak
z?Lmw-wfhlwPyW-!=e%u|B~1kqSofIN*vWY5bsbxL6V)4MNEXc$n+|q7j0|`jn!g;o
z^A_VsLQH96JQWK&6L(+MRi;~zAaPA>0TMl)gpE5u?%85<^~4{CdeZL*Z#@UU)$nOq
ze)jxg*!l||cxUm;*QDq8b9-XD@4OC@{3Q+OqD^lh7Rei0^YBAYvQe&r85h3o(bt`?
zgYSIDg>9Kk5f<STNob!cr#)aZ7-)3NrMTDWHNjNJ|CtmAcTbRtu(Fvy6D<d<a}LcO
zzv$wuKY1t4Hg=9>Cd=Bme+>9*spH$QNS~<jB1L}x_|A<O&0}1$r^7KI^QS`LhaoPG
zB7RNQU20${;6wBFJlgDKE<_^E`c-)AzsFH~<ibO-V`SsQi+Tgi5>tRX4tza$65f8$
z1~HRW9RXS6B94Y-1jI_V`c^ncAJ$$u-dsMZfC!<ECKlU^a58<_YSwUwNzlQjNZOms
zBH3p|XI!y0um*7<dv}_KP$C8PT500X;baOhgXFk+>;qA$C1}LYh6t(HXC!OYNS-Mf
ziO_m)1U4IXIFds)J_oMoxNNFzeaM1pJjjX{N5tC6j^4n9Rb??D0S4{8d&UEaaPmai
z&zRw{+_Ld6;ED#5)-1?%MIwhj*U1kiP}9$Z5gF>Tsx)fN5PNTv(PDU%QL17k+)*<t
zWMLAMLFJIVRD^)x0&CB{p+z;9TEP~|kZgDtxGK)Gt%sd9nv*FYR%=)ow!R8-a$)XC
z?VD8O*}L1?l>2Do;N2hLZcS<G$apdam(x#OogU51cp)2E1GiN3U4fcWl&%3(yDlEX
zSS`c~MmSoxkcr7QjB(a;x-N9zo}LM%`5~NhxYH02G%8%S(9n^E@o7J<hNJ-~ORd@u
zj5u2`j}jWzvbfgS?Ti84Vs%<jT0yM-h%2f#(`^o&ttEo2Bdda;5<4v~Jmb9u0DTbZ
zFvp-2FF5kaLiLWpYq9d&vYZ6&O@eg;6!tL+rUY}mP9ZuvI7Zkt-Z4!s`wOjMfyD<!
z8m#d!9F7#O&XZ{XWWSGT#9VtxhJGJer;!RZ+9kA4-<|~np@fc()`xxP(7v5HQw`M{
zM&Ur&+W~5Xdn7^iu*J8}bZza2d%b{zEFQx-rKTc>XcmLmq*8?FFTFLperVpvt#kJ|
zsn*Pz>#(2k&pjmIHd*U*SCm)u-)|4GpGY@cHFT@Jd4Esy`|~EZqM$UzICw($=a8V~
zTXc}RChH)GTUm%WXaq!X)PHG-J;XReU{O#79IAhC|0F~+(EoRv%X|Y4Cr=Sl{uo`&
zc6_O=^MG!=#jrFMzWQO(K*73hQ;~gJa%{m^L6WM1$3^S;#+&h_Mk9KYCN~5u)QKRE
zYqU=}HH!J=MY-c8qPs}iMk+5-DqFNFmR@#WbH?UEakiHN*LWw=@4l7Z2Br;h?o=1s
zn<rtdPcwODH-ftOU-N8O6uiImCbN)u`<_zBnSJ-El3$^n!XF5z_EjetP#7c_TfgJW
zw$U7*&I`2uAU3bSwXVL|`^71Bt&(osZt~5D#%_@omec1(KVmWNQ6G{;dAqPhf^T2o
zD<(!a^)(9AFOKYIQ9J2!7Fjd-&MFe8pSRt)osJV$uRu^+;Y-03J`vmUfH6-o3pom}
zlgw^Fw_I~mM#}6{?=Oc26G@V^;ct)|-;b}_ctW#zW+B(*cuSi|w)aH8zre-4&|T(<
z%2)gQbvhIcxRV<lTdtF@o)>1=^_9Q>%&a-nTkP|S#IppXh=g0LyVeiq%}zg5&U&0!
zY`syqtZx`y4Vbib@5fNjX)g(>NSQ7_Wc&4oLaRo`PEI_g3(Z9E3g$n7%0<|fb1AYu
z%brfH)YBe<G1Xbm?T7-!(+_PB|7;5<ffWH7LbC3pf)~8=HP@woG|yTbC|bu9UV4ig
z+)R1>9r4@?ge1W%fajnNRi4wc$PXT=Mcf~;k3+i*A{S&+gu1dzzHO9%5?~~YW7x@6
zmGCIO-xp9Bdb>xL(X!y>IoW)<I4*}8?!PZBDfoAL4af<|ZV@3Lk|Z&(Mx6p@^DbrW
zyds`R4mdP$1<ca%9nsy+DUD2f^961D8A=C+A!mk0srm4SJf3B^Lf}KglAkhXy^-XK
z6r-+2OZ8DcQmtSHlX8lye=|U~s1!%n=r?bfzRpnity%3+o-k%wB65%lVL!he&q|h_
zDhyBN4MUe-OivhuLveqzpW0x2Ba;M!vz|)~5K^W-IXJdEokcMtXK;~>f?qU}FCz_2
zR1;HXzl3b#Wu3R(I_7$}ckfFQaxAZ0w7!c}Ekj>=vuND20R(|X7S;mXsur%$g_!*A
zw})&rGkPt+z>uJPd9MTD1w^&n&^&nH1c5jVl$h$s(o}In<ZBX09_x!Srun1#QZ;=N
zC!Hr<v%fmjT6fO(R#7x93$OQteW5)H!E!ofdur?0io!PWZLMJP!Z2vLgH?2Hy!tGS
zkzi0j5ZD`h{^_I=oO*kGm8Ax$5<FR*I@=)3szje%pQFcQRqlICaFAZeDpjdQlt4fC
zJiEc?qx=rJ98f@+>OA#v`bSYx2KL!O%f_jHP6OuBLn3zu8k-bg{@cm1og{eO4EAxG
z+_1`up0vaM)ZT|)rw*cigF9q1B1#*ou0=P50SjWkGT(98Qloz*e>+M-n9G3*XO!F;
zi3kl~P<v?0n%#h<7?v|)czeXCdak^LLRk9md2Pc##OBX>D9GJ0ul#f^@=)%n*hWMG
zM<X?-HWniVG4SjG+)&)O^QCg=wMpD#Hs;jYFPFn#4D}QD_|DOgvF<casWhhR>$iO#
zVVEbMr^ULDCt%}3r^f9lzgM(!Oq7Vro(|f|?<}$QNhEj|bC`_jkz<&D@GSLgq<>!O
zGsAscj3BG!Hk}9>-w@iUenw{dolQXs`M5L=-}Mr4-@2c}!}dc$Upbdb<hY<xCa4d0
zx_sZSJizxS2;ytQ6IyO0-q0ixXMI_xO1Eaxf1Xn+tXSC3EaSJCa|qgCBHJr$gj=?`
zjPk~jC&~d#T3k(2Bv%IcGHBN6=YSo9C$18e!7l^~@{~;J53AiX<7Bs~y|oH(pLL>3
zO6_gKE%?zf$NsT@VBOS+`+En`K2qWskbR#fV)Q%fDJTh<)LQd%TGg%wBG~t<N{ag{
z2y+xL%&Won0oUTCRH2vb$!f#*lcX)zT_%v1TqK(n{=X$$<}i4u9vzan1R}0B8cDIp
zhzGl6c&Yq~@Ktn}F!XmS0X*YS+}VO?eBWTYUz38&oR!m>eE~Mpnyuo%Ith1OC&OM~
zHHH@<Y05Nq;-C>O(-Xp9lO$ENA=-F-v^j{Du%CSjI8c=z&QazgbQn`<G^>#Ix_E1h
zn$6B5$s@LPEVu9C5Q=*Vk%|~a-!7LzSaTlC^Mp01h^iP8`qMt!L3(UUj*{M1i)9<g
z<x>&s!FEj1Q^+`Rt~TSz#=ZUt946-mZsW0;vC_l@75ay*56EKFVo@vg`aBfzI-SdU
z)An=DctTrx+NXUDw4Z)sxRZII$+1}TkCJ-fzX|o~JI?6WF0Fq5;{_eqP^(_PDH!?s
zx%8)zdT~&Bu=WJNFS+c9?omLCBO;rkPhUB9XkFmQCQ8FFD?=9K?L%sb(vjA!b8o<m
zLOiC-sO$5Fhk}($4<^dZ7Q-&zW8Ah$5+Y!*)#)hPd>!m3HR~H%H$-5xBkmRLo!u5^
z13&UMoPTnsyL@(Tp!iF3`N5qF)9N(l>v)W*?!0p2_179=@ovXf|CnqIi=sn6x8LB7
zq473E@p$jD#2%u1C~edm#!)r&4EQD<!d~L6XVsm)jfz84@y~Ucz=noeKdi&ZIU=7X
zeu+c*vSy0e2<wN<M(C}fja1Fak)O>xP!D3Ebb@-c98moJgnyswS=Bwd*O7si$Bb=-
zfVTvFvDN*1!9%N_Pw&@t<!pI6S`I%ATPNr%Z<_>9k(1$KeZ&!(63Vr@)9>h?I82~o
z;y&W!!FQC?gxg1aEHKLQvJ=2Xs2Oj;`eVjpY}~6VOA2$izJl?ww+AxsqCee~V~m9H
zz}XMLsCOpp!Bp||#<H7D`D3(5g^`Z5vwKX^=X316b<%qyJYXy|%qwUk;o0v2N8GIK
zuN~3BRhrDj#_Q~)UJlgjVx}pR-R8NZDWqO<h4GE%knI0V=(j8?p8!pQ$$Y!<cIEvP
zFzogTEO5{9ky}__sqv;-HGf+7VOw;l-?`Xr+k9y}VFB=IAhow<gg6OWPXSTS!kqtx
zZz{&_hJEX<Jt^MR*C=Xqjs5?qTA`GLFn7#`ClXKH>r*TGO8(N)_WY1t-g)zGXzl68
zzVjas8Y=!n+5!kn*2Szk+8K_EtnL#%rwUy0pae;L>%4%pnBghEeg@S_*X|%Gu??bX
zLBYr>dG3pT$DKxdo9}mwM)%sQ6*=0gu_n59>s+mu?we)|B3Khz2i3f*-apcg%<I}^
zurX=@Gc-Q==}AI!$eJABrP`3oJ9=&cd;mw+-99lMf{D@knnn?0zruvNEQ&^mK|uFT
zw2*`XG52d{JyXVa6u094==yb!C=x3nVl}ie8(5!L(2R}ArNa>r5g2*j^Rg0FHP@uY
z!iE?dBk`xbo~8C)Z|Uk1VTwC(8E&HnMhe#nN6Qvx8U?x)337hfSUr_Vd?;5R@59d-
zRt1?c5tUfiY^N=gjmGa^3Qw%Ey;;p%@krxiVEB@8Av%+Wkv)r>K<L$;6zfaW7*-Y2
zDUqk%+pW=3(k1@{f#0sprBI^Ms=cqYr<VpgAInO_Z{K#ZH-D92!f^Rm-okpnQ2XU;
z>|h=J69i=8R-r>cx_80vhnQVqU#4Ovu}9d0mMSXEN;5C?<czXTiN+X-(caM3I)VW~
zViv~(k6VembJt<q$*%Q|bj%EunS+eUB8cCPCx;DKdcWTM(V>JIWi|Lekk$%OeW}ax
zj|ko-knEoC&eHrs`(^)kp!0DwOY{21%802irHED)RP9@`?OXI`AKVVKHuzP(uMXX2
zc-*t&g!a6vuS<3*-u)HWllVJ-%2hX&-_P32_c1jF+<?_r6$OvyeI~u93V4M`&}&lT
zC1OnGoh;MdbiVFO_q*f4NQoQgu12dUQmyU_N<?SG-6D*{J);T#!t4H}g6xq?MxhOk
z@J_Z+po1!TkfGIcjY;<YouMVmSYZMu(*@Qyvu>a<6rY-e(qx%Za0p2OqhKXim^G7Q
znD%~nz^P~|)+<&uVQsF`)l36ZN*mtFgX<f+1NCK}m#=TS@5x8dDMl3UbJiQ{<k}&w
z+ZePNtdNg}Qmxe~+e&dM83l6hj`P{5Xct#LF`@T(QCj-r#-Dqy;tH5~eHWeZx%@G^
zdm*_g+AkFl@_TM(uCNmp$$<2>trdO&<zksG!+|>iaG6`O&#?;&KqzPCeZ)j3T2(zK
zP=+;tNsD!RSb^@x;7GI<U@Rs;Y>Xsbiip)`O8Nm5yi*);0>mGEhpNsLJnAVyo9;o#
zY}GxnSDZH#OboG313`;^n}R|7lEeI|?(mMBdncZ)+Yri_fHS%QsE6OIXVQhJsUH)O
zcaJ?7IZq~(JbB18!S<6N=S4!Fy?*}mded+&znv1qh03WxTfnEGE9XlR&!a><1lc;F
z5D_biW?at4DJA_mm@&l9PX@{FXqDtRA#toe&t&k5PC@l+?j3R#^cPfSW&?t=-=>T)
z#A`<u3dN=(6-|G>-&Iv5o6yST8yq7-a2=)LScl`w69qcpj62x`x4W4tR#(2{P)Z*H
z(?(Tg)y#?SA7?cyFPl^qdad2E)MQuN5!mKct4wm-E-;F8S!@3`0*tqJ4@hyC3V@oZ
z>060>D*=77%M6#ikZQsOlDgDnbfKy=n?$TPt4`?t+2H`G*2&StJT8g)yi3-I@iqK5
z0iz5v?`~!R4unRrPk7y?K)Qp_2<TQ2t-{W#*j~4QFYwx)bg9*xXgzF$$;TDZ!MY>?
zzc8NnoY5#TQad@eq>+R9Vq=&a%@+;}BmHQCDBLD}oWK;z6vuDjIrLGo#P@d}EjU!<
z8*{OG9Nkj#lVK)-De^JRzW0CEtu?P4(A#K90d?2YV~+Q`dm?PDH>97KKTh><e4cCw
zW9zAXGr5VU_Pb%EU#-hqD+!s-qN+a;sSs_^E6rhNrY&~2kMyKo4fA>zDPlvudh_De
zwNIt6GdE96X`D1@3}3|HABc}Lydv%I6~da6oC|?WEeJzv+<+q3qjh4}!aq3Gie^t}
z>1|0T1E<5a_Clxv-3O(2C@jk+^U1D_(~ThoA)XF_znv#>D%70HgdKuT566rQB_hI*
zGnZg6PiOXFu~dyQALa9@Z|Nrx_todAb#BjJ=sffte>u&8dSy-}WC6@J_L4>H*PQ}^
zg-3R9OBe%c7-s+WQ^bbJ83~4*4CJrBdoJGtvx_iFy{*&(wY|rY69q3kU17g19M*z6
zl$pWF6NN9lz>{;YyJk5YDik!J5{cgYtf_L-xjfOAWSl*{JB3QN-klV1Vm_HiQu!#q
zFcm@=^CX0>Waw(*D(QO&m!l?ymxSJENG%EI6jY4B-d>$sY^+Wr0}3{PK>2=-Ag>R-
z!y1zhx&vmcIn?Buqk7J^Fz9!W;DY?!Gy*HEeS`<9gwitEFnBhU_ZcFAo^7hceByG<
zRh{5O&r2c7R|V)gVYPR@vn~8S<f^{s<hJn5>4?Lx_Y=%nWJdTwB4<mhxJ*~C2tIZ1
z90aXE|C*9=u+ZW<KZcUuwq;&);a~$jyQT%f32n+DM^TN2?-)d)##k{9=T-~buS$(w
z@dE%vCaZJN^F319)O*>Y6x)aU^NCD0-3Znb|HOvneb-Xsvho80e}S>5y~w*svu<xy
zO?Y)m&G_#>+c*87lAUCClrFog93R0a4<LM=0)Z|dwZUsiQLuexA^yeEx2LlrE&H*n
z;>N2&rJAD)Bv*BJQr@apmVfen!yS@+^AsR5Gq-+oa#sEw?yh+LURA3Fn2Gfp+2yZq
z4|<7koh4H29DW~~P<a)0ZegAKms9(<FLw_c53j_IWM5h-P^HYoChw2lJsW!KMJx8{
z?FqFrR+<yWKE$n2jfV0ST6GJ6;5|ig@xs$eu4a&-#C0^vqj!BWZP2qnG*l?{5!ikj
zyK(Jz)W?r7(!8xi`}=YV2fmA(<iXgZ-&U4NjNPo(of4ilrF_M>JU59f*<Z~P$`lfd
z0tend@3Ya?(mN!{c{Z%ayuW_-#Gx!xVy~e&Z*#BONb0C5zv<|3nC?`FhC}DuhWM-Z
zLJ0|+P9dtpMat)R#xgIwK0RzC<X6+_t&)fW`n+jg0X4Tpu7}CCK{FTDb8!IM!-A68
z^{Tu{l1e-&bQ_}FSEA<FnysYslj(X@-ynMO$~@mHw}8f}^))*w8=YQ?(&OXcf;u2!
zvWZAD=<F={CiJ}gbi&*ovqy$5<&F8d5~EkZyuq+<4<va1d`?BHfW}yq%0h@Dw}UOf
z_5tTJ7pr%CKFeMM+Iq}8_RVij=)95+(K=PZIHulF?}b++yOYK@!@DEgUz8Z>viZ2I
zjQewjAa08|)3&^)MUn#dmwFSQ_)6<y#fZ)Cz23Xy*ZiJpF@*PqwXFghx`CT`VDGpk
zy}dbed2i)mV{*B#XPz-67EPpe^hxWTe?7R&L>R2u+EO_QI0lGfh`Opj*$`{&GrL2~
z(b_r&R1l!Jf#hIenkmRSs26=JzuK5s7(TN+{WDSK{?BM41Nj(*234?in;79qg-_=Z
z%3+kYzEUE4Vwdj~h}U~y8S0AuAXUUL(GpLNIu%M!DbvQI7TC*iB-rUc&O0oE&SN3~
z4d^2J;1V0Qz#<cf^2fSbSokT1n?(20rK^>+-tZv{&*!j7jI-n(@l;jAh3HUBx3ocB
z73R`htX^F$$xjEz65G+2*80;J(2W7*P-Bd3mwBgVd|Snl!iXPOZnyYg)|4CX*=Ra$
z$&ML+5=1^vWi)dA{*?@lFZ)>qy$@~TA$BJwB+fST)roff!lt2IRXE&Qzwqu^m~lGf
zS!zxj7Se(U0e6!d2wV8773$fiB!?6sdJEsl(L)9!C_dRyK6#m-!?Kil53>#r`i|pu
zRup(nY02GW6vFX-fej)yq>W=k$8MajB>NXS4)wBo{*KFG&<^XsoKd}PE1tZ*{APce
zzeZpNeu#}5(GuJpE9@DV2s;(aoE{K<l}_{p!l7BOqUEm>s-jLaD&OZm8Tgk0y_{XM
zOpx8;M=toxvLe-k3v+$^e8b88GYb>pH?J2>W#WbIhbB$r1)pyNBGzu@qL837nI=3&
zwwgELIXZ%HU9%nM0}&gNYLBgSgW<59=u?lhLpr^oyeYqMv}RU$e$h9E94sA{hZ==a
z-{y-h<vXCgpBJ9D8<!L+G*!}-l84`2!<lm~xVHw(qI@;?*QFGEJPzN#!Z{c0N|Rgs
zUEvr(JJqP;qPwJBM#xY8ZIM7`p`@EW|HP`{3Z)dZ>QRzCl4sqB=pyC2-D|8QM)@t|
z@^op0rSOOK_2uJao8TaU_1CY}*Ul>aHS-a{vI0@#@y{6MQ-7AvwZ!w@b{YXs{6V3d
zGkwm??@zq6aA%c|gBn<afoWXRGOgtDG;Tz8cXBJ=e4}l+sdP-lA<&x>Uu401kwU&$
zeci44M3*IN4bz<;Q$%Q|W-9tn>S%leV@PSEX<?VLR8p?F>l@w9(Yp`c?N0-+isv~J
zE^?Bsbq%JeW0z7_zN%!N>sH^H53VMD{c(!ia`I**8o1S_za7Xl`}aaCM;{uV`#^@b
zX0o^166+s7r+5M>T974<wydE5x@<o)B%A51%fJ4M420)2f(OR}=ZF@Qh~uNVESJ&F
zPUJKgmfuKlCCV`T6Ky^`#OonGYiW_-Tf>f&v%F1ohlHNB+m6dXlB>Xq_r(cu4Z{$E
z7P^j!B;_!#M^ouRU_qBSur2X<uv9?14qhj_I&zCi2C1@)<`;fa|B!I%wmV*eQ)&kM
zMd_eB>}L&omJ0R2urR#@GYS1{&pj5jIwYYiAK_Y(#BY+C&$>!T^@H#g6f%aD+_gx8
zRoMf&gV}@lRDWW62$99cAY#fvtQW1;?*%V`ZML&^d|g<Jp{+yZxAu~t{G)Q`wY3Hb
zjb%u0`ID{Y;D&@0Pq;zrGm9(2RS0zX?3X2Vs#==f4!(Qz<-=~cn)Yl!T3qmY{K?V#
z2GKYHS<eyNUK@(^CU18>d#M_Uuh3qYM9t7Au0R0B@N$$}SOW69Jl{{ar{0H;It%7n
ztoH;4mBr-)(wp<Q!QuUGaFj?fP9;2=Xm1$KcV!*YTNid$6qluf7~M9C422lwHZ`R`
z(~%yz#fMKX+&F(i+%U25T?ILBQpF1Um9IE%KTDron%aD|{Wz1+tE5^g(VO;2f?^u>
zFgb%hNJPwd4t@<83}nh(F*RCJSwKFab>}$V%}Z%A#%CuT419Ezzecz!_$$GUfbTt>
z0yPK1Ev1JdqAhRkx!BOHf%n2<SY65i<1J+zABk}e2|!_?ah}cP>NlNN@}bV%*9|^w
z$-&2Ii*LS**gxUN<8T;u_@1kk$~6~sj(3i}Ll)Vo03=}ZV}a#VN?uL#fp1g;6mK#%
z=+^R^qzIplhm#e^)$VD{iV{3lJ{Iwm@*^TbcRk8zbZ19JdXO0Yoz4B%Z`9E+4Fv}E
z_!44J#>MG!EHX{Zk6<-&F=INgULys43MmZ3Pkr~eN*{3CXKw!9e^b}>?-k;&BvFYR
zofu7-JS52EA4%hXeVc!$sX_Fij=EJhKse~IFIN99``6Eh#67Ff&ttA3{#(T{(eebM
zOw@rtd*!dn8A8y}t6qT3Q$9*&Pk=+?nQBh_KTjte8@xG-hW7oE|La-)<L!HV<<Xjv
z;7)34PgZy61^|wJ4dHKO1mZt<e1_XVlHC%3^fVhzlV1k4YF7jA?F2(RP*N>_QuGAK
zGW&5^@|O4WXM$*AVtT+l`kwPsyYU?Q@fe(XOW1!N6-5MULB@tEKKOSg{eSsT&>#}0
z3mz6=svGZgDJ0}p>Skex)Br{QkNOE()2khW82$+f{&65xPP!1I6vc%9kwvOoG<~&&
zLOE(=4E?vu`ZdM(*7ogZBxzz2dH_1;ZUNvv^Vp1LtOTpnLec*t20_AjQPIq*N+yH)
zKY#o`Uh|*dtU*vLkQKmd{=hBx0f3hGYc^guV#nIuW=!c}Opzej2gRPDn9Urk<eEu2
zXw)m+if919eB)-q@_$^9FbjI%H(Qgc*YSV+>ObG|-*?J?eOYuYI01P~1MUBLV*fw>
z0=N_uFboDX0;tj8ea!!QVbmXEqS!?-6$Sq69Y6zzcMQT}>7XnnImUlIsDJ&DuoVVy
zj;@D#KmPSP8FEbLj>Dm!V5=)2**HL=7U?ikE@(ovHVo|7`%xyNzFy}i+ZF&E`t-8e
z-5t#6$AG-Jgm^HT%9&yVu)>C1@)I>MquEHU_=i<9!2_LsC+r3s-U_%&A*e5C!iBpI
zlK_!XKc~tyQMC(jdXN5M41v2_0A(D0H&bBie|<oB4v9hBSk10=XZ`zAF;{1F1-gwW
zl;|aG2#3}e{sWX#S9)#{J~O5|iHtfW)so=7+;a(mYDt}0qu^%!7*Gx#dryr6Sz`%+
zX6*+u5*k#w$8|l^02_JdwiQ*F#i#=BCUOA<c?I0;)>Y1{f>MYIGqXc)icS;Q+kLG8
zjn`-}YJt1~td;~<FbtBc^OSkhPWm`^A`-9%{({-A01x-vJUHv@zdq;xk4VSkiK&1_
zEjDoe@83`2qQ`oIG|cRK8kpnZMq#~Iz<D`>bHcnIWMCH5NQyTBM$~_Tpc6gYu^||t
zG<&j$qJyC20tM}b(Vyw;E;1B8ss<2v1SL!8tO)-)t!Csx3ZN>x9*wY&yKJ8_Da!Ei
zK7a;i)Fol{90VN%tbpQ(2gzMbnMP&CClmn6v`KS;Vnt8``Zwc+Tad>87C@Qt!F{Xw
z0->)I_4nx{5@MSEdTkIz<Nc)=hG9N>2#DLH;Dp*A<A$>B9|WJauYhiCGb9u@)y<u{
zT%8Hh^v(MTOYDYoC5RU+EEvz@MvV5Y=w}TpgNDTy=`ofe<U@PSOZ!aYzh=#(QB?uP
zORpf!|MkeAs3XgIt}pekBO^t9@GMw@AJRohFngSuzA|5edzBCv(0f?C;yrsFcenS^
zqaY9y%*|Z4E&lg#^!+vn^}~8@Y_<Qq<}ZkSfu_CBCFEirWt1KRsLf0$f+QizT(|G?
zc!F;6D69F;0Sc~$BE32X^IrUFU%KJ~PW0UnEy-Vx;9USB!QBBo=m}=ZHI7^Al&?-E
zLaseXiyh9^vk3t++U0vKSTKal51EI)v@`*>j^Ky(d^|a{%VFv3m8{Pwlvd1N#Ic*}
zgPT4e2qY<2&m}RvIiHQOc;4v_4F#fG-#@?Jh0xG~<t;Q}8ASKRS|(0_3^Dcv5TB2{
z7v?_TX2dx{|Ikr*gz<qJK=Z`Aq_;Fwo-QjDT|RK61+tW-mc%kg-3m=F<C%li!=;<+
z=1r^5RYnDb!HkS8X2H8pmUfPQZwPmbj)O|;paE5PnnSEd$mMsseiT9A=kx3iroZAC
z3d0nN<ePt1FId1SPRW^z+#Lo&OtyE0U{|MaL?X$JT2!_>FD<*h1VVf!j{!1}Wd+2b
z!&wy0G8$!hYxAOE^7G}kR>%=wSr_n;t#dMQyxBFoxsBq!-Kti>&6o$J=zXuS0HnBk
z-&r`6i$dSX<_@=GNXyTZ6BG}OD*-tGs9l!t%-#ij+wFU~cv23_UG_VYRoP>3n)#k0
zO^ciX+?pTT<sY6CSmRTcb*`Ry7cU)AM5{k~OSgCQdDrXw&S;V+N@cn@I}Rpalc|D+
z5eZy#j*l`6OC%ilA3!$&{jB0Ntl<WUWuy55V7amdgA&|J-)e?!o-x9|H-JMxfo(!U
zfMvZ}N)z=(W?abiaiJ;qQSfu2wBKD6;Lf5wD4hBAh;HV<6EotY?Q}MG!=&6I;WT$z
zj5B7B*qqb!G&jueYWYfbn9Qq5u+-{d>#E5d=^RpG?I_c1V&YgRL5h?E7}-e;pG+y!
z{Zih>NxO5tq8WDB4`RpeL3$RLkL<Sz|Ly*&EX)0CMHS@aT_Pp8+kR_m?-cwiaeGLL
zp(Sys(6U|r@`Wd}LMQlFe?KMH#{^Kj!Va0kc$b00)AA<iI-S+l{dXPHoA5Q<AgG5p
z7JVvF%i$2!tzR63TLL-c?2xc2B{Uj|;|{V(5AW&uicQZdoG?qe#AFi5guXutXeP@*
zJvFByQ9$lL)hKK_8AoIvAtRM#o^Rgy>$Br=DHk-XHU^bGyi?C$_~-K~NgpbA3eY^Z
zE$2IL@;k8Tqp@67C^**M0ZEePa@eEaQQU#W_cY1+QR<cZG3il!kf*6E@Xu91s<=M~
z>(i%&E28joVZQW6LEr+oFeFL>V*r%o(4CKddqo@=pF7bi4j%{Oh7ys}8-iM*p5|v6
zO1wDB0>2G6MV|dgBTTv`=rXF7)a?PDT5+RWI=B3Av>=&40%+t5^o;A6?R;o(s=Scv
zv*bH|40M;;%&EdTJ<j0LTa4Vx<_PGVZqx#<K$_mg67)CuzN>K>hwpQE8I6H{CU|eV
zqL4s*uR-?@KzM#RQ9nly3kx&`)gjGjg{;&EM{-iBV<6sg=DGIaYKE#EtSeie2P|C_
zq&LG8!x)tz_^Iz*ty;l#Xue03T&|wl3^KI2;|qX*8^=vuYvl<1(@*z1Fyq_zxgxIy
zlfX_>oK3I@4v!>cuvFg%WUOyLQPlqNJkgn5Nq4j`(lcXl-8<hJ0*KBN72m(qm!O*4
z*;yo6<b+6;xf>zzatC?}uY3kTgBGCI<DEM)1J#Z?xB-XnD#v;XfAf<dz73EoB)Q0m
z40Ve*8R$}<I2gc8eJ$RZEvvSn+6jO02caPQc}-ejKa;9!9RQeZ$89fHS4HgoKC+RP
zmOQ<JC$opu-GKFw3SK_t97eewUj-NE=yC%03Kzd#*UVUZ!(PKpY6gAJag1>)0mPNn
zj;rgB<-7F3t{-dZS_B}7onG;GG=p)Q)(bzFRX5-?`d-K{BFQ7~p9?~a{3sqVKYWNX
zDfs0t6L?~P%Kp=>j{k_ps3hWriTgbPHT^zqWtvYX6u(wch^dI5@qzMWjHc%V##-sW
zy}3en?VE815!JkkCQkT0`f2wUfe&{-l})e9T7sDD(pR=MC)j|-TqOwd+^NEIENd@<
z&z)QXo_qoHpL`w@7%-0jsm=VbFHv3^&JaWD?XU2@b`JkM_fa;@M8&ibky_Hj)*s?+
z%>T+E@uJbJlCDmdmVOBj^laSSsM)fCZy4YvQhI;$GGMu=|B6M^2PmXQ_eOb7xvUsA
zavb=7U{gIK0vx)Dq^DxJOMFAz&gDN?w4dc!m0(->Mm>cT)pr1P<=r%%4~uTYH2q;X
zv|n1v(`!@E+Ng!R__(>D&*FC)wBP;6XP@1v+zbo>)B>*{ip}xEZa_qAT})cK-LDM)
z)jK}(wIgKCp(4s_Zp$gA@Wt{=`A@&QLr>sW%P44s>(5@C_Qk089aRsCDhhsB5OJ`G
z+YVk<)MB333wNyGSDSH3l`P4bSD~_7Lr%&vwZHQL863=<(mK|}=jk-xe&}1x@?+j7
zBOywGW117kUZ9r6mzMH=7`^E0<fDpY<#^V$Ji|q>UBT4fUuY8QL|7!ve{BQW_2a+l
z)hH_N7yMg?ELhAF#xX=k{dCup2Ubayvy+5?bB2kj;2KJn=Dm05G>xkr{ig;aRt^g)
zX87kdKNB>R)wPFg%z%gsUP#1xt43i3<p&ZCg6#Yq3BS8=s9f8{s6p52nIQ-s`Bei5
zN$8DXO&j2`<&V3AJ?H1@WihY?%a@5zw<xPr`Em(hZu^mtX*^1-+)x2S7~E4<$l{d!
zOokrkRp=m&HFhFx;7?hW;7)MMK8srgc{&7#+BK9^MZ1THPyq&0grC7(G8pR;Fvt^+
zc;@(e5%`geuzMniKyx<)hq(iRyNOYyj`G;*?&e(w>F<gB9peupWoxWXnV@%p47fP`
zA^lfNz4#`qU1;hqiog#WgElCqs@vasC<UiW&E;xY#$c&HSu1cIvr5t*P6x{fy?-08
zv#3n%7Wo~w7|KBYE5~2+3gP^dcI_@CBozbqBh4Jc93VID7+inAk@|^l_6Tg#9P=&{
z2W62_sn@Zj(|ETxP-WS;qQ_+tJI`^C9yJCNDSwg)P^76yv_Gi)e!_?pOK{C#WGaDj
zcTCPNy<T!Wl^Uo8tF@F&S2>})g({_7as^QQq8?elLWti9$au|k9}z8Ulaas$<l+(U
z<xJ@Zh*=VFQ%&h_qXfr`(G@DMk(j?$0i@Q4fv(gU@_wd5i(60Uxi>Nk1(sa!hnD{C
z(4bE_k^VT08aH_BWx1EyMV*j~<oEgxHl4VrwEEV-uw~_(vq3NFS*LNrZ+Vn?e^q8h
z{bf4FQR6UeKqOEsaG9`6ZaBdbOd8&N<hM@qSDH^85B~MW%u9zoNa)InmNfwsE(HRA
zSkWDyr!Gb*Xo-(}A?g#n;+vpxN78E&SO+qIX|U1{|IBXj2+@a}PWn$qFgN(eQznm#
zW?qqef%Mpl4c+00l}F>Cy3souvhN=AjlqvebovrFOhKDJXdhiaAvM2A8)2E$6@mZY
z8IwhQjGQ3mB-J|X^Oxn}!0LJ5jyZ#0-rpcR6>_K9bE<QiLPGa;ugAS6(UTrkKAw2M
z!sH}zr%0T>C><d=V?4e>krwibP)RwdI!!7a|DuHW<L?{7U$V6_<O6efdszP4t74yS
zx_JCPU=O+<!9PCHoG}dlCFc3G%8?Zk<`xmEkwf<Ht*cvfD4TEz@%Spp=}3}(%s%(8
z$EHYXA$3|2xyy!F?;>VKq@5#(KOr=pBhoxx6x5FB3Rs#5tZE7LsW(Jzm0w))9Pd2q
zO%$SVZq4S(;S+j+)pAl_Q9_)A;50rxsUl`Y5NjaP=n=$eFBW+ReFSEOr&RQ(R6+~%
z8A${sb%zNV8~he**STIq722nVmD@Y@O5H7Hdr?+vpM_RJ?E7N>xE~o;pv35=Q2%+x
zeF^C?g`}H;otJP<E9(X~O6U$_)f3!FGXKQO&S;>>h~5-56w5rN;x`4+PwUw1<seto
zt!`1zSTVQe%6?Gabcefb3t`Wccj=X_%g8d++<mn-1$oD5c>L2aH7>)RF;kIos=P3W
zX1B5=D5&Or{|0P&eO^U%N`-R@R`#Nx)T!t#FG5v-L{~uZ@yyjFj=Sbdi|V)F6u*O=
z!J++H`w*j2#&KI{Ua5>^Q`Doe*J=M`0o>0HdV&{MUZmE18JjqT{k1Qv8SCSjCC{pX
zfZE-uSgt9gMew|e$rNNF?RWGEi&0tib0rU|MKTSOb!|oFok}{5+}A(*Rdl8xA)LIE
zLX=@2cscGTWRYWcm{<Fiq<uvY6P2VDGX$B^vT9A+mx^gYgKR5g=q#X4Q;_3D{h8|=
z%SE$d8oA#IK_)bO7BIP0z%jipFQx3ge>aZr`7l{xAeBSFHqMc`^^U#5j>?q)!f||;
zVB#ynUcfPIYN1Ps@vBljc6#@s$h&E0(g!q#A4SV4O5f7cd5zt1BRQv>{VB0Wxku#m
znpmRsBc#EiA+9{oQ`BMM&jGFzmAj38UMF2S{`*@Zy4+h8XG({`*H!;|t{`D)vdds^
zfH)l+Yz6*9^UM=X-@9MCrDUGwzJm?Vl%zF*ouut@tMs-YU+P$>6XI^tn${ne-{Q1|
zKSLAK@%RXvG(w2K@r8_=P2!X<H5oZ8c<3}1q#K@?I&j&V3L^Y|Q%*^2S>4F@81uHP
zQ=<P2Ay@vR?sRj5_AXwLTdeFR$jP!$Z_wScWF};P^8aD%t;3>xyLa!Qy95W21}Pc3
zyF)reL277_?nXL?k_PDz1!?Kdk(7`QNeSta^u6aB&-2@RANzRspB~2y&fIfd>sr^j
z*7;cz1-vf_kvFiAMwr?Dt4wWRB7uL8BBv4m9{FUrm?9e#^xLeG+aQ<wRm22@QN9j!
z%9YayszufVJE(*hSsYrL@v9=!$}!Z9kvKG3uo5wD^;UpHd@$uFxPYvdeHiY?O7-Il
zP(=651ZF$YWlEVzM_V59of)Y8Ow^P(FCdVwYum;yD$VA<FAD`l*%yq2NTDRvlWG2t
zo(r4S!?p4l+xndSXuP6MxS$8($uH0S(t@7x6ZU+0yhzrh))qVjd&{<Uc|k$ez})I2
z2Ri*x?RhaJT$vE&r~f152ddUMWb6x}%o%DylYv<@qcT=SedxA&UHE&<d7R8~tS|Ld
z^^dplc@rdNbg6YM3g6t@3~9zmcEMT{eduwle8#vDJYd6(S6M|e&+(ClDmn#LS?+k)
z>L(yw3U2!bO;dNo0GP}X^T>+7>e?#X!~-7IjFbQw$>LY}gRM9TP2(<VK6mgZ59f2m
zlIACXfMs!iy^v(30n7VzQXy3<x2pN>brp_r_tQGyU6&bDMkRkc3Sk(Mx2Vm5$owic
z5*Y60^ryYmSbPKK_D7NPY~Cf&rn8k{k^|X?C`kSr;$YYa)erwxj}G)Q^JkLZndc94
zULp`idbm1b{Tz{`AI4E5kC=?aY?LwPzY%9%<Fy(&Zc#U$8+o<#TkK6{b{Gvc4xrxL
z{~>Psq<3DZbF(_jbg8ckVHOa%|EZ4J^}szvQIrs52wGH9k^Uz|>hhN9@;qt6QD3SC
zoY*?mUlUrBxLx%(xEX9v>+m-YAH_j`FjY_p;xISGaciq_y_1P8$6aDue}m>W*NpQ+
zEC!LFbw^1bKJa#~>@I=ZQ8@5ho41u#2m51^o-8Ol4|Iu*%GNfFUZ<}eO9bFsk4UeY
za*z)dT{<Ii43}OKa%pvr@C-MJE&`296;u0%3z<BF2PcXWr%CbS_^!n1ZB1|3TR>wQ
z^vfx>qNH8GptINyao)}aL3VGJLf6MaYWo?#9Jt~&C9<Q}<?mSCy3A)SHG%V<cg8tC
za!!gec)c|EXdp(lm~Y__>ct$5GZ!dh<WZmJ)Zhp9YW@19Tz|suDr1)2@Ws0NyUto@
z7?)RQwBH&6va021IbLd~0LPl9FwESpgacvJ28_0xDzH1c{DJKH>$2?HJ28SLqk@B;
zEjw%haks`v1(1D{-{(3pz}z+Y>Puj;*o;fx1yAMn$G1(W0;eWY63(w;RJ12WDm-J(
zH!4XR+e)Y3xRM%;P*jE_3Qp?GY#09=6)C?eZ+cl0w+9CDoDS<gN3TETFS(bOx$b!#
zVTe%yt#(3_A^TniEXN%TUe|BZEJKOLQ5;lVLtmtM8oth!7dm1aUV5i%+qmCD|7Fz!
z#)@uh*Y(-4S=n-8zlljaq4mvxua!Qx(HoIT^$F4q`pP-djcuQ_=B|N@x+K)>mEpIv
zqE;omz$m^Aw`)<6zeVNX3HdH1WC<p#JIk4bQ2hl0nCZx5RV#5Hzb^NWg<N4`Q$VY7
z6PJ3xHD_H{2|j6a$W@Z0P#k#C)Lytsd;IzPE0g70!oy|YGPv@`*)e-%sl@a{vn<0p
zwK^H;iNj4N$%vW%?%W}u;bS4BFJDN8BFd~aAIII51tCO-(yCi*&Z^;@mn&i;p-T*P
zWb{*Y>nXq8akl`wK(2$ASvINvVRuFg4QUtbD>=rg`1a~HW6#gkhU>2zriOMDxe?pQ
zWIiCA#f-f(X~VJk2rwr_9U)DIiszG}hTfvQegDwAb!0NZAXtGjo_+Q^!|jg@DWODN
zA_}nv>YHS$nX6m_mJdM^p+02}%62|^KPL@oj0jgH09^Cy+zn)ryjs3viIYStLs0g`
zA=DW1dR^*6$Wr?Ev%59v)1(!e2>TCBtyHX4z)+{glR1QX<ioE)V_a2cfI9FUF0A)d
zTyvuYpZPw1a6iRiacAFE-e6dG!~4u){QzAg8t6k>wyZZYT5EqXZhLEMkV#ChGkxO7
zGl7)m;uYIg$R<$5jZzXo2-z?D>0>6i<lGCT_}(5+6E}shaE&`%xvK|ok0Y&Y(-H6J
zO$4(8uB<@EY>>)rM=rh7**Cdgv9=~<q0KnZ8R>6RKf47SZuHnXUtY8wHsqaui=g;8
zF(?hgBD*S);F?WJE6I^a5Fne~ZRno$IIaHz;~xDHe09;Mk&nl78lXZUX7fjf6x^&H
z4RDbd1K2j8?x`@R(HH}B@lsrjO1vzB6{^t_p4BQGm38_T<eZ&^vVZ<0N#7_V-)Ie3
z!Kn`<H^gV?Z8yG|^H`Zfs(;lEx2UFFM8ec2k|rK_)h#L<D7xB5E&Od1gj<*%QBk>U
zas&x8Q?Y()<YjVb%+-~zoD}>-z28;%=E?a398uDNn||iRLi$?}Dnmnp>l5B^02K(3
zY58i93}uNJ_K<N36I(@fh~!&fC<TAOvpl^xbfjZI7<5Ksja|a67DZEX+ZJ&tL+t{F
zAwIRQ{mPW{Zf&y|5a-94sUQl3n!l@|<D2yq!bFTulbeD2P%jo8m83w|D1h?vznoOz
z^tiU5ALxt0A|s|{l4W@{1E?mT>wF<@bF)GE7nwb+8B{6zs)i$s6>Kz3lEBee;-Og!
z@XPp0iiVO}<b}1yThPFK0>JD_CK;m#b^0x_`|~MrPSgp7Ob_0?;>K*b4^FgR+r~l}
z9_3K*^yHRiDoD@T2lpAVMJaF=g0xRidB~p{rLaT9>c|Gn&<==|750R~_DvswcHMr1
z?=`8AN?3|txN%PFxO;q+qYtWs+KMV8>WPH-LO(!GC|WypQKRduIE^e_O=%c)rMoc!
zQ9C)v2VB@D@CZv(sTHdF<w~4bkwKEsdVZ*8qahJ<!7ADy&(=B)_x+LNMFk9&u$~-8
z0b>_I8zA~2*M1;J@{1($@b6!x3r^S7ldk{#YSaR9NvwBl=0DOlS@Ce|9o)YnCPfmN
zIDq`v{COR9_Ua`bx|Qi%6}Q4F&>tD1FZ9MYf(8<m{n*Qd#L|WM9qA1wc#<sAc`PHy
zuh<q)uyKK^WF2Sq7rB|AXy9KG0J#39(fcv|v@{0QCTgGAiya{{okY@OeL$*tubC<-
z$dbm2KcH<dN5?e~Qs0lRP$xu%7|-#$4Xc(uwshoVzG=pRVG1A&2!Lw~ct@vp#_7<4
z#{pLt)ksLoTdxe%uJX#3XNLYK{T<Qa;cBON&ryBM?|LJoHuG81v4%zBqd`$)k=*=Y
z_(OSUZ-E|XV{q)e{nwti`q1|Qz<`Nc1*Odl|6`0&LZ0sA8WHm&F<QIri=Mo28<3Vw
zj>pVb8r%1EA3XuQuj8FMGa{7iOC^@<S9PvLLoJ^gy35;Z@5>QV@zJod_a+{Ycr`M(
zjhM#^5B&6oC+hd(+W!b;k<T8avFsts@WIP2Db>hZ(Tz;&SUmD><Cu@oL}4WnC$$`*
zaUiy!fb%2p(zA4%WvW%rALy91#wUikJb%J(VYeSw;g3<PzY@Z4^HIB`YbuIzKhy?S
z;IY1M2vLi%$CRVK!P-In7BFe7`KpX~$1)MmB19l0FzDumbw*Tx3pr9E_XIEpVC|ch
zK;kmjvo5l-F-_}5jrlsJZgj~RT+DNm=K^!M%Gvre1ILgfpmwr)DKrPoS_$XF(D(6M
z48qw>0iOJWP3UfdtF~$C!v_w?JN2)ST@kvx3*G?+m>SkJ&cHArx0)}W*rxg)$4yXj
zp53=pKdt<<VYyEBG~)y5EUS(uVe!Xg+ajE>7Xp8ZRDi04Tf2s(f7ff`8}6u&K2WjW
z^W%Xt9_1<Q8<^W%u!&v8d3z~?5fR-Fn$r_RgTNuR8>YWflq1!lBpTqy{UWv83^+i5
zJs{oj*-HTf(Q$y`6`=cXj@5+;gk-$R7h%xuvF<>WWC8RdmJHe_?l`!)KGY>T)RTRm
z@a@EX+S;FOYw{KKSq`d+$~(rShZ9r~;ZGsc?gsl$vvLv50ouF&Ps(C^<QVxBPd#Ba
z^%o2M@_~!2X^ZVs!xt?;w83$uR#s;chTw8z<qEr5Nd;=D%ZN!4<zawrGWxhP_xY>I
z!n8xV5HTj;KQV>~ji^<UHC-K546Iij-(_YvL9cuBKwC_iLdqilu@$o7jtyy&?qP4h
z(!~n$NE^)Ok?ts;Nl>4=@Fmc8s@02Jx9@-a>ihv|pNvJS>`D(LT_6D5Cx*zu<aBg1
zL^q6}8Af6j`WmucdSPW&`1rKo-VmTG{!j!v>7$0Rj9rLTXTXqwaic%So#P8J`q{T8
z<3Vt@(E3FWQ>~vFW{C|EceCuT`kwE27@qY^(UGLa?)nB6ud*?KO7X}O2}RHr<RKM>
zzZAQxX#W;F%}J34x*-a@v%!#_-;_Lr=9~jovlLh$9OlHU47+V}Pw-$ya*o%EkHj|1
zYu`w&)Cx!G+}>fSdA;0gN}NvnE@%bbT0#fKBjngI`wr8JAv57KgUQ|ys@OqYlvN2J
z{N}(hkZ@{o`e+6U`a#d%;gQ%sm@_c+wfRqUV5jXGYzjIILGK27gw9?cn@T5fQ=$@V
zcMzg^0sVmymM0{NCv%)giFg{o%#5l83OXjK2Wzcd7xJ_zIvhFzLMbv?&nJH#Pfd;=
zO{t5HV&&_Mx<2($cl~N59`zWGOva5)^)g?AE3I3Gcq>F5W!NSWKZy#ndS{H*95E~+
zkW&LkjxwsoUPLy6RmGjZFB%_Z-+r2VkNRcvS8HrS_Of&cy<4BM5r@;Ra6Ig67!n{|
zNSi}QK-VrkN&C4!2p?@(QIyR^_cW=4A9r5e#Y=P&pIe~+6!2xA;+V!><IgC}_o;Se
zql+kQ#<pOb)RCSVbZ(r8SWwKHhmyZ+w<WDEl&1==4`-xX(A?wM(pc=Lq$SuH`!+v$
zSZUs;UC}Cis^%u(oUbMdlv3K`)grj>fl%^HZ$@ier@0jG9*%S`D=PESx5HZwBjhmS
zF8}R->*E{LC>|IFEj=-30MDuJKe2?6p7Mq~Dy2d=h|?p%&Lg5_ArMn-N5&R|mKLCT
z)6{CEb0f`6C5l3FxChuAm3{;Wgw|d^Y2FODao>AlE-XNi-x%_1(9PRlniT1p6|llw
zN_v<kxhtnon48=f4#e5NjO@Zzhb1A*T&ww-knvB~v=Vz?ZQ|0e1C8`%xQ&^_Ls2(j
zE)?dk_UOz(11<G$VabV3FzNhTGyDFz*wN{ork&O^smwwhO9>gjXnP$E@5q{G8!y~T
z5*5gq>v80%s6<{4*o}DT4V(XZ-aZFR#d@-lO7e=zHFf=yYXwg1Guv;i46q>nW9)y+
z%3)nXGF?p236lK@!%*CmHwM4r1cKufQld+h_Q)ce-J%7Q;su`D`;0B&Y=kmrO@7=m
zZt3!CAm+aRwQECMXG1(*IsSx>bs>x|tF~dIPCz=!ihE+}4RHqIzBR>^NZqJ^x&OO`
z{hM4im)xl}cIQBy09Ks9^C*S)Zwx6u6En~&=-4;ji`AOrC{KVRmGihubof+~CQ?vJ
zuGF;19f+PSY(85EYuh4sv-@?G*eczlmjuwo>T-$HzP<~*^sgw7zr5?lCs>f1*{^0l
z_3(bCV32a=3vE4Ll(5Dzcz~&;EK2CKxd6Q-zT$%Y-X|}IkASL$<bsR_ACK9T_+Abx
zZdT<<FF2x5-^`}PF;*~FK_`o^a#o)@bO<&2<7>$$<e!{f_db5`{evRmBbugYi<1{$
zKm<A){*t>W^M-FrwcaTIOP`7%T%)Q$J+m6^mrbjB5xb{6L$d;dlxRbPpJY}K&jNkJ
zED}Rc!_#rkPYOCvc?P_OrD4vD$(dmbpzJludF%7tvy}s~N2!B5IRjEyc1`U%^e{}$
z6~y74h<ezy-mNh*8TtK{<YWTerLwwFNc6~`5>6I>s`GcpTX0P*=%5}T?I%eTHU}m&
z{P>=tH&8h;xGS1ro_Ud8db_LG=tDjJW#2%XayR2fpMl2vZO>ePMOg9T1pHBwHJSM|
z`ACRqeSdo?%kGi=Nu{kS{4T4F46Ax~zNLDE!4mg*o`a)Mk$zKxRj^>;TUstAfpS->
z3r1EBgKy_GS^5|%=?m0xpvMjc+zW-CeC12*LPeZ1uPn0V{FjUc9_%Q5@&sXqgmH$N
zBAik$<CY%1(zi^NFh4kQ*)HE1wC8<oQ5iBuM^HSWq$dGrta=bLScP`HLK)2Jn&s{H
z@d-6rm1$`PvHI^GFEW3g&@r$z;m*;!t32^-)_f%~*|_)afW)L~u-K$XHBOa$G{<td
zTtOy&B=no^lat!5!1?gN!3|x@Y@bg_PDMoc^k+hKn$sze1xE%<h3yd7*SPaJ_f@#z
z1VlrDA=I~3#_t`5d$YnG(AZT=z?4iEl&TN<#a8ZNrvD|^1Smxsm!9v#r8>I>3=|lA
z6&P~)F8o-jZ%i1<o2luH%5Q~|Zp;=)ErtYMh^Q*i#wvhMAEw}qNe}LG1#cFeP&aIf
zC7d`%=-q}z5{M@0TL5nlRh-yDA*Z~cRAyFcSCl{a0%gb6=0jqu=S^tKP2MNHr#X)t
z%;tqsZeej{l9mOP{>4qv1S1N*Z6jCZP+{1KYftfPQIZ>#=)jz<c#r@wl|;sZRld@q
zI&FJ}B{!H8=lqMpZ@cFHh14W|@ea?k59ivIMY4b1Z1&k?JGBusj1p5Vp0#W}E3pL}
zIeNIKM(w;Jn9x)0e`Ot@iA)ibt6Cn*LUf_%;N@<HD|#<WHZ6FQj_A2ak}s{-+~FS^
zs|s`SkAt3S*R=DV+nZW~V_K-$!|A|I(W>Bm{w3#+3(&(6cLdh3zbaxw&!ojo1~maq
ziLIxtzwgf%J>wJ?)qW0zxyLaOlYuR~CDq=X-d+oj*EYQaKR)&Ic+ZQaP&tHL%%1X2
z^=6R>#Ok*4bB=?yv*B!>G&W|OYmG<6Og#efY&&+maE|ktIYOhU&+5lw{#m?wPYKA{
z_x@GOb&xVmFWPSt_)-?5C<kE&BW{cgPfOQBR37ZL1J7kad@7)cin`{IPU5f8mr){N
zkMhYrJpx5nRi2j76YK6P5~WKAhKB0ruo!JJNbKb|)4OGoGa-Geko8_Qd#P(clH${m
zr~!9^#Cu}j1Me&6$mFtXoA9*0U6+XcZdKEt{AOmEavw26VdW5f>p~r(iD#QAZ@w^7
zE7dNoFE7qv0IFPYJR23ME^|}mRx*&4*h^zue|p~fBZH09B&y)zt?0W5{eGC$mw7?T
z6!wfl%!r*RvEx~F$Jd=@Y=1{|t|ypMmuBy~Lfn%;?><Bd<b{`cW;aJjgTjIaysH-5
z*fp$E-N5%Fdy3`~J#AtA_I2)?8{;H`IM?Sch|yFnn}dIE_^)~}cdj&YKH#}*>JK1U
z)l?q*q&1OAvlo1@yEO}^fU(X+_4{)<U@ifD-ToY}=oTugO0S=ZV4pKu2A^lJnaw(s
z)Ci1odj;qBM`)cHYF>UaDLTx2>^`e2Rn3Z{8!fF1-rpZl95OJf%~{{S$iNJm-Y@K6
zP6%x#I(u99Bge$iZ{^5aEWw6w-paT;!g9Mj2U!GMcS$|!>osK#%mj*XA`uu4Qk#vH
z9Py|f=cjE{kC4iq;Y6xdhm_B_)pe23UE_bsj?jx6BUxsrK+`eEP8ZL%q5FW#rdr$q
z0Lb)bp#i9_;x}*V3>Re<sKB|p3xSjMKnuq8YZTuq&tQQO$dylaPieFmSG?@tY&59<
zHJXD>=0pD2!-r>J;Pw?9?upJh^3MEYZpWDaQ~zY+H^qa){+-xqy=0uss=ne($h2hJ
z!Tjo*)i$s`2w<iuyPL0d8yMVC%53Gb3j3wuzb#+<>tx_dagn(WX14uKBe^w$TTTzx
z=BKHvn`S{4@V&q6csvK7<i>M6+_e1DB@-M>$mO;*q54vhmz_%g*C?|9E_5^?N@GQD
z(^3;#nwj9&Z>Y8%hdRBeZVb4NHq&$sYdx;nF4{7TiUiEkntAuLn8q`myTF1Lb|Lwo
z>6(IuzO*#_r`j;6P^fdjCwGv9v>4x+Hg6lKMv-*|Ou|kH*B>bxp`YtTmD>e}VDasS
zkSH$s1HBh)5loD@LuMpcdafaquLOjD<gn_L%JJGBj&FOvBR$0@iiyUzl&Dy4XTh(|
zAE+E8q5W1(OpS(_{PHwrJ%6N7R(qz<(7N{J(^$H6J$WgYF8m&|9|oHQ`|^68PsimM
zC0wf6i1y!Ez9d%EVT!r#R>nYm4bPKrHdhn^WR@h$u&*L}Hr(Q&cUIEj{c?LmI1GBa
z`9PN#oFCgE<t*`C&4V7Us;7>TLPR1>g=?|hu3kzi3Bd(!Ila~8D)T&xaB~{+MeKQo
z=z`5*#_LI`#Q|}6biWaB>EKImQ?m?hOHyznUy*YyFBkg+m!rC1SQFSu@{Lq2G8{Q6
z7Kaj{mtfU`BDT$jhkd>|dH51A;+-n}LC30W$|i0uYX3lE&Ro~bZ5ti$De#rX;nCBE
zb5W|{n+_5!Bc7y+vm{cbgXhs2ciUq_x9FaB$sf$}JJB$}y|%YqjA#~_JkrAb0ejzo
zv2Z_qM)nQ9duAVf50?MVE&shC!hf3XmF@ul`30O>w6BJG*k@%WNnt2{A99y|(VrrE
z0o0%^P~Ap%HXeZ=)BUXVj(a~*s-qEma)$S>e?t^g>H&(9)_k#_#j>q1GM>r#xR(nv
zXvF-KTmB^5;|~H_ygX@88v>_Ht$B2>?>HJhN(}RhNh{p%;Tb1=$~us>EMT>Ll#+8L
z3ku5*N=;VZc+t^qNNhhO5R_74nrH{}`{wlCV@7{ssY>20iDJGFSvt~g<htg);7FJM
zV=4s+e2!acIf9CztuV;I&@`u;;QpD?cHO~422F@{tD5?M0z^+G$GrN<g|DhxvSY49
z)VLwNDETJ?qaX7?lN91iCtbBzT4CjiG{rZX{^|&LUn}>+GN#p;raYn&cN`UK#Sc?H
zn64DzdoqB)iy<O46|cw3%|a_d?d)54Nm%=(+owE}ou)*O0?0*bc>)~1JAJrf);4*M
z8MISSqc<FZbNZ^`A@S7Ysz!Nn-0(ro$v4@6f_-l+c2{l{*{6QT1?xzxLf3r?hwC!`
z;c;Gh-Jg#kXxD0}QZsrBXs0Fu93OdUb{C^TLX#Vo9VTeQrm%Ekrgmy#q+_6Q(9svo
zRoqo?8j1J8Lc9jT<pe!Q?ntYl98jO*3+edXb~d&z)Sz-A4SH|3?^)kD!kyt2<kHB6
z+q`!<s(uCNk~()Sf(?WjahrPKTK_da3=knJpki(F(9VGbGHRJKWF#|yP&ChxQ7B5*
z%|yn^l<|nw9vE}ijoxV5jYeueRFH%jiWeBlIvmnmiYbtY*RVgl+K3hbO*`xPR9^W6
zR!=<A51negtNv<P9c{+d?>YzQMdDNqu}m4?yP?USp|MllOrzbNG1phmi`az(*Vl4c
zqeEF|p3Dkma+@nuW7z}mobq5u{e5*BTM9XA0t(Ph7;fogu+2504DF{Gc41IawooFP
zkfbv)<4N0?5h%`&?uwHwt&0FhtIJT<jSzM*)ND#Loi@}-)6EQM8UdA4HW|hu)#h&g
zh<ykV3^skr9$}smPNa`~blj079g;-R?$4sbpohA+MgsJEtwFJ?pg4H!fTISAT?BnP
z0JZoCeK{4v3J*PK!51*aQsBdDXcftDWks|B1BG{E^z7e!5_ZJA>GBLPG~TJhlOVy`
zp!dO#Gl;61C$JQ}?c1`|?N6T06$TH-HRE94&2Ng$9qtDC2mb^dLqB=OZhxdjPNYPO
zQN3rlEd`3>mQShD<qe?6zR~_qZ29Mkzwc)>!=80ef0~&{!^CKUDsxpWHn2Rhujw1^
zN-+!eCqshFAs5YZ*4fW$OJ2WMD-9*8xf@n6#A_($n_Dj)mdkx#dy{jRUHELyQG(NJ
zfsld2PrS6kc0ZAIFRFP5vzS<)j<(f`ym!AOsXc^9CzSPVC=uGd!~t$Coc}PhHfU5U
zSCHWsdL0v4qKn(a!T78ewXP*l?>vUR_^sHAVO*d0ywtWX(VXGeO&*a<yYb|<o3Z^Z
zn}5fy?oB2@a-h#ZW*{?tM)pyBZawdcwe*~zlncB5`8M+4aC!qf+Tz3$Rji-)WV1V7
zBMvb?S;c#k8ith)SE@E=r+lb*q^WoWpLUa6<kILE`keoWshCp##+<>QeLi2xAVl_+
zlgvdjLwjhSEdax#C#vP4gZ<7|)I3(Rn?$!8tW3pA|EXx1Dfg4!3o4!IiJ8QNXRf*)
zhO)OxCd0tB<{QvkN0V`tdOlS=0a0u?h)rU>d9$G4TryfpWo|F`#6g;e%x8L0cN2DL
z_)m^Iz=YhOxx1mi;4r`lh|v<}^aI(j?$fTNEQz4NVx#;jD0RePz7F)klH$@oF!BLf
zjO>QJ*hOrQlAGZ3jgjlY#P0K#kY;3j1^?q_9&?Qwt;($e4#=6%MEAF<LY%Tk_DMn+
ze)=?n+h_Z=LJm5{&<=;nhapG7$^z?}IFHI(E<4S*eUF@cJDztY)uW}AC}xyn87Mg<
zXMEfGu)(;f^^M7v4g*-^^|Or}Gau4C^N~=;5Mk+_UIFG74p_5|;633{_Jw5nJ)iM-
zrzyp-p>;2BDC)M)hz>bMq4hrVbY{Z&a*Ks%|Li_dvsFjW1ZFXnNnQUf^r=){>_WuH
z_;Kya9muy54;&BpWtLs#t^>zejdX~^9F_vd@#W{mTR8p={&xQk1GB5ttStFtZaSK~
z;?C!t!3<*Jy}sHv;+!gOowINTUyd8Gn--Ft-U-_R*?c@{FEJPwa3Q*3IS&C6*>LJz
zp#SNrh9qE42`~pso(0bM>e<=6Vg?Lo6XSV;r_LKZcC0Yg9YsLg4}krNxd|1y&3mSo
z=(%7pQ==zZfX9KeQ#6s9&p9lcbFFj|KjWQw^s3vHY-20zd6nY%FzkBbRiU4Hx4@Sh
z-^U_5ZL8J~J$hN!+SDV;j@6@+mDc9|Lqe&|IIbgYQJ%>z6%L1(4Xk2szwEx}=NYix
zan#Q?^Eb1QF*&mJtv)s<bJhJVH!iu~Q}~l$uUkt|3bjj%6PfW}k69NKlS{uQhf}@G
zbTSX@taZn=oA;);9XRy{pF{3A(9v5HQKM1TK`18cL1hBKFp`@>q#-=?r32(JYRnLp
zQcb>^CVSDL_d6D%3`XRpFZb^}0%aqVkxIBWvkXm>4pXV<@&JEAAbX>|;^$<To{)9z
z*wdVTqDd5VmdQ-e(-JY1hh%k;kNdRkTe9IeiQnvBxrRysJ)0j;>o&n73-QU*_E(?T
zI=UY-iYu$0=E0wb@A#ie`*3Au^4q1QWB5pCjwbPh!Qp+8oiA4Uy+7SCeP)>&sWqUz
zXZ_65OLw2S&&sTKLb(N-jQ*;FT})vq^pw5s&0|0q<mSY&U{|Z8-H`HJ^gkP=IQeij
zMf8rw0>4g=gvI1k`E?WuBlT`D6tLAR*)zsZyo*kGa4-JVg}IJDq*u*TPN~krYxh_}
zpOJ>}6YVyMER{z+nRRC;F;Zf<0GKz?WdUc8-LeKwAKPSxR-*MT{f+`LdVxz&<+^0j
zZL>VN2b+f`*^@+=rMYYnPDiH0Nh>lnu!&PtuL8lI<RDU_QAXlo{wAQwkrJ)H?5rOG
zg7>h_=`dNHd^<Yxr`iJi>Pm{=Z=@+h(T{_7!XJg+#8Xa5Xi-nggAKI=rd}RH3g#{e
z0G6CzAAx?1z^td^?K=GyK~j8>J$*Whl_v5wk~VL#_E5bJtrHwwIk#3bQkAeuf8jzT
z961^-hD1|B_B>Ju-=*ECI%h1nBh9}C9;2G<2|m>HmobxD`KTDy*75zj<7}{eDxzgq
z?Q!!bk;;Nju5xCze$~Z+9xH6vd|4=K%0>TDkr=H{bs5rHs)(||S8vqJpmXJ0xLYro
zM)CRXKjx+HgisWU05P10^P?Q(060^&{i7X?DQ`h>Ng~W3*zSBi{QVOu)ILe@njX}W
zfmU%K-@e#97k(r*9d1c&ZS?eGAZC#LYg79LM>i#98j9m@bX;SqVU6xp@eIF9n<Xu~
zn<yU9#(sn;PK<J;ha|o>w5i&-WyRkvhgr|{EwJXVLTfTy_+$T|o^t{nC2wk-5_iu7
zTNIaN#$V$8CyyI?^`DF0NT|r@+o<u1fSVA>>_p=E@@20ZI?xQ=;tppf1(ylcQ>pb3
z8_N|!-@qJUlnW<%40n+nARmkE%Jt#S-opAzh?&(D7}3uhETSk3kiQx-q{*QIj*cJ7
z_hjhL-_l?7BZ^~DE+}U%=U?slN~)mv;=eRY4nE}Z5C7>9PHYZn>{8Io9Uml!W{G@g
z*9j6!;3zDUEc+{Yv#pW=Rq)baDc-!BnB(gGuf$?bg(@|G4rAyFxoKOEVQaZ7k~LR_
z->?pllQCb(fcTtVtOy+AQ$RBk5EKT}PoKpbaC#HVJy-YK+tnc=0g6`x2rpjz=Tpr$
z|Muye$AO4vZsYnw7$I;!mX2!QNdBiOp3BHF$SDjELf(n+^N`8+!7l2hwxk5B3LkO4
z;-m2@`*<d}jPXy({}Acx5p%9;5@qhp%d4YE5E@1vBJh+f1WE69t9zx0-*HzDYLj1K
zLfq?mt6&u`fF2b)LZ7Z&hG2PMIAs^HiWjf-n0v@0%OLG1s(x66<(lV731O_n?r{DT
zJ5I~VVXp;ol#kdY<Xb~wQSQ8%bZ?;;W1Hmb1Y3r<DXhC_PE7Wjk^LA$sPo9NJhcYg
z)E(W8i7M6kZCyc4`o~hii>$B+)s&J*_xvyvoG3duCQmm_lV=^D)O+218~qmF_B3M9
znV05`6xSYDui{turpml0n~nuCB-5V3dK&FmB^GLFTfkA^%}gr`w|^}rj}>Hz3s5`~
zku7*mBWp=|RJmWUCx{@Lv{<%3uCCSxOlvH#c&dEx>{&_@K1uM~OxB#ayt-{yA2%;R
zYL;Ts3}R}|zyHuSyr8-uU=iwiNTT^$=Jnx>$&1wJr?Fq3St=99F4PCt`z<f1Es#1Z
z)`S}BfH(7M3l1U(v*AgQ!0nCD7QhQ*^cd$RS9$>9gxNGFY+e_yrgS%yC9jQ2jx?LE
z`_YrZ9_|_7(gO4}B9o78ks(8}>Gbl)t9~Bs8vb3yQ=mZtXXv$=Y1P?$ugVY+n3Bpe
zg>+F05iDZ{oj*QjHJ=AaZQU}08{$eyBS@D3IB@`-W*hC?0dQ{NAWXf<NN%`$Jv<4?
zC(}v>APYCVmtxi47;BYA(B5_5{m-P;s78WFiPw+KEe2Q<?@pUv<Wdh%z&EP_WVKsn
zo_no#zdzExVq)avfHnlq#jF^Atf|TPf)QO;zoH6Q`=Nhl3fx4XnxL`LZ-w~W{k<94
zaehmKGMbIw*pR9@=2w8`n`L684NtMx=M=POIe_8BQMWeP@QQ*{#QeNy*`4!#Kva92
z*J&3R3}Y@t5kKFA&6=snoTj_)qx=3CSJoshuz~DbXTF!%lE7LB%G*SmNFpX>6V*=X
zKWH_0PS}Ya@nWY+M|E99+CQqF2LbBLfa6X;wOOoUp}MVnmAdinAKJ)LkV|4<mfBsP
z4tU3$TdwPOtrDo>uWh_r`S&*NJy8Zx4tNWjPVUzLgU-T?Z_FxsWMb6qkbItei!7Kr
z3<oK_9kU{K4<Dp-Qx!TJ;cZ`rKJ@z<>61kmrUrai{324bbIHC;Ba1oLCG}rP^xx~&
zw~(L%!nlU1x~&$Df2`NP<+@1HCLk9I22C`1<eH!UYNSraL^1%yD)dDIGon4}<B%NO
z0s+U0!r~pFgI32md7T1H+}#V`aREE|x=SjDUMW67*OKu~($Uvctzz^?aqdEZ$FxXa
zHvxmj-??6mdIA#X_XyTR=5VH7FYhoO$v*kz>1QC}jEa*Z9=%Zvw9R*`6FSIPV2ZCi
z$^*zS6JTzJqd6J>-7|Ph61kW<11L>GXR9!YtBA;D9bqjIcwzugurlI&a{cju*JBCV
z5hJF3*`QU(r_3PMj~IRXL?5F4N}KC6>_bwEIfdrr)+EDt7UP`_YR%;Bwz0sI%<6Mv
zPSQBNwQH`^H=GHA;yTt6+I=2hJ!7h7i}ZOsC`XDE`2<%dz~ctSUq(w;(8(r*a{)#T
z`_(YVoYg0r*mKLqtjh0<iwZ0;3Iov!Vm1aE*VRO)g+xw;KB6|tkx7m8*Sub-H3qHF
zm$1SlzpNF`Thrn-8@X45NJ}mK2H-&(lcGurq-qDAk?iQ<1R@Z{_;j@8ci$DI=^*B9
zdFCetH5-cxlpeUSzNwSuXM)ViCidBHso3hGW`lY$`z2mAjHXPqsns96dcyfa?BnF+
znXQ|_lTkexxywbg1#-7oD%)41G8R+jRwaB-G9|NOr!cAep_`piOIlaI+2Td_e!cx<
zW}3?_7H2h`pAk3Uw3wkO@|OT$UL<T>8M4JSFpQOKoS-mvQClpr9o+F)7!r>+LjPqp
z^iF>d-3?{46T+3t%V|iRM+6$Cwly?YbpYp6bON%V%Cz`d)1!~fmZm^XdYR)myVqB^
z*y#TD7j2Lw`ai19zbg``TT3A<str<wI0#iZ*d-c-tqum<3lH9}9_pNn{w-`%Ac0?x
zqFM0mlc+cF*mbJKR#`0zX7y*;RDNDJi)9CM0*1#)+yeylt)gGPE`G9UJS0QEGGSX?
zFpHng1SGndC|juCYa^Ddz9Tzg5cKCnEY`^?xZH((nU`&NH?@G0Zs^#d&Y3bDFh39U
zwCT&oSUqM{&*WjuD%sF}_tHpDrs4T@_=~>6*Ec!wIq)QCO0J6H%{Gtf@H)3j?*!yT
zu1;XRm=05gwfY<8HIW(hRT4#==FVz)o&B6@kl0_Ovx+)Fi+4xK)E19=s4UHLL3g*7
zuBnsaaH7*kzcU$?n7?K{F1DG$G>FoK425u7TBbdi5k&&dlkP36qBsFpWVEepjI%^S
zeO$rvWEWr*O7ymwXy>w!pq-XP2f{XE5T$E{H1V7?vC`E{d&5$6t(CBY_BKm?2Q!T|
z5q;=lJ(e@x%7w|uQDi15;ZyB$N<Pxtr^r97vK_V$9$6}M(0!KLgP!E|djbU1Ed6he
z*V^ymUUUl8Dmhh>8SGKdZ5MFNmr`qL*p|~CfWKnbf3!-H2n6WtQFawI3VedwK0NQM
zgW9y-ssp~r^ln2Y?~Jx=l^z@L@Ozdy66vo5bxiwxMhj1(n(IGo&}lM#ggn1BRrsym
z8LL<qvh84NJifEP*8|+os$v9hgzG#hJ~GSCe{cxx`T}&N`YzgW7?lTB9^4Waa4J-i
z5!2}iBPFWpRE<2zO(-9E^wHeNizcn>^TH+X12=)@M1}YOqHFD4m6(*_ha6Mh=w=az
z`8MfR846^qV-%X-LCMs+8i_P;t$GSr<D3-cOL=74)lWr#=yIRwNu5VR6rPadoQV<u
z<Jx<c78TCkZ~c@v)1$g3W82sQ3ikZQV<$&H-@BUXpj$?({=9dm2k|fsSrQ@BnwJ?q
zB-trd)xT;M(@ghSh4A@`{}cu6E)~Oa+&RT)&D;Rp0#zz^qeXxE{Bj0@>nl0>#jH%W
zGREcdwiO?8b6=WE6M(vjJolzC<UeN9f3^W9Efky9tK2f?44Mq}di+K-=bp*2)ZW@A
z1F+SirT_hMm?^6BZCGnAi8GS8+qSBgg9Y{-^rbHHVgBR4Q#q(TLr}J#;m7sK)JqD0
zGUrq;uiz#A7I0#>w2CY>k?($*{&EZKB#MzwUV1)TsR0m4Kx4&N>Yhi|<J#vmUG|v8
z1CWug_52RLpi~NuF2uI>DZ_unOu?T$OkrvszWxT$L96RYr#@jg+<-Mrxqlv`QSin@
zn2M1D7JwcoBKz*~qDFbC!L7?hr>pZtv%p>Ja<gj$xrxCAZCfP_0eNFHGu+rhGK3Xx
zo|OSZaDJ%K<V1$CZoMC`i^3r7koU@I$N)4sQhU?Y<%CSMRw&K_2S!sD3dRa=trge0
z0T)hUcND`R5C0#>R5u0*^tyPGh9%cw=qAda_$1=cq}dStH7qL+#^Xh$^6rYKE#-i|
zk9QMrJXn&=Uf;6BQ8&V4j6WT|JN2P8Th_3bp3`f|MUkr(_Va!!@1gz!@1IKcakQm#
z^5?@w6!d=-7F@;Rf_5|3J*VK-hh}6~MGA96MaOOi6`l~lc)niw$OXgtCI7PS`=qDs
zs?KPNnLi5aA!N^44ZYXGyOo0CS+~aCzxPi>HymH9&F}}yeg}l}#eP1#Q_*HnX|6wD
zXZO15Xec|XWRF879NCL+bg`>!w&HSiSE1S<A87iKhOzhTkkchHH?mfs9N49~lj_TF
z|J-xDohCkKEgr=TIc#`x)!#HVAWrAtPVF3NYta~F{~4Gs{17Q>y7VIi#Um~d$qTLk
z^$;Fg<;<PxU>Ia|ck{`}r5I&!pyXIhazZ|!zSaYjBjx1&rASX!FQk!B4^w?baZWYw
z1X6Mot~>)^ti*DEtNuqEtH*-0S8x)Cs-)RAs$&0UK^WJ{?fTPKh^USj#fof>zVJZe
zY%s1Daf3WFt+r=u6rgvbF7nd$HhSZ|+t|sCMqc=$*$v$d^i;Z2+Ms3cQ@QUSSsHEJ
z6lQa&8}QAm&4`*P{qG+0wQ;B`dAtgyZSt#Wk)Q50sAp;Q5(DdEemQ*B@4zi4orS|f
zovk&vaLT&jo53es=1*@37^ct}A_OH&6eR|RyfantLM)Ms_PmArdErF3apy-K$tQ&7
zzW^;`rHDuS)3~`kiWC>yRNM(k0He8B9U--|mgCHg|D833Zrl6lP#5ykVZ9Ln&)=)c
zY@athBAbMMDv4$zn+MjF?wIo$qVXD!hw3|c3K!vR8~7KIDFXd1Z`~5z26}XB$TnLM
zyhs4<;eIkZ`>%uTpS&zX1{9ufaO%E}nQTrn^1Qp(43ijIx6M^LS<(0ROQd*_A%TUP
z>cI_^k3g|7LOD(Kf2fmPzokI6xy@1M{Wf5c(J#fELp+YwF222Zg19tyKQIh1eOSF9
zs?vr$q%PeRsQgMHkU*JIu2|_6!DC8`>?;2)Hnl#aI$x5JWyhGxgQTXBQLDvUk0mY~
zfna$*@Xg3GHHIO8AR?Yok_eq44B(h*(h~X@!Wj0o9%LNVu*;090xqIs1Eb)a5aK^7
zDDjs3qb9a3!vTz&p})42jNC!uj^NAWPlh@Mqm0+h;R$0SuOk`@<ur`TTJ7}+!(a59
zu@s|hqR;chT1w1FmC~+dd;x1XWg1?`AwAGtK^A0RGbEY}2w1U>c47&~Sku0iZh8&4
zuS1(hA^k~LnP*wwd%Ai~EA;Ey9D^BMocZ=$jK}q)wmC=Rd<4;QOID<w1|OpZhwv1<
z&lIiO4MDR7L&+0T(W1nm*jh;M*z|Y;84?MiatYgKfCRX9mTjvDf1X3_MII@j`Skm2
zQa_@yL@rGso>R*f7~LX&XU#1Iizt?*m{1A2vj_WUQGZrxVD*cF5l)WqWY6j-wWqKL
z-?0M;iuz2*dmS+CwR8lRouRt7{AD6mb<nvAMZBT(jlOd?+Hj$bVbQX5I*Jy|iNP+h
zXV5iB%KosuUHvjyeL+o0)29q&SY*#9w={5kcQDDt0<+fZO>81Y<&-+DN`0q}N|f<^
zVAa&wMl&bJ&|`DTJuPg)Fj}&bmZ1xg+v@Wi_loj^51rvla6SBau6;}Oj>x%sz<h!V
zKRrO^K@}b|tnytQ5=uTP(T}>6NV^P)V-ic><Vc9!*S*zRr$EI~`qp~X(V8HwodvXZ
zxsX|?WB#6ABe7wyVLR32r!lu}6T?gs2I~$C`-t7=(zuOqtAkXJ=tAK%NM^KGScEs^
zKOc1f$d6mGXt#mRM9^+1-JmfLGbG}nkO)KC#q)H+3)-7W<5F&6H&%g$Vo(j?DJ4=5
zzet1Fd;im~>eZuFk2ii8>oyN<&}_EhL(s1)SICVJG!vt{3lF+~j>vy{3$_WRc0)tz
z$@Jh|dr|+Kqc8_ly(Z|II2xF8^h*LKdJXO8f8Ml#g9K(jYT2(0;^zj7+RoCw&A^$~
zF4X7tf?(Exu#?aW!4!V9nAOz-W*j){{-J0*;O5gH#zbRGen{El&)>IBrd{G0nWK#K
z&dlgI(UeRGdv^90h$<%QLd~!o^<*sQs&t?(L_GoYPFv%TFg8o;=Q38=a)~uQp!vb`
zQz*h1xw_Z!EEVUqOZ9q{QPDANzCKbv5e<FV&V~%oCnOTGzEw)l6ER=;J!hV9O>>g|
z^?Uy6R=b#E6cn;48ze0;^5x68C0(yG)jJ(W{hDR3B$uIdd8Z0enLF8BI4K2lnOhFB
zr7~5CKP$z60dx#tU*0URXWX_p0|X_%hb~h6?SLTyWN0HF;UC)TGTDaKt<UMV!bpU|
z%5+HrLs87TneKAbtewU^Jvf}dwha9n)aOf8lW-={t~tIer!Zf9G01%91!RpBwa;5a
zROz#)8;utYC#7kmDE%quZS^U1`^@U!%BuH2(w9Usv~`Oqz(-g%`rR*XD9S(7Tb>8#
zaThr$%<Qt+u-P0{znNDN-@V8_$fY<6aN&4$o^!766QqREEFRu+TTS95GBa^G#rckN
z0owvC6DkJ6>M#&Zp6pKYm03(|B(M(@0;X)YRUtGC7YU_5I<O)Wh!SUj;KmY3%m#C=
z^!O+@+Ku;3P%W2@#K5F?rvF8(B@xc?C!Q;of!SfF$&n@jGy&OSJpSu;|27R#B^ni$
z8S9}Oi4^PV_4I8?F-5<zc0Q&Lc%y=(#m#6ZX1c`M_S9r--@w}iWQu`Kn!x%jb2Ubb
zeYX0&wI~z-a}<TPcKp9E-~Nj0OybA~0N9Y44C$}6Nydo!0NCDJF7%@g2&4Y71w(%^
z0RE0SPKFZAe>$!IOHuk?w!we@p{fLIl=q4KpHu<!0T(_|*C|u3sdK^qf_B^BVpgt2
z<dD!)gnwj{&23BX|Ml;~o*{d!26zFLT8gFbKLDG;wp{Wij}LGmLlAr@zMt=I{B^&J
zj_-Q>%8&@b&f%-Q9^oAi0~p#72>U=q$Iaea7_qvu*LfU<`(HFvLS%C=&{95VKfz-9
z_rCt`7yA1U{6D`p)QtW(l;)0>(xs1Rt@&r}8E(gpJSR3VV<!6#FjAcc2N?Q((l^rN
z%7|#0CS$;qI&KA;yC^;u;2A@M(Ru*e3KhV>`*0i~*jC`8@eePJs{wIN+ovaJGyexG
z@qhaS|CC&qD3SHDS`hFp1i&V(jUDjb(CMStH-G%=ri(<UiG!Q~s775+?684Ysi<dD
z5Z*KWQK#3?Z8Vi*{~x`41B7@0(sZoR|E~|=--^%wd^9lt`A$zxuJb=hMlg497Xs*$
z3$RL$TV~ris`pi;F=0%|bylON<HK9sHfbRp;#x$S_gUVI<vcmwmY*p9^%DJ$yZitA
zKf=1^<bk47je&r2=D+Pf&*QoPJO-)N*n4t`i;o&#mj{F1IbVE#6^TGU#Wf<h2ay0K
zsP4qCFjowqj<1zwJ7RMWltLU^c1uk!=dGPuX7&?+`C;scX`^e?IvNKTsjqVZEcap$
z0{m6l^l7t*web1n#sGyAE?j{+P0uWv%VDAZ?%kLoQt-qllXuaF;cAlpK))D2^?Qi`
zUyVZ(g>qF2tnbnP``zmQc)r2U!+_13TvnZ`OZU&Z+rSEb{8EqZ4#2D6bOLlL<fl0`
z)PVrcDY7K(^)p~Ki<vWhL~yTqlQksB5d<)wtTX^w?7SY$9G`iP$3}Ox;^#C6B<8)q
z6X5)V`V<b(T!x-QIDNjVpx^uge)`A2V)+yH7D)I}+9h{X25H{<Hv#GIvn>>rETD1U
z+8Y34^q4%Hk}@Abs=JU8pd8#-_J)kvy2g%+&!J2%OKo2Nfy1r*Z*VwHX5!DPisgZE
zX{u6tB{|B$4cA|)0vvAUT>&P@TXO2a-=>`sz`W6j)R(VW5zH#~alDP5t#p&a`vHJh
zC3vLa2ykSiZYS7bFEc$=(hvXtX@40}2<jq{imuPU)I}~NLOq1@s2UB6@ZF>yKH*5i
z!kcpE-)oVPjqL{xz!PlJ2;87pC2LsmwbLLjjstQO`;lTNJ*;cs>P3PWtp?cCZuYAP
zNDZB(e*-3YYU7*9bo_IDxbUCfgOcy11T|4nI#5jyfUNSQ5#uWc_-IUM1BQUz`97GO
z7m@-8+kR9Nud!vSZVw_FbpStQ#PvzEcLfL}1$r(g{Q_WabiIgAn619_1kf$~k6Kp)
zZ_z4JOdZ_C!lU&{i*!2`RG6yAM9c?Z3VyW!3MK+*#VCJ)i;ypm*9hLuVTA$_$3PEA
z>T(jVfB@cce%->k0{rm7=ab+ZfF!gApr;PLK0C5WwOzh`<^O-IzJH8+|J`f@D2Reg
zq_*~xe|NE}463@s_2x00Awh`M4DeCWOkv&~dUfK&a<teq9rxf2l5`_v<W8XQYOA^-
z_^U5qSH%Py_=whw$UhJh0Z^rsa*H0{jADRsTnYjZ!{G4*=4dMb#=GfSIH{*fMCbLW
zQxC91iGfbgJXDC^J@8fV0d(4L8%u{59kG@p0N~El5I}vt$&3%b2`7wC0Tz}xAWY}L
zBkeqIgqG6MFFd9v05L(d?UyCznZPdq5Qv4q%M&9h)~*A0Jz-N|H}7YHgkBE0Kjofm
z{I}Zle~+%PzbaGU)0@iwY+)`_!ruTCX#XX)Jcu$|eu2})(SDmgEbxrmZl>ZG0boOL
zt<;f6*=@~`2xE-__<Xqm86{_B!#*TCiQeyfrd6tKj!nY=|2v*woFA5H`U7A60alHp
z5a4s|0hZs?yE0%KGa@W*ra=56ki1(XSjKZ2mSu1*Fah9vPoL-g_m}#=e}#uwAfi0_
zj|K|x3Lr>`r<H5Aw9j7C_Xt=yzh%3Ssb)@{14Mu#e9Pk^z!=f+IlJTS{l3W>K(MGW
zKxW`s!j|Fa5@)$WP(DEuVQx!!>rCg2oIcN4wTpiObC_{~&sS~m$nPZ9gtFQu?4nt)
zeOHazYd7{!`)`2p6aj>f^c-p|Kp(SMY&%{D=qLNO4*;N#<}bP?f^&Me`sdz64{+-B
zoDOr+7z3~Og5_m5>4(7;CAuDY)gVZIOY~C)E2rQ;apkKe>BcQY2q@OpY7wq8lK-@p
zVsq-30hB-D06oa(eH7Xl1G(=PkP<n)0qQ2kh;wR;;o4!+F#!LWJmd1%d+}#^ca?T4
z7_H@;eSO;x?#vDSeE31~-gbkBN9Eex^<a_HB_7Udx1{ki*l8^*%lre#6VMWxBx?ah
zCw$&IZ+S}fp{(e+z$*a4#RYiUykb7>8h2j*9uVvxa%R8DIr3+=2VkFSE+sSwydw&@
zzbte*y%q$n$C$?3WAiWvD4(Y{+ztzTje+|9RZ&O|*r#$-U6Vh6ma;vdF_BK&sXOYl
z6rv;kvAmtFl>MGM>=%EO{fDFPE*0gf&vCYZCw)gRb}_B9^00=S)sW3oa<C6Zx8>D&
zue<ju2E{H4F+IflM=(a)0qqzfxEps?yI;>Vc6|T5UGmdU7@wZFz0crMl^(R7OKIm*
zZ`!mly2vvW)p9!K)xrDi=f+sJh#t5s@BBG}9@x%VI$lZ@ZV|=y`$g5z%GMK(G264U
z1b|Rz^5kGTg;fXnFrt+Rw9hgny^!U3J@MbmQNB1b>;TKjyO&}?vRG6mUjy}9?;UX9
z*gqM+BoC+l%kr-J1a+EJ;~$zO846~#RTMs@5vzzZ8XZ>>A`PNXuRQ&Su0@QEl=2Y}
z2NRN~Y3)CEKHU5z4id@$C;o2im4@VfJ#h>zi!_HXz>B7u0gkt{@IQbS3i$=BT=n;%
zn;5q1zyk9!>-OrrHIlTJ8fizc_nylk46Mn4u9|er$M*=$LS5(wwH)<IaK&-%RU32L
z2G!AyP}##oLLf=J#ElK}!Iq1gdW$A4;^AC*$6Q_GP3!@jNdlll`ZK|Ljr{-(nN)F?
z$2!=+e(h;b?Nu|dFo4J=a*e!>@|5=!|7$c~1p<uM7It@r0OJvj0w7^3=MrqXJ>Z=2
zTE~@(HuD~}Gu}Ab>Wg3J?ft#XhD3;rS1Z3lXYBy^qSt`-M;thokHRHwWZQv*lbf9`
zq0C~q^za6Pr`U8psge_gYD%0w)Kn>-dm2>I5{fSZ=l%ereHC~!mFM~E?C0rOV6<Bs
z4&5Ojf?l5f6kH8)b_ZP9U#3wGf35+HT3o;vooe22)N$);I8B6Hi%evVm`0K9*kb$p
zTcpG^@AHqV{s{7Xu&*CrO054Zi5R>qbWlGfR%K}+w~#U(cv<Jjm{?olsKsm(niva=
z@)5QN4)Yc87McH{TOJNDOsB+L%D@9qcZIpJCdpi%=afQf{Lz`Fg)w`98DtsBpndzJ
zQ}?Z(_wU-#QxER4b;tKvbT)s#Q{ixF4&qYCZ<k}Db2jIG*MWj;)XHU(ow;7F-KkAE
z%?;U@_%oqSmt7p&5K|j~Apf1NCTM>!$yBl>-vqkp&zq-v5roz?0MKWwV<Zj?Aj+Y{
z<R+x&+%i!BZAu);WR(OIz=X9EXlEt~WRW?_H;Qu}sByxe*ujJlx@S`zGPnb}0=oyB
z$W3oD@fT)rL)usVB~o)~5$2CuSB#TIfp*^lEZ&{0b|9e~9mE1MuQ#*Si{f2Ep?K~9
z;<+<nE3=XNtrf@M?{a(ISKJ)%KR_r+NRXer%Ra&0AcRLFAe<+U@2Pt+xdx%36Tqka
z3pE?Jmg1@Yq?l5mxwOkL8n4bi?UQK!pqX@5m!ZYyFEgKRi8pvU02bk>py<zN#kbni
z1Y1`gD>TR>MadFDf*%lW<k|15r04E}sCzrHtpVn3ONt-5GicF(Ml#+AqUpni!XaN!
zQ$CI|QW3E|A+^RDBHKm4Cw1Q4e#JOi0cPx|2@%d$-_kfMM#cxe52D5oXq9QltiJS_
zkSF!=DMY|>h{=JQ1hebzGpE0E4Xy%0sqXj-8VI%YTrf0JhO%4|0XZXH(o8Wog@t=F
z;p<8F)M~m+f`WelgJE21AE$xHK}FEsM?}htKz{g@J*U7hlrSh<6jMwuQU%G*3V}Hz
z<^nV`apqVl_fX0(G&)R|wno6qDcp>!4b=}qMU77Ek2Qmy5+v;a7;VpW=@W?<ZTL%*
z>;I3jw+xH&@7DeYq`M@fhVDkXyHiTKK|(2!?yea?Qo04{2I&||8l;g90U3}+pNspx
ze|!J;K92u$Jg<fqD8q1F-)pUNt@Cq^8i;*0RH2<@cA%+Q13-VdI5XZrq1&FlZbhM1
z4emInBv#Y<$vYidg{omF7uqSG+lt%FB89R;1|!7#4xZR_Zz44u2wpv}Pybr=!Ku94
zj{y`q|AemVx@xKsM#sZqIesc_)LzB~!Kn1FpDI4zKRt14I)hpzVqJe9&qS38+Qc`5
zpRI0foB5s07j8x{hcU1c5yE!<+@GE85LKBpVLlS<M@PXUKcQGAK#tJiGYK$sDAfN+
zoJyo|lMXF1(^jNOmfmXv;I+@<Tn!J887{Mvo*p<FcOg^1V>JEyz8L5T#)h;nYCttI
zU4g=f6Boh@MOBhE9I@$5y0QdnINa85yI$G|if6wH&0HF%6*T4wt80O)i3UTQSU4KD
zZy%_41K(~OIiKld#SRC;%=BzJZq``4E7Mq17;Me}2=t|UaiT6U`&{8cpAMZDCIj^j
zUsVXP>_-8<LA{c|5}^ZBnn{|J))tLNTz6&B1ZNquHK0kGWiw5Bgvl+!+QlG*Aex`E
zL3LUhriI<bV&#GWM4u;3WDG4BAZ&!SWH#y@lNwR7H8oCN=Hbn02hW)6?pbI%KLf8@
zqNhsWVV3!)oFfMa$K^3^^&-2Ay5PBmcmdllv(hVBRw9?i|CqyVWP;W4o+`0pv?;>c
z^8_A~7Laz5FzE3S98xE;9WeAU-f?CRVav*_*Dzezmr3<u{P=tDuP6TC{_Ob2B}gA5
z%FHlDOo9T&HY$Pumdqz>tZxdf=P!Uh=<O|v@Ffe3FX)4x#kiUG&G%u6JrO2P-h<>R
zFXr%f59h=8{y6Lzi}@a5RSVzvVz`f~5iWZUbeG~|%r1Z{`1xUb5GmVN?Iy13%X6M9
z_*05qdVjHjoM5Ow4qf@0i~RF|Z$Nvo^j#~o0CcH%03Fh`&=t<L(>V~Nj(0S&x6s9~
zEWV>2nkPL{%B`51FD%miU@0D{&QK-)3j2wn=kGt9fkRmtikC5k_8rOezn3*g?a?x_
z)tYVOJrc+#?o9b*6m)T%_x1VR<@jk0_$wu+l|t07FQ}XI&)k+=OC75J6dLWt+AY<b
zN3$%^PG;zp481gFkG7-a$g+|I8oD3T(P9aHpS}+;yf_3eBg*l+Yw=eisbBj&TB>Mi
zX5)iY&1p<aDWXeF68B`9cgc*AVgy>RJ(uMKmXWG=Gi!jt)2$$aI2MqvRr%}9#{zUC
zh0_$G;{<*<v`r$@8GR5*Xw~)&YUF$Y<#};lyFaOG27AtHoko_u&R0obw$}_MOfZ}U
zHf+jYQ#E6<5p+YnbyBApSN@38c?eXy)dSUloQLi*?Y%lKKv>8BCT2e9t8?Yt;q*9U
znTGq2-R$&oxmf%~TM5e>W6~If*RcxCj_vD*irtR+R8)X&f-hW)+z7jem2=d~)2%Dm
zbfjApH#i`DO;fDURG`1|Pn&le#rLU?$d43*mICE<e;}nWA4AHpYPOZ!ngjhx+S%OF
z`Yx0_L?mmAt9$OM^RshHLhB^C-q>ZL=xMaVQJfvw%NIVJFC42_kV|ip6j0(6KylPJ
zJ*cz+$~<$^cEsC14Vry6(A=ld;-JkHqz5U0J(b_7WYCy0=$ej@lJ4k--Dm35XFpQ%
z9&GsQoz+}&p!G!@0tyZ82iNN7M~@YMZ_Xr-aBUsqBA{2Z>>hh9U;PuN$ka1WPlE9?
zvjM!GQUp~E3widk*{L>H#frkr@2<2vXCx_CFYK_?!+aO@#eswWY8MHK5a@o%uk?_(
z=t8ge6;OV%lqu{bHS<)I*^g0KsacZO7P|0X`dmKYrKx4JY_A-$+x&4}fBL&~&RuGY
zGij15%nU5w1o3YfHYWJ}=`uQ)G@rf|k_r|^+LmwlE{UVFA`ugxG%_4d#@Fep-EB9K
zHUB39!yBleU;K?U>c75y7g5eQSJ`Y-A0mFCGg=fYVJ{P+t7`{wq<b^`G*&x5@ya!r
z-G81yqt(Qfs{}4=kvo&}8{6Rief-ATm?4TM+vV=<=38<h--Yz%Apu_l=P+kFT$pzE
zEqA?7;{;2BkizG`mL|?-;tw6ow*t$1H`CADumZ?(8jX(PY}?1BlYM5CndDdA5ltA5
zf@_W`X}GKLETVG%dz#Mu4KEnaTHoKqXqNL(jmmjHg%E8#)U;#x(`pniHZRRUl3D7T
zX5$SeZ9B56h<A`6G=kL3*VKm4r5KRI0#2b3y@yqp(W^1Z6;B+RzSXKbL{yt0F{wA<
z^oTS=?<00wvcC1gpD>&J=|k#^vq>?)S{Xiy`)ecaxI|8G_HV!r@vbyk#%`rU8YlU*
ztN{O*&vX@=0Db`YqqP4a1D1>&%h1YO3gCxvimSww4d5OUSvl4eZ1Cf8?U)mD0{v)%
z!@^Zz8+MXi3|AK0qyw>7k<sZvCYqA}9NAWJPQIB`U*Y2o0Dl`lDTSbEK=$FmIx*g0
z^!<YP&4EXem5v9$ugPDJASVwUdXtr@6-M)(3?5|?2s7y?;qi7MB@sW34YG{}Ebx}T
zZG$J6Bf(fqg@W{!XswUoX10ZSLM!6gM~bu{87ro*&11{Q^gHs=0Wsyq`jQT8EKCiB
zE2n&cImrC*p)Cz62tzc?zcPL&JNG49utss&`c*skt0uL3M<*Q-FCxVM)FPSq+?C(z
z3=qND;Zx7CUR(hx8MSf7$E_hlh9pcYtFoTfumoJupS59cC2Y(9?C<+Qr`v#>Bz2^z
zM{culG?83c;N2jRR{Toz?zFb}x};R%*4U-%WybURkcw+V{VvmN-)Eb_e48dm?|$nX
zGHf>c{S4*GZ2RQ$)AHKYL42`mRRHVfjf$+K>4TuDkT}qd%u=?z9JRrMJlfEA<m{RW
zCnK*PI#@Cm@pNF49sxGJf7*TSbv*8w*Zs)$OeCbGJ#N<_8BMPC%gfQCz0m}xX5W>h
zT}(svNS9n6EIRU6zp7@&!5)knn4frU=Z?fHaAVU+w~w)vy}TqntvEv`Y>bBqt8+Y@
zt2bGLGTt6H&?NujbN15tgqs7`<kk(AO`LuuO@tPtK`F7;{6=);#YUnH%Dpo)Q&1T{
z>B#iNME1v8&x%8b?bqI8Hu}_SCC-5jiSSza+*&~6My0E0t5n?@j^MAZaqEyJ+kB5o
z_|VESB#yAR(t%8uf7_61E_nU*Wa~%}hwd6_W$?_kZhXenvC)pDf6akOu)hS?mgHVt
z_UcUrz*yl9g?3yHh?D+?@^1;m-@T<@`UT+og@Q`l64}s>@J@%^X3-NHMO$NrzLHe8
zy`#xm<S^6V)Nlj7AU9#9aVkbIC}R0eTiii9_^3t&)HCo8ZX8x|iBFjiV_Yqt7HL*E
zpbPH`Hh4<_SHY?~!XkPq!s+}nvWeCbq8YbSg8I$eyd813XtXnV+sc|0uy`8K8|}=Q
zy@P)3mdbI$Q!V)iPy>l27q7f)c&7zIC0vW|Ldds)jgI?lpuhCsjtdHD4_@Iv`Us7F
z%r(++bVE>E+YVT5*%oNk?I~#oHZ`bba|wVlO~6|`zQcVLwHXtJ!`bQP@~yRS9Qob(
zn)ImWRCDQ#%i}*eGu&kO+^WmH#0}3L?`oe}`>hr&MdpS`M68&vd?FgxapPv&d^T8+
zJ_O(l=Ys%6?dB7X!Tar(`9V^+AuCIPUWQ?Nw}4n-FU2L5qvfDr0r8Rdz`C19M_AP;
z!aViLU1Z^0BSZ|NovQicJ9V~mR<SlQ?I1y|yk>Zje*CESlDcU6Z~ilTmx0b200tx!
z?^0FWk|4|caYy+no+YJr>Z9$yrEDN)j+-p_AlSXhxOPp$5|{$w?tN~!oR$Om-yN3a
z2x;GG{U~gkz}jP_m6|(L5aGtO?p?)1U9lw`Pd270{U#}j51LrB{BY*CK|U^{W563~
zg!irxO^(TfW$uP~RF9bJ2{qO;5H!WD&qV*oss*;4$a=|^;rP2*E2D{psSLon=Q;S6
zCc?ooXdHVeyctfKOqFWc0vznEo1q#OVa0Z?1)tDb0bv-~!~40wQ<kb{>SMWe-}=AK
zMaGA)%t|g*eeN-|#9bvkP#k(b(APSW^emgiM7>-vEV?n|9<$Wo3{^swo?2!4AS#_h
zy9Rq!oNq}8I(~Bo$GZBA_F1+0;w|oWr!inKlQ#|hphCXRQB~WA@u$=1y`$J`gY2c~
zU)@ZP3tZ98dujc)xHI1Y_sD*IOuWpSh)o4bem3{*Sdh|62e5c2s!f*6f%H|b89?*M
z9Pnc{_L}Sx(YV~IOCtL$uhNNzq@x0G=+IgKs(x!2>Ly<XQl{$js=<_fOmqC$T&mmd
zrTI@)EGHezlVsF*U5CB>D}wknbH*TrIlWDaE)afe4HCHA=@r93+IvOehdw?{8ixNp
zpBqS&<#!=ccB@s!rIh2Tg+^Zzx^_Rlr%sY-w4yVmicS{}?3iGU|DDktjOtde-MdOA
zO6;s*(>25SFRj=)QpIvKl_f+HF*^lDm#x^7;^V;MawwkYoT##ENt_+vKe)#X*EC8k
z$-`z>j_S8#of)$uo*5*KlcLxOI-0Ba(=jW#6U;;9Nb~4C7g{R-3~Eptz}wv{_fM*z
z3xC*7k*Q6*^~B5vA)uY^N0q$keV>$YdmKgpnh*)R0Rk(Ie0goGfLw<wI4I~_W_In!
zXZ9UAA^{AB%>2MVL08|NIj0q>{6%JtE-XC8AXTCTYa1Xiti_?R5$NHiQXr!zOUo$}
z>=MkGCKG%F8_8?+#O0f4LrvwdTK(|j^Q(;v2JgDNFUstkeB13y%kIa@a`*2`t-a=r
z^~yeLeD%^`)m--L?xD!J<Cw(I&tvv8+HSn+MNt{L2O4>dqGnvUdX_H4>0RaD@!|6H
z2mFj6x|Z?{M_C=;vW>TEKJu|nb5{s$nLcD62pq<q=wTx&r!+xR*>2+xuNUS-FDfrj
z9!N}fY5LGgieuU;*}JpjS?DtA-S|eHZAmj$0mBcU)c0zXMO`IQrUh#xf?svcAb(+M
zHXoAHVG?awSp9gQ3PqL;nCUmD<+xsj!A_EUNv;9%PF`n1Mknpb_*yT*SB7l+F&n$}
zvtaUz4KZ#Kg~*cNReH@q?l)_kL;bhqzhk|#F9$E5ANZqfN|6^nRIoWmtL<XocbE(R
zl6Yiv5pXG$tI=F3X8+C|OM$?L=!<@&mm}cBweVnI5R_Ax=g_8_NsC(^RRx{G<pS1t
z81KmxVIJBlZzkUo8u2Z-dm9Eg^61WK+hBiq7ym6o0?S8M)cygLGwiNVgts9h^1*-3
zzgS4&FevM45H=?VD%RlnmZDfHmAU_@xMV0=BKo*+5(K|qCgn<6mLN5sU4*>#Ok>LN
z)TvnanEDj*Fy&1{cL7QIdvW}jsCd}FOm%&&nz!MB6pJQ=P#YbcCyMx=RBK%j0^>6*
zvpTi;q~IubR0f)X5L!z%TMr_{e3L&&Z86;P&j?onfAHnq#zPym@<&{px)uFSd5o&+
zLClM_c0WL>;mmZA=>9=hKbLY8pGf&D!bqh+UQ?sT*V7O}kgP+ISdZ!6?|vY9%T>jR
z9pB0;wmou?m#BalizaWBvXrG5g{3#fBNbirY`4IbA|(VAot%!$OvI^&5Kr^RQaMFz
z51|CRSoNr7)ja#Rd}=5&vGxrw)ZQ(`dpHa9J*{)>La>Uq{7ZXIkl)-kqit+`G+Y$O
zQ|Si+3WT}QVS2ey16eZ0ynKV{fyd2UcO0-)_5rCtYiT5TQm-6j9Jnvwk|rJTVl@KE
z;dcWW3fZ4lETbsn!JoKU=|J@klR_ZXj{ZnMe=MgXtQCkxUbDQjAHsvA4Tplhzgob`
z_73_6B->LUDE8oId`6S{Y$cHjv&It#N7Vj$2@F$efy8Oj%&JjDum>URA^rzCyGVo7
zh(eO|#cep5H3EzjDf+(Cycrz<bH<i<%7=t)xHzQlMO;t3Z7dQgdpKVUbF$$L=N*Co
zXB|t#1czbyKCajYDt3T6FWLZ~yci7j8UPls2RR#*CJSkad(VKhmxfiGu7TlDN*iK!
z=5!yio<3n~o7|{IMB)eDG;0$T;b3WKc<<^VFj(Rtz=`W{w!5DzlXza+g<S`|CL1ek
z`R>xP7H{}!Tvp`onE4-qU*l4*r|zZ73rtuvxpO(bqTYhnpTx(*rs@sti|Oae@B2PM
z)qR~2*WOfqsTb8fen$i_qA3IxR{^thNWG;QVsA{lbWVQYjM*-{fd~sNI5p2?HWL*|
z$zwovtyBZZWq4cbBbeRz8Sz4Y=-M#0Y_(!+p!a?}U@0Sd&}~oM)^hGDL=0?>S9C)q
zLRsan(+!*|P-5<aueyS@!MOq=A4?dz^}VZM9mUpjSJHlf;fhp;v5`ptTXvx5PGlHH
zeOi!YyN*&VZ=|v`OZ6lp;oN4z)x4gGV;SaeOgy`pj_e_n;X^6}B<;$m;6X8nFl~K&
zqgHoDjdZngh-bZ9Vta3AhPzZ=f5X8HBnYqf4!Bwp!QO{lQzP0nBW!NkHPDyNCUgl0
zd83H*slcY`5L(|HDq=YAFWEghQAqY+d*&VG5qEJ5FgogByA$@<3vct-Uw$djWZl*f
z?=SXd?es1~BWn)8CC`wgc#q<<|00$$iN{8pL%0JU5tIL#-qypfDo+PSMmh89qkQwq
zXgPRI2Yv@dOYCYCjYfA6D{~iW*Ygc#q@I|xjyYk03dzeUSpkUxbugC$-{n4}U}+~Y
zJ-dENZXv#Dl~Uh4Yai1nv@;|_HD@R3Jr3~!aovp?b%aJe^L4cStIW)@f)8RU67c^V
zK`|g$D<xZSqZhzJ6SrZ61F*FA&d*-WrH0KkeA$h-3zWLzRwnY@`U<CZ6e0es*?h(B
zVrFJlGY>INbE?Y>kAt<Aa}re(ZhvbXfA3R-ga;FX2nqfiqY3@{98#k}%D}26N7>bt
z9P?6ki-BVo${CSHfMDS`5r`_e()y3J{J+nWf>z%L7dU(-yaT{P(VUfQIm{A~ZqJOb
zOy3vtgjn!u?b5dW{JPat2L=@y%J2-?c+F`Ugv<;eT9EHbOJt_jSwAmf#gI7?>S)qB
zDuBuMM$pFYC)dBhlPFUA9_O}@5QZ+&LCyr+s)VnJ5seyIjWbklW%2vBXfRlaczl8#
zVz%yq0W6E2!v*`OBIB~PI>=ddefO4JAAe4N?OV6_sM{=W39#AUXT^wNPIRpoVwyI$
z0fd*26Bo>QxQvVwncsbx?J|t+jYYwbQwl6@V3#5&J<C)<l7fd2WUqZ=N1Gll(NX|~
zj}Kd6VICR&E+>Z+{QPBKhk|UZXCWPf-2X(YR6`3^w^i3`e{84Q{PzdQO^Mg@M$p4d
zV4zkQ1JjCqv_Wcek9kj?-AcNrS4JZ4{UB(Gs?r%x1r72y^DR%;&lZQY;r@H}A^fK#
z8@d+hd^VIlD#K7V6!sl|pvrCp?z&054JASAb}^|eYaCrQk~Ar>6t`=z5NwNSn;M46
zUMS0K1BD>l3B;xvorT5s8^-k%ip!yHSq6@TJ_06Ht262RQ0C~8{%1HsU`;a%S*&cw
z9)c~*ZJj+`=R`#6RoVv2&^l)F5+!qtV#1$2ArU4sxQC%<yr<Tk8h778!x@zE(H*4B
zf~P#1*8WM=sJyrt(mGbNco%k~Irro0aRtf}v2G<EN^*vT)~7fEx-aR)U+h#^Dshc#
ze$T)tg}enjTbf3;@!CrHXyr}i6v(n+9{hG7Nu>j}ds9Ua3c{~c?!z{L)X8-H*KM5#
zg>h;@eSUL6#6vVicwAl}9mL<czDC3fF4PbS@QS56Rz2d$3pn+T5j^ChSRb$n<a!d;
zICTXUBvP)oJ^ajPu=7}Y5+)F?mY=|6q}(R`3yE$NpA4xESuImkci(Z0Ma$JWQUOP*
zxhR1P=*{%s!(2U0@G@y$egaaOOy#VPFv3v;Sm~UY6>Y4h5y?NytVc(kH9(1F%xQl}
zcS&i6eFA??MYa(=LV{=T$}oj$5XDyt_)Z(-8_O)va8;N$VphqDvmV2DQW}4mWCmt^
z&=mTCG%HAxH*H_mu6!@$8b<RQCDgNPMEtiXOmvyZXX4rKggeEaSp>s0mq6+!*Zqw*
zuWO<!mCdnOQbnYxH3Mil?_M9#wZFgIPemT!zNE*~jSKNr7CSg(37fl?qAJEZ5X`vO
z?GlU1rO=^y+~lsWz4<D#@AZ2zMf#MNv%qEQ3VPp&y(}ssx5EiX(AfoI_{=}Ayrp#-
zdVPvh$)#=^uW<xO)9)-&gnwg*6MDrrVeeq`rk=C=C@fCi$x>y(#>H~C+}8YwHfmP>
z{tBG!A|)I74wVSwp>coSh9mvrq34BBkI*^595{nLq+SWsI?LJ#I8tAR?0j@1E&c$5
z@3fGT%I1@Np%uFj6VXMxVByB;ll@L+{|U<Z`_W}4s_EtFS+-R7A=)b%u;pg&KTq*@
zNTsBR_2WeIEs8Af3Sj}-7@w*s973PJB#YLWk~<NK(^2qVo7CjbtZOZ%80ESaUy$P)
z)svk|%q)+H9<^@*<2N;kzq$6?<$H8>$ai*3Np9HZVHqxZAN~3vOYUlQlIvI1`(rs4
zH29ddXkj5Eo_0rh5#zq;bu5!kwH(V<QEJwpG1M<r(MZIPifSvA@q>2p5OaxIDqL7)
znt#CfYm$-=#Y_C7ui9?%3j~Gllk`>}|E8D3$LUo47MoS@!DIQYm9or?tUDK~G^b&4
zPwW*U!c=0BYQ^xTu7U|$kA&Z-5l;48aqA0t6q?wTIsmgwe%^G<6C1AQ)BT|c7|eAx
zoT(m?G-c?;fA**zj4odb)0V6G*5vsPn8z!q7t+Ct*{5=CVg|bC2Hhv=bbhXgv!r#L
zI=b~#tFkP3nPPfrkJeQ?eQIPmgTr<?dzTFZzx-_d(r!xZo#W*4k%d%M*!<RU)L7+$
z*??=Idk43ff$u^{xLUWR$%3FHKhYgQJHFH-QLHRzD%5d*nH#l#A;xm+#iiOCYFc4j
z-?>w#^YXF)Se9IBy<{wfP;H(Cg%suJGZ(`&qC%xbZIz`5{lz-sg2L401=o^?qjm!?
zoierT(X7w>H(o%M_OD{i;xipsM}Dr)+{CfuW1*6@c6ebm`!s+CpXu`<5~c^j?z&8`
zD>aTTXKmeJySDbZ`tO6LUJh}O=QTwOA54D>xFmiJOQ4liaIrct(L|*!`c9E`PVJ!#
zC7t;)W4>i|$`)#}>fA7~g)O)`HBG`DG!tq`lv#6Zy7vfd4DwPUPLk{!(PGKtzuJ0=
z4TPDP>=aaA&!S(r!jp1~%f~F(IjvF?yydBsYCIofFA3y7#RFRdnl=&!Sj$SJO(2(T
zL5@C>x7mLR6a0zj?9>#y<Uq{pSvvd6ayGY9eHU7XX%Xq=gLmb~N97ZW;%5$g*@t^_
z3x`0W4YxN94)g=v<UWFUB$~y-ki6%_TQ5pJO@g{A&p(`n`a~7rnfFoXMUA-7BrHK2
zr~rY2C+(nAwp0>qRfZfdu0?h7c(k(ggicd#CO!5$S-O#3CiB>Fsj=FmmcE#7>O>;~
zK3Lt1dqY-WM|vPK8b?07lNs(nL?hJg0*@Bv@)#n@n4}?$^^%f{WOmn|^oRJ4ujbXm
zwrM2xe#0#Bo7JmlTr2l{4mQiPm$8Va6Lv!U<5ul`Iq)qmjNgxYL+oL;E}e;U{bG33
z?!=i&ua$L;@&ls94BuCJd{Hr2^4UBL(>@ZYX*R0S$vRp%DYlmrWVNy<>SdI-?LOF)
z|I^eIT1~E*7rp9w5#%D!2ru@BbH82hWIq0+zdjeU_4u&VoOFCt!xXn_OQAdG5{hk&
zhaaQPs6T&9(XVYgVK}lT(#~@ye#hzgyDFO-W#rij#iw%PIksgT9{!J+gEuoiIsV1o
z@QrqtRPXGo!GoSdMMpx~i>s<b{BBrU^Ltv}fTW*q$;Y&984dYesFw|`y%gwQ7dkP_
zRS>%Twp1T}{9?08m+6L98wU&f=NZ3=rp5ylNe=tn>POm~d^A${#guvqKc(>VO@m7_
z(y}9SSL+Uv3WFt2{Ofuf*+spoOZ?=mEMub)xMR6axLc4>Gds?7u=3HE#33zO5~(FL
zgY#!C)CiDkE$)Kw6b7U*Z|FgEGIl*Jp;qd)EyjnFjD+O^`|C3xeT;LEBIQ+Tm)b(p
zz2qDFDtxv$$Gr{+x0X%)F9_S9(z9ae)a>OGbN{AGlZEH%KPCrA<ul(~rb!4MY}H~A
z%_wCS&CTK1h;p*}mSc*A>8N)|`4N38E=hocsZt>p>VKN}AgaV#@%2*(j@9w+hGbcC
zDsV2w1~OV>y*gtrm-%tFN;u%^C$CO3<5M(pp?=h3n2r>TbJ}8BxVI82uTJn1FEDqg
zQni^A0icSX?u-0!4wREz0RqzBTn3x@J0%)pzU3pNT}7r*`jR>n>?U;?gW@L?Ki?R{
zfRreF#KS;JqO7dl*`uN}1BJ?FOhPE+sk?*6^1EVZEqX0L1v{mXS?F(xuSE>1GKn>m
zO<mAXrCS;04WFO2^35I{FFlFlmnDcqik}gjr`h3*<-VgLW@#jxBbH&_*b2q+dGnen
zwslT;as`z)kzXG__0YVP3o}6cB<#b%is3J)JjL-D{7)4Bbkua0<0Kw(90D&w062hx
zHw3PXARHl+If26L;W)W0yppm|Y74+{iyQB1{<$o;De)~HCLdJw&sZCVg`#-GA1^%_
zK!x2F?4hT!`iZP{Qn<Y)S;F#AIX(XvWOV@<p5btQ9yJS(+$tKc3GK1<6y#DEA{}Of
zZAgSvm-%ng=o~y((v9cO0Du~%JHGXg&pjDh#P3yNvRrnHZd?TLwQ*@O)poVgL1+~*
zBB%&>YldXEM5cHb@Xw%LbV?;M$3*W8(XbPSbQ7KD@{3-3O5@oXs>PSw*AYVeYNRCr
z^t2f5u&tq@hwPn*gg&?o#JpOpx(L;=C}0O2$nsJ8evtop&-qrALiT&=@5-)ru~D?O
z7rYU|&QuO7u)yUv7-G+w=lJ$QSfp%n8LZr?zLzI(68G7Y3A2cg*{ORb&V6A;1p7!+
zN&7~&8e$zRfyoMMm{F8|);o~dGJJ{R&XH98syJ*a4|mE285&%*nAYg5AJ4h0D<etm
zomQBZwa#$eA)qzehKP(aZUlz&Uw*hWpNlDWVu`Cfaq2&+ne}EWS}n<)x`GfYOCl#8
zXrM;?=S-ky@$AG)Gm1MR;kJ;L^qv&Iz7$({Wp^caa~UCW%N$qaO3pFFqhkI^px4^J
zpnOC1eD=?-@4DwzJyI^`%G(vgL|&=`m8n^#qQ;~cMo{y7)JwAKbj!m<iV#b@cO*SW
zA>#jOmxK^?u@S~YR4wf|3Eu4oi-K3fl~}A@t$Cmv$Zc&wA68b@7gEGmndmPTTN|f#
zBsa~pFiNZKd1_3rBycSe$-Rj^cHbZx5VWRF6@!(8soxRCbbw5_w(1cLa9WAV$?w`g
zjGWl-5jo@CFbkAFwlUAy%&48lwOVYD{1{w~-!wzLSZ~!E4<oYQ^F_MYvtS+6f#}Sw
zS|<m>Y2~xq4wiuZnu#1acgy5qI)=E+-z4Z$YHMK;jE{EJpZGVgki9{hyv#~I_Kq5p
z%F2X|*%CQ8kD^tW$fM>vimH+s+eb>VG9El1B_b=aq4}Xz3z^4ctI}|j)2xaF_@*Ci
zdKM<zs5C}(K6(ekGw&Y2a@O>*Bbs%ORj7pPF&P)067_1hqNcS*DdetNzMjeeOMR=y
z<pQuaIGysnBM&3zWU-lus7ABDa#Z)7*tP2;B4L;lUDoafM?l)v^UZc!)4zsg`W%V5
z(Na6)!LUtGqBrea>u<VJ{T3wLPPF+mYfQrsTFhEqrmV8tsPl7vd_(jmeE8uSE}(QL
ztZ^$fwCPKM<v`e3gUqe^xe6t#1C@p0r;pKU?eQ?Bx5<;Avr2sQMdXj6gi?6o!`2qj
zDxBN?8-)k8v8S#s<Ms0AmD}rMsut~1e?md&R5IS=I~IbbkK%C7wz9q^TvvP`qG9S4
zQ0khqaKL0uxPL65EYiZzR%IQgX&e^YccVNwdNSGQP3m{xpA$r&GrXg3*Ck-j0+NVn
z*uOlKN*VVDB&7!KouaixRaSKdY`%_cmNiyzy(j|?A`i!V3mHju^T-vcQ3>43B^Y>B
zrUnpP_o|VX*dnZ^`ci&lwo+NDNoV)!aX)4gFRBaR8O-v!&^p08Z|eT*P+G&Y_{8qy
zEp5}BTj!f=VDg7&($-aX4^b#6ZCm6bcB|tPpT3O-yNuG(Nyy?m(;?DNKJ``jxc*dY
z$fuL<OkbDeDM7h&Z9DlLa3($|7E^h&#AUBqB3``MD{lS8aikG@#<z%%Fg9c%znIh7
zWsNmVTF%aWZL!W8XC!{g?vv!XeU$8fg&7~0`HY8TOxE}ow!K3NWC^=^V#Je_y}>&i
z-Gd#>N4oiqtRkjojcX>y>F%x3A}@<Y-12*2<rsb(GhTkS{nq4N8Zy*a9p$oeQ7o(q
zulJiL2q>>MY;rYyHY1bWL@&CeEmN<gS9;W`tJ^dAE-@Q%eWq%!wC7}rHY(|AmK~^X
zuQNq|utG%YDospCR^0zMi`bn;npo$?o)+bCrV2~Au07>6dAHHg<LpJ#Dx=`*aJ;$v
zTpY$f)oeSsW|CBN-78<&8C^`D(V@P>|A_D?qqiwKjt?C<_)t+fD3xRRM_pVIm$@+Q
z(8Q5^j_pX`ch;R5R%;~j$nxs^HbJ|&6t2>mC60-+cO!0B5>6#6m@kurujPb9p<=~l
zu`;zw#4Fp|W@OsakTxJw13Z}WcMF3oeQZ+?>maC6SSXr%Za25<hQb>6^7Fx8LwnLB
z?y5f`CARf2QxhJ(Q>l&6@si1pg4iCI@n(^Kn)6mQZpq{qew`YQm@W%&4o119#8#6m
z?$SDWwaIAusCHkv!MM3ccIC-=`2vb=jG3qhy@hL56nx4Y&Es)QvKzt#bUhbjqgzNJ
zsFyDf4i9wEtwVO1dHG739Pg#BV#i&#)1|lnUQS4NdmYIc2N<Tf{mp;42~f<(Kwtvt
z(`+wNKFP2^fghqAFGV{5b8ct3+1cx7L}h+$J2FhfOYU47!m*dmlC4G0_k)@UTv!0B
z!dRYn8hzX5%w3qx-jJCbf{to~faU=NCC^Ije$aaeLksSVl0gimihW5$#*Cm94Mhp0
zF@+MRu0MQa;nE7`Em!RA6`bW;=5RTa4b2qyYi2Pdh^fr5W3*Fi-|1Rp^kc3fxMB{{
zUN10|r4eJ6%Cgd7chc8S8(=f|7G`$bCRv3jPt1UbQ7oqQ-ko~24M_qoO2?x9n;)L$
z#gu|i{+#SND%o!1tj4k2Bq9a|no%}}`^emd3zgg<D3Co13u6wK@q(HAVu~IIO%Tso
z%egq1Hu;Qxi+O+zrX-e)c!ET$)vcv?9^5D0QH#xGM8_|~vWV7*uO#U%XdUM*rbx+w
z!d*OW^u5TXU=EpYS=cA!=P;5<N1Nz}GJh>sJdmyhG}c#b)0fqNsExT29mYpIi^kkT
zBaELlDhkI~Z#=O?RC(Vfck0ua(7s0Kn1Jb$%95vgo3CE<s!#%a0ltz>4TH4-%K8ee
zlO7ys>lshjN>CXHexti{ouK=Kn@F8@9qHB7izs(1QK8HG%<JbU9ah}O5qu2f!)@eR
zHr6;TS)&9Nvlh%Q>@Ti*y_5nW7)>e5sozqW9&v3aGJ+#0qG8MrDHNmIvf2~(mm!y;
zjUL2J+9tDqrxj>k0gd0+NUY!kET7b=Mzw6%3)SGi9{(2XfmWXdF}{E%s<vl^t9El-
zDB1_+Ne|+swxsA+$l(oDNyOf!R?>pJx3LuKp7$CLWem^d?>cNZa_f^}!(6R5G(R8s
zYic)55D;2Lw3Ko5_Y`RbYfvbu&RMUVwt?F~p~elEZ@q}AKY9#YTp3k7Cyz%b+b_%;
z5o{vL%l^54U<EeDRt3>ZLdvF+Y7toSymb-}Am1Uwt7P^1Mx=7@tg45;cg8A(R?96Q
zB)|quVs6FTUksg$2as;a5?k<X)gtw@pU~lJ-+Zhbu6KT4J#y4}m<{$TNIxJsI*<kq
zAfML+EK$Py;6f|kB;Wo|n6&zm0dIW&ysYs_5a*g?IZS?(Ne-c(=uCruY{BuvS);4Z
z#rDHPs4#YRuxoI}fM_p`$E=eCsC0v%q;QMp2oUcjSp0Z_>+}ZqqD`05qZyDyoKTMD
z$bKj_dkvF332oZ}&+u<R3<y@fM{3aGhApE`>W_?zQ6(xuAD>Q~(x(%L0O6NM-u)*i
zkFMA)SN0nknZkV*6wfLkfDU?nC|PlTS*9#$7GHUCVflOU)|=_|Y5grCB_q9&4F@ma
zMRMKL%n3I!W@x$SjHh+_5LC6}N|{V_thqz7cEMc0k!f^w0@Pc6-LB*=MO+GY4W)y3
zI^9y@bV2oq>ty#y$Z16|wV5@A`wRy}3*!J%fcrT7_;?fyjK}V1Hj#P)4{pT&OD0IB
zg%Dc7>3MGy>jas(mA`#V)|)J9wBnuS;%1TVO)LSw?<|;;W|=(IVEm<y7|W4LXviq@
zV}lDq&HhYEMXoJ8S?|<JT{`l2q4X%@mO^G|Xs&<0MTAgq$In&UK0apdY3tB_PP0@i
z!;&p=qAWa6l=8C;FcpkuC*m56P5j(3Y^Lwg@;S2_9t~s%;b-!$c(ts9HIokpM2|W9
z^AvDsRLiIpV#Q&&&w!>;2lb3LsnQN3X+VYfy~%Ps{4rr?I=CK)V3*{x9b0BTxG9oU
zG(V)~E}}ga^@wLT+C<7xjqD-{5!abT^4<8X-c`UdGA3Cg8%)`7DV%6P<1KvbYq<9)
zsu$_*I}COuMjyz=*fH68iXEaGle?p?kD~v8>O()$5yj1`2u80C)^C<IEqbN?xWe*&
zg`6^=qOZEmh6JSg`U$kan4pG&$%frVhue}S?oX^pWM}CqMDh=sE&TI@2(-*8W4n>`
zljK+^6}*e|0CmKi3VsExo+X(jX3@OES$>%@HPmCPWuCjttHQEgHjzyA2drhamg<LO
zgN7-t{Jv4Iv<Q}!8uZuezwT^0L`RJYtDx?lhnRCVVt6SXkiJcpom5NDu;}oZf3RwO
z(eLcCRf$o=?jX}>SshjK+45QT5g1{oU_=J2e>*}_j3{uDK(Gwlwkyu>WCaHlW-2HY
zXh6y4L$mSenI5wB)|`E<FVr$xV5a84dZUDv3O%{pyBBhxa52Om&v8L`K%RvxV(}?Y
zx<v|Z+F;!ai3KDn_62FJW0D?(M--7~m#&X8l;TR{fRr-ETuf1GhZA{?C02-q0pri*
z8ue+*ZJe+;cwI+48o32b5yu52t;zH`2NF!8=G8ClZ-*8DXYMpw86+MW@rX!#0Z#Z$
z2ZBp@7@|!b$WDFH>dveT6~1t(?K`rdszp-Hit)o^$xt7GkdDoWl?VKqsG|opo7NC0
zcpV9xahuywNw6Z|6%vOeggqjL(+cPos(u;=1{2)c?j%bQpThDlpUgLx4vpK5!XK|>
zR%}fKsE~S^cU>vBD6-3GvSh<47#bP)#=1<IsBO;{U5Z>T0p-Kd&oJ#L$y9&cy4+(2
zAhQ3QuIq76J$E-gu`WapeiA!|{<-yN^KoYWr`k~rj?IE-*1cNPtR||_-l<?_P<%{(
zOkhT~NKNKef!YNvC+wF4sz{@Chi+i~U4&6T^VTfUn0&*|HzJ=!EQX+!|CV%t`K(1d
z4H#2PoBVBIv|d01V@rX6CZuw>YqBJpHjO{apc8P|CcS>AHa-96<Gkg4SCujL?`i>8
zIjHCs;5-zcyXY4ymjA)5I@T|??4#hyT@;<BjRdHo+Bm>0wu4zmW5b|{hG{g^aFUNU
zH8XLiI7Q>AP4H(#Go0t=b?LGPt&ZX2oWe*!a<%+jd4gWa<eL<Y2f(aE$l`&&Rwxjk
zCczZyG&eLk<7Yw?GdV*~s|@@{s0uD73ABIZUC@r&@;Ia8iFcEEbFLmKO9Jw}Wz1;=
z>$J(oS+~n@oMw$u8OBEK;pARBdcgJcw;4vA#srY)#{!b%SEH9n$>i<g>SO1xjcj5B
zLKWIBPkl7xgzlT!&jz2o9X7c2T4^gq0#<{-R5HuVkV=zwV>~Q+^eSG=X(!z;QGU73
zJ%3eh=D1DsYl>}OId--9bt^bo_X(BB?o$75>0*44#N7E1@4Ru}>>LqpSUg{jT{LLy
zrgabgtH;|wy2R|&apL=N=2e;N1x|f<Jplr@{PYFemgAh=e(tde<Yi@fLZEiqpvswt
zcCkifw||*o4XgQ^A1HTD_<K=_j)l9`X!8psV_A-JDCx*P=e2Owl$F7KMd-@`9({KM
z^6)QRM6*WkIg{9+?-|^Ep4y7#<V;)Y&nZdu*T;?1&39Iwo3zO|6i&R6#L|MJFPxbU
z|5aKxe7T~TX=^F~vhli7hD%ZZsVjqKAMWn?NDxTjRR0?7|64SUrQu#f=VA75?B{)R
z?J)OkSH*Ugh^^U8v-*94d|Q8fJ(fjK6qWBmYgw=4cPxaq;-ggR-3rPQe1;h{j9Jt5
ziQ@XpPLY3hFaEWHft3Urz%WveWtY<D7?8!Wv5XnKJ5&LM%ONAkrNwN6Y0)P~U+9N_
zpW%wH&qsHI${R&ThY8q9$A|Fi5NEL?>EP~Q2#>lrq}ry|ABZm6gfNWW>EjB2`Qq;>
zae0Y=L~F!YNiR#3mshp#g&qu{M!`3z?u4s26}itEXPrnRl?5V689neM>q+V`dqupB
zM@7&|o7gt8&w0MP-ojK<4<X%kUs%Q`4~#<hLL+r)OcGNsN_Hn0A%>gj1+&v0`{8%m
z(A4kqC`I9I0U?mV%v{KnGfnJg@<ukX?z)sTe@EuuPIkZ~+wKGk88Qrz+fD=!t)3%j
z<<hGy{4##tW1@ePz_0JGurAF01+sLHzhpZxwR4|l>}nL>_pN354V$XslKQ*$yQ9hk
zG+l)(&Mv;#pKm+NO8Dn1Ryt=>LClygrMjsvS>xDB=pJR7rE|5Z8ATD^74CAnn4AW(
z-HR@70%lq#p+r=G9oS|ZMbF>_RS*#eJlm|ppXYdNsL1tq1cde4zP%;qZ6Gvyuf;^Y
z<doAd;?nq~i<DC6L^I8uZLxo+p#Khpwga0KYfJzd65y)o6(VNm1f11kaS4gXb1}_#
zwT#<;6o5b`dwxU>(5LCavrO;*Y=x+x<-fDoNADQ>PBt7KZ2iZOl(cv=x5~%1Fd5#<
zMQGi|83W7&V%@WskHA6oR^0lBG6lZCde{3uql`<|zg?1AD3P=hcWhy9^QXt}yM%<T
zGOOXP?)k?R|JghF&r1dd8f>D_$iA<g#6y^vL^t>#a-K&?ay96xA&=5~&C8&nlcB}p
zunxwv&<d9nAgjxdd|C0}E6HZSE>LP;x8Ag5m+kd!_IKbJUzGBs<w)2WVC7!RaXK{1
zz0zRx>_O?Zwb8@GyoS61f8Rw=HD;fa{;&c7OKBGxVne)VybVD%A-p+d8-YF+=UJ_F
z`Qf%4U5dgx2$gOC3M|U#lJzw^eS*Jzwo7wNWldK*CEwYIEGt))5vAYHI=fqr?#Xk&
zIPbc!ql6irxH?LDNP`V|04uQrAF}poXh)$#R#G)AS0d!+V5BVj$AU$cy~>|CBfeia
z0TFVQK$YDxEhAR{8H=Wp=#%v#HG7VEG23)W<HQPx!Ax{tbMm6)(xkF`2HsIw##;Cq
z>y?Myh&Koh9oe-(^*ci3mE~+qVD7oD=$Mm6ix`}044_wiWf*D<jEAZuV9&omR~(KZ
za=+ey(<9Emo(0qZ+V6|eZ^wAtAG(3wv?Xf?A?@R+4a0S%&2z^dUczFbwv|M|HBa3n
zJyE66M#5-+cl;NXJ2+mU0%|XwR{sl|!QKZ7JAE_8&)cFUo>iv6LY}N6riZ(JsU-Sz
zH24IP`p@D9_Fk2`o4e#1#rl+XZ2-}b_&)7}#Ky%tUV+eO?wPKSgaOZqSv{7d)=C4d
zwBG}6J#MG4ob-lgN0?q>*Ka~8!rrNoX>Y(}KaM-UZc$d5_5En$_u2Si;V_Pc53VUu
zhiJskHUr~>M3$2bm}9|^s?_W2GDJM*^$d(QVogG;7}J{-C`!zJ<#blXx-|uuEh28p
zjhVg+_l)|McCL*9(;u>PM|QFac%itk_74l--BawW>R1uBQOdHzihB!=a%><<Ts4e!
zp)7XiEj-3wdhf~k>XxD|p#07&a0Qat7}FWx7C*lSq_=hN`_|<#(5SCC%ytQt1WmkL
zZ0xeU-Azw2oMCfi3@MNMnIBK;{e0J*+)&_Srb{C(5G+-DjzS<z>~X;(Xscvw{i)a2
z(MTy{?{}d_j>B>kRH19lbJaXeuzL>;SVu>o(qh{A|IB(Bg%Mxupr{Xpq1;J7n*yif
zN|$zV%xBpk1HlSJzA0iPUq@1`L_YZ*RfIg?lV(A1S+v54l%%T@MQ?uwtkeLCY>%g$
zMf+&Sp`m^*(?5N~zeR{TS*d{*@Rl5&B|R)u4)VMctfL62Yy$I2;nmT>;O2Pfv<^Jn
zFddz#z7JDYDGVw|bv>U%acPH_nu-!dSYg0aK26D<IK3*sv)V0xuo9fCJVje%x+r5j
z)#r^Dl7B30#D+y1jBHeS)c$S6Uh;ma#~Zy~f!GFr;*y+}_IuN%=77+cy@+P_0eWLT
zRb#939bOw5_JMSdOfhP|Y+SCi9NqmNU?7g)IQ7X4_sJoCsFu8L`R$XibX=8m+(M4M
z5a<~+H8fh-ba58lQS!8PJCq$CrA?+;(nyc*F`nH7nL{fQs2Nb8Y+<n2w7pBtT`8Yv
z5Lqs;)d03i{5t8uvg)&+6T(X}&E}wM1I`m;3bps8kG>(Wk>`juwD~Qaw#NRIHh>hz
z-?%s{fA9eO79_1d)Tg7$G4GQX#&~-WdvHiNn%=HBwe<!vp9Vfw(%i^-EqSnDW6BY|
zN+g#-p78-{8@g(Tf?pB3dJ_rIDR|hMzgtOTc1sULKaKUkI^~Bg4vMR)A-r>?hEBK#
zx9<TeNM-sHb>sw?q6;9r4v>DU@bIJ!H9_?ZOkc_q%GqJlUW5av2Ab#(cJ)$7R+UNq
zdG(tnNL65iN@@MD_M@jFYaZdKm3=G%fHR0Y{E*P%>J7aLp-Q=oBhp_Q63O(vlhVHl
zt=dH9*zs-(+wwT<biBuFV~wt|nr1L!;jI~oyu=Q)jp2*-Off$aA+{V}mW0Q^i9(h}
z5tcZvOCq72xfwsxDzB@(XVHZKMz3S!$LM1H>eXjXZKMK^ECInvONSr7)hG!rT-5-k
zK=351bTN6HV(~1vuzZ9DI{amw78kHIZwCKVJd2k|lOm%vFQ0Wcrdlwvr}RS+R_>DP
z!4%x^_jtnSsMTbT)u-Y?4W0#HyF;64b&3b2?;v`exQy=FE+>g1I#@7y&whXNh2x)o
zw1Y@cNwrc}z-sxUFpiTIxC!T6W@a1BULMBDbMpyH<xWce^{>=Sjf&1>e5;%nZ7-cB
zD0?Iexlg6c*hgqX(cjbviWak?kouaR^M)>D1Y;B{Gj)LyE;O`t13*>UnwWOhxGf8J
zPl3~bTCTmhxlbsx8Oe;c8`%Y?wi|euuh@Wgd)(^8mLV^#8N8v$z^vZSg8ikM#j5)g
zkwcL>U~VYyQkM24YqgZxAHbxj=fYH>^Tx{_Jp9niInyJ_M>#5p%jL(n09EN?tOVYn
zOs)=1S|(}0VZ@9TST7A21xF?zJCs!=KvC~pmbtt%2Q>R{0Z;GYB_<2@gFtURcIJ}8
zz0&Qp9m#Qx9Ae`+wpWTf3g*z`W!Ytc#m6XtBc$_}z}oGL>-}4}wW5U~(kGA*yUw3a
z|D}+l<5~2D%e0YZ)E3EQ#)e3XVvH0=GvP9yYwx9VP^b-x`DOuHX1)aCtoHi6V(-TP
zdeR@Fahs>viqeVJ!f>Z?`jF{g!uatPz%VHsuoQ*pRy5sDQd*qw5KCte9+M5Z(N>-J
ztYoxoef~^CZPsa-`Fgq@0bLvrnnj}XvC6Y}cGO(StA6X1Z{DC+v`hUfqP@wyw>HRK
zF`Oah(e{v9!}xtJCQd0L<?zTESYDTsSiKN2nC=y_ColWz*qdaI1vsS=TQRQble?G<
z66n6Q7C~a>5ZCO}iFdgp(d?nDKjZvEz%*ED&`;Pg-Ve|2L)xG`ES#mLKFvy~mQ=*=
z>L)loi!xLxmhUk!E9o-u4`$nbZ9h%}vJp(7USGf(Ef|j;%r}yi33b;s(xPs~K}Rf1
zuL3R8dL@$EkSOW<z*e|U(weagcyaOi-K}u1X+)$9!!qQ*hSXw);P)YKq>&p`@(t)#
z82|;WoLtC5C{_rbjA#TNA~9Yn%}ugoG**9PJCf9v%Zx-K;DY%z@ix%I0uFa)^%ELl
zTV_FWg`v%1N89XB%ix7;MpNk_@TONKya>UiwZMS*(d=stK`P?lG~%c}<VF#ng2Ti7
z!#Ytz$jPMFWV+~D&>=4?QXtRKQ}e_c4NX%@8o<~6R}wzkNn+aFHuwV0z&<J?>d9cA
z<Q=0CJ%6Y`ae^F%{Cm0vXc(YE>*9{^%J+La*w_Dg<D)?@ZNsrE^%&VW=3%YfqAQX$
zSIHQ_vttyU3Axk>Zd6=R3cxy{su!9p1(iZV<3j3-TnjwHe#-i2;nWv9SYz0rhpXkP
z!mCIXSDxAdLJn`}1f<FaVRjiEiF296P=^un2u9NtoV}?uq;i2})g^ZT-?oE~NGazD
z8;#e4zPyY>x<eX5fh-;1jv@X|;8|zp1!Ar4CO<f2fE+V;RH?EWr->%Z19-B<NbP5-
z(Bn^_j<BFR*X5V+<d^(IM364PPBDDULaAQsY%Mkg+(kZ?03Pr#1eL1xZlIO%H-bg)
z5HoDw83L0f_o5vXd_%pk*D6^u7(^LjiT@69vsUTW<3G#m|JRouAv%yF;;|_^=hLdX
zg+inCHU;M82Mi_5FYa|qb=AnyNHQ9)gHO|)iK{DN%y`nZGILzD;Ng4rP?BeHz3f}q
z80;cUGP3gx?7yrE;*f=5Y<Pcb>0y3`KP;!)NP9xO4T(-jdO$L&uT83a(;^H<!v<f^
z^j`Sqc}m%UsRyOVhy^|2Vguzw;19JoS%}4mRd4l>(J1h$YKxKU&=6@!Edt6$V=bu`
znQvhmn#1gkGT(Z2Fm@FGl)fo<LcmLYZbNY-zVN*#*jL%K(3wWplA}6asox{jlq%K>
zaprlDPktitv9S9OHuKxqK}EkJ2_2FOOuPa3huGhr6Rx0y+BbF>b38Iee`fg}0~KjN
zr8f$W)TtZX^aQ^+oc0e_4oOxWXJHse7+#2N*daxRk%u{#m=&0CFmH9Mbe0f{pwi1n
z#CmF@2rNZ?>D*xMUTPcR9g%CaRQ$A3_glrsVhuP}0Qz?b=N1iDO~8p$hmE1$>vB}!
znUilsO|JpR3BgsZdPG}fk$?plevn^>zU2-dKyys+Oi}FOA@&k2p?OTysZqKBt`%FQ
zzaI%fP{GR<A0zEycSLZ#J>C-Wcr;xSYQ318){au3_-+t;;4uLRSjs8k)VhZ3LRxjc
z+Qg?Vo4p8n{#0u6>ASaZB5=>6!$ulCmW7J9=e|J&<84t!#J0%v?BMF7mKnif^3{6<
z%H$K<WKCS(?j4pnen-?jC(f>G9=*ZuBVVkG%rgG<tvio8DU?B|@uivTj0D_YtL6!P
z@Olkoen&ObI__O=toQq~V-9SVTZLMHc>9l-df-pAOW7Xu;K!+RzyIqM8T`-b|FpI{
zOBEsq8gPFRt^$)cEM-qdE)~an$!91pS3#B_o_H@se2Q*p+rAl!8GrMsqpvyp&m<sb
zy_N^D)f>)+wj%o2cOoM70LCr4pTOhqN*N{&`6PC!n9tFMiBDSjC8jF6Nv28__ZYD?
zi1o4h-#@tL105i5pDV0(@Bg!S{a*_O|9@|sy8^Ej@M}#pQR)Az-HIytLZ~U`rege(
z0ylVtwh*!a5ZoHJUy!v_0>$U#_$W)Y)6<fI)IFMoDITVEYvbU_1;EWBPXs76SCX&r
z>GZAZtUgsqTP+c^<Wb$xO{J{{{Hz=91MYr@wD4P6VudH(TP5nF?xy#DW_gTsPyKX!
z$(E!0Kho}hD>DDt=KJ3-q=HEATB7FL#4PZgfSw4(b$dRG=6}DkAuy)h8t}C38_yl;
zWY7l?Z~K6#zCV{K{eQ=EL<RBQQUQSNZWY0xTMyvRDuft@-wL4f;$LOwhWlRw*syS}
z!1FjufC7VQpW~I*wy1VJe0L2{;sq9G4n+-3{>6o;_)CoEV0>zIeh%<akMTga6sKP+
zF#nN}N~xZH9bel$miPb9zs@@(MywDnfJOF%dqe~3)(v*-R<$Sk;?J{ax<oX<0J0li
zG9mf&j(4B<jORn-qJoYMT)xwBeU$)JuNL6?L;%pWYL7`=MruS(Zzl814}ftAXjd}S
z087w2-j9)Gga1P<{ofzy|0+Os&@6<V*E@ZFRFvf&DIWr%Vr0oLdvdb_pIR$MEQHlJ
z`}zV;*oMgT&K(9zfXss(2k?x;OU5L{D1bK*)jyK=ufL~7DJ^hOX1<2!zWpD~(|-oJ
z|L-p$ED3<Cpv?2!XX~FS@Y6pg6-9cgW}qF%QFZ;_t#jMZ9ss-<`V&vFAJhD%@5$#o
zX#!65NPqc%D0>U2DA%_AduR}ZL0|ysuAvo>?i3MJWavhuK~fqC>5`C=6a^_kLP~N7
z0TJntmhP_a8h1SVdEWQ=zw29zwcN6isr$apI*#8dSa&S3_up=WKc)76{xJXr17{)3
z0i!>AJmB34=n96&$2<6}>1D`s1MlMRs|0nqh>aAieF0`$noPhwDOvweGZ&6RqHWa)
zt#<)(3;885HoU;+sWBdXtn0a1-E?7B1BRt?k>w;t3Es~bDd+sRWP2IU_oS{|MnMN4
z)Wc>~eE0Lr1&=szcqodWtr=ZCEs@APD7Fe-me}${%UO;v+|L;nl*_0GM7Fx_gczVh
z6*2)#ikqv??M61TB}F@krobW@fF8ByqZXR+-s8U<Ux@b2G5@-t*#6)#d4Q0)`jo@H
zSIca&hZ<zI;u-(vn{pQimEqE;e%t%Yl3)fs{!!{nMhh5T-RF{fsr|QXw0NJ7^|as{
zZU=TRvwn|4LDgna&@vsfsN1?K!yvQ&eNJox8~!9x0UCKvbS1BBSl6cN#+7;W0&qFw
zqYekQmoXBDIsoe{(SllSQmege9H#)o(~vlLh(a40E?0H*U;a{(x`spTD18hJOr}oP
z(?Dsv6Jr9d*Q&;*QPYUv5%BqExJQf3g>BPa?a<AVHytZ@926PRpj^0p4exxnkLP1T
zQTq@8+}nsB{ijRjUV0fI1cqY4behBbfwNw26hL6~`j-HRYYN3;#K8^43}e@6e7^k!
zrxg@NqC-dHB@=Q!&m!qW0q0Czn;|P&;&AkTf2)I+*;EfTux!Pn{_nS%9Q9Tks1EMU
z{&9-)F+q>}ug?3o1Bkqv!!(-!n<uJr4-hv$C%6|^NnD-u6X~c;c+5w&VpG}jRR({!
z$2{-t2#gOQXPcjS(*Yy%gZBwZ*0KDkO=&|*`W+Ya1)dhb6}<^SoCvYZzN@nMLsZ8r
zp&eLR;KqoxNWs2e`b92f3-M|pQsCF4MR?Cs9#4N2o|$BT5T?a|0_;}(UdGlfEvki2
zq?bp2SHkDOwJIQ76g#Em&xfLWg>seDFd7K-1YI2%U6D!dg4I+kNjfckV#05?(S}JG
zX5&W2K3b(!H33X$H*O|$x6BI_hC;z%Tymg)30`i@D=S9GG;o;B0_5DapIJSnzI67V
z?T^leJ*IM_WwbAU;z&_|B}2%a)AhXbKKZWKDtIAmPt7qw)yU-M9p5MtSK${gJJ<Dp
zMipaS)CpycD9+^g<Ae%W35O&NfF|DSZ9x5-nMHxqP%TFt&?^)j3%(z8TLRa$X1DD<
z=|XiEH4+0hQieUOs0?O*k?LezVRj3j;e`65C>9|T2?QYDo-sja{myqRzQs&bnd}7|
zynUpNU_l|<a8WA@*nDQdiPah<JpE*Yp<3h!yp!XZ*_ULkY2f?YsN-TB3PBNbLm$`A
zk(r!rR14qWe=(X)fajUAcdkpE@Tu==W9ITJTzr@T0&9}6&0ud7PVGY-fT8_9n=XpO
znZT!Lt~y@x7U+6jJN4K2Io>6HA~zgqjo!(y+r@m=$?*9S@BciUE#auZR<5u``B&e&
z8`pco#t0t;&qLQPFWXJVnZ!grQ6jJl<26G3tk+7C<Z>nJlEcqjiw3|O`_Otq?m=vu
zbP6z3u00I-iUMZ59|s<CF@NniSts{1;PmgrEzAhbl1#D3l%Bj~<k|%PY8$ZW%2TGC
zBpiYg1#L4ZMj}r-QR6->hJldc)Q6Ne8Tb=a4r5}y{r(x@0-ThLFV9a7zk;x6KiRZF
z_w0X$<2QKVCD{gMHvP+}21UQX_2*F0!EL)(BX8UeZq$a9v*O}ii}2JD*8{N2R=}^D
zD+ALCA`6+VF19QJBJpi$R9l16QSH3gN}^}_`Hvz)i_V(Zb`~dEN8#7Fesre;q3Ar1
zd21Kq&*S>|VN+$6BDRH$&tEmfe4b>>^k}jF09c$$JgN1@{5X#T`=3wmwe%PsR-leX
zgkkrb@`$708Gw0;or6<Pz#Z&ORogO);;Ci2*S~j`+;q<34cM@_HNVtm!GT=Wklgou
zOO8mw+#<2hDHYLndrGazRYj{NkXyG^571XzLDE8Xyg}|*9+5ABZeQ3Xc&ydboUq+O
z-pCg#RowpruXSnu_i+i*5c|!%cTF4*f8ec76ZvijIf6dzTP3k3fQvOX>BbR}F1$$g
zJ0UNk${^<?EJLaogr=ySHKQ|#+%HWw@;%|Ns!6UEEdELZ$kS+WZ5@-~k|D{X+Wso%
z6|0Mw-3u^qwty*ibqZ$g5G0WB0$O7$xYu>L=<H&hLq3ogQ2tPv@~)gxj3Y)SiNnGE
zk$@(dY;{kCo10BnzBXw(rkMT7jr;kQnn{vBi;OPb*M_pHBPkcu)3RR{cYWVEYPy<Q
z;E8xP<lYx2RkP&_u2mZ!DuGa6Yk1284GA(7rWmJqk8v4EZY_Nn#Ac10ii=UWsg<6)
z-@Y=Xxu;p2Fyvs)QnB!=)H~nqp)r9Y(8}x1)C|bt2Lo=w*9EJ1;eDz~HQCh>6bx2J
z9F-y$H{356(wxq>4ZWV*{aW*M&*b~V<yy^7E4%JX(G_k<v8N539rMe7Mu7(0Nsu83
zZ7RE`G?5Eq7d%(@8<Uowk$odI4Kjf=#XE~LSKyPkp(@(-6gOM)CKOE->{8U~h{mjE
zy}OTHUfvR3js$LAOE2=qFXv}K_4Bp4KU+yI%LSO83A{h8qAg*mgzkZ8cpJ5BLWs>I
z6H@LN8y&Hg%l3T#o057=v&d_a0WN+cGyU~{G|m4?zylb;4FYGJ`|J5eH<L;mQviZ4
zrZNlN*=T1AK*7x<a3yAgNk{^RJP9R{^LGHmw_V5tZZIiZu+Ur4_?PF&3V!Tj3D{ZU
z2<)w?TDKe2CfS*@F4nDZKxxF(i?JdF{M=E;*t3yX!5}Iu$2^NhsU|}K<fOty^4FX2
z^+$)(0BTCpc-+qWI&A`6-M7Tgw_9h`j85Yvu5n-oqI-^molPx(KZa0Q@R#=hVB>K{
zh`cHc{TM@oueo6vtYRNoEJ=-#CKUViCMZ(8>y8eM&;B=54mV%PC9RE7jRM#zK75hr
z4p2$15)Z`Rsl_v1fvs|p#5!U43b!p@DySrWy+A+MK=FFvA=2lgAj4_}(5QD14k7`p
zYqnwZsC_;yz<mreygD5@iYQq#%<wLgRxps2$&`JR;a@N78DPOL4Pm}Q!LlN6;Sgg9
zP*|u+mNnM`=GVS~w6O`H%?qf0p_*|Qug-wvB?z5ARu6UJ63U7_C!IPGo)yg)c(L^`
z12eh06TqD{b*&1+SdP8#C~<GX|6B<&V+Has#Zk{Sy_u(;CJd%2e4HF?F`w^cC~rl@
z$9$S#->f9jOnSy(RVH5x3Wj`h?aNOrrOmUb`leQWf?2vOu<+C2xJ@XaF?UY(Jm^Ol
zdo7F2SX&q09mN%XjeNa-N7)Q%us|YjsQeDeVb1GJe#gI891wvc9n+?5_L46#j^+Tt
zJ*?|aV^lgsY?VDp8I{rL<=O0gZm&_?h<G5LyM)?jQ?8D^;10%igf7c{w<`qG@aS=x
zwtTkmXRa=fT0x87#Fgd*&4i*B^W{y1a?_VN)<qR+D!a<KE-F42!&7l{<_dBF$@v?&
zgfFF2cJCjGg%?v+N%5Xt7LL4*?jMi8$Ssg+5PsUAx^?!akwW@5x`WsEG+V+hkD%ts
z!16gAOd(=^D=Wj!j_``F56pI^x~SRGE>Qz+!t%&kK9^Q7pm{35_BlbQ!_;=bI2xBS
zYG?f!L9Z*==V&xh)>=UGOaPqSbIrv9Z;iz}Ix;nXn?|Yh%iAIw13Y0~EFGn74IrV~
z&oJUwY(xPNr^QDjk<=J-^Z}kCRM&Xxw_66)3v%;DW}N;rG6k#zp-LSMgx9@4??g)|
zI-p)`fgHK4i#6>97_W-jHeDPV{CXoTpBff;bC9w>`#5+#;06VkB=p_rm*_xJ8~b*D
z0Wwo0h!7S`$c8zxQoa@j{6qFO1al3DiIMkZz4jopP!_>(bzg?Z4WP8>qFbRhGl6I|
zmt~`FIZ4$=aOO;qK8B#JEa}A>-!GHJgs>Hxl%G*4*zrpnnwMi?!528gOmut$i<=qo
z@`^so$C^LBKmkp7)@*dt&jD@7UcM29#zY$H*_H>#QHRe2__q`8g5&pxvK6YY&tkbU
zRmaS+OQeahi;cGSqI%k^h*+3@Xdi!pOzRt))QJicS|C~;<LYol#jz4XPOEmkqQ`a?
z*O@+i%|sTMN9T(lunm5OLKyOm(c<0EI1;=E+?uh{>D>&0_cuS>eY_ojBCNq0z!pZ<
zllB|UD4?{<OjJAHIDLe|Q1T}X?QiTvFdig+irlT|U22Ad2OP-T;97jHhUcHR1M2ZI
zH&vRB97<SCSi-67REIA4j_g7JsR(<dY&S;-qWWIsYt*gv;x?`(N4n=ys^Wm;e5BxD
ze)o*f0$BwtmGI;_sLlj$z+l{QW=J1geRyOX?SkY6WYC`GhMNU;1(mwFdN}r2B~XCr
zOvr8yN-|Ve)EpJUE$BZZyyx%i^21PLm}{u>a$`bg_^w^hA$AurI;q87D!QV9Jl=cO
zH*eE<X5__4-)mgtcHkFaP9vgtd>;lq{MGZ$4=O;Ju&sWIwIFEGamU=eI9s}Y3b;3-
zZfO`rwySREH5EAA0n^-8(Y4})c%u`&$HTW)`ngCn_Y|q-KMI-qeKk*(lmsy=s+adq
zUCT2D$TwOWOIv1!0l!9o_FeNT_&~Cc9%(nq*QOJ?mr6Uj-ysqfMfU<_jIZY<5G=?d
z7&SqQ{p$FOk3OHjDI*%~1|8Mzz@xHnah`;ASuz3q9A4brwf4qvcTi@fH9{mH^RvI}
zg5Y216b=<*NtF+=gd#UU;!;p#cJC7?FVcL&qxkDJ8qDs6Hn#smRQ7kf%K-O@*4#e=
z8gTsXEq!gZvQ3=(#J0$|B+o&R3!BY_VdsRwB>IsDW{Hga;mxE9hGC4}t`baxYKehT
zsymblyAfL6Zd1$hpC@yb)ij4aawgLbC&{~b!}m!Q3fF3rjYh#{kDRcOlYGTa#vw|f
zachZ(;?91y{XKSCiMMmvgP)E(cYc0TVLgmd=$U~A8zY5$TwD$uR}qn|nk?4DC?p?u
zn1eLZ)JD%K?snbG$Ec-?W|Yr+a7KpU2^Sv>{-8{(iOCgFx;7h5!i}kbuq~_+JFdW`
z4v8vGybf4FY<b8}?(!7jN8nT%d9L+PiDNn!MhSN?5I#%4^;Ky~R7T4@qkpwp|C-XO
zGi|fni`v4eBTdZy^CX(@OHcGFVu>BVh7$H3ZG--S=}JeKMJ-dfF-6tSL03Qcy@gX`
zw?a~ONVFk7tG3~g0iq;=arapR*KTwdeYZYy1yVQ!FL$WOv>u46s7_Mgc9rky<5|8H
z2InA4mE#P`!p$w5jWD^2xGH?d!0*#gd`H5Vp-V+JCZR>X_Xg`T``pvSN6qX4_P#|w
z%!rqA!xD(cv<WK7i7j)BGhhx?Uek(3lgxCCO9&rg%}S;{4kE<X#y*xcwX7a#oJe@t
zKu`c2eSD9xRf*WbtbE3Vno$;!=R4SR-v}qT4)n+`z3;!qo^mkrPVzG&9m{H(S*THI
zc)DxgH-hSf1*G~zTyPQbbR)$AQaN--0};|0wJKW^B<{hU)MhoQQ2Dlq3T%;w+KXQp
zyH)4z_rvH0Ly`-l_%*47a^a;%y$>je$d`HAMF_zVRyiX+e~%r1ynTW=BMaw$sinH1
z&EYTq@jp^bahcKSBPLKl<qH{S|K7JWUlqH*^USmNJj?G^K&p*5&@B#r6j)|w1?`Yg
z_Qx#w=G@5Z6}@IN@fLMjrfKjNgcqk&4pl-hqz3F_c(oKz*~3y0XBmDHyO_`36RNln
zU)4Of(RT-vuNEp)A8sx{8^|6YE6U+~670dD_F`w;rgWn%cWO;BHbu<cep$CEr~seX
zwTmP#SOF6^4$9UOSlmu&9a?wC76@}R@%+cF=@|51VIl68Rn&mafR}ww&Z{aagxxy0
z3t}x)9<2AUIQx3ujaFg^KzivW9<;J^*+?*Vh{qXM0u>=)1HFi9Lw8fBQ-<g_MM+jQ
zDs`aCULea8LHONDHDG?s;?2O`qnq(#qw5|dk1ps1VHa%Bi(fh=$zlu_@3}a-4=eL+
z#Nn@fmb0xUzZd$FN!cHOex%%9@&+1oSD)5=<HxF|Q*K~!=1mjeLIM-T#JBKoqz<lq
z2)G45Y}d7EN6jlcy4bo%N@IrhpgL*?ak-&s?sw^{*Xg?UNN?uA<BxV}394P{{K__?
zw#!_#mIN}vdwt)<cG#kIrZ7Ogvf_=z>e^-IKO6nOQ~6$F%t<WAWC?kxzk_)JHhLEp
zpFL3m)uSk@UUfWs1+dDVOD|3yeNzHl{rmJb+9n`RC1gxa?MXiO^^GhdC4L@=<XyTU
z&=i`cgGk0l5zcF?+i(8@37UP0N*+Y`NV<llCI2q}bx&~=urKyeg}>fUAz1_W(i$>-
zpw$0rVE!si4bdYYT-bHnUX`sx{4v6U%w<aixp;#o&Pgn2Ef0f8oKL}|sivJwW<Z~d
z8&p-ZAV8K2v{XgF;jf=F(6$9{gC0@u2Mf-llCW!Ru2Xq1Ln7BHw%tpy=^etJ&|(U*
z+>b7+@09apV2drFBX>Dy6;7VElxv^y^Q?w|jFL0VT<dOhM#K;$=&3CyCb5;&IO2z&
zP08>$DID^R$kx=8zRTa&v(s;LW%bPY@a-}bFY9vu?bol>$n<>s<~@<=qsF6DE?a%?
zxsWO&ihM(|`JxAG<H$~P>s<8(G68LEjv!o@rB4lE)}qOaee^}HGMLgX+^<Sv7Pr`4
z=K!ewHX>Yq8uS&u8VK`jyz?6(I;A_M)kj7@k|y-$;u&&$5TV8cvPcYTRL==xCGpi{
zxoTi7rNcP6a91Y^yuXIy0J1CkqJj)>fyOsd(yV-}Dl(XoYkH2Q88{tK@rZ~o3gia2
zfMq07Y|CPC6G26~Ub<#En|H6_X^|8!bM~;c-u%b&++5`4`2}mI`s$;Et<%qcTq&5C
zj%dB-dy_8xG@_r!&0*%XBE4BsGac~sd0Eox#{JguC#GiY?`y?e@Hz}?AO%;+t{<hF
z;OKs_x3}p9nxZ;SS#S-oW~??_b)?g@S{NhXrt`v{Sh(Z*k55Eh1oeM}KeX@XTZ(_>
zdnS3LH7(My$T>fDtDrfb^WsK;6P_rc-QADke3!=5IIzeWZ9Ti&t!=clyEA+)Bpngs
zx1R6|&@nX_D@z&z1UdFIv6>8vb@sI`Az^0jpWTA(1DHLe<Yu?KnZUM7{uSkbTCs!N
zr-nu9q)Ft1y*ep71oIAY5A6uwUAN$PKR6?_E7Rtk<#cD>e)XM@B;sbXHtfXKo}Mcy
zpT}PX<Mg%o{Zdd78Wb6w98{R?f~97=P-8EFXV6{dZwDeSWJ_iii(TRpwbY6+ogXcD
z^1n%CI}}}0?+!XG{Q4lH@SFOM(wmG}#ceE&aT|WWm3GbV(p32$Zaj&pkt&#CMzp%_
z8ynXYFJT0sANbc|?S||#Iizs0dbZ!O;O%G0*Nl1{Gl)FT`Hy><?;h#^NcTVN|2vtm
zV~oZ1k6pfhNphbt=gF%oh6gutW4Ru{yKk3pv@ul4+@v5i+<QGTy5}r4xW|6Y53z<<
z0U6b_+=Hr`CO&1qi%5y>Y0Py09^54hNd(zNT;8+Arnl+Ns{WP!==?-b>jlPP(pY%<
zK0f^68`2~NcBb9Me%84!KOP&=RN_meyAC5}!EA~E(CAHKxr|%M2{d&+K54EpY2m{s
zzL<L_m#Msy$jzAdd?oo5Ei)klk^-yIs@pR{ro!1@5sM-QPQV(KCyU8V58NVUZ{~k2
z_Hcz2c4yjn*^%DB-3>VWI;a05P(yVDG<a?E1dwJG`l&q0mtb}qP7Fl2b_-?|m4Jgu
ztzZcDnYgkVVaG{xIBaLEV@j`L9&7?Ocomiu%<<P=VG)udYzmZVnxkMKp?cP0n(_X5
zA9@DdAA1~z9}WG9_?b8{+iXhw!mv96YY5mA;x}1+oW6TkszCD22Xl;?c8$NeOxRUl
z2Mbqxn>wGyjZUogpRqn^{wmaIwekSmO#Ai;GX)3lyQ(h?e#cl#baWtJfZ71Xb9Zz4
zC8#_Cq8`&vm*2q`=wBcEh#`u-lnssSmK~>WryzDFd2w?|9-2?FC2gF*vf_XF1`cX|
zKU}bo#Zt+=xtf6?UP##tRI~StnaJj=E1H+&5$>OyV9Zj}7SzbmS6?4fe@}Lf+4)H*
zwhl!BdBr5b4(&+Xw(3##E@ATzQ$o{mZiCo6arLNuL7(t6V51?la)g4ycYqk#b`p7{
z98`Sjj(nQPM)f|$1uSMEQ+yb|_L!<)z##-PCA2`NpeEeKoe#y_k9^iH*z3-a6X%BO
z+U{>3=!L~Wy3X<7vN_HR^m5^8e#Mo8s_l<A>OXjwS;u^3v!?DkVh9J1P9Jfo$XWz;
zFI!rDWIbWKA388IVEBYB2z`&M`{hv<7*Ij*pawclX4DH0_Hm3K5d2aM3#HdUJM`I?
zX>VSdd>>V$-Xk|H-vW9srdj6p2PN(eT*y~Wv5({?`0yJOc{gxx(56u`tSn}$+@uA%
zUTHlGYCI<w;DRH}Ivsk8ZkKgJ<{Ws)RQ*H7kQ6|U{lugch02Lz)m7V+p639<T(|Wn
zbJ{-%b0<Y^AitA{JHh&ojLDc9r34WP-1uXLh?-T7%hD4=7Ik^wq?C}{SI()IK=q8w
zg?<?%eibL0&4a&zn2(5v1vA`vlWry~VP1{e+FJ0;Dh?8%Y+=Gl0%-!hqnjKWn_kZo
zcP%^@Z+5FDMtMpTclAAmap;iBe2X9p^t(FpBTBQaczKainWU!6MD6U8!q+P+-{hgl
z@RW?4xO*;Z6H_|z0zO?9CTJ^yY9v+OTHkZcq6)@z22&v%_>2tC&C2ba)PC;{(tVxC
zH_=D<s07*Db=B>U1Q`!4Huw&_l8n2GtX%y?$ve;u+JHV`n5<B4Md$N@YCrr<Sd``2
zjmKK~gt24#imXQvX$!BEMwf3a=Tk6x4*GhS%H@De#_W)L*si`8x2jfd7-fnU>q|5?
z7US59P7AXD8(@9~Fz<?16nwa>SCO!J?yab+7<Zw*8#`^MTdF1g7nkjRw6p;_{es9+
zf=->Oer0j7mCQX;!mC0dM*bu)ujq`D$N5$Ycg*;zw%=;&WlSJ19x0Q8A*~-|>k6hz
z;}@pXIx0-pbq-qw>9%AC0+p5P_uFk5+X4m5E=-NnaZa+-b-iQ6v&p|OeF4Bo^R(EM
z&~m(>ngJE*ig_w1tZNO#P;(e$?5_W#GVlYpr0<zAvCKm|#`iO98^u6X8kMMXS8417
zgeIX9s4z1ZuJO;Kr0V*1T;<=*N9P`HD~kZ%&~la<X(XzV$~|@Bo@Sal$JsZ<urM0{
znH<2-T#B?i8ndFCAbmYOpL{DFw6;E(5KO*%{1f{mdPju021qM=>J|7bc|O8M6eTHa
z=)eX@ZQq|SU6r0`L>85q6It7pHXL3<jWkxYnwuy_d*v?FsU#m<1Y$l`9Oi0_ktA5V
zg7y^l#10=@#@)i}A|2F6q<d%GxsK|qeexv0tT%mQFHO%KBGet^mMD~#lN@NSR6Dpy
z-;<q|L!D|PL)xar@ZuZk>w*9;x#=(Q{89&ao?qPK$e|DwI?-#mVL$bkD1)5mY1<wZ
zw|pOt)%zJu{bigfPZMy`D$m40l}DRQ_c!nJ&&Kk6v_I?0Uy=mWumm_;fL69pMvfG%
z%&ogB;ZC?MtSU`N^~>0i4@XxnsXbUWsIl&OPxT|aJGI0}Eq%LwB7Qczjhp7Rqe2?s
z0(Wu(Z$Ws4wvK1hW>gH=D-;st_J-(ya|Kh4n3^k#93jyC{z1PF_f2_@w%jmWiKA6R
zFo@;ODW;_;VANBH>&$WER*F;2A+6WchUFGBDR78|M(g5JX%6>?_%<HDfiV(Laomrn
z=IjT4sb<bC7rBLKcKk%2tS$#3R3ZDN+Jc6@aVN^5H{qKrJTS&2m*(7H<W5|~a$$Rs
z@>wc;7TB~t?sva$9#Dzjomx$YJ;H8u0SyJExU3w~1B@&hkVrCp0EuLGr}-0*NWKb7
zXN|GIW9c50A=VOm>j&NGU%l2ePqsRYH=YG|mM0ceOJLVB9%6i5C5Oo7_9^I6F#ia%
z&V?o16?)vxqE@0Aq4lU*uXZ|**Pe+^m+#{#Bf9ZpUvsC{5*wflse}1{(68fJm5=Bh
z?tjYP<WaOH-sNJ)_p8p(P3x^;)mT<?zD0_ST~z<BLlooL-jUKB`&`2)P@Q)EQJuc<
zN5VC!!b18gkjk2E#U6kyF{&@&7w7Z?k1gH!qD<)iLUmnqDjjhs{HK!L;aXj%w`N-b
z`>p}$PSKj7Z;O7Oh|o_-Gtsz6+jZ%g+w5Tr4*m9^6;&`{ya^Z1&KOaq>yND#=bJX(
z<Q+6hT&&YcUlvFvR^bcxAhodI4kt}%GV~N*pJa|oOW}Qc2UbUdaQ*mfM?EfnT`Uqm
zB_{mM!_z6Dvfky*g+~iD?C1tV1=;KZIvNIDB<PKdl8SL#ImN0=)$1v>+dxC|)XXHV
zjk9w@eV1YIDXGf0?gx40nY%E&Tba)WRWx59iqgQa+Cf9Khi8aV4mRUpw$5_&W9Nf|
zj}8B7<Fef=rBhrBHHrMot{kt#s`Wl<EpurbDFq*MJ?l4ygZHe7-eu%H<L^tXQvCiP
z>{l9(`$hLhMbZ<2@S2;cTx3DTXu4ubBelETTmY%H{v_A_NPeCCuOY{u5|jWZX@3K+
zf4^r2UO<$!No3y5$clyEu4Q=Xe0HM;HeNO1{xb(d)s@uUIYH-TnSQwZ#?ps+qa#H|
z9AK}4T1enCIBN{%roao*tPqKXjMLHri7Z$TEM;TA-*b(SEiun=E2H8F#Y*NJwAgu`
zPV^b7dB`)OsOh~Nf%|CZo;y0dsvjbb&-96t=DRRFc@`LJhiGI4l%dB^lczmb-|4{M
zJl+D!27~<<pm_b!e{s`0Ng=SS#6yEpzNmjyMXpz7GRs5+kL^B<=rUV4G-rkQf^lB7
zGcWnchimZH6>mO9;%?dDAV|v<*kGhZfyI70SA4^m%^)A6I#I?Svjd1__gGF!7i6F+
zd1a)E`E9mt=eW4VxQrlK8pL-?KS_3ji_L0&$)0&Sg~R2o80SgFhMS9@FQb2_(}H*b
zKy8ck>}6rtL85CDJ%r9K8dh&%vo1AMC_7MHcrgxOExij`x;%LvDZ!AzZ<Iq<aP`v%
z5%-mjOCsb++ehfhXzcALeXke;Ah?9&h8`w$LwjnvDM50UiLc5s&n(D8GSc##NLKnW
z($0|@ara%pg1uMv>ByNykB9K@lQb2y^Sa%SVK_{u1-@t*A;ls$HScTJaM1;`e*&7B
z@yQzHo7~Ue*-KKaW$XQVoY|(5+3a3-1JaozPTyS4|7e<raqmJV`|GXyyeGat+`nZF
zL|X(@upttUY>dvv*+<$md4ivBzQwO}K{T&3=zdC#+2{V3Gx^U8tAhk+h@#oXFS-B9
zZr(7Uv%edC)h8=<?QZdKmxRJ{Y4=U7osa7kAruUcxj%jU{G<v$XIx3@RFmln$mVD@
zON_%cQ>rpi+6i1Hg$f${Bj}@rIkxVvDk$xb*R%#XkTZKQQFcq*jhQpFCyBzPPnPg`
zNz1|G3zTVH$&=WWNzTNBdp>jXlwo{25X?2xW>s`dlG?PFP|k=v%wo8#wCuZnLi5@L
z4n>n?IM4k8_Nfsd*y$tm?7l{D(gAjvAynwizcH2KQ=0|bUtJt+)f!r!#0hXUUV-F!
zbg(V(zHe!<B$Tb+I2bJ=np&>8OYRg~^%UrLK3RqOUfTMKuLr7L-iyYGkTn+byF9-q
z`9#ekKuCVPypH`y)J?xFBAB4~9Wh&8nQ5xr!|Y-P-R#7I^N5+;uf_owYp?BeV%L+&
zcbOjRCK<wBiS%EE%cqf&FNWT}UfW%c#ZY{I$>JK}dR?bzfIeEP!l!-vQ+10>ztVdR
zjG6WD2jagoFM1l1DaDD{$3yL4EIG5#G--@TEpNtA39G9EGxjk?BD-hBxc+uScwym{
zQpJ=NvAV^rmAtT;I2s*$Td@Hv(;Y4QT3Zv2D*W*~5hIoO@D2LUaTr5-u-p#DvRH|y
zMw7SbkfO!T_hC9OyDTg!Ah?wDvvQZ^5L}r$kxhC!opeCG{y|x*Aw<2ByP~()!c_vc
zqiSw8cHAAV{E;Qhu`JYK^^Pug8U%v0ye#*$;i7_>qodz4w-h|wAr^>fTvlU}=S>%)
zum`;wln?k}iECdo8*9UuFq@-<#tS`aUHVco_+?Uq_xEEbuTP|)#m{iOO*f??EGLNB
z3^+M{bM!#`B_u5KjR@i0{=|#GFTj99mfE?Gufy@;e8}OW(&c5@-Vb?*ZJkq~Lv;X)
zJl8$}R6cdD>%t+CYFykVzcX@+-k5*7ozuNZs=L;xR;UYS_^Ryw<AN+CN#D|qK|TXu
z_fAr3xsFb|rzmv$I?4E5jDz@3x@0LX*;!nBoH7@|^%0qt_iMmFP`&!K)QF|zuE)hF
zD%AZ$0!YD&3HvGc8shn{ZwGMr`y;7jHsOK#8JLJ4_!YfmdE4uG`poI{g9gj>GhcwG
zkT%bQNGl1ub_Y-E92@Z(pK%JSwhI?vSk-_uWg6NizSm0+fNYx9agsnV#3Uou6r}$h
zlY0^eQHIY5!6c>+gxv_ly(hB}@kRmid4%L8Pnb9q>~Y08YlK|{O>$y$Bm}`?gWT4!
zYoXV)n)O5%yz=1TB{sU4xLWc!#L|Ib<OWB(B#5SKs1>OLo3r%pav=fw>6pZ-fk=hJ
zLR_j@uubtA7^PW3`RS32fU~4Tzs13qh^|zdM!<OI1*p8m!`jFfPDjhIG`N4^5SIq(
z;{^-$i7+Y0y*AuBk2kC>z|)Pyb^bysA7A#ojU%UER(g`r$8A~#$M~LT)vtJ-u`STG
zYXSO+n|Tylbv)<_v3k8FHd?;easG8BFN>dyn0wMAe>g2?B4uQ{_vrH-mv3&}2t{Jc
zXhQ;vwXkYS@E8@{GF3!WDXG8)BMY6KbQqOre<IB`4T48-y?c|qRE2?B-m*fptpV&%
zZgvC;xmMAc*MNufV`A<zV>#x=EusoT>qQg@24{n|`t-}kMB#%F@$US%_VO>-RG1v^
zuCI}1kuJ(4F%|m;Vfa;G!WKI2AO{s<Uj_5IZ%)!X=bOmWJQec>Zr1CG3<3MGYzSjq
z=g{v59Nd_?r=NKweHe4h-khs&9c0!idp@zM&%Xd$iG@-b9={#lL>ke)9xf7E6<V==
zO*EH+6dF&eXQX2k(7g)K&$-v#UG&4IF@T!<jxaqF;hlw8`asUo%A2o;@owZQJV<EI
zBd_H?TG8f>Q{T6CmV*_h>7JLnz3$2WB*3I*hZ3MwT&le|eG@!GA0ayYY31sd#vO7e
zA*6y^Yy?L_M9_l4)c);_h_nSc`~w;M_rswAwAn;6K$ii>2Xl72*hf)suq?QWhWwbw
ztDw{ibzI1&g|Fy%k-#l@+aQ67@hmmn{O%VIe*p3|tCumlzux506y^0p+wY#&>lX#T
z=Wq%8d-e%lKlUds8~=!~XlP4O`=PyRA=e|L<lhrH^V{qxc5Fmuq%8AWk=F7K7?F4G
zariDJMt)8G{tC;Epv;NI#qTu06pF`>R4+OhW86yd_20R%XE4V?Q*mfh7gTYycbp<-
zotqYQ7G^fu>=AU{HGbLrm}0^X_*aDbDEGI^_8#ruDJrfAkFuZ?&X4U-|0Pkm!fWv9
zBo7%aRIsi;B%`Mzplx<}&&M=6d^5!9W?i<g43(X&Z*RH`mFB6$(?zZ|KW#&m&N#y-
zuZa-;l_&USS-xlzcy$a*EzT~JTbiG7mYnr$H}W<|Geqy|nUv&5WneTtI)r$M5I@2Y
zAtD*)j5*ia#e}+not$sJK0jIPo)`Kq-7vKDHJT@PJr{|;ME2{~uD^GV9m86<SJ`>(
z`+1=~&4cSroBtt*k7NPD7HYL;683+`W(&;w4{gZnI5c~5lGNls8@ZKwTvzy9Op1kA
z!K*m7Qgw%oYfj2NjLpx0oZw-(BK(oqc(^UE;rpYE68*%yBNjAWxG$hLlDQe61tJVK
zT^13<m6c}`F!aiaX<UwqO!I4wxU?0O=9m%thCjB-Fr`T++5Igz-ja~1$|p9k>)}CF
z=nHpjuK`a`0<q-aohCdr(j#_GM6<t#?DTQ)eOxCr1!ljW`Y?05Z($bfQ!W<e#W%6T
ztSc%lt$7;pr}Id(#`yx;X^@Vd{@ZxO&M6HH3wY_~o;f5~7VP*pGF%9Sz~t9)KnBOP
zM012*D36}&zV>0P_-U{S`9m%0L>zqC*!B15_?ABn2;ey~Jlz=P?M%@kEj$545oqvq
zaNq9Mg=J69pFz#d(FBn<5?+bn-^(KkWRGAOvcs);*SF{-BvHUU352K^xdG|8D@B~r
z@fPKTH@KLoVe5tSLk_FA7w>*HMk}ZIeW$vqK1cvNFj(J5ec~uZ;=7Ur3LQ|y3*_C$
zT^Pa|ziqcWAV)+)f^Pg>ROU49Bb;8=H)w#p=(hIN6TNdysYT9VSUIBRgD#S_L=$6F
zrXs{#%_u-T0qdro{=&q*kF%z&4}69~d-{jE#`ne=6-jf2K_Kk=3U3L-Lo=m6!h<(5
z2;<qBN@qf+f!2r$FLquYELM$ZGwvLH;d|XvrQwBb16a26D#?oKApF^VKMJbb8~uP4
z8av3hdR)vsCiV2xv-ID_iGj`Rn9^l&i@Wsrh~cIR%>aJVwA0b`j!7z0nuXj7R4vie
zJej7h(razuI^L0GeJ)=&dfut?Ib1sB?K{vJaykH#wTr=8wNvagV#?=8T~nej>=FrA
zgknZ^L6A|_B<P0}k}@>n^D*=8rMp6<``>5zv`x+OCR?!)PkXWp#Vx1BfYq+=LTLAk
zqQtoD7;9Hw=(A=Wnhb&i8_^q(3HBIj-e_R5nYb%Q0z5d8Uf9GW5(-lTmXfQ!6IuU~
zy8jW8$EAVk#qP~7r@tHF^DiI^Oo<(_1l{PWb>5vGV}k8A>=%nmEydSlwM^fN!Pf70
zClh-Oo{8g|5DHLepO%%iGomlh&l!W(TuTwso`cr6iyZ%Q;d&r3K+m+$g)hr|?%t~j
zVS((blV06Hi>;UT=<Q<E(dz}KH2gqwgnyxQIy9;ookg(iCelDhXg!o|r<}fJ^BJ5k
zHi`6Da)q9!j2mAkem<=g6Hd&^6}sZNN_rgB^G2GI{AG5cUYz6OySzFNMzoCM@bC?T
zTFUS_J;jGQeP1ZUu64uZ0~vIdvQvo03H-o4^FX^XcA*TSc%sP*Xumjf%-0v}iqT<k
z!~hkkaw|PL<H?(-3O-E9g;#E=FoXAv-zp7>9lqCPytPB^ask(Iba1_@XP|9j4;9Qf
zPOy{CuO4Bo)?alnp_oC>J%m3_Y%Xn1zLC)`F@Ad`^u9I=??gBL?#&!3>+FVa(N9Kn
zFki^0_Jo@(8cbDIeIt)~O$6&i)T^VdcS$icCHD;IcbI+=A#Qd-Hx7@PF}>*BTI{s@
zQABF>{ZhRB`-1z>4kE3hN##`O!(;Wvm95`u4Za7^H%&Cd==4<QWlLWbD=SW}CApOo
zdVpqS+ABAB%d22kOlhGAl=ufv=cIj3a__?Q%~i|zxDE*<^}8{DNtSC_-UH+GWW6ii
zf;$;RVn!Cd<3m#Wu!z4{0O-bPI$*DoZ!`{WhrR7|4RBk06&0uYIbf2A1t-gSYtjnU
z-ciD{TL@BrqC?y^Cq;F?2>pPrl+1%aB^t9?s5>{zzySOb>nE}hbHfnHgj(As_2~BV
zf<2+VxVe3!zmDfW?JCCl!~>@B|Cpl#!rOK9iX@yM?(SXPM1{J-sh^rJf~*soQe%mU
zV{>2@#pmpDh>$Q{{}epLk+>*2Npv$1m0JlsOd4es=g#y`4e~GUYy)c;eV5EfT~_~!
z6St(LhgWjoABaHnU{%rcTG+WZ7V&1T5TCrATag!Y_naQ-%lEsDxZXk`qyeSNl(qX&
z&OfLW6xVAzu@D#06^(i)zB??4+-j?;lU$FYtvmDApJ6c`D8!xR2tl7oXq=kgfj@P5
ztPgBPjC+G7fkGJsw~c&2yK3XCLS%;QL-FOfYg|lSNza_?P)9?Fjh$&N0Qo2&me$%z
zqe6EIbE7L9y0u&qtB`#S=hA8KizVxKUzJ9pbRfP(%PCeCKgPO6n#L3*OMo4PUBLPr
zEPRE%YqbavkJIQjz-+Ei@+u*w>$ee<>u-=D2ckux=Z|^(aU^s-jkL#d;TGD{ywhN{
zPq<U~&`k@x3Xj^|`meW0*DS`A6fC1rie@i+O`=NTG4o8Jtpk9(f5P;j9w@VNl*=-D
z?3Qv#-qHPiYW&rb41;JJivRPwe0EU(dPW<$Pa+J-bWWf(72*6%#BNuZL?DpSt?0w2
z@I*Sl@h>!sr(Ue)Q+vN~?^2+{qq{OnxWl?%*(9nR;;BXWc`TujZmEx#GR^+DR06&c
z7s^4*<;cug5z`USf$%(GgmVno(<}F)?Gr$2>@%xsPv*i2JAvR{#2$!W+|Gf<{sZ7N
zUaV+n!bUg5*s{ExiJUe;KQuCer>-#SwuMvK9#5VOQrLOLCXZ&&6D)+A5Ednp@PVkr
zKHGL*O0^oVHs2_6;hy4J;*AO234sqxcszy?&tsA6Ll<`iQOe^8AZ)Wq3~s3b`bs@r
zN#LqjE<Kc>4T0d+LMx~}>sJcWOruR0A%P0_%!43EOA*BKns*aeFoy482i_ULb(3s4
zK>0Jey{djtBDgDUGBCw`0Ik4@l2~zOH>sM!O`(GMWZnQ*8^w{2xmOV7UG^=(IWfQG
zxJ$!VKoX2gGR$Qc&^7jPCzD;fYSv0V>l*Hgwm7{PTRB4QX|TR{&zN2y)N)qw#(HZS
zfMFzdmS#4X_g65+6;2!Lol<DI0Q=?r7Ky_5U3$W9KA#BD>neiaTjh5#Ttl|*^FB86
zaTV*Ez_7l%h=q0^cF8D^M}e~298WGwj;u}r&iKTS;X-nxc-U)TV#r!~O@FbJeI34c
zeEvVmKn<X#D6^dUXEW}U-an!b7~%4{<g`>n$+UH{*IsAcHN<pegELwUw8*o?)r3-a
zTxcz0Nuk6&s+f+gUH~rgOyAdrp-ul6%1MGI>=;Vztc+TJ#5r?B6)jDLdw;8vo!XJC
zsQuEvs2q5>rgA!>s5YBmh90v*l8gU}eBZAHi1PSb*zq8}jG#N*$Z@?`i_iW&x|3)T
zy(B;xPnUxK6D+4gPSy&z_XQ!J&g#vaYS(~%V-QlrWI4l~5asmnKq=~DtQnZnpKsO*
zCAHDOOSFYYjWJx9KCiwn8W#jB*Zn-eGI+8i9aB6Pg-9F3T+0-zz0(QF4o?{F6YIs4
zTg!V=h+`s@OqN^HZX2I*`$cYyWL@Wn@?yW#=L_OQD=ByKi-&_jccEvJ*aqXSj%|5+
zOCUba6Eowb>{5?*E?e=yh((iDJ}`in;t2*Mhx4-eXtA4w*7e+Lp+cLUV1J>_$uO(p
z**<VTk>9(snCRFGkm6t*IM`i!9q!~*tnQBH-d|&Y-7l^W6qZXok6*yKgu8*AJW*_s
zh09wzDb1m=w$_krr+XFn5e|yzxjba#YqkfI8*imMbx@jI9E@yvgR2voiqw^1ZldQ3
z(q3ja8f-pNSSK(7!f9?^#whH#@j#)Q*u*kH%9cqm;<jQQ*Bb>v51Y)dm)pQq)4oLU
zEoR{8tFjS%Q~IWT2&5EmqTZdD$>a&CnT;r9gza0Flq%SN)T=RpnFyhrO6y*vBV$99
z+whQr#U0@M!db<dcR^X9iZkX##fgmR*J6e2`%7IWZ(Xmn?hf)g0XIJ;@qS=2Lx?`>
z6abuPFEQRs>v3MC>DP>TARQy0CgY|_SNORXSQjF#bQ+q~%fgzL4t1&SwW0(uWkXtO
z`6E=zxg>G_;}|6I2hv_P$g``w{}3LajCCZz%^JSDBzlQ~9ty_giuPl+nn?=fAF8M8
ztoU7X23Kb?8ga{-k#mKUJ!369==gAB{OtvuBqKXwtzoMIUPz|kU-8md9&;56S?<rG
zGgs+DKC3vS22MqzQ4u0^TEE!(g<T<O(Fq`m$GC8!r9YBk>}^F^Wr>xjO_C{;XdRo;
z6xn;SJZH|PM;NwuB8#1|5H|_(LN94r7+z3^2}7UJPi|E7@MK))L=g-|!9F!nO_arH
z7I>B{m+M#QHS~5;<stRJC({~4BuTLV8w)dwjiQh0s>q2Wf2?O$MN1v-w73j3%^G4G
zx0oOWN9Ilw%G&921PhY00M77nwu^ry(L#(c>KPVZ*>yMrOtzUxnm__f=hsCpfc9z6
zOXv}-C&=oZ6!aZEpH|5bOZO{&+9qcs+ZU?J=I1h%ceA6s^J<p9rdg$_n(y!;8TI|y
zJC7KAWE6wT6lp97<&kT+X-W(h@tHCKB8$G`t16f9wpu`3<ZpWDMUbklz2>!9e#Zth
zJge3DC4@iX13C)chu(^wj8@h~G&d;i%UmOR08^okm9YS`a2_D%rjN)s_$E7$@OiKA
zHlpViBAy2+Yxz?mtlrZf()}^^n)S@P&93h3A3YYy2131=5))xnL*GrG#K)NG`~Hst
zl?_#(YE)k6{TV+0`Nv&8ba=N3F-8$DNBvJMUcN!HBm{`8^TGc5#36K7$DI;cX2cGl
zM&791{KuGsp5g?%g?gdh1RYjdriiVAZ}S{#N`WJA*~)pdO~khceSvwpAH!*}UygHu
zmnimHIV=h1^aAki*r9B&&-CYVZ3y3!#VG&w2Kl!5=VROF%Ti5yb@~K)bz;E(Lr)6h
zuFO{g>1kaDuQ^?qdTxjbuD|pAghD*#nS4(7*5c^WXprVT=cj=zNkrwxipC^RZKAzq
zRtEUkVzv(LiNq?&FeSrioF(p~g`kRNFXM9Bg)N~jC0e3ks&`U4Ld+LGv1;Q)u}>nd
zvpGMqD}@6?6%D?d&Qz{?o>y`wLOi6buA8%nl$UB{8%Ww#Y7Oh9CAzr3E*@C}uh14b
z{qpXGjH_$nE(E9B5sC&Scru}o<H$din-?Ge@5A3cY!+Uyzay=E7fpbFQJ<{PC`R(M
zu!|3lA-kV{52Y~45WRoEGosX`)O39QpAta6VU%pBnE3GcpXTI$wK4zE<V1xh7}Q_n
zQ^!r46fv!w)<Mb=9|_>?4~L9ieVG0Xt#Xt|Q#RA-Z0)ADxBu5pfPY>AsG|aPj$HG+
z`qTXXZ}5SC|55O_`xpTu4V$i(U*U+@R*MaA@;z<i6Qr{Pd!2#PHW(!eG}estSEv9{
z`twl3TCe=L%f?^1@W173#HiQFc>l=Z-!X^({F#4Psc#eit_Xr1ysjq~p_ZG89?TE~
zfax*aEI2;|tIOVI`PL!SCPBKf@e&{=#edr(@p<bV4?BL!>Ax}riX$3ydIxD9eTRkv
z?51^%|MEa)sFib_xH>Vz|It$ZFCPIbHX5!!2*Xy#V2VMJ3GiQK&FJ_qK*lz3<0Hzz
zssR@A&LaA~hTE9?n*J^e{{7J_!@z^;VSdm3KRu?u%a?$zfaX>rDQcAQ?1<U2yd8!Q
zvM~F8*QS5GWB>f<?kb?Hz+`G)zoVT0|NKAyKmW!zkJ8gbmZB5>zb}*j@;v_aXw47h
zK~vyC`H=Y6zn)I>Erp-LO(*Kd2c>>AhQ}@BA>ecLe!b*uX<-8lm`SrJ1E!_Lq|4tX
z%YXg4|JSo=7NG~3y%S_i>Q5gu>dCb-I0C5=@U_=4{5HLVJ<W8R`LN77Fg<Qf(2yJ>
z`yS3!LZNNsb5LR<c(MS_0XzUKF~$kx{xr&n^1x;6y$`xLoZtnY!4)pKzN5?!Cf_My
zc7>lPr+SL8gW>WtDioqD_I<T<cMULOtAR#aBaGk80)4Crc>i1ts20s1dfx7>rUec%
zM-U*~X;GGnRzSR-ME#R{m>m`O?)FhXfB(<X#(%jEEpLExK3ye$!}D*jF(B5x&GdRI
zvXo-hqQaIzigJ}ZscoS0u}yG$_8p;gS7pF5_5yg6-n-Zp&`I}+gQUN8%kS!J8yMmT
z!w{{p<@9)xz?MFX`sUqGM_I73SSC*oIPqB*&Q@|~jyr@~y17dABiIj`F^E#ckHOnP
z;R{;t!Ca;F0w@6+M>pOH4GSGf{D8{kw=LhJ&$HXDL_`2#XvCclEb&{tr(DeQK*3Y#
zw5s&#ajQ24C^Xozs4urerCtDM6r=Ynz7cK@Z__zn>8T(|ShLRXOVdxR5tG~nJ^gHi
zXr&}1p$&dx|IfQMg#&fi6eLzIfB);}QqXa|7ntbyTl!HpBS-K7N9%<%Jo5&5>zel#
zU`pZZl!yTXj_8QxDVlaEYVk9ZD#{8pyDW9(sbl(*1-qr<FJt>GsQfmXtu6u%iQg2X
zR)RTBdoEYeoy79XZhDPenQg&;Zg7@VOGuYPSZdT_-vlszL1&RxaeCs!R-;!o_|uL-
zCqULInUt+PP>-RVCY&@`mLp%`^4NO~VYFP<-?*Iq+&s(1bNLA0YHdB$YPfi8!*@$p
z&z2bFTQ38r%!ub>$LtkQiBh1RBHz=Y!e(&;;4)urZ2E<twxRTG;}`zH1xP6g%{n4E
z@F1vVO?+Pazx+o(8a4THyO$&U9`y&ID4zZ)>p*33%_`kVQMxy*=25;C$&<J03(u!X
z&g~o3$F~g8D_(9cmcEVf#|4|u|Cg`kV`*vh<{gv7A4<efG}PBG7ED^tg*eCW>cS|$
zD|hVX$Ismag1yiXB03&4FzrRz09t_RMz0zIorR%Gw;6F7Xcz<hr&G3;kdMpb>N$XX
zzz6dn2UOMDiZb|u?Bur@6_kZjJP_L|y6f)V;&o^YQz+|Z3WP7)X2>Z2Z1qGTy>8lT
zO$DMuR80F!UCfTr^&{Ky)DZ0AkN~`pOn$<r!T;Nf`2EWEEzoa34i~>Ynvs{r{Nw92
zBR!h2tE~Y;X5r<t4C};ZpjxHL^gUe}$UtvK0ToPuLvQT>y;=M0+pp0mjS)47h-8%8
z(P0(Wa9QS0_s_`{nVD?#X-JZ9jsNv=!=%4qGDG063s*^gzkkpG9_H<Sze_=JP<v-`
z?tRygx(<+-h!o4$)-AI5+zS)(hEyVs#$RE?KUPncvxw;Y&Yh0INK5wW@}d<$7$~r>
z-&}D0Q9pLinW!ZN3{H9LuP)DMzq}N>3;ebwT`bA_K3Z0T1*(g{9D6-4s5c293%IsR
zGj~BH8w{q6Z1o4jdbF>it$aYiJ=-aD>6BvkX{-e``z=T@4flG_Y4H6Ky7VBSA80&X
zPJ-?Tyg6_u#W)bJ1uze`H2W!9QQ2QTa4(wVykxzGYKOQizE%x(l3oX={RNjX2{>~-
zQvkfM71(~MK_`T^^z)$43PnFt6-z&p9WOP&$AZo4KmjZGj5X6=>d2}GSn4QHUhrlI
ziC|GfzYEvucOcP1+C4!TZ=bjEyks4TZt-T!u;K)cxZ*vp*<kX?Pvs8R4*5ayACh&x
zz$VGMa;n;7<h`Q+D3omVypf$?T`^OZkxMVXl8$+tf+!7<E#+tbPSD7i`y>Gl!RMzI
zwrRC$Vku|u91U)qUjRVN&WR`Z#RcCjnnv5fhDjs0;zF8nX+7ms)oVY^qVKSt@4uDW
zxPS5Rf%>9Pq2>26T>pKtvz&YDB(PDIcfo9_=JABCRz$aK9xEI$O(bbo>m7U+Z_=E}
zs^Z!NPpmydY9bMUr8uGOw>yrnh(`UmqTE-^EpriQ#zmK$-(Q3Xe<X(NjDXXzqEumY
z2~_@Dd(<;89kzOpzC?RQ&jbq|vCZHwgBgCs?BNd=jJcnQnHg1CpDcTT^>ek6izW@^
zE<TO<7v~Owmz+jQ?xiLtheMyUGUyaerWBW-G}wJ`odOaJm`*i23gOfwXZpP|_;&2k
z?Q@<r`Yk5GueTnwuQmev!}hP-80_h9?8iq8av{h1KlOD-Ml&sv5G_DmxxB^oiqqy9
z&7#+-jLKH-G3#{oFA^KE4a>ek!s+DEP9rCus=;3#m94D08FW8R3+5R?m;AqcHDvi?
z`ZxFy{J{8qyaendLRy-><tl|f-)a!{24DdmUm5+jPHoBV!?h7!`r<o)Map?&t_~II
zBWc7wF~el@{KG=LwDp}7G#tW^g}Sq~z##JoSYpNW<^7vq#?*#YLD30lM2Do9ne2~2
zOY465O|8}fqALAuaCf2H<O1v9r^!nD>FPlS&<eC5bcUXbWaMw?C9@51<&LF6+r%am
zpg6mNqfVX|ZMeUJYy&7HUiFk)TJ@{SPLx;6FHCVVNP7Ko)9E7*AcAelPc~fD{~7UQ
z9dMc)iH#dCA~84;XS+JxI#P_<m`B^d{l8bXS+|z-vc1BU9$kSQ`+;Ow>kZ;-ZZBek
zE0MShPxB#)aQ9!?p{K_Wgb$^~0O|==9kIU0Y3Fx1W;F{M5d>h<Sh`>F+v}q@==N7|
z=r+q^06-*vaD$r$#iH?PY^I|UMR_;@aEH9wVRSF%kWUGU7klZF;X3?|(_b3y2z*8%
zAiPa%<vWFmfZ%LO&Tu|Q-n7jx<MH6#z%qcfw2U!4G&U5deuA-=b`mVIxsagY7GU*0
z$P7eChYQg*=P+o^CoO<^GMHrL%&<8a1#}ZCD3t&0&Vw*v;<ZKm7W{JzpZ5>rA+e+_
zVNqLkYdl$kqcE(q7mI+nW%Q#%x4bClRP&v~S8AIdTS9AR#1XU4x0jDczTXr3aF;Ix
z_j#7jp_8V^>!Zj=68V&9#>AKJGe150%!&s=9g9b({%H~ZN~!~2hX#EC+s5s+`&FLN
z>@PhDxlY^8m+02`T})AKDp0^X=>Txn``h^|^G$$(A|cB@|MGYXOxAo525cG6;!D-2
z%?ZnS8~~iaVk9ra^5Ril)=Xwuh{eQrq+6VC_1K+p<>y=mJvToCTU~9Ig}5iL$p;IX
zc^Z7NFNDx^v#cN-qC;25{8!qm^p;mYgNRx{=CFB*?$<2ZM=dkdKdG`8>{9#^gu&FN
zqVf(4HNiFZ#xP%c`$eB<DQpwpFR$E60Ug-{G$#qb>G-MXYVLcA%jjVkF_ZVE<;1|c
z2dOnl<=Y=8aT^Jr6#j}}e}}Ta|KpSv4G(fQo$_Akk6UgY4NX$p9>lR(@X_F-Xtzr8
zOs2GSRxjya{8|QKg|G+yF%tj9uJZA~2}p+SP<wBG;mJh!9Q<hYN5c^BhH4Ikg|&2Q
zVTu93Po<Jq_eenD@@@|=QP^l0T*;hV-V_&h+^j|gmHbBA$3#%{5GqOUtcJBS0`gH}
z6jk$C;SyL+pSebt@ba$wYt1%!?<b1vZGyxfHQDCgc&R<cgLFnx&Z9Sj1L=G17sn4(
z!@ERFVf_>d!qXm1+fEtCTk<n9>s6o?j(wWJJ=!9JL$feP)n+KK&%8^g*$*|D^1HgA
zWr*Vs#iIN2F*xQF*NZ|U=inQRaj`l6my5SY2tGPbn<(giNO4}9p>V*A2>Zn@51Esb
zI@D7C`N|GwE)|5o4kiNPl^7?5p9(XS4>}Tl#fN=)q7jGX#1yE_D0m6zJI>|!7p&;y
z;$d3yo*-^!upL%`(M#=5eyV^No1ZQ~X;>=)ympTdloPvzOkU7O7pmeug>usSBgt#N
z1eWWA9!UmPiyWh&DS5BSc#EqI*ciUz#yHFnuF(61cO7c_kaLubF~*Hl-sc!yUVT(B
zP=)b$5q}+6O19p|i6%gx3}m0~J|~tVu8n=7>`33u`2TVDo>5JHTib7FQdA&Dg3<y(
z=~WPsE+m2!kt$UJD1<Ij1f<u{M3k<829VyX^bR7$pddv+K<Nlbuf8k)viE+U{p@|t
z`Fh51I5Hq33Axw3)|}V;&1({Pp+<Uqg(G@e$2p&iZpKtDJvQDr-U}cO4RKfl^6vU|
ze<xLnCv%c=WYYq2-1`Bp8kJKwDg4p<`L?UM^B(6Qz%u2;F=&s!-3vvTB{E>B4CPz}
z*NTLLm-L%`ix6}o)usRX&&SF0xcKz}svMkAXmN8G3mD4O*!f`1$&J~p@0Eq%JZs>N
z)oEW5H2QKL>lzvE6n{Lpee&|N3(Rw6qfg*J9vdfTL|0w+leOyvTke|OnY9b`WS8Wn
z?MFm*>yES_S`fqEdt|3p+&<@M3H1*QhpowjL;Wjd<}Sqhx?fEzB7NPrr|y(Xpd0J&
z7Q_vDQ;~S9FIsc7jZ}UehnMYy*LsOQYDD|AV>&aOB(8>@fS4}^ZtK5isZZ$ryK8bQ
zuCSp;$sKImd^1lL?*0}2{o|hb=Y!XDu#(?OI_STj@j4MIa^r@0fP(l9jS3Y9RwE|_
zYesk|p8A2*t66`vKm?a<1eZRJ%L2zW$WzxZ_w^<$2z|K6yqaNC!Nd1pepJ+4LHvyp
z(F{8y{8;K$HP(h8$F6*s1fF3RwmC#Q(lrOH(kS(%C#Gwt;Thl`x?|~KKT4Z`ym0gI
z<M^~mAz5_Q8-3H%)bf~x?;oNPTE*pn$%*efpWqmHa$k&bdm~oyoV=qJ7EgTxPtErs
z?}v<~%}kyL(D~4jUv=EmHm;g0AB_(9!Zk84u3CDJdb;-mr4-v*i&ZRDd07@RZ<N`M
zwJ62nrJWLek_PT+yNw5%`8H4|=G}WCZywu;4<G&&mVu^;;Nn?7z-rb-Q1={N<Whgg
zmS9!gmVv>>QcW~!+`sG2b`MS=ngS6s3jTm&wK_d(H<N?aYIp^_*ODYZ&cCCEBX{pd
zhWlhtpUFT|L~)(zPU9Pwa8?rMa00fhlcE8!o8km3ofx)jj<2R-R#WZ>4DQ-b*Q0w~
z{gPzUrW}Wgj8>H!W&2jDmmWv1AKS>3Oxt!K@6<ZzOQ*|E=A>g2n|iAAhk;HGFYWcY
zF<E@Ynz-#?!5@GoEs)eKwf4>nr^-O@udGYL4m&{<H0%=|PBlU4UX$Zb24@}ql3=8l
zta3ets$}VM;a+}PeW>2*!)RFH-UM<E8~t|3#}<Uvu1$6p%UHi0=%$_IT1fqDGDPH8
zngzZ&&&_haetrZ|fP$smjC5qt@V;A{_ISA7N!WsQ;{FSb8E|s4DL-W~^ir-+2_<OT
zDo<`={fvy0^D79!-?Q9HwLW`PSisP|F=e-Hl>c6S|Gv$?);z~P6dv(+wxJ-*>HoSL
zUsCg}Lpw|VJcYs53vBEUzMbAd@tctz=L8kaIyPCp&+;QQ1f$I>76cxA;b^{L5Qd`{
zwWw?j)01m9^6Ixubd{Wp*n0|)s4@Ixe3F6q4MsCpk2}$1g%xjZRx)T_CSgZ3mHz;J
zD?2d+&)8`6x(mTz$0%}ScJ{u_L0Oozz43kK-wO^}%RT)Y)024a+B;>>*M&^&3F{`f
zR^?yrG>}i5dmY1A#0=RAbqn{vZ=fn9H57bYy|hwFg(l$7=@pJ6DTsT6yir7ZnVkS%
z`sp(QUS^b{Ry&HpA!L-a1$_19X~-?~=Ww2DW&#JPr_V#F_4o^Mw0T3L3mf_&F1H@K
zcXNcf((`9m860$D3CK^5*-d>Tdilo-6<Vn#cY>!JhIT)-i?ex7#X0y&Q~0lL-5r&4
ztx7#2C*4(@X>gUV7x2m{4q3d=1b=_c=(7T&l`+$|mE1B1@(I8+1{Lp9WsPY?&_k+b
zioJd@4WC)lKXQb+R3uBxuib!jA6OE2H?vg_3*INdWsnKQ+j7a|{y6IW+2{31CA#KT
z892ANUCrTzddXOMeDmCXMxgF`alNrY`xn=4zeuJSeS7@SzSc0u>i?30|DBg_8Ir1!
z*#8zT{`=~0NswmhRq%MGsS0K>u1+=iv0r@|{zWGBqRnG&kgYw_)UZX^w}Eb(x!jLM
zjR&?(Z;gK6%CI8M!VngEw0}L1hzH}C$tqN=ha7ijxqR2Ris3#*b8d^5Y)r%G5chWK
zU5f^b!@)eQVU=-PgeMs<!{Y9F9rbBmi(3qb0K=k)j5Ao=iRfZ-Q0h=!l2P$O5B9~{
zN_@bTmd1QZYm687&K>D;*KASqYS0xpmzsTpYr-^3w`@_$ch%BeBsMeqw#FilPK2H|
z6vTx^qaQw0m9lD|f3Xt-z?r_~f{`#kO+OpN(XuLux(|G{Q%B{+Wc<eWdyzA<xg@vj
z)ctPVtp^;X+s@wG{_M|bnRAyMtP4p8KsMU#Y2k8ykm`lwvx<sLx*|P+bI-FOmo<bJ
zeab27vSNB4YmYy-c<I`05*B@>D|xE-(DnhbQxU%3Kd(}0H|E3Qcf-qZ4r$}fEcjmT
zR)9*+jB3c%m&~Gqs{_WB`TB=ixf9XEcojZ<Xye302ZMNd{ZjSWn^^N@T-O_Ml+A+5
z$EHou`sryJf4cf!^Vzw?t(r=jlV)=5w6by3wl8n)t&QW%`)l_14B|}9?8blWs&Moh
z{Lq0|AEN3mezN8DH{Q4_>v{IWm7K?QhE2wwSqiIDD~rnIf35G*9nmg3RkqgohZhGO
zx8Q!~yQWg_1RaZp9xi|ac9jy}it6t9(x;BKi*M76lH$IP0&z34W(|pU=G*zb=egvd
z-T0#HrS^9ZaXVN=#?bx^R=)p~n}MJE5Ysaszo;RlKhrZgRsWbGc=BZf@@5PnNVVZR
zXG7Mbv-|}mfsr%<vc`)n%*3#~1d9?fP(Y0s8UkVM=zv(BcjZNrwA&JC2L;JH_ia$#
zVb4~9HsE@LX@(8;EMZL0!00oE5P!s3jvhG;AI*9z!ZUSpbL?4DHQ^XZ?oTBpC}pVU
ztXrkw-CIo+-hx+2kXwpjla&I6Z1^;2i8$@ORjfzvZKT>$MGX;DHCO1yB?UX*mw^r&
zx+*i~V5X?($1exJS-UmTwgyaXbWSTP2$TTrB$4`R^4(qyR7xuK6%dBP1OzQgwgy^|
zwC*18`DE*nYl<rEg`cS@QD@*jA_ftJJjF?U6eTOur|$v8U1wcS?V%NcBq+<JKx!be
z1gs6hrN%*r!h;E<w|sQ2wageV#6G);ucgS6b3dGzv=)FM|LU58x{O6T6(bGzjcWAH
zA06B!pI1{OayR^_CZJlxsh3yNhXC?M?k`hq=r{D@ONo_-#PZ|Fl#ITL>wa`=qy^XN
z8C~`F+Fk(8pXWY>+vt<$koEEbxy@~vCq-lK&}D2sX*04b!|<iQX8vXdlO-$TpX$^9
z_Wy!0P^eanXLBe2+<v00$#Dk14wj24)kv0F8Ie1NJ=%;k^c(j<uOCeqtGmG{<z2Nm
zOjaO~#PBkaX_ZUyC!FdpZUP~?YT~(YyM>Q$gxZ-T><tz@pv=*@wY{9E3l`naID~wS
zT?uGYa-1X$0pTMH#A#C#J^m{OjT6xo9DYt2hiEDY^vOAm2<)S6ILhuJ-o)>CTL=#H
zbmg`|HDS|o8oWHmPjEacpd1sHuCP~t4s`SNZ!L~bP&J1jpERw(@@C6g=-;J+3`noZ
zIB5#9_X^_^3GVn!EgTb;_f;wZjAt`i-oe<QOl5Q<Wdk+H)2Mu)3|9usJb%PD%K-0{
zO$-nniuYcDtIWV}KZD6{edk4PT}keb9tDR9meh1q@XxDzHt!!-mu&L=?Q@VpHdK9O
z571`LfsZ0@mj8n1H+Vpo5c)YIq^y5fXC1fzq7(7t{uTKb|GcCI*Zz4)Nfq#Xn~qv)
z-Tep)`cyc=L{RF!7`TMP1bq;&KUFT5XsM}?lK*9y@AH=s`==cHfBhdIr}}z;>zaPo
zUlA=aoOSglXiw^|yvY)XBu?lcg18yf&++I+zPf%Sbq7%!G-}6Vd4Kox|9Gr!Z~+t<
zmaw_LA8sdiix?e@cmU2Lj#b#A3;7-uc{FIF2OJz9u!vw3#5c?;cyM~f*J3RzDoXVR
z?n-GnG4oA~FNj9oxv+Xw%!uY&nX%w;Dx&&x$%8J9^$UWnmZPF-u;s1`#a0nM-(uj4
z+(vPEDb>ch9oSezgDDrJz|%pBh*2GG!yANH^A7ARG^+U{^B6Bqsdi|~J@En0aIJl2
z$<?!IZ7?>0<;CR>zn2k-#f!ba=L$NYB{d_L5@k)C5B<q#aqxFnQgHZ{dTN6V^hA^y
z5EMr@DlCyNiSo=rjq$-p*P6C9;B&!1d6wz(w8vdGkq!K2#rb9p+2KLl;W3<~Wa*d^
z;r(#P^In{3<ycH!bnzxC-?y4-N>gX~BHi$r$Zv+$0~Yq}niZeSh9)p|BKy>lD|a^~
zV8Y63^K6E9%Y6H^Un#@Pm0`Ds>YMm)r_hTM%?|YtA*)GL0!DeW5nn2nh_QrzX<I6b
z{qP$6s<O1rJ=1KBxFwen<j3Lq$>3i<(^q3|=iReLdNy?;A^Htmes~Z`cy>u!G#V_e
z)__bej&*J#`LXDvyf*J6nUqTEvB9Mym19Ej-x1>9(ZfFj3xz2z`UCgf&(eRbqlK5`
z0QsoKLciL&{{&^S^_b)YWObX{qNPP4T^$vJK#JNc;1FPW(k6sJ3p$urY&8?`7H&E6
zPnz*g=*Uogy^^GcgjKJO$MED1w0<QDS8>){$&3;5lvP^7C{Bj)xhq$Nl4R&J9%K3)
z!-(sQ<%X#DAhKEoM9?V38}Ca_VG#Pd3fHCw$Ygsm>^$(dSaqLZAXTvc4T9#Uo=`4_
zz!zCc?mvnC4^_zoyi_NcT_g$=u>=B=CqDJVPZ`jq?0m96A49C0?vyymgPldLX|GT}
zDU6sHzWPAkFx0a?-8Ku5Ak$UXM3yc}j2NS-tCr3{zVbtTH3&F{!6kkxmHsZ5s%boL
zE&KY2;H=H!J+Z<=bJ$C?%s41)NFNW;H7t1ZQ$v#5D(4nK$%WZE#pZ2IpTu~aUc@)t
zcKD$*?6k`&a8rgaK9TMLK9KV#Z?z|IRlvnYz5{gUmgGT7^!~W>Rn?`Wha;v0EuOj$
zH+uDRLhh$=zMf^J=di0<f9e4)MhBqAc!YqvaO~qq+dp(K;)bqppz}wTP}m=1JB$Bs
zvZ_xc@!YnP8DjoRRt<(Atf1%F74`t;^ueo7*$4?iN|bE9-T)T-JkkUGpbD|t%Av*&
zCmMgz61wP_=qV_mpTGu&;72<Eq(V)m^&7|=S%{7BW!OL*1C|jX2<7hw8ve-}$tn7F
zTa>|v6VBQRAp2;!M{Vw7Ls06RkH+Ajt|cgFk77}p9>tLfHdp}5O4igC@lWcr+t4f?
zGb0<N#-f|M`~FYYgJM>ql&WHQN!erBbf$9Bsr+%rWzgB2Fz46lsEygS>nvp<8^kPM
zHpMavr;}*vo*ctHyA3>0;(S)Dk4mowL%=VKs}p%lJWDyM5?R%%G$p6Wo6@j%M_uCW
zBXZ1fKZ*9sdyCoW2`K|LEiHpowd?2aW!#MZb)d?~P!k(0iyAftGT{s`FhvOVWk<5(
z6uU;~WwD@y@stQqhF$OWFqE1bDK?Xjd44OnXXV1V4*J6i@h)eeux+aN?ayUP7%;7=
z#C6y21#Z*>h1W!+u3u{TyOZ!L0#wcV&qK9+>jea>cBx4$1xa*r%eb!5hhoa9km1_N
zl^XVGr6I!5^j{Rx>{48ow^8i{rVj;8O8M0o7k55LY;qV!l;!Im#5!p5kN<fH&6ue<
zth%uCACzu=cKnjy{;9){3UV>V+h$1fjv4fXZ>q>^_eCE^M4Qk?_|~Ylcp`cQEVVK-
zhsq%nldxx3&1$TLVCzMouIuZN!2RWl{Q5j@VPhkKPsR+*akwWstfRJOk{9E-1q8n?
zFav^c%<jW_{n7n`il<dz^8zC(m%JmH|4fl1`*z1Y2UsK@zAipQom_s-w+xpEG&MF_
zuIat<VJ{`hsh_bu&!(>{v?TrL31oiL&Nwf7$Sv;irD22i@O?KhaXsOAwY|oDQ+#wn
zW4--ZURxQX(s1OQ@=TuDe>w6R)i9*BR=<_A|4xf=3c^T?V!=A7W0S+!$koOfS2!J#
z9`*1Bg;NBEPX$t+4GeDxqz(#1t8TjN|4e>%Dd{y|5<R~?Sotj|sup}P1&o41O_W7o
z%-8^^c0oQhXW=&l{GOE1J#Rt2rhpYuL}JbhJl7Iw!H5vTzL+Dtf_wyMcc2nym?Z=Z
z#9Kswq8hjKrVwr_n8cIo8WsKGwiGWBGBA(6I?N;O2&D|7F6((}ct3$yv@IF6dPxx(
z3>Waih(J|EVq?W#%dfnI@Eha}VHTr;sJ{kMm)EBu$L<WjKJ}Og&SGNST8V=?)ma!;
zIG>CFNV^^$>B=bJr6FP-T5jsnclCG=AbZdivdZ%6ZfA3F;O)%9S4AH7lZTjtO>@%^
zzAs{6JB86&c=cX(_bcoIBkbkg$YJ#F!x*1}UD4XP6uVa7Ey)CM2{(kM+6`uP%tZko
zyOOvMjxJHL)J3htYA@)648vIH7Iy@{Zb7Aw+6GbUA0?r&rxSjbx%Lm<3I(P>L(3|Z
zARXAi1l}&DgUbw{AT7vs3RugYJdDL-y!2Yt{sX!5hK}_cofv2->9ILm5=1b`zt4gB
zWdOIyT_c+;jH3K}oL^&jCaF*8^;NOb!d-^`r~tI)rl$pOUJ$ijL!&*b%!@|R>Wh{z
z(Y@n`mi;LrUCSl9U#9`Rv@O~<nO;>&IY+rEEn2B^PfqyaUDy*TvpLVQQdg(gCa!yr
zm|Odk^G9f1rh=nd$l*U-*2Ny)3UNtbk^moxxdoXV3Nk8*UZ6`mUfW~p0X5Bw>M|F=
zJuS3$dNx2d1t+Noi>S!AI?%FJAz>~&FirWGw|4DK)S-npqc7z5z^f9FsgOI6Fh5Ac
zJlQ)wVsRMK{1}EjV%QefDdAp;yG}GDJ_Wl@dGt0sF$gWU<wo1oAr(AzzgNJ`QAAWb
ztZPHQ!!S%Rwfa!%=d~8^B-^}P@3e9@F?n?l7GP{#^$@tRgaKdNQYG0!Hq1#9j&OWt
znbx-h*>MUsUyCq#5U<w?Brs<Ed!7&0EZtdujG}?ART5g+BRVmaH|5~=^sRfvuR4IG
z?tiuNxl)=w;5;d2D6f8JzVYSbJCXKZrYEB2fB9Pa@qE)zStHr$qWp;{Tg}^YOzgi_
z_tc^e7#KYrP=gJff(;(lg;4)&u$>ewb)LK?Y3=9His<y3QZHflsBZ22B&`OWCFE&<
z`1PF5am2reB!B;8%amB7p8(P1p8@QF2nh;8#V*i#rvxi-LU{fYDfO)e{mXeH4NY<e
zXUoS>gnn&J=nX_4xwM=@xI_8b<CIJuK()cGo$FeMP<qvTjB6yyrP|Va-+CzZ(qvs1
z?)*;b@vZuJTdw4hnJfiKD=4t^2%#SaUWktQfWy<<7>XxjV%Gz+vQBSJAzykjzl+`U
zK0n7Pk%g1+Kx>@LgIb{r38r=e-{<K!qHNOUNn_Cs?xIiCf=jw33wIkGz-?8yHsVMB
zYMn5Fmpc_%FgqJ?C%W|Joj?&9(Tjlt1sB9724ueucpxwRAiHkC7AzQb*+Jn>__vV2
zk<HIKnr_Y)4{<-}#fCR`K?|J?N!v>KK!4%Ani#rO_^41lD9*MZ4ZN5=XjjoH!Blb1
z8UrJ?`GqFm5|(&ou2Dmvtb=kWjS5N*CIM&)Et7b6he&U{Ed`X+GQ@*Yl=)cULWn=j
zG<mDR7~QP*v?qL(&=CZOih`;Sr7!EiJYpL&i`$VH!)DNuw(24VTIZZ<7rC6|tDFeq
zuMpmX5pTxH6{!w}iUKjo?;<dCO&u&961;o-sQXkpQ1R@*JC8uK$?o;kr{>pir1MFZ
zwDRP}?U7qz4DC-x^!d<ZV1A^^D$i0iLz5bFNxHs|yd2|CP!i8OqzrD|yNjwD5%-<I
z4|^tVx>V@N$4Y`zMjO-+kV};+`Ild|&6oDG-|V6+KUG>KqP($)3x`f~JsFS}j;6#r
z&B^YX&yR>V6{PN+Ca*uTOcknKSD)k;D0^8gUm1UujDV11<hbzV#}}<J?YO3;pH~Zb
ziohv)A68hOUOAR{3GzL4HrYDOnR(tc_Fdm0xg96c;{?M#s<xA?%;tID12`{~h4DFB
znt?`VOot1Ly%lSRFh^0Buof2K#~yov34DBU*kxF87&a+vz%4O?2^wT_g3=w!>@te7
z6^@awV%>c2g#F```g_g)YhZpE@F-KP#@p(D2*4nFQ`l3uf3BSbwT@J!X%!&9iSwXu
zrq|12djn}j;*W^+Y9XdsUnKomHklf{NbWg(HLWFt4s?Z`{rjwlB0bV4OT{$)=d9P5
zsV2vP);(0W4+^Rzplq6u@=^Xfl6Gbkm%4!X%|n%j%SjMKS8y1cagL*uQSC`I_?%1j
zJ!vn2$(-7ZfCs^fhVb!ulIT#2vK5l;N}9LieA_idb#72?c)fo9I`q{c)b*jfgW{_W
z`CtU<WSWKI5c&M}!CuTd1;OOp{E!!Im63)?ge>9IfojFg6=z2n`{2~)#;r5<4GC1m
z`pkK$A7~(@8M+v?T<d!e)^5n$sF3YETZ+J^0BW#ae0Xu}NcfAG<9aB6>*?U4DC*7g
z0XJpwvIc}xOA0vLLV)Lc`I^DBBA$^gz?Cm-#(Jua+mInqEBwO2!PlUf^sHIWtC@>h
z?hTGSWc4NWixmdhe8zEibw<vy)v=?OZ+;_!<g8X<QnrY-(2Ut=|3U5%i(rrD$C&SQ
zDI0ggl`@(<Un+g?dKJj6MFI`CC?An<IsaQv)7;l`ZEvDIZa?O?m2BTYjMtL=cc!tZ
z*9p!KqMlD>`R&PgU-2ofc>3!V{)k~<q3CJ>^j@l|F%YBOK;++PG+$2IyyAwN_Sju1
zSicvc@mRlPt$$LjB>W~QZ=F*#p;_bE<C^0Q!Fm<k*a~`_q?_h<;>majm1QD0G*(El
zp+`{po19bJtN)ST@~7HjCGM%IrMjB`EORN4)3D#sfEXwY<thGEYw(f@uqAC~n|gTl
z&VQ~xD3EvRfLD8(O;+squWzZ#1s3iDAt}B;I1dqD`e%@bM;ENOwg2-=y<P)PDp`xl
z@<RR#9P*Jg02IF^zCNDvNwZxC>DpON<);E!U8X>uX?uMm$SB9O`mggtVc@cY*Jz}0
z<H)Dkr%YFSJ{iy<<i0Y0ZSwp5hZ*`UuH?*lcLl+}sH^`?_o5HjNCJ7%?8dkJ<$o1H
z<>_mP;3$|g!~nv2W_d|4fGd=mE2;T6D=5~PgUZ_Rw1q?XpLehV0_hKV2vjjDu#MU7
zfvUgvswHb?FG!637!gBcyAVsIwg38O{u>PA6GqI}tObT({}Sm2z^MQ;#AyDt##zX_
z`))aCqD;h-%njI>{q?mTkOlw~91buHMVhtt%8xjmP+|jAn=|YGB?|r9U%qgWDED2Q
zeXak`z31O|qyO>~1tstT_Ed>N@PB*l|NhQ@{Wc)&27n6xJFfTNKbklSyg=x~MT>v=
zy8gd>mj9oA1p$*czX5(}my7O#$m{B>rmL&L&0z@N3vwWL6#-K#g2TDfYF^~K8Ig~v
zDKx%q@YXxr)A$AgTWX?gPXzXq0A$$ZbYZFCKkG;SRNtvoAW39+(|>;XJkOf}Abc(j
zINfW<6TLE%fPa_T;TTpv06-Fg-T;T524Rfj@3yUAq07X$m2z%`IEPo{le(DOV)<cD
z8-1ZHka4nps)M`D9Eio{K;;o&UCw7F_P^}`GR1JXA%KiB^hVIBc-H*K*>&E1#Hm8W
z;UOUxEOT=Uw=#6!d(;HT%B(k&JrHOq0q!Ef;an+LnM;E%wcbQN9LPicT^6gAUw)Q_
zzUQhvQ^|JeU-q8=aZ`RI1t5*Ff41>IVVl=tq~0Yb$7_BFIRI9z6O<7l&a<5%Ku%~Z
zA9rn3Af*y&0vpFVXd8wb<mW9>lY^GM@O2Q=g(Tkd;$`<(TBJa%1D+!UFqH*e;@e;d
zAuT{DbCk7E-`bmqfZ-W%katy-nx4H5runvlktVu%C!o=b28<{4k=*+hq}IhWx*%k>
z3Ba9D5Os<X5xL2)R)-tlr9y$tOn*iJAb0MX17*tDoY;)TjH?<gHfxd-KH~%Sk~R`b
z7J9(<*;(&{hO{r9?B9vzpUk3Rj!g**Ik+cN095c-Ie+;5*fM|uP6asn^uvB&x}P^b
zZs<e?kaL4wy}j&MPUYbz#vi_HUw(<b7|Qow?)imM;#O0tlo9&ZZVApK3R0)~jhlY}
zdiNg@(2OC(zK;mq;SMtC&0evv--k`VT}V)j@<uUsVgX=7YvHs<lA=BiemX7pZ8{7=
zOEhMUu_2<d2zMYRJp_Pg2vFSz$FJItl!yQtZ#PMzDxU)dIU1ij(a-a3$3X*ld{GJZ
zZ(S3MZ+yLK)tmGjQj+XTw1%7pBEcC=x&6%39bBMfDl&Tih3A$o@=GH?{zytkfTiol
zGb#661^^7~WDQEX#JMFGWW8MX00?Ue88wzt{46!M1JcAf2ktU%py8uvFNm-U0FNQ_
zm61}S)#J~ri>}0GHbw{=sOn6=4x(Kud$YhVL3q%-#s1G5=zo5P0gfEtNUg?vQToU7
z>%2yqX!7XR2oP&8Lsmiq5La#EPD8+8m2O9mTDH;Kq*RNOfP^X;kUI<pm6)Q`7{1J(
zM6xNv@xiA+Ms4TQ@s~)x0sny&=#)`T-j^w<wliYE(`{kb1R*QDjc*CQ><8nZAJ+6z
z1AweOX}%L%#@H%%z%qA(UeuzvHf~=|9QCw&;Ov-529WYl<-02yqzcH%=kQ~o)Y1Y5
z)oTxsQcPAasmlUSLm&!}Tc{hl4iYiJYH?kFIT4fD8L?zb-UHrDw3`F+ew-?`t79Ox
zzVsC|N|*!k?ljm$9#vOnTpL4Gw+sgT<4^zB$mP$+wNc=}Gz~SPJNNgte<7V(1>o5+
zjR!)cztjIlxou*MMs$)?AbV{p;(V_Tr~A_?hp_T02ydW@go1e^SM9+G4}8(KaSMT&
zrSeNT8Ao41Mjd-Qnz5-tYh@c0!NQ1ev3Uo>-7V3+o646-*=>rU-p_x&tIpJ1yVW_&
z?11Oq&@uk^kNA(Dj46S`YUiO|;a@SvM|V=dmiv(*Jg1(l$;J#SxiSsbX#n<?t-%^}
z8aO;*_Gktp+F>UryG>7|aqVbnEE%CD0bKj3?84K66?ejRo@uYAKAz-dO$QCcq`!GS
zhuoP1O|psE6U^e+JALQVKW~EtN{KfhygfNS3<GjG!SG|6H^<W1MGu{1B9{&HofLPq
zrgYzf+-#IOrAmf!)4}#~BJI8~uhlK31F*e#oxhCd?gJ;sWIB=4{>%wE0%~v@s4*_j
z9)Ok(@*QG3isnW2CG<g;vd`cZeg41MYi}9sA(6eF|KrR-!woMz8x}SD7S$a@T20*f
zR?Z<$nOo;RV~Ugeahewv4o7cs&<rqLxv%1XabA1s571bKVj8TewBpWP<5K>W=+bWZ
zm_n_AuN}i4^kO*aAxI0p>I7Suy}wt&&`CmD*#k}q#67@%H>IBsBo3=h0+|b$Wht1X
zTf3?wOhG#`go0`4=J(Y06MPSDa9mz&4j^j-@ar<KO!D1|kVcMkt#d>@KX2Q8o~2J>
z2{cd>hw&)x+zDF+!yjK;nbnX;a~-hxhU}9(?sNf^I4#(%G&}ZV{XT*tc46~X?MT}F
z=eNF2P1hw^qH<f!G(<bj`|`V7M;~$WF`oPa)$ZQA%ZC+~pukn^QDjQV423)wN|e)C
zhmG(PW|Npl^nN`75NXk3cFk{aD?Lj09|(qSUY;1JS6jrrlZXdY_@K`j<XhdO$ghdQ
zxR(ZW+P7RdR_`7-=t_Rk@z!V*j@}8{z8Ff?#<$Pj874|p0?BL)_!*AmmaoT>mQs$C
zglY*B<_+2gU#aX`A3JzCVNNiXQD49*vFwtvow6O|8QFxyn;Lz&<rbN)5%R>~$vWen
z^obTG@OQ^OzHQNo^xqEpD|*4j@2j&WeYCH`3O4}A>|y}A@%Se@C7-ffAH2R%q+KY!
zc{{c^Un@PiA>FyF`o7wZspymXF^MXJ?}v}dCA12Cq=f$SxeI7$VrD=Uk*D-mm|L6F
z43Yx)ZJ7;J%p{<e+UdQq^Gbq98aaO_pl6!L>*?FreW14_jyjlrD*f{z$;CjXc2g>-
z&A`VuN(~xRvc&EqSM2FUX{W}0EBU7nA4UpRt#3twA$|USH8;CaQwxlU)mxk4BA1Nm
zS-*lAYsB6Aa_m{8FzPyP3g@q+s?<(7`UqUPFj33S8Uy+_)Pc7KfBXm`L3IJ!FeQ_2
zm8rKWwt4GdK5IXPT&^67i^_}GWW>nx(F#Rp@*&JkTNQaHC!RD(?wsKw=zOEbv0|lJ
za!;3X2!Q7HCNE~dGx6an_tyP@d{S2fYUp&WV(sUN_SCJ!@0fA_!P;-oT(N4QhKj;@
z;dbH#0ky1M1CIP7`l@O`yN4E5jqyDRck%;5jyZUd$L4eeB>~1x4MP-pfTYcEYKw5G
zpblugExr2vxL1ET&y?dx5{!~P-v9hzvjjHMTA_cVzKjSr%f5g+gLeGgxWuiq(pJ7L
zHej!$1x2aU$P4^KU7KJ6CR4RWr`YJFR_XTAuaOUb02jMQT4yZab1LU5b@R&8EGuGm
zIIpz5Rf)$TnuUlky_2*h4i_iEv1IN`^eAWOaGJQTTz*V+HfeM6J2ak8m+hBu5A8cz
zy4&>Vus>b%aF2ITdFC!XmFUf*)-xQJT*yk*1p@>(lioe$fv0!K&s}h17cv4DoXG31
zIcUg*fs{Y_1JIy!MAP}>P2ESc@-W$h+h`U$6a?qgJRnNB1L8<Vx@jk#5`XT-$3bT$
zsEFfSDO>K^3NOP?eiqgJd=$%OzZa%4v!=544k*BssMOKG99z&`<(}T^>@9F|@r`pM
zdsVRPPX7fCh#Y9d`;{2KdX4YZGIitI0<z!ZHWi2h9kNqr9?B1l>EseetO{Ydv+T11
zO)-ai7rigfzOQfB#|{ujtemjLdh!|ieYi?RQA$0>RptL1?dk!@{=<W`r3c3sMJ<BI
z$A~TRU^z>geeh(nJ0xQKe!|nK!G0AKb?r?vYr~)jaJ!>5H@Xkk5V3xBalh~Aihe<1
zJ>&EjzIb3<BTe*^Xw=u=^z3WYxu5<eYM^)04|K*Qp!Qk12G)H)JW-L&VqN>tH-eIN
z7`pne3Gn}X{E>zj3%|*J^4E#{`Xv=RJE&VY$V4!<R=yKFcTtfFIl+E3tt5lY00q!t
z`TFXNWbfdHqhBA|Kx{S>TTYK*M=I3cs1OPkqEf>BMBOh`CDhUXgvv$)zJ_Xj2KA*i
zJWsnC)matOu-A|`-#}RduA-MjgV+v#=mM7Xli%;ypk#q^?Ru!EH`Eoxi--;v>M|-m
zR7K#aRYU``cNZW5=Jz1o%8&57T6D-7GFr)I3ZGlSU0n~5i_Ihf)GYMl9ID8fGm4Pl
zv#8$L4AIzFKD3DvPgl~paOzW0ycQHlz6NmNCpG`#vlZ_gWZlyiM>Z=22k+tLKhk!7
z1JV07&qHr<W%f{rFoU){5^Et7UZ7-F{zKF6K!>;W9G=Ml$<)i($$EpCaEK#UF@}DN
zVv`*Uc}V2!#R5eoRCT42kP&KGRCV~Kt#J1{+QvE3h<CtD;26182hOWj&`mLiR3PWA
zpc}VPr}r{12aU6(k=o`r$=?x~>_p;lWw}#}_u93#l05@>VmE$B;YB!Zw^T*aLy!xt
zj`WW)k*GK&Ne~Z<FzRyLzKlQTa_%rY2GltpO+Tzc*gpm*Tona}5b|d5jeSK|?zt(@
zu=yQ1e&o5cR<|aXK0iV}BRzL<wzFa6<al>#%>`yw2BgNiJsvZ<*ejr>8cdv%TqjH4
z5qmv!0r&1{A*b$rwvWRa5~J<;><5U=bK}#D^3ivyv^mNy;`@pOqNyOIckNG^^SJk!
zWHj_Ttk9Sv<SuH#@q(zv{nmq!XX{}4pR=ETQ`;%1x9D#iB&?}5hDbp@h&3~htfA&W
z%qWn*!~sv7xzm;k=bHh%D<M$Ye(q;=zO&}81h@ko&0dV9{&uK9JXx(wJhbWDww6CM
zM9T*s&$R8`&Hd&??=|4y``k^sP`>`fH@o%yg3aDy-LjK}>;1Bs>{gCMGY=sQ0M&lT
zc=6)p`#B44wns*n(sGK-(m`>z-y3x>b=7`AoGxL5IV7|myqC!qthZx7SIp(DwmMs_
zReBfN$aQ#RRkfkld-enpSjptqb=1RnWOI_rZThK48l9J6n4bJ|p7D)HRV<A}1+hza
zHb!2KHE)0X>F$r44tOgvxBnuRDUe4XF$f;Le}t4?93+P&Od2YKZ+>tA2)z|4Z&9%z
z%-YZB)+|?{9O=OT_MUmHMHB6i(dSfbpg(pd7mL`Gel>s)X;G}mu_d$z5|AhDAR!hq
zsz5aLw)+$0NR<yg%-mm6JL7%6$G`>oQr6j1YO+`A!_oNgO?>!kd^in0>z$zVfS7tO
zy$1aqDaXLqSj7-5GHy^*VkO8$+*w~7YB=f)%Rs@5Z^9DdhT70O&J$hDDB0K(QLY`W
zhLw{FOMD0xv5*nYhNl)BH2+;mrAlb`At1%jFvhr4%dq*RB#xBVyhatTuz@XQb~$wJ
zaIFvHMtwp8+IPeW8lw&StT5V}Fp@dC$0OBr-zUu?$>AEh7X+n@E3k@L%j)MD-i0<k
zP$RgbC*aVY@#l!T6_?HN5{r6u6jbB0p8FWNSwHi5i~g?;@_Yi4_a@A2&*O^R=oQDW
zHCX&*gi~6qe0oDwyn`C=ijT;fxPf%<-^p~nS}-$Dd5!Z1o6ZxW%61Y9e%u<U@5kaK
zn2rw`nLp%O@wi&Emh}6lecMpQgkVhqs+T~XG5V>|Ay;`8G^JnAc*v@8baBIJw3Vfe
zTdXLG*WDyH3CEkl{%$bWQ96_^qYs{1UL;|__o!)7yFeLs?<Q;{Zo-<kJA=9zA70$^
zG@Er!w(35VJOkaO9;OCLm(g5fE`C<{k?xcP;of2}C81)agxbYN?uveH()!hz@|@x{
z3yV=~>x7q3p@+uZ!<2~Vk{}QsI%85>B{W3lBpek7x(!mW?}QK!@P}CZA<nOtarhD}
zOBuHLLvz0|{6R+6>{G5#p@|s8Rnewh3i<e9G%ULAI9a5rKnRw0;0&wS_lioLJMA)#
z@7jM79`3xE&HwY-mAB|U@Lt$;_5uMu{Gk|Z`hW@RM}JQ6iiNnSDdq3@;zr!#2)wsd
zqeJwV#4h7;<5!%<_wnNY!nBb<v5X{x(>ebT1h*cM;&SNM*a6TZjF^GC9&N<<*{cnB
zupu^~Q-(B6c>}Kny+^7)m+KYZ;-(Ka@X}{o;*iS8HUOOx&;UGIH{a_zq72Gl5L`)X
zAvpF_vK@(u|Ec`(0igh&MJ+%Ll?<1JzA>+AW_fcOAF686>spBBd?b3>Qd<L}Vg|NY
zRa8xqFqI+wMdZ`>jFI-=6{#cHFKo`dU8$r(o}y(j(EI*{mV|vq43*@$VOV8qQ^yLH
zW1n|{G6bigUyeqJMkS)JsPzdS4^HvEO~Sv0Fluoqp0dZMk}8a0V~158J~)Xe$ABUA
zIPBYDnVH6)Ng4Y)`eSSAknSU^6Ju13YIWjMLzE~27D=lwl^`L|niLrVtsml}vKJV$
zt`MJN5MEXzRfz01(t4%b1r6mq>1RYu;CP_MMitFRi-U5$z1mdDk$e}Rr{^QHRG-3E
zuNnpG$ae@MoxFDeePvqvRa@?mMM+#=)8?j0Tw!&-kH*gE=&IF0%aQVeBb3#ru$(Gp
zgE&F7BJRV9G6n8%NpMe`ypw9_h5w6MXtX|GJ-0!Rn_tZPw0SyGgG6oCSC*;GPDbQY
zD_Nq{kW!K(mr9FR(qqE}riQW054xByenNU{jJ6q%Odo#-gyt{Nit}r+Cqyj88N{92
z@;oKjG`$dibiK~`A7YzSDW$9Stmt~-YZkCkYdhorTx3unAG@W%>%Y1j@ORm5>n3R?
z+oe9CT-jUBR;2kDsH==AZ?2$VPjvR%%avw?kT~B=vr7yDHe0p~k?wO2oQ0S|i&jYS
zi_dwPXMPqWJp@x>cr+QXfiZbD)VX2eL0sOaxtMs>n3Y+%q^NexzkX7EB)CX+EBuy`
z%-C()TZ6&R@z2i}X1-7liA&EczT_|i;>;A<$9q4n?@UkVSU=Z~5f9xbnwnB{Mx4(A
zu~SvH4p*TVuepZD4|SEhDk**59LF3sFpUEJggHq+)<bysz$d5Yq2lR|37=bK<kLwd
zjp=f2HAM79DIV@f!P31_<P1{2y|zthS&>~W;a_s%GDR#J+t4lE<7aqsW4EFK?p#WC
zI`aHHulX$<8`65!>O;{;*F|0-4FR3a+B^o=M}guWqr{Gzuj%rST(XhAh3;*#&61wO
zR#x4QdEuULT~Clgr19D3%uKibw~hV7u~oOo99u8%=8t|($~<$~L2@y$oHzfI&*GP0
z)t~YnR$1?ALAyESpv7}|P4@XY9^U~Z#C@GQeoTdC2uK6%I>*ngDzrKDBzfO&aE~hT
zX%1FBKQ-L`U{r^-%5ciSOr<o<tjmB~oqeIo0eVwbLuQ#|F-fC-HMcSs^ptW1{@N<u
z*PUQx!&0tSwuR9N8*}rS-d!~Sd!FqowhED1(IsJ49XFhrHh&!W_-yA{KgFd7!@a5<
zsF%LNxgHwcb8D9~PfLhaxKvWP2DtSj+uUj<@YlV(Twxy*Qrr5~>*sQRe4_Ri_8lVk
zZ{*6J?RD-<+ZqHiualI?Mxy7_rq_NI#rih))g#xhR*nt!UXRSYAkdl-!<##82uo_?
z>fFAX;(4DC@*z)XCKb<<Za6G5Zqr#6>1*cL&h=~c%YShW@TaOEa(t`Ee6PO)?`U!X
z?f6K;y`=syW3?Cah*Ib|jv9r`2aAV%G~5wbIH@SP&mWY%xIXHJN${N&&{v<6FV)Tm
zCi<x<U@dr=r5minFDI-o6I16hyM0Uydl#Ph-FlZc`+=Lc_<Y+Qs1YnEHUr_6<Bv=j
zWP-vI@<dPxNH^rc_i&uH48a08%6w|byD!LJk~(twoC_$K08ID>iYnj^lyIS|JmkbU
zt{M|dCe0wZaYpR}vjC}SiOQ`}TW^6#I0VVFsE&zcBGA_1e_F{6b3#z$mb{!gn^H&m
zFeRbd1-y`CCH5)n&-ZL4I0KS5_d&DH2+hu@VQG%4c}BDZA8hRV+G5f56enfhCNUEB
zR*~Q*2{Yn;U_N#Gb1Gk``~&n0+_NNRoUbvj&n*a+ux(Oc65a)@N~9TS<|+JpsmaR{
zp1#A=OBpk6l8Q2Il5?0<UE+9?Wkd@v6eE(ZM3MHFXq9gBG=OJZ2vz8q(<78!OT2x!
z$g*ws%HsEO(sdKI^4g}jc1-jhgE^}mr5wNT1u!TWs6q!%D3|YoBFb??@Ny2Bs32Y}
zvT@j*gk4cVts}Me&IaSgYqW%`=vROhnQWxEmc)t*wtmT+uUbyrZ$9t?Io<w~0U$s^
z1x$TLXCpkad3vl9uJ*b!Rj#`KvEl9S*!^Ap`$shRuMfu1Kx}l6)ob`$Y~&*KHqnfY
z5vVLu&LNibWyj*OZn$nlWL0fbsl_9?83+YnF$@3EA-_Y%EKz{tQ_7e@s=2T%>=ILR
zMqk%q+~q`*!D8i#=Yk16f-JxXj>HP}gi{+q$infdd}hHfaTEc-Ca4W;UvshhgU4UY
z{2G)?6)#!w27a23Xd*UTli`z$NEIAW-oXapF<VpzxO=z@&^Z(Y<$dFGi#^8})Zwk8
zTE;8cA@lLUOSQOC?U&<Fk`gH3ZdYg^;P7-pLvZT>7B-mU>M=i@7dkTww3UUohxljf
zf0u5T2D0Q6Nub0?p~QNt8tUe-RcGU?q+t-`WC#gCcv%n^MDOu7a1YJ_!!PUi><K|C
z&M%W_u|d@~-fy4x=3myRqPL0d_9M1uM^-=|TP*J>tC=o0C7hErkg{TAc~qY^m{Y@J
zW(~HLqs>pgoX7smTKfX2KD7;P!d04|JAn?{`Q-+3#_F^;)Q0xrm3DLz?lqL;N*Mid
z={?ZHJ}M+*_n5pHRB8vNV#`)PI3WN)*C@0uht8ZZ%KBpwe703<n8&5v%Iqtk`!Mz4
zkhWF6GhWV04Z3z7z29qJH{#e{HNO4-Dqvy`aGzt;pZG@_-vKL$Ss>sSQIQy!;DmVx
zr2$_fR+x@J?$0n#?X?bkLAE6fq?+~f`6iqYZW+`Ys)wl#Nm#xV$R#6GBUbCJ?L9}E
zHGI};R6<!zn;TcKi$1(h#HgoM5?CCe)lOdk+T<@R;nNHeC)CJc0M?e1-D2e(D70@h
z`eqN?&k~23a4Zyde4T4>`Qaq_IRX-P6ISUe*+l}Q=u{JK_&jc9BPMO)0w{hDX@NY=
zFqeHY5ef5~;)J^=y5i=yiPD}lizekl6ZLyM+EhI(?_`Yf(ONMzBp)=Xf~{ocBG<32
zDg)h)BhLJv#=520){MOQ>srwvXw8U(d)JLVm*(xh%=xsM=$TTTpLgx5M=QBL0r?PZ
zU#BnH617>n^|DeZa^fn8?t5;hs0@!g@yhbwvt6|G`T319MGy!KFL$Tf)%x_OcL6%9
zPryf2=_>D7sR^@bm@qJxuC5&0XKz&J<n8M)?4|=oW8T4HDPsOQw$;A6;fP{SA9fiH
zdpW0hwcbf<a)wxy6Z~N>KJ6kNkX;<kB^L=?ZH+2;{C_fPumElnPAzuD@1>yWs$E`X
zd91Gic?}VIeq8(1i0=X7F)T4&HevS}3E3$W?DzG}vKPs$ksmJzo^_e)rhCa%vvcSt
z`RjMJdOQ3vuw^J~8<FvS@%Xj9sB;~{4CKbM!2RU8tX@^0l$ggl#_Yw1+q$EFo?Ut=
zTD#`MCX+DrJ7$|~!>p5$0(q8U%YSY`bIf0G4%N#(@1QTctIZDov9SJ^2Oot<-<UGw
z-}}3^6|h79jjTLNw~+5d$=-^ve5B-2NAB+}S#|y?VQ}fDTC!nz20Y+0&{#6H^2Pat
z@I_mt4GGQE&VlHu$b^0|{d1BVr7SnT0xO`4f_YjMIT4fUC^URS062j>6*?)`ZDLKK
zvs2rjw;)wx9G~`<!Ysm$m89_@s-y~P@+$)Sh5DFAVi^D{;MWImkp+iJ*G>Zd`VY;-
z9l;dySAHk^v!MOZSSn${0GP?uvyNKYFyBsa6uS*{4hI;ds;vm?Cx%!yhC+<uYOn-e
zPCUF_d*1@e#yLZ^@@9fmVM#eDXw^9QW7%7gYXW*6CfGM2CMqExEA+uBWt6Y+YFQm3
z>Ao^T^wJtZTc=Y0!1ychm<ecSoE{@Osip*KBoU2uH^931aCiC;1Tir#1G0}hT%y@f
zl2PEkeQ^`exPr3I8G3u$ro0eTeaOcP@}HA2$##NEw(SG@e7G}UZfFm7amX~vF1P5o
zjA+pRMJ4jr+(}+}CAaZo&KdlQk4+tj<GmbDs$+duDn8v7>M3|2KT?U-buHSzdWMA0
zV&BM*zWE91>?CNt81_=5W}O1nM45zcspy7pbp5~*mUf<>75hh=6F>H`Tl9$0cM5NL
z^W;LN(T(o#hwh-x$|}VBM~su~WWrfBQ3R9?d*qbwf1{Id3UvS`7!fMV`@1WD4CT+`
z?)~lN{a-{(KP~VA`J14u85d_Zguik$lb+bQ+lzBt5<;LQ0h_xi^N2n$+y1;<Ff;1K
zjHRR^w_4kRV#bjr{WyJYn4kCOAC?YdnoX6+Zdp*U7m6sac(MT!@&)=K+6^kLK>A;X
zDRl;gLTZRbhv3opyY~!j-)7~yaxiLr&I{*-9yAB|%7MW!50Lz6%DPB$0~CFxFaFdV
zT7~x!R51#mH*X4PnaDm6v4V@xZ9X&le&}%?aNrVZW&g#{rG0}OlY7ndE7k_ce;BF6
z;|Wm3FgcUV5XgR1qT3B%2hFI{0ww;t=HI(xp-(<A220s|tw}pMx53Y<*5nE1N2Ten
zM$~?C8C(O+BX>##<5x(e=j8B6Xm0eXfZA9}+70q%Re=(}IbNiKJ$7q>z!v>po3u{y
zi?&DI!}E3iT8$?HG5S^E5Xpz%6sqNscg~%qmd1EP`;b<~R>=>(n$ZW8X)p4+RA`S6
zmiCkTkBC6~aXl&(x7!7yw5mk9e%5c^MpUa37U{3I(3{;9@2{_+8n<?wbrLAFst~Uj
zT~Zqd#oNroHQ?1#9fV-7WmnV)zE#=YoO=bSJ~Ljz5&RCN<j|SG1ch9&E<vQ-iu+tI
zoZ(;+v`5?Y{^&g;@<rbRg0bL4?YQ|pj~<r$QS?v698orfx&~x6l$L??(&#%6OiNrB
zPoe;p`7Y1@Srs=T);And2jnwX$u!d`ZR6(%mEB+CLycUOmw%7rRA^$QSxOP!y1wcg
z)5Oa&ZghD=l5ihN;C$`}01#D)ZWhbQZ!R9EEL&NEb=;4?`3CQzTzVTnedI<$jK)^V
zB(#KA(OW%VZ(lCVKs8wwj)7ZXmRMnMSpqE~F;s&%E}TZmuwI16=u#4V+D_Xq<@oX-
zuXo80K#7YrrX9|b&rz@Op(L!nu0PoiMA&Sn;;GU{mskX6eBSB3>@FYre+c)R$#w02
zQ<nanu7X|$4x-jddH9hH_=CO01jb=<cLRM2&T^?xbET~NaV=4Ek)MG(bC>iLhey+_
zzVTV0h2P>%S{E`9=k^N?*9r`m3#2v-3_lg<vq!h%EdNK3!uiHOdKAXiK}DAmR!e@E
zwALlifH45LS%YR)q<tGozeR40a;ZpHU1ZMZD&f`$sEGwps|BJ}wj>OKUt$Qmtw6Cy
zs6T@X{L;_j3k2^91`)oJZ$nm5z%=*VK*7mUM2p-{1iBLBSyB<e`t#a!%<gK6T@iJH
z*OjF8uI190m6mq;X;2H<7QI4(c^yN{s7h39TTmFeP+j6!LrHm<9fg`SvpUuEmyeGR
z9+1|H0CWSBo8AWceVNrIKfPgv$MYIoDtQQ;koF8-@(eD&_KBj{6%$Gb6-~OpY0&<b
z*6k|*8d^{!C*G0w<_F5LiY>ueEvPhN$&i{jl*s^ag+W<{XSlr1a7hUaSq8qwAbF;)
z5pIi+U`0gZW6T$k<0T8REhwf@rL0wOJG%e+YJbZiASMeTm>ReSa`~lKgExL=(RQu^
z3(qY%AwYgS=$*cge2am0=aq>#hnELROE|{%9F2~IgohJt@tDCFR!)T2TrqWxHjL)b
z)HXNE+wsaI*Uo7;>)?1xxjGhx<#e~vTI74GteKKg`C8=mUK_Eg{hfa8F!C8i<QYbR
zB@>LyzBQrSo2|J6B_{VciN9C)$huKvPMOtB7c_WY3#3K`p}(KudQHuxP0i&_%{96G
z&K}AT6z=IG^kezkK0NuuvF@*_C`#XRegWw%iL=GSCN6zPs}Bx)-PA@+r9#bG=)`T+
z_RQ_cpNItL=QbXlk^sZqvrl;wTTgzR!2Qr>8us54UzYG2Y*jv+qw<w+I{BtKZKtSe
zR=0kcYvwdprn_HQ`W2{|2tmIjr(hZ^Vmhw8-fLas(f~vwhe;mR5Pf8P&u&uxrc_(1
z((PdV`G9utcBhc>Iw4Y^Oa)AvyIK~;e^8J%W7orU(8OvV;CZ&LKLXu!w!n;wmTsqu
zqGgaf40znhCSin=B*l_041J-4iBrNRC}6#ha|YWFTY~!?$9qo?-#KF5v;`DpBUic8
z%ofIGAt)=&ZK~(1G-zNr+Lt+PJaZgcb?3tAVcDmbrQ3W<oX2beQN@%^{ay}Q_7i^P
zKg)}ZGUffa&Fekk%af}7h2QObvmQt2cg0CCPTS>Vk6Hv`)34QU60GM)?MNDFXkS_T
z&O|2da#`$dyq7Nqil@G}6)=my?fyX8BM0>Vd;CNT2-~WlyruJhk}Uxe=f2He!(WS}
zViYKg^vj$lyWT2I?v4|rziU`4$5=#&1BKrn`Ci5efnX-MUKhn(+{5SV!F2mniu7O4
z5`E7GxdT*e$?#Xm4RKd(84j!#he;J<R6`D-OoOzMd-I4I8%o71s8dwzf`NBRR21VK
zcP+ioRCi;9z9~@w5S#^d9!|nV984{graE8N1SjyY`zpzV(&HpJg_P7u$jIS1gn(4s
zRHp$)=8Xasa~5dvddno<XrI#SXSUPPtdb-iNe(7HRcN6%*tXewOL4(MuWt0VxzsR{
zFewTLCykK0-Q_ch#I)nvJkwvO&^t5W<?U%&AYV8N6lwxW1mihlWUd%6^;Z~gC)Uye
zzAtYcKq0{h+>0dcREEw)V2b##RIN5a&hS!h#d6cbad051b(LJ5!|=r$JbYT(Rz)Am
z@zzr20%hdwAIS6VH>s|?6Y5LVwT3nUfgtx~8`!*O71`#fTo_6jkcf5>s=J>ItkcI|
zfHcx{f7OfJ|7*~Na8FyUeo`0s+!pi6dUHRtg5y$uMEs}C65e)ps#JO!PIpTo4uy<n
zSOPrj6S`}$aYfh#%B&O@`s~@hrH$moJ*bJH>lG)m!0bz?{>k)-&g^gL^uC{H$wS})
zB2R9aj&IXIpfC=To9!tyt^9L$W<oF*%|*Vv)wVefxb8bXI8^t$En5TuMDJw^mk_y)
z;<2NW?jpwls!swm7_a*!IiW(gtGgdf&JBkDc3_Wsn>un^mSUqkOI5J0f_p!uEOhiX
zCotpmquQi5ME2ZI!z@QZUtydGcgXM$QIF$$KsW9p>Erg{5r|x@8|((J9ittkqM|qV
z<$ZOO)n4;`onGbO+Fm6ZJMPWk>#a{9?l;HjzYWUG;m{J>o3|j{-WVaIoTbB$6Ia1w
z%<=TZYi%1hb)D40gA`p&v8KHY;ur;=AoTLqZov19B<m<%-%$CeZ&JYlL#p{}=L`3G
zk#jr)PTM~3cwXjo8g%^^_6zlz2iziXwD$I2H30=in1U`CB#vT7nR5o7vnU4`YM1;o
zm!~o5=Yy?=<UAM&_G7Ke?s#jCdL?zG`-(^sy@;|lF(@Uy;8VKIN*X*bIe9#}>gXZP
zHV8xghGB*+m_&rb*6eDFaV?^Sw3P?Nje6AsU&<_`dSc69(8kh2g6M<F(O-=9!?EJh
zI<C_M(jb{19Fj{T1)R^4o(krzi1b=GZBf)ue&chr|EW?&LVM1OMgTI_T3m%*-pvYB
z@b5N`;y%9Z9)gW_|D>E2U=WTA84XOzV7qY^NlVU`8gI2q<NJQcPL>Re15$$CMoxCa
zJ1>7=E2VLdRlN*fG&u;lIGf7qLS9031N!_yO!gC_<31Wlxtr)by`h`F*C0#b(D2=R
zo*fQQ4%keCQHXPM=mxA;r^YfSN^~=Wtqygem&MYTV^6~+>Rtb<h2AO5%dau17EZXA
z>o1Pl!o<v$1wRKZL`c!+{y4r1>@xWwnG<^}mxl3<E4b@%m1ZMW6#kXNh$9NOEAWB`
zdzj8ZQS-BlQI`qBxS2#`N7l(N@3GkQx<#3$D8xWPcJozX3v1skl$nwrhzHB8IO#c&
zTyn2)<u+$LotoLgRmVrA$!9UV1*!aU(W3WnZ?O4)jK7-V7gsQF{|Nd-tAuloe&_$N
z_m)vr?%lia0znWYmPkvdC><&t3qcVP0TrbY0RhFJ8;M0OPzh;ir5i!%5CxGgky0Az
zt}}1X-cOw8|9IXp&ewO0{XyM(%UbvSi#e}(#g$7FLGi*(Z&+&@ap(ow*SuV3P(NP1
z5t=&PgGmKnnLA8f*7`7kvUUrxHXZ7h4W=|GOkbur+AkcOb*y@|{WVL@d#!7RMY2Ub
z7h-lgKkYsamk1wB`GdR^rcXq`eHc>EvHLgOnmidkriHWTEWS{~u?1d$h8G-fIP^LJ
ze&GYy$AGT@)>nstBaFZk)HLryRIw*poNDiwmdwuPa(dRpVI3}QUFmK_SSq<EYxD=c
z*vn?cpz`C}wf~9P{12@)FqFS)A?=dZMa6}GWsFX*5?~W9a&ljHe6isK-NYEo6KvF~
zcPmho=%2uHq=s-qN?bmYU-6VdIF~=k#VHM}b8<4ux$iS*3I~+hgT+QvXa2^@uMpZc
zp7A*YmxZ5b#~Bc$It*I!Yzj^gqUKUTDe+k~EAfB+V1KixBjEGh3J47S2j3Ujqw;tJ
zO`%MT20%@|uAVx`X|@0qX=cCU^e3)D0*w$5>7u@rRVY+6{g!hi|CV!nA_qA~7Gi7E
z5B^zooG-3)&YD*29RPqKA`M_;+SG}e;I$l#3<udFSOT6b1)_PUlTEjVr&t!SC!Y+E
zVKU-?@AWCvGGpol>vP`I`In|*N$`zauNVrZmN!y_r6Um`b{(ddQ$9c>i$4&{*X;;q
z6aU0~hc82L3}r^}Z_nT>H$L%RF8I1!U!b}oq!0}T!bjfRcK~j(kWr0i=R0-@O4Kj#
zpcata19b>~J4f0XGtkW+Pj!j6E^G;O)Q;-)3R?lmD%l#&8x`ixx&}3tU0r(YgWI#_
zQ4&`uK*BP810OUo4(EZOS_j{5VxeYaY;(&2tmFKk!Y#Wql;Qucf}DR0OaJ)``I`u>
z?hNl7{l9+r2`(>SPRdcf=8FK_|9mTW;~4*XH$@JTWlKrD=|%9|AGP#bjX;13V_)hU
zA6b4;?;;2n-MI&Pmr$xH88P?yXvM!s3|W6jrnsL9?|PX4`!?{Q{8F4Cyv_0;pVj6P
z!9UUc;N!T3Ao3R*Ob^T7sL|i7-hW-?|Mk~i%ii$+Pc#Z;z4+Jf_W%9Y|NqbbKc4?|
zBLNZ(wY1#K;%}J|FR`{bQcb&wQK$79KZYn!4Q9X_G%!z3Jg~;`s{C9##Zgwn)gz>T
zYoh*}c1ZpqE3n!h(s7k&{x!69ruSN`+7$pRUy0vRh84KFc@Sz({Kh!Aag%|%Lm!ls
zWOdLe4$rJtBu!DttFFK~_L4yoQDuQUu@BIj3!Y}@J@^LA`dH%Op1JNOjNuWRY;F7O
z@A3aCK*`;VRLtE4LM0W*xN|8@&-V0Pmb)DBv~U?wWq_C?BKI~THAB3g2Ny{h5uSEE
zfN@iGQ{c`jc|~$2)wr;Qn^f*#8Fxgn-F}TW))n^PrzWb%tzi(GWC~;}583Ne*sLnw
z_Haji)32KJr?JQqq~xvZTO%Z&efFor<tZWDG?m_k32c9D7eY#T*Hy5Rjl|gP@2r!8
z&6Z+M40>e$r=Z8__(Wj$)w`~?-5h*4)dV41)JXd25?FL>RU;W@r@%ESUiSB?0tyq0
z-?0H7!#SirR8<NhQdWG;l(*CvX>f6mCv~E%TK8K4$7eqXx)8b|-lOd+wh4lg4dHmu
zr0D7S)L@sHq#bS9);E7#4h3Su65KBlF<bD0eBmIMMJ!{}Ohc?$$(8iL8g)eCF^EHe
zS08H(o%z}A1n1plP-{dT+nTlf2+qm$z<<3^y#^fer$et3kmCnwF5Esn_zsE;TbQ9$
zByz9JfYThB8w|CDY6+;yDQzL`j1>4;W-|Q@6_ddJK2CJXR$ks#Rl(=uY9cHeAy}?`
zIyXB?_4|~DTs0rw@fZw9sdQDB+D)yfOpr%crO(~T;rX{a{Qp>`@{63XI2TWyE4lr9
zaT-Z;E<pk_&^`M(_7)y`kXmzmtw6l(k<ndv6Ck#<$h9ixpK1DGy_XchFB+|aWU;Cs
zzIZ^w;=1HT>PU$~BvJ9QlK{A`&QZwiE&Wbfb#`+w^t`o<So0$c{Gh`0SNCGBzrc2H
zq_FINZ1%@G)ccW<!dJdC<ErsZp12rBuRcf<boRqqr?1#Dkozf+*p%}#sW0%y1xU_P
zrf#e&z7(6Z;^Vj&OpNK3JlLA(dhO@_YVL5;<B%5nu`j}RDuiA%ofDN^#)9Z?`eEtl
zcl(v|QxntyZ{Ig#`{H~giDEUv^+yuC++jJ$k{~$FZ$s-$%@llT#FOEvQ{I8{*AQY{
z6*=^l%5^YT4kDF)<xd=R6W*fd;@OHl2qxHAzJgzaSa<{VQL=``*hwbMVkmd&)k0Nz
zS&rPoj}HB+^mw=H!}=OY#D-n#IlWYWtoI#Gz}4jt8he}n_s$SvJ|Vy6Ke`CQDl$x{
zqWlwZ-6~Jw#T{9u4^L4Yx<pb6cj7r~cq=yO7e5l?M|k!ZVb8Bt5O{)I21`nWL`}(8
zv?mxK`8`Wu0xhEl#$*Da?6byvSXQ(U!dn$E+G!MYM}q`Nh*l=e%6V`RarT*n>dqDv
zWPU)$jq()}JcbGoilkPhx$Y#Gb=Qyn9x<Tz4{xKe#J~W;Z~Jw2edF1*1`=HJyNCfj
zkimVCa8Wo*oN+6!Le4vYVeqy%_I}Y%@XfbE+(vd^qLx5-<KJEatgh~TeDD5jx)Us4
z+&6ATUJ4)jt-5iv#-O4-+&@T!SLg98Y?iKP#Lqhi=2@Dv=s5II!AbMfEo&X~r<-(|
z36W(nO{CuVTTB~(@(u@orSBZ56PJRf)l*~IA^||H?K*<^0gZu|FtNv0J&WFG_=ohY
zQcy0~>%<1Se_^o79@C^p>)j(|uYt&(!RArKB=iyDNm3ZKmTA)voEi{`%9p9{q{Unh
zdsD^&6@%DUh<~UgD2d@c*&$M8<4TEoo)B;0w-u~&Q@^AkT}mR5@@Kad)o_QLj1~bV
z$^7@x8Jv(w{io~v>nc(fuQp-$|I!0_#Sj2ft9uaEU_D}DgpG;7^<KQ+0%!R-xNLfT
zJ)veNCPpT+ONiLz)1~CAY}|41TN#s^mj`ZVDSX?7eeQz<n+SL#bMt*Lm`WV!x@6Cd
z`mvoF@cHjl1@Di)U`-r@(a{aow=^@N)%zQF7Ku(fMBTpsCQM!Kwnlu2ftETddiJ$G
zMj@4nLT2q6LLA|B#U^2+Rl1x166gpYXAsNGC3eQR0t10ZW{34n08gy7NcF?lzBDT~
z>>tTaozI^i2xd900=XXg8hEZq{$3(jOPTfMD-F&d8YMDit}O12*=z5ZSoMD)nv@JM
zUn&slz)(3Ej{1ji4xN&A%->AA&NOpNv~(XtSbpH>Ef$s_PQ4RL+y5BKzJyib%YES@
zqV3}GYSkfusrf(O(><s=#<O;z?P1P!&d+;KW*3M~*Pz*695C*xr^Tr6A(;u-R~X5*
zNNWQ5rX$?_d~X7N>w*}se3}IbHmA#-#syH#<^9NuX2Gz=f%Zt|I$PD-c=TqpX{$I%
z$0UO1phz)g1J~Q7Iq>H}yMvyl*k>I+?`{%$VUieu0+sf86<=!Qso&jgu}?Y3oxU(Q
zzjMTcUH5vv0=u5b)HKCGptrtp9>!rldvpW>=N?r@sRl3I-M<Z1POfA~tgVcU*!rmX
z(mCbG8?Nf5OxYiV+<@2OrVcb@&|SUAYs|J0GS<JZGIMP4$Q#49?8cez_jbZdE+sEq
zg-nfK=C`y8qJWm&+w7Iu5R6{54kop7laLg&SJ!`8oICcgU;%Eikv@~AlZ^UskbDU3
z{`JK@c3pgUrc7$?#g{)-=JFbd&`*Q@SjpdVmsw6|8-c&?LML(;8$^J3w+q2J?@Z>J
z$)(Tj9gWy~E1O5a4E1Eyc;Fdnd}<R>jK-eQt!{9o(x|Nj0(&Lu_3bXi>dF?hpiw@Q
z--4#`@JLU)mJQrvq`lY;hN_ZXoT;N1uq;@DQqYdvaIqeEbUy6r8qC0sVXLN(u%mq*
zxF&J0^n+TYySKYC{us(q=#623j~{Q+^|5D%Waqw(Cbw~(@a9S8ZjgbX025?(s)x~>
zaSvtNW0Y|&R`K;wsEDfAzSdqYA^~W=o-Q?VLps6-ikCrip{?5Brh&El>=x=ya)Ps(
zc)`uIT?i!_A!>S2Fa(v{k?^wx!X)-y=YwQzK+6kZg7%qw^VYqkXnNLcE;~`e8<5(G
zZ5Ndb+omuP_pyYn{EC$NCPavM+7{lh)aFz^MD`kO@%2TZqb#r|AvUC9BUv(eypa*J
z4^vq%<&2U~Ges}k=@cC!(B5Ldbx-XC>cp!P%c6EB97_PZr}kf@jNdcbI4E|5FR*p0
zVZ=zSmDC+AD=N37m8$kI>U@AjlFF$^J~$qC6q?noE%2emndDd`x?omX39Gd8X!7W|
z9Er&QbXNU~zJzWH(s~+}H(tKTsYJ()R^8jz?|4!?uufFF4Ry~HyLQ8`xx8xka=3er
zB;V|1vnF9QW8$8jgT#%fMwr+;o_Iq$^7+F_f*Zt5QM&hD<pU{T_x5Es><ZHE8ghK*
zvwW06ZTNyY#3I;%WRHqnw3mO9!%D>JsJrJo<WQ*`l^DMH<DrHlOOjk$ZexJ5=>Eq+
z&Q&zt>8j`Nnkoq-5a%a9VA2fvew=Z$1X2PCR1-@j)Rm37%SFLc;4(vVL)VOY^W1w=
z<7e4UTdK2|;C7c|9arYi8s{dgDuy0r2>;1>5X8Sv5i^Q4^!44pI7_T5tYBGF5NBI=
z3>kxZQKX(W8G^cqfn0WL)D|WgJ1V;DdOrz0G0BRoPoZlrM-kStvIK1>1$p8blb8@(
z*Fto)Ozhensc_`=E8oTl+_uUok>*2t6Z-ID(v#h=Ta>YByhiVGw*<VsK!^S2=`8dQ
z$003C<vf&0YR&g;C}+Z|zSw9+VpDxpLo$_IYc6)z_&=S+oa)Z)e1!_5n8EfUwW<8=
zZdV1?WzeDzG?NK;L50kG^f=w935DZ{9De}ZS3_lYENFEm=SEFs>%w5Uw<n4tnUp7$
zW&5ep!ZR(2mC-L*>ps%VAL(Em&jQcAVYz9lw9|%L(>VFuaBN^tgTZ#i0E_6oE9DEN
zL((Rvj$53=Cm6f}18z54b_N=DKX4n&Rua$YmpBfUzBKI$+KE)SADZ?FcK^L+ltyU6
zB?L!MJ6qsGDh+_a=AN0^E*#|3he#qxIO`kiv=0)Q5yU1%WlU3Bn7#Ze{^D-<yPk85
z1;-w!f|xg+UAvK+E)iz$-#JE&skTnB6W!AyGf({d{`_>(vp6nAPIeRa8{8}Oc8J)R
z*^O)@Ag_j&MjspK$1AnXLtwh4nw)-*bk!R6;~QQSXT>+L(fFBcz9)*l`l7wwU}*?j
z^M;m<BDv<vWzpm~*Z7cr%$dK=^m@qv3*0j239dPwWSKo0E3_gX#wgXjIZC9r|4tR9
zCRt+W7UQGd=V}5o;X_NNTl>%g@vt_S%W!^TwIWMabkzz^gAFj~^-yuG{o8mvJ~{SI
zz}MX-#)svL_sD)o?sSB9Qhg`7ts6x3;`j0X{ml~Td#bbLquyiiIM?I^<gtOHC;ti$
z@VIHJuJ1VJJN#g|J(%7`J~(b~KHs?h+hQTj>i6bd>N)XJx%=I`IHBl4!-X<U!!D{l
z`;w~Z@;|kBYv=Jd@Gkx{UVs*j)P>0bvhN#dmbnI-`z(@cERvsFBzsxB(MnN{2&BOi
zh}jC#OYa~sFq0|9<#!M;b|hyj2X$-NhcZFXfO`K)^?XVYy|-J0mww7Ez-@KNj;c_N
zVZb`*Ng-`CitDymvBeB;igJ2j<*y2>ji+rkyZxv5#qYIQ74oRF(yH^0?YA+T4`GPD
ze3EsZ(x9$**io7s@QAPX((V-tdVYqRVg>gat<J8Siak;(&J5^i2=2Ik7^!cb9A$y7
zO?C^$Z79gajHb>@OZ9sPL?$aw1Vj$z)jI*bTs<1CF5NSxCuyX}TSgmH97$fXXXP!K
z*VW)_*JE~?ecxHloI&Hl&y!^{^C8@8f!u4sH2Weh<;BBqZM4u&uooiuxhDg-x%o}4
zIzA>J%XZDZo0n5MoM|&PC;TYAsK5e?(5>#6=@wj<>c20?P!a$Qyi+y5xYJ$NWvhWY
zCs(pk!#C08EmM)l3dW%&pj8jMV^6=@Mo_wp$b){+a~GeL8VO}mYyag|nCWSE^I`b=
zWM!YH)z;y2->S_XN_6Oov6+h1-YQ;83^(uAVkpVFY$pLO{acj@A}Yph3|8LnEt0oH
z#IkFF(tq*Eh^GCDhZM~vu^3yS6|3<tg=$ys{&Y_iPcT0PY(KN9p?6)u>e18PeXL+A
zYAt*xx`<mZgoauuCi9HT+ph}=OS0D*ZhyNh=crN)o#G~$b(l%VZn84&>4;5TRp(Z8
zO}EYGhAyV2+J;Q`&`oKoPfdFj^)AGA#)<uM8-u!*4Y|(ABfqjeB>#xF4PWwnDg95=
z?+;k{8z|KxemARP>{+x|l=72DweiYitIRqlP`97PSagi(&6}Wm-fo-<4^GqN?}`5q
zhASsdq%74rBh84WuJZ*RW>LEeYrTqP(5%i0{@Y@;*y!wdLI@4`%mPZaPX7e@jlaDB
z<S>*-I#V&$q$7Fzol)WQ{^$0(#d#JTj>&0j*@IVh{5-#2AH8w%1*?4nIi-T`-HPjn
zMbx(&-a3K<=RJxDn@zAVfbQZ(0fkuqb(0E<^z7wQEgu`LDKJYNH}%VmtP`wL3&POx
z|14pdkc;dbSYSjegCTpkcG?aDTEo`k1WKa5uyeJm0Ia&RQC~ZS9VnaAC3M9;TA-Ud
z9}MIv7q2*(YBGKfopLx!5=naZEW}l6kL|XBGqnN7cmb~!6DD}zsK~G1p#+z|7<{%*
zv5%-u+`nj@OzhF8HOUJzwl(4xXQ&&Nh>Q5gANY4=htuqLxY--%MF?DUeTY6i+`je{
zk9Fd*A$oxQb~`OW7Y$d>VGXNCF4OMtI<f-des)+OYoNHBiNFaFng7t|+9hI&_>S%>
za>_84w)!x|wpLs`dGK9#Zl9}X0Gl@x;4ZbURQ<;eiRu_HVg8#F+J9ZVN$1cX(^2We
zscc;Pd2Va#+L5Uqu8=n+y`G@#acui!O{yP952K~Pr8l}+VG_OUbc8{&wBU-7ci8;h
z$y4pRw#P1t?>T4YkGOgPg~wODf3PNb{@_*MOvz7GmVX=;r);Ddc8W{wC?O#p;ob05
z0yLo}LC)g)sfX=-U0tFMiX$JM^cl*ucgirm&H65B_m*YHq5X~i<+qBoMX!R$@1#HD
z%K1ieCfi5=6<nNKLUMcl>GR9a@2lpiNx4H_`o-Il*Lj>9(ky80oWv?e$26C3OP7>s
zrHXR?6rNcx9M^1YmCJ3eMrg}=+9%&w7p)FNYKJqo@pZ1fD?dA!5P#;urRAwQ<|oT)
zN`ZG1royRgTBC)-D_(UcG201UpQ<>|)4Bf+#X8H=XraqyBVzf|SR%~k<uC-obv?~X
zrA9b@?D}_$#G;t_&vYhrPPx9RD}942ju&tb5$gotCe~>!m5yWb1?gE19pS!k=3S7r
z1a^Nhn~{0PZ75l?u=io!Ax5@<HnP!t%42g7ch!bn4XW~wXM3EgWaeLdk8=JzxqYr`
zIYgD}YDKTt_3`J<Q7>}^ZiwIvZo6JFtG*N;FtSAKX?$lvd)|9ob>>oqX=iOI-^l(i
z2aYx=@NlRfeEanNO*6X^D=MAsO7~YO*_B)Bxcj&KeC0|-X>DRvtc#V%FQoor=;iM`
zQfVa6S=txhy1(?%yP&ijCDchcb(L)J!8wUq7ox>uTV~fExrnKAq{(dfyX7C`Z-gn@
z=U<k4PoZ5vv)M`U9qmMcprErtMY*}<<AL^2rgev!CJL;G;!fBwFTAslWBSgv#L|Cm
zF}#h$vgOj|yhjq{VT^wB^40mbEc^CZRWosax;2lB12F3EOt9wq7pO5ECa;P5Nc0WK
zpO;S7#*SUYG21ZP3Ebxg#<v7=r7Aj(YEIpj9p-5FDgTl(?c0C4vR|vtm$I#^=br!N
zGUgs$*#nOBm5)!-`c6`mp5Yvm^^rtmwruIkPam;4t`h4`S{`jnsrB_FLZ4K8FJgRc
zd0_b9;>TyGaEe}C1q+O!wD_Bx!cTna$`lC<efOe`k1BQhc|J@k>^BK&Tn8GET78-;
zcGE(Jjbwc~v}Bd|nlx9Y)BS164X004gYWjP${j<=R2X1YR~9~R;4Vzkt!65*iM50l
z<iJAWp*jJo`G`q$)jVRX;zd6$szG?l)p)w&<?%14tTLXO44JLlu_ufkTtW4deJ-3b
zZ+2c#DZ`i7c*uw!na#AFqe3L6p`t6<j_!AU`Z;~P)>Wnz5660+xiKqkn#)7u-0m<j
zYMG+-Aud-UDQp>BPAD|#$Axiv1BC(cgF3oq3xgP)gIq&6&T0xwn`1IU1nN&TFI8`4
zC8NG_Z1{Q5s}g*2tS=iO^n6NFSqTCkQ<;1FD#7iyiH{IiWaU=8T_z$pl~=L$c^9=m
zYsUYpghhu0MbC3kZk$xdLT?S}6n5ivg)WlI;``HOpLbGC|9Ex@B5+ky_%*4&4=;6m
zWjxk0w%}#4M7CRO&WxI$dCl-^68GoL+4NRkrca*!K3?H<P}Fcj%?Hb089J>+JVZ%T
z!Uv$N0&zQ5;yfLGP0<(0Gd=;3MC*!mxfo1`W|wa`iamfb-$n|`DKd1c5h*HQ=LrlT
zO6oM7LnWLURHb&NQ<U!BaMt|AYTC5k*3Ut#9JT^SqH<yYgQ^&7vN{@V0n~SLwiZ}{
zky-19_WqTqe3BtQpA6OD3qLH)7LdkcXa~Z0P5OrbJVW2nsJ{;%#0qP9;rms~o18t_
zxmO#gy4mnX55qBAfkwfZjth~HKX?YCwqP(aV+U_dX4nOjfzcBK{%D!&4`{@VAXdVd
zxNB)7=W>R<&W)RurTP)ob5R!tgHU1X<M6Wzhrx2YfPdms0Z`rqk~heS@K_g=<%33P
z9*K=B_p&Buo~V66?74*m&_}AB;oo184YPYYR8rpOu6-JHcevi6G%zmEDBJ>v%@1Pl
z3sx>&Xy{u@8TqpFI_FQ@c8v?~kfrONl}0_*la%~<#=R`7f*`g|z@HMNcdiaUqOIlf
zz`K<2J;(l=)*Wuzk6eeh?nw2{gsN1iC9LpoH;+Y@{ia8|YT$iaso1-O<9s3AsdI2`
zrYeZB>(rChaJ{`KX|l&N;-}WSlUllpBxDR5kKqWSz}&wi_w5RAPzzyfOQU3qvJd~9
zt)^iNI6=Qqak4J|(k0G=Q|z~mJCa~8Dg*ayOUeM3MoYA`_>&nsfkEu@W3=VW_5*SQ
z+gd`+43&fXbM}J)Z51^IGwTA8SD1emjJZ;m#_J%7v}w>?Nk&)n^SyICYy3n&L>;k0
z>W@=VxttIvR%upaQ&~Ekoh<t0JV(Fd-JE+--As$VX8Ov$EFU&fv|=5>1tVod;Hg0I
zkHf>Y%nbFZcNtvj%y^>09y@7{4ZYJZR6MLIFCm8Yb(4BurAK9U7ppy1PL8%TUN~Yf
z^hQ5Ir$1L+`{$yCl3c5%>s!*ZpToLC-#PL*`ScAgS~vT~jp6iP#!H*n*q1*b6{J1z
z&B<rFnWhyxs$ZYd$^BjVQ(zE&7Z~=WMEfTgRLwt62wm^ml^3*nM7=-C6uy~kq6?3p
z3&@VQ9ikncjZ2x>dmR=iWEU}a2uo>KwC%k+^ir7F-mCrJr+Jm&;2*L^D81Y{jP`q3
zOa2jZl_c&31kfU>Pk?AGDd$VHORpa1<s2JCT6KlE8h=-erfaKm(zP3uMC_rX6Dvlr
z2cBzVjYB18GYgxWg>e?NZVTpktYmGnPs?_`h6>fB%%vjGV}zT<CUICxN~E}1Z2(|Y
z>`nDu+))NomM<svpfb4bTowE5)p;X5VpLZCLN*S2TqgdhBrpr8@R>E9(Pc<Sn6FF+
zuzh+3*hFlgK#}vxkiGKtZI_kZcq~u327b+9ihutvi1*gRCJ>#_VAo#3zH*v>RyYy@
zM`UdpGB#zadB9cW4r#P?)d2bL*pz(Gs^rTmrW(qv8|m<Kk52gjr|?b0+bss7=IhTv
z8sDy+g>JRXb5rbsuF9@1Y6~WjJQ^B_{Ss&1pJ@dWKAYo1d-byB-5Ix}sFLzPWQR8V
zqNG~+vJCUIZZe8RPUaRden_oni)?lv*Z$uJEYWP@A3?~PYcOE3<(-)u`0E({%7&MG
z!3J!wEy18o9UM>$_Rdg^qSdjh7`pKzzl57lQFVMSY<A+S+9Iw7e|t`hc_T#`CRxc^
z@o5pzTDW8YSsBPFSTA;O(W1RbPz9jn+2sz&Ggt;J#$O0W?1m5w2EocKMW&JBBlc0D
z``tV2b_>AlC2J1qST<7d$wHD@5n_fo8+&_(Oai&Y0u0eww8;Hxym!#r+PaxI=K$hn
zm6dD?>!2%lWTJSDh&iNx>Q}&Imm%4nt&TI0W1K}JBTl;Hv63BT?iEBLDGFUbIaJFe
z9G#Iv#te|HNB#b(q&DI-N;)(@S){I)+=MgCcDUKYOYTUbQJ{DOZ!+shY^0e9f35<t
z5`p}ATD82=_?{@`yfv|EOBhw=DwmpM=f4?{zCQtPYJ@ek?bl*s>6>T5pNFV+dE#sp
z^VvF$5A<HiXxEC&sXZDB=s#0=t{wSraqdWdZ7=S9j@f3u^2wj4uTMC$2U9Cw-;%rK
z===?v-XvT1(sk^d`|;O3NT^Iz&X)9KD~uZGJH%ZALSsB0JW8Cl7$`@ZQH@;DuAjeE
z%D>Z+&AfA&^Rg@7LBz)lr^7?A%kvvC4T%PvgoFK#{ln6Qp4?0xxk%mEq4l9}hQ8G)
z39sW)wv=|NIxaG%fvsPYyZc!S!zKp2Wf=Hf?qlg4Ky;vchcF`BrxRs2V0W*(0VZ@U
zzXy`UPWc3$x>7|2cZUM<Ugr^07nmp0JI^j3EudSS1FH9>EpqmAH}rSL+*s$d;55nG
z|M(ppU~p$G0)zSn@fdTxI(tR~QtWQ8Spdp>_<m@q$^nKDejdgllOL90xTmhhU!wgU
zHy#_5cC&JpwpxEa%KeX6di9BDPrvQMH68Td{E_J=_{<TX5Ui|nXEv)XMebG5P;%q%
z0!7fUDVW|=;H=Og$_f+(+3)1rq#0g1-DOx^tKP%&gm5F>$?&~aWzteuQ8A?PtzGpo
zr2g|TNGy7%TCW#pf=mvWz~6q7-FJ_p)%V6E`^hJdVw_HLu426($!+csnE=Mv+{$r?
z0KgUp9V6fgePos5)1-GmW;5YQdw^tPgD%a^$^g1=4P%}{(7R{-Di30A(<16+T2;M%
z(GWp)pWx|BH#Y`Mj*CA*`#PmcOq7wDwQ8BfWVDkV6-auMl`+I3^fWxp>8{L^jx;B+
zjN3X?>SE#GNoYT*STI;{LkbR3%iv>s=fK>s9Jf*5rA0Cbb8TzblGaOWYoEz<-k~}S
z9LES_LB!bFg^u+oKZi3ix3yQeyJYM2;@wuAo+tGb1K;uyftMn*Jg<UK&t_=Bn2L^r
z*49V;D^0@UO#0^s56|1fcJRx2TuqwK@Wq@+FWU&xqGF7vfMb1CXsJ79()+}dN1A$*
z-k<Yd0TOj)zMX?TMx{t>Cn18*I<~@M@oZ76iNM=J7;|^R&^6zYV71XV#(1QaQ9~zA
zrlJ(;V#pq0dxE}}bJboC%h)VEY|}b1pGq2Rk}G`U<=b2L)@jS6iS9HD7HjF<H4(Tc
zLYvX>6TBVo+QI?5e>OHKE4%P=zfF~FQw_((0Vw>enM*qLRIMT5)O@X{F~6ij@8$h|
z0@-H5$CHY;o}eF9@vT}UO?EK0bxjS5d_GHcC@qrdba{9wV=EXoRI{F*r9g1#yIVIE
z2<4Cv13IO*$ue6ZOb*{V?!`T7su<M2Mk8DDYt+MI^dX(cVBoN6P%zPff2pY8GJV31
z^d!^%=O}}Dh+EIMnVsnw(QLl<MRSt->~aJ$O3XgfPhI*VqE*8wRq+nA0;ch1WM8ah
zE5G;#8;@ka>rYj23ISbtFpI--Qo+FjfcBF!BSk}eLmT(0#2S<<#)2#b-G>M*jH9-=
zdy8SRD{KA2F5Ka{bjDjrk?zGKx^m9?L6r+ci@pO=$wAwu3u|&=LbOLzx*YSo<(doa
zBGx-O1|;5dJg$d{F7u+@d#j3m!9C8!lX^HCkweE3BYxhh*he1J_B|Btu~nJ>q?!F6
zqzd?LjV+LVMs5Fv^b?0-QAFFZu6?9!cnC!r)N<H{Ul6EMVTxUn_6~BI{eC_a?(H7~
zFcLOn4mjh(<a+92c0El8HH(AHirU^p%EJQI!c|_c$?&n`k7(?k2#&{rX1W6yFgJ%_
z!mHHX_yNmGety(5)6bpNPhPs?s=^Pf5wD@NSq>W=hR*k6O-i5dH2U?cj5JZ4qj|#f
zukI@^5*uvdfLzZ~kj1-3_}WZY!Epi|uh@FazRK||BgJ4~=84y2tuiE6iD_jzL!H)`
zS9xU{?@XLCIUCJ>OUd(92<`ggZz)`Uo|N20QN+YWm(l3Dli@UWwIWn#O(I2&21e}h
zq0<XM@)l(ETj8}VxsT35$RQHo!r36)L>D_Ym~FJ%>gC#UYr|xyd#Zz`URK6C-RWe_
zAcyqZ*G{PyS5>V#CI<r!e`y;J5zC?#OIOPzTQ8h%lU{A@@aEH{*=<_9O6VC(<SmV{
zZzI7!;W$O=T?({onaQX2aaHUjn7B&{slgxVsw1c|WlOxUal^UVIE-RcE(Bfhw&{{>
zXeTB@3$oR;S?AtT5v+biilrvvBW|H|6F-B!E3<NcjR_s9Qh(e7?}+OL9Pjj2_~Th7
zGmW{KStLX03D%+ZbBmeJM6{p0v};KDzHi8JuxnWE!mu$YyyN#(_)JbOi#3&(gcSq+
zRm^MRvV6ixajRY!QA$CJX<*}huuSfy%COAn-2`V|SH4`+E01jR>=wij!v?y=5^*k5
ztw#II5u!rbwF(-CA&Q3D_Z;`Ec&o@dL}f$jD(Ot_fA3m)y>-5q_JH@9FkcnNJ$%HI
zxD(>c-Z(~$5BoSs>~Yz*xO|7ehs(`HB$dg-N(*I{&;B8``p=c?Nqn3L=2hhWn%5cs
z;-`#O6HuD76@R8}&?6>Y@tqcQoo=H|Zeb%oO(Q#b@XMi^G@Q9(h4O$+c61^?V<`l;
z)#$QPjZUI8#^HJ~s~n27nx0r*Eb#(DVPd1-bo|s!Npx?MV`=Qw!bx`Yz3|bMzSUMp
z7+_5{to5_!AZKT1O+MyByqDrQ+iP*>;J{$RQp%0@jH@fp_*L$PGnMJ{bT-P@wh{YZ
z)d1W|{+f0ee>1)UmG=dL>lj6)<MO+dM4Cj&yveF1UCityGEm>vl2v`fWxiNeLfzs%
z%Qr{bJV=dV$JCIEF5_=@(h}fap~6jVTS!qewBbyjFXxqpOh+D=hrcK^UveF+mG)BC
zo#`nJ_#RMR%!J2k8b|9#e)O{|ixf5LvEWAk<@wKX{1>s9uYT&R%%3h7<mpYER+%I4
z+&Z?WtUKnsM?K*mMm|($E)-epRYi8AR&K1Gh_Bd7qJ(elh-PSp90f~59HyomGAp|q
z!p$}}O>XMCeHgUAK{51TU;jq)wKmP(56MX-(-cc)TPxYcl?SX%b(?0s>!;+NG@w=b
zYl&Fmm~yS}_^N-Uo5w$j#aq^Ib-)a@S`FfovR_qazZF&5mJM5f9avI6<UT5hp1Hnc
z^FGMTV=cUm+B!&_{|q!ASBOzTi&!T-BYXa#WVgxopPN~QXJ$`bu8v=PvGGTm65>GH
z-=30E`cLw6jSx>5aD?7p>}DmvVoLa^o{+ERhL*MQ40D{W78PdKqz(8r@PK4=nI>40
z8-KzRW$70XqrzDVbfCG3(Ew+7^M@yE`s4q&O0n{XR6vkAY*yv}OEd)KkC(9;YMI1R
zIN&2__L3<=m}M#=BW?Tb=RGpVFt2ud%c`hnGG`<-+w8$AEYkA+q5o257P}BH0hUvR
zZv&jC#ggA4Lr-*z-8JU)%gRaeNxB3zb!I!K|48Qj%7b7vFTAda{)N>*hOv<=r~F_o
z2u$APT7fxuXdBoB*fp=EcpMA+wFWCU7uW@-$HM>-sPbUUi43R_VwbPjK<tN0cw<wg
zxqOk!!3s0_4Ft5-pv7mrD05xQp963Hm%rTJ>u~QaP=Xghztvj`_LbA0w6)aD40i|n
z=0y{?TW#lK2G9QK6HmwhxY)eQc7y0Y-|bFB4UFU;AXGczJPVGQQiPHd3yp*|*hl6A
z_VQJBe@0`|YPBbK*$0I$KfeUqx)@<GNC0cGAG}xSJ|Ns7B<i%d-F>5I;?lz&vOh7i
z=&FIpblyEA!<pkhKD1CUG3$G0aB!RlXUNPFIM*^HFgLL%9gLA48ZnQA-o?Pm@f$#f
z4?gpk>HXh>04R`YeDnD~bF2Pl_WutE-B&a`JE^*~p#OYy|9tzf0RG!QNpc8ni{<|z
zi9iKeZT}BR1P2Hq*(3Pb?Y}v<|HHTZ!vkQ&XF|iy<It<P_20kcfBW{o|9s~eP_Ore
z$?yCA2}=7zef1QHLJ0l0C<HoY0-!U?xPS|+4Ty#d-hel)quA{n>q|!9Yx~F@cvTio
z@mK#uhW(eza1Tl9Sn@E>?$P>d<yu@Cl{?(V?SnscS(Kss05AdEVl9#btO4mL1sFn_
z;4TnHES##ZZqWXAphYZWhb&S;tbwA;WFF!IB;f_nQ<uHGJ}hcUB#~yavFLF)1am!!
zT_9J`pPz)D0jO(@M=?e4r}V39FyCfcx~&i1!1!P`9L07#-E}$0usGtz?Kh4A?i5=1
z@Oj$FMq&k_k<1G@bwR`<Gix6qFe11A;0Mrt8@y-iCSS|1mwW_2d<e|%H#5y(rdSHd
zU4=h`-CQ~A*98U%fx&e1KRjz}%^@GG_`{nj29OZ?`)PQc!ow&demBD)kwye1zK(MU
zCZy+O5ybl#&|m{B-jg!RGTgCK&<)xH7i|a#5oX}U$zaak=#H4u?)Ft5P`4v)^)%tc
z=&&myEu0hCO#6FDIN6<#RoiV30b>3@a|jX(QD9cBd8%Iw3tAdS;>h5U2%RTD7;mv)
zV__3`I-M?#s919jyw**D<m<=(eyPqzZ~sv6!l>tw`J2@{5g&k+T{@i{Gzi~&=TL*Z
z)<7=*l=lOQPaHr_wsnC4;C;Jl09{J!$));1R--M|yFhNTqILBG*MSgf$OWLv1yaK;
z#j>8B;pWO_(6camxO%GR6*Mu}!+F%J4+LhFq#^O_*o3EZixp0hzy8sR{}1vFudyqz
zc*mrj?^ph{`n55m&|CW2z$wpmU-xdfvNc#Di4%At$UP>Q{W?8<8h{~%$vkgkYPoF@
zLuS+xNNb9YP)OV6v)s%&6!h3u!(I}_8%3@R)2846VhsdpTBtbu$%JtQGM^T40i`87
zN*W?Dh%h2x3o!&_eu~{W!$Zl@l8Pi}e5@`qy^(_a1#2+M$~?fc>2+-3is}bG^k8J$
zk1PlY)-+fqLzwf+E}56B^H04!xVf*#jyl*VtIpA>ggb~Ht%Q?1*cvQ&;D9A@1@f3r
z4du{oSK`6RjV>Fv?OiO&5f12n<Zb~9W-&hmecD47+@xTKJIje_d$PIoGs8QT4-(Q`
z{_D{RhlkO5xb<z%;}Qn{w$r9Y_rn!A4@M$E*lP$YPbpPSc7xB&N;$t-Qyd5J4cJyB
z)eq??n>(}^S)RZa7zAgP(|AT4G3OO`M)F*Lp0Npd2e!lIz@L@ZqTeJiaGv2ppn>@e
zx&1e8>sf|On`B#yRKO*S#r}xW_&j*YwD-0Z|1K!J{k2j3f@h`1B;s<p8drgLYZ!nM
znmdQ)*jDo5BlOfAwuTwoqgY6A;6$}5(I4H7&o&-@nsL*kuT9hr_ooB!xC&B;H*0Fe
z{&m$PncyQ7Y!0(Cnm*c?q|5NHOVcm(j|cQ>^bK4krj&UIP!jWFyipJ-*dPW}Hs?uE
zyS`>T6LN965%AbTdY2x|8RZ#&*r^;-znSw#$a(`Q>}1dA|6TgHsf>c$znCY%vF^L$
z1epuOn=EK0?z~i&S}2wPpWl8(QvSP$7OYExgSF*)z;ixZ*qgi|G{s;<Ge}#t<eR>}
z=IpH<X4)JYB8%krpie(Q4o_gd4PFdAMUO6XcDpFG30&f#4Y<x3Y&?GgV`adi9-s6q
z<!-70z<^G7Zh(~~Jjy0yI)cpy)rMB1<5!BEuB%RP-px~ZY$+`n9YPHh>;^e0$m`;7
zdil}48^M9ZwKoBRWIXM->btwuca^Pk0*;t_dGA*)W1LK^5$TUAkqkt$DEp)rPL;lO
z(L08^<Z(>!6Vx%*`9{*wM=%lWC7h;Ph;N}G`|VIV;kv4|G(qur9u~JX&F*@~_rUqx
z04)yBtyXAHFTS0JYvH|4-^8g+_U@h4#KTh(!V|%w44Ylt*V)TXXw)3S;iNMx$_<Dq
z;=&tjBW0F3NxM1&oa8L=M~DIW7s$29-UqMGV2?OVCcz7^EPh(UoAP0+GMjC<^+^+M
znMeKfX-}NwEbkr5w1`4Jt2?f|`xZYPtY{s|>sLi<M>eZAKdAn6ff$siUXif&FDj_J
z!O=VzyZZDc@z!J3#X2*+xjSBF1%?<UOxQ=sKpQ0Lg~B7gVD#L+8NbDiisk)n@#qr#
z>Igx>sx6b^(^3*}hzx=IdCb1`yM>D=tVhGhXp!T&t#4DZwDvVSwNIMO2;_T>n|v-b
z$F=399r=LM9xJ<3XY*J@PUW4CxC!VMvh~#~Z&hD%iHi`CdmeAPV8%4DkFD5NcS~L5
zzRkYKu*LXYCx+Tu@{WgN9;jh1S;K`c>`GvzAyxG@N<3S)@~C8j`9N0n=LDU`o$#Kt
zKTqU1;P%>mq}|Q^_f7r`<^2pPwC$l*@HYy%Ln-dI)0J3?(ElW0X{+*o@?R}{{E^Kc
zbMo{Tq<Qq)%9=M_gIUmhgsiGdEyJ=43Z;1%Y!;6I8=UR?c>Q^nciY#3`XMn62n=L*
zz`D!D{YDlE<pBs#nxK?7_<hMIOEtU{QsDHvM*657ftNe)mm``i<igM4?o_z)%n|2&
z<CZbu5U9V#Q`{^G4`r})M4A+J6d0a8(<{c{X#h@KMX+vrfp$6FS)od{8)Uebyg~fS
ze*|dDaC6<CLTFyyhp=ktObDBfFcT|LzRU*ju@rWkKV6EB%=VLs%P2><xCRkHstwn|
zGPEXIRKsjoAW~FYRqqZj&a24nXxb_t_fp+hs;8i+)v9Wn=iq)1%xiXaKJTlymT-?6
znN_pi!~PmT<XG=ZHe%m7fv%&c$;Sn4Xr%T{<GvDxz3n=RBN8#mGrf9k204%NTVSC#
zQnrw5`#p9a(~1_vX63(z5n(;pd*f9CDZ9?FmL;JT-B8eqgZa8(1Hak+EWm{3Wvi|*
zkjZVk<Vt}5O@xi6E{<hyl*pLH%x(H-KcJ^4R+;Z%izz>513Mv2hx|to8fBt}qzoZs
zMmA^D;CziqNs|F4`~F%)AJDnuQjq0t^lI_s{y~d|`w9DiX|hTmvelpx;-+ksEv2T?
zTh7b-0}RKXG<bl^)Mfp3BzRy;n7r<kTvgAFJ3abc^C}tm@~)?>D6CK>X7k7KSgS~c
zd1rid6PuJtj9kA1soZi)Zl|q$$)G;kC^mJ(z$Ad{a;1Uqh(Jzx$TMO|T@48FSc3y8
zkIIfc`w*^CCRyRtpHS)Dv~24?HAhi}XfJS^#5?Wf`3FW(TRu?yTxf7Lix{<nbjtKw
z9+W%AS=LU8qaXb2{2eN9;MW=S<rmtQJjuT7(Y*ThmfWCaS?(VTk=LO88ok!pSM>Mw
zqK1#Lk_7W7rEFi`U<)Cv2IX1G9Kvvt@F$eVA<0U1dHMYoug^8&HeO^d>q@{~p;N$0
zg1IJ6M~Eh6=S?1L74Or#=)*xkgfV8{p)esx7TSj#a5B6>qhc@IIy`t^facsvVeq0X
zg>`Hoa=pb~`VOiyn>T5ZffT<QC6*Kpz^f|%8M!3yU{w=?RX@zS(H&!0Jq!Vh><R}1
z1LP19R|VQFF)r7DL6lMT-cP|0$w?6B(rFR!VNm1GwxDETK$D=|-50D3tul|+qS(u_
zS#!UEefnh%se64t$^aOy8V17{Ve(YTHtWCw`QTc|Ydqsg_Mgj7pbR|r*V%p-N=mE+
zz)Bwio-vil@doqY76d&<>I-810^tTKh@R;Sai1c!51^ZZ%Ph#k1$!0iiIty)F+(u#
z)mGuj??B(Du{2)YENZ{C0I4hJ@Nj!dndj31{REm)==&ma=oTB$RbH$NBHT4iU{6kw
zhWVjATw{KoX!J*UGu|m$6E+_m%sou2k|8>k)$y)1P;UGz4q(uJKb5(;3XF9sDWz!$
zPwQ+|8cbfO6J&qR_Pj2V^2`aRtXmeFpl8=%h%a)q$*2w(_uZmhCciW907sr+&OO$|
z5LDQOko+Lt$Y7!>T0_4O<0r)0g5WRK$s6^dUq>##HHj^(3cf$tvQWp@Q54lA%|yte
z3aU~1hqOeD;D?<9YqE5*><h}mbgO<&7&a1yIiloR`^!A|@)&z;&#3%U&GSnt1n)KS
zH-sL(q3Aes5V!qt@S5G2%u$x0yo=dE@h9Gd%9$8R-{L?!iEbjL*U_;%)b5p0vbD2F
zDQD;Bz_-URr+&kcXX3mbTIq@G>cW8d0F7LGC0dh5*CRTSI4$7oXUQfhqN*2jquOWR
zx1R#yrKX&T!mNVpRZg|<yN(hQA6-c2E_@403lz`3WnNjGwdSCc&9YE>r|+L&)xR`-
ze>FihVJDU-X#b7Eq4MYwGaJu#^S2-9k>?)AbFDs*VxD_^Q)iAp`_56oNxPzA)eI(#
z_g?)#ANhQhY)AMoGf}U#bLkW=>;(|RPCfov(&O((GAKQ&>Ga%V@2bG;CFbqSF@ZzS
zT1+@SM&GT!KwfSn_%8GGJ0ouR8@Gw}Aw|7BOiwRd?^?K?MYME`=CG6dk{+!W7p+1%
z3uL>rq+ILKuaB;4;YcN<p!MsHf4EVv#gz{E4@m<veiH&$)dX%`5s<V#?;rn8p2PS|
z{7Kp?%CY*^T`e~H`g8d4j*%+<mjw*4l15YLj`4C8smlVB0TZnpv`bvH8)DL*l?7S|
z1zQMRTT%qY;~dZzB_YPjlpN7Z9lO&;OM8!#){GNJ9<4|DT<=PZT)ZCfbLkN4m5)9x
z9Mb7X$erC(?EbXh&ui_1v~LMqc4oXwepsGite8qJ`Vr3eg4F9p(9rCcn?Dzw<EO+v
z`HFVX(sK5dJkmQ{;<ardY}A+Q<?!%rub&znS@Ct6)%;W^Afs=2pgruK2h*3c%PGvd
z;z-h4OQXBp{sfVGe`Bq2l)%Ra=M8-qGp;usG3<WpNbvp#d(Dlg?)bTeO+z4aEa%qj
zHTX}GuHqbsX(RkJ?#kFJ=o~!%v>~(pv_C#bs0+SZtX}qwo9E_U1pRnHivJ?WSL>1b
zOwYqUnQVU+q*vTJ_Psua8iupdlD0)%EreadDxh`v(!dZYn9XxQ#QLb3bblqoBvCwc
z6HR+L9U2I3C$ZWJD@T<_1(rK2QPpuqA&2~twYTd7+byRY1P^lGuBXJ?&ocZG9=nf^
zalZDClbr4%!IsdRX6$nnf6J}>*u|rYj5tY&q<l{ozw4;zhoUT{hwy_M{M}@oS8G{n
zYr$6`qB^_N&BrETR*1keA+TJbwXt6MzQzi%O<KG1*@=#@=PH4GGVKboiZ}*wKBtPV
zADU|nxvQ#{En2HwxSDQT0gsxLqg=SUm%o6lu{V|uWDxnlh_Ck=0<l6OUa=}vS3v(`
zv)mb3U5U;q$<A+Vxq`w{avD<Vn5ZZ#_^HT|jHw;m#+{^9ywZS9xGLiD_5q7#BW!xQ
zQmvDk9p(g5%nJFx680#phf^pnRvy(t-Ey5QPcVE1q!J6iCE<~Ty($%wD+n!k<>|<F
z{Tn`yzB%a(&3)F}DNG-TP0Cqn>D^PW1h+Ufd2f%OHr~EqG}H8uLR<}EY3Y|G#=qY4
zwfmTo6G`$TYiVEFftS#8>j-XxVG(7fxECCxf_J=j6^L(dSYDfIu0@tSEBV;<H3K2j
zIGya7w|!gpu#a!<Q3eSfYiT13H|yvGjyUiq$jylr#7)*rmW#hH=Z=MY2W8B<yToiX
zNa34&GNI2F&)$l&ci$Xee0v)%;*9^xMa-RavgRN{;ctzE9~EBb{pW>|V#%pp);d*C
zkX_`{>QsI%o2$##c<S_pebXmJYDX8&&rn|~nPw>tIq9Biwuajcr-^=P$4q~=lgfSW
z&IrRxi7@YW7$`;yo3f{!u_V59VLDlU6*mz+8@7fM_!=bEvVE~!wU!3P;HrHtP`ovY
zZ?A5<U2g^}@{|to)E`{yr!D>TPg3oF8HpNkSbnzMyl<qqdz3Fe)hktiwuJu8rs~ja
zT^Jq3j{5BFS$67G0MPF;7uY7L^nH0D(wXw3B)+uN-jr4<^)sg#=jk*0H8-=o3fRm<
zRz15TbuU0fIY%1T3|EmNiP<bG|2T9*@VIAaD%UOI$Tr>q?>bh6{CH%()eSvQWAFj+
zeuv5zBeY$vdT@(;eUCj&?DUgMZ;+mgwo-d@NLeu4p2jfjQP!uCb&>X>i_ayZy8Bz7
zd0BH*nhTXeB$QpMl3A|3mMVRqAlnBU-CBy8#Nzz6y$5;r;@xMRs#K<fiz`~kZZs0=
zk5#Y8rtyumKcT0XFN(%LKf+CmV>|$JvYOOL^L-w|3+|b^ud7o<cnBAb(oycvsLlJ*
z8emJxbnmD(wpO@6I3vC6v+KSi4>sOk#hG8ZR1~X$7ncXY;GqG_cfWlmnLGOgw+arL
z4^q<;Oaf1j%Y>FhJDj*48ufu_>5I<@7gGUg$LF_sOwSTY_b2v0rxf>wQuH$1aFHjH
zg4XYCFOEDWOYKkav(ZNcQNKIAZuHl|G98LC5yzdqhmAbCQ|<yX(4mWrHg^-qd!+Gi
zb`g<BpeXT9vb@>uUzk~H<zVz%YDj5s^nZw<ASS<kOf$y*?K3K32BTRtsx76e=hu(z
zgo^{<XYh`4uT`w;O39P?_n`Y?i(A7gv$1DO>8_X-G|~k9He0VK!V+wK#n~o(>I?v1
z9#54$lJ8g-N(B*S*mpNY8t{k(r62b?@`E4x+#AidE*`(Fq8>-13~au%NHSFG#wBWK
z8nz9LPsdwUd*$f4Z~`J_R6oV)87fra`pMoL+kOi7r6WVo5Y76;=m2~)<<vX<nkL6=
zVOp6|iRs9(FVcjL*!<Y-VZQnMnY!$yN?#qI!a9lzwy7L?>*^sjQ7it83x(OL`WbmA
zMx{%%VKIk9@sfs{Y$m?Ej~~LM$ht@89#zuDk=eU<4+@l@N>b-_eNkb}%LjgLh0H}a
zSevHeDel|?a(rt_8iS%t=oIO=>Opi(m&6oB)GtvAnNN96KHbr)1Io`z{LI!J3Ktqj
zbl$l??YyUd=0C<A{{p)!DF96PIw{xuZ@D@J0XF&PY4r(?J!m|?6>@7-TLKIAHNrY8
z1=6qB2YX&%6Lw$RY9N}1<EbABPmF!kM9-1j2L3@OSkh#~O)(}`+!)Kj*8*1^Z{1^*
zvd?z9{DHC@U9bp#pA=qE8Q;Y<q%08c41N)M+|tR$j!6UoMC8hNi92b6Zl`mr!RcW(
z_otF9yOd9uPUKa%QSM#9SR<TbVxoU?il^t`Hf1#FT)u#xlwl#E(L%2&L7!(&^27>k
zC_(dZ_X|`@je!$-m7~>CzbVRB0wh!ka=k9|pi#SK?q>KkqypuvlP+1ZBGuCzHavsW
zY4-s#lWrfB?oKYrMvnc=6k0qZ9c24Lf>Tv=5pioD7&;!e8Tywu`yP%FrEjudrSNT7
zcUZ&cj*8;5j?1!VmGZr3p=^oR^n*-JIf0zIhh@(9c&6T-K&@P9aKdhH)e{^{KifRP
z&S_GO0Xp#HKEL7CGeUOu6L=NF-b)OOpFbJp80+rGTe~bf=;X1)7vJgo(`cdQJdaTK
z1?>9YWPVDoFT{xbLJLO~CvNuKXvlANtLlT6l**F#_OHw$FC7Sf_$jWF_*7VTCG&Ah
ziaQtXyw2?p8Oc`;9>3H>-Eyu9d?p~Fny^w@%!j-95+PL-LpRPv%gKdv)p3XqqVBXV
zYKzpn`%2o|aQNLg1uacFfXE`qNYs%Jgq9Cl8l78w?Z~E_FN~fJexXuBpHH&Y4kX8x
zMvs;hR*4Val|x&)M9#?=+K+?RAXa#d_@Y_H8JS5m$%JkKetE8e#u2(^8A99PN4UW#
z={nP4VGX%hJ>u}H)%Zo=`u_0GFuH;`)USwr;(xe$GP^@S<HNv8#r$2Kv+#c35ntGu
zwmkbHW9v&rXTDv5&11CfN~X2l)Y}z@53Q@&3IYuQRgNU1^IyZP%Leafi9P=DF*k7m
zzpog6BmQB8u)3pNr20;TKAE`Me8~D#W=iP&Pp^9d-nz=Q(^*JEb?dIiY+0)FV)jeP
z+^#+^%_3Doa^<fBco3P*-=OrvGg^MiRjqFuFP&y%l4Gso4yFeg&{7J;JB|mGe15w$
zRPxFB+w#i}VGn6S+pZW#ilym_w5eAf%v|T4Vt#e53Xoip+v&iWEUak%AnS9S9X)@J
zve}RE3DHq2vd=Qf3&RF``Z{weLm@rBf10le`4CtWC+4X9gX8y9VHCEG8d6b>AF(;(
z)dB7R;Q(1D6oUs5#n#K$l&F<(XJ{O)z+Bwc7Vd3v>g417ZzKfeU8g!8-ZpT9*j9{`
zwDN;`yMbnVjh;SR0V&U#FDX1}ziS$?+Id@>AbB$38US8nC|$(PbU(9b?czRHnflAI
zlARKFkD5K~h6x?6pIR|uSs#U+rcxk(V$K@Q-&eKPsfJYl4fM?MqHO(E5(%yPK54}F
zk17EKj-1IRyKvGx<_i=WlK}JYA2Z5Idq`^Nq-OftJ^j4ZxQk9}m2`P(0`F3DLnd~b
z5t9o1t_O)yVWZg*sC>RkhWnANY8tEY()=!UWhH#YJt@B=h#P+wZ7_Vs+P^HRP4P(w
zhKmTn!)I)$1_!54{gfjj6tNL7Xum)(|BN4%>W(8A)ZW{GQ5$ga22T&xNnO1GZ&O$Q
z!vjm&{iTytX*!58_A!O05Tr3xpjO&OvLSW2x0*4qM?o3*6LtovVIbCxbxg)Yu_3=v
z<d)T~<qHdYzu?p^aR)Z8{(las|Mk>XM^0^KJej`(D<nB<{0%i0BDv$Ac47@M4W;it
zVl$l*$OYMV5J~|0Qu;jFRh5UG_7Dpw+MrdRyLmI3<@gOYeDY*g0-Iv4GguAm$Ei09
zib=KpYXFhunyux}e}7PlVTysF{0YT0Kvc?dNi@Xi&?p4({xP%B2?2J74hbw6j&Y^;
z8N+Ibp_EdwGJy%^0;YK8je!=u7J(y17Map>6l!R4`l>!tRqrmUBRP{Q-g_(MHIwUf
zdlUcAiQxw@KCw7&5PU<`N0nU}FzU#bE`h%JW156E^$C>vHAa9UhVIa1n0VdXn74KT
z*9;#zZ({WU;Zn0xf$MfFjE^|IW#)Y!ecTidLSglPPLZHd0q1ps!)9ivm2MRvBx;{{
zEa!@E1)7GHWDv9%&okOv6R#$`6>fTw8Hg1(j;4<8ByWg}GiQDX<FVSK?z%G(X4S2E
zS5T^{Zvi<HM39x1u24V9NMhaKHo$0Oq|?lI08SegBFciZ<ja#LLa7=9N<%9Bpn_Hs
ziQ4kcbIDi9?)}ahcYu-c2Qqw?WGuVLMOEFClxWT1cV$4D&44NE>5bJ@ABiG&%CQiq
zA8*<it*iMeXNe*aW+}$rjND~sT{5w@zmS<VT~Rvl;l|TiyZYPEeJai~AI`lowOXHO
zPk;d>_HpxjV`-RtMts_PAXt@1G@gRBRg+_Z!RX6Pr7g=zp$_9IV3t`BS#`+Q72Z}E
z>~9%dv@|!3hj_pXRN>q~bz;dE3lmtan{hveT1=hShit@3a@3;bh-;Ruwo9Jbd6lfe
zAkP_lNQn|*3F2CP1(VPx>~C!f68nDw9!GTwra>RZ7nW71`^Bup!<lrmR;mISGbex`
z9E@!+>v)iLtDb?O_ovr`eI{Ji{r%~}nF{qkP4=5kdmVF(@woPHJ?Y+&BBrNb73{(N
zD%vCZ9BtyOj4*N9z*nR>tsL#qqzbKCr7@&sty)L>*1yO}#%4H(9p;~E*3UN)Xe#0_
z?N9zCd5E<^0$1Chq_k_!+FAw`Pp6*kJ%K7w+Ui$9uUqAMV}eFn<!&`sLpb{<Jz@*y
zf=~VCY93i<yEg&tjt*jM?J^mJcH9b((s%EKn)J%e627$WDZShF-!N)=XYz>cQgK(~
ziP9GC*l3lX!ng-?rlW6E+ERE107<Y)DEuldZoRy*weY{#JMVa^`~UwViX`VG5n0D5
z)UhRd91bDbTOylOM)t}$G7cdtBO`l{jL2RYk?c)2nc4ha@9T42*YEfFe!o}W-}k@Y
zAHUn}`onc9j`Mzv=XgBs&lv(CPy}5A(wy=T{*g94HFM-eE($bCNRQAkab)0{mLaa)
zW3Ng~vIeQm?xN*$o3-Sh@9g{J%D+ky4+9WGl$bT`L5E(L`peE#Z7BV+1~cWcX4P#C
zdiaJ&`-|sy0w$3Hd`fq061#`=IE|2oIp1dyRCe_WYj=f^p+Uw&IT-_a3EPox=fi(C
zJhi&uI#dZtab12&kjpTJxU}Izq5f<1o6-XJHLH2a_(BGX>XuE=oErlo2g>b5yZLZ#
z8_>qZ%zmOZeo0HfYTcL07fbBbDb3wN$@5X~*cfRTRQddPvo)TWXLe#$EI1?RxRW4x
zKK!(k@nXx#*oq{b-Fw9yx_GZLG%gQCR`ZZ5x&XShp|WZ3-NQ!VI)8nX8PX&0-h4Q$
zQ7jJ_Am<xQ?cQ7a7XrEcq_Q&k_B$6nE#pKy0qfv<uP04HtCCr)eGa;Bv|b?9c>dn1
zQ}tM3UPy#FyE5#%ULAG0A6;`fw{FHAWH2B0fhlb-?<&yBf8?jo_v@JvcJu@p^e|Y&
zCP1=o4XX15N6pxW*`+z}O_vbWF-OlHdlc6idJeH{N0#*y=+--Q@eT4~XDoR-Ii;Sp
zovQl%{0x$<h9nLE5}VEt<}0e)@6vln<zWc!&N2}HO==i8n>14VIC)B5o3wZS&=K%j
zMuB3olVHBSo*?zH(vZ*!j}?<u2RJQg96y6IxaBaRXwh_~fp@#@V~Y@xyB;VnHL}Et
z0g*C|70EX7sNHkDec5Av51vQh>Nz;98qZM5EyzU|Hk5DJ(aQB{QD_}e-ks+f3;M~o
zFB`gvY>4yKpE9e!?zZ~9pMF$GHlIcOojPt=`A*1gh?$e`!u~?9YvW+wwE>3-x}xXn
zW;N_@LABq=ef)HNVA7-~$LOSOxliMFoAxjk(%c9OTE_p`axSI!60c+2^Cb+#&c-JD
zV|&~((S?PexzUWCHR2<z6z0}@Wti(kLl+}5^S}**OkCN$O9!AQPE6E)Bo6lXW}SD=
zg%i-#9bndb_atK)wiik9%Gii!?%=~{57>OQ2`z46;P*laNjGWN%vmO8b$k)DUAOTc
zxs7Ge;<I2VU>t~YiLtM_b?3sbB0fP~UEw~}V4AfQsC~-NEq+zPb<58YaZ#e;+XIP*
z(3s}$`mY6WpI`71L7fNP9*oP?_b{41PwcI0?sGETNuWQ>@V1jpa>@?}TYE{3as`@C
zaLD(yZ0#M8zI>+h30l>`#*v>)-<7=7tqPNZikU?}v-pfJZ#Jg#`4Ix>yk{|&v@ph<
z`$K8$LGwT=dyBA?SM%r2r^QFZ2-@9S-VAe(h2i}#I1r4!>cSD_oYAa=Zjk1B=;xoe
z5+;*^T1E{|$v2-<riljxt&2j3WlcUfIN~^RC<!&YbnT^P(9eSQ2aK3UDG@MgHm-MK
z6%0cP#8+P>pXb^%B)1N$d<m2m1ZY(I&^6;wx?S_v3^ybV6DZ^u)}{0!Ih^BXJyzk5
z!Z-+dc%~hqpn@p;aQY)0IV)D;yc#}g6gs=DscjOqLDG3()P9WOsHL0qad#>0-KdNe
zVF)#I+o&NtsLWq{2?DDfB3xR8r2xHKs}h62Ea^EzFa3>)x+5Gv2P#)d-x2xxx6TyK
zU_Yc%9V6U#+Z>7}=rv}kQ01K*SPg41TSxPo10yxo;MkGlq<!6lM%rsdnfMih(fyXo
z49q{dhN!I0rSBWO^4wW0<ysJyEQX{wT-SC&Jcn>S3B4+yUo-C@X?1w`S~#8Ya(Ixj
zTak@*kRYtY_z~fygH*qiLcW~TXujJyse?8jPmyP127_2P{r|>m#|-3cpF@L3nEy<#
zZD3Q4ix3f5*#$G#aKTk{`O0?-P_4JnK1UZTvitOl)xK7+IS7ai$q}P)^~=w3d*?8v
z{!?||aQu_~>@W{*VMufWeXA@O8!jNe^`cXX;dlJY>7R)UH}V9=L=#Erh`b`*McGJp
zir0)OdBW~+@~>S@apL+Ad{Z*mS^<6y4$ssU(Nn4(WkuQFCM%&Y7Qc<wCmetg7@|>v
zQcqB4`)J^<cHIJ$AISpiPg@JGk2Y+WF}DrGzS>s&@RKJjDZUpbb{B{T22jcyDFs(o
z^KMr$N|Go!Uq^`K2T(iIo##TAY_^RA2~bdfu-W)=?^z?VkQE;h+eS96aSgYC?X8r^
z=PjVsyv4qcyf3uP=Bf1YG{3_+R0L{ob5T&L&flAD+2>NaP#l;mR%0D&TxeMP*q9K<
z;@sIrQtr9tHy`&_^*ltM-4<&tv4T;PAE`y@@1B!Cl=C!jRAxKHndZl1oxjT;-_;(B
z9xm6M#AXIzQcArmlk#2>#>O)4!{!xYC-WNizTLEby|D4r%kU!IYmsfzO{$;7?^3_A
zq->iQH46;=@Z92WJ?xGPDh$8nh^m%q^WVqLq8h)T-`M<WGdOW9aS^)Lt#zdase9L4
ztTu^{NsQv2fm^|2h1mEF)6j2pz6ZfXr=~#bP>=`4nD};ae)tr@`5~CXio&5p^osVg
z<USyGdsvh3?q|y+`O429?v+~qjk!t;f-usdrEUI1?ER^K+w>9#YAdI&d=3)7J9GaQ
zG9CIrtXZy0mcIX{_X+j{O3oW+EovhF_U-@jas3DJ*dhm{Tpe`E{P(ftoU25_AfIsr
z;3Mg{5$HMt!KS*rd^J$CjRN{<fIHpyfB267@drS{|6h%_GD-d|VdbQSSAhE7S{oy_
z6ezp|T=)Qlb_AbH0d@wCJ{(=pF>Otq$Y%eyfAoK(ME~|Wr#<v53;uE`{`XHJ8ejBy
zL+THy9ndMhlsdZxgm5~%b^Z~8|Mac?!|VTFm;3*uF84pIvETnM2q6AzarP_*BCDDK
zY;wLLupsgS-B(Cz#xmEePhBgWr*%Q?VYVVra6xw2&>-~kb;Q?5FE>S1Ucui>oF6g(
zZ4kR}R`%aJ75c)E9uJBapdwibAgCriUjnI^MG(kR#6K>%`QbkWUbSj~jk5-mIg23@
zc|)A~_lIQrfR}3pbRuu=l=Z7XY8z4qJO3Zst)VnlO2V6{yfv~10`KpDp@C;)J{^E{
zI10LH9`dZ;4|^`x8l?oXWD5;+W8ZJ_LHNxzkjQ4Qv_i=YBRbSNW4MY1oopH3rk<z&
zyH;UlFgRzT-m6Lfz2_RE>i$n*Y-1U*<sK(uqZds#eE0K#BDlc^XC_VKZA??qSXvno
z%2P#v{xx0$@}~0CP{9sxl<*x3rP1~i>OsPYff1viJFBnR^D}9FE+l|Yehf%2JmNtQ
z<1i%_f*f^XL0Oj>WlK02!w$+kPac)pa#C*sh43(_$JheN#OE0<Pr$N4=s*{;NrVEo
zTq~BA?a?+cHPk~&;<&HJKsr+d33V<DE+&T_0`V}NFJ<KJMdfKw_O^bs-#}pv=#V8K
zb{q8o!pvdd)tgXGXk44gwlS*MeY8gY&xak*h=aq2cJ^-sEHFDEyutfyq6G>*AxIs&
z$(pT$e~QI{;{cHjiz`Unq@KLOd02k;X&z)?CjAAFj4?DoX)yb1+GQm9is4fm@NQZ@
zRqf9NUD(xDR|~DP1~S)DXd2yHEu%oT?A6Lpz{33sAfOqTG+8#24v_3yK!wp&Y|q&N
zgvG$f{%`4S5m1Hw<Ok+h$FW_o7BIVu=xY{^KoeO4kltOPKww*bBt9o(Jq>DS*i$m1
zE4cK%aL`%03{5vf#H|4ZWaFk_l*@Pi-d{M2lD}Ili(X){(A4#%|GijdZ+hF553{5U
z!KTsD_$|!1eP<p&*dK-CJkyihGlLceLxUceXJ7+#yR@xJgR|!&a9F(pkI|?>0PeNt
zq*Iq02T9f&kbmj@0?Ob1+QaV!8Y*T5QBUydK$&Xr1CXX0<WLXnLG)_4T!AhK)b`^z
zJT`;Mf$H-kXai-Q*8Ftu0TU9Q=z>{D*6C3i#Q+oCE^rDS^9WFYS^yI+$113AK{B2R
zs4V(eq&tDPRx^HaPP^*28PEt67gnd)C;WBu)H%r;f-JKflJc;C>s+YGeH4u9K5{hg
z=Y&(<#P<N67BrhWaLN#^hthG^Vrm4*aY6w|_Zm<?>4M1s&`3dI>aHm`?#BB7+Sh8o
zJ_F(U>dVhZ30+{*4KHna2-OZXIf*@7PX6m>{wKM~F&3C9HaxBCzW-}ydrvERnX}V*
z>_4_Q0#a`*X+EGRRKX<{v7X1N;UBDle3Dmm2NItG;&xTv<%p4W0IbK#BK06>_z;*-
zW}EhMb_{S9ZQPDcJOQ2=!)pueGYkFF1>ngp1bAykM7`UxYEm&|9|Ftfnl5NwrK}%^
z=qZWaLR~vNUJ#?}tpJFRJE=-wPN{uqJche;7I?tzj#&;Ba%54Q5Sc6(BTWZH_-U!B
zG~kH3T?6!a6FEeum=~`IQ?;ww09E}f(EM`DRreO2f#XKn^=u<RjMCu@m>r6G+}G{+
zK|6mG5p>%~o2uELnq63&O|B)R$O*Uy5GHg>hg0V-q$q^sG&nW*loy`G_PDXMxRei@
z5?9l+U9pJ+GbEPdngnV=!L-a+uSe`6f9ON9czSZiJy3rdE3*m`N*KY<x)Mu)p)U5p
z@p*px8j5sCN2B;4=A^rvQ(WtQe9;$x(C4!DF_+tU3}kSiZmkjzO5UGiXFDOEqOf%$
zKkzOkb<qwNuGxpGNO*W`P*cxWYS&L(Zxm*qH!o6KvYKiYZvIe6fOugdIw$|=Q03V4
z@@+t&Xf|!KJth+6ezf%~N-S}*=jb+NozE@1p+N{;TwxEytVBhf;;bs0#Nf^9jVb9u
zhN=L?DGsotPYd0b)2Hl~22o^+S3kQRkY))xKh@YM**^gycYdvgh=vjG5F^#DM4?DK
zUMtUXleGhIRu?Cc=pNI#+o%_2SchKey0jjA(<AACD8u{y-P2Gr?jEg1Ib64XWwdl;
z*-XiY7^ByeSjQ3NPrwtBfWpar;qi6iL8~$6v16j{U84ez4ZZ`>!pSV0jaISK=V^Vl
z)Be=f&_rwviqCSk!j}D+8;qY;t`hg+Y4<z=qWX=feX{}10kssOVt?{U+^t`c2QLTW
z9!$b>Tpl5R1x}6+;qN(Ej*g`we%J0?gCcExm8u7S2GssBI1S+dO??xe(~X|l;YW5(
zzNK55JLo2?0_`rEkO79rFsJ*Xo!JJmr2!{&kwhItwH^WXjHehU*y4u7FHra-=^X;S
zP|71ryg-1s2{_I_f<t5&raDI~2}|^Z1i<@PJB&=l1&Mxu%)GIr2xfa^kJz{oexQTy
z;n`QbblyLqu?dWQ>yY<tz`*qnR+3bt1U39RG#3~0-}!~*YQ+BB@ILrk%)6LMi*JSj
z<cD(?M0pp!0Ewy*a6Q=fpX59`Ve_SgZ26~eMoz&sKeBU9fMWA;Tx7(~xu>Pko~7r)
zDOhFh0^<wlX&hKOXw<wSoCjAQudhXEzyeSH(8Y4_0`=s+%#gzO9^^p4S3)jk3p}{R
zCTVpt4JHU6fcUDxzIf&NT(4X7$hsP&+^P51r6{FF8bFtE5{B4NPO9zYK;54N<_)Q^
zC|jwI{PnE|5OHZ9cbDPQLHEmAvY8>blT}8rGp8O*{?Y2+m@6<gfJPGY)wBQdD#W-H
z5veFKvRJ*vixD-_gxn)(T1|<uuXgCF8w@PYu~Na=qoSV9-`b@w6$VO=M)Z^}%q2{P
zkG<1)h_RzlsqV%R$J=Z-5=0V&@pgp#B0t45o9(8(6B_Xh_JbgOFW5*w!9|+Ff9*zH
zZ44PO8mKPq4%dI`k*o;BR)LaeNbRQZDwPa`eudZ%3^Cp7UAB+yLA@g2v*gsf16GtD
z;rmL8_Y;7-OqN2~#IfUNQ}6>s6s5p>b;M;*C=7DyRqqS5go%Z7bdG+LbI$vnKzLlB
zh7tw)tH=9SF-u)>t|2L=4)_E*ioZR-<O;O6@eN4W5SNVgbyd_(r(%;*9)3Bnu7LBe
zdY%zF24TECKsHKZBfT0#!@1C^zJ$Wj`hJ{jJ06Vw?or)VL0vTX@F>oRZhstXGQgR~
z?1@g~#auA?@m=%L9|cJW1l)ITt0d{<_JGTTa>>hk$E2rkenB$s&EV`wc<g8O8jk4Q
zYU=pKmhvx&dA}!#K|x_x+uxAZo$I)w=3B~%=4@`yqz%z(f}$P;mGf~7NXG1o%ITP8
zdteqr`$j>nCI-AKz0<H*`U!@a>oa1$DBvw3dfuTTe}~+M#aGd-j#;u7(vDN$2DfPf
zZp;(6G3l><VZ`t8jBcbPGMG<^Hu1S(q_4_K@Y(^U4w%iXf6ZdH!VHAcm_<!ct}`P%
zfK-#1e7{BPHpUd@!C=%*oB;EpzbFL2yOL^Ue@lRW!TMl+7nHAr&j9S5{>`h5hgT51
z6~)YW*HYSwC7<x%N1ON&J^)(n*N{!5-WvQYl68#hB73-%A#N%#cIBEhRTUqOipU7T
zA{kf^eUm`YOslX+q5O>!jgm==oohTYB=<E2_!^%Rpr7Myd6j7z%s_V%_yCh20vKov
z#ZtnDWd-he`K*|GNuX>e;GYavzN%29i7(PG*Dr=oKuGX%)aEY4Vz6@9F(T_EMv<as
zBy$5d08P(m#OVDR<__dWf;@hbX)zz^4l6)9)8eKF<{lw*Zia^z%BQ@2+|x;d(vG(-
z4rIt?yEe~kd><u5$@u|OvVjZ6jlUQY#k-WR3sBFmKcbt`!R{jG(&$u9Y;-$0=<%dA
znptJ&v<c;zgui)C>#1^EeV@wLg>d}=7-WWNlPOh6QDnfRc;J7++}RXDSvQ_VkvB!A
zWp`fuRR-$nLm-MsqG21KcfZzPg8I_QV;#;_A7EG_c05fg6?tD5?lMp8=nN@`a4+0|
zZ8KC;6kWe<A?72~61{@S_M;>n&6mW%$dcePhZDn_9=j5%z>*sxWI?Seu*jFeu0Uak
zC(^F8Es`&sKbm~t{4t_jPXip)qCy*W=Ng~F!?=4W@1{gD82F!Cxv}yEz))a5T}C1`
z`$8!9A&!*i_^}uyoR5NdM#{InS=OR?CHr_h5U23<fy8Jv4VLZ1@p|LaCIhJ@tB%9E
zB0z)~;<uvrL_G&4j;7<Onh#pAuC0764JXYX*!vq1xnS-Z;wRHS<%Il3V`o9`+!~SR
zc$bh-)Xy*4)mx(vY*KVbyTw$0JF=3K9`}qg{qHM{vy;$CNA^V|3k56l+1Gb^^s0mB
zD6*K&HsbU$-yENN`a;=o{NR}MEbO|2O=+)8`?tp*etPUxx2O7lM?QbAH$#OTQW2V`
zbBiFBdK1L_23;kGypo+@+$vVfLZQTjb&1xqe=F>x07SqQ8|=>6&2y5`b2EhkKU4DO
zvR*_GQ>UKvOFiFwT)NpO&k-^??5X$UtzO+5;W(WwY-c-x>~Pjl6(_y(T=ASy`y8SX
zKe~8EvT+VkU#xnL_Em_dmJ;&UecHh6X%`ABW{N-pc8YB#iUVFf)$h1xhDmxGgAZwX
z>~ZxuZ&L^PQ&$O3Sv-vh8h%3ziwJuE22Gxm6S$ccg3JyouYhH(yKP?36Uo+jp{%);
z8MXu=k!YoFA?>&fheRFUa;Oja0>&K_ac_TTp>SfRU}rjvA*TQmyYiNO)7GIXde%yM
zj`@tB#pU_S3;G*TdQW16*Q{PlqYlB?h4yLx8ARTkPtaV$P>kvUaSRr@E>BC&Sy@5v
z$@VKJCRJ?*_*Gi9$nP!CEr#=^(i|9@9F6Z&E!RH~e(~gw{&LU>Rd!4F#!F;}JW{|$
zU~(15qlLh^>%rX{tx3}_H?xo%xk6ad#+y~i9II*TX=imGA8R5%0NtM37D0C>aZ9|}
z)t-fdHIqr{=gMM^3qIfHEWS}^+bKJ3wPEVNHTRM%Y+vxV44ohSYCYLp-|mM_&(Ob3
zwc<~_-cT^5wB=XU)I7Tcvc4BDiY$EOiU7e*&zjPBW@dGU(SP+L@%vOo<OAz!lA2B|
zTT80u=keTBWcQ_O&3n0~gV8i}(X}5|4{ofKlH?8|ACeyJ-TF#&c_@>(HiCGYYGAK;
zp=wIxfX=9YVsmS7V`9-$;LS0q=vIK$k@_b_6>VsF&k7a=vRMnc=9mAS_5BxLe0`0R
zmG~DmR~+9Tx-k$}H#6-Nb-P=2m8P%bF=CP5cd8U3U&X+R^n~at;Q|c(+lx=KZoO+0
zBYfTt3K#Lhhx(__R=1xIezV}mkilJ@t11QG-!Qs~;-cls%BNM5h!Yl6u>3CO%L^!U
zaT-~i0T-(QltV^p)nB&h+xkG+owIW?5QbojFoa0rN5w5psr_~*BVsp9z%&pw^O*y4
zj4^}PmaNU4ByX=eRectvmC-H1Ch5c(zNUCuzJecR!|iO=4XUCyu}rbTRr+dQuHj~(
zTND!e?z*aN^}nyb>TOv|e_5Pz$kKyb6m#i%geji*VS_Mw9}`oi1*1pINIucIt=_4;
zeB%<X;mz9f4=^a$p-XnJ(Os+0MbF({XV4VTP@89{N?gLbX<Zz92eJC(lQ6=(F?BPK
zwmwhoQi6VPW*C16F;EUGxW<!XuTq{4f9t$k%@j)d{f5nO=2@aOGtLg-%OVyhrk=Oz
z&i%ubWLW>9XZq@=Pu;HII0}hnQ({5gyAkJXPyD7)fvH2V{3BI$>db(b7+~;ag^a0+
zkR_#o&IvVI3JO;KIOC~N<;?hL$cg5C#!J{jZP(A+gt_hC?Cq-Ft8IqLrmNR-n1!r1
zv1AvntyYfwPO~P?j&~n-_0iB@>%g!APCx$i?a24kMLrwRC1SZ689}+zC~^v4X(|Pk
z(*yMQhpe%!J%B7Otx4l%_Ir{u11DgVMy_>b6x(mArqwdJo7dhh{VEx585voiy%`-V
z`&qW5>GpUIu}3H4DeAR0g~|0S-gt7Aj!Ias9{t0(bvvjD+&3P6ts+9wMNs8RL~XZN
z*vC&%yy<D*>>0w`fy~*`+2-R-IqwcTSz^Uyd5W{^za*+uRCHa`3MAY`S*7H@$R8>s
z4(Em~Uv*6tNy~exp+^bJa+1rtZ#n3FU!A{xTd(kXR)h$fX&QCQfUL1X+yj+T71^cL
z8H}t%oYKK^@5QcE34f8Y`=fIAt@_xjAPw>c5ISEim$U@Ve*L?B6A!qWuL6v$Ihlei
zsZLAb>wEX<<c1XvANRel!ts?N`2lplTolyNA8@)KjfE62s533a$W3=0Td3>hQu6=w
z8PkaV-d+C;L|q)aL}fuQgpjgC@$@wg+Vx7|<%6jyG98`Ad}+?KD_Q1pC##71<E$!o
z>(*5JxF8c91I?GJ7sO;+14$yv-9YI1)H#Fr;TjkgK;+ri85&?2T5i4Xsr2Se{Laub
zj%#kDw`J`T(la;=B2N6zWhHIq9H?ce9|&2$+sRBbqcf*zAK@L+J?61UkyCp2=%cR4
z$LIs)W~ZIykMX-*XR?S1AuT02p$Wz7;26?A610)3rNtUjSy9;qUfucB6W%a*Mu?c8
z)8-JNvN?S_v?u0)aR}?5MY06%U77g4akKP7yjjAwc^t{!ufxz7b*&#qFEq?<@+<q<
znDu=BZRth_0#1<AB!HoiPyyLnl)3JLqG$?&<$9*m-1SlWZ;E#~GQUhl1ik=)^Q-+^
z@nVD4liy6^(06&T_xY2Z@QNZcDEP5YO=ummHB1WVUcy;AIU6tif@)3XJ>n*R>y`cG
z!4c&0gkNfC)`F>v5zgNqFXoke!TUY>6VbHRsUG<erX5zY!Ic1d+V6ozX#qhm6$1s#
zq}ezyZ^|A?UJYppJ>Ofq0E1$lQ<ZIvAaitC*51yvg~)fKAk#2s&ntc7%w~AhW$R&_
zXfP_QlI}d$D)82p-iHrAgD_Z!9G@9u1yh2?GNngzd&0OF*5_R2MSPfB$cl2&Pq<OW
z(zw>-c9G{v{Ba7&9?e>4kx0;863%<Un);N~kqsj;<%`qut)AbVY1W5wD@^KgR<&sp
z1`ftr1ZxnSc`4CagX3t4H9(iI8K(CW&{G(GZSpRs4_^L$<BpBOv>`h#y^n!!N9U6F
zMLZF#j%i{sFaG!8T>#YHl*1WXJI*u`1{$Q{ifbPeFg36DOh31#7qlm8k2#L`&zu}g
z=DL3RwrQ}$>ahmO#DrX~r&OpC$N>N8Wub>6@M~`xG#n8%NHmpV2t_(pl8JFZm7x1@
zFi;TyzJ~3}ep9`Ld24|3t+h0I;m>M)vJw$J4pZDkx=*^&95d%u!R#vQNszr#;j?NF
zR>5(u+<R><greNJny0OvJ_@wb_55Sc8ev_~sKqiX7x19(>p}wQH8A$=@^g*685W_$
z^T_<Y(mWDPdv=`T;U14`2+kA+6;C08^H9bv6kYC4LN{1~aY)4b4&>GCAibCtcDKQ3
z3cFjP*lu(Z`eJ(b3$cVnuluqpGTGwkB^koc_yNyQ2#|}Vn3~*-^K$_;qG)$#aNZ(V
z3a4Y!je%_*f&iM7zx{lfVGI<<4k)9+HI$Eb^VK5iB}mal=v)b30?aTT1kRRQ%FYjY
z@azx$b#P?qVmUCHkmA{m!gb=O8thvT4rfr}qD6sjIrV4RotFA*I1t8nnn#Ck9(dOI
ze3APU2&K1C@8eQUY*=Znm~r%gqR;4rNz$g8fQt#QCl>@GOgOwb|M-B<rx1}%3)pTq
z3kLe%zp9%;-^lt%D;o2r8l>W_RIdrC;Pkr;hsHIc046OZD=-BApFG^R_&Pr>4BY#x
zV(op(G-XsWr>yn{-E285PT|bt371VY{kY_C%D|KVoJ-}K)$5zcs>PH!xcVEk7i*&B
zehnbW-aK23*9+Eq+1<?BE_P{<W;Vd{8ITixT(8J*4B>Y{Wsf|kXk((VV5Xp8!3t^$
z8I;emuDr=eQ%sa8U9|_=;=Df4GnMZt6I^6vi16%ufGo6j7*0;w#xAy}E-~#p#f|hW
z00gYzB3@nJw)%}RP<MC}bm@&}XF*C;Ch?KQM>PvFTADj|fYu05U|A?CStu?N=u&W{
zf!tmMfbC+=A$@Jk6ElwGp&r1CUWvxaV{f2g6l!Tw#aSS2ttDo9gU(SvT3QIS;!+TN
zpgUPaw-flbfAi<pQ*FMHuA-DfZL@Q=j-;pUGH1aC`drjgw=Q5yr!98}1wIlr_tpgw
zLCY=KK%2>hUit`PCtZV|KC5%ea@6_L_h{F+oblWJ?S^qF9Q>6U&5XaiQq~KjnO$N{
zkH{dhJtVBC4ij+RY5p+y%;!K<&92m>@tElp<E*72Oyb%}GvNPJUj4MYxrU1+@p9Sd
zhyL+am!+BPIk?nH=*ule*Owca**22{U!XI>+?>OsS{vOXFdIynWZ)Dk|L9vTkjZ=R
zNeqxv7~ai6hR$X<ecZa$<1n~M%iVgzu1Sd2%A{l_+~FG#0IIrF4Zdv3?PHsaE)f(e
z3d{x}Yx(a()?hC_q4%?Y2eT*gUZim#Ui-?hzJ1HqA>_ww4!ZL`@$Eu!Obm>ek9{2T
z0x&=-=fab&NM>jtG62^C(FFuU79hp7b#epL#S+q-Jm0Xw4f}T%Z{%1k*nGzT87%&y
zG!<2%kk{9SZ*?+s7p_RXSOX>UNMT8Xe1A*FX-<eUL4LN`323Wq)~1{UUbL0BeGQ@M
zaPqih#y<O9s@Fc^f6c1EbYvG3VQ<F@hrIX$znp?p`bcB5yOUo(lA4+h>Co|gnjg@2
zO7UW3lVihzoL?FGJL|FF?3BK4|E6&hh?X6C02$eWpQ7O!bRr1GHkS6P!YSzBpDpN&
zitDf7^r4lCXl@Iu=vYt}zoN}TGPI#ehaW7VU{dZigztpeOt&!UM4$J-p~=fj3a9kp
z*4i$J9p@bzn)U|nJ+YnsfhEv#FqR?#ybhA;J;IbF2!tV!B*qCUW0Rr9CmMKlz<_+G
zSFCCD98tBF7#wvCtd7goCY_Lxc!p8%G~`)~3)IE42Xp(Fuv5d&X~sD73B1&UFRZOM
z+uC_*FQHb|9mQH((hIUuNQQ*?mz1B8c?y5cb&9OhS^*ud&sFlU<SjHJ&v$g+ZX&4#
z_-<;|Hf!HG6qChgc9fTxS^d^R(*=>Xf`dc<)YgmBk%{970K;a>1w{le9Jl*hZ%eC3
z(=)=tw_t7|tQ#hss|E9gp9Fd)&VU}vl`f>vTbrPUhFsbvRun28C_HD+alj3>e8@#9
zztJk8<LAK&!aR99{^hN?*E`0B5l~)D5+jG>h%nnG+&Z^;gYFrem!9@Z2L!f*i95e8
zvj-6)AL=^l8Cp%a>2+cXk}v3m6JYrHif;gIVV*0pharvzA@9p+VsslZM_;D{)rJtp
z%GPrBl1*BK5ssj&wF8(lW?co=^7#DtBADeZVfZV+)>LzQOw1{~UNNg9bGn?a?6%WN
zE5i2_mBIcxs8C2rm&SO;;#@?=HyfuL$J|F!yxz$iM+r9ZA&gP#CS=<7qKCyB4a9?p
zP36xS;~`vNKI@V^K|<n(y8<v=qHspNVxK!aEQE{#JvaWqS7Q$j-V&8T480vtMdQb~
z1$IK3OW#7@OwH!Ll^b2l+q{zu8}w{u6Xi>cc4Mjo{`;h0{EO=CnnSAPYTzP>$Vd)3
zcol}E+ayaY6o2A96|^aNtalspiTkL%9Hj)g^{GM^Rx)Tk>`0aKF;_c0CMKVQmH50V
z1yH2!ch<PQPv}%dc_nfS1HcALik0*IDcFL`#h-;dC^jGiJ~^mz8SYtqy=VYR{*OjL
z@9`Umbf`r`Z(%VblvdWD1%2vRct^D05$NRTH&5o_d*w3j8>et50#g2T>h)B&`uuf}
z#8Q4NzEz;UFO+yesR={494H%Od!anwvtcl1`QAf-+g_pJ_Z5#NtQ;RwxrtZDHjgHs
z#f>*O=HD~h7TAVcD+x^d_PtCqUHSU(i0V*TLlDkj6fgu6?fdpEqX>Din*#Ao6m(Um
z&jL3(h1)46HU%Ib+6PSo=XBFoL;TzRWyZk7@VYGW9^t~Lp4?@caQ@xH+OZr(HIq^2
z+TY9IxbeuIeLY*q|C^!+2!n;7OC66M;5qZMME>HzvY685bsoR^T#3`ES7B(^#TV`T
zM9A(vTB-*Lij8&*8aG}_#8o=A=uH_S4YH~S7;U2_dZdw;xUt+%z!i_JG;h}PHKQfz
z(AzLZvhZRlO_7N_N}Jc#cFZ}IKce3j0TDz`e(j;9-Sn%<y7ptnqRNMy-5ekVoL{n@
zrCc~Z$Lm=LbKmaIneWB#@}rs0dO7PtaUqs?tW&IIlqKF8h{JiX{uDjG7I_`PgB*o)
z+DLm|r3=^huAa^I9(6K~H`O2wc_>Vqw~rZ14_d9ZNyq1n$}^3Dt|7lVp?uQ~duWO*
z<;41%A|<&gZcE1-PNi_!Jr99Y=8lj8I%slvd2zQDAhf<r@r~Pm(;n~`k&M@Oqde1t
zQLXJ@gjz&(Tk@7?d}RL;h4WJ=#G*5KDDYx}5S?Pj@j=6@z{<>teQ@3QCtk4*nSuMR
zy50NqN?H0);Q5d%2V757k|o8#WyfU%v)z07(`PoHFUA{Q?gD(}d82{F9*7gM+eoh4
z%0jsUdb(M^#*Q!F(3>+uO7dZEP;8UO*U<{IX^#C2KWq4)7xS3fy3o7wAw!HZX)wT1
zAI4=Q@KJneIZpXGbO_X6baug&F93o&j`-m=7JKq^!MqmHf=4%l&>95_npC=dY7o=!
zg#W6*X^WAfwJ532&)(oeCUat&%#m9m3ypq7FMA(vG)-0{H^meH1G{OdG3X&@I=uSj
zTv}aNKO!+%U%PP^l%O?E3&~GtPka3W=2>;t@;**nJOgu)zw~oKVr!oOG<^vWa@?~%
zC-wucM`wJfAJ9zbX3uN7@q2VO^VWa$OP+7&*vJ{We)?(o1(<p0pVv$0#^^XdQ({PT
zv*p^O8UUM#3;IMouO-&i>pipzBH(!+C%a9RrF2^FPT>@@A>?ek7*^)mNx-Y1WFz;+
z+L%ZC2LR1G`D5h;m(fsk5Kg06@!J`mqU!^s1Ka}~1K&0LcAY@OE=ONdWZ(ON4Ut&q
z5kb#OZ;ia#;)HBy>-OeMni_mfa%-OmcbaUU5E8)tYz$}=U_Y6$D*iS1u{7eP<b=d|
zHqe$c{Zz0KTBm2&HCq!=_D=K#ZBoEw>jS-=8$b0AT=YNLBJD$8vQM1TTG+7dWLLUF
zH@H1|=G7z|QsK-ty^$we(DAa#U^3dZj3@@AJ<jF--A=xO^H#Z+>(9^_;N6EWbp46>
zW!-MGXW*55=hJ2mdh(ZXTY+v}vh(3;DEbqRl7)`&*akJj7f`YP`b=2JE2f9>onJ)=
z=}*EJ`vSog1h@#JcX{O_*$=c9vyf6(F^hgaL-_FXh?^bYVgsXbl-^q=ns&jq=!>Nk
z{BQYV+h{kx3Mfre&d_1?$S*mG8eKq1xjmFwore?H)DutyuY5U`ay^&|j#)9G$Y<z$
z?~%CDJJFf?So#RYwsmaOJ~CzD2&N@a!%USqbWy<@;_q5NKWL_GULg1U_5fu|3r8j@
z#XBQsamDaeF%_WOAHPVp$AiSs<{(S`Nax-5_tY}xmEDbw;eE=5imQF81Ad%5XKq+~
zbuzIZSc<(R>T+-Kg4`#wy+Mi(3(z5#>Z8$#yjVOf*GP2hMXoq>5ifR3L&#uEVs0v&
zF2tiAH$G(zGYHdSFlw5JMMoaHNtOC8!f^f3Wf$uSe8>)4#H=sDnSJ_!x~0@UQPu3O
zUF6Xg0q!L{o%wup+@*S|6fWQ5A-G3?hstY5^{4^~fp#ZE_q6ly4M|kzvE+@i4z?U$
z@tDa2pIg_tX!pugHGoD6_&UnuH%Vyk75s{)aT|p&EkZJ*F|V|}N%yAZxBQBv1YW_e
zanaaaot*03Z(xBjBjV<(a<K<TuH!vmfb*-^Ai|+SzGK>+rq9lijfK+3W<@)rH|~^k
zoVn8D(hnkJB-56|DZ5g`2<!%!v<lxMvQ2Fy^%5zfOnbogFmWh^y6BQ&%kX1K``|?+
zSr=oK<OY_Ii(~!EWYQXArsQ1n{tSnft?<RCj#X<m(8acZE|`_L;JP`AG3o)m)Dgkq
zi<Gs;Aa)FPj?^Dis%5cKEnh?7w7GO0*T043ZeO@GOw~OiCFaaAgJrL{Onv7XYG$8j
zfE2Z|kHUVS@KG5+PN1D$HosKiP6{+6r_=f_WK{6;OU($XCH>-P_kZJM=m2q@mW*`k
zUv+>UqB$}1`S{o7-q^l#<f4V;jQ$cEVc4gt8XSnbC=9BDQRPa?ITRV{sTIpt#34Oh
z7%S!m=EYm0LrkRP9OI#ceSR)IOR6VoPl{9ZHY!cxoloX@h6|F{5Sw>EDC>aNG~+&O
z>w9D8+%&$C%*d;UL7teP#c$#SA9hH3$t5rO0v|4qFC!-1kSB&f8QyE}#2?<eC}xuc
z99&Xg>+T)OF{l{D)@5Y>f`^bK4&jHNaaW7CH8@2ViQm8xYpGP1;PvyOCu`%aF4v52
zv*rrAA1e|-C_QxDSeJ#cE8jTi4w&D8@7SgF)wAWiH$_U7f|hqOp6i#3t4`<f?yO!p
zl9crpz%0yMIPW6r;J||};eVnF%zmyNt5iqt>4pSa&Dqn<HPu1v)!Je~+#@v?BpXyI
z?VBYml;1^>pOFA4+z#{7zQGJLA#SXpB1YuLN*b`~h79$`dWF-UXvj?d5IWVHpE~8Z
z2;PgPz-hS6Uh2mkwpchhZ$5BZe5ZGZ+uLpUV3pF{ER;KAbNrdvnsKmp<9VHPSsB!x
z(IR4_Dy@LuOKJ4TjlhmNzJ+7z?iV*xM_Uj1B)EQ&KZ9v=qR^cLFq0;ct|k=C;X!@W
zGHAN;g#Br*l!>R-S8T1nhdlyk&gHQFhM|Z|W&r6NohaTJ#?$H-nq^F>J~WV4ULP|R
ztJ%rwMml3YV1BB+C0g>YGd8C%a54ETLlZ2j-rv?3|MQ~aJxU&4=+fB!*D{NP(Z?|;
zPr%d|>~943Dc}fD_E2>Q!oA5%BFz0Y^`19B<Tue06(YmALIV%Klft3GwBT3NSZsVI
z72*L{iHr?qET&9??8;`Na=5~xkNb|c_Icx$gZ@z%h5_|=z;{eZwfi%t{hEQlBT=Gs
zL8c2*i+f)tf{BE+>7;G2tZjqAC96ONv|jl37m^OlCqu0uUQdF8gAHWmPkSu0pVoKX
z8&V&I&&0{vpsE-VKNovb8EIaa@hvUCM~JJ<+k#p;oHdFc_TC<v=5@xcW<&LQVcGrQ
ztmq}})C=|RlLpB!Ns^lAhmypKE~IySi1foIa_E!8yu2jDUC2>_eXnjYm--qo?@Tbs
zBGh%n17G{B+tz$Ji~ZR-7F+q!w7aY3?%v0w1#X^<)snsHy<s!^iQMnx{SJzMReb)l
zZqqv-kctc0GgspN`UvoGRRZ*Z2qjzv8jKUv?*&qLvwmm@_$WxuZGgtoCdD-2UtL+w
zO2S}}8>onx4Kfn!R6L(u1-DueNCQTJEM`^qC~l2oChBS;#PvyQ0_7TcDZGxpIA~@O
zoV5V6qR=fUtLOrhjxv_Lt5s=!ECj2N{%|MAjM+j#or*Dxt20U3z5~*VeU|%K&4r5y
zL^QlmdO~nLl+W}&$6>8-I=1le=09tm9)@_~*IBBS{^hl-V!-wIItGF?sT&e*F_zg`
zh&yzO`aqf&9O({xGWvI|E902{_;z7%@r`PbC=_Fh-a?lBa`5Iy!S7^za<s>>#=hJI
z$gqY$o%;zI_^!;%s_*Q-ao2i?a}==pf~xwT<OI~Ng3p2faf5HMRX1SlLi`BSP6?ox
zR9Gj80uGkK0sfA=a|fOA?!$jJI3(Y}CGxU~HUDc3+w0;L)cn#(d!1`N1&{&eP>}e;
z)cs%3;6>9v6#k5eM;af1q7$R?5yZt>Nuut>U4FG!Or8Nqv%j!`Hbg)Cl7g$TzOtS)
zP1utgmJOwud|D60PIiV(Ye2thhZ_&1Kwr-5jsK=r%t_=GhyQmi=zsRb*K>e;y6XMv
z1Iyo@{6EQH|4F?053l2kcm3zZ{D1e#qBp2<|EiNvjrbkk1&GxD@jtxe`TyJZ`5(`o
z9dU5txK1T!wg1mPvH$JFz<p@<_Zv$dg@JUF!+NGE5G=70?av-3Rf^U9Ve>@6bYh|p
z)+axEZbvODz+>=;%xmKDE~x7-YvAu3FKINsZ~2E0_RlwLj{plVflyQZ?+3&yW4N=}
z`$vul^lv6o{+BFP=dOi%K;U@b24>B9T<3*bKK9~El+8*PT}ueH8PF-K5Y>NuFO{d1
zB%0n+HUZ(|egW^vAhLFUMkm1Bm6Y)r=%|fADmj%LV9`eE0yv|5QbzyXTsWXqXQoGX
zRJFP6=GnUESnHxi?Z3OAM-!c9h#)fVm$+eTu?{wig+?G%w#8s0LD2LK@?kr6FKSn7
zuYjXiY}l8GY!z@*f}QIH?*Sc}f4}Insk#SC4|947Qx+UE?27~M?;cy=#Xb0&{q<kE
zCl<q@CDZm;HTBOWGv$fTq8RrCh)drAF`DZyfZFEs&Am4t(+DA9blo<a*T7CM1nF62
z>y+S+NYz!)04WEUiQZ74@opbj2hy)@0NCE_+XZ`%2BPb%0`t#po0Z_ZgaY7zbjh-u
z3j+%7jGfA@E(oFv0bHyTfy5mWM3w-7T>e(!gM;8)Xn4JE{}v#tjtmtUhg1$2NO%CH
z(y8&J+qR->fMd`JG<!nsVfH{WP;w8OCb>H@rc8q=So&*WPs>7zjJj_Fzl2|z6-W^I
zm!+ptIBjF8Rb8eHL;$IJ6*3ChlM}tcq+^QQkAB%y9O;)$^nQhSqL8)I|7a7ap+|x2
zvj5>DO|8d%eR|jVt+j-sGmQZuqZCjStKI`}UwxY=C@B>~rOW{bl=1-aMW0*YHc$yO
z3^!;U{jZ17dty~yoA~S4qC81Z={K7X`XeSB%gcYRUvW}q@uo!ZjK|h%kDeYKO`g$m
zUlDU;q@ll=2l_z#rk#}4%h@?D3WDRHRKkH@2WONS+L&&<IobvjSpve<TlznNpye*&
zLINnsrQ0NI`pa5^nxIBoeE(;NRt{w6I=d>5JWtBjKA)X3yj1`lHhxgy^tUeqW^{Ft
z+Qkr5X-W;|4OCJ(`N82?1U{JR@h|kez}Iqid$(wutZE+zU#&?y<t4)%z+=cwzs9id
z2FhcX<?KKWrM&*th0){fo<}2-G?SbffQOU@5mo*8Y(c(P3UI{z(D9r}2Dgv5DQC?+
zUDxRaB4%12O%@!{ov%{-zyPb|nE~pF{_*4LD>NiOL_NTbn$i`rbB-(imdC-7&wb2t
zK;*jxul*n<XK;N-Ax@qn;aiJBD;JV7B7KA*vec*oPuCM!Y;G5!7v^vS8vC*e+N*^?
zIGDbH{CIeY-Q_98L_G`pszpx{OLf+_e?A-kcye{%0bTj~0Wo=L0jmCS10<Clg@)+n
zi=&K=AhS6T2PZKw(sTWRj|Ugz4#q%q7tEjsQq>`do&~XhqswC3OraPggV>8u07QY(
z7`HTK=HeTyGR}FogKsK8#7<Z!>^Z3NVPUxJ*tO0c)T-o*rA!>m@hhBt@DM)n#H(wK
ze?D2Wm3k=oPHG6k*57xq^MoISx-LVaf18@!0iZ-b$<-w;2Mi2d!jOZVuJp@%2j_8N
z(?6p>ADcw*DO}Ic2faQBwSV^xj$`z0ez*q^(28;T*56l-J#qteV?LEvW<NmBDafoJ
zOc}glF<!Fv2>?Vs`1cDl6L|q1Ja!E*_d_!M)0$Y;Y-<ZWcq7;ZAED8c;1jlb^1!}S
zQUAG~lIy8G0i(X^Z?}M92MF7qbx@*rCI9$)$<c61F%P#iB}nQh*}yPvtAW9hwZGZK
zjL(Bg$31ucN%igsE?!;-8BGOh38un<1cfAqZGBLX{1rcW+yX{i4fdU`c7X`b48w~`
zgeJ^MqQ3l6>>iXMsB0fCM--^bKa`wt2F_2ZF{Y|{sMKVvIad~x8&rvMWyZNCx9jPk
z$ofW&@kAWQ4Rq_*z=?!Br=xh0=I&EHVON(R99c{26sapA1v&Cy!^_O0H22g8e}_l^
z<2L7%#_4DJ^Gx<ml=Zr0k``10R5%dtp<skS<@Ockbr5Hj5M0wrUgok!%{u{QvJJQo
z;o(`2K6}XUR)HWcUoo!LG78ol7#RwVQEzuJZ2(RZv=fFZI)a4QNpj~ZL3pPrG;ZK=
zV8VD}SbG}`IB0o<z9>2Hhb8XQ_5)P<tWMff=3Rs`9PINSn7G04s$yfEsGG$uf?IJ!
zUTIgDb>OVLa%J&K>FtzxK<o?W3ySnl^I*B6U)CT%?+SMhA`AmZXc;KE++mXHD}N(e
zVZfk6$NKCHm_g7Z1_Iokl%y&;<u`E}lD4-^0QRLE4DMs^Q(O2r=%Pd;o4@1d@n$A&
zXP}dP{s-`gX{_`&ln-9PDItd8ZdOm8Jng*WJF4>oDE9+^VA!sgqUF`qhOtM4swf33
zTj7{5V~iX0A-9s`815gc&GYZA1MPr06!T^g-H0agL|y!Gin^Vr5vC+JJt~{U;;Jr`
zR7+O*MlxP5=zg}%JkRm1PQ{VWAfRkoa5=LueM*+cRSia@)p(z~P+ob-cpIEx(d9g5
z+Ml=f9$PG&)mX^9u{5c9H4b9r8=8+i4t_Ak-LU!KnfrdYrV5<LZ@U6I95p?jrqQKI
zdxG116y=%Af76_O558OC@k9Rf)QF^W9)GOF3|+IJp#iSf-GT}+t%8`nw54tV&Ple^
z4U+}(-Pd~1kPE(SMjPQJ&z$topKa~~dh%x<asFfhESCsOW@scm4(SX$c5t3F7km8)
zt?PVxwbFo#G<I;TWGo)Q=tBTb*sW-t16T?-69^t$P3+C(i(nUgqNjHD`=}shhVT~N
z)YZFb%VB>+P2#rn&%h}#4C<)RhzKI0ES)4^8P<ooF^~WvpY0WW_c)+#D5F`386GJB
zu{{GTT>lx+Y7@3+>YP|JN-slBiUz7H_*Y2aVPR#ksPiOLPmVJ~(<IVLE&cinW0{&f
zo)kd-3NV)^$ui*MFhqkvJrPb(d`^R)??a;anJ^4e{`l-oEun8UeguXZ1k7^dZh%JL
z-Ii6LR0-s5YctWl2kga?mr<8chfj`kh`d&B`f&@xzfc+*gu}xUGszzFas-(SPLUH`
zRlf9r(6<ovLBdj^h2akOT3$3B<#gn|)_v6&($ZOAt$6JWr(QWW>r0b!1jSLAo%{HR
z5pn%a#=|0j_-Q>J0+jXT{=UQfPswUI{7!X8AX3b7A@sRWybcm5*xvaI49@hg=wG~%
zjN!`SySuavAbE7FVNb#?RDLw{iD^061-1-SoH~ALZs8g<>wY$j3!W-at}_f4jn)08
z@1bV8e8V{-MY5G?(k3?aBnQArVF6%F3GO?+=;sC8iDMa%tOTX32fS|C>SUvF38Mgl
zSzz6tsmySf@f8F1jbM#RIfKeF7@tDELTdN+o+pqI1w7NqQSOpe$;*8S4VoN_=RATS
z<}dEr1vON)*K{A&0E)*jXef?9-mDDw;QY`j&-B49i~lC&qXG7LgC92*PJMV-OW|?h
zt=b=*&-aS?)&uj$7Y!YY^kb8bi4xIgo?$XEWkE+iV#`yf^d^U3WUU$G<{WmG_7Iql
zF-A17IGj$CdSc3e6?VYNF3@F98{RY1L0taFRq<r-X*r$rL~5R}=WCsUsSObJh}J$T
zdTKVA6P6sVy9cg9CsT-i_Z*8T-#)=;lK*oF;E!|G{4L0oP1~1*|EhpX;-P3H=Op{2
z#ObSO#ke)Ed0_Id>I<Bg!{eF-k?N&hlL$rrqSKQjbScvlmJ~Gxb(jswL9p!oD1vGH
z*ahG-<wJZ4kGm#!X%p`r$tK}Fzcz){g@6M~0`GGNEbMKMF3{LN{C?~HFxP=56MC*v
zTG)p&D=aYBpHrrvCqJ-LCZUaq<_>gW)m(1<K_HHJ$xutb0=h?v46nIp3)4Mi?NJW7
zAO-@?7k;<>4Ojx~X$%=kh!N2TA`5a0*1)SRF^wrNYUwB7d~Y_Hbm#_ADzBJ(7hr``
zl2^s&V9P!ZCXYYKnJ7Gwy_h8lthqEkDmCVo+9GY1PzZnumKsQzY||o?3}&4TMB&nn
zeA7-B!G#fB?7e6a!IgV^;~t<wN<Dz)^GKz2&9em4x-7oOr1>wtW(v5fr9E|?xaEyA
zE-@1vVelY^lVSaFe$VsLyX=HDY5LN2Gvtnt>c3>%q@e9SwN0t=aQQD$wTPPRgjb6%
zLFhaL?lkHv^zeJ(hFZ3Z#A}Q`G+B{>cx^5X@A7Wvb*=&LBhOqMq-j|vZJ7enBbyRr
zL-^v`Tvt-j{bZWwO>l-!^UWO{j(m)AWyKQ=+<jQ!kmaQ0#!^;;m3s%J?<)X8@zNlU
zT7i@%jq=Ko1=B-iZ6zT5a-`jkI*K(rZ)95DvT1G9Wk@111%VE7WHyZx#4(@d-pCu(
z%$yo}6y<aMgytYsv{vjvRLUOz)e-61z6S%E7wu>&T|igW(k@Wvk7dhGu}zXE+DMf^
zmba7I=2{xHpafqbzfqwSKtg@3r`0-7s+;)*%VZ$zk#_67(1m+_>%c&C8@F%-HNgMd
zh3g$kws97r^B2_@l^g(n_M9aFysP#ffpb}25$LaSNK`s7U)-)#h!(!`Xq;9Ac#zxE
zYv+8VUpONxbJZO`5xcdU6wcvkjevRf5$FxJ(VhUHWMgo?UtJ1}XUx$jBdZxC$OUE!
zk2mNb%aYFn=atZL$F18s)uYxgC6JYcF~Tlt@$N`DW~GQY;bN_L_g#M85q}qlvX4E?
z&2y7uMmGco7>$@Ie!L;R{XAQqxt+Ot4x#Dh=*>}yL-DiZtYQadymUpq3fRaZzE}+B
zJDIrYTi>U){q(%^I)!z|ukHHuhRi|fB2@fVvvl8FFH3~{?e^w3<^{x052P$>&7Ss&
zQOH=)s#Z1<49@lG{FtAcL2YHqDbuYjb<Ttzy!j%f&Qbr)Zu6G2TFD$9I^KBdhcYbU
zi#0%FGS5G*eRd`aU|j8~5d_;h?$7UWR(_2HwvpPVdj~VheQ${U+f(saDCV}RJ0^ZY
zJaFlWVw*e<!|_eoD4KkK5AF+kIu9U<>#+0K$HEs9ER|#0<ycbbPMdY2<(?iZ3gcOq
zVfVNCLqBHcUOWhxx8kuAs($cmWKn09HR)A)M)(8d3Jb+_arVsjZ;Cf5-bWBSn2fHt
z-%6ni=_Xp|nEwSB0r7fTcdTfw{N&?`*40Cm9FvJHi>^5{HO(mx6nCX-MnCxmjk4ZL
z{-R#w80qNPN1ez+`@Q7-NRZPt&1%Ec?d;-EnFUnN$>M`ObM}=eG~z*GqR@d@H^w2}
zOrjPH%;qPd5#*g0wKDo~dR*kf@y`{-Kk6_cFnYYOznlKzq)Zzi9L^}|7G$?6>EgEl
zv9x016+ZW$u9NbKiF{4;SfnZl=kqsYwc1mA`&aLYDKOr3wvXdCy1WH9qPg-uqbo=I
zLSG-aq$k6!_P(wVj}IpgcMWU4MJUq79TKTA^0m!Dpj-t9W%|B`Hl3q|Gz-6oBSH||
z+=-GrE$!dB25LT`&&%(0slwjUwCYJx&FFk@s$Mn3;!#SH%t)GD^cvoEGRf~1O{1SD
z+oBmf+b!@+5X;@zBfJlZIr{*BR_9QvX*+T-kZS`I+nySu9)%T-)7#24ZBN}R3+LJ<
zk)EgGG!$3t4LiCjck$kN){n(*AQ2*ZOg5aQv4jM`*IO@i%4WigQj%zU^SahG7Wm1F
zWt@GRQTo~v{5x&V?cd%-w>Kt+3OpHEt#By_xv%nl+0s6*9ls950fnV4>H|TMAx`7H
zDk!lypzet?PZ9gx0d@d<uQG$lwq#Vb<|g(sm1GH8#JWUm)QfN}D=Z>|%a-xFY3tR4
zkzkym<&p}GBkBars9j#cn>uUgmsMyuPP~4>egGK+DI?sD0;p8kufbUSx?qgK?<;?=
z=^<G7*{7VpQdlLHDY#i@dXen6`<4&;+^f_~k#Z<DeG2-{RNtto#Vb+HzEs<pZofbn
z-Yu)BnG<%Uhk$t$geYHBw%$}d2V=g1cj{<sat4-)#u+D2hniSX&Ake1v(zkNx#s&F
zV3d8QG#4WullvCkG$RQj%LuZ4gLA5<SibaV#GHC0udo~U;A$Q1^;@B1KxxD1uma91
zGuml3mF=nxvZGMJ<xGB7FJQ}CkbR|){oOL_$#(QhWyZduDq)TLuMVLg>b15?5bd)%
z+=Iq16Fe{VxI=i;%2aTo6Q<~>6q(x((w;N<@!C8Qmb*Lt(RNanPX4vlp?uX0q>YS<
zW?$uDy$^r_*X^=0)NWdY=*tF+_3>59USHCbzi>QB9gW|`Xck9otNzVw0eiNj<VlsK
zP-!bi9Lv$Ed)Vd>aOwJ$B^9}V&gY%VDpnCM`Ux_QElhqwgIE|%1omW=(|hd;khth7
zP3#qoGNMy4O9aUWw0v;n{J|@ogwg)yPx@fAODFSE1t<$Fn^54*(i(;s3Cl$*Qfc=+
zci9oRoTD2TI~qmKUo}uz&H=`rg)c|*ZhY^6um`eYU7wa@S^qpvRNlY2Ql|4*-;AWd
zF)4yOvA;8o`z3De2&nwzG76_=e^(XEOun1OkO)4@F2C{1-?f@bDhB+h{#aPtH(xJ8
z3rknKKsd$3jbe6KLU38f1nhYp?6Ku3_S!P7=8Bumf0NXDLqo;b0~bsFEVb@D!=r__
z3pWHiXRB@ZS2%)(LVQLl`>rL(Vi{QJ1D&$R`+n)zLb0O&<Ys20nOS;5^4=CyWBuEd
zpID&uHq1Te;VF``QdHZKx%1(Jl1?bEo&HJ&DTK`~6Yf{losmCK<QD~+?&6DJhbUvf
z$w8O^=7GO;em&tBfagrs+y1RjX8axJi#gd*fGm)7+HMPVSP}mq5xRUE_iO4^Dg@&e
z^$go=AffLq@2_)kR^v}Xx{Hqu7^4u$CIppl4gyzs&A%qc&Cu*cWf1aXM+Wo5mcj*X
za`MVwyACk+nMAYI6ek@8AA}GFj@=iU@*?!__g7GnpIbxXP3>>ahB)ize5LapG!DJ@
zVKq!+!BOM-EQYW>uB~d?BHUPCDp}@@D1aG|z@O!n4HSa_?>kg!UhR)72$5AOW3zZt
zlNzdk^9fW{B`@Y_`~*|MZrrY}#*sa8daK5sNF_LsVYzqDhbAQoEpsO?==-uE^~(9J
z9JEgvs0Q{6=7z0>T=bF_4zl9=)a$ISt;A?oo$2wvI6D(?D%bA+BZ?*zndd1K88d7{
zY__Om9zrGaP$4rZl5O0EWC~@744KCeWu9l*C}mERIb;9zoO9~^o%i>?o&WW_u5+EU
z?MtU;@B4Y~d#&&Ktna!L<4HxBKfGo$UVDED$BmDYkHaxH=n`F<s_(P>eN^%s${Tj{
ziI@Hwn|uEyhrBiiJ1bj9KOcHTVTVKYnB25wgH|RUszsKFpUn{#S=5Np%0S=K#wAi7
z19wm;6z^4nFlYa5yl!|}!GwX5<~t$N3k144x*5vyA%fte?<(dh3=I1ep6wAypd!jy
zf115{0frW`t(e`01y)f*H?YSWVcOUN7&YVNkoIwOuR`}SRi<;ce7C}x&xZ*qT8_{i
zI`04-Kw+77?K<qE-k_?m33y~R$My}>Kdn4+>oyP3HJNiSOq>x<7NjazmkC<_+R;#z
zo6Qz$?>Ow(-PkwSsgK1ScL?1MC%rd%pcylus!l_2C2@rAxVp6shSA}HkdHn?pjito
zB}p`C)~KyYLel$TWCSp_G&b?dd<o_GEM2D19M^s}!tVCIvk4iqVJ8z%$x@p7L*Xvc
zMWTEJCzejV97tB4(J_|^ZC^;)>M>q~LUS$B8g$v?^LN!W^JJfIek`;RrqqP_hl7#o
z+Hd|TR(Og<y~$GiDX?8zri>M})?6+QXqif`q*-Hm+z0Ifk*G@jVF$abAjL+h0Un-F
z=pD_1O26Ij02aKD7Cb<+h{6o10n=<cyv{BQ_$^}#M`p>DXbe^rYJ}<nKhhCFczQ`d
zl3ShOijz+X6O+ASi4()vw1WEbSI0Rx?FFR={L}6L*2#%OYnP6Zq>I9yA7j5FT2K(W
zWb?cn6^rsUl7a;EbJC&)brj}q$a*D#{>D4I3^gW)DokFR_<)zQfynd#gdfSYA8Z=5
z{FJ=~2y{n`--*+N^um~@ciP^FV)|`W<;g<QS4~4lYcUw5Q+UJp+C^q(5%>81!iC`j
zF+?}+o}tkvw~01(8Rh~8&G1zMzNVJw5GRd_wAUh}rZGmfS)NrGu?x)6#T`PqoeMb%
zG^cL7cN1-xd;w00OH_SxV9O9(-~ll~l$xY9H2bLKL@@EfY8kGHRSs5x7|=sbQ)sIb
zXS8^US*B=uua!w@HMK{w=lyQOj-_#qOR~sK0e7C5XZ){v)01V4-5eb%-rAr1WMDLJ
z8rqrN>OmiOTxl)!)o5yNj-Hq664k;SeCRy1;ntUB7q0_1?z-N|<xR{v`wvyM%YOF;
z-o5Pj^;=_)2$q@T7(LC(9~{=n1}1v4r>Xa20L%O7@C#Yz9=_0_$VcI?d>$#g<pGLl
z>(L2}t@`U&l+5jAD|_>8J~=n-vdM_j$PBIAQJpm>Oalo{it5uP8M25{w)j;vVLl(@
z2mOl#j3O2nmQT8~)fe;N*}{YZmPrb_v~r*8=oyy)zb&^2Ho=G49&HZM5wWiH_U#Tn
z`eBuUvCHP&KQ)5?yuD7G?y&|X^LcBScdMGzE9?zp%f!f)mW|+0>;*rL_tUiMIt&{$
zc_jxoJ|fNm!sl~uAWl!k$h>p>SpT}Rd7kcrc-U44&(&L_+kg|1=m4~xOQS87w^4rq
zBx|-G?tEXqJaW0p5Rqo1GuX2xAbpkSnkqA!*r4-b;yUM*ur+L0U`c9`S}GCig7un#
zQMUS@>ykJE3;E}|Bvs0~_v0ruh>vgBm@lckoP*#l#4?t+R#j^2x5Kj8yAMLSF}J&~
z;;)tICLzWyH<CA@_bP#!9wEJgKI{pKD0@8hn>H9rIrKal8I_1tOfnHc1U35<9NPs7
z|097dTBJO8WAjew!8B4Zfr;ydZ6@Y>k2K3giaAp_Ps!W^SYFW>T0NVb^X1Q0X|}oQ
zwWKu6vI_*gUu&tug_^s3?56a)itm<+lhReBj2I-l`_SBJ%<2*Jl<t{+nVQk*(c;>J
z1GAQ#YQ!fRSIS=R+OJkTt_>*1u%N2#n(uC69s1u&P`G1H-X*oV-3%gZg}i8!O+?X4
z(0^#v8-q6|cwN;pME1Z-;|lr;rvm)~m(4ZvE0P(GD5Q}?A}P~&(|cf=%3e{zd;I=^
z-k7U*4Ju_P9?E}yaZ?dBwFPOlVZR;{q!<Oe+=M>&o)2CLbbet}^(ghGC(oe{gpEA^
zpw%mlr`~v)r{~QH_=MP1Jp;k2xfu{96R^kPSjfZ7Q9xKuKbRG{Guz$Trv|1tE6E4?
zkZFD$ZtjL(!aczXVOd*U^pwuJ&n+oF0>`d5xRYsUwn^`#eFNrB4@vJKKEVl|#n_6(
z5+6{gf=f=m`cSxl^b)H`qdT%sI0$^`-0`w9yQziAwmq`Z4z*9p72jt6=na;k@z~_u
zFP4;q;iWEVqI{kjJQJ~&_gCwUI!Xs3``l$64sXy*)NQwxBsx)s(D9_u>Aw1kz4s~~
zQ@EzxNfuGPO#9i42`y`B&sk|?Nj3N)r;D^7YGVd-PNmg&!<dqEtffQt3g4>(MeED>
z*-?QMmNRp$aH(rSGXh;giw6dyBRDY1#TM#hjLVf7<>bRj%?<5eTi4Dc(0Kd?c=)3&
zQkm>R1KFOB?9bEyU<nE$aLGpbMsJcgPxmymbIBrOa68N;R#@^wp)~FDx<cPVpf*lT
zY+7RgzuHM6>-4PE=u$p(H4Pd)$4(d>6vi=IF)wrs8aZ8*h9IYd16v~bp@C5vEHOGr
z6joF&OjGu?`=4_PY|MP?FX+_LVx(OJMMD@0RqQ^KSf)*Gk7yblQ~tf;@1KC-d?`%o
z#e;Q~Vd)sh(fACh(<9Rw0qs85&f*?gF_U!|JuFn0+wC`sEH1dwr0nzk3(;fuEI=lk
zr+pJDVyV}tdL9<;7yDkv#k?{b<N~H5N-X2E8VsqqRTfj=fGw(yvlXDJe^>dG=8KHu
z6L*^diS|?`L_49GSd7RFk;8deCE?FblU-nj6p@xD#=`VV(U$ufsezF4Yj9q(hj0q<
z2d{Ul%)cR^j|cG6EXi@$h#*ABk+KIF(e3oC4r%+yY*o7KG$(R3x5DlC2+Hx9JHE45
zmZs**y?sZqQiY+&+}4V2x;6K#kL5{+jw$~zo(_DCXj}MBuL+sF(L+kqi)1mlFI)3@
z<lM~X^0Z4bs>Rh=7&cU`Z-k!hKbL&{(ssG}M-t1$P>odO2JcV2SrfA6PnM@Z)eWsV
zLVJd_5ve8NRz5b$Y{M?A%v3!0Nb8bNW3l0Nrt!V;z0O#~_QT<-@XL0t?~ci@s<d;1
z#79nmQFS>Dpzc>ohhshfzkKP8t1Kvm<r}dvo27-i747B%9>Z`^w}7DFeT8|EHvo@N
zOZo+AD16>9@x1d=U+~?`Tv>-!$_ja^Z$PloFmL)0Y~9(mp}`jWZ{gn4&6C)QZ`7!{
zPWRp^?wcCW*AjW2Q^8arCs004w9<?%IALjf<Hdy7d$4#GnU1uv7PP;y+uG>TNM{+c
zJ;D+7IPBw*u0uF4x(%9*b;u`dVDOUy%I{ql?PZ!3fA<CJa(kqQ2G@F`izMvXZN`nP
zjPqwdUw>b2byFa{X}3<KIbGBfsE<$X?%<w=OmAFy9d&3cFW7PVO=g8WjhM&hTTb_f
zjdu865!p)4@9$c$NsmUJ$e{INsj?#3l3rX+*vl$RuwkB2XDPIHi4FpoFnS0D!Md;s
z2v43~>K`|2N6lq)e_wxrZdD}Cj(*ztSH^X}<w9*L$1pOYSn<NL)nzt`{LnYJHd1By
zBGj8b83SnbbHBY6@qG_f7Axi@!Op1wt9n_q<Rsr@i1}lK=lHpTBQ3rwn;#jkb8NGX
zqzd^z@|BRV>Sk5cZsUy35yl6GG`TbJBHaP*QWJ%WGS#=6tyC;jG@7}aX_?8o9>d%M
z@Yf$}=SWaukc`-k0TyJ3wQB}SA~GE((x`}|VK4T7s1iZr(1>GU(PUldh=CnK?w!R*
z_)hpi8GJZf#u;X;d5D<l;S}axisj2DjqRN?jGdM!nV=N57`7*DC1)P1D3C@oGx1$I
zw~ih>Cy>e-$}ZwE`61@=*-sIes9>i}@|hl2oSo$=-3i7fXDdPB<Ht@;#2ep;F?JyX
z$At62w;qD%nsU9w=31cq!>9NN-<+uk>Kh>u3bO|`PHK{}bD5r62yife_Jcz?gw(y8
zgZ<&VHir-{pZ>Ru82mQk@;zZ>a3Hzjgwq~0W#p3`=dK^csQ9vY8Suv*$HtHrGx0J}
z6I9$ZIhJ~J%M0vb9tphNs*ANuUuNzoEt4vHaX_e(!|dUBtAqu~`M9-5CHt0+U(YCn
zd*h0+ZFqHreFV19SG!*3(O6r~{hSw!3DIGrEC437w}9`@*=aqO+#UXT8e76+jy3rv
ztYy=OKJjcY`MbABy~QU9Vy{Q1+*&`pQaoN@a$7y$<jjowRNPzdhu>STu6wl(jkJ0W
zD&TZSluh$;qDu2E*jIG<4dmk8Pn1dJS*d3`NzZHXgU|ac92d=N2T!avq?X~~j%2Wv
zb3#;{!_T69X7BBEI4E_sg{1$3isV59F-^5?)xecMlEwSH4+tphQ4K~rZXx3j8Q-$V
zadOD8(o;0#3TNq;EFKr??#ph(^^;yxkw<H?AJi2SdSS@4(;RAqenmkV(SbdI(Uwgi
z8suZdp?e|}R?}1i!&k7ZDpWF`*@;dFAND;qq&)e|XGen~iH|FU9ZwN`c8LXjzI^IX
zAyY&#UmL)gDL$JE!2*?53<hMI{@4JO1Ef~%gt2F?LOrC?gM=_(g|b_MB`$LPTfh?~
zPT_Tp*jvrFhf#`8*)^|n&kVdeDr|JR;?X9WA(jD)ZpRR#w-QPOU+rIJdt;;FZ!64`
zMGaon43y%uIGEqTn{rZ9aD$4STce8>=ixs=ch>DKwZVy&V4;t)T+5c2G#UM#{h#I0
zXoAU_N%EHl#7!RdEnnv|37SjvW0w8*CPzZL8@HmOZ8oe*VTU{-ZDu(b(Q0K^e1Eg8
z(jb>gsO*Y?Ev=&ZD`Rc~_Qww7<{K;v(=}a(?!i6?0wUd^MJ<zT>S;+|8}k<~K02L&
zJMoE$>c%BWm{FSi>(wcD8)KvlHr$R5NC{@Yiq>1aI9TuDzjXZxZJtN^u%!b(Y*z4h
ztnAv9vh?>vZ&>xr>%iRQ2a@5h)dTke4!5~J-dX>`h`^G+$uzO@semept6pgOMbRO@
zEicG@kDweHx-mg|XIPI#rs${7P_T}Pf1>p#y`4sjdESX(UXZNTP*Zw3Lm|Jzqa*%?
z?O5(Fz4^cVl~m{CiLOyn=>+`kidK&ir5)y8a63^Kp6Q-etUV`5+zID`CZCCBg%79<
zh=HF{<=3RXLnYe89!CjN5|2?vQ*ETV${`-LP_$l@WV-y@L3r}Sm11zCk9oNDOZ-h0
znKdDP2~#Un2I5MB9%Vzt%4uCAPv%}NWFJ;(xULvP_|@v0n$An#U_X<7oEDNKl@bHF
zi04hf4kEp8xezkDK%IL8E;wpoF#B&G?0<jl#c22f)V|;OyY2O_-`nvaBE5^KAl7fS
zwJw@FwF?Cx9pt4DN9h09+g00MRsN|6{NgMB{=yL4P4jhprGCyG`Zov)EN!hzRy4#j
z;w<P9UqQPFr4t3R-dotHf4FceRk-kH5;}4J@O!^L%D=t1zzl*S54olBuQJ%*ulwJw
z^MC(%5cy!2pJZM5hcDoN|H(h!;U<S0AS$5f<u9+Uf8nbCP)hkT-}47&O8t|Ubh!V&
ze*0JU{_90l093DA-WurrL%{Ig0O9)z5J;S&%HaFU{p7#>-hX@lev6_!(fH)YoS$id
ze8#bV2Xm#?46}@oFW5thFgH%dn|P$uXXpD;L4nVYJ$qN1pM>|nt?mE(@j@N!CYS2`
zvS0o=TX~h42rQ*b;a(dq#oO>R2%D*Voh$b+Jb53=aew$NtMNt;v{h{lmeuH#=9YPX
z2TMj<pRK-RyEhNW)?xV0ejMDN?*JsH?Omw7zt58)l6maIOG6PB_eshL*|rz}$KE-%
zosRj?pZx*HaJJLLV=n!e-!sJk`VB+RDah_c7yF=uTwbkMa+MfQQ4a19M}%U9L9xO^
zRz5UMYyTf{+JC)ye-XL&U76@g)c(5n=vi-r|L!}mv$Y3pErkh4v?EVmC@PKrIPM0@
z_yU$AEFX*|KNKw8^4Y$EevEFM8!7wxLoPq2T9J=l%;pi8INR^%AZ`KLRq`aO@-g=n
z#H)ucKjs+}Uly{&R>~*BiV$+YegQa|0tCwlLoGM=mcHNHFS!CX<YYbTVC`E1$q^n#
zqxT4$;H*Ost#!bJiam3V&N31~`c#ZKkc$I7<lIRjb=kJao>ts0wi>jGb;S1@VXF`c
z;=s#6_W;omT&Hb>&7X|^zdrik@1@yJSX-Bw9*vQ|9<rx?#2^b*Q4nB4Xs^EwApbgU
zAa#MbmCOsoAGS(!qH$yrEVu#}BRQ2gOect884Z?|fk?L$WaEJTLKrpnPT|mLDx&eU
z^HJ=k{ct_zi1i}U6I+c{?OL?a%Z)=5cmemT`t2xmXmB`=0kda<9~}8WVZH)c&=CzN
z`{xxS1K`DBKQ0S4hZ%?kh*W^^&Lv|voaL8@DH>hz);!yQ9N$Dhn<biA`Ow_r`=s~K
z?&kK_YL+UpRrz;kv&82se?6`L@rAJ(+z3O<Lj!-6=AKdzsmrWA8iTwN8D@`UOjlJB
zv(UsymLm-oM4r_I-zQjpyhDT)ENRJN<$P)2DyF~SeG+>`#@OjG^A65|Sx6&!3G-Wt
z*@+lF*<(m*`z}eJCYC1_jc!pNUnu)f+Hm4ZI}n=OBjC@@?XzzuNU8WbKyf9AA+#Dv
zE&Emv*S@F0zdz;g(82C!{LqBvUn_in5oXkl|7f&?FE^~X8GTEO@;*CkF*z5DGzjxu
zF5*QbNhxG*0ZjQ-lQ17Ee&+!E`|gD>A+Rp^?cQC9?6hN+r}~fmb<|p6Yk_^XOS9{U
z0pR@I_jxOT2yTUVAo<0M3!4|{wR%6~g0Jd2AiemX$R8%#?z@EMn8Pl>BDJz`?_kA;
zGSpd+VRiN}fG!xTa6}JZPX-p8fif_*uje7CLBd-X7$Coyb5plzfo(xMH;xox_mirD
zs;0~hlC5(D{;L!;Svmbq@O`rH<f5KM*uN1Fur_HAvHw80X`~6}9rzW}QRV{t?ywDP
zE~Ccu{-camNFDkU?Ili`1q74L^rMO`UV_vJr=o)RbhlGBQuSb+q$5>D_*Q<1@)PXA
zIJTgg?rBf~-|PYWv_BXC&{~ye(g68e#LV{V=?e$&TzW!FnYgt!ZwZvEUORtdsR@Fw
zs=Yv;?~#HhS0}W$6uPP3!9%W*m-T!@Ogb;V=33(MD;08uHdpA#KeO`Dm>#wzaZ2d#
z51Bx`OgD&=D>lk4v!iu~+B<1+Er7ii3~CFTGs5@6g_LVU#ENcxs6kpW=r-z8f&~l>
zQ~A2uWQ(5Hz|K%m8c`av7AfXyNo!p?a-4w9wzDe+21Mcq4Q)?PlyIbHC&`Vf|H1wT
z0y*#T%>#-(rZ+`i^S0w)Q3aFUE-2AWNqz%iw_&%=4I#+G)@~w@IGDbcjxAZOZxcA(
zb55E$%6ku)h5%KvHC6An_xTT(y?tk{^rSb7Pm;(tVx5CeK5BnqmwP>FkiVldn|)g&
z&l~@nCMr0~&rE?kn8Z}*3y(j$qq@~d!>1`W`T3g#VUH?81mxl8m3hrXLsxyOK|C;C
zhnw)>se5iQnuV~ZyhkWTxTFV(;JC$hr$0BRB{7QoVj+0J^Z`s8P4Tf@{1}nFBE4eJ
zQ+3F$&qe9YzO%;qAln;glFc<+pZOD@^}+AJ8}eR8R5%SHykt?ohkVclCIP&I+~)9R
z=-tcYzK4jK!r2t<fVvbTF7}dyU26eQmaqwx4?EDj_wa(>?1SPZ#5RBLdy^a<8QoWS
zT=M{X6QW+dIUb^-9!y0d^7-BXoGl|(hlT=#h<KLYAimgODeE86Sjurs8K&DRaIAcW
zVGHC-M^EfHcuPd4cmOe0#1&&Cj7#a>2pNG+eH#P-Jr~$oaaa81d`{7&+wzXNI>_ue
zMWQqSzkJSo&fzZk0u2SFu(U&2&(jxRDIN<LPEG25#j!W~ZO?&Mb}`V$XX`I4`XdKk
zV|+urQ1cM~BH`j|Fd=xYD%fslEFyga#T^6mgP|7LBL~(6R2S>uvBp@vLC(Q;#QfCZ
zYJX%?grm<&!Ux(5l%xh66a)sA5!ouiAznIyBxedo;k06!P~Qfc)6bHAC8fGIx{GCy
zDv2j&t$pF@?q@v&s&Ck_xVduR_~0o{y}<=I{0FJZBl`6>SfHCnBBqAp@Ff@*@{qO$
zOr4uzwPpQ;m2<sVB);Uo{Q}=Z5O{fBuJ<PHZ1wf(tZs37I;&~MKtT|vCbtddA<#TR
z&K4!~iSCUEwU7^VwiUuq$QL$e4de8b)>CaJiZy;4kzNae!1Z<zlcDpUk?S>8o_l8g
zjE||&m3~L1GHo6eeu>?ghc-;mWL$!+40fJYLq+6)l|1JwP%YlUyB5f}69+~)tgAAc
z4TU|c!WfX%C+y`3*tRG&85~I|T@iktVooac2&p27+o>{u8<v8~s3eV*Zy?`A=X1R2
zM|u{a&}~Q>F@V1fqssAcvM?|3kg%uh(>&%aulbB6guGY2CWKy$kGK}W@Y1k=_l?wS
zRDboae?Zau+UE!Dp$9UPBIT@Rdq_vY%RoE7r?s(5bUl7nchJ-HD7V`xE-GT^pfslD
zhiFa|G-fKiY%fd4Z!5zCjl?Vkz>adJ+!o3!XZ}<V?Ko!0*~x~rEC2XV+_WC6LSwK{
znQNyQe|?eJeKLaWxIGNFt&7O?Lyxf^WOQVYYy~?c3QDqUWAeFUFZ>;WKzx8ce<!>3
zO*yNHi?X$g3H9*fKW@x@6^H@52&XuEn?o{dAlorYu}7X%OM%<byX|W>=>uX#B8@j3
zp_cPf%snY58Uhz`_ajS@*kJS?$7i~Y!!HU0V;g0f<C}Jq1xYob57s1vr)P^+eNbTV
zLP{Mw%^Q-y5HW0JmXd&ZRfT%Q>`92V?x6Bw1(_n+E0;I@MzP(hW{n7*SMR-b@qxBm
zJ8VEbJV|xm%r1Q`Xkumq-X;1^G*iLSUdgUA>D(F94lXH{9})!D8F}*yi<i9#Uf{MV
zUPR5DC}fY=<2$q<I~lH+V<<Y>dV^ut$c*88&D){`v4=DJC+aph28E|-osLHALLihw
z-%%Rvu45)gZ=N~vdw|G8g7Al2q0X<f{5b)Uz}@ugZy!3B^9bzYq!H<bB36tMC%<&p
zRN17$xn(Nf7IR#ox027dleR4lrQ!n~#7RK!(s2FAibh3VN!V4g|5jXGSC{TT8Ig(C
zp6A_+JS5P<l_$Pkctqea7fo9vhOAXQ+Qj(z1p%`^Cd?L}{?XVFHao$WofC<Sz<bx=
z=4Mv$Zn0-7QjM`-EcFT(4JuOY(k9=h>I{;w37J9@#~$7aVl9`DjCU6V&T!EzA9Jpx
z(l33DKM*6C(%D*>#ndJq;~W_*J;x)rd44%F(*CHEhEO7y`Lviny90$^p2qUoWjF-6
zW2_<%(YK_xPdZ*6<}%VgFTlb@6BZeqOeU~E-Pp-tRpP4uar+89;$qJSwo|c7x6kpp
z)A#>mE$?)srZ<ebL`WbJw*C4LzI!%)m~NuSios?rons^zh6M+gml=Ii1!~`Ex!J)y
zX8NvEow$9~Qez8A*5n)Q;k&X%;NsNMu(<k`W$#SSbFb^6H>zC=#0QTJ`p}Fn2lRV=
znDtOnGTz;BFpwP7#|_e;Q0DFldv)C$AGxQ?F!85H$OUn}w6<B8@(ZY2%l?*uhaR+~
zzTCfJ*wxEkLtef^=k0J^?!$wYts1Pj+iL6gOZm`}!kzSE^G(q=fN?K|#2i}KG&cWw
zB~cHGqdDicexDDYX*fQ9weV-|2(7_{#PrnXzf&?0QUvbf%_?-Gs-QR)UdpX^4u!rj
zfpMuPdB8v%jyOCg4iY~g!$rnW5XewF=<s%M)N<i^hn|VEnz2T^i{@w=Gsg)bqhfTr
zy3!aYE4(V8^&;hKGXs#FB?^<xw<ctWSSRLsHuzklW#9Jq9SEozr|eT5v8}d7XD+EY
zw4_Qr{61lQ&sK--O*do&Ozr$rLGe6d`F`#TJ$ahddqT=c9dLOyz9m(p^opk**}ct&
zVp4g_Go{Nlu(ktB*DlIGRr-up&e4e-0kbi0#T1KFTZ=usQYePszPT)e<0~vE>XX&z
z*`%ALi(E1d_9?fROJ07F7<9!&(kt8%TauQ3m-^Aae{h)Snv0Cdt52IzWcQe_i@QIn
z@p5W?@0nIx3_&HQ`jQD$yNJ1QnlY%zn+$X!KCGr6?-A}(VC}lq^`2(Q`>C{wtx_=|
zz)b4TZG^e2Xb(-NXXJN&8}PQbFrn_jRfhIn`kQ?9@0r?t-mrU`1lvD6^M0xvT15Ai
zwaz2Oo%K|f-IKjMN9lw)!Ake8>!5*U=xT6UFQ+*RubFgFcuF``Qd%yTb5fhnAVSw2
zx?p%7=S4NgZf$Ex)r;|%L^DV_nx#GG@-~GQ-jojl`Di{ReI#P8OEqLb9eiBTeqxUC
ze3_b@M3Qzu$RXa4j|-IhN1j~bYlgZ7cdv;@e@^e-huuZjX$7scT`n05U8C{f;@C8!
zb{c$Q`lW|ga}_F8dg&?&Xnab-Zg_I=-aC)ED5tVn(IGz4pDo@vlB!EB`%>WkFf!bR
znWS^PiMF3Fjnjn<=7pNRvMvHlUiDIpxvkC+@5modZe}#{<g!v{U`j&nbhO8Lm_v7Y
z!QIKPsUA&dD=l0a-4FtL*ZS5~O4}pzLUWyPxZ|?`OSHfJqTT3WSjia2^M;p@;lC+Q
zR`I-BOL$<CT+p*?l#NNsy3ID7KtQ@qPPEX(e8y#M<-p^#b}r{F@Njx8mi?I@bzZuy
zNLVdZjsCjkJ2?pP#QNeaqj#a*H+25nlX1ypK_xGiaTDhXktEjznSIMc0?Xc%_N*lN
zn2gZkPXqWyDFUBf{h=+B-p^HhDJ%o{m-hcLBmMQJmH;GOT8r<pEdP9~Iz(QCy#)**
z>;S=Q#0_9wNwttZB6l;<E3ol&q#l7DFMD0RG#ip4OyI!?;?3H=BMBJcAeKRyN?mMV
z@jk+|_v8z?Msm_Lb@d!GaTo$ooJoxR(W6OOy^sv#hT~D*H?6O(@2_j%AXJk6h()Q$
z&EhZ(BK0)5DsyI%%sx2wzk<Ng60f^LU9bcRP>p-gRD4r@D&@2NlAT)I@frJ%SQnZ!
zgb2P#gHks_xk(p4^FGdV>9EFEn%CczR1y~wt9n)W8<r3Xb4>A-0)d^g$-u&8Kk0;O
zxgc6{bsk$%?;I_xq|#&M!S+sioHD-K=0h+A0i~UcR2;$$CivA)MqKVlRe9LoL53Yf
zMu+5RskE&fe3OxJ_^T*kzBI~5m@s?@i5TBrho03MCdXzrb4`h5sV;#Bx0A-!4t8yW
zH&QEVpvt|zPCTG=>#wKuzVrU2xRMM*qz~0p7FJ5L{{OsEMeJn0`xOp+VmL+^(SvlQ
zBW4uns+HAOlm#!Wo-{iq9UT1r%o7<>1w76G<SjMWstwS39!PxRmE%Y;%4NX^+eJ0G
zmo7~-C1jwEYLFjIiWEX410CAe$6T*M9D7$csOph1gRF`jbZWb(2<<y?^e`EPJLOzg
zAMFUZS1|S}Sz-t{w^$>-e-32DUI2^CdrZ##8ZJPMnT=&`;jsj>FYa_`@hFS(sbh>v
zYFa}<O9?tcFPiD-)R%m{gHZ~`mDzp&_HkiQQCboyyJdRX>5<;c?aZ!ZvN5$3*}zav
z)JfGdo+D<4VDb3rJa1JFTMNrj42ON@SX9^ouL1%3p+zVg@PLPU>F3#?De@*+9F%X9
z#TS2g_^-`PMyW3X?f42=vTs272hUuWVMh7rM_Ehvc<`c+tlnAr6aBk%R{gBqm75jm
z12;ajZropF&86;n2rypCv<~HNrZipgKQ*p$oN3?^(+&73bHyC{mU*)!mC!kJL*O|V
z&f_tv!+2oAm`DX|cP}o_y0&O->c~7F2o-4H!iB}4vNV>lQaafy+MX@1y}4)zvKIwS
zQ!`jD-Kttr9>_u+G53I)ecvC*TDc1IaK;x2{20X7lgn3^XYZSDFS8gat0i?xyneCz
zK6Rfm3n&=xftql$X?nLfN8eWZ%8qtIv$RL9y!|oi^R;Ca>*!alvahU;tyT8x16&e-
zq(lxmn1CGk&4i5f3j&**G>6(s)jP?V4bGf~u2CowPEtPkY=T$n5^XzGl!go>Ty*$?
zbe~14Wi2r4^%E+!XQ$wMTX!|;-O3*@%-6%jFn#S9)wMf!I=-|o%~~<2hTi;mk<vCa
z0~NX!u$>Y2X&J60y+!N&pv+im1~+bpb{>a$n&W9#S`@LC1xses8Z3*PJ?@{s=kVK`
zRFE4*l5=k+{z?92-relb`M?4#K`k8p+%dIn>ojZLwLjR-tu|=Aj~S4Yl%wj|Pcz&H
zw4i>`@VBhB>a<LE9kiT{i@fx&0QNt=c&<VuuC%V2_}66B`z9(+DI+F9R=$+9C%R;q
z&HD8egm7%y15(eU7qFZNhJSIBoNyARw;6Y^;ICA;?2SSZ=#2Y=texlLUqSiUvI@x9
zIz2K&*{M*I$;0E$nf&1L8A)RKG-5OKq%1B)e^8f`8<+AJ5u%ELDdgeoT33ZBy@i~+
z2n?s`DSWW02FBb43JFx2`D)5g9y`Hod}?iB<g9+l(8HF~1Sf>@@bg-jjWS1{+o8Hm
ze$LH(s!bjZ)!KP(tGSas-8KiSqi0-xE(G7m`}P@j;##2_jcJISjQ+4*e{GPqzQpTe
z(uU|8wj%fXft{Cy8^DTNl2+zeC$d3b_>4L?LE#BR>3E^}egDBHC??0-08_QKNw{k_
ziNal-4*8;_JMp_h?@e-KI$2H!>^bgtovH*sZxd{=OgKaPerZMHB>pm*!H|+5I38li
z8;(Ll&i`={e!W?F0Fef7F!mq(B@GgMBI)s6_Ce1hqkdXT3tk&?7GIG1R$NNPo2Gdp
zLBhclX*oLX*a{vU3^rFV<lM{jy>DSj!fW7yd#%w!Z0yMyqlcFcJR1iuplk-F7k3<w
z-@-mQ3#TqmZ1-2M`-dnAU?fkOC^Y7!L*6O+TAH1gQUefj7uBSf5^Ve$6r<q}QKxU>
zVYb?*QDmmXmi&)GU+lwE7#7f{1`7*X*$LjTXUD8y39$FH=o$j=9tQM=Sf1UIG7vj2
z2d3VX)QIGD5y`O6aBbZ%f0nnoS-%Yy6dZWvJFj<Y&PQ?UBryXsYD=2>Vf+2Yr}*ed
zHYdvKB<F2msH0j^P@#=Xt{qr%&?p+MX{Xl}M;`ygroj_o0TaCPw6g`<TXMS4@9}m;
z9jeJi3niWQ>AO$V5+KUSHIehV%fE48jY^0aQS;H`W}PUA_RKfPHt07_o6^IC7ZBJ}
zY-`A?ZkczSvOyVIE1LVz?;7<RD$0cHo2S#*DNm;J^SX{g`l+xl4NkMqB8`0$Ep8xS
zX_<GiEb<PTqNhQHJ~haavR}9LVeU)4v8l}{nc%?USI}wq9GTqJ(#IJPy8}&^XwG*C
zE`oR{&mld3D<PkT$LC`EU}|PEMNc3)!N-I`DQ1R%@&eewFJX5QSoq&t+FPw{6U@tW
zahsvp(bF>|wsKv<Tgv@!I{84R$2R9IAnk{Suoj=Zr#5)p*wdEV*R@yd(Oa;igK~eW
zG)wQdBFt0I<FP^0n?LU7{)~&+yA?^lllM+X__q6F3r{5f9Z5^IjB6t?A3=@fxIbcW
zy56p9U{1Y)y)BLujlePVhQPIV@H|+9VTiDLk1aB8s-j!R<t`;llJ~lSlS8LbjR?(P
zyVhKM@S$nS@b>h`?@re{jEF-@my^U_(Q4LDsX1T%0E%F{`SF!vpE@(Zo<t>H)mwtc
zvRI5kyOPR^c70<Fd_8htouM)UQDr*^cf4BEIvj@ahbSCgVkASl1T11Ru8FEk+X~dg
z9%4Y;5#liOMF3AU8^x0zg_;yb`)d!mJ@_1)jgf_T{b|EG;^=}5)m*_|p6+>Ky7R68
zswWAU4<a;2JIv))tS<KqB4V$Rg-DEJ3Jx)}NZ7316wr-R)2=#AO|xn`5`kg1PI_Nf
zoapWd@o@bUUi;OvUb!&u#QRQ2dPa#TO<{W<PgB(v<k;>0XPgQekjb#pAm7_3IuEw!
zX*7G%G^Q}MPl2vAc*4Z(`VGeNnv7yGoPY(Gj6P2UHyOkzwYcKB8mrKt09}JDjc&WS
z&Io9qZD9+288{)6)%4lk)+)Zj)4A_iiqi*{ZqJ>Xex+6@)MRkLXk1j?o$1DB->iyB
zGl9>;<8^)=snU2b@V)Zpa((>pXE)#mH3Q<^v?dKFB4|UMo2SfIW@DFTC@*zrZ8uiF
z#czNDOF!|Z$WM&zD(c|1kRGT;dNN4deu{`uLP-MG;)469dCex=YvbFUkVb~!5(_w=
z!)8dFODavAV`jQQSGx&9tkP<LbD(9B+8v>?IFX!*+r|*Wb|}3be`ML>Nz)r5Mn658
zCm?q~P(nnVgrw`I7r=g0o541U+8f1^DR?z#M|RxNDuks=Bq_ivZ~WCkP(AMseo*r2
zNWYk<6K}t7J{Y4aiu=ycl?VU#0kF*ScYD~an{*2)hAm9s<ifrxd0o9`1igu`zArWi
zDRpLBY@d_<ImhL{)NnjlV2iH4)jn|Jk1hH>^QC(%Yn093DJwb5kr9hPS5|$wtj=r2
zd|tqe5Lau-0yA@)eMsGQ*qGh2^cAK4ps$D&3`j`4#+&XyW1*;3aTCWE|J&Gpzbd~;
zzw+6hjZXot0u^lnuO3+G)lRADus*Bn=h%+Lw23^BJjnQNG2)%ViBu-+q38y!HUBi<
zQvtUhR^6AFrY!Vjn{f|PTEX6R;M@Mgnr8)3*(nH$q-2P+`x`9l<=3^srJ!Oo?&kCM
z$k|v-chCki9UrH>tQisO_Eo41mp{k63U>FsR|c&E4|49^*5bW)TXva@-4wHDPii^~
z<4P%<9dFjuLet6p2nTZkXkqXR(!*)fv9tH`_|SdzV#=-Q!V@^rcJF`br}FS#;Dw;k
z0;tevS5pyXVEXJhhIN@Qyhf>-s-#A58}5#_R@toKH&?Dy%VlN{EImC+%-wkWd~}Im
z!dYhCwpgvv1i$@9g0}B>ZEF6!KfRLay{}FhzAcGBa67&%wsv9W9R3io`6H%h9VLlw
z0&a)zkbKhR#%|~&$aaDQFZ?*ibu?h~Dpk^<X1(UBN4i7Vi_1GL5RteJTH#XY=cmD8
z%9l%F6RPBvh+9pLanB~-k0tH4B_sKl%B6lZ^UU)C2vXOjPOJBr{WUpjABVjx6xEh4
zOi9Md+;V%&{({4wy~|nGy7Wl9pjorm{q6kgZ%vOoJ#=p|3>U=44a;3P>1ZN0g7X?#
zHTnHcyg?C8@XoW*Q07mHIo6kyV&H&e3Wxu!S+v688I2}rH=-#dTqe{!Lc}2V4m^?`
zE#Ze7v?v&QmS)K{mawz|1ckFH!4E?kYc8lN?#Yky6|$yKj2$Ix=+jUZ<-+^QpZih4
ztwAO7F%w@Bc>6*A0WI|<hBs|6V3C$G97~J6*1|&46VV(@#r}B8oamjC$&_p(FT){x
zqw~o#t?b99NowD1`usT8{Or~Ob(+%jxyH)(OZhLjy3A*Ul2OhXY|pzbrdfC8_p{?~
zh*5G$mu2e3dg9M}f=`g5rZ3w{o7PsTr*|4HJI^3L_Vn#p(hnYM{rI6TRarV$MMi?l
z<qGFZg_#YQa%h@0M)#2>He0mwq=ZsZ=j(oSo4$@w%@r#eSUQ<<gvr{c<VKK@ZQ)i#
zn|^Y|>+-G&mh#sXT?S3qJ>9(g!P~MM^fN_!Y}loMlO4NFvIpk{DQ}9z<z%j9WhcwI
z6NXDZp?;_NS;n>SMv-3UxjpV6cBfFW6O>UWU51^wq)V5hCxW+RYmK%+fTZX`Ez%7d
zFJ4r?!FWSja&XTAAMR6X;kDdlQaY79H7vAqBd}GLS}ZSq4l^k54@B9j0Ae3E<d$9b
zFLmPNeM4{|!5cgOflzp$;KysYud!0U*Bp_mo&l7kpX8J~|Fk^@$qVR)V+7Ix**3^6
zbib1wz@Vb$ER)|VWALYpfnHdDxukpRmoEOCl_<@t3gpHR(O{%!A<Y1SIGNOd-L17E
z*vDmi8Pv}i>*^ul>(i#=Zktmv3($8H_NW@*!50Yi5%Dr6EEiG&q|tBfyKL6JU!VBD
zy@<jg6i}0)-d_m`@(C1Re$65zc?jn=4Cn$&0=PcfK`Tj^Eixy|1^!-Swvpbf*kAaS
zye%=a+c=>##CZ$wK!*kCml?d?WB|GYnL?gxEN}TYmhV5oc&mKKA`RTmF8Xz)TJ-ma
zOguY4Lai1bi-&<Y_Rz;@Q)JBWXZT+ExE9LZ<P6d-cG!{V(*`j3TD6COe`D@vkLdrz
zfd1{ZHV3Q`J+=M+6<xY{3BZf9EUaSxWV-e%)A+X+Sx_6+4-G2)6H5BGtNZ64<<IiK
z2S^|fkN<rQ`oG`K1K?Ghqpb9kocS+@@4sF3zkYn0a2&SHjsq$k|JSe32JU{{)Rcda
z5r5qb|Nf%81D?Y#)BDn<|KnG<iXwt&-v8ngi5`#@#pj3eu6E7eGyI2!Dh~KkzEHkR
zQ~nEHsePn24gqEX;xCuuw{QR9e8d>b?uy7StH!y0Q0ff9E>t!7io3}5H^iR6UaILF
zW=6fUfL^Lz%j&r#VkxIfce<Ffb>K5pN7o_`6oZIG2n?*(yMGPpRV7IMw(Qev<Xs%I
z4E`p#xR1cRAG_voHsjv?#@fF<zOJ>xjd1&a8lqN;-T&#6sLAl*Z$62(>qq3O5c6jQ
zW5a{}oejkIC1AjqI{dk{Q_%>82xj0)v<@>@N5)(~*f&%2jU~uib_TX4iOfI{v+oLc
zQt`XaISFz-mF!VFP!n9a3qwTOj!KLZz*D;M=@wGcN93De!#-(2&wq!~s*n46oc=AY
z*gc0JT~emTzi!B<%!KGl7!IF$ry2ITn+KTPp8jb%*ipl#`T=$+2H%fh8EK7lkJy2~
z;?2^^!j<<vKfD#WJ&~#0aLncdVKIixQYpN#`fQtWwp~(V_{#Js*eY4mK*&EO;Juhm
zEXw0CSL>=#7D(WRIO#!NPIZi)=;>f;i!?1#a?NKEGJOCo$6m`OMtZ-$>1V}|ojv}U
zF6JkfWl0eWU%Ndh7pKq!?H*pR^dpeonq(5h4tPUCG;-pK9UI}9eZ7h%(oUDmNI{D$
zqd>1%XBK&TEF>b=HdF2zU~>E!;+F6dHf$6v6sxsWVj&NWKsNlS&Ciujev87&-(rI;
zqb#I&i}$Y;ka+-GKeO;<nX0%bMS@WvP>>1pqGb1Eyb7t_d36L5UC+vT2#<svY0S}o
zaaw<8@cyd!+-FvT9T7(_AokPD8mK!!57#r~c3Y4;XB+$?7a)0g)d9UP;=uikxYn#B
ze&2y+zz#&2T0ZWS4E@dap77NmR&>hV$e{<8aBZNvR%G(KT2<|WYisL+=hV9Tbrx~v
zk9)5}E$?**_AGCDZz(byd$p$eEVMiz+#yFVKKaLv<u(vcHjwxiozAZ%Dn*(iL3=Pu
z<n^55AP|6QYY1_Jha1;2jdTyLjp3Aa8&b>BQkQWaTYS3+Pl@;zp=D>+7SRly<&%QN
z&5lLzLt^k>%E;YZhQqOev90uUVKaN3*tXH;7U<t-IfNnG9T#zWu-XQr5FFdfU<7;*
z^p#b%YN;j1^<kZoz1QQs;RJ=6)>UZ08@7XwB7}bY;(2`59lMgV#$r6n8%>A@L%AG6
zplI8gu9^b_lUXVFx}e8X)aPc=SHx6eyGd@3GNTJVocOicd$E1R{w32uxZs4Q?lMW-
zzShx?{{dYmh@qgkW`rhf8AvG7izMqpAL4kbl0kd34;3547KHin!)OEm{wiXF+XE4R
z$HC`x`M59A4Obd>oKaow<Zxv(eib$Z25&>yu~NrIi;Dd!8X3*2KYce$hY&bKvgh-z
zM)obHu}&v{hg2qLrnE-cZmrH8A9Zs%Rr=19DK~3s0$_y;Fy|dKF7c%MIX_@v?Gy11
zK;F!UOc@Qbi{32lXv9wBz%{@H%XY%n-B6@BQY9|M0~k|wx!`iw`9bvHb}PbG)~`s8
zB#q&QXTDJHAC7M+@`4Z}c!sr$YpXknK)=57;uU)Hh}UuwuZO=<DZLacP91h{sL0{T
z0Z*UgtD6g(8=GRmmfWY_yG?{%+X9<PuA@_u%|;(3jq<+jegrEu+=_J@Xe!5@eAgUb
zq(2U-=+f74t65H5y7FO5x0<iy_ecIy2IOE2J%G9R*B)`-gP4N!CMfN%!1Gq|y#n^*
zf!myi?>4iDT3seZshC5J=Waq4g(Fy&>B6ya;ry)PWyK`<9FYB-*-K8p(XC-sVsssK
zP_9<dMfRb%GwZ;(A`Ywc!fXp4rJz~TrzWWKD+4tFB@JZ`;xHR7M!^?;5;=duSyXu1
zapQ@JMBl;sE_6PGSw_n2NZS6bqv{V@;pS_62reVO#$p7``(cmt&QDQa^^<EPcK7SQ
z4?9pIGG$=k(k<%|Xw*MmcxWPU#}%A&9}OV2+F`^ve*`h~g@+lp0Ob$!CNmZa+Sw*Z
z{;7pIodPA#U!6U$vJ6svKCo`m#N}VAcKDsPpb$gqe(3t)uqt*@)`W;gN5qwqm%?T0
z$<te4>!7*10Z=9>VWg}cKMbz&Sxv>umWp}e2D9=~M?UTVy(%%O)ELztjdYalu9v+o
zahPaq@a{s%kDJ|n2{wrZy2w0wd(J8_LvFtN{VhwQN1aD4zFx*x^dTx?Oz?~1<zW!8
z5Z*0RGU}FK%E2KYXc|@^HD@V2IKzrYd`*P}DfElp7^3CEuNdfoVzgKOdR+At(g*>V
zV#gQQh{D9e#X{X%M-eNviT53!Ka~tGopij^9j5!0{Iga1nRUw~_vdWS3vSkhIGDdY
z<4VB~58X_xNUl)^$4p^q(H;jcN9{^e)l(+Y+d$?#8-rxnL0wn2O`z(V!YklICxoAr
z0eP#mwFyU2|4lep^?ha;b`{D$RXhxj7VPLGGiy*hD#^LH@AHH^)WtCVj0t(k8`%1G
zw(`C*aKUW5gj%XEd|NIENZW_#2YS<IQ)t(gzp$=}+UG?Rm1CV>8Jtk^&Nyk4OK1nJ
zJ1YpI9MkCA)h@V5mXx@rdbdMi)GV|T7~znb0$HIv_e*(&t$d|W!h3MWI9Imu)5hx`
z-d;_Lw=HD9856$<_hEfgfzYqUz~HbovZGdH3X;AXElA;i-=NU#jRak?H=aHIE9{jI
zAp3ShSngml>`Iga0a!zHoMOI0&N-9<DJwgLh7SxTDJiUspr>F&5OfM+@nN&$^I+PI
zj-9|<PZ?repcI<}AKsHu%-F3vK!f5DKWm_P^D)fs^%OpZ3X{^M91l<c=@lefp_xvH
zlHMbwB*osA54yKcm}2O8B38H`Lsff7pu86QRPjL`yV?oDZwS6~hHIJn4aqW@ztBA{
zxi4UfTIcfadO%5Fq2DR&9akt~q2-_*xg$LKdruO}Uf{JgkemJ6g-z>8vuOge;%X4%
zShnsn8CXqiUc2{Tfyaivd{+3N6Onbk;kj7_akkH#yR52|nuHOhu_EoSz_RFUeI7F6
z<svoIU-J>W6XM=;0-YM<?1w0u!o6Z|g>X2z0wZv+fbHsey)H8L+iZ+%Hs4e^g^0^-
z{8gNqod;C;H>^sw=!ca@kmhvSBBXP4k(ZsC{qzxLlmw?X=am%Ig3?hStAwtqLp3`5
z_41IZjDqm>Bdo_*SN6qOPc}!eW#mAsgwPCSLb#iOT=>Nm4JEp-Am_=tvEN~HbYO8x
zcJgt!b(`7o$Qx+S^!E%X!LA_=!8e{1K_sfYnhk8d;G`w{QFa>kT@9Yz!y5a3d<&3!
zM3$8sg3)mDl`Zk`xs1v7UkmblB4k=U!Hf~T@$z`YV+XV_aD)k_yY_1yC7l>Xgi}7e
z8E;9e(Lt*?={D#yJq#fnT^D+?2a@BO96sXju$ju1vWc=Q6<7#erTml`RdQmETH_mH
zsUU{RPK%bT^L7BYQUB(E{?Kj|^znyV7i|^k;lu3QkyurgKbI=f^zOxHIFwI0%WGf_
zm5MK=FfE_rG38!~^PB(&@DfBQreNV$%bs`PKNYlfNu%CyUySYywKHp7(X^TWrgQhX
z1H*8NCD~_f#a}I=h>3Vy(G?p!YMIl~K8%cCOw&#pMzO5;7tliE#SsSaQ;Pv#j5s|8
z&(|=p8?6{km&1@b@k#lbWEFz;sD97isR*;ziFbam{;T5`el}`W6^KoTa-Lwtm!aot
z%tv4debi3C#H1!VVOQV+CvN{EBW*Q-e6AR$*03d(-l$;b7W2x<Hy4&!Qrgy6t^<)W
z`Te4RS;=sW{Wle>7LHnT7U|W`DH_WxFQxC~9HMGbikcKgJ5s**!ic*Ug{r*n*uqyq
zOl$8Pk@+Q-UADaYgtO<PSCf=?0hBWz*dkwV@qa!U(HMy#x*+iF(ZY;hTVRiJj6@`6
z-n!_+RK?^nW|?Ix>W515PSdt{6IDT7sv;Ln&S%{zM;%A9{AnS9cupFcs6)DeH8a5&
z0aXFVh9{Ygb>=T)qnPda$3UEG;d-(&C!}tC=>lT@=lHSYi0#qEqfYF$#cJj)T=*|H
z6!VluQ;!(uH`XPc5S}hmQ*B9qJh|*J9FcP*M@%4#GbT1tO~rhBg(Y?-hwh4fiH>8B
zC_=_u6!;K*)ImdFk!1S3ZaEH)liAp9W*z<tbo@e)-ZZUkIHoD2<eq0uVfpuHRD-!=
z4_5`_iC4y-Rg*NO3*L9{*}Zhk>s3}~FwwoJG11fd<m9Kllk?h#Sd1c71$Iha+%IV%
zsl29de+vJkdvd>eM}m=sngA<z%(G)%RZ#x2PE~X6Igu0x#0M!Z>#<QRt|We~iQ{Dh
zZ;oR7ZmXp+rPCXZm(N+iPuu*WQypAZ-0Tfqqne93)u;o@4d$OG*_}V_-RTjV`pOjM
z^pW4axOUR|YLr^(9GC$#C!2VDXsx`+rIdeiSz1ZN_Q>u@4-*ZZ^oXxwj9ng$QuXnl
zeR&d8{l!AWdi-l*dD5AFY&<Lpb)V9BG9lScA1`=&x@KImLyT`@1~>5Ao0Q^ylH$Eb
z7x{i=@G@8exo@7tITYVfSw<a?IpkSu+-%y#o80)`PDVsVVtlQO1V?IdZ%9Rw#9w~K
zwJYuECmTl%h3#DqvzqLQi^P1(;!81e_MoZ_5BUD<l?aJEy3J<UWbmk42XL)I+{5A~
z5f{Ss0SHR8xT`np$@vJfL}~ZQk_RUT=+a3E!z=vt@5%eA7#dI*P>4s+^9N%RgEyxs
z^8o1cUZ1rSQb(EU4X-Z>^AswVuzONF;4Bubrl<+SmsV#ViDh5UT`C@DqaC=SCCq!k
zfUK@ulRL+SkeCN8gOeCiChcJP0R+x-q}_vB-{62uq6LE_<#4cMGL4PFRRew$lSR#z
zcE4M(FppuiT&tA!_<>l5uR=-fBsdh!VHn7HmBuZ4IST^DMvAAXE!)koAa!PcaVULH
zT6^=aKk`8npg$?QZDhJnN!xnynO<_E|2Em<8igqe$^NFwfD*P?epHjr@2R?aE$C^y
zAY$9Aa5%@+M_5+nVf#F&oc#0R3%tHc5g&Nt4QyvIJu5*{yiB_$E3Z?3@DK>|*mfw@
zUm7&WF+NGu^f0$yBMo07XH{k3f6q$Y%T0EV`r7Jm@p6wdgy{t9KZdUSY^(G*MyPoz
zQaD^RhneL0&A#0O&K$=1ZRY6{M%`y{M~}}PGd^>s_ssxbTB_5FMk7tDsLHU_pfoj3
zE?ZfW+chs?gC90J9JHFexv=(r8}~lqQ+~C!`|#R%%UMmu4!$`-Ys&Lls(c!n=o1Eo
zRJCdKTrxMS6}B{bzeW%ZHMB}R61@*Z$%V9{F_3{iCo;cCq;u+J{xOS@%ks1nkv`v|
zmNnQ)Rl8Yfb@qlKKgl;Z)|l;6lfVrcZDZ5x7ikz?z{Rx&Vy9(Vc<w3Yb3NndTL({t
z+D-`#Thd$#p$be(=kh+4{;EANClsPi+q5W75<8CcOwZy7+d-+vcU~FT=;2cI=JU(T
zR5Z>@o!`r%rg_7DgnP~(j`xk;avf>$k!gi6hY#Vc{i%;F60&PN&~+;57d=;AoHD)}
z)E|?^*%bq=D&m6znl%;;KU|H|qFo5{A1#(NFIl-<>aGF6JO`&#73mSDy)x>PMm#O~
zc8ArNm-jqSlDS&N$#<&LrnpW^4CUykIJbJG$_xxX`GUfA(s$`-X2b2R7ZiWosH<U7
zCGm@f5Xab=F3VopTi`?$krG`RzO|cE)EQ)nk)6uMJQkzr3Ke@;{JP{IZ>|ix5sE)M
zo#3YVm_U$H>0O$o#FV&_{Vqyf*>)(Yt!fxperLdIXr(!5_<5n;Aa7CwGj&NvQLJd!
zgba!Ji)7<tr=$gc+^?9>km~qRfEk-uo|&!wy?U{MwA*fng&O{hCUwYIE!kE>K{2B!
zr9Tajo+u7YQy8Ctx8GO=L5(*InfKTlc6zyK%d3HD_n8+&GNlkuH{eE9RM!snKFPTB
zMVs&&Wbq|1cc9QjMB$PY#PW*qORKFwE47N;&EF_4TWAg3xs022fJ5D%LUY6P;+Inj
z1e1kOXb;*9<a|K;z$~acAphQ-lDhp2w9Dtxy@I~`c28I*jM&RK*29RFvn!AC*_a&^
zT;q-#Oe92C2UTK<T#A=ls~aTPqKfrUDUZCNO{J{ra*P~Llp|zQ+IqUJ$#3bWdOF%u
z&zJK13x7pb=eytGUm>+gw=tw92xi1t%n&D9ukD>Av|c(QOcyE88=K*F7S~NX(X#d6
zdz*LI?Xfr_)?Ue>Pm<c`a|CEr8Hk&tw2d{K*pJuBC>6Nl^`s$Xvj;M+V7~|7HBUfi
z+WTx1Fj=BF3R}B2nmr+cNZC(!s^3V3PzBq871!jb#n3{EKGo(QtGr*Si~n=lomJ0{
zA7SE(up4;-5?6H5-o&=$J3xFPl69`Z9{yh`Z$}d7<09ApqdfaF)bBzxshN;yL%zQ>
zsdM8X@|B+R+wXr($nD$s!xwHI%0M=3egPlzgFR2H528q{7{Bt8wjKSousa|S39(Nv
z?b}4K{k)V8&y)em(0maNf?npyRF8%}9Xt|@uB?`^Ip6*QHvh*R4`_5=8;elziQA_X
zlR_+y_kR-59#HICr6k${$t2xM3{3PmfDUhuLDDD+Q|#nm4^Wr^k5YJDVk9F<stty-
zGd)FZG5mR?_mbX=HKv<rE2jtvfXL(miC9UXO0@B7I^G#5$BU$-Iehg98qCSc%0t|%
zRLY}m6|#x0MNIU3dGhovAOkwj#L6jOoDJ5WnVYg7(9sK>+1Im%^0pt{NQ#$bk9T=a
zfJJ|q>r+&)%Ri(;eoLF+l+7iIXW`90(qHoZ5I%uz&*8O0o*nQ2OP}ZRI{Ih>PIj()
zvGGlV@@W;_HM1jBg{ef)svAZ&WcL_{+AEEzg4thdi53lZ|44zrRdbKU(~{6a&v|!2
zDV&)9oS$O&({n)w4qZM;kyEbtjVg%hbc2ip{HJVu@nszuJ?Eows(=)qgHLW%&rgib
z7$sCVa*Mc5mnB=4V>X7j#~c^mdKbCmez4v9G0;JF&l+uUbzQ3C<tLsMc1m3?0^Ryl
zzhAMye|XXDL2T%@X-aMUm+<Yuc^?YstV>`N&4EJ|KHJ*HRR<nA5QxcEL-}eZxR+->
z{=OU(D0tS$B~eiD$@jC#^)d7O(lN~AvN($9#lh7%jrwD_t%1c7fgmpCxKolaSqNV>
z$Q4n~!L4jy%x<%SSH4tvhsxd4?6vqUS4r-rOj)wEl_{n}LtDq<12N0zjh^r^&gUG#
z2}kEZSEvq)`UeNHQ%mV#{3a_vA8<1B3Ol?=|KuK{7Kx-+4wPID?nPhduMSFd_I#y3
zZ=9~4X8r-v`zRRu@cfmS&$O7b?cHk(G?kC&Z@l9OD|OI`d;;Qt&{40rv8*9vNgYDl
zRCMxqWgMVUv0h>$^KN=Z>pG04OHPAM$-p+O(hRLD&<b`vLM~o&3-vK)1*I0-;*{yU
z$;RI-p+-`BWY1lXXJYX`C|2MAf=&JQ0oh^qQwdX`6|o#PCTM5wboP9ge?3r5rv);m
zB(<7Jr5`WEriOVCklGAHFI6PTuIC!6Cz-uAA2AxKqievmbn>3iexnDc`I7|gpM38f
z2~_hlpD8@xHMeOMg?J!(N|YufmaUjZ%(w<APt#wU9Lv*uS8_M+eBOptcwq3}+|7nA
zo3Exda_+@V^{Gm_RX9I^iKY$tnzuc7RsX|%@IPu-x$3osuMcM3|LcfYJw>#mKi}5C
zT_bwyQ`^&019#luBi19VC(N|?B8$1!SY6#FtIR$H@?7FOFN4s6q$D+V*Tm=Fti$>|
zI!q^=_Lf+_UiRuSktI^?K*4A|x0`c~kk_3!#vymdHH|<5A@%wz4F7Ouyt})_#oX~?
z_SgfCdg;Ym>W-iAW$!)spDi*)fC`p4s~5ejj9)hAi!!P9(4i)HA-7w2`sgYYUJ}nv
z`t{C?!izx6u|ssdF**N_v+oY1y6^u+x}s$zR8BV8J6mx$I7vh{mA#UYEhFVv9fV^=
zSxMQF-I8oMkx-Ozj7av@^ZwR#UH9+$-S>T6&-46ONauXdXT0C9_iG3$npfb@;P$6!
z|6CJC<UH-0Wv8wyX8J~V`(+r;noyo<I!M8~NXFqLEvB>PUfhCk59k?Y^@OB2#t}cr
z2#g(;_U=G@dNQ6_72!?ykESt_wTKq4*yrg;$}6=#m%K`L@XlE|2ZMFIm=CB}Jgdki
z<Jv|(jp!N-9!j3fY<6w1c1wFwXh_CYUbhj38PyJ88E+&M(7YHR4Bjq7{P<sN0O{zq
z@6#p4PBp4tjT?69Zt4aHqmPD(H@R_xxvOuQ``8PPuDL@o?NU9QbJx`}a@<akZ4f_3
zV0>=}UHuLi?bZ(p>z%K0zN(%%%HMfmuz^YYyHEJ-OBk3`ULAcn`Rk_?3C}?o8h=J<
zeXQlI{A|K^D_80G<MAJWp%=#%McdItIJo~bFZL?}6jHYl63!syav9E8!xrtMQr8b)
zS-ciqZR`{+4enlI;`M;)zDB|m?Ev3uT1gz)8U#?<9Jl<SzA$38f}gyKlx^K&opxW7
zZG(Hr_e`%tm(SRunUFFWfWyMn#%}Aac39O)W0~aKiqd*_x;y*pc}}^mdz}r9F1nNo
zS(n&N=v$~|Z}VYtBm8X_^P`b;YO}QQmM1-|bbC|oNUw6A#g;eQ1mP7|a8!;BhpgC@
zDN|vh#W_EI;tuR~-f2H5FSyY${MrtZjjY}u#aI;mJazrA3T-bSMr<sh`G&a>cDl-0
zOmzKma_=(BBNUMFJWk8_1n}5bBAjLWl|VYGa!+kXaMl|wteP(mWEp|IcVJx+7bMH@
zWHK-#V8)<{TolJf)09-6sXsSyZ{b$e;@PlrP@%%UIT?s6Rq1J4GTJcaKb*HN+KIzS
zw{YD-_ngnVZ*!BeiD`U(-2M(sV+1(P<*hDgN}lSZyEF?Ib4x}fRd55BtM1_2VZLk;
z77MQG@m^HNQiMVihA#LwFWhwGq1*a|@s@z#dn9bJx@GVe$PvrFd(`Jl?bEfga=QM+
z)uZ$u0pis6q}jf`p0WDVJ7dkgMs1@1Qik^C@zBHcrZt(@FZiwIvW)lg5+?(Cc0W{%
zxu><$VCGkM6ftj%9632>+GGQUG^*}l*wo3k3hM4q*)HY*sQI&Q_s_qdsq0u=xY#}8
z9ceppWI7J>%#q}OV9sx22n;uvx~=*<JG~R)Do17L$OP)69K4%O(0_)1niS0}#?e4n
zi+|2Ou|Yr7-L-b5QAVgE(~j@&YMNlG3RK8EU)c}5CHEiLO~_%6&m1`43N{blE`{kt
zUrJdI<1VvG38tv_-A6r6G?`s51$k_TLVPAN6H$}$jCgDIUX<yWGZ5TFnclW@%D>+3
zQvOlrJEmX?qp$W{)<y2$qQ2R@39^>9Z0a?;Tn1&0MC+HhTkY8gQ}?-$l+UL9NOnQ>
zLU1+9_MQn9rF8lAu{#|>f+EMPMxm4{LzMp)ECZLbMp4V683e^i<u%L@uq_JL-6Kfx
zn5$PqUJfBCJb?bTzuw~^E3%NAj+$!O+H-Z$KA`{c^w4#fb9ZDPPi=Gaf}*7#q_7tg
zUX3#^Hiy3$3^F=t*(u$6I&Ga;O%m7ZFJF^WH?3;E{PESc@QviQm~N$thx-nb+Jt^4
zs3Eo@)jfJiI^}y{B&y^Q(QTF_j!(61=&A5cQ1pmsG?PRo<Sme96oi<BFxz;}6V3Qf
zD6GKTDxzWXW%ByM%{ZF6seMZlMlYfRV4E3vd?~%^TJuF$sDsl|?U^y~>|&n5gsnB8
z+Acfj#bU@F6i%sg`)XTW=L|!?dx`e&Ftm79&eb^#f!&;ZO;tHTH5HD+!mpW4L4%;>
z`qcJ~z45IB$BdjyUte~fgSw7XW?2z}5s;YvLXlh@%QD)=IhYuugYt&KURspVyfu?y
z$7^d3H;D~&rA%Uzx8KlfHpJRK`%{Xlnb~c(Go})K-4B>KVf34+dv;@QL<4}`6HQf~
z{cUDvp1>gSquHTouq9biyY1FFZnfj4-_vo^e9=Bs{rD<(&FF^0Kz6@U_skkDQgZ${
zNzq3vC@(NoHtPD!%PU*ev`0qcWiw}q)z2c{FnwzEC&B3T>C@rsgJ#L%ctOVoO3%9U
zW@F-n*cjQC?DOiI;!ks@;5d;WfU>bkY)Br|)>nH@)Zfw}2@^h!*2gPmT$$+RWmbjs
zv#9F!Bg?WN{pIklBKD>5`tW{)Ei`2nrS#=<d}ei(D{4xlNnq|^Q&?1u+cV5gly8$7
zW>KS<blFvuGh2b=vfz-URz7@UA=xP(Be7l3KPT0WuX9@e2Ju0eja#qeMK_Da6t!6t
z-K~4Q#R4O}r93#zQp!`_tX)B~m{Up-Fjf}1nPQm&|F8q4fiqh8={o<frt%aG&vX1{
zgPA#%)YiH5o#3d+(B{yJlh9wp8hsGa-kOZ!(m>&MZSkE!;T+GO5mUn*MX|OaV;`dF
zzD+7cs?dfUGo0&I)eN@k)?@{6K$>Vipq>a9*&S!$*KU??mP)8k3Ja)&0UyD9I_|29
z_(-@%4ow&;=KxxPc?S5d^QtF9QPi3XAPCrZ;*nsv+(EYcz6mv<r`%y1rHSO)qg;5o
zr0_(m9=alwtVqtj>C9HmT)fucD?%6IINB80bMyGzdd>klmbfcx*uh%KOxu+c0`B)Q
zUJ|XDNsRq_rn(3s2VX>fwYMANZw!^xy)wqgC4_ghK|&EKc3&ICZ^pXCev<xp^_9xb
z)b<o_{BiMcjG*`cp+@iU!YSwb3RJgnk8&>y_<CN*)M(4RPV^F0`FL6{G!fIWzonhi
zKLxWK>BpwePD;qD(2<l-$1XnhR%RqqB<9wGX;b%`6cOg96|W(A-5IP^0;sASf?AYb
zaSo-3IHW4-yQS?s0n>mB3U5wZE8)K}(+j%q8%^G!v})R&CE>B(N?E%ds}TZmP_5x+
zTEpyOC*KNJM2Or!5=c!>o3ekUA_Aijy=az+@s}Lr<-?SZ-cUHA{r&XPf<}gzPB%U!
z)1Ib@qwM%)8IRq4@mdo*>Y3(#$JaTrViOyB4@kOH+1qqe8cW)ZSMkzE#Qj&U=h)`x
z4OA4OKTRH`@=Mf11%?JF60by`z$-YQc9(q5M=~$bL5*&W%Z@FaZAa!zrg!hzkhGaN
z{+p;$9P`v%G^d_d-MEzN#EYzyxkJ7kQXNZ=Y@hXgZ}^UjR_kd0h?og6-JVyvxIgb9
zu{PPGVow<xhu}*AtAy(KInE{O977VO=qqlH!C`&r4>HDoE{{a7z;Nu?$@Jwzzh5Pj
zq7++IQH?5Gf@6v~$Fhd#4n!RPF^JVL(b%<hDq7~|lFCvwcal+-M&83Dnl7FK#%}A_
z9vqhjrA(!)dbP$ej9@!Yr<=A=M6-7yTK(Jj9)HUtl3|A^_gJb(XR*<-#vOkXp?_=a
z)JXUhjv#~h#F5Y=KW;Sd@>kM?Ks8Tt9ga^ikX=p5JwO(*H}45oBR@x;E<WL*$T*Yx
zf!S2XbjOzBA*a{^oTNrA2f5xAwAS<<d71NejmHP7CA;$yn1;}l*-Gdi+1it)CHY*n
ziKJH&K8LsI4_7dS()n_Z(i)c(_OJwI>G|^Q8963M+cyc1Yv;2GtPD7UbKQIW1O>=!
z=T09vE$S#jS6_U%g;42ItUbk0!sW-Y9bHPM%<<HKp@JE!A<MNv8<-@`MyHLZ%IUcw
z%dXvQ-Gj&FQ%U4L>|bflMyh44uljJ`93Q<S8_9QXN&$x49*0O))0zd6t9_ewl|{Tn
zq3uwm5Xy{|eWaj)T4H+3NA=-*cL%jJK(aYoFOoXT;9z+sTQARsWBP;#+oFa4BG2MJ
z7nMU68&WQ(w;hg>u7rKizl~Pvysq5rG?e$TqwMpicg^0B3Q&b~YdjlKGLqh?za@?S
z-jPSF#MqPMO;Zc%quwnBh+Do5FW+V+PYT)}BATk0QLi)jpBKm<O;RCvsa^b9k*M_t
zkn-!tM-xs-kzY+etU~jPjnEuJ7MEpwf%yw<ilOG;63gx_1HaCBTU!6Xzes-w?Nk++
z*3v$1{Ygt}k|odDLw_ljyN*V7j2010&A_Pb*u?nTU4L>~|MJiOV0A43YfUsLul283
zvPlwIo0<i1KT4w~p-QxR2okGs^}vmlStl6nlxj<MqlHYB8GjkVQG@{4FOhWp?AKQM
zw_qGo3TmOjBy%rgq|%8QznAzQ|KmTuXp#x!r}wU^y??x<U*B=(1eZyeo`uT*5$M?m
z;E34O`z=RqA>p=2QH`WBS1anU{JMX}bjV)Jfd-Jzy4_5vL-O4RB$@)jWtQiLpItqB
z^dCMm@VmO`^fXiJuhB`>49rme3o;s~&O`32E%5RA6(0xw-2WlM>YpKZs0^X|o-2Tw
z>#Z&HW@&)!6nf7M-5eeG-}?K%a<%LOfn#g>?0~_pKMK>oe*f?PHEjq=170wVv!{Xa
z-!7fMKG8p)IQqXVssHxl|Mqo1Hn`OlOS%ng{{LS7dxSH)Ja|s(AKvGGc__a}psO1u
z{{7ni*Bk!pJNUO1SHu0&a}czbrfJ|v;0zfVGfCI<b4|jZLn?K42_#~;-c@;&!||4b
zQnJA$=l(Qb-ay4cqnf{`gr1^I<$^b=8C>%Rf`-0lMRg2vvyoh9`RV-Z#}M+$x)^&U
z<14fcR$zq}wC_nec=k9$^(r#s-4_G?F+JdULx+ve$&Un;cSelz&YIdA$4?McoxBY_
zsTZJ1J$_i>K1I+Pu0ZbGq4C}`ijm^em*f$uLifoshv&iM(pNt}pJ8|LrU>agVmZw)
zds2Gt3$*a1SHA$vWK-|IDm-I2IQdw*@B)hUpYngcoq@IIeZ{<2`}ehHlKdRZ!W@yd
zs2*}2Zsok|eFqkVE^v7lYXfj#o|hj4<R8#|DU~C%v(}2wq5B3P@{DNH+NZnQ7vUKK
z7q1lh#}lX!%5|VVhBb=A+c~*p6o&Y(y}k~vfw#W7F69BC^#ufFf6JRb&DAlH;ophS
zOE=*$s1Bt7%gPyX_(OW*rhYI&dT&o>N@@V3!5;P7$Bg;r;8=1(xKv0G_D!q8vhGg<
z73EEacFWW`7Nm!gVkhBYFHK$Pp8(6HS%4FxM;Wvw;}GW6`v*FiWnbk+o&zL6Hx~ma
zIL{C-UU{cVEU1DY_egtCfYA95y!un#$WoB6ESUb~LB>Y$Dfs!l(>GK;KvbpPbSw1>
zBop{4M0U3$nFqq>^^<fmAWz5{;#G0t>@JUiT56sbz-kZ+^bsOr>l8``Hi9$*I6BT5
zcxUCBJSjQ%-*8O8<b;O(tyIG48z<Qx*Mkt&&6{(XZR8Vp+qoiU2|dpYi!Ip6=vm{&
zqv7;&Kp0N5K%=T`2;7u5YJFc7WA6$~7L)rWK*xMamm6)LU6@c~mFD$+ZThj&Y0vk7
zX+0FpbcR=eFp%_TZo$~(hcgi34qmzn;7>V=Z8CdU9RPAqMwWe#BVo!h3*t-&7ik5U
zpbzr$>+bgE`Sw|^|9$6xJ0*bVf%X>Wi2u$3horiz2;B`K1nV^dHuNc%x(?{3Ked;@
zneEG^l&3ru_6$Z#h1kWflP`a220`K`#Z7h)q3oRE>gv>5et!dZQ}BKu^f`lwH5%B*
zhr)$%hOXuNPm9dC`aW1iESzP1aMN7^(TBI4LE-rftaUqFft3klAIt|U1foUczEn*!
zXw);AT|JEp5bWrbzirbdQ2&)VTcD^aOB)Oi#KKuU-=3}nXPsMAGQxX00Y<wFlr`-R
zii?I0@7!86y8fl}_>bq@AM%y2f24}QlDYL9@yo>N@sXD}Njub!nDsrrX<Wa|EbsLg
zVZS0`lbe*OF*%x`K=wnJfOMBi43Y*q%UzlhXA<pF;K36GyPV8csFTPtf_BzbNk7wl
zU+xpB04s}v$LD}}bO*@<{Tadl7u=Dy?@8v82M_iLklnt)F-7<H_b7b#8OSN}i~cBv
zgwVknr-osVCM1t9|F)m`A`)Ot7joO)S1?fS^`Z0VL)oCKE14D#gspc+5yYT$)Y0um
z7ig=4$4|?y60ssIS=57qC!59ZLjs4JnqI;Luqw;pWE16x&=!3`sgCB%>Owfhao(?w
z2FWAb%6{0sgEfud=a^A&?Mu^-&1XNN#Gk<`UjewWA7uKD&o)2JIU|0Li033vnsb@-
z;sjTlAjg6(jZvA|48$a`(4}05W&K78yWI&!O*iY1TF3|G@|^xTaM!HS#sX&F2b;ik
z0>yDS{f$k56a|}$#xO^B(8VUbhf4xU{#@GNexn->FGq1@a6^Z4_B3V9lJQu_;^eE`
z2>WX=wd6mR*F{uv5t=nEmcfuJq}p`s7`iSyI4TWlm2ARq9*Mts*rd4$x<)>UWvl4C
zK#vISq!KW^tw=QJuJ-%t24D{hHfjh8ol1~`GmR&HOTWG`rH&SBVjIoX3*Eg)Zgpe6
z^~2<+TJLB)3_R4A+K8K4ETRwIW-{+WA29>?iX)4Io@4Z|dc{eXl-qaeNVUYW+|8I~
zi<qU^C12kFQg2?u%@rS*O3r{yE^ZyqeJ`cf;XJ8fMUT_NHq}mIRG{OAJ#OyVkA`|^
zw8Xq)C*Tq|HX5ZtfK$r|-GMDG9xsnm(#19;7HRH2mBh<yg7RR-&e~3h58z7607gGL
z7m&1QyZ?e#w9b$wwE4buQNDH(cn+4bLjiI{{GZyyl4R^UtrV^UgZ-wjV`aasMAFeK
zI%MtH+DB8I3VJbb{-Ff8#|sPc!yOaTYrj;5_6`&pE-+Ft)NrISMQln6D~MXRXvJ;Y
zlh@{Vz~kw_eX7&c)%KuCE(O+k!%X-OIOsp@0F^ylCVVVh67<I2VMbTMxUo@Mk7|M^
z{p!nDoYzOUH?h~tXyX`sc0l8P{3ZDWt#8#H<1acv8+W$n@j;S}@s3k~tW?0BlFO+l
zI`Q?vueAr7_b^8$adC#z#N6#QqZcp{ZZ@k}OYB?>)_6gTK(y@|VTs=+n7gXrVW~Ep
z3OFU6{Y~R+!X*e+W4GY=fv|A-tjpmlJc515fmVbofxVMWD<KbtC9eaGOS6|!Z+|E@
zAVc4~@2cxP_82)Pg`aJ?XJ?+X+d#QD!@-a0E}&I;l&d}n!|Bcv;K!dpbA@Rq>Dx}3
zY@C;-d-x4VN>ro@I`?sedpuL#XY=$c<3tq``!|Hlw_=xR$A0V+1Qpqg1~2w|BIG>~
ztgyE4fFwZOBASjqa5tS@Rs>HOoYY;FkVu=#oeQfMD7Y-62!Hyq+;z~EDI!VejMoae
zIK>~N`;0`;amQBLIE(Ql-<V{=3Oz}10VaePlMPEcWaD~4K=$rkqyiA2ur+e!h;49N
zrB-q+u$*D`pn3&{oAw&PzMbCv?(jMj>yOM1_S;F<-o>O5@!+!3G|5C|#YMtya8aD3
z<Mj1rduxD==}d9B-Vt3aNz=p}H}CTUkc^Wjm^>I=<gJ@yCx5uq2NCOtU4ENrKTFQc
zhh)w21h*<!?ubvOiR1;m%m^-a7u_y8Uxy!B6l@dKOb$uwH>F8;+`^37{F$k%h|CXW
z>gFO&QJtq$G2-Btc*gNtxxVzcixc%{+rf(N>O8uY8i*tKYl-^5{#tVsamOfBGdTXM
z`?!J)KE4pKquR(B#!6WWNl`pm?r?xRvRBRDP^4>GV@Ivu;?qDMRw1_VP=+F$Z)*V#
za*iKJk*Lna(HM)keRAAow`8ZJ{c<oQeDbOTM`}L|{81E5Ff=+=MwNBcot7?)cBIM`
zvh%2__n#=yQ=w4dKpXLCqj#xr)qu`cMS)LIr$X8K5fW_Cip}Agj&2XFTY-yNO_pyM
za{K7pmXb&z0bw&u9=Lck7dmW@wW16i9kp+ioExgrLFKoxBV*K*(m5<iW<uH;OwFvw
zxNPd-!B4PY!+POjxCp5q#u~w*IfaJi*9_7=aqe`l;%dCX3j?$LS<U((vFD8cM^Nan
zYZ6^oVejJBs-PrEpbo}~qq9`ja&*4ITW5bjJ6>Q?V3^K%eBh|;flDz7w;49D{v5m<
zoXav`ZW_w&%l<;Fh&dZ)-K+qRb@E*pHL-taQXN>=sb7WwJtmcn(~}!e<e<6*SdQjK
zlp6Az<n_JH1BcpcOM7z8Z0ypb)fJCOZrKtATa2S^mYh1dC)&s?q91{EodjB_Nekv7
zvp|eJc-r?S;|1Int<uSU>pRKA3+Z!2@x<%zIsK*HJ;i1YwqGxIiTt28{1cyMWf`s5
zO`HX7*pWDW7E_I=cg25r0iemzN}NaR9UN^t&n!CH2ITsFN1-#ab;|0lZS35)6xVro
zGS2WR6h2EyUdmsm7-ntXckVVOr@fT9%2bB3idf~IeNlD)I_?d6`Y{)o($ZU1F*|X?
z175~=`JS|Q)hW~UfNSv=T&zagt1Qohr|LNA`Q}b>q%WLlu>3B}yTI$tZWcS3a>Hl!
zRgI+7U|vk8RPOlOhQDPELC};2#SUj|{2xDcDvIvAjX1-Pk@z)YRRrc4)~uPYAOz!%
zHM^~0Hm0R8tuvx`v3q^y_K$nVeSIa%8VMU%1#!cXxb3|JpRoKeCxrsNFUk8<BQO}O
znYhM5==B@bGG59`n+<82$!RDoSd6%z8=-5C?Ch%<y=;=uvHN-C({sX14D|T2q&507
zo4zW+lw%;&)$Ih!b)k8_D>i**VCLHbhhXvypu-Vu*KH;g#f?im*@aVp6wub`@tiw~
z??Kd2#xC#or<qt7m}s1?E6l1r%j)v6ZpzwJD7#QmjY-rpdwfEU@J(_Mh3Gg`c7*40
zjCnxf1bkw_V^&shc4-Hs5X%}YJ-c1yh11pQwccpvIz3{2bdtu>L@HwnTfigyo=f=Y
zA)DDYTV7!yZsEa0tgcQEhQvh=;=#i;?D}UcIqyduFjwftFfh3?Y<LN^<d!$yd{%~K
z(GjM{C<?^54aDt?#c}T2ju4*(iwn2(>m=g2v9|P)tSck&F8v;`@XAtZPu)cthkjLq
zr)jf;1`<V!=yDgtor6bsnH$9-<>o8w=I}V5Qdp8#_y)IdFgFrX_Ododa$L+ViDc|L
zOuEiAMsH~%cj*ehcHJF?&`TrTT3DSY)s{M$_7d4g&gPBY@M2>Ha(+ZxnWkJ{b;8}M
z<TRgzS2A-HmUTK-6HkALUbn|Gbuo~1@a-vztg~NL>=we6x7bPKOes;F<)eC+#=h!?
zM3F2y#e`V~g{PWoljdhA*d7dM$5c3EduZ3*?B4wk1u2MAOX~3dbJd-~&GRWmU>Y8o
zxTEPT&q`2I;4d03%QYtH^Pso#)#4ZUoFg#NS_<MuDOY0yY3X|q8^)OyRPnW1A@a9`
zfHMjl#<&A0-9pDCD!PVt$C0=`dg~@<6-PAvY23Wc)$#|Cm6xjLn~w4}<hXUudp?k-
z4yH<Mc1VoJct&5J&}C{nJJF4TkyV410?~#&$XXSjPVn)3&C$K*;r{vDq#hhjFV(f)
zH;3XgI4ST)J{ZG$MjDwo6+GQP7PaM1)a2zxoEk>jLU<dTeMaKe=q+1;J&_TCnSNwU
z$)()1-&b5mR5-RIS$PL+?cHnmazhzB&Pe-mwN&lVlu6^KKAYa<(a@$;e4$?e-*7kZ
zO8H<%#Y0T*83jL6;(K}WskZD6?gp-_DKlw(>Bg=1;u`LD+z{s!;Mo5l>twtr#TCVe
z@0E0QGJ1=&-(dL3I7-i3;E7CD^uBbZB8=ArVrWQO11;H;7yQp0TeZ)ozBUL4bW~Bc
z2s_8(W36~szfuhO{<wzh-OrbN1KvyYS<k(z{NqsgPjYl1s4bx>Oej7Cy*~GcRo4a1
z&qR|;)SVz`s&bq+F2IlPGM0;drq%MA)kBvfiOK6k{<Pk`e*fZz4addFbEGVz*W4WC
zeL(0m$|G>ADqc2S(XSm8Ol;M=k(fkgJ9RNB&_+|0kT%dBWfP!y`I4Z=8pzRduL@z+
z{0_Y~nvP#|$%WzQM>9Whe2%^s%X$eVH}+(uVNzL|{P$CZSh5FdS6jrAt}XLjn!2B*
z^P|jJI4v0y7|m@{&bw$&!_<n%VWkN@mmgvm?5G`*8f+s>27+_NjR)dwA{7g=eZ$^3
zi}7En4L;mQygn;5Y$~=(*~k?oy!wG+pCD~pp<hmIWIqa|MWxYRHSJelM~UzGDx^w6
zRBF-H7Y||m%A#dP<HTz|9v&U$Hcq=3N?rE!P^Iy}9kC_2(~q=yzKrt?x5+Yg^n|*c
z@2Kg=bYZtv`mBC8(82UE2gZn(^w_7G86E5td~{bEsht9x%)d$fTpSjI%atcao$lW}
zERMH!gk?MU*yK>sZg1Z~1(R%6fU<6WT}YMK$Fft|tugFXJ!m#viC6`gSw}T!-ZgT(
zTWO<&l6B8PLO*lyk7_KPGV@bUv+@$>Ux+t36+XG}u^N;qshBN^rcg{}jN3T9TRP+`
z?`nmQWyVdp?P<C5RnrLWUi0{9w{+q4ZHNMIB)$?fYyi@ge*N8sSM)=IE*37-pNVNl
z9p0%=q!R6%{!75kpCy&w5d^&~d~IO)dogvIjQzLnFVHVKJXJz1zk$)5>SC^(fJNHg
zV5nDJa14wh{IZj6wyBuolIXiDIlJ(pwheV9>+xlzxQ|-yj{Z-PlR>`Xfb@sw3MzUU
zp)TWN-S<@vo%U3K0qtJG1`)^3^D_zpig>2meYM?pDWAhEF;#@Yn30@<&AoLPB%)bP
z3-&bPX#05^whqF|eq$R*8kByuUn1=JtsF0)PyvhpS#Gs(vLFF>Lj0KAyXxo!V!(6<
zCnFgcU|m2GR2;lYgTR)NvdxX1Ea`A?>FZ=`i{K|B+*hDqF`Kd>#l2DyR>@JJU24u0
z(-F<%JGo7MqB%R<<U$d5w*J*tRkC)chPDeOa5NOd(cpz~e%hA@aFkj!cjI}|_Z1=l
zeMkD>q0HXIF8=!1b?4SptZPBhn?1)KgLu_#@B|)u$oL{gY*Kg3M2pP&o4HM*fr`F*
znc7>`P+kLiGCHv2DO{(oe-`(6=Y{`X-vI*VU<g3~NeTfTi*Eqov%i`twI=FIO*Z9o
zTT*IeYoW(K9rFyxFoe4BvD`1L$WxwT>zqy_C&&3G+@IMs+aZF9{b0gg(-R5ml+OWb
zeHXyeMN96t`t7c*!Tcg<j}L0@D}6z5DfH2Pa>Io>xIywU(zQivgf*b;p9n_YUfhDn
zN~YV2P%D+<pH~bxK4OH=5h9-j@hic#BD)<C>E*inb9FA2th3#HI3smR`K+@XnH%oO
zKQpR+Tq&-fPK@FoH*`grpC;pSJgzGqw^YE*%(+9Sc}LEc;W8tUdwY=Z?Dk|7c*A9{
zyAP}^b}5ISLCd?jV)h05c6j%1^zxIDdY_heUGc0<kzO88jaflLHeJ9b&gm&t`y(@a
zV`^R<sa^8fQu)yQYyW*I{-Cfru<iof#r@Q=4Y!vt=3TNA>9J;}Cn(tBR-j*alsq8~
z-gmTN=hYJ_PoeD__Ru;!eP4l0O-Ynl2k4(vCap>gv&SaMgTbm-(L=~7Ir~qoicm;K
z)wj~=KMWoU&pv-0-N?zxjjAr{C_w0y5kV1}D5zcv`l@~C`g|y#7xX1$fYf;(J-#{7
zhq9D--eyzSgBx}Cgd0ogjG6zQ0G3w1G!b$$uZJR-)T)`dszbV7Melb3mUMa~E-N14
zqP{ofx0s+y&Ps0>vwpo8bIdb>+l{;|&UD{u)mBQ?jP3_lSF7w8X)UvO=S6X*vkZ8)
z3trl&+0%K#S(Mm~6(9zrodLfnYF~hx$Ga!MLnQ=8KGK>H=-Jtr{WdW%n@CRB%c)~o
zQ%Rgkz1^Gk1~ldLU7o%LjyuAaU&|Evoo<QnYkltHRm$sj=@A!U#GH9UuOR|6C8+ep
zU-;bXHArmOiQ8Hm#oJ8Cer_vKS2QORl%J!wPG6)7R+|W*o|pNCC1YaszArFPJ9%$z
zOm^62ab+?;0`syBRZw-;aJVShmvv8?%lQF6KtZqyj<zG+M+XDSL>O4NM<7;+uSU37
zww&+$@WThL@0d1M=Q69d2z7w}<{C4un7f||h?stPUFhhOl?ams$GKF`fHnFCHA2l;
zx*2eCA~04$a^Hv^{cGRz)7+w09J#6$m=~uJjGi=$G7y`$sk8c6L-v96WU9q)prXH;
zCooUxpOBXa^d%W4ZoQ^|j0sZ6RQ6v!PC@COi5cZDI2am+d4h$l*YK0(8qohNYq9^X
zI^`+L0f`jAx9Kg5q~@E=`e!05=~$3JRcla?3d>o1J!?vF52K}}Kyh(wI0M)G1SKSV
z`f@W*7_+5{%zUme1KvGJ(V%jcE|MOPCe#5m#@N-3N<_adi*k|?b?nJ+B$(`%o!VOL
z(mBG|*OW_^jPVctt%drMC$Pe?h*Z0gvun5521Vc%r1M?xA{0Ij8T;`p2St}_lDACi
z?x$eOIvhyM#@);Tk42QO@H}5e3qTkUQ0d+cj4l&{9V-Q2>n5MGzwcYE_x%3jKliP*
ze6h4zfeRKjvHhich~6@^@(ilWsZ+2pNO80+d-G<+`=qLlfX>bPjJu!fNl#tpCjSly
z2~?bRxbL&|ylbda$6I-tN<=u*1;I1H=H;c88tC$b!p=2a!cYZwbV~I@^vgf&Pd^Ui
z_ABi*u={oUCGiITBqh?4MYO6e9ka|6_l3(wWkv>i**geKFjfhjb^YUpt;rt%puQf4
zqKlb=Mz@G)HD&kEDA_iQas+#j=gt{%O>`IMUWPYBw7zKTjXLOkJ-8EFV=V^X+nG|Z
zt<rsRmd4wUw>7HQW7RNOX6ynS-yjL%ES-7Lk}nc9Dh|`)H|vmXd|0hXWosRfDSECb
z2p!-q@04@GB|?*H#<f38gG0>CGL>%^@t(+&WLnwMq_MW5Tc1OnL!P@srhT8gX}{Ey
zy}=1_(!6^=LME7%{9uOl8OzT{ILJ@{$5Yeo#h!<4pMiwNmgMaqu)-HHq_G*n#;f<#
zX}x4&${RnkDVhmcQ!1~OEQF?Q%6(%+tl}0Js!QUNUrFsb9&okiruh5#>A0N9aE0Uj
zfFN}8-|+|N^}!f8Bu3gtQW1cD<Z4WZ+qp%l{sRf$-8<pnzfmFb>WIhwE*7N#pC$Yf
z2bXxGS?4DhYpad&$B=>9?{!%FO}Y*hZ*3>!ThT|YHz@~8`T?(NyKNYGn_xJ6c3pBI
z1(Swk@w7lwm$pVM?H0~`OqUz6hI9Ukl@E_KHF<A4(Qf;O$tJIhgM5>|=cZC)UDUDW
z+C92eFfDIEfbCgRhX%(=odoxXcFOWf$)l6_L=yGF*!IhHxE9Nu17d0Bq^C6xDGzz~
zuhOh89<&OO$PiF#uLm275!Pn;+p-}mnhc9QFe*7V3`rNQK0J;W9gNdl-oZs(@p$9V
ze;j4}`NfcoGZz*3=YDWn{>gc`casY7rXa;Fw{!3D=<ib4o2d;#^7oGnMLcxhZ0!kJ
zJpR>X^6h<6TX0fudoI`F@%hmzk!nZbs+1^llDlE|6BKLl$hO&-Zg4v%eLvX0UPJ+D
zldlv;jqlBbB?;8CVER!8!1TuemL|RjA}nSJ-a_&Xlts&5`??g$GF-|#q-~ylmulkX
zS{$eEP~(6)3Rj(DC%lvrwQoLv$zld-XlXo;9YT);fs(u0dex~*i2F@-&Y?4kjtSLU
zmm3=0KkAs9@re=6IzQ>XAc2pzLHvSsQ8QIxml`NbswwS-KTz%?aAgVc-N2ZXBXfbO
zgimJf7a^-8TKqw{xt9b!HsRa*Em;uBp*QbBT#(C<>Spa0^LTi56kFvI5t|m3coQyW
z;K@+82K1X`2b7TH2&eY-jdi1}N$w9}H0joT5T)6rpl7P^iqI-)*^m(Q{IYOCGNzDU
zLH^S{R1a?&QR!^ZvQ75B2a|KPGiBHcgD>}7CGYGRr5Qpwv~?<NIuBmFaiE=GPzk@S
z+e{|vO;xltZ)QMQ?wP6cOY%*+rY+gmQZee18wN@Kx&CGuF-kfa?KIIkqytP3+IOEE
zRp|286lr>Zy-Vuei7fL&n5o>4m~RLlNfW6(GTKh~CdCvp(xo=ad?O)qwgS~IhPo@s
zEXq?gOz^zDrRaDMiNQdoijB)ok4x2$DMPt6n{0JEVt!Ir^~O=&)Vp)}B6VMqw85Ps
z{-=`+6x4P!rn$Z%#SP|5_vE}6?9Zd7C@d>jJh-;saZlU(lblE=@GG|PRHLhpp*$`)
zh^5_?NMo^dFixgHDo)fFagr$s2QixM5Gf8HvBLDLhPEkE+uJk|4|7S&@orvX8Eib(
z<2@m()tHg!vV|0$rqq2qBt6S$Lz3c<VgLM>?t?L<uZ(PmNL{bpM(?(L2&bNN7_w=f
zs*@;>@Rs&?3zjcV;9_vE5KJ+|R%^JNg^im0LrS~6_cldG;l%7}@=+eUd;v=9XqW1*
zvH!;-oF6&DpNt#0{(gk>#O@U{wDA2&09=8gSqkQg_?XbxN&J>A!U5b$w$!Wxfb7)_
z(&G@wv-)6!Zn`YT#fwKBDmy}CCmy@JyNsfY?Bz#La2fL)zR*!JhW4jmBXI7pPqH({
zu*yqk0PkQF!?K31$r7Z}*kadA;sHB<t-~YCU1y5to3U=nBST=Ss#>GLT|`u9ie_PO
z4-*-7LXmeXaGa-NHVgN>`8gy2F_k3@nN;FsNZ22#nDWR<bj|nN^tK|Wxl@G{pJBG=
zgc^UyCqHlnJQGEZ-Jlscm#Uzb=5~C<L3+F!g>yi4@NbB4TuhA>jWbakNI-@rHeDpS
zG43^X=O={8;~LkZ3C@l@G+{>@na5Hj*;yCT^^d95sbq`|NQ#}M4H?i{rV3tqq5NaL
z>qT2OT@&*-bA(vrk_1d1!zW)~Tk0lKst#-cp#WbxPS?BNHriu4bSC^FGG=D-d9Zx%
z6D`?0$u5P(#<h!?qYkwk!|YP=$z91h8bE>ah&~Cro4I^}I+npFIS5fYKy$1NyEr*$
zlk(n;4KF?J=Ag*@uI}U{O}j5H5a*6d2aS7Thw_gt18iS=dVl+x*1j=p874^On~j3l
zaWN6mIb{0S&Fd?ohtZn~NHhdO==nKbak<NW(qawFuEx~_?;{kP;~jZ3YoBG}kL!Jb
zTKF<bNOa(ix3WuzHwD@cZ5`I-iE1UiQT&Kk=<pAByA8G+r&K9yqB->aBzvo3WXZZ&
zd=q+{)1)Q((Bw<M1`K~5*#Gv2$uEeb*7t+=rhgG0DtJO3)7~j0S6*3s73d!^8Th1G
z6i0zxb`oIUWM`*8fumB@*wy?SFquo(?@8T>oep?7iaQfl_4Y!R5?Ttah1NgSvp9ZP
z)r-wf*TM6ZmM@|s1^cPl{5|6PM<`e6yYczKv)q(OV1Z!PBRZGpeEpB%_BY9luq-~m
zt+R!nWBR6g<T00frD2B2?o(shl=04?=0|+-qf{;p)=ppOtr-XMOdRtRY5i#L*sF4!
zv}Le4wZyJxklQGm^{$r#@GJfJ^wgFc6GWPF3O1|w3G2b{z`+)PKGh+%_~&{Vd*zeE
zn!*%&-`;8`w*De^^M#mckye+T8LeiF+s$-zf_Baa@NVOj@em%r`~HSfbWS2q<6}>y
z=8U!#iG>I4TnlH%iEyhga9S=Db9YY^L^OOHJ~Lc2zV2dv$1rVSKgVA2jL9gfi_zW#
zNzJKGIt^mlQ(zz^c$8#ht(Qz|aA%b<FnT?;L3MkL(;&5Y0inwGmc^KL*7Bsq&L+=c
z8Rkw;6SVv1<4_ak?M69@zTTommtHd8!1~X*H;Jxtvbrl<B`tgpKc(q;zB4Xa6_=6H
z$?E&Wfzw0VBq;EAe#L)5-*BgNc!0XX`?O(N_@_4XZ5d?&wV06Ueg56gpliGAU_Qp3
zC3m71LpEgvI8mg{m1+i-(m0LKWkmzS3C-YKF}g$zbpK`9w?HfGEZ=|kuAxrKoLpLY
z#$V}n|GCG1n~ErsY-=C({QACP`YCUw%edQ1P{9w9?>R=R8;-bUrE=;me=pM24@?TM
z?td$}aHM<4=&aGhYgEWI(F>$!=n!MCWrVT^J4zyn<b2me(vbK~X;=z_@%{(y&d&$4
z<cZi~`q7-*{?+%G+#+k+GzVG$-L?}5c<DNjB1LxxSTqBWedi3+A>FnfUZF_{TDjc+
z(KFMA%D7+=yrRkwO4rJut1=SWNqo2Q_y!r>A>*?rwFRAOz)3x)cEg#Rf=1!{DZ$}1
zp=p4CYwBqS%_nM|<^Qp?|AlZZumPi!CbCua_eP+JoIETU@hr^>+6F2m#6^r67hBYi
zv_M_`=q`{z5O$N>x%hzNZb0H0fD|&P9()*?h9Z(Z2@qWpVsL%akF;q$4vO!$2GpK?
z>qU<iE>!;ssQt|!Ob7_M(?ve<Csh=EZx20CDl6(CY1c+76O^*fZ||M9d1LzJIf(G>
zU?T2wnk=UstOlytr8<F>Q+RH7A}~|$?V|F0vtG1I^aS?jjrVV-(qDLlU`&m$Bj?Bp
z{%I!lyZHHkE&NkXkmvj%bv94$|7KnM`|nt~imVi$J?EDHp+o*JBVUs{$TM?3nP~D4
zt=xZp?)SnG-?LZp!3F;}+v^Xe&c8O_|K4!_`p&=q?0@=-VmiEh)DXG;FQ)9@e^&qg
z-H`VPQBG=5JNI8kham@fMZIP|_^87mResCZ4@Lr#6rh<(41`gzbg%w-<Kc!tf0N~?
z=||OhRN-F#`N90p$9)fRBL4a^H|o!is78()9S=N;qI2z_AMFS3loLehntw6!{^SEm
z^RvKc6e<{kVURQ6vj35R83-H|gu!7P&vkPv^&y7+&Jk%J;md~|1{@wd0k6dSw`xr$
z;t?~z71)R2K1fcp@!U{7!emC=Bs*ypA$&dcqkw8=z*rn6a^Oz*V-UDXL*do8R9Cm@
zrA;&YIqF;0=)qc`S>AXBQK52c4(9TkovBVW(u|2k78JhiQp-x^21Rqx)z1t*<fs4L
zXZMyB!AuVp=F9zB!b{8)K)|VvIU-fy7zpj}OB|0B2%iR_YcyW2De@da_?y7IdVAQ!
z{=Qk{H5eUy0Qr_Zh>@_(YE<e#jLH!LSfYAtI8j0tDt8y~dAP{|rDkm<j9B|2a@11h
zLf3EZAArGl8Hyec#ivq%vrsr$fu&m7Hb7%>hK0*((BjO1%VF`?bpYT}kdXVE{zy`h
zKa3C%@A^u$!^Vxmz{-4CePP%KyiR8j`Wlb|B`Ke4^4{&?Nwwzy>fXEx<wuZ`xxmDu
zxjq#pK?O+8>#eIADrokFPrzrgM;K+=U2V0leM@f-HcUcPXiRnF`qaG}2d6(jxHdZ@
z?^HB3ps!!!M?EIi*R$Dvp{k*rv~%HN_8)}A|8;Ks`s*og#7UvHB4?WY*KNs31;IM|
zydYvQoOv1g=CBG05Of}zK3h?FA}{$RE~&DW|Ir6n5tbfdg(^IwKBEA#mqFJ>>;!{O
zvkW9Mdl2T^7m=oi-8=4rsf!dsG`Fgi{}ImT@RO?<`ID<@{1u*|$gxLuRscpkg&F<}
zT=Pj`hOS!h?%GRnv2fW3$e2jGyYJzFY9Z|(t3f~P5OSVidE`tIPuL6KZ1xJn^IkfK
zMs0zooC}Z-YM@INS=q;H>T|<BQp2a`8W4}@W?gTw*iyY<V8`c&!(buZrJ`HCx)0&$
zN3WdO{Twiox9ZSaeedeDbWxO<YxwlOys1}qgSl)0SRt3;MGRN@ITHfT+I2{o)8E?L
zD;lCeO1*V_NX>Poz4}=chi>WL?cjT+$oVXkW{~)69ynP<i3|IweFR2^4J5TxBF2Hz
zP^BL#h&rb#sTd}gw7^wxXu;I*Y)=Tz3B9Y0GnaymuE~UZ?`70-QNkYVhl8wg6_?1b
z!bQ$bh5|FW#IoadZyXEvbY~(d6@viNxBwAEI01<*D@F1<n^dJoJ^~)b+V}Q{);sVS
zx`NPPYhuXOfaLlHVB@oc03fSysN*_XW1C;i*x?h!TbzD?EGwD{4?Re3SJ4xlgrUB~
zl}x|s6SdG^sJ^#=Zl+eRyUEwMrqxWMQ=n*%7a#Iu4hnsiX|5CK*dD!C;}`aK!_p-q
z<mz#L_h{tT&M!Rk88LkXPJ0tBQZ3<`^W~J!faZMy=f^6Xj~6hG2Z640w>}5!;tIHh
zy0t1qAfCi}$L8GAb5hhJr+w%M=EuA5_2!&Qstp7ppvoyBv#mjOZw4~s6+-n0Z0>Tf
z{H?C28TC}AIq~4VlGs^+yl`OldY(WIBl7~41Bde9l^-}{`Sjm!%K!P59z@a*bI&v7
z_xI<?O#y=#2iXYP=7nN}xSXa-7Qus~RBKe=$bdOhAD}vNkiT(Jn{Ek0z|S8A4>#Bq
zc1>jvt-2I}z-|p6Y^jZtxH@@~j%we0mVH~U)yRDr4vTgRy$?KDN4xhsY4=(eYps@>
zeS_^r{{T7IjD%BaXvM;FDTBbPu^&;<VFgRkJ-K?QDY2xKK*-0PfdHh4as7z=q#+&2
z8c*4gB|jyYgJKZ#l^E^~Vg;m>GzLyx-&ltYfjOGHDVSp1Dy2xc{$4<{^pLZIWF2z9
z0QEb%tCAi&L+^lZ6z}O_X;>2RiO)-Q<j=m2&V0n%Yl263hteM|-Y`0wx|5dY7#P~H
z3YFytM2~M_vev#g7<bb_@pJIGO;~tJ8|q(f@gGHx<rH(0#FcQTsP=3uLo|5zeY(}M
z(P(=QIE9N}TrR36@$=6jywzFag(c;IRp_{*sIq&kLaBLhzL%4XZA(Fqe6PrFjD5Uk
zfH#tVT7LOG=bTvUu@@iXwc+GQoA0w|^=Qc24}ghtv^Iy>OOL!1+elnl-7XZ6X^c>*
z;$mH*!(%C_DNbeMYd|`;8e#=s`1aZvi612at4KcaoHXg!5SBF#CM`_<0NTb(*0p=|
zM2U;8zPclBP2ilviRbuAF9@HVbE!b(2FCAnn6;RRgA>FXW&O|c&zs{oM?cBHr8tP>
zHZ0o74SyfHXckXg9e{I(c0B-SF=7_>=QsFPiIq?CU?i+QH1Q<qVaC@Z+t-sOK3eMd
zVvAiWEE7rDe_x=4tRT9p%($Z8i^~vU%)P_pwCv;$#}J|<c&k2nd~i$De>CRg04xdP
z*!kErxO=i<{Q*RgH3glL%0<4~^{2a<-8t2xOW?(cT30SvERe!CfUDma8--I()=u~_
znT<a<f8a~>@ke`TQ0etKB}gRwAY?n4!^|o>b;e+-;rv<I*NE!Dh_nJ8-;a<E`5R~$
zst%p_?R1D7&=CD9K`3R&C>u*vTs0DjE4rl-@%GJeg-g#DG7t)Bw|j>IrK~eRDm1?F
zfkv!XpOYi}?lRMa*xt`8a2nsr$0c|}$#K~He9ie{F12mwKc3SXHGhQ*U@tWe7mp}J
zIvVUVcaY(JKM<Q(N8Uh?vJ<E?s@ccc&$FR9LIoSvMi9FD002<to<XMa%RM)!yqCDq
zcO(tIGBgWRFbQHADA<;jfupLLs)|1STekMYJ#;caRFgLabpX~?Xl59wgr$uFq5A?O
z++OesG6Va0^mw+bcS1Vax&xFPk^>B~RMfFGNy0HPo*uB_FqT4xl*&WGUJ^&64~!|(
z2OxzDt=A5i2CA9UzH!uy(UR@20}k9&U<^29h_$*8JmkYxC)?rB)u6S=dHr}Jr7vK2
zIG}TEB$p;0>KA);UBE!J0ua~}NrVTPoWia!Zqscl*zNQgE#aZPA(JCUGAHq(Epqfp
zCrB3XXL1tKvB-uEE;T49`v#TNttE?DUCtQoZ)YgU#?wc50%Lf8sZ!TO5~+8xR;J@Y
zw;w`doyYIbp97<$%Sf}=+__~%G4}b{_N?m)pihIj1dAhqch`4aO7^9WNUe@t{p|-;
z@Cp|m@zWw5O^z3tQy^)OEUW_>2(%rNj8yOB^f#QXiyW2$EvD9EP);kG#ugqy+2cBM
zP8F89t6%><*6^MeTpLbyxdOje!}r=~^VP`eV2pwIAU%OmClMg|J6UC%CrTv0Lm_vZ
z$kGSOI%R7bG<((>n;xCoEQ}y=_E*$c)((zPU2cZY>QLxM7@o4cWLN&cu$!iI4rZ+T
z?uRYdGSGF$@~q8g>Wmd2FkC-&?;tXb(ICIMYt~u+PCfr8FDSfk)*)Jt-2$yK!NHWO
z;OR#uLwDdSk`=BEKCFUA$J83r5#A|itVKDL*#+31v757Nl+8>-g`sySFeX0&v1rBx
zKn1JQd=X)qXdKL?QF+d>_cL_3M~VHQXnZ7&ohvkq&)UHqX+Fx8#V`ciZuTo|Jr#-h
z0(#xx^>&z^ZGUX!q-nW-qw>HJ&2gAX-M05=4ne<s#NBI(Cn_KdtDY)Y;{-Yw9jelr
zgNkOYdQM%s2o+ob^QSnsQlLug#-h16F6&uxgw;b_U1`8cx>w9MaA8Y?ItJZtdm4^-
zU8ZEt)DD@^Xf?b_Q9pQ&^1}dY9%MXtH?8A5SuAu_&3ATGhIvew);Si~c2^RP#JOuN
zK7OhrX@akN!}s!tihe5MFicwkH^$o1x!x`Xk^sqs<T*Ac_1=h+OOV4v94|J<d6iw@
zJSxja{9vrt*SL~L7uY<qQmT16ZdE&gu<Ul}>$<iz^c8N7VJHw-JyxNjWb$BMp2RI7
zxdK<J$CG@d<D}k|zVK0XM+axcu#B=Mrus5HuyL$F^((c=hx)ohM-V-?XKQp<gW=o5
zRQ`2sQ$bWI0lZ&Q>X?!<9VAxb;v&|fd}@Jqa8t6YHW`=rohwUhF6qmw@IgJSP~zcd
z({}Y9+7fe%1vm2-(Wv3Bh2+hzf?luNj?Nk~O1wM$x4nNzCH2~mix>VB)!2(sz~hi{
zYT|*O<*IRudAp{<OZ^e$m%OMDUbpTnu!Y(XbF1cb3pwHD%kK8&(c}?=1wEcBX$?Og
zfxRAqxh5djbsH0aH8astIHEJsmX>tayG6up>V(_<q8YQ5=WxoWRM{sh_ViTlyY`}q
zv$i(@FA{XDZ?o)81V%#3jEt|Qi=e1&HltzY%qN%et?FWIRdj+6<%`DbbGhOBGKHnN
zglD;hgSeslF&)tz-#bA(V0m4}jUANTI)rh00w(f9&7ruei;U}x=4Li;)FK-Rfrajs
z+pmPLQXa+Yz2ABL%>2v3xsRhS&n>`}e)!WwTq!+XG}&h?FW<DgMcOO15K|u*n&so!
zFge_ib5+H%lgFn@l;Cryu5owQ%)<}1?P8SyQ{7{5;OHon8FWiUpJZ~rqrCNnCSu>M
zh<!KJCgd-_;S5x-^|8)REnnr+@3vSn$D0wVn$LR{x5<m`=_U!saVume^=Lff?|7P_
z6o*l)X=QftYFN<7iioPF&J<QVeCb*XKPiyuaTa7$wq-4{R=?B8ygDV>l;8G^I#4Xl
z46S2krmet#aYX#eihj|ca67N?S)O*6+=%&+m(C;IXR$itHJNW}t5W=`l9{$&t98Fj
z*YWpR8hTmFwvBpe<QXS3Ki<{G?1O$@HDt5asMepeRR3y~mZ_C~gLjT;!^!{Q(DlSO
zn|1Z#PrQ}zb9Vi9lH#j&bFQ|vDc5`3J+2l1Z9R>qK#S@A&ou@7fBYa`VF5-UvTTdK
zlC`l5w#k=nHFCa7jG)tJ^ljl}<Zj5Z>rGu$-9EjCHDQguiqS@B>^FQnL)SvHC*x~r
zJQdUU!wp*HZUf5aFy1zJyvt^lTTC_eL}56_LQ7%c63z@IRI=oG={R{if%c$9t~d)5
zyOvK*4yEs%M&6NJ496KL<5Fdsg21T$JuR1lmf;s8)1%88dWj1U4KvtitVP2#8e6nl
zvR~h<P#u0OC1r|N-~Ak#JE_xE1qlJ`pY5X%IDV!AQxWMls(ZAp(}u>!x+Sb;J-)gB
z6>Hfi9OX#&J9_-}WS@j5;t^t|8q5bE)Uq*SX5MiyX6X^JZhDHrw<r5ibhkd4vvq??
z<l-b=%#P4%aKe((27iGmu=dfu;JP=lZ{wWf-eVbFR$!8}V@uwsjogjMA?YgZn~R6y
zN$RxlTysKSt&EUW{jvri_tnk|_Ql<JtsnP8F^UH+$@mKmQUzB)a+;8$!3T`s()bP4
zhAUq;v3{zG9@-Xji%b16*Gco+f}1nyq@9^<UzfhY;sw8=1-gI7l>YomNQTNOEZI|5
zHeI_vt~xBe27%1B|CJ|;!H<Q@g3sFMvjwx?=v4|wFt_>J-uV%H1=Z0?g1f|{x?SOj
z>56B!bX!f$2P@%cjq$PlcLW)}c8@zZ;^<22gs?YOfd7`%c}<o#No9#oIWEY7rEDD;
z2$OVMqHRJJHI{`}TgyTrK>3s)ld($*7N5k}b}tV*nm2BJq6>;;<s@&jpzY;%uNak$
z>HD*XQ!p1Qkjy@{z7oNreW9gizJlq!Eokf`q!;q_NOXr|Yz}J*vIhVwn5E4mZ`L9~
zx8-boTTar)E6c)}&D(&3xb?isW=^za+FUwM(CXl9?MJwVk$nj$=StM_3Ux+T{C#27
z6f95f>xe*toD2M>!VeipOrhYu)`i+xp@^?#7{QeGcmhhdWTbpSrkR-16H(f<R?KY<
zDHNscPutQN3FKX*dB_hDN_uG<p_6pUML>#JthM;)fzFsNzDQg>Eu}cwN|8UeT9m1+
zk|3_Vs_6c}w@1!{Tr_FlS-kg{9cSQW0E@&3;vzH<1f29WBVeH=`$xC3&yVllE>ZX2
z)@t8%^%}cO(nA|+<DrM`2NJ$O^(oOhrRy$VKvmlAQ$9ChJ3S9wc$5Ekm<EX99Zm9i
zhedpji}HTI%cU*+{dMil@h9aGbF^k{3oQunv!V~5qi@%@HpNZlbGT<&cev%3mmbv(
zO>nE_JDn_sb0S>&|LL57zWef{M-VxNpgux6Y>Z@a69Fm1`7}iGtQ(l4KpeJeczi{z
zx{J^xg+FMo;i9o?Hdl?wT(SuGdxLGl<KaH|_f{zKw+a*%Ulc+KBGisxI3V6AV_8;w
zwa7F35mhkCT!bs0xbTzV$xKvvNQToG#Wvp3AEjdy_+%nO>ULE*G><#SzQA~PO`RWw
z$Qej<OEYjcApJoX#6a#|%niOFvOmyx;B$MaiJoG-e(u(E94~Q`PO0fOt5d^c7r*?O
zJ3HiEMZ#pnp$2_d<8|5K^oSfi%}I2LKfndq+x$c6vbdShim$qL0|^%5PbDk*ag;B&
zp0BC*IOtfT^~`~ay5F~*)3DOGNqxh>;nwoftip7+L6bc{nY6g2IcV?1o>lOM5Qo0c
zPE)W2hF<rE`Z!rc%n8*iI^?@#Kw@tEh73>N8(GCf&b=fm8O$|cHYM<6j?qFHc(OS<
zi4FoQ@G+QFL}qU$564lkjb9;|e{@z{I7GINf_ks3V(`!w=Zym=KfVP@!;Exsp<PkS
z*T5m$kOA40iPiEY-I(_6srrrp>A~Vu-^y>lgn@q<A^eX61A!%UoUir%a$2}IPMt$>
zPo&TenX8*GuS0fi2O;FAs*kID<%&Abm|xf(tq9x|D$9oU%NhqEdfdiB)yp56QWvBR
zhzrAMB<cz53x;`zw}x+nnX3jNaoW(&8G!7k2*s3@hj=XR2aqz)%Y#-+_1GqW2(yT8
z8sQGiPqme#ifA5v$^OPk^XGy82$P{PM<BCjZ{UcgV=Vqj-3+hL|GK<v(TAffLv(?0
zbFoptbp1r)1l_u=DL^TzTRHfriASakbJu&{JzdPl7_wwTTu<|3w7oz~&xKZL#u^it
z!^}`sP(9U{Y>ukCfy>%}2n99DQzUHGIO%RPA^$PrdfDC&ebYsa;O20RpgO0Y+|tYP
zCS*BRVj7~-3NZtgTAK!(bY#ABV&0Y9$2gNZfQRb^;jk3~aeWefW6Mged70ze+tFok
zC7(PWzBFXJX^@Kl2V$Wk*JtHJ|1y57U=Bf88}o*NBNP`uaj$B4Mzq;Rw8=$aWSkbS
z6E@-jWaW?rhGc8eBDV{J<x0tn3@}P-#{jp|R7f=+!HV-BMrE(R%2a`$6c4JfWP06C
z2O+9LQ8(4&DKH+EyYIF^vA41KeEwORiNOB%jW+m3wXiaeWt0(g7V)`BCj7}Y&WTkD
zNj?eHjk)oS1dp={t(Ql-<LH^n96JZHxwui=k4ay#A_^d~42{9O=T|himMQH`{&?E=
zY}Lm&AIetg;+BlxkL20z!*}4-$(lxIxiAI5kCO??b>0@4Pxqm051W5yKm@pfWqf(k
zRf=f}Fc6&5gzol0Jpi98nhF#qBkpHw`=sIgl_-Ss*8pE6ik!b@W6|p$a%UPjdlETf
z0*7dXchoGuc$nq-ZBmYQWfOUXZ7&FBVMAJw#&h&-_S*|}c_}yAJHYuu$80C6y4u77
z)gm=NKJe&n=eUo=+=#;T7d^)%CML3|=~1lf8#l}MPbaIece;hu_5{B9QF^-0PBF`B
z>S5Nlr_yTB#dDr@#IuUq{9UWQxn(`>g2?%5C-w7uwU)<U?``(DhX30-%$fqd_Ye68
zIUUsYuwM107!w9I?%zFrH8A3RBSBW1fBUVH^M{hg?4&1}-@u=aiu^f2^ReG2X!Y&@
z<EkF{Tyxx?SJ9HuvC_RpIHfs?nvqV!V4PMkav=V`$nlh8K>THIUKhvdUbEysIX@Y}
zKE5>lf4F<guqfBIZFoSCGG;&lMH&eul@O#vkQPA^q(Mq)>24TWkdo8^R6-g=KoCSg
zBpjrX5R_0-y1(OEb+7k+?t4Ak`~Lj4ZT*;yHRD`!#d#jbe(cekpXYQZ>?X2W?)*en
z%d;qUvc_P(7J3s)&tGh~IQe#!W#<74R!Xe=x{{aTvMCH56hN)ki?{(g3LNeO6W6?c
zQ#HZvqSB2reYLU9d<K2vNMTmg1rmX=>)?Y+=C}fL>6<|(b30i@o4q1Wj3f}h1$zdr
z<?pjplKE#%^AH=;#~r~NwU6ZQQ;4oZo_0Y>1>Yk5VrVoqD%G84T0KUbSbCdY<LsCy
z;+Mfv>BRxY<i8o0mWn%Cc;;rjW{30d`ThP+FX*jGw?zK0&OhvL4}Rt^{IS2u`!T)<
zE0)Vn7rZ|L;o=F|zP&i+gxm3)Mpz7NTS~v#fWFYj-ai1|v&{Z)Mxb`}dzFF_*uKw2
zhV|%cN@kaRs^o7in^`>qQ9Bz_QV;IU`U5D#&Grrf%eWG9yat=l1f4J{5|)b*<dgh~
zLFFd#JwL;ZB&d{5KVs}>YD5#DJ99oHJwDQ43hk!~`kZnO5ssUM3G>?Sw-k5aWp;C$
zz`S`e=E3HL`VSC0b_;7Qr`v!IGXwmn=p4F{cb;UHLwyE`$u6BtV3F!JpY@)d22UTf
z8OsF?kH{{iJ%4dgK-qrE$;ww^>`S@44`Wcr#b%xmeSML5D`qZXD(U-`c**9EU=~BM
zX2h+j6v^Cnz2`m5{Fp11-Ms<hB*XW#QgIcOcVILvtz`27(K|a-zv>?O7}{(RoYfTO
zO6kk$8<}8~&fBKkb7Q^CHwUMhLNhPV)&^i9+Z87ka<fX+uj3!h*?qS^XU6?z09u-}
z@htORszs)jT0KjqweGX69%MNu`93ozkkaDyUauVdAl~o#f;e3+l{uc346l`>Z4%a$
zBzT<cHE1iuw~25plQEVdq?S3}$;-|#zYOJ6`qo@awP?&@0P%-YA3z|eOG^{!_>H6a
zJ<Oocjv_0ly~@6=!qHMn(IUL7L)n^9;Z>jKJe$0+>E}O_`pJ@acG5vI!b9FDY9$G<
zd_co?shPg{>qA4VtKecEB)SRR0msuvFvZJ$H<bbv*IK3V`nGsoX3nC(wmz$)Z>4yY
zwa9SMY4>Lf|6I#oHxfr}9u)s8Fx8TPkk@Vimd<-H7R+$JC9%?~CyaQBDJ@V&zjo3Z
zu(O(sQ|2u}0+yyU!fc8g4o?L<cFz`dobB{dyaJ8*FcJb+0tF&oT*mOuZHAl{pt9Ws
zLGbe}joc}zFRAjy7im!K#H5b7O?vhhlK9$*3odCThB4Z^zeSW3ZgmGUc69|f_w36n
zj1N2BlY2r~E3|Qu4+=_}PbZXeicOV1dF^J*>z5osBR5@onPIODe02p^W~^LbctLW=
zu(}VSkw#2t&<OulxdCwBrFVos25ZT%-1WjUW%C@dyrFHk-h*O@Mo3onno?#Y(RJHC
zzCAUOfcTiW>@K&YQe2vy4_0ft47A5&Wla~DU#BWlIz)@>_Z&I)ZC?2#&4Qwc(RES%
zc%|d-r(tMn3N4A>OGtQ=YtetgZ%AD>GXOAq$Jr;iyS(R{OTf}#40v0OqBNV<Ajw9k
zwUzdfo*&+s9YU-s+OsqU0ra)uy+&QAC|WIO0UjtO#S>yQ?d1KphQ>u1UULs<iWz#e
zpFQLFm}?t7eJ$$r$uH}`nQtsRe4OFT9>C&B+U=}EkL492*1sT;n=ovP-fyTV7ugFD
zLk0`6El>y9F>LW`Ki_g+<y)DF!q<O&n<-Ct(Xe??JSAi6*V}clBS`+3(14YUcddq<
zJW`lVo#970RKCN2*K4nG8U|c$Gjv*dMNdN?wf?+d)pMI={;P89wLv@ZvPr(^mHV02
zwTCY?WnRmD)u!#UwM1*y#THA0m6L1su>zxv?aHf)F~3glrC9LC)TCLpW1x~v+WNKl
zCC&;6)i-F9#_#ezt8E{y^YU0#o?;)p2Q%mpya*I0yR^>--8qZe+_Ht~1WaUW>-AoU
z&?F$^=!lYzotO=-9p26qzE~T?@UlH(zQc7bv1riM;o_RT!>G89VeQo|$;QiGU|ke-
z4pOazDnSA^3?~I+lTcUxm?lPaigv~F)s*_xti38@@JYBB^Nel<`_V^0?N@rh;-OX5
zX2vlQH=6jm`YNdypM?IMv|oacQeYU;0xYzx*RKEvnJaUy|Hu7T7MUylt-n!zn-nY*
z({$2ABS&d74(m|Y{kr=u){G&iRa_~?y{R6(vH5g_bvx+Ac&y8jt$VN^@x2jR#|%F7
zL-&cxsO>6rk^TB!B`;6zk7h23lTJ1dFK`BpWXO{>phksq4%E0VqBX>SrMvgkpozr0
zHAtyvy?Z6Cps6Z;_*7iuI`et9SCF{)8NpsIg(w6(X|A|81kUbPHx)LvRPL~dLm_4q
z<|Z?;Da$FVCmlV0yb}btx<^Wmh&E=SsCk!PgoGqd#rEOagQ@Cm-ljfa$?1)|t#j|}
z)sT<uL29zCIuQEq9Pk;s3BA?;;nf#<!QR5i<bA~qN6|84g*|%j7v~N9x&nhujt8;y
z&^uPf5bRAJLn2DIB=&ET-*DD4x4%G*KZ*7N`KyR0fYGH%$?|RIDX)r#eG6rp#PD~p
z3dq&4wyH}KsmBbR^rj;anRk?XEd2VNcGcGqx}gqYQOs<<j!Yavy_YeOYh1QW#<$Fl
zty&Fz4~ld6(+l9SgTI$%%5V4Ve7Q<T4{JwOS+ixv%1P&^b}W{kV`AfDWiYL?6p>cV
z-wYw9^B4w|I!_g!=}xV9dh6)ZWgt{U@3#z09CzqR3Kz;QgRY6cNq6Dn#fD=aPJc*r
zC0=<f&vIlr&0=L?<!tyQ{YOWu^u-@McYo`+zpArcMX7ughfzBb)jH;osG5w7#VFfy
z4DBC`SOq`p=irHfq(~aIxKfQ;wI{7AoBl*Re@r~5@I{;?d!dWhI_kPU<m2NLpW+uS
z+&lK75;J#FX0<xRwKo2VnJBG>1d%+ito`wOknN{9qK7e9LagZxE*KFwF}G`qY0cA4
z1o7c%)QMx`u_m%amkr5Fai1Pdy{jN~Y&DXdWo@L!M?VZ~4a5e>#^$lt<ZLY|zS(TZ
ztY)TVMrqLOKP18A+*buXiTDb{QSwouaw!_`3WuM<Qv+3WjV118PS6YK4D7XH>D{XZ
zR~A@Gjn7^=mwH@%2s<I)BiR<mo}nh%N>Q0W8T-x6oF_~n=wAH0RgdixyI;Eva8b(d
zZwwvH<xExj7-6kws7w{WfvFu*UL*JWChu#ljAa`#$svv94>A)@=nE8hXDH0Pd+WX-
zn<$GP-e;9!n;MD|u0zg0U&s<E<D4p!IInsVvaA_JQYKJI_9QqcI?4V}=Mfq%YJD|W
zm(HV`wWhE#Qy?&eS<%$tFwL6tnpD<l5uf~6CLdq9T3su-GD8&~?9*^b=CJ94NiFMy
zK;dH!Oj}+@l6!CP)@giHDgP*~d?X1AA&a{CbhyrRjmO7CKxOeB=v*57dDUXI>A<<y
zb7xZ0?}izrk&oXY;O~g(>fV1Wku=hoZ<ymQyd$Z$a&M{(xx}4&9r5*eKwmHZFX-zn
zWYxL<F@N;;s*~lDDs<mWvHjT7L4FFi@-@D`Oub#Ndn@8Q!<`4B379Zm`D{)WeJ+H4
zz)g_2%zmFCb^sY%MSCYdpz$9gy+ws0R`}mYbx#m)#$k)>^M7$rbg-yj<R3dJ{lgTT
zWh=#?mo&-krOvw+!+TDBYX4VKeba*;Kc@TRz^WLSx=<l;I{6$Qp(vUu?hBAV=MW2j
zOB}*rkbTg>Y-PqwIFImGcv<xS6MOEDU;5ks(u4tYLHIR}e;r6HO4(@?78E%-L)Zvx
zka51)v-Cn`zwbf70~!iIDJaHX1?$bhLS4%Q6(TtBkHrSgy_&(CsoYWRanB*X;|a#!
z5#C!0DGQas_p@=8ECkylV7X__ee$x`rak!o%vvK%)`Y(^S<l~AHTxGG?=N)Z?=Sov
zl%xsAFTH<>us{Az?7&J-QzA}v*1-EYz8oG`{AC?fa`%x?VduU(Eq(f~`l<GfWS3t(
z^ze8EVl)LQu(T4HhtYBgY;;{7Yd=n|WxzY<f$;kZS_LQ1V|h>h&u{p*FZ3f7R0h<h
zjOc%EYUOBxrhJYJw?^=K9G1y_p7ucLYPAgEElPg}IiuAX{5muWNfNfecrAnIn*{p)
zzyHMl_}c#g+%6v<&tU(TL;62X$=?px$sIml<3*M0_xDTxfARVT!gdv<qW`B_`9B`{
z?|<~?RfEqb*|D6w@xMJ8e<x7>?R6X6rrZ`d*8huR{`;N(={tOW?w4vd;XkeRe>-FU
zekP?wK@Rg>#Yy)6@%XF7;q&@+VYj6J7%<L=5<w|^eh0?A)_J+lO5sUZPYZXQOo;{|
z%>Dj)-ve7cWz9eBFaLTv|Jw_(P$0y1el2|d*YCT<WDj8A0PHDth^OaK$U$j5N$)rV
zLbFPshnI7%Kxo|n$QL7`*S@6-+-)8JzM{5tF@(Y(gi(r}Ug40Vbmm|QnU9zi_448!
zFfSCw!Lt%kFxA@If~u?>T%n#77e)1*ee1OYXn}Sdn_Pm?0#u{R5z#`WpJ%)@KbHNL
z^9_9mMN%QC!G%GANwZ#o7b0^q&74UQx}E#qe$wxMheSEhA>~$RYW;cQCWxW|O1lg(
zpY!mQ<$=e<EYlL!0XnQebe$O+!0jA@<;wzz6aK2LHYVLE4jDjmK-DS%LJ@aUfsQFR
zTf?80pBd4`H-D(<GI#Tof09Z6wY&NhIXOtQpmUhUZ^2*Vi1Q_qE?ihOL9S~OcD9@Q
zGC0n8I4NDJ*}Mu2Va;KH#pZy<5*`P<+k>c?B86rI{{Ng~=Tw*Xy@7NRh@<65@cTNr
zvkkyp%FMA{K%VE-kCf|dpJKgn<vRLiqdYuG>z8Bvwq>VF`XBwp82<N@^W-^VSb0-N
zXYCIaq(3#2Jzxx>OOCC~p#z|45$RNl{Im?*o|74`fLuWr`!gCQ_KhVObys@wLBzlz
z`yqn2x)r+91MPaBy`lFPK$kQO^J^nR-#zD={vQFmWS?%UNp59>J>8ux3xm7St@?2C
zUeg0*u@c<h``wa)Ej@seIRG9pw(mlZi!{$d<7Bw9=WaM)De;R}-nDfRpzCs2=5v`)
z5C<P945_$R*V9bn6jKUKvQ{*Cj_<=_edO^RjH7-lu^*m_)lH}t_K%*U$`M*5U$MRS
zxP0t2Q4fb$QhmnnijcoQ&VRm^@Q1rT<5OE+)*lOFIjR5i439P>U0W9%AtJv^UOul1
zev`sZGdCd4@boy+nL`rxGjjs80|fzkOZnaxssX^}uHI_n9TP89UWSCYL12K~0}rM2
zmvHktOYQ%`ZhZEc_^tU*b=UWvpN?Xkkb$}_p;36@=+?P6MTu(Pdf;ry6B_XpSAP<s
zALQu1?-J}T!V@7`bC_T)tPp?fT=0KgyqzqN<=OFPp6y@D+fi^7)TSS<ddQI{rEl@$
zY%wH;48la!D4h$n4_v%T0B7>Q|8_uj!)w_;&!X0|`xd@bM@P77YbCp0PAbhXP?Gk>
z#bek1>%AmR^b15c{C2puKSR!&nf(#v2+}$$oB6;F39=2w)CX-`Hc0Ba4NDQsXrFXY
zVtRi7{!zkLx7mo?@CY&92HM*)!~oL3W^|6@HHp9!vQWTN!^v_<Jp?q%Lw|T(T4+2|
z0)nH#g~Y)2XM^Uc9h>=Q5y=R$DGtMvMyEdO)(1jb>)#ACf%jJmf|Yjmbf<2e(sOzj
z3yGu?>GuH-n0QUWQ~t0aV4W6PV8r?+`h`XLR{#{|0k|#!Zuc0qo_)i3n@|Gfyx-g!
z+-P*(CLlVmM1nyOzHKTP-?A(!ZRE7LF(~W)5>K2Do`g>9%)9`ot#v$q5$(Q~N`=C}
z44IHgS}A@L0Cm93O#r~}tUeum<0<c(ZAhtF$VxdG_FepQd5?xNaS=XCUViME@1q$;
z#9sq(OU1L#t|dqUk<Jr|{J`;+w-8aOpeL>VWI>LsKjHh#H@*xkU%9ailW;DREwgRm
zyGalD*FQHo=PtQI!rc4KUEjHMl3IPawMvqQ$4-Z@QY)dDna-B`5j|8O$#Ero%E}`s
z;3O2p&ZM+>w)=|Cj4IY?RwJ7t0LwX^DLLlRH+^P(&^6OhZE$*OYASt<FHPT~OT5gy
zqi=7t0)alAs(;YbBttk+esn^`RXR%wuA7PkUf}T(gyhZY=YebT?3?2i)?#!759>?}
zd&0aM32WFYmQLJM!RIW@`QM(;EATNL?p|=I@kyE%`MFc(@_tX5_jUG)Md~j=#ffoM
z&9nyx1Uq(kH?8aD!jNoO`LpPPs-?$8Is&xs9&Cc!Izcvw0jFfwa{A`_f*0dl^7N+u
zB}d<1Z|pOmUQDqyWEEjLc71m6&dxonO2l+xp{Bb8*iK#3x<g$Dl@mq3oFf;7xW_WC
zZIrEKan}#eov|r@xmJkG#SZ<XHUo;?imQTkN@elkUv(j4X7t~@O@aQ!Q?LLgD8~_O
zc$_{ez?V_nt+<Q0Gsih<9#K#i6gQIZ*J<3kQ}&ay|MPJr561T*%XM;@+Qaum91~rZ
z>Y60?!+%;On73q63F8-bU-XLnle=~D5)>krzQ#QD_<arIkEN8}6aiyUdytO>ArIes
zK5WG1K%4v^;<(6EAPULl6l4y78|%}+_*54|!K@)IdsPDXa;48jWKD17FMR<hjO7a~
z{k$yF<4rAkO**6mAxBP4Fvej-cBe$@xe7vhif6C#(7?c(!x)NzW3XuR)k^v<A#^z)
zTiUq66RGDHuo>jSeB|q)HJE$Txs1A7&<Ia0zN{F#N%$7w9wON<qqn;mKz%Z@&wSe9
zIAKZV-kb%XXa=N+&$pvLAFhNo`ck8u-`4uYW7#)CI2ead-SxpdGNO{&#>X@Q1<3F$
z^~XMc(i2Z<RkoOpDa&#WU}UJfSy&(DClS(?d;#g&k4d$WfkX3EaA8MyOS{ew0SPze
z9Ridij>!kwYIMbCC1*Y#m;7><G5=+y4u0N--gDfa#H;QtBb^wSJv1^NY|6i~8(4{1
z=a)y5J=}K$21`DdpES2U7@mhh#?i;c90%+rU}Ha^Vcvz&!MeZ|x0|<SR)CB*HHSXO
z&-Ylj+<&WMSLTc>Fcn_Ez1Z7OVhAmm550+~EUBLEEkN<LUWMTt=NX0pemO+QG9~Ck
zzRYg&#_NikzwZ+@xwB*_MdrwXO)%D26%DjeFILkANlmnnO#yw)xqL=BZ>#UxH9oI2
z$6~aPj0H-y)Mo%!Q!_w7b^x*XiD>*>Yf*5T>m@1b<C0cX`O7{bMbsupM_wD&0_Lcv
z(B%~<;6o@Naj)o3==}~yS;cloZ9K=6$bojuV=4$8zdrz?ly^Wv*URVPVFdo8?*c~V
zlv<N-J)|n-{}6+ruebgLlOxvI9c(FbtJ&wve-w4^?XHIiaSvyhSAJZrdD)3KAJTQ=
zv6_SXYT74p`P-YIibUr~SME8Y%hLAI-p5W0w)QtbQeaL*@*bG69t8k3bcbS>;O8Yr
zB6TT8jt-j65_VgF3ik_}PbcxbEXdkq+4iqW*bv>P4ca2s8hQf=R0ycnFQ>u6v%epq
zcDK`b7+B0puTyQV=1;0WQ~i|TGA&=JHuMi&;fyCJH|IR><}v)<5vmXoZE6FLP&Wh_
zdRT*IgaH$Ub@xg;8-T`TBF3}a;I6=nZe#s=b((R1DbyY`Gc|7JFxNQ(bI1)&;0&{}
zkTWV{=FWh95W_5&Aqg%633_BQVJ$rqtB($2)*e5^XCu8S>v<`Lzu*Ovm-M$p@`)mt
z!d*NA{s^--M#cbj9hPFJdW;p!!iH^rioAifU@_0S-gG4@F@q9i0PGuHlsJ@K>2yc>
z?^w_u7)5VJ3b=#!3?XBGn99{v`X&RYxZIzN4F>R>0ZS3yx+`;XO>!&;n6)FAK6w77
zF_fnqwU;pBx|=Uzo`f#5=`h+bYaHValWYmKsA3wIDOiDx`OS^ZW;a%CSy_2zLNw!c
zNs<m>k@aPCmo>Q1Fc@PMd?U{j@@;n}U&xS6l09<GvyB@|hbgCgyo`J{LVrdCQ#F1i
zo}~?rWmdeT2nxUk;+R2FFTF#Vh|uAEZhlc+wLDIxl8lweK~_`dM&Jzs0@=E^N2w#f
z!)SDd8_htuOj%=e>@#(tE)22Hy`m~Prp|40%*Qwso%O;<3o9<Bj$Zppffh3EB4rO7
zPvR>chU;c^?S~4XbkMDV3-fYE<i5$^m>c{#BV?b;T58@JD#<V1R_`e^kE!_SI=sBM
z=@0u%lg%Qym8^*R*(niwPF<2b^p#tZW6R%3_2rm3v7gyizRvu;Z*G(*gwbvrxOu+G
zLzX0mzw3@Xh80e_0Rb}?;kb1JFJ=TOc^6zCPXleVi}VFnC#2|Y=o#=St!p-0v~#61
z>_{`=nWhg|H>dF7my<i2N56OodpJTVfB#%>uWny)PlmbgM<6t{<IRWD>NA8Uzc_ZS
z35B>G86}9zq!)8~v7^7w>Bs3;;JZt3XkE->)skk~;^(`KOo!%E3x~QrQFUojdS7GI
z+@!wi2Z}aSM<0yGFkyL<mGXabI+7R$)lcljRV2Ai!CZ?C25$8Gp?tDuWc9E!F)QFQ
zlW?$1<Vz}T`sLQpshCphfXedAQX0Oi=YNxk{aG^qRbe78K82DE96Zeh!0nIS$RERh
z<%*!cwBQxhas?ZG9Xm$)tgsLqnOSw(q@NsjIQ#u@_S;x?6D)i8i|S6vFW0C7<#smd
z=CkgVbDq<j9uzQ^7JPd33#_ELu)XTESC;qil1%bD$?y)Vq?L+e11(c9{3huW8~I*E
zMx6!!$-V#api@=@&q8}`=hN_ay*GzCDxVtYkEEBy%zXP8nb;_Zc*pv@_6~}ne%ukG
z+f-*MLn&peZiyqif+IVZpZ8t-?!AdTB>Yk~;>%6vjAbEXd0?0N(yC^kn!oGS>yD&X
zq**prTgsFuq?QIx>twfDnv_`LF54w}CwCcWI1UFH`n(}ax^M7E$~d+e7uNSHG)s;=
zoSk!N+fwa$I9c27DRvcmhD7sMAG3@sV@XHqR86&<84Re21yWEKb%H)(>CGu7ttMVr
zs=rBde_5A&>C^OUGHR9$(}*zdEV3h6a>SABM_M!ThJ+oi3NqXEmSl_y_{XjY;?0#~
z7({H-#$x7EZg6*RSbwdRYj^TZmdF^h?Jpdfm#>p}Irr+P>&K<Nd*a1kb>6ewsJ7`7
z%aj@|$ef9<Z+n$L>hp_=x}@Fv%4pFA3&`ABu!J8VN64M_eBh#FuWS$==jGy_7Un$(
zAmNv!Wb7wzqofF>OGbEOZ<}1Lgs}LmFqD-+cXa)a4lC`BWBsD|=rNwV(S7%H-aSu*
z;D(JF_R8q>(9;6-Au9&0movyvtczUi<OO#`Tsl3Y_ZpN0pWZ8<X~s9XCj4}_=-64m
z(1-639_GFdK4%WVwo7o8jWDI<$#MB65wrfhBH^_XovMPtJsLW*DEZ66>dm_^e#HNi
zl6ISoKqLLdUto-i5fS>tofFhFwc2?-GU|SEQeo703X3fgp9$zRJ+j1XlXp2jj4ARG
ztPl*r<H~O7hoVS~=~-l&uodKtx3^GMv!Yh|-QV3VPh1{qm%kU`p=E4>C(`%Af4|~<
zto(C`STA*k5}~xDkHtr*YGxR-R#@mF#;0*5a~0GjEiE6qSIv&Q4pz1b@8gGYeVXZ$
z@7~C%IXyLvUE<ZZ#xLn%8|8mpCr5o=*?iw;$vAOnj;w*B>tJNjC+ST=&pg{)38@?&
zN7-92%tG7@&ikA?;Z1o~+~+bWj&OSz(g>KZ%LD}<I(*~>SFLdd%c@*Y7^jFK_jWqP
z2@F~9xq0ld`@Ms6%Pzr=iuNwh?K@%4Kd;o@Znea-e71d6|8hXHW^blaeMIj@$e2z4
z#^)wM2KTXe8y^m0zw}kDK}n}q>AOUP8!XEP)r<zz5<UHmPLYNY-{^SmKcOh&IcmRt
zt5-N+0YoVu`n`fQ(mBmYQNznxZ`_Ys1-oAMc@-%-q?Ed{!Gtp1E18Mkx|!hJXR1uW
zfce<(nF-5vQ_!x1qJfHbE`OPNvsl{>hn15-S}UWv(aNs+m2;a~qdeRbR$|}kf4lNX
zq5R|FKVErmf*e%kL6Vl-ABY>fAYry<Ij4Ju5>F1ze1DN4lTd0{zj*qod#)UhP2(P_
zsorO_en+Xy{WbSfVeK!a`e(Nf6tm0;bBypac^bds!V^xUWgjY6u&A)zm}K`^S}mgX
zE?jV)WxJ;zdX(<=TCG#zOn(!`U(hPsrXr@+KR$fb6GYW(CrYxzy`e}5lvCoaezT-q
zX2V&K-4K|q#Wr##*Ssu}mTt#wepXlUT~fHW^~Q<X?0#kN*>yi{+C6eICr-70*^03&
zFpN#Y`M#3;V~v8ia5va}KN=WRM9n|MC71hVPqEq<xW1R%_1sq886UerZuFzfQY1qr
zWwkJpV?*ZzquNmgox9u0)f1VZI2w_t%UR*7;K&tK9KKWZb_J+e_89|bf(0A1>DYL9
zW&?koj0pP1C9U{b*<;MQ!FHB)CB3IIdw`w(LceiVyTg>oO5N>2fX4#m>Q+QGex_V!
zY%zS8k@{NU^q9GFshS2;Txr#dhx;i#32RgCH7>ojQsF}*?IkXIwx{Yc*Pku(6Fhm&
z!S~}`k<VJbUs0nA9aqO@*y`@}&HQUaDTRtY0r6GYEibi}NZ23#iXUOm+fY_i4PWFq
z>*>1Eh@0Ei8Rn{&CwFK*mq~bOm}96Txlf7U%kRri;+3r7m)j$yQ<hot1+M9Pv7z<8
z<F4AiRwqY^RlMluj$hj~#5=PO9A9stUC4rj%N^d_;)n&VyYwRf=1WxYbA2c?gL>0(
zOfG!l*4)AJO3!e%7l)ZhPW!TDFT16D%}Suyt$zp9{)RDcOA?TwcJ^uWLx0~6z2i}g
z`aBnui%I<&a3q=@ALKohzn?}1xHKoVJu2Hog%NDa2#?h(_QMv}lmY~>wFV3;$L|`l
zgxFq0&f;hPq`f<G^BD6m01%4-f=v=wCY}o2;_?>_xbs``<`Km-+@;F;#y9HrI?9=S
z?aFrGp~E=oIF+VMys%E)Ot!;N2Fk@2c=a$|G*=k_)<PrLi=5t^F)U)*ERT<;r$Xt(
zH~Kxq2#`&-y0H|L%|!FwXsE+5xhRod9eY$9ItHTOB8F%`Z6jc-6jI*5P?VgC6|A8$
zMT*ddZkEH!2WPk^jFK!IGTcMFS%YN1D5cv=HyPnc0&{3za(%Orclwkgz{6mvbt#F+
zaE-`<nx0=<tR-co_DIeBSByLp`Cs}9$9Udtd``aBob`zC?e*<^^0~vpahX*RCff*U
zY1$jdnwp@jw|{C+>7;AgRSW|c66KN6m&vCOLO;=CdI69&*tqjDjr%I9)lqlt_XWoy
z-{)1q?+eZn?^};|o>yFhrNm*(rYdv$M$#J&G8BNCcMZGBrtjM%l|)ybM{zrf>r{|!
zLd|nupwHpI%Xs|j<fIAT)(K_&i24nXdn<d*4|^7IaV|dbgx8DNBr}KG`6&vc1|9IU
z{eW?(rnh~?i9F+I(J9hIoB&!VzsBU=`0h5)-b)^E#a1DmwoC!!_S}w6D#1r?M#r>a
znSpUKH}MK4USH(=uvzFmd_O`Yc(1v}68L?1M1}6xhH#Bx2nt=ofzQUW_gh{mKg`hv
zmFQ0sJXjpCaf5N1b3W5H)kMa1;5mIW&*eQNb6%N_{_xS$-zw#UYLI-U;E2QrL#*a&
zL8v7FESJ}N4mIwwc$~7~5MvB};h)$zcuEyWe`~hK6R-8Wzq&TSIv_h9_K8NFd1V*y
zH}dwYET{{|Z|ga{Z=5b?72_-@4i20rpFWLA>!m^^nrY3iXAwJTzAFE%{1AK*yF8PP
zDt@$|NA4zaSrMh+iC*qIkMz(3K)LgJc^$WlxI}87)`+2Q6B|}&kSbeG)Ni9ibr;^=
z)wDBsr(-;lH^yTXUD-XgD?y41%WJQZlVHB=i@oBc<7NMH<{{GUYReCwL?wzcyc{B$
z>|WSyjd}S`=;b!2zhvqcZTjEC4$=OSDsQ_9t|`0g4qR2fLk&RPU=^4WT%%%>i}2EP
zPM4z$2i2=xIaUs@v9Fo_aWT1=-)H*rSXknZS3lyzj%4M;L;;UWrFZ#)3wNAKc`M(@
z?X}tGna%rqJMvTqauEUAD;`v$*OcjWSG@3H*M@I6*_E|$%NUpUw_9@g<=a-Qy83+4
zcz2s#q~UHkX^dtIw`u|7-C7;4b$=<H>`Y7bydR^s>L;`0d{$i^qly<QX1@W+X{uj$
zwc6>$R8g4FNFjF~$SotnbG+8KRZ!Ya)$w=P+g2>Jhej;9uvUE^#Q9@!7e5#{!JR3p
z7V#rVLa)jjsLrwxdZ$!n)tyI)3h+4FW<XDCf4B_^#H!W;T<`sb$J<iu%Xe33^}`O4
z9GiPk%W*u}4ezc)KAy;6b+2j9CjzCtvLAR{vT914?;}!ob~KB);q>cuLl&RsMn#fX
zg~%;g7h4DjPjZOO7M->&)=%CIl3UOBObtuQ%5jV-IDe<!oU1b{tT>D8`H#^tDh^dF
zEig9MQeqliREVf~dL6zB&2Q(A)`pyPsY(s}eU%sKu$n0k)9^l~OEgs7pL%0-uWS#m
z*zr?^ZLwdQ=xs{>H?{bZ=nIQ(^6fOrZ|GsT-Jkca_VNmE^N##EvuOEz;VZgB>5REj
z+y6kWpBW~S8C%Hk;P|7(@qY9c=0glT0lFidFv2$57x=94&E++(w@Sf=G0oEeIma+y
z4?4H|4t&pgk;4<2$Z$#LRsrvt#c}P`mzMZX_sOZt`b$S%Y<9AgRkh}2kj7RrThII?
zM69;2MZ`_Z#4QB5khY*H%_tfrA(GZaID+%Bgl2^|$W31}#m9V0N25%V5*J=6u|2Xu
z&DVlg-AJLQ3<^7+>C~+#z}J|qYG3Nq+PMdvV~6lXom(xJfM3@H`fD*AqMSU(!E}U(
z&I_|D2LMaE7}k;Sj-hFq`U>(D`&c>Zu$7c2uWmzdpLQBB@$aWE#7=Iatf!z%u$*;|
zBJKP^hWc2V@3Mo}n7$~I@YQvAZBLM-#!Q@N#Oj&zn6*h6*5dM6)LeG`_A{ChyR!sL
zj#Z=LZ-C-I0T!gEk-U#QCCr%;Ne5q76XQa}t%1dRQFag<9lwm7=#8@c3DALe8vGP1
za@6o|Yx#MT?vWCgUE5QR<u&U<4;ToTR?ZO6FZz+*DGPRO0yhmIUI=<|0BkO%R=4Ne
zU&}yuW@&R4eX6CaPtR$Ze}YtJbP)(hcQSYM_dY>|jcAiU;t7hN5g5KVdaWe8DGAUe
z{y8L38)Lq{&<}#}N0H{(9i0+4xTT!KBtt&=adKjlk$ffX6BXY<SNn@J5V5X=SP7DR
z92v{)C;DFHvIUobg0vYr;dK?Ak1^)iiJ5T$Oy8ncpmRFwB&9%9>pgJV(=0D9#yA`a
zwqQM0z8dZTULT!C;A671jn#<fn=fZJ)GFKYl9u0WkauZj9&R~7@K7Np*RDs6NV>^q
z5DPl0OP|zmo6BF%92;zOLt1=>juOs^9`IwBnEc=-0l|Bkj60`%t&cLl1>64d`LpC&
z2G-blvz9o-wq~@5V@PvdFIE*JbOb*FgT|9mVBW2?a!TZq_US7$8I4vzKQJ)-j-wQb
zD-Lb-fPs?z7bK`WXUhpgq&`DSha-XZ&DsGL&)Ewkr}&gx&VXIe>6kdhptd<$%J=A<
za={h{@cLd=@ZFkvF06feQ6avyb~9kPb}Ge6kG}>8%-`*b<`~_-#de;+X!z7}^Hgq-
z5!{s^xS3N*ErJd)DkEhH(|BB=YMoIM%!kSRMSPh_jQ#F)*`@-0dmg=#^U5@{>mP+z
zn04FWKHZMBQM)`mzddrnOKQSU*=F96>HK!dYL9JyqRZ#^zNDyAF<$rj)ovJ`+CMi|
zcx~fR2Xpv4-sP~>K@CAIxjEAe)>1>rbl1}heJEsDWnn<IR0xh_J~oLQ{THN`rm-{2
z%Zca24du9YHU;`_Dlh8g#<;J-G@=;^f!t&ArWbu+%B>@h+{tD~^PTECqJF=VJEnH#
zYb}n^^1to(W;vi#OmL>4YbfM->8uA;pO<R0$TWpXZ~4^Ss{?BL5`HUN6+FG47sT2w
zQ(AFPH0@SjA(B@9pj*LS%8S3IY-mss5--FsKOlQ*h1H?=AGD&<V`urv^ve#iZvNR6
z$r=%|7|9eY0W+4Eih)RZm}ZXECK<*h0b#@eVPgSgM+3s%1ds&=$SJKjZY?~1$d>%%
zTr&E+HF%&l$dWeTiioL*iIoM;a1ZHfgELH^chynqE8M=G#mD0sSe_756%t1no#2jB
z=Wj6L$A|VCGkOGib01sB2p^#r6Vw)L;0M53>&MfV)cQQhoVgDoqhRnYfOzd1zpUuX
z6SKvT1d^5JMc*+^<cA?gG0gIaB7&&7J&gD<O8l|07oSMYrJ~Mj=k*%RMh2372_P%0
z&A2dhqxT8BE8-Gl6zkl0$|4P;a^~b5t0xp=^tIFxPBh#es{F<wWr)seZxbVb1~Yf4
zqU=?xgE4r~uT4_V^V@e41sg*S(vmMV(0T8V8L7{teLy0^XlbZYHwSj24t8R^3pNBm
z@08yN*PEeF&ry|8P5BC)iqR=<Z>!??kG77Hv1Emft}urzhO|dwsw@v4cA}%~+wHgM
zREj5H<JT>AN=k4{ynX@6d^eC6Lm+x$Dkl%`;&FF&LQ-mfzGTwogU91<hoj$=bnvY_
zZGZ6Ys0dL6_T<d*mA6Ee4Mf(q!vjaezgB}e6A{(6==0qbRko8e7u)Y;hS|I+vNs1<
zAf<~+S)E?x#g8?!pW}~Lqh5E=reYfwL+Kp*f<z5Mb0&@fN#U13$7p6{3CR@6%4LBH
z1p`UGQ(9D>Vun-hr6o>wkL%eD9?&&*KF%LF<~SA<*+4|S;J9?rUNP7)@w71FUf#g+
zI0SL;Ylm5&t;pu-Hn@ry6>-{y+Lh~z6qEiEDcDPzRBh$?x3|^rDUi|4Z|kRgDqIwK
z8o6WQMdO1y=t?Ul4Qn8>A)qtMF!J2o+dX5`ul6bQ!tTkna*bk+xp+}GSl~&hL`im^
zg(U^bNv=8{tA8&RG-le)?QF|0ppn$FEcO0QC{OyU9kB(G2Crn8IC^K4F<g{VaXW(;
z<lDny7z|VA*UA#}3`YCOSp43RpT~UOv%Eh{HC@f9XH(JJ2=hAn+u*jnc-Mvg`=Hro
z{%CPX#YcgWvBEJM*XU<1<tsJq7d~!&G55YulG)`jdhw2w%8lu|RM#63*8Sf{BBy`(
z-1X%gS5;UfSZ^U&U@V<K=kZE;T327bWCWItkJC9}_XEjh_eL6tHVlLolLsy>(xsJ2
ziGMwQ_l+I<z|mS=<H_0doXBO%es-u#!g787VfMbwi*%a{19=vIbeqX278DVh+L%*Y
zkov$<(G^2)|Al3%%B>5BIxH(B7i@)@xg%I+ggMYFZV1(HA}9Th*18iEZxpK$Zo6?8
zOI5{)?+5T)>mCWC-qw@0vQT6Mcyq9edguem?FFlBCTUtAhWF->UPoswJDgHK!NL>4
zPQ^`WP+achzprb0(T-K*K}JlYSJSh2{NbV(?+=jZDqfVzE8TL@#p63zt4nHEj=Y9>
z;Z~(wIyv=T+bbUO^s+Vjo#O#ilGVCBE%3B@iKFub`bd?~*O=L_E<HYZ_u0<t(4A9i
zLD^nR^6i1e!6n&8NzY+wZ=gLhfLFlFdzZ1~j7jsg{kouP9v-7)tHRe+y%ZbEPe)jW
zgQ0{J(k4>2JJj{`4QUwK|3cyHDTp_XqWGdgZ??-$z=XRx6#61;ok=$EtjuR%A5_zZ
zyw=EPk})bkFEW{DF_LDXZ!o7x(<<*6r}2q*;UucvKC!Tbz|gM0XmV>ei^2K~eiXtv
zyXB;le@xyP*o_hyotbZ0=kZ8g2`rKma7vXXy;3ixiRQkjg_-;MDmiO_o!B=Rd&Kx$
zz|<|Y(gN6gNIw58Fn3se916DiP@@R9Sf6JB1DIi@9F>~;IgGjVW_y*BuJ;bDe=b~X
znQq*JyhoP<zZ<j%;g&S1%|*IBVv7%aqAP!LP`0@2Y34m2G?aZqeXqjAaZ|kSRMX6+
zQ6Ezu^%SQhy?_AA7t=*ZxLKB7?WmoWoI8^`l~h6H>+hH3TkzS8CgX!=u$;@v`t!<C
zbEfTg(=0@j)w^O!x^2I-COwaykFcuo5gC-dyl64M4&zDpX*Up!d}(&qGPPZiY#73l
zW=PR0tjBpz#mY0BLXtD)o;KRd{RF)Ow|Gm#?LPHg$_bBM%HZ{763W@*>~2ZM5*9qB
z5?}WMoJ}C~BClYXdihhAqwlFne4*;6XqRac=YQe0e*w6Ey;#);%FFsliDuXDfXp^O
zLC|qo_A6(kiTH5YOzkp8L0A%kBkG!`G2~3#ENxxL3<@W(z(I%yS1a;hMLBIAxm;fr
zx#B>-hcWL+uV7abSe#Few5*la#LBE@)nXs+U|B-6qpP`vp8B@RooOn=T*gc7tE0s+
z>O8t_mRBq5%qzp9u~>nVvg(Qh@e(4?``v1~vxMt7fqQyZ;BH=EOuMYOsrZ(CXQ-W|
z?M)JAD)*NLUs*40u+EJ83<X_^@FkVP?!FTxUzcM`UtMPWN`ZUL)|tO?#=0W0>Vzlm
z=xTyx)WD<{yL5JauJ`Pz0S^g`!5iTov80nI6Zzb?iQTQWZKHv10hXSTbS6@=LNLfO
zH7odBv1M3j@n!s|RuB!T{wb;{F2bm5vYy4bqdop;4F1qcoUeV3goJwWEZD=pon~E0
zto(df7CWG2KuTuEqQESsbk<HpFYrxv$yc#mqJ0)asH4nxV(sMb>_FeOy6fUCMR*#g
zBcNv7sn=9rn!a{(OH?N@^9kbz#`1|G%20FP72(j<7*S$193czkd-D>TDbD@;*v8%F
z(dC*jt`)|E4v&;f#;BgbcA@f9S9m7dXwTLwA1t|Vnr<7ENB>GI-96bwYiOl?v9`LV
z@<{#g!kPKPsI^WDV!y$k=y%HtgZf|JoJd_LG#VRulVn8IQJB?!HmK^vsQQv>`%QBE
zs&?QE$s&KQI8+wA>#QDvxM8n*uHtRQVK5A?Qycn+=<u1#NGEXlZf?e3_Zhs1mwu9n
zDh|ri{F8vS>OojR<oBlT>-~SE_^M>Xzm}#zVleXe`7<X_ri=t==zs4ylP`h4<t2y6
zG7qjlXK4R((Dum{-ZbFatReAt`mAN(Tkz(jUwE@&peE!7{K!zCFW&B+EQFfG80<f7
zdJ1*_^pDt}WO_4)OeAI<<YO4-?-c-IJ#};eK=1+ZYG_1gIG&fEB>p=o`tN+!x3Una
z{x-~Qsn_{$*qFZ^=@(#@e0`A>O%4KzU`C{Md&l-jW#Zr%*mD(w2dJ-ON#!4)>67Ca
zX;c*yL%~3t7WINm^}6+vV;V>bvOOZK0cus=@ilGzzyHjCm%RJeT}9xxi-|wN{>F!s
zrXxaFZ6OYm&99f34Lg9QcPH#{veffG?bBu`h!6+nC7|ay&$a=*fd+8c9Nre^zu~w4
z0or}y3UtY0*T_Jf-;FZ==f(No{-O#*ykaiBYVH3=;;#n@Z1bKimtLIwW5)Wwec`{q
z?b8YP2U7p;BKseH=Rcf<|G%CBm<J(-t0J#`5B8~aKKG8=k|JheU!e^?2BuxE$-s81
zd9CwY?nxQZH!HD^S`}(vh$NwbhNhk|2$~9k)t~)#6CqDW|H*+_I@*uOPn2osUl5c3
z1+rR5%J>SL#=Kjvq2CRL1f$Z@qNq-)0bpjlLyYS>!GoYvN=0o(oe((KdPdQE6EyXt
zP(oV8NY)gIAHfg7Q5wv4f=1yqB2I*u?8LT`x@rd`n;EgA(;T7WH<G3W=gHoZYp%^n
zMO%QY&_g-iiN0GPSds2?>)c~fk>XI`IM;`tr4=5s%;HDD@gP|bwIn+jJlMlLd{ess
z*iI6C{DZPFPx}J@;kFq9J8vVpC8zv<J(5oHe#iR|zR)`$FP|6%kbvLzOBFv@U%0xa
zb|xvH8oon$cCrq5c$8L}ns5Byg1y$s-y*DO4BVKR+<wdxGcF-QyR(VfxPHMki2OYQ
z!O1V=)n&(<>VbU8e&n>^i~jPGn@17nghs$0wR88O03!w8(7TslvU0v9WsugeZVD*u
z$B{JcO)$cE^ZMobB-BvH0E>x)Ee)bX8;@LUA8#CjKT4u)93fk2cBS#tbf&N58So(d
z3fw-|M1%sBMk2Eh*R1-KBkf{8CH+Z<MZlg;qQguwYkj%FJ=Yje4X3f!&>%!^xCjsV
zsb^h#UDPNQbw;!lKRi@d$TuiE%zw-OUuWp=dkpegx(KvlMn&|eg;f4rE1vM7-GE@y
z2))$;%Ok}9Z+CCeS1Y@efnfl+uRObB2YZ<mOdgwf54}*b0!RsGFs8yV1tg9dED;=+
zSiNxxNJ5^fH-Q`Yh2Lv)$sUR515CSfSjcJS6aSdk%36xBOCGZ<YWJhzzDBMOcm(-#
z`p70)=S1=u23Gw9%3TpDN_VyelsMFo$^B~y!VRl#SjQ={ligGmN|0AO`(PfZgp}wA
zBq!<RlgCC2Tac3do%<0AF}WcF8z8bwuU}ylOdRpi=!)pq-oPL97!I2YLp|h-1;*5I
z#8;0N+&8PhfS4oMsT%m#c@!w%lP}qSx<axxaAWmk>WqMq$HV`A=H9~nLNM;uP#F4`
zdO^jS2!REQK|!O8y2NnU2Y@T!O#NzQZF%>3t@S#kw893k>w*B}=V%8-a=O<oE@wN0
z*W`RYmAW=QeE0<$%!oF}55;WuUDKr6eL1$b;Q9`%?sce9>(qvyt)<X=AjP)jjT;5<
zQk?h~SPR+2=pZVFDs1G70K}cB3Z8S0O$Xa&hY-(1iRy_UHWb?g&po`F*lcz&0A3*o
zJ=PfbC94WmvhKBzGKq}me#C|$V#MN4G896ArnUMzV6ztr<01h>o^1?fhxB&y$!Hss
z+J^4He`pXrB4O!~2+=wImo->&GD02|h^gD=B|OM{3rA0fV3{#G1G>oysLc)q!SiW5
zQ=Ov~z1C6+jBp4vp@<29HLM2=T)b=8nSksPsUI)0b`gS3;&s&%^;EclpFll6iug6X
zgC}(EZc0?N5|MPb0qZ8)$VyP0<QtJw!$Fo*Z^AN%{-O!}&n@iJDJbV%Lh|qbu}=C&
zQ*Czn?o(~H?!Xe@8P9<G{BYC;0YYaa5NE2E;kSpt$}?gK?zP_GD{Q<&SLm?K6>z?H
zByIG;sI#U#G(G?cocz8Ciyqyk{8Qeb3kSQ?QWEo{pjr9(Y%{wMR2RbN_u;~bdo(<@
z+?@gI-ze1W6k1Z-K#WaG4HuzYcWV#KPsPugh)JK8*cJludiK?Q<?uV*59^?|VUP=(
zQ3R1&(uM`FOUa^0QSgX-T!M9GrwUkKq5gY|UelK|E+qFvtj5G_XPXsh{#K>{2LGic
zu+sOZ?%0FhHXCj0lJ|u7=vRn4vErQeh)5)lm)QjEl1QF=W%g#>!wa#N>avWur2??u
zAxU=NUHgq(JaQ9_5M{RRd)3p|G>+0nSU)|Q9r68@fj?txEN8JHSQtpiajqxK5$hVC
z8NX5F1(5$&7$k4(C^BCH9Ze__2*OVt!Ci_SmR!>q_tQsGt=96aXU@z&XFQ=12LL70
z?!2Av%888j^FuOYXfop!WEbl2Xqi6KA^Zb0_s3yv(ObWfG!Nxvm91eF!DaSs2A|+9
z0@vmh&#@hYWC(Adn3mtag18-YIsP#54Ci`g?;E;k5{=0r_Z5O=WBX<M_Y=!8FZ4D)
z!fGpJ_;T@cG?av|m-L<=1N#v+eO~Wp;~d3WxE2AC;{0zLFvPsoujX(7*KbOaAd`%B
zO1V1IJ*gYxXmbvj=D9nxvMRa!G_Uy#(IO=b*>O`cLwwsdw=ebUda2g&b*u-joeCj+
z7w(0ge3x7ZibQ|G0y*OPE|2Tk(sKmD1;#zftD=3yf*jp@cYISv>5`=eXeTS+y6&Wz
zzcJr48l5P@S|;^5Ym7`--2BSW$FNJhAMJZ0dEB=o2`Bk%XQlpS_?xyRp{+{6YPE9~
zIH7%DlaV5M$X{(DCqZG5PPKThLh1bCiLL6Cwa0@yr(C_1D^15vmbF;f+>J1+<9l_V
zs$%H*<iA&Pe}8Ol9EPgy<C7O+f0ovvWL{vZ@HL3)`qB)Km$uxIhENyD9v0mAp-rfH
z=|L7=5nXvip!uYh9>%tQ&xPh8^NTBAg3+<-M^0U68V>q2DN?_snpF9)ZUs!Me@&8-
zyc)WHUaj5@j6rrnt%6imqkNX}8G@OlS4)ZU4S&(e2bq4!#0PFhkzUFcl=XVZ1BNC#
zRq$x&b7MadHL%V8`qhZSO`dts%FJt;Z3i8T_hxUA&v(MiTZ}AJL=v##rc#`+yoe(y
z{Qgdi>nw(|N0~r{F|5D|)FUKK$I9uWdx8hhiHM;{Z4@>roZN&*a(4{Ap4WEb^mgFV
z5rOwHQ1L8fG48d>`3&wLwiz`8lo(bV15U#1BN7A%w#2qlLGwo$CB>oJ5gt_b+WC`k
zOyOx1d=v~wczAChbYd^M+i6*<q$d#a`l3Z;smnb1*A|3S-u2udBQ*Uj^|Lkdi3HeF
zy5=93W5Htm>GRrby8(d}itC8g6oN17HEG{s?5Y7=_>1tVF4IS(i|CdIe2obp=y)e5
zfW4E<s@S7(%YwDVl5t+%G(<J_97f7dAE^jOU7FDtE*qJice~o4JtBW^H|ZE{zOw_$
zv|xlj-YMmoW;1srb!)K8XKFoNDsz4LW_C<d5fj=Zax?gpjKb7mlgA!}x>Ev<^y=J_
z5s<O_Lt3vK(#Jxrir(}Mr+6KMUjJee<{8T~6G@r`81Ga(GkQL=eIZ(i71bm%KAn6B
zsRtuMmqbozzoi<h!C#{b20B-6#Q6D&Bxed=S8xcZRyZ!a`;L9tHG2bzE;xkiw7zny
z0t)9EwXc^XsShGQT{1-LHZpixM50lQpi|rWnCBs)M2bR*Whu&DxwW2pvUBaQ;muzE
zgL%Q><9)d>9k0o>nCCh-o#q%#F{H?)d#jkjno%mS$W;7s(g6AV9hlsXjImx&JNwh_
z_2H*X!VO=VsF&)oNC~H6ICk?*ffPay(M4GNjR2_qv=Z&QYjZHkcrdegw)W9dOr=rM
zrq=EV$PfBoYG0}-Ia#80s*YWvKEbd#ydq(PWmiYt9qzVw0RPTgnek-y|F|r-?*4MK
zeQpoP6sB_VT@JfP7GKqd3TlWg*L?|B73{vP&-&l{btb}jq-&)2sMh+UkZYzQAYub=
zjCA>BnJVE-IhL!WSF-7Xq`a{5rcTfX?}Tn#7km^`3}#CAkcC;|0X5+fSt6_mwlDY-
zu`ww=9b2Gz;tj=D27DGL>1mAIOR&)94VS46`sCN+OiGR>N36H~Zw2Qrsb`Xkz}&KU
zBl9OlfNAEcp4>OuCL-BMF&O0KHwE^Dj{ku5;3x^b)58b?H0>0zpBEd42jd!301|1f
z_bEx%%}b9x?${Z0@>_xudK?=|ClitMfOM0%<Y}r?xttTHWkj3f6;$|*R50h!>a##P
zS@G%qou5q^(8jkGhr<vYW?^fR7Y}veKMdAVG4lSCS)zmjL~gkcw=q|!CKXNTw^MlP
z5B3(%V-nH5Zz=Icb^HsYL)cMyHtC^oltp{80B?{kOv2JIktOQ5?uHZUdgsgrPVVKX
zl?TQR$&75ZM$NvSrFl9AhEp^Uvq4Eiz@qTfdw}|Gz$c2-H&99H$CF=VX*Hv839DKK
z402UKVS^<a*VRrT8y3mej|nDoAg|O10?ic~m%yq+x@@>Jjk&<VseVH4a1N7@{AfVR
z`wWBtQG}N58F_kj;;nVj6D%gOwDEkht|t5X_6@|mr5%|T2jCC0<{queRChN+!1hbT
z{1{-vk|N7zp$Ex|i21dK=2`*$$#F1mOCCO?e&+FAY;^L^0Wlf5DFT}jSdD7QaR&l?
zCG<YapGnOk$YVRuHtVMMuA#7ROfHgMEwP$9J#S`vSBrFK=>xVDZNvPlgQkx!C<u((
z6L${n(hFn9Bevv;F72N5nS&~^D14V(w@H<B3r)f@Aln`cLrdC2I8^j$p>y07JsJzI
ztQ^!fgwLnZjn%PNTs|1KmhyphwK~&W&v)}3BlR}(91UZ0JuaaB%8W$jmSW$0`To-N
zfTD)!vd)Y38XvZGr!H+g3`l<M3|(oJpA9k~o5{iMUy3O{jo}k+7K?lvDJ9D!cGl{o
z)tR5XzT9s;RLbv=|KwOnt6VDdSEx-~;;D!kGk<SjwwW@wqC42EB_OhSh_vtiF8#`R
z!=>k2t{G0MmlsduM&I@}ug)|KY8<$BZr~F%5~q+f2&Wedb{wNgsTE<>M|%|i21ET1
z={RSEpZAknf8j{UD6)ajUQ3xcULFx+bwLdE7A8y1clJSD!J+l15;{%!Rj_=^QtdeB
z_Td&}sfc8$)Z})O$G&^*`;k%O<TNALsUxYEISioG!{hcg&Dak|u466@wd~{#t6-_-
zS$)eb)$u_j_$CR>X;p4Y)ifGaqLZq(&^#SBz3v&tCLxvqer*m^L~>N;WT~FsiY_r`
zOv?%@!-cixMW@o9O&Pe5Qk<Q~z*`#gQ}QtTkFeDkfs&XXnj6pQsd!{}q#0FL(W;iQ
z$99V3ZiKThL{LBGN~wDC<IH98mZl?kB2_%@yDZg_+xl0%i_5S&qkTFq#<2moFx$L^
zj>Tkq{uEMv->Q#w-b@?&I&8(3xJFC-AWn`dPtOAPqE%=W(Co_dKPs^|MiX}j%raBQ
zLQYPav4mUV3iCd*Z$20uI93_Q$F3bkttgPP+1T=E`9mVWY;sgQTW%|nyIXS=^VYgb
z#*2YuIbQqUcDlTbqEj~8cJATcR>#mWI+oi8?z?u&%8sg0p{NaPS$H-u)N(z0<oj&8
zOLXiWQS4sf>?YCdv>h4iHp&vGMIiahGT^7>L>JcgBlQ%m<YNajm6A!E3ClCffN!{)
zlk>Eak;1x8ogqU;xvhI1$|Ve{Zl_d@ZaQb?FvYN=BG~Kl#4=oxHwNP5(2>;i7wHRk
zQ{047EW7p|hncWi>D2Mp3q#+V_l~`NZL3u~$E2|z+l8yU79M7nMdnKPx_l`mV_hgX
zsyGk|MF*qwMqyR{DPdcgu2$_7%h$||X1E<o+z!^aou-eCDo37b<-6Ke9eg%tGMYW4
z@JkG%kU;$gu~b5z992}!-edmyJYK4dZ3n9IEsx0bDGJA7T+5bcSeV1gbBc}BQ=cz*
zsT#AUh?Aa=2%p(+m#wNhJj56z8xcvXs?z`7_j^9yWIB#J^Rb<7S!KtbSv{Y~U8=2h
zE5knlg8ziuiOKmnT{`zy@kn)vK=u~a6%AC9q~pt^Vm^==;+$<E9uu>LUE~fn?6CMm
zEjg$AV=$#Zn~GG1TOyxg4NWho!EIt#j+Mr<<n`939UsJ!b2sYHlu+BqpE1D{pMx1j
zspV}5qP!hMnkJiu{lu?p6#a=<{c0@jMSUw^w6-@U_W*g1hR6>~iQ%21kfD|h!QTD#
zo`vr9jYbyI=GrPQPo4lQSP1l-*m4H`(&iD2xq(-rjn>dYkiNR>c}6hOGkH{DjZr@`
zX%^eAv2wnC)fM>EFVDY~yM=v7kP39dhr&A96-vXu=-Tn18)RWg$rXvIvh!-kdwE8z
z*HED@)IUC;#uU!1w+TdWPTe)ZH=Jwf4w0jEEa4A$<w5KpCASlq*j&3*TE~Ve#}))n
zh>Jz@pILUTWBPGTHf*39M?9sbwZ42eQ3gm{4Os0<hTNJD%E17dC2gC_2jf0jQv?L8
zw(p6*2B&FpCD)YBPvETINX%Q6c6Hjm>OMXtCF2?~Gh>x5=N2io^+64nqqi<~PWnr~
zWy5o3{%bcfiPCzaEX!XkWf*bw?@6$GW=N45=_@(gM>`YN-@|y7B{j35nhq~@RfNlt
zuUg!>zp9gTi{8#$<>Qi&=5f2bp&O477r2&B8nyD<HIT&A1VGW~!IQT*_r4Q^iObj3
z6T8}3E#Fs`Imbic5~*sM`O@Z}mdzP9gfw47^5@;d&rHd*#&pj`ab7X}B9%6A#FNQc
zB}JQi>bBxSJh^*qyxFU)dn~y|idkAk+T|DR=<X5!JV|xfEj|18Su=ar_d&`HIBEAh
zX)3QC6=em9>{EhYb@}lMnnRa<@Z+x(AD_&)9cQJ;sGy|MJ%M-8P>y4dKar4ac{bhj
zT=p)T_t|??u23~ORej{^DmniDvG?BbRR8_|_#r})bCQ*P$OzdgI~^e_l)YDkRA%-{
zM^;KDLLC&cXUNPbk&Mg`N@n(6zsKu6uIqhWpX>Afe7?8OzrX)Z<Hk9!^Bj-I{c#`j
z$x;uCz0*WmpCHbQ!g^`EP{F#(CW$&EsZu!B^Iui0Q0#P?pq!#sO|-Wv$r5kgoarl5
z@lE!BD4w?IR+L6JvAKDyw%=Rjieqp=J-nJx_nmgwvTz6vNpLt=wC+})n^E(~%`eR*
zT{g`}@Z?(XF9#pNF)MmHI=)Z3^$%Wll6WbVZHq*>P4S)55YSz(NT79d4Ho|#DYjdC
zy8C1wmxHvbQfrNoEi6GJ=MStra{{hedWS50IWkpMXP#;JCFO8BKRoy~RPC33QKpD0
z<Hdm+oupTF^<Hgv=21NQ8nDnxD6ari>0qbT1G+b5$0m6F+p#E*MG7Tnt~=9^OEZz_
zuL`6-6<W=Uw{}!Z@3%ZG&PfZBxM{d{`H6GS=^lEW?4P|NJrp$L70$+?+FCU2-U?E7
zG%h~ZoKJGy$q#=nv+F2##q$Z-0Fmz;fLU2R;e?rMEHe(~y02Ms)<WXe`X*jK?o8-j
zervvvTlZ#EOQotc!SJ$r@znZ*d%X=;_skS4ENIS4>qjWi8~hXyHF+{9n2`EP`81L9
zm$S324UWl1a!Ix8WBMAXNyl4KpO~-CstQ$svgqZzJ2kDYJEXg*eybP7tU(!<+51@U
zlOb251%}m6U6ramp_W-vBsZ_4l{4a$$rX3T`D3=5tyy9Hn*61vYumTPBx7A_{*KOz
zd#;O&?~WOM`BSmR!etSg1xHb{o7`8U1etw&)P3?gQmmlbh(7N#-c&>K2K34hG=D@L
z28^UPF^UC%s#m|5M!#+aOJ<@M;ptAZYsp}o+mcrUa^lyyKSE@KnIo=$uV+&Wn(STd
zct$-AQ7*xhMEWpGx&dUsJK#)~e`_D2RqCPPrbT;mL}%!u3)xw-PoTb=MZ6zZ-!zDN
zPP0f7e4>elgG5mYXU*JBx8<Te9ISL?R_|Mbn1G;5KDk>R_Jr;0#R36i6#jXdbdb_D
z9M9~%3gb<K7!(>4nI5U$HnU``cDNi6u*z@UgtH#yJU+EpE7H&hNykuD{C!tF8wn@x
z0GN|@(f(o0Ro}J_U+JaoAqo^GjN@z3u~(bMm9w0WaSIU}eEbQx(1x!HD}$O-YQzc=
zxj(IbU|DOnzT}zkKBGs8@}S=vNlhqnxRsuoAhi;opir!KQdLw`D`6Hex_V|R?O{Cx
z2z<A8$_#u54E1q>a|A1WK6SK&juD35c4p4O{Gp$RN9B-zS_L}1?)V~qy;RQ;Up+J7
zF#9#IN?Kn((v#=EyS3><(W^~nWrVv*=a(}M(#^of_oUrlEYL4LSCVuS`3i-yo|lGH
zUu*p1QO8bK>D7<E7Q6CbQ@m646G9LvALmw2-?V+J8}~!(^>I7twgC^MOe_fZ-7fjt
ze90s%Anvc;XPNzRC{Ch@<Z@rVSAJ~N%h}x~(#*{2+|tQj;nk=!=pC9)c&}<JJ_8j$
z)%mc)i!r<B?;a6MZTX!2=6xqK+14%~=$&`ih9s;&48Jym$+3w@zeP#Ld_y6Sv11tD
zt?b4}ICNCvZiY90<_HKs762~wvZ;34M`-)n+30Y?d*9sNEUf>O-@mPJOjir_J_;X$
z=7+f4b8H=y7|!}<<bil0yi~j(PMXlx?d%X&^s2(iH~~j5LRV5w1D1Ihlt+Co>T^r;
zmIp~7IUDcegFOL`yLU76aYoZ{S*n<jdrd1<pMNXXYvOw6{E@S1YU1yl2Ag}CK{07x
zDNV#aX3BuxE$j$BLmqVXB;Exh%29FsmHI5P0#sS(ADDt;DPL9U$_j^VC}Ziucglc_
z_eO>3CQoT}Kyi1;o+DpJNV>I09Y-^0dhSImxvv%HeUO)*JKaV72J@)Bv;Wm#<&M=h
z&iO5XBfczL*z2XW^sy5V$ZZ-L4oxk)D|rXo%qk#%Q}sof$|xBC1a?^X2CH}009Toq
zA)eR84dm!2{8ltom5O^uPd>#Bv~K0^RK1;N-?28CvM!DL{mv_mn6q=L)Sds>)`zR!
z{kxGrIw%@5Y2$DC<#&HRfY>g{5pCmH1rs!j$t62FS?J_ybrzL;ZsWO80x~1N>JTJ<
z_@EkD-}Nhxw&zl>i*uOS)>Kl_XeiMH;v(MD3$|Py`Y1SOT3npep+o0c{wegvrxpU^
z1vNwZt1w%#cOV)^?5hR6joyN5YAQv^81Qh>x9v*fGv4PFyjSUeXIiQugAueM$T#D>
zO+O_nV2y0*?Fo-h{XAnqsPzI+-e*w2+3txWj4xzd?kcTL-8vP3YT|Cib8oez+~aNy
z2UWHAlkq9k?r9Fg;*5x?OKs-c3cq$Gh%DM1mw2Qs^=01Sl85Ocvf=}5+WaC;(rI{I
zUY6mxR**lb+R@3c93iE*>`&ufu`8;z>~UrJCAU_5?MWk&T)cBHKX?31xySxR!?ri!
zP^pM2e8~M^b<Ox<d%c>VUv@joj>2nOi11q9OZ3Y8a&o5{ta7-G;yOyF-w!)YHKvBY
zwh}8%s!Ev?-=n)C5s-CbA21j9VeXfwWZKWs*j3K`5UMRp;N?r`6A#Fm@{3ra1ERpD
zJ>S(-ij-6e89xp2MMD$d;}VpJ+-mH{Cc|b*RpMhe2d_;b7g$o;$~$arBhd2G2i?1M
z__p69zR8|0czbsRQlJtx+MnttwayxZHrYGGiz!$4i`SKm?6<}4*Vtl9-+-yH9!t~2
zV&dBawh`fjyBmmy1<at{&1Pmf_dErkwao+*cS4iq?^jnxWEgeP#8R?F@K(io{;vfy
z8*P?K6a6M>B||x}kvISA#ixlOPs%b_{m0nKEeypLF9?2)i)>b65oSNx>VZxF{ar!&
zjmXD(r;S=#D_Ba%1rUHG$~)cgCdNHIW^R>46Hg-m+!?pz0u}J@xJ7aknRXUuT<9R2
z^cKU=8a5Q20oOeGeZ#DgQ*C!k(sx`%g3=@AZppFauY~WtVChhdMSbq1mR9q3q>UE>
z*O^ntKbkMY`RrpXm~^w)UUReSkza%z13hF9BG|5ewC!mK{X5{Iyk;;E_Aq@g@}Z`b
zLo-QWQa=B{$nWP5-HB61ZY}r>(t^v?2BFRPGsHfV#V7!PCM>q;IEu6QTB+1%S?NvP
zwCQy?#B=nUdYN;2xj(t{pqOD8xU*JPGe;=;6?#ipfjv%$F0+Hw84(Ma)^V%P!&l=4
z@n_#(FYH1#O(R^*c?%{*WSAqI-o=w?9fXJrYFb~NX#M@P4xunE9O6yX?mPw)MiCXq
z7mJigV@DD8hhSAj#7)63CFULt80C%Q4%)+C&E3n^Y-X-sFo~-Ya$YaaQ*CeCm+XoG
zJD`50r>nDMm8JN5pCoF+I24ow%;%|zy|gU8g7V_Mkqh68aq#<k*nHIRRRR4@9pn`b
z{_yPxvTG_nY_8iKW*0qLv#<@l=u$I#$st9}&?3MqCvU0#Ug7wTClt(SR{iffJD|R3
z#$KuVqq+OcL5wRHKX4lR5R8&CmX0;Bld}lWrbH}npk%yPCM#lZoIG8Yf3ri0UP&$K
zXKV;s#x`|<<12f@F}8OYDZ}Fz=#T$k3+W3}k=+?jJ;mnO<-e6ff9@BUjJat)OCO2=
z4=d9r86J3L{L#?+?VVp6job~lR9rY)kxB4HF+q)V%5pqGPfb1iGue@qWf0=!9yMH)
zAspqX#Rjvl(ntskDPD|?KwHTG40y%;GlKzwpnSq-=x+w!h@RB%QWuMg5XSp97F<RX
z8BA-?pEgQRj**Q2WE-&Ff0d2?yxkpka(MyTTQV@$7z00&Sw99K)He(-k-pG8emeJ6
z4Q}URbEk_L{&CJL(@Mgxm1iqHFa@D6O@GgtZ~cq&<k8rHzE6c|Ia7={lz>&xu9y~v
z$m%={r5EkhA0O>&EPq@-cqg@wG;(S6>h0np4U$^uNZ;&Yf*#nb^bw6m*wcv&LV@^m
zFPJ<|DZQ+<gHW1o4Rut3CH5KSQLKFu-k2fP$ufEflooQ=l{%)-XY0`B7sNlkeM3HY
z_4>GrK#pB8q>oJV4<gaS+w;IddKJOOUcs)Y5vsb{P@5@=RUaG7=Ljvwop)}0oPa@z
zUd?AqR*ma^E0pSN60&im=EKehdhEkC#|~q`p^8CEGp`nAn@DuD*mazP;K}Tu;t-K@
z_#|vb;VFX(w}2+3RMJ&Y8GIO{QiBT|0wzl(b9=5!rr-zaG@nhRARBnF?|gyl&U=>V
zK5Ds>@-H_QVCpvFoTDP6DQ<{c!G9ltB-wjZqRsKAH<**_;)TgZ<4on>u#L)gFc-~>
zuY{8aK9;P^(&$$6!BmNs>8j4~cFY^RUW1fggsiajtOCk?Y_K?i!ipx_T?r#J$b!He
z(^{`<a8h@uOMSs^u36tT;90HAsOg>eO_5CgLgovcbJe<Wqhf<eskU#0${cBKm;U{=
zt%->by*IbduJ=!nPk~$p%#P~hiti#hQt|9345q6r*c?MBbb`aVcT&Q=gDHqYl)h3p
zEjwDJ>iF2RRfH&YQgFYeq-*<-Vrdm;L5Q(Xd~87ovS8Z_@%u^*^&J)69U-I{cqI82
zj+wg9Q+09^+ahX043AB2sm|;Ezzzz>UdfMH3#y2?N73{)k6VfYi8B(mR0Tf!@WF^1
z$ie)ZFoL2pSZVlV7Pjy1uN$)k&0AsuR>s1Mq2`_lrJqJ=fsCe3E}l${sqcqB&7Ri?
z!s4^Y<O!h&c+0x|o>l!YCKnW@Y@9TxgIjOl`5uf&qf;lDuvWPPLgvb4u>!sP;-b1x
zv_F;KD|7BzN?;NB<#)GRc4uhUg^HODSZY5FRwAYZbh<32Bph3HWT4sLhWxmd14Br-
zNU)OAv`#Y;*r9&xTYK50`<_e$_c3*Y<LciTk-^~&gk5+=-&DdNq0xfSY(aQ;QI?zh
z5O?sBAI%&IP2fQq-Ax9)Q-xF!tc>buv9g!@!G6M?r@ifQH{FK8R!0lp#xEcneBxvI
zZZ>W46)I@i`SC;i*_RrbHMy-0(FJ95Q_?&+!~e#5oG!C9CCW5Dp4SNyhelz(ye?wB
zmk`J=@uGMHNeKSQJ??B(W6{A;#Uk@DR5D1?>IvPW+#539UCcRTcW&0RnE^Easbtb9
zLV*st)BFtApvCVgN}S8LQS>xz8(%GRg=-d4HtXZr+`F|$q4cKYMpMznip2+V$E%!(
z?>iCi)%&wh>GEOjiPiGBypW&9HmBBgKLm)eZm8~%`X4lpt}%k!b-7G`POMkZ*7FC0
zKdJoZY{AjH1(?}75~D|3OH6&O!00-<Kymfi&9H6b9a=w$f>`tG^|=-uy^0B(Aa2+S
zk-w8i!RI$u50)B2@%IOIS^QgYzI8a`IznL>us}}5XNZn4yUYV8oJIZH?D#F<5^PJf
z^6an)WDe@kdX_&qt0K)<=~SmI^oLZYh4s{mVO3-AWr>vl=|1A+;~qn_5&ldDHR{Z)
zvoVyx=~_qN=+zglGZVgPCoGbc$Hr}cgu8;0JAPg#J$yZcB5A1fDSh^fjnflf-rOAb
z)vz0Bn5ygiOr|C~Q~jvi6O7I_mv7URHf&QlzYwO&_F-1@2-~$7FYY1<H>37XS{)$t
zRqWdhT{SIH3E0!D_gWU{sRDPW#@-!*=RJ?+ILRlYjv24MSOwdsoPN99adm>&k*NC6
zk!M%^$0Fj_xvkb0Kgj1r!hiL(1EpW&`VUbjY`fm}KVZ!N@ZvKCLK+Fqn*P(u^2pI|
z<EIIyBgjD%9}$rD^T649X%Y{-Am|(_<mhud;-DODl5c%7BoKO>?n*PMKK{mog;45E
zaz*r4@HcIc@8k(mxpsgcN@9qYn5Nybo9ZD}j3LzTl2bX{RoYpm{#ty1gAsdx<UZXw
zUIVTdguz_~ws(S}@st~A;+Mg^WSBcjffNuv@HFC8XZdWNlK3I~2s(y?gwTfnh}Nw|
z1(A-CMI8B2CZb2~facd>P;TJ&%`eN)+Y3tGo?_bHKu2egpZg4g0v+DrRMyb5bg`fB
zh~aO09a+aJoS(ddrZIm0i%?%eenCL$+4|XV5{y2_*1MPJ!#7y!4}~(359*u+f>TKP
zsCuPDKqkVy&%RyFQmMERbSC_Er>W&L|Lwt%ioO%3YAFH_E)T4lznQCAAAV%rcidI@
z5ti<m^=!Uis3m@ifx4-Vpj%E(T7U^w&sB5f`C{1Qauj7WJ+S~u+~xQ0o~<`ga=P=L
zOyqF=7xA9{pKCAACOGT+S7M0R*V!*vB^$*R)jqp(b~j93Cf}9BnLND6>c@(V(;*RZ
zMV`=l{7}3a`DfmR&~&9Kw~m_?Lfaor8k*4Q99%Yg2FEqa^hxyTl3E7?zwuGT7Oh_`
zTjbl|2^}duaPEZ?d#hRJXOC23g$5k?6HLL&;Dvi<<GeTwE-<u~WFE_r7jAYjuJB$D
z*f=)9>APWDWs&=tXYWe_8(sgGIY!F_B_j2h!*pLhH%`$Rei}46yI4aN+EGk1=#a+l
zUwS$7%9Ay2m`n=~A_KS$YD-qE8t!Q;jZ>0HFwLys@oNglq9=1R-a-X@>V~pY6xBBx
z!_%0!LXQFucCH3bod@kWn=op1$wpsX#z&d7d=F8rc+R(E7FUa@h)&H}M;qAa%BPS7
zwvtU?$HaQ{erfe~QunY8Q{B5*X`L|b)Z3fpZM50QHU~58UrYi<uWG2d*cxT5)fdHf
zEJ=Df4rgB(@Ox$7eD$%2@@@ah`H!~!IKd{V|K8mcwu~*pJ{EY8rToXhz&jpAz{(A9
z5_244!o;By+d#0j7xc{4cPEWq{Pk*GWu#pfom?S}EafPR)p0D%QHl)SmLu&rW`*K2
za;?4m_?Kw(F^agtb<CH8bw?HX8)V53SSWT^8wHoS$p@2$YU4HWXC4W5s~Th0y7}|N
z&ITWlOW^cn7sc+QrO~nxuE$7dagr@-6dpK(vEw9%$#w1NS3lD?m<&+_XD|^_SqTxs
zwR!xM`*3cJM%atL)JTI+v1JTeTS>ErI*!l1jFCA|a1=vXL@J`p8GIbaOn$k7KWBh;
z=fPaB&%x41^kLX%xVN!_k!p`+t!QVlu&Ad@w(c;FC$YGa<ddJnJg|*<<ukC`b_iAE
zF7gv?=&p}k9KAg-c>myx;h-6-^1gGkHf)imRrif`Or*$iDy9<0(8_G>&#|H58d?77
z+2p4Yi^Q)nNmefS#D#}D@9v73GYNKwOx}}1Xa3mL=Uf#(iAo#VnC0prT%X0CjVm$l
z0}HX$9=07)cTu`*+YO7RU=d5Ispu0OJ@~OF*v9m|CEtPAl*%cYnn<>Gr?grLA*-wY
zD;Lbq2S7DgXvc}>#B$3$!xuRmzU@-W8h&wbNSb8af-gIXUoJ|-Mstxh@be>ygn+p0
zcAFiHk9I!i1bR(fbvz<Tq$!iVwlM}XNB@HN1V8Hg(@f!?d?mQFDJg#_$(&4K2<zn6
z=43tCCl`q}D}1>~eXvg?(vhNOx%~y(+*{pC=C9?7hR!Bzb$;5u^;9x;Al3YDhX*7;
zNs>Ny#r+@M>~E}A7Gm57t}g<_!VMHjyr7#%^6J^hUItPgB}k+rC{59nuf>(wDNvxL
z@nVWa^F3G&E}OXa<%_H;D(Ea`{=vf2j0m71JD%qY^eIeA^3@Y0vfxR{toR$|Bf^P#
zUWRZm$|6Gt{>t3{{F4GL8>OP2w!l5N3{<RlK<*ic^<Q@<V#l^1NzbGA&mUPNPlQxn
zuQJTI6tOpP-U>uEt{v;eX>gw8?&yh@+nT(1yochg4~^1t_y&y6_qsV3IxVqr2)p{1
zIrkw{k17R<=W(o_{NKODd!Xttz$rEV3$OY)39)r^G!^)B+)C%xaeh#^289!42pRPe
zo+LM&c(}k${u~@)HyuaR91hyQsX1&BHOH4VU@<y@N2w+7>+IJ>Wc~LM5hg3xrM0+*
zyIt7QBP7R$tv?SratN}M{eHZaJkFz#@CET$o<2y>J@w{W@RjD`E*}cCYTU8Dv}Fxa
zB4N3gdGtD8$H-K4eV_uYoD@m#K#cm|0fBN`ig@);*WZ8n0DqZC29pGe693DJ0eqFj
zA2AF~Uqce3T7l*sbh!lFM5%!ZocE01`T)^4NE-T1cn8${lM6CB6~Bo?6e@ZVX-K5<
zsyz!ipE9@$PqL8`vF|EO79srkw$meTJKI^E<5{_hvBG8)|Ds4KI}mVpWPBuujFq7W
z45}9g<YZ0$Zh8tI#yc1x&Ym-D|Eu;So|gi?U+Ez7E1BwYo8N!4a{<zjc|52PyGC^a
zQha`$)e?J_=LSZ`l>9lxI(rKw1^duShr+P@3Q3^2H?%#T+T$QaJ$r8DTolwSyeTN{
zD~nX*$;-9?XSl`Q*n9K0^9^KG=OG7rUR1``-|~<CL$UNbc~DpzcEt?>gPpt1|L_q0
z^OySn|Neh}{xrftz*QGUtR5_z7_T0N$cV&1AHvHBg-sGO@OIBD27Sp7{oLb97FeF=
zt3R0z+|Sz#yYr_c>Mz^nU!J&yB1kH}g$iH1_NO&Dc-Mj9|7&4zkzN2-8SjrEQP|4|
z&nWl72;fScKvI`1K}mTCpv>!Ky)(hXS}2YWh)%>|@_(rl@c{YA0;nbK!YHSx!}wzP
zFW_~1zJX<bKC(c^TzWU-a%Wo;620CO#iU<;4XEf;@Lyelm*51{>U|1CR6&^nHN~<p
zSmq;ou9Ln{avmX<mF_TiS^|siyyIF1W1c*TPK9ClkDUjpWQ*U?JM3;AT-aEW7%5gT
z_6t&AtbQqU;}YAyJdS^ROwV}`MI22WxS9UGM|p!#V~L<=e2&dZv`cWPzsR3+xyY`;
z6Vbx(|NIH!&-`z}lzMB@<%@tg-x=~46nqQLR(mdBt3m|=6}Pe&W{GQ;&wy3rI~e=W
zT!?1wpIaT@4u*ddM@wYsoof&Sw5t3*+2~;CC01~!$bXNG2!-#VI&Q)YGHm5u0@6|z
z5rJk9sRqOElj(cICLa)I0ajQU7ikbkB_A5QFvKP2R`ye>21~pn!WFl>?zlYw&ARL?
z^rEyQUH0;DxI8>bWX?(T6l@skp-cYuL=FshuS<8s_b?v{M5S!4{x<kTjNrWOyWWM9
z1B#}@JJ9f9xS!e_r6)!_a}k&3>m<gyy)y7!j0PS!LQZ~!50AdR2j(=iFTpK1Z2IJ2
zDU-MTU7>`}Jt>-Fpp$v|&`)PbR_pm;^RDHb9-6<qye#Uukkxti`n8hZR2j86m2Z&#
zK#usCY``4sXy1kB?f^U90g+11{U&HKn2wzNj+VchVI4vxNe+V#m(dbiHe8z9c+~0N
z<kX}x<S1~jUb)WZ3cBL0)w^u|US}qshg4Mg`<e$DEP@!obpx3ACUFw)4vSq*z@?SC
z_{|v0$>H8INx9j_!>P*u(`Wpk;`5x+fs@Se1tDGzg6ZL4J#(Ew&e!V<Ji`NHrw$3A
z)7DlM<HRQ%PXY={j_Z*EXXVZpFa4j+?5_vx)8pUwKD_hO6x_{di7Zlgf0N%lHoiVZ
zkZf?yWJmAK2kg%{RAVaBgFm|<oPWf*n#~(X(tzW9MVonz-Vp}q71a*C?+H`6bNkrI
z3!%#9h?X8Z3QKQ@G`aPBXd_5wZq<4|HU)iDz@`^B(Kit2&_IeG&9|8Rx3M~<&Zn1J
zHO<9d<j5y|7peBu;Mq>A@JU9t#!urMo_{H@{wmUh*R`Rl01{KUe(k>Y+ujHYzmFow
zlLR0sGVHgG5hs8ZcR#7pZDxEso>^%?QzsuuDX;{Z#T-7KYkzh}ZYm&PC6!Hr3*`%_
zgXx(!O~)NcI{ntxZXx#9h@xv0F%7(VOq*Kw97%ZK6u6YR5p^LXY{c!Xe3H+!YtY{u
z_C!8d%!I-)N^0jjk33)cZE*bxEyz|VjBDY^zgWFJG#>`=eTe`HM{*a14w~2yh*}Go
z)Ee0yUaz-HbfyD{?Zd@~;m;6XWAC_jr-pLyDX-5vcR5iIVT0F$LvY_KlE8rd?Ge9R
zBw*kdJ-WfJ@&3k*wW8)i^_H)Dpj?>2u)<t)B4s2)<NN%K#x~8sR+p*=03~jPF<F&A
zDdI*Qy{o@?6aIT~Q6RZG10v)VZc@$p6KFU@M8XMKAlb?&t_O4-i7FC_f*^x<_KFPH
zYf<kIQmX%Q;>ep?P9F^|=hK|1*-XhtyU$8`j6SA{V&AmV;R2C-jF>v4h#Sv`I8hs+
zsVjqcMpYUpZm))=3*pn);V5{3x^Mp+InGJ132O5aY<&kC*il??mL6VYM238K+BW)l
z&&PFTXhg>BO+AHMzbXV-Bc3a`oWIsC|Nc5?SQ*yH@M-RAv44E-?h}D1^ot%q=2#FL
zBZcm4-(pD}wPcV-S<dzqoFV}N15N=E(A|Z@YDib}sQV$2QMht`U@CQ5dc3K|hC#P=
zatA^|+%R96U0X-jg}`GfS+fwv%I!QFCZ(+~AqfhII)?K=R5|dNP7`aFsHDJ}GWV8O
zDMySoLY=lcE=)VEB^VLcUZ$SI9>UIf!L`~oT5nER;0VTAv1+qzug=gTYDeK4-)h6K
zlbFRtVE8XghNN*PU(3qesfMuf?y_+9(-yUXSy#pU)0lS$NA@Q80wI7fej$Wy=lS$;
z4;ATMP|?+7X4sT8n3<j|JCOF#57a3;-C+KS0q{k1o^mHY)+Q0o@ulixUt*(6tOh|>
z;^6f?R+)IpQ1?a6?I6LQu=yF;Qr%ei#FkCE%2x)nRSp?{iWr5Y37wR#ogRMM!dCfL
z9sc);nK7bMH3cJ=uk|+w`7ct&#H*%tIeoZ*vXQX}&UuiqAw9Ag{fdg)b`_5B*s7cx
zkn=K<yTLy}h9L(8@j}dHeBQK8gX{!@Ym{Pf`2lJf;acJj=TdRec06hu$i-w^=jByS
zN2kQY>H7{iLOqw|ThKqmuhI1T7Y=3D@cTrUR_xWLs60OAbf5f5>)EZ48iIv>AAbi+
zCI3cQGpkSQ{E4HSGKMvDW?q)&2cTmh(xTRo-C*3&c$&^){?M7sQAiz$sWI2>yW0*n
zUek^uj!fzXPo%WH-)$>Q>+?I_tSEcrZ{z&HzEZ4B{NXlZ&*fQL_-5B{vY!;OZjfdg
zuQy7G`V?WROMfEAal_bY8?ffZl6aBE3vhhi%;PMntQ^tvYui{E99mWV`FA(Ze{Jii
zj;avXBHn?AS!Mc^h#^%^7pcjs4$5^uI7M@=6|ue|6k&<j+3pBc&Wgq;-03^qL9>sh
z(}4Gy9P9}dfIE?5&PyaAVH7wS+2kB==sOM&)9Gwn2X)G+$c!h%@oq0wRq9;P@@NVt
z;8B*a8oM(eNu%)trUx-WXe&^K<OLIC4a=0m740d$dfk&);kf#=KK2F2A>Na|3pHS6
zX;S<Yr?`OY4t;YKOwW!mIpqgM@ZS6NlcBx_vXi_nt1zVbWF}ItM)F&`!hAsQN&%q!
zVKDyg@2Al|TF?WnlCC?IG@dm7#I!?k_fa3(q5T>2eJSDO$RVh!*zh*tQQG;G4TdWY
zND&B`HM$y>S6`P#y+>IJW+w9GOMSn8-ZyzOs5JonnnN8~IHnnIIxq5*hCW=qScK)<
z3YdHW{P#!RP`<g+6Cs|ck2cL_Zrc_JWYs+NM~_~Af1l$M@0TkdxzU$KBP^Z@!g+3U
zv7x;?JgPRN{!FYD(>b@LkpAtiHAJrJiVg@76cO4`+GyRW)rvh;r8~w&2x7h2fRRhy
zAi@Q%Nbpy`$iZJn1Z)Y*hdGW+H9uv-@AKkqkvVS-J>3fF;tkI9E83c+zwDfXw~bak
zZQDEadf7;R!Ujo+xP+;6fegw~aIT`ln;G8C=W~j4kI%?M<fL@^UnIv>)69H@)^*SZ
z91P|lC8$4DIXmXax608n0s}q2sO+q?wj>W<4e*B$8RlEi!nYcf#m8iS$Y$IinplFk
z+3&%FZst#Gsz&_+_lhg%c<6FgI2|9hDwCvbT%0&Qf3$BBGLjIbwmLob)p|O@bckdS
zsqfXBrK@ae7fp7|F1TO-u0mf&7NYM^6;6zDX3~hr&7Z)DIpbu8?&UF#_)ABH|DD4R
zU>GzAh7ogr;7|L(v&5<>JpKnFDd1F&KJ3e`ff+=ixseotyP!v^lW#dXdQu3*b(EI)
z1!6fw%1eH5)1Z~i6v5iioc9x!KqfGM^u;}HS*UeSbz*OINX=xAYlou+*qt~;Y3I&2
z-0c%kBKqjw=!rQDvS!Lhy1;Q$?1S?59Sk}3&+55^sN7v~my0OGaS7tC;aJ8FND>ap
zrRnmFJ=c8nizLujMY(uwBGKhwCt-q;I9wjg)a9<@pMpbH+vl2q0D7O=hxW3*@0H^{
zAy}m{L)4s%Z5b2x6|9H$vPtO-YEyZa>XLb*a2n@%?>#5MTqw0*!w;j)kt%oZTnewo
zX1#I%$*nfr<?It{Exm1f(6F!}flV<CWCsSKP(hM3cc_71j}N<pw}g!INgWO8QmPOr
z5S#D#;)WFumjY%`{+?r>ox}h~gjD?o*D}*hzLmpC_+9ArfQufN(mq5JS;dg3D&cs3
zOUyXjpG01C`0<sf_F3`4nAV)?{z#1*=Z~U}P91x3QnGFX0Ww&Ls1s{{qR$E^UOTgG
zGSP_e;sh{Q42GsaG`2ls83Z~IpuC~!OIn>jt9V%)*Z)DDgW;X>Q3(m-_^Tj9*Ls1v
zKY2@;#XT+}d9fXLMnjI;r_P0w_?20x44$}>KL>wn$z9T#W^SXh@TdLj8R{7>f^5@6
zmALE_{Mv$Xwi0<JIw*FF#k|G9+GSG_C*dnK=F(sY*%QUe*WbmAuET-yD-fNsdY?BW
zTLOP*QN#8zY=CQI)f64_t`VMx&ahuUp*Rm(v5D$LiTg8?$~+fcyx8oo3TmnwO)2+q
z-+omuw4nK0;{#%hcl&Qp94CWyaRJY2fY7yL8YQs`g3shZ!M&#<sZ{vbhkqm%^jHmV
zd(~J~HLW3N73QGsw&SkkjH7qRNbZ-vHD*KynJL%G)n@!dD6g(S0_K86_5^B|0wv$@
zwMB^pUg3lct^|8_qv>vY?R0`tU<CIb_KAadkp30wIZh|>x(mg;jvd(5%SqPFa(fqO
zB$a3c8bI^OZ<Lmk<v7dfUrfExfjiSdg5G;JOUkVfPDdqZB-LUcn>lv9a_+ILspp`a
zq_SF+;d;WXD$RQ(nu}fmMa}(^9YIBfF~gg>>A#GNbL_32)k`^~Pza_fjo;($5qR6A
zBY%Mgtm_jPG+G%1S6#lib9;x=VFU)6H|*I~QYkdEJ{5%s#@q85i6`4<B0&ck!5yKD
z9SypspVAL-hu-ojP;s2PG+4u(u`H!w(s#-9F}5?W%<bhOBv*MPej2t{%(GW)x9<=t
zy30Q}Hy}wXayMSL-^wc4-m3FtDP5F6LT$^>i)#Z$gTRHXW~I3*#S{@1FMs9D$f=L@
z7w0T0ckO$)p^5*wTiX3ScCj(fmh$JKM4!L5$jyiCf;3i*;G@Nzz+N^jI*~}>G)@ij
zp!Z_x_J;1&G}4cghqi6BiI{SSvA2~X$fx}KvAyL1ABUX?MfQqK_Q3`AQ}Uec_G#jd
z`bx-pwcV&37FD)vJMI|#)}F9&QarmZ@`=&~*AZ>^r2uI<Rq=!y!b^^0pR&}{ZGQNt
zNF-lnhNQSFKc&mN?rFqNcK5Q=Bq}+H7{?cJn?%vk!WYi(lhr*pydZa}$>^)5yq9t@
z5Vh8+obFq7Y!j4X+(w-1t6MhWYrCC4HguUb)iq@JN3YA!v^4Ehyz?P<w@m4iUmGxW
zXv+0Y9r=|TAoaHt(eZ~UzT5wF{pMyTHY8$iD_zOyz!u_2%;?%0Nj>Q#T752qgkVy<
zgYd4ruMsp8vuCl4YDGqw(mot)H!*S?POq<cNn%fiQUXEOE+s}GET)t0%WJGrv3)Q`
z#E^#)LIWNHfnxVj0G=Ik0`J*l8VO3vu5~uCeGfbgi}UO|ya?$l+5PHUVS%4@zR}%N
zMN3<e;hzhLnx^mf9}={>g5<#z-OVcXd5F3J6g;8X15~AP@uw0puf3g%vmER$hel#6
zo+B%LA!4BuivgeGx#o?SbT^Of3%CvyhYMU+E}M&uQT);(uGKT3la*(2EMv<#gc4+N
zrN8NP?V(agB;8GcIlWyY9DmAbx{`uvFaKtOMY%$gPU(c!qw0}+DD<cVmNlZVN2wv+
znSERiO7H~2_t_Gs(~kO^+x;$0+${^9RkqPs6iz)xxIBYBOgY{e<5~I4E~<Dk=2y}3
zSeBc}m`{Ytl>f^zN))61*1h+19p2UK@uj?(dt?VUU6cCQ%;yi-W;YtLwylMD;?&fJ
zl<uDfghP4wN@2if?Um^G<Th!lal?68>KCulCDzwJdCv=FezNn`a9J75Wcxd#Uv$v0
z%C~(Y6l#Azz@gp;g;GEva#bWZ?S#@|Ddi>059|+~v_9~lc;I@J>!-k3o)7&ecs{7y
zkNmE5?pB<@gA<gz>=9g~7;Ust?cQ`|jcfH;*LqpYU&F4wWdSum)MuysD=NI*hK(gO
zm#qF2Sg;d^gw=xH`Wm8!oC7lh_u@^^VMhr<$WZAA?!@0B65-s=l?C~xWaBa2K4`#&
zm_%*xNzj^=0=Ut%`diQ)I69Xg;t!zZg}mJL2REfz=qrnm5}+D>w0^Fs-|Ck+uxcm)
zca=Ks1bafu;#d=BjEoi714qsQr4F$xr`75m+V3f=wnen6Nnn8E-cC>L57<A%dhD&0
zts!m;a^RcJ1dR#7w0jlWJsGc^A3_0q!$t#p;4V)$UeIt<?eg>vcw)|`?C)INF$OaR
zXZUd?z!uKFB{zuJI#>hYsHP1d4+z5QRT?aEZ5#n>-4P@%6tT^vVoUY}E$;wijyR=1
z(kp~6xsMx)!bI?vaz?UFA=#`65Q5T-fM5!4UOkrWzAw;nVp_Kfqbg{-&m!c&KF|x=
zv-z!P_t_+=WgV#ks11>Q^}~a;f*a1`OS#ieNYp9<yN!WK*J@;gctXcActm;@z3c>z
zRHZPc5NyJ47K5y&4<-!EF|yz?GYcjsMZmZjeEIBFi?8nV^)ko7#*3feqr^AwLll@(
zJq3?Ia&G^8n5~9w`GKLeRMSK+V5B+2TPg<k8eU)1T)szbkU9E3qik4ae|Ww6YT2;&
zYf~7jtN|8ObxW%%3gEIqc-l4nP!e5*M%k=N?c!Mv*wak@<KW&vJ$Nm>6G=`h1xnH&
z;IGz*BSx3;5%7DF22T7eKrz-S%ekhNS?k~QoCyK}z5K3m?({XA(7yM+v)a*`bMYN#
z@_=N(ero@PkSs?7d3uu@=PE$}?}TKXNj(|44Ms7MB|o56EJK`j0<uO*r3)yArM<Lj
zh4LUeQWAd80N8vTUAya-zn8z3&%LTYb!R<Dv}f%4R-H6<J_wNzWC<7LLfKvDvA6kC
zKiWD40MP4Sq^&$2c#`jst+n~sq%OUZ-~9mnJRKA8`gR=MSege5UYo59wr#z1{cndD
zFv<}AOwtsKy7Bz3Li-<nZb570$3V+mI|Lo@Y;NT*QMu!sXv765wbPqeq2VAD$Cl*O
z=sObpNGpq<0j3!B`PsNf@L~8v?eW(VXhScOV=%g_fJMFdqy@P+=hc_*%I|QHxWF8(
zz;Djcya<fZWk_W3&(ULFXbws^`bhsoFR?3p4{_H+ALsC9fPAJLNc*Ewq9z}QJuP*<
zJ1j6SuMlip8do9MOSvuG7heMvHI?=x;w*hO2K6IxWap(ID04Zso>_)5p8UDQRK<mQ
z3ko1b+yfW$Qa@NuePF0)$cYkoS9<#bHy7fz>I`e56r2=>(D59_a12}x5ju<<0tc@~
zm*^PLs!C+NW?3Y0w*juGf#kIV&6RV?uZU|A3jq#`%-&L9{x=}rI2FsxL4srUO?i3s
z%o|{yb{!1A?Q}cT%vV?7nkQo43)<WTO6MA%4KD3o#^FT++0mQTescp?_J*A*uO#w*
z99eB)RzjVD6e=IUY(<Yr)NGjVT+v2(fz8O>fGIVS7O!_@y`+I|%$1eU7+cHQ?J7sM
z8<8I%u?HlNut-M2EL6GDvW?oEwiEnS%4T~A#rSGAMk67LKENg)%m=Q7Q_h%<MCVN@
z)tKgKNx^jQuND~p`npOHAT2XSijwnx47QB8E*(RGZEZXP_LK1A-f~`JOixBF(AR7z
zkw3w6tpYkIojWrl5B4W2_`bp!yxg69DogF6B|!DV79#CKb~#Wxi!lMr^$oz-muiqj
z=9CzAs{^<I#AIR)P*8G=dH^LVe9ILEClc?p*O{H4f$!`b(w@Z{4n^D|V-(AEX<$9a
z8yx33e4D4?5{`=NIJ)B{o*d2*={xt+%)U3(okG6bbAU+gOHX`;W6chbv}=8I9VZIP
z5D?iP&IiXtIPw(LKQ8KBe2vY#;9kO+{QL+7YCndnUZ2Do*XUS+j1M@BuB~o)<jvqv
zZvdRcxm|2fg7J@km-5kbu5i#&Ssvf3G<4|Ao&NGA%lXNVd2icj4^EU#)y5~!YHMKV
zYVGulw-o4`-9ABBJ93bSeNIUs;u{@CYD}qzose;~?jd>Lh8ouUv8neDKlT~@SB^K!
zc6oJC?<?-kJJFq`t;2S(Vl3&Eav39L%UI02BK)Oi?*4s$%R_-Cy7W&k>IL+Tx^qk2
zPFJWrL~s=_?0Tp?k}35I`{uh4`v+Z=ezecEkKc*Qzq2{=HxCRgsCa(Rtv<1T=Icri
z_*Td#&+`Q)nmv)9inV#5T+zaoRXI1-Bo1M$-1Pe=V!ZU%k00^=Uiy8?kSSScJXASY
zE!m?iHA|+&>Q<fItpd$k7|SjxlkO(~02Y4NW_8S6c5C~Pitb_l8Te{QU8rhBXUX2<
zgRUsIR-4JX6uonUS2@1;*QVwXGcZchex<y3ZOUUHyse_PV087=V|CUa1rxrErAb#R
zAP_qpFmjprioTQOUILb`r$jXn{J-y+vhu~`d*YSp?IU{5mD)<6chSsWd~XzWDpP8h
z>SNZuL(;!rAb<xEMozh%D*nDd@7>)Mw*0m00e4K`(|GwR9T_Xfl@F;kC12B!kU>jV
z)-QSOK1a@c3;fPk`!K?F>uo9A1NxS)CAA;dq*7H0D=%iA%Fn)=cB{2f_XFo!y|p4L
z`deZ09@-vD_>%t7EOW2&1tF(9BO+G@N+^aUWKKGdnWmZD)V8RAr~zwMZ|_a-CvNW9
zPlm}V3?n{*(<c5cJ?%#rjVD4H!V?HsU^+H}DWcqop|VO^#YgwQsHk}=zb-GF;UgE2
z&r&(9RFgCxQTn$KUoW{qo`lNWKZw2dn$MF_`5i2)fGsR2<x9eITI1sy9RRI(>V;>u
zcYSf%zO72nFXgcx!7<l0|JXyLyty*b?-2XpRaUlU8G`9b$rI8*aq;cXL$l|l>rpYe
z4<UBt!L*<KI{Ez%sgEUW>cWwZr|Df6cW>3FF5c!%WX^_syLMxNA{@vLEt12EuDEqh
zL0P3n?j=KI^=#>K@`qHOFy5<xm^<@Uv`&B6FFpC7W#r+=b!C}4zoV!<HrWObL>hrl
z-ae{@y}X2fdV6-aa3~!0nX@h(%;cmvYOixw`$M@=Uy}!2sbEPJ@@+XtgEz(R-&Q3C
zHI@eykV<hV{z6@l<l?N)h5kNV&FI0R=(R%B0xT`{<Yc+U!Hk7mVc5lYzg2ZeB(ct?
zuLoW4857;oeu+X?o>zY_-3(x&p@huahhAJ=;xMTVHDDd-*LP^{T_KG8di_7NK?jhj
zQczmn)j!M*6%tAC6UD%UdqqlbBJ2kx&SG9ND#l*%8BaZ$?S|Hw(NMhEW7$u-i;c<~
z5<Lc-98&!LZ%!O#v7#X)&cSk-f{)*FJ<42b=825ECGH#Im~)gZrzkihNX3nO8UOh-
zaQ<H+sM>(;PGBSh&y4b#T3W8`T3_mfmxvUr0R9G2Ql=r*6BhS<v$(md?j?)oSCQK-
zb!dZ2^1=5+DOa_HFrynP$*Vyl*e<*F6Neb;$+d1jfw}6-gbY)>aQGO0qq&QZM=Qrf
zm!-oS!lDz<k;fW6ZYUp~aBvK1h6rjCS9vmfF-1O-3plk9TfJ}XZ;%mBxQb2AV-VS`
z%s@~yb>azvt!x2ZFMB8lG5fQKbmPHK(;o@?-U%b((3|LEjYahOx?5RGZX)t^Otam(
zr}bcWUHp%^@Hxa@tq(mV9FUtV8&@YfqqbPaQlRhQlH3dwcG_JAS#P&=w%f)Ze~r>>
z$GuXCnh6)yJEl6iZ!tP`lQZG?Ai3_+9lsq0#n)gj#OZ6ic)Rr}6pg0|jvh2D68m4r
zBs)3^Z*sP9S3k`N>o1rso8&3-p5%|^UCi9j$EHl|A|6_F{xK3(kWPfhJM~xfyOen-
z$!-?34t&iTnb;HEVOc%=ow<>%a~;OzLQxDN*wSU|bK9al<ap(1W<$?i%Oq2%-|F}G
z7w*bOig1pd7~^q5M!y;`p}q|!%4(?i5-}z<|G`Cx71|frHFfdxTC!%Gbk_buH{MZE
zjGX)4RH!EOmfFVe_adNQ8r$I*QT>2C5J|)MQgM$hdp%mke6ktT%40)e>s7zGAW}Mt
z&qsmhnw=3Cr;Yg?8!-_W8n}|TF=tg%{I~6V)kO>D;}iR@^6vdP@l%MQS;)BT0Wrh_
zfF!!G?tU=&N+)OrVFpnFo1)jcd0j1oHpvX6VSe03dTpZPc`IjE{Hy&R)Tr5J`MrYM
zT=K3R!8uSyPAKv$ujEhEb8N!|-0D=PF#W4Ou$Ba}nr~sb3dcUV>x)O_<c%iJ>{v-Z
z$Pg^PwpZ;1^&tj)%kKj`pn^&WSSf@2(B_?x!g34j^i85WNl;BpSUt${){>qRhX=ap
zRPsgdwbU<y@^W|3m1da-aUH#ox6x{%K;M4zCZXJ`%>-x3n%UOp`6BZ6MfK&C^1515
z@hhut6$j4K8zw0Ch|!xSD7$7|mVi~<566n*sbjC19ZTWE>NmvK2EYtEa@$^BMmGKV
z5ajljet)2LP)n=EME4OT8VNcrLMGTUECri)eU`~R4qm0l9i*l|^E5)JmDJ#7?#rX{
z-xT>=Z$2m|7moJ_Z@YUUw}_PcNSsGEA%S~%PNEsCdLHCo`Q=HZ*S0wu7(qL$EWZ_N
z`4gBd{=y602@}CbU2kfeOay<-e+nD{q1kp$Y=Ux;_*U{GQK!TT`}xl=&bKLD@c`U-
z5aCwDSX95cBsQCPdpz@Ak+Q!3$NV|qYjm5CT+Vz*ML*UgqUFi7d`5bTI5plek~Ur7
zvEgo`d2&e3IK2_yDNzx6v8NHbV@rq);G4Oc{k_L+jkJ0O9J7+<Y}Vf}sk6}aiN_c@
zoZ<pjU83ODv{m0l(^D5EoQPA#iO;%nWS8U5m_NELerb6!YGTUOIsC+9d;lwBw@mu=
z4PD<R7}Qyu$`z--&}@>O^*l5QNvdhOpJh+&Z+(}=BOjrTWO>lX-O<XC+8ANkp&HjR
zF7R=RU98>z>@jgd?wxK*^V`&b&CV!OZ7s%94cVbB>1l@dXIs98o~T{|8VTN=w6fl#
z@WCcu%w_wJLN&hGMcH*v+phMO-xrxpl%P*s;ms?!AQrPYeu`DRrv|KLdu4WgJd7P`
zh6YHeZpL2Uuo{RM@gOqlJ5e1jfj|RPgEpq$lepS5v!f5MY%BD{T<;yz+^+MnT#KHx
z+B1;erRHjFtnCmVe*LYz0O+GtD=MWD^Z2j%(9<;gu*@f%TBDTzAWhn9o!BZreA<BW
z-n`vc^DxuZ;E|U?|3p0gLlG~`0DPV^3>Q?L|Cq27@zmbEgkvF>T_Rl^1p38O`D54u
z#MWw>0V@;p%__O#bTp^U67OvUL3pY>cF2;9?Jj0!A!71gOeO2)iM_>+G6Qt~?Kp+y
z_cy}ML{ya1-~0qhaYux6ItKzhj_ENo92yuy7MIV<qtCM^+o!YN!U@L-B^UOf4`U-0
zZN7EPD6Zf?w{~?tRemd`rD`v~-QPYY`Dr=mSxlepv8~|Gg=Xix#xY5Xtp@YHiw2Uc
z#kATd6q(|Ejy*%4@=jw}o|K)?_PtluBxfZMk2N?S_%w=s_14$a6Jdd9?xJoC+U^k+
z8_I_joDR0-#uAQ%CgV?|8B3vqLt)Jn4*$^ZJoyEy2hf76GhA2^){!ET?@A2o)}9{N
zd{8jkRGQhIDj4SHO`@^P8~8}s>+<<!m8gl>9zxqAn*AZ(ccmC#Y^c)@7x#GA8wrpP
zoJwk2D6kf_>p>bANfs=F^N2b`eAMeFWpF8&;mJ<8c|5De9>P6SJ(m{wo}uOuEv{^V
zmo9le_Cy<}MS}9#h_k`YU2+ZW<YX8jj@mY-Tm$9P+pVMx_xxx>ZSd8{X>o<Dw<Wfs
z$(vIx!)gMF<a~e~79~Nnc%{J2TEF{Bn6s{yq_Xs?XZJbAjI%9LW7_k3d6@lOp|7%3
zK0g@z&QHcavYVkU4Od8%pd#WSTnaS?8B2nMBj|u;=!q=KB+u=8-BsX<+Mqf&4&}Z}
zAKQ4Ve2-YqcUZYYh=FlS!h-_7anI)?WQB5Wm2{p1)D3lSIqe*ItEoD~<f7^;8LX}i
zcP+fwIMYMF#~UJYvYOM5xE8U-Ls&arPsWBqJtcQeczo~~UKo8h6MdD@L#dK%z@_UT
z?Z1#%|BX$-a2VJj%!gCcB>$l0)CN$ne^Ru6B7`qT(|xDxFyHu`D1>`>o>@o0w{;i7
zT<Nj0gdZF`=Y4CNEwx6O*cKz!zprDSqb*}%dT`5}iuBrncwRjD99&1#9E|H;E!}7`
zyLHZm(Wg1buo%18Z%e()t>(Ma`k38AlEq9n(NHOCu;ae5HR_gK7+hNSEYcX9I$u@a
z?1}wlqxjalw_EyxG3IhL_!e^LjHSv5$kN($6)1cTwKx)VEcP?{z-PIJbKOC__jq-3
zitnm8$H*+H+z1g`=bX0LL8E$T=)TNQHY2MNS9?Tkk$4?kePW)XVy$m{phQj4$Fj^k
zy6()>A-LO)<*{x5wpko<n>f)MsvvG}p0Q3nlEBEDW()&t0?U3*`l$C+eb!<HLawC@
z97oSFNd|fG@;6rBih9MIa}+hkvp~OksCPV%K|a*2k4`QGvOZRFY7XsDunp_KpeoKZ
ze8GmlC9`{bgMEUFxCv&3&#}?jo&l#l9)0P}alX2`Fqm2Mahd^^SB3tuV!B<T>ci&A
zk6>opV6ai0ou52@HCk$#R~nGf!guMrk1c7$4^{>eSt<C^FSB$=E+?#U@;ZJO`Z(RH
zP!076ueIQ>Vzv!6T*Wb8rg^J=Ha{Cz(m^YYk1fz?9V4zCleRKL#hZR~>5yCgxio0o
zPwRdpZRFFs>fnDk;r;<ueEx`-l>F$>9{eZ4(c(VMdy+t$8Rs=LF^*vLg~s*BQf_of
z8_y6YUyYauR+~QjSO2v`-wNm=0R*uQL^*_oCn$53Y5>{cu#9#LI%q$In4wblA05-b
zEz#fqIO&H(X6!L!-udVI6Sk*dH{%fG#xRBtUs!noSl`PBk8gbTm<`An#Xa5LTd!f8
z%9xVWZiC_LtCtSf_U|rayvBYOBEfShpIn3bScXtO3|9;0Ly0--%Zp0S6`pI8)YjYn
z#qa#Tc(cK~nZzJ(<-I}%<6qv2<WcsyZ-8N&eJO#4;oU^Mb##{Xz?3O*nsgO38vMix
zM)%Ix0OjJ>(1q8wSYbT*{oGx^4;@alaPw)YzS#a2LVX-_25?gIK?tZN(~jxxu2Vr(
zRs!}Lb*usS^A9wDY)}e#kdi)vCl7+*`7G1OMw7*C!hpcvflClSMuBSH;?N`euZ^qs
zquR>(YDZXA=gITzIj2f_66~=E3DpV0;0u6M(pwFjx%F$yUvdjVQN#+f2*b7v`0Xt~
z7iMMxU@+AaVa=F=Td!mt2~FgQsw=1exi<cd`3irum<898Z+-s=ApI)=u$CPgxG)V(
zNg0qSi#w<JUdGp>O*x1!V}eAKz5&;&6mY0l6E_h^6TuxN{>K1>(+Ja+Y5cj~Kc=w%
zZ!Z*(bUi1^C#U`zg7F{M=YRgopkc&Y-sS(0z~GEfwgvwW2@D}12dEM`a_!IjkN>v^
z5J+B(1jd%$n6mo2YvezF`rkIsBo(q-n*I+F5OlD72md7k!r}&c17ONIn12+4TD24Q
z!%|&`eMuo-P)DkA^Cq;PB~$KFf0Ih`j}r}EEYu*O9Q#QHdRl*Mkmr0sh}-rW;L5|p
zzRMch09+uUaYe8;N4S%{#?be{kN54=u2jD<2`FbY|C~M)Nq{>|q{J^66BQwGg^zWz
zyt(IUUtVqn_B($i3^8ZT_qX)FAyOEp7kB8UY1&{=Z`axdec|(VNB1D|V-#OO<{Ty*
zzm((R_YsQ6^05;GY-uF^n;*@OSD!XV#!Nu{ea|*+zm4=XFc5ojYu?B)hj#wpqXVW+
z`I=whL|cO?3g@+_P&he*<;pza!jKd<xm!h8RCwvsUrw#RK84?3FVtK_p2e3~{hB|X
z1-mjzy91gDCm8BQQ~`{}F`NLStnouQV_a+{6wl`m$5(13%Bp_@bNymqgSJM?t{K2Y
z<U4~S-<m4~wDuw6L9Yp5<FDg!_&^(AVp(UiRP8*6i3Bo4)As>q-m#G;CKo3pfi^|s
z!u^PaF=z^AmyJw6%aZ`%n=cXQfzIIDKb1mOd}#*w)%J)ZHzEKOG}+x3I=fJN04a=%
zkVzaJG7cKM$G^A#fv9hq^znhubN)c7R%UeoqY3-;UjVv%EFJOpk^U`ZL25`r-5(CK
z2DXD1K8Vw)hQyc4NqYM3Z1b#eG=d`okz^G_)b)Y2kD+^%hj)y^D@Ow!*{SXk_a%X$
zzg4n?JcZ2{a(kuQ<B!e58(y0iu<JMoK)GUfAux>l)WzCt$Wg!@@4_=ErHPVUe5fI5
zsH$~YVproEuoVvxeS==n5)5xI1<2>>?8EBmML&ef588>(7eGsdN#=fbqrSy`|05HA
z1J+V0+KQtShw_4X=xVqP+<PSm+z<1(-5gk7=Qc^uMF_?IxA_@1LO^HlfKm1$^cF=t
z$zFsLi68V;NfZcg$Cv<4I&K~C{9;}{bH?72T@x${8S3R+uH4_-X2P5Y2JrWm0?pI-
zL%zTr4y{EGZiANW*jkr^-{v+R<pmb2eaOWEUh!`Qp>Otp27E%O!-M+hq#M;t8h_GO
zrDk%v&)?n@!oVN9AsmxD@TW-h1f~!6#$49x+<W*-0C75Lq{+*zo%C~5&2PRW0dL6$
z->PSJ+t%2B^GEY%T}L)om?RH?|0*Amg<!%N#jTNf@Q)M>FV}<w12hV&pv$;S0$2i7
zBFv%YKmt<koL_g)x|(R;Zw#>=kKI{vFdTGfh0(Ab*y%S09zyxtv6cRMu&X4^YZxB(
zMGVSrm)nXffcMf%l<@1s5oFj4d3~A`sO3us>pY}kp$tv+z(P^~9~Me1vQTbzcia8B
zP@dL$Rxc;1^j|#2L5tyi2>;yXuvX?p*z9sI1yN#qB2+sEO;B&X=o$CJK-wk55Y@*o
zFvz%H=avIM%Axq6r}4HA)lAO_)%A*NRe$0|f4y%1d|j*wTfWC|>f#^2d-5zXB%nT(
zg|w43;8gbm1Y~6nIu9p6c}l@%cmT+iS7C&z`A(9P{ew>UgOfL4XJ#@9+9Pv(MC)cL
zJCw{$`+c$_HI<spf5Q@Yk11eh(HXRT`54X%xQ3kTDIG1lzs;cllaUxeeWc)yXCDv-
zT(?0oQy&z8J;f&pMi7Ciw$`D{rWBe3QZ$*S5dY69dM&W>EfB-sV6s7|eooIKQ^+=;
zu!+3Ht)*6nb0=iHtN>zehe}<d(AVY~y+g6A;Iz!*1ELaa3yt+M)Dah?CK>JGIk4hi
ze6AN&G{>RtuRK{GAo-66DVsvhDqQB%Q4<YAPK$*TgdfmP%>fVF<K68g7+@>QMmEE$
zc0y8Bf*T=f!9J}8ZiMheciB%jg%opYfKgyc0|U;goAS=s6Su?`Np?V9SDtk>5_o5&
zAbDb}$%zX}FmHVY!>aVo24uQ&22s<j0U{@|ck}3rqv~*k)ybtsdW_E(hPjEhGs-um
zh^MEa<Y4aW40#qGv$A)_&QB?e)fkt*X`@O`q1ZzlhE;xSKTO$4yDsRrIqOTJ(1gFO
zc=@!`A@g_7YA+-{(wxhpH=x^A633B*XQ%c}$nP!2Ec7x`u1yI!o=fcvpKs5r%ckAP
z@x2Wb&{U270z<o}jZg``bSj&2;UQMo;oLiz(pXuC$gw<Xx9VJ*ZAcV)ESo~lHa~02
zOZTgai5o3TtarX4Dz97~Dq&Ulf%~<IMx{ZGs?W1W-u_(D%>y5$nzk|SMOuDo^V~+*
z`QsZ*!&i>eI!PG^c;ztyEd+^d4~wywb7<6eFA9^mrf=T7EKs@X*<aLRAYWLq@~1W8
zKkM$l++bBIu#C-^X-ic9oT0J@e)|R`cIFUJlQ;~<>2Mmz@#%~licrPM>qIP#zBSvd
zqxMH@r=3}c>=TBuJn#RHz4r=hGVRyC1A+t*NdyVf2}%>`MJ4n=P!UABfHdh%M0yjX
zgc?CXihu&rrGp?<q$)_2s&qq_jv(#3?s;ZrJ<qq+n&aJkH{P8&9FG}9a^KhWFX#C?
z*)rTmEKmjjld+`TyIle*(E*_H)!OfJ@}r%;EEBmIlT1^Ae~X~@UG<(Q1Azo3OIG6f
zu&D_k_*nix!B*~nxMa-nUQv~j7HbBtK>RuYv*iK<>Y?v|+~fwJMB3PEcpJp3MmGrU
zFb4}hHJ|b#gAe{R=vMiGCu=8qDUTU6_1I$plre4@+==B7*3r=@Gvp8`EQ~zKetmQT
zhid_td#*<qAD|cg!O42Mgmi|4CWr%m14?oo1G(x7k-Y_}2i3wCVc^XWT>+`z(Uq|~
zyTC}Aga3(WX7!qXy;C!B_@xjaU?;;FNMn|v<io&hpX3o`RSyuKgiX>SODUdYn+cbG
zs^bam?7@&Y-^0!#t=bbzt+ZQ%yphHxH@~UdDq_)5cVL|HP-kkUc6$N3?Tb4gjU}RE
zNBiVMzbVZGL41DkYe>ZEx1Dc10@;O*P93e?k7kdy@GAh$5)Q}Qp(4ja7FV}x-2k4H
z_UUCaj=k|Mce6r*3xsd+9avJ7Y-0t7NWw26#mrziOF-|=F|aU}VIpC#d<3jMV?Zf?
zHeIq6zJI@vGaj1%I)M5r&j9q-U#a*vQ_?4(30RW=vd;8Hv8i7P{fC5YxPSXF{Y}Yw
zS%W90(YveQUCzln0=sQ0*;e_4nKn@He9p6DTq(Z<Yi=vqDL%=r)A|Jgo6bi(iB(J3
zwo%Q5wa!5)ydlRK6OmM_;*pe<*MF}l(d%bA4t-u5bG_s^#U7|ONNXP~EDb*BJZuVl
z31khdu7xEEHUM4J;$UP~$<gEz&~$wDAiO$`q*|hJ3sZ+;j}$v~i_ZG2K*2>zGzzlz
zamM4^9z~|#r@j{^J}w>;FSu88_o_}-!;Ggz0o&VhgQ0&a>l@|Hf}50mVQJ*g7pI4V
zBwd^Z#9VR=D$ps5F*I|->G6YsD#7hn;U{;%L~YmOo)4V@PmtwvXfG>6O8Y%e4?&L3
zStJDF$&qv#4wh-YZk$cXz!T|0$U_QYQjG*ZP-cX}-pY<Pw*3a>;y3V!ae%4ub~+0d
zGkrpaA36378p;owMYK@-GGp3h_y2tP-7t7zm>rBkV1zwjv_#E6zzU!PY1sCV)GQ!;
zfn%3%@Po{B0N69Nqyh^|Ml>k#q+00$P9Y!fvLAtZ6C0!ulm=Gb*Kdi#Pa>)<$en+%
zdM|58mYTcWI{F2&%uD|mTnA`nC*BgK#|tyiEMl1lY|m?wHiCu3RkFvR^1KJX5D$_I
z3;>x=Au?b$QW%O;Z~*II>erVfLOjfX#X|gdis0}?vzH(-Tqu2o;~FsKljed_Q#SpI
za9e^mj@cQT8Qe9|En!|U;kU1ZRUVb!Qc34Fqs6XZb2<*N7v2GN661;h{W8Hb93Q;_
zI3yzwu!vL{%kCQ&@X{C`jL<<QB86#um7WGp#*{$r$@D#^hQrDjFXsJGFc45&)1%9L
zKDGKjaSq;qR|x@r9fbmw*|%n|zk?@G_xt?I7Q3<U+~KHc<7TM?vj(Vi{cihvV+0OM
z?1$Ij%{n|Lz1J;<86-mGwn4p@&-9~|sRmTX=f%kSK$ckr2P+pNBeO}<H?!LHL-?zN
z(3fq>n{Nyz(%yb34-n7hIcYvFMBu^7n;*)F4d1ZA!SL;8N*I~(0;edB@Q;5{4zdz2
zZ*zFHq=~ZWGkp@MZJRV-|B81l(L{e;$DNL(u)BbEB8~%Vf)=OGrH@YT{oHzxZmi_X
znp&{2d0oe;)%(n$m&~lcSk+6{aP7BrZ#rTJt`D*ixfN{CTcy$KUjkNVuU$DuwJHi#
z!vh4pr(TFO+K$*?GO1lxHpZL0&ipdtHK)EbII`%McvKj<o$=Le>Dr$)&Hvb-GTbIT
zno{^L1qTAT#v{<dIeXzaAsI7mutE5s{nkA|WrI-Dk@4&zs{110hOTS<_!r=`iouwf
z)M+02b;>Ssqsj&_ZJe{2pbSeh^t`<NIxCQ##zQN0maXAggjXz1Ca?OlESCG&Bh`>h
z-zEBl!0=50+ZzH2*9B5<9?3T>UPB37K<QpUX<a}uauF=GTvJ(6m(azt==MiI)W=Ow
z&~~LZKgCUWN{=yQxZ_&i8=ZYa6&n6$^ozF{y^J?3PB46==W<5MB|y1JXOveUAzY$n
zI4foB+WSHBYwP@kGjWEs2-n6hoarsj1h?mgowRL<IDyHV0?WCc@ezixRQF=3c4KkI
zc~@7_k&l3Qu9X?Eo1|$glLQ4iRRlh%u(AkMvtOuA5c9uQjkw^+I5HbU`3&cw3Xv_o
z?kP*96f@87{ZNQAQI|C;j>kAy-1=qsC?0di;vvU_ht39H_eu|p-fg~Au71GvdpdA7
zPa=AOg+Thk@b<TLbLkD;U5w9}qszt#*{pB)m;3tTi0ax7?smF$jb~*SW3>B$armX-
zvhG(|x3!M~XU^0-l6ULc6Y-Ia6Jgy=w`}2@yoe%^sL5P3l&U`l%FSqZaw4~km~n!C
ztzuP<m=AiXgeq_2#q-LQ-0Y7(4TBOPTEr|sY`@n}GrW*yh)6Kxe;ICf<DFm4qN&j`
zL}8oa_MoxT!h}0*%r{1<aidh*n1QU(o7PURzBqN4?Q!#=z}DxqaEHMw<2XZ}L`Pud
zne8RZFPvg&X~|jumN>OQ@uPQozwM|m$z;bJIvyD5FX1k&42b`ah0CA62~>gjtyh1?
zjDZQ^n(1JM#6`jx%}rCu3F;sTR@VY`Q<7h&Ghsp#mNNFopdg5%+4tEZK9LQ4!Uhza
zd!KK7R~{zjo>7F&=i=>}4KbY=I~NE&I=`y%78u4m8!8&E-7ZfFJ-~jWn)*}Blddjl
z33KP}#(+-;r_k4zpV`)O<U2Bi-hc-CihtFVn-uz~*~RknC1`Y{>Ljt3tRI|QMz!S&
ze%$F5S?46KCITpP;4D@8jpZDk_O*E&hfjy`Qp~ghKI}?U|8?kd<jjL|g__*L(R{xn
zP1-o&I^Um7d>EAeZBm{uOa8dUDtcirjuIvvXbETAT<LJ~&TJF|5*aS_Un#OLKW(^w
z|Kh^<l`36)v^R=9WFPr*3bz<0e6?Fo$QW~xzs5IS?O;c)EVB!qgmmIO<DjJ%iW#*v
z3h+5!pEh?U6Nd1g?7cU`@+uV}J2R~TQ(XHtK`OV1o7bbOvsf{?ydkz_1Evuj89A1r
zn&<s$?&!Z@vVg-%E*}Yzkg)k<Z}WpG!##K&G6NNny&dgNO`tQxXjiP-jMA>c@w^dl
z8U~A}UG9#owob%V_?7pm{d2I+iCc2Kj}I?B`D=TBnu|2^>jjI;HrVUgBL&Q$CX4y>
z{83CM!A?s>7<WVMc=+3Vte=#cm(rcfzbPCzgV}7(AI|t?l5<(4;7&$t1wkW~{WsK^
zlbbxA61D?5DhoFl74DpQ&(m`-<B@so!eIv07W&r1dZfUeV$izC_HEw-(0S)uM?QYZ
z10|UO+t!c7c|EWD-bgY<wO|`IfBkr8bYODiESu0%Uw=c&hCTPQl7sT1bYJCrp$nk>
zdiFm4y-LdgaIz=i_?MS@<2ExRw|G1+XTN>>ZTQ2<1(ECLuYbCMTGO^Qu~$j##IRRb
z=tP6qLKU0qUL6Hb6%wzhL{xn^$#g$->B#IIYgAdQ47hq+H`@a~oOO!|rxLOU=2&4A
zZ15I|VQmp@{0Zp>72aE)4Dax*mUYXt2xza=%wO#LGEF7oXt7u0Nc0?3=_AH6CVTe0
zaJc_#t8EqHmM`2+%uk`a6Eb~M#mS&(l44T*?y$1oJA)cV>v+rr=Og92M#KkY(}9I+
z#-3?+;#cRYBPn+DRY_htJ&S0g_SbehWpd|p+0nB=X4!$99^*TS&jR1cYQh@UTbY1-
zwWwuuEa!5jndNtB5dL;LO`i9olcd+Znl{m_%#p5a*q0#}AbV~05c#=%YKh;4unCk|
zmYG?p_hYY5OGj5ExSzcfJ>*R@nL^oc^O_kdeJH>F;b}jCEy6EBLf_r1;~+Zv2Z@9I
zqnXuDcV$Pd3kCiwd-cEXA_7P_!7%;t>!H7r6kmyw;=f`_6FBZJF3tRergos~P4&E{
zwFx(*X1j8mRF3SnPbSg)UG_&&f;~eMEkOpwR*#fszhcdWwHeNpP-C}#oQs2r5q|4b
zKb!Mrv@ihnGKDo!2wU)_WF0i)!4LX4>4Wjbi}zDfg6Z$_6gB{1blw;A*-1s&rT~$f
zQlxTN80-2a56++RWE~zt0<1<I2nzPvwRU~;3IKpF$W|6tPh*(wwwwt_K{su*;~|Kx
zmVlC8&@e4hxh!z_oM|EM>@wJz3U(H$!rF_niycEgev#R?9|P;oyr3itFgu#_9GnlJ
ze`~QauZ&P+@umENA@{%+GwppBS+;$pkFHegf#SYS5+WV)8!YyYbXuR-{k{pqulWK`
zoe&|*3O|v2E+{tjcdaq`V>mS!cZ^te2^|R7CF5?t-R%-NaO)4PbS^zj!cLeuq~5<T
zZTVI%vRjI+5=tFS#N8AR@?!2rHOCwX2V^^ZVJocZ!ff$05RrGxzgVHRhZy6(PJD5a
zJSyzm&Pb71y8Dks2pr7E&ycxFYyR1`zGs)CyR2@*$TkJW3<p`_p1*GyfVURhUa6%}
zZ!uMIf&(_2A<@JwBUc3I2bJ^b5u-pH&xUk>@@RK?R&vjCo;Bd)xbg9f>w~6`B4c0<
z)1|o>+!%%T1ziV73dU>~dXCCVfdTcUm=2GiY@TQd@JFq?L6N0<15W<?W6p=5_0Vy>
z%(j6X;Rxa{E9Zo5iGAiUGw@n?qj&{MV?~glCFHhn#)9BBMLdh{Lc*k#$~<p;!IyS;
zRNP50fUK-P?x0=Huv{AmpE0bSY+!<H=y26A2^F$JaJnF2-WGlqq>TW*afRqzCL_`?
z&*{KDdhU(A|4;{}dlb`?UQxjE9d4GB!9-%B?<U^E&jyajgfe)iE~p$&JAD4PPZqM|
zhg8)Sxs#2`ZONmDwyntIntic44muvP?6H?tsUcRg;oXo4$DK~MFL3}%orvE=Y_5k=
z9`sTGTdPX5Ul`B=ckLV0uTfB8u@^`m<&#f|OT@<=-Theldu(q79PxF2<^7IE|2Tv&
zf7vKE$3D3E_oi)>^8r};aS?W7F?q&mlB^(=w}Z3#BH`!Y)+opZM(?$4Weq*|?Q6N;
zk@4vw;ddNnTI>6w^f`--7a6mpjqm3^Uzs<1GRy0<r1PT}5B$P)&n3XfayBq?Hl`@A
z(4|Bn$&;A)EQ0?TZY}?s?l)&s(<J-Sjy}MF(Zs}o#^K$pS>0^Y`y`j|%xiS59h{R}
z`s4>7fopJfx?xrW!E|pG-8O5J$dp>-J`M0rVZvK*<!!YWUY(qDtz{VU2ikS1CxxQD
zPfg~{uvQ3q2DRmxr}{}$BhV}$6;ZowY-F7-(k5Qt@_MRr^@2lcp!n#ITuGp0PO%TQ
zDC!hMap2gbqbxS)CiHb}{fz9B4ar_&_zb>QgOX5k6Jz(3$~j#ymGp{~7<&U)c33b9
zn_=6!-2&{O`3K|pYT(4Ym2`tayb-n7&E=@=xnXmHL;LHIuu>n{?s+?mM?B8&CaX=Q
z{<FL<6^r{uJoKxysp&O~N7K<%1LlsUtTnJ=kN3}gQHI>=HpzOIlH;Ik<R~zmSOBHW
zTSJ|Yi9)TxA0gS1(R(wzkC_IWy&vO--g>9*&1nQa%U&q-9#;xj0LJ>riqD!cLxwZC
zPi72_ne7}qS>}tx`G4W#C2Q8*E#9(SmefK0BiRIYwwT1HlJ##=e-B9jBvyx(x?wcs
zU1Z8&_^Fx%W+eUKlKDQTc@n_6L9GMEnXh+gK0uL`L$~pLR@rxJVe2Wl1{<$!uxp}S
zS);2KAF^N@lh!uws$PbXN)b|@L+W5ml8IZ@MuZ%-a3fe;BqF@U_y(R>clL8hzRz6r
z$VVdxmgmq<EzKAR!{-#@N{nFmd0P21F4=-Hmxt=XRYz&ig=#*}nj*?EY$ul5X%zGZ
zZu?si%$PSX(ZbC3v(x;%_`-#zBmgdEc0oz;zy)}OW;L?@Ec89MY8Z+j5prIXxtA5#
zN>ByZWOuLJd@!gA)BEGIB)-YsQ%jAPo*GL!H@c|dZWlCdvFn!#<p5!&7Arnv1Z9EK
zUV_XC_!yeqJwFX2CPZkM;sCx$nr!*-GAz@NKsI{&U<V9_Ga=#ZXp@wh&Zo68eWUM3
zeFDFyczu?1wm~<*Mb;Zd_V{I+z@CFv@dPwK=vwxFe48JfsJgp%u*<%bt4LaO$mMwO
zo`-kaSTr#C>kj_zy&>fq^-nYN(_&yLN#RAd<@4v+Zv70#<|kcYko2I5lG6)`9NR#1
zrV~A{)#B$R!Bz`<A<>I4sdx=py^EWD3q(J4rD-fUpC032_!WF0qP#c0fL(?Iot6!L
zQYYE^porzH-jAi;QhOH{dqBeTGXx(P*Wk3ZJ@P)_VO2oR%B}wH1uKZ@&D&f?MB!=H
zQ{4oi38Q!~Jra-n0c;xCUbnUr(2blMBVflw7vmd_=EX8%ZPQZ$iJ0qc=_fw%l}SZ^
z3C9t^-WAaE5RztEtb?)m3wLJe4eVnuD-T;z5!yWx%apz3r?V2w!gC1DeDod`-Cz9F
z!I&1Ih<ME_)#@aDrsenO2#z_-gvQ&HdBe984Kvf-t5zVBL#v(Ryzk}Jg3y7IpD9hb
z(P2KX9fQd6fsk2mc0qJA`&qVa<obm?wAn2NOY*Gp6D1et65D`B`c`p(vCeyokv|@`
zv$?i~)x;Zsn_#?Ap1~1xQGwpP3_M#G)E5$={Y27Lz4(SDw^UJqVUnG|-K6=ukrq0|
zl&x{~NY^~s&l)(EbM>2G&QyThMJUn52rGO_e~ibQb=Jk{-n$A}6-!dmxXTl_pTM0u
zIn^ERneSVSXbTqCf(;_N^peYPf06Fq>DKsdnDDu24WF`XszSy=(0!gW`?YJV&SH?u
z=D!0Ni^|WXh}1K}4`<u8x>wz1_C48U!r||>Ho^0$>M0wPW_gB|#SM%y32w#o+ZXxs
zA7utd6&$9zNep&C`b?xvDUT|r?S&=ECbt8X`V>r-LBn?iO@sGAdrx4@4e`cOdMLe{
zKrz#gIOc%em!ZE&6omrCuv%kgvL&B^KaArFG)Qe3n<4NO+YH*GrAl>v_@0xz-3IZi
zkYP(oHJ8n)^KJA(DXwk1V+sqlGd^Z5N&I7Fl0nuw1*iq`|1lw<>2w3QhA~JWdSp6o
z=wq3xFlfkQDNrs#z?ch=39FCT3Vah$KX3R|>UiEhb-efB<3ncm58>4e-G4;aEgf)H
zI)D>F0xBt41kLAwmQ(`uGBIMcv&D&M#<NMNAWlne(hROUfjy-tiZlsVAX2&aron=K
z5lp&<I$NlTHo^|W#ikNK)SOIZ-WhB_TsY{IeV*)ZO=J1;5<MywD`muQ3VtT9J+)Xj
z=R<@~rg{xvyken6B$1b@OY<YW6TufHI1I{J2cS#NdJ9&<#B>E04Se+#^yriVY;&Me
z0-VPQR3|!>$l|=L^KrTnc9$r80Wn&w3>X)?A+Y#?>1MkwF|Z)doVDu{*u`1^Q>Q*=
zOo^U|hPk}f6nZ7Q4DyXI73H4S%w0a<s~^9;Lx`~MBo~eIl`v#RwH52Czcsiz$9uVp
zb0YY=&Kb87XOO0vOqI3_!Yd?E*1@+-Yv>3UY`}aoth$HGzP+W;`2cJw%4yZEe8BvE
z>6xi&a6d+A`btsTa0*WTRPvVfEg~tqOH8D2Yv*OuaRSw_nd3{~gC#a~R!Q#D%q*Z~
zT}W~~Gi5jNfzGnO-uJW_TGJl3aha@WG<GW43hxZ`UjBS9(HQb<ZABdk(G;7adJ`jz
z&7wUoG_R>c=XrJUh#^lhN5O&H(rQjlq^Fv+K!LD23A02ibIYm-&geF+#0MX%=$R+`
z`BgpQL|#9x&6L!$Uk%%B1tzJ-nh0hFiEsPqQLRZhZc&0Yqgx%=UnWKrHd#5d)*SZ|
z{O+^)tOcZib=nlodDJpWR)M#_@MO}Ph-I^%AA|Qgeg__!CU<HM-a*-u4*{n8S4?<5
zq$B&lv9Er<{um%4sQjU1_0Y79foAX+xyak>27nz+FR%=vp|voz@$+9b;G!94njeD%
z^S~7+iAtcLKt+Z~o30jg7sM3sLNtrvL&!_J<ldV8%ovrGh^rSLzdk<$oF7aQ9S5de
zEkE__5{!paBW9X#)9==k_wIyDd>I;AknlD0wU7=xR~ekPsS?}zzHzjzN|-Qp&!^}E
z?qQ@ma`3`3sq9DJ0$G96)W?A-Lh432_xiO6wl)P$n2(&8Hd(?rQ+<48apC3>KFURQ
z$*AN}O~|>WrAG~#{x779BDXp7Z#MkX5b3`O^j=Ytkdd-hb7NI}n7hfueSC<D(t*L?
z*aH?*RK9%skC-W2GV_+D-Nt34`?TVtb;Ec2J_R_;RE)j-)_gCq_ebPhFAbL72R6s~
z9ydW{da64^7C!KaAO8*^=P{d)u;j>m&8|Fq1&wL;F3*VZAlM<pJ;cfoKI_hDUWf$t
zdvdk-;gima8KgAWJNUJGUJDmDRn&#u$|FR;4msq?2@SH94%BEE580I3C35)=fCDmi
z<((#LW`XY@&a?L^urF!^DLhg+mQL5(ieb9y54;W`dLsGo02C?acb~3l0T3X3o7XiS
zs(g3^EU%_;fl@nZTb->?-ynvU+3X?EDARmQ)2@}-tSv)47%rr?>t8CYgu(lYWR)lG
zwG`oWBRxZ45&nta@BiA2@EFJ-3wav4j=lZ44j+loSyvrxGR$IMgCr*AH3so@A`K!W
z?CJOH88j#{$oEYQ$ZkBy>t5}QQ+~tbk4v%WbXk<=@5Fe5!~Ua{kgQ06cs*$eO(0`)
z^PzUJJaHV_=`bTNnj1u3^&80G4S4_$aJ_dSxQfPEqkh&QdcOqcBVW(n*%>rsF$(S>
z?0o2ia#iGUC$R+Q7;!ba+t@VtT$7~GT*Spq+V8u5K45{{tNd}!awq6Y>iUOY=-IjV
z=@XYf{_5u@OEk7;h5g9_@WAIxJzisx4W5jYJ^h%(?SuD;)f04sHp`SqTS;;QqR}i+
zFGkz0MQnZL9BT1;q8DX{7!hf}+%18;6VyXd_L%qUW8b&9YhJk0krd)CpANzEo|g)G
zUfX73S3v!I+MD{QH>1Dx<tDsi2YiTadOeayrW6Mpu<7wAiP4;C|57Wf2K)hoCc%(-
zCc?qoU?3oYWnaOho$WGR@LPSB?**pdA4hJBs~g=bZinqdZ~i&$Q5({gEw<+v{rNuh
zkdnwm)&oHj@=d6UfKTyUhJZ#&$QhQ=@{=+m-izO}hSo|LdKt|gAMs%;IrEsONz*%!
zHq;*ivPn+MQ_Ue#ZX=k)XenL$l!IunSs4VZ1?Sz<4)T+aJwmL`3JcPPh<Tr6TMQcw
zN(k_(PCH`1Yhr(w7rSnzdR{$<VI%>~k4E3l@Tvm7%scpbJ@wAOcG41NR}MWzy2mWY
zh0`&}!x=X6*GTZNlRbggm9BqEyX5%sT_Ss_hp+MegU?hi5SA7$?Ea2`v|EHvpHwpD
z9X6<&kA{n5epukQs#qNdNSK<Lqgpxb1g8z{6DmKvm}{i2Y0N_4@1xa&4D`urOZCod
z9ekpKm1Go5*6_UT%s-F@IvbI>t2<lZ+GsX3QB9nSWSVKitbFDwJ$khHnfVF%CN+~H
zOoeGT^!PptPke2i)!fErfcf6lRHG-uIm6kd%Rn^Hr~Z30E8@;N1^gXSMYy44#9qx3
zvv}yX1eW<ojpq_Vzsg(@9HK5{Y)vn7=iEQ99d1!3PhYKNVozuaHS0c5tZm&Vo#QDF
zoA*i76F%d%n^xOZQeQ?y+g}K0>HUc)<v5zT(z~`1bDv4N6x2OA<8INS@z|ozgVdvh
zRV(U_Zl!3{gdKw>k*hXwW{bj9#yqe=AIRP>E$Nj>n2JX7u3m`booN|W`d&c5(l6%&
zrQ5c)lhmBlx_is-8c$<^!8gADRO!qsLC?~3PR_0WI9rA25!OH*PxU@a;Gfp3f9YuO
za_SJZn>$24>Oa;^0VM<=-!Y#^$iw}|m;X2I!oUB19VECMXQZi6{&OJoFOk$q=>SO8
z0&VUA(HeNvMgRf)l8p`UMS3y`n>D{$&A9fze8qqN>fqOPa*$AhfxJTgKmLGLG$~kV
zHb86P9iU0U7PYdt8$zs%X^1~<4Z5)%y#WK)zl*{DuI>G|-#Hl#{>mrX?y3KWa5kVM
zwh<VXEFqa4*m>)U9~zhhkn{Dz?9>+j{we<RFaDn*_y0;o?!SHX|D@^r*MIivf0JcL
za*%>KWC_&iN&p1&D3IOCgrw%RMN*Oi{atV!@PI%?{}>qENOQ5~TH)msNjJo)d*_4x
zsaMC#g#i^eKQb%t?7xrD1M1;E;2DG@IU^7lw(1dZ%9VkUc23dQqk1eUq&0_Jo)B?u
zB+0H&->fUo_7x=*?p6W#2VI~`7KWrYC*NOvvay$eBUA!QXI9@1;LHsHLBYZ?rTh;F
zBI6CzDVN`9EyyzMhJ(RM(W2fZsS`i!{$Qj4sjlu<0szU5fEwbiO=V#Rl4p$Tp|;&$
z!{Gn%hj{}`{Y`9mwEyCt;u}pat^wG^D1=-ATj*nxmB3MHJlqG6ODcipcofh}X^+9!
z&|C`m5>9H+ob=aJ$E;%^#m+M5{3~KOac7Rs<Wig?m7MT#!B>Dr&83EzWXIR$4cv)f
z?hNK8V}Q>bn`%u2gDBLd`Fzi74ll*sRzrP2$nj41JsiP<X;urc?!)$fH5{`L_JDSp
zwO7H|V~QLAe^kZ@tUSgLk>tY;sK>&44nWIQ3L$@>Fwx525EDFv;raF{jVYk(cGA2h
z9tSi?h7w?M0<rt7fbimhf$N#dh|iT_9|rv^bpPYz`i~zH(1ruC%Y(>gcmI}u_YlYc
z$VJ2(Y_^v{3BS~qonwpKl_kag0V`PsIZZS%)TDAF;|UeffH`z&m<@m)p<#H`Na&Ky
z;f=@D6`)5yzq!aQyLY`oCO`5MGQr{BMYPIsMnleja(ftJ=&rv})d$iB2QU>4TPfL-
zg%n}bNE<E~#94O$p;nMQ`y-^=t*nu`U4OEK^{^e?^%IylA+P)h7#Vzf$cFa`y$ngY
zd;QV!XxN2o(?AVudy~Hei!FnAKy8a!ixy+p&nI3{{XJ{`kDtSzuNtJjuOzwo;_umH
zWC|JdMB~jAgV@d?8J>};90U4_pDt;mFy;5OZScy4z+C|3vl8ee{Z7jo5th_RGhtjC
zY)B(X;8Q#D(c*Q5c3~2rO&Wv1)?+45sBh$!jolC-nFI#b`<kl4O7Ff>FYZHiIX-hJ
z7gnt6)Ag=?`!e{;t=y%Dp|xZIwR#z>EFeTk3(&Y<O$Pj!$QiKBc>f(d!#e;CItm2F
zwBF$Tbog?0j(K3_o7zlGMVMRa8@qq}>q13g<&7IdVSn8skriZ>qqdj4!K-kKw6Vku
z?_FsQ?)v4P98<KAGW-CTga}%BXJOIEZw|8e?UCLUmtf4&Ugej;TB8zReTNBtTN%Z@
zr&*CO0A%V_u{Z{L(^e>kegt^BOZ&|aSBK4LfhZ>T$^Y+~{jYyrlT`*c?r{l;uKT~f
z`IC|)>2G~NgU}lW4iE{MD$uyrD3iU0B$>)dKscG)x!WTJQZFGV&Z1kwfSWpyAYva*
z1p-BX8>Cj^tYex3=nkcs8No3iP%_H&gx<36@2$Hbq!P5X#s=<<4%J(Nld^&utY84r
zvOY)3pL3|6=2T2@{a%-ovREOF%x>d+KbVrTu5hKFUikxlzK>0_p$~(f1k8|FWXU|(
z&fp(CDmQP&BVAV^?&_sL+4OJhP1er^9{wr!<(r5~T7l?7qx;Ycp0f|4NC>yU1oIEh
z5(tPj(PwN)V!DZu<GbD(Rk91XOq>>M3%@5%QlzU!<u-XV1y&XxMmhO#wt4}W#F-j)
zzOxR1@fDVr)O@8Luy^9j4ZCGeNkreKO=dsG`Kuf3Hsb6em{vx~a?d8B@JywFus~;i
zdMwTGU7gasgcYj>AvwJp$t9n`sUl+^XnlR_wIInP1dvDs)9(FFXW|ZsV`P?4XW~vU
z30#g5Z&>-F!PnkRvr}T4ntX0KMYyxfESmmiV*N?KPEGq$2JC2EqGCPB&TW%%F#+gg
zUJ1^fb?)7EpN=uzBM(2R)iM(pReGnXF@*M>?gN$-k!BI!H!3ZHKwfzX=z#Zr4>hn#
z1T?&aD91awF9N&Y7(;BVW%o_CJDwZg3MQ@A_o3L7;vnweiW5vNS&Y)|>~q$GN6RlN
z=75;E%gxSj-G2)>Xj4nC-VhV55>4IpC33&~&6lz(mD+(Zr{*SK1JO%N8iV28NnqJ*
zhPi8lTOU{ZH5LZbBObh}TZzW`@2+6psWsoLo3z{THK{QonN$bgu*w@>NQnow4n_q6
zJJ+2sV1~KIS-V6>)k~`1J>j+Vb}eq#YEWoUE1j}5<lF?w#?JFS_rWB$3FK!lV}YHp
zLp4CS9j$Qqsqs%1z5pFc;KkzOHTsJNdqNd9)eatfCQ=5Or?KxJ0vErSOkX|NvG0Im
zMKlNj4?t{O2B@b>2x0~#YX+Pq9spJE06cP%SPcBpE<0o#39()VTKy=8=36fnUzLJ$
zdIV&!+{g27kaNJh!I8SJc+hqg%lYjIyDiW)6SlHIP2u@NLWZoL0Hq<X%O?>UA=e!q
zY4Zz!<m>>N{Jjx~;p6!9DheN2`R(D&F(4mSRME-pl-zCf0WaH#k^6UT<dz0aplCB5
zTNH4DEot)y<XQttU71CQUZ;RB0oez$g1Ppq;QOYQee8l{%esE`y6mK~_a06P;TdBF
zWe5Brsl8)0!h|=0B*uyf%HAF_GvRMdr0^#7zv@<lyPh8|oB6_{5Sp=Eh#30gVye-A
zDr<-;2XY#PcTsIWz;#vz?L_isWrBn=FGJ9KYEVRNFEAY{WR?TXQYibSi4Zl1K*A;@
z6YM;(sOaI+&Jh}PMZ#Ei46j_mQPuwrgv9faVfL01P*7&7(ptDQJ3kn31P=7$I~d=<
zvIxUp!HNaROJ8C|WIt>xEY=;B3s-Kwv3-Nv<+&H)E|e%S#kqDFF0GQAbdD&zjQlah
zZ4!Nmx~X4(GN!xRZ%4ES!2X)t*$1Yp*BZZ=k_aBO0F|$iz49`YWEfX63EOFTISeYN
zrF&^Co%{Sj$-6n=gBYD>>i$@0S`N}GhQZay6u7WFejpfG2)5p9y6~5xvcWkB4zXY8
z0V#kgQKWma8(hCO<^XwpmZ3>dB#OTqTps&hi~P=EOgm(kPc?LGt|i*Nt<W(-?#X@W
z#ZZiXPJ(^*BKMXf^8;g!K`EBD;%o@Vy-XUwV%{uh9*$3~4%^L5Y^>Due+Rf!T0XOY
zUceAY1hkdhthadIk`u&(1ilqG>~j*^*4t<6JY!C%^X}*!Mr&iu<rDiWr(l4_JMFPL
z^F7@2c2O;;VBMhJ4QQft+@lYsQ4RQtRQG{pqcg`(8ZxCJu?Qw(*&R!C(qH|&{rh{^
ztw&~K+o^BO|LZpS?@eA09yD7~?ekTCR>4~gBr?_;8mZQNJ`a>ka9Y=7{E9##Cx-xM
z-YJbY_{B%ydB+9C0k%&r7wOsFDMTXsILGw)Y1trbhH}&?=TCq_M<2))bjzplqCzj2
z^xr;564|Z`C5`xeQ|bwRO7)u5j5$4(!+KJtS=<M&M5iJz<^tB-qhKU^&Ss{{3Mgm^
z^Sa>ibAs>1>zRe&3xnqgn|MovABR~7RSM!U`zgqMrKnW_IjcZr!`Qon=tBfkXq=6i
z1~iuhzSj)BFA%Rk#2KSaIKm5b@%AB;rS~`j`_EgwGzFeFQy~N3IDH~D&HDSsETdlI
zJf1V!nz5wt-r-x;3>Nq&1TAvG3DjSdTw4)2Hj$iY&OX(Rsk(+!O(lP+iOn&1+g65+
z-n99k;Snfp3N6)~0cnY~H4!c#HMPw7v^-}zPl8zy+$-l<jz9bq;uI7F3UMoS7Ux^J
z5>R+nd~>iK?Gz1~r@6dEVGL;iIYBA`GcQ<0*hzNF8QA|L_Mi+3Y`Gem#q<UC^2byF
z3NMxT_CQflOPjy*X6tjjn~;er(rW4ia2j?M`;svjz8kg8Kj%ci-%RP$;4>>o$iRf|
z{RUVTk>=(R#e^|tt9s50Gnk`Hy-#c6@NLAm6rcb=#za}a!6qvpln&t@NP}s2{BX!|
zzAiX9@<c8T4*^xQDTF3>%lYXvZpC4!T;7TKNPnKz2QJ|aM5duUah2$;PSo-Ss1qu|
zCoV)(bQ3fa)$8*ki;5H+LTXb9vjVIW3=zXfdIOC@<3H{oGTG&Vxnu6}4%3=bDiZHW
zn|6a&3avSa9W^W}mY779{P6k#63R*eFOp}Oo`iQZ(wXzd`Du$Q#a~M5p@dY9VgAAX
zs0PJ>&_gsM?rGTkSq+yGb{TAfVdQh2+QQC504cKQq!}Gw@OZxPZb^6<yDa{+KI<Oa
zmBB(7V$70ymcRoLkZjEdlXY2oh^~!7y`r_kPi%3M-1_gz+_B@=*Mw_dOok>J>vxRN
z9=(6&-YhclqRj1w|5!zsnuFA_s>7c^sK2Vk{}}VgrI8(&KmSXthi_!(oV4NS&`2`O
ze14$l@-RSEo5nCR_TZ!4!`C-9XsSAt0<`BgrH(j67w-bydu;YcRlv-&-pDbHyJ^Vt
zqN<g%QY(6oVq&)G>8x4UY>Z=UL=fNH@EW10qlLC3WBej2<syMy6R3Yteizj%5)9{K
z4VmfZj0WN`+KvxR&CAogcm!l3B_Oy@N!nqXl4S=ue@8~{#pCvuhL4}8gd3wyULY98
z8}cU_-cOjkQ&iO>RIO4xo-0H6&?;NNjq<)oz-qhHMO4H_GzZ2-c*aKf*cS3GJ@zkF
z7ue0ujkbFKSb$-WDI$sK8fMbQ@ma)3EEVEe#Oqkxx%|Aa^~@-NyokywMDC)GGese*
z_T}oFOP*Toj{tV(0s&3;qMAX-o=xiW3laM!&hKx)Rz}R&;bDO8!xDvbg1+;*!$VWM
zNGG>QW7LI<sM(f-M@XZu-==D*>SH5LKg;&LCGb)O$cZn7+J#4y$5IKzQkk2p{SY*f
zd0!>ysns|;9e9vCEMOjer`MbyO>eF}`_lgvt}oJUmhg4#XY<f)8n@{f&cCF>n&NID
z4K5n@9)7Ut+Eb=j6BTH;d-R@o<>+)ngt}{h)mKj+2AN)RUsG;vZeNOt5Toh@uGVMk
z9jGg!f(%Zn8kn?(%47#ri_+n63YE^Y>{~^-UzDagI5|(Mqi{nnyV(GxA<0QG!vRIX
zjbi;Oy<*m@Cu#Tq<wx!xD_C$*fK`2B?~8kE#4TFY@HhPab7mhIXi-)uHer)}fuLxK
z>MJEbBp!*%eqtav9=c^jzrP}?M>kQ%Xk@&)o3JImo!VXEHrD91Lb_e-PasX1d>r+D
zQmR_0$kMmymm#W_`&cpZlhnN<#&M=Y8eHu02i(yEi<(GxEZ?O~->Y_pQW6>=4p!Y?
zi~eaSBE~6~$Ne`NBtTyIF-;O&3A1SUF@AG>P}3u%u|4Cac&z~K4|Xmr|IjC#Bbgd}
z8*@#c?-ch2je?<w{c!PY@9IXVc`~6URjo|<C4TsGf0s7p4}=s)_&WZ&_O{2VzS&;W
zJ7q^4nezANd&?J3x#SdUlCYbJH9O+5m=#Cuu?;DqQE(N;;;dESd~Mi(^G)ilv<8!k
znA1@9)vvJ`=_Wn+E6g2(ek*qNHLVvN{l6vZ&N#<pdr+@u;kxX+9NbDzC$By4?%=%6
zMQ}uY-H3QRz-i8<zK|@<=egKp5-ulh4Ah6G?ImQ-vtdIbuXB>H`<kY5`d~Zn^1~5G
zXAY6e@ghVX*PT#UF^lmDZY&z^EB)2FBstF4o;hvWrtMPO?@?=A#FIwKWc&LOaH;cx
z1o+)eYoC1FgZxV&Uj&r3RFYeRU<iY(x~Op6snf^HHbf?)vuzZ%(Gm^={^}C6Ft4B*
zJz7kxRev64(LOQ^r&v&3$nP&H0Rkse*eoJ(KzVNY_}KdufTv|#Yt1a;&8HstDAf#F
z!+3~@T)tfTnd=|`SAGkUM!<Wd@7ml{kXxKKij-|ygD1!Rk|H!FVQxug+ub<7ir2bJ
z+<!RO?q;8UdhAPT{y&CZS}Q|>WUA!9&bZSOq?uz~cWS)o>O#7>eK~T6RJc1DyjsXj
zV@^JM?ti55&Xn)nL0x9-1qZ=g)$1H11s>QBDls>`H*@)o4(M_Pud^iyg1g^1lkwsY
zJDmdyu0Fw75#gJ^76aKf@v<>ygG&0!d|k@H&mRh4#Zd5pTfeHo0{HyGPZg?fY8=um
zykXak=h)hl$8RLv>9cu{Xov96l695+&obqHT+IyT+I|ype0rbVh&T7Lbe?X5Oxp9;
znLjT2a&_BlsknyM^|JQqNb(7s#%tAP>Rha_Gk(LLJn1Pg968XzdQP;D+T1QvnClid
z?_J)^M{dBh!`_wcvuzzr>}CI$n)GJn!W;30b4D;dN%A#-7GSJYu>B6EP@;X~Y9bvM
zgb83<uBt+FmtcC{<+)63Y2wz$Ohs1`c&P0sO752?0uqJEq*wf%x$8r;VlvISf+aVz
z^_@%1hHI%FCL8HorT7TQxP^u4Ifl>29?0JqCR=#Bbnbx8##@@tHnX$SSTSj?H#Izd
zpj+T<XIkE9s~hG<cC}vCf~&S<MTz1yHT#6}AJA&LU&`pQEr@B>-+iX+hPmxhZu9Zj
zJNEgv%(f8kl1Xn9`J^R&Y2WoEd$u!~eyv4B&6M~btHsWJ8L_E#MXnd@H2sgK)2`Fr
z@|r$)|9!dodnZApH3QnZI33~6u8)84x-q)0_6eLrQKLhSOdH!CWc_;m2{$ue&G|;t
z)gR#y>qqTg3%%Bx@0u%**A)w%WU{&oo7TQqD-4l+Ej^4`dWq{VsD3x*-dxZGUANc9
zO2S0!rMwIOnWlOk2&$5Md$0bvr<vCv_34@A<THz{567jJfDK~%M&SDc=|mo=9gZTD
z?zu!&N;_nEJ`&x`x$hg!jkxLQWx6@9L)6IT({|f;MwvDNIg}OfO1|t(t^cGzb(wBx
z0(2PSFWmTGTkZBC#clm6NjP*tuB1Oqo`zJB;6d9|a!@T&iG0RkIc0v0ZCYBvJd(Rx
zNiIZa0T?KM>XNXN3a18J7eoeD!8H~k@WdTg5M#`>yfsip)R_2Vq6NXRp^4U}R^tvd
zDZ<oMdS>3(Y$rJLPsowk!PSrq;+Nnz5JT*#S5|8`7w>&KkZ%sEdbX7K3m(s)-dxHW
z(}9ROM_z2vslhnFM%w7+6P0Z$PUb7LR{qKLysKBSk^E^=ZZe_NGnG@E=TXg1Wp_W&
z<Q1L=DRIMOG8y_GWGA2P8^G`mUZ18cZwZqPL46<I7Zc_6-42{dgAS@wCZZ(lf%k$w
zbuQGOB9$iGwM;D)Bl|kC4K(#7*I~m82R!DO_uFG5RrLO2Nbhs2T{~@=MIEX72M+J3
zTlP2i3%qOl*Y7Rx`mKPmIHj=n5hJ!tcCSNyy6(xg-9`sD;n&%5b;KETmJ)Ve<K8}a
zp$5T`)bM|LB#|~+K6<3~=Y-emIgC^;J$LOxjs6x|rt!;_nMgO#zYf6floq;zx#&Z>
zUN^kZJn!AIyv3t&5ROv`;)+77En}4Ea}E37!g@4EA#t=5&{F&0%lOVo4zGzlaIVXC
zcw3#WNiqpCts)QsNa~p12E0OgV6%R=y8MUTZdIyM!6sb}B&}994m&-!V5e9+y`af?
z$JAe$9UIvi>Wjr&Ja=nRvKS#lo(em72hAP8pqvrSL1c0Pc1E2rV|9Or)YdCxuYgcI
z-3(GeFOJ+jx)$_c;(#b=ZpcUpt7ctCnq9=A@Qz{D*RffkSZv3`+Bd`cUI>B@f04-=
zTvhk$+3!I%xlN;GuNotHRY}rDey{ZGI>%AR19UGF_@F%i;3Kr#dN2X-jJ_e>Pa5Mf
z`bDP)KrVFrbO``OK+XN2I(ne;E<zE#)_Ft!NR3pk?2lJICCcs(V(0p}y}#=3q~*kx
z3AZshF;g`2oI|}Hckk^mvzv7Hul%4|I;-&zgiJ5!DI(=LNi$`z{slu2bRAB)FGXqs
zI#vDS(f^~DoHxL7FTnM$&ldYilCo-;-;*wGR-+_IqDP6f!<Z~Kd&I#b#lJn$&M<Od
zEonLoa-e!*<7Sa2;n}u3G!Pt^&G%UZGXD#x+V(=H^30caDduK#z48HJdB{2gmzl@x
z9-#U#M(3fk=|+xA+|0HF#=>ahd3i@aeo!g}jC4MWnKk3z&n2;JJ%8FEYsrn$lc*_}
z=-$I^&GtSz>w~%F@O{pcuij+-UjpnZFcG+jIuC4t#>%c5sGWXW-hM93wYzr#6XIF6
zfU3a5TLN^aDh13`ATsg>&tqW<L%2adMt$!El=e28{v^p|?*v$``bUNVhX`A=14`ur
zK_}60;0|O5Jf#X&Cl(J$OgGA3iw`{)#^nz5FDf%hlRxMZEj`hB=|IEy4XyDeID$eP
z*Z>vSS9@Z|lMtPRxnuVEm#0|!Ea`aGPz}43r&eE*okujzTa{J5lxXPIsFX6*zcw=b
z(4P)<zs<wz#=@*k6O#<-Pk{?+tiB~v;xA3=X$F`bMulX<WX)IyFQtY^6VKQwW<w?G
z^f&{pdg!Q)X5LTpx<}618~E3d*t=t#1(Rp{`JF3Nz0xp_Mol;l^a9Z`cWX9Ns(Ig#
z*dk8KVbb@hYYX{sR9?i)F7NnT%qN$xuC<|^AgXCSy6x^rj>rA)Vqm>DDE)m!%1p;5
z5j;OBE5lg*$`omXe^xi)AB$gmJ@fZ0r0yoDTJcy!5690<9w+UnA2a#5&Ktn+u8vPY
zP>J)U;2%Qqrn50=XBDz7(b-@@$JKRp?xhWw^%Q#goM&?{qsi|IdRn~p=p|$XqYi!w
z1oK9UuT^e+bfG}p^<W&HH#as@Df9)?x>#KM=dZWFa4|#q6Cu;GsgMsoxN9pL@Q_~w
zBbGZwyY2Z*%v2E#o}Z?(V58?y8v6PYQ!y8xWoxJbXr`XRz>kH>jJK)d;(FeX%EiDA
zBt))wm@>!{e-cS4cJ`sL6QWayumsF-Z_PH!zlKdH2fqiihM!_(RHhMcBCp;^=`<Y$
zzP<<o-gh)Kp`ZaEe0w$RDX0j~b>2~r$P=_)Anv0_?cbDh;$PQ;dL5KJp3Qwh6W(An
zW1AETKP`X0U~j11(bmy^Q;rMP%;sggv;|>cvqAtiDfGKHS??pT#TtMboOaOQ;ObVu
z0RnYpy|T??zLYh3vPdlbP5q-YH5Q#9j;?f?b95YcMO7LgIUZaC#N$}pntja~M}~gV
zMv>MhZtgShsvGXv;Fh&l{i36;8*3^v6}#K_WlVA%dA!nC=Lskogvh1ZFWBXN5U@#<
z@7CvPWuqv;vBx{(dv)nPXk_9+BQyBVjSS$4nQ)3+`wJ2ytqLQVgmAl0xR6PB*xZ8k
zjJd4Y6*4{G{LzD3;n6z$@Fx6Gv#h!^Bb!&4<qMY2ei2Pg1+2}ySZp#-Z_k_%MBdv7
zJSW4@&#I0~hDdJs`|;mD3--dU;({swX6AxF9~$O$W1LUp<n&dUNwl>bCjJj!0om<T
zv&(;|ur*F1zah`jl%!|p+>A<RW(8AD4g|FR%TN96g0~E9Kg*#iEliJmh%k8z)X~V8
z5}wKn*Er1hgRX;Rt&0~xt&|d9?A#=iAd|)~ZeWm4*h4n7d`zi?FbkQ_l9Z1Yvow6R
z+E}H<9kJP$!Xt{M)D2k7r)^F=%kF>c$$TKPuP7GqN^q>(k12{XF3y5h|B=jZkw1XW
z>lzK&-=l=n02FWO)HUf_P7l~g&!TZv_sI<`RT0V$Qm>W+Dle98bMrE-E%3OhEr3m6
zA#(3*;!+5cBixiyqbU5%*wv)3b2jyglk!?$Y90D<#H#E9>UB0!xz=pG_uJwF?qHgK
zL0z6{lY+xYW||vg8qiMq0W}l6TdjxiwAguecvjjOs6cYArWU1n4bQQr8zOn1uhcfo
zFt$d|;FWOE%!vu!dkRvC*?5?=g$E-DI>QcaJa6RqvQ0N`YK7NI$Zq+)>FQdRGIFlB
z4y>vR%YpO$4Kx<_Vb)+pRtOIcL*j52`-twPxT|;Vb)6d$VNYGj{8hDD|G-U)nLigC
z{tAhsXLJNV*_}U%p8G&6j1KfdpCyh3tV{_qaHf0hXyb)Cyue8PHkX1UcjZ7dtT=K{
zqt&A^i>aaOa0K(khpYk0rv2`osl3!9x6k-nqo|xyq4`V+hfij9wt9vU-Ah4imM<uZ
zJ|5FfC3?mR*3LNlfJadU3KDHOS>iyXwrd0M768j~L45H{(;t*VJ1p+_RvUnfipou&
zioXpb|9$SuPPz?`{3w4c3z#B*2{29~10V<yrZjbrLd&Hj%~MQx)7pZ_cF;v`LTn<3
zNJm|qV~JGKwq9;vyP8Z@GF?U0*V!UUB&Bd!zwxz#)ieL+q2X*3`zDX#PU8uhgb1BZ
zB@c)kO)g6`Q`p3t5~r_j(mD025l+&&dfz_}F~C-SFBM6mWD=-ecFLYjrXrs?fEYok
zUnHWPzsSmMQc(3us@K`F-s&+NHcj=CEOd8%THV-nU|#D0tkXm@%%-)+x;;V79E8)<
z{GrlTz->Ny>a4yl%Uais>AUr9iB!VHR0q>7tB~wpD{<W8AYd7BO$UH8^2ThKx_{+3
z5f!L5t*wdZ3oak-lvAD*U=&3S{TIM`_ibLZqdQ<oEKG`biS_FZ`i&|i8~HFC8zLB2
zw(wG|5)F|CBQ4}Hf6$r1s?<gxL4<JIN`OoVrIg^mD~T$Ds3qpCDI%@o<oDyapg}%S
z;igNB95V>3?6(gfr^m}^3v5s?4=CLrD!e$D`s8+A;sY}yh*hah5dZB0JdW7I2Q7mo
zl&!zP5IG+s*+4>p+DnYM-No6~;?u%aOhNc`?w9v$LoqMPav0o|<^o!9%YhS|t?j|}
zL>sHFmQvw{d8x`HU^ZxFHZOb$Ce1|BfrA<In*OUvMfj4&&?@*K*eR#YQyjHs#Gm)g
zloo)A5TD{k3rg}ue4?m$Iyr@tB7A1{9+)m0Jz~JDOA7D^6PEx^i*Y3Z04Mmqzw%{W
z5wI_Q*dtjic+dr~SMjp;N)4g@kEULGhU-9x%BEv}kddU_xeE9alT*lcdoVnv8EaIp
zZY;YaDFLC=TdF>sON(N25xyn#h$E-ESmc1uEkstZ<{MZfn%ZqqOyE<hclq_2!b)0n
zp*w_#1`dwFn=a*o65wmrRI`s8)@uE_Ly)s%?G?3uu|AuI<BT}^_(AD*Nxi9Yr{IB4
z2{6ZQxK=X=FMn2gJSD1sb3JkR^b&>L`6bknZCR67EyxAc7I@8iAXifv#6~;}oU4(`
zzhhh%RZ7I2pXq%&VR^47&8K5L#$%#mT<Qv8^<(JAE0^k~?MioM{cX5cgKD2_X^(@Y
zpNH`jOF(sFQkyro-BsUNj0{Z!A*E{m>Uhn@_Ir;B!+GXaomz_GFH;i<x~X?%3auuN
zA2(y7syz9o*4|=@I^CNcCtlQ4P$qzYCeO;z|B%A{yAlcfI=~A^lnQW4ZGXPVui7Z@
zvxx!8-Cb{>cLvb*;2ya1DRc%{e+ZK?Ls>4uiX~F4>t8T$Y}Oxi$tc8wp?xvY!wn{!
zhPy)L%)~a1oLzHfaHb-~HeZ2^b#cW#nnoc!h+Zo45=!)GtlD$%K#Mk;rz(8KbR}`j
z%HI!7IIVG-Ek=6!wR38daI-9HH6O#8PxBdk$nLAi;V#Zoqmcf6XlS2i<;N3IvqDLu
zst}>j-=2jyZ&5-B2;a<U-;B6u_au9Sx`fg}E<vg80BsgTF^xH@xYWucQ&gocg{<?$
zR(Z%c5P|3V0HMk;Jqn(R4E1l_RqS5;H9hn)37b<Bnkh-R?XY#63hskM$xXstR8!}{
zrgWdS(!q1td*8C0hEI~#d0k{8INAq#N!#9s*JCy^J?R=hWnix9;CnIQZ#-u!K|a_?
zGzH<ndW-7JFr5RgHl*<sl5^(^CVh5}Dg7cg6fT6x(}oKmUP3Bmb36|+)3Ll)*>&!!
ztD6X}TF@BL$o|UWr&neNqJV=@ktc#>%6*>ujkT`D9%In6NbYIo)E8YG*yc{bW(~;c
z3J)6TMqxYf>s3W@rNB9qlwDq681A&z?g`8uYYtEk{)556{EQ{J%(qa%4pi%zsnZI2
z&5t@Ye4gNDsFLNL<E@jHN@a+vt8P-XF!X6l0kp}-!8LCT?J-f~%t!k5w3a?Qm-6BI
z13zO@xk9DLAeFD+b;fZ}5eu#tn_O1X>KYBI!U*zprGk~p=f%liK($RJpNCtc>J?6}
zQ57T1pZXw+Zmz8o^!KlfdUYxWH51WbQn$`iHL?&7`g~aYRoV9}G}nfsCY0$Qxs2{$
zhh^j14^V+Rc4jQUzqcW-pIkZtRx$-coox1$AFlZ~K7Bh?AieeM7y76waH+W0O#VW&
zERe#KN2_mHdvWkstW5g`x$F9HU))TD>#PS@)N#7)*MEC^>%QTq;*|n{!@osI|FL)Z
z*D^*<8c2BtGt&RseD=(cZLo=L-cl-W#DTd5o1OvkBfCdC?p>~~AX*ZO$A03xb2j}1
zmKN)1ccBOUo1N_`FP<6yFrM-^{<*>|=WqE(g$3+f8UEN`ib9(F*i8ACfjK8R+K$av
zF^xa+@*3*oID#0Tjw@1L-Wrz*q%`tcWry>Si3PFC=ZVQ@QubwRrS?d@*5y5WA%^er
zt5@l-_6=4y{E1Sc?v_7O5?x%zoMgK#EY(J=R^0)uV7^wLDkfOP2kHnDAM1JE+c_We
z#E)H#U9^uCd->>T#Jr|AdDa&7!YSklWQV3-$dy2T^5if4DoWV?sZlIz#&NkX09c)R
z9pJn5Od+$`Hc(w-g_Ds9lSqp_`b;E8Q_UDfkw1hTFwA+8+nfqur@e2>GPd}#ooGfM
zCsK0;8J=Jj=^)M_CKatqJo_z%BVj0#{LsoCQTudJq=fW!;9mT>VP|Tx%fySDR57Gi
z@k^y=%H4ezZHcU1qfhJ5c;pIiGWei>e_*y^a(JWlB_BP+xM&sKPhS{18<E*krxcWn
zk&5m@MQ5k5@M0G`Zw5tjY$(@jMN5~ST2j|}`d~`4NJyY_>Hx#_X;{#I?=wPqb8P8r
ze|vo`3|>~yLx01&&(ysJSGJ?;eKF}CSG(?5nnT9ww~TN71>YacjkGarp0-;(uOi)5
z;6Aclrn~3?f`k-yEZ$CzTdI`_6g%n|h?co*7Hyqh2)AMoxwKi`+lPngZ5{S5dX$$s
ziZ<xmmCfCkcEp6{9exb_SvA;0P$rX$&Sv{|0z*J<>z&v$k23MS@0-6(P`#4YgyHhy
zcjfrtV(hG5vcrzsj3g3bjYT64qK6lM{Xi43k9v;#7r%X3?9KnDt%?><W)?G^bD;g#
z@gCr)_G*wO>3i*nZR#KYn;bPMR`kLrCWUk9=2p+|lA2f<=3iREl5&6I>OBob(-qp^
zAk?~X!OpTEJLnf8=sG-wD{*cGG~&M@Ar?F>u~X%dPWl(O3*--+nle{E*V%}DMg8y6
z_CKzb!Hf*wKU?OWTa@z@U5?j;Y~!KtM@;4Md;oS39V}^+hKpn{1^0?maeOq(_!zld
zx+}6}6_j=Fz`jkQPl!_{S4?>lGyr(IA*?ND9}ChG2k@f{Pnq(s<J`}Yq>KEKp`wyo
z9sqv>c{K|JjlO0TRu$F@x|6#a{ezzM)g=IFUOu(>KI_k)teE`c0aHlUo~b3ZJ)p@c
zK6%f3<(xP*gHR)cZA}KvQF4xc_P@>yRRL1-w&<vV?vZU35^28!j~XDQa{%Jeav+2g
z?W?!kkMOt{Jf}z_O~>mDNim*2E-fblQLH2O)1~Qa&&tl+d3FzkKgHY?s}ZnyeVxP`
zOpF;0r}eKhYR0v~@1NeZLZ&<f)5H#^zJ164kF)m<q`L3_$0Zd?kqX(zRvjUd9fxC;
z>=7z^mzh0MIp!goP$-+ScakJ~XE(?m*&+Hp-(A(`bKm!McYVIUf4WYt({bMK_iH>~
zkLP13`N*Z<o()s#T)b}-GCeQzKE{2UINSR-An~6sdg0=7WMzha?d!BhKOAU<BAlb1
zqTU}2u2Egfoju^q@?LJ>JXEHRucEW%|N0b~Jwz`OY%937sjW7DazF3RLy07K>qBi<
zP0HvnR4C36T{QIWnm4`1gRUquV((Y?$qK>sqbfu3Qi<DCGaP_yC^;^luADes*kTxj
zt}w?)AdLH_^=Ut1oa7?MBN##6!M)Od7(pdan3R~LocT3^Y)Gmg+bo>BpuiU~xkOH{
zQ3-20Kj7L&2^7HUI1-HW(229m%0wxQK+K%^&emG9Ak~&a<V(A;2AX9T6CJqUXM*-?
zm*-b{-)9jml*$@}q2GM{0#wqb32WShsE#+pQ``28w6~o8!1S$UC9pjDH#^7QHuL}O
zhmuEFVWu8993THj>-7J8p1*%`ltAFX?^D<R+lxmBz>lwg?}1B~{!|kFTZQ;9e<ksW
z-~0QU{J*{E*b6+oIbXl0?0@1u{M+sNm%kE8w*LMyzP8dIfN@?bAHhC=jIq7@4^0Wq
zt4hanoV`(ZSIG54t#7MV3ct)Fs-qol?+zZ&(4WyHTKzVy`RZ}QpC6qgY7dD|BC(dn
z@2`1I_IgV!uQNb#gCM^(k-mS(Vs89IC}`mQ=Ye{M&c4y{=wb<Y#>UvpHZwNh8E+!F
zcO)+%cPIDS3$G{b#iMkN9T(TZHR&B1!r^uO&E1c%+ya_%KavltnhnwvXT<nZG1+~V
zwhxD8aIIHWFwr-z-hxzFjgnjzMsU(iwE^n<UBC_8O<JuX_A#nx<s+~d`RlPw)`hL6
zh@q#5*=o*6juS!#wf<BefT(HetF9*^>59^pjG`{L(j7B;&9plmD%FGsFi+fe&+WPN
zb2<BO_heNLsv-CBXPKQppY|6C!fRfePoRc(SO;Wy@P{#UfoY<X=2ZxlvW5lgIg*vD
zsb}|^!EY8=-6bDi`41s|N^lR4wVoG8DA+(+<k4RQ8h!~}&rqoN)>KzXn~R^%3Xnmg
zvSbg(0kW%@eCj<i<9}%SCZNYLsHtOe93duiADNA8Z&Uy;5ly)>;J_XbLVLa~)@uIv
zc+3y?bXbUrZkw0(e+OhVlIV#g84~0&5AJB86rH!4xYBc+<w^$|w0X^drxjN>SPYq&
zG;63^DzT+4A)^LmT1b+|4>5th%pF!kE<@|@yF0WCURt@g&#giZ=DiVM$CGW00XkK3
zMB>}C$2CY=>yvVXnc#v}c>}#5Gc7$k8EGe{Nx$&#w&kBsRDy4Ukv{gAN%bQp+|hw`
zFo-0jc{b%$v~}o|n=RGI3)9(Ii}qyG?aY;bn$d)hCVowd)hmRUIgpksBFU`<i2W%u
z%aYOgu08f$h;uY-PsLznUiBueTMV4dqzLFP*r(A0AQ)j|j7wUDEjA|aHoyba;eNne
zEkd4vG;1+}QH@`kK{}T65ZagRmI>_mK_nWHToK$DNE3i6C{+>fy}OysaBWo<rQ2zM
zc?$~d76?jx(GP*boeL)9ClQn4=&k+*5`7N>LheteJ1fIZ00=&z8bt1a@R|M7VBxhI
zgw#kw1U0wtzU}c!TWoVY0tvmP>Z?Me?k9B3nTJeb^iwWlI>5I6>9eg$D;aQ1(jhj8
zBS4n>LGTwc+DQaFk=E83c{-8XUm<f{x~M$di^-JoHIMhTD<mpAVGma5$i2f^fy&;a
znkLc4CdJ@sw)4T^CRgJt=xX@4o_%%b{%Yf3Faa*(NHP*JfoSR>Tte-mzU>-~=ISuI
z%I}h9e|M6^C1<tt-wgl%GG*imdRTwT@O&BM2%djJ{JjyTTnUI5{Plcs_3I#Z>0SV;
zytJ&Op&KkK)&O73Q&f8Z_&T>fbhDJ~{BTDfkbTis{Nb&L_t#}A0*YH`u%#y7luK1G
zzMvYstgAz<9IFHpbY?o{PjjG?VJ6gM7>Y{{=)X7KrLu>7^T3`jW)j;vW`fzzD!jjM
zzxRK1SIGaF14Kke&rIy1mgzToz`74R1P9@H(BjPZ-I<5^%pheX?&**zeW~Um0p{yA
zgj5e0lU0uu0UN&k@P+ZPr~7Ib3PuIG@bu38FUj|xk}+9WflgnO99dEnSQej4Qh~3e
zmgdx6vI6l-2`5z@ZXPA@UQJ~jKq#6jWby%*2H$(!*fD3<aI6Upt+}~I)7SWL$~>Xs
zQ(4mbw`o$@-{sif-(Sf^NVa%(J2=%1%uoTEgdJNc_AGN3M;>D5`!Q_c{3#3pYX#+}
zqjzXhx*lEhD;RN`C-Bw$9B^1K#gRCZDn`%#ZP+IG4}-JEO>Zl3gt%(qy>1Ew@AH_G
zsCFbWoH@m0n|;zMmGHy3_0{lqS!|rlOI4umXSO}6zXCF%kmskC2W9$R>!nm5B?wB}
zdQssP#M&z-K~4ay^)v1uq;2-mjF5VRXJl+oszDKeQ`}S-W^8t*Ylc53{6c51XYbmd
z)FeMJRV4}n0g_+U8^=*51)6C?1iKvqgfrXO_gD}n;D<~F#1Oe|hiJnuK*#BL!Z{07
zjPc2{*L^5LY@z#N8!_=*mkncwdh#Rv2}}+yVqT%dSXJgUq3Qfp>0~2#L8S~%uP-<A
z&`uv%F#$aB?5?mXc>(0p=(|Hb&B9Ip4&`(zlPVyv4Dbcno~qEED0*NNh4*%czf&S$
zjpFPKlzS|nHR8d2slgneU3`X4+Bw!W{_v@+;|+TTf~KxweCFQ2c+m>J)66svN}^U@
zu90?b$@y->%(&3CUKhUk$|bC;Kx_~B?!#mrpvgW#io5Wq!Y``(+-D^1pYW_k9DA2m
zYHGmC1AEIy-SiVB)zJ_l1PkH&orT**w#>tQ+(?i(G7o}k&N8HtPwYy?q6~H2$6o49
zF!&y@z-%g!s&u9Em6qR*L3BT`jVN6DP^R>iMU``d!_cDmj)wy2dszj6f$YJ})x^fA
zV}nY%>Z&&U$se&dY`b=myiW0V+sgb73FaXK#Tf~+WDKHZttQk7;OkUVlz5?eE9%7&
zMUn4|m-rv3pPHqggH0{9YzwZLvUA*=f5(Wdb)grWU&^h;)k`Xu9awgl&TN6^=pFl(
zQSpU~SIhAk$!n*sEpJd+T^Ho|kR&EDBUV~a0s27i*E=uWURQBky4~QrsY!E>M<hQ(
zYkUxq4yw>wjLCn$ddA5EyHKc^{CZQoc<Ydr@#XRS=55(+m2t7brj_)@wLsU#s)?k}
zqYt4rwg#C%sFOD|@9~SLZ>4+*HqDLG>tr5F*~0(nfQggLXL&^IQ*A5P%lxBg_pOoq
zaQ?++2yEhBg3fdOVPRQoo8T?{9L|gV_m40dXzaiEuoBksCr2)2{_+Ay4OKcJRTB#b
zDQQ(y#RpI_Qan&8-Wre%>_JSI?h3FHvz|mTF<UWnBi1Ml_!+sA7=>(2IkE(KsX6TW
zAZ!QLkQ$xq?2=#ukqoyn4J?Dl-oZ*RFJ&x!hZ>|v&!nv9%q=vMrly>P(3Au>E9V7`
zS|H@TL(VT)FFHykRrVeojvBZ6_Nx{p!GJQUi3=*4x1kEf!X&1`QD<&&qB-y7siT}J
z)Jhb@Hv*LWdhP^E>lzPU<=J<(j>{o*Dg7*?`5U-SniZ5_-)1`;7cCG^Doj3xDTS=(
zwP*@sc3y>L3m!~9UR*wtyV!J4q1e*zK$XPq*Rd>rCKskz1mU&2d5iqP0X||B2iIMi
z2wjH=@oNsP0xu8C?{^JzKgf==7`}!GCDU|d&Nh(_T^3!c7dlKs$s1^t9kutc(4(gq
zy?K`sjfm<j`k+WfUG={GU12xDpC9S041tF#^#=FYTh9u>qh*f02BKT8uE6u$f%2W~
zYdcL)k3@$hgue{K$|o9d4Uu!LA?m^$i58uSZ{U)Atj;1=R)9smxL_2#+H~x0)**qN
zc8wz~XGmw+H-u+5kQx}_gZaOnl-+t6%f}NQyZmJE;dKw4{8;xWhlkq-+tNB@&1D(P
zMkZ)5-^uqQbbQ1EWhI8oy?v{8=i<p5edvXQyBU7lev#AA_)a_n_S`0u$2UJh&oh8H
zp=T2|q}RHopuZ_9$5e0U;LPkk@4M=GUx1IJ!0P-E9<|u1nIs{Tt97dkGOot9kIy=<
zSCf*QEO&yUI@qiMW-Ivyr=iGuyo<)4v_Wv3@HK^FiFNeDwqMW*npLug8G~jOBO*$f
zU1D)C`RG5T(0R`&jFp@psZ`1J5kV`h(5>qjMJZF1!g1Aooqb{LxHNSg^2=eah+ia2
z0bQm=cs)5!ST{L7Obypk@8uPnuSqTxcC+BlHz(-EYK5dk(nom9*kh8ecCg^WN#h#L
ziH$H~WATmFJpoExon2X?ot?9{^bCM$MD^)l!6?)O<SEo+d#-NzW3cLPED1jVv|x;-
zcT&e)CO+Rnl9EV4F%6Y(a8V?N)Trf#%7ts>i@1|-U5@7Btx%aSA51wAYu{ofWwIxB
zdRb9$b~KHTosadN`eUPpV3LcO2Iq=%7>tXNP(xG$vsLF!<I_aU4UtQj6rUsW(>Yy9
z1URC5=J$%;Qfr?0N(Y{e%;~m}S;E|9>Od(Aa=z;rRcS|=2hyZDr*HY~GkRO)v+n^h
z1rNvNhtwNq-%Sb)g6*Y2S03v65@{vlw+{yf77DbCpTI>uv%c!inDy46x9#F;C7hv<
zloe8vFx*4AvpMHi@2yk33{`K!>@gO*(WV)#IkAoig6vzNI6A&@z~~lu{1)p_eUx)l
z?5hT6>RQC~mE2&((8Q!NK0HlVXOWy{`VF8QjECZ|!6V$+IY(2_QgpMzs6l8}QmYuA
z8q%A0;NQu}Nl$e^iQqSzc3SxN^73CDa}j)Q`s_Z_7kz*XG*Z7lzs!U&0LAfh13T}t
zFSOsq-bRFx!$!mI;Ak9dCha_rx@N&Kd+J_!s9rVHN~PtP+#ka>A!zfyX7Y`YwgAno
zpX)Jhjr8zvIv^%NlSU<d%#%kiL@Z)(`8i{BZ&v6mHo~M2hZAVqpsW)9ydJ+H`TAWx
z_qT0x$1CD8p@_@ikT$-ISyr!o%QD4EmC^-t(s;e6iVOkyw)iA~f5ug>*}p`Ui^k~n
zaf^)Zs~P9hw%>Ca>9)URafzCre1#;{jC0L*fs;zI-nvy`vS8W67`?CZ3}>=hcXeq+
z=`EHS+pXqp3rF^qkJlM4PpR0fm$OERUb0+nG_+ao9qKTro$2iz6^}U<;q4G)_dy$v
zi_(i!L_e#~L`Sj42t@?xEe*ggCe*8H-yQ3c4t}DjJ`H&jO6YWx%1=>CPS3e)pd-R-
zruU3%sQo|;X{rI$>!UVJR8;a2*aRi?b5<(FlxlR@=OdY$N#*k2^{8`Ql|CP)(u+{E
zib{wG0!ahAjAv0ge-L&;5q(6Xn#iH;b?UfE>w|-IHle;T*o#W&LN@%@=ve>Fx}9wh
zg>v98pJ=gKd;Pg~TuQA1Rx8TM@gS_cBAS{VPa3Ni*1X8={sNFYb}HBAlHx62r#CDb
z&SfXJ-qPK&xPTm5OeXE#*EQ{n8@D$b5EnnjTUySMAt8~k9~?4kvZ3RABL*MQe1Seb
zzqMt4PI$m7)hZ<2<8XA6tZ;cwg5F(-tk+ULpQv~DGP)<qCV%5G&g+~oN7;+mA)tSO
zdZ9SSHRm|mqM<_Tg7|<hI^(tOcsO|kR;O7o#z6nUB}82_?1$ERJ*YA`Ab-=P%|*Ta
zz9r8m9liNGA#K@B!Y30cbJtTvMpAQgv={<}T-_GNzYDSS*(B}T<g*<vJLfpn5xboe
z5VgJXX6BE1Jy4LJlky>!{GU)-u`t9|BN`z`BHWEMX#I$|D)g*YFM9I0y7iI!iB>3G
zq98A?W4XbGmsKmA{B|~Ec)ZhM5O<*wi*u5@n<SWTN-w7fN<vR*Rm`bRgf_C8n1zcc
zQkhnnn$M}~T1d0J``E9rhdl1yx?5xowM5#{`WmH~9CiXX(pG8?<?~lJmKkVHLzn-e
zwlkdri3QF|gL1LJPmJ_Bu>kI1A{T+_yRoQp+3m19vC56N^}fG*vFo(^tnNJvXYR8)
z(rE}E;o4#T<k~b_Pwi6QGOp5+)r4iZgmzUg9Piv7Y>eH!k8&V8c`oCrd820_>peQZ
z$g4WY27dQhz?^!vBh?;qyJHt}b(b2oJC*TMeWtgw9TG@&Qbp?wQ=scZ>o#6BZqe2}
zenS~c$7kcLSN%?wEz{CTprX&1@WaIe^y<1}+_N%GvQ-P_Tq%sL#6Ap{&UGeReunA+
zzdfw>pmbn9b2I{bfb8)=DnrRYvW?{3b@j%#-La+NhYF1XJ#1tNtk-n(vUiXt?e_ag
z6Z^F0X?=SY04Ff?-A#q317@t|4OqhKvY7logLX3HS4gja_a6Pl>v+UL`31*6(WRYc
z!6v{tEw0baAH#}@$~#Ctp;D)y?-<Y|Diz|YE)HC++29_>Vp6Upy(<MsiE^5aCzF;6
zFd?}kbDfozi3~$xpDv|&uT42P;hUWW<150{bJ8HgzFzMbzp8W>$9D?xE~W?E>pxQ%
z&sAb7^FvA=kC4vNg(>fu3-|rh2-F82)lvq?ew?7lLkZ*`lz6^~)Gi6fY5{;zOuNY5
zYgw#bu-ROhK{2hZ&v5F=MNhC^K)pVlDRV^hOYOIhCW~<*=s4MvUAkO8&6mF%33>Wp
z+@nkAfmFO8Us77a!uuEAX?Ql7(>rYXOMoQ2YHB{d9(U{sIPwLT_o!t>6g8k(8yKhK
z^}&8T`^nzj#Sq0i6XE==_e!&-3s>HJUrm<?QGYsIns%>+YT}INcm-gE0gjnz(i1`x
zoSDV)7iR9L?G_bmz@alCobRD~T1J<7g;tCY%SbksBzutPJEK7O{_a#ZS%sBv7jxTe
zW(-HcKrUtCf#a}<HOpGTL0`Kg&t!R%(wf;rPEQMoxzYLeo`F@RiIi5~N_!|q*D_OX
zP4VPc%g=DMa!pG%9d%4Q)`^#$B-)_Lydoc<oBJ%U@V-`MyF87urKxIMCBe1%)h@w4
ziX@fZ87_MD4%#mAc2FKn)Y<x5o`xl4R_d>T6-Ozv`fCa4<f{o<nq#7O%0*Pqaab>b
z@7RM*_-v<|lHZW&e>uCe0kj&0S^W#Nmn3o%+Jna#Gu&tp)yQ(S)I*#<TDm)<+ypGT
z(!!2~VMKdvE!=a@BQf1{-XOV-KD^{uaKGccX$!UVIA_5*6xyXoVue_RIgL5QRY7y5
z(TZ6VCFwRZ9L5}8;`V@Gip$LgRAT3p_tAZ#y_1BsKGf1nBzd<D!CypWZdOYBRs~Fr
zTUeypUZw`?g?XSEFju)6zo~Fy>Sn?^MH5V^kdlQ{Y?9eUEn$eHA|=h2agn^EiDy>N
z=7%d#bQIjtgOC+xM0IBwczakY3DqxZd970FDR!nPw`7oeK?g4vqo1|f<h8&*aUR8L
ze`?-mFN#`~mQVgncxffE4<)<O*$JmyDQ|KEJcv&Vq;(UAbWA`s%4&t0y`L@<>O|&y
zQhp0~<mWN9NW80U&8F!P0E~rdRyBes`QEIw@U>83zCB+6Hafypq+1Pnt*X!F76_T<
z$8=P4T`Z@xle0!9iJftp3^c_>Xw7Utn;hE4d<K#}*-$#2*=JHE=eY?^ES=8l{UST5
zVImgNGM}>5slNbuw5IQ}U-}(*BN1<|OlTtRcI+xVJ4)r`KLZ!>?10W#I0~!$q|sM|
z5~zgWeD0o8%O<xuxP<4F{}!8kT=}sjxT};Rev3#Tph<#NjoZl{A%>5vr;doNULn2$
zF1ax_;72hqA!R^ILG^VHeWnrg)(&lkFuo5b_;?5L{{|#IEE}G(+&R|zGpNS-vawL+
z72{jkBtCMAnAfOyJ!~_mKy~9K5JU>oCDpNfEW;^VvP0_Z%%MHxZ^>v5%G-fUeQ1Vt
ziw0`g6XZk8Xt%7!*zVr?M-^l@Z*dS_7lXps3C>ex`fz-CnE_z_+Tjp6+eXv1eimz#
z3u*&({Ae7dUC4GAvCr^<Mr^NwG8Mtq#333=WeZDy@zZ#{wP^#4-%?ZTiz3b%wvtOo
z8XW)=lw4d%Y~t0~VhZ^zP5O|W+da&CATs@YLi$uFXM$+5u&VaEFlUj%b3Y$V^1nQq
zPEDd(WTU;}HuF^>%_RAD|BVAIsyQfv=o_5}{(D0WIqPT=9}CB7JN=%WO5zBZZ?atE
z|Mqio1=Sj3bT@eRMv7Kw57=2>gNEG$MeN&s=>?uqjG|9&_3%Q}LxDbG^07SSHhq`!
zGUt#%y+4GW3b+y#J7+Hwt5`ta<?ZJetDu5^Jh6nnz^+Vt)xo=%p58{f*nxIrLP~jg
z(S90uXs1Ma@1;3{-3vP&qh2zy@5yT7qDQ)QLV7@mO|TaK>(4L8DaO0tqct`h@maB=
z%SJ`-_SIIrF%{9D!HtQz>6c_qWH-n2=C?s7qW-Jqsic$VW*qiB8(gR>DZd_x$k(3E
zAm<6cbnGYV4Ud?v2a4HmQ@f(*X9=yAGZNl*r$yVei<cdvUhylhb-R*12ObP>Q%8CS
zM|##>Clk+?OAdZ?D1{1aoKSOQOe_l+{b@F#7EZQi$1ASWtTd!@u`7?zZ3LyqUPhuM
zf_s4n6{goDk2nSyhRn%X7~8(f%I_?m0L&-At1L`EyJ_71_VQGmz=qZWX9^>n2-X*P
zCM0GJi?<)zIlL7qKcqhph2EK)u8Rn2rni-*X#b>3W#wp~(bwJhkTZ*M<THbfN%g17
zpmcL(Fg@{L>*_L#DIF7`Ua3sAx-@NE{Z9wolSDZU&Njb0e^XD1!WO@N>}Pq!W}Kn>
zI+chWFCHB#^a&^EZ>&njr<8lg=X2+%D)c_O;SizgOf8}7zV&*1+BX7gb!td!^JQuU
z8@?z?G|}>NURJUQH`=A=G5PHod0Y`$XBr2okH*KZ4RA{Be{K|keS47ksaWS>yWk6v
zJUGuJOMzVQC$YKv>YE*G=2+s#EiWM7EG{N#q4eoDtcaN5)l0mb@B1|0mvV6SW`Bpl
zt_U$4?#q$U)9~MHSI!5M&_RhuY<&gaxMU5My>Q9ZC_4f+bxUFLmh^@14BVdp1n{hH
z;VCR{?!~B{BH_})!>G$Fx(r|*Z-F?e1n4qc;%%@5Fo|rlUG0WhdeQbEj#VSOAT;mx
zz~)f`wPoVW=Xnl`MH~b6Iao)f9Qs84lg9}xl(XTvcv}8=E>pwkl%LOKxo7QdY9Biu
z6<eltDhO+T3CW5QeZ@+k+0|+GPm>Y=h<izGKLa;>(M*z&=e>m3m){}ndD(uHVLzDZ
z9Pi8_H-M(Ws9-LrKd7YXmRIAaK{Rhm>3QhJ9ZqXGr6eh#rB@y%<$%(59pY~ZTYvBU
zy&gZhX?{nJDuVrzLZisHelcKTb0|~k>BvZFZiCxz^X}9;u&CB62{Bq;Q=_2ho?@&E
zO<y}L)Aez%(Km^mk9o}{um{&DO^=DnI-OhexFAnff<*`Q>}(R!Z%S=1CK9g~+xjqV
zBE8;>4Nhff@NJaFEJ;iHNdpYqvM=R=X3#^B5ox`4;WKDqR4%_NoVfG?x?wDXD2gs~
z9+kbPhMF!!jcp>4f}tFiT3*+RFUSm$UtR=${wnk@BlFtJn8Vz%hRBu6cM+k}TGV0g
zDYADdiD0vQWA*%d(}qVZ`*_>CdD9U`qZH$#kNFs<N$g@a<%^y{Y6UB5N2BlH&8>1^
zX|ttZCf@iN(X=CMZRV{*^uk5Ut6h`HJ9B0)SK%&L^BW|c5E&m;M3=MnQj#e~?YvJ2
zCG{CD{K7)Yr9$kQJSpik@Q|7C19+wl4yVer%;)SNZPy%`ft_fZ_wf6r2AM4lig6B`
z&J&Z;EDj06y82dSW{5XKEENWwQ}w-U{~J2!IZ^BhD{)Rc$L+4dF&1m)yIQZ;TGIQV
z+g04|&<54m%k8G(L<fps8jchi&B{fbPo)pGU`@S^B)oc=dd8R@BW`BHTA^oltS>XC
zowx^af1Y+IsUFmgDSiS-S@sHd<$=QW<-s4r^Xzn!dl7Cea1!0lyKtZ{3IrF}B2#4&
zy5)-WvLgdYjjvl4o7T=IP`Ujn@CDgCWJ#1>lI(hmXaC+lR>=t&o;Qi!iP}8UIMMmK
zg-!e_cm%YX2(2(a3)um)kK}S~Q*>>dqQHbo@8D|NK5>V~O?sD~ffH??-c#I|9w{H5
z!JJ{f6$sT%B9K1V%7QDt@cuO&t++>7xuiKmz*|wn`0}&n=uT><vM(m+3voe~CE!vO
zi=`|=Vj<O!*dS@Kt)K68364I5E-w;uI`A~XR<~q?<#2qu@R^|r75^6R>*w+J%^65q
z_3jsdjG|ueK@l)QWW2x55k|XaX{<yxPTWZ-k3GJ2tn+@seu1wq*}%O!(D~&)Dh%U?
zDHYn(+$z)0V83!!8E=KAs_(COL^UU7|IpR4SEwbGeh^xvRAUR_8Nm{G5A&qK^AjXW
z3@#shsi|U24d)8CT-S(ocx8bsQw=SHek4SLTA9kZWuGDv-rb^xPB`epD~ksG40Q3)
z=a3dEg+@8CCE9Z$TY&1OVFDI?PI*XYaI!9j9zInCTe1XPYXK_ZRB>eix_1s2>!w~b
zi!POdp$TRv?q=0wl>6h1I<b$-;ra-qT+cEUm9<{<ZerSK%z(=#XX(E16n~2a;|5pN
zhn8wPaA2zU{KS<tCT{X@`7?-q0&3v6%h})pb2en<V}sZ>Kr<NnqG{;l3pNz#-k|@m
z`(d{&76F><uJPryuzj^puB=ke4HO)L)u^V73K)0SN(N%y*XmureD?N_F33h~%w|b}
zW_gPV+HC;wI)Y;nS0y$xzqdTG`0W!V5fEh=8n`>=Y4h|?PM5Z-rK}uDz2hIGH^BAe
zo#v!-wpB=w>HeGeQJUF(R)QNfink}a9E-X?_iC&0Kg}tpFPDI*Hqn(v-18G=Rz{g)
z0;5Ry=)?m{3mzMzou&q7=*Z!>hgc$_LqC$WAH;jmY*1~igDIRTG_I{h{rcOfuTZVr
zbDg&+vjK&cAin8D`@Yb}x9B#q(&f+3+!!r0w?#jBv-6U_@yR*FR&mVL6+G|<ApLyp
zMa03a{2=>@x{Pv3Dgn2dn-0DMdf&J!39T;7?r)#Bd=iDs$iQ`E2NEyt!smqMRbhj4
z^Yz+FkN}!Kv!sFHgQD9m(R>E5Fkn5DKW^3zn}})t({7HqT3JUuYNLXwiOkNoO0}sh
z14q2R?3GqpHow9w^em(f-%KnEObW9`cRGJ+?c6V~`0tFuW@^Q-#MAcK-@INNb77o&
zJQxy83kIj3lfAvOviXtj2Fo_{P%7`e$U8!1P^!|T=(1-TM|EB{7HylA)A8O}Ywf0q
zp`V$)CoL<@Cmkpyc(LJ&oJ&dlQN@1fAKNkvoE~fpn0o;V)=vjq#i=^?&sdZ&Jz>fK
zNbU4FOkfp`9|$^Ogcdhz=pd8krQp#0aewux&ijWh+*^`7ZInY{M7|aVM%c@Irq1G%
zx5US$RgRIBANqJG^WJmJQK^j8sVjD*1sny3?P>N|9E*F-^1bG61qEI-UhiSEw-CO@
z-B={G{n-bhJZbmvxG+hYcXKBJov<RdeWIZ2!4Dh_a<XiVy^2J9M7^QMjL+b6sEvY;
zNlBDp!==c%D2yy5f=?aXVZ2~(+~~(Z!CP+-YIJ%tm0?1tYZUw?cc>?uguk{F$F3lX
zkxLJScqhcAj(+>lto!nzWEJMoyxepKL8H^k18S#kg!@<INYi(uLG)DAhIvVKZ}}xL
zoE-x*Kao}g?<u*aob()eTr#mp)RW%wkr`dTxe)6hOyu!2)WYLA15+~NOP~959LIOf
zWeJQgBiJqPivp|9*K}TJ9}Ejyxwl4G4i*y<<%^sxihVSF>mC;mWgqh_G>c8&hhMwn
zerN5@h4xeJbU+Z&+AWw=s~OcCEj-~?G$?q~-6XN-o&aTBsqWjY&X&TmURN1+<lvdC
zJB`)EYK^AK<Y_&7#KTbC3%%t;?d`ar9}>3QZ_)?GZ-$<7GcY}kaOM6^7xeRo9i<4h
z^6FUb=wD3h)uV(M@|9x_OJlruXzjscBP@8;$6WGQX;X$_Z-sKbq;mdA2~`#!<~<bh
z0i4>s^=Xz)Us%rDQ_M>-VFX!p=~Omxq}tiHM32f~NjoFB*`(T7?l7sWW&{yyH=ZR{
z(4!6a-(A99bjgZ58)z=AK3eZh&h{~yr_mYHm34aT4IaPyxs<1<L(0aOZJY|lfTCx5
z#Cza+ormv1m8qHIM09Of$uSROg~3;h_9wy{`>{7_m|vN!9P$m6^Nf3_wJ6+pt9V>>
zg3{)?Ez?n6Mhf!jn|(ENCma2Vvt^Fwy?vOWImNMKgXzC63S#9axD>hSVKrPVM$AZb
z=6I;t@CPz2V;xqWhP&I5jmh-k>(^yyGxkzI5wQJ0qRf<nZnl{1MSdzF(>p``CaSH7
zoBUF=z+}!ENgb%soD`f3OVq(~r#|o8KfyC=TC7?>Y-^Hx<B-_Aa~uIqyFn~+#JebA
zTbzf$`LD5Zx=h(MS2K2!I4kPO;Eu)xj%6!nXK|-qHwL?FwfENow`i`-7>|ryP@k>J
zS<o~nKd(bMd%VHl&ai4G$~5_PZp*1bwff<<jqCxhD?dg<z7`GT0>y;nt~lEkBj4%w
zI?F?a4y%720ZfD;2-lF4m!9WeMv7-F0pSOvDb8i^(nn4%&@d`V+Q)GSXLJlz8*WG!
zC(P(tG`T8!=D2T?G{I7``T()NT+#mk&VN7Xt`GsPBy&Zb{$Kug2bMUDFSck*he$eQ
zBD_oW$s{?FaIBc=BL8_8%DsNru4txf+%NO=s8UQ_Ai_B-VANHIBSt;Jb5anJ#*(`C
z6s!NlG5(&G_2?=kTx704Hsojb@aJN@AoBqdvqIDpb+YBkD`1l`W?UQGOAHv~UfHd$
z9v7w_`3Vc?+wSVR@eo!g*1{P{I;(Xn;M~9Sh`T2s8OyrV;Ln%~<`El^^sd&WNBsT!
zK76N?{89;6umbH51{bt-Z35V$p8<{S-TuRL+Joum|K-O0=fv*dh8vuFv}g10-}l9v
zNG>}OUvQyUU2OpPF*XO$WslZo-zh)*``h+QGl+R~1-|&f*O^WK@V?*w;NM<cUk1-z
zAy=W<e;OhGc}ai$<O(NT*u78cCVzin|Lvds{`px~Sm+*#K6Lu+X!O?z_&@*28R<h*
zD`l6(_s28hfB(2xQJ_g*W@xMUWu^1GIR8JsIHn*;c>S@EM%*8}jlU>||M{Ux4Dbj!
zj+p8H@@E&0ss&1nEu1(DfL{4}{~_v*%y+7vksI96O=8|UH_g&tvIEE`E@bb4!`78*
zSO59pKcYhBS<nG3{lDhfln7BnFpbgqQI}Mk;$BTb<HoR>Zo9RjD?e<!#g-dcqE)%G
zV_esAZ+d21U%j^d1>xTA0z5xZA3!+|g%>dcu)_-xpVqV?Ksh%ASj8${=og!cLpEW9
zXEcxBezJlmF$TgfGL>I2N_bTP_fC;*zqJ2hfraArCg|K452qt`fbRa5FNlknZvl=#
zV|dmCaq@=gmEOub_<M}}mlq>Z@Z=YS`7C38$qyty{2N`?{DTrbH{ti>S|G=mB5>jk
zm|KU2KuqF$Q~@jBWx!^>fA{sAqgU;OU=>4P%w)eLzG&Y-V|IuX7mgXNf+9?6|D2Jz
zoAvEV%U01})LT1**kuDaUI`#O6-BxEFD|x;!bw)DXa~N5^>CS;ApET)XrKHS#8MZ5
zM&}u__X_*h0`uMm?p_(fnM`!>J=`;BO?#h=))c_lJDyM&BOVt6NbZeqw$mtc>w4;n
zZd&}`uS&nBas@w3!4X%-o?n?zQ%t^yu>(;pxPuQuidOH-LdRI*NXoBsh3!&~%|x}J
z`tOj%GuR-kMjNQZ&q$OV@kAVatwGsMoM{3P>cs$^ND$S6vn?M=7IFNClv>UNi>Gm0
z<WWf@y?MI)0R&$W2J8oVu-uj%uK;89T;xVDH%$PbNSQJYOe0g|C}Rz-kmgj)unt=w
z<l^~!?;WekH#0t;NSK1bN2hJQFnOm%6G8(&@pSl}ylSzRf*@eIX|UE~%t0$kbE{Ls
zrqAggZa_2)guKuB*+07{lFUR&ZZlk!(C$LI7e!!vUK%E>5kUM9oZ%4<npM|5U$7Ee
z;fOyA1%)M?SZpb$@fk**Vlwv1l~*g8<K?qDtJen=39m<yV2A|khXBP$Pja&)_8~Us
zj!3SBY<d&GDWCmtnTNiu((eAR$AipY!NJ+`n)i-V45<&adldIeQvDednBs%EUwNn+
z{lByfBMERd-$UAEX>E>Kga?d1MG?wCC7_*Pno4+mKEBAzo}9^T6cu#3>j~j?3Y)3J
z@aIk}=!&iWb8Y$GKXGJ4epkD82HVf2x`W&o$X5e$1MxfuPe1cnde^xO^cU(igbk`1
zw;XrwY1#_}j+h4k_W9@Nt8GugIyF>t;EP>a#n=|$OEri(p28*%P<=r-E)KztO;2&}
zWL{AKrTFewBSearN(PXk%xat<B0gS9>UjXmLY6K+#fRVQNZsR}cRmO6x0%ByC_j4$
zYUaU*irgJ|@h#30p28MB-V>PJIKx0LgyLHa@q1z@T`^#Sts~UOlokXKK!}(^_bKI-
zx51p9Wrk@VyNw4-P1MmY80X{;cX?&beHso;>k4lACecdj^Bo+WQ+SFt5%B}0Fu!<d
z2t~Sp>eOY4oz($0;kDz@EgwM?A%jWZvv+?#Oc{oiz~{MI;Mx6Jm0Lx%s10VWMrsiv
zc1I?8Uv}2CXKP#bQy%Tfdd&rV&LBkA3gepYQNPlCIm|8~F!mLId5#4Vafd~_cFj|{
zd2xxn3=xV1rwwy6jhlq3U|n2?W%Q(#DXd0_^E~fD0Fq@;B&(4W<7s-y6J*4OHY_$g
zvusvG?Pt5JOF0Pm`u-Qsv%w0hpe5#AmJDL6VpU2cV)ersqBS2j^-_yvgg()2QU9SJ
zz9xqc2)lFJZ|N%}A<)l`AWDD5GsWtaW1VvwdnO2GMce6B36jA{12Jxn+={NgLH2o@
zsJ%2rG2Eu{pieNx><JalIJE)YSE9LE@rhb8>!}9YMaqA!@HQx5=h<#h7yB#2VTy%F
z&H?u=3bY154SKx}gehya5iMh1pv5;VRIOZg2R^sd3CSu&t^pv`SRpaUVh|nS49nB~
z4wcdZ_S?`(ia`*GA;#S={^TFa1dkHhNS~Are55>^|H3OmYpUM@=Z(ZVAnRPs*32nH
zc;<JY?noD%Xc!d&yks4R@i*d(SVDhY8Y9~b@w#&4&OyRAVa(Eqy(vFYk;esK=*@#H
zAaG3X{Wow^xU1whb+>4#J9)Qf7l^WhFezvXC{+)j-XQ2z9c8eba?r7T{4+kii?FPk
zAAIM}OUSWY2LYtk(8IqM-cZ7FRF<ERNs47<22>{pmkyu!X6GsNAjJ)O*@dr9Ts@6X
z>DUM!f+=VX!UM940Zrvzf_aGL@hwUFTn$)7sig_ikdS-X#NM=Mpa|3=_`Pm#QBn1P
zYj2lK_QT<s9hG{QTN3&(5S)R?cQ^KJ^%wX$L)wMf{Cz8Hf$l!ri%?$Xb{#C#eAm)O
zjepBVP*TySPG~|WC(_Q3E56G+5hV~Huu}Pc$99;ic-$^+7*W{lS0j?@BPNpU(;4UP
zLa|)yRooMN`aD$l)ZT+{JRZ@DcE?i9N9B3!K70B_^hU0k?e`h27X@HDd~V{rLxM!?
zGgIdB0^1{i7eB2HNoxQ){m#al@>Tj9x3V&y^3{I<t}56kF4IKvT7KTHg+L5j2=@a&
zWtaP?1(sWyC&nEXE+y@ii9rqd5H#uu!#$XPpok`6j^O=7Q^$Xt*BjZ2K~7R*d0saT
z8J3n~MZX%Te_tFudWW2@to$|Ue-=dZL}58$uL|zS-TuKYRE-99Aw9oo@H{*$iZels
zMPQrtlAH`@eBSO829cx81BaD2bnL)V8z+V=g$$E%6>^PI$%vUS5>l4ZhnYXMA&f(m
zzFE<p8|LYW&IO1~7$h2Gn;)Ra`1&}R*clUB0SVmJ;5kOZT%b#fuK8*RV__&|CruLL
zD5;QgM2>saZ|UO6Mj%|5vk}$8iY1YT19n0850CNsyX1VJD5O4kS8SZWjLCQb)FurA
z;Wp&QsQJZN+{i;h8}$%ZWI1O!0Zd>TedB!$t0>4arSn=!pA2rx?vrd-0eNBps0%q$
zZt>=qF?_0rR;@V@^CFpX)<8}a1Vec_DxLX?Tx;q08w(5&qrgmQQ?=CaGCac%GlC+e
zvqwE>%1}u~p>7n<GV7y?uOJ$WFwOv54IOH#`sOxHSQ$}6T|dSp(|K_H=)GSrYkXmW
zl9d;OO1W>3M~U0LC`c|h$StZcb#8D`LhpIjTg=`bQM{ay8cOO)w)6lOgxO`8q1iZ;
zSr8E0Al4ZFW+$14SQP|0mxDPvx2To-(0xiNFHux=`vQUma1~iHXp^dP<?ty^kW)B{
z=Ew5ThqqKLZ~6`6Iuwi>5WYL}LibHnTAdN3lLm?}9lFoq;0Ckh(8uShS$ErkKf`!b
z*y@y_!}_5+6jtcY&986v5Rb|EMm>pI+$-lf>@SxnS{imdU{y=Ud;=Qfmc1z-K{+xt
zi5jZipidgf;T1hkeWZr*c{J~vSO?r3;`rEWUIM9JsO?Tp>3=)V|NFVA;}y^n!~Yj8
z@gQ;8*C;3v`#vU>9SYNHkwe(P_Tf2mqA~9u9NhO-jhH<~tZDuD&dbysHmadP^;rs+
zE@<9`eaqHdmb0~4rz(Egy}tiI%!vos6ic<-^l?RW8EZ_{;9Rp{nKqaXD>ATg4yn62
z#qpJgg(LBpT_xXWsYFunr*ex_-C&D1q8j@c{@Jh+8>5J}t9QoRM(KT#LtprujoU~%
zAr!jN==N47ef+4>0B@2Ze|c@BUZY~qwCeSNT}wv237qh$*(~SNLwbGk=!G;8%)EJ<
zs!(#FMdr05d^AVQy)I^dx(C!_Db3KCOwdb_N1L<8@I^FVG>N&T!dC7PsVAg>_F^w9
z(xMi@Kk6O5ac}XaiNW#SSj|N|s^QhX%yEk=-9NdO-2(XN6AnW%pNc?hQaW(WN;=ib
z)GQdAYLn3$6Kc(4n0w>Qr&0Ttr$!)k{4L9yDE;{D)_p@jmZW8)%6M>4&18G?3wS@W
zm-i<k_SCF{kvuteIUDG{2Hm=?#XDDb!{Qfd_46|_vXdpyr{0{L?cBQKpMLsgw*<8t
z>u~+Ib-6;5?T?>T$c-+o5yu2fUaI+W+Rpc^YftZWgIOKj`i(s?5~)1sxygjHI`0A5
z#T(GPCn*_m9;#oFxaM+#!MXoNz>@uAu@~8Mp4lr|IIl<nImW+Vz<yQ#Zen|O*(!e@
zUL|>$Ix^(y@cqO-a&&SkQLj@Q?}hmk36zt5Vm3S}QNV&`8fNnO1~t$^5j&HTOJvnu
z0&phFR<K~$gLaLqIb6Jld2eP*>W!$G<;GwOaHYw2^J4$Z)c&~{)>j5k>}VquKI=FN
z>qWarK|rI7uhA;cUZr`OBrY_*4<D;JXfS(;MJfavsL1e|x1LdXuU>W&`F+IP=DGow
zR*z_=FMW#@-YaTm4bO`thC>tfZ7P4sRd+2CpUwLM!uiXSCCfFv+o`=v*Dpyvl9|EC
z=4b$s6X7?en8M@S9h3>dp0a5yO=npQ(zX))WIlX8$+l#HP3*Nz%1wsiyg8x1+xQ13
zr5pv8L`3>iKi=!<8z8*yEM`#oX)}Vvhv9}GT&9OZ)2Mq|Z4sRB*p-(IpxFi8d({}x
z2ez~{{N7B4A3^sd#ngVYeUxg+^O1<0l}yo|JsqjaPc8Y_tEdk`2KKYIFTI{NrUu75
z@f9Imm;Z=9_!?j(!LIjvwbsw3gzu>2qfYe)c!}MzoiF|p!K(p3cy71T-Xl6UmK{{)
zIr#xB=|J1*G)~rg92QpciIdC8Ym4tCx(FjQyI2@yD&z(SY3{9tGhFC|lN^1fg?zHm
z^;(~)(u1dgSakx4^7)d{w;zAU1YIa2jP085Iugp!%RDAUYUJX6_oZNZ1S#0?#{?I}
zFbHH9eTSYBKUF+xo@bi3@Dt&+wAZYk4Qo0UPCM~^vpn7<^PsI`U)k4eC%Pe8`8`U)
zmRlE_+7U&eK{4+rLvhv@J13GBs9b4QIB)HjXn3T_%=Y{s??m{Cp-TyuxQ$fpD^6d{
zuDCof+Em4zbuo9CQ6~66rjS^cNt%^uThFPnhLE5&7H`I0k(y3{LjH!-PQjj8W%uns
zZ7XfO{B1Ajq-k7YOStK$Xfna)m^077d&U_l2^)4WgDb!qIqU5*;68D9x<me|!Kr<>
z21s@vhp%yql)@x^#%hAj>;{qPmmD39YaitYb&{fVPU?3NN7U^?F`OOOCYn8$F0o^w
z=RZl*zUjgBX`5gpt2uG03LGrLUO0y@2AS{W@(~lZNjsk3C)l^=YeYju-^K&YNezmf
zj6UaIei8o#Uz(~0aA|$?xkme6lX#2T*A(*rcJtQ=`tnUX#Kr%t0APM`SlL2E1SmQf
z2DjDJNhVA4tb1ezxqR3W(UE7atf^mbjHFuKi<J{0mz>$@7;RcBbf8cYwNG*-l3bZ`
z8kqR-A(PuCM4*p2v$cCNdfntExt{3JD@Bx{jee_syDsXY(jH5VS0_|_g!KuGzX5$C
zF2_RP8U-!FK}bl~&T~B^a+1O_Y)sT7#nrMSjD1$4<ts|b%@{)qeVl4(+QIX{C?sF=
z+FeQZ?->|<`<qeNCsuTds1|O=$YFu9!h!8n+|n0>V)#b*UaXq2g>ma#&XRc=S?+o@
z@gDM*eW?*~=~FhP7vz)}NFV|jf{z~hm?M0Dcz=;zK9h%A)0fMX1eLE=I4B8R>4hE5
z#ps+ta5ab!(b(^?<gd(05l;^2=YOes4-(xLwbleIowb8`_WJ`BenpxXT5U7(WF7ac
zC{UqtPdMZ*vdO@<yQPvlKgj+sF#3ONyY7g%zS<}|=`VTeDlcJcAS>}nVtU+7IU<ZK
zEkIbHfs&?JYU^H<t$&(ghiZ1^)*ka;fZBCX6Kp^eAR&llc@(ITl%H_v--2=W#a`w-
z1(z5m2iTQjuO}@FVZ?-76dK5Tp^j341#JgOa62Po(3D5U$XbHfRj0)QlIBl@tHP~L
zS_IES)X|~oltLDQqn=@M9%~qK%u{G5z4Ka*NbZLur3k0g87mgTN9kPx%8AsS8KU%d
zy<6Pp^NP9@keo0LTiY%EoH%KYB$pIKv^?J@EplNOXpZ^r)>_0qxoz1M^p=KZ{2N$`
zv6sU2bQg(6jUz6_o4d}doz&qMu@XILAw8euXst{^$i#Qf>QLvnux?nN_7N_G4cOQN
z#Z{5r_jjjqlo7^cI<e2lXJmPO0r^y)T($8B+8O(d2QL>HkgDtz)Au+#^Dh5db$4qw
z*${XvTdeh$F8^FsA0OZIEW=d;BQbM%|NOE4GLQa}15bGpx#_DKIl=SzvHNXt?!nJQ
zi=MpV@3v^DyPRS6wtHAZY?RICFIt>ejA+!gUL)O+DOJ7YDVMjYC)w~BYQ@6J#WCN8
zwCwjho7qJPx@~<#gxp%WP2E?Xf}1NFP%E6@-U}KJ_iPGH9RYw<3H_Z7&!Anx^Zlhs
z(fD-mN`k_onDF)0CFu2Y*^GsxI|t7v<vq8qY`)KnHlf||YqoxydYB!oJ<GG_3WL~H
z@)bi&hO#}sB4IXf+#DM%E1OGPb(_W?DBC}OX1zWB?L+v!ZV1~7)3cXnc$vBIM1K`H
zgy+-{>%@_<{&F}nu*IkkiRESN4UE;2DSi@#%(7b~=lcR6PyBHPI%;NmDt;httg}-v
z^5jTe0;=z%vg^s&BOA&;Xn^4g5$RRA?iE<&OAMR5eT;ZXosE8LGnQG6+pfE1kyM^^
zp!!?xS<=xOtg>e;V`WrY76Jc-t`E~HJIe{#H4=aZM_@B9p?zcKd1khKzDwVk3QDiy
z$@aGQ;!Gh!7o?L3_CGQ${HQVOF0X=AEj@ES&i@;F`p?&^3WR;9+@D<jsRfs0WbMGo
zErW)ZL)kkXxM*RwPXKU4o@Ghj-Q<jU0Iiq2r{DYu?7_IT_uJ#@D67eKG3Wt3z-Bbv
zv|QN*7s&u;m#p^%o(vyPj|aKgKfo@_6Yq<VCcr32%3{D{HqgXI#&DU_(gON6a)`iC
zYS}vf9IA&&v8!E%O8;CjN9wIPfc2uF5>;g9?4@MB@v-wYLf-S!aJPva29Rwft6JR2
z(ficKfp_LkqT2qVxsEwn!-{Kx*!KzyWVuf}uDaS{;Z#kB4=s6IFyHEd2FCyqXKTbf
zL3-6j`H*SIMZIRK1y^MjI=%#~UMr|#yuq{P?xW`TJ{rR7@HC7GnXEeF%#z}`@7;V<
z`?j0;zy@%s(t;zfPD%)Y50P&^!1|mSiuwKp&hf7Y5uN|!`8V4sHAilJ=QTi%8VOcP
zd1Nw0zvmS@m4iO^&k&6}G=ciX8ZPGklHK1YW-=f@!}%RsR0FJ?GHcrHY#FKMw7tq!
z31L#)CeHyM`{CHhoQ`n79+x<+UPt9Kiy_7~(|~SsCdxX}qn2HpTFJf1Ng*Mtr=l}l
zf#dwr{YF=Z7GCTflASiHGV(pQfZ?PJgxO1O5d5@U$@1VRMU*q3apTj0lm$oky(F_}
z?Edr=kec&tNL`eAWR>R>&b-lnbqSa<Xxi-4giNOk+%DjWeTapkR31*fL8>;;`<)2K
zmF*(IKq`A6`Dv~Qm_0;lDH56|w&IhOtG>VH@uIL{skr9#9!ZvSb@Nzv%w-PtT6BLO
zWTmNazeVQman@YOmXvViZpvCKDgd_NwL^BjYEk>U6)xlaF^QP?0T3Rm<<^|#C={o3
zwJK~CJmuL|@=mxkYTW^CV{2ki?6&i56AGOkwR#sY@;FZ68*yO;GWcRfIEJl;OK}#B
zfpD7l#^r5_f2oNQD9l(ge?k_uW3lR>Vnw!g_Hd3FW@*gJ^w9FXr7}40u~e8A;~oLB
zrxDCC21j`1D8T#DwSy?)5IMKDH}0f17i~#}fm6(+^7?7^t_8?}iuQsa4i4oAI>Ti!
z7^e%rju>#CgJbj4SBFJ%TKNOb{L+ms18i?2h2c1Av+H{Cn10e;%vpWoa2eF{O|Kl6
z<l@J@i+gi6L=(a&0?6Z%sd9Eio2O`14vJ(Kb?rx?3QKN+38)3ATmv?u-Y0(u2y5aW
zQfn=xm=229%3Vz@?9Vdzb6vb}={jlr<o5$5|8ssh!u&x(E&)r*M}=zA^?*)HzpCOH
z6R=aV7;wfXnV_j6ZB`t?wBNSn9`99)Cs8ddbpr_T3rQX=_qlPiYU%z6@cSpY+~yI+
z6;(x38_%a#FnO%dZE?-FC+gx|x%D0FUWyVvC|5t~jn0V+^6AtZ?KY*74lMJ?JUoN!
z^ySuuN$WhZN&pG(H;eg-G;}8Tg35_4k^=ktV|W;r-XQH0k%h8U*LsYDMHdoz>97aY
zw#MndH1PV+%CX=dzMoVMH(uq{Y+0_Ddm>!8SVhX0Hvnh|o)O!6Yy&vsL>G2-XkSsk
z*Rb820gw(GGf*^%#;XN0ih%oz%*vGeMBU@^IWq<_?m=N-c*QqcPpDpAwF7fWOMXV7
z_piAf0s>2)e7a*`EdpBk*6!`-5`N{w&?JxLe-g5LKR+rsCLJI8e6=WgLMqyIYdlsf
z*=sAaoZJ*$(hmeLBS59%bHVjtlbh6au}-_lue=KIr$Hq6F?jF0jmQ$<Bw9o7rxdzY
zBPRg=5`5=K!rCy%mfAmA7g(NS39ct<1J9um$lnlDb4(w>-b$YIZu^xK2{WJ*Uwhn^
z7n~59^*Q)(f-B!hN&kvK%)=HLu_354T2e&@^&DC?>aNts56rj#Z&1}g-bHO(cPyMb
z#HsPK$;wph(j>*@R<-ShidTahfWT6Z)fM^)FkcfTy&l{N`mURP2ac44oF5fx=-zkQ
zA$(Bup0(%^?FiRAdN3Ax-UrCN_)^uQJk5963(Mci;liweg@fCDhF-FE4nR)`I^1`n
z%Y(*XN%<7j!{yVii&F8=WnVqq?)^OZVV_xpkJ$yQ<n$E$<leQ<isA1M=p7DO8gp$}
z30pkwq4ja|)nNNp<)i%aqRc{tLYwQJOL74d6y9T-#lUzUfb2lpc5?6zn2&k!?Fo{K
zyi|=_JQ&>UJY+-lzFO#Mqie6Vgt`CkVgI-7VafN1Dcxd7x!YguA%KQ%wS+_x5J(cp
zV=k(HNfjrgk&wEbmIJAJ&fr56Nk7B2ua)KPz0)5kzRtm<KeTyd4_=~vukvv#Ij}h0
z-H?{#%zor^<Xdbg^?3OHkad^4)#Qi6-NJ5%d5}|yZ5e#~(#8hLk$2?tq%apqS<l6~
z88fb!yWT&G_1!xpcJykaVbNZ@#|17~HDT&upUK<8&Xd0pmC`&rE~@<8oS~EFvdMSY
zO^?}+LdM2X!t0vpSf9X(uff!fUZkp%0Q4MP^o9V|IBZAY1_KQ@tE4#Wc-z;2@>BK{
zTN&hf$LBn>1n=kg+*0H6xh1|#f-=PJ+7KH~LETH`WN%wj3!)R>a61#UT@}_ZPz$e4
zgF<z2c+)I;(C5~)4-z;~ckIvele_N|pcFX#9GrbLtB(<;W4o<c1~eIzUWh6f%B4na
z>+TFURaveLZmwLdmdMENU3zqwh`oVQKB|}};XDIZbFA8M0`wgWhL3o(#i06HG~cA+
z1jn~~rnbRKF)sw6+o`CK-H2Ygu74v(=O%e&cr(SV6+Y8G+qRD$4V0bn?j+BtJENZ)
zSC5I0@1}gwldO@NcvAG&`piy%sxFjuKh~>#9uaq+rf7nI28*k3+l}S}+q5?E>AInl
zg!Tdj^i|p$+P5iyQs?SbS5|Or(~}^M!D0m(vv@P^_Lp<6uL_UJ({mkj&HFI|y_lsZ
z4i5X1(Vj?%<p)OF9gmAGExR|Q!dJ2eyvB!~Zv6X^=U?0Ep8n8wtl{sQ$^6kxm*hc7
zsxyw`LXV|HXpr(g#L^>uq{jV?2jm2Zt-oEzJNjH{uri}kzG#hklPZCE&0<JL*ob7c
z@SWV|<09MK>zmzp2PfqyLUc?SOVb%?evokA&SX4BrR8p_<vaE|iaG`*vDY_=f20m!
zBvFP;ci6m-VX!D3Rh)Ko49WF&*QjmkbnDaRwC^6*at6u8QV|$Ue&0F3>|ax?6?P)-
z0L`r1n^#YxMPm}*cr5oM-j9_^i1S<BOM^4V_hKUFhEoh~ELd<LFg=Mp$^8}{<>{e?
zwQxh+L+yN~lG?{C9<#k_R@VwwvNSzgt$9A}Hs%dAxfnY=Gho8n9bs}rCGaNW1ZCB0
zb?=SE`O|Y1{~QO7X*<24T=uzdXRv4^&xzen;*(w<b|=Km=vlU|R@dpi+8nle*uLe@
z)WDc>Zi8`~Gv@`Stkt+6r7t_rYj=E~<ist_>x!8h!*4@d4K5tmOyCS`0c9FSDW%_}
zJ!-DMNh=hpYyyXQ2!#zYW7s$}i{5ye&KvqXk<QAS-=<OGD!=%q8PCvi`|=$1?sJSu
za$vkx>ne*?%s;T4woU*~z|*R#_v)`r+1EhfQ8YonVr8~1Qix09Ij$%Y%q`KY{j)zF
z`ziRah~uT`|8=ZRlm;fu<SmU{WTbWI0QBK}LZ|oCAJ5%Pgq}gjz!Mmq$^Av3LB?Lw
z8-N&w<qBQE(n&V{H8kT-xdFgMFyys8&r)s2GgDCUL%A=OR10skMnrFEbJYO*+5q<~
zJTE3R7dWekrQF6>@vMJ#+>5}&mw5X|SL=U*=6$6|m<SeN6Iq91bI^Ntg)sqXLxcTU
zuvLr1_ixTfZ1}luCABGmpWGiwD<X&j_Y#5j^gMJTLj$!-bzW@J3R=h7)ND@V>Pwpa
zgS^v+c$Ev9^xHkRvI~FD*6NU22voIPtl73ru)PEoNeQG<77gT){q+ug#SZ|g`-6iU
zx_-VX`ZVICX6_TU{aYjY2lzM(TC>_Afh9jrVSl>`{o#j_=Q&`1N;nc0|0@~z_m}=Z
zzW6tf&0m=IAHMRx{I5p|2&<{}fL!apum}GLBK$n!{PiAROBke@sVO%9$M5^^Z|i79
zP@{>xX%qi{Z=nD@?`)OMU(}s{`F8(@>zJBAX4%Rf)xWpBb7SxYrQ0z3rFdhYQUrk!
z>!V30$uqwtY=8UfrYI255_dzD<tOdt(Z*4iUsjE?AK;8-4aQjLp}rRXrLi|dkn*E0
z>@?N!xG(?Oi>^t{QKg4Moe0}n^*D4N0}!IqlXkamF%rj#hf_&nn%VFOx43-~E)dur
zX>S8lf$$}49L5Nyc<ps3+YSS?U6Bq=2^jATKuf+NjjST$*W&QE;}Hx|VI*$gpSyAX
zTx~|;YoRZ11_--LW(Z(yCG(*CLNWqkN(8p=Gmf?k&z|0f{PZ~pm3|IdmI3I54q7{z
zOfsVpwb42>I|iYDZH(w>nYA(;ZxSY6;NW&|d*}|WYa7T0${c>u?iuMbfnlWoik?{q
zGbtVsV5iOlpCRN06q{|}f!8aOJDm`N7+lzV{qoVdj)bEhluujiL~FW#K9F}K!^*L~
z;`haZr#8+Jd<5O5!&pLv>JAG}F(K&%3PbUp(vdZQJaQ1d0+#ImcnBhwf>>SNS3(lg
z`=Nd^gJVPd8Sv3Hc{3%w!O{`ukWx_EQV#AHU2U+OHR*2+V@b>}uXg$_rr1*c5JZ;(
zaDJBc9Fb%$Hsbc#Hpt((E@3||Co+?f%D(v8N#NH$^0x!kR~*?6l4&fReo=tk_WOd~
z{f8f)6j5!?oBxlq_YP<(-PVT@aYPw~fJ#vVp-8VHic}%=CP+6Fr8f(p5}KkYL3%Gz
zlqOX=(nUoG5J0+sf&$VJDbmYtZD*W&&OP_e-0%A@M1o27yWe-MXFUZx9v>p7+{1(S
zj6MPE_Qa|1-7#9S(=1YBrKcybkIVrcpuQlQ{OJ7P^S)XYS(lXS!5()e&!FsIZGM_p
zB}*VZSb$#mp>GA~fCgZMArY7VqUG=J#O0&N2HUKhll<qs{9Ybm+GF<iynP}rPEiN5
zxK^P0C`6Ebc+08g)7qE7MU#<j4>`?sk8Vtv|LrH<gPPPvZk^08Iu^fforF5jEjb~E
zqO*V~yac14x*Z@L4I-ICNYtp~!=w6yc7%}V2*?zdjPU9d0Ln#xM7~4r+U~P(<{{A^
zPH@WwS33JO48jdvt!`@!wmMEg$l^J<0))C2f?I?RkJDvv^i-W6^ymsCTSJoAS@X8R
zMzRbsIujcWJdLCPI3QH7R=D21aYY8w<exzXYLE2HHDJE1`1tcrLR5Ge7{ZE(Awf@b
z(zB+ZZ%$>tF}{^6cmv7dm`NI*)nq?Yt_|FectM1xK`L|{1iN4cTyc9;K2TCLmMz}+
z9_XN!KCPtusv6Xk4W2Oa`05yMmysn|x*R7SbZ!+GPXiz~y|TTz`VwKeMg#aMZ;24S
zR2%g@y7XLF@7~>lEh1!d>E)|!r_xi3W?+%po4vVs{CJJh(uCON4&Y>kNZ9Dj4TM_=
z#IncY8kY?nkN{|gOW*ncO7Bz8+eYXw@V4#Pj&eV|c^L#<qo(G}ONsb5n}~BUGVW%W
zIafX+Q!;^&qbi6AKgm(6eZR<2PsCp@EI$a(3&_=A_F4|>cn&s<X9ilO)FfgZNZP%;
z!TEYjCqvp|>GFn{OE-Eu<|GXm&JHXzaVFOnX8<$qb&yRX+<_lEvW=yK7c5)T_^XqF
zjx@?h+-R>hC)ssD)}g^|{)Vv9^Sme8nIDR`RT2|o!IE=y3sGN6l(agex{N>t@OYDT
zH^EZFvL>Goj5U}?))SoCzctRftQ>sc*g9q;4SuqXE};{&Ubb4mE&G(gA5=sfW=<HM
zz4I8lmVu7h>JeNoT`$qIsXc#dWovUQ{@rQ1W5}K&wB2I1L+iJA!t%xU6E6(bcD{pS
z=DYP0D<J1Q7Juzt0cBO=N1I$ONwf8~HG~mt8a43uNB;XDWCv?_kec!5?FbtfB?ZWe
z^FBczfw_*U|GxJe?Qb$;`M|lwd}oPDggwkDTaH)(fTq*zheD2L)E5rn$qTIkpLpC|
zMtOPt{*jvilYq~xQ6jKgt=@cq5sDL%&PDVs@tgSwW_Ig5)3w)i$*?$VK_x6_`Ci#f
z*_7BTj~L~=SrUb0?%%n6@H4pWh`l&%O^OXY6UsCxA&KIF1i<6R0RgAH-9yH5959J9
zuOCZyQ139%jUCe2e+~8}@Ky`Zlc-5|YCPehP80~_TpUJp_4>n*>2U>oQDhxA0l9Vp
z26HLcYL(8$&h91OyIl@bE%`^50>YeARNLmJ(!V%tWHE^`cdax*PCEzRL*VN$1K-~)
zPj?!EGsXCseQ#bQ&@Io^(`g0n)IvNDMz(Xp{_DNHhY8-ggN<d@^Bu-NZn;V%Acmk1
ztBG#mmCoBg`d%Wvy_ombTKRn3e#!(=JCM`Fdmn>Rtt!ZO73fq0AZvSIC}>D+_@rDZ
zv?0`{c=MqbB@Y5%A@+m9Hj8r#x?eVspgYQEv-0|1PWWaboUHME<PQ3V0h9Zkd1smd
zF``?g$JwnS6CBXY#+d-&sqnX=3}LOfL8TKnu^+39Osd5s&<bq_20Es<FPzLf(y)VA
zZ`6zS^2Z9lARQe)eDy1M{nEN*M9sL4;d*cP-etF<aNCKcFWI~2+Ji%41&D9@x29&B
zerR*ABieS{qa_3qR+iZ%vW0wrHwA+YuT(Qi1Ia#`qQ&%nai3hb$!C5IUb0@2`N(L@
z_r#O@<W)jF2~*zn60!WyJom-7=xw<mkohXZzc+6w9!WJ#I+>P~0c=S!`oe2xTT(LT
zk0|7wx#whMdUL1Ud!@M3@O#|)JE_;j#imX9MuaE1+F(&o+Htm*bNo#ERFuAP(+a@v
zG#NQPjfOZ-^evpEejT@7cL$Xwx8p-O;SJ{Q2;k?-5b&L}TJJvMQaWS%4{l_u7^q-2
z10NdyG9){BsxG8yi^~(Z6H2<wtoA6Y!><n({<c7gxkr%2ZsP%$YnfKGm%VWgus?a4
z4^S{Ufq4Y#N-*<7f71sgO5JJT)flBhjrFRW&dx#ntZ&01G61-cBSFRpvv_h41xCKh
z9UrtrQ-Y%v!PkY70(*`{p>7W&!J$hN%Q($;@H0|4VBcfj7=&>Ikgbg}@1BdLT|fAR
zeBJjALtQxY3bptcjRa8s2eJJZj$tdSfKUMjj`_9jMAa?ZcIsVhfjjW&<gFROI=LqR
z+oL`V#<9&bwy~G6+}TIO5H#o^&4o`yF4BRt*Y^9KQT6*~Ml3(zQc&K99k|*iG><6g
zXkLjyHk+89Bn)5?yOW2ql}Q0E-4BRz^}_7T<HtvdA3p&gFYib+=<IR1Pl=;cA~03c
zs{Fp^g)ye{gtJL$XS0-XUaewT#|olNZ<vM3FWBR;@<U~{pA5N7<uYv4Ri_H@D7#t<
z-0>{T-LLuY+<h(IFv+sZj_()g0;gnoO(Vz$z1qemhFK``W_dE6`dDNLNhke|os=6J
zksg<x*clo6EIG#HE0ZRBaP_bQYNiJvJe{)vp!M;j0Ycm&wqs9V1+W{bnrVzFzexj=
z(dxcE@2GZn!c#3cLApEauhfj7Z2}GF%NtRxu`}2-s}o^?*8pcXynDIEG;ayj2*Iaj
z42x&lg;L^$9}yRFcS*%i#EB&N6H^hIjFBGYcf{VpQQj8&_-gCbTGo1PP6z%@-W0?y
z>xhG!z6!6){ml{Z9@QXtDu1GE1yRPKII$cVkMucWH2kR|{#RRz81C`DBOAmezs&`Y
z;Z0NQ4Eq44-I>)|B)Q@&P-u?~0+e}Zp|Gf=^4iNcKX0JiGdJvRnQHH+m2}Y5k2M4G
zZ2@8}-Jg-2zkY`5CZk>4W?%);t6_kYP_wnzBCq;2u60S+j3EHwLjb|;*sA<(0a~dU
z_TfXQ7kB}ek5DVi9Uj~Cr>K)usrgf;`#+bOXvMl+>Z^ZAh<xDZhq+_-jRs&K7184D
zWIW$N77-xPboZR5fI><C{2T0|t!$G*=@|KCXYuE=47(Vq3d0z%`4b?q?_w$)Wtba+
zjw=NZkICsj@#_CA#5Y=!9Jf*^pD)0~vHPV`zOhm^3@550AkR)8*4(y0d9Eq-_^FuO
za5Weu(EB$BbB{cra?*aORH26c!dWq<tDm=Po{Y0qMceV=+3Uv0wfeoK%M%!!T>aVb
zB}rnNCP``nciDHJ$*hMgH{Gqi?xK!5E_q(p`E~X+Jplub7%~U+Xo`ySP2mdjs7a4n
zx3QElRs)j1{R!OoWAUkuvQB`(V{vbqlF>`uU+e$ivzZPSjDP+n#95?Rr5lY)X_XdR
zHm;W%c)!o##H=00!n-Oh5@)X>z#Mb3X`s${G)2|fpS!|Ww@hZ&2j3Q72)!2vy+)T7
zl&)ioCQ~%@Uq?*MlPn)@x<M-=yegHg@9Zx+cr>3R+{+RY%;>OZ)zj+c0x$k?%UFt<
zvpC=Fs_rPs^xIc!o^AKzzM7^7C5ct%#-}u-Mu}_JZ)^8v9g{9+kHW1vT-xgpLFM|o
zV8F0M{8`hXr1-NHq-dP1x_~@K5`O)STc={obPB)Z>6xQRD6}4bg%GpDnVAqJT)-tm
zzu3*F_Da@qUnQJ|#B71QQ5WvJs(|*Xjo70x*XbMt+kHH5#+|QFPAr!(an)G&@EgUg
zsiOIX@B@8g;ZEnFU0Zq{GxW0P+9eef-{URK4wA9F%;O}3ZWX&Rk3)^cNri_CGb%qf
zW$0WvWTD~uOmJMqu5q{Kij##Y{>X&e;o?W*R0$P=ZkjGZ&D-l1KTQ9=tDB%<v5NQ$
z@|UG%t)t#T@8PA84&gB<{z^$h5KLk!ZWT^y7Q&~cjfI&b49)MULT>_}2jy(AxrE8|
z+VqW;rpiRJG2vY^#!%JSkGYHz(ffqPFy52j)c4SI6vbD^l<qZ!8pa?{r7XyRosRuj
znwZw}xM95&^Z_gIepf-``7x@u6uITdEU#gXvAkyq*FT0m=Bts`&^)3&akUm-pVMq&
zl`M7sGKntagI-xp70EwWJCA6p@G;d$!WgeWW+^Fk-m29exOe<+l!`xkBh{NP=*tPc
zh&+&ELiYNAb~{^JDN1gs@zr88uN?$C=<U<*y`c5HUr`gga<y&Zp>eN4??ElUxP0lO
zXV7^he;{%e9X`6Y$NHm3Ag9Y~w)TB*kxYbrrs>t$h^a9JSICZWZR$vb(lZV89x=S}
z@MEd3%qX(3a@3HNz3F-zQ}W|jx}@JE@ynmaI`~F;sylxR35=6WzdqHRL@O1`nmOK*
z@BJf(X79|W<Us0h%u*$Qp3RjS9*@nn$L>6~?uGEj<?0koRi#&bTl15<2kw!Yj9pNL
z5x4r;_S-y<G&Q;`MQ*vZm(bbp_^k575*2dueGFcI$5rt`tN$eY{rKfV;FE;CcO)R~
zclJX$N<LYGuDcgLDbvq78%>>~7bT^>#a3eJe-A9VQF?G{5|=F;b7Uax1zQd46s<~A
zyw;wC{F$@}O%eJBtlKu;9yO~wsdv)OIST5T-uCKbtn9v6th%9QjO~mJSB*RprV!1y
zPl)@bO33B>Q18fPAUWPD6gkYH@fis7j8l$C-nIL4C3~BOgpx*?tuNFw=s}XE&SH3W
zsmQ(?wE_dJvAwE$wqR7PG?-}v$+}_j@yk!sj#)aYs7OJ~4Fk6&xbs{Tbqg;#qc&=c
zS6}IWNXMqYYG<dytTrX!!s;*zwh+gr*j4rnDsxYxoyOuzgE>B3iyr7no$NW!_l;ji
zH(WFPmHoSi1<DUp+`mYzd`K0&))U-XRuO9P@obvNkIeM`@$7uGwvdip*N`WVT1+zQ
z4-W38C9ACT=*Oi2k@-doyC(E@=_X`+=^p&flJSXjCHXy@Y%><9{4~>3J(8=;+^~fS
z$tWABs_G037|k4+<1`Vp!>drWopCs-lPEH5t5tf2Q=&b!xWOf=J3Ci<E-Gc5qR*vJ
zLWk2sm!y1_v=k@W0J05*G~0CkG;bMbo=5t_wAQD?tXki>SOj6_uII_AI)qTtw>R5!
z-@9z|GT>@h-mrq_b6e{?ACrzHp5r0G-uX?zoF-?6e@GTxEz6pXy7kzmP)WgQiSJbF
z5`~tJ#<WcgMN3=DUGr~lk(x_)XSG+$x5Pou{NF*wf1T63_d@!rT8q@^UwxGxInjvH
z3p|L0ee)Ds(o`ZyY#k4C$Fpt^N_l}!CS(-R*cDj>Y!ynI-Bh~E#r&sF-vZm&aGjd_
zq7zwk5_)6^kHiVAOmh1ILn9Cmt59pn8s`%;VT*yxvr=<TRnDw3-E8h#GAnK5123jG
zssb)5hekfVaG&D9v0S$TGLC)sj~>$ee&q2H%E*x;2WZnSrj_sWd;0p*CyvO^NA7Un
zom}t5F;4`$%+>HtaLkWyuP}soxF*+3M7!+2V?76`3%_ru7;|r3lO%;KTEcyKaxz(2
zj&MOagVKwVO4$GyJJ}fK$mZt+K}xAwN0+K}rC>yU*c#(|Xn}D2yjrp{)o|u)R3C#z
zO9zS5BH*^%DLX+tmU|G&Vt(Fkt9h051Ox`uFyi*0c8OFvdphJ)|BEB4ua>?(Y)!o!
zr*x6mvMZM?*2Y70fHPCQcr=wPt?7`Z2)jv|u-V8bNjZxW&Lz4L!O;_6D%tkQrnw}&
z^WX~cB_l+-L$Ti%X=3WddbG~V|LT&JnbR0C43WRb?)Bm}_W<@tYYbST=-Qz%z2~Ae
zwk*ilz3QjRXqSX^_=>J@><RTap7?U<YS_e`$BN@j3}c`2i4nRptMR*cOBu;oXorEw
z5{Bm4%sn-_3IyN{iedf%RnL%n5fpW0TvG{uhl`dRh9ON}mp;!QGrfD16h6=ww}WJz
z;r=_oai=pzT22oj5mi5Juz#Dg`mA=Z;YTgSm5y}$j$v|Zc`tKwe$}`0Y**1Tp@OB|
zI{kdjW&Zj~qHlr&$&~S-Ap@$lxO4<s9d8+a)cv{?nn@4hNv!yTTm1eTo%<Ix@dxfj
zQi)T?QQPgq2iS&#vgyo*Ac_n{O<;-BfXpEwUk28So_MH(L*zu?wRLiK7J=c|*E5P;
z&Z0~HWRopbid7{?qrz+hy*8E$fBK@b>TPT(rX|2gsD%*T@ZkrCaaS&n_t1^EhcI8F
z+YW%l4_~H~>UIC%SqW(#xtt{yo$E}6Z80$~5jpY@HMVc*tAoYI&CX?1M+sa?_N!*n
zQtb8Tb1p6{y6>y2-k?Ao^xrybNu-g0U~D;vx)s*%HM9s?TmA~mP57{O!e~EMmjsOV
zPpZT1{SF@#D-!`0OEt+pa1Yl<tWKm4kbHGXRHX0XZuRC-e7OC#>C$fgHBB;gHj&Mu
zKD*mTW#Z`l3wZVvu%b`??IF_3_Wn%^ea_@Bq>|%6_3qB5zaT#T1b!^+P|xt)Sq<Dd
zWl3X`5me*=>8(O(eT;#+XT_JI8BhF|`d91hwK%i>ie<7@yl<xK%M7g$)@Zlx{M-4r
zwD}f@mQBDo=^if?@MlL03pY>GYCq`{bM8z}B@VJ}mg1{gg#DF%sAMJ!&Ks8Z#=KXT
z4A&YN(f#tlci=nMr6wTX>i<|+J{eO-Yc^dzat0XI@*`oabTP&}Np~vP5y*&+UYj=N
zY|D!UtDw;n+b_bpjW)2xD~PS95m1aeKgG)|F{Pf7DF~u6Q4JEFMJWxI^Qa&oGsFH`
z(zTu$M37nDV3n5L7nyVpcCo7o9=fkOL8Ad{_<CoKL4LheeTqw9)lq%JXOj#Pwad3h
zTAt9KT>Yw^8dytl<BMC;T!(G)Q=!)25V?o(AI^bTb7XsO@~mWZ>*D78E9yrAWY*LZ
zl~UknQ=gta29v%P4ztCS-R}6sSvvS-&7#?&mQz3^wjT+17$J8p_@R8Q38ocEZb#}E
z_6od-KH455rs+8~SD*g6$fbX8?XqV{6MziyQYQvm+XnH4#3cy@HkU&}i{A{mXN_O=
zrS*A;66+OqU0;sal->~@PL>?780lc;9csJ~voPwxGW^9>hU0G$W}jY8^E|VCJ--HJ
zmB;pY=l=M5mCh{w?1I(B!Q5&k2i4HuXsC*~kE#rWyN60hI-82TJ93FVDlG|87S8*U
zn~4cMxpid|+}o9Zqeg$cLe6qES<pYujUu}~rkPjrrRyk+I}iJZF;`0HkR<@TOFNC@
zN4jg3`Jo!Rka1nt18-0xCs$u`ir<_A1CY3jtHH_$Sz<Xy70>3hE)u2k^4h2DgZH|V
zz59>t-<StB5zXR-DHmY8N4T*UhGM*$nmfA`Hi^E(g8rXtBIB97-=l<xZ+%N~jSP7c
z_?{O~WX`g@7p7(@vZh<|6<m$yEB|~SBitd-qr;}t2FaY^Zmwk%9a;QM{_6zc67Wur
zmZ~+{6Z-p-eTel27IyF7)%WiIECkegy{KJ=2MBY{kLn(}f^Uc)tyiky<G<s-HLiB-
zrO}}KK3S2L94$A|q52qYP0y?inq04nIj$KXoHOMDT>m-tj@s}j&i7)Fs2fP-^}g1X
zU*NJhVL(gvB=IG;5}I!j`UWCSUJ9=Am<m%e`BTBxM7PL^wf*}M678;_?D#8m+*z#B
z#@@Ry+qhA}{dWgEzdFoJsLm!m7WnyF?{6KzfoZ-<dy-TH@EF9uG!8kk%di(he=m9A
z7$!bQ@Fc^1&FEghROxA>1-au8Fv<#597-a5_jM}<0X%0E>upRI22FE;mv?*3Ip1H<
z;mO*fGsc!?&vzKOk)&%2g2vue%rbX;frkENibX<uuW&0_*51S2OGwCAcm&t!j&+(C
zP~Rr4J$piVbI#w&ZfMSHF;)Dl+~ZLt*Q5)HUjBnAly2C!awY8Hbj~NVgH`EPNmJU4
z0Qb(7i?u{JWJ$0d5#e<Drj`(G9oDh~#A!xUw<mXloHwN?0zi+nTsno&A-jJ(eQ;Qn
zz@!)7X-h!3We6jfzI1<IryvZk20}3ecfEMGV?Ws3QDG;uPH&4pA&!D^Szu;lRLUt$
zqI4KP=OBUzQ<e~u)3<|(R+P7p5L|^TA|b-Y#z~4RD6-8TGxVbb<#_Bp3(qd4$v#TM
z7QSM2<tFT1J`Kt%Q)?O2!ukaFIJBCK+PgQR%ngjFj*JB7W`et_Xi09to3#|l+7nJ9
zup22-G%uS<`~OB1XU#?Y9r`b!_Pmc7<F`M)aolQ|<81)@%j$vDIO(v)H7@e^b%(W0
z!rp{U7fHrZ({y`I^-NL;odEe$iOb#)yVYZ6x5Zei7QN-~BX0Tu-<}N%4=<q+?LD2D
z=F3S*2~x4$uKatst7@xV?W*&&qN&vzSf<2~N|w^+DBY&v7`2c)Z(*W@)7~2QYAK%`
ze0g0t_=XCdSDYBeSZIP#9`V8N7M2)_Pkf0tMf1gPYj(^cOLo$r)ieK#iPrF-`GZ5O
z?tFv&>%k(Ow@-j`Z~7hKxwdl(<R4)W5&=NsyH#2<+H+$#vo{hOjKe_;&(qhJe*dT4
zicRQ%!piPK4*Rb)|6Yy)M<E<?%u--JPN_yDgvCZ+Ca&f7_|>oqa|?lYu`PmPw(nV%
zl)c5^vK;Ao&mrex^cHucAV_S0h?6}XGJ9lI2If2CVM|DU`PrD;q!<Ri%9xY8Vf~(r
z1}^?hNZ1#bpAwTu$a!0VlDOS-PCFK5c-<;DX{%-C1giBL*Pu8Xmx-niF1!iW<^wRA
zaRlYLb;r#RcZT*B^fL1n+dSZx1D+LFmnODz<+4}cVh1Bmfa4<%DiY||bu>ekEqC^o
z<3(#7^>1C50Xpt@2X*V$b5$_unuW&cZ1WfAK>XO-&vP=lPXuj_e|_5OIfZj!%$haw
zJekUmvlRaJ#XIH#>1%^6^Fa4Ls->^di0G#nw7_jRd+wI!{d<wAb49(afLffA#%VMA
z0m<g13hY&Z_yNfFuty9G5Ye!|NTr*dK-3X55`2m^bG<_HTz^B{KeW7WgE7}!u@<Hk
zffe5k%g=uLo*6wJmY*`r_wu82!!Qh-Y-&Ficf9(H2Qn&_dBVHgs>^qluHU&pW0)Bx
zA%Xlct|@}<6z>>fmbGH^Ti>NKSRgv|skz=+2L6@;pS=L0S_aNRl1K+`B|z-Et(Bq_
zj_fqOh=nKKvZ`pIX%u9@IxZ;Qq}h?|idaX1Yl_LlW6wL+G~25fEaB|ETO#2?2xt)w
zwsROizj4IE5-=r2@hM*Rjc-s_ogD475_A#nfw>Eni?&eP>Gt>Z;tk?4A?uCW+Gpu+
zx27Io^7<Gv89cLO8)Cf)+dUgkXZ_vS)z9y?;#Ta&`8Eovsum{XYy;KCJ<HT4c21$q
zTzTi?0Qvq1!We#+E6tKcnd`Lg7c_5%wczB$#@RzO8RuF)MrXhjsSj5hw93%vHqQtD
zzxMNGCh52;oO8&k>WW)sqEfQ?@2xOl92aruo`9ec|Jg69W;RKm6K{Z*Kza!OhVKJ4
z@9~jZeA+99{AvjXI)UkiZS=a}I3xYZR<@e>fST8Vfs3&(o1PmTYxRCkw0Q9h-MXA(
z{i;t$X7R<>susYLPb^W-*UeUiuZmhN6T|d|-z{?w^^k&YShZQBN2^d)uBdJtItU--
z-QfA+uN&)Mo5a5!zaM~cblL6AtmQvSNRwDHnQxoUw=mt`<!^566-c3CrRp52^VutZ
z^~x*Jg;c?C7Cg&gfSm9w*%}~0!CZ55<-2VA?4jy){#mwV7RrrqXH4zFXp!I-rj#z2
zogS4j_vo%HPx7+N6vfQ5^j%yrrd{lZ@8{C(ETuOGnh4BoVbV12p<*-(<^@o;HBx_~
zbZ2L8f`~GXwA>=n^tpS!<J!{)$(!JKv8#<t!Ivb4c1x;;Xe*UN@<XUV=3O4nC}|!i
zYBT9BfMn@cwduX9CnChKme|TKf==_Dp7H(9bBPbp{jFSgjrA>=(`3wQ&r$HcNKFjO
zp`q%;R)iUZmPlqtAspO}W8FbA-{c!E+(x=Ws`clq&y$tCi6_v7L|e0IRU_iw!iS42
z46HArALoq;KZ9y@XE~zm0czxMIBl+^y|uD6y1zv-+n{^zvxCCVMbo`f;)IRkZE7~N
z$R@)C<`810berev@^t=<<tYn9Jxy6|cehFUE!$9>pk2I_?#(AVr7-u?lWe?TDIM~S
zftJsiYBY9-!lGec%4wJ1Eo3xEYRWz&HTGsdoa!(}`0RUu5`V#GOkM{dzvK6n5`QJh
zz$IL=l85LZCzn$3=mSY6jqewh7nPHQwK_uLl5MFQSc{mio^#pP7p^+7qnK>s&b-d1
zCOomB^L$*JGIbj@rM#?Be<MVUNyKBtJIy*xXP_hx`)TqxrF)bP+UvfjROhuo?hr{k
z^uB`cnTUxH_HO;n`XBa9Y%NUl#?$-uZ-`n=F#jeb9mX&<AIPS2yZY;y>l3e2;^vRw
zThIKllE2q}Cj=_Al0KF-qCJiTFG!pJo{`mA73)&6zlepE?+vd-WJUNq+wy@lgK?!7
zGlY3MyLI5b{T=Kf<P9W^6EX=uj{I33PYRK5l?T+zvGa{drJYC_LLKsF-W^m^Hdfxd
zB^!9|;j-Lv1+;sW${AVF)a|FTk9Jswb`#i@DbAIND^@6<P~sz<YP+ey?@zoFg;o6a
zb-#xhlc?_ziv4DCVks=BlVQw_{<@FGWJmn2vEJ1Sj~cnpwEKX#*F#!mD>K-2j~45v
zppz0}6dVI&{s%Kk+3R^&Gqgh@vjo4=J0F^hKbm0Qlt1JUnoST@u43Jza}BFHy<g(o
zC7XKYfeO)%j7a(+EJdm`b~jaH@^V2YdsWo<+tY5xwrP)*(BDVl*+yxM3i3oT8`$*4
zi-+sQ4Ow1vC{5CBy)KPbj%*F?Q2m<y!%&}$H1t69zInGPtGgM0xS-RZ>61=QbCqT6
z2BXCuZ|t?4Q?|m<q0Y+7m<>KBS=M+oN;zx4i(GtYQzkP?8)dC}sdz3-DcSb6XJl{I
zkk`0$1E<RoZP_mTRnS`skQ%!qbnU+7ui~9dw9+nb?DHZ&AUQ`*+AX=0E*?=4RC__<
zWmITJX!M|m8LY<6+E~SO^%g9gg|g!{E8`mD^aH+J)yob8_cyx4NS`PRvrhPXCTx3$
z_El?CJh6IR^*jugOxv06)G+UNUImWJ9*3L=pPy)?JB7R_-6<2*^Pqn|v2I!Wl-g!f
zn4E8yf&7Z7honbx+231186zy_UOV2XU!0>q_<{|*KH#M8#_x;%gRUBSd=INaajwuC
zj7Wa?-ytXj*%^dlxZ#w^^lu_ISwBQFg)xl!iL3<PA;(g)zhO~xV+MEh+wU*z@Auh{
zJ~!0zZ2uqjo&WaQzerv8D95y*+}}};O8vL@P{Ow-Isn2Ms4N&eP;83zI?x{-4%%3r
z?rDKJ=%pu~i%plyym`uz7rE{u2kg&3J-Ms0h+-Lh5a&SQ`4U><*AP!($G1Q4A5Mr3
z<b?1(p8eau-G1=>P}D&{DW*OQ;tk%NinFAh>CL-5CIdCa2jIy#NVzW)=kU#cC^1>?
zl2dy3ncOkKSd1?+TF1TysmO&vpx@<PAFh4qGVnjL4F7u4)(IF{m#emQ-T$riB_<-o
z-V<sH(bkA*$>6;`dsV3&%zjQ-o<^Y7pirKJCYCGG9#SrffsfGx;U@w!Lp2^?2=s3(
z(O=E-Ur!n*Ba7s^{9kSKU;pxddhnDAv5why+(hs1d6a*B!vFTb)(iQ-uMX+${kISN
z-+t?V{n1B285Rw_qpRMp6qx_hyZ-B2c=`h%WXEkW-9!K1Uia${RI&i5Xj0IjXZ8Q@
z?e`*7%^!mo&;0obef-1gpz3jigE0>&t;G649u#>?H{IB;uKea`N`cqric_LoP8#cb
zRtnKt<FsefTp0r;hYc(L{T~0%PZN(M#jD(_H2F1_dul;>0yrf<X{TTWc&)}`^;@i;
z`dGyfEY=l}6w%v@gB-D6l{ZiKL$cvnMAqd2dB1!r;VC>wTA$cdTEwxFIk#*+6kJ&W
z?R-1<WaS!PZ~9sZT+wbM)+(hsK1)ek(Zq2AJW0-OD2joLXZtY#yk;>@T4dyO_MTZ<
zCBq-8{gNR7`prXtnSyTtuu@>=WFc2-SGqX)^RE8i_6vAkodkERCMhqc^4A1_iq7z=
zKgTQ-&OrF_Xbp8e@8J%ujUyz{79>e|+>TE{7>6%Gu-$D97Vlh_K-4-11Ep!vuQQqV
zfsMgi2ITSq`1d0rviG+j)Ik0!Ld0Wo=X(RU;MP@O3%ba!o^|g6K;Sxp@Lu}C5<ZSN
z8G`Ln5@HUBnI{3wR}8Wd24UK8eAE%R@s1?~ewSgy9|u^ZH=b1^Z4@GUD+3IZKDe~G
zp}sc*z7CB{X`k*E$*5D~2*e=%iGR}*I=eZLof1ZYn_IyTRK#8|^D3G-KT?O#f98V#
zlsYxU2_fSI;kzX<r*xl`83-i)9X|j_13vy?aMSx2>gjR*xol%8DSu}WV)5l?mt{|H
zOQ0bk?FLMd-#WsaX<AglYJCiLx<SN{k&CXg;nDGHAMBd@wD+w8cf@gdvSZ(#??WE*
zrO+DpL)TyfC9Rcwh(g;s(5>X2I&eMs?Yf*H0{)MFQ=Gm~Uj-0NA56x|GVcOB1|rq-
ztSyX!(2t2h7E=}M^?3&QriY@7&Rk;~7=-@Qm}?g91q&=U+kh=^x__YtXZKqCMX&pl
zxuEC2dPSO1#F8)v@9${Um)-5*-F}2s_2x5sg!)zZzSJmx%52}wEJ?5RUcS!w)k>i9
z^L?sW@S@bx(8hD>J{dQ;`}OY`d@SU7varwamV`I|;#8Z2)**~HX79Bt$|$(}Ue8Z`
zN3;=$l&}xyqVb$|mWB+qgBYx7w&tf&-9bFsz|wpulI+~odIAkum?Woe_DeLD&+@<=
z;!Vlqkb4EdydrQ=bVN0M2NFYp#C&yElINldTTvfKD6RppGVGf@K!Fzs6ZSiNazKat
zbNxU@;uX)11C0jBveQzqs^rEj>~0U;<hyJ-WzxuC0vsL3pTxyk0F#P<i7kMwTM)ki
zF<`6+XT{+dnS|PtU@b-W=zlwO|MQkp=YoBj^9Wwb;Me|Po`ON_)Fe`lsFKYlXc*_8
z>l1glqJLcqS`!<@)iVzgo-sbwY<9GJLJ6jbYdV&}F)F=m$??eohsReV>r%TU(Uv8q
zWUu=i_vnpnpM|a?u3r*x<2V6^D!|(faF!z+rHM$m-U>vtH;7bQ(FghM{c-T#=rd(^
z-+y?|gt~m#9<q7={4}yuC>GlHK<i`Xu*TPTZ*TNN^L-ur6k=S|k~N{jL%qwqySt+o
zGqnkX+*yClB!aTJWkb<@;yym=<_ZuOzoyanU|n$``amM=2t45_l8Env7f`xTDziv>
zF{*<w>4HS|jF~Jnm{G$Z<T(<B{qmMSLNeByc@qQ<BX?6t$m|W7NB-P3y@Bp*oGk6-
z2z>8i$Rw$3-4bsS2_SfO?~Uf8M~Ll;*bKy|tFT??6<8sgWAecfS2{`Qcmc(CNE~F;
z?mbAqRoPK&(@s<PSyK8^9*WFPe&d+`&fe3}s_ABhAKC06JU!q%30yS_hkNfb*-^8;
zPj6?vo}|Pqpn=WtfW79R-&Apl)@8NLUf#Aaai`YBuKxOB!}$32GTK?NuN*D6L>wn!
zMk3m*zRi$oH;e<BVxAJ{ey#LeUkte;vhIe?l(QdiJV3NY3ixyhun5gYX$J}u-LfLh
zUwhVlFs~L7tGqLh9<>NlIOabYlfD`yVy^j2M))^^XgP48l&}iy*arSU+v@`&Cpq!0
zcharYR*#3(Fij0kEiJB*ax}Q9<`%$LvmZV`dWl(0_Vk^u43gJey5cxVwB5SFAbx)b
z&9KM3ln11{UH!(=4#$dk1AUkP64FkHk1;iCv0?K{iT30*i8y$VPfN2za$G2Spta4a
z1Z80bfncIT>wFWf47rgE$<-0C;}pBmF9FV=AY#eA=8Ds+jIrRm*_P1gwgrQagG=2A
z)0*$>$0V{?r6*z{eY!az!T+FXk~Lt3Q<%53Z2Aws^Y@(;p#v+hbB1=)lp(Ptm%)Vf
z5XKTXzYC+=aJn3qL*1u!?tqLh8>EHqr`|<)*V|KCcMsk?+(d0R0z2BF!q1SZ5$Kdg
z)20FS0tN+3MiaAbiDBjZzEw3}E8)Ah4YPQIy66q?@02#*GVhoZIC?rQLZqybNvE<$
zdf&wt%y+jsb^uFLlun=MS1PH5Y0oT-Zw`h%BwBC1CJVctPIne+kx)wBPwm1<Hz8{4
zk*CkvJxj$;VF_Gs5!9p_yR#12&e}w4fTu;Gbrc9l_^Eutg-9+9s4apwi>@k7ojF-(
zmP1C6l^;&|Ml^y=t=0q<lA_GBwq>cYP;mX>Nzudj>Z=8nH9`Q6H5_$^Q3tL>a6}nZ
zM(;#wV6!DY-`jg%+{qV*2MlBX=IpKhIrn$s-xLUl-XwLqf5M1T2@Vpq{&Sx_#S&dR
zRc?bh@&4jhTVDmk=q-G)W*6$?pk;7E=(SMOKeJZJ8ghpVDc(>H28ZfB3rny^|4c*A
zbl_kN6aNaiWCSjs5`K#S;nydyuM82X0dNBXE4W&I8Y}~v%+DZX@RP}X3KP^J$k<iF
z5!P=0CRUz78Yr6b<Rq8cmT2B8ll<8p+u=6p;{G|wfk<yppeif9l4W2MHtXa>mHRU^
zRqge8VQlZ(roQB0?U{}jY9ItQZie6=*Q!*+=GY>>(n`D6XSO58qr7{e$=qaGSKlW)
zR1x?VSR-%2DmEjf&>e)&zDOWBQEd~3GUDwu%EctX65-@PwtC~Cx5{-<4d<*TytF7F
z&YtECt#;h{QQ_G1BJS&#)yU2cW(2jJ2`cE5`nT@qFZ}GU2SSCQlsi{gnmNt(6Q>aB
zp#ALUnX*f>KzU{V>uZk~tqGYpP#_$99mG8AFVsN2Ks<}Rz*<BTt|^A0NdWH(%w^O8
zn)Ue(CFV1%D9j^7V1b(Yef=Qj9_1ZMNvxk5T?gqf3{h!>5J^I)0?nt;2Lu*hnno`m
zX<HDjz<!XHmWCGdM$V3viH5ZU4E|_FE==K5wlS9zK_8-R%F3Z_V|i?>6l!V7<_qhG
z#m7PoGhT{PkFi;vO_GmfXT`8evdp~pCnysycl$3DXwZqU^s#8M6ku8l#_a9nDrLXL
zLcG1I)~!kJJaU#?BoU?x6x4F$z8I$(&>Viip%+4GmAY$SyBC_?FUYDEm0>;bb6ltr
z?kxd?2f|u8%v+)l+#`r!Q{>jtv|wS?VxH4{%{I<9NMK_gI4*wZN>F4l-G-V6D<>=4
z4~V@zeIJ9zycT}JkV>#5N`j6v6l5Mom=6u|1M6BfKimL}klVO0xiL7rbF71x@i^2%
z20X?YeeW{{9g8)V_ou0E?A4*s7V(d5+7g0vg5xb_Y+C8uEo7#FIZzKvVEG$Vhikp&
zpmaQV0S`BE@D`2qS$J*i@M6MzrwFVpHjgu1FV&I<N#4ayk3T334?-QEdXjOpWMAhT
z&|zuAbQv#W_h6;jm~8Frtzog+Ba8Jr8k>}=RaAXzW9ONfh*sCxFay}VHkYGs%gjvC
z%~|)h?K9d#b0v9|z6fpMzsfQ9y8M|pp%FWs%T6`~6{E0?h`}LOqeQM!QhT)wsv8OH
zB7`bGd7I*TShkvtmTO*?HDOV`*6h0g=OU*w1{XRQ_rPuSZKdd$pUXC#YWd<`vOm25
zzP_*hhswZJg?xuC^e?bOHURbAN`!8ABz&#2++XRH8ji@7?W5$Zh6`54jj2jbYmMlb
zcdX9_?>=S5<3)?=V>Z+zMGQv5whzX*`DXe$NM`GNiKUYFS5i{Li6|e2!ns}{>5aG;
zZb$`pURiR{bj11M2#WZ)t*>(Qs&QoNzPiURFAC7p(S{U7uhAA+^_BC*K~j^jb{i6z
zzQA?Ost*UozD>YJiJyfg!oS7aYC=lHsK6B?U<5&m&227q906R<1CAeU*UEe#AaY6o
z-Mi&B5a6wlyg&?8SjAiB*}8=AM3oy}auxVwS0?oWMuxsyCPOO4>bR@^6BQ(*a}5np
zf=923q+5AGzK+=)+_?bvsNe|IK79%O>*(4Nd&#aCT@75W3ho+r^FU0iCeBzDSH^vQ
zs8_Sb%KQWx0)`H3*Ts-)mM2SZKy`LXVCdBHoN>6CjX%|0s~C|a9DA_4KGSFk+jbM1
z(y<LXE5xy9<lHtX{TVSv{GGz5dXB-Oi81kEtJ8OU&i6xAROFX^?d;&o9BbT!zxCiU
zR70w`Y51z`M;37-orc;6aJHJb-_&qqYKYsf?;RDpW{Wny*5aYs)@=bIM{SAgn~%Ej
zZk%|ExAmz}(Mk3$uacxYtQF5Bs5kDPlRdxo`2k&wRfOvc7jB80!e<w_{BA;Oxm$8W
zTmJ5qt@qIaNt~o)blho=^8<Yk+r}WG$BVRAK!iTgrZN2uu`iXJzO^R70w+>t(iwDz
zf3$y}tY+yGCc7@hiTK-8sZ{^jb^qHl=B%7`i_)rpZLj=_WSEB3HFos+3;QONf~R;Z
z4W;CLlb#ux%Q|s13#rFcno|on(84?mM02Y*7gjDPUj%m%szOeW>fX6B3R$y7;RS*v
zM$8Rcnr+02T@%UiS29%#Wh_3Bg;JrwG{n4aP2w4N^c9TQ<JZC)6dJn{{Z+`YnOM!7
zNXZn7Wr>hV!`tUWR-PV`T+|y06QEjWLn5Z@Q+bd1ODkIu1WVHtv5}pgGp%e#$-qlN
zUt5ynZrY{#PUA~8RxD;?2pLPbAE-ujW``vgtlrXv)W@vpB>5|SSC`Z=Yh`6w7q%uw
z;P2n`@qM?o02;iAE(PwR39%MMO>8O+Qp#`obyaSRcqT2Gtl&$VP4UZfvo23QUCYpr
z%C;Wq7z{5PI-qQZb`n_N@M*Cet{x}p{1&u4=ZKDEG(ufDceEzV7j<JsrF6yLeKGg8
z(|Ku~Jv9D_i|!EY9YG8rcx7c?w(}G+=+s0jyaaAZxzcfsOr+^Ca6RdD02ODW7w*{=
z0MXur<&*qm8SSIRNfoF55mUuJ<)wX3Terje=-2L1!n7x*_n_JhP2qP(^a9#-y!e=j
zojzN7ssi6)1857Y3)C*?#gMlSdBAoXujD5}|Bgoh6<5>CcCbODhL3R?HCXg~;FEQh
zZZuPF35Af_{^D6)HKxjs-PWdjqcbQfi@3MYs<uCLFyiS<<!aE0Nx067WKxWGM&Q}g
zi&}_&99%|~g*8QypFaBRw>Kd_$A4cfxG)zopw+04dTqp5x%AgdPFwJCKaq-4@+iM(
z4Si&CMcOenlW-X|<>KWtYTcc`@kKRtM)O^o)rQHOq;-JyO|arAk6v-OzCQGraqYGH
z(v6D7V~D-dwH~%ldZZQT{5<$l69rD=7E(yB-CIH{dK!XHItVbl3s2%>YjaBr6nOb=
z?T$*X_tKML%j?{Z#U-2;lSt}a^pr2#(%}|tarY!>r;NnP%Ee7aA~|4iISvC|kqss*
z^Q;chih>gfM*0#@;8-eJ@2&jkfcBEu{8(K!<8-N^57B9&o4RfHhLjnVa`91pTP(wu
zC_-&8da&>78JlAwTenifYUi9^R&+Z7WhuLYc+4x}d|~%_rxs6;=THxV@J)ZCJbM$?
zPBvrrP1cjld=}LJU)3i<zZHS6m$Be;vm9hF5I{9nAT&G21Lh6r)nZSHmfv1R8R~rE
zXkuub3Lgevep@~%lJi%jx$PO8lHmK&+Bx16O(sxN5hmv3`hwf5Y~}AUdhF92#1g*8
z8296kbc9$gN|t(-^OB&Gp6hK)-T45uy<@Ge%{zxnd%lCD<<QNP5e1I@Uj6DzlMfzc
zb|txfMm3OP=*7%*dL_dzt7tmSmvNA7IQx}>k#C%m@<Ek?yabR>5LgMpekYujzv!4T
z3D>8AyIW5$==<I*xSu_BYR?VC)TsrluW*lykgy0M)4a)e$$<yIql`Tnm6y_`<n<An
znj_en9`HG&G6zL5CF(}qy*;;~iX8;%shMuDszeZhD}BBDi4qUl5RBEu+`fPAoV~g`
znHq99dQ_f_kDm-`=%#8i2Q$@=VCnR(vxUUw_Ymi+{E%fbOvWO~TwVJblrN!5)TOyN
zHN~vYk$z;ceV#FS=6N6ihRZ)2{NY-@26WO(3~Geuq%U*n9MOiXFJ3_$!@-#GKd<}H
z!b9S12wPAZmS>vj$uU^g*x=IPy?@-|eJZgul8pGVYGXRbFEh|G;He*lRyI2stE)SO
z@e+soT@It|2dhUGO9VqskP8|FGo5J$5G-I#;PvX`GTLy-H@B1y+-oJBaf-S;@o(Gu
zFD&@Ko_xxS3`C9^8{PVY(il5Q`CVQL6+%$&iN5E@-tk;3nP$K0!@MJ^Y~-JT7i3i%
zXhR_TRCPzIIC~8bDILCP4?lcvDbugc7<i3r^h~{A?k|I7mosB6F+Qfyh~I&Zi>eW+
zC_gBfGl|+>{Nh$~oR+M0(U!V=Q|BpyVa5yYcf_)?Pap9LB3NMy*rwSRvjqL@5RLNk
z{K`O%X+@@v_mA-oMhe-K6%+@ZO1|-Z;SuIA(JTuwj6TMs^Oe{7451cb3dRdQcS3Ma
zBDREX^>)c0SH&!k-2CTNQGEHt?vmMye%n{6!RMPFQ8N!+L&X|Yp#9Hi;9KMDyr*nx
zMgvxliL4&v&81nq41g<G_vf>1Yxfnmo#$om(|=5+oL2KwVOfEZf8@)OAc9;Lxh1K4
z%2lNq2q#{Mlg24Dv$V82dHt|Y$ww5qd9jN|-mSh{f1RcO^8;v2CLhnBbK3s==l9bT
z;2AF2sXY=lGmzU1^vZ~}nAkCh@5*&g44}Umq(qkMt}*2gYE{7n0?uEFhF40jY#vIz
zjyT#FR7+dwow}H3fko@IbyhJO&T0NFXn!@7)r7s&cR;u~(cC=ZuroW<q>tP}`*1f;
zscGs<4ihiL7}C<S9Iq(75z>+jFL}7VB8C9_^&CxNf-YY~7bve<^^J9)EvH#k3(~G%
zM@%<CDViL*3R7DquWvvRQvp!0*b=O2RU{9Z7>V<oypbiexSa05oA_eNum+FhW-O@3
zJkK7VXVniJ&8Ck`$b9~LpbfnTEq$4)GxyNLXM$g!xJ9;&(A7?m>jgqK$S89Y_(2%n
zPTE`hyF+)^zPh3%feQ`dA!LTGZ80<<qd*$%xJh4`QWJ1^Llcf@vFjnh@owzZv(Fw6
z;(yFNj_bQw&@Yyr2OfHD^{LjU&^3jvoSPyY9fAmBK4oSx{&axIk3{zkF|R_HeEGNB
zZ30lxFkbJ<Fh^s3Os2qWUy-&$awbRq$_tf7Kfy1TU9;5Mn!sH!!A-O`WmG^aCWtzu
zlCjY?UopJpfpC5jH1keF-?V`S)*#W4M|IE1^hr|gbB!m74xKjBmwv&L|9Ln5e5R6T
z&rJ4FgPLDf1W^ZImtDcgvaI4oU!+X4D7)!aS%OMcJsN^I>SMNf&nJJf!;&qhOGQA?
z3-`iTW4aJhRO<`d<6&1oo2=Fg;^vgm(r+akWbSEc^=xE{J9GX7+)T#^E&C%!Il=;m
zFP+)6$O4y5|0SzI^fRsnW!EMI%3e4zr>&yPS5^LKZ@c7@`iL6;qXbvz>rMqdb{M@x
zMoN{EeK6)LglM)o!AkM+NZ7=E1}`7!m%xa6v^CM9-z&SgYo-%I(BbcLUy6NC&XRJ>
zt234GFwG#(6s;|=z?IbWlXG&wxkLAohMwMzg(Tjx!A(<%>K<Tf{i?mdDDQb$+<whc
zB411)+er;QvHhiEh}P_WnL4`Np^Y!^uGDBt>gL0e-stjH&$i8HNB7m~h)sEO?72aT
z1s?R+@2J|ERf^%glk)xb3@cW?5eT_J{T#{k4pbI?G|lH`DfblgSDd=6@AQv&7&4XO
zEvuaV)6{K|g61bHk(@;vPe>YV(S;cC4O_xV=scZ^NFpu1A8!Nr64?{OybZF<mbIU(
z^v{@PJ$JrszN1uKfuXXVO|LBxU!qAo_R5H1JmH`sKJ?`{5JuB<7&{#XHyqW-#-vl8
zv`(S&JmWrO-pQI&j^wpesINgdr70~@((_PWsJgGrGEM-v|0RWJNRR;_5r5xO7|79<
z5IsU!SR_iF6yWPaqr!gbn;M;!EYouOF*;Vx-f++U9vWOIKaJrxNXjt2##@!4zb90r
za~$h)8P3nE5`#&W>SmvrSjo_??=Kuc9aOu7p1Y6)u)lLb#tvA3D0~qH8ZmxEEnKm?
zTJ;Shea2VyO`&8&So==elz-Z)7#}KV5w=KIQ4k*c_6*2MZ*)Ht=?$L_OPlnQWbU7V
z@F1=T4+ME0g8@~D?PJOq$a%6cLJ@WzQ7G<i!ieJFgq_DmiQro%m*3kNfp)nZvdqdZ
zQjcph0luulC{GZXGP>VekwUgX+pgrj^@Kz-_PFIH#b7}{*a0^O8$1q<F&Nfd7hQ}~
zdV&}&G$D4qYn_E10_nM^bpJJi<BXbwU+87yUh4^xbtAo1Mmy`CP1^2<()HB=Cx})1
zwaWoGZUT_<_}pc?TKh7sNarjY842mw(aDgg@dJxv+i%vPUCnXy45gkIn^rj@Gjyka
zm3npQu!ZO8Bwhtl4InWiCy5fl;yx=XbW7bxw#YD~T{OF$vNyLgdg9b&Jsw|@yYY8|
z0%Q#KLfxE}XMUH>Oy$?^@$?}jA7mMghy5UN?04yJmpquP0gkOO?mz3I-MF>??h^Bi
z6WIHVx7>^FA*Hh~F)xgk3s%?*l*a|^8Fx15C`F&lAKAu@Y2PEU(e^|5mj&AYJ&vmo
zW<?!10nwNPPB1M)5&+@^-!gn%=6DdW_`<3&rmhrb^-q*~i{UFSn>|+BQ0{A4i%Y3c
zwgg{Cv6yAjLur?rn56kP0iOVm8SfaIEguaohG?`xE@pbJSulhYjfuRUr4%F*V@8TG
z4BpHNpyC+s^oR6iqNGT4D`aFU;0s1+AHm5rPW!aNoQ@9{PAGf&JKC-lr~)rO>)du}
z^3l^b;%H7l-|lafl+&rayflFXKF!^MyQu_EJvYB-5y?C*aQ_21irNV1ANy7X3ULw#
z*OHY((BkX(85}u3E<(v8G_EE6_H1FJqJ$f})RMB(!tZ80U7qVYJ9`2uGOr>A`LnJM
z+%G0Os&1a!?uiK!NJx1#4}LWr`VSw}cBrERW!yZ{_$43#%PBQTOvFsTA9T~4E?@Z2
zqaxu!C&W8-GM2eFPL5{gjdh%FWn*L7=4q+vloDuHOy<-Y-`~Kw^eEsGByitMny5;n
zAMHxlT(Z5ZU<6n3(s;OD;05}i))<qZo8xM$@d1hp6Dc<X!N32uG4(h@h$hqT40Ipo
z)y^yJpwGNV=p}<NsT|z^S5V$yX7|Zwh-Xrx+j8c!*FkixZFjoqhi<fFKo{Hrt<e?4
zPo>x|ud^C%4*~a!i4yn{?hscW!GS>W@SQ2zTP0FUXO<$BJc_|iX$W^XL_DV?wzVy;
zH88(~kK~uC7$g#MgK>%9r5>~i_5vycY$1}gKduaabzQQWY>lZ9h|gb~q_Z<(S|<fl
zPuFdJWV@+ow60EP(Yo)hOO{{*?-F`GXLjwotzW~3D}1M~FHMY;FSJ?BzV%%dofjE*
z9g6F(mv|yR3A2l`o$bv*m^Y$a^3~pq-?yr!XV~<7ZcZvEFt}s{Obee4H7Ir(AmJkN
z0{($w1nmpL|IoW^`r{A&hpioX7L5zcwg!R2ejN$&!;hVkACmK4B3>`+RONjavL01m
z1*W?s7;HA(W@K?kJU8_6<((HNFJN*oCs|uD`YgOGR4gPUi_z-pm7d?#G)p+ioW)0&
z@1u1(fCT~36`JSmB7L;dc$@Fvt7i_kw3G8($h5=+FOZg*LaL9fvVA*r;+#-gfbIxS
zs|5^GFiLBSHRC9nRyL7`_g>#pHlRck>2vbRjHmLfHVGEB%lUZG99y+CDzoN_*~*o}
z-8^K7Y1g=R(&M7kIYf{3MtF_oaC#G4DdQ?8$r^TTpV;w^+>~8b@{9uq=(2iJH6SO@
zM1RHcg}mu~<~RBt{yRRpjc!dD){C;`7XmLNsnB&|+Ew0Z*Yl34ywIQ<r9fro>4uOg
zGjBnP){$ySrYnKk!C?%k$}*?hC0z)nI3;F&4WYDFgzL|68Li!F-f-@@0MVPW0fBmb
z?if=K9NHtKfhYw6saM34%6Jz~>ah8*2d?{;fcaIun?)7y*DK#VDKt?thN)6(3|62@
zG^Exf+V153McYg0nBE7%tsW0{>Jt1u_;L9WD<H;c5Uj^^OI=$@KUbw}ExX4u=a^{&
z-4beveCZovYvdKa%6f}Es?7lZ!@orGhc{~~TZRqYXuv{rln4GfMo4GKu{oGJm~9<%
zR2n8S;~nIwQux)bUQbThG$pL!n)<{uNAHhsw~1ZVcNiupJJEYY*)KM5$)zh@3L1>v
zOOqI;q|#buciKhci0tk^rIrh5BIKWWvg|+bUReiP;z8klrtcOqG9O>Y5QYo_et*Cv
ztNW^qwuqUQHS9h?$XE29V*zw^jO50r4`j8)oDU{q2~<uT)AG9VX;|z&tdyC6nC1Qy
z`Y6S-xsiRq_;9>l^FzOTFXL}F?|b-9D(Tj((B2<_Ebb#6J%}MD&o}uJIkMdJ-z6V-
zhRSDBtzr)8(~1xBYQfZC@f5j=*lU5H>c*SCVrE69Hx3jXbh^ZUrVC9eMNlchgdH#T
z7h&0;$;#FlIngm?4=Qd$9u>{jvDYqBejwlxZgMGdT7xF<7z`0ai3U#T?i8|%S0p=w
zNmRs(CBk=!rUmKSpJVwA(kD&}5YigT<uqBFPwC8p9m~lC@t8%;nG4K{ovL(K_0Ka~
zwp(p7u3xgI#XonpD_%Oe9wxZn7e49IgEjkhJ>z=c;#S#<rW_6YZ>W8YTFVL)LJbm5
zYL>0W6WK6bh1qKKAs0RoZgoO-)8=F822YLQXWm<3Jgo-!6@SMH`+h}%36ls7;%(Au
zy8kN6kCu$tHRX^c0c*a_Bw&8`30(<%onCWC+#+@}>~KZ0uo`|%{waGU_k=PDgKbxz
z2(BI|39C4j_e%Po;1XVhf*5t#?A9L~7h5Cp9^csLyo*uylcCTlED>0w)lQ6@X1ljs
z2=*%tA}>Gs2Yygyf4z~EowLMV*BMJtWlwP|JYmbPMmAye=s==@LEQW)Q5&g0*mVC(
z`|xLP|MUz(7i+zJDeI57S$WRK--LiU--%di?>})v7JJeiN-u|C<(_v{irPPn7Q?LM
zyNC{04-4@t>Ae_86Xg5=N&(b9>;4{K0=*vs?zLfj@h$rfoC@d4`U~;%m6w0jB7gH%
zp5`FmLXXG2=|2?G??osKhvq@P5jbharWi=Dy@rq^7gzt4B$**?cOe;uZ-!!LKiM$K
zMNd})Nbu9~@5@`i!Lq??Pm+3VPErt_5}hmUt{Qbg2CT@Z9vdneWQ=uneaKN8AYic$
zszSOM&lo4Ae=6<&Aa=^4fq^l80gCxQey($4xkpnB+dp<W|6j?Jl6a`A3lLfmsGWt;
zsz8ZGLSYcqhPDhXIm?g9PgO7T051S-$djYfaq|&TU?C)H!y#1Ib|*3FWi!X$Ql|fy
zUB0hG_?_1Vv&Q~_1AxMV+~C05GBs*N5>1Z_nalPs=%4!zQ<#1@esyTbxLrVQb|~v6
z{J#zlc`^gxl_zD*uO`f&CyOZhhx!sqOmM*)f&cXO!r`aW|IiEl3lsKV={F^(5ZQA7
zHJ;Lcm}&j%i~j4eEIq<747fw4_kUU6|GDJ;`-=ZBkn2Bx_kVh(@*F~ttlpcs@joLp
ze@&JDmj`(&u*Ex+JyGTVKV0Dd{2TAzAr=86CuMv7ffD!gTKy-@)K&wbrxt(Gwf`@E
z=j07?B&5k^d2aCK+vN291X%;N?}$Yv&{D2}@+I-dbWcu>-mA>oc!$R~Z2vxP;^%iZ
zjj%!laFvF?ZvCglbh7Mt*h#3~&GzO!LHJK%z{NZotdV;h2u!b#tXYMgnZEpF{vsl)
zLZkgBMljYz?EQ-29?9vw>|F5;z=k5grCY+;xySm{r4Gh*7rb^s6)T@^JC~XVM8`89
zd(OuoEQA%{@I7){0S=qd>bFk_sb>Y|%e^#GHeeU`kyk$16Ih>*2O|7_rXT|8vv;|=
zdh@mh_1x!xYj8MLfz0f|+t05bNUYf<N^BlD)8~4PhsUIbVlmcX+RHizCxaQ^(Z7=F
z{CD*4x_ThkeXf81IQzlRbuJr&`3_xDKNLfkN^2uD#Eh^J(_PO)GytI2%b!u_l-^#x
zv<Xv3Bnr1R-4H~rQ77F#`-lN+#}DUg?g}9q1HctZRNF}ukAU;K8x7689BLRp@&7n`
z^LVJ+wtswx8YxT)rN+KwjZn%KV~I*c%D!)rM6xH5u~T-TY=!JgcEwmiMD{g1W#8BQ
zj!)g!^?bk2{oL2{`u_g9y5^N;obx%)^ElqedpXk&ZgbY)T*`ks!|Z_QEaYC~LW5rt
z-yc)C4%KK8?EXQ-wX7kd%h?rdH`l>ztz&H$rjdOxu_&n)2TDnTIK0t1Fc^Aa_3qQ^
z*sF|aXV{bjYJ-5!{V?dDBmNHfiwQ#8ye{~em5G?*q#ywyP!42lK@e&W00j5YKCHOo
z%){aUX^Vg`x#PbU7LLQHsEnutFgJZo;jTH=;x7F~I4;&tQG@@f*kp`zLhs>j<##1j
zCwhLLN&j(koy$cs?Q;u~e=z;tKwsI=V$icY0BOyKZVe$67MurfOPU`y3W^}mx(**v
z1o(JALi)Ay%9Lgwa~gxPF9*ac%`HGs(2jQntg0{Q^1L|`rNs0&^x)A`#)!p^m2&ew
zlVTqNalIS;&DQL?gbZZrfAmJdtuqH<n&+6816v6zkQSjsg}Rqn4>Jf81np3;Zm_o4
zhKKjw@RNNvo_xRA2zIYeumCl{2OM3!iNo%~B-R=sl&#GVUS22X&~kic?x*0`@d)vF
zo3la25}&A<g5UXYXB8(}0|&mOdl}G1{Q5Q0kD%7>2K!l)#nu)*Yw?Q<m4jXRb={)d
zba~6*!EadOL&=~QJQvm)eG^EVUT>s@uQ3Lfq>lJJZl~uM9~izG`tHtuTwOSD&Z$5H
zbWuelgX_-~bT0Q8<FP*I%nbl(Peqbiih$AC14LrIcYM-E+g7Q5+-ZFIPPM@%PE_R(
za~9?(PjJ{_4;<aSa8A5aN0};?$KbO4^w;O4A3^Tmg>vjM@KJAcFF%t}h{ZhtmZY~8
zoH2!vk^BB;CBS0sh?5^3WiQ?iY*J0MW%jwRgM>2`Xs~<3I2P0|h=$IE4^fW*$RT))
zVZI>i^0}Q}qm*mQec&nZjiuU1oLGM*|K1#R%EqJAJ%E}95He9np$0g-wWt*==|;2&
zb_Z<u{d{^x5m_1;@{3r(g==>Ej_<Otf1av;KX`w?qJw0S^YyQ$|KV{<aZ@`&18t3n
zRTB*{6a2Av*f(KE<pLE=ggG9`4;O(?WX3<E)(_7{A*S~x2tvLH3Mi=FOe@5uq3~+L
zFjOM`YN^WEG*}6kH`8kxzF&`w45?9f*CR_tVJY&!ste*%#~w*io_*UTA)w2h>I2+x
zdoE(9X5qMW>7&O}&wtrFdR*_rf8CM))6Ls~=n4W~+eiQF*0pCRwkGqi5cQ>MSk4B3
zUMp4w<;#vIktK*R$HRf&WEx|^j$nM2ey=yc@3aRFE+ZshB$vlF7%M(n2^^(!MR%nm
z*nIPSC**F8EWq<#mOK6fgmMWL$vM|?7ZOi)#iW1M4WBnF831Vh;tk%?>BL-Z!ADGn
z6iQ?eJX}>M;kMlahuWD(X<_ot0Vbw;jw7Tla2gqjsxzmjY)}XC=>%P@h;K#MeBj(n
zX5?Ebgt#2r>XjzA!+-tT<Ypw4Od}p%Z%s*9Q&i<4)jD-FTkFGp5pdQW%g-bnE5r4%
z<Q@@z^qhZ>JU$tsrdr-Hv`ziETG%y~6fBVk(qqnum#O=yX_^VI7b<>j|8#)!Hzkrr
zA|JdH3^XY(GIso$e_Z8XsZ#u=@gPQY<|n=+fD*`NPu&3wvwLh#sF=OT(l-D{uvtzN
zRP6qW{lHu`k1R^uZ>t18tv!w|1&aiy8OBE;R=*JVot1%gQV@k%ZKB`yjf%AfmjFTe
zb03VqA^gkT!pP-U*_<`El*LaNy;KNTd<F7h=0u(a4@em1jPOal)_lJCf>SC0f>xsL
za5P6ft_SJl3g1q%B+oDcB-WJ$iu!D*$|ocBC{QuQSuXLfoL);j(!95re57}x;d0^;
zi}=8D|8FkznerA1$xqH}2u3zTwA~WriA{<{m?zhl*|uNhC<IicemA=1Liq){$mG?E
zbr=cBOCSXC0NTDViN=Zo>&&#alGM-*r%aRdsZzPBgbC4&?ps6c_vq&|Qnt+hxk7u5
zC}sN1lDYrtMZRI)$)a!yQe2`y__Ts&k~`yoWV0@SxIm+i@(l195P~7H&p<v%vHO5Q
zN=q!$`=tsi6+tg-8HSixFy*Igytp9Qp$r`L=h`@-YOBQE-`+PiXtUir;t;1i%jP|Y
z1d}7vj%s%o`cC6)#f`sMmGy!jo@_Pg%|hu@gwRa+Wc{w(e0_u7D(C4Oq%W(>Qm|G7
z62X(0!KBK(x^zgHE~k8oJ=RLKK>Ix96^ee?_0sXi47Z_g@<1w_SsFHD;bJ#<oSC?o
zC(eMYi)?q;d<XYcm3<`Fl7wa%3ZevQV&+6+s8f{k`TO2hA#OYCof>-Ho4N2T_QN2;
zQo*U=I64&12x0DFNPZo1<(?QE?t<hk#%f@Gw6R|A+$LE*JV)Bp?uAm=Lu@TVCaF|H
zKAGS8afPG>DuAH~5RtTNk?$a$t2|47-=*~owHL|A>*|q^_38dIY#f-oD_><O1J`jB
z(!|MGDN05xgRj6UvrU8CyVZ*EP0s2?qXS#PB*UNCV|l#dB!4XhtI!g1HC2;TZ-^{7
zZ{wWxSda%f@09~rqblFzNoEzeEQ2frGTbMH^*)UN=gZ$IHutJ@j<lMVX|#zIFq(rV
zCT6DaO~{QImTURx)WtSH=p-mC0-PeZ$U<q)hCBI{L^?$}d5GG9te<Hhbz@@d^Ukx0
z+WMd~^ed-ZP2)YzxP>VsBVH2tH7AJ=LfLY|wH6YAU1Jaq74*bd$_<QNAC*D7C+#rF
zK+Sk)e=YmTZy6184^RJWW<b|~Cby&6yZpkYJgR`h={=aK3JiY?b9I^$kQdxw9$TsK
zGEYsDit2jPZpLDFJ!xS@vDIevPX2J~a?B1r?-OUeq&=<*y}$g%|HHje2;J?S%iKDD
z>0slYOYstahxv@fz4y7#b$}W7yA2FKbM<M;Q>cgcLt2WnV93bi!gv@{dnnY8(-N`Z
z#o2K;etb+NPbGs4%BjW?O$`;r)~9$*N4y*CUpq8g*o9a;{<+Yn;C)wS4TM(RsH@te
z3(QEF!<?th&5~kf)n|&@lH`5LA3*vR33p?lhX;ua2?q{!5pxtK<1KkGu?{nNR*sLd
z161Ef+nY{qoWq!O{6aD}$<+uQAT`VQC1zuUm*B7>3$=j*i1MTrkUGgezgfM4zJntY
zn~XA&zC;h<p6aGE6>uhZFsK}GdWN&FhAU8*aMUZN=-4A!uF^l`QzBK2+daPEO?fSr
zG;MGk##mE`hA21|COiLBi7^v*V+Lzbs<Ync5QUox(H+ln571s@a?pM7iS$I$WkEie
z&0~2mp?LgNH(#oCAeIo82XRM?jxZl<*IT}UlT8xGE6C~S2bqh&uSO$&xWp7Z5#Dp6
zRO`SkK&<YpTNK_D_nldJo7^qdTxZ<B;o0#XT}xc!SzzEvw3&-9W0TPL&?**}5U;4O
z67WQ~%cx7PL}f?ai~bf`)vb9iGb9IZ{#mT9JQzGluQ}mIKVOvU%apIq(tO(N3um+t
z{bVW};OuoYD|}<rCE<8i1b%BC1gPSEvq47^OFNP+P_|S@557}f*Xj?B9sP_U@)I!)
zc&uhQsq-o#t<BkZhjjrx`5EE6Rx&IGQ*P>B(PErk;(NQ#WV{^bWwxxV@1FXlM_!J7
zlz(&%Ny&?u(z!-ZbUfA;^|%RqR;>Waz3#8KmOoUI8olAvDc#}39K2xW)V%kIrfhT~
za6Q8>dgk>%w&PO$B)d<}asNw|=7BDtdn%80@)9<B37V6t_v3t%tXKoCC592|;Eykh
z%~c!%`S+rV2SSTKh835G6fdgV2P*OV%9AMV)^8Hnst8G?CxC;<Qpv{0%Z3f)Sem%r
z)KWGcS-cund{e`L&qW`=Y(;IMW~`-UzjvwrSm>Z-V|}D#&`-DCF+$QwHa^tx!KtQr
zc0LDIK9S>RlpkxKQ_z-E6!X*{Re9H5<E>0fe=KtSSY*|U-vrcm554VXp7Tq-Dy9mZ
zuY23{(6@<2)=I}uK|9B9*N-nNZ;V-Xw5js8X#Xz*+KZln@@MbukLqPVaN?Y|HQ;oD
z>5t$1O2Zvrc{<4Dz15L9ZA^dpOIOm6#1$2>OM0s|da4)S@S9PKX-6rE9o8N7G+S4*
zIzLILRk>>ZaIUbM9ow5eeX1^&yy!DjNjn768iIFq;>wxau~3O59Wg<*FTF0u)6My}
zv*fdsw8zhuc2C4bJu|n9Y&aILqGXw@7e1+_oY0VjjWY=lxvb~J+D51y@KTOs_RlFb
zKCx0se!ERWY}=%5D`>zXS-IH$lu?JthF4|Ob-qLnzASd^iF39mFSu}ZThmTb({fNx
z7PS)`Z4E@0%-^QGeK^=UrZaF&T)Wx-*Jz@-LCaflF=v|*oe+0Q1D;ZaSDVXICP&1W
z^|coST}KnGRJVslT=3*3WP2|+$oC2>)NHC6$J~Z=qx&u=)Nka^o<9}6lfQ8xfACtB
z<zz|e-|DLW5ldkvN{r(CKc6Wz)Oh{eA*>_guaMIrTVgRf3QmaPZ8UVIk$_J3D<2Q_
zJEqs3Z~0~G%`e1F$dzosFOTNoGx<qhGbiIxKf_pgFaKpVH+;f05ApXLgBq@`r+GUs
zqL{vfb=9!&vthT+;rN1S?l_m`1fwRE(1)^BqXYb@*u+#-J<G3-QZKc$VAv5qsncVU
zND?mk7&ay2!TZcWXVx^utqyMGkVArW(yXJ#DdTOU(tU4ix?*B7<&%iN1KBy0QCnTH
z()B?>EtJ4|4t*?|wZ%*C0!^m!4vFO%OAoqU9iJVc=)I34UYfS>m6yCG)~Fdxh~i}p
z*aq7oL*;eH!a2EZbqn3KRbR0mZFlR;xlk}JWU6&qt&hrAnn(}h_so;e@UI*Mn(mO1
z!n8t>!p{^kgFcofc7>iXmZcV)>I!Fq_v6n94Vp3Xv2o*1I6!iToRa~%*A{Uq=$@8O
z*}wTIDgODnR}~S~BvO6fl4c^}xn@~twwFIK>3vw+W2<-rtWhuq*q)V>T-kJwaC2a>
zd$Rruh>CL>AeP(ur!GTEgXuM!M?Xc#yT1aCzMdsZRzl^^8k7d4KXDR>7&h@$iD)xA
zs}+e0m=MqZ{6y+Pu&D`>T2Z2HiVfs-DK(EZ(b1U|h}1K@+Aw=H!gp=@p@?P2yE{V>
zDOX(9ew<L6Dt$$*#N-^7&`?!yq}LU~kngR$16A8cr)1|y7n8f6KsSCd!vB3ybBcVE
zpUKT{DYoAp=7mkcB<Oh4WyEWHX2dUn$9J}Ep_zrdF9D)z67Sui=_dVjC-o5+)D`3!
z)yvU4shwRx1KRG-8bO7a`v;oveskb&d}sW2$V64?{s-qs<?&zFFANI(7*PpO%=dTO
zPE$&m#7xZEJlWBzy{AYiq804CL(*#Y`DWysX115@ql8MOI1$9M%&HYZq9aI6wg%TD
z2;f2+an1CzQZVP8rMph2jc2a?Alxz9?S6qcso$CGH)-)}@-SQ3{HeEpO35Ifz|oXb
z6n<(3%InT=9Lw+I%}yqLY1U}Y#fG<QNBNudD@CX7U`xJ<S;m#1=$c61)?Z%E50zL#
zihAwFaxrK(?_`ykwZ^yYPcxK1aoou9vXW<;Bq}bYI^EQf=f$(2HhHk|0cgiyn9o^q
zM&0X<X?gEs+w*IVK=&OLcLX=0D0yMtj+YCx<&;s>n>Svt({7FqRxkJU5+0+oKv#P<
zHHz6K-8^=pd}?_jevg5Zpx5^K1rVpybr)7%Fn1KjcEdR#L;wGDP5|bb+o^8$8?<~H
z!4Y!UqrtAS*{0*r465x@($Q%7b^*QH9})+ql*VG@4+$lUBtKnmKc>Wk=^lpZzDeY=
zOu76|fnbz5mMHrH_ES?NNg>gJgLH<k_<J=jYLvF!C6NlwqCIxiVyj0I9QTh}yn(2R
z<Rvykne-*RU-*6A(<{>#nEV`Z@2Iv8bGWjcpyaFCeYk>l7J<kddI{HJ1c97XB08i;
zx1Cq;DcT4s805(4WKmAuU~vtgztG}(w}16!FUQ`2_&6O@a)n#kP~Nf~N|gF)%8ReQ
z`WsNa;C9r@MeV41NiDjQ91G7@7~i)r&&QB?Z9QEI1Y&FhKDqSML3+C7DETa_{D8pK
zjP@fxXZoEwlJW!E)KFH&m8Af68M#|16Q@*+^usB!O|mx+f;wS6!!EIOgYuQ>jAjA+
zAY!+Q1cRO!_D*q2A94#`H^UNj&}oS=5Dl@I#_(UZ!d~a|((NM`r~(Jg#=+*Y;$G+R
zv#4e9NdN+@KAhc@KS*y=o{L~X*1B4)+Aq6z%D)g7+6lHUn9U@?1@O%!cFU?rPk6;@
z@E-%>bP(BgR?m<0{683J6SV@Vl%8hmZB2`m(jP1EZc+RPEPjjnz(zLUTEiNeBm5X<
zkgC!(x;ErA{?v3+`SV#sN!xeh^4xJ|GE68}%554HC@<DSmR(DRLio5lvsZtI%tF{V
z&!ub0qRjh80k=eABdF@33n1SCKnW?s<y^M~-%l#FKY!4$q^T<YLM>(YM<}S$*if4I
zY_t)sHbWSRGh*WHbat!g7G{1r3k8|n_~r4$c^KtvHPGKL{?w)SWRFU6X3RPTImq4D
zR_Fd4<mVw|og34q$Wj=dcilVdHjH5|6Z2N{9=o*6yug_F4p9tYh(fkrn3p(lEVFU_
zM3|*^4!@bDI0@YLn2t-=T{G|F{cXgIkM6DN3|NV`|5EMCKIMyQa0xES=N+NTeXpD#
z;qZ~@Pyu&n_xbMeH3^&4go(YqfIK0=jSIJYR)<V?)MJAGmvN*e>vzfH|JRKSRUYe-
zhTABy{vg|ig1v8(sc+LIU-^soR@@`EUjV|&E;@>5Y0O$-v!>Qt&zZe=$n|8|IP|TG
zaC?s`39=*MeYFt6gILaEGjo1yIV&Ub#7}iBGadHa6J9CgLF(C$?+Pxxy8%kR-Y$Yk
zFhSNU&u$HS2Lg{?00}`Ps?;jHSS-RZtb~yAf)HtUL#)wov@P@)ZJuRIUlJ=DcJKY<
zPbGdafDNE@>*5kuf0Q8dnCSfvryEd$I+8SsWPgsgD#EU<lYwg}%}e+wyjlT?*0iqh
z*HJu|T7vkOUOAF9do6vt+a4uglvs}4%ba9}{>-UuN-SNkY-1MsEnj&lCG8fAjXvbN
zO@&Lk{72i#I`i`@;r_1#7|O!p9KA|alXF{lKGjxDbGrs*tJx&z?>edV5o(+LWbMzM
zLu^xxPCFm(^XnF}3n?l2!y*;9pwZOCQo>%T?Na!y<g^VG{qo}4ih;ji$yMCIfb`;g
z(>BeS#}<NpDbuePHYG*y8}H@qdBoKV5?ZEyQ6BLN(&$Tjt2ePFk*yY?88orB$FuWZ
zZt~acUZB-fTlOiZtoeW99RO*}LxJf36!E-1NfeTv>Zz(FS+H^-i@z;qQ}FH86ox(+
zmx9A=p2%lFP)wLmn}WXT*+`!!2=)N#;(80dY7}$~K;L9Z>;YTB$O4Y5zdc0ns+V68
zT%BNf`5W{Q)jKGVKCd*vjujV3eszDaR6U%J+csVPMY{bH@nxKI70E}|E5(Zw#WF{%
zmE&=0%lPQCl@vQxP&GB^`tqNRw<?*4@RgrT*DOLTPwPGc(@JYCX4V(Jm38+b<UH(5
zIv>zxN+LDMIN1-PTM>i%zkns^m&4rY#9AK}9GZAJAGHayW!6eDw`#LOuWN-LiOE^*
zN*CiFFDqCx)nfsvQO{qWs2JP(+D5E@eqmJXN9bFWLX}MV_~m+V=hP-dEU;p~@ZE(z
zR3^e~-ip5ycHEYPO}4gmT<h4*U6K9&r}YP9R^z@~=N~7N*AwW)1Hzahe8Ee(OeA{f
z<!ZL3P;$H9G14Ofh9FV+3|1z3G==0UbBO%8K@y1tSq6|N!W7fy?7T65rdlD*s-AKK
z&sAw<-42T5aj+}mERz&}0LZ(yu_KI!*;Deu_+#mqKO!C&snem8wPxZllU|+(do}@E
zS99HUdF&L9ju};5BD5v=k}2vvkaozVP13@o!emKUnK$MD&-n~IF_jqESx812RH4wD
zyl@xHOvytt&AI8AbDP07LY-8)CuNUukM-~%RHU~Z%2mhZ`*<q&_rySepL7i;rx?6+
zH%&5xV@jk){~3O|L4^2*NJ=XsF%o4G#(26DSfCZ{yLn>x_Ik&Y7w&7n!3Lx#`U>EB
zeM^hM*reX!Y&PgBC75DPIWc>O7l|ewwhlCZ1jwcAPNYPzfE{2=a`s=um=%iDX@9^>
zryJ+wd#cfWJzK5d>Pq4s9tSw#Eq443+?(d2v!`8qo5z7iZuHy#6IZl4w0hopWp?Gq
zj0Pn9H3=m+Z#!VuUZ19R4ww|s6QD)hAO!XqKs4B9ik?uAUhRaHyR^jiMMip_ljvNq
z#M5^Tn=CO*#F`cirmi|JbA?QP$4Uei2rg0hNW^`988o=Z>qy5=Z~5Bh0vWn);ER}4
zRpz70j1TW|Eq_k=sGlLj7UY#m##4<##EhYy`6~y(UOQPS$y+rT3?IBv(@N?bm97sT
zjGy^Zwg#xz4Hz9o8-p>)LeB3pzQQ`HLIQgI8J7U}?uLN`US*d4FY6g{@o6sqH2DTT
zMidF^m*i~t-+xQf!=s7b0U^S{S$Get-YO;-`2`?w?-ylWMtp-O*8BU4fuJ%A1@+q!
zHA8x9;J@gbVl7&<OV8`6euV%mQ!N0x5UI_%;IA4H;qo)NP-_477pYWB-=f@@#LPC6
zqU?yOwYdanUo}sV+YxcMgqw^W(#GZhDn(e?lYg>ea^efiVD!F+72oZiv0%yjm0tQ`
zMo`V9hsm}HM3R1~g`Xm5mvdf<99){#Z~&@jl)PR(3@^^pZMGLqO8b^*ACD}}F<_Ao
zDnO>m+DFa><IS;4t?yi7Hn~XU_3pQ8Kg<DbdaRP`q!Pm+Eyr!j?a9$_#Kpt`BFdu4
zFV4()D>s6ACqGt6-2tLAW0j8C28ro(9HAJWQ#wIAfAxa?4N;viMatSK@Z2y5L3K_h
z#L>yt>%2avN0UF@gM>8k5uG?=cUK^24urvzN@r+WN<Nyw=gIg2LrHI=0VOOSeGg)s
zgbi<miIS+DK;DDvjcqBSTHt`<d7`^S|8{X|`@z<w29r-&?OW|ed?Qg!*V=q8Zh%uP
z$p;uTyj^_%^uQsxD&C52HkuW@!z89XQ5^Ev?!mKHoA~4b@7(pOH$h0~*?2DS7^CGS
zj&xbLWA~cGoQ+m(E-p@mer%4k6$wr?s}##Y+yEvYnw(3doQS($z9g4)^lF&TB?J}!
zgRC`1eh*rvJ+u3$URFM7?=1ITe5Ha52!U1B0c)&(HKG((SDfVD+hJAJm}GM+-t^}c
zEcL?xK!#N)X-|&reJaN#ta4_zWjh^QDQ+sMJDm$o3$gtBExaYEraL6wwqz2MhK)_-
zb4_pMtBmiD#;t=yi{8x+wtAp#Oq|bhdVsuTKH_7$)8y@Jr_u|<^lx<i<@=5ac?+ZO
zw~qCSK)MB?yl2Q&!)7bSjrPJVO4_L}No%#2!xmV|Yd(PK!Mke6{xgAuvNXhbr|tI4
zPOQ@sXuU*&9qwU;h5CZ&krcxf^HPD};#t}i*5R|85>ms)yGw4*9$%T+@!wgQ)^Mt5
zJ*#{^t@k&!_9ZOAC3Z&Ri-gw3KQU#bH2#}pAdkfL?wu6{Z6<d0bT58ix+(5BZt>@L
zCRO~JlUAeTcgbyo8T!N%mGOtM7WE+sG%!MK*D_)VJxYc)_`<SE2wr`~s2Aj7pG<|u
z9OEYQvPf;v1bv_bS`&dD@tz>dqGYONft>aV@+%)%c)j&QhcxE$Zxd0Gk)Ld2nK+&&
z{qQoZpEVpMzD9Ufw54`zDq`&J6>Oi-8pc-rOg@@C0?*MAsO%+8ARR~l`mTM}si2ms
zDTGDru>Q!gU`unn1=fPMQTyW3&x7S>IaD=yEjraTgpG@>q&IP1?<a<z-{iQsHuLRm
z2&wh7O3C3Jt0Tbcc8#8{7T7*zC=(-gSgk<(g}hdpE>Xz(VuKIUs9F>4Q<9++vJ%P?
z&e)I>8*u4eBuPMs4MQ+_n`N5}UG^i~yurN1)Zj>RZAIZ)+dTSV80D3YBnR1HxsTU(
zv{h2C&!3|~hRY6iS8m|8(<4HR`GFcAawqhG*eTOh+wBW`F%G8MjCx>vu%iz0Ox|1P
z*A!NwZW;--Zuymdy?`m0saI%Jc!M5aC9M-STIPSUrR0P5__9vc*ed4F8o2xX>n0E=
zWcQ7%mV7S$P^vxE-V0sb=~3Q|iD-Qv<3Qu4GS97M7R+<UmVBp;tz)%IK@%l~EsxhC
za(}nji<~c*P^l*IU>^32YzP1J6$jIs^ENLwKM`7A|K$ZBf25fs8455qR;i*}L?J+z
z-FM9yD<fMFo>baq%X<Vp19y=bcF*tO*C(6zJ7#xYm7lt9aq`rjdqnrZu=!R=cC>qE
zv)!ZYe}a4M)nPPOX4#Z6^oJ?Hdlj{X=k!oG7~X#E{Fmd^+9`IOK}<I`kGWygsjm8R
z%HeWWQW&`pi(5!&`=9eUA{QpXY=xOai{i!boC?R&Ec3KtC=)vT8Ftg?Ju%S3j8HmJ
zsib$QWuO|)>%rB?OcYYeC}x@zg|QakMyc{Mi>fQ)50CqrlaLc638%*SnCWjHOU~P(
zN(tn2$IT`2Sd4CkJuBDOI*Yj)WY+ZpkBgEMy_Wq}`4a=*99v(p+L4sO5Wl>%g7Sul
z)j}aA$77!|D%7%33ht}*MP3Stytn<R$Yk!A-m(y<JHKl}U9Ft5F8hLom~KwK>a|T9
zL&HnKn;(J{aXGAtW-h*h1ssg3ly#gcq9ZaoCnIt4$=FWS{5?O{*t!1RJYmgXp<ps$
zxyy!?ia))q!s2XPul|fOb9GwBw@2tkcAJ6qYTj3VI{5~VW}Se?II%k66;|oHg<FEz
zrRVfmN>yk8{Kqdz;iOm_XSG#+!V%TG>_~e-lahIj**S%5ea1CJJTguu-;M_paEnNc
z8<o1Xbhl43e<}}zYZ+h^G(2S?@x@hSdyl8Jq$+y(LSJ@_Q|kO5`R%_`z>rs-W+@^?
zr?pe(U+heFRpK<1!nknaK_<pyo#|?hu?iWIDr)`{*t-u!!!f>m^2wacI%g5;0T@^=
zv0Gw8dk_jnkXzJ4+NXMCYSbvwKZcS26Vm(Fi1)vZe5FV<@u=m)O)bZs`YJra6<vZ>
z(lD))CS5L)k4t6p7evI?KkV>ewEPb3eaEC=q0Xr&E`DNBD#jNB2p<8F3*o3Mzy^%@
z#XX9<+~gSZNI(=EFf+veMQ{5bSKuA;`>P9G`kU_Y)sN&(;tKQ$d|cb$GHwluNt-r-
zIeT4V9u8M9!J{RSQuI5WSACe<en>jQm>Imz?>t}g<Idv5^%`3kM~ESL%z@}vFoyns
zOlf)5SWXn4Ap5$&SH381l-2{>mmb(V#t4r!{r@ui%(dkD1AqKaw2eP-Yk&sq$9nGj
z;@AEn)cv){>^#W4aM%G#d#Y?svL<U~;ZC)#vd1<sY+JJeGTa9k*jiQ7s46E+L*jiq
z<*?(}(H)p@TfMl-o(t`lY{gWR5yfSoG9W}bYVkwrH~;&4{9OY6`9lmFko)d>mj2sx
zAHK`cP9RlSf>Da~n~rZcE@v|7FcES2e%xeUghr+x)<$E!7@?gVJVJNVLHp2u`<MUy
ze;a}A$YYks@=t`N|1fh!e&p*6!j@dUo=p5tRK5Sa?szFtV0%tnv&a1}Yx3`({Ld?w
z#Np@TDt*<Y{+H+Re|$*cMNWk6`pP%xpJ4ZY`uuN^p9lAv!++WQ{lBi{1h9Q0&bMLy
ze}4YyRfH|P&8|xHmr(U5)h0X(b8vBeK-f6Wz%_crC4uR-2gSZhS*6>~4ML$>=YLwz
z|NgpuH9`k8DE^uD_b+w=Jpkko2N=8c9i+F?2fbO=1KIl1rQl3C2js=GjpY#BT-3nq
zfBekP#><ddR0urRN)LS)V&?*sN4uV3eS}`58C*5@dyypIxnpN&pZbvitKxtQkT4HV
z-M(vbA{K&&5AXDu>kv18fRJDhwyUaYWRt(ZP_AJ54$I}h7L7tN@cbLfLr8uPW?(4$
z9zb_*H*5PF#AkXrE|~?YbN%CzKFz`*C$yz?r~1F1y-T7`je<n$C?JyfGKjM0VE%Q=
zCPf@9x#qx?l22>_@Ty#d?6D36g6iC?%BQcKpg`^gVgN$ivcTUhY`CM~2{KdvJz($Y
zoEafWleG3$_yAqojgYCgly3j>6@8}^<_y#VLkUExF$>KKqQ7a5&~=^>Y9d?iRS;)a
zxN><*M$c*DL_YwRbAZG6y>N%K2D5kKv!mbHRP)l+-wQxzu=_44C=bxsF7Vtl0)NbY
zD_~n@P2AQwPef*KoWPlR+bsIAQd7zvnY@&nVRdfz%l{~sc%6ZG5U6WvuKkOp{*;FC
z!H=}J&ll|)n0$H`M{4S|d@+}F!LhV@1*i)QEDs04_*LU#SLX%1^COsBcyT&`AiC$t
zFVH+f5nOi1G~UT{2i9&DNS>A(nmQqpM14O1<IM+3G0T4Qy=F|wJ%8RJ=yEPMM`&86
zom$AFx(<2ja!|2!TM_V>x{Ep%DdCotk_+TOr5c)uqFdhiHv`d=c~fDxZuHupppVP8
zbXr6_8XJsTC3fB;!BWqg@dN1{C4Uyf$qeFXnv(a(l#s<(?*8_l7vsMfGEhCGVOoTA
zihdYL8N+W=nmaJ%rHC-2kDj@tMy&6B5b4dWAOj}zE($jwjl%s>J+)7QuuiDL^O`OM
zFQ_6RaU!7L?g3XtBLK4v27{mci`<y-9sCK_-j$8&OdrIRA_MgcMCW8R7F@%JG@3Xb
z{($YqL6k#tCvG|>XqI|ExFX-<Lq_q?dNmmdn&Q(jRw<bWM^mmS9`<IG^)I9SAnkVx
zecRl^HZp9qmiNlx<BvI2%Wrbt^xIq7%QSNve7iI2(o!P7UQ)8$vhcW90_+^4g=3pt
zx)d{d&&=M}q_HkLM6~8R-$udZ{U3|iRa3Yz*C)Hw2md}XVt|yc|Hw**DgNCr2OW#B
zkI+|JD2uO!T~{Fb4jJ2zW1;SLDM#gWIt%qVxXrkep>|a7Dfg5*)Sm7-Gdu5<^8UjV
z**oQN-|4@3?SEavf4>RtwK=>o&2_b{e~q=JXpe1nSN$?k7F)u3HSU59f;A#P1>)Bb
zVn5SUzXHiOGL{kO?AyRid4(hytOJX=2OO~xcamI#n&+~*!VK&FV$aSYqVHVTaTmlc
zsys&oxX_V)zQ}IRZjv4bRthIsP8*ASfx4(pPkaVIntTu<_74FRyQo-SR=KyX6$`>s
zYebg<zJ!gbkc8}<$2Is(omi@x0#JyZr6E~boeupgF4u(Ga7BQLdUc=gTLkbh=faR<
z4Pq0H@IgJfH`Ejh4j5)5a-V5xFmTF8T)`oNoS_G?2X@kM<`b{qL0f;`LHCo_gC)iJ
zJ|o!mMJ)vC#i=c;7{^gxYe)pCD~iWqK@6z5ng)zD#_g}iZ(YOeg}VD*C~;t|qaG3a
zRscjd(rm;VbX#OE7Pn7#@xg0^{Mp7nvlm{Z=ep!A4nU#`ZJ!Z>11!OFYuEerjm%|h
z%5A|tVk~e*ZCB^uj^V53|MuZzS=h7O1W>n=K0z>#fL3rGWvtLHO3`Z;`gYA^%J0(a
zBVpblX<jG^nE%bmOKz+9%)SMygm*H95^-xN1)Uka@vHk@q<b&ER>L*o<Wru{5i_SD
z+sUF8aeG6nu&lgFr9WX#L#ztqc%)C(uj;w2g^$z<UwyIvq<UW~twM?R{N*o)QCfu>
z(ab{Sv1=%Ng*!xc8}cRExC41#pd~l1XT8OWVU}*HN9W>-tEJ1xK-H7|=S>=|2k_6n
zEzUOYg$c(ZT&^6QnUX8d4%Rv<1+Vm*&Z<iUlZpFGG~HUTb#Ogy@|gZi(pp4emm4?@
zyc4hqHH9?y&F9I=w`eA%D!X&anxDA_H-oeFiToWvQ6MD%D^S^{(6a#)Z-F}E?sliZ
zJrF7kMgV<uQ#UQ_Zqf?p#w|+KZE-*AQ5XM90v~SOuj85c?&51SdtyTaoGs-qZ{sq?
zk~9bH{~;px^%QX6a;84KvHVl7eU&F;+@-(l@O}{B<8>%VpKp@dV*E3C8gD=a`#fh7
zPP1$U{0=nim&!oNsbt86JB&APnDUX_HSoHCGgjkW8j=Ktn*(yP<55qk{Y&tLRK)LL
z4x1YiP(1RuYDcHm3Vk$|uj642qGSP{VF`j*omgm1D2MQ@IVkfBK`S9j!iiVrq^+i+
zc?X0pUc}m7rYGu2QJ<{_84At^HRRq(eD5WWm2`LI!LiI=VO($0`_-=Nx+IHOD0-5+
zVgk>A7=wwpIWob4(MwS5uGQPXUS{?;%DdwnlDU9Oh5Od#IWZE;avr=zLoo0~d%Jsu
zc;_^r$1S~4ay*aW0>kaRys-E5Ari>c#-F#4QLvAwMWCKQt_mxju3HDUkCAV8Jo9$h
zuII{luR607(Sk_l1$4hFGh70=W)6WPR&9D#<elhP@FeK{5|FWIoRDIq0ic@O3aYCm
z9&Z_fDEPE^qYdD;4i>rMc<Zdn6iCLzdYIJo!3B=<z*y{(_r-a*ugIR7#T=Rn0<)T8
z%G}e;B=ebJ4v^M9O&y+9=zJe$2qGl70E^k<qt^+}TV9LAJ5q66DQYMMv;m<G0**8S
zP9l!IWKFPkS(Xg<QuI!!2P;0hH>sWB0J&_zcaHQw6R)pCNZ`EG0Zi3HkOeG#JU+;!
zFFn2^0F{GwQCIFwAAETjg&^!X{)h!Cdgi3fxw;#SsT{MyD>n;mgE)O{DY$<SHYbC(
zq67pC3Fv<Fq(CdU<KN#7OKnB;?ITQK>$u&G*x}t&o^05)qLXYl1fNGmet<oi=hFVn
zbWVL+tbhKTk4e+F2KP{h54{3?3QpXEO{dFX*}dXY#JBhivP`gJ^#1PBl2Q+D>}vyg
z)ex%<*Zw7ZFXGJ$cj{oZu0CZ{{Ye)5_Q}gLjw7e+M!}-))Uai<ZEr`jm>+K<7O%Ax
zKUU8ez+bF7|EOE18?36XUj|a*=nidcT~w0I0-V(X4*4eLMdE(5)M0NsmqIgN{*%>~
zGXoJB%0(|WUG7ie>*j&-4@Z2}=D=Mi7d8yPSmKC4aP3_&G}CqooPr1m@aWT<STgcB
zsDKiX^6~IeGl=K!LxoJiFXMQnhUKt-3fdv)(vNqGt6{ANR9a#9YKre!<nb}1#TlY$
z?tD)<f==M1?8ZD-S41;Vm86g@bwiDKpX1d7Y3KCGbr|gW)*6dZV@!_=V939qy&X9W
zMV4`=)ogxn5q*cb6$9}{q_!&lwD?wX&w?k4MIz}s5WF85*`8??avZ)gTL|qc-C7I#
zb~W5PbffeHT}N%g3WIJdl>x%=9$^X<@gep9-b+Z7@Wl2}+qi|1DvRb2c21MGKDhGH
z`_2@MAIxM&=;s3UF_<%Clu|sa-}KVS%JU|q#PVV8>+QoNw%3-0E`N8~TG3@x=a>N9
z#92Y6`?#-uW_1XwnaSbJkmGmHL%Mzn2wb|D{KOhuYnU9HoxhJ^H<4&grW-mSKd%Az
z#|s8tj1<|5XiF+CMvn%L;!mDM+6LYyJV~o%#B|&F>1x8fcaU(=_A$rzGROU+p7G97
z8LBC+a^tXb3i$0DQhdyGE1G@jd!%v>V$6`?{Noy_2Rd!KDU4R&cs%*ibCN^lJfm^P
z-l(T^C{}T$!!V$8y5#oUeeNQL%Mal4kQ+wa#f^^W$F}31Op1vd7~v7&C_FGD-dPlQ
z80DX*(b;&bbxRFZ417~L27)eASIuSEH7#O(bQZzl>*u_Z(H{A}?!MFW;EX|0)^t(+
zy_cJuly|qF6{*NL9G3o~d^@fgJ3W?l$L&OiseoRb^DIq6t-NnZ+`~La)v`-B#pzL`
zC=ai4Qh(y?eQu+IHo>U{vzFhHo<Sq@&ulz8T7piC3eNIL9{xzqxWXW*|GFyZ<au_&
zR_|`ko*9(jJszu<QCcmO8k<XG!*5;honw#c&Sls|6EApLSUO-$X7&AZG+gxrH<rJY
zf0ZWhg(~Qzr0b_K#!+e96YZ)W=^3N-)Voxu)?y3G6mh>F)L|*GeY)he{&R!%RQKA)
z{VTWr%O=GK^We&rQR1frQej^g%bz!$_3W={3jrB5wMO_xaeX2|_PVeCiDR?^tbFa9
zWv#ULscE0FRc^*^CR&W@sx@ZWeXsT)cvE}^oGPriC%fj+MnuFW@xxL75?}sOhhp71
zFMbgYL{MGWbf%?DU%>X0;izuzC2hM<wNdO5YQ5>X2bo>Ia!oh6vs<#Zm@h2}Qd1hl
zh(U8_TgG3_F(1v_QdVC{@Mg6gs`$#vM{-sB=G_S1ZVM$^4K~D`e>C2uG3#3;->iZ=
zIAL+{wO?};sGPlY2)OU}N2^N--iBl*jqF1|<C&>x{n_|R{U2^syRJv1@9^#V?dTM%
zt;O$>ZU{vboAW-oWUG9|m;5lZWaGE2anC$QA989>AD`a2MJ~Ehj_T=QzC7+S3jPhx
zsAAN4TgphXy>qeVdbILv^PSvL8X5ln2aH~o=`)Q)TRqIR86~bER1mZC9cj?z>VIr2
zGb%llXQ$)#ojB?HOyZYR-95wve`QsN){m`2Jhpfqd%gy1+B&tNHK&o;#iAE)8U%%v
zY|pO(@#&h*7cQ>@D(}uY|8h-oNYq*9(#z7UPm6mHY+^vZ><}(+*?o7DARfpzPnI)!
zF-6komJTkeII@)@+b-AUAx!SPd2KNUJs&A9oOn=OQByad%FRuCmV=M@y|WVcdiv#7
z19figF==jx4|oC1B=bjlnh!s-)y$XnTv}qVWZe>dwIE%&q>YMNj2lh-nC|qgb*p8G
z;XNy#(Di{b+tJao_C449NoyC($~0j9uU3C{^Pd!Vwx6MI6F<q0RlPiNIe7W49k)@q
zU};FX8lQuoW$^879A+%(z3Qd^1pdJ5)8QoOM~Q#O{CurJGjcxU=V5YOx|w!DmS}#D
z)wcEFiF=a<r>Hj^UljXP`gMzb9_98=#nzp(UEnd<VETCFjt`znz_5@|%|)JrjnD7n
zJ=5}mxiYDlN3pEC>=^Y1JKd;11%JA2Vz^*OL(wSQ;tivZMRZ`ApGy}<+gp67iYm}#
z;Zd-7+YM|gze{ONrAv{otj&g|hC`sY%SJq%^k>EDvQ^WK=d{wtX+HO#MO_=NkEoQ}
zqGI@(K;^pp3rV8y?LR!I>@7b3hSN41r@rE#{PefYk|Umw=Q6;phdwgkqIXEZbY0HN
zTju$ZG_G=kSIn!j9lo3w5(aj3(8n=k>Pne-iu8%z>8;z|c4^;d;nA6AynUl-bcBU^
zgC-j5P->m3SEw0$gS_+eQQ4ls+t!k@*ZNs(V)JTk-lB5i*15ENTGnq}e)_TEg(JTk
zR(D=pXtH|swARLWY;pHTDpMD_@3Sd+ZX|;UqoR2WozfTTM};af%r0Fy-F0N)_UwI(
zQ9-oeRKgWh!|vG7*_BAe`sUAA8nG)pdOuwRPoS9D$Lrag1PspBG~d9ia1JLlw>jA|
zOD4W<T}kp3GSJ&9dJvpFlh#AiiOpQ~y<IPV*W3P|(93&doFqB>+<%8ytqKu)soV2A
zVHA0f@?Qdl8;ek^e>d~1lLn7DV$7MFeAw4WwY1=SV>#XzuAzH!LoGwnphNq$@qAzw
z4=ewhJ7k}(UX<Ffzdg5fabeU4z5ioq%5o@Dm-YQmuuITduUyqk$@FIlICsy>dQDcB
zC7F2*n#42~9M2m=2FLG2cFMlt*lfqFHbJ%K<d&CS8mt4S-<akHnA7W(2Q6lLjVQh3
zFOo;=R{u=-EZ!zdXIwp~ewZzGsJ@Ff|BLJk%q8KY^3l)lupM85K$9006;=tAGI+!4
zUznKS9H%%oz0Kgo$ja|}8fg7xJDUp_>6pc=0*=;F+644bg*!$+l!{07c1LOBDk{@{
zT9_+K+Z`*)zwJ2%d|x`T89UZ$H1++EoB}V(tndM9RWB6VHU1GdxZIS^oLulKVk3x*
zgp>qJG}4*5o}bT10S<2az=}>lsO3h%j@Y{A;y{3G^`=Nn`>(5LoHUJwqE{8Wy-B9+
zkY$x+A>6xz4amKVDN0yQof{$kh*s83qX6kf_E9a_;Y<!;Lyq>14>h`-x4Ks2J8n=N
zi(D)&JhuLkR27rgFS};T{pM=!o|o~0m3DFDVAQ>SlMOV}4Xp_ol-W;QiN!ZdCJ(J#
zjX@`hCc$@)tL~O*b<kygcvU#R!r(;MIOO27^+9t~>tk*Aq$HhhmEuN;{_t)+ok0>z
z^M6vj$>A2iMGz~6!U<<?ApOGZ8}z_64w%G?*MIT5W1M3{7QANJPxNAuIc3|AO-9Op
z$aQ@>1es+YhlcCY-~3@@^0R$R%y2Ok*g%=ji@#G-qL~p7%Q(s)A%-Kx#%J0G#zB+W
zFhH+@rlG<(L?AM=;(cU@GU%s(%I$+iHbfutyB3IFpvuz$wiJ_s%)m2ryG&WMkF4-U
z)^Y#ld4tx0e5LCKH!F7o!<hE1lImsAD2#pS<Gl3IcUIANnjcg8PU^BM`EMggEwC_}
zv2Lp>Mr~EMr59(I&i1pT?D@T)hhPNA<`i6*-`gz&G2MLgghk3KlInTs+iGkPG!_oL
zr=Q=^2Y7{t+EOr!E*39znc5i96;2vpzSKy@x}2U}<jXqWGChoD!hCqaahgh?c;m_n
zRg$`QIh;9{j;DU6&OAJ&`bmTLQ?$z=A<C&ck6zKXWou^2zuZ&|F0_2?pR;#7HGbXv
z_c@bJQWg}~aYGy0pdtvb?s6p-Sb`xnt@Mc6*SUCcB&y*Y<={|S)TtexuV*lw20mwQ
zujqWy=8*<AXKW$z9Nc?umhbT4oj~8U=#|z*-jA*|kFS{dGQ?W}eLdnthe^U!QjY|;
z^}Eeez4toT{iEcyPM~Ohf5n>xmn6~!x}5%@_8F2zzT?jL^uRygF$0T7{R<>qN{qxD
zt+&`S#iS&|=~4d@Nq}`<%g_twEp|q$(*yeAgqN$iciZf9&Bxk3z=VuuZj0F>#AhbL
zDba_<Gp-o0%w4#nCU+zkTF&R2){u)rhG87##6kEoXv+@Us`Iw$Z#}%A`g}J(5KGI1
z!p-_#I9vX?GnOjzldnCXs?YDN0w9qKRH{-_8pkKn1K%hpJ@5l!9Vtu&%Q<aDV3~*;
zuBjP>c8R`PmQwT#HosIFh2#jxIav}JxQ<i=<c-Azdv!}uj+_oijWYDp4371|99_D}
zyN(vYIKrKx)^PC&(3k>k@rR`9J^|DA=8`yk%Uq^CgwJrZtf5;h$x!}3pa4lCvo{GY
z`19?)`}^BH;>d9*4HdG?IGPy_8ca3BeAzJThU^x6rZOkG_C=aTF)ih4iSrvYael{8
zs#$>0JdUr4F&KP5$ZHuOa%*z;JQ}r>-uyw9m+7)3;p1gqk>Z%g4W=X>vdu}EP_|n^
z?k&k&`;X0r51B##5d)SYaX$N0mww>R3F<#)A4pMKL&T0<E;2kSZhUtcT$9*<hNUv<
znM@GZ3m=?{_j&=8Z4TRDSzn)B?y?BU2L=X$gZAIpus=S`zVJUnP00{@8W2d6DgTQP
zlm&fV+-HO`AR|ohq?qG_(a|bG1>^0CVc8&GpTJD#igs`3rkgXpa;+sk)db1&)RaWp
z>+gK3{8zB2ie}5nvD6!jcN|)32z8=;8PVh9yS^^lR|k~Tztj(A7>>p>Bx&=Z-aFYi
zF99SPyff&c<DCAZt2U-Bdmuwsdgrzxc3E`Ujz~YH$@`U^if(yL%{)8Sv)D2|-~jE}
zgD+>V1Ek}=XrJxg`|ZRABAFzJYq0Yra+b{#cM&S?vjY?RQSa93M_Ca_E7LwUH$Nit
z?c13mgm@<iwwDxBWy@avw9nr9pzi-D#L9`qKc@6hPmr8`{VY9bC2?ClW9^;h<|Y1r
zefx#+>-sGZDB~Y>@mxKXOFdVh@rbs^=GvT>%X-FElz${O<y8E+Y~#5YpNq=cUU9__
z(kEARY}>r?pTPoLb4`5}Mo6&A#iQbxeHrs&KI-ouaB##zX&cyf!%w5zgf^v)&VjN(
zc;WFI*B-O|wyT{x+N=y4#fLtoZ;bOF7YG)d-;`_Spsa{P(i2SHZaGY91uO&${p(nV
zM@8XhQs#d|!|_rJ?D`dD&gpUg`YWq9h=B-mZxlOVd3zqlaly6MC`0>4D(wYwyQ=x5
zNjD~6AQkLq{@Fav9|q*ql$5WEHQrNfnJE`<WD+XgsGs05T`;NAU5axtB@0#O#W0)5
z^nC3n^*{(_qq@t|3hmKfDMsj|6Ro7-Jeibkp2V`{M>ek8fD;3iSDwNU1rg(3d(B>i
zuW=sf>sN@?!|Km0DpdOhBHm`@uTH-9VEIAAr#<*}n`qA4W(R$SRFxN2U{&kua+9kp
zLfjDyO<t1H++jQ^Y^k<xKPw3aivpEBx0M2FylFH!qw7g4I@=K!D4-U->zm*c4L<t1
z6_J=;ugJ!Jy30$Xud<g8P`~<Cv&^4rvG&bydpvet_*UuuT)P>jwMYrobHin=O4GhF
z@-XUTO<~;(%s%1}?DR9_xTr^*^<<gkh+nwI(_8_r?3U7l{h>h9kIT6bzARU;2H8OZ
z(ExU5{KkCYP|qMF=i6oW3^Lc5n@wa)T_TpM6D7V>NrcHWr5?ruXFw|tqT)Ef4_3SS
zgKQxJgW;VD)QK3^&ZdiN&QNCt1zXM~uKYdd{P#ap_<N9eRAh*&)BisHpGru3c;6)<
zMN=V|cAjRwg8v~Y!a*!ns08o;Xj&ILZYg^|XsGdHLjpbzEf}GXE_!<6EdcRU;&QS^
z2<tU92=^jP47~45hDb2nH~DCIG2AUc{LH=iiXUw#EcV`HfARhYFp7zc8*?7gV#0n~
ze1!~_*swGrS5C`}4&o1pmmB865RK#+j6EclYm`(Z3RR!G>IP!-J%SdW$vX5gY1MZe
zV2sCpD-D&zqA)uq;&q{vb5i0<qxIFIr}KQ*i-C_(6Klj4d@EAd*`k1BZVePok-Y2Q
zeys|7p}?kBFN~I7+bNgHjlVx)HUXC1R&ppSubXS_Z25DhQhHzyFT<Nh_$ilt`c#8{
zzIa?0Nmkueyl|rU<zbh__a7iaj~#vqn^6M^8Iq^8m--GaTtWPb^WbKDME9&9(#PmV
zYK&$2;tOA&Ml1T!+K^u}20<@D>nAC;H{qWU@u|j<!_l|dpWnO{3`^Ka@{!8p*HZfu
zM*cMB8ir}2zp2JhK91^K*RgZqg7+RCGC;f0Tw&Ga%6EGyq1<0h8?3yAll*v@-8x8~
zU{&>9x41o9Oxt)>Z~Y`?ubNtwWeT=;xgc*ieUP_$o#ofs*DisBd5H{^%4hcaYVJSZ
zq$FaT3suhSBU|RK)}3flU$E_)qN;7<&-Lx-l@W?`nl*mKlBWkIl+xNhA;JlcrdZm=
zU#o8=16}gPN+Oq@ip%4I12=AtZ1EfxtqT%JaCV$hbq`2_^*$>KCP~#4s@*{5<OrW^
zyFI7=IJ!nL+{X6pi83t$optW!(#7Fu=-Ygq7+R|wRZWHxgTH2rzkiemv0HVo0<2p&
z2<bDglo=BD$Eyw$$0zi+yK2VCZ^;=3$-at8xUh(tW^!tp#7(j_PLDJ-O@<`)C520<
z5H=M`EdN13^wovz8m@Df;nIJxl%1)FXbpQ27c@SxbC8Qrsc>&^XX64bd6fTfv-iPC
zLw-Ahkdn6$1(6+3ezloYeNvjIU;o+BW?A&)P<ONJl}V=OUn(kkp)Am^d~k363eSG=
zmS=LwAtAewn36**cFJ;-o>%PJ<i|(ID<=JGC;f8}1(=<UCt}nTx%}q+JXCUFYM1I;
z5UmujeF}xU?j+5YyPOuHH4cxnYvn-#O8X&C;EVX1XctN$+BZ@D0iHx_BR!;dCg0rt
zQgbqve4yF;Zm2TGgp>Q`vqO<O4$&oMXAlF~FSRO@{<Dbx4vuerre(k-u;d&|PZp4i
zZGXXjxS;o%Vtrn;1sT;}To|GVyvN5`0HYO&Ujr*|oXqv?`0YO#?`|&1l@<4}_DNuO
z7oNE0O7u;`9hE0bbC&>tZENPrlvnQY<p@y@TVfxIZlAUu_vDU6gkE4{ae+2Tj*pK-
znAXiSUzqkiL3HD*DRw^3{!7i(J|v>jDVYP%dI7sDX5)v*ho+cOc9gD@%u$<mpRR<(
z{i>7Ds*?$wfvC-M8+elrAKUN5LUu%2tJd2`#70UH3Nrbg=dE9Rd@lrl9I<PhGVFiH
z{mSCNMm)_JK}Ic>T%D!L3S0iNRG`Shh*p5G^1!IM^c=_X;>`hsmn|D>nxMQfLm5C>
z_qee`+^M!W8FZv4T~AI~FeoMPKKwbFGA*-eSI6b%le-`-XZKXL=k)9m?RJ7N<<WcD
zRliD0IBxaBAln*-9D(l7GMU6d^*K3u1@Qz<<SRI|*CtD_5V>t!Wtoai@jRaMEqyQr
zo!RG3+vvqx=U{LsXFq^!BgCA~{QNfOm{+CGU~4e!6Ou=q)w-d|=Y4<SBu$R!La5**
zvl(QYIra{$7&P1U@J=$#vRhBSvHMc1a7MR&-|sx7%f*G9M`8<}PO;#?WBBSk!+8m9
zM=Quu@`(et#dS!#%aM`DjMv{S#tY0IS)p*K6|ar)x2vJZ%4_6u00C<cDE<?JdY!@^
z(;8%#)|~F*s$boL-ZjHWh*zGwAIYSv6Iprbzh@Y*68Cv`#IrPK(DhlVSsBH!X?f%V
z{r(%N<4k)Liwq22ZVeF`?H_cUE8t(NbOBODeX+etW=V)dC9n1ahPx|)R~e>Up56OZ
zL+wsS|H$rB){?a>gL2hfKl9+*J1`SuF-iZ&q6C;^`U}=toquR;uW6~JNIX#0;zy!g
zjC^I4(*37BR^~r*5M1U-zfyIoDGvBcwN^w&yX=yGI!IoIKS9;Vl=YDE>}0$umpFEc
ziephMF&lA{Q=_2KzI$_G5V0ZWNqzY0de}?S+wmD3Oj13-MQOY*`@oYfu+{{pf-e||
z4JhFdTgkDP+Z)Wk8y0$~q$Tm3e@`-VBh^XatprdQNRW8UjLlo~{iqQbJ(cwmyf=U$
zlDnY$wz)#Dw$kMdd6Dh+{VNZu6U$uOgf`w&&po{vSR{AT^W0&9lb)^b>CU^wn>og>
z;}jZ+MGdiDyaj9~Qa63Sdwcb)fMYnh?a6xl1l9Sjm7QbD>p31G5I?YN&Khs@W>CDp
z7j6*K`A>q4A6+>4rI~TB3+`<P@fP(kliat!%39qbVoa22^m*sI%!qr&CjAYl0^W4b
zb5(hnq+e%JEijrUVbMucH<?~Z(J5^iyizW2tq_0i_*I;_4<;$J8%yrL8hj*z%xxf_
zW0sNe+RB*D>K5oa1sauyA1_#cqO6>+J-1CFjsDS@<)t5;>!(jyR!f&+^CCwB-$0eh
zY<`1U`tzVeKX0AAbCCNp&IfP~`Q6HqEmsi#=G7Bg8J^0#qYez}CY6q&6J<&#rzK&#
zSn&-2Zzw#6678Lvjdz&&*7ImXlUBV2*(|umTy>Z5$#2s%D%mcYylbLX*p(xO_EzWY
zsO-p%j>iT?(P8fQ4X-RH^&e=vFcv>GJ+*34)Ew7CPEx)rpiCR*$T56kzL&Q#AtGG#
zKpTSyEfd|w#FR>-YFI_gSGzkZ+5je7B$!X^yzeYcCYGucjRywK^q4ERV9{2*{(T%t
zE9JrVUoYII;KnvUeEQebYTdTGi4L8<xpaynHd;Pc3I8)!iM6qwlC_=T+9Gd_Ym0V+
zYxM@{3IUD|jhL~)F$3+XE#|(YPKhg`nr9fs-~0{T{yWeI*OvmS$EE+a@Sg^5x|W1W
z;(IiSEecz{_`>rX`BMxtmgJnHdYU|^M-G}_@;86K&JjzAYvRejCTpO|_8yxkL~{!R
zU!Sop9TPgfQdL<eJbui*QAfFYM@1LcVCRUD;gQ$H#+;*hh&i-$pKzliSn`-BhgiDF
zEsROSWOswa5C$WVN~WU5W!0z)O}_rsSET%n*Q)6?xlFIE=!c#R_9@^XpW<=NFTRc)
z51cC0^p$-OFaPz^N*I|f%@efb;%<%bu|0C-OEaUA)wDRfw2RqYF4GyEgp^f|Mk$@8
zE`=wj$BA)2Vk7w74!KnzOqY}CI!%XBGx8RmylL^a;cf?)NQTzUFA&`EmNd|zscOC5
znEWsRr5b$}nxCq)oju?d$*DE`y1wkmh67B@iHGXgo;*kq(+Qpa2K)U<fP5XV$=f4%
zsu?{eCf4#LM(=6P>s#(M;;WGb*2R>UpH?)yg_bU9tsye$-I(dz{q0`>4ild$*A%#6
z&~k@RnV0bYvG?ZTRPTNJ_abGXvW$z23mM9osLb;aLS@QK#s~>Z=2}=-NMt4%B9eKY
zhmd(nri>Xglezpp-+k?UU3=gAy1MV@{^vQK=lK0+uiat!4xj1$KF`-#a_2zcUQ|*F
z<6Cj-7NZRM3t#4s=vgFND8)93u}V=ADwCEtv$|zvoWr50>otq^{Z?<mSM#lk;_kwa
zNG%jJRI2(F-hRpO9u~8^;*xhYd@DJ1g=#%k!&#TLTBNEBUNyE%v-%EK^4{vS@5m9N
zX>)Z%Z>B0r;)`CYkK3-H%q#21j27Ie=8K!5PwmYTUu?W-rI!=m4^SQeeL<v|WVIq|
zWml&5SQ=IW)bQFqE4TiA_y5mNY678SKqikC#h<D`e)w>p(WZW0Quuqj!SA7ZN<1uo
z3~6xsSKL=W2_TiOml)m1F8$N_Qpy6zoV4dte+NByJp<q9m6oKX@~3b7F9=e<30bJ_
zG!UH~^ygnhLI6pZAV?F-;1o~`nFGfnuYW6pfEfnhdEfbuf6w~;pFcLn$HdeC9UF_-
z8gS{d;J$<79o=b0eiL8&>dh8m00(oMB+k13&-?Iie&`K^##|tsvDTmYsryWRPO~i|
zpz>q`V))b&6CsUKY7l#H9SF;+2e8HObqOVZBx=`SVfgis&Jsv*sKAL~-E7UAkqxrK
zK>~>pNb473`!4%eie0I76twc$sC@dfusG#3{%@it5bB5WY_lo?8ikkDc=LHnf5^?A
zobm%zj|s@}a$aj`vaSvkVym0nlYY@Zf?7xrvUr7YHRblN{-j@th-JYqu!F}u{09;L
z*WdPU{;%JfG<fj8*6aQHPrC}Vu=N3sOP2rn?fNe-+h70l|EIUWFIiy<KnOxuNEyY-
zGpP)Lxw0OTmx9W09B{iDK$%P~kIHyMq&$e)hKghI9<7fNKuX*r;A^u3xwiR$fw5yx
zlU?(t0r|HV;VWDT7JDjH_Fwm&^TltRX$Wq})dBv1F{sYkU*I~bZUX8ECBTzA0kY@b
z?bfL}EZ)Wk;*h5TEkJfilK`?`t9HKB%voTE7=4a6Z@aLAxO;uDD~7ygTuI+@MS+PT
z?+0L=T*{dMvPe*<GqqS)7LU~^od?>M@Gn3*6V0alHY;~9ElQbb{I%pbkV9H<H{^p(
z$Ov1cYCmDn<5!m;<cBL{_tk#~Fc*f}{NICkWpk1L1P%TDWh(}--g@e(CCz_b<RT76
zG6ZoOR{%)31V|L?89g_o69Dl*#6$Hn37hZK7NoHR8_LppgO^%v%=yjaD#JiyWW!gx
zpNc{+uMyF00E_Q1NZ%Q?4}JphDWibTN^V3+c74+g$UMhDhRJ|^`|$gx^guke3nEXV
z3VlE&!yZ&$ik_Vp$$I=gB8kZ20YIvsgV=&A;8cs70bChSEc*bZIcyX{q`(nSKV@nh
z1(|W(8>Ke@VCMT1(?#GpHG#lfUO?y7um%1oB1ZVZ(f4nkmE2;d_GTEUoYp~;(;h3g
z^vctRyL{@YFp!Z1t{T|B&0t60_{LKYL~zJb;-jm8UjBsX!9nfBAjiFc*JuBwbN!q0
zwUP%2lB(<nvP}Lt?bi6!gGB6y5Off$3i)6l)&`flh2eNUsH`MFNLOPxbLIZ0Cx!_`
zXF*d*%#H<LuDAaSh49qtK*mByzdT0y9h3lzK(uwkEs2Bs-zMvPq8<Pops;ms%ELP?
zu2cl1zCIvv{^kfIp$b4sim3=kvPb1gkGp_tk0J2w8l2r7b0`BK-9|tV4MGH~0d=b|
zFn$N@-4**>3y3KJx_01j>u^Rb+<hL>0p|k|4tTA~LK#GdB*1+^NbN=iPnCofcmb`K
z{K)1?5DdW=0aR0^;A}nd8!;LzN>5a;f0+b&HWW;3CBF$uuvsa5v&iF>+Tpt-|Ib_U
z_a7)QVCkeFs51ITf;!|{`3yJk-26}qsUBh1@h*6Slq&FBfj5f1e(Za@E4h@?9$0op
zp`bl*i_ArhwUj16uCx~VOMAd+KDJrvE1*3Da@94_o)1ttj3<l06Kn@8uJ!|98a8`y
zA(t?<4zdu+^#cXkEsRt$&;>U=ScURgSguqPL$q8#c!>GF@A@zuzwYG6dQ+8|Iyr1+
zGdh1UGx1SyF1TyEX-q%b0O~Jn0}~R{D!N`h0Qz!$8>+`po;?Uq>V>}`a$Sd<&Y)40
z2JEA@!e-4_BJHr`VGed0?wUW=w14wPrF4K9XV@a^JIf!V(XUn&jL+Kf!|W>fu@63A
zR%bx*H`68gRs0dK4IiGqEXOq;b=SNag>TaPQ!2a_&7f#`_2{6-hcS7k{{o<y@$UkX
z$$f(=5VAF(;$=2Q@xcg)kK-XdF?6QpAHDjGfr``@Q=_n<3<3hkC2B-nO2OY+Ng?@C
z{y)2aQpBOPkyK~ykEdtPu>lG{G$GaN)!VnwOAVsuroX%a_KS@PL9n$bsX1Fc1%#*t
z&|TF$FXp6qzTNlQd(fe{62|onFe%1gi(x1|!Rx59un=d(I2$f!1h<d?8K3Z~YhX5v
zM@fAA2JFd#)w-H*z~ns)isIzloaz$O-vb=0ZQ_9|#SS#xlu1>_!0dcghUE<hJWa6%
zs$1^dwIdQxq3H*YgSDD-THFSd3UG&=VB4Y{4ugU?OMn)Oee^px@(PD#0v|M-R&`N+
zxiwY7lr|>=%xR6#lx<DgE&+;xX97XZsUzS#gDu9TljAz;7IT9=;O6pFoer1YDY{B_
zAxdP90!6V&?Z1a2U%v-t_;Z#9D&JauLIF%DIGBlIdf3DCmvnZnjj|+go|-lr8M?XA
zBuGxQ3k(;!7<h(H#t;8m8PWRJ+J^u;y9%28*aafgb+Dq1fT+mOQe6O!PF;`WBRap6
zqs$(ct{v@Sn07q)OirmEumXl5DIa|LV|gm~;JFBdmp7=(=f<yN=+j1IyZ|bE7Z_ly
z)Ln*+K~MU@kU!yKD|)i&J-@`%Esvj5gp&F03rS@&`)zfLKlUcWoo*Ses7oAtHC@j)
z#0ga2HO+&x^XRC8#3#p1myx?$04UyOMb{c-{wB^c8=Y&&K}QB7=Zz7G%%|AVscsbz
zUetN}1)`qXj&dcR0Lf(|yNRG_Gg|qT)p*7<w&I!^&{|sz)5xjXm~n3zNT42?NhsYB
zq3)l(Ii}$6{N=;Y*KhsS;<2wxvHb5|oJ;kx9ssNR0KDlQm+4+y&nBgkkMB8Wnu8gy
zup8VA6ieHB9tqyp_`Mg+i$8;<5NRHN5P$H@DWs+`%}^wdVwqYrfj!xDt-;7-eV?^~
z;7G;i?OjkKIxf{K&><^jH;#T1PBe!rYk}mX5PU5-ye|yw%Fo~O%(<)5<$34pw(X#$
z<*Upu1w12BKeRiO^Gel?xwkwjMkF+IJw^Tz;qmwL?r)#S0~b#|Jh)x<_c2u&TMzoD
zmK;fsZ=)SdO{oM3NLINASyD~#fD-yz8A$NT$h|0w+XQ;ij}TE7@f?*UJ?@W)*n*Va
zn(P?Hx}z`aB2oSuNf59vQSo{?=sn2qe*imkKI->mS2V9Je@GQKl^_eaPm%fzRP(yQ
zj6L+C?MwtRmA=^lU5`pf%{;UFO+pRy@T3LIG86=nn2rvTQyXE<5LW(qM5UTw*VSFH
zEfFD*u<^#KB6c9&GfE?+6U$g0<L@GKlSLzo3O`n4MsEqIAB3CFM=^wb^&dt=`+&_%
z8V-xObg}s!_i;GgHQr{maak`!GU|;k5Fo5_NFFF-j|+bjE2B!~nz31g4(cP1K^(%^
z{^sl~rI$k>gKIui^YKtwsRO$saOA1-&nML0@dLBqMS|>W<BqZ`O~DVMi2ND>#h&JS
zr{$1soy@)y(3CeHeT>=bQ#T<QxJR_v4jA?nKjQMS;1)c@RHr}OF>gU8f>N=~7;p&4
zH0yGKR65|&m!C`*ACL&zIPb*W!(wm)RGnohpT+tMya>K`X-==_M!VtxSzKQx{3Vtl
z#kd+wr<$&#2r@1UnZc%gT_B6T6g<iPm8NTG4_d(?5c<AVPm#K0TF>!TOdalLQe85m
zc_y4G$WWSDj;_F;Rz|7ij9J`0eEk*ya|TUr;Sk9d8h2)1DB=lH=!X`*5e{@x8KLcf
z-yyprkbF4<$UiHxO)~p`$R|FKgJJFKy9Lhdg2o(z)YL&K-<PA*4Lsi>a6qelHg%bY
zZ<~5|4VVasBN*-W)6~t@;p-J3NSHRY+fL?g2{?5Is7;FrQy<}aufIa*HzxnQMTh#$
z0<vB9lHuB=#Bb+I-@L;=5w|X8?7RahrA2$j?{Fn~tW}__InwPi<myt^&8_lZ7IVzL
zsFiUa0!Y~WJ#6PdR7u(RGmYmG4a4btCO%{gpmnS#P?vdd<zxlyQ~Exy)tz{Y0{e<N
z?}At7D#V?hslqsr|Kue3L4FimqYo@*qEsZA7Iq3NT9{nr$;r;>QnQD1RYwmgufbqa
zFl9OEyYMP)E^w{-P}R5DK(A|F=twQX%UDoY@aPznur$f%oEb%%_IM^y-_u<^`K}Fc
zEOX`j8V02eExp9z;(v5qD8OO&1q6SnMxzYqz9zRJWTuwROp}sqVsY=mHb58cpeUe+
zsR4bS70^sP*P{pLN02d3)axT<kQfx&t+7b^;<_PxZ-Y@@rWea;>rbT(*QA`eh&h9K
zFRS6J;RsF)LNZiug5<r##_2(bq^8fBVzX}wsS{PQIf7y;^c%1cUZNm<!Spl-y!u_h
z{#cJV#mwJH0@Gt_^<X++emdrJ@<X;aCU`^Fpe6D=JtDRquYe(yO{|-lKx){xRg5}x
z6L<IH>;;({qm=5Z{3&W=m!~wBfEqpg$!+R2ee+PE+XNXr2jr<_cxPDHniZ=9AN8IA
z8$dU-pLrj>9}Rx>RkA}eX^Pb#D~9EGH7IIUg)tiOsh+&gL;`RoMmHt`Ii%WyVOCHw
z8%7`Oot}~5meSA}&$v>{uTu5pW%0SB(x(@5e@)sOkFCA-g=GhlZLfl@(m68br<CoA
zlc{QmCw*3Y0FaN!Xgo{$Xhg0~&fp2axyjL`_ppSecH>Cof{OX}ru~VWh`p-dc+5|E
z56{h<KV)H?L)-!>f>#K%uy;1Inq$*X?z9S+w_qHv>yz$sfRPw|8pX?^fRR4HxN)>h
zLTWjdv16C8b+C<~S=b?JcEbsdhFUEhfyu=Ir)KvNB+4Ua(mW;5?P8z>0x|DGe)=L6
zTWC^QTSUAVUbD>l76JxLOT_7k6PlPEhSFghOyNuN<SvB0tA(u`Ovwi@oX%z#3l4Ge
znJ0Mr=aVMjI7#}bStdQre#$7JA*{4`{X`!jnWw<d4On1JV;I=on_46rn}(DVp_pc*
z2okUrh{AaadtYQd03&rnva<ZKPvx0C!j}>gdP~`P+`PxE#l1H}j4?@X450_2Q!Kk<
zY)9ksl6?2f>pkYO$j-5)<ECpwrDqa)#CrB`-Wi&>(&8R~XvXD&y{aIA9XL#RqWR=^
zT3s-ou;?Z8&L(*0J_>(=D=sH?HrDN+0JbeKVX55t3g_J3SaybG)3&j!*wc=V%7e=E
zLE*ZCC6oS#;VD8y=ci%O799yw_3cTqCGlL(JL&FM9|#UCjIdC?aYzr6C<g1$7!gO}
z^W^Lb`_{cjX<n(8crKd%c}FQl5X>?~67?@2oA`5IzIW`rY)cxCkkl7>8(AdGW+se9
z4EUBYe8n7Vo0moa<cu77<kB`&v&O~?S27TLxo_qj`0ap#S668~=iZE2L$ml#w%=sp
z6W%kz$W!dK@WlNYuMu<JW@XO^Nhb%*AQDn?<X)7aTY=KdOm(xoMo<VbIjOV0Ym!@e
zEXqk*YN?v!DLL{I8?Ovkl6fmKkpgLTk(cGN=(<@{^M~jl5|#F?X?A0SaA8MHb(tU5
z6vlV?JS4;&`($QFc7j3}$&vTcGtIv-MuBg+xE9Cd<Rm>JbRHcZLQ9UkdU1`1!>Pe!
zMEonE=w23;MeY#a_#KUz)PzE-Z9-mpRRe-p$uHBx$hNfFH6m^&P1Bi)5%N9uyOlli
znnZgVWp<Y;jU&oDFu{di$9Rt>jwRk%^C30ac$Y6#tvt<}0?}@HUn5DhDfZH1%lBDE
zBX>W(0ptjNr@q_8*VfyNdnU%(YXd36M9MWTwW!zr#1Xb+h2*6s^Af*1s3a0lctE{a
zhcuL<{Qelv<inU@I+bu^4K~2AA^ARqYm<Cl5Xmdx3*g|T=`srQA{CE^QiXy@-culb
z*oVvP%y6cWHaeqG-UeJ1SS}~&M;R&<EA+y+D^+gYmF-in+h)B<3zEw>t~mLrjW{zt
z6?4MVoteog++M<)O|JdQNI<a{7MILxZ&)2)Zq2t67h*;+U+ih-35CK~WE#fO%|?{o
zj3hCpu%kZs6luVnQ0`cktbci8;8W=vGt}|@wC^blv6+hryfh&aYcf5U(#L&|429d}
z_ku#!C|1lR1UM=<#hnC<f8M6A(iPrx91jY4t<qzzXxPc*(r_zd8$`~fomBAGaq}UQ
zTilVwA3RkhyEs}(J;FoQ$&s`3d^%ze)73)qg{VmSrj&Inbei+6cpF1+^!gxPlv_17
ze|XHWq9bv{Qp*)w7<W)8G+O3fySP55a41)MEb^;@qMs`nGkw^LKPppFq#V9j;?48;
zOw`_bSjFB=-~)WkBQ)#9J<nRZjqui*l-AQ7-s>&WT36;{hHq}XLk-G4)=+;G?iJ)g
zaC1EE-ls9WHif%Wj6)*W5>C+>#GbF#6GrJ6kD{AmswT?Vdimxc__-pYdAK1$=q|bW
zo4v|wsfeOp_D=-9gQr2OO%4-sfO&p80yPus`KU($73E0!RM}`ess<_D3~E)`!N;RO
z(C8Te7vz)Zh`#*AL9vcy-IDqn8LXcp=exX|Kyi%66_pU<(c6;%fl>JehSWsKoH>*e
zJ?$ISC!Py9Uz4UdDAb84HSXRlc7eT_rYokb-^mK-gs#m_iZ^OKsearo9xR_w?v~$Q
z@winS=dVlCpr4DWZ>jNOXKbmSv9D2vEi&p|4&KhX-sspod4S2HBfDs(vPT_TXCTrN
z<_@23t|4Z}`fq=fy~we)D&$0(=@i7n;$XK2^x}GzUt}d(X=JZztJ^9VM0J(r*sJY+
z8Tnzk+_0~>xW>p}{@$51{9}*zZc(!=Z)-R)V((Kzz6~5lHkgTjYfItkJ+959Db(j#
zQFf0cCIWuG40p#+cl@d)M$e)$b|9IQ@b!8hJ$}cex<fWirz&H`lhytnCYQ^@orKWy
zRU(K_?F(R>5r16^**YTuJVC@;V`@SyhZrwp{|#rm<viKdfwCfZ%FUW_rB`jW`bU2+
zru^%Z-dp&@cq=)Cf6Sx36hw$8&-tq`yKK#IB=0)K7v;)t%jj3eB*4{Ftm{%KP(R>x
zN}Rfli=S?;t#Y?3=r-K@=#Lk{@mh<Bc3kh#^ot5)7MyDFX?V6N(Z@jkXS|+*$j~VN
zP*zeTLo6RoQCE5Gq9-#-fG9+9gosBuRzv14dFo5Y7d9+kIi@cYC6oh*LVBjCoqIT6
zwc?^`<1$$b|0_nIst39~%}TIul)rA#C_+I#<ge3!arDQGbaFH@kqh^Iu(dD^+un0p
zN#18UW;a#Zp0*_7={(v-`4$}@vWHHaQvntEy7Q(DX6ac5$@%O6+D?gqFmL&vU6^mI
zo|orP;}0^vGA)voeir9-HJMJHVPWGoy=3UCv$T1AA(@x8IU~fvc0xz!*)3kgnRwSX
zjKcx8K@!Ttw`9n+b$HZ(877x|WE+!Yp{hlKhvQ+p%4M2V^g6G|J^cuS=Qe$N_>-Nr
zs!Ra6aH3<w$n#H2&Pzl7H%%qcRv87ApeT94RDg(N+>qRyvTA`5COV9{X~BkB1`>7F
zK|WK3%NJ$*DPGWfyimhzvDsZVH8C`Zt{F6abr5>skN9@rj&pB3{j{&_OOrJw$V!5l
zz9^)OwOE*zTxM8agY2geQ+8Do$q#~3uw^OZN#MBjMQ7DOXDr@!&bpKlQ=I8gJCDjG
z67ShNJ$a#+ZWq(r(HZyhMTqdMp3G+DK6kAO9Jyk)IJL~Uh;#22Z*oD8*W6G*KSuUO
z8>YRgK=a6SWFqrxhP9Bemc+zlE8M^?CI$rR)C>$zmEoLMPJd7HEV=#LJN#BF+VXt`
zhddY`eQz)UR<}&fR?%FWw{D5>K=O7}VjY50?JV^gj#J`l4R`h)v3ju>do%o!3ReYf
z#CZ$b{{ZNZD%Fc?`N#TycuWm2Kt|kVtZXuleo7rpds>vYC^fkW%;1L)E9R)d=#`T(
zz-X*AQ>eeYEZTetBzab$u5LaLO_C@A73$_6P*BK74fil>o71`i>Ua}@bz%yc=f=Hy
z+^>uxpa(2#5RUVkay>ek2^h3Ne1&B|FNu>}E|~o`O@#26dny%^P1?<bp5AJvfJa-a
zV-RpEvQQ_9gseIbcfmdS;kWfcZfIR2XVS;>$u9~=<n*a|U~(G$arG=ba-xS+?$?TR
z!(j+ATp&(0<g=;1+Sv+#W5G-f*4%JKvwJj(o6*XAbOjCK%R4>YFFdpgK^}U(s&%A~
zrA%m_=Z&q_BqA7D0BU?2k$IpfO3xZCw-K_>bva%R(>yG-Gj?-X4PR>SJ<@FMrkFQ4
ze+Oz@=Y!o>Wq$tMv6ITbcchvIiS47|fFBfcQIS8%vT$}tf;f@GkmJ;T<+blBa9*G5
z11p^1At6YL>cWN;9=Mj|eykO(nzR+~uSNtlT-dzaC^CQEPTPn-z<L+df3WN4=kq{8
zuGx)fFM`nQo}AU1SMDml6vme^STtAW1J8`~o*j)DRGY}15nuYKFj@@PkE|^-)NeVp
zeFAb^XeY&i7rny0{IKvd)ux(eR2TQ<Sf;C^Wn=%C?qZC8)$wKN<purN(vWvszf-^b
zgvm^t#J_ufeeKdD`vp|Yn#Q66rO$2#2bP{ARe34KL8bW)oMBtFK@(GVg$bzl$@K1}
z&9IO<+YKIpc<6J}idO!*zyUYwEq8h3>+5klfGD^YJH2{N-3uMln8R%3ypbJPz6M;x
zg&<@{NNhW!2LicGe;|&C4TRrP*BNI7{_6nFH_@=?gk(%8ZL<hQX{HGhGS)_Vpy=n@
z5Q$Yoi>7vdT{Vp=iV<1yh;2p)XkT&`6@v-0!0Z(3@TCMpzUKljh{SWTvvzhlKM3=5
z(D!UJVF+~tINH{|R<>FYh-F3_2${wk5=={1C~R?-9U3-uf_Tg`hBQ+@3MVkv;lRa?
zm}I#x^Z3in<!|^>qy-PLTN}E;YVSEr_2V20iPS^TKOr%mY<g&UWNnz&PH#!U_vefA
z)3LISZ3j<or_H2#CL7v6*jiT)r+Z5{zR|ugy)w`$YPWQRwyy@(Q!~b9)`!9k$cTBs
z0;6G3mFvaAXl1cKWAl1I3k;E5GInYToV4gBl}5Ja_I!Yy;zM=)9%ZuYgN)zafjc?e
z-;J6)*pr2egt<?+Cq>~W);i0vYVw|UC@lnw=FI9MYQMbyA8nXQ$*iqBP3Qg@EUF=f
zXDp{ao^Wf1(T<$dO?UG&V=&)lcwqBsU#h&aCtDCcII5uuuQXn3$=!HR%>b(-vS!yR
ze<z4UihQpvrab0te3{2^RdbUoJ7ZyalNT?BM0>4^fAhJ^N1R}HELykzYheek%DBh9
zE0Fin|K3MXvy0j2wtJ+hIX;^lQJ5*ZQiW!53J=<3L@v3Mui*zZi&L=k+AcYUQt%S1
z@&f!KXF_1tienFu(K5|B=1DNTpHr_N_NNUBsojbpN6xbWSiNoF_7c(Zx=dkn4ey)G
z?1x@{V&GZ+ri?F6Gu}LpbXU+D&oTS?QY-vRj_yojTK4A(4d6^vzbRJzXvU~DTVg%6
zd$A5T;8~Y_sPPPGVDhq@@I9v1@N$I-`#yTV>t5MK@bV0CM;Skku#1|MeWG3csj0PP
zxZ(WO1Xs{=%&H5CiSJ@b9y1H!A>X?)dYf@1P`nyFGn1VjQJdtI{G<5CmmJ%lw?E8+
zdJ-?XDz<XQ<yIxX=%LZiGj=&4FUXMz^L#>{D@GG9z>3Jb`Me0687PqGmc+%GqWd5f
zUS*ubWLO@l_&KS}qfwQnuItVFQlU@n0JSuJ%AgQ{)LcZupK#7uyCGv<)<myA2Y6Zz
z-k&7%7Dz$Dwbufh{GD-vCu1%RE@CHjf+s#b@^&9V=h475ULyR<S31rFgK@?MD6Hr!
zDDHOyC$3d8paff(-Qum%m`jNe=4l+d%cbxoGtT+omZP*+7r+qh^FGQ8V_3U9laR`p
z5XWc0>Q!Yp*%7)E<ayx2{j22KRTh)gcHM)2^45w2VI-Z5w@b43NekjU@<RMKT!~<M
z*LM-XMe<2a?7n4Ja3x_h0jRu$-g~R*MCP6mvRXoL3rjJ6WHd7^7I_n&gSvyP87e~r
zc?jJE&zX`!Gr~hm<u~NriF?-M`4535fyX2Qet{vYj^g6e(-h{f8Tw={;-v&KosYK5
zxfsh=+$)2+&@s6U8Z2^p;)zZ(6Ny_fDy`sM)-&-t{Q?4XbX&MV+I>#NU4u$e4$`tX
zRZ#t~r+}5F8EgIoyLf#vSY=Fka}D4{-$SU(MS7rB5sJM`qdYJ+AIyy%uRX*)ZOWE_
zKw@<W>w~zBx|;iv9oYD^*x$Rb!>U1*E)rsq%ULB6&6)U0alRVAV-6n${-MG6r)sfb
zpfFqk@*lVLnm7OOEX<TBsI$;3>?}?wQ|$4^qtP=8McAXN+b}Ge<h6inmSDq`y~@v@
z2t@U{P*+w!Uvy^rlPvDr>-V#zP`cH;(%u&JZgx=xM}iycOzMjBk~i2I?bICDvwk0j
z7$;GBIVZh&U((DRyLwFF8e|=x52ay~M8zIDa9G?oakj@>pvgF2z}7jigIs9<9g)w|
zPckS%i*3kC-`u9Xa3k<X<64zlCx+?ST^0BZvhFod*JOCe^}l#yfEK2z?UM2D-9Bpy
zyz1eC`p~wopz+RAb6Nl{LrD4oIU`2x+lr&TGeB``X`0;?N^onIUGQIECQ#So^Mqg9
zVE<rtU510SiAE;v_yJ;U>TRGTwTs?6weSIZ9@0k;$OStiwqo=M3}}}P+dwn{pEUeH
ztkdu|4<9KLtp#QT0B|%fy$O7K7KJn#%nsrrJFNG60`=Z3ST0Wl>aY;#q7iI9)o14A
zmEJC}U}zxpsvaZdnP{l8!r%B!O9GYHbObu^YE&#|#f+VJ)vz$J9ZM{oc&9m<PBqV>
z3@ng3SX7Y=0>i{ed;?O|zkp&im-#`P9rH6<9q!WaI_te0(wetpP}xP&S9nl7n%5@k
zrCtfCkuOJQ-1bRf+NL*(*cfc_j=pH2Ht9!Lf0KLh1b0Wycjq<0X$M?mCnchvF(930
zXw*DBjoUh~axX3nC#(leF}ahbQx_+mFA^5>rRp`?f%AynFo~Fk_RTLAIPFXNPu_!-
z(~E}G3T@G>&oM(i%9%i84k8e&iuEt9KgK4SV0oJ8tc(ogCavtu9mL*Q&X`&kjl}GR
zU!TF|PeS!lLQ!8gX-qsk(l*F%pmT#i$Ucn})%k(WR?zD$86LBFz|SVp%@^@y7Ez%d
zh)L6Hw2}uHt3vm2kDNgLaPL@T@|80-teDwiI-`4+FV0J~OD8Zl-xANd!iO&3LATl!
zb!lr!DUIBri+H#Rs{(a3x9hKxCi@H1U75Y2?ZL}k%ne>{7B0Cp$Mh4Ev@QGKb9KPg
zk(#GP9zLr^O(xEA0QSB`gM_r^I6AxY>nUn9&m7yknZ&|37#+e!94;(l2&G6947J&}
zGODpgv|~0m7c{FU1$0({(BmV}xzL#}gYCZ>B7Is76!r~tKF(U?Kqpw`Zv!jT>X{p_
z`Cd7-(0+VS!NkFTYhSyeNMJSKO9#h_QqCl${BWmst2NnVNA7NrnErpa_g9r``AORR
z$Jvz`UmQ~pS}B&by$_){><Z=&VmrwPX1U-8gR5jdyvqVe{CJOXf!&eH;jG>Zhcaaq
z91PCWNfp;J(SaeFo$^fcGStp}z6XtU2Q#&xTHHTQ@~R6ogfgx`Gt;~ba{I&a^@Zuv
z_5~lPLquLa++slL#qo^-nAc5EII65%YzBz_Q_iMA6@ks1&pgTn_G+yzzsrAG!Q3^z
zt)0Q$iL6`&z@~7w@>XHzO}TM5pec?GVcBoM96YUX2hfF{stt~1SI}OH3rxO>{3$0<
z+xCz*ARnOWL`K_A`zkO%)n{jQ+j*9JYuPK+pz8bdRY*x^Q+GTGYKC$7R*iwL)n{X$
z8uvvjQP2kL^cQ4kpEio?6!#keeX_J%LrPkA4T;gG3pPm7CTrdn{=MNrR5<Xsbr?`0
z*4@@!HYN7h@AGmCA>Bs-KFJL19K%9s1RBLUQ!5tx3?A#Yx1N@sXUV$m+80&<8#GX*
zve&Kr=r-<=UZKS5kzG7#TbG(xeJ!ECu)eh^Jes$`4cIb>ZY?P%UnCt}g}THdAP{KF
zqI3Ey3%fJveJ#E6ayH$DB#E`;_L(m`BuI)85X5881SCYow6mX28`zK$kMWia-WCwH
zaX=o$c?9AfZSW#TX)Mx>D|IS^SNoD$ta&(#DRzM+jPLCGmuoJh_ve$RJ8QaWS3kb;
zU=4ACA(36p!=~@}Yx7?n$R#iVG!2ABPau)&5`Pp=V%;f!Y2FO!y!jAkv6p4dJ0DsQ
z)$?qx%Ibaoowq;Ui04NRldPOArS~YAkAE=F#3yR`&ZBLu|HcZjYf4`&hT%<kFOxxv
zRy$vv>?<gc^=yJ8fI<e;d@$0~LDGF#<z2b5S8KUM)3CO%z_h~UX{bc&j4W-I_nenm
z&UYK;LT<)|ly=q>j)LO132$~MlA~P|Oq6jXiD`!CMOX6612-{0o?UeJ$eKCdbL9*d
zhkcCN@ZxLPz|jj2-bIKT_F1|h_3vo+o-p<zX+GzF<F44#H~;y_sr(?Eq_c@*46uf5
zc!vd%R`DU4=06>!iz*ho4frQfD#-N&lB6^TLEc!|0tZ|UROy|+A7+^NkTGCFY{za`
zyJZ3#@EFz?40kTpI&hK%m6}Y<8jSWVO71wdFFkE;BdqH7Ts5+5@)led%&W|-+h>1W
z<Br^(02$(x&fXag!P;X{SCFO;+A>GJK3$}_=DDqD)q?MgS4w=|z{=Asgl3P|EPVH4
zT^>00Yui}@XashBj^$7;`Ily1xfBLSN<yeL_eV;oJzM?EwbW~SaI3YEi_}Yc%!=s1
z`tt>+XD*R`Mm7os7lO>}=0?xH4>DQPMlLdr{%{Ve<oxI1%mfWzKp;DP$j)0SHl9<d
zR|xd5Lbex|=ck9xzE;d=-iDv!3iLP*IsLHR@A=sXJba}+=KE{PS^cw4VILg#&Qjw`
zV0>+4qMl(J#p{QG#))y}?an>dZ;YX5H|b^sFj7&pPPAskG|c%@$5RupX9PxxQRmF$
z$yc#x|I;3Owp1iLV(MS^%~AB4WKUC|vwJE!+!HZhuGom^W}crG7Yl)%dySM)J;Vp|
z%Xo-Mb4`<c(wzwAnM-m$J6r^(#sl;5r_VH&US)7<CYv|CKG2k{za=u|dn((=A;TDT
zES1=NeTlp879^sPU3oE~k~AK9_X&-DIYjQyaNZGWF^RuaFm=IRaZ-jRAU13_g<)J&
z3#);kL$!UaeY}$HPF<jO+Ly6%1Tp|1w)hoja<I9G1Lgqf^7vAh-I^~p;3)ZWGEWiU
z$*~bMnQL>MG)4TG6B~hqKj@|DWM<y!-tc1bA_iL%4ooBV3yliAS7`&2;jwhK&*{*>
zwL>>O9CcGxscnCgKe5l4%%H-pkn_~j%)8h`FNvyfwFU{)@f&(L@t0GkyU~@q-yn2|
zOgRoKqz2EUk>%ne=zu>1s*soSOtI!=;s##Bl6J(iWD<69d^DE_gX(N&o5ZUWd=z5m
zjbH;h@g72Fohm(|L$1W|9=j2mUY$!{AB;pudSne2>Q_bPazS+j7MR>|AU|ou#Z=W6
zBCjOjrxYt+=(177ijQ@;YC!rZ%2Clyjj5jfiUSL$#hTfgloEE8Ekl?hf>xM+^$t!E
zQ;Cx$cjlPI2sFCEK&TX|KEo47wr6+YGjO5I`W0aJaTPBY20Y#pZj(-)d3ig>K7M{A
zJ6Czz#xg{F6hfp4mzQ<#Gb-6r&f<hWbOS-tnk<bjPsyQLUvL+<8E)&Rc3n)!GQ}sF
zZ`FU){Qo9J1;R~A2wsq?bcyNr-QuDdJ~mjlSdri<Qdo==C7b|uEarZR54gQwAmX+3
zH+WNrqx|JvtF2JgvZ~LlWW?~awhEg>$q1)-!u${J5%p2xKfQ^cvEqy_k<Q2T5<8JL
z!eXaN^5Um>yciqs8#(1T(24ErTj}rM9<~>Co9G+xLgwL28@M(mGs0>sr%+4^eviV4
zNW*zjmGNi3c&5a2EKJK!V{-(h5$>!h(kuSg1h&%G;0YV=b=BlPI@`!F`!4W|Mn1{>
zLELv??oC+rsRD9Fk|H7jMe(N><mlmdMO3<WFYjk~)eVU*>bskAHYzvMjmTBqA7Ryg
zZJhp=JY?ryr9bUzlD#uQK3tmYg+)(hPjMw6OBT5c%p>$%qALf$v114#8y>RAus3^@
zkgrNHwcO&rHGRt`J(QN|`;H2x0VmLcd^-6UCon`yku*Yg$lf!czT|*S$d5cWZKBQl
z&b}r6Se_9(xV4*;?72(@yZzA4_0kZF!y%bD=a`Ra0u^j1-X|aVRau`u#`KvD$tdPV
zx?nU~zbV9cnTt1<Z#|R#pdi7>O66#?`w6+hHj8VEqFb@hU?$iIIt$|qh?$u+ZC+0(
zMNOXDQLOHF-90BQiiUri3RX)yM|;PUkomDi*nD_YNnEj{86Wu%tbm*kd&a7YWLoF|
z28I+3C^N>SroV$dcY6~gpq|anqPA{6W}9HSeW`Ki{`Q0WuWHpM1IZdy=CA%Vk&+B>
z5}J1VDL$C?(t&vE!H{LCh{8Z+@m7t&{Z~pIzw*t4pj*h<u$q!B_7@IA${D=s`{cIN
zcr-H@*K}L_de`a)GV|^r4S5U*4=H7k&5buh$&^m}q7~(;LzS?IkM0fn<`v6p@7a9o
z>BHULZqVk1fJU^lA<DO^FmcsWM+nbEbn<QWU%usU8e6%PDo8$6Rg7u>C%7IGjD91U
z5<jP|irHXa*<S2RTYWBj-bSUyqYBhm=O9Hg#K}hwdjiM2=y#gW)V<<j1JKS(#$Z<H
zTwW|%UvxPu?WU=&2m1S=+~(%~VTkcDi`d3(Q?V#Wa%F0KNC3-A2L_Lf3@!CPG?{-@
z*#9p3_qz$!uy=Cu3xAPyroiw47a*qwWHx0a!b_CfMuFV$i?)H2yeQD&+JXA=WkUy-
zK*vy|;bTpVJcFR6D&{3{mf5q8K^jNE)3O6s;hvMXw6y@N-9%LYR;#l9VSsj8fAfg~
z%>)P*>jgQW-}+rv2i1M~b08tu$SHJYA^`e!bT6eGk=I9ZqVkCr64(jqhhry_;Z7k8
z%&M!vZ9<XSdKI?|Wh4Wsc@2Z#@*ggbe|3`nDr95FOA+z=%LG*%1jDZbPQB8oBRLAS
z`#>ocNtW#j$+ke5Cgwd=!84d65Ok9EO@lL%Rdx9D_vv7n8iVhNfQzTz%;xPng_;8g
za;w2`4D9(6<{dq7Otg<w6XyC<q(bs02FTV`HSlL!LW_Vvw;uoy5;zHW&sVih0$-X*
znnGxN%G^N5fA`*8hS~{KQQy!1>GAnVC{F?1l#tln3>(l*E&_2uqX1NH2MFFu>c~nj
z@HVq{I!X-u{Bbk}@Y>s%9?hE&NEK+|O_X6uuOT`vLa1Aym0mpc{H1g_R%(-ABRkk$
zY9mn{QgLE2+Y7%g7J-{w6AF2ia^<!Dn9%=?f#l!5YXA7qPeK~32NHdq+2H&C-Forg
zd?@9>3aSqd4>Uh1{z{VZm)H5v!!ku38j$K&i!S}KYx_?x&cFY>cMN(kWB2ZU|5Zlc
zf4%{W6kx!uP;|Y!`Ok{me{;ouy-7R5pl`5SO0DJfA20FGm;1M$4+((k|42?S=f%Ij
zzP}EhzyF*6?>~O)9N5M)Qi*)vdZ|_Sy?gP$8Hs<pgTLQ*=8OEG#4`pWLZTQ;EylF*
zra+XC{VUHcWqsoJs{hMhno>o+2y9{yoY<bnNdL5N2w!bE6jWlL>wD$T2DtwRgU)YF
z!K8)Osu-wtOLITif~yCb?ov<-tInm;nE(3|{vUnP`vneoGd+;&pp-f%7r(wV+92*%
z{n5X^#eaF_rI^^kkY;-5sr8RQ#s752|I_Cw3*=x(FO(+T<o#8l`#)aSzqc6v&+qVJ
z1vI29>@#2d?{2`3DKw;~TXa7C@7|gxybxLV;n>@+|M3!kIjX&4Aunk8rk2Z})|&r3
zgZ<YV<Cm-gX7uDS*Gm=uyBYoeKK}pv`2XJ@{dGxN>Mxo>4&j{yvtX{7yrlKpzlW)@
zXt4Q(Oa2@NOeR(+M#c_gy2@MTR<k4g0BVS50?KG7x{F_<P4I*dZ=`c#tW><lvhas^
z&1o&~UydlhC&1gR%lB5|5mRanXv%;32{LRzL|2z(Qy{5B);wfY6bJ2$UC5OP*p++<
zw%Kp(KvAz0s=5I&vVElL6=gkdh=u~V@8%iKP)H^)JidASj^X!jUbMvrDE1ITkfsXl
zxJN)!@smZK)kHM_N}VP164nemK8~xIh}o?>S`&r1{i~2`(Y6PibxMKy<Z<v{9HO_@
zKb>QF&H-t_X<4q3PnJj?_Jwbi?JUg_SqL6oeXLc>i(mM|#k61T*HclAuT%^P>}31Z
z$uP*D8Z~tiwYXv3%bWoA{{v0IV;sroMyQVy{F(ig=Oaj;4P|h=L)Ty<Pxf0SSAZs(
z)1m~Dpy<AT%u==u%t3+7VFL$SS~(r}Bh@Tf6>x2HgUPbgeZa`Ig!oYdxLCVj)oT%W
z`i$b%_19}L6TqbpRkue#CD;Mv;uffht`_eC4?+Mk1KR8h1I^8EZnV$+s(_tuBnoct
zDp2MW0erzE2*(vR7cl+@j5<ZwZ;jt)kk_=Q<_W;si~!51nIe-{aVVpBWyht)vJb!w
z<p9C$s<*+JDCkDijY<$_WT+jscAN?zWbNSMkh6q@$M(g23a*2iiZ1u=i;=T3eyx(T
zX+)qwev#d4f9m)Y$sOP_C!I;~1g<T6cq4R_02ZfoXm3you5RoD(s;agTYeHGjR*58
z#!q4j<Ccy}sQhdR?jPku^!)-a31)h~35Apn05b19FS3}@KeG$f=@1{(K(DjG6-251
zmQKoUMg0|2!yC#MPJXVRG?H{eHr7`7iT8HG<S+;iz9Sf}%{u{(MXtaLLuaPY4~Y7<
zQ2*=$ciq7}-5BUc%JN_DlcN}cEM;@D2*vgxe{(c*asrSs*#jZIu;Y}}tump8E#Ofs
zfQa{G@7=)RsR(iafflPMct8WoPp&1OQVhL76_O2%s<k|z9<S-T9G<cP%BRJkS0uPx
z3<BSxw&YZUJ|n2Ld`^xg2|1_(f5%PPQY@a|R$L|D9nY!zvFacTLxU8y5uhR(wxF7(
zn*z0XJ}*`%-a^2^8=R%4HOo4P`$LT}*mZECJ&U4dQ6uoRN_0}lyx5W>Fo|0#)6XPb
z43HsWw)_GH?FyDP3?AO=Z#f}li8MmLhJv14zQ@^>qG8RV!lkC5TP+mEao@I-arM$R
zcs#4w;ajWX;PH5<>w|j8FPG>NqQ@1m4RH1=0ZqVn#pPa$uLDP+>|-chHZ>i6GKQWF
zeGq-Y>4MRBC2;Z-6zgFC?FA=X+d~j)^)Scry!05P*pR6OI)V?DiB2u>kyhbX8?IOm
zNrEu$saa)f5IRDHV&1d-CbL*~M_v4rHh|`5Q9Qlg{A~3WhOQd)x|_CVfS$q<Esru-
zKSO7yQ3l%pgO@l@9wRpfMx6-y(ghH>0sjq6Cko(NqIV&7R6g+0DB{<jzyDYUI<Q^(
z?HSCy_<mf3DO9`<Y^FIban~kp4(^`^DkkxnBh&1+Z?Ye&ypjlPoq7f0I9!4%T5z-!
z_)-McHx1RNfPq@b+vE{=E5{&R>wCU>pGhdH{zLe!jqDNCd%zrmV=7a%#xu$JZvAGh
zBzcKfxJaAUCD(!c`RaPY>ic`Y+N4Ptv-Qs5*NfdJ@W8}fEY=nznr7-@egCapF1D8G
zc#04;4+V&<0R31#K;;(7k^v(wM-|BJu>U}%Q(SjE$EVQt-j;}L0yvl*px)(b?a`Vr
zuw(_+M>y86sw%%2MjCsPv3%GO2NSXmckj?Qyypv9znvm1`%FWwKX-kyQqU@55DCja
zMgAKn!mpA@HR~X44S%`i4LEx0(&@9}*N@&<vaJF&q;RGKaH-6b%TO#5BzFfF?dEaW
z=f5sH)=~9PHt}`Z*x%Q$P>J_P&~6i0ciW6yXl-ji6r>40#q*kk`g19tL!>AO4vS&W
z`5%EL$QASE;xW(k$3F)WVfjmKtY^f=Kz>N9kr>HTTwrg#9;rB56gAE>Ap@6u48R&%
zR4^zerxbLnE*Qa14nNl^C>dcwF<ecTjdd~sJhh~KAF{&(!=u%Js-PNLrE?NgDS7kj
z@pr-6CO?K?p%F&k18t)lO`<5&i{A(FPf2LL<#&S`>;kQ1<lcWtXX4mdhSHgYX=Qf6
z1&hgijBHJhymg`U)q(Y5@9P%QGa@R`xeuzKM-U!CMJn`C67bQokXGM?=q0Kf$@2%m
zeNBTx7X>Zs=Wvv4scG_@yeEhKTy^jfL`Q7kpCzd^HR(?5EIVtz4@~0(lfcc|Os#jH
zOdqK4a!N_IB@TKSkJVXzM6F+*e^mS6aBtl>cQj*LEJWiQ(0QEAdvJ>IJZ^!DVp6|^
zz}Is?9AHo~s)whKOhL#}0S~0`VX(h9r)7~ztL|y_CTN{w6o36Qx_>^rnN;4S5KQ)T
z_O~xn3Dm~ZE)+AZ_UzH4e2EOZNoBTz)@7G9d#?9Lj*SSctN$Z639~g9DEg9Ia_#MR
z^XPYkQlnU>#uM4zvJHan*p%?kH+Q<AJ<JNFWb$;Yz#UZq`C-)+ie&Q387$v4xlR;%
zN^Aw)nNB9<B{ovL>D57@j?B{iZANq<B%C@t?R+zWh;$W4M1B(9axBM`Ax>-Dc{=Ku
zKtL|*igWH@aLqpxi&oAB3DlFeF5WRvImnj;v%?smm`*kfJh@kD*~0mB`07=WkawMU
zBRgOwmorxDIS2BE=Rq>cPQIOfRF~ZwMjPq)k{^)mw)m7Ju7kZ+`V1+ffym2M(WQ^Y
zq#3PCyTE41)n)4W8RDQ)7ZnJ08n5QNEmf8LD^}I-349!=feWZ3YGNGBf7YcZH@neA
z_-{>WfJxFQ^CbKp*9kCbegIso3$q7+M3K|a<26rz63|Gx1cjc=>c2QySU90OD5I<$
zJ2;`t*d^mkbwEZa3JRt7+-Gpa0aw@1*Z7YrJ$E^nJ+zWS-(#iv2#i38dVk&Oi#=eD
zWN<d?Ti5h{Nz^u>1+jQ~lZbX~@TS1NgK^>@R!p~i+DH4K)rY!5Z)4i$D&R1*2G?Cz
z0W0%ouy^BuB1c#am|rjUKL#5G|LX|TDJVS!$^r$<QnKm-YcxnV54<KvU9G$Efr1&o
z2~duSVFz;)P*OIy=iv?a>wfHh(?|k|0Vj#|(M0|Wx@YeT5RnnGNHw<f@g7o{5Mo3?
zr5c1isV*Lznr1TF2T}^rd5smJ8F9O7C(J<_Bu+h>txU!x{s=<@U=yGI9_=&Cwi%v)
zuD39S=oR4_c_ns|&V;Fq&BSoqeOA(+eUuS{fLM6J`6tkX(czzXn!?X@G<wIQy&=Oe
zu#%y~wVcP=ugRC-s&2y2AR-XzKT%Ig{v*Lj`El!BJ(QK?vH1acF1*$8btuAOtE;h0
zMFPbR`B_<2FM+5Nonw^sYe^iC2ndf_v+4t%_;z@7SZgSq6~wB2tA?iLgY8^Q<{-Ev
z(+KeGWE~8T+E6G?mOIUYJIdAE+u`%Tb6IR7zFq^0)#lX*+vxRvpl1@VR@xRP0w(j1
zy0X?)fRfTI>3Nre3mSe53*cpPvr#o;)m6%7V@zFvjLq%dN_R+eo~%s=3gJ9rQN(j&
zj{r#I{r)1@!`p%4Z#FtK5)NBZsXP3XHxnq1{b0>{$Z>MwtLq9;L@q|_rkXKAIG7}`
zD}uJH0gKUBP)}rW0l70{OjX+_!yXi9kKmc(Ds!{57edR=T*JVnrC7-BQlF;8&O9Z4
zO7~O~@0_ia+%w;&?JW2YBediQ8mKf7WN>DMC=*8&T<TMDp)6EL{HNn1uiDdI6B1sQ
z-B$({CY;yvmbWYRI!6b#Rrx+?sq3tRFd5dp&z|8C&7$;+W0qUp#T0|ehXRM^mf2%_
zZ*7F47r?%`FZI1!o7pVSRA(%K)dlyIu2T+&%q<vC#afDV+1hVVVxP+jkvdw2)ilS!
z!wty({dH=$UJ6rZ*l9A-5K5o7@QiAaOH5PuoY+pbnjI))_;ycma=+!;y}POtbX$`k
zPwOfMvg4U};S)CaOKi&XJ?ZNevdL6P82K)Ko3XO=$Mf$-+wq@iXvyE*@hJf!qd-7f
zX?>Bo()*_G%^Q)9`_hNfZCw?;MgZ0HVJJ{~C+%c9=4x-~VwLp++erOwvHeokDfo5#
zu=;idSQ6F=uhNoNRoiq%mAvKbyX&&XktF6kEsuD^>aP>f3<&n|eI)hSm!<2e22ld)
zfh|AH0&jqgxjnAgU$v$FOg#ZNyJY7j5Lkx`o$bqoWD|N0#zLRd1dTw-!@N4a`RFP3
zyDY;M`Jstm>u^Wyoq<pa_h!f-=a^9c;?^j_qri{f+q^osCY(s=K_!2S<+#qRC*OiY
z%5)4PwvXujM#$=g18Rs9uDcvmetS<R>gTb`vClAz*w}qYFXG)zrmIRYWQ{)s$hi;0
zs90$8xoNu<6M3&=92uvY>y-mp^C&6Nmg}>z4FZ@ZQgaEFgy}@n0Mb4>CAwDXXG2Z|
z_0Vw)>y4OdBRM5*xq$d_Mh0J#Gy3)+IgEBu)~wkN-)0J6%7qda(-Y2W1i{d~(u5;Z
z(Tg0{_fhuDQ;y(WzgOG>LAVn6&Y}(J@f7RHbO*?EyMLbAItItRV@H7b89r-2`jBN_
znl0r^fSohIePdcBb_*1~S|4i5^@xC|)X+pi<?t-^x03AVlw{>BpFW1!Jd-G;WT9aI
zL<`|2m&_zrVgO2y$nYz|wB3HLgw5)Jk-cbT^jOhkH<hD2%wgL(dA<?UA{x5W$2Hn-
zd<ma9z~byvj#r+AWDdQ?(2<KZT}p!WkW3JZsz=nE>!(Oh1Yixpjfo1W_o;cUG-}1-
zJFKafWZ(EM%1|oQa!nqB-93L2N8Ky?TB<{sW_cSS7*cvE`eYZ$Yp`3ew~ql1y2vxe
zM#Ghvdj=tk&G#>W;j?~CwYkqY`8a+d)!ch0jp>}6r5NNRHbg~Ap}&qHPM<Z!Ck`-t
ztPQ_}U{txH9D4XEGg)>yZDN`rm0Mc7lOdXjbgS!rd}-U*QYCNQVmi{m(Qj$!__jC7
zK!|9LPDVFD3U)tHDoT=tkIj%N@T>*yO5o2Y&`CqYJd5eMPTw~hK3Wbb+l%Vy9mCI*
zp5i!SaZxH=du;=&mFL{^)Dx7w2uN5y^xyJ5jnPgm8kpdB55CgvGG(MfV@~!mX+fhE
z6*sHYN>)EiVs)ciXq0K0+Md;(hN6(!dj57=O`3P&qjs(#?F`po2@<m(;8<?`gTtO{
zma(1i2a0Xg({4C3?6xJUTTnag#IpJ&i9N!eT!ui6-C6pA-HTd#%U63!2TP;7+YFpY
zlZM!&P4T{ix3f_s1ZJlfI4|$S3?AuJ&roRxYzd!ES_Cc90te!Vmsi3fsROSs4LlqW
zp4DybmO5FSfkv+{7{pO<{xDOqn^#Q;#EDGpy5w+da$6aaGJMke7M-1#UEzN3%FjLw
zcst}RNy@Rs>hAGF=bv`t#PN<K8+Dn)U89}co(R{id+$}*8-(P-ylmB|>=?lmUlvML
zaEd=DOn?}}yL`QrcW28}HUEv7P~P5q&C&<yeRDr(1h0g(u!k5%Y|^{-3mN7Qm^j?$
z){(8Q$@WXloQA6g!dzwSW>xR8DCC;`!2(F(HNRt%0GBij_Wr~TM|)f&x#ykBn$HYx
zbF6vCLX&#tWzsftnqm}d*}#uSY>KqEQ_rZe?&Zh^v7qF5impBiJM{=&AfZ9&)YW~N
z96Ki)W<LfI)zj#OG{Y3eV*DiAUt5jMNy;Oz7dqu{l%*!#em3x^O-`N|8_dXmUHf61
zI~A3`$+Bd`m3FoZdvn9MN6o^?3G*rRpEbLyITc?_v$g6)X}<ohedAz7M7{_c>`6Wc
zqpPGg__`AOs98=PANT0p*`z%NFu)xNhU5qicbJ4W`d2nMTU{vMa@fWVMX<^120ijr
zB(qiMc9QDe;q1R1WiHnLYG7|(W=wtZwYAewAUMIj>QWyQ?x7;$4pya*+;1RN=pvx2
z47e(Xz!9Hw_<?IXE4wh}DhChB^Z4>@^j0|idfFWvo9hA|*&FVjE6HVsw^z~q4@v=g
zkMmh<28nAh*3w-gJ(bhUpJAaYs2(Zw4E^=s3a<iLOe)+{Q?o{)JFU!mDAvoF>@bb~
zwCrc>hWX)hrR$qe5?;mxi>&3{tG-zu3Eo(()}8#6-78yt-0+BTvY5w5tu^u?MuO{+
z@?djtL`QE-Z^He3rp3gi)C=E3iVugJ`w>eY+9R;`E4&G9L>H+1^S8IDN;*cxM(Ib(
z46-tVnJwq8gc5FT+a&BNVd{!S_cM60aLemLB`(iMO8WN1w~B4;dqypTn;+;x*kRdO
zrY+eTq=49^OH&3e-trFxt8E<QtZ^3lmjj<rzU8?R_WE4ZDJ%H<+SM-_W}_(amY926
zoh=%{wYKKfj~Oh&9<;j_C@^H$4ZIJhh#*Q;^39Feu@`($tr%CKH|8C2luG|@h&1TA
zF1yvyjz3G70-3I2xAQHwDb7^q!7UduaUNvEjZ~#*P|$J1;mSHdNhH@n^dIBw-3UB|
zE*Q;7b$YU8FnFHwU;tB!GLeH{`N5o|Gbr>egncPeHfTSL)sX|gXnE(NW6`qF+>c?F
zjRLHaZt}gfT`CumEgs6dX3Z_95yU113eyE}zOgd9?ejwYv8O}s-yz3yj2_coErH5M
zd>$G^xH6rKEVb+KP9{80$>A<JSmH_3G0`c4>KZz!X}YlbU}@jDZylF9lwqC5^c-Bb
z7`r-|uBQnN^tV|axY&r^Xs5tKa9nG?+BH+=N82oY_9O_)9|p_HxX`JW)kebQex!w6
z%Yvg7JZVIn7S5kP-PIWaHWKTDKJosaqXydwY6+y2g05##9`M_aOk`OHm4==wSMK+f
zD~h;4bZUvwe>XJPlY**-i($+7AHzYHpmLX~1okXH6IhsGgD;|={S;JZlBu@TnGZZO
z_PmBtz9VrlBhu3`LYP^Ci`anm243Th`|dpE36oa}&uL%(4p3L45zcCahcvZTpSa?a
zdkQkHo@cKKs1=R5n$N3ztoG#GeIjfug09ze6yxxtKc;rD{g!g-OcE-N^Qp<$s)D!X
zjx^myPb4hXDrM+e*Fa)%(qmTn11y4ROpc6h2Gnm5CAg*Di$cptqTSqu^n!S}Z!xRd
zebtFbhlXK=j))>rP3A|^m!6@|OjF}i?bWS&t<!m|+cLvsqA^nEm2dWaR+vDtP&ex+
zHEol_eZQI-cr>~7^;-*T5A=V1_HNL6{Y=o8iU_zbf?TM|74uxz=+*9n!M+Ow^-)aT
zRmE{a6RbU_#taYpu2oN;)6`UEo$Y4VysDJ2pJ_p>c<XYo$>7!s>sQ>C%$j_U#?H`I
z)cWsl-*tRhIKOG)N4yHDpVEPM+47Ga<s?mCh<UvXVeLKbC*g`n{w%+@{G#WRH)*7G
z34W8UPZ!m9Z=rVA(5ses0DzPTLlfGDSS!ZOc^sC2g#*E}|BM1w!}i-M!cXf*RY1V6
zvSkg+cJ;ceEBriv=TTy4a`iTRx{pfQ?7rUB8!6A|d#IO_pO0fSkU{m)Oy(mz(UUCc
z?@~!Z)72JNiV5;liAUU{bxdpoZUosai8Q8?3o}KqzIGupxslfdw5;|WA}>3<3dWm4
zuT5B^syZ^9-L2#U@-Jqupza)-FI|f6O=9zkfprfcZ(Q+QF=ry-J_Y49j1{)ek-OLs
znM6mKxxaO4d=&Ypex%UkQ9iC}u;&IeHuJN|7Ds&s@%=4=C@=Ozqm?`MF3I@usL9l+
zak|O}71fs5?R`Aek=XyH=-uUxcpSx{Xmqs2+s@)+vd=|3TpeeN-c6Lxsq4N7nEOG=
z)3tSNO)oJ`()*)|(UtH0*S!7@V_zK?RokvRz(@#4ch>;Y-3&-e2`DMuC8<cKbf<*G
zkiN8p<dBj|ODc_oA~n=dXN~Xs?Y+OV&)Mhv#g88_!>skJ`?;U%zQQ}Uxg#3SDp2(1
zrg3z(xm08qsJ+<beCR8yo=CS$4NRZR5sk1Kh1-eA)H+&@HX?QE*z1A^Vz(bianxsF
zT%CK*YN?<1rN-^x@7Ql-hLia+y8y?YP)YtT$8@H3d4pnf9Xq5nYGSZ&T<G2NQi^3C
z*a!`C1lRP7gN`QPr+LDL%V$JMuOoJ6(?&4o#?^&4r0Q(Mw<_c6NO@dw{2bDr*mZYt
z&s{cg>pZ^PHW4^04C2y_Ycl61c}yenJ7{X}irzQqPn{JUfc@5k=btr}FMJr`j?%c&
zGDgrLXt-aB@L3LDHeL&^6lxk8N3g|;)DPdb)F}HK$ZGd%o9rdQW7)b4&wQxg>!@%p
zmFV%J<v|NPC|rY=h`R%qK2sm0Y-jEBx<W@+C%c~T8X03=$q~C?=1sH8nCRouDDcjh
zFTPN!YuchX@!AsI7abLyBG<U89u*E)0k%<~UA8!7DRj6B-J<O}&ZxyRBbQOE9GDk8
zp3j{&Op#Sei3^7<V`;vneqJD|NZ82sW>RHRL1ePWn@1D<<53G_yHtJqtTN_&4KTH8
zzD=rUZL8A{7odBR(iRR$Z<3{vc|PeC4xyeGK}-U3ri8U_iA;5fYqN;2w1X#ncR>>;
zjLaI{<u<3oZLTOzC|^@FzU<10VY<Jy`<xN#Qtak<uc(mhWiKSxg_zVxIz*j%V{7~%
z+%{heE>o#hurm6J;(OddWMT&H!T1WRyc>p?4d5Fo^zJ_NK2pk2Ftaegt#vDtC8$Wy
zwWd7ZgKvwNyFxoBkEG7aGF;VGo8)QKFBV&4(=)1a7!C798YPVg&nFEk*ORo;?T7pd
z6D3V3URG|R_AOaU?}+XTZS}cw^WW3psz&D(#z3qA&CL`^g<OSJn|P@9t{!xVVuR#Y
z!VKS@M4OcO>NzZE%kQ@2d`Gu=RJgrlvOJ$Xm-tLplUl@tW+hAZ%^3rC1;Y%TW;|De
z)Q7ITt=Kk|WipZFJ&u-WJhzv5ek1en_wPz-)z26L66ec=Exk6si}P!K-<vkuBG?bJ
z{c^@Tu3oU8aZ7JH`i$Z8;_v;%<*zddcCJE3iS(Ts!s$KEoE*BvblRaTRoag!)~yTq
zc$7#=zu3jTP%L<h^guFe4vEs=I}=3iHM50?v@VL1?PNFJ%#-h4H9pR8J&7-~-}c?r
zZ9WOg3(9I1p$ITQsHoKm9cz^y{xS~qnDbQ1M>hJ1ELCpJj*lE{TqSltmbX~L2@&3m
z-Lp=_#FQ=7PyjteP&=;*<ggZ<$St%34p1(In&I(9;h`W*qWHtuNRl1AGB(iEf@5SR
zewaIR;I6tkwf7@eu$*KOkZ}JEj}L{o>z!eC;Ykl;ZMH(z={J>c%gBa^HfW!7zRLm2
zO@Z~4xP;kbVf^<<L!VEDY@(zPsNA9;CmR|SamPucA2tywxHqU~T^zlVG8BHJ!sD=0
zPA0?2pR68fFr^g<AaTo!Ll%@ak3*%0Iy>TE@Qm`PGnCI#;SiQQ(_lC=41df~zmZCK
z*UNc?JMsW7sbm6gr^H&C%JJ^NM0=9b5tgakBnx*X%s&oIxyM3FLYD{#p1kMdf8zB{
z;tt?o`eGi$vWionuXez9TYMglvwe?V5nd?ifb18u!vR=^p4n-(!kqI)7N|G+Bxvrm
zYWFuN{9e(75|k75GY+}MOHle-l$6OP^X@~e)0(fQ&kgaxa?5XC-2kw4Q$6KNYAn1!
zH1XABou{K9&e!CF8odrV=6T?fKtN=p(Lu{jLz3O7(O_y)+{wiNY`gLMIe8RYr<Xx1
zl8ad&!^$teuIMEoE<XDzO&ztGyoELE2u+wRML4s{JUFtfUJBDJzD+w|on)G)x~nlx
zyoU7B{jVcm=SM-k|CEqtEZwhT)Qo@=>5?3~<XQwtFQB=0xgzWb+GO0m5P>5g49T_)
z($&hZe|H!Kb+HLnl^1uu3`d}4tEkUMZOfjwD%_X_1t++A=lOJO&|xg0jYoBd_E8Dm
zPn%_$F2)eWq2zefxbW8FwHIF->lt`J!{y{fo<cba#G!awMbN`f0qHaa@Rl+R7%nXC
zM4es~NG&TI8d{?oe?6esYnX4B<vOr>0hyqicyrmuATRsYrJGe99pAy%p5^1__)eT{
zzCK(gj*hj9&(&wLbNTv*7{Ttws+`2@g;BH2RM(qNDu=D(fn(QVSajaWmrtf_s)gYe
zE}=X)OY*si2@R%|1`z;Zc5#G)pg*ejbiSYwsVfKV*Y2lGjH%kJFXJD^JO-r>MQTcJ
z0lt9tTMfGl>9zq}#^?cZez3%Fv<+>@hgAk^+8I`j8k;a4^dP?1J-9AbRr*udu@L)}
z7(exQpFZ|QWU8h4=H?51sh2$sM_iWe{7%QhX2d>?YeowuZj*jVSFkEPRqB$pJC>uo
z;+OC8_=2KGs3|3Ror5bKS2aXq%H~*)hCJ>`O~rO&6acNKebpIZI-}TQ-6y{+@8nwv
z4N4JB|Ct><+zyOk&bG?yZ#-UdeV-`rWKNJZoOL>wYDZj}@*GSyJwNTK*<F_1H|6<W
zp`IHPOm8CWM~NPcK@Uz7dK?;}9#~>p%rjt)$o$x0yCdi=j$nr4=Hcz}08i|jDvd(S
zI%*Y=4RMIReB*7NDD!D>x{<fEx7ieG=pB{+?iDh9e<o>~*OCS{_9N^z2sJima+e$L
zfj@m*4R6tIW!4T{Y&YKj%DiL0b$%46Vo}OicL<i_k+8=IrtJItl+%_Poe{%M5*zl5
zt*0lfFMBMyhcp3tpP3FFUzqy{HEoQ>oCXl4!i{8WJ0fMk)Q#1V1w{Pr4H`=@UVl!6
zJb43QEda`onoo-~C4Gf}QlWPFp+YtUq{l#&0W0OpUy6qk`mKgyQRBj6o(TRngKqzU
z^qk}^Qji3gHlG{qUl9jF!nN(vA037vG;@Oipgf7vB*qC-jUg)7F`ciAWI!2vJ$XMZ
zJlxs+ft=u=nz^2xK*0G|&c$PyN{>38caDid95&6)1=t~Y69R=+cym)Z)6%nW+ZO?S
zuj3ztDcI%lKY<9MPhxHpp}Cu_cz=`1*iL=n4N&jiHo&9{4)nS>_&z6Vkk~oR)Lv7W
z{*-VON8_LW*s}{J7pIAS6lzUiO+Ka`E%({3T%(o~EH{FAkT6YBF08D8B#1`>U~;Lt
zDL38xHv<Lt0D-i04#^GGo#Z7-b-{e~$pfcGpDn4w8u3fVmBqZYEt$)pf<MIs7)zhf
zOVFc=$s{LgSQ)3Ajn}tBy&9oIUGAcn&Kf_T&JH8j=A!&uFW>SHjv|h(i9JC;^oDZk
z(Vl<w{MbBAxgq^KiVwJF4g>2u*=j8mDzs;Y$cNrszupsuPjoU8nenWXoO!uyoy6`b
z$ujE5S20hm^pRag)T_r1#Rq+&Od|88Ov(zY8M48Hj9E~Gd=3+$QsFCMk2daLNVJ_o
z$D*yie2lj9Hj36Wy0k|;-GP7pTbfoZ1Nv1zVB+aFJz5?U`|O8TPGGbdFxKl(%2U{$
zcPLKajs}f6xo~w*q%ru~!?|=Wy>|b%%E4F$gR+IP#%g^dz2P-Y3(9`d4}R!^zB=^!
z?J8dSy#b?mfjYLL_O<Pzp!7$VC|Z3Lp1>ux3Fa1Id`yr6RT!;$P@O>`fmK=<tFd16
z-f#<cX#9{NHo6(-Mg8kR-H9S!gAv}-`_h0|P{vv8(Cdg-@e1O}aImEKt(aauU+&@o
zYVOiRF(aiKT%vZ>{Td&}m~uZCT|eD7_hpycP2qzLq5ko7eJ;z?hZ`?Ou=nZqF*b1n
zG&TmpI1fxmUbTtc{HuB-TXWJhwkLP$-GRu{aefg7sJN&$7PN9dZbf;aChTZ-$x~^p
zkX+sG(Q`QAY?#;c?BUgv=yn_BJEddxO>6vGeZ45V|ID1*xzXpc-TS*4&q2?X508Oq
zy>nZwo4<jA*xsVuv5U(r;pJO#wXPxXt6%R*3r7j1*HMg&5Y^<%<F;rzJQEIhu(!l(
zn&c+OkCSnS28G3h0!hv3(pf!NbqlL$a&zI<a=9v-au@WCKFJ}w?Gr>34UczCn)RST
zK<ekB*C?52^hCnly--3m^E{V$Uo$7fr@&FN%iZbauMV~3O3=}+1ORE^x{oH}H2M&o
z%L()~2Mhm43<Hgbs$+*T_a?{etBqFgCqmf8%u^-bD|)<VKpcOO#O!%v4vykefno8W
zEz6}WE62-JW*Kj@&8~__*}wFtC2M*NW>!6R-p+68&{@rNb2%}Rm(GaYI%Ymj*JFGr
z`Dx<j<XcJ_*9eu)DUt3Fz7RgO%G6iJ{xQxED?_c<A(Gb|#VSupA6z9~O02y4)2eIA
zTj(GO@bh5U$b`Dl7GkCAM8R}rDu!ab8kV`?3M!Fl*^xRK&l>4MY%Cqc8xH1y2~H~7
zV$2Yo{=TA(K%X6ZQh8S^K@s3s71eRxdC&yc5vpC{eg8^&SA_ZrK`AHf{g!-qP0@xY
zUc!z@&ZRvvR#i}8ZtO!b4v`LKCx`CX{E~3o(s3zU$e>x4)>zn<beH4ExNw)6*p!Cd
zG{MsElr_nb7UC^z+O`tq8|EI=32Jkw?tsmnzI_yuW8~=PGlv(wO1&lL&}pj7=C4gv
ztz47R$2szro~*k5SyYBz>FKR>5Bwmd{F$@=qG;N7=(N4;Np{5!cS2GY#_WJz{5#Tw
z!lN8&tgUZxIU!_JZiysR!u##f6S6H;%Kl=izsB&X1onYbY-ahu@41TNv=LFiC|YYB
z<&l(*8K-~XsKwiBGa}q|U_wS~%bnO*AB~MRJE#-2q&RA9vcjROyu|h{QO|y}`?q)P
z%I4o-Z`}u?WRgilxbW|RAnr#dGvGpwDMM0A%fqnwIkE)F5j08h0Zd6uYr7P=bszZ2
zyCSb&6K=yy&Z6!qa>g$kQ)r}UlNAIrt<_WI`TEV`3FIb6A!T_8h{WsaF7iv+SVD;w
z?GgfiH{=AX=dE&zIgB|=Ny#g5XSsFaZq&$m<KeG$W~v|Mm9Ez9-P%lp-ZK$H6+1Qx
zj@IgE>;agoOi@P@myLS3qv`Zr4LgX-K*#pV-x<HReP*RCe9+mI1ml<=#K*ro`Wnq|
z6I6ID;Yf3nDfy`u``&Kdja8O~3wv*sOt{}iQrK2yF(ly&f<wtSj``GxYS3KCcT7nU
z-=Bi$Aq<<CrMc<axY<!i(L3IlSSps%@g<~*XNHalLj%ft@Uwm8V-Wuf34n%FEz{>#
z;jpA&1Pdyfv#zx1kkIs3KnD<!EMer#iH+V=#LwvJYM1HmbyTw4f%*-Rqa8NaEArCk
zljmy;cE617uNbCXF`wGyeeO#XRQG>|F$%zr;sIqcAB~s8&#QH6%05-^w!!yJv+XFp
zS19McckuB5;+ttl5{EW*`oA#+<G&LA^>`<*)WWA&lbn-(y?E%2^pt+_$&}dScGiu^
zT-;|`5f7G!Q~MKMdd54EITu7qFzu+wAm?;f;yfGkjU3S{>TS>6DT+B>15Ri|qtCbL
zkDbPfH{6zMUBFRCE`<ND(TBpb>YAVu>nL|dJHMap3i`!G`5B=3ObY=0v#+o#PWXbg
zYnf-s_!pI$>#zY-mJ5wgAsxol_yc2|9$z$3TXISl!1Q$JO@{No6;m1Oe7>!qDz+sC
za7Wa1kUG2@hG0lOfTVm9xIZ<=6>&9mK3T;sG-VY{t#Z+8p?+7GZNuUy_awYKv?vSB
zGS`ANOt@@IY>AgjQg*+vEK*AVvFcJD=fU3|SoVb)EEgZO1Nd{Rm1xy`l25}M;%v}>
zb&C(Te+<+fxj$e$@>Lo;<m9?+SEl?@vyDir14UAt)Opna$+P&TTo3>-+o0(#2}q@|
zK%kRYb=mNL2N10Q@Yx~qJ&{Jcokc2<JuMocRWhtigt}^&;HCU%JC&9uxL0~SaD_#?
za`vpz6*AuE4si2wlfGWQzY}koBSUjGLq|*Y94_G>3XC`KELwS@NF+okScGR~l}f#D
z`k3nmV!dAl?xuc8iov;3qBca3Xpi?ke*5wJuXeMqlLXxYkvd=6L~p{7s=iObnHAVg
zM8z+2Ez-+mBC(+{9ooK+H+!2A;)!dt7!qm34%_UdQoKBWw`u@(bxasi5e5&_3Y&mV
zvU7Q=MO2kngy`s8|H^}vEkyV%wVp6YLKKA{xCtz23OhQDPCp0TlX>41N(nU=yD5Bp
zRK99TX7qY}BW<}PT1eR`r!2UH!8WWDHW9B`FdVnj92+Eh)>gw_Z@HS)9VwtSp-r|6
zak4G+$XaNXHT)4X4x8RSFWr}IGk9!0E6s`})Mm2;X5m?LhG!t+5x6g6BkgQLn*|kf
zVjtSsdZplm7$?pJK;}Ca!SD2ml&b0FK2x$MKdAd6N8S(l_;l)H&^I7rY?gOj>Y*C^
zQtHVI(Qod3^(5EFKb_O};f|k;$I~=pJYJGqN_DOPDJc|lM)q{LZ7GSRD@xo3>DG`a
zq5Q`$^KS`=H^hxzpj+T`Up4w{)MamdbMj3(e%GE^DfrIl?5f#ktD$X2S!7Xzr@$3y
z|Ka_S`bv?izumUR<%|<_Hp`4{-Gg^0jPR7{NK}(pZH~AR1jlT%0-gv#{xiy7Cg7mQ
z`wDFZ=8#b?C-)$SI@4m8w0YJgfw|9*KzU7kHKsK~eCx;sI`NC(KW#lXSdO%jbU$$b
zsp;l7Z2_gz?gCt?2CL=ASwQRYYTE)e)1b1OGIKkwEh3PwtWE^~df)wWFO}5#P_*Pn
zAu4Z-lTgaL(uA;g5!)N{kONBm*LF*cL`vx=_lsQLX~#bAug&z}FHXp$bb9wvsm(jY
zII~+)$Ido$>^Qx)rmG`3h9=yce<*32GJU)kt=7FMMeD7<XVLUPQBq$7m9?+r?t*wm
zH%T#R!@aj2={^&9^U-5yQViRq7$g+`8dZ_MzYCE@<D@CpRYSM0i?T#7rnzxUo6M07
z$wn$3zeyio<vtw_Fj)*j+Stho5wco6<rK$px3i!uujRywMPwUm4ZvQYQeSk`Gt<=!
z8S>A4T!en&Vd(mH<SBhOC|Au!t35U{CLTg64`E)m8Vxh;Qn_}unuRNH2CHBuAjS!A
zPTUm0a+qJ$YSkR4gKN2Z$ACCatD9F^Z`Y`&BHt>PV8kD8;n~=gp2)<^;2EVI<bq`q
z0-orLYqu@{6G{AB8=KWdxZy0ajB~{_IITVt+d;fJdcKHFJ-#A>CD|=J<qg#}CFh8w
zVR2B(IcnSIz)GhWg716?7-dbzfpZ;1GTG%n0)wzKG&ne4`q-(i+w=*%O_CZSkmP(x
zD3eZA?y`Lv;p|Y$FQ;Q3y=<Bp$g^)TKERZwaw*66<Hp0AI3eOHN$R(_KjyBe+T6Lr
z<H}Kiy3r-p=5kjdU%;`-n10#r?8j-=NxCiUgT-^^<Kd2o4++y)=#2Xhbf|Z|+$)5B
zcXP(|q#jJo+#y^KUh^R<B1$v4om@^N%&pR`b&C_HBKc*1d{3JYTjgLjZ<J5hq>a2Z
zOmuc`$_`v9c3Ub~mb#gmOJ!$%F9!w{sHKGV#KZAv!ZM8;Z+YDgRCA950ZyjkVuT~}
zk4FO`te{)U_Cbl4ddoL+|3X?BLxj>~<wj=~)Qj${%}b$>5sK}JTtWoo*I;X*{Ugf$
zfjC*A-v>bUQZoXgsXRPg@G|$@@^j4Z(UUqhp`H-zcFx5$;c;!P7vAR{e!BV4YfzO|
zq3!U%I!IZ=DO^L)b@9>*;wluWG_EvA!@d)<`ZTe1TBfN^WA{j^(o3u>Yab3~pL#1o
z5JT8US*%&d#P88a3ac!#7a*r&-x-=|Qep!Mo~;1n+Rt~-_VTy<1?+5HWYuclfTHme
z7^&1b1Ak%xek{_kIPKI;18!GN&GWAuOQushdgIeAxvyTOI{3OcpLlj{FlTVN5ME$F
z3x~Hfl0QLWw6nVfX^ZciaW_+!10`@mn(BT{FSF4iuWK^TRVimU3^=QKy8@TmkF!m^
zi>X~9X`MLN3jS{_RDis|E*vU8oIsA|fM2z(Pd>IbC%nz7DZQ4j*;s_2GXi<-COdA8
zju+1+3Vc<Vr8j(fUzsd6s&?rwK&ltP-9}gIi(9cikDf!@F|h6qu&YSHLE-U#^*k5U
zY0P3Q=j&JSFT=F31fDHi&oYr8@;ZsV0CAjRGa45NIC*bc+|yaTFqTNmCtniAXs@_&
zLz^xn5G_Nm2%&vm%t4!3BN<j`#4n#<VDZIIf-PJ!uMnHAxmJtGVWFoGJ3wA&I83sl
zoBFpWAtsL|VT-wPC?he<bkL1Bg)od(O&+7R+wTA{tqFXRPHxg=rWH}xWBa84U15*%
z6D%r}uN))0LfiM*k69axLuwA^aU7IGA8GmJBc$5x-93`<l{Rs!hVw_;mrrXLKPXLn
z@Cxsfzzh%fdBK7H^hrNPpU#8E!ikGWDqnk3|N5)_x9P`)wtJW#%k9-UvG=0GC2g;?
zJjWIrM6Nh0PMQv2wi{u259K3^7zKwI@mKyH4rt0Vo)*PT1xE`Lp+bKck6%rEcq^|M
zTZj#o?|;j<o_SjC>4`;U6&8Q-s^(gsy6&FRFusB5%Z$?`6SvNn$wdM~JEV#M4Yz6-
zeeOWu1XzM6EWW{(xD1yTpo`06d?{oLsWQ(hmUUosyDY3Flb7N6>`%DH8tovf8s0JT
z+Gg&-1E-A$5-n`O!pL&zEzZj|MrKt)Cg&YPhIh<tiVWJ&oVX#6S9k%W?=9RPH}q=|
zfXb;;e~K_W+(QED@wyVS6mEdvOfQeIQ;6n5n6i>MIzhf!C(x;!LKO}%nfE|QNpOY@
zy`2p%R*BA|+H}BS)&s*!<A2G6n?ubCL#w{$-WDwtG3YLn?T#b3e{Y*nEo&A$A|U4c
zgm^$jq7l5l#3E8xb)-3c<$aT(z<2nfV;l(Uv9dLcL^2M(`6{4ESdTo6V3%f!R;967
zrSyH80OU;OrwOE3p|{c6;(gCOR;SCT|2`6dhmaqIChe54@O|9~fgHa5&71mHh8K}q
zcpiqlR#Tyxynge<3a82YsuOVu&Ttq#SV_3avlAfG-b)D!gHRg1;5gqIi~Wn3^!@0f
z0vvZM>lIjjJ7ZcPT3F__s3h;Wg24}M8=*?*j7an4y*u4DxJtHG)(!Bi?F$?z7sxLa
z?fB`_J1jY{Q&RAYS_XjgnMr+y_xk-CHIke75IoFR_x4BwDQ&|9@NCm=HHj!x76XZE
zCzmI0V@qE6YP!l=E7T2@zCTdgcAn80_zRH`fV&t_XW-q&I2F<)5Af0ylGkPi_8WgX
zKF>RTQjEf!MHX)nvb*O%$ZdnqS5Lr7EehXaBSB27oF;gS6(5oEJPAJxDp6e3Sr%{U
zCbum=SmZKopGe}uhR9Z%DDPhS^iN&=A_tP<I5QlGVD^Ymk=fI+tgLPpT#|g{;xdI;
zo3EJEo}{X4XD5=F@YYW(j!Go))>FE}&B3NntfBOAEccr46&W7&#vKseF`ZxW<&}h3
zn<+^OFZGnb6K~T3U{a|A!Z5Zy^q@#l2E}~Ln^p+oZ2}t2OHIqX?kmCO{f8p;pHI3e
z@xN1C1yT@QGI9u(CY&4=d8Zhigmgv;P{5<UC8a=VmhK^`TDHS#qXlS0?~kbDKA9WJ
z8p4%N6JO-5e3d7es2zcnO5lI7<`)2AKGW<_>LL5(mttyC`w7lS9D3<w%$L&_13wKv
zSBcur&9MGUusvR)o~7uc&@DZ?$jDihBcghHxGX{jqiaLF2kUZi4n8};kz?bO#rwFm
z*Ni2#;FQ4Fl-c5-^sbPh=hf%=j8}3c^bIu4o$Rk@_atwg^JF)7{`JPT;H+jqsy6c+
z6lc8k7eNiGA}{=1kwtlavEDzn4ty9yel5OZVm!&%T_3_1LPyxnIp`UB7utycv=`X(
zy8}4a_kTsqx91{^6Q{8bHS#mSbb2IN7(wjUlAG*V#<sb_M{GrI7WU30$KhsSFQ8g&
zh)+$U&qh`n_f6a2+Wt)yNFzV7yO&^nbaZtsP|2EY-O~pxkkB_k;doGF*&ibsxB8@8
zA_r_sv@M%R9)C~#sYoTCLo=KROj<vr8Wme(_0c^)5o0S=-u*?yHoCy#p9wg)2Z6^k
zYF8phI{BQ^aCL}BU;=rj@PK#B3vbYZH+GDoKeARC6yXkn$E0UNxmLymhLRc*^zwVa
z1GuzG9YQv${G~#D6R5NXEXjiB?m_(BA*?YLbz#5M%9FezuGDb38D-b8qr4XsLBrFh
zu77$zeHL`rz#Y$121HcjWR8sOKC}b}j3GNHFk;bCqWxxm8j_6X4xacZxcV5(j>H_I
z5MLJ2Yxw$F<?hW!0bHSh>B~-dQBnYt@$52$NKTOz+M28#jdxkd4>f<G5Z29+sSZ@Z
z(r`052J!cNKm6iN5KA2f_&SPKEFvY1J8qLHghLPVR4BWxKC4j9oO~fvD#A_%r5)ao
zDRIfsD^kD+*vA93gQ(n3Oqr;Uvqm&gjk)z|%cJt{=qRn8PP~9gw$~8fqF;=Jtf@uq
zn3*E*_qJ7oT{cy3C<i`9QsU%)jNZ9Fpkda7Hq^dB>XHUOt<;x^;?`*Goly9$xUB`G
z!`~ZIAL37r*=9?ZNBHI`tVWw@FTj?IMy+=ulOxw~0@CDuhc{-boBe1hI96>Pep9c)
zeCUC1eJ@WuKqX%#yR5U%D4Yq8g>>K63CrPp!e4zCNpCA@dgk1i-2v6It<w>1D75`q
z6<GfKL?su^l7<+v5%pB7ZPVwQObkfXKawOw0K4BBiJ>|;Beh5aDQem(a`Qj`A=b4J
zpSN*5+}hERF*jRLY@sC?8{n`}i-1!;^%dPBhOy3xB;WvV0<?b0OXgPUMj`Gs*_)!|
zZ)o%CIBUP0Qw=`)Z3BbBt_-1rf$0<@<SkRPa?EA7NhIDF5IMp6FhMn``#RJAQKRHT
z&I&w8NM@KWCkI*SQZ|g?UsUnu<Cs1PWIx2&5JYJ3d=eGjm44@jEKOPjW<<GKFJ&WZ
z3M-n?tmx9(vT3z~(q!mA(2;mH&g^}|4iOLZ(A~ZbRpaZA@cFeUaxWKMQbaVKSC?i~
znxH|_=fr;@KV$Xj^7w6)@6LwBfP^-kRn?$dE<YsdTU`ZM4xzZ0#ym;03)`2O@<4y$
zF0-*Ny8B4dLESlXBk*UG{+gf(F>A8)6BRvrQl{}}j|(iv^8vMl4>Ic1B743cI0a&)
zHQ&1Wt3dkcUayA!NK)^AtHf!LoRfM3dPKD01>nSA=5dGF<qa7<*MSq>7At7B_k*_o
zT;wSCpuwpB;aJniSMH09xii;;5kwmy2)NMm>Z>mbbFGAm@&zwbt{vDQSD(rhLqIN!
zwBv6?@B8lvy9z`|sNMk6;i*q$vEPZ>@m$eGRa`daZw|c(X~S-%3*>YwT~4mIAfQ=Z
zgAS7adWzfx%Z2o0tC#b~5i+5xP(9iX^{^)I5y-)g*p-DPPOJ)D(c2Pjv}fwnRj`VK
zcPTFvUW@w5P2tI*Fn|U$kLQ_T4QY*DMnE7}^$DjyH&wom<Ts4Vq%n4iJWNII_iU>n
z4O|j5OH*G29NIClu!wZK6O2eXHdaR%{|XSB>JE)rAK^pa4e6meC%M76b{7`d=~s(N
zUH!u(?aqmp(_?6ReuL#Ii%Fr%t=MXN_=(#Dn<v?R$A>9IjHqM8O%>WLhgS@}5d_Q9
zZi<WV<dzd5>84hLkYnkXH$s80mwuVou&+#GNAhF4lzM;V)RiP$7I>(=7uQN3z^=sj
zcvtHL_XQ<In%w8dZ@0Jk3E6c?Ux(6!nT0Puz2A>R%Z>M;mHwM5Mkh$7Y~R1cn=$;6
zADi3isMkf}4thE#OpG;^y`RVXUI}?-V3(s`Qcc&ep?WK$m04<ukfvVkv47iaZO&=l
zaX<Gfg!Eyl^wIZ=pqXhkTjoyf0`ome!w@c8yv{V~LoA(EM@&bv<>`hKWVgU_OxJV9
z`iz5!W%X8wCd}$dzAx`qZifeN6Gmf^>*g3Qz-_4>f0ZDbCfrR}OuG_x7N%y@2u;UD
z+|Dc`hJUvBd!btd(XsL%n~X}F{Z+b#JJTXwkyx9g&HPmy-^#eY!tBYt@9;1;e%!&0
zBexOiYTXNvfLG2dzMf@KvUlQMUp|<=Zx~AT3KuDl9r8`PwLHo7D&H2_m>WPbRroG~
zRR$lYUSsf_4#g^W2kY~qYgnmsmGb+B2$0|4^gez_xVXa=T{UJyp?alO$5rf155!ZE
z3$W3d@0lNe>Q_>Qh@0g`OU#c?CY8s$TKY7MpPdxFtQhhboK-pyx*dv<qB$WDO)OnG
zGawjzTQ&Cm+N{~0Nj&4c?8fM1B8sr@0{`azR9L3HE2}?jTIYaqmiHlAvb11+|E=;u
zuiXq}CtXp)#)|7;8WNwB#e#0Jf|M^29~uC^e@pr`Nv#CXjz%uzF|s73QQP`w3Ok{o
z*L4$x5l51NE{!l7Px^@L8CkwZtzp~=rf*~kQq7GmivlQMo5ws986ZR4QDDFuW0ZCC
z^_X5{^ce;1v<%P>q-!DfVKiTzgmCae_#%g|FFYrta5lkwASo`A%Itnrj^YA%I^8~K
z4b9>l;>j4%FB+;XUt+GG>%<I#tytFfMn1bQ`qU)}u?<#SPl{nubINd?%2fny8f|)+
z5rLeuDkD;U+T-A$Isj~Rr;F!*IQ@T)BdD8zqAZ0HPPiZ{))do2twKGwj6A^l1n98=
zL`^o!d%jG;B!6z!pT9E9z;=k4s!$6)8rBZ-d<xzw@_FjLUlD_=Qv#)ta=`gsbmA$f
ztOSZ@mQYesp_VlG?OqLuR*@JO)$2TODsh|XKdfL&pqXCJj%9Q^U_`gr5V@2ma{Ohi
z$v8}q^gXjdEOwJ*<uFoAQS&*ae>LPzg{XAEZ2GiBpC*>sK`Di)I9jZFI3xpGLMP=s
zs^)!yZGH=6RWSv2tRUnXr)a+dpuHQ!??ecU@Ac-tcX(Aj{7^98(#%XQdoFcQCoLH^
zRCGkqb@6F}jk#2LV2gQjQAqJEjU=@8-5`)(sgojT|2W^XNG(f;sMF-2ScB6~m$XRy
zA&2tMmwCy>(1&|F^Ugm!KDga<2QeR81+3wkQ<5o`h{WBuu-u2Ka#hc5vOA^k8%<T{
zps>L+^U2>Qb5x6{#M@$QNgblL$i#+*Z(ENeU=_pKI^b_99i<qt+|D<7*k5;_Z1BB&
z*Y;e`&au|XVxhdV@^ra!O~xs+Qnm$E)9rmG6P%i|5e@er5H=~g%_#0vYz6T6Gq+~n
zs@%PgDb-117;(v6Zbc<ADn%s%%GHyhOK-VKSw+p%=37H>xA*JAO0s+y#2@c0S9IrR
zeN)_pxM!R40z{|UiW2qfDF}&><-eY(;>CT0pH?T#ZDO_m-A9gd@6(+l9Bav+LDdZ}
zYFX-t2^o7M;{Gsz{;>A{89H+lq7NJX1b5#<a^z<`dlNzeGWfcU$nrp~us=^b4|?n}
zDH+vxUVdo<;W}*urqYaPlwF$jX^wME3@Rh>n{Ae9JT17Dhe8RhSxqH&k@y@?OjoeY
zA5TAPPLkZ?V4;(D7**3JtQu;7%_=JKHTZ}Fc4;tXj3ntSpDpp=SUcOfhkqef4JFGf
zx>?RWLB}KqSPMS^0qF;yfgZpfH$p`zGiH+lwki!?56#R?h;62YB8mu<lpJaO@DAQ{
zBpt~d_-a3WI#{uF5;->4oeE~!W&1qE2NvXEk@3tE)V(iWel=CgtScInB?_}!kxs5*
z?pzRR5`M)}pXSU+FW1IeP|ClIX|Kc(3#eIKu0Pw$lGFkODs0PiPjn`EbqAUv%Y3lh
zo>Fg$z3l0GZMt9O8?NHVTCo8(2ZhnfM*!a{K9*Sq@mI1xr(W_#<4ki3VI=Rw16Kt<
z^2DGaLPzLL;;2HlS=J5w6hpTf1~bmZanvAcjf!qIR$_K?^ky-XT_;MaJ23sxpuF%q
zSGmCN0cHT7DumFKF<Iy+0WL?>h*^tEPN>Y3rJ!H*E$hR#tys@0vH=(~sd6=?&_!nB
z_jEEi8~C--^oQnm=Vj)O#7egxg0Oq7)pYeY<ybz!0c!y_HuW>tTBMaxai?>y?X30R
zE7j;|wDw?WIYr+bH9=C7oj7oCn{Z`3DQ%8CesrTqpk%<-rP*tPRUJUdbmW-v{_RiC
z^gx$Qjj$q_4T&8!TbY3Ytn6&($!$+28Ei+wcf;d=19Wad^FG{GitQq2u47=42a(3Q
zQyA;*fUIODU6i5)Ea?;WpRK@Klw0?f{+T|juS^8fg=R!{f8a)|!d(2PVf?4mp}yP)
z3kyjN@jH*spv*|+F++PkGy=M|cRblozOA<dOPZez-IH`al~vI0o<x@GuDud^UHg_H
zw*Rksm5nKLJ;*`UAFl2n(OC$~6FRg^O=~q_B`uuTb^FoxTLeVe>>R0TPguJdj}@MP
zGp~iy?l4AsV_e8Lcbc}QHw*DS8!;tzT!s}TWL&8%GY0~JH0>9oI0S4{UfW#em$5to
z!-DS2U2gz0kMZMP;!mE33k;(#Q?-@28%w1wGa7sjZHPKtF0CI#mpF`mj47cYb+mKG
zr{NAq=50&c?-(xXlyI1M;?t7Zez`S_G$1MqgS}Ze&H`YhXN2|q(fp2=YUi`pwjDlz
zIYlHY)KvPX(fEhbg9)Wcovjvp?vvM-TUkp}S=`y&$78EUB6Xxi<C%*azO8xgy*SPN
zkkPrVsfb;mjw)SYUcTIIucO^|yE}6u&^k&uTM54FKK4*U{*aH5FwsqBbJdpEO0NTQ
zB#pj<_=u?X1-QAsq=o^C&eDW>vYW9H1GAcz<WlYT3NsiZ80s(d++Gg==UxYEPk!;k
zZjA6~qg}GUZgky@kjCJ9>VQ0Eyiq#(wKxJV)9yRqhh>~q2Y6~XRDag3?NNw`D;&zz
zA4CZ=F_pCMb90V))-gXIC=CQCiI^z`=o@SiAG!q-XXc(?WCAqQf`sDZ4UvymGX%#y
zQ_1P_dx9%Dw)C9-%=w1j;Xu4NUYf!KiT+HD^4smHk1lEYE5NAL>-QF&k^Tt_!(}uN
z+4$6iBj+VBv>RMB&%C<nXO|^X)WBiR!cuzo*^<(Q&!?4`NF)z%M{Ry8HX0F|M;!!s
zVwm{zOu;fd&XmsZVvVUtnYg1BZPZ5y%pTD+UU8pmcB<YG-uyLWA-~znIHgml4GW46
z$z7pFHXCr(*BH*a9O#+>7;ofx#j_gZ>QE=pKTx;ntca$a6Rz<}1C<P)nLIbhtDi{4
zeSVR8(PEDR;Ex;E((&C*%1*0H>$?J&qkI}eRe|M#`5wC~Sz9r7$SkyoV!EbbQ>u;a
zk$>>Crr3grw%=4*)txeX*)7au(yT7vaJt~u`=^(BV772urt5Fbu(8nPvz9@bF_-wA
z?VX1?hgN}ofK-yO?%a9<D-BRheWf_(Gywj&oc*t!KkavOk*IBh$Q^z4@wHRF{eSio
zK>HE;ECv*Ak|&SXLWGw}S#mk;sUfdEQ#EpamgNb~L`S1Fro=|+a&wqc{gmhw+#@y+
zQ!cBeL$T1lzq^+|Pj={g!Y7T)EUbmat2=MN>9}K0v7h7(dMvUM-jQ;0I%Iz@-1_QL
z6{<k|J=z>wPy88^(qR^&xU8{_HeBP4-f3FFb%{(t$MJI|Tij$F#w+<nJsWZsv!i&5
z&K)J?S=DhpL|Gx**3^rid=H)tXu6{B-lHYE`GlS3*l|xenc0KY4X|-nE@XB)Av_FW
z1ix1zazs}YiGC(NmJ%hs$@Bv>sD^LJl|^m1?l8D^4O~rO?Y5UR_D#$>58g#-D7Uge
z(<0oeOL6u)r<wv!K$aw(^qz9Ws!`JHfnw3_9qe-u&e3O*TCE|Jhv5v@-9wfHeiPwg
z%>7hq=?7>!t3W1D4lWwuaVj$(HJb2N`7Q<U>`$uVYi{ew%pC{I6Wvwi>%o~`%g2V<
zv;zp|8$h=3dh=B@PVx^`>%3$v9$RO{B#*gMiMjHilI=&$PZT*>L=`lOs(S_=<*Jx`
zJQ<TRq0u9bZtZegs7XQXDMvb!Tv#?*dz>Vij&2;=gqrVzSBq4i%8x!XHZdV)BFP*z
zt*J-@LOCe)%A3;LJHuhE)M`0F;pfHQSn{BN5sdQCVSs<)wEZJ^nbvecm`GqMEX`=C
z)*)P3P~%|SZRPA?=~uGR#eLHe?YWh&hxAP&&kXRU1A#SV+@@0?D%6QfqC3Hr2v|B5
za)j)<i*LjWg)vW455Em+*JSJ~F5pD3B|6)2b0}uwgFY`gMG0Dmx)9;m`;uukjHk>R
z7DY8Mbx!26Tvlm^XI>2X&eKM!k~vn>(O3E8u3Wx45zm|}b(dm;ViHGA$XaI>S3msc
zs$K+E^`8Bu4**J8lrpV?yED!{Nq7C%UI^eulcV90eSX}(^W?tJ;q>Q$<2;o7rP~{u
zQ0E?oE~hx?ZT7GK&^@B)3Cv7|rQPvWARV+1%AADEj906yftxH0M)~xMg94Sd)7->3
zB`ndwgzp$Nh>S{O@PyLk#_16JFySK?9%mO4e)ry*u3MyX?<%d9Iyj2Aq?A>$B0I8f
zpnWst$JEUwpCVztQVCh!vbcRK0e7^8(C)2twd|5&J`(Wtc%d^Kt%Y$Le~!u=VDU}3
zq7STN9aogtl+s{PjMUIMojOg9j$VEXH;`OL#HGPwyl2@7q_aNck${+R5A!x%*!coB
z@<4`}m1Wn0<m}a~GxD%R`9^VoxiY=>R~MmIOpcJ&YrbW^ag*s&fYqF<M7{o<>1nbL
zM9R@|ZB}yZ96Sz@QoO~3v+ZjndIZ6LyedG-%T{d9@WE7cW_53RYaYqC0f(-3W$MjZ
zb(!tX)+F<YpzzypPx+^E%r^2*COs$C1d@E?Gw|Epd%L!{4HUU<dMV%*D@47*`ywmJ
z(IN30!!*IzHnPk<qB#4Qwqeb<RvaxM7{LgojbE;Ti<N?W4OSB4XEmRfxUaZwWSlZ8
zjrg70o1Og)OvLu4?8<zYM+D!#09;!YZb$?9Z-^wxZzjcys)`FH*<l6p2qpG)f)BJe
ztlNaxzoofORS-TJ;KC%<b;kL^?FqM|=g45#rIygJ?d%6rIEm}8ZRnlv^vW*5PPn7B
zLz1uTfhvt@vq96<dd=2f+WunWXHF7`hT|n9oTW~e5LGuXb@R3Vt#1C^RR7ak|8HMH
z+2lh?u>kYib9cZ7iGXLYgO%Vs>OaV#e^zQd<0^6in>DDLh>{Ul!f2%MwZEBJ|Fkv#
z)0a93i=1!~ja!EREnsv>uCxP$QqgdWv_Ll24}SlbSLya-0^Y=v>0sPc4O<+|oh^`O
zyz;*t9RB_P`p@s=$iRk>Ltw8RvQV?m{9!%3i(-Ul(gJzZ7=6IDueW8CAN_By)@=iR
z4$X~EF<-+LMm?91r`K=)?HB&mE&ThF`mgW0|59F6fg*KYqf`??Ky288{{Us5%x{mR
z!)`2De(96!@jp_CmY=2vZGQ<IW_kVVWCL7%ixfil8<!M#<hCu_pNfEJlCJ&NEB{aL
z`uESC5dz>!793G}IBI_DnY@g08lgX{i=x7Dcb^dZ_qz*3H_eS_g6)Eb9x8U@gj60D
zBf0)-Kl(r2xc|$~s)E3IpcZ6TeOJ-{_vQLOeyHyM)4wF=1HL^qzZ3Zeps;IzG(o!s
zph|w1Vd|Rx&kbB(7#-!dv=xj>SD!ps%dU+I$%^qa^n3jOxQI^((AYt<eN%d#|GFEQ
zWTAiqp7ixq*WDE(0m18Tj9!xb!-$?+r|%aF<3Z##fHITNn%3`=z06OeHL>6mv8Y7N
z2Cv=HtVr#k<FwGL0zgxy>80M}4BQXCHw-sBDhj(V1Cxpr0|to<6^Jg5l7eZ0eEYVZ
zB{!g!tpRfL6^j+rO|l>jy*t0VHNRLshf@I2keP>ot#>bSs{Q-^i#4C%bDsptuHCZS
z#O20#m70#@5s)uKD{z`($iL`k*B|}DWp)$`qA$<)b7~TnDH_HO0NO=@9iJi8V#r67
zuxr8{V`>fs>lNj$tp9yU{I~BcAixb3CwU%J|F8R5*Mj{Ea28(!3Qwy<*D%`yWSW+w
zqleIQPWu()>_?Qqg!Xnj_;ye?0;~q^stvyRX<6Mgn{&wtcwuOuDtVvD@^Pj+V7T@r
zAh>`g5qf)pQgGCa_<v4KqIuC`DVOiic3^hiX3&CCnns6sXSIJT-6e+p!2WzM)Gc6x
zW(}3I#kmhf7DK5U!+{Q@`0KX$&7Q#;JM>cR0?@J0Hh=RR&7L}mqYD+mK^uC3GAF;7
zDVHVNF=v{a#$rs~i%IlfbS$T$suShi;ZJ@g258{2Q-i+$G^0T2evD0yc2uB}d4PY2
z(0Ax>tLUV=^WZxQR3c?kyUW_x%TJy<W}5)-N$+BvqRW9*&N~U0x5Yx|1G&Cawjm`*
zhUNdU*!U?>mzBa$Mh7UxKRK#TKp;7gnxg#EzhSgpqSreL4AirLS>PeC`ApDjrB`lC
zZ%e2bq_738cU_NmQ5rE3fdJFMJ$Jxg>mD}L2k30=$G)_$+`M4HLuGHQ0WJyIm$$NF
ztG{l6nTn$tA~4D57;TN{7@U&NC_(7Tj$@O&f~;P-53KGk2EfQYw+`seqObtv{chu(
zB)XRph?sY>^H^{gBj)UMsq1D*;U#VA1Mn&k@biKY4UhP9IS}@H2zX&FC6YXI(u|Im
zoJ(4dsdWM~MgTsE%WL`P9rhs_S9MV`pRq$whG^9Q;U@UTKpddf_a59VA3Pget7;g#
zZ+?56)i=yuAg#C9NTkSJau>9U!QkLJpn((L`WAiwy*q*S0q@?G(CT!gg}U;~5zR=3
z+ck!G7bH#B5zU(0?GHC{Hs>;e>VCm`evqH?e_O@><;;$Hx_D7r2yY#r@A;?pbWgE>
z&<J~1P=t-r;0tt_Bfh(1zCOTOaUvV^8n7~o0^FCyr7wrIA8=E&J)b(4u^T>Dkv%42
z(UzkO`Bo431QEQ5dU`KrPio~dF{L8^=9hgG;PCY=GnDrECLl}CBb)~un=owV=Q@6f
zC}@XlejJA6MxZWW_{QD!_#Lg5^pRp5>8GiM*1C=}K$Jz62YOq)zLeO_fsRH0+Cm%%
zJ>rg~w=v4R`SM0KR5j!z8HGM7E3=2WQTUGN<rCim3-lv^K{sQ5n9E(c{s9P1oBfei
ztu9o2^c?_#6p`x&_btK>;cC(#oj}(7>uLVK)p0>o9Uqac=t}(iH@kD<037b%;!IsF
zFW27DB=Ot<`0N)Kz&M+H4DftB>>&u^@7R_(zyVIt<I?6U)33mIXxJeucem#%Lkybx
z-!;X)lLEv?1$L0%f(O`-u?|>An9O^RGEJQp03u*jtr08ws;>2upT2?hRT)X}U2bNy
zx&x`pn}2RnG4?g09Ez_<qdeFJY@CvQp$q`LOo*-sI5?gy^#%U+Vki)A4TCSi?7xnF
zg#|N>n=^pldi#8A|IYJmi`Qp2Eq;Xy{8lci3J5ef`+4`zBAJy&ebOI+UcuIXcSWJk
zV&zDf%66aKZ9s2X>iQXaiG#S01_hmyU2b;^Zif@CD*^xm7drae!>Fa(mS6zxO~M%S
zgSpj5`~oahZ6n8kGtqn|tKDw1ZZsq-|C2=iN1#?`W&yMF<<}dl;*lZH+YKna@tK$V
z@mK<He=Dk_g_cHm83Eb77{MDQqM4TsZ5=8>!CCvF=oXLT+|Jm8kJ;Dg&s$V1TGyX5
zVwbTI1C`M~>Wk<SH<S(zoM|yf9utp{tE*=|lDHlWz8>sZyOshR6QKg0*tQ8<?-))i
zRNkmauM@r6M+r9%tK9;f-B&h%SCm`0LTzu|f!9pufY)bVI6Z=qYh_xoo*4>kGr{(l
z0nF-0FHn2muZ67-jp0%DG@EgP8l*~lg9q1R2lJlINdcQIt~4wZB~7B22Y?}{+~qZ}
z7A!{HKCqs2{(f0|m>ebDzxRUbT_q1);>pcxhToTc&}#*unuPWsqOnfEeUkuX9jnZo
zE865E{(!S@h7NYKm8t?NakU<Ho+|x3^+1>WBOq5a(HHNk{Z?}+GtxM?KW#U?%*d_w
zIGgD%Kzl@VpW;CNvW&-;YDTvsF*EvGeV6LXMv16ty3&B_{gvD5riMll?(bSAN~TQ4
zc6#P{x?`gQz%F^;<Tk6#;QkHr=C@DRv3qSGtx?S3`RqsdSJ>pADx~2xb#fvbe;b07
zZ;$%zzH+rZhrn7K^XL^^ZzO1Sze8z#(u81cts@1{{a2dYz7WGz>B56Df~&8!LAE!T
zfegT|eBHyK@p%#02%UfSsYYGbYnv$1oW32D9zYr_+8_rYj?ADBwEThg#F^pj)xJc^
z`kvXUl$W)C9=y0jkRMi>@5H?v#Jtk@-Qy948zjIH78??*q@S4uj1_8&XtK=w94~;I
zO%?t4uCZLk^&#m59m+KV=uO)4h0E$q-1QQKBk5(g#sH^EV!(5&7Ow`_HQu>O3y81M
zM1q?0P^$^8gR=I>lu6pMpl58xhU81pPu-g1tH329!>3@t=C=6|a3qET^;%MNAs}zx
zN@v9~<@k+e)Y*?2J#Etx7hf}SFdAatCY(jRs!GB9sXDplxQ(X}Qipebrrxm?ILA@}
z{Z-A6+MfqM{t<2{GfgN)JEE^fL9Z<>VxZA{G%j`W^T4O;1AKRBDBTraSSCkXlL_MK
zEYkzbVlg)BK_IBj(}b78^FD?*^QR+qYRX#1IGGKqaY~P0aG)2&2R8Z?p082$xa@-J
zKADkU+=8!`*6`h4hfBPRQS~18UlyVbL`+&|Ab9&FGSAdYRL$_%&vDH!dJjKjbyl5z
zm(U>@J(f3JF4nM~y~&fFe0891c1(QrV{3Ic!7ZRT<%B~j`A$o6x1@+wOMETD9VLJi
zfJuey$C)8uMb7gQfP-Tnie)x<`2ZypMIY~{ynU}7I07e8_88(##x$fRCkAwEr{+<B
z#p&z_=_sY%0Q~R`A9#dg;Mr{p0SvVH2BoVJ9dh;J*U9ATIP5(Hdd*k-{E^_^8$##i
z@2^bMF{b2m<*or==9Av9#=V)+rT%KUNVJ#~K}~or_|X#TdYX)-Vxw;sK#zMm#hvr7
zzHL)y*Y4h4Q=aSjP;M%~=TWgvC%_JUo%-fmS>jQ(*rG#m(X~jEPoq?N3SdGQ#5(ne
z85FcmQovt)m0@7!xKF!(zt_jY_b2pb8tN(%Fmp|>z$auGbq@P7U<4xP)`kPX9XRv%
zfV;xGVlDZ?A{Q5+Z`Btyf{Tdd`O(qc^Gn&=arj*ZMW0MKJ~RkEHJF|CV87vv;J_|U
zcSTy$s%xf>U~W68I~Jq5*u>{;D?wLA1+t#7LA1ReOUq}X&P$BT^2Egt#D+_@r=xVr
zjmqe;&^HYi&Gzn{g01*p0s+QiZjYA}<*A`SlbKL4fHsAZG#>CnsM=K_+i=D6S((^=
zkS4dI-ttkRPlOxbRvg-#1jjn2M|n}P`Xg8uVWMlsNd7wOb*w=4NZTnG!G=_tJpwxP
zPE(f{G>mEoEJzGm^3~WIR}KQUr4qiaq${sjbG2$90)4vvxHg2AJ_><T$%Gu;wG8X1
zXB4Q7hf*emH{>Bici}nP?LW|D?wLl_i?$I5iflw}rPdtOKfiqUqZQ>6XL${1a@kOW
zi0N-uX;i4BmT8&RP#o;|HL9-S0}U39klow?;c?E|p4LR!q}%A_*FfaLg3ao2n}{#N
z4HXT?*?l+(lAlD1Xuwq?Ru+K9;_>Dq^|Cybj5YFm^25den24Jhz2Fv6{za@OLf<7s
zWLQ=G6dtcvU&`+hkXHquK3;ANJp?qEzMdZ7$6&^^jfNiWum=Mr$vGtQ!1Ek1ox3bd
z+8-v0*}e)aVk9y3iVs+y3|$NK{8_=Y`M}?&FbhrP!WhEo2rNZtrl$<hxWt1mR<WS-
z*z$mHv)=0#*ayqEkBsc!8(`}vV+bf`cmh6f5h#Ql^P{*&jKjc7%7-y$kwnQuQXDfJ
zv*Lh`@X(6N2rs+@4F{LmK<!T;@885Rjv;GGA*S#qiCDERT#pCEVq=l4;+=QD#&5ZW
zmE<1J9h%^dN`Gnmz9lo&X&TH<1oHSHgz@emRb^?zmjO8u^{ZXaEP4^E=V8{#I5gOu
zAcZbba<2lkTSBVplManI0O62L4RDs{S@mO#aO&JK%IYM(UBIRaxkfbrBD$dQFbnKU
z0?$>^B^PTUHMO)3JyDOHG|{mn!DGS!L%fD`)A$nUv{Dkr%0~gL%nS#EE&|Kp_UY)^
zXP4u_L52AtRICQ>p%IUSp+J{-SDzGEMZ`FA#ni*kIEbyW$;n?>qnpIxgUZe%!Yu&K
zZEsi-#jOYd;sg8s$=LHaBJ83ENA$j(8gn%fD~?O;2;Lab1u(c79N}b@gfUuecKZ&V
zL;R@sHTBXX7gt!eoO8?<ZH9n=QfZV*bRy$e)BD~s#x|M`h;y#7GCc$6$i`G9cd5)&
zNQix|d5!7l#=ax`5e+Kzz9l??nsia=8U?ZSS<RT}f|JaSBuuHIN^0u1;pH%f3CJTj
z?>$|{m*r)vGaW#3tH${%g$e?gL5&QkX!ywnQYsq3t|!f=&x+?J$~G?6Wtg-IfPqK#
zm<azyn@!8MEMAHDFjKW=DtyaH7r8Mu*PkkS@QE*|em4qXzN#8LbdI($4)l2B&xmwI
z5_2GEVpDR^5&{-{?2AU*j~vH1`@DgIb4U9*C9I%z!^rL=EmcK_-{H@@0CW2X$jdTs
zB(2G?Eo|KXs`F1j$<2ly@3S*V$x@1apniF6RPcuPXp1j+0aE+L&#n3=@TA)F+2H<S
zh^L@r7xZvwE&1usZ4gxM`x|w_Y7pYr8eA4h;Li+a%!E`F^$BIW8`GuR&!}4GbY#x)
z&&m}P-&zg!it1^ta{VvH-a0PIc6<8<L>Q0;=^R?RyJzSIk&u$^?hfe~Lb@C24oN{I
zq!9@z0SQH5C`sRI?!E8(xA*ft@B2J|`58iHuIoJ4wa&HH@jZ?W<}b5ddjL9nu4LkU
z%u^8QQ0Z!k51$>zDnXjSYMDycd&!B;B&BIpxVIZBF+MW*GXgj!820p?hh3@huz03l
z38zJ&^@kjWCAzAOgck-%=!5Gq{pJPl88jWI?cNmlNlwh0_<_vg2Tk9PtM8jF5v_*P
zd-}bD0+31%gKPFN2h&@_nwqr0-~)fe>;^l^VdQ_zZkRo#rH%Fp1nN{07r89%qz>^A
z=zct}3GN#6whj@`vuFOg>1jUulw$5t@*=aPpl~gfTnP3|bbZRUcpU9h1e#W{Z%3L&
zbQ<R>r4Pgw;lZoEd<JnQr?ilZ*k-A8OF#A4_nl7y_b&&ic{t7AQxbMD^#PfogP4SS
z4{ip1?3!QENI+}(n)gfmk`fR@VkI;%9$X5A!s2p1tovf4e=JXNCO~G4L(TFw3V}Y0
z;nkL(ge7PM9qQD5ig{`NMdv%@YXbQw_8OOLSWrlcQmG<?*2EP~zD+c|Y80w9@|orR
z*r*?eDWO$*8!zXcVww0}$)IEUsla#@0Tdq3`*x?&$%t>fWYkLw@|xw*%&4D8XNe-d
zmSs+g2PJGv9R2{XB{X$+%2FI4MKL+Z)X?rku(Z4grhNSvZVVEMz>|f>;`NB?(a>u~
zHv53ZOLac-qqQzvM=LA}f77XWUE2+Z{o#DEwtj_qz6ueh_wl{aiXnOr6W&S9EU9Kh
z$TKV6B4P5W7oz^?4Dosj3X%&%=v71KpIaZfA416BQNIZM7FHB*SUAEcp3Zvi2hJyz
zJjg?_97P4LEaEUJptZm5D1jun9-Jq?vh=cwEyot`fI^vyzQ6u!K^xaEqx$hZpS9z7
z#_7*<Gg^`TDeN;FG3hQ&vz@$*bJan~Z1tUBZDNmNY+^ToXV1)tv?QejTGf`JxYD-M
zM2y1Ec+(FubGW%vf6YvKtQb@S?(?p`&3|npU71ZFbu`sQ-53>iU>r3GIQpz&Q+=8g
zbOswy*!As}U$<5&Jr4K;uPyH;z9ru^VCb7l&U?OL>#v!}pEKtqmf^?)Mrt>19-#(~
z`F}iG_@424>n@bzQ}X*c?CARysdUW^Q_U{}m{--QE2^;9GDxeO-pY;{yPa<4ydre;
zz!13=5w_RKN8ik=xw5vwrt+$ILo23dhs%+aw+8}L$d4Zy;jAc|FN^evBw7Ny41$7p
z?*Sbrpf<_EKK_rsNexn%0R|7y&m(0$e7?QvK-?hW{PeZsPEUyzbkZ!|(A{<CWw7+t
z@#JW^qQkZM>WC5!enM-iw4g48pMFbPRyF*L<oy=E_Ry%v53?fN45wq=ZHbjRXE7W<
zl<<^{RjRwK&KH)wRkprbS|^`eXCaCzc8g3T=1Ad|U-Z=HJ?33_MRk%`hMY>4O|?|a
z6gW(U-7y&OCxhb+)jf7ziXxo?ER2hyScWI~3Vhvl><$<o<iiWVPGs41<=Me>BG101
z;#0Ops*8%l<HHWhxq_-CDjD9Tljs!E62gOM?`3U`#&{h9GO=(CwF4t(vR!9VN44<Z
z&i&f-q2927{Z61*&;Y5!+*VN|SVtsd)YYt(e(Y0?au`wpgLm6WZ2pG2W$&t39Oxl;
zcakwPu!<GX`gI#yK2nt}dE&6kdWtvAqdRoTld!utADyRH5N%rvKHlvyfbw>if_F)~
zDN|OKXs(T<|82*X@J4%ujVX{HUz8~5bIc9i$|?gsQkVS1KD6%`0DY6$&1b403Td1@
zcU^xivs!sn`>PgX?!(+@en~?>1G#avAHJsCX`>fZdXR5lHk<Ce+GH8;hO)^bt9qwf
zHaSb^AzAX-K-I<+b8a@z^h28I2hdA^LZhdCs#VKmjy0K;jj;oed`l1|cIYj*fJUK+
z6QvKe(ckP}+CW5^NPw36M`{*(%?Ph^ocH4I8*LGGRS6Ynx=PzK{~QjAA>Q|D6Byqy
znp?Td_YAE0OuxOr*`Y<z!$W2RTor)PMHQmv1;#I9v{XPfQpHF+YnR>N`*Zwhco{_p
z)UnxNx2(XvtFHoL>DgEwRf&zhjy_Sb=TZBwNSIfCuAc)zg$**YN3fK>zCV@F7tFMB
zSdHZGtbB<g++VPZ3(+KJeQ_<|D7A@<+qkP{kYtGeV|+gGmfdC;9?D)=W3ZU`r7R*{
z?fsgL;JiCl`UFre_>@t5z4TGiT1B($$AoFdx2M%iKY?VQegYa1R-(AlEMaG$4%KM2
zGSTmQb8^#@R$`XK?lS1E%04?Wx6E!Jrmk-tZIxhv^Zj{)|L0B>Cj+t+{3Abh3bE>6
zNotPS8dEFl0D{0EHQNf38@Zcp8v<t(>vT5RC}qV>OKR$IFbwwqYQgrsd5o0$&N>Xw
zZzs6UfuKhoudi@+%s(kOQP8p=Jn=$^16ojdN)mEPvdF0&tRK~;rvKIpP&H$Ko!X3f
zJGSGH2#LI=GLU}4@x}!wDq?@H_`Xw|UtH9fH>p6iK8n9huQ~rj-XGM&fXg1FLL)~r
zHV55*-YUkfR#BCPZ0t&z!OkkzkIKaE#qh12BS8LIZn^~5(vLYq%7O5lEu$x7v8BLD
zrHHNBQyB0Tcr-!4m07#M+w{`++5?z4EfPaXnAm@Zm*HXYbx=DOQQ}ggvFy@Vv0PaI
zs$8vJ*PJ)<8LeMXnN;FWKLW`iA!M#S_>5xXw|pm{v38=M3T@gHmPp%kaF^)>YFBlM
zHVH$WYFdFG4@t%ItOi?ALU&7#XtvYcR8s8R)oG{Cb<Q;sHuhEaZDykRMM^jh2HkF$
z9O=&8_T|sr&Nbq)992HGOwVPrTNFmxG^z5G*j{?~Gd(Svm+Md1B|ZC`k>jVhO`jn@
zKaR7wr_<NCrdh9FlFlec_O1iNN}1ulXe_@yk}``mH9H8`Q8>YPPFPH%^n1Y0!Bq2P
zYR$rvpC(EJ#Y$Y}SZaP6sfQLTdzrjS`Jb!+865VVSBk+!Vg-=;l(Me1TZ|D85ezKd
z@VhOHqwmJ~D~riUd{~#g<x}ro^vJlLXHZtj*)uch0X!G#0^vd0Mqy44bG)U-XR!5n
zH|z@xbGw-U031c{O+S_Bz}20>w({(dY^epck!g;9MP`(6w;<SjT{b|aud#y#v*dy=
zfz|hqCB?+qjIM%6GKXSX8hYmUAS=-&;H8iQwYLA=+|}$l<Jb?Od(cL%&pV({=6rA7
zmnRWa%XD0BbhIYW<uNK>&=LQ#O7UPZ%3k|?v#&MGITSne3R$JTJUDC;z}wB;NRU(q
zm-%eHNSmn(<Q*Zj6Od{v>0lk&*V!KVbny6gU+=s(Vb!BJaSc#;Gq<-1Oh6v)4BF4y
z3G!3h#Yo0ctt2qo)jiV%XDy))SF<A3G_4o=zWbhz+uPEW1(}|o6Gt^Bj|T&^IcG8t
z;<f)MtUeA#IP`I-nFe5NMmk_iq_y5$eqfb4@G&ejUGFH9ET0f&mjL+1Vde5b-%cg5
zQ)>F}23Qs$N(;d&*SD`C2lkhloFE)%7FYu-G5mi1q>E$;&z?=i&?E2hJd#K-#FBR<
z-}LMRN|WK>+}8-WmprQF%-izy_drW3hk32j7<iN&nbBGS4>Q(8c*jUUx}Ew;wD_P|
z+HopsDygq#UpsI_myhu6&e&MlDFI6UVw3=s`<i~lnYom&UnAsZ{-en)P-9@WU3T=i
z;m(rd{;7lJb|}4>%K#LkyNnykU&?b@;ngWRAxF#b%6i~mVD1s@l84hwrM#qFx5ZH5
z8&2{FcEgQ&oKR)RBt{8zHT-yM{DJ(6Q%KzQdl0=~9F8q&S}Ibf;>IIp@)18Q{4~_c
zZ2;Fw-Kx7cQ*NU0Cq!7!pQ=2uByq4&Bz7hMa7-eLRaQ|*WOQrrM4`pG7eu*XNBga-
z3@DudSf9Uj;syYCg}UOX!b69n<uG_3;{@JNa#zqG5P3|Zd@Mddt*x3YpH62cSWB-9
z(2}P~7u+mVwu@HJG!Qind*bD_pZ@?npFC}B{IMNh*k@Or{wjz9DtYWwC@?D=@e^tK
z?5+J5=PV9<Ps27w#)U6}uP01%=n0v*EQy(9iuW=gI7}S9K#7hYo!n9yWY{{nkDHLa
zWQ{c}vSJJUfCIQ6OvovT7U7zy*6(hhLxc;|a@9?nj@kj4^~{S^R0+t$HW^t}PgLOS
z3qCs;{F;gR_USBTmB)sr8U=JNrwWp^#<rhBi}srYV{3VksdmUzOF9APgh{k8k1U_P
zps}G#^0ROe9EFj1B~lHTPeHWY(`K|Uf*6FuL|(lIx(yZrh}7=9Wyx`-kmZG0m^79@
zKyG$Qn;UP2naHM`3_T`!X=X*xGk_MTgie7%h{U2IeC|<wP3_m+;wZ;7KrAgG?Q1#`
z0kcpU6qGKg_P`q_{U}#8zB`#CoSQs56VO=0J=PZQIV&{22Lu&WkFUYaAGpUz3C!$`
zs@Pg-TA#ww8^bgO$K6xre(6kSgOt0<u<ReB(o<HB$taO9(x`F{eqVxUGSURWVe1jG
z3rk}7o(|l$&BaN7MHHr?S1r5wK<CW9rA-jux>0C$YBqRE$JXWP?&KjTh)DlL#S}d(
zUl5oKVz$y&n2&T{DPNef*fx8``~JR%ZHINm>GHY)_G4Uhg%S4>CiW*k5I(k*s>0bM
z;rd$tIyJLrc+?>-*zQ=KLQ#_TdFmCvQnK4qON#XfLDXFOGHwt|Y5zqQi`vRLbOCqV
zL*0AFaQ&%qsJq&q*D$7gA#b%6rxVAwrt!K`?AtyMr9h3~@_9^R4YZ<3YGI}>OMh_K
zbsSq(gx>h)g44av`%b-V#RIQvH7CX+!3QRI`UGH*fLGsr_)H8UrP}{?fAE?y!U?po
zm+w>17O+|9se4jXS6o-x|AlufBEa4S5quQtj>D2ZF+7EbC+4<epwJ%U;1{;X2a^|0
z^k9Sr0ngDOx9gzF)9cTa(Ql@&0H#|tVva<c?r8d-iR|wQ1GSM^l4Sx|1&|N#g{A7u
zURrIwiO*ob0PL$9CrLjlc201<LoMmW_e)%Uk99{UWP5GE=gzH|%^;AFH;;t|4?G{_
z)nMMJ2kIIA^Nt#FSUKra{C1i+6^TeoZ;%vqm9V<EcTR?II~3zTMOok_aO<4N;QGTv
zd}2UCO5E)35+k3O%wgEG8QJf#y4?nNEU@O*N%~G6d&Iz!zn*+TqKcqO=~WCNrvtE{
zkL+`9U#tkUwEK(=LL=i1Fhb+cC0wgy!;9~!6n=+mss+Rn)+B5mdU<xXHtdMTwM0k~
zMyxRmajIh|p2HsOxK@TSLWf2@kK*5@*K9Fgt^SZhD$^+Y6Nwf22tym=o)XO>4^#D!
zb`Dp|zB1N&;6j*|XZ@RHC5>eXI=+D$+>UHZy32v!yasXZLa41TRTGx*bY`e%>a+h$
z&;cDfc67Z5RiGWUpM;de4A$X`5(Wx3lhxBZI3RqW?AwC>)i!hXiQAp*xV82yz~v?Z
zIyoP)C9>KxG3f9*XCKk-7Ex@|zMfI=^clIr1I+}G)oj`_V!kA}5%9zq*-gkr@Zabe
zrSWX;h9f0Gma6**-Wgh1v}ec-H2QUCD1AH8O+12@w8aj!wZYQIWB^l^E}@Td1)(2f
zu}1d2j%j)kzb||4#U#Hwo2hMx$2KK#^)dhlzmhhfm~R^ojTuA=d$#{Xq)>s)@(yh|
z(PfAWzmQf>5QBk6F}JPUntJXxh?n@D|C;gU)iawwm!CW^Jz56=3@gGpk|17RIVRZ`
z($Pp_<AiCVvt;i74j*T|O=Qvp-+Jk2cxI7$D~ZP0({A)rP{J{j&fu$Zs^x*qTfNz3
zLPu{24*JelpKmB$of*+5iuWq5#v*}Eli+gvT#8-$DWHouZxo5tr;?--GrxA`Ituu%
zBQNoqiI0e;%N*yYX01z-Q-wf&!2HD9gE!Y_!0>x?`g*kFIbDs>&qBZN*%eo^<ZM<i
z0|45h%wcJQ9LnG?Oc~1X=dAnRmh3$Obp6yef;@vA9nNKzYGd9%KIKTnZ=B^WC=V&F
zD^gP!mAa=rBJ_IGJvL0Y;&dYu*%#>nXtF(So|SOOMo#>&{UW|3_v=m}1fvU79DGbq
z00stm*~+BGDUN8jDPzKSBHWW^D&d^xQjp~jmuj~#V07-%t-wXSYxmtx1gYMF+V#~-
z2u->kxzZ8m)eQPvv>faefs7rWVcCuOz}IWsT-G4YqsPaMhL!}#jTz1#Ix2gMljxMb
z$n9lAe${C)-2^u*=xIO1s=rX*N0?M3){c14?v&bOc&ktG;3JEPUQne91>qgYu%ps@
ziXSrFBa6=Tl+`#xg6#uw6;qe&W~yg1?fC2nXQ!(iV8y2aGMEY)SwhH-H3bGWT%7%9
z{F41S+l}yR;2h3QhH_~6zZ9c}ywbzW4uxhv{bm7@$sbniG2|~3UzReim4LILA`J3V
zlkU<H+KWc9?y?Xv4R$p$=MuCatrnTm1O^0ZDuKKeO6FamY&gQ9qVs62y56N*QU17U
zR9UT(Jo#d<nMa1I_LSPmu7ir?fPFk>!Q21scG}#gSo@nAI5K%UROEVM$l}WCQ`<2Y
zGjzeM>B5lgLkU)l)&vHQvD||tZ47uE__I+{Z^Y6=L`F&yaCfU&hGcd5Kcs9^&Pemx
zFV`w|r99#o|1IuG<t5E9#sXiQIqee@e}U7zROx&^DOY!E<8<6w*7@cMS*B47ZG=ZU
zviHtiPv6JUhzem^SzF!6rw)7{jXCK%dm9uh=rjGOdLN}iYfI^A9Nr^~q}Dd*^5<i`
zkatjd`j|G3QuNYLgrP*y)S@B#i_+w`BruMrda(9##@3g^*9pZsQWlE{FlUaaj?T*^
zbQKY~(!3_Eq#>3{==S{14tYGIKHE2G-Se=K+uf6R-t*jb<_Onq0m`z%GSwNOi7>KV
z@q~R__bRB!nBj5mUhEq~rth(_l}A7e?+l7`pUl&usH`;{if>%dXFN5+!uygamNkxO
zDThwvhMhx1-sw{*Mv&Z5b&NrZgY8F;S)c%-D6chj&5}AjDy#%aRQ&S{-6H+WtE4Zb
zGQS=At)(XxDGLYtK9y@E8cYIJBc>Gnh-1Nr->O<ja!}HQn@T7Wmp}xFc}yu9-o&!x
zq?(VBx|AjgKGVRxyghG;fcPUWFgXoRz5UJiyf0-nEZ;O8)X9wE?v|^#Y#+x|jj7<T
zx#t11Fm84l9_q7EpT4J?vZZPe`SI8T=yGlg*`K88O?H-gx!T#^Yj5QJq;3V~Y?V8X
z^qW7}zS7n`ABySaBKe8`Vr+z&;Ds*3086TT@JhSGf`=nVYqq!?Y7G86SWKN3q>e><
zc=Sv}lOEUVecOnBO9orY9@Z16*|U~bhFkEOem@=Dd|@0jg$Kg9^c8Z+^N!L<%fPxb
zqno)zc>x00plJl(n&s}z{OS4qfs2|_g+c%hn^W#Thc3-S4O`%!-j(;9CY^sSKT;YV
z7!}ign2;1KS20!29qtq?&z$0+)=n4tu}wtAJ2&!O_WoOTLSf?<Hba+S|Hj^!i-`6N
z(RiGe9Lv;ubCd4RwN4=$yjnkU)tdKC^4uO!QzlA5!xbGn_deq^qV=Frt+EqOTYv%M
zRob`lD1h53RB=WtjVB7I_(ZnL413I1SNnjGaEx^vDjNz!lKZlJa@%O?L~v?e*HvzV
zR3ngD!2i9!x!LF{aP1NY!>y;uW$XyDK|t<X!!%o42BJfLB!LqT5gquQ+@#(@oOoAh
z(5%)WM4jtZw3<Fz8xba{FDOxHA+TP`an!y&kJY~VOYNf@o~>Ts3f0YSz_zEk^g4u{
zs#{+y2o&m2*QO^9oZd^peCcB&z=2~K%O>4?@;#MEA*dctXBAaD(wb7fU>?#Op2r;v
zi`Rk;wu8yBCD4YbHeN+w3&y0yNW6T4QNoXd!&pUyr6D{|+hun>T(z9E3lw=nfI|ow
zwKBcq!<-Z9fL~zpT~~`E^U$Ly%<+6{SDZEVF5TfreD<t2DrNT^AAm|)47(u}NB=zk
ztzq-fjWVU<cHX^km#szxPt_N9b+xoVaB`?lY#gl)b%B3~HjUsI(#ClafXI>DeS2_i
z41}#zNlh(j^8D+aXaGW(Q}@T7UPl&9nq;#>jayQ#ULP18S>@zG*ON4O?XGT=C}T8H
z4$aF^;Xou40_qRqKr&GkjdVk^%?#~xP$LR@j=z{VPzUm^`?95Kafh4@6AYZwmt-s}
z$0{ws`ru|aUrz&cA_Y@0;0v|9Nl1ztLuJxTF0A1%9Om%2h(0VjJ{9)&kiiXwp{3w6
zyXA(KgJ*CY3Dv%E2-ziyU;MQN4{Dvh<(B+Xugg8Xrz_=sP$Wbm+8X!#%!D=H1aKth
z4sPNlix~ycx1Wu?fayP_8$HiYuLnOEt#v7*w*T+8AVQ@okm5NXlrF&$W`Ds>wcQ~v
z!JH2%kmk~`O4;s`Mv2-tQmzwEKUgHUm^7@<$VH0jO&qI>D>0DeZJjRDUr&tPQj--c
zw|$+`JkJmpO!4%<(bApc%i_4)!%N3Ec*b^Yqz)Qv2^O6Vb;*|DG=lO>+Hc$e(T_ZI
zO(4}msq9lj`JGGWR#aWuJCdeg&0`QssIi5>7{l#Qk7td+>Vu?J$@i-q%jqqHZBQue
zRnwWpneM{+$Yr5ug>A7wi-ceuC!jev4(mk?7YkBPlVY&a(?UvA?IA)P;Pn6AbMtCF
zWHGq)yD<CKF~Bm^MvPOJOD`AxGf&Nqj51H~iu)G<S)#!o$aC{tWF<~g36ZG*OnS+P
z#h0|%X52=AQ%Y&ZRxAvUJAF^KjF@B9dUkdM%4@NR-y%Zhxg$@9#rkQAdF6R-gvf4M
zPeuQBvef$As|;Jw5w4vUron?0(~p8lRtcYXPv^Kz9!*q;yE!f*2N;RYQ|{fu>pz7@
z=%@Gl5Qf>l%pv8b_4<y=H~s|8j3do_imA^_rweRckXbNYgawve(dc*DNWD#)&5n@;
zh-7%hyF3Z(kFS4}(%e_2pk-01ny^-u2Oc??@@a?5?kaGc@emB6yr)V@p{=%$O&w|e
zfP~LXcFkHS(`6++{nIib3y_;QWk1fv#$}aNiO_HDfQxBfR{0n{%pbm*S3MBdDj|iL
zHhv6Kj2voS-}}O~A_(MRg{CV}nZWE}gvB;i5d=QaIrz>((Ml>ay*-1rIVPb@BYX1o
zKB_>!Lct~bYC2m&*MbgZ5#;c_uWC?4qjKb<yyu@9U3kgQ*c(cHX!~rvJTu&3_2oYa
zy3lQ+Gr0KZZA==fNz!pgFG!JB$5Fk&pSi%L1aD}MDmUDYoX0gbOu=NECo|1b^pU96
zu5$h#!gNx{U4N!%+;Dqhwog2evu7=<ik-A~^wy<N8p&1e8`a;u%KCL2$9P5fIpY{<
zjNXN7_Gye`4P!uEBC<-J)26ZVVZ2>swiix&*9;c{YecZg#L;f0W3FJ9=j?M26k2AH
z2Lxf}Mq{p$%W-^2v4jezwt{zd>*Agm2m$(dsQX5=6AAM^W6zQ#V;=9LI;Da_&nvA~
zzk1_#e?XWyAdLA^ctZSegE*$X7?d1T@{iyn9RTodZW+8F>#sCv`TM2H6eti<@YfP_
z)l{zF(|EyP+jq`gN)9_UF4RCq$%<*@@6u<X6M61QZG?FcC{4JM6<ArGnHcld%H@bO
z=&th?@E#gh`C`g6rp2kv&+(HuVU|_S?|cX`G3oiMqDoe6y5RHR%L=U_YF+w7UeoQ%
zw1<z7F(s5PqFYq;2L^#d5j#dW>a`u=ijreu`NFtd_OBdeBuieQ8WUTCiKZh#CsD#h
z7DBZk!W&6%Lq}V+_C3T<-cR~nQrNL38}7BxYotz~W+KnwQ$SPDnO}nRYLD&cC!P;&
zwZ$ImkrqQjZ(cAg<}3}BLi|SBE`FyET1&I@Bo_^TLN89*_ha8z-S2Z^CyBY|_*2t5
z9XV}B$2v@N?-qX_sUQt=M)sz8M@pjo!NYH^7K4z#CW_%SUhF0Jo1WjVp}4rFR_e%w
zxR>393GI5U<0HM>&ptbhzgBNKv=R7d?k92zXo1#2cWAhpj6VMyuqnfyCWU6$V1OlF
z`}F=Ww%CBo-iBkX$f!u8P8Q(iJf*Nqz~q(2`ZYhmq9}vQn-mRvI8IxU=kL@X!3PC+
zJyD@#pg2wItF)+ucQP%P;aDUxDkqUO_1({6Vr=Xt%*8u}Ht_uDC;?@v%%j#K8Fd)g
zxw5TQ6Hh62408*x!^Q<O?sO~*0+{F>FgTrR{?m9-c145^m@f*jO_v(g<P)w4^M@6S
z+PVq~kBlIS5IZ{{!&k5NI~4_@+(ZFnLAuEU+n%bXC1k=VDIi#_&Z#dEEl>vkARMbv
zCM5z*X@ZoR|A|n{9lx-kb3}8gRc=H{@?)qv=xYNACKDmV=7S(r%^>wOy|j~efbl&W
zEwV}86l0N<gCRTarv$1@E_CqndCH2VQ;;1V8Dze6SHe4G8^w5P`4v0RWt_?C9N|4#
z^i%6s&JHl4in9rxQ-QTitefJegkwF(pmj*E7&)~)$E)UHdG%S50HBnSU4?_6qjJ<x
zLRK(y>Ky!L+>vA=;I!e6Fdc?b5vx5ogyl-yRD}Tz`%@iG91De|c$;_%F8<R9#YCA|
zzKA?_8MIOH*kY$c@U^0>#Afb9Xci8~`7yT8S3JuBnw=DHb+m6X?;f`=xb%5(^7^8Y
z^@mf6$|O}c0;2-x9P_WfRJd@tTjDA4YkJ!`u#kXV&{9Itl|DaEM?~6ug~=?f8%<5q
zHW$9JS<gRaDN)}HV)a-szlmTv1sW>PfG%=w-)H`hRUglN7TngAhOK_=*-jjU>^e4i
ztEg0>jbk5~a)Q+jC>RvtsTLFVY2h#MFr@m~N5jm5kdG#NP=jQn3>{qTuAU}-7;jJb
zA(P@?8yxr5?5JKbF2XUUii*f@@P>P30vI!#xWQR*j=WLH4Vx;pTCL?M!__=D`SC+X
zHpmNe?!c%X5Sl)Ce@NZ4XDy*s$Zm@`5DR5nnz*(O0h1fiFqXQR<(jTbU<7e}+`olv
z^5usHM?d5SKcDhq%iRYBroBXG#P&qy_+r)lOcbx%nA3gHNKCz7-Gr6RiW>goopS>a
z-JU+`M-;#_XpqmPZd}f{l-}$617`%>$~<@F?eYY!88^FlvQ<_3{Y)?|8q^gR90_2&
z@Xe0=@kN<Ta!sYB5^>ZilBWvd6}JVMEX1b-wj(7~X8-(@fGSeFjKfjwuTbKn-@Y4!
z>NpoUa1(EfOfUwW6=6UG3#`OrO%@oyDN9TFO*X9iHVF+xL0@ki)lgb_IJTY7HE(>?
zP(bdjxtxaUg_pN|3d)r~jynLyd0M#UjPm0OidQtUE{kQU1HxL7$P_N6Hk$Gxf{8^}
z+)NgE>rnfp2>UV~mGv|Zv^7ejb>Ugaw;~KGL6g}X)pxBx=0i75btw*Y;ullak%SNl
z1~7G7a<;)?m1BbK^eb9($MnFd<7ayg$Xe7ZH>gM>m}#wM*D+%Bi(Vm<v5JYn>1Riw
z1O*lR>*=4a`+GS*#e_NBHSp*}V-lY7pE|jUJ1u;YQzkSbxK9VkAIvQDx7IAVH?rEJ
ztZw!Jv+Q3+e#Espd%quy6%*PId5R%yn{7s+^JecX$Pp<=(dyEF#J-FD)je&<&I>kT
zROsns%e%(agpQ_cZuO@kZ9Iiw_Z+9kxeZJmxW;@MYxaxn{TkV%{?!Fyzs+pZX1}bU
zO5ygg;7?L;4NA&O6dzYitG^ElWd3Q@`ueQd;qy%5O<*!0SD5}upFl<`l|abB>fBDs
zom*y<tWtUOl19SAB#6f^|F46H!_auL7|NM6&?-5T{ygvbkGuyt%nl<IE7Bc|ZLkK6
z!$xGVQSFcqVW3tq!0drZOeId(9k(`1xIric3FGk|s=QB(G8V1t-M7e(OP?9AesVt`
zM9}i6s$aNm-<x5wBEK@ccB~&V3-@@m_rNfSG)b$ULV%w6Dk};p8y6FunL?1Z!MsdW
zn{XoC`X=9Npy1aJcsw0OWpLVw1fA_G>3n(8T8o{LCcw_vl5iZwQaiyLrrBA~{sM+d
z*Q^;lg0VrMO=8uTZt3|w==UO?flJ@DH6J*YQ=JFe*V<z{mI=Qcl>G9{J%bGY>FG3F
z<$5G^D2Q~{Rd18smVmn`(9YZ3Y$i9xyKKUZl7TdAaC-|IRd!C*ZvbgjER^gRM<&%Y
zfdtm;YI&MjFq=-35V^qY8WuWu5%+4z^FWqbhv6AtDwDfywh{>3`WOcD2tz727}g`6
zKW)C-JP#kuuEgGYBbuc$2_WLdCcME&--_OERVmtr!IdcFb5*iGhA9N(9GYbJeVG-*
zGiVhx46Zb|q%$Z4Ox@){Bd|Qsbh_ON=*Bk>yl}5=OF@g04*3liJ;cVr0U4hWz*YL=
zT7D6ycpqHRQO|v3BTyL-iBU)1VpAig<$~SP=IzRcq$y&4#Vw|mHQ<B=dCVD_=7+aB
zq}PZi8+W{WHFh@egbJ^eEeLn%?4I?-zPs=MPtJS@<9$5*u^%7Cl}tC1>52=~s~N5!
zHcoRigAfZMr4W`VFPoSzM90Kves>+&D<M0oc=Y|_xcJUq_x~&KF!9Lf+8<8n;E+j`
zK9S|L^+@ohL*n&Qn(c6MryZ_RDp4f!mbkxW50{+CYKlb~?M!$;w-Gzgpx?(@1x#Hw
zwMSn6l+y>C7w9L{4C#-9&SpOHOq{W*R5wJkfBYT6!>R7A7?Mr_WfMYCCJck|!}^Zh
z6rQd9VS5Erlaxs|R~>z)x#8QN@4Ye>JnNU05N#a4JwIBD^PP8+{IZDLF8!LkqtKB@
zwLan#pE|8Hf(|C=CXCNxzu-}@94)TJhP(T&p&C_sRPfw5;a$a3$N6i^U!F9>QilwX
zXO8Khb^%4Xgs`>**V464Yib~<;~~*OUKvHr0Q!fJuOsj9Xb!eXeauyMJBaG^@mo=1
z)OM^A#MK+G<=(%}lQ+$p059UgKU%)@Bt1`jLmMc@MD!}zg4ZyOdl^NHs{O@J{H{-b
zXHsyV!*|9JNQ%^rT4u=uHm@JTI8buQ1a<44;kU*~YwB)PGr#`?HxckGmE4GUATd(Y
zS>xJ>XhO044P`)@s^LP5Ft}(zR2V6K_tA4}00IwT4Nr-Gf&UPL9T@mo8AvhakusW`
zSdAVwqVk%f`bylAts?hsynHQj`%YL@uHU}VGO58MUrDHw;_;b9eM-j)Iq-)g#bQ~w
zr@7+D>bOe_8K)&}bvlC`Wbnd~coJd#f!3x@>}TSiM9rp;^8r3EdNk8kfb0W>0w>$?
zP!n!7CJOon$U=lc$oe1^XZN+cP6k{32VsziS>6c;9Zq^A@-HQMoAD3w6i)n#QG0YP
zJ?LL7Sc)jM`{=r~HGFBn+DRS*ZXEI|iiD)iq?!?Ye$3b)Ytp-lG2AI`*W%z5G%MH*
zRAOj`pOdCXjv#@G6q%d&l`H!Xqi$#Xkn3PlnK>W!GZ0#Otac~$PnI;L@%AFtql!zG
zHPasfW#`t{lNh1bK?`K#E3MzW`1ONKT17>Vt{l3P+*PTVwW*Ipyik}$wRCew=lsY?
z6=G>c)NNf3<ZM`Nw&xtg6hfu~W}@al&Y5hPZ6yas`aCPI#$~%8UvYnMFP<D-ohBEB
zPPR);I{=n=`roIEC(HZMOcfWaxtQag*?Djtn4a4)Ru{OB`uzU0Bu4mqwF3T!s~W<p
zJDyGlG0sUJ&9W`RqLYm(8Rc^Gx=8ECI_S1kjDWD{L^Sb2x=Emna?hT{fw38$GMe8@
z9q-xXUoP7mdrVPJ`2;ThxVZlnblG@mSl9-PbzIa=`x9Tp!y(=!XB>c+*di`?y1Cot
za#giw{+QrI^Kps*1#Rs&Mh1$wIasghUf|o$v7)5E#eg{^?gtPp*8v6D-j2kAjZXBH
zF0mm`S2km(O`-X#=C|n5o>-P4V74Na%*P{a2~-CBA9w8E%p@blkQkna0%IQ2J#Tbl
zg(Ot^A*S8{a5pjnN0i05-iy*@jy%>mCyEJh78TmpIzNNmjT1j<?mSM{QO};CMj9k&
z(1G#Lekl^unbuE0azk57`@qBwj?BX(bIsE93KyZVp|~s({5D!yvvF)0UwkiD!Tuz{
z*kEP|jPIBh`)NU5Y%f=Bl+dZ1uJGzW87cD9$8cxy3c)t1;>;E?>`SWA&`}!V(w5FQ
zYjSYZ8eKG}3c3%=kw}V^co!*b?1HZ<aft81Mknjf<qJM+KU+VAt@$+svl<_a$_iU@
zJ=ui6vs4|vOqr9&P8z~PI`vPc%6r`NGd&HvbfNC;_ydy{IJdH5F9H;5HX<&YjV`@|
zb=5*GtCQ1cWR@n4ZYjk4K#Sf%wlbHPlaB6{)P+Z^DND}z<kKyk#&WhTm;g!PGf9NP
z^ZVLKH+=)Y-(Cu43?Ri5zrS>n{{nWfhKA~mL!F9=+vcVpEQzvEx^hemVDXpg*LLwn
zU&-aw+%>rH$iHXY;5w5p2_=k~`k*pVqriO$tc3G?F=}97m(r&LaD{)P<Ph;;g+BD*
z<8S%0J{EUVB8-wJgnlf^aC;KyXq-cRmvuUGasBNwO%73eCuDTWgxZ%gSw-uKWr;Z9
z8%Y(>9W@5~AMz?T2N8@ykcE!cSj*36kxW6QJ@g;m=ozxwul5$)E<X{kc^suh*t9&M
zRaVMEV6i~_L5l%iA8<_rR)eGrmJo;Ux-7f+=_P|1h}S7<;??F|fngi6b;HgVIhOdp
zOT=rhzLFSl+Pf)=-<=Ev{3?H9rkee<x3FMqiJxY<^RR#hn3d2fs4Q(CiZ`SrXqv!-
z@WvR%;6beuf&apgCa$jrilfDlS7Ib*Jf8$q4b^_iI1TgYzA6l4Nj$TygzK8#9QOE~
zmBhT!ima@+S+U)N{02PRzc!|k%+>#Hkz(;k$!^K~*;B{AL;sg*woSt6!F_k*`H!c~
z2@j=;^?DD!i6oHbSa1;(e;sW_@ZcQ;e0=-ViOXrXg5$YbMZ6+;+Y(3Irtyu*s@<j>
z7&Z)?Bm2*DBb*W9ZhVTq(U16k-BY4JAV)hhB=nAXVCoSn?<G3>Cfx&O=blsfA8fJj
zSb&8W_JG6$z9(YzlQki<n_tOLtSUc9s~oR-oUT<?``$*r-01n6d$mJ|`G~zqAOeph
zYB>92zYeTzXnF@cxO(<es;CM|K>e?;=AihgbzdP>YMF*&?d22#thk=0P65=rt+e)Z
z)^_U<$t1%8JN+0W{hOC*6l&+6v@SmERN70ed>5|8D0rx+*L;es=NRz6EX=a3S<}^J
z$V02WALfp|zm?1?o8hunaC)U)tGpAPSg*8g)~cnaH;S1Y?fnHtETFMuwruh(`L{Un
zuYX9%Q1(OMBQB`Yexh;kPU1EJJ<CCwNO7lfvR&-aQ76PvS3@)lD?QrqHthvx?#~GJ
zl=^au0{yH+MU&?x;&Bi^l~_FP)-~75cKlzhAM#TOy@4W+{u?tz%X<K>_=UL74N#J_
zL6djHG(6Lz9gI-X++30+E@R+<Qx3*5Mqd@2KH6OE$~=RsW2y>{-?~P!zh`Lfc}(8K
zyEaX|D>D+4gqICR<!}<RK~?m#{HYQ?KJsF(M~QsrS_ENWtw&PDcQ=JLn!51BFPW!D
zIJzhWWw+rdn0|-CuTT(-b2R|Na6PQ-(USN377M>~aI-VoJoo?7q$Jr5wm*?8*vEP@
zo)_!kH`26Jb+r@be+5XQ8>4k|B$cQXk65VaK8>$hPhU9Zs~pc?&IKIAIDx+D_0?sx
z)d_x6JgD|py?E@7W|y&NZPY5CtENxD+|IG`3wX#{GT@FPU(dAWN8mQ6c;*3W1c{@x
z<5R3Mbx`PLKCaq4>|3dEIXp7q6cFZ=pNScBv1xp>HEf7^B!YpAJD1hNVMwtn8rBUO
zGcb}L0j%pT{&2S6(R780%i~Z2Q}IltDTK?RU*C|t^#!!<WLn$-Tp>6n=dthwvbNoZ
zv0@p>YWo~gM)g-(nDn}G|9q3F-}f_=LGtU=m@_y_W=Po&p(9lzRs}z6A$VI)laD1A
z?pz<N7f<cQOScdDWw@}u*g47O?6f^O`UnT3lbEmhSjZwNE5z~K%~l!)p;Fxx{zlos
zpQD^V>t1$fF|Ryp0Oaf+S3DPsA0LsRFy7RN-)1{zlI-c1>CjTpX7X``z8phy4cA9M
z&GY_}+?gHJFz!dutwpix_C4Ds=T{VG4OMLSA<bz7ZhYK+%>mo)^R~u@+SIGN7PSl2
zZw{qg&ZRU?qBs*4+x=8wO|k<WNEwl5|LBWVAER0}Hr0LBHd%VFEnjdxpPYsi*5$E>
zR+s)uxf;#U*oeq`?VN@RbwDjbZk77SPMSu$Q~sP@%8!QrKBwh*za<UIFr~;)tetkM
zH&`U*Q+XooPv44L>ErO95+hV`tIqx<He;*%5H326!heaD5$bML2<no*GIg6{fN9D-
z%4*q!kpWDhxd@}ZNJ)M58P&ZjfFwU03s>ZBw*m`6;#!=hmx475rn(EP;x|gJ=H<ZC
zkK6GUcg%f1iml^A;+maSpuSHKgb63BQNc{d`!Vm{=W_iVp|L7yjF<O(Q^NEyAakS4
zrt$_*PzKXn@z^<SvUmIE9W4c_r#sK9*~oGfao})b$SZ!;-Qp=vhjxMQLG2jCVPIPV
zAzLjIw$Z4uzgYlyJoM~M!vw}|CvqmBAz;tikT%}$z$ds|b_WB%!;jTE(b~BlJxK}n
zUw7G-UE|j@-JxRn3g{<gs9oO!7Fv!*;((<0ScqM56(PQcU}7R@X}i!e!d1-7qf#Bb
z>ksWUsOasvgmg?_L;l>im)FD?>k0klNL7Ya*3xS5A+JV_vaw={Elw)DV89LujqQnf
zoFe$XIGUOq2au=5r}&Q$hov(c?T(ecoX&)lG}m2{)>u6<N^KT!4_^IM;yvPB;i!;>
zU0v6l2nSeI=TjUj99AEQy_aX?DVKl0&D(Gt!rm&0=frfB_zehV1!8(gm#B5trLl!p
z@bsJ4juD0rw$boSB8cwcwQ8DolDMVO^LFr`yFDov_c*6K1ehi?D4ljL<_(D##Vx)Q
zr3h?Iuvdv(I|>4N0@o&ih&HJ#ISg6=ls1jzrIe#&A0W~d9J#Gly*Q8qZ<765ingm+
zpgLr`pw5oHlegB1Dh0vO?2WwV7I!gbA9l$2N8_t5Y!;q47LS@9*;}~xnTjIhnpx~J
z@`TK*>-_-O<LFA+HYe)cai3Vs0#)NAp6!MsNz&ak+DfIj$HC~_&)&OhDxa_TfcAZe
z79M~nvv5Q0M>#3Ac#&Htnu4(j&4)^A6*x*@2B*^nE!P^fsM;WH6ESJuD%vS4tgyOA
zatB;Gu7X9@a;p?vE2tZTT0r&fZ^T70yh3l_Z4E(_=O_gRF1;7ZEwj2E_Tbyqw}j={
zHNKC!KIx6Yt)0Bp{WSRfI7^zHZq)(?*OL<qB}E!^W|(%A<~(1uFA^iqpIhxz6EmZF
z61MlOIkrryN|vg0Q045)JH0`NhxR1~`;H`#nFI8H!R+Do{{5=+A>iqWWJqOQW3~e_
zN~rWpXX<P%w-G#wUp-XFO&ZEEJ2)0aQ3Gbg%m~l#jDfDhe>@T~!b0_qa+?Z}slzaM
z5{&Qcpp9e}Ob}TK{3K~8DmSl1%%JqAVGOUF8?}Lf7&=2XeYO$N%j5tGs(?@T@qcLn
z{=)ijU*jthYQXEQ;`8-&i#ePhMT(5NtjKfQ{*cbynNYv1%$pc8oXMy-usKM<061$_
z$Pz^ElxZu0fI+t?t0ncQF<h1u7=o${gCsQnMh6UvxXQyYe4L!D?aabSgR<!vXqfov
ztWGjLzyB1<Ze_;BtN(57FTbAu^L`f+xoj4J?Rb}^D7SnivKIi;q>Yp*vmhV?<EFjY
z--Q#C(DIMU72t2fk2_A^Ig``I{si!oRIdo^ki`_kV4ENq3mq3ck%%_X{>i~PpRI#d
zu5eTp8mn#rj9onjZzV}|bRvnv8Px&hYLH!O%S4DKKeE*mJpD$GM)EXP1Jw`t=dk}c
zYbB{UQ8T$kbD&uX6#<HuT^3T9`S(xR#>w_<lJvD&A!v`s<J^nZY~(21EVpaiambX`
z*F-NK!>QbrO?nf~(J^rHU}V77YMm^_w~gO<x>2_}MW>)652Mr$6~{1%9aS+d@9`CU
z4XoYSxwK%9+T182^qm)=7Qa`v-Gxc%%#lhl&w!2Gt9&XG6BdcuM3Tzp#BV+wE(`jO
zcTD>!;h8i%B7w8P2-%&%J(E9_@1C@b(2{8fuB&HdS6n9(UY=L2G5R*Wu-T_`odFvU
zM!TM}9!$rjaw{<*v4eRZBkoOhYBzKCHF|UWY}77B5O2xOXjJ<~Dkf#&Kzu9&+?E_D
zIZL%OJA0!6TEd%PmMaRH|5!|+#krs~9zpL!&YRo*>R0X?28Y8#JvGaaT(1%rrsy6e
z3G34+@eetYDJueSWU8mVvg@zw`bZ9ldg6aVrpDsU6#5@o)YBj9SILU+`<H^$JWB;9
z9*M+%f_`~@7vm_)Bn@Fc28@?FI97`O&}q8R&6NHSz;-&wE~9Tt7WEG?y0@Ko)7KHd
z4?-wP)RAiA^ZNgCp#2AP#BGje7CLR;2acd*^H|mYAZX6}10&lTgfI!SZk)2o!{FBo
z(umr0^5jiVt(^Uz?ZZTlvK(9KxI%ixT`a`1$zx0EPEZ!#^P;Jw^HdL&3<Xz6mHXRu
zB9zsadz=s_Zk_8Ln$l|w5Ah6Xlmn8{<aYKJ3?JY!USt3XBYo25XUi<lB2&3_ixL6?
zdIqBP4|0p&;<?t3&%AM)7ZIcYK=m(3Z!V3rBk_>+k}CexCZAuL09lM_MaS%$&<~k-
zn}KKxOtYQ^ZI$vu@5sB*D-Q1A7IH5T0Uar?dU5@ULMUlAv&b^`eN9^OxwDaicw`Dq
z&!yGsQ$qNH*CB~lFAFT>aElum>ugvX<i1d94hp?)R^D)=Pa=Bp*`lvcdb~tJ^V0Wm
znUl_jEyXC`&Ij&ddogsWL+=bEMFZ^)!j#Bk?Sga%Ogw5@9+KB@?M6qCmB^CS?@<oG
zLCd0<1uljS+TxU_>X?IZyu}4NnXI^@<9JBXDcQ@v5A$?1PC%Fsu<}tc-iY?sUr8(e
zC~cJEmgM=_;F)vx4mCNG+nO%afdecAgIAbQ_Z@>K)~5?}aC|}KPMdGoHIk+zbETMR
z&ZT_MN(wZC&qbCb2~))WOhFuIme1+_?5mvMy#vG^FW!SGrhW5uDh(QQiJ3Hs_aWDk
zEv7fNxn7<%OwA!gUL~f~(On7wp<m;I<=6MQW~%pJ(Bja`RsDgAJ{DyybJ<}n2$&5y
z$hFHH6}OZy@tuF4Y(A&6m)+g1wYn=4xC!lre3yNYUq3?+OMQq9K;@SCLniNy{XP<I
zr!uX&@phG@V>VfkD!N?!_bfiC1ELnT7;}5|p9hTpiO>Iyf(t_irK^c!`#R07S;wok
zza91<Ma9JZI|vWxLXoCJ0B%y^(SB%${@-6G#0@*(#mV&@4T#GD1M>?Npqtyjf90Po
z@K09wAB?sMkm^M&bNhNPaMgwoY@IWCci^d$6f5*LBYv#(A4Xb~Z0H*P5+=oT0wSC7
z+a^DRhK&AelbEUEXB}=g@;siG60ATlepI(ElmD-XW*<O}L5l!3jdAs@8zQrOz72>V
z>~Xo8N5SU6S*guwDsH2o^My<I0n=&}Z#nE}Y^mS~aLB|&jG|=;IOr__smi>Qlqkx-
zKWiE_vf?jCw_KFrD?r-LP}{a>`2DdfpmcBLu$T|}`{xp9P;|T~1%qR5GCw{7@EA$x
zhYxE1^~C(|JMzyf+Gr#~jbp#*fcJ>hf*O!ra0BG+*9ae`%LRZ%0bb=j0SwmVf7QkQ
zIRlt_Pg-oIadwiCkYRy6n~%9bxlZi{zI$RD%VT@P(}|Tce+L=T1Zk~}%$vdMy+WxT
zfC}<N48US<#@u22>rI8F&~rckXikqlCB<Egm;#S{hNF@2Kgst0+|K{L)dQaLn*b^W
z`|7KQ1}^Hf|6<E~fJL_E5ybn_7qH(R0&><bbV#Y+Uz3;!OWVJ+uP$ejBNk8md)-Jg
zqQ>3yIidpnYI}X_Y^&zXyRZMg<COSWaode<k@Hd#IEXt=DSf5!&rSII_y7CN{GVT`
z0f1RE9aJEM|7Y^@&&B`0|DqWs;vxx${wn;Z>x8H;Er7&-Z$@l|p;Lq6IokF0EmNTK
z_aez5NGOEG;!LvSXahSPRAEYq1NlDdb_b@x*SpMKlH#X-paT=<{rak}`Hs;mWm)~H
z{|}qE2m~xC;^J>#{`2Ruc0e!v56ZXh&ZcpV9Gjp^p7UzVe$Gq)VMx|bHIiTTA@!V6
z7-bjq6M(R6#WELwnDafgujPVmwCT91t^9$E`!!GnV;UxX*H|J|81iuau3qg6O|sO3
zZzmv<c~74#{tB<Q@jd3&n})z$Y2h9kN>GZ_|7OGg*KJ560^X|nT=<zKz&8EYcdj-<
zLOOv~_|RcZ!!KZGqavuJjpNU1B7s%X>6T<YT|fm0Y6e_^1r!a`Qmfnosb%<34>O{_
z#9DXW0_ZG#aN_{V6Vw<bpq8<uAH3KvAd5Ous_Fv3Y#IS#S2Up0S#ZAuut#g+H|y>b
z9Oi?MNhr;onKc#wMPL8WEg<Z31?GvDHV_g%fJ)pLMpm^-f}5~J^Zykw|5+ktc?d?a
zFiB|0`rlitDt_RgvXDCX_Yh4K!ok&<=wTgIHWcFlG0=gawP85aHpy?N=L2YLc_hRF
z83GOi<RAC&`Q?h#{e1Inqq_Q!KnVgUT!3G|Q2)b4!wJwzcB+*++MvKk+19E6NPDJ3
zP!i(>I;5#wA+9{2v$nMrO@s4<qje_6U9_v@we2lHeqy^j&<No*Mn~D?zH!a(^L3}&
zuJ-!7Df*vF<$pe9*g+gnnW}Uc3+VfInUqNa<rv@<4owPeP@aIc&EfPX0CkKBF|CRN
zh)qnE>}%@CZ2*YgO^RBQOY5w>IY3c!p)6FG!3q=I-SR3S`~J{Q&%Frxl+E`M6pGg+
zp^A;lB%2M;sS2?X=@lW8%3Gq<KYa&kF641Y`juRq*R`g9nQj07Z9g+vgrFDWIb{Lq
z-+R~07Zs7{p8#ZLglt1ju(<H<q9KGz52&Inf<g&JTM9ah@lHf}h*Te7e}ys~a$qrD
za;-l@9eCim{^tZ^<_<pDjKGZocx<%n?MtuBQ=r7zdD$nr8cL|Ydd}1*z5qvaDaNzC
z?0*Q7lwnOW172}WgUkKA0*Fwsz@jky`d{&KfL{h_$Ql0o3aCPgOXG<GycSH4KXaLV
z{Ng}hXY0xAjmc~(1dh7a+z-<s4;E<KA4p8aYui`s%>fzPq~yxVb#lnv_6hL4`EM2h
zy#S;~<PKQ3xMjPzA;0C!Zr>q@fm;aDNEVNUJOCpy(foY`ta<*IfLTsh-6ja&<(T>c
zYlfx(o?f}M=m*f~G#*7fDlmE#%lQ5W0AOqcKpvOFAM|R(eE_0g)a`NqM$b=nU{hs;
zxDN1baBl&B+v7Jt0CYHY1*kGo65>}~)7XI0$I*9r!i4xlTrLp7-w~VDDoAI$wt2D%
z40wqn@N4<E2bNo}NpKo`Kd`YU9~eF;1`l#l{1yy(E}I`{x%~&&lT?aN^kE&$1-@YM
z+vRB-6|P=SUqC8H4K#3UtY}<z?@|<~o4WBVeh#CX1Pak0){Q}s#uptb)&AADFT5r!
zN)}>G@4k_QbxZ39U3?w_;vEgbDDedF=a@Y-cs7q9@ogcVk>L#|ioHEsQ4~Lo+ndUq
zxY`!9w-~wnV#o51L;l?6dnbq&sjlESwVk~^uvb~P>W{B2DFA4OAs7IYN=mh!XDj+^
zG*$#VJEqfTr*C}#Z=_a|mQU%CSE!@~fQM@HXn$YM-hw>mze1fkrgZpR5a4A{F-gt0
z?PeHp_UuEsl%?J!jot_R>SDeFC}b`E&}+%dC5y%H)kX~fJocTX)0c^C8ZUT?$zHDS
zw@I{{Lf|yHdQ<%K;-_Cq&$kEO33kHfG>hf#wDswU8qZ$me8uAqH`G51$Kcy*HB)sO
zb-yA9+RWRrC!P@iXm|Dy`Qi{DQrSy0BU9YWHh1T%g``Q=I%iw=Z?xBvv7yi|>Fj}8
zt<Fi#iBHPb>$<+@=w%>C__cW{Qt2F`D%pKYQIesYDLYjDhk)$O?q*n-t<KU2j0;9u
zJc`%JZfA7R4h;V5d1uy47h3NLBt4P*Q?Gg?FpYznw|L*4uLBi+s$ATSC7{}X(jFVU
zHK)Rby;m3#y!0xb<9$W0^UCpT-_?xyoPM_3qYbEw#n))yOx!ozNkj43*k{dZ`_V}9
z@ZfKnHJa6==!4e-05ZVtvIs#TU9r3yXKUaWBPAR8rDeI1D`=)#lwbnfe(Rx8CX9f`
zf-4nar|$uC5Mt6jLQ(+J{-(i^{HV;2CP38UH&J735gT`si<sF}=dbP`qO;B0lSO^i
zXD39G*9fBB8Rjy33E>QGfGv8M=#7Dx>7_ma%Dh&kqTf|1qWI5@qzC;}=<|0Z9l=LN
z^HiX8izM5iqv9Uy8$_cZs@O-+GJIE8*Y$jp<4IXKdroBew7Hv<&0sioi5OQv4Vt+J
zAmVi>1-x6HlS+6Gc-l3PMP)l$$FBf$W84Tx3pQ2-ZK60pxU#lol7Pi4e(e5Hzsh&|
z2Y^YkA_@&*7}i(|Fl_E%ImK%P_;ar|a&LXU;Ig%G&3pAVV1v>-5jxc&39u}4igT5m
znFM2^TT}FydVH45(N!#@K!2V3EM(2XHiu~*Iejy0z5)1Ya-BdZsJ)|=Ejua34Pq(t
z`97IzGg1kD|0L}v0x0FwEY_PV)qG@`)3ZlyM|*?-%AF-=Jk;kV6VZ_t0EK7dQs$)y
z=8*JV*(V&9tN1Q0lY;Vh%Ezs4!=Q8E>duT8{H9e61~}zBr8<(qqb;qB;|09o{|{B?
z;ZODdzklP{^F&70afl=P5ZUus*&;F`dmP?Ma%|cA$R_h3WTzvt701ZVI-zuAL~@Km
z^m}=KZr{)M`}+@Y^Ljp?kLz(=_p6S4A08{WebxV?czU9_a2oI&{RgO&ur2%qSSI@G
z+g7e!(hsj0&Wb18!qKG|Tt0~nhOL-G05yJA{k*FF3WP!<CJb;Ha{=E}dT-9R)-a)X
zAh#Uyvh5fsYG3l&u*{sP>b$+$y_O{E6XLg|=o%@ufosdV<1A|E5&<5rt?ys}9|Cu2
zGyFZS=#*hv^P0!^);{2w1HX<>yJY%)cqH!1aFVW=!*B_>%~N{PU|&9lwg-Mx=i`iV
zg~u%U20a>Gn&W7;HMjEEVG}}_nk4M;LB<u0!pmlg`iali!WcO09u8l(KL=1rJQ;TZ
zgl2jMKT3Fb2>1;ZfmfBpF~@qMSm}AiMS|Qx?cJ{0+YbBxWGXuGd4J*a)IQF9Lj7P9
zZ1~$;Vm6+k;w9t>@Zhl)sAUU#)>%#mafkc`qo)p6nHerQ6falLM_N7ktlA7M!gct)
z2a`ys^V!bnpZP)o>W0OO8Vrz>Kn0E3u3X9MMfe%+rpYNj`ev9(PD7Q3+$5ikz2y1w
zEXM_?2kLZ7Eo!ay^*;@_C7B1le1|3UO%aLzj5bo7;8MS3KvPEg441#oG*MkVJsQr-
z9Y(wDaBku@H5YC|6zQSzW5p`?0Q14tOT13K&q~;@Ww(h-=CS+NphM$;&_T}4fCcj=
z@hpmf`DBR>7{JL~rJm&s=xya-3QM~<y9D>?v61Ig4mYN&k<Ic1===C&iZ?9;Wm{TF
zcfB%LQv36T2@jPXToDUDCFb_ntv`+k)TR#Iapt4Dbdw7Q4ypo=LEBrOBgqx#w~M7D
zg8!ba`q8o4R3v;jXc}OA$+cV&lD>Z}5X`pAzjuOB>n~mn?2Q{}Ytmm+C*gVr>z)bh
z(B)a3(&a-Q+jf)Z^MxooA47$y0fmcKX8v8d-yreu?*awnp{9qcSXvjTU5-bK=iEne
zO?E`t#I5K-)|;SDY3g7=Clb?q-}v8Am4%g*(vO)}s^Ota6CG&-tZ9Vxp}kozN3s)}
z7FA|7aEv$CZ^nfV&_;AtN^OlA=%uh3Hej#non2O`$fQW2dMi{Zk<{QqpO!@MfE4Rj
z>6EDT%g5pYtDO5dx4>7|nJA4vz(O~)SAc#{fzp|o9#W<58)FA*r~69I^2X3wT_P$p
zi%if!2;Cm{qGz4pPixql^YMFUJp1!Y+Z|>1R@V-=(TX0s{|{vj2Mzy@hK|#9LC@1K
zOc5TGp*RMhrv<t3STBqa-FE`;3^@gq*cht#57hq|4QATWCK71k(l5EcMWf7IYCofx
zVlZinwU6po^)o}6C9UB$QBHiH=I?oniz=mnIMnt9K{WJm0zj7bvE}`B+ax>0F$fSq
z^D<}}A2OHz%rSMLE~o$L)2mIRlLA~2uV3=KHfz)iStjPe9zUhwURnjl7>}cCl7=|e
zrK?ERXy%8Yy)b6YJk}AaB``eZnA#siyt13v@$`9_$^-7oG^qQ^pKtQ<H|MM|k|>40
zsZQZ-Q{(X`#uxc=>pcGVAAhuTnya%BI*?;7qxB0htaADstv6O1nDQ7Gn6lW~<>P_z
zb!LsJ(7y+XfXeT~Y4_6<pZ5I(#5HX2)1A>cU`aGEjk>I3Q#zDc{0YoS?}g~wDa~>_
z$UU22lM(8NI&QDR6ydpv*9mfAR4rVb)QpHSeO2>NLay=@P+_PRBYkb!)~1wulS)K}
z!1gfdxbFt)RD+Dg^%KUgf??Y4)l#;q41Q0(p2gL(F1Y9-w>wKQap#5IPud|6AbpcX
z*Ev4O-|*NtQSVX#ZW5!mvh+w;$q4K7Ev3x?{#Z@(Xx2O?qcf?sm$p?R;u!QzX9P(_
zNnBmn<t9*q^rH1~K*1#e5$gyt8DYN3fA0VH)1!%-4|&f?c=aqiwD{Eho`$BW{=Ous
zDkqhbm%UhXBD*4cL|VSDw6f3WExdKM=)By0>;2T7xr(4ao0mY)UKa$3!q}V0i|5^^
zwyURB0+Nq1uf{hU$lrSPhP-?Pn4Nq(RKk3FfbmF}ga4_toO5g~sS3+8;b3uR`p)y_
zXtGn-SNhY(Yzc5*QM$v?%bfJ0Imdi+b%wM+vOo|Yp1LD!j?IEC_kM{YwN#+f^SsVp
zlbq*fdy8e!v4RW2wV=T_Df*tdt;>J%+6?S2d_t%lMV)e!S*?5D{SLMc@eFR<OsK2y
zb5W*HV6YkH2=4&-aHWh|Pv!@P_BBKqmF`B0Qz$WG{l3w_?s8Ka{)!xrEGmrLjX`B-
z1lEYTE!=Cuo2T1EMy4hsNX8ISFP?VMq9_+L_it*~ZLw`{y3a6R%r>61Y_nnp-+z4q
zq6JTqOho^$xMHXYl0CVGn0%T25d5(`rNwN;qMUNvjQnjwkZfrh+zKHwXO`JxHrUZ{
zhCFBHOgO)Xp|&+6$5r=v-klA-%hG6uSI_}>lne27{BV!(c=~DW3jg(_lq;8Lb9t|4
z9R~@x%nWyz&>g$m!cYcnaQ5`XxLI+2IdvRcBlgjz-9)hmU1xRT2JDLAA7D&jy=gkW
zk_II~tg&-`SOfP}eh)w8hbAnuPf6K(g|aroy^i)x_ggcVetz8Ut)@AHF`25YUIhy#
z1+XKmlYkNL31}4?-I?_MGG=Z+<rV9}$_aZm#HwAJ2t`{5ProjA&J4~@Z!|#t{1G%y
zGhHgKRPW&R-d*|P_g9RN-<wY~P$%t-#`1*QMc#`n8E?vkP*7cPXiHn2ih#SvD(*i7
z{rPijgskTVvZAu4lSO*i1xwv;W*dq?*?l))duW$Qw+~=loo55_W-hG_XD57*zhkXN
zbis5O<yV2PYgpA<tOT22Egl`SkFwetV9)MO9PLmn2_IF7C4g8F(9{CM9h7!lJB-YR
z)uv{{p<_}<)!c3icBSp(dg%R}9~<d7exo!U@LD-_;tBq^@Zk$dw)TdNaQx%)4f3lJ
zxX*q^n%dIef3-AD?<GkwuV0l)YRc<!-6)VsEPyq4Yh?Ol5JxFb%{Y%SLgwYv@0zRC
zd@)R;;W9;c#VuC6ZRck7wG1MPFl-{)DB1)fWY+UOv6b^?KFik7<S2~gsop26jrGNf
z?-oW!q#1~M`%=Cn^aAFs=Tqp&OYN-O#A=#ESUQ*e>fSGm*Zog-<_>Qsul}pEI>PzU
zgA@ha#fHPht)B0(y&NKHaW_WLdIxej(-jd@yHWE^u^#&dv;DVTl~shRb|p0A^yK`g
z#3<h_EQa_^s>p$hr!<QWtNC7l$?jEy*HOsnb!(8ScZAW-qk$BZFprc4tj}@?G~i|r
z+BR^`boyF^oACTn{+eCfhwqikh*Cosi&9j)FQ|5jX_A!i55((NIjr&TK1<1O?b-5W
zSWC$V24G+|_OHygO1j74k+-qizkI*s6d+!>TkxeefB53^QDl(q`riV9pApob^?U#S
z@2G&8<n@u`>_~le{YQgGX&Jf0_v?SdiFdZ58GO|fqV%BY{*}#+9)pfJkYCTC5S%Xa
zJvu4wx+6i2-i)-IfA0V(6fWHv@#9g+sG3(#&++R<MEiUBhj+e^n>fr`De4@)`Bvaf
z%g5!{a(5<F-60A@bD~%Lh4ui4(7U}PAveOM9s`>VoSd5nN81)Sex^>ebof3Nl+iu(
z?&>-%s_eeUONaf`TFSxw29^p1qeD;E01+n(+uzrQ(>Gp4VPi;X%B^ieVn8>TkXCFM
zA(C%1R+UBWtMwr(cUC+tUk^K@nsy37sI`_3EHI28hBf#u$2~`!RocDof79fEOSYa(
z`Zc}BB*O_$AjwO8enH#+M)_vIw7}jA@S2y62jG2v^YYp~)P37*-TK-zj;7?NR<H3f
zV(SG=$oa(3KbGN`yOSka#;#xuRrSEEjecsdG9M&FLn4}xCyaF~LGacCofNTzsj4vR
zW2MCdMc0C`o?7FKF!>^RBdsL&1p$yc{ht#AyJ%gLX2q`C@EIY!`%ba9(o|a+6WZNH
zUF{t0r}cZ|3?oPq2(QXiY!N!^F8Nm&l*$L|eI)~gawCznb&!l4?WqMD%SQ&WiwyRv
zj>fZ)wB8uaQU=>GDd!cM>@ClnyxQY-r!N2(u1W7^$T&^p`ftl-KB?~5pF?EodXY;;
z7~u!hPl;KauLq*HXi-WU6Lc1T`mML!Ew~LPih-jyg_(4f%o=0+Kidu9meN)iZF<@{
z2EtM_T+#NTBR+Gp?fBARPk22&zu2vwxvNo8YlTJNA`-+eWYMj=(y|L<t3Up@d8=a?
znfa$K6y@Q<(9OlbHXoJVIhi-rd~!|ee0Zytk5K$fLLI9f6Mm+t>l>gpI^v%I#Q=Xy
zOtPSh3xT<C=HumY9hPs{bpik8tH!EL2x`1_%g3KUS3q`;{r9%c{l#PKfRE>tET2zL
z?{S1rn*!|I@dK8_C(%5{vwlqV<QWh*0%^Up@VR}X5n;8u_)stAI=(wecab-7S7hPY
zGq?nu(YB@P_>J|xw_ojr^~h~&x_4Y9U+pR_TD8NPEJg8Al}OVW8JrOI6&pVen`AV9
zpo>Mh*y@8xkhax-&avV0mnUsmO{Z31(?=Toqa=<Qd3~t!qTFp<E8NDq&DUa_DCT9G
zR8fDA-Npr+=?yQ(V<Avlo7Mh6engbGW8#a|Ks_IMDDlY_TQ4-p=I@Z4DqOu_r%KH^
z*>vezOBT|Uz}Wz!IX<sC!Rd;byP}`W7}(aM!?AbpfR2qUZ1X8G?0zE*tw$D#G6mb0
z_`Ed^KsfAESqn^v+$BOKv#=E0coGZVS(<Gvv%kW`V&u(HY=aNyyd>^zBl2eY5A{zX
zvbzK3$|uf6aOA%StSX;nc=8>0ytwxl&3W%b3ZF*730%?l7RyDeuRiuB%fhW?Oil06
z$~3lIFN|_2u)I~N2_&};<!mj_)~Z^n#CF4%FD2lpaB4`%f_b)jgfP>aFPHT|lbSDk
zkM9CndexR4&H$w?7ZzW?2y98Z8ZM;6v{*Vd1ph#oaM(&QyscMR*3l?&kZ023Y}j^9
z|0@|b9@8l7D5?I#og6<jqUC*TKCf^p@)Z;yi_Lfb^dTsv*=3slt%Ju26{-h#uOoLR
zte&QvFOD!ie?6c%S&q%=vF0vH*l~+i_j3%n6CijfJlNz@BD&Y%Hs8ypx%iEIJ*^>+
zp>uRqiuA2+%XPE1$>ior$4p!tpuH~)2H5c2Tpr>@EYbK>4!=TRsYY`AHElf1g<HQz
zf81q#!bX0GnAWq;jMoxf4ZWBSx!3E6OeB(U?YPw#PMXYMbGA+bv~yqBUuk+JeLT$w
znj43eunu`U99)};oe6Q(U4t?>_A59;m()%)*)et1FI;zD`>KviTO29w(uJ7Fd1RPc
z12JbK_c4cw68Cjp?5%&O?vLaJs;gT$I`^vHN>VQ@{j<oJDIU4*zeMo5usvTw1j{LX
z>jF}diYFvYBX{UP^&ehrjHVea>DoMxZg5tcNh7<JzNJk21L3Ry7|b@>#vdJ;Eo2--
z5nX{M2K0Ff^0ni{DcyEpMhZWpPfbN^cuMc1rKtXGmS5;tT)_BZ2RGha?W_e+9syiy
zTPjQL+XEbDa4X$FZ_?~x&Qs|OwRqmbiXh>GjP7Xh$Qr6e>C;kW&%UnV+oun<LC0h(
z^J!NCjn!=K<R<(AI#@KU0e*oGo#VF9!}!44B{6?W`CQL@T|8_&YPmk{M&RV({1gF-
z+c$&R&e<YB+pweLKiP{%=G=?USI?r@o>+$=c*~wR;zem&CB+wH&m_Wbz@=ibc-Wg~
z65A}u6j^@J&9kI!_KttH1@u_X-=%QMm=hIsw>i!q(c&HuQ^U~=Mn^JhnnT7diK?-y
zv$qcw0(^<O?Kz8viI}vpv2EASaSqQI#_ai{ElMRT>j_%R9YyK!t6BU9T%oxgsVhbb
zRpIE>O~%%<P>*tp)|!$smT9*{mSx65Wn|`lnzC~NFRdHslyBQITdr`T?QbAx=`x^&
zEZy|~?-KEAq`EyBgdCOvMF&a|8EJuilISi9xeni)8u`U}D1|f1<T{7;I(7n0Pfv`L
z-Ahbda^iV|t7kb1!c>c7Q1oS#L^4Vk{aX$TX^@m&pY>%~&D*THL6=(-PN5!G1`BsS
z%JQK*Td%e-I3^3hS{5S3DP>TbaHzwv!_0xU@}$+M1|AUykox7u3T}?3VCfH~jp4(w
z8jR7wfj{pb?sgp2!W#Z_PZP>K$|XzW>)CFWv+Y8-E^$br<n+uajMlM2N|>XJ0h!}W
zV6l<k@<A1mkzRUEs#ytU`a~)VH)>zA|5kr_@Qu)f-CXf>47;ZR%N@g?>Y%FntFb#S
zdQ)Ky;Jcip%$OgZFN)k+ILVdc#9id!&-Eqg#b@p5pSXh5yMaUhT?T_Cef{DusR{>x
zb2`T~oQ&Ma=~SUaoiQ*IF6)rSj{>Fxi}d#AR2%}P5IT1%0JF$ck242GEPr+p@Zpia
zlW1@!@1Q=Sv5`vnG$IddIFd>Ht&8oIrhpSs5xj0(o3jOTG5RP=XfZu;w;*%mIwjeS
z9Ol)VIM;;pYW>Nbp9)V8{9(D3jDL2sDE%z)-_b=2N@C^*EDc{L`ViO@uhI!~=t?`)
zh%eQdsBBea`q;%@keiWrT{@TMez~bV2E@I1ph~HiO}EmWKd4j`tQOOKsaXcB<-SQ8
zq##!L(nY;Vzj$9U56p;jkj5UnZ5qH2i-=Um{pJgc{PT~?gVf=BgC1gv+w%NNES|A+
z3$&M7XJCM}l(Tt`zSVTEpfa7<q>T<|({Smixx{8|1^EL8mThMJ)~k(1vMo*Eu-6`;
zVL(75V9{)oMPErU#fi*WHULRv;v{{*sTdmmkO$60_lw9v2>p7=G63(~(xoa_P9D4=
z1{1e6oNG0n{EN|L=_Q$F`#t=!O#A_+AxhG9%teW=l1OLFF>zTcH8%+3D&9!P>U0Be
z_b{$`x1)fw(hFdf?{AgXT6vijSxO9SXu|9B>9FfWAnuC@{MaYU9G#^b4DB4{+fC{i
z)|N?b^ccF7m%-FK0{Ia}#EitMDUpXy-xtq{lD%Wzra0D4z@BSb{m`h=9QWih5V$7L
z)=T=F|7fQ?_6>8_664ZHNAzF7<Dc@6xprEqQ42{`wS9%G^tbDO7Zf0R8lmbx%o}l3
zgzIdR4<3or8R2zpI-?u3X3_FeTqXNrE&9dZQ-g9N*zxdVQ-bKxRRuDCI2S}m3R{U(
z@8|z1=B(yI?`%(q0H08oD}>QLCQ+`jy@V$MZM|e0t+qeFtCxn>^TpCrD_27_QS;PB
zuf%5|>$;1;PYq+K2Hoykq*DA=;qTZ&#uYX|OPsr83W+LIs3=u+JlS33i4BK7(3<j-
z<&0hN%5GAy`e*y?tH`=K3E7XS+sA5$xdb|{8>rl&?%p+}C8gbHwNr>xnTeb9nOh;!
zs+x{9%Hn<arLWSv@pHW4Hv8sXn*fvUSAm7tC75dju$hVj-pzHd)T88bwkhfU_LAYf
z<RbY3p+DbA;4~S(k)fp*dqYVdrgUcqYx49JGc-%Z${oraG6c|`s^Ai5y|*E*23;$F
zOVH5mnGTyCYitKxX3MpeVkdSJwiqb@3V1>W=PJr*)>56V$DagMuP(^c;1{f_@s3Qq
zxQ9(7k_4qNFICwok)`R0>fA9uQ+bhzJHq?{nh!o9PN{=s>0GDwJzf#6H=1l5d;{Fp
zW&*(0(@bDz|G(RF0nJhgQszq5ulzb%31-n`FRg-jwZiaG8eF3895Jp`=_h+qdcDl!
z>ZyH(X4P75Q5BJ3HfPc}QY+WYyD5A0+K@@dn0Jn*-|EwbNhxnIQy0eU3x(YLKo!T5
zVz=ZwZ>7i(O%Z2H8G@vwd{&n>v^hjzWMQ6_-X0p{^7GM6dC1|<FZQtOezis8$l6OU
zNA5jE4b>!}Q{suJ0LcUO>6}Tq$qj3^yv^QC=^MaNvfC-PYuO;5{e)vWBRfc)OU|)%
z$l8jvU=Epbx2|=lke(%%<J{Vxp`HEf^-Nrr=ak@G3o?l3P_nS-U(c@`3Hf4>Mvjin
zjXdUf!MSu2Y6;+j7<!#lcJoZ)oQg=p6g=E()I(;>0g+~ANAY;RHFlY~^Xqwdrv7aq
z2>+p(<j;4(mt8I{s;;^6Vlq>4{P{FL;9o-xCgi*K84Bs&(SgGBXpPRASWPahD-Ct8
zd)HR*ms}|J6kZ|0Pg3RGa=v*4S!Ti9)VX^on=(5Pb0f^Zu*EH}^+e`hJA)u-Qw>;2
z+_9~WyP#T(@n>@o0#uK^1p<t~E<~OG=BsL^1^UV)7m+1NxFobCqc%751(R~=zVcW5
zL9Wd+obRA9Zl=XlR^oGE%G<n2mwl?QQp!R9i1STSG^t5xjB=niAjX2rUjVO!oIRjs
zGd2ZC_00wL($P6<2<_%UD4B#J`M*U{+?s4uhn4nlAWHZ8vH$$pAL25j3}YjsUyt(r
zASklx)|td;szc)PFuwzcXbh3Pg))V1<?m$pH8aCMk|$Uhf}EL#DRFt|Z%GNp?$7<>
zw!)N6!q#o5gH9s58^oU;<Ql2@+!M}28zrCiMkZPZ7zsVdzoRo$HI7NwWPNAhM7(GC
zw3@~6$}0|v<$@>tpUnSy#4`(QB$kq$r0wgjL_;db*;2~me~KS!5Jxnx?ig&o+iYyT
zJt&MoMhB?llutJ?1HdB8aes0SAi@`0SfTAX@PLW2VhK1+Q1vVY-!Lp@11v<7F1=*(
zsYQMdjW@7bt;Gq~bzV93kti%}L#riqWY*2El(^@!HZFRFQTicLZuFEjpcb0d5vpxP
z=iTyV-{72P?uQh%Le2_HiYJbLXe2&^Ja-OfOeDF#r@j~iMOzpvyid?zzR&+tQ@!7<
zfN55jZAnw`=@>Ps{B1JbxMpV%;bUvG`gU}+WUJ&3HI6S~4~3WVEK{*~VIEelqRxvs
zpPBP`iw;A7c}LL~+c6_~hm(bwX_x7l<S`=QhnQ9QpJtWMi|C77WvSw!xm!C675A?R
zk9O7ak~BXqoiQEa*PY?Rs5y?78Gru3gfdz*pF;q3fKw_>YLF*2FK;nL_9lI*wEO9m
z+sU@^)$`vMV^fr7Qhy;w@_dc_;dx3_2FooZq{*3~1ANLJ#zok6V$-|R%m?|}%ak5)
z5?g@kMBC#$19AW<JN1gZYaC;OL)@n!jBHD$T%5d$&lvjX(hnm6ReMaaXmmftI4V@b
zZ<vLoFY{Dg8J7S=xmF35)A{)6iOqw_n^TXbM9^pGzNVs;O2iT&v-KzUxwfioM!voz
zCZfV9W8SB<+w3njx~;FpqqyF8c|1Hhyd<D>Z;dm%!zdy_LcOT0NUOH$!=sYmVkV=2
zB;csszP{h&ksq9W2wtKMW7u-%O(L!<1WF7@q!w?1dP6-=_oQ{zuP*%ms*lTN`nBJ8
zjN7dmv>C|Ct!8vMO;5WU`}DJ>1ijQJf3<hWBP3Gp(OkW`xy2PrZ?dBVM%)kyGbXnS
z(ILl9yA$PIB2%w1>oBll=wAYIo~@2^@?D+N{xdj4mou$gPbl_Rd{;GTTbOPp!$3Fu
zrvj&HIi(1B$=9f=6Li#dFEk5cntMA06RH<C4BKNc8_a^i^;w-pM(zrh!7!#-{&24*
zHrHFcc`1rVwPMpZ)-|{<$MEPBbt_3k|7Uk6=KO^v|BI>=HnX@w7}Po|N-_c;E|9K$
z<(1z*+z1a$QCmhs715Z`R82FSVCO3U+%52FYy1s^Xd&0Saokx|c9&2`5^pk;PM7m9
z-JX9&F#kYAmPYLcr^{>lD6>HU?@<*dWt2u-Pc=<h3%)rOXm*<y%=qZa*+FbOy{M90
zamh5Q^UDDb7iX8vyWBXpu<>|noE7Fi`IvJXV<oT4*-gv$@g0z3nBH!rKUA6U=wEPY
z5wZ;au`Y?z<3b##6tM1ed~9X<o+NgFG00Cdn7gB|CA9V0<Yt-At|(s={Z}*I&<=^2
zZ&MR3N39HD>bxoC^y_9UKK84lf6;v(yBozL584bOm;h7jb#DWfwaczlpmvyy)TO0!
z8({={Jopj{M;ImO5b}i@1zPf?p8T-OOf2}clW3Fq!YuPe(XAFH4>;Vb%z(8_Dx`fw
zSxsV+5HbsLXLV94e{!9tY6D?E336bax}K!RvMwSA9~`F);5RbY5x<sHz?}j`x%jmO
zOO#u(erXgoKIF(Sq}M`~Y_+e>QTT9e9zt>2MuCBdWX7>JM4k*6d6<}Uty6!~xviO9
zXr^D3(X#saRQX8ReLh4{uLOnyYQ{2b=J2Avq0%4r*tlXiA+3lZB{9h4T`_DzV$n8m
zeBt&xA+Ea~nr)`5Mo2ANRvn`N-HiQ?H89Qfq+8b2_xB-KL2z8To|GE~9sj)Hm*>|k
zhCs#<eU{($zfPufRD+?&FPc0mIn9D6tXpE=hNw2(p?wPr8ZM1&=GNQ@E-c4jC-m99
zMXwCI`FIY=q|yh+dKrdqf9yN6?qW6swCvDSHP!Dm>`jtlIQvw+$q|3u{}|g2Q@UJR
zfB(kL;kLlJk(_CpXJ(edLx1#+ksN)-pwCW9y@~d3<E+nPGASlhu{YorE=2&GhP6*D
z>DDrf+IdMQ@4bkG)b^1D7gI&yfpeEv?EZn(1h2BL*9;jG9ouHX`43r9BRg$?DJTfY
zyyibwcZc10^aE4!Gqq(h&NpkVVdifFwHvB*yp&}!mq32X?Cku_O}qTVs>C0A#E)tD
zb$pOhg_c7dUI!7L4plE`fK3#u99TkA5Z_$2Ra2np#|zPu;>`!{M>lp%akZT(A0H&b
z9$JmzcY(*2_Gas8J%thvk4j&>&lUb9pG|0Q)o78kE&LJsLpD>+qMcGAidbR1{LM1p
zs`!mWfR_}%c>iF({JT_InXmVh)<xi#y&I1UsSCE<Q`$GF*2Ca{L)}CU;y(Fd^ZS%7
z@z<+~)Amc>ThrRR|NnGqt(_Lq{Vu4IL3s+^ldAK)iZ;4+D6sPJBs`c(lb#l?&DdS`
zabqioM?c!?pT%B2P4?)v7Bpo(aG!lHy;>N{!XHZI+>CwFB%1>!dF?ep>B|v(CUQ!M
zP4r4ZAV|sitUZqWrgrbm*K)KLd?O#=AvZ*RoOZ7GAYAM&J$>Cd^dfv_m3cakbzl+~
zX(j3}Zh-W6S!b9^>Ei_TX_Ogv-AI0AfMOoH+{K_NM;`;%sLVs!0oqLQlSF!YK8{WZ
z%Ji2mg2V??6pfgfytCp&pUX}ou=tI^@}!kt^vtNA7`oiSFlDg9NG#boH9}cbE%n!a
z4?c|oTdj~zPc#U?xwXK^)&!D-47LGY%HtM@37t^-DVPM0)4e#eOf_5l^}2uV^vdbA
z(<>+w>FAr8BCch-ie|?}8uR?Drp4T&WT^h?Hoz`%tXzH-mqBAek3XlrZ&*A%B*l5T
z&TI+|-FlJ@=gF?Cc@~%TLujZ9)GK#LN{wOuJsgtni?G*cd1ooK7zZSop4~JW>0kZV
z6-knBxSG4i$A3%uc;wmMRQ*&>OT`Wpl<|^zYl?m5%rI`6{TH359fXM;8_5I*g}v=;
zINCfp2_o>|*yL~)cP^~v{n*kWFV?rJzMocyM{+Kd+s;JrpY@sYI_eVjyWhJgxwJ#C
zJUFdZCgpUBYc;e4T1>V<nX{WrX!1!eDvBL2`Uj+V#Nz(eS2We@M=`1IiP{nl2FU1r
zgqOt$E|0|h*;`oY5WKsuYF5%wCes~PjpTUC>n+M>pMgEZr12()>)PaG^MAD)Od@_-
zFuxIk@k}PsLz$^1ksVjJNFTtKbj_|&+B@$vCGS0}&k@+YUA#DERj@MTpzQ!W^okjn
zH<~_ZuuZV^yqbv0P~JYW7sMS5$B}~5{SP2-^^Unt4xENSow_Q6SeQxKiHBm(mI=f>
zN-MirU-+qhsChOCHJBC87Rb-$!9HxnMl+K9K;#BMK!F~@Hz<j36c0P~?U!NKo4o!G
zgd~#ag>1X`XjzrgdCf;3ws#NRc*wqNq69xoZ5;W~fn?QeRjym?)L;As6JUGyITT_-
z5`V{l3lUh=E8}=u#o1Vhf{nJW5h1cTSrxA1u!(_O*R&g>@(E>DO=f%2c3lQ9L}E2n
zkCP;#-y_h9B84rHKR#HyFHi_tZp*Fu)Rs(5Nck<KKOfhmIAw8($lt9M`wIxT9M=Jh
z2KV(2-nReQ$4Y5}tW^@_Nszgd&WDcIg$6Y!RS_)2I^IF+M&uP}MRcby56;GrTZ(yB
zPQ977rokYG|D7WtiX>s`>Yf7q<WC_I1u!`*`ld3)XL}f*R6t8dftnDnDV9^O(t9f}
z#q#9Wz7KUA^Fa$ikn#q7lk}eh<XVx8yYYm9@0#<o$X4eC=YB;$oN($=!O%~a_vlZt
z8<|c$`H_7jWTa0uAkn3)Qf_Eot1~B%2rWmqdH77m)mFoA6{}m;F5I31e02`dpUQTn
z>i!U8jJoNj$R|QLdUe+q`B9*9L+hIC6#dx{WUiy9%SVW-1u%w*uKC>KeJ8#0_-UYt
zZN<eXG9vvsf2uOMKa)xEY-4C?XU#7h7c%z2E{7IRv^)LrlI9>ij>Ajm-}7MOz;fe>
z1uDKTDUiJE%1*DGsIVT3XoF3*UU71#Kh~^mm?};oCHv&+3ho}g7{_l<?~~!07%ocb
z&NpT~=8%Y>W%V5A$rme5pR{=0$<MFuX`%<vS@dF_(>ba!SKC7Q>Z)^3ZY4B&s^Yw#
z)ofKG-_o9yIh~po^BZyC^8N^=M=QU<nhCco%&C~fd3T1jtSPQ;n43t2KVWr!?4xQL
z7vwi-1<~1YxdO=+oq*eDw?ANYneNo%^T059->X963aUJD&bCMLyL8WO0x~`(Cr~j(
zUX~f&y|4ZjQ|~4ETLKeo5k{h=&~cR|P3(DTsKw{PO>J~!yGhp~5DckT`S#r*v-^#>
zP3H8j?!U(5ZOCL+R#A7NvNn~PQL;=L;83rwHxx+5wA$C%V<LXE{qr<vTJYK_EHfCb
zXx(Wgi#1#|(mj+@j@II>;M@B(Bq$%4&}nGme^W(MxtFrvy5u4mkuIj#chQ#@A71aX
ze}BeLxL;GvoP1TC=>ak4e~uLp{a~>e<u##F6Dj?Pg|hoa1m4G$b-oqYC}TUeNm~vX
zc8WueLqhp}z%gp@;Vi!MyrDkSrRU=u>f>;5HytV3-^@mx9W+}0{(<Y>ovsx&(k4D9
zs&+f87dlg_wGVm^a6z4!*C)bK<~iLnf7H3(^ikL(*fd6VGHn&g#r-<hAU&fH%<N3%
zKXZVxJAL<E(!Ik^q8St)q~>9l&co{i)Lx6fUU}rxx)8JZoqx+ow~vM6h{CUTfxL;7
zb<yA4r$$YjSe^s)Sx%d@d1}cUtky35{prwxq1SSA14}%jjNXRqTV=dm805#izPjCN
z7Eqkqwd49t{OZYyWGPFO){2!5cdHrhROBDy5be0@l97icn~kUt&sr1Yru)Ld=D@=1
z<ULc~v1rqkP+dIi*Wp)@56Iy$EKQ_kfGAOukM_qc4-VD_fP07Q&vnuSCX`S>DG!u#
zYJN1QYLjJrO1-^Qx{<VU<ek=A&77B>U@5KgUqw5uS5Qk`)bJ_gh(^rIQvK;t{Yl7x
z!K9;50xMYd4Gk1Sks<OFtF{I7@levjc_|rqqp@7gL9MmdagHo5&uQ52hx%VG26n<w
zo7<?KRJwv&)bO8FoOaQ;a1GIhr<4zJn2eOEIK!V>xlp0Nbmko)sv_iJ%<bb1$%bs&
zH)yZfQDh``6kknj{2A7s#{{pV$gol5FtCyTq}@RCCo$6Fmgi5G4Q_<>PSSnKcnJFs
z4V~uFde~E(u?8U(;;cDY@tBz(>LJC_8mqyq0rjx-m_OT!7qWPYC1He;-s6Ttthql!
zhy~enu4Rdthe<p)x30)$I8V6waX%W9)`;FLjnX@Q2phn7s+x+{SU@!?|M>(g?xOXA
zG%{7JCaP@-TDGAvl{xorc=U^SFfV3f<-#L#vBA`E4x4t5r=#W9ji6ok@DSv^O1t0f
zRY$e;M2P=aOJR?Q5}=0VsZ|*zGS(HY`BJd{BFWn8%`k`ikuhb5`tlUP2Ij>Fb)htp
zySF&{n@zjJ6>R>Q3ra$j(OG8<G{$smfC~EIo)4Sr1A#DVC;rX~I9}-0Qz371;{9T+
z9VWwd>3+fs+<?=MPwg7on-FNjYu%OGWuD7Iyo3_0hSXvmGPIsP13w!=D5AfoA^5H|
zkx)8pk-4>Xwxw63!I<R2D%;HC!qL)EHT)9t)nrmzA5D7evT{{OmtZ8|p~>IHOYjuG
z%xuh2LH%4inV|GacI-j_!vdfnPW;*${VhJ=Kbh04u_C9bnpSlJDLPbGM4y(CGv0!D
zSU%hU44*=zWbm&0XkF6VFR<iT_#8-+d8>9+H1)goDh>$0{?)<1P!SM@-6a}yPf5#~
zxtz1D*CEU5e7|9G_~j$Hj&fTxtXSfO6zf%I_NL-Jk~L;Hs}BH3#{W47&DZ6L#fm{c
z+BYb{lSV5Tfy4mMsZtZuI6JM}GD-CuOo%U#_s}TjtiLe-1&CuX6Qdj|aRZtHYPm3Q
z*wbJLvFT{NRrJu#B~pk#ntrz`KI608>6)C2Q8xUi`@i`ToIqrEOsI}9j?#q)*qgnB
zgFA{{S`vITY_qFI-T2hmbKTD&Okd@U5devJ@YKdd>8oWxh>MTSXU^_q<3Vf2E)$0c
z!yi|4drA>D8ktcVw)o?au^(l7(@fG>w@;I9lNCP9&7Tg9h)j%|sVOX5@|y#_G(0(~
zfQReCK<x(8k(}Hndr#+L;Yv(eOxFWbb-#Y!evY?ZkckC)j+YizkIAr@>p9j?kEtQk
z#`JRI1j8%ZRdmoy`fSSSBysvP93-VifOx^)+6xyv(>&Gl;#y#Dl3}F)QzN-`RYs|n
z;PX>{+c8+8*#<s(1<d)50LC`&1U0ONq%Z-s&f~ocwPKT2ob1lg90+yBGIHk<(G~>b
zfzcA{`p7~X%tJbN_ze^#lXPsB6{j)Z)CT9xjt`$1YE}XkF*)9%j$D2-d{S&#>_%^M
zQY{2|xb92IP2fFWP56%oF@@Z~GVU}XNJ9jWXlbFAxBAsknb+!5gB?o+t$V^X8~0*3
zw3TV0SB}C0Z!EL&&kSxZJQ)5f71f>EEG);hZA{yW9L!%J_XwG7ERXE0Sncahp%<qe
zV{*KEV;Y&K^zWfu4;;$8M~Tv6x;yUYWF19%k#hI>ryY89wXU|mq^Q=ww^AxNZE|fi
z$$xJ(+h#b8)>|gcmxd*vT_=*CuO7i3iEK(gjB)B{qz*dfKEAyq7)ZVoJ1Z^@qPesu
zZ=7R~WC)yj<xd3Io9^?I*Q{<%FDU1~9L2w0dX^u3$_H2VQM>oTcV<ueE)C_0LJi~Q
zP2<{GWVv5$$%j1Y0hhm!<(%-uj`W6vu(4<XDez@9M;k_A^52Qfl$^9asihCDqm%NL
zqw)2<X(9X9P#GzR3zU~!h6QsHX?`PJQ3Id4K)4Ynl65D=6n$Ol4Q^&sov5L@a-)ZZ
zpVIlt;2la&!$bIQ_h(PGDs!<>%Ct|yP2W8sP)wz4M0e99R(LQE3({!ov7+cxJ`WNo
zGD+!Gr5`x`gO(Q^$llv|^EFLJaX36PD6iGMxB4SO4%OewNOmnM6ir2W$N?K6i#FE_
z;g}9$AJ89>kI}?ELkB~aUC2Q+IB^?{{&D!l^C)1cCUcN0o^EuIv7L=ImNTrm>u3<R
zp1@CKE=;a*p>tE}*@nRYxmHNvm1zANl&8dhTqq-XnFGo^yS!d1>*0ew(%{~3=(WPz
z$iai8l=xN#GM&5IJQG!>v#M{Xr0L8#2ANX)IP#>bwH{!?AU($b$S{MM1n;#WlKwc|
zXx;u)!djD%`J@XX37(5Xv$U!Zgfj2q!i8R$+f+U&oI|cTnP(`brSxwWEJ>|yD7Mfl
z#DB7Dh*3_ngYc$k+-qOWcvQiI`tX){0A>6#q`nc8R-vw<X$2S2Rv~b%X}D>&ehWKg
zIhEHm<2E`V)#rUEcI0iUemoUe{r#K_&aKI^o2E}mpH;)O`O5()l960-+M9}igK7Pl
z`Wvle3;i<{`}o%N+8sY%X76twfcm2JQ7OMUqPCJzxfjvl<PQYlRX)UO&-<TR4e{uk
zMl(`A`Bzn_q@!uN1b@|p7}h~efIg>5`Kg<)iAw}Jlv{b)L!J37VfVKa2X<N$KIKe~
zrSo1VBnli4XjiLLoMPO{b@9(^I;xdo>bmE3?thA_YmPZl`}alY1)ZcMMbp&GnrJOK
z+9h?z`KJ5C;^P0r1(?a|7e6Z&n&Ox#cZd6@ON?mV8nimZyG9LLB?uCK+aBvpJQnKE
zceW+XAOB7HfN3jR_#s!OpvskLemW6OT{iH;CB1m%)50)P9=W4N*L_W&2VzQ=G?AEU
z7E9rpDT}5PkJ}cu+NM@)i{G>Glw3aOuGyfcoM-gGbl5z_vii{3ljG7SwZ?NSB`hS-
z;5&o1Zsu$$bk+gmYmWx24*3@%Ht)8+P2}MWdG#|lJnH*mX+6RyS2wL}=ocL^%{AtE
zI5Fe*P&5g(eiCBPH8}xjAT)A3sNnM_bJVQ-gR{P3OLk0u$h1S9@;#5w@e6tKIKjRJ
zy|b7_vXJAO_3-ZNzhAnvN!}T!hEGdIk`r0&Dc|1jF-81n`6m)VvQS+3%uC;*NG^#t
zDb;#qZH#wH((jY=6yL-KUO<kPjRZ5=d@J*W#fN4=z&fiIxo-5t^^3?O6PvS7Fkoe8
z;3Kwh0@1MO>C2Bz8&iw23{R^T&sSIhsS^_#w3qwS=B=vD+fLlBy_K~o)lgnoJ);=(
zUMOjfy3U2_**Gs2kYc-x9gCSB`~CNoC!iVG2Pbz{{wQserV*<l9KkF}S^q%DhF(3-
zx(i73e2AO?Nept)ueoOKAYFxTCkziJ_K$(~N+ZFZ9MfEa99B<qC@4X7iZ1GdwJt=a
z@P~YU1jZjxH%Y&;gNDm{X*Ea4_iX7M`_kc&Z{po6oPG*2DzlhM)Lp?S!%V1)-dV6w
ztqPtGfUHa(WZj@c0*yg^j*0tm0sAVHn^_bqH%5tOIaRgu`3s~ygl1S*dh@dm<9kU7
z&C2sSU0!dM2Rn>SS4DcXL8R<Uf-ies#RxWyT_HJp&9nv~;N^9|BV-S+6-($`1;R6r
zXXft!kvKYj+IG-ItMo_=7SS~*WiPb1RE-DDKF-@=y)$C|(}VekmE1MtK~o$qhmi{u
zQ{N@PNsztk?STFE$uQr^JE^!oV^mA>I!;U1GUo{;onOJsEf9E?+P~kmM!{By#VXq5
z{;~W5J&}-G{6ej2*({FWxnSC6OwaYVbOT1U%|{vfPZ^UV8OU7d=;`Twp+~R$Gnr1!
zQn2sxPp4$Jo`qp|I(_-(tn%eDZXzqo=nFOGGtLavS<s@bU9vX0r<O~qVcxW2q3R7X
zP9$bBd<EO&n{7C9AtjGrn{E|!`B8Pf)x~UfrJ`jrmFhfRwOlu|Z;FWjC+1elNM=M6
zi3`yCeK>{w(?V1#<Q3A&kK~Ob!GYL|ZTNY<bQM0C{K_AZl|jOPPz3D%zvWks6`IJY
zU~a`XRU?1+Y>?_v)I^Mi(-mC_>8+-YG*rjDTtg)5ue3x;BnT_8nU;AoInURueYo*=
zdH*NLtt7=LM$`A?5G6OqUrzkj9BRMERk4iC-d2{9VeEYy@^R8iwKn`o-o{Jdn|%R%
zv#8T+e);($?2@Ys^8C|MLYS@!geFcXFs*lH?$TIKw?7@iz28LHOw(eOf4ApQIT^bZ
zKY7CM#vZFkC@&9sV})>Qp=K!LPJYclnM={R_{y;m5eDEMZ6YVeIb4qmTlNfFzgw!1
ztW`2_F<?tVd%K^ikaj)gVgDU11?+67Baz!WBKDSNd@Hv>>Zc)WCpjN(N>#Z|@%&m0
zb26=lKCy{x2UhOZXO+cFw1*Z8e@2vL=RGkCksM#0bt@#rX0GdV9h(H$Jin%%G)CDe
zDwF$O(-~Rm{kugfFHnEQx^4vOP^`r_x+DhdoIGYinc}Gs2fVFMI?2n)TY|unuNEhB
z#<;af46$=g%3E_a_*t8so2|})=boZ>YVe`2YQhJ}Y)-FlGakKyIi{C)ZoPd@VjY9N
z&m|9<EernNAT$u(w_~2Ht*_Qk^*_Nz%mU6nAxbj%&)$YqB1lpCifs9vH=SJICf@r0
z(TOgT?-s12V5f5BonwMA9bI{6X`ubQOYzq1l2F4V`-)VU=9Af+?IQ0J7H}t~6W}GL
z6Wcn%lYb)jN0W6nI)uu_!=P@usDX57!{Z;WCbCg{h|PqhG|43EqOOmR8s2M|v2?1;
z7dlMlbep*xRSz%rS8G!z5U)6+SV^0H=2x+w#C6W-8ky>f7r;`Dd805wr{dyJk{s3C
zuCiD*Gf=@!W!Z!X-=H(pmHyoLLIV|;POQu8u{#Z6eXJ>10NYcmTpe32{b2g*^(1_R
z-0dQ8-mf@&s8o;ho0b_KrzS9=V^$IO$MsCyynb2fUw3{wUC9^dW1ukAEG8A8M~>*;
ztq?y)>k7TJja~{negx&vMco^7*j8X~hI+ll6@N$CPil8p+E!qf%*@-gDMiEX$2gzq
zWOQ;O&R|BkNUm9X@uuW^b)0?n1kMnxci_w#=of9UXL*8#igY6GDuOq0sX)f-i^s%y
zfJSk99Z!-^p2SsN#D&yY>tx-}@a$wumv5Zz&%uam>9ZRxat0UGZ*k^e5-K5tOFwL_
zc^C=aq&!X~7vLVE(ndFjdViTSt(i22nhE~wo=76`_D^EA{QX?{Jw8nQhIr=$LfjWN
zi(HybogR95tkx;qn43eDu~p7Ssh!bo5l|%1L55`FNeYG=0{_=jZQ4yK(2oezs#vN6
zU68<ONkRa4XH;`pE-6iDh}4YG)lyy0l=Xx_?QoC&(*gHG4bXCR5nVZ3VL-qYae6t0
z3VWS$HZh&utd9r2wx{{j@n*KHx@oSYoaOH!X4)ggy1<Kf>E6=x)B&kIiSW)IZrAB_
z3IM(<CrP<`IzFG0oLo(EbXbTdtiCjFL$%d}3r&ofC09egO0SJjW&DK_>Dwpn+x$gG
zOhhP=IY*K3(kcJ<A_J@oQWDg+_}%Orf2Avuv5=9{R~R}<)zoD7m=%8(((cg7rp|qW
z(VYAGfB6B9+x(DptQ-nR4twjM=(k27pk9DVU6U+DLX$8ojO4=ew}YuN%M1fY-uk*f
zNOI37$z|`s#Z)WrYom5kiD+*AN&@*l4xjeE<T?rQdFSk+;A)a+1IK!wkR}Okh}7Q`
z$Tv3mT{0FkXpE=P44t}m#2X&keLG!p<B7TY&Vzh5%KD=3uA=IjLnFUDS9nC9EZ1F_
zPEO3h?!BVXLd~QRFYu@seH&`hl=Qp`A#vh3-f~=-K0m2t<U-oNp<eoJgMW$sSviS~
zOqV2+?6G(P>()I*x0&2@D(;y_=xeLQoVI}Q7vEjAi?G|En5jc?Y85o~^j?72ZSs>T
zxyN!ejrC-yLF%E4B;m7ng<jMQpq_&KUtl%PC7~Vr^nxPPkrg6*9?3*rmUVGqP1%u5
zb%>BD6XaLJt8qm$+s1DgkLT&d)ZTWb4T57c`<s|_Y{zv!b^BjQBcM8zNpVBD^2j18
z9Ig^Z4lU@UJ;D`S8<q%-4JP?QxmJJ`-hFnrf+fF9T;OZXi{hn&a4%afWUJY1PJ2#z
zs8ZIiZR+-1ImSMT`+G}Xd@6}A{Ymrn6>CbE^H`?1`lZ6C1ldquBm{@d!<SA~Mqbr|
zr`A8dongvg{il@gsj?v-5LC@QtU#ZZY-eKmCF)E*?Fqju7%@b#w#n5l8>U3wE;Nej
zPB|r+Dn27v7VC+BIG+}rNYI(2vZen0S=mO*k_>{CMy0o2NSIs$$t1`D|GnF}*p;2A
z*zR}=YDe9o;o~i3!!C?5ZnWU2?fFcVxyNmJucg0&WS;@EyQ`cTYF@UTtUdLD<KeAj
z6_5LuY1ys2B24FLmip$7dP*xmtb7ds*zm0{wzH^MLtl0SY5gYwUCJj7sd7omQLI#)
z`~7s1DOqN6>b)<RVodAAv8nHJx=O|~+Hp6w<2p&IM*RU`pgffQ{2%aE0Tq67!aRRu
zCLrrqJ8!#o7?tzk)t}?>axOyt=JzN6PMMgOGT>&Mw9_6$Q(;=dTRoS*70z-V+6M?H
z1T2O#x6`npm-#2n+=x0w&%B0aSA|a|nv;O)xo{mtPBnG13dZIInl5nryK;WQt#fFw
zQjmE#FLzYp(?~zkI;*X|OY``DxFUm$q#_g8$1(I$8gY;Dwa9t<C$B+G*<|>zrmhMq
zV2H?5q;8@Ru$A&)cFEUdY!W8`_CBA{OdwV}k1fhmRJiDQ{-(3g>-zS|{}#&tH`*a}
z(|b<axt6O1LSM*i=aSyBYonXeU!8N>wmeD=fkxyPe+YDg7J;xnvoQrg6>S1-K@%-N
zMJht7z0}vm^V(PBKqMuwu}U3^`1;Y10;KfNTc6n5lNB#hSErU26^%gZ_~JR=Z9ZEP
zZ=#|{%M$jfJN^My`omz~E1&Cso72O@a5MMNZMV+8Dqx?__m`>%fAhl2Vi{OCMVnIX
z3IQ~OuT<wk%d`3p?b)XKjs;aqkmQ+jvO%}#ggw^&x~|do>-?0`L?VPq&I&8zHbxD7
zly9V6Q2=>Xg{0D5(k7t?lG!OpmZ4*`4@Q6KW|WsLwX?}EWpSlL)2^*tlPl+XCX<f}
z3|?qo<+I?M<=12%AlLL-*j9BTTId1trHj$zk~p}`=eD{LI#6ZAYSc&4y7lPMjG{k)
zf1<(`yNn@7$g!}>&{t8&-`eF;&tJ_exQ80#CkMF$pbd_jU>;dd9$A#nQQ_{5npMJ=
zYD^j}^oxxwv}ob@MeE%qO1$IfBl{OdL^O4kl>>maa-FPb&JHu{T3fkfZMMS3vB54c
zJ$9V4Ug31GsXVr7EtT@f;=mjRz~z?uVz^!CH}dVc5BIFJ(N8KY=fu(pj0LM->HDJh
z(YwB!1RR$Obqg5^t=Vr@p@3J-AdZX{9Qw>)^yqTfGDwU9pIMD)RZ!r^|9mU-!-`id
zj!ScCJJ1k4+>r(@k&qVb$J^UfQ7sSfc0}CVj1T8q0FH1^3B&MWvNfzCsDO8P?-D_w
zJ@-m6_-7xLZ5AQ5-Ui`<RgLN{gEgfptfz<S784n*p>!F)KI^LWTO~s)Y)jCo@%3hi
z$k(@PREG<FCXNeSTN=dk^|`JNPn>E4EbaJ$=a5|=T{^~D&kS#}^E%~!D8S2gD((lF
z<MZ>!CAe%>o!3X9NfB8?wYDS{Oujwmf!{Z({}Q3x1lQ?Pc>+&IA{z(<uBa>d;y^pA
z3;twiHKJ6CJR3i5cAT)P1Yy0{G~^+YF!l`7`oFCD!0N7GTL*cl=1J#BC|3TvblmL|
z@qz8ltZjfmfFOY!rFZh(rJPW)Emw0`b_qm6-@WCXJi81qJjStWI%*a7uQY!+)ghEo
zH`wDh<d7H<E@+BPgyCv(C#%R*z?~%yB3sHF5NP&xu4sfs&H(TdPwPZIFJ#6>Z!o}N
zkw3Xz<oNP)s7GxHnD1ORO!@J(F!kgh!<DIx3`y)Yoo{6W0)Jnf+Zvcy<m)TXh8=>s
zCzQEXweEs#7Ec#1(Gbsp8tT|{53a|BvI+li5W3)#!cfV+SVhD0DnsC!kEF^_836jd
z9-K~Hxh>ZBlC5Bq6}jz><8CVd<ROX^IOIGGToT!h3e;57?a5odr?~z;J*_wBvVm@2
zeYt1rGNF6PWioUIndV4~jT&J^2I<{0AEuOD9f-2ZFm{en{+`x5t3{$kzN>0<$LVm6
z&CV@m;7traWi;*%oy1<%y~N<~|4SF*P$5gn=GT&H*LEDEN2LvGKgG(wqOrkQ8pg;t
z#f6Jc*vJjWh4#8g{CfR52aJIjOP$~d@CtF%9~pQ?=MD&I$wu%TT>GX<7?AzKC~dkj
z>hmm_icx`pHdg%25M^AflfvEa2k7K~-(E{03~4tjyMw4*7D#biEN;vVkT)5+LgBA6
zBe%|lbbmTMv?;#Ze@WYtqY~22PaxPD8fs>dGZkO=c&Q{{EO+8N^_IZH4N=wzgz%)q
z63V2+Ug#r7a)-51ip}%-0x|qT1No(KKS3yuyk5aQKz0k^egA_@<AG^p<aH)#mGlf%
z9II#O{6TLfJC5tb5#9+?|5IlxoQE@JX+ZeD@UDmoBulJ;nf}d-$&qJ&$z`}nWOP<Y
zWA@>9(G(*{E;d5kr3Q_5v)<&h#)W2VOCr;VYm2N&+gBf_4g2Cc&}<M!b76=*IL9G)
zi^PdllOB{%ZYs}*@c-COsXz{O@lrcqBf0wz^u7q>2_)Zzyn+OIb*Vk(c{$GuQz7qZ
z;BP=}fHvS~F=NCT60VXy6pEq5aWUt$0FFr-K`uy$XR_~G(ElUtEug9jx4rL8cW$Jm
z-GnqqcOxYtB?y}aX^|2(jdTgpp|paubc1wvcXxNZYkTfF_l)m5_q*>HpJNY)3Icnr
z^~`6^`TXbaFIM4b>cl6Z-wlpTsfe<fQyE-$>@|%P=|goF-jQJ(w3j=GMiaNjN~5R&
z?@0o1)&Tx4qL3vSiY#08fr>D?6nt1@<<o&bQ@1=}ju&=%+_3_*x!n}`F_xBDXQg9i
z3T)98G~^nUYYDI7AqT@+(QRi46-YberfJ=zKM6s!uQk{!o_ED6=F+uQtKLct3{#IX
znj>X@hkH<+C4Nlf?&@(1{)*4_!;pE%o6hMqFe>wV7)Tpv!tnEvqLqi$4hWZVS%=nX
zmNlZ%Pg6Iuv++c<o`DEyKZG2PqQ#&RXAG$_<2B-3r^+&JAuHnM2Jiw4IejdMc2n0=
z%20XUM?sEUCmZ4WTZ59(=c4k#_9wiy_s?XSiD;6RJrVB{Q1BExRW3(ML%ifDp=gTa
zy6na2lCyXz`t<g9F^nZ~oogk)sLCZYX=GT&oDgmCdzzJ!S+5fd&zTJ{l!Mj!VsQO-
zm~Y3W2Tex22Mi-=c<~$HWY#{}_W=AzO9GYd20lZCWK(X)tTOlV^-gw3X%t5Iw(8at
zhfT##QXDA;b41KfS2q9Sx^+T+mB?kfj)cUi<qt1teSZXkKJ#OHDEnmlAUqx6Kn?yu
zslTAIK5}0FFAN1d|0hR2a%*pY?i(VE{w5j+-R-lV+n;RL2@|4rT1bYDWn23jOpyY%
z*}hrSY?Lm==<i3|9i*ukFCWlZJ9=Y(Ct@ozFpBK2F{&oz_wQWGI%ZdJvirR0LM8{N
zu0@<JxkvJviTdTZ&WL2DoWjs;+yqTUT9Fy9jS*8lLYk3{zV&2#eEpbf&!vlv8y-PJ
zblhpmx}a1o59N;epbD<Wr%`z1L@0K?R@7X`qLOeRCt`|{G4B;ux>2xbl1NS}%qxly
ztabJa@tnbMZZw2u#$#Lt7*3ZftAeT55gM1|8W<kk%=%(unWWRzZ2jH38L#`j!fMl>
zfi4QPqlxTl)`_hP@cpkKt7bbW``j6yGwa{<%d1m57pYU>>%P9-CP6d%!ssZ0-8+lX
zLQSZQ35^hW`8}ghagbN=H5I^*yHpXQlUKwt<*@CxGJE}oYV7HZCEwa<%-5Q2!5vF=
zPv=?r(5eWN9j9pN=}P5>T(8YjZQxW&a=v4f`6|0MVse1{1N-q3jx}j3&wfSWqouTY
zu86wYHX}{6I^&kGWy;yJuJ`D7A=OHv%%?MQ^M3iC>%YUK8Txx7&KkNM3IV$hQxx6g
z+5mn&U_Q=Y|62qt{qdHLO!jvwT31^3oah8b5oyCgE|fL*1+Xo6K0o4tQ3)hNtwfts
z$Mm?J<kcyJ+)PAmOU~guW`9Go!?7kaBUY=Z4#<MdS7pWMHe=Yvp6np+V->?>xw&rv
zMvIV@x;vP09W|-Z&@wZFfa3%;DL7IlBlwQ)-In{>IzajU{q#1+*JI3Ih@0n+5`|=k
z1k<xc#l^r1xp}F`;|Nc2;BBZ{Dag;zZ-Xe%(^zjLi!&Dzx)ny$wev^=z2Z+0&AWCA
z>&>|^*StUO?wy}SQa|-Qh7XUe`qp}n{&%|k@0-oPa_Il?kBI^3K#W+>r@fG1Z`IHz
zxb`G8w>MQ}rw2`3F9BQY5G~cJAP&b6KBex~5<DH33j!@*AEROmQf6kM$wM>Yn`Fpq
z!gSofef<CM9scL<4)vpVSa9E*)|z)aZKa@2zlY1JOFGD{m5<`0>`!JGKxM!P<EK<x
z?Z=}a@kIU}6cT+0Po8bS!z7mp(T$<Q`WUEm>}SQOKMqMQHzVPtX4ahBsK_A$m{&5T
z{?WaE?%V%f)Bk_N7{ieeZ3rSJYiye7CxEIuv0-gfjqV7bxAW=);FHRR-EPJVN*dhK
zabY;p4-dfS&%~<XdIQ`eJDLTXDZm;I%`PbgwlD~s<bI>{1-g2D3ct58DnB&=b4_jQ
z35HY#k2jM+@S(br{ks33*AJ+9UZZ<n)-B=ygMXG+a8BN}aqk*{kR5yw1K>YFRSo;L
z8Z1reWb#lJ8c@rj9~S$TX&~Qi{iY~KVeQV(<%UC;y$(Rw6u!J`CrCtxH}mP@{PO;-
z?E1gIp}CZZZn*IDxD{@y8(Rf;o_%WXBg%j82c%Bn6o@tjY3F6v6?vj;tTk(2jB3{7
zv9JK1UUg2d_#S#399_t5dRGg4?*IMm`VXJU=?jNXBC}JfCI9=Eihb$}Fi}R!mWqZL
zJrqCP0QU#VP4oXo8e%J6@Twvr*Bhw#Nu4nKFEG{S3OQiqjn-a?p8m}b|Nc0y0-B#_
z>k~WxG(uN9w#KU!BS6W!f0zD~5su4+vk|1iOP`0q@uP`<0irm!CIDy?TMg&{2>?@p
zpJ6#)H@%MhsRQ64^qu32Zh38)hc@`q_QeQo&PgyJ9}s`M^QngqeKT6Ji0a1T_#Xe%
z);5GWOYnDphNBSSpu}k<uY>>XQZ`q}!{0z7rU2-lZ=m^#&f)R?G6-Ol@E11T?5w6l
z_f0Ndb?~E(oCD4Si$w#8T=eknQxaqD#<p$MJn_Y%_u~=JrFn}=E4iY$6R&570Sqxh
ze<^bSu*Bu#PHH?jk?YsxZsW(ZVfmn5j;4ASF&ca^fW3nx-qecu9Qh&lLt?P)N0Flo
zp<~0bJf}1to$$iG&v2mcvS#ro&4kVd04PQecXycXYy}u<L<YArlb`d%&VRg{oW0}n
zxY%=8@<sa$CoEC|u|QfL2CU8?O)cGLto28&(oL&)s?^9k69u{Guz%sNk{9m}T#c&c
zU9iw79V9lBGycQF^#6Po1D22_ic$V~selBb2tC#7W_vkE{LV-Kuv_nZ@pL+hqr8t>
zZfC|Q=;_|cY5HY-$+(uD8FkW^&Hn*W%D3_|O}Pp1Rlo*9LsRSG!$p-)@uU}8YxAJj
z1&>FQs+ic>cf3Hu`yK7HL-CU?`uHd~+!CQo4fF8RWL+!3g-i5O?Ov6!$y!CIm<)N$
z6ieGhrZSEaIq8A}GI`((Nxs8;{F_Lj9VYsrXK@2!w|fo%&1hzjK^(YUHAgjrFm_hJ
z4{W=1B8oB9!H)ohCw8^OX+2T+eI;lwf>n4eTDq0ox{zy3_+jWw>YHlAdn8&B(8CHz
z)0WH&r;R=z3!|A7co*h>j`n{)=KBc8Vm}%$Tm0Ka#6_<MICi*jUlI2til9LfFG56`
zfF?EAjz-{RUhf~^GL#zu3{E>4P%4i_B*7+4e7JUgGG3*G*luAtV*2OT&9}4sP1pTC
z{Fwj<aC^JV7DsfvDkweJwn77B@^QsElVWRIpOJ=``qS3;WdPQM6CFdWXTwF0;@=HW
z9H$X>0G$n%)8@eM#Z*IJfOt8WJ`x?6vDfxd5czoo%(*go?&>XK!uyCi06sM6rqAB%
zH020*_X4S|3+|(SdtyXP53fiNop&^~k=gE>c}@X>_6~A${Lz`^{}Mr=JMbO!hemul
z_@9>#`kJm}R8=@I;|@q2L8(4D+Fu3u5z3*8@9C)lqbl`dvbds<DLJ$pch&=sv_zc`
z9;`hguvMU~hYj4!oR~>CTr96wF8(h%RWG;v-W?PFkw$ve%=zJZQ!Rn9?|8E5!C41z
zCrSee7`<!<zy1smZhNDTcmMmYETLC~HT&a<ky5!WA?O`dbkT)XP}Fp8z(Y!NEP8d}
z@y~y)j*2kgzt8>|HTf5(TrRX3;Io>{x<p5n{N4eul$2foTn?@N64;i!IX@tz2PV>_
z{TG;(lT)<*7VfWw?SW6ix%W3@{vOw&c%K3%Kytt(a?205hj?(}Xe&?pF}8>OKhUDy
z&(fxwmmYv8xKxQ6_$bnvKXBRDZWT7~ayVdCP^^Vg?6yxJQ!Is^0MG%x<kU}GW%w7F
zajutcSb0*6ZgVfdbO(@*apCaSHhb^eS&$qMh68N5YR*QK^yc8K?oXM}l$kR&yA#nK
zF={4?9%huvQrLKb!MLmSq7X7;<(Kv>6TYKg0DH2|dUtN`{v)o7_rqz%{w6SuXeUE9
zqWnNJP%;nE0f4K;rL|udC_Wa(MtP=cu2_D-n;8Y-=fw40vUL#|(0*!k*!@-YmLe|=
zY`N-2k!81zAv=BXr+lopu#;<tY#|jGX=??f8VRcVi4j@-e4*jA-eX{qNH(}l&SILf
zc*jpwA4Q5X>P+XEPzpgcir&wEqb3l^)0DajNgB3f(-sXFg<b^!2tBTPfNF{hkL#_o
z3m26GIF*+u(rfeZx$=Q?K#xnW@6`X|ooz}=l-P}B^~23>8?dX}4USA00zoRJaE<D|
z1>P|j9M%F*xszlyCE(*&Mv<WKzRv()l<#PgV+c>9Xp6uG!4poXu4i4khr=^Wakh6!
zcFv6MR|-DiQ6AQ8Qk??kPrj+{d!?O)hdKcIys$6*y3ax@qH^NY`D4~y5HL#8rFZsx
zhXK|JM0cB;X{U@^{oTiF9JwuP+3$7@FC8N?MF8FG(0LZH-7C8y7geud#x7%^bsT1$
zy!2ZCs-4`}Tov6WTPyXl^P^y<hQE@FNMJt_MUN@w-yPGU`!mb+(z}mythnvJ{}Ah)
zg{zk@T7DY1Yy?oGDR$EOQ|P4p9+o7&0ECw}71*$~(o^het?FXb%BDzTxlol6C`eZM
zOzme8=AYu7FfWb63T-pDXTjTDMU>%s<wj!wzvmZZ(_L3_z9si?`4QhzX#%iY>O>9h
zozl>W>!w7hTm$k#FQ{-U>fqHi5I4%Macq9I`?66{yp8!a-}^5-ePZz6{472+ult{2
zmPx|LcdPOCz}jCu*uv(p3TPI8hI4b50WngQtrigOXes85!~Gb}FRGg6A&8-n{iMV{
zVZ*>ZnxI`hrkYIwwoo-3Ac{sd;B|jzl^77(Om%7nI_Ged`G|W$H5n;K#gLR7r2w$+
z)MYk1_{%e3S4ID!p|L4xL47v}sIfg3Z)?$sw{MuD+6Y0@bn)Cqh)@7bg$X>*HMUQ0
z;bc~cfNy;?5=EsQ<VFeaT)0l;bZcoF;Zgv;7@tx^o2xA4W`eB0pmBqfF9{!IkCO&H
zA^HHQ!8gP!oG|()Tp&*6f6L*0-Fp-h<=Gy-Mk+<~(Q^fyu_dgi9=H@oh@!p^<o~=Q
zD3e#j=gUA$=W0{@RJ7#=;D!T%gkz2q?lK#Gb}@hGHAxkiFCm)iWxVDw=&=C){OrS6
zx4ak(z96IE%wyjRDff|FrbRf#)mKqWA@xq3qZv+}{>@iAZJy#B1Ec*^s>$UXed7wi
z_&&xo<X6EDVPN!M@xVIMoIm+6jduEY2C;!3){s>N$Ss<)UwB4prdcXoc%NlazpyVD
z!|3uJ8?_PGNHh`Bn;1!`zic?fffG|V35vl_nZon2ThcGCeepb-?S}Yb2JwF-(wI;z
z5W%sGdUa-*+!R~Srw9`(U+M)Eem<s<64i@zN)Mk@^OzZJ1qOjhTmaw%m2MC5Zb-Q5
zT~{w)9N5ysc?%dZ1p#>k=s0$ByECa`;EUMk>DoC^*?cr?b(<Gsg=Iz1-**BP7F38J
zk?wg+)|fjwRvgw<#Ed_DK*eYXII?Mk!@hJEeIuprvm?2DB}Jt-r4o=Cdn%d6erYoo
zsmBF@LEu`rF2r{`vYBG>N^X=SIt;mE!F|4nEgW8w%qBi<2>c@t-=JngKaF&>`870L
zF3g@`bGKZ*hv#D9wm*>-P`8Y?)Td;I)R{Cyc%-$UrPRn->f7w$WW_+YOZ7{^hJWBY
zfXCgmO)7(UjX3{R76MN=85b(mNU6U|=LO;d0nSkXIE<ft=*td|`XgIVSQa|%k!%jV
z*qb(=HHsOMM0+~MTdR2g0N)7P5cO!YX&#zU>a1sdOOg3rbXbP7m4YKzgl+RY#*UxD
z&W6naC_!5nd57QBBq<3&NkQ4GV%8(d$|@oq{18?<dInv^Bq?K&>$O;0kqNmu&DC#S
zMp3<JdpJ{Rw<tEaXPUCUj*x6QVs>IS1CCFO@2sUK=758#4ILluZ0c5n@OdI(ZmYI<
zzldzsp<<|Br@5lLx;G6jq2(cK<s)GD_e#*-Z1MD8yK^%WyvAV~KCk{O7=~8{!~{^z
zn?D?0!}YUJRw%Te<&{>DU0F9oP6X-@N<yQcgO79M)>i3%JwAkON;YQv?KAr479u1b
z|BqUbm2(s+d{dt`T;V>E;Pvv>>|-E&P#4wiu+W=IinPt*gZD|>X)oTvGb6xr%hpHW
z3l3>OQZ}O+CD(hSW;RSWQlCA6cq1J|npm2=^iMwtk`tk5_9DN089V4sf*jYC3~#=u
z#KTPn&3U^vQIIgfe57Dbg<pU#2r<c;c%yF>Ul*CZTN_0yKslW=3qUl195Ns3mYb6i
zV1FP%4Yg?|Z8gPN7Xk)adBzTbTnzy)%<Ma63-cez=?upXGCB5Da#TwNR?=F|>T@ho
zy{=9;^OTg3#7ymJrX3HLRKuTO*AYd=B!oOe&uxP=&urJx(wCGfC%)e?Mo<lqn*bCx
zsV~|IwZWavvs94{^>jCzZ@&)dJ}{#iYgH2Jq~Uf)Vm#Yn90b~RP=_@k2K;aU%k7Ye
zll4BUA_C3=O;d~MgrgkO7Gt+f3wLDn$}(KKk?B96>!vZnUK5XrBN|Z0+()Bo90|ol
zqszUQn~;9Uih4E_Z)u_2yb8lDjNMcbDRjgm%fP|E7kj)O)VozG`|?0&EQ$;>rG_D%
zP;x)9CAm&pQE}vAM(!34RP#ECjjsg2u`KUcL`GbrQL)#e$#i*}+r7kS^u(@O@Z!G|
zrJNaE+rPMuutRqI@{rI?9vs#s2nmY$3PtBr9*Sb&9ga}YptvEsXWLH=IeMa{({h6=
zvO0vOv9M(ky}NBvgU9zDcUaP5y0%eMJ{Mo0`SG5i<QvaOLRo%r<BMYH$h9o6$m@=W
z@N2?SQ$T-f1g~rTKNiQaoq;=`;kQFgWMSV=^NzVc^#Ms685|RI%5op+T2J3FvlGM1
zd7%9Ly;zB;FH#sb`V{MsNf6m>p8bSiCP!&A{(NF!&XgKyuks~`8MAnSbp=ZBepDi*
zbHE1&Q*ZPm?v$}EO=o6}j9n{kZ*|swG_gNdqe-EnaQ}?c%)^3d2woZY!aO>#hSykD
zolPk>ry5H`;@ub8D*H375RuJ0I#?5}fpr`baWEu}_xq3E>bfR)<^4n(gH_u_I6Y{M
zUpOKTWevdvi#Xhm@en_{`^>05%<d$YzoA>9Kz0B<bj6Ov`by07qm2QD<#&oxiYZlH
zoy*5*VXSJs(32=U(tvMJ^%0x|IiY6^D3Z~*F{m-|jUN*-1485UYc<{?!+#M)hxstk
zqUxacJzk4>Z&q6j9{iZ+KeQLixD?&A=^OB5baCC_ACzwDazl~B)p=t`jL&b^a6XaF
z(P5}d+AYBAzV3N&86LCVY233i>Oz2z0DqCbMs#s{^N&Bn|7wKX=lr};I7^ozluwev
z*5@}f8XwJ5+m#o|G;GNpt(F{TD!0_`YcZA_6~kJWvMH=le54~4i>RrY4L<S|q#=-#
zu=WJJN+FRM8p#UD7|Q~*)4Hq>*~}2COpNNc7iRp4fj=yz+V*m3AML>F!|k~5^4d@d
z>P4XUfPn;GnuR~|FsfoEj;9Qmns;AGTv>t0_O)>Gc%5az-RMR~UJY+}B^~!nXq#m`
zY9>wz|Hbj2xiYsqzL3Q!JAX<qQ7|4~<?$cX=1kp4>mNKSXh5$5LrTQ(`bP7k&2*)~
zDyQq1mrD6Ze3g29+iy9{@15!hZ2=`0X_1vj9d$$QccvnPRiJnT3kjitA7sFtax1-P
z6mDUB+a$IimgH!@CgGWH>77PM9c;HFs41Kwe3l7b#9kBAa{VuJ^FkiN9@<bxu!;o<
z`EXs*d=$qUGX!in`H~Y8i*s7Eyyd|N@dw&z!NY}(a(c20P#JMGYTHbq;#EGugIgOv
z{+T7W3g3&4Qdz@lydn0!Rx4g#a}fYpl3C+pOW7m()kN8JaZ2mrj|^zI@H}qcZraXR
zw&?<B=}eaur{HKF8qMoxrI^?tNLQd>$K2Y9O_?h3hII_jF>Shz-_-4tt3MsI;6tFY
zJ$aVAyKQOQ6n~fWLc&{aZNZ4x`0Yi4?bdGt92xhs<j-aGmUZaCKIq{e5gQtdCt?=u
z(rI_A2Nc4yCU0YP!C{qMZJkcaP{Jl4H@OzR2j=)T2YtdvkLCtF$Zye}pCw0m)_#JC
z_hc*%+6zqkm3gT?b>2*TRW-r`qe3^)3!?m>z+0;6HHmzA3jp1G%YO%I_{U$`bvU82
zakcG9&7ZvpPkC(@&Y$vR+;U>6zL>2i6ywj`e7d9VLO0p$3tyMv?_Vo(9HmZ)Uooe%
zF!H!`3D9`Y8Fq23N{5Tith86EpQc-3b>CBV<Z6g`az8*3g_?@t8%m%_oLP+azI9)|
z1TR8EUIZ5rn4a%l=t;l5wNSHHv!BPekY4)3-s|S-NI&=mfL{Y5#Qf7IG7`xDc*&6>
zpb8QUCcgMS5_kH+n*H6~vjx@$-lrDCE1AC*qOSL+{oN$__^HJdUNh*MaL;SC)qsYy
zj4Fdif_n?YWy?3>BP!QD(xkq`ciqkn?FL{XN}z_~=RHBuOFCFL(|UO8lJfhT<pemZ
z#xc}<xya0W23mBJS<~RWfi78w(1PHXR+t#d&9kd+Ag0R|@C(^>+2Z1^>~J3Y=hw89
z%1d&2a(o^v%s}T*H-Z;5`dfU!-OP0Jh{$>`_5~Epo&6ApqSwDC#o&at@IiqesRMfq
zCl@rtDsqF2(+VxupJH%$-`&i)$T{VVNU3@2wX$V?;=+mIL;qWqQ^9URPJ+C~kdEEt
zTgpRXi$(dD<=d@EvN!xnB@cCbhJ+xxJ$WBPS-1J!HQOf$(5IIr+z`!54G61&1a^kZ
zDf$vY0-kC9jm?|$YiwU8=NMV4u|H+aJSzYeiL$@;)if>Jtj90yY4jbIC>E=(T=_*=
z;G#pb!CU9B-;sANr`Wok&T__-4RaZQtwoKc*44e2?L_`tV%HLZQC4-?a`uhEXb}o*
zt*vL&vQ|6i_xVhb233lnD{*7Z^zBEIu*N3@(V&vxo*~=aoZ{s)3!vR`83+wyEmK%V
zbnQ4KHQ`k{UN|hvx2KUJXQX_I0Od0@=Njz1>sJcGy6k&tusSd3p?@K~${%jOMv{E&
z)s0vJH#r#A*Q%2Q5pUwi!9AHoRjdfi#V1X}M_6qyoPG`v{Ef2szDDvA>^}QevXcLN
zj>r*G=A5f)s#jWGNa{A@(6TctcWru8T5znx&6cpu&!0{8oDK~zh#y#N0xf`&)kxvW
z>hhLAzh6gABnZ{H&r~XZj;O57Hka0+wVsI|Gfrm9ZUZnoHh3*g<-j4RFIC0vqw&dB
zqI;e$X+6E-$M$GC$p06gvo@=4LNaB59^Ty1%9oPZS;*YQ@&|-VbYRraW0gGahi(=o
z1QCmmb9>fN%`b&eUw&nw1r<7Xj$jvRn`6!y<`;3=+n7l6C+gQqYu==QgRG^$)4|T9
z);uu;=Q6vx{~N9g_>+pXdTK+5`U~pn()M-qqVKlP-m8BhpLMvlnV_3{detd*QQG06
zAF_h`Bds^lBvKHP_^O(%w|17vj?wX5<g5dDm3KsVq!AkkqiSCp#gE^aM7*f4pRKXa
zYRWWNyDcCJ+L1x8e$^>jrEBS|{Sz|hl<=y$ywkZDBOTGs)B!09i@iLRpf6cTYU;!)
z-f&Kq;J)%#s=edt!Uw4ooQ7J(bzMYlCv$|7RTz`^w9tXkwZQGYt(3cWd~-jzDBNMZ
zjR*anPPvP^D68+iV^hMGo55F$ETQy4CZ!#^gAg|ZOA<G}r3X7K{F$c{Q<K&W<iRIf
zFK{B-cf<Zh4Q_bk`~sUBm@)Mv75A!Nq`$WHL-V%ZBpVIDFakJg>@e3sG3k}ciNZbP
z4hn~-SAg>jOCtXBM9xw+EB#|f*aC`!Q#Z`=iy{QW{JU7<SBSSbQy7T9yqao_EUZ~{
zp|JJ#d^+Kz!&Uz=X?(Fm2V>B5E&;vT+cX^MvUIJ5746yyiM4tV;8%7eUA4S>*j;~U
zuS;XYK6X8KIeY|;8#3zD*XHkmVXEiK817}V!2_eL#5KZl3&r`=^%_+ayUd{_37$l0
zi@Uc3j&FW|Q`P3pSdmGZ$C%Rtne@P!S64(L`Fi2%&X{9pGrqd+r<cz<sz$EIhRd4W
z>%rO3iFzaeze%$CPQ-g<C8^?QE)8$YdZh$~a-nERisBE-GZTR|p4*}f(Tq~wbdPR+
zb|1&%8ld5Cig$ciy_X$PRX|cfpPp4to0742;j(yZqzD5{ODv<Rm)C02tGBGt%^l}Y
zn*}zZs0+iSCRuhB1z6X-n%l1BtAcS9tG;nu>|M+LVXqf>++8`aGy$U&+)1yhgvJim
zfKN?O0|U>wCzHF^jU<<)8?ub>b<<K<jtU0d=vby(k3X<PfRjO-3?$9rb*HLnIMpRN
z)EroI(pHye&(H4ru!i_Z^;B1n%r<lP>phA9Iq!n@>q%0JKM@VOtEZI-k-gqC#!gNA
z>eN<QvcAX49o5}gWb90ZS(>-xH}$+x&~$rGFF%*$L&Mo~kF)*?Mn5(w*UKi(F0|^(
zx}Y;XD$N;fv4~2}(F}i5`#_xa-h2N|?jGhE_Q4E!YU?m97>|=|A1&u+*^J%>nm;Y6
zQgt<HL>!JeT!xLK;>Y1#uz*d%p3qz9^nKx?XZG%{#^0c`^oXGV%1@?2r2G4#j|TaB
zW6$KYl)C6HJv)DHc>R)7<H3s*Neesp^h1<Su|~<0FJARYE%t(OBvJ+&{fN-jz!?7#
zEGi^M>_SecF7XcWm$-L%3HuGI`7|I?I)99ivR9CNuFD1^fJAmCsa)Qp<5FvHaTJMw
zTW(B*3-~GIn)E*@#AD?NZUv!ojK!7flA@ana;AP9QDjHXR)$|lXpv6J0LRU6@3*(Z
z*EYK^A~O+6J?G|wC(1T~t!%Gm+oRYK5E>AIPHF~(cC$^1ytoCrp^#WCWy_rCr<|2v
z3$2_+E%;f3Oek!7+VSs6_k<+*3eN5^G+a*jY<s=q&$YJ&#zU51N5I+B;r)KK<7%s_
z@c!C8&&qR|%d;BnH6G?(Sais9)RSH*LQo@s2+r~UZnn8mbt&YoCkxV<6lKi_GVpd<
z<grF-VI{os;Q?1-e@F=&ZKmNynm+3_r?r<r@yV`R{2MLsFX*=5W~WHD4Ff|;4dnSJ
zThDNuH{Lmh0LewO1O3qMElX>17)PVPk?P|fK)|^tSbmB%j{ode#mGxv*C04g4Fy^D
zdcCpL^mh%i^D+4$mATadirbu_u3LM6_$r#LuLWg#)W{UVq#UL6Ok%zPAVNkutYCfr
z95a|f$Fc>M_R)d;Qdq{Yp658tgZZ{E*?9_pRkZ?D{NW<pb&|x#fvx;xb+Q`hYTv(0
zs)*%-R1)xTclS%NE@aGL<dj+MT|BfqCn8%QJX@C(>P0+!$fh_jH>+LeOlRx%cXxI`
zKu0^Omrkt*Cx100u{u&7v)$sOHakAQ@p|)HH9C?3WV(z*Cy{;`oM1tq^J?%%ibxcF
zJPT%tc{uX18e!Zk5;w*b-H}3L_q3(AvMu!n@D}v2EA4yPfKFrD@|Wva6l7IdjT(>m
zA4+{VI8r6wm=p;%sJ`wLOuvx5&(_Bud6f~?16Kc``{*Re78Xg0B}EyJ7tYV+B9X#@
zWzrdGOK!m+UZK=(KOextN+AAvlUZ0MEnN2===S4Qz!Hbln*Be_0r*S(wON(q9f4eG
z1n9UIf5c`<aLDD0^wVx}7$t%cct6!E;gbQ;&#~t#6h-%T9559^P%GlF!0KqlWIML{
zGZ^yu4q(`R;(AO6BtQ3>!vf2I<H|Z#>aq%TWMY(9#_c<GC;=U!)XkFv^$I~TD{Azq
z{<ro}E@#G4;z308OWmvPInav5L0dRK&wx5{Edl5YWCTy`<_+=VhV*zN>M@norY@+3
zzm>bA?iBIeK@e*u$IhkB(Mu6znak`t1(68#fHIUF&7Q2qi?2I&#-T%z7MM5a>HgK;
zN1d>6&vN{65t3`EmXE48ro)0}FBi)Ir=M%lxJ^vNdvWJ5kL8=);zK7avsTea3rBPl
z?rpF3>oKdH4FK<VI4Cx^*$0kTtP^CU__;<;5!D#>H#?p(7OW$HQfdD)t&xEu6F`?@
zzUl;H&ESypICQ)(wukNN;V7xrlb){p!AhuM+Do)XMHWa$_?+3Cg}lPA^}PHhnnVZn
zm0S1Y^q|*<f1EO5Tmu8HJgs6em_-xW2f!|js_>t;B+zwS+hQyKGL7Y(hO%=4+)gsy
zKZLYM0VvGnAdmz|QAYBOoDOkw4?U7_@GE@lgQmv^cA0I{G~a3uvNyfC&u$wg1wv8V
z%Z`s5D2Cr&q#BGyi}J$`Uqt@ARzQ?s6{1_RA+a>fj?#?BM$C&w4G&5|X^v}*c@PAW
zVFS&Zhju;0W|_<xM%!3LfdlV3B{I&<#!e7pEDW16!y;a=Mtk(I46`on&q53fsBo7`
z>ld$-4|S^>#2Grch6c{Y*5XO*FCUz?$k->wK^;Ma_Q;D5$~lW<B=t&oi%Fnc6(MPS
zcz?YoOT_tVeOQV7eRl;NJbUytdiibgQnTGr&-S-wB5<PkZi_QachLAts_PX%y%?5#
z$<cK8K@UN2Fgo~^Ta=9HplE_y^=2vlfIh|z1J9tiY`nP}F&&OhkrZ}wO?@2=1A{?2
za|xmURN#XcHU3R$K7n<eew4^@W1AFqyPgiz{3OGIYsFX@(FkQyqF&md^eBMTNA`=(
z-m*dW3J&YsQD28J<>f16I&>O*>5t&r2JvInfm4xI(GGkelyn@o^zNwZk||Kwm2x(t
zJ$7&%zSNOwFnahM{fyMBui{O6weuTUdv^wlQ9mazMiA6wa9Jx_-V>Y?S*ti?4bU8X
zQzlqY&)w@q-8lq~^fw20$G!q*<o&CexNR9I@xd(k<3>h;9CP&SJJgd9nh*0G?N;ca
z`NhG_!Qg^7tOGJ`0XfX9!sbZMv0R`Tl+N6fY};2HKStXlrk+#dKabH4pbp2Pg+6P#
z_Mr(5^fxncqKpb=u#O0~rbA~b3;i0^#G=$e2=u1(NpQI06@{~#e5m#DW15p3*-wH5
z@Cz*b2>|ex*1qeg8}5oG@{8bv?c&#uvJE+ET-qx*)*sx7;kgFR-&_NN0V3#2b6W{%
z)WI9pBK(vF$N?4}th~{i+$}f=K@-D<Pa?nuz)35+j(onXE<NF<K}4`9wcDi0p6@M<
zNh_s1MKd?8b3m#10baQHkoFLNJW7Y=HkeF2cknyRQEiP1@%fB_y29b(J}c-0m7mDL
z_q8b5!|PW$#kA}lM);&37PB(4#h21M?_~Lg4ry!IfqiohkslF;*GacDfVP9V&q4h&
zUcPDnxgOt`l8k5g<tcW1Lr`OES)N)Br5j{P*Z25KePFwFal^!64Epr8HM93g56_c9
zIUw7$##i7@fBlAIxwRqwLKa3xRj=wuE#0N6JDg$k?u!VCD3WHse51{o9z6SixFKFZ
zi~gB=&~dh+Co*x&yj-(HdTFa+hlmpJFdX0(v1%i$SAh^{SzwigL(W@C0xOj~vE+y!
zMitU?neFxB)i`FGorWTz4Y%VDYbDB3$V6OZ>bOx-Q~^Y8OaC7^w;uCG;QL{NS019j
zw%t<3g84XSDt^u5omgeoy9CKle2g-8E?p;0aSMjL3Wki>Oi%WrrXU}e<*H8kI8}7V
zD$Mbr6WabUsOXNeL^2$P=+QZUOM&`#Qsw=c+1)s5HTe|b|4BOSMT?b@rw#A%&&kIT
zqz}SCv2*b8aoU<v^VA55b)ZG-MxFwl`Q;b;@{lx3EtX7O&-a2IbHbk}`Ov4zIBdV!
zjwV!KwUtIxDjh=iDS!l}@#oonNZ2p*D88!%uS(Am@i-t$fsTie_Dd;b6?0n`3`4ZR
z>6UhGnK&+TU>EuFE&^LnWohwVn2NERr@8L8z38H<+3Z&pns@K0?fn<LY*QQ#Z&E)(
z-(8jv8H#7{<x|>@@F(NbCTo5lVX`S>m_b(Deh_kb<IeYU^*Y?c4}&_h#;br{SjF+9
zUNLH!NUe&rM(epW{Km=A`Wh^1+rAc=F?Fa1bZYnSM8g6uEx@tY!oDiz7OpQ~n`B=E
zZ0N*iGJ|qq^&B(Eb&LKsWi?iY7eOlc^58o&{%hBA%O@(V>x1+nCeE1KHQlTuS5qAb
zXq5Q<bOu13*U7cUM&TB}@orRxceovm>oMIrGt1JC%-e;>T6|c>cY0K*$<@P~yVTEc
zv2ZP}W84aD5X*djAj7K9?=>%i#VB+5pQDFPQWO)!`6Fe5k10_fJs-*^FX;!0-!d|3
zq{A%15Y=EvfYr<RkgLoP!b}WASSo|z21F|uvJ+-b%bX0#Hw*7;#~dz{mjy@Ol#4oN
zzvCpxiNZ;u7_dBek)_)kY2GlP#>Mq*vM%$|X~fF7t;^|+*sBi6$MB5_p{4==o@a5B
z8_X^K=?B=kH2Yvg)yf}2+MNv%qTUfV(TB*qpW%-dADQ|Y!dlm;{5QFHRTXx_bCE(S
zlDt&TtKvLXKy>Y}<La?(t!%TuALKdZXLYzpM)6`fG3s~H>?pweDb}0g$@hSOl&9@(
z3TF)0J26y<o~V!>u6Nb39DXos3D8|oOS&4reR30P1NB^N?-AouNHYqrW_4h{Pshkj
zNykve-?S~5cPqsBnMQEfu#ad$2jOE+Rbvt=PH2R9dy%Jn_q@S5D>hdz-yPj$#bckb
zzg-bI5H@e%m|1y@KjSHO0#4#@c>FO2=?C?$qjz#@>Z9_~T;BGQR$xSAVW7e4pLw%4
z+4|;)Q*bXr5pAjQdZ{{{Fw9IAZ0ipae=qFWJsQZ1y`>vRg?FA4eK1O1s{ME35I~@S
zG^y-De_*miM?uQFmNTb?B&X*%$&FqMC_QeW4#Ehuj){VImIc%)$N}3zmG5h^6e1;^
zQPuj8J%xU)kLqX1==IN*HDusal1~}h5Qt4-&c`F%+pX?7OXY-uqDhP(8hw8hdzZ|`
z8Yp|)<;W!l4&-@TzwvRL1*?oS2mZPWt8eGI>guijW!{%QapA$?|K5z?9B8SzS@JLO
z(qe$Dyf;D%95QY0u8zg7O$xXh8aI1=B*|woer@#~o-8VJ(s1rEfG8tvD>$Y7>m8YP
zT*e!zWt8y~tt+M8o&kP>Ts3+OI;xC6RAoqNMjKrKUr~FDYYZm6hC^@+_Rm?Cy?~-K
z9-J%OKu~mx^@%aD$IDUe%B&&}9o@Tm8w=ww#b%dG{&LKv%Nbu{kQR*Z=Z)uQCLcxG
z83A`Y5++yFtzKfMwZVOt*XA_6S+l%rqeYAGVvlnM{p_YOR3K_oxLpA~`2<0-sb)7!
znJ58eLAU=`zBtm8r)~GyqjlaFA_2tcd-$fuA}#m+r~RQe7bUHS+cg-p)GLMc`2|M|
zvt8x9lzp#Nr)krPot|%Hay$8Jf6{l6uexMBk)xVrJL>7sJDX>HAy=-tnHf_LV(<_-
z90$ZA+OvV;)2r7DwqhrnIg1UA)H$ak7UIry4b;a6Qlc!F=u~|ls-kR@%nEX)Eh$*%
zth6UY44--)wKO{LJoYUTCEZH|8y%3nW9_<i6ZhSv<(276Y$rDl9PD(S`NKbk*j%f_
z-H6)B?G>DD*=c_M^J(PU%X4~A;mjcffhH`lm)Jh#MNv_w#_ca{jJ1EPvob`9lU790
zjv@Te6lbBsU4ljeAq>#{x5tOwyr!3{Y2m|MRD(Lcx-Y}bt;%J2zcvsrlC7E_SL>?v
z9=^ZnNA$e;K>RrYWrF};wHzu!p`&b1e5NmvKA(QS@Cy{}gr9$6Gpv&T;gJjz0h+ds
zkl{5$;)E1_T251+MvI*Oq_12#YhkmOeq5=8dqkAc8++A&G?)n!kU%6wb<$ATkmS3H
z$R~G(=eF(i$ZKTZ(89}humA!Ei17&X+ulFY?{H;A><`?DVZg&JswKpO-@PM=O5nE5
z?X)k3;R$5TzfHe<)yC#7=^38;dbMPHK_g>c=q4jzZii$XL*vWK`6@K0?2bhTIiy~q
z?8~t7R#p9+S@EBr&N8azUmDGp(lt%%X-<qIbbe?o1rig0{CAX3swmoRNwllLEZD7F
zZc*z9g`_8GlD};f3SO4x|M|w{FLUja6PRR?E|@%P03@bAos0axf*?EY+w6MXR(2S7
zX32LJXn`eCe`s)}I3Nn4QZE_Z6}$`M7}+5KBYa|EJ4lXN5&mvJ+Vo?6^D{#zBe@c)
z(n&@}RJP0<IRA8l6!`|tSi79W#$uvO5%38Jz<Pxnyd-K1R3)U%pk6@kxD}61sxwd}
zIbvEQ?s?+7TIL+8D1L*bC91)AW!?EfpmJE`wB<^4%-K&wYyURH|He-zu`(Xf&t40Q
z=SUfrg#)w*QnQ2z8u~ep1J1=!3r=l=L(C53=NOX`nK7|H=6az5WhdKh*i8}<^}T5*
zJ<NDNP_gw)7#6QWIV{eJQ3llex9#38Cvi@RLwS!}RRY^(DjCl6ZOL@|%RnJ8W`%s9
z<E(JG%mDe;B#k_H8eAj%kzMw3g^FGGljZo|zCj;OqMqS~J6zgkpUfG6aft;{Bwexq
zr7Na-=D#lfM~i$@zUgm^yn<ls0`_8^LK1Yzw4h56d*sz#_w0}|SXqa6qPvU?u^F!V
zm7P9OZ0avpp1y@JLc7C{(E=PcyoN|9>E?|0d@q|+H}k@BLq2q1%!o4Ptd25bySZ31
zrmpG~649-^ud1f`(``}J8fdy+GA{$;TgW`D-#o>h>AoS{d_A$2we_x)>A>z({UiJA
zjr(EI-%S^ZAi5GOT#eSP8<WV*9IlO;|Db+xUByJT@XOucz|@Rlc?RgZO*+!bJ&;#W
z&(MWg=z=VZ$XAUZGA}0U8gY6R(}>Hkb3i4k@#rAt$ubFDuFu1EEpMl5?(GicK{~$|
zRS=t9;&p#S?eEAwJ!j#fQ2kq#8-Ti|xDNxuW0MLrs43K7ri?&2|G<Lr%T2BAC!?_)
z@$%P(JjM&plcKPor|tgXr&YHmzJjaQW|#6O0|FvI*W<k0oAbNsvB5i{PXr)t-2Q4A
z{_j2a7EH)K&hcoTH&%vm6oW|N&J_iCd#gzWdt}eYmC`YK5eKEu>ST=ol^#1KxR<*3
zie%9Hdb7XwB&I3d3%*>N1C*h3-1bkK{&}<%+>A~KMM7yY*z#{!L!oHKuKXZ2)ht;;
z(gZcX@W(Mg)<Yrgd#S;Rpl&o^<0mi`rp;OUnv-9w1=+Ht?-UI@k3yl0%fH}e!dW2O
zB7@PwdiJ~BXr=84(6C-R_obO~N@nJ@ptAenG@?=zVs02SO-Y!QD1sIz2}5i-6R29+
z3A|un9#p5*&Fd19Rwxo~i#|FEh0o>3M69owOysj#`i_5~EBPJvMb;A^$(!o#jSGm<
zW&tY7!UYj9ILlW<)n{v5B6$am-ENsfMFyX|7~mnnZb9Ai)BS~&J;+*g?FjfAyRVW`
zbuwY;Ig{HgjYX9zUPJwb4aDL?LJ&pl&F49$XdFr;*E_Da!)nAZJD<Ahm-wbbIgVOT
zUwJ0flwoX~Bn46rjOqg?dE^;u{&%MS7&x$5U|CSx<XD*`gpjoX-8&*1hq}@T4$YNs
z3$3$SPG~8l$3j+iU;TfL;X%xgcc-BGB`rtMiOGRIP_Vjh?G*cz#7oZJ;1*PI>3EN6
zK?gdviY%fY9S7TyXsLJC-}?Gqr|{&kEdY9{76_#uPBrojsqI3$tPx<rgBK@kmX62A
zd2l`@Q_-R&=#kOlZhqdq%doDXU(w<M?>qKyI)C%V3lfr~+TFUf5{Eosn?L27g1VK2
zaN4`_ZEc+QLtOl8#GdaLD+KKX^xNr)OD@W!Wd53w5V0h0fh@Hqa+&zyd_Z;D1V@TL
zTy->_{ejfB0it4x$D|w5lzg0uejCm(yTHp2GaN8Fm>wT|z_kU!<F8B9Wru5a>=3OY
zb=A-(<I_Tn?)Io!f95~u?Z&>OOOv5C5Gecfl5=H;-=UbA-!4^LE!ykA8{&d%-rbcU
z@7B0nRfCQT=+)Fp#&_8+SG<bTM2jOWvVIO)K(=r;$k`H`TW?_p>T~o!M@hse^8uRK
z)$B9@Xju>Ln?h*Gr}MK6G})qrMuxhg8{wy?%}fkK`?v4;cIVn`OTRT?IYe!~^RvKR
zICZmIB8$GXtt3H7f4-IB9?yz7=dJu|>F`1Ja0&7mV*_zAgTIE6%d6!+>C$lJ*;5EB
zlc>0dO1iM}o2A2^_F>y4-Q>HZC)JU8N36$R0hi!K72&6S9$Y3I^?6lTmET%3wryx5
z#lxzED{V3`y!_aBk2jO=cuXEBd*+Cr-;~#};rAcfUC-N!y;Z$C{A{pTx0?@+V?J!y
zz{zCQZA|+l_uz&nblS%QqZq6aK}w2dZT_vniPh^>AHw_+Z#}8E9lwPbKb|Ufv4tlI
z<FWqXw8+j+&YrpgyztD;(Bo!14{4QFr1vXI`2%I2N|yL&LIhHLnqFnYtdKgR(t>(t
z6g9P*I1APp2_*V#bE?c_XU@*pgxdRA283!?XouyV{T|t*EZmX%K!eE5b#1;O-(XC$
zc*HqKqiI&jP&y^gHrp%1RU2%7t&<nh(ncUpNkDqakbRpNG6KWqT2C7kx_Lgsb*b0!
zjBM6yZ>~LOL2Yzk7R5=ElGLOmaB&~^6t*@=e2oHTy@3gOARe}#nN=!qCT=F~u|L`s
z)i*RT8#g1=%b$34U-{G!*a?c;F5R3J16tuK9@W`iz!}^+bYOGZ*Mg{nZD~yH+pidd
z^0>M2ZRfA@wL0SaVHMmNZ30njA}Z=nyguOZ79>3EnIfL%n!Ik{RTB}T9<DfT5?7Yk
zg0T-;UrRi;yWA9d0qbsSb4E!1Z^`)_$a3VZyIBnXaZ-_$EJ5g-SRm)^TmB@xrSPH#
z9cMUE7x@T*t{u$`>8k$jc*QQh{Z8Lt0w14ysij>2sU48In<d&eUSsNVSNB61kRbJI
zj_kf}(*xh0aSWNGaQ3~Xp;3<E*69WDaU^4_auPrb#@@YxPJC(wtWzoQ{kSAzk!`o<
z?igFSP&Y_rd=co-W!M`&+cCaTsQMLc9h%;KW)FC-N7x>jaf_qK=%>nymD7Ih1hl#b
z{%LV&ND@cO;O`ijsxn^yeZaIpVmKE+6T6TkdLkl{x3a?2&LGq-Nw28iF}QE36RMLZ
z@rl&I-;wW42ogbrt^@Xl-8X_~{<Zpy7-0MYy0Wz?qRJ^_EE3P`McIr7Y}^lG*ohUH
zVlJ5I&k2}wUugqJ`$MeaLKZUF*nY8)9ZcI6R<~+}u#30Eq`dO&y%MzG4@Wt;bQOV5
zTQ$ocmh%!y;Ux-d6Vh*(z!=7fmi(gHqSRwQGAB%`R-O|YtdJx|;V9Vo^<4WW_|b3_
zzn?Z&iq|4uE9ED3c;a1_8mCfiAYx;wNhcqKE?ITp{+oU%<XO$+2;)d(9ftBA56F-R
z|IBI7^d#<92(0(R&OzK_7-7W)`BC<Jq=$}r{k)w?S=`y^repE9gV#;nodDHy_dMC+
z*+F+eS?&D(c~4})>NHd7-sN?81s6uwHTvxj>g^=o5Ylu;H+DU%oUg5b%5N5bI!|O=
zy%=a;J+)G}9?tuF>?(&i-&nkHIso&zL_EL=2KAj}aO|wU^LMp}d~*-tzy$iqb0lt)
z(s4dZJ}6e^WciLwkaw&Pt4TjnWIgYd<6;@pbKLn%5afi9DGvwhcj=0ZzQ4&E`sLuu
zMM11ehdT=l@d?Z7Rc8dgt8Yq@6{6;67TIc|AS+mcechz*cq0@Dq(Q`L9EbDH;}|`0
z7L4so`-p>q;cV)U_{s#N{Fzi`I<Y9iX@0snr>kUjELivGpdW+i+&T(=)j^IT=}lyK
zl4tTl{kuU}f?)v$%xu!kpu*U7OG40VzWa+hO%xyJ@q#x^pZ~^zSQ5||YMHO~@;q`K
zes;v|;2t1DIoDRS{3GQapcjrq-!6Q2)Yf1xb*4w$=O49LrvsX^h1w*0F;KY4o8I{~
zmF*^(HAA&Puw{pbJ3AB~*LmlJsHY`cF~@hR|IB#4%CC&5-M%zy(IL9bvqKz23`a7R
z{+V!(2_mytxqq4kx(}#586N!Y9QpSKXdx1UA}79RHvX4f=7nz|Izx&&s+jOE@;<6T
zSJeB`yR(yCj@xFWR=UDLE~Z7hSE-VisEE~6S~sLV=7hco$SY7#pr)nOh@kiUo_+g^
zLK*43_RzP;KG>-zoea)-Uu~G#QFq9?T0$v}Q35a(o1-JQH_B%*o{WS#f46~x1fRc=
zaX?ulO4KnLYGUOiYYRYj_hAH|w}+tP4r|qMOP)1nBW{)?^HL}v`7dBYNKh_XnkW!x
z%=8(se-WZhK3mj9pm7e{HSLZ5l>7wrdT<6~smq$f-#-LrT_Wv($nrV%KpwE_fC92m
zH^(PGcu1x0<t7bzyRqI+HO?&D*J@LreL{vJ*Bz|0nC<<q1%Xndeq#KR{vfM3+iuA2
zN9Wn^jfcryVy%&U>;0SOz-c6SL%C=D*7prZ5>#Hf6PX)n>so?Zx?zWC>yZqf1GB<#
z#*yFm7ID9k)dBBM8Nwin4(f_Q2Q0Ds^}&z*199d77NF55-;|L}7M92PU4N<LK2?lp
zSoC@&YN*KW(gQY7a5|m8xLIy<L)Mp<))s)lMjUd)iU97ZWF7mjCB)R5QTaDZRbT}6
zY?kZnM+>@iSyS1|53GqMIhaOdVYv9xM1d@b5;C}Tdu3rB9MEH;!`tiT28sd&BsP%g
zoUbd}i3ahP-yG@Bg>+LGBkcmQ!Y50D!#25yACAMv*^g{KFIiVkn_E+U&Jk#j3-%Y%
zJvTwXn8?8|F1h-y@%PN|>}27ccG}W?u!UW^`-ut7^vBl+It|99a&Ph5ZgFkMs%lq4
zpTGRW+1_@A9bYGK87+<5EjNFbwpn1n3ZhGfmSUQS1>aT6sW(fJnF;Cx%cYDG2eDb`
zW}XH3eOL2m)IEBfKmy-eSNKM%0US+=1c)WN$`oD$v<FaOLDNgA=Q3^^J2NEfCcXgC
zW`Y=v4wDtUPP<_u$**-VbMDN{A)8U5A+tOOJ^<oYs|DhKKS!nv^1C=gK88{)VJx3;
z-x3`{>mTStx_t}GZ)zT~13@K`Ctcjzd9Y=+5XX8zrT&}}a7UTm`gD`NW7dQyk$zYm
z#C`aR_h6iv%XvL()Vad&Z_Atsk<T&f-&^KWJw=xOr1Y%g^o%u-iD2_>rkPOp$8z7Q
zm{_-*F^oiyD7ou>K+uu+A{-(|%JnhdBYRpVi5}Aba;q8qYEk?s+!^J+bf0jo=K->t
z;qUIvqAym&4p|r(bQwSs(Q3GTd6*da>r=x1b;Al2mz-Ob=FLM6$da{SM`3}y(>R5S
z{%~0o4n!*i(g7r8-d{q%cRG5YgvId-{=t!Ylcao(<Z6x?sl?n>#S9FLOpLNjcfw%v
zngT2DX*@k%0u>wW$~vn$l@obznhY2w1HO8(-}}JnD+dfD(yvl^?8VIQ9R@+T!^|6)
zXIiUsVbj#`DWeJ$%HhI$bN;IP1tqaiI-%8Yo>Y5&c$f`&lMCqyt4u6m(jAs58fHzE
z*ApOJ{(_U;hMY89m}SAw56d%1eAZ<-(OSR2yWe-uNuV@mkpcUl*Xy0UUT=v-)lk?Q
z$=&O<@Sv=t?h<UDH}I-y5SRz9>^v_kPY_ZkQ3A`!mcJh7OM;k3Zq(N7b2%UI9_N(*
zmTi6~t0&(CgfhIV7q-g=+b>i|$-{PJ)6QAmc?z7@HJrVGFhxL&m8LCklMYNa9xAu1
zYi|kw5qwYYSb$c<m3`elUlZdr-MoO^T=j@oA8VaF_=9}8@~opxtVcpfJjJAfp)%{f
zql3K_NjmI8_mHFiwQaA_9R0S1GH4383;MH#g>v9Cpvz7|rBP7|Z{v#DD8z=)pVrkr
zG>pAgk}b!dJwJAilfN*&uiyZGylL?DXkfb;kefV7GwFx=h}*Bb5|8)ZahWsoCkwo2
zn)XilHhu{(>8xTlyUUsz#scz=fkW;0LfJ*0ys|&3|8^67!OcBk|5n7((-3qu6O;N&
zW8b6*8$Onge?;<-{9_=D;8ZZiB;GTsH72hY`1aNrl|>2zgciEj6{!t->|z(sF@Qhj
z|9Ane4`CqA25n9Pdo6>vGWO_iff0kBU2AnU&ukQ_GFi$UD71CWzFHnA-q-1>y%&nS
zWG%PZ10N|9ZU#C9&*Q6c9fy%VM`<TE=Yr6M@;tZkdVmW&CKWof)nxp#{hMM{g}S3&
zSeHiFS{WLQB@ArFT3E>tql5bCjCsSXhYQJMiSKAZCMPq6I{<+ZXjP0=E_x8KM?bPv
z&5bOo5f~+|??YP9Rs?0<19R*JSTFF4%#dPtHxNhuJlVe63lKJwOvmT$<Laa=Wu?|<
z<WkB&U#_;hNfSF=SNXJ*dXcBL&_ZCe)|;b{ulrjy`wTugMIf22dvbcx7iNH8AI;tq
z?3m_xFTv?|H9jZU0e8QajSGPya^X*Ml8l!!ag1HI`|Q#Xzib+{7<PA%L4Hq#l3CLg
zzJXdtn=XOx8yr@ajhlGBBXB;Z;Za|ak|`mCk4}sGLhVV6!0T_=R@U%|HW5JaF}jgE
z*v^}iG-$9RqZywaV_gKW_J>)`ZZ2)+A*DYSXh9-ZOa3NgVY%VsM_Jl%F+_qFq7mtS
z2w&<xe>bi^)GS^vOZV#HP}kS_-TC|Ki@JGC(6WSK*Zs5cUC%T>iyn*GsHcFdfr{Vx
z&-*8nG;DL$OrwRMXjs*vS6yC=*sT7B+(GYkt;%(>rsyG>Dq#Oe;%_l4aO^<k7-JCN
zH{E<51&lmdF34o+URZDMj+CAu1^4!=*zooqFC;VvNhG@&(BPqKfHlPW^};rzJg)jG
z6aNo)ZyrwN{=JWHQ?_JEhC-%nDmIm&BxHy}noOC=lsU7^<Cc(<5E(*LWS-}#C}hk$
z3q|I6?7N;$=bX-Yf8Krip6@@u>vvsU=WN^F+q0k7wAQ`u`(8TW84!qWxF_+U-0yUI
z4oR0_!)sp?J?--5=45~SM~gaI4@QAV*|Q##AF(o^)ZlHK<Xo4x>6%_N8GHAM(1m@D
z!EeRfTQjmVx3`y?*EaQO59Xk^W5cI7>_W2N64SF}%b;X$E_vRiX}&RWkX_4h#J81u
zQ-DcAv}r74NbUt4_&wZgcXI=FlAKTPw{08X+9FgwkbD0i+mjG}c`e#a8eKA;LwDO@
zARoGJJGWV%zZbm^Y#MKH>JTJUX<0E)jU?3Vxyn$nVz5qK*wC-h^Kv@b2RUt?VX;x<
zjMdCZJ$pp^rSH9pA`^T*VC7ksQ1r3dAgtO)THN)kkJ8|~S4b%K`#)Q@mOxS+>QKU*
zK|1+$+Qyg4d=2)q*|ZLorAtxIm>ux8pHjHiYu5K1xzj!l1ts|h6Y0bXr>W#iY2|Il
z)AycfnNY3<$mg(@`;ML1P#U2h<T!%^3>b%GkdVvrc|NS@8oo#Guw3igClnpID%RgK
ziXW1@7}K1#r5PPbF6jBiq~=+?+7JoG%Gjx*CT1F3v|GabmkCt-43Ba~#=je&imtEH
z6pvzIi=<?Ubgb5-Ir;t?t<1gSTRo}M68#@z?z}tdnSCNva#qi}_;WHT<^~$mzt^y7
za6qJ52b?o(w0fo=%W@3w4|mX_f9%-Q(1Gdk4wx$yu|E2?=;H_OMgv6(40?XLPs+W0
z74WDY&6bG)cDfU<1&$azAbX)9t8Q<0uQ(sX!A^(T7nG(5=VUOQ`!G`I(o@Z#I@jW~
zHZW&XZuVB&T3Y-ljCr#cbBq)-MS^LOyJ+m9WmQF_wPs9FILyCri7)WcY&f-01Omya
zI)8rVO~T(lpf@J9rI!%l(O}#r<-t1R?Ho(S$Jb0K8XPh3-hq1Go}6tO!Jb<Wb@ZeZ
z$3PYN*{&+B(&^^9xn8SIClP^@z8OJ8&xS{zLcilFH))Lyg?EpOgQl-6dN;_b8b~tP
zd9A<9OcK4CAuF~+;*{iiYbyyBPEDPhyzv)o1MquhuF>{vEA7Pgd)2i1L^36&w$0rk
zqb~FX=E#ZRMTUsB!k5WpBYIKu-TEO!)HZt@MZkHXN~&tGBFdv6G+%h?gVl_K_0^I-
zixQ^+>}_K4N^_y~yVeSqv=y~wRh`Bk?KHb@uQAVr+@*5mOTVEof4I^4Wf#w8_OTtB
z2tL-Vkjd*A&BD`Z4;=C?xi_7wld^N8s5U>5eq(8-Kh4t*WM$luc|w*9xfH>PA)F!o
zVuid`%<%(OicMN^!rmn+voZZk*2HGjj$(f9ABSpv-NceCS`WXFfjq3BORPUEC&=0J
zxU)t?O67`m9l6Zx!6(FZF3(dlia?b-P{y}VL$p@E!Jid!;zmEuB*7X*j_MaDL*^w^
z{uNd;m`w`)(6#)CZnSy+QQMM!OQ$5#j@EL=J&8{g94DKVmIE-gq?moHMj}(Ru3~ni
zU9I4MAi+l!%jr3JICjYRc!%7qy9j5mS7@aOx=Jc;xAeiLR*qlaqRSiW`}$*oQEq`-
zJZsDoFAKj$inmNN?XY3z0^)>Y$&|pH@OOT@IN@%b2cgGHlHZ{YHa-b7UwnN7H*d$j
z@FlNiaRJYWvwM=vl;+WIgMK@T&vr$sGEHwip2S;^u=liijF`>waqkDxlO}ziIO5CQ
zm&0G(!F=a;!Uyh9ev(F=cH3UJ`t;Pl_{x2UAea$N(l+;x&ez=~rp`9pz(C`8?S-{q
zv#3Z$!!|(%)yUcs63>h3KJ*iA{C9Dr*M9Q|(Wl(=cpRK3=?lbNbUJV0R}1-A;Fv+)
zbvz^l_YIhkdL~;J%rsqHxLoq(RCNCKiHj&lutHt;I!>$2+9)1Yb*o5C*>+(RfX$ny
zKaRcIomE2B7om$IGb?~HUnbuj{>Dj1ZxHPg$~hXGQ0Y@v)Vg%hl~6J>iQCB`5~RR8
zF7#|74a7cPXt{a^A-t-_!Mk$<wp<CP2?@f?Q^eKSyFD%U!qRn;`sA}7>@j=*+8EHh
zJHWcPjzvV=Lnl1OtEkbU*zFQ=oRyH=-Ku_iXIBsQxe>#Wx3BHd*Qm<b_#269pP&y1
zBx-u#C186kL|g1B$=Twie2kU1HNw1F?&n`DVF7MfrJIk!#3f;7C8WuV2$oRKkgV<X
z?fV>*I-V$6T}p`ov*(GIDYa?BKayk_3bv4Ql^oC!*Zxi7EbZq=VbQxZDCN%%RF*T@
z4OlYW{hyn=m}^DQ*IrB>Wzl#R^X!UD*&$#S3gL^_s^Q{a^!0dvarFrX9VZ}b8wnLr
z(p53wXTrJ)Wg8SAy*((i#E_1DbxnMq<?LWtQ?j{%9YZK{^@&*~Tz6X}t9i&)H-EeQ
zm^J&vsFp;%wwR{3b2v?R+geu8@+|_1ic`%Z?Y`dV?R)rZESvP37i%9bYf3*&FZUA+
zh|jt%e&lWSl*{npM%N>}KF!QBeW#9Rm*dzfLWXw2WbUN|uS#pY7%JOdY}h#%c<0E#
zGKCA<#8?qZSJ>SI8*rCnN#0Q51jpOgTe%rG911xXN-MLDevE7GzRYl*@m^#nw^*^&
zgGq{6Ny6E$BI>MdGFP+@Y=V23M(?-DOD&94IhPN`Z-1|Ha^a_SjE$dUreD|pPym5v
zWj&^H7@fL(aJ`<Gn}jbig>Qrk(wmV-MR&OWn*9cZP`=qu?6n^=PxXw?DP%tIq9)-5
zRVhh&m0JgcY#tw(+%{uL|FVGFyV0-*^L1(TRICXV8zrZ5v}*n7KsEN7Vw#6_U&bDK
zL=wpv9o1P$XQ(>#uoJy_PeK_LInZ1}_weJ6pBtfvM&Mf~LUpv=y8gA;i+3V*tf5-E
z%^B;+7CEiT%kH>TUVm3Vw1u02+2GO{Tk6b1VQF~J#<C7bYhD_E1QaK>eMk2<Tbotb
zv8hi!=j@>K;lR0N-hNwnQ13v#{>n+y&`ho%Gxp1bu4q!K&MS2<64B@-yECuxh`F!R
zmQ|$aulwMcvjr?<Z><NT$Bo3c9!Id#Mm(<zV^t2LpKF+)!{ebn*WgvOxeuZI$4Qoc
zBxtaP!|oL+s$ub39(4O|r;zhW+gNz-qV0Xf_vEFg)Pc4dADXp@JTx+f+#>81VkBxG
zN|cjX?=$yrK46@f`@!t_%?v-&ioPHeCHZnjZi-pFyR=3G*@WN+o(O{iU7`c6Hw2ga
zw5f<!1W8^MPL30m;#!M%Qviy%KHJax8EoU)Gg^#Vv_1*loAvCOPBDETw)mI~_cgOS
zoD5Bx)vvxR@*p@-`)S6}VtZkXYiDcMmGm%607-v~<Z<V2mQt-UrPFEHNgX&PAvRtm
zFm0^ele=_ZYUq4SeA*}K=hPqL78}-f`YX<eevlhwi~A^aHzOg|fXHJ50FXn4U*|O>
z4qgu_#58l@nu{#Qzq3G@Ka05D`xZ5AhrovXRsX5w_bvI)4E!Gt_{>Q7sw#%PNQ>n}
z_XiXbhdz|Ji*JBmxoVU*E6wLoQgXzt<`DU;ruC~7cqiW``UoCNco(H4Yv=21RW16n
zn|HL%$=I!w6pl#WkgCK#+POh_I=l##9((8Y4}XdMAzTf6PJh_#zcP17-v9Z}ljJ@s
zogj-bHm!6-nT#%B?+i%?hg8+*AawAW#p)HYg+Iz4a@&#+@}9Y#R0iLS-4iJbe@@3`
z)#C(2({3?Y*HS51Xvj#2`mqHeSM^z8_S4tp6tSY{-pM$3QF(|`Btv;Lf7cRxOWXW-
zV4eD#w{q$JFL!EQ3?E`_i-)P=X;;k4HzvNG6(2f8Xw~EZ60DME&k;+oR~!~4Gx_Dr
zbXkZzUu9O8fvT5I;}JrRB*SCbqAcPD(5a_pSra5-VU(45{K}Krx*+M>3u%Uo{loIF
z5tZW0B>-TmI4-{BjHX9gMQIitFc-ck_GgX>@1%Z!vI?Wi%;x-SoA@8Eo%#@-4;L6Y
zYgS(C05bt*Oq4O3!w`x8W5cMU!@V{j7msPl<f=g4%cn!wF-J~+Q~HOkhY6Gx-Bh9E
z{~V?N7&m$o!dM`M&Vva~?vcCqjwYiUxO3zi?9G3F=kp8VI~D?N!rv&AzkiQS4i-F}
zijTIjf5rHJ2K&GAU>_?oXq4qxp8w?*{Eznqp4{0FDmv{u!is++@A3Bdh@C~Xx@8%S
z=*C+`zZc0-FIU%Tp8jx!;+BNVE}aa#Oh5gr-U{OQl~tJZX+#}?)DH>8hoNISMM3`^
zxpgpO7%z%3&+n;b3*>G)F$uA`P2tqb)M}uuGj~Fy8(cSyqLL-l0w;p`diIc{INg;G
zDiolgFc>Q+-zVw`fiXMFAy$`(Q*k86(=w}vIrR=`Er11yiQ`z1P}Cj4E)5%BB3jjW
z6ZIa7_uxEnSMq+=XUfe13fE7P3VnN~>xh|MDBZUDhW7L>fMv+c=g3>XU93KEU4BGw
z{)GOUeahD`_0t_!Ij3-m%<EHITTeT;&Nz!s$2@-jvw7fIb%X_PSf{Ns@B2j6!DpQZ
zuaUD~pcFL%YjmL^NU)cS5!NS@jXN;#{0JyS_!vQhJpz*HS;tcH=W&%_f5Ql#U9nW0
zv62Vu5GOStigT7i5IoYUZb^n#lyRmfA720;AwPft1h@-_z0QsFI}EwjeDh7|HO=q^
z3!c*&kj|Wo$?y&&VFYs8H}=UF7iiH{-BrGtm-%&fODI2YYY_}gd>|;}ECpVkTQ^ve
z2^ExX)>13_yv0PP4ZI-8rdSvP%OeJ`t0@-mrNp>xk1?;AvdY%7eK|AvWDH7+7d_bQ
zW|gLEuY5>t?OjuO6%NEwgRT!tY6lOR)BVRH`Rnd^j+}=XGIEvtlJED)b^jQ?DykH$
zZ8X-s<0<fl;U4sy2tjPpG*ONa_4>glXN)Zf$7|uEJq7#Ba`^1ObZ4%Q26i!Qpz2n^
z-$_*GDY~l1p<+vuE|=vN<6GUrusTpQ(;Wu(S$eNkxDYKWYwAKttYMT1eg(}-YVrBy
zY3;AYRILz$GTDPgr4=~6@FVwSQtWFFkpslOaMJo)kG=7&__{Nv3)_*U!<5IDEimAq
z${%z#rDH;{AKm1~hIjB-^t8BT^)AxlZE%V%1mG2wvynXw7QQOOxD)o;_{-MT$OEyI
zA-A*4ikq({bEG8V{@bENp63&RUFWn&gY=92zu!X7tCFbWX4>1MbS3*KKz${NWz;Zu
zq*M*wm8Wmh-w(SF_IISZt35^(h_QONjyKtbOWDzFPKZ4ouj^;IL<!Vsjd%KhYuwT0
zpdH;LF-^<5fH+CA<LQXa7oR3xdk1Nuc^7`AOBmYF0k%t4Y81r9u_FK(-XRSIZylGw
zXJUlolH*aQn{P{OXv)MqXAjnL3`s7ph;WHuQqLQDUza+WBSN)9`_oaEMn+xf@TH(X
zMqL=c42_b}oRCrV%j^^c4}19L^6$)H=n(IS!|vM)u%`H4&==jhZ;LnlGBtf_7MErA
zoKFPZ!8(`<*I*1Aw0!FAiUG!Gx9_{nACSpEcF}zfP|_k_+kb8I`}!QNAlzLDaVON^
z*q<EhJT;F9H+4eCSGu$*Ag(f31p(GT(?m|QC_^j|Gv;A~dF!5$S-sb7+da`_bA7Gi
z>rCD5tV4y3N+7e#yl08lcK@U(oz?5puMb@3XVPBwwxQZ%eJ;&KL2PH)a|tY$E<14c
zTm<aNnkkYD!gibCjA#$L=P+g*rJ&gwLA=x|&8MI`Ke-X9>vynPFOY{8^tB;&)JlqV
zr3-HuEDW$QgkJ+NQCuQB%k5yYw(k8ki^7^*o)%E7^fXo7l^JM<vDdO}C4A}0?UygA
zA&l$+e@4=<aQ#aJaY@rzb%#lznr`R_%^zx#9$ucBYyyCMcPH-awoxP?mDwqd8zZ-=
zRqhp{L;=PpSivwSKe=<MnRtV-Ui@G%vzTb<=2U7dg~mf9-HxR=t@A9C`<66I$N64@
z74Ook6}lI9kiDk{b`cF)1^v4j<9KyxSAy-s>@+V$Cl0)!XXvWWnb>}c4KWi%j0zpb
zx4?PtdFaHK2w>asx@@+`y|=i%8JmY%huw)VvQ2qJQ7r#cLZI_w{+Ju{-kT3L4^6Ft
z?u`~Xq4uOJik)g%Imr<q8sQeQJimi{)}*(nDIy79d=^W)a#m@|zSwQo=Ez<YRoL!I
zhWkp|aCv0;vwC0alwxVZvi#QCkfl!y*XjoyKGj@gUlZtOyB>>6BtCiLg-TMOk&g_s
zYsWOl%vcTcZswZ9srRQ7yBf55Wtc|G+(w_biTV}KDcU6lJAat7QHa?Q+<39+5^m5o
zZ9``9-h`TPwo4UnyvOe@ar1q|xo)R?VWZi7FwVR!Z_i>*T)E-t34NA>M=ky%2K;jg
zc|K)_>EDvo^$6hy!oPS<e00CnBb+flAlUiHLCAOY{sTjB2O)*gGdvBPAI&4)E$Glz
za6Y739lUEv_u>|BeeKk1I_)pkEP`XIyExJ+0<5D@uW=$DV8}*_rkie)F%Eu+VJEBn
zM-9R=xHG4j>1@&Qy+x?QXYI|#T-ap3<iX(hXMnY(Lf5gag@<w5M>JelK8(N?L3*?q
zF|@>&5IX$|Dq^}Yu_(}=r;sE?`}Az#m_}0F%KOt}_vdrscDVb2iG?kd^uSZBJ&zW&
zmMa12XCYS(7gN}nK$iwJm{EuOdF+dQ67oPF9xk1x<PRsp)YSzs(9MGXnRrxNQSCXy
zMtWYAcYd};HHaDHJV3l>@4lD*a<X0ilmRG?;eQJ2?;Xm=7IMF&nbg4SroJrvIcYvs
zd*>bJS6y5@wKXZes84uD2%N}8^Y<<i!aNTt6kmcP)&2xWHTrZVAr%WY_yvQ(O2W{h
zIKOSTuD_FJ5;+nt*e^(2;+}B?r_*`oB?!o!qm1i%_!!@f!HMB?dh=6a8Tr@6hzi^D
zqnD1q3R5r~+X5adeJ98V-^^{(5qf-FfjTFN7+7R!GRo~Fxz!lMnM|$?Hm67|^qW}T
zLbSadXl@;~Js$)tjyx1TKad>4N`}@VKFOjZg?`*ix<gZapv|*ojZ}NxYfUDzq2R+y
zyAp0~_NypIu)8X%3>ImPxEAZg%DvV5;gwq-*TVkyB1?@{5EpGayrWt;l^zBbn}b$h
zR_mmh`1R@B`AUV7aW_HAr<cmu&Y>S-%a+?J&pp0N;px1Ac(Bo-T_7S{6};9tv_bON
zeFwL8GG(g^tW|oyP5q6GM6H!Ajq64g3T`h&<CIJip(b|QgJ2B@Q=x%Q<w^xfuF9tS
z?BsMtg<Cfy278IR+^opHUXT;LE&i12`#${p)c*T_{Z1g!t@Q|($@i1;0mnQN+yGoC
zgm}h{z~Bdwz7-s>ZRcwiS@MA}ZopSxjwdgS6pwO;wG$~XYt(vF;>%q+%HpSYX9e00
zlR!v*pVW%-09*Ztuqu}%+Ane(Y=C%jU6vsL+f3pPS4H|)6ddz^eVn^R*B5Q-i(99R
zHZL6U!{*C&HM_A%91%OpvfB%CHquw;pXq1EA?CdV)L+31HZbL(3p;&{>#@%DNO8<%
z!V7kL2pNC`u2xyxz!9M_>Li`yBYvyLAc73&+%ej-fWDI#<{#mKNJioGd?kUp)Fqre
z<&!Un*fCbaJnutwTF(0^TKbUyn!{3h=%wJ(qc{nwy`wTyHRsm)$EZhD-_|71Je*WL
zE?Mi_Z4kf_j>ZvY<6UK@C7|+RyCs!|8|Gl9RB;_+7@}_pu%E(IUrn0`C4NWK%0^qn
zb^RT6sx=%u&qzg<Oyt=p2?_W($ZMbDa?qI`91^7i_I*4o`I1}@$&}s^H2CJWBKpL>
z;|3J~6bU8N?Pr~d6uzeuwxjyK(Qo9+Vbxwq)xgXk#q~MB!+LHBmb3TYg!tr8#gaB9
z>YBZ(R_#l!8Sgv3OGtE!EuEL(O$0f1M#~hJ9b%bna~VgOWf;&xZ>o!4(7`Ej^-_RF
z*Ng>!^gXqr$O#tgv!J%Mwjmb7B#;v<2j+6rsv^1N^^-3<XOinrid{$mN0gK==v?AW
zT0+N-E6N2gb6ai62)l5}B>n84;;0g1!ITWUnRq{I)4%XZIklYc@<BWQmn-**-NCo=
z?%DC!pn3iQ;?Hx15$$tUVun<B36+<hsw`+-bph*(JlB18Lvi%LuYelUi<}{hm{M%>
z%luZ*(g`9>Rd>J0swZJ>;zkbcnRe&a8#&y(#6zl$)B<i0=D7ujyL9Z84*f}JvQH6T
zF#g4#NCBT0I`a9EsObhrXf-249qaf`A)Zn}mAZ+{C^ultIDu)%4Z0$Y81W+|3xzYe
z0UO2%p~h@4jIr0ER~Ksf+1?<&<5BgPsA1NqmGoqN`3_oz4(cXh9saEA{Gn>?Zqy7?
zHtY;dD?0qvO8q7&F;S}pTjvk24mb?i`}5D7^3>q>)bN(JPYK9O2w+VhVb6&+-Lar<
zc^~N#C4MX_x+HvE;p2g(+nW4W)%i~-HMmh9kftWTK>bph`h*m<j{uXtSAuc)$Alv1
zU@(y~B+51s6O(VDRaifuP)gTiZUArCBSn2#I$GSaCqgiirEW=Yv^dv9R#btKTEVLG
zj``Hn02`4Y!=qUSVhq=uA{%s!$hXYm&B00OrXjzaM{BJ?&%kb3@S-EQL5sv-E}I63
zne!dDZcRIz9LEz^^Grnk3ZsHc?{DQ6LXmc&8POQR;EozIKDo=S+^J4ty47O{IWmF5
zk`5yqq@ro3#BXY*P}ia?-*OOs`e2M-6UjT5?0tMYULn+tI_%9-vzS=aqrBwq=PDiQ
zIg6jiEfR*4cRtHJDvlYgcVF!_nrIek%}9BiO@B}Ps~PQ-bFnRxIse`Bv;JTQ;K;Ai
zHYjwZOa8`;|CLF3G)-ULxS2+oOUBKN#*=FNSBr*Hn{KLqjEx}ckZUvVbV(#h(O0W@
ze9nK)t|Q?05JPS#&Y;J#?B!Td^6iS4CYjAaAiT&hj62;kQl0&P=(hMnjW68;r;3Nh
zYy-v#i^(%~yYxI8&mRVieor9&3@CnhKp#t(S^D(HX~%*>db2004VqpOs3!6eRLeSC
zErL;ON>CkBl_JBbQZ_~sJ$kdcO(#V$q0AHY>=A7%IrWyIG*=W+)wW5{kj54qvM;sW
zS%^G>NPFpODRE<sCsn<@)u@->j(xGVb5l$p$hkyKB^pkYbKujsVY;b0`FWH9ZFE9W
zfhkOd2SQSD6N}g=n$p~&g@AeQYzdU{tPR74C8LgZW1BMnnoo~{!jOO9Fn2W96qOy5
z^?p5CAF7Veh=*rS&7q;|PLX!`64s`|l%BC_CiinwT%IS296@a}D{XSv9M$9s9_Q-a
zV{kj&?yEzad&T1hos1>Js~HB`_TFlqckpu6tQ#!SXG+wz?oZrlljfRX6G#n?Rk303
zo8=qoG9ADMX3|}~e3fwrop*2avxB-A@Q|R>MHxDel8im`!VP}ar?<-_?uV=or6uR%
z^mClVWg(`q0pdpFfW~GSB!dFMT77j^pSN<ZOg~rI5$)QS2ZSQQC_)cI*PQvsdpe;n
zf;Vq^{hf@ZE#tQ`F(B*u0eJaGQd2p9goUnBu<!XF1IQlW5ir#s!#pdQqYKfmgGyo{
zuYA>Ev3@2DR{9~-UuQ<i=~W&xm}Q<9&Yh<&$SO<|n6}Y{FwK%tzq8#-jeK8Wbj^&V
zTBSfA$T{fbQL0n;r`^*PBK^U>INy}65_4nOcK9BhLv0}y|I^@<z(OmUyWk;DoRb??
zaf7GgVc&U;V`G=|H8TS~dPKJaI+0h^BxR16AlW)3fcI@tN*OnK|EXvk`xl@#@wiyA
zn}$=@fNw{?-6;!}l&IUsKzxbM>Of+`b`^!m2>HVcAcWZ=S{hI2Rbi=*T1xR%TQ|)<
zs}-36Z|ZTA(Fz@9-T@xIbnT`nJ-Hb9uFK=r5C-`oj)~^j?OKEE04#w;l!EH#x(=no
zh2F1YxModqupT)iXTX-V8qC`}NAXWHor2`Jfu~@f;p2urwM*)}afK*B-RvB!aEPG1
z*l%-2r^B<8<*ifu-`X9WX&|iBvRt%?2~6*65bUoR-9uZDSD1n&{n#!Q7;DjgV?q-4
zg^=Srvw6+(lh2y9H1m>|rZ$sehEEQp&DzlzXu{TTB8QK*F~OW~I@r_gPNu{MXI}>A
z%@fo-w8ehS6Js6(<mmnstI2b~|1fgMbIC}&yi6_gX!V4~<}P=m+q4o>vw+gpjG1WL
zn#ERBaqp*gMKLDFd;x1CRe(*epNtYMY97OnHi|E<foU|LaUkyJ(+F>Po}2F3rSB-<
zC6XI7D7I`qmGCN!+r?aI%M$rD+|7lb>e4>>N>ZH5wdRr_@iCS!EFeV}P#%ipsKTQ{
zv8|p+)%_ixBBm^TERdvQq2#S5ybtFQDK01{L4Za1_!fON^JizN-F_ofK7zl`W^%f&
z92x~--E{Xe;Ay91g7NtJg3y5N^@gYJfHHZj8SFE}<BPPbd*fDXKCyd(8$vAXmuSd}
z4{c>@`*{!VYvt18m5L5y9}hTvzqp=I%;{-qV6EOqjsS7QekOC!dD<YSYmw;$QevF(
z$|`P<#IO81asegV^Eh?zt9hf#DU}KAJomBdwOP1XWmR3@#)Gf5UsM?GoQf2X56M`6
zzMIcBdW0g(wBqUQ0c?pyLBf;lr7f!4k6wRhUWKBH-Rk3`u=oK8nzkAZ;7nHVUR2A|
zvTlxV77|)uB0@Yl%n={?8pK|`QO~?@`@>aEXE1QvqQ|<r)0&B`yiA{rdU^oW`QgGt
z)I2l*L_<VjPisDQmy^|XtWvc6ez?nWs9g-(f5&(E@DZd4pHu;WCnSrE%tS^nYO7lZ
zduHHv3!pc>G>wq@iKKxGO^`m_P2;D0K)Bn3G3fIJT(QpANR@%VD`$q>{L8d-qn?G1
zW@O=)?1o}-<HpDnSbV7?exI8AU;6<B(nmx=V2(GBxcreaF&sR)ey5(*pHPYlN0S2h
zfS4EmWdJ!wpg#mTB!JpgYE98xr-FI)Lq60kjMe!=r1@GpM3s!cs}}pQ_V!1_c~({?
zbo|XzT2wVr7T4!>$eI`tN7TDoSxJz#&J`6-Ll2BQ(~qR}PRMdl)qhaYp3#dibZ)8u
znNV&fR3U^@kK9X3{kFePBV^W8Ri4+*Y6W<q=GmXRXU@a~pXwUG9=j0PtN1+8B0{qu
zLNo__<a}>RTi@vEOhsj;3q~o}h)KNqva#a^r*Wom;0KA?p^553^OKngeq5U){><e4
z?R~z&Q&POv9j&5%a1M|6Dwe&RE;8hT;m>P@i-+j$)^yT$50!u5O~3zmUk|^g{Z!U9
zjz326$^{+B#)2`9rak5su8Bo{=DilP4GHJoNacuhbz4{{B?jn^7L>2NdXukJ5vl(E
zG-NGOHbs4_H*7Mdx`gN|%|jYy`5p3F(+(L3(YOhv>RNZI8QIYqE~L-X3r0Q1M)h^K
z?I<}!KCfOf?leg#O63UxGmTb4b89fvHy3`x-S_&NtdoyD$~P<7b%PnS^FR_f47l`$
zW-BndY`pukW5_zZA(uO5k=uHq)yKb;n}_DGQFx~a?*x9K)mL##{wzcO)F3VGqeBTg
z=~U-}FevU`COCi%Dynx(5=r*5n+I%$mnmzGF9501QMif7s`yp|=eV&lxq5yInWX07
zqfAI|6b&6uqPFH(F$F6|q&G%btdZV8+ev}0vDP$|3sbvS{J?(xOjHXt>Oo<~t&Xb&
z4cLD3I5qoS=yW`?-K1yQ!%9%>`r=vmJ<R&d<{=o9FGNTC5J<bk<Vv~9D;1<##UHWO
znzm^t&%1Hzcu^~Nqn${8uyCs!qg`&uNZXpx$m-3EDVOvYy_QBxq2&&JT^%1TI9ZGw
z?4%V*I+E+sI~tI`)*4Rj+|74$u~jnM?)gu4Q#6M1f>rM$+wX`@B+IwWbdThNQ-J>2
zA0V^(Ldsi8*%V_zJq-`|qVk@?#zmbZjc3i`bqZri3;|ELn#Q`-$50ts&B21l9A}!R
zb<ilzXtQj`cSlod5#)g9^Fc8;7ZFGq6V<W<k!RFbIYZl<gogUTlR}){3JyT&bp7z{
zwuv{K68m#pI@dLq$43KsR~PvT-b$Aws_8E70OF%Hvg<jBCeD%XK2qPH_0ZW7L~(Zy
zoh3oN7*Z77xV5^RJO^b3hbI7+k3v)054kbSc42jiA)-^C`3s_X7IHjuSr3_Wq8O(Z
zWl4qr*wLJic3az2lSVnhS?XAFKC}N~yy2`nyF{NS8c{<>JW>ndo>?ly&KwxYDax|A
zbb-mP<Y;wch=uf-$Je9ESJ^sl3uh%Wm+eIrnLz{{y$!3i$0iktm1CVI9vdr9uqABU
zd%0|~Z`MKzUXc2+6L!>q#NgFXLJ!G-H6y&WABS?St*fK;g8xmFY`Kh_=(U4WKU6&b
zi9F~mLov+iFV+p;UJ{)mV!4ZQ*rHo5u%kua!4ZkuY;+#^?R=#_Y7UGJP4+7d02pCy
zm?K;OsDZSU5!xslPtF!i+L)))50OPe(Z0snU;tZ_)w65xnI-5judWYAsU}E9*6l)n
zDG|id*Mf;6)wIVWo;cO1M=Z^C2IMdbecf>d!NjC#2cbe{*xFw$D0N07ZaFe_Y#q?c
za|o7YJ@q1Gbv=_`*+Dh|y5@DWfW>mZHMpkVn{extc2;$l9QHJL1BspBEa1HhuL=fl
zUqsq99iD`BZ#;V$!P?sOo{~^ulo_d<BhQW-r$163<?E`%8nd6<Fs+OUu&uS2M@%r)
z-y0FEphDU7_H(CKog`Ynp8k<6^*RAnCeFW-3QsBD?Yf0Vp9cFIr0c`<GQ!g}h-&ks
zeB_KYP+s2!i==Z=C0}GU-670u*QFx$>exWq^<cB>*{I}PH?GJyMao^KdwXrQXH~lI
zsl{)CweH)Yvk811<DqT*_Fq(a>7*VyQ5~jTpiE*jYa+#tTprXZ18c$cHJ7u=B0ADy
zz8ops$J0aNgGF(IE^u3%E5cqyFiN@ts?axZ(vYShxmJnx)Ah{3Gk^<96Sg=#<sH3Y
zqx5CQ79p7;M9=Ft03jAkIOd93Xra4;>8Sw7p>;!_NwB2sdEDFNXYU+(9g1nguR<u_
z^ky!%Y+z;DII$zomd8f{?b<fC+=B%YSls(pH#0u1HJ-~KGAf$XBNICQF2%KVXuh8h
zDjK!WN#dwx@q!u@O%5S(T>s|ov6YwaC!^f9W^SB#xhkO4$7MbT5Mt4qf*4dJ)Mw8n
zJ=90qe8SEX7Uw1g(|5C};yv{b+1p>Gv~N$+pCQDsy$~Ep{v#b=<7*^|G)T32e`eC>
zczPjQH=#{8{5kJ{AuolgyYGMqu&XnB-IZZWUe>=wOkQN6wu<m{757)doc(l-I~kTj
zCzSYD<uczc>4evNk*IqW`I8AgJtW^jZPs=w(u1dwuV&wf5A|g<=C%pjt?HLLUfw#R
zxjNyyUTOmiOJ}+;<~0y39%x+DMw(rD1st!AC|`c2!<Lr`mCU;$7{a}VCwRM$*jz8=
zjB~le!iyB7+U=>`h^Tu6%uiKapck^xK-}%TNIq=~V>0(*v`8^u_8KZ+=li6ns`{Dt
z*u+imlMm}Yp!kTnOWVC{at7%ggYY#Afr{%D@X3p$1mB52RlZ7tA0fJmMqXi$#AVFi
zf!=cG<K>4dpwPI7+Sqw>#AMd>+$@9|y-rbE<Kd?pd2A6zm>0>ASCKjyVt5g$ckieO
z8Rn$*Va#uPp@+t0zG<iO437PMI(n%(SP$<2RLl!CzQvO#LmzdKv{kP(Wox#6ew2R>
zU}Y`$rjy!ySxS6(FS-<|)rqK+Hiw`!?Z}hHmm0%o<37gNTx>_mO9<+!lYp^2j^SF~
zJO_}blST*GqEU6WRmW8sBI-8>l8bITMani!rv{8w1MsO}$=pGdGeu{7mU*LIrqry_
z&WNueILPHA^{FseITD9V^gG_(z%C!hD3f5ySMo0(Z0%I2e{`7?V@PJ`#2q<)(Z~{i
zJhMfD)M;meWitg5h;!dc%)lL8ah5y#RO~5*Y!$cDH&*8Z@*LVpcf$|ueKhh%+o2MC
z9L7KLdC#<raKU@UcQf-f6coJW>j`W6%V#hWtE=Z35ZP?m4x+pm6pOCUvEDdlcUJnE
zOx#5=17(`)VY?>ymP7eagWlR<ok<FA^KU2aI4G|%31*FNL~c?aAnyH~9Bt8SdobrD
z18ji|6~qEF>vJK;%$E_&6myd-_d5O>(-daqEQ5{?+hCj9*}G;HL#Z9d(@pSu#d})!
zd?Hnloi#8_xM4}X9M~@d!DfX{f}^ESPcEDnFW+2!=DIJXydv0Hsmn&8Vn-0_H|2hg
zo92la%GJ9B&1FV0cHu*(XtiCe=Tw~XIl8M+?tZ1<+PfXwJ1aWHd1GDZ=HgX6&&xDF
zS&n#wU>}3$n{?!NCK3)s80)_46jv#fdXt>+>cIqhj5G<*WTHv0Qz}GGmw616c#?UY
zm2$()BVD^H-SGPwPhR1Rc&Upx)QkjrP@X4@lh0<uNK#fYd&Rkk-!f9h_f|7ndaF7*
zp&?~~cA~ue0E@9Efu*k34afG**-=8m0GxX<$?=TRJKa!Hj3$?s_q8IxBdtE;<AA%=
zcvDr}N1t}iOO1mR+*g|Q#n>+SSljs?K}p50LZh(voCzZv2?0L)*MLY-xfG1b;!av*
zd4)Un*u(#fgur{rFMb~HxX;mE@RX%aHe7g16#1eQ7cql=L`hjqh^MV4E`@HdJAAis
z%!|V=NSqOUj!KR7WZKy)+K-JZfa_6uIVw*gLlJlMMKf-P((^asU=Fja*R<HnlG}Q;
zqldf>OA&;f!@o+X*Fk!^@&mU(u`%^k^>^x`yg>I~e;$Dw1z+<g#w8u+eS(srixsz6
zlU&wA8Y`Z1hlK6hB<yX1o|eNaQLB>bDf_tCb1XjZP<q`0n-9u0R$P;x8FS^vGDYr~
znK)PAgQE;wk~#P^jm`$*;um(3zi@H-G~h0LCZ%e=rm~0aftJtZj-xrxKf4~rKhewR
zxhe+bC%T@t&TLPctyg_VV}&>6rQYR|wY48@iExJ|jVms8!ub(9xa_8Prh)5cnTfk(
z>H{N-xNevv_UBMGQOT5uA1aJP3Su+lsks@K$}Z7O95(HA)XWW0JxkJJQqo_C$UElr
zpJ%}F`2E0mvbiffrKH&`0H<0<i16m2(a!}zY+nYUpt2o%QG}2-qlNj|Zj6y=EbAFT
z-|Q{p-l4fk$o(rex}1JOG+_5e7-@-5!Y+M}SFP|T^x8NLH$2ygbfxw$>vE8qI-_Z`
zJ98*s@;L<vi1?s3JkSA-RFo`@;_E_Cf!;cDT;ds9mAD{z?H<)`{kEv-W}s%oV!8LS
zwVmG%BRq-r&8D5t?0zj!LZZqhAFiEUXQPmoMny-f8_9Sf&C?F+a|z;>sIS)Kde2Ut
zxG4ButK10qil>M$LnIe*5%~_AN6K*M%p*Q6B<E?kn-mQDvWHbM^+(;)c=c%Z=A+N!
z2&T<duYSRtQ?&=%>#ZjYE@$HIByLlp>|avHEOwiSg?Bu^p2Tb#^wFJ`n&(;p<-JT^
zlE`{G{ELn@TBE`BA^CWUY4~hOlG`$(NVB+*oa@5fL%<&|r1H`WIGV@PCdZV71#-Ha
z?1t>5ftY`>h%9MooZ~1-LA7kOKM(gtq8a7zxTK)F3A3x5@{vgMxX<*pCDL~Hv!8u2
zVOUmK_RMt{O={{ge}R3*`SqexrB`;%BC`p{*IY8_s55gk*lo&QIHLKbD@H=K_T7t}
zTVwCtQ|7p+87D(#R)pA0(p>UL!+5(=6iw+iO%kn!%eyY3&;(QLCI>%Vd8QRA*h{>!
zKlMq+!yCL=X5E)7RS{+e`C^OW7X4*ccg=4YgWIP0JI^b585>_!cfY!4n1|W6i?(~w
z={;qR7pu>go+VPl(<#|97Cg7(#>}(vN$Bfel(4@S<&daZ#}T34IYxwB`;&SHp+*=K
z>zuM5iF}_(|A<chhaWw!iUSqo>J{9(zw~D<VC!!6seAg}Pv-33e#Q3`xfY2Gymu9U
zBxT~-ga}!~3Va_68(t%wo)B`CMd5i(ARyM@lXADHlnP(`gYovCU++JjDMn~|1xdC~
z{`d<X=v(N1;1!J^2Ol67?k>|nc3?z^g<ty+A{GSk(PwUIf3Mj8#|Qp1^?(<|y8$z$
zv-!q}@27<aS|0U{n&Lxz*pa0aj!K>)P2;D&hdNQ}_P4$Nx|079arp7c4k@_HP_;Sq
z%LDWGLycY@;mlynHRb*<ZrP9D@HQLVJ_Qz<KL|fRUfi!d-iLuYF!96u^gr>FKO@{<
zdmw=H3cx(Lci~@tfjtOq*ZZP+*H20*|HJ(L^EP@$p=D93)Uwj-e*i$gcHMtYbnlBW
zgjGqeOn)pP?i2O|Ty0ALVwKMG75P}MU;S_b<kLKpQVai_o?z&Sn52TaDd!<ZK$2qk
z-X3|#?>5~~N!<<8yawS)B0-5~MdV+$pa0u;@Ro%?bKd8b+@BXM??`9?jA1KC()>U@
zc(xQ6B_p21htYAn%PE{9l#-G9gb{NUU)>ympy0<xQ=I#&)4W3w{?_K>V;BCK<HnSZ
z3$N*FkiNBudoa>}+84>JbfZj?O}=zRcyjX1<hJ`EEgVM>R!gO}yvol8@z2qO2cG8<
znkP$j=d(ZOIYt8?$t${{7|elU^g3p>y--(zvij1UC23`+dF@wc-kxBXY6Nx4Cw^B;
zAj?A>Iikr9Z2klT{>M;Up@s!xivPm+`_<`Wl5Eddx|5F8g-~RaZ{BuXY7(kJXbZ?N
z`v6-^7`ij#BjiEmup>K=gpyQ1OT=?b?&~$oB1}mm*W)&Cr~=_)N5}-QI4AYCCFnrv
zgHQ*uT!E-czD<RjM!NsH19PkT<3q(+rzq$Gaj0&hg0kzn;yD_C$6tv@B}d(UIS;PT
zoy~wS%!5CAu|Rq54)BO>(VsGu-|SgOK$aRPTz9UYgATiQ%Vn<%g3TRtmo`|3Zhn*L
zhRE>wH@nW-#Ks}qZCiGf3rv)1&;foP--Oq3%x_W0t^1Vh(}F-!#2w%@t4Oe;&xx?u
z+y!DCy9lszv<@fjAVe;kNN6Wk1I!_W4xW%awt+YK@i3hQ(q69pM6=k}(z*!2_a_^;
zJw?WAAgW+E?6y8K4+-WLHa!pE@ja1QyyHQ0z~j(Fnpsvuo*EwV#$DxwN{7%pHU`-m
zc<Y$STZ?9Q{~=zt1=68?TfVe6xwe5OMc%>NFGT(Pco+d=uP=1pWdU-|DYspq-%TPt
z50A(rRKe^>qgbcXS!!N9^4ti7vpkc~A^O+k{`o{TMmAh(@}2Lm@(evc4C6+^8Do2G
z9Xiv~fc$fXQr|X<ew0H>27$M_U6z_^fVM6Ye-JKkmj(;`d@G=U7J2NeIyj2;!Qr};
zZ<;(4kJ{atbN8QvHfTI1XW*)9f`kS?^-=LzK#ryhuO|b%$g&VRgKOZV2XO>dAHV`&
z#4F8A3fF`bWUK%b^Z_wVwFB$w<|i4ksnLS*aCNN9XF2V?=ug=r_fNe?L&@Y-AW=%B
z-d;NSSWt`d;HM2iJH!#_R~k-w@NHk%^~(o?&rHzM(I&4`GVQ{r>=gqRphl)_6cy?Y
zH=FpP4`o^msm>y@`u!2uh2^qTA+gM;gc-U4xyPRT?}2UpxlJJs_ZCUItpEqz^Ulj7
ze&0wB;uloDJ<jV7Ll%e)ZjAj#s6taxv5!{3L>4GHi`a9G02g}EyrT@+HFN`c@{NW4
zm?46_*$vG$OG|a7nF?WF(~Nfp>g|?;YE}rpIke|PNO5Z}hU}>TWP(Tl>0-UP5)d(c
zRZ)ku-F^hnhrQ4UC#7o=Sk<S}3I?4=>cn=lCWD;ix4hZ`xjl@e!E`5RU)QZ5jZZV8
z!;D@&#O<Z3yE6+!7%CvqrOjuBxHkHh<c6|k-M+g0yZq|UBjhI!C?BA%to#w9TwuXV
zs{!hfvIP^loVzNUq?z9bk+cO?Dal*5&|}rF87TJzzey6_MJFOv7(U=G6+IKGwZMJI
zl*JJV`WUO$>Wd|$xri4=q)oFLOC87qT5Ac{-t+ypQ(w_$ZUW!1x#$XV)5#(ERcAj`
zYmFva8*+DV7OO+FrFiF=M~`v8AF!@t!gih=#j!`#dZQ_*GzMS~(M5<t$M2%|Ezx#s
zxosE}49bY$hLLTHZkdDXuva0{*!Mjisf}u>q?!LZ8EFuX{U`zVgexRvcE|4^v_N)>
z&!iG(-W`w&-~(@RZ(UZHCECj_ztIM(LsObIlOV;kb|zTrBVdF^uDxDOr!ik|CrfN#
zh0wr}EfYH^IqZkx^+m9Io$-*xT6}h=CO=q2wM6&L&Q3i~#OJ0Wn;;kujwhRTRmDH@
z3P%gv#Yu@mj&RlLSMybo!k3(N?r~wErbM&Bp_t_TpiQFRsUhfd@n>qPF(6YiV2eMo
z(c!%ZbS+kNiq^mj^+5P8qOIYgF;pdN+%v^7+6EmZyQY<d=>l;dAKQC!g)WX4LWQxL
zD7C|!1eJ4S1%x$!jOWiD#Ixf?$N$hT&Xcgw+?2mcTkf5L>H!5Gni*D>D5cWy9XoTr
ztE&7C$=3c%_BhO{X+S4;%eT0?wN|sWKPLZ6vn5}s27l`9)><#?CV{T|-xxK4rf!h|
zc!_Ftfk@Nk=*xGZqLEN!9E9K#{G>LmgxU&Q3J+)%p3o^gpo6k+yRn5u^r{URwShD>
zr(k~cR=4qJ?mCe31o9UQfbJB8d4CVXX4<>!&AiPL)qOH*yd>H2bUi4n4}t(>Z|L?_
z45q_WSZs$c)wwmT_BB90=c4MNbCY7uX*z{|x~5|~rBDlsT3zs-j(lBWERlN}<O9ZD
zRmu7|3VUez9h#DtrWI*dX*^n$Y_lXLf0|x_FmkAWv9C5bb_f{ZVo}fe4qH{?wQ?U2
z3_fKV6slhjPV1T(xW2S=&NmkgBFZdw&!%x@KwUxznyPy%lr%{^T}qz4qOw>;nA*yH
z^g<3CR2MgH)qv=OvP;c~iPthcbSUne^0)3-zcgY@cxtFzHI+Y22tE^9J$}b*VF~4-
zlMa%BABqqn3|^{}9dWv^sb*TkNUz<SgGSt?KIeQ2qgA-G9(SkDN9Ha75!Z}wK@*V)
z^xpxyPVV3AAGtxN^PzpJDK*s^k4yBq^P!!r?l*i_sNc$4okpHY?<xi9J=IshR4t!~
z&g%A;7ED_o89ZHde`kAA44gu<k|OZK>3eoRE4ugc%voQ)%6C<JeM+U`kd=GjvvM|4
zkB>xwK7~O7Rq^Upjc<JIWYXqFdTCQ_dpsVAa}!qwmtG;hhp$X;O>qC&o6<*O!FAaz
z21>cSuVt9LjY5ekZTH1>oTFCogEi78Rsdp!;Lf!@30k4F+`M|484RQzjEdbL{vj|3
z@~uxOUWWzjTvm7PB(N>R5=jmZ;B#Ul0=-D&Yk)#J(|J+>^$F`07~C%S%ZI8AgqZfW
zxqnEdfSua^WYmz3b-x6Etpd8jD&?MB8&ZMOY)6R7d6W1;HRuv6qCl2cntY~4vH40W
znt21dcz7(<qc>?Hs$%e~MRT%g8;f-cfl)~?tJHvRq4`I5uwqI+CK*OxVx<HUH~ki8
zDJbVM;xD%Sv;-eBgWjgZuoX*6*@Jv2xqYCFXVw|muZ@|YR}jdbtgwoK-gphKZ-ICb
z#BO3&l+p8)eynXZuTEigscfomf!vJ@VWKa!mRxD&X7mGgd+ysN>juw!Jl$T11*c{c
zmSzp142f=ls!{RlygOHW;Wk^xxIrhJx-@xO4gnSm>3t@&bn3Ma7%U2(mbmXC?P!&|
zjl!|sYWn;3qiESWe+a~X7OQw-Av5YGI>Piva9a~p37Z=o?12d$#*O{een*odt6#OV
z9+43D<0EM>nAq)<Hn2-!CbdRyLM~xga7aaF8;zp6hjeDpl=c@W1K{}mtI-*V*_fKE
z?lII2L05T&I<_!=hj$}TbK5t;b{dIx-JTR#lc~Ns3#AEsiAsZTH4M@RT^9;+rcXk~
ztC|G+xLKBOLL7OhFR+{?OiDx*gZlUc*cqLV*=D6$0-m>O%sM;07yG)6y&=UPPIvaV
zuP|J7uijd_lvD~d@>uc}<m&<>*I3w=h0p^OjKN(^$)u0A>2c1lpKa2h5FLdGIJ$AX
zAk)H1cT(2fKRD*qep#6}>jCbW6bVX>%GCeYvLA6`$j|LjNdw=Hc)or#3%O*ZpWVs+
zBE6>;0K_{qSQUPJeRBXBn}5Wb6JJ6(Lbm7gNcrYlX}{Era%Uc#GYzU7v?Z}~P}>#$
z0=XGednO1jHWv!<-ur>#fgQqi(f9jIEmfy#_OdDy#esoQ!At)_3m|_{@RrVdsZYpF
zYJ(hL)}Ai38^V=XK{~h5lWRzk^f>CH)VsNj$vBFPirr$;%}gyEJ=KINN{VmQ{YwmS
z=@IFog1{#nI2PG%Y4MY6^}i|450Yw0<B{w|UkZP&CsB0#4EoZQxiS{DEAp~eJc&q1
z(a$|tSoXAN;+@aFM4MEx=2W4suc)||lzwKi@yyrFo%fF)lYjW2IR4VT%=bf8sXpn=
z>idO!7HUt#;=b-e@ZTbR-$3`UIb?8cG;^YD_?0iNrwA`1v0_O1_}M;*oTgkt4<2LN
zU)ho2HSu!D0FYdY_@3xis<L{}n<IAyVZtN8yBv`=lt=U;!8n=N5buGxe}uZlnW0_V
z-dvN5Ez8GHx70QBUqfr1p+m^wU*Mxk^FRz>B|K8+xXq3fXbU~>o7|8&=#SfP>Nm7A
z>rfNow#;6z#}N^CY1;R2XaMPJ1X}wz9Ct_KgB{lDo1JY#(osl^&ahOc-htIF=Au3X
z>X76csP#5EUZPlkrPn7(-I_(W($tAR9z^=wLnmRkmgVPA&+d|b3X-Abn`xT@n#N+8
zwDtCN=!=1UBo5FciIGb0@rRO88_HvPBk6Rc9^SGlKBh@T9_VYj^1ZCC;b){`YlT3s
zrE`##=_j?yANPb$evsb|Txk>dWB&S~UB2iulOkd%M}P`0y>`bnWiIA}y~sX!6pr#f
zk7dt2+65+w(qsMt)dx=*`W(HU>{C!W&Z89J3#2SZ!axYlS_&jDlqyd^m^f|)63P<U
zIZm;Si^u^Hjnz*bfCc~NwVUd)zPd)f#cO=#K0Hf^d9m-=w}&=63!-Ly1xC-O&Fa22
zZ%ewUFgZq%&t#aYlTVNB!(Rtx5*hREynhW$Y36Rx{KhIq$PeB_6Kes@#lE54PG3*(
zesQWy7(ukHwxYX*Ky)Wio}}dY?ZUm_nILB8Ppk#bsv^XkX`G9PWti-4R3_-2?FvCz
zQGUM17o3ZFz>^3oa4QoOn1!l$;Id35d{^Fs=he|C;N|6}fnSLc`Q1YPCqe!9U$qbb
z6mZw$ncR1*M~9nIywM)834aj#^uIQOd^et$^!OcB>Y6EfWqhRNHX?!afeDs~bT+(v
zco|+&$x{kg$z@&W+rE_92pP)m6Yhp;Z4mv;2WoFu^K?;lf|l*nfdbMu*-Qz%{UA!D
zpATGXIJ=FIdlvl<4FN81L6^$|4kP7c^Jj-P-eXejPZ_<py}Z6LxK`dp6l&n6t3+{?
zNl!Wr<u**fJa5EeudjZy)<bpU?5B9|L(Re3lti;|Td*uhKXn3ivSx7Say!(@xNu&C
zWm|H6rDSO5J(0hv@qO}CZ!}rfCxklhkLY_wh`W#s8=HU_#Z`l2(XjH~u`-uxO1b;m
z8}C%A5#7$vA!mdbrFvY7f4x0!w}Z$C@{|x8M@SrE)~YM&UiqXb#?5*ufQ^U`sn_|@
z2lGqZgdVB{jb{l%MM(XLMDms2ywu*+JR=hobbmd@!LFS~PdJTyf@$Q<qV5T91Cv1W
z+yJ3A{)N|q<)l2hSG+lP-x!sSkXg>ZdK$<r`xPhx_tk@sm{S#)wFj-pw&A(-K+6<F
zrKg5!lnMVCx$H*-bs2$rW34{xeGh8rUGW|It^L(?)q#xRO4|DNA!mulr&bweGLTJK
zZc`k$SJk=ORWdkEJG0JxvpjVRgC<ABrB)zG@<YytIA0nuDJ4}-DKk6JUzdD1>*J0~
z3cq>##T)%Beh0|Bgy1enso$N++le_u*F%?ID_x5lG9XWNDm|W%0cChBH|RyM^hfsL
zSg4pSKt(gspuKzNnXjmn4Web;IamRLyc#$#W<w+*{gp%rHE{>CR#j}#A036bS@p^&
zj=T8KoM5azlRwQ7JP1HQ$p;}Rdvx6@q><u`qnYxujI7%rR$tL@sS_wYJNH_-2H_b6
zPEfj+PoA4Qx~In109q8|)Q=xY2fZ7+JNXDZwvZWPrdR?<L($5c1Wp8TAU&34k3;Vd
zJ>18suR9O2QxY^cRXgi#ga)m-9!a;x>L<m5gUqjG#wKn{qapu|Ss<jZeYDJT71PPb
ztKZrzVf~0iVe@ntJK%!o-X;%F+~IYjv_y+hIj|Ksp7N)zZUJm7|F_prX`e|O4i=|w
zWpBo4mM*?KKElqtA&XcYXo_=@dW4pX{t3+fEPE3tbVlkD8MU@$-vfpFC-G{Haf98-
zw4@HIx{}X*=gB>pBu>^UnH_D&D?|<ogmwK^GC2PJD@!T`*6`};gR@W{x$C$ytv%;e
zf&@LKkU%UAZJ=0=Nj~she-Se_(A{Ydk|mz9`&(3p_JfJ}LI^gUjS>Ae(*;L=2l6tq
z*XF8Av8<nj$HT5`RQgl$3Fm)#{gPREr2F>io9g!`$2dF5sVb^Q0ytym*Tppihg_D<
zctZx%@>)xKvn^tqc66_)D~`sxb_^I(6Ko3Hdqh{e^|4W71M8B|7E>&3H7k!>oj2l4
ztuulIA(#_AW&8r;th9$8ii#Cu;+gsAOgaJ?)*vPrbBuA`SEyQ<P4+IZbd}B8W!T6)
zbi&9mT1yo|>9nELLfB7R^<(NCyEYoHW*CdI(ACt~D(g1dK7t$~TU*s;Zi|q8Liy3b
z!MPU>-%r0!G>JRg)(Lygr227c=2`-EUBwvT4@6JVCyTHXF`rU?b+Dx38IUkR@?zKb
z6<bMv#N+uVI8E$y4=Q)VoI67o{R$FUtlOx@0TdfQL#<#X6{)T+sU~+5YpK5J^|ZlP
z+3$`$3g3dRX*}c$*U$PJ?_E$okf@{?XiR>UHJpg*qwZ@8jOtA4{!;0R%xr&0%GS~7
z!R2JfZn1@jnR){mG2Q}=R7<S2>Hy{roMh{yqQYejoS3x2l_^8;mh9lg)K`0lES!<*
z#(wPfh$bRh^fRco9})jj*IBR1T&p<9gvZ=TlPVSF^cCvpEJsA0<|5N6Ur|{Mh8nSm
zyw=kl%qrs<nf*xlQg0;l;k1xu6>sv6eKU(Wjqg)5Co=#m$$hv8W18QnRTtPbrBvRq
z#RkUlv6D@<pTJl3i8qUWB6fK0cl+lb_qG4_*cJggIVFU7|5o}4AkpD?XzokJ$YvY%
zi!lhLryG)pPzVhCy+q?mS?SC|yQil*+B_~rA(|HHdn)SE#`C`)yuU;8lwO1$mx%uM
zYCabH_7p`53Xys7nl(BtN7jD#uRGu1CPdJ7dgJ568it|uH*cI=zqgzIvGDLeKK9f?
zuAP)e^Z4Ix!#PUvVF%#g@y|vgMjHr*s0`YsOxW*y;UExt|M^>Y<NtPVe|&!V5}Yeq
z0iT`!mMnkztrF6OLx@Shn{R!A5TCr7q>B%$;k<1>ozDBW@BNP&#g9L>P(YBf^)&hJ
z&$qk(`dghagaZGLD!%Tos>wgJGC%)OTnxB6U;LwHfAyCCV;nrMVxS(xCF61U?=$gN
zZvf)%1|>@7+r{5J^Z)A?{qG6Mg~t#?Co<wWm;PGP_#gLUSrUfW?fK`!e;eW-<Mf|D
zrZ<4yE%*-Mr-fht(tms9{~sm+-xa)J2!yG@z$WjioAkLKJaPZ8#`HpXB_!O3X*F$I
zJw<Kf@hgFPh7|!}KY=Tr6ZUVb{BQU0$3^uYkM)DV1cI9~Ss_}mMThJLst_p4z7<gI
z_<#IRWMcsxe7WZyQ@Bz<p?DBu+W*H7MGAA|r-wa$|4>gqjo5#=v?~dbrAZj<Qda!0
zRz~GK!U}gByUhACFPnet2S5B6SqWRqlgiKxm4EksdM#L2<d*$MKmPhi|M*JkNZ1&t
zPcN=a{rK#!KdzC3ePJZ%tgpte)fE1*Md{!p^c;Mda;;yiN&NHuj>y(tYMt`n-;KY#
z4&1)W!}VRM|L*<!c;RUB7n(hA=U?rMo+R>c`$!l!obCSI`$r(eAD({`Bk}K6J)b9X
z`#jmTGydiKOA!+5VBz)pUrb5<vCU!l;P!D^eVqRA@80hh0kgM;!C`OFzgxWjA6fj`
zD^qIWz5qW6f}xGbu8N&z${3)g5|L2ytpUMx&*y6~zi$up#Du$0g7-^_a}(wUv_v;l
zn)V>EWs(X#DPdUp8wu_tvYN)nrv_>uf=gLZE&r2S^JBR7&&RL@A$#Ec<(DcDf&14W
z)osL84XMSmH3EbUK!l^DD!{ijP-Ue9S#0o2_2}TSI;5UD4;8o~<_2tuS>#EBOT#&z
z44a~8z|<x9%?Q9bY}{af6*TmnoWdtUm<|c86;gVKm=j|L3X2+QD_>iAYWi2U((0dp
z)V-Jce*M`thXV72ZP&4dEd=M}bKP8?2Uv1r8p(fjfzR}L{UxFybezfSss=*BN^sW{
zK2<%mJz!PTH*=|2?KLHkFlThFb0{5S`agu7WmHsg_wR=W=>}=g0i?T2N~8n?L278w
z8A7^6kPeX?k&y0oK)MkaQV=AiON61Dd&cLv|F!OYaqlY@Yr&d0=j`9!-~HL&ZNmXk
zW{jpf8AZbgftRx3Kouxe4&VkC${d%OZ|(qxBxP2Xy5Ap~l2s$ndY1sB3DE0R6P5K#
z|9{<8v7DG3{Ge+<(W7w&8qcW%llY#18feu^{UTODQvq#Ccra<8o08SNGLN>c+W;7g
zi6d^3TYv{7LF+5`Erg!k0JE+&sOIx^pe5w4$J8IQT;`Vsyf+@SVX<lz?&%eJ=#$pm
zLk|rNbfZswi?gQb_W%Gq#U};W)xTL*X1RPe15EN<n;EjEv)&Ep22r#frc{NY(dypd
zs+Pm2=&`Nt$r6KZB|N4{iUut$^h%0<#yklMw?1fJJ<O#3H~)$G71&%3+(mgaV2c{S
zI4uG_WQcOe`2%!lgYT1Vt}U1@9;WIGdEI`JUIPC4pV@FYx+$QQAztyi3g9X=syL@#
zzOJa;AAe67q1e7%%Fj0IhX%O10cn8rYzNR2nj2!!Eigf#(w{mBGMm?Ba9IjGBj#ma
zb^$ub_-M(q-&^@W=|aWSz}W*I?FeL*VrWF)^j>21>%Jh(O5T#-KW9Hc)=oRSbBsZ6
z9L!59?jJWWm{f>h3G!(KxRLwM(XMXQz}FC_Od^%<27S=ZdJBLnLtW@A&Z7B6dng@?
zgpKuQ-l*g%5iM4{ZR0WT|0@d931F+4fzHuFB3|aI?GEs$uI4tP*>^Y^Tl)u*4$)e5
z0&TIK=@#JI;}81}71h-a7!y83D_yXgVt0zfIlxhpicDn1Yu>H@0B{#;0oHNTvD|E_
z2Y&DeFb+brMNI@CcJklxUx(vi$|8nuP8-^pQZ4+PNJuAZE^mPzZgz~oB7hr20L|p-
zY|R<8W6uEq-zW%LDf)=dyLJy&3+mMXO;FzebZf>6P>BhlSV)xxwA&p5@NPjvcH%Ie
z*mQ~^d9C|5qj%Q{?$*NkI{!}boD|UdA$n&sP`fcJ(C^BJ?V|;fRoop;UZ`q}Yydm&
z6418j1~{OJM*fqus-TlHevs^+>h6DTVwJ5$@#nx@&Wt8N{*N2MHj>&p!?Q1zev!rx
zU={)smCf(^yKf@T(YArj2Y4s}%ns*6Awmbo?ta9v4@5|7|82S_0E3+RVeu?=ul3n1
zS@HI$X9_|v$jY3Pr_gZH?*=ouPb~tHAdn(RLq+L1Cif1@<!|HlgueinmfS$B3k)v#
zII1X^U|*u`id6g2Bh(OCYz$UH45y;nMjVU_09TXEU7qOvzfPzB<5x%v8J!!(Z*)Zq
zJQFTrV1v8e_FJCo{@`nMpG}#iZLONyvk=mN9KI!>k0uH+Bn9ryX`J$DCcwS%mLZV|
zYbcazge_VI=v$b5Z|I@W3`0k*HdV%K)qX{Z)N3+HVXIix2P$vyG2AY^t;$%i-`t(L
z+->ZKv67&E=PaOgK^o_o%58vefW#PP0Deh}s_}rf`L$}W>K>`l8~h`n(%!u#qm=|F
zr@I2KkAwk_9f3JwXVi+i!lm8}0N6YLLD}-K&`<h-046@Drd-)qQ;woHptmXMas!YR
z)RaL!z*>t&&w0T4?`YY4-*38IR_i*B?>9Nt%~kuIzpfD!Jx<z8#yIrq4g0ePV=hgP
zE)C8kV4u;7-JCmMgc5AN-?wmM>{=MGtp@S|ngxv@K>;&nZ^ga?!W-ajCjzz`W^|JV
z9bj4Jh=hl}q_0O!deaf264{CFF0${oX(Atj%C!)I*-!NDs^!oYTFh#ipzrLX=>J=c
zxAO)eZiZ#B*Nrj%KkCg^LPsPaD5g{8K_9EADc;E{*{z#`bwr|vadUSUleK&pEOOBc
zxJIe05Q6=Wn*0QSG#Z}>BPnCVS^*AThep63YQBpy^cbJ9W>(Zuy>aGo?I0UT7>dbz
zUeAxfRJd-P64;1f=B}-oPVOl<h8`%403NpH-UO6r)b!yS>3=m=6v4Aaoh;d3yJ=fu
zltsoK1s21mx}ARXshGd+RnnFPj4vgg=igqV!8;(ABAE44Ysc2d7rYES6`Y@r0lCsG
zI><(lzlP!W1K=n_jV#RyhbkTLmu7xX_3mO3pbx-~B8?g?uK!`}fman|#IBxK0NVcN
z@rK>x*6#tn6f%PkA!p_g1AsJRW7|a4aK7_<E)1A-b!g|z01Vc?<TXFeoa=F!x|*PV
zAcc^NMD7npb>!e{IbyI{M*&fFrlH{mQ0G@tcPs)=440xev|ap?T;4M@fce!FsX(=S
z{6<>U*xf%nd}9QS^&vxMVC9(|qGi(*&;j5OO1K|%){JEY3O=pN#wizLgW77UAimRU
zp8*TPB7hFvxtAQ;+rjA>;P6I6D}{#O3XBNrc->>3Y4M|vpI0aj0gxGbBOF?iPGCS>
zyd#baMR(@Q0it2Y6*2Y$sH;Mv)Ic#M$qBxCr@#x!(B{|0U2n>teNziO1~MA#oq9B#
z=$t85YfW*Am~$qad@mUm8aeeX+Z`&RWs|&b%rFl?FE4p}-mpy97@9Bg0Z}`o-79`X
zlj>S;Q%`}lP{1Ak?_AX6w}__b>Sz7GB|Py=hxELkCD8B<C(9L$)JVgkrAPUi(-1wt
zfxitSQTc*4vWiheLr-@jpuYe|Y3B8Ml@6AUZji}?lpQiL=yq1n1GhJ?%w{%*fHB;^
zD5d*rS3nPA&nQWQ0o{Y|sAaiL`TF%z`-PBXXk-EvW-MKsS2pw)z5wGMvoRUVn%?Y_
zdcG*-S&MvcUT{TC-~DBni+_ft!G?sJVvtQIu9G!zMjeHS9ny5kUH@J4F**(v4fzT?
zq_xVunbQG9ify4+-E5O5ivg&U-t?aIb&l4Gxub5K;QJ_i?2&v9-dW<V#B9eib@dE7
z3uBJ{TB%^k9x*|soP~ktDWIcD76jP9T{e@kps|$ADgbu;d88(R8rPNQyyapkQJBFz
zM|(f;9&$GUsO$HoybM>W{k>D~1=&xv`(wHRtS=KxjSb{KzBu{|Fy9B~as6QJPiv-G
zeM<rH@6Vsmhqr(}$SQwRWyxrXF9;aq8jRK9SM{CRreFA6We`!oU!^~(kjDl>ZYBZG
z;b=eQATOEMtId-nbsNSc68fVk{w1}7YjEsj+@#Iq!^ph#|3>#5OU!yVbob75ZWS*u
zA8-LHvfTGrdJ7D9f4U`)N9da@Id&*H`sm)p`_gVaJ9Gy2M{LiuBQ3KJuYiWl#*s-Z
zfi<?xhnw!?<y1@dF+z-#xz@4!$gcqZY}W{60a)t`Xz@Nh;`_87TL;}4aQLsu80-ND
zb{<8zRq&SJCMnI^iM9cmjeI(PdfYTpg(K2Gi|{TX`M&ve^btX|I6z9y^-%B&&Reu&
zj;U@FFzsP3sfdTRzc|z|lnxXXFHRmbVp73iP15C>Q2^{fFAxzI71RlUdnPR&I}s4E
zgXz!+=u-^VcM@RL>aj2}74QNaBK!L-)_lUm3`0n0c}pWebYPmE9>_!2O*dwK!YjDo
z`Suh|$eM*CG-SPREQ0#%?M6W=L<lXJ^byZ=y}ulZOU4WZ8ePl+c5WZfbpLT`5LwC`
z=kwdMb{4wW7%W$h!kPQ&OZa7gHL;xf+hWE}GHW*Zywyd(>v?HcK_v+PH@83Fbr0@;
zlkRSui{?68byQrFyrIhjUT5QMgBN2F@>D=VU*%JLoaD*`WUlg!QG$_&L3P98JgujI
zo2R4C($(S%zxYtC2}@v6PusApC*Fxva3o3k7(fkb16;r#G9YVhcz&sOrjjYHy5O+6
zuYTUY5(cmVJ{jScly2mn+$SQkWM=`W7)#{R&+`$_Gz-5H-IuIHot;9LbwHyX8@qV7
zve|hxt5#rg9bcNf^8@%?nM!pNH<1chSo<Ieckp^g!o0Vdh+Q6t-PY%`fq+|V$ueL@
zdw9w<=VS>G)%-e&9ubPM6)kzcX)E!+lRcjZ&*wAos8{~=DI?(Ng(BmThS>+OWAn#$
zvS=eIUVVd&9$*9#V#n_kb=@nzhz~16fYf-sfoSa~)h)vH0w2$)D#q@q30;=|Unv)5
zpi$JCwKTAlY!dm{hjbR1EQ5}8L(Wy6e*oLc(`(G&OM<(brMo?&*5!NtxA>&KVV$7E
zet+P*t6ux4nJP~!G>A4vZa5rnl{~~<=OqkBCzvw*Ua*CrqkopgYFYWQkPbVbXnRe-
zX?2I@{MJ>622}8NHpZU^Mg*;f1`&6$^)yD_=}4_7gM}vt*322V)1(vI^cUmMk1#G^
zZ@DKBuuLeg7ugsstL%%xIm-%MQS~{E1FChyEVuTh-ASuZuKzyDu@_-JIlKk)0vX0G
z(<8eQ34M(Yrk7|iAz0smJ;<KN7Cb|r_r)TQs7PL^1D#t0OwvO7s&@ew2Z%Lg4m9^q
zKL4-`O`o2HsyagolLME{QGA;~Al|pchaFu#K+kTvcN2e%J^~_Z?hK`Xkc6YS`ue}W
zKu^9G<0PMQ+W6HMx$$yY1;n_)S7ZmD;Cz&y2JYj%mG8jBOLbY>)sy!X8JlBmp*)1Y
z$yf!&X=sHjffigGYSQ_7LTokd?cm;c!W`Q2zMse>Um7-BKTi7L@rT$BAUJMxjR33H
z1~#2P`pPOM2&OABb*7&gyLNw8NH2rv2usgW#1b+C4mzl@^xN3wB4WMyxC)&PIN_K5
zzWMOa_(ahiVKgAL=LP?p4t6Y_R`?(GZmai^QBT2zWaWCS)J&OpMx{T<{jTPg{!-r8
z%B@YIq4c4`H`ecR0ZTYF)>DF*f!;s=u`)>AI2}0;_z2I}5{6D(>@mgw2E7^a4>47w
zKWhnc{9R>&&y?Pm?YhT~b?u>5)A)SDCm1`$_yf0&c%Md#ZqQ+pJXr!YEjW2QdATvd
z(DykRFp&m<)RrcBrql{<{i~s4@h~GBieHcos}6ZVEBZhplm((X+(~Zu+GP7ezp00w
z$8#Kazv7|9^?;IHvkfZChJ&C0yyU%Wmi%=xVm=}Li!lpZOI~GUydPzVHT->=4{`RS
z7=B0b{<#X*Th<z+kV!0ADYEnJ<u))RaB#pb*s;R)4hz?fT$$ZN$rT_~BZ!f;n!v;V
z%m2hZR*6S?5>2goSC(kjc!VCj^Ggnafjgs2SDLocZLNhr>(FJ+qghQEiF~11vB>L^
z7;<8vC_TF`bb3g@rQl)1vnUpiPfj*MzVC4)vf|#`Ot5YrVX>Sm3USGEjp;T;GZX($
zAP|;(2O=BCe5XB|wC_9WZ&n_29!s)WZc^6;l;q}5loq9cp58(i=YFPlJ`Um*ob6(d
zX<DwxK5rw8;RfdURXwKq`N!hJB`PL2)hi;+?6A-jKV#SfZAfpt*P|8iU;)4xjC7kH
zhaP(aK59OalC(pNJ#;NLoa*xGwofjUn+gN9@2@X7H5jN_S;cxuvZp1-i{G8Kowat9
z8uFGLRz@a2x$7bRN#@Fn|Bf>-aserg`K~{=nBu28vu}QyZ?Z<DrOh3dTC!pCSJ}{B
ziDT~QN*#IiN%gk8R-KoR%hR(J<44)&d7U5CjnG$>T$xBD6mH2(Mv+d4_(ld%tn=F0
zEG8N~k&!H)*SGKUld_NiGFUr0GGL!=&7pcL`Bm2>-<ogwcspPQaC9G*_Wj)`5DOOD
zgW9Jnx^eX^3W|e)vV`u4z|9SCs_=nFf~vMkXMH^n9RqM-FI9d>&K4(+)>AuSiEwMl
zyhqW}yq{3Beki_5{7=ykoW*u*|4~&)8gBE2Aij+;rXaVQUC561gfw1v_Cstt4J}KL
z|FM)Lrq)Kyd2|i_B^EwtdwL*F8?%1dsU4_mEl+|>YP1zTekp7pHU`q%2?^VP$~pNX
z)~;x~``8t{toaMi|16QLz)P-ekzpu|%PnJn`#*^a!121o?lKX)ZLefqDmai`qoKHd
zy46o;C+G=YJ&m;*jAOy9k4jI=MOMsSIPtdft**zKjQx!76H}830HQ!~D7|h136{aJ
zsq`>I->&+VNO-od`t0LaSfMK&2-3#?mp7?P_ks<R4;!i&^gI)Xs}8y7O;^r30$r)D
zx`M3;*M%e_!TF|t8+0DqC%u8gd}rRoV~Kx@kUK?!v#(oBqz_!v$2x-NK;HJbQAicy
z+c~h2NJjecesA#4kU=PgP@RstJyT+z6NBJr`LluYk5ERQAfXy{d+UVCyYgq;e&dI8
zU?~v>s!^!$(9(Dq6`b0GZ+e~77>ln?qwD%h>OT09>m9_x1NGUnNx1I!M-!!44+1kS
zpO3%92jC`{Na(xb?Uo?48rh-34ta+qpj&~lSEU_j-<1y7we=Xmwe!VYpY_kaT1KjH
z#M$%;h}ZnNg_i^hTn%@nPxGLpDR=G6erPjU(+~)osxrCk?fd|Pg(*`!Jz*&JF1w^z
zI6<;-U7@LLrk?KX(-NeLE?j%dUOavFjdnDa=PVU>Rssw7<EU0g3cQ%J4r7GIB}$$0
ztsS@!D%Fo+8uojSoYK+fst=s`sZwP5nSJI6Qkol1KrFJGFPgcDH!4PMY|mct2VFMX
z(~Nwo{30*mrfN?!Av+9zQn(#3m0<B>igl$M+kn}g;ceAD``Xo-Os7y{IKjcZ<qSOK
z0cohopCK*5jec0Qu!xX}{hX$lT=r2<0YQuK+*6LcR=9UhmA0!bn0Z92`p3lS%1g?x
zzE_jHk+x{e#dHuvS!RcLEK!Hy;c_;+rg%%9vUFvo(InJq8I!%{MEORNcFwtMJ)jh;
zrsC<<V$UF)FVzY4>6B;BcyM(FFOrsAw?2(KW~+T{`n$|Z+9`rK+Bf^lj`_}pdzMp|
z*rpvN`vq(4)>)IUXrFp}V8nBjm63ql-Pcfuqyu;qh)00_U&&sbLy}9917xp*+I8g{
z>neW)Hb;xNst#{YuH+1vQkz<Kijcg*TVI*^7A?;tRv+0f*S~hrUmw55auU$5(zg_@
zR9v8&>z_<S6c?;EsxK-8qJ{1dC5oK5*Mub^b@%oHCF?$Cp&ST{KQX^fb!e?(9{q7v
zMEYzh720H?fcvzA#@TM+Gy(TMneSqvq6F+RKI=gntLjyi^9M%;p*z*=nA@Fd$v*Dy
z{`J8y3GV$E6h8u=?+UQtPlJDnK@$1ff?XMB=$5Oi^(8Y1iszj-g!J3#&TMtM!$vty
zp2#_b;m>p#Cz#F|+J|Ks?+YQgAd~%?F)o5xhy9KTn$*9%t}<%;`abGQiR6BOW28M_
z9AE*m^$su2yd5<IP<pT%ZaVzC_<QfXWfvYkb7*Qie^3HtogOH13uN`lL>D#SjN1rz
zjiX%J!+wh}4Al*1-TE#wHz6qUT3enxU4gk><OwW#8wyan+n1w8@x60A3glGm#e%!~
ziyFDgJ>KB(QbucX+=5xDI{x((2e-$O{woidyHbQP)Q``AY-JcZ@znn3yA`&7D~~$D
zINm-9yZliecF~;SNtx{-G4zW6jmVG;yh8HKebSByQtuADx9@l%FPK6M6o@YXtY2%E
zT~xV4CDUZRSO7hrh2)(?L{gOgEpcy|7FW%5Sl?HD)+56Xs;|&z*n?k-@jeqI#J*6D
z+m94M-UsgMR`Gc1_5<g70%-FOWO*gEt^+SS1oaMPGxvr+_$?yA1ZxCJ*mk`<xPOyh
zwq;U!PB(yYJ~Q&bWrHdA>0ywYm70mn5g=op`aII1*tl73hVgA+t=dzU_&i$8Bp3Mg
zv?0|U1?u_@XiA~XmCl@vJGE<Yd$!J@M__icq|Wu=yR%@M>W{&Fh*A^tf%Ip^Z}970
z0P~7S1c!4Iw$MYnIul_ixy^=vim{^+4G%Cta3nSQl<jmd;@jV|phL&lgrF;ul9jxS
zfv%_rpnNj{%KndP-x<~nOiva*+Uh62;LAF7gOGU|M<13t!wI6LPaIjeOr5t4LFtbY
zHn4yREdnTwP&&@sHTQq+ZXwOLIIFyqLo#!%eREy*dji2Ts>8e=8_ZvRir?+uBKz>c
z)}!(4XwrH1_~zVnGmC1kU3zrWB36GasEAmMmqeL`x228=D`45eEh`FzwK!C+cx9G_
zt=>&h$3dx+GRM(vqDvl9^Au7ZmlkiRa})l$zQ5p-*!vZ*O=x*65GYbZel^bAKg`<@
z%!vA?TSxPp9v4ws`|iuJ?3|;U!p2zH=ZbJ5ppJKlfxn@|KBoywXYEBw&pjYxMl{#P
z$7K*W1iwLs(V7PlLh7tni+)*zz=*WJ?V1)?eWG?Rlr%Me{tc&EJ2f1T1D6^ER=@f7
zdyzs-x)FiBd)~edl#9MIM`!g{@^*vCS%~@?9eqL1Ac%kQUc=I2!!WnVnKO@Qz?GEN
znh|XHHP4^G*OCZY`>VB<m^q&sF`7BsukKJ6#}(|QSHsx&d;C*UXwjZ74SVL%n~$)N
z0h(wiwT4`m*HQ8>v^FQW3keaP=B+sQ-s#+hT`fLx$faMh_zwQvtuB9<IQXVz=l0rd
zrv`AgZsjys1T>J>kRdtB$j#QB^GrVL65W}mNuv`)YnET~9(HjCsq|J3V*1p*RwpAj
zdoyLBTU@WcXa)>+7I?LC4arq>QLmXQHfcp6bZYn;l*Xz7hT_7jb@Hk{qLgutzJ|Em
zB=2>U1LCI*q1>)u1C%!NECB|4;$qJ?+?}TI;<R<zLoD)KP$;Y6i({~79Sru%-@~65
zA@b{0>RxLehx3duyuw9@mW2tX+ohf%B|S6ICOi~aU^>vZdoGG9%E$}`hUh*WwwpXT
zeDl8Lm$Qs9C3(2E_^+O!TxUe%PQ=4?=8JT}342avWLH#?2tsTl5HF?OifXPDDs>yB
z8}Lp!g(M5={?wR50zqC#JcJl9oO4hoby_-NT#<4<qZyqV*c>NCe6t?XM5ydM_v-W4
z5B@ovCnSHlCzBsdiil&^nT64+1;3YlIgcqR%=F(Q0$P-9v=QlksB9vyt!I%US2t}_
zvQs^|Zm3{qk6Q4rP>(sAz8OX|L*Y7C^$cmiLz8VsC?3k^MF1{5JAHQh`5)liVf+vh
z)BKK4p1{7Mm0abr6I$n->K)i3b*k;?Kg-YGSzeB3n)9PkKBxxAqpd*jglf;}^6fG8
z@cqUHU0BW27SRh%n|*}i=uN@Wz>Lu90`v^<D<_ID*?2|YE`ST}_3wz5C$YIj0D1!T
zF*luMWZ=e%drbQ>-A#_1W6XJG0Uz$+E&FYHHSab}2l$rAo&Y?9`AN;yhuIUlDAe|G
z$tYF|hZ3<S;1p04r^bQ<u7;4jmrn9W!;=#=AuV^YQe;KG{XU>u)&BDoKB_1MhHU#V
z)|pD;xFrOG-fvQM%g9YK^?bKEv92q@xERWyfrjnv7)t&LurJcf<DQHE+?=eljGVA)
zQ}pN-;>ICn8Ph60{u@i5J5=wBNEk;vg4Jn~$Vl|$Vl~h%8vzlSA>rtxS-hmv{Uv6z
zm(CY^S`i4_jOn21Gjxw$<3Qv;*qhBAP-~YnoGrBMF$xi?qTmE5!<r)Qk($;kKm=~R
zQ@Q=xD{MhR9L>2dU`OMWynn2xF}Egh_|3%WOqQl!z*<tjO})x4E=nrXD(Ldlv1Xv2
z2LfLjy?Z9EJ2p+o*~Ei;5O&CTLc8jc43Di6cD9kJJoh#Q8D0JfFy^vJ=Xi*mWJ%6T
zZccXZ6=MCHH!_j{7b&!r(G_jUw|GEuRQ|o$j^FOvoJ$kC*&4>qUCYok#>HyyKKC3|
zSco`{o&0KSa!P%{mIQ@Btao34)~CUQUYhQyz5V3%Q^$?*m8*$V2ubPSFY6XXrm2wv
z$JN9ly4Bp4Jdy_%121aiHzlNT51}4w0*CZ4X9Nhb;M{T>v$<LnHPp%slW?J^syH&?
z4DQ*Su#fd5-tTi6B;JUCo@ok%av8+9tDtU`0#0T4r(VNNi2SQgbWuSY*2Q&4rl6zD
zLwUfDsd1M5z^53c;b)`6vF~Vuc1m63|IQ*1q-(M`ewH^(C#Dycf?$IMnL_qqZ;V&d
z-M`$J7MyX18?^T`hX@IP@>j%ogypRsFTeCyzz5O)E_mxO_CQ*?^N*Cv&!iTVrrn4l
zUZ&Y2dI~Y7hr+H6n2N%(nS{~>wj|PwRE3R?Fj$Fu0Y$T)hc@iKs`_fED{Dtx1z}84
z>too95*nABx17*cNuO#78A`Z)O+QcL$_?6U?;O|!xW^P(8c)6dZ)Tilaiv4$bB_p*
zk=1TR1Ha$WaRz~2moiRPZs*k6xBn#YWMmMsu*3QCAm5wsAPmZ~N$*J>B9O>$mpSW8
zMfo`25yNj>pnQ35lGEo42A(b1;Q{TK9bCMdH`als)cn%B9vfsoS%5m-Fzv*%#W^Fl
zD#B=9rFo$ID-Le^3N~R{U#FZj4So<E^1BRfW^o4ja)W6y`Tm!^t`mISBj=sT5_un0
z2CjnLhvvkO<eQk@>Y^68vELHeilFqxw$eyBu{b=`hMUX`WD+q1a2p+qzPh^!_VU8K
z#5$*VG;!z2glh#O5YjNO!AFN;STgx8v&VK_xn9eE*Ug}1hxD{}&lgzcT8Kn@5sxmS
zTQum+#pynr?E<ST2<^9fInR(QklW<oeTi)K1pUa?T+5q)!teh(DkfapT2T{DY^sn0
z6>nM3hkEm^h}88TK#fa%XAxgijIMD{UfQR&x(l2*{<`r@IC%upN>M*rw%3n)&yPmn
z4<r6FvE3(%aaqNSX-#hQ87d15CI%3tE@_?5r9QtUIqu|gaS}O#dmj_S2(_w;wy)SI
zekTlY_SL}VAhg|Mw=7{cywkkjuOzR~b51O~@iDktng{&#a7U~kMaB2io}id{YlR6R
z%YYT$+2%yi?+dMHbQght)<(~+8Mxd!Ex++d;_>m}&2I}#Zcb~`CovVn?&TBo^|&3!
zpgY-+=11rI=Lx+DvB83U-HdpI&2$X{(%=g+1(ZC#iL04z@cSSlcj^|bBd~9O7fqNx
zGGi>BBj_kz^gitu3S62>n(pr>yv#NjZ1El)`7(r0$+O_SE}NRtb}OlFEHp1YUuvyq
zotz8s+je$f5K?dVk9$3|@KrygB-EwHD|y2`VYs;OJQrLq2jh|2r=_&7W86BD2#b(h
zCk<b#l2lL&M*N_=Z^ahBJmSyshYq*@@l*_d)Au8v8b;pmZ29|txH7&sG(LCA5Jo%J
zpZTg`OXCYby?I$ka|DLG2dzgEm%jJY(JHYf%?h>k$a1YJuZi4#-%R$^x~_Xmmj_0q
zw8YRJj_RH;-z1AAyCWY8hri1Uu#2Z=vl|ym)nK9<lI@cxKUg!5e#5ZL5>x4bOkcLc
zzK9ljv_rl_x?|9#(e%jOte>v$c>4##JNZHQ7AfItc(<C$NAES&37(_+WR2qzZvQ-g
zU~Z{3AYVb}(~6Yg&EfNP=S&aP#spyXiyG6Q%fv%MtP%m~Z_i=(CJOeFS5X9}j7NO+
zN^~)GnP)BX5mG07F@6L#=4rL_%)_#uHWlT%Kg;!T4>F!e-nh|}JO{RD%I+x9Q0n6e
zpFUxFl9dB3U|fXO4Yz131&DYtFcG%e&}-b<S5`omF+DlY5~uODch*R|a9n0@#v%CY
zR78=ST_kveS{_WQHrk;9RTe<r$tQ9I^mlHO+OzjnF<6}lT1g&DGumV(-#EgHU&N5a
zk;2^ncK5#dbiVz3zv$6EMB^$_Ix^t?F7u#lPfasikF;k1$!UN)VgI1wh+E<_c3%r<
z3?WH~c%SO<0Gn8KIu7(fjkk=xrk0k;8V?&JN(ZMa2vKuL?W3WVCdf2=T#mbA1bUK3
zPmaaJet%DNFt>~Xq)$KY_&ecoBXWG(jtnsRbUrvpF&EjgDSr||PMPMR7EMCPZj8^%
zU>iL>0V$%4n{AOq(Q~4<pn@|$M)V^8?F&@nzehK_P~?torweA`KdZ8$k5w|mRfIA4
zR{8s=$ItzAk~db*OTt)L=}^z=t-c4M&$ki$L?X$f-{hthKQ^GTy@TM^565ZWLe@Nr
zN;$5#q9cOfFi?W<FG~99XGU~O$NsWa2>w0;TyYRKAAS#?XnkH|f{9Cy*W>9oSx85Q
zMkhGq{WfUvzK6kzm1T_A`N^yy9&sPGyv1c|jAInL=_4N`{QapzqDJB*4anrn@B`Ld
zF>p@xy~ndQQxZ%LVKGErhkt~v^oVR0MrVxV0P9j&*}ZFy;6w|287z<_ZW5U_$Kvkf
zH;39793fu*Xv(HuprbaL$CAef(n?gf2{@kfenQPc0qT2BJwKragmkL)srJVccMVOR
zGdY80R42MXP`kq}+qcC`eUG*9@?T_Q`v?-$RYo_9!rz0~-#Ps#yfA*0G%9&{b$pUO
zHYgyh_`TOhG*Dr7Xk3V7XpP*aY4Th0Kx~mPO4*W}oGmFcHe19fzF=!I9*3WI=1UdC
z)WuH*u6hi}B$W;mY2t!OE3y4Y%&KWCa&)Fe?V1_y-7w1A1vhlE-$hw&usK%M&3wGY
ziH(l`hN%)8iCR3$Iy-;Iefjga1Wu_)75O%#5ddfQns!*Iv-M6gTqzRJbAYmciv{u^
z4}wW=NV*7I7LzpOR1SRm{e`O@`+sUwckSvC59FN{%_o?k*zb}I2;yZSOg^w?yceUb
z`h|e<`q;@0-yK+HSFHbfM(Z(qac3ZUXO`<nd3b>B;ivk?erN|bbFS|?ns&Gp^P%_e
z9Dn>!xBeU#&=)ZNpzh_9>==3a{TM}}ReT|q4VE-^^Yxom|AsvUNs;!R2cq*)0%4@b
z>%Iw&VIe|~CnS=_^6|jAb5UjlSgTKsYu6SuNNpB`zQsX8y&Z`1c<(lrFu=wwO7)NL
z_ZB>zi~RYdKUplF4LdlJk1QeaFZUH!0{My%Lc>B?bVw!OX@qNRuK*oAn*yqB&0R&f
zN5Cq#oV8_=Et2t&zt{iR^ydMz&ysFb?Q^^!1L~2@V^#GT6Z}B5zZHa<O$ZaYrJp<(
z>nVJz736SBF_fMhjC^QdZ@oaEB;rOmotr27B`Yt;>T~3IhcJ)by%W6DCN52K=0s+V
z{&DgbywRC<F<#{5iD@P+G)&tOtI12TS8uZFUu@_nHv?0Y%a1?zq}zE+VjwN6=PWtX
za<xOo7)A~hAB;*$tbVE&izGbcFnUZt>ZT;udCs_9U8!l}oM6v9?9Lh}a;E$3<;Ih%
z>W6%3=&+}fU#~ev`pjDoPrr)8CzR5cicRn$<!CZuoXun9e%LSkZSocNw*t~T+I=HO
z?vHw3>t9zUDfq~HlEmmaZcbF@YKMM^xy4$?M=6=PF**4II>D;G=OFT2{^q9w>h|1(
z3M5SW`2LsAIe*@=9dSp7+cfhm4skjG`_*~VY?|MS&S_|U%54EgKdZ-5#X%>ePVaD@
zO|q;Jszl|JSLW>xT#ap{f&^KTlusAf0l{~9u$-cTD3n@}e%iRe)Gv<vi^vdb!RCsR
z2v3Gdgv>3nJS4+p^-<kD*1B-3RUcFgF}|%h{xjJ>?G<khTz{<@{P5k~a!5@SDf|8@
z-n5&)4p-%e&~CErPsibMy)+XJfx4%a$D`m?ks&CYm{oK^;iE!1<)8i6OlcGSFY^TI
zbPVl@4=0X(I*meEJD0xx!7Ou~58ic6$#iC{Q-j*ShxJ*Ydh-5xB5FOyM98VwQ%=D>
zKZr2!elM^2si@TYD)b4QILqPZhaqU$@4j!f0^-$McGY4QLpo<3v4_9zC&?|?@O1L*
zMtCv*diAV|Mca1iZ*54H=6K-~YQ^2}_TpQnJDGg?-?fn0zkaLQV`*oJAJR^MYeLG9
zD(Y}gwZo#9ZKe)FbxH$B+@~)`e!UH_&-{iAi{RVbVcZfoCl;26f;x_(JV$m~p1wHi
z^C~2smoRt1x?;v<D1IgFOzFwDr!}`Tn?P=PBW;@V(Ii5!ZlD5TVXQvhZ8w+L(8{0n
z=Y4@T$j{e4R44_eFbh8hHsa8+BwmXk4AEPRU8-Kly{(<W%?CtvSW^$KO{**-WV3M9
zPC;uX^H*0s(x?paxe|z`4&o-8El><ySyAiklK+vLx>dH|+UZ7zdC;!|6G*T;PZj;}
zd1S~_@^BqqsCB+@b|F{QO7rAJxoHG(Y0FLlvfS^^d3U_@YL~fa;T8Me%2o{^izLkt
ze0=ff9~;!I(g7B3y^Xz*?Mv8>`SmUd8O8|bd*#wXvI?>fVHZh!yh>-ewQqp?$&pVa
zhE#j=Tq(Fyg;E~ZnNkj}4f<-9v;w-(oBad+J_3D%D9|2GZ;EOSXleW7>$%hoUJ|Ah
zwiEx70cVxVBTe)^4%H4=-t);21pbykc(#4c*2z9803M|$Olf0V?{XO52C`X2L=IjF
z0^;pqkjGJ`FYd-Yt>-4bQmcyjcy*#vZ-`;fDc|~j7HP0f>eQYV3<6Mn76piW9+K2~
zYw~vWlMc+PvNKUl+br={et^9-Q9@pNY7|cf(V5`QUD=P@O}X@G$nAF;9dTcsyCq&y
zzkpB1h;2p%{roZ--95#sz4jReYu>RiUHMys<1|rimTcjku<YI+eC_npY@?rk7p%`4
z%R(Mo@4Gkb`crqqQWJzlUg^nx7KQZ-y`YS{Xw2Z7Kx%a?_<)__Uq`QNtiF?PCkxJZ
zqAaxo-y1<v7$xbhxo;e&XFtf&R-E-{JI~8Ld!qL<y`UNC_08#CJVAsArH*)<oUodf
zuX13jcgY~k)MsZsEK2@U7DKkG?57#9h?~Fc<gpVw5*cP|K`K;a#^^7GN=$HqpG+ac
z!V6ZF03Wcje(H7i6dV3uM8nTBouwCbzhy|9_kvDlEi2{<b3W(r0$Yb33(?&U&~dHC
zAdI_we{9?j(=>NZFk;>dC$`~>6?h@lWzkUal-`~YB1~yz7n@1LWXB5QSjiOkfNKhg
z8*R8O*tBktniBn9p8_bQB)y-TOrCLkAk>>K?BNRoth>b_|4BX(0KFy0KXR(qb4w;g
zbb#xal3GR~>?<^VsJ7Bki#Z?9AsN#rNqT9sxnE*MF)~{6oWJlXnx>yBq5^JT2C^_?
zmfPMpRJXqGguxp65TCuY=M_llKy9<?1qOr~DU89I5>@dA%0O~$;~M^Sz-Z3U*K$`f
z^hqJK>KJRBiI`3L167k<{NN8>Y<n~>Rbol?NK|(Fck;KH86lp86l>kTbyN0m?qm^z
zsx;p`n}12075><PEm=ee4)&gJzaqY!cwFGlFR#$9d0LssyoaFr6gW&K<idL>xCl$x
zTzg?$<9YX2qUv6ZwH5`>wjxJ^N#CGb3+W$-dwjvg>oTDm`2=&fdFy6Wkm~c?LAJ#j
zKD_|;PNNirS-4`{WfnoES0>K2l4xJPhD7BY$qMUg-t>rx#J-`YHjuw1iPnpdfA^x`
z`}8mhOzuWR8}kzRGE-Onc`qlE<sip|VFRUEYxM*3*}=i5C>GyY#M0qrLKx_ShUgqS
z8&9;LmALzcR`!$7=8Awgx|<(H=a;sel>B?xDOv&gg@)R-lf#?X;Tn%thIoxd-@82s
z${)=0M624NoxyVyWmHU_kFY)v<<w_{-3Ou!hAt3gC_8|T1ovRsIlrg;?oX@GM^%2@
zB4vy1bAI(Zw@v;2)S$w_ybD*6FTjv(RwP=YBf`P}UmOMcf#Hs(>7HMRYM5g%hNH9F
zqfJIoIQti8xfRGpj2Hv5{^wHIM)sa_SEQ;K&!^;CnG=#;)H{@^glo+T<63xdFNU5<
z<x16286LN7OAvCGNDM!H`NJ*f87GIJ5KjQ#<_gBFY^0&zHcPL6^NlS#X(IdMZ^Gkb
zn8dTi_RD!5MA%A1P7$b+Try!6+?7~@ky0Fi%-P`csTk0QHqI{EOBIf7AJRI;7mq%1
zFd{R_vh7y6fb#KI^Xcqu%ne7KywPk!!su`FEtlGy#w3m<p$Ih=FFM9|q$>mEe>|DY
zE*B!-#qitB#@kflN6^=8iOL%gX|}g-&yfYkj(&BI-AlZM(bjsUiS@<%`zBD?#ld#}
zcrv&rOUiR|KG}`MQFSn4k-G)lZZ3VkD;T8A;TO>FS`mTrgzZk~V$#XNlcmCXkgtVl
zUar%5iwtR6ykK+XZBoJq`SW=j6qTLFx1PS(t>-S5Ade<bGrUMV7t`#>D#Qgcazp1c
zgP-|;@<1{dh@b=&I<Arz(Xl?ZjWKQ*6jBHa`BK^&wW<nw;;S4Sf+G+a6GPG|8iTV#
zN6dBkYk1${(04&t1o&8(iauX>a!2wjbm8*}yI<p86t!>!)XSxtmb@J67DHB0DAebb
zZoF;OarVKFjRujurWv1QiW)%hI9|v1uNLiT>=E$0V0_S6^~uuEzR!*F{>3>s--4r<
zm0%bLFp<A_LetZ^ui?qLZtoqtkl*a>7bk8;ybFEh;UU!`jvXRp_wQ*Hzgg5e^v4o<
zdHDMW0`uVU3}S%Cd|XMm%Mwg5Y}PK}!3`-UGmr+qh9mn2iL8cYz!LClyLIcSoX0JT
z2$lnlJ4zkNFHY7;8qo5Bu?{;{g?-4V2%IsLGgP1IGuP-*oFgs=*#lM&D^T9jJFz?H
ze>yW(;I~s8yubUVUR`*9$$}VCm*H}o=6!v;B9`2KZKe8Q*YUdF1|>pLy|y2>RFb#i
ziAtJU){^a6Le(|a3SD+`@{EvAu^#gOEFv-!q3g|g)y7A*U({y2$UFV(r;LT}v;7UG
z(SpBLUwM)U?8?rMrq3>00kvy-^K0{>R=mIYP=|}&jIQj23FK||srIi9W}aXZ42QW7
zwpzTsfXNpa^J#gtRZWP}2jNbxJuYzC7XEk;S9Jc93`B3}Rd*=|_yGWCG(DweIf!<W
z2%1H37hL4WDx)&}E;3ylC`d0_x)n{4KWc8GmyAKStimh~=gOao_+Fa^`O4>1WO(OO
z(l|keYmgtoq1Zi~sdeP9DcZ-EY3&G^EJqBOe<Aco_$-|#?KEb%Sm+)MR~Z+<Mvh=I
zCM?wvgs^8TugUmziqSZrGYX6Fk+F78fVFjMqZQ=Sn>3|pYvu=S#`svA=|Ygt3Cgog
z$4YczU55kIFU<6StxQK!S^7F3Ez-Y#aY@vr&6j9AbSAPW^71@#ENxRRJ!h@lej!Z2
z!sMHPV>Vf#75_n;zGw7|bM<RLynR5A>m&hIp!gxvjc7R;O3u<5MKmE%wqq5d`p()s
zGq)vrUw^aCnJG25rPS2^@YZu@T>3itOy9m=nrf&xmJ@yLN=dX+)HPnCOo;5+Ct}-I
z9SrHB_w$zi9TNAk$rzsntCii>OVH;Pn1|r7jR<{<=zfE>+V89098aLw_sz)?aRn^e
zUY&1O;wy~;Vp&-k<KCL^kSY+#d_m>Qb3ct>QXTjqR5yzO?cstpH>8A~nL8wx_&wSU
zg!!dD^)m4_7#k|lMyuZ<P^6d75KML@v`B{!{DZt}R4$S30=>7y(*;x-Y?BVLpgbZD
zV7;{0JIPt29tV@fCyAAViw)BQ%QKb*`%42guF1EMB9B;wV@q-ZqR<#SB0|e%Pkdkz
zgHnOkoL+KwqkWCmw)p28v;K3UJ<HjaJ;qg!*s`_$q`}JlO*#E!F<vv<MF0BdPc1Xt
zv=1MkXqnjgCU{oA`=*u77cObuWjA#qN=E58@|`!BYM6)UNN*NvgA@W8h>8T&Gg*QM
zB1D3Umb7kp=gb|*08@aWg~{zlZQcT1tbd~}TC>G*R+l1eWK160eU9)Dcf^2{zVWXI
zQr>Q_Z1iC>>w|yB#YyL~vo>av{57-Dr3D{kg6ujMYSx{I&(l_#`*8hOs%zzoj$IIE
z{LSZm@$4clbmD<xA_ocIJz-LmQ(jPhxjlU{+gE0=NUpXaJT}$wH8Za*AGzycE~DV#
z{D&+(F2^`5KD};W6@o+Dr5s%7`KKG0JG&39%@HL^je1uxOCY|@g6#TiX%lC+B$B}*
zj$(mQ%$?=)pLCGSt{GNSZ+7;C#c4u+Pv3m)ds(d(RQ<01%%?%48iJFvzdVw)|I?Ka
z@l-awgWdSy`juN<LW}2GC@g}ZaS0daooC~Xw0vHj#y(^Yg!6h5=QTF22qg=$rEOo%
z1;@1-H+)|N{d16e2I1dk>C>_#jota^4uZ8A)NZI1;yS0IUyXs?&S_z~nIyNXeX&af
z{V0*{WD{(e^*T2SQs!`-Zt{vq+|}}eYjHX(o|y$q<jVgu1Yewc$NvjZ45g1%ZwJ+J
zZTUx=cFP%uFqSX>LD_~<b*hB6SvFCNw@I|Z3}~QlxYBXrAW3=Jw_E>A=CnjH!A+;a
z)3~tEsEN?1g$%N1B-y}-gvE;)h;H@Ng7#*~kXGb+zZTWz3FsuTV#K*=#Hb7R^#gIH
z)d=mscy;fnp{(`W^XR!J=ra!E(3)+Z6a_Pt^Ag<(Hm2eTEiEb^f!#=K)356%oSv~!
z-0!rRB`!4myKqb2MGa$~;CMzvjlLuPgU5g*5<j3Ux7qh3KVk5M6FZAu=$)zmm+6(z
z{LsmZflc`X&-4omw4kkD#?3Al*JUYvhG(YdS#`*6s%VHMz;?DBg=G?J`F@_c$F}L1
zgw+Wq>jDJ8EjKInAJ?l4OHLVZ?h8n$dnZB?<&{ccD_cQEVoFC=p2q}z9a{}J7~@gi
zO8nJMtkmzpgYnO*`O+RmY1d=dF^)^BW_>ZYr=*Fq@oZb0gM-HoWvza4&TE=3FhUEg
zt4SQHF}8ubrB}2Q{HG<;iD!mkJw!VTdnB>nk?_djXGJKn2aYLZWv>hv?U7GEmq?HN
zx!OkOwUTry4fEMO@Vkn?R$Z*^6dcWNQMgDtt@nJ_DyKQWx>K89;h6NOljfu7Al{+0
z9p>KiTj<yF2qtjAX43VCiC@@d?F{?M@~p&jInZi?rvjwQ1k-?}9WLc14g;05Zwv7n
z?gS~4AUg^sw=iY_CV>i1d<z1`G>FADc5kWWG!jQqKYG4|AjF(v+sL-l4(GwAEyxG$
zN&N%{uU>E_6#kLPk0ibHNwzX5)IzmrWRD<%2*uxSJ1v1JLbdPt1gBkcpK``%sf);4
z8($KK5~PHhI1z}CZ=4(ePRaM7Q(vdq&b8&VGkVKOaOUC%b~z6T)(Q6B1G>?R@ha(T
z&z)+ZSpuXcs&X$QEL+kLoG{`(QP<dkv3`lyOel&w+v_mtc=4CxQ^R?P6&XRV9v`B3
z@kdTc2>V67gp8cR`uP#dqW%8{b2>abtxw_)Y)ZuwH*bXce$qGCYkflC+e{|A9XCtS
zW=3JP>TF2vZF|LwBTViH5jg@gn0a1lr=Iaatey(1$oiN;#=!fR0+SBDfW8RbE$fg~
zo?=wF^{v;`Qu!PBa;IvGE~zM6dnqR~&m2Zr^UlPO$ey@#p&&4O)!gwl#XizEowe^K
zmt+w@ZUeM>dSc}r^a3IYG~byB9g99^#Cl|*8T_`5De&p#OW*!RQa9C%#=&uKuuhCS
zX7BE{0=&9s6E}Jj_on%W{|8bHyn5Ky5YSZcNI4S-l3-I8<jG!CuN&C#X*?7;eQ{xn
zEXu5VpBZw&dO3-BTd`2Ud=@M8A+^8BhB3xN#aH9gZl4gpU24b63D+Xgkfnt}SRXI<
z3j&ta*Bt9x?uiQY4I0ar=p2^VW*TAWE{Mgx7-7r%Nr-ar%^XJmrpvabAc;zLl8rCW
zP$&$3{A=K|&}ejgJ%n&2Fg8Ny*m^14Ewv!^<>026)x-e&S$9<t(WuPIM<J69)2Jok
z7lsL`=5zn4SY3s{b~4x(je)X1|JLbbypv8s;cpOfbpt0UZ<ifr2PEdElz;ciXXoF6
zApsi0O%KTybl*Ux4jI&=93#d?H|cw&N>cF<Qi@ljG<H`$PQL<-GfeUPN~f=tAdc;w
z-CoxR2yC05^%q=KjtM`)ZL{K5Io6v3flp^2eaZR8ia%V)!!&IDh`YXy>ot7nPyc_9
z5&3$<W6STa_ZVT+gXEWC$l*a^i4byH(1c`e7q8cvjMW3>9;6CZR_BQYW0IX#pQ`$o
zevn|&H_<kDhmGr}5h>!}x(}<l%5@vd4X#;zOqln2utaMQ<8{Q7HZDl$Q|MQez$d`3
zB9sxTxX<5dwW;;{#*fnv$w)sWVixR7?h*<a>LDh60xrWBz(O752tz(N0^`qB(B~W7
zZx29~T79#nblC?X+$jX_)oYTvvANj$CAmi5Qd{e)WJd}SU@eE>#Cd{K$fIIy3zCEX
zi`dmXh5&<@{Ec63l?0P%b12ntr??HIvS~Z|4>G1&U4a+q&rZpXJRZX3<jNoyUc-}p
zBv(T@iDmz_`j}z|v}4>QJEUSW`L%^K7o%(R?E_486-A&Nh*v8y8MzmkSL|ORQp|x$
z)5WzS>m4Uw6Y#`B0MHCjOa}dsXS0CzmP~J><d(bLRD6i2#`~0J3^v#`XBF_<AW;f#
zD^VE1JxoMxhEglU7+9X)DxTd#Tu#ZPhY>oMQv3FYj!-NQG*fLqQVkf>;kY~C&t!Cp
z^?J0Q(glC-Xz56ClY65)cds9lYGjpe4DuM%uUKrdb+jKzN0*Zje@bPCO=zDEOlml(
zlTYJcq~kvbW;?f4tVPyHhDWK)M)h!BB%Rq;f9O=m(l92IUp`z{Gf(ZofI+28)?5a$
zv*Hfo4}`xD<q&O4_e}#Muq%7Lwcguj#*FECahhfv6zsCeCe`^+>&L0y4j&PNoso^X
z0-bNwsL1p!^mK$w)X&SnEBb5hT{Lrlh~kqZoT~ay#q4mB297n!D1nss&~c|tdEq=K
z_JI!F^SmVr`j1sl2=g_2KAWZO>(ao{32*f=18z?szV#8mCGH0SZkEWimOK{=svSTP
zJRIHHlv&<uAkoybm_my(c{i>+i`}*aH+L(K6IVQwH+eB_1#6L019_7sS|9_p2KTT|
zBghrRxp~*)-}J9q3<#!sSraS7aheMoo#BdK#m|EG*tewDruE8<kLwn;#)9T}LwdSl
z3fl}&U|tzkwY_PZIZ#Hm-?_pN_Tb_oL}x$bNpuZr>-Xpyiw*4_qdRe|a<!oQGfc&-
zQFfiV#UGpW$oqA8QEhDO_h@EC;w?6eJ9O_zgf!iFOMF^QV%+j@kYA{1vkIv+PB>J!
z!@n*qWiIArf&^8Y9uaz1C3V-d`=J-wQtb*!MBLSn(o;Ha3OXv$t6T^Z#rw(~a=~^=
z<vjyPFOMQwjmIL3dH1CY2r3JpTWa&O30#~ksX^@`B$`Wa+qCaZkNW3xy-v5WwMh3%
zcwdBnXZ%*f2l|5iE`Mp*+O_YiWR`cy5bQD_pXss^S@!869}A(f;%GV(Vdj9%h3_*O
z<)jFtG(3j(Xqo3qCvf<rx5=ItAA~N`Zx`-`uABgvu^w-qB<^1EqyBT2)um7xP~oEn
zb)Ezx)DVjcCrE3&&n7k%u?x6AudkL~IGh~@$`XKlU1;j6Az>FXeU4mJbXTFYtGoH|
zm3b|J7Qa@+TtHSXQcLfy=S&$CsYgs?0V&7M<C)Yr9xmOOS!uP)ZbhJWlD%aW?)AYL
zUTqpb@B7SZ<LG8lvv49Ufgp%&1d%Zk^-Q?#0=$h@NLXG-|8F&c_^3uM*4Q&4I_*57
z7zQnS;ixBg7(8^5q+CQ7-Kam3Tud)OYQg8W)O3LVy1G^LlCOg)jt^GN_qQ7D0o@<`
z$?o)K)|r?@$~g+WwC>rohIsewP$UHlesQ-XdeQ7)kjQ$RU%MvXD9HASMKJ12<d?>B
zFK4`lGcGeemJ!Bse;9oz2$-;{lvrDBy`=g#$f?EpQ#Bc$oT=LXRs@qDw}lSGBaQ3V
zOuny2IwmRLKAD|-u^?%rAPlltr&F}54jU26rCW@t%~rTSky%dtTVTKYt$b2%Sj?sn
zq)X<Ri`V_@vU{cy{Z}vpk5%u6>1PQB)53|vtbA@nc-U%@Nsf4IJNtzgx^i0nai=u5
zZxlkEj4}VFa=-{yl06=qhF?Xh&*IdVlwNBjR_^rSJ!YO@->yvxb&*lBExL-?Tu!AO
zWJTr%Mh|6kz|@s^QA84qd&t~g-C2|R;Yi&>Jobl&I6=R?4Vk@3@Ag?PC=~D+0(?nO
zPnuY@g>uMLvm(6)nj4?<Itd&GET-T{$fnhXxRU99jzOuubm8o}YJah_&7o8Lfq#of
zg?6rbVXf?1ZjBh1II<yKVx_~QrSgx6ojj+URm$gUozEY$A|+N50$!%54|PrL)&w1G
zv0jp#|8fa9SUff)^B55p;@Xxg)ZVe5z;Q})Si4HHYf%HTGoMQW<+kgn+qEIi_{6^R
zA(>gQofDCfV3~o%diIZfZ{uTw2vm#ox?war8JiVoxDM2cttnHKQ4dsMdOBAyc#@%R
zrO{lS1p46j`Iru;#c+Ad?H`@VhpB}W`Wrjd!U1Jtv19P$BOw=iCqBz=+~i=z6hoS8
zV+I#LLX#;CJq!s6eGdP^Nk}?rOmF1Fn4j>b_0Gpkv!lGZR`V@`HIDD!i{i)Na)V-x
zK--&j2T9*>!!+w_svTK2^`n<HZ!-@6w-<oz3Pp@_q(-#SP4sjvE0r7OjUr9`nmK3U
zb10TK$H!b2;aD1wL4KL+kMq|5l+d3QADB(C6lO-c-5a5$MfEt49&sk2l0)%dkcU2(
zngGjq*33|l^u9JI+zV2XG6Vo2$#)EcIDMxx1i6*FKmna1#{0(3jC@&m6+G#b%!4!)
z9{HXty1@eMx|FpHOADTUzAo;GXZPwCove*sEVZ7x!RcXyq@?r&gUySY^1?g4b-Fs`
z^g(vNFb7{W+Mo(DF0=WZo3vXrubo+GP1j6~+q(<Me{>lTBq0VVVQ1+^MU<`&$xBdr
zDVpiHzgn<UhYpnt2~BY|k}FWaMB1=(%O^k<D5dKaWvJX<|4L`D5jDnBVeMZr=bej7
z-u+J*J*7~s1bSjV=qwdWZT&gKoWJ_PQ<-lsT6L;mz%^%D?tjG(y8xcikzKZUYNo*d
zMc7+MRT*uI+jK~~k&p)6fOI1uAuS~#Aq}ztVFM!FB`Mt{NC--Yba$s9Aq~>fUB9)x
z=bU@bz4!a>82q7oz^?aQYt1#+oX>M_40qQP8!5~o@VJ^!dG+6534bX~Pq3!YP~MIV
zU^ucAj~F9vgMa<VskA^7xpjFzC4Z<MC|u`%))HWm9ForixpCE<V9nr7ug0z%U`2xQ
zkETzP6f3QViq%Y#5W!$v-&~&b+E(E28S{q8#NQ@6<_rW7v8}5&1YvV?LrTqHd>j>>
zOl-qFG^OZ?$swGFLDyI4q<T@)CY=LlsL4tZ-_zE8$k}^>c`osRj=T}NCQ;#(n#}aL
zLcVdJL-cm8Ft6GZ&GR9|>kZB2NRUXwE7H`@NGih6Cg#QtR$N%sE3?g~x^XGVd~E|H
z9d7`ZS8<c?!k?cxE!l;|+oGc>R#vf8=R$<%&*FilE>g8#EWQ4Bm$#U5ra7OLE&1jt
zt_q+eLjx)LFuzcjZwHoPJlv2*S5ZNLZlmZfXi5sE<ypba^?EhOGt<p%ZXIn3o7`y>
zVZkAA{VC<0jNQ#$0ZSlimQLh`&q_6SKXx^Fl&*7daQLGzKr=xrAjsv*b!qy&8H^He
z()4QbrNGp4EtO`dMX$9i1wlaw5)Y4fOtN@?|3ipHWrO55`&p;fUd8>{W$#;`kb!5#
zN-i;CI}7lu{Q;71=2h<E{k1MWHKn^_JcniE#UViy8F9UFIM8`Z%6{*%^a*ZAZ2V;Q
z#rE=<S05tB6WB?T1B2#-hzWa5!qt;<e+`i6nRg^(XQh0SO%_jni^G&Aun97Rj<wHH
zTq_|ohXf-1dZGN(YjxPxjY=l;P37IZ5y>sgJY4TIH4W?{L|`V%T_>hX0Q=)_?^8vd
z@e1Fds^*z40CpF!llL4L%xBvqCx5#|M%#6RqzNRAOy=DvCEL`~n?XNWrEULK_{Uw#
zrO=YhM%}O2leDBYO5}COWSP<YO;q_n^oDvCtu&7AzhV>q@*3!0O3>mD$RlYZ%LV;c
zDiy}_mO>;m7_u-=m_FO{Cmn8z`6=k;8Kumr1C{V54?hA4+&0t>xU0FQpGP!K>}POI
zRbaY}zf~A7BKA93NnObXg>=5=hgc=Okwvo5EYba>d-jY_bKgtxJOhY~Zif_w*3)&K
z4O@qm3=}ffzU1_d*n?dN@@$hO>s;EEebmW$tLJC9FJL9%^+}K?%;}aL7GL#Fe~SaF
z(mG6&RAidOkp2Gdqlcy@!>qDyb@UA|&sxj!Vb*%{s>g<>pdVXI7x)b;fL<kwcM4&6
z;He1Mct^qEQ?a`hJFS_XtkZWWC}1U$OV2V!|89^bwoUQ0YvJI1R2FkZ(#Ei)p4DvV
zrSqZWd~2JN*8Z7zfX09r^Vow;h9=u<-tAAfiWP5vI24D<?v#MynW~{V)*Q1$J^PTb
z;DP5BffhmXZ~N{ohgZ(`dV-GBA;Mlt0Ov{85!-gSNP4OwlH4ggyjLYGqdv3J*nD@h
zXAVmrJYv3R9GFm=jT_`*@5NEfCtZm-@@J)B%3vxPc%m7IRm^nn@kf{{KSDT?kc>V(
zP&E#hP(4FT!C1t`J@p7I1i34zY@HNRV?&6x+-yMghFbh=6<bM|Os`evwWJ9Bx6-2E
zXMnF`!bL`{i6)TKh(NdSu&62hq-|kjCp4cuJ<!oH$YXsTplmRdDBZwdRs~<!aI|>!
ztPgcNp&opEM~wlR0|?z9pyiC%H)unT@JqkWelfC+bWll4;^Qb;M>yPZRR!Q^v&f9I
z0Mo>cmIq?MN#^SA7u1gpW@?>K#Qzk`cC|->UJ0>^IY{G<?0%ezaOEgq#BdmYW@Tni
zX(c$s43pNTHJ<e_zi^_z9UaQH;oG*=6D8CV@a{PEfy;hMUaLuP!@y@H{ycXAfa%V(
zdp>!N3;u_V%E`%Dx1Ymh$;}dCQ_*}HXu3DERY9Q`_XksHt?KZ^tOf?E@oEV)xmua4
z2oYI*>C~P=Kli!5=R9C)=MrpamNmmMQV{<K9SV^{`HvsYFW6I|y+-LK3-^uzwCuR(
z(Q?9S>Wn6~H4nqT4=I28mdQZtakGRUemVz7Yy?Nc+@Hlctt6Y$>y$*$D`Mir2AA(z
zs_dXh2W;t6ZzJGi)-k+OIcf;DzX!nY6Tz&0A%yGI5SzoCE`q9uD^ydH`@m*E0E}Yk
z&HhRW=&Ae88G|#*oCRh}02FXiRjng_4;IfhC8e>_3sUe!a|0%E?-5)N6mRSBUjBX9
zAj8y~65G@+O?zmPgH1Tn^#bW7hBdZ@3EskG7#z(QTeixaw>8@K#wshrD7K)LH3S_J
zab8s*-K=)QUl8cKp3K){q}d>%mwwGpC(pV4__oOoR{x?3#=|ACcr#5M;cPO?Srs3s
zI!a^+6~sp5#db$sL=UVXtF6!le;qBH<Ifuc(SNMRo`H6ZMpw|(@2B$_ho5<~c{(({
z=-IRv>)uq;PzSqj(GzuH$P!vb{+h1jtkk<Hz<YXzo1;o;8Buw4z7lywGGo1MO|P)t
zPm(@5p`=yHt@`|UNK7esyE{1h`ZR9e$jWm0$=)(a%ef<<fL&EpSx*mWW<y`^A1hJb
z_o~9%3uK%vm<QB_+a%DZE*doy*h*xW!<f2=4<$J|9FQt+g&rb^a-O<+6%W@p`=nZX
z$N4`_1t#y#yyIeDlD%G|(rrsQXY^*;bnYu)h!o)-Na5@6r-i7WpRPd%_Sqb#HKW*~
z0Q_V^JXz6esGfK)I;)VY^#!nzx1!bFiUQ(S5@xK2!nAa)0=X#(?5Ne&Lq*j_T{&e@
zC`sC2Vc8RC#ZE61t~*aVpwu`}COl|9HjJ$@mmSIpeR=5ip)S$BD<I6bQckCTGG=uS
zXq`v;-$Tv>+Gv_z8pIe32xV5|<a_7WPnAbIKk_7^p1EgsV)4zsTDgwi<HuP}YXifu
zT5<V})BJ-Yxo`aF!$!i2Gb+}5in=YV9B#G4Ji&%lF>4o4zf*Yp(b9{9XVSILG!HEp
zsduZ8e$Rutu|Ah7r#u@*+mVQz;%w?ik)u^yp9^i5>N-huU8r?{{O`4&a$`$X4HN}8
zF8N1wj#eMt*ge<yYk6Vj?**x1SOjK5d&&>LQ0Ikr3BMG5c{)XpM<5L6-*TF%|EeE)
zMyu%9KTvl<u&dFhHHNyYwRO}WJY1EYr_^=z2O#E}9@v%LspW+ymNzimj6oUa)DlMw
zb385m+z6W4Uf+X4&zg`OS#FeTmk^v&yY7tV{|!M5c+<F8+(ydg)UudKA&^^Z_ZXvD
zW4w*tL(P99dOlC$G7Stv9%u^)doggS0wF;oz}gnH)otNV4--r3F&-N)6sbW9tYU+B
zD<U0)5E7`OF0<UKDPPhmbN1gCZ6$7R4~T`oB3<rafy4y2*YMZ<6F7g6=l#WnK*~!o
zDRm;u$B4UX?|_1iCd@O5Q#y^P{U<TjmG4M9E3|F|?|}XtiYzpRNSjVuc)tZzAGf;^
zYjHyuw~vc1Tn>2&3(}1BqV8Sr2_Y{1?Hd!$&`vp0Pzc!*5|m}Ze5oY#obzqMV}Bh%
zORTR(J;JGctfU0?R3??cUz1H2D)A_|2Jg=<r5?P1m-Mc(zBuE3@YQ0x^F__`C~*<)
zi&ST<jl*S)p$U1dV6UEbk&;+{Ojk?^-Fa!c5S`3}_gD^u^>o*MTqYVMB?n$GC0=i<
zHD|PF;1aAzWWy>By(l~E;<EGOMmHu{%hPFkUllAVw}K_qvk?5gps!BIGWDkzu-W;+
z)n}unL>nC&Y9evu)Po5KB~w1siMvXB$;bsUQ2njQf1Q2JRHw10T3tw|7d%Jg{Fmdf
z^U)YlzN@!yN}|wM9MT<@;n)xF6G9s+mE55k$G-?$f#FT`8}&WwkT8p|`WiM!pBL66
z4~~Cl9vK#cN(6Q(17P**vpF-xdP(Rpn-I>fE!iM(@ixJ-u_2J?1X8RW`Q?f8b`FcS
z>H=iQhVr?6MfpSxuURUJ3UfV?joJ{yeI6X0R@Q|}6lAd1C9lXAdWme5Ztu>zHY)Qd
z-edC-(rQAI*jDAM5XnQ*EVnW3fjrps@NC`cAkYd7blb1c$C=m>UDC@n0$kFw=G3(f
znX5pLh#JCC$}kkQ`nn7M{g-Fz=D;$dcvaR-ePmD5bglY__7YH~7O1L#52dR<*CUa5
zPr}6{C0`vj`PEySmzeZ9ZN5}PnsZ*dQwmP?QFkb)mghIvI>i6PIxle@Nt?z`wCh<h
zwR=N4_6*0n1_1l7>o>3L5R&PdWrfcV{>zpH>|My{P;mcnzf@eW-Y9)ZRD?`2yjlHk
zgvcD)mYm1*)KW0T5n`(6q=ln&TSm`I(oMOE3p5l1jn{m8RiO1mb#&DE_ual9Rg6Ck
zKoNmNKX(XDA{1OL>f<7cCq*(L|N0_d14M>3(^b`?7w9-`$f(*zX5TFI*#d>{LX7am
z2=R=_VILl7+F7w^YrPr6eJ4XaJvAApucOz0wb1W8`)3ecN&7`vQKMV!gmfm*O*<o8
zCv!shmOi5Ro>bZn_KJ@fXdi{sqpaQ}8!;BOA<OK`WcICEI_N3XF>RZ7TAjbKk8wy>
zVYxVRY>LG!#fo#zLGG|0<$5<{I8XGcykX3?&JP_W$_CA6p9zAw2J}9^vKSO=yrA~j
zkc6&Mw~CWVU4f3&(Y><U-g?vY=)>q^Y}CUn`0i##dfS5SOqXFHz9WNO*00Sk4V%Q1
z%IolYorMXX<B0EdNy9|tQ5^ka@3+oqMYrPkIv$&`X)fnWW@XQ3-ZK53a}mAM$@?Y4
z8R3#{Zu%}t@Cz=TiW=;tIbFveXo>0<+AfA)=lwTv={bYnYI>S6H_2JN%c$3<I~W|@
zn)FyEv%mJwd5u5VjIwiDDOF4(eT$c+MAR<f8K<IK*&Lk8pg0m{R{Mipx;DMM@N25<
z>_KY^^+krhrciAiT;y7IQA(7qh_2Mk7jsu{tJ{WkF{X;BC02{sSq}4;EY<VcX(iF!
z<4>>Fw^t)K7`<LX=`nKHgP3p7J3cn_19i@J#=)h8#}}7Kvf?^3ShJ<@Bu9XMCtvW|
zVML|GT%5DnnoTpjK_bDK^D{i%*06#zsFRD8q=#m>IiraH0l^PqEGAa_&u#R-_S66E
zA9m4nn!U+WouNdrzw9BA;_*>$mZL`~5Qt;(hbY!pAs_x?(AZ{8^IY1bn5&6uO{mCO
zDuHdl_@2X|B$^a07c$zgc9j%Uq>rY}If4;<{VreNzrD);{BM#t8bUr=ed`Rgzn4yt
zkd29xZ%<Jm(sD8GW}b-A<m*c)UfOk9f>3AvQd>e~JxQq5#jG0=XSLHh`VOT@^;cOD
zJkIaxYK-(3j3|8AMxWD*Mi6?e1tB+(%Rnu2dCikl*GK@s&8Nsp)&Cu4`#;F_!7Xxx
zM(%B3RvE7T9XebOU`K6A0&~RpV%PgEnjykFrT(_+B*WY9Js?KHofI}Tt<{=RXnh@Q
zx`HeGfJj$e!`p<tv5UxWWO}1mOJ4)oaCI5F@T^I>{?<owo;HB|qreA3$5PNWaB}Ap
zBiwCTt3N4fyscj)+X&o!jiN%pljxy4bN`XhBhV`H9PNur1~s+IqHou@0RUFnfqtVP
zP_rmx-Eh#g1pLUSMa2}P8~q~)vbaUIGV=dP1}ED?GjLst|I0c>c%}><ls!){Gmd!r
z8%u$;OiB<SC))ty@|R!<$4Y7o2q8%$L=JnzXIch<wA&b@@Q<1i-d5k;&;OvJ^?`5Q
z21wv=5vF59anCw{hOdcoD=4bE2Y{1~NkA3WXZz=9!-ciFe}-{U2YB$>Ua%Aq<Ae5S
zUv1SLo_h|dSc=*xSl8Mj#8mJuAW;=5tlRo#T+(=YFP2dyBJcR=DgqsSh^4XrUpLwR
zGm8Ws=@Re-a4P}umKkz6o&ritAUM6QJ1;m!p`vsKphJ7mm55Qu@n8-F|8u1n+9G*>
zD|cv7=xDR<K<^9*AY|2IM1XsXmlYIdqq2x~Th)e}wIkGqt<fK_0dG|!iPusM;a{WZ
zK9L6WsAH@s+1`RtIy(d4taz3&1X!;JrJ~))U<`IkJlcBjzrN4^^Fb&z0(WhYwjKU|
ze&+x6xw`(B*EcG>pY-(o7fkr!YOg6@@V!LDiu8ys^N=08i)*u&rKI6LwcEz$Qh);u
z1ddq~#lUxkm@h6&3%jG*{(YnNpPM^GL(>4_^8B7s2hzR2aUhUr`~r3~gr>*hUMGMU
zQceOR;uq6hublnt^gFZla#kD!;?M<fU2l$o-Wu{nF<H&&YsAE(Qm{II09p|?58x3h
z*^IRTH?)Q9BCsu+Cy9Exxd5Bgb@>BLZMwpWA)#HH(b^6)jrqqILV*oeKub=cEVP+Z
zdV9GY1dw^|xF4)t{IagOMBe2_h5pa6{?A8nX%Mp|#}M(qR98G`hyjmKO!KxKE&3A#
zK0BRapag)eaMScA@Dav&Lf46McAM{R0W9xa(gifqw*hb(wequ*AIoNt!!v{_UQ@>`
zzI5a{&STd*K;B}cRdfT|tJrf!6>z!wup7W&H%5?*3B;~T?{-Vu@t5+mBc3#C%VItk
zUz-yzjlfpn00iPv5b*>6_!qMt82zU7in}NWVAQ)^!HT|5zIC1s{-l0Oruysuk3QXk
z#(*GphuJk2|04j&dE|2<TDUBA_Ep96iP6$p2Snv#kdW7x{p=OVd!$Cl`@p$|g^<~|
zgW!n*K?vSY-}A_`o9MW@IBULpwgdJOOTi2gSYRv}+()wYSfOYS<7gtg1?1t!2P&Vy
zO%(?bx42rNxf2ut{-q^={l>!rsnOla!!SqSu+*5xL)%W;k8Ai2WE5^qLI4!fEtNEb
zz6_oi2SM@f)*uIXQTDF?IA#1_?{Dzn7Kpt{sm12l{+VcqV%QH9XrS8x!02b#Nbjq!
zeiC~T-&miy19B@KOHpMEnh(Y%u(zAK06i@wAb`*WfGbr+RaHEk4pepQNKzUIG%6=I
z+SKhC0%)6b|L)^JDxd>v%!jkzNr<L*;%ncwXU&7gSFs-1#|Kxy0BKhP^4x8J36b8>
zO#F$-J13J#?EmW>BOcF8104D_JqL|{ZkaL(P?$U1nXfk<Z`?<^4I|N|F9A4Nd32w<
zL;VHt7c!yX4;Z9h4+Bhxyb;2Z)qFWo(iMkfXC0%K|DP{K%jOFRkR?l-c^UZsB)5*a
zu_K=1Uq3P*d}ds-{qzZT{lSxZijV(OKg6kth0uo~k4}Pp`F(27<I#QK%?LuwgAL$G
z2@pJiYk}FiVlXquOLT3`0%N%qO=gD}22-N8Nl5AM6stPJc!7j8IlB4n34*Fd4+I?^
zn54;W4-=cw=FcXy4Dm|b=eFAr7Ac<=TVHe%ChXliu{csBQ2;d?iY@ZZ0Nzz73AmG7
za%m2vnpy7vr7r+1lx+L#HPf(EHpH4|3`B7RY4t$9+y*iYH&<#3Y!CQ>dl&%-@9gFm
zb#*Rgc(%`;69mRw_e%76N1b7E`*T7nviVdtpGeWmjr|DchU~R5fE8tG0Bq)y{qJel
z&8Pv$1(Kjc&l)t{--=$W>^3#;x8W=Slx1#nKVr@P4Fr-Zi3oB%0&TC*-tr9i1MA)E
z!EXftVC`&mKGZ;tLJ(ugHAp@dda);QM?tVRYFqbWuu9og_F-mPunkD55L_=Y2@~~+
zocvHX22<ZzIRTw@{5BsjiJmpv=7?QC!Yc-C)NJ#yLtx<=>UwvGw<c)(qz>V{X5XP3
zu+u?Ek-&7hGO1YG1P;!)xf57cJ-uSh6ZGumIe5N8dKf|IVtX(ezHpNR7wq0``D#`A
zCBYd8qI}jbTt0f#{j%@x%m^RWFoRN0%4<3H)yJ((0LC>5!q@ELdIqv{0e~pqKaRGx
z_N1V2*)qc+Ec*|w{vZ$}-qsI^ZOMP4n2TdpZ?J9x7Df@n=xVzk+c~OZ8WZnjc_O>^
zyH&P6flI@FYkiSpu%<uzbjx^4^N`7v;?L^Mgr@G`5#xym;~B_Bf!Z^#15bbqKRu-a
zfVD-S5y0UnF`)=*pk%wHdOyH5|LN;Yagv?Gc=0B2mlFt*30T<=hXnhu$MDt|9G_aI
z@l%`41D^hA)tW~A{^sA-F<c(xK7=Cm=hc2?SbnNSwW=t9DZ4J3U2|3_5S;>A#b`Iq
zp;r;O(G57(BVT#u;W>^g34CDfPhki57yQZW3nkLi3;%VHfO;jW>_hq^H+_cJQ}pfg
z1t)~B9^9FNgc-U3IvfC#5c|N@58@}+)FghYwFy5Q_Kg*|+08OG9_vAW9+Q>YdGz`B
z)pRV}fR@sDErb^^at+m=E3y3DwXaW19vmCDeBjNz#pUtl5`xUi$$452psSU5k;-2l
zXYw1o{&WbpL#(Qlx)-efnBGtyi<{m72B*O};Px;3-Ge5rEQ*;#{M8F{{^j=nY&!j(
zP&#5+b3CTdxI)PIktmI`_2bmrSS5aJUHJF-Fp1YT#eyRRE+qr01~)aATLtZp6ZP)@
zX}<IF*24I1d<6VEZkS2PKO668+}Dw9<W-Ay58aT7RHYhx{BZs5KZs6y6C5K!*z4zm
zCSf17h$~^QhRc~l=kNC>is`98t;E>G(u?~VQs~lBKmr&u<Gm+3C+u+c>FN6Z#8KLK
zqp*fnqyE=|rq?GYC5b}~$|x-evTGu;OifFR82tk&Vh3gR??2R<9{tG*V}DKjhV;*b
z@dbkUj96l*z2@omA_?wEc>r2&?jDbS+WP=YfGj^1B&u4)Is!}8HmU?}N;|T*9IAuK
zNLe$uq+sW(O*c0-^9SIYk(coNTEWHZ#AFYZ5(xH}Q;_z{StY+V>aBPekX-2mWb=ak
z4*Gd)&!Z409!tRGS?hfNJsVYLhi8Hg3>!gL@NgZ!wsw=s1d5{iX5j=!1Hf%>M_}?7
z6-0edH~r3Pz+GM&Rjw|$^@#%SAV0Qv>#%>`u(|=^hHe8>;I2W;5K&#Z-cwXtExmUi
zaAbRVR!OB^AaKe!auvq;uVJ@~m||FohF-NegJc`x!S0|pmVo@1J4kT10l@;>aL_6R
zS~VbSpGWyh9d?5G`N#cwH5&%j<;%51L($S>#^s;ok9|TI6p~{Yi5S)9)e{K45BVLt
zpPl_$F_wCyvHI=HE-JoYOJ~BWq|`IUG`r`iXWdlJdI)F44Vk4aNR;CtoU^lh*A`g=
z9T|0--;`Z{YPd?*Av6VzqU)Yt+$bd&W09G*H+)3E%&m_*{0O=4U6tb;=+__zn*}+Q
z+0sWv3m{KC&;D7djb<~{34J(QsSNOqYDakQc9@f8D~#zHK%AYo9=>x>sd8%^)A51S
zAPFCDdW^juC?-TA>5o^a%5w2dqGB6r0npc(xR|a`sv&U;*2lIbaV@S`3wY~pv+E#i
zvid&WQ+*A^$2uTFi^&N41(xPLV<w@WKgxtI{L6j8^wLS~zC*MhR!)G4;}TO$ro8}$
zx{&g*#S_uPzlyX#9A7YJ4Q*p>=`|4%`Kd$I_{YAi=tl2}*0vIniFtV-7Wmf$nflcr
zRS^V8u+@%*<M=1nz|6I_(5Dxa8S5<HtT-ywtn792KEes@2H*1jCYU|%H(kx`ovw7|
zHzr^wE2b@_ek&FRX}boMfRK0u_EdjQX6GhHpvnJY9y&NKBc1axTfbjmhdD28*`$St
zqSu&uAKI4!y(aseL@YwIzqiBu%HUOz_vyU5?cJX|?~h&sU}e&^wI?Gadz0Kd)dkqa
zi=b_CwFxi^0)fWj4OxfVzC_C+`tDEVh`mg(uWDF7$W!);PgkcX`S>w{KmehGsusxF
z|Hr;qoB)mNT53wloUOvMl%#tH-a!%EMRsnMN!-tsX1>U~z2TA#iNJp=B)ECrsfM&n
z9Dt%7XaXhITse;tR~!r5)@smNHE;Gx12|*$7z)#btkRhea(D;iH0T|v$ZrDPIrNY|
zMf1Ue28&IaY3joHsz>~5a1*o(O%g<C#(~NsE~Ec)98KYqzaXlDhQ~R>ttsstm|5*q
zF=;roN3vt%&W*kjMB}1Fu99uYp0{(M$?PxAnfPLp{0bsl%@e!%io-DOX-(u$h=D;x
zXh%Cn+vFPbDg@+?b63QLE&l*J&OwcxcPN2nRjeK7Gn{Qv1ezCzf>3bZYfQLvPggNh
z?wSaL?o^UEi87tqJnfx~AP7zM(Sp(CT%wwmBrtjR&cVLp@H{$b!+|=0f(<dkQ<h&>
zH!x6QIm9|X>+>&CB>qsE=n<3|#QXfhH;U}@VubJ?GN*ahbxq=})%Vv7k*X00%f_cq
zL}W1;6%nm^LOV?Ey$iVa=+y>7_y07W8IZOcAzOv>O=4+3G#*K`rzkMfZh2%&+c$$a
zl*;<nknQogu!Wlyw}&FQVnf7jl9lJB03k_nw#Yy|NDY@GOF&!+SYRD|+k~%1bpqS{
z$WI8eS{tmg2CV7zCz$tAm`dal9n{7T6rLu7q%TrjY$YmT?4eg}1hO0&O3p46ocyss
z>5YoQon>bvuVqqwov=x)-sLc|uDxW8JtGWRFH%tM6-iMt<a`FmP0t3VS1dEGC2Hl?
zO{43<2zi#Fp25(>kfA!vS*kJ<gz6f6O1gW0h6XgitN<GjHi77QX0tj*wrg`Rt1J61
zY9GoWvdKk>#~RBY^ekkk)XSe`xwqy~%{|6cM!E5tPlP6>5JUO+MY24(OkEd8n}U0=
zuPIb$pPlftauR22Cwn-$9yLN<)WQiBe<n>4&Hvy8ccJu0b@1R<tNiSwo^GGaUyE2T
zGsbV_BYIIi#3(^@yZN%44eq)nG-D~)^zPzx?<BQ0d!0%-#w!U;N;p~`)dR;N<;O^C
zfvRI@LpxfTKfX{8BhF;_FR-QE*qwW~(&buTTI|hm;?<;t=-=K;T)y&=|5CY<v>~k{
zJ)5xCXx%(17rt$W=L|7Zf_UL-iGtnsBNX*-4d2x7Mlxi)#T@Z(&@XPk6AQo@IeaA&
zk}J=WtzUT4-@L6n3%m4WwEf69Jf3tukfbs>ni0vwePYY~N6O1+Y$r2q#>JnW#hZfB
zh{o$b&srHi`Z$isSPgU`Ovx!zwsISTTgtq~dX*8C7O35lL$jIomH0IllUZ3hNn!ji
z=bXavp2%8_$W&C32u#0;wV2$W8A`k<U0cn<Ch7xS^$mt%is$#0O38|j?1Jy0un;J1
zSk;?9QSUc>*glXc1w)j?$>D@gXS$t*)^x9{ea|iYS}fW=Sfn4q*Hd8^U}F^J=#!=M
zJp>vRg77c#TIm*<&meiPKCB2G7fNta<Ryf+h}~Gj;YM^RBfU3#>IX$D7gP|&b3W+(
zfsAk0vuGdwn19lZ4mzcy*hS?}(WZaB?!J0smLBpj3$IG*swEr>`)KxVNd$MWO^r8N
zY7$3<QKTJ>dsHeKzMgW)^?Etl<%g11Z^pOTx;6dVS}1n}44pegP^h@+a%(ifJap1{
zF;Qn-f_K~H(@^=@#NfJ!Kd(7TenK?>TuA74RXBc3M{K-buXje3cS&iCqPe~U4DAr8
zJ(U{9l(+#e3uH9G^Mrd4_Mz$`)hkJkC1rzY9;7{@MdBUfSfDN8um_weQOxdrXQ?k{
zVaRMLEzZGr)2Prt;ZT$T>L<TSUc>m)i?Q^?*7QRa(+~U53PxF(Whgez&U&wU>p+Mh
zoYMKsdm%35;%M4xdnc8!zh7@l4U#6U``j(q8Hy1J3-6hI3{|GE7j-v!u{yeQB)mJj
z=*2r~z4REWcH|U~z0s)mS)^BJ=Jg~q4MH(s`@ZGTo^|h{#qylR^IHp(6FGPG?VgIb
z-Uh9C=FOU(<h22dSiw+br83i--vnbs?|8$)MxhN-6ihZa^&y|Kv8%H~lx8Eg|2F!3
zgrRtBPcLd#%G2M5C)xI^{#=g?hf%jbwz)YqUTvSVrAlsY?z6p=;GeO`eo`G#=dLKe
z1dsS~DY4;yj&KwyO+DSxK7r8bTGGb|C^`t`D20jfzpJL%bAY3wr7&5g>}UJt_uvp&
z#I?jdDM`uqcH1<xB2Ez54r|HmQ%pr8?!WZBy0?f`o7D6^wPEbk9HN`8#BM{yv+Xeo
z{e{mf^L7P~xJ1M@3h7JrKMf9GH%*9)_Aq#caDQ5jCzmd7WudT5N|)WShY#lYWM-$-
zV$wG-O#zckgwsl5v=IG-HSXK&!C=MqxkM5X3w@!)aiFAP`{KLg?3GkxSOC<_|M9t5
zKbEX+1mjF$3x7!pEV+X<*n`mcls?T0XGB;`j*)+RV2{xNi#%jK((5=otN&&yjs%I`
zj+-TJv@o`_!H%B7LJApnQK%svkcokez3f9D^zt5Z92=LmL)~fF=>14~!dGLjqQx4i
zt+E>W#lk9$`%!9YYFWcq_r9hMGZzq5CYzNd{0zPuNO{E@mI_CKb%X?097QNIT$h@~
z0I)^#Ql~hzYhxipa81-L(PR44p?Xq7k5Oqo7!jxU6aM%r_sz^5Eu<-x<fp+7=!H+H
z+#UN^wy-LYbe;r46&;jSOJbm)WBAcC^`?ZU6>$)_>#2ppXrlgdcb*jWE?MS4uQH1G
zp#U6O(U(|jx>2lCbmZKY<JZ6w7RpW{vp?B8T^>1Fm@lQ<)v#Ya`Uok&_3ii?L%$Rm
zw7%!>v|axJX-(8o$j`oZNitKJ2BR#0EU=Km9UJLRtX1OJ|Jr+Dq0t@I_>>*Yw0j0f
z?VP`r;`o%#xo}6B<o4_lbL!iy+s$0>UHx0PJd0z@SFom8G#_b>QO!v^hWpa%4vJ{=
zyJ(eYr6(zWUl6fW$kF?_Izxoc)yR{s{;UpMMshHiPQqR%CX=xyb!N82*dsh@MqgXF
zgXG^OB>nQb+I4?D)P=F@+1qE3*u?B+)9hvRCl!b;{4ja8Ssy=7(<Xs5YagD1E~R_0
zS9u%nnmP(~rjByOidLHTAL5rg(ES2|K#dW~!)6;oR;5f*#mWr@!g9SAUdbb4NHf?8
zJ*F4tkbObS1VZl7ERL^j_XA*_y-J_G!N#2RIIyylJD*b*m?3aVHqSD#A9~ixw~fKE
zjS^yb!(yA_m=!u4>|x|wNj}4gf{I_}$4^*;<n**f%X+3UeK`Dk?~6(Ihmp*U4<qg;
z;`O@CCl~5wy44LVZeOtd8gb2-EfVDRey8-TpBuO38MWov%~<ALFzB$y=gRc1#7%%~
zQWD2p*VqK>=vyArSwGn1y>^6RYK>}5H`!EPF~N%m*=R|PFfJ$#nN_t8<bo~_4_HZ+
zMY9G1u=^{@Mbzi{;;VFCC#^fb&I-tV+Z{=>z!#~>%!3(SY&4<bnm9GdT%!HkHHHnl
z-=0req4$e<l+%F8Hx7cS%{m9qX5ETMI||+E(Aj~Dh-!6jgvr1_B|dkBIa*d2L+E|X
zV@Dloia)NhR5&;Zp9PSuZXITA1*Zbg*Q1Kob?B4L+TpaBEA|%9;G;u-bCR_Qglw=h
zj&nDAY58)Hrm~(y+X&fk>nH0aiHv`4+u+-;iHsjeH!>0F(vK)89t&Rptk4jxqGrz#
zX#uVy>poOC;&k1#UjGxS{@1S8uUuhIlkYtE*4pZ#urR|i!>yb3r<<;ysfTSUx~8qS
z-4N@c`&^Hx!u?cquRDCDGl51kW|t<_^FL-8PT4xa1Tcw(ZG1tZyz(jxPb_|-o;?Ap
z`qn~2Va$r0-81Jc2If$OW=}F#vCATc+6*&P%gB}%=!r{nq>}yNncT0z6uVn4cr#q>
zt9M2cpK)`>DVGB3Y)5O95^4ISTujib@dszqIoYe>&0_O(LxuWox`AGlmqfwpqJw!9
z1?9Ac(apgE3P`DZCq|QRVycVa9Ijy^;@%pQM2W;1I|s^Fue4)6T~N1_rLlW5A4e`F
z>a-oK9Am1Af)lMSZ-OnFU6q*{!Gz90ebu`~gl#u;Of%L@sriBu7RM-fdFztLnIp2O
z`{{87oId%CL+y4X(9o_G-9N-qIHCTJai{mIai{5(Z*hFC%_;i~d*&t0d2gl-^BUd?
zMXLQ#_fi#A3mV;1pk_22%^3fRUjT{Rap%VQR8*H?AGZ)E#wx9F^IX@TJE$pFqR#8c
zm7mqIzo5|Y=8u5R-J7eQpFmPkb6RDA0M5|Iq<fNYMYx?)Q2Z%TuSIRYl4CPZCBi}d
zNB6;$X{75FpT2p4=0(!0%k-*O<5ICFD)(?^0ioqYv2Xo2qr8x9FZwdriO6-XBjdwO
z)AV}e?;BgsZQWK)aTcm0g?A+>(6O6Wvo1=BH_Re4q2e4Mf3x!;ruZKsa9?97q*gRP
zQH;!sp6#Eb(<*r){$FsZ9o_>UQmLVBzmr4#k4}avN>yV_!=I@OwV4@1?}bPhmD9+;
zoFYVnl$kxl+T)7Z#J0IfWsgfcF8@;=DOhm*a}f0wnN&sE>H*L8>!Qd4iBd(AgrA1q
zjSekYOqs!6gjBSDjFC|Baa{;)M2X3XePeY#+8?My(>PqJ$RpEkjuxg~AV~=vbb#~$
zwRQ5U(PWXIPWsz0e7}bnk{eP??;AtI`u^-WcvN=y3-U)73zcm-72CLKt%~`l1zW)_
z_p-(%BlzNniMB>s<+_p0Nd2|rI;%cfO@Uo@8zL$5yagn<g!G)I&5S`B9*%?4W{WnB
z^-<_)M^*5}{!vb>@{@+RYe}l-LHB=-ODKL~Fr=-K$Pcz)U3m9PV`j+dB?-m|{%MMA
zL0q!GoK{Eni*AGT!%$?#u>hU-tnC}!MDgt*mqt-2;$vUflxDeD7eF=@VH;fbuXC6p
zsb!9r4Qq=Lo0LcD5vOZw%<^4n5c^S$yD_bH21miJh*@?#1*wtP=G>BnY<VP?iC0>+
z+O=3MOF=Y$o*U0R{SPt3YwQuPH)?49+rbrt)uRRjDt=i8Yq}*CV-Mg+Gl_~=Mw~>l
zsXnJ93lWOcTm?TwEet!NLwtYudYJNvztFZ&pcl!$HP7JZS+`KP0ZHsu)(uQFOC@29
z>P^(lp<Jqdn&*^Fo8*-Gpy;+8d|MB#lS|vqX}vhD8e26b?UvA%R=nEqBy}ZPF*P=I
z!9n?WkSf?{!{sjTojVzEQcB>Wt~Wq19xjyvRp9xm7yy#8V<RPNgknW`L9W+?(McMF
z(Nq+1d2!Z0%!8QNqk!`Ox!YYTK-N1(r|uXVDR?HB;@~-93tH&4RI3ncQccX*C|mWf
z<ppz?v8$XOQRPS=Tc7Tx36vB{WDrWQ;H`MM)j45oIr+OCIQRbS;Tt0@L<9Ar0qsJJ
zxjf`J<bb-mycde)fVq}rB%k~to5R=T!T0dwH+sCS*<Z`Qybu+@`898j!BT-i0(_XT
zJ=vV}5Wyt*e3fS>8i*rkkw*H4<=Nyza$lwKV&xfj9ED~xrD!pwbJbPbYE$*Pof35e
zzGPY3nx(*&Gom%T8@^Sps@3<Gn1pE2)qMfZj?2;WL}rH<>lrj;j7Bj}cCof_`>gtU
z%S^j5n#>U@Q_{y@u2SYTd)KEqf3TI5d!5uTyt$FQd$`M?>o(6AOq(gIEh`*(__59d
zoO0oR+>5IPnXrKcn{FbiF)=Ck`#~0zs-&@shY{~s#~zq5?DEgFVVzYJE2mg|yX_4O
zx_M#-g=>6t%5g&9yC2QrVcnTz#KP$6E{D?c2c<>GlD@J38@vJ9e@jdPm(J8e6ezrS
z@e>ga{oTmY*@0KR?B+Eo24xT5_g->Yh<6K8EhFm0e_n0&EV~Jj)z#KfEu1ez(qKL8
z3a41733PZeHZ39h43*qt_5M$62((775S$U<Ls-`x?H%&-kda}eOBk)?p-JdO1ST|S
z=(yTO@&JSU@o8Ilk5441N(ia|zfM?K&ZW>eYcvXI8%tEP!PkeV_Z`vr(#OP0)@2k#
zQeJ2^+OyRCodL|7m$SYc*0Nm@f)pLcaF}+U9+2URG>(6j%wyJ+BW#(uk0EL*ASRjE
z%8o%@7vJd+$km^CZ2(0Iz*$2<HNbs1HZ5aLY9f_1uA`_UC2KqhqJ+FkBs;>*d&;kb
zPPgK;f;;VSW1Q@<ewNM4=?lmq(@wH|4HJ4Ql)EH#@qFDZI#tcJl8N=gG#$;yL7a!F
zXVIu6V#sEOwU_JVusEB^tq{mD3mrR@Aih>y4irUelPuB&M~^+a9SMz&GlR)dFG<>X
zQf4mN;c5t6aclmSrq1K3a$`{-zhRG6PoivHjYDM8f?Gn(M>PLtJ9CE4Nd3exj5%r+
zQAkO+>B8Kikf#xTO}1}zpIRTjjKw!5igmG2fDmZ%iKlgn3#4H|ARSm7^kOqQI^E9D
z1XCg`dSWTG?witFxE~g-mEFGd%MXObq(4<?WTL%qs_4q9Iu<!R%8muP3hNGR$d4Jr
z_wmoETgt_ZS?oK~o#-rmd!s6c%c!Nj?_C=yWkA~Mvm=Dw1_x({IDBB@Q|gc!2^I|B
zBN%&#0-5K65MRhhFa=9LWz99$ry>cy3}#t=)igoadVisT!FCkJQo+=F5^QXN36=Fo
z3Q&J@!>%7&<*0K=<=jPX?()>v2NHIegLfXyPl(Z)+(fc0OYRqCM2P>6av@2u4NW*8
z-BtFEioPDCmkPh83@1O5O<AW%u_s2px3E(=C{ME*Zw^1iUoZ;DxDIRyu`T)`n-r;%
zzTi0KH2waH%_{WTvMrL{TbPOh-^@Gx@Y`3B7h)JU=fP^-DLn{J?omV9r4wio;NMfp
zZORkN-<WMy78w)H!V{<JIAX3~HcyK4k=8oIf=3`SxU=_uu=(4)`u7l^C8;h$RO%!0
zOQ!?nd|itQA{8l9^O+a8X~XqUfB_yh@K*ml9vKi7DL)*nm8{WMBQrl*Y`TnEkV^}D
z(}r}z(>tLAQTidq%sLHkHiPa?fm(p}qI>F?T$_ofm~f}da3b_E<!7MY1hv4=Ve4!#
z(3JO&a|M#NfYKuvA{vOlVWtGpfFWAp{V2#kiv-Ce&<Z7mUU$1yMrObG%#SQyX(pF7
zcrRkoa%7a?$|Z909T_5Kb==HhTOt~Ku5qbjIa@$8*-KM|%03eWuOw(0GuK&{s~j2}
zJA&GMgvNz+e2=Zb&$CEp92K(HO~mud^&~Sn6HN)uJ4v$<hT0$V8dG5ru}cpLjm#NB
zkVw)`O(#{ie@=-@Z1UZFDveM(8;2b`H%<rzXAkOypO(#@yoRVrQ=|n?wmw^(0fo|>
zDNJ@f2Rq0V10nqGobLH$H()FAW$YDIsbhbAmia8L_M@BQt(3m?$%EgntwIXfcs9}$
zabf(kd^1QN^V}&Oh4Z@*!U;JJ2O4;A?vDx`WU;Zui%p!)f`0n<*4T?xE7zRH4wX7H
zqcBVY;MCa`jF^17Va#JiBUx<u1U}0;Vc}mad98vO>|Nqo`2w+5m7cm{-c!@tj??;4
zJvmAZ7w2#lcw>e7#b8~%ggCc=PC!G@i0b;&Z)^EY3)EM+qAhbm_SrgI?2UmC^(T=|
zxIcs$;fvEp#-rb>Dy!DhZFWPl+{d}D8E^S5>8QR{&UqXSW@A5nGAEp$3bqMV)^T|(
zkY|Y)z0eA&3PDvj14hKLx=|$&Cyti^{~>&)fTv8aGUU-HhIh8kHpLjG5$dOCTdCBY
zzX`Usy@H(zwb}l{GIxshfW6T|t4ek*9RFx53upe#d<uBx_-r5aO(@K>V9*`%RJvxW
zrHqHWzzM!?iJAOec#ncpfqGQx9(=uap#u_KMZW8xeYWMc+s<@UvLJNdJGcNlD$qq<
zcmd5}C{vws1M{<CFB&R1!e~M*AHqn<4q>}VO|?zDKJysg3_T@S*ssdb=T%r8Wz~~~
z7`R4l^q013d5cYRez~b*#TX?RQ%SK1eQcS?-;B)^c>HxlMd~uIxf3l5drrI>W(ho5
z6Me(^YLh<(*J0S><S``b&BSK@I;_@Zd}XEGA?6243Mo`^1*ug>OW)KcaoZvrGSa;W
zy33p2y1rtJjQNw_t<<r-#;|5GIofP%;&nIqMKAB}Yihwm0tk)k-8qo2ll9W*$j;e5
zcfXV2G1fWNFuig-a#a=IO3;$_I<+qhxeU7H>teBB7{%xp+OkQdTs~>ki5mKJcq5Th
zKu55-<Wuxz@#gci!sdEbQ{QR$NYD$z&ml+^iEf5kvG9f`Cn~x=++vr^&Psw5N+aY-
zxa9D|;ixKJ&$k~jXIAaJ^Q3Q@l4T<=^&!HE5f^w_X>#IMmMia6(#yr3sp}GLBqBYr
z`>PT`)HV7NN%>%~(p>Y;GSA?aM9{5EMhqPQHh%eT9j{zCVeY`FgvIaumCA8tiap@E
zFYlj;IGaVXslQ04X}pvO;!RIX4~W`dlB`KZic(gqc%ktm)Y#4!AHO4tfA$;Olv0Hv
zP4W#E61M*!2FruH(RVhQT@CH!U}--_!AX5EMba%Kc6l-K{qeIb9j{QVbfJ(m7tE^4
zaA#}M58(s_%*;rXYzG!k^dB^48$|M{&aY)iGn$KmP9j_SjMzM)qLhhH3r+IZ{&f5d
zBjeA^aF@j@qM``O?_`q96S(x>U?kGZHpx@rNK;9`mO8ibtIY752U9``Hh!0YVVUZP
zM<6@z?bRNFtSiLnPPF+Sn|dsFu{ElnD_{YKJ!Vfi$*`D&)0=#4$mSuJAYT*mHrs@H
z8#XtCk8tXz)5=8o@e{9Q41o}RjTksk5EeCgu{Ijv>$d)e5#SwnI)-6%1mur~uLfy_
z)2m(06CQe&b|lblpEp@#)1Bt3*Yr$VpR0W~89j+z(QdwwL}}?g&^*Qk*CkpeGZZPx
z*RaBV{X|B`!;*m#)KQKWjyXNR%&j8zkM+>~{lVC5INv=@C}z)lt4BGa2}ie!FXh8(
zRX>>qU%$X;AqgTu=T3JLsiV6osPowU*;OM*Ztf{GUloC)Cd;rzzdfXRbQ9dm7Zz8^
z(t8fQQEVvU#h)9~N$qmNh#)g+L-UMpNX*BK7eK1l;_jfxAB%5cee86RUF=(_#dDu%
z0!=Z?Ean6ct(LNnG;o!sGG)P~01rFN!}{XPFE4b|1v0>wFgzr5`Kp&AIG_Um0NHP;
zcj;Tsu1C#Q0^j-10GbOs$IEAw{kPxNUZsol3Gg~r>PTz{u*{6j8ZgrR`t$o+mp^w{
zIofCbSgb?*XmFCfEi@|XZCKL0{rf0u9mKw&*E?(4kQ0Ix>Pq*}%K$7?)8}!95JqLn
z`#Jv@O-S@sA4QU+Va+LQyT}!syDQ|qI8e;{iQt_0Db*o-+%oIM%{WQsK%`fnVzKn1
zMM3);4~vCdcB4Z<%Fs*C*%IoxW+7~(RAq=x%uVG%O^Nb!hG+4qbS4cV?v+&d`H$+n
z7EU@lP~Rh~yZqRudd`UmS$r~vCcU~UAY@qEJToOyT(>KWZFPWNdeON5Sc>OP>`j6g
z&v_X%h^#*WkEv2dcAY!HtK8ecW15>0-YuAcXUI?Mc-?YHVe+tpyQ%?hl_}!r!lQsI
z5zh8>nIr}alSaNb*W47(*M-|EF?sI-*;TYF`9gVRUa_=VR3=tTsZ=U6QP9LVDpF8L
zajdLTiKcCH+A{2eOaN&xa(Mz_FbB=0!!^UhPD9<M?i-~uST)vtLJr!Nc0W}oS|4Z7
z4Wy^maZfv%!GpQb5~n`voEV#u+@;MgfUSDDO_@~~hR%ZfW8WZI+C~An_Hl)^AQr=L
z=R@>_$f@D^;qS~JJznzk#4AB`nzwc$6I^)Xi?>Qw=j+y}yrQbqQyy_JY$!C3S1~Es
za}<YoM2J;sL?!mFhFEvAQR!mUmcDrf9m_R+&~NDwN_S~&#PRg-$pu7$cfD5<6hY*6
zbLrYV`b>iAifi*`c!|^zGrfkh&F#fJ=%MTt6Z**93{S3d=s_Q-A}-g}F%u`3F>|f7
zN$!>0{!*ZkkaOH#m2p@cau|B1Ec$piD$ts|sfX1aqk+hC*1jP+`G<Dj2lFNo>(-Lm
z*s%wi46e>U;(t8AZQ+sY%EY6p(<WJ-iyb<&saYNMs5xvF#zrDq8)a{Q0wO3ZSj7pG
z7m+jT<NbNUd%vkJQaqGr@kVXu%Yj%ZA@PWR26!|m;Fi@5q_-6PGL<;i%<Xh-xUi8U
zn~j~lM;PqKHQoM>N0@6|Yp9zfEMF_94<oFS8YF7vZ(|X`m#;IbFR+yTxWQp9_0j#%
z`%~W4ph_{p8^@^vH^oNxpDcil3=HY7vWM{Ah76_KFK`8l@4Cq<`n_ki^J*e+TQnR{
zv}yO#NL7|WhUIMq@tk5!-8+^#4ytW)JHa`BbjeHm;B7<#Ay$CsXf#rCPM+%ZslG1_
z7LsM8c{6j&!Ze#73M)G7Ln26MV5wTnI;5_q;OWkk^k-os%iL3#IZc74yy->a+50}A
ziZNodWft=)ndXMFgeCv$-$fVXDcbP_Dy8H1$PYD|;G#CaYJBe@ZGYLVlADTrVm3&7
z-E%d)d1l9YRgj?Nc*x|r@DX)NZDO9ZuMfo8ib-EG>UF8~`KW_&nXNXQbDJLN)5YW(
zE`qImgrfc6&(`$$cg^cti0NVCKc@$j7Eny{`S_iOLgII9jOgMQsAEfLfmjw+kG2Pn
z@wh)p)+CFkQ7+saJHv`l?29*fRda9UjR$P6n4QaRyU#W9ZY{6##4!1@b&%v1V9EJ>
zTz?)RSqmxUn3wN<4^RB_RsEDvtnz`c7j{DsMPcT@#md%EPm?T<jHsGgMR)u?chIkD
z2qu$BcNmzGTv(G{KAFOF)&?b8{yK*^peoo(7gy<;u&_dKN>-Ya=iqN?J4||LuUh%Q
z4wT>q0mjsmBwDtq@>uv;iXB~2e1;KCG`u?!k)B7zZ57p?iAX(XP2$Yl)rNQ_1v4fs
z(jh7jThYQ>A1<~^akq9@wF*&*h=O)b9nLVrR@QTLiUxvHBJx61;i*LlK~UY-j>v*4
zVv!XxJWkr^Txt0}xkO|AG{GpHtX0aaRcP{t98Oe5<EXuy=s1E&G2(0sPYyIf6>b$$
z^$YoNt$nH05&b@3RCLB)#|aBci3LTvHyKS?_?|lnsrAl!GOgKm?B12N5^+p}4P2>#
z5Jb;L^y21ksh;0{GB&2q7oVYh*79Ue@#B{9Fty4-S!B6U#Vg&s;8f3ijO$?`3YU9&
zRuvZ_9#FLtD{I>BalYZl;j?2yt!STG3&ksKf+&n>uS*&ZLz}I(JFu+^%Vf46>fUFV
zl6_NZhT0Cwv1qJ&Uo;O15x)M(TgLNndGt-)xAsaKSM!QnKfIS16#M%|FO{PgdRcU2
zQ+T6e5_3U;K(mc#Hw~mF5EPe=Wf)VD7aevanBr8i3ddIXvGo&KC;vE}s$#SJDdp8F
zHPb3Y*l=B~Y%~mkCLFl;YQ@dQ1x>1w6q-)<bBL7t;tpTICLSfl4<NE#F1Pf5exOhG
z{&t-en)hS-I-20DCcc%kyNY=+0xP^c8g$cNbcHtILrsoFtkV+>_!HMg>&&1_`KiTQ
z3C>FKT5*FvOyk0H-c?wym=)<*WB;h&H(EZSUwd-mhv$*^b7U&vSmCulHXL&-Ii$By
z={eZ|eR$#oC=E|rN1kEP5lB0<;MbGBg7wlSovd<Nj^$*Hw_+5vpC7FFvyqeJd!O^L
zbWnQ>j);Et%1)P=lM-g#>sha|Rx2H{9j>pmwf33cHlfHN;OE@@xDM7+BNKyR1y`U7
zfHTZEebsNN&(pVvg*FsshKuVM_wTIPmfX}U-<F2pE`dtGPl?!70iUZ?nM;BTlz0zH
zlU){OKYvxuOV^z&{z^se`nUbMJ=jNS=g1b`)UUX@!1tySvVqCjs>12}`sX%09=Um5
z3O)Cyn)7syn62=bH)k#7F4oM0O4kCxWm}S04f=6+Lqk`x^`Rg>Qg2GHf7)a@cxfYe
zxshG3|Jr39BgJ0_`>d`haZe3mh?YvY3g@`=0bvfi#dd4(p3PV&HWI2QRN5w5n&hA=
zo2^XCq0SLkEQAJr7OO1O8%aQ~KPs9t_ZIT@_3$6FJZzH61#7?0XAG-XMQ+%VqkN5D
zLZCW_`qC8fqDkDcGnqv%W-Uh|Zl(HYQjIh>RbTa`)Lo?A@?2tGeCj<Vom%~dWH>X-
z|Mlww8toYGO=s&FzUZnW<E`q#-i6^k=rz+}_0H74CPl}lkWN3PCv1`vLdwZ7$j-I<
z81%#Uh;`?0U8vsqj2G-x@oyUbWZ5Rv#d8O~L%wu49~wBF{+(fX;ymYJaQv%o?NxL}
zBc)Z=tubz3U54!nz3c+%W@f$;XRyO7;y{~jGRWth?-Sw;C9zzLmpAfzmFi?a%i=Yz
zQ!{RFJ!W(NPLRxDrebDal=AV8e~xue_Ht31Lsa~gF5%w?h4jVl5C1iPG`T}O6EHH$
zcrGQ$8T;zLj7v@48O*wDeiAbCK9of4MnEa|&HeF{f418zVvW>%*U4wESzckp+d7IK
zMq;Kf0-4W1P^nzcXOw~`L{%G%c7V!-O%rzB8br4&AXHa9WrA~u)IZ-XYQh@eAW}`*
z9drj;VWyodAzmY+K1o6N${j3N_|}2h-e_E$HvLy<6*epHR+rN9#xs;1G1v&H7B^l#
zSA>zcU%O#!5~=M=bY=>hq#^ot2l!{Ly*zIXsS?i365PNTjXiVfvF+gWXj!GoqU(@%
zC>&b~Yax=r^G9~5;bu~n<rf+I6^T%4cOttsgTp>@^S<s%onP-0zP4kKU?~%^6@vQs
z3tj#5RwGZht|%OCOe`9jR%t~d6s}C;glxS4c@epc28b^Ggmh7Wn3!FAy>wx+BbJe?
zb8qy#alGCO2yP1Ru8f;VilUa|^d$2hVf^C*^*p;MC>#b$@;sf!;V4KQK8VAA+LsK7
zFe4v)HYI}X$#`?Fh3>A8JPA3{`)!E?W_R9__S5-sZ9n?7u-TByD%sJ4M88Rp$R@x<
zGw{MU%xciYqBME>Q)DoO_bN}Hve4U2Yk5~On!s<C!WZrD$!K%ilLYhDX-%LFaj)Wf
z^|az-w-b_?DILciJg*j7(~n(yuoqGJf%BtVICpG$%#h?5+R3BzJiWx3dm$2#*+)ok
zT0U}9AyZi5AyW)XeSABKd_6bT)Kk9{CwpYvfA<H9wCX116|6CMKU;;@=cy$n36_)m
zI@U~*M|4J7>rGrE)<)8z&&wnaWl!1kkN#X{SP&Ajd?x!Ji8)-l^xX@nj+AETUgI)G
zi!q#>u0Chul?Ab47XG8r;lX@QF=8xu9vM_#SOnR_?ag*m91L_3t%hWHXqOL==qUvh
zT178Hs>B7lJBW(&ZkgUI;fE$X%ia+5ymw6ZQdaXFpnjxQe@-5Fo<7xk@c-ER?r^H#
zKK^55WE~wdn^4(%&upPc2q6*KB75)12*(O#CL<gUNtuzIk-cX|_MX4{)A!lG=Xt)*
zzrXAHT~}Avp*Va#pZk5^@7Mb^G}`8AkE!4BeER5%i@9^YCm?#80dMbG76vM6!-u>U
zH9#rv2Y=*e*!I@3R7Q#hgV7GbcIfI{@xxkVlM~0l37n#IgRg6pLH*Q|VVhK|R|}3b
zrX_zpEj1~^hQS~yzr(jZ_4M<y8^ZTmo#zNKfFsmA6pdEr@3JYgyn3W<{UQ3xoiEw#
zMnBxIUA%LEv+@i3;~~#kj}w+jrqkjjjrKpMp1vzDPV0%xFsQ1%xs6IP2e_aswm*3P
z&&?8kL{sOtA3naq5>G~YzE!MswOp6PTV3XDal-W3b=sa9ux=BN*=aJ?d}S=y;(sj1
zuMZHnIvn4kI>|0hVtev9Q|#*g@=jK<+J#$6({O9blMI9FjSk`(kDI>R(C|#?<gaP@
zT5({W0i|E5$)lQwe-U+rQe)tpW>(5QFKT0)G<T#XaMX{QG5?Y_hn3#orhuj2Y|fZW
za#O*<BS3eMCu3-pQ_s?xwUA5WBtWI-cBa=U^|dFC$uX?6J^mtFb0o^R*a8KPBXf9H
z4!=K^4Ty0lGUfA+nSME^-?lPfJxl*(?g=@PP=1#=@0z{$;{>rjOw_D9PdKed44zz<
zcP+{BGwz6;aJ@J<Rvjq|-x!tBCaXp)^5>y4OdrWjMf;{nP%4}AHLdcS#<ef}VeT_=
z%Jf?e1btU~6a?d>LnGJC_a!&DR;dWvUeNr!H%-Tyq=8u0k00xWLr^2as1T@&)y37Q
zr&s1x;=`w2Q5CN}*j7yRj`nluW30}T%rWW|KfPAu;_UJ)$8}o(3v2%Fwx|V-2Ynf7
zdZ{X>!k4bWOY`j8moT}uIV?zK9_?TE2q8w=X51i9PpnAaBXDA1Hd*1BO>gH=uxSoB
z+=|MQ9KDy4(x@W@F(+|!A2vh@zSd~PGc9|kM3*9!F>B(MMh??HU{lt2oKOgBx)GCC
znq*fNtv=B_AE+5!9Up3K$dnu+;VSA8BdnHRBYC85+==>zA|%fDjP*=DeR!WT|D+)D
zW2TFm=dCqagL(!#td#*du}SM6r*hRdP$*1U?9mG9@?}c20fuS*^R73uH~b)xovAb8
zryfkjM?r#AeH*ZO%KU;7LsE_}7fRh8Uwv$s-6q#tUP?x%9IQoaC17q`Wg_A<t3cUp
zCRduK3*zPhug!XLP)UENYfl26eDEQ&!3_Jn)Kn3xXpSO++Ipv#;m+e<>{4mMGE;;V
zK|52d%U97Atfho)^XY;G&)@(;RIQ_cr9J8mhOu-+ojnA@a?Y?Z*fV1}jl}Xid+Gc$
zaav9opcd^cmuHr<`8ga<1T(aKc!+K)Tl=8>kaS4ggN%<MUa<@|UeLK7VqKG`CQ0R1
zpCv`XANfVOY%$tZwprx%`r{B}o-x>L$uj=F*%q+CXccN!SSJ;SOzQ)ENw7_J&o;oE
z*}e=JLA7iAJ^yR7HnvGP2I`Lvq%D;F)09?9TaHqTId6M?^1T!ml1Iq_<ffh=V$q>3
z?ep%fM5$-a3>%8DZ^8HV@477+_ggcs`T<8M7YOb<dRvUQt5Gkm_jd-_-^5hlg%ULe
zB2Lk8T`INO)s#<wL#$VvIm@+%XpOq5!5opMxEdUlW?SCoh!@@XhZoK!_cg~~l|!EC
zHkkey<__xhN@?C#y0)E{rG$B@>{P#7sW&U88R>b_vD-5CH<=<-#2*grs-ouU?HV-&
zXACPMq;e+_lqeUvt}O2{p$eVbYjBNDhJyZ%aUh|Iw0d-6h1F}3L?~t-|0F8{CyaJu
z&)<N_El;>Cf0e6N%Wz%fN)Pjs_^a7mdMoECu+pDbWGn~i&#of8ln@h?51#~m?JaUv
zP7|e?dF85+&p~thrDYrINqEVLMp>z<Zg-=?CF9JI)k|kN5qbLdVEEb_e1$H^Dlpwj
zJ<zQ-uOMEX`{QH5j=03ng})nBw(z|QQ*~HXNZNMT2Av^h^{;A!p@Z$pU#3_xscK_K
zZ+3MC=QlL!(l%L+esnw+MV{%lbId+qkI}wJW0V?wl~iVEY=%6_ia7o>TYQBi4UbZT
zjD7?uwqlh=MYROj#O3Nw={|spO#krj6&byvCCI10JZG!zpJbwF7YIPCK7SW6r(v@&
zMS*$v!Ia;Him?R;>k%D#pr*>p>cjASpA|wb^l-o>^_}(wA^v~_?(q$Q9<im@RtO2^
zgM>;~>n3gF9d>ZN*2;?K*kU7<X`O<rbhH3mS+bzK+H#}TJpR>es!a*CYMo*?p{#=8
z(X~{nik}q+TEt1u{nE?pc%8+sV@kp6<s*M9mHtAw|6|$<4ScUN)lYxuVv7MTaZx0T
z{VC$tMMf|ATNd-D=C%W}ohV4Q{Obc~>P#y8+5&le*2B4c=$~nD*Pc#&sq^@$SuC7@
zAMVgm`xVI4IQhjGS3AXBPl2**V^-8qjgtcZ#7Zw`=?(+`Pb-S2hOK=c%XuGX>bXMv
zN0ca8*Hf1q?rs7nyljKmy=qRNJncnB*?5s*6`)-n33j+|0S)I%%YrI9#~(#cTA_L8
zCG}@LQf{??>@~WsB?Et$bC}14*#T9SZ%(6d58P}QlhMA19FHFCaubWq!yRH6odKFg
zi<Lw8BJD@26$QgY9hh6TXC{`Ir8PPXx}K(|g?tk=YYb+3M2aH*`Aa1j1cR|JPT3)r
zFUP4ps(%KT%AK+{Y-aqL&mB2WG9#-WB$|L>0Z50Mijz!6JfC;_+`-JUBKnEjm3}fy
zG<Zg9IPqC}j2u4iWf>@MpVc}t=1M4#L(+?7e^23uy8Ei}5kbW_ujqly<-rY?+J1>*
z!@+^JrSPX4fQH{Ak&rlYnuhJ46T)c-{!aJ)_#ni1KZC9Bx#v=Z#F%gU)yDo{h7ot&
z5eDwx)5{1+wc~Lxt;{MOhii$>U{h8-?zlOe`r`nQf<=`2NaP!s_1NhTlfImFVcIG1
zrhTBn`>nsa(LeOSrZ|gnK>9v=<`V|+D7PQ$y4Go)#gB#o)v9&{>$A^AT}Z>=2=vS1
zH!}-YedHy1R>36RK&EiLzvL~Kg2V<Y^UT)98PheVm~b{w+i>QZw${8c!aB?T>{jv$
zG6kI#U*&3N!ISw-{wRs|C|x{Um~+XxCw1Ym-^P)5^19h4AR|*SRBaeZy=0epEnZ5-
z=m029C#(ve95yqI=8V)NO#$*jOj6jukYY)MU_hPgsJlkOnW9raAL-8!e<TcjRq?ka
z@IO{Zs6gJtNc;G!qdIpJg}n=u@d#T+iR4T5jzHFQ!oR-kjW9$u0M|1^AM0uzEh(X4
z(?>Q2BMWvP8IVt?t-HnXeN?HD-Rj-qU$NA8jDYm=OH>?D-ih{?<y!P#+Z`25h}&EZ
zLJ`kM2b4K6tb#mltuU_`qj>L9B><^~MCt1ke1}FH8e4ehO)Xvy=b7IUV4zR^9BqCi
zfjBtg-}2DvhCDYj(&vIho`*#cgmPZAYBzz$-1mIH#kYT5^l&;T2!AIiB0UhF#wS9M
z=Jk%noPPHpr`siz_gcpsl-6w5gO=MY1-i)N?U0I?+hoffZ%5uZ@XBTL7~&yl?h#rZ
zmDnwN002YML1(MB2*6Eq-bas-GaWU|^gu_JwA1l0DftSzAU~tV^~D0Y!Q{)_mRC@O
zE0%gj87;j!?<k|=o3lrODkhj+Mbn2@T5BCpm*-DxBX~Nx%3VtF12FAGkm438Qb-%Y
zx1p^KMQ4`~j@TLI@+3=;7#A1uEQhfI?b`PPtXs;?D=)+FNz7FA!*vE5OaVckgi|-g
z%~gN?jNj}fkFGVdtd<_{l2lV;Gj4_!@cy1FDnAg8W83^dtiH{%{32J_1Tv8T@@*WH
zCHbJf`OEuv9BP)?9IDb_&0{D_-PRB01wZp>mtR=qYlKffOf}_C*6lSNM2*oml>Y!)
zMBN$wJiy}B-4W{%CU^%qSrF%7^)b&hr|4>!d;zu-ZhPJqjY=}ozNd$>=e&`+0Sg=k
z@5}Lr6h=!72jMYnr&fWj3EU^ENIi{(yQ8oiF~oaMdiKn;H?6B6$B^Q7{n{SQ9`#YC
zC*$f$3s&7b^8y`dpriHjG<g)kS{t2Jl=pHA7{hsq9RUwKk>np`Xy@xR$eI&CAMmtC
zEa+|*uA`rxnFBiPpvPxp6$w*7g5C5?cuwDq&HTlddk=FXkkvfN&|b`eBh|I?eklOP
zk;+1D?;JEX$$xfa6+gkWIf$$7eF%8`Cz!57fX1VHm{Vg7wzB*P{`l=~fON6_Jp9C8
zIAC_XoPQvE0txhJt!Vlz^X<?;3W|+dS^HJV%)8q)()@Pt#)@^CWACw)kz1#;2gqf}
zXtB=j%geK~Y@C4MroEf{WrmOpD@Wg+5WHLUa_HWOjL}qFqfjm6#!@>KBmeKr8|d&E
zHw8Mb72DW`esII#hjqh*gPJrX(%$}}L%F<|u@$3W@SYw)6K0SE3Y~SzohTc19>)<7
zl=nO`4T;thS0mN<8<~2Z*<a<RVQZ2UB@^t?DRb-ejd;w?7l|a-wWsbnusADkRs!kZ
zNj+a+DYM&mIowNzFN%NO8{@;Fw#EW}T*+P2E1=DW)BBxn`I|EIVg!BtE%)ZEo(|-;
zy!Bmw-qxVmv;*tJ^m<Six~P=k1l~HL{m2YZ^&q9<5!{Kzmx)Zl=&PwGk3CR);sIsh
zld;^}*JVu=jB3QN$<T!9tsKT3KH*<G8?_<x4&Fl1fYtC;Amv(VBveR7Uwg1h-TVX;
z>E~tX1~&=AWgYP_tk_e93hJAQv9CGG9tt-R3SgtADFBHv3B8o>SXvxk()r?LB;Qtc
zJc~V=EF%(gd+3K*+(Bf7Ksem4KvGM~(xQ+R7INvAEVdBd^>V0W`nl?70P{}n!OqY~
zm(EY>(Y0DNs@+9{m)21W6#Z{U4N*p9DkdHQwv*<Q1qNBqFz%ik=~!qadiw%@i5eJh
zh&8}Mwgej-f&*T+GVM+P-Koz7uxi=e4raj9CB+b)NmUth?ErvmFoVWc@+%9!iS74o
zk}tlIyzTo{HMRa|eV0g+I#31ANmTxxg6ROhfu0Ecq2lKtUV^(~XBa$nr{zOQl9@82
zl)G|_ug-vMfBdF0YR2kmotm^(k9}6Hn(vQ^`^A1=4nA*uWE40#wUvl+4lZv;Qg=VR
z(s{MKu5PVCeEH_8)`ZXTBP(;JB4G}Ht_@A#`rtp^i%dLF<t;{hOc(#X1Pb_fz_fw*
zNxJKO#-6@GMNzGZw#Rs*_5&idhO?0!Uk+%mI6T|tG8@GHETwFSJF~9hSkY#+llH@&
zQnqg~Kg#S%yur!-`Ny?uDOsXFR1vO#XoLLAZloZKS+D8xZ>=Lj#W0G#@H(c^(wQ@y
z47ZNIAW07iawJ6e^QkKPO`0`?G=#<doW|>!|7m5h`8dkJ=SG^r>yjUTb+=1ptFLDW
zYMP+MGCgWIKb0aJ5J(P(5w42p6PM)<8vRrpd)sLymrdDA)3YHA*V~WsjUIDOvk#!+
zP$smx@_dvoyrNWX;0CuOZrm7Tv)~KtO=}kFTx5Px7WKF|OCn}udDekzhI?tQ=@|D{
z@vHZbsggGq;d_Y@{AJI-`Cm7C>(l6!Q~XH%xMu3LdXMQgro0xdwZ9B)t7t2Xox;C#
zJQxqwK+{Ui*!|$zZrbH@T~$5DQr0wrvA7ZCdKe!*;&!`{E``U1ik{n~NB)jnv(mX~
z=>(E?w?Y?LvXU2JEBd}v;aj^`hd3aP+^4ZCQ0X~Yz8ai8pO{BC%UHDI;#{lVCuFx)
zYg1mdERwl%j*_QVcvB$xEu*8H^<wMXY5p_YrNl!6b~}-~gMu@-TUPBF1?6@29zR^@
zS?zRMbdp%?nY8X+yb@MLS?oDZxaS<Gq%IlEyR8U};_6?p^xRqJEdR!Ax1d$y03U<c
z+=Rc`+^z?O12QAq>{2p@(oUR)OZhhMbRt(!?}3nEtQm>M+vB0QjB>?Bkg04O9g<6-
zo~Bx5om-c8=8Q9D>hLrKi-&rq?bA1!7~)h3VT3+ci_3j{7>h22sTjyfq(l&=kKR+f
zoj=dV#&Neh)?_(9`&T>mlhJn-q+U+C{n)-k)^kF}`v+I_j9D-v0N5H4{MW<BA5Go=
z{x@$R9w^Cy50Qa4|7u$Dzkl)X|M~yl|NmeA{}-M=EY?k6u?5LvggQm%qRc|6EvbH@
z=8wKHz7gc(pX~p{mk`gUv26vl*&8iCIl3=l-+M?d+sg!o^i0_I{0QZ;u)qmFJ^-w@
zc#Lv-raxOR@WvhxMCI7VniLHB37#V0V<OLtsfJL)s>Fk=3ueHEV}Dcze3!yfEK>{i
zK6Q6P((LoI4ZkqJmLAwrI*4Y*eBTth(Q!IdcTQgNd4F^hxH@?<${e+6g#v>D_d}fO
zN8fxePNu@f9O|T+sV2fKkOpj>z>$UAr{8V@PjaVI0D?5FjnTmDAjCQ^C2#MOZGOb@
z9FO0Ty8BvLtM_atFhyJF@tHcAq?A5vAi)V<o>&oI`CxK!JTx=%B$LPw_+B+ZE*}ST
zSPX~$T8#!kD$^18VGdjjmIQDySa!eO82(G|^Pg%5Zyz45Q!}X<aEl`eqxZBD1rDgn
z4>rjs-SEc!F3!w=9GwUVxP0u`)DV@V(N{^{=M;kf_pKvAWYV3b7TAZ(fC)AU<RZj!
z8}7k)(-U+EG{D*^${X|MhtLg)(^i>t(&qqCfn0)4ra&ed6=?EfXK1EhFnP^Jur~s8
z1oG=Uzb)&4bwwzcKHcBdsB&fT|9KFM&~(I@U^w5zrF`v%n4EbsaXj~NDT4q|jtJex
zox%uU`DykwykLwTc2K({wSLe-<3I~$NT|MPmh(^G{xtz=I;n~Po)RhXEFAMu#7TmV
zCVn^u1du2}&WFl9pYGlMmp=4gKQz@xh;B6_jEe*Sd7`vO`h;+`^rqui>{1ofa$Z^}
z(0gC!tofPsm2%JNo^=+9e)eJC`{m1WEs1&$sWX7Tf#bkk<q*i52y%&g8Qb`N5YnA&
zLZ%_47xQi3*xzhFJ^XU3x#qXe8Swu3oE`w%1Y(z~Uoc9j%KxzRX?#D{fXlGvgj8I3
z!Ru8`)GadZc<Z;F6hQx}A-B>H@Z#fYy6Xx_vuC&@SSz%YZjXLBK!G-L;;yXfagaw?
zfZ)SL$Re<G0cm)w@Dj2%@}5OxI$}(N26p|x{01UmElhg%EkBQo5rB@dA+JE}FM^j0
z{_YF)ub1byWt0+gz{&cj9^68`2iCdu+Ud?LG4&3>DTsH-=vheS`EI`1+1}6xI*i<i
zQ9R`kT-W{|poc*qA3QBk0ydwQNny;-6A>7A?`p47FZ(a})g6B2iB`if^~8ews~Ry0
zI2kcaG5M!|fye$=V9moz{@~Zlkdv^<ekptzIQvvJ0UP1*ZIc(wgJ-4K-%L`*fcHid
zcv5KaM6wQRb7inm`j`oT)k9?rIpr8{=#>3MY6XZs|LIRa0O{f02S2(+Fx+gx_QxW4
zCrN)7dnK<BWacW9$zB3Sy*G;B07-HAm!QHM6JSf*TE51U^%*>oH=wa&V(kOa_%=a{
zIO|Z2{wEv6|5OEEKzcMvX7!)n7yLW{y(<0|pXoj3<L(dWqd7iF4#}KvkyObxZ=1IH
z%n+EUe7c2ps5^V^zT2zZl%pE&-NxfP$Iy|)1r9_rTdfQvnJ42*jVGX8WjYXc^<P3g
zHXekGnF8uG*B$&$3{CL`Xw5>u#%oDamy(6Ll8r|RbpM=b6>pz+yf|Hw!IIAfg95(a
ziyJU(@xTHMYf_fpj@vxy@tgERNsa-&^AH6sj@btA12XyY5}|KrV3}gk=yx{ccLcNZ
zR5ohmcYpxiWPtIFovO&V3tP`aP|D$|$;HZ017vAM5>78}5Zpl0sO(wKfYf3o?!PVY
zvUN@%^C-$NUA3p(-h*v<2b9uHR(}$y-S2l2_VY4*J^k3z8}ABf!0G$yfd}EN+{xM#
znN<o~;6=!!g~&mX0H-#{d6EFCM?)@Mdys|16hRoIr=Ee1%gIkNJQ@iFI^mGZ^`OMb
zZoUkie%}Y{p?mbKCb5|S-uodd&U$*WBXfap&(G;C>R2pJ@GY~&(VMM?I3(1+k$@w)
z2icv|K1*k6=i0(s4=*+@&U^atDPHt_V%%6PCdoa45H}1)i_s<$nKkRg1+N=2`scB%
zh54W!?EYjj+{QVup!0A64u$)*sZtqJJ^aOn4;$8jW5i~K=GFA4?+Y4RrT5Et0ge_1
zdN&bCFid_(d<CZ+zda+5OU#V!#nLIz#qXVB(DdEhNRX62<C}2luQdC{Z^dk^i+@X4
zHX$Z~XoA*>xhvr#V0$84KM84E%hN!P9jc1OugJAaY~wDR=S%MuYy#AGYPFG#3uZaB
z;HWajd%Ya?ZM&5-EW?$PL4D-z;tKjk`{}Pe_X+RG)n_%vVU>FAeti^sO7H8e3bp-Q
z5U{}l89YPunU+s2@KFo{Tg3#s(c`Acf|q+8Cg<gqkI%r?`Lp8!@`)Q6^U2(wy)Dos
zIPWfghh5`4#oKii?y<@IiQ81xI<Z>TnY?dvzUkWO2pSLGvpdKa)qSa4WAcB~7XIU2
z2297JUNXo?^O`oB&V#E-1s7utqcR5&k!t|SBR(9>GPw3URjD+%oJ}Be9s%DzJ@C7z
zBKQm^;olT~zBcS{wFoT7+_v=N7&K_jk2<f-H^~IwPqRZC3_l;Ulkn6sGH8qn(8t+>
z5Kl+2H(tW2o(WRHQ404&k3Qv~>~h^LU2FsH+ybZb>@xVENC$=CJK=A@wrCrv+#6q)
z`hX*WgV$BF^A?dWGfUNBt9~F6^CN5nIA8Ai1RK7-wT9Yip$P+)d8_MggDe?70J;#<
z!hT-shc49z+q+L&6GY9pcesDN`XS6bMoJj6P2}EZ=7Li+zGT3wa~I(#DwW?a`%(wC
zoREvNsf)b>H^4zUtmg`v^$5EYbr|0Q2G;;%fDE0gfdS^(1T+-m4m<>ldp(BPt^q4<
z$aA%r5-g)MCgxl+Hk<-rof{=<<o-ldK|DN_!o+*4bmI9~$OqrO!gJbY8!)ztK&c*x
zl-V)Iyutx`%5rFKg>G&NA2gNCFabzNllW@UTwpQTqV1$wmU27k>FmSp(Urmv^lZ12
zr+pj4D$0SSR2bM2o`r`K&oEjL`Nl?nM9Q2p)0K`NKq;z??PQ@9b~lY#F&Nc#So&Bw
z6G-N7&}K#O_$;fN0sR~Os}I`x@PRepvzfsz+VYwAEg@^Bo3lY4@F&P{+xkj*B9S|;
zx$&wZ;(K!WI&h}+5e0r!QpL<l%yo(Q0umt~`^C10Aw4BTBI#kM)oPbbeq{-~NHut}
z;i+s%7T*<W-d_nNqd<dyGH#Aj41C&7FdIU(s4R1QcYhFl_P13QtC2W%sO;g_t4LWn
zN<L4jusQG4ezo0Bp;hBD47K4b3hegC@L>RrT66@6?X#N1^&x$8q8?md-h&VJY0OK;
zxtT!0kQ=sWpEa2(>XcQ*1<$|B%?P`u!Nhq6LP-|0)_~x$?^18Yv+WYivjCHVGp)2J
zetJ^I*&XkZXq8Ndpe*4*kQ_37K5dd(e@F1HIRx5PkVhI#2s-ttvWEay;L*={QJ`dK
zq)!37haR$5?y1_$Jqy}Qb-221Gx)<=N6=t!!1@4@DZce>`}-HX-CJPi^0iui@LxAB
zu){K9aH7%yt_B@ZU7h-XYvy8yf3ET)AEMqXSizj6EgnTWWMbHsgu&)7?F({|m(4rX
z=&>U77wi3SX2Dh~MwwX*){^2wL1#j<JA=$bm=%z>3Yh_Gn>283cM8E}mMzj`3Iv|1
zx{};@gQe>(;EZ#`$UG~n00RikquC3$x)b9n&fq<1oDgdNuP^BDo5NX6#c(*h39+Bh
z4KWVS7@Oh8x&@goLyl2zNdg6MJW1d4QgwS-GG-pKxq;^?6i|Cw&!}u3-J0as`np2c
zDA|XJBt4jLGoch?dhtzyvon@$)2-y=qH{E%qj}<`Ua}sg$HC#3<8BQy@iWsCCR_k-
zHpjaLaX&bVRM!C!Ik!>bv?!IW`T&q3NkH*1HjPcGfbAsSf6%7rMK~-ltu|`8NrOT2
zsJ)iuOBk(lI7<0v=p~_9B8DQL2Kt7p)(HXD+qm-m<rq7)7oX2d(v+zc*h^}kO|nRq
z?gyKaVqiGQ;e{pA&8FdupgbaLBa$%!SOyn>gjk6Z@hId<+_fy3ZWgs$-w5=ub=n%D
zFb;Gy#lH6QS`c|8?zTRI^D&k`rfvdeMMP!${r76rFzfe*Os00QtTEN2^~D-fNneAR
zu}L*Kd$WyVbcIW|*iuFNK!RS|h2o>gA_jvz-42dP_GOw5EnY1Px~%){*zHsb5eaif
zZqos{HB`@l;Fq5HN?T}g`#Bg|T!pj8lzCziXat9OhB*grzbp?dr<JpCHo$T1C-=RN
ze+!7|6WCP7@HWedndU&c4Fz10<(f1Q&Y|+h_5o4ZfPQ08FiuwZuKnB)u<u>nAm~WO
z8!PR6VU@NWeX2qIbslqa2~2UO_Y^MQO1IlLCJw96FnWRQpU^+xk)W3lqHmI2eCHmC
zaBtqbCcz0qAf51gZ+`;|KG6Fmn@;J3IF4?C&dwRzsUj?Sa}yX8mEMEt=btmK6`W<Z
z8;riru_!XMq_g8xc&6xjWsmZg9PO_sC5rw@ZsC{(EIR<fZ3qs2?2#yr*(?4FbF@DE
z{N2zy9e5H8S+h2itG=`A7@0<*kn7l)>s*{WAnN!aN?UUN?X>rAV_#_U%ggKto<LNK
zgWz2~>vzDzt<;L*i(1cEy_q1Xrl6dbV42aFE1Pe)C2Pw@6-FKUS<tyz;A<<)qzndg
zdCn(hvxQbc7j_3JuF+N$pR({DXW?s)b_*TN0bn>v7>}mwrjcOT6#PTiRnrQ&HLJ!N
zE9Uo1ODJ4h*vOnY#Yz@Yck9{y7*Fc=_pT9KHdG%&-4b3CaVn#1$ut}(c|Oj+vI=^)
zoZz6LPo?pb9L(7ktQwoG8gtP*BMg=nY^gAAuWtt$`eb(ggH4u$^P#V`aBUq&KJdkv
zir@1Zb4V@CwU;79H``Un0b)Y2of$8aN(IKE71ohmYk$m!_maeU$#AL_Kiz&YzIs*!
zMpne|DK&KTAL|x5%C@JkT!_q$IF__a*5`OdT;7eHMo)<lP(a8jBA6_+S|QJ{Ue0Vq
z#4E+^E8RE?+4y)Y8^MC<?ZgO0aS}tEvM!Ro_Yo{v5%01LifklWNf5NHEYYIOsro{r
zjDFYB$$_N%W@1Xth4=Z4$LlH!-fV+goSxn@e25y)*BW?t+`s%ZP2aw}Z@DFu`a_AQ
z3S7f6mhdJpt6f+S-@64!oL`0flHc4+^{+_eE}u;CmmIxP->Z-{(o}rL)<>T410U-^
z2bsFiq9WZUz8w*9!MxC_L}N#$ftM*Y;aWGft!6zdqq2$|t7S)+D<k%m(x|@QSKh`(
zv?&>;PVRh+Se^sV!}mSoeXj;b)v;iSEy2QidLl%v65`P7)UMor=X#_`yxoiPC%c}t
z_R<@BVirSpDTcBAtp??@BW{jyXGi5qJq{H6uyoyR$1W62!!PrGfMNN0MjOiNgRW~r
zkL4bEX$&`x&5CAtKEeuNo~41_+%GgB+vn|#7INB)-FVG5N_NBHo*m1T`kG(oY_q*P
z9sME9u}uE>I)9VD{l`wrOo1`NU<D7>B_B?p2#;dIZkIs4Sis7_b79LbWX>>!lit;f
zGyl%;Hrs%SL)`d;N?x*=8?V@wx?cMNuEs?CQ+T@l4I|d4a4LA}9y7N%(1vAmEnoFJ
zonuIn&P0IJuMk5^r(7VCz2<alQ;7;s25T$03)yr#iEw0){3+4acjDbNw!Y~SgXkON
z2~2|uu_<oN)O_snrTqKh1D)k5D5@M-sXhIIFRGKGi{D5(wRj4T3Vs;pyUlFSS7@NL
zBlx|h;W<@N)q=p-aO2AgGm4W3N&K-6$ZN*#Pmz>{tveEgyJOCsy_|wI*E#$$rd<1V
zZwk}_S79Pj`MjkR@**pWUwu6#V>dC0n;(K=+Kz>?TnvmauV)HmtJ~xTsg#}4nM;iH
zLN?jOf@b%PjAfHPbN?EGOv~2LQ@oBJxE83}SQ2d7?slgn3R0S|utC^LUqMFI8B$LT
zaUo}zKi-17J$Fw>gS>0hYWY*vYxx1ybE*Vbsb-1++2&1a8ow)AD0hmk)S^h16o1qE
zL(K74#6QYa-Vm2UB|XzM*qwFj6{_${=JeMuv3jIX(^>?p2~T$sdT<G_8W95_;iW15
zcLqbXqDF*&<f(XMtI@<c-iZcB2oZ9UMrL-vM*M!I{=Rm{UyKX>v9kT+>v~7#SS0sj
zgCT%Wl<-;s`%-tvSkvZX`>7%red`@AiqtJH8En5|wJ~5zSYWZc)1CC*!h;v7k^VaR
ziQg|9lhqn*$II~%s@@~2sM!SKN?;q;oYWxs^G#4A6sPvQB!Yp{qfI-D^X=0?90^Zr
z<{Z<ka#Q4?+6Murbe;g)pkgy{Uf#9vQ|IwrY6jjUiycnowB`7W;t%ZuDNXoOEVK-?
zMK|>GqFE!wdahUm4;9rPmDh9L=tSug$5iOdPd9AaaUwcFN`19&zrus#SPSD0<bu$u
zZ$pt0aT|jrd1q3Tu$e6{M^tXZlLA#_gAm4>x|2Qvo7eB+l{LPRbv}5_Gv)=(kC%`U
zT+cX7+O$&ui0EVZMc{1|4_AlttU(nBPH_L&^vb6#u19|`iPfv$*@rWv&AFZvYmAF4
zcJZ5xg>|8QjhY3UI#qeCS)4h(-F7;D(+f=!G0R8BF&rK_Rv*7Ii3|sbFOaJvHOpDq
z4O++sEut+7fq<O1>~`xa@NroXRTp}tm*KtLR_|PxK%T>gG-R!Nbpd!woM3`Y_C?%G
z#dmWZ@0l~F(`NwwcztdKrmCRr7XPt>{-a=7ttN~e%GQW*WM^=rHUuwb_4CGytZ)!8
zVKla*XN#lSSVVrr1lBK>c5{TPC!PJ^Tb)>fDh>3F%M9OJK|+`ZLyXNQ!5=j2=`OGG
zlhE-g3<Fi1Vm`ndCXpNC^~Z6y!U2Xhk_>aq1yc&x=M1gZlacbZbug>FYPGsMixbDz
ziJ4iI*(1Knd9nDDB0BGPg|cS3(gG?}-fZv7r0z&rQh)+RzWbT{+fCyiAH;uteOnIG
z`c8Q?EOt80BH0vahMC0^4963N$EFCC`hR-@X33j!dz!Ov^Z4$OYB9bjv#?VSlJ|z+
zB9@0)sjt?&A&DmZHt4DfI|OZ86~2!6QRMdy&YZ-{Y2^p}j@HPUK(7Af#xHAP!KC4M
zr0b0E(%gqqgm@9Q2Dh~+2_=&vDmDvC?r_j#R7^Rfssw#P@0!VuKbUXFF1I}6NLDN<
z!CJKx>JQNjB7<F7fA6si{Db!ze1fCVInPN0(dNH>gZzMPaYghHIf{`}w!%_f2N$I<
zyj^7#CK&prle0AhI{SUhv9DRl?G#+~Z$JuhlNA%;$I`uHo|awHt*m_jx=-VGBO0X`
zx7y?6@L^q}Cdr^Xi8OvEVVP`(!^3f1ptC%VAhIpUlC>b#V#4L?r*|}=SqG|g<)9P2
z)cz3l>UYpVR!0r}YE`Xv>LE#=_=2!BW3%!%+m3WMZ{-Tt<e#SZmBVsjr3Dq~?#!BK
z*3}?`8bb8zVuG>A<_iNWjmjJ4zV}BQUMMC$dDa6l0iY-qO5?PzRne1-rd11T%EiDF
z#2OQW5n&8+N|K{(``i9@+mXL-dt5a{jfMkiG{b|Mzm9AFEvN}T$%Ym1u$j{WGv9j?
z9>heCVim{AVk-)e0_Z)4x`94SK_xIZwrzne-#sfrw8469xQsj00O-qXybKg!khk_H
z*GzC|`mnUL0h$2rpXwrv^K$Tj@9oGN&pY(cWF>};5MFL9Xr)g3NvNuq6LuTx6FSF4
zS^Xti@SR~5O$99WK!-Jkd4Kwh4uI$w9nUAm0dfq)4j5bWE0-&6RcB1qFX0+a@=W{j
z*n^s}cbZUlAG7{@`9AcGm$oX-7C)Z>EP^^2n806-TZBB}v&@4d!eOKjSdLkKX(aP9
znX*2+LjhY2eM^*l#wf#*@HLAG#7))fPRtiIy}C3Vm%(Ogte1cS2^Arrsgru#QJUoz
zUcbY_a1=gdq5s^b4@AP_tLs=xY#G@UR~m)esk3~_)#RMgZgnfi99Brn!Y|P%dv_HH
zVlW4ahnR`;>{sTKP31q0Az|w`-CHwnVs7uA-Y~-ecP<Oys5*8p5=7nul_L5i?S--$
z$!|-HV`BmRAt-9-zy2Xa`g5_;6r*HPFV*+qj=ZYq#c?O-NmF4-&HZpVXy`zmA<}SH
z61WCs5xr72tU+z2%F(?B!<Q_4KF|Hh;H$kjmJ$yg?h|Dj$gjF<{D|kN-IN!!5r-&^
z#Vd+qKEl`6HU@#WLaPMuKXpGdmI6ppI`^u3H?{;liD9D<XzJ5KhDNo$>4kxFM#j>c
z2-UNzhL{z*slHm^gaLRfP9#A@>qzPLV5g9yCY+6It`IsXia{A=dEdRx;C>|A>CTf<
zy5PbzW&AnXXPu-F&}218BY_j=vh$Y2=eoA#<8%YINVe^E`fVS>L6K4C2D?@}e1#>s
zW7TeCTr}I|NJ|4Ay7#rkX(gUF1VF<uWz}JHK&xb^**|AX@zwK7zNUxbM=NB<>@5<h
zd&iOphZC=SUq&LqU)TT0E})E&+Se~QU=!!EFc2{oA`P{z*EppGs-H-!z!4QOl?UD8
zsNv+Zk-r~HL6tXEPiWyYZ&5IORWm1X3?WHJgW&GE8z;s$-LPZNV8hvemu%I!^9#VP
zG~lW>M(@cZ>|WeFEk6)U??L$<)HtYDDxv)802?wmy6XFVeSn=y8TW?%{Hj|Z^QsQr
z{7I$*3DXQap5z9tMsM&NsVR|gL9<88)xa@-k2dsgt9<y6;w~U+d>xY-lL_^~_)icE
zcz~JgG-cgk2+8@#_M4Em9spF&=JI)P-EkJDd+b%_7p~oUVrF`?H&mp(^Dgl^lD!JG
z(~s3(i)*2@2Yc0n*&u3)Acs!wkS-~VpX#)M#P@s1{fTCka0%+#P6-@0Xhx;I#eg*q
zj3VRmKLHEO+3>@kAV8s0CTn{G91+qRjGJXlK;W#Al`BBTTcHQSZbm8D?JK!n_4i<R
zVl~*X0=)*-K3dY%khWX~Zjii?CHkwPzXdP@S-frzlB4STIAJYXJ}WxCMGcil=@$)^
zbw^zeYKlo<)M5pPjrNUmYtMa-ZP_7?n}-0|BZoVfvZ(c?z+}z9C+%{RUMb1t_BBtL
z2_w=CzPz*~5@J>m1O0Yb_^epW@DI+pJi{9M6Rr>7jIOMn3VWLFoOY5@!P_O+uMn(j
zF{{@cu*v2}sElD6Z~ZAe-vI4O`i3i!*3M&$X_+p%8ngx%TLs4#YzFs;F8!i6JQn6g
z4?l9!#%0=sjP98|`+iuF!-$zRb~#|S?=|iwh-_Qu*uKIW`e6r7k1^iKdxqGaOkf1W
zli>35Pb$D|B}_tI{_ih&G34}|(ZLQCxw%<fx0YsQIALFa_tdhBO{9NBfLhV(5pDfm
zq`O_Z`4adUUV=<vRfTV+9j%q=Hru$5gc_61npUO(BWd7y)Q$&-<>}%fgo(=Rr1Tv6
z`hL%JcOs&Ifv}AFaCaN{I<V<}Ug$JtS7oda8%M-}<+J&Clgz1`U3x1boHe7!JXVZP
z6u`7+6Q)ep=+wtfz0`aeXJ<$BahtSXay=1ienD<l4Pf7icEdPD;i5Vw%qDT|`@1*-
z*pX<mOaeMp5y<g8j7v)96RQ-nQu9>=av|K$*|CAg@IFt5Qw#O7;}}{qwVs8Oee}vm
z_a<0lk&1yTC34kzNSW!XwepS{7{y)qAU|pB7V%6ZRXwtD&qe4uhAF~>Y*jYg-ER4d
zJD{t5h4OUlib%)J%6gD9kzjCDiBhL@<QfbB0Z$R=-FgpB*dso!4&=e@(Nb3Um_iQv
zUPfd%Krdg0!}$y!&QMxoPU4xXHlYWY!@{b?mtX8gZH8=aaif;gaPQj3wgL!Uq^ET)
zKEIqhk@W^oxzu8^`Q@FbaeLit?Hp9`T4nc?k8jmunHZk``1%nYTotKXI;vu{vtTAH
zt$UzKi5_<t2ashp7yhItj@hl!TW2WXV69{*J=wt`YSA$POqwQmbE@O>KwcQ*2lcwG
zmk!NY{0Cq#{~T$7IVir6*r(u$=t>x?@GET$4QU7D)}elLPt4``8r+$iOVCud6+%{b
zd<;IPuK#kCc-<##rn+Z3*-rfn#>UkN@aZ`5S7<rmm@8A@kp|{wpGpdXJd@x4)eFG;
z;iBJ%R#kQVOL=F{L1I&7Wzvn+g2EfmmQroFc{ob1CKuCqXigks$&?PKuP|_O>a>_&
zk^~Qg=5k66kFm0ZB@mrEoN2a3uJB|`-AjEluF*3~6BKcI*=-aV65u#sr!(!=PNO>(
z9-M9esm@*!FX%PiJ*G{1`*KW$e@ChQr(=z2GQ6~5%~X8{2Q^H7yW&b`<9*%^;fA8e
zmuT)?W^*TC<HloqfM+?ukt~#Fij`3OJ&md*CWSoNv||YeafFR(o;etq)^`UK)F1;&
zLUe_lD3mN3;r8G_xboIvw9NUTDORrCLnt@Y3`F+efmpW(V@@=s`=a?lY%Q1B&;-Kp
ztDR~(r41vd>Zz3>dwFU0&h(aTESz<x+YLp`$F9PMnYGrymMQx;8-cBS+YQ43p?e9D
z90JClF%7}24Z(a3I2BFr3rkiu9s|_ZPHGH?DTwImi339R(yS%3lpjNJq%C_7Bk*0d
z580Y)J&_2*^0N%!7~Xcx5WXEndAoDgY{n<`4Ud->lwKJW;n@i4Fzsyx(%aWaZ_AQF
zxooDGTo;DkTVKzlXIp)xjaf-FzAbEgRq$`Xw#q=e_pk}#v8nyGbtuifD)lngYj{ut
zAzV!daUCzA9H}%xMp0=44vqV<@LL;RvjUm+VB(B*?a}M)3`SP+-*|z#iiC$&gEpc!
zs$Srm2;23&gEG)Q;UX+PPHCLm+>9AztgI8*7Lo*uBia;6FJ-(<-|2i;A4L8J;%l9y
z!*LJO0XK981h5bb>oZI%kMcHzE61Ixj?|YmQwU$|nOu<H2UGY`Q0=^=Q44<Q)Ye@^
z&pbStNmB8Ilix7I*|u@58-odd&qE=7c2EQl!QK0+dO2b!Jc<mTeFMKtwqyKEFu@8k
zxY(9#53*IAFwGi}P2~+x9tLA)Nrq(jM{?r!tJ@ZXxwTK#adBr;X%~3Tb@JGt^bc>4
zZFa9mVgPVb>Typ;*oe1IZ=K%8BgaD=CAsxg2s9}1ZkY`Op@WoPT6j~YDz&-1ftE>d
z0}_uIU29&tkSis1M_DMb<b5cU;l2)Anvkg0oWw@$GhU26_&tywXvAxi_i)H^^$vHD
z;lmpmYgFzGV6a)?cYB`l_W0Tc5>0ABv=9~HE9n<6a$`fK3gp*n9)hVhgDrDItlLyi
z02@d?woa=~dowzwI1t1J+eK@=AmOx<(Oyio;u;Pnb>R3O+P)Z5WA|dWR(!06b45~O
z!UiE&+oj|Kc*X>w!1)ij8?_rD<h%^S_OEbS*Y&aGuZhT^hgc4Es?O_Uz+{^L&eF8I
zh~?YiOtU~o8RYp$FqPH`SjMT<+{t^3rkHXvkK=i3jm;lzyF@tHIo}<AT3B2wZ7*Ra
zvyzA3-Ves8Qg_^DrYpgj+as-?q;@Ape1J`~a^>fm*yvSVfFC1&<Hxgc3vYe?FIo40
zoR3oD@xs3`qPb3f2DuX|r7j3j+S`F6tMp+9BX9InxX`OKKzXJTv40(jNh&_F9m*|y
zi?@-XF%+C?d6HdYHy3#PMid_}Pv)Ts-}6sLxqqOnYP`%53<@L4^@jWvRBhjJ7xApP
zVRxD98oSJqsw^y#*1MXd_jZ4@oNGqdyC6~486wP^%}?+qTcYV!XWv~@Qa9t}xHl7k
zORplMUKLISX2=_TJ9v(8M5<zGy(y2GEbN<vB6dJENdrLRieRK9)KS5S2EK$jB&g{H
zW*&toqsU>B-g>>dW8FHW*Y@%Y^J*~SZ^3;|djAZ>{5>-xnyF9O<E2z(mNx>Psb|13
zLqvm-_tE$C+F05~PQ&{j$xRfV<VciglyK5-3fic07cHFee~-o^c3QASbuBJif#Wb$
zIA}>F+dnD)R2WsDlAmNV=5>F`OTd(GutQ+^gJ))w-y+jz-`Z(@f{B~;*EC$6>D{gC
z{?*aDa>sHCO*`<x+sV8QJ<ohD2XQg|vX10sWK_VeN1pn-W6s?0yRqTuFbLSU;SOCj
zhU&Ziev<KyfMv9{E+tRZ-gixHno(Z0Ps16Hm9!t;V^4hXF-h>15s-XKl5(W(4f?+c
zl)PLA#Rfw3qKpG~KO?;s{*c0Uinn5oL@SKjh=rS3SFl|MXc9=#!Ik-9r&<do1|>uq
zxrOh|z9G*NT*FO0o$z!}9>HF~GYvq61(9joRitnG_W2_HFd9S_$ULsxl|I5UBekH)
zk}|6D{K2!b9rJt`z|43KUHrvKd)R6FiZ?$Uy?o~fnw7npl&R_jWz<p7oa-prxXF-E
zT+eD~z=^e1FFMQT2~tx^ib>o2Q9YH-SX>5NDot14kJf@ep7fovqw6ZksjO(&ozH36
zK8@HRwJhG@qdW2T+e2}B;kXF87e?uVav8U0cbV(@%e?8*qx0>42rKqmXrw=vaVoeP
z(=2C5`c589DQ^zA#QfPQ5eGBev2?q&o}T$C4h?mMd8T=|i6gD@TD<G)Ti=7f;}JN7
z55k`@&dS!KzA;B03VUhwp9BXu?$4GIBgwnX@FN4h6|qPMONj-W5{j~gy(_l8BVerK
z7I{7FKz<x<z*O5LPKexuGm?n~^L@r+B7I;{{DS-_Bpe?D*2suW99~sTB8-<&Gof|(
zG4mi$h_R@Wum0r^!Dg-3>df!5ToN0DE1N*k>;IZdD7y17{DKqJIg2Ta(-R3@gzD0q
zWO5jPY?LEAKHP)Nwj8k7w^Ym?zoDpmb!%AlHC)g1%n4os5&DyAWyeVc;Lj(Exv)f=
z`>r_!uwb&&u6+~1vv$<Sm`!iQvhi<|IzOC~e)BR!4%-?e(6yTMS=>aaB;{yrs~Dty
zJMMR=VV2pvtq|76)cx(n$Ldg*Odw<)yt8;Mtp24lf9{z}tyS7VrVY?MlUhz{3Hz{=
z%6bPL-teluUZfkdbzUZ`esw{t_EtM(yO@GPj>x&=h;vn&i>?sijc&z_Yrr8vwI^q_
zwovCGxl~<*xNS~ugj~W{X>N(UtdjWK^DM&b&Rb-WrX)_{(c%R~_T6<Gnss41qj^PI
zdyGy!K;jY9KdH{0rrcjv!72p?*Vn)(4;Yrc@9COS77xL)vs~dFZP!npC&6GHTo4Uo
zd3jePZ@bF=EFd!h6+Zn*Y}7rWRWW*#v|BTpS4YNjOyX#@Fb`ieR?AAu)p_7;cpIXw
z+~&kihcK&%$yRstY(#XACT8vPC?)V8Qh^Qg@EP;`8Q~NZWZw7qrMrTId=R2wBJXrh
z`F<P?A2vMikud^;EDKwV^iJxXtAFF7`2+T*8t$J^H~%Q}rr4z!t$i(=W#QzXK>Ack
zWoITVqCSDyzmkVD3pkN?d_rN*zv0x;ED&A!cn03)XG~HF#UG=XfL=|oUV6_%$z}em
zNrqTyOxW<lv%^*yd#;oAayQN&&mmq3zbDWwc|rSE=y*OBGp`jU6k*3>o|@6*oX1Jx
z0f>LUxv#W9v-73GTKvPWpX`9wdpuS4==xu-n=4?t>xX5m|4T6X->$|#zMzW3hNz;y
z!%hJMwW*20oDd}Q87G4h2+>Rsg8(q${c%s{<+th^-&xXNm~{YlsvwVm2%=_rDVCRa
zx<pJMHct}>c^VPVPj|xwv_!X7ult;T$)tUA6JWxHuljxdm+Q}hgi&08bi#~H2-%(8
z1Tn9ozbTcZrmJtBE12>lpdi}D>=@}$&ztfCF+OYNW}V4=W~tVB)^tPNpThk6nm;WW
zbL9Qy&F+MNTd(_Wnf`ycA^&ckmfQ#!j;|YeWMlp`-}xUe>)&4o+yvjCNM7UMzawq`
z@n7r7*cdUU$$T+9YyYGO|EF(p4{G(vpVs`2T>GaM<{#oo|I6p|c)|V4-z4}x_m>;`
zzkS)?zvG+B&;xL%U+PaD0A6bRT5!sq0RRG#K)0xQGN7)r&BgCs|L1!b@DTrYs$d&W
z`m=5^ug{7@8aSB1GfT}Ktc)gr9MI>bK7HS|sURhf&Bg=Xn4QN#O5;!Xw#RInK*q~y
z<lM$;1~jn*AUWG|>L8|>#=d4b&wrEjpf|=>&>JvCsUmu*T+JZnI)i=P**@IdHj>7A
z=ktb-!&sTG-zlJDUIxq6&K)d@vq3H2?Rc}95X$u-8e^FdJV~%E%(Ss*+DiFAd`)QG
ztX4h8QD@8)Ndk0F-VRO_;r`ogA5-jypk0Ev300>izDEm9fZ;cATuev$PHh^bDH0@5
zR)g&0CNqF$6|MXT|Kbq)tCP-cobU8x>Vg=-cAKg<O385gcj79PDhkc|%|OySy<M-4
zw*sRI?481YI?(?0^G&boNtJ0y_kmvD@#ovHP%IKj$K)`ylskA^22A--i0=Q3z2<Zj
zM1D2_PA2U`@MCU%?D64#4TwI*>))5#<fZ{%5YOvZrJg(B|C~Dc9d($Di!VUht31Fz
zdLRet-oN9qKky}PO@m0YVUSupJe)to6<Xe)B52Cseko2>-oY3HG`&W^244;%4FXJB
z8A3>@ea?SI4-ZHXnF5Uvh=%0}hBVDlbP~^HwN}^dbe7+r#D`vx0|x@XWT3xl0{YXb
zXQm8*=KA5lBNHgtwM8Kk<D*Ef!JV`0J^9?iV~6Lbhg&ls0m7M-kz$1#h}@4eH02oH
z=S9e&1@eKAp*|-RMA3R(e$|`3gPnY@vFa;$PjX1=x}-(&X-c<$zcex&Q*Z>*9gW;s
zmHU&Y|IW*LaLInk`-90*2)K;-sq8vGOMq}&U>scL89-oP;lS3nJc2mX;S%@754l7Z
z@K2Yu{4RscN)L6IJJ>yf3An2n8MQqCE7F9+pAcyRW+&Xo3Y#xrrBg0C9I*IhOPmY{
z!=|{E3`J&}fya9X_|CWAGy|W_y!{sc?Nj<L9Qf8E;D}FqAm<>`aWya+7_YgvFaC=p
zTDCn6oTY6cQMMG1hH7M)MowWvZ>qO6%d1I%B{j+6(veXZyYyx-L~UZpaV>nkt4v2;
zI*JDAw|m|fR-f-Izdui7<uWVYgOmds9z9-|m_Opy{<x@RUycElP>S&|iAEY539y6u
zU~&cj_-9#)x|krS<j2zTmpCo3!Ny$>lVkmJP57769dK838bnb)<HH@S(8qDS_7UP%
zN`G(hHzcnbeOed-MQmSz-476<UT<h00|NB%a2PL*5%S_}ma<Ot<|Tjq9dQr@`~b5a
z1a>kOf!JCzFp_k{CRFoV;4o>MrTXbRA*nVENVs>a2HIYQZ2;NG_#<{q&5vzb)1Y@s
zjN~Kj*Hi_F;PTV$ew*_M2t95DN$ljl$Nib1fD2Z)fjbTWGc&;0dj;F1b5_&yVSn34
z^X>_oMylW}AmsNqGmfT!%o7;%X(G5B$@k=GY(YOj#BlUYIaI4&O<r5XF3G*HDLWZ#
zaM_0oH=q8=^TnHsZ(H`2;&-&N;rP2NNIAmJ`}kq&H^Ba4%9rFD{U~7hXLbxRr?M!k
z5%}$rK~}?IDR6i9+-yTZc<wLOxB1wSy_=mg%rFpW_%ie$fYngp^A(uzj&Nd0>P$FK
zwjyM{{w{sQDD{<A*Ohl^=rp32-Zc1-NC4kK)r{<4Zqq-1=srYRpGBcAZaD!BY7STD
zp;Yq_PzLIGj=qOUlTkddVlr7F;!vmq>7J*D5XXM71_<Q)+X~y4fa;|ZIlUn<brj7*
zAnr8VEPt_h;kWkrQmxU~qYD|)hRX@ea%zqQ+T%eKZX_AgWG{5^u2fW0uvbfvx$)eo
z75$nk`*up+>PFv@%*N8x$kfQQp7n`|XRpT7ychQ_d+mRFopR7ADUfuu28NYd$$(RS
z`GXVxC%zL%YjrOMP=3cgyPdA+)m;|o<h|9j)E}ZHebn5s3&LGO0XenhtE%JlnS0u;
zU|rW^pobbqX-1VH_W)y(iVYBcM-N;!LFP59H$W_xZ5X}B!6mqRl0j@W1%*1_VOZe<
zo&}+R>n9FLy4Lp!4WvvDdX*V+i)wjt&x4i8f>Sq{5*#kh&kQfe3>LsPLA>vWN?dyS
zll@<-zeEA?{p3XX35kCt&f%0CpwiE{l~M9qtj&bPNVim*o@(*8T|3e{_G#mIGQy{}
zSXff?v>HGMXP?6_Ht&zx9|H}<XL|RXEPC(OP*O~9u%{72HkD&sl7#mYm71w?ZKH(R
zJCgJK!0$XStx2t?bDK}GX;zO<DH7WswDKuoue!ceo7DN*i<B?5)9YEE=opAwwCzM)
zS|n;r=g*7=M^Ag3(xhiyLP=w{w%PZV``LLj!5tPompaMRTqJ3sl$1U{G$+CE28!%Z
zz-8f1D`wCVU(T8QKr^ylge$v9BCq#~{jDN31E9mlhf%Oaxw{p_d&Vw-lE3kc+H6!c
z!4(`Fi94H&n$CIc#WKLoE{QEFS*RveY~rfTb3DD23FzA0G8d=aQD1P?7o}<?XZaur
zfuo_O%xiE5Yqco$h(-gCPC>N#W+mO*yt0(xbCCExX=izvW9nV4BBpF3hE8<CPrYH2
z7R|dWkI!$k=TAJ--H*6An6mlA))Fxn6pQjkr+xvFI7qyOEANkLG5=!Toeb!`XUXUr
z4`|S-e!V`YUG<}{5(YDdsgBugZrQS<7olm|f$HMCQpA(@51zczW8b-M5ybG!3?&Fg
zugAw(BU0(3owgZKUV~5Ozw()1-UuqH0~FCoz@Qw&v97&8QiWRhzLhKoYA0WIx~}?v
zubmH-{7cPLzbW^E1I`$)<8(c*`{s0>kePt{htK=YPO!9gs4ohmQG_)raRuTD!zm3O
z$CZN7!LgE0KcPf*xaka3NARYG;T8fm_5K(kVqmtWX$DB_+_f#L!j`1uw9em{W*`B5
zmx_Xf?K^D};V!V1o|FQ3PN);h?8*%9Oq-H8&60OxAklOAF742pA&O+7U0<LC2odI^
z=<jp&SR^hp*i$HcQi}0V5{7k5GU1n`QI8R6Z-GQg14y!Pz0H$t!m)>wpY%TmC&Pmp
zM&X-k!P1|=?{s$M%o;Fj@Sp+HZ^=(_)#Qq|m4I*{X)2tV2*Ut15CMzbHUa#o%5x1v
z=9XX*rV@OQ7BA%?c)dr)?pIAISb2)~LZY5$xNWm#1;d5ui&#WP@T^6)MB01W>~N(m
z#}?66C1Y@$3}Jr1cCypDE4Bhsncaecl%YSQ7;&37U?{|skoQC`WlJ#Js_Ol{J*uzh
zaUcVN4jk_z-dGa|pOUU}qdNJ`(p?9g<TUQmy{F1n#H<Ioxx9pQ;sYiTU9zt$Tdvv!
zgZ`}Y9Ti<>VYx{nSW-{`RTZw!%}OkJie}1Azi>N1E~ce?$XG5m7bQ@GCn%9F1F7Gk
zv&-1hI;1_kL?6Dljn7B4k()`$n#|ubBL{Z~1nZ`*$EP#$+8P@=(<7Fz|GMh~l5O)d
zSrap&1~1Z7$OJ(eScQ+-`9_`1T;1sf%q|K@mzcL;eXl!fN2<VtJn=)~GoaCVS`7gC
z&%fX*0&HdJc)_GO=j(2D&*IiCl8hI9aJDFC-!u_RbSnBExj|dv447kF2F^jqDuU;k
zm87f8!4HBj@`$b$<9_4ef#n#xP5HP{b~@8Z$X#Tuh41bZ^dGTjYOzlht|4@b1j;_)
z(z(RO5Un)n*W>Wz3(+Kv)UcdF(bkyzf#};iI7k17y!VW1y8GG&Ls39P0#S-6Rf8a1
z4AKz-C{;uhklvBdL`o>q0*0=F(m_B$dN0y@4Na6PARwXF(3^8|&s*kw*35JNzt1;y
zv0_NhZ=b!deU+l8ZAt^I4S!EJN~1nVSY5|WUmRt?n;zjW+|K)Lr+&n68sf^cVvbq_
zN$ATiqkw0!va)#6lN;C4J^s3T=ZCiD;WWn&!PLctO-o#HA1bDM6c@uy`sy|01D<qv
z3fB}9CR`3evXAcJ9pRXY&%DQOyl-SWt^vdKNOAl8g#Y2V{{O+d*x@9_ni8{2^+PnJ
zwpxn-*XP@#uWlv3{UVM`TY`gn1RH)S`gsHZzngB}TuglL99yCWN`&+K{Xee6_4Yvy
zdv2uW%po?ycB+k=6?v6PeG^bFpEI$PqSR)4FKSwOTu*03xy@w<_r64G09~4MT2?8b
z1>PfyJ?gt6v6O<m*xt?I=HOqTS)ZM9xpeP0gz-^Plb%j9Sr_04{g#&@*Tr50#!39s
za=i^?rffzKBE^C64kxZhq{WMil+~lev{Q+rnGPYc#n&7x%=#RPEs1~rC!fNU-d;0Z
zmc#G9ZU7`W!uW_%UI5`kcY(<07dYP;d-u4`kc2(41oXR=UNQImFx)J)DeSFHr5v2<
z%Z=xW(iNsrHCEur2;BMHvyxt_+yh<OuO6s<mq!+@ImITgLc^zqYo3c~uB_K{uVqla
ze+vAGHllVC%R(kKYB=EX2VDoBS;L*l^*rc3fz_13rz7ANUOhmW_(&(`Ww=>1^t+iV
z)K(~Ev*gWWy6@agnx`H|;$Sa#p{-mee2|lbUH%?YkMaRx)P!tbirlVGCq8HSnWf(L
z2{E(Fw7fXl_RKubl$(x&YalJkdHOAff&nV<AjE~|L%b|>cd!B4o(zB$fdtdOUT(}?
z%C_kJtdrdvkk6BGrE!Asnwv>sejjAstP!PxhC3g8Q#LG@n+QJw(WJ#!2?BE;UUEE5
z?<X$%9Ri^h>}r9+q+-LYI-gU%7k2*$dL0aTD<0w8sUIv3)w+2~_S`>TLAR|M>k124
zP#2pNP4YcEG&JfJZ;c&Q!LOCEn$%%vK3$JnpbJK*puuk@NZws`eM+df96R9l%UOAU
zHLf=ccM{#3H8huVM7|+d(|3b**fdX>nd`Ht2Gsj5{r*wJ-(%WS6EMl{7gtYtt_0~>
zC;wa%C6(I7cZlK&!|%2m>4dmS$QGCue9_!8e{b=8b2Ti#DffmEWa-(C=`b@Wi3Yl0
z*sZyov3l+4@Lm?_);?_IMJwj8^@vUDwyzji86XaO&VUKm4Os1qRHtStq7G(PoU*Vq
zT)Xv+huW9xdN6Zip4)Tpo>)%rUEEC-4)deoy!qeWaG>q<XVqv#{Fw#eq9ZSPdRG{o
ztusRU((Q$QN?Rvt4oZDM!W57bHwtQ=?#6;|RwUu~mpN3gq_XIJZD4R$re0j_jWWNR
z5L<1N9%y5hRc+^8tUhEj+h~&>W@Gv=BjnfedU3?K4Nwcrez75Kv0<rj9mzd1n>kCj
zpB=HZmY1@2Wc7QoppK(cuVPdu*Olz>w>R<higyxDu<o?s^sip2@Mlr)WU_T3psw^D
z>rU>b3*tF$B`|i{q`A!jrPsPQifr65D4Ob#bbq>a>4I9-@6+&#MS*y~+v7@JWi<0P
zq(5y~CS0T9Y;GiswAf6&tyKqe??UxT#tLJ5HX4Ge@r2ihmt|1#W}xq_Y}eW^dKsrr
z#})A;{Lb!?PP!$jv{ADm!)88u`N)jB3z;g@*L6FU;<UB3a8L@kt;B5HICe<okyJwa
z+ixoq43-cb7;(De_v#g0Y(#TD8`3Zv`6wF>gVNZVYwi+C{1)j^4+);j=!8!8gsSfy
zA41Z&=1>-&MNV`0RAIu(<3FSXl#VVVL=%YwSsOFdhD&(EiR*g)i!-6Yl*%{g%^rrd
zdGl@+=J*UI0=lNl6_rU<d4Z6Z!^-2*Er(Y7e3>6(DJrNxSUZ&2BcsXqq0~*GD^P2!
zP=!G{!LNHvct6vXd9P23;}|_`SH>73V>@e~nh`c-mp*HgKHKRjkgCfS2bB~kZlym|
zb)D9`^N8U+Rc(3*EF)ywN8LicK%nB@F-W-5MD%M!_3Thh=~*R@eZ{TiMRc~*y4OgX
zU9%nu;_)4BfR9JQAsSh{e?<qg_7#S0@<<3P5_mosl@bhcG|jLRZ<QS_o|m`g_%!8J
z?`!s%T~QJf37vhtpx)0khc#`Oq`}utS9&w7CncHo;m{p4rI#m|2}TSxjl1{OVxou_
z$%(wi!+!y^ke}Yb7Lkxnd?NZhZyqxz6yu7&pSS6$!`<UxR4i;~3gb%un7#ur)?=lV
zpwZ=*W3>rLf)T8@R4LYO3`UIU!xW-hdr}HO74RXf!+M{mE_88|vJ?K$^IvvYlcsbV
zrSt|Xf4hZya`&u_S`AvG8HKGJF?5Oh*z?t@J2cd)3-dNVl&gCJSJqH3T?9#$3A!Uf
zcCGx_n=@eGR8XTUhO_iwpYdNbZ(&&y3_MR0if_sa^Fv5KaSKo2{)(;Ys(q5lH*tpS
z43n1s8umya721)YjKh9)Mt&|aj}g(LAik{{gBDKCWno7bhuCVEF*s?_kO8SkMZyn!
z`g4NR<z5zDMmqVHfTvDrY7M+S^l4wvKkRsemHdJxJBER5FgHOqhaRGTa3GNW{f8R9
zOE`hMYNvd?G{^I^S~LXu4w^*ZOnWtrJ-YVcGzXRZ9s&&>!^mM@W8caq;u61x9hEMG
zMlLmAOL3)o>WlWQ8FQdNFnr~;W<U=ydqy{1I&5~P_+*?<NBpm-muBurH#j+3#FG<F
zJUL{8aDphQ9O1m7T?gt6KRKb8Fy^Ae$_24tb*NBMsIoH6K2^Tji<jvB*+ndmmCYHF
z0Ln!Wz|6VJrWc>pl&R=R^tSjV4R{AZ`ybXOg5sC|*r>~K_}L?L=>O08`hR26|M$Q9
zr+^yxoz&E{S6f$A@n+2BwBMA{Ix(pI7KNI4<!?kaCFA7du5Um-*xh16_dO>XX@iiz
zOJ7oaE55Nyfja@paRSr>*)^8G8~s^1qntVwIh;$yCNeYLsKGyTnqJIR!#qdOeFRUh
z+BUQXpWNNzN9PbV;CG|BEalbrhyIVUFgzMv2zqJJYo@HdExHnh2ewmjQha<fWTk0*
z1hGV+Mui$J1c5ZJ;n%i1OvEG31m^#W1>Yw{bDr>IVW#(LP|cq<c;`j@x7%@$M;v~C
zN9ZxwJH?Ol4JjSP`lSdS9?J582vy7_(GBl3oA6D8G2!vm$$TTCmc#pmUd9<vxOd)8
zV+Oy_njbG_U{r2ip20q_?D!hONot+|^@T6fU=r+ODErkG?O@jt<qOG?TFs&WE~;g%
zfE{ZZ3k$0>vL-OLDYfOCcxRpFT=<Bg^&OGtqksI1oS$YTSY0G4n-wnNYpg?{AP8~j
ztKV3eiDhBPY&zGBq-RKc8-oevSHXM}8&nZ0)HW1&rV|yj1+LI5Y6kcXE^n)!1WQlN
zjDAqWu71e=NdVJV!o(WrjYe8Bvlfb&^Ty&TshSerQ{|Jo@a@T_>K+#rl(60R@OUb}
zZRQ^Lyf}VGZ1sg!>I3wahI||B<;TrJ%uDW7p&9GMFT4w>RvWgv`dL1l^Tt-m@!Org
z6{Uz-J-iHId(E_bP5D}r)V*K;R{vU%!oQa95TX6vjtk=A+|oE=N91az%Fq~_n*9|j
z4N8H(lK6Y|i`zmGFUmKg#G-#93Z7)00d3S7RXe`_?hOi#@+y6_c;XYdhEG;pNi6Y^
zeC~KXikS!r-5J*_|L-B8#d&fGfPncG6;;sKIZSPlvf!i*A)W1CgO>U8-UKxeBDy)9
zVFA|=(^o^_CCnu=wo@(vXj9wjt$3fzpL1n;4jlCX=wDkjZsaU6J@O<J1i&x(tS=-e
z>L2m2&k~l{jDOdr2$`f}rh}ElJ5J>XpqX%1+U~e^zS#(v#=W$4?;7Lru|J3?EdlS6
z&9CalfpO3o#`K5O4@Y;V24^@~jwros7t&j;YS8btoI`I#TBK}~a&l8UuCa_M_J47~
zU+)g$#JWM_@!FbWCYc)v9n~f?`=mY)fgShN%K;T2*oA|Abk_T4_A1jUxYwHjh{_l#
zvHUf7EsmccPtE^le;N#qV?(YT$&nzzJ;^3L<X3_2*8!j&V5~A^iZ{y#A<<X(K!Alz
z<=yaTs}N!TPZ7fJ86i*a`aZb{5#4Yy1yn@-yht>~E0SmLMsq=BW8^0Jy*blji2B$U
zl}9{Y-NVJaB%jd7)cD<dzkt<;gCJdKe5?uML4vysqoo-IF2kXcR1OosAV02%VPKF1
z;FPdG8?UIUPM>V=)X;Ld#xY2S05;hxxESTL(&Tb2&yU|y%31B?hXZ(oDb4_#Ppisb
z9utPu6O6=SAa4(^%)D}^(wZe|^1yhS6kJ&$`9{a|n1uoElJ-xDgFRPIm2WhJgOX-B
zm2d6E4K_Jm)-~Bb{(F?uoC*Bg1S6zGFSLlF&?7&Sh!~tx&Yzm@_VG9{QJ8ENJx|@T
z)Z9SSvI$5LMb9TLU_o(-;dQ^FAthhQQf0%beWL<%Vhclsm{)es7QhCsf*q(K<7)xn
z24Yx{1zU5#WiD)|f?I8_4w~j4B++NBy?@=5Dsd9Xit5Pyl+Mr4&WcG%_X5&MUw1S3
z#(&5^q@bR7)=@afc+}!YLRM|C{*VQSb)s7<xgRxhyLKMzoSv_rUMy4&r>RZTN$!r+
zLb~}J@6jl;`Mi?*HDQ?ETh%b8yZ9;`PQq@+icF_e*}Q~!wKz}&{05>#1^trua(Q*(
zX-PpksPW4T>!QC+%fSlx=ZSF6IX|7BT>h*1fW~p8V>Ym4twSm|4iVn}t}3{;LPcH(
zbPFJ)Dx(0vj0Jz8pB`LJ=#HXjB%06Iw*ai+ce?yA5BQs7+?Z7DT@~<sM@$X&9ctLC
zdn#+vK_0W#x)A2BQ}?XSs^c3$*D$Jh!XSWHEo!+m^H$*!Gsi6O@6A)!6B9cwrXu~p
zY!%1BF6FRCmJaI0ySqv?FvwHWJ9?!Y?QPW7@lProW6_l<YhTEfRv;ZY|0duKhm+$b
zltgJw`A6ENdKF<OqIJ`VgLHJ>3s23lFp#umxjNn8+c6gRtkz~`;Hr#c;^bh3O*$H~
z`Z6krJQ98*tm57H{CKB~saaALfaH{6&Wf-tMc7Rxf)SJboJ>L4ca=e@dL*HiUVU-o
zi%m-~4Z#lov#_W9gleKH3Qmr#xS!bYWPg(U%R8?gI5q$BNw4|yN<BSnsZDx-4GT_v
zqui-KB`XNx;AvcKV>*{^f34$b_uG`Ehp9Wp1juO71N@bKbS<z|D)-`h;woCKTt;&+
z{!N3nJ91ECIx)rkVj_2~4QK7Eob@53{}z~$3B{<H$dv+}!!J0dyCiy|?;J(9>_rP@
z)x(0MOWyizVYwo0JdLx=cffXR>&&bf(LY^_QG{Jkq}IIFOKn>@Z>RqiA>cM%m}Amq
zU~7cTKx^CUn!!GLCJLW<Sj*@THMKt{TBP#ljR(K)gCNE4sbU;XEyuL?e*1x6fVz-)
zLwyUWuakoR(8y9+UiiBD%<}e&Sp#cp4!@NWv%K%L{3kCa(XhjBlRYUGlLLjuvTQ?n
zELU&BIK@_BOU%O|dHsuj@jm?97wvk92_^M}KFOsDfmCp?3`I76Bf))CmxGhIGym={
z%G$oQ6wo^Y3TX+4o5^=Cxh3z32sUm3ELyNSe@7gasZ0%H({mA3#cK^8=TgxpPBKg+
zx8xL^s#EVYgu!!mwx3h}Zl1_3iVx>lb|n4P>|W+sf|_>Nm9BQ&;qVL9{g0Yx@Z<n!
zvro}Dd6vIa*bA&++PdJ0IK{`4AuUkm_}C8`VPk79cxFC*Y`q7tSXIQtGMM$I7DI8)
zpv#EF58l+L>~5wpAzZ!O8?Azr8&Q)2z8w9E1Slie6t6k-46V1LGA>Sx{ZVQSI(TOt
zN-kC#cvRdK!IXZ5gnd>R;Y`=4yVz_qGko<*c}L0y5`ty-E57gL&V6y-sP!5?BJ=J2
zF=72UTNZ9Qg#LqHaaZo)*|HkgO%yD|(jUhQtHi&z`4sO#x)q653-vkL3DF7TrjNY=
zZiaxRCrBp@r4tZ#VToQ@3dNRd{ueHF@8p*7^{zY?r41XDO8VvmH@=QUQ;ITqo!l-_
z*6C$q_+Fhodc(NlWFPRPa+cq9-vvb`qr~{;`Q+D6IUvZ@zNmANQrn%uZ9;|7GP0ud
zL75Z;*~_M9mICu{V8D}9+al>Qu>naxnUH>#skvDovVv2xw70fLvW{{Q<R))P%K8Fq
zwVJ<pvtNB<1t4N_s~j%p2t%O0=*C_lFa~S`e#nIHKy&psy=5TZ8Qv4Styw*s3@N&h
z81j`-cpmIY^zSu%E@_Nmt1|$r>7^%=)?@XCvGMicbi~GkD?#tWzce1|<j`Eb&h~0w
zROdQay1On^ac|g%ZX_84Wb#)jbqCs+EYUMb8H=#!e2c}tKz;#h@Lz2Lz*^}MBdxth
zyGisGz_(X9&ze}6EL$a~UiR=WbR1Tn0fs;*1H*_IAKk_uF|IM)32?hl!hY~%CDGG;
zFLq8=_oyp|daOs}UT}OdR$JQ9m75%N7(wWl^;P#H0FJT}wf$9J+vZH*;!yGEv*PiL
zuv2;nd|TbO_RM19@+F4%g8Cl)_kd;FH?i{7njv9tlWtk^FQ~H2jiu7~XIVcSzt?%E
z02rl@ElO^Tpm?bJV5Dlh`*zJfCqj}=<TJA7_jNigvZ;u#E5P?{4viN}8@l5?J45Y!
zBXOAW)>h+3%jmjF?(1`D!dvO9l=p}BPZq#7b7QxM`VZ^t1JeyLkXjb(wc54s+fMDn
zTKGOZ5YR_6q4%@Q%Y;fVY3Pr>C`=7Pgi~5<uk|w30g*;s|Lhk><QEYz(FuQ~<Iz88
zB$6lGx0*vJ319)VdG=x^-+Yh>Kn+XJuqI2|V0ks)(#H>&!wy_TpOyWCpD=shD}2S=
z9=b?!8`BcQolcM9w~brXD+OU-s(dLKK5*`-A|fN_Fc6`Ljr3#Lh#C{e3WrYC0J}NP
z6eGeCurhYuf?Kw#!lz3^UYZN?snvR&pNo5Gt$k%zPAX~jeNeHx6*{qh_hhv9V@A%L
z$X-sU@VKqC4MppC+08wUnv{pEkR|G|dMt(9k;ZU_V28)yB%l(c2d(8#YBrc@Sn`7v
ziiz>F-{+#;x}sdVo6*A|XwAW-Y3o#<6EsT1g>qp#I|snnj@;9dYeX2E$8oHj4$WSl
zXq8~gN#JlXLcEi0UI)F%_{2F$1Sh?v+>J;>i+n57&vmlGS?nlf8~_Ew4y6L5c~vaI
z-75S)?pEe7A&cJrDC6xZ>EdjL7h@yjfMud)0NS2Vr*@5FH&ZJKaV02InEp62<Ho<y
z?gq#xM>Cy>PCtPmkQEgdqW2BaPAd=i%eXG6576=PS*<0Z3n=lYw{)wTHV+q4DG0pu
zM74_vXWF<lk$q9rJ7JRkt%&Fa2^^z_Htr1Stfk~!LbXVufI$**Ic-_gTVyg@U?QJy
zk~rFgC~N`b?dvFWzPPXKdG$xj<s;+g(LzyfO;Rc36Hr>Bp}DS!|G8$Ot<io_o%6VD
z7Wz8Cqeo3l16shH$%4y>Z~?Fpcaq6$j-2@KMp;5A3>T;&<S5oe&O8HL058I=oRo;7
zGtaW`niB*6P>+O)jhG}}*%U?i^ru$Gn{<i~T!N4Dco$M#4+KuoCFA%!S@S8K0DDf)
zF~@K7A58+fO*l>h?9aSTOPGCt<$Dw6V-u3}b;23B=J<_7waZs=1Wh3=!@V-Lp~6Mu
zF^hB&S>-Xl8}@7ThjA1Qa?GV6udO*7Ow4XM-X@qZ@_kN@okYYZll2d>seNVw%8X=_
z#P3a*3uCAk9c9<;Q|UJXulVF;!IvG<rIKEG(9Ja66HT)-eb-B^EmUaVMDEuluQakg
zgFec&7dU20`#Wk?N)UC9zf<13W0IyoWWTWS;hc&h@Z*G~PeZ~9DT2qgQ%eD8lM8Su
zO1Kgg1MkE!jStqd7BJQQ&!qeA&5^fz4uAaM<GyzYu$NlEhI#A=>yF58UXOpH=}1kz
z#YCh-75#ay>iM_AX>@GlO^Lba?(Nkx(A>{eU}WCW@A-cVI1>Wmfnb0||FGw9L28zq
zX4hKkzNc7213r}3bklk~#nVbHpq0Rt!i6ruRorSquM3nggF6S_xGO#eBqPZ5sfPGD
z9xn5k`uo10QnOm`5;z$E#l6miwV}2aY|FwXep;leL$0eyF~tqVtE%5(RYP80?I7i0
z`OEj578ZUUmg`~Eg-<k*5mGAUX<+R#%3ZF?tK*$0l*p&JxJE%MgIJ=#)q$-`zO{Aw
z50fe07T}?@R3Vffq+1hbJz<g(7>M3;J?<ofPwk8QmtQ#z`S!jZ=h-f$RNXsMbE?a}
z3h0|n@{uF?Te{S7)+!s(;XdP{i30TpW*gSwFxb_jY8sZ_ajq{p_|zEV_<k@s>8#**
zhn0%Ti~cwy>J168nU&g*!J^SM30TQEN|~hBM@z5sFP5fT0~^$t#)R;@pF>W)Bh;9S
zBH%J^UTwcjjh2?r)*qlhB)2@4;qnX@Q8wdVTELMhW&_Y%bGRmNzj!9klsOkZwcD^^
zqkY_@U~+2CdL=N#-bbS^_RCOxa=-HQf4rU|C8ex}S`^*?siF@kF#%}1>`70eZySOA
z@o`@hP!e6`E7Ju<AkjHZMZ%QTinPCY@R_T;$NeeFiBa;o>kQ|I+cfcq0&ydf(?6`D
z<JDJVp!>zbAJws+_a+==W!vyXd-C<f(J?QmoP}8SxAx3?i5vL4H{-Zjrg5P8StQ$H
zX7Q(tg08BGO}-ji<&{Ph8c!9V784m8^`{sOobH=&jY}PzBdF;}rpgvU+nS^nzWGr&
zV}lq8y{1g1JgSn!C)N<j{Kq^c=!JZ&cm3D~HV>N2{uIaEs!AF5-ihB=`Z(E}?}-JG
zfLM}s%H&T<dav6eC8So?Nj9AUnM4V-eZH*PT~U+Tc|!4yEO8M3rE?|XYjbU$<@zG}
z4OWEjPidv0<d;D1fD%+XZDL23pWh$Ay4sa8eVF=tNa~H}v>)}~P^1Q!=~3d{*R&nT
zi^HE2ni+ADQI$7c=h#Ro5uJ2xC=hL7*qU!SmF!!w%A~&LAEEeGdCb&-HOK&CJ{4_%
zlL8pmA7ew$aQyKH^ralCU=L<_J8~uLTkYkL3fT|&7kf~w*g2>^yGE0h*=@iT(5Spi
znMS8hC`d29FQbEFqxAA%43IuZw64&XRe}aAo;S67iVMHxLnIpEPjBmVbi5jt;ld;N
zdYLq+Sp4%lvqW}33F?g=d4Dvl7c2c~=&-(t)br9&Y*_xhNL;BJLgG@s(0=EBKuPvZ
zV^*0J=lhxHj<(<(8P7;Z@*cVY7J|~P?cwG}RHql7+osGK2t{58<~-uf&EPH^F60oM
zQ|2Cf1)`{Hbj7_Q%g+V0pj7yJaI#}364L8snor96^33cpa;dwrd;V++H@wKr`^>qa
zQ9~8jAZ4Tqsoc>dpeO(F!J@+(lyz^HrkF>x)6(fGENEX@QQKBK&51^f8fT_&<WoY>
z1>wDa(8JwWg+r1zXs4ZBf%GL&iWb;Mtc2}E`VZS_fo|1>_KU=|E%REQ{=cqmQ72re
z7dL0dsNZR-P_~Sgb;GFxb-4>#$mm&Ic{p0lVIIhvh^A3B<Y&c0D1>uwmW?Hza&tT6
zsD&eOdspsOYXi!RAi41YyMjvRczd5bpeVdowe|I%kWv1PY=J~;oeq!7O)IdJYjoCM
zFx>Lp9}11DQ%*m#7PWeDdVjrrvWb6oV_dbL^HO6Vj3u|1o|Ku1!G8Tch5aaqV)q7z
zZE13cY@CdPOl=dT$krNoon_jiPBrl}78%`<yW_nFW2NL+d`4-;43m}?K4W^bcH=<k
zX&61rc$uJ2VsW7QC)vS{t!h0{c*)@_5v9`pSm{hbZKGDd9pUA`qI37Dn@uoOdj3vh
z`*%I-LxokQMsJSCIY%tUl?+t^q4r>~y;d7j@80e9Cq_Ra;>Zekc99dQ-5+HnP2*E~
zUz<r5+v!aNFFZ!dQNAl(DD8JPXgl#Ando5i{dIgE5hdeq*iz2{!9{6Jb;T(m6CO_$
zyri8~lF>F>6)W*i9wsZy3GZ5`-?Os7ebE6R*O899+IQFS)I))rsOJf#r6VOPeyb9<
zWK)(DC@G4QcZ^4)*^Jtm_J2hAem=RUX?Lk$Chf8o3h_>?LMRdXarR-+tq;k*q1&i^
z18yJA9<e&Xo>UjT$fcAMiPF$=Is&;-j2_EuZ?v%TAroQr!oi&B9V#+yy_^#qFeN-J
zaljSmf&a5Z2WL8VqV$7F@olf?GFY}0qObJVi+TOZ?#1Do4y=;9b=wHW`jzJUJjzph
z)~WMV(V!YD{KwV!e>1)tnKX%(!)}FM*mrb;JVaBV#yGW4?;uXE%tcV6MRK}(fF(NS
zyD3urQ*)QOlo(H9RoU?3KapDe!wNPhCy{cTy-9S0-WeED2);-KcoqXpTJAcJiL0oq
zbWD@y{IK=LKuh2R$TvjF8=N@kPI^w;wW=&Ae##=ICuUWIci;TCHuHb7kU0m0{cF~%
zdlW*mE>iMqP<8^E6gMB>;Sw`z_JnGes%8EO<Wu92!%hIA>J8}LVbxqDq391F_9Q>^
z-R|P1C3<PTfv?+3rTCtqX}_g(t1}tO^U9uVwy*}f6Fs`Yq@-S*r%A_TPo(LNi+<td
z{Ey$?TT1p#;t=tDL<r>Y1Z8I?bSubSVwJkLvoccXrv8uKL<mkN2f1f6z|Q#aFKD|Q
z*g9CfP|8BUoHq!#$Ox$cdxpA8RLvVi>(NKD^rUjj!oURpQ$DPJ8wJR@96w7(bTrN}
z_`|{6iXv5x{okzme`Dc$a{Gf#tmr@ZQgaj_*rhp6*At20mjM0f{#xxN;!ByR-Pi=M
zv#0Kk_at|`#rvWkk5%V2_d_>4M8O`G+m2N3g7nho=<~=3(1rUHehs|9Lz+x9dx5?1
zRQk6a<p0MzR1yXlt+F%O;{S}z{a;`BfB1(1D)0i5X_?ml`G@g;{G`cc81Uh~EaqqQ
zA0Pk!f;#{I>m$erZ6x09LQUi_4B)wXJ-Kp!gQzX?H@g4(@7c!RW)L-SB{S(|w-6W&
z_d`W;-x5ErfkdiWfE*j0ygohJ#lQHv$|>}i_>4UkTDkdOKF<G_zYdgAdaGQzCIf-U
zK^`LSKLH&9kudm=Ht8R=5pYTPS$bhq$#xw$R&D|eVk9LFxWf}izURkAKbru0Nl)Yw
z=(_>RlsA#D`(nD|M<CJueP%97wjlS6Xg~pC>yoNnj=qcgr@lVo`32IiW<dW=Py_DU
zABk*4*OasrRAM9GQZfVY&vv<KBS2A|c>Vg7Cq%0ckky{$(sOPDpIZqNuUw6inl_{|
zMa^*=qcH{`_oA&#E4*WUCtICdQRLfQnZQ4ZYiDCx)+=Xw4v=dFr2s3p+#4WoKZU>}
zHeA2*=_)BO`&B*I@cqlu2SmUxM4kw63SA{0R#687oqrRtfHQuTt_pya%>epfA^MAA
zfL(W^Gzp@*GAoss8rA|l8Qcr0Et5fYZU&IkV{~yUAO$%A*fIzR(GlLjj9>;>CtnN`
zsJ5BkSh8m(LPaxR0@^a;YHd{=2ffwhqn)Lpr#jFlwSY-E1Ar-7kLN8tS5IBQ_)hKd
z7Y)E0LlaTBr}1&W?-|Su%m$l)r+nA|jL+{8lM(E<e)i5&Uh`TlmN|<8Frg3t&|b#x
zZf6zNtV>u66FG_ljKzWmN*L*wN&a;{ol(QpZcr+rL^k-h$@w>sQ3Fg8WjLK100VMO
z!5WND0NGUvmZSQkfEw6B<ok-diuhgwT*Mhb-~IXlvx@_x1smY8`Hg?0bN1WZB+^q`
z-q5<c5xInNanhOG!)91FKma~ddG!0sRgx1T()jRoH<Db*n%MvK<>yxai_adoaOKEI
zJ+$vXY7PENG@KEd!2lbB(uGtdOmq8jqd!?)k;nXmNgxfc;60WJN^YW1#p#!G$2oEc
zSM!j`{5iHJX3qKpVDdfdhz6pYW{_Jp&)F3MPx$&pthAW6$F8sr!A^My_-eoAG(A~K
zeuI93t)SuTsEEh{r4jp8^u=2_D-kMpvK8ff(PJHn359GD^-#$<qjM`S!B;DJ3@nO7
zi9A2)`dY3MKaZ&EQXWqCz~O7Y5v?{`*i1GK1VaXgc9!m8?>?5-c-Ys0`G2wi&{D-5
z+(9emZA{I+NX!Ck5gk&4LA^NLl(B%$PE_(ZfSD=*QWSj{_-zh6jD=V10qvX0BSI30
zp6nKsT1ieeFNFGS1L0C&zae;rMz9__h22C%^O7d;H?SlO0g5ooKan6raYkKe;Dy_y
z839CXLqPQu&%fWv8F3KKeEy2jX0hEdgSB1`Kd?7G0BG}&(ghuPL88eKJ%nLEiIQjz
zQF#-B9DUQ#4uZG7D{^BTEr!?}A7T0Fs@Wek<+6iB<&DU5I!nb_(3-QOk%j^_lR@*_
z<1H8Vzy)5T`{K)orx!4PS);Y-A+u{Pp*)#PJhv>|S<D?q)@Qe9qdtxr)gr*WKl=}u
zB?rN}PXGuM-pIniJ}V(rat0js?F{@v3(B5HrH{-uIe|cKmKoB<9f#MKi$MC^1IU3a
zJJBm9j5EG4_rb|%z};o=of|2-bt{l7*%6Q9aBpt+e{dXy{fSvAd7zbGU5|R3TPOYy
zGU}pVXh=fA;Y+2cUiBN4m?7y2%Q~&u>rfK*GzO|WIB6nXw?BLH)~yim$wR7*^ppW<
zT;f{BiK?U<5a*m0dKKNYAOLgJ-zk1zXSiN1!49njjyMK4u(2{{5#a+XYd<my8si1&
zAyy{g0WE+$!b{A65g27VS2%K8(IlKGXJnc@$@%Z)jB=Vc{T6b;^C?E`L;N@00X;n5
z>qX0u2GBUP08c`R$Dxr2c%ZJ>EBQ?lyy4-`!S@SX5Z8cF?uQ~^D={1tHD8~%5Ot10
z_fww>j(Dj2xW6!&=U)OkDTB4DMTr0S@kYaj)V+4qUQnuB_I2+7GO6clr+AOqkTBN$
z0XJeG%qhBdq7nA-&sMJSX?H`5R0CB?Z+OKgBM+Vy7UqC2poofqXGV(kb%i9YXL1p)
zK6Os8{(zI1Ffhe^i%gzp`TjTYcGU}0L;4SaB+tfA(ua!fsiQ_RC8VcJc^G!(G!^QH
zA-jRRK4(r;(}wYtVE2cV45*flK-mle1)QC|;?rNPOt5s~Qv=L8nfj}{*4rm%#~YKX
zu*0{Dz*e*F6%#0@r}|ebe*s8jDCp)#-R5Jy{WD-2wpY}DZO%4IM&WQ8y>Na13^N+&
z$^JPG)PL_<7suP#h9n&zz+!@!-9t{U*xQT%nxDD#-AlB7#2bGlRQ{W3Jbg;ufwEaG
zmy*dZ`o~O?_aDqCh13?=c_6qh3#Z3|UCVgt?d?~40N+4MTC6eq1i0b2K|yg;3Gq2f
zktufV8h#X9?{f@dh~hYKU?*<$W}V#6^2m*sEOo8qYxiF4(!|)d%f$wfid5%TkIqcg
z-|m!tR@l>IXC*T}VcOY;G@0S<^aKi@FEvHqC@Ux<dI>R>;4&u2_)l{Ex#g4U{Sc>$
zv-NObd6%tCViGT&$31Fz&)v-KtC)QpJ?nnaSYiE9&cv<N%T$Ry9$Cg^L&0m{Qek05
zE^)^idD<$Nvduvq9YBYiG132!J6ZHf!+NQqq}IOEx~k-Er?md@cmc>CHb~lqU?X0G
zT38n)lqKR8FIzMXuD^!<{_+5ZYbIhuuXW?xl6Dqg*mH_*m#~{uTEXR@tg?K)EbRXR
z9o#%(A5hA9znX5|B>>-89gm3U82`IEh~ff0U`A6tAM2B;{c&_`CxdA`+_|%wr6%Ik
z(jiNFD=Gf8SblF(g27*ln2(~RX!K9__1}=Y{LhmEO$4h3Advg%V$^S8Zc<8&;>_;6
zi<t8#qX-71leGJnekS34{)uk|QKH@o(Gi_g@VLgFz_(UFvBAo@PMoo5>UN&CBHmng
z)0JC%!@dT}-+n~@tLsfLqgtp{k!=@#1;~t21>6rUhLR$X$QP9R2OYuJ_(pFT18a~L
zfb;%gZHC^&!$^~fmlsjSC@^5A{bBzxW7T8YWfe3r2jOJyEZO=mP=i)~3Z&8P#&KK9
zvA0<r0fyV8)14yik7Lb6Bo9ZYfKK7oBW~o|GAhfoi)B<SJ80)8h^STJHB6^TS`foF
z&BAK=l-n=xMWMWIaCC#~@Hflkj`5MiX(&=M58RTbz`k++_0v|Equb6z<&<a7G3#8C
z+f6p|i$Ls~>1F-mkLwbMtfdAMXRd6I*HRBLrY}r^?uJwNXt*6^Fj$A(>5rNN(NsM>
zG!)J8W~#uh{;DALM#uT$;@(Wj#9tUO1CY(Y5=vXLYp2}U<OS6h4lceMldqMYs=fa6
z>2v<P;`XbqmB~QcNfPz(#wuA_J{3twC!l^f?mAN&FRP|@oUwOy!?zp2t=j@*9Xc=}
z?Z2XZAN8!57v5G1QwHvjn$s?5ye>Nh{w|AVIMmrSuJBJAj~r1<0q1L-EQf$j7Yg(v
z)d?QqBJ7vBjGOy7JJVJp<RP(Z6ER=38Fa?ZG>#Dn=|r$3$0V!C%WP-*3jZZetj9S5
z9XOYrXUHaOz~)OdP!f<Yw;QzlYSLA>RSR0mYPm{PxLW|=oZt3=nQARZV=vGZ5%^|H
zfCjtc+$Q!(d`D<&j^d%aX90}FWQ_O-PHNcGkXqrbhNX9Uh$>cMutvRrsd}?4>hfN2
zMGJm5Etc(h+57S08q8VSx{xDXcfDS7Fy}z#6#DpAcCyW0z1#BOtyH&jewGWfHz3|%
z1Tg#)k2gX(3#F^mk#IU$8##(*aecO6wYv04_H7*Z??~OGyXV{*77uw*R`CteOo_lN
zfR<}2W1S3TZx{ozbGVHK84c&IrH_J?sm>8mZ)+Md<uit^36BECFCu)J;`5jH5#c_>
zg+iTF85PxW|9C3G5Nv##Kiv<!dbauXtAPYJDOUpMO1Kh=w*SQuz#l<X<Q&`?Nb|b3
zv=;=F37j=5Tuccc&9j(e+oEGh=He|5c__Xq&;b{oH!;7uAqE5qDb#i{Cz}e4S2?;L
zQuPM`%;wwl9f03#f*GR9<CbB^?F$QZe8nVvu86iU`hfR<qsCEGTn(fkRh3+b)-kyS
zX_SNv2=Ef9H=TPGt&Yg<3GS6ZGGZ>a*)gzw-~8pTd#OJZ^`xZuO$UEL=~kR3t?Y@@
zsPEs?x^R?DGui!1Mk>&ok6%j#UUO<B5jc?l^iDImMD38BX$*9MWR|8w?_T>mh0>^W
zYe+-oK)YqucSCw)754fEv(}&joOiVsTq!1Fdk^F<BX*X^u#0%$3gP#Rh1?X|w>cZ=
zeiLyL_8sg6*`Sw%3N{}oH(03&wJ*I1cfmAbDZ}Vb83$?27eMmXmh*#e&=5|BbU}6@
z1vxpR@ESly8nCCDz1}WtDX06kyACLNJf%0C0<&1Uoq=|_M!6Z(vA(trMa!-f@c5B%
zk-I9IizQl1C@o?RI0TVt_1LfnLJF6IdIM+!?x85c>7I*d5L0FSxSL|2`~z0urWm(h
z(}l(jLR~katEw&tI$<q0Sv^|mlz)U!*uDYE>c(?-cStael<rUsJZu*^8Udl0;pbFD
zCh?f|K(95jzT-+^-XeWiXBq^hnPHYs2SsQ=QG97^M-MQYOtn6&ZIn6QvCe!Fw8kxn
z$#w);65YecBOtb%f#~)&{w<x1TpZ_u3&KLSmLDE2J`YEe`|aqp>R=zoXT?1$38|<8
z!pCNySy<TPkPd?cZQ*_e-3vy?9|vD#o^8H-<faMuRKjm|$!jLE^Jm^%Jb{vIn1R{6
zJj(pFu3ucNE0vM{!@O)=Z4<9zbILUREs^2KNkxOs+8}PZqsOI4w$FlhG%d85CExXO
z`~K39-tN!bSi1n}>1%rk!LC`cn)<X?hq8TrtD7P0%YMNMHM{!}EuN^Wa0%$qd@uX4
zr=W%NsXS8dUNvVujw-q*O99ATJtu4a9r~{a{0!%!8=jT8Jg+G5NU`euaZAg%Su?AF
z`uoxO#J=&mnkOG`R>Zi{rawao%%|0YH_5Wq!Z*&WwE?IeE3yTP1!~m-q+Ahx%+~*b
zwnM2A<2Z;IsE+AI9A|{+W0P5WwRK`wbW(~U+HK4#Jb;&=;>3^Ko^g!zAmVlg`coKO
zCAa2dy%fMNj5H>28d=re2_4Y>jGl9nrW}$PAqMI8nM$ySNCNh;JBwvNbcp?lbVat2
zd$M3ssro1Y9VL}uU9R6WQYsEv)Z^&*W6WY#Dz+Hyxx^CWxb&H7ZOqcEy-<N7<LwRx
zpg+x^qRy~?GEC(3my|`AJ5o<nxeM@2@7N|54q`$E5j;cL4ze?<vLa27EBE(ySWFy^
zM)d`(z+>f;xSwFcA;MY`vI18LFMF0@kzQcaTw-H3EKnKOh)962>yJ(ejD%`DIzRrs
zHEj@+wna>D5v9FX{(Ol|alQ=t_AH5NB%LZb-M(lfWu_6MDkGy-`{%S3W9vE+Rv%}5
zyZ=t=v%xxe<hGakHSJg5GF;$iSiBB@Le<>@FmcN`sC)lG3%|~`o`T0W-b;XXbX13k
zbIwU5su|GrM%#+gsfltFxO5pUjLM&D2Q5I^dKpniA-K|>M}^D?5laWuf%UsHnxY)8
zpZR~k-_Rz$e0k6DhY;miz^pdd_da@D;rOhHGa^cA;faz=;@H`$Px9VcNrmk5;ML*~
z%M7aRlB@()@c6V(brhGvH}h&@iycV|Z8-ThxLq07!l<6s8OQHGOnuW!{dAJw$?D8J
z#p-0CaAeY8ixF0LB-z_zqbmMP_|2EsKebmKxGUN*3Vjj6DAH`ewG%keRgQ~&XdD|C
zm(#PG4g-xMOu@ZaM-h#Xd=`{^tGglD^RjgRH@6O(v2XXvvq6%tTTdaGUT(s{)B+cQ
zeAKcRSJieY4stJ{W3cliRb48IzwP5_ly9}+vHOcVXzMG!UMMgs54LWP@y19vWNAKW
zd+!{g{-th}0y)_F&ec`u{%ebC@g9}ywe{G>uW&yWjsM`lt3fy@n_0;3qR311y4rsL
zzQA31ReMtrLtVd38WJ@b4hvJV38|2I!NdJBz3FVq*4p_hp@7~>tFBtgc3@?}qyM$K
zdg6C5_JS$xY@iZVCSF`f-=iFysjM=awEA@P`tnPDuVyL#gUnL&YFA3Z^;pZU;371u
z^Zj@<G>q~-p{rJ1k)Zotc5+iP{T7CslE%~P#WRR*h4ka;ibsWpt!5!#f__Edmu~ym
zA9ZEkHIpRp>e)Rfh~N|Z)!9y_+sbC5JT9i`-=39ZRfJ$Y7J=ug-NDuNH$gj*luvPU
zx-G}wgWM1+#ipYrCwR~9+-OatIHfRjx%lC<O1f-Z<(JIej0jvy!UP%8=g`8GCId|q
zj&B<j+<M52vwKwbw!VCVKc;Gg`The$6OE}!X+>iuhX_WG3tU$9hUdaWWhBZ^^#qc4
z)&PyY;^NqxTbm-*Z819Hh_hLv{ZmjfuD(5rxd%^DlKq=Pi^DQqb3j@7De*UKyY}%&
zN3*ylc{Je$LyN$`Uc`I!g>VK534M<}rOF@4jvecqFF^jIYk%eOmkW|wJAH1;T02q;
z<JU9Byd~npv=g_v=&6U2dkfa%&g@cHed^-t{0=67J0hWKa$Xreb`1Rz|7_HTm9#Vq
zNU6;3)L58tOPo%aKe+(OHH#<r!}C{Fz<ch@zEM2!c<_+ilb$39Qt|Wc`+so@+7Tti
zByr~5o<UOeSO%mFWNEU|ay*EWgV~>hQDCNlvV$vc5lB!m3z|#<&2b5x-7V(y`U8l1
zX>Y!?CCU(;n8#HzF#jJxkiID|oNxy@jPa+uk7Xg<S)@}kCZ{LZ4L8*y%mJmcy%3Gk
zqx_=sKttY*Ua3&8Btp*`m2$G|z@^z-L)R|M_A-B6>DY{AmUa0ZqUVkK?Vk4t_KP=W
z47EM($)`Owo><*`<cuNQOftSbL#jPSC3wMlLP#UabKbg!!>^}8d3ge30Z%g2;zWs_
zewX5BFx(p`!}N09xWQq}yimqw-+aK?DmIVW@kVIFzMpG;zg=%%F_x+Nw`VF*8jdr$
zz#r*j8$+3GZKgyktP(e96;h9kQ$Rg+pE;sTwU$TFW-BM>K7c!HIcwVJv?JZS$RnF-
z_t75Cu;sfnArEh<HE$=37OGO}PKkjn5A6)fh)MdF&>?+6(%VGLl^fTA7?tSVvi&G^
z>}?@FUntJw#sK#sSp>Z)MK4#93m4K<=EKs2PVs}e4DE^3d~I%@nhO)D{0e(<uT>WD
z>++B+Yf4p3G^M$5jM}~5re3{-XGHgv6$cI+p@tzRd7vLopI7toRDN1Gs<Ek6RehsJ
zMthIxx%m`45Xk&^emPxn(OrHf-L_s~XJ?Yq+?McrNTCT~M=^XEh<}R<$r8uNw{1Gs
z=^f%Wca!^$b;d%{t1Gpm(ig->)|qaqdB=k~$1}J8H6!xH6FynTbSK(L^{>{)&LOE>
zqM{2_1dRuapo|=!Ke|M4%MhAQ9>wkS@t*Aw9n1pI^sh2`rbSZMc0zJ==>xm~8oc|c
z0bf^MbWnEJG_lxTg>yg~ECoij_2uXPt3K(!47mP(>EwR<|GvoaCTNrf&)j;B1GQc{
z*qS*5fCuU3bAFd82D^zk{pe-nK%g8I$2Ch67hLTSc`HRyf7U-$&jNINyGQRexx3^E
z=gECyrseJ2-Um?!LfH}YXv;f_e@t0_1)wEf7F+FI?Xc9{mg3|iTw=hHeWK31r<~5u
z$7euvjrG=^ZZ1Fj%cy@)7!*!}a^N&rCN{n7uKYu+`8)Q6$5;4jU#0o=*PaFe7`?MY
z*qmCJdoF3qA%)qnvEzIh`Ztfa%=A;+7T<%-8TB^v!v!cr#Ep*0H5QuS02m)l@^8iJ
z%9DFNsU4Sm7*fIn>ph0NpVwh;QPBON;!Ya)&P5pqZ7qtA>a3_K11pNdQvf(9Pv#T*
zNn#vlKu`(|ewRN9;ASLrw<=oXNsg;5QA8kfZ>#4UN<J^V3M%}t6mwnapec@@9<q(f
z@~Pov$PbUQn%zL1N`k{UdWK>y=~xR513ZP<(xOjesO8Aw8Oj*~X(Jd-`;Rf`nfw~r
zzvs3DlRQg64nk~li~iCxuAzU9*xOb!Z~X%+jk{{rNzO_|1A78{C7W|J8*}oKZjA5r
zrt5ut?0<+3ooCD~fhmstkCWENly4@WVebjhl{a7sQ>j&aA~jn_OpQ}a&Q_0i*PV#Z
zT66HV39z@Gv`Uc;M8D(ne#VW64$nFIe^kQa5nqT7dG$~6yhXiam_Hp6E}>I=didXS
zH?VRP<WRDkf{z*ABZ$ev^;%(BV$?iKq<5&TSt<bT_>L3u`2uzv+C=p2s=pt?m<RBK
z+rNV-99>WdBC>4wVxrs1r<6vtH3*vqm3Mo8YZ|CL3@pmQkOEFv*fMnQx9gXmt-;a(
zk?B8pdSV!-(xK71qn3=pgA5Ex1Je1rxBdXlgd|4clK5v!x>2xfT}!!6N4A3}I#C?C
zjf0c5W+i{Q{bI^*V2otTh~?P1PyV3c{*JY2J=xUO=pV-pG81l6xmw!q9f{~F<PCp{
zN6RscJoCrulBAIU%IG+!f!zh>YK7iUulqPYiy^Z+`U?QYwSc;HqZg*pr>`+2V;Jto
zQh)Up7n}fb-Lt0oV}%h>9y25_b)Nbe>*i|vAxb_*LAGxpvF?reI8&3LiyRjT`@ZY<
z%3QO-?gJaEV;)}>C;8vQ8q)>Mds+JnvJ-TSTPS7{c21A?L6!UcN`H%s21xp!0SVWG
z!ycSz{!Z>YTlb0byFX+@+W0UpE1SLWFIRb1ngIwG)4`JIwkp8NcSuLIQnfBc%}`gw
zE2$OZF1dTlGsR^se2^DE`y-QboNnQ+Dgdo>`d#y8zDM2VyXMju$VaM2Z9Ki|y#L@$
zC-nXnXBgjO4@Yi9X?KIP5=cYg{Rmb8Zv4rsG~}L-rOZAk{>vh}{S_(vD1<=<k+2YJ
zlg35utXoxL;{_d*hcFry&VCMgn~tsug}>mz^|jt4$~3~@$PfmVf=yQH+?9DzTECzX
zaoZVm23q=dZGDeG4RWG00`A9KJ2m+b)-K)yToy`dp0gNLLVJvqC-Uq)G~F3<Ol)U9
zWdw!tdw&R;3nRMQi58j<B1s2pt1eH8Nu-K8>x<4<s_NvWOY1L9P+W78D}xpqcDGio
zj}#h<ZW)U@MiP}uc<F-8!&lPBw&_%CcJTMzDeHTqttgo@oT}B3mwMjpVPa?5l8*r~
zJpI&&@ML7I4*nt%Y&G4>-c}6CUv=|p2znTaw=G@r84U-ru*xOru`qZ-DEt>tS77m`
zcY!`i*sZboUCre3%R!Ae<wBd3i#~Y?Z@ChjM%SvSq~=s^=l@x=XJyO?k*=0<m}$=l
z(u;&&T?M`5EP$|_>KP=&y~d+@_SSNA%S2=n#?DY}6923I;HmB;#AdH%XLgKQ!j?tr
zI$A3Sr5?F`d$S)`*<mxqS!1Bet=2k!8he*yO|0Z`2^F5E;L9^G0GEF+RLY#R9~p)$
z6S1|e8em|iw5peZWUFBFLlVq>Eq<&ffX;r{A@9y1>L!5n?B2OrtALv?#(1wZ+0!Dm
zhqY2dDK`N$X9?B~DB<dBcc$an?>(+xJ$gM_Wcm)>ME=(>r7P9Mp_Gjp37cOK{8<k=
zv)q#jbFf?pKRw<r{E)SJU(zR?_hVjmaW=fuQP=x;P_CuH_^QljYEGd5s!t`){zaZk
zLbnJELX+Z^KUViPp$2hWMz!nJE8gZWkQX3CTR@vM0j|^t`ou<}X;K(%=sFsh<kcoI
zl;kazEcC)XxPYQhblVOu{3W}{KjH~em;v!^T{^}QY)p4NUvXa*+29uF?h0E4FIr$T
z#h8J<5;{qh097opP1V#XLS7t8Zd?Pq;8#-&`Uqo8^M;dx)c4TM9HBtgS0&eh3VBUS
z32QbL?OoasFZAK;MVgs1;F+mAv|_Qx&|4NTpH?ZwZx3iqS$t}1e)L&MqvI>nZDyhj
zpNy#eXk-c+1916R8D%o|<@<rkcqlys{#0mhX{bZ*$`#W45La;!FUFZJK%Ku@sps*m
zas@wlI}&iZgUpOt?_tl+j6CQJx(B__Z2uH^Yo>!~F+Fdszhn)GdmOaZ|B=J%S5DO}
zX>BEl&6HyXI_uRSG$6)Xq`&svH~+K72BlN@UxO>t6IcELOGIlXLh>J)39n*<s05RI
zdwU8epu?<yYip|7`!jOVGPwj$V%mr__XO^5!G~N}B}mXG>b<ciLCKx?*vN@&Wl#9u
zv1O+G7vTPq04sel!#lD&0}9T-ku(HPRZ-AM$=91`fh3<&gG3t6Uk6CI+a%#o5cEmB
zfA@rFqXn$`V{?Z#{%MgwFo~_*(l;%2BF;Za>e=uwO}qu=?<*k@rsVe1-URiv3n|Ia
zERK)&f3#cu1+tQnAAxFXGnX!X56%pViD0;p8zw|qC`(*(c!@Nv=xCKVSqa#T0U^mU
zM!^?22F`(fAdDZCMtuT>Frb@FT^KgR7(|AofBp%2f#u+TmF=b48W<New)!|6?3{ko
z&8`99gHBj+xP>u=@Rfp7ww0=uqF5WjaWtb>D?~Kaf7h#+(gWYF+NeJ^Xv91cO3g6|
z$wD9^yKZkdQw=gCen9a)e}K1{;Up;DI(ZGEnH3Q)_;+3aLl8#ljU|5W2|CC}d)oGS
zVkW6MR4{(b`GO$w-i|f3;(||<`H_|i_Sfd`(l&<YNa_Vc*xEqhA|X7YDUibJ`S%Tv
zl&w_F9p8eje)&F6G_WO$9{=`=UvRkxuh5WS9w}}B>C|nZhHg|oN~9CMHah`c1!sTB
z)&6FAsh;NR&C<920<YwMzj@Z0x1fcoq-5i}m>1@38m&nQ)AphBWw&}TJ!E}n1H0qf
zio=>Y0M~bcsSkWliYJ(nR~bCT5Z$-Y3oRdUz)q~(^R28oAr^1z1f&TOWri}QjJ9O8
z3Y=D%f}fGojb|-yR$%I)TEBpQQf=he9{)ro+V~1z_xKs_k(=HGE%lMnvg{J9uyUtu
zN7~0~fgxpHa)j^3SFe`Ydx8TQYxkmbUv<9aFl-lJ>q~&p{RZ(Iti@vtQ=|4$vgAyn
zHgEUukk^76SJ&;*3XfZ5Uj9S#%l%ydwznqkV={swUA2U#-XG48*}+%I%@$6)HK%!1
z*2WE2#rLqgEOy`WvMi5y%;SVxJ&6pD6;98BW{uBhsqN5IhD-HM0(KTIJ>i2TF3t%n
zV9WMGkN{S*%PacrHg1<%yf_IwMka@*>PBi0Wjw;#3bMq7>~4Fcpou2N`xUnm!g|Ju
z55YY^$2;9%B9A&`>l^4lC?;thPG@8K0E93_p9_o6H)k4wb&}z7<G~y&`Dw)2Yo4na
zL>NA>QMa~LqVV4aU921}JCO%*9(Mj5%<CL1kR0YkJ~40s{+b#tL-2)zy%+gKJO`f>
zE{h7n%U?&8BxMYvI0w95amUYyDiO?uCPaNR5D78kN{x$zUe1h~N6`S8lTz${q6fz)
z+`Xu`-o)&yNs4L!T8s+cb$j%D_JSp6h>2-`0D71=k7A$Zx`Ze(&b<k9xd|!J)T-Tt
zxDrTZ5fB0+vn&@%Z)y!~LI%0c3(e>VIT|z}rO#gL#$&lhGYyL(I!ywyO;~c95CQ+h
zbj-&A-@4^(AxBh`lm}Wo7D#IvWvSP$@>wK~bTkJ96H#(P0GWNXI{55-dY>B?(cp`K
z8P)mmHmQhn2Te_gn7|Ndpgo>~aB*p*#PXIzrs0y~jq!ojFvZH<P#561m$6?yq#>C<
zvHj7O7G_Qk)A#Td)eLJw>Y9Q@6cs+A_xOrnrBiYf`SlsFgr)@?!&g8i!`2VKn)9KT
z{6$l%lqoRnA;{RLmz$6l$r-2f_fmiL6%#fndh6TN+;<+9eI7>nW4kC~$%It*uY*xi
z0F_=H4Rgkr1?eU0XzAEUV(n#Pgy&Gh_XuZd4I+ZBP2^(!N8cCJZjB2?#XLI_-B2Lf
zmF^2C@H<u;?byy%m`oLU^WCYx#W*+63h8T=S~>?ys){HY=yYJzmSF?BG$G;pjZ<$m
zfMH+}$a<_gRomaBVuTI<80YbdFo)F~jJDO=SIq=BwF*yN8l|E(zWP7qEazd*&r|<Q
z^!?S)5>{^$KvVe~Y$Cs=b_6F@4IBrerD^awxtr+KWGcA;s!M_BNJ2$%(Ue23+S}j#
zLnF7ye5`<d8x=h7<-~(vu=1{uv($?zKh@a*a3<S?N~%Rc7yKv|twBu;ciDT@8Fi-}
z_z<;##Q{|2?}-ar!_L~OK}f`>-d-0AU2x}<`EbehEolR$Qhk4=lchItZ_q@p<nXun
z{^%uYpB1}v?dzQn!o^q5&)-RVSy{+;=gn4Riivw}(OqEGmKgQ?=sY4w*e>PFTU|cY
zJfd~nj5rd13*1!!RDc>)W059zp{;}oMiiA>M9Dy^u1mbGa9o^toV7&i%C=8#Ua}v_
z+>ONV21>q}sacrQ_Pydu>cF|+W~#{lRo-{UQ{Df6A5tAD95SNJLPAC;JG;;jvS&hw
zB71j|tm?{4NM)1}W$zKPlN@_zk7Ms~zdn7h-*x?5<6D2;kNdv<I@iJJbKdhcp2LoB
zk{wr;^6-7$$aBNNN?X|B`!_I;ADfQwT1hU5IH00##1-EsUQ?yI70Z^p$XBoZTSmFt
zc|Dc(QA?%X3f)he)w!=9-?*j~r8V5gkTR~-#rrU!#AHk~<>RyhnZ)Bc;X@1UfK-i1
z_3f(XclW8UO~27oA@a5GhA+~BY$HqwUE3)C)Tghpao)G-P>qN1`bK=3@j443>m#IH
zN1Un8`V$aQDtW)-hSj_03SqNALD{!JH_ReRWNF35Vko0#*4yhhKcuDYB-CeRd16Nc
zKhWcMz1{E3dyN5T8ru;Zii+Gc8KP=F7Uf89oq099;qxMckX^_3k5?WPpY%8PK6Dcs
z6hd$G=|s5h*K~$&PgRURsg4Zuq|a5!^Aghk=IlWNnQtsfANlCv6GAEt$-y>f7K$?;
zS;x4!$vjrU6YSYXM3Su0)n7MSv5$v7Wkm0EL^xd&G|V-bzMN05VPh_jEA&b~5C$Nv
zLK~uMHdv{*q=m7EjAfsGi}g6~u6<$T9yg*t-CJ#LDP-kpc}oGMnIKLchy;H;%1SBZ
zWh|HC4jx|^{O&G6{1vnd&k_cOz?;X7-I&grC^I?8C|2aYI@TZ>yrL5`unytSb@r(<
zr$q^jbRCI}nxQ9k-ARW6#X{%|BAc7heoeIOi<VV1kg;m&+g^!?i?+J%_uKq=dt)P8
z_dp{X#YLOmx(Ml)1%PVFT@@0qU>__y?g6k5#?h{mp3#oX`>HfbQS{YkWiq`{lo^61
zH)EVfY9zH+dc=rCO5pZpi=Zj(6eNk9kiPMWaU?I*LBi7ubvR0~DBY2ox_uOdmCD@9
zB2N^CKCB2=Y76LJCi%!6=`xV(OOP<Nl60WrV=##e%|zT*5UFK{Q$th)>O##AxkFs1
z54#7gNyNXp(<s=vCxh6>QPw<VKl#Y>{pawM?G)Ja-Qgp?A)|ilR#5!C`QuJKp!1B#
zYxCuv4S!U~oq+z}WtsM_5>HRYw)SX^YwQl;OH_C>cex{)l+8I~8--DMevEIJbh$t5
zm`)QBD#~hDaCfDowcAXfKZbzt!~97B8djV@(-*<HbN7sTec3JLN9=1q>y~TE-KY^s
z_B&x(mJqda-9U!PC7#<UEDJbQ8)f<DPS1~aR4AqGH^2XGd--KyyCimI(Neaq-lXt;
ziA8-u{QK~6oy8R^&MtOU{pb&}UUi%GeViwmW#jSjS>l#LBl~7lEMzy2h0Ws<(U~$3
zema<s>;FVWjrIaj`&jh`%`0ed4LsOuMAKW<E<2>%c}7n6TciA)WOc)q9%-`mD&DA8
zG;{l^^qWr!#|GmzNA#}tTA>Cp*!=$U_vZ)8f=wFd7*s#H^q(X0rdS!A3qEv3*3pUT
z`1@Cfn%%SvFKT_>bCryc9qZ>rWmQ>68ui>(!K-Y<n1ZbgJ$D@^pt9X#s>LF5gYq{O
z+rrv`0U`%0@|Jz+6?@c02gy1Q;{zApBNe4fx?!h1FYok%=WM=xC&MB^;Y`NKSPJVW
zR1fZuxd%0cj7=gC_w|WbC+ho>rsu~hM}X@j`N1d{%|krUekL*2jMDuj)g6j8p@}Fh
z0z)Ai?6JkHORDVBA&U$V1LG7|_w9@|p(WDY(H#C1EF_&+eqp6dN}n+K!GJ}sddxa+
z;}OTe(^GiB0CoNrU1q{z9T1+&CG&~!`=g?<#@j60rMObB!t{+y$zd;IpNZo>^2$%M
z450I+wlig>MIHSkAk1cn?)1<!Hq*Hfi2u&)zhNOGE7cKV<0{Lxm=)q*<(wZTafEeo
zz)Dgc70|{xrMT?oa#PkV`xFXpf~%wdxe2rgn6f3}&lqG1JWMoO<1%r~P;ttr=uIy!
zho5H{`KcBtMRyWAeIUJH*>MwJ)AVt?m*3>faPzpXa+K+)%LoVd_N*7X-4V;ZPW)}g
zB?U^~YqV)DJAdO&y*GC&%HzhB5CwIiXX|QnIs2_Iikll)G3x^BcEk>cmwW3k@pOI~
zdt$f>cVn+t{ZPu>hwRWiF|a9IUQyzFEFouj1IPuoH?H7BcOmxwjY&XDAXGJ%E$=5q
z5mk>()wp#D7%%B-*Yw^~NRVd`5|c?_!(`Q(Z=w6s&P@aYLiZHWY^S`Yk*tuPqr3*N
z#l(hK5)^WM6LVA0cW;Eo3l}BA-#-)iJmVm((uA;UO2?F7FEr;^s|);OGeGs4mK9Cz
zapbXijSMNXC2_#pZ!e@kZKd3;4+8CN{_-?nawcY__cbWJOa{bzG<m5E6~jKHk12&-
zVV3d50w^WvS&E3u2ymHaZL45XMCtWE1fLy5lTCMA^)>k{qkaHo>RA2c((7ADNmD8S
zA%$Q<Zd3lHRQ^x@I~Gagdx(By;O({XS!qE~^<7ktJwk}BEC9}Kw+tdOG}=p%({)?4
z#N-d+OjM)T*XaS!JElbh{_{5wxI~!r2u}RXY4`#Uh~W0%p=nJBK*xaJD_DtP5hOfP
zc=opa!g>lbL1z;SMaA-tn^D}Z7pr$zz1-b^8!gT?0SFMhPs8DT_tu8P$)H?f6BRmj
zyX*5Jd?_z<AP&)JNR7H4`YYdy+<Qj>MbHlb9bIJrs7XEmkX2oU<K1d-O*o96J_j)D
z?UMcWKyK%)CH@az=hm=c_a(trHN;hb^!|P_QXnWK>$RM3Vx0w9VQietLtls)%SaSH
z-{^ee3Aix>O;8ti)sItW`LPJR^^wA|eyc8L%1`q1{}8l7SbuPAR=>FMJX7YsfHV>2
z6PNUWeRG(SDa8R-Hv(92n)T&rdgPizXcmEkU&ohvFr_9#&n`<GA=evi0X=^5--Jsy
z1>Ju~9eHaYZaJJMd)R*x*8Bsx;s?)0+z;X+@gzRA!#`iy|Ja%_4}}j%{q5^7`MUhW
zQ+&CPbbjvB$;e^;X~F*Ezq#wp|KEmyc8#dleMQgHqQqlMe!Ff-{;w0qaURws#q46z
zPnKmM7a?S$Hi&t;<@=~VSu?|BfahccM$dNB$IZ&q9+)tYQR4iF4|U#Jh}}78`-l%$
zD_MQi4!gaz(N?_j{k{T&^KqnFqyOn-LIg`n?x4STUM_2q)h-%Ka9_jfCh!_45W)r`
zO#cv&rc4O)&FDi78wr{u5e!=roENqc+Qb0VdPtCJ7A<s-#2#N@i#)Cj#Tz3iuvo4a
z^8BTf6Bvmc6#`v)S-(3l{mVicGbI8g;NjUH_7b02nfHkdhg_3a<OW_y$q%;>Fty;~
z0!@(t(8W5&j2~BtEZ*7PK(b6S01HLaY``IihF^9E63HGqjUiwVks_3nX%TB7;27YF
zN6GYo7cmQpd8^}~P$bC(&^R8Zy1urW73k}50TP#zhfp@q0lby&{iZT*27JV|SwOEl
z)R@5qK!R|T!WTLy#i|$NEC45Z;M%2g@e6WC1Dw*Lv6}eASdRmc>Y0c>Zk^ehtstwA
zqtQFe;5x-egF-;Q-a>D5zZbzV`vBm*xS1wKrtJa*Dh7HX6Yl%sPL`jixEFUD6yf!y
z4krCg=0giO1vf3mG=I_sapCX=vJ_uEIGzh)Fuk8Z1NlBiRa>lQ;L#KuyktOW400`|
zXFfe47dc2k?yW(Ant|Mi;QbwhIZ~nBpEU=4@B?B>(nQ(79H8#q0m&RJc)`?E8u(R#
z=(h-sQzP(79<1cJQjmGQFWJry`22YXz<?IQ0T*W^!9KYHPD8D_NyeR31qAV40rf}v
zwET&8Uvg~vI<q#*pjS#udhR<pP5Yd(#=(b9WX(ONJ=c_IQq!Qu&-1))w0TUlydJ4_
zW@X>9H)wYL<br6M<pO`992$1rU5wKTk+TQ<gd0r5wAid*V&oog<RLb`Mp4$;#Lq5?
zSRh&>gww$&TWH>Hi*3z-<Q7_Tqoav#B#z`SXPgNe&=jld9j>STci$wJBLqZV=xs1Z
z&U~sdj{t?4lx4`1)Gnhg2O<*n*TAm9@FCpY(h`MVw-7BoAQB{b;zp?n33<i(nT<DK
z8CQVRbU6qWO-h$;7MT{?z|KM%<Ht-1OQCXgu7CWAtl}W})3jD#5cVtBYOUT6V4`m&
zdmn%Z=AVTYP;?7ycddd@k9D}7A41QN{fGRAC2~UsMrnLSLIlE6n6$#suF09TuD)$*
zq`Hv~4A^|BJ3uUTX!0CqE8Jkt|8DR)6U424h?N&vQ*?@+y8ZpgC+xI~pw%Kz^j~1J
zv_%Lj&?PJJ)_+gXd7mQ;VlzRjQ-o3%mC^m6qg`F;=H$n!Jf)#y-5Spi=H0I!4(oyC
ztau>TK#YPcT`OU!RWy0wA-Xdhz_dDM1T~O_Ssd6*OEDd_{g?@b{XDZ;H`*b5$&Lh5
zzVeG2%J<H!+GiEo4P;%>rq{vX@}``PQxJJy2;=3Ti+`D3mM^4;)?i$`>wlkKxaBb*
z#k&eehA+aw7B9y5hTyT}CsOdg^Hr$G!bKoVp~i7e5|A^~_qNVD$gipwc5MBRo5;Y&
z@#D-_=FYCb@j@hP?UWN`=tEq7_?j|ieFcw?Y57{TcA4lZ_Fd)15n{4i=<iOFZK^$I
z*}hLuC~;wcuQ#+ZcolE3vmVY!*?SM<zR--Gn;C5b5G@_T#yr?74Toz~Q{-v`_jV!x
z83wH9_=THgDfde6S%{?21?!g@lufM*n0|;P+l#x08AZ->P@u{qW^+yk!J*VENk=>f
z?3$gk8^FEmt7sV<2)4~y+*{r@Qq&gj_AyrMBb@sMG^IbNmjENVYvT`;uaMEFMZ%ul
zil~U0`;@&q3?L?K3qp({UH07WcbkCl_oR58eapXaZHk942E2eyluU!!1}Q{iu{l86
zZqQiaV0H(&k1}bzC-%&xfjDCJh9WuUc+3-u`GTvQEewlA$TkTRWDZ$VFn)362ia@m
z)R3S-$)aJRsbL_$#)UwIb+Wj3&r%0A!HsDA;9N?(UODCQfw~Og5JuPgsTY?0@tIp7
z)}|Msp18zOugtz{rVWo{AXho=jkEkouJZDrH-OmDmTAz(T}hKy{Dw4V9ys#J(<~gM
z1@qEE5$I0z$mm_|Ab-Q0Ci?w}jBYpf<p>Z_?Ys~B@X1;t7mucS+yj3GGN3JzvBx;Q
zay4ZYJB63WDt3^2aRb@R2sC&Rk0K7Y?cZ4pmQo=y=bV-SQ)z#v+d8Lbq}srJDbzqF
z1<LqQ4~{vsmx+JIs8TVL<H6lHbn_VYu{(^v?v#87xdh9ywtcf4;{P|K-~If7MFaH*
zSQx`Ut{%8{Fx$Y_@P!4{mNhl2`tU5d5tC0A0Hj|ApwhYO7C#iC$P?h{3j)$|_Y<6u
zJ|n%bJl!QMuCejv*!Xp)g8SmuwUTF9ZXFK<{V$uN2t>S7KyLry1)?{=HHioqC4c>u
zakg`Rz|FU!1WZj(30-K6Nbf-OOTwqnGHyUub&efm*&>NM*^9;`IfRo}U3FN+)Y4*@
zhNZ2-RL~-b1i46=S-uBo7#EZV<AKkaTjB}0OQmqjSVV#j3Fs~RKJGW|8iLjUDR^NN
zm_0gs8(cwVzu>U%5>B%;zIH#NfIfm~_3;ok@S8HF2$)bYe@6(Fh$W&eLXoJF7-A6h
zP6hiDIcSNq^vM%<*@Mf;`dO9|U|3akg)YB|PIvxx-wh$U@N|T{RS2)Qm)>>H%CjJ3
zcowUoi@V<b`-&1d@g)sK9@h;W?dg~_@`(X)k~<W=#lTe%r&(cA9Fo9#$UR!{8=IZ-
zy)Imce@f=VR9J<%MOD@YbY$JJr9-DJLI|4lfjE~e8IXIBnR%@jI{h<|laOrMUUpmK
z@8=_*qt-8SVl<V@^b;;ZZjB#7Og_t}Q_ceeWbbVX_ndb5wzRp><oMpj=}N_^cEmV*
z@QfI4U@tjyJ9673#Co{U#4t-+p18+?k~LS<>A&+tgjHKsEG(Brf}a0|VjQwS)DmHV
z7%>kj<T8JL-7tqzR5UK+yh&+#7J{;JC@qe5za{A=-(hZ-CR+I-L*`7|4ZPLNxSX@|
zh>7=$6<FYtE6j_ptY<m>#sVc;II>t!-#Y|xniX>4PX6`my)rxTHcwS|q=~|p8Y4V;
z+^dsMFb877n!_=v1NV4OKbO7XZ=S|+xpxOHPS-FH_Ov_CNm%G6eb~=WF!r4}Srx|k
z0)=r8Y#Q7K#}H{oH)nFn;Pic4>NuvD1DL=Bwd{k7^b@2?vaXBloiQuMmSZ#__}`Rx
z?hvrNm_tWg&ZKW6PiurgNg^R#$F@4SfBNa{*G$p2BP$(h*$f?p_FTJiHr8mbt*2>!
zi_89Y-&7DsM4WiOk{hOEVf@CtJmm+DowE6xxEVv%;Fma=*Uozb5jsat))dWx>Zr37
zbTpdFqi>#;qwJQY6q2LtlcV%#)vNR1yKK=+<k@k}!-}Ep)uX^y)`6c8+Ed-q<|E0?
z0WuFfLXGE^wb<;zd(dzzn@<bXR*Qe*j#B<M(iEwXZyqyQ*xfP7CgRUV5y&RyKVr)L
zoP*ajZy<yC!0QL@)=uJjym3!U`ovJ{Cl?ZERYKHEZVs}WIYI2ZwuBg}Y3F`jbm}pq
z;%X2DsLHLF)At0)Y@5qjt3kK)+;iF#C4(AjR56=dAQVm<&yaa^D0$fUNIRoY^eu=M
zfR#_36R4icQ<}?BK9cLq<%mq@(oSkKH?GKEiDcq=zQ4+xfg*q*Ia*-A=nbQNc)d*Q
z$HtXMuSOfi7L;w+*Y6cP+E$WGa2HV->EJ;a)Yjtq>aK6Xzh*BUqWmmJ(sUFl%9!`g
zZ}MALtvZ9_)Ys{4E%hT~;k-8vYSWJ8J4Tq5nD^2Z*+hKFnbBUq3j+NLlm_yYY4VgC
z@hw@G7xs+}b$@zr2N+`SqA%|U1>WJr%e#^_+DM|+(3oFnU9uy){tg47l&shYL9)_x
zJ-vSBY(A!0<24p#3xrXB>X@wAFfje;krM@(ovQ;&n-{Q}evROneDMt4aeEW6;Grd1
zb&f<LXe{}>mSQW9bTJRvCZ32;>TE<?zI=|87I^^k%`>Ns4&m$c$J9`MwzntVCI$mE
zhvlO5s%f11R>Wo$G3p~8A7V;@`Z@zS-<upf#OrjiQ0cK+)(gh?MA9hG9MDes#?#h9
zj5YqD<!_V`V1A}&0eYq9EEn3tF0HYzu?)TnY*0-G!F$4S_XI8H7FU-jRC?{Q;z4qh
zc5ZhrL?P9X8Sq5kUez)^Vs6*lH&!&#(^8$`{BlZ_p{g7a)O@8%a<Y5aMZ~*-k1RqD
z8>Wt9EmDfzwN-fei8+YUxB3Ww2Am%oC+a*zu=$Dd@(;W-!=OE)2hUrKtvYcqo!DCJ
zAltwN;s;$E7*bqeuJc&^D_1l|1QNOZl;771{AZZIvkeFnmncQ3NZ&)CD-~`OwRUg~
ztHfS*ey+$n1<^P|rz4U$V7IJKMDiw4-;G!ix(IaJKC$l{%uZ`<dOic-pG1Rn@e*>1
z0fuA*^&jF90$zl)?STiDHF^Dz0dKPruZ@eor*!gfq++wB<>aNh$v!8Ty@*>a85YIF
zJ~(fafY`-{hvp;4g1*w*6b>x$4j4mnc8IQLWYofDp&gg5hj@S3LTkpQ{pQ)5l-A0h
zug|}1qKVg#K~Zi;Ua7)ru@TvSyg_)4%vx!O>p)xAl$NxjfJXfu!wASzqk1}L=>|fG
z^Q?S&d{=IqR_kCmX+W9a;$&4mO~IB^pk@kN1FOn*-2Uo^hHGRhhydb4NSJeHsAB|-
zY|ny!PGMiGUcEbCTyhuFy8A_#wBfgZ+Ype+M`kYZ?Aj;0>ODV_`>`id1%y)ic8J?0
z;^_Ur1Rcdyc#9Uctga%$;6~`J%$iA;B@NWLWK|&9k;#QrK7eOGNyE5tu{>cZq@!Yp
zDl7rts|WO?P?38?vd?pUoW03$bFP}qFOg0>*4=^vMan4Az3^Gz(jgG_#0647&)W`B
z1~SL9o8{^ElBje~7nu@yf&QAZCPYuV=Dr+rmp#|7MSx!CW^@o(4rj=k--AYOZf;{C
zUp~iNgh?<lwML#rG$uidW6ZGkfTgMRB5K-_IZ`hjk$b@P-0bD3Km^~7uaG$FI0l_Y
zmvUL%g}6aL_etCEKGEiPD01de#32+fJ=MvzBt&4SKK<0x2Tq4+fCWDh@0rc5<^OUT
z$`w7O?dBorG=4W1KslGvN5bluP;!r&sd%S*&djdl%bOEYFU`Gj!_2$xqg}h{N3D=G
zLJGj8{)?dO@DO4pQBB%!UaJ#Gpb6wi4JVlIJ;3_7=frsGB8-Yt^2aObGp!_XZP4qW
z<0?#cJqjZWQoPh@0{vHS%aWil`2)iwb0)7Bfq36ibeyHMjwH}O<|HdB^U50|=!G8u
z4Rb^fl;_q(JmeZ0H575J@V#h!NsrO>+YCe%9fZ97y{nY+@MljXj46|r-cLeC&y{j_
z`|iMs9JSVQ2&`YubY|@sxH?i>wk=i@xMh)Xs8jK2Xn%Y@;myR&>_b#`pE<$r9g7?#
z#V0Pl_&2_?S28UgIfs!mk1#W1BPwaFL!UDcVcu>Tl!U&Ba)<Kva_@3CZY{Ry;>3%O
zO`Wdqx3ng^Y5F#CGGw9e9$#JKla5OTfNfFOE0G&1&1iSQoH8PmNl>K|4UcZbe0E(!
z_}D^T3!&Ha^T#5euI%xeK3Y)=zYzUNB!G7gj8oM--<&V98MQk&;VXh+lJY_k5E7xu
z&2viMb=$NTp~?SvYu5{JD#~`CW+q1pn2}C=hs!M41yb>Mh3+<W+vg5z_Q2@0?xVN5
zGCZxTv;64(IVtsN5CGZoU|gx$>3da6P%Pm-QXhG3*xDsow<fhVlyJk1asNDZ`BqGy
zn>49;XmK8);llMj9*ss>ockNz=^nEx^#EhM>v`LtHzr|NQWc)3ZFmok7eR^TjpxZ7
zp_3!rUam_u110?X^Cq<34<V9_k>TrQ4%%Rs+T6-hFr_-o>D~(U9qnj1T-3#sdVO-&
zp%gipGrc5XGoU-ns~@_@zbKe&Sf*RR=DzMsY|e1%eApN69X%+q4eRIKGO4Q@Exi?*
z!+U#8Z>S$?XSzYhM6+wV424M#t-8R;5aL`L%D;cAZK4iT1+!m7<)CgaExine*j!(o
zw3hUpjs2}^w=;cQ54Dqz5?M=h<3}BWnQ^A1fy=J%S6TKuC4&RzyUVea0W{$yf(S`I
z7QfH#Wpoo?#KySH?-Ro$^|FWXZ4ZsTI5JJU=H$o3!Oyn{Wr$H8ushUyZ)W+oZEIsQ
z*e<=$98NXZnEaqhP>R-1g?N0+cV`=rG>FOvPuW`yg7T5`smdr`Y5!{p%5Pb^!%tN#
z<=bN^?S|8DL3)(53>gV$`P2Irr|3~rc!pG1P?=Vw+QSom>O(P5D0B%;mdie>ou;jc
ztAHy=2k{o*-Z(=l--o8GI@+eA>k5X-t@A$GdBz4=VA`dQ<W*aXF)MW>D0%YJ*L{-j
zR@QLZob0mem9i1xMzsg8nJ?4cLZwHWnCpi(m;^L!?L`gQX!SnN-)z_bNQQQFrybYz
zmK16(%`4DMHxsY!TstaGv*i1q1q>ro%6T|_uWf-6XFkL4L-WTN{{Dh?pLd)U!HeVu
zL`PdpC00Uwm|VRRsIjVPRc!+A+9`yuD=H;MJ80L2R;w<)&w+)*K0M(z4zy_Hxoyy(
zxOlF<&9V-syY=2`2tgIR(aqo<exj`nOVyR<nT-B8>`WB|tJJaWxH#54TyQj3DI<Ow
zDuxl@tQx!s6OgF&@n-x{38iz3t#%%THB<+$b+c=yeF-;?>eNj=l?o-qVa*}QJJah6
zg2C03@N8jsTT2h?EC<sGat^<|+6fZhbsC?7QoX@BtN8AFkQHD%233uvmJN5w2yQPs
z6wffPc5PYtV*h=zFm4MI`oWR8Y84Mqp$oaMIPQNcO>tqQ;gn9>NV;yGrzn>Rmk}ai
zmnauN9e*@%(mA_*`|J#+tG^2O34)~Co!}I*QC|(}n$Hnf6R78ihORZn)#}?Zbj?&b
zw1A$IgUe8maP4}4wb$!Pjc-u=(gAiufGVkRnB|$6_hw9MDk%jKEA2uexC>VMkK!_@
zY`1sJD#xJ*rkv+unh&?8>!k+f^hKDGkd4uaFo+-YIW2b}3{z%R2%Z|Jv3X^a^2ntC
zH2klnjkTM(R|-{TEJx`K2~$)fLV)u+#IYBiu`NxAeCZ72`=t4*QegOs#d|aVp{*Xt
zVx7n}zlGO=IK9F?fIikzH)T*ySf8yXw`9An$2H0SaHZC*y6@@w3&D|7xbvjRVchv9
z8}lv!=Om$2S+{BuR2??NH)Ec0`)*dv2OcC8ofkeiQc$dCE6O!FG#B&d;(K?P9HkrD
zIV1PiMfdxj)wp8FK~Rcds;~_#t26{D=$CR%Y~^Mcky<L8Dv=S(85-sygqYEm6a&&Q
zAvsAZn81Kp6>Paf&%HK^yZ*tLq&$dSTfW9C)vY;4^=CpiA;eY>Tk)z0uBmaT{9T5y
z?2ZVu4la)Iq-EWx`E3zuSz(?~!&b<L?OCTe<lcMttXxt`Goog;2=&+S#~95zoz#TF
zz1Ggb^07`0ttG0xwYeTiqwYi~yViVrh|wVi&p#<Qc3N5s!|OkLt#_ZWHEec^iEnSW
z^k0t}X9?dYMpdn{PB}o)ED;0-vja^az~wq$<z$`7Thp}}hBCy~ut?FI&x004O1*(7
z*Jxw34&gmSr-qmbx5tYwcx_=c5y1o!NtpDq9Pbcg<iqnBV6y%5!~o*lvN|`UgyzMy
zf{v0Z;bTk!o$zhld}s=Pg{nmKUBS|j<g|Ds+8XA5T6eUnQxpP0)!tWmWlvZRI>6El
zb0&TWhNvHj4na(@!Jn;}gOp~873CuYl<Bz}gG&41k9Ci-O3@Cq6&|vEt<+?xD~wYu
z62x66s%NokH=m0l4v)aQ1bnfo;PtVb(}@(a)~;3GKiQ+&dpmEF$K_Z(Bg>PpQwCvX
z!FZO%BTe>D1zwt<n><cvJ<!puVJ|_~rj=`J<bMxS)?)tAtcj;&VYLXSLn|xg_&P}F
z9U~9^y2C*}GCsx(o+12&UjKaTZ<C(SjJd%N{L~5aLVz`g-J{u>INyoRsj^|)cBsUQ
z<8B8EOM7MTu%6w7b2=VfAU}%-kt-hNwyHg;Q+8+4hC=^bnG@u&GLKEnT=<uj>C3?z
zg6fAMdIq=V2a1G%i>^o)h`2${{M5GG(ucxf>$eV$->xdeV_4-E=~S>AEV7z5n8HT~
zESg!k>_9g>v{b$5vYL)1(dw%vD9}Vh@h;==&?A36%(}nSCAIJXuS?kFSFdyzgB*bL
zvwYy5-qYlPBG}MNI+<<;j5>A?<x#voCX8;g=WDxI!lksaL9f1Q_G;&0X^@mcg0!LS
zohJj_R)n?~-{c%ybhwMtNw-(|Mrz~9IcPmObo#L{D0({+n^rBc(z-xE{LM!F-GW?J
zm5XJ}<5eEqWumw9%MGUbCft>U@_t<UulWs#I9X+>cUa8VfS(i(Nb}5;g7qb{`iSOJ
zZgg$_CK$hz9CWMPQYo}B->Y#3@@3w-`t~#q_r2zsQf@lKhm@7aqk&g$n)I}mSb8uv
z9S+5qM1RvE8fE#CTS!eMbQ<_t5<MQk<WkMxdDnO7+u$~2&+d^QHMTTvz=Ejh>JRgG
z*>V<IucV!)t~t5AFYDGR0pBD-Ix6`|ZbBCUv&N}&u=vSB`;>>I)Pe&S=+x!tt69fr
z+*#FVOYrhkWl0i8ChtNiQ4xU-tf^Fvj%4#UHM?=g68{k<q+BE*C{=$ut^Y5fgW=x2
zz7(iH9ofoc^d-4mF3uS_CboB(sSw|pCO4z2D#Ko<YGa=bZn^|mwr(gS@ou;3U=O_3
z)+~~%%t3oq3bHOzQw_2BeS~&WC`!Yv4Q;Fv!5i0>{V^lnGSJnTTiT$3s*`T?+a9kV
zlQ}x0ShP^8IX%pyj{<B(>-Rdcp*cxcaNUSkb<A`yecywt;@>m*!>+FJau&-4ywt@m
zPH_8}U#&^&eLnhnTd!V8%}lq@=}u%9TCI}qQJ&v~cf%x8`OS<ho1#kE=Q(Xbvwd*;
zs^K+!p5JNEWX}kD-D!|aU6!8f=~-4%b?%ObZeqp^UA0b;yx~1?1)W+@w^BJ~mt9Cb
zdA#b|yGk2@cUMTivAm#PG3eDkHsSV3yn=be)|sG`*RRjP?n<t6_bGngq|zfgs%<(8
zea|(8Gd$?Hr86<c)9)+s$2}b9gt<8P$bS*WHGO(SsiVPXRov(=rm!IB#`ZsPHd8w1
zULw@>_-xldWA9x-oj28CKUDvl3(smD-rRZA-pil8+`5l?>@@1^td@I8&`S*~nND<&
z+`Yo2FEr}gXLbjL>G?$P;o*fF6uUn2k7&r789i0JF*DvIxii(_E0CQME$+HZp7teL
z4`?cz07v}_wx@cW{ts@b9gmW5cd47!E4bEqjq-ngTInwwCpNu>A0F5YWwzg4!ctx`
zcU#n^T!pSpi_cyXRVzzTOI(U7_Z_%E;;Y*;Oj`J=k3Ius+n~-Pb)NJ(p%T!76(=Y?
z%Ub>bLQ}2n8kaxT^N$zaMik_zqimYtC;k}Jzr4_REJcJ$REY{J1_OvptBwz6j6eCE
zx%>Syf5hBSbz+{~XaZC|ESC;8>ixM_kk6K)hso*k8(008{o+;XDbo+#;R^xO+!F#3
z&94;_Yo4*MoSXUMu>H@AF<IDmmtF{({5kJ`dBnej^4@7Q<fy6j6R3YVqW|*iFrgw>
zr9d9ntRZ6eH~7ng{I7|7ISJpbz=Ig|zux<=|C;wDh-juhX$joiy~zIkn~b@@*G@Aa
zME(77|M{Jll<-WVcW{A`|Nh*&pD|w_zV>7|3iZp+^#90I(Be@2n`;X#&{J)G#8aib
zdszN?cz^tkhNj`Sq^sl5dej<N*l#zOz7NH&9PuQKON&3^&e)%6=FdKG+lEM}+qNRp
zX>QQSJ;vw|>NZqP2)!yq+<7kj>~H_)@pk0A(I_6he0$A@LMa5LM+9!ZLGA%f#?s1P
zUB<8lRSY4uv}k;Bu0kXQo&!m3EIK}zR;;6b{Yi+nEeRpJP#&rZqQ@B>vP9WxD$ET2
z+qVDP>U<f1Y$m69agjgH)}OqvPxC?<Lg4#!*J1j%5B_r`rD#adD2KW4uJ*sY)OBI|
zHAaiFvj6)Q-L?6>d8lEB57w-V{puJ0l1E7B6)oy0|I1UZ$HjzduY87Y{?!S7X@{H%
zjx&KT{>$L(`cb|noCyK!eXoCcXh#pg(9U!pIro!E`{UxV&w$yZxm>L9%R}2p1w-4N
zCvf+dXU`aBk0-|M@h=Z;^eGtHJUg@RzdU=DFnfApYYfr5#_xZ2=RbUm$aNT6yV1ta
izdU>YcWw;t9fFEM+1aFzHqj{f?}FSV*)(ZGpZ^10FIEWv

literal 0
HcmV?d00001

diff --git a/_static/IT_sample6.png b/_static/IT_sample6.png
new file mode 100644
index 0000000000000000000000000000000000000000..bdb5354639b0a6292417d2ecec9c7cba4ecc5c4e
GIT binary patch
literal 365253
zcmc$_bzD^8);0`+AWDcxHz*)Agn%%VA|ObLbjN^n4-F#SAYDUBhlDf?okL5<&<#W9
z0PlFtdCqgb@BQa_{O6tD-m~wy_r80rbzN)SdqR{Hr12h8JVrx9!;|^&{u3G+&LtWe
zrrIMcRE~AmTMRU`$HnFn5=t@>5_C!qwx;G*AT+cOA@PaWN-CP9UlAUFbpM2>p>ZRr
zbm?fH-hIjXh)qe4^Het2H`7R(MBR=fIQT2WR#EUbOLUXcU?W52pHq01!7RH*ZGrvw
zU+L>^hVSoBGLm^WJvP%l4(eSFyc}uJo`cFX($Lv3qF?pLyub9|U0ER_yF~wr#+Uiv
z16t2IC`@yCxeYD;)s2gr54QN~dESI;nUBxCzGMKPcLN<QASbPJcmtH}Ni{(#T=4=M
z%|u{FKqrlnW)mD|{;XO&4iVwl|AHXGaSYSA^}XC@OiB<18YWL7n+*C_nD}kDj^5*x
zpuK>~h+x3;EOx0GMqb{qsAIyo!4b0G(lo=va@U-S{cs1SK1YZ8R}K)J0sp6-{xd7M
z2V?h+z`I%#(+~L@TyNvkSaE{IOIaJtF2<_9t+C5K#!EG#_hIn<rt$-QR0TF*$}Gp7
zu}k!VHkFV<xU27-SfsEbKC?yW!<m*C752W+=;3&F!>YU_wq%HU3Ju3a{^K{tNVh`e
z3Y9draGdm&z0=;zZL*qqK#WxHE@?2bIe+skwp@hwrt$2t;Bkdup)&gMFUNe;Esb%(
zs@1h$M-N3?#<@q2UF{=lJvr^A>V);8mjM@Mn^OHf8=2S+?pAf41mktS%PP3jbWcn&
zS<!KyhKpXIMKfxA=WTfkPH-)JpY%PY!yrpQu5~Hcb&{7H%nv?C3x`kjtsW+b+7r-U
zPng5Co}nFNS5#D}jJISCqft#^e|?d9o_vm$84S4@(02VKbC7Gh!f7~a`}J8Sx_ssn
zTwn5^55TQBb?AVlClB6|1wS}4Kwt9x937(8LdK45DNfIVmD<LygkID}V~0kCCH<E7
z&BGzTk4qGF=r(>q;#k*hAGvTB{gfYlJAC>u`&~*jjR~&iJMdGY<FB4s1}qOkgPvtQ
z|9~lzwLYR%K*s9Jnn_WBfBjGhCq1Z4d|||^fRw_2Q9Q#Cld0pi1~$L1S({waL+m!?
zL(ex4^)NTv9v({7QI!P{w8<a3T~c^bCbR*T2{Gv&KcXI$E_uq-F4Z25`GP!#jxB~E
z7q?K7B3dmMqdufJntqK4D=5o|jEpUuk_n9{V%d<yP|b+nfRziIOFjt<la~XSlAYjv
zY_l-ts3_CG=f_X|rW?-DnP7NfsAvE%vNgV`K-j;bqBaY4?0&ekVsBmts>7>GualY=
z)Db8n5ct-!%6F9D>f%WT2^(%BU2$0=SRP*SIJ~`#!hHCRrY&+w`U73?qbq-&_P6ci
z?+6UPa>;&Xq9D=@F?=s-|H_`o9#9eQnm(;a^|CNp4z(^UpD4`n3e$dzH+*n^?_Qjl
zX^^>;%_aS|;HOrW&T?=&C_pbj(L}ZKVX{0#I^ny(_lwx!7(^`m6U%d=P|}ya-z8=~
z6z1q;3TJtK&SlkNb9jCgTKt@iIQ6;5bNyF@@!|0rvQOfU<ND(wU+Jn-d}3lvWvPm<
z?`Ml=i&tcYtI{ZE6-^iU7n&4VDG3%hNUqnj>P@_#P>(Oy%50E6`({B)oJH|z<}-~m
zhcn=e;a5grPV(5B>epJ_`&^}A{3Urs8<`|IHSbG5?N_ISM~=vEQMqGz5#6%f;@?&^
zL<)V>&1}(Ye%ma4>(`Bo2f*`{ijm6cwvHH1!Ijq^bQ<JH8BB@##id1}<ycK(xj)uW
z!1vmZzrz~KR%$dnhnMC{$1KS#;p{;5#TOTgsI9G?%iUmS@RrU9`Ph$L_wBsxgdNBz
z_3xY=w=JAA*|VXYW`hE5jc{Bk!JKsNlJZj8Ou$6>C}ND3kBcvuzgOVPYgeleJSG%O
zlnE3y<oJ}^<N;(H0!0Gu26C@WDe@_71rE6G42&$ESym3^&aBlJ2EOWfJzTwb8bjU%
z8@GzlES=$SOcNXzUhvGn!h15^KFTnWJ+{*D(rVk}RrS32j0h2M9F+IPbRX8JX2O=B
zTBh2pTF$p1G%W;gn7N){&(XxGs)6k!oeg%6GIf}i4@zjO&WX=?&Z*4_fJwl@jk1l}
z=g&RFZdA`X&daIUsWI^zsSX4`QA<!w3DyXDJ3D~Czu|j>Cqya8ETk?72aP2y!|-4O
zoEoIIiK?9WI->PL^&53G&bI@D{TvJXnR6#Ic~uXqE`|;q1?)B(rs@WS$!J1o8bwSX
z*ASgyt>JvVKt0cLFFg!)M>UGwr@QgB5w)E+{ukIaY#tA9(g;>o5h`or>%}{i``XJl
zox`nwuJT3OY51s5@_jv;mAF!C!q-MBYWygb9&&sqd>z7Rn)mpjxJ8fLDXa+DacuB>
zB_2rB1PVXq!zT{l5Ax+wn4bLEvz)ga*6x$;mr5j>?=V#LXfAl2kUHjfZ$Jz-BV5`|
z+B(#o%$kHF!X`{PG?`R|^n`wrx#oo$i2$38<oqw!QddLIj}JuYnGE&Ydv)OTa~tQw
zj%n&|5JR5hS06X@H8X$F8Mkc*$Udom67|xE$e4(L$$L4_-m;!iCM~xp*N`T)HvQr|
zq6@H2@Ob{=%Wu9W7i-_wHeVgQ(pJGz5m6cV6y$GZ7;xyUC9dT>8+7{f^mf~ryn;L;
z*&o&ki)Bz%ZYU7>C}1h0Ry+Mkhb4i-LATm^^yiBYl>o|qNXEi<2?@#^g~H@hY;)T(
zX&cZ%PZxJs6o;}VgVtF6Z=>z&9~f@lZeoBA_Eio`?J}t6;PE1H2{9jAR=f<`oEmu<
ztHuO$*5}NZFPCrI(tDt`576gR%MZ4;^i^ov`PM><^eF&*W{-QmsJmHHb}bEyYNu_S
zoGXV{+-Va+w=62;?KO`Q@tW||eJXDYk@*!7eUy#-g4VOY$ZIvLuZBp5Ik06v%O1;K
z*92Nwvx4GC;#=lw?B?Qmf+~0RbN9#gE033s6R4$t{D{j>F<SPGV~v#+vW*sA6G7?7
zbdkm;FyH6h&!p-;iyZS_<=xP{!pYnkD*_wE1BjW2@i`a?L)P15Ot!3e>0gAM58%J2
z`cm##RtzlgvQrV!0hH1TzebEibchCNQ<P9xP;5tZNw*Si$B;r_+&OH185ylHh8g4a
zL`12lXF^ykzNIAa;|b*R4`{7uHz;C3tSp@^MUINk%5N06Gm3;u!S3VaW69rmjm=Jm
zBGZU6D1|Z{TFR{r=Q^I)liK_mHb*p&IUOCpgnlFy(x5g2H0p~wcb)b36~l@Lr_##!
zoQ1%U%fX{sB3%Vt`Kk2s+IhSB(OHqk8@VX6;j2_p0l>n_d;=JG3cg6+0$v;pHL^AK
zKrfKq-M3Q&4z%H-I=7E*3brbaS1Xxk6)k)O?wgf2UH}(%$LFqIA8&Vv>cI!ZD(^4H
z^bf<2ORU6gGY{$)PE`E7{RH%2UZeM)kPc1-hl=rH#d^ujLU)N%`sp3^9k=B#%X=2B
zT#K*3_rM?*YsO_p#YM9g+FPYl<&#xkMVKOWhOrnyIOpBB`%iZhNXy;YOxvh#o}TVb
z9UPZzVsWK_*jRMeDq1vIVYJ_rU+%eFDl;HYzZ}yVd~5Yt>4SK@OGEo2KIp35DT>`n
z_*q^pr(T)bHYB?V&B_)qRk-Mqtw>8sb=^}rXO?zg8^P%7(`!O-ue)$xZ6I8XahZQh
zdaI80wPEssI*Pr9f;41I<>k>>QRzo$nCKK}4^b&})G3Ni`R}w8Itv<#<f8H)prHks
zqhbE*83ok!&nFsn{wed%E5`RgG;GwbXQ<OP^TB^TjdPia@n307HPk({w<;1cGN`MH
zu>%NX<M_qa>9a<fH!1_i?t`Wy8XC#VKPS4(rx%B)`lro5YdC4h%Ly6VT5}qj*cyR2
zU9Ii@)PpAMDuhZ}gPaWMT&=Ba9EDs(82<5u5GwsA8^A#Kk4Ky=MHn>XmFOgF9YA#a
zoLrn-45E+e=;(wUOiYD7y_fn|anvsnhA&P|c0vGvi;D}V3ooawgBgHZP*4!S#RK5s
z;XpmX;pk@LWa!Fa<H-2WO8%>!_aH}O2Xi|ob6Xp_KlK_K**ZIkFfjaS=-;1z_7mi4
z{@*RxIR0x`r~v~0+yQWNasmEbH>#-cpIjj&b61d+=6iE%)SRK(5as6P<`Vu#f&c5)
ze>eFbMK%7rC=VYW&wm#Ek6Zu$qN<J{2MJqiRHII!|Gi-UD*T@}|5Z>J@MrA*F&F=w
z=zrv*W?J;IFyP;-Ci*ziNlgN^kW}XH6+fe{C^7r<dBBGHV)^G4m3}Z{xMJTIhlVDO
zCiDL7XIJ#SbnGUoUlX+;u8BH>Rs$6)A%D|%%-o01=uT<gAr7C(zVpF&gZ}O-<~aRR
z*@7(0N9iLTIL%W^R;>m?SaRIfq9QBfQb3@#sgy@^%LU-R%nTeV<^DS|GBQ^$#92xK
z^S$_A9ho_gPQUx=OaBOZ!1#cH?r)CBueiuK!oO`F1E;sO9D_*m?K5=3r+;&hU%ny;
z{`%-|4sq;f(Z2fFOrhUi{LR2o)5UYBLi1O>CVc9P|AGz!iZA#4q0C>coH(Y*QoitC
ztr)7~)E~sLys48|9)JI<nfv;vw^h;Y??g@<Qz#lGGD2L+&tLrAk`qSzW*q<Bl7D^V
zM2Aj(>1>eA^5O5qoq-NRZ2fNp_0L$h#8D!%+xlKP`Y-hbwQBtsf7$U7U(9TZORtLs
z7bcnS_~ChfE%o~^M;cua1+>8yOBW*&1j!*p%u;xn!bE>JE4~WMToa#Fl8Mzc=q<Q%
zC@%5Hc{;->1vo9x)|3BII-P`kHBFPdnl2VSl>J^|KnvddWmV#eFGh~Aycwm!xe(BM
zZ$&xEF3rDE*Gz08vNv{_ZcDz{+0TL-zQN5hyi8l}?`{gv<IMhcLANpUl8hdY*V@7h
z+QKcCNcl{T<^HO7;+XDSD783S>Qw(N`B!Q!l=w%d1+1U*{FP4o(#WAiTjb3VvGd=R
zA9{(hrbF_S0e?3EsO3RH0fB<2r0!o$V5Z|A9CI9(qW?QB*lI(W^C>?fePz<W(xNS$
zKZa$#TDkY{49oxf<<Bg|@`n4|H!lOCb{865AU$VRL4-+MhToP6rpoojcyCX~T{Ipw
zO51&|X^<AG0h07H?=!LA|L>$Vvl(;NVoAUhM%JNEBj9;y$7|5~Rk(a>2A}?LIfyoe
zt-NV$-UHD*^wV+w*A)M5IkY9&Fpym6h5rIG8LRdJGshe`(RwU+EFS*9)7cU42iAOt
zz^k>`!2;zR7^%Leq3GSEaqJ5j)9}1B2U%XbC2`*U>hY9V1))Bx>QSlWVdwTPMX}r8
z5B(;#-Loy9-^VaYr!YndSAGM9zRKg;OmmE!v1#0AopO2>^tE4&Dso{V@gPaDfnkth
z!cq6g%S+;aC&<FSH58Z4CquxE?feXW0x{%I!PRhH*mQ;QtxIua7-vJljCJh-j3JEO
zh(z0pn|EA}S9JjqN1NeuS09EwH}q@~vO&?6Q7UehYR!LgcY{zjB<D8ruiu1t6qRjD
zyFFmlUMD}tq;1d^-41<i`FYs6>*5UWf2Zt^aH=O-0mzZUe2B&QtY!J(r*&k_oDHwo
z&CD<>6nM1Oh-l$W)N?PKF-^6G6&06RJaSZY7+kquPt<QHjfdV{@3OKFaMY%`9kyW?
zmpNb1mV6Fr%W)}Dno-EFoKR?3=t1;}WH?F6zAoNuV8GqBou2KyyW#ZGgrek3i;gQ`
z+uD7`$Kijv_FRHQTRa%hAR6~xb|~^VxuiRa*7o`Wk>OUZ!YO(J<28-oo8&494&XLY
z<UO=}d54m}%6b$i;7Re+p$0ii?o8vcSCRfbT=!Wd3d@F@j#05yTaNm*@1_mgdBw;`
zuk1?V)GrsOwhqJ0x-qP!za4c6BM&KwVsSQ}@QOC&I--as2@Y(;(v&woxuz@sEa`uW
zs7_42oV<q~=QCyt=Z%_VEms?q<0#ZG(WbVyp0Bevts3O#MNx^2CyIZ%&UZ)l^!9oR
z%8zKdDKngC2q4kwO>-O)G@aBm+YffyD-ut0p42p*(se2IJP07wvtwr^WymP7lEj*o
zZ%$1+kx_pkQ+%eQ?*9_mo!G~`;dr9#Dzk~2XO3rEbC%J`-KXY9<|0i0Q<|PGgF+Me
zvGv;vxBxbdYo;Kg>4a7B`~?QI>F9p1Y8V&G`u6&a_X%fR=JrJKrga1}22@tr-1WPF
zE#Ik}K<XdA!0UZ_8YyT=U2MTAU(mN4de`kKa%9PS_F835)j8kd?wx?gJ1p-7_O~A5
zPAzxWmb~^W?|BDbkJrWj^bn6DmL%Nz^ni#-j!WiyQj6COc-!TS`@f{!_drsdNAf~e
zu}kWYk~Vz%>LjZBRuld&SHkx|F>@6c*>b(ZcG5+vZ}QzNUJSRV<@Q|r<fw->by)0v
zKHbr4);ymN)n06}MbS0JG;Y!n4ZM{=E!h~YR53&JKtv5f`T9%r%`IvQDAk;Vj(Zsg
zQHrfc42f6M>eF$aP>r|5z{2GczDMP%l?5=X$Z-cOlT*zR`W*K$2dX~7*=_a3SU|%F
zMaOJn??J;Ht~}#FOgdUQ%BAfHE;5IEhFzXe`pn2C9tO<yZiI*-j{pO1Et(f~QYq1k
zxi~(!b<qj_0Z#N*=2Jt$-|y67;OSONQ=H&^cI~)~5kl;K^!2Vy<6}A>QGjBRXKa3W
zfc^32Z}Z=Thq<`V2>LOc#Wh54PuwC)DK5K)81=ocek=n92%+SrNo^DBzXx+)IisCn
z9p$ws=7V4jJc=UH)X76(n)aby@>t##A?NJ(5br(mlPy2R6k0&7KY7dPiU`{=e+>#H
z?PWw~q%SLaXm|-kFEcf5>epdo9|0mWJ<^0pPMC{so77Xak0r$Q?Yqg%Z~<p^C(ygI
za)s_i6my=Vak;LOV|ZKBHON0_rO;x16Lf}R_4Jq{Q=_)sqv{9dK<U$HlY_A@EoYP3
z3hD#WqtZh>1MIrjUS3vXPsJ7QMy2l;_zZ^xH?0b^FS7PzliPh%dJZJ$;wDCY@o9wX
zj2VLV`XAtL>61f^j<C8@n--h%FPMdxvY3>`Lno7CS?Z`SG`Bs4IAxe*zP9tXVvAo>
zD6RxCT92YEbvhn~IvH)(odl;S`dRTRZ~DAfoT%gRhqg%GLu1N$N}D?R*u{}7i9ek;
zhIk&!I39J8rb^Seb^c+>3ft`#_H(w-X3D1nDKSxpZo?uMNj}I`TAKhtAp}_zOypwT
zw2N5XslRnK5;%ms8usD0EUV9RTa6Haq3o1l8Op$>{Cpb8Q8S(GkWPHpd+2jDpwIdk
z*)Q+GgTn9BKM*#h@3WxeG%7iy$SNO))VBT=KV#oRJs2Ud_~q$g<@`pftxbZ45l-#d
zejjtj9@fiB$Q%lHa193>$cCFPSHcc}_jec2L2O#jrEm)_uC;Vj$!X*ovlw^PuxR6M
z-FnEq=gFX%p4)nS$#4qo{bkPvDfCiuPz2ZV18@a4*-~<Su~S^$5V`u`3A+J_u9Gq;
z4VD0o@z!j-bh~YwAi@L#3jS&7GIJb?5x%+Tc_M$Zl@ry#<9|#nA=b|Wf*ik|(s4}2
z6PO=ar#*J~5<L-F7j*T)mi!64>t{sW0*XCsWZWb3$wl0&dG%f>#FbB|>n|XO?~zf(
zyP;*FGey4oRthc)&WZI`8>t;zYD0y!D6^=W2uaihC%d6&hYmH@J@TZ{y0pG_uX3QF
zVHWCyn)Ua^Y$v}Rf1ii~u75gpRmAwhJIk3;j%X(YuFscu`6L0nCz;7b{eIA?S3Oap
zm(!=a@TpTskxM?N`S$=urZA!?(FH2EUY3(Nu|3Ok%KFV0VdJ^icDNW~Ys5@nT`}-h
zzUh1;blBor5x?4}z8F6agXZTv`psx}Ix3y&M<6;)W#36yhEPHX>XQ1JVn%C6CpQ~E
z!?_*OzrUWk`xGV2^Hnw@dvUFrokh;uuu@h}Fk1A0K8R{%7Hd%q1$Bdm%N<psS(Yf9
z{_xQriWY2KZDzQvJNCUye4U>T)<V}m4Sj?Z|5}5-!0t|d+PF90?vLnOAhzycx^UgZ
z5hJ=(6kLBkrDy!wb6j5F*Y7Ac29-6OM93i1*SpgW0<n^vj37mxkfbAsf{UJ7et0&9
z2x=CA482<--Qnb`BsJpF5ku6B?sr@z;t21Nx7hU$8HE^HAHPo|V$m`KyNLr&{X}#(
zJCa@pT1Aj-Z;#}LQLw|YTosw)Tk>MZ{muzJ2&^70RgFqfCd2Q}YiTrp8@t3LhILEa
z?i9=vYP&LdZLGzFQ=2UX#q=P&FMwuC&+y~E5g7xpUS1l<7gy(0`Nk}kH6Cd3%|&WT
z``paIVJMb_cz#Vbh)c;2!y7VEIzFGXRgXiGq1n{Tm|__|fhLu%nLWHg7#gf%TXWO7
zBaeF09|T_<Mj-+sH5CBOr{8BliMp_KmqqeLhfQ@ow;yenSU<$nw#_%Oueh$S)&V}-
zxDU;z$0CQ8P&_HMW$*cCvUF+`rBJw%39>5uk>J2vuB5}|8qx9d2nCB^DQ~wG=z4;7
zN^yD1YyjJ>eS7Rd?r}4U!wrQ~@b!~hWPdO=W#9goVq6!pOv2`}k0M-H9g;%Pi$4pg
z-Cj^1@IZkxtpNVUG8Sz|7<bS>4w>Jl8ty~Mm>QJo0#^mXS?dDDLm4wlP)tUl4Lb7f
ziL*@0Tb#7o@GId!hsUd{_qXtS=QC_x2JWpc@go!`=Z9@&;m?@or_X;CD-I{I|0f=h
z>e%^St<v6WaBJaq4AC*02k^vSr#A3Y*-3I=1c@Qrx)#{B3J8oR;@x%xcifJR)ElL+
zycLCy;{;c89j&h}4MAk>=Y#Kojgu6o0f7!)*K)RZuI(Sj%>qSrsgVgS7p0{a1R_gc
zs#Xu?lH5p>8h^Mn>`GK<rU~t^TQ0eT3m*RTf{9|+`uxgP_?SqB<=xBJ%Rr?JW=tJ9
z0)2YFrR2b?jGi3mp60s&enr!c*lF@06zA`5MI3ezxbz$l1H1CraJUNhP$E$7c|7kp
z%;&NHD^YMG$(VJ@eFZ1>+m>zirPai!kJ&dW5O0(*LmvwI5lr|zPXn98j5>pb^|mxo
z+ztzK<S2{L;1g#zVJ+O-!!O_YBPIor_>eT9t#fMt0G%hl)OFwf5uu-gdq9FxCBFj6
zCk)r`Pu+5Fsa@umo$N|W9%;SL3ntm3*&}>{JEDHvAmLOLJ~)Ei!(CIc0G@LerUw+m
z9R@gTp5YkNr!OVV_g<1b;pRH$Ls6=fWHZPN`L%tN3|n=rr1N<CQ|W*#D&K5p)B197
zMy*HXP>}Z}IyRMohJ40DJ@?8J>#t^9uTJ^WAvDz|%xQMtC{>U*@cRR%k~|l!Q)E9)
zV_1{UYnmPP#KE@<2(N8k=GfQ2WWzb_hlNjYNggHp)NmI*_T#JjNTTPuoML7j#84qc
z(-%!I@r(`g1ybS097r3IZvJ^aqcroGCEY4y{R^dkC;!ufWR#c!E9MMd*6?l}Xk&=E
zEU+#SSw2miCf>4WF~nP&7dgmq-zf~)muqm@6MtLTfe-vGJS;G!Wm(!eB=^B8sA{8y
zZ{8j;r=a^X&9JHDZCK;5&*U4<iaE63mu)AL(PhVJVioD-CMY^zAgoBX#~N@@!<Q$s
z7ar0tf;=?rqx1o8#Y}Us7a`llESEO@J_%|W_pC{aT`R9^UQWm3uw9g9s@Dh6f<LbL
z3lla>yS+dez<NH>qQ}tv=ZR;P%QtTQBs5(nQ>vhWvdKh^AMXO#u6m{@<3flOg)8#*
z{&*K5e3)fMq`U`<tJxqJ)RverM3qKu#Z>**#}eg?MtNPy<R4UrX4oWb&WS$I2^9Q4
zOC@sVajuDycCGGC3E`V`a*pClDhw_687k{mA|BLm1}(amu|Cy3T@6F@Fe7~KwKm|T
z1p4VE^=&(SJG)55b=#ss4`zvtq|)P54@%n#J)Y~>eBRrFO&4`NpS_7j-NiQ-q2ogQ
z;ksRn{9z~M*Ory-{R2$-={`D9CO4`)Sxz#)iZNb*yA6z(1M563@Gic8<gCx_)?Z@=
z`gZokR*Oy6m2Z&iK1Jcpo_}&DMGN!fo0gfJ2Ec>*M<cd*Q4`|}Sd>=LB@H`8^;{zv
zx>glk4$VZOa$@AkB=st&O%Xa#UlbN^)(<!%m3n&)-lK{9OxIc9$o^b|w@dO=D;!hE
z8RbSsS)OYVCTAySG{DnZeeO;&Oi&0TkMdloFQiijd54AL;X%}34M#<xEu6+(Wg}AI
zt)!<jGLs<-5v>s}Sx;nu<zb5YQwWGOoo)Ke$s<rt9rLc|Ui+J}^k0qYppNY}b>@H;
zQwLds#wW%`r05pf_}iSv4l}L=`U^Ge{4?(CUZ!Tu)WtDQAp^!b@4oKI>Xx-#V9`+R
z_v<Ls+o(5v+_TUPNhd^<1q|ESa5S>nSF#m;Cf_zpGpp4unBT<o<89=Ly@3cEo)B0y
zpH1>D|J3HYoa2*jY_=Ob@yB7L#+}#$xAIAOgV)PK#mHFlR*5Jsk1}=nN;k+CWhgHV
z?LKsnUiU2d3)pl58i{Qtynq-3<b)0ciOhEc>{W8{`kskWFP*@Wot-v(T4z*y9r)S0
zcOq?D%(%6>BKbxiwr|7YrJH;zBGYY_q=k)t?)F!L%ql*wV;_evJgfa~LN+t2N>>b_
zzgMCe=aian&HxQ0=@Rz8%5HC>NpK+pRQ58&yBbDYniXFUZwe_C2~A^<4Gk4_cJw3)
zEew^(?>!Da;BH)SG*Txmb;GBJxYIjv@AeCy>5v7My|Fb}N(4Fa+ij4?ik!nil{_2n
zYwJMWlz~>nulMYg5t>!U?rc*=WS3VCHtNS8_KLD`oRiak9RSoLS)j@=W7zyorqVhn
z{m&MXcq)$E;{f~AoZp~DM&|z3@&UVD35Mv@uUHHYjN?z_A332s#Y$@pqR$IEHQ}mE
z!)x!MU(^dA!Wgy^3+7nfTVH^EOLmp=(=<=*CXz+yP>eT*n`6g{{Bq)3`t$3QzE>i%
z)=N{mYwMLqO3;QHtMgMfluNQlHDa%QIxDqU4@YhYbeHXH_%tvMz{M_?{JP5)*S)Lo
zD6|{3Br=LNr+3E695%~l(A7H<n!LBcmS4=K8G0z@$dM<J`<6cUH*m^?9*~#|0>vOG
zJD&Z_s>1T#Kw<wAfa=hG=vUKtRkcFoIkOeBJEHl5!m`zff6>*r76dRfn(N~odec#Y
z;_wHff}3eorm1>0xm1D<xflJ`ggwV6=4KCfi#Xh?+S|{2tlYNKOHXyvZf)B;67sFy
z1`QDTG*Y}9B<W-i^Fet~g3!yTQcjH>5vR`E6w5LkK>faY>Dy21Dd{2`HOZBsHT}Od
z+;Ue8b7K`2v!|rFu^P+RwKbQ2lvIjTyI38;NJO%Lry~*(>CNPNJeuCbh1!SJR^&ib
z&};s@$`Ea4=WFE}H7MBHrg;t(N1q4l-?<d`sMT(r-F2<`XIQD4s}r0pHs<J_jZXwv
zN$;iWk{7e;-a2c{5P!N=B&sTdsSK+=7?EU-sM#8P-2Tlkfk)d4S}3@e<X35<UgmYN
z@3N6YU_EEvIH@8p5>L+c6MA(TI}F>*@Up3$zcid{FwaY#)wKm#!b9}sU(@=$xeKyR
zfL=|?8P>)_X0PXL#l;%CPM?G4#t`ywZ{E)>!+n=gqTF!L+Xy#f7^UeTu>c=4<d(Ur
zd-i_t=FYH&5@Mf+erGlqV+|KWwsv>fs?HB2C1Ve+)ogAoGSm7PF?pT`Ec9OjYl%<g
z{ID9V%a67!Z@dl~`FQqi-?Vl#BCPAgkV`rJUdFc&&}8!MK8%rFG~+Pf>NpIrPx6&$
zFs`N<GFzGE9k?y;GqB1?IeEu>UTj70$6TTD25RLwT_ot_*%);@u*x*ubhv$~I5<TF
z1<#3hfKYJ;?py5h;@(H*_ePs+h<kN@#b-3f@1WjZ)^{oGa0}P)3Q5~b^qtHwK+CiS
zD6WDH>vPT5+Suo?7-<^w=$A2Wd!Y}OE%`QkeX6spY*%ZazZ=R@NJF{TmXOVFd=B7Y
zhsf}5uwa%%OEr-OYvPxw!*booH*+}52dUm3&L%3rw;f&daTVi+nL4u}Lj3w(2pdgf
z+uAMy=`#AMRe?<__|Ul+_}6B&z=44MhFiOMl4~RvE*V#*?9i~M;Fx5~{&;NlOOE`E
z#e|EyTV1VYl(QRO=MeEGU?@neshi{sIpKKfYw)EpSv4vbzDf?qJ|=j^58ZEub`vhH
z@dC+vDmrbi<wf-!AnVgDDNW&w`rjijC1#_c&9rKUngY~SNDhYxE}y%r&4b~@axNNh
zwz}#9rFTwaCu6vZt3pPy5c@DmGw6gL4)_GWvQr_5hAakL0oiZjJa&dkJms6ky>SO*
zH8x$L<I=~RpxBG)gxf#f(cJluj8y@6-eTeAh<?e>_Qy71dX4Rts#lq!`Qgf<W4AX>
zkW*(ycorwUeBqM<W0u;^u%FSxbK9YLubkyd8lZB>{@+BPmeey&ZOL(Oe`_+gZ~Wn&
z^4F&8mI(Dx_c^G3f@C<)PRD~M$K0S)M}Ugb!SdThdY>B(vRe<TnaG$*1+f%|)peqN
zOlux%6(ER#mP8+Zc^>t1Y{NbID>&qE`GbBx>rptzYQUK>%}jwDxjghERh0>Zt4x?^
zau*r7-oIfD+#a#Mfsr)FDFV!51zzR!I;?#>R_9gE9l{kPj=3?eF&tb;-1JU%7;*{y
ziQ6HaE}I@^Mc7}kCb9GA)Mg95XTlNtOR^lWx&}aK#k_LmU&uS5+1}d?6Fzppy@CZn
zfmDVeo&z;ek~%Y<T>|DVd-yzMEt#UZ^VjD=CsL8z-$!$#Q|-bQooG3&cqy(_PxUl5
zOBxRv6gVZzW59zZi#+Vc{m-}UHm~FbFB~vI5pUd*{VxT^9M`TC*oGAjoSE#|*@|o0
z3ukzL&$Q4=E+7V2tOYB{K%|;64kSVu8E){}F1?}CYl7>qO&@_{y~Lwe(RcoM(jwA6
zF^p4RfSzPccYV7??RK4qjHgj23=Vna#(%uh>n<zo22><FzZu++1HJ$mPw-WofC;_i
zN3U7I?$7&(wmb7_w+3*myl6vPxV%^!Kb2Q>n=x&WH19_zdNnVf#Z=SADVM9u_W-*Q
zNk_pG<&nCgp?0iV7eR1V%BXBm9M|D~(sflfkEF7fX|M9;J|mYV`Q?oDC~P%KOivO+
zLIij7ompa^0Jz@R1;>vT*XLbOtKagl`)=u`Kcl~6<6s^yvh5M*@^Cr6=Y+z0y3{3n
zL96(<MNDulS^|a&uSj2vmyvzpdLTh=TRY50HseM*{sk1qQNzWO`_2Uw&1i7Fv7GNQ
zc(ja_6&Fbr_-%1L#D^C{jCEo{O}o_8ZY|RfpGTV7Iqp6eXd00I_EorN(*Tb+>vt5L
zD5WC&y0ygTKn4j2eR$l73hPX5#fN_Txh&yqv%xpto=UdYP-!(&`cpT_eA>>bq0Usj
zzzKR~i^4d2ESLBv=bdHMCU?h)Q2)g2&IOO8c*T>r3ih1$W8*f;2V%=u#1hLBWpU;-
z-j~b$w`z*V8?j_=FK{W>k%~UoDz1Ig`@w@KlcwQpy*FK9x;2ue5R7A+JI-UOo171M
z((#*kbD!tX!)4xMV>k%1^?c58SO8{K-vNNF`#}4Jj}bkCmVhWQ#w=N-_36ObpnGbG
zxoOIe*(8>5U?3`%cxHPN^`!i@vA)k71uR0q>w0JjcbRK?cM4k0`)tne;=abY$!BFW
zKXvq6JHwm5P3L|%6&6~|ZWf4_%6mGSKFr@^P<7ucQjCje58@iRS*6c`P3hle7Q9`y
zM<z?p__o*%THcdebKcxcY#d2Lc=SE*ek8m28k*KNyS(m5)SG{`t>?Y0>Nsm}R*(Mq
z`Q6Pd0@m0nOpFk7y_ax2?Olb{P6E{LHK2-uE}z%Euh+(pGLjo@8i$3CTa7ow04+wV
z0^M?OGk4zI=(ig^1Gu%qQ`}}5?!Gf#(DcEdZb52#QC$xR?%mtruqVS!pO0>|Da(KH
zYb=yB9;L)vMkstK?rFJ8s9(PZ!Xl;?EEcIRTI8W2rju<|DfNMaHvGtSchf>p;!ml>
zUx8n^PvQv}u>$WKu1Oq+_b*a4O;fvgbf(y=dmr3^JB+O`#IQV~fqlj<^Y%eU-F;4A
zHGMB<fwwPS_YOU#I-ZPG6g5{$GUjThBTBzD><A>sA}Wm9WBsMzOV%Q0@tftO{RyD1
zNtV*Li9!e|vDliW)Y?0uNQ09RX0L#Kh6>v_SRaflidpZmmM9SVjm66FdCrx+H^3sG
zNxKq^<L~Wq;(VokSY1rp*yh%_{ykMg0@$;Zs!KT9S6H3MBK6L!Hr8d^4i&CRu3dB=
zsCa)RK^A?)-JUBV!)AesD?(x+c0po>(lenjGj+H#58u`L6!h*_7@O>?HLPRVt-bzL
zrnKMhuC42q18v?ZUFx_h(0&V%fzV$ZX1uYj>|;7m?&rCso>kou@IHZ64sdF}6g%R&
zqhmH`z5Wxv^d?+xts5&lx2~DBVDk$fH1L{D_d1`kd4><CE6b%6*7G`3O%zc2%p+(S
zC30@({WK6<|0hyAkMjICe6eY`!2FfFmV^RW%iHjMRP~&9ni$p#FKN!EH&n@bdgml}
zfwb7yg}Ka;$Pcs`Ar8O0To&E(v-V=4%6H!pg<{*AL6M%Vm_*0SkwQTY{y5;Xa5c9B
zR1}amP9BmUuA&|ki{{o1JSN8SJWEp#CM0cfTMi&?xXy4@>i!cBo-(T_##Eb7H-zNk
zwqrn92EdY@>exyI&!Ef@*LE)&scHgTv|k<ZGUFP~J~OWIh0OilK+g<eN$t>i@uVd`
zy~jDvb{uxJPw`KED|M3H=Wf*hAfK$p7htmZ$AyTJlj8wE<6BfHA+Szp&L)26`tD}G
zHX|7|3a>*pX|boiG=b-vs9=qYHP1dC54tmk!&@LWuK5j(`xd==i@9R*>yKX5sBl*O
zc~D3Qt_1^vE|IRB%9NSGi2bQbHya~m{Hq+?e&q{;M?rqKs37;jX?nbWCG&6fEs4VM
znER2SPtYzf=%hyAP=Y{$^6NZmBX=%3zc+|%{ZDM$Tmw#0$~&+Tl4Pw>QvFJMD6H~y
zIbZZgX>O!=VhFpuen5>&{d$69<;9=9ia;!HNzmi=r#0K4HPD;5eGgZrh4l}*pMJNN
zTeZAn%4w-bZG_@AEZMJi*Vm*r@xjg!h1lC3d=@@qs5lmc{rGEm*vDG({SN~An@a-r
z^thT0{!%=)nh6yqp*)`bJA(b5z*`%a0$$_ZB6ChRA4Xd`m!Gar!d@G&zZDwCmay&6
zU?g%KRm=){LcLDz^?1Jrc`}TEbkVM<nyB`RIRMwwT3BxP-OGZ9u$yrm9<9fSbt}XG
z_75Cw*aME9w~Oog4Vm^(ySxU+R~6y>;6^@L{rq`I7$wNmyld}Wc>|?2#v@;6-VKaR
z<z{eP1EqAwBb(CI^;Pm(LX&AEPu?ea$NkzZYaHUh^|vAQ<8vzN{ZcKt#Xv3*ZoNdR
zt;acsA1HU6))Our?(Gchm!_c+*3LV)_=A`{D8!80PQ7_^X-uA+T_8H{quq1<ENKr9
zsdQY$puhDFL+p7z3zj+ASY%+k&D8eu0v7X1)3XgeWM)n}XvwVh@nBkVG`<RZ4)ID@
z<<lqEBlP}(7*9Wcex$~dNLcb`)4FKpgWC!m)*AIQv-cE(T~w#U)vuD@$sYrwY=rx=
zEwj$0l!A_LfN>G37FqH&_$j=%8C6V`7m*Y*W)(;p`gTR9bDeK~g_3*P88q{*!uJ!+
zOR9d#aMrTAmB&-`60YyE!au!8JNh6zE@fD;5-A26AftY7>Pm0tkMElhndd^*72r(e
zC*t(s1V4InkeT~9I0~bO9c4d-$vS0#JDiwNBc;D4wXgz;T`G2Dkpn8jR|uDh3SBI3
zV?B?K@cGS0f40?<e8JAb;Yf%#QnTvxu^4|Cd&^J5^@c`79bEW!y2><cpP$=Dc-h5<
zxWTDOnTN`LgnI)PJIw2dOKr4~?qgHcYxG1j=Lly*(9-05V@qIA+cuI{OwXpS{tB+H
z*4(mrev?{Z+qAES>xs-_9@KFhHr$}nDpgtzAiqB6_7S{)Q+ur&k9~iS?Y&n!*Jiq+
zuAgde8&QQ7TArE@cWs^b*u>aSgo-UrlB*<v)Q}Y%yecU-2c9ExQ)WVx3B_s4%e;!h
zW|41P<E{>GqjZqvjW?gFkij6f47aN;^}#le`?k{8Z_XoXbzI?BLXCI^#4RU7b-iis
zd&#C-^Nz`qi5k<m>1LcKUH-#^<BB4-dhX|93Nfd5jkGCM0Ff@MpEE-7h)yHlAZZ$h
z65ijZ5HY9woDCnJDv5Db>e~}zJ+F?vn!!cWQ?~NdE0UBcc8l>?Vr+P2T1HJ%ck^x8
zj788OY!Mq_V@ma=6}k~wajd%a>dGNvU4W+qKUmwlb8KU_@ostc+hKjl$uyR?Cq}Y4
z%4oUx#jtD_43G!+C}EI~P~L(9g2+9=18(ZH*-eJ3MyB**5Q8O2kwF7AjOV?dv)<03
z%=^}CT|UIU|CNBo4<;uvk1HGi3#aBS`AZOxD%ObXQ4V$cSNx7wSRBz&WN(b4{e)6w
z$jJbC3zW7^h8otC<j>df5QZ;;+wdJo_K~4%Wn0Kiieo|Q{*gM|Md<ZJIh7-pCbxHx
zh{FRz@sN+eHgTclj^XK{d}ecA)LvSB8|Cxwhi_@JN=?6``(1bXdSh3@T5b-siJy<*
zT+IV<iOv^25P@u^ZP9IS!yon%_2Q9@8e%Rpi_z)RX=h))6h@k!aXp{G<#Ct3%kcuW
z2HLehTPNQRle0`*b<x3L_GTOiibSVZ+uESK^riC4P9+}h+q|?#ZCf^+cDLfJ>oigs
zcDnXco?aaX7^S||G@a+gvFZ#a<a@itc~lTGeGiS6Hl$*-^$?4`Kb~|E#s3|!MoqQ;
zfKmDSphQYh?_FfBPjD8uto)9M+T0UrXEspWHP+@XRk%GDZpkJW|2wRFcxXwe4#jy^
z$3;>}?YIIkAhBqu`5i6^E{l@Bk?=d}07EW#c4W3*{+^pOEjgo9KgoKeH-;a*|Lvr<
zt#kLIHES&J7hW#mth-HM5UqFJ1=DPWRZ4fkjq&?DIT6Cun@eA$oKU*cm`oWpXe64N
zc5CW;hGt(7IHo2684CvOS|ju|Nx1u2h*-Gme&9>dN~X#$$luc}5L6QO+s~?Dz1}{_
zr^X<>#E>?jpC#YHdZVIdAloRrFxvZ5VBfYF`iipJuezRW$_T~{!>@T}vqRUSs5REY
z^QDh48H?OE1h>MZsj^ZznWz6NZv*KX4hm|uw?O0WEvWYo;C;fqB)xLI9KW3_bZ466
zsUsqjaG3Jb96~=TmeIqN__?31_8JKJ-S<+PtLF`|r!pJW7;d`~U+LDQB^t`Oz6z_Z
zsGjO=$Ynd`a(?CT`V2QPuX<ojv;UcSIv9m?$PjuncYM$d&9<xj0W>Bnl<d+C0y0rG
z0UQ(G;;wy~bvRA_ozJf1yNq$x*@jK4!;919s7QH9$a(tY)GvnoC^5d<Gj$mr=633g
z@BPds$@FYyr>}`yqZ{>)Wu8~Xn=Ch8_uhZ-uCE9pGs<}TX6^kcc*`o9V6@(zS_Fqj
zz>DoVO_)^Zl&m+ZcO_187Ox-M>m4#H<HOuHxv!B=Zw;DN7h*mh?|4R=$LRJYR*UsA
z_R#t?-IQeni;G2vbo9cu7Lz`-nJA0(@deUCWXJ_$C+rJJ3)zyM`oW?&x%bJ$K*uTx
zI`Q5$TZl?$`Nm@~jTfFhKt;bxHj2lsJ)hR=k<*(kmyv)>X6D;bww&G83Z-Wb6i<i(
z!ZKaOLTQZq2e@b36oRE^yqz2SrS<qhJKinCj5MHr7DJYCflV(&fB$pIv*f!Qp_=5c
z#F!18ns5F2_IU5D%o>?+!6QW*nsfC9ZbL)4<!(c)--{qsp@*Gd1q;Z@az$usCnsLo
zG}ezD5%1}tkAoiI)}jqTpElMBRD1_mQ?*HJ?piHdmU6LgAUOe4Q`S7eh5uwjs6wjJ
zVKB43OtbZ5HHG6&c-af36L~hS&!Wa`lPNd;bKp;G!<R&j5A+7BY9vxMeCNf!cfuHn
z6;snpenbsRgCfC_iA?%sK0w=<_I!<hxPs_WofHi(!NAQ<{J?r15Oxd^Pzj>345&y?
zOc6GYhw&{Lbv0X&=P2~{^MVsT83A8A-|wO68=2Mo?kt`OXu5hq!c5Bv%z-kss41`8
zs#9Atmv3lV)+wFIscEu*QeQEL{9{Yx@ZVb^13nmL5bhb{dtEJW?_2$neGP(ikD)bW
z@txM$;%<gWYDv{HMaVYjyEbi0!w!UCJPaX{GQIyP_D2FGM40y+&NqxanS4TINtM8U
z68ppAz)g(C?Mi_((6RhY=>hNEjA@TphI`$KWjTh6yy=#J#PPVQ`iBEg<aH=ieewLl
zq?p<#Cj4$Kxt|Z&wC&3_?=jh0zT<w`>~~-1!vjLR;zdBu*2DT1vG;ADGt4u_(Dc}Q
z&qLP<M}*gfg~Qb#>5CJX7|d(4<fB4a0|H7E8*ss^2yyMp+hi@if1PaIo4FReD$Qi;
zb!MdcCc}O8&g5jDUm-8uX4V7>-;WPZz5F!cbw5~n)qm4*lx;ce;0BvY^tu_2ZaNtp
zNQUFVSCPdg@3dtL)7Mj#)W?}-O5KPoUwofdHKy=>GBZeF@nTY(vEuPfC;qvb$go$V
zyVkqr54Td3vFm%`U7ZvTCJRM6Ren<;x_k}CjnvF)SKl={nkdTmGi_EtZ;D$>>N8rt
z;(nODgWuokzq^r|f2TRYDo14bCU!cxC0>0$)2DL3W{wFAO6wdFE^X+M=nb8$8?mo0
zre>;9(RCBp2;n=I+oLlyV*dU@?Mi$7m#WjtHKO8N$D^viFx3mn+n?R&k%l>DNRc)A
zb~MIfgY(4V$ivG1EzD$&ff*Um5Wc5Q7uG$^8-&G|3VWg9A*yQ3;A8m>mlFH<uzXck
zX0`hCr}rb80remCi=&?aq{V6<M|YN0#%jFuihzUB1S&9ULv%R`HXzUu-Ie84BAS!6
z;z~E5nipwcMx_a!2@%k4r=K4LSD<l>QWBef8=H5=MGzZQ=vaCD;j;OVN^o?Dh5)cG
zR2#^CkXlVVDPEB`R2iF~BZz-9c!vRQ?QZ^ft-1Qyg}h}87#zfBrx1N)KN&1g5!flC
zWeOJfct;A95Bq)lGDpd+*-AwDsGt+ao%y<Ka*M4(DxXhurY*3O;?axA4@Da^q(;Bz
zE{S3#V|YSg1`#wmHwJjnXF%U*168Ukd^Zd)>Z=8cUbTY&=3c3-8&~{#dZ)=Z0?RFY
zT8bq{KCW!DSMln4+t&wR2MHr=jbh{vYbmK*>htSRC(3Po7RJ@Cgow}t<+X>W;9mSQ
zs;kt~pMK6~G_Smxtrn%A{=yOAAm+Fi;4vxHq`gdIBlSJ|TlCqSY6YNLx+at2C0e3&
zN0XRj6llN=H=(z7GLx}_NcWlLo)R(f><|iZIVzrDywD?1Ee}^~{^GM1|Ej+~8MI#I
zblEsyJBw}kGlorw`<U3eUN?DRo!W9*g-`FTK?&i{T~cM{**@10YW5*q#<M&=A?J&&
z9xXrAq=W)}Le!f@_43RJIX@|oSz1nX86@AS5hgS2OOL`UPN^iJaNQl4Mx}*v<ef-2
ze>CAJ!H<DwULSzuycI;2=}p^k8QNDlzj-HFO2=qVhUt8i^)3U`<JJKJA{*POa}#eD
zF0KO8AMIYFZxe8dN{J;`N@$~?{5JTd*>wfiY?XqY<W35BeP7tEy^2j+KVBw^%<X~L
z-ix751y9+M3a|EQ+pVp}tc#!B^<@YHW~!jt97tkJ;@l`cNrX<a>)~MK?G89Ts!6zi
z^0s0{xhDMDm2^KfG5*!1H@R0H0W@--u{9jQX&*aQ=kNOt=vR}iB*F6Wp8mxQaj^x(
zGuE!tDoI?VR20p^vrP>tw$Jaj?M`m~yubB|>Ml$8$kuep%uLZN5QiSPzX#iZnJNEv
zBb`I*yP4bSZKeC>lOuNgsTTr|fNavFg*7yR!cbg!DjxL2FyyT}1#$RwQ+7rh0q|sj
zT*H4}&^F&!e}<<U%}J;US;O0=a~&<E=>wLsH{w7znkAy_wi?Y5`k?VrPpe4^p0n$g
z(M0^^KcaLa#GAT?O9F${<1Z2y?G*Cbe)|FDH*k@6PW^_-7YLsN5`oK#6Ph`ua?ewf
z<Ad7r-@<0KIr88|$GMc<MKZQL<y3*g-8rajW!HHsQ+b3>(SghDg<t~TVdHYkfDUBg
zz^C>L>b>a_!kmRlfvX!nGmt20Y(SJg4F(i`XQ*%4Bp`28OFx85>usbHw9G8!?ILC`
zvEfK|8Q-Tax<|$Rnz{wH(zi5YCU2VIXji$wnQG?E>)Df0JJ1^SmUfmJ>l;^ZO>xwV
zYUPMDT|pRm?u@FoML4ki;)^zyQ&kmF$3XlE8HN|}Fhe>QSGR#b?R=+(uMo1_=hku@
zaH+uhm>PFw={L+!BV_M4EO@l+G}gc_P>y_#&<xo4x<Wha1sQ>fpwJlj$D5vPl#Az?
zD8sPUp(KUh;rA93V_Q@9bKJaKCWF}A1X{;b(U7ut`JXp;46W3&3mSR|#0+Has(IBd
z4;+2%va9!4rIStQBE-sJwQRUdU`a^B9|GMQYEb-mmFyoTts7$<*v%aW#Fav=xZjZ{
zwjg()FRh=%vdJ$ld@^qp{P7{m%OA`YT!Jy~la#kt7hK;f+2k%WMk6c6t3NEza`5;{
zOYVd$oweuZvUN*<e|_t;Y5r80^Nv8~V<Ur(YhQX#QSCTQB1kzmd^TQCSJZ}IV>LpH
zEDh#REz+ydtE3r;uHOH#s*g+)!QUvA-e>ZF>C3dL|E}mzE1Oh{e-7kGiJTlzWA(W@
zToifO!3iPb?O@uEjtFLUt``_n5F@&AiD`=Pr8Vzr6+@T_0ygRWelWXt7V!?O;m~eW
zS|?rPZ&g?c0tUW=6Yof|yftHEJ)6F$C&R`H!mrJ4xy$90rG~=kGp_7hY8lNFrv5}^
zzR_2bcYA>6HXA<32!#8%a$2Ju=pSQw%ZVDo0{bY*0?f`9kidQ64Zg#$HXiB>1|G@T
zC^D!FG}}Eg8;-ztHf=PSV2~Rv9t|>iz_AaW<u-iQQ71)2#FDcWzsK6OJ+)xF2-=I{
z&r+STfK%N!+0*Q6<n*zwQJ-3*+BBHUbM;>nejH*qR#GD>d_;OY$+r)ib31;o_B-si
zI%teh3UcVpHK{qA753gtMjKu-tT7y=R`yRzQuA!1#H_y)<G%m2curHaa%FZ)E4cxx
zd&Kfd%8M%Od~l%uxhI_Rl6chY`fztz^g*7Np8Ab&VZ_Iu36e6#iLTD?NF&}|6Bx2D
zaOZrI6B7_5yCj4=)m=r5F+>dz-J4>0gLfNXTmftZ*FP)1e^k3nM&H!6!gztO&0(yn
z?30eTOHzFrNNxN>YS<AV@@1R_ciY2lfCSNFGDQq}@`F`YrSV6`?q|0aFtBCqX9vC`
z8!gx2UipnN5D(d(P~nbM4xVdrdM169E|7s@m+$GKd};gHGXn1}$20D#L;Y}(mFi~$
z#;^dw=Ybkw9Q7Y`{a&d-j3*lWBBuc-<Z5NwhU=hMZ;b1gP!Z3JwR7n*85;Ma$h8V}
zgbc)M_pRBfz<C&{g&OPxOM}Xo;+Go+bm3y5{$R&px%Spg69Y=^0-q&sYCardOin|(
zbbB9pR-P%Oz67+~eA#g1*1WW>Nn!r53-YOSI*wIM`1s}mVMa~sb!H>L74g<!^0xAd
zb$>HSF{45%#*9gQ!CUzq<RiTNq7kLapQTDB+n1UaS_-oa<I3y4@J92FvkhF{EZc9U
zyMWmOO<vI%<DLe<S>1<VvT6WBXH&HuiJ87oGH@7MKGlx{QPjUxQr!{2LKdf;Qj6$w
ze{u-!T2A;oPOH-%3pKLfDathYn}r?Q@a@WLyzL9yhHa)P$glBM)L#G&95KFY28QE)
zD5)NkwV@Lk`1yzt<N0FsXv7-_1nOlsXZyaCq*ZKoO;A0gYHr|-M2Gu@u})9#bBrfg
zcA6DY4na);_SjR=jG9*!kljBvMF0O|LyDQEIilMK>W5G8yH9*8Wj~caHU4q$j&EF6
z(zK0-?Cr@imekmpIW1Z&c{+}d0i8D2$~_=h)!%4BQJnw7-CIXR*|vMb0t$kJib#iW
z3(_TBA_5{MBHa!p-8jS`AR#HzEg`6s<VbgS#{ffjcMR~JqtEladw=h{_WkVreSdvx
zxm<$Rz;(va$MO3uAujbMT+i}Xvc6Gj<pp4foK=#kKn?;5A7fFxCC!;`LB_EKMX5|U
zhR`JBmE5{zoZ*J;1YDZ*H6ir?b2X0d%d*w$CE*If=3j|y;i3++<^iCk7hU_5J`LxF
zjL<z6mSm(Z;la(M*#~mcjaXUXI{G01Z3w^^CxvpcQ#StqnbQjs1Q+dSdTY^Q6`Onk
zjwL^Lprimk#&_1W=>EsPRXoz<LcIPSF@#>Or-})DygTyBPgoKQR&<IwI9M|U*3AaW
zG6cS5A7}R%50F6^$d<YMKbbxJIkp-HbP8K%?ha`La}kQ!fjuTAqYVzt78#1?HW^N*
zx^O>;U!hb{k0+Bx;)&gLW^s7vEHx5lbLklt)t{@kh~gc>XyfmmAiolARX<-6T@T$~
zS>|)xaFRTgB*fFxl6cb|Z-U>^aMC2)t20`{YrEV|EPiRSp2M?xhe6pe<VDr$IlsdS
z`P+4o;kp9qO+yBn(5>j`^UmM;fudPxT|d4;!c?>)Muirh?zvyYa|6kl(R(mETIhMb
zqS~AGK0Mb&+Kk$SS?4`hWm!s^ftw4A220@c_v!o98M5Iv%p?%<ydAN7ijQHlNb&QH
z!B>JPsox3JUR!Bm|A2i^rJ3@|cj)55eyT|*vRFv)iG`LE>aY}Vrp|4)Q<>bunmeF-
zntXr7Q}7MCQAxTrIFG17kmr{&PIF~G{|>88izYKuPRJ|S(#Az}dh|J2crK5gZLt1_
zbFMrg<pNWZoczB+pljt^N_zIwB&;U{25yD^*s9gpYkH$~M2$ZJ3~G3_OhWIp>o%Gy
zV-utZ1(e9v{1Xcx-nPDcDy9mw%cbz1L4qeonM5^?DY-)@s=R4)YKD2wIx6%>>TH{S
zurda76x!w6nu6}_Tiaea*f86%#|b0(ODA3dn^Uhumr7TwwqzuFJ??{!q|=Uo+OEQ6
zY*>^R{+X6@K1l#im{zTpjQi>_XZgZi4V*0a?qWIc6IQnuFS?(I77_YG$R>*EDV&Rt
z6G7`k2q->^%)@q=d!-}hZLWXcM)+eN_eT<CXUQ_Xh3S~~x$wKbUZm=xqe@s3HGcTl
zuMY5P$qe4Svm20OI;B&iu1=4&rcW$1cGuG^BC(h_#pndHbK;&kdx@L55sQdLF2iTL
zbqj=yc!nOgi1!$ar()4P{v`G>ccO_@T|WkdE5EZ+L+999e{47GWO%gR%+Ku<A?rQv
zV37lPvc?#<uy$)LXt63j>P_}{jh9Q`&({k(9mQnCB9`4lQ%s)tu{Ux*%?^7b%^Pw<
zbMsz?gnCEs#0z2dYU)G~uYd-@oQjI_JK5@6e37A_FtnrPwNLd=@QKBD&Ou)-5vN(o
zAZ(A#7h5Y|OY|KMLqsmd!WcT6wWhHPxlJgR>$)IRrfuf@!zJc|3`&`D=tLqeP8FZz
z!_81x|AFIA1hNzLuLwjdW|}Sqm%E$lY){7I@dZpjK0PFfTuNL%ptcgl*iAuxQmFmN
z*;z0?*QzGnGt0o~WHtc)A%Y(Ua*{nB_varl>T)YdQ-`PG-&AIa{Ug!u?xQ8T-JLxv
zg<%_hu|Jj86Y?;}jnO20h={Bwz4DD&{%m6alVw45*&wAZ>~zs49g<=DQs#nx;xHwL
zDRtPrX1GRZ+dYLUQRqN?=;Nv$=|AoM2-}wQ)^eVuXlv0~aqiEhcH%19j2f0m8t%R8
ziw>p3>YWWL&IDpiZrw_JJC?E{nm6cL!;2ieifd#uKn0~!mhgJCY%n^CxRVib)^_jz
z5k>TE5}@LSd%kG@GQlji(pvN(j4|+`oehwH1tiy0`O|K4?MT&A3Y+?$?D)6ZJruq7
zmbZ&bs@pNw<I~qigcn4;`%Sg+-n5tW&jXB>L~|H=Tpq734<PP0^FP6x*ITN}COdIf
z8>g=wjDN8lU|lwz#S%|8sU$JjL-O0iZ}a~2-8GJxKLd$V&12Yz$mPVso0)(wf`>z@
zqnQsI99D|(Q9(ohJ79?P^`^zv%sp{A&lqLQ*0b~A+53hSX}!t+gEhCril<fAHxdKu
zoGIJjod?g>`js(-Z}Ve|#Mjs`d5+3wMFYL+T_M2zFO6eEx$5xkxUU_af7Vw$co8Oa
zxDzv1IH??~6a#;|QA<qY9^_4n6_tB)BnYxdjz7CLyq!i1pE8BFl<nQoDbLm`?&0rU
zR^!O3?Oze_B@o#xiHziq)wW<}QZ}%wzhQGj!t`0O4}px}hgx9nb30lMS&GGUE`qBG
zl4$$PZdXR^BS_!~5JpIZEGvOezp(qn%XN<&jD<x~vk3SQ-<!xJ!Rt`h^1mS1R*T1w
z-<BO`@Z<_PTd|}QXz+1*<1ANZnN=&9(#?~;huq@W-Py$r&tnOJENp5X>XBVmbsX<Z
z?TBt0if+fKeU4EHbPn=1*m%tT;|u;^Q5!VGExpZky;@2lR*jjnYNufV)S&@5*F6Sk
z$l&*!OEAPUsxTI;X4`5CoOO=MjUu^>Deg6Uc>`NUN<NCf=jYMK0y^^jVd9_(QqkbB
z@^wjL9ojQ53mG8fCHKwSuFBweCL^{OrHcAmMg;*Ga&`?}`{t}!sr**M_b&8Kg=<}=
z&_rAQ$n(xJBcrOIq*(T46RlUZuVmgTCklm_LCf?(w^FJ`)*}Y9mQTF{v)dfCa>cp!
zH)tiwm>S=6P>}+I8}}is5X0GoHH=!$99c|t_)xeH<db^xcv>`jw9JzE+Cxx#6erOp
z@y(fm3K5ISDVBwYRL~k~aRl}6*uU-ThEQ<Wq+KIHepk^oj71wz)^^D;9T?&NaSSaZ
z#z)OebfE__Y^=_~kj*!@A<Eo)hYT@ES}k94S)N9~7V<17;^8{Oqqh$=57pui0wb9E
zzT1N4PDNE>9RhXF?bl#jsl-<?Z3y}$cM_Sq-4B|Id-hbADik?Mzn?>z7}UuR@32h>
zaI&iIT=1MJ(88z@vR^v!QJQ`qlr%;R%wH~NP;|cQWSuOEd76C(!iOhpBCeYhPAf5u
z!!WqRFv;?n{LYE5!~FoB_Hz=N#xUdN{g6oLQhfH(#lZ{m+{?RXo4tdGry*qA^!w%V
zoYf^VDhe;3<21<KSfCE|IY!d1H`kJ2NvO~6U7pN?VSpJ44IvvaPRiH&Rj8?hAv#<}
zc+~<7W|{<>w-|y3_h6HRc=FZ#V2Lmaz8|08L`mS!qr|+dvb||(Idxi4EVANNR4RR@
z)@HtYal>?4vz*n`8aJN#MuH@O3=OVtWT_9Wk4MF&S!vp*0JDZ`o-1!gzz|3|7d|l7
ztJpA3ubpRmYpu#A6wvDPP1bfkSN+dyB`cWiA-n`Xb&jV~PIVPB%3LT;3Kh8#Bmw?b
zmnc#l>$ol<r|0>bkc-M$ZeWdVfi>Rt*e~L3nRLrGVBCfw-UaKcpC>~5>WtyYjZf_v
z)Yveq?(?2ePBU0<E1##;F&+$<SS8WjXK)Kv#00<Nqs5ndX4cZ9xy6d^V^&`Z#*Znd
zYUhGjbw#<JeiJatR0)vSbd6AAL4+AD=xv{`+Lp`<?C~UVw|z4!k<&KBmlYYY0lS(b
zDcSvEA*B-x?5r(h%jBf=2c&He263A*gqdKoQJx~!#6rlm(!xE^PXCB!-ZabGOTVqV
zM6{_pV?P=MWv0W_kbbkhIUYnvg5Tb>jn>1r<bIk!r#34M(mQ*_g-Q27hKZ$Xl3@Y3
zI%z%3RmzDDU|_9$^K9%D+Z1PriY?q1nAjKN6v9sC@;Q_NAp^_Z#u=gxaueI8wX)$f
zn+q5Rp>}!8V>aXKtZM}H0>QEjzhi8WaZ*F|wbuEGUoj?H&6R2QPPkRjz^g_fi`Muk
zbAf6FHBPD(6E3P3olqdmpw1CcZbi9t!R*GS)6-eh6iA_knhWN%A-l0QZ2<bbw#MB|
z9&bwiZ)1zR%9?zstdt*iU~N3(imvxLnG6)Wn(UKwaW$E}tkpN@^9pwRfU5z9_TaUz
zoe%tFPTI+cbMg!qTGf{}b2)P`@1WW&XEb6=c2+yyf*FB5hx*7dw=kc2yNfWMB9-xj
zKoy`n)PnbJAu|*T+vKZxg6>d3NZ5`Vo-!Qjf=#?w)2MyXcMc}^_;RDVGFwXaL30K=
zh4)$DTXFoC30w|ay}_QUsUJ1iHic_{yy#B`<LNajHH<a-ur(mwMTZ>rA&VUy7u{zx
zp=ZOJY_!mmsJwrO_AUnyZ3NN$xyeDiI7+xI{7EcuE~vx_-i;Wzsblm{IrfTKrhIKs
z7QsIlfDH(~TMqrCxknN_RWh%Ab$P)G>-!lHM*jNkm6o?3*FFc{`-ic*MQ<`2zPvhv
z!Uk`b&K0RYn5Tp--YF|GG&s(+&4jNRX1$3qx|egp-53`}|CA$onkm)e^Cawi)5b2O
zr@$3DmrCNo646)ajGH{eg>uqYQFoJ#dtR^;Pi~gP4Lx0qy<pfPV-0@|rI?k~pgdV*
z*Lj9+0B6yGJ<Q;eu<OBtE$DP$^EMXa*Jx_tKH%QyY7VUIWdv;uEoG}}&puMhHxJ?-
z?q<|L*S=?e2Cy;p!y2HeDu?K4XDd(wj7*sECNr%Ylxy?`mntYTFc~Z_MpBV`mSz5@
zaSx;f!I*+;AMKH17-&!+2+i-QI55^oO|@I<kM9*bS|jq`HPd%+Sh^_bSI<C76QFFM
zD*>CO`M(x@y11(LW+U5`Ds5Fq>$X=S37vD07odJKY2km^5ZFKOXe$vVy0@nnkNnu)
z5Rj&ZC6RJ<dL|+7-^H!7_f3)b3U+qVBT(>{nXX4VWmm<{8ZU2Z@Y|Vy*;y&hsR&)W
zVS~u^?Q%Bg?vyNVZ9E}Utw?)acr@0hc0!SjpZ*=U*4))qbxh+;cyh=laIgc5Ck2om
z^Nh({eQ`0k#VMQ)q3mseveg*M6FMPx%saKYDOd;OE)dAeCG0`Yik0*(Tvj8Hj#~5M
z^WfMP+PuHoR2I~B@{@?#`o8aFmxZ6S+l+t*zh{bUA!$t~9i%|N`ZJjDZ7Iz`57Fz0
zF<DdHt&wsoOlG%l=Xpesx|k&QBu!y)+uaY5Igw5ZA;%}7w9i5|(&;i4%#gCui0d^#
za!y(PPbI_Av94ZbfnQ6U68kT=;ni_~LU0ni*kh4^sKpTf`eTP&(HH0SEaGXw|4=bO
zNl&H}y^zN^jla$u=RzKkbR4m1n9|XA){2ha6ll5KM5{*{X+KB^QNrhYu6(&hP81!n
z`=(O{l)bU$$i9EF2^T;6mRRM`fXkG*`|N_6)fO@%tF7nI{xIHw$Xn*9@mrvr(2BVf
zX-*GYB$LOxjxwX^)PF?i2YzJowm<8!)gs3ScO{WsFZx0|%+He1aXM589j8mwGV>}F
zHIMm-OI0XG9x7$)F?6FngKiBjO|Of?sRjBF&UK-)4?_$Gy{j1}<t4kft*_@Gitt!i
zKf9Jj80cfwlCcfy2DROLE3tPOmbcsniq0N?nygY};~7}3^UIeK2uS6=*c>8`N6ytN
zLQKVCT|ZmrcskZBhVFkO9$Rb-zY{R~M0vh7uE>f0ccI{w!7DT35hXk5dV<Rms1(qv
zd#W*ty;=Auaj~kb6jn1hp*pmFizk1*6%X-tMngv526*=+9#J20kZgkmHfE=7jTu%`
zv<x(b9pA!T`)i<D@jlzJ|4imt3+!Y(p);oPLha^bWZzM~D&$cW)WQEs)>)jDO37vT
zq#<Q}n%i(0r>DV8%c$oZ=$>D_FYNL4PqBS}8f6L?x7|Kw$8P?1U8>bWD-C&>L_!zm
zDa~Y^>Um^eFhh&@c4F4R(7BpT@Va69#1YTygl3tHjt8+bZkFndrynvk(7TgJa-d1x
zLS(!r22e(`R+qW=q{w#xDAmKqPeBKnu1~F%rS$T2rTTpM?1LQt%-Z4>3d^CJDOTZB
z6f~dnXM>DylG+`@3JG-F*1O}NO|5(*g*hM^h+>~33)a(!&dVj4YNN#OE9$tDjESA)
z3BwIvxJtCGDK8?C%Kdd-^{OmnR~$QE%Opza(II08-C92AEiHE#G=E4#o4=#^S8CJJ
zvv*ZoG+9K;WbCitMt$$VQ2h|uHPy>rk?kt37h9sHdgB&i8pMGdzxvwh!maI*mup{t
zNQ9KTAlzGjSkCjPi|udcSGFg#&&hMlr7UZk|FxL9DyFu*P}g{*#8rZ~?O1aXPAcc$
zBPP4I5O@9=UgK7{Cca&}*B|N9IgLd@;ah!h2^rDMb}j0yuVl)_Cxu~2=ohnIpSx-t
zX=ANg?&VI{rwS8ZJ<jKNx0u1iGK#8pLZQy7xXlyays)HlIPh-&_3>)zMV#7v3i<h$
z;H%fF%8U4@9ZrwfeI~0D&3ap-N^P|IN~t<M-B;_q-7?xDv027F{Vw?M^TSxJs<oQq
z0=8QN6Ro4P#e3I}i{2T|j1_nxjeWaE?7LbObHmo-lvCtv+!rj&u0OAVo;EW5nDb6O
z(c?FO(om(A=-~7US-HgKw_#Uql<Jv|HM*jppxbvrHtN<TQKJLCKWbaXf##U(`V;0>
zI=Y-eH)zz459e0PFZ9L%ZV*$3i5fT1oxX@QfyXU#diV`s?nxi+eA9Uy3>Qy-o{T>X
zl^Oj6JwMu9S0}xR^Zq6l0MoE3DC%J0s3vRjQ=I7MfSNk-<8q3k_`L8>Kfvf5-z>Dz
zQSm$-VR?R1k9Si+|83bD*%j9mNBm`6EML0Z8ATD9iV$iMexpT-3HuNq3i6ctKCpO|
zjr#GcqgFRjkGNF^BfDP{7CA!Y3Al1(H3E(YdSvg-7WEra^tECSe7GZ-$&Be!(ijMp
z#ZNQkWzR&6YIb_<3A9qIKYUcSlJyG1aG7x2lWoI`_b+OWc^?IKqv1&kfu8F}9D8~W
zapVaY?^NTrZRD5nl+{0&l|4;S<0#roRMyz@_CnGs3$JE&31>%0)J$j8Ccwf_hg+4~
z6?=Eqm_5h>v7C7isjzrfk3GpEwpWRXl~M(m-4%9b^%!S%5tW@Bm0fz-+q6r^NC3l-
zubj%!dRd*`O7>Q}Y>Sh^<w`%qtpR{b2kzcJBO3^ROhmmZm)M3^c!-bLF@h-6jAGAd
z^7^J%AO5kbgnJc3eh+_-AgBH3qKDg_Z?7uwVJimx>Ur3R>>h5<;Vj-v*4QYj7kKG5
zcv<wB4N=Se?Md~MWN(|`WSl;sJ^a~kW<X=L1V5mJrr5F3r`FRu&9cBY`Xsf@+O1|q
zOnN()tk=xvdap}4#VGiEp*a#(rxnMy>C-b5^q)-3M?i++J#rLMc8u5u{h<;aRGl&V
zWNbuY2QLMLRzi4t>>$^S#3UzFOu1*igiDbxx(X{o1FG&^E=Dl#6y#Rvs*88NkY!AT
zdj55IuhL(Eki*LBd2vRV@DCR22B(Uxign!zf7X-cHVuU;(<=X6);(y2XIe_#uXvJ{
z@2Ue1^hR_^xi$&YmxGk2{k3UCQd8~4n+^|@57yguA4v5KnluDF#Fj_lHNJedbv(=L
zcHsO__ybpd6VYYfick<}ZZ4=CiZ?MC{v_A$Xgkz0hWD0i<l;n5f=~!;D1I>eP97hM
zn?&&_2t7ZaO|7phNFfYEMR6!Y&z8h#R%pQhy`%u4jJV`TW<HrUQ%nl%Y*nepSRqL?
z30-1-V0M?0@o{62-o!u=37Z_#oJ{Q(y?iS@Vg}pTZsLDBs$X=Bt0MQL6WrYiQ;!yK
zTAU-`{SUW5dwBd}pM=5EO8ou)(<Td?&JW01oyRmaw3=lG)EOQpzwU{ho#mTP^}2Sp
zG-gYY$2Vl-f|~hn<8oH?O}yCrY_-}0jnpN}AZg}bPiOl~ww3h);C1bMsLorp;{NG$
zIIK|_ByvGjOHTFlU0Q}y{SDIB<UXg%t5Nf`$=Q83ac?lkOg}P5u?>Jpi&4J-cl-vi
z-6mqS>slW%v52=_-(?bq+$Hi3^6|%HwM^8W)&g)LQ*?jrS}yElq+rlu-bN;Y_V_70
zY9md->@>(rBI!uiCNqGR;(9)mol4|I>~&rHmtTJLNk*ojsSxY~%t1+;>nBBz;JgP2
zGhB4^PnO3keDj)~$FjpGi{^+hIay2nl37j3TDlmFJRnJ~e%=NVMa@qn6<l)lmj?@y
zO@bXl=$31v#a}VLjDaL2)U?{}(G?3pTEt#=MX^K3#@|pSmVP!Mqs2xFOqgdR5WdV1
z)dA)u-`Dd5q_4gWKBvV-aJM}BNZb<dZnd6Fvr*q02l5t&REMZcyHF7Bw{H|B_Q+q_
z&zDEn)?AoAADJ{P?Xca_b07VxCSyq^Vbs@B`Oo~;)V^{^4ZK71RtDd6wuf!aW=nhb
zRZWp;@N4|xtaz6M<7hrug09B}@^iqIquJiC!+_|OEw?=bTvpVbtmYV$%3dlFO~D?L
z(#T%dRJxL2;7j~x)m{<)@~d&3&#eZPt*6edkkN+$)Q&HUP%k<KB^q2&r=Go&x*xa0
zDzAR#q5L$GRmzu6Z}VR%odxdBp>Fji;-^4H7E<9CQ~p;e3&>t%%uKH^<wTXxdKZVs
z&-aM^d7Av)b<Snl*e>CBF~;7s-LXl~UAcBp?oSbw5zRHCXT44%zGmp^HB#^U`NHdX
zJ>l3hE~uz3aOceZQiXUsR}_O-y?GyrAMQ^5%aUi14J&d{mMu~jwtu=M^iFhVGOwLG
zn<fQ6#dE|s<uriNRm@(h`y4dolF<y%<&vHeti-;xqNthgg@A$TDnEzWtVA0JuM#mQ
z*o7^l0v*waW1FB=Mv#@SXz*n4(+CoVcd^1W^Wyk<D{Us_AWDN5`Q6-}WnxVmu>VT^
zYzflP;o@?`B0m81{dqwPR1EIQU{7q96tkVy5M7~%+Ks($+_&o>9|+l$n2DxoYx#!Y
zML@#EbkFEVPc1J4YY(X|Zei<1=yCfPO5-m)f5*eseC-^((;Bv3jGoK5+jLY74B!y`
z5*+L`qPq@_*)H<r#l96Tx);cs(wQKwNHM5#w0kIfQq&Q6wHcoGNQm1`ii4i2jf9K_
zWG`NEzy3;K`jt%m)tt_wAfY#!^<CQUimP7>A`*JUb>!T}W4C!v0*fX^qVz<IFOCqE
zM}M&@pO|%0t@_@<8EA@S-=lyA!*F~pKhTb2W*MD=MuGPRxgKhaw!~pYhtznGz4YEI
z{4cM{P_|x~G%r|vPQ?iudw%vTCA&E*F?jVc(FF`LeN|8y-8lv+07iU@?UF&I9?e@1
z4ObrFkR2AjqI9MwjvPSlZsPei;?Lcc>7WYp*yy|u8jHq7Y-d^pR3@MS-l+zPcv7)(
z`M+CfJ6p09#8i$Dv*%<#j6|D^pHHw(RB}UFe(~e$k+7or?{kew((-S13?ZfqirC&n
zsWkW~5kZ57RpZ8F%J(*jF2p(rj!zl90>b!%cpBA>SMCWZ!=uTN(1%xHB3p5WpY9wj
zN?mm%soszGP!RdxEg^!-1&x!trhsh-^4RoiJ|(&fwL0o8tgMu!86j^T`=xxAz^ATk
zAn8U#u`-5z3jYLsUU`Q>MlhamiWEB78BgZ&K-Dmcc%<G4)iDx(qQ<X0Biqj(;tDC}
zSxn-JGLiF?)~1)|Qi*blaY?F(b%~dQlPUG}sLE(Pq^frN7Qn)w+7!jx*<rT&`0LhR
zJ(uCa@Ih(_o<dwk?w;(}ciW5Yg9!d;vrjosHTVr?NrNR`Zb#PMB(}bNc%7B2cmi);
zV82(QmUj!kwp*yexS?X;1*-`KX}_|u!(HJ=XKyywq|$1i=pRm!QA#Q8)?VBp7qtE~
zKmR($ge|@yAiIs*ACx9ti>FsX2LrwqU$A0|W!qSkC!_d4Z2DHV2fd@}Rpe5XM2iu3
zrlLs7*U4RcH>M8EZGS@Y_v=xEY}C3THRZ7o=IjINPZE7n^g8JfNqwWq5C$HnnBZAi
zEup~xOEg-Qy&%8=N~Q_tiliayr)O<Q6n`nA8&Z*FH60eg^k(EwhXWmLU$2x`P*;aV
zLzI|P)k<8sETyv0(V469uQ5<l)-9tBAxTX~4_I>Oh`+IWq;pK{U$h&jA_F<>4(_;T
zav0A&@Cx>rsV`VI_vhe!GlKJwDf(QK+RgV-g}#s$!luU_t@)-VxqDrj53*(*)!*@|
zKj(GfsvA>^E`0G2qCxz)>;Yey3YYqOyKRIueduNvrDdk+MxA+dY8e+L>lgBzn~%#F
zARAA<@3*oThmvfCSbqrN>#>JvVKklUElREhBp({(_Qh3bUktj4hpH`6iVTj8g~e31
zAD<Nas<sf~QCc3nnK2P@Zy}s1m$3Y6ey(&>-<`Z+F1gfSpltAMu1~`l#%;*0>_R+j
zh@Ucqz$%1#g~x?IuaH^>LU8B8a%=1b%KD*EE@tLh^+7nm+Cq#;Hg=eZjG1lV`Lwva
z3*Ahr<hp}B(1O?5LZL=62O-I5u_3Y8cp|;1m$Sg*k|n$7IwF@M`t+>DX(2<;b*gmk
zE44X)nXm3C^?c^@@gm<V4x-jKX3?cBJ-02~$9+gm;vJ;ozu<p9W`;-McJQAf2&>3N
z>XteLp-JvhM3x5@xtJqkJqn0D#dsbWRPKz$Blw8sNVG}^v&n}&xoJ47+pj%Hv~|AK
zjuLDBbMcRQIOluLXy7MdMtpek*ZDP%O6tJEWHt?=mz*k>rwEXv59RA|I@CEykqB4Q
zD0|hb(yNSjY_j+9bkZa0bk$4WAx~%y?$z>AV!x(V5ELHb5^1VB@DrTxd$oi4meg(%
zQa#nv?sj3yi%BZyfwX8#CK}HQq3r@v?JKWeLvSwx$Em3+E?z>amrzqM!&toN;B$JC
zv1@wJ2?^o*z>9@_BtTr7WIhc&*VD-6{I%k<^i$#@Rm=FJAM5s4(3E>3x1X5oWjXDB
zjFq_Uo>;z=GU##EZf06_2zT9P?pE+M<rcZLU@cxGtbJZ<;c~f{vJ~MZq3<Z>X*8T#
zV5KCuzH+eGu}xV}J{&T6e3N%K=c8#@nq!=+^-1mNk`Aw73v6i=XKSg!vEv)><UWYE
zs>k==UybbR{zaXC*?PDXrDB%g+FES}+J7k^p{HH4v>wLdXLeQHgM;yD`&}k36A;Sc
z%d9JFB36u(D<)|zkJIK@Q}o12YS`7|O_QbV)V%T{q0269njtaLB>Rs->qGn9Od|yt
zSZ4_^{%#~}-dakFoPYbJ<IbJ)coGf${1&9<Da*mT+>5OW2L4i`)2`u*TbwnY%BPna
z)pe2jzFm*Alv%^e4nnBkxoj#{^T<~ERz06m?;Wp~goJ(x;X^ryMy!A0%!>%HY|S-_
z&0&q$ETVFFuvx%IDB!|XUrOvE6KA=<@xvv*=NWcpUWu=!(veh3&|;tnPfYUdxe1+y
z@C{Rf+q#tsly3+sLX%Goz9;6^r@OZX)aSF9ZCv|trFCqM)0S+dzl|Yc_h#cqUkgk0
zB!l`cB-cKpke5R*?h~=0&m;s#eGRI5COrG3r_@)*zhp+orEZjNU&mL8$qX^>t^^~J
zFvTIDW)flLtI|F800USK7rqrlW0VKh8+B#D5=Gh6<_-!J)s13%!F-~24nnh%TXbF#
z`6IU*N*i~+ZkvQ+=+XKd8xlaKj|t10S;|?vpve3Y%fJHHRnggkSRru$z{la?lJ2vf
zqvfjs!GR>lj$9iYBR_0mIn<*nCyNOORe?DYUISomK458N)f8jLwNO;6twqp!<X)tT
zd+Sa0XwH90NIc-pd-dx(ZB?KINYh$PsI+P|K7o+<De0H87L-?vQuLMA2s?+Wg(L#7
z+JR5oRmCs#A#hDcXMm>Qy0LXbDYl+Z<WA3Uyiui3SUrVjVjg^YVyyo4)sMm#wxk*y
zx9^MC_fJWU8fhpvFCQGEgq%~oTBbsF4B{V$svXCEl!<ps?{Xd79*{^kkiRg;DGA}b
z=2!s+?@JgKga#5(mP}Du2ClWiBZ%qZKUVtbG#qh9<kYVsZts~pgy}~)f1vn-3Sp<G
z3x|FEUP|kPQsWrq(MLF0(+`+l%9v+BL`2%wCtiurEuI~h@Q%D|X}=z2)*<rNmH?%j
zkX=hFg&yYw7FruiUPFSO^Ye0FZy5sYT4pRYmFub$S;X6ZBf9gKO8wn*lxgQ_A@tjb
z&2}CHEp020Gb602jaV9V{sfS<@+?kz)w0Qp7!7{SP8OY6P~ZE@<9k4R><Xn29Q*xN
zWC-mSsROq!JUr4?LdCfrOYshqwzU`d!k$jwhTsP{eT~jV#l!)+dM~I`TO*EK4s(8J
zbS$wtF>x_morvG_w5`O;jJ(>&<P_kDp|N~$+pa89SofXz`lg|s+Sm%`xc6Ct<v}w<
zxS^thCFk|wauG^aBbcy`;>u(yWx>JA^oxUTNZNASZn=NBOMr0X&6sF4%BL2yS$yfv
z7jg#-E2}ddWOvW-*4)=jJ6IvP2Z<}$euU(8N;5Y($wln@Lb*?aLo~Sg{gJ<3w$^Qv
zc{+-ub+!K$!l^wkPRcmvUnNET%-xh=IF#>|>+z}0MDq@oL}y#PxPl92IAv{yo=jP~
z?dG|3Np(}dk|v)E3FPWU2_*MYKE8^YLAfcnpd(8*MARM_c(Rt0Dt#(w;@(FFTLxe5
z8hL|Tqd_)9VsZ1eWP1);@$|BG>*CsvX}HgaO*&+!0vCU&lx-cSvJi)pnLJiUCjGqE
z504r)a|;tLi+!R420g8~$D&)u$#Ik-Mpvjsq(+%9H}LL+((t;h6Z_6xpnC{?zdj~(
zx=#LDn%Q^9(9BeLkr_|?dOBOc&ksC=i}ylrC41e=f3&>8!nQ?%zqC>ITB|@qPwG<%
zZ{?4ClZz)inU1oO>2FMbVr0lvRs?(B=};7Y|L9AgFX_u-)EvNA&|*?Lk)q3#EGceC
z-n%F1n(RDckM&7wTOe4X;f@S6l**R|Up<qWfEJ5vf#f=-@XDiK<Yaob)rUzu0_)WH
zX%yw6(58I;!M2>q<BEp5g6HEt%x}qqsXOekOSti^&Bp05+G?XXh5y2z{On5JmZ4xM
zxJJ_aSlvY?vapyiX|5%I0<ytIB+w_XA57x97UNXItlARDfLX(t;WV&fHSj$xLkj2f
zuLWo6J-H3xF~<&`oyL>*XMx^IA|k`XSY%YHu+xlu1=pX0_@EEFd#(JywQ|js0ZU?U
z()U>0Dw+|=hp&_>J{WJJuKQyR%vigPR-K?a)kc?e>}Ysfb){HFYtHsq_xmowwWXVh
zV$Y5?Au-yp_6|=aVER1GG(V`&U^^64?^Qt8L*Ypc#%)PbUE%i)`f|_qqgh($)Ht(Y
zRkoq$`Q;nfyi3EQ*5|1B*^9KsrJvX)j`3jd#%AM%_jdu<iO_`j)tRNO>GUnm9(dj4
zh@I`x&a3WIM`dwan_p+@ydy0}J5;{i=_7GgZz-4S9lJ~ms;%{*t9-{j=_8u``GT7J
z;-rDC)vlt3snpifRQp+c2Z<wHLRLzY0S~))SW#mNV%<aW+kT|MMehzJjuH#54Eye$
z#X8s2HS8xm;14^TT+cv0CYm}Ags%F<cDrwT7@ERY3;WV<#g&d2<k@I9b!e<|O?e_b
zm$EEvc?S_yRn14WTnoeMx*!GGLFG)6T*VfkqAyrazfPV%ON52*!XHyuOsq$qo)0g!
zXO*EnoL&^h;!?=1)5JZE^ky=Cie0Eu#jt)?W=(aW$wiPcbc}>ENi_6s<WZcY)r72Y
zcu6y?&+lh{3e?O@r^4Izs&YEw$hVy~l!^@(hv%<N(IQ-Imn!+Vglc5lHvRET*yjNe
z7(=n)fjJy!KbBQ?futbwhi<OpQ-d%11`j=MlvBj}{5t0stU|6(K~pXXt$95k)+|FO
zw1)Vwk2CpN5<GfE=6dPwzn!Jtdv0zR(L;P$UL;$oU8<`4GC%Igv$56Uej-U|IL2R|
zth$=w74o8@wJr-eRZv90S94F<DVgcH%1|n-$A=5js4qQ7&MLfKMc4)xA`5rctWk}2
zQF6~3bM0}Q1)OPHOBlAAyoC`W4V16|vw=-K8K1T`e78QP9YQI!H9p;U;bzrF4;Eu8
zZU?YY57r@|7FVnZS={R|+8FxQ^uuQnw!`iL2as3S4j1`Gb?4U}Vvn`cW<HEm+FV;=
zZ|{Y!Z*nP^Xf{2{Fw!*i*?4lHn8bCE5yEYB?Ud1S4y)^(xRVid#!Yj3Ke6niFhi%r
zR;$ASJ9#EPwajCXo_WXbA0kW6PLDP$uL4wLCpLML+3xL`FGKsQx|`$$XX5UKg@lIq
z62`alwaKoqGH0xqwr6+i_6rZ5Q!*&ud)tUQhR_5@?eCnMXf=2j)z{HlKWu3l;zp$*
z%-#g+Nw#@@-jjFVJXA3;4sz*;mtN%UjjngX*POw?sT_D4jlZ3`UXroBy?{$<jZ54(
zm~U0nATdF)AhdQf<k#EIRbRtL3ht?kWCl-Z-%DkO_J})?;lDcmkgw<Y$Z0Vol@GLh
znoBg+*qoiFvq3rmmd@g-o=k7VvPhn&h|p*}h)~;pq)ABktdeFBr%jtt#s7Rcs(O{d
zX(T#yoH1}YxGdFsPdMa5<E|p(@=IEk-ub80zFq}ZNYTeVQay)Zd;S?#{EpnH9pp@<
z^n${bxa?k;Uc_{}Zb)w=cH0G+^x*yB904hbprqD(%b=yrB4|Ua7^rpegIX7E2jwZA
z&fSG~PUH(WjP3UX{R0<BsLF0@ossr_UE%hx(k-E8qgcv44r<DvzF13NWMEZ-7DRtP
zA4CEp!DI5+1u0sfAgRxh(W~G%w0(5jJjdI>-G|LU`b#nHSYBrrm`>Ki{fqkMbt0iV
zEA7`oci*8gNB7(`V1yX21Ym_?$xp1XYEybtcYBrJtC5rC8kq#xIWMEcR|)Q{uwix}
zlA~$xLIL>PGO>AcUBN!2-|Z&1LzCtTfP-~y=}`D93Xn>eb7HicZ7#AR(ToRQa&U)u
zdpe23$Y2YnDJdlX<cu(v04yG*sf9ItjW@Z7AL(kRnjC9Q0eUDm9&a$=uRdZV<EQHh
zs3LNDWn7<qAqP+MF=U=^(${qvF4Ut_rPklubH&u%BZ$4*li%+hEu{G1)!q!O%cSoc
z1Uba&SUEx~cL9fmHaR~%B#AeT<dZ$E@aLw|ufRLu+zm8JB5e$(3Dj4)|L2Xo!x#ZA
zN0<`Ljo%Da6*%5lM528FO7;f}kCkm>j3*2c`6Kac_9mbcBA2`#geKvzcX2Cr#<WUq
z`FsJ>xgTa#=?-wU5_N3Kn>VWVdjLRc*xM;7F=kBar%#}jQg@%$1Jv|5E8L&fM;~yB
zbx+2_;^l0u<^|3B=-|ydm1T767KUx&U}`1EXjpvvlP3pghh@4|*es}1@D<K})VuZw
zlfF+F>}_80KqMO2pR4??VIs&onIZ<nj>a0<8F>Ku=&zM7fSGM$xov6>atp<_D>eVb
z(XY#V0G24%sAqsa1#cQ@!}UBkKPkYKR!dUZDfU5vc6U8zJ&8;CnVbdQ{jcO#$Xmkb
z^NfFzs?Qt$z59S&q5wzw%S~@UT+p5iYP|l@ymn|N3|`$-n~K#g7lgGLmtCr9!rP#C
zd9Ke4Ki_>hosxNSXb*my%@ae=io+)`Y3Cb^yNBU*;C+6$ZkAuM#!`%i*7en-K%>&{
zgHeVd#p(S!do8dg*PWSq8|is8Jq~h21soe4So@9oRX#L0eO$;*U60*nrPzaR0pPa*
z#O0_gjRO6x?Qa=jJP4ag>%p<6os3o4gMYCv)&2E*`96BIzn7mPIRjDyyz?!gyq0;T
zF=#kz3;JsC$#Oz8o_uMvGXXWEh8I~;8=_MIShxB<)J2)H@P55uV#3_~YiG(88IK0w
z^dIk^=FV^{Y`)J6^DnK<&$y|ebUg~rls)uH6Wo8{h&lkwWWEIk4HCW>9h3nmsraoT
zAKE*{e&meXCgtLuXK^6!{)+Y3tiGS`PMaLKrAmfwx@fR{Mk&lx*xC1T)Lv(fy6jQE
zrsv_X&H!KYPNSEq?7sl|djQQ-ZO;oLzdr!|BpB+9f&h?^H}u@&86fYl_f0ckE$&5$
z2nYRZC;pv(JW-|tKR4x5M^Fa({wUC60w8dW?vW}jNiCCTwMiCitz##}{~mFN8w;Ab
zXDCXe3auIb`+tNJ(9Au;(0hXaT>U?_$v@}_{_8KjQ(mB%d-xVfZ2xDU|IaGozrBDj
z*=Vu`WkqPS&i`cf&%R`!nR_xKWxfB0%st=H%sr~A`n&(odH?lx{U86kVvc6+Sy@n-
z`EM`%_gVk{?)ZP;dYr9ho1#v3m%t1GKhpC!IUeZ2NdwTX);*-EY(4FD<>{IX2T_6L
z*}*DA{A`8cSakHA=}et>k~$R(2C8(?$jj;uzaiXS7=TPSooo6C#!_pzRh5Lk#sfXt
z^&SIjQb09a;{pXh&jG5Zmwq{aBR((LrCnO>37YmM0d{ewmXeHSAL^}~u69oZA8S8Y
z8}2l4*!{~&?(;88G+l-}_SM3F<IoHp7=ZA%vp3(<|1BxrG;-H`OMmkr03Ebcv7XNh
zFtx*Y19c3&Z*)Dq#cn*J?`~ZTvg+{wMpl?XA_O@ChysdtTPPD%GGyXO_21Q#U$TSN
z>LqyXg*8bdAn)k~)<$e=0$_n#@=rOb76YIs4Y7HzDo}x00!0X}-y8$LKy`s$_?3_y
zK$F8?Jngm~56D?s@u`J}$S+P^8|J^p>RSL>jQg%1ZLLAi6GuqXa*xQi(`6X7$cTc(
z#er$0B0o36s#vj^ZnuNA;4#P~=bZv@tYj5rZgGzVA*i97HLcX{0@$ZSfHOEvPSb!7
zX-0Pp_?s>MzuPgxt~hQ}fN~=M`e>4PS183jei;qdfG)T}(W5blO|&>Zf>3bg8vN<>
zU3oSmnGiCg%bfs;IJDa#3OG=f0ILoU8f3a0>_`sTn`^G19i+n273m2VnhRv;0)z#x
z4)G<p8qS9cV2dFbplMdWxQnMV4zwIip-D29J~E|np^^9F0lQ5%z(0R`)8JQ8%JBy9
zOSlE5-({RbuP%-g(VoKwI8lxyvP+}f=VQR+yd~9WC;~|O70ZCY#iQ|@Y!^**$5uA>
zX35s;Jn@$A@f8?F5x}8#`gDtkm6QWOdS8WCpTCbxe||1_`5us}%Dq;g!%R24j^$(4
z^QI*v-Asw>ld9xy7w-Ibvm+%TN0~*nkic{wvfJ0AAJMUp-EQyExvI!XGs|*yHHEv5
z@2vOAR%Szn;qI&XYuPy);7AY7@z=Rzu3=oi{p7#?>&uUuw`VledhV@0LqGVf5OeM6
zaT`PUfwzZ+r?RBW;rghuyF#k?6#UgOK+-$ea#t4Fd|Y|HRXNz%*aRHpO`Wn%!kt^4
zeCnh%79^Y^b;?s~DF6bN-)ZF~?Xz!nqNQP^*v{mlpd~Dou-glquyr@zD1I=cRXhKh
zXsN<CL($D*TVzd5r;+ob`^DiX0+cc>WNAH}+}v-lYyr>ErHi}uM6|`ZIw#+E*ey*P
z^Yx6;5KA=!<KM0A_ku7nub-%SM?SlI2fY&(Wp$vGipC&IvaLq{@Zjc`?~^N9slj!N
zmVKXEfvp}}ZgGc8<b*SU&UwlmXc?Va*oBAu`+k%0LA)JRLxW0}%?*Iv!*_#jiI)l>
zOdbJ%x$_w@XO$$eN5AX<$lgx?K-#GUhAcOlQUBHvbfAyGM2Q4f`u7`u$HaT3n&TeH
zs8}TIymqgh3!oSxi~^`PfR0qGx*_2lc@7RJuA(jvcyJMK`!H{;eQ&O9k7R0f8KmS_
ze{YZ>6S@J&f671tR=A79o=STc5U;_vVFD9DshHG&8^`ecY`tI;%+rAhAf3FXi$YAy
zn{@B5{moy|YVWVrcDR{7C;ju_74g`cVD?<aME)#}cfpkR*|ircyTw*BBkvOJtPgw*
zR!kD?KLToN?(<A+C^>Hn*ejfKfPNdiNuz-Qv)J5jp|oYQPaS9dMp6qXE}G$UqkvE8
zVD1>{g>;#C%Tx*GF+5@$R&OZ${kz?ydwJu=&7B+{rqI6_&?^~Y^9GLMfOe^cNkPq-
zMIm{rgjY?ed4IztAhTkmnHep2G)v?$2|^QHG-o}i*a1om&Ie)(VOB^58wM<mMB>KH
zX+U6;rH+c>(6hipodcz}34GxFJS<bHPqhEWhTVM4je&{!F^0=f{`+Y@<tKUvH3tLo
z;&zIX_V!XN(R=Ku@9q--#3JuC25cgr0!a;KcnHY2=;pcbGTryP0sat7&e>fj=9<Jq
zrFsuMH!2aA+)~TYo_hzQ8L~u3`WXP|9T{=?8OEkxHHL$GfAoff+_|C9-<;&{w~@Pb
zdnUHz(YX}*(mkOw@&<t0BW}lA)&MZ<2T_$Ot7pH_c#+8waNWlxai|w`@>yBxK%V-j
z*ud*z7)XBg$`VZbJdWQw!MbX<<><KfN{p#)8^ACfc>AJh=7@4xIc_Vluddd9#>D8P
zm?72P6`B1&Yz5*Jmr489aUkOh1OUwQFiEtfTeHTaTrov-91y8(>qxx5)G4zfhqLVe
zTOw`;c9|t935)*s6@D342dL{-u~0)nNFW7Ul{0|WtdO})iM^MZYa0mbv9|@UY1gBo
z;Az0I&<}r?&0!^VI}vm8KAYx@4vr`=)m8^9{j1j`oZTt0^NN8D90z6yg0@T6fShe4
zQ!eu7Sa>X()*Hv=o>E+5#<3(J#|;ojb!g_(;-OUKy(xfx5EB7rYZF+C{o2MWi_!ee
z{rfOd0d-bbqd!40So#nSjXq!o$?{kLMF~!?(_qD9;bHiUD;Q^;0D@~#YBAj5N}3z+
zcTx$?@<}-^P#`uX8)U!rz4yZPdt9yngL49KK3QAl7WzAHO^(9_vt!POr1P{pD}k-b
z1yo|<x3@uiV2{u*ZKUct42|0L{g%#+LQqapJ_0r$0p#aRcLHBAXU$0i{6qGW1`fRn
z<LeV>Q4SN`d#*~C8bfPWhoKJe!fR=7Q2cmUh33Q4_quRK-kkw%NL&Q=4O}qm(=nv|
z#h)QeumNlckqFQaJR_<C^=LjSq-%>$94%VZBw5T=>VttQ3Dx3-jojj_n$Z!*moLpz
zNUskyB8U8~83T8iMH{H6p!>`3{zzujt0$6H%A&g;2x!G8(Tr=r@*q9vrebt#-lCh{
z!8oDmNSQT)9t(Bv!)T5Mr|L|y&+OVro@KFV_nx;CpsIueUafk#X}0>&kT4SH3Cr$8
zf!^nN6-Cs5@YQxz`4-SFi&!Yr-V|#Kf}fJUpp6*FZrudjHnlL%=X_{R-@Kj^VzdyT
zz2gr0+B{jkBgUnbewawce><$Xyf`N<*PrxR9n97VrxDwTHGpwby8@9vJ`K}}h5lK@
zf35glmt4buk+Apd{*jXB&!4c*eovfUd8zw8bwsDcJOb^9fJtgz<lR&sO!(NFTk9iv
ziMFLm?AKZ9l=ZIVkW~MY^!5Ptg&5{BWp95A+?Vwp)WS|bm6C<Ez@zj0`>m>$-V<K-
zjcCe6k*!}DM88;V#MmS2`(<f2bsoH4_Z-Lg_|Upw5swgL7fAC!k95tACP3UJnA)w0
z^Y`AzHE#z}y1IhNk3#>r<|IrnU%fYI{+jtF?6J+HAOiTY8-T;D6z$o)A?#53xAXxW
zS|8j@X#BH#FYjRTG1n>Q9iu_?&^Vp(+S@viP(Bu)ILLp!zyJ8W5!!<w<HLFSM^Qq(
z{ce#?41HG}6qyD-d7k_67C*2XmIIj|^NM?fS54GB-%M9Q`OSKgocNb|k}bgyAYd68
zGpaePiD3un%GePj%)j57FT!ZAfVzhs`rCs8j+ueSuHWIt_#iMxYXn0-^zyVztpFl+
zAzHN1b9WYyeZWD40NEJ!=ohD=XrTG5tH<C)TLCkrt>^jCVeKjHzhfDOC4nTZ+1+o)
zLO=Z^lmxAu#Q<moNL$t}l39V@HIk<TPdZg|ZUHW@6-c9nF!$O^foq(rnyq9X9QSW&
zt>}B;=wGDp-wpjYblkrlc{zo)^IK+ds0V-RvHS-idM$@G#S3p4`v1c>|Ni)g_0U@Z
z^KbwAjk9@mYVc1ifdBAjPyRny=4*TJ*CfhbiZ5V!2~k>mLI2v0|L~js^Y>n2ps(#U
zo`63Az?a9Eo!0ReZ{_8SnGw0yy(iR1)eD7LXWOXm{u_6(0uS6py<WD{-!kxR5A&94
z?F43`#^Sz;Do<aVw&sO)&S=!eXXU?F)ZTcv!Leh061x89Dt^5KzI<%KYD<ZU;@LC<
zx-^)S^N1I4TpEHzqxupa-_9uT2yKJ!vv;&Xfm58TRcuPWf8zb0m-~O&hwI?PeXjq*
zcCf@hvB!V~ld_i?EL~{739=l{H?hneZvd8AZ>~ZO>Iap$u*wm5YvsI_U&Y}NiMMkO
zoCmA+&kv<j!YZ8CwE?YY9th6WbJX+bN98TN2Qug$Q^P;ag1ES7cfJ+T;CIV9C$Fps
zIR81Hx^osQF3ykPfSw?Gzs#^6Qv{^9Hvn&{TrcGBWj_$?HtDb-#UE?_;xC}jkz?T!
z#qKulZI>wCXr<kOW_d*n&we_%KY>PZjt#jv0VSV8fyNYL47;9%FGtmgl4u6pddR3T
zI;J`k6O~`J_kxxTZZ*#13cD&%)6nB%P-fRDD>>RkAkh4dlvY2z^8kQ2d1oZ5aA+KG
zaQV3aj!ew}LTbE;Ll9yc0#bv*G^v2IgBp;exM6KTtwF<M92iO^#FJc#w)#Ed=T?=o
z{*(jzaeotV0wtObZl*L}+@IL!%U#@iq%+5>E4|h`!>$t!q}T8J!wqPS*hh66g{=$a
z;!Hv{Y>j=dprw2klaIs-Dijq`c^WIMN+QJ$4)n^d9=I2jY&;~HaRu{-=jzkwq|rG@
zO;~{dUujuR(u>!urv&(e<3pR9UnG4ufQIbF#Z<6;+47NDHA@pwX1xhw{X!{1q0HCr
z2;kHzdiz{IS=?jR-2XtVL|4Zd>o4?o8e>NS+_Yc!{@A8(nV8KtPMRM==lLqfR6Gfb
zJ8gdn+Sr#S3mhWemMNa={ZM>wAgCgJqlH;>>u}Tno|9kE(Gp5&hOT5BuJkil#jKnF
zJzMIsIdNwu_S;%e@|_uYgAeoHiB=%6q@2@!e?pC$xlZ1Dcsn{ZR^{D^=f&~sQ}C@H
z&UtV58IKSJ`_}cS8)bI0^r_o!hw6v>^A$(Q$Ai83i*<{z>a(2T<Gkw}C`j7HW$k7x
zhyL8LP615Ja=-~pM1FyVEpqO|O25j-oSi-(=#O5yI;vzpo3DM_kMa-IC?3{zZR~%r
z0XZ95EjuO*7|%O4FmvCkv~(ho$m0l3EHzKyt^GzGoaDWizNv0dT{x`&)kP)8v`hKN
zjoP1s>6oYnhZ{)dpSD1d_9S6OH5G>!KngvMYmD@P<a-2r&B9yum%57|%Ex4y@mUPD
z9T+%`0}BVNkxkZ?3!p7V9#j~m+e}rAq)P|y@VZcSw*rD+i~Z%^V|L<^A8MUzqXo9W
zjRS%szx7l*kF+OQWK=KXmS=?}aM9tTWx(ML7IPV61Qt}C&^h-L_NJ^<wcDWw#nWks
z>uV)mdCD24MnGEIe4kT4X{$v@)6{j9D)wOJRp8frX6tOpsiz>^T)PxyVV?*eNQbae
z1E03Le0px6Ui9w{@R1g<HJ?K@vHr+?H2q6`449+omEkB+#W0SX^;mu=wdYe#_qG1z
zjd>#R!gd2??i*62x|ReYCxKYG%W3X9!7<%~r=vxb9JSr1-Ek(JF`V#CcPq#_)vthR
z?hWemnJJ>pb-76EkprzV^_)_xYJ!nny(JeU)V;R$Kxwi`!lr7pCS@m-{mQYtbbGAl
zOy9}PsZi`_XOMqwDjX$xgXB=uiWYJHUW{j?{BWYuI!J!ocHax0){9?E?K0gnQKBUQ
zp}x>`uT~$iozW6s;W9HC33<pNxEsA}Q0$5%d&@p*vx<KN4@(v{&Lg=hPRBw;1)(#E
z{sfe!j!Ru~Olv^d!vQOK8<;y5kSGAIsLMix>_m$DR=mC}F-WRbqw@+l*34r{o#8@D
zWVJn@-_;U?q@~D!ie)u{df>DhwH%!lg6!uw@P&dm=>@N!pgmzg2Vz7CXs;9#9=Av3
zgHIvQ(XN>L5gY%@my*Wl;t>)CveV*y%HGD-6D5Tp-Z}An#6DhcU2N8SG)2aGk_)`3
z8j(qK_=Zkar%LGcqN`!YLZF1Ch~Y{ibdO>t`#=dCxj!Mb1qhtdnY}u6fXS%x>)A&W
zb^iJqmXn`h))i;eIfbmolqVA1dE&KM4bo|s!U^J@70R-}@Ah=<sW3I}{J)sA|Lm}s
z5ChX`1+)6&KY=JPO5;6psD03S2C6Z!q5{b)w;WfJgr$1%^i;SYCWNhPK3<fgl--xR
zBaTkv>|KU-118kqx)wPV_b(nl`n5b)%n6q>@U%<~yPPaT*i-6&Uf#`SGyT#s8;L~Q
z;r0Z$m61eO3M3V__f;{KXX?n_8@k4?Dd}r;;(ldmy|Kc%(b>zWxtzyey6jJc8;m-c
z=(H&$LY|K)*IxAmxb%}OBTlh;x&TLy_{D|@slqpP9EsB%RfJ*Vq1DC6x=C&ZZ7796
ziO3b-7@RS#n9I_i>S*$E#_5(HVtZ^@-{Txt^_s4nlr+u>^Um(C`LJR%<pz$M-@(Pn
zd^p--ov2@cG?X!j-Hz)u(UVY`&EcSMx{Ngiz(%lWzqvNuksVVMm|BQswtI=8V-ddz
z2xbrJJ@F<xXeb11tk4X^p261Roej1(PQI5uxiJ<>#@zrH{MCG4=Oku%Xm|(I6l_5J
zoUaEu)Uxhl6ytoIO$(qY3g&bf{dB7i0qAw|tPy46PYyvHb7$%vYu%<lrOi*!6+Jvu
zWI)EKoU(B&2J(5d2U%PG5~mBqNS3sLFYBiuWcmpv5;RlAJvTsvy32m%x7q2gYyLh}
z+Zvt!-UMZiLXf4OKr>9}a1NEfsxj?OFaaDH>F6=%8nuRvgV>Q*N+5?*rWnUn55POE
z(eYZXDVYCyi}Wk*`M8(<UP?<To~PNsQJCrasY&V$*zZi)f4)=y(bm0;1M#IFw2=Rg
zr$)kvR!9%9yu{|%+;_ig<R1F3BS-{@etHt=_BSACZ(aW6bxXD7tldedzMS1JGa2n}
zG+o`b(Q~h;tK-bGkCbs4h;JPqzfg9bI^<X4Og*0$ty~R4ShYzhBkY-I-}>g)KFTJk
zZafy61~E<y@l5RW$3%0r0$;s%TF9hpwLg1?Y_rGtH(d(@QffNkWMoURrl^TNf!yQz
z>@zx9>w+rJi$P5tP;Y%PsV9wn!fa;?2<^*|u=9A#6+oZiwU~)!X+37I4h8iNz>hbd
zw3v`K!*A^c3<pAvDLJY*4720jy=-+9Az(pDT0HIQyEJg$<f{b^#D1#yO-`%qEe8_K
zZ%!V-*%(QaGPV3@A*3W}jHitNJhlq#x6t&yMvlmd1)$Qle6fZdX2CG<yzkWj>;Gc!
zJ)@#bx2;h?F@TB*5l|5WNLCRP5Ks}xL2@pXMv`O-BosjeR1gpeO3pb*h9VS*fD$F=
zoU2HZbKUjo?(Vb4`M%xf+#mPX{nLZd7L2#vC#<#Rnscrl^9N_-1Xly8l$4G~X+%+0
zy(Fh}qI8AwsT2c)m^%T~`<&cg>#=_$#k)VwPpDTw8?rE=*WA5NN1%ye033o8MCg_{
zVXS_G?)s;|1Py7W>oHi25FtG`-G#VUUNY#KK%z))g4pWV*0CgQ1Ye4qIYPMcPY|@*
z&i+G-_04bc|CPU3xV=5JxQz=L{OlrVT{u_J+{CxJZ9Do$pL5&59JY_E5QilxpS@4&
zjPf~eo)qc$Qk(CE?%Gr0Hhh;nMZb9yE<u5Y!Htcpggnt&pH)^(*C;Z2s{U2OFm4lH
z^TTHAWp+7X^TkY;8%1}o`H={dN3m3J99dEdNylfZRBGWp*I_<*9$>$m=>Zo&fL>c}
zqMg5C3v`Hy{Vc}96gFY}=v0<N2;MSdTJ=x?pyaFa+I6hrT#oYI7SOzUmqLjm|15fN
zLQZ?1d0xvSp+oW<2WpN|%nA&XbS2~w@T%I2&!{LC9cL0kW}*F)4}w5u$Lxitw$S~`
zhEr43beB6f;ahmnsqn|*u0xP7>vw5czoV*AqV%?O*XSj5q=hW-t4<mxTt6bA4p2&0
zz5P^#f%`$KV<pXJ3H%+!O1jBS=C{t%vMv1%=KrC@l0>TDhnG}-r_H((hm;<=4h5Pg
z_Jrx_ylsru*ij(Z?;ebx?{kPG+bNH@RbA6~ZUS3rdb)(FvHeED&>y!M_P?m(!`$pl
zdrs;s+5d8grrXgRqYD*5o~hk*z<lJL%Of|0>n;Q0^8@g{@|if6VYz{9S%`v28?t1n
zj*VL*PAY2ycbL4O4A>Jw7uibCchMnfoTkK|elFe6Sqm*oAx0><4h3A5ngs%hkD%vL
zd-CJUCD3cb-2U7W)b6nWV5WYR7ge+2CKp!rBVsW*mtA`yx7d1W(k7SfAlf7?cxv>3
zHo;hb9x&KK(tQWMW+-qjFCAS}+-m2@)ENJuXQ}-t%~31A`vpu>+Em$j@a02}l513^
zF#4dDQbmF<E%pQ;4`FjqMjigH>+?_bA30L9UOb!j`;wD?Oh$cT_%^2Qh~sV?&D#bA
z%YiXixhnd&v3>jF1Wa)?Dz+6d+ug(Ior|n`uBclB9?~IVu0zR><I(}<o%y3ZO|@C0
z(9AS~ADuJiE6{ZMqDud9jh}5;ig8Nqqin<Tz#q;<3PHGnpB<&=^%;<Xttl14xyrFv
zeg5IV(X`cyA`%?z+8!7&uM15;DcK)7Yvt|A3y;Ol>%$FW|LAtPh8(Fgh0_PI$L?f1
zHl=lEhu2}8fEv`kvml0Bb8{4(U~E|E&W^)9YJaJ7HUN1)o)9-)Y18njgrXx5Qy9_A
zDzbk}HFY4>9L1>p#t|wcPbE{3I@bqt<`&kJ&z9tJP#DhyBoxf>_w}~aGCXr(2=C#i
zBXt%{`7~FyKB%IltJR|UjPtQYA6r)GsOC?Tq4F|Raz!O>XM!uSv!jWhklt|g5>CYy
zqSpjHzfE#uIA3nJiL?abg7+Nm+I5b8WY*MsEu~!&94+@VwJ=^&Cz>w?;g*kp_UW9Q
zkL%4h3Bz(86Hy4JSZqV5n8l)fVh@nNF91*+r~!(9sFsVG);s;v*HmM?=Q>0|nfR}p
zO^HmIvBLaIk;Gc%Mpp*9@95jSZNu@}f+1hAZhxBjTtUbA;A|ZUQfGXqO~e*`pGU~*
zyE!H`fv~N)TYZVUoNGVY?0WXRW+SqRt$S)YFW#8XnFaBujhA^Z0nqz~1os2_YZ>j5
zgGm2D6RT*KZ(NpDWTv8O>pEl#_HR&<Mj|LbZ+G;xi8Qkj(g^u}HKZHv_mH-nC(7oP
z=9^0s1(TsBz0j^6)vRLU{4ATXi;M=dg&2VV@&V<}$e|c2vK)K<<<(&Xz+TzQWq#y)
z&PdQ<<ufptS{&AP%uS+t2#fUp;nw-}MkLR`)$_;Y&VOA!35S$CNN~-<o2|$AorFu)
z*6K4v>n9PwBd=p=!ljpLh-jE}4&#g2?I>I;r`-R1Fjn26n8~-FxIc_YSXf(nopf`O
zP@AJ)KU-qNZMyASqDic8=C}zR;UjiEhRh}F*?n`vm)Dwm<QTb6i}RQ+9SOVR93W+4
zZBl5d8I|R<_E;Ut0bBB-vMoF~X7ZrzN?>4jHx$X{kaC%2t1A{3P>d?97mA&;e3(r*
zlm9tB_>eatb{$5M6v9aJMo#SXr6Iu#Y|A@(1b79+KZ;6c2khyrR8sJy&UHUvB90+N
zofB+93<*k>Y2b)94hhPLPum(WMtZ@;%wLCtMT&fb<vjMn>FcRxvE0P@kUpqM7Tm>g
z&_!T4l$_~6`+GzE-;s`J8-O^1n%-sn{@^t`w^6d4_8e$5zdIJu)i-!dculut7K=+T
z=(5r_?EG%*!TH>=?gqvbOZkj`tuj|7$t=R>V#MdB1MdWiE3mmv76Ytle-u0`)a;7n
zt<H~x#;;&s$<(|pzL;69?U_tQW?$FIE{=t8^`XATu6}(6pN($Ly={Bj(NAS<GZd$$
z!n=nC&#o1eE%&4qJgO%iAl;oI+egE`QB~t2RCyrMB?!a6<!$4NvZc$5vbT0AXd*l}
zFwlPSHp1@vr%v7c15efyop$#677+A6CS%F8yIjUWz5i<iZEpIMtXQ{C3O9z*nK{xj
zT^mEtr{%TmGtz<EkD|5JWqSdKv(|c=Rko(F<xdfx3YGenUHPdNm7U4PFarjsMK|-Z
zEQt?H)$<SBtnttCkyHj<3VmsgvhwFZ7r4C*>u>sVwE7H|c$6(U;U2jWD+3)BJ#Ryj
zpupMZl`qK?ZIa2$p)o6P*~Lme4UKnRhw*c{(BG-aqF;-dfU5GRcGn=fOWPKR2i<`7
z&;|@Q<23Fonb>|{mg7asu;V_LsVE8M$wK4&IgpKbbt_0*=Gz1$3?Cf0`Q{i59Lrpt
zALxTwf!OKU@LR?siz8KY$aN3%yXwtKe?D^2iS7Ebv$<^A8ZTiBLd4BzfloXp1(%`=
z(t{b@fHhF>yH%n}67%vQ`dor&7Xcv5S$tYn0cd`^sa3V!TMB&<I{OdFxa++=*i#ec
z)uF#rbfshj(6n|5H;D2(Kjr@6L98tQ=yO7O{hI9;quN*MozeIA^(FRo3b}GMgimUe
zhV6ixMK$?Q<469g14GvD(9_*bohD^wYSraoLiJ2C8@;}R#47}JgPZM_ncChn+Pt}j
zA^8-GCViRj=JFivN@nW%GM>bF7D!CJ!+6_q<F*HD-tD$7WHlSr_7>AbMGo58jQ;eB
zDxj2QFszMmijS~rIxMtkJSb;97q?#9J5V^du<rn~(~TD2X^h2lx-aExbCw@vHlKxR
zJ6BZ~D0ek#`=c;4N(KGSQN)t6?V4@g=hGIN*InGo_?{VdU$MPA>(P|~nSNqO#-I&4
zY-Q|YDOq^{#Uj0+Lmta#S+{4DG@N-*Ch<jFAb)HM<N=|x9D>}<hx4B%UF188FqweM
z!j;8QVz*#6<h1;bij~ltpT_b{uF{y|n&R;ua5x-K>PXeJ*#L3`Mm}OGX80y~b_!30
z;l*gOdElFgF<i49eZ#z^GIaXhnWNH**FRxjbU481YTxrti<O<9pNQa^8JssNfz!qm
zbqsw!ozp08pevR9L4pXW0e%uVMhp2+!f8N--zE_O#lE?|EC)&IL<wSS9OwyLkN7SY
z!Yi`v_Lq?_Y2)0RzJqz@s2^=a)ZFRdl&e(-8JvDhjUe4HUen$Un^fwrzslb`_#Gfn
za1?Z3dLDcsr26LULtcI_$n#35h81<kp@BB(PJsXpdbPf*X>~s#k2n_g&qaX(l08Y@
z;r-1Hh%P06OS$McnN2fS=bhPnXwh4$nKiAw;7MF$E>D;7CmpAW>aM;v5gl=2=U3;0
z3;JFCyw?O5I!zV|yoj5x`-m<EJZ7!kIu28ZO*%`B93qYn+YOD`bqiDt<(}a5pI?@%
z?Q59f&XCAu;xx3lEU+@YFQb-o;HHMO)Ix<!vU1s}W5R7h02Oq-L|HSu-q=m&L)y-L
z1&%cIHoKy>t|pJoQ%o|ha~^$y(i7fl--@((zk7X(xhp{JJc2G@(lEG&?dn3^?x&5x
zwxct(-7U;36k%U|sk@7azFs|9xQ2DNr(w?rTRgE&Q`}dmhm$ysI!<c#M+Z-N9-(~-
zMS))b=+mi`4k2`734$yE4iI_bp=!S;erXLxGNhTa$dN*&Ule1WFULX3ddLj!nmlCg
zy6l~iS3U#DJmv5kLVA%{LK)bo1CmyTCUiCyZk$n3*gG9>Hyz{UsqUPfse_n0D@wiU
zm9>}u98R`#S~qdtg!*2x%0#R?S;GY=6{p#O^#R#LaAcBhEiALZZee`q{*I$&fQz;f
z0@0&>!h4?)C)0L8E1@dr{*Jky$n8?roe*5_AJb%SFAwE4#kZwg6fJ%ZwUKRyvT6@l
z$Oe`fZP^)dR2_Q&94J$jw3if1O7c;y(G?9cyggqFKPFdH8%n)-;#l)v@qzynK=t1-
z187ETJf!Cty-c%FP};q6>-n_CVqc6@4gXMcxE|^qkzY@z<XvB{1AY33$U?)BOWPLX
z)6}HS@8~r`<_aabx*p!D;KJ&DBFyLcQtq#}Q!6#hro7`cScBoM48~?+ZV#l#p?hQF
z&}a^-RH?!D<}Sm_J$IOnw&?9lzueY&v7{e{#~$75Kz}c3+9IeCcnRuZ?|46R_L&;3
z;tdYCh!(z+WmB{4_O3GJF|bpL(qRg7@`RpAJ4NWr&!Y$Iuf1T)OdBm)jaxq5UN{k~
zxur7Leo~9k9b1W6`p%5$O2-G-0c&kNqn0IF+RcRGTf>FZi#(R2Uwy#-ORmrJKF4ed
zy5jn^i_(?eJ#G#fnI<}<`aq*!-tM?T{k#Yg8;vB{h%cNsYmiO_6c7dteJ{q|nYth4
z6>nB#dvjULSNh-vk6C{=0%WW##WS7G;{$ZE)E2Ha@*8J(zR740%L3$6PTX@e?*&@Z
z(8;I$#oq?yjM%G$Vo9B!MAqOZ@AT?h*wD89<LWs;CqvS_Hu(1k1JjTgGvP3N*G@;t
z6%`q#TG9bGE*sU^{GCKji;mt<vR#cl!^aoAo<#B7i%urN(fOluUjOIHtU?kxe&z|E
zt>u5G@4vIqxVpSY`o|Wxp*G9_C>{yH^mi<D^`BnmP?m;bh1K?pIOl(CUf&Bss04LU
z4{rY2m-5==NFB!-l#?<foeNNS-{CUZLbPU<iR!@0kZU$Jtuy~9uU0eC0MwR~$lgo^
z))&7ReoZA~%>z1FW}j;>I5t34%cE%!fp%KQqVesAy=gDYr{7x%<)8grl@14(_r;<v
zt`W!)_!7y?6qohtFWJZcbJp#)AJ7&llK0yGA{SgB3%XJ9EA2=5>F`?DurrJ&Id!Gv
zzHN3U6lBc3QN57o`Hn95pjg6?f&Sq|&C(ay&?Kq?@Gpp$veKCXZ_<$(gyhSekaw?G
zF&)1qdk^3&Ag@4O6~9kR#MKabT)V4KuE&kqR^I%llIdCkV15a&^tgY2uue1jCID&8
z3%KPNB=6_zG4CyXbH3XI7rx1({|KJw!@8z<nLc!9-X@rLe~vj1R}@b{LbclIXTjGN
zfr;9!ioyit>cH)U&=vWXKmB9l_B#+CLVPYe-v7<Fuym`V8xDE#DDC?(Hbqwy593w(
zf7-^tSwQwe4T_B9*K1Z%aT84$nx)!K$YJuc(La1E@0m)L%2;~w&M(sI|2>cY^#su?
zNHRa@%khgrdi<KIn|?ueaN?O@!i{cwD>Af6#xwNuj&a*>k^fW{z@IZNh4;&|car4S
zpZ=fzC;BE5j%k}IjDC|){)fMp{1cwMV?&CB<e!%PzrF&ux@-4bQTlh;;t&7lz5MV0
z8eIc~Le()YQj-6{Yxw&kta0!{f_|p^GySigk&X+V@f~hI$$x$Fe}AC=!%Ir`ffu4&
zllvg_?+5rF{Ph2N(XP-9uS5CL2WOL2Pe|g_f4so{+qF5t^cfWVNEiO(fD7E@pqeOe
zj1&}T*`M>T<M}^$_Fo@iExA`8%yG&uO6pdHIY<Sv8C-U<a|8l}{x|QObpJ_e!_M?i
z&=R@Sf?<L~X;{}y`@j22JZf;4(>X1D=l|dQ-n|KGIL=Fv^a`H;?Irz(yFlI@p$b(p
zN2UK$ZS`+|RY?beAnQUgqyB&R+y6cH`TxJ#zr95uj{SeUYHn6Xf$-o|oS=|$(i#Ox
z)yb~!I`_9b_PpthJ5u%2(sLv&)2?8j1F5-sUe4%oXu#RGxf$9e1vA12kaOUc$|2v6
zTeaj%o8H#ZqO;-BDzb>-0A6)4yZY#Y7=-(L=u#s@pfTPQmF2yiWw|yNk0S~8$5{U$
zrfcqz;fQ-BGr~s^w@ZXgGvV@sg8zX;b76waBYoU_+A<%<9G{-RUq{X@h-(p!bT6F6
zCkWN)(khN8<%pa;W&(v{yG-6u{9MewY^$R(aOHNa^&WJTG~f*H3P#7<#rL9C&S&7C
zvMvog5n8Z}v(J2uoy+MLQ|{WW=fb@p#t*JOT5i7e=Mj?+xL15&18ECRHhfZiPPfwP
z0_Ab;1Jn|NMy{YN5r)G~IUC+>72w_6Q3_>55&D$N2?42z7a2l`tDTz70Lju6zWz;2
zRTjS(wQYDyfFc^9J>rr)qp^uHVWEj^Pvyzq%*#bNExkd=-0=UNVQC2E?HQW4kzzXo
z5EN>S_ev)=eJ_YKGi_!0KdI*^ycNOhnaTo;Wi7|G9s^UD&n#4A4eGG`eaG%p(D4{6
z@o)zQ;X(N%U?SV&!a8ghW?I0q?OSI=)v+5Me2Tv0mu%u!9JppcN-o*&wA{!GJ=PM)
z)}7Q#*uBz?^k38w^ouo8Ckr~t;H~b;rJ13xSMMeugx>OvQOXifSj_>ualLXd)$u5)
z6P;L6Re9`kVXrZprgbAvRjd^QTO$I|57w%qJCrf!QwC|$Do?*t#wGe0eMlY(7FC~*
zJH;%!(zia>*VUV6^r3ruu+&ZM#J?hvaVfF_5iWb1YvUi}dVxj%3@9mn)tw@rE=Yp7
zXsmm}l6zIT4mu_#P`|zCe@YzgiWQ|K{#Xb8%a)4!Q&s5~i5GlVR~zJncgHn3>jrlr
zuTqv5$}RNU6Pn9wn|&{K;V0G4md#KhAh7R<JX>@6-QHE=ARQ2kdNF#Ks2^SagUeW6
zRLuRrQ73PEXe#9JbgEUkNF+yNMW_VZfw#P{k5HGZtcfXm0a~X%1ez+JV*4z2-wOdu
z<hB5hcna8Qlqpn>qWTGS&8p;m(7B#ofY#rRHTEfofF3^$ywi0dfMdfDRHP?ODRjue
zw=KFZFZ51eAkl5#!QKv4R77Hn*HsXrM4yKS-;BL7`T=#6B|}~E7v;(ljSCu_4NZE$
z!hSx}nvm%7AxGG?$EHmNbKPQ%H)z5REtC{g{n6&{w)JVw=$nc7AIik^oOKYmX>7>u
z6^g3YEnW?JFBy?+HTao7xJy(;xmoOFQb~NN>Qv78iB+P`{f?AliOIOb5rm#xgQR`i
zN<;is<KiePN6-dsIH-|7r{PK7dxZ8+t+yX`JNO*F5Swww*C`xA)zF@Y?rnOJg97=|
zjGp|_R4D~a&fw2f-x3&#P|z&2-3{e~A<XNs{p9ePg@937cmWW<9m9(kP;GCLDE8d^
z>*1-d%4d=~D<(>3TaMNA0Is;}-4v@s!s?2cOxBfbYhX;1Xx#B+*UhH;i*vy*M!hpd
zk$LE;y#0Y&5B;)KyxX6)AZh{g^37*9@1tY^Y<Kg9R%ry)&Nrdsq|QB;eo2s06J6)y
zw(5QYFuF_}WRqudl#g}-u7ULZNvtsIXPvR@$+0Hz#aB?a_`S0<>8Cn5>eW!fIxvi8
zM*G^Z{EW&~FOAPfMVcc8=R9b1zHLV`6r{0^E#QcQQMq8g{teUy7vRk-JpGVq9ixj8
zOnVNU@bXqx+v=dZiqsZ$Uf6JD2a{fTg}sjI53Dq+FFE+9&1Hi=Dr4jmQdwQAq9vbT
zW-_KM^3QNYfpp*wd(&BhM_i_hAQ9<Px{#bA+g(vmSY{iOA80nxzr`%`cWB-`JROJq
zAScw5)n8z$iCQ~CdSY|#Nwr1-M^%W@BMr+}b+?rV>If$@>!#w;RmCOjv*z^$>IiI^
z(RhV}xtGM)5>NLhIShg@v<F-d&GF<9VN7XrC{I1krK3bVA5XKsr#$d&9`2vL%k98>
z-rfpt9lWg)&q`$B*@@+Eh4b3oCBr6v!%RMl3t@jQ(26j;HNj-0;q;EyiG^@#JLzr%
z%W_26#;;FKVmYA5FST5gW8*s!^mKX9u<Pp7Fv$ow)@b!-WVlF6y6&pRSOra>04o;X
z-Dbg^crC?7`y{;zSJLe5!`m*<Nm?3IwNKvwk>=OL=CW_nD;B~#iz0PpHjRA2l#5a;
zcV>?ucaQD_MNzPcBKTyBG~w2<3&r<7`Ft&d>pTT0hg0c<m){gJUtZB0h<7fM>`;PC
z^QP|%c4(!92%`mpZw!&YO+a{V?J;bt&i$%+wDwfZEOaG)46NwhpgS}^K@QNCQp`Nb
zF;PCBu`(mW@3uKUl%Gz!&0YZQn5z-tK{>(`Tv~;kAMZ?*f0Qi)(dtZI5y|+<(r2Bs
z^>XfxGyYr6E_)wZafh+NdHwmUp@U-jVafATc{%nUF8ujANPwZB<lB@<W5KaERcP6X
ztn8HHqdI=2gkANHdcw;L*{Qf6lvkrlRo9WTO}#*uCUsuAf5>gy;$}YF?AzEtz+DrW
zq(2@n)Tmn>KB*}@<}fB+e4+MfVaoI;Fj)|JPNyul0syW4l_|JFBr|Y(TYLmIVWc@k
zT<6krzw3O(o&f`_kG2`W0>>2QtA_3>RIvWJ3-9%ix96*P9Lc1nB3@4naUZEh**Jj!
zbu#E!SG!Fe&;1CiffF_JwfWNHs4J8CxmFr8Fcj#ujAkpe?90gnVtqrIDzd_5*?+|{
zl~`AkIo_!u$K9|6Y<jbVGK_0@+J#1KO{Ir@^?`ezXX*vT{#FZ+3O!?26!l?cH2DG|
z!DOFOv4yTofp|%|=wXg7{)lD~nInt!w}~#B^(ckR{OOhA&RUvGo_a|a=Vgg#!u+)e
zF6EB|JDYFJ6<q1U09n~gB$=3AX~+x&3q@JBt9Rb~U2R@_T#@e@xTnh$$f0c_mI4IK
zz#`K%3S=oXAJqxz=<jI5Qr!Ht6ed@O{=8-nY$zx3f1JhbE2C6i{J7U>moc9(2+Y1&
z<yOd+)^SM$^DlxA#t@}qSyNr)wfQ0K2_DBHAQ|f|U-ACLZSX8hXLRApLL7fLck|s3
zR)XCdw}7ptKoM#1SK$X3#TQ_XcF}4+&lZ)u|Dwf9S|&}Is%usE&E{?wZf~Mc5T6`p
zwHC8`4FDx7QNiS3P<r0Y<p4OPN0p>j3<|{nnj&C6Snri};fT<2cWSECvfGzUz)Q*H
zANaUp0RZ3BIdh1K>t2=vfr&_fx6)os)F^oV2`tw?1PD1L3=Bc@_5Ik%`5xnr^qW+f
zVQ-W`579c4Wdt0maa1!VlO7*Pz+XwI$j7V?+@8QXr#2+rUdY=8S>Ya?yvHV@n99{=
z1RdZ8*k3Gfx8V*4*(Wzm<=rg#!29=EctP4F`b3gp)??|9vuVB<P)+}&r{*rarN8l@
zPQ3aYwv|1#ZUW7yQn>v1o5y4E!8Ge0-zA6DH1{`q=aUB)fm|j*EO7xz?L(R=X|MIH
zk-g(J#=BWLM~pGo<Jla~(G1@P<w|(P+UEgK%Bsj01eE*XhkKI<d)-qK>v{<i%V+<r
zI^C8|Cockl=h|1EjP!%K0oYDu65aLr0gS345|3u9qR9@atU(LDPH5jb2Dpxroyib_
z6TQ3q$cgi5IWc$3nHuIjRcns`@^%VlK+0-Ea+QLB+W8hEU+{KTZ@)cO#ykPemT)|@
z{XbY9?T9DAwT`1uO>88xK|viUH|4Z_rWe$;oobJI9N0*ikVLm-<lNI|NG0%P%+AMC
z{e=izRK2Gkh*0JNTOO#R7+R>e7*R3$8?Axwt*pV2TB&0Rdpd1w^V>T9n#^FD$=~k`
zp?Z~f2<>WS4K~1?NuwRktql!$h;@;<^L3#=fe!O^ur{!^h(tI+F=N$ZTieD7+?P)W
zzKdUo8yPmH{tS^t=pEi(w)V;2TUIeT*mnMcPaVdhkwp{y59T1;=WH%dg_+#spZb`o
zEaV6`Ya{u^arQPJXg&{VncVD5kfITu(>V2Ay_tNYr@z?RdQ!{)k$q~dyH1iF*mb#N
zLG@%ZRFQYR!q9U2omrNV>DUGSU6*+iI%{pJ?K$9B7()*zA7pSsj;E|SHa-y)ZW}m%
z9C^vvP_}U^p}Fkb+KfCFXppn(=R3W93;V64wxKB=o7<2-Yyb*K>zcdHzu`1f(t+F1
zt&?TaqI6z^)iG#rf$3+n(eB6UJ|;0>n`Z_S9915RFs@(9NOKIH7eG!rw=E<ej-lID
zEzN1FS&=y0v-Dx-z)9y)P$gIg<@S^;;Z}E|w?7{kkPGUv@atT5n@{QyVccvSEQlY<
zTx!(PRhl)HR0aA?YyFCJtigw(`-I{%Z)NQ5P=;Zx=|+uR)(;@1eJhw`)oIJ`SztZz
zUE>Zj_DnxbHo~abBIvEn_Sk|9FO0ILLHsSDvj!-|#VdFD2<i#!Q;6vJc2|UwLsZM*
z2)2aAvgD<y8(4P!fq02iQBl*S(|HA|bZGVoZCpM|kibH<6?VvdNl|<cUH(?mW%1~C
z(}j@9Uan@7D*~92LdPe%5dwC+G>b-=L4MZ-1Z}yhpM$VZj_51<%7lYou5F%RY0bg;
zar+{Z<9i}K$9O5TsX{*wtzeP_!)^LYTI<-F@nqu4LfCUH*ZAVPT^|?Wyt2K_|H3**
zO{$2n8#Dr7TIKnq5|b4JojTAoo`#4SoYC;)%JVs`niOUJNwc|Bp9S@3`BSrvGFuiQ
zLX>ezME^jL`KYH8LG~>h!_~*nl1d{{5wiV!?WJ`(I;wOZ*ac|p1XT)`uzP}c!3TF1
zxUDDSvc}$Ze&770PNZBw<vt6eT+9U;wxFYJic3hM5Zk%gf2kEbqG<IxMpEaw9n}!T
zTw8n1sSBN5JUEzRK(i^3vTfK;y3*t@{_VZ<`@#Cu$Z*m>59c*5x1f{5cHbDiHbUHb
zJfzG1Q1*C57-eV1u~x9Ye)7$q(H3H9?_uU}|K+Dr^wc!NNthz8YYk{^-OM0r7j{7<
z{8dNDU3Ms?yB<WN=XyI9n3+!y!$ve1a#9qNdVw2Y!UbGxmmB4fYW1jSnockjtiF_S
zJK7`inAF*U@2+0W5&9cXUkNTZaP#WdohDu~2Y;hAs!~L{hcJ~UvAlVRlzcXr1b5`v
zuT#*W675fdt8_nz_ybt6gG6X@-?;UeBHDyv8ic6128CNF83mf?dyL*R@BJ!xj5N3I
z{)E#a$HiH{`vOYwM9EJ@3#U$R+_V_J2d6XCMdvJ#rLy+7?zW@#p8&V_DgTepw>D)A
zE;he;pmP>+sX)vIYUkCPMavx5*E)%t%`US546HSCJ8=H++DB2|YgRJ%bl=(4mtWBz
zp7X2nivjMwO#KXLOd^bh#3>qWE{efkD5vHcxSMYk>r+X(3w=H0F4QZ(_*lKE8ksg1
z_5F)b+!$i=3KkZWL?1NpXtrr+TPB`tffc(q#;o0noCn_{Rl7OtPBohZZ1BqW4k}8{
zD|b_;%7JFlC5)1Z5-N+{K}Z@}l(vj4w7&v8W><_RaH5wh%j8OE`K(cdo}gn|jumh&
zzw10tIJ^joFaKf{LA`zZC<PZJ!meH8i?R+~v5fn?C2s`0igKSA7u0I&BQ8<uA0;sB
zXM-x@v_)FQ;B@r8`l5@I)BMtZgM3PYq~waz_FBy;tjl=4g4q$of~AWUwKf|*<PkhV
z$#J<@<v?34dtO6?lDcq&>q(|m4SNjiv4RIpUijnf{twjgvdG{JE4Exk9bzSpz>*@H
zR8(J-6Ves5cBQ_FC#5uVOQ3^|<(@(ncn0c0=+BPg#JfzKhmmqy>Hd=s_U+gZ6@Ns{
zGxvy;uM3)njAL#IoZlKKG)E~8Bpm37NNJK<H~9rSA*Uc(Yrhs_)&tKU5A#ean;DwR
zDwZ41lt}CQ3`pU+SBgSb%lV$IU*b=gFY8xjr|t9DEea{f?knN?qJV$YKy9YTNz!ai
z(D6wE{3I$cG{qJruw|bq8qzLGqaA2P>XfaWUhGu~_7Q?XR+VJ5wW{0=^D&LLYiG$$
za0h$j6^AtN8MW>8)7(k{KG~LOe|<@yNz8S01?j<fznB#)m#spkM@0DCK>b3vUNP50
zt4_TG2)1;u<4jRM+LB?Oxg~p^uceaQB0+0WtO<*6ngYWUzs%4`&bc!2bt{@b0$R(s
zLz!gYfp>3m3_&s6R?AA*eF6SGECbtG7p0nRpm9oeRGim@3_i7vlPzQTCBuenn3lq?
z#n&^e$6lQf;uSAGgSBd%7rF~SoK(!LE)LhP3)#MR2{xSFra_!xtCj@{k7PH(>(f2e
z`Paj(IFb{1yLH_IBl*7AxBb-ih!HlJub*1k`P`4u^%shAaJFJ#im!9v%q$vx8;H$1
zRQuW&G+2^RL13kk1m$a3pvjZ#wu0;VX1D5RjMhglqk%B&KtHG26|=qZ1?7$9t;+6~
z3TTI~GgZiTi@5K=5%J=Cl`G|6gm`RGCzIMvGSxZEh1Bw#Y*p#aj?9SL^pz#33z~^9
z!F?aFz$dd+R4JidTr}LXg4?%MV(sS0LK7Ws<<>fbn@G0)Xv=WeW8lAdxFkO!<GfcR
zY;6wM9ZSci=M-*QO<!XDvG|)}&b%WiLt2(A(YN&HbAq|5vX{@R<}YEFl)r6SFl+@C
z&tLhTqq$NluhZ<rprRBZ+n#=_jjg0K^A@wZZKlKox-80PkA6YuE62~1hZazAeMbCm
zI34OWqNq=z3~9>d<7$t1`QI+{29uN*-KKBGARV_n=(t#Ukh^t8=#`)5yT8=x_b!ob
zy-j=Ix#&tuwgWVYjBZy~Agjb{(UkKUDGWzl7;XZK99;Sj5@F?&MZrd`R2mdy7AWO-
zKWj)J5<_P0Q21I5cqN2*9TvR9Dt_s%v~l|Iy)NolmrsPyk8Agj!X?DdS+M}1lQ@gv
zjr6hdUP%n3<$D(MY$Yp{LL9j904Xb-!BfLDO-B&rXS4(5Y0w6O>?2;wTTbYZ7WLI0
zLa~{a0sybhlGM4P<o$tbI`OV22fDOx=gn-Z6-CVqChBiB=ktdM@k#Xi4bY^sbOo5h
z1~pe5E`K>5TWP{5<FZRY{k=uUJ=|<lrjD&A7|51urD?OQR~y%%rAFLIRt3gcd{*7u
zg=AkJLxev37sYuUG7oiCfGFUDRmNUIl)BPPDh&xgYMyN*e3$JWXLz&A&KLvB`f9qc
zl6rGBKuE!|df&Wj7%SusTA~go?ZJgV881Nqh2f0*A@XI9-5Cc9j?-H!cu~cT<jd?-
zU7|)y3mOMj7@}WcJS??bQ75~cTf1J1u~ZQym%rTelKOk^IWRhcVkSOIQ(J1Q#5QAd
zlRG{{m!jclm)co2@y`5jpDtBo9ih1y_S;nh`W-<^_wG|~CCf25CriiQOnm9~C0EyG
z7HiIag9a{DDaA($mn1bOIZZZm9=d*`P|hgLjHhuB3RslRnN7Saue0D^S12o{qfg5!
zVy`_Tmouw%s*aFaSf3f9IL3VS=*X_gN_q6m6|$Xx<Yt8}%EOD++G~ljf~)P}+Yv&&
zbAZxeopY~*Kvr#8Yb}`ZN4zz7k$C_6=40cikfTyDPT|~kQ^du&k<=w<L_Z#mV!f`l
zyeTvWwMTKLtF^{*Oa1jO9&@_K92fKp)#7x&ERt!m;bRoIe#9E7kbA$}R}~#%(L;E(
zn!9f&5BI|grN+GVK=To#o{|i5T=A>N&z%Pbl<{>$ccp}Ko2@9b7@7uBHWi7gRMuQy
zq1+I6o|c;o;xb{*QgH3lm3dLn$QssIqGy4rDm>*cG2oZ4*A-#fbVDSPI(`8g)p&SG
zv0DBlj2KWm!A*gA5<M_|*VCH5#-UUGc!I`7n}LQzQ20M1zGP;n(#!9pIUp$VkZnJ8
z^c2?ou=lPSwM`7UAo+&JNVlF}qeK<t0B6oj`A9i(e(^p1_4Ly5nwBZXj3htdFFDh9
zz@Ulix>jKsEuoHcKd&(gGSw!v%uZ73hi5$>EN7iL5NFhCW9y0v=yF!*68~ndUXuXN
zqsL@F_<{HPR5&Y@#D&}Q3*rpNMGtm)*f!-O`G3-ia!K9IBY&bHBxD^@x{z&`uR7#U
zB8`i!22tT>Z-Ye;_XQt^+0iRY993SJ(+ZDRO+E?C7prY5?LM;9-}Ucws#VS}&D^23
zLbgrn>?3<YB7cNa*4Sug8v6t~;l}qIYcy9D+eHP28BdoX>K}>x<pG|cbCt>wN`nFn
z)1J&P$KxEkRR?)llCb2lGh45<%jk<ZeWXUkxwx{n4Yx?l9tgU>x@DZs-|f_-Zo}|*
z*04#Do(DENi(q}yY%kun&mZgSeV`K&4hmmPiM@0Lb`Ro#Kue{@?}@EwypW@LQ9-d8
zgwYaNi6P%w4)VyPc~{JyIoR=3c5P0gDC<uz0EexJ^TRZSk>wUX#~kjT8&I`fF9H2?
z{xAJ=YSul-KvVnWkC$rLDUv_P+kTuMyAG~z90^=nRrRjVV_@@amY);HJEz}8%-(3%
zYb~;r?1CL)qB;Lio=FN`fOM8;Xs!$^E`ABu<0-n!*+}UQ7G})MD2L7sNEh3@TDawH
zO}^CO$j$=Jnr+DD=|CIHDY(+|&;|iob`PG8*?!8wHm>j1$u(E}qk!&#|C2jjbOxZZ
zBjLi{m4+Xx6GXLFA00&7(G_&5b-mVDqZEnR^dp?)FColc$=Z$2eD6P$<5R!%<>L3{
z$UF(5DJ3(B3x(k7Mcb8-*$ANjv+zjJR^K*=0M9*x^eU>q@eKUsXp)11G>?QYDzG`V
z4ExC!wA>Ypn&KYU^rYHmjrURNylJQPN!Fe{C7q!)bfG9UxGM<vph@gxq(sobXzK%e
zjHq2%%l~1Q7#y5PmAxJnQwldqdNV6$h1@CwX!7M&XWE$aVT#+cc1*I4>bunB+aZ86
zO;)|vNh+O>o7q=%xI@It(emu_iDyc$jsEqZ;tIM62R3FZla$j`_^5;_WDmKa9Y#{K
zmX#z<U^c_eU88gkEkKsZS6j3(>TP`vrY(I7UtRfe)^mQKWQu(Vpp-A8D=&bB!$+8@
z<P@mj$^_}Id~%k+7Z=E&9cs!?ryu{fTLr|>YlDvFF3n1Ick!M|L4&Cqho~v&vatBn
z6abBHCuA|^dLf&m^qlw-Y>s5_9>WV&E8u|Gs@{%Se@V$H&O$PYbcL2bd^6u7xhj2S
z^~*g}(laWA%qY9pp`mvcRDuz?TBv~`u%b;tvYvgrEBfhJ`wIfL&%5)YL&$rR^QJDH
zyPg6+q55<CLDP8l*hDROoPDI)P_jsuj^(P)d%(ygmFFxGR4;+rwxk|QD1H>UmUAp<
z!sC7`)76N!;ox!0F%8B;+EQL)DM8DN=GtcqO}aSQpZzW1k$<=U45!v~{rZ*#0pJwr
zDd~)8{q;kqB9AKj!N4L35h4AJX(0-h%R%`K+MP@Kw+5<ZQ~|JAQu?+j{)Ic`rT~Dx
zs((nJ`Pn+zca`*6ve*tv1V^Mf+6xMZYcQBkaR-XIqFPtg5n+BA^Re{;gde%C<?knA
z6R}k4_sB2KuA4FANxFmj-x>Mc{VK+;Q8(j*J-&#@**LXIFmF=40b8pdb|Oh3s;L!+
zM9N8>DbKLK-c9`p;0;BIq+c@#U&8aQxyrg<Xl>t;x%Dn|zAFpmuzmf#oNLt5NmF&^
zS)@TF$<jQ61NisO*hhTg?cV(Bq|WyvZa_jUUq=3xVzJ(gMVeGI^_FsB=E&g58VboM
z0Xs7~))dFbLR<#}unb;bX_O^0xU{|L0_w~kZ#PGxBDs#_L{ZuJ#(@;BQqC=6q+hHS
zg4L3nbE=x6Dm|bmT$FjHMopqCA0^7zf8=V|TBI%Q!aXNO6{0{s1Zu<3*4#;?EqBcJ
z2pe@MF@QDXL;2cr##yczy2G0sQ1_Hx7w?Uax29_dxSPqE9_Nx2+H`Eu+eF6c9Z%<>
z?@#`1E!*q8BM;y1`Qt+%zB10e?nY0u?^XuYN3ehMnoiAb+YB6uqo34#BpLhfiy64E
z`2#EQf(Zd7GIO(^CPQL6!T<Zl2N7mIEwMbUHrUB?3HWnZ<hsW}g?uR=u$^b)g8fsc
z$Is67Tw$QfP>ft}F{B$ER8E@^`M2eCa2@*k=9C_)hSguDkt0pwnc%mXN?!i2g10>0
z`b;~VVv>H1xdxVi0&!`ZGhjh=-C*FrQM!#`7?O|(INVRAm;k}?Q^Tt5Rp~l4X+3kX
zdy2QMKp8H}5I};vMScpYi1N>TyImo72=O*Gk6~AO^{}>A?rZ>-W@6duV#>CqxZ~rX
z_BzGKA_|)v#>Adw)k(^I3Hs2@@jbe7q)P;5m107wdj@XPEalv*Woaz_Vf#dURNrlF
z6+pWhp)^rV<|DQ2W?F`OQmI<9TfK5QPfPE3IR%4HL|=lfmdV@A($Y8E(-H%1551rh
z#WqDZ1UYzI+OJIIch6D&=>GOxDZ@GG<~Mje*VYiI)y`~>6xBH(0N)!jWN`$(7y1M~
z{8W-YQE!O#4$Dl1bzm7<jzND|l#XqXwbwS#tM-?pwXE(<2unr6ve0)2p8&k>OO=4y
z^lwX8Jn4h7h?=eKdINFR3a%CQL_H?$DCKs+2b)Uf=Ei{uI%N*Dn^b4;m6PCk_1)`B
zmf|3+*n{@gC*lTYhrn{LMfqReAOOstJf$`R%AN1vKo_`75`QoTu?9GZWl1)z@u8*t
z)O5RI0SH!p1M6_0K4uH=o(0nAEEeCo2ye_++ZLdUsafVI@N&yIeqk+k44ff0$^0lT
zS`t&z?ki8fawEZAtVC?nk_QoJTH;dXC8>3K@*jjWhoce$-b?m;{_VTn%>BDr$&$w8
z<1{>1f`($(q6A&^ao6Q_H37AGs(JHR^9AFXY>0L#fZ!&5Kz*wxGgS(;elPSf?APk4
zw&(T|cZZ0gRH~ykF`HZ|R(vxi!t>@OdH40NJ`Mc*K;Qm}f1ZTI)UD4~?E6QmytZAU
zs)iD4VAHD_g9FW;MQXT8+J}}25DCl9?7RBct|oEAUyXKH{K0R(Fj%Gt5@Dkz7cj7>
zf5CgC`HgY*F0>}Eow*Wq)5+%GX7KxfYt|w8PP8(1mTgFub$7y1D9;SSXkZUZ7U6mN
z*+YxDN=AemsgCx$0D0AK6qdzmE-FgdkBIcd6LKX%#)4S|4NT~JqJ?>zTG_c*`~lQ1
zGCvh3FbwZ5@IX}tVBy=$FMrWa{6F4HIJ8L5`&{k$df~{3j7R4z;}8iB%H`W@Zt~h%
z`c#&v(Q8K?&MMJ~IK+;%9@wB>wQz!J&VV_sduaa#b$!(Xq=f4|1g;7+U9E~mF7FtG
z2TQJ?x$34S?kTK~Sv-(Zl!#Dr7NL`*%bKM*AUT9oObV1JFx>!hgE0A~;Lc*z%<O46
zPV_;OTJcnqOaU*{qjy=dM(F%}8D0##O^-;1bKW9eRpw;I%rVm{sa$$=m<{-+fNWC?
z6X5!mWZ?B66oRo?3aEAB5#H?<800CYc2m{k+f&9$Ov3KkTOBrdQ~h`Lt1L4dYLfZ0
z8H2Y{ZFUviO)~hh=ydix)p`EBZuTifPrvz#LpK#p{;JjEW;o5TNKTXF)a=zM^M(5j
zL^E{6JL-L;&dO{RS(oYwz_>uw;0X8v-YCX5XLg+74YdQ%@d=pA5yF2!Si1_<V07xX
zArPB9?-${~7v<V2^fD)^-m@+MOXmcLP#Yx{fEjd7+vCrK=HA5{ZjvX4lXS}(dHdL>
z7{#f~@Hr~vEyxlHqXi0qcsM?bbo84B6U=X1zsv5K%_<2|`Xf+{v!IuT22Y{7@ZIsY
zmqXycqKC>ufeiuQHk;W2Qs?TK&q(t*S%?gJGDyyxz0NUE@&&{5cMRj!2$jUgFVVNk
zx4zIXzb-!!r?Jzq9PrY8#i=D*ul5ooBInL{KF~P>*nYUT1^c@X$g(A`J|buHG$^``
za9r~4EX4fS2{}C+vXT$+H6cg1UE)`a;!KS9OT($o<%)WrnIWigS^M1?eccXbqq8MX
zOnyY)m!#eSezY70;t?Z@&u>yM(wRwp;|flQJu(?QX@i@9-f}MCK>ToJxBqONFda}G
z_E^Zzo#q*b^)m}Ir)7C_(@woUq{W8jwzhITp-!=UN`$~P^6BBHh$U=O&0DxxQb9(-
zt@?Okb=LFI$_ZT|k&~p(BA!QnuAg3Iie{YIN_PU0Z$HS(ns79}gI!<pG@N%JD(7pw
z7ZL3ji#v$=tb4zy>;;%n{`-vxO*b69dFc(_pLT=x2D4H<nc5|ZD*^nDNbKqR)w`xr
za~N*C0Ip24;j8>M)5EHY{-^;g+B%qURcp;3URP@tm+TjJqq78kZ7pZ?kwzrRsIem3
z$ywd}Ro2S60fe~+^6&9}J)5k@7O&^8(jjqd+v?(*ZOc0)2`TZitT@J_SKmPlbBE%5
zn=?E%!!bp46texTdod2HczrE~t@GI8YfxAmk+kNFwYB7;M6kAL)|8@oP8&h7ePwU?
zafn@uWRU6RF^@^%DBaIUHQ`>jgnpyRHFIW1(t0o`GhSwiFKNP3Yxk&Z_?;VQa0GxR
z5>%2A3hIA=vECvn2JjSXJc6tbK#@VN-DWxE`xhhyxP{`qjJ@RaI4cDS?YTj#dJftN
ztUNBAs&h46r`_M2R_J+@Ic-YvO?AfA0eudX=dF-izhWNSUolUoJ^#M_K&wj?Qqs$+
zKNUezOc9CT9!N2@wZ}bO!2S5MU2+!N{OC0k*P&k&t%bH5+v^KaLP`Ov>YGqHXs6?a
zV~@2MY0Fg)FtCjZg&wdvMe)ifQ4XCRTKgB9G|=mw2l<A!%kJtP;}o1U-$t*OM{~k-
zRM^;!1NhB;h?uU#ut?WpTCAXhJD%<csM+M~8rT6hsG9>hWM&mqWP!P*z(d+guy(k5
z`zQE?Or|HTcyOMm!Xi8Xt4_t^F*>vQR21`J;DT=MCdM^a1ypMPUxIoib2kvEPxN!!
znDYV6!RltFZZf#ja@b-93S&i1uE!RI?1PYsTDk7bpR%28tD<xjeJh%Snbhpj$<23t
z=KZCktcu>L;Zmus%LvzKN;os&iE?s^=t1R^o>R;+M5srF=wTnkcD>Ws-!5bmVRUV-
zc#*z@m~Yz08{z98IlUZ$SXVx7;@4OQE7>C@KgZ77_1~~9gRcDD0&>cTgWB9cos(Ng
z`mr;mSR6i{AJxti&}2c&RO5S|o{(Xk`{L*5)AjIid%4~#QP!yqSM9a5z0FI1dQw#c
zMPa9Qfo~#`V3DnQg7VLkj|5SbbtRRyWnHZSw5?gk+sB~K(Tn;HGup>}{JzHZO+qc&
z_ZzN2vLgS<fEpx%I{l4PV5*hXPjGA~2QOi*BL3(p#)fpk8g_+u5>_hBBt%{iQ@6!d
za9~eW7B8cS4=qIDoM+~gXij)uA>^(Ma)%a7uK<Vc+_9Nj1W?GxuA$2&q}B3vIM(n4
zDm;et=qa;IXyE%Mq<M?|`1d9L*o}g6u>>rIcSU^3j=iO80+pA~K}NJNwGOX=`G;NQ
zhcC{yS4HTMiFv`dxzbT;zwCoS3e_2!HVsuO_b@Z8Q4K%j3{n_)+!^hZ)ynC5i}Q*c
zjG)=)E^7j&V(`!F=UNkfLX$5-NA8qEPx!z~-_AaW;))K2T~(#!4_mspqcCggh_W|@
zNhUNNae&1!l&PtPfC$S_gPiGCzJk2vy4_iqh1C*72SU3Gu|)x)y*6z@w3bP!?w^+L
zbbP-?t$HJ6`}rEAU25AHFtoQ?Vy42(X!QF(boTCZvFU*mP80ShL2-Sp2=U&l>q!KA
z2EH`U5TO>6Sg;SASp^AOr0HlmUB_KP#aDBy??unHJE(nRs4<q1HtMv+b1G>+eRZ+A
zHPYLnG{Ux*W$Fe0qYXM$>UW4lxrtx05Bxr;ZO3Y%gcnbR-aQ{7RhcCJbNbXR!R3@y
z5<vr&Xh{aCQiy{yyQf2mMJogkDI4*nUO7D7o5ZZVHiv?vUE*sSK<Bt-TWD#PfHbqd
zCU9(rvsz1BP$k-m5}7&%La}$kxD9-XeQmY!TXI2zTVdB&TyxZ%&isY89#WDd4Wb7n
zV`K((zkhSImF`+@ku%#Omqyn40o8K-jn6WS`tdoz#q}EHM~pa)w2&aA$~blTaC6bp
zW1~8=z%Es-psO7->1w$via`mbndh#k+xnJVNYk`0gN%Tdn297LUnxfxT!@Y}yd&k&
z(S+Co9eycIh@+9@Oj&Bd<2W`r2~y?(e_%S{{GAJXbY69pp~s!h8)3of6&gWE!NbOA
z9K50on{?Vu#M`a%Zi(I7Y3$#)nm+ZlbrRvSdl9$Ib@2Wo$9_^hbx==mEh-98Fs4J1
zT94eH1`15`_o$b>?Tx)a!Hd%2N@u8{aBY(s`H|ye8_PKp%jHBpe%$o1Tt?w5FSR)X
zuq~I;7k)kCq0vucg1{%nz{uGXCYSw6K4%mKzE5sc{gn10@5nCZxF5)$hg>KE3(g^m
z2yj7snhX0)Xm=~%k^53&)~z@lMj}KJHOWRb?%(VN3@OGwxq@|o<_m7Q1-U>Y6}*j&
z-K0sS+LGk_Ov~il=CL%-Tn5yF9pBraagXf;T5cI(!5K{G_|@$rS1g59E+9h9a6Nnx
za6!|l2J3{(Hncl;Ja<puR@t%<vSDbpJ&Z%aVES;!d?9)H^{^)q*E_krBNM(vPesRS
zP#;?-9)}k9wL40cDkH_>Gm6oJfGmvpCDfG|O8JLqSxLx(Odq#9XU>9wNsb9nvNkg(
zu}@tZPl6(ZuU0YoL-|>igjfyhPBEIw%lw@oeGUgGwHNU_^Wb`x-K;{fq}bH~QspPc
z(2Y+4taN)fwd6feAcU{f76zXaeDu+`9!Hh|gOPH>^XA|ppCPIKJa#2>_e$1#|IAjO
zDjw%HncEZNf-h$1nw>BBU0_Q=V88OITQg%iF!+Wq7$K13q2aj7KsB?Vz@f>Syb77k
z`u!AE-09UDO6{S4U120{bsW)CCP6*zqDhfS%8wNJP|M!*D=k5FEF-jc9n-CJ)|Q{f
z=)X9R^S}92y{=y(kjs0yHAg-;1*uCJhnC+58Yt-Bt|Ra}vY=cZD?~K8bg>T*t9W#1
z>j-uC1N5gt6?9ZFLG$}YhF0o1OVYo2-*-`UFt=}iESAtk)n^3b!5pYkw`j{k+Bf~<
z=J|}jE1{3IT;yM6IzEW<bPvuKVk>O}Uc94TTe6MNYP;f|`Q>(Pd^Z+F-LaHZ(BdeM
zIh1n03nMxpwo7`M=v@$*yk56?U|P(!^2(nKeE&j(|D8h3Uv{sf>vD_NLuHlev;+A{
zF$c4zKf@;3@VZN$DT+JY!tb!5BxQ^6W%M*NDXB7z-%NlTk~E9%^;r~0!L@JKd()R`
z4Z=roOY*m^ks%3NBMO9-`2m3fwnDG|E_Uz)u6t3T;u`sqer46aJjcg`N5V)k@)-Yg
zv=8Tt)c|!aPj3o}Blf@+d~eRkQd*%xBKFv06HL=KNDfYX`bEiz##1CeZ~+=VG&1R4
zU6?=mB%Lb>M9cGmtP21X7Ow0{uU{EB#oKFmnc?)wj4N+akrrAonKJ3AqtR4Gt!i#d
z*bughf=Ber7}5e}0A7-;Mt_gd!|SB$awKFV$K(O{;nK26QFBhv!3GZ{rR2dMqB37r
zps1OdTS<?7dq-J(W{f-{>$9v4yA)rYT=RX)E_V&XADi;mb#GP35mL$Z1x02)iOsZ$
zo$kd2r|{^a2TmTq(56$QO#iIPi?m+}PvCO;aEFFJvB}^usS`*yRx2GK(Kf5eRTcNk
zv_#)tg7D)xzFm54k!Sw$>(3E2b=wE%pAR>-(wDq(TF%1Bku#h@bDoYqi#fwa$d6j#
zIoW8<*j86qj_t(pL0Hn6MPEW;EHkGpwe2i3#F;DysULnEACf7b%Dda-v3K52Cu&og
zkl?RzmMc~C$Acy|!rwQ0`dw0T^>A>VqHT7^$A~sJr0#3Fu);VDW6qqoxR7MF#<pcj
zJV-ec3?>4Fj<m>!-YhMsFZ78=^8_XWXJK;*$MTx!g1(l;X0RDfRTjcxG9V!@9Xorb
zZZ_31x>Uz1LiRJ4jn*eP6;ptOCP{w^P(3dlP<w=-PsPSw+XaW0xV+<-=_g+0*fh-(
z<I3CNh8L-b1^W#ua$*X)1Ipb^{?i@$Gm&xY;Uj0C^aJuR&A4Nuj?X<Uy#cjRB4Veu
z=~r20nNy2*dq^<UE<1S4eQ)>sd;|wJL+Dj|{$mIT{+ZKvw#?KQqW_{UA(X*{h?0Y`
zR<tl7;}2^$C(-%7vb=oIJ%??xUMXw(HU8^ro@s6)?;7x<<EKoX0{LT-jHW@b;aLM0
z-fL|^C%g|b<9Ss}RPINU;%Y56$H@zlje^$@V};#4ty4cHb)JPes@+=ecw79es*(I^
z%{<H<isGfwGMM)$E#^WsD5rj;{K36`si$8gyEv<N7N{^QI3BtBJg&LEvS=XCnL>GG
z=`fTTP~E>Tzt~IF8kMj?3{*)#^%Vy@HlX`VI4!D@9@AqxNgDI#Lk7kAzal2NoVqv7
zPW&EOkm~AclHZ}_FMpo}O)8g2w4t)p9tfjMEkuMCS_BXvCEZ%TG?0ZJ;xv=z*IK_Z
zN!H$)7Bt~J%WB$y=6XDu>$>Cwg#38FPlAs_cb+X=o-d@FU0&ZT(zjgJst7L*-q%3B
z7?u|JR`aT0bN*KY&gfc%AsI<?dqijWBjrc#^{Z(v^FLOHHBPTQ%Y6CckL3H!*E02=
zT2nhtvUazXSh+>-g%_WBkHM~l$6<=<9$1rFPO|nlV8VyjKYT6z;f7EltWis$CLXQ*
z-shvvoR=dOU%ctEd(OJy1{0UyZG4X;ai&MozSTOg_2dCnTI0;Cl-AUBd&QE?PMK&*
zYp$Z5Ax}!rm1ft>j$79EW1eseH{B-kGe80Sy_<L2%fzTrxXNQwzj$pxjOmTe<*gF-
z#*3<5`3i$!*sP{q>)gn9xn3}T=h%7vk6Lf(DK<+iNx$1iDAEjvZWNydu8=kea}rJB
zf83uS)2`nV(Ep2Ec|6&H43d(v;(O)YuhPs^(WNbwn0tD_fp_Q15}l&q$=EIFoX12s
z8r3ZVEEZeK^_vG2>Nz`um)B8^T;gNM%yxZ<mwov}z22+^X)g%5@z|#GN5DUK@!4r7
zGpX1j?1n}(j?U!Fv(#gW*uXCI`Q*rZXsgT+N^1tWEr)b})pyEK$#$L+$5N$`&YiWb
zdhb>`h+Zn7Hl833)b$uO9O>9j5o%rCAEJ}WKY@4*&J6XO$_x^=E^}1+hPk!fof)KM
z%}^9NZ=Y&%4o1`z51!3Y9lQNkSow&Am6?RpZR6zbuXwi`TQ?=C(h)m$7t8dJ?w0#p
z2HI@rThy=0mT)myk*y+Q5Ue>ULrw1gICwTyepX|Edma52@s!8Sa)OP^8t^KvcTxBj
zXE8LQMGc`fTmZ0YH_k`JVhq}h`k6uWpv{kPVaz;O3U*u+dqc$CsA4Xe9hZ5+MTU<t
zu16*qw__}z%f;J4e)Y`l<XADT(cfo3HSwMhu~W=CarkN%SvW3fraW<&S=nb?;Ispv
zc|uuw1RX>(FmWaB*18_K(M<J233oEQa;0LwyrDBOs*KyM#59#ijQw`{MDbOT+p3c+
zB7fBR7auwL4Eu^x^(Lz2sv?a<;17!v6x0-=e;gBiEzkNu`@yG#_C1%OA)@(tV6v|B
z8AO#eFNaUr3lUZwBCVS)bm|bwx=MuzCB2)}%`HN@miKAJw%y%szS^5Rx9`#GB^Z?z
zY#Ty;K=9KdA5h4(R&p!IC|?0$!T7Nwzv3SMCH`KfJk8?|E=AY#e{nAgbi316S4Hlw
z)qF8moqw&K;6(G{8%d=4r9;p3e^kFdGTCw_-c%Hu*eG7|_B+d$W&utOA2%2FLvKvk
za?M8EG3;BqH$2r{kAR6yF;%&xX(##=(QL`!|6%VfgW}w_uF>GG9V|EmNpN=>5;Rx{
z2@>2xXe_w91$UPKfg}(-jk`<mpb1WJcfX6h&pG?l`Obacs{QBQA72%z>T1$-_w%eZ
z*PLUHF^8&lQBvigm}5RcyIq4IYQln6XH2K7xodH+dr2C`_OA8xL_DQqQ&<y-$u01;
zlSUNR!XkHoZefun9V<+P{hNOwHXDMJb0ybdx>d=VXZDfMg>ZNlPyVln+`%fqj66sD
zq-=bAmNXH(n;#XAqmUx*{Oh_8rg<jz`eL*F<LKoY+TdePf!1`VjgkTjUT<?nbKawg
z&{r9rPQUTxPq!(IIdeZY4LtdEH#4MbrF7U&GPcBHg+<B!GpW@2LpN9#%Pz%|E$ar`
zJJYTTQva)U74O%}x;y(|*-duQF3HbYh}a1)WwNGIy?5gc3RRiy1G5Joz>@CkD~bYn
z0{lt)QNsm0;+>i2A3qI*YlY1v6B6s<fB3Hbn*|>!3o?wHmbyBN*CH0nj=;jWg?lr(
z-LyRN)#C6p!-Rp-2t2SyuXyfXf9l^K5TRphcxRv_ZuR>)3KL%$Rhj@da_p^m8i6Zh
z&Z|~BlYqkzg{kkjmSOMydfAW)BH`I_T=CU^x&`|3BbVygf!h%rr!3r~;bY7YXXSKI
z#9vC6A-N1{uge7h7F%dqowwk5%@YgFh{_cFS0}Y+_@p~G!Z~Fi6%{c!_WtEKfdgg_
zDfg*2Fy!PaSkmo*Q!L=_c$w=u+h6irt=Won;1hc`3+yQW@go62z_?1aT?_u_6aV}T
zHYEaTP!|vc%{)6C1EGq;<G4`Yiu@iNqGu8#?LnR@^gEgU3UH53wx?!h^ZsyC{pUUS
z*A>u(kOtgbtR(4E-DR75VCq^B7<9;am**G)m}*BDx3*Lq5!QDJ6p%r{d56nm+fx59
zFy)TKtO<aC4fhZa!hhXDaMd2}St7~)AJ-x~3;E`#02=r)@%7`YRVc7lyK|3S=+nP>
z=Q1-hOKg^6nwT#O^gEjipx25C;;q3Q9k~@?1rOjLXxjqQSpnSw>VJLie}C`;9s8O0
zH6Vs#fXQIEpdj+r1WlMWh9n)r*@MKYmtpBU>`L=M0f(LiK13U6nNlX|xvl$D+ki;H
zvK?jpm$Rm?9){ff1dGrwroYzfpKry7Z%~*4n(*}XfDdB`L_8;KE`V-GuAgw5FVN6a
zwPR{YqyBylaC8^a=C1Pf>snCB&o-`sfox&W^KPvFT<Ujk>C5XYxn2T9n1_a-5?*kh
zyVR=&1#es3-1UwF=dmX``cELb0Ulyb=y#P_GPmioWFXN9n*Zz;@YiCfmy|$k63|>r
z`|GJcR}Qui0(^SP`}VpGNLN;gFG_z0XWIbDpOb=D$dw*&e%S*$v;e?8{iX5VxF*Wi
z)T=M_pKI%{<>!MwreZJzX7C)gxls^buLLDFgUMHvX8p+oiD07o2hY+>#=oxBt*ith
zlmX>Gzb&a5qI_%spi?pb1W4`zFzrJx8d%-^ZsTRI3EUMQc*AbXTQOO%?cW@`{O4=I
zl_Dia9@D2L-_<I+|JPEJWTNq%%Wk3#RQX#*qXvqE(akoBjZV_9|K96|z)BHVvyof)
zdq$<CabT;E$Cc>D{AcL-*H`=>ACNkL$c1x3{^IPv#-0E5$<Bu$taL70YUKSrpIX!m
zA7n58o1pvuM-=~SH~kM^z?VxNEL*PgX(EgNa>M`kKjHiT{p}xnym#3Jy&(W)|1FMN
zI?>;|5C{aH5}LswKyl}Z`ue)))x0wh5d{M`(GTPIdU^iV#Q{6U0THmt%+D4{nlvpr
zEs4(9>bA-kz`6W|&*wn#Z!LHUWYmmifca5V0fqQ$f33piT_87gasf&!OPp-czqbqU
z!F}ZHqOcDD4xPE<2IG(S3i(X__6O|byj#<IeHvqbb4@4Rt?5ha``mx)viou=-mU4Y
z10u)2_xctF5Ypxw$pio19Wua%fRKjW<RPv5x9+fk?OjN#-j$*KdpDD`?JlIXap<d(
z{0}?-|NU2_I(H#$r%tu|e~n)M^*{XI+y8%W|NqZ+zZ}E`$WNZcK3w=}<7s(O+6d}8
zY;<2xKg=xvR3FR6wNLvl$rYyg@PhNdF6#ey^5X8;^ctaj-UOUa*7rbX)*1U(+U!w&
z>0T?AfbJ<!PV@bdU>o1X|GVTWbzt*fr9*Q!!=3&-==OTo+Ylgh^a6yK=MoU$SwPYX
z1-t<nQVYwG-|?G~&A|JD$Qn@fOF_ph_?5cW$GS9!1oUujL}0>c`+;RR6%+_<fUYXo
zYxHUD-?PR4@>qh$w-M{AlzSiOQfv8=f6Lz9>H`=^BN<3J50&{MxiHS|hH>tTT$_Rq
z=M<v_WkS7cwCn*JT0Sa(tR7xSJz0|YjWzKOFVglqNSGr+`*N!BqrmhHQ;&_oL4RCf
z=AQuBv0Cym#b;Ew17KxnZZ79;gF+!pL7)frbEyiX;{jj-e6gLL{-2k0gUM%d1DuF&
zzsRnFqn8u3P58vH_nqz=*!vyoFfk#Tq8ku}U)%tq8i##aF_CXEhK}3%#|6M6Iz4!Q
z@_W$hgeT#Z?kE;;M^#1#jYAhfU>DX(AWNaTI52Ft4wUZ@@O3gd`&cW&BheE=0;iE|
znx{bvwhh3J_<F!(ycC#;2ZC{<OA&Dw|CU7mdu{Wt$GOoGb0}VSLX$RN<M=C?S!ORd
zc^hQV6-z%&;*?qeAU{pK4eT)PdmX0Do$)COACgq%J)v>gD(gCN{Bid+sPC915|egW
z7h?sj#M-chxr*Cl7a<^fYy;Qd5_)`6)%2_FG}5`g&AVW!72zt~B0ohZq=g53FWjc$
ztlM#5*t1svy?%6OCKwn+JAlq}*Ypr5HtZ8;G@*7|`(O(uqa!1G&4H)MSC6ItIEJ#l
z5J8vRLvY>)2{9|aZOmH{=Dp`n4Rf1}f239qD~QaCG;O^EI!DV|8K_h?1OfhAn)!8Q
zY0l`iFb84=`kjaRA2n4KllOj@egIRJM|sKineqae#p=2O&0el|$ExPLcqK4tw^W-{
z*C>>s^Tn80@msw5{ocEF8(NW;9`JoJ1e4eAB!%uvd5!<I&iQNQHZwmQa`v4A0`{#q
zm7?M<U9cDpfya>H=Z}d8bHMuRZ8C7P@aDJ(HrN3y;)Bt%;5W9Qw{|=if^tVc7#ft%
z1b9;m)9_h49pQ|j;S?^5fL0eV984>&#5%>lytCd_T2SZ;?6GI0pG@7lD~JHC(sdG4
zM-_VGH|}=~nNr~PDmG!$3RbgaFrW!O`U$$|H52t<^PcwH$GCd8zm|Ph^Nq>e>ni~q
z7lUlvyU8B9)pwfQ06>;5UMIVg!cLT-S7>NkeJcfZbj`<t{IH{^$in!OZwtS+0^FXt
zObSg|ST~~?m|?7jc|6=>>59E+)o46OC<R2aU?5no`T3pxYD1RFwf!;%dbNPp20Dy9
zt;TBNXK(Q;xsA(cT2(NvUQexs{AU6B*98S-W%C*xxTgBVB;y<SHHMxaNP@p%?s$Xb
z(Y>Z&)g9!Nk|aBKGxNoLeFE!%Jg}+*T7+>BY#ZY~Df6Y<@S90)1V1MFcIiL>=@OWR
z?1U98eL7}GVTlH#tY3W3^4l#~&aSM37z<tQJ$yc)nS&$)_dR%9i8=JEc{HdkRyjz3
z70wQlntql`b3krGdi7lvP)IsBs^|SYvBE^}iZI_VAb%-*X1nz(3gw`KL|-)gm9M_X
z(px+!LRz``Q&cR^f6t3ZXOKC^n+K@3`e~RD@ks-CTX4+)86lONBDAi2eLJL?20yz#
zT!a?qiOc<<1tv~()8Bh)&(T5fkwNt3V%W=Vl_Bk>CPuN)c?V0>GQ+rw*9IcLsvT`l
zR-}FBK>v2LRRAtizm^l0THLkx9fbPbMZqGbGRNmJ%~7j6z?h1ixSfu6hRti&ea?O}
zVAo?+q#}xXWDZ6o3bePjpx+e)unw`)DGOg}90z2EJFhyMb=jvNberSZxX;YIMrN+b
zT>oPcg-B>1Ptx6h=8c0L(o7M3YL{UHsw6+L)gbVy-YbjpEc)s_rg7({j8|&#E#VzG
z);A!I&+;CeRrCZ9)6M9n;(%#ME{~$8?5`kto$LFg>YoQOsP|Pk?k8Io*C2WeZg4L?
z`xQj$HS4;H19o&(R^N?3W$QCx?|OfSG{4I?`bRVSig2rC&x6cTm?)?TT<Xj+U4QSp
z5b9G{QA`VS{jv?Wg#@4N2zH)3*Dp~jP1uQUm+y*ZDpV|!t}ov4i3_gd_Exx*DwuTk
zx?Xc{hvp!>z48KI7EIzX9$^#>v9F|xpr(l0t$xJGcT4kAG_5*12%UO2fHRu${i3Qp
zC!z&UkbfyYZfi5Oh0k0b$s}=2H$OZj6m13B%4~6@Cq~9aV3{wIXlAA!?P^VCenKI2
z?S66jY7col#@<7!vi`iUSPBPJ>tRQ!PI+y~s6r}@bR`4@yivf^wP|Ix66z6U`r^Wc
zzvScKKBdA!08UqhD{dAAJvJAtUhTjHJ9lo+`*!49;J5Zp#CaGm)h6Buq2hFnUY}xh
z%a&?|(Vc?Yh1wEy*(_b}TZB7v6aTdRhGsxMhc;(i(|5XN)Wms{|C4_3wCy8`SJ@U*
zMC`b6UH)iA&r(A;fyt8m6*E(hrN`pP`4HYDv{gp<d!)!wDm16*;!juwz$~AtCQ*Xm
z{E==>pQfRrX-)=Pw=W8YQS+5^m4=k4KWu5|uzP|-GxehItPCtnw~W{0O~Y<t>j-^o
zSgfTIJITu7&akF%ashU#2PcDf(RX${3gLsw@Zf&lS^C1wmh7}Rkn1-z_0|tOuLUHK
zT8r=qb3A7%QyAk(u7&xictKIlpj=JQWVfbmTLR+<jEnyr;I3VY(u?Hwgi0hzR5@za
zb3`_bFn$;?z^)57W(|X*IMOZYVX;+4Ct;0=Ra#BgNEXRjBoRcdg&^4(Dw<S^OpA{d
zSaf;Fl~|1|Jmw)gOy@F|QFE_)CORT|1WF|X2SQRc#HKqVlsi-x<L-g3To-@P{ToF&
z!|ywzSvQq7mfhJ><6vK^cpX4U8V$wP0p4g==(Yb^is;ODZ&#z~;?>!BRpXDBWu2R6
zdH(yer;P$zW^_$z%0eAq*{n)iP}v90w#=bvjSg~6Jen22n=AmhEI{u7TK06?aKt7X
zk0o*{ry_!Pc!_+Z4{5Bk&f1v}pU5!^E&rLYe1T+-mIIp>7xTDxNx00BzJ#o;)?g|W
zd+H-MLQ_%41u*%IzzZY@?p?mhY^?_!{qI<2Zs?v5D&s##d$Bvb@Ow9eJ`!DlwM=S>
z3;VZSI4(bux!^{-SIy&NAwwf~*J&g{DMjk*HCfL(*)yXLB$l;`sWfk<0DHibT(O<Q
zLiW<v4X<Sd90}`4NM{|{t$=V&4ou-?=af?hG@B@P<J^)^5TTb-gB}uDoQEFMe6Trr
z?Z7kcXn9XjFo>D)14<Y+m@YV;O!FQo>kdKI_NBA>#Nn}kH5>VUH{F~@G>|NEZ@*p;
zp~I$_Hsa2<MoE)&j7h1DN}(0K4Ig17|L%nRrH>j?LHAg=Nz~s#)UXw~>YXcqhuyVz
z7lNrzI5|H_PBZE#x*YYz+}M<UtpqH>&OLxs@QdD@EzQp6>dK4a&)!|8)2s0k&Up^k
zBS4(OtkeR$a)c`82%1&hUALqpvb&jJho+oi+f9cQ{w8^p2V7oz;-sWG>Df4P({87A
zbGm}yK&%F5<h2*7y*x)e4;YG=j<-=xbOfkqKb*p5^+u)iJlw=&GH*b(O)aNZuYua{
zGPrEd+Cy93n)(x<;I6ANUbxryi2L@$!5Bkm)}9ZOYhtD3>$^Pjk0J1;D6YBPWemFj
zz?~3Osy3LkPJ7+UFulI4auW%?Ig`**f(9P7P`k!DKTF{Py_UVq)A|D1k)@x&oM^?6
z0T%%DiAf3Zd;-55;_O{PCV15lD}d4GW)@NZhP(J#ifbzd6Y>EM(&xS3LIHf6a&cqQ
zuAVhp-{`MTL13A80{SV}nkTiIZ$E!*1vvD0K%9f}gO(WeN0;IUF;Tp_Y7<e+dv)`E
zSoQU^gVzgq-g~t{?~=;j3qPDGmWMV?@S|YeGLwK+C#p!Ps24_eqJMia;fr8&0GZ$_
z>o!>gH*6;^lZ)cjf$Y)KtUq&IWB+@OsKz~&B?O;%P@}(wT}t}ZhIbI(=_KyZaoxSa
zJYny#iZIr4xjRIBM%;vBAxB?2E6@oo5-#Y8GyTBWaIsd~>_}_AJ!cR!0m}FIrxl^s
z>nd#^+ZX0sI?lRUW@;EDm}4>~>rV)Gnuw}Y5{auPoGEzPv6pNdxnIVcDS+-8U%n5H
zSS&!?3LUg*36O@e#^yFXn@cGBQ8pjMqLeo^l2vf~RQN*^I}@T7;;%8oW#ivtWF&6D
z_tFFV93+}p)E7g+h!#cKT}B$6H&YDJ=Qh2p7rP-UWJtz(BJ7iWhxYiJ!+_xb69@Wy
zfOf|ihR7H`7wOc+9uV#=BYQrWj=facPz@rMd|1|bzqzOw;m()+HHMv9SrD^K2tdq&
zGR`$#{@Ue7_5{qwQzE06_Y`d_Hg3Vb!~&WMr#@{(tgGbk+=?1N2M#Rp+_uzP;Aw76
zY~bbGX*ku(I<jWkGhQ?Xr}|MS6n^;8*Y6nmctP8`#&nV~Klz^7^Pm>GMfK%36r2`M
z+E(uhpB~A`VR@sG*d8766jqbPhK#`E9P_p6KQccF7KC=3_+bZid{55mh)?f(yijX$
z*Gi5D=35Cj*f&H`JwH2Z4hP6B2{fH6Nj!gA^SxQ&^-$3B;1;)bTAeQY5l4n61nuiZ
z{H+`}Yc!{QuRsC;$#6maW*35hKl%KR<lkX|8m6vkh(cm<z2n?;D9OyaEh8&Rz(l)I
zFq=>1J$E^~O8G|E1p6WP!V^J)@{Z6O7jNAx^uR>}!G#sik(gD_y-fV&=LJ07D&&C(
zhwGyS0vP@HIKyQ5l>Se3f}?{XA3h1B>k`SY=zoZ}O@*vd8HKKLf0yf%sOy0V9v&T<
zmp5>vv^BhMmA58W=U^nSft}YJ=YaimgbGOvsLyPhTRFE&j@~|Vxsx8SzcV$$7$&mo
z>LpQpqt7+}v@XEoY37HDS-fBa+`caCJQ^3eq5zD{b|iuxP)9syJj~cVvf{?<RJ;%N
zDpjVw#%l|P!A*1}i=^SsaAHC^@{rT=$QJey3|A_GY0$l)FiJ}~@+&1OqdI>Bn7}q<
zXLP%5>E%IJlG-t+Pa0{z+Co=IcH`xlXT8<$t1~E>EUB#sY<nY1#aHanDG@kF#to9b
zd|2{4(^hwy*3amPZ>5Ts?|iIFc{W;cp6(9)F0J5iT_T}gLSklV*&6BERm%H9jCzko
z<&x?JrZ*MiW0>Al(1)1jB=Y(s#~mP)akNSXRO-myFccppFdq^Om5H(L#yLoe;!njk
z8Te|Za+=PW|KNtGJ~qX*IE)n8v9+UiW!)@o_c84*8`&j|Jh|8ZnMZD7*?ffUTx=WP
zJMP97eLlIH7U#4rW_PVAHq(`Hfi{co?KL<wcUEX!1M8ogysDbu@3=dDsYS9q%&~oh
zlP;(h^zYv%yan>z@jhdhGC{wKh){pQ&zgMRNB)IZ|44Rp5GzmDM~6;~2r#$b5<4Ev
z4CkBB3Rpeu#9XT?^9grLwyzw01Bw5ZGy3RfCExnsA~e#8%`;W``iJsOWd7k%ljc}6
zG88tgv^J{zp+ru97xYf8H?;9cAQyp0<m=jtXG&)>b(^x&`A{jhkR_3)9noGKeq=vs
zmQ_cDJhi5m<K^Qu;xkw$jnhNR!^Qjlk?)X^8>Xjl2dfAX$U)ZNxVgXl1vA_t_mwVY
z!z+x+;u~G7A_FnU{KKY<G<p#)R>SElTr%yT#F?uR0@Al=WmKk2l8fz=S`>YEswxlI
z!yxxcFS=@za9-igibuCI5lNkvhvLD7{dv|oMR}fKg%6_YtJjl3ea1|001<h*fvs;8
zXgN4o*|OLE_;cn?Rp^KIqPB<Xa`2xvNIRj9G;w&|XKie&f(CK=TO!Gm%jsEjrb)Vl
zX{<{9k0WjyL9Gf&vsDJ>B)qqtz#oNiJalh&%6t5S9PAxOSx;~~BlUv<s~6v|OXbTu
z%eF{bt<_)E-YjXErfVJVaZXAZ&&<GHaFof%whPaPD)lmrOTMk0eV`Gl)hFNYCu3BI
zLTLGkFJl7l4}jfw79;rwGl%|HwbaS+dOeQU!TDc<83z>ld-aXf>j~5F<rGUt92y)W
zArJ`$I3l#nx44}#J#FByEx=NbBo)7KMMqS^MRESV_sL4dGtW$%XpEPH6+3asg6!{N
zPiFo(3xI3?{e%|0Axqy3auTX^#iRu{`^Z)%+1g$t$6zMLwP|Y9n$^zwje+r@f+A~H
z^7H(a)GuhTwC|mlyk$DEUw4%P;*!ysxIRQSRam!4xLE8ie@i%Mg(b$;v)$F$B%Pyh
z%)QcPmQ6v~v|h&0bg}-?NQbw6g5CmVj~<ux<Q}beI=euuD>f1)jUqbj8uzd(U<=#i
zkdM3>g_mgCOpv2*E|y4JJwSP}Wyt$Mz3;phcQX5>(UYWhsaSbE?E%((21fPcgSLPo
zwy2N5YF;b!D@;XEFg;nbbp63V+K=7<_Z|*LZY$wIyn}$yc|wRiBV)CNu!^wQQaDdN
zML7b`s-aP1zU9U<h*IXsS({W*rnQ{^l&I9}Umz<W)SgP~OO;KPm|S|ZD8!*(VaMmI
zZ{!~vsx`<?^dwF3_-O3AlG)cEZ4yvZ;IE^|+*BTUgJVjwLTfl6Jbk^)aUmWb3TOVZ
zPDmbd;2=%?z5ONIv4XB?e(0ze1tZMT?MCy4OJ^G)`7V<SRceQKsCW+}TU(WUeizFb
z4S#g1_;P5O==kBPMO^ap`gNTP*&qCYIO$Bwb{lPU=IU4VyXhM26A94wnO~RwOfhnk
z?s-yc>+n4!;4Q14a6Tdd-h{uNM_eXumzcE)J@B3}Lb5S;c$-Av+$p@KF}lM6wlyF9
zMZ!^CXKf+9)r?<Z<o2PLeK9@NtJZE8L{FHuM_jZ(`L#>u!;wR~SDAc-JCpzqAkAPC
zR)%rS%1PvN;&S;~Z*q3hS=o6S)m1!%m$r22DWz+CxiJ6*48%Bfs!V%?>Dmnv{Edws
z73^X?A1)DX2d;J-n>{KGvwB{RBmd`SgvDR>yTj5V0eltCj;Lu(`a)KkM+dRVLCT_=
zUT>a*|LA+IjFI~KND^P8If|E=SSyPMaxk`0^0%<|qF21ALem&m$>~NpE%&EzZqJR>
zZFTZg$3A~VGAiWpG>Y#Tw!SHcR4C!~pO*nTi96Gry3Fr%&!y+-YsN+upVZn-;98e*
zipQDB39$m!;{tzf!Dq<$p~*y~Bt6eNQ-4&pmJ5n^v#wRU46=52K4r@q<3ay`f1V;;
zka*}%x2J$s1oO;nOxneU(fb)k-z!sMl1*!2MHcej*rp3U3r1kQ5(u-9PCEAZ$x7(1
zjW{e3^YmWtJhD~E<oKgneyUwl&O|uUI)02)W(Av{SdnWEMQez}m$-WYDCq4S@WXER
z0IbJCHi<;`gY>mh$WN|<dSkNinLB&fD<ao+bAE{w67px76C--!Hy14vxk<k+wTjjC
zNoojbVQ)`FN3JmHJNkzHf|OSvjCQ?^N`FUsK8cc5LH|vmg6=(q>Hhb#U<0hL9$#df
zsfyE><msk4qPyXD#xx%2+&`n`e*9KD&YQIQ(NZt=k(4j>ZeB)rk9vKr&mdgQ3IlW7
z>tGNV6%M^)r=rnSB^mCLpP+V3Od&jzp0@G8)o}O0$9~@&RnDwZ{fv~rQEk8^@FKm>
zzM=9_a&-1Dlf-cQ86(Ae$<h<;MeCIz`<wmLuKUAWf8K&&B?D$~xA_FP`hK?IRwmtS
zl%mx}A!L_WB#t)VC_}%H2jly|#jp<0TIA<XZ|7Ry$j`J<;t^Af+%y5`43*ZE1siM(
zEA;r!ZEGUtC@|Aayo$!v<2fz!j0Ez>0SE9=)mFN3C+ndNRg9}y7j4r{g}G|Wy~c<{
z7i=o6KsjU?JYV0RIV3!b{PLDo`|81u%=-&v)INUIKo+R6qI@-{_t<lmQM?U$d#*C~
zegbvc6APi(n4IZ`^E^u4PCGQY@q1GR3WD}vSwo`M)Sj`~_Fxrbsr;&ex`js?*9p9f
z(rMDL<*~ZJ^)}jz+ex|*sN!wvaqL(M{#_YzHXaI-H6K??9UwRX+@+zmIftC@copMg
z$Lctow4yNTt`qfbir+U=@Mx=5&J6u!&XoR>r^y)nSP<!?^?&59@`(H{n=>EE#c;KW
z%{JpIoyTGO+eA~D1FTc<UqLGbJg$RI6)+7$qc4>LjX00Bv&^d2l=L0$WR2rn<MJxh
z#mDXm7oz0&@XhQ-(;0b4nTbl(mDwe2|7jW@|J-!dir*MQ#m=T9wRZWqE}D_s95PO$
zC)<n#vCwQ1j_rk08e<+yZIRv8rv9y)fr!zx)^L&cl1U4R;zGqXv*P4?uI)*;XgK=M
zN#SwU%Y2o~3dN!|s~lzR)p8>G1tUwfX&Z|!I3`R?Y3A`@44Mo74Bd%kp<JeAVTWhh
z<LQ@`GyQ%qo9m(4hG}~}RegMxgN?cpdV+-%JRhQ}dL|<YD*$_PJ)<-xawL+CO{o$h
zS7#TNcXWQ_49_sd!kYfR|EuhJ2Nd6{+ArWsFeLlN2<{mV!kweU;=Cr|5h>xK2CoAB
z=9A%ejAB@cyzpXFBj#?Rvo#V&O_agqjA^1)XH{QLMAm9I&k<H{d<rAQ3HQe|Hm0?@
zJy@^9h3YoNjBBYC=Zs*SUT*F$8zUL0a(WnrYc$)y6lOA8y&c#pFol7WW|&ePkUrl>
zG47|N2I*q-qd57hDK@uJ4xf5%05LQs)CKsu1;Ldva|f~`Z8Ubqhx0iVnj0hdjZvq*
zVw3<bZ7G-$=jxsOb#b(K;40=sWA0C`fD9VKbjr~q*l@HBROL1t3aN6}wgQ;{tP{s{
z&P2fGtUwF(2JAvA9HW%-@inBe${yCFAQooKn$EuqAkw9#(c3+i6SR@8%l(F=o5u^;
zjRkY9>P_vTja=L3-8pHciO2ruI2SsC7bSa3amGg^OYpV;|I^+{Wq8)iBb}orMHe!~
zt@Z0|<=y+YFmsx@@_6&<NST%*2WL0M9oLa0HhlZcJbCTCXm{NL|9+Ms_^w%)bS6oS
zv|gT?%YmNpMUZ2i9bxO3!cD2P?M)FW(}>tQbhFzgAWSrPsi9mKRvVSjP>@ua;i>FF
zygjUJ@mV=>z50kS>gc+`WA&GsdMnUUFjYu9%*q<Fb8O2|*9SXAUP)3#tXOYd#?7AT
zEz7CE5c-*LZ}!D(!}k^h=6YDtH{)f(GJov2B=8Oo%$2Vi{Yq$z^3rl@<nhehElZCu
zcClsp+RFah^MJBT<UHZHZp$V!CDUIy7#evVqwEqDj}-5(B`dn;-=ukK%bNIVyEjny
zka@5jI^tA>^cm?B_=irt-=V5ZPz)n`GW}6!^P5tPY1^*eMrkF~bOUdPXA{<#!D6i7
z>PT(;5&kqh%5i$fVO^HbGnh!*8}FIegQ~Ta>*XKv>&+7>#cuMDa5OdbSVp@%o&$Tm
zO9t)8W)Y68k4xZyjlcX7pK{Gw(%VV@UE*#@bfWtDLe7SV&Zh>wej}l|2D%A=avRIS
zHl^phQ#)^=HID-l1m}ws(0t=D{Gs5*2URg30+pdJw_CdP?17w3J)_z6nSO>)65sz?
z+Q1e?<KbTJuz8}lM|qfK`3|Vvuz`bw@j^RM0xywvRiw+ev=D{MwmkNn4r|8m$!oc^
zSV>Nin1uTe<C+UH-QLzII4b|yO?!wS<%wR1AN|cfe!^h@h(s3YqMRn);rgQH+~4r8
ze9FxK6S3*bv8%ASM?t34*oZtn(QcSg%3UX5(y43Tc8AH4=E-N!f*e9osUbOb87c}o
zH$ZHSbg~8wVfLefucI8k)Pwqi#KJTe=C+Jt$p-4w{#Fr$FzvOVc_vD|X$b4v?Caqk
zKxNf!QCD>nVXnhR_N88YIqf4|Jog0B>I9lPXcce@QdU19Bx7usp?roWx=1S~-ba}3
z0$o-jI{GgqQ#t!ae*GMIhx3F&${72?F*T2;FI?O<(iVb1fh$R|>xV)`4f;0PS1Qz;
zmbRJf<%mtSBT^SyuNu|(Ib(WgNL@q9<LjFyjlu%)JlCGK!0B+0Q0|W5Ve?Y9s6%#)
zhLD^-?4x$>KGt&-DY<P5i_xO!n&?5+8$R}xjQMW=3<{5aOoLLQVJ{00FG|yVWUr2+
z;94;WIaKL+SzYai+$KgD7?nBis*3qh=)@0_v~qERDc;E~NpsCxCdPQSW*qon#8(#b
zc;{XJaQ^cc2J?cr-fmz_Pnco+WLjR&9o;e8M>Xza?Kiel&TNhnj&4k`vLM;1KUAg;
z&ezMi1r4b_)QXqeUdu|hvTl|BOVu3PS{Hn-Vd|uZrs)REIvk6%jEiIy1xhNO;X_e3
z54T;XYd|-=qH1*})Y)wxX{VmNdSABg7RY39NijLG=5M-?1s3Kr6w?KV1Kmsm-j#MW
z-!tB+T&i;fop$rH-vUkR#_|p9GI_UyqFCz?f8-;{z5ubHZ{ZUD)#c66jr>^e@jveA
ztux(0rM_N-ImMt=)HUiHlU+P<WjLPI3YsjtyO1U_5#tZW>ga8g$h&Dyk9Voq@4DYJ
zfi(?zR>Yb>r}^ErInI38{3A!AKnZcmp%zJBWtPV-1C<uE_a;C60yN>a_L;=EPi3{F
z=Xkf}ELlwkjGyXV0X8rp^X8g}8Y)keY*Vu6`^he+CvWz7E(Sd)!zkpn>(NHq0bO<5
z!l&b1IOKXmqv?RRueD{+bH|>2VB0L6T|75!@UcY>IdGy4;%Ww^Rf(O@F`lVZL7<zh
zxen$D7#6e_7*Tq;!r{Gc99hpq9C9EID^wpmj0hgP3gqG#A&lIk$ECjtH&n&d0R-7Q
zqN;5SiQxx!$xrc4d5^AWuIkA<apdXylTs6$vv?SESAIYpS5%W#Wy&RzT%B8^p1*es
z=ES$&kXux!36e;4j^3#|d2N=jwd}e34L8gy;WM;4xXfY45LbLKCMpzq?e_aQXjgoZ
z?L)uXc&5>ik8#iphJQ`rZ^fRFQ4e628x9KMTEtz+raaTO2a|5CPhZz7X6ufA_j@Dy
z<%QWvMg3!*of+Z~Wzgyje~pO5w3l1v-yt9zS|r|AYmMxsNXRpsNG!_vp*Mh@2o++d
z*gfS`Kl`B|PT4pNm!GoKn(b^jrJbYo8B&z8o`W<9aQZ38vgpa%R~_zJJt}8jN4eCF
zd}P$Sg%|RxNs&U7i%v+QqQMCGs882kR@pneA=WY>h{F3Pad=Gjo`MW^-gIJ?mDooe
zbBPTA8RDTg3JcBX{o15PUbPTF|7U38VA%^RJjXBOWFCqz9`Oe5X6wI4ic7Q%1BvRt
zs_`-th`sk`xF$XuHY?i}YB-GO;Gq~a$wps5K`Nf|WiK|A2!)f3N7OoMk_{+RU{^Iq
z5fy2=<g)chk&j1{Rhk;0BVZIlQrcTyJH!m-&76_w2Cwi5j*LsaoC>3F;LQG>?NZ=t
zHp{Ytmss(pYLV1sC4<K<QSVJ0Ng^PL44_R^{1I?wdECW1FO4~8;VUOj-2S<W$wupu
zjrTl#Dv4Up(j^JcxP-#@iZhH~1ju8|db@E4I6>pou#3JakmNZsN|e3`(ytcoQ^dG>
zKR5fl3JtyI4gT4WANkphxw<}r)9@`T)>7jaVOMeHL}WpZzwxf({_HQ4QX=E~8vb~o
zde0M!^*cSP!PbLeQd~P?Ix*HwQxvbtp5oe#$0WF%l1&?53TvT?KaFuRq%{H5v16Q5
z+5G@B=eXMEz4T$YF+FR)@C+UQ7EpmswRxWFs*48D`U1@JQc5iDW2uQEmCV>^<+?=^
zH$gbcVLjZs#A}G7?S;V7m1x*LDXZ?Nb+)IE3(5kDC+hT6cm@6V6XekdQ;x;vWRcxi
zamPAc%5o0>8xg<zzj=ViQ2Ew+v6nscMyhL3hB!OX(G>YXi1PYGH2ZoTo!>|5PJ7p;
zJe!LSbi8h;>-x86H_g2?*RGjuS48#-N4^z)Tux$ix16}Gby|QLgj6SPrxL@G2Hy{M
z=&%pri1ea3e2Gp?uvcmCSlWNsl;pB}uJ)UB+ooMGq?I$Wg-$9!AwJC_m;n2C!NKo_
zHOm)?v?mhPq4*&kg?I9Ca-0vlQRpNe%w{KvC%}+!y&LvLcLP>pT2dq8UX>Aq70xc^
zG&1AwSPH`ZrcE015Ywz5J8l~!zg=^UabvhDlyk@rh$mWtxG*ki#-Ql#q-Y<nOjzG{
z1c!^zJlv`j<O}++#bo$Vw1vRyw=#jFYOz{lv@$wmJwkMF2Yc+71sd9j&i$i~3t*^I
z+4<x}uR-TRMRBp{vuU&X#3MNu`I$V~PG-ZtGg?QCrGz`@@(Jm&G`nH*1a2Pkf}`EX
zj+#dWir12!3Ym;qGwO@OVTmdul<HT)-@jqM5PK~%uOVCeC6<wyg-IoIYy0JNR-WvE
z=ypue(>J}zIEQLuRpGAm&ta>*m`ndma&jpl8DO!Ienc)ZB7b2{3bCXh5+8k_G-;+(
zWbnP=_ov**XGQ+8?(9^REywM`Df{L|n7i!{I6^~j2rVT9BF&p90NW>&Z>S>py2=qQ
zrZa8_6IO9i>QB;?tzQ@P(lX9oPY(S>ID^Sn7HMwe&mF4t22^0}E=Cy_?GJR7PEQYm
z9g!iBmyGb*&xKu+K5|BJ<DrhEGkP|UPufu#2sgVc8<d&;OwD^Coc4R6I5dv+F>cWp
z5yIW5Pp9=%YOHgr+J!aPCZ?o(CIZK6-U^Z6O+Kz_M=e?$d=1_FG+2R9j;r<#7`vO|
zk=(!$lG{<QA>?Ch^=8b<lgQ7E{MkY=qA;n=k)%T%<F7Lu-FAl|<jZ@TB0qNbFU~BG
zK8Iq|5>_w|zagKbiLEl9-sfA$ts3kV=MWe0nx0oM(uQ<o)nxInWk!o%E-2>3Oy1Kn
z^7o85GtV-HJRN}RTXd-hx|722A78E&d#A0So%%byWit2o4rNQ1hdudh_Zg?$jpF+Y
zxh2h6l95~-&WmrA!K1~+yfG&gNxSg^l^tORxUNy+J~)g_{DKLW=;X`Cf~jw~Pesv`
z)--GT^Kh&a=`gPqvp6@As8BCRmPhGisd%~G8O24gB@OMY>c$n!2vz)C8(d5o{-I^I
zNOCPz!A{;#awSjQ-5!>3O=gW31o9n^??%s!APa%21-mnY@2b)`D_NX3t<{o`No}Yn
zkt2CNY;g`t8N~^OYB_bUyfnY*_VoFKR|C!)K}5JMGZ%rRQBG7lGD|4iGn@DRT2K`j
zA#Yiai_Oy)6DAo}s_UL^+6I})a&+#(W%AbnH;`@33vcP|Uq=z}PYhgzmFrL24jL4_
z6W}3Qn$KWKS?xs@OpL8>FnLzjtCHo~F7sTiehFWNY};LgG49E`K()F&kUX6`nbw{<
zEQR1%886D#<%jba=FVA$jr9H;clfH6@eYcPWM?TbH|k*g8|{$KEP}(n^2~h8n9JD+
zkvdCTUjMsJL?|jDq$wXypi$#Hwb}F5B}=;Epe#lwxihQ9pPgKd$)x><DQt)q!kMSJ
ziqiviS;nT<pL)4DC`{X;S7j57%|7&xezb!A`X*3c6~RBv%%yN0iqZM{;(It&^f>H;
zg%OrB;jFWWIh`yX(xHEz^o1hh6WC89#BTvyVlxLdc18O7L6@HGE(v69^Hdwmy~Mv+
zl<tu<?5k|ZZ_xKrLN56;HYf0U%{mpB3m#P{mZnWw3*89L=3<=3VzF0euk16LJ%g#j
zb_8iH#*p8Jk`-Fia8P|pm3`!HG|g}LQbD&xGe4Bs^M-iVb=&ugsad$fNyRft;_Jt1
z-qALpOG%AKA6?vcn1m6vDbn28J#C{mRA<@jczo@JLjvat^9zMX0@$AIaUD6(eBX<m
z_!vGaEqjyG@XU2Of<=4W<a+C4|CX%lD?X|zm-VVUY8QR;voc!8m%Q7;9@qG1s$868
zbIK+%(!P5iikeH%^)j0W6|ZWUBQfAkJD}^cC(o-kT2)sxZf};*PL2TPbL{r&GhVnP
znN15-Ed(<uChixf;8(t4(>*T?`;+@n%QUndATCZfpTA^T7=}sRoXxo++zOj{Te(uJ
zsWF+j13D5CvavlOUKm#!cUvFi7sjRjEqKI`yg(%_?<E^XlLJ(_#dAF<S%6^+lD}Lf
zM|q@EgckZ<ZR$ns4Cm2EOvK=CJ=94#5ygPXX9w--3`~koh=jCF*AVehTZKbmq~VVU
z5pk=U2jNfOaT%+g&EuiXN;e^UE$b7<Bp)|NB6VPg(6RO?^G<#-i9*AAJ004BUloJa
z|3GfE(9|amg<VnDjER1fB=r>n9*gKd3L4i24|AGv)_4Y1^+)gW>}O=?wBCFOe8&;$
zZ{|<4N6*1sN`dxa6ydk-loPA?o5TtI_uF1lDQ3Ryht3yXBUFft+KxMMu@(z<8K#WK
zBo;xX>W%?qs;qG+X1X5>59&8~E?sXy<`EM1HB%YJvh(weYNlG}Q07`vX)2mCAsHNU
z_TyK6_LYETPpDRsoJ+CRxvd>)>n@#!`3<vMP~@K~)Nocr(0%_LWq3DwFA$-i#5}R5
zUN@fTOts1ucOg+L%X%|dYhxcy?x@ZrK#1bS>SCOpC63kqo&xeC#I~Q#sE@(%i~u}8
zWH}*_&k<BwL2ihE@DWdGhLr1xmTX<19eL-LMb*(_|GT^}n~)@}R2U&=!hT7(1d_Kg
zb(oWZ>Ll=SbI<bc3CfMU@+Cn3okAKHiu(c6(}1{lWH4la6^qeKp8W=Psb$K_=%zCs
z^_X>ury>h%)`q5yT|HzkGuQ7O?yQrK-qmYJnTA}I2Z@)T0*Ll}3jSV^SvIH-=kzjU
zpPfbBFdUDdL@PH!kJpp#rTJCN8BU<_R&pboDdpirgJ5SA>BMqWIl6hG9Hc@~eNFd=
zXzwCV?}ZbVuZ44S>czt#7SvCX+&mWv*F`UarJH)ksKf0DWr_K#?86<W)%&;E_wD^(
zMx8v~BQJKl?-R~Jm&~9)e^$vh*1-Nz>^|j^V|`W8kU-|3S0bBb`ROT0@N?_GJcxqT
zA8CV*{}BNxp)Yj=@#7)QP%7ZS7%{i+a%Ix-0L5wAj~ZvqB~za1!3Y>G$HdY-LVTqe
z_%t}-m73z4&S(A;p%2oD)D``B_NBY1p`#gG54ejy(dh8H2<r1<P%gdy_JHSnMsv8$
z?!il5QLT?{yQ0BO0T@_<4)-jWO~}<<PhW|ll0SHtf$9j*81vGQO#-3jAlrOOBqmK4
zqu3;(T<Q9e)3v%kSpe>cWBTiEgNm1Rf?`w(6S`Vi_$yr1CfX^rL{l&^sTwP7<ZAS2
zZgHC*B#W4+yqepIjnMIrU@A)GL=`p_etxa)j|<VD;8IyL>=BP0`z*_=$gY_m+HjPx
zl34QuA<%2a%9Lz~u*KZ(1~XNLNy6$y{hdQiX;A511%|ozfCI`R9XG0QX{X#JLkU}q
zl1lq8c>4Y(0geo@2YiX3Cf1DfXO)$<?)?rLBFYn|NGCExf&dG+r)=1KxVx+PHe^~w
ztIw7TxAIc1N+->F)y}&fRpCY6u+w?Krn_Ap?aWR`TZ&mng&Y=qJFqI)g~XMm)!o%>
zRt!4`Uj_9cezw`1h=*JS{9{tpLV4m4I){<LbUW(+4P~C|yG$pZ#LZ;nK5|7c>Gw<U
z-6*VgSg7llhW@sZEqE?1{&N;)uSTV)lJp`wLIr1J`sNIUsm^3;WE<Puo@E}(Ew(p9
z?JB>Dq%M-Swei6l!tkmOtPdzAY}v27BvkZj5^(5mI$WOXbCN~~;$lUhA2ahXb0nwJ
z?@o`i)_5WydyP@YGqHvkedJ#E;gxZFa`ioc`1(^luc}5$sBQF7M5)lsDb3C@-=fUa
zOi#=XzVW`-rwQi4!!ySN^XdV}xI-%G#>u^Xv-0=v#Mzt*!VM6^BsC9EJT6stUcd>w
ziwGu(7jDEZZ;Uk2{jJD;z0jEWkuO((z$(;!BW_F;uKoO`%HDA$%4u*&=pUHL(|dL#
zRt@#iI_sLHpWX|zv+U)1%FYCx2&)%NUn^Bf&z|{8mn&O8#JpM?Y<H_syLz)Uf(CG8
zClFC)mNaiRmP^eBNTqANUpnAjKbzIB@@p=F9vN3tBgmKF8m3vLY<j-eyJn^znkQ4;
zi>+ej3!*(KTiiX<lD)}}kf%}#vrW<PW$;X3E;3CIxl!eh>MS*Zx)Tt+JkG`J+p1XF
z6<#E`-dL$9YK{^|Hh$5sZ+Z!tS{=PzrQ|`K7?Ad*ws<I_H@Va3l}^5MO0e`&y{I*r
zz}Q1N(z#>O9!e7TF-xf?oKMo}EAtX_-#SM{cY@H%jr*OeWuo0&PAebMncuxSWKN{h
z0wyFs22{J$f=s*cH<`g$DeZ5#ZFAhmXiI4AF=f6^k-BUn(|%49xX;9<3!81KH&m0k
zZzZTA4^>#3!<X9ErDM0K+8o2hoR)eM`Lv+!k9aOs^<G8D^6Oo&A<}VPpdabZ2OIbV
zVMT0CHs@D7@HU59(5MaPqj)8kMos`LTII%Pk8cJpO%5JwWAN;$JPnt`y2_UvSiawD
zbN}3qF#Q=abgGrcDZl0hs5nk+hb5Oq<A|q%2p~$lLxL(FsMsb{&ztYr|9mN<q38PX
zR$04>#$uS$)RJY#e1r-Yn1~gtU^;9Vyi`gpI?<j`w}#sX`1Uxr2A4(!Q6~z&c}S;c
zvVyHc4S3)s*(t~-Mg?hBv(sW4tAC#VH&6srl;XyVw;aKb!U#+_YrZ?*Xf8+vf9p2X
zO{|SBPjkZ{KwjZzi6--D#ry$A@?dJ@zSu3pVR2+MD%AX~mQ)@~6hmwZ1;ZQTe0^$l
z6E4zAX{c%eTY<cP5%av8(+kYMI}R*t`HYux&xe6Hd=D=%D*1Vk3aSUb<O_|~TwF^v
zR|*ULp`ZmaO@>JTeTwOft=40yM?oU%B_CWp&5l!6X{W>r*BKOvU_FC%(NprH!nMAO
zTZ}xcLUme-5Xe_q^^(=ZWI>g|m*w>7IFaM>9JV$Z@pJB=i>zFFuviM1ooCkpe7K*w
zHs1u>raNHdI*@bdsU1VGtk7pm@84r#)!@>U%}fWDQ`$oYwuEracRLz4&4lFV<m!qK
zqlq!F(&sT1;E(BwDu)kP<!r^y(_4_I8_IrBEIrj*^+HyiYJm#Ib{UbD!n$b~cjGP%
zpSqKBjSPtFZi#As>&uC*G_3*P*+yOXY1{kLK%Y<tdC_^f3p)0C`w>bkH%;bvC+tOI
zjE5)G-6WO!qmZEZ0-ZK>sI)1g$GUyqA>G_UR5rW2sLCWU)BUp9u5pfdk}>k4`>Rux
zt6I@*Os1NdPv?U*Ner2_yQ9U|3ZLuIps$d6qo*F|fZSy!l>qHRF3+O1|58FK@YsG<
zzAn^<RJ>Pk*5>LdYsU6Whb7v+ef10yXNq->Pqj4Pu1~rD0g)=}y6=xk#V@D1KN-Hp
z-H>^ht5VEWtlt9So+Y{tlXCjn^^u?p#hk?te(577uaO-?zaTim3Wd6aNR9BJY=QZD
z{U3)`^E{+km=`NwSZIUyLRV}5IqcHVH^|Ao4tjyYI%~Es#U!gu&=fS2km4f!T=r_j
zYA<K6A`W5u39_YyMlkh$#@<v<2a0nB@%$3rwPI_yL>>vrWr4H^@6n<;$7wvln|bg1
z^+!ewVvS*6&i&z3*Mt<tw#(WVgZ(HSlqp7n>s1a*y?NVox4jnxZIf7iwXM&WRZ6<n
zNuoK{nw*!C2XoU6R9^^0x^^j#H{>)&mBlPLnT|>|wQ7TS^YiWOlU|22d~Mv1&3OLo
zrWB?~o-aCO$qr4@6Rolz9g+^3izk*v064JWw`zDd49wx9^#3l%aLy>wsiBdQLgMqq
zk<=^kGn91K@M;mbgAHIsQs>%4X04Ib+WEh@%U3oTbbll;YBKk`5G*YT5{n<C)l2Gy
zt}vM*EKXbNPL(rm+HjbfL3;}UKcQ2h36$+J)V?Vpv9D%5?fLD9V9%kNE1DZ*{905+
zEnBCW^baSIO+R0L%_ZswkUG_7Hd<OcCO;XKq6rCXgm$tWo>dj}s#M0KG|0OYwVJvj
zbl4y+qUC{R3nLBEzMRO_yjz-qB?UhVrElr$r=RS?Iryi@*=yHwh*8xFlyMmG-{@E(
z;kKYE`hEUE-gRc=7iJLW-tyq9QrG5%q>EJ!don|`K}5|Ho0`bIsrEp9V~Cq!Z~UWy
zAB>Ko9*j=taN#ks<X7PaGeszy%fuI-A%C()enCnlR}JXI+RtkQrm|m_woFN!r>z$8
zY&zzKs`_aOV=}(ka~ohKbEP!(+cxm0Z$UP4t-9pW{<U(9;y9iY#{K{{FO*Cpp4{Eh
zW|rvgXEnGpk$R6=v#VyPL{ItU=Rw!ZP&_{jHB7Qj&M6c5QY#Y7P0mF{5r3zrCi0YW
zl-6-<EvPttj?Wq*^NIx!RE}!LRc0SKvRM&BCf#*merP2nR$(6uSm>@Pp9ez<D(ScQ
zvqxf3PHjtY;@#Y7w`EW2LaqET80#rZf!`7;Bo=iu+nmGl+=_10V0_gNqwb1gt9lSF
z>)6zf`J(jIj~2!(ztHF8r=z%Kl0Ka_jgtnJrJ*sMTo7}Vis$1Y=Cu(M)t?qg4jx+k
zDk4FKOrenDAK1oGp^I2CR^$-C=splcdM)GSNNOiJv27{6Ay&*LhUL&E4-Kab0Qa>!
z4|!}f3R|CG9*zz88~S8sw#hF>9ONNUI38#aB9t8gJ1x%rPX&kdm*0ics|YLx<20)0
zPzA(no~u7cjZsm&w^lV;442F?@Q=dTbyI@i;63&#Ly^%_kIsCg;o4Q%V<<l<pLPW4
zHqYyGxLKrFPpFpd@?8Fq?6uzpKfq>$g(n5TQs*fwg1Ga^&1ZPPoAT?v9Um6ERu!L|
zFIy#9l+k*2-crXk{PEOTFl&`r4ZSKf0cm#(c@_csNyH`Vh#s&K&;~E1&l`%lM^O>{
zzdpgs<GPUHA0qkWJTEUyhs*Cn)nBWl_wqp(dYu-l1H4`YlItooU2V|Zg#E<bN7M=_
zWdYQ~=wsgXc+<$0(sH+zD2lf0=h-{&oDY_JTBK5yY<k)qTBJUCg_ofnkSv5oR?26*
z!9rM~9Ykx0B;Xb8@%W%(r42t<ox%J=l$XNt+Iu_!`r31x;2#N=OtaKIUpUE9_cnzi
z+A)jq&XHqIE^dxjK2Fc1vp{5{uaa3+L%dspc&b*uwYzSU%1AI6c<b3<!wn=jm7eKr
zZ!q{>x0i0$;cl}^l{QM6WH8n6DvKq)KN7GbBCCbhN5ZG0tXQsdenME<0dek1qHo40
ze?fms!0cX#{`|pY#W2fsEpp+(n76*s=%phi{QYEMMl9HC5j5wPcHy4@zaR?nkbV7z
z)C+ucQUT-i7S<<d7X!;$o$4-K;n$Ok#kN2+gfbEpx`t>>pP`ddsk^M91^rT5#DXn-
zbL953ozB+_;n3-5Us&jO8F2;I%T8u6$nhOKdHC`Mp8Rd^#Y_+Vq*1j2VuC@UBvuLe
zd&nQbQ_4F=+>L@`=ZK=yicCMx<MF%jvgkpwsP!u2bl;0{<a%A`d%M?*l7cB}rm)a`
zp^Lt7a&8$}T@}%a!<K#8SBIZM(LMVM)F>sjOzF6&yA&sq91~Kw#McM^0Bm8`lH^#7
zq9TJ_s9hdV<L_o6%)7CjWuV)WLSnGvYsb~OCS3D){Yx=#OFiL=ABMR<TRu8G<l!XU
zx$U$?0SeE-@{(Akjf;7uq?K%z^6<f-X<uqlK?CShGnahnq$lUN2C6kuPvi^sJtQ&|
zd$)rx<l*<H6=@ceAM#kb|N1#z-T88sEy87HwwW9d@E#_cN@jg}?KEZT|F!x6L->K3
znQQYkr-Ik)-XdBCU%R_S7tMRE7<uZ0T}5H8o_HsRgXEWe+}VfeUu_9`Q%s|ew)KWn
zO#SkW9dBv8UM9x4#@w3KJWM7chAR2ThiWcL8>jV9E|N{DJ>vPy_U8bd=XeNGuhAPT
z{Nsn6rA!s`o!^Z)pFK^9c$0?9bxce55cL5#a<Li53FK&S4Tr*8FdOZ|-H2riYUDUa
z*2%U9PG4s&G5Il7$Y^(SiYFl?3)+%It$4mLpLwrn-qTEjjo->z)=t=>en5noeQ@=~
zE&RyWhgOW#n0OM=yIsUH-P|tRVdy&@wo4_%E5D6u#uHs)VdyLUmDvIt=TI|^kX<M%
zS#wuYk@2Gc=4d1eof%H4-5O|>AH(0=@64ZZi3odp(Rz={j)rYW{o&8bg?_|TtoCHi
z_w|<vP@|XVSpL&yKYGy;Q)uXv=;{sc)x|A&()`ZBQO?YVNhq4C&x`71>;wz2;`{B3
zr$`v(#6Yz}*N~FU9KSb6CQD!R<`3(QNEO&72;qs0aM~Hf23-v_v)Iv!aes&xuzU9N
zk@<*~ce_eE7n_f9ITM>iuL2~%*LTjE;)zIac4Zjkef;TqD8uApEdg}_ikC&x=dfOO
zi7&bRQa_#dJ4oEK0=!6eeRw0}GLC*IiOGn{&7mu#I7luNktBUw68tph>g_*7W$=8w
z$kt8kX~vVeXa`08%t}XpwpjX%G2NZ~nFsrnF^!|=T)(PT9~@8}3tN4xEF8&~yM#2w
zi4zPh;tx8-buBhHE=R2;iT{H|!-#OPJZ8x1P=KetytrJXwPy*MWM&GNU;?WO4*KHe
zJ9^dmt{jel)3K8E8$&8cfFrA|QQ7^tFT+0;WMNSipE6Hh9O<o#z^Z;Z%|b)4=8Oan
z#Mc!#BcGBB4wx_NCOKmh+`E1{{b;Y}g!W*kM?@wq9<Eoly(<cR(}a`ITvB{Rfb!w!
zTobv^ov-iI*ZF%ZbQT9ofxY4AMb^YdzSv~gCz`&BBI+^<Npx*>))Y^tD_wr<{l;4H
zKFlk*ws7rORIT<j$~d{;Fwx#*^spJ1CTV5FCUtJpj>p&d{gpvqF6luNE{~;Cl@wZ@
zG1v-p=Si2G#E2DgU-c28Qu-RYjtNOB(m8g&d)ApogBgCVi$1^0vd-^<LJ91;47hi>
zXLQxe%+~Xn3LEFcwSA)qI#!oGabJiD$0Ifgk0P%}{y&_(2RIe{|Nmc7h%!!Avg=Uc
z7=<|Y&aQ~;o$S5I$_!<PLlh-@9HV2;%usg6-s9jH$L#+;clZ7I+`r%N`~Ur~>wjG?
zsT>aHJzneic(z@SzRj1bP%lL4-Q+Gq%wSs*NE!q#wYFNjA$)NNgG?4FU|!Jr>1jUq
z5zj-*iF>Ik`R(7jW}wxFBDd@kMhT?B<cP5?RrrKeN@KCrIj1NzLhSqNKS9l(_Xd=_
zcgN~`^SBd~Y)v#a#cYqQvd{G#RoM>M#AEBJv$(w#dqng$PgjC1<vv*`*Y6(Ue$ni{
zuIBc(n@M%3!&7+=1@rlM#`>+mh<@Nm{+FV6Vu?TSujoYf+g2LiTTrSOURN<q1s2Eq
zfj~-Z-hGI=fQ;)MisY{TwkLo=c=pJWA^cTFM}8j19(<Hgje9#Ct1JD`R!YZIx?y*v
zr~z`R+#J2P2fe&qa!^)jY;f2T!q|uAx(Kx$dqJQ>-as$Eb+2Q%5BVKta(Bw!05(Bj
zdCIe<K%F#Z)iGt!LV@|RR=!SC2OkOfS<lUZDef$-<IwX;(-Gj)v>LGRpCn{M$j{Nl
z+rc)m-)8)o`MfQv($+hYlKHsmeBX^iZ<ou#7IH`y-nuW;kgQQbruW>U49k&8DKGEv
z-3^Q2wzY{6G2vZHL>O8>){xINFtD&3@5j~Qi|(9y0VUKz+aJ<9MP^@ULm`l}TTu5N
zg#DqI;@+}`{eYg;JL9kB$WMZAzAVvBD1Cc?0S(HC4Ut5k=bhTN^BL=_(tC<5_;xg*
zoH`b}lm7x<wms+o<jMmI=WN*i1+xRIAmm(z-6iXyDkauS&bAB@np6UEGK8{OT!F&x
zIRlM!oofyMONM!cc&SUE-V>>0np6cTpHJsKk8lT4AbnGse5&%+_!R2(R^(5Y?Pt%(
zJb}E911S>~mD|H+(A->Hk<$<^fpbX8r9z}GRg?3v1`tnEG__7`mp|2fipG|8gg^FG
zJGpn4NBlvo1}oE?0PF^_GqFH(El>itw_UMTOBagT;z_FDKBR)Wl^0!=F!h<_8my)x
zW6jiER_j@HFcrOX)3TJW)Gi59PK79_9DWv|i|f0TvsJZ`I}NwB%4oLPQQJaOK6pZZ
zS5i3Q<;4J<C!;n2Ibe|Gz&0uQBo%4ky6LlYeWI-qOJvQyw%WyPSt-It8p%X6XyJ;y
zVr8m*&*0p1J$QSIYNpnZAVO;|=W3Gbn<GHL4`<&wx$P3o&z(d>ETI%XAK}U6pNkC`
z|DE~!WGP~!F)oD~+MhPYAHRLSJb#5vvdfYXp__a)*p6?-q%-MpNDe2N?r-lygZrwx
zJ3S~LMZGs(C$a^mG+KHwR$i^bw!jG1kbIFZYJ|mJzC`Z9FY@G~3zWEp(9G(^2NRTi
z-RH{5(pYv?I))0SeSQU(#2k?XLbiL_b7J5RYi}B`CuPH>TOZaz)|dwGSd{2;?%wO(
z)5WK8d&yH4t0g$J*}GYn_n|ZoZEZGkT@x2l{EIdS<3?k=BXqg!o)I>QS%%&(%^-Tq
zjT~k3Q=U2ad-z20soO`Aa_6;`nA!wE4O?s(9RXkE2F1+Am?!+0KR@9<E-^W`t3qJz
z6a*hPx4P@?@WDqX72R8I??TPLZ~H%V!S6%_wvTNt#9Ya=+v1!A&QLRe9k^de4CLos
zCBSD-sBbL_-!$TAMCRBw`PbRSqX0~s5NTkVMmkyTG#7N`1v1n)Lw;(t`;tgYhxk6<
z)Z3K8Zb#d@?ZdJwg?i)Ox&I5x&;TL~5FrE;`=uEO)y^=$4-`p>V;v>Tt6CYo>Waq(
zK%bEc9@1ZhZ#D7)3WU5~totNT@2=>Dz0<071b)dlno68W{QIVYwu_Jo!ZeC~{|~BL
zVL<5{uR0@OBE!4p7N|Z6Na)PAvlA`x^>N&r+PW6hQ>NU0s2B{?l@!llZzs8qF1>hX
zBF_K>SgI4v^$Xk%B$gH>D|AAoiIV`5O|N20+4E3;6x<iVBzJrnq{|(b{Vb7szwD}h
z`F8iB_U*JJW@+jKO&EKm&5Ug%Az_E^uBcc%G!K5V)pB;VS2~5+LAJUj*?K596u6st
zPLA@s=B8S%Ix1{=bP0x9`gK;t-M{O?i23%lF+`0i*)wLFXwX!5WP>KEgsle(gc!X^
zT_G8}9mgH1!ngh+@uj=;ZQJpUy`vfieMMt-9k+RMxLNlCV3!OHrR(zA+Fs~&Pn(yx
zacJ9eMQ9hZ65#e><BtE2Hh#uDTwU_UmASe)|1!Dzhu1px7cC39G>S?Myv9p2+y+MJ
zEQ)laE-UlYnDc(IL6ED;GO9!`_`N6}US604qRd(AI~%a6G5ZAWdMCMsYfIyyfmkj5
z#Cw_xH7h=<TeF9|fJv#wyk97!da`;5^$X&=yjr}hvbDZ216I_M-e&|DO*>PgF>oF5
zRN{T3%_~*vaJLS~9BVRd!@7_2yD85AO<0VVDJUoA1KOf{qT+78Pqifw$sLG9JHM+<
zsSUG3C>8&}S|;{%o2|MJCPr-e>KV;yS}|Glq#VVTQO1<2O4Z4nmIEooF&(AXOa6O$
z<ilE)xixJVi&frL+tq&8w3i1B<1fB=on&WP$875yVev)@Wj??eZoiZt&1z;9ot7#$
zWjtHi%$=AcQStiHXQmAx46Ys2GO!M+s7bNwRTj@tv+!}b(Qum<xDRMf@uya#xoLG@
zKl$<uKixG*Q#U5`7w=RTps_3>mO-}_KrQ}yi)JCf<1Ufjb(E$eHDnqrjSm(Q9ur)^
z;BN%8s7AoMq@$}gKSyr$#mKKYR!IWN0aj)*Qj%`;6c91!da{+xG(@ka2oO8ySrL|#
z4GjZt_J6^-mS4ZL`?6#z0<Exi;TsS#YbQG(8#u>yJw(Qmq%YR4b|v>btE_9qdElSO
z)zDY4SL1k2!giw3csLK$@O=1@{^%>?K&0t7YnF~f-QIP(Z@V?S9-A^LFFoJeE8Ivu
zQCDB26}uo6WLHC*^z6mwBxMR-mP-OrhJp2Ex_2*!3$abag6T6JR1{k&4f(nxwM2F!
zxm2XClJj;W$WIw<NhKYuAWZ1nrWg8>orr0BcV0*FZ)JK+ek-N>j!f8Vz~gx3dIna`
zJel~I*{BIMkS;ItstS?gcMzLL<$8h4E173X>r-3jXqdLqIg;TOs*@B0vSB4{$n((>
zr{yos8X-q|VDo*7tx8F3fA!lkb)YSGH;*DG#jU8l2KO<h!k{fyM7LeIO})7iaQ3~K
z*bT@+=T6^Q^u)wws~iJ)YMKvY5LK5-c|MJ4>bI{sTpuPSQ#^pM^A#u`3my0q8WU@X
z*5J*mSO_-4zPyF5UMBbI=c*TJu|Me<HXu%-)cN=euvQ>F`Py^eNjTCedRDD-?0MIs
zW`J0<tJmgr+CzV-NvkBx^?uIUjM}hLN?8|O1^R}xhz?4B{v>JLuq+ZQZS`PXW-?%e
zP&_hpL%Abo6aMf>p70XK>Se;;;cS%cnCiJ09!C!51(FkDJw^-YnYlr#h%KO!n*gIU
z#wn$DX8T?z^63itz251WmV7|pKmLsPr!~aYIaoW^{qEKm0eorjxkPj3J|WxmGQ|rQ
zTVN26csjd{)2%FM$^_E~65V%6PG-}F^8k*85)k?#JO5S;|5?Z92nuRjuS>eBsFXbN
z0?gj^s9T}98-8G<k&Nejfj9qHm!$o0=??rEwi}`RyB>tke*6M{(HIq1U!daeXBq~v
z(0|r#;B^uf!p7_6b_}#7eLd#zW|D$^)`FI!eqd38QE$CNO^PbJy2b&ri!}A+a>omN
z^roCgQCx&J94}<jTjza56@-Tkee7sSE4~&byQnm?OUmdEePi9dLxpyPWrqSy<#6F`
z6FWPAeswWqKHCaw0ac8lZ$1)JUzh=AWw<H3MM7U&pT6JYIO6B6kT|Z*a_*ZG=ym}+
zJwptKzUzS8ds-<&-*W^?A;RIzO35nC9%fR!Xl9A8)Hv2%S;<5i4C|wry1dM&TlWig
zWWz2tu+5k!o&=^)*f`-O@18*7Ov}8KsnZ~P@-mC7|K}Uz7u*1O5+9<4b*tsh?Vtj+
ztIC?1WY1i=X;L{oD5O|LSpC?tD>v%(xI%JI8-ud!0^W%y9A}gX*}djD$YvtfR)UC$
z^B)Kn*vfBvZC?Bw9A5T0IXegKGaX1YtN#0Dz?=T;mRUwGuE>T`o+yn%H6rCMh8-tS
zg<mp^DV5%|nD?8x1s~BS(XVzDyq|&LJwN@>-B)rF`XWrHJvl!vB&4qJP9PTpC1x|k
zndT@Zz9gg$#I9RaxzRQ2bxo?-1>NgdFQ0Xwn{;fsf{w|ylNFOPcw3Y+S2@-XI3D2F
zvd3<qr3yiw)>JuthJ7B&k%<ncwnCFEI3m708lE=#$zPtikVxq7<)GgInni<v;iEkC
z{m)GHf0o7n^uZ^^EXC=k0~75V#Io9kUKUfPl-n|*U##_`vMAqh&L-y1Hp{cz@im!c
zvPlS6b_|DoEtS=oc~e_yNuYD-b#pg>*o;(Fw>`A{TA|}YF`ofecPt?VpJ^b~7uSJi
z<5<HzSG`n`f;^%hzC$d-Ihxa9KNXRj@{LMj<Mxyay*l#-as=}-vI_$y6(+zfkQ^AJ
zn!-P*;}`RDHkLpm$ko&TuIvBl2f+Vt;0wYd?k#p#+vIL!rtQ-xi%)D}Zk3=3K3ELE
z>~>K+41;qz_smR1?VTbll#tSi3=as^L{@HAy^;&3BrfeH90umu3=c3Gtc*T{B_UxH
zv?e$}BHjFY4}=B}S_rZyL~L~klD(F}Phcxn56%&v(^xGft{K*Mfd0E6@P4FU>Q<xs
z7T#a1<prnq!vip(?Fgd2D=*@xdBF2CfdZx_Z^JpUe`OwjdV2r+dt+A$V;lK`(vn{O
z_=>0cPI7b-{wM`|5E7bz@?Rl$wyi<9qXQ5r1aiZoZs^4tdi-*mAvMBDPuS5#ZDKmM
z0g<h|Ud(7WZ9@cjXyR~>GudCB(>_*wY8^KHWH@I~7QkL!C7+wsLz2X7f-uqfiyxjP
zw)FFEF^Zq79yhza8|Uw?7k0%u>oSvI_#WrPB*J~%T9p+6%p!GwcT!r+O2;%64!}ef
zqZm4q2k@tW35e&TB=&CO8OW+&?b$TH|DP$<|IWABzB2^ky=reb18utv7l>GpUMH-^
z6F^Drjn(n^JB^oAA|S5Ek6_>5Yz|^(n>;`*iOnQDRwL0-Py90}3|0UgJs!GAk5PL6
z@5}ptTsU-&1Ue6ELG8bJt+)YAH3%lxKAH3uFMPQAcTgeOOE@ZWOxu8yYkAu0XZxle
zr~_Y!Zd{`Fqx`e}0^dkU5;VNoMqGRNuQT`mZEgOa{~pLE4<66A-%M@r|L^ht=l2g3
zr^EpOFQXwqemCwD0+ganCW6i3PdVy;Ut*H~&3^y;Us-T)Cv^!7G5+Tj$_f8J_=gtP
z+Eg^@FB9MmHavkbiOk!ZXCPN;4+Uy6J5HuTEAT6f)+7M*H**|dF#f!D?6sK6(F5cf
zsvWyIDUH$e0{WBv)aCu+K01y~7o%Xq^#32f0l>@#G*m3<_qo4+mMlF2mBj|YV|>?$
zq0vbm$#85YnF0QZ<e*Tm0?abmYf*PY@bwp#PboleS6i<@TQ7C)O9V)+))|QSrvOdM
zvMOH~;ODQVTC^FWSAazyrTFi|{xFc(N+2ZgpH~82aZA%2Jc^}m9;VsN;fQC+;YBA-
zJ0KvxigzI&1@>~+JKDRF4TpY$$#MoJ0M3C`0aV3D$98(FXlnCFaMAx0rTz~eG!owg
zM*^|0_2sfZ&DYVJjvy`%1B~<Dw&!N?k^!cE06g<x2zU_H+91Pgu8QNPc>&R^q?Jl}
zm%}(`as)^_U@E{Z76pLuJTX2)Y3OUXs2J{eT0Ps+5)hGxCO$PBE4<r#Z%Ad^btAK`
zeyD@Vhb!H=>vmv2VDR0W1ekUcFb7^F=DVXoD9;~gh#gUkagk{8Z@NeuqUTUoxg8GX
z<imiwdKg4K%R7ku9~J<-)XA31_w(DLdOTT>3-@obAOmUVHHH7{s~Sl85UhX7quFb=
ze|=tHse<u{fyWilaPh!xI0T69vrTo-;SbiCyGDl(M*;Vz1vKa<f5tO;&A;LGKf!Ia
zFrV?-0s@%9ubilNB$EK|jt|uEXlLj5U6=0`P@n-BMTo~6fZk7TXrBoJNAH&(+vI2s
zDxB&!VOmqSC*Z}vsE`Wx^VD>H>Jy8`{eah(iT$Em2}rpp#P>d{zkU#C#JCc9bn$%_
zp}(I>*3SQ(kDkR3cF!I51fsup5AQuficGwtCs@2!Hj(OwfSz@Y#>m%+<YfHO>ZZA<
z$t185pj4AM-EBlxG1ygT;y|dpHnH!%<Wln`zkbMlo0FExfSVvER`Tb`{iE<73^}C$
z3(anzH2u`!S%Sl#?~{|<2wuo*Z*P(~_^s(FINbaiKGp!ph+Dv7)T_aDSl4P}_NoE@
zGGPLe==91JKU@Oz6`}^g`Cn8$+jmCr%xmX9eEIt*FxDaF^*q6|?1%=zeLS_*c!4~`
z=wz0*#v~P3wtl6Or~}oD^J~w^9{vUn5Ts-e`$2SW0#KM(Oby~G90%ZeA^72!8PbVj
zDs(|hWFKIvlE=c9M&Is%7uUxVfH~KbsnB4Lwzn|$H_S0&w$_A+?#_1LFHJE=p1O@M
zKBm8`V{m!D^j&(XaW+fD44C-=1!gxJJKOClaMV;zf^f6kvszEh{3?aoiTPGswSY-;
z9k{CVurodqjQu`9e0Pe$cQC3yw!~RZ6+61JM~lwK$q?w>#=CUaTR4!BPJ+$iDxET<
zcbJQ%80m#$d;C>ZXx-|DT;bwdf#f)!TfJD{9XgCFsQVo(b!^f2?Y~vX3^+S%6m5Nz
zY$7>nN@rTjM#i@(*N<7i0S6Np>}hqf750YsZ5v`4G6suY;5RftY=$lmmssJibw#%p
zrvw0sUME0#GQ#W*%}JjZNVf0x9&?cK#a2$nJ&=4riRmb&I|FBC$J8309xCtE%GZ^#
zyc$VlalxO`qgh5cT!Bjg3)L-OVNH4+^4gJ&_!zth48#&%;T{*2chobAicS)*K878#
z`vv^OB)oaUahT?u4)6OK#%zs%kKWGwNgoSFAW#mg>dl$%D<^GEfH3xv(Mu4PdF3gi
zwpHyz(vwahzT!6=rW4^i+40KBs`I*N39nr{KuS!Zm3B4F!~R)%?7@uy>$<XmP>P4X
zjy~r>`EXnTockZDUVo)t!>b|#>!*O#6%(itCGsGsP~dC>7s5n%#g|pbfa8fNR5AT&
zHg_$cX0>PL9L0Oc{}$L-o|}Ko_p$i)k+{C~G0^kbuO3i|yK`qTjz{GfDd>rdoUp2X
z&t=@OulPW6lnfZW&sWP6l_us%-M@A2ALf7fGLYh4%Y8hjvkxvMdG<?fO8?w+#vyp0
zgOGaRi(rNn+<(Tl6u4g)|D}lz{s_DuV9j@P0E+B;eFC_y*0dQ5h>BpcEdam^Ai;-V
zIN&9`9xNX&hd%7{cm<esT=>Qe4<cW0TFMh|TZ%uqk3-I}6@Z!!{|!Kf=f03^Tjc?C
z@n1(@3kxXX{hEc4mU2(F0kQ{8<no~Z<*nggDJs9gIY!Y3P`+DVj840Vl#+E!BerV#
z{rZT0QMP<w)Z0OWGx4=dI6V>^=P?r-Wvc4{$9jWq-oyvJH>?K(p%BIz2WcqzxekcC
z1KPab8W|>j7y^><3^*`%rG%RUIh{bc;|l)9(y{)Ss1Fo_g~#!`jsQUtVsy5XJQEvm
zk`v`eOdAvez<GRexxg!_61n~9dTs@;J-8eC_O43ZO79n*i{47@_sw<c1@#Da^=>|H
zUe8`+gEil!>`}n>`^cIlOSTsZ^L@{Es5v6~diy1ZsYt<<9otLqyz*?Ndd%^_x~H5c
zpWiAp>=im3YIl%5*#y(f2%<iq8h;u31y3jDD8&1b;4K9`Ya_GAF7-qjDYi`E+J)e;
zSn<vd6&-xui1+v0k(gD`l{rEPJ7qTvRj}|}ngNjG`aEU-cn<NSfmm;J$gPD=KrX#e
zve`L53Z|ge^laEab%}(C989{Xzr27Uc+V8i5HbURbPZDqhTQLZ=KQ-1HuHR?szVq0
zK+piS1q@Y_47XvY;dy#bJOJ9>njcJEdukwmf&6R+_;I_g6{3{8q?&tJu%I?ux&>Cq
zWcM~sUwmjKuq>n`PTq?M4?csF^WGM=8^1Sol;Jb&J8BYEBOE{FR$aS8Yx4Q6#A-OH
ze;gAYgY9F`B5FeL+C!F!!>&Y?$L+%26YuWrUF<O>e!GT0xP7cxJg<3U#U0>A4}4o4
zn&s&qo@H~*2#sivuLD4!{-9%vZrIP%qU=Tt;8Aa({K0be_UAy)^A&L54PNZpisx_9
z7-<Odd8z9Lv%Ed7p>0>mccbzXoL&WW=)VFLGM4ipYZc_r$*+9OsS9oV$Q(8Ge1s89
zyZ=?X=8LqSl@kr&i!_UQ(0+XKYf{F?Da{VKEq=x!z8z6S<L9<>5TAMS;Q*WM8FeOl
ziY}ejIj8trK!hf2VCg;cWE4YDh3fU}fGGCYyy6PK1?q>&oN7*Xx?Av-%sJNR!~w*u
zy1|3Lc6S6|Wb}(K%6T^?s!rhfnRO~T1?@lI6iw!6?iWv#Ws1+L2OJGK0tn}@h#G&F
z;~kN|FJ2Y`8NB@f>!qTOfAR-<f)P;5zzrtX&tF&po;q2iD#JQvuk~RdtI5Nohn5g?
zEYi$HQk|iV_Tb_;pzZ@|IUO9JN<&hz<Xu~g-4Hst-{*ML@YX_C@-p0?8JAqvB#TFf
zZ<ht{5?`lJVJ`|l9oBbIWsaL?x?%DQ)JOEQ8gGi2iW0)eBJn{B4{na>690o&-$^w|
zAq3W8I08*QSCwTv9&M%-q8=UQECKSkr{oug8%v;J%h5{51v0YkAmOXg;PHu>D(I>M
zB1e5m1v=1I{kc_tP}kpbCFieD(LDsBDTi4201xP9Y6%E=uK8Wa|HPA=A$Z{fq<|h@
zhJUZ<`E=jQYed21#)UDtepWFSH8Tq_jD$!?4oaX^592-ZRUaOv)fP(gjMH$|+6N4Q
z@t8KRLWcM1fh*V7>6oZh?>s&xI4K{swC<%r;K^Y^8x@yKZ{Dz3yLMtAM38dy=vJ)h
z%+_0F9umBfmzcrnJcoo6<s9CXa#nnh`a*oTGX*OfQ{m7rUu^G8l+j5xsTO5_StUP2
zg^a5O&!py|nKfXoH5(*wqHi8ASPu1gsN*!RDjNYL+j+<8qqiF<RQNqrQ=3|eY1D>d
z;39rUyN$<^Clxzgcf(TZNtkD!wf4eoz%lHQY94j)n2Ecrla0qwx}@!eC!N56qR2Qv
zG@;U^cEjS9US7ERoDgj0EqFn&My}>Cuam-f_WD<_5DY5nR&vu`;Ss{B5Gj>MY=UeB
z3CudyGI2eGO&}z4SyB^MXp}oq@*y6cb_0}Yp;-;UXgZ8?eDaN1YmO3=$0beCKF_cc
z<f<q#D=e+O%1syVU&GVC$IkSN_~esB>aHxWA5bvYa=vj+{5HdhHf#0=BTh`TgI61F
z3sAz?uD`agUMmq9qQ*qj2J%NbN0+@hh$McCl71j_{BnnT!ZMJk*(_w~hY7|H<5MS|
zcU4+LiOFIo-1}u4@x0OM`9A~YARmq(T%Kd2c`w2DBxb3UH!k!M*Yua686jzZS<R|R
zXXOz#&Le~9)bEyy<qQgGk)eiMKTOs`>ut0Yj)O`G$9IJNhmx2{uMS)}ZCzeI7<l#Z
z_s=MXm<m<ll?H`CX1^zctbP%Og>ZhFGLTZ>=oEH?w;|)moL{e#?nyBR@xO9?31q9K
zXB|4{=IE<M1`~bmMV=+zKBjlMsov&9wYMOzDXsD3pNR7&q(q7FpJz@HNGaLa>J@S2
z^1-!7<sTAlD+c5vuH59XtmENxl8%>vzE#E>xdMgUEz22>(r?GPSho}2^FYbS8bN}+
zj^4iE52>#x4+PpJO%6-GyMEOxUn6Wrt#}{oewTqjN^T04WQ>}7&s~Q+x&`%A=@~bi
z(;1+hP`g(e5u825-JyE2*e|H8%Dq(kz?i0nsvMV)G7aT3(dp)fUQ_D1lPrK6AE{4D
zQOsX>Cvr7qnw{^9OvDu?q^A`@Tq~61?H?}OD%{vq-pYD5oW?=b_)SXOozkYGRw#^$
zA>Fp7ol6iZ=$VmH+0+VEd)DXD2z~z+Z#D+Lhv|iknvo}dV6%m^wLa%w*(8#{1$H!r
z+^*>_W{-V0+N=A{vkzNNqW%tHk<H$J*V3<O#WO>qN90pjAj;l!BBTQ01{GIPF-}k6
zyv^FPabqOKv9DMt(7=?27>H1Z99SLF6z`Wdoj_rVLQs2Ul^;nd3k9u&BHT;8&bfI+
zC-<7PhP?das+3$Tt=ajU%!j<H-+dG$m#~P7h)U~QI$gfSG{6kG4}~sANibx7u<7D%
zHV#^7-^R7ICiutf^oG>>(+OIB?vG8Q`}`)aQxy70MP-ZHp?D8;67tyteo(SPI=#f?
z{gSwom13bdl^#>uX3=h!(@2F>k)8rg@9|v!o_5w(W5|PtDnFLop1_mNZX}2lyr^YU
z1Foi}4UZwXTNMn0F@nTPs>W$48Zf`Eh8Alen^&^c3`tfU+rMPTJ$QMc=aDjZVv2E9
z8#&|OC7{4J(k}&X`fa{<t!S}c+aJf+(`vV0RY4hvs8xIcx#zD|n%zol{X#l-(w5-G
zIqr&YUqr@fD^wTWeo)a9jvaCQuwg4qcfeY>^t#u)nEOcedJDN8>XGZWln3J12MJ+c
zUcgrspXP86B!ETklg-@W8FS>E!+ncW{GWu-nZ&qK>JPjGAAb^3d}7-YAbrY%E-!?V
znngCYAb;<z^)zeGxYZ#fZG*PXzklpX>_De-e@$3*maAMODdl6Dr|zcJHC(EjIPFAA
zWzFPT?E@Y6?*);c&Pj#p5<l!AkTOk5xEvg&@iOAVrMEq`rZaAC8OCqVH{A-*+zou*
zSHL~I7p@<W&D+Ptag#W2mU&i&&<|<xmJ~;k)$V8^CQTMm!lVAOsS<ey&GEP==+XXg
zJLV4Nnz|{RSXo<=fjMY2`T()C+KI2?vV2tRd_K{6#NuQ~w<=5n$Xh=Pw*0nklvl^a
zdDumbA!BcyriC`Na_K{fuYGk&xJYc@!*3b&o!?J_?N)0OGUM@4NwOTgA4$b?jh?+o
zASzxwaImnk*cSedU@t&|Gf&9~xjGrk5@$)(QJ{y^X9Ckkg%twbSN+wmowm6UF}iA)
z4xgU`BIhrQ#UeWjjFfDJTTGOpZ!eWy-M%raB|Gj_WVFqx7p8eN<>T8BXTknd>PIjE
zEOdzI^VKi!io!(~ZY@Y6=u)wk-@ap&wH146Kq~!X;y%7cB0-vAr-fiaiGzC5<>CR7
zf65tw+5~Pt_6qBv=p;+ui<A<3hZ@31R=XE<T@QgCMBD!Zp}zwQ*949KTwl|nb{K^+
z@T5a8Jy(RZ=Ki%n95fgYr|1-J>+Q!E>oj(mKK{?C2aaq)0)SYuY@Pqlu}EQjOo#kw
z_O+V7@A{4Uz1w=-sg+3NM2nc|1ZkO8LE#;8K}v%~KE~qvnlBxCZD@LYyNT_Mqekaq
zCD2uMX5HW9>_{IXm5oWK)O}|mGYnHW`S%T0=}xa4epwUnPPGd;tf89I1UOV=ul2A!
z9nq^TX=Br)d52?72`!Fx*a#?FfKD8g{-n^0MY`CoJ**z}kMFO|hzRvfV?oYVCVs95
zLp^FG7O()@jV2_Y;Cao19EP*DLc8Al^ac;Ll+o{1V&*n75CN|vv9*G70==dA2gtYP
z$C^9D6S=unzW19<d1vw+th<IwoW8s(kkCTk_HXkn>9wAGByvoKet_ZcE-!+2?N`}S
zcMCpdKI~K|i#?cv^6nHuzge9++Ft1?zbzwTxA#!DQ9IX`ypN{BtB=dzXuk|fwr}T-
zE&9e}rl$(oAly-(EW6{QdSb^myj6$s>z|6yy|UM0d#tK^?-$aWEn039<f;31I%7=f
z8=@06hCu2haMz?hl+!E0uC6CPWhoSxLX6auTJY&*m2+#2w-px7n9y)B<aG`u6x&Y8
z=ra8J@%gdX?~+r;Uquhk^6!Fkj$^cx@vdG$soh<@wqB97g0l|fP=Y23mKJX9r7tn3
ziM(p2EP%x1s${Duih|}SZf)=U#+BJ*_uwaCT1rp()NBWo)2*2o9w0kRqt;OH_Tgz;
zmws)Z(vwbXuo;}FVU#DsTl=5mq>S)+@PI_ifEC~Q+JgaLYA82l|HhwkkKgX<(F;pI
ziM7z*xv)?p-`P(d&!cv}@r}phzImKX>(~CC7aEf|BDFPBXMomL7;Wo-8RhP3xewf^
zZL2xj4c%3PqfJ|wa@uXon0~h#l7*)L6wLw)6@-9IOp?M(EVqXB1N2Jc>A`P_{k{vJ
zZ*?>$qe&<gd6{5P>O)QlTB??%&f^*{1qIu!T^ea(vG8k@Z_g1>p1&wfDdX_y+7roy
zJC73YL<Sv8Ow9%G_#AAYJq|YO5A5Gg`HEwH6_yK3JrLDftXewZA-jH$ldGhW;LQJH
zl?MrLaDa$YK%-HrEU|8sBn1!+6*z)MUD8-(2Qp-S!2l1)`DJF{F7<O?aVhNAckxNA
zDf_5`tvW(|M`yev3cb(AyqPvCm{ukjRyR;D*B|NhyX>HOY%Wec0BDR|_LC3jCiR>f
zbawmLzN)*6zgqt2w}Z>AVt$@BeadqBWZh_5-RO;##26N~AG>m7o_9P1U9R%#kTF=!
z`Ap<%Hl;dN;+2%VA<#9<-u*t(K+_*B_Wi&K-LXc#*D20!8QbkU-(30fY*HMzzeej9
zfbbh9a`lJrw=q?VZz<DPySLh09+9+t>0DB;a%bWoUg!qbQ6O`?$Ikxy7pXEY^!_=I
zfNdV$fUuHX>($@)%y%I|5o+8vTS#(t%d0Cc1F8-mWw?SAiMOBq`UX!q?51zMh??e8
zcb@?eBXZOhqqwgpFZm<Z$b|N2dI;C9fe9rL*3fpNSn}U?)!+NVb|;+R6TUDoI&#b%
z)f;6c`!JlYS4N}TuDd8iBAc+4ghp-9=ow#iv$BvOlK*?~InMfoyJ-PmCqT9lM1QGN
zaoYk764K>t;#Q@)x@wscP;(Z#NLoe7^}}>-i>CHo3WqE+jt=hWXy)af{V?tA(}J+j
z>+Vvff&q)<5+_@Z19G%`fEVKy)9$kLiou>Z-NuVVid&k;_6zJ$g#&h;maD?PTZSdY
z#F#WN)9a-B7mcW_xzLLrpC<RbNIx?NTcL<IRb9;xT|r&~*g!GcxZU#1Q!c#x=e*Q1
zC?m#u5F5!)-N(D{xnK0RSZk6Jz7*8k&GVJZI43=g%yBy?qlr8ud{iHSd+tmLo{0Z7
zLGy3aqqCrOwi+zBo2+lN(!B)Af+cu^nndkqw^I}nnQHbe#JUVyI47S$Bjt7aSrkBk
zj(YRzr?wt<Iw<coHP`QbQh;|59bK1i%HN-u58R45``(P@pJ3<v9)e@V+yMCT1p;YT
zQj00GhV>x4?M%PbB~JQ;M^u0JpzfUyVFf1Z`qMIy&4X&1w7g?SP#ZX|V;W(=qSg+2
zCV#W8&pzv`doJiyI3pucWHKqRbHf7BbLwq{esx;E1L`Qwtdz`mZ@7M!;&Gp=hi3Hn
z7}7m({*WBDFyAKV>zT;Wtc0;)iY4ej`4M_dBC9+p;k1z(Gx-(YBRp$jzz~)Z@L&_>
z4{oD4rt1jZ<f4;RCtbEpOhrxOck6xUW#}iU$DNGK(F+W^ThS;0txrB1jU~V1HP!6L
z&irV1zhF?<xKEZo(CFuWcvwenwFln8rWtQH#2nHXWF?68XX=@#Y?%9Y+1Rtn%y945
ztLHX!Tkn+s6G|Lj?CkKW&WYT;l~MsGpw1<Ob{O-S39Q&a?UWZpQrunCm7f;+F4*M4
z(?6hD(e?n8<cNG4DAphWHRMZUH}qO8DNZ-PV(OjTHXQ=S2{FB%M1d4B9-6-C-B?jf
zwkJ9}Ag>q=d}mL^E?C$IB^D|OPN}U&?$iT+vA~0esFC;nKVEwX?i&0s3#Hb6e!EEi
znnKzD28+0S{${`AdcVLmOw~%mke`s>Wo9pzO{{{^7AP7KVqY=!$khMV6x_*b7z3U2
z(^`$2?)_!P^|n*Y*iq0fI5jXd^7Cbk%`y7#7JnWFw)lp|px{4S{Ni6*oFIdiB}XCl
zY9XOoX(WyNpwZ-9M8Rf=;_Ku+)Q6diS77dR_Zsg~&6a@_nj@M~bl9j?r&=>$Tkpv$
zC9UZ&4t~<P>l6xGkLoi*H1ZQiOEchQ6D*gduXT{#C%jGkcx$l@Z0YuG=~!B8lzG2*
zdiSE1MLna5zUZL>%s>27iQ9Ta;Q7_gX%eGx<k7R*e2oto-oTG(RYc*d(saiIyBQBX
z!C#<=yBQoyV~N~aaO5D6FdX0Ef9G3%A#sd=byZ3-@_?uB{M*o#ohFe>;@mwB?#b=&
zW?7Weeqtl}HoxJlF!^>t?SMQ(O^0y3xk6S{$Au$JxZPqzER;6T$Y4n!XXZu9OR@dH
zr}SM4znhFX1f>GHSS;Ne>tJgG-x|rKhk^Z5pB+(g$jwVLH6zogM*pqIIMrXyS90{V
zj_r0p1a4*=VPlR^VBP6co}5dd59#Wb>HZUcK_v+EgSXpz_GsT+=${W~uHKw|a0alX
zwNbMa&seT69R;6@i+xshaD(_f=vy$;;+!O+8$*sM2(>4x*y2FZTTA~*2fujIoG$ck
zm-KgcKmqp_r-=2coCA0THczH$`qw@#;uKlwo)nAq7h{p0JMz5k9hTuIgArCDtAV&B
zz_E6vBx-KAZT8_G(S~0~rd7uxKnKRale`1vTkgf0B3l4~%*t&J%@_%C*X<7S(9E|~
zfZjQBee?AOK&j;;iHJfOn12i)M#=h>;5jU#Ibw&3gab4M6h>{a2cxvGV0du}BYM*d
z^YYxCyVn*NDFi#bBqxmXP!e}`;?&U}z?FExdgAw8r~cm=#HD3536!`i6{?Q@LrW%)
zM=vj3T)KApOpp~ZcS?=nEM?rRqvS+BJhg_5-nYnOUYLU&*^1}Ng?ADK?_%Ia10G;J
zZ1ap&>+3AKY_F4lrM1Yjy^I%mo|Vf~fBY4B^K*gKHUZeINNbf>f7I@TtjsjKOIQD1
zRIJB}jPQY{y*!uVZ>82yT;C$txci;Np3zpK=(P-|cKW_pkB)E-N<b+k-Dbh?cx6V9
ziu6NlljJ!30(#i}2_yaRLlRD<{{L-<8ygdks=j&qPlvYgrevdG$<0NRd{?E7yCbcJ
zwKu2E_NyR1r9Yeld~`EYvY@5OkGZSXTt9l{u-B`b!a5+Ns_~=Bg{>&!o6udQ^^a<%
zFlWdyKYwYmB*}t(;<{Yok_B6ec3Nd+AE7{mpA>o+X=@IvJsDk8&D!vreYLJ^uyN5t
z{|Ba_q>}b^#Fgu}1V26Y^c^RjSm`#uaK3NQa8z47nuG%a*}f$hzTlO5j)U*Kndybg
zb&CccRllLYX4@`UyHR<b64V)x%}4L@dRi};TpoF9U}EIyL$AV{F!~f7IyL}G0QXC2
zD?=wrg6bRLKFCSv*Fm_`O+m-h6vg{r^E*%xbCdG6y!oYlhTk~^M-xVkADLuY8zpc+
zXLoAHC9Tb-vLn=1O)k_Lk{`)>F5L3*+O+6gQ(teZDlVs}Q~io}@T;G^6rPr(r?b#{
z>E^qnpZ6axoaC>ae0oOPWfD9WDrl13&?<}aHz;q)X9`>E_D-mKr77Ik5BexQn7i42
zW5#Nx7jAL)*VtFTlB?EQzq4pEgMu*VM3#&B>t&chQ3q?!=SR$}xc|xJuM(%wHDi9N
zI3}m)+$q(FRjqNHdhs3(A=O;&-zeOynG+D`XN6{O+aO4~uOFr^-y8!4St@M8h>rF>
zC(sf0t{Jb}Y_o!Fa;yA!ivptrs=8^7{oT8G8$oL?@7x$0KM{%CjX78`QUB}7Ot-fp
zH3yL$rLkEjIJ`Zb{*M7z|IQ*~>4x7I0A===oWVr)q?wvZ=Q4FvlE+dZI=ni93(cp*
zQkU)qAo<=uH#34v-t3tm%6e^~lH%Bj<%?Zc?T)g{uORZe8e(U)_6SKfcA6?QkT#JN
zy4<Vx%WO<?g76}LU}3`r71};=UbnBj=N--X{H`J8jE2=#9n!K?MkO$!y%Wz(*-COv
zchJsJ+{i7L#*lv1wP(G%9%90@BcOZV3je{eSX0p<%7_SC>E8(c7;-?A-m=o_LxmRl
z-tg=R3C5v(t@pxn(`KoFnAt;tv{}(x=#`XCq4$_?P{?ZzD17%oZq!Ni^TR2JY*`0A
ztoKCU`-<29?3tIuySLBRdu#k1kesJXAtdIF>rswV8wA53i!gPHqHDJd?SF$(u+n;W
zN2|0njh(jj1K4#+ecex?t@ZnA>rvYMvCiAGnY6O|)IO$0^ZVZE!a95u1QK1{QKF=n
zF`EbF3|rX~^~<U~!A5Uss4c3Zu#2k29o^g2MMFl0DQx_<w^U{?-pT&1q%<9t6&qLt
z#*|?*7A0O7<sKi4SgBIDz1DcbNj~w(XhV$4z4|Q7B@iUKDfZkoG&NEBZ1}i2_ko9}
z)Ku4{vCj~bbgv&<XRXcYd!DQrJj6j<eKJ__H-_n3kKQfpYk3ho#uAQIAnJF3C!I1^
zLe0U2x}Sj=qMyS;Vf)_Ku4E4;>r@NEWU5F{@)1UE>CCFHN1LL(xD<OP^xjg-=ax6S
zb4}`-A{d9F7>4|`V%3QJ=@iYULu|w?gko=#Eb+K&CDhJbr|H)IS=5;q+1QWzP)(O|
z{>WB>vYRe|{^X8G@|@ri_21J(#4KBLN47I}NCO?k1l3VzGcjKLcP4qn1=wjSO5WIs
z(NbGzYrnjfsaieEmeQ{*x?UpF?5Co+N*c&KVux5boGu*tu~J9dA%kk)p7470T&rH~
zqJKsO(Y~c24WyNOv8lF12PLBe<`_@Mz-@EiuA@T%6+F$Jn?f;}qQFI)pWBwSxH2$B
z+P2|+TsX)yFGhjPD11D|Z+>7^|2S|;a*WU-ms#dhTRCxr5$V?r_a+xVmBnAZ{b>b`
zJVpKBY*lsi1%2p|d&MDRvp6`L&XN0g?XMqGd&0no(*R}l-i<V>vnEo9@{Q5{Q7}t}
zM>xvN`qmAFim>I$X72-GwQIO-@!X#-xA&VWuMD{27iquC4d$*L@vqC?!)o;ZUZ+2#
zKu1V~^Ufc;jPsPsL<E18k654U>&G!F%J-O?kWDfeqDN{y9cSB!f;dpmEE>w@+oD1V
zoGDpg_Sl_IUm{V8RVD6c&O95sz0GDvq-KUGTaRwTwoU2DFP^YIpJ~H^9rsg*W~41S
z*vqSt$DTI8%lgukFmpiEjuc%;7z}>LoM?S44Awgn?a_=fpLu=~rCneB8_-?{bHqMz
zc$GASvbh&)dL6pfMi)EJ^h^^Fd~+2my;YxG<w(b~JH2oNJ_wr>6|;AjSm&+()aD~Q
z-v=WTQ9?agfJ>DGz3h?pu1Hi(6jU9#psBZi5(0B;hGc}sb}i`c-B?IL=`W<0Y_W=U
zog+j(09LIhZ8sNIqOKXWKO?50-DaDg)6WWMw&`$zMjr$Js^!@le5Ug4>WK!_Q(jM!
zF4@E0l0I)OsF3Z0^e1!H22`x#gz#Fa6RL3g(F4rIuc4&sT+uqFGujZ+i}On?LHd19
zLp~7KQRr7ub!itWtYb|QieL+Q0VP|)Buw|7W7kLa(j{n;#z$rtZ~2MKk^5IDzau5E
zA*~zAk$70X;ab~u*v9t-!JLsKT4z*YcXSPxH&)8%#}OnzdYFVS1=QJ?)y8-mlWw1c
zH&_&#luGKa(F>RI{`7Y+VJ-HJ<1T_7F)rKaa6P5|I-CBIndwgTZ)1^(+;khVM@}&_
z)+sI`8P)y(AexVUI-N@QwG+CtaAWa$Sk>iQf5+X+ug?)kpqMME6yk^syAOq7rz#n}
zqL)uzU1c1f6G{4otU4?s_`tbkv$t*RD|lPP-H7<4+F@}VUBr2_*j7tc$g^w?2d~mr
z%D8+YVoi>iE4h!z+?xKmiUD1+?IX>8*V+2Z_M@gMn(Lik6FG0p`$u_D&@2K+i8Y7B
zeHLtB$%gI6t|Zp45?JsRE_FdU1x{(3KjW^xjn(wZY<UpGah$S+>UW37R>dDpKsg~x
z_b`v6sQRz}nE4Q|c4z!U$Ie*+(Zbx=D-W`<rEJ?K&&>r?Z~fPWQ&Kj0grB#vuSSn3
z10rYWwzvn@A21}twon2O8(_D`^Im|RB<}vLQuRO66TtNTEyzJGC>eF&9YVodAQd6-
z7k(Bn&;T;!>gf=TODEgZoHHcEg)*GPm(W<M_gtRarQmA)x_oh)4rYOS`;PXv=Z$CP
zZ{)mv15=ZqCR(6AEqo+4LbCBjtbmcB-Gu@f!Z*ppQGaEz%!cEmNifg3G8}_Ug(cQl
zl37Z3lIM(VXQq1_zoiJh=aoMZH`7CM<7wf<$-+F@X_`6mw`>YZ<!8FZ43oC`qxKv1
zSl>rIpz+!|?qrX>nV0Sy>2owkm*IK2XIM-$HY;!rJ9~;J8|%G<!SO9IS|4Xy!=mP3
zUib#;UN=+Ta|CTQ>V!T0&W57!kIoFEu0f?p!;_L?{B%y6!N%f{QrZqbm8MYv5exws
zE2Fzwu;vnwwzpkG65svIY?rQ@fyf~X+Pdfp52NFh%1ZLr?0R{5f<i5{iM}iTN_-->
zItO;!f=dPX3HwPXP<#hn$Ej^I_7rw9u7*(`LDrNR-*6;FLuU+jTc}$&jXXFogn1!U
z_(Iq~`vZl;!8BOk0}SO@Zkxl3udvY8BPq}8gu%f?<okh*pp)Av*A+5svUR`zSe|jO
z5yEcKT28X}QLXfZT_I+-jPGURp^W0ZYucl!=j7>nE7;qYPpY*)92*;EnPsLYkMZhS
z^JJ)bIQxH${kd}ZrZ1;!_q7muxqz~%>Du{QBN~_3t!c-j{F&d}Bv{FQ@{!5Su*LLn
z8Ag$BIu)KxS+Sq=*G)G1*}$prqXespFd{g$&k(!abN#piOv?;mRA;Bn;Nr{*(}~IF
znb(7k&u{jYyarRq{7v>06yK8zOPVh`37rD%1AZ_;k`*AkPAcOmq(92>j`@N5l+RH9
z=Yn(KIp=%QV!)MlFfb{B0YjyX?4Q|crBvFsDF>hY*?6&@5+TVhh@5Aql+LMEPU5e6
zplClwjQlJ+G2E!-trZ3OeO&ghEyTxPPd<tXw@Z$?Bke7^<9=lxg8T4-H3ZclMbTc)
z6=C&c@sK!zeA9F))HCKFV<YzrNzx@pA0@sUxgdXSq>w;D+;DMH3;@H;ob=A95<@-0
zR$4_b?kF}yJuFsNhe*Aur%*?%Tsvm)_faO6{l<=I8Cb8^IzGc~eq8o*+xL7&`}fTb
z_EtT9?%n8(7Wnh8K|KwD^MoR(kVQEj!$!}VMq@r$4Mi4<h}4M!T@xxqI4b!SDr81Z
zm+VbcNo=x}=aBTdA0rVZ#6L_xsm)I0nkjqpXSS}{FTYSBMAG{k8|2&ihO?p6=nFm*
z5=9CLPZIF9!~LXhZk%{YsiRdC$koP*hljjayQB$&Xn6_QPuS|`ZwX;Rbty9{3qPzR
zzKsYIzQ~Y$QXzp1;oedZzS11Yq8-jftnlmy&x&`Y=xa`DHNM~Lu;5D>&=3n^ZO_1`
zP`ANcVxO`-V{z_J&5NaLu8eby^0mHTi=cWwFe@&_64h1cgd$}MlF*AlB(KU`pR|JG
zr!gVzm=C?1Eq&=WNnK0URW?QV(ZVgl`?}gI{2kVG0=S_&nwQqY{e)qqPV01OVIv1%
zG{Vpb?m`A%nVE@pfw1x#eP)k_Qsa4;L~M89`WjDV&6K$-6OseZ((0+4*z7VFsV5RN
zgx*j$c;1MPIG<j9^%CFAdoG@iZqr|%<FR)QUa9rRx-x0yQaZ5D&(@v`IC|b^2ESoE
zF-kq6Mfg!JTQkP!uMNOP8uI){D6*r^2Rlr0z<|Gf6!P3Ff1e!-t&5HO9iamSKlF#M
z;TH6j!xEz`vNE%ZO~eA$JI)C&@r*oKw*(VqHmmV-_5Z|pfn4V|(`qm_ZW2x6#2+Yd
zm6sRp1{x*V&yF|~vci*H-{psgWyjgXs|v2{Zq+j%K;h&SY1xiN5WW%I_MCxi%jp$n
zo(gO-$M?AhFJ94GZgCR(Zpn*o^~7*SqH5GZ5ijag7hzT7{y1@enyyr}Y=Ay8#wgT6
z69G0rvU<Sflhra3K=@bj13HF77vlbj3JyfES6<8UMRe%SNn4o1pIvbOGwF$EGMP&e
z!^TZCD>hJ1WE<Xg+=DJMT5k!<*gfxi<DJ@e29@Bj-s~?b=Fc-@byck{`gwj5dovfF
z?=Q)y|8wPndxcJs;IWhAS3b>uCl!uk$Yho#)c1>|eDqxDQYEwoO|5sD;NNi$4nKK|
zb?Y1Ic&|Y7Nd@T`RZ?hrWVA|P+&&X^JKk7-$hw7nNg&49G(7jDr(C-vxweTAqS`$d
z?bSvCl@BiuZGf6zhU#=0IArG;8-8&fv7miNRALP4E01EpLLXgTymD8Ahjea<%bFab
zcgaWU7mYgh?3xPMwddrdZgU+1>XCsWDd%9L-TCd?tB8`V_8B1gzpO5C>RLjf5jbZ`
z3VlW*sB^7^8vAZK?7q5x6RPN^HlZ!OC^GQ#BwRIX@rKJ&q|S3#KnvjC&~fRTmcRn*
zn79QorTMmL<5A)gNq+AqJG<c&ApjLr&et#agL)z+0^28>;e=eNT7Pg^dG#2SiSF%`
z!dmLc_V;|uQm@DoLPG8@a_hU7BqzK+6oIN#ulSMdrx!7kDK3hYESRX=n|D(x@0vK@
z4h_GuPaxvpW=7xfvHxDv$B74iH%>oBNCrooUZ1EGkV>7t4#6&(z;r(4I|994d)t|w
zl^7WLDKcg{K?U%_j1BFNd>Bp^%nFTRt-%pxC)K5@rDp3AP>8VLYMP+}gGtoQBW5h|
z+R)<9g58=q(fys`^~1zugE1^SELJVq=`_p1AAI2;uWw|#Q&96O(6{?GF;{*FdhWYN
ztF37+-RDYrV*@)S8c|(+3cWH9{#p+CT6Ufb12JVf$?Ydv92${aTv1=~*eUAXy`P>3
zJ=?>E+s##`*XJ%HLVOedZltoPzC7mLG^oj3O#Q^^bSfAq$IB;M61(Iq1Sa~iL0<Lb
zix<1680YPOEJ{QC2qB$B5UGn0>%Cgq5!s?ZC@1M8`@+X~b@4GwPSob9SD)Z}-|EG8
zHbGvIUESHc{D=t8#f4Nc@wKGhR`GRW;5DyM5^K}7d0i!ec3q~y7(v@|kyc;B`*-z)
zgDu`t^8DHCqz0|=VE1XJ0n~69tIVqkxZ0OQ8l=D|(XBI-6@pAm0t;vId<DnPE8Cfw
z+zXZ~#>>VZ7WV^rEqEmSQZv$QY|defzx8ym0UARFExC+xDJO3o{NkMVwF$~ecQCSh
zd*8%7WeS})7FO3p7>oX02&{bl5hj>X@2~qKQ-V@roI?HN&~s*jLfcHi9z=^X{2qSp
zCIQ32OY~^+f#UB%mEI%G<?I+O&x7^{3=lhWX2l~HevGf)t>Z@C6rn?7ljI9~(YNa#
zmvLT1D>QP!G+%90O`)fpQlIXjRNp0OtsFnd<Dv>ZP=fzPdaF;mYV`hB1tDofQ6k0q
zuZ<7(ehL>c#k+(!mX<K`Os-Xd&suE$c6Jxu#y=ZSPEP)wzVDkf+Km}JenIr#$~bsb
zC;LkxiA}z$34754?~J{oD51L$+3Euih+M_4eh3G2C(?um8#A(5R>st>qB7I5*=nZG
z1wp*J{TqlQ=xMdi+RRc&<f)`B-bsGQ0r3P;%LLr57F78~McgMpONil8w>reA<}5IH
zXLUKw6a4g+vzTY9V*6NHNjarvkjir_#0&|>nwdXwzeooJ5oeO4f*q=PWfqDU2GAD_
zHZl;D8dIHIK`X*)m)Q<jgn(CbF%_jsxS_N;S@S+sUlmAu1*K}BZsh(yn#51s$rsb(
z<sxh-X52RBdPCyoR?M4SB$Br~!;{CdLpsEcAlYH&kA3sZnJd0}G@$(7iN9JSKp)45
zA1&L`)L^xa{ExGSkL`{OY>occ0A70c>P<h!Kl9IjBtTh$1dE4FEi6>dI1t`_;x@AQ
zAs&e1cXy`2ZD+sxUcF!GY09)m)ogJ0-$<2kN#M+YbBa|GmTYExC)Ci2I`~?0*@T&m
zwpoV?L93}D$SOYVpV$D#@jK*57HjHCxr(|k6Ac}f;IrahfyefS@L&r_w?7a9;P&xG
z+3?*LXhDv2M`?b5;5;fVTT4|S45H{q8F=Z8><T7qgN3EmBN|}J6f^Y2I|B>}f2_0J
z2zmB_=Z(~f9NgS6-<!*!EkT-<m#?^<=OJJJSY1;2&r24B{%#c=pdh~Z{s{?bG~>ID
zo!nUyj|w6?69hX1dlg+F-j~v`-3sRv39!AB3veT%e6T>Cb2k$lH1o%qewj{;9qm;$
z7N^-Z!v$-_jA>EbGjLYQT53vq)^3tHhGeTTPz{$&1Ebl6R8h?{r_xAR_{8^2MnENI
zau?Q%MIG`>w|Pd7s&DJ%PA%f_YFnd#9)|Y^g=5%Bwt$<y%Zk5c3@A5pA+r*)LE%`u
zjd_TT`n!v3H3;2nA+6c0^M);;D2Fv(lQs_V-Ni0eij{KSi||rX@yE3=*A`MhQ#=GV
za{2uNMAVP&g^U3ao}PDU?!U_1lEj>{#lKQ-|NHe*@g^jWmwMiwLY!EG_FfAutU`~=
z_)#pj7>2k1Jlc>U2=Rf=AIOqwD$(&uJGI4`8XXNLk6)(z5!SXQY$FYW1hpZ4z{&kd
zw#6)aQjB2*VuX)4-q(om#JT@le3w!{LXn-!iwWVse_o&0M42}*DvbBuz!Oh!<fG0)
zU^fJ9b;xN#SZ~g=MreeEoT0$g66B?5GoD0I8Pd?J3{BbMZQ5o^v5VYcb3J`{x9~^j
zRI4CU(9aM<rdLT6(O@9ViSFKW4fdY8h^A(cRzLvVXpk2uB^~iCW{7|2gT<5_1t)C-
zH!bpyqAz}M_6KVqpj0G_O}8m_5}o-@fmQAPXw~zk&Nu8Cj5E$8Fl%&^v(SB$$dS;J
z))iS!HOPmH<187TXYV&6-$qykd2M0iK7bC4)r?crzB<f7NY=q%BaoXyJLP-rqT5~k
z+<zcNI#lQKjbOnne-N&x5Obzf-%FwGa(&}>-4Zm3>nFO0)Fn2&N}5qsUS}%t<olVb
zZWRhbWA3&7r}WuX6SmESMojOLvDc4TDug4v`Zy3BD@4lzW!71CK|3kaTtK8E(3Gv{
zsoN`Rg#*?<*HwBZI5(qU^~)wBBv0$HeWE^MD{HFLlcP|FY?W~I9b|?$G6ECirTYIc
z_8#C^@BjaJA(8DaBe{+0rm~Zf+g_Djh-{ML77Dq|WJfe)kD@5b-g{+dgzT9R;kI}E
zpYL-{XM8^2^ZEU+>s+VXxz0K6_kF*{^Z9r_#_IW~dssV`=B84*kqz2w@t!V7teaix
zhGop-5#L9JVYekuY%v(%#yKXwZ2&4P23xUgaRjti<v&0A6NPfPs*N$LgSmpKxq~$I
zcx<*2N?F1VSBoWg-O91#j<G9T)L69RY{;r6W!cQXPZ&3B5C7u=M)+>*<QK+7AFn$7
zuK4-`L<l7pV0Q%9q;qvw%Z9#&)loFc(?|**1I9V)bSCQ{mp$m<#q5lWo_<$xW6Ltn
z3Sq2*Tg+xB+Nm>w^1N-)xW9K~$%*g5j2-Z)dSqo5mbXz_pZ@z=N6xAwz)~6f5U~TF
zuS|D<%yp1OQ2}DE*9XjVBhv^jRe5d8aS}J)LmNebY0LEH$9-vUd7F$l^L!aWRKV+Z
z1b;$OSrHSL6k}Ga+{Cts)^2^kHIU2R$?6Dxx9;25Cg)V&U!RHRp#=G5q*tF`6wdnq
z0~u5o>g+TLTdiGY2*IHbl9-R~tcrLO>dO2MkZ?`mhdw;R?`TR<9A|DlYcw*8ZV?%{
znS8&9i=RVRh{mLC9K<MZDNL6OZ0_lo0)jc{BT^1}8AIViD2l|{@0%T=>tn@@*sKG-
z5>p<Z63KJd#*3G-YN;uv_Rd75X4G3t4asv~ie;Y>q<Q|x$ss9peFifax;9ZLeZX`0
z2CAE_lbek)6ehzvA`wqo-BS4nVI=K?uh<!_>Hg_mK>qMS#{kwa<<gNF#JC0M0T%s%
zbjKgy<84;oPQ5-i1R|ee=aIf9Ofw7{`<%C<R(0a0VI;j?5w;Sy^2COAQJyv+(ecwo
zGbHeK@w4zFG8%_M{KYko)-NV9+>fegI!o{`otBJ8O2fGc$#yAT0ld9ch!jLIaWWrB
zJ-$F+|IjPSS`-=Wyz|3R1R?L6WER~RhE?hLQ6;lC?{C8hQP8n%#}<YDZn^hI=S;D`
ztlkJvY$>wGkuUw`$W(Z;6-G|;rbq8HKDzlJHnO_k_v(k4kNl+>gVW5WHT7BfQG^SD
zjL$FTt7oHsbz;A*))^@xSQI6Wf$6kGYY~KmRm1?gKtId=jT$=TlLWzmLe774N!!)P
z`RJ1IwHa@25Pe@OQ!ehu@Ls6vVn?T%*Z8oRhnD$n&H>%bxm`1GhqiG1Bl$~mqOx~@
z2xD_v2$%f=mqO0AkulsBEiDlY<C4R=6h5Oh;go7pVX*F3v~>}xLou@{=v<sXw7Z=o
zNqFP@b8n+mj6U;eH}euk#Lu2zBM&wAkd^@HSq|+U!F(k*4Zl|+t`}}I*KFSB$3&!B
zr4=X__B+2bfc~!JBgWL!%w#V*wm@U!ZXlibJwjl>$LZJJdF0JL2WzBv$7$3QYYUd_
znV<PFQup0Etkw=vQofsy=7K$u`EsB0$kLK(qtm{ZtP?7m$3qHe%WT(*%T#1vsqZX}
zpV$*7vMq~(+Ak6UtQG{Hu)VbgtDDjGY%vU*^K!-<(kp>KiFoj@LLE6LWRP>>#io<}
zKdL+ga{T9CWG^p_afz(a%akpzWqL^xmZ%?`l~Z5Wk?YPK_1wX4UX!~t(UfT|heOqh
zJ!70;u9La+jCwzoO^M@sl%NICxaQKG&u@8wD)(VQoKC~a#YkQ=SS^p<Q2l&fO2|{r
zNhMXruACOWIkc*V<XU&p&77aSAQshG0{ZG8r_v;`hVL9GCY*X?xBjeZRr^7M9-A|0
zlK&T$^UpW^`!mcCfuwoO<Txek`R>}oD07$RI=bI-iH{Fd*Rm#+fl@1(;R)HuawFni
zxN+x;cvx06ymxN^IEy7U&C9AJ;STVoO#)cB{t*k~`tUuILQAa81o@&bN}UWitgEkx
zEf4XZGP5VY+E%PqGsiYDHeMT`A$WcJ+StKH=N(V14QJWOI1kpE@267ebhLhbwBAxD
z0e4EA(3$!Jclvz~{rBG=YwV-31Nlv24dfLZE*NpOp^0|`HTX_FnG>0f7?>PRFTQ)t
zigE!v1fw!R7s8l~%W%b|yg4mp)*}VO&ITJt^%@+&h)Ra$*3lJC)rBUm#A8-#66Ca1
zHls)rp?%p`A>MW9c<Vf<!@9ypA9KjpEVuBF=Bb_?R=J<`mCR`r6u(LwkRACY3r2pT
zQ(V14g(I#l<&5f(`lt9vbjV&F5tx^kxLgAnm%dH9YpI7c$3t=DF!3oSBOjaO%#c4X
zr!<fTW}cRv=l`23_P-yRorslobMs4BEiBukP=cOh+<Ni>4CCj-IqahZ%;chgh#=9Z
zU;4*#kl8>LA#^67Sq%`3FNhh|sICOK;DKsB|Krjd-@iWL4jNcBa|avG`~G#7{kf$6
z{$i4@2(I_qD;knL7CoSSW&qoTSaeJN@1?v935Jmf7#6p|F2Y@YG9jZE9JYE1Q<26Q
zw|;%W&|}df+v{XO$t=(R<6-}LgMa@hI@XXkyN$WL|F0TThGaXoYG<*M4OGZyKMG&*
zV!PHbcbO<Bh|iYkNV?GeerHSHXGWx;X05D+ghVEA?I|IA)9`Z5Nl!ejop2wQggraY
z@vm>k3Hf%^W13ZdNjcdgt6KOgkNsFl{JB|zm|Y-xBJ&03d7fQRm>vNKgma80uL~3H
zDz<90uc$pY7XG026V3=K%Bp(dz16M&C<~9bsTWjwn4zE~^V~!Pw4^lXq0AvzPLBwP
zAo5$34$kdTKKVXK%1z%tcp(*}uHvo0GX17#t3ea4X0Kie!Xq^g<7ijY2L6go^W=oH
ztpf6!HH;M<7X5HDjPzk7mcWofl2(W-1zE)N!ljggn#O6#l>P2>ugifA_vpjw&((X0
zTdh_{TQI%k{mz~(&6wG_MMN(4o9G&$T_UqCHWxOY_XDF}kKDRcPW{=#PFPb!HWvO2
zw*tjv&&Qme?7Uw@;6K;&pD!MgfxC->Qp=y;mo*1rsevs3>gBU#tDjWHeQ2b}AZzV5
z6EHy=@L(A7#cZbIrMLS<vV%)PcVWG%(ba_Nj)96&V*1ztJ{b4l#J<o`5WoH91rT!f
z+PskdyE%$jnK!<VJU|t-*E|E7HcAwi`l`iev%xEAYWoNs@-$Sy=d@4Wqg7vkZ8MIj
zuz1d#71R`2g3rr;(Q~7e>WO2&YsVrCmIkQ2#}M(by4X003Bz1v{ZSiVa5dv?Ti=uf
z=Y|?+>OSj9Ie+i72mfqeSPo^`IK&|Z$fCBbsa&(%@<8Coto-3x8Nq&3c}Ns$S%13Y
zO1@ts4ERdFT^HpaY_q%Eui&xAI{gWRvvfrmR;O-AZrCxs)$@Zd^TL%`r>q6HWbYjE
zfjsyVD{(9T-R+3?p|73(?|lJ#L{kBV%$!2k_<pTqO^W?BaiM6*^H?|#&7j!kBb36T
zk!WMl&?VFQ0YYF`v<7xbp$JkThGGvVyiJ)DZpW^W)bRmXKGee0voFi+ndH-oEONOO
z&a4*K`*SvsVWKs5HNaNmt%$4yhg2!t>?~@|O(!FL0Y&nKlOaF*U2C08e;qiRpp_mS
z?IC>ED?q!n?+i^#_4ct<uYnudObhME6QHVQ5|Y{W-q0RwP}XMX0v8rq@cB#h1@`3-
zBS)46^Lj)ie2OBY&{~MOgIt>$=`902eu6s&DnkE8^!Jo!!(}*$%8Qn$wY7V3D`dN&
z(mz~&e&jO#`6&%pj6o^gbEd-^O2n3^yj$=m-uaWY{xFXLtIFJ;Yx6%}ya<FLqvG{E
z!(Utdg)@<hE;2KeJK}9ppeemsO;ZB~g#6x#-e$CJh;_PEqK7j>*Wq&}0ctp~AQ%TJ
zECuL_ckXAWC!=$MG-AR!#t=lVfvEC!7P)oTMM=bFjZiwe*Jc(6mFYOl1}PDHzD*DV
zFSZcU9!8~dBL$-w=x5h^T*yuyo&{il1<qO)6D2YGw#Zv4(J9=)n)#IdA<0p}gz9*I
zqqZsWw{7>#QQu0`n+xJI0&o+kqdK-kc7b4VkP_^54+Pj6LJQMx)7D-Tl~Di4>`WWE
z;ux%^{a%?BGel+iCK=rf_N2eg@qfH9{RFe1(N{gjr{Di^J*XnqPIH4(UF_>WH`V!w
zGo8BSMV6YxuLfvNJ{uyvo%YPtea>L?5Tc+h?{FCz1QKg45ONqimx4_;g%o;yvw1H4
zrf=Qutz1YauMJlIjKmJnQEz|--P&cji7VK7<Eu`~<Bx{s85)6(=BUK)!;)8Y<~U+E
zU-72r?4C&Vj!#QQBTqs#SCq=6XET%Fs>;Vx`chJd^sl?-9|fEDI$S)xQfF*`a|nV>
zw5LAyQFN4&&13ZmTO_Toul|WwoC#0Kw06b=*!&S~@BXk$xvfGfRl%2yIkBAhB^A!N
z2{<SgDtX-CtHEldH2F9wCPZ<!15r;@`Um2AApjd~XNjC@FqZLkUR795B}h7r*=k-~
zrU_M(e4|nNwo6t-?6O_6EVu5g#IR%6))M)LXf>ouJI+IZ8}+kky)$?|=DPK4!$nU#
z{*`e1fQ5@Y+%)YsdaPFFsm&WiPPhvVy`@<Tc2?YPR=o_-OWD24=Atg0Nh4V&Mm0QU
zub}yuNL8(_;C71QzI^Sc*?J<1e>!%k>sD&K$il#vLR4%cnhSj<flHG5>9xoCRLvKn
zvjy9pG)^EEPUHCJ%^8CI7bTBM{QRj<DpEb=^f`GD_h9?^k%x}SYL1s_*2Oo+e1=&V
zyPy>^)q|MW8rg%ahoM#CVemsdGI=U@m!;x2!Yb%@UK@!*H;G^&#gA&kEfKh+2wdY<
zoNam1ulXw}{gl>HYCYFc@G;-p#=e9t+3VvR%NM6ppGA##zPOJ7E82OuH_Xuk6b$A3
zU6#0{1>o(a&WOa`H~GA-S1&1Jx3S+Tcd|%OC!c)z*2a0gdy>rqB$o39ZppXvg47ou
z73n%8s+1Cx-R>?nv28qWz1v=Y_k!KqXF?r6e|};6Nsvw$7t`JLEG~2#)Wal1yHKl0
zS3(0*jE{I}H;&Kdc2FHS!VtcnZ;3fXs0@lc&z~l%$MnK8BYux#`+I5h`|DLnqytRI
z4*gZJu+o|Cn*w7$3aE*n(tLggCJOs$E{Kl0-z|b0uLdv!R-^$CfE&dhF64Ik+zjy8
zH47D=nsLCA-XUvhd_hYrPWG6uj3CEou?G@=N{|{xQ(&1~wwte3Ow@$Q`j8g8@W(Uz
z$X0HBw5tJ0z)>*6liu8iGB4YjP8^jKr>ma_icb#B;ny3%U0P!sVEE86_nX^8trBtv
zP$S@4jT{B<VJM6Tvydp;s1pmlvxU!O3lJ@}w}lB1EXTohXx&wSNr_nm)@@c3CRVI#
zb?h_Z4QUU6w5!2)xKhzKHoXs`3Dw02k_px$g5K`Rp~sjB&WcIp)F7oOnERBLF#pcx
z9_<cb-{hckdq+b}FBF~ix~~cvYQ_Zo6murZ**7v1{2`nDv=kc}QpolYFG63a+r-M?
z(vZ6FP#~1{Mb<WGV5c}69nlc>P7e3HqnpCJfP-1`)Z%rvXUm5HMqpG252!*N93vg~
zONH#Mr0yWKT`2mN3oJ~->#}JxjiEX(=8Rj-?kaFVyl2OFr^%c@ZUxJjLDxeAjYDGY
z&6Db?(}Nc!Kkb|y&QQy;HCmfdihbc}>Bl*I6BXa><AFGD2wZQZm97L5st%(=7+kid
znSJ8ZyhIhcL2H4BvMiR5bCAlkUK)X~SUr8%=9K*QKFc~oV%tHo$}j-T{VyC7Jn%qy
z+Yk5Pc&~mP#xNb6YrbYd9eqR8CT3A%Fk@xAQ=2^YltI|{AFv17TpjOdblQsSSCLzm
z0-_q#yFT{uSYVH|)aR51o!7syl4*^a>|Jomf`8K-6`nu1@TjB{bG}$~VuZRq*X8}!
ztpj1XhNUn*$`AlEiU9b|KW5q)QUp-5Ub>Q}8XrCmS2P*4GU=l>cfDlUs`lt>%And2
zLC7HvOOm|-Np<Hok+ckphES@>h80$}lMd}|xfE>2ICfb8WV+1|*M*h7)9wZKt0jcE
zFy<mF6R-o>!;itQT(VM1nEo3o_s=V*Bp9yW;U4tqUldPaG65#Pn-65UeXpcvR+I!!
zdqU5*%}zN^&%AkEW*I;Ra<Z4S;qk7sZzUBxf9JuoU6AYW52P5UBRZpSFpvy!2;O*D
zFI2X~VQle&8&jrcc&F*48I(4jhM|=JW|eFlB=x1kbBIiCLfNBE<G{YfPLNj!aG`a$
zc$id3KoS4uU?pM#K+&;X6HGc0ffR~-0t%Xey{Abxm~P5PHB>SBRFknY+xt*`gI#b}
zb}Z9t&t@aUTbYIAc_61<>AVOVNoqXpjmutjMwAKFb8xnAV7IqTN`3mkYXrABvQaT@
z8zK`RD_LX9c>-!1OT#*(b#-$?BBP~d!UCS*YE}*=lre?Z-7vklsZN4B_65q2dl`W$
zbZ=fsK8ynsUOqDO(sQ5nYUdy9Z+NGij+pn1L(RLkvX7-fT{(nleCK)#YFdKX<%7RC
zwa@!kE(skKE*;j-p-MYaAI8R(MWrWjKSV7PnbKuwHjtw*E+do+oGv&nr^{iuiunPO
zogS^Q+xg_PS9Q6;v##@^WZ%J^+7`6O`3L*9&tVrlV4C9Su)oDs<xECV5=Avwwoo#9
z=Sogv_t2cQ+u^!zLxo<g@E@N#dUJ5<XzN<skEfpY-(E*L8yn{4k}sX(^vqy0<i90K
z^6>p-IQxxy!^*b9LG}_RB%*ay^8D{d|H)o!-yecYTEx@mHk{sHHsV<c)#|}C<_^!9
z!tU(GaN86Pvp6NIg_n!Ub+5w*9bgAuKjSc5BDe`iogQ~>Jlg=9_6ysL{ZApdcuV#@
zF1Y!;Il{ZlR2+bN{}=pH@(xsrgn-n3IHB7TH<m*`25@1om>vhc+`;pPDw~@Rw;jS%
zJ&JT!C?g|XQV_dLBa#b7#naRMkG_(zH~H)sXY}GyPyDLFf87Ot)a@OXBskh@f6m2T
zw2((dg>JqIqEb33A`{-nu6ZR$EiU3a*m4AsKcQy{QEPRR@7i7sp}K@O4q$Z#8-Aa<
zx|>1dgGXbdvd;X*E2ltx*1O?zDF|Qso1Q)(u#WWU(YYwoa=g6?u8K!6j3y{(Fy?au
z^Q;+BMqQn0P|$aJ3Vo)lTbcOe3#~SW<|Ak5Q4S}kbdqj)jD+ByE^qhU*B+Tov1E`F
z=?91vMe%56D|N2X-+l*fpzlt4U`|87GTE6|iQgkwM02Irk&e5wI*C5<=NGLuHBVwU
zxlR1AiC%^0NmM{e3U>Pr>H5@@^m%#2F3ou10E6BOt;C|U?+W_NgX?@{g2;O(CQ?jR
zIG;xKhSm9o22o)|-dog@QlLW3CaQd$g*vUfM6xhmsIGGngZt9k2ib8J%&4jS_Lb}o
zGqskL8_j(RBR_b~#0HSl#lObqsj421?%U^e<+F(1>~t)7T1-S?%qau^va;p@g&tSO
zbhRq~mCt1gQ)bpS`u*!~nseT0#&_5Hp7G4cfK%@gPEB*H1qg9_^*OqzDY{1Hm<##Y
z%zI*A1)tot(C<`WwY)HGn<Oed-AZxA)nVMRXZ?-!;NeyU=B5$1#bIrwmBvYrgrH!S
zmST}kyF2l>xtyRZcW)UG+0iZcd=<u)n*KTY9a4GsxE%@|eGik_<^pBW0g^RtH`aHT
z!_MQ9VizBkRAZXgudeS`F*h%&Bl&i5d)v?U@r)1Nf1=-KI?DNSztH2<AXn51I$=^W
zb9cS4QcX25Ca(m2*94dF0fi!*8x{rj<#J@D>J0#&>dIa;C`x+bmBuPqSXxG&-5u_S
zxdUotHn!!h@{Mc%u|WU)v15^hAe8*)I_~&NLR&A$Zo#p4NTpfkBUoTWMs4KP)$AO(
z&3M2w@Bpu{4NbP`NGRDMx(|Y45l23}E>zy<eY2w`CxuZoC0X;y{u90qc=9Jc_%o_J
zliCm1UlnTKcsvh%;F@>(GxhUtEepU}M(2U%9E6@*5AYq@RmU#XrF@+_*XW1QT4lWI
z2`)t%t@6fk^DCq3q}gv%p;qN(A(H$~UEl9p9?ex!CwHE!UHYfHvy%nd4T-Nut<*fz
zYb>e1gBfd*Ny5Lz@hD>-n$o{K&Dj*y%3^GeKJo?EG785_xK=+l$>pXGl?agKEpHb9
zTiu5ObLZ+5508O=9M2^RmbC8VSX&|yOFwe`1j|7rlQO$&CI0i5-IaEY)+<`)(@)sx
zcagqjZo9hD-FtnsTd<TTUnQ)m?8{f)xonHJ2RY^$r{}c-Y_A;1=$vwR>#+Cbe4~@U
zRD0R$bJlopxtlh=a+Kvd=6rrol$!qF4QU0<cw^}ahulh?Ijzr_w9m=2k_Q?Vbz@)P
z;J35bK5$)~F4tle>HN5%G;E5p+&;5#9E$DZg_bMAq2iO6?ad>)dK~)Tw{@r2iZSAp
zytB#Fx1FsQe$t_8LNpygc-ysaWnO&;*e-G?31^l*+WR!PJY+Uv-W42^f|rs%(HNk2
zOA9|VgFsOx$wfb}{z-@WM^gn%b}4|8olIiiDEvC#eD>9dl}+c<_hvuy8(H&ap&0Ex
z9#+Ku8*zD-Sj7RKZh|#tU#%%Mv1^huEe@zC^FBW7eU-CJFYi(yOHx3P?qGICD?8FP
zP7^6$uj(_UM|io6=F-+saga8c+f!*g)VI@d5xrjQr*b1#>HJibMTzlPz=ce1!F@rd
z!Wy(e0q04{5#?^gv2k~RCdAmV{;yEPnUo9`!Ghy&X(A}imB!hLUy!MS;BV;NS3Bld
z?&dc*v=-J8+<O=;qvOEPvn|)&r}bfuW#jjSTuVPHo!QQhBKt#fYJ7RxbLO@DEZ6(K
zZKbVUm_Bz(Bl5^yom;SRSw0)ae!G&~D~na;xRKhf;<ldqm{2i{jQt8?-lVC8=i46g
zi%RdBQ`UM7U+Dj@LtyplP-NFMr~4f5BQ|NN;+dpi*TL4_$W!vIw3(AGhST~~VIEk6
zC9HQOE@B?|r=FQCh3!9<4L=+_3%zUbsK&LMqmv+kTn~G>d7reCjV&tARd*8aM-xUU
z`>1>o+e=WbyUT28@$=omd7QZLzX8w!_R#XPh^?Rh-6`}wS)STHvtwTKCww(;y_9iI
z<R%;s(JLi?c{IUwS;4Iziy%2gdK@hEk@IilCB`#p9qDdwf{kPAjZuq=R(6&KF08;y
zE4Q6km}Qr@-Lk_yORlo(cp)npmBPrL*Mi#I11HoCXN<O<9S^xm@E)!4`B)LaMz6!5
zV(EkL1<Wf!;zr*~Ea$Al`NL}B1TE^X-)oQOw?Gv$_toU!>x4$&R#w!<AWWX?B%ble
z=6#@a>Wpfm$aT>^1=h93&o-Ci`Oi$H-3k3{IY=Y9`Z%~3Ov)G<JOL`%+6B+$=vKjo
zV;9T#jtg))x|Y7Jw2i%^Rj;1Ornr`NEi6YwD<7@1cuLH(5IW8?gJ|dnx|31fCaDER
z0cx64x64ImNn^S1@>W0*v1@#!{h?EekmXp>(C7$^4ai?EAQo1Oi_<YSe1&S;KM}*X
zXei+f4(&D`6hxc3Z|w`(gGdD}nx^1@n^K&3AS~RG-<NBrnwsF6HfjUuFhghtrp3NK
z7EAfb#sKk4{{{*UUo<`a&=O7jSIJ!xRRvCXjIAs3SskU-axn}M-M$Xhr0zAKX+$G#
zphNRu+*f9bR(e2A{vc@hPGJ*YUY1rhi8Rsz-p0GHpXK-Z&B<f4C)4u?n6%HQyn7Qs
zUjD$36L&2kUa;Mvn#8%{C-@)cq{mLFWdW;DXR|^22c7ESzC1pOT<|^i2f&f9Pe#{S
zXMDFW+oE>0OJi{^(S5nelqk;>Em{`pz{NsK0p?=10va6EH8vkUjKk=fDtr+D8ywGE
zOAs*Nxp3eqSB)jgcVcR$q**~<E-ELVFGFM6YzFFUW=UCEX)pweV}<c~h@TeN>*U|M
zf}%knSdtYcdY%({RZ2?GnwnapiG2JHQ|0X!xk(gnO4OOM_qT!~NZJSQvAS{{7_Lnp
z*C9Q{IU=rQ{uP`GkX)sv{T5s(NN_x=j#*ig)ue?hnrkKRaBkXW{b>R^bDgIbFW!px
zQ-X*YktoqJ0q8B(udTuz5NR8GVPw9e?BW4b+d-7_&Kdz$8##RBcLl;Od3p%#zo3Wz
z`n(pf_fDJgz8Alp8=?ANx2u2P-&+5MaJ;!3mbP*DSfzpOgPiBK5V9_AY<wCj9Bi}k
z!SwEA>ev_5!>)GP1(Io5tu`tV&%b42RHMc`((t4->F*1!X+^w_j66jg9ox~db9CV?
zFJ+eWD50$CY$vF{VO@^<+4tyN-I6TEJKO4`jmKji^V_SP<<G33*=V0y8%UflO3<5G
zG)<tOSQ#&_tNTz9M!x;P?>iMkjaU?y41c1BCQ8-OxTsS1zwwdZ{|P0n_d+*on)aV-
z_7;%~Np*5GRn;rTjVKYpaJRIoK~1e3*;At!*F&cV_{w*Yd~SYuk$umN<0RB>$~Iqm
zlZs11yZjUoG<MeM!gA!3{d7gl(0JQpPkm9Wt+QO<O)mU0%^F}S5;gUGn^S?GqyN(g
zQE+pIHQvkhZFO*9kQ$2R0#4_Zaqk|KNscLS1)JOfcSJndG?>MP1(9n@dZBbmlrOPp
ztDAPKpSD0LZax2){P12@Ni%Sh^<&{nxLrWr7rurK|JqL_qNOaEvlgR0-QsF9M>^=A
zAEY`&j54EL;<ib;Qi++UwgZ`_Dp`w4;I@#5jn|+EAPIK1mbOXOk^sfpyI+P1@at)t
ztG}Hi-<`59NzUd3{dtN^c|QHI3YA8~VQUr3=Bn~D=<Tf9AQTrvgtwb%Sr+3-xe{=Q
z94t;ph9{HIn_o~ho#k*LAC6TECth^N#YU>GMjrFrggh;&>y3}g)bdpS4eh;8&aJSx
zPxL<Zud865R))!Hl<36N6AnDMani5Y_vW$qhy)Vp0xU$rao9&#@hu7@jYBK6>F|tk
z0nw+MW53YnIDwC0`|yd+J^~er#2~!m;c!39Sq&1kbE1i>H9|hPsbMaT$?tHcB-qF4
z1_@N_bC{Zt?=t1dk2wQb;{;{%nr=bTosDUBk$L4P!2t=4)@a^jJa?1rU2w^=wH=wq
zc1PjgFB>I#tm_ww*T^>z%=)pxQ!RZ8MtE&HnjRhhGIuEXpSwEds#T@D6X1Y7Youcq
zE$Qr9VJ-Oyuh!gU`N2?UAenE_e}g07*aT-N&=ndLTc^yLW6j|IdJ+2Yh)FoPvMDZ;
z4G-K-vqo9MEucdud!#2F*@@6jXqTr39s_)Z+n|3P!G_S5nLawRT9ImwbLCoa(?tJx
zIDNdX3$a&Th6C1|+AoMEaZ~f<1%=3wY*pr>hH0NDT{d1Vk(CRBm&fi+1aV>9?R!o~
zGs<`wPi{X|LneDU<NhodcLkkrmkMExujJ!ye`72B2YwQU;3vJ9(@*|@>Q~wC3*4TH
zwo<<VQwp|yIRcl0{92sMj?_r`?i=0yT<C5ZD4U9tiz5K$z0GFqO-oS$rml}H9gK9v
zm{!d0BOTjaJNdwcKGAaCxl=R=Y*HJ=?I-(%*!8|Jnd}Vo8s_;4d@rorv!&V3N^Hnb
z-(0SNq=ZTO-ad9V;@ppGVgp0llR=YAWd>l|?~z3#Yc#@+-ntf;RwqZr#ef#%w0Aga
z>CgIrGOFHjt>@~_9qh4`Vs`c1jwj^i)w<;4Cw#mnC3_9;o^*LXcOqXRkX8IP_@<0t
zy2Xudhq4yR(&EA4cG|VQ^m|UrQ^3Ln#mDXw8=A(v=L#_mIe3JEYCb`JdDhPgRTir0
zHE<R~u@vY2^^A$bes<zm{Hchu1H_e<xChwuDew3Cn>hZT$TS!2+#=XF1yqI$JlA6e
z1pv>mBz+vbY~70i39qZ8k6K;HE87RRe8~|BXPDFpo1O2!_i=}K`mFA#0=hSReGm*k
z-)|gQ#RCP^s`hTWU;e%Jd_?uJk*<g~dfOs1;EWYbGnQV)<)qn0W(S?0)%u#iX@qc|
znlFFIm^F=_dXkU<NL~*hd7-Y|mZ)&I7tj~>V46o*kV)YW2b!HF*-P6$Sld=pu__!7
zzij!2*?u}Tpja}>dp9wtP*>{OdGHRS|L(JEo<lqQV^`R9^`=)nP=Ff7i%-Y8Fc_M8
zAD~Y-*N2I3d}fsk7vrYW<n~;qM&|mhveLVW9^fIrS&lC`hLh3XEV=<kplYO4N4qV&
zeu?1!i&(<sLUcuo#I5OXE#n6g;B6IWHF5^~+PQyWsaBAyR?w+d*jHV)1`a%j(Qe-b
z`S<Qs?hFgy`m)gH1ltv>Ni*HI_+e6HYP?$m*aM@~N_4xhZE1%b#Q~Nf>uU^FJXrfa
zu<-Gr(msH4^h4s)^jxY<i7l&1TOasMQDGiZfUAm|7*JGv$UnPZy?qXR@x4ta%CkTR
zbCBg773L@fCW!)rqnMiK@2k3+Rh?!!Q7z!UNig+MsV42)6-6hStP46EXX6%TVC=Di
zd?kuA@AIHDms)N-ZU(3v4MvHAAT*w2XTeccMk!Ksgs*t=si~PuEfOt_D4C=g8p`39
z)dC>Eire>5ABje-xDE4HGq8=M!rVSIrI@X?09-r~#0ejE_iZxF?bQ;|LkTga@#J+!
zRZPVlXFBTLOHuF;8i;YR4;8jx=qt#+DPTSulIsb6;I&AuRK<2rCNen4%ud8_lVOA@
zV{thb`@k$L{)qEUqWpT1JYp=4y>oC0(!|$Gx++8jj<wn(-WBfjE36Ds@xi~ueP`pk
zd5|J;YTgu0$+C;f>RGXp7$DwhPXQp}61uyqE~#1}#8{W5iV7LuWL0<Gr+O-;%;WA{
z1f%;}DDrC5M28RK#-yUIM&q1fro=VGty-U5k23RTj`^G~(UqartrG@=PnNuO5fJ7$
zgvQ+LwiPdT2Y>(7ALq*7K@B7HY0*Hh*EB)BOEF<0<>M7yfiUH2tkD6~Wya5ygSf#u
zf*Cw74Q9;VKzJ!PG9I6?vf6}E*pNDI@YcJ)w)sl$+HnKXsq_J7m{(=brYWax91iD5
z@h%8TbJk&OX*|AX00?3FZwTR=S`7Dw%m|B9E`ksWX~mC5B9b|<W_2bT1#9gG9>QPS
zPhnoIOKk*5@0w-Nggo%jrf+R2{P~IW)#(l5pDi4|z~*U4G=b{z+eQmk+5>!js<EX0
z*%l3?JDM$tD%>>fho?TCyS`X_6~<!(;34<|EPEW`lo~576y7?sP^ebLh@h@65xKPO
z)pZY+2tQiNxi(gjL5#cCf{s<pEzES4)6*Thg)?HL?J+}M|74W@E=v6E?~aU!cYeC0
z_R&Aao{cAI4|pQx_cR0SujIP}$SBNe&4YOvoQE8g88g|nTQCYbfsb8P7)nmDr=7k8
z)G}?Es#S2688{f;dgftp80{v<d13wsd^dw@Fw3$-z8wpQi9L?4n?p{4EU+=8`YpHi
zbpu}-%1;J?Z_po1HN&d2^Qtl%lzvw3^B#u+{W~saQ18vKDO&P3C#Ce(;LG=8{oXGx
zBh3Y0mKf|JmKrvN+X#*M_?eZ8!>!K}+ed--Vm9irvGxt&f*%NNjfxDZ7sh|Eb+7G~
zu5+g(L~l?s-Z6rajTxu_HEcUat^;^ZP!5Sw*EeJ5R3gP5w;`2h-fg9365_sqzs0ih
zdD`b`UaH@TyW>;^`%DH{1lxc)_)O}ZMr>-GJz|4R_H=)96wk5zL2763eL`-yg7jKU
z+NAeK_Ihu5RWjXd$-46=NG>qA_8q+YMI`V#FONms$Gz8vdnXC$xoBcr6lILT%+10E
z!>njOhLU#0jBg)?tx@6G)YU2_8P8@NOU*NG=i0g)$4$pvs0^DC#$=#)erLC;KMH24
zgaOduS8a33$>uNx?I{?{4=Mr2AW6l71>##Mo%u8Xo7#vS?6d>aC?}zk7*yceP;#GC
zNNf&Ja3Aw&eB^d+#eaI^C$^SAhhj@3;GU`LEF3hK3Zy8eNjc`u?W4x|{bcm?ZT?t1
z`G(uIh_AzU7lA-C9)E9S$?Z}?;^qb49diZmK-%J~Q~kcvC!{!=FbOPQ0;t`EYxm0i
zCY-2;Qt!M!Q5o+%aLS(dA)B@3N@v->(8H>#=)l>2tZ%OI7X&2Yoma(Qj*>e{a;N-c
z+2VL~MxpQY!#u{qvUz5inzJP6dBXaAMi~m`BM8t+D>-1f@ue@^yY9u&k%P=w1@6EA
zEcSz-W<DG-rB*T|&iEkNVmQ<XKltJ2x>}fu3;^3hkgB&0bwMa%pDZ#)x=oJ82m99=
z@5tcl_Zg{DHh$Pn=^ZBz^Om3=InorXz%oVNtrs9eU(GtmsVX=0gsc1dpzhmYSsdw+
zxA)GEzo%^v5)?Fbfk~zaiKok!%p3i1!g;>{`GKddXJwA!CteEhZnsWVS8<~4#H7An
z;A*}fZ_qfkI#KV)Dx!6ghV%kmyLr2ygSxrfH<7wv3~`pA;HEW4;TPP39hXmEf5<?7
zP{o@D@T8z5ylJfRkRbPVbe2IHxNYqtviY^wFZZPw30^Ji7s1g*8i*+__I~QMCt)x&
z=<0ofGoXBBdnKZx$g}ReZvMQpZ4&>i)_MZxXYn7@uhcaCPdoQ~6p<Ys0A`j^Sel4n
z2OxshD4m5zo0<nWSe<HfN2jHK;(r$T(wy6@;^x;W2CXZENFY64F3R|0v9i+R!sgTI
zn=AX#CzfWEn@gvs^nI-OGMHW3ZWu_dPT&u+Ofy>pG}^fL7!%>+x)SHpN7MdkD0NAe
znr(tG<P#vh{L2&WziNURbp#%?Dt+^-+v`w(u&>^{CpS`Dh$x#s2((q1veQ=ketIQ;
z-}^H+!?CkBRK>PulFtOpJXcdaKc9H@GD$`!@##4jd-rESQFTgT@n^t{qKlfCJW2lQ
zz4vu>c=vqV7+JUNpw<{yEK`xld)LVlkSffKys=KF-2LG+%0lJ-NSo+iXd(OQxHW(_
zwcMH)@T+E)K!*6NTPN@`Wp@O`(6uaI#23NQAm+rJ&}#<Q?l)TF&|wi~zo-QY;Oxql
z`~z+;_y*yUP}alwG-y|{Z>>OiP5%n;C8IwL+lbMD`RPv~ETV0-;8I0KqvUe#Jh)#B
z1#3w&DhA9wmKiSKjh)HjD7je(?>g<<`94LeAv3Mn3enjtrUA8oY16<CAsn$ocxZC)
z52mlTIg#&?6T6@O;=?*3d|2v0X0l)jV5Klxy*PK}m^U#Ewxd?6{vpbF72u4}-0gT0
zva$Td^+7wk)88PRraSwpP`?7I`Qjb_jy5tUaxSJWaE&`w0E8wn3B@1upH=~^xRti^
z4?gUR+k+8mj&=rx208EyUuMtu@$D@PX8>(&;r-i7n130Bu@ij&Q1Q%-<a2)<8v-v#
zq~I<_7zqB6#;#3@o`-LMKmM8!Z_(HT#G%~`O44H;LbGFkA_xC}UU(?O_gi6+9`^?U
z@r95#fGuYcPMqX&G(?BhKzdA<MKQy_;T8YqKUbB4Hy`UhaQWOHkzD_^^#21W|BoM|
zLVy-2%6u**jq_hWpa1oA0$a3j=q}$<=~4LCHS!N0*FQT&1ydn})^)u*D(&IF9P59p
zGX9TGplVVcKGvpGULDlG?yUcK<bV9=-Gor>o@rU6o&H~s`R6d|zg|3}fo+OOEYMc_
zUyu18PyM$aF~{L#Qd4^LHto-+3+EyL6Ia3=qEgI|vOn?1YHEws3(4(}Duf6_fZ%B^
z_x0RFD&RRfC9*0r2g`WTYMOMhg*#FHzdwYiUE~|z%71J9Yoz9(GzPU5Rba-aThOCZ
zBavyYohQ!i0VB}&Alilf9q_l!02P0x-EVMY9Xyq(^Lil^?!$mzFvK=^x<Dnw5|9Yg
zgxXXHjH-7xLLsbk0Pq-V0Kz?%mD?nF@*=%4q*w`nTdp7Tji{G4X*)nsKF6*ed<%{T
z8ldPX$Hn8kzRe-AZ8D{=x*#2aF(t-2Z8fPN<zP-G{>vlL8Uzqax#N5-EJf8i3K>?_
zBIE)kb)-9qx)B2mdJl2$`<qa?-FdPBSt2Hk^coWLc_WEP?ib(gLd%yki*00(I(=s3
z#`odgy-r93W(Dp`G6BhJ$*o!MGWKxrN?A!xJgO1!zqXUN2z-RSq$g~D?U`Y@`?6C=
zItO33jJ*%p#M#6x)lHz9PJ0EOKoZq9^c?`(?;sFrX+)Qm1#*DF>qks2_9FY8G=?^k
zV>3-&Tg1FB&>lQIeIX4?@ECN~ds@4m;`InM3!HY2s>6d20O-#DYynEe_hyrgPZo@5
z=(^#HW1Rxe)oWq0HxMAC#ty`!GDa24&ojFdyoQ-;CSbK#H9WXlcNjY0*NImUG`rV@
zcj$KMd<{q>COxQHBj8R$qT0G2WzH?T0g$nGOpZq-$%C+tu=LOAg`ZlLSwFXG2lS?Q
zuhEvXLu~~}jtG+^fDi~2{++YedHsD3Rc|Jc-=#xfK!n}-p6l&gZ!iNIo*5$Dwhz#G
zBsX*9*|E{}!QcU?Uxb5;cE##(P<(sbInlNSl8!COUcz|o&LoqQ<+Y8y=(ezuYf67f
zO8(p&f4z{AgrhG#Ao<R(FhVJ=`z~NEi9W;#@nSB3VGa2lMf+Dkcu@cn5rhMk-1sLI
zek!n6xigO*H2=y696^LNR^M;`PwH?c7n8j-h|X)qs(ozeJOo|BsumHl%xDFlm)9gs
zGJIes4ZZf(#vwMG9N4#p?*kO=?}4G7KSDM|?uc<va=bHAxZu6?2FZu~>~{=FYkluD
z6WmIx$-z{UaQ-5hOXM}{QybjtpuY4ko{C9-hO-_$7-mhD2(moZ_Sh}6TK^u;xZ5qX
z{AOg=>1?0Q+!Ifvt#F($b70>&s(=FmbUxat5XCtHcp*1(kBncf+<xwco~NOC0(7*c
zImV<-BuI=WuS0I(E!F-lgbYpFkA~gmboJ_1kIelYcC7it*XK61pfxd;&ydU=V?xAy
zo1Y6SzRGnZ7&x{+Em>@L^mNo1Ogr(v&c}Ul2wwj)@{OF9U11Wy6ADIKI%>vDITuU$
zMn7dq(w9p!0rx7ANJ5@6Le1XgGc}b9*$wEu(*!s?rfa|*nJJ@q>ga~EZhlhWSQ|(@
zky1;1G;PbN_6OMwUFc)8K!&gpP4IR#e+H2|`ibPb6)W+JMUo~E@4|O|p0wb7a~3Rd
zKg7F!Z0miOue~L-b52jjfN9XH#oebKW|YVFK&mV{9_Cz%v8r?}UIZUD5^R9eX~JAY
zMT)iLc3O5rC@YIt0NJ^LUPw2l-9HbDW`2Zt3!!(h=+wrtt;GA9-@cv{B*j$~Ks9u5
z&9924dM%@LVRqxF=&<7zJ@DAvDU)(Szi39*@p%I|{6CbK|8b-J_c}f)GcycDkZ>!w
zGLJtzlR*Z&e)lgffj`NfY(rL8ucX(Sl~~tl;>JgZ!sDuJjuF+Dka(V?8C{D#eo4j=
zrZ!hg8eYQ2ENgEb>Lsd_I^)7-8PcnOrPig}9DjQk;AC<*xGUU3;3Mqcev>zEb;f)2
zr=C-~K*nCg_XXG0)T%{r)1U;r>8acOL6|!=b9S#A$-Uq|;JNx`$cE!bo$)844!*aZ
zE#}(A(p_7f^zU{5@F@C^hx`4<s{2TlAd@xs;*Uk78M<%8f|dj7QNHhGD{aLRY_2me
z$6>s(ED+#QbmkIlN@Ik1_M1-T2ZUO=ijB6rPo3S*(Fsz}Z%M;<A`2z%huo_Ms>~U7
zr4yE6euXD$IP2G3)S6xG<t8sk0Rf{dptZKEAL_=TWW&Vm?o=WUMuU^ha(i93;IO0I
zeSR@MOW}S%s^aV+J3V+465eIz+#N8cYm1EJrLp{pq@umFWz;j>&lppkT-be4vhgf0
zLH_<W0O%$6a)EbTH5Xt+H*h5L*?X*6Cqv%FbYZxP0$|KtVUNNwv&b?9D06LI-f-D<
z7JDpCB(eO0TJ`ws3(1=lBn5q^)}kSGshsE5V14Nzavsv0jk&71emrx8o+Fb+;=6!!
zA`O;MH6p@q7$(_S(7kszamNt3<qj3v<|i3>n+zAIvDDptF7kSpkjzS^kr0_ByB-`g
zKc1jA8BqRe=h*h6zKs6)=tYGw#+Jmk%;5@uPmXo+RaBgOc=Wc}z0QhCxc0)W2i0ev
z@Oh<nt*)$kxExf8WJAZL&V!^eZu;BZHH2P?a8TSl4S*gTn?}+*lEI-S;V9rUY{n9o
z7=Fe$_}dpTk=~mFWj*49O<Q!q%}ciq{>Tz+e!OtjO@fh|--Dy_1n!wY!)^}G=SR0@
z>q?SJvZ#)|q48Qh9#^=lMLl(jZQ4WR^#RYN^Ow-!$47ci4J=U&b?N*g7W~DN^c$1C
z253Y6EV~S=?8=Fw8pUrV)-|LJ!^bom&hbo*F6`+KRm8T1>0y3ZWc{DZ?>?Ek98`%K
zdN(iqQKXhcRLyiI*X}=fj23}qk(kq|h!v12c`ylGzrap;Jz>+YdIzSE)shSd2Y?1#
zM;_8Bw1a@@HxLjihw=a?-s8BN_x5eUQBRqg=vhKzf40#BMBILvfbBnWEd!2uMTo2e
zY4Js9Bi{b3_d2c=La={N=0X@L{Cv&kfC8?6^(((u|BCfV-4iDk;o|S?dtU919j5co
zi`Vu$IfRH4tSSMuE-I8%Ac2A>5x9tSqMy)^Os8|-AzDFikZo^9**2&)t=mUP(VLBD
zd=EU`-;Gy?I;VJ4tmkLqS*Jl*Dz_gZ%8_Bcf8S{%ljaM?Fk!u?XPe*iDg9^{*%?}q
ziQ%V@dos({V}j*qpYs5WYzEn0%Xir6anEUcc408NT2OU^{N5umU(Ezx><H8jWpVNW
z)KnC_d5^6=-~%Mh3gOI)q`)Fvx}n+*5#$zGC|Q-O(L_CnU7_c{OlH)KQNTbs5l<0S
zvHF@H-;X5x<@9f^gMxQ9ZyqUZ`NHcYzQIxV-H@(9GZOIJ?gQ<8+PcERV6GtGFri>a
zQ|D@rc)=JFo@HP|kZfJXi4Nz$$DxpRE0X~8KQqJU75*!m0C+p?r2bS6QLGGP>_U3D
z9n!j62$g#9-Wdn;67~4*<K<plU46%=#lLdB+C}cgqII79?#rHP$v+hvM@}o)^{ilx
z18&;kW<ML)t<iqzw?VJBol7&)ve!o&+7a?pDu3e@Ylhme?Hw7pwN%x7ZUO3L9!j;5
zZ(DRN&E@;nrb|}ftZEu0t_bWzN-P-@#l)+Qtcq32TYU!;$Yv2Ty>#La;3T*Xqg`h>
z3iUJAV3hNiqCMA`Ff*9wI(rI4TcIhBXp1|i5r+3|n!u@<u5b5p!S4-9mpg^LZH%Gq
z#WE(|S9P|7aV%CVO>K8W9p+7?2hv2Juv;n!KfG|HfG{o<Vk6FGWj{ZD3=!Y2Yf72)
z6s!o@Zv5a`804K+<_Y_o9rWKilRx{J7(}tsDu4Hnv%BLt2`qqu`!~s69zvHh@!XJ+
zjHRVM9E4@cs^4Mr*$Ol!lL)X%hayZ63&r*R>ME#`zHu^Q59U7`V3kqo6NDkl{l{Sz
z9KP3p(a#cKdOy*1T-JY?5HQ9OPwSa=%ue1nOdcD-Rd1#lFT0MllNOZrB`0GsmphEa
zmhg{>zlZJd6=+6t4RY^-T=vNxRHgN#i#bw_)G-U3Q4rQT{J~=WP}!I$Lao&!-hTz1
z=(q=)#678sW(3xDOmG(M>iEYnU$IBh-$@pqB8=;d;ZF=mjzE|Q7fqQU$6+ulrMixC
zQeyIDKR`#yaYS5ddjC75;hC0=T?81=4;@bX+z^3+0gJh95NJs74HB&!-JKHUW3R)|
z#Tde<5t+2CC9#uXzlL8a8>YR$PR~QX<nUwsNLqZ|R*?PEy8$`jYJG>zO}9ztr&-LC
z$~=<`%Lb2Z_E-*PR&1vQ97|1<xo|O-b6xf@DGiBeIJqnx?vvAbD6ebbDEeZH=mCow
z4|3?}zVj16Wv1paPXmVbj-Hk-S3_%7f$8e&%xZ=PU{!nW&6$*X+^yhXx&}LSFYdvH
zFFDHax}RR~-OPBn+6I~mZ?d$pYzu!+yz2<P-pHcE0{y3>4t$)31BdAt=M&e)t~KD%
znet@s#YVryW+i^le_%s4a>ort!q`(-__gm?jc~K9ms(|5-;=L~6xqskae2IhmRFvy
z9G~(ty4=7$lbq=zNOWs~j^H04Bjynwy0hGp?*ij(he3IsAf}_8rzQp<;CZ0^V%w>0
zvxv?G!<A2}-(qk2AiQZ$yZd{0>}o_No!}JP^dSHG07X=Vdq86wr1t}E<Mq4y5(24<
z?Kr(M<`MIpFe&!ftF8%BzHamjK?H`O(+)ffb~A28Q<oXG?aJuyxDPzr?qI8BkDUc?
z(S+7QH!p^b`G}R{5z3YStH@<1VkVk8qWE{tbl<)^<`ckku0XN8QCHo3hW)v_5)|}^
zo+VHNcfZJzN>wlN9h4|R_zxFbleNuj9;D3M62t`s_05gnJ%#wll-}KmHin}@f}Lh@
z^J~3zx^**Zszq0d=Ajirp$;6*5MA<~HbaLVDm+!1q$)XUUi?vH&_z=9O;pt5q_>Tj
z(<-VR(NXrN1RXTGN_5X+Fa#{`q^0>pkZvr-sk;^{yS=1Z_9TicqmJZ)T0`CKkIC1}
z6K3dVZzVlWS(|S){n8ggzI#VhH1lE*zT#4AvIYF7=)ovcjCX#}3DgRGRMh>P29YVr
zQ-X`OLHIzG3J{lZO+9*hiv>8`X?4CV#u0GT=iKs$8$aeJ&w2~d3R*;4rG@4s^R|}8
zPw$uB4jsWI&B#n0IV(tWM={-vC)0ee?*|miF;R_Tg4IUmXo=Y6)WvOJbjN9?AWrBz
zRVh?B|Ft*?4A%NxQrLBLsRq{o>!5(y4OfQ+;dNpal*!i&J1ev33jbZzz?p_T`{WVj
z*rN1r0vMvZrF&DNmHl~TJ*up0H=lbsvrfMz?|X`bkqFZ1c3I<=q0n2gc@~shmeGGc
zN1k;Wzg-17s9d%d68~=>_Jlqb`+AtC_WvMrL(-8^D~{71>IE`jVm*1$WEW(Cr+XvJ
z%*|V_14FYNEp=~%B)RQV&BIdOn8y-=*?JZRsCyU;Zoz>smaKnA8%3D<{M@szCV^`T
zoYuxoj#sYMAxV%t-(tnp&pW8+ChU--4y0aPJ8wJUUak&?q~DzkwhwP#1LpLaxPvmv
zn`>QAF$aLidPTKLr$Oh|GMou&<I6pEFk%!J=KY_D9f5qm+{?f0j#P<AXeFb1vq<Pj
z9-JcW<i*`GPv~-gk+9!wB|O~V8B%<Fa8u+w<t%s8`wiujL_AN+N3PPW@B5Y~*$L^v
zCpozlz*~hcxM!L!g@Z^or0cdErT@MrEYn(v3hP=Rk#wEKlukAwDp1fW%8ZoJ=J0Lp
ztRzUUa0tPBT;d^lMPW>lV{g_3+MDPo<>Wf7&U#-fR_wg!3Zd7+$aVjkOzVa6mCNx&
z(!N%Cukp3b655$9v_u?ng6;SCKEKSI5q9Dv%ykz)k#dbuG#&)2yumichxc}~HPa*u
z&T5z>G$2~Z`tx;}2%&_2Wvb_dqp+Pz5Z=KC*K3_bZD!anb@tFM?dj-^PeT*)U=BmR
zRS-M-M!B>>?r{(+M9}fu`ISaT3Nv0aU(rWGHSpVG#HUP!?ELu1&eEijPfXx)Xmf+`
z_f?8WuU}!ERu<79X(cn=3VtBJ5*VWwIQTuKctvZh!Ktc-Vn_L4F-r==oyu$^vtu~s
zU_*Dz)Wv3tmk*4N4U|oJ-Hl@;%cW}q1JC6?ZT**yhws*w7AkmIpyl?qZ@k2_P<HKb
z9Akq@)b)wRhM#SNPx!pegy$~x<kl0ufWNs~{a?4W6bAsRTs*=j|5zQ*NcQ_`GaTaU
zw=EoY9~+d@+h!L>x6*b9>LseTei8Ry>v#G#A64);OiJ_7W_rA~^FI7##?784dT+Lg
zyrjBp0=7H0<5aoy2s7Q;vAy>?Y3XRKvEEpL(3~+@dNb?8kMmBOxEh`DW|L#ba<Kxr
zi`<yoV9yA|O>Lx^`u%wT*%xO$x5&Rl_Kqj+y7^1<yQ1k=#+1cxseFIcb)AFLC?qg7
z1IwXRy5;=aH++kNM8ahj*OTzhz&z;WP)g3;(Fu-i48_L48}q}eoORVZr^sbd2}Fb0
zCIy-DOjirqLx?J&f=l3>#FUj3EW5@GzYTT0zQPmJFv#OPWFV~+%RzE|@Lg4A#tPP`
z-utkC5mX&PepVYcLL&xrfo6+=tKSS}ta1`C7X5Q#(v87k`^6ZNRIuygFT3RiN*eSc
z^f!&9MJF5#_LRKYrdg)7CBA`d&>>cF^Rq#YKu;62>fxo#M3pzR-eG@v0U*l5d#n_6
z=Pj-<+$pq=Ejto;q(Afdroe*o=)?O%HLRrpmK1ltqlt?X<upTQ!bCp09=H5iku|TE
zk>jB@6gKLOmHza#<8QV;e*s^A_Bd5sBq`>fjE(;?rBh8JfnuxJ)cD2&D$|1$ACKZp
zMWWbKwPK-LWsiztr(_Y~(c2uZUrr*Y@MBaJL$w;{cVZrWoucE7YDhp*ghe9qW3n_K
znE{AE&5$Je4tyVqV-tWA`y2HYBLfzp#$h$!Q~mvGu7GRtV7O7=ZOPf|w?>$VZLQG$
zhk|WhlE;4lMVx+-n(KjiQ!{#>ptCg~0ZlaS5jMIi$wP_j13K3Ozjstn8yR&8=A6w-
z#~V;EI1B0()qJ0=!J}UKBE}R^8%?H3pHt>1mUxBDvj|;kb2(@Gp7~v(#K@6jM69pb
z8V{8jP8H|e!uLlbcwCeoCaSO~r*yq<r0!WYw(S6F+8>e2HEjPdG#x)rJRL(GJa`^=
zWY8D_r5^ZoQ8J+OdMzY@2C#oeV85C`n77)XT}~6GTF#MlrHx6|@vy4>%wS9-MvsYL
zT9ag2b9!-(Cj>aQRcbraAEO?h(rfH6<#6@EziJ(P)oVGweoio1ub$<05r<}KS+_xA
z^l<9$%TUxow+fwQhAY-N50Yh{pm3j{gx$EX!Ou<!4ZEA=;KWb^)SH><bbO0S=4|9Q
zz(Vnf>wuh-P&CZ--rN>I$sR<#J3wJI<j5N4m-c0M%sq=xWCkt|)xI9pzEv=_G^R$u
zw~a@u-e**G{I+L~b?(gc)1$^<DKRM&Q{#Czb5p)Dq=_S$b)JhkK-3X^f99+_%cnvE
zmyw}iRd>eLGufkT(Fe<iY?m4Cc*b(Ki$`(QBIcD2CPLs^V>-g<gMUwnxlE18q5!1C
zE=zL)@l>Xox>cZD*~wVIN}*ej8MLz`(J`%D!V~j{H~PQKqBhwOw2=IdQcHlGb`t;|
zbgM)NO%>yj4Zv$0WjlZfNv{;b)Cre_QSnXF)tOGIeZzWM0^S7b>S|yQ=x2T43T~lh
z*O&!6FnwJ*S&vUdAFl)oeRMPF0}coAx@I^qa1L{^(+kpgC<d&2o$@}V7u9E)kOr<@
zxMEvZ%QN1Iex_7Y<9wj9^!=<Zck{`SyV*Zg&wiXFs&ZcD9!OF5HaTp@F)KH8?(tjj
z2?KA0R}?1M<IY_`6*sDTHiD+(+IxYQqhOd7W)C?ke&@Q3y?+#WHS>->_a{^1Wuls?
zk;^y~`L65nwWe)w3*oUhX4h7j=|0U-ra%%0HNI0qYS=IpQ3})_XR(mUTaCR$x_w>2
zd`gfns#lb8lLM<b6sHcg2if=HAMEVP7Xn|<(y?&r*Lm_o>zGB`=ZWTQT;PR^F;>9r
zfk`UuHprE&%UP4`J%GUq_711)C<Yp^Dhz1p@&}6;Hek6Gu2ljY-WFC7u&%h<sLVt*
z{&DbzMIoF!QTYCy+#<G0_;ueFFQmfg#c}=r@jcG??nCeXjnN;u@8v3)thG-aFU7^E
zODiD}QOfx#^JE?um@df$uQQ-~e9r{o7cYV__2Rj&TEbUENDL)7T*<%oogO;zl5#$Q
zOHj~2EBg|^T_M^j=fvvl3t69Q5zp<D%KZvdtz^Vx`)CV_k?TiLArnuHa}uMWvQ21z
z3V<D7b$#<IT@Kgm1E}N(>dGidsh3U&?s3Ts*Oc2vTRvsmJoriBN&iGd443-}j|<Pc
zF9wdN+YZ)6k~f^O;p*RxE>xPn2RQM`sh&@^?PWG65jbbC6|8Qft3E3TkWb4?ybbLr
zfIXg#h2I{8{gQ;P#Dv)NTiv(^rWFX^_hU;^Awrhd$_wJ39=JZiH5!ri`kQtyKB4XI
z3%SSUHkhe1Wel-1?U@^^&sX>NYp?B&ZMN{_h9p|RTm=LvQPXkb6))qj*Icy|tr;(f
z6=PIeeIAwC?|iRmIFz;==};HWGteK6v@!3eEjLW;Z!bH>FvWT`icCn*?j_hDGnQ6-
z36-En-Eh^rv7$bZSHX{yK)?8{U@V;7lD%mzC*4q)^tkJrh4r=nIjenf&bz}k|MQ=K
zE~W5((o4)Y3kZI^e%zEVuIUxm$xW?u3nfV_oJDlQVNMSFiGNCOHEGbV+<uPCMI0Pj
zMstW8r>Gj{?!YPkb;>3Ioch*!FTayx9WygCySp-T-7MO*`$G;tiDw{l>zOML1UhO6
z1qwi42a`hV*vUL2Eg?Il|Bt=5j*7bN+J;FffuTc`kZw>)=@5_{a_9j>lnw_(x}?RR
z8x(06Qb1xzkyNBhLO{Ah@;&Feuj{(+=X;;a=l$NbzW=_p_+yp>!koYJJonzmK8|C5
z1}%Z4V$d5=N~Pp?X4_Vfw46(!c;TFJLAV#r(|cFH6nSL_PhL*Z65l3@{8QzZc!Hd0
zU@X@u*|J!|-PkKXRG*F(ediw8;Pu-o-Uj%SnU#6P#Juj4tP9FxGXQEjHKQjPm9;7o
zi5PR@wBL}R;q`H>00Jj=YWk@n<x9(qZgL{S3|_$OjqoDmE%b<8MBp!U?~72vL7McH
z)F}M8y;XxF&0lE+SR<+Snxs$q7|a@i=PvD`kEwg7O!ym`JZ)vZ+`z~+5i9)&sQR^*
zK#;zXswX7_dHZ$rI&cb-+I>O98(r-WEdQ}&PB<pK-^PgE7dI~oad{~%rlmM)h$VCE
zlIi;6m2wStO4(f?xPJrMv_*>^OJra7ZfVtMOdf<fUNU;yVtEYLHtpcOqKtpZqS5`5
zJzm|W`{jdtNV=w;W~8v<YZcYU5H|*;+w?bu7q?X835nz0D6)q2KH#so7&sT-syRFv
znfX5BUD9Of**qFwJ@Rp8=ScH34dFkqz}@9h=K9Mere6Bl{!hzuT!pii&NwdJvyQzC
z9F>57{$j$zqC8&2h%x%}o3c0wWEnzZ-$BM@nP@0gXR(FaeIdoNU&tSF5d_soQt>=n
zO&HKU^1cxwbP+%bJS_!k?xcu1(sPdlQ6XcliLsxz;r&40r(7hDbt2v|l?DcE11qij
zcG$(yzQD9)=1aCh3F$FYv|B3{FM}NFIq8-hk3~KW3&uX#{Cgy%nh|Q06Z6vqba|wZ
zBnyj!S($JwA;tIEAn=M>pq;O(jB@tds+~uQjSLd&KOB?-`y|p<hF)cwFHXG{KI>s!
z95h8yD>#1n%4Px2sJQjEPY%{>5xC2u6TuRJzp_eM`<^A(XMVJ5LIp7v^D>)hGeiiB
zI*iS>(2t{0n}VKui6SeJ)fm(e%qt~JmM`84aLT09lH<zfM;xLe6YHO<7vB7ahaX`#
zTxhcpEP^LZ`@vNHJ(i4^Ku)h?xEFvs@hL7*C+11{c=lIXm+Qbh&`r<MSft)_p^ghL
zaNz15X_>u#cCa@3q4COm8glj|sD4T&F0i5rs52B9W^g}E#M2<k<Sszu`L^0_OjLyD
z3TC9<A|sx7ts1V7pQI|q-v_8Q4g(qDt~u%Y7hd6q)}`{>(bd>ZPe~3%ULal|);r0m
z$)Gfcsx-@`7l(Dy?FdB9PuGRc@hOs#P+zR&A83ChyE&276O=RVnI`b3eHm79cJR^t
z7?6C9x)TMQsn1ZX>t{QaA+mQI%$UlW&zKYtpB(nE_Q<dt<5oz!V)w9f#PRFSm&1xI
z%-SZOkrWYabT-A*w7A)BJ8X|qH5iW8v4?~}q{m9(6TrM{tjO$R&H1M6-80PdPkz~(
zt-2RGiVW`8)^HK0mo}%Ty+TTK&=EZGV7%G|#zV`r$5&Zpv3~RQ72hFv{lvdP_$3Cx
zzrNzMvAC(ptAE_I+qkF4%C=ggrI*EW^2&UFp*$bs-h0h-?l<fj<W!wqz&wgKo?D;$
z%QeIv588DyKU4_1b~|UD5z3tYQ`t9J%sC@Jmj_A|And-R{*|;s^%pcY(oqo?p}NCO
zTUE2aIenMQyrf#S@!7%a`Ry#mm${%BoWr7IU+ITKeV)X-TGO8a#q&$ak7ouNG~43q
zKRpGrGVQ)9gW<I|R|$tEN*MP_OsUShKnQz$q?X99(M@kClY=}13<RYT{fPQuO2kKA
zyX?MBMO?FpnnIM)P-uRY{ix>9Mp1xs<-<OQ2J#4<sc3qU`=w1Mc2K!Q;OxPd;{?%t
zNBOy$kz0=f(Z~@+%|}kP6<&Xbs+y9+F1dSE-|(yG3-04euTWPdv(-Z(8xx&I?YXc%
z-)aoj>3Rgu%d3012SvlG(7E*dTs5d`biTNb-adR!wj;VmTCa<n`yLqr*Eov-(eY?Q
zVX|CW&xeZ}St*R{M7s<^UxhDz1U?yxp^3QN7{R4xAG?9jf_atPM32_?@fVV4iJ4~L
zR-Th1yfT=p7I?;wwV_YsH(038lPu%!HzD-IE68wL9N6Abd>y0EW8mRFRBN|+ru+6D
z@yeI%;CvLHV?ymx4Ix*q`*MGdJXz`~cLiB>y)6UMA|K;EyH*D8?7NX)KR{ha<VZ6T
zQ;$*IFlH?%0MQeC;C|(Z^Q}gzXsrezHxnho%W-e45!IBlsp81+^#+Zg7KzlpcG0H}
zrD=OXTcnd7#LiJ)E+M~KZgNYT5$o_9U`Z<>MFsu8Z6ueP)#i|{3Qr!oum3W+GdrcN
z=2+qL%S{5H*zdfLHgj@J?>6h9Y}PQnnGX^B3MFEMOALpHZj0SUd-Jem@mt`=vR<@&
zQJ^w)h)w0q&a6$doD)>>uD14u;JvX?V6&nHdOK<k>iT1xcKQQ{^u4+WFMb|_VhEMf
zggEZni+bsy+>$=WG^(*%`n??7b~U#U6IG9M{8ixgT1rNT-WcA>+h@?8*Foh9D%Jek
z{MXf}XpJw01oXuD#taWg$)a&rJJV`Gf#A5+L2GGVTi)sy>5E|<j>z)A&D8S!x_GAE
z7tS#m!DVf{lR&8^WLAzO)D=0~3#)U>?fazXMI!M7a;<En#AGrlCq~>Ki8uS<MdY20
z4}`G%co?)NTvbUaeG+;^KbllKg12;Kj2T|0gZcK!>d<HXHl9BZ>EEaIzr2urfy24o
zL&i|~>R*Rh7D|S~dx((ZhCG#J3W?o<*Z^%SP;Kh=c(R}Ac)npbC7@`}E`gv5aM_@1
zlI(=-%HBSYo+n*p4By%DI^Ldh1C2hapY+lfZ;2syY{;3G0$libWuLxF=r7)YRLiMg
z3lT}zD`32*_Eet@Xn+)$fwnwuBqnw>AqOM{j3EvQ>dH&_8jU-)zw%9a$jnt|ruKv^
zl&#(Kcy@zSZYR$8RLBxr`s}9cK1QICHtnoH+qA-$fY>#Nlq_p4u=Dz8h+A5Bh|L12
zZi<7*$T1c2pDRvNPSnabV7XFvdH|lPkkGYu?v?iExb#QO;jK7+D$jof*AdT=FL5r3
zo)BkqF5xI%87|bn<w_RfX~`mnG!TmvJz^>hMdEugMJh<Z+#xSzY2)t~V=S#&8E5#N
zJPv${yGOo9VU1-;-z^Dl5{)SWOkEf%DIZsIp}^qXJKxho%g=-{ZS`n{N(^4Rg5CJn
zCGBiF1@Z^n+fnXL3VOj3?-P_tlxT^18Ka5CgV}H_A;%#R5wBEBk|#2-%Gnh~`QhIK
zm`E4%LmR#zLhEoa<4l72oyVwrgf2<NX9L>T^1X>qA}%bQK4L({R$b3KrVK{g*!$fc
z>ue-%N|h>s{(L(8#V^c2iW8h2o)MgrfJVk;@=fMv3IWFq)>q_eiJ+qr{p5&^%}6yb
zDbStLc$6e87edpMz(>Im9ac-ZM}*}U&36*(`bnA;Q*cRi$p98!A)WSx981cj^Q3bN
zKV)^dDD^2FmQyc92yzdP<uSfC5fQ{!K|7T@ix{FN0}h$T!`XEVqUu7swd?ZQ>wyrb
zyDhi7TN&@n9=((~UTbxxFD6ct<A+}c{Jef@&yKm#DudtDhBIAy6@OrQixi)9=h{_k
z{8ucipQM9t(VePp$8wm`L6<y+8au0*OAj!a&p?@B*)bmr6&z^mH#HO}`%Mr7ZzYpu
zp#*Mesw@2&>^kma7^K0et}q5#VqRrD8b;*#-VPPVV60%iufld^TkM8G4dCk8Q24jB
z(o1hHEL2v$UesD^R=&3VRq72!cGRV{k07a{$gwyn8Tp``4^9j5CK(>jJ^W)@l5%rc
z032t8(DD@l<LK8|<7&HPZ89sx=1@NW3-QxeL~8h4f_@hmJP-QoY3G@*!jNz|c~cxB
zt=Tt^1F9v%35JN@MQ{=IBs$74!2)I>R=07&N8q@G)bVf;0_tI=nU<$$r4oU?bZGU>
zbtxPsS`(P_9kAkUuUY5eJCEmDLq$IT?WuLD7DT~vLp;hSqMgf0ungHLhzGq9^57@l
z66Hv8Wkv$a#~3<Ab(l4+y^EY*_&HN+hlp9guy_^KNDn-nFH#U|BS3}FF<|j_Z>1^L
zF`iPs<zl<k2cZF`8;H~EICq>Ruibvc<X4nvxe*8_9O*~a1o?hCLEm+^cxAI9g+jFR
zkr;A1NrHE}&9UcbaP%NswHW1`t2;zCmV7+|R`);H*1JV2%!bu}<%s4`bP{i7nrN?Y
zl|gXW$s5}us9JCBi>#~nAky1X=nwN+u||q0pwtPXiAS4~czDIG`*<u?+o-hXp)$pF
zt@3dPm_!K3ZR^Z7MzXBt<Ld#94D+|s2#`OYLp&~Kr#k#BV5k|Bu{%n1;#y4HIM>48
zT9(kBGAWkQ#=^HRHC?vg>0<+Z-|^nH)CcMhwH~?{-WZ|6c@Oclws`(Uh<COZwMAU4
zb%u#uo9#PB^X0}_lLRO~ZcX&KgRA(Eel$;f1aCw<Bmv?Obgg}};1KrzdP9Oen)@6h
zgcy&}2U!;wf+Ap?P_qx{5UynpWg^1*=tBZgMqx2cIm!StSUQuGi9vNHoxp1*FiXXN
zB(`*$F23m-37Jxu_+4o*2>-NY{MAx-wY`~_Qn*V@f$R9DU`WPNC!M$O)S2)*4)7Q*
ze&Vobz`T0ftsb6`|4R7h!UfbcC~$TLUd_h;`<r&dWiKzz`$Sl9+>qUR6IjJXEm+q~
z=UeM3cv)=hyb7ughO23OpR@7duc9sWqIsq38ccRqi^I@769Zyi`<8Sug%xJOhrP6^
zx{ri-`C8*SO_g<L4WD(s<l}pWh3xbMQ>Ky;3b41<Dq8T(sXF#6WqJj#jbuahRj!aT
zVgB=1;8T3<SsGI11o?lbwQP|n0BqPA)Sv9H+;mAceT^OB<Kq64)TQsPb>~%gA#Wgn
z>jG4m^Rm6&8~m|&q&IkN(-8?qa9*zT29PV@@JpEm@N58CX05Wh8YU)eBlx(JaTC~q
zHPG~(oc4H=cZH&E#!6ZRjR9x0JBjNi`ZsDhHzz-o8EpmUCk6g{;r^Qvkx3d<h8>??
zv;7&A{|L=LWZ}C%KC|Boi899V<TETy15AGjVIV{#x9;N)+Y}=DGtU2z3IsXVXYgCO
z`HZVx?k@H$`}&W|{qyo*l5^e%zmJN|apbS75PflK2VfMpp+iB1*WOb5M{fPXB$M{@
zqwg|gpRWDxKl{%oF*X5K{zG+)_^H~zm;U#+@_%_(n;ZPRA@}o$7}noB-+zAV|Ns5}
zpR7N)J<b5_cit-lO;Q}Ud0<E0<&Kpq>i8PK5ckI`Y;1sDvukgr!u!8G&Ht{d%uUC@
z^~?3e8Tb45p6YsqI{NM}0Okl*t^-Z2L;}RW28>M8V`_5EHm!5hskYNUO8PJgq-pxn
zmuK!2KYracs(h#c%+4Lppmg80k_6>|PK9*kTb@>@*SyLx@G$5GKOd*37w5-H6*t!g
z%KMv1-}&{54EiQ0qk~aAyaAK-h&A#uz&zwY_X7#rH$c`Kal0i|yc<8^^T8Y&^Jj!P
zpHYP(HJ>pnpGi&JQeP@Zd|fW%{nnGQLEi_z{>KlxLk_YH^9<gYKTp0D2SzqwyG#pm
zqh>OJ)6xLxYdP@qlR)zsOFWMuiFaQuFMfN=F1SzLNVqR`0q(xtd@gU{cPmviviJ+E
zz762yRNYCr{g`E;GiK82AU-ZAQNV)#{PZZXFICLs4G=hqRSF&gjxHMvi0Q&dwPyfH
zC&O_>WyQI6?g}Wi%^l9Wa)lN?#U|3-ddOO#nJTJrv2}T&CAe&sB<)+j4s4P)g;ogX
z)_~K_8;EI*-SGYA#Md6W@>{Nrl-R6~m2Z}#L|zC7g}V8+S4soHEm!arV)EYMmS8+Q
zA?L5luaVO8KW7aseA%dY4B`L<N&3ZnO}8-9!%LuJr#Ovf!~!E+EoK-yy#M{||MT|v
z^L1cJ64)aqJF}H*|61kTU=*0p@Boyv5uNkFg1pb<Fn|)>lYmz(J^|zFKx-(G1`KLE
zVX8Aphu2e}n|FU8LuwJ2g(qYyge^I6@oU}_u^SSEwt+`%-!V_O(Q}0KJAXzIC5I7(
zwCW&4^Mrd>O-t<=O1+pVuTtPUt7hD??N`S?oByfI@7U2K0Y?QWjsSO16%e~H>BL>K
zIxI<*7?uoS0SArfmYZb+5>%x%-&sgnr9y5<J?na>`lhwQ#FW7XhiLgE+}0QFCp#PF
zOFEi!mCe#uaYNuR9<W6Qi-SkXEM)9lO&l-4{{Oy|MPa!R>5F};o2Qxn4piSJeAhhE
zejhlJ!AgCWJXZYkwq<*PrgP#NG(eb!lJ9z!M(Ch9!`<uSmARlOVJqz`pdTtR5i_cB
zNJbgcDcr_#MTR65gTsWSrtaK)Uh~dfjJ~8gdHH!j0iZRaA#0lqE>hhYKml<Bc&#$|
zak}5h$}WI)#xkI$;eV`;KQMaHU<z)4C|rVFx(kefJ_7*waYxI+tuY5M;fgEDc?_9d
zLc|LE<ZDfVQq=8>9NuUDi#^63cc5mW<|c`Hd(|JiPWs{<8Nz&Sy#X`=)!&U*+Z)Qv
zvoyv|#IiURmsJ4&FzwZ$f}1egu`)Who7;feIDTV3Wct{e1hzj3wgc{Fo(0I(ja7Py
z`P~0Aj{cYO7erRs8FU@6^*ru6%O8QH%N7f?k%Ab&j4olX-3N7NyFH6QiI4ysZeg>f
zA8zB|a{;9ZAK{r5ZE-DZ1~}Qep>qW=_lA6=&+3tRzmhg;pQC<|`;36Z$MpbPejbg%
zrC^IS6d9BVtK~zgSkUf~yP%Od=397~2JW0)AVMFrhg}AXlmO|HKIkf|MtK9>IxfHa
zw@~r!?ZknD9gdj!6vK=1@P($TL+id2kqTfzt20aUiS^im4mso5O2Cn{P3vFl3J5rp
z8MtBx{{x7}Dhpcn|0?-02;#`GGL}CBTLRmW+Sw7}kw&IFExy&@xH|~U>Kp@3FI5$z
zqdCALWb<jMxeZHsZF&-jy=`1M7eGI#!XNoGxPt5hfKXFSMnRgOk}PbavlQ0B|HCs|
zK9n%XAK&F;ga<J1{961ho6qGF9kuo{xJP~fs$iw4ESx~BaIq&b5#Za#*Yk-mH{?Ff
zx(zVSKHmb3Dxn@L7tfztc&qm^gP1iFN*$^you=zY{K5skUYKH?V6#8ES=%CS%+Guc
zlVa69Ar`)aAKgM}9pO>yRSwh+xz$6@J(h%TfLS2EU_@OvYN}%l7rT>Yqm}k%R(fdP
zuxW@)%QI6mWG`)+{|Mx}|NMg$9s{?PFd6^SBUlfuXSDrfjpHKNzU4w=zyO(V6NoYs
zz-PYWxC4Q<bzf_>FKc8EV{i*M1u_daK=Lx40*niOkar<Z^)4g8J|0glB@GXS6I^w7
z0x?lA3QlWX?D^rLphdaU)Qz8I98WVz*V;C=E}DJ~VEFrNOMf2;YxYkfO`;hSDEG{(
zw2Sb)>b*8mZNCVJITZNq<q1KUHkPqOS9)*4qC19h(<EB<T89?~k3SLoN>?*U``GVT
zlR7EhCbVo>SE-YGAGh1Xrrw_4eR45okI>!^sMfN8wm?(!&I;?}soL`W)nT4;JR4)0
zli*+U2fY1N((iBnb2bI;V@`T>rbui%@>eSTia<6EgPB^<?-=9iheou)Q1tb?2585P
zu0`hptl55Y7CN6NfNF9_Vj=pLd{Ho)MEDh*Q1s;qdOZ?ez*I(K?6`x*B5Bgfr#&qG
z+K(`NJjSCoEM6yO87v$uWGU`Z5MV<l_oj*s6f?IT2q|J;zw1E>`y7nX$;gxxc)$|Q
z1Kv04iQ0FWju2%U0NaKupJ-k>Jo@=nI;(jZLlRTIHnvTWxe9=mNdnM6Oif@jQ+I>#
z5p~;fqAFmFAM7l?+6!Pg{u1m-lTDzpkN|we^D^P4FmX#zAb{pcrV2=Ja5gQ2b7Is9
zd-PBt73>-*tVGnjFF~+XYxX-=he_CUQ(nuAmAQa1Ed{~nf!m01MnQXK-HT5{1-gzi
zaHqEpuu$}X2cqy5urAc(%qRt87dp@YsDrC_u+msEv^TxI7bARwlLjH@bH{bQUArZ}
zd9xOF($ntCe~zE<YVFp@jl*lvmq1LV#d9U+W_Z}i;RX)~;94fPcbTrYzZ$!Vlm_><
zN2>j7BqHk0;2)Pq=E@1I*{^ZqF68cTk>S9|K{VYhhpl$w4rv3BNmbqtZzQp2)9*?=
z2y}nc5lI!{G=CK(9Ul=j?qkY`5%_Bd+~SjrB<M`+;j0SZXHIahCH2)!Wx^%W_CA*Z
z?<!+5fH`{zCa;6h3TRRUmaUMf4Jm{8m))BbhZ?ZXki|RxDCtAMN1dn(KL|S#rx9{4
zGwksg`(D1q2u7KByjbAv+#>p6^ua0p{42KUl$FERd%Xw9mupibJUn&`ejHOk^;q%j
zz^=f&vHpP0;GU%~rN89M$Kh-Lmpir=hnX5j14$nmI~hYSQbF2`o}C&<y7G!Ax*z~d
zd@*;bUtzZwkTz&rHRC4xj-DT|^7XB9cS;#hVNtx2kh4g?gXq0atmp<2Yq2A4CoD0p
zCgigQWDH^3eg<#67m$m96{KaG`Y~Q0b@39fMUyGh`;Wj#rgt@~Sup%`Hh}^Q`&|dE
zY{0B(fV>M5q;yH!yTDdem`C2FV7-{PwUDGZLM}Qy_K;Ss783ep4up`Gp+q!<AyGfq
zfU!){10_MtgbQg92YPT~Ad*2A{Fp4vaSK=zgqU_U%(|Eq-PdnDjP6a7&;@-v;st`y
z0*G6dIer3)Vu6?pFC-nJaYZ3+rxxU%>a5S78Gmr%gLKI%QdKpD*WHGrFx&}<%D)58
zDSk*dlUCu&C_$2WwxQq)7Fh>^a=c;#`H&R;_7>sdGt(0?15JKckPuVC#)FyY6OJIr
z8Ki~mIs*7czvrWN#~b+!x4epFU+fM$zYDnDy~4EfghZ}!b7eQb9Rx~5T!5>xi?^U9
zeHHX3065+hx@Qea`-%sj$thxjm6HyJe2f{#beTy!F>Ntu=S(?ye=MBHW(sj2HPe(N
z)6(~r2R_u@=zM4L!HGR9cvuEj9a2jwUk^un?AQQ^zDDzGtLbc44b%d$>49T;f3ygH
z9&CO8o~d41J5P9xCoK#@G~e67cndxw|0L==IqcITff*S9)Gi6rI@OgQDa=OJ#hg<w
z(|3w<?nViJctd}!4I4H~inD#KJ-y`^8{>?<wT_V4?mcs9rTDj#q!+(U(6Ykue_FKM
z`a1`SbHO^<nb!UKI<NJrR*4&%agl+OBvty;6Xv^QhOeFXt*b}DK1xf?Pw<jNY^@c*
z3uNh5%|~>1qVf{_2pfo^w8;WSn|!^saGzxm;z&}NdaX_jzb-_uXmv25Co?9->32lT
zfN?%s1&Hffe0RKPY5cWc1?I`qYnd^$(iG)c;au8(^ar_|R5i|0fj+ZVhZycmk@lCn
zK(Qr7HTl%;oTuvQgUIedi*jR%szZ>VerHS{ZZy`c68_@ct4gK!ApSdBCoEq?`Uv(@
zzr)?2tloNms6IC@VdV90KB`-n=0!$i#q0TLg`pkoqOcJQs&92HQdBRVJ~a?hV<8c<
zs~+g2+RH}<l;iav>&#*1N79qRITcS1PRhn>S~A5}8wn#MtC7nEwYQ*pTQiM(n?T#_
z$-+Mz6ZylT#H#OOmb_I|&y!3LYYoLp$4g1q*^-pG+jsTEe+YEz+S_Fq>sJ3NVi|wu
zxyhV4insLKc<PB6-+qeo8uoPq=#%+|R!OLkeV{wv)vF_zgGE%$$kEVFOY=fDUD9Q<
z!Yme7t<bFu<3wVO94l#i?1|_-W84eX6+O-^G)E#MGF6Jo?U`MP;EL~1UAi8u9j}^c
zy+B`^w3*oywFNmdK+zMQ^34t6mbs>Pw7C6J|1Aol-GDQ@^!#sVC@X{l_34SvPXiu#
zRX%XTmiRXG--E>K$&$&mbKMXALMG3B=FZqW%$<>EFeJQ6h2saAFF24SWkIS3)n`?^
zqGyhE{{UL4B)oFEcG6NaexZ!v$TU3W>z(}is4(B^i{<Q?Owhl0R3UC8ulW2|ZfT5E
zbuXo{h2<-tKhBE?AJWMdGOKYgvNV$aZp#CA%SlCxux%!R+>3z%fy)zZEoo3{Y9yPK
zP0b`#O>dJ!-i%MlVpcs!>2U9-J40)&Lr-|Ca1{l8vg0+K_%~{jo%i^@vGoz{3(+Q)
zIus+SU#7&d^73=Z%`)@)!c3V=ep=V8Vas!`E8B_Ou9CzrX6ot?$5m)5i#5gdt6LY5
ztJ$m*AH|!%c$H^dy$_~M2vBqtDVdB`RgLCj$y66VZZ^}-^pAZKU#N7KG|g4|?p&ky
zaw+84auY4G=PyK2xx-2>$A2AynVsx2pM*Y8geo<(I4AE7$b8o&3KqR_G0&v)95N`t
zbkt1xF?NlYgeTSY_mzEzasXEIy!Ake>c5yCOx9SSQW6oOFCJ<+NxaOg|8+acP3CM{
zp>vVSur)siP5bQ&*1v(cyBdhg>)t|6Q>?{YYFK7IgQosNkC-xo^@|*e84FM0Wgykf
z4#M%z1!G$t<)cVP7MZviK(U#}@AvIO_d}Dx72fvH(0txtn?e)IB2uh;Q|u7ORpHT7
z(93q`W3eZ)(To%IK3tDO6cRn8c#1>OE_4HuC(HScpPizkO@|RDW3^C>`9}zZ%3XJw
zpJOohV}!!-8%H`Q?ER`*jiZV_$-*V;%P16S&b%JxJWIHD7%8JUlj`4319$d9AcwV>
z^vnaTDet|*AA(2@0ncFwbBC*Vtr7I%a0`7v9B?p~$4}*!oi)ss$0ZFqeuCLd?6O&?
zcA!`xzT}aE3>RBno$P-OatQYGnzpS5FtBN)|24?sB#MmZ#iJ?<Uq<}7w*SDTsSyu!
z)DHFwc<?uNz?qFRRFal2tAK`wpq8*e18LCT7WTjJff{-Z0u%siOHC)l4@FJ26yfL7
zUvI?y@^!9SVx;j`3aZ11NODD)iKzN&it5grtKFL5$`jzOeGO)X)qmN2)dV6<jsY2w
z#u|D4T?=SnW`4Xh4b?OKqPsZ;Pm0YRWfHBDKsAAD*W>Yu-27Avn(NQ>h2G{z$wTLo
z@{ixkrf~8SQNAbR1|y8?Ik=PJL>(tW*tlMcBPt#CU>&q{!+8c1??95ZJEW_n`q5)U
zelMC<NFHReEbvuQ*BHE2eWB;PFr=)YR=K%;I5)TCE8zPEMTauijkG(6!ZR?X>PYP7
z{soYzLJ$I6c4kmL9Ez0p$S#K-7suXEo5Gq0lh5@Q1uwL)#!F)XSxQaMg@-#xGeVos
zbK*o>EKl|EuCH%?m&uS@O8(uQcYb-#W>1czdQ(K~6{o){pG*fz6qK$y%IIakO<&H(
z^nL`u35S99txy0oU+R}{l>+qO@Ewb<g~PEUpn4Sd0v}Z4pxPymr;g(t0AlL%i^|WH
zbM<v&o4qWo0}>A23*gm2RjmI%VE83(&>+xqZRO~nAW%4Z$e#ma%NjrdR$uT}EkZ7@
z0|mdZ{Rl5bLn!-+#XF11%)*8zzuq=PvE0{<lEx}0ly)f!m0T;UU!X$BJhnm)cu_$s
z3@IudxB@KzjHbl*h(IxZC8(Y-LG=wg733IAgU&ICQlG`k)@08Hm`QdJl9M<1ppZ%c
zI$F<uKlo=3K*w8Hl@8NN0=Oc7vX|}zV7LPlid_IYM%GJ;2zBN|-zt{Sr94tHYamo)
zNTRW0(>eqNsk~Pwb97HlG$*evf^$;z9e3D{KY5pjuOiv;;^_%MB#!i{!ce5u)_#Sj
zk0=sHLFP_h_2?@Lw5MIzX-b?zBF~Pk-38)Uv^HJU?9D{m{>uY+Kt>q1R?g3-OcqQe
zJeIr=p9A$ObvWg4tNkO<v#&uEE7Z&53=`t_UjJ43|DA~Z=L^c~n6Pc@f1jpgoLT8|
z+kp&o&=r@WLVCII&>WU|Zp~y6PWLX_*Jt<utCFP&>P^QTEnL-=*Pkx<4L?-41Q*WV
zasPhDd8Q$5Lq|=_sP3nJj~!PJp*JQ?BRAzXdTt_el=_2Z+Z8Ro^JEjL^QW0{Bu>P`
zIhkU9XUFf3O>el_&>18WvXnHsSmTej)Ln`G2H?@x_8vWnB}6*Q1_mYlxy6PhKSS$(
z#@I^^4d(3V5~<x-f&=`y5fnP!E=A!xeqhK=ga6qA<Bci#U?@8?$!Pj`0vR|3(l-A-
z1*Eld`cQ3S9~ZXkOF3fSN0m$YpSh3Q>k!MT8Zo-`6<O8k_3=U&!=lV7PXKb6nrdGC
z%-;KL<W2(^jr^?Iew2UcfZue#aQ3@yxn<ztubX;s$J1L=kdKhjDUZd)XG)XPCq$=?
zVdAxu`aI7wjGy$?rik~?NAp<A{~q{cVP;a&A>v{Ft0eD=Kmkm4O)dlj)1rPGOlBAl
zb3%xv@+ad3rd-ekp7ZFwj~27(j;mT(@GB@)m8v#RK$JtW<Tw$)8Tfr~(WbV>)dB+-
zo%cJ<0b|@IftkkVLPA0H75jpsEdioa05K-$lnZ3c7%s`q<9{ZFn4|ACpEa6>c((|f
z=9n7QpIz{TJWis4kmeQ+WJ<5!unut>(yuxKmEA)&>Yc<y^bz3XE8RiQ#CA9@;DLi$
z^l*;^i1yb8K@$_+ItlF7`|StcMdT%<oSA8OGlp$*fOHS3KAh|@`Ld9q^HP;*@_<Zf
zHgRh$7)xeA5dpc0_Ds$&ChGqQ2BfynJioO792lY-ZVM^%=;hDKJ^Ao-$7)gyc*7^i
z_W-{MNA1y?3}d0gZ9K9Ij8~Irx0^EkA_Ml)L!!Nfl*x+|x4LFfh&2FikeB^vsYdEK
zm<Y5tWaXNjqfALYYLr$Z=cD@jEbRwi%<#DVD1P~8rZ0RGLroO)8jglyO2s@kO-TnV
zF%3N7zIlL}#?AuHMy9-yaTjTkxaKTg)Kh3lNoX~P*%TPrw6p=bCo_N}NraF3g6H!t
zYz!la%ReAj4z`L$1&)BD5W8=eauX1Aw}&8*ZrFj}!8~6DzT7~}nK?!rzelm&9-~hT
zqL~P_6NAii@5f8}B!zXg7)?WIo;^$^#Y&Sp9BUK5=_}ZqB-GY^_59@E){dNu>8oXc
za@-=h1k?R=jE8#SZIDt=?@1-m045rdcRF<HVAR5yMLkZp@fo7hI1$<@o8y3^PRoNB
zc;sk!G=#fE8F?=Se6(a-nP^wRc99z}EwnnjdJ*^+s_DHgt!xCJX!5KhjerR9zV_F0
zDJM2BGb_Ti_I*kwpehC1z`k>C6TN-yZTfz#tq%{TU_Vh+(3hQs(VZKJTMmNr^UdU6
z71Dyu=WF;!kmJyFOgT;35+5^0&~(^CGe5nFd28oQ3D^8xI^}u8ys*pf`next?=D;o
z9=1$d4#jPzYVJ?J>n&nGQgs%(7>p{j>WK81X&u}FK&E3eE>lJrl5|m?HtajjSXdBq
zY29^Ru{XBiq&p_>LEN1|HwD~ptse!MyoKn@OQMZ)bS|i$9CRY6Kkhzd`|BwFeLMc=
z3r;$m8o68hm%I=E{#Q;poFTVVUgIhiG^XGM(xC^3>*G0$-;`t?se$-N&etJSYVb`_
zhO!61Cp-YwMyjXZ(brq&uoF_(V@Ogd8mUvnRXHr?S5%U5QUk%3d@E}HUI~usXsOvE
zun~fsGYN1Tl-ub`G6QpvjM;NAsgsF2<TbxpJ*$YSOi+{`DiIKPRRJkcvBPH<q`FbY
z=dT`rdmrJ~OGDKY5lL;c)SH~cBKdfvI5|I8o<~^;zf(`9c(M|F8`yy}6LV$w4rr<#
zH1X5L&-TVh0wpC#j?<-LD8AclAl}4S);R#SUxQ|}^3dRms|@1W`@xO*i!F-gQ}Es)
zv!WQT6Ib9h-UE2A);0myofn{h`=Fk3TXl#f3#3qTCc(^vYW_xW50FGoNInB}GsTkL
zUlz7LWh^Yvax<25AS=oB2jiBO)t&8z@|mySc><xT&F#Z7NIK<a63XDdsH!QttJ6Zu
z5+Kv5%BLS0X@f>=$*ATd>KioUu8%&=kJm^BLXf^|)$m8wk@rE9-pQVEQ>YYxe=UwW
z+N0y5AI@cZz<dMMIv^Bq&^TVUD*Q#48(j-yN38mj{tT{9pe+8wYLk~>p@Qk<%87PM
z`9{HNhVe+L84udvA^$*cveFd6y-AK><lC62ra=7SOKQ82qFC9r^vzf4J|ygw^Vcq~
z$;TKy*@!LE&3iFaWGFxuKd^?~%!l?}RkUz*DtxM?x1JcThHwXVS7X155`uA0`A5Y1
z4bv=7lnH{QuKvMS)mLP>VI8=KhjzNk89`8N=}f~*t<sQrh2bVcgjwf@(4D6IA}@4H
z$EKqJ9c76qDMqLGwOy`|iSvAI+MNE_1HcguZU;)LG{S!9rSp|@YkZ?}tI+e2mMxJp
z5_&;+K*_i6c%HBk%26W*WAU56ProB$Kw{SMJcH`Zzmf`BJ1ju#&&d}N2^4H$1exn=
zRu$bf1p7VRM2Cw96~MSCqJHw$*CcOO;E%>3ouV*=Q4VEi+%dZlFNK(zJ^8A%3+7Fv
zazgnCjjs*MXXY!hK@AaZ!1Rqskqc6B+y)ZO4TW0cC84i3W)vL4vZxmn`LsU+=X#qQ
z&h`QuMhfM^gBy}5wWKUq&EehVI2XaOYwk<ELUZP``txKgXz|;p?l9Ra^J?M5Ojkn$
zq0>a&K(|Q8y{*6uiB<>W!6yn6UZ_XoL$NxcB{V6r0Nt}2vA}0;>*ONrQB>Z)H+?s%
zKCaq7>6V{spp2R25oAy9QG^6RHr~lH%?rwe?~Ma`D4aQL9Cr9hkg-=BsAxo)`!Ut_
ziRHCvnEKntjujXdTrba$dh`v=Y_MWDGM=(P8;o<SyaIQE^4t18DPlDS;*+!vw=S{o
z!hODc0+whf3&nTB^AD1TIDJ8GsKPqo{(cQRh&{aaoS8E%^yMn(CaL#+oXuF-O*lLV
zJScPeitFvjp6xhU!yf+R@t&8j><<97D$`$E=R-rD?tKSR(^NSppgD<uD4rH5J|N~W
z2DhaRIk5X|fTqNt=U&%atxtny>c)E5Pexl`?wyooCXQ|0HIH2|>L4{Ak|VKC@T*)I
zt*d08-@hymZjKgTs%dUe-*Oe@H>zOel6cAoLo%(z61lvXKhoZIKr#B`T;jJ<QI8s}
zcQ5=0cj&*Nz+^puWTDC)L}Q!C{TYpM@n;pvERsTSkgqAuwuyUOkzan0LZVuNGKr4G
zeFjxIK7&@A?K}_u`U~lkM@RK4WsiuYI~CNg1m57?DTtD#z{~1`5WFK?E!7)RN=M1l
z7f-f4BP}MEr^KJ5EH-G#VGOFr619xyAxc<<@AL`ayKkKthPQ)fA7B23xRoy)@2piE
zM$sxWv=VzX8tff?Erb|C&^ixkC;s`S{)tF-Ql%nI?{dalq$1s;yCLt8Pw`9Z@OMp3
zhWK&_BH<knD|#6FeYQduuh!CJ!lW;r(RuzcvJcRFS+-%Ko_M6wG8XT#oC+CrPBdP7
z2*d$JzW_3_L_Nt7;U%*|kua~I0|@0lJJoJ8UauB_P;%nR*<kgD(Ai~`cK5y~zHn!=
zIyW1g+@!>$On%e^*AjatBXg~7{gX}Olu<lh-@+c)Xnd6+ep?#Ch_(0VAaIX1=RsNr
zlAUI%d9Wp@mNX%F6!4_yrF*r~mJw?s)$S-~-`i#qkb}81Mj@*Zqh^i7k;XluGgF4r
zF?(0)2NXHI8xE_kkD2L1m8|?x&b0fV8Nbk^5sMd{W7@9se_WrBy+8E{LCrz<gZTF(
z<On9&K#K_4E22Sv=2SJjnBjBqVbW5cxlJl+XWE*pm_9G@tYMw*)K@;JYkXA;WkR<i
zmPQ~}b%JVvU+q7a;@~T$^(g%ak;Mtpe?F*yk-(Y0P0rtj7RG}D;qh^;v=x?7bcc&C
zYg?F-DSqqR2#tI*XjigwsO-eupG{<uc>DW{P9;l^0_;gWQpRgqf9}a-GE|s*_<Qib
zJQ#t)4_+{vV<DRkfj40?)d!D_33TI|OZUBc#D~{OYQYGR@G^AlSpfm9H+R{9p_oRV
z)eD4BV@yEUil1u>TQ;C|ry>dj<C40fRv$88vKYeXLV<GD>0$W2fVG?S&97gkuHE#a
zgxgS!M4)>SbilV}x97GN<#BFco8j+~Dlarr)o@M6BmB=bCL~Xi@~+w6m>lKpkuya>
z(lJ8RkkLP}if^{seFuLdZYg5!iWk}?u1xI*Rp}+uMZ;$O+F(67e|M(^kg;PK_VT0j
z*{4K><8ELfC&IiINg=POm)s^YGmhqPYNcE2;@*#yc=@8HcYyz)0CHlv6_61Y!35fu
zV7`!Mp<$U5zh6%r*Md2R+4--#h*egQvc&=5XM7Ol>jk2yTCY*Ukc;$xmZtyBYgt9m
zl2pR!TL1IoYt!R9OjN1P230FXQR_7M{2Ki7gZ%qFe0JNul=w#+;E17c0$YGHK2E|i
z?cX=g1uN2Z;hqF2_ZPwA(_nJ7k_kU4)GuNOkwbMQ!FO+|&kcm-u1}2qX%K=tLU7Ol
z;&l!sq$0Q3d;VL<KLBSp6hz2h!mroy{yGJ+ugC)?zXBUAW++FyuS@3mJ5`E$77gvW
zzs|{-5`3n?3-r8Jn^}B3kFi6*-6mE-)%0s`xd;4JK6Muj(}3i#*n4Y$7czkV+cFZ9
zyz398SsP~s>F>A5KfU|EZJocD?f>Wn6C+@uC6d1F0*|s1BD?jM7r<Yu@mnzB|MA3s
zT*o9l+jUYIXZ{~P^WPcxe|Zt;gT85`qghV>J2Kioeg1#{`%6!>T;e0MP1yg@C;#6N
z%iTcqRuGI_$Nr10DLV_c6y?C)1AtWYGkab<7_WN##$l=!+MU4113cywE}le7aT#Ex
z8FoZw#)BM-QV8?nJ*U-dZq}Xt2pvHc2bO<cN7({q@No<1xJBds+@tesf|ji~zd&OD
zPAgM-huX9~cMf|OaJ1s{5y18E71&(e06I(xSM#g{O*o^fN3W_J#&uGFFpK|jgf|}9
zIH>M4@>l2$%dM0OOzSGVdspe>Z4Bb$k?qYFCq+JB6oe}8*LDCG5-#c0WY;d~Pr0*9
z`rVQuc!?TO4QS@Ne_88)IBpwM`Gqm(>^*0xM2q_unV<%af3B>$Xut_y>`|#;{MUo<
zQkz9L`L|n!5n#&Yhj?IXWd@4RyEO4TRMcyt&NIA#U*nCY{)NYW0#>GK{VY}RYYzu9
z_{3(OzKsKvvk(|SNnU{__I)7f3Q|?beB}YWo!r`p#kzpAslqph0u_9UogE)wjW$pa
z2|OZXpB-)T0Wc$V@e<VY?|^V@u*{-4tvjk)IPboi%6q_IQ3Ld6M~y61IAPBVFiu>j
z(+cWu7-?_bA;od#%iVoidlZr5;H_HBvudtOeyV5c$kOAo+i{S}vbi&sGSGsFeWh!@
zBZ>l>;Xq8uwJ*X)0*?Z_<+n^*ZLln^^nV@@Spp6;ciO1_NkKJvkmQ+r=>q_eFo6ks
zVo`{34{X3?vm!LM|6G&i@NFPfOpBV|^ifW|!`}1UW^fz71MuD;V1kW<DfGOmW#*0V
zfMn6~Gf{gW5F&$yMoftYE;m60#wnVo#><K3sj(`}K`nfCLbJ|~@nZ!6Q)&f}o1<??
z)o=GQbLCWA<*FD`Y$2ZqfkeY6@?IINTpkS0rm|jJ(ictVD)pi|O6z(*;QBtsJ2Sk}
zF}I{x{4iy3WSMW+yh2^0M%+)=A>xm90c`{=YF^k&+8_DNT#?qASv$YnNZ79giT(Rf
z{_jUQ3!Ku0+kc+?KypnCD?S*R0(yMF55NOStZ|&2xCH`1TwIjwIw&?08U2psmip7}
zZK#I*nc7XzbT}(xh8ExGu-rCN({HWD1AHH*SfrlqzXWZH&adyHyN^MWC08^3&Mu&a
z-3PO|g}ZfOEHHPp&(%$$GdCL}*y;Umr*}7N75JNu=hHxiBvKX>{kA^>?xl3Va3l@Y
z+u)BlaENGp&w5zNIiZ+Pz(w=PIg(H42Gn2o9lPGijG~kb__!g^7Zb5aVLIjCVNyyO
zAhhSO_n-WEDKTMT!jgk$2aFt|wDrnuLc=!!f0SDYG^()5ANJ9Y1Lix|Hcu%g=ewDF
ztY-O0fr`;w{V@pms~(`#;7TS|Wn6*Nysz0`*}Q`H*S-dS6y@=GXmeEtqyNF|z(=1I
zWu%1<`#&Gpv3fM^7!f1p{V&?Lb~gclwOit^^HzFgBxWIjS63I9EyP^=kbZXX2r#A+
z0B<-Y>N}t#^qXEWEYwqVA*E>r%5pUwS2U{_z=$KO&-JSSGqxMR8!zKJrBH+x0Oy^y
z&K@<D?tIJls96RzEp(Af1Wx|PgX>r4*wGfv@x<3eli1n;53r*mXCquopSlx65?!+c
zDppd~Ioj_y&<-G~-UWav6?puF3@T$~+a^pxebICVY8+AJvpqsyp|aT}0b0QUbSd5L
z)5&nf(}n2y?9;YICzlsj_I|Kz6O9B5N7<R<PHttq?I$qQEFA@qW!nh;l4MdMk0HMe
z)ODRApL7q)kcQi!7?x%Jfy9hnOmplz;P&2)>P(ved}9ZHMTVt(lVz}9-qzuKO5<_3
z?DcT^D5}(0FWp%_I(rt-VYrWLPr1fO{h<6_xoubpbo632DIoE#=QhdO6`O?{d)Qkp
z2_d*cF?)Cm0iw08C|YZ+=k}Unhn5R3gx%?4#f?r*Lu=-*etzJ4ycZvG%Gu2I#r8wO
z6g`%8G%*a$C}7dlP|}R7L*{m3n^sETy*+G`d67i)a}@BI9yOP2-)(b6(G%UBYyvl{
z^-{O%+E=NM_AbT1X5chOJz1Dv^{2?>-c1-O;X&Eam-`nqbR0c!sqw>`Ntpw`JE+>^
zXo5fay+Fsm4(ruFwo3IDXq%`9gspB1(P71a7SeM+aT0s-5GpjZpbm)2z@Gdj#Y5RK
zD4;Ms9xY9m)yBlUmnrT0y5bUy(-0D?&8OIW4($g@*jHe;pxp@BecT5^%_@J<ZgmH#
z%>YJntujR<D&Ewx)NK^B_!ORBEtehvU6!f2{v;s<bKe7f_snS0UHI2JcUzFMzny*S
zCK$~iF*rEiCONft1DF%&fc_j@cpR99+HB9Zup4u23o9uO#g2_5qk}`&Iss46E@%Kv
z8<qa`gCdm?61)L8RD*!6k!k`mU3Kzv*)Deg%(E-9K#oj7gBlH|)mu@F?ks%j0Risu
z!qsxjka&<KTF(29Y4C&Nn`UU~D{DF=@v3l2ePPbP2UO~!Xc&vExKfX$S^SV7-1-U$
z11HP^AA*ESWq>)ksv4c<#{i&bU&5L3Tp3iN7BJ^UH}^MU$I5_@3^O<zxoD4-3cF$M
z(=dd2kpZ8?&#Yjys<y^Y_kb(0E`8_=KB?=52xJ5PJyXWN$;kX1Fv1+SKBctLZwwW@
zNdrP_9RQJh8IebMg2{mR?Qh_bB2U5bOAdd0usR$hwGR+wv7RayF@G?;RTX5?8$P=<
z_+(7|v7k7@-<5kpz>`l*fBU7bri2ojiUfe_%4dR`%W3W}&@9{w7S_?7@F*J{>wV_b
z1F~L|sR5{;E*e1uaRJ_}Ecg(4KiV}^NxD~H-DIpsp#CSzV?h1EWS&T%Cl%~lVu+p~
z6?+c5h9)#m0hjd}JYSHc)QHKgNBUmzrVyVMOOQchu7J{SyGH2l2Y{JGw0U2^PukMx
z@<dBMYhtTrdH~4JnrhHgfLqyJvwb}FiOtUSe@?tWnnnW>)yOExwLc-_Kn8qp2g-dB
z1@tNknJhc}0w#V`7y@>ihj;evHfVwJ0d91vFXeMo34m)KdOiNNyBI{;bu9S3!bT0G
zVNe}cKwWH@0;AOqg)V{9kC5dRMFYz@XaCbwKs>v-uK&nXlySif6Ot+$w~Q5X1?{a{
zC*$F2If`~<9?bSf`eD0%1~e-T!I4^5>$UGwb?!EwU$eP^wqbHU88zZ88}zm-oTl$k
zyl({D3U(60(=L)Z){r<PJn?bYp|uZk$^q8df^~pS-v&Q=I#X-{^0=XK)gxvwoFLI?
z-WbHq2qeI0ErYVq?8p!z_}B+$3$zfwSD0amrE?qfT8KP)6q6xW<g#$&$a&f2lV0(I
zE0m|-E51+wx~y2}b}J_A3j*l1BYun#fkd>nz4DU%g)f$sB!0Az08@$tn3?CGxeo|j
z7V)O89aJ@Vyr#8Z+N%J@R)c1qq7|BSq7F&2@L;Ju;HZ#@wqBTU0@ybEdoYUBKm~vm
zxDw&G%%)e3>3a}P93r<{6->Rs6=)F_fbUH;!h9%`#t3^0@&4dAX@I7JE|570kJy2f
z$U$<dF1Qk_l{Nec^YLAv^DXgdZ*jYJ2y0Fv7?}Y|cn8}t0CL|%^MPZoO&kKlrljF}
z#p9rK8{Jjd$8!}_Z01*_i>-y#*VNjJuC*<4agBm|7;Q}e^aWMr!TFbfXnNaK^6<*K
zk+=nb<*nV9z?E^!HMRM>iJhdjvgk5=$S+G8#@p_POOkXdj(s%gEMr1C8z4m(<dG%-
zd4F0mFn>ggv#x>{$kQ)||1fyuf+?eOOXlpN)mfxiES&WfhgsL*8E3x)nXSxVe23?6
zFJ}jweQc_~DwUVAT+A+jUGB}@8I@ao3Cy48ibiz<E$%HAxOD1bM&-$qYAMm<?&gbs
z(iL;dAhaWF*i+6A7_bLye;Mw<M#Ac1#?=Y_Ww#;XygUPE;r_Cb!eY(phGq1CH`QN5
zK}mTd1X5HbPq`&kpDp`qtBDyrH{RJ(R<0kYVQlGkv!n~fg`K164MR$A^Q^G7N=n&8
zR7>QZantV(rDS`eF!7#pFR4C@#iL^Ix&-dQujhy5-}OPMp0dqDYa;m?i%RRc$r3l&
zL4(o2pk@Jdfnn$2JvGDy-JeCwLt&x=<)d&7kigz~o0GFi0WUBpl&r2#tdL$e;x^tx
zEBmk>XFQ!Juos&swxZSI8>3Qro5QLVnwt>`m!$f_+8Ar?#6d+n$NZx4o{~WZ;R#_C
zmi<+4j!r|RZ@4pfH}yhnAD5Hp=l{6wy<VOXi9_*Ok7=U-^$}$r;H%T2m0@gjpw}Cx
z5_^qm(+i-lO)=OL**Pumj-H0wOU}broJ~vD+AQv=)(v<Jw5PN@El?TJm1~H3+bqDn
zT`7?>LU;aE^~g0Ak8t1+ec`S1*%b`v+Banon!Gs)!&|o;8*uRpEU(Ygx*vRt{50;;
zGdc3g23_;ynOw=Ck%Zm(AY=Nj^y)7tgF8#{E$Atk0sF%5O5Ch`0u&bRf9?(KG(f?E
zEf(n0Zp}vt91bm71;+Nu<{OktRu?8DJk*=hL~c9CbmHwTNKD2yACHkhgN&yMc>1Ng
z7J~0eJLM6Y$S1vYlnHX|3ALf5M$^{voN^igDu$Q(dcSxR8uOu^S2OZcu(A2ASMkP4
zRi3ph>BhTzL&DZ9jm8ge38`woOjd!^el|#gbUDS2zfh)3uspcQkET{|j6OwpuF2sS
zZJJ0`-;ZZqx^MV!$XDL1DTS7PegKd$8fEn%@6!Q|Y2-ba8B&Mx=Ga#FI2`XBc*Kw8
zuFluzwfcTUX<Xz5;P}Bq!fKD`w?mE%rhSczn0SM<?ZH@!(3(MlGt||b29R-5rr;e|
zZsr&12*e8#EQ|xA4SVGzltAfDc&bh5+FR{M9WR+CrM?j?AW4Shtsg}V_f>3IW($Z2
zZ|=xSjb{WyPcX|f>SqpyM5@ssqTclFmNlsIbsH7~#I5u=lIA{_7ic4ah^N(s4z^jw
z)fy9*L-IL!)Z!E?gdZcG`B8pK!(Y=K@(YRoEdD!?gwqbq<ri@s`Kz`d!+*fC9sH9;
z{R?@q=p<<wM0*yyvOfA?C^Q${Wv6IDrI1UDE9{5G?#uBwv^;xD=oz$Vb1EdA#5geS
zcNPjEVqDb44ey21@us^ts^L>qY$zko>yXi{e~{scG~n6%cL8-RG1VO83z8f^+#L;R
zLOUmnTtW_{5zY?sjLkMOh-SMLv`DGG_;DE^C?Q)buE%onwyi3gqA6*baZkfFhOXxw
z0dn#~Q+0+@oTshxkCSAz%tR-BRM_sj1Ix?OkELHk+X&f;k0n~J3AYP<ASw}W!*N{~
z0W`eeywayqh@Hm?CrnPy9~}ZsnyB0@9KTiWsxQ<MftmH{hj1I-H(C6Cee&bL!)cHm
z9qV)n&*tf4#bH<OzUn!_Pk@qVR8HSY7jy!RT&~pE@ATqox}l&(`q!gddxMOYbeTm}
zEi?9i`uv7t^6Ytsv_EP~YJCr;%a77w9AUlSY$L!cpMmokPM_632sMA3!Pp_h(F*Kh
z7%gpte@h^aOhJj}YhMmn{LPjM)RBe%{F0r+&#yGS`re_WJ~W<?iaWgiwnp+50li1e
zb}<9pI%Mh-Rhx?dlR<+%XeK?BNHd3@T1h6Y8{L*YsCNEtK13e#j;9Q8B6fkZdqgC?
zY5A*CU?C8@zdo+3H_NulBA!>Fmv+-mqkHV34$8<josHXo9H~j=F=mNJC;T8BE#=Bq
zMY#dMm<l1qwdx^!TchR`C;?!8{Jd|GkI?Pq<NKQ8b0KhDLP-Ijr38gB5tO+g|De6y
zJ36pFhMMXKCV_VIhrd8)$~jiDn6`)wu19&1I!d?&5e+L&ZbpnPyMJtmV8q8|hsoJp
zsv{AQ9k?q@n@19ED~7V?$kVSS5Sl6SZAKV3RW#!7-J~5lFHHjzN^<07TAC90^N|`r
zok9-HuFUU{QKSDfrD<p@x|Opj0`uf<eGP<=W6yMotaEaBOf0yWUV|lzBxJ{B7z(nV
zcn_??FUhGUBkFsr5y_%Em*)K`B5HR$);XP<e!ag9NLe=2Rh59DkVAc@7;34y0ya%P
z?;HAfThj5(WSBqqR6<_Hv-?2ILv!D{J)F$rMd4G?{qqFcScicY&U2U3tNln|q7ioZ
zRVgU2o{ENQH{xO^AgYGvxB#IBjmil+*SN%J=~p`sCaMD?6qHJMCjIbcfH<&PN~j=!
zk`y`+JD)^hX%_K+0~iFV1NyMq2H+s%7?=-^#BGEb8q!{fqRCW;<*<_M<S&kZ{gq>2
zrCx>kDy&{Z1T(y6p}CWJ3ebKwK0Wm}_srb*^j$mz_ODyq^m1B+CIJe|@6kw=0d;nT
zy?pj4ARVaIxh;~f3{||#`VJAyuNFazg4BNmc;y?+Dw^#Qc84YH0-@6z^t_~3V$&Mv
zkoOQBs&kvoFMVqqx0m_)Piqee1oiFRYi!YKmU=WN?(uYeHF|`fV!!42lh$k1l8>xV
z$lLszfQBT>TvlHQcov1QCk*bB@r(!bHf4bjt!bag`p?veH$0$0Dncd}G7TXI=C_7H
z=N5bqoP@wdFe*Y;Rd1F6<i79PF#{LD_$?~n)nYFRxJu*Ol3uSiGU$|OmtQ(Y>~w*6
zJ4lKs;<0-?q__%{r7qL=y+dbdb8P$rW}W|}0!};&$wB}-{GfB9)+MyDk&A~ps7Z6E
z@oYf3ab&CI#dvVW`9gwhBsu>apXbeQS;yKxa3UujFs=P9)#^%@W#@l3Odg##@`A90
z)nxRsi*F&dm3)tWl;%v7K-15H@K)VGwR|=8L?M#V)KXs4N<`lBz}jXd;`^1+OrT)$
z99JlFiTO}n`?D_|*EF?|c33wmF8k`O7)cOmq^N41aXW1f(WmB((3YrIFD<5np6`vR
zT_-SpoURUnKs@{?Z7O|PExr`r>^EgQ`5b;Bo-rp+zWK3|z;%OyoA8!cqn+1~X=0lO
z+u<j00P+wVo2L7Ro70oJ_PJcCFY1bdu4&<ls6GbNMb*%yV@O9#0pUxB4b(KE5D{5E
zd`Xu2^+cy7HdlWR-q{#<?df_RH+<J+Bm}p1)1c>pL5=Y?&r4PiVOuL?-(JvwYg$F$
zz_ru%qwSs18oL$&o66x6(*oD8scJHUnKjcMlhGrZ_f<~h`A)~D{K0ca=Z4)pCN;%*
z2?<hEj3&QaN@7Hc<;5oPzQr&4)BN#h=A8c*d2bn()z-CvD#E4|5R^wjx>IQokdO`)
zrKJ(+1_3E40TGdwmJsQZ?v@sj4rx%3Qt3Klxplv_U%&6W&cAd1v7hbr#9DLBImSKi
zsH+MY#aF}ot*y+hc=R-W7UPM(>gtiRj5oX{&+v^;O+Y@GDQS*s;uo5409hw+lKB#w
z3DeK-Tj)lpdQnKQ5VQz^5$?zX0i(SPib8w+i;`b5qQVr)Q>`3@9YtOg={ld(S7r8k
z4`hYUE0xQ6<O^iYgfO2fMCL^EDp|_ycg&;cZ+j23e0ZLFg(_#<;w=qMXTjdt=pXcZ
zb0Oabf`kkfGomA6-k!Wf{%vd<m9lyAX6~i*kvrQX9|7Z>$-=3*>P3N`ib!C$fh>fQ
z*+$g8dwwtI8JUC3txy@<{sD}x!Ik7&{d5mHN-!G|Do09d6mstqk5+2I9<nA@p0^qz
z9MI@(kA)`S7_*wWuVgN5!`hos-U{F53V#J0>P1fk*~f&r@Vs=#XHP!<ad3KVrVbie
zdVqhRxIcmGjTdd30D%;`D{7OtnwfsMV_u?4yrXxR<#;zAn7N6XZ`g|QJiRj(rER;g
zJ@+Tv`#s9-e#_4CngPkHu5I66x90iDlL06_F`Q#cty|k^fySt{I&!-<Dh$AdDW`Ax
z?1O%^yO{v!`j&lFFP$5Ckc@8LY&8n^@1d`pHp)}nB2MATFX8pX!4UP>vG4F-34J{M
z6It8f?i@FN(XF6|_TD}R`5|sm{af8)mJ1;syFRm6xOy=4E%~0=S3eU`seHqfO<MbI
zZ21Zp!81`Nip3N}VGR~EhMNZLofRaFvMqRYKA+|pf1}SU;eh%^RtW=z{>^lRZFqBH
z`EklER3dM*z8E1jr7;^`BWu@N(AUP+ceY^S+PXu4K<RR7T35E-WC{X5k6gPRar!-7
za}I2u+j_Fh;RK7<rpsoA0OrcS-yNH5-#y9Bu$~PZCG7i@%`Fd`kG?}#nhQnN%P&AO
zDc=19H|#~~5l~yVqGOibBWhZMPwg#=klL{8*8u0NUl1HaEZwMLn@Mo~zzQ25^~NHw
zr*dAMa<{p>8J)^5Wtqeea8nHoGF>S=4+=5G&gIw}xZMOnD$z5;JU?ut)M8{kJha`e
z$it7WE6-EfR_J<#7&0z2f-*g9A!?bZ5A3?TSa<xoS<-@Bg>K!4w)oN^SOPpwxQSp#
zyQev-MDz~za;M9XV93jH`$yZWwEzU%m%O-)yBLl8_LBWmwK@-Xq7yUtJhw5vx8&Y&
z>i%AKwu6-kx*yLLcf7|Xfxq2?pR;*cJ{LQ_srQdMj8IM#*mh6kavb-l)RpIGAqifg
zV=$|QPANG&!@Na;<h*ekFc8n^ds44Ln}4n(offdw@(jOdiE7PM!t+YBEz_@XFpWTh
z*;nakE=IQ3psw6Zzo(?upY=3@QVM|yfR%CH5yfDp#HN&f{Kux_%BV^Y|Lp0{BX3@3
z1gW~_Qbf<g3I1j)<BI(E>%|VA`&yECzU#1YX$UZ5MIe-}blXsqY@BupwuTDEszgXi
zPoZd#oUo;D$u5_M+9NvPj#5a8(!?Jfh~yX)DTF0bjLaA0(hzNCMu055%CL05ze9OD
z;dZ)0j0A~O>I|<^wn3}n#CQipG14-}10m0Z=;_(vtbkS{Syz>9*OXOLBLJvOKfHPv
zbEdOgofRrSWd}NDvn#4l4rO%<J7%&la}tz@(uOnKi~_>$WJ**Z;wQ+`djQN=354?b
z<;r7b?*eYUz>Ow#uaYOeICB8BZYuS`c@wU*=v$YgwW@-}GTfwfLGVEi0uBkMSGU&O
z)C<lg=wG@+RfHf#Ov|mfIl^6%UnC)mE__oAgPZqVff_U-9b0H!`JjxhpYg$H<tXzp
zTG)$U*7*i=p`gl&(nAGpH#wk8eYvD$n~=-!<)YdAtths;_m{WfAE0b*zqd-sG@DjX
z4lsU~*+W*{Jd!-vMY!wDH;j3On0b)sOEkY=FJ6Ccqth5eK}a^ElJW)OE_T~s={#0k
zZW$UoZ7BKA8)-!d6D?yd7T^q`+nd28G9_x>LSbY7l$fL?Zyj7{YM>XFO$s%*%<qvd
zI=}2vc-K(A=QU|G_0sbK6&twxfD}RtIu(k2cfYIaq$d`WaPa4$h(euj?|_GQr822}
z<_ltmoseXv@}Z+%zQn?OewKW`*!066WwKOD7aupu^~K99#a}zOEpWjCKk0W&7|I{x
zSn*(&pZLH_ENsy$C-ud7Pf~O;+=zeChDJ3B=zSQvM_{xUX~`p6khkVsvZ+5QYYiqV
z^rXBN1X!Obn|&UgqP|x0<UYlN4KeaY@qQotjZCyYg#U*&jpa)MwE9x~5d@ybuB7be
z2e=J``f7kI4sMLcQ)z*v`*->WQZsns8=sVH|ES@T<lW#P$CZdW0-IY*C3As5h$-(c
zQuB6%2WtLRv<}XBp{#i((DC|X^HW8Ehx`TadLT$&2MpaXrA7k7)GEcJx#zJ_JS%A0
zn0zqti5-*7YVNSOp#e|cg*zD&d;u?+tz40{`7C`>-&a#L9O~H^^$#`I$(tS5K5F-;
zqy*ill+vc4v((uYL^<Br>Fk&%H@LR0wj58d*HY^%oLc}RwnCd}@i8?@oV&GdF3ARZ
z4C_AXsAYm>a&gW%Kbv#&mjE21S8d%_rUI}b>x;#V{e(H?wcDs+#1NH;Bzf`zsUw?t
z%GCf$OuS1b`&mTKv*|cHXGbwP2+_6uFiIMY4bmJw*As~|1awm|de;*gh=Adqag?UN
zBGdZy3frf!@S64_m-nz(vKePnzQha<6((J;UhuB$E}bO1aVuo<;?n)W9*byt{(LRB
z_*pMbE@8~sFPQ3j*63;n*OZ37(#thnJ%x)!VauKfgM=;2GiEV^jv}@#6rM=16Fo@_
zg;x0^5rPH^OT7tk3$m@ZD5ZB-`c=Xfe3z@Lh_?kkjW+z&l2QVpES!7T^~*6zQuG8x
zsJ{E<xDy-NAde^1&?8#`aSgAzYhc?szMWnV#BkC%D5SaIJ0=wgYj^$EOF08n!oJQK
zYhbGpQGoxhp^*~ZD^jEwdvHkaWLA`>o=<%)6cG>3EXZ!ge}3y6O`+=3KalRK;Of_Q
z)|(jBtBNIYa8h+%&+;t>7C8!r&jdSSHr^r2nP}zLU?CR443%ixi%z{h_Iz`I$MTcR
z4x5cXM~7C#1Qdu<!A|2<5-5gT>u=xl`QuDj4sGSpUOoY4*4RxYr(|B~3BfWCJyVWI
zJ?>{ecCls8eu3$Z_DvjIbAq?}3LHsP#m&R_31`L`D7%_IavC&a5v841t<#<Iub4$!
zU=pX_j~BTaQL)ky_%OI=NHp!~wG4ADFO!SrghB(3RV$7K4p=ODSOm`dk|D!mdVy~W
z9+j+?J=Roz7?5H^Uh!xVOln$Mo!XvLu~*jfEa~{5r51eWW9-;lkG|8tt+b_(V~%x=
zirN0;Y``ShQO(IH3aS|gDp(Lgc_2aS==gqVobd2biAF#u6>F<z(;B`LUgK!$8=HU!
zf1L7tv|~tcK;E!M2#i3&9Q^o@Oz{Y=(?(In0@G{8wn;umrR%@-#kQk^`=2-WBIFmV
z>ZJsM4aF?gl-B?$7Wz5|--Q1<;rsOqBwqMi1)7U!=zIlO7X~EP2HW~LSXYM17bQ#=
zy|$WW$bJ26T!SmLK^Gxj6)88RW+Fw;$cqMw{a6at<c4{sQ`mFjdge!pp}6^BFZ>tv
zWG|gB4gB??|H0gmp=5@EP72|@zl^H94Ct=7?>SU}xo-C%eGYabVJ~R4!&om6o1ZGv
zn=g2YRpz(zBKQQ0u#M;G?bZt4+!!PFYB_^M_=>V$3`;G&>8@baP32w8d&z};ZpTPG
z^In?Q{FQRf8nB!(>WLRvAe^vnD7{x4mdnV7?u-`|6)HDzBhe>wWc?RS?BB>y$bX7{
zMy}syGQrZntb#<b>v@)UHfF=sf$WFOYeLUOhv3`j^1JQW?1DC&XhLvzE*$aq4P3Oa
zD!m^yV#C~SI-5vDSCOx8=`*HKj_*Hk4W;+~>pyY#*h~5`9)Q!!_;MP{xKI-_{rG!v
z)_p{^h%n>2ffdXI<N6zN@BCBf)?vhtGycLL{;Z8(=O#EZjqU(;#$}9~E;&XtsCpZF
z{Ss>?O}I-{>6ZN!``Ze?5#Zd4MFz&PQbCsgCdH7E)dK;a9dMj+(8<~P3n?Vr5ffJy
zYnxdz;#EkeyLd%Uw2^rM7(>QZb<=yUfWUl(U`t_E(%B~g^&p91VMuTYt2d))gu0Xp
zs{h;w`2KKWK@J2@x$$M{@tpsO=qK~i7hx^?o<@)R#ajN^2Lib=XDwH!=4$avps#lV
zx@DeHP=Jm=&tTHP)sFU3-U$Mx8FF?j_;rG!>R}c+|6ZEnyt<~t$*A*iWVRLBivG60
z`N76$6GWI&6VHpT|J8r~pZ?et1tN*t1<9?yMCkwd9Q@`L$~A=O=XAXT?{|hv|DEnP
zcM2i*>7N_H{2gxnFND5-CZTv$)4<aIc!o*vm;crO>BrMP$4AJ0{z~^KemD8}@4x@+
zGa&c5@TGkr_}yUmZ=TsduZZY1SUi_0q(a^Q@e92!{Ik;@M?@?2SFieX#khO0)^CF{
ze?;G(>}#?-&<UHR9=R^)GPciLCa(Sq82ZN>{mXSK#|A%4KBcH%z&~83R}u9Pd^iUk
z$@6ROdl7x)2O}0(iU{XLf}(WfK53?AN!FM0*=K>}ZZ*4K-n^Rw`#r@a9jL8dfs{_x
zHs%^q*Z-3{o^dRj-Bm1HpjKFW-%(SK@F?djsxV;7xm(5_Y6MA>VSvKqijD&fC{%{1
z8lrcs=b&J{DWR6FstBH9uhDD?Zl&XZQ^;kYJlAe2X}R8)Ri`)-@Z0BCKY+pRSHSLL
zGcGJ!BMd-JqmDLs&ARcM!!qqK`L&-?l>fHUM6V&1UrAEo?mwlKUTg{{8k3>eq%$t;
zCAK0D3XwT669~9iCs^i@7p8qFdMlj&PXt08%2glHdw}irK}-6DL>1=_{)H=|z^#Zk
z5oo@rUFi_&akxh}Q&e%X3|Q*zfc{WILD@rZo>olZw(bQs_ID|F!~#Y`v`D@I$kO@4
zVDI8z277tcVD;Sz%()3eFJ$93g5EE12g=R%<HO10cEr~sAnvC_U*O`D;w7V<&bhq+
zYxi?Qi5i51{+>6cuQB81Y=C1-M^_9N#$l6&uW&VZanbj#f~UJ)4aa|7gnwSL`av1k
z2y$)vM~r_yiQvgt?egu*DIs)`)4S3qfK8Zzqxop8PdM0BX96bNeQ%6!ejdz?t6w4!
z|4ARbTglR*bIsQfOp;`Z4Z*v#56A!)g&#Ulv&Y2&$NfdUH4H8ubkkc$6#pO+S_OCz
zq-@bdm)3g^gL)v$<ofpN1k!~T@dq3eK)41K#0bb~2<@GC{|XnuVIT+k6sOIUXpr}b
z^R@``DvP5v%h;sLI3(@1oLaus1rCG<_Bh*08o=)u*vbMJ7%RsBuI*6sLP6`3BcuVg
z10FR-_B_}B@W@)6hrMBp`Eg9bPbX3^l@+~GVll$00foE}Mo|Qz<kUY=3Gb$@JaN?#
zI7y@xx3CmjDs$4Jq2s}%yAxt8Wly}*FggZkn}K*K@CmINqS7y4H52KJPTaGExQTgK
z*`??}+>AEQtB}IG`iXM)o#&CEZNa&i8|w!M-w;%QtZZ5mO~R{&A;R>K8=u|+XKl~m
zIC8kX)Bj;SR-r}V`99!k{uA?e2icgOA0=THSFHu=lt<6Y050syP@<^-%eEDq?P~V=
zSc*id0HeR(E&kd$ymCWff4e)I@#Yt$>p9IyEVn<+@a*Q^T1=zQB&waea@bNqi{*J_
zrnu_VG7$3NhM?k)u{l5Mn&axY2I7t20)|>_tDxhJfUH_Uqqmd+*8uoj5guR9(=3sT
zV%LtfLDd`|?RN^URfltm=rk)mu0zWUsWUk@RK1stu`wRG<xve36*~}Tc7S6Z9f$>j
zFmdGn5=#I5FYywhhVy~^41Lv~?qvm=1c*0i*U|@M?^vb&py7#jzC6}9PEhNt9tI}+
zx{c&GumjEKr^$(5t@lJ$n0}ylU2{p)0_&F|yLma6X+Km>{FJ92Tdu`OQNt&+O&T%S
z5Odi*4b-)+yR$22+!9;IXE&@vh<;LWPGgpQ=7KX*%tS&2dQ`aV8lW5;5@$Q4q}Zzg
z?SZ{~&{4_?j4C+Qd{p)RI6nCXX36SR(>+87a`KM^b*{V}uAE9DWoQ4g2ODK4?7x;S
zPB42j#KS8%@u~Hc5C`#vCcOYd9gI$gcOzIYQns1~KIe_}u&Adujyf2gIMRDuLU!7!
zw$i!Fv8%#?3N?MRfIMPt(4ew%=Kkk=mS~TiZ;KB-L8=*taUERPK0m*X7YcDQ3RX3n
zsw`iD!xfE8Ls-dK3w8CAOvXxcg}@`IbfK4`IRRqu5YgW2t`HRB?L-;3>BX}$_-0J4
z&{!|g@Qh(1;`6!pspqt(_L1v%l#6}~fh+z(b!W&M8jGIj;!UM^*YqT#^;-Tx>2t&;
zb;qZj!I5Yl430UL3jvwYyS;to@>FT`|5TNf-iPwXvOD!%?Vkrggd7?{8bUAGqH=?P
zT<&Ehpqa^xp|B*TGzrB80$3@j0Y<M<xK{`x(HOuFMKjR&Tp%8X@1#7?)Vk@&wUinq
z8NSwh?~|urx{4UsEhAAIb1<V~S;zB!=5-3Q1FSl&*E%AY45JE=;bvMfG^t_0uvMcx
zCr-Li3smtU7@k)^_(Sd)leb_KIR{`(1bDi%u^Kcm6E{<X(IV$|%tVz_G%^`EH|7qf
zAN_4N5IoL9N=e~0F-DBUm9a8DTycao7;q_^<f1b|WxfZ<g(^5|SfX3u$OpBp0kau(
z#XsPWlNQr|fO%YCvH_{*&`)Y3hu1xbOzAqqy6_$`hSu}m$bbVq*dDTeFwo9<!m8Qr
z{5EM!qxWPK-?OHoD;s#0;(0&=^9{f;kPYbYrbhL1c^o*qfM%0u`IPwz1RZ|oORjfk
zZ=}fC1~MgWN8`ihYB*i4Q?q}ezY*2&)gD;FPqh+{Ve^uktaY~(*@Ln{;ovkX1;J=b
zz9QgwIh*zf^wy_c-%FyfJFY*{!#3cH2)1Qog(p6nCT^3%w2xM!K+kLC>?f|C0ALHY
z!9aEn#0#a4!4*$xc`Qej3(EX4;)QJ--C&M3=7DGRLX9B2X`W-tWzK(nj6!a7kFiJi
z!#g@>-Al_fIJm_WQ7ka6y)WowgNH;C+$ZB=T*y{ReSDf+sJxepsm=G--u9oQ@;_gB
zeiJDYU%$Vn`)83@ii2jeHVhm+BI%b+fF_0kvi-V5y$pDCp8-(CK3JzBLt_o#LfD01
zqVpL~<iLi=m=cdT2A-yS20_2<=7^l9e1viq`~r7+qv{S<tou@>@yF>g0yYqQYA~Df
zJ7l@jpyyrQ-(E|KSbf5hcg1TM#F<905O_Bbmi;_zWl^BZ$CjWXAQ#i)f_b$09gsJm
zqj6z%=xGz6F{*#{u{o2H_ELo@zyb6OmIf`*-NbI@!flqPr1fG^z&cIE_~=9tU?Zj#
zDH{a&U@yn=9)JDnYWXbTJG$LJSTRpu(EEnA;$VM@nq1vq9tGR>sS0>FEE7ea7lI*a
z?6Oi<FnAP<4R<97sztGCP=l%^F;L=gkZksaTIOhU@FONO5iL@?1U2<I0+)s&Qi*iH
zRyp(3DH=|Q(Mgo5uMwm+?vbLae-?+872rkH?{E_B^3zQaj%h1@;$e@{_gm}>DX=RF
zj^i?ZX8U;kw7BV)$bxdG4yM;HN^UlvCoa&vE*dP1J9tc>w~60q6u{I_T0kIuG4Gkh
z;9JBs7+7^*ftiPeK5nCUdYXNU|0#CsrKbh?*I#0U>RhS6O}fDMb=E>`aysvVf#}>@
zkB~#uH#~2ZI+c#F(rZub-<`JG6hXb+3ur{MeCy&0kYk+8C4(W*e~iW8vQ!=wz;hlK
zaE<Zk`|D|ty!^Kc^vCOS!Y2<e^!&BZZau-usBgUUp2RI{ROIl}N+w;}Q&$_X)KsrU
z#*=rb{G31>|8Or^I8;|sBk{?1p_0<*0DC^>sc$5_7N5wElWO?9>g>9Xk0%M`UNth_
zr7H4e<=MjKlDfm)Mi6ZI0*l9Inx%1A&M1a6$Vh={OfRkQ=+;@LbXJZB?O_3-uL|vt
zC-av@wm)B6wC~xC-;e*MvG={2JY;;^ePU>@I)C}9FnLY&$ISe?otDegPGv#kw)uM6
zC=W^Yg8*c&ATc_Ehy8J3RU5sXn8BR>p;^RP3Y4rQyK}%e9s6>PAhwh=xy!w|WcpN)
z-VDA-t)h{F5c5fD6iVu@RfW-<cVo`pF1gS><Au8>TI}8(Y8Dl3_D+RYP^g_qAszQS
zxspx_PscSlvlE@G#tdoM!JhD8v|EOQHr-fXP0+{27h2q5`&m==bl);sEA8I4nE;fo
zF>60m)+QMp8uXfTcQl$k;%Tv`yzJ@$G2&eO>KvOoy(MX60V{Vvu_0ssjZrt$OUYGE
z)Cv?PF<Glx#9zntxx>}Iw6=7IwyvR-D{!DL`d5RO$rX4poy4+v<@E%`Rq2wq+Wce{
z>U?$_=HtrcmGzeoqRKyUu!wl{3a}x1bK|+j4SqrUfi<ter-LOuluqZmtAngbmXAy_
zxVRPwbxF-Un-7;Y@q)^aU>Bzt9EOi4HMaFTflhhiB$nt2G=fL}`6s2As1+aWum2lG
zgGNYHaq9M5o_(lb_W%LL0B@xF6odX0^cF-&{9B1idv(5Yz&yKZs9=*H$08A(Tw5_b
zvr?<8=MX}4jK^%8`cxT6Jf&8PQN}~%m9y$;yJT+b4H|1eZoTA!H45o}>>VBN+}C^S
z#JXnmPulAaIB)efh>fRnn2yU9-V1IF7D~P3_1IT{@TmN{+Q$*uY8tPkQT7OCZMBM(
z^z&CTJ$~Gpt1todxiZj1>;L&v*5=|3kI+H;`0;K|EPcC{h>iVma~@w2)3FR+qt%_+
z#bU#NWtG{711f5gKnhe>y|eqZ2#1z@T!Y*0b?ql$H?+;kaYx5RyTmLaH>5j#f}*hg
zrm?*po~>&oIk1z&3SM-5sA+z<x`cB1v~AMhjZu-$^^P!a3(Rk^-U#<Q(u>rZ=Io~?
z_jJ9Gce89eJR6dHiP%;f-Gc)=@9!@BuiN2ufDQ%ZqB#4Xcj=`P(GDDO)PWPwphwNG
zs&u?G^h)x78!Y5tz@)-ylttAXwj)hl=Y}SXFG4cDPn*~tZQwNtSWw)WqiSOO-rUS~
zH{ayFuP_Wvo8qR!WXD(1`hF8ml9XWHQN|}2YB5q@$DK_Y+M3j*tgAPq63aSostTZ;
zaleV}j6=5(PGiWSWW+P5QgmF5Mn8~yk9<H~&G7BBGw}vPVinsiCLNRu@>lm=+Lmsu
z0)JfgM#Yr-uM3?Vm_~Lz`9a&wHJgjL2QYEEO_1b&3xAoj?9uqq9KS12xQT_W9mzu|
zM6TsHRb(urw^oXfEOm9wf4HzEEx<bCLb)C!YxCO$JI7EERut!U$$%c~=b9FMT%UA&
zU~DfIRzdU<${^-&-Hz_F(wFqf#ouZ5s>*Ft8wn|AyJSe&pG>Tvtk*g?+fgyxJ8t`K
zJd^o*sI>*T#}KFyjoH7yfu}~qdf#>dWW8Vb>UM)GNFyq$blMQ3@@_xBg=C&t1B`qC
zI7?cBP-9_q@?B(0I8o+um$eQ|5q6SGfR6py`?7Zg=bm5h@(WI1(`3x8T*^WmwMJlm
zmQW^xT-IsCy$Cqz{Pu<Mz?i$Xn>FSBV?7fh<{<4_p9lvRgZVz1%=8r+O52B~)g0|D
z>(`mZBjywMWOr@Y+fnRxJWfm8_+$xPHyo2ngXPqTNz_L<&YAhUy(!+mZd`<$xB_C0
z(pvZZ1xbE`=755?+P4xmuU&gnQaS_<7jWtEvzDnwyIU3)yr*)8f*n1%`(t7--cpb>
z9WnY#3WXeRA0H)cLQLMlmUY!isV%SCuMOhwcakIN6J8YZC;!}F3cSynn0O`G$e);_
z((Y~=8HGdz;@Wz!i(ioz*$dSMjf&X4sI@oE24|f*aSiI-jAp8lz)7qQL``U(TcX+q
z-C(ZIg<PN++7@c&LxO8D3`++!^O(*tO21S(!!h)geLcIi^6PrPSlYL$$-|!U<UTy1
zb<uCV)%3|(f|p|b=Xi3*6!|9l!BHDk1tGR_nsx&OM`@O+B<9pkcBXHi-%VOHLSMQw
z<pk=?IKiubY-IUI=<)_oI}uvTpGZ{rc}?!6?za|*bm*Ubcgjmw!JFlGJL$_4UI(on
zbK*ZA?A~GQ@+m5o1B=YrO;2BZeS(RG{C?&by8)b6dtj*c_(}~pb{KqzHvRFf2FP0o
z;RTc3Paqrb{vPd0<o}5bRQcaQ(J^1!BUE)tdyYD^En$GurBZ45;~Cyz%D#Ow`tE7i
zxH9gCpXpN-pdvp1tVBA5T=H#I2Xi(nit1r!f^^B?04xOZimH3xp-+}r)w@{nO-DT&
z%x?mqf?Su;cTv+cebpxEDe1zXt4iZwJj$C@NFQe3>psdcH@N6>(W{s$XO^I5b*d~U
zG`<Qxjo&&z$L~54-kHS{_rYPIkYhFE@l`IaTSrd6o?(I%KKh39e|y;cyfn{~9?id_
zNB#UKUafMX21P(;qqCFF{{(n-u+USBtYlT{_3-DWL2RWm1G>qLgeOa&77jbneEpde
zc`Hxl%o7%z^Z2{4N6)+lNJ|n3*Dk?d5esjJPPQ_M?A%gKhaQ2&do@mrSuzG(_@0{C
z{C(H|h^9q902qV_Y3JUH9BqWOH60Ivu;UTQ$aMj<V6%XVtcpBe?%YSRj?6L0<pZ5A
zc`iicz6UA+se}^3D}78%26`eTkZq+PCAQcUgO^2SCP`ybv88fjQEzqMv*cTXbFU{U
zBC!sD?_brX%|L_y{%)UjWRLYgdo;=HkMlbErr|Kfd_lmXSN-Mx@p7F&t2@wY`?)W2
zYob|0GhO^%iWCD?3p6ePlB*6T-;u_REFYc7uKEK-g=a;fb@9FSFa@21ywV3Z1m+_$
zfG6}mMKZ_?Yh;db=V*Tob7NYX!d=-x=wXb*1}Tg+`XTA`6LdxFnwN=pEFr7?J;15t
zzHUgT^JyCxAwfub>!-_}yYj`z`EKX>5H81?(Qxx=tEv^^_Ops9)ozD@eoKl8S9!i(
z$<WyRCVX%N9;gPa9NO@WQ&75Y9J+;4(XXw^Jhyj%@;Y2?J`4P;Opc@S+ISRzB>TOp
zt9yZ+GRiJcN??|?lALjo;u(q@y-Bkk^j5uHD61q7DI#`U+a?Ph8>-0!YLpB$zjN96
z<pod%=`EB`g1^~oED5$ua@NZ2(9~|&yLBI`+TM*URyq^4!EcuzELI;Om_ENL(0$bV
z&>7C8#>#4NY%+`AH+ad3(y|e=S^u!tp~yFGb6ve`FE<SLRr81Tv`L197Q#GNtldxW
z#NMdw8Rz?c6fope?eAa-)z_(cPrl6>5oV)geTHKlYA2b)$49{PW!}x{FskSJ?XJN@
zq%IrILQoTJ@oSGIq4I(zb@(LarRNu81YJi!{Aa>`qGN9UWBIA@iUR6;lVIAg!0xet
z_h=CAXif@1#T-s|J38dAW%M8iA7ao?20>Q~(^gI9C}$))ie%J*BTO5@nt6d$es7?9
z8_^e4n9x*{3fQF`t~%9q-evFUAKr(9aA`J7o}B}gs1CSkH%3T0M>X_j`j<f0A2R_K
z>Qo|mq`HeBb7p7cM#a4F-#fDs=;MEl2W2sQu=~yOU}s|sIBu~)SSx`Hi2>OQHI?OT
zPiZ_5)3Q;}g%gYLj%Ej4Sb@fU==e+MMZ@HhX{&%zXn2fsQK&nA43fuwKoogHTK2&e
zRjC!XxIQEXDZ%#`1l;%AdXHC`zhWuU26uc)A6@Ps+icu(NX|qK((L?kf3xmXl4UbU
zs?+bTlsDF!H+X$TOW0g7=nbwilEu@R@VH{CRjb&H+~ewU!qzZC$Q_lhwF$K_;fhf+
zB^@w`xpc?v*oWevUg%z7FsOKXW;E%&Zg2bdDCaVvK8`?W`j?zHJw-6xc;R2fn}99H
z-iNbn3)&4{02!(vjJ7G}3-wahCx_OIT92xSTVG5OdXgh#Xlm_{_E<Wt;4bFk;&)#!
zjtlBt>AjBM-5+92^3anzor~)}q3*J)r*en*?1OC`SI|f`opbO=bl)189&=h5<w;a6
zAKcmw%V`w)FO~eCn`<oNNpugz?@gR+KbDh8B|2@aozqqOw>0L!owtP3oY%1RX$n!4
zy101o=xs2=?r!oSW~JwNE6b2@525N;n8#RFZ?uY~ZNQm<ZmR|9jh3Lp<O!_h9k*%n
zUTMAz1i$kg@{Ii8%Bk6+4tsD)Az<Kd$1iq3kmcuFYSfG+Z{Sj3o=y3Pl0!J4XDF_X
z3vt^p&OGxJn_BGg6)Q+_k&&Tm@bXn@32SzP1D@OIN--RE6Of0y#Cr06Vda_K8`SSW
z!)SzkfD<ge+;yFCH0rKZ3=^lta6!FD_%t747g&+1LNA}9=%n2B{XNL>3si+qTc$s?
z%BsnegEnE!tOr(u92%oj>S=Si#%g|3HnW7oCAB(Er|VDOaH-*|S(5+36OS9yMa?I&
zKl09-kxL`3k2kBGQ-FP9sO%9_*>U=Jt)-)0mkj$`;&uEkjGH#Z&5J1##ULsVGF@qI
zPuS~f`rH~aoyF_GyEOjZI@<U1UOM^Ac>iN`pR4|04Eu?)hmBU<y6y1&i=Hml@}pam
z`>D5yN$LjM0Jskmj3jx7^mPOwgU(`q{BD=f!=QcyNzZ%33V$A~v3Onx_9vv3ixj+A
z)N|ZwtnGBtJq{&w0*ka{!(St#NyGPkgy=kzBWaRaeaTQb3AjEgZ+B_FGd$y_Bgl63
z0rS6KOReD|>y<P<Gehh3;x?6%j263#N&-0D$4O||9fkWONw667=H4yS6lk_&)2Zgi
znt&gtN~o8a+nPA1mtoB~R5}_aAq==+2Xngh<p=F)j*P~6aIG}r1nloiO)X9*_z>wG
zTyx1?sAN{DPdmyM4v`y*lz2^kZH(3|v`HhLwlr)uWzd~!BPV`P=$@n2t4}Hh-NRm7
z8U<mzS>2ohd3(fH==QD5wqLE{ugGFIh0h4*$|hj6bKSrzv~vgw*HXd7&Y2fy(l3)3
zC4a(~)37rhNow}x^o6oWSLnmu5BGG@#_fHih%D+Io*(B2FY7t(uOFA!P=znKtxs}N
z{rBGZ1RBc;c$!%1rSCuavxW0&l@WdBO4x)Pas2DKC4-tJcyFH9=MO$Pr=E<}j7E<w
zGk=brBMkjYNua+fqfD?{k)2z3xPCk>`n9cLL+tmYr!uEHjxtW@&}4QTnb1;r>;Lv?
zUJ8Vx{~?+p->(kx9?|pVJ0ax4;up01I@#!`)>-@rjX)-Qk0IVh9P5~MTdd-lHfRE<
z;&~jR@X>H2^KRwaI;#taF$<M9dWJL8)E{L$)iuoMJ~BD2>&5Z=4-bJ4-!oLa`SZ7*
zk9s67T#JZ$hids@a~!pT%D@%m<`y-ZgIs>o=3q8-aCJHB`766`w`H^g-4LIuv3uc)
z;xS-xtGU0m@}}ckfdSF23-5qNYb167^gf%w;9v>yZ<QCN(nU1acPkxC?>~3n`(6lE
zCifC6m_knMrwUmGXA<wHa=4#qMf<Nn25f&Bu4q^XONku9KYy00Gn%gZX5<P~Vf0Gk
z@dkxS67Gnw3wZPHR8P%dQPx0f@CE!Y+sa*XtG{+oeaRBfe|CYR@fj!<?%A&Xc5VE+
zCGcU2-f*SpH!jAY|7_wnh@S6%1o?NE(@Q{ei-7z-4w!S)_8}tY2Z{FGs8f1cHc+m%
zm+!W?X#KB0#|u6uef@mQso!lDKdbD2UDw<KaBX788H~jL>3je0{_H)6MY(Q{qZ=yl
z+vW0it)XZ*&7=DKcK6%#=)b#he}hZGMg9N6k0p5~dQNf+7Muo*f=I0hzH9NILF8Zw
zq7y0|HOn5!%7$s>15+%tU@yd$*&Oi<$Sv-Z3)bgn`F?@lT=mntcXPj4A1{gFzM7qE
zHvhAF6g8y!kzy9D>k-2f4`<?!$L?#?A|`ed_dAM%DNILZ`aPJktpH=73?57NcjY7l
z!*pn_F!@llKr#mIdLC}(LbosoF1im<?^v}fxt~(~LA&v+73m-Nv)gsD-EROgKLk_;
zOTm@ma0FNXST6hJ4Tz}1^|)^+u2cstwv-%U1L}u4ozq8R4$3OT;QC;-7*4AJJ=M84
zQl1p=W0J;Gy+JXsSk!ay_^LVg{apWJ=oc!x)9e;QO&^i|s)zq}VV%dOT+gOfi1q#x
z((I@6;_BgV0$e`)v-t4^VVLKWLRUm@?Na9g!^joz40t1dxu(B#*dU1HL+B^4_t?F8
zM$8LG?%+-q|14nn#n7s~J6iV_83s#i?NkcRzOy?ZLAF5LF&=(5i38ILcbHq1wvw2m
ziw&=N9=k3>_7e=#ShZ8nM<j9@Jwd`e^o3={8Tc9inH~j<M_nwtQ+mKGiD>4*poZ}X
zWbT;2L_VJnOH&${9C+rk3A`=e@|0f?-Zl;4q2qAMw9gN+!$q+dL*e7cCEv)Nph#+2
z918Qsw!Q5&3g7luTUFFaJg}5->xKvySU!(O_qqbD!zeJ5bSdZkuw(6KK84Ia7hkCz
z{NL>mFFOEpRJq95E%xVLvGnATgekZOba9Koc#f3wo(zGC5oh!)>0#;~(ohg82*K&)
z(xMJNh_MNtN4x0-Pvj~QkBbJjPf+cb!e$*IF$K2fWzb3WfNq3%lt#!<b{K~^R>;L}
ziYc^<%KsAK_<rKUC8jE%>Ewdo@Bt*9{h{{S`;78#3p6ySR(s&!jzGd9xhCS)oeSn6
zDyx|rJZdZU2H-s{Huc(_Az9b`6@YJXd(X5>Vn<z~AE)|C!*nW3?Z~vXOej*j=w2XV
z-@gU?hEeFB^neHLkHFRsV1g8TX$~oR5Y(mJd+z*hPUwDe^A2GWvoKV>(`9#Ed9OKr
z$O%SPVk;2$*l5<X9mq`=6f}Oj4m06OM#??0`*}u6Nr~!P&?!GR*SsSG+Ks1&u%J$R
z2LW4gA)ycH!z9=XUy4$53hW`Mf|w}%`&Q=coUk%0&O4j)3+cCt5PObX2vlev8h=gs
z@_987lM-@ar+5G*_(*!3RcxlPX<ypZlTBzZM_Zqf-BlfKRWM|-GI-w&?R;@#Rietn
z8!|SoE$7vSq2nP%FK4=TT_s5MYDgAddLn7~txbS5cm>i!nQ3ewqH>h!b-V#w937jq
zr!5nEY|o+otJqU~u3iz*tnub)U((Tc287q?!&r97OV?1qoH^kwC-wNxb@O*m<_YvH
zKm&C*)%ys49)>#D+;j~=@jJAtND8BO__O#(Gx@A<ACivxK{w0=_Uea<$$E4<;lnu*
zwEB;IjqNP10x3W~Q#?ipo&H9=3G&0Vffg>-6@7Q^t;mrr0Rz)P(V&Gptpr~uBVVL_
zN@kbgNHo*ojh=Ln39zEo^@Yx}Wv&;3Ql^Hfze>Ul)D3OSe3ql!%M(>)OkJCjDD8;N
zyxvpB-JG0?k{5=Q7Ye_0+F<_XjQY3J2z=W+&Q>$V=W2BW)5HuXkW^C1rrI3HqBd7q
z<~&{#JT5d(9RE7^Fzo@vhkT=qx+Ppv2&L0l2TOkZo~w)nB!S_N0iNO$@T!+T>`Z63
zHDFdnFw_?c;5_G{qOU&Bt11_Er2r=XR1q(mxj>%4_KL*F4(k9sGw_pB;SkxZK8?h*
z?YCU^INP`xMJTSZtS~8Mt{2QR$Y?J7>g<FJBQr{7#Si>}dCJAiYRpe>#6}Ay54wW~
zy8V6KZ%VG>`l597edHfJZXJS0fCjJt%#;zIAMfzAJ+;1%w{>^7M8~bh1GI{oGt`+E
z<J(rGSIk#Sy6Z++!@n^bleh1g!qA~(Kv9j(9VOttlqJN`NAlhsI>x3dabw+4p5!B1
z@jDcnC_714pt&fZ6}x3Z{Dpq-eC>yexep>{N)lLUwiScfZfGW5eQO`SaM53Guxd_I
zrbqGHkb0PT5&MwDD!;jF6z&Z8{}VL3f<eoJ;F-NB!hoO=*v^$_-D^+x#2P7goNeW4
zeUl#V!kSu7_S>D$3Tl~b(!64vKks#hJ(?-TW#H@?*|1eYF(`ha5lKy@t$l7{1p8Ad
za5MgJN_7)20|{8ZSTh|*ACMD#h7F<Pr&k(+^+gsxLR%{MIW1hA6qtTDc{~L^XNHIz
zOyh6GO&+ADRpJyngJ01I9P0+~emXYaT!ED0bA={JhE6vXSp7v1{y2EFRaS?A+2CfB
zmGF9v0Qaq&WfcUVXv|aqlae3HQ!Y2(j?hQeM0KSY%*{q0lUkX;mUZ)EI^ERs^*hhh
zf$r6Lm%b;0aTT<6+<qpjU^GSc@KZ8y)XH`o+lRPQ;=sk!&n98X6|}av%Tw?U1z;FC
zPD>qGp9RKz^tJ}zI~Y_+fZ&y4zK-xsM}ZdLV{^-Zp<9wY9PZhfWcY~%$lya@F+N;_
zh|nMb<nUtEv4Yc3%*Tjpm`}=?lmd=FjLi`L{4VRVz=}X}2Vwg)_@*%r`jfPH@UMP)
zsi0FEXI9i@Sr_jzR~ovYv8+mcWkn6Uu6Z96i`XfO2?CllAlJIbj|=cBNn@>CT24)@
z!h8<)z%j6_$!dM}5;+W|)a4O6PtoVDnx31ve&gKWSok3(!`>B;-<)aXEG{iemd~Ut
z5|$9CXcvdwlZ6kXf%JHZ8ckf_IV373U7INWI0^RUHmxQhZML_lRbEwU5heSaJ4T$_
z8_5TqmPg(ge~cY<wWEp5;HbFyjd)7|R@uG(SE~&DvK?(Dpd#00Gy^ATt%d<gQxvZT
zp2rxlKNZAjur+Bf$X&f+N=^6#Vc>1$dqQZME$sRj@53AO%`TbE!p*@oxPUEo?vKL$
zv_im?n7$uF;evir+0v1&GW0_q@m-6v==#i;r)Ln5Has9Tf}k#o2=^MTdT2q;wLT<B
z9-;wkvhwB#h&@r*@8S4}JL?-c+6(yYeUaVvn{sSAw*>+x=Z5K4k+>7wK@Awkcn2l<
za<znsv#voAC1#Okha9Sb(akERw@Kay53YiD)LlfrY8oGBRcVT~$=VR1MMd{4Y{~{Q
z_#PDdQ%6zKCEdDh@ES>UZitAdDtS8yyH`Qp{H1h)7D913+Vm3Acb~yru!!pVW8cF}
zcZP<v)>NLDI9jtvN_hoZk}n`!BWPWSlE3y^fMMq0RZLo&ryqcl+Vnyygq+NHzD703
zsJ68;M!NQaT=xg3P@0S$l0Y81t%w&I-n`hHn}-j8BWLrN`Q~EQ;H}$~+N-g>7bqaH
zr)n-tzHu*H=4Q%AH|lGV14n{T<v7Mgo-<Q5F1;k@NN-NY@Bv}j1CnFX))!j|B8?_%
z(W0lTn-h+;ET(KSZEUvRF;H51SJxcz%=1CBa~X?h;F%U?<dU7DFW&6~Y1eJIXk39~
zC5c~O!5+XFFifY=PoB+YJIG3-{On2|M<UOBm*GW>)xSSVf4=G{QGcu-^{a)dx09*Y
zexCxPqMEeOs+X)V!8QS6+EXpx+o{j|X`Tu_HJfXE)#j%e8Wd|5VJpqlLJ$mkjk6WK
zOm-V`suxj@B-n2~0k&dY&cP_T);Yq_&4Q57DwzRQp=u{F#km_Aq5huT5!*K5`p;JF
zBApImyaP$`&s^GjfSD&@HSzJ;B>_%5F@+@&_Hvw4_Mu9}0N%dN^8~7yn!5pe&*=lL
z0Oev!)Um`Wo@((^ok|;SyP=%k^ugGmS8_T;?X44MIE$yL<ibZJ*Gi5CUi$|=)A|yl
zGI(86T2qaE>50~r$jux04?RJIA+@SW{CVGujXpexlg26VDethJPg_DiHR=MLuKzyd
zb4attc64m_HiQOsQlnywz7CXbeMZ>LMqi&yqzN$Kl!PPWQC0KPexpnblNciUmH`z#
z<uJ<-w0nO)oe9+=$zIU8w>qFsWQ)l@6$BjhxE+p7M{plGW5}-h{<^sIYLP~cjPqh+
zdD{dd4JzgFX3plS{=+80Tt$?HBsjqu20vIbQ0)Jd7WEXWHWs@=hxfNZf{HgeO2!%#
z6{Kr7*i4(U?u40;Oy1+7alfOg)n@S1nrLJX&3rKv-fs#Hf{e&Vjn6`VtqXNL?1(-5
zX~N|S*E_Ul$GkCcae9<zp6b**lq9rqR&OT(r=+B2YI1|pR#&4-Kv)QEjii@7o#Lqw
zjg=EIN5ztA_)LKQYeA-%dSs40q-6YY(qBU96)pDJ@s=xJ3o<YkmE{CYJDqplGImo0
z`i04Q7jkc^Y{7U&mRBj#HTe2u&vh1*<vZ<NYW&X{a9hn1SKUw*-?ZnfoOsZ!OL*7o
z(B_|5(dsz9?yR1WL1%$)Ls21{LWzCDoJa<7+`>m2LEOoC&${r>P{#J{srISNkGEg8
z8!A?Cani!=vX=~qK@;G6l}IEoP(=Qrl!`;Gg7wYS8$v)TNHJ?E1Dx6WPJ%F!u#s!2
zYN*(pz|HHGBvHP4f>gF&+bk=HN6Bo`RsjDjInrx-bh!@1JEY4x>&VUIXh^|)uDp1^
zntf1(XQA*Zj??du53eNx$eAO1R{Rre%5Xw^b#%Qvev7t^AL>$?in{arKx5pYw0b_K
zZuJDi1I*dm-6LJLo(GFviEd|{Zf&A}d`;fg`HsDO6f5=(Q|U{#5_9@2Z;lY-z~o#w
zBy)TdkB_Oz$@%-=A1u+xUScw?;P&KxcaP8qE;{dLv}lX$uzmF@HPSVh<8UlZ)_KZm
z%Rb$K(Ir<buchXQA-kFb4V6VQZ9_uy%UuaK@*-~jD#UM3ayxoU=NH)nYbVUzWq7`~
zwl*?sd>`Bli3acYjjZ*_Up_Mrq=bZfo8H@SxiLuBQKn<hKc#C$`L%{V`0=cW5bwA+
z56R?3zDhI82`o=)rC$&4eC40>%wMgqgndv#mW%<Xz2PR_)4~AW4Q6{5loIyA(r82e
z_@Pp;ql%ibB9E`F>9&h@PPSPW&6}6yUjNBF`Ugy<99S=B_V)O!&p&oJsrrPiu8A}Y
z(0ofuuX<xF59KFS1jWK3@ldj}vNKK?$Uh6>#P(EoR7b$uvOz2U9)Nmfn=4~{%i0H5
zK<fys?{5k#5wYAJ`0l1q>{KHiM2Tt~zikScW53eEGvkHE!kn*r<}d}R^)KdeTIhV3
zv<6C;2UqaROu_B&<>4HINV{5EmP?4x*MJIVHkCMN;406M3k_?<8?g6tIdyTNaMH~T
zVG3FXV;TmPgi)L-!8AbRJ09baY1)u2#gzy23FTyJL#<xIqY=3cr+LDU!JL7BIj~t}
zEfKGX9~OW9Q*7!w%d*R3P0#N~zJKV7U>2`M4oy`(YB5n#BjD-WEg8GO?i544q~_W$
znUzfDRLYGi)CB5nH-_vMbl#YQB#ZF^_7+D6yD^n5R&v4&StC+(thm9cB;OeOA6V{s
z9(_yd_ETBF3!f(KSJhd|E~KCFD8_>clZL%o1wYF`38b-2D{S3r#8^JBIaYCAsNa>}
z^X=1%;rm&i?6q?OQ^*!vGACa{M$g;v+|ns02it-}h=0P{@AP}>qS@X9&sl4eU%><D
zx8nq7V?Ew<$o^UasqqzeFn>o0xb+l}O#=t#pDmb4GrHU%P;ag4CdTLK1N?=)n96dE
z!qc<TMibJ-d{+&v4k=I;R^10*-=wGl@9CnwF8Ajz73OKfG{Hj3Ue|vSdLv$%{v~L%
zK62Wo``!WM*NQ+(R!`;bZn+NK5I`&=AXd7~H=XwcsaszNmieuH&hy9|eg0f2rO@m(
zLLU*TF8VCEAdj#M2<RU=D^x)R@a=r&Lml@Zt5RU86f6|;$}19Pm4bxhj#i`M%I8TK
zraoA$SRD1jBsU?C9ReP!@#@rl@-V&}_l?1;<<MhrItEQdJiMX&Cb*JdLfMeKWkLOm
zwg6fkJZR1eSO*4#k=oYf(~l`*BX8Nk)U|!=MGuoxJGT>IH{sL*+n1Plp3XMq>ldOQ
zBn`D_+zE!Br}6haj~x2kl=M%aOjK`|=9U?wpx+rwAaTrzCqJRT=>N$Pv(9@3Q6ywu
z|Bm#lRI~=s?Akb`q;)o^HztuPs@VFs&p+g!q(E6qtTLl2!f$EiPO0eMI?Pw5@-1w9
zJ*Z~KPyh!9s@>>}VzTXrel-o$wfiMXxtLD5L6uO-&p+d^nv^v2Dr^*qjy-o{SesIw
zO(#{Wg6g66ltyaqJ*n!gFqcXv4vXT3KS(83wIcVK2i{b%O3m!T5Ym_D3tmEUHy+tl
z37CaDrxjGQ<QCt~s?k~!P_lSwI(BQg%T9xsL>l2hy^eQ79r*I(xGS4w4aVN`bR<hr
zqcX?MT^Zl4Ji%gj*39Zb?mNdqdb1Vo*N=LrSCZVQKB*`hr&W<jB?vn8WNPe$zPW!f
z&dj$|vFhEjfwqbvd1uCNkpY4Xyp!2CxPF}$k3`*x7&j>zCpOu^61c6&Q;M9Vcu>F6
zUMcNrMP?ql2e}#V*Y)ROdbee0i9=e*rl_5SgJ9xK0CnsU^!6b_#inz~7rxboa0x%P
zCRmk;kF@Tw6#!F{cmUqKlKrD{m1Gp?ZhznlUoR@9_3DPQlCa{g_IJR%qKb9Cm1s<+
zkUDMPEDw<pnxCilfI%{b-y@hiJe=!^_>;y`0XvrNi$wqYXP^~8`N|S93IoKmo1Vaz
z-)~QCJ%4PJZkft1!DssRmc=#G$w)xL24j2G>4?V-g4p`-1n+!Ipi2ei9ox6~>d@yv
z)ImrnSg*s^zD`_zhbDquS@I0TLFZ*@6XJxr9_+A@@Z0FEW_!Pbh1>goTDVscn9`pn
z_vhSUDdLgD1wvZd<4+WgH-XcVBUBSjddKQ(udfn78mWL+|AD#^_%uvboE3kUq*;ah
z8E1#%n>)vdi3)+;1km_gBo&zL8!%Aln00P5f%iNswgKt0G#P7}Td2i(N9fTmoDKH1
z??1Y(kuAjEeEzy3dkr|nbj90<#<wHEyy#T73Z92FnWHji-2ORqgr}AP&9M?}=5Nql
z_U=O&uGJ63O%rhWXUUW-DFKp>pEY$u=%9J+rq*3UQwhF}*w1jB7lUfbe=~f{SWAlC
zn2_7dsy(fQ>0~6-7$gHa9g3RKz^S$3h90l2X?dX*?;TQ7+HbGth5A2xM(d%16afT!
zF6!?KLS_%vIX9Bz-eL2o$dYormQ$xs4|j}H-3eEzc#4^DdGG!n&GwxHBB(^#=UUe6
zhUCUoN(OyLKM3TzgX>ueiE;ioJ1B?`Z5|-wV%u#EwLl9x=NmZHC67g5RZylyj-@O_
zTU%yRB_waS(XDrGSEV|JUmlo`Z5z_HbRns_WJ%#X<#pwCd39~BmM)!?2k^^hzG&vN
zENKvV4moRo`oZ+sPwO2u$uT~IR;L*)jPa|g{hPth%ad3+Y1a-Z9AtjgXxz8aj_^M(
zq)>fZPtu?f0(PYZ;Lt_TP8!c!15XL5+KuKvFM0q`o7br09CYum5pZHK^A3Z3L)flJ
zQC!v5h!6fX6L!@rPBmPp;^Aa}a+1+r*Uhyav*ci?+dIe4c4(b5Kxku(vj!LJm3$Rn
zTxme&s?f3LgE7a1)_MfbGjV*7<rE>Yg}9Yp5cPx;Qz!G%jJPX^ZC2Fk_9Dr^&>Qh)
zwr>cjpzPGj31wPFgLjJoW-7vgUN8_91pSW%(DAuZ$wnQYTP0aa;Z8rUs3>6WV#+>&
zPe(*Bi}3W3h<OB6YOGveZ=XJb0U|f*#VF|EHOyT#lQ1vY0H9Gd7_(G0d^udLBe?4I
z2&`^ncIT{(3H#+bF!tL5AU)PUn%C)ZF$e%#$Jx4t6pYg|aPe1KP`cbGj#z*BGp;wx
zvmkgjh7&lgXO?d-%efZm)_EYDqJk*x1A}ji@5;bM+~2(v5c@F*?GDW(okWb1wWdti
zYjG^gPw-Utz1b>mT@-ID7@E1dQ9`mP@yb#;$Vlk*+<>3af;bN;2sJsYdR8Y2`-N!j
zOpZZ@$?3Gg<rH;#l6xf6pP0`{M)AJUDwHDqFZKpLUIi&;Q{{uV?|c1w?&oUPIGxR?
zXv(bE*GfO$ky`u;$Pz)1bwc4)-I9EVyyfbFkqr9}325dymWQidiNb<8vL&x4{g6`Z
zDUFq`+9B~5h{ojE>7S34+0tFAaW<tePdb{v?1{ZESEtLaRd$v}%6HNi0g3w;EF^`@
z(ciXoY=LF}vL|s@0dw)quergkZi6QIRl{5d3Fa`dd)F26`>E7*LeZ(A=ifh*{#onX
zLOU5aAXJ@K!J>}+s@r4hO%g`vZ7cmiF@)wqsjPID@TM&HtUj&}f!s0HTAsd`D1)ss
zYgU-ajTGvEkZ}Kg9{l#$vxIUoB~YNC>uYOCi4)=rkcbf1c))0FA<u|D+0R#*M2%DS
zJ6)jyCo(tS1)eagSl|Lfvg&Pzmsg{;>}1bO2cfHyyg72_A7)L~E23C6XP_*6T3g~9
zCI|3Nl<d@EUg!g$XT|y}<kg(LOhq+;kVXe9edAWYm~h?iV$8HEx&=^=?<+z|`VV55
zWcOC3BHKv>?CueqYFW+#WLjT~{0840DZXD^9~i?CeJx3%P46S`8-3tZP!Qb%NPpMI
zcC&g4+WPZVV8#E(N$Ue0#|VbdY{A@?kr;EHo<O;=4EoNl<|C&&lW>TP&_2p$0}k?-
zP?}wX0d(k^f`mOEK_tA;^QaUSIHqlBk7A&7N!|}|XUh^QCZ5(5PVx@!0J#>WMl0-M
zTG4?K_V;rk9^X@KS{L&y1NUj^LxwzpTaTlY50|K3S;FJq!F)T~oH{%89RAgP;V+KE
z#xqoBQ+uuy_F4yckjg8v%xKW?5;n-b&>gG1rICB)&|M!u#(qPY;N2XtDU>JO@2LH~
zFOa#pg04^D&+`k1;JNEZ=%_WOEFY8}ofUrI+~M)<Sfm}p0!lG4nZ;!1>lzh4;K#+o
zKhu)uW~+U8s@9i%4|A7ama;EfEuZzZuz-#%Hz9+-j;ybm88RgC(VP0(d6^GC^@WNe
z8dC)ZU3jyKosv9#AOYQAI3;}y?Q<=|U>;75him#SfE`<L;m^j+lSDH$D>9=Q>_Pm<
z7^nkHHP!ZxG->Rb<in95>wOrGngVSB$KX!I#4ANOv+TOVu~=gIUteOMoujh#dBIKm
z#<{$IN|m_orcR4KOh}Wv!6=^TOQ^G_OTcGV<NA9wJ32F>_$V{i%Gu;CjrbX1%EVm>
zEoQY~`;;H7p2VuWO}2G}<>s?&V0U_V)zON6g8W-#rh{UJVy^JYnFQ;ZoI|QI>A*wy
z)>Ss$Lr(FF$!;sipcaMzya6LZ<MzX;g#7Ydxi#6}xSTBIchbL!Fs(luQ_!K?`v{U|
zGptw3c8aw7H(m3&<o1;tZ39U*NvYnd1=4*;JHMIpzOq*{f=Jg=*8F-MMnz^}A`jMq
zUER5NzfPUM;X3qQKyK(i^JesL=F6M#1l6GpgAi}_>$91c19NHZLY?Uzu*`9BpFu19
zbRSqT4Z-s44+#V2pQy#PTEs1|(lY+w1Q{TB<1(9s_4s8qtaCK=8h4)deEoLgM(f6L
zGh&jCK<(E9KF%+G<GB6bhmb7j2uG>e_<q>W86>3DB1FqZZRYqS@f(9GVb?>BPU<G}
zfap(GZKBue)Y%u@reO(?>!5$8U0CjhGeUaB+8q<Uo-!7JelAZ#zt(V(-)~m+g{I#W
zz1Yh_f~g!+R^5~$+ou<~TFD=v<moDIAf^uK1jCNIA|$>&|HHId0w@Y{=#m`2e#25?
zJv8K(e4Stxo5Iv@qfO^WjKt0(hM+v+a)T;K#dkf?&n-L{FSpeo!@J1Y3`EM9$hX&L
z>0a7QqO=`C*?^S5)dnog;YTM=alRWDeVL4HGsyiNm2u<PZso%yc9`t(sS|nmvee_(
z|A907`z;myx9$3088iR>um2Unxb+AcRy4h~$N&2B|K=^YY2k*Qkxi=o4Xxw9eQ69>
zfh;E;|HT#lO(FGfZ}I=@AA5`5E5Gh&6c#PZUCWXI3($eFy*Z!Nf?j*Lb@#5ze5B>8
zNYy8A$`SE~i32+SKIsI5zUXhZ0^~lw>_Ipv_RS7IZ+gZxFCepwg3!ehtVxO>6*Xc&
z4A?)Q{TYG8j1}1Vv1fP##~}udT*3#$*q5^>_VUs;5ac|@imq=NTs2A%4vpHU1*ldF
zZi*cCBl2FD;+4X9b_CyJ{S-1`PG*jA8vsD{IL>8I5t!I#vG1RqK8Z%$^TKZX)7KuD
zaeaIP8TXv#BycpB*u5Jvi$B!xVcHOD6i*1>n+Ui=_bp&->gNFTZz}Mg3+I*~EYlLM
zlwX{-_h&BvS3Cijz7JrQJGwsIbSLB{bub1l@$g)G_zDO&N?=aX2KgNqk8*$+iIm^q
zIrJj8sDn9kj9pdiaPEn=axDxt2MZ0~8mQpb#{DGCh($2iCZRu>c?07&;+#~}PX^4$
z_s<^`4TOXBQ-gVtoK6jxQ6j`7$>pG$C1zPlCg~P%+K!-QWgT6|*|}{FY@J%hyWfk^
zs9E^uTic4pg*9!$I^PHVA_iS7lxR4GiB}AL;{C6E<H`@WUVmoMcmRMIuk-qp%^*`<
zp9}4n&c{_c)yVR~(YlaMQp{G|-$C?`L=VpbRP34a8T)6!$Abp!YHYMRU{{G~ec(L+
zp~V<bUks4MA%yWU_(~pi7lKZFdT*}Lmd3)ShZJq5aqOBpvmQ&CuQ`q8aOjj0$G7qz
znDZ{M#JJYvYAQf(f!OA}Nqu&e`OYx}=afMJGs?mMf>$1NalA#ho&8U;OVKZ~3&G)e
zvpG0KJFyO+e-MD_-COTZui^#HEg4F-N8hr8+QTS!{=A=3<~NI{teqpNuX@S~qpP6A
zvV=M=M*kUw$VUiZ{(rc8%ebtzuWeKj6fn2}5v5B+Kte%UTDrRhMM~ORN<bP6x*L>k
zrI8LLq@<)9r3Gn_GuGbje?0Gg-0wLb&o|_UcdRwnoMT+$3ZvH&L7lT{6<I%)6+Uac
zT?R-}&h<ln8cZ_2Y;QbfW#2)xGH#TB_EhqpPbm6J2le+mVj&r!&}12zJHPLrGY}@0
zlRQRxV6gcFKHe~JjzUPhM;GXfsz7_N0!9gEX=)$PA~w1<4Knj;Gy18QDn_rjn#s~M
zfGtipfZ61q<F}!W?Hg;Frmi>@NwU2GOp9U&18Rnu`1{Fc3?V6@@i7-bYd%@N1vUN|
ziCX2;KLBEFm4|sVrdwKFo66nk4!}u<L0u5Fa6`)so8xZe+;R_xy?$aFBW8J1^BnFW
zqrgE<b9S$jss-8BSSP9Vd&9U#Zv&2ID31%hF7Rg7RAmVMb;|wy)~v1whoO2mURUuS
zyZ^mr`t^57N6h|TfM4LwdD$PtOkWT;eLGLytv(e?02GQrvR9I6p<h#<PK0+od=;@7
zvBL}ZS6KCEePfGM$QQLYK~pPT!{``PS3ngTsxSElu}OeLoB;LjBCkS>hXu5s`nK<k
zq|>~#VBD0fiq&^NLXSWQ5dLl&mW0#qKQSXv&kJats{T6Z`Y2EOL_~1kh=GtL%Y7B;
z(8dHYOKDKi8DO{a88&#Y#~%jiuB*c~`#`;a?KQ*O*Fkr9xp$G_ME%hpX6j=FRjZX*
z`!Y-vp(>;Q!Rx3&V!@!;TfxXdVuxA|o?gK|+)%onm{F3+nO8!!(}*Dbbom`o<~m{K
zqA8+V5>sjYr5oy6#ze3$Mj&)-aq{b1pXSChb3)0;i4wIn=FH*|QCPV(SkOC`!I~Jn
zoV~A_xquKWUZ%iaME$6^r5kvXY<Egt_klSPTmF#1O)J?WvjW72_~Hc)9z5IbcsmtN
zt@+X$i6rm#^`M?+o_KG+G<zrG%j@ZkI}sP}=ss<lfhQhMcE&n8cslK+_}2706@E2|
z1s$L<<ieI!AzA`$K1(i^t%mY3iNu{%+bUaH+^!pUo0WMQ%17P$+^(C{#%msLCLMco
z7B6P!_&WI))R=T=GIP`_yGu-0MBEB0x8fNzUs2jhcp{f6e|S)FuO4Fr?4o5YDnTdd
z91&40|9E^>#+KBX*gfb$1son6FK={y-tYSo68Gb1TCvqRO80DQg<0NGhh%;$5FRRK
zX0QAXl0V_Gh-_{Gk;T8pGMOqExUC1H_FcyN-GY6T;Kh2?4~Oj5!<W@SluY>vBv<k#
zX|UyJFD9NiL<R#GzPlAH`2@L$g>c}LkO}aXplI0s38^|4sJ+}b<0jrL-hAaKd42-0
zi-B5qcNdr(<Uxj{(?RI#NnkAzd9$VY`!DKDUOTD9EY!y5!5>3F+1w!KBN2#o$Laa}
zr%+8fFg{5`?0%^`ioN-8N0_O50eFut+Z!I3h4a=7s2|S|v7azqrQ`Q$6oxhpO8}?Z
zv7g|fA*z3NNoYyQNWr$pk#4baKGwpsZWWNsBFIXn&Ls{oM{w%~z8#3gYPVFL5M9+L
z$Q}ovY#Y!)EZYglw;*q!GN}d;MT=F%TqM<Mn2;QGY8VE^7OO@pP$8~+2OtbdxdUAz
z(UMyryPCWcb3hL2l2!@NY1or)VcYUOSvCb7M<zHj-_UyV=77WH2hcupK7pAZ_6p39
zNW5hUb^V!3k<c+I&C-5#oiCu$xu_}0pv6#)P<1#MO}fFK%a@bBC~Q8}W)iT-5*Uk{
z;9!+J0r_Km$IoDt%!0>$ha6bFot;pfkhHklP(UcpO<krrU=K^ruInd44dP5+D+d(K
zr@KzOefU~I&AvIeDDUI1_fDNYhi#JDqHYdkue8#H0$tAS%;6g*a+%{FLugOk3Ru6G
zs1f)Mw!tEO<1|E;>g_sM*(&$lx=`#50H<jQ5Zk3CUfel@+SXRVB~Du7(l+t$<hvSi
z03Ht%KPi%LnLzhgkE3C@hMr$4lQ#(xi<d@M(2JZo{O!2tga<D*>`L;5y$rudhWnUw
zLZNK5@&xsVK&C2zsm}*3@d8{PQomVOI9ENX#8b@eKO?*tr0KGMm$T1I-3W3n&tM4}
zJ{$_f&Y$XuGaNp+xJ4W1X<%7yU5(@c-mvU+mWXF0fb6|=ETNZ0jFN|yU2zZT2Cncu
zd35)g>d4Ua4v4DakYOQC5Hcwg0CU&xIidM>`C*RNznxZ7ZpQFI+Uh^pCS_6Us;S~Q
zZ#L3~2i|PAypam?WQxOdg+oNj#s%C+<WYyMC9)zV;3-UNu(=J3><f5nQewIhcuF18
zvBxysByf>>UjnnN(}sx~bZtPqc@jCPgu#I`sagPxSwn%GL|v+q3B)CAsV_?fRfHH6
zc>Ah5E)EqbvpfYj(;7va?Yi;;Oal0S@VNhc8VAAzsi)}`^KO3TZ)Guh%N;<cM@6Gt
zvun0us&A$oE5L}~2l0}ZIvH_ptyM0ugkj5J+*{EoDu*qtp)B;{^@%I#MN-QgP()Eh
zhem7igR9h0ogzyj6rpbsTt@g~7AIksl!YO9#ZKv%<}K;=@-X46S%nIzcrvs1@K8%|
z;aq%?X6J1?mw059R%DUarY4_ayd_uCReTq{3IVD8kgH1CG{8%GAK#{k=c^viW@g?e
zBN$;ywCfR)3_`FGQ6~=0>m&$Yg6EK%gR2#74dz8t%x$j3$ETUdDhBd~{!c28n?hty
zwcBRNneZPwCj#`gT%3{7z|;i7qpDeh!$5Z0poBI`^$&X3sh_8QUkZ+)Y|8L|ab8bg
zv{RpSIPpSW@aV_eyEQ<NkywpXS09sRs+dunHA?u1MITa?W*ke(X|6g;6Zf&1KIXl$
zm9FAz&Dan>LFP2a$G{gCXHjR;egc+vDRK_7%-Sq&d%%5V4|nh@<Id^KXQ6uW<4CtD
zWU&iUj1sSeRVk%qlkUu6e3m8ce{XxzpcwK|!^hsdaHqjrBb}RBGm8D_quZn4zW+IN
z*rDj*ZshTf`!R7v9n&yGPtp>58!uKeT(dj*XjWc7hM)M_-Gk+K3dJzs5Z4aUo@spy
z2fPGJ{_|1b?yNLq&o0@ldA5H{*r#zg{X#cbZKnDx)@y^qY2uKxg*16(Pc1W*)0o4^
ztIAn!JzQaTxRA?U_FzwB$80C_MJPrFG&N$Y8&@jWm~zaBxqPf=Qd-_^%$&iqCL?#6
z4GF1=6xUvAwNAUq68Z{ZMvAYNUaAPwo;8{%H=kde&ig2?5@Rw}-u_?~qx)mRrCWRE
zIul|`Y2m7^735_}rRVcVT?HNDU+8`%;Udg8N}Iq)sAf6Q<*czxy>j_^NyN4uk4rI$
zg~-egwhfQ8s@*j5C+Ub#o3t}R_HH66-WDlhAMP~%z@(!4rX5$?N1seX0gGyv=y63!
zi_gs|J>bp|t=F3L2kpoE-0zF0?7KoK5GeMm%VPG!WW?9xcvGi<frFQGr9wsH(iZej
zVid}3b_?$jHdgO*KZHH+^mQQhle7?>@Jl`_q8r4uxe6Xh`{6?0z|lo!YXJ_n06)@v
zvn3fS6^*ReRE~hVW9L~91H^4sF^MZ0wWS(^nnelTAx!v?s^fm`uCY^%vD<~6Snbjr
zjP<btYW(@-g^6Q5OdIR9A=|>hA_=TI7AM$p%<OsJoZYHCF`BSO<4|*!rd79IXZ+-B
zt8A47n2>lYu0}^@G&Arn7dFx)c<hC0x=4Q0#!^O(-$t%0cb+=FdN}%u-9Oz3|7|`a
zV#rdJZroNiqrID6(NH1l5g#Yrxl*20Y$@51R<dv*#Fh|dO%hkvKl7-YOY`|EkK9Pp
z{qxq`w|EH%u;19qt=={DtSsQ0WVxzb_<gbR0~5Pv>9|e$*PNR<2$K;fBop-t_q;QV
zTlx<Ai0&%7Uu3+=)no{~P!UNrpRPw^m1wE^GK4=>&NGgtoz)VN9O7m+v64WTBo$1d
z6<$e}plqusu;8znr1KVfe1^w9#OT#df2Sh1&as6-_CwBm?!$Dx;Is?hmprp_gNw6x
z3)0pR!33^OR&eMXokPdFt1^+%BHt@r)Ua!Ft(QbP*B21_jiJGqN8Z3Zl7HwI9ECfx
zY^6CbuU#_0114D_AMfYs5RY7BE^y7T$zu6>>-)>F263jRxqNoMH8Y1#HGF2!Qz;dv
z^vdTWv%-z7Tv#^FraUj9BueMj7p<6kE|4+yNAL<16`#O^#fX=1<EjsNKhJbSKS=>m
zpel77x~1>-_;|HT+;*Nw6n&Q$5O{qQ$8lebLaMC2pW=w=>sK8<Mbu=ApvWcPuU>xJ
zzT9!Ei-Y)~ueIiVHl8aH@gue>W+U7%iEF4}q}U7X425c~-woEm!g-vy0(o$cuy{TC
z-fr8JbHKYLM{yJvy?nv#Nw@Hs2u`d#p1hG!RCR014}psq+O-<Zp;d0a6VeViCFD<r
zQZ##V9bi2FmUr@(oygz!k0(5`Vbr3b7SD0|&s{lF5F-LR3fSF$c>yq!NQ2baeqP>K
z_}Nt@S*uDQi7bamP)J=(lw_hHVK?$l04B@4mFWPa)sIqpn>di&a&SGHN{MF?Qzm=!
z-5o49f`y|gJa@?(-6LR@`l<1wui=X#zhXK0lmQnU%rkftk0?q@2i}^URi%-(FRP)N
z4NYEx;u9N#IbfMCIV@NnBN}R}7_nz&aD0~%X;W5gZmvfo2!MT5;i!gI7Xp7C!8|I9
zdVN!|T=-)t*j&c=?S0A)JD@TQ76XrPXY8)gL3NyVl2>ZiX;&y-!DkxY29DDjLvhg6
zXKmNkN1$|;Gm^n;j|U0KXHm~Mv8M$7mF;@-5q>$Gu0coyR`HCX{3#s=PY#F8vm&f{
zcsOauc&78Y>>Ldjawp%JjdWk0aGW{+IUYAn8OzxA=ufjGlCy`s%RT0~IIpVNaSsD|
z7a8m&MpaFhjO5;sobdTQb}1<i5)ek+i@$nC4hpzAf83BM=093mx!Ygp+HWvPsaZV)
zFhs+KShh0$YCGN9=jVP_2=%2eL}&E&$fK}FjP`4WPv!Eh?UtmUyWPaHvjWSO9q8zz
z{WQ*9@aAb`*k^FK77!kO_Z5jTA-yd2j@eJzLGHW+^9*TFG(GIVreVSf=l}6olaIjg
zTh+INY8YkCr)YZBAxsQ#ODuz;wT@2lMkF0cgd@jby~3)_emX%oQnfV4etaL+2rN{J
z7^gsMM#C#*N-Gx7NoUn5<i`;dZ-GX7HcK|1y9Y_Ez0OLV*E#e*$v?Bp3x>g;L^pi5
zLE9r>`&o|5c=1wI*6e8Dh}_6sJ^zZ{KxzsX&ipy-L-1YUc!KHHzw23i^7nOyHUJ2{
z1h4sj6}};;OtE`CC=5X#4>E`sZp~YUTo~rdyqf@WAmiJSKU7C5B~$WD6wcrdE@$;5
zJ3}#}LTWk(I0#+(K>9&FxwYe3v#@`0H_N?ZV9uVUovV~v!L#hauBOjbxiGp?S={8_
zzP!9g6{#GqOg#2d<=4&#21smPtq=zXg%b4qWCjXAM;B}CZ|?`RE#^o?Y+jDS1V-fY
zv@S3-MfIwlobP-I+1RZ*wS5Rx$HJg9PMZ((60GXElU2no#mXK@WS?dguM^iwi#%L%
z$*S=#!tG*8iEnU`Nd}(R@eo4v#*6_UsuJZ|ovRB^&T(6kVjp&7m(VYYPG(F<`$V#E
zim_gB``-C`VAHe#^N<qtp9k^)e1)$dh-9^xU44A1f>uon-JUl~%|`p`OXK@XmcW!v
z^dEh7V<Xs0yuF!q4FASPZMHc01}nv@{a7?AWD7(nxGcKaN(17;DaW7}8F18KXcQMT
zJjW8tr0RW0gE<<&G7Q9-=iwiZr6+oQ{6hS6I}PgW1uojpV!Stnuxn;b`n`tc*eM<7
z-L3~udR!bPU+7|AN`OIj$bgeo@8x`uCU@r#WS>2J`~2&d1YiuUzu?z>G*I;vZ&g<F
z^{aB-3g(&p)*L_1Kz+cnKaMTZI7~2`J!;+m?m8^64@faz;RdSOmXI?aJXmaJJsU~(
z+H8aki@_;KjHLtKEUrv8{+rR4%<h{0y=?T%ba!DV%z$eYR{(1aLP$atuZ=;J$|TOB
z85ozdr45x9S7PaT2fvqKsEn`C<wC@Yw->*oSJP2^|Ks4@3MatL%<}?3-QZW^1QXM6
zzQd2dI78O1d4q%cE;sCiEu|(Jn-fE-Walk&G#>)N(CFy<KOFLH*qX-uS?7N}AVRLF
zZoz$-myw6G9?c+R3ysJC@y@a9IJMMAg2AZZOkgv_EO6J}W|tlJmHDVTv08Yki4CNO
zuYSa{X1U|Ny5jV8)D*h6vkh^&?5bAP#naUN#9TOy6hLUHqq0(&&V}hHk6lLPSCG=Z
zcxO>fvz|kw)Pu240deU8P{yi?2B&RJ8pe5bGix*<R5eWYc?h*6k_wK}J;*MznQEZQ
zMId{Wxmkitab#R^%e25GwWilI2@C4Z7X_``VU~ayzQO7l1a0VLDa;>$E%*phFy2;<
zqiBF$IRPy)vNuo+q#r=r$j_7Wa%`}5aBllAK7}XQ!^@B1)wtWi$}H;U&fMD$j^J`9
zQ0~w;<e$ACIVpLcsfkJ{uf;M}Ap6#2j7cJnD`I5EIp9U25p$NMoC^E!hoxQEfG4IZ
z;s-Tabjzmw#%pKj>8##t{s-%5W_>-s2MlgWT|u}+Ju!CD>x9fgc#T$k(ujY{m4bxb
z*zM8$H|JfMwnl2*k4>RZz}u;qU}XW%zN(D*F*x*k?Oo3vGLF6O*CwQN+ciW5Nq59C
zog<Q2QNlY*v2Aw8s?C03=wqY53C=?_3!>H(mSkd2_pbA)k=Q8Hg(v}cyza`@sOdF5
zVhb7r_u1cy52+N%F}%N8HmtssJ^x3+7<>BU@fjx-%`c<D*|=5Zu=Ha1rj6xPr)arG
z9=*%uC&nBCDT%15M)mHBwYvpN=1pQ+;|LXObz)-Givu@vgxzh**b#k;o?Yuz@J>00
zOf4#+n~=m(8Q)u5wZq;{m8iziEA5EIaTU0=eG=!GHxjhZ$Q?~@v>#qi%6_w6izul0
ze{Mw8V(PO{u@hhX0v5iHKxV+{VB9`pP}8x0p(oDzT)TZDUsH@odu7g>>!W!urWzjW
zz9FcI7-yQz9?0QidMJm6;Z7--y^bU2oq)?W2l3_?_@~vXr!knQoSRB6`Kjk^jAX!9
zY-(dK#hv3?3bdV1t?bShkz8`gR`B7JPX}Bxns4RmyPG>K^c=)>F=s{dfd7MlOFroQ
zl5HqYzW@vRA<H7W&)ETOm5(zIdiOQ3EmrV8%|I|m33TH=x0{3yzQStfAa3`usbwK4
z7R3|b_##$b;B9PmpQIgeE^58RN6c;P<J|e4*-mAVqMsi<P+@WJ-c*B}ec3x3`s)v}
zLWcsr0|8e=^P$h()$!`8ifC2~O!y=vASU(}$iLN#^=Zmwkqq({E3j{QV-SA^#A0-m
zha;zk;)k5x9~7j8+_OxgR|J+`>Vds)MKIgTYh`wvoh#BKHnwI%f~!w1w>ZQ08qH_y
zyLBaKP&}TRtfI5g0B(a}BCboqD5dR}Sx(F7$p>S0KtYlK781u3SNjCk)thm9)qSzu
z6zF<WY|(-Pu<&7AQ!RO-0M5*XE$D})a^?nRD+VyF7yyqw)au9>x5ZPuBLSh>j70@0
zlRCif2|GVMFd1fD?bDe#m($J}vYh1d4~?P>R7_~6UXowO`XVNsy4?o>dz0kQ;h$f+
z+#V9x(S?bT!Gt_RN=SvdLPi56EOk#c?+3!+K!TkKa_Bw-r$6edw9SVSD#`W2oK~Y!
zSdDRhXU>ac<PsmdFsSi|>u7jV;FXxI7C$7&k>sYPj&35=FmOeqJm&_k|9DwffQsqe
zxH=5hkzva)EMmvu2Z_tfdi7WetA7z4Kgg)`<V4b7<ieKI1I)26F!UqGbSAj^t=d~l
z8up&VinEadfE)-HXS@ec{d7u}pML*+NrF~MT@{^0(B3+g4|q1yRrWZC2X6rp?VNqE
zrnCqt4&l}D?}Y_USZ?O32Pzf{OHv1Vp1gT-D>x3U905;8rC2&NL?l-RM^$m!M@t`D
zEknC*$V@;7vw~7X0AI~rSkOV0UiXLGTkYE#(Ieo6mV|%fMI-qC6Ntsz))GhsSXep3
z)bi2p`qXWc&iHldWx8}_!Ra^Xw4+t4M={&>Bi&k=w?vGTaI$CK@U|WF7q}4(XvR|-
z<k(ti47WfePe%gv47JY^LVtoX<kRdcGM-*KyU04#>k~geUX&m0E2}HeMu3F|(VHW1
zTv)6!lWdoFGC53~bx}15=0gudC1JYS7e!UNtd60A7eXqQHqB69RV8(rJ#k<LJQNUn
zL?v!KyIQ{{SAv}n6Bb%IUmS)DD3p1zUxitNqKxifisJW8@FLO_=_MBb)f5q8*cERS
zx?e4+2IcibvhGkK{^c|&HZ^7`w&D*IE|~SgMRsNrtpbWh5f4IDlH;F;N?Dnl9VKFl
zp?enWWsstp?Z{x@if3;Q3Op_zJ}F6R>NK(`x<jcj)NWczL~Sl66P~~iMU=gC^0jMs
zU}^HdUMsLjnVovP0#!#9l0j5ICckP<q0Z@#adkG#B7&?ed{*~*z1~g4*RsCnMhq}r
zAud{L)nu=&$%!CXN;dNG;%4F7bU}hp`pp^M?`kgh5Ox$@CMTnW=qITom&|^uA1~VH
ztJE8AmI=6qA+m-k7fH$MWmUyleFE9GGNv*7OmuD#M5uA^$_^737f$Ll9n`DY8&_S=
ztk~jI&A!RBCV00D>e##qm1&_oR-8ASk=#;r3=5Jx({s<|XKwa$mob@%9!PG=)4K}t
zh_*;|M)D@zuh*<O!jF>EU(%V>_#~kHEf#eq;9&M<j;+*a250i7G)vMxc`N!@_U-3t
z%oRi$hN4{J4_3(=C~C^tCy!*)7a1H~OL;p?D~12aYQ#B`O}%!mVnJ3fKj7&M72ZGG
z*#DL!Jrg>G@82wHXBRdTgUL^j)I=??WUgta_QFRJ_v%wX{xON`pn0#(uY5B$7yS&A
z$BsJ8{=qU^Ihn$>{%%t9pmY*CpBDt(`w@sYQDwx7gwgqb6~+H6@db_SQDYlv=)bnI
z#ppUfu$R3F<*}_S?QVp&N9y)olrLua3?A8@?V<a@_heSR>?Fk|HKO{WYD<Y?wWpHi
z2)>kAPoQG$K|<X=A;x&wR&=EY*H^Zm3<GyCSMg(!eZ21RMRgs+cT#`BX8&(K&@&O!
zKvFK2oG8$eYIY?GyxYh=q8h9L1BQp3(5nm%RM_8B_v7c_#UOA&5@KR(5iU{rwx091
zG($@mZV8lvMC0Lj0XEn95(y2=(k)I2AF&!REtXy4k^f~()D-m3mOYF9KoH`^ka_nX
z--3zaWPr>XO!7OL<|Z?2dXT{h;)hxaotVkE?e78RUqqwi+lb){ZPS09C*hJpDX|$~
z7<~PPt-)Zv9^&&x`_$eS_g}_Q|LZkGJrnwG4~yS+pZ|mH<l#&QV;Bm<&qRM7TmOS|
z^gsUkzh6076`_S)73cYDHU0g-{O^7%B!KL5L51INfBV$@_xt@tYheUhOIm0u+5d-6
zOo0%7B!(Vi{p~CNFG$s&f6EX7%RyO6>GuuJ-@dM55Xd0<X&+VoCE-c_hQAAnuNCO*
z-$L2;NgbqZ>0I>}Z)sOP&JtI80L_UMM7{4}`_BNBX##zA*G?=ya+%er>#Q}^*m|*l
z_vMr2k*}%pp5vE1HIu~yJV0BZs(H7bVcZ^B+R3{V-tXb9iB$9HtI|w)51@l6M5sy|
zv5TKzoM3_&Dh&J>F154(np0OB)0QD<!Ed{4KN<mMVlLR!rm5$Y$BSKgap?xGXJvVj
z^gMo<QtK9&A-#pHq;Yxv?@e#xdPof3K9EAmW#@t6AXdl-7^M(+dl4wJ;C+KxGpFlo
zZnIkRp#Dg@ORg|yT>(Axd6&mPo5b5x*IW5K+zSTl_l|)l_3k-D##Fqi)QWI?CPYhL
zXMMTW@Rg&}O^)9!JfVK%BJe7cc>lSGRimEizNd2xokYx;3lWwEOxG+C9Rhg$MD~9P
zzpk<j1f){+JZ-G0Al-((0mT1D-KAY6?O*((w@r%w!9VgJflkpFC=DrKkH|0A|Gcd;
zNpRvcz={M-*VTA~o}b>VZ6la4tA-A*%$PHSrz4iL9V(vNpAY*Y2TVcn;>Eq1>~-NS
z<<VCO;}X5hP%%d&C*!SmsI3+beLpx+fWc%5m`d@{#*V>%Tn45v>u?`H0ES2sER0C~
z_LfHSi&{v8x%Uqyj>jt>e<Q#QY9{I7RV9_qX04wDzUR#u2KBP*)nWb`{Ov@mx36C6
zkA8tUHH=3veA~C_k1J_Ne)IHiucRwNBQSXL?bk$Ck^W@B{vSq501;in6eT!Bg1hRk
zS}qS28J9RijroU`0Yo<8c|h1pc?F)d<CP9;`hQ$Z*fM(_cE-nC+1UL)V+jLAVbzFM
zSzCbmqN&#TK)Tk*Dc=A5Ig9@Lp+1M@0rhYEj3gb$j7%H_V#XC00@6xdVTJfv7XeO%
z<sgkSOm``~Tyj$YpzLMX{BM+0gtj46m2_+37X;`C`{dCL;0)bE{BfmLW-krfG@ke3
zwt+6}LctGyu%a9_()vT2b6^J~HC+1$mkCG`GKQe(cZlaF(h-J{8WhplK44yl(tV+!
zB88J+ZPN(N<eK4jF};%AzphEhNGZTtDgwVd@y^kY&yCjduzxdI<Vu(9ouA>#m6i#3
z_;*|2zqZzPEZEQ5n@x;@|2Sw0X$XVMB{1HAQg+yFBP_{V5NR2<AkCPOXcyEe5=`$K
z`w=D~-+c7_woALMQi~vDDTG@w0Gbp3(>mD^;F@_#aPrwHkzG?B*>pJqKSXs~ia3t|
zJK{<{ZWrmbioI_Y;cO+g3Cj5hSpenHx*{j#$@|OINOlL&qFvPtOI8T<N?TRpdE7Rq
zz~T={MNyVuvJx|;5X;ee`aH=4#8gGpCGF-ecmX{C3w1k`$ADj*HB<&3kQF`l2+r&L
zeoMAn^}OXFj(fl_D}|<`UwDa0J&Yg*k^XE4xgXNzersU^vGfP)Xst98L3p+e#c%z7
zMcTs2^L%sa#;;zh9_ORnG9W#z1k-6mcysa?woNx_BA(Mmmh=@mv>)&z$x|cHTde>l
zhPbNV#+I$vIRf(7d75TuH6|`z((n|-S?f(Nc6kg_=O=KLM-ci-)6<HhBZTeZSpt4+
zJlM{56ALc+)*eaR)i4o$0?vuP>wnY$AXv{|Ozn7T|LIfM5W*J~pNu-2{&E2Z=>pjD
z_`4A&3n;Rg*G$}BW@dHJCgT0pj$Mco*}3TDhX1lgBlIYqP8l^!!{3$qK_~H!;kn4J
z98a8-%`Uu=*0MAGAp0C#Y~HdQk}eRsR@~5f+k~*YQJrWoh`Nw(On`B{$-*ksK2D@K
z+x==qR93-LL!p@`Gq-28(}P=p!IbI=45pfS{gJ~Q@artFn-9q~=&TlTXcK1odD~oc
z0${9x>r#DNkpARXlDrQdB!w{QF{IzA1!pTp3-0{^+!38YDJ!?d1K6U<U^3+w+>fhw
z!>^_I`U=GnLKP8_kdf_&j*?_-47S7V7Y2tk-b*()AK3u3H2VWw-lMQW(g%EXZ$O~&
z<@hU`D=Q>nxoor*iL<)9kTNs`8Pyx5!*K6xR%czm#b}4wMf6c7*qiG(X}>_E4ogL_
zvV5}r*t&Sj=pZy;_3YfYb3IqG55RcdSK#$KHm)p%gfhtVw7u-5%Z$!9JEwh(BxzWw
z*Rfc^j+~<3Z~Gfid>dop;{CX1`DFSN7fr_@vaS%x$o)~(10v<>Hfgh+3`g^h?2JvG
ziTCe#KW+xkbIBe~PcR-m@ei6$pP>1~`ZL`wf!Ub(&{u73?NqD62&&ew^=F0OzaIAg
zs3rgUk>epOW|GZ8!$*Jq#sdQrhRAktQnZ1Z5Yh@64Nk(rh%vU0O(}R<dm(Hndg1UU
z8aT7r0s=4I);yLp_>)a!WB+gCAB-R{2aF~AJUa>%^8#~7Gw?lEAWN$in*+)XH9jCG
zTKL!XEg)G504}oQe50EVW|)IkfPo`E*dQ=d$g=<@ZY-bk_f;IVeIDG11m;s&60D7p
zyTc2B!hJ#@ECNGDDM`AMz0j-ZDRP7_&mfq9CtLq#XFy=j<N|)n+4NL<4NUV`vpzjI
zFk36>ysGE|%hHoU&dQBs20WlHVyc-9%T!4a#cHBqhfM_Y2^9UwUU2w&@TV^l*Si`l
zH{0W5V5B_)AGxWQ!6-|J3@imUsgFHq-7eV{Wg8%M!Ws7TfzNQ{HSE!F)jyS*3Pj(z
zD2=$EYX<Qy6nCO8xPmIbK15O8C;PaO@+X^r;$}(88XBHgKVFic&ex!`qsLw@u6gWx
z2ZWQt$%7>p8hmr%DoDMfW%vqRD!rb?wgL!sEix<-9?h^>A@S11MURi!wcmh$Zl!QC
z&)fD!BWiObv)~b=_ih7$*l$ybMz@6a<UBAxlQdL$?PfpPv@ks_9zEFp(Y<bG+#yH$
zs8Dz)i8szdhTsWeY$W_K$Nm@&e@>sr@^QU~zDK|<JwL6LprZYPKwN&^rq}2IUdwNS
z8OH^Q38_nswh5akTs(56+KJ<8vesO^?D$(<1BW`Y=9Z!ib$_innm^Z^G;Lrs_#_@7
z3@orvIy2P|j&&prk<w(Jt6g@eRzsJDM=R__{guFD=xbCQo%|IDIY5mg{6NGi8v=ga
zP@Vj63Z{3zj*SvyBPeKMQ3dPx$n&59S%s@MxlKnM-a`A~-6);y3vU6STS=nMw^>S0
z2O4FODS3GF5&Ud2gYl-JEU=XW-)-pUG~YCsnd_i6;4|Bxv*QQxrKi8t7F}=HR}~DB
zikJP7KgO=Ahkca@)J9AdC-kiXTj_SnAiHXgi27=~njI^3_`uj7zDknjGO3vecK8J7
z1ehmHVz)mpE8$ekzDch^7>$>f@2)<w`>t|%LRnc9=aA#}Cn+QGgfj%uh?D*kU{ql#
zB{UiP=*~#TXr$Z(&?Fealc!`u*;Mi!e8ip_-SDdyk#f<skhf4;5}An_i4(IUIBX@7
z|C0-5O~_j|`!f?ww01V2HC-Wz0<)cW0#UwEb*LnYIq}{PTYW{4EOjEonb!inot220
zflH1s1a%<gaTy@p5|~oOG=<H42r2^%kL#t>On=37_SE6Q7{rdz?+*GB#B?nk-(Qoh
zOJm3{b3f-fje!w>eOGLh_@dX@Nl$`TxEPG9r$_y1o|Ae`-W5BErORe})iV`EL@O*F
zifx4Hps3`Ps$?r3&fqWevS+{3WdnU&<x;uuM4?M>et%$7H>0YuvT{+QUu~0{b**TN
z*wqKcyZFQz=4-h7CwKU~yM4>?FqTal4gyf0dv<PXUsy2R0SMBJMl{}wa=^m6T0%cI
z+!)Vm=rJ$`F{Z9-@`D8k{cZ6fi`=+-?9f_}GWoxxc6=!_+SsHSF>@e9HZeTZI6Yj?
z|0{g-@8cX#5@d#9{`kN}|JPr6;L;`sO35eqOIwvZA4jbXWMwj<$r+j7x~R@hsSXiT
zDu@d5qTB2kdu<p~vGp2p@^_X7yUZNLGV&GW$x}p(RVU{_I6T{306_ucS?#J{dF53#
zr{%#ZRk6nR!V|_sTgKGyn@VN4=f+^@^L2rUPjQfm>%qrhBk}C_ktKor%wZ&9CFiBK
zW?MCEV?W^&?HcFbC2w0E2lF+Pnldu3%#UqZ2g#om%iYk?GK_cMRphN2xVgo8F|TFf
z7Uixn$}?$~Jbyn5gKMo~*>$7D@7M!j3zygE2jXB$=Otp`MVme*xji`kJy-T!u_)Hl
zd>Bq>S>Z&e8OD{|vSPU(nEmW?R^$DziEjvrw1;<%<dpFfpxcO3vCT2i#p;!B_Dw-6
zH3zKLox*zz+LYUFiv){NA5P9^J<m(<4KIPjC-WhTxZ1_{3#$B=D=DurUVN<iqEO!`
zzp3BDj?uK(B=ZHOy!}<8!9B|siZ`BjiC1)D&pa4w>>KOZPo|M{=#{$0L$u^W@ft&r
zloTC%ZqDV}Xd^$t@j-5Zi)MKg|33v2SBjokC6WDCYBSMYsuzw7`4(Eu_?f2!zlxHj
z1S;C9#cE<=2x5z;nt^>PePZu=YL86HCbX`Di78W>qI=v**O)RcsLMC%4RbzCjpyCt
zL3eMqcR_>ZO{RpOYCvVDc)?J2fOn(tCC7x|r{UAbEjKA<@x%Mm>+tNqLB7iq67%7v
zOkp}`_#VF^<+@F6##Wm>!@S0TnUCT_3}M7`)D~`K7$(1n#L@4#MN#kEW_ymEecT<J
zo|-}B8fS)8&|D^OY2bT|+IwlylqkdLF2eehBNC(An#|va2zP}$e4Wx)-2;rP&n{AY
z<xtI4DZ9&>GfB9_IPjo@yg*N%IzERm;q!W}6)@lRpVVIy%U%4r&?oT8Z1;=j(np4j
zIy?~`G{NJ=adVp1OjEwnlk72CJ8TM*4#3jkLDRj=>OtJ_It|HR!4uV>y0}DfY4vK?
zKV(`buOiXqt1q#1{tQPI!pOpTfRER#7OO_)!MpBt2a{F`hvqW2^0e4fAWo-O&<Zo8
zPSL;kxdNbAKFF17g1cPCVJejC)`do`x{{FeHr)^o@%POswnH+H9V~jf?67Cv4utp_
zE1SG6rOry^-*dt=F1fyb0DOl}dT*^KYA$q|)e9ErL4t%FJ%`q~lES1$|E)>S!3>3@
zzLk%+D6%L+-%f%Z_bBv|hO1_%wEB=+R(X<IdJ)YVxCj%FZwDw8(b)Kbt|XQH0{dTI
z&nw~@rut5l@TeGz7C*bT{u$LwPatcLM|Q*!?i8{51j6c{6YbWgTv&C3nuDow+(=-c
zutxoiuIh|350+adrTr8#E%R$iJ}@7B7q;<s)iANn8qO&xwMxV)cpP8jnTuw$wCh-7
zL-p*He@Oi6Bge2vxxTnx_pg9*MPW=1{x_LnQJG)!-l#)iv(gK6dZnUx4zMRyBS-vG
zK?%>7O6^G4&Sl;w3DO{Hgb9%O5;yDt6t1b%H$?YruK^lEKs4b0JYUR5$PJ%ggS<lw
zOGCX2fDUmityC86)M*i9!2AF@XyRgfNZHtcD&k=05})D%qmsrAgb`h2Bwk>gM~m0~
z#`l4xsIq|5hT2CusQ)8PHC43WB)_S)HCX#o4e6<s*dsDk?x(wKUP8>1<AJ)H$`ayP
zK_a`GotvU|<2WlLk)Y!{$Mtr;Y%7b~4xGzRruoJAz{QG|WL#|(z7L07F#k-*;>qi{
zS<hrRW=pHfGz<FJ<yFJ!!O}7UVNi>oPa~=yepc}V+6t3=p@jWe*l|(l9zvq+0Nh@&
zX1i4Cc&z(rEZ%RQ?%DNpzfBdHVO*+((n}1>D0LNmyHjmE8EgGQW}QMGPrj)wIyxMv
zmtTf`ww<Izr=I(YmU5~z9y)(1N@9E%&SO9CkBbr)kHRu8EH&o_N=|j@+Qo~RJ}+}!
zh9E3Vta)YwW_2=P-fyJC5026!wRN<y<$gW^eS-qbqcVyAkajOLMlA&B`<1h$-k5(r
zh025<X(*oOaO7}fplA1l-9dDC)Ak;uw#>v5#+~Vj@VyI`r3{$(Bj8X-6{YtihRkhu
zDmBGv4(vZs=p};ltxTDeBH^GHXovRb(B%brIJgS^aRBXy{&@g((K6)R@Sw!hiRMB@
z-nJg+FFionIZY2B<}anP9JiMEo;y_`=1KmH+Qgl;Kld%boH*bd<sIGp^6BU0G+0hf
zEw7W`zPxhpbjm7_SD6-N85W&QFmeEA*MO4mI!Pzr{a0U}FW~PAd~YHw9T)DvVV{PY
z*IL8~JCJ(Y1)OTc8v{n-0fVHR_pB6Z1GCKMPjp4GUQ~V|!aiKBJ$6%y;w0QPWJ#C0
z90hslL>JTQ6uezD9ku^2FDp6Ci8F%#KK<Khg0CK$=ytvJ<q{*uZq1lZFJC--51iZ<
z-!au5f6=)9FP~U-+UkL*vC@atjl}HAkL}0vu4yIQ&?$Y47>$W)>JweyAv#LSuOUmZ
zeD}v*_2I9*N~FO##RmUE34Pe}sCX`G`~i+t1-;2fHCcRyi|NL$66^)gtC41i)C`$q
zM)x;l(!`q&hnw*C-l5f&#|pbNybW>y2gU|8I2zj^37Vt8%%87;8qV~)IUg&hSOG3!
z!t<p^)gUs}H#+F+XSes)N3YA|vO1rncGf8|pHt7*y;41!1o7&_S)UCa`RjZ4t4Y+#
zTBI1*&L)piJ1Yr@Y|@;w%euZqG4JzL?M!WxL?8(#BfA*uP9a<Aex;v}OY5z_ilu+W
zdjIzinS7XURyllb{3}XZQidwWkwM#9-4=7TYtyTS_jIhkbmOA>^~LV>=m3#A7}i8*
zJ_PK9UOxt?7U&fqWH}(#=W(D^$3b+96{b;yUG3B_+qESqBBjV%hkf5TZ_F;hi5pp`
zFt#4gD_#lEG|`GPB-$kSdOaaADmCR<Ga7I}o_(&AOWb0FV)m$asR=03bUCLM+Lc?L
za~!#)pSw$~>YkmHC%-xd63L~zfn?wM=2$dy?*>wT90ttvrr=piAGTGfpyl2Z+Y5=t
zYvj}^knhI(Ry-kD9rdu_NI747Ft+xn^J^`+eKc*ow>{eDG0DIvW>md}?T1BKO+`Nq
z`No71duUB5SaW6-uo)1lWkz6*73fa0hA-_D=DKL=dr<Bat~2~!uGs%$BQSz|;6}Vv
zFfuuRgaB+-lXV^`{Mgre)0wBS*Y;p&WebYC905Y7`q>B%LK1K3p3=Gsa{#aHY=&Jx
z9Wx3PSkJN%_WfNL=vP3$&wIW=$A%o#wH<6J0lqRE4oqG6ppBuh2P=9Tfb<=Zjx$x|
zGTG}?X6lz9X4=Hvc<uUwBs}9nkP1blK3{4Q=Z*EveS%pY!v;)JO6*Y2Jt=uT6t&ab
zWG+@pFNo#V7Mun{TWH#-+Ii*~=H~z{{|Z6E54&DIf1{W*sNf&hedqkAMJkazMq{E~
zq+!knwO&57($qbK!+Rj^EJu`|h|RH@!0pG?=~!++WGLa8u!M(U73j*3QNI1^4SeSx
zwnwqDqKFy=tRPjLo&$y$T_SSce3`!UXv9h2Q_$pSqbp1@QgD_gsj{gPEXTzL`BX2*
zkJwyn9u24=j-IIB_c~vHL~^>{yB(F{nYO@^HI@1)2G9P{%qy<a%9;69m_}GGO#zx_
zxtiYN^und^lBkUXkUcA)*B2bZJig50sX0P@`(<Te>QSa><0Tz|pK>^7q8AW(_U-$h
z?v8|Brc*WiqP5MIE67^?BUE4$Xv7TBc$p|;s=InHwX~Fd-8E0KEl6{jsbgfn&`<nB
z^uoXD+rR%oF$yQwr{x=-{J#rfba)+LtTc@B&lo1GQ+$)c;fNGfC7W8v@=AETJfx(W
zJh^a{Jd-pe^HfA;xwxuoxN66Snk)@KAnB&NB1%!WdKBYQJ2g{jD30$CS$+iPYI;U_
zg5nhjP+>4_2PkEB=zb6pQOwX5bepZf8Dk<*)=*`mV!Rj98g{K6grJOzhgZR&9e^E&
z<Y^KmTmQ`DmPJLS=Xwo$Aa$}_QK3#@)-0^6YCcvq^vo<*R85{<m=Y4_czgWwK*zQI
z?8<P-J6<m*rzHmB_Iq*f@1tc2(N19?CwJEB-_J6Vbpn$f(ajC8cM65#PN^@Uy=B0n
zj&~`S7D_bSxv5y+pBK-ESTwT<stVCD`VyR1TN*P@LX>Kmq*@HDYdGUsP!}mdVBSR@
z;J`ex%9d1CZDL%UoYyG8U{~@;F#W2_v+<Euy!qM9$6poWCYr|+KxZS1*flBjyf)BK
zJCnHsIaMKmG(N|j#T-@bGycH%mVoF{D->lKYMejOFMWK8N_h&bY~0+?<o90oSfhuF
zx3Ve|!_xITgl+oR>PrW*E_Yg~P)FsNl;=0xQ~Dks4^|y1jO;DZBaE8iu-NoeJ4|MR
z_ISbTO36=A=Y#4!W*F#1m6vFc&xnco>QqauVX<6_8_`@VY`HJc=v)OBCH}Z8xkuoE
zP)P>CaH;vBkfX(>RHSb@o@%vx1uBQwj(DEW`?f##7G%H%KKFu+;@}+}Yc=L;i`b@X
z;}HMrrE`{GK<$S+cygqs3}bELEV00yD@_jKRzLH>)5YtGX*<xy=1n}v=S!a-9lqx{
zW46QggK(+l`Q=wWxLj!<P8&G$-azu+nHKm2exm?Gz=aI*_cgmj^u{l?K|uD+ofX5B
z`Z=6`Dt~P;F7Eb!KWVi2s~I_w`iGN~x2mlw_@Xf?@PGHt%uyYkQ`FbC#k3;1^Ryi2
zg(s1ij|@ojLy!W_`o6Lb5-SeoNCAkSRW_m?#yO-)_53Iz8jR~rhjJD!B#xpg_xOq<
zLW;fl-rFVkOnqb}w)2;t*D$q7@|F86+q#&Z*nFh+i;ia3q_-xYdvN6uaiE3v<eBf6
zF%^ZU%!=j4c`)rsoYetmbYX43Lyf6o@?KGY6i3VwH6DIT_yK0UJX;oNEI&Yn)YQ{3
zVlkf{FcmC-z*r1J&BA<TOgB`mX8^C7wi+Dug%$bv;Wr_hqlc}&UP9GL_cMmzdrD4f
zy4<%j+9#H9#6lqjg8*$1cB6Et!P|1N+r-l-mlJXUi;9{xvBG?XZ}(iobVkX}@HRb>
zMEoSSjK5De(?1l~A9X&HLgsC(mIjU!njEu^M(qAy3-#|?`uUSUe_TPG?#=ccp5%8t
zgxA(Z(@^Dx7w)s%Tk%T1VBzCSbK8SU%u3Ub+Ro=IpOontr5UK%85f_LTAt_v_Kgnt
zJ#PmIJ!82C@k3XtPq6Po&17j=Ux}of>r6Z!BXqd2Lg8DD<$}xYC`C?UDX_798GDS+
zVf7J>e8D;I2uI@A0WO+*HF#oQ27XXna=p_f_3T&a`}b{VG7BMepDaNaM($$#xhqPa
zNIi!gGD)GqY{o53=z5W9Mo)$*P{;{ehsqqAo_o0-r%lG@ttVR>p)irfS#Jga0pIaX
ziC~=(#J({(oIa&2+WAcXs9wSU6lY(W&=0Jl7g1ER&z3HYV!iexFF2<g&k_E{LHh@x
z5l7b+lO~%2`P{=Rk2K_7&{Exv6Lfcr*d(y@PELpEPW}8aJ6UZa$YDJXO;78R3b4!n
z(9T*Ie~L4XXUp8AQjckf+7ECdl|{Wkk_4lPcv-|J{czXgp`aYqo#uFu%@Y7m7J5MS
zGe&LhO3vzrW)Sp`_h<YZ@Z_IH%b+sQU%Tt?gbhFAyxYj$4zq|)V;43RM7`n|P@9<Y
z4{eG?pX;nm%HnABW)lolCzF~NZA6o5vU*(zBT-jh_uFBQrpz;msVqQdkn{vi?;D2a
z^Qvs}evB@s?qGFMBW&=44m9EPoetY|@3Yk7XSm-&#T^DS_U@Kb4I<0ev{Z%#&e@+3
zk?~ik_I5qBc2UzAC-8;M)hz<DLn>_qmTTK_rTSQwzEe2)&Pn#y4f^}TO9%s<a01H>
z?dm;r=U1tjc^V^;8$({r=CNn@Nx}RKd?!^=dre~;Oe%h_u#_k&P0vPkIiIgviS1pE
z7$`w<P^Z(^h~W4qtp}<(c;1S;_OKF`-5eBhc}1w|{xPlrO8RG<C*(b_O;&l%`O5T7
z+*&n^5G);SWOP6G<<xHrT#r~<XvIDi&4>M3hsm8Nq4LkbTa6|q8ttDn#Uy6%mU)nU
z4g(t<z!r*<qN*3hr<Rg<MWS|vJaZw>`xa@-LogaHOl-Ki0+meewtf&PS0~wt?^OOZ
z<4-r`gf)jN33$tfPaa}ZoZuRS5o=q#p39G~)W}eL6j<DXo{7f>#1(!=lrN+OrYY33
zXg@p&JMp`Xv+wjt+(UCO-Xy7CSH48{grODUGpoo1qgXn)zb*_x*U(WX8l*(nwn3`D
z|7sanxn$rM@{<S8Jg5%j<o8qFR9~W^TLiLV^=I<~Z;J^8&fijq=MGtor>2QOW2DT(
z`8mZ!vc|7_&=N#xJ;MtK#253Nyu&3y-&4pHZu|dqh5z;7QxTr@F95^IlgAX?yVUXf
zfX!3GG%2r+JT&09o5P>4ko=PXwuqmBH(dVOB7Utv#VJhG!#yZ*H(gHJQi=#UeC_r~
ziE?G+|KDKafB)WNU*;JE<?)p+{}XZ({I|DA2n*S^JkJElUHk7g>fdki+uvu_Vlu4f
z)ZTCZCkWwh-~9xRDjO^fM#uG`$G=_m|J!Hy|M%DT*cY9IRGy0q@sHqH*~(B^{(XHS
z!~)gHT*HUBf1g#!b~uMXuZbjLQjZR%8bXx!x<Eh4yQjW*-=6@;t+XhHrvP-QD?lF1
z1K>PF!e~?q4s_+gW=~FuNf?VoB3@8QXA!TNTxjL`d!MQ0sJDXf)?n|Og1h#Ot!rxg
z(7Lq(8J6<c4!j=Hpc($@)O&KCWA~o^j5Q*`NjjXpCe_@Y7<8M}pk|bkls_3*Oi!cQ
z<o7?QP|yDWXJrOpibZV==oB!Me~7KMo0G0(S*CWZ&9)%_PN{n4!kwSH_v_tAWSz!)
zpkfvWFzL({aP}>1GZ&oKR$1}iJMp{q!w>^I`qkGnuYNsn>oc^^C<|-%=1Fm77F)Y(
zfhv5{asA#NfIoQ>s99GaJ#91*DumOuJu`g(81_w&Uc|s~@hQw(G4P3gLZSsV5Zyu`
zQ$Pkl*Cm|#{m-w!8c?03X4i<&c$&Nx7%Oi82{c}#njeBO^(@fkLc#UBc_7G%kf=SG
zN(_ynh>Q*7nZN7={IR2oS3i+nE1+Mvu-JLa+=J>n5Kbb2CEYRp3RJw3Vf1nYYfL*n
zK&aB`<eoW4O4@h+3B+!C#)D96&P}ys;LLYoF@>kcoA*1bh-U=&r~PRkQ6P@6#`w7~
z%Kbe44G0hu-eV|q(3M+v!4)^Uwh{f_7wo9oAZ+JIecn}KP|JjTB#PQQzg5*>AU5_v
zi^=a$K_NMO*f`Q9H+lZr#PF~ogQUxw1v*<Lfv@MQPG}T@1TfB&kOPEaL|UMd!10&1
zMSGG`HPIkhZakv(-_EgEq8?3=c;5%@MFgU4UthZH0sVGYlhv-b5Ft$$e7%h2JY66Y
zJYF5ua^6|J0o_JW9wkN{;>S~aTH|`TM?KEDhDPOidPdOHe8QZ5Z8cV@HOB2X$#GpS
zmP3k4dt(@*=zM%7Tara@zy?rlzKe&LE`e>BF?jagge34c>4OzdvD*-zslW&D8gdC1
z25FoOYx*xM5LNOXfcuti2up>_6JQQKNW)`A)U(HdNaFBYKp<x!*HeQHy{PI00-~db
za4qp;iPvP~yT7<ga+3A=m86pt7E1qafo2jS+bCgU-Q8c?w2(2X4sQo>?yB2+mGu+9
z94xWsApMz9YtjMY)jq;0D97Ol$8pR@R!N*xf^uH=emx8%^$lk|TXfRz`x=B69K!P%
zDna2S=C;>v(Bk>b=6sigXB}3mqW^dBqB=!5iX@<J*JR&RI-}c8FB`3>rc^bKNQ8S9
zhf8wEpp1F@_?xo*ryMg6gW@72rg#>nOZf~Q6M(wVBK>gDhs$hpw2Fb%mp$J@34)Dz
zfW*?vTxeh2Y9ie2-#(Zs$+#`JS@Rkyj%c4#ABHvJu3HP5zgu~F`?m!H7t&^x;TyaE
zRpgN1{8$;$xWuY2>tVooGyhZeChp+{o>`Re<xbEb)!eSCH(8)kOn|n*3uvbfUx0{!
zMeTuaSC;KvSdL_pR}3K8fp1|x(F%g<sKs&9F)b`NgQwqpctV^~`zmSfWsCKE0=uiL
z;2uSU<<=vmHSNT+ZRBq<mG7t6XJc&%UYY@UdlQ!ExucK$0XN5V1J+D+2bW|7r}tZa
zJLk>hfN!HV>|a#<=c6JhjRASs>>8W}n*e2bOyYfmUAC0X1wLCIhiP}6hy;}3O-v;F
zhv2Tk{GQIOTVj`MA)e#WP<Mkm%`9EuYn&-|{1!apsnya4tppxl=~+?&!5S}PQn~Pg
zu6J-UY&32a2$$8{(D5`ugDoTk^wzGDCm*4LrwDQ8P^&~qA4U1<aX-Ltu+-w-bQ&~F
z2UDg!r$^6}QL~AW3S7cwB7NUgRXdvWwP5feKjq44;^JyxtGtB>&HHM(H|oECh)!j0
zxd*&M0`I-{P-*;K56a=d$6r(Yv?a&+R?8bZy^f_~N$YPq1-91{9QGtJuSpbsnhe_W
zW3NBJ{=n-$#76Pez36-0yUmW3dwhyfgBOaiR)k2`{hWO4UtXD)<iNsmLtv9?%IG+3
z+ZQ{uueez?xe}cKFl<90FCzi&J+%n7^lJ%^LwR<Oz`Lq=YXJy(uP4jlbW$7Bw-Lyx
zW7QVcKS?N}dSAD&^WH0#ZGDL(a{dCko>ygem~~1!;;HKk0U4-Y9b%8ZXLU&?8V^bA
zJ@U9Z&$jU%O}gV^)KTtSN3r9Q)#rL~Z5s%KtRpgP1^B!qhOd#hx+SiL_-S%65sHCL
zQ%2)FVfW3|=AI{YH#QqaSKTX5oWRSH5))QE%Nh>1S08ai`Z_1Opfw(3x7wUm`0}9B
zV+}J}rU!HWvOTL|ZdlEIC*!v@uZQ^7^Y7hpxxRbU(*ELr`C7Z_k=6lzb<#gzZgWUi
zu&4R<uVF&wZ44w&{Sah@@Ajz~zJmP8C-x8&wZ^cF=S_XqBX*}Mz`F~>At{p-Zxli$
zw$x&7TgCz;p{Sm#X(&R8Jqyf{YdlllgbyG(gi~0gkoFk3CbQ6x->S!GcqlSt1KyEm
zUhICT-TP!M+cafzc<cq$upHtZc-A3+a)7mBmH@b4mWM`3Zu?ntQxaARm+`L=G~%s5
zonGX#BF1xg!VNMeV>G|R*3b668wN8bG#C2Z!BVdk@bCE-AqZ9XBW&4bE8GB{=8+|y
za>xQ;(_sDnHkh|jA!FPk2Kr0iNe!k?w%}6>Qad7KVykJEACxmEGnl+5#TZn6?0~1R
z)3ORAqSF%UH1sTkp?PtU$NIf6x!d^O>?=L5{T=~so8=>xhJ6T;8P_fwyn%vbsHH%o
zKp$&GE|wGvOXefPnRev~yn>3$RY+IdLLiThR){V@H5>9nUb7_6V&-cCA2X+g(!mSw
z4G^g74~9rf^N<jCZzbB|3;5k(r3ie#Fmpu7?^x&%cl-_Kuy9;x*$%F)mVl-O8`sn>
zCj`fuft&xLlTX|CK*IzN?6aE`BK6q>(T~96i+-aDxWP9bu{P|3D`E{DyRs$jAr)#L
zmi{9n@eV(TSi1WJ4uot}lRm8RPauPtg=+6&Ikl01*QpukRg}6}b8Ux+JsZ|cQ(%)E
z+^z?4VVhVYs&}f|N3jIT6;IBGK?2*7xv$9%7{ok>&J<bS>+--RUE*vZz*vB7bFhK?
zwcd~*Dx2&=c)jYPwn|_%_s|70+Q%D^8g~sKWlj@)|7S9CT9|Hg!!Klv11jvn+k4;B
zy@QY2+vqqb9qi47^_{<6J*e?sBZJI~((aZXyZX)CH3qSE1VVS^5*FBXJB7amm=p;U
z@3}iUy3Af^=MP#3Q)3?#es_XWd2Aq^%%Q9n?XQz1JmG}y6C@{iOC<U9Nnk-G(9=o;
za@3!4OZHno-9Y1i0_KFc_Ej4s7z;^oN;YAlPy<x5^Dv9C!0<*YEjZuJ>hZ==VdpQi
zQO(LjNG(|f_~R}6nfUuqnbzTr+J1w9a!JKw@dCt#?&i~78;{^M81u4uG=XAahS-Y=
zXI0pA0Z}Emqh!Si^An(ApkI8Je7*r6JB!M$DQIi=do{#t4zSLq6rh;v#c^<qiox7L
z2V2Wun99>rg-T3xmztEh?n>%O?0722+<eSPfISL%PL-H_ZfU{GfAH}P9yaUbZJ+IA
zT@GOzbjuZ!%QF7SZ0RT)L$Z2O8b<6`s%PH2SdX%$VReJ8+oIoW^UySp?P%!yT{TN-
zdjZyV{x!eBJJ|Ev5bGG{f>QJ3X?hg>`ZAt>7^=as_l?y2?%ddd>wLmU<{1%*+>=KK
zJ7o^%XP%Xu+@xG<g)A`4RK*{}_7|+;duCx4R3n1&xdVuP&pZL9oF*Py3!<v!q;q!f
z<q^B-_yM9bE_$FC<UIXxm&gYAJFNYQ@yChRIQNxh&t1}Rhb6skb<geI%izMF!b5TM
z)i1ZxLdCdH;YL=Cws?jy)T@mL_;NO7@i&d7b93HkTsk1c%t%Zw0GEf`QPi4#Lb&n|
zQA($GF0JW5ogb6v@dNDt>;V6JK@_1#w7&zTVe3>~0Ot?H#gqrPhthih73g%m3t7iQ
z782`(YZpa+7$r1Rs+qTef#nDNc)zRjvtZP=?1C@${hkfX(S`*R{kPql{P9b>rd@rw
zi7$ks`iW$;CY07cyl-&8qgy*TUvp}b((0nj+`Sl8_W?J${>G=jU$-|=-E3_Co|F)Y
z3Ax*#8A)vRy^2+AJyiVYG7P!%&Mbz`s$x#1%O{X5hVZ`NK~LB%CeEtX5mx#V5{2`i
ziR~B1XAlz{KBC@>;ljSpi{p76n)PMEvP49hlI$0sg=;?T%T$t22qGwOAy$<%M4LGZ
zX9#vzL9xt3#NZf}fEyg|i;2ERAj|ly_<$of_49g?ypLUdyF3<JJ*kr2XQSOU^@68-
zvw%F70T$>Whyu9L>$Z#>%#Ovr!Dy1{zATy7uiT4+7B4;AGIDzaAtdrx3|QzU<9o~O
zcNiJ{t8|SMeIKJ2S2c|*?1(=$5?EdzJn$$|7B&#xQ7xWAs_#eruDO%W9IyK-@@Q_B
zK{Lol)Y#lfie9xRc{q~A)|s`=y06K%Gp<Aw!A&$s^j;9*v#NL`s7$FZA4$bP)S35|
zfnIj#bMFl`>;+ZFCDrofzE5qnz6amOnIVO@OUC8d&Pn^UzBj29OJ2M*5A#Kuo8(0|
zhL0RSs6PJI{Xsd@SF&_ZwXM1Y9fZ4{yS6y3VMnMMKF7iTba}tn@w-t5hWt*B!4ad{
zQQGqB|A)Qz4y3wo<Ht*66~`V4#|~N9+i~o9q>PL*ib7c-D=WvAL_%^<l!PR+QZ`vB
zGbx*lY%+e==Xvh?exC31-1q%_f4{$f|4EK<KA+F^zOL8x8cqt6%9Ewhb?TjU%7mh)
zRBLttXRfhVEo3HNdDc?R<Ue2cxGKgnJ?HDOP{$s&OMxoO3k?(J3+zsxn+n~1FJC~U
z5a~Un*KhxGm+Y%dPj1bk)8|@Q(Y9RWiFqxP+V;N(WSR|COTWHT_ydf-eCGf$F)BL$
zP*zm;lgpI#H)h8tB*DJ9+{MDqedp#E-~-bM<hPl*RMrrPj1)}`LpZ`EP-pY~PrIb&
zq>WI2BR%ewcv|bxw4*{_aWs{Ri`k`~#AfMu)+$LJo}~u`Zq9t)!#5@FgVyE3_cutK
z<c$+vK_EtvCu5OXbrevU;dz?RX=&t2zO!5`JoSra{NPW<@RV7s^8hMVb<-120OFVi
zc7LKb!k))EPe-bzzhaG6D7SGSyqQ&;6g%tQ;;)6?)c7@)bStQQjc38qggrcRT4mDh
zKCv2e`0H6i2Ell#Bj4Yfgx>(dRhR(NJ3a3KNi=KyoxDf7^k0(IN14V+x*y9|3}o?Z
zo0*+`Om(HsE~?3516O0)?<v(@cA9CZ{|1`u^u}9kh^<`T{ThGw;zP)+YL^~1@kCza
z%&1wbjg?3WtKiVOg%LK^=tnB>hOf`E&HDJky1Y^QIM;5C9jV{f?m~Ze@4)9RLkV`V
zzHxtx`*9uvbgT5Io%3%MAm5t^#piVIeJ|D7YmA|s*ZPYu7{2#n+)Y>XE8WE(HLI*h
zHZvWKiflcxbPU>rD8=uC$5N+uQOXdkQPGz2I6L8VB^lq@#P9FmFrJd!DX0|0NIv{#
zavUpmt)NG&noqemv9UzT8Gq4}1-td6(eypG3WN){th(%kKD8GTLY<UeLW<7&(u~C>
zAfaX4`tHpt(O*|!^xjT^Or5hL%$SZy=IEJY4QH<{knfbX&rh_a4rSDihEh_@#$0r|
zcUi{$d%aV6rvGC8t(@4}czwk~3%4z3>!YWG#eZHXehpyN@aL1VFVu_~c7~wEum<Yp
zao<ZiaK+RWx<?Se-fB8@;k?E6&C<I9RVO0fQrlS+#$9WvueoVI2|L<ZNyVZG=gsxr
z6^S|JXm6H|+pN-6)7KEAsuDl;sZq0zmwbyhY(I$C{cF-uYm+*86g#>DF>MVVpxY^T
zyHS*JZSmu^?nZ8bcgIzNYVQ7mfOr#h>#z8WUt*rl`5TM`<%Tw3enTu&|9lTnKvo!G
zvaPb3_Q$g&7=H8CAj8pw6SFRNpER(tnn1!7FCuNq7kD(SB=|7#AN;Er?3Zs8t>BPl
zW@?TG`$><wmy#^3?B?=53&Mg+bl&pOwT=)eC;|C^YUiT!z@=5kbeeBAmX}q!#Jsyp
zilHYedSo0%K1OdYO6wKFp7J^&+gbibV-jw{LKxwjuLeo5w5DM59gzLlnu_rY>I&w+
z20H*`M(@6sOwvbSVQn=vrRPS3H2LYas~*!<5sn13m4^CibX9IdQJ!2~fBEH4x`@M!
zWJ8<fBh!?L8!B(jBLqDF0gS1Vh?dunlJAO=$LIj4rJ=4iSYsBFb%a$i-|4Gf+K?a*
zJ;UoL`SBK5wVo~8vt#;LlOxE;$UqhyeIfb`L)ERV-1syC28&C5+4XWdN}e>jDn9yF
z3ZtUXeb7+y3aZi{O%r4lk-8!G(@MR<(vM+{C*{QTZ$RTq-P`>Y6CIKZB#VaVvvX1m
zkE5zmTKh}vpRUPG6tXSWgQN3ZzxEpG{Bv1OWS1+!ffO4NNU+nveXDww?%P)^YUft)
z(dLO?nm0$oeI%97586)bdrkijf}8d7U8RnCxR9)TV@0o8t1CwZKU6c39(zPMjcbx0
z03LxuT_{PNURwN%*BpN9-#to9-SY!wOfEc{f6V%*zuzwI-C&$w%pwy)WJIPTNl8yZ
zzJFE$_LK(s<X0RGvy)PPf5f{0y3^JellL1g`|vZl0|Ort$Vuo>T<4Z17I8KKxp*E(
zPDnIXqA7atkv-9Kx-@Fp>YvuseGe8Uvj~|LC4;2JP;xat|5z3|Mel^Zq<$&+4sa0-
z>0*hWeiXaC4Ax}Yz`zR|&`5Z!pT-5ly@(|rQj{&At?EiKF;5xWz1h49QbUHLQNpZ^
zrzK5`oU~nV8{8pc<ntz^DmW&A0+EofNQi*4>hvqX;j|+8*8J}Kl+lOS{#u_WSz-0G
zR{zB4O1)hSV1*7COk=G-+O=VJNNI$_%}`B`bil^T;N0lf>%1if?<KCJ)ex!R-=Rw{
z_vYVz7Q^h1FXpto)o+w}YJT^9)jdt$``#Q8*BI@3o=l2<c0Kx@O$eGU%Q5#(I6T)z
zKc3q_`$jI+_L|-ydIOh*-yj5UV68I&h}#_);jAPjpFjIJn<Ow>bY)JkzM%grWS^x_
z7};FNkl}M~SpT$E%Xd}9q|Ofg_t(ahc#q@2fdl*dwldpnzZ>}d$RzJSGd9?CRma&R
ziX!sxtOMQzy%L}sc}JIU$156-#q!G2JEOiaduY?fA8~ArJj@b(8kf;*8lG5^BtBQY
zzv1(F58S<;)ou)4ncw*GX}_|l&t~k!gQ<-pS8Ugx|DuWfV`Fsw4p%X-o&OP+G~$TZ
zHfK^W2+GQQ@j5-S5pKN+BgQ9ZnA-biGMc73vlj9AQ&&f7x~^yeU(B@7Tzs|0|2xP2
zX5_@npxqlA#x_}q5ckJxpxa2SZX>#&=Mv!;6UXYo$a%8R@@4Blg}eReT?Fbl1(^3^
zkBL@82sb!YBnJYO3?x0i14MF(?jju_fV4OC)8AhN-;wc*faMnV=<^$Uz>e#H0r$R{
zvbzPJxH3EFurEQFkq45p)Fr(I6u~%G<+0+#*SoN{wPbOFMzE$L!aBK=88LIr0AYf%
zaLn7W#~B{cpMqPHPlQZ;3461@R}Hu>$7A5`biji)9spdBkYKpsi(VkwPg#$}4{K54
zNP6CQmGJ70%Knzo)AyisrMqEJM2FkHZf>LSqd4r8Q`f`dkJ(KszkZy$1o;F7Z=|k#
z4!22PC67JTzKaNU#tX$8H-@gNq^}@8gg1hdgILy-wxQ<A1MVAl3hvV`zhi+u;Cp+v
zQ>dtdD<?yzUCQl|`_~3H<;{A9!>(alLWa@u%HL~Z?l|}X+$9faKuy?B<7bvqj{z=_
zPOztvhb?v{1j*M=eh!#)OSR3m@2yptt(K+I+elN(a-X9~=1=%LWc>g+y{0Nfb)3&1
zUzE%gl5a=EQwp4^_i_+ZmWd>6Tq@}Ut4={NRbc=VEjkq6iTJ|$MxVzQrW7L45Lr(^
zQ6_l%7`W8FWjAc26^k)wQWe;uJce1R0}M*ki>|RXbHUOf^#wrK4pyFkM5MV$fOI&6
zYZ!h2tzhQXsdy0XE+z<Vs*Q)_O9yUyO@T(P{pp)V$F%xQX0h`(TP|L(<9Sd54Cc9C
zTiRR^tFFaii9G<CUjlS^?~%?>Bt50tB<y$f6yh`O$b(Q1IB5{m36{=P_D!h#$F+G;
zo27X$`hO}gEl4^AXuL~mUZ#*y-SfoPnP+`F4YPX;fK~bIXctvt{vz-gjfxt+gmXqt
z0e~HLn`0HW8tJ+0JRd64vOCp}d@xgt+h%VFi1nEU&5Oe#>dnpAb?Xwlh!;@H&MAep
z-mkYB`z*6i-T#3!D{-hx8jPz}+2d=hQ$lT%gwbt;ufqTwf9?d4x3l-+(5vUCpPoHA
z1^2OZ9rah+WrK)#oXmdy;&;2i=TYaVUn%eK{MBOeAD_n?-3+x2xYZu=Tay@Vcp&L>
zr^>!dFp5s3YMRjd+J9J-&6g6%{FVa@_SN(myk^)=4Y^8+`g(|!H}KtWX*Rs`ywaP0
z!S4RPA6_5vEiz}nKvTr&=K}Vy9YADCJ$rcg7i^2<&8)re>L?@{8H)yE<i2cUl0%aZ
zpqrH@gD{O`x&kYR;9vt06oh!N6;*=9*5D?&pt|vETmquXJbMdvHOG8{{pt=C0+u85
z7#v(#2GqOcL7c8XeFIdFL>ecPkJD7cQ9gm3Tr7U0X(FN<h?7c){;uo%hxQ>E^xF)t
z9vzmgi`8G#nH*|?xNOPNWrTA5(Ans(R6bUUWp{6iR1kSPbk0m$h1o2IN%T4bo;~dY
zHs$$%C9A8^L$7Z4f!~q-PB<G;8k64*x<QsP4lc@E+h9qlGTwUK*ry=vd;E!Ht0M3)
z4Ho^HQ>9ltD_kHF&lA1(O*GYC<v{{jF#Z-kXfEsV@Z4Kwd*W{FxiY5<L|4ooEGTWp
z3T|e*E%d+TCvb-zB~aW*@Xw+tx7Pkj)%FxMb66N0u^lF~j-zO(alCDUXu%U)qiRd5
zF~!qn%>4L%yicMEx|G$S1D0m;_kPuQhW=Wp0P^$6ZC32F;vn+FZq*FS?19jExO7A`
z?;y%wf7-Dl_7uxZxvpeO(^G1p9J35QMeYL?o>V;Uiy2`n=ac5Fe@uoc!y@WX4C602
zz$+ptw_~|Bv)iHM=#&Hi@72W$Sfk2(*Q&;dJUK_y?_FBjZBbf?46CZxZQZSBKDzy2
za=zM4d7)-@B&Jhf@jH)smE+G~Mv1=7l~LutTe2joN)VT0q_?@L>3;t*)hJqH77PT!
zatfxKlN%|mnIfjZ*c^tuyXL8$3=GuR)Mxq%@yBZxytkCn#O-eZcVF``6`O!yLKFoP
zL4_(tbDl2l3IFz?S|=VJoS)$mTX0YF<(?cZ>TXB#`7Ocre%|IWAk@-j^l>ye0*umA
z>+iEaPoRyTPMBC%v~D#vm8IDgKPOq0cHY)eO6aMkOA^GQSejyJ;&fA5wQ_p#t@P#l
zu$#qe%D=d@Jy!!d!x#$vf?vfU5?sLy+OZL>TpsVp`SBE9MAx+ivUJ%9x1V7(a^6bx
zM~Lr-RRL2;ZMN^ZDC_Qr*6i=O`OJ0`#0lc|jm^N#J0W<ib!#ds>t{WzI&@&+hs|T?
zcDh9Pp8T@571wA=qLv1X@RxIN*lKm-FFE_=$MPWp(Vy3i!o`oyo%3CX`Nzdkl^yZb
zc*yP)eD#S*UbK6W<;y1<*BYg_FF{h<>83MWbPQow-k4tTeN9^7vG@c4VRgMuXk9rS
z`@-7i%cqJbJNCOe+#=B;y8;N1{E}hzp2OGcY56-2fzF_p@c<;V1>8G$;cP1l%JYaW
zw5l{gtU0=QOY>l>PN3nOr`xb+vkMrtDI0^`b1p1IoIRtX&z>diCAGJn(8b|&A+9F)
z6<Kzc;P+?>N$EGpHiX$vw2>{`&v3$4h+~^*l7HM9<eR;DPD06a-k+K+mqF4yFdX8i
z;i#gvPp#*a3SN9st*0)e<{sDbgYcAJy(NIgYr56^)byW(3UaCGr(*EAM{Y|l^gXYX
zUOjdVtrX$D?1>$W5>~jUP{%<&W-%dv=+;vdl2r0cN}a~FyYpiTwhk;xmDN&YhHjA#
z3D#d!+yOU-@~z7-H?32>_Z=+05H{fhphGmpb!qDCETeR312Lb6`%i@3{CIz|YiQm(
z-L2KVzqIgH0X4>Ee{6-Vg9Gx`2_C5*zU)QUXI>fJ`TjcQOTf}dA6bw`|AojX-586e
z1b?G`{z;MmQCywv<-Vfh*DdT8zvr6Is)!^)2^;hguF+fLkTW0XmH|b8-3s5f$m?U9
zcUt|VIkJQOPJ?5{GY(FbtCxo>N+xEM7^keQek@8wv`dXtxu^d<Iv__xzmsI&lJWbJ
z6Y?a|Mw{$*?q!}-qxzUPflc*3T608&7|AH%Yj7^(q2pW{JLw?5!%01(p30GYs#Eb}
zP%d@S6VA6Yv{gje<e`U%m9%A1!JK4j*JR;<oVb6bs0HOaj8XFn*?EY&h2<tC4;t1S
z;3T#J+XB8`RX6n>GB-loROPGtUSyU0W<A=v1YB#6$VY9hP>z|O!ZOaL0wHB#96xWr
z%?5a%FTSF0vu%_;VTGKfp5Xwe`9oLl6E<po#)p&7aO2W=3TU3|cTwD6$DW8A<U^9Q
zK02w8X;pBzMy#u5-i|z{CBxBtJJHc&O~@+YHHjfILi<+-UN)6Lc+DU0J+fZ<dVG~v
zTP9KO1@E{Iv{w|{{0}OD{dfPz-T4Q(=8Y^I=kY<&@)+aL)kAyC=bR2Tr-qSIGsa!J
z_kDiR40V}ue#k}{bzX}0ks|Di_!QN+G~V<~JAIasq`ix%y!f`NLPZuLeZ3*MiY%N#
zCi^I-@oZ_%b!~1nOQo8?!vry&&H&oh2$eXui&-Q7ko+J7PM7-r${o~*PtL{0nuE7l
zTx!Y3v~s8i1ICpWzQl8FhN!otoOD&Zu|hVu{>rgSDq?m1VZgOQ)XXFpB{9X-I9Xs+
z-`CJdHbp9(iW*B?@=813J?-(?&^K81VSNTE&2zwDd!E~>++o`74D*flq1%=z(m01{
z#|%O<@=OKs#2C3Md$@#wOK1|8n>&)zOt~WB;7iu;8+^en;zwVl=cRORv|J>4P=!i|
zRoJiFNxn?#=hf51UtI||NN3s;-FUZ`{(Llbyjqq#CI94z0c+Z6&KNtOCNCPY{TlkZ
zmm+@Y8>{xCdwoAUTKQ*aX9;@f$JVS<))_o)Wtd%`$BC7E0vzmrFMa*5UPt!;u7`8a
zx1sDIDdxKxJ-fEt04>4yLzapy*v7hAyu5t3MH@wr>ZZiR9Qz<5Ox>&mol#ATBuW^i
zL*9Lr^9pAuqp~Nj_5Dt+n`$>g9EfjI_E43(AEEh?h+?7aiA+Z*3~CNlR^;9xiqi%H
zGP2)E{bbY(5$a11ff3!qkb6{S3%OZ;MQMi$@Dod64`ZrTEkko7OS#{1@T6Fg6PhyQ
zJoV+fc^RLHxcOt+mp!kosJ{tCO)yjqd+r&np1ek(>!e5aV1O3OwQ8``akuLgiiK*I
z0z*xpp48PTsrURIOHHQXL>VhYra77*9wOj19gKIISzh0GDjTSPr*0+sM()Bn#zn#2
zC=+fJYZi=AdAE5}b&4;zdW?0IWzi~J`AJ(+6{s5PD=@iASP$%~w%9TsK6K<~lqc%(
zPWM~XN&5M50~GG9{JChZO*BWW0XDJy?v5+V^+0r>g240qo}=6BxtFxF&Qq)X(q`~%
z<5pWx@6}lKX%2LvTh+o^M+i(LG|rynCaCVs+3uHa9OC+5_dwkgqdkGzLKeLQ!G#cq
zNT2BRq@aYVEjwY~DXAaaRovk=%>@G{8I^~(8UvZ^kJuT$Wqz+9P_&fxeAFz~$K5I=
zalHGDm`Ke$c?ea8SW`r8f_M|=VF$lj(nj>B3jU7j3TENjHN1SE@8ZPYQaut?ONj+G
zdE4$({xR<SlL8Ztdb|9_oQRF43&rqMn2tCTY=Wd;dI>s6kO}8K;EY38-Am_`;G%tS
zY^GO<eVv?ziiQe(gUe(b$mGn|Pb4()+|2ukZbw^+q|AwYA{IZ=?JG`&kT+w-nO+qZ
zjYc}fm~*joJBYvb*v$#4=$zPXp}{Nk@tDz0{-JICQ-pAoU$w5&DE!xdpokjiQ4x~Z
z<|}v}vlMcgQ|RJ~wtL(+1|X}bt(1&Ji4tc}azv~(K3qF8@kKHH4P|GT+W4d~RTyH;
z;!C<I2MEPzqPe#rha6!tEmsud`j~PVV~ZiiYCkt;S8!WZ?)8LW*#l?<=2#8vK9;8j
zC}o3Ot0{t{DMFd5!<BqWG7JCl7XAn2lSc=}ktE49rr+DWe{wQa&l17FUwl683n<|n
zYDH%5K=M+tqwJS}64S|-eKvCoDL?pAq;KKzWC3pSzz4{MnEC~`&z&IA(gx(<<c+TY
z)xEHROGtj6<%%$QjDWB9W<BjQ$c-;|O#HPvbH~T4Vc1Bhg8{IB^drZl^&#ErvL2rQ
zC%v#|%yh;`cu?OXgJX+mw^Xr^3Qq%avp9{vD>fkp4-SyLF?daG@n6#y8Q;xmkLNYZ
z(NSl?`I_=IUTcM~7^YL_zM{;VLOsBV5n&KCO%y!yL>mA%gpoZ_TITn0JmNnt=70Zl
z3-xd`m73Q^{l7rt;s!dP_~j_V5$4>@IS0)pLS+OZ!PTDuWPfmn|LOxyKcoq`)XMfR
z&gNfz>7Ra{Q~Dr9PzjC1-?PU5>y7$@-m8tgm*Z&?T7Qd@{q^hd25?ikgt!0cFZX}(
z$s~B1MExGM-~QWH>94<$9o@rGreo++fBZrJ?>-0_CpTWwdvL<@?<e`c{H7uOP`w&7
zn~(h7!~KB$SDWZEhz#{6TJPqW<mo!v-u&<*3c2->ksUEjbiF8L?+q`|jW{qhZ{kdz
z6869Uc<GaHpGZGjw{!mOMtMsHD@|==Rzv7*Wza)#3dWB$gO1Swa`2Xcb98_BaLY5h
zhG@Q{pNp)k9|p8-^Cj;?)zDR9Oah>7%>lSK2FEFf%wJrLE3~dYd-l1_NkAb|Eni-4
z0ZF6)@&r}$db4fpfx@#O&;__@H<Qk@(k!@85d+wFDkNj_5j;u#%$^%EeVxGZ(Rsc3
zowCkq25^!Uo=MlMas(I8*U^}mzNEikfB&mMsXEg<95u6?uj~1p0BJ)EOk`d}O;5wE
z*$kk-a#fQ*P59lrLT0US6!0P3$bgj(GDOa#!2nv_23fcPERneoAaA3*f5?|AmnHa#
zadv3$bK9@Mvd?#+Ts2GeS=pPgk1LZ!kA4K*Y2ijNXs9JX5TADHQoL%A8VjqmivSRL
z%MrFIf?BTy6*tvtqj+d=?f}91MMYSg8X-2q6oVA};`d!gIxJ0fb6)c@n3LayF)MHT
zI`|kk>}>ulnX{Z_5w@-}MHs<=xaKfOEyyM)Ty6T+ur<3s{cU#zoCRASaTPZ?dyeS+
zS55hkgQpm&KCiaQc(44^+M<|*hFRfyA$bsj`2=b{lP8!!a{C#iu^IrapChbuQo-e{
z4y%$8c0e3d5Cu@OdkbIKBzTD0)W~`tC4{QAs+ZM9(edlTRX#nn{G|exD!D)i>v?gh
zvv(I<S?+=_tf*uRpvGrm-R0;3o2KIvkcN5(il_~^0EysBy#a94g+L?yZOS$WFh@91
zw+M`*WC+k~;o4>}i7dXv<Id(mJ4i=E5p$bU3G}<ozEZ%OW=`n0TNXZ(f(V1j%aFy&
zOOkc;fvXGTx0QgCcxjB(J5HfvtYgUC_p0E^nGE?!Ptn{T5a__PI?~P`{3kW|Ki)4@
zq6SXYy64QMY5c#Pb@XWm2%o=zgJMs$&(#^QpB#UTzE8&^-7JLY4|Re@^x=Jzi3y)l
zN~vE5>XdNn@nsATuXcE11Q=@3sbwOzpv?*HBq~_ab{Rp$4_+9N@PggIFc&9A0*hHK
z19SUkAX85Occ~N~v_(ISpz;~OI_2fvMC)>68{wEW-UG9*HW=T}BugW}q@>%hIlJFD
z%?ub_C|CFzkcF5yMu|t4XNq0GXt%tIeed4nH|(cyqSbxwwy9hzHwS(c0R=eax&Hud
z{`%Q+5rpF8DtNNN<F^M;HI#Vp)zzyK$LgOi>%xUjfxUwv7zQCuG0EHsiYAa1m|cil
z32kLPv-IAi3S0%~vDD%9k<%d~;U3@=Xed&<YQh3h%p0cD*9*u`!G4!cSS<i1@>Zpd
zVf*er;1Pa0a{E*nhQ=^psgAT1DD1{dF>i%`^q4s6)Gh8?7WOv!nX9yV2Y41}72P<T
z8HsuW99D`TwQIhy|Emf<bbw+lH1Oy`-EY;FGxGqUM9px8Pq_O2V$W0jHPV8kGo8VE
zFHW{lZnb5hNS?!XWAA;`0(aFg7{xi$?2JG}M@-NL*awl7BUyvmoza4u=o=pq+lS>5
zP#!(_SqL#6*1+(;PzVTmX6%X@sjkMoI+Gf6J7l@Rggt>x#tN$(v7qtRm8jfxhAQ7v
z?s`s50a41bB=iH}F&D5AmZn<LV#$-a^+%6Mn{#Lpx4=%vTm>6kh|9UPv0Ua_#>6H1
zosPVlxui@|+5O1aVL$t7-MfO<-2n78Y<vHuZgU%^M^~M*$p@GDQ>;LM{>b!+sm^OS
zU3xvQ9ogEVn;nfm(@WX;&KC;VStq3=hnbVq+lOZ3rTYO5Dij`B*>al_+6-aYybut^
zl(OI0(<gAKHI$f9_CaUO8L~FGnwBaabGN2QG}f_0Hx*i=!}mgiEWt!&<8y2$rm=E}
zsU~eg;c8fB+tGGo&14oUvD6&vr)QpC^z<{&T37D-B9hC#P#Y1zC#QLvi5X<oHeI)z
zR8kKOvJCAJpLHU&iqU=wgE1X`Snb4*-Kluo@YuTR;_hBhq#UK|n31~${ZpgOwwA#E
z)p&&TaKW=&HW-%oo25gDF_C%F20VfKnUVS*L4aM8s&O33VGUHQ>-z{>lmIqPnaQo4
zU*4J&p~|DvIL{%j{T`2WgAfxkNB;!SU7VBf1@nd5?8nb6`GaX@9`Mwb2d_NW0cI~z
zRs=p22!Pd5I$QWtjwjnUDUWuO*nQO#t>pn*%v#u@>B90tr4mSg6Tl_guzk?HmygRC
zN03eYN%IFxG|Ye{C*R)S*jpk!cbLUQ-xxChY3V|uHA|rypD+X6D%S$xly+?yhm>eu
z)1CU4C@J(Q%!3eODfjQ#JHaD?8`cQ!QTPl3+DV-U%L-0@7>1)gR{M5h7i2{9Z{)8=
z|3d77V)QFuw~>zW2QqFOn5_U|llWHNGn$Z|bmWYti$q1y($J%;)e6;>ig&yssgCDD
zoH+}@+C;e4F{DrymO6Kwe(r#C>>1-!jLGZJudo|(9k`-c4Muc5Pt8Qf_CY0RZX0s`
z%yS#18L;Att2^#9EM6uXQJMj?7054X+9nz-fb&qosf%-tX(aN3oP+S`_`S%WLl&N3
zcKBI&WUgi?B6dY~WgS`E`GJ$6Bglk4Z*)a=tI*8Qj1lhuvsF|MoE(G&ou%c(-IQuJ
zw7wW#j<t*Bo}C4{!_~HY`;PbiD9j@KP23QjB}0{XwRQ%+Tf#}_gzJsWO%;fixi&So
z{QXGq!Y?}464)KKj@E^p-Z+0~KD6exF`w`n`)W{J8^JvA9Yx3fzdDb<D4)r~&{y*g
zNeVOnc7o*6-K;0ZH5#B)L~52W&`iYo3(WBoDk}As5yRhbCgC%4M!~ytws}5_BS_6r
zzlCs?4QxUjNS*Cq!}AiamDu~vq4*&N%do>atQru;M$5;H-ahUiLZ+9P<B<nMJuAOS
zPj(2Xqx#N28{z1SHpFa(++?KxAea@fQuzqtiI<@SoeLLNWb)t`mkP%42njR9l3iCL
zpT7p4m6qd@!wK49_}s&OsBT6T;5*kP^&oPHMvvBJYR73<Xhr!X&R*jf=hwC|vvQAE
zTY+J@f4NB<`^ggYIT{n5ZD9h==7OyD4v^z==Ma^wDkmSoh56L^SZPF@J0FDiK&^9z
zR?R?vn+kO8J|2#;gWGG6bzz6wH%<V$5G7WwIWnIci<RMR==ngM#XL4jF+fIsn+=V-
zk#&WCdD94PF`!pFdD3nYj+S_xYCU*?`Q3@_Iq0Yyx(OP58-&io8kBMG5k`*(7+H|F
z5}lwMV@i4)EUdbX@mIfb>qn$AQ8(>Icpj$FFT?Mc$pCC23aK%Q{@~}QpBR25u9DIl
zfSpHcyU@vf9xpF0^-o|jp<dj@y}HAkgMYUHD{M;H^GEChc7G-Wm$@IMTAAv6RdRS_
zD#J5A>ZO|;VuFd}-WTp?*P<71>HN5`sIsveH4*r81go<C2*__H5M8FB5biDwWFESN
z+P?D&A~AMr<iT>4M*pjqGqxj^Ogj9>>HU^Z`#gwau1QLqQR7vaei+6UciyzC*8Qr<
zw`QDL1MXW>H(s~k2V{`V=Ld}E@7$43>Aa;W`HnR`UL^Zo=R~~bjEAJ9+2<=9{y)Uq
zPB&B*_Eqv<t(;)2w&8ZEjEvqUcVmrc?G^}+*HrT|edHneBsr(_agLrW<J!PpySIho
z(`wD1D_O1nwi7&B4K=ZIQg*3p?q$zgpRr8slA2nTmLgDbyk>`Iyz;h_)<jjWqu_+B
z;B$FaZ!zy9#{{Z%)eT`$z;-vHr=!4S(uqf2D?+gCTwYz-N`LnRCKE9r?{+jjez#s?
zxetFmo+3J4H|(>aAl1e3kV87fUZ>8ow0BS6IeT&GWBx^|MjJHw#My6Bi|3@i6O_#I
zjI?sd9(^G_d(U6&^O*<w3Xz#S=Alx{X5UWTeM4ba{sy*@z2YwCm5p1va5F;lzFmg2
zV~Vpi&$tLBy%Pl2!W@+wcPN5B_jwO^FMcZ=(&^=%J5-t>>=k=C*+`3fO^)?MFD^uK
zjJztNDl90n9J_1E_o3o2sp+UEco?z!ie!aBW^I;>P0RIXt+^e5<-D-Ao1$>U#VX%^
zQC#!aQ|a&1pcT1xrE(_qA5>0(d#a|m`d<&1>32BAXH1gfHMy1<TP!L)@=m|J{7IXO
zMglLmi5QM{J9#*Ft}m+I521ev{I30ix4Wf`7df}4X>aqnI)^Jx5n@C<BrjW^ABYYV
z(>Za}F(Ujh+A!7dB|*Ihb8b4RMegZ$ic^ULoD&n=3g7)$$=kbmmDPDg@wZqHcDB;+
zY^sqxGtR~s7EIahZ>Ju@T>&a$h__A>FIFzjW73IUO^bU;j#d2BQrZ@qFm_2@N$R`h
z`SrvS&e03Qo2O}oQ%|0|IR6vo!pEge=6h|nd*7XUKT3Afl$L0k>RYF}a!0wddyiaq
z6u~>PI!hG$vOfE$239USv$q#t?DR26Xrkwwdlj`(`^z`NRg8Q%zAqrrr4mIVKYy?r
z{u0)x6KxHMXr<3F%2oSL^?%}3|6Q<ETae+=!{8m!ZvdET5HTT|Ys9MU^inkt_JvFD
ziFi9<7`QG~23ILQdI4g51?XH?7-i<Jo7Pd&RHMFpIHuIKe*mizW_tDs&5l>gIjvNc
zm-(J}ClbFI5dE#0<>;^9N$lwwsiJ?L({{-+R~dhyYB+BNoMYq?>$(`Leq8R@u2@bK
zsr$iccTQO+wOw*RKQ*r7#uhAT=ma-slA1q1eoI7fELX7}>2rp6`~1FMXx=N@(QRlT
z6rdNs&Uu3IJ_XGVr$Yp9tdjO?HC#v1d2(mE74Gejbg_l*1eEnr331wOGJ?o#Qqh8=
zg~}b&7@=dRs9g%YMY>9@q4RD$wMz6gif8s)o5bQ5Ort>fMQl!ZZ-%SOw_(UiDa@p4
z5Er;h7-e#NjmeDe<_|Uh`TObim*=7C462_(qno4p-f#ExO=4+rqAG`cr(}t4M+@%M
z8rTb!LU?Ek&^KAPMr3829HenSwY;KCa2DL$Pm%4Wjqf*-Zr;-xsd^t|IlM-Tsuz-!
z1Gn)a-9pfWQdRNH_XEr*#2R5BIe2Hfw9t8K`1sw!h}x0Xy6B8o8gAp-^kZEi58wpF
zz-w3|#8#_#jD&4pTS-z76hY>c@=RBkpl51Xee07@%T%>rAWR;F(hIrLo{J4r;C_BW
zP_zXIv;56^Io?sNwglRH`LVn(jV6Ncb`(Vr;#au%jAaO%=7BB8$66~==DI6$;Ru2O
zKmyg#ZpWP6$ESY@TmsCba53Nbu6%qPL@kgHj=`lfQeZF3(K9}md8U~@&Lj1@j#Y|t
z8gE4o%>h?M!+T=r=T^Xy6BIgIh_ViqS>1*fzh&v$`w75^jH>xgq<|4NiXn0W1`kGt
zEte6sV|K%~o*$&Zo;MY*xlP9);s{(Mf}Zf>yWCFv|9qy}HbI0*F-xB;_BbR@v-XI5
zrn_*|4F`w&Qz4kfe2P8~XntD|6DPr&mjY>-6(N^^(36p`df(j@{Qs~M9=H?oV^+$f
z5UqGA!@F{^tL^=E`Ti}8IcI|1DtYig#s5CSEMcy^d)B{L@;96`Q6?mxfR`CtrrnVg
zOn0i6bB1N;G;QVxS6(@K|2V8*`8Z}v9~X>WIEEup!8;@>a{5Ywqo3G0>qav~a>aV^
zr20FGs%*sw7O{wF$S%Y{XiogJN391iAnwPWt@N09_?_>DaD4J%8lzNw2J5P`UTIb!
zma3hW$ROZ4^jm?@@*K;(X>pETZ|NiN1p_1;v=5tFcGNLPFDP2X=a|6e%<^l#vP{I!
zXyccW6TM~ufOCW2WQ9LU==weuCV7)<j#*%ZwjDZ=#%)x69iRk=IUzQut_A7*oetq^
zSZNG^H=T0F@*l^H^^r88s2k(q3?6D|lXQ`nF3S?cDyt0vJ&?A=EZ}Og)E3G9{0|?x
zzrqj&<cRX<^b>ji_(D8}bJiqbq^8Y2%0{@Y;PyIOT<FEN61#zhhi{OctGoGPIepHn
z_6tL})ag%nn{u8j{j~AsBd2pD<yhrJe%xkhYV&N-qd)8z1S`}G$j5_2u2~NDVr;^v
z3Qg|k&^(i9O&<@DZEJl}Veje=qILz=U9n<wpoit0tYmSFvSA=Sv_c`Dk7IQmID5B_
zEsmYpE9&Iqr?`<qHr8G7G1j=f$@45j#qtirC=X7mwnSl9=aGuQhl#v50BAaXmPa8z
z&m$~l+8{bl)BVLdLLbTil36=$;FBL<Q<(rxj^Nz;Q^GXgy!Wl!Y~kk>@5t#GTJR`j
ztrIQ7v);aT=Nv(%mARopHGSd~X*gzkG0IM~J7&Ym{OfwEfrj2(4(|We?V4K_{r-1k
z@c{X_9_JH*?;l<%602Az^{S<zmmKxL=j2I_9-BPtU6CQj0#6*t8^BoxTMv>OrXDez
z7;~F8dAh!{@TIys(j$luy*D19&CaAM7DoI6rZjmtA!>8<^5j_y+PeiV`VCw6^P<LA
zZHr~8!E1E<$l)W*=Or%PV<!DZH6OCGa@(JR<|5~~lY^L6YN51yqv(6JJ(#;Aa`Y-K
z<wA*7;<$hg40w<y#oa0^mwYnj$UF|^)}z_3U_KqM>F4F;XSvVox0sqYno6%{T{_fQ
zB4Rt|<fW~ZlVTZI$4vASqB3U6oVwoup~BBo+b%;Z^HmJmlNXmzlf5M(>=0|m-sSk#
zHDmH)g4FWm{6fkQ(;O+NO#33+ZRkigTHiyHRLyu+w1lzE+E-ID%5i9SzKI^&_@5gJ
z-5A&<^jxM+x%C^8Lqx91b~AQzH%x%gR`Tfhdp|NgHFCS#{wm>*?ik9Sc)7^F(Pp5w
zMelK>6Y*iyxRMSJSJ>{u@Tq<1(c?ujuPh&}e_ot_d0`jo#xSWHo@@=+P^^b!M78mV
z?mI{C@qOkg_Rckn2k-Qa2T&~TFmUoUaAH`n@@4MgNg_E{JcbZd2VImc8U(i<V+l=Z
zno9O@1ly%?MtQI<oU8FI$<dQfH}`peREGHT!QJ=sZxRon4-t-kv0u2<U!j`Dsm4{x
zQ{RUe4iL03Jljxy*{rMjxa2Vum@IPI9aIQq5B0KYy=?NAEMqFHDRv+DtiN_2yQ)y>
z1>b_)hLVrP7u3fNewz-Z6d}tzwzILlvBI$PmU->RaOI1Sz)l}p_w;$Lmi*#rQI#O~
z4I2|-TqKf@+|3&tR|kf~CZBqE-yXN77$klRSa@!{zs#^l)h=mPGk1P)#D{eEDc9ls
z#$#9ePeC+rx2hj~pB3~1jCItG?fiGw0`Ar`f6Dum31j^^mBxzIH|k4AVinp?RUNtJ
zL~QQUVNrX-`>;`szgLuR+o4b=WB9t%(*Wv9USM4L1Olt8Tqn8A2nob%uH#Ip8YSCU
zV{5<vjnsLOxkM1A6Hx!swejkzYS7`FwN0<<31w?uqGLFN<^G;`{3@IjlHfWuf3Foq
z(ECWCd>a|BaV5H@ZaUS54bnI@wI^qv-M_Bb5wCeViG-4|890yp;FWa;MsR$q65`IM
z2TE}hi!mYc*@Do7>E%j-UPrV@jG*AJD9ynnB<92WnU6YmP?X57kf4J$FjH$EV_tpr
zUsRgu$`1Qk0Sx2V)MzFGZJblR=oZAm@Kgo3Oz-Z|u}0~b9x0R7P0HMt!)-vg!SQZw
zJOM}J2#y+rQrlkP<tm#A>F(w;>%y=0^H;f?{22zZsg&ie4C_71h_~f9+{LfTyCzH=
zfn3|5qD*sl4x6W-HgxvsSp%EkUGKRmz^&--mSZ)lOqE$b=$dYp1S*ocX5;3tIc0f8
zb&Wth6=5Y2p1<dF36z+hf!kzEdUvpp%~!gSq{F_ZfRjhi)bHxc`QHd@g+>?jpQv7O
zRDE;ir>>+&#>7t``$uUezmRTe#!}O__GB?NviMdOK=(o*%Fh2W#oqdhZoa4BG(WEH
z<eU=gjImg`SBp6^k3hIM&NP<ZdO_YhD>@P}TRxXQ=TY#v2og(@q8Wr973oCt4kGG}
zSJ%6=6HABOss~OXY6P`lb&Nma+*Cflo6vf189HQ(V6TM8HBsw5GgB&qIZwD>87f_r
zO)q?Cee$lzX;pSE-ItwqQaJVw5Q4N!AEp4~Pu@PD!f_)g<sY|iL(NTT^*3%GAH^ME
zlkFO$Tcj<1XBAd%`t2MnwuA(KwSx8A?6Y!LdpEGfh(ws`>mzSvwJ_|{)v_w!&hIaC
znchrcYM0tP%d<o7EaTma^R?mb5k9H9&FfS*V27l^O>%gBURF4IfWO9j{O|b+Fn*}c
zAoIoz*LSqP&l{{E0Q2HChw0)BldmhfJcj)V-)G-W^(WohPJH6N3BO*U_x01IKM+&w
zS7J60BoJst^kkwVlAsx;6CSo_Rchg3vBvZ%Ib>D;D2<t>&EX??l)1TRhz{j&i&*{q
z>Z)zGlMml(wTIyc$m?cdJ%=L0Z9RgSaQS4a|M77qvdjiu)(JzzA%%6fb+7<DAP6+i
z^sB-v&b)H7o&h%yr`4xrT&jaC5Y)aVmrPT7G?#bvptg77I1>A{%hAWB_M}e9Ewuhn
zC7&74;4)Qzw5C7;&Bag@@o_R6lq33=Ggf{88j!dP_eLWF-1lK8=Kx0B>&({ceZ-t`
zp!>vRcD<1tyl^XgF6F6l%EImff8!4LVP0OjrO8fjT|Pj3ZX+PMGl1j`Fb(jUmU?Z2
zqKCoYSC{is)jR-*212PKCo-?w(h5tN5D?gb&`mr6fS1ukLJ$R-I?#Z7JZgWS0%Z@u
z-`xcskKCyG4}1mIVdFO5Uw)w`PC4u~Q9+wQl~<)w6GH`BYOd#f=%vaQ73JZvk<QWw
z&y!elAgY&TqUjgPQ!>}Fo$36+pEUM}VQZO>USfN+v<g6w#2(@fHtlB;iM?4o>D<LN
zkt6($Y4_qTdFcY=eA_KjD<0dhE4pbMyUq$AAmVHN*-b4_fhxvsojh3S{{jH|&+Zwe
zeE^pM?gTk+J@>*;w}<nwBc3P=PH_i46c@N06dR|B4Al;Fpl>`h!_>V5S`f_!O3gV>
zb8neXn;t*PltRI<llT?30K%omxo)X7-qiL3Ks()valPi@7@*h0ce85})b3+?8{ltw
zJFeU;UWF~x=s08+j8(eoM>9Y}bGVz`DH6czp+PnrJ316iND;zwg>r<e$$ay}$EOO-
zBo#W}9`~@OP9s~6PcME0yVkU8WO7o%YMdBr&U|zz7uTj*^Sc$C=vB$;V#V*ptCNUp
zPI~_>0es>Elw5IQ^7GLRAgNo{)^@%<^`+TQb`SQSmXbKLr7&NL0TF=)NAej;`(vFO
zeaAg%2kF0i8i)R%Ea{*3QR}N{;v6~dUQe^&HhZe0=BP5ttQI(!+z-a$E=J?8q1uPX
zIXD*F7cW_Kn&~9qYPcsmv##Fe6u&eYI2AA6oZ%L(el@dp=zL5`$ht;W!{%Enk^$;P
z)_{w8iQNEP3F@T?i=SY_U93A^$~%&sIN1Fvbo)9RW5W-+f!Q|C|0aU`$7d!Gj{nxH
zgL;3gAW>mAg*GABvzJR=EhvQkp58M4r*<_1F2`!?upT**+|2t5dv#cTd2gTp0Pcv1
zAQ!%}t$J^N;DbM9C`U?(cPihl!YJ{^vKiE9z{{C320h*Xg(<2M1pzx?2m7zltyqWV
z9^z~d^{%{-i_00^%7En&*0F4ty#B*HrZ@Z&Cj~u$nhRqG>))6y^GWMQVpPjfkKW%2
zNEt3Azs|CliBFB80#P(A2tHrUi0c85>a-PQ8)JE5S=+mY=Zbm$<K6g^f1>U}Q<z`x
ztMogHqpCw=UU4mZ7%VZI9cRD3c>^;R-@JK`0)9{iq%37F@DO$ow5as`Ieln(Z|2FA
z>^!ax5LPaclpfp$qvN}P9W=v=mNszmQR2YOgNMeU{Wf4=&?6u-UqPT-vhu+Wv%IgG
z!sy?B<bOj(zkeniDmjzpt)_qPZOa9Sz_&T|`urd{VI92-t3i`&3<*q&(W^fOP9qyo
zii~rn3FjUE{Y8JEBmeoq!yRO4xi5X5{NMja_vTxednX_#MJML0*_)t1P<U0sGPr_R
zx0&lN!I<BF>wkWMi!-v`bw1g#{lEDY|K}@zdp#}kdfE^5nE#Kj|NnpfU%vk?jKm;w
zV%h>dK_2X&P2leNKrk#M{$F>bsywWnW)vYs@wZ1QG2$%@IwC5&tEUGUfbyRVu+8J+
zF{LXoL$|_?@(S3w{Kjg7uB=sh$c0<#a<7;(S~sYs+=Z+Vto8?w7(IoXxA&>pa{=lR
z;G#VIibS3`3O+}cwn*HbX=eH0Wud2IV11G<&sZL!TnlbzzYe9t#_^W*{Z02~Wv63J
z?k0Xs$8!o|D@qz@&8x3Jg9J)MN|8ktCOal{n<&|y&o?sYX9JV7J$-|7*TazW52Efv
zE-Md2(Ip<DRy4K*WMFOGyzjZi3r#pQBg+&io@rU~4gZ0A{1rf=%?%W>qoh86JU^v0
z6{|gZj?jA)=q<eu0?q3LOjg?Pu7vyog{0`mM+SEgt_7zjY{m_qm_2usI0L1A9N_AN
zB;MyIxxw%y_%(<g^n-SOYB>*-<_+JNd#2&Hygbo10YR!M?qiLJ2cpD-gW2=&K6fNK
zUPISQuRg}~Kt$U(km^f*p}(lg9|y3g3@ow46jrVAPf*#wr$r;I3TS0oa9YP5H%LZI
z!^4=h5FMkc1U6bhGsd{SrZk*5;x&U7d&ivGxn_;TGve%=hD=G|>>&sSGsz&+!a+V^
zB&Jlmntygvx7v^Hs|7^sesI2bDpBk#O$cZ|{gM(?e>_*7%tP$^@HE^G04MK#)cDeL
zY6@(5JMrfd2#jUTX`ZhICC*>`do+)eey9Img=v}ZK+716%ZAn}Fp|wcZHB*>&s(^)
z5$y8j_vF22&68gsCgA@d%VZ!<t-7Fpn1Bs&>_6XFBxjS;qZgL&i20s)+Q}I$b!On(
zM5-Q!kn2>Ui7Ldtsi<mJa8AZod7GL`4bwW;lNaL!c>F`V>>$VsxcxxMA7Le(OaLSE
zR`^Lw_c(Hun~emfpv7^3aOpb<%G=u8YjgZ`eEMWHkC{ou-5TSMr$;|w@QG#2iQO6}
zcP3jFt>=B@ODq<TM12O9fG&o_@m3}4*ab*!l^&splAH>>*86j!*5~wB&2R&<HkA2o
z#QZj0hI6dS(;!*iuEsv&JiGARPa?Um3_IUHbL(d6&0E5;@|jDR`tKfMmms952D!Gf
z{yrO|(Y!4!0O(jydnv3P@t<yky}b|;(;8^!XCs5IqFEn{=5Y(!5vB@(Gsl8h5c^b5
zz7`DsJl-B8Z3eYHqaf-vEcy9fR+P3w6yiz3R~Wml>e&F@E`~(oq$f!(*yQs)7x17~
zT)Igv4O`~~WB^<D8NaQi3&Td6>C(z8wG8pu&Zoi~=(Df)w&y~fF6sXyQ&o0_@f%^T
zhUVu-@WhH@pJRF60uC;;UiuN&WQKMeF<sbdQrSb}@s|0cqzAVlz9kVdGSgAkHf!<c
z5|#yhXgenY?h<QEs{LNE{%C~${e#3f=mK24eB*!T_;3n^$S!vQ8&ME39?`ft*OjgQ
zd1}dFhE^bDG*UYtQ42Dc)Xu%D1m2k(dh`Lbi#DI1o#RKG7k7X|#Y7ei-sH`G&-X*~
ziD(Ij5#6-21hQMWj5%YWxwO*9-NC#zbtdH(ez@Wv3?p*YeQ<+t0D~($?`AeLGaV8Q
zWCbk)SbvV1^Tx=(wo@vszdB-7U-2Q2J;IBr!z{vhOzBTrnt%JlaAZJl`WM4dRgP{G
z+FW7z0Yv7B`&d45;|d~d?R|1auQ)E>$DCDCSPg^0mafh94eiE{nR~ss{OL*By)rQJ
z5C6jItN2856+~psP}qfhAvoG|qU@vLi)I)2EzK<V1V~fnPhJ$fnsWXo-e6PROSt%F
zIq9rziCu#+abo|q=jl-DL^jnCm6W@tirSlekotZOI><OpS3y05&oHm_aBQOvq{LMx
zY8m_0gmMteBvWp2E`NuzFNDvH++&bJ>zc!Zm<PinpWnns8JZ8&fzH5Y4E4AmXD$3R
zX8J7dG%b+{TU{r^;cii}M^8-)XkSuX59}4U>{6$#YEe+(B!FFNo5T8#fso1^6+*&A
zJ@fmY=M$JIQL0~F$Nd5&E6ojs&77}ebmY60z*$|7vcK1qC{x}oEgM-fX~g`InF**8
znQ`ul9BsvR`n`n~9~;G%)!uEstDugyoIU<ApXouwwtc{@Plr5}?iGRb4GH=V9U<q>
z7!@Vd9$HSjiZc|>iC<QKjoqR&z%?=rA{D%Ri)K-w{ltRRnC1d8C7aLGTbGIGR@%ig
zt0ZN-w1cFb6!-6RiFWl36Q3OzX{jjEPNT5<-3tIE-m%X%J{CxNUPXZSL(i{R%<MN1
zbYC~Ir|;CT!M!=gD{fn9Pw#l`mvs4;vg`Af{#jMOnCH|MPse_c{YonKV7Mgwb)0g1
z>4n0F52ojQ7htAC$#zywjQvhd{PTWK^(Wk0LhR=zfA=eoqP;nK;KTwr+24f<*i1~f
z(|-;)Icmg;&5S5YD2Pm(R`_#?kC0#@3+S)|;4tQ^QPr<2B>-h66=r$OBxypcI#4>x
ztsn(%-F_R87C|Hv;~%<R36A>w3-51{P|%0mjqxp}Hzyp$?VH*7X&|JzE!bEY6onOs
z;XU!m5fjcobApXA4hq8w#Li=)!CNsJF-Qm+f{Y<bLC7!*u#=!liP8v1?_66nLM1!^
zPK6NZAZZd?P;=9Ai|-y{*Z*qRBgP7TO}>rIMIVwku=7D^gxsJ$FGYl%d_tdqyZa!<
z>BbfrcBBTk4;I5&9*}&}3SEB<V0e+#9B#XmomyLPqk7+%%Jd=1!*|85V(AZ_5YCbf
z<rusGnu$>NP~Kpgx?)s|=q&Tmr+OA_`#=%2m*8_qhy7n#uQ!;sk}9+A9_q0)4<(M=
zT1E`q)V$yXlD=f8m~DG~rbhz_<rn}5IieRgRlR|~G;|N^$5{gyS$DnD>T~#2jbQeW
zx^o%HfIw)W<OQhBVPf==?v>(z71#iCGpD0oi+#0=wb!>}xg6`*9GG~iD{_~p6BsHN
z3EYj6+lMAawe9ms8yP3}ks}K)lcO0|`tyhmS=$PGfu5%Ny|u@PH-q%oTOXuG=HQyS
zRKig*0RUk+S^9`4t~@99Igq!HG1Yv%4=Tk@lHm4CX8zIGjq@O6b<uHI-6}RE(`)|#
zv>y%Kn?#R84w2s1{}J3_AoIn+HCm1(@NnQNOp8BlQ_5}e%%sj+2)}M3Z0bUM^GjmS
zB;nK4`-__BwWTQ<M-?EbPBCnM5s2El47us+@uAo%XmKuL8|xJqJkhj+%qxtqPi>tZ
zUOeR6_x($pht;(@66xpP2JXEP^Rd%cdw{?!Km6?ye=Lg9{g16e5-;?A77yBg_gRX@
z90+YmM5`0)1!Bf;ju*rB#ge!nPcO9<y5ApK{Yaw;ocKiCVeEN7;C36ZhAH-dxMXXL
zW8cZ71u>civX4yN5e%h6xD`$0Fe&zKBm%`gCB+`aMyd^<t5h$q@Zn<;)4yZ!%ac=I
zL>N-e$H`Y=Y=N?N(u8VU4+Q4<r+_vm_oMiIO@omuVyKK52K?H*z^+oG{`w!<`W={L
z(=>_?+2hH$HDgg&jTOE!gg1vyJi@v6@Fc7WR>FLV(ewm*hc1LP)&u@ri}U42>!`OL
z8X@QU`j`sWRLBazZI@gFzC|!uR)%ufI6CB$cG>!XK7a1c^w9K5B`wd<WRA7NGCa+N
z$ak3lA8x5FUNiQQ6&x|oV&KR;JOsJ=$wLN-VhHEaM`p1z-(GStHs%_7))x|7(;SGz
z>6V@E)kj{RhZ^B@qU(}42L`LQulDt(0|olX3W|6L9F5ZRZ)xAzTSd5Mt6Hl<+TIn^
z1j=!XW}C!UNSa<7N0ZodyM?t5i$r_Q=I_d}<n03$)bCED#4AHi>a#2e7%)UqYGQol
zAfioj24yk|VaZpLs0mXS7DgOXtzBFmvct&Dw0a8RjxGC~25}C_nyQ0{a8$fFyJU*k
zR}9-m=X*Eh5u;U|RRgky1=*DcojwY!o9`4TJjoCu*|6AV_SKt0xrm8S3lawesGg*b
zq(GH@ws2i>@c!O9RYNt+Jbw9Rh>BWQcp^y&JrN%lALc=83I)yB(%#L{X}+36#(XtQ
zrFpsLhm~>?ZxB!1SuEjcbjn!N)M+;ZR(k<rrCyG|w@J!4-8A9_Tg^d#hNiB^OYB-h
zzmVH~lS<5WP*R9;d<NQ-cexfdo=a3}Z|XLVwnuJ8k_ZU>!TS2MaXe9#g{})v`fnQ(
z!}Ab~8!yr_onQPt-z5oWxtl%E&C8#uP9?`P#zy7DBlHgrQ(X&Ahii<L!H_NP;zU0E
zH>Wabt%#`DE1^_Q^(QWYw}9GX5B87vv*wFk-FfVuHtFLedGAanh_zBVyuB=HV`oF6
zbQR(?AGXOdd8r29^pQ+T&G{ac(_O(9H#{-jbK`n}=Awrw*)_>`HiOT$)oV{d`jA!x
z!&H)<h`Xeq`Jf-eD%&plE6%-hP=}j1FVn|o5Ax~d1Q`ar1^Fu?=TDQIB%L}_>S2o2
zRY-N!%ES6U5|Nv6>Ls|OKI$lNm~dK^oB|6L!{h2-x)-G=tVRL{Yc2BFgu+^57P6bO
zoG9aN(QhdaKkTNP-(uz-eM+RIwpGuPL!u;>o&FnBOGswbmSII=S@mmE4N1A6j$_6K
zhp~H05n?&xQo{yovG`W+g0`;3GtypNUx&kA2ZK|p9IH-`n(gw2{px9kPcAlEa-ow}
z46grVtTm@gKdRjtlM~ccVAbhttkK-}Fum->k+oj%uK2EP&Su_mQ3tI&+cv!%Mn(_+
z*$0?LrV-rvJOUxd<AR{5%T0zxxyERU(?DfeakqHFHj|IdOp%#1sJYYqcsTs}yK4Gf
z-mGvRky(aSk>;r&D?0CtrVA`*fgeSa;!`vh5sMp7>-EX0SYa5sbI+`lmti~;vN|tT
z*LN^12Of}_RUs}8%76DFhftV<aT$DJy7Xde5e_xaLHC6}Zzg`f;gXLY6n3uvL*Quq
zofz7J$<Ze3xVMwjrCuQtWIQM1B$9D-cy3nLhLFS(054)!77l8Yvud=~CA_bS&Gwg_
zyJTAMHr>LHJSeZ3Rw7fAwp7JzwOc?wUi0QM7v&etbhQJY@?)uj!aMn$>-m3)+j2`^
ztYtGv)a3VM58nCtc8-%zl{Q>}CYf_c)#fMVEw#R?rfVk`x~HucvJ10D`+#+<l{zEY
z-Q-^)VlhA6QyyYhxFNrP9Zi&^k{}!U6=DZ3a@`R@Uzwj}Xx-^-Dd)Hpd$H|R{}fQ(
z_LkL(!%&ZDPR!Hi3a^#`H25Xhx$(6SwX>J<+$C1)Q)8_A2je=N?XHyXAe2c`iKDIG
z;{C5G%ThZ_rS|4@OP%Df{ygmOe_27Nj+&M=+}irl=kDg41(&d21RMg?bPtKfX&&2@
z#m}O4IPY%(6;6sZN2mAR)x;(8N!FImx`0`}IETxwv#F>rD|?+h^%ePJim(UX`mzhW
ztVxEo7m4Ec$B|2WrYF?4ck$5ZJVlUzUVp#xeA!2RzPLf<ng_qnNPo&N<WvTP=qHN=
z>2JJw0r`Vi($Ij%dIIuQ#7}eDDlD;**3usw6t#ml)jo54BPUU(XWZ#m0Iw&l)adr@
z$8GBO@7>Hd-pb8436bdp%8qUXFE$5yt{$#fv>gW_ZYHUmF^~e+w7oum<#u&V`@3@V
zCi=-;5gf~Pokn--p1%55#rd`iN2kqTS}!TBYO2XfJQy!OE3lYd=8F9s`WBE~nY`nj
z<({xn?^l8(sFv#%vuxWcgp9T6?)Pfrl8aeGokft?*}JyoLeIsywGzj@i5A65^g7aI
zLQ=)%iw=L44(1QP&>9tDMPgxGkQ&v|o#%VpdZCYNP3=fnK!ku-V%_q5YLTKBG`Z>F
z=|<i(-A1XsAMnFdZ#Eu=QI&cSlLs;9Bwj*KPu#DD;rZ%gXl?m2x3X^Mjz-paZ@CDa
z3V(<>jUJ;kVRcw5CjLgXd?68)M%+eD$7jChW@?08{5TO87a>!cB~t0q7YI$_1q}m*
z2jR?2$g`S;coWgCEjchb|K|n02^SQ?3&|W47PFY`|5eWabdYp3IxnPX5<G0j^}EDU
zz?}$58+MYEK&Ubr+_<P{4uZ;TOe-Uy?Nqc#w;*et5wa0Kdk4-KjG_DtPxnNiMrw?{
z!qivXjJ{W;E$+64Y0^?wN55H#zNE+f>E&#eDUaiu;0wlq=wS1Q`KroXirVOuY)6Ep
zEVG1dYRn~DQ}g@z1-YVh?-9DI{vY<<JRHk)?E@|pQamUm(?cX<B+5J!A(D9rnWvCh
zMCO??geXmj%=0{#F_L+nWu9lg^KPy7+WXymulN1u`{z5p<FGpP<XE2PzOVbb&hvNv
z25|}M4nsr3&7VJ#6&uCrBNOZ!H(_4+0;CqxW!>eYu?C^pWk{6d_Fa)A6S0j;+8?-9
z+%GUm(9Hs!E}4)4Gjo+%+@o`fRf1ls1IOg(`^9~@1(`o|=`3!4QpxCSoA|`GR%4?b
zdY@}et$#P`6)3@El`YOkIg7_s;+L|_+h`l91KM<%>9k?_=h!O~!5#Y1%&Nmo=KX`O
zBMm;wVzz(cXJ$(Kg0T9?mdR9^KKRQ#Nba~R5u+IeMg7zdb}(j9Q53nOta2g5=;99l
zu}F(4(@ub7=z5aOKx4w^4)P2)GBC&&{3NRx#P(xRfTAeFHuH5GQ)!@Nv25w-&7P^J
z5^rrZJnrzRSGB<ifSp`XNm)7?l!paRDa6lQ)P>&qLX@_|TQ6Th;tAKCm9gaJ$dB63
zv1Iv4mldOYGvv5BQ(h;Ty}LiJIvRZBX`ANF&NTWl3T+`Z#nmxemfS&jeRxD~dnZha
za>hn&FKKb<iXs6|W=fJzaL47!OZBfZZOf&q>C&zD)sa2l+2pZ-qM}-&+uCh}wIHN9
zPzggBC=Fk=|6@1*CtTvM#dfy<A?vYP3;r&l^$DM*V`XCq)tNn}c}@w&7_*(F3`xSt
zUXxf6^>T<5+$4{Rn3#rQ>NyQwqD{laTkIiB7o4uMC{z<nF)Bo(f@su6v0Xz(>48do
znb7}M_#xNN*x8);?uvS%1DZ6GjjP1G<%>l2d@-?i{YSA0Z@=CC3jH|^;wRKiD>7JP
zk4oSQ)E&)7(Q9+0fYFp^&kg|zvJLchof0~(#O$0>IK0o&?@pn2nsykvF%tvnn2czo
z4m~)FCbwM#yD<7#o*SJS9YEmH(ctnjXgm5OYZD}yi+BD2*lg7Eg|j2O3|~H@%*wCA
zF7EY??UE%Y+Pl!L@p&PCFWW=6`iRcR2LZ)rm&;_?ip4*dt;yU1{fI=&biUv_!a?zv
z$^2p*PQ!r$%PF|^+p2tzgm5|75D&NJDw)~p-={Q=`)zlg!0Wnitcr)<7WsNnGhX`>
z^!t6>{>x!pMRG^jqsZW&pLDe|sOHeq^el@v9hbDEPMrpnnz!I_<~A=&drHI`TGoi4
z<w64vU4|#ior`5Z>O1$}ygkLLnRs)AzbT%+>C;-+vnfr$e2FzILU>6{19YO~MS-{H
zNq_1_;LB(9K@;z^WbyO^bPv-gQ;+p<?+-qsm$##jilGSkV$^J_|F!<L5R3h%h<u8W
z$SCI0u3gb+l-$t4NJx?FhUDVF(TqXmH;zDK1B((56Ji!m4V9~Hkugk5H<{>o)wu7!
zO+H0(OQurU=Ib6^bbK3+9?%BID57x1tdSfY%8(6XX9KG3D^cl|i8!C09q(lwUXGcq
zKM2SUFdWIYc=RSUeAUO$@IGV9p0p#|itdZ^%?MMm!#}VMD3%uT9XJKujX*}4NdV7u
z{OBo(k)WK*&P1{UlE0ppzLNdhUA3*uKaJ4_Twj+ZZKj#KZuGdGx3dO{V6IZ%_kcZk
zXyAjGVF`Ieryhb4qOa`62a`vCtF>+dABLX4ZsLoZo<O*|$usO^(5+-v$3x7HWPj()
z(}bFcqowEyZk2c&^cTOT45699`pO{EBN=t3%=t2U2O<v*RVhPujP&2KU7C3!KI-6{
zEB+F#r31rX&#BsKpgr@J@LK?<4Tq7@kL_#1{XtRPZ&O~BUr-`fV`5Pf?`+o_1;{hL
zc!L2Ecaa~aP)8zpHgF505w}DSNOPgkmigXW(wOt)*K-p@w{xMh@UXqn{f)r}Af*LI
zCtu=ukmI&{Zm6o@imJdY<BZE7fF=|nz89Ic-zdy?7OG^riKmNBqNROHv`XH}*7KHF
zSV;<8YXB(Oh#7O|bZasBR9nMSwo~(yaRXeOdL~coQ`KDCKrj@m7@C58Z6vvDc0k0i
zji*wyL6zbv>(yvTkjinsxUAli1cQt`rg__Oc_0fVfcxJ7RwX1uu<D+dC(_sIvZ3-b
zRI86nky0Bj4khK3?C^NLXLa7VLUf2r=A#|tT4JdX8&EInO6P-L|Jkwz*Kn_-nA3%S
zH*I~ejux)AUN@EP{v)9L!F7f#AcpaUkl(O<Jf|`P^B(z6S7}RyX)eOd=I{hRQ076P
zf=KKVV2$ySYCZClAoDsqF1PdM5+MdaklIo)C3wLt-qbHwEtaX@c?GeRFfNj{`AO{6
z*bVLz(w|QDXIqcuole9n9nmzhI0i^s;kGp7;^eLaIKR*SwU^yG+W!E{*%BCu__SWm
z^MNKVNDfATkpIn(vTfw?G}&h?XV+>7_CyxL<<1~pA41~p;-&i%ym=RBgXcO6+W>Hy
zMuy`d?+2nADIF8u%AVC!FSD<%XXIHE=F<fTFSK&xJ?Oe@=;GgaV9`&WbBCybLOWo8
z46G|Mc!<8N!nGxCI!VZ#WGj9ld~l}7(yLwDp|2k3K)PwK9<ijb9vU{dBStbwDdS=B
zewrTF#5~FY%Ne|9xE64fE-M|c^>ewA><W^TvpIlF8Q$_Ql_x#fQja6U-in=pK%NB(
z=gEj}YOA8L)8^D33|7Zb{$pTax8x3Q)fZH@m+p?bYoGq{j$Jw>d|_}l5}5E={OGSg
zGBkVE@PC%hj)c3TDDm~R@19qEadI;O_@8fOYf8ct_QlVOh#z}d%!ZxzU!&7)!H&hY
zr7xqdN(K>JF<Yz4@S?8B;!I@+r6$BkIA|1Dm|Bj`Ea^OR(Xsz=|KbG8)o1|vV}DSu
zGthF0N=n8-vUj;hJdd^2vc|TPns(Wt3F2IM*gUd4R0eLI+VVDPB~vRN(fOmKj={+X
zx7~O|1(y;qFAk4b6;CsRtS%h&t^63L_K63A&X5!nO|Yd{PaGAEVBejF{h@Ym4D=6i
zirD;^zuA|EG5@f5{f~wGYYSWppT@ak7NfsT1U(FFibqe&z*W!|CZs`<A0xrkHZbsP
zvQT%B<lu8K8`XkQ7jwzGR1<M-MGlq>Zl|Gn+o<HVq?8nTM8pluoYu8()RwGaa2Ms3
zOHd&!^h|kQ0BntG!faQp7FWi<O<C)Eh6Oy)5`b=J+IWe_iUrBZ;`5~p5v416mqw{-
zb4U4Rw(%RmVQ&|chm^~2sw|@f#64cfzk;7lK=9hs!`C5eMgg3u;S42{slikN{rzHw
zf{N8`JA<$QlRJ{cJ+ern#1T(`<cK$j0i6=)ia!zc>w(vDGRD)zw;9TL#36ge`%4~7
zs-Ibq5s(Sj_3gX;GcT^*6OX}&hADTD<oWX}bpo@*Oov~}oY%$c`AcUIO-K};(sm`3
zlgXWt`ELSAYyrj!s2oz1yg)wn?zwy<5<r!l{Q3uGdvjEk5sdqimC{nY=}Q(2`&mu4
zLB2b*Ca~Ue{XG!XlR4vmh|rn8|9pR}rU8EkfZqCET^Li|YJNjl8T|2#bi<V>y)V9j
z7rRNzbu;YyB!SyVIXG!8yIpjg8(y*4j=9g|5!gh>38wh3!FQIcCdTAEFJG9Tcg53N
zAJaQG=Dxp&&~RJm9W?n;&KieFj);%cI!C36Pu-e>DVj!hqRY>#9aBp0!RCT{c@d&@
zcBJIIxRFUImSqr!-_f(;PcZ*VZZ--Ee4Og`xLBrRO2zUR<OQe@lu`Re*n11RsHT$c
zEvFAnc{#AfUKD$A{K|1Sd-b(a*jXxmwQIg*huy3=)3-M)Td*0K8;F{ouD*^838@Bg
zFYdH;NDd>ydv`VO{}|}ST{$Z7rg<<>k~D7@bD1Ct6T%w`n!-%CogMF<4(T;r#<(yf
z{x;NEL8?ELvqITFkGp3|M1sz~sV;z6;w>Nfz3n&THcHAX1>B8An9R!w0rOB|aa;Cg
zy~(b<uI!XY`+%*1k6;SZJkR;A(#7x7?#ujy-IZR`%SDLW-svCb48&tZ3YT$bb?fDC
z2X_>WgbeaFyvz*uJJW^O9`(BUY@cc<3@qmEUOB4eBOW<<L+OF!Na+5P6Lw4$BdMe2
z?L8iD55V*fYmbI#`pSms3*wPWNJLxkt?z7;zmvrN<D>W#D^L<>Tpiyy4gTz^r9+=e
z$`XZmGlYlG_z8=9)?uGKTXVwQ?T5c8L$HW4uFMYMHjpr{DV|y*5vThiRB}FLDu~+l
z25p3Jv=`%t=IZx3o)qt858^hWcw-IUB79K29rli?a}so^y9CAy=8swT32#k!FQ0fQ
zh+{*JZuGv;MUW$3pghc?SO9P~xiOA-_!bR~2!%+ev^PU1VglEHrqQfB-P9|J*-A9w
znbusC_@h@sB4nNPm*rU2l4=E)Kwrt=A>#f0y;TN*FpF^I(|0FL2|7<t-n2$&R=5ho
z2EctL)-LJ97ZEodD9G4U*d~|83%|hS^1_u7f`n=Ixc0c~C^zm~i}H!}t}=x9G5{=K
z8~4p$@Fd5V_A{c%x@Xs?FGvX2JPA+tC7@1RrLa!y-70*4Tvn<pJ>syu=}tnsqsK3D
zLh{mU1>ub#Iy>j)=gZvvhfp;Ue{msyBFE*G`qhS!xQ_e9JPczC3hH-O8S-vSlYhw$
z@20_~71}Tz#J-KML^;h>s4M)7kIIENN3XiJwfou}?f!;cBwSp`_cHG4+0M&*)D4)4
zL2IXQ-M`o&8I2Gsj!eA-FxI&mh#FXy#9q-viNxX)eX=2bHwI5zkfdN9=hKh-^3@dH
z*>5`-<WA@NVI5$v(KK)!aiyu5IAdD1-aK#UCbIEE+T7In@z1W?Tx;4tV<oeqfZWXb
zBT;XH1s8nJ0PiVxr70@KJy;d=YFUVSGqk?rjfL&px9B)*cVxMNt=*&pSvsKLfQ5Zo
z{ll5fk&qmc10bK;R9q@PJKm!D1=EiFomntPOJD4jl_7ji_6BY&%L7b;7e7V{C)}y4
z_`Qs2ta(hkROVfm+`N_($+l-btk&9JF4?&MJyYwvA57WD9~plBO&(%ADSdJ04XMx`
z?HOv?F|<CRySL9^a`Q`#hxC~*NL5aVlf^--<yQu1$~kvW360YRCSIy$oz)R{>@cKw
zCJ+CGk4g@%C3;&tqkd~gFD`BCq1NWWN^2`}Ak4+`g%`z4Q}IyteHGEv_bEZS&zco(
z#S(JjPtbWH8eU)gq-9NvFKefDdD9GZ9MeRSOwuhAtS#MIzOvT{tcK`&btj3%#l+OV
z3DKN=bFS(<m}D!ul{vpX9&@o4Rc6h!n99KXEH?0(Uu4&L-KG$&C*pe%LLFv_{Fn6h
zy<a?K^dH8u3jbJ!8%#qdQnkX+WeMbvbt;*GP+FQXN$Fe^KAPJ|_Nd@&X+d}e<qK7y
zQty#?6TAc4fOZlzS5MIy7l%QE@Ykq_^NZLgNffugv7<lcgKkDJR{&FK2qffK>lE+E
zwA=cYz8jXvWr-#^4&do?UFP9IcQVo{Ps_ye5WZqh@)pDi)rnAuDLa3lcgHnsf4#Om
z1rO^+i{qEZSA~k1ZcR^Zj$kCJe*XTKCuE;S0~o!@2##jGhdD=g!{4IiM9;L*E4eP8
zLpIfKTPcM+XbtpbWgZa+zrp!HyS@f>t!NZ0hliN>wq*X}najugv(bb}CdM1ag(J@G
zyM-Jc)WD5BswFYa_~vNS$x`maY+lVi$DuLnNEahAxO+x_6Qo3^Y=!m$*tRVc*H-Ww
z97~JsX8QR?el0fPVvZTtDS-|14T0sdF#mvX{GrLD>q8U&i?uZ)KA@2GV@+EBT$Sq+
zJ*C09h}Rq|n#devNIgVb3d{^v+J1U*>@T*vO#xPdhKIOs2s`Qru?X=a29jAiT#*?{
zXCdN!J*Ao5a_!z4);AS@QQs78E3Ed(q{us=WI-rgG^NPtCMkyYtw}k1H^Wcgqn+>-
zIkY(}KpI)l#TOmS+Fa}+JK4tj_bJpC&ImKd*|LrNub-=85IZ6JL%&a#<$l1duVK^&
z$W1_800f95@=zT(5#1=(+!w^O%=&ZUJSB9=aeAqmEcX(uMPEn^A}F<{4)@5uleRk+
zs{9<$`FB6?DS3N46F`xw;La2C{l0&j17vmfs<jvgENaycDmSWrcCj>_0V5^zh^f2Q
z(rMtElgPR#S8XjjkDJc^ceJ8#;cSFP%B?s3&wQ@C*G{rL!OKp8;0UkxNzF`p7uz6K
z<m#dggmlGlTlT-BWX`8#d2`ctQd%82rfDWfWGqmvSo4$2F=fMOlfd?>e=Vc<^UUr7
z7L8zrb}6?#VgG5xec+}V6QYx#I_U$%OKC>d2D7;d`f^#+5XXR0&2IFQdz8An-PW?j
z_=@}A*voK59a5!uH=|<){sNx<BN~-;6>+k<nn%<u{V#kWL-A4Y`G^6rR7bP{>su+S
z?l4a>pu~rL`00=1j(QEc%*){2f{++KwAuV^5C0FE-~rt4-GE0){}WCpqJ3&83vI-!
zrgOHsz!iKC>Y1ANckZbb{XK{2j}L?rg-QX3jEdnmVEf-_i~sp`&$Y|oX2U<X-uJ&;
zssDKS;Kl#{_y50q{Uy*VFg5xLx+PYKg1xV}X8hsrZq;291YbR6DENC(auZ`92|Nd&
z{W8r5Uz`tJpSAfG)n2@UhW&oU;g%*s3Z4QQWDHnBN!JHnFpRe>h=2ty4OUWbA2ymd
z`&R)bk<q$Rg!ohfb1-mi62e^0><vN1$r#0SpQ-7(7pHcmd%4@@57pFrD9pQ!FcRao
zXR*M1;xbfXzlF^WsOq^=uGI;Mwhnr2Nzr(FiqbLo1Lt@HiR=R?0(10gPp9fu9+5P^
ze!-NuQA)OcYMQ0R6$lz5BK|e-SE~d^x2rY-gH2u>#fZs<TiEK$cD^(e{?sPh5lAqx
zmT&M#>{guQjxAl02a<6KFx;9eWR*Xuw6=aWNpd44J;!?ezdj<KtkF=Yuoldz+x%H?
z`^%4l8E2Fa9_a@wB=+_l1hd*g+{e36iN2~splrkdh4UAEW9Q^Wp|Wv5wJ*w``iq1h
z7=D@PvVve7#Aq$YqK^|X=|U37ZNW&3Wa=xz<PT!6P(xJaFimfEI_)lg49;5jyq;}i
z_*?oMg$qHrcky(b`GWfwHk3AEItZxMnjmd{NClqa1<q<9mLPmKPV;USB<6FWqRU4D
z<Ym0D&C7LGza0<lpe5zua`ms9)jS_0N+`j5QvJfj$??Z%)1O_boOaW9ih;^$2p%tt
zm+^+5(|BGUsSn>*`K)yn0Vv7<OjrY|O>Ew|h%^9y5gi@EZ#lEXhpQ5=zKc@^^lV#%
zapV@ZYdc;S(3vcOH*){$LUd2*&TrSjI*Bf8VCmgr-URMp6KLQN8Ww}dICU6g2H%BX
z7K=F$njT_CCcLaGdSK^F+9+J5`OgRUKeo(2z9C`&(v9BhmfvJeZexsaejVVmD_^(*
zW^u^j0+#v6t%(H(s3C%Y6N+sPU^7&$WEQRn4rQ`)5Ff4OJQbywGno%NJ7V|%d+gVL
zW(CtlDuWvVR?ZMe>(4Xxlml94`{TZ<dS&%`hX)FmNJ%G#O&f8VJ#_(ZbN>6sFhqL+
zb1ySzBnTQHw#uplnoSXR(OYOb3ot8M1M16dkgYyMV4dyr6EqF2;7TjEAa-QaXu{DH
ztB@GZNplPbG*?oyvm6NqnG^z7R`deC#-i$*RHGHLdZGhw;@rfiGU5i_cN*f`gPlr|
z%i~}(Su>oEQ09<G!!0V`I@}I|8JbWL6}D#ScbXcmBpM@OLPxz1dE7+5CGc1$QUH6~
z$B5KL*SR7H*wCuPi$K;hVW!^VATX4TB}KdxYs{bo2>A+I06V1K6fUzu=bq8(7k(5N
z#U0~llf83I_s_@TKb#?+v#{St7&rVC{dXjLLjRPzI4jt+Jj{mfUwv!HQMVW{q{Z+1
z2F$d{dX?wJiMX9^{$dwz(BRPA{s@Sb82-)C6Z<xTEXRPQv|qj^%_t3l90mUFsQctx
zCJ%&GNQv@Wz^JNtl2~F1>?E$OngZ?@0p&pV5t7;wbY3~yE2;9_c(9^}j$tuD_8Mfd
z#6asBJ!QbtY%KB*YB9-${BRk4H?#}=<UC@rarq&8@F5ZeYXvSvuBzviGFzt-Z%(*D
z8U)y7D=I1QEpk0SDAS@*qV>9ZZ;nU}oT>7_WW>Y^xuoa~be!FbZN`dwI*&UfkjT}G
z+!ks-T`y0w`ycy@hd7@lJZxgBzxdYhzNDbv|2+EtTIjFyka}uyOK<Ma*IvKFt=p|8
zl5M7Z{7r=9<Y$0n=(QC{*b;=Zv4o`QfAf(qDmvNH58559Ha3GgViJ-lUQ`13%W!ZZ
z=b;{$$tYqHqQ5X>lF%|trWtgt+ccPT8Phup(DJaU=XiVRV%!x1+j69+f;1e)(+zcC
z1!>6H3r=OkjOt}9Su8u?IcVI$G{hLcYCHINIrMtl05~>|Pb0rXj<kwIXHEZ%`_v&@
z=pNM}A})cQ4w#vhv2}$YWB#xi3i&8_6@?F8&I4cM!z<=#`2-G)7F!PMzgbXnNY^O$
zF+)x1&mRg0oce3>^D*o@EH4WnhEw*vOw>Ej3>W+d1~X1_JivRcb?;Il+LDhR1W;>b
zKPFG)gK<_6iT9tx<p@Q5mI2Kcz&5Hu@V4qVtM*{RlAudSv+J_7MNy9&l>ej$Xh4dp
zZ@x`>9VX1%+|6Ev@dRUA?@|OYB%fLAzOhFCJhz6!6b5?1QD0j$wYW|l!Tg_`<r4XS
z;VgIH{o*WdVG2fnug@v*U%PNa;=+N~S{M_9)A}PHkqx{&Z8${BKCTak@=C6^72%M&
z%=m$!nJyHU4B;GJM0<35CI#NN);#+72%Pu=e}+-N+y1aITq?{+dmj}=Q{MAUmS%$_
zQ2HFo#gEbZu5#5k&`dr?iNz$-;9}vNROa<i9z5Qq8fqSX;<{YASnf!@%#);0RGOQc
zdwt~A^<3($?Y7&C5yemM(&t+)CBcZ5c^KbgAyJoT>yE#Y`&R?Xw^pGSFMa0dSqr$;
zTDAka6@$=qw>p!S0C2IqHGOR}JXW`5pEmP-->p}l4>yu_)l;d{H?KkDp_|io&4>N%
z<s@tUg_HT|g?Z1f{mi3YrC(hB`WQTA$Q){TdtYG)HE>aS?C4`I)b^gFxM<#8PU|Y|
zY@?vCRG(`3Einn`TawblW3_%CTbU2LIooA?F7bcUIk-_Ixjmj<6(*GMz)k&nmV(9e
zO<xY9!&IKl4IQmz!N8dwFrK(o=L(|=Ex#qsZPa1yG-5^ZNFYhXx%*Mmc2xPzcHtRX
z8jSz;wJ0KtQMpBioe#d}!pIjd{YqiwA*;9zkSMcY)1t#Vjl(ycU?*RTNu`e`&H|7(
zOL`DNqgzV(nC~T8{ywld5?%4S=987ox0rX&Ng!1!PchSJ7KADVAceewMTqR=LjY12
z%zcP=K&`Ke#AKP}`<I;I8@O%6m-X=M+%Y_xMw#VXaiE8FMSm8{s%<ppHzZ7q4qKPN
zZFd+{5G~4sRC5WkQgGJmX$Zb(0PpK9I6G^i-TKTvWt<H&@t0r2<!D;=INmqGo_HpW
z&v9bxx$_Z(qL8Am!F+uMwLQ74aLOI9=qEVhR8KY(_rXAmNdG-lA<T(EJmnf#;(aA^
z$L#K_czodAvj|98n@*b3otuISK{Mao%XT}i2#)tKb1s;AQcpYrpBwUZ@9;{eYZl^@
zs>>7^IE9zkkh%ihAbxihkO!Yw=OESsMmNqa5$PrwT*5(2Hd=SEND`yr)vNJVyZ%z4
zmtK#&u+QjHJWK@eJ_%R(-1N48zC}|~y4J)jMi%k}a%t?h!1lto$RdC~+W%2Fj#4fL
zx%UI7b~?9pFun{f4nT!$M&D{nxV%AcchQC#gTmsb^(UbD%HvSJ{N9T>Pbx~hS3RhI
zc!k1V=PvidFX6H6(CgVaEz>>Z^6GPogVB<Hc}8CM!{%j3JK_c`4>9hlsWsZtaHQV$
zt_sWFLilaLLIplJ=QRze>t$(mq;P$L7pw9a^J{hZAJ2YMyJR{7r{6{HHk|~8SFO2j
zyDshdq#knVFUxOU*jbgYawT@f#xU)suprLnTlR%b)6LI*e4gYyHL`v)tE?x#L;yfN
zzn=E#vqo=fEk0ih8<fu|)0v*ugL6-8*xRmuE5`AE`d_|;T~XmxnF0+W8K{$l$d_jt
z7Ah7%_+GZ6Q7@D&3(SoQYR)j%HGYL@<S^1<<EW|VKCzz=vlbk4o&+X0eh1k&p`Z&}
z@nj^L$@y@zXAx>y0?&)d-rxq3nO0b6F|--zw~{U@grXhS^DUrzEYcM%heNt6tR2>S
z&DA4+!a)z$d@uIQ9*O$Hh!qqPr-nKh<CunCCLb)eO0)P7&C3vwifzs*Qodm#96`(y
z&<C1ksqe?%3I-F(zp`z*I=0B9dC}kzKo?~C1l~fyrxvT?scY!4-uSDco*hl|RAwXH
zhN~+D-O5`ki#4d$@go@ZF>a2)1HsYYM`0_S_erIiVXLGH1J7Gdf51P6SJvOiMCj1>
z5-!I)gj`v;tz^>IS#MnwU|kV=EnNED8^`Y1Ic{kM^P>#M&~7ao((~`aj1z2p#{$ml
zF?hwu&<Zx#w+n2i_nxtc-b9U2Y(2B2;gm?P=Jl?M{`L;%#~PX<bpxvCv&&(<7m|6G
z3_>LMT_FX^dvqBXyY5{IRP#*c)()+rgMPx%`?197eK@BG6A(wsf^H3SpI1=W^c8^N
zqa@#FkVA5IpfO$n8X<%g=G1PmnCLj{Y+&_`pw0;A0gAe_F@3x8B<5*xp%s*j^y6u#
z4lVt1iJOVKenF!j#xG*k&jB&1htgR9y;94jzTR7__LfXlJUKNycJoY_S9<G<=8q12
zLAVNWzLWJN*_UeKrtcZAgO^#N-D17&cQXfe`BUy?x@a<^3`i2tO+SYdsSY=TpJ9DT
z)!fXlyhz8Bn#y(e8P?W`t;%GmfJplwo$(>$*qwdzSn(xSCi_yo>BuJYf8AnfzZqc-
z=6Xl|Ud6)k#7-ccBHreVEP6n1lYQH-d#`bzPHc`I7!yW*<`0Xme5xxiy1Z0<b(Nr+
zvZ-kFvW>M$fS=nPa5f*WJH<`ebl$M2F?I3_T^LJAmuaGFp3Ff$<Lce8SJ9S+o(_6*
z7|C_DVGL}i=e`NV?no`x34F6JQh6~(olU$w8CCR@(I!(dz;#K6pI`U9{jzpI5H9g{
zma4-iyqHMXiMeeQ9()CBz=)zN?{ntaDx?Q5-5swfC%Zy=?uE93g8RV)^%XV>e_J*x
z>D-B)QyREE+7F|OP^k+V`NX934c7%GAB|Jxrd?pwwu`twKKwlra=waRNR4E>ZeytC
z%6;T1-L5iuhR5ESUgB{PkCd?M2{ljAcAb-;9aaU~tVK$q(X@#efklTL8;i`~fbzY3
zy-K#0`McN!?Iou)w_o>-xd_(!JXG8Qq$+eb2ZVK3*gi8~s9n;wh<#vfTC>G7_^7r{
z;5I|A(r{?@z3m;6dEP}v4Lfa<_yw5u1-K3e-+GmiG*k42#l};=W%qrI4zG=f?A}g}
z?ly{i(Xrjua5p5OXqM-vZKhAi;P8q7o0I(DC;dJlQU)o_!CF0a!Ku#8%rR`0qKMrV
zH*Z(>7P2J7p0^uqk;#}pp|gP2HV9!cvsRrpWj|scQdy`!k|O6I9TYPe7d_6&_y3qR
zT!eGWreoQIT4TavK#%>iBlT2yL-~-(;uve;AW<23_xYqv74`OPkISD-{coEeye?cw
zm}TV}Np<=6())JpBtd?3(RS`6xNV<Bi|MEd*Zb#zxTS0C;tTtBeq66$$Q*PmN;IO8
zf5fhYtIvT&od3?KPE3G)#at@de{S%e{DEv^ckMVml~-=Go<E1!iJytiMB5bizY8pF
zvI{;cZEQ%JB0PmYTLk+fUNd@B?6Uu(_l(6cDNgb_57d;oTi>!IU)c_gGP-~0ljPTM
zhhNLyi;*dwpGD7Yft{$qsCg>PRb$k#&oG95$%fTt`2ojgbrJQ7j2~_BFW>(7x%ksO
z1GQr~;9rF=lHmF-*ka-2rS)qeefmC)sa5Ue>prj0-198Oj%}y{l~2?W0j>nW4)4|l
zHQ9RzGhnit+P8q;xTJmk=W5n8<>!bF8Bc1jhghXlZ9F#ZPkq|W?HLSas0t&&J27K{
zI@ze<TP=Ca)V(%=?t+qwib;Ms1d1PAjqjk0rk=aX;=Q?%yz6GxQ(p8D%e_D7*c_eF
z8+6Z;+9Rc!SM>$|3&%2S675efdhlIs$}xnt^<#^8I%Tv2-G-m5@u99cMDOkC3|^$x
zCjH}R`cHq(3Syk;m_IgFz5QqH)&J_m*a>ZE!==7fP^?{>hY=#7N<{ca@0iEd^sBNr
z@2}+sTr>23{5b4BVZ#kqR*ySmJft!by)y26vaR072Sdy5r0BDA#BZ2One*{vNi$+y
z<2XfEOA1qQu1Mhs)I3}Z%X$SVx6g}>%jG+YVtDPW9jA3%EMO$Cl*fNci-mk*Az5MK
zNam{q%XEz*AMDn6wL^ryk=NXpLb9^X)ZFhZz+*4F_VX&si)O^-a1m;qd@$eVC7Wb(
zyh;Cpc}!Nj>VfxgGlM8;X9H6HP0Md-<WD5xcgPM5mfrP#pluj+-{OrANm}<{U{Bo3
zw-=ZfeeuA5M>gHjJJyPe&@Ap-Obh!oxkJs~?65pt9@|)gd3G}(7F^x(HG(CMEozQ@
z=HEg3Wyt0dvTC<#SwHSg(;%sRV8}Mj%-l#pAIyVR9x65K%Rj*7ZquTh<eiu1CS>X@
zKNS~bZ`sUR*rH=L8+)zN`x?W>HfJ--_q&vvgLS>dvF(EsnthhRGf=3Kj#aNeabORc
zu1UyQ%ebj0$M7)6HHJ$xtn|^a)&kjaZ%{V-LYq%MB~{O7R}OCJ*f9&~75tE8ud|Ew
z$3Fdm<u}iDATj9uyz}YzZfcZa5=E9pI|k-?IrIk2pX->X3ThQlY;;M|CGB({g61>(
zLslaj2=h!G>PITdNL9U)@%~%64zCZ%L&a_-^%*<xkF7!201IVW#KUPvAkyGxvmHl5
z>vf7mt%{%Dz%5<eVieuqMD+;zgWeDER%TSgJ^We^k!yn?BJPf(K4`9dWb3#7#}4q8
zSfG_igd&-;j*qIT1Ib_z!}6}eena3@3aJ(3v{f2{4flq{mS6g%FI<z1;^gjq@GNB?
z>_IY=-+gfNmgo)fyUNNECX701#wwbYhe04^PKEyiK}PHdOpwn8#6I!k!T~Ej<eXK%
zbzJ%~CCH7Skz6z(2jhZ#zsRWS8&i8jp6gQcE_Z|Qc$injcI+WAr4Q`KNY`IEPbJQ@
zlneL@6GZd_ZBhBT_6H0hZ_TCFoDYe#8j_EH>g=RN8_P-gt6W~GH4@0ncTQPKvwyS-
zM4U6(m-KIX8})UWc~aBrZ!?+#%{%FE#ZRaO!oWh+rBG%@#C97gZ$oo{ivGBFTj``*
zf!n80bz@LFQjJ@5FTjTs*isQM656RIXb~8H=}I}jQsgic<SH|YO|w{U^vBL%-Vbi}
zL)BYfZcfJ<fLDXwjq-c%F67lBrEZvz<V^^2YAIJz)yf<a%PfwWM2VMv9Ly`#-W$-|
z9%u7OZj$)bfc?>i{f|F?($aY9Y8NT1{kc=PM4!;7_r5^Y-!|+r&x(+QoEe2{@h*)M
zi!1vlpmN^fW?E4duwuEpF=o(@v6_Mz+G3u`b!rw@rBSIymq>c*F+lJJQ1R%~BIn1)
zoWx?E^)83-3|}KCy*M#&A43Q$$^QZBG57Nl(NVvT9A(>MscV2KruFlCE4~mL<$}nU
z3(*pcU$s^rRr~UkX+&bVAHSkgSBWY^8#d?tEV}4PZH^xP5i3Sj(9hj(@^&T9SM^3+
zKG5t%(&L@Jrb{^W`&YHs%KClzbbVjqp;zxa=}|x9s|zIipKqWYB<{w`Wt`tZht~~U
z!f0}tzE>cS!MpYCD-@^gKF)fspAr>QsVrte@)}%D>WL0M2JzENzd{1f5hQJfinDBC
z3%NN;ziv+Ja=kVjZ+{eW$yQqgWtHQmxs1+aBP)?D`zkspZRLs!&aGwa*QIh&`L@Rw
z?Ce>7AE^J{H-D?eI`AN23iy32=iNF<>P$}+nw1sBcJ?;^`G@2B;dKJI^h>L$>4Ok;
zA6*plrp|}Oa5v+oP4%5|y%F*>FhW#x%f7=eZY{R`k*C;+c^3>YI={<%xW>^h+OH~%
zQzb?M*sE-j>6#_0m8G)kvgPG*6Or@zTc~_&1j3`;xK0bvcJL3=KC@H&j~|bZU1KKE
zvml#sS!nm+v<Y(+{-kE@Qtfjiz%NaHaN^+_G&&?l>jZ}Zt~#>(=k;AQKUT7p1adq1
z_sBP^FjbLJ1O67p<`}1&X-Rs&ca~zRucdW^+F*SAbF$Rq+f-X~<H5c^9nkm3G5jAw
z;eK1|;J%uafWw?<PSz<>e3I__FF;)A%E0C=tA(TT_|_UfzvFn~QL<I{s+(<ueL8W}
zUO2n%T!O&UTeJGFWF*Ymj1S+jZQqP5%F3ADFG+swHx^vL)%>x$faA)nU5A<?w!xxK
zClSHN1Shq1{2vALIt#Sx522x+{Wk1!{ik+j9_ew+Bj&AfAm{N<dmXO|*Xw<nc}1J?
zj&f_<WgB4WCtTeg1X2kOZx=}UHiho^Zztv8-IxEFznGlnGOwf6MLPA58~E?t<;K~Q
zNBTOy&t&n}t+)x$qJ0JEPd7hmDT7IH#RaDz$JU>aMAIU>PN0B($sUT|jsRD#RK^Ei
zVyriw(QoF#V<TKu``P!=DZRITfvr5*wayB;?^<!#1iZEU3f^C&(qQX@<x!$M>Ngtd
zP=sBXR>Qi`SbOS<|GhvZRh$EquH20u%ILxMtOSWL)+vs}rEC@(_UQLz$<M&J0KN^t
zwJ^U8-JE${LD1fhS&AoMs#hzBvRiEz7^C;6C}BohAw4F!%V0G&bwZKqhrn?`i?~fM
z_aC(Umg1v+U%r($(IdFgxSk!~1}0zGY;sWYxxBa1T+=$24e$}Q`m4{DOE0tZJrLTf
zQ(Gp;UGW!Fc%E%WZ_$=qHM1+<ym7Ch^ptxGuD3|KUlb8-uVJ)0!%OJh_|AK(@-5Xv
zC3ho72dd!AfINNnB_>Z@zJ>bDGY44`Tjj8)^B@9uz80}=_hEJi2!_G$3FIWtUQBjz
zIVd<&_}dHM|8l!<4_tgp6Pf;eYV8TtU=Z>gR17y)zhzJgs?jnE3v&VhvXb@~juZQo
z0kLTM>n+#BoQT)IRp9D}TZlS6^{c8pm|_p`d&{!X;_Va~_P)=<EsGztw9RWpI$yt&
zw%=Z^1ogKV<G3D`tn*yH%(n4B%PIgQ>ybF7iO$ovkztKn1pbUb>l!q&VPVaX!(-tk
zm1Uv3`e7S}Cw?o~VRlX{&u&L#1#;YR9;?nYp-97}%6@-p=oxh)^~LS7@X+j<+E~8|
zImC?SAxvgi4PJ&$f&E9UlmD2BYytcE4?|s^G_#-?pL^hFcwt&zsokhoN-^1LhZQQQ
zF%c9=-qj=%;vN)F2flZf^L{%pAImW+H*U?vt^|EC9<B0xZ$q1D!V;7?;%`a%<9kX9
z&7x2XN(>T}%2M{&nrjmUydV3gt~s>uL+r9ruVsW^{e})fj0QvSGlBWR?+QY%(qWvS
zM*y;aonVCuy3{Avjc^`~R`JGD>V~d!Cw12zQu*?|gk?bFeaC-6iQqXXxPy2JKSTB_
zg`#7`qW%cd;*P)35&NPwSs%sytsNYnuRzBphV)<j&0|wU*?PkRl#3sAx-q*w+lclt
zS-{Hon|cRZrM%`!SM2gVRRJW$KCB`MfU{~Ap{-u}i84SvE>oQ;%euIjLu$NDP<ITJ
z3{&iZw)WB*{6qHKh)o&SRgbY7rKH!R0StW9I}LY8w_tkJ`>}ORGy6QDxn#yZYJ2+^
zCKbk=i@kz<7LZ8mzP&c`&?~w}jH6&0_QvF6D&YWIT3+E&y~Gp6#NXZg?}Om4eMGn&
z6&u{RCq+*G3zR}|DQ<?WezQ+~P&+@EPtOf4c4X4j51?;3b!nQ~s#Rd=B_4qVaMq{M
za;*Y6FY#)dTxuPb$Cz5{J*Guj!H%%DMFL0B&!aAk7j|+%JH@bK<EwTeL4XN4yytZa
zb+zIoReG^@^U=wm;-Y2<nm*1skRVa7BLNP14!`^$UV^jqk1vj6vBjg86D2%y0^Ex}
zzO_&pTa{tXc!>BK%=JToV+l%qr@6A(*B{k883;_xsrK61_}5#=2uRmo;w^Z!gX_@(
zoVZ-aCe}*=jaDUd4C`TQ-4LQ1uOE0dSvI(qSb<;fC_=F9A~R|KOV`7o^$j5^rp`gW
z=fym^T%Q)Up*XrhFkqb9#`~#QO61v<?FYncdmIGY#%wJ?u3K=u`R3j|m)Wj_@$0mH
zG?^IbmA;F{33|IWA4AwoxJbi#sazNI*v<#LuavbBSs)ggRO8)R-iB<GS4(|tJf>e7
zCl4cJ?1B+@*L8x~4(#JBU8ecr0{x{bN0?0GZ%I+wD}P*kv}{`a{nW$ikIH&JnA@Qj
z3u2$-Z+#2$E${H-nrh^1Z9nGLAK2x+>A3oF5p07Ty48&@Xn&3^gl#w0^-a)xFRGv_
z?z{FJg76{Ce*VWM!Ow#D$2)o%3L5Q6b7O;MS_C+vZa})*QVswE9o*n(ueLp%<l%ml
z@|BE^)Sk?sIic2Ft(1Gj&2BxbZ(6iONdU>#D<<P@TP~Www*wfu<v=T~K&4oXry<d}
zB({4nSM%DtLx|awX~EuRGNtN<(n`Hl?othJfMqV#H<f83%Ljr*cdty3{!rDmOdnY?
z>^Oq@_FZB0VaeTV-K1N%q)9JbNrN|^9oE)+N>bl)ST*Q-(ns7sO2ucuUNjZ0nGS;e
zzUCo!e%r|}YL#Tnq;DK{*FmhYly$Ws_(|+hDb9_CbES{lA8@qyeo_$JeQ1YP^y3B~
zxaFn`+4rog<W3+X5f_S;?VF@ciOhJo9u#^>9ZqTwxtxyj*c(qBuM>z-PkiA+EuZOo
zk|YqDA8H?F|9o3w>lCS)`_NM+DS>j*Tz803iFN=Ig0iXME`7IJw}-`HPBUB3wjz8V
z3>%O@clKis=E~d&b-IF~?C9ZAQqoNc=fCNe1jL|QN*w$A@preRjWwtlS$u^{%%aXu
z)|=6P2lcdg`vc0a=t==97_EF&1_$-IXgm;=Ml&pGQYWQDAVjiByATJv&AIIJqS^2S
zlqsJZXWU*RPR>*^ih35p-TwvDY)$y|$cWZX!HjZk{o;~#^Ummyx@e0ZO`E_Kp9eli
z=<d{LmpFRM2?PEFLKM1x0+U;xCeoGmEE>UvNev{1!99Tj<CAhxz$1JcLzO#{#crb4
z9s>mHoyF)8Vz=ZN=75$F!SQ>IdFfG@t{$oXQ`90_S%e}+4yNj7vW2<ZYgOa?{6ks9
z-<YRoOZFOToK_?PfFHb=Thn%e7*EDezb$h$I;+`AVw6BAWcb`w3~x+E#oI?N0S;eE
zHMDdKUTu#359qoc+C!|5gOhxLeCWEt%gV$=9{qv2^1cyCEcccxSXIF%xjj+IrS)3D
z`zNs}9{6t#K7Un={T<cV!643>H;=%-!K~|7;<L=V$?LNHyHVIm)aI1i=twc;#<=J<
zw0P{`a|ZET9AjQt^>}usARJOnQJMfsy2v}HZuU~Id(L!L(0WqaSAgN10}Oo!fK)I+
z*rGU1WEZ_ZON=3i^<ARI%u1B23~J6eSHWOzus6M@_AHif$&$0~^=i1+F%8v2`So@I
z-}@HU)fw86r!-?}jDA86Fw=#6x|BMHL0MyeyuSF01l=YLQQ0Y6iLmrsQ}hhN-c$Y{
zT0YT&K4N^G+x5leeff@gTF6rUK}X%`OBN`Pg@|t7A+)6ZazVsYqJTlUY!5G|`Us(O
zcchC8yOsE;RHa(xuXeXRb@ix<T9RhRAY-!pspF`I5?z6|vbFV4P|#j7|L;@b-ytE-
zL^QN{?T>oy{n-k>X2bC3M6sgrtMRC*?RTZadT0gRU2yVOq$4j(-FK<lS)KSbu*m)j
zp5>wkUW{1mE~(gs$*ux!Z=D&zm78)cTE}HBg~#J$_PO!AwoXwzR%OMM$a2y!EvmI^
zWqRoDYFVw}@0tbtjf+*04o)dWsGhM++{}RBlV057vi{=y!N$SY4OfYxE@ZDnkgqBl
zW1L$sfce*FMy@wjM0ZwvB{j1Mu*8B--XUD4IcmAG5VqEr*QF%uBk;kwPcXc(_giA}
zUZ-nLrGM56)g+puRcy(S-6!@myMG@RPu;yfEV?>roHscE6|G#~@AReTOt_qK_ctP>
zptVSkUNcZnwLq?%DqS*$1s);R+()59Qzy5d7CILMO8fY+t~y;dJU~^A+Y^+EClrha
zGB;S~8`Gp+0+ZnEHgvl9O3f3e&&gI6f|Tsu9sYxz-X9hDl^vqhcvjm~Wou54#a?h9
zn+x%1nX_=O5L|lC?ti#)BqcCcB3Ef<in)FO2JdIt)BN{)`9=dsMW@A7$TT<Hu+(uX
zP^TOj-nES!%iS`je%L~|KlU<WlW2@FS+?Osrk-=P>H7*hYlX#nCt-=A`>d;i)h^mW
zGi{RH%!~Dub3LV`xolZFbWQTa`$qss$0s%nH<HIa(XEo~%QTl0C?@^h?IN)A{!o9>
zc<7<E5y693wa+-0Z{~lHv?xtk8ZLt!PwMY>ydeeH@s^A~U-@&#`&FQH37_z!(Pt6W
zU-P62^BRg$@}wRnS>1T{I*?!^KajZe)2rPXn(-<F#W+;nSaC)os$6Klubp;d@$B^$
z)R-pj`_wKMjNl827^_@SE-My?_cZR&*%6}0pn|9NO4g{G%y7;zBRoksYl;D%fNo`f
zjKG6T)Z`|OxDmV6_Pg?;At~ua;(uT9f4t&3ECug*n8(O0`>(QU<06K4Bi+I>3_EK|
z^&IUJP#yb=g}VI2VbV_7HQ&jHIR|cysYg7@O)X@oxkmIe`dWQHLz#ok2-HYaV}|DC
z3R7luPkV%GBxp72CkDW#)1cOmAaik=ipd`4^1YysM%iUYqXq=J?g*%0iN6lQ+33;<
zZoIRVi}W!rz8~ixqAx$lrLV*+pR_Ray&2Jed;cd1ci2zj9z4YT=Yz;n_UX4moS@VA
z6eiy8OP>)J;L?>Eyk;sM2<_7Z$z1bV|KcQ!PjOdk!4yoxOaqd=E#V`wzjAcQ0$Atb
zCP*3t2rKe)!Sw#siQ!$G+E)w>r!${`gO!7mcNr8I7TNVAvh0OZ>}WwyJy|+Arjy7N
zv0loQg!Yz8?)As5aU^Z-2r}1R^b3u+t@i<2O^4UI2XnE0k(bjf=C7;$SNZyDy=>@U
z5bbOZXWwi77f|Z;6c*yTwGV&=EroeA^YXI`v}Y-A**;92aay74TCc%lw}4&w^H_w;
z@BTRlE05IJboMuJ^*?Z}Uw_3A>if4U`Z1v=XwH9tvD%T+qW?Y)Uokx1{a|yU+SKFu
zU*GmWe~R;jFdX&|T>DSI@lO=(@BZnvAFlLhPc7!_#-M+DbN~En{_FR`gs>z6Qy3Ng
z55Md`0e>#BPqEwu_)oZ${hvPWzrT<FPhJ3Fb6htt8H>Ig(0X$sI3q(Xe^UeF_+Un8
z5!eNJfc5r&h%-ym86db(W{>rdIKv<{?(gpMuM1xi4Od#bSvCJ3ciIgTmxjs}o<+-o
zKBHXVt$d~z5Pz`+NXm<!aTYmWEkGTYXZqpghZoy&0#1M?FCHH)A2*fofXs0K6q6d!
zs`Fog0b>Qa7Y_?)yz_@UBW-bf8RQel#%uCwjP}s+Ca>i9H@og_0iUfGl9a4IrRT`-
z@@pRrn8m-Y1|CPKqIKMdR}TGZ%$@7jKiSRefN0Y+P5ML|tldI3bB4R_af|RlulEGZ
zN86j?22F&Tm{M2Y{a`V7OAq`Tc<Yaa!X*tm?&<0|?mw%MSt|^;@MCzc$w>t49bE@@
zom+DbdMb|$HLZ%@gHMip0g&DL3oAA=K9ss5pgbt0ji4ZZ;(z}`tp1rxoDcoMU^8>B
zWT4#ZZ76ROXl6Cd7-~p+sQ_DiU*>|p(zyk#NgqP0bFnfnSpbVjtNrE5N<Hd#M*F|6
z#<iGGfh{$@sTcPaAmz*2CEYyZ^$+f9J7i)I58=$W4@!|e0H8)q^XD#>#R#VZAGkfS
z1CPC&eN`L$s9aL&k`voDU&wuvTzJ$H%h54tw6Ca;^gSoT=w<gO-Tz*A{do+1M#`_A
zNB1-S+{eG!V)&m#{F6|rd1Y&YDtfsrgrTg!P}vIeIH(1oqiH;3A;dB40GX(B3k=&Y
zU3(z>m3b7pCi9WQCCB*Br9;d`Frr=rOp^7+I^5<0h=T1y!l;`*%{={9=5K_ubue}T
zRSE(C(6?)~Zi#Vs@WwyTLhZbI+<AMb#J8ZIA!`P}M<MSjxMuFLJTuR*36|vVEk_uS
zQLe+i$A80j4o1eL3xHAt>~{NK`H`4|+wTsJKv7ukty|Cbe#&JSsV}eJ*ZJReqSu13
z>;Gt!egC`sY{a2CaDm49Y&)M!E1}6L**aC^4EFbN2U|#<^R%1;LCP(g+m*-V?XIQ=
zZH^TNU_Ibki%|*8kU?aEbV<PFF^NqfHO~M+UtNz_H;Zf;8ew-7z$wuW8K!IY^+Rr5
z7D6U$UV{AoI<T~d2KTmH2I*t8^s%X&9>D*~Av?fmzavo5)Q>wanL17EfB6s<saTg+
z_CJKC@6rFl16M-mo-$6sqIq7Gd5Uh}Oth0<ZRs!^xa5iNAUL|<g9ZafPtT49+H0?*
zAd%0kfI252v+=Wfv~EvVu2jaXA51nVYpnwrGbr~u@GWQiUZ#Hh$3F4zn=bN-uq!bf
zJ>`v38GlqfvlCcHun$O*^<KVRbFTx2nGb=8?17Ggb@1Z@jIC?TA1XD%sZMeX1P{Eo
zR%;&samLw3!fXf5-@Bu@>dv2R;}K^Rt9N>W8VQ08a{#n4h8ql2{IE-28pT~@)!1sj
zj>OOI19S21xsko@LfXrVpog#qD@;57XT}jG1dGVd)%0@*vEp<p5z~@l?6wdZ6;4`9
zjbiy;f{R8nnFnB@Sz}YXJ&GIpW-Xx4w!`6j+1qEu6MI5f=sU;ZR$(ROgkpwBNY(+Y
z<pn4%aF)7S*G;7e9c#!8NKI*97|{<%H+M06)tsIcatLSW=W3*|dz|%Z!(~slXbKUE
z+bk>Jle=*El$|@4(@3YIvTOLeE8l)bA@%N_OA~{Ml!JV2jPp4^b)%Q{IGza#i$=tG
z?AET6hCQrpK}viBSc@6=M|*y{=f%r$(aM-h#hbjN?mcPFGk<=fMyMhFR+G?{Tw5PO
zZ0@V~lKoXK11)?K{S0+$>D3Fwa~ic8cS_YA)il!<FXlb7@i}$qgSQRb=e%l9$8+7*
zKb8Kb!23ln8<s)JR9;)N!e2+Sa1zZ5?3j{jw*}RN04iQ3bVFEw3;W;~Yl$@BV>dVv
z&nCnSYcac#Di=uSt?+LG6Sch59f&RH#L*8GO;*8R()(-ky2rdrGvqa%2)th-4qHem
zw^0jODJU5hVZweUdss!&@`B#A-IPQ*H~(Y!akI+3ADI>~Oll~{sfdPs+5|IkGZBm;
zCe;1+Q7NyS#skp8cjn);B7v$4fFoytn^ZrXBFO^&#VN)gUpeGNsc2lS^O_SlD2O!F
zST3@eut<F$o#R8I!N|`{n{LAOgKm1Sp50v!;nU+#<MYnGMyUf@%-;(+i`A%T*4!3O
z2XM^B3X-X5l~mW~1F@Uv?Gt{n<5{)%LB`Yj9<VJn#Oh!7U&>Mxy`(C;d1fMO23F*C
zr@QRyr^aS4sA?%P^0o-Ng3LnCz+V<YEOE}cfxkdKzcVOp#CCW<8`uwkuXXPY%TG6c
z41~Fr3+NH&jUjb2n!|2yv2Y~R9^123%dXWh-XbT!vTU`XVAKmQ8rP`5m0=be?W_w?
znD}&h&%CV}$}ID{+PFnZ%`fca=TI-25Qq48_IEIl&tWGQg6nSm5`b@FJIBz}_<eB(
zP(K@n^F26XVJ4x6NbqW=(UMI@1UDDxnV*?XfbZ!8;K=$rp7(aa@mIG~H%1-iauZP8
z%*rBYL4vZ|8{!Uu1>n0v5ka$z2`fV_eFk{~kR*Uy&<bQ`;T!&QkM<un6Jcu_IFYQ$
z=<^kSZF{fViFUyD#2T1-k~lFtPs6Un$<S87wdh0S&w=SXRZpNdkH&fu#F8WJR4Q`W
zPxwrxAtC|DDBtS$u^-J?N_1TcF!ygpg)IW>(*b-E{8k9wTtb_JfNB@@{TCV0{+i~+
z@d`Na+fZ*59eggR;B=6R`oNai57a$Y%?F;R7U0+^fDum5hnGg=r2yDkE)~y&nys|i
zK`rD*Mi}M+xoT;w2&?k6+izCo80`}Y@#xLH?=sPSf*si6;_dwFcl^)i_WZo{S*qm*
zSX`DurL_P@t3j6VB($qQP{cV&&O*kF<>t>z$UlAWIt%nLa+kPF?gpO<)_VkiqxbCf
zV+yP||9gbGG$@2GgT9*{{qsUY^)XEI1CNy0ley}gonl0VHCU?y3xrndh{qOj6>$o_
z@YhNbb|N+!h#v&hGee?WDxbo0hutR!`i8ix*9nJ^)`P<cOlVA6IIzTUcCu{hxNuq!
zs4vS=JpP?SE^dDv@6*?}<kB^RZ<wq9wX)Kv7tQ`$Sw9!ClexG0;MzBXqw}HKNS_B`
zZFZp`*uSGY4MU8DBu@0Irxl*->RI4BCAv`k40`|q<xyUGSltWsPUCMqaAxP)VW*{<
zw^O8cV&D6!;7&v<yw?=hx{tA2rg|59|HM9e>c#(3;lA!6dZfTB^N;4V{OMD?ok%E8
zE5@QrPFz`CsR-~`Si0K?-DXUY$>rUSF!!!84MktP7$V>c?lkVym5*N*^M&!Y_=EF@
zP@54SuOuM@Lc_w6n;d<QYR?gGUwcth#cUHH+P4XGzgspo_s8`%{BvQ#o3_AvWohn;
zO*{WrNPLJXqWM&7fBZVhBDa-0052;NgknWm_*Q1ohFjH|*5&qjJR>6(FEv8-L@Ncl
zELiJ_x24IZw6R_mt+3b-%J)5CGm+MEi#dRSKWnpsf<|>`c@aI^+MQ3qy4zpiEwI_x
z=v}3be{hFC!RBgazo)C!9e`!Om`SAR7%b?iKb()Us<SFGhHXh+YgR+0MNH(%(yi+@
z9Gc_FO*!U2<l`2l90#xt2Vrd!Z<k%Pi?eBF*nStK1?@V=4EWK`)s;UpWN)-7?Z*oF
z<j4=3$1Jw_XV<04tPfpxg!=m!$Y0JlKUjP~yt!3->PtY-7!hgF6}Hhf$E$kh1Xrps
ziMP|^idHgfH<N>1281bFEAAED<k)g0U8zruu#O2@dk!VR4Rwg4U5}BOX~FC^?0$d1
zhdJK^GDP5)^@z&9mwEi}q=bKcrTpm$y~e-Rgv&`!mB7wHJNfb9C*!(%ii2Vs3`=?$
z3`8(w9xvW5!$PaGel5plpkm@KK3HpJCaHvJ01=kC1cp%mT;gmbeY6@am;kud^X{Z9
z4<8ft&+5_RHRnf8!HIThaWg#Mr#<aPt|CVdJh&5+a6cK{S`feW<PQh+D%p>%gVAes
ziP8s186DtfEkCHKF_(=@;jT5|&B&ax#oi3GvMuy0$|dxqE|{UBUH?&kYh2O$GLD{^
zT<g|bK1(k_j$>490+9Rem+r)m9575@HF%`}k62#r@TxFbj}^Cw5INz_js7r)f&fWz
z)0O5A4p{*%d!E<a--d9X(cHSb+4uPcWo|<zB(8kjuGlWmIgq)snfE3oRC4(0VD1A>
z=s0}II>go;P_wHEbfGtCZLG5;!h*}7`MNqRG06AmWeR`!7+9{$E-0pt{Pi6D<AM5@
zZ@$GNz(Kr}g5@97zu9|OaTnf$RrIk8e+$e$Wr_K>#&&1mfX54aIJpP~^H}jGk`(`$
zmB-V1vgTK*4Efz2jiZJ5oR*qZpVGVmVvHZJxDV7(Kp)vpJ6g}GP{h`G5~!5Rl54+q
zwm*n0<z-h&>^wvYeA-bXNTNt?a`RNl-uM=%=<(AY)vgG6@CGYiT(;K3n-!O!lxX64
zzfz{tAy*M8S;rJijnOHCdlC=hM_ZkB-+nx-|9-5+t*n8u%?>dK)mOSLmEac5t#0zh
zN3`qZq8%)J%k)1Mz7!osrKDx)E6d~vXZ%VFGwk@3kJLbm4z|Y`(=`5CTXdz?_uPEC
znpWn=bEZD9)JLZ-a2&DZwnS#BhEVcxucAf+{G#z~1e1m8=A9Li=hk<T2#)6n*{W=h
zmPU%}%bt&;$gpD(wSwpWdesMT??_@Fyv_S(nsqO{^%fuJEwIV&8B*Hr{(sne>!>RC
zt$$QX79uVh6zNiF5JX@B5~89aqQs&>x<y(*5Trp2QV>BbK)Sn8O2Q?b3MkE@^UlY;
z&)Mg_=RN!U#vS91asRk~ZNx3{Jl~k}Q`3S4*EtlHT$<B~^d`=^FG`CAoz>)~j2+qo
ziKa&U+LW<LIJGU%n)ktM3~7Nx`X$GzDMC5ts2f(A<BKq<YaU^vT)V2y4m&k7BZiiQ
zAiNkFpJ>joU$#^uQ}L1+=Il)QXC5<4FpkfjgTm$9hPDOE#V;%n*|qy4p%Y1x8~80>
zj1fbt_Wkr0yd%t@(+^j;70jV=GTwrodic7`veH8piNetIpp1xRM#xX|aJRq<Hy+mA
zAhr?=7Tfr62mZH8Bz$9Pg2H+HV$pHBD1CWrSWD$eE;t>!)?4L{fN{nJy*rEUvVl%e
zzc4l3rod^a#OdP8Gu&v8<1!dxmvj!Ztv|Ndqm9{boq9$zraHK2e;S631aiwZWyz_L
z@hjW|UwJHzDkr-lVVlXkD6<2CZp!4!b?IyEt7*SqeR9y(V-o6WazjiXvVhyojk2%W
zn%Y`8d*48f$^BBs$mu@LKKyV`vIoi$pR+QRs)`g4ZS@ZZc@k|+77B}hNp*O&z<};~
zNb?GHJHfEqDD6-Fc%D>l({GEFA}-8?$vlnef1r^6j+Uh`qhQY+OG}aXvxt(XAWoo4
z$3Rieo~Mw~&}=X3$yulHFCsU>w;!oDy*J87BckRI&*;|?&v!=7oo39$KFFMyQk5*q
zXY7vex!I5P!CfFtYxl^kD(@DfU(2yv{j~Seje1!oKUv}w&i)vAJ!c`e6K@Ad?9?ld
zc|btVZalGEQ2(^OvFHFrUtCsh(+O9I<?Ri4i)^~?dsn%ZTAzw!aBaW-ez#SC!RC;u
zrp*ZvlYnc`iKI`!!FD#spxpW*qZZ*4>378xrGt|N(j-3fwlooKc$rSm<}^Or)MOv1
z_b|C{I97FSvI<v*tu`x?C6*p*R%tA4Jc7C^oE_^Xl%0Fn<B~oT1h{wuWS0BVea68V
z{ef<018}T9a0VaWC<*GKjP*j9dV!*2B54{Kvvhv?MgGKI+5fy!{>K6L`>*WOQi{vj
zZteU7DafX@Je!&6^+Q!pkHMI^IsCW=)i(e=^tFOVunNTHtQUhvpa+*;Mlbiz<(Y2O
zJS}-!javSdOw#(1%C!@-%`M2n*Xr>7wR((lCzxSx1U)WL{vp+BbM|CL*bzQWZ_o!R
zOE$y7mUl1u^2Epv$I(=;S=AxD!9-nILnG}c<btG(am#~X9h;$F6p8VRhMeO(WX)xt
z&t!bhI3xWMBwSv~!6O%K{Ugg5C2(X-+3~%wrhR`EX1ErRZNr$^k$*eP7JdridH`3H
z8H{)nHEX2^iy8HEAKW+eMM--H;LyFKXLN$yooqN!t>B^K_?BB>!O6uF$Vn{<$R6)p
z&wo9&63Mh3>?;NY|CBCyha6G}OOL^RSP;ZhF&hs7M{#3$1kBEo%S$=>XO^6^okD_C
zugO$cO&DXO3j!tdz>mTngRg!(v(%LwwYfkhxJ+3Y5_EXp(O0$}L*x36Q=sQm>+1fT
z(1gSQ68&PHKP$U%-OrFu2RV_9&%qH*F71rg=-Pl_-*bp`=Ls?rnfUaA{(EyCxSLn#
zzHw?^UbIsY(Sn=d<WSfh|KKh~0bUl|d1Z~wS#~MK7M**!%y#~+I2$c!saDheu~?lq
z%`~L@lrVepy^q{I;V!<?Czipz;S<X?^^q&JbwpQ<3A=XP>GSEgTma?y6dh-d9L{%~
z%bpY3TAfOC;E~(<r1k`q$5TSjJ21EI=e3Epa)AWP;j|j*zpuyt@tH47$Q8__a?OtZ
zH?lYLl=v8&1$UZ`TZBjTdn`u;HD2YpmhfzBcp5B&UdLYo93!{<rQ#EmG>J>KFVQ1b
z6ICz}NIJm9-4HOM{wEO1C3->Wu^P&=q)85K6Za)~Rc5cfm`|59RTS;Om?|VY2jSYY
zpWp;$Isl^?1Iq~IZ`ITCh3Xc*PB=nh+l1?negfwZZSuIKhJ+K^<glb3#LkTm^Lk-0
z_Rq{beYbZS84S6|1xt+FQ7e!Aty~qW*MdvOcA?oi{1(i`^r5+@oRN-=RIZ8lCAM(|
zn*I>iF)K23gfROs^cSDNFqq5__@s!rl}{COfmq?oRfeu>(cWfu=dyw-R;yMwU}{^i
z1UuI?{H8!&=juKp574nnf1Y}O|Gd@l1_$1xdpE~-FAYB3%hHm0RvA0FP(+3c0#&UT
zu@ZzQXulFhMY61m9tT)&f;|?0EfSwxM0TPpTt9ts7<SXBHFVagtn(%`R)%@_(qxa3
z4hi+Bl1QJEQ2ntt{E%%}*s7$y=zd+9Ht5to;xjFN4t0#~+Vx!UJ@SG+%5eo)*M>4(
z>lD^67jcU)wYEX_F7#(mbT31%&oa9U#|V=R-Nqf7#d8%F{5ds*NuCz%BX3Rr3%YBm
zLin`nT~QQg=HY)f-F;9ES%Y*g?Dy$Cvfif5a-lJhe;qomCKra=@nO@pfRdcyEP_`%
z2p(NtX~6gGK$$1bfbWMjYrinud@-ESctO8ikKaPDE)R5PpN8-WfP{AfDjxc-S<!U>
zkwppahT)NZS^~UDmMZHPSdMf;9lF>LocGgqLFV$fFM2O5qGVQrONBdzKp$p#W?+%<
z{2)Yj*?ugY5=CcN2x5gdWVFj-i41g+xMh6QURM<2m)xE7?j<1I7_dMj?WMw>hsvWp
zBn{KDwtA1Y+Um=+VmA9hcV!1-Xj42+4b1&0iyYR_T)OA?rX9=)bc^XP1*=3EL=KZP
z*dz2~yQ1TX-OD)-my!Dz{#_xg=RujP`K8luZ&qfRe2Nn+_@ZFUgy8n2IiksaS%}3_
z!e`eVRJ#RlSC<Z$OV~KK{k_eel3)0JgR2`#2;{YJs9Q*MHkum`_G=ADFO<|HcFQ;#
z9bNuBHHO$ndmfULmKk0qXE&Z5Uyqk<#whp0nQIH>P_Bpjw567-&+8v6)q(-yXdPQH
zQ)=$tyf!^-M|1WUuswOOCS-9$n+|0^--x`-wPQFIPi{U3?=RA$==&4Z0aj2!9rONa
zm1v3k2;JEO44761VcS><I?mR}CU$-BSuMEs<*oU&2SAH9U*Sl%%zU`;Lqa+ruvxGy
zT>Jp=&?t;E;PXX@4}4v9-4Qp`Dp=me;yJ_qTf>LB+vI@N$>)Fnw%_f@FKxTRuXCV0
z&4DVc83whCS#mBCh0}SJl5Zjae-Vgm2aqENFDE}%bkoGp^x9OC-%da2UKE=@(a!6j
zwTQx4i;_MJW2`;yztNnlq6kNuN1);S$0fWiNT8&PgqZ?-rr$*BvvHrMPGaK?O<f{z
zHMlAuUYHcPiP`!CD2+fkD7Nv7CKdCMvyQzT0J$lHPjj%b$kk~*f=MVzU$gd7ZLpVp
z2Kq8q4YeYpWP{pR`PS_mV2Fx+$J#~9x=N+WepD&WeZ*rrK4_Dlc2k{{bs-ycQ}7~Z
z$wfF6&Y+nrgC{ASN_k6XgcL4qaKy@%tqvY-Euh{WXQ@I^%c=1oyQkMAE(?jT>Sn~X
z)>Az)I(E$JE;^#?7_keNWG0kN>=Rd8*)jucBnthx*Vs?EKrczJ92d)jZa(P&)Am&Q
zv=d(LbBL}zZf%mpJ(1y0XRm4HX%InaN#|c%sFgjQ4d%YGTa6I)&|Y~wrsI4|*xRA7
zTPX8xi|^Aa{q6Bj7pON2yLl}45lDW4yA4Pyrb=U3_NB@5ZOph6N{QhP*3R+v&Bkg2
zO8Kr|J0u8(up~ZNb1G^B#NPI<$;J8Cdij<fl6sxrSqJYrlZl>^{T$H(h0xRoud)aX
z8P)ym=0ig&)-l{J02*FkCOUjH2cXnb>wNo{2neJWPX^OcvFG7REc&!ZsoqWZz>A<+
z<eobpJg@T-L+jvZguYXmDm%ibow4F8{5V_f?~MG&1^A@{xDT7@F7?4b5g|5}2k1Wc
zJP$~=RZI!um92QozOD<2*puhnRDbr?5BEFx;1iduAG%2goWC!lLffL^Sb-VuiV9#j
zN$h4-SI%lMnU|Wkid|>$BN&;p-h13ZdOtw1f5m-ZoVBxU%Hn0ch(YLYf{Y9hVRogX
zt7K8iEY^kJ+Xa0MJx7kJ{;J{R=&W^S8CeMp2>6+Ay#?sKO8IeeVm>Pcu2S&-9&SCm
z2)EwRXeG}``MP@6w{Oj=+w+uAkG$b*_GIt7%BNyPz$-KHL;0=MaTzL(D^QZRwT9i!
zp_w8wcqygY$NM6W>=oZcx^0~pb258Rz}R<$YYRc)@tgv&2lIk{hFX2Z&d$IF)*fC=
zXzRJ>gNUWgSdQWMrYd;t{HM>EYC&zinx*Myn6cCKRHAE6#=#=d669RImry^brcD=7
z2f)p&_r+3<t=QjqhOK?#ndtF$Y##z$1Kn<{3stPF?E4kp!dcng2dp%Q^<cC4_J-zm
zdva7?ly_X^(OkLII~Bx}<JsLKicEjuGXFbhmr{&)+Ka=V4V?Z*P59stp@PCM>UFU-
z;l9Y>i#Gz#DJu{h{L-pEqUO_HBu2M6emsQo6Y2wHaJsQuVwT~C8-`$xLG6*1!o6~!
z*(8`#wIrgc=e?`5ipWrG+S`9fb2`Wr+p({Dy-;ils>%c3^qeD-9P7x$gJ$M2mA6#?
zOTQ&gj8z6IL^Jvp=AVTZ1x?M)06ZvjWaxnV5Ck^Z*}{wlif`Y(p5qv(O+2oU2#j9J
zhoD#3!MlJ`4<K_!S?HFH0<<_~j!!gRdhbSAvqSspaW0zL;DFMfTB1CfE739l>owd1
zl(?Hae|=L7&_Xl($x84ode?uIU^SlCK%;KDm_ljds?b{niAEB`_j+zusxkc6UtpUD
z<QdyjB^lrT{0+XO5mTO}<dM7<K;K|93LpeC%PcHVK`?nVcDLGX?7+D_7%8{!J_xZA
zc2G*&U*17kc><Ie^HuzxlAe3rRm5;8CQ)LIJFSomSUbot`e~(KoflH(SZFq`IY^u&
zvpM#VQ^tD0xK0ydUw;04@8d;4P_Z7tf+@>Phl2cRGftF43e;48#v1&G4=zEG8~fx-
z_CMqyU<Z~jkpTH_H-2eM<<tONLwy~X4ZV8K=l=vn{tu)qt1ig3;zEAo;Lm56LW-h{
zT!H}uLKw)IFy=*wtH4wI>Rtcsg8GZlkT>K79>cKs(3;_Yp>O}gt`I;9`k4AuMA-D#
zm;FEf_}_k~e}@qM_A>)dKp|pxp5OY}|C8??{QtRMKP8H0b6W1e%Ft&k{Tl2giLbXb
z)~rt)Y=5h^4L$uP=Mzt2rq)K<?TFuV8{zPazjpq|GYQaW7JR^d<s`&h97`-PSYuSe
z{GG1r&sXhv9Cn&#|AY6X|JZ3yiANyrf2aCjr#g?P4NL|S;EIU(vG(Ta+hotr$?g+m
zy54JF?VjJ({CVwKR1q@b8qwbL0R?4^q{*~HLn2XsRICrIFjipZw$s?#0)B3O^kBc(
zduMRe{zGzVq7|RNQa8eYk8X9ig@%eh2Nrg*9~r)ga1MJjDst8zK90cD2UcHOAiq5N
zE|eD>A8feT6L5%bXO29zRWmytQ)s}>|3cC!mHYCOV`+sw15v|JM)?XA+U;*m7b9Bm
zdn30=mo}mAGTAf3)E9XOD%cQDR(d$hoiU%kDYPD5^0L9;Q$jD@Pnf$aiX&JraXsMl
z*UZ*RVAVHP|Df@ocSA}xHCzPQkrmqlzb}F>d}M4z+k>O~^ah_O_1L)Nxx{*|iZ#j+
z9|IoGD|T%jv9AKb8}qAKdf>YB{2fSmzi~QS`WPiSyI`%DZoU-+v%AjaHV>SB`@#f%
z>HGO#U1s-Ky;q}QWW#5edSRYjz3&+;L>0$oe;$A{#a`UI8sgl~aKP5%;0qSStdrN)
zBOVN72(-umNTQ9hd=66}M4>I1&x)B{`)Xk(!K0Y<R~ct)a7D<MLoT2>fL|A>jm<gt
z%ug~J2E6Fwo#35<EXBPO?>}k{_xuvie~aV8#eyfNn(OoY`@-t(u~7iVTp?*gd56+2
zSFi)Lo6kdwxmcIGLGTQ$+3r$3yRg!!Cb{)-#Oe%IkfA5uAGj$rxy^`=$t=lxL7Mx~
z=v~C#0-TLi|C(mfE(kD+Ai8h}$>u|haoF2-vAsQ?*RJ5xR3BY>{?DH9-%poSW8^>_
zuhshY$ARd{Kp-=X0qw;)BKtwo8kT~^C&R@C?&I4nAZ6{<-gjQPT*z`71hwUenSlu<
z8Uwki+>Rh<ETlMbwxt7FubT#1tZXvUKm^)>JglgJTJ`Lnq>?Ra7sgAy-n1*Lv;S~N
zozOo9!29|by}OqO;W|~gfi%Z`pslg}P3Z@7M&<x$TE?zs<$grhG(sF9)yRe1v*zeo
z_pNsLvMZ=vK#IGT?K}KgMM)n3jdEyDG>ag*VGy!u23NLq^^N&#cM#YQC`vxCaG&)r
zOTM0|Q(#r|fB?Blm7Nhm`_6(}^<dAn*_G*H&-S_ZYeyX7-j6X@Vsh8Dm@?tYmk8*{
zJdH?N+^^1_iL$kl25{vL#42W00-?mtgMd_ux5B<8+hl^>E2`-HR2*lea<!e(1D6o*
z1xTkkDwG|5JA=k%8x%p*CU_MVkJt0c?OA%+XXUVV*FOA>f-)r=k=kVIKG+uebC;ml
zg3Bf&r6&|cSo*=ble9!!-)t)LYdQKvl#ly{7l;awW`8!uhw839^vsKzexr!)AzGO}
zcXWTfxCqEEiD9*hFhJygtCS9pzTumDb7Y}v$9@pxr*>eOEPw`(zalL?swtm!r08Pg
zNvrCe;YC|$yRHd+T@8EN0^v>KjQEqp?)uM{CN83eKT>Y8U&MgAAaxH>NNBmdGFvOE
zE25&SrB!<nEHijP@7a-EN?Gw=>Un05y-HHezqPsQWCIteMNtp7D}_r}-gSMX-yC0~
zyMRgVU!qRfU~G{JExwYz&QrsxvLyJG4xI3BGqQ>j5Sm~ArmxmhJM&~4pL=UF?|lAq
z72#KP5sJ7fiHTo-)DZI7Tn~f~3)z9HLp9)pK5(9@d-A((F#Y#Y$}E(%K~%>+(k0Jd
zLBG5ErLB4N%va6IRgegZ6NhI}Z{GISW-=Ut)7chEZT6*xr>3$<!X!IkV?8Zljb}Hg
zmK^%}j__p~`EJ&;&gqVBYhsGO-7exI@o^D<O>w07a}ibVJakd)vQl85_pGo0d>%uL
z$u;-cTg5RK+Q^k@j1_lXWD0zYsP9Z?Yd*ZO=lz0M)suG#h$q&N80NZ0H?Qf0`$l4T
zJ$%~b%`*M}e7?#P--TwWm7weWpI3STCqczKL_R`geh=VqLE8y+6=pL8nGC=**6XLQ
z!g1dR9yv?o(%ggFGsLk9PT*_qtROgje`Z)S&TzKbu!5XJRw?{aU61;@j-mJ3Z4`I+
zYye%mm!ZP<OBje#j{1@t{0OLl(O3E!*UzW7%kIoEujC&u2;8^m0j<5!ZU(P--IJ(&
zckJtI;Rhrv8}G;$ZC(S?A+?fGIQ^L~y8wwStJu74jmO$#(Z!-Dv~5zo-Nh=PROL9z
zo4jj%IU*-z1Ty19wt%UhhsS<S;by%B;O7*^d=emopcT6Myc;eo`#zh(hWj-gNG8*}
zFp;|CH&Bc&$uv7zgh&}(VBI8DYkKP2$35LAVlsTTu)FB}(`g#PXO1OcNqW%^clOV;
zU;B+b@1v?Z0mVAp?5Tkvl$%nJLdP*>k}Uag4|OkZf?P~rL$^dIX|a!V-hzM3kSkmO
z8)m7})%SMojm#KDn)Q2}xR|cW{&~F&vr6l)*r#@?o<o=9y1f#&%<jj{LgAqkkEwa+
zfeow4Hb`tKI+v93eGj<*h{qM6<?o~qF_pP(Vxb?e&1qa)7C#_i!&4NIRR19XBT+t*
zD=T_z*R(@aF5iY{R$m<bD3fm}@lED6i{{{}ntZIGQEgef_estxL2@p2L{V>g)q9gO
zYlMcw-Uk)(dY`K+`gP6y#_zNesPH9}@X9uqimvyGd>bchUEK@EP{vdg=hENSeV9LV
zGx1%~h}lcSdwPp94sW;Ads+X^T^=m&z=LRdSTKcutj?59$`XKu=8zGJ$P8NuC|j%Z
zsfIK;3Fs;~15BeG1EZQ9DwZE1Ss|{VzZ`_|ABv2^kcK-5ex5rJ%<3KRWcKRGSa0_%
zR$o}2X40UVQgD1QGX_gQ2rn93e12t<tzE_LromYeFhZl`!Veu?^afL;!F|cjqM%7o
zBMJTbd<2CzLw(u-3hP-ORh=!+c;Z33c8|D+)MS<$jMOZ#GP`VV0C#Cl_3e!)iC*r5
z>^&*E^ygC1(hq11HkYO@xldzh489{;G)meriaar6MKOL$jWbL2tz!_`@muv65R=Y~
zVx;;o9_R>V+)<Hh={`szn#4%sOXZMdWlC3264C2N$|$j;M3HJ@DO3_B7*kGCP0+JI
z5%M&vp!>@0#=Cv%$n{nMvtzjH7pP>wJSQUoqUnn>H;B*odIR1OJJ5Y)@H*cxTu(?8
zhby4Aff4CjYr7mL%OE+Bvw+Jq7u=%#K<MhK(Ww`)_>Cgzf1CxP{5p8bL%H-uXsC07
zw{n1X=mVb{<)k`S-({vvxUK>)<KwFUJRTFF<zU|i$|*6x(i2RJ{4{M9P-I?|oW6CZ
zMd~iNxQ}<Xy#-MF3#vB6oBc2ruIQtk314q!*_anaUi_BvQP1YYPn9CIh?AyKh#0Dm
z$oS`RzV&R7=E$xq;~X275X9&TyC+j_h^=nWG3FEa1ex`l>wF67(=;q+bAvu!$>c;s
z!Gk){|BdQV?qa^%QuCo;&v9`uW!Lc(m{d-dB1YypeMncthLf&wyhG5R`mj^uR$xYQ
zFPikBcD96~Aq*3jTQ4nuUq_UNuSC7Zk&BqH_<j&&)!zO_b(Cw)6r{9e<T~UotlY<C
zK>;m^(_4Rf|H>bcihcC6ore1>v6Mu4noD}PFwc<)_rnr3o=vO<C+{!SZZ*dw_4uyq
zj*zgxpD4LjWE=ML`%J<bPjq;p*{404PhV*ta8*xQ>%hClv;N=A)%tlaJN}Uc@NfOz
zsw+~OtyBGjhukLi1Knh2QJ)DpQd&(aim)zHJ^N0>8p20rK-n_37lmF{6)3zFVAz<}
z4nid>HMzZlWhZx_HgX{PtZ0eTb6@Ahx9w#H^(H_KDcD&%6<P{G*&X{qiMEl{u>$vn
zWq;X8I+4=uan#bIoLu+D3oo9k1)p4+2xv*UIl+0}zwGStg@eb6$8v!S=WGA44VWjS
zPGXV+wP|V(u`U|>Gkz~0-Lpp}UIfDqN*l0*bBaS6*}SKe=uuOQssR-4-!sJ7u_pXE
zBmp`}>E_xu&JoHf(Yi}1w1oCOlL<J3DjTa2n!muPbM|Aft)&=YWop1C-Kz;GVnPLj
zhS`rQJic%~AUdGITu{USm_zqqtsEv|J>}%rabE6LfazD(bx;wBcE)8CunL8?iP+xv
zBcT^2eqd#tZuizD^5l6@;#5$j{}g|cqe|Sk8qx4V`Tos<^;;R$EPkwnRCPMk>(7?u
zUxW(S=F}d}9T8?hW-qw^R5(dIof>e-5`yib3J<HJO+#C#)fs20OcDS=A6*IvVLUxg
zV{=r&YR=Dz=c>33(Mdd+*Bya}WD?;j>cp6T-xF&r`K&pDc|js7Izrmg@oMhx?<{cE
znO&voq$k@wQ3(5C`B~zl?`A$m=aeNbNzdxI<#*1^#jQo|ts)#3hVqr9Yv}Yj92@RD
z?&Or<&Q;mk%>K1VsoYk01Vvmk_0GaBQJ_K@5nFoHJn(dQHs8*;+D!0i-$SZ-j;j|l
z+lS2e?x?+1%N&3E;%+Fj`<P$<83lQ7B%*Ql@FH=W*?`|UxrX9ClA?$2w$D{>T?=k(
zJ+nkZ?1Rc3--Kqg*?mP%62H)h*CDdhvM|`}F6(hgUVF5B%2Vrd>q?k#v@6@^JHBIE
z!CIfD_kyJ;rR&4)ZuK4+?D+P{qY_`iJf)&8_l);X4Dg>I+rNG=iXvQaQT~U|3dt$i
z44P^nAqoi#q&{eQQL<6mQ2CsDze{R4Y9_?@>c;^h1{^o^C&m*jcGL{v-1}6enH6CK
z)C^C#S&j&9^>_N<th|4=FVG1%bs2jO{LCg#5AI4VIrYY*A9`#7C*#|Vvw}CmrG^r(
zHHxkofY>L?q!iT3N^*M79+#j}Gd~GlzohpizQmo&N#t^WVaIj7L-93Dfsr9WkFQE*
z^CHrbu8(j~=l2X^YfS(_yRLUOy2O&#>4N6OE85oNS%XyS4fD7XhIb8^VQxX(sd9r>
zvb8eD92518fafJ|UpG`L@CXL~ywO2IPd52Op|yeJX^2<s?2V>I*${FiYKAA=g4cgK
z4rA=$zYucZm|pK9inSUPv<-2(hmHiD8h1~lOGwbeaL>5nOgX{zj32l>Q9;izDLo~P
z7{i|pjWX&Z^`=c_-zT2ur3&7?taqPbUEg)wm$!`-Es*3c*_v~F!(2x*e6171#Ir!g
zrvv=0DYF3I7_VT<Wj)(lG3|_HyfqCZ1?l&HH3>$_ZqM8_D17LNOSrPutE-}wDZW<p
z>ZYM}yM;m7*y*j2jjGrZRK}cy?Ft0(N9;v7)obD<dX6NX@_P633q9`V!u8DZHdThD
zK+M$3#gRE`hDF|mwX?pbJ9KkL)T6_UX*NngOqf~ktamTIq)+rO-R39ijVG}sH$<km
zZ7PEjBU0N$7-&HAmp!LB^(w5-Q~;8MIH{*BB$h`Sux(}G*Gige=6-a8OK^9x&*t<r
zNB2!x(gkMWaj8Wv!6Sm@n)40He#=+URC+xLKS!>lQDmxa?FH^bQ~|tNX2Budvbv(4
zDsiz+FKeIIcHjO9NdDV#4ZZdK9#r4CEX7|@e^%Z1HOSbsmu{q!5gu;1g7{ldNP7O<
zW=lu&`jrV@;IJkzXcrwSGzVG3k>K4S{I$lr{p4#tyL&vSU2G$D7X={*G@mJBr7qrj
zRD-^tlSL--GRX4!keE@h+K{PUR&mVdbdzAj#q><;X9^uZv|?5AH4oU*PO*P?6uQBV
zy_bDQS^2}%ZJJH?H;!mMG6hMcd<e4%0}}1rh2SUCpNIM_(;}#+i0aC&gjjgAv@j;7
ziX!P$Wfz|L3+EX`TMS49aH8mk@%q6oEB?g=dKT3Y(9Yp*S??e^$cnJ!h<T1j=Rpw?
zCK$Uil%~~o0gX2JtcsS&v<Qy4nfN{S?!qS_CkIojT*^e_YL969_woxVBjRFiUN3O}
zST^ri3#Zsy8jewIdTf3TEiQfALa7S}A=a2?i*)1JgHM&dk4~Ma%?X;CXG5hWp+<&i
z%&I=A7xk}wR7_8<-5o<yY{nkx9U`)rNEQeSb!#!zS1y=I)}SG-DCC-AhrFfR8T2Yk
zkq>r3vL?jeEcMWcaokj0;ZCRX3SoU}o)wE(X)|UyUFWv%@XnppQOraPS#VFH%pJ!}
zl?sOM8d`m>+u`O@9%}7T1@qf<{{gZ7ZK7~E0dQcI73YWjKcJTTlVtAh0}YCJZYXw8
zxsQWh9(_eV(oFd!-TxUa@wEHOxaU3=%d@8}6L?NMK)3L<r`-O!E23cGD7#>(etMao
z7@HLGtb2#*9{U}mt4nje!IY6@OWluzUnSg#xSM6-SA#}dqHbkySGcmfH`J?FzE7w@
zkMq|(F^xRWQ~T7c&=-Iimz+<<#^6E_(G0ehQf2XxKr-%)b2X3bLN<_R`1%A@cGq`f
zIYq{YZ$D;EmwEoOf#fFb!kcwZt+>@*viuSwmR`z~NY>MJ?Ke{tPBLWDF?}z$CKc=^
zTwx-5E>A&)828Kz3hp!Fh5Xec;`wt^2T5=yWPCf3pzgpFx~*=*!eTQ%(vJ=PLeI`-
zhG+TxYS-7FJqYm#?(3_69;9N34q+Pkp^c|v<=j?e3_ErHB{p{?-DQQ8gay3%vF>yA
zy2(KW6xsPG5dbzumr~<nve(0E&<;f`Pb#swc{bsveIGyIowO#SXOY4TU~e7s8%6sp
zJxo=N4*GIDab;<ZuJyan*y(t$B5)t?rq|(tf)@$c0@9HFV#BBH#wwKQ_JirB)Ly!x
zB-yv|9wQofPP=lSzoS~l5-%EwN_{`T>-=?3#ZoguUR^0E13s#AYJBWB7yOD>wQ^b5
zZHi^sc4@43YTQkA>-jO6`|nC;8uj=!-(NaCD))I1jNSH#000qo2k8`u<FxB#l(I`z
zit17U87@N%FH;kKb+ZfOO-MAC<;$R_0{j{d=2?4|+^QQ*mf=ZWq#ye9Hmd0PWn_#S
zMfDsB{7X;O44FO^@qrAYho#!7zc9}QuR}uSlnJ48S3>DV`jy7a+x**WvkWc>mPPln
zkz}rxL}1F`6@qGJV<mi@0KS7i`n&BX1idN*c)3@P){3Q0%C~gxg@pU2sx*L=?Tr`H
zZ3JZtKor0`H4fS)gU{I;DPN1`CoX$|yRJ?7+<HChe$L2LL1u-QrWiwCT-c8`#UNK~
zc9^aSMHz^m>;Qxj<TtcSs(B;x;*E#<mpZTfc)?Sm3}52HQrj@rF@hpOxnwwvUrtea
zMF3AVyl@mr5z$<rYl~|Z%n@bKg5WaO8?)G+&r;Y@-0OL3Wx4J$sT~(}TSuSo(JF0?
zzix-ILhjVm#2IJy#C8w?rsrklgxoYckTw!>&h|%WrQLKvx#oTCZ$tTsr9q0$Xn08c
z&g$77eO7qK;wgj65o<Duup-r8di?@vt35Jd1C+LUC|kj25*sh97HT_XzI)8E3znT_
zXf{v0;eOTQAdQ4c>YnBJT521GL?7EP*Qk2sSeuR=#OAOQ7BnqT(M<AuN}<lZ*(559
zd-?iz<MMXpNLR4FeY>l=I$X&=2vKjeCp{cY)$T^MPZ26P*Xs3`>tn6pIKsM#;VXEj
zRR(<a{s!9(t~nP1)XwPZu3LtGP)}94#FR~J_7jsH>GvQtF+M1&fHO*_`_a+kH_uEN
ze3&3wR%Xzr^?%SEc!U(S=#ETh)_&FStskzW`FWs3^K75bG*zDMhqyEw=S8Qf@k5G^
zerzdS1+aV)y%zI%U*7K?V$y81#x@c<5!RhS7i%x;=*iy-bSHU5l={T;rnuiIg77Yk
z^Ec=LdgFWTD@QUCfvIXng<fAA9i`mPD#T&acRkVA(@PH-CO8{xg1>6&m_7GQ?8#dh
z7FdG2?uSC#r?}n(&?ZrFBp1k}vf$25(bZX^4Ot{YAA172b?kz`8mGj?7XNoeF}D^I
zbZ%LI^};&qQ&IYe#GS>gjZH|VAa~-fQ){sK+KpI!S+0~mtjVU6i1xKtBahpZ#t>IH
zI-#E!<5}vy%JjMiSLNPDi`y+2=k=~?X1d*abcwZGAs})o+~xQMNt^Q~)sWa<FGK8F
z@IC`7WXItQ^WcxF%(i1Ge8fJxegR&KVcvdj`Pd$?9#VP<nYOUBCfE4PB_Dr?2dYPd
zM=yQkW6pd*WqTXmd|G9H@y75&hXMY-eX;VRTn|X){)*sQ=ay%bsSDaB9sLl&)$rP@
z`N|NJj?VcydW2n|O6|bRKGSb~Njdcu^OFk+RNucj?ol~DXie6o%Gh+})j2DrjQY5n
z7QfGuI`gub&}OZx0jeMAW1idy2a>~n!XSZzIG>#G4DR+U$-obX#*na?4<p@Y=mTZ~
z<%+Zfs1C>8C_~WSLX*I69|C6@ll8t}wKf}f6^Ft!1JJH&*S1t$l8U>F$WFIP)@ACH
zvvHl8y^;35i?VW|HhY3ozvZ@5lH;=3guzwm!uO%i4;KBy#A!B07+#4C>l#i#S$2Kn
zXzU!+WK=P+IbCZ33z;ri?{P(*vzCq6rw(eBb}gfZo#%~pbIlmkH7g_b+DYuxC0J97
zZ%ahm$*Tpjnc4dSY|{VcS)?%-RnT05+o6xM7e3|Pi+BicG;>doC+a*n665l=^LZC)
zKYpaFEDVtB#4Rce-T~~xR`Q-ZmVw<9dfmO%f_kBGr`&4Ksq^d$r53>B%-=ASB}&fh
zdOp&rx5}{iARx~T?Y}kn6kh5oJMvPeNUx%HUhF^QT`c)q&Fq(0GHt_!itT?0W-I+3
zc-e=b8(BPs$I~p*SAnGt`a({1MV5VAacRPs3$fAk?qwM6oguyeq9@(c)dLqQS$&83
z^i~@cAJJ>hCLei9bWAcO)iEe>hKRLe*hb%&IpPkkCrUHdR`XuH$^oFX+!x#nVAcf>
zYHa0`;sy_3+^(|azs?g%ZF*{tAxYgw=ux}wEJa$E#~zflDYEb^2%K4Ve4JFoXPA;u
z6J=4xP}ER)fqP+I(AOr&dE@%mS4nNmN`_2!1?m$lbkajl+l{VL_S2HxgI)y@Svs1H
zXiX&8X?2P0DOAjizogl?u1)hY)w4`@Y?_6LrXiO_vslA8HRwdvDdKw$@7QmgfLbM3
zeptiwUT3{M&w{#mDNJ%u==)TrY4gD%>pil$&z9q83)4m<Xuqo4BnJ_pa}&JYU5oR5
z-VMCu3H;+VE}<ox=hyDM91eeQT34;IIdg~QO~o&}1&q*?ToI(PYA;eUa)f^5DKckX
z%^3nIvZAAkw2ZCi?PUgUbKg6|0$23emlSDRd>W+!_B2&txP<wH0WtZ;(yDBOT`Pb7
zRZ&3a-sq%{BKvp+Dn&6o5l^^)bG)yrVCkK&8>0o4|D|jcMasrcX&F_2NC570qTFXH
zpWGm2zicp8(`?h=&2mvx_7sE1d4(HD3<EFAO`+v49%QL8wiBkikRhc=T+)4oj~g|=
zLKgis*3d_1Ju0diG6O7iSj^6NU@z1JsN@3MmX=B3j#3PE?Y_*b%nCXRj53)q6KgwF
z%8$ybWxW86RVw3~>fjP@&wCxQY6qOl6(juyJ^@8J<0lNR9MtvFPl6P$Lb4m4GXI9!
zK5HF+Q$_N3_LE@wov7>xFmBJe)J$-Al_;8d#B;?o)Xr0~g>n-~RZKck6uOPQHm2f%
zCP?_FR@k6_1y6LWk50k`C{23L27FG6aV&%3eJ9lT${Zupl&C+`_AmC6(q{{G20Jt_
zNmE@s$|%g1sQ*obgB-naa!uG_T*Jy<E29L!8`otJ5;uvdM4r@<9C>yg2uwxP(Qo%n
zd@_MjU|(05(dK&dUACbID7ud`#~1z(7gMy?GxG+!UP9QtOjDK)sX3nZ&SK8O<++MX
zw|3h%k@uLb%(FgioW?j@+<b~t?_7jY*^KzvP`;1BdxADW!gZQdqMj`{MzM}hY8n}~
z{)Oz!YD%@4Pk(xBCwkj&?`rOIgd!ZO>C4IZeqsih@?*zEU#DtU16H#^(O`q5XcB&#
z(|+x>h1|On&K0CRO}i`?6Z82ow6^2rO5!ndc=sh!pFD}_z<Zgsz-O!blePdC&A$8?
zuX8K6x=Vt1g~dd2Aby@a(CVB^PE0c7lAA3Gm_4U^k(pC>#OY<wC83gS7aSrl{;|~5
zihb@tbk)Ocg>o&>z*pC7Fz33R<X)Cr<>yE%F+At6*jQv11*`0?jgi#f`NjQYuNaAL
z&mT<Q`m6H2MpiO?+&NodilxpLRJ`ZVUFQ2D1G{N&w}OoD=sH&b#N&J^#Tg3wAd1;?
zLH~oIgG^g?Y4d4>ux`UF#*w?V1>T!tjj{ud%@sTyi9w!)h~`*k*Pwq-ra4e_Y9b@=
zK0}nUuAcnunCcdeus$%T8^kn+_&T0rXp9S^O=}mtDSYtQmgS;whsWV;o(1+6pf(pQ
zN%uN<w||bf|IVQ_woe=0XP5ouQX7f*oScxsim}53AAHW%-vtZNRPviP#qfk6{UX5N
zMVlG|kE9osNj5?Ps@b6XWJG&C<%#<{P%=*{zd}+lrdWL2>JQ?>j@beUB`}<baqN_x
z8j0|?kf<KW4O46*p_!_aLMVG%5LIMr4xKQrWkj61%i&<L{-UR3Bxrt?L9ZVfM&rzW
zvF}{bILnUz^Dw41Wv3>GN-L!P9{F~3T{&bO*0=0bGT&tXf`C}EVxrzEWzlz&rZ`ha
zr%v^$SMseN?nVc$K?#)%at~*b5QJ@<O2s-0DNPaPFwP>fUcF39RsP(7Qn%7gfs6*V
z6NU4(7{NT@jVpA{IA!auk8^9-sPS>a?wd|HItKQHu)QlDg(r=I7TR0&B;lD>ObFzd
zQ#j^c4-X;_PMvRaLaQC$$bAG3RL&`XG+wxy?i(4yS|8$P&oSCBp={N=WX*E(9PT7T
zTOShK7@px3lqa`4j%FVCkvgf(tE2m_Uiqo|BWI}JgeH}}R^HRuT+4SUFo7M%hjF_Q
zQ(CO;+?imd8M~hcW}db<L$HJ;B3T1dsY884`jR#aUq7B@=_A316_O8L{?=@6N2u!w
zIS=)3W-^2F&N0w7<>x0&d;F|=*6#3i8TcG?p^7Dc#K+|d<BFQ4>ac;8ImKm0{OcFj
zNpZdm{m(IFbhAl81>)#LYsfw^&x~KVjMFn>*jWLhy6u3GT{3rl`Fia@^l&|++WMg>
z>X5bGK-b*scbHl$+|(RJ-14`6ikQxuY-nq_XFJupXVPVIS1{#{<f%N-eyT70a^%F0
z@a%(wT=|N`O;#4|{oz1~vJ?><+BUK<$>)}Cb2n*?Inp84Swapr-emrAwwbTTH{<M=
z5G-pV)31`lw|;E9c83ZdF5+gKpKCFtkzX`ChgI~#o|n_hwApVy+;y$#rBg%G^Z#9G
z_J<5CMF^1<2t`%>S!jIWL$RH@t9I&9Un3^3F~~W<-MLLfB^Xcnimex6RN5;a22$p{
zGrfPwZBb<^*=3^s9P7~5QdC@cRCzGLbr3}Ag2_yJs3XX0poSEWrB*l^V53ad^%!DO
zi#V+ndmB}3W#jO#Mk>L~&3T`3LpYxK!v&cfC@hh{hKV#Gg$oOc^2FrOBq=<2axLqZ
ziWvXa*doiby!Jd88q$pnJi3|eWhmNUgc7-gYTT?zIwFp!7l9nyyYeH0S-1m5^y_jk
z+h&jv2Vd~14{22N!3CR^i^gZut~6*lUMP9E8I9gT1@l~uqpTxuco*Ctqhzd3Bh@~8
zbxeh%j}1ThL{;!~J9^+jA8omU<%GddYDDnYP$hd*#8}`ds$3#}ihz;@t1}JEs3#Kr
zcW4ylXI?pz4-UVxv*1(AENVJo(;$b=Vp51kc90Ge1PNq~`SmqmZU!>rJbkerHT_sk
z%!!@Dl@rF!K|uNwqqy7@hqfPQZ)15W-<4S?J1@Rgf3~bR*6f^fAwXLO?WY#a7N1C&
zX!15VhEnrwh*yi^s>WFrY%cDst_mM6p`dzebb)?z`998TJHV2H$@myM;q}0@IlGgb
zo}9T1Y_5v8lHZuy`fnjgYtQ>Funz07g#_wMMLl$%D(}XuOXF0PmgXpefA*>6+9?g7
z+Ez^V3NY;--B9s5**f^K@VF;#RNX*>(>T{*@|rxa82F$SC7W4Muf0y1g;LKRo0z*|
zBe&rtIpJ<$WFYo7ei{|6Q|_UqP!M#=9&jK*FJfQ*TIt&o>@TJ<p$%SCHWXKT8n*qk
zJ1b4z=p1y8#MkJHEI92(p_HFwG*WZxQa^b-XVd@rW8&*z>ex$XI8Hx%9QHW9F+o6A
z@QiNgph6LUw0M8MQc>x_=%{z(d3=Y?R9nU3`Nf|b_hlW#pdjj8@h&p<nLg-r4-lr@
zd$9k&h}D^gzb)}xr?!!|-x#A^U6fSLy0J`}{>*4clJp>!gvXe@qc5@?mm+()9lbk4
ztl+bx=&LY8+QmuqcZ0JtVO-{tZkNiT!9NCP-w;66q|#;jjHgwU1)XiRCa32Yj;4|&
zp5#0mKpHQ*yUj$Fah(#PGQ<M-HibbU(NXtYF6-c=%Yz%IU*<qSiW#l&xQ9!v6)SG}
z0wV_nDu`{f?)-#N?qPxn;^4J9!UmQc%@W6P1sdrTW=j7H?@%HfnJ$5WT!ZQ;)nK+I
z_7PU23vCYZ#}p{EL+?(IYU2r58NRrp_*cj%Ta9g%$SBXs(-JumXP&+ABAQRhjG*=M
zj5IN7{058C;a?qJGT$5DS19q%?E-P#GrEs@Yp8Pc1s38bGdp?&n*HV-FRbuHgo@#p
zgR6xPUbwu|^GM+j*1f^Pzf-%ylN=a#)by;@se87ZhQXKF4*A!h>KLj?s`P$58V-s(
z!f{%A#y^GBRNT5nYb<E_7`YE4si4uW!o8!$5#+%5Urd<c8I&Sm$+;1FK1iHqt7PAH
zqylaxi?ZE4m0%*{uhh;h4i}E3GPO}?D>V+#&_B_(nfM;vfTh6+Hqa@K1>F_7_~j*~
z@+&#^sXEUnIc>tgcda`1flEhCc}s+vj??s&v&np>XIrxQ_ECSp{ZowWFH9!7Z&a9t
zJuXOGcP^1Kz?k2pJ%FTBnCfm$rsOR16bX)T?%O6i;S}8nw<|Jh^|i5@{4k}z7bEeP
zxLF?&haCHoi-|R&eL=MaM*eP`y(D)2!bfccnn*^XUXa{4aNc*LU|@aGxi7&+fkXH5
z^2@~r9iYgIDX|}R^mmLO9wZ-}#Gmv!+9b~JpGv{l$~dZ>W;8C;BQ@E{3yU!yS0cqF
zA}1`Eqs4(x%S8v9-+Z;`qyBDObWSQ&N{xf^tW#0!U&BdP*bZ$xbyOz*)|qotQ=ce+
z3wYz~5ISSBk3&Snrzt;P5GH%2B+K#gv-t$E4`q5?Dp$h^%HSV-Pu|aMiyRCQvo+kM
zKNo$1-qcvR@Ppp!)JJxW$MQd^YCJ2zl@%AvY=4hE!J_5*7~%&m{+8n|?egoEpQ`oY
z|2z~F8E(EuRffH=*}B1G_4v@gY_XaHvYM4{phZmS<Hx8=%1BDI0=OkhDgx=)N?ty)
z8M=#h6gm3sv!hsi6w{{<V<N_N#LlNFS`BhqNLo!HxI;wM^tQTeYp~iOT*ugvejUAs
zfn6h&Uaeq{IU(mfB~NH}HN|1Hs)$9=C{p|}yEK9JrH1Y9EZuW{k}v%JdS-u(a((cR
zPmT!@(r!9&{p(K~KwSc{(cb}qeh%=J+^WxerHPF*lwAAJo4-F(+b%McQHSABGIKN|
zV+j7iGUmgi8D&IrI###!gNdKd1My3~Gd*vk3hajmSR3)wJco*hDL>WyC?KC7ERF?^
z9;d$obL=S<#=x2;*5#6IP<aQMup^giPstIKC^mbFnSG|U8Yp;wK;QW{^Oi3e1@g%I
zjUE}^vz7VlLErKvM21%v*$sZ4XbjVc*qV%zYa_8uyBP5ugqN=iOK6WL|M{!RTTwsI
zW+n74iMRM)CI<h@p!&_kLnJ=F`D4fb`ZkZ{#kqy;2F~V7%zQ9RyGUV$%Zx9u?Y{}b
zt0z-Exc=oI|J|4NvsQjUSZ)1PKa>AozVyE?o4^0KCR73bJze_S{_B6gDF4IX{{P?q
z|BdJGM;8jSg7!}(Pk(|EsZgOXF@oDqkJx8#O-rZhera!}jt1m_73zKRc=?Y#_|MJw
zZy$tSfj68wVKM&K@pwaq5Hcex`E`q~A~ySqE5uiR50{@{V!&yeQow}zs&f9pN$@*!
zsK)S%Tu82tj$k?$13_V*Sf6if#fg}H@02@uW>D$*WA7?88F-CVw?O9Hofp~@SrEQb
zu?g%3@#J72=*h83{!Wzcfh3MQFskonF;36<nvO%|+L?wduu8u_Tl<u21vT|)YdoF(
z^gX@ycrg{2de3b)P!Nqw(-Y|YTy-jNOTFLj&i-FNy*%On`tc{~lo}zO<hF=E9|<+@
zp;t#5<Z5%kR2=(Em`}Sp$EIKsq?|D@K>E?JxgIg$<(vvo3E%nr?DSob1Sh(mR#<)>
zvl?d_84XYW_mEINQ|P6hs;6DNjrB->vb?*6Plq#d1(fsJ&%wHME?*eQ+)CG0%HnIV
zX;}n-Bx>6-b2V1)ZphV_vY1>~+MDqpcWwo-o<>w8tDFHuCLW-gD#$@r_E?~(uTK9B
zGyDBvP@~-RZ*`p?KxWYMcW@7Z58n*tqAujsiG;}b$KUng5|ev`W)%c>lrsD)#7B~d
zX#H|I-4?}lViS7C%Slo$$gq%W9OZPP=-__DhTbbR8FZ+q(E96&%JPf9T_#2(zmLJ8
zEnBJIhlu<do3Kd(Dqh;n{zKj-+tDxsN_mihUn6Y2`ITFKRm^$nDvjo?SgBZR%u}QD
z_Cu0CyV8PQKy=t$@MtjDo))}(5n#>ekQ2x}hgSM;?+i<W8p9Y8wT}{N2LK~<4WJ&9
zd?~G&lR-8HD3W^Oy62yqyI)^E?Po|mf9mx|?Q;Me#CooRQvX@`LYXQOqT!^iNENh9
zM{nfq0JgE~Ivg@EC&IQ;`DB0>YPj4j7V5G`7<;G7->KSO;5p;B{FVv)SaFz>8KTeX
zc7q2EGWnp5`lv%X4Ep>@^Wr+$1H=fzc#p&OWUjrDw-hwzKl)6g^o{#-{$->5Z%geb
z6t=3<wS0e`h^NjIf)^^VCX|6NKuY87b#e92!0vE7<yDJOJa+!|_g^pGE9jTIKKH;+
zk!3y?Y2bWJtQ!0YTWHOl)3CZdo2@G0mbPr2c+f*B?DDJkGB~{xc3IEVKLgq+=>-_W
za*2IFZ(ITdBD%<&mRucqLJP3Ys>$vw@i%;t+uuw>tYmk4)m3y{9run(AQ7@nbGikV
z#&|-qN5`1Yz_8|L5MOW+CQwj#O@Yd!{q!+a4zw@HdAa>mDmOjj#R@=*ufFb2S7nf!
zID?5Aj69Q-nVkc6aruO^Rvd8_+S$jCa4rG!I&fklV0cr6lz%%5UJT(z4;%wB0$lWu
zV$`Dl=Ax1BhvF-HuIw2F$M1XXlm%hWn`=+Of}mz!J}+?|A}-B9QTz<X?rDvEES4fh
zoi+g3^pRcanatPPBj5wM3yBA<ImY!jaqZOa3~F_5)T}a;{W`Nt=6XMOfPj4&G!{l*
z{g_5b$m!#_Z=tYMMMHO0J)p&!lj)HQVs?#_fd}ZyggY?0HVfklpM$%gVFHo!`?U_)
zvWLj*ZaMqxDyL~reja7G{6g!SE5BaJ4d8^tj_f{RBtGaQ!tKCF6<->b$si^M{d>__
zI*uBh|J&;L*P{uFqND$)6vWxbJd$4kIc*+6Za-`^OrA>f;v8sF@8}Z5){Xk?WTouC
z!-E*{H07B`fp9=wX-^QZx7RtjvIz(49Gc$Y@wo|2eV5zLv)F6{%zHSInx>6&$vew$
zl)Kv1vJwv2rzS($axTN*9r=*cBa&yO_kP~of>)M#j8eNU!a56E`1*R1l;+{jpkKQo
zR5FAoEg9c|bGj{$afo?)&@Zp+jnSni>8(s_woo{IWt~v%t$zGq{Q=o8lzq-)%}c=?
zMhYGsD^&a}ej;(HLh9&qf-wA~GBal2MyJB|B%}eTmD(t?FqY4h8_6G}8qAmy_lkYf
z7Es!x;ctFU(i&d7Y(clY_moT3u;}I`>D^8S&&bknAC+n(8+ZjI%DSvDRQ4dJWPN_H
z6-vAqaj)Qa#y=>=p+0>AAzYWiFvDL8`cr`gP{zjKvUCSM-uOOgsekEKg(os$*Cwow
zgFEqS8wcYyBYwF3c(_&@Q}N;LR<h*S`d!D_@@l(%Obsimj8O5`xC3hwQ4<MHHFLN>
z6Kld;KTu}b#R20;^@$0c+R}~W4YViy2_{kWrcU<=yskS|JN_s+yYEwn??@L|*PoUc
zjIf&<XT;|WmBJ5ot7Y)V&&a@VyA9hXD%$u#2)Hu?XZzhTKhi6$F#_opoyqUTPmSh%
zVF~|M;^kEJLxlNnOEvkgl(1Q>CG%<jJYj;b9hx9U<+82RlNy!GBU<J=a5dZo%|KH}
zb)M)0Y1QC6OI^oZ9?iTqZH|1^So`Fp5>b2|oEoQKM9XVrBsq`mE<nI-b|SJ9^`9w{
zgHWWb8})?z_TwJJC}Jah@&yvPgI(u%S?@9(#eiAU;KMLy4(0b;24%EM;D20O$iHL-
z5d9pBXqJ!C>ZCGwuppp5`<I|O(jcucakKLWkyfi?2ofOld9FX-?9&r&O;%9xs7-ey
znSoY4o2^LoNqPo(B%-1mf)n^I?9400_QSiyE3!e7Fb$n-1(x||z@YgY(M+^<-=FGK
z5%PQ&hq(NB5z|L70EG1}<hI1+tN?0SypNa~NK6D#Jf0g1b~0iia^&+-PLoajCAEDQ
zM{<t1fqGn%vZpqQo1aVJK~TY%s8P)!=i0cC%j|WM!QN2T+@18huTJ0a76@t6cSgO2
zt~dGjJSHk0yAtD1t)6g3&r)n3AU;AfqAsciMeps^Xt=~J+D~+{F*ihOa|+8LQfelf
zu~3gN(Y!#em8I#Ii@Vwt?zvggl+xlhxVEV^xbf}g8}u5bG^>5Q^REOK9&_f_I2F6)
z=RE_qD^+nCuS2-^_x0LhCU*~^!lQtLYVnc$l5&A4|5AU22c9-<>=Wc~kW7+Jn3qfs
zYh<C2bc@<MdB8z_PPy>#%v_`6^Wkb=UoRH4@o<3IRqc;>(o9pt7&#vn=UWoxO*OgB
zQD_^q#yq<ZhphRI<hn_P;aEIG548BK9%3k-(e0%&#qiTg71ghVG6ih}0FHkXT+7T>
zz3-w6k855u_NJ9%X*tJL$S|$T((`xc`ysMh6a+_hKRgmsW&fjb@#`aI)AOe!_vg`$
zsB4~wXic2#ZWT;Epp$*wEGYcyD3Y69I+xd)S+%K6k>M7vbRWrYPxsnbIIhYf>6jAI
z@**W0%=PB!hdIs%CmIzVHVK6KC~yjZR;`w2r#0qSBo#J^-XM+bI4dzWCqc}2|Gr<`
z0w9zk84vf{*cKk@UPm2$Oz(Y&z~7nRNNp{2Z?<)<nDxH;OjK=pUVa`bLn?>oun33j
zK$_6F0<A(2T&sqF<(#GzOw9rls&xWgEtk&r;3_cTaVb>(DocFexD2?jC$G;gA8}%Z
zPM94TXW9bb?01MX&t49gy@fh|Ly%Z?EM)&tFj3w0V4jEU*L@+DR-A`g@jNuE<??Jq
zRJXrub2Qxq?_(P`e}fNPq#5$39$JV7Dg-!D3cWg6z(8Q48jLY2q~_?gWhaX<-eTyE
z%(;Ig#HQy>M9pUw;;x*Epa)xz)(B0IaeRh4`Hg*kE~46u4{J`~Pih`O;)qfH(MKhE
z$h~IvU0P0<r>jaQ-#ngMyhI~O+U;XjrFL<S@6{x!+f!}vrm0R#sQhH?bv%*HnTBye
zBlIE8Ir374^Bw4d<EDi82Azibku4xq2DH(p-Gw$<G+%0ev1VcwhDh1<g)>2gA_~zj
ziM;z>h$U+E*Voo{`AiMF!6I=#gV8E)h$8}jZf;c_NTS)+^>*V3Ch1dkD^!hG(yB^)
zN4+=VTa8M7G!3Br_?jeex^p%&lPN7i9evj`YzY~R{8&+A%nkFkn*@hOLUOVdf)gAr
zSqpU(%e>+I+Xnd^2jMoK`#*6OKWg3-)=><dfXA7(rQS1IABwDhG--5&yk*VPa7w?Q
z%9>G0FMy@hQ!Z?qeD7tbMxsi$a$_SeIn+?5a8+a6KszAxPz6+T4H|yQZwU$1NVJm6
zTnk@YC$(sZ1iOJ5@HOVcgc9t%b$oAG<XieAnbk14cP%v<>=ADTkBx_Y)e!MNC`yt8
z05%y*i0}RI`0YgEk&WA@W|&=i)ssW*WD57`5=dD>Te}T&urvkRNP6D>%%y4T^Ia?Q
z<sv+{Z&3a8h-LJCh1(o5>qrZo&eU^ho_X)0swR+?j!7frQ`499JoEWVm>YxBj$#%I
z*6a4w>Acljk`v?DM*jDbFW!BXoz)6Y(<AUk#~7PpzU43Y^vX=+-TSc}r9M06E3r^|
zBt)h#XZn4K*z>Q5)W-=4a5y8_6B7E}_M=Hhd5VHbSgTvoWe&8KM5OvODULdvlg`BB
z%*hptAMeG*)w13XY~GHx9l*G$ci@Rk6$R|e-b~5F4DQN5yiJ=lkS%=uRvXsXkeukD
z$AeYtMxi;^4#{xU!aKi=t6OPraV@-G8rGQ2dv*C%n}v&fQm$b}jhsNzksp;v5~7Av
z{#%9XYFTt|hNGm5dPKuKUeO0&3Xl96XRafu2pZETa6UVyUzbkK@^m2V;nqsR-*!7G
zCB$P>kN$~hr8JO@nM%pt%b<HzkRXUHXpuEgp;AgWyV1ySIkcPYUclocz6+DZ*hYMB
zDApj$zXq)!Q<yw`fjCC(215Z3ZQ2z6DQ*{x>8orK4>xPSwsJQWVx#eFuN6nlj%?;+
z;|^qcmw}QbzZS!#8r&jW)mxbRHYQ%7wUy2T0s-TKFA30{=z;FwhbPyn&n<$dG{Ywy
zYY2|d`V2WW$G_v#Hlp5JhB{4?()=CCCAF-J@$VXh2m>!^Wt}C-pz=%#5Eis%angPy
zp}?7jW7m)kY)EtH71bXok!O17)W6d7f3f%8@mTNw|9C`JaY+{{xrmUJk?c*95m}d2
zMif#;vNxGU+1Zzr9kMlSG7{M$BO@YvkMHAkPM!Dpyx-^D=l8qazJGrIIp;d+*13AU
zp5yVjKkmaohb+@Vv1Y9EC;@j(XFxq6J=vR35lMz;H7?JzWo4Irt4MteRy24bw|Cey
zM|7I>sBX;Wg$PTk5e15MXQh``A37N;b&YGFBFjv?0lkrFrsC=-?Yt+@iO!;)6SHa9
z`|h_spCk%Vrv<Js&aTWLRR}dM-n+Ry2=~d8isZUJRWs^kf**WC6%NZ?H7TnBlA?i;
zjD9O*O}p#0AK;DEE}7H)z9L#ARvXa~TjPxT`C^w?Y@pMOr#+%QW3=0{|Kg5B8}2jM
z$%Z?6)L*96iG(8IopXM3+8qV)d@nO@?3mB$*}B#5{hyEl+_EKooqfI~x_&FQf@1CI
z1?v;+mlkLceq{jZj*%fbu4Zp{+jVA4a5lMOwOGuEk(;y5?0Boj5UIX1N(y~~;kHIh
z`7m(Q)NWBf=86<nANcw{f<HQwJ62_cQaN1CQz`*JHM}BL*^ENT9p#jjJscTMC8HUt
z;!v4zIcAQTWb_Jbz5Ld$=nRLZNQ&|O(`tcL6u2>Eys^`Pt^wqy^cWS(+rwlCJ^XV7
z&?9@2xMPz<!>8yyHjH0AA~QCWV9cwcFq3j4opnJ-NdBiG1XK8A464SQ6t&TvXE^A~
zAfaDXiMM9xkePHj`p@StXXrjS{4rfArpQtB^^`aa)=OdxARrYYdQ8qe(qy2D_)Yr;
z*Sm;!l$wu+*NwlhNH%r+8EsX~6h;7uDA5nJGA5vAXMr5aoQ!O-Z(W%h{@dxtQZf1R
z{?0f|GLw!H;Z;7zc>{G07hH{my1LJ_f~$JzeO!-|oMHcaz**%9b(U6puTQmNzE$3z
zW4A=}-qeec8rlpd-I2>(DbM(rrZ+@cj5A%8*)>pe478h04Y(YC!CG!;r(*hJTJUjs
z!(+jujdJyZ=*~K0T}J+gY6X57-Qw!i&%VR8Ypj^XtUAOwD5TAM@!CXY)CCI|p@@MV
zzcW=f;NwlesaPOSgPLbG;Mvx}i7Z-Kptv#8zKDIvE{8@ZH9-r)4r0njWiF3S3aY9K
zjS_eY$h5z#oEK4Fe}d~l%gghmTG+{7Y^&`qKt;#R94#h~TQc6cPJsIm2HJc*O&bn&
zpAUg*T#f@fYJx8;ESlXtGMe<e?4D+bE*9OtSuwJ+5#L(lPbNRq(dhD{DppfxwS+$F
zW5=3(q?f=-6H{1`0M!pNz;d=KjlSq+Yv+EyEWh=(Z{d^6>Wpp@RIzDiE?M3Vh@s}X
zzq7Hp#*K?URA2ml!fRj49x6KB9mJ41Hv913o74aN%)tURoOY|n&Fnuh^DIhlK+A?D
zd;~2z<+@H8-fCmq8w8~4D;T8lRO{cu?JH~*@5U{)thpt)+R9m2b|^4tR}<YS4!=34
z<@g#m&#Ll$9egsI#n(TZnnX8)%@*L>T|UuLnsHni*+$1j@82f$h7w?i)v~w=1}rmQ
zsmO)DhO0?q>7uLcx`SgftPH8rrSBC0NOvegv)@4zPlK*Q<?1dT-Xzq<6JQW`%)!~x
zBosavbUjTC((bCyma6l%P1Rny(ZW^Q`ktS}*x=92mhx+}p&UHwC^w?Ecc95y(2eq|
zqDVJ&%i6w06(b0BJNtR79!~X^IzG!THY(wpS!e2OP}bSALpeXhQHM=DMPcK}bB*fW
zSA009?9d`eFR6O#%iA|n&oZd5$zp25s1q)g#=6?ynxIwuXOQyeQ&)2!Fscsw*Bx}8
zq{`hSj2;}1S7%+>jh08WxxfE_yq#suX>;!DuZq9I1;XE3Vske{63uL;l(g|iW~jO5
zNLKJ{tNZNAw@>^%N?b#7b{*N#-F%bi`v<vQwAe<#Jw?a3l<Us7&h#X`E&fuMtwjm0
zQ=d;{$xGf{neMPy^fx%x%$LK7meX?|Z9wJuk{y|T0VXKr9~<Y&Gq^7$>gyJZ#1t45
z<7j4Zf|+}2N+AtbxHz3@eWAlyK{^GBJLrxtfaXR!f`6xz4cU>4n|{m&ju+dQ$Xk@H
zgY^QGH+lBX;X}FbIVFpYjYFdB3AufT{`_*98qFIx-Qq-2G$)A+mb0+j8Uh@3tfM%Z
zxVJ!&o_wy8J3KIuVIuYe>@5}$4~m35$5EUP;eJx^5st2QOe(&We)8ypKL2M8B`q&M
zGR@3SUhk_sH8yfbCzsa6vP0m9$&<d2W$sXt-_3u#2(tXVG!@Ts`u(aHQW)?NFouW{
zpTMoZ262SLRCEvBF1bNZV7bMQzn!g~_P`N3+M6oLD=H2uGnuP}P5m-nhf=%1vOj!L
z*CD{t<WjLu1n<NBLv`BfgdUcwjdjE`&7HDe2oBEJ0Cf4BmWa@!OuO_!QpGf|VyqSx
zNqc7=^pQ=D9z{2IivN(5iG5&;x2X1fY3Ez*hYp1%1w(mP<e+CR2$X!6f&W|xSM+Ra
z9H63t%@=uIZx9f95Ri);$fn+ZOL7;eJ2H0xv|jGI8u2buw><*2C-=wuco_7r_n|=N
zeZ9P$f^<iWs#`jA2M{r5OD>_8?j%f<Rwvg&;lvbO>#C#8HLNTl>KKBS-kR$TV1#y)
zWZsyWlss$kVw3NqM3j369uET8IaX}d`=PGe!`S1h>EQKij8h2wT?%Xurx0qc=B*8I
zu$;`SA4|X~+<4M;EumQE_5ZRvCIDj1w)@H%{_DPHI-qm`5c7quhk1f9L#607#?y8*
za(^KoLd|j-C~(O#@_J76+O`Q*6x54jdirG9S4zr&ztn;lBfkdhSqJ5ndSv4gMFAdx
z#@+!X-~29qzyy8(j<38O9ySH`2O3m8)y+o;JOD(aD2_aIs4PH|ct;K6IQOVOhgSMn
z3OC^!#Llrcz&VgEq5G6W>doj)hALgp-PMQuokX?J?yI!gAQ;64L*GlMNqB_(<iLY=
zzZl};UwF8OsF#gs90wzI<I?5eN|5?1`vtk35U{+VJ-g(>6V4XwMdF@5e>``7CaIz(
zR;K3?|BWWptC0W+@&|AwEn9#AZ9xUnNM@RJIh2%wR)27#K9+<G^}K9x>QU1{PDW-2
zTOsok$+pXve9W+eYzt&*SDg&rF4sOJ!paW9tit**A)y-g5zatEx5fHYEA#SJiqO0J
zVl~199>%Ux;Z!c!22+h7rTA(MV)l}rDj$xu^!zm<1<pD<T0vChcw#MPL2Qi&d#b@B
ziDM&o;}n<xT!U!uxlMz1=@Fe5s;36;8h_h>OT5$(FK~+SyX;$3!)W4xSv4MJffC=%
zN&6#<jKgM61@F!yoVyDbYKLh))A@%V-L8fUB4W|-V^n<CweL&gkAH2({}^Q<v*(oa
z#5sPJ)BBv}|Efh8@h!_$kik)m;r%!<&?56WFWu+}`-a05rNfXa^Qp;6ooovnCU^oM
z%P0h826tXo_BE<vS{8P4<#erC0~|hcr8W%?hbf(a>sEQ_WiVCoQD(!6dygNT5)JkU
zr$J|93)qvw9lyBi*mw>US>R;YzI`sY*Na+VDw`&9LsApc9cFuQCK9?Ox8pLpF%XyV
zdb-g~y&xgumO6OhFtgW0Xy@t$>9H5NY5Ex{uhv!=W_M%yVQ=Tq@A*`j1A0d7;QT#I
zH8DF@O_P9eX6=e*AOm2m#j8hV-WgTn3xj!#9M{|>@Mx8n><iJOR~8SUT8XQ$DwdEJ
zMJ$c_fq2P9VI%cL5_E=q0G_Fbc;V*V2px_*5CVY9yB~?_9iN6~<ybM_)Vsv>^58(Z
z`umwHObU-8-@0!<&s!EpbdJwpj)B83%?a~DH@qdi6#61q5*Fqo&5r@ZBVis_m7;Ja
zy3PQ~$@@FuxL;~Gvvy-F%`_T9kDHEP<f7F}#)ZT2)ISqozcLGuUdX4Sl)P8gx3{mi
zV4S~FB8HE*RW}+<o-ZC}4SR-tE-<rCbIZ3P`dI5b#ABF~Wc)O$$<ZTYq_0QQ;R+X=
zAy0tbs@Ua{m8+X4gub|w?EA|Lz%i&IESmh3NMOYdb*S<F)1Z43Jd&EsW>$m^DUumj
zeN8W`xwS^5et9Zj6&=hzwdw5`0%IkyZtd^!?88*(?esCe#3Jj#p0(@6bevd6-ggD>
zOej8K&Jg$TB(x`r3TjYiI;{|{Vf?_#DMYVVWvJzzPeHIp&C{&Ychot&*5=d3+|_lK
z`R)#jxgHzeLiAwaDpyS_Wg492j9$v^f|C=oFj6)6Kg9WHR7=cw-u#G>ZGtk&{D+9s
z^;%aoz=3Wj0h}#Z^z7N%*vHo6|I6Wc9>^BxBjRK%KasXp0thKOu5M+CZ|oy^m|jds
zCH#5Mta9jc?*5(kHy}O-<xqMP$Llx_pslj|`G5>OlQcJpvC63kAsL(QO2@m-VCJD5
znL$mM`^Yl&N*C@)Mz#o|H6p|vBi7w9-rLz+V>IAej=e>(&R8O=7GjH|&iD#LM*;O?
zV_k(|K}_8)6x=GGi736V%b5w{?~jac0Y0Z(MpG75;TJdx6!L-{#&GYAqb3K54>Bri
zuKU;lK>L7nC!B<U38`JXwYW~j(Nf|FR?_s&o%hxQ_ru&bD47SV8g$W_52k}EU|*cI
z@}%YWu=SQS%(D;NWGJeO_4{~mZkbHMPW;FG6o-+|0v<k9{yphjv1agEetdMw>&(C*
z#tbslP2Alb#-P12=<g<8FMr_oZh-fn(0ncTDzwO+(p{qsrtIEHQUpY!mU#t1H$O_=
z8i(2XMEhxNdc{KJ?8s|sBQ(rMe~5h9#FM6y`@Dxb_Ce;719a}s8VqgP?zL;LZ+dUO
zywgEwWi2rZ_APwU&y@550y*7|#L)K{S@$DOk=O05+wc2jyajyIM}XXwVQm?Du>tx+
z#&7frqu_{A1TehN%yb=LlDSIC?f1>{2leh87xqrpD~&)xm*o6WZK*0w^j2R13%`3>
zbwBLXjx#fF$?oQa>!UZe9}<Ku-d9_}7POG$u965|PaPTd!gvyR5{&XGS)RMUxd?oj
z-^{DjPQaDR@zBGF>nFb_1s~6CdX+39{D8Kd59Y10Fgf+p9Xu0W?1DWWCIGP;afXqc
ziW{N#5QIs6yQA}s(xJN1DeyhRN8d9N4wSdX4s;zq?X4_56ct>NL502K>mZ8#;jJYT
zh|ied)6I@ukyX>FV#%>{uD*JT^X(>~@qE%g@nbABe@oOA(q)8;Ue{@ftYkK1<My09
zg@qK}N%tLi%)m)5{n_h6rjia-u&~m->Z#8TbkR)*9Rk9SQBzbGqok9vLn`FYOH=01
z$|xUfk&BZ<&&TDP!YrZa)b6)rCR7}(b$W}58oM(&gXIUvJWyBt{DpS$co_*!<~Nkp
zQ|mu73o8$`jj{M(oM0qDK={TYZf75VoXYy}10iS4LrOZ5q&sv$Hy&+%pvcKZ^euX+
zv@x{F+*Jc-8(6k5jAiB(_vW#QQ$(EZSuNP!$Q%XY+5`t(W=>coyR^6)-|O>PXbj#J
zR>SZl+Lwmg;MSN#=#}qND)ASJ-K*}ppVyZ_ts3uRF5HgQeb*M<Gqg;$V?g%^YgIra
zM`GJ>Khn-XFYu!A<37v+x$lL|g9AQq@LY=}jlv9F=ZbgLGne5A(R?8G6NUdT7%3YG
z93fG1g84^(wIk&O3=e#@B+HikD|N81jWg)*Hy9`y@<XRM4{NI`F53Gu6aqpu*@H4E
z=TY)rR0@hdNO1bY#OWkIDFPp@2Q#t<qS+D`j&j$YD;vSDm{!Ty7Lq)QuUmIpNV0RZ
z_B1!asLd={>h(7+tDRIg%9_D)D?Fj&ck2<Bs#-1ypgy^kVNt9<Pr4oAky>Fn4X}xP
zr0eV5E&AXFN!8W>xVPNU{w#8DnnB@|L*!?ubly`xvAL6<cWAL2G?bhQbhdIe<uuW!
za%jM97+in!4*@*%=I|+zqWsnkO?Bc>T3(qXb-{6BRC!G^8<?{@@XLzfCtwygqG~=R
zV7|RMz>%G=DrPexbsWetTWy>=$2QEoYxRHs{mm4Bc2hWe^+Mp!a)$kAteA5lG;58V
zEz$;6sN@5FO8KGF_&H29mS374GZE%Fj2?0ZA0GQsr6uBCPn0L3h|x79(J`Aqe1W_<
z%xOm0sqUOSo<)I&ovSTdA_}}@U!Jh`V#-+v??adkclX)*@<|vQ#~q+HkmRhD+aRM7
zu%SDHi~-MEb~PF7^kk^l4ylT6B})R~AsuFBcHqFFccSdv7vOg&nV(b27CQG248Olu
zVhjjOV1oa3{MX_+MZ5z%27H>FT>a8$AU0?Pu*bNaVvKcz3Fy^TKSD|3SOy*>zou%5
z?k~=8)pdgCbJw{MrGgv4hG&^ozZ0onhm+q7*f__B3isid>nyR>M@D@br=?jB-R^uA
zB(4v~gXZx6vb-6*?HHw?zgb}(l>T+-sihN)+`k89TP7dlyzh1^*i}BtyQHD%B=hrk
zcv8i}^b;-!QBQ@94+{tVfb}PljeZN$R&8od?1<~k%TI78>O2ye$sk<}{j2u;=MUa?
z4DbzvS56Q9=FI)gFR9%}j^j#hGvEK^W&Hic{OxnIH}DN7K8D=>>v{hc-upXq6Z!n_
ze2stpOa7PB1%5Ld5puxDo+~`^`#tjCe)T{0&i}uskV1#x#pG|AF8upn?El+Sh&%-=
zyg5&)pW){OzA5Z)biAl92hE<&Kx-E@(GsHsGz~Mb)8zQ1U8t+2;x(R?74%B%D1*1j
z(GgL3-fuDH{d836{pWJf{~h6$T4jdCvie=H=hxuEgo>8mJah+4lRC3?UiLYo`475D
z?LnZyRV^&!bUgFe-6{ko#EY@Q#MXbK%;|SNsU)K82SimR;~P8Dm1XLe=2eE?3vJ=M
zYJ4Vo14NIH&nQJdE09B<6uiSptD!Di(4f_6dfKuwE{WihQ9W~&WAMzE9|#a+bZ#X;
z=*O}86I7pxcEy|Q{|gBe7Q(-8Ikgf37-7YeVhjvTAym_Bvr%%75tvEAA<f&wAr24(
z60ieQU}wn)IT<%GkbEVRQKu5l@<88KzS60b4vi8^T^4{X5HxB4X9`;Rgv-MsV84Zf
zQdI8HkrT(BJ}*4Ys9e{$tfgaEoN^kH72Sm)qWUWsoW7d!-1^1}-Px-I+lJ$W6Erbq
z(*Ds+vddHLG0^1<BuGV79g98%xOw^OdfeXgV^#)d#94N%vznKm^#*0-xSowhJLaCP
zzktqGy>VRiE%wJT#r4ME*A#2By;scJlM-h}%KmsN6NMDdMSNKQ+cHLeo&p<roag5T
zf;r?*VqISMQ6I4#VVy#NPND&Wep+9V(dK1!Eil%nM-MkiC9n$@g&B-zS+YX5ZezDV
z&M)t>GR=rE$)HL4^kn6GzlL7P?c>p}W#~3!MjT-t!m9m()Zb6eYZo#ctf-Gcg4J^P
zvGOz|O;~z&k&=EC!r>w$+_#z_aP{%_X!tQ(-3K>xt#L*qw!kN0k&sWn?m8#n3{Hok
zxhHkhPiW-WMy<pKc1|*XVtJ#=tXn!0t#i~$#lgzntUHF2KZ2`4RNia_c`n1C59EVE
zm?mqZ`wO;)SgDB?GmNqW!Ij_bTPe8S8;yS*f@f3@m~de!Wsw~934jM8e91+c@{Fq~
zaL;52Onx!o_yGKA$<EsIM%^8L?TjB{p-8aKOCL>6W~)MGC9Y%W5OwdzFf{Gop2$Tl
zHIIWWLpXHaFlhR64j7WnvvCGNMr{T9rFRNox>DGx^={_gKA0RtAVF$I-q&Fxgxh}R
zrEGnfl^1c1H7jd49jj`VBaq(rAuEGQ*ZjFbsAF*RbQrw-I1;UQqPFW{CZRz*&9;M-
z(X8XY`^%<eA;nC0oi6>)bD)ml0EazlM>xcqw~&1x4b|}mHuM>$1rF*!e=?e8F#5d+
z-$#T&q1=|xL%8r|;b6B`U-*iMx>=P{eo8ae>T-UZ*3~nGIY(QaX|5jMmG=JfWg~lb
zZ_$x#PlO}pKEYtU)o->dIs`5V4sFj@#eXi|?gS}1Z;U~IHDCOc^I1}k+@dErKqB=q
zeT&EK4j-%#-f{T`VujgiwgA?}OhN%<qd;U4<*Ed+zt-E`HGPhg+8x6X%^G*XT2JqU
zBOZqv;KcW0gv0#ZLKO_vJUzWCO?S-5-tF&I7N0si%Hj-OxlzytJv~zghfP1d{19dG
z*>fv+PMms0m%u@XMtXO!^zQZ4KG@+}Gu_!bj1u~CpTqjpu`Ja!U!u)gqWftV!#MKm
z%6#JA8C3~wBM0R8aZ}pQ)m*ebyGLRwt9ehjJx6(8&!_vUt2NG!^>jSts9feoDvlgD
z7Zi9VtKA+7^fbk49i1g;t}UzT@E3sv#@5>=8`H2#)l&)*h7UO2=Kvas>tiSbcxNUz
z;s0@myP9{>E;m`du_RmLbDEXuQL7_s6j(QTeI5ngePd~A`(sx3`{*22Db`4w%H4_9
zd4Rn4fmgzY!c`F~)!*m&ks`x)mS}~y=SJD&J$FpckEU1Qd%(SWXPCjU7Vb6~uwQ;v
zC~R5P`$A>;*GUv@+TAMrStnrd-QM$m=-)g;Ve_z%qQN@g9fp_sV;;tHr(-yGA!$^|
z=#l@NHA3v>wcjk;oX#8K?bC_%b$XWp=cv+Am(U7Nf)?rLdz<unP5Vp{H;!|%T8jF#
zFaNrA{&D?YMk<N)k7PN&Dv7U(43DZP^8R2^QH`VuxhX%88V734;awX#?K8w22pEAt
zUvA(d!P4rLp9Yr|OxcLMQ}FRaq0>zl--F*Bd(t41#Id&N1u)huEI!wRt?um_LC3^d
z+6mLwEo9uz$Wi2IAwyaAq-BELadE7k&uPKP%Jh|aTkENL09z*@FXa6CduUUrK^>k$
zXIn?<6V_bK<F;X=6hg6V(~g*qH$RWRb3&N~vEx1rmfLb1=sV_tf!ma#q%F+YGIONx
zWZ6L6Iw#-&C|<<l#rhsP@)hxIEa(O+#q%8$equBRHqw`w@on2741x;*<615_(22Vg
zK?E>IPdn<Q6xJCsh@Q^}Bpy4xz%f}yYISgI>`{>N=Hy9=(lG$7^F2TX<4(BD8PRku
z$N_&W7RJkDlQj4QoUviRn)K9*Dxik#GnmQiy$$S~=>AtXTaE~RHkxJX$u_Qi@E$Jg
zvjAy^abokU_ITO_L05^yYJ7bbp<1{ZpQ+*>I#(3jP<F{O*ZUUH>>f;ttfF;*7Hq1&
z%{#@T0D)4Mgl96+p244$TI}z27o^hoi_Ebm<ik+%vzQ=8Fa%c9NZ3-e#ps>AoL3*7
z0QbIQjn(r33eRM}489s+yYxL0j*PZ=!5GBp=KdFVM`7?f&!Jn{hLfKa<{`<FWM_*c
zpP0mHAAb|bC+X8py)E)Vv>qH6%m9;Elkk!7)iPP(;9N3xGajnZ8wS{^A38M-;tuv~
z$*S|s611hK@+f!(uX*k|ltU42eMGPDMo_K_s?sE}dVql3sc|U)x|R2ED_UPOZoLaG
zf(VGDjXG5u#oNQzYSt>9ql1v+zi0Q$bnT|anx1}t|3*v5`adq}pFd2&B2_lEbF|XW
zCV&tX$h6>_l6(D~2aAVP2kNS{ffvPW{;g<^k+cOkGR`3hDelJ}R9Qn9nLvvY6cjXW
zQVju{7pw=Ga&%vH7P`VPVglS`UxjTVb1@QrhC))ZU(7b+Qkij_Td3o4soVJKNo1%K
zJCFy<x8BIj{e1(9A$^e|ik`a++`zlfBWZxk-C{0Y#32i*@Q^G}DuEi@BguPn16p7y
z$VB!04b(|2<fPk8<fF`fk@-e3cLTS0C1UR$i&o5++C<8?g!=;)ET0h_6bPyGF7z24
zd{rj^KqZy2by3Gf7UBWm1zxQo<oOO#E3v(ear5Kkrj+hZU|CBTZ-!GQ_GM7>VIZNV
zVr*@qCZ*g!+Z&i!Dt+dCeD;gmIO9Ecq`iAzK#l0_Gp+U&;S8;FLeZKUNS*%#joV2o
z_wBV+nls4UJQt7u=7;mVbG(goObFBrW94h$rF0tAXVmr@j8zf7?1fT&;KXCs85Vre
zR{`RXCO45Lrtyt#AU!o`fqrQFxJ&OKHMW<L0o3TufvXhTM3ZcF!;o(f{uvmz4NO?G
z*6t*6S3S=%cOq9pKSk6xnT68Xyh^>jK;MZAIRzfPR5@<|lCv%eK>WQ6!Oc?lUFsbq
z$)U}XY4SS)tF_Cn5ZUs`z2mO?2EcShMVV>j=4jH{DANST;GaaloF5D`o?w=xOcz#+
zXGOghb5`SdU+f=P*yqW+j%KSw&4I|_Ow{aKFO#D);#CkPZUYIR`I`sGv1tsq8u##x
zE52-T4mUCwDTojo6ugt>Ia#zuw0M4ae__wxQJ_^Yka<zL1m&>LHW_=V@KpcOnD~Fc
z$f4IdfwwDF|Er8*R#PEHoc5D~7msGn0V(nB#>}ebNQ{mLd%~h-)mY}oT8+mS{cma}
zjAxz{zCtK~p5J8BtePsm)lq{%&l>pa^^iMQ>)1%lJou!ME$A%yvL`D?xfm!T5C&n<
zo=M1iC6~&DU2jp->B;?e&(!EUoD{`lQ(=!B=!;KP*N_??4<N)M8A3@Fw^lT7k7RZi
z7qMpwT)HAJo219T0^!RXSL($q)@@ULn0`cRs49l&kUhR|QEMbBSYepGWCe_sUuzna
zL}V;8alXvhr3^5`ZU%-RgUe$lUdnKuoWff1u;xL;*z3aJN>#6Pz09SqnQIM8`R19o
z+A5*r;CP1QT;&dDyb3lJD>l+$d>H&h3?1#l?z5;V6>(oOs30-gI5dR0=LA~UnSDB2
zjjWY7Cvo~ZT1BS}6RJj;lwL2Unfjy@E(KTiqsnJAyo+Ao{~)!(>rq5YI$N3nl~03R
zTvfAd3WhyxTX?H#9YzE02F%?U#rzeWypNls?P;1H((>YB`KQmJ-@C8}mU^s|$`g8A
zxY#9<CMhx}_H9pUQhmL;Y_4hm7riA`JUN@0FOchAR(0aR>fvn~+HBXYqVs)B1*)50
zS~Sn!>UFxiOjJB@b2DRA$@5;ra@n>5Z-X4xbgYDl^Gd_-Sh-SKR9=Ln{~|$J%A~~N
zOBQ1%GPP)bfSJd7*My0GSgTZqc7a%NR&<oP!Kf<Fat3DHw|!h>rt!Ml?Ep6E+<8B<
zo0zYQ)u%9}{Qk`-7CCE${5$^np|xW$Mq24oDH6Ol{n1FPGp}1bYf(F<-eR##t;LF^
z@G4oveF}wV8cH5XdgjXMIxV+ZSV&0uG)VAS)%{7l8|U~eJ?o5*YfFsb{-CWjWLN4_
z$Y345K*V>}d0w3ovGF{rnWgT!k0?(>HHr@Q;hFO~NS-gkrC*X$<IcJu`=&eUuKzYS
zzNs=-oAb=`9{GOWpRe~>?}-J!4ecR!y<7%PA2lm*N)fjDM&My|NPW07UCCa;cZuhI
zmsVzsVyxfZ?cTh^ZYJfC7)?*+w!)8_kE!%1kIEVHC-RUlt8txw(1Qi$&<Sc64W@{B
zibyI4Oz0=_z}2qRycc<yZ(ss#X8D}t+IpWU1yw;-{I{ZePJ*=(E<kHZAM4I4ob$Hq
zzF@tpK^ByxH=>;0;&9Y)A37vrD~-?h{U+WE{cdY#SOScqC14w^@biA+CH@Nc)AVWO
zk4HH#%mnBVvn_b*8aBAh)O5y+yOxk#H&P+NBrtB(Dr}{oON`2G{Y9&Ha2}x)5@E9$
z7*EE!8ih?|ywm*HLoF-@*@ul~y6zLqUMFr5;zWq_)K9Bq4DXxoYQ)43LoD-{Fk^Cd
zWd+;$o)11uIqR9Orc_%ST-%x3T1x9su9YdgqEGt}mfwn7Q=p4Bo?zm(JjFLb#Y%5N
z<<R)e9gIVqZ${LK;3efdonZN>5z}^oo-e2~q3Q>Sp<8utVpILllB`a(2Ngr-5(uZ_
z7=nr?q03w7Lce^$Ezx!rQ`d%SdkZo)mRk?1t{D1H<~*j_$ZWYmq1~`VGBei8894;?
zgKh-k5p$f-LvdB%D5vj3wi_P6;P;=+uzF6_ppqmeY`VIb=l$lqMn>pfU<E~k1yRnG
z8a_K1puU>-?85+0;=;)<`F&FqPG4$8!h|RAIaiHU#}tmM3I%Q+vMQgM>^;-s99rpl
z%v9M=fc=hW&C-qEQWZ4?fG3x-*M9wPyH08Zq!-i<wr6Lw1x>BHxpKPl4E4LxEgH_3
zrOJ@0Te(HY1RE7TcMybW02?6WPatzdq01*CBSgM~;zc2Zq3b(?7hPb+nG^4$ug7p3
zBC|I%TA=K<=_#m#s1fm9MwL>Ryp6PTrjSC{)5wB0R^LBpXF1G9iDz@*&_#;p<3ct@
zw~pq2hI|+~!z;lRiHB9F1o)r>*rci9BM(7F)4?xqfOR*8c6!Zfg0THpV2z>BPlENX
zC=V%MG6rv{+lEcp1KXLn0jZPa>b>wk%nWF#T|uV@lU~(mHy1sNJpPaYj4dp%`@KLc
zPU2YOg__q?cIa9#>4Ka;3r^;QRWzxg9M^o1-rK@r;&CS0uH8I3m+4fTjAV61G*?Ex
z$_D_u)l*+()JrNot?T=sZ64`r+k2u_6m9$-5miWJcJr!tM75nmqZEODAWs;9Pi|Q)
zVAY)mQA6qli5V>h@iDqo`}>mzDa056{r5vk_0B@Gme8=2BuHhYDIQ^EKdht#`@POA
zPCJvYGO)tjWG~Z3pB{XpzCqLx+e0Xp`z5l*`af<*bztDM7Min}J-n8G@!<*ok`8$v
z?+P>xdMRknsY-s3a;K{zJwsl-dh5BQRQ~Q&Pa8(a8_LXZFw_50<qA$oB`nX)!beZf
z@2Q9KQuaia%@Mam*!~VPQtO7gG~05FGwkQ+zy#H#FhW$&_2A?kHp_`KR*LG9uPa&?
z2EAY?)8A-P{J>nM6qIj)dO^G$hvevIE_4$x%pc$9B^iALlp)OcZhHls>z)xP6Rih4
zjFzGX+e5%pw8dt$sGO7pNxh?1lQSLkY)IT@<~>W@s~_(W({>F!z4yJZLynv_z>-k4
z>&#0A&%m>e(3T_h`5$llp2x;=DDt+qT2_-4a((yvO_ErPRlPFx;$I15gW{m)8T{it
zh|st}J<s>p9xIx-aFL7RDJ>7@!;ngND=JOdd6o!8gF3OU@BU_^c!RDLL!E|TGVV^H
z{Lt5CNJ_qO`VMg=BQu~QF9~Mf={4n?_fD?p6`(AtFhoRPH|y}cCm@9i#lE1FB?+`z
z>>^(|r9)dNQ>c9tdzvI}-q&^@^^Su)*Tbim+#!G3uC3X{Z-f?S=IJ$Z7orK5KutE_
z=9jX%)N+m{=Q`@bJ5Z_kyjg~D6zXCjHpDY}e=&6GPpBuqq0U+Xn-uyx^?^4#t6De9
z!M~86aH&y(C0n588Hp9&k9-xLk@sn2OOvb4Z!()Y@io{dHE2-!zrc@Ci;11DW;VE6
z@!BSIMzwuEM>K5a?U6i5$IrG3>rT^@d5hY|fHha>Wu<{mbFqBP7is+VF)}}SQKNmO
z{z#IZ>2lYJw291~*RNrAG3={940`Re=V)+wtr^R;;!~2;rUsiWjm}*n1j$#%PH;1i
zz7@TE$udz*J+}OtR!iDs1>GI?0*N~yGtwOj5~X<qsmD)W^jzlGK4q#yD40ZQ+1Uw1
zo=<#S3#T8Q)Ic-zfA+4NRbD>kp1n^Q-@9|QIxJml)Mz-@`=nFO3wEQW{9Zr8vU%;6
z8_}hD3)$h!nR|h?g)-@T81<%|rbr#$aj23$6zA$Kv_N34gpRer3Uw!W=!+<wYr-$f
zBPgE?tF?GsSiA1FC=Vu2AC@h9a=vAzn_a1IjFvR8G9}Oc*w#~h+j;Xx{g7lZFU!uV
z<CN{-=+fwK=|jPaQo*`HEkC&vct-;-LQV7zFy*>RSYfZUvAJL$1#J7b!loj`TwN~8
z4cqE>HFdl(avZ37M3t=`pTO%5AQV(UO!qaF6j?s2msTWIU|RdQKYQ_TL{v?K9(=XO
z(9w`$oZKV`x&!Prao4pKNTE7)C0F<9B@&aDZnB!jGT683+dz}dU!tTz*urPe)7KaU
zs&=7^Nj&XWUvaR$cfdF3wk{{)eOO_!*5zK}2=eZ7oJ^SN2vPNPk;SLifPZOSS_d*i
z-;RGQ3mdzV<!v3dBCkLptA`czYh7pVH<S}qCsy>A6f5VNPv#gh;!6guGZ^TsH`2=)
zSMw5hCsZh}ckO0$GxgDAXkJBXZQ2BxDFy7c`bQxZ<}8d0N)?h-JTnrsFHXfuS{n8=
zlvc!k==aCk*U04ow)`;HYNoSKLn%H<FT40P7+UG%4!S^JocN;Wwte*I71SSv&O+~(
z#766YmH>s{RXS|QOfgt7+3Y6`U5cK7;hGoW*&XS!lzBysUhZMf&!9?S$1eCv)g&L9
zLi+F^Mg%n}a+O~6%N|4YzuGS)b{1rnevH}5?KKfmM|Yh{5QKTc{#?=W|D*z1W>jk3
z1&yM)^C3*mGKsomLz=rZ^~Oes3*mSNx2FxAG2U&!G)h<EI;+MaD>Ogc;!-fwe_Dow
zw4v+xo2Yt@R-e#zS~_AYX`)~pM?i*Mv>YtSB~o%ldHd@I$|Yyas@w%1kEDv;Bl4--
zV%D6sKFo4bAN1(_bZNu5^on@AV?rp*#|FlJ>}M>eX*Ou(TiZ6u@x06EMV*noN5X3b
z$E6tdNi!>Rx$57TcCw96-(6E3>3yb1Mb#=57_u3-`VHJw)+N-W`w_%w{0KEO@5Nol
z%!0Z0S09&QA<PmLwZ9WIx?bak=)Ot;64Tq4IUgi^>@bSCyqu7dp+UyxOZD!E-b}lK
zVd#KJ0UYhF2#J3k+F@6eT(t=6);g%)&W{Q;=*3o@L$+||7kbBQ(>pfpG_$mhZFN`|
zm>}Qg6a%|h3)7Zj5({YTO{jNj8ql3*hA>V1%@VP4qAaQjx3)hWznNe>e%$1~XVp%%
zL2YFfPzCRre)0U0SbDp^Gl$L8!+77$7?hPu;GAhsbzm&uL>&IW*kq>Y<n|puF(or)
zzv7pD+eX+#+Fu3nKi;D1BsqRWE1vlLe@*`)+3<FWx!KtZ2EGLtsXvZDz>ecq+0-#|
z{;Ie<w1$Od0wV+!f@x`H87RZ$x}H7}gMJrBeeLR^<-^OGa*wG4;@RmU5$^&7{E|Mt
zRgKlGW7$&I>B<VtV7Le#%ZxzbKs(T|TB>tJz1Sz1aQN0KFLM7oo4Csg%Av!ubalM&
zHP{PjW6<V79)ZzduC<LhBCgY-934iApMGbkBGtlxc_vGSPhH7|yZI3iF^i@ZY}6eS
zEEd`49|O5DeN3~<)Iqb<yAZU><=>vi%jHrAlRmg=fxtoPlo8N|Nq4nnOk~cBrbC^Q
zd4685-WsP}9~xg#j?9cSG=)sXv_7&-Gv#YhK6GWCRUi5hRPiE;Bd_eVRfV!(TE22{
zJ(PWN@;W%&AyWrXSPb-O#8lu;A5<rkvxA-eUgL`R8Oyj!*C6qNw0j>~NHfUp=Dj~H
zi<ZWo$d}tseOaX^FN02!m;1b_ZV6iYrZUr-WF5vSc0Y8EF?T4P@YSugxmj+0dYG>8
zG-?rBPGq8VKY-DAsM+|TN&1G<JvhXGaUcc%#L1i3Y#6*7S>ZK+6~d6&V)fw&+QDiu
zvNzoKauLd!+>pVb%X4lCGq}_X-R0^XE+prhXGF=y3EoX=`vT{pL6xF>WoA~EKfgL-
zlHtt(M;j{F06l&saT>8|5wEvu1xZG4>V~~tQf{N)4vKg?7{~|VYVt?MpJ;j>Yc7=W
zKacykNg1u9)ljd?&+C~_k!_^H^Zd|ZS_Vtq4_n86V2If*UoASoLDAwzLiIh7RVdHE
zHR3zP*SH@Nv6`X_*A9_y)e>6_RTL+<uk|B!{NOgaxk~iFnk1UO8KkD?*q=l}`apw#
zq(j}Xpm{6ppk#N}^9k>G%l@9FM&YG?fQ@aRQjg6WH|0b&$mxx*;9&LVOT7N;_WFmp
zytfQ8`z98Ukp0;+YF#?uLqbf2BXHw<2j4*>OsdHdywhKh>|7N*$ZCnFh$OA{gXv@E
zJc2Doa~rawI;8`N@QCExEYWQe-^YB#{QTL7Fk}3L&t``zOQnSHQyTzSx5N}8#xGR&
zq%#TtXCsxnhl==0y6F+BATeXE(bb)nA&_9uKd_4o^5opO7ZLmvrMo5jLN!q?cj6h2
zf&OaCwX;27dhr;}C2q}yjOZ87fFH*);)gxTNoPuokI<*Tb!W!}tz9L&AsqB4Zm;_q
zP#!)AMN?nT>=7xhVMT3ZMk!Fa7%5BDrS_12B8c%ghPs$xCV&hff&_-d#eB;mR7DiK
zaJUdv(S}4!`h~omhasXYJI-i3a$d|vwu35FzKLPY(`XLnH>#}(!k>E3fi;uu$;ldr
zzk@1e)`Z?<SoYF`n~Zew62KiU4T`dr)VO4F`)0G~Ww%tTBXtTh`1(lLr3_QiN%yoo
zGZjrD${WyQS{pjx+ZD^1x0h{6hZ0EZ<fUkDD2GOJopO?yCMXalkVj!<zQ54cvTnaR
zn-5KFlVH#uTcjDbk~9%x?5t&GkRU)cE7D^h*lVCx+P^kDO0eqGt|s`Z<YwjhYTJpi
zUyJRAXcfK{Q`y+AoMB5m+mYqNoD6PtZ^3JVb%5eT7hkTG$4nr`GXeTamJ{_9+y+$?
zatmokTJ-~SYdzd39VujW)6cUQ3=S3aZKq@t{nkbe6#(NP+v!E!pSO7iKK@4F;aOL8
z{5nU-`m~;*#45PLbggaR8arPtiA2#4;M<|+$C!9IcclXlDe>ov5m`B0qF|(N79H_A
z$ow4=nD~jMjV34@d{zapUITJm`O+gLNG2QahVt`SH_3*|DPw_4hSn7=ISPSvjJIs@
z9Q>u~Fh*MBJ@*Z~Mr>icaMl|`z`9YEbk=JYCdV9PG~DNHF<2kM99`+q>C(K|vlKL#
zb*B3OjKA{QFxp!A)<DzK0XTEt!RXWRa9Wt3=_u~?^b=AW@(8XVE9RY<FW)?<6cR^}
zrq~*G^`uVTu5F`V%ZrapoH^2T0TT624?Z4skH2!sF42pJ!V0k^Q4WvJc6(PvVcJL-
zs^RC>BfQUV3&IrEdEO6t2Oskdq~zHW;>=nV7c_h4N_wh08FRcCrAFpx#7(Sy&wMbz
zvmd&9L_o(S5SNcB_nJjwS+^X);FP<fhId}i_wYDCi)<fu7N#`1r)+zaTdjyC7}@Rd
zb_3VR_XiHNgrRIJP;@%n^o<}DaGt27(7q^YWKC^ygE)k@cVTaIO&aTZkD6S)-N*}t
zk~~zsO+2i$gZG0u7So5}_ZM=dUE(a}9ld+AU|rhN&`I^@!SRoGt3Q}5y;QSvd-RK=
zpTbOB7Pg35sG2!KfWHwNKPT&?s43=sdBj&dr>Fct^$|||28WKy(}njAXK(X<<5eGW
zPT9*ZBQj?Ee9iQBZ??|4o3HptcI+gY2;`X#*iz+eRv$fKki-`*#Ymw2)iGGcDQb-W
zGm%n(4wm)5SCHOi&@qQ7g}47DV>}jB1*h`YE{!V!>+$(4#7a`&6rlLZ2-`)fZ8Dml
zLR{*<5&Q4}hq#UB;q?Z8KXjZK<Lk?danBm%&X(|6CK{0(t~pm`^?+mA4WH0sQ;q%M
z`Fy%VI?sU*R-zooS8ssZ=IV>S$fX+sUaZ?j4&O3cSHDWc#vS?bSJm;KA2hEcfZmt4
zrfq+zk|GZnym~}0F7i-5S;GApV9FM@6Ha*R{;~n&Q83x%?S75;p};1JuXtovc=}an
zO=jdL7IBu#<+;WVzJtnT6an25G)5gud6n}|K#Ow~G$qXO8D!t#-(`aq3T4g@w3N%z
z{IP2t3YaK3rLa$dJ#z`fxSu+gah4<?TR4X=_<Z^%0o|ybJD?Rk@1SB~>+`YbF7Zo`
zV#k~JO&{Rj?6s2rLbS|UANh}~;?Ey;2f^9Uomq<2`JXS!JM=HApVY{x57oAGS8u+t
z?5m|ig2Tq6fcQtkh&VmRB(JcUeRAz&-n#&hV<I65eWsz=r}04_Q9=jFYxvpj#NaXY
zA_w4{LJ${S0=vrGAKC=#-?jYW>;Lm%|JzR@iJ`@~{EP{E<lnyC27!_L4+#1tY0Cf%
z%noD?@ob!&)yM@M!J=~c)q&|wFa`HGYvsZHs4?rz0%(T*Ae7`6P2+QNl-%m@aUb_u
z_&vQE#xwtq()T~<u)iD3{*B4_YgGGxo6?7{6hr^U>eIp=a?ASu2a~nXZPPAO_tkR*
zI5q2)00QqAYUJsrpm^a%JYoC)^1T@>V2htoiQN3xWCk&Mgz9b+TD5eYm)E+B2tBn<
z%0#?;yT#u905BTGnckP@Ac4V!J<s5Z1o-ekd3L??=}+$79X*4$H1?CNj{!<k1jX|W
z0AO3!3;Mn>a2kN&42i0tx$VH{QlMR(!3JWn7ls)57lNO$R-itJ(=;r6SdnJk4S?Ou
zX%IR$17BK`ksQBS8L<a(w)bEblngF5kFMES2s@j=T0I79Jp%Szk6eAMCrJ1UVKys^
zI8RX}-Hd0g-umkLaMcc)XA7$%N!Fjs3XFcR1t5EZc<lis$6=z_yBa765IvnPXzdAS
z&-~(z{M&Z?`JYmDaBwR<I%EH*F&Kk)A*&a~C}KubZIW6@q}g_oc#**G0C0v}qe_^+
z*$VufAo$Ts>INWqompDh?AAKGq_e#*lUtL-MYYb?-9VTW5ahswrkNo4j1=vNCJ{&=
zY|Uu}oo^wN+75i^NeCo~Z3opanMHHyvjBl!=kms<JVVjrd^ZB?%Xg4Y@3^qtv&tMB
zgz%x9AjFH71g{eak>6uNP&xKHU|dv)hz9a}SrxyIwnr>8m2ax&2)sHr8Q`frD#lIL
zx){fQvl$Fx2<B?9`3bxR3mcY?^Ie${MsLrYf*SK249Sa^!omC<%jFJ)hUSaM_Q1VB
z|73wSl9ZTA%*ae0T>BVxx&=;cKMAT(0?`$!wFE%-2a=9F9~JoTtWt!=YmD$4LQf|D
zdbsE(RY0JaJ19mVd3ZwMK%VoWx5RfQPqUH>-)tQ~FntLm$j?Bdc$Py3Fsn#-D%$4!
zmrvg%ZUjC2<E$32@nJ#CkdT%U9_)>}KuVT{O7NGOqn+p%vVgc^GU<NQE(3t1j1Jy}
z1qeS(D*&m!0aauP@!g<Eg#sUhX?>ynX-80o6$^+w9AT*MZm*ZJVTolwgLB~5Cz?Vm
zj{_t%K7@UpR4C|ay&8+z<Mm%I;{?^-O)D*!kq~I}!d9>}vB#RqyE4*N0%Yr3-|+T9
zx*GA9g>k~aOa|lECye~bU0q;mLQr$9h$G_Tso7Rg?P-V^pE){9#ref#s3m~VgYF|v
zp6pdk;A^-vBb&p1OX!rsN8BiA!Z69d1X$?c0|Xk&lWkO>oM4m0{~K4ai{09wbG(@(
z*64ADT5pjSY06PDU!?|U_R*XRzHDE!vh(;DPs*LHCxumb>kXr}^X_Wz+#>Kno2$GJ
znt|r?G3cXe$BfSM#gA;l5Hto!q3V=jJGfp3?tCe9;2uMb2yOck^nsc8UF3T*;ziZ6
zxhVLsY&;8VHTi{WW0vy{%y`gtrveE5UZZ@i0W!7&9r6HW^tgeSfTV$rf6Hg+-&9p4
zcUKB2H{Jj&kB}osIK#~c2O04Lx@8CgcOFyMjDw1xuFe}7Ma@3SQOG$54oO+LA>f9m
z9<`#|L%PWFyTuquf^e#zG4&t*)j5#jqc&ax=n54P;dgXK=(9#Np|U4M(pA*nj$wRF
zdvNz7a8Xu9d@r(vIDa5=Q^fV586JU&Ewsj^bYQ=K{Ny3&6$JcSAAR6h?L%mOrbV~I
zr?M5YR5-?$%O1<o@8Jtc?mbt}v0ZgM@<cd$e$EL(CT;-fDH<RaO^}U*ACAPgKg;Tk
z;hd2kdFViz7sgoA%BmcHK?OFV<f{Z1V2QPW#_22)qI`|pk&)c5*dQ>@_=qj=7u#N3
zdj5N{O_Ma}K<K=j;(u1mmmeQ6cwgN)0Po&%{`4zdPnq3O%cFPZZREVrbY)ewl(vWf
zeImdf8(AYB!}_^s^zD?hdG{npS&q5w93jh()&>$BX{pZylQeKzb2vRsum-Zu+{jdM
zA%LO*2G5D|X&=B~EvfsItxeQ7gV|~s>d#H1zINP7l1}7vdI*y<UEWoF9nmuz9H|ib
zDuT0Cis3GHsZC_v(26<G>u&VutwXis1p|XV${KG##s-oM!CEg=SjfnrV(-9}6~ZkZ
z+9$L+)nY>q$M9g1UB)-WJ5%Ok#@?#ZOEf}Z9y?e)(vTg}h;58*0qW2m@DBIxaGP^*
zSVZp?l3E1pZ_2bY-ti`sB?t^RA^}0qhbM3W4vZn`AqZlU7_fUV8v8O|mF}lG;9!ER
zE2{UcyDNhH)X=NcpLwjn&MzLsC&}|-$nE|td(Yue`&Q%7Prwz|Jvo##dalLu(b||U
zv1k68QETN?furBU&8zX4dISDlPT#5eSeRO+?Cpt<#NAAK4E7QqavH4PtoB&6)s}e;
z4rh`2lkVX1c5huV%ZqEm^e7;weLv9U<^qlz+aLb6gSWg?M|Gv>t<#S;>)+o{jF}@|
zjvtfHj|=DEB|)EEq6##~xZ|G|5B+XWgiGzg0o{^})_C%#FhNQY<;5_U8bDtd_VUPy
z^E|WaK)Gwz6C{=!fwTzWKu(6!FrA54zOe{Za}x|h31!e_?+3w9Po>Hyti~pYNGB3i
zHX>Yh*Eb^@@dBJkgj98)qg85_5lJfmBrySu!%|lP%Pt-z>>uRA0mM6F*kjvzd-|cs
zm<M9mxAwi?4)l%KQ)*#<=O60U`Dk&vMGoJHP-%6aRAf!Ofhpn-ngc^}eyX!Ir`tfK
zXld=EMk+S?fSlUz1f)A26@IS8XR`?Nzh>ygIFn&~(#!ev2D{+Gt&Vex)%2~5IOzlW
zE1QsWL(5*WsN?jL=1Fzw0$6e65E#?Ffx88tWjYLOG4v%d!12V#3zlT&&D+`BY)>PO
z9V@B5eur^<1_ssPWZY#8XK+5)qqxG9Cc=qOdU3sxJCrz7D@Ch7nM^CP3XR^@E<GV*
zqliuhn(tSbr6?92rdGQo>f?OO4rpNe`Tz}>zD9}<D#=>^X#}peN4yMK?zBV;5_n(*
z;qu~YqQLVx6F3wuxFw)-3EZ$QMsD){v?<J`yv4s32;wqBY8d51_g2tm6{=Na4G<ws
z&YUrxD@EN!SR5~Mr$!#w67mWg(3>VUE~7lQWi#8tO!DG9xT}{<TEm1m)jY9z$k5vQ
zOSQxf3?iR8w5G&T+%X|Kaw5|H3Mt>&l?q@rGg?W`=)_sEEH#QApX4EUzezf`@V2kW
zLW64a;R%7};cYm$Fi^nU7<LAed?NhWXHKl~lAa#Z6q0K1o*nd9@pcSj!;^7duRWHd
zq53i5F1^!D(mef5iI#i6>GGIS{?O*B-{R%^xwIn>c$eExA!a<vblx*XR=ui8fOqCl
z&my?=xBgdvG%hfn!siU+e|{zT4tN~-{}CB*lYyK%4q$CF3|I0`FF@rHZ;GW2uM;d|
zvW9z5WnEsM9|8cBp<xz4K;4%$L|tIYPHJ!+I$}YtAwZ2y@e7dh3dZ8M#^l{JJpgzz
zYYJz?S10yUK=qB(^7~oXIFNWBmzb%kU?5P-|4KA^LQYXS`zq=i!i`+(IDsiJcDF(_
z(!`1*q*VB_#H0*fpMe%xXeU3xc+C}|IL;k7=XUrdIUXmN65kxhUv)4&c$(gA<pVV$
zIoa%ew`Vp^;1dcF*sTU(Cbz4^F<a&^`{BF!fsTq)>inbxj_C_?gVyscc}kvo>ws+8
z)NxXKs?+DN_kEl0AvDswSWNW+=`(UAu|H60HCrd^PWH<%)?fm?8N2S`;h$R($l)_1
zQ!F^lkrob&Au548AdqV~j!#Z}Jzu+KY#3Nrw|gWgV>pc7Jn%YUMV+(HXf)()b9@>9
zwmmpg+L+13VVEw4^7JMlR7Vv~g5EmRyvak^<dpVm;#m;_J=8bCgPg_2ZnC%i-Vj;`
zRGI|hEsFKH=DeeIMCEMdUX32uVVit<{Oim+?uMPfT6~8_4+P<hzeH!ELtLFdRTG>i
ziP!+L;9jy*YH503PqmU9z{0wm96}ry?a6#ZcWyBzAHe%@_C<n8&9zpo?-PA7JdW4F
z1lpTeJlX3e&&7}J)4k2)8;dP4r=ZeJ30PO(5ju$H%8{dZkz_+!!Dc9UN$k_lL-Ow$
zu3NDre!si`{`$GpFu{tc&|iIrl)&i}e6lO@Ry^t^o?9J*8cLU(OQj^Q4cgC0v}rK<
z|3UTJ!n?5gI=D~_{mt3Z`W$*n^(QCN@S+4gV$tVuc(rTnTEk7Pjfu>K+P@JE=h+Jv
zHN&UB1@w_xR4GVuy0SYhbX_(Wa4%+_{*wp!bOnsNucbcR^6sky^{F_T#sz$8qh9#b
z6iwA~cbVPHI;VCHE$x3Mo7&dWuyp{3|L*l{h(tYBlg2AaFIw4^{B`$08WA@;>T>1O
zyFk&r9p_O4`McVz=!XP%&0atH^oY43D)^ijTI*T&mE5<^>=i4TVo|}3uQgMj1#HG&
zQERz742q!mE1jVhxJj7uKEEuNEBo9H*N}anQG>%C4*$6r7fr4h-2^>am3lK#b=y|~
zTiFM=qm<<vsMi5OO9!19^W%UP$(fxhkznDx8jCB2jxOxC>~#CV;ie=eW15$1*w85P
zP%RjFe9{;#3sF5)fH7({GB7=IRa$)}N%U+t(iVNPwQV$MqtKCf<ezWTjxe__G$20D
zG+ITbk>GE$mX;b-V4h{v8Jp$0`uYpyB6sGP;T8F<B)t`tRqs8cnl)SW_ZMS!iFD3i
zA8YRcOZ_p2#p(4`1&(3355H$lR9*6fPSh>S?KRCsclvKysN`nf-xl*~x&%gZfAa!7
zL0Ls)R-84K*)4p9t4#|<IAIw~wINO02lfni!AGf_UU4XlqwXq)c9PzMS0VvHM;UL3
z0FEtIJpUDFiJV$_qPF<&pv!i#v}l*t5z4CPrI@QFGWT?X16bVQ#pIDwB3&56Q3J#M
zr?K)w_}XPOp=!O=EO%rR)$YE-^2L^Py-XuMX`s)b?6si4b+0_5h^bB#(#n4*GtQTh
zPM*}N@|zkjVhXZ#Xyt8;<>?M$x2f9zjwOA=D>BQrPB|)R-t<b07z^W0&{Lr0Rc}?!
zW;ocipRuI_LE($&?9M{c5GAFNO0^bTlHSa6*QzNyHLl#qV8*;j@u+jz?X(XRtP8aR
z-|zaiP&6WS5kbhFP=ajE3+-JohuwLYU9&-o$&t|+de`s9HMbocp9VX!ri&m5f(Z9K
zwsmuKMva(k{k8*o<4nPtue4cC9I8h;maO2~Tf8x|ohlhF$LHcc<spuO4xbXKljlsT
zyzsQ&R;08?b9T6HJZ=CJKh(D$06{2a%XUpIs($#3*!)i)w2<05E2Wk1Cn&>UjK6V!
zq+k%}5uG`%6B?mF8FV$cVpuk@e2HD<s0^oDQ$xTQk2ml&io;q^?S6!;3rO!+d*?-c
z^;JyH#-`-(+WgQUr^+bPg`77xwynS>uJk=i{4?DGMUE;*JSAa(1edywB%LeSN~Avh
zQOZ-MD&nkE#-VDd3+iS^$D6N%gCkYa!$0<91@(E=4>JrvowT}4bqS$Pwpji_o!p2|
zSQPuq1~B|jXR<W?vU~XN-bb+Btb4)TWo>YXs%ikk`2oUyC1+hiiZ?*K_iB-xBdc01
zYdH;TWgdP7-2Y{;W%q%da{Z{4B&}w{QxZI00*`8ph|r{QBh*BFlN_99JwQ-r;fYw!
z_GYLvQMLPvpA4-PySRBvUyINoZr+&U;nGxTI2oRO<ju#_3ADe1w^G=_dPio(^EKaJ
zU``GL0Yba$2<yaYD?9`I^s(L$*vNNKyu<mFTx#E>%FfnS*LjG;7gLw^%GjvAlijDb
zB3qP^&d$M<b7#gwhg^I`O+7W4&aR1Z?>0vQ01$eA@c*pTa6PNvJ;3Tt&fg!GC&2NH
z;$jAe$R=<p<s@lC^fJhXxm&Ezp3PwX#65%4(e#VcISJmzx73Z(Qm*s+Lt>Q5Aqsj<
zFuy;n@}#b8>89IHfJ095trEcByQO?djTQ<=$4PfU7S6<euV68=Cx*ZP0x=lapG1NP
zZ@^7n9L&49W<)NS>hB0@er31J)aY_3k`(E5{=-yGv*6EHCyxU+i1+MHzdA}(uuWgS
z@r6b$wZ!{ha{Flq*~u?g(!8#EHG?zMbmL|xR;@Jr^Ch03dxPX#5R>&jLPR2nXZXPI
zI}oQ=OylU%V4oLb9_EA!P+7gLu!{k8L-5|M)2vFx9<1y=F(Bd%B(>vs<_R6qR}d#f
zK|OyR(XWkq@ZkTju=Lg*D6!_S#1n85U9>{aya(1O7kg(&<T{i$EHgesNz&UZ0sN1X
z8Y`bv7{nk@O{cxG^6!Q}>Y!nyiK6}Vi=-pKs&?BOvoLB}QgILMaFYYH(EK=PPLCF&
znEoLA>C&3BQmb1zvm83CG(H9;?PaP+cTmV&1q=I-VGuDXhxR~E7sEMa3$}#_sQ*=2
zhnxflC!lKtibc~AR`xh4<a^x%jsTrHSyX3%se*d%Yy8wKJ)vU`?>T5o0V9b68)<Gk
z=;$lBn_QG-HmC%`m>cBs)z$wGdv6_8W!v?ODx!cWsbGMVgs=dmMM?yb5(`*}bW2Hh
zHwcK}LRtw$T0pu%T9B6R?rzxge&YSk9`F0O_x{cw=Z~|;!+3@+8L-xUU-xy*`Kvik
z<QUuILYge)22JrG<FVGVz{ny~qmY0ljR+IgNo*{)pLndavi={4sGlU_-+d%So%;Y^
zJ=yK&DJEHZqR0<;9|Jrei85M#!TK2P`q6KUacNt~!(u76)S*St!Me3{dT2dvczGyC
z$J%--q8I)<F4H0_DY5Osgqa42y<@HE$E_I8P>KuANZ}ke`eouhop6nehH_ATL{V&z
z(Bpn*1n%mNzPt4)GgXNejLSBa8Y-um!{k$huGD7Gjwxs>%*N}2Z8%#6EK=fL-3-GK
ztMv7_y4akOo@v+|ulh+j9iU!<yMiW==4Z0->#OsfCgB`~nf^&sPkkn?uj@a4X*2x1
ztcztuNKAC{k1w||ysR{|<8JTsW%bYu=U%Ry5yg3WdKYolB*@4N-qW#%Skfzp)56EA
z(x;cRstMX)iUcK8`FctF^t@FhrMj&29EO|?9Z60(BIFSdu5--uCutdF87)ILe~j<C
zIDdOZyUE~SS56IW({g(0R-b&RP`^=*RRF#HysX6I$wm~OxULFEq>z)nHW`X8r)0Wm
zjy3>A0#AYoanoAl3N1&Qn*7jes5ojcg8MaXqnN;XBfY?TYxaxz#Pvm<R_OR1F@9sW
z>v=F{v9pujUWs7PQzW9%h<2vkgj0x0MP3i{;lMQEa%`|Y+SAgqhu%IQW&i49Jt1(`
zm1Xv1s+GGIj%6zO1vbwaiz%-|;VOnokJ*wKZbDtzpUh)9vsF$C;S}rJ!R)t&+qmso
zJJ=IVqb7rolm~PQ$;u)6TIayrIN+ZzxvxkyZh^N~nydJq!zD$quZX%f)PuS2Hb;PJ
zh|&u`qszTA+YNLpS{x2u7ZvI)%cWu?egf`PNQX_GH=(cHRB#`35+?M^N48K9IA`m0
zK4DaVW_5iA9LxbzFtug|vNG85TTzPQzCW6uOZ$vHQ06FLpc#Rhh(Zu%goG$H`5EmF
z6I8PXJPKXxr2EMv`u>r2A(nx{Nz$GPNw^Pl5Vz<{hE{i*fk$dRJ{fx1m}K$18ci<X
z5X5!q>+uhW>bvLsiH@|-(-AgF%{E@~ks6(`-I}Hgu2`PWEDvg))7F;W=<ZZ_>|!As
zX;?d*kUH0VJx<qFXikH;y-nq`vv}8GU^3_KjH~lSr`RAdeNNrx7222Bj$H$Q?2T2!
zG)2x-yY*?J)c&5&ZIHa$_fgI6uXXz8YW=tW&%AfVcBbTf`}ca*XU0OJc{;&7ENsRP
zbEm~1t@oAZ=DU{!mDm{Bo$5XE&<vRwZ>`b%gjWC_X6XZ3;|HNBrPiW6sIz@*#;T4x
z+vUfl&n*uM3oe$cSohLDhvB=l%Fom1tnQBJN!N=5Y*IqLreW;8uW395LbhNw*s}Fu
z%6VYc)&?LR7%!i1@8I=e<u7-`!mWkTuLgkypjAHeZDB7rhV&s{Wx!YUqF<T;;z(d)
zWd5f+_f>ry3CqDnFUfk<07F~7iB^4~70B&~ZdfBdpuWim<#Oq_S}r1S3`<oj?FznG
zb7)X}V;3MR0xrs1ZQ-Np7fEihwC(}0I4Ayf)b=g??u9ju1KI;O(_e(;4JBtaCh-z*
zF*Mg2hINT<(vNqlZ0872h4}!C+ZZmbjLA_cB|LQyZP_k4<W4ooGMfSORn$xjQnN(;
zZ26b;<B|i2Gx|x^E~a}}ZLSkH+Sq#UET-^twgfM>I18<1Q@5--<8{**+@k`ZHgcZ1
zs(ya$6P@WG)l%H&!!Nb!NJxfR`b6-w8Ia7X0uRd>cOAXwZChri+X&Lj_U4XG`ATWN
z=wsTmbktvkQ{J0;g9VG0`k%kvcvQ%|F!Y!;|JOGSdR71?HVAu$vVXeLmN|@p4&glc
zXclc=jk<Z8RAHwYG7)k(S)5x?C-{|LrZ=auPEIxVJ?Z_rL(qc+L&3?IXOR~jl{0B_
zXb`nym07apPuNlOU<oi7Ybc={4%#~xVHQb4gno709Y~uPV0YPVdgd}pZAK+vn>Bt7
zFt8%dQNOD<)m=F&6X*zdLKeoq4wyF;Lf%_96=9F}@Rz5b`AIMEN7Y#Crq}x5;3)5P
zMyOZAj%c_9OuP;JIJIa$+!&=ayzoy)kp(~M=%NJxA02?a+iH`Jqe$PjUJ_@=g~$F{
zN!c2q8%qwu`>INIw{9ir8B1Ff8FYNQozp(vT$GR_dHBSZ;l34vttHZJ=1j&2nkw`3
z{s8Mfy|B>*m5k<;x9o8YTvotJy7ZR5y(6BK;A_4Kebf$zp#r}k3+bKRnQXZT?;N~m
zgAnKTw&4}S1v7@{R(S`Oy|-h(%JTHq_>c>0&?-^s9-xPA`fFYee)>yQ|D#tltI|RX
z-}viS56Z1bk~eohJWP^_SZV3yh%y1b$OJ}Qv|zHWJMs8PcXrnJ06Np7rLh4ZWS74e
zvxoW~#P2<CP<)1fVunEY;8p&_TWNmMqdoinw5{E}Zcm>UjJ^$ns14WzEoS$!+PYdL
zIJ+X;BzG1Urs>%6$Berj1_4Q)3l%v;km9|P<>7Ld2bkl<WVI(#x5pn6&(=TP=FM)w
z{yx)|FbICsahVtM&7#@M(>27a3xY5ts7%-js)b#>uxWFc#BB+%qIv^M>rE>+yelzE
zZ>>O+R$4sf&l2r-2w#0((#H*`okCFG{v3q;W3eS1tK0h@&t2q57WsjFeI@nsI|2Vc
z4+{|sA`vc~?%8N10dqL%b2#x?gZz455`4uD&7mwr(p`JkK3puNnc@S5?O=NT&>W*Z
z_)84f+mXN`^12^@gP8zyZQ)ynL2LT?k9v|43U0jF4t}%(*!JwB*B7y!eQU{7Z8h{T
z9|uC^gl03vvBmiyQSl2ojd|Hy1j|9@drun~s4ss<9qr=j=g7>QW=T9nbJ6{jr+gQM
zW*N}H?GZXxc|mqv2wM3>8CRvnz6=4M`p|~sf{M&1j!b%ap7F>GWNB5R^A*BNdBe^Q
zdSvNRK19Zm;@{f^!Cto5l)`#Xd9C_>+hhuyH`{C+l;UA5*pGAr+rw{o+EqR#yP=1u
z>o^U4M_<kMVAL)1&ZT6MOW~v%837Zj8g_a6E$?eurP}fD)_VBw-)=8U?tc@a>h$UK
zz&&)&p%->iA(|5x%LOVcBVhsC=Y2h<<9zW^_|4?2Uu^V?i9UsX&Z{Y;JBRJjSq)Tg
zIQg^!Nt(^%dYg=u?&TN#vwyB1xywjA<y%REd_*kiu&Q57dYXcXmQpNXsHi-|ieWY0
zRF_P{>8f<OpE-#QV#~h@WHGsKU2rhP?RH_G(MyL{#f}6r&GG@cE{2MHc>7ef*`B6u
z$j=^9gdZP-p;Dv|XBDx3H+h;G=Ojp?Dms$>+S-1~W5c^*(`)J~0<Vl6%pRCZxt%tl
z9+5%kGZ|fQI*;oZx2ixmv<Ur-mcZ$U`!!8rRN;y;w9s|jP>7oY(`D%s{{6HwIsl)*
z?NY{Drt3Qa$L2pExc;z40CSu*;-i214BTaN9|}=4mGjO=@*?6hSS;9R&#nameD4Vw
zg?@f>UwfWD{R9rv3Bh>pf&>c92o^r&PpyKmEa~0iSeF;Xu|^#a^6SmzQ(gV`$zy^_
zCeOBgVU1fP`lc><gRe#b{UUKdk)&1A;0Gp+ApF^(c-m#ut6IS+yaavBs;yO?>ka4S
z*gTc_b>cMZ@82E5B<I|1DSf=qsE?<8<h!J>4AUObF5EZTfH;m_j!SbYX&=utxzeOf
zm?{cU{ZP=W?X}!m4+%eQM%X<*Hv6FelM69+lEL&k1q&_Z=+S29VPG2Py^<MankixM
zq(`&%Tq6i~Hb@w6<zvihkH~cnc&8gXT#zu(<`ZzI??$XnRYji>##u>#;<sg;D{dGv
zB|O?^>t8pN-|vKiDp9p;Bv*s-NB}#I;*GQKu$|u>%-b50;9@agC!){Zd+kZ|k#6;o
zVf)LLXIiW0rM)X@XrYI5$`vy`@pfFXRf51)*8EFsFhhf{3DJ4XN!^A)Nbcdokk+=p
zN}z~WPZP))*n*yIo^_XdPW9|>m(%k=D38)H<5GY31YCG0(}Xley^<u9*z(C3-PBma
zCiLJ`1~dHjc9XD{6H%%bV~pinwU+0WMK~uhw0d89vII~HpNN984)R|JVc83<tr~i#
zZaQ+^W3!CogN&vDsF8G>dj|mE%+w(MTvpphtDviNq}aWG#ryfm-_*sq)Uec_-|Tp&
z{Z}(GV}yma+U*}cn#$@U9{>(!l(l5QZaT)~>vu8>R5dRghTa`I<tQb8Zgxzk_=E+t
z^#wW@rP}>UW?Azah1mh$I?S=rwd)YY$}YVP;>-cy(wl+|KHK4CP0RxZJqC$L0Z*<2
znKA)yI=sD%XO8Z}bb^q{56gG4!McrsQ^8G0s&$a?ku8{;AJwfb1e$NG?x<^lBWg|I
z)Kx1guius?=?_G($`zWel9c{>%K0NI6xEz2kpu=GCYl23t$UsM5CL5T<@F$Y!j+af
zm(0ScD>e`-fJw8{xah5;xO)|p2>lkVg5yKqSpT^O`nU2(&>ilzAnnz6xPM(veP=B6
zumdQHp3?P!*fJkPx`l8q+}=EZ)iz~PFzU^L?^-zymM7KhDe|EI#m4k^@ABtbx5t8e
zXD`aeeEwf0*#FTV1cQ(_%O?Gf`tL^c-%9)675<Rg3fDRIO@NfgZ%^ERYRW`{>8Rb*
zf}5VuKltr#{6Bx_e?8`69~g<?$!DC1Z~wR3@V|8#evacl80i_$euhW#fAqQk{^l7x
zs3hj`3X{2h)9U}`pEmfgNOm?0Q1b47ex~PM6)s2XRJ_nI6-4MpevyAWkkGLUIXeYR
z{sYjKbtH6rsrh#8WFxdSE{nP`R^8pwj-;Q^RzG+n)_^|kM+Od<1l^u`1&JUIq*Kqf
z{olr}FTGRqPyg-x5!t4N`>@@}{$1g(`!E11r9otULQf{h3#TkXxU{n7{uIps0bR_)
zX-L$u^1bdJWzlkb5Wxwo9Tt|$k2cx2feo|0t11v8&HC-xpnDSnf*wLAX9aTFK%Cwc
z<Hio*qt)Q-edu<TMEVR2uzWw;K1|L;s7mC-b~wpmC8nTZp4!cQ;R7f@haXSZs4Nea
z4#H+aCH!nXZ!XgGk!%cq7g8yjz!flq!ruDF!Pqt8Z%$B15wuMl9}XYyng_02yTuY;
zX}`|o9NhAY;Yc8%a@BDy+Ie1i_^c1}(Ue`wtnklpH?3E8jn=sL%KsPZ<!1-7Qaa3^
z%@Y4{^Jm(yk?{+>_8a<$!;O6-3e*NFFl}oP;<o3lA;9Hf{!t!yxr#vtlX@Ks4R;3N
zOFI(QU26oP!8QTe+z|8SecU@lw+BE=Lr+H0I+jK2^^Oz8Y5nQW?nJ1Q-B_%6Vgl#5
z>P9J10ff@P1W|v2qRFa#K!IkPx{!@o6*8LC@}^>JRM`{xq3(PEQPvcjis>2h^H+8d
zj%Ep5Wb7YB-xP!KsTh)j3M^4g-f#)^p@%X+x`816m4fl0CikEF;$e`X!^I0j%WdEk
zDu(2eWrC19E>M1dP&RL%M@sdlwoX+}Z{6i+Yt}nI%f&U%j=1}gG8Ta;smq1QwSdr#
z*0CGLC8vx|I<<VcO29XALQVKpLWxfCOKpc8$wutmHWjg?AF-5kgj@~;H=O_Rn*X$i
z=gP2DK+*qCtfuIO+htDOg1Ey4J=F(U2sn$B`v{J?1UQi;nif%w<sV_hxkAamlSrdR
zPoG4pYeIh~?DK+y8fG&fZ&?eSP27Y7-s<Tmnbk2*!0l2qCUUsgv5L7y@2i~bwrzu-
z`G{oaVo1;x(|*e#b(b1qi5<OUm*XQ@^717z8ZkNH<K5QM6L!cM)V+_Gc5Fb!Wk1>h
zPvdgd1*KHOuA(zSaLjx|SZk?dHkfh<qLiCHw%OM{7foG#2~Oir<U2873pwItkxqO1
zBoAR=^DD6MZQJJZ%MXp&H^wP6Q(@*_p%=L-pW#pWLsHY&q>(IlA0|RW2gK8!`uY6c
z;Ii{{To27X@34$vkckeso;+1${KRElXd0;-=>#kNjfS#gt~8L{uJ@sWPUMk_PN6ta
zSG3F~&ddAs!T)&Gsj*^^WAmYBy0YJ2ozXQc2;Xu8e5QH;@^MSQeaxz%SGe!I78+-Z
zF9-8tqscob3gs|{#DhjKdd&nvK9tzz&76^$c*Wp4VhXPyY-5lXq{W3GOfmi<<|Vu3
zeMqV;MyiFwSvi+A9ez{WbVotyHO(KS(q;%apbQm(G#}^;r&)uJ#>mT8Ht#3bNnK+$
zO3pAiRsD9sw{#w6U}fxj(<{uk2yVZ~2eV`Wj1){b|0+-!eRK~U{L2CCYMZp>J4tA*
z?<A^iL5_9LA+?&%_)KdZocD{YtHzM=8)14G1SL3WU05&ZV%JM_?pRbg{5<W;8Rxhr
zX{u3xL!T=BHvUK?T)jv3Ux49%ojiY^Nc_~WapJgIKL2aqjC4EERGtAQegbJh-<_;Z
zQX51k*RL7Cdhi69y(x9@>|Q>7(e4>jIpW&QFw4%TGit!UvWtaQWL~qM>_H|%nScS)
zoLtTEEBX4nIIATb;T<>f_!7vPBDgEZ`p`=DYEmq<`y8It56E?&CHf`2^-Ie*qN71M
z_Yhpuh1F|=An9D<fw`JKR?TS`$9du%?y=j;csj+Q0V1>3zRwV`2i}2{*xc|Vmk-X?
zMNj_u?EPSdJcty^UZ}q+ATQzy=B7x)j9H>T<baD+h-zn^RhQ-jjI<!{kCD4pA>6b-
z<^r3e6&wwE?b5RQN@BbN!SwiMC_UdvwzShuPzniMgF6tVwiHfJpckQoN!eyRda)1Z
z+>VZM0+acueH!7{7)$ikG1W>971WO>#L_v$JQ`!Z)9(0e=q{3{f|~CpoT^IEEnn$U
zb;Ywc{0@f8?c0gV%VLew$h;>n7dnHGcgw{nR1GB?Sn-bIuJdEnju(PD?yu?%V?6+(
zDTE;59vVv|O}be5(I+Mx$+~Yfx5CI9w;)SPb!`itXb0j)d~!o<+4ZzZ>lz+AbYx@X
zKIM<d_-X~!zw@-ROpmonCNzrqs$qtyj(6&uC6&ZTN*%UMY-ZuF68S*f?oHZ5RztY;
zbO!+}-GsOD*hxj70#%ab`hvoGZm*6T^nWZH8Gu%5Q9rXbh_b(h#p<H53>~G_S<&aH
zox?>_mvsegC9J4LadH)|j`TloW9#!gPWo&d`=ESbG%=#Kp^R_kQk0HJ%=i%VaWQG|
z`&oL;wx4R6cFR(1rdP?Tl3sMgIzdc)neo&S>dqk4pTn^~)u@B`P}=QmpEo1M4sqJW
z9VGdo`V>;UK{>jckuOZt{^lTQSvq=F;-`;a{-9)eANk(}&wnTe-J}pSMeKgvUr)&g
zKit<;!b6L|Q7nPs#R}*@nZV244q5NNU)Pif^+6CltuM`>!m5J<O*e@C>OAqT4)Nf@
zR(|<rJQcPaAjMhaRy40gu#rB~PFfX-!n^}LZt1rwDe0@(*J%{aA;o$*Y8Ae9gz+P>
zKsxwVOn_zG?qT#2M`nRXLC!b%1LCc-ENLS!NY~ezv>NrJHakED7w1-&FEQP~6KU3W
zjWHgZxZ^|`AXMwCJXKfvzP=2wt|zDchdPdrE`H~@#?Zf+NO$>u_aPt}ErZYj9dXoP
z``$-y?LwUC_O2H<GbUEsJx=cgi@5L1SjG^?-AhPf!?>Nu3IENqz#mxM2uVXYWa=_~
zXLwawtM0o)Gz5y155O_BxE@j>0MkTXfFw}5M=nlikP3g&kJP}O+Ec>UMiO7=WMvgV
z?{iI~fO9V(NbvCioe8xgW**M`Z_=2eEAk9dcnw)!`(CE*(%*xqyY9}^2P{=g*iCg5
z7BVSu6L-6~jm5J4KRoNn{i-E7gxIDmUHN^sg84s_*}&-HJ-ut?`2*$wE($n=*9MlJ
zU_RWxIq;y)qBB!Xn0@UIbm#}!rRX~Ch7p*%dHsJ{fE=R{MLdp{Z}{ZR!g=_I@<lw!
zYWgp-OGSTRWsy2S6j@#DoU&1s**_2o9yC0QBsK@S`3W47SFPaGC;*hHNBLx;+WpH3
zKjtC&Zofg!dCaT#T}^^r|4mIi={-+e`Q?Iff?+~ea;G~Or_YBjknl)xcxKNm7m2Ui
z7%R7p2aHgK-7@i5M3|fncm4s6G};sp?lQUX1(xYW!kTfSJYp#rxE`qL+jI!4O)nqh
zd_Em}7k{^wBk>-LC~`8}UHJ2x+%o7~nKYnMBe->ybi`l@wx5-%LQo;M%4vSu#j7Kd
zqPcQz34C8`O}gHP&s}l=D!32v_2->$-(!eG7!cEjb9b=}?4eE=01MO7>4&&w=(l{c
zSqc__x8BaEi^SKg1Fy-D+=lt#Jwm|=BtL;Ytk326uqmxH@G@s@3>dDCgoclgwgu_L
zo=O_J$h?OUE+MKlZoiOwzD|2H<IBVxaV-tUsF3fuS2Ch>T=y!zds=fVVmIZk%u7l<
z=Ti!bfe>b);le2x8@GbhiikjCfMihJ@gcGR^@r*!k)dc%sSZ7nFOZ0ZMGQVYGkEi#
z$n6Cnqh56&oMEbFj8OWC<{{%!R)*h$^EAUS;bT;s3}QVs8t}=uiTjD<ma3^l{S(J`
zK+81UbdL!i7e$nG4N|66S0t3gIxB899wUZuk|?eend{@dZ9-V*QR0UipM|Z+0T>lu
zMypi$Q^ij_kDiO<kx~+Wlw#QTizXfkv699~wm>ur5`xbH5?TnP*VantYGe&lnT}y-
zy&UV^?nRuBH?8OmKCp_Zzn%#y(g1ay4o2)wk)(o~T0-o4noe<orTm@XH|OsvlhyeU
zeKd7@72qtc5HbRl$g*%+z4@?|^K5a9^WLn`GI10gU)pulVzNwo`3t=mE61z%uKIom
zvX6AeTO7D*S~7~p>><3=es_LZ7n>9;S1hpspA$iO!PaqW&7Ak|G_L2tyv_pp@E7d?
zXx=`3Ix0fU=b!+BT|N*g5v&;B!a~Oe!bvx<eO?kry1FBl`U7L2*}RM0{;9<ezfO*w
zSFEYw+rtSf-Elg^J0$(xVhzImHBu7&cVj0KN_rFDJVX|y7&r-x6#6L-$1MqYm*|5f
zoAUX?v0;YGbfmnNwz82pIWNf~xXW^%eyNlBFu80PS{=g5dZ|~;C_<QYM_gBjK0Ar9
zDaeO1>UD7Q6E$sxZdy`auj(YVDC{5^t_u-X(e*XrA$ZrMrhjGm8N{iQK6)1vwZg19
zjQt!4=Zff8MpY#XwtWQK?n!>@?GJk_QHGc7ls<_w8&b4a)mt>zP=+Y=_!&JBjyH<n
zj;Y7=uHTLHDy^@%-3Iaq73+W!9a?N%g%-zakr&HfF?D46Ngr<NihmiiLdHRBW0m?(
zSL2aB-^+qA08GaMMxe$H!6U7jt;bBfmJzs6d41{1V}`NOp6nm%H96|>p(#VMCBzZp
z$tlibjPgCXZ8uZpLzJkD@4PW%nYS47HGjuVa5)s6vn_xs-!5Gi*JVwga=19#RwEkX
zH%=EzmoubrwlA$}kH-WQrAb=fjNYGqbUe^UlJw0s_{wJVbVG0??X`8WeNC*jW)}vc
z&5!SKS>w-XpV!JCv<~<lUm#1mBA%L2^fKeb;<$f3w$ox&79p`?%)P;D?ZYik7eoY&
zFFB=yu|VLt(bw;^jdVH!OGXCU`cO9j4<5^&+Hgea45pp{eb_c%H#fq1z25(TRA<kb
zda=QuCF4>=@-63pcI%-dksH+gS;0zDX5Mj1u14W`1unk@O-1OjP#(05f8$y|C$O7E
z(b|~mlv)xqiclY9+6Ang5`f5PmA$z*QW-;(Dp*Jft!&Mj@0G?!Q60`mgeaBzd3EL$
z!r^#dW{r_Dv{3iPjqFLu1gNj*<-)fjesVXl5o_L3{*-PR9SDJ=LfQ0Nv<&_p9^tI`
zS0gTd&X!cHA=MqEw@uws_C(VgUb<`f6c}y4287Y>bchzvT8rh7v|Fmb_?V$eaP8p?
zFO0Y)8qpKpSK0v0k+;6Wp|My*vVC<}C@xd74-jW1VL8S=&GC=5o8t2*XVj&kR{l^z
zo1E5aN7pFd&6k^0_A6ZgRs3W`!pRH$3uS{j`O%mQKP$p>8~cs?NKf*#fShxfTe8_S
zP;$Z92Q<_Hclfjmua)aoazkQIzc}u}Sjb&<i9J!rT_X3Q!L*Fp(j*fcH#H285rDU4
zGK%<dgi1Rip0<g22Yhl=tr!=%Kt;YCAK`;a9}5^J>~&A493J%53MNin@tYhh?jcG9
zubby|4|4N<(_@Hz&;^*N>$TLize}unTy<H3CTlZk!G>Sz^q!KfP%xUYR`Dov@Mi1i
zhu-ZfQOlgs9jo!XxU3<a^Xd8R>OQKmOt7r+SD8QjCaj$OBd0=BZQzzhplb!@4jZL6
z>wwXhp|`Jy=EcyiUz2vEj#7*-&?q!qxlc6MS)Kjld3No`Cw!aNBu0Ezauw#Eiii0T
z^^}nA1rqH%et09`ThaY6zvs)82d77ojAlxLq@xUwBZshFjJ0Q5`~W*Tvz_=OrMUI`
zVX0|m`qEBcmF>{hNfpcB;MiCkh2-bO^k&UN)IqQL+R)i6n%U1)O0ml0<JJXs9~(aR
z5EkO7(A2<`CyEaP3ZPvS>!5KylQ;v{jY_8$GZtu6sAl?Q87Sv~B`_(@fY3G|>y3=s
zEC5E!H}~$2geRS^W=?1QqP+D{yf4spi5c&J0EQ<GD}EBp?3S1qfT3Izo1ka^rWTVg
z{N^<;<bTb6ngIcU3Hpd}`HRy(BcpDy66y?ZwNEAf7XdCSqYiG(C6IyG=*F~c&ULYH
zf8x2j?7SMFRsh*GKO!D5cgeNk{J7%B$?^i}m$^?HlWj!HrW8TU*>`aGD17O8(;Y%5
z!Jl*Gsr3-{YZC||XhFH$R<254cwz2ixQwt+p-nBQ!myaBG9Jn0YqldN^~P@KDV!Vx
z-`h-YA4$qmc=<sR{%Q$AlCo|qSUdFv$?78gczrKPX-gVolHKeIy8EUm#k_{n4GyL8
zM>nGFgZtN-OhCEclJ=zQg4kmn_b4O|zXfw_ktdeP-oY(SJ=K7DZH;hyIrR2}Ao#y@
zrz214i&6gWj_y2f`EDTRM|aL9;frzG6Wj~D@iK{npZjj*q)dz(N!#2}-n-Mql1YUy
zmpMs3l&$DL`dE6>D*l`WbcktG>qh$D4{74(>&I;ZD7&E_XEUw}J3AivYU5Ul<(3U*
zPm0Ij4fU(ofw8IxDoEa^DqF?Fhb@O7+p}R1u|R0_;+vs3e{Rd#6nFZ3Lhv&9*A%Q{
zw+`X#eYI8T#rZLoQ#z@|(_P1zzpq!_<9sx$cz;NK^E473s^>1JM`fyAhQXh~>BTv<
z1Tv1bMH)LJ1=<)V^w`2F-2Bcb(z{zfu$@gnTqwW%SnewLlT&Y#?%d9)QsiC$(;$D2
z=%R0&(mcz$VU^i7RR4ukM=L%$bsg3V_psu0Ne%dO&Z4L2O<g{qZZ7@?81@yp3tbfp
zZle32a3hi1eG>JmcMZx6CNN^5t5dm`qihL4Epy!)iG$iYbEi*JEXcG4vZl^kgHMWm
zWxVWL0d!ZNVnhflhat9W3XC61&9s49qPCz^Ye~DzpF=eBNhabCMIbwdt4ZDj6i+iN
z3gW)<+1_oW;=D>A6MBco!ME&LhV5(|j3?S~^3)kyx5xc$dFkxs2krRU$vF>&b|>;k
z<@2D!sBB<yf1>jATvG<gPJ}pjK2W!RfIL6AI%JIf{ET64F}RjWc5dEq_qSs)*1fo6
zrtt^tQXKZ-3XlGFT@a<RN2@uz?&V(rg<!uk9%dZ%a6_3O0suhNtT56q*P>wid5=s=
zwaT+hS9ac=97F(+z3t`Y=J@zyqA}Rt)(+|S-UpKGwCTQ2Qt%P^7CC;8Oh{jnbO-4W
zVQClN^*AGsAHP6xSPEb)tUWk5mDKzGy2$^r*m8@<-4ICdlnVK)(Ap5d(lutw_B&uI
zoG^NDL%J2A#X^+Ocs~3GlUFdwP2+PaJRNB*@+Ij?ZPF_7;q)Cc$&WhLUR^wuPpaG7
z%D%ic;z4*HCS^V2f=*2gE~a_S6A#bUc1yr$O|?Zm2*UmGw{K_`Hl%5I>Tu3$Tpo)#
z_;JN&V$*J0!-VXDA;G6)j>_}vYz>M{B&7^&vaJ0L*ik2%!XJ~WT$hs{H2vdDJj?jQ
z{)zV3n|e>jFG9(!3n~mn0~m-1{*qxp_s-^W$KQ|J-)EwxpK8u0vWTQo{Sa}oRp~F;
zGIo7&N)Z9g%n@pza>+E6!e{84?+Z-ZttDVA84e%S>F!E*8=xd(C+!c~w_kcT<%nha
zO*&D9mIW)(J`B8eC_cA{lHPbb^qf+9BT~G%zQ*$+`z{1l(dBGRk85C+TOT!#nTshu
zzHxq$E%_C-sS7nL8l%fI-)`{PQs$o#DG}_Kv?DLq=;i(eW2#7vK-yx^`<)?}n@P(M
z7TSWj9-tfNw51P-fLyH9Afo`bNgjnk&9}{m{MMglyc=HRT)F@|VF)XIcx1-T-5l*p
zNwO#UNjziUv4&rXNQ$gaySdwBWT3_FyVI5Fd=b$`g($g#wa=;;_p4O*P5by6hSZA8
zWRAO2Z;yS+q5msNu~V6yfCV_0dEBIS^mfk6T(4;<&-2eC4h4OFkTt5FQ(2XVBS11l
z>AJB^S^G)o&$ZW({`^99HVd|MUUc#@KJ`fI)HN8<Io#4~KHXLFuBD4^#jwUAHs2_m
zq|^PXZ-4XHE0r6<Fd@P5`19MFLH1E)nsld~ea{)=P0ak#3JzqNmfVIjSRBD^gHNqK
zm>-UyZ>-0V1~rRn`W!-Ixj%f-%M~d?3g~7g=-xt4bBkj>ihEOSyvI=ach?=E8nN&W
z{}#=JlJWs8T;f#-Wmk^OM4i|SM1@YA=FBTGlKsL5T!bT9<}F*^FU_HsDzfAZ0s!$)
z(>g*Me=xDIvpW8iQEk<tU6Rp`_*vweHVkvn6bLP6;K>LwvMce$>x0;`9!%4E=$G={
zAEwOD*|U>IXq6j()ak&PnM7?QdC%1D+}A`_%IEyohAZ4j%mR*y&!8Ftc2=dT8cH-S
zRC2_A%8HTSA<bx7m%esJ7x#c^T0{6iVXWRMfSDA1Ui6Y{0rd@Y>T)JU(~?zhc4dQE
zaUykT0SVBL*((*+yljP!MM~rS@dcvS5^`#|$y7{Jihdx)A{WibW{~k%hjvkwlG0)Q
z2X6wzY6-(Rklt8x4wn*f-u)qF%!2@kAom-0hW)M1pQ!J7#8$;lo18m+JsknES?=2@
zkvu6W6EzaDT^X_7eVShOGC+2VbxF`<fX~#xxIIx)XH!OiYe}v&2c|yTf92m)ioC?^
zEGvjr4i6#8&UXv?ejY0<t9b-U$#4<3%hHK*+V}b;FkAHvW~P|*e(UVbb&`Xj1=>xb
z`_{7%iD~@c>ix@ua2UiRRNi*E{!<UDM&J<#l?f6l(SS=i3A8tIk)48<oJ>dG2BUo)
z+8+nMeHi=EFr$#cu5TGZRZdObp6$=i4CZvg-pe$wvO2i(`!?f>TLhmzsqg=zKzcX=
z@uFr7@6C7`mO-&e3lLP(?Whq;28@MnU<>pW!sfaslRiB|l5ksUtm(_43&&g)a<nxl
z?+&ELpsS35X=uSVk!QbE>6X)=pr;@c&3bYg^8ZGgMPro#9E)KMna`wWWq4!>23#u!
z%~#NiMRC%^<|N@tYjhZo?jf9&+bNtP03-%j3n$*Tcq#=syzdgyBkPRzyBMwHV8#$;
z{+GT@qdRy2+Vuq^ynA{I29B4ai$Uw|xMM`x0^<2v-H04N4PZxGS3|^Rcc713FjA7&
zBYW=neiYk)lSNLajxU_?bv)XeuhvD<!w(7=(P*W1zqpt6JljZ0evBWk<IY4Eps!0t
zc#YZn@?*ZA;~TLxB5qdnwmY_@sqSXuof^Ae$Z^Y+F0Pm7>#3n8ipd^q?xO>kAT^2s
z5t^PTmRi}b&j+N6Ik6GX^M24KaJHoo<~*?*(*`auS}7Y<Y(v*1)*HyNfbfAwm;1GX
z1*4VNHd+h)xDWFD7^CqxWAN^S+U?@}6;ihy^J)tc8wPhXhWpl?T4wpp98EjcG|n=>
zFmw5<OZFQ!N$>N#e}u3~dhX16nk#(fdfhjqHzj)gDDHIL>3qjPZ4KLQoarfp?d}&A
zviT$K_!!YoK_4@2eGJ_PvS1l8t4JK)8`Q=a49al+c(!(R_-2z{*%0??IH#G(!LzQ#
z_&5tvyE{26vxtT<WxK0k3wy@p(Xa{JMj>|DbJK!tLO)$54uFAPY)VyfV8b5)oY)G<
z;a%2huZu5BLPdPj9%fH)+?eZX5(Fl-YzUbE+rQbmma9R-!QgGf;JUv30_4O4QY`5c
z`!+SlaEC2n7f(gN$U7jocREH8Qi5=dGi2q>A(N1fqn}7b;`VG;GVSL!Fi=>6Y`k!J
zJmR9BUN@U`Pc84-2F}>%Pc<*aR}AWwxo!+w_&&c^&vFxkbb&!;?fPFm_>a4;e)SNi
z+Da3qK}#`;6Mu?<mc4L-q^?GTmEhJeJT$Y-DFNi>X2y<K*_fLvPg7@VQH?u%9BmZQ
zN}LvXo_dmR+_Ne$O;(5$er1f&J{W-ScA=}DcO8inBCN>bbkWq&wC8Hn1K5g^`++q-
zxNg&TJw?gb44K+_Z^pJmmiJv1FFvp88y__O$q6ic)FXa!o`Pl2X*OXykG@z7Z_sOp
z!4E8$m`2k3A*d!=ojWnCEBCV8(Reg>xp5G3ZBTCO^Daxg3{eJa&OG9Nmbfh<;PGG?
zdb~~W7%bn;p_uR{pqx9+WU#7EJB*^)ojz-ZHO4&Vt?s>cAAPsL-Iqo+kC>OXdA8md
z3M*CLE89Kw9K_{v7~?tmc7&aFxlDU~jy0!ZmgoWyFTyP(FQmJ<JH2%2TY4<MRrHxR
zu?_-o{(MRv)-0dDx&fwOpq4=@Fkee2>p1^jlm!^;ZHevscI|@(LEVG&JOm##2LdT>
zXUh=DMg=8DZ{1uXQ*$)aYx3B$XLS}Z>S0-7*}ENH!9MX|OKp$U_Y3hn%}U~7JY^Ij
z|IDKiPyr-%;LGROx%rMAN_~$nnK$x%C5drJu1)&R6E)DJBnX{dv<fjs3$Xdl-3Z22
zLRjB8Z|IPl!VNwBbnpJGd$*H>cYr$&Cy5P&3q?CX1eVj!4LstVHU)D^gZN=1G6!j=
zpSBdgTKcJQM<1ISWmTRjyT(|Snw}Rar_7hB_Rme~N#f*pAoBV4_O0+=-=t51g-pgl
zIFf29Nzhgtdd@?d<QyZ)FpyS3DeHrm_yq<dK)ILwF-|F7N~smWcp5d*W~<joTdx}D
zKJIn;dcuQ=nGdx?lOko&@k!PKk-Ns7@Wvx;M{_R>mzR?-oo;e9=%K~-r0ND~BZ&=K
zU$nyQ4c!kKeG#UR3(pgG#cLj131ySJ6sUfAtfv=CnC}rAKS^_i8Qh#Sz6h$ZxX&`4
zH;J}2po3Oa+jJLbv;}QN>7+pH#WW!W3*38O!)g+Rblx^NV)GhVoC^pk(C(dbMw89e
zKQ1FCtW|*>??6Z?pZE#2j3`qjr&OffH3hVWSK8e!)QTZ?fWRcPoXV-|X|a9nKe<;$
zAIRJexC7M~*IM@V>#?TTPrE3vgy&fL3TP>fpO&1bH%c2}U|rH!e9dvy26sAZM6scZ
zWrePUHqHX?fO+zfaov|=L>xLSQF-N>!nS@n(SFGBfr5qa+>F{2wT^Ox(>=68`(HOg
zo$xM`gN2a9_c)Je>bsXCqVXFKE`n8d`ApASvKr?ND?G*xVoAkpN?FxbXXs9FLj^{)
zNY*@p<=!5)jL7{~r}7Aq8FtOg{1+xHkL4SOuEjJcP>eA$%uE&$o<wb6x8X|PMv(cK
zJTcSlkBRTCxuFzQi8-~qoLR&mTz%h*2-_I|z}NZy0hj#qN+KVX`4&(w)n@Cx^8J|=
ze+3oeJq|1i?jzMK>i}*;?AQNwxc{pr|F1v%MCd`oqydHJe}h-*SZGqMDyyV-3Treu
z6N?}Re6pf);RPLDG)QQM8n0}tS5eH%p_74SF1%u=?8S-|ft#Gs;F)?uFg3%V(P%XU
zNw(+*isX<a11<A?kGjeU^@gWiqbn>))#IXK+J)?9R8xJUp9r;UR|TT@9a>vM28tuC
zF)t(^y?XxJtN4=ydB=Xx_W%8kZ($SMo*?1{W5|;c6yfccoynkBjEp;>un%c|&&b*!
z-6L%j6V+&vOi4WB>N*!maKg%?BgHV1Hx#b3nGMji;o^uP(FEeFb?W?w?BIwzlsJh5
ztKXP`mb4JGn@PEuf!%Z!b!VV+QMxf2Mwqq*>L#T!Q*#9G&<R(`vsAtMU)S~D1#dsk
zBJbPlg~FBpj3s_(z@uiP1VrTq2EC$HID;5_Vt}J3Fl_t%9ej!W+;0?$_W$s6@$itO
zjHRXGYuXF)e`f(Kvwc5jM%9!H8EbPU8HH2Kls&&y<>!(hD%2bK?-YKwmHh|)`7fr<
z<Q_;@OhV=Tlm8|}{z-KBA72+?L5|~h=E?@Y;|KpEK=8+lGdysS=6RHoeE;f?{^M`?
z=NJF)$N&HL-FFi`?YvxmT!yHll>Oq`ng3>H{{1=s?JFi2C0#QJh`SLc_7STbG-3q+
znP0ea_cX(Q^9TNT_`zm*$OPyOulqTV+=fBkHwh$lwJ2S3#lOp;{<mKF(2Ix4s=!)8
zeB)e^kn_YK7rSt@ndx%z|JD0cM|7_s`!o31hI%%lsNa0y{eNpKaK+8NiR?vu>+dZ8
zdwbCoJ&3~Arvf$qdl&s@6|&)uHY>LO=MDF7?b$zF?8tB^10c4_KjOdj`9<&%JB1gg
zn&j_fKmWWH>Z<}<1w#&*bbohR{qrhjm4Tg{ti_RD<#%Y*KYe(y4|bs8=qa&Fc>kA(
z@t-%{{|LbU-<$t`Z~p(+Pd+>U?D4kC@w9G)4jC_q#mOmitNtYgnZd?_!qF=rzWxWe
z6=PoU4vJ!nQM0%q(C!jAFU=v8(I*OE$Q)I+<3At4%XrDAVZ`rCe&Yh-@`O#}d}W^#
zmU!4U^r8FZh+3VsWTh&C_Bhz(pxSApO(N*-GRRqbHX)$p2l(DiV8Fb^@X=QOA3zS-
zwWAfP0q_kk0Wy}CmHMxq)Q#=>D-eekgNCWxWaj5bl}q@8MU137AhjGogiTLK!qJ*|
z$8Ha1U3pSS-owCE2Y|)~0e2nm2C-X;vlS>*KkSn}iir>=yuAMYmX<fUL`U7IM~#wR
zOjq!|mAj|hXJ2OY?%j9vU4_KRVk9n6B8m%DFyh|z_JhT5wjn`Rk!tYdMeG1!Tgv^O
zOBE1-{3uY1bv%3*bPI1M)8#NVmugu3mGwt2vX?V8BLMr%0_zppZcLm-<DELO(!|Y&
z>Z7q7Cs(||nwS7Oz9G4U<6A_{CwN0o!}};gg!hQ=d;po}_YfS*k){xRyIg!{si680
zB#%}g_Kj@SIuqrvIX4Ljvl9q6oe!Dd;eH28CF>6nhd~|n9Xor-B`q#<+V6Ih`$L`P
zra)IAHUR?2N8Lee$Z4uPyr*#4f41ZEei1jtguu~+7&ZP$hQe7!phu4z%p(r`!S~7_
z^L~V_U<6A|O~Ws!EO9O+a&@<C4-rJdJi&z^cvF4s3p7Hzk9-OJpRb;x*~GDbK-2R6
zbxRtjo$xW@I}rZWbm_;Z%@!O#r0Fjn^}c%W{<ITCEz2dJ=h`9S191J9%fLFI80eC-
z8Cd@reAP`(a65xGAiCCi39>9QBG1J9rLX%>hsewk_7%33eZ+nSGk0s+#csKmuYBj)
z0VV;TSG)dA%5_AXvJKJ!Tj6u{Yvh<jP4z99W4h{*BY&#F%zSO4VF`#=s~XDfoVdeg
zBvbMW4acf^?Pa|Jm{Z*xH!b~gj2t!oy~7=+s-}@r^xV=daMdPxD^J`NbAk+$*eDJi
ze+Ou~N-EYM7jdQA9tIpd?MNq@S%ldJOMsM^!<ge&NbCu4_LB#St1El!amNemo|63_
z)y=&yCCy$s!3LQE+b}Wh4~JPPBpe|GD^tv0O_0cKHwc05q;%dH4N?>9%ymJg=qf{4
zbd$g}cznxDk^*a(8?G3peQYS>5bgWobwWYi9k-n9c#RHA2~S|N^XU@ALs+nv?Zh9~
zdlKF!QmXH|?`ngG?OdX^+AkGUGz&2jE6XSd^)*dO+|NxB{%NkXL^=>*#jw2UJ8J)A
z(ER&F#R3SIw8OV~-LNs=(;6&GLw27!A9804)eh$r<?$8a{_Wt7Bg3?)3{|bV<Pn9f
z4H7on#Wj=o&!vUPHsjA%Zo}YZ9mDb6C84nSqkjd#e>z}?&Mwdbn@rS)I2X4IV_P6l
zv9PSI?-^hLq|1ZRD+s3@7(D?Z4u^y5SITi$(F7nlSdvm{oP>aQGeinZzC^RN-xlor
zjI7-D`OeyvCOedDEeQ)L<3mdbG?Ar0=S5i#W_~Ao1tpz}?9KP0{~)6&t$$T^F|TOd
zRz)4rSPB3?Md2}I25<n{rs5<{0cbC|x|l~01Ru#Pa#N#-{6msM?R5g9=0ixer0?-b
zHfqYt0frML_Cq(`=!nDxZ5yLP-whuw**KNPerGlWD%n5@3-qq(`=CB?I!@G~VH%AQ
z9RZ2S2G#C^k?`}p%lNo=5L`4L)}nTCLDc>f9sLm9F$f{;Qk4e<Zbu|!9*O8`3(mQ>
zKt#t1NVnrGkp&;D?qgL9LD$Xcpg9;aN0q;(I$bAmYhk@ohr6fQiEd7;HX3x;7!Uq`
ziMsx+#1iyCb~{O#bFsAM)qA3X1cn!1w5l8H@uPmw4BR{C!xWs^50b?M>&F$(Ax@mU
z(3`!Q<u;~v5aLY-L8ZEs&PkK4)zG)|!6M0UokGIa_cSi%3IRa~O;blF^U64X+Oe4m
zeSQo7{7#Y|6(n9-t3n~`BZh+%Yaf{15aZX7`vMSsvn>!>YAwCNvD%7F)s#u+A1<Dj
z!uQ8MAAo1+jZO*(P3|X`Ie|X{*oslp*#oSHn>bj0-Fu6#`AOSW08Jj$!ghwi>c0C!
z+hx=67?R~f-qdD279IErn-%D#;1@RFWnF1ST%W=Rge>Q{?ZPe!BK7<X2UZGa@j0iC
zp#r7G8t_?+DoRUT$uxNW${!jAk^Ji)iasgpDtAZ}C1E?GjD2su5k{SD4;lFdV>_qp
zVV4d!)=;YEB8Pumb)USBxqnbibu<!w_uSD}+iJy8Waf5vsPlI5u>6G`vteZa3poAt
zLA`kZx?Af`S)ws+aa;K&?m^0TzsynMd$g|{VL9~M1>zrI)ST_uAJC|k!AzexMzqv*
zIU*YC4RNR93adX#etV{q7~k@#NW+o>H^0V5!r4Zn2<XgOt9f^mQFvEU9xGQ7+$(-m
zVfQzFeBjz#0t9~~Ka{*fwdMV->YmnU>Ai%O&t;#DtBEu3pFOvpJ6#jJQ73%iJtW*v
zy72YUv^TmQLjKWM|3}ce_rysbc~DkK@y|}&Es?vsFMb@~7UV$N$0*#uC`9rez&XdO
zeK15m?qPdCLU0bk^E95X^W`<bwX3_e7Ob-%x;ut$r(YL{4$hglH0ho+F&4CwbSp52
zQFSE_hAs`jqUvdR@(MRNOZaF<i^`?(mJaE)#M|g#hI2ta0t7ugW8t<IrtPAaX`LtV
z%s$9ic7Gt&;jHl}AR-*Apwe-5YQ#5UDXiexh>(;haZ#8pdKJwfI<%5zmicp-fof*F
zi&o(apve2=n@u@)V@}Am6VEjxHBHV$Z-V1v)$t(nOO4dicCohI51vNXf&FUHx~H~U
z<4&jK>Ue`5t8<Vs%8QhjVtY#H!2C6zXF<qkfl!*DBlqLO0O9h{!pXiB$H%qz_U|gW
zOL8lk#^CY3m*r&1Lj+Rel!lP8v1lTGIl(dCeLm&rgwZ|cEXp~RJdg2vry3<+bvg*F
zix)i-UUl^5oA|On&l<JL6*9Gw*4^j*J(E4Y<+YFEJWngxsyDqBNvrU2iJFk$qJ7`3
zzK|daD{QOtR$|GVshga4&ayY{+S+G1m%=aIyG1V4{kr(&rg*6nT@`!w5~IN0`;RXw
z-xl$8)@{sRU-5Vwjhjn$a?Vh}?c-oYTZ^^Av5&*kI-hN|J<NvNmtg&11IFi-e>*FK
z2>X>LB{Cd{J`AimlOk2!&<!n~QRl(caQia>`}nN6Z(M4xzB@BVrtp0%Cd~OEXMwie
z$1<(xP$6SEo7eP6+^x>IXQJJgBuonbL$sf?Vpl{7c?qxzI<xB0xh(z6-x7wy-exe9
zHgMQOb-2IX1^OulIRiCpfs?h|ms~VuzJ;*8PS}KsCM3V%0#7R6li)LT?@j5Z_p6*D
zN5jhqjWq@)4TOtQG7M_1Z?-(^-^I-ht5}93`xO%LTOCg!%8$7<L8Bx35?rx1<vZWm
zOKt|=^-ILZ681D86j*W1kKKT5O}kX*O7{4(RU$jvP@NPZ^{ftiP$gCQx%MqM7Z-JC
z%K;^=Q&!%qBy=A6ei`D|9}F0NHX^^VqD39&=XRdwSH)Lt?MV3w>Xmy3ILoW!b>V5J
zSJn!^v9+(m6j3rZNy93~fBLlvErP+win+E-f`r^j_=EdJ^qK2s=h(D}t%1E<kB^Xg
ztV`{dM<mU_J7^h7HS?k6K6<F9W)`*Iv{u%Dht(&NcIJujIz(aD6&@Y#hnsLZ?OD3K
z#(T^_WtQs~Xl~bMF=h#o=(|TJ`b!cIfgIaXFd7Z;QcauXFsQ1aZR*i#^LDE(NH|`B
zQ{Ap|5_B_5IOVwb8UxI!X=BK=h2fyKUDGf+;jjANaj@VQ`B2%?w+zHNpI`lyDQX{K
zIjUm`hqsO>{@%R=iC{Kaq7WC{LN?&gQ#=MYK00rvs_i;ItG8HkTe+aiH&%C8g?(Xv
z=i}=|P-g0&Z4DA-s$<<$JT-hz*nJ&@Ho>uA1LrL*^O^YqoJm{A7+F_f%LJ>-aE}9w
zG<2k}-AyziWk2)j>Lk<g_WSoEKuK+O2WMHzxC9HGn-#Gw2AdVn!99Z5HrU47+Ir%J
zYu5v)eRR%#;Y7O=m!12SeP9!*H5QazZ6R}MM~^znUNik$_nOZ9o#EYw-;XZuR&oTt
zXsw`}^s?85;hvkM8YmlEFlNZB#ODujgkWOrgI8tB;XVd&h?jBDhzYGnYe@z<GdF#Q
z@XfCT*Uai1PAs3Br~}XQf>+SWNJNs+BQoM>WL8dD#z=4S1o|!SQRTT+cbe{emfLgm
z7^8{Z2~!Y=1{FnaUueW}5eS{QVO0wsYjJ=0Zem>_l1f$gj_YUU=ZwURN-5NfA;)#u
zxJ-el8Xc{i`>F<6jB+sz=Ia8j=WHvZ_E$3ASWcW=e9BV7Yboz?VhZt>W-2m~!*Gar
zgvD;kb~k@`x$?2XmwhN|UQ97AKMb*QiOs&;9+F?i9N%)ztYL5e53|m`tl-_U$HN_Q
zG8V6`sI${2l6mK5^%Tx##*Tk%m9n3ayLohPS9;6%^6`hV2m6;2hRTRnS+kcuM|WU`
z!R{i&daebD%PwQ3mo-fL>OHO{$=aqYC)vhL`_l>Ur^zjh%yA^4L2^#0u^HVH3Ch+w
zGExSaS77fhcPalcTu)`S@K%~_7682Wo%Y{|S7#e<&PD}Yx>bdQr<bwLrWL%|c1-Me
zHx||+Wk|PlFV6KV;n1{t!>?1zqUP`LR$A)bS3)m1iEw<4*DPXb<ZR=#DU894^>SxU
z;hUHZNTHAJy%c&TI#9RI-zqdtBRGhdu!V;0hO0w_OEOn79tBr?g`6}*qf7a!;-t_~
zq6beXN9~S`BHy<J<Ids~ncVLqQLI9$3DKM;G>ll1Se;I<T^`%0qZ2}wpw733#MENY
zjI3W_V@TtsL_|?kqy(En7P9(E&nq=10?al?*76_*uZK(d&L!Kq{7YD@^eb9xFu=u(
z{w|f;9nJ@h9*xv&YRbdlSuz10)JTGt4Wj6R8W%nbAAUhXQV*qZ*u7D`O*N!VeK%)S
zWJl_1>JYm*;}N<gJP5n{*w)}ZhmO=|Iho0eJGb0^P@S61;s>nuu!-=L@HtUN6`~!D
zQm=3^29l~1+|G6(Bf$-FSH8*e4tDy4pOe@}`sAN4v59zZerWxG9Zz^hiAerdZ>zdA
z(P5I=U9@PwcvRg8#Ug=6wJ#Uqey+pjrX@U+I_}Qn$$ze!j-Zli+b{net~k->j!z_H
zE+yF9jgtnge#p?#QMJfl_gy_Iym@vLLm#e@R%Ts>fG$&?_dB{H*HU>ZLmt*h%2;-o
zbOdo$Ho8(#PWxQ5ua_$zlJhh&=$;cr+c`J1u4IclHprd1)<C$1cb9;aH8*GDu`Nw8
zw8~A`y1w4F1|O+Bv^Xee&a<<1l|?fIU%sIA*2kI9opDd1hS7KX67^*sqxZqBAb**{
zkwG&PYFuzi9APRxSM)ezXXj*Wm=;4ilfyaY4!J^j;GK>GxX$1#LtD@p@FD_DYt$Ze
z;v~aHCJ=rT;<^I70-WE_RA)!^cExY{T*PTS3BSd^1}`@8R9iJ7N_7S-riYv9x%3P5
zhsS<|C|lK8<;Stak+jSS!3`D<FMl$k9+Ku~2xMXuSl_5dgfgZyzJ^TfxXQ%ys}ya*
z_lU082<+edoH4pvk#c&sYBTI&#^)nixd;^XacF#&fDY}o@RI%wgM4U;)mp0pIK=yI
zw^(ced9bhNb@$T2cS;WtUk;}I&WjsPl8jUv74Nc}tOOinctRB4QCQ(oF`bqu6)$8Y
zWWLRy`lW*@e4_s}`HK-8O<RWdVIo9LCpdLZ++y53MXHL6L)<<VJRP39mLV1s>>4zw
z(ID4&dRD~M-ItUM_k`b54&n9g2;V1PESREDg?<IWdZ^hf_3C1A?xwH?!=iO(#SNSh
z@u<6nQ+A&Q)=KIqmn*}BDXP$Yj~ZWvWF!{}oAzFpwVO|+oof~_LbUBRVCvu3i=B>P
z?_}}YN$x?!jQ;tl`Eg3OD)=9I3b7n&)hy=v<<>RWyY0Du{g_L5Yi{u>@nHO%bKk4I
z*j&PmTVE_BPdWMB{OJfTn%yl+Xg{~ym-t`Jk`NQQnLKbk5?bckDBZ#Ja^|M8;_7Sr
zEM|L8xe|SH+1OF!zDm>WJ%bHATlR(7prY7Yq2P0CJGNG^-@O#_^~@4$c+83Mv(4i_
z)kXzhXAeARd%ksvr|06v<6Zp=AFqz%242W}oVj|7+CuvE^op6L;+?cw&Wx&J{YzIt
z8gqE?UPQ4jy={A%GR{nF&WgfGt5HN^-1FU7m%J5&mj6JM?f9i>&DRNid{RkV7tCHP
z>R<g1^^x(d9o+yM(Yd0V46y>4husVnXhto%*H#>5{SWFILvnAw{@zjky48tmt*om`
zcOcfxsrJZhfwWxV6)xLl2SPT}{Wf_k*_>}4Zl9g;Y1K*Z#LS`kLagHF%04T-^-X)K
z(zD5P3w&Mg>-nt~GxG8L>zR{hz_OxIY(bILz}`*y+Q_^mZWxd;x~u1cnwbNK^W9eK
znFnW}#ogXVbj;eJMb3A2TUDRaMhyfQy}ikC_r+|kY|8G<v{5EXX%ixwoRlZ)RNE=v
zdp%XONjJaQ%5G|fc7@2+OE5%kL>pDHBlt=5T)2ns?Z*9_Gql;{1tdMMPQFNc9Izl=
zRL`(VyY}$mA7sGDRV-a@W)6Ju5Wi%)efRqOfbP{a|2l@VbUAC}y&_85q*x)cj+kBE
zRXJ^}<odFTU?X#(Y{}*9^@U&R*_p+%zFkW%_(U9KMlM)iU5olGD0km?s}WTr&JYx!
zkH5%Lh4|5iWOAt@m}sQ3xmk=uWM7EsiZgS}`!!J0ksgWZeqrUHQPMtDAA+aTx7*-Q
z!7w7dYoJ-k)H@V~%g90eP+OETPVZZ-+fMti<UC8#cVv=SkbaYKRk<d=u>mO+xh*~A
z^vu)V>mhoC>jeqX?RC_q%}<_bpVD0INHVr1HQhV6_4WQN-_?O?l=6-c#_vgE%f(VH
z($x^Wy}r|i9^!4p5)`B3`l9L9(*gYI9-|?6`Hpd~3$DjFJtB@!ujrxTi3kx`+)vVN
zpw6CDqa!7dV0g{6e($&$gBM(Y5%}5=0vBUN<~=SO$1Oc=+PLiXu#Vw_Dc-Vc9B!Ct
zjj{I}!s74zvPI?xKAOrL$8x-P^kzzh@>E$HtBxs<JD<xfOD@50+VL<_CRG+C&Tt|~
zCOYJXH&4V|(^FP*FCpGpG^mSc=Q0z=554){?7iV*jKPz3<hR}a_%Js(zOm$&s~fvw
zNkYU*mdgH<XE6q2mQ;kKdG7|x-`NzU2%oUXxkA<}q9%W!OUcSV8TK?~ufc-DbTBU8
zi0rZ2WIE^iinOXBme8?Dtbl5h^Z(%NtfQj*yLK-~N;`B7Ie^4a0@Bh*i*z?gOAg%~
z(j_gRl7f`d-9vYGcbDhpdDi>9zw@rM&ROfMHGl92jLytG``+KZuj_N+Jq&0cCUnMX
zkzcr=m<YjiC!16je`kL(17}~KK*&U|OhituGeoH?WkY&`wwmC!VD%2AC!KRnFxv6q
zn3jv5H{Z*Vq5{SBOqv~=072Nd?!&VtEAX)7rIBn=VceB@*YL(E9@>#K<hY+m8K-$x
zI6Z8;-)(#e`r`STb`(7)B}q1v`PF^N3NE-SRleiP#VF;Lh_wM9dJfP|m<m=#<<ph3
z8XG))na!C%4XQ5c6jR9_wl1iM=hi?zn}&!_%w_&aOg-bK5OGQ=_tnTYA6+}Vm_#;Y
zD)Z{B20ZwFYF$CvZKyqq*o}qP`A5FVyc%M&yBJJGw4ucywSsGpG>$cw@JfAF@x}ds
zM-X^g)Ir9~ev*x*N?nC?{Tt-i<WmF@;obXOd|@(k<nBoBSLg?+_|@fH4;_pN`PKLI
z4CeiI#B;qmT3op{>e)0p{G7&Z$>9bNj0DidO7e<XdbpUT`X~8i4y_*3U;ow?weTY0
zhEKn&*QYZv@RKlqUD}SoUX`=f>1qD9hDK_ZBuF{zejra(z;}@5I8vI$+LN+5*;fAx
zZflxLBHB*{=@{WaEtb=JE1)Z$6+wAGUt%d#J5|4Otm0q0dkOUNj-Bkdew^syOU&mk
zuRI=?!aYz9-B$|8)nBg7y=)aOv&ti14HNxAix9@jygO#xzf%A~h(on{-5>FN3r~;6
zfyO&?!#xX^F{GERDrWmEPdjTPoa~AV!ygZ7ZBAcqtel9f1DU}%j?coiqpn%ckACOP
zSN0Y^kbRIPBoX<oM?ImxLHaO<drkRD<Fn+-H2g*aHAq|I>3zKLE^1Iz7<U4;KXV&j
zpGG<c4Nj2q<%MTDk%*y0mYJiLevf!|hTvQL_q{VI;rjB%{oiZO0$O6Xx>=lS1&FYf
z%Pb?^yTfb2l9OpKuaReMoAoBZR4BmvR9t%q&9m7oEsyxdHTO>Zu~ZPTndUa_Mpc<o
zOatS7xAacDo|CzD13ltkL~0CNKykq0sI-4M3(V31MPDY<F_-m3oYtmYd}6IR_vx>(
zQdt(EiIoQ##Aat<ax=`p)2=@h5Aj0|4bV`>A#OZJiw<fZ9Q6BH%W*EnX;+ss8#5AV
zD~Xbc7@mM1ZTS9OxXZL}qpXLo%=W)=0bQR-IdQfnN{R;8D5e$TD{Q?5A4UpEylt%&
z?v0m?sU6k+9%RKfN9;-wBpZg>^^G`NDSlNE!%!%Jh&?EKahzA~B%Nj!O6@0-%)f#&
zHkPVwRzLl<-9|tT{<DElE@ydQHk)_LE@5b>Wc1pstK{@zD+XXfMaZ33>Js{<lJjzW
zxdK08Q3_5&Gs9-wtFm5V-x&-f+5-p)AEuQ`kecg&wyJ{MYeW5~Ja9KatM&b*wr$bC
z3x9*n?h^W2gMpN)agF)bxL=h~2GVk}EomDE?77*xC~TC4|C;pXB%w`HWf)*5Z8&OH
zvd?67XlYJ$Ldz`;miTwgXPe#~y6EhN7PM#@!#h5hEq+_#P{`8F<2|$jPIFK5Y%IMz
zUL{Y;SmrqvY6;K&$^;hQ$v6{P<mA}b);2y?OCKM#E{95(Z|JpdSVxU&79`YtZe~r%
z$eM9(B%5aXEfs9Hfn}3*GadOuNd=~?V8bt4`Q)m7MtRK$jzi7I&rN9anh>~t%v#Yy
zqb5zPu!^+YRwdG_6Jx6`Qxfxf#^fx>7w&GZmd1njsJ4`ds3myG;kDV1ZHrNRlkegK
z6;=2(STT3>egm^r2847zdU$h?^OMa}t=~x-Y(L)g2{`7HcP<z_o)TFEAW6?)L9W-f
z*5+6Q06G2Ct&dZWK=P8A6$QvfeM27h)uwWQ;UosS(-$8EI6>WY_ZBuA{WkJtLAthW
zttqs9FOOcWv%>74x8G|GZ{`5y@(k2kqM3K_WZ+oeFoDskNHsRiVS0+SnuQf~J3qeD
zAg$)aBGU0mjb%+9h844sDnGCto3QyV#^SSC*%NoZIf(Z$0$4#28Nv*c3pW}i56KYq
zM!?HM<GqP<>n|HmyjHYwm6Xq4Hm*>@b{}c_%olHegz<A#Fs`PUO>26jUGhgV|Hw|;
z$f~}sCP4OjL$3hoD%*(vAfk0A1>5tvvus^3@fPe7jhDmdrnN*wZq`+1HF<Ti_X{BZ
z4SS;v!`Cz0kXD9-ZO)D?8uDjCd78@2i@#krc@{XTl^LPywdD>LHkr%&ZPd7n82w3K
z(_Y+ihzcItq7ZN`S&IEjTXHOnE24kIbB5SZyzG9o)&z_cuSzOx=pG#1*&*>>BA8b-
zY7Tvyxb9atYX>WtSWOM83E0zx^`CK0oK7lMyVvHT^w{CJi_+oDXD>SdG^{H*&Pjv;
zj_={%#kqi`3GtY$ljN58J*8w}po6TvmyO?5JCMcDQ<QMFZgEe+;p6Lm66#C*9)*$i
zGWQ1;LId<{o$1cJCwE^sh+;7|b<b!&(wf*JN3bRB6G2woyNw73hZD{YE~z&(cE6S;
z7Bk#=!bO_-4%?^ko5^Y6Otdx$*h6ZhSJXqTA{)WhhW#!zzY~4!mi2Z#!2TT~x8EQi
z9UwH94{g7G(_}HsL#<~N=SPoDeXOOpo=}oofS~d-H}o|{zHUpb`<c9jtY_U@&Ab$k
z&lbh$%;}PopOP-`KZC7ql1uu<GTyU#Yu&X4B{kr5^XCt(3n%F&J0`v}G8NMQAVAtj
zY?iV0<;|$Oo*JvZWOd-gp-O+fN52HC(bR6JN4j9{A|{*u-w+e4=3fiB3!68q{B5f+
zdL{QoO99K+-S1ALhiR@+{7qq(&j}N;R+TbjN%D0^LR7ZYKDLW#1sUPFh+GwBBk8nZ
zdkb)3uD4r+tBgji5!>pFnw4@TOfg)Ahn9X=ADE6en4M!`WN5s8$kX}^ez?6jdm!y%
zm$4VZ)=$*vqqj9At<r8A%p+ksy`g#()X`cr*zQ-s4#3#p+YenbrD$=IH((M*p}g<%
zt*Qa{-r=fqjlb3!!?Vaa*rXkHMihDV>27ouhy1IHm^I!thn7s}h<ti-m}(lwPJd7?
zYyJiqlvJCPmA|!t5m-`sHsO6&TXqAuoY%eds+@OH6D4YVC>jm`1eII!vY<M9zfEa$
z!DFxV5Ewm`@;OR&N?CkUm;U|gb=L`d<%h|Wy;rtp#y09$3r`b526UU7RL=4hv)Phc
zcmH=Yy&c%bcoL9xIceyJaJlM#fjEWIpXCV$snvLyPB@RY7F2!g_kykHtHE7g`lICG
zOW-?BVFIB_2QWb)n=wV2k2^BB1Z5ZpQ%^g7oBA~ezKfJ0K{2MPmQ`e>?^(NoJrb{>
z-e>|p>VcWC2epuQrnbcC1?Ift-{=EzR7f5PT1@SP`NfDWAcG%D4rW95Sv`^D>yX9@
zyAk-LTBPXGFg=E|8Z2^%3!_pjz|rjGI3AYCMR&~LbIb!j6AeRh(#M5y6e#tdMex<A
zAd2W<Po>eD89=rJC)SqRbNCZ18Zb4S&Q}IJ^1;W_&qBp<rS+d8$4i?ZZztH?%&~@2
zBELQv097tFNfh7nMpcK)qcFRNe3OG+-d!H-@ywWlEadP67LiAFabd?phH{ieFvp9q
zysC!Jq5K%sVXV?wq@fO<D|;ngAc|C76JZOSB6V|3#|6e}Avizf<X({;8c1X^`5Tzs
zW-frKApfI3U`~nHx7a1PCYXQ^P_FH2I9Ion18uJ`-vMUc-N%?mT#(F@nKTrvppGHE
z1+3rF6P}rj|5E!!V2(boG^!#pe6-n)*OiRNaN?+#-hvZHeq;_T!K1+$lG8&HbVtp_
zC6y6eaudQ)9}_~)1M%Ljd>eBwm%XzL#__#&FJkQ0<g{A#xgRm8DOjcePiJh{{O#`G
zd~h5!X6|#Rem{!@Ibs=jzAV$s8n@AX3PVi;yas<BZ}%3S_-hQRZqr7n(mFjI-lHeL
z9<U0Yj13|@(xvcr785>=>GWy=JRDmhqegAlz+m<K2o8X-=h*)2uhxk?+}CkME<a|X
zV>ifg^4Jx#bu=c{L=J_%Zvk>zDH8Jm?B!fD?OE0Up9ok9S^Du)6bJrj#2N)l-Sq=r
zX6xybFaR(toUPG>fqCa0#6^zU`I1Z7miv*6j9I9K318m&tuvS)0hu3A?`6?>wCt_O
zSneH@HHN{Hp`ht_pdh|mWRHva_W6o!@o$~5!uDITAyjt3qOU2k2BYG)Hx&__r@#`y
zV}<$)kZQmVt&1!Q<Ws>WQEC-AlMNF_3*7pGlP~jJuuh}EZSvPImE?hVWwl&AL8V=$
zIjgGz1GqaKoCo~HnHv&EhyqV%1@snTG|+**2Jn*nna+VG(3Zxr>@sOYNm2j;e)<zU
zv*Z)Ps3bayqx%7)_pXd?;|pFa?*}J<j!w<RyxQ56MEX0SVzQIuulz8d9xh->ochh0
z7_{+b#XHe8a_i4~J+%)4sq_n0K6|VnE29a|sMz<?v6KgMG;KML1^Q<9uM{G2n{ixd
zz;$Da(=WW5d>~51ML^-G^|~A6asx1jw>)2ITa^X(S>_~vfq#Lnrcup)k&Rb(WIHg4
zk7q>foQ!CB>9TkH?75ApR9HfZ`Q;rYIp#XDw9)>utIkM42VJQvhl2*!?P5IbWN)md
zSM@d}%6_zqW?tS&vJ!3O%+42>;%smdw%%j%6W>D4{kO{tFXGyTm><kXqVO-g#_t}V
zqzu<SLU3-g;3lLHM$4_$7$1})g4|~G6g*m~53j$pU1-h4nwMHv7SawW&J9ZE>;O86
z<@b)N=qF-H&edr5sdR;^AR@tGW%d;r+D-{_&?jzn=MYO703U@VqTKRak%4Jvl;wIP
z6-4N0=I8g0uI?S82*e-_Ng6G+M7OOP^EYsJXHS6IwJMD?YrWdjJ{tV`2)s*M2Am{(
z5r2+32dU&LRbuW;GEkyJqTk_h1pen3<-U`fZzt!YF4-1I3VnQX?TQ~y#F-c-UWO7i
z)>)(Y8!TQ5W>MBuy|}ON>XKtKCYT-4CJ+3R)O7h{<Ca?lkfI!WQwU>z+iYuL@lMvh
z#eb@kzPBuxUpO_xI=RTNOmoop6wmWVA@AIc3Xfh}4paFmu>O4-*SkUs^;nPBiaA*t
zICeTDGFX%L7+6x>r4#!^1rk&=BkdAfaQdwszLtl7tE7~M8J)(w^^57lxOw-`{uSjx
zog9R*{w6Ty1jlxC^%G$1y_Nm)uc8>|3ipVO^Odh)_m}YQGsFQEV$z+kwFrpLw@K6~
zaiQO*lAC`nOE(#f?ggJ_`<E;ebDJLrzTNn^5l}hwY#ESzZ7>|yHTCeTA~t`fJ@6Iy
zBO;3V`%g=GSY_J3_7*k<q`M+w4|=i4&3U(J7u|vneTRVj09LqIprkUW0;L^T5l`ZF
zRI0Xi0meo-_2^WtVig^!Oh<&8>D@HMCR1Tc4pB**lYK~*mDo8-fH$!A&F!W9J8W{k
z;n4OVqRC;Ba!FjVW*j4F`eGMA@nO32{FQLJ1v%M@>gde-ti@5;Q}@QZ2Dsr$`iY~2
zW|WB`b^4Xuu5^b4+S<1Dn24E8Ho7N-J;C^uu>PtLdUuB)=qQQ<B<z*;Y<(de-LAt~
zb1KApv^C;Q<T6_1RdWo>$96K`sDk0<!kAF{JkmqCbocUAanjpiG^*7+h%m9>!}(aF
z$l~|3`!c#h7Jjgp_RU~M41mZU^>uhR19^zrV?V?vKkD<cl$eqH(hWY)b%-Ky$eLhg
z_wnO+9%QeXCXS0k)RC;2`u7P3wS|-FDcT`HVua0D$Ij4Nhj_F#@Ep07#fK5&o&!CC
z{te3<!<K`4sM&m=GrLABSJyinL5;5jZ=bD6hFz)yz)IJt1YvbSW10*%=w|Y^5;ec%
zx6X%D)s?gjVmOBxBU4X-{4j&`IFPa0hks;39?CS&-qyn9*Q`wr6V?10yobkBSPy`n
zSmk3*fWQ<Xa-)<9l_uhH+9StAt}dPg*tANW7Lt$T&u_K5Dp^37-{zk8t`qV5BX)x{
zXS5imTK1KCLriodsn;5CJ|keWJTW>#&=%5gz}V&RL|rxTed&Aqe1Q4v%jp+(woWc3
z9SOZ89(~mh0-DjbtXsRRpz0vKVsL64!+2wz^56GL#)FmV;w8Oxq~PSS<~&v-a@=-|
zb*zhLF04jl%`$sXHH_v-pooIc%fB&LnS5LOZoQK~OsH}-x@4+L<}G>p{4Q_zk8({f
z&Mf(?9XCY>R##N}`>TBbO4EcTao&CZVs&GAx0I{>BFU${8}214rlZ_T&`Cone8%i`
z{PrPQvyrm<0+1wm%l28Rr$}<W?>LLwdI(Np>g7E}^2R$LTeFp2Ve|IxBY&W~<vXv|
zOiHl+QD1T(T{Q`Gi7L_IL$47KJgE;qSjfhdsH&V~O`;itc`H$=CgA7{UvRZdGCbJG
ze|0fczGK;!n#_&eM`ox5=Bz1sqbKVjY0qM9PT#<$kLGgxV^dzwn{9=chu_e6YOx3?
zYECjHQ*d24n9Ai!P9957%1BK<D|np2FgC@qFon9BP4Rw0<tYq-o)b<IdzxQX&a>%O
z(X7JNm@#ybxX3iaZ5(JTW7PjRIR8t*qMhc4MJL6>I8BW>`*xOhmWoWYY&ZM&q=bDO
z7E1M*1qh+(u%<P~b*l|QC!MR3sG~w>+aB)##wj-WGuj6TW`W>n%=7+)nLdZaHFAGA
zA0uHH_~2z4$QRdmtlp5RMILEm2;b6@zM({{2li!+e@kr3uo$Hw+YtkK65ktnIAV_A
zAJ(kfq?bhcJ(_Wsj;-_hdr+?0?RLu>$0ZZtQhcBY2_8f88B-Oehu*vC<zS1sZ<Qtn
z0=Imr%e=6?^ebD$_RzN?)wI%ohlIa`+blP)Pe2O2NC)4`>}bxSY>Xh~E<ztx>X?7X
zceIySWZZ@{xffW3TEx0NyZy$Kn&p8(#&>MRJz}~G!v^L{PMKD1BRx$LCB)|U65CE^
zKCMpLy(gHnhFYcxg$E7ORF!b+teuZTEWEeppWw^D5g0x4N}@^(F97~v9atWxK)H!s
zRQ!B<|MA(*>&Ulp3N#<&`Hvi<OWf)mvwk=KSG_DrMR57TZscd*U8S%6L4<41Tj125
zS^S#6T_TbKR_Y;ay~Jq0prmKBI`YGX2M_y-!%=bKdJoX}0bp=LFt#K_EK<G)l!pA+
zBm533=Z79jlW`tt!#RFMJ+}L-s&92^=ivESkHlNEsVb8;@KVTVY(Y%%_)@fiCAeC`
z{0*DWx0ShEaKC4=LW|^A<{jdqt<01wkKN9X*dpUUuPw;uA?GwzVtde^1&uTyu*4Ja
zrJCy9R^@qjeV8!DgC>_b?o5y^6FE<h({#LI#yFb*(bD(Y%$U$PHvxP8k0h!;E`3bF
zmy(dxDEaB#`Z!9^!;OW1Jw6EGwjL=aV;auIi{ZJStjnkYNk4nYTGz$#*1Yp$u8cMH
z_DM_4b7m!5C-N5Unr0MF%&c3>?3ci6isF!JyO()wfrtWD+KiX~vT-*BsbJJIMjQ@1
z$zJz9Zqdcw2)U!rHo4DWZ;LfZoBoQgFHHuqJ3sk3@$>$Rw)Ep!a=7XBrAm6UZz<gP
z)Z;XPv+b-4?84+dQ8FZ9|L6fqob<AyBm^Q@K)cr*-`#^0T1eBvm{uEk`T9^UM9j(v
zQ>vj~2+cLd(|X-afuyzDhC8h}m0g-$XqIFZ8tI8m7sVEdCs*$8a(jN#@|9VdxmI`G
z-I>GPd#WL=epmOud5s(@P1$p?MHHpaF8FLeSkws~pdJJOiv-Dp-vK95yGnNzxeGTA
zR7&2`#8G=A*cp5xoGR*QlYcoCZ3r#BVf2&`FQ@=q5zoc^kLX`FKR!o;JNK{gDV1w5
zt&ZnhBW2CC-W|9iU##Xj;)O9sMZvkWSW*QoRUf5Jg~6~TgKdmB{n>_PdG772gXakj
zJ%7_DgqN^qQEAM&nGe=W%hL<WdNX3)c{w*&qaAnZ6!E`G`bhAO7RkriG%ZTW8cl3<
zXPU;V=hZRB6oc;=l8^Z&&nkj(Kxz6~XCk_)s*uK+`gVOu(rW^P`%Wqn68_Xn$-r`D
z7@orKW6o&es@ceR)&$%c4W<z(GTTH<v43y4Pmw-m4x=Wr7OWXuCqNY(!!SKK(E1Eh
zSznWiY}5`0B>BFrfPS6JTzhu$CIJtC2@#BBG0<kIEqCX9_jUn~_;dy<^g6T)gE)Kr
zCmq<}&;U+hcf*$fkCLM!<s5RT1w_BK*FIka)A|^cZAeB`ii@-~g`Gn8<i*rJUDAIc
z@?&?Oi|YwVseC6Us$U4e=(;5kp9?0fqWba;;{Yueux*W%*=Oz_Z~f~|$&MRPM}I?O
zoy_RPf1Oj9PK@bD2KUo?6^6RU&~x1qEZe<qYC2CFDAPeeDeG?lU`c$MJ!;Z=IG9o6
zSnHKfKK0eJw&JB#hkOt*a>yq^^1=L)R^2&DXD-;bp2Rx0ipo2>n1Xv9rxC`PmP51P
zC_R%8x0}Sff2OPJu4|epdnYLvX00#%1+SB{^`pA+#~<*s#j3a|H&<W3U#ZohJ;|$z
z3WOJLTreNPrO=E2yXuI2_*^?AKj_Dg!>Kihq#{%n=3_2&RqCqXMsS^001=_&hMlgv
zUf9Y&ZMkWjb~}J>?YIbV4EP^P8@2_Z`(Szk)Kvmn_@6AT5ZTh_mP>ry9XVRrWuz!3
zr42Y4YP%7SZQ&(tK@NIZOYE)#p|%r?Ot8?m9f9oM(rHW26$V}zcif9prnm$|31b{Z
znrm@f)P%9YQZ_Sx$6EuZs4QoE>f<W2{kcZ*^rW;f&OT|E){`MQf!;cq`QT5IX-3$E
z@#0=BsTd)st89Y#6c>(zFU#RhIuJe|*q;NcVJcxX(re`^fIcGVohCJ%z@yZDhaI5a
zr%@X7lcdgwC7)13VkTxORMzb<`%<6uO2aC{%i>Fbn{yvB*l2PMIJ!^VdEu*+xf?Jv
zx8LzMCyI1gd}9b)dm!qP{prY;eLu*hNZhs?!Ot~+_jg7B$6YpJ*Oo|DQ`DjNh0GAi
z+fTrHPb*-Z4ucs(MpKj%gIN{<9Rm>&F<#}<XBHB=Fe;;7=9jvhjKfQ?p{4Y4dnNdM
z+D?||)?vfN-GTrY#gF2wva%m#9(!M|XKYAUVP@Vub20CU*fkVjg$v~bxSdq5@`blq
z6KAk5T}u})udW`bya$sH_`14{yPZ7~hKlRsT(CqvbpA9ES&58V8GA)t_4BdCibn9~
zEh3)MxQI;UL%Z^7_UoMGiEK$tqN-;DId=!#OZSVkA};qnMN5g{5zbtvX&CBbdCQb?
zCJQm#&9QjiAtK+x9TcIYJ(M~}ZdK`VG;nE>BPj8X7;48CEljk0tLGf1V)c=&ETj<g
zdd|oFiz-X+2__eX5wG(X?@-T~{Dpzx7`5E<Rxm;3O<mc2%-kA7u?-)@wiU|5PA<N+
zcyBx<m~EoWSj2jrJ5(UHj_j}Kv%fBu9uj~OnIZ&f+UCZSzBMZPWHiCMllR`j*Ji)c
zjd_#6`gI^Gj^ee#^4NWr29mv_=cU&wy@F;-Za}Ws)~y-JsS@R=Z^L=&To>isC}!mp
z?-Uw(LQ(~7ma}1AoBS7uIS+5zpgMMeK+egmIn5+2UG^bJ))eA4cC$V!k6$SkEgQ?p
z>hg8$$I+7rhy62J2S4}Z5?T&kQTNgXkcSDwB(7W~+0&hI4f{Pi|4@ha8Qov-<n`(c
zl#Voa`E2mqXnqWT-<*!VM5V1b$(P8QEe@=&LYyWJCIBV56nGn~tIs-&AWh!L^u+~M
z#iK0KN}gR9uL8dVk}-1~meWsIhXwT?-wi~6E_cmm-LMBBOKdc2l9<*}=LVy;ELL=;
z{U8``!lsX4UTaSP+(YmYTJsm=etK$Yr=y2&wmGbai+{D{Jqi`43Yd;e@`ph$b+cs|
z2r*4r+dBn^kMid~vCNH9QwGvA^;tG<(Ru50Eis9oTgh0Y6HL)4Z9*We0Vr2(^K3Bw
zjPT|W6Osn{>;^s>(ZNNxYi0Anf)ot3st4xLNSQ!kj^Z~=s8(6~MMt6E{17C{60kN;
zJI5S=H36Mq)eeAacZg}lac>P=vND(Km*5d9eT)H#oWtrvcf4>JE+}1SCRI;63f9pv
zme%8Z?D=Ck6l}Pz429j>8+$Rx+ndh8&~?~_I!|;5(b-$i))wqlzS(C%9zUB5k3%h{
zbOp8)k7<tcA993q2aP0|HvBlHU}E=!e&|OkFtfZ8e$Wcl2APqQy1d+Cz0!L0@7h|?
z`u10-SD~)aw}P<-_9y34KBMVuAa7u(w<It69$wUWIF2VT!EV*zgs|i~{)gkQ<lbYR
zSyj~SZdiZ|U;tyUF+<C!av^k~yg-3#=!1Vo`IwWyau*`uC~E|XZ%e2<ZBMEu&nWxV
z1!Lgmw^9zw+E*q*$+zJ?SPfOr@5Ws=eWtr*x1BN~8Q}({f=R7vj)3p+oB~1O<N3uY
zp<{t+Yu>Bfx{{uhM+$GU+r4pyi^nI6Q0((aqQ+qJVD!e$sh-E_*r&V|Pa*O+{-ev8
zOJ;AHyPr9s$P0P7Gw7U)zTc*-k#`+-&pq;fD&k!e^h%RRg2itBiv<8oevu(8zxzCD
z{g~h~to{a<GrfO!0m}3V&zlGR!94ZFV6B1oM>ob`C)!^-7qi=p!=mFE6gqu!svcyM
zjj6mwB?vzW6Xw1~x>4M2pHo+=TaRvSd0R9f+BKq^uhp9+9uN~{k4x)jyX#=F80J43
z`jdH2tBVx_8Bp1=oP7k|>br5%8WwTR<N#Tn;-ss^v7w*$WjX<5y>kZaV$MS`SV-;Z
zGqN;TlONSfeLuvN6|aD?#vQeGp}yE}gIZ-}dFkf6gflv4GZTYYD437N@z*g5`e?Uh
z5qBBr5r4w9c|sCvfq=Ni7j;GbqKLArR<u!OL5WOqrU^KP?5&1lT%IB=nT@AavkhG{
z+=NQf=AuP5T$5OA-X_+nKX2o~WRL#&CQn`b>3^IR%eiPk9Q##Y78)NBi9#$FgOR;$
z&Ec~kBf?(EtZm@(PEVB^92Y7q5GXo_?1P4~9i>dt{UR;p@LJ%mm+p0vhW2u38y{tN
zKg+B6Yk{ML-r`>^IO9*KnC9)Xc+q+J9+kIer-8i_eCUcrDSVDbxTwf{J^&Xoi;dLs
za%@_~=<9Qk0;Ww>_B#p{YFJU#00Udpw&0U4@ln8bA;8^zroEbAi^~cJRlbdjN%u&S
zJu^hAiHjZX$37MAD>$ub6xMtr&(+f<7mBZfupbsF1uiW4)^Q6eZb&`uEknx7+jNI`
z^njs6@`UdpxXBf(QA8?ypDD}yKTEv*m(GXk1mU;yLo`(&Lbs+{6-`2Tk2b<yWy3%f
zVuWkBtoW`S3IdEtVt?>eLCM?=G5COzP0Ess#Bf<xAL;v^BU7MEB!4D4@~CX4fh*%^
zRstbp4Zz9Ayx`KGC9aFYT-(w!e@!otOL``jF|drzVpSZ5lia|4(i?<eOO*V$tP)cF
z-TmO4;Q?dTsH`<&=*c@YRvn>bysTfrAMQN7Scv-7gdELzbR)dsEMVYde(Ya0(b&uD
zbX7tmTJm|_eH~z68vWzcJoeZgL8PfWe~c*SOO*`0RhXweB#7i2^*8Bh2V6n>TM3ib
zZG^EVEdd9Ka>NsROC3(>4-yhBR3gl;jU)|Fm|X-DQqQ{sP@XqdbjCaV$-+UgbZ;{1
zjC^Gl_p1fHtfE5=$x6~I!2&&S4M;R18$F7C*h~dyxeQ6;CTWzpcNvpb*!0kl95QUR
zhdi<ye+9yGhG2%&&uW)Ogb4Q|6T{7q^h(!^Iv?^x<m&0b4qv_Ehz{Ru#2^dJ$kL)<
zXi8Enf!*1*=L%7EzAL>!simymmA^h6it{17!S+RRnK>92nZad{N}JcTF0F28%2^P_
zZL8iH<6`PyaM9EkT@HQyEGSp=ZI0J*nV7dfRif?v&oE|aqL#&KRy2RC8(kl_)uzvJ
zQ_zLep=_gUQmXJS-=Y2eS|;tC9Z~E-hZ1Ce^uo>H3f1tFe>h@Z<i|q<7t{;pJk7r6
zleEOYY^57@)E5k-W>~Zhl2j~`)uglr!Y~+jEoSX^?H=8yET|!oN4Qc{Q3Du~cWE$3
zo<Gr*#CuY@_a{+Y9C)>zN-_NylDAT36Axb_`D3<7i&5r@X7{F($I1=CCXBySPW1dd
z3!d)C$LLfaj9Wk0BOst2>CN<hpT26J<$1XD@NFP6w2kNl{dc8IB@*_;P>IrmlTsfY
zvGhP`xdq;K3BFy-%W>9D^&&?DL>)KVbp8gkYuxAb+i<B?{HsVe+92+WtkSwz4(zO-
z+9q-xM98G)=sM``eG%##JAo$b0gkV_OjGBjEvMykvfut?Voog~kR~w`pm4*7ZlS*!
z>#n}@{7<#RIASbsV0gJ*b(Y~nYZCM`$`QUj#!LFIayR!s%3avpcahOGnd(toy?Bd-
zYx7uW6~#c;!v4`?6ZmWGOywXk{srxtf3=tp?^*Sbm5XQb0qX6uE@l!D6Sgq7cbgs+
zp9L~V_KiHfV#|_hfW9wsFuZy4UbO+TvtB3YZ$RQ=-k^91dakNln<brc{zg*s>st8I
z+WlKr9PN`jSKpdf$JQG+Vcw5<N6>R$VqOTl^t-=dzaa-4T*BQ^_1vTplNy=tr|<`*
z^BbIXmvB@4`ZdiaY79Xdttc0PSXsO-SQ#uhknuAoAn<0*5R|BpX^-=Xy`MVCsl*9n
z^OAYm)SW#}uYxGcFtE01iL}fxpvn;B)=Db5WDd%2g^&d@!2+=yP*3BqJ6cb8TGeN4
zu-;RZ@{nlEbP;9jW0@y6M);Rb;y`e}ky~e08&WhH7F)lXyEcxtspxGqgx(6()ZVs3
zOc19I4<?I9Ov{m_GbAt|xac)MxHT=frXvH27W4@P2MZIipUjD|s!$-L+HlhB)l^OR
zqG90vibnIT8r9#)G&CM|N*mFxnRKfi!-W1;eJPKQeDJ>%+0jV*!9vsPRt4CDoH=YY
zL%tORZ8Lh0a~&9qJn6*rsTZ+^0VwucHXDhxDx0g&X}?eG{+fq3HosbH3<>ochH6rT
z%gYsT!Sd^?R$To;tti9~UZ7{e&$0P(%ymomf%@nq1Og+03dc=}$4nFGTjwv4Jn_OZ
zJiOv9Zb{C`XynKkPh5vld;`NN1F-U7OMx|uNP<E`&^^?vaa|;j?9PtlMUCqxbIb$!
ztg?>x!IfVVI_w?ls9j)fW92gWP(dS)krIEzk0VG8I;8Eq>2v+fnU5H)ZC$v0qpzVD
zdMwEGiv@IyGNn-$m24S{=LTOsNW1<Juq6jzX{-C3bxOpxLfl`?$$udj5YwX1zPRO0
zN=UyK%Z{kDd%oQ&;o7PnAhdA4Bl9ObCqQci?J`eWR$3qnCtd*@?ZfKLef#0ObYGGu
zb$BE#-JRA3NL(}m7?Nksn`RRo<gF4`ttFJO1yLUxYB_clm1Qqb3Nk3x;|*M33B?HU
zD&$cR{$gt`(q(xjrz{71^$+^92j|;~OHS}WfD8R^^?nKZfB*v<|F|iFM2lUuLk{mx
zn(?lEkRx<Ki^Tr(vrgeYW>~EMnR|%w4Smc;vJP8n1qbEJO0xj(M3UW9sepP5Y#y1Z
zHD%C-`%S(NyU`aBD9|`2-UfP_<jMIi8TsnxC{i!*pl@D{BQ5&#Sqju;k7JzQ7^&^S
zpv6PKWFrk}@<mD37J48M!~<+WCh7H7NL7GVN`L=5GcqgQ$l#%5PYYx2a%pB%=j#;3
zZJcBIIZ<==DRckJ*t=E%a&Npo5pM)C;>)d^eUI9K-_22#TECiJ<qxss_l#<=uh_Bs
zJ1u8sZAIRFOg3o1E)uG09W_iK6wZooG$(l&u=z6hSjbvfw#!=bd-q4HAj<~|^{@6H
z)EB@Yvof;)R=?6;je-`sfufBH+6qYP2Lf}av(IXnfgTi-1+nIsY52;Ryk0-&A<x@v
zAk!_A_?v)b_KrB`<$}!eLW@MZ3r@jOpFp}+VVS3Hnt+K&vKx!eu^R6#5TiVEw;pvd
zx0*C>vcu38d9t79&!6^aPJXw|D%?fouCe*>Lw(uX|HNaF^%U_8G$1(~dsME>4UXzo
zPEK|DDQQ$cQz0onME$@<fb&aIS~%MzvS=0Zms<`-nOzSmvDtcv_7OhH+BC1N6iQ}Z
zOi`dSn8~{8B;-TpQp5$W6~l%o!&T4blmVgFq^UT$;3T4fy}+yHZO=3Z_M^}GwP|?a
zEuQhLQW%hJ49G2x`E=Xe=D_P2x*GW0IcXe)VH<|;Vh`B6ZBcB5GcG#&5$nnt@c83E
zUiKUwRxL$UaU^$pLXH}ugJ@X#7&Qh;Rd~}rze(j|Ngd{%&Kom~)C(X{K(Faxbuh59
zaK|1Lm!h3H6=t{!;-GfYSr~pZ69~9HS<*xHZzO$r=VE5_<b6>rr$*NlU^|9UAy<ex
z(>rUNu<nY|0oRJj*Q5)JHffgJi{>$r7Yg}i8CO2DWV?{vZw<=jSp5P`JvHjHIJ2Z&
zXv+`s=R^(Svw7!+ZD6o?p^%aMG>f;#0*B3X#XKG1_ahsGjM#-SyR_DT`RkdU>z8X4
zkYlH4(c2c_`0wnsa}=d87IK-~<gSQE$PsfL9uxA&vvzb&qQFh*c5%aSgg5gh*(03s
z)GP7gv8@4eCK_d=5(6tTK8$oj-`w058D+@#d7W!uc=#hu`B&3!;_pIBVN#<4liQf5
zUf47^#?<5VN8J9%iV<R53l)UmBNxI?642`%Kip`9mpTvR1C=>SA4YK5Wd@NbptB7X
zcmwg-N-<}dea}wx`Psva)lEk;?gf&B6XbV9>D#_)CnF>|T_U%t$KUm9-g;5;>v$(E
ze%KaB2ttE+g7gYW#((E%S-dcBnwsI@Xr<$XqGMv4ok@|)lODf>FN+{tHZ;tBbSBbc
zf7@Y-w~DVLy9*uk$$9)YnZ}(WVp+gG2P7EQmnwh6I=qO7V+*?Aqi~%SmynSkxFWO!
zkr0bf`SS%`CchPLs)}|^Hk{P;KF+DVPG4C1=sDzqE10%(la6VkO{{CtQeXQPA4S=g
z;CGcN6#)vGk^7q(0*%+m$|wA0KEOh~Oty$5PSAH<pfPi=GWyf7j<27|ii*;8F9%15
z4;WFIkDF9Rq|H|!4Teq&jiZ0g*pY&T>Fnsm&n{}jyuqAdQ{8YSxz-jU^;CQ_?!~~N
zg35&3%<*=&4mFj-$i|p8l8Xl(!p8Bk7$8yeiNRn}d1n;}-r_KPm(@W8cY6eNE3ME@
zbl*sQS<s*{6USbY3Te%UQ?ylr))$kU+K0Q?ni8vxrB9<Ks^&e!Lr#P1&PbXBf`f^8
z+>l^*Dh4wZh1#zsHmJhC-T|2aJto@Ex(yMZt8=U;p^y6URmJ%Y$M7>!tRfL!czY}(
z)&xm)bX)lw@Y~JOoKDCy0=lux_IZ_?4uT`S`R9Gy33iF)QnMVR@;6eNeSdmlYH(TD
z*sLKU4=yZM<)GW=@-%4#isjaps>Emo<Y#P3Gr#8udsP(s%G~Buu83GOB=?GeE1I*9
zb_cjGc(tZ&1#U&&W?x3Zfv>ZVx&6u1Tx<{JiKxB5pk%Ad?KF*dKIh(JD&!Es_Pf^D
zMR3x1_qdawwV=t)2}73=X}t)Coi{*^bA})+_I{TpsokI`Q4dS7G;b{rZNV|ZdC(wy
zKb4+XN?_4=F!%zr+c~k_G#&dIV>>HbTX5UyDfnu!JO-odv`nS20yvUaazcFnNG^Aa
z1<=L{a=Hx^%KL*?2Rty|TTN;C1^3{@Y2fovzuaida$B!sEpK@JyOh8m#c}M)bO^JZ
zVCKs5@@z3eG*B`-3?C2n`lY+Hnb&8xVm>AHqXy1CSAL&V+CI8Lt2K%p6VjWZ`<LGI
zU8LIR-mJ&W#kw3Zvi0i}(T{9b-7KicXL-1jnkHC8^k<I)dDBvJVZb~A)^L|UG*Fm7
z!)*$O-u{_!P4w(Q^juUOGx-AB5nUgm(IdQ%him8lS6y1V3UubtsWDlyroy<#b*#cg
z@&_G=>kD-8;p92bMr&V4#Qo0pVOz5w(0DRl*4q->vSXZnBRfO#F`GyO*qaU1Xlu7&
z`JNb(nx<=o&8^o$2Sjim*C<|3uhOM+JmN@eXD@Cj<!nUL%j=|z$CsT$QN1Q#gRZE%
z-n$p&z@@>VpJICwh4H_Tp#z_~;&<za&3rs{{z4pdP08>7ge<UU;Z?|fbR!E4_eOpi
zWz*p<kf*kkn?0**YLP4t_y;mW)Dt~5$;OS=Y<01N5M@JuAc#3!3)St)NQ<bECpQIH
zHga${7*C3Pf-Bi`8#%YV0nQhC7?*?ww09;VlO3Dk=s#Y9PSrA{ZXK&XMp8KG)fFgJ
z$d)1;<PlBn*BCNn76|{<AIHi>A8VYS{3KujnhVzgDat1%W$<Q5DA}}XPYCN)nCbPM
z-cRG}7#shj?C$>5$<=o85gF7+2Z?xR6Q>?4E`OagO;gl?^dVd!>uNgXp*1y1E2Piv
z@MI3+EtF}l_PzUZ(Pw&f))e4We%}+rMH%t7w?FB8r~~nElx}exrA2mk!t=~%x<Def
zZ*$LDyiJ)~qu$Zg$gusC><wqtwE9Ll@JG?$&o*nZZkiu8hd3l0U3DS)4-9g$f8v!&
zl(G&ujNED)1tX(FfLFiSIai|7QrS;n^^pe)cEea%I^#M{sPrdNA0+svxL13MC(?Rx
zGlTw@cIbzilo@Pw8ru_~f5M4hu$kd;{5+*pT1RDzr@AnvZ6z|05Rp=>Z42@i166nm
zD=cfA-AZPu%tB@jbp$h8xaX4J$HtOyL4Hb}WyJ^*H9lQOB<mjM;DNkPi-$jKfeVS2
z%P8GoF320}0$#OsHiz_xezN=`%`<Hy!%cCKXVft7u0gMUE<zA&*4R8;o{wXa%nbs?
zK_DfUzta7gpjUOmwZX;VT2+IC;-%65xvU7mwnt!qkE-%d&?T`F+VOkS<cNenHt&1X
zSLthh*b4ThRZ;`eWEOuue|mp|s&(gv|Gv7;g%GNcvQj4sI#}%g$RPhz;7`>^UI!T2
z3Io4GBXb52BbDxpI{g5dp{%jKqt|=vxR*5f&dc6$?zpX~7ZpL)o5(}r@;zYCPjkJB
zQvo}Bmr?d3j~Uya;K-f`S^gj|^BH~)>!ENM&2g`RlhQt@127yL8z=v)QgL52(Z=IR
zef_oYOq9}u{3HRrvY~GD%D>|{!-;3khO4>AtK0=M?<X*=er04V0Cf<mir5b*V{xmh
zS<|g6x{$Djo$yqamUywZew=*&GN@$dlkDx_$I`9ra6k`I@eVbHS7R_7_I%fot((gD
ziNoR{Z=7Hb*;ab9fp0;2W@uOFEx>{|q@3&yG4>OIW1XCQ;dq;cjdp@dl77XQExM)E
zhcc=N<CfzLuYWsuc)90@>5IlFQw00*?Ult$Wqb^yFnB$d+cY+qsbTO`!&v|8Xwl{E
zLW8q=`hbw(pb0G;D1Z%U7Yk1?kN~|iL|k6pi}2$(e*D73*B$6uJkb)<Q7z?|gl37%
zr)KKmf=A}TBYR!bAyn}pp<4czm(+If$}m{?MuJimU5buW)rE3YvFh$XYS#4(xGjX{
zfYAl-fQ`n@iRsWq;x|iGwg;&7?*~<d4dd$6sdEc{tkRmmoonqI{jc)BJ~cKtMI_sv
z8d65zYvoY%GXJ&VDt4~7)znYdo{{n-$cC%W#!jc?M8DheKb6oi&Qgz8)6knZvq~QN
zF}vFH({jOH|9u06<h2J7F6M5K7|ism8SOJ(Mb)=fmhnxtWya#6fj_95Y?I~vXDNqZ
z(^%F>Nq+s?jb8==_^JK5_Ro1&TeroF*IEbB?)%(@`#gL(%4yJTscwoMB8-4LlH5RF
zJ;t+Gz|>t=znjp|VZq|VGke5PsmZol@5M6~liXVQO7-vDF!;c*^p~ZlBck_e-V`Mr
zP&)s^h4SaYZSl@PzkS6qCui5PA#|weFz2-v&5JT$!vvfloN*9z)7AnOf|^=O43Tuh
zmXkx-$K<(!TLV$2f{n`44&!@KZUPj(*2@{sH{m+OhLH6LYACzVPALBK-_S?bbsmvy
zh3qd#rYMLagP_PDUu2M}dp5S4Dlnu*=UF*o`vZ8@ecc2o?>x@e9PGobUsY{`0MwO;
z+mSRmj}^Xyg&pB|0J`SA&`Tm4Ko^P%LPDyd#J{G)1PcqMQoZ_|<=hgS5gz(Uuc08;
zdc)IsBvRl5C<r-26NkrY%Z{Np6y2{4h)UmvH9y4*c;51xXLc>rjAhe*maaSay;ru=
zexEr5Np)ICouSif86<Kncz!r-lgGX+%tuj&YDIKx?%y>QHm9Xqnx<B12%221joC`0
zVwzsJ)&!_h3k2kjtMNj6+T$$o$n)oRmA#Ox(Cr#5LI*>Uk~(PsSY#X3lnAdpH3Dtu
zn==JA-z^;7XSP9v7ZzBXjTCM=8}|X1`1;jckTU}BKw>xGq2@`ge&zP&U46d%%Yi;6
zb~({jq<DQw5>#r+MGg6HL=I+H<v^6iY(Eg9O}+w08<3Y+E^~*8tdcM04xC071#cba
z1Y-X%4#TjW7d!KEXm%)#CTl>4q*8qbgC>K4(D9aRu6bNe;RNiuFfP16`@An}zoUPd
z>g}vRY9}?Zy%@O~=F%h!4JR$Xpu!5}m&J&AJL(GCu~q>^d#0k;!co3d^b1<4C<nH>
zN?L;%W%V?fZn;KN!h1wJbA{tL&xP|7*wXi_r>RjNTb1^fS)mHA*GY%D?s;-&o@G7F
zK@`rAT^a2A6=O!7ALKaWceqCn(vQ}T*G)oqa0!tWz?OVrL48=PCIuU+bg^KVNaJ`o
z8SAh$0-yt@TQqjoTNWe-jCRa`%DDX)2ot+KlX<WY5OEu^HD&}~b<%GDu(fSFhAI<o
zvtNOdBm}G=$#O6p`o8A7s#UhU#E~uP;pgv8R-x|LXus?QHGq)J+&``gwMHCRW?qE0
zBB`{~06wGPA*VcyOnp|k4+KP^=SMw-9q5NDXOCXxu|g8nmhgxU=pF<%ISge56!`0S
z9^@{^rZ`pdY5`0OHn5G@xn6baQgu{=>tfiY83F0Xrf=`ZNq}U4nv~bNJ@31;(Q9OS
z?gmbDH$&YuEP1$h^J+`B0zYY<KLiVoUS<M_!NsCHD8Am83^SeBY?E#K5+GIG>j1Cs
zTRDGy+0LG81*+}?V<rF=ETB~jdKzZ^`!Vc>1ZesQZ7vKq0ve*&Tl}7Nt2?3U@8bhP
z_1}?xS`lcdl<ql7ho7Aw+^MN6vFOZBCu&4!feD!R8InZVn9Iw2y*(WT^|$EeB!zb5
zvrPZ>#K6-JlOj}6u8Ofj_;;K*sJQ^XSFFtw{^cz$NY#8~sJ0Y{d0?ZA)G(<$_~JBl
zvNk{?C%P_1j#f0#meCBQ2@}KjTV`eUdFxyQ<gTdS#&az0Gi>PrT_hXbz#;Rkz>9m_
z=0Wy$%vKi9>#tB6){a;5NbQItWOr37xuKLV&`wg>ONekKhHE8d(y-9_2Rw}YjA*Gn
z!ML(<htAW^V+w&<r{8S&AFtN}wP;GdIsv*_Y{0N#&(52XWO6O`_e-#V{16w((4BCe
zLJ$}@8DjG8xgu~FmG*8^;_;m14|q|v7*{T`4Ac+Dg)8?lBMNjl1F`Q$l71$WQvvnl
zSmv^9sThJ5qLr{BqrBvB2ShkxLS0m#0~C+bacPkBi*I^oJiGI@vTva=um>1tt(|{~
zI}lG8^hv%ffwC%OgGM&U5w^-x*8BCZQ78o=06O9D`Q7r_1g2Q8F^{uYG|*{)Gs4hK
z!z0Nusw3(pBYcL47{<tLmY41|D0nu|L>y`H#aTH7*E2q(<d^|ER2R(|T<7m~e6QkW
zDxPK*PzNg<q)A)??TSb!lp^i2Rc-IUSztWYL63s3$uzhSn>n@cnKU?J(f;85pdpkd
z2R+~Lvwgy8L>jGKo--J$B4wStkv;1735CLkpE4?DCS+aA=~e#TjZU?NRL-h&?4$L4
zLnS#XI1sNRx<eJY9f{SYBM3I%@k$xNE?fhu?+E8e-|Mp>Cg97a^c`+hY;3r~yde^C
z``vAj;v2^38pltp$mafQmRX-;o3XM{H*p&!SYBC?ugsrdLkdR$?`%z+uV*Ue-l#ac
zf`nhFq>3Jb8H`0{pYIX|-B%;|oZPF=t$JJqB0s-R`?HvSVK&2}i78653fCRrov<r9
z3rLOqoz)Z#b3MKC|8*?nZPR!Z{o=mQOE{vb`7^kyeMgBCY(5#oti_G7%=WIRp~{kR
z4!#KtP>L1HtZ#o)l(bBx)ZhY)URF(l@Be7~un^Sus@{vBptuaIyK4e@>~qGl)cVrd
zz_%__&M&DDbDBdX9!rxgoKU~=ZnXJldW+#F9(J@gq>TUVXmW>U7><tjG9gP7)cS)A
zRJLh6f(+Yi8|YO2`i$8Ct*LTFt!nZdNNV~m6B<k=zs?EkyQ2Xazykb;Q4<W%ymyh2
zd^G38EY)3QFzzcnh@OL-f=U=U4r8;!s&Q2c7}nRC&KVNU*K?Dfy@(W2kx3JV9tk9j
zu*dj&_N=?`8;-DQgNy`{Nq#C4F^r}SbiGp7Urv3(Wit4LlQ9caWmK~r3LX80nNdNr
z%xjY}ary`e_i$TUFz^Urn*wvVdw3P3_?;PvCPC3ihXkNyd0i0h%t@3ow#I9tkC*_z
z)J;-Q1}4o%vJ{Ie-iBlC8D^NODb0uue*!rn+i_%RXD0%OfkTcX1t9P9ELiO=QV0y8
zSXZ~Uw#7-C8GER}<`br|2n+P2fp%21p&zDsZLPK+-ts(`C{hYf!&H3646)d8o;HYL
zy`XkWAS3JCJ_`?WQQsVv3foM_dTN%;M?Tg^z@p&K;bOn-q=FDEcE_|g0z5!Q=wg+N
zh%Q?Ijp@CMCU8&L>9`23$SrYA<OB#Bk18U@2MM>7ktyS^tHrc{uIZX=0cZrdJ)#$5
znBUh}dG07tJ%;;RM4o07`kUzC#A##h@UYl1-)gDfz4es2lU({7jGRb&i}#encqN@`
z5q2TU4>p@}v!Xno8g(IT#+x6bG=kNtGiqCGcJ_OJZB`Ho+`^WcQHEfg8rs3U$*%zW
zk2sFpubI8eOR^Jaq}+DO@uJ^10X;=e0g*%d56e?ATlqbA7{)p!0;5wyHGj`*nKX8q
zj0f!RfeoF@C%rMC*<1esKYhyC7FO+K{72>f&+Xx#KMW)KYXdasyxmz<eUF`jh>1Vy
z)a>;!hVr?8(3U-Lb5WDA(imfn`&^Yp$^Bg3mGG_Dk@YJ)P~^4XMCNLWN|LvBxC^N`
z`xa?U9mzh!Lrums5p4+Sx6{>lnm{`XBg|4tENOo2a<MiDU1<*u1@M=Ha}bG72bjO=
z!&vYLXdtXKpP<Z{X!jMLqh6aJwi(a94fYooJy>q3*C_q^0od1n+ado)H~F8|^`AfZ
zk3PL#a;9WChgJ<b8vvUt+^R>1ohiVMzPyP;)LNI@v1;@O5j3nREFteV(W!{bqYHe0
zSDwH>TV!%Bnn_~oe$}$U{%@aI=hP7vc;Qvd2Ytrp-F)${@AfkcOU?WRfH2bb#IKxm
zyURZ`i|YXBp(}*mK)P914PtvK^P!)Wlfc_`Mf0Dy@(=X@uFqrfXC*+f>aIx&q=BaR
z&9;7mxoX?IympzTQ1Jln#Q(QPPCwf6M}lIxx)|}Or_7e1C$Juqp@I#NR;nlU^<v)+
zz^)Pyd6srn0x6=QLYLF|6Jsum&minTz8MiYCIvtDvd?clgxcOv3RlhPr_7#@hJVV&
z<QwYhJ3p!NW0LW(ur!{J?kL<Ywjw$F3@4`leF^yv_>T0afuuElzKM6~XwF3b006aZ
zq2Hu8sbvdjTL_=1i@`uQ%eoJ=%nu)3&VUHd7r^fr6LT0#=-6YHjR1hAAqN0Q8w$wc
zD@vz6a-RZS48t%(tPudtM+c(iqeL#=JVX0NLH>;QyXZ^$W9DZ{MgPx82Yd#Hor08*
z=Qx5D#xCgA`gPswvEiLS;Cs`xB!1w3m*_Cj=9fUw8c`eIdjwU?N?!xPU%JaYmcuy}
z?brhDzRG1IcNc^^tH7s9@eZJ%V>fcVOa8~7|DRCi&KUlH(U<WlWCo#Fe})}E<VL)1
z-WPr#{KpkY%!kwj*zX*GPUxxvKeT#Nl2A+Hq0!%oEa^Qksq&!$fb}AxQ}q9Tz<G%r
zUI6)iAq7;Je_S@CkYfN|8bl}C_QX~UfsgnDFUwn66mGfnb291wc6I&xNpdiOCrx^*
z@+4P!dWfQehy`42u21`+LcxJ3STDGS_)JiX)Md%Gk)Cj($H2uO1o%a<&r7%yAK}7x
zEDeAMlI`O<k}kSD;K;<vEDlU(@N+*0eh;(Gk&22g({<J=BlcS)e7nZtSEPM)CENe2
z9I<Bs_g#Cz`nbM-hncioD=h=ZHneQtM;Ni7yMt=l0uad84zO~zw%C4lyj&UJg4^tR
zQLrEm;-)%}%IbLmpMb$@Cv9o+JK&%a4DdVo<vDxlu!^=;Lc!Po^~&=4$oKCad>RPd
zYJZ|G9^dd+yg?if1a5#P*8r=HKp3zBRrEqKf&y`b>*{5G-*5ZhM-eu<l-A@+mKUG>
z^Aq?#zsEUfK)S*IV(&eJs!F?VQAMICQ3(<yD;WeNBZ7c{h=H7uq+}$=jYLT*B1yo2
zpprpy&XS|#oO90c&eh#-cb~W4^L_Vxe{R*SQ$=;r;NGmg*7M9~&N;>$<7InSAv_M`
ztKA<R+>6rJ;6(SkjVGA+Ed>-1%P)qBpEv<hN>K$#9Wf|Tw<fjrR?@X!ILr3^s>5wr
z-dAX5Vqj@Mv@r(18?^4!g*wx^@meyv7${l~Sg)4u`bDT{d>h}i*Rw+>2#t4H_@)oE
z!f_CK!$Atpn~SMAF+qt<VbovSOv&@pemVRIe#VOT--S{TyFl!$(%y7A|Lgke(H;8q
z{nw_#mHeROEc(Klk<%)~QSzoDgV9as+buM(xU@k8zOmlFx>VT*SwDF?n9^j<C&ltd
z4W4@Tfymb-5bYAgIXV%4ljk|<kp>?CEDYpaK8Sr*H0Ir1`?PFonK<y@9#*ty7EG{_
zV$ZU={P8hZf+}FTj1)fnHJqbHwC_$*w9;ds2!w~|jH7#8V~RTgF_(Mgx8owq;E&8s
z%U&b;+qV6y5%RYm&SAp7u&>ru`NIwzlMRJP5s!nEx})qHh9#K)9tZtoT+fru1iMXM
z@R#J8PYj<~F*F?T9z`soReF%2TdqoUVjoyXY+%B7L;vNqq_55%LzRMw^G%%P&x5>(
z6>q%O-9>Z~E4B@FeXPp%nGj9Sall>IZ&aAPkz-olsz$BmqV-H8E3$C~e4KdJWMpSU
z+z5N2Vf9$>9jDeIT+61CN41q7x$&CPPNm<oHFr=(Eek7niV;c$E>4?M@2%VyH6soy
zcQrnmiao6zTyoGFwB;<o;!cdNv2ad!TMhcBLypQPH^T}F@ZTOgDJ6r&$NeyUouLrR
zNf)g}*iQDiZUsMdKy>K)J!W<|z%7356kN5aw1UCW<=6Nt^lc>jQPx#Blh+Cp-l9~R
z94vj<Eg7iZa<5=`7Sw(57{|vu@}p>TPiueLAG*HoR$t_|r2rc5C-uC7>ovs1wY4K`
zdtVaF8lQT0s}`hqoF4DGc1OR(>3bcI2)uufN8M6ylSye~*{0l$KI2+Y9e&toCr?4J
z$hslv9*H*<TNF}}qqsjCZc#&=;0{LYVfGS2p-a06(QmNB+-2);uB%19XWZ>Z_74`~
z${!YX<|ooLyDFFNZr@WZ?_~T(Y~r$ptU|zgHSWRhajP8z9of0Q50ylJ)fzxinGqxf
zaW8!6J>rhealdicjN)^}>Mm%H762`=8jdRhP-ONjxNn>kZG2vhe*mN$7hgt=fc-Kl
znT}2E<?*!qk&P|dgE6;OZP?z`(Ddx#=O%H!ePaWrw*5fo{D#O*pW!$n%}scouGp7U
z=cYMmMd=4JzQcygu;+LQt4U-C_COi_dSjT!7QJJ`L7V3Z?~r}3Hmal)oJcut_QTu0
z$E^w{BcP;mnfn-zSHpy@atEf6C2u?F*fubZzm}o4AsmM=L$L#<kD1`Y{TQb>iqaXh
zNvbXi3=y1w(y3pTVf}$rr6aVixL|ergG~1Iqcwl=RA=#*GxRgbbF*-@m+pjInWkf$
z@W!Fb15MZD^!7J~VDu`&>5u*1V<#s;khhVkXA4?FAzZMg{Xr9tYbm!^$3uzGK7uL~
zWQx}yB>N${8fm*}z;@6F#mcWm&7he~1moVBG2u6r$7K2?FE}A~2egs4(*0WIS+!N|
zA8%LF1+hKY+-K3qOCcJbz+9;Zw{n^jMD?5c1Z)wC=hj?D4|SkGh(2d|k1%z82~kXa
zta?o&FYwhYYEc(Xcb#6Y*Lm%dxPs*&`~yFmszE<lwO87Txm(q`2oAbx(4L=a6W+!O
z{{#o)=ibQfhYAD___IHz&VG(CdF$#l^;+h@PS#lso@;4(2@p|+E^gy;mJsN`la+9n
zjtkR#7EiTt`P}v<BBg1Mwq5n}HGOsQY8eGsn2S5@e+sstS`c1Xfrs6nP?~z5tFCte
zs$k(OU7wWxirRDhXeJ5ndAo}n2WjUbFpjlGdYx9yAEoXh0eIvXvMh4XHY<kJEoX9Z
zeJzwPJCefw`&avU#DM2xlHDIC*oQD|(7dGr(e``R>zC*#*l3__VOywr+Mwh?1n!)X
zL+<8MWBx{&aQ~Qxpv2>Nb?dp$+5~iH9ec~G{qgZ-XxyKta|NMRMR$j~YN_m6_*B)P
zp3q*}1l1i@5$Kzpt*i&}`2~Y~@B$t3wc;c0n!uA110mjr63qe9H#)t3VvYBLs67!i
z#6bxcg0W+GdM&vxx54k32NAQ5N0SN-s7FL}+XN<3!9ruSRt-e{pG%P0Hgne(Y}Z6R
zkz$90+aLFcEK!EikL;q&7^lZNZrd(oG^=I?6!TBeXjyMc)zpdJcB|$?g!c2vWRDOJ
zUy1iVW7Ds0<E;mJIUVn*-=N`+|CE5J!rxr0I2VUyBh92Hw`8#~0j3Dg&-NlV)Vr5%
zk)nO%>rty_Rxd~%V)YRnqVS0|!m4@7RUgv_#<xpo^yu>aUM#q;RoH{<mB$1OY686W
zh*%xPfys3gEwR^-6HK<%D!jVI6B`1Qptz8nh|vI@YlnzUYEYKmz5T?a2d!XP<-r-7
zxO+HyT84Wjrx(^MAxb~%t!wLM>C_u}AxHGAfI4j8!lnl~3-sxoBYkbu)sC}$*`9TJ
z9Fm#bps}y6&aGKqSd-j_bvGX}VO<mNzu&Xj(a78QWHwX38VF6zsv2frynb#&DsQXW
zGepcJNoc^+Ag4nh|BzGqHO=ycWgN--oz@^5>HqMaVC^7p{6#0O{y?LFa<G)LEWDJp
zR!3o_#pulroRa=f<JH$5f4QJY<P%gFP`THbju&QZggiDIkzP=rx8h?D*3Z;~GxOet
zWBi>7eH7X1TrhFwnbg-_(iS8&nPfH~tdgp6)GJFVbmu>mNl}lK|2(+K&Y&w?co(&u
z@$TW%bT(pQGDfEFvI<YS>8XkFn3e=SY8kv365Uy^9$IGrX$*G*)>dMQ_Iu3(>S2b>
zXpe2dIgpv@j2wxY9;ybNN^~c3yP8d((fA@ex4d#7j=}d&d_ngcNHon~_f6eK94;96
z&nnw1xV*r5M3f^E6*X7ab#J8EK9NR}T9G9xs<wLTnrPYSj6iuMr-g0%Ei6Xk1}=5?
z?O(*i!YZMtoIs8gr)aNh<UCJg4M;kOa@wORIH?q5BbI<W9jKmt7f<|Xt)_;3e)#cd
z(9GxSRcI7rc*Mkee3ya(S&p6Ty5^nv^;N<)6~<=G6@UIRbnO>cFrA+-fC^Ca<v`6B
zkFvlk#KgfW!*;}X7@2ky6z)e+fVzY5F5N}*8#dp137O&*74%f?{nQiX72G1De1mFy
zJRcthf%EGX;$bHKz{seZ9NgSmVbAW1N55j@w)xnr$~3i{l|U2sK)13=YmEuS7!~$3
zBJL5#GtOX&sO$S){=QRtaoI^M#{G_B^Q=*<t7M#M`U}I2qb{jmI!r8wOhg&q+nOc6
zB`6&lPR#mb$6sVx*>;I43KlWB4TpM%!;M&uzZFs_%d1=<NqFGupgR+}JBE~wj=`Ty
zg#r%i#ZPRiT)(oLn%xxAY@b-Y3O24<)w59*ZKHAKR~1%VZ&ee{OZ*ik|5zZXbR=O^
z9rZBM`E4QT(Gs#;SCFXi$7Wa3I*XO_X@)LX<kldZL9tPTzoz;O!_tD9wKxyX6sA4N
zaMfZJCR~i>%c>a~VFI7Ur!6c5Gilk4*Ks5XbG%hIZ#r0CK%>kddPmFfj9&^xQxypX
z&$qjK4_L)Cmscl5?PN3;U)8YDGt7kchw(0|@P!%5bhveglr+S(y!8L%6}VxhvSpi*
zx80-34Bo)nD`naI?-<u^>&09)ue%Y0`;4=E-SNm=ra#n1U{fwy6VwdjatkZdov+XH
z?RGSPL5ZflLPs{$;+334<;#vk6}<Pfy<Xi=8Y7*YYdd)PBgUwxhApIEuSIKf=6ooU
zt}+{6<1|iN$f0VJ$!TLj6R&XwEjQiA2NRf$=){QuH96bhsSqtZS`?mRto+M2F4n5d
zIk-o>30kLlpGAq1ek08OQPoJJt39dq&MUiM+%bH3l*Uo@JyL~arbE?=U#VByF)mCX
z%6`|SLv<O^Y%a-YMia}~($3YT_zyJ`RJXPrtgNAboKjp8*ydnmgLp64_J*c0kb_Q#
z;1Qx=>MF$X*y6Fr&E+3mLwtq@v_nQZ(8XC1BIy;`M(e%GF_bd#=0~bmY<;vzydWj0
zOs}Zq>y#8cxlyUjmEwJ(zNnUiIp<ur%)r4Mdl4qHl9Dm<#R?Y8YAjyaw+;<6*4uI}
za91rea@@5b`jCE@@M+(ae{?%kpyP9~szr@7=lWQ@9rzTTM;zSyOh1R%ey%$vOc%4U
z=uY?2?1!2_xNqupvC`8Wp$=)}AS71Nah@LPx>o#FF2}pq&{Me4Aq@A%-nsm|g8sV4
z0xSB981dv<ev^#?emsqNI1DcWUkz{DvpfFOYLhZtAtF&70&jK|1#eFWNq1>Ug*kGC
zf`#q8z{x6*J|VT<ZqBiKC+V#|x24p`H@GP&93N(!pgc4`VSFWjJhxMB;B&Ff(Avq8
ziFnErtlmM9!EQrDOxDPCgfz!`-m<WeB)u2MNQ_2$u%qP0v)I42K_pzUFS*O4L|h-b
zz^h{o3akEbF7uc}rz;na-Qp1RBMBiUX6g0RRf9toakH?bnGGx5KR4eB()Ms1GMNT{
zDd;)qiO}AE6y>#scQvpuq~%Mc7q54?2#N*`-!6o_2Ye5Q?7ApuKz2uugW`pHs_pt+
z7v0yZ$_xf894`v<(kL$POHKK&wWd1#5+Tv2^nNo^qmWA%fF_PP-*Wskt=h$0fV4t?
zz%2RVj7K*Hr>Mis5tqPM4zJrzDy%?nPjk}jA4zp$IUi*SRkt3=iY=$0?bJ_ez)vq=
zqeh{ST10Ct38sYPV;sc$(}?zRKby~Lm*-m`9y&q9EoeIO;aWy-Eo$OLMnQ|*b$IGQ
zdDP>2y_Huf=pJ$Rb!%&QuL%O~L;CaT9xv<ncuPBO%H}v{F4FPJ)Hj%RRuoF}>;2HZ
z?Kim;kG`-2ZRl9_G2GercK4>+)yfEs-i*4ScIt;R3!~cEPWC^pAMq@lK*?JA<5|LA
zj|PTnp{;@8V}gF*w?#3p3wNu&k}d9V;x$s{orJbXl8p>+daL{6*3>eiFpNq#HyZd4
zI%*mmC-f53b2DDPbX>p}+dH!=9^pP;aGk=g%nh1#@x~5UpG=BO(227NJ*mW9ZQGjN
zOmE4eG<kPe4Xr~`4{(b*Fo<6I65xMgl2e%);s52G#&t-#D9gOwI4xjAbwny*k;rWH
zDsb(VjRWp{zWm@<MoHUXbD@yRe>lJy)*h1oD?p!>{P30QV?c!!wC(WJ!mH69f3DQ(
z>3;v*^hW5GjoO9kB2qg1)L=d$f^c&nmGj}VUk`pJJ+=!;hYw)HqVGKp<~feOMHied
ztC@X?@Gg|WQ}au?3=}ezk~DKG6_IYdvkzJ4Ii6C{X{^BudqcB_&!4k<xZHMAvi!2D
zR@OXABg>$gvebR!m9j;h1RNQ=kY%*tRE+X_Uo$!bO&)Y>c@^e13XHmSEq2Z(c#bf$
z5n|Cl<=xG@9ArOlH@NP0<1pA(e9R)<u@t~0&f_>aH7RGLn{*SBUe@AI68MWO0jCQX
zt-@RFx2rX+N*_n3vO}dZ5ZWVOyz`fbqC^V%94EE&)N9nq76h1OcN@Dv3NYSzS8<3D
zv<yXr{pzz8;P_bDxPg_zMWi|Ri;?hiM}omw8alBWd$5bxIiXaE22cApYd`733Vykh
z#4JI1!Qn6&$4iyjD)O|OMG1c}Fvw8tXGbh;>rjUuz`zx4wb{Hq!<=A5S0PO5jNHy@
zZ(NSKp0!9dgTU_|Ts`aHhF81RAwyH>YrpZ5jFSR!&C7otww4v035_O<&c3hhE8z2C
zpW?cH1)DqTARjtEb&40=d6KHo#Olt)!&|n;I{Nt0{krw<g$OO0HsULsUiho=cVFQO
z31)~9_o3K`5FTpa$-3Xv`KWlGhr9wE{l!|f-$mAGTWMdLG?0|a94}J9<Pv-nCo5-i
z0T)Z*2~@fKC^_Fh`W%fr#<jPX3Al3Ktejjz_Hurem)s34N&2WkU{}P1ehSVGVO3=+
z%Op{?`osZd?WCo!iO=FU(W^1s3W&O>@QN<6CT^`S%di?z?aUa>uCH}_XEvS<`xix{
zau2C0{h=$*8(Y(8#biV1XUx7WjT%L~m$n_#4)3}neo2MfWne><N@Nv{<T^jzd_8sl
z*i7$HGoAf6MDWL5FENbOg01{HtgN57>sN58)XwBUa$m#}>Op_}<e&-rOcr+J<^8MZ
ziJV~-Vba7~>x93+g+zOtw0zGO_XM#uBK?(hQQn13rEy2yhTauq!(2qBSnGV%GV;04
zvQ6a_Ov2;47%)GF|E3kGSvJ>UHV&PH^mQH5%%|Ou$2!=YI@jcjFMgC+PE2p!>}Tb9
zIEeD~Rl8w+Ba;Q=*?<$3m!VEce*u9)r*0U^V|8u3I?i*bXKwaW5XXYSEs`_rWqT7L
z@kx{a;e4mnGuHim><~Ew&Jb|PKMzdXu@w9Qkb(n(U-XKMVz#Z9hDY&z`)=6Gc+VhZ
zD;zN{3=AayzLPA_=jCkh;$&X&i(9^2SPhc^QO;OopPpNWP0Zj&AG-yN9LiTm(4fn3
zlpS_wdTV2B?3=P;(;vrVaM!$wDJw8!DZSAu=lm-zC}ab1eBLZ(l$T|TuF7|~!F1|E
z9&DuQ=o=cCNNZ<*ok{Pza|5`Za6~&z1U_5gG{oBV-tdGYF#pmpG=%47V|_znW@Cku
zYP23bUbd3yPcM=^N5Nvpzy2;2Dz`xOrmwD;MkA>%43l*si#A#v+3r<1%nVFt<ju3W
zxe2wx3MTRG!Xl!I4QDOT$_olN`k`721PYfS5ix^mXT)wT+hrHAW_!DlmBin&w6BUR
zu2)(b^){eXd^Akj7^2RkU`{9Wk%K4FT6Dv}TU>mUDU6u-B{S31B4p~_y&|s0wG3H-
zv($&br)~2-rC^8{sJO*6BPPnd?wVm?c%bF}m-Q@41KnV>3EeRer-{hUad7hfh_A(K
zSp&M_8YyEO_eqVbWf>dP1{|L5A{9PH5!DF5^CVK+VLo@Y4X0)2D$r#~BpHmy1^2<d
zP`!{t7pfDp0lgO&UCfnE$J}2y-Z;=S!pM0)YZK!-?Bik1TB(K0j74cJ!J5ou+%b^R
z*7x#nd*Kf#!VCKb5*Cb2=hgnmD&}O+W`vltE-wV84P+1z^))(@>O?ig)ui6pmSvp4
z#ys}FK>j8btlLgRS>NmH#Ri_kY-^D=@RsGIBG^pLtUAwea;yE-<!x)WUMI+TiJ5RX
z0RGyV|Kcv-D2a1!eP*T}gR%el8`Vxsteo`7s4w2%pV{G=P2+Q)*P2}2Q&*}XbGBtO
zkd2C_Nc?^?JGcv{@OWpCn8`eE$hK(;>W7qxkb*`wfdcIIRPsl|4=1k2ggtlBIKD?s
zpR>2WsCz0DDM$S<FM3c({3PaC`nT)7i9v=Sz-&soXA|OpAte`}7A@8ElDL6+`lYLM
ze<S9Kue#x;8?9dB44<u}!qXnc_v#Kj-=I&RsLEQ7588*b73kFA{%?1mk5SXkNWStL
z;}`B%`C-1<9MkmO#}@yYPyKv_djZrI;bzHkJxxC9=r-~5vuO_(SeZ;((2^r#&HEWk
zyD4ToQj-GmgJ<dAmF~H1^f&XKcip(<ZuCwvT-$M&kNcTNpt|+?#w~R&Qma7qS2pX{
zr<o`|RkWp^G8$j)^mP-F55S0?qvADIFi@c<wp)#j!jQ0%$+mgB?e=3J)VQnK%I<Sy
z6o!iDO8IW-Dx<Lt^D39cpH`~`s3h$6U7q}+29LvK)lHFuIkqxA!b__Ub0shBbwm!r
zm4C1C&TVZP9Qj@m=Ihrl2A(hgQr;tI`gMMpVDD<`B$Y>2ElG-fzfL|W;U1Ok&#NBC
zOP}89aDG4c&q!1h@1?ahKvDXy1$O}*4N4X5iqbJK4<%A~GS1;Dn;$4Kw2XIX^ZA*|
z=PGFHYBmsYI9sD$Xp+ZbV;f&>8fI&SoHxSIx?vMoXP85V2y|14IdIOB91A>v1Tj@5
z3Y2beKdksgS3mKtor%Uf@XpZ{&>g(Rg$gU)uZg(VU+H404@o7yaR&#2!B#Sf4wZ)D
z<h*~Oj@npE0B@~IDceePr%EQV>kSog1kQXm5mtr?I5cr4)3k>MDv4E4`Nb|ebg&Vk
z6X))Y7hHL4rW^e)Er6<F#-$vNE3+}rI2@DTh6)@Hv%T_#W~g}7<2#NKPd+g@hWo^R
zDsFz<3EjuxZNl^r^Np+SPg&?`MCr`e{c(hb?U{EO{WKJ^6|UtM2=Wj{jy{Z?!$y!{
zwWr^{ycb_r)`70ni8bVFegMtg90JZ&E8U{sb)Dk5<Uq9-igXxScVFXaDJiRf9F&%Q
z?*li_i9qJ-k)M{r@pI6}Eo1Hp%`<p9bWZ!5;F2mBtG70t=TK18T1d|`kdkh42u_B#
znLHAEMYC{L>uJBwhvGMN(mmk2yTIYr^5A3f+gzON>%H01jr{ZVEhIKk0d0#pNF#YI
z*v?(+*F;k6EQzB|5L>ovpXcfExo!0UcXMdriXV;R4<Glyu}6uxL(WQxDLkJwdTA@6
z{y81GFT&pHj;TXs+3{g?ZKetrEY7-tT5F4*!Jr=9v)Oy(b;A@!u@c2x(8F`mnC2~-
z`{MrTI7`mdGxXc$&sn&;s}^LvTq$2r%fEHko(jh(;+NYn%IFx3^(V7k|1+uuND!iJ
zl{(3MqVdNa&_L~yj4%SR`mG-Bm@R&ew3R*}dmCae+1b{`>{n11&8JFx*{}|ZO|3Xw
z0#^{o39*GMFXnZ4MM_ru!BZ}Qi?ammc~0(G1ih>R-_#gK*{(sS$V_RPO1Akzit!}+
zOyW7LSiNjU($e5)u9<A`xAZ~Gt0I~f|H-L9aUizpkt0q<zXK+EV}ue*HDgC$_(n!B
z9%aNr^Fc<$=O=*axAfsa#Gh%s(F{0y#05MG{$^?Eh>JT$FsOiIjA|0N%C8QYs~Z{!
zl9;YSYIqUMT-ArB5-xPc`ex|$OdsW|YuTa_)Gt5hCNYy>@QvbPr(GhHtE9Pbk^r~?
zM=I|#ol3S(`)7&^<u4eQw%6(>HVvIAv~JTHQb`XeXiJ^ZuPn%BUky?>IzPk1Ld({O
zDTTr=a=|sLbFQbOq^AL#(j{PDVr&x}dy3CsceG{IQ&WInX~g%grk^s86WPu6-k5;0
zeCY<m)2I=sar99bBx|h7#w@J!4U~2*3-vYG)yIkEG4gG97otjd9AzU$jXb#JpHyl+
z!pLEpX(MTiQqSA6HR|E^R_DS6P<*@Ly^}Um4>FBYZel~>EFJvQhoq8Cqezb(8arUv
z>kr8`ZTv)nG&BJM&YwtBxO6Ma6nf#^S_OdcQh|W6eM^EPx(<b9!AD&4mS4|*{dx(q
z8~wm2pPmrQp_}vOVd`MvMYrQyX%qM6?UdrmuhIuii4)P2#npqSK=b(emd2+C`=N6k
z;dLH;`8Y%w<B<35<@I;lZkL?z(ATr9@sbvCNJ%Zd{2Eqtp73Vd7E}|&&625+VNN<n
z!D!++*0XJ|)F(GL312@@VYxl@5PeC+FjtZsz_(>nofv}6t9PW|wlD~6??dtR-tlYR
z--?%N^f^~TwaiXUy;~O~`3|l+Up;?BCLU$*^4bQTnUiJ;nmEZ^DyTab_UbDS0%f^l
zGLFk#PZ-^!{Ail<MIjp=H11_?0<hE-`E1XX5JeN@)_jm#j++yu_-E?)ckTnSR6i#^
z|6S%4h(>45qD!i|@Hl>`)M7ZG73~QnphGg!lW0XIx?qIdTW_p#kSy&rL)XC`S^I)u
zFYQO?-OKV%ZG;oB88W&2oiLl^DT8RAjX2NR-|B)=;+<HD#b_484xsj!QuJ!CB^iji
zP)3Ywbw=k57;Bu(l+}00Jm`(-PO)>dQF54UbUP$t9oV>icH#~Gg(+Ky-gC?vvM_!_
zzW*~QzuzM9#SeO-W%i02>?(<DcAT9PqN%Yq7mTiBvN6&5y*zvMktpGPk@R0mHSTH;
z#>D7XCIzF1<GIlzo7i#`BEH;-$J#*sWcD-H5yZSVZ!1$G;_v)1B0G*$sl!5T-W52m
zL*Cfo-U_rE=2x42<gAe2g@y}-7R^8>3}>dbhn?OQKXbs}n)DXaZh+p3x7}j-Nood*
zyxLBpmZAM6ZkBEC=@*CPbDx4R?bK(S<z{17jYMgr<2#~~epGN;PKIwNjKJ<f=@vEH
zSz((II#}6cf!ImMUE!EWj+*3v%Vt>Dpk9>g)pnjW{aY((>yB3$=`kA{#%yU=-RAsk
zt*-@$s@e_>O+G{`r@ncF)pLeW+CVhEHayLT4y~N!t<+Y)p7Q^jC;n$leNK)r*zRz5
z1EavDEc9==g(aQD=pw=+$IyeIV$zaoc+m}{j4!D9R2Mnu0KYTlJnm1m`}JrXBx65_
zUIGP8Htg~#AfTm5)3qv{iT^^!{<iMwJb@b!a=NP~`0ozesCU@1W4%QDaSVT<ZR~^}
zR<{8&@}W81s~`W-bq_b7p;N^>%R6{_7m(ipb{s<-YWYED{SQFo#18$r4nUP7wX?Ii
z1I4@E<an2-`B5Q!Yqwrt@=^GXhC_4t+Sp2dzZb?TQndwBiwYqEs4em7lm8^86h#4v
zL)&dL!v5deazBYGLfA<ZfuX_bsP9+2K6h}Dw7S4IW0og$&pM@fir4bp?!Yrn$6F+6
z?aep36!d{|*59;{|5p?0x#(GR4^EJQMHn~LApLjA4-A}-ljNA98OGl20xb&<>CP9d
zmK|0>-MM_6dgBN18|n_N&4r=gg+?+t7pVvYfca{)^)J3J=Hs?7Y>40E0j%lJ2mkp#
z@k!|>5Cf9&w-*hs!N40SJ|dm2N!Bbp10d(Fh#O#0C^2LH5~yod{Prr&;a@;JL&HQ~
z)geqR+Rj6`kJUh{2^TG$S%ou?4UzX9M@nFM_S{D;lYO^5$&%)|z@X3{<XSl119@S-
z#Xj1WcA^S_6%T(ZiEclNf=hOJpYgYr_n$cG{V=$>+jFLWYs~$=9G{b)WvfQ2jd3_M
zzbX*a%c)DbK<mgZ%C_;^;seC^?p%15jjhwCK&B1QCwn6kN=cqK>Ty^!`z-lyH<%t}
zD(c+A*ZX{(!aW_RWHo7Z9RjZ4RZZ0d4F>W%!DPz5*!siJo}V8?tgU?!gQ;;K8R3R2
z@&mMGqb$5TWVhQ#;z>(aqsT>9ME11Sm%s*q+_V~{(N#{*eR+WNPsS}x31?s-x^N^;
zs-d~J9Ilp&YXN_598h??-;mko0+YeKbxn`O=EJksreG;=-6Xh9S9A!^p&D>%a#_;R
zESEl0GQOv8b8a6wK_Ke6Bt*AF&xuXYr4K0H1yDuO7^6-+cRb}HSq%uBXlibchW?ib
zmGf|cN-c{?liiO0zCxij7=X^O7R0di_rZhT@BSEi)N?mL?NUsCqtyhl=Y1WBANG=t
zfDUyw9f86s4bPK+VoiCguD`!6iU9bk`9Bc9)u29t?NXrXSvFX~k&g|cAtAWU8*FRi
z1Ek6tR{BT=18}P4ihckZVdWPLk3y`0J8XNykRqU-6V`=NE#gW~(BbFl$AuYw+FOQ}
z05*I~tv#S?^evyw8a|6fN#Lmh;+#{jGr?Nh2eSEk_9=b2D=F?Zm1u9tjiavqM$Y8P
z??9$j{D9HX=j&H+v+km;4y`*9?a2fzwU8M|#OSzqByB*Sf`HKyTzFc?5rj6y&$-_0
zZvpntFxX88Cn7t~z<)=vbC;rJUz67*s-(2Ecf~E(bq6C@7h$$$fNAvMuQZ9`sa6!6
zX)Q`>5hhz7L5R}*zAOYNeT@>CB0(m0*_uw;cW5A|L}O4|;#IKvk~Q7Bw>IF=7;YTG
z3&BbypsQ%oU~UegFJ6YT(}8djh)YBWuViVZ(Ta_*46NfLMz_moZQmQcQ@6#hq?;g;
zXB$_+&e936fA;s;w0}{uO#NWDwX}C;{j+)azx<>f6Agv~ED9c}$_N@aM`c0_%`e(G
z=9&o1f}HIwUp^l|4S_Ag8jg5hcr?xI`?-r6FIE51pVm8q<YIKs)hdGbKdj6@k!J$b
zSxM+olocz(wka>ZR_DG3VP`ZC*U+7Cw?>T`Ax9dOW>_R%&iZR-|NgjwM`3wWl&@yk
z{_nlyk&o8?5j}Hl2Rs#g0e!rV(3)3-c7cYs(>;Oa2-FyCu(Q?oPx2UJdIm11NIOuP
zOz633MX6;m+U_<`UU4-o3%_VyH1D;)l=1Hu5m2$G2-m-UcyitGf4Pr;hPi)zLv?oy
zdS-a6shN_%+!4^4lHp;FcQQ~y27JgZ;i?eK*1>ckg6%)uFH{+V0_ooG;rnMn{=fW$
z*1rP5%;G&y_P>k5l%ZmuK@j`*yk~#s_x?%Hm2f141H5=8-}X0ql>hnz{D<EH$Ozmn
z3Cp){{uyKb^|}AU%l`aTyEZ(+#q`6pq5qb={O1<^`!^Q|Ad$HEisH+Y|Jm!GlLCXM
zXwic;|G&D*KQH$m-<<P-O?7VS&b8?O+3Sa;VePCGjW(wJV_p6EjQ)#nD$ZPmJ$9y0
z`EJtx==J|Uc>G_^ABmkFZN6e?IMYz{KU{^hXww8ydfqryA8x&93|#WedV77XcqZl<
z8lCumill$Vu7AbWfBs~S2`&7dkXauPuH;*6zoC^Quy!z&a9%h1b)NsUzyJ4d_2-+<
zJxA+O5_fk3ZnnIlF}x=-31Tx@7ZD>N&Gc8g{C~6(<L>Ew|CV$fra?8nHLyJ%Q+?Wa
zDf)l3mY{m%MmE#l4C~_m(Ix!(kK^QFGqjcp-2I<jfCMQq#Ihi(vj3hR|Bro)Ttj;y
z5{rghKfnE}ApbA+-amfze?R{Jefs}*@&A9j_dSFbp-1RhEk&$&i3tkQt(l}UPl%V)
z_2|S3?cL5oYX8bpKk#C^K_z%q?#<e=m1i+8QQ{*=EVgLZ5>Ahp-Jmh6-zmp@arB@k
z-({%_|F?*Z+Q9j0_2|DI2W%IC7#L&1Kk|NkJ~_b;*zKHLdH``*`6Dwl7pO&Uy3GoY
z;vhUV(??%-k=g|*w&7&pSAZ6_$H`Wg5={~`{I08L1wE_#t++-S0w>*n7Pb=NL`a9P
z8VX`!KRl{|uy5^O>P^tL$i?@@Ufd-?xK6sX5BfiF+T$QO`)~8!NJWI;4hz&NaS-dy
zC3r0~Pwc*`4F-6)T$@sn#$P%FJjAoYZfAZtTlu`S-OQby#GnLt@8giO?mP+i2Mj4s
zhvbLSzwY1Px;cNOpTJ9McK}SEX~gr?Js&f?9`{xsOjtzZr(}o7Rls=Lrw!p_-%9nG
zCu#}LRFP7zs=#LQdI!BX#SPQ@(~dj1^|pAZ=U3ka&8I|=Zg4~HYV`T(;3~qn8;8Lc
zoY^=>9Ve)>j!wAl2eZH+MGr3ceyaY~VYs{x|EsAFz3B1u;+6B*$3N@g4^RWmc{j7)
zIQFsnnIh;>964e37bdG}+4$0o<^^9jKpra@)WQL7@ZK&zJ|pRJlzDc8%T%yXR1qKM
zj@^52z)-#9ESZ9{8dGS|mT9el&F=@t@bDrlTSo`YW9?rz=fQnYj1n-j@hHPk9Ar{(
zk~x%FCi{MXwA^rkgzPbDJ&`~lI;0G&LB%0gX>;$;4R2WLt5)xZCx-6_HFsyd=-edN
zOy$k)VBS_T{^93cnH!C)y^5a6+aU!?e-2gs97PZ|Jh8p8BIowMi-Y`I_JBeueON{-
zkJNFwMrq%z91&=M?rBqdW3LA&^T-HaH={FPNq$g%RE6nFyepr3^Agn$e2yM0lf^p^
z^q41dJ3h-`b>8cPr(T6gVg8DOo)|h+)11`^Y5P!JPVvAvN+0tyt`(OefLPuco4cS6
z3aXX*+)Wy^SJ2TAJc5CZ3(!_jQv(h=(nOkM9cTdbYh*zr+WHvGoiur`rF#3<y6+B?
zBH|XAop)Ig-UOEx&@<02(!q%1b6km`RWcoDq*w!G*e6ogLlEJwkHg@ZYZO#BlfOxk
z!KQtJeLrIbxMVuzbTvuGgUDptdDhhS$)8}W<}`elBJlAW#&>Fx;qIc?^D^}4nP|k)
zdm)q=hEgBY(L9eAb*mvaqd@AVHNf5lGw0A>`$`(3J4H&2)gN9?7{2s17VqZR8PD=4
z=t2}<TZ~uT#wuEQ=jwNu-TH*>4ENC9qx>WIM7Y4KvBPx?ZfNG~)e__PRU<<U!LlVs
z=*UIBDP0G5*s5-BcR#F-;x?>ejMO5^8+I79hDY+N<%Pq4QNCPoL$e^=E8Uw25|e1Y
zYh8r^Ni$G0W>%o#?yY`_)DfbzsUbWm?GZ<}L#Am~nZ%@CAZ?G+J0cDc#C#p@Z4AME
z77MInmJ}J%ABGWy!gQ@n(|nk&EwaYX8ijIxnBIN9By-~c=n)qRn5t_Vw|jMI_(2pz
zgY52LsPG7|a9X5aA|7KA5v4GW!L<Z8O&`oJ&az&G;VM<=Oz4(-su2WoxRbMu!egq2
z<wOz>)MG7Nch7^agequof}AjW!SAh=Q+luI-s9a-@pjBopssKX3B!m$wTh<2uNNJ#
zffp>;!Xpr{u^PHH8u@suHrXBl6{py=eW%bfEA^yfp=-Z~2D1mRIX7NcWC%8vaMCXI
zfzDDD%<%LMHJm##tzj#X(E-x9rTsH6ljQlS6!G>O+3O2FWnX#Koq$vJ+Z3KliJrfG
z@v8f(tA2+xseMzHTM{WvHjCe=4drK*B;i6Uskom};piT@EAe@M@$vsH^{d#sjqEnI
z>zGP}h5GIvBfN6Yx0+<;ah`iOL$IR1jHgVUGN_86W(<01Vd6eP?_cz#sZ=~|cy;x5
z6=1GoJ<$Y)W7+(ob6Sf-p9Vp|{30ny&r6eNqjxs>Ri5A{1^t~bS9(5XN(2OOE?$3A
zq8~>{cwQv?7W`4d-+14s1N8Kwh3ok-MR3;m!|+?-_YtFs1e`KBTz*QgX9qk^NZx@p
zw#A7mBS_h@!n5Db#?d?L25{(#21<A8+ld52hjxOyvI*T44`@HuQJ_X)`hxPEwnINm
zTM78Ua_}r+<ASbzWd7r582<Zudkp>zjuU@H(Nj;1x+-$gdSBNqu5Mkk4tr@@MK2S2
zmu*?A-rW_(b{u=--Oh~G!j<|{Ze?K&ZT=t9f1DyQSKA+>iR;)BU~c$J-x4A+kA3n+
zuIkc5g>g$w>t|7<&~7B<4Ct{A+4C1asvM@Ro6K^m{=Noa#Bdt(F))qNuB}ZLL3sQe
zgpnm1P+W^Nq|%6c?1CsVezhv;pcC@Y-E#x<#+I|T@dWl<XRRwTy!wk<jT4z%kK^X(
z%f8CK1|8sityfj<`*3C)#nX~a#luNm(0h5izhL_}6t_+ltw6dQ`p()z6;eh&Zh8+^
z0ovWlUT5HYnf3cN+VU?r%G~b99K_r*TNpM=ie_^3irFE`U;3f#vt;vw;s?2n`25r4
zidTz22I9j#Hrn>Nk>egEj<!tTY*gWaG_oB`p%Q!Pzq98<RdpBBv;ZrSb?2IC`QFG%
z1)t5}&*LevELlBBS$6rK!xP$Zd!viy{vJ7&A6WL?DQdV~;%Gj3GG52=jx*2b_D+2L
zvDrKG&st-5IjCfHsk?5tc4BxkDVA-<t#-e!xVi#DQKMZ-f=T98FGOS(J8CL-oWi=y
zt{n=`&kRiY@<<&NFejpTI#h|3-?l{IbxI1!p7VTr*5jd$o+hhQURIleSm!b55!Eig
zC0x1KcBS`a8|E%bOo!eeLZwlv#6vwngrXZ~je#bsmGXw^tDLF+&u+>(T_{_!CfY;i
zc&DD94`b?q@}O#ODMVdXFLbR`^f2|y?mCtHcw$IJZXsTRIsHX$i=l_{4N-T*$30Iw
z+pOFy1qH-(->&z0+<&L{fb)ljN!3N;jF`)VFeJvY7MeTaviMm0CC{Dx6~A`f<FkEr
zFs`Y#x*>{5a^?O#gAXH<(VV<B2Z6J^NgriFn6zJEHc+JAB)5Z~&D{e8z=g{0ly929
zwp8O`S_kTK;o(-vrdhqIj0<izVK-klVODEuZ`%@knek5Y=8J9QEE}lTQcI1-+g=D#
zIlNP&Ydoyvw#98&=*?(vlNvVPD>sJ4h(S1qht{n)f8bRnxP4}u+`@u>dvi}>0T$X$
zp3z+U$K!W#^)#L$BEV`z)vi13D-)pD&K}sSeNO<sr(JztXMGn|(5l+jN9aeSCxwW&
z&$>Xa;^yJ&SOB18{~|`+^g_tutI$5j1Xy;R4GX=iTb@5<E=-|H+7l>8&rSO|?myQv
z5A$$h{aC)Ba4=<9AG*xlqsBm!HdRX8HlK0WVM$VQxE|VC@R0r^Og9rRrrg;`G5jN`
z`(Tcy&A5ixt5;d%50hlx_5X?|;P*T#`juobpk~E|Tw!wF=0G3!F~b*G^U_p00Fu^%
z%JD>2!1aK0<D;PKNd#v~2{kAE`b{zNx8uGRGQ|6Ffa6$Nn1%55SliUWlyA7$52f-=
z$)6uipb<`Hy}O4ny><@(uTty-5LqNajkOsUnX&lL)!1a_(8&w(Mg&!{d!KU8=y=aY
z8RF#={G9p)|FUPX$$DfvR}qOXbpmJ?wo<`AR7}GE%OlKX><b)*gbsl0*mNkQx$#=d
zY6dWdi^F88Sf6BRSf-l;aq{F2>28@i3f+QM<=nS~{c83@U?pi5HUs^4aWLw+-o|T*
z*uM1piK5Jkkd74YV(qn_srhYObQfEfZOcG$7l!t?a2^c^u?^6A$YCAet7S@C2g<u(
zJ`BXwAF9ITq=Z9X)2g1V;;HJGhe5|n3^bS!48o`7op0Kn?|o%?aD6JWfd*3}^yWkW
zwIuGrZO{kXIJ_%V=&3*w2k@&s#>=gbkWpB#%GA2H@)5*W1xp+hP_tbVIGQ1A>{IeQ
zcqq<v;tKal`~I0jz@(ad^#jh3KN*22E~;~IuKxu6oS)VUTM-z`D~M<Ug_W8^7bG{=
z4M*t#7s$Y?R2#Js7BvO{na!e%ntW2ZjJUrWZOJfnk&m38sa^>n?O{v*kRhEAI!k&B
z&+_~?UhB$$wNRd*!_^Xr(U9zFI|WM0Ha(&38hnqnbDWwv%{=lgC*->?YH8oMRZva)
zx){HfCRq>U8loD30AR1Zm8Ze9vC*}_A3qNagRA$G3NjRUtP^0~jzLEJvSSvT14>3{
zjk-ek-0q}__+d0I)p#YZeE$bFx|{t#eJx2J>;HP$x*=BQ`RyXT0w7KI2-Pk!YVmKl
zZ3w_<f`oR9>-p&y;_-4T*X-+HEiN6evUD1b!nE$Z86$oP#eLNWy+e>Rwobz#w|m%_
zBd4Ll_iQS+W>|Wm8A<KeM0lfct<tNDCHbxKO%paornBquwnmSC(Rd98GRMztpPx2B
zKrk#CH*5U&Z|xefQQ3Eg^TEw+`)1Ny7zpap8S&11Hg$h|Cyu<Ub7^+;WJkd9P*Jqf
z{c~Becr+>5qQnzT*85gJI@?k&>#`0SCPW;yC+HpdZ$vH+wgNdwN0&CZ!4C;C4<X11
zCbO4Abs);A@Rs98yX45bH_Rf}Gu0k&xa2tAFPGLkqOb6dhKp%me4(o2TzeQ9Fsr_H
zYc$A%Y!ZE#fPkkH>tk4lE~;$)sy`(W;2&gG@&~L`MV7NfV^nV(m!WTYtr_Z+LLBLu
zx~eli+#;jF>m8(%$HG>(o>(&Ez4p}%HXL7nd4Sq|+$!VNc&KINU5W6F$3Z3p4;fjw
zLNeQZS6+P(=FVMoYZOHN3&I?Y;)Uzz>UO>K3G2uVKb)Ajz2^2+k@Wl({k9gFb#0=b
z=ZFbY*%%3&-ReCai;FD!^!THx=(}#gU_$FKZn1V*(_E_MnuG7Jj~2-O=%2F0N~nTM
zuUj#s#;o3AD#E<7cax)UY3Y2246+ERC?4C~!##=$Hk>$O7T=e<2_UQ+-TQk)9Inw$
zl3(i*9Ys+z#5AzEi<OS9zB05%Tl6d`7y`bVyAXHn6tCw6?{Fm_=MoWZh<{>US{T**
zTpd4yu7_9WGBqJ1v!45UdP2qL*}(I7q^gO77{|sdW>FMC@)Mb%5<Nt{AwH?($L~ar
z=fr1gE(TPSkA37Yen6&tIsEtn=A8~*YGNP8AKE{hFYt7(tl5UiPQrr1R=7{Up_eJd
zZhHGGuR#qqslPR4uUvivwlqO<R}vAaXUSMT>m{=`IpT8(7xah-#%MD?DvBM<eRiiK
z<A<6nD*v_u#J8}=v1C7)Cm5KD%<|3ZqxFp1U*BTGG$Fk4`ks$@%Z1juXE!z)j*&ub
z35~L|={WIt#=|;+V9M?EaCFz<o!EJ_G|)Hv;9qf}gWz1imLax+AAO>}9%hzaT7{^w
z7h#abmMeishd7<6a!Gw+6O6%%h(fygRjF$1I==oT_e3(5;tNO0>11wx%+q9kZ$&Zd
zqXTF}U4HDA|1`X?MlI%65Tq4d!fd$_dgd(CRt7LY?~z8Y53M!-Xez>Rr%Szg9}nJN
zl|d2GhydyyAkta`WSws@E!aGNJ4^N)H>o>oBd_E`DWffJ=hkc0Z!a?8XQOV3H{!I3
z9CNYp*NgE^W9YL2FNXfW1Y5ETWkNI4%_l;mz-Fymz+~jR`K`Pld5f{|R9#u^HP7pN
z@*S&NLA~lGR2e4Y($RFQ=CgtnyvD)etli9%k$4xZ=^8&QJydPqWlUh3L$Mk~ZaVD{
ze6Y@Vim^aM!cr?yDqT2B-F+UJbLusFE5AZ<Y)f)^U4e_+o0fzd&*%&{^Y%FdU7r`&
zW`d!!ktG}HQgLBkc)OWHN{tx%9}>a>h)J-|$UWTgAW7pT{PJM-KEF+2U_qD(8%3s0
zb+NQ))7ySK%#GEPIsA$s{uj0_n+k=~&PrG#FRpo_`(*w2_;#G<wz|D%N48u%xRHQv
zqIBQLAJgb-Ch@azax)rqVhh`l(SMnl)>t;_3C{GT%4)lNN@gQ$LAPv^r}Lpq&z!~T
z>YWsK`ZwTLqeO9UG<*jo#$?ao{DG5@*&Xy(wXBq2K>CH(2TfoX0mSPiKz?9UqPaFw
zD@=UsL6Wf6`UA2t@c!MCA3L25K8VfZEh|4c6?NYYDT?O~N9^Ck4_6wCIwfc&_Mc=i
z((#RyKMU+6BON2F<HQ!}+G3qAzd>K6f8w`$+HWyVVt>}~T1|uG{Ske2sp#F6OX?5G
zt;JSP-_CD2$%`aqh-dZPi*dpao_v5zte1!F+G`vfxL`*)jfSDJ$QMpsR!`RZO4?4o
z^P0tenhSb<xkW~21al(qq=+WKi##Q1`MNL{b<H9D<gH*7ACZQIjBr-jQ-<<C=cVs5
z>k$jPUtv2JXfLZ-mXs!PH`Q>obDy94ZQs+cHB4zKt@$`FYnZNt105sfDGi1B25z}e
zpknL(3D9@MvMR@FQaksoD9DtSFjxd(-tuVo78cdpzKS5vwv!yvyc(umqF4q6g-5FW
z7U55+{q0{~AI#R3EmnVQwiotnn`FtCC#zwrhRI4aj9)!+2<8>-m2z733`oQls0}%z
zs1q)}Ql&C^_t`B$@rVih=c@Rqu}K#*8Fy@B`WSSoN+14?9qoyu8S`D|#+Am|PE8Bl
zbm+iDElgwuR@I~)F&fvha=(6PK%5{NDzL6(fX0$BXkjasCAUg_xUc`>td3(WLCRCz
zAqf|Su}hwL(>b!P@rnc%@#+Ec3Y3viPK;TPeF6(*)~gkMBt)@XZOG<hwFp#9WiXb$
z&b=7e&q7Rm!eIQ3o%^b~0q$8M;>Jf*51+qKtWaG^4xD~t?rKH%zDDY1AddY}JYNiE
zr@SX~M#kI78PB6h8XSv4#l7C<<TQod9~G<=eKHA+=wDi?XXYgYza&(ZB9rBT5uL@X
zqvMK2^SpfqA&m}Y))#Z`Ou4fWf7$9&-NBCyP*P}Dy_}{f!3dKQUdbIk8jlZ~2K3JD
z2jb+H=*nnhSCTwpU?N*PsP^&fA0FY%#qp|1MKLf@-cum!m?S$eG0x=~*IEe-6m(gd
z#cJ~j93vvOV`AEv&q}a$fvVC)^hx-I)@mkh7s`h!(Q|}mLE;KE3)6l)QSN>$DTtAd
z0a`sn7+>TKROW0)jh#-p@#a&s@*4Dtz(C%ZyC2mjw>G+e%4M}3pY{n87c3@apXw?}
zNxgabChHZ!Cm#ak2}a2zsgnI!oYz0!Ub{VWH~RWRmfMfMj$FW}yY}GP?e8)V0$9k2
z4?W#8G?Jm!a;V^%a9FrHdy>{RTh=D%s^*%us^OZJ+k3FQI3Z{RTE|VM<u6gQ6ax|i
zw>z~iaIgistu5V-o#T|wrJ4xeRnBm_V>hlZbK3bRc79?@)*p9pAYDA#v-#`#ndafA
zYBO}B>3hRM8V)e*=qQ@h+FE4cIb75)<@z|kCAB)kDIIrjqQ9h|G+|URnD?--mA1a1
zw!@S+YVjK}CKbgmi^FfKCmV9j0{aH8&Bs$+tDIs_%Dmp~e{v5lJg$9tWQmN^L^WWy
zGjZu`MdtI6aytggk?YRRzB9>Qa(v>ylhM2Pc{0&YV69}Hv)$;#K(3mbwI~{OX7`%m
zn9g=L26wcpK)|d}kG|SNgL~B&9`1%?g~5UZ@ohny1l>yUx3wvIazpe^f(l9EF;ZqU
z5ANN+jm1;X#}eenA-eL3e5${kZGG}%3xl@(rt-V|zygaq5m=k)sg?XxTJ7&Gpux;t
z|MS49T))rbxk#Ss@r-)?A4s7@IC|y~R}!jnKbUB5nzwNZYXak>(>qSH_vcWj)=6g1
zF3ftQ$<*8?^1hA9W4TOc_v|t`ZnBRh@OhI7#P?#)r|s=es8~xTRvH+n>vh&?`_xvk
z1bsd}x=d=^G-1Tu$IfrG(t6b}E5cur<ltq{(!O1^>5jn9%~+wb98)>x$S+sf`Lpk1
zH+`=_qYm<AyMxFE!q-vk89GA8*A#T*o`;-vDQ4QtOE;U?=X{*n*ng@RBqrR#_VpTM
z7qX`m&_Vau_qjNbFxxgQ93B-gHY9ZSPWdlZE@SH;>B<C;;_L1HJo8FNmfUFznIH3u
ze?^nYzXz?X2pifnMv{?bH$&{N<DayJ-XhHs3~8WT(2pP+)kc@mXp$04Bt$*pJR4Aa
zIi#fEu%(lYc7ry<Qa$Ur_$PVvds9}aXuHH{(shC8`JrS+9cj%Ai6M)Hz7<h7?OB~8
zl$A-<WF@>*FR&5N=1tXxSx!nwakJehVl2yJ`N=s;x5moBHHE(=uyOm0{4m?gk(lqq
z`->M3<BmGSXrD+Wf)}e6VV!b$qa=|tt&i=ZosoadarKj3TK!iI>o;bS6bO7DVdN7L
z7!Wxo57=O%iLcP)ZwzN~T71yyAYBg*Ulorr=Ve@z|8aUK__^*}qFxF1k)Oq3?AKWP
z!4#%}=F=X(gV6=-qZl%ip0-=A7kD+s1k48$79unYE!ys{kh<TPR6bC~IkMH;dW>!?
ziH2QS*b9C1jc6{%#y8l_TIfzYCL*=GgvZYL6*-PEtYZcGQWj(~W5`i)kG1p?-R*qL
zsC7w|{EFN_e^Q<L+hl&VPO`f!om@Y2PFd1NAHC<57mJzT?pg14Ze4t*qLWy42Iaak
z=Q>iT>2|zCVYU2N+^mQD4}`t_2Fm8zT8hRQ`psjtwaM`S=~e?hVf-6xH+l9d2I?YI
zb)xlKd@U>&gh}{<X4i{Z)(k1$y~@<7@+Pw~UWPQ;JhWYLY<8yAFzeL~Jrwh)|4+Qd
zyK3{Sm?L8bs1|MRtz&_?C!7-(M;P=!p2+JSJh&#)>5Cyw;!s@5_)_+j+*eDIqXRjb
z_gPc*biuz*4^@=oXn}T8$1W|$D$|Wg(}4w0$`WO3wSSf`*fTsLUBcE)Qtd>z_iMf_
z$XU{>o}z6yb#|uhmI7+bUyLTjfB2*ozy%ojJo)v!!hm`bmA!uNL?fVJF6eOiwm;!s
z;b1huvxvY|0#2^HTM`Ys1{IPQCvkj)ht;yJpuZZpE}>3!2MkCYACFNo^}FKWq_nO&
z3Qr0+CPWs-R$q{J$HvPt@E^k?Br|%QD!NNyQeIz}(*1K1>WPN<<buKAX6-|J%KE;j
zKzGN&{;Z6GO6?h+?Yfe+5N;ahrP^qhAc9whL1^Vk3sT&r)kk$PS4j%9#2P5-MDXvc
z66S0uRPy4gq<oho+ACcA^8GfC<3;Le6=#Wt&|6IP;T8BzsikOE_FYUfq%k-BXYo^o
z9Eja*Dcc_Y2m_dMcAxxX13xgprE-t$?gV4WR>F<CIJhIAEU=iTs`w*%Irrm8hCZoB
z%oUR!dH31=-<M}tdyK;g400?sKeEJK?+V-DO@1S+s0-w0rZZ>bW9O2|MvG%x3VO*C
zHib`TIb2+4{MIQ9RqM=3mbl#ywgNsc^|R5Cw}q)~-tDCqofiyt*NQE0BJ?92a1o81
znWUZ)=A2<BF)|}(NEfQ-VNOU-5Nn!Q3F_6;it9`WM$5#j*ktF#U9aCX6t}rNNxdat
zfAvdT+Yw*ZSXd-oWBC4;Pqi!0*%!{tzGiMr5URD~n9WMtqkP>u8Sa+6`0(^WKMz(!
z*yn^IDWN&$hgZIimfWxB!&WO(=$&w8|25pVpWxYTxI>6FEukZeyB+`IA~D(Ec*EIA
zoELT6!Cx(ac^5kS>+PyRjPr^~s)TSS`+Wf?yqdM!kF|zdl@9e9OC_~s<GDM_Zi!*(
zcRVS*l63FBZkOo2xJlQ?{iK7hzaO%-J$+^3*($&L`1|QKMWbDb6CpPmBUe6Cv6p6x
zy8^q<8(k;l5APIOXLR*gCv;hDfBy{6!oKmd<#1~lBMgh=WOhgQT@v5ibUDsWD&z96
zksk~QarHGgX1=)Z@wIo~%=6*$<CjW^ATsjBMTnztwL`3(CTA?jZ-(TkdH8LFP8Vro
zGqXX9_T%WF8KF=c8?E=N=W;^b&0gT{G?RX8ZV5_OoOyB0PTqco#HyK;vjq!(2KWBV
zIfBi~;=&Zs&0D_vABK)l#r8S}6S%oMNdlz}T0cn442#P}%>Z9JY?1buA*Pt=x!ycH
zRCldx&FX}UXz(GW+d)O68)4Ove#gxuvh;$9^H66mHm40q4b>it5%73=v0J7eDVAxa
zsuq@;W0?=reMA_)GjvjRy_{73RKVulnOE{0aFZ}viODb&r}$Z}DSd<HMr;uCKq%_7
zxKE7cHh2C`M_$I>i)V$Vw2if+LaZx2R!F@c^#k2)qh)BixX>$=^<2UaF4AvNn!0?X
z&d){P#fg!daU{JgWAV(?#OHFPH(2@93ll^kN9jF|14}?I?mN0wo}1RdX@9tBpfSgc
z=uX(B6xC{fDSvpH+ju=C3U@<3OpZc-jN^PWdu9}FeCNy^vEiptg%xc_X=~~0ig7}}
z?m@qvY>{hb{<7APl_g&<8Eyf_u#SKJPVt#Ww+F*oKaKax2BbgEtRyxMOU7!lSW+dF
zrNVT~4DtL9po78|-;bUXCKZ0YUU$nf>)UbxzT4QjnQ;3$vdrsynqQO!hHFiKUCqr4
z=hrnC$j>t`s~~LFeq<{4`ih~(is}j#>0Wek!#Df1>~L*aQ;wuyp~{<<Q5B_4H;0*t
zT1c6S?o^VsrU}GoKIM1ZCaga|^HwI|IJiN2*->;ftx~#lEl|boqwQiJ#39zI7+q6}
zp^b0*Q`RB^?eA<?51W|tfKMD7WICz-VTp$WJ2>yH38tg7mP5eBF5`oVpu%T8EGeC`
zb`%d|rzOSh9H$CJ<8;5pBJ&t6%*8G2C%JeZ6tIV0l%lz5y*IFwro&V0wffXDRc6~e
z8PO(J+R4wmO0&3rh7C}p<GDp`>-&KY-OHA7dF6pc>iMNlKeeR#h}VQ7Qv~s+X1!QA
ziXpsMg~sCbMa9Kmlbg-BEU)li5aRznxWkK?oUk6LT+9^hd{vh%7lE4VT(t3f01QXr
zNP7v+Aq~5DdaWOqHv5`C^0YSk`;us|%*Z#*v_-Etj;l3m^Wr^ejTJPja|$-hA9%`Y
zmoh?SleA}M=w!JeeJYE!xsMFJZ5?y?y)=nWIGSQ!ADSD%Fo)9jM*bM_<n$#O;)D}#
ze=7H_j&*ZsHxbQsttPb`woyf`95%do6YY{;1GNWd+q6<~pW|B3lH=~kvt>yIk2^y~
zTRE{GZ9AbJuwi}HkGqxT@z{O{A0gWOF6j~dvB>|!*?U04*|qK8i4a5?B2glGLPRG@
zMD&sb(IQ6l7KBlPAnGVlMv3So2!aqTVMgyIdM{ygM(=gBe0%OY&;2~>{lEWbec!j1
zm4%gLxaPX{KKFSZ$M1+#O%~}oU0`>LGU~kap1Os=Zlap!{vhWp^H~g8Pu(Ep;f@(j
zI{fXDdLiv4>PxHmcDGpCKa^j@+;r<U4cqi<{rSL#?;KCz+ovsV18U=z`@Am@!ngW+
z(9OW-N^i|P@vZN@(QH~_Ch5*9O=e;_0_SEOmV#!(iFf6i=Hk$7WE<`|+{CftP(qf}
z)D!gLY<ccZ-3JD{g~`*~;%4g{QrViXt!${~=*KP&N4Lw~;NoX$_B2M-xiuK!g<%y^
zgoB7XiHi>uD@(zu(DHDWQTB4gD++4Wp*)jS{KE@2=HINZ=A3l8sK+DVi?9`TmvE6-
zthSFvh8uKzQzqr?e(}@f`|Ss>dOb^ImUanYjkx~gI`L|Cr$T>s@o=a9O{Yj+w@B~z
ziy$`_Be=V(fmIgndpXP*S>ZBnBo=;O^jLMLCkf@T-ey_6{|SjW7JHP&kX3DNi#P5`
zuSeZebY`WiuUevGvHj{_+JBX6cymvuTQ9L9q9rJ<DVRvFD^0hu3Zz$7O~YkKNiL9n
z*9|t4>7D>QiKgH)H))qc`(iP4VO=Dg)iLV%WyE=wY#thrYq)C!9ps6AO4o32sXsnO
z-;kT5VzU_G8gKDC1t4f@DiMi6>Pyj-68wuaE{N8xX*U20B$%3&^?adjqFKkYcU=v3
z($#p|U*vLAFdZ&n0(_jSUif?|`sK#X53Z~1$~60rPeE(8jly&PO7LOAJtc<}$ehfy
z4oa$LF=*5O$|)}ZNITg$zJmsa40_|G-|)#RPL(^6d<nEnk;PAk<n2Q5sPYTFr2A4A
zDL~@q#m?;tj4WjmBj(?HkNg#KXJ;=E3eW}c+xwG=JbJu=bq7)PP{$8t5kE5)2~Nc7
zc<@I-$O5I=W7Ko2jk_bzB~yWm8h82pj*=S$cyGkk^%$rSDzzl3n3GfG@CElJyx-I|
zx}!6eYEw|o-jc7+7Y?Bmoi{4sRPF%2MM{zn3UDJWm-MxyM!SqtXq~eSwSaD?JUfXj
z(Tp4QG+mmlW?8|{jJwtTDiPTNqOz<3t`m49ZQS#k=U4)Bc@?5K-LR$Sd^xM!H#UKy
zrT9))DJa!vc;QCw#1Z>$zRy&g;YL(0KkPh)<dby`+1`ymNe!p9QwJ|P*EyqU;596N
z{m4;jU_(%9h^LpAO=4Q%x*<j#x6!y0;Xt4)Ei%{MjRw7;10cs-&*?zBrh68ww^~%~
z5`{6E?r;bTMLJ6pS=EIeu{Xe_b?>lO^Y~+n<Hj<3lU^8E;z?Gu64ASa!22*vIHh`V
zc%T=Cq`m)x+HTqu<<R<%9?2;3ZdGf#y}%NgLOqET0YMvTPUdHG*vR;d3IfBoi(;ev
zHCf?7QnFW4mfc@s%0p?aY30~_m-f@OQL8TFN-x}B-g@sH;M|F-RM@6i*HC*zwNGXe
z?BNm||9i)=0WKbo@Ab`oi_|yj0%qNvvz;e3zJ{V5p(PQ$rODXyNg{iTv{^Y%VyZ55
zkEO)4`obk*3wT8r`qF;Xo+j#Rq=bEAoTW+F^&<9SK$L18i}cX3s;-lNJJIi*34gm9
zw-9dWwjyY4tcu@9?XKXtz)0Bsp|3dHt(<pJuodoW^`1XCC)|xv$WhHHwUXmqyRyb+
zLgR|C{m+l&4Nr&5?lhX<EmEmnsoHviCBJdmq))y&SBt}n%N-squ3+$4@^w#RF)c-s
zV0N26*XoB(f0o%5>dWp294X>3%$@N(nccqd$3v21TJ3&VrLkKWq%Iq#JQ6wbkV;<9
z|9zB5{7KB~bQpJXk2Tm=v>07hMGmLH;ON(qrEvqIN}E-3jPmfBRv^;Z<E>0gG5(=#
zA-}EE3u=3+b#2|MvA>nnP*!7)f3X0V@u=SUN|A<HJMbq$BJLJ0LtdMRqY0(Hv>xfv
zCf3GkSMb616b?I??)3LKP)$QTXP%C5FuWK?7&+vvB6^y;w#R9fSPjm%ej!u$X0v)|
zs>YD<<x9=iL^e7->u52DUNS<Rax_w2CK66E`U$J{2xn3s<A{{+J3%x3d9!h&OesI^
zqWsU<H8BXuM5)fWlvH$}OCHHbA|ceRYnXMfoveGjFr1ST-uC1IO@3MY=5?#+<*6lq
z&xVZEnUYf8w;i6cxpQUS`g^lU100GEx?!f5usP$YGy^lyD!Nj<ssrwl)(jhjo~&%J
zB$m{&B^MUeGR52{7dGyAz7{y%HC#xUrEO8*TRfiCD)Vqqw$oiH^qPIxA~bO{+g;Cr
zezBP8OnMxGFhtSz?cYOf^$xh$ju2I>?GZQ9(qQo(TNpYlwvWDg=0b>)4eRoptx<kb
zx%pc85+$Ks`dfmQ;PY9PvRu{H<6GAAeLO6gs$WtEm5gFS%;*Rg<+NN_nHKq{tnEGf
za=v#p@zC{LC^O!9Jm0T;i>8`4iF=|)kw%;x{gQlkl0CfKy=Gc6{|Xw#tJtmotlyD{
z_bsQ*#<JZ-*NH*Lf<)P97+?(5>WQoMjhychG!+yyy)rTmW2FVP$*+}#^!UXx4P<5J
z=cnUS@)O_24Z-fx@cAqMH^gLCOQPxmTeDDx!5;be!2oroTZv7wDx($?8%_1srhQI4
zAr837n4tc(<Bq{Zy-QvC{KI`I{4Fj5x)?rH3%i-x{!iwtD{(4jq%Ha=66klAVO<U<
zO4G4V<<dNJ*lQ^F4Dv?FLXl4eGA(oZ$pR#EP|=s~%iVH$?w3Wo<UQHGiJ3(qy)usL
zC5X<*b_paRnSid3$2B%4gGAu_Zuf)2xgFk@?WY8V{g#G4kC}JlnT15FdL0eN-XHT0
zKg{uOwv-vx%`w~f#rdiKAf%)H8OnfmXy!)J;%k~6^NmtYxA-o9eK6({SX{VFD@bFA
zZ0>>RyJ;FqENV=(&n}6ziI(vT7AzZn5%1T2<LO9`{Twmf@;!Ary+mx1K*9iGN#13&
zxAK<s!5INExX0+(n!D}LMZ0Q&S!SkA)x8&oUqda`sa;mJsa*Vo`7!=G7g1>5C*5us
z75E9FLcwE9jkYA8v*o@@^8L_*Tsz57?8<)febSZBi0+Js${TM7+<BYAg_VXA6Mt`r
z=rr7KL?$l3U3Ti-$<ad%xCD81D=)@-RS+~8uRVX$@#YZyJp1XrKEr*LH1fYY;1qC2
z6<?HB%RT<06_GNgz`Dr#620V~r(})oRf1)TNBlW;g?AeS;i<_AJ+x&-b;CS3sbO(2
z#@^cbf$fr)1nD9ZcPM2s?VSB!vp;S0J&TK3^UYHMTnrZ~bumBv^?q|bcu0MoLlP`d
z`LL|Q#<z3oa@cB&VrKYXC%6#ey~;EQtLriq73y5^<!u3KtBJ11D(@=t`Xc7ge%caM
zl^sP9Ubblu4^ef5)Oj(A&%YOkoLAAVr{PJlyC(8y{Wg`&(Q)9a_fjJ|4CSg#4Y6TQ
zPjIvJ)F2JBL7SS{ye^%3yez|!yfd<(&+hp3LD+3{zbwYS$uf*M*ROVg4v&Hwg1)Lc
zWK#Tzc<mM=HzQPOUL(vVVQOReUQ?L%vcJ?#+LN#+)2(M6%D3zu7A?-+Ks*Z8qou-J
z;&Uh(Z6FbzU>BoAIYP=x{ZE&R@61!<3Hp$TXKd&Ype|gs4pef~$o?v>;j-?H`r+2x
zargX<>jLMXlWTP}PuaBJm)6X-Q4<BmFHy%WGVHR^b&(!X&0?-!eo^vRENobse*V1X
z&Z!!~);@omdFK;k@L|i^`8!V)Pc11;tM6Tzztg+l4_CnNv4vW|hw=>%hEhhVF?V`!
zCkl5*1XW_?EN>U}`h6HFB9G{|()E>+Gu6SaiuLP<W)mVm9wUjUC*Jm`YlsMR-&|=h
zczQ$enilm$nps8#Op0VbI5%50dAM&fYp+j<O)y(ErjAdeFPo_IBmO(<4)o58;6wD8
z_3djdvcc^X*y~0w_)=o<y@_t4T`ZqpdvYz3X<<`ld|u;=#rthBYvaxar#L<AUDd1U
zf+=6xrl9y38QSdfg^IA!1QWa-e2I+Rcnm#S&<GRPb~i)W>lR<=7M{NDD46Taxo6v^
z>OsP=S$2G-1_1e7pCl>xg{!&Q9z~kKV(PTIX6@9<vCH|*y{;|a;LRG}c-_a#Bl7B^
zDzwiATcEzH+aC-;?#~0+kB0DaQo~^jQcZ1<>RAUw1B>Z`#j};?hD7L9^y#^b?%-D)
z)gj`PSjx^UOAARL+Nn#RH%Ylokn)fdmEJ45&gV@&HhiszDXZ(If`_&!DDn4?yKY{F
z<yI;eNW^^4M&}$>AJ?;wsazGbav%IUjtz^HonEwC%-|iskW;M;hOm$2ne>1Am+RlS
zih`C3&IEZY;haer3FJWE_aDI2QlUT1-Y27`TQlXxdeW()A<M46IP6Ilc#gI)tPqZk
zH_T=-LS!YJ_yKlRj~iLVNQ{V1gQIP->Wm{%y^m-vQ6K2=$YxiW=C^cZq9*;!A3#2n
za#MC~S8`MESvTNCR2l^b+7?Wiivbq5AoYzlQTH}NYC5#J6S7mV!yIvxtR&I<JP-&E
zgaV@NL~3~$11>`+GCmf~M?s1flk61lyHz|0fy%#h6irn-`l0@lK8^eBJ?pP*sf>Ri
zzu+PA%%z`8GEP~16+fnUFP;>b)=(@6uxu8nG=|z1Go@F%(p+Z2GjI5z;u&7%_t==4
zad$Dh257t>;pcH}l>LLwVE4`5dw<Xr_V~u}kG%(8&-ZtmW_b4nC)+21RyU)t{fDvC
z8C?Ts6B{5s>^$P$c}>>dZ!l-hJ*_UX_vKVIRq~;{c~wZHuwn}zRx<A1)zJNrDx6e4
za&EbUibS$<`^eSyH?^B!)T*7H+_mN=$!LoSyg5O9R-ys)f=1AEm{YlA^h#K2Xxxk@
zBkiGv{rxx?_I*VChifh+<6a-6?B8z=vb5aG6zU1J(V#(VtG^4p(4O8I+QK}t@`ju0
z=X~^IIsBaF*FliEJm7@yV0p$9QTT|Dm8_*HY%_<@J!Q@EG&tu%#}x|_^^G^|{SJUB
zYJvWN-8J!dh#)p|tM{LBf9bL+FDmK3rraRb(iLMO$<!oQ+3K}um+&NC=>rFQNE_SS
zq$i`6_$Y?A*i<(0J15D_#+p+ms?Qed&jrGp;&)_x>RCx<#klXW4l1t`kO-hFH@b73
zFCy9;k0E1J$2TYAvKZA{bJT)sDP4AMYCq>ldL=|ju31YdLepF_|LHZWeMxmS-hQ{i
zh!VE9urDQkd-Zxx*qGXjT^G}Dt(WGL3B1R!RimFo{1kks3yN}|oG~d>Chr)b)(iDS
zldTHL3x2&6(IwO}dRJ6F`mztxXXoG(8%~)_Ybg>%b6o9g&%Y?c&~!s{EKV|S<M0`(
z-_es?P{dDf0?E|zqC_psX4*jCJUid%uu(v*A%R|iVEJ3S7M1?>Afn*9FgCC&MIwC{
zxl$g!EhP!Fpwruy_{DPzK4s{!J|qI^UH_=DPXFwtK&~_CU6pBU(3|SJ>d#A-Gdu-Y
zeP9@wBKBb685z8@n`Tmmqo(-*mWnMG_DXFV8|7oS@e87x?ppB0WI1OpX6!xq-6npK
z56O>Nqn-r(zE;|_BZ<ykh?!4{<+6R{*(ZRjr&NkPOrPC%@3utuMV400mY*^{8&sU?
z*~i5Vd|!GoUeudn#9_032ix$2$5HcX<7I9-Q!&$u9%VB6%N82hhQit$w0y52*^YvI
zce3wCR#`|%usGc8={pOM8Ba3(*jSR7VO6+-$w(+PtW9h*o<xq1u{-wK^|7maPc#(4
zUJ7)}bYwi8=r;1Wh9`vyw&F`*y1_n}PX&(ijhT9je~-I@>)e?6U&ERF?=T)DynhsZ
zwTdpqn?@?}nX&PEnxk!`ns`J#(eP3K=65yf!cYh=j+|KK(kw^kC=^mZ?T@l>LvZr=
z32m`IRVm4S42zHgS{^n73G?r%3W(XpTO_IygWpr({OPhg-M*jH^lABhYiC2bW?d&M
z2x8$u?)Z)$d(t1?#HIW|uWbl(3O%xzWucL|fPHD2fWn`FThMWfZfDUUgsUfSnSM_c
z6{f)3xy*XWbXK;m?Lr0ga%NNzdqJBl1Uq*zs9AOezu~IBkIDK)a%dp)r)rfh$Ol)3
zziir6i5IB8mVDl$I%WA;_Yfox>wQ+)(I)uTA;<}jy?uOp%yG;I-791rdHwF#ey%9Z
z@?jiEC0i1r)N|xbcQ_`QJXgw8Nba5|TF&gU7av^ic}#-X)-iC=M$l1RRjFftx9pUp
zr^?BjlU3Er!S)({+w1yv{Rd(d$;4aPn+^R;5=liOu?3%gn|sO_bqg0|B7x{!sRG9y
zzC(gmc~5>kKHDU+&D`5I!m$%Upcs*@^4Eg&Gdm=nlQtVPQ-#05fttq!JMm>|`L{{)
zJZS+uq0Us|{R+(-dwcNr+1nAG)3iH9K=Rsx%`81)m-J&oYVn!GFwHH@<?b?2Ui1_F
zMphSEOghK9@ffXb%;HPHFlyT7{_K!2ZnUh{Bv65Ah1D8PrEd~a$<`lwl+{_m+Uk@l
z^sA*zG(1_$yKLk>lgu;S>RCo*>$>kOBvLUPi;eHL`Asu%El4-x<!J!Cvj4%Yll^p%
zv-PyRD7}i7N{g?J@a9(|Adb$S+?Iv#U0Jx4ZiGI~TqN8eX+aX1Yo<M1pw-vXE=}B8
zNM!Tj+hI5-S_vJu_6cPn1_H65O71a|V={ae{p3YF!>pNTk#F4m_amS*CU@*mIZP6H
zy@t$07W6(6mtq3h$2)=Hf@h~r=&DQ7JuSvyt77k`iOzZws>>PS8J+`7JvW;zE1lQe
zPCxd|T}&uiO4NU*b-9G0q%hd@E6|)yFCaBaDpgDWu7k9xU@x^?V&duCBeK1+erK&?
zA}u`)o6I#B1THlE{I2f$QI~o#Z9E6@71l%`ekc~_G=SV|bF}!}J(}|snd;ovBGBEc
z&9@-y^Xd3?%k?!g@<7SE4@uOY<1#UXm@ba-j0v`+pW$P&im$a+*+{|f?->qPf2fO;
zv{4L_X{OA0vYFO^KTPa)zCul*7xASIGKcSvk4dttSk7`AUc`RFGNGk^WXaYxy~XRf
zaYl6N7MFWvZ|YlW3;`)CaAUDsRE{CHJG;EJJPlQQ6rrVa<;ZcSo-VX<mHwbwnhz!^
z_+}8Xf9ROo8x!stV1dywj40+^XiWH=`X0j#ol1%_QKx>4FRW7iD);GDKYfp6=RO53
z-)^#pzsK#;)DnWQ>=QD7q;RuakhPFfpW#rKl7LaQw@^^MkrZ5ni?4zEK@fIIX1w#e
z+hv|eahC@^XBtGO{2Q{#BVDTsl4Q-BE~`y$)`g+i{=Y^Q;K)Hsf~Qp$?ApXeQsZDD
z`PKehTA)|hvl#99fu(z!P;{4|9o~IN;Of-DlJ9i7&a$h(z1!@t8zjH0K)BmBKDr6X
zz)&$^qHklOsf)X46uW5Xy9nJdRA@!X9k`g-B+clHWCIG5%poR5y>Xy^tbS8ZK;z86
zh1%~Tz{@xD>g(}z>^db&9Tur6n{M!Q>;dx+*h!I;@id7sW|rKkgBt_M!$5908u*L`
zuVVEQkpPLjR~1l6CF0{COyigUaqi;y(8+=9nxZB<uYnFzw*p45nVrF96KmVUK*Qa(
zyG~9!isxQ_;@z$!V)erYr%8Og$x;-h%^81@e)#E#%ZSE3w@ZfJQPcINoXCGHABvnU
zI9jLj?Gve_ZAKpl^(hYYl;6%+$~aZE+9kGdf<u?FwtJ%aT|8Ct$?JIb8P#uOs>*-f
z?#IM=tN~+P$gaT!bID&&x)J^dD|?IjJ<WuehL|VsI8}Kyg<Q4GHoS#kuTF0DgbC4#
z-iq7v-<vsaDZt9-VdV|O9)g)e-M|}fEvY+Nu9UQ9)HHf6(E6#~h;4kYoLU7&i@MPO
ziG@qnhmV(1xW$5;mJ1C=qa)MjDiN?wg#-cW*P=cG!%Nu^*C^{Si2fZ)$A*Yq&-6R9
zJ0TggdPnmW<wT_}6JQbS@tiH$Pv)zUwm3h4fqBZ0PM(<3okXU&Qu17Epd*>f#MCoU
znqP@(=}Hw}A(;~j>X_$voitZpEYmih$%xP{3KUp;Y)uF&j;(bz<#i^_;+-tQrRbow
zA7}pHUggE`(|k$PGYv+<Px)J%G@=}95FMS(85%tu-L70}iof_%77b@SjVMAr{E>bl
z!6!`wzK-H^uTr6+Mn%>NqPCx`6x4SSwZ_+A1pA3^Vg+BSkSq5;oyY{<k_<!4Cf|Vo
zHTx!~1LL83Zf#RBNES}g>v<XkEq+=QDK1AppyqJciAirj${EB^?MEKe)1YU)S?o)`
zlrY7FBY5ZBgU&4AB7x48(EII!1e1z4_ZVV0*717CigGKal3xjV%{1won-2=4gnK$R
zExArVe#i>p<rhKm>cqtki}Y_$llb#kESi>BDd`c0?efi;NYp)CM@Y5oGM_Om)F=2B
z6YZxK`;zd^Nm^A*z`n2Br`rv~3X`mal5@U?UDq$O;dKzf5R}#4(LuG`fQg+YB5b>8
zCe0+jxOK6&BsPl?o~atC=nZjI0T0Z*!=heC=%->0Ett=fL6D_mdyZ$Dx49rWcLRS6
z*1z&RI>X0OPnW6g4In_kLsaXUS&;yD)Oz#YT&!U|&xdBu3v;j*`D(w~1gbeG%^MxY
zmFLSDK56z2%2Tne;mzb@l7}~_df0(ve-WzxeeXs|TbSqoT4`CPJiJq8>s#lpw)V6;
zyc92mN0EV6=!aqBv2-om<?4=%xNfcSTF(!5GpQ7~`=t4qzbEyGNWU<xQ27Y66H6Qn
zVE3(^#<OehUAwu{qq%&6hO%Xs#4OgX-sq(uO-Qql`}zIwn%Hfw9_Tl+eJZy%Z{Yep
zaueb;SVTxm(zC+7ScR4<awjE;k*%;HHHXh%>H}=z2~s$8D^1rr6VmmJ*CX6%Q?OsN
zc^|F6IeK(R)}H<pw{8{{C{#|f@;_Kg^9OevhR-vp*2<K(K+vb2d*!l4i-fO7_7Y+y
z4W=s4Z!*@YQKADhRGm77Q=KNg?}`kQ?JtEc3RuQr>?Ur!baY)k<cGanokz$2UDFwe
zU_785FBx;b)}4N$?-jsUoqpMcXXI;fv}kz<JG8j`i!^&r+ISvMs{g04NUhtDS_R$G
zci%_GSEK72m@n49aGbp)Ko3beKlo8@Q4h`85luJykVGKPC_S+wt>40fZEn0PmC*Z=
zKLNt6S#~|~H0!9__nktPr?GTf`irQGkSEQAB0kY*wBV!jJvT82*J{{!(h>Yl56`AT
zjqbDQ&mV(G8TAp%D{mgMM^db?Gz5@9+Em1TTu>+FzQNB@H!lSzz+(?Oh3!!o6iGLk
zQqErU(dxd#^fy5S?n9MErm$ro($_{A+zwUQyuP#<98?oh;~SznrYHYMrTgH@UWm$5
zmiN*(WqmsE`KEp;N+}zs7eC?eU%fGL02xoE;)oXAxPts(Qo$QNY+cL-<2uRgIY5p2
z-#%j@rD!N?7THA+a}lUpI1WuHo=NLRr&h_{>S4lm+b_Pb&0yTyy?R-qf>>(TT5V1e
zKh0rD^u2crm5MP7KO0}jZN!Bp9nz|A_1oAw5>`91qS7`Vffne!YW?@P#p~4)iP_B^
zX#-;w`3ky5Y;q$Dg%j5+p2>^~Db44PNT}TGhIj&Q*nQzyxz#}5qFar(0GEyw5IejF
zyJQMM;n84MTqLVZ;1ppy=KV3?baI|$$9-HjR2qEQgnB;@suc8<s=8nJr}wBvP@BKr
zG(tcep*<gX{4@R8uLcVD>`!fxXOwO4TdOW~PqmQCJbR+0`-z#1J8qwfVyy4-f}G)W
z(2@wj=+$FNrbPnpWlzav#Lp4t&mkhwi*dlK+KYanE#azhdC*#ZWw6luSfF3!OcC1t
zgNV=XOMJ)Www^HYxE*@P+(qi|=qr{b!h4)5J6e(MaoKhTt9)%5YDspkN___-`F})T
z>?>@`)01i5^@%6`Cy0a8^ROPn&l_StbWSG@*!ZU{A`IaYx9eTbN5#!D*vHRf*xP+;
z2J@F&Iuc7-OPtU)XfrI91M~WrGxZuu5~sSk43{DB-cyqB-|eU3%e>w?g4Gv#{$*lX
zfy|IJ?+OS}{6P0Cf&WUqie>Kclwb}t@T~$<(+>(f_kbM>c>EZ8<?ttHNlimL(v0S@
zNbVv`qdvJTb^oVYLO1VeEV<kz5Ot^bYgx{I?pK7naLVKU<9_BSvb(R?HV&&6A8v=W
z-!%9oPTG?x@w;+i-EDF?jxivR-@>!Ltn5ht*E;%CGC}Bk^DvRy$L?724nmwYhkX@)
zpL|B3M=4CWd~?Dyxt#El$4KSmbwA@219z6vsbf(qy=dC39vK3QkD`C|b2Yt>&oZk`
zc@ggRM>+%;KKJcN_lv{wc@1f>m)a8kaY~;mKB-;0Y@0-Vq@m{4mMD3Vay9Mq(@$2p
z)i>^SiT|BC1T0--7gGbC{b}j`<#`8k(IZxuvXAw!2(U=h&EI;jPmXOm=)Ld(G5MLT
zH{_3o<}W3U93}0b0w2vb&0zx%M?rO-U#;7xa)s&6W{E_-<w9|8WF_-OQEncZXafl2
zcDKv~eZ<}S{uR_W13Nmz-JMLDr8n3OM&rxXc~1Z<(7;`nX4rzaj{Yl{dLZM<eea?%
z$B0x0YZfxQYQYsQm*M7@u-2_{g6%}af)MK22(oZ3s8+0(Qt6kY?Jf<8kdJ`iE=`&s
zP0w?N=jC4Ns1o@AF1ww_4>U;Uhw@ck@}73+uR+lh+vCl?l<_1d$K6f>S67I)%M|K6
zjjnd;FX>2R7Ebwp&zWl}X%mN6nv1f2DE}$EL34QJDz+K+!$ZBT8yH?3##i~Q_L|}T
zmti)cl#qd!5$V;r$&sctZ_i$m`+JxH>R#t50r)R%?H&Y5_EsC2^R{<dL2%$b<XJP5
zKQgg-cVP3cW%}vJJeBCgd4r|5*81iiA^*lmxcj!B%Zfahn+ZHT=J`V)q6)b%EdOVM
z{r4RE<yAZtb;06?u(6Va(>EgzgdTUtkbzxQ2A2zYM}UFv-|QlxxC=c_8xyzA(XrOt
zX_XzF!^o1tJJ3g8Yk=WN2*sXK?((UVN5-=1C>A2Yg(D8L;5p~*Ch9e`ZANTj@0B6i
z9$oUwbza}ojeQqE6f`Q0UKU>-;4zBLvVgK^NVTzUIG^o>@S6ZV$J(gjeAEt(4q;J*
zIn6brc*im->n0c~{qWJU(l^bx^Scon6L*?y0GG~(=;5xr1EM=o?<1MXC%tnrIMoOz
z#=mT$Uq5CF*ormpp3l-AO201;JbYvmb!q>@+i+;RLHEmG=meL-X*t16SVgFGS3a!S
zMXZLl8?+res}(JPpwffLGxU&f@e~k9I?x~9Vur!**liS+zJu~W1+TuYJ-D*<c%gm~
zphIpfen>q%u-h&p4DYPwx3JT#Q7w9ewp-wL;u=fzc(g*TV8i6D@J-9|d<g{b5u4C&
zZ>296@g<V6ZrP;cnlCs`zwuJsO(E1VQ@YxPN{%23tx5t+jz_Re2cCOd{KKgem`yNW
zsVG+LvbE3v7T1jlE^mkKkG-n{k_+U8nPR<N#G5N7d<~NFaho6pGQC>XS%-S1B(W*p
zja-jWb#Og~Z_H!t(qYU<-$DO7Howd+E_z6&SZFFy0j2k@ri`v$vQ!Na!L>FYf*!Hy
z$N#ux>JrM9_GSJd%Hz5@bwotdHC;5;F9f6j#q?dr2`)}TAKwV%LcSXp!Bu5#au;+k
zWyYzP0tTsr;>bika+Ll6H^3rbNMsC5jxNE<;&z%@#&Ix$+uPf%OiYW@-|1cL++B4H
z*arTSo&m2Ro}Q#aMFoIZnIW`zM7aZ}J_5s5f8A=xiLpR);zdk(GousNgBjj_1uYB-
zRzOl>=O<fZ?5&M`kQ##ZEZJso&EWM5-UNXTi8If)^rnuV2aR{xSuX`U*OB3xvW?BG
zbN$zH`vn2GWqcg?|04Fj?7lm5hQFJ^?6q;E+9L2q0zfwkxemlSuyIF%MAwUv<HsHw
z<*hg;sDsB_0BcoB<o^DXHGfw>^Sa$F1MmlV2-V?-7#%J6-dsaU@x%K37r($L!LqEX
zE6DC9Ffx6*6A5@7W0L@4lZ3#MiuMeSfO?bz-BQWiqF;5*j@}?(q&*1nz%PU3ku7i)
zi_lyd1GEog03><csr&viP4FnlEj7jwWy(C8uJgG#0smDGM_<Bk2S>x3B(0*=%c_{!
zpVOWI|Nm=P&mJbR=>Vay+)fNUmcN01Z81Mbm3kDw12S<a374JoyfCb~PEop@KO8pe
z^HMJ)g98&^f6(|Ur&iXp6h-$_0PeA0cjfY&5kCS>*^>C`HH+d;yB)M8Pj~phb&D()
z>vgDab7Hdi>BKxWaW}EtEx-TE>(HRrjT3w>6AG_1M69QnQ+Z8=(FECb9|u8~G}imz
z>{lU}-mYkCx%gF-=Tu@qg)iAseIGtmLULHKYNpSw)r0s9I^>=%<DD8v9p1Yls5^9a
zp$K=TX8SSq1i(hDz@(cvX%v>%^*;(iya5d&J4A=(Q}mDKUZ5<=mPpDS=QP@zDxm~{
zzt$xuZ3LntpCe6cNi@0w2aw=zXAM+%3oZuey~boX-VQri{@yDO9)a;^9<h;HQ-6rV
z#_0gVXrCUN(ElG8#`}0+-%iHTZ{_o-r0{=5Tv!CTqWtW-*TJgFP6Nlux2Qc5if)&k
zltPV7qk;(xT>JX36Cq5vCHb+RsPP6^*+}9y?C!9)RVt?As6YRFT%vLijXVgq_sm#y
z-vm4&IU?;h%yTG^N!*1t_*yP3C}+D+0POd4JqV&*oY`2<Ccm`(DMW_KG}X}LY=1=5
zRq5Ai(dPrUyYl?8G)LMB0l_5jK|t6vaQUe|PFYm`V8nP-4IjtVAfq(8LZn0iU1$X1
zo8C@&FY;AEd|C|70iN$!F~nI3PE{FIF9?TD{SrtuO!?v*%s)n270{x)^45W{hvjs>
z{DJ_XAG@EPjSxx_;1HVmWJ%kRo|5QS&m{CREvXK6`zb8q@T2VRw?8*<8KNNjp6dHP
z^5ipZ&Qx&S^!)Q#*^fbKgi#!`*P2{D>acII@%GoB$uKA%5;ico3fN@&gXw!6HC6$?
z?Z-2zdLD*%(}AlWcxSmu(EgnR4*7<kLuJGv#U$<u2}(H9knD6&8ZrEn9;b59bMGC$
zJ8RgJ8Hmp)OXxR=;-X6eOr~l9O&d?{)7pc+Rba;3-)lP4Bj&gR6tlO2F)^Wh9VheP
zOF&fFpgnvtmLj=|6mHT&b=?H|;S7B5y)U2(8*2clB`hl$ka-KG<4_p#7khh-O7QsK
zH`}-IL9(%1^0lB8R-WmkVGGOZdQIYwDX4-0cMplc+gSBTw@~<-R4Kz-(?7WHX`ptu
zMTQI){*4DtumY63WUJd*d;dnxNJm_-*W1amxT2zni@icNT;XcUd^J{}_SYioi^?hf
z`Ju@j`_U}_Qj}q#0|-r`d(IyUZ=T^nj;5FC*8aJR{}Qd;VD=>tXd<Y_i%b6?mb(Jw
z7GxQfu?<)X2=w6IC{wO37~UviUfNL0ONYT0%Z2R23raKKd_?h&9+>7BUMz;jLW@Vr
zs}Iavx(yf(#kmjIZaj{Q8uh(E-l2%aMJfk2mz9tqTtz3{%#=MB?XJn?#2tuk6D#8<
zyTc81s~q&r!iDb=7(#6;0NrK{@U_VER0pM2s2V6zD0A$gdu;Q&@S^vtY<rI`IucvN
z(To`1znZ7RX<XfHV4IctC6!1gn9cqx1YvL*wRw7AiVuBO2`uHa=hMyX?GX|hg_P3*
zC_$^b*USyi43W;LgxG8|pcC|{maQc%0D0_1>5nAE2$9gOwd-Ig>OS`?>N|tt`}l&H
z4K=ymcA;Ah&?Z~#j%w@vF}N{AL*Htno|c_B%}L7msxqiutdFY<2^ai&Rei|XrtIm2
zh*Qk^P}sGjFM`w2BK$_YSTF46Us7_jYc9!Rln?p~HHV=l7B}IekytZHLb8$(sG@wY
z@|ohO0rkr8`ZddFpd;7%4;?wEw<wiWt*QS9F6w{#k_eB;&EYE|Hrw39fxD0SwP%Ok
z<W)OA$KvoU3kkyhN-#$actl9~vUd22QeyV4t;SckhLtFZSt;lh9qzX~-`#kHhYkvE
z*OHb-f^{Xl95LzrZ6@p9ijx=~&Z;@neiMZWeyc>+V&);-etOpqr|||KIXM3EbvoF;
zgO&cbAMWqJON+uU+B*UypBw2FV4vdWjz|Tv8lhX&Wx35R!o0!7c48@)?BpE_AHNj^
zs(Z(%!9uTfn}fbc?|Y*r+E)=aP;<}S<S7Z>{%rX@c3&=iv@%EDXDDa0!Wu4G;bi`D
zM4g4`u-NS_n+q=^?-Y$4gK#>oZ{@RiZZRgn0G82NIQ!>P!tVkOt#(x3n?JwCe;!f(
z=jU)oxeKeFky7VwTn}-n$<TEGu$gP<r<aWut(qarKQqHFfZQ;>U`*E2Pp`^gj6D}!
zKj|2Gl`=T@${0-0MOrXAI3W~@JVybMV=dp_%f>spCejSA%<kdoR%7^yU?7>uXdTFK
z%G>*~t)}10Dslocj4f~>q-Ee$kLKpYi4+`V#=9(B=ZKT!DRJSSZ$Ldf3c`a#`}HGR
z4_;-`J)i%xTd2zbDCZUaG46kFAOGwl|ML@mZzATS3IL3LLEPZS#H&xajbp+i!H5ib
zu#5mDU8mYKAj6}8#b@Fi7l!^b@@5af8$<y(y%D+wBpdw#aWTTX{_5oaEGaMFf-Bp#
zr|^yR-|N-?;4l62WnB}8Z)pw?oMy#RHhBTIMP@#L$cq42J`%Z#d;NJ9+~V~M%$|m#
zW^}4%ug@SW&4NV!3=eTdC;lw})Wzq%cU$-W4!Zk4(RcrR#~8x_{(YxWOZ(^lUl7&*
z_4O5HrU4b8h`YlU;F=f}_CmG(U;k{z(l|ackxI}XcvYP|8{lj~Mu7sAzAeQp;=V1Y
z=u)k8tt6^Jo@LrY!19M}<C2*~KtLH1mTW8a6fpcfKuD48%HUVFo?|VwhZX?S=0J3%
zd=!s_+nCS2O<B>~z`)oUe{eV~VqHF5*TpnF4n6-^)C?ZP_%>&V=F3MuP_~L#w8!i@
zY5K69g4Cmvb?~YX@+y3OfC^K6#ok`^->z1eFkmXSB$XBX3-c{^)tD00!>NqW03-mN
zjWM{QDHnkqCG-uzBS8}V5*JH=8@3gf`uO}^is#PbaUkf$1(4dqlEaK%rg|Mf>wbf0
zLRkF(9CRt=n6a}owPl;T0K)z9!L;us9NP?0;SOG690!s+#ELuo`Sr{Zx@SG+T-zb>
zQy|VT9ywy@QQBm)_qghsBPfd^zb<(ucsqZy8sJz1waY<#?P)VD9}JASkRZe%Vt%Lb
zK#J{g80+=_`<e!XYVzQLb9_eoB=pZUac1L(joriTLbkup4maOGTWbov_w(KW0@2t)
zypbI+7zE%-28LD@BpImUQzigjJl{4IOzm-mo?jqebYM(`uGTn~(D4$<+S%oZDI6ki
zW$J@6@2^d393t@-F~>)WRDcV#4(7xeXKH#z;2CUb$d}Hx9D1EAKJ0oSYE`GE7CRv0
ziRAPBuOLBmeEl809TCeo0MXrNr@Mh<Vu)_SDo~uIg1|o!uVZIia>86-FvvQ$xE0TY
z{7v`R^xD_|`YyA9TQxkRL0$9ju7P|SJsg<mC+#ZNMH<IC`Ehk8WIGOo#O@jH=j!|d
zM6j91d|Tt@(W+}WBZ@SuI`t{Y)O9=o!Qp?Im>$d=jPA@a{T>B%p*Hd9TZbDR6R&Jq
zJ{-)K#T}=%X><$8@z(Br;iHS=D1IjT9Iv+GYzg{0yVj@2W6~_+>e0};Ps@kEg4_RX
z<uiK>-mT)ye4Bqcfywa3VW>}M;D!P);>(S>TDQ0ISO+{#HoXrYcr?;ag6+-z85M@8
zRG2C&hh0VX_Es{#g77kxI2*pV%Sm4`cZmkc>B@nWS)1$#z^whx&Euc%z>9G!6|{f2
z^v~r2k2<gfO05HD-w*DNhV)MGRG8ohO4i&seP|aDC44y<O`w6|&RGy;7zpQxQB4$V
zHr)9TKgwmeeZ^y0+31D4_gMY?%w<4WBSZM&&^`Kx3w%E1GqWy(7pbAcGn?SC6QM0$
zZ|3uM%=7`!`XAt!>5^tn%C^P!8wXQlBP%QPXBqYe_IO4RSM_TRXoC>dAVevO9N7hK
z#V^(krjy_RjGAuS{+QhEzo`x1HGba!)_sdp${h=Nth%V-!T0zta(by{1fD>Y=WjIj
z2r#enj2QM=LCbqyVib4#n!q7CAL(t2WLiJRkXUG5-=RLmNmrEKK9X{b_F0+dGQvLb
z=tCF|QWLl&>~N^NevJk_`eZ>bu-;|dOaV9PPsq)RFSnl-U?VHuuMoI1PsDIwr;oMa
z-+ktzlwP>78<DRmW9J^Ygb1eFCSS?ahN*I>j>R=GDlRjAA<a=nR}R+NSwz(`l`md~
z3EdB>P?acoFetEcef^<Ju3qK1c{OW6%pY!A(X?pT5=rRiujkpcoBXx7aOOLr{ZHs+
zba&#%HiDSE#gJ5@w&ku`9<3r{x7#I{f~nX!Gk5zpXF0e-qLCpxU56$yoa-;Oa|RBa
z9-eCg6VIBJz2;98fo(>CqY6qqrJG^?$5Y9jk*|${OlrrHVRiPS9`}awD$cB)cuXMQ
zx#cjGxR(eQS?D)1nGvmY+yQ29@u2-CakGgg)6p8SPlce=|L4~FKR%(g#|16e-%j8A
zvr|e()D`-i**pM^8MCN;pp}5k^x}fk9?aGnN$o^M2K27f>W$UjhBvz}ezF8NNE~iW
zI!K;6Wv8~Kff`?IeASK%ydlA^Tb>_|;JmC9e~~^9vZc;A`y12w7M(W0`eKC(pqrER
z{hd{}2Jt!aKbt0$Dq=cYQaS)V6cK!V0I^aDe)1e@Lg3#xK6!woFq!XJJ@8ToA@ca`
zmm>J(=l!o&f_#jY^F^QY)?x54j0169xN|Xc4WPx2gJ#J{96mn7`~rxaK;cfp2Viq-
z@;kOM+`Xaa+@}BoxhtWb%-`;iOdYj|`dvDK{`8(YnWW);&E&ZylPm&D(oh5-Js<8)
zdl34eQKO)7<rJh*F2yATHhVVR#8K6C;KU;mD>)f!0H`Gpz*Dy91^B<#fYL)-Io7ah
zcqonoUvhP@5ue`ClIJlx85=QmO!Z_o)$_(Kz`&l13&t!R$}99-;L-@9T(Si{fy<Ct
zzg58Ogb^}0zt=EoV=Ble<{xvN@{;2pjr;k|Tb<7vw49WxrPF+SA&yK7<$#2wKv$Qc
zQWlan>wAejSX}NtbJ=~MZIU~cW56xcz}tm=94sEM|N8N`Q_n*ffb2%>ykf<^vJvNe
zZ8QO{S^YUK7Uv?nEaS@U!v}WNUr%D_7hX`WAQSaB!H*(b5b1G%W2v#!d?xSPz@=Fr
zL1GZ?x}vUCuAyTpE@v0y#>O|Ffg8g0o%DKXp;tam8T?L9b>&13^3WV;&vleEla_qM
zSs5JUUOCn7=i4htTA5MEJ4#h=huf3F1!1D=*IA{;?8cn*BbmS<LNR5gXs=&c>A{Cu
z39pM@c<{G!<+yZptlW#KWg4(W8vO!ghCKkRDf7?WTp7ao;MRZP2y=do!;j1PT*l5p
zjd<%gd9Ve?yLz)#PZ$&|VBm*b1&tOfVhANyf68my=+93jHN{uHt8C9p5newUDpyKg
zQF`nHo+>fgAcsgJaGMCU$hozay*rOTR;ktpl;s<)edFf8<s?`3nm-o)N8jat*2~Oz
zb5S5aVy^gE4Evww?+60VE2wCfJo(l0F6V;dMcMBlxvn&(hfwNB2AAXc<gl0$PB3g5
z2Q?+eH6Cb5N`teZOE?~woWysKl00e>#PIy?hAe3@E;lE5X1~y<x*A+dG84kM6$*8h
z09~ukfwD0Hy}b66If5E;YtkzIM5J6A`h+%u+_G#j(naqqMDcIhd@EL*Du%Fva~&FR
z2CZ@ZH(Vo+r9lquU$g)oQ|n+~3L;!sG)t1P<IU|k4$18h<lgj$NI|x)VC6}Wq!c7v
z0K!{(ybpe)Hk+P-mWLItl^#@jhgBQ|^{#;yvSCvQbk~gOBw-zJ_B6bM7lNaA6s|SR
zVl5?i6N~T?)tW#Bz=)N;B_mxDVg>2&g|01AQ|!He?uL`+vuwzrbxDKez=-{$W}e~A
zWPEUfK^X<_QQTvFm%+<oJI9#D`&ToC`_VQNNG)P94Y<#$^^JyA@Yed^qN>QRTPyhK
zDd5c5IvxHfboJ)jrO%v6Es|Q*%-$;ldXqbG&nE(e{21k-G4#taj}jq8r#`<oe+Sm_
z#a=SyZAwZ#-^=!|$JzVR*^nmMq;H@MHmo`M%bAaol`BJTb8Wb6aJ0W)51z<UOJ&>-
zKY2eFiknHkv53`v9`bteJiTj<aYpjXgu<=U$-|lv+_;7!ntGsDzFO8(@xe~*?~_^<
zPq}%y6WNK32pqt#&#;fYkr)mY7+m)W-6q?m=dymVRwP#vl*tsw2?4M_o4rV)@dFTT
ze@CZ1**ico>L3M)!R6WPeJWAH`d*0ej+K5O=b>z*P0D`^Ov&`J&KoAj8u(hl#K81D
zHLfZ;aV_1Y8!WI=FJ8HqJ{|V~DW6L8@^mcS@@0Yh+VV-9y}x@SG-_#ujCErr$rx4N
ziDQ`=RKeM#Ck?Mw{Lo74%!&#|mj3+PqJVeW)BYfV?(}dIPlcnM_}ZIz&jJdwKfZoK
zSMwPu{OkANIwGju_vwExa+s$PDL9$g-yc-IzKD83qcA4%FPrW?9)dJaEO@s%^gs4e
zN0JZ<9aR2gcKU2MF6mv?fn1YgI#$O}PYa^NqRWSnNHhHl(dBiwoIdMehQkGow(Q2I
z<{>oS%r~;coLF>}N_cC2>eI$mpG<6SjE=zkD}B542m99F4@;s(W6Sy3obsL~k_f&S
z$Ic?w67}}m^`~R8V!Tc!&&SXA8fp54P)qUUb*xS#;dG}v^$J3XBpQOQ^oivTx13ny
zyj@#e3l>=$Iu-mhF2_XaWrfb&;i5s9+m#p3Rn5NR^WH{8Z3rTp%?9jg?zG%@pkpsf
zyWQ<7V}ikJj;L?}hhsu}8d+!gP1abk_#Gz=%Qe^fSD3hC(@GKzA)1lqH#D3UyvWYU
zy)03eCN0-XBhHABL<Ff2$~3W@HkaqgL1k$HNEX$Q<Q1_UH#(ici+=PwTXO$vXEP#^
z1W$-&yo7x2+(p2iFmL#xaBWULR#Y0*KnG)oCF|MQWVCxNuugUcCMA-XjvZ`gw24F?
z8T5`glF?ukD;4zSs8t^v(1?=i73yzYKYNGhd$PL{cOpo0*}_hnf5mOP;<RAkcTBVB
zk_e5rC2v3{`iUWiwWd=+8m3Ve%$X7HD5Hh;;4QEjPxXEe*=V(^=ZSg=3L>{<7Q1kP
z{%O<sPJ#-C)-s2CPKdC?%7@)nk%;{Qw+4P#tsY*NmHosROtBJ5oudbifp#&LQQE(@
zT~K7vky*M~5~Kr=;cBJTJwRU&ZkpjEHs^Kg(nlwj6I2DTNz&2u7gqR+wuERODvdvu
zJI@*TeQ$yWl1T9Bzb2`$vG~#p{gVj;Czej@PUrX~m*5Kx29)LL;SkwTb$0wpuq(;^
z)V%%T(YAp@JwamStuA!4$=JPqHEJiy+U@uNiePHA%YyXCy3TYEZ3EA^`G%r^+%;Wl
z3dORaY-8QqTTZJ_StgxO5f&OvA_WT@33FnuCokM@KA=zneK+nX5fuCI{;1u9AWD%z
zO7{10x3=U4HLr<hm9Qo`i6$*ut_l&KPA5X_qV67X^cjsEwyAjwI;~nRTbiW?2T|0r
z`rZvS+NGJ9oMUv3Z@_jjOhr?W6chMq%rbSPAu&@WoE#fm_=)I|=4onnxo3zwd9;c{
zFZ+2NQp@yu%QE#m!YXi(XzJ+RZI6Vv&HEHBeTo&=+atsVE{_?xhNqO5qDsE5|9JX)
zfV%I!6vZg3u@YM6`gCXxS-p9z@7IVrh@#yY|Ck_(9U2>EC5ICZTa8StzgloyqeZS~
zS1ZixurF~;S`TCq@VQcT_e2HB><V0o<BV?H^Iz7N3@ZQgu5yyQ7Jlk^GVXdTvSN=q
z$yg6y@flq9N!maJb^1O1cZB8tQP5JrqbR_@<reh%zmBB$)Nt?QSDAz&07B4VCne$K
zMtjeY7%hoi?%*Ygh6c1$=%b{*W%5)krPIw;YTyU(Y*34Fzc49=bll(gAS{;_9LweE
zbi*`HreLD54zdb|%B-<>&Q;*RnEn=tbr4>Vz5MnaLAGI75k;_YPP%V?f@l*Z4v~+s
zV@Fh8K7JTh^f`hb-gFzDT3Px{g{wgV-?{lczITj|q_Jvb#VIIF=fR%Z1hl6){)32g
zi%uls_hL}xeKJIVg<r9==T%b{<dQ|Tz^8%v?6D-DflNjXqcR;zgyvMvX33=z8K_9(
z4NF~MsA@{0rpBO!9iS^tG=+(k{$%wGvkkd#DiaD!R<0jXG>C1~8_uL7c-zr|mdhqC
z$Ep29gy^4MO7w{VQQFx9jE;Q3@)H!OgP#HRcl!~^nDhNkp`NhU&2P<7{D6P-?WoU)
z^-%|5Km_iD28G>hO0toRkg0_vMQRkOSTklZee;MMURFxBvJn9dAc}3tgHbbLx#dMv
z?b&)xJa{CRA78Tj_V**1;LPw}!}!>4^9PQ4X*~KT?bP@q)hK8hbP~WY8&W|}mT^*U
zUORpome`a{KQ^cMA`G4sIHOODMd(_8iS6V$OQ@`YjT;^`D!jzZ((DET6TP=Z5NUN+
z^Ha<Q>$;86NfK7J0GnP>Zrp%Y%_>)5aymp{{H{6G`<0LXtSOwE#dA`&e$y$?>ilpK
z^i+z0mbMd}p27n_PqD2ZeuElxN`KlCmkTlOxFJ5Gf6sj5au52<ZoK9@)5h?S{ypmr
zJd~}gE!)(FVSmKKgVA=g<6kp)%uYJVm~?Wm6r2I3#zg1j);m%8V>HasXmK9$S~M03
z^qFG-S!-{H<*&`#pCAyxG<=Q_bg7Rn?=WzJr<O0|4{!6{`;G&R00U2wnaxQ(A(QR=
zHXz?x!)#_rwl!VRxf$g)T?HE6bM#Y1fM_kgVdh$H@OBM1lI9)br<q8q^yhu}GKPL=
z_%xpPFoxu97mu96gU0IBg3>O9PK=+m6T$^k(DvJ-!KX$TtPyc#IW0JWL)nPcqP~MA
zFS!WK?=e;c_tnFvx|IehgZ%sLfdNP$>`&4=`VK0vOj?Vn+SNTNFQd#3c*#f%9d{+c
zaXsk2jq6ib;fjUU&(%Bi#2dMzR#a%=x8qpch#dxsd?diO$TPee*FBOk@*2|CNkn*d
z@hnmwoF7_#LZ*uJH>YDZYGEO_<E#``Hd4BHh{iy;gL|^L%PX6cJGG>L-HBNG5+>VO
z#{7Xuk-lO6mNbJ5zIAHa5iAKM60}U>Cz&R?*IkW1oo6rClug}R75lGeWD;Eaion7^
z?oV^}2D7?lUv$UeRd#2pqjKdJ$&y|b<=Uw=POm||8em+sNp&nnK}nmAkAIEGhGLX?
zTJ}-!XivU=VkL1zNd$PleiBe7>q!Z^w!TO-b<MK^ELP*MfFdUp*gKyY{d7fH0=`6c
zF+;B`uWB-FHaIu<@^ujkH3s0U?;*vLawUP`2-W44gLk-gw6B9TSg{IR_4M;N2GKBz
zJnj?|)Yp7#jv(e3kI9ML8;5)@lM_QX4)3q`g4Pd%-{h$-7&H+{9FCerS`DCu%gXcg
zopgp+ie%PQD$stZptt1srD4tU?UAHP_3QYlz74iLkJbm18e!izr@(?y!`Hu1cl|lE
z-f%Z*gciF@y`)6pxB;MMU+}UFvCJEH;Zr!x?ZypJd~cA(u8uM<Ie?Y)DIR==c;)O@
z9({Q_zWLm~D_`rK*vE6CVu=RXV<?zt2Ktp_lJeZzqD~p_F=ZGh3W_|E8XbK2BAW@8
zo-dObss4Uj3nhK#FtV*-*^XcI|K;tY!ec9i<ox_+?r?)my%k0mQl~I0e&>#5|Ah<p
zV2|Zxo~?e<N@PT<9#3ZsDio5C3ZgV)!(%IJzRS-l_U7{|h-axLKY#Pc_GB<yQhTMl
zbO(Q`59kl(1rKvFBHGa(o?FM1O}Nyn6>&TDDN8ztB07HNifJ}&y_u22u%?!L-J?2u
zcUIQR{qR1@k#;1&MW5vL@_hP8b8>l1iBJCBf$iD!L#Nae{j&Y0@K=S?rr$aVq-SPJ
zSD-Wr@@Mt1gD|4#>U{Y9L(T%tybTSSk-$7{)M!F^^FV|tfb2L^Y7KIs;<SGqys<$c
zLT+Rko`GIz$UPj$uX*Y6u|f84x+U4`gaeZm+CyufgBWwgSV`iSd1oU$3fl>+3q5<!
z=blgGZ@B0OA^V%hzIfTT)_d9Ij2~1^^E*)*m7x|tlx=+RdcEr9m6ev*c)6?f4m9f{
zX+K27<<D}j|0t<koxky+6KV?}-YZtm`qt6jheLnu*(a(${v&q!f7!E(nT^dHDgSxE
z<kKWCQaer<toBmw;~uh5PjKQV>uUfVhWK)V_yTK(4$WB8m;#Q2(Pe6^>`e{!B;{;L
zF#lrtC?_rYXbIF_ZjBRMFYSo$ntiVPiKo%Kw>%)^>Z-%J#b&UPB@hrwbU6>t(F-&+
zd(gR>7`xkf<cM|uP_|rFdil}!v&;}8xru$-v3tFXoy=R6+b#4$DAq&<kOO2ua5cCg
zN6PN3m@C610BXQ?(_zuv-lon4wAwA~@_C%%r(*Z4AZP_=%&_=_2!N?L1=BFb{oE(e
z@S|RX?A9}YU3<{4j^<aXPG^v%T^zn7G&W*)QRc4fZCC9Hpj$uBX8X<}Qij;=oQSXl
zDj~}jFDPlrLdit%PwSpsd~dsE5j`*xKxDVt&ocim`DKA8)!xDLMS*_W&OmwYr@ga{
zUo75$vp(=4$}s7pZ@XchdkUc@<IB0`It?_lmBZQ=)3CGz9Wi|}$Msnoed4%dT^A0b
zLGfm12Y>yAuJkDi%rxFqOSN~?JL~<mM!NNuth)WzHS#tX<S0JYar`rllJ>`^VDALG
z1&bKxl@4e^t03iY93vnrA@jadbT_Pl{#<;y^7xUVew{*7y`N;mI~?6fE;AlNBuXL7
zymtgfxzFxSPv6-1F6cy)iR<>HN>xW~;s#XQ(wkt=M7P|w2tTd#80?%NeaYvwtlTKe
zxXfbIeBr^t!f%SBdePeOO{`6(n;lTqPfr)z7jjLee_lgjgM)yZoDpeI$f|G)Yll!4
zf*1jSLJLg5%d=OQB(lbhmMAXbMotVCPhN=qKka>aIF)_dZiy_({A7rXr7~7B%M_^;
zl@gi9%tOYWSV)G9Nr*C}2o=k;tYsdevJ5F1)*>@yj51H(b!&Lv{eIQc-hb?4AN%<F
zYaK^x-R}GTP1ki^=Xsrb(^#BUbeBSBcMth#&crr>zS%yNRt};X`{m+Vf1RzN7oY>J
zVh|?UmV1sTvbE8an{IMREc>H~tt@%;{%{>|93vQUTA#EyS?11i+c~?W+m5Lk@sQ3<
z92+KlO7W+yABJ96wC`(OX_oFC<{>_sM6gg4dT}f-dFiH9zlG>E-9X<*tZ%$e>T<UU
z2vW`s4G7n@v`)Qx-&Q#-+q0-~rwS?ZxZz$VN>H9{0)o~3XYf$FKHRes(~^FjJv^l}
z$~6BDL2DWeo*t8d%Jg%bLI{<0s^KorUO(qZEl2CIU@@+YxB?%f^mVi{@UBRJ>#EGE
zX2_GKy&pR)@7br)b%XJ^GAq@^8UM6nL^qGbMB(y<{-}1??vZj$Lw9wNZto+x4+_?V
z4C<~Ds1I6Mue(51GqAQ4tc{P{-9bXXInnvXuvmTQxZc4YH`1D^m(H}Ao!>2v&^xV*
zxhS{KE#&HxW3B6H%`h|hC(r7%Z^V%}N_zK0?^7nHtJUPs>;;r&P1I%nUFXyMrUblT
zSFg4m%jc2p%SR5X>L)tR2j`G<;#N<8seS(NJF@TtV)*wD15uTKF%E^&{|kT^xL;XR
zJ}V4mwF$eorzOQbB&{Y7NuC?0PC+W@tjh+wEs_@-F55wTX+IR`EuwOe#W+5cvO&1;
zZYZlA5{>Noj8vrb6)%gN%Y#$21q-N+9ukMvo_S{GP&TVg0zj#OWyl@^ottLeHt|S}
z#>+%5+Zie<38C$FnQ5_kP|L@=IcOPm)nZmhFIMCsEWfU83HX%yq`yb^yF5AJ4l%}E
zJ-(1Hi~R)#fKU#nVqOM`9#>=ZI!M${v2tK|B7DN3@TL9=>tg-XzAGn}U+(4bHtE)X
z&0x`|%m5fO$qtMG4$Ltni=HO|q8O`TyX=}9m*`x_Hf*Zs!jrPtUmT@jREdQgtrLkS
z9d!54^l7INoA#{Ui5IneF(+IEoo^MHW_f*5Ld6>bf{#h>@DJ6yZh_wC2{>@P-eO4c
z9fikwa?i*5oA>#HS(Jq5Er&P5%oz<7$;V2)Je1A}vbD0+_d^Xe4Nq<A(q4F0d-2P;
zy(Xi~A-tzWiVv}f5fO;)c!}P+n1_+g<e|>4+7$fI^WfyX$EOT7nF4|th(`%x0jaM#
z>#z|F-Kx^>=sc;}Em{FE4=>H5&RSXL<Z}x6kqq`hI%a^<{PC;={Y^yo6GbR%k=r5m
z7ltaEpiVqHtX)ZLvR?`!?#zceUJ14@kUb$3-|);eUe7^{MM)LkGF&_)CySt(VD%Sy
z4IxBM><cWmAKr0M-$dkOsOy<#mP=MVhD{+$gVzgiq8Vmgd<C!W+2!4A(Z<o2#KiJL
z;fP~##4yd#$vG0QD6R9l4Pd-7_xk;h@0N685$mF^H@054gVUFiwF7OuoEf|d#G(3Z
z-?cn>yHy|a*Zi*S)<qh!69qW6?tbIJOiz4%lMF$Bg=K<vUV1e-G(S~%pgg_#kp^x@
z-ae+KmezH8mLWCW;=3YC7)`H!rWSepabm{Bg6J;~j27!JHUzvMHZ^+at@+?4R-raV
z9<lIBpjHiEi<*nMl0L;@Va3bkrh6Morp8S3R{e6P6cU=Ak=p3#%=Xpo>0%1w>z2tp
zD~f;dQLKB_K1@vixdSlG?}Lqu;lT*=`>p1z*!2_E86=j+o<DJ?|F=as5RL%k&s6Sk
zemht)xP8yO?Eb(;U5^rB*0ami@lri=LuZ>wCoCIFH4bMLl+wAwh~Odk;TuQ!?)v-q
z!nBU(scYaPwv)R1BdZE^@D8L0XTg5?Dck_9i>=0R+faW|NfwQtzuqM(9wuw!sGSE*
z3>7<=PlR)!=;)90+)zI3*IO(6P&EwU4>dmHkJodfj@rr2@PJ#!Ij+okyQ!i_l#N3;
zKYD&?#9{vgSU8Fv+0Gn%fZm0hk-H*z)ffwaP!Mazy-K>x9Hxv#GOQn!&M3)ri(8?D
z>964LsUiGIne0d$-DW+*`|ZxR92HfD+a^pOl;gN~xRhTRIy-SrBeo2YUONW*B&+Ka
z-HEU0Uvr~U9Ew#1U2Jck)j${lNV#aZ4O)To)q5dwjH{>F@wC-hyIWn<Z}CXIe9VxW
zRMM9Zt}On;%zwSj4AcoKWyGn`_t7Iscm~1CgK+-YRBMat_T>Z=sI%kn0Vj?gsPPU!
zvG7oXf$?wx0buI+fJ*u*^Ml;|N}oCe2dBCXU(l8q*@p*MiBS14=%1zR-jM_Ouie=k
z>(4kN1}&|)kn4@l%Qf!35(i=K6#XLaN~-)>iOFm^;x+P4)D^3@Z;o955ax0Y%0w;0
zZ_0Tml?KB<F)%yq5nI|cD#@!yZhOK@V`sv+&$*;guPa7|fTy`by2myz$SXts5$TB7
zm5hp`7lH!+2ixv`15^h?P0Rb1HNN2VsiR-urfmatayr%@x$+G@oOn!=E}A;ASEN2h
z0*npHW6Eg{R+OxqejJD?0X#TFXg*-2nmo5aPpV;pt+Abd>A*Sdpgkk)6O{LGY9}a7
z?xMf+TwYuDNyPHkx&SO2gcF_BZB-}AkcsM45`+!vyqmEOX1nL;m;xt!X4<pfvWyJ0
zc>FdZ(nKn&nrROTzE@W1$b|UwQB&EQj&=mj@(We)WGJUpUOvdpT0=Z9Z!AvhtiSE;
zV@*10yN{A&3PQ;-R8cx!wbdM<uY^*s2=z+{N%AyYGLKR{cbIR{bv~EpHMeH47*3)e
z+9#nLX0BHCGc)(iRCPBOGC%%IB9!yB5<)N3N0_klZn?p^{}ag|H*jI=6X$NSk7;ti
z0MfVyHL-{KFX{RdOLvrcaiHwz{hbpXHgBdK(&3lqrOgiQO}ogQbm)P}n%r~m$(`+(
zB!67wl4`d~M)0Q0yDXy&S1V0Tw$+X~ZMSY1iT2E_(5I`qybUE6ec(Was8M@2jY+F+
zQ&pZeBRX9wVF6&$hdK$5B5pfQ5<Fm!e)c8CfYL#SIKBhde9SvCYHOjmWx<Y5b;az=
zpE2iyP}I{qEPMkhyksw3aCak}QmHS%w;HLKwYm_m3F)5G&~Z3De~@{FzBT=|YZ<R_
zaE6KhYYtl@of?&{x>dvNr{@)F7sl3(StS->)V#XM(c_OBt55Q9SgA2kRPTBqXIg68
zt;}t&t-D^cxc^KtM)@R3v-3^+qw>GfEmj?w?0t&SOE)&h@<yw^J6jO?@zTfqk6znr
zZWkPWN7sf|V`7r0oKX8bGE)ZRa*oo&O4n)#)L}&*1V7efj<h=72>von{`6IBxUAXb
zaG#nl3bh0p5(~>f%XbC&Ps{PFqR4a+-Sfcp2W73{Fu7A)khsFr!+jBFzhYp<rt}|x
zR>F$^pb!%r<$N{riKoylfgiv4)rT+z1i`2{bT;H0qXcN?`!+w_Y|=+98G{S0&1#1F
z{V-Jm`<@L;Jbo^}S3vvxo6Eu1xhzbz^<w!<Omt3VjM?wzJ3Q6hO$Y&bQ^2F$KL~yB
zmvPq5g}Ji_QC@3M)WiTiP4+osFh9f~cc<f=(vELnDk-YHz&?q<%Gp;YN{9qG=R~NR
zclk66^|h#R(MV$UE1=%A>p33T5PbaI^Tcpg;N}V=hxd;^^YaI<lmg0WkCednJ>Ojk
zRgC-Tae!>vBTV<Fa|~cXN`Y`m%v~A|2H6Am3EM2c+yM@5^3I1I-9av2xqG)0lpaaI
zX)uR4P=Ao~iMv+%sa##l_*y97tII+gBFmiu2oVAariD#pR+)Wn0CPrs=~*P0U5K4V
z$1wcUd(%Y$eK-HkqZyv>gro!M<VikjS2qYxkw@lq#)_$hFuo9ZT?q_T`=&dc*7U`$
zAaV+DnJeYX#pxKiy}KZb)oi8ta`B7aAYQ0+VAm7|I1=f0aoB$}K5szygpG;|#Rb2k
za{c^8Z_qT)gU}%-B<$B_nj7Ex<pnCYBdn+GH*(Z}>Bjl{3n=b@6g2ClRmuP1{r|uD
z0t%QG0Nu2PY55f4c_Hyl;Z&>WKg_5fTg+W^WJ#J+WRHJcE2reipf8vrD*fdmhf9MA
zC{5M}{SO36qgSS1Ou%BZH*kIsz?=uOJ8f3asF(={WSWT=3TwH~^6CY!Kt=FS(7E;l
zN&B&z{QQG4KSG&3Wms9Cv@s@?4M%%{agYmbn2AWv>vyQ|zjHiY%V{~oD#82KW^bbF
zZ9ouQc%H6+m9=AzXRYAYP;*8u@7A+woomZejz~T>4z-i}uDdyb?0N5U67=PVn(HG%
z$)P_XHA67^8X@^<@Dq8zExdu-l?B#s$AR&*ZTc-Y)e59wHae;$D>DQHaK<2oo?CC-
z#vyVA9u5gI7>g2kgmEx%%mrJvGulj)#-8K3FOKa}fR134*<)MuE%iA=_PI`#&0{mQ
z9uuL6AT9DXL**Zr<JWZ(CXDc7V~euCmoASoAzW@msg%Ews@LX%$R_B5G}TIAwdAK0
zz=w`FYYDvIz)7G{VH1H2hj8>efg~&0@rr5X!^gLeN&TUs4zS%zP5@hQoxC~x_P?n#
z|2;?k&(E*P;bsz2GNiuK61Stc-Iw5n^Z50!VjTJz85hKroUhbQm-MVQAtepi)bJq&
z_}_72e_(-r*)UPZLeylxO)ld)#*K*&o8MgvT}#kvJwe=!`{oLTE_^3FT9laTy}#ZA
zbvTb@n4>2DTfE0cm|7n(Q9klOjGlYWq^ZKWu$IbFw(reBU{<nOgGyFyJ3k~^w$_){
zfN9fdw7zul8t{f`n(ot4vN{>&d;+`g7mJ$mi{|&&ffFj#G_>@*C9-lVfW;7_#^rr@
z#_yVTy>#u-oeT1#mE!v~N4Ibse+#S{OXREpGf6Knxm^_=-8yui?k3lWi`?wqcY=2r
zGa~JRP88i5)mme1`smJSp+Q8o_8|xG^ec%`tCL$SWMGVs$w-NCB?R+varFe@6*9J9
zDj)*4huB%X%f=B?wY4_mq5K*Ce_<Q{Ig<aDg~5MC2f=iXe*82>YhzdO1MMzk#%?%4
zRaZASCxcQDJVdzhm74ra6g_3k<d^Uhz2KKU(g~-W1*ml%B2Y8c<}M(O(@&m}Bw9VB
zce_wKjW}+T2ZHb8gH><_2>DVvZXqhxG|~lnj@Qh(<_v-%RayP)?y|hu2rUZz<snxm
z><QMe;5CCs4S_931m2$n3538Cp++mINP;=H)Lv0Oa45JFS$&k0D$%G_`s+yD77IK$
ziqQ75NjDz-F{w^TZinXvhd97TUVE8fv}T%fwb9lNaJ~Es+Pbw8o#sr-z>p!mQx)yc
zGXj*w9hjpeI9pBWj(oYrlW+~dW_zGmJ++5q6OlL&!G}aE_Qo-+a~UXmNl*mSN8AyV
z|GN1a+)_?V)^M#k=|m&+^6i|98zg+F#&HM%k?W>Ztmi(a`YnAuFj*Y+><eRnH)}bU
zb+Jsq=WmjS{>*z=&Cl!~Q3rN+wN{Vc(bAw&;+=niz%GDvVi1afI{*%~T-GV4`tTr=
zC7Ol`GgsHTKH%>^?mR6$zUmB{Aw@poGwWFa%}>G#h6jdaEu4D&D7D)N{nI~$xWAtQ
zya%Qc#r8DejP%!|<L*r=Zqw!Q%L=wSp+{c^VrZz1Ec*PpR#GOQ(gW*`9Ws;vp1LK3
zr`DefTy=T!bKpp6fm`bazz3no3l?(B_xjTjMM}mQm5-{pIC_ui*+Va*?au6`k~d7D
zG;}Kwd~gv(+70o-`sb}3N}K?49F_%W^(c95A+wR!H0FAc6<rGluQ<6&aGCrkKe_?~
z8iE6C^XWhyL-WX7LzGJ+W=b?sY$JSa2uNd>=+;I>Krf<#mq(xENO+oMIs)U7_8`Z|
zfO7n2hLI{VKb|i7;-M`scsHrsIl+NJNPWt7i=be9k2J074Z%!4<XX+(N##&Y(Kstg
zFC+rgxMSYjQ+8^*C4u~2qud9Xa+#D!^42#eycfY@kt6jCxgWsi&8cWcIyYtw6t@R3
z_2{#-b?@#@Ms3kj^g7vfN9jTet=7f;m+L2>z?a9QA;ps&dOZ4X;7e@YJ4?2C1diTa
z)m6I{?Rg{al*`P&0IbOoR1HQtyGj>_?>#i+yCb|+$lR2Tkvis&m-P)_QLlgH2B$WW
ziR>eEDR;fz?+q;YXng3Kh5*fY>KtOfbDwb?+T-#){FxkkE`_cg4oIzTpNq$ncs+JP
z*Bc$uK@oZa!eMI=4zo}fEpr`CeS%#*z3TMD138D2@x8*!gL@U?@k3?UjPBcdb1+f3
zXFeRokF)nI9D9K-lD~@|Yj1c)V_C5>Bq}>sqFOc%5>>OoAxH%cNN6-JMc9Qwe-1Xp
zBhW00gf5bcOZ1v~6|V=~5K>~Iwz8lLchy9Kh+4u$@M^rCP3S@hB@QP-g2ix%<++8!
z%>ypE`qQ_Lp`mUud3Nu<l78`h1F+KgUD|7_3P!I#iuNjYD;-C~+a!}#_97{^_65YP
z1e-&m^C-GEQ8xd+o7Q?_xyhrPsr|7-4k)auu!Wx$j9~KSjW;@?U52&E#fn07yAt1p
zKg?VnfSucZa0rqnOSPSaKj;ZRv_e1qV0?w~7DdGg-of({U(Yyxc|SPFMTDN?_t|GO
z7-!NDoJ1`15W7J>+O8PW6SOSlBI7ossp@A=`60E3pFX=lp+9dL-cai?7vJoaj+k|s
z4cX7WyGZxM9e4TJHsJW)#B;Zl!*LZf*6Ye4a>n^>2!--ff(gSjq26qF#2?4SS@s@T
z?EL%b2#=1{`5vy=?YWmOGV33>to2xfhTy46jmAaSj&j9L>xZ$5gcJHJ!w)z_<|+nR
zC<={<@4_g5;C<=zE=;^GHQ<Q;1i}SAhBr0A2pRLVP`GaWjI4}ZLB^des>voMQ^5Yz
zd9jv9gZo^2#7UAG{ZTIav~JalfM_h}$kGk$lUoxJIUA_UkCc2!4JFQPyxSUzG!vRP
z6ljBPrlLKD3O|o5|Jw_;v_Z0$*4ifxUn8rAQ{)Ns3p{5^03|w^%Pr+XaGwhh^{RqE
zs0eH3-tBe~4w!Og5uH$840g*ZrLuB=b-)w73xbo`v-e!NlS<$~Z4~akP&X92#!IKh
zdDR@(cbc+4q(gYe&g@9%3X*WIJd0n79+e$ck5k2%_Fog9YUri}T%R8i&c4%aV-OT-
zs9(>U?n3R>rvB-e3`23$W%+xwsBWi{Q%O!YL!}(x7zkC+C-VBbj=2sq&kBf!S2hbb
zt9I{`ecAKi<^rws;H~V2nGssPPe@;;acD2wsP=Yx*Yy6$Ktu}?Bs=J;2yq^}n$~;}
zIXYA2_gL${cK=(DYQ`P^@LgwCC+){ijz$t&ooqpZKq?N>@`vYpi{LzEi|u&7I_VLY
zNW@OS7!@LG7xv~zHr`zRnv(ak+4Nnja3@>>M6ltJdRmQF?~k?05#SXfA%1#+wJPLJ
z<mApMB2s|E?ACD;At_AX)a(GgTL;f4b<;fLEbfYNEZP?~qdZsyfOkL!W~}PyBvM<O
zfa*?O#ro>h@T8|b%(76|jZ+?RvMJ$}@F>$~la)iLx2APj{`54}<@aWxm17NR1#{|V
zZH>d`=}`y=r|B1lgn82su}MUvPtsAX6vrgZ{ZZHrt!5x?85Hsuy0igK7?UyHjUubC
zh%=B-(Z@+mNFu^09@g>s1$4yGzaaQlajYSKI%5E|{tc1CbGz>T%77_g)w46xF%&a6
zwW~|_O0X2=cUOb>)P$1L(>SPd3Zif<h^(t$5|pWhTZKc~<ZsyGI5;nOLf|=$koI+s
zfmP<Z6{-=V9h@dhj)wXhwh_qeX>nYlWD3vYa7>fTNQ+i_wsUJJHm&1i6($2#qieT-
z!yJq>ANrG(!w3;QBh_~trBZ6|0X_;ePNKF@RC{F@9+U0ZwyjXzqvP9ErcV*KXXbPA
zdg%mEczDe9pn^%DwYERH9SfpA66_^CS3WQ2C$BbO+K_lE$gQHSDnL}U^wV{FV3Kci
zF+{D<JNIqdU^xT{4k{n!TfNEN2d{(s>8uGpht=t{7g(nR?5Kt?P+CZqIkImaT+jM!
zSkrZ=g{YC>9#O*cTq*?{S|aU<Z_mpbP&oy}mO><DZQSFmH$3~Drt=Rkz*wINh=Teo
zk&NHPhwFQMgHeLENNnpPN&5)_iRj{TbkYT>yY;^$L4j;TV62bl{g436azbjbJuYU*
z@h{~{#JmM&E!}bz9ol&8X?_G$_R<dTLXA5yt$>J`2!x2Gwu1^wEs@(LD*%4VLBf|9
z(*w`#*&T2X80Z?bJR!5F8GSDc8#88D5;Bl9*dKv1=NN9r3EXvBKQ3pH!q#&pO0*_y
zlU5}Gdy_nQbz*ORDo$*-<%(URpF}$OVk^n0?{Fqr#G542&X=Kf5vt++m4OUS+|$&V
z{^HxkO*MkqxN2MwP`|mLCRQ_3GEf>W`Mx6%In=gXd=V;Ij_c$68|pcB>H{eX^iFG#
zc+=TN>j*w-n$5EHHyl_m^2mAM6XC;AnNt|NQ=;*TU4}F^p7-&6PHy#sa-AGRzr7NF
z)o$5;n?beUH0_XIO@qy;+cS#NH|Tj@C52sLCLRu=e7aDPQ(Vzqv5H;SEHCQvGV)lR
zTb-2GpLZ+E9$Oyo9=W2hnT#Ao&#maY>1Q8&Dtvu=dK|}THolX><Ol6JGNGo;G(-@s
z5CRCDiz!b%Ci@0vCeHS4`f2u3F~7WBvhH_6MT#@f;N1Ph0+LErrU-S`eI_`NJLWvP
z(hX!gqbcj1uL==g?A^FIfYt6mxSjo;q~AOOX6Kg}!_+q&tin=IGZ%<@%&d;(REPlP
zr`?_IvAQ%n1l#j2rr~-Wrs0zcTHYeb6TXIZS3jo3#opi$i-^{g$45M*+BW?6QWAIH
z=({I`kR5{cc@~7aFNky^W8`ACzAovcXVz8urpyqJ&yA9CQtJx?k$b%5Ve7AJXa~`*
z)|P1XSfNb6d^WJHo^Fzwu~QD|jF8g%V)@j7^Ypu7SVbw1xL|5c8FO6KY&(zra0K$i
zlS}gupG98vuJ)n2l77zQx;YXc;oRPJMRS{D1<N7)uo1_*A3aF$B)tGjrgKHgNRw64
z;}?a=p^Z|EVT9sxNVW-}PSJ+FH|O^J-aGO5;~AnR%_GS^<xOfOeI}$2qw`Tcnr$bQ
zgEKihX<u0ik%}C1Dxj(BQ{8;xQP%k*o10d&CG3gk8}lQy`Y^gr1~{Fr|8!u#jk%xU
zHq^rGl}oN68{Z0J@q~hm?FZ|%mBk?dTG+_T8S{uZXR}wcb`P&Xb`%4X+Vr(z_8#9D
zrwe3uRHyfBe@BbjbAJ|i2QlXls&@$XmD;J2Xt$xk+}17hE%{<zjP>N{U;)1e<PkFw
z=lwyT)NWIVfhTu-9VAjAH`}T;1Sj4z?-H0D1URsC@(dg~kYCT#u{*ZjGJSs)e$iHW
zae1|KtGFLNY#sz{`=(C~VLNu8i8D4P^9^Md&LX{?)Rz`D$%Zy{7zk$f&L(;8sf@jK
zDXXhi;X#&IlhngUIaHr_xp$ZhK~Avq1_M*u$=OmGc8S(oCe`;k4@qq<4CCu7m*RgN
zf2O^LPC?F9wFs+Jj_833Pnu-SBf2|__oY|GrQ)a4={h48VQ2-D(M#4eFNBTCy-qP@
zO2ieqK@!ovLBbt^Myvb9E;n&>#W|UgC1(bDLSoeB7VY-#yl<FpmOx#B_alK`*xU60
zSnIjkJ$#9&i9l>70;?*L0Q{0O&@JYZOoV9pLh78@n0I4>_^ek+V>S^Pez$1jQ8)pN
z+D|E?sgx{Fn;Te`E+$Q+uJ=S0y1m*Hx7{Hg*qMtk7hg~aQA#c-|9oM89e+hp3i6e|
zrG;fwH@;Ozz1ufdK)=)G@P$#NVbV^N`WfNPC3fma-=y}+3uG3JQ8r#w>xGea?aOel
z?OIg2Y5lrch2m<;*FgDhj33V<Bxuck2COqDrbB+A*MDjKJUgK*^HSVe8XeWXJ*PTo
z=_9PoU#{=wC~}`5?vy87``q+}uP^PqqB^&L5{CSZ2=ILk%7q`~oIkF9p*T$8-8ADN
z^-58bGGxOHtYzfrdKV?&)7dZiZk92S@HyW9nO;#_uZh5>;cM4jq#3QQ!B$W|-kDF?
zID`jisNm_P!mO=KOqKP_S!_p>8cW2o$!VBA+?TsL)7vRhj!0wEsNU>3#oB(p2@&m$
zEt{3VA={3z8g>~;F{@jLGIZp}lUaQ705bZ!AczS9SJ{xH<F`do8K-x1!2PO&siD^B
z!q`o!wG)^_O4V~M<SUH3Po-6T*>9%4NvsOja2o4U4n*V8*}442c@g4qJ%bhbbIw;j
zzDX;*ZBCZFnE3e6<u;YJvvr=J@r$VqWs&Ue8M<5@)5$OMpuM5{zO+l$s)(ce?cc8D
z_sCjyw{Z>oww8~h1Rvx9_?2(>z7(&Wq@I>{!*{qTQBT1GEufTq-prOOv!IQ3bNWub
ztUzQ0A=$U2i5DM9XeM7Y^jV$VX?gOD$UJ|82jCttz!^iIq9_`UuNV;HDLHRX=#W^=
zF*Ur7&OL4{<X77r@HwH}oFjJy3OY9Ina5aPz#OQEc!+oONRAJc+I@;S=*INS<ch_;
zLGsxI0b44|L})HBpnSzz*4qcfj5cW{c25!%FAa2_-`i!t6?i0hB1z1Vtp<9DU08l&
z$8_(cS7vFu;ogaPi`6}^+vE{_=%t$&>9`5!9>8^!DjlI8!T!fdf7`g=(dl$XK-;_P
zQnK6(I&M90Vmqm?^y%2}%a!2;v8tCMM5_vqjvlo(cPyhu;j0_VhbVPPou*P_g0{2G
z9J)PJvFXYd*Hh!UD1UY&e%m=WldX$v-dy3+?)6dl`_~lJjXnBG+k~0W6RF`;M*UlV
zceIkDRw&>C6~4m^-@oREM!7OHjirX8tv*nEzvxegd|e!Vx=sw+N%!q*#^mH!dy+QR
zuJc^c2}7)%?o}|XBLKLitA+-EDi#<<r!!uwvNM%G|9Y1Ha$8_kR;+YH$we}Lj|o=9
z$e<*GV0wD>juorgvInlt3?8z6^JuBav^L}lB*mH?xgx~7pY}L>|H;qeiAs7hGW5~j
zt9QOXQg1RtN4fVk-698r&Rnw>INJLV5Bl4n2Sl(Mt}Vx&NE=c8{7T<0@EY&yYc%p`
z=k8xbFF%Ir$A=nf)B=rT2F3R`{@a&~{OW)H#IJ+HA8+;lrx&I8cyk4u?3pWYf+l*L
z^Z#ES@^|wrrI{?6|Fi%UtUnfnCT?AUjw!EZU=#EpDupxreqnrH55AlH%BjZzA+vyk
zA<;emt#1#Y+xn1{-_wxIIQZw2*28ZLtyGE^6w|+6r6FF75G$-FGm%pFhr_AptJp&U
zX}_U_<NGdaNm-f&VSOCNuDiJKEv?Ph1Ln6&^V^&HVqgq(sv!t<#tQDiOOcs~|MNSj
zM8Rqe|1`)h^N-8)`=7h0SeaAd^6>JXj@p50*pm3<76UbYxgY+=(5>#E|1>t$^?~hA
z?<oueDHKYx`~lN{T<KT$sOSSh*^;c}-^c#+p7?~gdnK!*9Y6f)lK+}uU+?<05gZ?D
zP2ov@Ix+ueDE{*m{$eZmpP~4(Irx8NReYk{d2~H0Sw*sR6Z|`&s-cptY<&H{0M!@1
AdjJ3c

literal 0
HcmV?d00001

diff --git a/_static/IT_sample7.png b/_static/IT_sample7.png
new file mode 100644
index 0000000000000000000000000000000000000000..6cafa761ddb478f1e70a584c3fb9f1c3c01f8bb5
GIT binary patch
literal 578291
zcmeFYbzBth*FTJ+fJ%rU4XdJrG)TvSfJh5ScQ;EnyCNywNViCLcQ*@2NOyO$unW)l
z`F`(sJ->hM+vm^cx38I<YiH)nHCLSXIp=+y{q{*t0_QQ=V>C209LbMjifCw$kZ5Qa
zwpbXznO9;&9%yKf3(Q4DKS_#;zW8JhF)_C?Mnn7fEjAwWlae~|mqQPh6yLa~K`|pq
zFH+DH-<xGfW0Jpo^i(R)JKaEnNY$1pF!0N(t^B~RmiLT`0uA&PN~dtj1L<}R+Wh<P
zzPzlxf#2PorX_G}dTge69MriScsWp^5gHe(CEsH}kD%|56ua`^SXm(;xw`iSjWhlJ
zN3@=G<6!mW<u<ff`WqKF7^d**#g7TsVi@dBSKN=KcjF$KUsiG_e8V`?lVXBcpzJj!
zni2O7w^lMW)uwZd`Ljyln8Q$q{?~Y+4r31tTg7BPKOi?ILwmp;&mej4%b+kSL`&y!
zV!)nXd1xRDVFu%eX=)CR;P4atn4uApaS19oT;`ftzJJl4rq98?j@~|1YryxZr|<L%
z_QBYl1OIJ}k;%v04OWrZ<hPFkg^S+Sn_iApd|hLddW@4~@Dlb4`c*0C-l)>xfC;S(
zTiPzcYfutCnLt<HdvKV5JT9$8(8KBGNF~O;pa^&@qkhGY?+gj4s)<xgm${F5PLOVS
z3S~;k3?YwFR`$+%)3-^g=U5^?^zIS|BAaqI&!bBOIc^%xAM>7+@#ZPqJE?NW1um)m
z;;mR+t2%xt)clKW^u*OJtj3es?nAABPQ)_HrRnB}{+^9=OnY~$T2H)Rwcg80*uP&q
zF-m`X5Bq6|5CScNS}lsB`Kfc9Yo6HmsKgGv3~rg$r9jt74pL_>=L@uu#i_p4qc|Zu
zyqDJ#=8GE7&<--o%F2{}HK)VTD5fyKyiU4ExIjw}OuZS<bXAl*$cC&i>yJXdJWIbP
zoBjmbo3!-4bL*qpdn`*&?u(EF-app6x8(gf;+sk{2_vSZ@Jl+3q&BWk_ww7QY|$t%
zBt$rP9uE6RFOk*Wv+)TK#<*^iX2o9cQNa3o^z>oo`@{$;BW(5e&QA$WzIbNn(cKRU
zc$Pu<@quK<`iRC)lDFP()5(6~UO(h}loC)ZJU`;~lbFnRK{!qS0Zqp{HB2sV(>9s!
z4>8*mjy!oD>O9zNdwBGrmZI1XuTA#I?TXBkJg$vp8UMkH$5@o35`|A`+CQ{MJa|nS
z`GO(xRW^2>I9Y^BHhSH+-iViL1Q-Ds1|%d5A>=e@1fk3NMEWWQTzYR=F<E85I}ggr
zu$YjX;z+ky7&4U=tKo9tCVkZoVd{+2ztor4V=;gj-jp5M@la5j203&;Tw1X+uQjg4
zsZFW<FvqLKU5v;5wP%&{IL_6@lO#15-bTFQvV^w`U-39XU4=h*_?4<HY)Rtdi$E-d
zFMGR4JL!8o{V%LipJ~Vlw7=<#3E9!x5!kVm#k!{amZx}=7a;?T3!NgFIZj^kxN!YP
zcQN;Z^mM)SrA$@{k)Nd+8CuJM?Z$pOe)2}j<qs2NQzhb}xT7wk;gN^Yi;8rF1VO}a
zyrV>?Kjvj=r3++ue$IZY!C+5_2r3|Cc%DS)L8wcQ9~%;@CiNudB&I(mj9y!nOp)eo
z5?w`XT|YxCL#+JUMP(|5jQrpEzIjIZR-br(+KaE(z15izn^27{(MYeCIR9z^dY(b1
zIQ^N*iOGrO{8d$&e^$a6Pvtuewtd#3V6MU+`5Wm(S=C}iiu;v`Az>r3TNLgXUIZvQ
z6fUZwK8#OVJH1(_Nu)^v<<pIg!-C`eA@W03w{<8y5nEPw$Z?1%aVRmoidBP1!=aMM
za(}G;C+9mKt`2Jq$cNF8EDkEOjv1mE{F#BuOEVXX@U5+#tKC2+=Pj)f(y^Rf_w66s
zaXYD_l;c@DZd;GerOt<Un)H6MsfA#F;LS>5D=aAjrL#<wj2@1GI9WLZxq7+H-nm+R
zWH%zCA&(=gCdDP+CiNp>;?C!G*OPf?LY7Mg;XYu!)ibboW?4R*J-t?!=TG1B4qmx%
z7D?JQ_{%C%y=a=rA({7=z`SQJ0_Vx^_R&`pnPV&UZ>+YB=qu;Urv(Z4e;NNUGua<(
zP%&bNQ!Z9+QZC_~=ljjKSU-I|!I-7~sG@ps=ll6k_b5$=Ny(6?rt+-ttmmxCEVna}
zvp|DXgXRUH2lz($g6X1!l9BQOZUe;uuOg)=#T0KfFVxB2If{pq2ZxWGmzGbJchPw4
z`|=>p-~h84F(h7@Iaf=lj<0T`mdXh=Fx1a9zn?yPI{l;KVa4U}fdjYgX8ly{fB*^A
zH>w6fqtxqEEw~0eSI1w+v&2gWozX#sZ1?GIY)xoQ=Z)_rW;KJy!<%HhmDNL~wO{K6
zJLLPC%Qv0yR+g@k1<3EkQCPxV9h#N!r`EVH4HT5P;YvNExQ@75_`j*daD%Y(vE0e5
z@EIT3;CPGP7p?Xec+83W+>a~3n^o@jWNFXxkLBQYSc*>)fl#jfa0S+E;5t5K<an=N
zBqsHugqwtQkUNPr5mTs5utHD*u@dp=%Sqbm*D6HZ3^w9(Rjx&@`kvDFg<jI=>$dl5
zE!NF$T)-WYRe27FJ%1sjH+0q0t6mtkZTm?*se2Os#(=<(0FMT`>~CjTM=hD0ou92w
zl~j{*8FkpjvX1w7?%|uS-bR;eQEQv@2lSds7)pXl1BwB@R{DNNP8z}*PBQ^#rDv#Z
zL(($R&;;MX&cWzc$_n*A1*N$yB~@yED{9fjG1+TZT91~#{#edJ-k+K_A0;YEo+X!;
zaE56PDVDG?KIrLU3l3*eP=BQ{RyS_2eVv2u26Y3obTF<mS!x!ycn+N`JTH9C$&e8%
z$uO%zTKraRqGbkl?#-Faxors@sOe+rgVk_3TU&a|we5Uu1|eZTSw1ty5H4tMRu^AO
z42HLZ8YX8;7Rzom@msbm%4F@-kK=I~aa3XDs61qDS!f@50~fFLOciO3dL?3*2+o8l
z^;zmf>blzB%KEKw3{h<JY_;udEPFut&VKg(*nauR(n%cU2Y#-@E5%3+yN0ob@-nFg
z3$KZQ6l98EL!&e2=iSf5s;~v7Ij@rLmLGYO+0|BfHu49lrXGeD&d5PzolV+g^NN@5
zW$?uSt{8<`i9>M#|4%PlB|$BgB9Oql!;#Pqp#V*?LNW`o?a(fXR{ZTq;+EI9_M2uS
zqt%9khB!T;;i@U=sdN@!6XUpWxO2G%G*&e0<uOvNES)R`j|<LAZsfPq@&$^V-G7me
zC4A*DG(8;-OD0Go=S#D1F0t01?Ra8GY*Pg{KddKlJU)5TB2CPvMrq2@pey9mb>81s
zFjz1&m0Zl}#OIuPHFR7<pe?5@JC#yWGiO^jIwRO{BNJ{4M<fYxv&^r|)jRW_IbWu1
z@n0ScH!w8xv|J*g-KZ%%dr*jw77FX;=T_OtYB}wUyakN=u1R6zHUIqXuUW*qlkE;6
zoy7rg`Q6o+?or4|p_MQs{h)6CRLKYG!>u#uHF~Frw0Hb@Bp(Yd&`D_GyN#dHP3f@f
zK$W~H=~=LHEkHQm@dvnAQ!i7?FPJuiP@m2ePFKC<2jwZ#48eFI%(q|f6mKVxmb*3S
zknnEyp6*VqM=qJqg+KX4N8fX;0HH|<ppBE8-Lblqr=>nMI|1o^ZH2A$rFy(iMl%y0
za@Fh<!feI=EUS`Lr$7n$mf47A1!0-WTX4yg2N6?T_mt0?CLcgTsl8#nMtFDH^LLec
z0tM(-xhP_kD#n-k$@{8+_ZnoZCTSuoi}n^MW1&5`M~3zgDBS~gp?l>2S^jX34h{XE
z^Y_uv0?g4K{QDj`;P~eg0qlQd{&hr;@<+o2emw(r*Yx}UarYx+I{JT<F>HZrXd+6Y
zl9Iqt$<W@|*v7#O;<z*E#Ri;sWcyLw0S%4l&7b|Aq~hx%;Q7DJKdU*a$;$8<Ladqf
zjUWcb%&yk|JO@p{l@BOd8$0U1aJ9Cwao}?meD%*Ad_einX_i+n{<+1`Qt*|U?57u^
z5PRbnT+FP@tgnO~zj*ONz~0D&Pf_f{zr}%Hg0IXR9c}qoSX^9Om|ZxSA@-&$Y`na@
zEUfG-?Ceay9ZU{xHjetPOg0YG|9Z%OJV(sf!O-5^*3led^Wx8Q^$j3Sj)JdV{i*0b
zKmU49V^{P4s>#OT-^~IV$nxh33mY>l%YQx_5Eb}ymhY3ftFe{3n7K93XFwf7Y+Ubn
z1pX2DKQ8@OmH$mt?Y~6XIXSuhyXb$r^#2xBb}+UVg;)cXItu;Qfc;zezhC^fpa9FC
zw*R-j_*X~&a~9}np~nI&|2b+xkDo<3jR6BsVJ;^B88`yd?9T_C2l%4<*AXZ`2xQm1
z5L7@z6GoF16Z!0VZ!ZON?lmW($HDf@`D~a{);;!BQ_vWGM3`fYiKqghxzsV9)N}Et
zgil38#JZw!A7hP0+Ec6x^CU{eqMM(>xvv?gr>DJKoXZazJPuhd-5@5WQG?NtegVl;
z$qz;gyuA2N@BP*0ol3*{9s?$r0)8g<Kuq{=w)^);k$EoyzhM2%CJcHS(WVPBiV`FI
zy9WHJ9Q$u-2HyAoR{npx{eQ1F|G%AZ-Xd&Q0I%bs6=$5ubpO?hzu7v1S_AQ^=f;)f
zQ(h1G7#!6c)bG~E>$uZ@YYh_n8}tD%>)Nl?s$yuvreVMO^$KoK+oI>`&}J&^7V-K%
z!?V9a#e4W$#P=PBc}>e7o?>(@LC&^D*{^ov$KLRY{}n_=;kPb9Ryw~)u8<S@w3He4
z)YmLR`TxH)8R)`&E}*IZG+FnV<nwQT1;Aghunq6CUy*kwlw=406^5VT0T_HPuOq4W
zSAFfRFC&Zr>)<(!b^JT|+m|nVT5kIoB;WqcgnY#s@~C$SLSot+#r#+DBj5){M^j+8
zH<EuhH~iUv?!#FfR2};}`Rjl$*tgfTC9m`TYW!PSX@Kt2Pew@9|CPM(-5&#fw-)>Q
z`Cl1ul5U{;xI;K}C5it^ehYM8)Z{4MSe;&q`a=lax^AjRa4ROrb7=u~hk7v{qa?Te
z<GVr7!ZTa;s1Ht-?EQnxG~(;o!M4w<{&($l{MH&lf^0hJX9#Oyx5@A$vZs~^V<ZrR
zdib__95!ushf(ep*KLpmEHCDQQRaVut$(ZJ2_mb0;^EI|{g}73(fU8$O8&24v*YWZ
zIw@V|><+#ghCS-mt=`eZw^li?m{=MH-+9&yFgFmo?S8Y2!(F2cD^|Xi=i7em@>@S(
zqO`PraXnu5pnBRQX@)Z;`(=84tpDm7)xesX90oO9qsqbO*i%sSe+Rh!olA$)rfE(t
z3vRhLztV2khIuzBE;o`OL>rxR4)A?kp)0)_pnn>aRn)tNIH$DyZvC%c6-;V-9(UoQ
zcWJxKJN@h;J=?^Kxpmu2nQA&}f8u0cRI}iI@FU44zu^aH;?YkVWm6ssgYR?weR^xt
z!uxZ2y(3p4g~w6FfQwftq9;ZUT@(MGzO{SwpWf@DlM_N1T~3rhIh#N8-=Pj+hksym
z=1@nVuY6G3W54PkIt`gUYukY{?=+=76>Q)v`@?%IoNZHvV`%M&xFJ?Wqi@~;E_j@y
zZ%5_ODqNW)=Twm7NSjuB)0B+;K@NL-7{?zQVcNL6;CyJ*)qUdDat)ZrPx5**_#6%N
z^PkU&&!zvT?|kjs8bOXkt+T-FEQ{+*UKYpeLZ^rLb_@M3-Pc!wS(-^yKt^4Kp<FX&
zze$l*(YMBN>fM4Tz0{mpVC?yeIR|-QbT-eWxNw6gqG@F9OPa4O_SFm3c<tAEXAZ|{
z^ELCn$)z5jd+?HmXqo4xkYn?%m^Q{XZ7evl^u#*?5I92Po$z6B{f8Y#<+kR+|1=~>
zuff8uYg_ce6WWT?w5fal_OSWJMinRXF-T)&h5xW|wR+B8+69;tzMndVmP+Q&Dx407
z`49J4mpYm1)}n0oewW!aBQ~D21Ps8eberyOZ>FMx@n196uYazo7k54hs!1CFri56-
zpDFPSEQ$eR!$}Di<^@a_LyvUx{}4Leg8{7M8w|f(@PK=*Gw9morZ{Ub`7awCsdAx8
znl2I>fY+Y;sgN~yGQe_J-;`p_q~yLOh5_3*{xDrWp;LG=5du6RG19I$ejpPlY$hFG
z%Y3oUN33cKP%QQUcC<F6Hvr4gr_~nXQQuZ6mE0fx_KdJP{*er`v4_2+1WWw$KAJyv
zhRqvK=;{SB*&v91Wx&V1xcW|Xt|<ZplFH)vw;3dXJ~(5hg1Rl>(Z@B<;nlq+#Y_@j
zBtae<l*LO^J&T@DlH)c0-Uz)zUHX9EHp&`RSX&f4vPK&Gj)c74S|SMZ=+vi#s!$1B
zZ)MFN_keV1`~Asnwu+mtcV;--pVG2F=Z#MCxLSs<(Fh`|q_{7?WleAd&LZt}*03P(
z<vAl(1miqW_54|q6%n)0GgG03D`=z)kw3NFW=hUWSju7?(aL^y2w|7{g$4^b(kYm-
zu|c`!1j&R$VQ0r%jRfV}D}7Fl>{_!>CjR)uy!@hV?D-TX(h;crilbQ^Ugr2hv*Elz
z*)xtynp+}y0+q>Cr0lt|xW*ubJ~as1+^=?+pNk{sP2?p-a#CmQ?mu<)oQpn)!~{7>
zGoRwuTqy=u6Bw1BQ2Vs>z2%SnF7%4kW{xLdp=^+Spf)E?BforYivpRDWRaiu_y)_l
zMN`5N!x5XJzW>uUhVid>KCp%$NHA2$yDp=j!2=4WxT(@Qzdzf?I6E(MxE2*=!~T}s
zWlr@?;dx&XSd**S=313}M7BQE?lMSjlrEtdo5q`=xc1eS<E%}Cbxx0px#jo;EKo=p
z>PeC{{4cekp}pUf&9d1e5ONB5Yu345ZsQHLE)_+@*DRuMe9qPvnBt?Xwy@4Z+ZIz?
z7PkK=z<b40IJjlju7`4-11_-2nt%NVXaQ%|jT<g->N~N9%e(fp=4vGv(=g2Q!Ml%m
z0!LqN`sAU{RMI_`{fOrnD+Zs@WU7^g*R~voLTV2^H~DR%Wx4E1T5jua_DsRM%z(Ce
zezqy=B6@PU*A2bd-&}CtWpWu6C!Wr!U&ZDaT*Bp9Lc6?}g)}^2u6@xdNk*imch6<X
z8?%1DdX|~wsnFSosMGRZPp;58ea%*8a6JsU3OX<a-y~?;@Enx&P}u-x!K!xn;h20P
zOAFH81$92vV4CJN*>v1Z{&GG$ptPW*x!zC_LzM=V?Q<1GLEx2SQ*S<6q-S<|)Pcue
zeZLvde$m_X#-hqGse>#K>L;~soSMNmf8?GDDLrAiPjHjf(G-sJe6i-Z5(PnFOgNym
z{Novq2xxq-?{*4G>UlT5Q^y$Z^c^%D9tfR8hSxSm&4*U@bY)pl26sepRmbEoBhPZd
zPFCo$N$=g`bllBr&L=e+I49Mz#yagAFkpGS{~$16nk<v&W_YdPM4DFkGueO<m<k1f
zg=}98F%*`5oz)MZB%5)pek4Dk5@)_?wpqkD`HU5dwg~&S950bPaZ{X}ccK@WlX{K1
z^*C&x|G~(0azHPFDe^srQBH6<;Gh<gy4DQci=WSspb)?syPTv}Bt;${e53r(L2}bM
z9I~)aM$nHkz&2vvrMsVpm2BJoC`0#r9|D`DwClu=%|Zc2Z5?{#1A3tSG-bP=p~SIb
zn9p+EbwErQY%`_p;&gXC)zXehwq+7P<!+Wh=RXL^MTj&I!CsHb-TxCVMQmPpcPE&p
zl3s7``%;EMuj{nV-$R`=s+D>dBvED;a~U!c90}Km>ukwr3IY|+Y<{~HG47#vziI)Y
z=^^e#PkqKihqN0lAFB9QO(}7Yx+y_kXQM}9nr`c{^YppNQST|bMty}@wg_d#eyxkV
z7H()Ip*pWtQmrc5-mC_BQE1}{viV6zg6P*iQ2?k2dbahs1i^}tV$R5>O}h2QD`9@0
z<jMQ+)?>M~!$iwsyI*qr*7ag|_A>6oc<e{d!gu{{2R`+Uz_0DUF~+Il(@C_pIQoyH
zfIc$h^=${2AjgePyA=~)Y$njm%C-`2*FI#rHrK3G1?L>K4MZ&-brRn)=$TtV1fsoN
zPeLI#!Ej`!CDvx>OgZ4AaBCVxinBuqKVKF~voyW)72Nrm_4V!*uPc$=^rDiyU~Hri
z1Um{bY~y34b*edUvQV0{Ol9Y0{NkpaxH;mbDn2z>oyh3K=Su?jRzVK+;S@PK1`Kg`
zdzyA<l$+{qv1C|&Gr`5)vTief+PJL05gH<u3j-o-T#D0^qZ^Jigp5w|(JVQO3E<G$
zh7vQzYdcR5l3Pp^yS<_2-_DL^2jeReOK|spoJW6qdvzQR)9RTZhbV`XIUPCg5|w3}
zaFaL<C?1ANRfK!(8G19FN7a`CeyPa<U1hPHE#!2F8;<lXe2CLCZXQm3m>s6>I>P{I
zjr(Jft18{7HwN)8re+?@M0DWt7iKW4QS4L9GSTMora-T^WoI2vPX9Qi+ZWv&PS$Mu
z0cqbKS=WG!e!gKd)>N-fDS5;eNE<EW%6Zd+DNl5(?Xnp=J?sJHn>7#Lw}A0OO=f56
zv=?t$=PPm-T}xTROziMB&f^zgulV=2SWGTuYI97DDHy5p)G6$c`WZTIK#<aRvA?$%
zYsqk(ltwa)48UK805-e7QQuLt5h}iwm?n5R$8oc8**vu*K`f?M`}7E%J!8L2NM^!u
z5db<;6ycr7rx8-5gp6_)Bjp0@d!*OCrS1SSKyCW$mpo|3Z!8j9;fve}I*O1o6lz0y
zd9@mD%D-Q!G%`B6j>L2~{))%ng3>B>e_6ptKTE%MPzinKaq%f9D?tF|nQ~kx4?T(2
z_xxhlwH~L*JnOm|TJc;}i!B7iYH|g*$8Y2fV%hCZ`e*}O27sR_h<UN7fgf1AkIufz
zDX>IBaUS05P9hCB7|$`Yysq271R0o`Cfj|;y&`K@9t!uzauRtUmj2B9t-EXvru+wQ
z&baSf93GOow>OJ-sfch9pWET9h@r9;#72@ifnaahwzfcTtF%DzKek2`ygTrwTunzj
zo$y}aV8D2lrPbGn#&<m-qy#jM2Gr~rt*mHmtiw9O$6muIDP7WTqq2s*@&OAB5fXI7
z^H=N~>}Lhy##s{K#exIax3^b4V3Pn4)Rp!90N6+5{lRgM7*V$n4<B+C(&EL${3M_Z
zcgl?m<~$#EbI<_6x=UI_{#d4huJo}J9^Bou+%cS~8Xfh#1~NGWgEtL%d^;j+x=Bq7
z=R4q=;5&qEiPbfvcG;iI(1WPwl~0SRj@ujT+C;Ds-^Fjk4hIC}u<5cm`rEzP#`c){
zka4QirR;c|+mAX@&OXXDbQ~mk8VPMY<2$TQ_dA#xOnY#+ey6tyGaJto{dCmmVQkcO
zV>vCr8DHC^Sp6`kZknPzW%B7#5<+W3H@nnB<-vGb@N;s^%&jx>8ho=?VVY!Jt!@?^
zJoB+-%tAX#sVdfZwvO*5#<m25Ps`%n_L!BQa@;_ix_;iwA|Z8duSv37_YCBqsO-RE
zdX;5ysscmT1#IU)N4}}+tOD=nQ3H&Nda_g7Wrbh&U|Ndn^=L<w^<$rwKNQ$aC#_H<
z0Io6_$=9*j6%}xyhKz^lIHwF@55ds&)%1aPu451JNoXE0Hy*W*#7pzyWAR<{ZQ+XQ
z9}%0*w?AQ-Q^kO#to2083z4ebOf0Wv*LWSMy5cSyA5C$U7cTZI-_Y--!sYs%ADib-
zhBso1#)dUgM0Y@hNOHh^BxHPA3!^TF{?ZCGBiw*U=p21mg{ota!M0=7>;n~b<JiGN
zp3!4N#Y#@p+y%#ExQNqRRZg&CUxq8-?%Ybh@YCU#Q@3C&<A0M*d~V!QAVPA}3)giC
z5WX~+XsGqDQtFy}f{Jg8%Sbz4|BNVh6?-9{UV?r5l#06cH(lp2+g6r>+i@3h1;lGt
z(+|qJ6->#1uHmZkFjzIXJ#BHPhAYw%14aa`nRhPmtKvbbR5&**Uo|YEirc`;B#vjZ
z4*q>5jUv)kgxnV6KmCu&+CX?~Pw=DP2(`a)Cc*GR?dq1C4&?jWU8^gwg!#mF82}zg
z_09(h-DIv6WAR#~RDi<}rz#O%;~#t;RNH3=!^ChC&5bdBE$1+Cs>6Ln?UnZ%M`gZ!
ziGdl9<Uvvf)AsTlHj@DidShWTRrlGPb}Zj1q~46-CSeY0d7RidPCSFtAVGGDF~kW+
zPmW=`qRx*=4h(-@&PaJ2IYC4t^ZiTwAOudKq7!=PF#I;m4!X(>VCRpN)oGt?4N~ND
z(u^Ulp2BGn-RY8&kcI=#z3xw3Mn{)MG(=qm>O`^U&c!*bR~6S#Amn>E@5_}~&3nlD
zUKa>yeeP8(3NgbpTY$jm*;A4|?;Mj?c$giGDm$W>(QUdFiV?Thwsfl7c{s8m%DX)y
z9|Z+c`!PuXuckenmP<TgtLxFA2qteJ-rn;RgQwAj5^r9!ezulvfBiWqh^;aP0~YJ6
zpojm|d+mYZ7H6N|&M&M#c(m%Spqr|NeLJ=hW3Y?it+Ul~_7!_xuYgB9Jq;^FSPg6j
zHX1H5Sobt7JwiaE#t}JZy9Wqs@|4BL<p+tn6L3V1XbSh!M^h-@f>VauI>~672?YOE
zx=%7})ZY@)c<g^#Ai<mJeX4oI*Rt~E)aT1^Nh5Lb@nWO|)pi*F0na8*@EkzCY|os}
zyDZLYJC2H1EhfAAfC1)VYC^jL1UM`enlRM)zUlr}sPi1aNUS%C1Z$e6IL!fWcR0RD
z#4;eBH)|ZP{dMsnvDOG;qRC>J=eXPEAdoujr#5vmcVGdknqcuaA4Ja-JnwZmzp0{O
zo-+fHYrdW+;cJmn+n=ePwb?DMT?v|jTs7Wa@8V9@vTtANQa6x88q0`?X+8GJ0;s(<
zkQZIU3?>&G6o`G*nf!(g9M8F{<5kNk<fQsAIU_HmRIMQ{KV<xsjN?j;nA6Q+3yg`Y
zC8y-{gLc(oo_zl#Q)=4RdA&)d7F5Nlf1}e@qh0YIf7D_MNmP5V;Bi=l+UP8y*lyIf
zov$J^ufZyqWCMTOLGGjxY4a#b2;ClsPZcA)-)+{sE<s&%(_|FoYnz>BTk%TvJ;HA~
zF1tB!Y0;!*+7DUGpF=3CSWbCK{N4u;scO;rKDs4sU0it)<FDZa0Usw%29KepZ5^8%
zJL$o5g__gU*@mpU!=pHT*YG1;CmjwFwywY<=9QWn1E&QQ0pIiPZJYi`C8B1?Nu;R|
z_A<Cv2+PhM?7Jz*-+p(bAjf8m^4M*G56mga-_946jdvf+NyBZEAWC=8pIxi#X6uc|
zR{M1VER*yLutTrQW_e<&A0%wz9|_8r4RDb(P0rzqy{Ltop;cU{{asr-IeGwir3asO
zTjJ0aKERO!4o{1Y^6PpY+*J{L=Kov%Hu+sEvm81*H*3)SZULnjhTs`j|4sSh7ItH9
zO$t*ow)mv(p^<ipI!%DpFdk90saq$ix!x(5x*u)R5|wZ>7T>%!EO@E4S%WH?XW(O#
z%O1Fj0wcTX%_+FXOFo5pX}>)pUZkZsju|?=azjs{u33K~;qWpXS&@{}fmASM+V}mo
z%XrBs*U(eeyUdby;cQG7RFYzOA%mQ$;7Oz?+glS!uJ%OQY0U+n4jtH4ZdR@YwFs!T
zAA5$mnXZqVv>o)Y+cJgyLRKna94Cbjv~%+(%4GXFtZk@YxU@Ya_HFjlU4|oK7n8&Z
zjH-E#jE-N(;|F_+xy4<0hWX}#-`}kVfRXw_W<4!T`*9$F+=NDz$vypEzs29{DBF@0
zWT`DwIbZ~a04S;LKXf9`j%}D{Q9qZL^PPUgDFc!hf~{QuvTLq>ym-|5n7CK|CKy3h
z<J>Ri*E99?2c7DqkaHrg(Tng$pdNkWP)kJlXK7(agLU64Th>Z;eLLbVf*RV5#muTo
zb<d$0W{+yWRi01N)sC8TaXP^aa+vFx>eKJ;ly@=h)wGh6$mu9IScG1Aq!yH|tF84?
zr@}{fy!!|$zMx-Xb~r`hMh_TmavNITQmb4R=%V^<rox8$KM#pw(q8Nl8n+)>;de8k
zB!a6>MkS$7yVuHJD-hA?B5~q4NLSveO<PUKefLz6dj#hVJ25k!L~`<Pbevi}ftOLH
zWU8uhNgFmyF;RA>*s9R**bC}%tvGjmCkRL3`<D{oK{4znhY5%z)Pj>vowah;gG`@I
zrdq=xlMBRs2Jj7?Tx(g10Ku$Z)!^YLE9H$Zgw!0Yx+mK0-_fA2OAOk=1uq?;#}8b9
zqZ<287<yHWofc8VW_%ys6A7}F6Da{A;+|K;n?8cSieHj1_s0xnmRqwwg%mW>ML~WR
zJGDPke<Z;-)MY$#VCVbj@=AK05xP=w#j63eUp$D+jaytcIB4Nvwo#T0h_zc;qv$y;
z((-#U3Zr3(kBD6CCgiRCAfPgNWIDy$({<jO+HtxM@2|FNmtRPrpg$uEo;Bja6-Wb1
z({3+auI(v?zKM#r6Y611VyhJJD~S#M?&O)aYgZSJI<{hnc7r*GQ`<kqgw5MF{bb=V
z?VHO1rXO3t{9u=xsoOn{8_V1<uHc5~U4d|cs(ML^HQfT|@-5!DyG}zpb>5}kDw`P0
z6;4EGZmg+M&@ZL&h4o(s+oh#M%`$GqWC1zfTzLXO<+F==XS#~L5vH0KHS2~HVV)#(
zAX(?It2ZHC#s~{z&-B_A(_O6x?xC$l`bPoB8qFMalNy$F08vvEb~&jD*+<5OFDJ@!
zby6YUdbha-_C2vIP4hV5tyvA@uLr!RI-6%Tg)gz9$##osbda^S(5;LB|5}7d?)S|j
znU?wD4&hNPDe5@X#Vm`AfaV+p+d(j`HLKhz;+^L^X8`8yQjWC~&LF@2+dz*G4q<mU
zYOgnINNiXiv_~+2Ki*8kp_8aB32?q26Z^-VYyjk~R$r?OCc51LM9vC8zkHdo#(vJG
zXSKffaJ*l>!owJ+E|B(NLtCLHNsvCPNm%(#&4}$dgTrn~OLNB(mop}ntDaAWUQp#M
zNB(44=0z}4%Ot^g4?m;hDLMZ$48!R1;(vf$9YwTt&`BU)NRZ^tW;0h?Zwakx;p@<8
z_$B~Eqxq<qHr-G$lq5ia+_@g79gGtUKeoPm6&z~k0jh1dSr=EWwY;;<YPc)L^vrFq
z>x$Fr+&{0IpQF*VTI2T^;@=_hn|HZH9VBxw+lC7vYm=@99c(72P3)eq@O4|a>`vN@
z7gp3x>A1~XZ^8{-hPS~E@<Lo~O}8_<x6>_l<TmxG@SD-Hcw3>n_osE}&6V=3?<{M#
zSsIV8$q&{H;kJj!O-e{wd|jkY0{D6j@}TQ>B3{>tIN9M_!)-Ws1FYrvho;-aIzXpD
z7lyd+1YYLTcFWs>5p55f9fr`fHiRy_n&!WYgJYH_RCOE{`7a#7=_~xLw;=2D<6KJR
z*B@%T8_rT&77lHiFNj>GC@<?<=5HdB+~N}EGNl<;CKt`DYfg1$Jr2NUjV=zSxAGYG
ze;9~pZKgOydhA8^@Z#XsaW7b**Bt-Yy*uo0#Z<dJu9>QnW=2+d*dwED#*!cpnWb53
zd~XJ!{0(p7bQ)Q~ig66lQF1esHFMj=xAW)T)71@D>l@!4oEzp(zRxU$<FQ}qbwZjB
z4r*7!W`x7;u6aH6+B)H&+oOOu_R%)ajRey<fW{{hGFVQ&mjNvKEF4qK_K*u0Vh?#R
zJ9|D@t&zZsM^?8V04}Yk=d6CrDG*}SIS-Zmm^<<o-#6(ZIC)^W53ERIP$Dw|sO+H)
z&o*lWoqIP_199qUK#_#JLX2%M7E9*HY_fv0h?QD&pa|mt8>u-@wbB`j0>5y3Wc3k_
z!Z913`2_<5zfkcJ1A~9KXM{JN+z4;58kgATllvfIy=0q%t*mfdbT-eWlc|>5iFtrr
zS>^`u=^y!~;v{9ZJ-B9&*-M&r{3V;B%l2C4YSDeSXcz*b9Q&~Nu-G%^K}n;}Is|qz
z%j!0S$$|auqzr%kv9|GcPP`3C<tc+j03OzewP8&kP0|Br=Y1e-#{<2chJ=A`e)r53
zHyn`Z+5;}0N(5)~M~#n4{9*RuR41d8INWnfBbY=QW*Mv*{Q*x&e*amgV#2xoshb_l
z-qxco^ZAq9RVi8)$Jr^HVK*WBJJMr$L>7f&@l|QanPSVbyt=Qog-?qo^pnnr*GhPb
z%!mGm?kROK2z~2NU4;ci5y$)6$c08~2&=sf7l{?h7oa)mD-*WLJsl-dBQVLWN=2@o
zhL)#54o~3uCqNi~>vHuX)ZO`V-lYLh$NnaOlN1ThmKljp_jFXkSS!&eR*bRU;lu7-
z1%hv$#$Xd5HvN=}XzeX|eV_vnZvOR@r@5&s0oB0*8}P%%C1iDAM#}Ph=V5Ldl&=xK
zQkpk&D>sw)dEYtz`d+-;82;&6V{~<z?b_%g+T^F#_tq%AOoZ&4cJd3_CGyueK#XH1
zney)nHalb6of2_9s@I>ikcwtDvsTBuG)j!FD)M<aPjC?=Sly2f*H0J}$rUU44sh*r
ziqcHa9l=6EHnAYl9P%!if=U7+`P}PrKlu!$v-{H0^09`w)?n)ghLb#J*gJ%Wjx&$0
zMb4y&bX@go0Menp>2iVp%3Os@sj(wJj^PaNTDR8#*Ns%wilzopku~80b(xvWHoGBp
zNVl%Nn6d2e`0nx#?cJS1`fGEQ<0QPg=_lSqo{d=QX(YK@TLtQX+6lByM*GLC(x3Qj
zI$qzVhvvZjJ3ji{J~v+_+ZKb1%NtRUQJ<K&zhX4ILu`fDi`@A&qhIr%5^VLK;qzWo
z=C+@@pSQ@VN2PxDnvwgRJ$9`4BP-g41jW3=8HQl#C-hRJ2zH`y7-Y*x8_2{sb0FRP
z4Y0+cu*ZV;;xp%~otc5)n0F!^t+pLT-~EigC@U^EP}i8zj&9d2q?!19-5KXraeVZ<
z_{1sU;fmb(V17A)kZ~H9--0eD?#y7mVP?|{(V)it&$@EzG>X}xc7eiESXUT^AGN3V
zU!(MnVWx{6*L=|G8IYH0-b<VXz2des?yCaCX{>-%uJjcWbX=zV6LiKhmm2!9-{yP^
zCyXq$!vO|_#8XBVPUWrlpS<&hL$-JMadDuo7z25Ur!2Ep04HO0w29+4D9v*nX0{O<
zW%Y+JY(}oLm_pH74xkBVH{X){HxZAfG(2_KhCFHO+rmM5d5AIsFz?I~WI|cBYIgZM
z5X6Y{wc;6_6XY`8A$|2y_MrBqIk1(}=pb}~I>dgtQ}B1iQu4w2_2(m{#wCZHCP+8K
z;Jt6OsaTpj?&367mx-HlDM#p9I7^DzDIbA2rad~&vtbRBp*?Q>Xg%Dz%n)p~D^31s
zG8W(_c@mNl`(2U*v!uklE%uh$iO0J*+&hgwEEv93%J8ewcD_JvO~y6X2zpvq&q)m!
z{erV<R}Y>@o6PQ&igVfLM@~zq>eA01MhLP)BI1}>-Gy5GwgdNq13=%K&Ta{0YaR)m
zbe8yC?v4MFOJS;LSDZe$*^98Qn*TBVRj4ucI964+>?P72q{~m<)cbWB`DB>j#G`&Y
z7pJ9;=<em_6>Q<x(}ilsrqi;d@;7nn0=@}O7YZ+nX?2^x4SR`CzG4HcvU2t0b%E)w
zO`}SL>0ss)WBEpVSRiDtDBc*P?p|iKfY`)chPs@-ae~)vFBFt_#cRCl6tt*TXK$(3
zp7)COUVU;X6z6r<Lu*8xOO-Grv}u*Vtv4ui)74Zye{t&=fk)&3=Qnj7u8U7rAeMhi
z)LVAb%s+XYWYCYFQ|*>|G_zRFDabO$2UuYfO{<$et4ow~|KzmUkkt=$lfm9x2-Rim
z9>S%2D#zP2mr=F*0uH+|q)VE_EpF+X!GUwOAPoDNyo&16fD|l!%f%9aWfgJBX1a_n
z(ND62d^A{i91gXo`9m<4Z~sVQMi$j=BqpRq^dWb0^mpz119f>>vzmta&*;H+PumfA
zj3+f6)!TEMcq)sCVHYxFcKs42Lj_>;xK4cr6SxU`8<$l_UN+tPyg{4vwYZ1tAiJiw
zOfZDo3a|WHxz?u<D$ILIAd$e`*w>h$w-E~B0P8>BDdgsQtl8XS*K|6}pKwR_Mqog`
z-DuXjMms_uSQG9?t{HC79<W1Rq#>SHt?Ee*qr^NfBEvmjuhx}bbjQ79>J2U4!Cc@t
z31)iOrv(0)=+@EM?oxBN^~U$j;z(rEvBJj!U3xy`hV~}r6*}?3&9|Ta?_8-9TZ7--
z^yRgFJ&KV&=^&3WSukS})eWLVl=i<ZFoK@YXZs+<aKTvWrx>Pp8K0QEpydpw-5R8`
z-`6Rx3u8s3!*$gK4vUVs-EpqNxrH7h4sit0dw{h=Ug|r^q_nkOqy(pkN=v)0s6gj3
zPQQhntKk4P?iszFNWywJpG*^zJ?jZUj5n(BbU0}gzQ(<7=TB;awLUL@9|I%iI3^5v
zRG=JiN$-Xu3i+aRg>|NRpmZC-U<`F#?SEQxhsD$>kN9>vF}y8tx2t>iHYPlH?cs@=
z;X-;19LV+{^>PF+s~60UyFeJ{-h~lz_ri95aM$Cke_341yW4uLl{@H`wN4mrK_)R_
zADd}KoqYG@Z4ArKy?p`-q||DL7n@h8%O{Qiipso7hpSxdk;%y1HKq`A2_E;=Y=1Mw
zN|e})q-J#fyqq$aPgVlE>Q{9dPa~T{FL(wILXzqxni`S22KEx(d364ZoXQmHglNB1
zWt5IReY;({sfb%-o#}ZxT)KFC?6G*0`cTqYMxACPi{#f40qP~#mDV+D;D~uOrIY{n
z;pFH$u)iWVd%E)Mqm{f&76K~L{2juk)~L#C+muRFW6cOS58No~rkegsp=JbTeoCLa
z1=>2%y8NIt4KU#IFYJR{)Pn5Saq28kcw@$1PwygSd<KV@v&}5|+yc5z(mU9pGR^57
zH+Yuq9=~Dlh{k$u5-#pj4ju+Q!z@a1$tTwata*kmI(UQJEP92DO9=F{Nj4IsyZ|~|
zx0Z=44;AVKp){@yL%{80C3FD$PhzomLJKu$o>YAaZjAKYivNY@jen?c(oSx(?sGAI
z_?A)fV~~(_uCU^^Yu?!CHh&1414DGyhtS9C;X-sMr+)lvG();Ao65E`<DWiUxp6=z
z=S|5lpJ)9lPr&QpdEhqPr)Ch#XK~2f{jwnB4X&|ZxcWtA^W1P;bKZ{GVCIfnkc%Cy
zyNPlglV`q=_2Z@%3yP)@nB+^Jb=e8S+-v)Mb_HdqAF;O$!P1k$mrkPQR$t%CU=~b+
zE>TkzI0X7=43=19gG=Fk8P3V{mt}=1B*@X1p{u2*0Q1#C$z}oV&xXD50h4^EIL6PR
zJ8V9QEqQm9qeLK2mCN<9f8f(u108Io($Rg`*Ce;tYJP<mU}{a~+eunsehH`*;ufrE
zS!^2ih+qZu4+P@f+4JwN3N(6HvFYEkIR2KDE&$)<IBl;eRwrk_rKT|>^#Latg(0q3
zdYJNU8S{fvDr3b|exwu-7(9hK4lKd3v#PJpOfo2jSy#`d)rmb@h9fN6E%A&BN-FYo
z!E?qicRj0l!6$cD1DD3ohTm3w#^9*=n{g%_C^K>>+~=NFWLsL7-KHTo2I+9iUZMZd
zM(+Ho3tWJv(`nqLB~Er72StSHZJ!1~8XAcX#Z^<Itt+RmIeoyvm*F+qMD6Q1w|hc8
z5YJ5pS)5zeK!J-j2nG>9bVF&nHvS-AFuHW^Zu&X6pOq|IEA2G+*Y3frgJO9pk7<(O
zWHruh>K#05+GN%$N|s~)-88WD93|_y8_CCe2wmB*Ldh4dV&;7?>PfjS`r)6hJP_MA
zjKA7j%hbwR)%%+j*Kl*5^(t$I<6diLoUT6KDFa@B2!Bh}<`<^>e6PAWNF-THpYiie
zc6uvtV<;o=SO-~rG{t=!oztkJ34l&tCB5BUcu!3#l-U;OD{#snxm=aAnni$Q_a5qe
z?(azQIN`Kl(yy;mRo82R`Jr}%r1(PgR<N7O7G`d;y%cL$=g%ColQJT8KLVqQ^1cVe
zUf&u;P4zPa`$k_;SN#S@i*5Os$cWA$A5*kFG4FI8f~}Ka_K|#Ue}RXNc~=D!)j;Td
zW$$P$*PX~7`Pf=QNPJ*e`D-u(cB6d%3AovUsJmp2h$bLuKD>mQ_V9dkp>++&CkTW1
z-A)4;8N_3O-t>dPY6g{l3Zx){Ibem)$?nKU=&3*9min6XLS#2}7|*L1hJIFe(>>(l
z{g;r4NWQq4C3vd1x(H@4Wpu+D=e{qmQ1+|Ema8t7nVh&Dc7%u+iD}z4kZuGag;XS&
z97#;iS`-wKg*aQcqx*Q<@CiM=o<D@fBc6+rQ?06wVnsI~!llRrRB9tKwUn7=1Leuz
zCw&vrerGrITFBa#!3W$vdf8F>`;crz_~ELokYiRD=grTa`%YsQn!k}{GShMsHmXG+
z#Ji&g89o3~1)`CesfIC2s&UJTLu_Sc)+VC?Se)U{<rH(MiP}oUh(fAXEq&}#Gi8pN
zX~0)kq2*-!E%)uepKYR)J_ndvx{Gj2jmmwVh%W#TvYmbSP7bbAi6Ti=WOfTk58gPp
z!EQ5`c3nKR`%qZ$!sQ=aj5Yl|yyht+gXCA%RQj<Wz~t|^_7H?|^Zddf^1jEl8A)-Z
zDWmPDHN~wDAsL8?!QjlpEsbg9$K%^{tM+f@yQ(4mta3q=pq;kFlleh?ijNd$g68}N
z5_c0b=Q><h(rxT=F0@`VJf69md5!`e(A@2xC%C5!wgD%dYSG)q;~qjfS;-loIUw0j
zL@XqLg!^rNfby(0RmA&}1d5%2*nrPHa^26ACPrfnzokru6H*ju@4xY?+v8^xr<84D
z2T4pjW{+^3vOAtMNd589(_V!ED;=L>Gol+Uxn)Sb5_k$?zBwC5f$>SC?sovg58))B
zL468KLb0*#*Q{A=rN?qUFgH2k)^GQ%J^{z#2QO;WvKB$753AQ$<&DR{5p5ub8p31K
zs<1}GvGU$SATqp6oEpcFtP0djCR~?~Og1hltC!C$E__K0m+Z1(-6gXBm0|o8LllF9
z%LvW?kvxX^$y|f^k!n>KV*eA_nmAOYQ!m!f1U4?&vQN$<Rad_LTA7sX<p2p%eW4ri
z4PS3BXa`7VOC9VdanUFA)1SnGG@eK6rmCC%HpWnb)x)gla_=sE44)iU+^0H(v%pr<
zOWRZYg2VqT<L@c)TPu)I!RSNh2A3dG5Z~U%9^oyXkYAhK>?)jB4TxLt1368b%f2}H
zO&eSe*Lp&pnQ0>>$}guxCV>aQF5eT|xjYFjSxg4f;^vFD(~eN5UjCj%wXG71I`(N^
z16RJyxe5>Tq{y~FhhIftovT5fkQA4JrSRtmYR~UyVPYh(whqYKbnQRiBuN<+>0)IB
zD-hjunT1?xa8Q4b@o_wMC`gWWzM-aSk+u7Y2b~?gm&KWdH#YeAdC_k*BhN1QjdyIG
zCa+3Mx>x)6Qu1M%=d3H8B!vm0mnI{r-|AGtM3mNYWZ2rq!wkppP?(0zCb))dEpL?7
zaMGtSKNgO(TYMS1Qeuu|(vfYPKHYO~KxdDk43S#u?bPJB1QJfcy}}E`q)4{oAUorV
zch6Y14HX|`)i;WO5MpzMBfkgP>?E3cbZfA0S(l971pJPKZ^7(%JJfJ^6N?s}1#C9X
z10=jMts7t-WBHOu=gYr}jY`?lh;oP7X^JBY13RR6=I!bGl_$z&XMmKa=1;I@#1Mo_
zdpT+Xr1qw&w=FdGGuvY;<G~?4ZvWY>lOiozYgolqf6P8W1x#Ek#Kf_0f1J2TiVRy!
zEdQ82vN1Be7CQ7q(8ZQbnQuiRm%aVls@0}|c!F?GR@ied;V;;K5~Iv6gF{^ZQ~i~;
z00*f&g?%b4E#DdpFO}R@ac8s!W6+ai3jz6uPMOMH0XcN4PWMrV64R3Vo)mdAiR{*3
z_ZQae49wN`o^-NzUzB@i?N~S9TKx6+?&gshkzEQGYFuE&M)048cTJ;y_Ga81i70UN
zSS+2Ls?oEXnB5{ugYd5VnVa>NF;~BTHD)yZ?}|^m{qq`^!w0OQ8a5veWT}0+JOvf8
zQMK28UQo8K`Nr}mBp(kKn9XnDCzX~_mQV_N<E#kIFRwTPsOxy(eSc<kQ?Hj<@hNU4
z(65z@<;rGUs)nMYGIqY3vGhi~VoB;!_@GM@_SQHv0lS|0ev_D<8aXl&-*dhD^;3c%
z$B5wD@4YL#7r$fKT&cB6NRZl26Uu((*{izUv66SdpB}iDBTaQ*6AWl>qK^3&m|3bs
z6#2E(p0~waL!6D?w5vJO*d!sUcK3^i6<!;vi!!f#?6WD|Aws8%c@p=6yBVAA<PE<r
zvKSbRs$J)nBn+7HE|9N5W9o_AmWOG?48@iJAu;^`R(m|#m0>2UNjookepz(eVzl1}
zdE<}<)}VIQU?ztl;kPg&W4>x<LF8-z(Y2fN81*zjcP0<<oQ6#InR@Y)--S1z@@-ek
zzd^u>f_G`hZCFpi)qXZYP-yNud!U0uuxXnGi$86=BF!g5dJRM7pX6fFE-aiR$lu*$
zSgj6m_fokZguRE+tE=%1#5}6r*LJIVo>)9*YInOHx%+jwHIf$E#7YS`8-hJyR->L5
zaQQyy?W_Ywf3!Ae>o=1Rc;*#$+6cM&^W0eP$SO^%Z7IzQfG}HN<`g`^N7H=8?iS}l
zkGXwf7mo<FDrD^rf)?@(@cW0et3~39z8!PF-p<7%$szU9Gkfp3&!f^TTC>9Ixj(lP
z{*A7BeBZx9@?kX>Y43h%2X0$$q1mh6c6PJ=7Z?QB`+PL23E39US?of|xtU{Bt7qx#
zJ!3@Os+qhEz@0<+#3bUJ9CJaaSvNrCS9`+o)*p&ytKfEspV{`SEG*%P3ee1k310k;
zTlX&is{h0e3FMcRcu;7%Dimv2D@mhaGhr<PtvkU&O;&%;3eyki<5*QDnO;Z`2)_B?
z^r#<Y#tmjvn~<d9xXOae4JyQMKXLk{07d8pmJb9Ok3S5-S9=~{IUko{Q75HPy)4aZ
zu*=fVQ_y#E{B%2aK1b(M%TdQ^5XCKhycIf6#HwL)X7y{_Dalz-KsbEh-|CM)0NFGi
zi_G6QeUYpjki@%#XKh)df+H6jrgr=J1CL$uesV4y!Io=C9;d^WTH}+TvjUxwP{m3D
z<@0qyK5>>(D;11>osyZ0QvsSBHZki9g}^@FV?uYy#y>#;5C<8vclVS>&)BIAD^+nC
z&p^+FXV$!(k-`CU!2h&5hUwhu5Gn7u$p9*q@DPvVY5<PjHMphO+L*w9Ip@&fC8TAz
z&WVjF+w%S{8=ibv^Dfx(XK}6`>A-?y8W-;3w%c1uc<Q1gUw(53fD_Gbd#nrVD0onQ
z89$H}pG1aqnz7MD(w`;vWk5;7V|vOBzX1_Te_5y8j>KF4KZ-s<pTDJ&$3xVm#6?Me
zHu#FSU#H;Tx}F7OPthCEB|(gl@uTtIYNQhAsN<1At)K8oCNW^xdVtk^6@_YH>$Nbs
zx*QFMRx*e8lQiO&s$$<3M*a|edRJzzLkeV>ppzm;`;5`(;sQ=HpX*&E(jSJG9;>?K
z;Y9V@ny*haH?kw9`J0pe81(XN18y~^V9!4c_-_9LCm_|Ar8q+y5`l9ItnHX+YH_5G
z<EZ`<sKmvwW57H`EYEww3SCL~Ns&1mSDJRQ>l$RA%KR(<y~HlI@!CqXI?f@&DJWTX
zS$McHEOcEK3$Q@bFG=nRUBErci5EQ@w>cEa>Jr)(e?bI(pSg*DFi^(1oxHi4nT{G@
z)Y(Z8x6d}xKd!Mn3h-WY{3TCbeNt6g6qZG-&N^!nF!bbCNSp42{lw1n{=)8$Hqbsk
zLwL0Y4x>%Qcg7yIulFLoDM0T>E}n1W0Sbx{l{M`}!$R_D(I{HBTp{a&pCs;BH>msd
zpZ_{Uw3xl;1Lcx>{hq8s+=#cRqp(G#L$)xklK>v6<1z8REsMKKfhAX7tKAEioxEHO
znj)Gll6S`GKeYoga0^zd3T7+njc&Ecz?NOlRj~`E)^S*@?{l{V*-%t(HO}87RpH2P
ztT;IqK8O&-eFQxvFeMp#hG`5*XK#D#Ug-W+=NW&<qbXe&>Pm5$mjEtMpa~$ff|0+y
zRZ9QqPu&<(M+k*iC6HfN@~JyBcz+|nuZXYD5Cc{iI6rv&Ql1;<cI&QUg|0h021e6d
zShNzz#;R3MFZvJHK@OwAS63M2NtZS(Z6%r{G1J@AFfx;K)l@H>ma@GFaSjKFg`pK@
zrD~<BS>;^`1&tY5iT>&|?;~VTrrY#S1|aH^P6$)m_%Dq_-*(+%Z^ZjQ)iHOlYByJE
z7S8cKW}}3%N6N(N#*(=U@LucK-Bt_%<IBUi30Pm}S!ilse=mAbdhq|l*;_|N*|vNC
zw}OPUC?PQj3ew#O2vSliQqnCg-Gg*WNVhc7Ie?UOH%NnY*U&J)?;P*v-uv0_-tSub
z{jD{BuoPzIx~}s)kNO^;qe+D&#)=E5dcMy?j1MNZeCk0Vv^cvNQaEg~nFMdlVht^B
zho$tH&pOUnF17}%xNt<Dy(<eHeFIY@3n;Qpecj1PDj*{@?xgnGr%7SPQr7c+gMZCs
zEtDk0ihPDbR!&_CeS}lSq|Vio?u6!1r`HJk!%JKBl#F1jm|%mqLPn!6$(2w3w_6Ve
zR{h(qr!zp3``%y>x?&OIzo;|+*O#J~mehWK8DY;KEHI*ZX6YrG=3t_DW8a^KFq&|a
zbxGaSGe+HbVc!HMDHF*P38uQkyDvJ8b<SIeA0yc53F?4{Z8NJp5D?P5p!=(+;dHtn
zUFcvbY)+R9p<HCp-%@d+&PZIXScF*JKM84uphQSmszUX((L5UrK$ZCs<DHwuSRJE<
z7t8TG-^9;)h0Dezr@Rf;@FYE<1S|ir06u3LCq26w=N2*lVFrrG)#)7P1~F|oBDGfO
zhZ=PqrsuNVEvYlb%?opiv*goLN>F!qjZHPTR_=0SYI$h?6Z^Ra&^0AlH6&U4Orc>J
zq&Oy9l8QL)Orq_h;8+oSn|t}gziX!EGp-A$TBG^U3(Rhe7IOI$bV(3vMVHNWQ9`SP
z?dvlzHxV=f?3X{K(S07#PETKG7<}Pw>G8QRQN6Z|y5Dp)w~rRuqhjC&1bIo9%z__Z
zd!h|kXw8NhY;TUM>NI>;y&4hHz6f*TAgm@K>hSoU6ZwjFnj%7vn{A{yt3@<yH{mde
z_gvh#cy!P-SME2Rjxecx?gD3o52i<#6)B>O0n6dsmrwyCqjZT5QRHdUio@%RH<gDD
ziuAjpOTYH2lOROjww@}ok|6ZB`g@`)iAj7oOlA3V8#3nE?zCrFfc=&qTkz0m1|Chb
zO1zATcErv$4?^d1P^?3h{G=;EfLNg@Sa{QE)pUJ|C<=o*jhynjy-6rf$)79!AQC(>
zVkPhnb}i12pUOeOt~l)kiy*Uz%aFs)TCPg`xpQysxrE?bWgP`%(W-ZBwR#i=;g63t
zrnaN%<Qy_u*5Y|r-})}PfO<+WVJvJ}X8c#>%;}MZilNc?F2jFrr|;M|U3izYK;CJ}
z7O?>O1byaBL*&1efV4vh%cEXCOQWkwn@I*_q#BhTW&a*{om}VJm2<G;+#z$OVrUVB
z){{4EH|$wXVJ6CELRQ7x-<B&3CEMWn@oKq%#G^TeN1YkYJvpsQc^M`j$d$vCFDvPf
zZ}M%9#crSZVJpKweJeFOHd@MmM~n7V;!NTjF$T7VkBG*jJDE-O;*rJ@B4{PcO`y-L
zcCD)GH}kx)5A!Fr;yF!apC&(|4|xhiEBZlulk8KEd(GD`ibPe_^!&s;Es8X_A_lny
z(#u}zCLC8A98?2o5O>%6O20UHB8_Rj@_eZspj6a(473XvF+iy{&iv4UR$#gOr%R(k
zN!{KL9OzQBB%>hxh?Nc4ToQ{3(H5{2a!bOFhyAb_j@sZbqh6WU3SCN-Ef9^6)tL-%
z>#CnX*!$LGqb9R07<()2%CLQGgg1Ruv2gPH;u$?H0DP)NnY+(*ajp{qYiKVxE6ZPa
zrkWB5#Fyk#<aYhBFAn-F{Y+&{S9so|d4AG#F7r!#%>8)ZD~w)!vSON<aAozFQz35m
z_TmNBQTNia^!&h!{p2ZtTgo1?9=ClY(mTM;usR5>w%a+q@41URGOSW6PiLuKar(ka
zQOy*y>4I6~fOF_K5p<d&tJPZFWM;TxYY)gl-jdXh|D_=;N+$^^d{+&+9w#;nE}YSq
ztK}IEpuJ7KOY51c`yeGhH6qJOp<~6zD)_Mn{jWmWW+@I7e#4&!-(w;lUz4hY8rc54
zyldeWjN5U4(?|<=oQ3XQ+7o7i%x*A`zrOOGS9C`L72`sEtnj<)4Psc2EVU7q){13&
z`Eje=9TgCK=mz>7LT215%$}TlLC~_Q8Ab5uf3s9t{y5W*ho$83)g%w+rb`M|=f9w=
zs%6w|#^>l3)11G`ATdQdTuIz7{5Yd=6gCL@l1K|Ikm6{ZWQ%tQpgXpxndtUYn7%O{
zr2jt*nCEl!7Ns1TwS*oFy0D6rqlZTNR1_sK&~dCEZ-X5(@*9)Ia;5vfo)KvD%;<1l
z*QVI2{8Ai;rLD8m_oFAEe0ixS?GJ7H6}9$fvoS`|2|EgWx3tMJGoQA+a$&JN)em9j
z`BrSku@+A6keSYPE_-rQT0Xvc-}AswAunNBoHg`zXewmw#nLD7afK{>(%ftd;5BNj
z3dMw(aM*th%im`+=Tx%ToH4Dl+$_nT$SQ=;5=t$tmMm!183fU9Z-~0m$^i?ulLgJ{
zsrc`9D^hHb>1*ar6@$n0Ls!pnKG*cyuVVh?7b+c|AQR*aGC>@+-6a^<kG!7np4S1h
z`Z$!21pD`X)r7YB?{|h4Eh<VboiI6~aEb(9E)z)^U*@eqcD=S&pZ!*a96>DawwrpP
zXG9O<Tn?8ZjvcEmsY)b8AcK)G*k-MzkMZD=H2N^@JfHCSZyy|vMLh9ulZRh(3`jfM
z`6YQ2MXv3c6&@OC4|R35VW8Zs^<*`5{SGfl((mjwSm}y%;IRC$nF9nj3ea}j3J=DV
z!^vz08k=PL0n2;1QB5CZ8y^mq@d2%AOSUd>O0iIs4M-sgAG2rQ(UAyg*>*o8Lr{GB
zc?XB0`JpNADF{9bLLQF4rFAqdVe+WtCJ8DPz;j4Xk+F#y{?~4P17!ciC5l3azCk4L
z;768A?rX`Gle+kI(UwLDjx4cyYy)m{zxQjD8e>vyZe*ChltSUXoZ_j_eW}hzmz*@!
zN6-_WmaAP7$Aq*noSPf?0`s$y`okqLQa1l9I4^Eb4mrQk;x$d!t=XF7IWztKmH47Z
zRQU4IEoqv(ySZk1Q#8{~!tP_lqUV<rm0qF`&R#p+s{5?bcp|I~w}y@vRmA?V?PQ%~
z?G0Ao{zyUx>n}bz&gpnY&7B<=O&#qEshrEnTD$K&^JxO&E8)W6&ezsOXwOZzCsm@A
zbA!nX1h{4#qn+9D#Rj)JPTjhWvROUo<u0od3k}60V1iBFAKtnGtIM|<tt56~4tt=C
zAdip7rO;8dv5LA+qQ}msb4*@52NKIq6ycEK6{8Kqq3H21>`<mHS#<=PXq?zW&B|BV
z9oxl2q{zj^WM;dZ`ti&Hd{|U{V(&Bd3_oXS{dRF35l7`Zd>92SD7vV=8Ir7?68huq
z*g^&0j#{*;MV%k{aYZ7(+BAa?$jFPMX9|@*pxA6cz<Yh9*{D-;d@8Q3>xx0n>D=Ng
z-EFhq9AtxU49zo|xW}phQ>CXlt_iVPTnv(nn%U21RvtlKuZrOIt*6IS19`BZC*#KA
z#xAlCq}TPesA0$Fc&HpFh&UIRiaCq~mFu*spbxXNLF&2cTKolmDgy=k;D5Tb%Cu~5
zFL}iRil&hnWsp+(t`0v$-KUvfN0e>LKrJ#;IDQ5F$uAj#h=ufu<ty1-;NhnpR*wj0
z?(K*Br4Lv01GWAwy<}jtadfue>8YLUzp&AIxP<B~{hdtj04X!!5vm@=GCHDZzsk&S
zh7_b(gSG-ap&!aW5;&Jy3sgWiNYF#3<w}Q<+%K(8uE4{sSa(7m7p1^WPJzwD-sHQ=
z_gs3()tl4GCq^UnOjF1_ZOd6?_dH%FkaI}O<_jd+GL=PLMl$FCVLZzfZb61d_LAB|
z#Xkw;kd^0dMSl{=I^4(va#lpTMPtmN4#+6YgN#zBCY<rExP-1W2rE`Ne_EiBU(onJ
zKW7hkP`#e&#&d4sO+%?mg)KY1zC80W^NZQxlrIK!gS>Omg|cZmCS&p!t)qoTY<Sv&
z+$lMboTpb%kX5Op(?B|4I>E~<0-ZIi5XWEYCO8`RID>j&df?f)`9NAMKwqc><1u@e
zcBS_dDEp<so3!TmAVNBS%gI0|a8~qge2X{dy<U+Ru3&2nJ&JW!cVMPGpw_v#Y#}FX
zFMCS`O*jwM>fw0MnY852hI6B<kz*Cv_<jeiz>0k{xr7k0CNaoBDCfx0tg#YMM7Y7m
zSsKH!`-J0GXIwz?JVW_L8Vg0r!mH(17+tEfeg}ML?A~<8AN8WLq&|NL_Kh^1WGIOR
z2UXA^B&CCoRmLuISh@#RKIh1yFsPe?XLw*zkUB-MT#SM@$6M%YJi*r?`NnbJL^u9q
zcr~#&**`5(=}RB7=Mw9hsR_<8r4JHa^6qY$iyqjk3j9~jx5dXZ3Oww%ZYzF*t#!wF
z`_LDK<R<W`$VaGYZdb!OlD6j+m8s4h!|4d2lgkCOiJjDTRr`96*^_47{S^o_mxaQ&
zU3C9;K!_*FHSxPoGh_m8GkY;9KNZT^A5rvMUHE!GATh_Ma<8mOXVd)R8QBumKt2;w
z(P5cfLd|bp6SvGI`ac0W+hA@?%r!D`X5lq_WOUTWt9rU2vS%fYt_S?y_<d07>mgEi
zx9?KLX2E;06ya~uakG461?r~4(-0qf3n#NCejdAxBspw7@{Sd;d3^8EN}s|>KfSZ+
zrt967yvg|}JVpIX+Li|kzL<n9r!OnheXXA<>aAF6wnac=)r~PPjmew3fBs}a3F`VZ
zZJNs+bLSQ@_4xssuVa5$f7@b8vmn7TE1KI86Q&e7+hY-^dy4i9(||&8(!k(-=Z*3D
zl7xe&B^S%&T0xkUJ@?DkQnF)c_C@BU4!H#f11c8W{}OeXa$4{pgQJSn->TEw0?LjT
zF=K0$?QCvnMpI^z(y!?$fxeZ{@C{Q7(#JZ>^<5P12hEEw4v_t1=_QgRYVi5$8x+%?
zPQN<@XW;@{c@bw9S3+^+Ew7qOfVOETMpgSe;c7QsTAy*M%UUS&w@tg%Rze~J(Q~$`
z9>-rUA`S#9=Z7IW2<ovXmnvPpeS9iZi&KRyfi~2(MEol~w=vy#OIhJv6&puuSGKbO
z=oeqp<klz$VXVFvB*lS*6S#sqrMea-<qvpbTAx=YAcVqCy&BmrWOORYI3i-fX=FAf
ziA($FV{RAcNTSmUorhXHHz#(^^}6uv$7T3H4TpB~>2&mRJ?8)j|J#ffmkmI8k})S4
z#&AL-fh^Cl8B*1c_vD<X7j4EWL{d|@q;7n0e@GrjAlPYLtr^UrM;PCg<fOT*H6+)x
z^=K~7OfXVRr#N3J%NW^772r;@|4}@og9m<*vgH&E<Ps!lt9T7zhacwCR~|{(bA&b}
zp6CGA3h3;dKM3kd3Aa!y#{aG~ceLnO*t6ToDhg_yaY=9@LE~e3CtCLfcT!Fy^M#nA
z7W86u`|H<ty!x7H!cUAoSGI*MM2D+%+{Qu$WBbGN_wiEAL+HO%9|-Q(bV%7pUuRt9
zC~1Cbz9)Z`xk7mT(-?9z#TihZlGpQ<>u2YUz~<=V00ybv@bsM|RgZuKf-RsXQ3~*m
zQ8Fz_1r+fg+dCK?n%a%-DwfqQ(#j1Ta2kngQOZz#QWS3#2e`(<B~jc}k2=MonlTf*
z1jW7g<7I7XH|hE7T|-!xwte15<-jsHa7ipGMuz1ZU@MQle4Kx5s<d%UzqH|ZBdbiB
zzn1Mc7$*w_Ilh8e%qVdwf(@g);`#>x8NF=;1rORJ9iE#ULMV)A?`?b>RbMg3fkg#f
zPx8+sv%it=j2&knd0*=0m^CWnj;+}#Df8SH*&C^L&nJwoeAtDC9EH`_fZKag%XKrm
zs{2`ph%nrN_C7BTW>hlCT>IH>vhvP>)a3_$TC9+fS3(vi-H|U<LwJS1qD7~ImXqvL
zjFHx-{_+v+H_LV{4{m{-QN`+&5!^JX5*6q=__Dl0o<m|0NxT-XtKo4nzKy83yz9~e
z@!==Jvw?!f{E}r!P5R!z?wBn;5@5K-m}{~Pr5eJHzH2yyxPEvK+nFN&PROJR(W$WO
z;a+7lnhxsK92e+#_1qy^;3Eey#rM{+B$Wb&1Ubi@tmG^!ANg2nACVC`Hz_HCWjWY`
zkkNq`*mqrcSG2$(oZ`qLQu;M6SEEUT6oJ$f?7#wwyRAG9wx$=DGK~jd3q}Q?H*$`1
zanM21g5=eUi!GEVJN{{!ykr#{-p#iQ*9>N5Wn-w;TvE=Vg{LBh;XBPi)x2#L2&p(s
zBc~{4-X5#df|)W~E#M|z4etc0?hQ7s<80)l>HR3Z-bQL4LPVgyTIRo*7rU(WvDB@3
z>$33<2hbIF-Z)r0t|=-GfVAd^sdxz@CTx*;vJ_GuR*$3KZ~Tj!j(VoK>v$Kd(C@d>
zp!)|^4DxHhe#c*MhTb~gR76pB2d<CSc`#JItQ0T(d~tN^Qu(BqP*FsHL`uK!tS*Ta
z6Rsh^wgRe~jl_JqqO}uu>Uey4DV5b`w~wYr%sVqUnj4G@&UecvPK6tkjJK`J9Z%x|
zN{X_V6L=xr17`(n1Jyr;rsOyW<F)LrO2|ec@-<1Z!q&Af7rIHz$v{sjT2s%$xxY!L
z-i}aJGeqFN<O+5hy!y8{7Fz4`tLr|_K`v}qhb4+0`#dSP*?WIsSvM;*PK)#|^rw|>
z$PVU3R9qXqS5h<Gzl%5J<>qy%6v)gVeh2Bko4IA!%7dn?k&)y5Y$H6R=guqcO(|RP
z>;`56LW~-bYxR4*ZgcTS-az`mJSdr0nLCERj55a6iVJ_g1{8ifbRrir@yFiJJW2A)
z;P+;lGRPtn=RWZS_J0;))cj^`QQl3f#}RGmIPt-}Xvk5QD_~J?emHLNM|K>Zd&CrU
zhQKF11M@`)o;cq$U21>+$O<Mo8@8?kPpV?=GD?Bt>`QcthnF}T3ei<SrH)GyYH1~}
zE%<IgJ#qj#Q}&Tt44(MX8orI}MEBq>Aj?X-c@)L#j{Q>O)AAQy{~`wimg9E7p6UJR
zeMe;SF;-Y%w46x59U15Jl68uxgN<nSH>W_5UPYAJEzrYzf;Dig#Qyw=yoA7O;ra5W
zMN?NjgdV8_W(p=6$u+l~@=@!O(K|m!zi}=-U6+;sE)meoAqM%P9B&{c%=W}mGDHqO
zq|)Xh0UhGe&3m2N%HCvVKyH$|UYIAQlfq%K%X}dr@^w7rdxcQi4FiXLX8)_xKk;=3
z-)9a9igb_4A1_n&>kk3+;(1}8pBD$AnMBx+UKMXP@seBxb)jgyrPb?pL2qX`qfgGq
ztAg0w{9VVs?5In<#ey6k=lsg+m1hq4K33D(Jx!!UNugt;#iQDevblMQxGfq<K7|Oq
z-{k`W+rD;n9X2h#m=3!U*_^9oqa4lWBpeyMLksIdx5ckI>w$wQb~R|$8!mkL(Pz<;
zbwd&;;ozBnN9)~R&uUH#*!*+mD460hc1_+T7WJGz$D{~~$(A7$5V8~&AegxLnBA4_
z9H7jsTA%q`)h0>T1K-~J+?BIIP(mXrW4AN!ulY)c(4Z-|cp5Z;4Xt{gL!Y<W^VuU`
z&d(FsU(*K2s4{f|8915Bn;g83Vp^9tYLUQ-(K9L+8Q~;_pH?#N@$C+FYXQO(t~jf7
z3%N#=zCYw4d!VZx=knpn%<(E{J7?`1#!tOR^Px7~f-a&!Db`nhJoi~K!vNea?`kcj
zu{B<)lIu<CQurt+8nwUD_xQs8<k+_3FOpkc)v7s|v%t7HMp`)(KaIWgjDDHSCh>5z
zH;jh^iWfs6Tlh}ui)>6;n|9uMLk4TEb3k@}W^8o|xl$I5U+PrQ4iT;L-^%U3-TG#D
zPET2IV{*RS`$!!<30*b}?4afa;FXZ0jck5|9r5l~j{1A~cp>0^?7_>THzYM3zr&j?
z$%agRezl~Hk2lDOz-5?-XEW5N+kf*>jU8X){22?Dj}f~M3LkAuU`dQz*U-S0rv3Av
z0EH)m%i(s0x?clI_FRZ24gzckt)nGE>5d5P<jM=zs}bdV%oMRi#ZfoTj5^Ch3YY9d
zG+`>9E8;sF&yToqT@-#uV=<;gvUbD;z2HO*ZB2FfrOuq@M{d+3P3{zY)Xa9ion^&-
z5cQQJ>N+~zYJwT=+bj;}W~o!TA=S;|>@y=QgD^^wgt8w1ufCGGtl(W|V88(XD&#^Y
zZrp9;r!|S^mQosn(T-GoN;Dx_VnpV+4&6xs;O|xH*aYr|EZb`|MjE5VE{}`IDDAwf
zirKBSJG4VTs*eF7+ORyg!*>1si;Zg`kjQgDd*|lvxb81BuUBokg&?=zfI2{pQ#^iI
z%!Zsux`AdQZP<tzBdRz*_7o7)#ar~Bs1B-paJaCDRW~_YlrlzYsl}GrID3|ySHCoM
z87CS^-U`Q$fC{LFLw`)elO?e!u2BWTkwEtO>L$-uDegqjU7U%$s0#I|Irk*RhyK;h
zQ0GN@=%c3Zh}RyY0bHO}J5To~Zzivm;9_Ma2W+>hrfH@6tm>1ap^+erSyDx<kA1>o
ziov`vntA052K&ROC6aEL?V|(NR^n<QJ_fm<()(%5g3UTp&RImmJwG13BQ)UZ5k+pA
z&E)}9d#Gfs8otG7%o(0YpA|BG%~HEn-@=u}pvsP?qUby4InzpyF!MCHpGFZ-b_C+}
z%Iiw`ev*nMu}D4DS!1??CS<z~-;fmhv>sIqUl0=HX*IPsOglaC%~#yIymZ4=PWzjP
z4bJjYv@j(BS`oE&6m=I7=7!^uX0^Yx!pti-<>K2yhgS7F-S`_9c9PGu89NQXmKpwx
zYo>M9uNP0==p-?zn$D<FMxe}9+;N#-{3Hx;DJ@yJl89(OBod=_qYRK4B+ZL@qHLP+
zq7eP$0^IvXDhxCb=W2bn&&?Dz2X)e&)<h6V)KVHI(?GhW;1=D7Vu|weUSTBX#QdS!
z?7~lkt{?Z*j^n`3@(YUZGYV=T&dZ?j?VYG(gl45bA*RUY_tq6IzzTW#oRO{*Vy)&J
zMiTMCbpt{w+hPU!SM|R7VY)?Ss=F@n3G(OabaF~N?Znrw3JAQ3l#PM3^-F1t(e*rv
zN<da5^Y)Jye!pyr-j&twe1&!viYEo_G$Qd?uZM(&y%W4VJ)Ve6vmcUt!y6f7E4t7O
zn_d0DJ1r-xG;XV0WpQ-)WcQimR8|Z}-~G*k;vZ&R8pXq@c!ngQ!()J%M+pzN{k-_s
zZn4ek;{vGzmY%wNOV8kPt*W`O(J`58D)(S}e8m`88dS)I@>UO(Oi%?Lyzf*1eJ_I!
z*Vnb!FI1aH@EcY1@ocb%I1UOj6OJX+*OSB~WtN!j`%dZ%^UR4Ckv*}CacmjU{8Poq
z(hb5jf^d(-HS6lJX$8X;XtN`el>JWu(eI$~n?l(%&kq7ib4HJ{Ex}L|ReDwT4ah`_
z%K8RfbW?o+=(sVvT671$j7|}1&qmp^HCDe+B+}7IvOysMsvW+R{lUMxMAPt5UiIAH
zzh6Ae!Wq}1W;&nUg$D-PT))(bfxgnj-H7Y-kQ9cN-uPbho|&qD^mhH(`7oO7QhP&r
zLlsCU(XVYo*GxN;YSXRg@=Kpu03|V}@UC`A{&KdQ6x}`>QC(?yM<g_aUwEm)4cSMp
zg$wDKa!@hi1wA&j#R?~nTXOOtZGM?JIjK`Zn9HcwEBQdvGd<rPnjZ=(D3CHaqX9Pw
z9fd<&S!Jo-gWfCo$gqd5#@fXkO-iT8h(LK684=tk?OD;Aw^sJxH%jR0%~_xxl0}uI
z3cbCr|0V@8=>%;LdpCHTqP$Mm|H1!Ir%_<R)D~Lh6?@~JJK-V@Sy#>f1eHKvNB@wV
zc3^wM$hVG;m3+(b^TKF6MKvRZx6STGw!P&XK}-Vx00W*f%ZRwD0Y!4YTYX1pM*D>_
ziWrH==S9S8bic*r5CBcNP(ELh^;?HhwO>4&OG$fW-)OK=&;G_ahquFq4esx;Xzhc;
zgJDXgy^zil8WPgBsA^~j)$h9~*`G5Af3b#W`*MU5Fn~8LDDraKeZ7(mA-Z|?S|_lC
z?ckQ>GOY24kRL}oqTTw*&K^MatqZ<-sUT9jhFD2Hiki3Teip}sy-Y{*xVV$JCz?Pc
zffT*SeXmK^9vag0L1;zqt$-P&0N-YoRdaN`T>Ai7Hox_oUnf8AQTCKTh8&?)`0GUo
z@hL5v_D3l?UhOWKZ0qPTcO@pTb_bFnc{VErywFcKAxi6lKhtM(Ukh5wHqO5PZh*0`
zN^S!UseL1LO_h99pvxyhx`_~d`!Veo05(K)aNIh)Q+QY1+|ga~*CqnhE#-U+Ybzk_
zlLHb&mZBP(k{01GrgQU=(Zi&XMear_A7=Z_k>z%_`SM!jG|_FeLGzriXBNk^eyZVW
z-p%i9cPW%};P<b&kIYx3VS~aihF^SI5O<h})pnFw|FUB?LEFIJJ?DO1<j#g)+%I$_
zXSIv-D{pRn<syi+W-}*^XRNH3<;Td2%Ps?}mMBlB(aW>p4L@Rg#-bm|Gx~M%DG=0R
zEY;&>zjbYU;uN><WFmW*w@~Q#7s`rn8ap&tb?tGQV@0($owK0(QR89X7ta-&v-5L0
zto?T4xDUpd)d%{S%d$A@XR|hGb;m6_0bER6u6NFc$5oC^Z{jsR_1)e3PIt3M3H_vr
z(Rxz5o>K5ch*-LFys1I>++*o5x4w}hMZ><#Z|=i($=<K2Px9+uNU&Pli3N=gyA~)<
z%0(!o#*f{nR^x23E`-e1C_Rs5oJP4l9q}%0IfiphL(l43dU|Vj9p`?SY6ENIlF)H;
z`oU>xu*9CTpRZx&7RybrDOi7Ac5(k-*vN<T+bh%#VXz*>PY=<g@Q%=A{mzy`e9uru
zUY;E?Qt^wt{f)>s7$C#-JNvLxy)lb7#&6w5Ilm75J?LV9wT{D2{)5P&pA6$3X+vbR
zxZ;k#56a-7B*kJ%D9bKRW->3veM|paw%lKiRAl>{c9mE2Y$H&f7p~$D)>Wg-T+_%(
zUsbProc%<>^L6s){a`bX&#}>u;%7~I@fWYmvWkY~qj`g}R4M*uoUy%OCsFjZ>(QPI
zuU*)D!2FxRi$^%#!n>-P_Rw-xwu6G53jsl`0x7Op00k1fZMPnp)8miGBv&Guwt<S3
znx=)E;keH3RxeN<^IXzHN0<=9>lxu2sw%kQUL|RUEMtTl>px?mXdc2Mm|M>kv+hxZ
z2~6}{jQ9GG9~)dV7{8=@iQ7|~R*&T&a~a8Z`E|qWd&;dCpU@0mhCt6}{>zMLr(;HV
zO;>gg-qI;r!GNb4VGqA)PLrG~{SjMt9P`%Goofa(?%{|u`*=n(*>0N4y6+Sh>mnVt
zY=?|fezu~*R`rJ+>57UCq`Z4g`=?g0ge*ph+O0~9QCc>rSQ?=BvE0Adq+1l%qEfs@
ztvX`-h&ySeesh5Pt?90+$n)*`Lz6rWX!*zmCf|{YlicvBD)Wr;%<d1=3y9HnfL}=>
zZb*Ouou!CTWLezguyDsRcRfm8FXR;EN3N26B5qWg6WTJgh}`4#Wvq?cbdP85nE*ZE
zm_!Gwew_94GVH!*X#08i<NFIH;@`>IO*gVq0$Wx8B>t7v-wl#fw4{5^0vXhk{Uv+H
zBO!^ILBS};IZtP8ItGu3@{{kQPHKqX3O$bm&VV98%Y1O@9KP<7@~ly(sz_$)boAFy
zTGsAo#(ifv=3F1yBryypHun7w0*BB;zDr2xs4GhCH<}E(5!NhEjg@z+0qt6+F9K(e
z9T2Zgi6n-?hm=h<`q_myzo6`zY*=@Ce6Whht(-VkjL*yn-~Mv1RTFFfeMdpk<~Gsc
zn`%N@^v-f7%%~rVvdJkm`2yNo_R$^zZ*vnWIHg45Yo#>ZS%2Yecq*aUgr;-q1j$ZW
zIb}?sK0$cWZw0isNAiS2L7DP-UpGhb5NmPXEm`mUo|-<L0~b^w7s%DNrI?iJ{s}b-
zGk4*=SF-<Ux6VupwywLh|EVF!39ntN496+O1=U*s3~hs$t~B<I>$AhJ;;3j*pKrg8
zxR}GIX-Get@(jH_$|59)@58L*I8W+}(x!I-;6et8!k^S^*ce$n=VWMI(QOY<6XhAA
z1I4ueCaj`GexBZa>02Zf^clk!kCGI<`FAv4w%|peT^>76QZ%Jm?Xf#Sye7tho_$j_
zVM7G*$f0oGy{Z{V`A`DJG4-ACVeP(0-A|DK)N65|fW3}V{aQJQS$ZjrI4M}5q$`Au
zu6KVur+yP88rcl-FNNj5&hDaH{t3QeD5t8-wg6ydJ5U^!LO;_!@^?68JOpp#AuYP=
z2r*84MdT$+`}6e-@0@+qkp@>CpmS!hDrLxQHumvg&Pk!t1~_IAf0FIIVj~{2^6U~b
zp&klBeJ|z4uKF|ziGu#G9P{`c40q5T>$Gsa=d?Z2?nq*RM>Ua*guhk{QxK2@f}hZ+
zv#wI`bp?rFi$=1`;203L2>Hm!U!k=os&U~6m-8d|L$uyV2KQ`WiarYroLEPV{a{9a
z1`;;DmXG=O={i-cEjHU#+CrFN@pV8M;B#)_((}v57F?Do#<87Xq`zpDTGp~hm*Vbe
z0FyoK-3Dn<he~2Q2@dc80!42LknHW%gr`K{lf1h9e>Pt2ZBOnAIFd*LL>Ly*-wk2i
z7NR~VcAII4k`FH4cn=_QT6ago;=AQHbIc~{5J93EsDC)l?X1vYWH)5<ZC4}>1wbOF
z*!Iv{0l2I+Bc;oOKUYFRe%GuTJ)k0$8i_aly%+#-L^6g83d>(#f1$iR0oWn_&<I+E
zWD;T9mz@#|?Rp5OUgl~fEV8y!>B-@%Zz34S=7PhF_SwtVymVt$XRm{z$*JGSfoqBn
zDGmmg^cf?Xj>@MA-uF=K4>I{1^_@o83KnfFBkOcyToDwcBsNN^zp6;&V`w!0y(WJs
z{093U2W<y?a&-<#_lVWB?M4_~Bk#!_0unLsodfV2DFKpsC=L}#>85o4?S(eA9&dnK
zJt%y|r8#l$&oAD;3l?iuvE!9EkpQ8W_S0_y(;>Cg2}qRdw_A*^-zB$jxd%u@zUTts
z`ALcAZW!MZ#u>mG@W^B7)I)5$o>+kzqWP8-@BbE>{P!#YT|V#>7e4h^W-B8v+12+a
zpgW+SwL)$_iJJ1v*()L8tGFAAl>e`93&89PpSp~)-~Mmm)BoiIzU>1pe#w5t8_fUr
z5gC8(CoU4~PHwAi`u}gT%Z3|ZJ8xexy}$ckzSjSG{(?UWlB*w}sq*;mWdh#iG8RdO
z$2Uv^(B(@WoCliV$=0EkGua`&kg!uD9L`LGwf-0PO7}T(V!-pw?J2NL|Mg<~>kkc>
z1a1Maswn4UyUnx*H&^?DH+}Nl<L^0O93%p`3|=<N9kt*a3^7;qaaeuNj(t34vaLtW
z7LPg)NMZ*HD+wem;pF_bY!DHP)b>s^06NrWZiu?NvgC^?aG(D`tN7w%%rQUn{|>JI
zZxe02wCR5Kypy9+;9reyU@@CmHSi|KEv$@bk>~8^YR+Rybyg%OS;|xRt=e=MbJdIo
z#Sc}?LjlOe>?Dx;DUd28p>@0aeWs64fv?$PnQY<H!p&**9Dx)$`_xe{9h*a4Qn*f&
zo6k>B@dqxVgX$S5yfJQ&Ygji_2?9{Cn|KXnOvWf|4M5SbEHQsjc$%cT<Y<8UksDXc
zc-v#E)h%#jVV!gct9=VSK=Q$BKw~%Tv7Fyt;OBWg3OyyNf6=i2k3~pwid<bgA)o&O
zM+WjG>Tnmx#lsD3fe;3-m{J|m`O}vkjTadulD%-X{~FD9(UZkgFxYiM<R!RQ)0$#G
zBD42BH`NwYAglkNlR^4$PDAW<V(Qa~AN@jyZG=9}3jk*<xPyK3V|ee>Fg^0k;*0Yl
zYmZ*Q0G`(Z)s{U#Xbrtu#PfX4!Hi)){R{a=>C7&Bwd=EKH^;dL)gaee>HO|?UaaY{
z!s-?{25mrTs3wWa^j#Mh{L%}Z@{@JrMNSgHsb2BYT)nq#XuafGehT}apYQQsU_qaV
z(n_QL>y~Ud-6g4+1Hw4HmePgMcM>;%fLtR%q~3<C&#{aX)q|`<zT3&TDl!gKk>K*%
zrS?KHok?7JewTzq&{gCqRVNT@*}r8|)2pgjG<Kf*;^@S|Xcc8->pTHnB1{A4$6vJ?
zRYkRcbTs$aypUa;4AG7eU{I@{tv^r3aULSu@;halg2%b=<mvzXnBr%_!D(&V`!-wZ
zUmMH+^G_s}d9UgAM$h`7lYt(XKDBo~O;5TMUX(utR~UmzyRTqr{6p-UBQHfSG}x5y
zv9ZHf5#^c3!1RkXG_`;?{CFh0Uz)yJlKVeTED;qU1a?tmJ=O4^nG%Yhi@qTxQqyTK
zN!j&%lYnV!P16bqw;+_|n|-JxH~gEmF&whglE$AJ0to{(M-#SFgY6n>UfUJv03j6q
z5W-r%&ci<HiyGqIF3$6Q2(C-5q|l%JS>*V_x#v1yfq^N_9WGN1<T*cdZs2iN0M63g
zzIPYJ&}=o-D;v^OO7@)m1}?N@-X<l~Vfzj7)3c8m66;25&}eZm8$B>$wE9T>yaFMC
zYxr>ugSeM0VBMWc3PFpq?ptQtT}I&}Vpz(z<I?6cTV%-FL>~r#T)n)qD1*Eg&bYSR
zjMi)6nYCUcdI-J`7o0>y;V+`DP-+#xpD-6*)9#+i)pgo@<67*z38m9x@<?9S>eCxU
zwaj4;O(V>uYn<OowvR4?999$+PI8a%<A<ogE8PZDK~iNkAy(v|ROp!qv*JulLB5E&
zlRTOl9NlC@j<R!=aX>H%6sNP>mp)dzKkU}E=^V6*`IT?8(wn_{Ps(s3wmB_^am3g(
zY?O3i-8I#e7fu6(3a+cf@~c>y>-#hT4P4zRC5;vQj^r;i?57&z9PDopyR(XKi+gft
zbc`bi9oczTT6A^}qDuB17Z7$_CO4Q$9S&v(Cac(_e4Xkqn7=j~cBFOp44<ogsQFc>
znV_>&$}?PZC|CR>WR>7%M_^rIV5NM0xf3Ae(a)H7$%|6hrkE&~L@D=(`H#}JC{&I^
zRVJ9`&dZK%xoVf}xq8F-ZfbTbDrc_wM0O}UsZA_ggxC2}t&>AqR!Rtd-7w*%N<rZP
zLgY!^qH4?Ff){r`6lAtHA02Vv6Hg?@u!c<3v@c?s8#O6EX2PZT55x2C7L3*!*tMkk
zBh~MOe;gFxfV15>0B*z&<Xj*if9`()4$i&oIBi>SL`P`=DA61^BgwD_b5dQm&>jGo
zkE>;p8UhL^r6or`!_#TEi1q0zX4oXPMVblmr2Z*LXJ8N#O=!CnU3H4wEILjZ5!Y!s
zy$XAzqL#k*;Rc<M2B73aL0NE=oUkt$hy;(VE>`25fS|2$27uaU0fAPDm){QH<bO2X
zoRmDtAB3J({Hd-m)%=Y8(MfI>dIcy_q_7|L046PV{-PnhAD9ni-tqTJ<3SPLtL7yQ
zYZ0cS^1KcqEI&SKArNXG)BR@S8jrKt-j04c@+6Kw_qij~$6jVPB12b&9iW3fxCivF
z_@TQ~@fg(MHImRnciJ7~XV<x`CFF)>zX+ijyqQ_JSyOm%^4<xM&1=`wJcU<UK}m7d
zaoUuwdhDHSEr3+_U4!q1*<91nFcu~R7z(Ph*3hG*H?+@*SsTyw)g){HF$&ixb^$4j
zd=8Nql<Gj5Gp~L?-fWF^SixB}16pyrHk1WpTvFi1ZVZ5Qj}uU)^TD)zmcbB?Wh6H<
zK)?oQE2?Kl0H|pJ$6*uP<1)9ENGUOQq=+gn)1tZtWV?Byca7opt|>*A&mnP|wn1))
z!_GM*UjLV-ojmB++pY!4EbcCJlUYr>{zqF~m7@w8XEh3=FDX+1=3NgS{fG^){Y5*8
zDpj={l%MOcb-X$KjDq{AlCub&sYJ^3B<t{UyQDfTy%_1`hdEm}oTxZ~{A_?wqCptH
z*wqDS0sNsScDitTQ(Xth`{F`>3?1m3nFYpql?cc&-ZV|BI04ZO&IzMKk_!-n-BULY
zI;6|;JI&RaFuj%rl6FRuND;Sy?2k@@F1HW6->*JC`CNHGnQ1ljOD>U%o&<(O^e@gG
zf*A!|<m<_H1Jjv#8USLbd463M?$Ngkpk>I}F86I_LHm6Au8#Af-&}bw%l7HKn{_~%
zu>&@4J*R1bvk6={!|F|{i_;q^fTSLMLub}<9`(@XK<+pZ4B?`Sw&A1V-DPpGy}4M&
z+n=<u*fd`a=5oj&%kB%0H%~Z<w){ehc$))~>cRmwlgPvO4bO8dBwx!}XUG`Fk^22Z
zF;KoHEab1|kCtB}^*$RLWtw1prXYL7=D>a-VL58<yI4<yxY=y%>0I=v&Vu!*;gn+k
zEiHG1UVt6cz546H)DWc3PgS;eSPDXMiC<~H0pG~4^Z_A`=A`81Hy_;l*9nulEZOtn
zR4L_h=@V)iQA-pmJ|C8zN<b%<wLPyLoOa4Oe_h@?76CJ9R&=w@K;M4=EScx}3-A1=
zf+E<5=U0`>E)z1fJ)K_Zl&e9F^t;5H$za6G>;qY*MV@(QgKDs?@1;D{s+9Iu=zYb6
z?(e!0uyWXnCqjN%S<%m<;cN}2G+0?7m3`M_$NSV1--X{ij4MV)i~sE<w}y%k^-lih
ze?-$BID-YmX&oKqB3{!+Hev`}qPhZn{NxLngFZzoLukvhtJ!toxQp&^|A(&PnSCV@
zQ5Q0u7GQ%|)F|@p9j%Y2?+7kjD5Rt@meg|`MJ+Birj_;~URn1y`Bf~RHJ&y6D7}Dm
zIZYYG-BXX~B%?NhiqRgS`6i@&>#n!ahFxwIn9O4ovE=b;_JXMsUQ`QU6pOcrHF<nd
zc@INE8dYDtjTj7!_b{BVp`QVnA#u&F%`fX)ON&yEV&@MSv*ySUsOt6U`cJT|D<eVl
zerx9J=-#a8jF#{vg%=yk=0?PA{SJ$iCCpL}dr~*F85N^Hr+O%4vw@6_IU1iTU3>#^
ztSyEvE{&lp7vM`o^>EbE)jB7^j7pa+1>ZH{wjo9_X=*CrMV(;9`p(qR#*fp!P{~YN
zSsK&uR1YKXBInYvBQOG|rmgkWmq=!o3%uQySJIJ~y+WC!Cdup^k(y@YW9`7~ZMP<T
z;`#teRha<j|BT~_$}4TTj}H<8pJoimb1n6{oG%9CcEog!5?xahK9&T1PL+dLJ0B}(
z<BuI==0Bd1%YTkfjAj;<tVD;KQa5}A5B|tV9W?MDu3o-Kab60FWZugbR;08@AL6X^
zJL@8%ry&fL1l!)hFykepB<OW!B#qRiz$NgM8I~?Um_jr2Zn!Cgw*1)THwAa3p&>On
zBp~RShdLuAOBV`mz0xN(y_CU0(ELRV%#40&QP5{0^m#kDA$8XVfDh5jzEI?sTOoPt
zSsy@K87db6270IPq}v#VTt;FBBLv;RL@J3-igqq={3XtGv(33I&y|LA#L=Kf-S5Pk
zwzd{w5zQoB9-$8cUv;KYEW=@KRY8d!x85TADM2N&i>)$|9<(-zx>fIqF0tUfZaM|M
zN6?4E6Jv>kg`Xr&>b``Ds9%_kE%>%)L;aw!!f`gvwX3m5*UScw_w#~iU&{<R4Qqe*
zoSp{<#NNzb`vTWE%c((&G=$!lH)NiKK2l&ffF*s{rxrI8vEp~0Qa_HbE3K#b?Qn!?
zUkA(e;ZrgS5)q&qEqP?TNE)J_2AzutZV!GB3_A+KN(ow?BljMTH$Np(Yu`qmMCLx1
zIOVRyoK6XHq9#O0&|;kJ2A55?aW=B%-ohYBS;n=2JduxWQc_)Jlv8cvO9z`W`;1!o
z%tgF|<hWEA+AXW+i2A{}q-=h2yANxmhi552B|?3&U)}DvL5F@HDfbFRNj=`>v~Id~
z)RDZT<(tF{dF?@L_hL=8-m}!}(E_{Nq#6(1bKCi}03(8<#EgzoLyupsTCY5OD|Z5(
zX795sx`kkp(h$yFTs&MR^l;wb{5aFsuv9$k_$b_`tlM`9Dr87r1ra^p$2rYpPZODb
zw0wVeAIU%w*%~~m401LE9G2i)?Y?*!H)A;i_nZ>!>%Z)NGfi~4Q!%n!{EzN<CGQ>R
zOGdhXBIThknDOoT?`?wAli^s{OlI*hUVRU!apedR?GRlXYWnk#1zz)ek?JdfZsQKn
z1&=d{z!Lm>Pn<&piPVwvOeBqQnpqE$+N;lZyO`+C{m6W=jW1dkmPt|XM>lxh-bAW@
zKe}N$r+(1YW3C;`<Q%znGNGdr+bKkR!qP^~MU59Ij;tdEN@UiEs|7{06h3WlkX;{>
zz}E^R<1Jnp#LSTPrNKy)Uhbj^XR6a#x##_=EfM3y3QQyj&LbdxgNB+MIQcb9LU2Z9
z3hRBB$-CeBu?6*PL<+`l%Pd~<w@Hwo>M{+@0fW~(12L6F7NtB87;k<@K17r`2i2g>
z8)wtKhH1gM+|NG^Cf#c`r0jiQap3JveVML6VDZ#XAkfLVx{Jn$YEVY?J%1kSh$z$J
z`SA#%CC-ZZ)oo7too8}yT`)Fn4B1HzjqzjNNf|c`qNf+*3nn~jIH$Qm08<i(oFvr6
zH2!{;vw-8~T2xb`e|490bcqK`gB$B6*jWUA%57@&v5nQbDaiz3`m{-XkV<&<`kSJ3
zV*myXNkMJe!@vB_!OR^@&ork5SkjI){+lD*k@Y9<jgvzKw+PvZuQxY_GncCo7XWgo
z>MGT*uiow<1n+oR8xC?}3y>!fqHGVG3_8P3_RaC-sRAuxdj)xWG$0Z&jMh4YsGgNc
zV1QYkSXL`Z%M0IZQx@wa>5vNTChBQrf>qItLAB&$#<t$py7qcn%lXkzn~2Ko6qB`G
z&i<|P{eq1_d`22aIRU<d<%exvGTIV-S0VZdCfbBO$xhf<hg`0G8jx=Le*LTx<b;-S
ztbP3Qk#JNgSNAyS`!{eHV+R*hw}YfB{6inyU3vskJ9t)qK+(aZ9XH3CFVA-R6}=)q
z>2jF?bqa&30P_YufZlUI2kU=|ZoOc%RLwdXswMaTUfV;X7^U$9!qIcE_LC46n7k2f
zaDB-1t7Maxq>+STkBF|n-t}|NQO7sD^Q<<Hoam+$o5%?^p1|t%F8;FvccG&F%3Z0j
znh$C`2E)bkeriP7>aW@#r&#*Le~@FEd$J=f&=X8D0;>tOz$#v`^!yU>@Cl_Um+XO*
za2E}h{&-tH(?D5-G!(4NCaN{rd8~=R1&gHD*4+0)-r7h};(vSc%Fi@b@TiaTdPzv=
zYsiS2VsjdgYQNqW-+NFZsSrmDp`fdkD8+%$X_yIf%$*$M>hqq&!1`*6%Z^t+^2$+S
z^+_s=x5y;beSf9QFjxy6A5*vP%bdjCQ4`~Kjqu;B`&Tb1;m6$SU}cFYHYGXzy|Q$>
zaTf(tU`xC7zMdeYsOKYI)Jnv<%jt`&9(G2?E<}BU%r-H_TPAGZ<LPc>qYPwEWWP(G
z@GexUrru_fr==0^=ZI>o_)~BE<i-p^jGM<r;f3&t?X>C=Om7h;JENeHviGq@ro7fk
zdDZo%iy<FXC8Aql#3x4&m~OK9=M_g5rBG#Tgcp<8aUWQ}IO?Om?4x2cfVS79t>*Dw
zwyg-hZ>w*59_zBkoj&W*QHn2>YA}B(okM}@>2A{6qFuG`xHtt?X1ut*Oi7}_K<^m}
zlU#?^TgxS&oH((QElkU8=*$XwCX4!0on;>!gWCU>V-Ru&^Jcn4L7Do`V)CMt{Cuwk
zjGJQ-3*^>oL2cS)p0UyY9(>P*sQnfUr!soCY1Ryxafs{$D&S|F;frV*i1IJSn$Zu5
zZYFsc_SCwbXcs<ir@$&w_in0ZN-F!PIw;1picL+<ByCs=8qEe+Gidh>xbg0c0iQLl
zUfEH8w$w8A_x^IY$f@^3*w>w|vJ|UnqW-`S_%CuWO_pr*+aowCu2p6mfE(5rN@1zm
zr%$OOggm$O*m2oO_n#1r=X>A}mwIRo<NB-7w~gp<2RO+0;Coh!ooDnkxA1Km@^1yV
zUvrsfAeRb{x5z77ycfrBX6st*;FyX;d1Z3aWMV>o+@y#uMk2%f+Is5b%9?Eg#Hxpg
zw1}ckRP3$fHT<~Fr87<PDEV~Ok63$SM&vGL;Ry9)&I|U&(RpXyC=WRyN>tClEWrn=
zZ>*19)CO>H2`Kg{4aBRU!EP9in~cTv@ScMj5n!=r21iYxVBU_A4IOJ~-z&-wS_cy!
zJ)ZU^tqc4)gwz&#Y}Gxq2`s4@`&|)7umbAdW;{$1w-MN&2XMeV>IM=kcn!#X4gOg(
z$Vn6eg&n{JYQYL!wyULHyOt>H<1E4S8|3u&<AakQ{gUF?r8*ZM>o`{qVa(7f>*rf0
zS@0|1pKnMP4^VLzVak_3@Nfr*^N|7RGSh--i`oE5&ly_FP>Eq}{gGFp^D65<X7WEW
z%ixD0C@O8b^y#22|3BxDmzDHSBAjH){dJ47fhm=*1FxV1L~suoiy1Z(v8N_jfdfQ%
zbJ}jZF(1TcPRA4Ce<BB1em^^`lrVQX9+na$6v_6V0{&czI)$yV){_$`x8e-Z_|&QQ
zfG=mUVw&){x4;bASVA@)GC~oYaU2R*bqB))>}k*KjJ<M>>w_L7Gd~m{MDSNgzK7?M
z!$?*w?&Pd=$`CMF7ATs;swZ1+ZiCXrX!$u=162u(xv=6MgUPQq0-+s5VkzqNcC_E@
zVc%jN-vAi-@rg`BXNu*M^&`t)^vRa&WKGJB!eGqU;25gs04lZ7S#Zs_{^HV>`(XlH
zvJ?(drdVbJu<s7AXDad+NkFBQY`MK9_v+p25P_$=p!K^33?I#2Npma*(OMmd9y=3A
zI;V*2v#Hul_1R1`;$jfx3Uc;qGYluQ!)_6eq!L9=KK|mQaM0-A1zDB>-_(Iu5(7+W
z(f641$?L-b)5Qs9>7<T03@L2P$rEZD<zS2(Aj(R?lw6)y9=zBxoH2A$F@7;N#WR8U
zoPQi@NxCQZX;m)yw_N3WxMJ$cCTEn5pE+S<4T!_f2ACw!Zni9A|6u`?j2RH@qZ6!e
zn!n@x_01#|%DaC)svFA*5nPE7N1Pi{V|J2Q{ZhG(*Z0<~Zbl<RPBrXFaA+J`2N?yK
zeQTAsO8cEJ<RiA^@|OB8OU@brG1>#s%i2`W>GetIV>|zDCo1(oo4{UHlmf#XD>ZJ9
zX^S$i>+#0BdQ-Z34PVaYG|rb2A0lR<F6)vmTQp8<1_Ud+=6wt);LKT_y*^vmmn$R+
zx$w-b)Tnr`F}(h0nD!>_^|A@@UL#j^nRu`|8s`$*x89|bNLhrqH~~o^a<KKxRJg|L
zvlbc7j+aZtCaLPL(p&&ISa61N`izFmsz&d*_SWG}#h@cl>6UO3#Bui43d<23-DIe(
zB~19<T_<f7BX<dN*``T=shc{U8W7&5*yn%Gc|EVtz>=Yb<T5m-#B<M%aS6b|eu}`g
zj6O^Lu{yU_cgKKnp_i2GBQFXXc3!-7=TW?gFn+ky`hQkw{%uq3%8Pn8ofhlFDEvuG
zfBS+-^<1yn<NRPTtlf=>NKIOF)?e+fw?udAk>OYNJx>fPd@}0f&8~!d7|0>FZjm%0
zhc(nhH^KT-p%QhjMEZ-7QoeQ|+b%ex-xI=1tZqn<A?xtcVwkBJqW@K|)t}!NP23+_
zreB;}@s&~FEI!o^$oY(WU%DCZ_F*x|u-1TNS=ZFB!U3idK2)!9a&Pwe_N?E7FtSsu
zGh>&6>E&#<KnH^2Uq3`YxR&+=SypVE1K;vC??I2tTY5`o`2%3E;Y^{#F#&wOlBk{-
zd;p^8z-sZ*-7hazLbI*LxTw9Fc$)Rvy5K(22mBqMTM{%OK7S;1Z8d*LXFWffn!6j%
z&26Hw{R7O&84d9#7)$b_T41H-k|0U455&OVRv{dN%i%+)&X-d<ku0l$!f*WZoNI(t
zd=)hsr{dn!OQl7w4L={-cyddT;=g)gY;M%7lX>#CXy;^p`*pJpfn~-#JjZWvrZ|`h
z-GE6%?Pbjs2m=LYPRC59nC#qa_Jyx@qUgn}JNTrRPWm`QHaLBLhSG?wnT0{P4m0ns
zsfJF~{Ku)gI`D6&F1TQcO2)NmjP3I$nl#0EJOm@g$sAbU_mD*wkQ<BT>|MNZ<6jzw
zUSIMz+8tigAl|SF5sUB$Kmt3{?fyAe|72wURci9$q0_zfjrcoD3;u1x9~H1Ec=}#T
z(1r##f_B-l2c<^%dgU<m<@DnDk+`DwQ`$pq?B9(7-O9Ni=q=A2CvLRGu@A;JN|af2
zz%)-Rs?Z8b`;DuZzxIFiUpQqLh_!<~^Rt%HxSU@4W-?#M>>*AncO2QLFF#=sXB1)@
z$(-y&l4{A(Zzk9ND?-n_!AyfDpsPPWaXoU@$_`|mLf1gs!<eOQYem0sInJp?MZb!J
ztd4Vn5upy6^HZ(2tGe5Qgx$|FS?hQRUTuNz0n9SV3`4n}Mo}2^y(!v6%l7^yr8NWO
zbU|6`O^vpzmy4}#)6zOom?d6hE|+ul%Zeyd@*cjhb`?+c0ix?n3dhOjvqANl(2DGR
zP&Rq%kR-aqTNdg$F?A|j!xp9ulGV)xdj<qHY!LQE{Q!-!B<Gg6^zrIr%Sz!bf1OpY
z4A?O@mS@6u!xI;Y%Amei+clBZT934FD=>CuST(+*b)LdPJu7#aCXik~sF)EAo4m7X
zOp3o(k-sl+L{Q;MlJIDs<7v7u{2#afevApo?5yyw6@?p4(^8$Bn5(QuRdbX8KXA-X
zUA<&Fhl5j>H`t7@uk1XFf0}CnoQf`ym})dvWf~Ht)N3e@A`{UM;gSFGod3Pj_>uK4
zqDANLBLb-dAbT(gVsFF}G<GtaI-#AuY2%&zfk`Yp0{|1aWB$TIfwkum2Ou-J-}Md*
zUrrTb8PAtEwiaGV7~iAQzf(Ca=4&akePsHprze9^f3f6wtllulp%*ju8l;ARGem|i
z90UEvxPJhv&7yZX_!47R4|L=n?dGd{i)>97&Pu?ttpfD<i3a=N%I!;umaW_Ejymp#
z6131BJCKfcBlu#uu0a&$_nK&-e@l5>$Q_W+hJo5-$K~lgPh*eK&i)MhDwurBtRmq|
zj8VQ97OGIa4sy)kJ(>)Iw<Apc7l?|#(qaQLdLm8q>Vdr@_{DghS*(}vtcKd^e7EWL
zh8yz+Bp{ZJ0@G%Qa)&_)ljN*NXz&V(Z<E?c%V~f!+cmnsM&bFm%W8`T0zKC=RB@%5
znHD~Q9#rm5H;D3Uc7|izI0yWH)V+07lwJEae5;@!NQi_=2q;Q-#~>*pAShju1Jd0f
zNT-O>p@gWUfI~BMcQXSFoze{he0#j>37_wM-rx8A@xE)hmVhkqx~_fgeV)g81SX~W
zxi7m{xaRpD-T7p-hiov;>x_v+69}YPjfA?nA!x^zg)S0<=^s?W?imHcW*!MzB46qc
zn}RN`J7RlKR{EoZ>s&%wbgt&-?cCeMXmB{$->BX-n{-xn6pSz3sY&>*0omV~R05L@
zJ=^E)j_sHe>Ft-FC(=zUQ#8kGK@raw0(f7B?8ycVK;zWM*8bXE0^0v=cR?CX_LO(g
zS#!ygj)XgSx&&9IBN{=1ASuQQVe2z<H%Oz@LKT3p-Xqe3(0IZ|dXr8x0O{$A&u%B&
z>l-|JC{oq^%b8FGIp_^4)V&x>|2V6j(iMW!QFh=7cZ7cJV%kxiU1V?&_9e1h;@^_J
z_Qw+Z=viKk#lABc4v?S(HA+;hzPB4q3kZ2my9qK3e@ySxUwkaWJ{MwvrDLAXvhH95
zAlE&zF`_qj^XN83wh*u{(N<n$AQ=}gJqEx{-0M^Z)xkx`VL6jCSt_UJJhoG{MV@3$
z*Dqj`Z4=RFvc6E^!y<2!{^WX@(FRQd5<$enP8?Dl4{6_h7}uEYxcb_aq)ry$C16iT
z{Mm|oVeFag-~|>y#2PTGeRV-B!W8%H0lCvfd9tQQkmXikxAbJ#wK(#;riK>@QTvA4
z$7)sd{u$ndHw4Xk@34<bPQ^Qd5o~r(1HdvgqObI4vwf=wz`<W=B!{-sor&yp&dx%k
zS;t^<?4Z#*4%OIXK`1M(^uv<(VsE!`iyYZ@d(@wZr#uTRan_a+d~hFcb|}|{_;R#~
zmi{v->Weo0)32vCz%XTVv5uW`VCBL?=<rFGLy0VcneK8U<<Ly~0F<GFxc}j1P*c4v
zRiwE|8(7~ilQj&i`Voio%Pww&+u@LiIsUbTJ7s0SOMvSHxA6d{C?@@Yh~^^qq>)(3
zC9Lxjx%SrRGu$n8gur9UV4kIpNT8;oNflBQI_B`d1e12%1^weao5e`rkBACCKG>jP
zEgiQ%Yq{Je@<wbe(k0x*4h*{X)~e~~v0dj<ZqMgMT@0PdXdM`-g=kcMxBB!<ljQQM
zBc_8{V`2{j9ctfvZ3UIlIwnED-R6aE7q1az=GpIiXy93-TN|+3Bd29h=vDw&)@63<
zPd!FwH(W6A$qU>{k)6b+pG1r_g6VrWZ2($0U3VYn5i#sC+!o&;riBhcS}(`K!QEqA
z^(3i>E3h`{n^Ckh8DobGv6Uar@`?FKiFtd9<J`0#9^0mo#&(O0|Fu03??$AwlMyAh
zjgnrniKpS&snizHP5C88=tuxw0HL5_mES3IM85U7I)Gr|Jxz_rk5fz(z2ydG?df4%
zUq`0uUY|NYM8Bq4WzcAR(Fa*3x>RHvNyOIZTR0CAkax8SXfM-60`V9Uucn`jebk-w
z<8K!kVf6IWWI_UB0REVqAw`Bp0=`c|6`a9_Wt;x45mJvVq<+WJrDVbxc~oTmeMrp&
zPaybKLwzggK{Xh}as1$bWif(xL!$m&jMe_=@EdFf9c7jfJYRBcUW1_s?Chrmxvwz-
z-###rB<&JhOvcKy^$)@hOPF?OB$jnO(=f&HrqryBWH(i2Kg6cia5ReM(bK2J*Pfs|
zZhc9jQC?e5O$OEIY)rp7bEUm0{p=E<$P_ktmkv)o7mc=qvN}t6@w|hXx^2DRl{H`c
z!h0LQ=|{hggp)c<8%$ea1sj_pex$h^bdFgQz;Dusv}dMnh9<0QZhF)xDx&q-XC)1r
z!n(_PkAh6tCW&+Vde3I}wgLPv<8`th0kWcE&43+D7T-e=%W?(5(d+R%uCj^Ktpjgp
zB}MRHgH2ilYD*yJ@ZLsCteb|+NBpg-mYuiRl(rdpwBmvH*W!eNg(uz6UkY+-7?h|V
zg%#1};%^YI-QJ%TqKw&46G_w1NV03bnE?#{-XAkImQ|&JF}D@Z(Ok?)g;cnLXRz_?
z#*1KPn>4wUKy)De>3uTB1}6bi&2r1#Q=HRB#9j(h(FgI_C@V*FThij;jzN+m;;zw6
zL(${lUX%zhBAVo)ImIBZ<US>ik*%*-afw^#lTC!}niZRgjvvDa8)>2~5;OGp$89Cg
z3Hmp7^8LkP=)E?9BidX5epNOO+1vbJfx<^m#K^&6x&*dS_IY3LwDM~X*N+|Bi^25t
zWM8jmIJ!*)s=N5}P0aNiT1A=ZOX|{nz$JS@oW?N{9z|S&H%a5a7jCdCC?{VEsvGx?
zWHK?EN~Q|RPAh}vO1^XSjz_eGR!xkzTT30-#ookKd6`lOANu0PTHoY%%O+;xhUgju
zka|KIDI>fetcNBZkU*(prMoO7#<s%^Xd`8!K$Di)7G+wWOz&yq2Cxlz@P_QHu*8+>
z60^^!HWM~_;>gV!QAj|JIcvCyE^|8e&zqCbvS9^z7MO~dMAB^6x^|CZ;zEx$H0CaG
z>h+Uyg8KJn4Iy{SV#7h|<|#k5;;;LqiI`VdJ`=^0?=1ePl{e|G=k66g4a9i?fPr1Z
zTVyj^NZ`g)H5Dcuu_=cPz22u>DAnLc5@zFnKT_3F`*|a~#&m5|fHN8LErA7gL*TU?
zgH}8|)<!wkR3l}F2;Sk%u%<R>1Fv5o7<e`Z$yx?-grNPa?mVU|unTgHJeh8wlefjG
z@H+?D^fXn~w#<h{A%raw*2F9!7M<~&lL*?qSFV@vLV}Zzcb8`q3{~+T>Lxa9>$Pgj
z%C1~7t)pF+)FAUAh}_B}*3_uyQW1Hgm2^kZKQS0xm?~RU2!{Dy^5)l-xUqfa45~Ea
zDTN$AFJ6W;s+HiexIVp%Zl$;4>7uo5(xP<B61$wh3-PzX$#6|w1xuNAt?w;O*f!1(
z4&fDqlUW=~;sh8@68nYLzT*gn&k)3ZwA#UEVZ4&gkve<&;+wFZ`qpBYKu{CVfd0rk
z?PDYEz^J5&>2M}d14@wUuTxPXJX!=&ZN5qZbdg~0aoivpsp+MV;dX}=E1`;i%dx*3
zcLzjPw-l=j$-eTxCyF51lhqB}Z=GQ!JFDN19%_%y?!iYs{4_7xMVOOpp5J$nu>?*p
zo4sRjddcNPZ)H!4sNqGKupe_{T`Ykde+Fz51iWCku7tnYps~=dPF1A)4o3qa9tc$}
z%Sa<9Vn;?6z|W5Lh`L?l=85j!d4NBQ<uw<|oWIocx`cIGMmU@O<F)K^x*2iGh_YlJ
zV)-ENq+P}<eg<X!`>8|C1Ka93`jh3##CysXdN~0WB?S*XFKyHc4Ej{r%w4(F#?k2l
z-ax+S@Z#vzs<p36_IFyZc^*y6<+1e$#DjSH9{p4?Zk*1;>1!y)LZ%8pt2=Z_-EstQ
z^8Tz@zXbm>IqwFciTOK|=$%{JZ^0N~-h3`TTTH^;Qi%#-|DArP`z4!FC7oU~Kdf!O
zlI-Y3epr)+H&%<lW1ps|{3X01oeV3wMhq+CNDyNF_DVAoixu}L?(^oTjUSp2J|;sk
zqV0O(X}<Y|AU99hMd9$jhUWh`R2M%XiuX?+EkMux#rLUT^aaewiG-!rgr&AI2|w-&
zp~X#hPzi4T({2{*L;5D}zUbnn<IuN=dKkv;HicD8fW%jTg#M_Iq2B=}Fg05$Z3H#E
zR|>?Nsv=QW%oAE`6PgI1!`aEWmy=oT4+)RHtwpG$g?YRUWEBxa&U7X;@j)f9&y!gf
zgp2n{2v>TnIC~SyzRuZ@^(V6ysgD;oNkZ{NCTSEKBe<CXI3#?Q43RFpOTBHPmYB?H
zDumqcO~|^}Q<qyNpuKCyaFt#opt#A{rG?7n-AWFfcxMBdO)@Lfd(SB!&~FHnNW4Q^
zDz^o3%9GrKXt&xkj56v%zlW&&Cj3$1jh=0$iJ=6By)0DhNnVxYoC&7f^Q&EriU6xj
zW}On$K6HZF4jEJwRBeaa#KGPybzb~rgI7XQ92l!fzV(Q415`va)Ly&F&|x9UwQ%=2
zqv6S9R*gzoYPUv6{fVL94Je+))&oi_WMe>g<8|vFqu<VISPunBMvwU{hVelb>P|hD
zWF#MPPX?}QGVYM4*z8%GCyz=b2BfMJTYpfBEBRQ@)TwnWwmA!?pI2ahm+)@miS#Yg
zvCMbY2wba)#w#YCGt-^@3EZouHKK{%$)FO?Q5*Rnu&4&|u^RZWbwgD0DBX^q9CI8O
zX(i(pdeOWig-sr0BzP&PQCn=;l%THJjR>FlFNO-KWQVKn-ZP~gPB3AH(pjNUoB8p3
zHT?pyu?$9#(~nlf8qUIFK5-MBWv*-AN6?iLKG1p#XgxNn+s$U#<7Seuz1!>u)CUwV
zvvQ<M<W{0soPDBgL&6>dq~<Lla$z*;8`b=m$rHB)8R=o!>LbNKo6Le~uMcpZ)#q7D
zZ+8fquQx;MFUx9c0{hSglQB-4qz7TSb(q2t00m+aC(UC~w>>NKziYTP*U2m<v{msU
zpxf2P84-Nj4#o>jb$zL}of$<seLLR$=Kdyq;7yBm!KaUu(MdhVS>Eyr;>@9M@%)l;
zWSz0bl*-3)BxVlDw5uv$h(DO6Hwq3R5vFXo1r8&xNsInj;Zag`ooQsbrtF^iwpXve
zPdCeJb*n|U#Y?p8(>QqBhJIR5H6E%RiN6r<e?n&V?|{jh$6&@|+P~rQ6q(vw>B9i-
zVuD3Wk5NT=bzI({-wcV0dFCvWw4lf$-uuK&OaM!Neo^yc@h66_t=zbu#kl&GWC-uS
zxyP|Otx=JC_%(`6vcUhBEt=FFeo&&?(?7@h8&2%y#<_#vu+jufYL+d9S!_bC%9Gkp
z#J#feS;$MfqgdyD9}kX^a!f?M!ZOWMg?^11{z^Fi$qDDti(w(Bhl<Z>@>FTsBmN}d
z{NrBo3TIFf0F>@#jJbdR^DfcF;FHmdyaM@SZ?EQf1L|%x5`g#<P2~EN)~ln}a0=gB
zjdDf(C=%TU1|I0)&NT1iFQ}q;tH61%BUKPyI5t0q!Vo4ojQ}Gaho)LpX2%<bu0MOj
z*Y;&4LxD-9=|S>uJA@0gU?<E$f_7m7Fxplzys9(;WSSbXy>zqlfdqu+xz}gff2R}u
zWMLm*W}*lAka@koiS&sQQ-E6K0>Z`pN|$6WYWU3#<fyZkf`hXhgsqP4Ee~Wfnltw=
z5Q%5Ho$igq(Y0vNr+)z5(F9<}nCNjlulPa&<SnuHYlW*oQe7E9_IvQmUDG@c*4K6-
z4mawq&^Vr+9M~M$m&jjMrCC#nVUp#>38&z!Ujq(`>mWhspl?9E54v?z%(D!W?VT8%
zJJWgR!*T0x^ftdj{s7h|gLU_CRpxzG?(d?MpVtN@uDy;&h6c>77?roh-9kA=N%i2u
z;B(nA>ylI8Avl5tPM)y=htFp(N57m_GO!GwjqlM;KrYkkg%f^@w>^SN7)IBCIpmR`
zTxCYj_vu{x8nv*k`ePyv^!YyPGLe*;$Fy=)cR_tLFwRAgNz$!Klk%IU+<&;=F}RO5
zM~rJj|KWE3b@==@AB1AwB;p&SL4U8ZU;n0mUYvh@4^ILp3l%QsF8%tY`CqCvKHyxu
zjwXGmZ1>ks{NMXP{QtBY@ZlyF9XT((R?`^P#h5a=M53JZ|6c#H6oOfHA>Z@4e(#bU
zEiR&D^eb$}7bhz4MHzRVRhUoQh}m70lUCD|{%{XRDq1dMjl1nGL*@4<P?Cs$!}2~m
zg3=fi=v9YCXngyf^_O}5DLg4Oi&p8IB~B9c?@F$J#5lpnX0+fbe1B^4fca1S3D4Ii
zoqb$=eL%~P=?Nau(&FU`v6x*O-{cY*GVjCbE9~`j{J~Q02kR5JZd!cQxubHjoSmkl
z5sYuh8yTZ0^rg{)cX+7GZgw1xy=n?%tkRZ(H+S&xou3o=#VUGFza&uc?pk9}q&QYh
z2j2GdsR2oV&MKi3kekPQG(PBW&)=*gtdUuiS)sjhaU@TZzUEyW+!${D68qFp6Yt(1
z%DR8O3KPs#*sndX`Qs|yNPI|~Vn05DOMHCS!c^t1G<+tJlje<X+e<7_j|&w?sBQ+m
z_(;4p1~Ot-K!PeM`*lsjkbC2s(+$iBQ>FbEbtgbs<^-C;ZytalJpl$P^8l#X0F+iC
ze$aMSf~ih<6Pb3d6)`U&zyDP-MpFVpBph03e5!<Jg6aGHO1|qyvuG0!Vn~00m<m6v
z!HWK&z&|nfDa9G_$@u-D8~^go+?ni)P{vQ2=^bUS)_h!?Vr4Mey8Tn)wU8db?uM4O
z(uHMWuib#lAf0pGXFPks82#xw1@Fsu!oWw=^H$Yubsf8av;!XAK5L5d<B1t21(q<Q
zFY&*8|0@E)3#;3p`{j>~O}qi?EACmU$BG;OHoJQKZ!>V@k7%5*@;V8PM~>}ZJ~_&!
z*%!`^=11Ct8z*R;v&QdId@ZrOBXleha6#zO5#A?j6CWPFUNQ6ld7))N)b6dcWVPD}
z(Sz}4BRgx<hoD<?(dekRAA5D-Qk98NE}desKwtIw*@-m>TC@jzb7yVC)&b1j1Agp!
zn=XJVyeCkn-^Vg===>dpaWGI6UgN>Uc9&yPalJO{u2_TLWce7D>V0gEg07V>rYeNM
zD`gPbN;D|-KH0V*nJEAqZl3l5U&Eh6ObByciHa2=Kkf4jtIDDFJ}jQlq8Qw8D&gTC
zRzNV*#fbQhX+%3A=jfE=uwf;|Q)@e?yRl8*o<I5fu>XhIp)TfV&Y#YI{>P^G!QkQ$
zmQR%Ep}(-_Imvhw1Z`7U7N_soqySV`*1eR)@B4p1x(3&`Z?oGD;G3cHOi?$_*-Phi
za>~LAgh#9jmn*W0T=wOJubwqW(QGUvJ1Dw<gV4iHi~b3~0Cj^NN%_Ig`(C$T_~XFJ
z&=ey}DFr;(L#rssVCT|NHL1$0xdMh%OMtWHc!kpT)vCEfUb<#;ISg?G&bx)CmaZFU
zT5E5R7ES4Qqx_s$17KAXnDYj31iqiUt>iLX#7T?=Q(rc6^t=@^PbD((1M}0;JGNkO
zTM4*YkNn`_2o-qJg-;i3zFGd3Nc{&+_m5Q>*hUyJFW_^h2i$+$wq8!(mwxb)?8PJl
zxzfE2K9`nccQeEqY|mpgqI9Bv2Y6iTlf0zp5CiOlQY;_rVy~Hy`&F$Qfw!$~F*)jW
z-%7n5rAu;#*-aMb3h#gWfnSFL^yXDia0vpuF#Bjf__>z!gSb+w(#L}CQW?v%0q>>?
zf}WRzGX7l{{?j9TlLVH3IWg14KhFInor_)9(ciG|B7-Yh%jMh^W*AZ&ReY#U(Zwf6
z&>fzZ;o6w})RN8>jR&Vnkh8f`!u@IYy`n{rF$a+opB`;AneS>-%gnlo`CY1IS=zkw
z4oQ*D9>^(w&D2Oo|B{J$lWy>A9zV*igSGHe<>rdRcyGi=$d#?dk1VA8Tp|*hXoqTs
z??uK~Ph80jrlU%{CN##yyk=X7;cwXAHW5-+dmJdv&A{U{f-_QIXG9Jh5<_jbIB2Ue
z6WVWG2w;>L!}NrnM$-*vtC<xY;3`=$ePldYU%X{xRw{D4z1hLD)guwF#}<EjdybYh
zLhE$KtsEExkN19*Jl)R(o3pLv4PUu&?zx-rr;)iQhb^mI{$VY|D=df6rM;hAV|TQ4
zT*dU#XwMtl?(6Z>dIGGuk+cV@RX%$#&@EZ*Sn5pLu8-AGi+y>k=5+=Bo~A_-jvcT4
zlVNQ0K$m2rJU>V~Q9%!pk{Xg=*&Yp&>=v86EqxLfTwPuoDBDoanBQByQ&r%PyJfE^
zCgD9}@{068LWGv`tmL+se~S-5MdVqPM8x{*16<EM-k<W9C`kBMU4g%0b8Gt=W!yPv
z;wm;y&o0uFeUmLguhoVt?{4-R_Q<-X0DGYRHvGb3N9Oy?m0^`)B@JqqNwQ6*#4SS8
zqdy6oct4Lbk84G%^X&W`4f(nq{$6kHK5;)itJwG9pdPbs$d=GGCmIB^Tb;5u;`xi1
zCtUt5kVaxc5_ptPvICb_|I3mBxvOUq6(1?*SXq6xs+xONy&(0)nX-twrjFa|c9zbS
zp~ctSaGh@ETL=b<kkHWRv<l0jL%sPWZr{m!3Z0|(d=yh6DTWg>FLBmzDzox>_w&G0
z?-{zamtI!ro7=L7PxCif&KhseG(#*XuN@K2Pg79#DCeiI(l!?SQUEG`Li}c`;B@8r
zz2A>`GY8Dd_Tt$xav7FeJZ=gVT%`WtJZhxv(nZQ&R%v~EE!D$5mo-v={{Zb!f!f&D
zF5SOuF*@_;$WN$8F?sXjGB$nbZjpSMfi{5?VT9Y@gwtyE`j?h82b&tcNf>GBL>Iib
z$~uY48wjqQw?eUFiG3;>>#OHy<1JDJC%Qds^wpC1^&G-He)!(6rdfUyH8i4G&e9nS
z(Om%^DdgucG-^pr--Z<F?fg`eg3UC+%e3^wTiE@AQB#-yl`v7uIM0@@gEIYYt{CJl
z|LyW+A5IQprqd#vIvQ;`J<!BvVMPiqP{1{ufJ;poIC94gZbpjGT{jA=g`lSit%1~q
zO?&VZ4B#e!USqB1bRVfAg0a+;G$V2W=MKyR?6LBCfI{@?HlJlY{0yjA=1Q&6AaG}V
zK2CofoH(NVBRUT6>jPsj%?J?qdz0?nUU$IvgJVD)JZB6YK%b1P6=)SMlw$Zsyu%+-
zvckwiUP~8Yvb7^c=!K?|*K#mikWvt<UP0wD-lxFXbi>;&a<fhWoP!t@YE7Qe7uzV&
z{Y@ks08n}B02SP2A?5*YJnPoL+2?a&5D_$^1(S$*I?wL2Wdi7YF5{5%Bc;+IHSKax
zcGa-7KYV@J?bDWY?Nb`8WT3DWDL$0M;i3UL_qXIGU_xoX#carkAdIxK2{{KUoh@(6
z0Xf}%83MsMB7QJqR4rtYw!U9d2h{ML81wNOvCkwL4Xzkm_h|MQT=%{cN$R!v2kE*g
zU@qr<yOxN*1(2#yr9)-<-*?u>^FxhsGXRybyHpN<CVM#rHQO5^8>8LA8$VXRBZRdg
z99R#`o`DHY>d8*p`HyuVYi?0%DrtMKrq!1BeORkTR3ETOR7-cW#2{yfMlRqNmx0a=
zNgHT<VmZl(;M+I?V;^R+QS<XEytxaU>#{Unb3oXI$~5)B6cHGe*@~#h?JmjSO^n_N
zfR?kKUQ;a|nBcDyu`|$bInm36+k(JbdwuVttkUc5gUYhynx+X&aG#F_NnjZ5<%O|3
znADmE>Gku#)|Z3dxRvPO0DeBvvIQt`m7u#DZ%Mg96~p#p$%awMcJLhNl!px<n#TH)
zr+uZ$*>eeeT((o<D8eG?HE?{zBMyQ@_}$SM(Hb=<nEPk(SWDi3fa!_HBh&t3>tH6&
zVwVUio{94lpYxtmP#WSt^rUHX6*cXofJX?=*3w6b%-kHHR@b)T#XxAs-+m*0lxR_+
zOF9O7JypQmP8`p%03^$z)BOhryC)?vn93^!jGk*!IP$mi&j3?C-xVcKxF~Sc<8wAE
z6GgqYvpx7ssT}n79bVvA$H>NZysjZT1~!a`ppls!lY26YyawnyG<)5QGdo|aycCGg
z!2L}=bpjN@Yp4T^a2+p?OxuzZa=F*#U=7J11I^fw{T!rbt`4{&Psy-emdLG8<afqp
zgZdbfdcI!@cN$O>&P#(^0fowRv_xb$goNSArX6eLkD|y8s$J5~QW}pRoaZTq;6=8D
z9}g9bgIhm^6T_SRAYJU1fu=E%;A!Pce7xn=2fsScdu6aJFG&6aJA?S}c(ErV%G<km
zp$8LHbvk{%Hg0ejq+h|T<+>G+&X~P&o3*4B-Pf#bNwk-m>ryiJrYR%k03n7jh(gH}
z6C?_9b$IsG1g#T$9S+>i5&)}d$ZUyWRqFLZGHascc3XM`pEf&Upr6O$RUbaL0!^N<
zWtUl{E-Z3b!)M40;d`m-T+wHg&3z-uE=%CSfEv<jWp-(xRH<ZUrQ`j??Ad<hSu^TB
zRZ^sM5VeoEbiC@YR$owp2`s@gzUJ9JqiBP}w4|o~>EUwd;=ribCVDLu+k9KoX^qsv
zJ+iFk=qPMFuS3c%n)QPWL)rFn(i>)sLXMt<SPHPz9Cs^Gr)*r&lE!{9tym7xtb;NP
zfgH2pkhp$t;NwA0LauSTw{16iQQHwj1I4Zfksv48XQ=0sUw-{Ur;USON=Gt!=xDWt
zK=Iq9z!})LqHtx>mBfxRcXbdUxIJ_vYf>P%C8uWxh|!40W~@p&*eazMLW3<IiU{ou
zKI>>FifYL<lfqQ2#$nZG#egCu3qL>IpeLD=0t+rJYQUeR56mrTd^*#T#P8BYz3cIp
zk7gB*txxLZ;Hon=4%<$Bk88-a^OV@`5;X&L$P@g<qk!WSFgomTt&c?~7G|?xuOxeL
zXn@gh$GV+?#5)y{!A0Er$xn$Fz%Vn?G{s3xKaGJc#PY(Zk7XY}mz{)h{mq_0k%Zu^
znWL$P+8Ggj)aI|};?N13++1~L#uo=$8{^K*{JJ;eA!K9uPyE%GxGkShkb-zVT^Ad5
z=^~zJP2>8pRx%Vl$(S@ybSL;g6oPMKyc38w&O`2}QjFC*<+-k2l}+OMR%{cs>(s$g
z9KNv%0~>j(f3uiAAY)kafMOlx4FT%&(8G`hspnY&OP7|7Tw54r)5N^ci{QA4d_ttW
zkH&O`Cn^YKxc8}NTB93c5$F;hc<^~g-zTJ{CNYp@VB0ZID(s3I!77TzCHB(Crdesn
zLY$qx?g527HNc|t2S>K_R0xho0HHu+1qFvgpp+jX73L_%)MiqHLa<MFXisf<>}(ln
zY;oK;Y0Yrny0LsM_^JPDW}LJ-eu9_>S}>Y~r-&%scP0U(YMZCZrlz;53?$kyG>%lr
z{Cz4CU{+%qdxxIn$+5F6s^i4Sud~RuM1DQNmOBw>bu(_7#@8LZ-%a>PeKSZ5z1dK-
zSTCLGtoj_uz^<zqSvvznl1$ZcC3wwj2*`#IV0k53zub|ZZTpVv){qfEy{zs^esg0<
zzl_iKV_M8q&*3q+J!FZd!MVY13aNt9(Owa!n|1Kppi7i30aE*8u`Vg^vudc~WBn$N
z?kY~J$SW}-8?rADOlwR^@Z0T%5;l-P&V?;qU*|S~TH(4;as9^R@<%P>uMQ$>-tH^8
zceZ>MKDJ&6^t{4hmro<VSxS>Cwcn6AXn8}KCZdp~$m&^PK4pVSUHh}Y@yx&b>JQo^
zwT8Zb4THq_>3Uh}^oc60c31aK>wBJ`ehTld{?W;0fIMVxmXTD`LhS3?_njKazMsJB
zVJ~WwICDc<^CO%W^U_BkUY}UT!$_#Z9$lIqd)V?rn}HJFE*IAQ_=_LhHjUn}k*7`{
zXQ$C3!%K{<W_=ul`kcB)lAoPY(vagBdOo@xZNzGKgZ2=wi<`yo?G>`#PA0uFRiQ-e
zbR0XfFmv;Z6#iH^6GTuG!D9Qx)%)^|euw@dIzFd+n#zt#QBzR@L@7`6%OpolhqKXi
z#<8tGK>4xbNAGXKG6h5i)tGXIzF6aWLRMyLZg-PSTz6M{RL><kiee3o?M19_Bz{u<
zL7I5&CWJR^eG5q#Fd|^xZ*^P>o8(dc*b_W;GzLU>-mq<CYU+h|YN?@>*>k1tUMOIZ
z6{US7F=`Jaq|dP=OX_zvjlYR?k@n@k-Aa_suE6qH>(AdVGJYr&438AqZVknDEq+VV
zw{FI5x|PS6uF)_4`SqO()3W2VcMky-cB#qSw5z=oRdN;yi+B^t2aPSAQ9g8-{Jxnw
z;9@E$#eh?G%MOqrGl34m(~Ve)K3yC?a*0Xc_!-s>oK!!|_!?a)3xA9X%*fjj+kW^!
z0Up9_;VX8gb?j1HwbU?7cSyfp<#o+Z2a9%B)|9g#-A48S7tnFB`^gNxy+=vQ9wF3B
z7N0QGs&b=nHbpK}pEz?n@@^ACgu>r`90g)J+E>C(f}fTL-C^h=9m?M>vDzMDn#T$E
z1|bAe!G5m~lbsD?571T%7lVK6x^9W1x3q<kb9D;o8prZ6&+uAw>TL!E*Cgcs{x$$&
zMEojUa;8{}7xzYz=GUG2$H#D<N$F~ukNFivx&b2oM%AZmYpG%bxl37RwwEq&7|Hed
z!Qbdd26HDs-x5yN%~d6D+%3^P7CG+Xpl+HCP9s;P*(+p8w<@itAY1(WYhLWS5`aI}
z^@Esve}cy1g!H{P{orz)p{Q1ivt%Z-%JXvNbJE-B`Le;2hnuBMu%ygM*}ME(Eh8E4
zhczbg7OAORl%&hlNYJ#7=%(;X3a^Hd+8jDIqgteH-Pg}~B{zdi@_-qFDa;cI>(}lb
z1$8!4{kYHhsrSQc%(sC6;LsrYQr!!UyGk-|)t^XtvOFlRjc3rBtFdZWL?&4AwwWYI
zHu}E7>oc<N<Dh<1rm53;{fU256JW~looOsnHnTkClc#+8c<EfRLTfV!n-_U3n84~2
zLCWg?%si5>!E0pz7A}uBbE>gqY)5VIA`xgg?#<@OZjLAds+D|W4NDdMLWjLZV==2;
z^{Y`BPS@4pHxR3<wA5Zy^${#k<R|wn(hc)OK;FKxdvWRIQ3m0|PT91ep(>s@*&C|Q
zsU>RsKvZcLOtk=UB@jx)bmxsLn5MptgBZ3E_Bo=fWr=Wlv~oWT(Z}AYyK@M}<+xsV
z$xgotN`(b)6G@Y;6$;^|VLF)`yfe=ck5oSUZE~6S@rc~;w-4Sd_7Bnc(dl0E+4+68
z)w5VNM%k&y!rALboEOwyRrdEqZQ)vjVeYOPZ-Wzeq;#jnHsDk4^FG{^rGX;hriK+@
zS49g1?Db|cubkp+!Bi^0mZAXV5PB+X2UIcY8b#OU8!G74RFdZXrr6g!<Q~Nc(#3Kf
z`2zA))%B`pQPM-QqAXz)vf2jt^4LsDznZW3TFZkKWUcMgUiSM6Qo09L5Cm4)b-nQ0
zdv@(TQ_T^{%@$ZqfzPp}DP`knFP@9Lyl-}zhs4ii6OH1B_^=6`w@>%IKd<6&KUj@p
z-cD}H!TgNxE3Mx+0$$y#8Kq2p_~MmnC66yY9lyVr<@x*h?a+|mIP7`QHnB3}vhztj
z_a%+XG^(7*oT9tQpkq{im6?;}E&~<bY_M(-<&RNAmJTkxVTXw0pn%%^qRpW9aV!{g
z(v8gg;G1L9MOGxa^W3QS9$@jyUjqo-V;P;Ose%-QDo~wP(6>Kch0KJHtq|BLd_SY3
zt{L9Jba3!loEzXV25GRSLmurzHOgjBv(^l~@(@@0*zI(HW5t&?UZzkcE<#!(z&;eG
zw-O#BV)Hf@d2uq9X(YUrc_&dQNE$FyBKCPTEF@b4IEuBr0O#^P5=p*h<`B+kc|c7$
z%wQ0>bcEsEP1<Ij8{{1Eko((fXm#4Zi?ezWli%t*WQHVgd->kpAAJYQPZ)A*$x=4O
zXrB1O{d&z~S%3KwTy)V&ajPd8XF(hDVFz*qt4^QSPUA)3lE_gi`H{E3;~#sAC>?Jq
zh(onO!}@nqG#;3A%Xk>0KNMst(c_ON&J&^gAZjPk1m^Ttp^6X1a>~<JG60z=JR^@I
zP<8gJ$3MYxe}bzMQMGrC$Ib*<|CS3Y;CPaFP1o3`4JRXyOf~wgPAj`6IJ2p>2DpW-
z%d<V8X!y8p6zbj0v$$uSs{|r1X-dzheLmw9H>=!shkK+}mCxw)8?F0>l^Y}#Y;+5p
zHx_AY=ayEV0Em6!nv1^X0uB6g?3McJy=J8KvYVO%EUN(dV!&xO%6Y{N!r!}>km*Bf
zj!?7fhMhUUC{`@B%f8wrny{6EhPMLaxP&pBFfmXVodXejs!+AFak={`bCz~pD5#d%
zML{Up#S~2zdxn<}JlJ$bVNTt!K1bMm2Dinly2|_vD=G^o^y^AsxUZ5{-1KOq)J7lb
z2f>&;ybs{C3*eT2+)Xy93+lx7^?U4=IeO5r=5>HgIK%RwV^-Q#4nOrt_iPh<(Dh^w
ze!?72TyGp$x0hRKuS^+&p9CfcgUK?oEFqt@1ohmP??KfLk~MSfC4&eQ6b>-zE6UHs
znI7Q9PGQ{?CQ&HIl#Xtd3#)(?z{RP3!<nJX&UBgCEoY_zENtpnhTm*qR50PGSC^7Q
z=9vWUi7q5sutAYfq2lI*s-aC@FFdQZT;Xzz2J58=CbE&+E)l4Viq|^j4KNNRKUGZN
z0w%|XYww+xkhDL6zchS_y3f#n@l89c=6UG2tv0kuw?lNDD=eeC@q0H*p%5}v)Ky&+
zD3W5kVJpsrVuvp@1tvX;GnW9sQ1uov`zoYH^chmD6sW!oXC!4Qko_YGyv-$dOuXGr
zZXpsvJ|}B-Gl%i7)nn;*!$Nl5R?U(hdDJAPg)F3mRE`6xXb0hgln~9IqpBww>z~sA
z4YVu3WfPEDZNYDFkah&FgP*~EJD;pMvFG%ZxK;#%oIr=KPTWuPRB^8S{gt};^OcIf
zrv)|CfYxtM4Kd0hE^JV2ytR6`+L`#t<~HqC-RmpDB*I6|F}XcNpMs5mLSMuol<qi@
zW2AZ}lzE&FY59y4&uEadg>v>u?!uGag$6S4zBN{U)eYF4ZE-)*O*J8~jE(n#OKF?r
z8td@XRcQn2In|meD9>v)qOw&71<j{AuQ~=PC$^3y*bZUvfCPiEuTkzAWPLDg6Z{UW
zoZG*C@ge~(K9S%4&fw>U?VGCj0Z46$@Sc<4s|2wKuyPEjFdW`qQ9P0_{|tr^yK2;A
z4+ZIrzbtgV43NZmjT0E~Ewec9BF1TKq4T=NvY5nGvU8#VRgyAnT1j=>GmTE!2d#eY
zvIKUs;Gp;+6LD+4$#H8_nkBqtw!>pm7wD_90<Ya(Ao9k)$(3n<4?ut8nLV5{#24Lp
zgAklySjWCZ1{j$!hLp@8axb_mlkVxis{{tN$yPqgC-<=ehA8W){oUp2FwvFsO?0q`
z5=rr=VUnB4Hz;;=GSWYTCAE!;!S)%P$5_>2JNwFhr#IZk^s}Z0#^1|k4nTeR$OZVO
zxLqLMxMgUMCK9F3yeaBllCj?9uJ`GcaMaU3woVbPEJ|94G_J3N=XdG^NM6nHie7R#
za3FHq67gg5FAVOBx8PI*Q?73kwVD>UzWXy9>J2dY!L_hZu)G5Dy`+26vgfauw@-08
z|9r4gFSiyCCr!u%Q>^6ILGQ>4jdi4=bM~e%R)haUUnqJSqF&APEQrc(wzxFBgcQ<7
z+>w#mloc%2QQlJ~?4ZN)$>Rz%v3n2y)f``(|L%+Pr++J40J}lRNFLP`t5voV`@Q=9
zs;-kfsx3QS#C^7SMf+H6yI<EEuPR#oyqqpiZM;8@)nlwXUTJv0+Wf3*w0|V`?D@pY
zm^BLwUoLp<R&lZ>>9+O(wfoXXF*aB_cU=rhtT4l>dfS@(ajFo4D{`&L@?}az7CSPi
zXg*G}t&K^qEzPP9$09cMvG=rp-iD=tn9e8av9GXf;Os+orgC$tE%v=-RN}&@Uql6$
zMW}hH#XhD2Bguivqnux=yJ-|v(`b8d0CI|4GU>=JujxY`t}s*5gKgLNp+~4sU#(7W
z%|JVzBlPX%^@|x6Pl5T<3{lCnQ3f)VEFtKP8_9!1`-A}a)X;RFlsN16ESn6k2wb*w
z!G(Bwr?7nj+8Y$LEWt-Yt^hqejqMBPG@WMW`c_4wjsnnYD-bH27))mKki+UtSi0>o
z+9`~*3YG(gv#ky}@906Wo!A&(F*nuDeV8~j2a1r1?fBZ0yBTpv&RMW*<}FiG1Snd&
zybR&H8{}TPvfYi7-jN4jQ;hY)yzP%F({%$%Vp`l`ALliXKjeA`bE3h_E6fzcS+`0#
zMxVZ6NYJ4jv0N5zy7A1h%&Jh6Zr8~NI43?=DE)l=8Tfu!Rcabhco8`<{&k4{dAwp?
z{oWm%O}xi{@v3~SrGCEY`j(Xgv3(~fxE5{X$!0cx#zBqOJ&D1Q%7e_l4jK_<V$<(V
z>Y=gvPLBe7h$gCc(#SL9*2W~GAa{82r&C?6O7wXumJo2`Ksl17tRvd(lM!`xX%~pM
z(e4i<p=P;Wj$pSb*idE*KP14)UZa40M7023-9ug|^1A!OdI0F&pd~WAFandw*$;-3
zH~vJCf((i;I+lFU0TQd>)2h$)u&rDs&yqDLJ+eYT>xD12O0jZdz3a=yYwnX^3SHBM
zh^=vIdNJ(kDL6oFNZboU?yk|wmr-DOM1<N-S!##Lz_2Y|CWSAC?VBTcv)RPR`|@YZ
z?ezYv+F%j+<Z52P+e<Dp%K{N=b-va_21~JNg9no|TJFliQjCW9_5j{40}^nHeQe(P
zK@f?W`#T$fR%1v5SpV_BCQ<r>V((jQ7GjuXW~M}jzD6T9UdmsLQ2)LWq{@v=grPR{
zTrqlM1z(3`nyWRVM~&6)SO3<TYcid(vW+tzYN;LpTYLH$RPC)w$`T)l^jT$(X^Zv9
z5=`bgrv}FWK9j~Y)#>@!_co?<?_>-mlE$v1*(qiDc*`sv8Sm6(2J~Xg^r~|w#B}SS
z&ZPZ{)T_i_wsJM;BRa1eawQg76f3T@C?jP`JK>-4fW-lS$yyivLb#l)&naL)AJJUe
zn!R4^pMbC8toz(+u%RdzHWAJl?$QV&iRygm(dcqtqy3j=QoEF3_Pdu{p(gkHP#obe
zF6#$>5x_>7o#c%$i@M~@q?wI(rdbb*@UHeQ8GVZf)he2GV?k5dkDq-pAk#Gb_*s2J
zPw&NU&?RR+d7mEX4WEOr<ZX7IcJ)Sm1*JF16okIu?z2Recv|i!UCT&-ru0G7D{Lg{
zLLgPtqI(NoeOw*v%jA(o+qK*{xZf%qb{T`DNPIl2=~4(;@^KkNJOL(4dhpm>OC&k!
zWV6kz+wJlZ5AiK(+eSSqF_IaxK+vuQhk<&w2B$-u<;^$O=Aa!QSFY0@FS6pG&&5}D
zsB1?V1W}Q#0SmJ}z-FK$%NMq1X{_4^u(|^2Y~i7l9+7jFAXx@zjI0c<!U1C%u>%yX
z#mr;{4^}b0XL)aE)#TZ5c0?}HNRgKm11p9ay4(H+2IH`sIl&Mt2K;pT)Ws^z$g-nR
zSVK)Gn8)P83C^gptdMVHU``*{9a`yoYE96GTYX?Au`#Q_d}X@uNs~mb6={l=V*MR5
zV`4sLb}%`qv>{=+1!Y}Oxu0pEoha>df<#HE$zx<ZM`1ohXoEGz^*Gf)%vP#ts2Szb
z0^8l&uz!2g?`p0o!7ZLvymcaR(zX3@2PM=ejDvhApS1Uqk@Kj*750*kr*G4Y#fV#I
z8wHP-GEPsTtCUjRl$LCOF`kc9n%BPOOHQCai82Gbm0ZMY0drwZiQ)8ZaUoAX@e5&U
z7`+u&PxYPWQzGky7v)hj2aA-cKF0%T?P#}1k1aw!{(8FJ@SoDM`8A8P(+wY4AlZqs
zaR|3*tAWf#*EGX&!il!{Iyf(Ldp|uRQ4;y~t`ifR%?kXkCliN&F|G1;R9Y>a#<wa^
zpEC)N#Byk`OO~_?rj;>&)txuTP#bE4=p{Z}A`tk{$l}f3UC`>#*sSS5*PkuQ#kXw~
zk$cQS2gy`ztoqa^h`qeHW)b{)E9fA!Ch3e#FdOFy@l!5#Ig(=1E$=sqhmO*fSFPyg
zXen2hT|zE>qdFxuVz9eQb-IL2t}}(S?VCA$L+17&|B&N;!&`w3Y^RRst=n66!^tVG
z5=Vl3TIvia-aownXto3d3f|u{*nwHK_l9%+zzm4GUYmkD`bpFL?m31=`tN`8WI-Q2
zoI^7I=nr)A>-EnJhaMYAaT!7SRow1N0m6^d%rEao@tt?cWLd0>ik|si{Wh7D7PZ%t
zU$frcH&t%BGmYZRaei3VY=?%Y9h?PEI*(bLECn1;=LKJ>Lp5vTM??zl5wXYK;%Uk4
zBSTtWUvoK3?PJikB2s`fh;P7K&V=7LQ7J7p7DsO5DP5r~k0YO>ZTHwZZOPclr=afQ
z7})JAxpd{QsWw-{zifdcdJ>=el&)gEyCc1{i|I}*cF5{?rPPo!&$i`0CW}0xYpD*f
zPEUVVwyI-z`em{^N4u*K)vtT?L}K2Sfm0ix=HDIu!~ncj%>YY!KMZOKFR#DKeAvX(
zs2@!A(oW+D*l*Bj#|qH0QcICrUI@&(mz#^_OXF7uCL}ySp7lK)$g_%Pl(Be&cUxD1
z!U^rxA3BOLn+(poZcq)(z}#}=+OYLmDEBu|sIVJjF!C+=)=S!N!R*et=83k|ch!Ig
z(YvnH%zQZXZG?VlW%Ocb&E#>=?PP#<Fq7?FbDH^_$mShZG95;4?F2ij^TJ2*^F-2g
zd;%a3scH#ln)sqW089Eu`f7bIc~!O9gG@oN?a0(6b_1KZY3cOCE9CXPHD>g-Ms|G5
zFNA<Dz#|$@Td{b8>*#}`8I1;OCR^}p))yds@5uNDFei3yYQQvCH@OdK4Vxb8B;Q+X
z<0I!9U|S2_o-=6M30^ik-{rI{CSQqh!Cp<><*s_I@!mWzqie^z=-M6aEEVW9RrZ86
z(1#JvEOzlBJ4-kEOvm2wy`tb0Kw!8yOC7#Mf!>_&UemAbg2;$%L!9<@;rNBo{99YN
zbW;7nz!{gj$F)NjV?fAb*ccqz64rXjue~>LIeYnjDj0IU)h~<tasLIU^|%0P!~h$w
zBg!tC(hqWYClzl)RCY%82}x;Ut4HSvasLd(_BLos-yJ=Jw7Ea{THprgUKurBCo;e>
zTOLeOTKf+rk+l`XU~nliVEsA9w7l>=&)UWOI)ge~Z_obRPxWz|!Kt_oK<_GTQ?(^?
zlN~{xOFXVZpwY7GR}yNb-Z)@}w0>{j+wEE)S(yyXL%mLeR&Afa<l6NzwSi4(I(P(q
zcn1xhShOq~&7knQ&8eTkMobWsoxXTOy4&Jb!ZYAK#{uhICtlsMe!&}m#5nwQ`H~y^
zA%|=x1S)0+X0@NxjDQ3{6`=|oJzhPG@1%^bNj>8X*WXRR!K_~qo2-QLn`H2$X(IHu
z6jn&%N5!$k6gMFw90y4lNi)qiYXCTk-7?XXLPo}YS>nWJUNGpKc5P?OZj2cy|7C%n
zL;|=UUCbIyzpL2IAQ(bg_W7Q_Sy9{loiGNLWMglZ&QTvx9R^3F_edUdz0DRfqND!B
zH+Jg9l1teU%?h4HEKi8gwC<HZns+i#>zk`)lh_J4gV0d3PYKXq%y@GtG)d+;4)!cv
zxeMG|kd7LQLSs1TTzw}|wNB1!ELPtnVd)G1df2hz*vR5NRzlkRB$;#8ME68*vYyO7
zo`L<SSUqoC67KHxor~GN@OU<%vrhW<ZJN{wnlx9daHXVxQ@Q(VCT_RJr@5z(FvTLu
z695yFFG*v!3n{q4gwZ2MLSxm4wp=a9iIyywXnoKq3#0t<o=y}p^rd394|c#PzsO<V
zYJ4|omkTLLY$TJxB52Nf!IDec|0qzdc*^fy)lYSqiToJGOeHMYHMVIluXGZDC;mA4
z9MuM(W;J36-nLy3_FGqKqOQ6svmmp(sfxo#fSM7Pz(W`JLKHiiTs!LW<($#sTpTJQ
znQ3zND|9se+*>@;!KC?98t5h_WsxVtumFrOJCJ~CWo&?eNc`#j5{1;3sN?$(P2jc{
zQ0sSaC$;D0Vv==*xf)rzvzhvheC5~+`1Ojp5-R6ai?2HqY3OvN-WARtgEfd(mt-$h
z6~+;_YnD^ufDjnj)x=Qr$yov_81OG9l}QfT0$}nzg7ZiZL@iZMy;@viCKAFBDoqT_
zTq%R+Fs=1~v~fLO{#?L8hyc-cp%-&~9#0%UAs}m1DBXX@y;|1Bko4|e6sdYUIbjs(
z=oA?+2KR{)BG?d>NZ7>J@e3wmPTO+*nzAFZp=>LCz#GgWiBb|4@0HMh9UtncOoO8;
znIIt+Dnhq@mo>MmnM#{E(2hev<n;%}&1?AOGIP|Fbo<5>%_Er|$o$}fKG+1)JOwaj
z$Ihwc^?t&oSk&op3Nr0aSm&0R&4WtXIQF-w0U_>4v-O!LOLr1q54m80xv`?Ml>Tc9
zmdSNu<^h)-v7L`76Fhwi5^FGXX2pF!K(XKk(aO>VBDYH<3sJjqudfj9uMskLSjL6(
zsw`}j$Tgz6u5ZLd0S6Px-?eQ@F*Wvp6&X3K@8fkoOVXUiqb}ve(+<jBs^-3@FuImP
zOaRDP%9JqE#$~#{4VeCs_~@G|`Qm!$%do1y59ta2TK+!rb>;)}m(~g_^Xt{8Vh0UT
zkWZ6@si(&Z^J;oV0k+Fg8#Sk?w@-eAnRw%!z4V4&EPsm{(f3+guUw?{=9oG?K1f64
zl_Bi2PLgxK4F+0wmtUzQ_JL$~q8jPD$(GvDUv+^-qjY`7f!#$(^elOH3`9vavetB_
z$8OB0;IfVDmQJpAB*X@}R?$3(U&=4re0i`^#y<|YPwSVjrFM0RZctjM9xd8;6jbD{
zB1@G-To;Z$`%KQ~S8g-#JHZ(z*iL>-k?9zpbrx>jJ{=byfxbPnD+uhd5`JFVCv#g6
z_&;gyvNTCn&rYcnc$}=2rm$WVDsEp;6k5)#pyoIDh>0}u^6peg;SQtfCcEx6Y023)
zx}O)^gT>h`MPYIVJN*cN(0g#;0M|4oVY7^2i4h>N5zwY!ISL>lVHPNUx(Gb48KG*k
zo)pN6&V&ygte};r)fkKBN}|Z>?2)qX#v;83DaM+y6mHef-q--~PT%`a(-k`_Lb>})
zg8CBnM4%TdI&ER%8{J?c3B`k*d2w!q6=Ul`>F%P~o>0ucFqw&05p)I5;=~%N=Ja?A
zQD0EpWN>9?%F>`e;l+KOV^@XRrxl8+<DkXtZX~P`dlJ0(1icK%vVu@Gnv{_5Z|j)z
zZ=0)2Nt`Cd7t+=65Rwoa?k74E!q+GBCCYD^+)Q0;Juy?0J{2}HWH`@9%v5uCgnZd>
zMUT;Sn`fIcc8P*Qx;vLdkOWKkm|V<Mxg&vSL=+eki!mOHlMo1KALc(g2K@1z4IaCA
zfuRGBs%KRvB|Rl)_o#7?I6nG84b;w(L*_hvWYeG$k7VHk(H1;}yGN!?xfx4dG!ONO
zT>v$$U{<toOE1Z_cQM5RNTNF1Kh9PMY1=Z)Ihv{tB|K<L5a{<va&5+lsI$a4{S}0{
zh6kdsM#5_XUEX;nKJ5m=mEn={5H(M)ez6!z3B6FEMDa<CQF#r`M@WcpvFGZE0f~4^
z^{L>IsU;_~c9w^nUik8EhW|C|!iWKeUi>(d78#^!aAq{-#s8pNAG5F3RX6H=CmO;3
z+#^5Lqe|ZN-e8n;_}VTz1KPn<ZftCSyN3OmL(-v!j?yhngRs;Q9<J<G<Vd{haOnw0
zc2)sbUuGI%;#5b^dsmxy*?3if9=<Tyad`BfCoji~aTtx)q4yb(sct$rFU*9ybs`lG
z7dus~B~i2zx2UMx2tV^9xyo3~@=({pK1YeL?7-GIBAfB_2PsUEw8F)q%PkKVq)l&D
zTN2PHJ^QsTdJR*8@8>*w|9c7kp$qpp3HrN4Uok-B-jwFIN<x8SAuD+p8m5|9%<4)N
z<^#)8%TQka3xN4^CwV~s=H4MQUgj^oCc!uPRsgAwIzN}_pN)VnW{8$QuWz%G$8uP_
z##P-E0Z8`)V`?w=$P}jkRgZHAqu)+2bCmoQk3k&E9*{@b$;Cuj{^&ANls>3p`=ERd
zs!=PzE}Y-Np_8{Y#p5V}nm`*zQA~PG`^Z_WZ{d@M|5Vgk7mx>7X#z^J1N^(`sMy;)
z3X6&MxV7BE6>>`CVJaw7y*=1rW+ok#hidt>$&vrs$UJ5C-K;jknD!!b0Y7j4Ou5G5
z37|=Sm&vgG*YExF>Td$2Ovm5)Ei=`Nc@GXxfOLOi8!vx((#IzKWGg5wIjOOHviklW
z&SJ7hVdcE9#rWW9wzX=eLYF$RADoz8rVIqHjCcWdlvn|vTJP73kaY$U(yRmBSsX#f
zCZIl6V3TRo0;=Y>8?o+^zXhAw#%h|{0cZACN#>st9v@(|U4Q<Ue*a*PXiE4pN*H7z
ziJoWNt0~?*#pVYUd?kiHx3__Ur#>K6`}H%By7Le`Q7j_eKcyKtBxcfeG@a`=d7G;R
zoOQ5kd<T>kVPke-SI<*tl>MhLCJ-+4vXPD~srM!d4I0LQnO54aVs#DhpC1onuZedr
z(1&|lgVC}}wfo-VYW9Dd#(oZ{1s2=mqu-2}{S+}l0AMtnyBYVN)L`OX;I8LH#Yy~O
z;vpVKVy0TBANSh44<B$(Ec((Jwg2UI{J(sf@&sJ2;^q9gKL|B<*nz@w`}&XZ#e0XV
zI+KqT{=;qhH&Gb^BSKr#sbos}eVnX){gCZA>ZJL!lC@(vGGIlxjmP>bzp5~qg2^xC
zPyeDL`u)=-V*qG5aoOJcC5rmb&-|Ac89@YUa^jcnGXL#m{LL%+d4)olt4hJwY5OJK
z_P>_NiPM9JW#ONg`b(|GFR%HZzUZrn(Q@9ur2cnV^ULe}k6-M?0nhW+$4}6Ixv>BC
z(+@wjeQmYx{~}oZ2NB|bx)(eWV1dbeUcUJ&62hPR?mu1gKY#K6hb6FdqYgy~K~JSW
zczM_7f4mebSbgOf5*iiw#*72z;Lospv@O2{VBC7IwiVL-Pap5P#0xMsTnGGk_PdoS
zbceTj%-);;8s-1DpE*SiKMe2yTMk~EbvDi!wGY}3e1al<Yasn^TjhUTIsg22O$5|}
zOI;7dzp3)bSy*~g{gEPR$gkSq|6+OUzQDlLQ|;<~|9U|F+phA<WeP;pz^O|~mofXl
z{LlZ(w~5Asms+%JTh;M@v?hJmG19a3XX@=={zo@JoDlO)6Xtu8{x?Bq|9(q8oM1M!
zlfn1)KuZ75PwBt<wD&beul$->U;58vrvJU$uq1{N$ci~-CI79<{x6>TUmL{#{g3~H
zJeH60jUg$sw{i?$7<IUwvTpI;a^cF;$`@`|igx=Fs}{XHsABzxLe`AH_obJYTBxq*
z<u_0<pl>XOyj9xfS<Z|(Hd3oWwqX<}D?%3c_jOgt5GNoOXD%i??_gypC#OiowPFvp
zsKwp8vR0F_S~4u8BJdB-%6FK)_JCmFYm5S)CT+s1Icj?K)O%nBrG3>L%Zge~*rzn7
zRCGjZC~tBdS*|Gjm3{>c6grNW8tWXIS9;A+dIF|Jy%V1CE?{w*{U46l6gkWl0_VVw
zCqTFG?0YiW!4OR4^R%%x!HBd1!^C(w?sdFt2ZnGCLmbkRo~wDKXoybHOtKYh?CJ@$
zvoVAOqttz$I0%A_W-k*^#=9$}{=PaVg2Y*8&8m2wDGFxY=6Lan=GoB|g7l2;8;C8!
zC=8be%pnJ>fxGKEkhi!1Z!9N(AUK0ymPD}OHjRT_h`Prfnc@TJaFI%)*5I!ZNa)Ii
zVMBu>*;Py-1=28<7kx2c@&uqcjPs?#K~-FAJF|PK^IUuAyGII)$z@*t9}0pGMi+mk
z0|C@<99WeuAJW)Tj||i_te&{14)=?uSsiRRAJ?1}Ta=#HJ@*(`D#nCWxYXm`F9WlW
zzVoQ}-mEw)0?v-zx8}hVN|!Z3Vgn=YSO?y{1%ishjzAuR=szD!+&W(qnO(vJGqF`|
zO1OYT4!6j)-Pfo@X1NQ0?SvKn;&C`<m^qJ26RG-PlM+LQc_cQ(-yV*hXCJW6KISx^
z0CG}vBNOXyG2&PRPY4p8?5~cnvuaFxW3sAFfG5!@yJdZ&P--hi`;9T2J9v4Bc|y)W
zNTu+#))wR|?VRtL3)qYPuIT8!j#V$vPSIYVUQBteGu1Hq`i(bG_wLTFfY6C`7z#8F
zWexBI_{`lXj7>ZepzISsk3C)dIB{+jOuBYPK+cbE8u^^=yVjZBt7n+Fg$a<_r|q;Z
zDb}EajdAq)cG6|zJ#ckQsY^1xfwOcaBtK(24Isbk7&7|DkBVSU=Uxg-zT(gotGHfQ
z=jDVDHcLuH#|#L(o**TWNHEOltv;Lyvnz-dvRM@Rf7pBLs3_OAeRxYqDIuk_BHbZf
z1|m`-(hXA5-HafJN{mP<rKC!Cw=k5TbobDm0}R9Oy4}y-+vj=T?_2Np`|n%pU5mAy
zwYE4j_kCa2b)Lt0ggwWP&hX!MUpYgZ`g4r0DDW;k0#+22fR->0JJ06|(wUK(o|Qa_
zq)^JrR2|<rm3Urb>re4hOoG!FtdIVqgY)u_9w&^-Wu9{&7SIXgCq+tv2>}}Z@;K@M
zsJ4(G?P}ly2Jad-76!1Jod9VoFYd{wBVZWHn`$({X|9KbhI63LU%TFe33c<oIA=T>
zvn|*V0eu+trLV!~3Yo3_QN~lw`#>S>3JDYD5#vBRmZA^DR%d!{3M;Gi|9p)FmLDuZ
z4G>2Wys5@!8?B=oB;?4_nFSuR&R|`*eYZe^1%Q?nE#{sOAcN?*lYpGRIizDAp_=Ff
zOwxsj)fuDLA!S{flsq8@_)rRXwqFme1g_jqGMJl;aj$qo8?ju)&F8VTY$;yNP)=ba
z?Co}MZN_{m9sWw3p@%M_;1Y1kZm0n^tK}5ou((`cw6S@m1gjA>!u?D33=~DNX>60=
zQ6AzcY{VjVR&xTNPb?U5VyJk>H&Gy6@I;~p=*uTxkueMA?cC^uR8o^rfD*{w{gO4L
z!qASsyZJRIe|Ih)Y#FR-PbPOq^*VduS~^Z)HGsOt;vPvw(of%0^zylrsMrZDMo|g`
zrlF2Lz+yn=+WDo5u1GM{h11Fg)Y3|N<bp@?s55}Hg~TL*y5g%XctA84JLCVkvajS2
zIH;W?02r-c_kb-Jqab{n>E}*A@;c2Kv2mAY#%_$>($~rQg1hCC!wOSz>Pn;I$2H0I
z1?L60WnJec811xOq1*+?o_rIGIuB&G>w#s$1ejb+0Cw8^urbJxw1mMuZ+z#eT6nif
z0z<t>1Twv|O|E2{SvX_JG%z-cJdXyN>8i-$3~?Mcv{SNREfD7f_)|uiJE3Kt0jY;m
zODM_FxstuxI+qz;9q_cQy-hRK1lhAd9x7v1EKg$oY1_Xuo7@QGbwA7@dqVgn70BpV
z1lMG5E%oBKf7uB}fjf8jVgA|uM!=Y3xdkgP)U|f8tm7|!?I5qmisBNcmuLM|JNzia
zS*PxbB52_}@okenb%*ILI@BEuwR78Q2H#TGeHwIJTqtk&wgGrBR&uNChDicBNf$c%
z!?UmL_wLBEO!n=J@O5-`GAwe?z2_>eE`b%x1A1;~s_h5~yPkAS$WlSv?m5VVQ%nvU
zg{u<ws%hV|K7YkiX1bUZ#$Eit4dj2t7Mq#X_nmp2E>2)}R$6xI&q1zlgdNfJV;%BK
zadXyX(H;f;TJw46zY7TPV&MMsEjB;Cu-+QlGe)oU=~>(3$S3+X)iO^!&+CbsJOJBH
z{~nKp<DWZsLs^^0F{ARBubgyf*&uMa;1#^fY}0h(RKRN#McVcTnr)G@sVe(tl}Otw
z#0*Hi8-~13xg`(HcRs_;F_g1-m#&gZ1NBh-IhXIpmti*$KLKVh5yazN!RV&Is+xNp
zkMOt3I)c2B^HID+$=ni~tPR#JR(a*O&zzN$AgAc|fF^Z;QKIOPe5xxzOu_ucn@evS
zuK<vmDsJ!)V4m415;8+~W-{#gnO<R9gB6I6v8BY-1tJlRmNAbSO4G1y$b5&ZPv3LW
z-G87w(`i(?go<iMZ8&8H5;+zZhk6qtKowDnBuf~qe)k}UjS%yUr$M)noBT{*_zaBm
zr*{wZl`b}br|&Us3Xra)LmY?oB0U_~Y#@~BcV}Dalr3M&W5XajuiMc3&vDkDEm}Kk
z<9TfKxwgVg;PX#L!^@s1y$0sN#MFHJ>B)wImfYmk^eK@?wtp=iZNSz)f1y~9hkkoS
z;M%!@sq{XWq$pkFFdQd-mU%w??fkqF@CW0X-3}DX0*1LSXg%F}#T@y7Ct8=?ZdqYN
zJ~I93kVij9o<<2KBEjfym7yHY2l<a(xUeS9X^`9(E1{?lXxp|Ycp>&A>#4n0g!o6-
zo<W8ZS#d>xkIuTjLm;&TfCP7<M<-Ab@<=L>DM{_{7=QKMCV*=SuF`tt>z*$to?1u|
zk%(~<D+7C>snw76xHR@4aj0XxeB%~Y_EZ0mHLY_bTU<O~25&sZ_<oovNZExmNJ{{k
z#d|-=>^yVa-tt2GFpr2vxJZ}m4-liLp`PV~1C#Ot(~T3<MG^)BGHovJQrG12D^^JP
z-zyLgEZ~p?NR+xLqt|~d0H8TX`rSFvXuBuMr=EE*`4g|#M@g75o^}l&_N#nMDX!S_
zl=L@Foj7OAN(Nz)FcMN%2(-%#fPvK;Hk~g<vQ}W~hp#ZcS$V@w+q)J>NHceX(IMs^
zW*nP?3Et1z>(5t*sV|5<oE!KFv10bxz5GCrIGoJ(H?|3`RM{2lYfaPY3iOnw9w1$8
z4#Dj@@ae7cM<_qBVL749p4T$3W=wH=c<|Q8eW--_c!oc~jH&^nbPa?q=J>~@H6Ngf
zwU+k*+6I-vK`gU<U`)dS!c7Xs3S0yq4j59Ck}_|C)7rv)EE)$E5pxdUw4)%g#o`E^
zk#_4lhMQ*qPCWJC2k2e)Vy00OV;s+@iyx59Dn%5;N{yGP>c1}9{cY*pctm@lUV(&8
ztZ-dY50&b}EQ`)RlJsMqj(WzWI~=l_KRtYXC=ys8Dp_Z&Z=6i3P7fnc8}{QXF#esR
z=Zx>nCW7lz&X;7)_WO6nDVKDEWQR2LTk9|<YhUKCT7ZD8sI5YKhpT(dMN>ru%g)05
zkc*0v76#>Pt9H`hY4Y)JUjVr!;6(V;PH@1YCqR6eSYnyU6L1-8z(Fva3}T%ry+^^K
zp9FQyqgh5NVA?KU0R}Zt)+VA3yy(gO9IT(2PJo$!=*`utk{3e!$+j7Ljz==8_64|D
zZQz=y_pdDxp-bK`#q7xXr7@`ao`6wxvMFwK6}bEb&9Yu^vq_Dmdfrw5^@0IZI)fva
z`p89O)ZOxSX15ykzsm8%Mo_A)Der5m<CknN*%1W_!pK`COER<<5h1!Gu7~i@fgK^7
zd@F83^A*U}`hNITvb1dk2vY%LrnTOZHi5QkQoJR<J35gz^8%(<x%%m5uZ22y2+*86
z@nP(wkrA@mUE;2ki##?x7{!e><_RLYW_R?}PRdSIot~YgyLB$zF2QB?a%iEQB`otz
zGQ8o1wX3#Ppg1wpb%REuH6TgnVL&j+F^6}X>YbFG&aYw5EVA-nrR&jrqi=RjS@e?0
zXc1?eyL_D9?X0bKe7AOIsYQH;b)NW0EDOmc)e?n1gB|wZPSt-xpZCk(b5IYy5ql+8
z37=B2nj=1#mmT@|fv9|Y!Gq&38GVD&Z#wfHkX_-7awAz$+I$_8wtrfI>GfS5wFK49
zl<mf?CEr8wC82_0aodeD%kN0(lA&&3Br9Yg$}Q7NGTThsU}K~(80BJsIR(z0_`q@P
zju(*0)5)BU_MRZ8ezJB1<z;Q1sj7X}*4&x^fu`j6kjCAARXWAj+ehZ26#6LWwCnp*
z9WNu9o2Cg3bQbEGvbCyjBRTT}8Y184|L)z~ZGdHo<X8t0pRw0;01F*5>%OWikg*%!
z^YgUGzHQsS?SYL!fND~>`ttnA)z;G?y2;D5*j|jZ4es+wB-ty@tdYYV`u)2m-GAh8
z4##8<$A9JbQx?gc@aj3!3~4H1z*|B`iKrRt$$Hh>OkF$osnp!@Im@X=lznwO7NS!$
z9=#mayLY$J)uxs>J;~r@2of1B+2bO<SMxQ%E18L2ylw4U6Ww%_0>iavlcBR9;EW%e
zG<7Q#wqc0cdd-0j0n5F?&wG7Z^Vk#Pz7?wA0sJ8F^dGIto<?SEQv;!UQN1l%j@6<Q
zjy(*#A5OPp3;Zl*d?rH-Hmo&K?Nw0Jkk=YGSP3ANK&+_YKc`ioDh}(zSrV4uY}vCF
z>=!*CmHQ^}q=4bGZ)l1KlPQa&bsv^j)a5y*O`O3Wqs&dLv(jJKD^XD*y9E{LxFLf{
zQ$TMbwrpbML~6QsOA}wD23w>t3it%joi97_f)iA|W6Qn3*AP09TOg9olrKUeL{_2$
zij7q!e@NSK{N1q`^nr|ar}OPLAZ8mMB>V<g1Y~Xxx8wjry$j%s#u;Nr`)|6Y5;=Nz
z>0k8vahN7$jBK+Bv#Xi<{2T_-c`vi47F4f(mA-&yPl=T^5HZdGF2qJ4>MPwD{_YtZ
zhGtFzd6Tm-8ma`q?p=d|_-ClsV2;tCMl%gtgQhca&5$M5EEvGygFejJ!f9UsoWV5-
z)D&Fy2iBD!sx-)+nKm}>5cDmTSJS?31@XVHC&pqck)fWc03BB-B-J|f#y|Dn@--F%
zp*Uu;mVkv7Hql9Nn+YN9$=&y{pf8jA5mRms93p0|<3Rzfv%%!5+tzrI?6oqB0`O~<
zJ-EB70Z<Bn6EK$ehRe49vt8(Tb`*r=@Nt6<V&@p@*9G7MiN!gcvPgGu-Q()ornlvp
z^pS0Ory80HU;CDayKh*p`;H&>G?n8Vb=A#_5yBYJGkEi!=9BKj8zJ*-_FdAV(T;xe
zv=N=cp0z2kNcM*iB{JOnQV9)%9Z{l59Bp$t!@=k6N`y48)KOLMWj>(#Y%ym%9oaAP
zDp;lcv7{*bJLG6dccCpGKp=Z9pyj4QlxRLQXzv0$_)(O%t>jBb$@^xxEBPA<lEXUF
z^Nj9EHuLUnPwhKj90x<rgjeorjCO#wPtwBNBVstQ!;O<@yMda`zgAVyH8Odr&x`&A
zM^0|>!<d?;9nb~833C-MaNWB*>uBH|mtB?D%s5R@uU#Hdf%jAL23y7Cf=@#z&@?IJ
z%zZjLLodlPI$oC)vJIC$c8!Jo{?r%Y>ttK%B`3y+5WxiKMF{k4>W{%U)Ul8l#kxd3
zBr>Y$<PdEd+e-o@I;;_4U|cEWz+y>Cq`#q=3uF`_?kT>*P%?3|1R{ik=t)*f%-I_y
z-l~kkSQh`Can2O$>TBfvS>Yp|@Jf_WNJAh$K`(1fShn<r%}<iq5uAXcAw644kqh(%
zqYH+nLx3all)gQ%a_mD1!-E*!KWlhpx(g-MBHMM@_5zui^`(6>^$RR~*S3}P%{tBe
z1I`eY_E8)_dKgpQHD0~{K(Z@UMb+!c&Pb-T<v4xfz2&lT8^47p9;OjhvAWGA&`leK
z!zR#&7HLgz5q%tW5#aLiNWHZ$21BF~$+bN1Wz?5W(D0@p;2;NEzvF{Rt{5<nG`}{J
z8W6&9i4u?iT}-v!oz6k_2}I;fBAjZS`(P=vSI6OwU-UI*`%bb1=?iZ90eU}0){o(?
zyQ8Tb*jjPH1lU9+tcHy*tsazce4?_9$cKO82&BrYyxL-&cn$?iyKrZ}p0@N)7c?|n
z#XK2^>bzH8IpU#Y9Mq$`S@m9+r+;Ec!i7AAI+Wf+$GA=p*nx#n*UsSXNPj+&@j8>7
z4;PZU`fvH=mSh*UT!m5ua&G(2U)v)3U~wF@GtEBiacSBkd|?MJw74lzzh9ZiU^V8I
z>{rG?@)Y0Df=wLZU{hsXi%G4z0fVhQmEb6G?l~Ku-N{gR8mh_b`F@E`vNv<5nkQ!c
zbZGp}(AT-un#H{W8*$eZTcv5<U2Zk&aUOqrt5-6_DWWAebrP^X<4Nu=Z{3OXDCHE?
z_nMAZ6%x@vsto_LZa$J<P*zdP*(1*8LMMa%!k_&6ly1p}Pr9J7c#-8GpT~!({VdT|
zeS1m_20X~rmZqPFcg&i@BTH(E8*wZMfc>%KQ6xS#1Qi&GWzgbc3ymApF3a<AK0@u)
z#=tcRVdWnRb$}Mkr`O4W%gf9!^|;&W<Z2UXpyF!Ac6{HDUj)5wBP@497(3NZ)-?8@
zxjyI=FiN~58BB_U<PG+4$EtH4zftl~e;phvA@<tS)NDI9bQT$4_%ahnYk0>J5P}Jj
zvTX<q+sx-snA_?1C4@Wgt$ADf*a-#7C0HUWb!;+YFYcBB3pk4MJozV)=B8Mkwn&+a
z4{zg7KRyCYY~2zy1C->E%Z4h71?Ep&`6EPlA+jZhq+MaHq{wYDIbZLz@X+TB!)4rd
z<4b{+uf{1cuU^!{kC?kwzCK0?<#dGTzF7-Vn$i^uG{21(rqJi415^iW=hQ}Crf_E(
z%h+b)$m-OZ$Zz4EsHHpo!dn4GM|Ib$laMKiiv5D$>G{Y@p}xX+(g37>6V+>r=@S@l
zr$`#hrU`e6PL*fB#_X&8)`pmPkOA7bv>D@D+|_am6S{d{V}9vAI6ZPKau#uoIqIb|
zV!{v+50Ja5RIu0eO_fWzY<G)$Fq6;}h{mrp=7c2%??|llihHCG&AUG};YO=8Y_WGn
zxVXEN&=JDK7j!ckBAOWmGocu2f4Za&*Oa{^y=$J7*-6l8UM_jpSVGr;`!^)Iia<H3
znK4?A{m7e*M>_3Nf`o)8O5SFcP4-x^wzX{knruPwwt|6$t!(*aEjkk)sdCMs=Z(P~
zi%*(}E#7cuUW~z=y#B-F@l<~_Ll+@zp^4|TN_JeU_r)77A$$F4BOS1_Ei^u@r@;7B
zy8UV+N<PRXH#1Uo6-I{LiMD4RvlX(|wiZ>tf4(T%pxec&pdZZCGA<z7w;5-cf0*FZ
z5siptP~S}3xaXXoyT8;KWjH?NQokBdku1kV7jvZ_jS_z&!(IV#dRjqHkIo8SNc})2
z)}7o1^`S3!QDFy9=j3L4CqYd1N#M3Is+sy^BXGXbV)g=0(YGtuJDG$qREtY`$pg4W
zi0p~|>1K^%sxdcBmN%l<mF8kxtj4g&IY8k&jJ<*lwNPEUN3KIr!qf9faLGV9199j$
zWVLw6;L_IhO<s>r95&izbYECno04%0Qm`8F$Ni&6gGv!Y96O@n%7jQ<hH|8RH{Baz
zWB_x6?tV;XQD~I4#LFZ8vGAzs&a2gteA3_3BZb^+RqMxM6o^x*+8k&KDxz1yzqX`r
zs&#rLyJ<yO96LlXuYK5<KLGkZTN7>arw`L7o^YU%F;<r~Q0CgAi^j(Q)jE9x8bTYB
zdMa_zy=s7b9<bBZooWn59^NP<kJ0P*crW~$)J4t#i#&T<;UWmQeN`X%*7Hrm#6w&P
z0CsxpBfgbM2IFnYN^5bTO`By7T8=@)ycx+?Pj0n8%r&uXX*R!Vm0ou;y&~bRv&e-&
zY!x>%#(8#P{AzBK{3_v6Q=eQXm3@qIeJlZuwHnu1A7yVZQK1WWpOmA^0al{WWexbA
zw!Ovr?;l5&e_nZd8{B`XLg_nImyx2K0l8AiqK-yIfChi>`nB+xmcw%KU42WPBi|$f
zoZRoRdm)+)shIujI_qWZp9&nO=B-~Par9O^s`j;Ygrhj?XQ`35QH3mpH6qb?QbG>8
zUVbqe;(X9e53!I%^Kzk^zN|X}A;Qk*Ie{`vsPTx&aj)Sxx@7X1c+alZxcFK?zQo+|
zhReEO;M<Z3S0S6nSx#No9Cg)WB{VmW;wxS^_43xN_hsTA{N`o-su73WL{j<0FWhtj
ziY&(t8{d8*H3LpRC&#;R6~s2fgG7PH!vdbeJDJ(u1I}5XS<p`0ZEq499#NS;I3MU_
z6--t?sOUITyjBoBlw`g%Dr1Cr$2RSDZjOjxb@)nwap^31$$=Ko_5;(7R{Cds^olbx
z#f}~pH2~{RoLN48|J9?EIxv|j^l*1~9>{<beSZGNu6ASPk0OM*1y2-*dj{8H1H@~%
zRavM>Nr9M4dPPgqo&wW3boQ#H!b7Q}9ynJqEGK|+)<K$2r$Vv6;~3z?=qZ0$jzp(J
zt4=}~7*aIJdDYv;hawlVq{A<ZPW*6eO{c(|I4%F;+H~td=Y`#OC%`+R#6Pncf8?h?
zI=69Afue1sZK}dl<vP+j?1L)R!ZWzyIpjqauh6?yGj)(}bQo}0l1eGLUU>KR>31ze
zCEf9cJpbIyVVvd<w)x_u8k#a7u{|uMff2JEitVQ(9&w9*HINH4>C(m1>Y92w5mX=c
z&wa0ACBZLvHi&{`2<ol#)b-mTpD%UzHjDf~$}`#TP2U?yD%sQ~=tO*B_J7od7T_Sq
z2gL-3##>~#CjftFy{yPQvNySmk2{|;=EA#vC$4nMVCiV;StYxtDJQjVZ;oV95^c&{
zRE68YM!wymZ}HUu?V=_S*zIoS&nv0zeV$jQFNiJ>K}Nl>$``0<TO%3j{CadVc13F0
z6<q?|@)v*Ja;vq2{hFx(y?@rX5;afu@lulbPhbI=cF@HiyJ7v|>OwA3m^nJ`w+<}{
zhxPF+{1Rldhn;*~LSf4bA!&wDe|LA(5#-@;kXl#Te%)5E3hS6~30>Q6AvyZ!gD7^P
z?`6_yB4ea7>71?IOWs<S-eOasz~C2rCGUDDTE4a?TOO=5|5Sf^;N%kpCaL-tX3k$j
zFu;M1<#{~`24bzOiFSR%9$u0`+i&y8CXmm<J}5)CReHny&KfYPo2<J;=Zf@Xn)yJa
zAs+as{ISBd>}JywG9*t#9m7XD4@d<NpgCC1WY0>{ej;WOb))N+x7@MJsydwJ^J%=1
zj}49C1WoV;9I_>#-k;Zb{;s<Gn32V<$M7hTuoIvI!oBk;YMnTD99Yvk2fUHt6<^hi
z845m*Y-|-VUZZXQkDJCW`eVtwHYReugX4|=6r(bnd-vpY`dsG9{b&I{ng{&?k0YTq
zYVl#ybP3v(Q$EDu1-Nc^3s??&CHHy#_@22VQPl%ZWw09;&rN~H?#QI|nu}II&-g{E
zH}Ex=E@+X``>q!`H&tg#N*BMl?9x;2z&ZX?n8$CrZ{I~mSteN~=Qk%B>xU+0l$Zo|
zZv7i+cfT!ATEd~R*NE;I7(ATp>J7f0tJ;mL*CFKgEJtm`1@y^53~mxs>r;{FStydM
zQM#GKQCz_b+_w?4Pn?FzJ?S6C*5P^ACN))vtBW&Ef2#kgOH3BYt-%wZYGeMDBGuXx
z@*iRdH?UWa?g%p2C}OKuwcel#nl$$5j-3|nHeKo$50tBh_{W{~FOOs^&+{DpZB2GR
ze_yz;6RtrewZw~9xL{h87rMkFU2dp3E^?k8{Jm-|K=pQ2d{d<Y2l}!@lc9xuQO8qN
z%_;_`#tZ)Pt{|zGWWagK{CBQs;HW}l-X~}%<9uDB*y-FnT%ASDiq5TD01+Pi^XaA3
zi+yh4G3MQq?tbsHd`FI<Wt5$T44kHuat$B7Wt1(<;HJ{{isXc0%kFX8T1Ne|TFK23
z@0d$%HpuW2&fG6F1(2}Xe?DNUfz^Q%?4rMJ5M({caX1Y-A8YL@)h`XSpLkN9ZX?0}
z<6Y-<aW&-D7vtZ8y&5ZVShc{~(sb`zp)egZ1Cm-k2C)B7?=5~`>+;nIX#mg*>|=>0
z8(M6;5u57_1E3N1dSDf&J$1UkCOaC>8z8KXI-NSI*S<CdCUtstpzAM#U0Q~cu@B~A
znvF?9RH35ZM2p-yeH-tU1W`5nK#gX_-?xX_CHY77gt5~(SWKdqT}O`H)R=xSdM^O0
z+|g;L+Z05XKkFN@wW}p|V58xG-=4vP!TQt5EuSY5k@kDWgK!ld$Hn|p&7ehRTx7)a
zOFP29HoS074TX0?Q=rebZR9)#bG=ckQD_Lr;FSta32n8=e<o!B=PU{SL#y4NY#Pkp
zbp7VYQ?H`*pMA-^c=9rQ4|J<}^*e@??^w-I`&C)y8Fc1g_v;d#Z0WID^-7ky#Y2K#
zoOy5FoxdZIx$`l)*_gd8*1Y?0=LEVm^^TH58y_kS2w~8>?jU3_0{7qRr`&ot3#Xq-
z*$KDWM@}Q->2z`4Zw82jcRi^{x0wL#ONQoeqYRfT3DzMEkV&Hl@*;hecKvQFHunck
zv_MCq2jN)~Pf=&SgCvd?TMBRhik*F;OK6w29V$<+rsN|9K;NzoniG9O*xjAwENI->
zO@js6sz(ph#^d)V@@D?t+Un#GWO_@T1=*$p&ed$B52gXpL6;i{y0G8ev+<W(A+ccg
z1iP7+Wm55WrPyME6cg|$70)j2(LOldHeMQAgvAt!drtO#d8&o@xZ)1#xRUMAKAtye
zI>(1cu0Q3edGbLv(Htmo**9zVdgch!yzXo<Z1E6IK%|5Ixp%U^6IZU?`NLbt|IW!v
zBJM)D4MzVvTasA_P>MgbR!t0_?f>ILWRy#IRZS#czZ;2U>X&)<-GNf6?Zg*Wk<c*|
z)AmA@5GJHF>lhF@EDK=OL;vBZm0Ql@6qxoUsHHH=jMaLhz8Gf}@fgM22W2sPePd?$
ziD(?JfSlqh0o0W^^GdQ-#X|A}YudB?6WVFi{Z;wj$mKuYH{Rny1HZL%PAvoH)=pKr
zpfCWRd;TG($@YfWz*}h=I9Sv@e!mUX7;*{7L7`v;Gz1Q+v*DnAgvtn$s*;B^NVtwb
z02N*{t;cObSj+Nq%CB3*0NtW`fOlz^=akw*_ytaM4$m_|-M2Z?hgawotdqGKQeLVf
zs_U^W<fgMbkxv%W3chh_a8qu9QBCWOFz&u=<$DrnM}U%TvSfqZjU+;Ys;+(gCfMF)
zukY`nGZeVRDxB^2WFOA*ft#kQFy{1Smn)>fj}W$vtSJO8yb5cJMxpAvzVo!xIwrVD
z&II}ure{fOwZ8vGG%X4HgOIw{Z&7G|WRZT$OTif^oCYzb#~r5;D^Zi?FLiG}6MU)m
zMTDLidDYdE-Z*Q8!9zv^v9|EV#(PE-ua^!&zBnTAk|{8jam|b@q77|p7A?$5+(qAB
zP+GO?5{JlYk4*u$TXW?qq%JrNUR`bBb|>6`KvDqQq0RAM-_WZ{UiA=G`6G^?UnIeG
z@f8%K@~<Bzw!qg~DCTsiuELX$cMI3jwTSJs#|MjtOeDvolRh6Fu%AA$>Z&>ZRN!Ol
zr+oyDzIc*n!|9~I>2~_-X2um`07v3AkApchUb7P4w!y!{2hd8g<Ezj`XYuUiU2`du
zS79S^DMc7H&*SI0B8U6=SH+S6dJi<(6T2N<A4)NRG{FbLU@AVNw5gJKN`Kr5IT9FC
zW+2@ZvuJYNbOEol43CEnSfnw#MfMnQ;E)CMRP%*2JW2v-wf?PfiHgyJEElTgEgl~_
z!cO)1405aZjUb>S&1@ZTnGEc}I1)Dq4K}(8cwoTq1MzNREU50F>q=j^J?)fm+K4Cf
zNPIJ!(R)5ZbVz2*?b6l-VS47lh>o%L8nCYZ?%)2d6bf=^X$oX3NnHU*{jt5cN>TTt
z!b_LPWUEuJGb`Cs-DyvH@%*l}mzt6J08|k+n<N<(*vr@i^*yvzQP+TQtOMr<787{%
zZjX>jMcLi~rRfaR3gx#|`$M}aEfjw{Kid^e%qTFA)^l8xAMgRqxIaC!gd*cP3XxNb
zP<0r4y}S37l8?`KwCiH?Bw)DUHCxc~kRaetIBc(REM+riU@J+$_KQF#{H=<HGkZR6
z)vqxJ{MssbMV*sCO3;dsIV=iN^z+m-Ej-cqbE5^whIP(%=$+~cixp>Jcf5)XddyP$
zf0gqcjg@!SK+%pZwt#R`Wl92Or$uPlo!4rYGPZ{sFz$i!4IsT%RTf+OmiKeHLHhBq
ztQ;Ez4-pC9WY$3cKDg`CjtFKgUbNA#E%_9!8Ow(meVNhOj5~}0*1v}YgyV}1uM5Sk
zCVM&9w_b?0!HbzBcoHSJ;uutVTzbD}D87FOPuh7*eT@sn>Bed!3$SU$2kVsg75>>Z
z6R>fX$<31C0SEe1yKzzC^<c_(ntM0<Y{#>1&9JOj6cq4({3_s!MFd~|*8;vsUvu(@
zm;WjShSj20HTnCW<sk5|?!_M3=xeRr&UG=Qz`U)Z2OcugRkU#ylb}@})D1*RZpSyE
zThqf4UA#5i4pyn<wgEl)coPLKdpYzf^C_U~h11*I%V+*~*rujAwLpw6E4<A^bINAE
zeaKbj*x=$FJpO~VCAY)nflOjtsr(woNIw4+B@MnEKM*~*yth{vDxHfOEC?9f^K86;
zw?D~7*Vq1r+|w;`bWt>-tCz0<?Q%o~H%8A<<4J6xfJKx&5o^D0)m<;;gftvTJozmT
z1%l~vjl_Dr+Jr97`2a@JR87IJK?`Z%=<Scs#=XbFhwBf^u&<#(1<k;7$NO|%VXSOd
zYhQdU%~phF=r1pTuN^_ub<6VDE1iQ3<W8WF$u%%my%&j&13}U%HCyXlh+?L$md{s$
z`eZq3pww8Vb+=oUT<9Hc|2M_vnJYk}Jx27x)f4Ipx2m%fDbMy5Q0BfST4QPKp|lm;
zQuUWbMogIM0b8*uJ6OSZ8C+!KbH={5JFf=(ff+;oy1w+m7qti~UVI`2<EQ|t0}Igo
z@{E*!18u)HK_cY7O2W(FI&@~>p1~Jrj*)9zb<gm-_dvQE$$d1T{2_+JyX)E(=yD%%
z14@!~it@{y$YWmx{{z`oXYX2}@UqQ+>oP83?GxytZcl#@xJQl{shOZ&Is(ixBlL<I
z?CYhpm8(;X4np@=C4NE7!f=V6fgOqCXBrIRUW+s}HnkX&JUW(s?;lGjt7t(ZHau}e
zkE>}drZihRZ>t96;Vi*(@!ehDb#oBAw$lL>;Dp%LvV8}zi5m31nJu*juvr`9dMTMm
zD7=YJRra=XUG3f1vU{Kot?K1}BP-M}rVfWQBj;D<+T*_cc0|2RY%JwK9|v=ir@DD0
z)SAszJ$ttbcFOf}ks}pH9}N~f)~X3^*RiPGJ!q0Fs<?ik_8`x3gXr`pJiyHvXpT-d
zM#Pn4u>@kj%iVNmpUfn8{xw~rW?M*z0pqHE!t&NV-<6%WcZpZJB%?jtxyDIrAyIED
znAx7qF*6=N*on!OV#zwc?C5ouPm!G_%_*U+Njx_Pb;vOACquq5{CeXN-K<iqT#6$G
z8u*S=jolxk92)_fmFGu$xlu>(C!;lEO}YHH(`M>=Cngz+X8f9UaMT1JM>zVKGpTw(
z04@;S=O6PdWp9ca6v}8jq@RgK6QLC8EmS180?-dqLnLM*mjE;4`vq9ci_jH9Sa6^_
za^Jgo*0cKIIT)K};u$kN+N<nm&Iyr4HIdvm)&KfK1Rzbp>_?e9PO7l5*90-z(7=*P
ziKlGfIV0L0Y%%^+C+yRlYauB4l`UK2Q$2uVdfvM|2kiAITKu6(9!!dGQ<AifLp?S9
z*o}R3y&A<=19~LpR%nhMV1T7~LZ?fhrzXAHe5dALLk|Y;g(z%p(J||EcJ@X3`&E=(
zOLoHnJ>pCESajMLG^a;nO8zaP+w$UtHtPc4$u+Z7Qe3>YJLz=(k8{)e&+|{rmp-__
zqA(7L0#Y{CLig8yJ)U%};ivowplIxy&VumiTOR$O3qDmlqh;wohWGBALkK0^0vLP5
z?)Z4e$M4Q<lv*~%d)fl7Wu6^gg7!VFOOohagt$F%o7N*$eeeZ+d(+%fHdk*O00{fr
zYvuf}d2=+o2I!1PkHTDZf=q7(hyDDbw&aJbs6FOLv9x%;#n`o6lDO3V-=fW|@0#dQ
zV3r>g3(kK|k6c|7HuGSJp3@8>G8VXxScU*?V}_M*GqO<oQNNm;(iclhI?q07xPzYN
zFa^=s4y5nHRx5CO|CQQZga`pf>J2P6^5&TZV#uq$ukS2ctNgI_851Gpm${Fpvkwr*
zU>9`{TzTfHY$+6FT=BRym5sP>%lpnK{IzUDDg&5!BH9#(N=w#uR{kUb^Sk2zAU=b7
zz(=KN{)!C;E8<kckG3UtzLAG%*UnK&!%BHd6-#AxMl$L1T;|K=GvKi)?8=!dJA1tV
z{wm5-n3>DB^;KpV>BN%fgR@EZQB8x-9BW^hZ90KI*=vbhpyVRt{^J4VYM-zRCTU5E
zt@syTG(@9x6l5pK-=;M>>~hpen|QGu7Vt>ZzcC0Kd0Xd8Go$<SR)YOc%X@ESsIHUp
zx(6=>pjsFYi`_xC#39c+_G~Kv6}=93jlqp7uKdI)`>kt8jjqd|PmO#9wi9MdM;+vK
zv7N}D>qK?4V70v@7`9%7I7#0b5k4P)#gOHk`17BEMsAqRYhCZibvUYCxdL^&ehc(4
zI*1h1H?|?cDW{gp<}h>tuwQ=O%<&V+?%CXaJO##gU)aL{cQq@{Lh}qs6Mc^@$>|Qu
zG?hGO!Jo{UKmXgW{6}V3@f-soW4QlHdK|sE09CQE>-SsogC1d1u&22$0AM(Z5yj8h
zxSCR#7#Ss%JRe@=8o?)hSrakJZ^UM-()mLCeSR^yRf75Y%bkl!ZEN43O7zrkIKKm(
zkatg!kYEw3U_z4f6)J}S_%Q7Jbo!`6uV22}CV3|D<)xUAKqz1)PuUc7?_6Nm@vbuQ
z+I{UFV2L!cC<UhDj9xG4TRZc^q=>{9!M;qMH&Y}wF`m5}MgOocCe(Ng+tUyBb-@86
zMZ|1FvZO7!y7CbI<*_go`)_r)2`9dCSo1VEj;YzK-)H2Vv`8d#6?$huly!v$<tuUY
z<5i!YJfhSZyd_cPMmi}lBMu;kbX*yU*UG%+rj7@L+VJRRgZXRPIADaLko$gy<Xo9m
z$h13ZTA3fV8=V?HjGyWz@IJh_b)Mm<(hP7!;GK~+3D9O`Cb8Wm6ue}*o45o8w907Z
zj(q8K_1OS!D9Y*o!jc;=F+Z2D2N31{#5g^UsLrYR6_Xg+Y@X(m#?Qh(jfzjC+zrR(
zp*hM^<LBsXD(C*nGruJCNl8m2<Jq!;d5!d&uxSJ*|C<lcpz(gO6kvj;JN2jLNd^Gk
zMp&m#{*U{N!7_wryeCronYGi!U|c{g3oh<yaWFq|7>VXt5(xK|;MCcX%P(fi1u(bd
zqU;8qvnmb4UIKz5JIUptvgmNTsP_%V0p}RD(SV!o!@vunC`wnMg|F8O_We}bfNg6<
zs~+Rmk1o;4$iIY^coY9+?J@Mg3LDz@Brc5a^+pv-tQWR$rZ4Ml0`}@syP);nlyC>F
zg#@L=1)|k#qP%fB5}+2B3G`yIGyFXdFF^+j;vv{KP}!bac!KstK4;mc!FQyJizyti
z93<{P>>u^$PId)g7PEBfWzdcHXGpp1txI?)A=YUMw0mQZCBv1i42GX7ukAP`b?!)5
zq`0oH3#JI4A@@dwji$kb#zzp#qe*ThfBmgz%r&>O<jv)%Hqm^kHBctK*(*}%4@uew
zqBVIMcQs2L>Wf5y;7J>c=RXbX%hbPq@9<S~&2RccW!%6}($6Y%-?fi69AT0P7HOZH
z>vtj?FAb@$9T!EK^GBb|j`jw4=m4W3pJDN$SejWS<CDkJW$Lm(agc3-9WUc(X~;h^
zWZ%S%e{mcmI>IDAlHuqD?j^0!-iKkFwT}nK1t-pfn%cNRuNmz!AA`1wHYiV(pgdE*
zkl4A9VA+H!E=WZVucVVY1q^jUW3`RM6UyShew@2ivP}+&culk(2C3BImDH0f-{#j*
zxRua%#|seGr;l>oV_xVIuGeT`OrsBG71ewm9~TGw$?57Q#Zsw1u+dU3;z$UoECi#Z
zU?3vqO-OsfZ+dstV%dAV{IjB@T5zca$mQCRo0u;}l&6U%$mawWc2Ywyg$Y*bF0y{w
zesBE9={v~@Ub$ZRqQim32_q{LVL9%zve0v*vR2Y23icOyUi7QIRMa@*<sx7^WN6AA
z4gm}okpwgSLd_!6sao%*;9yhm{Ugq(POe&J+A~S@ncrD7(SH^U%u+ZK(u~*R-@lbj
z2x(ZBwbCV>0YV8%yIF<8*Plj2OXRN!sGevI*q&&jN1tjc%UT$0&Szfx1ha~xGM|1^
zeZQe2IHM^Br=<&^<Aq>e^O7Y_=Z1fV=hV$}m8st9`VlerR4bPuGX1j2F2_XjlP~+J
z<6^5~XS1BcnP9T@>3lK+_Pv7HQ+EOY2SpA-`B$w*+I%}JjDN$_?wKr-nb6&1Ea@Ee
z2ro)am@?0G1=s~{Nl=VYzO*rz+&5b$+EZ1B&nAXPbrM($E%r#)b7zN4mw?{L=zc8X
z$@aRS2lMwY{a(_UKVgx|RwCiU9hyA?H9Yv2Eu9M(wN|JF6GjD%MDMaXM-EUtHRt{Q
zK#?D}!Gs9cw@L(?qbaxiFknf{R5h+@RAN{wCFD|H`EDLj{%$o#btrJX`RWN5%uLR8
zZI|9@F^=sTjf`ut;!PBH=xhxyzbF256#lPg0_ySQh_U+9w&R&+3IB_6fThp{R$kH}
z=8^IZU##eBnuO!**t-ZE-LF?)Z6}5({Z=XeCKNozR|azd+<0!EIxv6@+886FZZCeN
z=^%*zh?m>J;=)pNDtf`XJtU;<x8LWd#L3terF=s93ugQsdT?WUJ~su7oB_k?@bkCO
z1J`i}?h&d9>`%pJ^jyyobgs%M&;w7^z$y1ZGks7uE%D!~WtZ=pR3p|vyx=nvYN1?@
zVtw49-4Rn|2OoluI43DbwQO|G-^gaw;XY3$G%WxZCu!;)hsR&-aiFe>93iZv+AH=q
zoB^V&2flysoX24#h(5=^vZ8Loeio$j(S4vy@opwhA+5-5y27M?QXKpH?()x9lViY&
z<~NcBY@rrlNkqkSW)J3Tvl-E<gb>3h<q~Ji5YEWsGpGj;TmJZ>8H!|41!;qYqn3$f
z#0n(8{Y$i1k5RUFl>+{jdHoB)GLa!g0bvw5G_#nZsssDBl>rL|w87j`3)cbv^uiD>
zWWW~+$v7OYy?8eRCRbM=0gDXm<hR`Ae+rHt5sew)!?68y#*pFK#aqrrP*xvp;V!&e
z&H-};gi+x&0}?M8lqp}iybH8qBvAFX<amSPPu)H=`cAsZ6w)BByWUy)n}7!T9~tC2
z7O|SmTLDKkpZ7ysp)@Z(hw&Gb4FgasJ_`uOml;ofgf@eSEi!P&mO_MjrG&V_?DLE0
zZ<7h|Fa7oNl!>2X-aYqnal|?l<dR=^I93Qum@V6rgRJk)yi?*OfV`PT0Nb$z0(A-j
zjlII#;J2%2Vh1iPEp0Nx@8{XC`~HR53G~Y$P_I7=1SsC6GodW96ChcNk^vm)3i6GB
zmH@JU`CzZL&Vgz8U||cfWH1m(OLG^u-_VHX5BbZ5<DYBGgyovhm=Q4yV2ItqHX7St
zK;CvDj_=%m3NC~rT#ITL?1L+c&&??RlCUR~K#RsI%;?&GNNp3`u-K13ZG+IOr4Zn6
zd3^%3f`OE12oBc)z8AhQ{5a5!Oe*$9_WcH{{QXf%f}Xdnm84ybCHQLr0QN=^^j_v+
zss#V>UeM+EAN(XJLBAjWP~ZFSkN-dUwZH$)|GxkK@2~%2ShKM@MWF0|?4kdIhv(0y
zAi)-g0jz(^j$J=>R1H|j&DbPTZ&{iM@BM|?{|(*$&t6y(1_J9Brd+UAb>B%EQ(&}o
zOhcSNg=(Q#i1pu|B7Z-Q|Bo*M-`|-LfO{)ff<;?$)3cr(T?eQm#E=1M@lBn;zxSs9
zZ!a2^uNQ*Uj+cNvAg$!G1TJJ;3@7RTXyN_WPnysIJxZZO#=8HPOIaNHeb>&19RHU~
z`G4R4f3N@lK7aqa{(jqk0;}%LfFwrG?#;WP#;c0ic>O)6)7&iVKEc`plk$O$(Rc93
z)-Q<vEO7s0pEsGqe=zGf_vY!zm3`Rx5v+;L{eI>g8Qkk$YCIHg_KE$CSA^nzC<u@B
zKRWoW-N{O91cW%UQ!ahSI`_B?bi?`4ITmaaq`n4jzYUf82J4BRznNh=5x&R^H(mJ%
zS27-+bc=~aNRvkjRT-&ZO&{x4WD7QwurO+Ld{!9=kN2+d>)zPFMTAPPo9@+}Sx0*u
z|G#gR3EWuyS=*t7lF4pd=C^GweLCVt?BzezLXu<(QKxAG#>(Z}{t_KZKB2nd97IPN
zpD=~B{MW^0CobU8kwU=Rr|y|LAa;ETm_0S|H+RdRf)jk3TdGF2+aY|Hpuy=@{Vzt1
z98T8DKRSoL<U9-Q+#uRWiz(K4CXuRy1IaQ*JZXpU>|5p`Fdp(UHiN~;^g;c1oPkS+
z$ZfYnNdlXzqU7^2cu6ZS;52M8{rfK%(M&&vk-8X}vlX>aE1A9b7LF%p##qpWxpO^I
zqrx@Acj73WcXOpv!O+KY*S%g6NL=m3K!R(tBZzq22BQvXsEnV5h9+K6=LGQ?p%S=z
zftz#V@3Qp2eX$n>jc5#kfO+=EU6KfLhlThKf+Vixy8TSP%Rq=^U#oQliGJ~+QKWAL
zK})Iv#|G$mU{F4~?~!Y&f!|K+EU+cr%Jh#n^dJD`vZ=i=Gz0u4%GrW~B?SDz$l?jw
zFqYZ@D&J={QcngR$v0n>dkN$7Y9g9jXt_vh_lTCJ_8jO9LZQHsCA>{WR3o05E;Lw~
zW#RKy6hp>n`26K1_cpZ=FtUfY0WvHKxhx1cQS?lo3{|`YqwrDxs2zW3>{)xC_Bh;1
ziuo4_aW9?m2ER5wHyHY>&lcozOFCenKEVmN=Y~QRJ{ruj2!PbUSI6?bMcZYE+$GlQ
zm(5uw7-+Jk@`1YR`R){8CyjFIwK1@xi{y}CIvIfWo!Uzo<_xb|@May(QDo0{UiFA{
zgP#Trv`QNue^%;MuE#(Kn2BLrs|z4i+|vQB64>@Xq5cbWtmWH)vo1~trY%C^VGnEm
zX}kWL(x%bxCcix)49b7SB6NY6EZv(>XINQ%28bAb=<hgmZyZ_xC4Bn0$1{zZRfjB2
zc1^eUO@q3h7|?p9nY9%1S5HU+|BPk;EtAJCymAEE{>%8ghAaG+X<Nf7wG9NE7G$X8
zRi$>m6#@aUf}@R*&e=I&Y2V-m9Dt}^d1^xn5x@lzuH+C`=y<0{U+(bLbpg<X9WOy9
z7@6@v6)h=+<WP%hHfEv!1z-mH;Xq^~YeUM`c)*De4e*6%hM}NwD&u<x%~NN=Mq>J8
z$jCH=@)zC%&7)*wLFeom(g+RW08E-P7SLTux7-g#eX0OuS+Yq97WUHnz<}T&5rdd+
z8~XeG!ZK*kGs+SIS_)7K(A-IUeel(EC~II`%w19;sJfmJDXgq5)47tc(xATXyKc@%
z3kYddBY2QR|CDOq)Lo<uBip~1v;Vtx^S|B%W$8d04HTEvWtPhGv+45)LN}zZN`QIO
ztEEvIs{o1ed2djoG9+B80+sa~&eK6HDRw@P<2+>w2n}zvOaKrmJIEUClz=mHje@Yq
z?bYtw6Ukf>!Jpe;*pe;C$$e^CW(hK`@pW$Ix;#eZq@~$H3zpv<XM~10hab`whyh^x
zl=bcFSjDFmGjC$GSQ8<XfA>lPPUdID2X~FW4u0AIjdEbfoiDLl&r%1qGq<(BXKt<<
ze9?A4PPG}Xsb>`8X%J$mlc~2dodg#_T;pAy+kaydCC7|)wYbTYPT^nljAsOer&(Km
zdKRzjo|$VeN8bl%&N&a61TzLf6f}?f$yLy<td8p!Mb)&R?vSv+XE0|@q+zA2wOrb6
zbnml*G$$kwdXsY!%7Om$=m_-n;uC=9Itz_YQ9~8=l)cnIOM0H=;X}-p?DlvB%rUX4
z@SquF=e-O<nA6QICly%<;M>u{mB*xY@pIUpdl@qU7=0DIqO#-zDa?hysObQ`w;}J1
zhR9H2z%tf6v9Ctgx%r3&x2cpqLptZw;nZ{LEdDHZ%U6d|-Ku86zN$hRXCgxy!Pxu5
zp(_2%^Lohi`~$Cv`GTG%=Op2Qkl_U%J&!5a^~)Fq@3$Um4NG~P>%!b<jq*(?NA#gD
z-5T9ad<*Si(h+^%_C$j*b9ZTOWn`*LYMue!r11wv64#NRE!T~f!0u_AMuD_hT<+<Y
z%H^$^rnA{D9HZ+(&bsH&z08I)_w2R(G#rJW(rPlSxH@{Z(54kL>c?wU?RtJ*XVXOa
z0`Ob%hc*i=y-$PKHEDX`b@dF4;iaqXV>a}AUIi;92{E>Sf?>KPg-O;kjN$dmVtq}F
z7PslS8^(WXtfZFg*LsObbul_W33+k8?;FIp8@cC87aAMky1(x`k-qvabPFTxkC}3w
z-WnYIF1(q)r9h(<K7#QUM{Q*(^i3^}t5v2r_9Q18Guvn2688|mG&b5g@N6FHLe8ZZ
z!n*9vFkq`q=j)i?-5|R6mq^nD=R)rHK9Skg6AI>_9~*Hvi&%T3aSB^Uv0*1rQE}X-
z(~$j6TYLioR@Noqu|KY19;x<N<VUh<CvR`Nt*btFW`=Dp?n=&FT0Dxvb=bcJ#RebK
zAB0EnEge0RxZWM-r_6S@byDr)1ZFjHsyEcQ;cUaPg%5VzRS3pS^7I+E8LI{3K~>ZH
z@D$=~Ju$192_i}(#Q+rr{?RlHX5wqgiZy+Z@7cQHMb;Ld%!T#`a`5ms-geO_h{Vke
zP6_D(H#Y;aUtFUn)t@y`rvS0KmOA+?C$AYyMOr^M?i9z@am$qK^++ByCq+hUgME6|
zcly)_lJEME^CR6Vo2Nl$JzYLfa5@32E?kO7BWYg9SZ!k4{3^@Rdm)tfQonJ8ewzOC
zvEIOke4(6`d~=T={~u>1uREy8tpMtVnqF|^n9<(ielnH{32D`<<^iueg32{@KZC86
zXW0{IBdoudqbJ*bxk6Chz&`~CknhYjPo0-%GkCIB+0a#O-<pBmd!?d;v8HtC<{}ux
zUE0T;C>oTu=YOG<I}-rLAS9)yP*%2Gmo>Gz!~^;lWMe2kzgePLke%(YPQHx>Qn{<A
z6p$Xe`5{-H@U{U>TjdhO@@oM$@idGGy)9_gIKBZWs%%4HYxy<2QV?4UP&|A-ss*OH
zyx-l1j&B`@0RlR2G}yXny0(BPQRG=PAaRO1utdnuLG%s(vlP@1*m;CT0pQ#g;FNz=
zodjTR`&E)KHYavp(@8;he1p-U6KA-u!Iq%E;3^$*E>a5EqRqy7d&;hmnNo)F|0=1#
zgq6QLQIt<=mOu@^R-1dl>(}7<*`@KmcnPNtmiTWzx9M-x0@$C^q4m2f`v!02Oi&rt
zF`dnWy@ADA1BOdCAW?B36BP8a4bIzIgoKs%BTsdsm)LKF>})V&!}T6amkk={R!UM1
zayxwu+<-8s)KBR6WzmMXdx$4_*k_%QMd^ao)8UH~a~F>*K_np77Mc4Qcu+K<RNkw=
zN&cq}TRXN^W0udA>k4(9>J{g^@eSXKh^INH3c%i9AnG7z3c0_-->kvCyUm|*Fpr#r
zk3zSGXVG2W3!H4dyjC?*X+!{f|MPZ_=q}*auQeWVA}U}Lj+a-S&L-5HAE%zbK1h;^
zX5?;BS7m=)M0ty-<uhAp_NGfc<NRVv{~W@({!A(?2>3W`fRnfAY<6;q=6GSe3q_aH
z2H0imF?Hzz7Y-dK{I>nZ6i!+gF{GzDfN1NFF84ywtuQ|7?K*os0RM{|(nvWo{`RB&
zj?6E3{xo%;Ys5dxVJS{zw#!jE5ON`#{Uy!^&&M3f8t@ul?GYWQb$3ga<QWOXH%Yza
zy1Y~WCk7irrT1Cmx{7<UiJdp3R1&^%-sptwzf`rzE}5IcO#7mR2#E8Xe%!{3h%n_`
z`!@2LX0NFl*U9{%s59Ju!`3)#mmcQ$9@xlruw>s%{&xGU?`l3HpNq=W?bQzkLRis5
zk&*fpY&?wLm%D~<l0Ry&bX#Fs3Yb2fqW1KMq8#tbGBK7UKKE~AYrckTE?=OR<Pi~n
z3{^(v(vIY8Au69KMZ6!6Fnye<P${JpxlK_Zd-4;`Yj-1G&`x8<-2161AuL2?hH9vT
zYYGt457Byf+5EO%Au8-tB86c`c0m=M4Su_jD7CPobq*>nYb}z~n<+?$5})wqlItGI
zOb3jA(_b*B!V7J)xsz@Ng4Q|gT<QvmjZC96Lv)4MJpi2E%sNh(=*4CC+#mt#!R@Jq
zKc^VGr2NS2=HTO^<WC#3v`UHP->xW8WVOo(U(wj+gi5+*icf^(_*7XL3=QK_y`G7O
z2cQ$+Q7%g7qtkf8VW)rE95Z3Azic1xE7+uCO65SLf#S7RB7kuk#vdlR)h$VHy|!Vg
zD_vq;=7B1<_MmCVRl_ugf+#TwexyS1iBGb$KTC{a?_oo8jL=+322oro46UwJH7?ir
zvNXzLE{jJbGqZgn<_2M+3A^34VFH%<ZQ0;rev3CFqAOj&XZ-R%n`5fYbWoY8=7i!Q
z9D1#_<T3{#9+O(ylvxYc#A92oRT#a;G7V2Gd_5jRL8LLm37VrwBMQs2N@yfv_I@VR
zM^(a(e`K%aHOK4yT`bz(UeGNc$-vxY*)=rUS7N_#h*4SI&hh$gYeYW(VjfRsJFuC`
z?^`L9nudK_8t@@P2$IZ`KXH3COo0-cDhoxncmFEX?3fa9UQ9vWi4=Zvabo)h@`ABg
z%iGcqSGpVfvx0RK9zT8>VqDKs)!Bkm)~b=R8*{aR!@sRw(B{ht;P}Xhye;3y+G1Z3
zVi_8dg+Zn1G>Jy#%An1Zvqqy5W0#M^!?7KGIpgCFwE!2>=)fY?gk5X&;%=QlW@nJ&
z@sldLgRi!etd3deG2eLQPX#03+%TUy4tP9NY8#_n#hYn>N*TNJ1+~F^Nn?nYy;rnu
zMP=t~N#E>Eqd<~|n)EGy;S|raL-O{@=?~y|9N*Dzs|fT1QheixgAZHlP!ZlBnngj^
z5V-K-s4*t=v@Jw@8t%Lob%*|FT<vOjx{a;rZQA?OfwEI%V^1|ABR6^kySOzH-Z|6Q
zAMWuUGYL#7%t<1VO4VJ{Jo=>C34;EI`&@FrfY-oL`9`WQ6#E2M!!lq7IW-P(7QUp*
zr(<)wLFLxKG;_spxMP6yW2$n)#l?1w<?-J8qT{QvJbkauR68RYpgy_Cc<f04e{Oho
zwJLhuN;b@eNq^*8f49`2fm^t%N)ulk<F2wk{y0I)lcCqD1h&Wg25$``-jP&X@<^*4
zZ*E|B6nb%UhyL@sUgSgm4O?B-B(C-Mbi{)BU*gVspUP<3q?UTPO;k(U9Dg>;vlR+^
zesU|ltBV3NN7n&N45h|O{JMRxavy)d4}Aoyrocc7i_e9gfDuMq$lpci{4TH~gfpHK
z?VL*Bt}lLtt?l@G27$LEu;@UEKbeI`itg~6ey@`3?~dTh(<Rte!d76#8AoT%lHh1L
zUkJIs+Z5c#AKJ`F{zUmnn%7&^o;|vUIP7w11J*<ElU;FcgULUYdRJARiv$PWitYwV
z#ig{gZ%_;TiR6Zxu+n4o<Ap|S61qo!2-(WwK4fP%{Mc<c7*L6&qC3kMfr5ad7?KlE
z1=#3`E7;&<@>1|VD(3HU4$&nIpXzooeX?-cZ6(uxbTth75rfZIe~OlY+fRA_BUF<d
zE^fe>T*(I!C)r`#F|#qdF*AxyMaR`g+(K)?MapC-&DbxeR%RWi)}>qQqv}(e-JDOK
z+_)zo=q|oWN$6}jT5y#$;{~(S_YW>K{u@=gD`CeroQ2t#biW|15zqR)<a#Id;tZqo
zZ?F5Pk~(cN9bcNUfNGXA9yjcEclG^lwucdD-kN+H_E6Ct$9iTbk+s-5#23*RS6#9M
ziPodE*H}v9yG*;#JX^h}F#c^E8>a3?M1dcY3ok}fy&cIwSgf(q{nY>&bfZL&fto=*
zwNG(u!e&LUz-7RGH~0#*H&U9Zdtk@#)7fkQ(3>BSvZzfJfxf+ne+_O!x^sHbHl_`K
z30S(U)EHv~dMX;|E;fu0+1F=gz-1|#A3fuhtoKyzOPA8Z3{_6&Z-s`=Bg&84cQ&$l
zLa?`Hm5g@ZeWqAlnQg}s?@aWt`)UyBmtk|aC|c?#$=iSFF9V(Np}gNB`={M+F#=i^
zMeaTkJM0^(gI8bI$5LS0I0AQ4O)rewX`);?OUjJOHi#tQ>WDA7<TCjQIfO6{->2q=
zl$gG4vDB{;>XB?mj&0|qM8N7)Z}TG?Z!Nj1*!9IShwdcirgjFMZO&n}8Hkb-13xq#
zEMUc9&7|MZt6*bcR7vRS?PmHIpyDqMnBc~ob`>``J`OW~(D2OI)>fZTrpt_;)(LIo
zycf&Gg&sXLQU9YQgVNL_!X!$bYDGn+sJfu1nKAhh&aK=#50mYUDOw=3roXGsb*^c(
z2joOfp$dB5DD(a3BcLMbKzt@}(H1P^**v+L)g%!0TAp1?w@gKWCh}hu#R-AQtv4bU
zyYxV<phK~cy`kH%b&kHH9;E0VbG%A76m3UTH$wb_H8z>+hjK-;vAE%a^!M|^!&)@=
zLq>M7CNg9YYh>&IyZIA=-6eI@9robni&}eG&EERmOm#h<bY2YJKA|r!vrevP2FjBB
z#%Ef(rg@u9bO-;3w6_k6>V4aO4UkegrKKAYX&AbsyE{Y}5Rfiu=`KM!l<sa2kW?Be
zDd`$YV1RwkH=f_~?)@Ie8~ZO0WHGbW%v$$-#d&?s@TsOrMb=?We%L+d#Ql$Y3U)LZ
z1qa#LVNxLiug3C+>Z_t$kXLMUv$y0abWTZsS1L*SuyFW(iwiM5=4lH}l7P7Z4(<ho
zm`S9Qs;&(_OT5mx-RjT8(9jvpoNpc4I-g6`DrqUMsuWLtZSOdpIwbj3*f(z%4$Rl%
z9OByT*YNLX&NT0b^*B8W--&!KbmtqhhDdQ5Q;U^UcM$D!i(t(O)6x3ysAW9XttZZF
z-q$7Fw)zc*@4sGz2u<<NGmF5AuiM>xAcM#CBQntr9V~EDRRWe1P0Uwc7-yA}hT>wl
zp||amQ)`nI|Hg0G;rGz}bl8{$2^oMg!(q2&IKYt!eny=(O{xB6;bj@gt7lndcyQ=-
zH(Gp7k-9E}w*hr<M{aEQa^Ms;4Vl^>uUMIxnuWE4^6eidFS0uGYx4|u&-2&Kf6t(9
zbfSGWUZZ`7@4td%1!fheXk)D_z9qK{852C3sZ{FDr_{fIpEq<wf!piDucw5%mFTJ)
zU`YIrsZ)SIDPE3F4@f?i#=YL9c4sLQ<3aFdwkg928Q*FszBY#je`8Xh=e_B<GE$ex
zVC3ZX2=T{HIMLC4`ddZCBGwUsY&t3Y$6TEatuh|xD|Nj%S)u#yuG~|p7R@P<X3FSV
z#UL*bX%?icHT``vrJT2BF#XhIr858mfDQemjw;FEadiD@bCTDy%Lp06>?4GYTau|u
zo2<UVoq3`Yg*-vxHYZDL`B6gxlt8mC%GOZNd6K1DdL_F0v)<eHbg!;(fs%(|H~#tC
zwy4e}hQS>|1NUF230i=&keUn#2tKlj(y9k!S`3Y$&lu-j3fX0zjvw_2GsHx&H@`jx
zkd6u>r0FLs%7)G*Z(0Lec@Zh0*KNqBS|%*4qLL$UXCctODEx29rY|UugnlFAYk3su
zpgIHCFd@iWIne7dy5kwo3ZRSj36XzA66dEBKBqs9=flL0foP6|pq1FKE996YWAHEu
zG89+lb$Q<XHg$~@267Ycb+f$s<=bXHu<j9=RJ9#3_6I@a*EIKCXD;%{+~&R4|D3mU
zFLMsTYU>iOL8NjiJ3$}Zkr?vN{nba^82|$eX}@1Tn1dnsg0C62jS~$8a9&YnIimV-
z9QysZc(b!r^27R7)~22&Wr0m9^d@ep_+2dT!spY*$J8Sz3h+z%?PK3I=cCO}5_lWs
z#X^ztUJ12z#}Xp4UV{P7I&U3_6AaGYWWbKBWd;347f+1Zv1SU(p~ka4BdQESrS&qD
z+Kt_m8;??)ZtQT&XVCnT^Jc>6$2N~7p7!}b)Pf-uC`KLpT2m>iHWUp?*awZ5#7Tvw
z7j5vk?PumD_cJ<JwyQ?GRLhsws>zZMOfQND<4+UV4e&7h)+V7bMD`UR_sApTv`o!h
zYX~E4P;$Q#RvHs@ylN}HcsUa2H$UOl-Lqok*~SjBanDRHbosVRKZX+>n$-77$Hg(_
zvhecUfzT9ZOrwb#Ydw#heN2Dwb4G>1KBEUzzheB<rEGp84U5CN0%LT3_vOWn(bV<I
z*rpJ!@jPq*$4{@&h@-8&@K9!uK8n|8Y59lg#n7S8mC!M+@|C?;>%yyw6N5QL`Hm27
zuTQI69#n0<4!)Bt9R1C;iI-D1=iGvK-1`eO6isPTzo=<v1iU5=?7I0=H4`cUj#Q)Y
ztI?^+ydLrk46-AAS1UtW5r+i+4gI!d-i5akm1QddOciJ4aW@EE)a-O<SG^p%6*(H+
zz>?34P!hXtA#4PIX{zObfyW))s*HU$xxo|=xnok40a)2&{@$bcbN19&x>U<>lX<3I
zFGT6HGOWx*-jhO72zL`e8%QS1t`{#B*HBD<YH0z=z$k{5i^Z>Y#ahfD{$ET@CWIJ0
zK<yQ0@CEYHGSmb;w2Rd|h}?BHXDM&7YO1g+kv@<q?eUp4Wu6J%udezg49!dOt4jO7
z3!#tcOwd=m=u3J)4U%_~a@lM?L}WHOyo-J;^lGCJ%1>z*F>j*(?ELcOE4waw$^a%f
zvikR(mQ4sIQ12pAP-N13e+o+b%lAF_pIh-W?t#kn3V<VUEEmIlz!XKVZzt$>VB3us
zjK>%UwLH-ZU~&yYR5pG2=i};$WjbsUYmlFG)k{6Lk}DlAf$EbX%wkRCWxH5!w62(p
zhczouXX<m%t-dj(s+m=?X?dxtE}hnG5iJ6}Wmn=4&x?>xAc5NR`fkkqO0TE&(fkf=
z>Qtr4bb608DU^G%LU#%k^k0esnCQ>WUU(QQR6ZTvS^U1)B<DB8N{KRVHIF27$3M==
zwHiV({LE+%%NVp^sw(WdPiFh;#7wd?XL&Tc2)&VN8i@h>f+NTftcMl@(7~REc^q%F
zssnVzPF|HoE4FK&&W!AsSuS^ITD;7JekZ#=<I%<T1zCUp#ZW(Qq^5#DcdEE8r{#1d
z=w?j?aoukGwE`)rV>#>uppmTVKrc5^V$KE<P1f9!j<a-W6e+hqIRLb1SK^JC;hy2%
z+wt>xV<OG@<`JFo`qXG>4~x-pc(cixOU(f3((QRezxRv6SB5heG*2Q<crFLd_8X90
zOzWn!FsxL$Q{%1p<&G~j&Y7GxXcQZvXq0wdtAF-Af4|5|TN?ital-wqbM^ofrYK_3
z5OQ&>dQrjekhe0%6NEcK2$lHv3VHH<9CIgi=@_RaP;b$%g2)>Wu^OQCy&S+*Fy-_q
zJtQOHcY<EQt$g@q11^YvBPwSb{q;w}RLZ<8$39~lw_SWe?X_&oSf|G$%pR4C7Tt*E
zSm*3lLh7%_=lY<jbssZ7*gMs7e;|63-#SuOzYX+UQ0A$rX1lKjJy6FmSmS11xSB+F
zmg1q*h*NGni>z6RA^S?-xt2Peb+VRUnLn!1Pwe-dC=j{&!MHt!eD1=BOB{8fv-}Hk
z=cy9M!5+dFUqx709ZwVF8U)9qS|fy3hgHOuuUBu2qoG&bx85@XD=U#+r(nK+;KHB3
z0L+2Y8i3wkT-Q_EQk-3eLnd=&%FyIB8cLylFQrAR+hJFHw0ha+gM-91k)Du2)B2kz
zvK=5KM+b5|vC^#J{kedqRF2=pv6isKFGCNjpO4!{KzpXNjmhoARM$-+_PfD^7}@5k
zop#B+1k>eK?!~pqux;-#kD|1q8mZu}`df_p@8yR#N3GN{^<Ur7@N-(pWPTH(KBR!L
zEb+Fd54=xhfDVqY**m|R<EX~BPzOi_jC^ukZs$U8Tg1#mzkpuyn1|5XG?ecdxI1bE
zhx<P|C8Er`<r@iurrAzQ^6i!$!ziGJ{xNb-O=NMBfrZ%NkZ>H9u?;-Q4?h3Ea|r~R
zANHT^GdRyArm*eBtZA&XpOR05b}{3{g=-YQ?Qe&;u|)EH$PL!rty30ECRwDILM61k
zaW$Nn*10zytKA^*<;I{_bn5`2&~*DI5~qDy-+&-oBJ1AVC&6{%0qG&kraF@AZO7{1
zi7vt)AsU$pCFJBqe#tfV%GM$o*NumvkgVO=x+lBw(WizoO3KEwksQdsE9|OV;L$lX
z;>P#dl3~|B>l~8$me2HGFx`BV`=UMEb>yQ=zU=qYzp&i4qr*lB#+!wvgJStC!_v!y
zw1(SMXU#;Ae|NSh#cNZM6`o@_>gxNYhKR;~Cc*_bJ)fgetD70S^sut+DdFYpcu7<+
zWY<Iz!!Lj45ya-*^VVWEAD@z1Q^G6MG5Ut<?sQ5><*%+}KYE>~B;+hb3XELiK!YsG
z!**$t$mfADner!5=nP^DW|D~-zMuTs#?R*PV<YhI$*=Ukd#~M)`tlRuZFBwRhJxuv
z*m=d59z9A{C^lY0nq|qHsvnef;-w<W?du3dl@~A^yaCBw51y@(Gu!UrRF2Z<mo7~^
z1jlptXzk}-8Am$1(X)p6B=A`F)<Ct)DV>L(KR?tNAl{e_Pl0jhKkG50^chxl={hv8
zqp|!%^CWM4>GT$%3SbKzA3K{`L7{LrF?|rwjsI+|Sf7=*Cuh4p2`qOLb5jAj35NtX
zcg^h`Mdk}MCF^!(&-8lxo<3WCgZHMd3I#qZKbaUs)Boj4di}Gu;jU{M<zcj<G(jkt
zlWY~}74$o%F_H8iQ)>0;I_hjF{4q!st=gH^nfd8^z5C2Yzrp8gF<SLbz@A}I`)t0J
z>4Vy(myqo%Uw1xVyM!2nW#5LQ6npA}?PD5qS3KFPvC@8nOIbWR3_qqP|HygQX;my&
z9CZE+^i}05QR3@zn9H~p&biDO=6Pp`JTfQk3)hU^JuTX1c`<4V@2;trE$tL9XEO-*
ztN&B!0M{bWO0kW65@#O!=*mGSq7U}M$uIqFj7Lq_>1<J9jIQhHHv|=wsPG8?E)}m+
zo<A81);y7zy@xCdF42y0OlrNi%#Kj7;vVxlIbK!qbUk%5<lN}*#=rL{APqm6Hu4i0
zqu1bAqK27>-LZb+qPjBin&!6vA$7xD%@kNCtb?fxztDhWLO1zUTCR4+{Bz_A9f$c-
z>iN@P(;sz8d}Hh&ewz(QXnexAVT}$AW_!#`%);p#djHVZ@TMM_^9RIPRfn)aD)p-%
zXp5Tx^UWA%BcTp3ED~LtxkN`PG~^cSma+KgPy122TeMr^u6HQ0;3PjxR+uL^zmbJ-
z&7rjYWQ42WS2uo;`KNwQcV>rTjF?^MZtuH}%|aw?l?%no7wwdVlVpC*Y&Yfx&_1OQ
zOW}jazKbXruVtwYhl$)=XM82pRLdJH<PrC8MIEltoqep9sn(Sz4j)b;2D;=@NP?LR
zmQ<6+aG)v8QOKU_RBz3-$=lPIdQQojls*lqGR%^6uB{16_4SUvNJQz_CZ&=wy-ivk
zz<xUbW3>Ir*#2Za@I2xoXi<)D8Nhv>y=o@_L6@bs%%65#-?=m=BRgkdwqdUcqts`w
z=XN_C*6s80?yQg)lUXbyq|9MGtpYX{oBS4?F6FP%ri15BuR!hjwmWjCSE%#-p;^-2
z4nWI<w=~B>U_;mD5y)4$+drdSzi2n@@Q{2Q_2&ETxQyi|GE-$sI!#ADoN(D<k3RPm
zG=FAx=YOtd&St`VOd6zJPjYVa%WapLHyf<fu3jdnRVXyIU)*fA^zddnye6HgTVAGT
zzx?Ef^nLvks*FUaXwK-5d#gqERO6B>G5$;s@!h<8w}`o3bB<Vb2)!la+19Qr;J8o4
zZBo4h8{#;`h^Z{bzqLEl-7|Uq#g3(r`EG0PyQ6)s`wWpyC?%)3hrgH#UcYF#67j<`
z)9A8<{!8p980r^)7sqcUvcU}F1GYyO^)NqLf%e_U@DijcMf}%=PEH~+UPh0;Yi;*j
zKUA<&Xvkx;x-XW*Nv}v62)WX-+g9D|Q~fifSs|o|S}4o#P{{3ZH8y;b4aw1O!<=-5
zHBf(*yuK_ybZi~lcj$rCqe;6J?qcKhmBbkSaTQGGh2_(vZ!`hSNUNy4BjGOipPx~g
z2<p{H69osU)3Gw_8xnl{=v@(VcGJJHIhIC{G~fuT9*jUwVAQeTW|3~Jql}<SdQQ~=
zT5S6ws`V}zIyAEAvn>KBS-`9Rg@QpHjV09fdcPu~W~3p6)8M^j$qWh-)P~IamCn7s
ze5<70E8oPPzgCjlZ2%`sib*WMzg2rG@VBTHP$0-a7b{I+uV{^BJg7A-*gUdkhknMU
zXylob#?xjXm3ma&-A##)zTJ_wk?w!Z&;I>TBJDvF4D5|MsYwOxCQS}ne7UfD9(CmC
zR|h9+U|v0%%avY2RY#V5EIs(L+f{;5W%>n%<MzJ0QdT6gYW{s>+M3L%&HmCjA}gGh
zB2H~OnKYzi+K=!!tQsyO-Gj(b)$*r+&0JvQ!d?>p;9nL18y4(`MaGO4`&1!EMxIh^
zO!;U(!LZx&dZv!hNr%3)1?$lwA$mGWD=5n}8lg_|E=0eb989aIsuFIf_$nvWkRzcJ
zN+wOndRr~NXkW85C$-UwjZ$^{vdjocB2svc)xRp$u5I=0?H1o~4^*AO-AZlA88#^R
zX8Kl`r7Svf-taOaIBUbn`Uifv$UulnWq#&{Yp+;(LoU;o?C&215}_U)s(R9MA6t_*
zocVWMQykq61iHMUBj59co~K_-6XBlR<|4f&%uzG=iWmpk6^4qO)bh3Wv|_+>;0Kl;
zoO~vlUyNRh%}$aMI&pDi3jCh+@QVXEb%v!Lfp3ECDUpd)@|ow7X}_REn*urdm-V_g
zzF8M2gcFO9UA2a9yZP%qO))3Fe`*Y44CgD$kAXoo=2`e9qz3kri1x7kVs7`;zWKb&
z$=jB`%J{{nv*HTueUu0iK)&ACx=5MGUQQ4*QDTArs2uN{R-{sueD_qRE9l%O=>+YT
zewx_59dNwM{1Vr7Y+ep~6b;o=Qv7Dy6{dHx!QLu;=?Ewt4*CL<o<%)9@}vUd^_s?I
zkj1}M)x<HP>cJ3knZ{soHC@q9eZy|)hXTv{%%)9HKRn=Wb7pg;a|k%3KlQZqm4b(U
zWTs3=^|wWx@m*n@66ov;DENO}>$=n6cZkqERwrb~X4sHh4iw|FR3r2(+<q#v60->k
zF3Kp~hO=g-@wG3gEMdsOdG+N{M`-s+y157q9<y-lECU@1(XxNFD2^E4VHEV@2VU}2
zL0W6l8Ytg2|M_Vo<1S^S(rMvB<<|(B_4MaO(4pwY*pR73MZ<L15hsqwbCQ?kw$Dma
z<%1!tG1;;DFKt#l#C20{Pc;CB39;|<c-$a*yhP?r@8?d8Ma&A2gHM>P=X3i#=X#52
zmw&^0CHaRzDFt?N+X7v5y-z64*O#*4aDT;V|5>`uW=oV4^q+?{J%()AmtOSq-Gt=|
z)+2scTs5ej$xHl0M&d_{%sUEw@Hj%HI0K4IJik)x2)V6Nh<-d@?^og}WMuyATm5ZB
za#zSVk~a^D9Uoc>T%qgGKK*Ui(#fa9#PK!{GO$jp1p;>k8+T6eAK7m%kk)DP@9(O)
zM2iS(c^?5e5ic66oFL&5@g{Z67yZV%1YN5BX8lhR^_=#gWlfP^@{P}dnc&uGGEATy
zT7KOa(5%mpN7K3%0}*Kl1J>%BH(I}YLiiKkf9CCZ?EcuxH^XZ{;g55u)L66QGhYLq
zkffGVe!onbh<zP};NWH;KV)mu^jLEB=4Lwd&<=(qmKvs2;6@s*Uz_SzO7A-M{QS_o
z8ERMrS5fF9Mqm3XU_D>h=lRX8fi-JLx1ohmYNH%)fYY5w6<<QEJWK*DbkiO{6GRSm
zU=lchuX|`HWAsqD0e7}V)p7j%=BNrH?J>?bk^THvk9Z(+N{vEEw}aD&+rg_1BE}*$
z8I_|~9m4hdR<(Tk1okPeiAPRTCG0Ng?-b-kWHIbBF4iE*zweX{Wp{f%^RUXGtNdA;
zABj%Z9KD{jEHJs42f_H7{);dWiWh4Gbdl5Pb(`-NpDt3N=E@l4E!j4=i8NPRa%RiR
zNR0`;Ha}2F_=;)lS>|n2e<V{hhC;s~N04c$_5q6;k3)gm%iRXYvnVJ=Y0Bqf_o7lZ
zFc{*y##gNt4X@$)T5;{{yAC$?VDu<A{<Z^-i#TZAYwXIOT*Gl9x5Tn5lqb`v_c@9Z
zq9NoI?q-C|VQRgV?k<0$hJQhj;DW<A!4R#Zf|(t@ACIU}Ei?t>O@tBz+SgFqbi~5c
z)SJylZgP$!yu!sjuY@+eH&vfXbRp9PpSDAm&r<V*8J^^i+-)`11i!yiigDSd%c}K1
zl^ftiBt%r=kzfSFXKP*jvWcg|Cc<=tQa(a~JC-8xj!!M~@h_%}KrE3)uGeD~7`7@(
zFk9bJZvbU?o-E<^E;ah0eFfT}xVv~6Rcbtg9(p%h(+^bL5eK11B<)W^StS#b7(Alg
z9${GIhn{vN?;c=B{$N;L%;C^{=^x-XxB;{>Y(U%bz^~SA?_-FOD2dXnfHk0WMSED5
z5W1DpTB*fcki8;u8Px^u3?A6*`YJS$6`uX5DNbhNpQG#a8_a<YY$)3HbY}PJ*76SO
zS7Z5fo5Rl`EQLD{sejOFXF{Val3@gduhGr>;-vKEyMqyBCa))Er&|V}JNmnz@6h=U
znPX*A+u^6DmAXhxakI`|FY_(?cEdJrX6x+*G3Ytn$7aJL1;l-R$2b$pKhHsqX|B>u
zxuB5|Y4*Kd*~A#tQ2}t=wWsEjUv|{7jc2^wKembdu}4OohJU17toRqJ$BF6hh<|Hz
z6hJg7{gp4w_w`gaUSCpeht72&hsYk5hrkK~l2Ydw<rRp`oxcV$SMsMwJ`(kZtSrX}
zk;HT8_PV#Bs1AQ_h3<iQHRI!ra8-SiTt2kW>h9(IO4?;c;H*ar(P0+<-Qym9%V4Qs
z{1lSkH2S;?^3O`JoBvL8$$(<q!fUYHC71cuvHY^}!EQ<(*`V$Ki(2U8u;%yG2bFGM
z0k}6fR%GfTT-33*=GB*Fhw}lBQ<6EUIzHIgHhL*Upp_iLPxu;^nfY#9uBY{@7j&nW
zwrvyiwsdC0)CeuPhE=5pzNc=EY{#akO7}c&z@}VgA^fn~06dy&>|;vsUlO=HAbEsN
z$+*Ys!njg-V>g!r*VA$;8eZW~Lida8*d_D9{0aZ}X^#?+f%rpJ7jq0SU*NASqCY2N
zVsQQN#mVsJkBqZ73(fl;9}~UZJ^MT$@VWB-a;VPC9}fz!FuCuN*{t7)_i1hX5BsQr
z(%(<|kj?kQZ&U~gFL#g~U1)a(%gFQN^^PH#MAX-`U<(w09>m;yPqRB))|8=lA7}Gn
zdcZ1O$%R8LlrZ<$3di1LlrqALdr-lM>$yxSD&8N|)z_sr=7Ep6FVeh^uYV-mNHBJq
z1YN&|;!$`out%p-^anitISIF-Qg^j0=saPx==AqT#S2j1Pwm`rNTgZLk5Qt=vsaMJ
z$r$LZ!$4Dc%LnlKYvGTP)E;yHBO7VP`z075Mb+K|7Y4O2j#>6`zhwYodk9s3%7Xst
z-2v~sUCk_J!7@|}1vVo6OM9tssV7^pC~wG>x7MxJ;T=T<@cx)h#;w1LQ$h-CLTE*5
zziZKmI$7z&nqwby+<X}K)SW%_$hj#}WT<%j8Fft8$%5+_sC7nlzk^p=4ruTNebnjb
zS$%qg!OHh+PYNeO^RhH@T>w2w6Z3w@W{UNen@Y0pNDQh$XjAza^OJPVqSLHjg8}}a
zsQ_s+^oUjw@!^nd*bMc}+zd1~Y>Jn;;_e+8R0vv{bQnyq1|EPXjlY6!=`z^4cJ=LE
zWK72a(HpnDTaUahD_H`2_t6E?kj3Q7)nAKjEo7k8%zxmp+?*=n_A{pEC+X#8h5afk
z(1hC8Q&IW$J*iY$ey%lb7IrUMhfJ3Ikio|470q~9JW0S^Z7FJmfP`N<EZaUYxYs6i
zEn|X6dM;TGvdW-zJKMa@H;<cf$au(TBWc2?xjg(=5mY)S!W-5m3>2(uw7DImoMQaw
zrqr$2uxWmzG4BcD-5kdLn%m6WEg9WB#!yQ9Rkf70`K&zq>r@!1g-77wM~c}b;Ftg<
z!jdYna*LRk!o1ohXlS5v$cUg?FE~IB8J9J!{RzTgvp;V-gFK&W9?GP6AF*Oah%JqN
zUs9i`;#{AB^A)q*D8Ihq_{YU|3JIh;Jx-pV%*m9~H4BbpAX#33*(|nz87tS?45|0c
z9~<p1Tfnq6jJ)f2QFt@XL0geWb|VXmY!z`~Of*^aOEk9|Ik61xkTc!hE>iybfb=T6
zGRMgD?R!TY`S+R;S2w!Rr*r*f+Wdoo-0wam(nnU?<}FOsDJ24`NW(=zI%5U(nkYX@
zUGF!-EaHyOUJVnw1AOjd>8-DzClVc$y7cHqI%}uRayRQ(QWEq{fZz8V@^bLGs|DPc
zf#y|~gE|BKw(w!@5eg+?VO5d?o{UFS4NZS#Wt`#(8@W~8cvk;Y5#OQej0_hqgM!ar
zKI09n#gW!Jw)Q9FAEBfO@D6j?j&hYnbbNEcH!}pa=ov%-bQ6>h#zR)&<fZ8tTjSBc
zOSM5=ag&?r^5``3uZTOE|7Rup@pa8`-cjab=h`nX#8w>QuU<}QRum6b0UHt1cl#IG
z1rjrbS%^W5AE-%!J4>dNt`=WA1X<6Yoann$EX=_XN<S17h3xFf(CVj8FNKDBu~~4_
z>`p2xM>oS=6}tbpK)CoA9Z)*o*DzNoglHJ{pT65sNnCrl>i8!PO-x`xeM9A!jzydl
zdi@}x4chAzRTM!01(neV^e5|B!SY)75JdYQSiWhoNq;L41|;qk{3$-~?HH=H?ZHP<
z>FgON6&L>5Loo7@5?bolaB0}E5wrPTi5E)ccsx>6BL`<h=h79`U)?6Ddp6Ny=e#$b
ziH8-g*F)_2^Vk`{DyHb8TuHv->%boG>AZ{@C-3mcy!|R80*jN6(5k!N(71x`C)CX>
zM@X6-+wol-<LmiHHYDr^FaDG|q3#HG^h6`vYo^xPY(X=egHuq<=Ep?f_$1>d9?I=-
zCCx^FlWLO09Cn1Q(A|KH2$$Xq>BeD)oeTcW47{Z8Lf-Ifl~jQ%;Z~yC$2W?P7EXnm
z`dTD9xb`;-$2Mgpo-Q?RGX97&qW4n+pYIt)bi3!vwlR-#Shiod_=|1P6%;y-9U-50
z)?t-pt>IZ+tlK>?cgKQ0Svb8(H`zyN6UR%`qBIoaV=Z!I-d;^1zX<3UltJO|k5zmT
zc%WY`_WgWz_DISGrj*_zPD8J(4M3O!L1DoN<bh9k+6L_XHt6DCF1e!x%OwVou<&cs
zldCj>WoP-8%w?Y8cdCL_sYk4CxRlBu65&4c^b6vI=xU1>2iNyg$Swg8vEydat3AtR
z(VF$6FVSVY`}NLk;m!>)FC}MZ4%NO5nX5MtZ(L=eoErZ#>Dx!e2_d+xoDRiBzTxhB
zsCZk9kJ3M2wM?VKZ0gNZZf+2ktA^Kah6{_#c|Tdz`9b7{Q3jH9tDw8$pr%u}%}}QE
zK|BX-BMw2st%4{*K8Y~(p&Y4iG%Fl!GGdK$FQWSy?n)BOAa?!SFf)9N(a+1Q3ofCh
zfm{wS-?dOjqP`bSf_+j0qp}$CeT+8<z@}q4W)qS=PF}u%{7OZb8M}>)>&3^JRx1l0
zkcTRcdmRi)BdmG{gp!3T86!N^j5Ku5H#_2(o~e&+6AvXkf;}Is8sYW@J`()9=Z*5J
z>Z%NVszPaO;OC)t8f{Nqv%YQ&Fr_l|57M1sA?p&BI@JD<uG+U)3fg78D*vDE@a+9$
zP@R-NH*<4f;8gDB<94yOQTetbD2!wtjvp$o&0MZyyQ9igg;?ojJi!lSs;LW~=iUfZ
zEbp_nOH!_sYt|Y|uhy*$%Wnri>v4`;nM^+nOzU`3FV8tEX*UbD)4(>?u1CUMPj~)2
zc1(Db;qo+4S1^3@c8tC_;N|LVjax0W$85F@!_};WF+ef?2`A^*_a+I+UuH}Q?*+KL
zzAjIDW?1R3xm>pZ8OVB)={daH9g3ClmYy4pCyd;#l4zLKuM*3#GFsv7IZrGW{oZE`
z`I*0G0QGP=F{D<2Bdlv>*zF*LVj5$rXe-RfGw!famk7P^559swLu!B|r_@K2@P#h=
z74?dw%~}#GL&nrxMNX**6Lh7jk@XB)9vjN6wPv&f=AJDl$A$9@K1EfIRL!3zfhRd~
zoTlcLUAC2)!xA^3-RtsP=x2w=$!{r&Qx3Z+H%|Z8oOCHUV{T1ylzl-IUG-Qy7=4LE
z54I^c>{OmAh`mc}m^H4R8mGB&DbuF7?=sH%q%W)fr8=h+MTe$GzCs5;%U3oQ04p^d
z_QH2@MmvHGkM&D&0Z^9mt4I{uNg!JE8s*J%_3vmrUD#RA>B>#ylODPRT}<Fxn=en5
z3iMwW?h))F9nLW(^N)W|`SRR1H8|WPL9`2BCy?pqS+HnFrhxREM$^KE>0s%xm!;Ni
z0NQkbqEH_Mj}>J;{tXq{(Iz?Xu0xyGhEv5=GVI248q}`5=x5ND?bHdM&<kN%t*}Q6
z%eE+5^$OTdA9}jD6}P9Y)5)~$s5OxIrE%$d`+b)1_EPx5aul8R%%bwEUKQJ$?#0BF
z0L8ao>q8tWnK1ess1J)31vGr!oV~A|C!vG$^qLQsjq%Vgs$YHTJvOx2VVtNxL^0hQ
zkonjLjG)JFz?49PgPmQt)A76=%bJ(i3tyg%&ZWaI-jEOZP!S>10fN~3kxt&%+m!!8
z>^u38wqE>2P66l@yHh@=`+cpa5hY|kmVLK{pNWR$)v}1Lm?Di8<QNfU>#Y`Eh+k^1
zcw*6ngTghc;)KZ{RTnGPqduNOR_@d1`UWJ!;_tGvnt(m-IlMb=6Ph#goqIM~Se4KO
z-LH#~=qxh}RQ_M<_)XA1Tuc2KNr*I>7AkgN>}t5984aTnj+~0C1I<}?Fj=RqWCr)y
z$_ag>uk%9fU=54XWm2_^bqJ*<h`gWBw{2&(PDvw}fAl4#cC72p0jIURbC+ufs+FTa
z^p|7X7I7qf71GTd>2GZz!HXLdYY%?6x+}-pQX7A9YV9u!B8)}W;<qp#h;#P~-T3Ny
z6K?J_&qk#}8QR3-Viv#TY*8I4sIbCOPtpU^Y47I^;8)gHAQJ&o42cdGH@^i<!?(H!
zZO(m8u7skFA9|2h<MuU0rib#kYuBm|FFvbq2D!)l=v!5u4|Ww%>EQyV7!*B9g8TT(
z8-+Gvqunj0Ej*{f-HnNd4A3mWC7C74d0lE>tQ*g%s*&rilX_0x=-{}9>2J89J`K*v
zHk!U4jU=XDUD?ol!`OdZ&A!B2lm9eUA6E9VVB;+q2GkyOYp%`X-<@i7nb%d67ZDO(
zYo)y1YwS0G&Ndzwjs38;G77(a^GSk7=~La;+J{E}$x+D9-ekYM_3BE*3S{G>nDaZ{
zbUwWnW+i8ceDCA^PbTnprQ8b5f3ka-^;>-PUu2t_kMH$YImPx%M~k@ri-8iG(h09U
zvc{AA@iCdL@B^Wb%m&h(`pT^EdnjzDbUokEtZF8ac;ZzslVBfIR?#D8+Ke2_<V$F)
znN;bXH*fYNuHzh5@{5IWjk&+l%#p%QbuvHr@y7bajTH)Q`gpij-L*aZLWU?p1Vo7z
zv-$57)(~z*#k76ogq4q@kdF$>qh2<EjML_$^yP^ovdavi)s!bMUEgb;@G=H<w_Y5X
z-V57QDVYmp@~rTS1$?trL4o0T-s6t}OR^Y#lluA^>v9x)@hGMJ$BO*G<Ll8u3P<f1
zyoA%tmS9MIA9=rNLFA`b<NH}dE1kI0mE2MzuzjVDs%K$S?_3E<wKhKs?y*ZZrLe&_
zO6PUN%t;S{QNvCzUvi;UL~uuj<<ELn^}G6eRMTEp1bgD$l%s@LeGX|%*Ix~dXQT2*
zU2J5Wr%N1n>o;xHU@O}sn>R<+{ldTB^z=nRe9B9LJIZ}Z$d*6K?ae62w8;;>6n3y(
z0SrG#l<{VQ#Jv*C|3-vDVD;Yvu(zBTGbtD=at?WqQ3-Z2>u>{-e<rE@tI)DFR{R*@
zX>{#N4&^ia2bYp|Ns#h6B_UPSZq`Ry%spvmnS375WYOGlV@*Cr5*&nP(+CcDQRS~j
zXJ=Fmbb%lkqh?7nyH=wBF7_0mx;FzP!(GKzuOLQP6#>)y;as%9LwVJ8G$}Fqt&u?|
z$D=Nk>%2A^eS@FuqK<;<<0`*J;fxP)Q3g5k^Y*N9b>gYlQLL!jIBmxs+N_IsRDFtj
zljJ_2o#L64CnU=dPEg4+q~UXU1Pf}|_r)Q&9klE^;VUolV|2*o3#wN2Y}>gcb6JIa
zA-;obb78!GK4KNeVn6nHd2%K6Dyo`nFvikSx#nEEQjuO>-AdEz*9{dbR!^y&H1bbw
zIN2uS8`WgS;pAn97*aZoC64;jji-JGa9$Tu%=P}#dze8|i-ilQfn~faYOWN?l=G8l
z-5(P#G7e42)yL5=+V5K;+Z?QX*7n9)hX3Z(_GMu#LX<G=qZAg!?lU*MuuXrCWu56^
zr7@t?Q{*k%p?eWwIqLPwv+;o4dXZ;0?JEN5JBNwOa7fO&qUXxL6Eq_e&6wG`Yk0WR
zhO*t}v;hCz&a(106L?qa?=|n1`z#ziRIY)~yDj#}ui2jX9+B<Snx-RH_*V+UB{2E(
z&s#p#D5ztClrwsqgBoI$R7n)H&!Yp}t*dTKcbAz_BG~sCO@DvZ4w@*iSC$tU7nnQ0
zc`v3t1B3-nJ0Msn__v=^GPL>2^Ar?~8wQnA+6h)%Zl<yO@=tALBkmMs+H&DSx2GSv
zm%(V*fxnmL#z<ZH&295r7oRIMKoYnb5@q@iKgE9(37Z~)3`b_EZ2%B1_t8TEf5S~w
zg1>oFq6`ld?j3#V8-Eb6f}jCk?3aHe+rm$yApDnblX{G_#XC#ruB!UP^5uMaW#xXg
zX9uu$|2>E%h<|v5=<E^#ZKIJ8hyEKNpG`-fk7hrYTRM(*lSD3Nw#-3*QXOHR<RtNZ
zi5;4AQRQFuQ+U~FP)VerN>zP%%SgZ}7Dhlm-(AP?gs~oB+W8mHax3)PZgn6vx>voT
zrK$ctVSRx$2Noy5)sg2pML>!8vT4^%bKs&+5p3V?deQxY`^cTZ61Q9jcvysw@XVK1
zLOa3%f|`p<#&MWD!6l(HKFkE)Ulv0PuT6+)0zkzWn7^}ux%fEXHL$e(EvDb#j9JFx
zyPMSmSA<i7OUmg<y8&>}4u+}{@3OkYLsnT>+*IwiTfSZgI6d-tH;-O%5$#A$S5^J&
zsc7?bm`Ed=-=Bv21z6*e3Lmb*#5o7G6IuLPH(Og69rgj<AgOAWCo34bZ#a8C^5cr!
z?$@@$7lwKt0}YcZ=vS7WEzckIgi*^4U>2|8!WVX{LYP<t6(e4g*$(Hbf_ix^i{KM9
zg}`vnt`T@d)A>2+4atgSM!GD|oiE2Xl8#TMR;xkdntnr=!GTfa=kz-AG-;c{ivLUI
z`8T!hKO&4YN<n4F@^RZT<HYgpk)e5x-eCMOE7@ha5g2xCvv$hNUXaqwZM2=wm$nU@
zlpF67L|7&hAwa3e#oBE<cQtMlzmoB#mN}3OMdbo=Wu?C(%gM)5$%%gJl8nu$c`u2U
zmHj>*i2w)f0Y<n*Qe{3JgP4gJ-G>(0*vH||+Nn;aD!g}FoqvxGqrKQ%ay}^Y&GgzF
zR#0|3>u^5nC@I5K_{h!lBTplJ=Z&G|H7KdPFr%iGTz$*hxCHKA>CXlBTX7Q-RG;<e
zUwdtNF|ENzz@Us4<;_xs2iW38fNy*lMa!FJ^FbfIL(xw;9NC!E4ACr5blR2Lsb1!@
zK+d=^UXe_Y@+dlbw*EmFqm;D`+BO_(YiHtW5y+&YD#b=l?4xO+dW>dKTnN&JH(#l8
zmYFO3fX*GHywmtXDZy3L9mwpX6a1Pqi>t3_RAt77*M?~8Vm!KvW6Wj-VfIt#J9*<G
zSYMo1CT2AL9c%5QFO&8u?^54&Qq$vU2XLx$9+{gqNsb+%I5W4BRH%M8X&ls=RM@CU
z5Bx>{`UgtWDgGE;A+cIvQe_X84spdLd;*(~rV@Je_YsrP_ac6=%73>%L0AUqwv5yu
z)Xa7LMECzW0K8JC3C;Tr;ScP?7qDBZj!NQ0mvr#FV&BDnU#N;j=BFmLn!X~srPe`E
z5UdO_?tm)Bp?X}NWzI2Xfriqe1=t<dhspkOoN}}@O|Qa5;#dvRXaM+D(z%G|1YDv9
zE1yT7^-D(pIOcSF==XX6pd+R3n?Ml$NK8w&yfX1brn~+;a>BQeCtUjA2aLU|zD@-b
z88!OvB0Q$>4Im~>z-@#S6MI7z1?EJ?sue~RfAknvmLjZZ)JapuS0R<NYd}=zJ*`8P
zX8=4lmTJwBjQav^858j}Jh8*ph0ZUIcTAHF&%?C-B=VFxWeD?2s3$sqIAowor|RS?
z&f0~G94nu$g7R1YgXb<mJRy9jc&plTRnyx?0yj^>lHn3YtF*qbq(&R#%A?JG#*@bZ
z``|t|NuH|Y+~li76|o$|r=bqSqO}>95(c&ka<-WzB?_k!!VErSZcma;<QEa}gh<qG
zGJLz&%G3-5{Ib<L{@vBjOuwIdrK2P-7z5;lRMrfPOq>Uaa8+VCtl-&)$OUyJL~$0W
z8~5S>CI0o#Db;%FE`)gnA<DY$4pFrVW?0Hq*KwJ$(6OYzPTwK3<%4JD&E;NtHQI2|
zRb@d`*qp<Jj4lVL7|}yK_FU3$r#o^E#9DhbzffYMf9t3nKJB4kSdrkAiLTAEYl(sw
zkb7J$$2Q+#HR0)2QDwM}37tf4dvtQX`q*#$>E?Bk(*~EDe*HBax#mu&xk`WCyhv$u
zBbqtjiP#hiCMpiVY(g~~-`&7`x%Wet7bDAdUfWjI?nAMhqSqS>DF+FV6tkT3Ps+C3
zTt`<f^(+?-O?&GgIPOGdOf5nN|NWl!@4QEoEtq8qTmL4ZmMyO~2gdtCg<??tx#ejh
z-s>si7~<n=7GvT_5@AacZG`;ix8NX7R~(b!w*YVX0r%8N>^t&1Rou~7H+cbULs<vU
za?&12Imak#PrH;YGrcvVvUXh^`A0TFukPI8$Nd{%;vBo~oKF5T4qarJ9(SCwFDxN<
zU6ARDom_fq`OVa;N%7c%Cm!!6f*3C)5-aHu<I=B4Iui~LTMk7$Kq55UK{3S8EKC2U
zse<ho3>(O)mI&u2uS>F#h3l!-{z=NFl@G@;JjJi`HaCr;gcv2N@X{{C^{RbC4M}Gw
z4rkv9*E_%+#JJ?WgP%m;-bu1qg~jWI3>dmK71KZxQyJy|RB@}&CoRyGl+}Ms8mny;
z(uT>?D9R_shT$=ok>wqiZiTXoMzE(6wj~P;N}B#c=BS1~+%QIS(kgPv(F3hjE7e_d
zEy$!(fr^Mef49bi6h@nYcZzSC<P+gGE(YHebJLadkyW>a&*XSzbr&s`B(DP&Y)h*`
zvNz=$bmh;hz3Ih&K3`f%{J<iwfnSHo2pk<t7`I1$veI4Y&S}|Z=1S6>q~B;;r%~<2
zm2PJ8{2YEU%Ga~QzmMyZ!4u}euutzJ%Ysr51jS{H+nTQ-TQZDT^5u5B*IFwLp339}
zBW+;Vn3W@`)V`Xg16M#QEXxV(xq6(n`tt3BgZ79AAx<LZBR6Z}YYus(7pz_CVX@-I
z<AW~h^o%K|ZOz9m-0nY_gy~&ne*(m>wCd^ypH~oNeg8eax@7Ac;+y-8O?l*0B6LV$
zxbS_)q^tX`V7{U4Gcrev_=f~ng0uyuIORmLt%m8$qWS$kj2l&%%)*}C(*`}YBrr$4
zxRZ!0nl7SU@-a6^R~J-F(lebC4pQJ7vN2=LO!GJK(Zy#yC)v@q4L&;LjxR8CqLc*t
zHs9=59Ltuy)6*CCT^ql7-7|4_yJcmkSy!eCvzy4UrtN=uqXtI8K5tCeov$xVVE*X0
zHcWE#<0)-C<Uk5InlZk_8UL>}|DWgJ|M;IKRHPLSqGWgC|7;)s$FK3%ADj|B5c{)R
z0Y|m@cN_Y@dmV5^2H@&aFkT5g{{MS*R~Y_O@H1`SAnyO`bNPEm^%rFaS2q+)NAaH>
z^55V553AJw`af@iMX_(MSRmpb|JTR<FV30>T-`=DF4n{U*Kt=kvhk9Cyk=UKEd9><
zzYFyG#^%%|_N>uV5>sT*B>(Tf?-e#`x3Q0owCulc&?!oN1X|7{$EZ07BG3?+kPuXX
zVSTIwh$+7yh})sZn3r_q(FRwq%O4!~^*EVzz%Nb)1R{;T7|P|kdx-^r%vz)x`bl)a
z<)cj@(~|~v5_uYu*cAV-?=3Vwko<4ogx^I$;PZrWRp#GMpEDKJBnN#}nI&GC$3^h4
zwMGc^;G~JcxP%BD0!%agYGLL>CpZ&Y#C-o_2iP7^^v43eS6v_Rv(I9pzf#Sk<3Q#m
z87;8G$bb4>q6POdigzg%&=VBsl`<F+8($<TuaebA8~QYW$JhYY@2Q|x#xA|ehTe)v
z65f4qc49x@@~r6*NklI9*Ar2b?>L`ZGO+|GsQvyu$^RCk{ridj*Ow711+GX}a#iHt
zgNQki01tTi8yH@t1`d)kGVxg;@N@!V*!cz>(iV_130#Q?xrKBBxGo!T56MvY1T1M0
zI}VuTw*W+@E!Nm0;UiQs@W;}B1VL=u@(kVv;_+%QhcE<MtqxlJM0&5OS!4i%CqC|u
z>Y)g*y@SQYc;{o66m#J}5_-(h3_T2&B(-1-AIBei=r1Odec&mw<sJn8|D655C;i{Q
zKtgbPTH=IV|9goho!GO$xnkq?`Y`2N0)rdfCGcQHUU6Thg1aCa_fviX36*#z@=AsI
zmh@K?JbHy#Kyp(61mKQnx#oLs!A?8`#NAZbkE+HmyRdvK(TH}q`lV8uin$Vl7kAw>
z=UYm>10$t)*c>TQL3whgd@~O@MfSh$kpI~!{`<xJ0;}Ly+y~o#f9A|awQJf)hW!e}
zdbHvo3gbZadl;|k!~7eh?zYKr3c!07Nvdox(}W&+vKR(YfXjBF!ai%o#HWe;ugi7E
zb_aVDYc9eUtq0Gv7x%w^I8@&+P<x514r>^Z5X+ou09X~I<)o##{#iZy2p`DB5BS}l
zP_8tCRFrrwhUN&i0G>CNL(iEOn9*EKZ7`(Ug6lUPSNqzvYvP6X+On60H)Mo+Vm1r1
zdp5PPi7wJuO9rIH_21Nk!pzfeGfz)s^z@qcE_&CNouMg5p&2$LLSz<<@oh<VG{+(w
z<O9l0xvdrxfqrAW{X*$Q-eZ)@t3^pe30*z*L_?*Id$+l$P#i5LR-G(&k;mqf$UC+V
z5iWs)2pCuskd?Ymtq<;a3Ov61j7I_|EvFz%{UrqXkOfp^l_v%`(UzAhF5$a81oMIx
zPL~rfmq272xs(}hOsNfKa$fg>4Qd~6?GPzFD_@$2rZ6@?zHTaE0}ri4qTfx*w=RJN
z5t$<o`;#t=E-s%?Sh^rVKJoQuI*g}!QJgF-#}(N$?MeMMHr%EfGknB>*L%3Zj0XiC
z4{>o0mp!Vk?$UfVtT#;_<9;qCD^AX3d6_inLwkI&%0`&7c{6y*G6w%yL$jTc05OG&
zl*O+48~rCgGULB?p;)o^>KNYerBNZ~6OZM^D+6AZSpW)J+RH0H@Cwt`a#cz&5qU{y
z_=(Tj(M4#N8IQuq;F{9kPFM6!yoFcKAQDIBm6B^_w7~rEsCf*>1OKW}@QwEwWy*yA
z^VZV$AH+az8-cGOgp+O-X#E)gfiKq%6iJ$x=1oJ%EyC!M2YmV^7~qD$=0sY$1t`4N
z0otz0T&zEak^{K7rFa9_6bl^-=ZB87Y5ffVAx}X}KUw+Pf#ItJfMOYlu0ZetEf*Rc
zg|#17y^q;5aRLmuIo19bV?s6FV8kos<+P4O{i3Tr0J@o5fP!EG!s$<f(=_*HQeF8Q
zKq1*JpLm6kSj7UPW&^<Y)<gv2uO|G}V8Hk?4uKMK0?v4YiA?GinI6A!Wc7gbvd|l#
ziBd4w8?#SkjV;`_3sN2r0py#}Er15j#|?qhv7)qgT)rBFc!L0gF>^snQTYcnxah4j
z977QGlaQ~#n1LFwrrxvGO{!IcImQ9N(0kER7*jy!2kbj_htd&diZI~KqY1{Z=`q&8
zI!HKp3qKj-K%};oU>+Y2j*nxr(_Q_hNdPEj%8c?c7C`P#gD@w#Qe8PzSK|LV9)<G3
zHy^{Gy1cYpJITLsU;-GYYqA4#v=Yk6s?97vBfv_J`38I<&7jB^$|9HfQ%R-Xz+Vqy
zF`h5y>iq}6#u?O$l#&p{O)g+xs6Pij#X%1(e*<f;Pz4v0<3qnYeFURJH~kXGWxj#V
zv9d0(K`JYo1x2YRF|AH$D;L0euS~24)VFkm3)wbFw)5MvGb!JF7n9x_HjC@S=fdMt
z69FGHL}&iYAIrOQ{SSZ`(F4EMl8qTceY?2Z(nDXj2s*S&Fb`nguaEZ#^8r!6Q67~K
zfL5&oKd>(bK4(*~tKEtruz9hBQu?~v&pS}M{W(M|%ZQgbp1QmCyBpg<-G6$6?IHh3
z*RJZT1(@hd0sD-h74Yyg*nk>tZxv#JUG_x52)FzBO6{CUi22Gb7<x-SlOvN-%91q3
zq;>!%U=C!IMks_X(<k=Vl&{{wG5!<KMPr-Tb$W+Y*))n~>mw0&JWsY1f+Ullp|a}n
zzh$Ng#vaYKw>MRSZRN#0#Y9X!t^<F7H^?%^9h>hSy(6zp-~*G);LFo01SEU;>JG54
zb$j1!q!`y<)qt&0&mif6qR(*a5h!n#3+MPMfpWk*LM9ILRR)hx0?&6zzurf~0UUEn
zV84wO&k$p3BQ$VfIzK^3ZCf@4d64kz$<2?~j?bmZ*!bsk1$OLT2u28dqdYLj1AAK3
zh<A(z%HQksW4;I2iU^OC)b1@%p5UxTqx~#|^34QLr$)Ri$Ee_9y#bXEJ?=Yo1F^ET
zDPU)HByVPVjnZfIrpSo+@1fveJ|pcd0~E&}vVx4_RIkN8(zF65)GR7%aDaw)m&NK{
z%@~gwGHnMiy3s?lv52XW%y9EF=B%>}Z^5e#f_G&Ep5)pe%`D7Dz^DlypR5fwGIhLY
zvA=t0H4Yb&MNO%MY5$KwMqer(Hh@bFVOo~BeUF4}YF6I#o2#|c1%wK((L*>pBhB^E
z)+VxUYd26@UIRN}oF+i>9=Su<g(lQwC9SOAf$=`O?GSfK5~_U@5F%merag4V|Bxs~
zhEDjwj!`Jn_sV9@B=AY(LTErrE6wB5dDX&BH1QR%oSGp~CZgJC6D4a75pW@pkkU@h
zmy8&b-wu5ay>36BqBlJfXTvn;LNn8uPMSI^=6cUyAp#46=U|Ek1)E4OzW;RvxjTV$
zM_a&19_USry$yUXWhB;~>#GKlC0(CIK*!G0>lycV><&>Im<eP1J6C`pKLLnH)RT3s
zNc?}nd-0S9fUL?EX;A`8coJ1rcbrnP9auFl5Ap_m0n6V+Jo;rqwW1*M<x>M2UKnhy
zDhG~D;XbZK-4AXvE6EpwXwVlfA?Fpwh6Jo~c&Rm??>i=*zYoSNH$l0deC``epv9?n
zG{U=V=RTe}A*11f$DqU=&tf6`^912mo=jP57`sPg_)b-SAg5i0lIqPw|G13R^|-+J
z4=AN8=&j6mKXDoZ3?80LQRyoxR+O4wu0w1M0~&N&;u%F8xfyD=#6-V#I&()omlW^R
z>x6nQ{*A&qb)@cfqdU|0Wv)E%x5qJA@BrZ;s&ik}-fVVIJI#J-g!LxYus;ugGS)Wq
zIlvM9*!mxdPh0|jEAg!$P57u%q^d8z?RgTI97ON&P5HOMmZ9VD4ZU)v-3dDnv9@Dv
z%N9vm)^mn=0M0}5F*U^GXL>bE@PnnYi&S`2$9a;8r`sQ;nd4n5f=$pIl9OVYYn#*j
z=k53Tr^+zd!k(cA0(3u{5v`Sa`d`&gu3q+i2+Ryz{ATQyKKBX#Sd)sg5BqyK5&M5a
zv`57E>(&z)<tldZKTq-3V==rK;a1PZI8a>q9Gov1<{XAQic81zvTaS8;5v-i0*@|}
zuK6hI^H~Kwq<NlOwOWho=5&HV{=IDDVX%6xM6SG8f&_s<n4XYbc*w<@Bvw`2r~*xb
z7m9~Xs$8<UgY}YOzF~S+Z#>5?in(7UHYPLLn9B+rm*y8qhAD+HYv5_tzVxY(Z(IlS
zXiB+wi;`tIX^#cc9AG^oU`fAT2Sgr~T=ZH=wwxy(6&6jBWyti!dy0P%`8<}DvEHH7
z6ePflGV(vg$K!Ii$^UuP@dmEtRN$jLrxq}cJ)AU<;C(q{tn^OXD>u9$NJzd#pt3a2
zpuiHgEN`TcxVO<jZa9@|iU{1@P8BmZl`OMwOR9xcAL!-w&QgmpAJ4)9tK}|7oQr3b
z5+ySjzu6Ah(!JKb(Xh>ySSp~@fF6#UULGn{s;HUgVz6kXXjX~jE9Q}B+o`ipscpqN
z$z>!e{xpDp&gD}r$z~5XOd^9qNj6#reoU+JCc3B1^B6mRB?yb>z@vi`AU|In+wJ7~
z^a0`E6;guxgB4Nte2XZ@@U(vK_+mK{bY7$+)40yy_oj<mm8D4{pr}nNz%4g_l(u4D
z+3!14p!n2OUY7eZDf{E+uMoV@b7i!>^NeHvc-?*e8<HE7rN%sV*5c`t;u~cF$g11F
zF*!DBBF+2u_x$!?$3TCu<p}Fb(d;Ixbs+KE{eoX8DB;(am8oF5=J<Q{HnK#oj6;FK
zS?xDHAjiq&WQn<VwR!?KG)g2@!FW|=q$>Ci^#=l1<w@3yNSRw9vsz2do`)wwba0om
zo7Y#Uh6VN-<uB#hnDOe+)9_|r+t^-Flb17SbQ@!i2VOIRyEom<MQ$mAr-ps*S%#<y
zH}(4||F{1>j-c4siK<~pEED$VIc-e-J=Qmxqlu~>wET6FmWPvix=aCYriNtSU_bX<
zd6!b~?YMx$=L@&Hu+oGq&zrC>-zX&|Wg)RItMk9s%5Vl)<PpY<$3#s&#KG#ON;rMy
zwi?Fp6qC{TM3yqPi7I&OM~IYIsyAadb@XSt5^+D8n+Ix7*Krb4b-5_2h2cj%(%C`1
zshb%`p{pGOVzfcA52!U`ylr4wC}1VlOc4E&u7qm!Wo&O#7W;OfgTwbYVp{35OKqua
zlo#w_JV6A`K3}(D%9A?$kKVwPlIf=QjaLehJHzGus@ZSe9;U|%Y@GC-tPU{H$>(Tb
zZ+WI%Mn>N}1mPt<3W4jdPrk{hqeDFQZ0$mR2*7wX=<WS(+ljDtF-cSUfVc5FZ_3Qf
zMnChme8<5Uo_$j^!sz5lR$doo34OZknh;&=lZtnX$-f>;f2<6-dAF-ygXDbWUH4{g
zJte)l^0VGp$vH#<mk{fJEIE0Gi%uOID+iNyb+%AH0Bml*>gV1dz2aPx&pn`5eidNb
zir#JfFg){i3)!PU+cLLE^!2a2g0k(<ca52EYjJKp4Ik0KZ)LKHpjLN90<@2jLMXw$
z{{-?I$Nb;BQHAA!CP5zdEy7?(Jnd8mqE49UG{__YH<`j3f?$tTPHUW_$U$?037_Ih
zlVbsj5G9S;)^p#2E-{gbH7C5%o`O#lOmA%Ltj!QPIRhw%nkP&N2#fTzhkC33rg~r+
z#}acCI>m(QBtzk0!4z%c4b0$mi9NYr>%OD8Qi>6&KNW55;==Exmu2z{vJD7k95D6B
zmm&IyMlCtX^?=@2HQYoK6C?8AubB7$VeGBrqU_qeZ$Tsk>6CV)qy!P9yGua21tcV-
z89JqVXsMyQK{}-sX(^FzX=!-ZT=#ul`+oMb_vd~71?FVtT(iz~#P9b#9H7<HV5Mj5
zr2P-h#^l+_id3VqzJ<M5HHa=$L<}vW9f96!jBY8Q*yJb-rFvu~p~&KJ6!8az`IA{W
zV$EH^ns*Yy4N?n}N$4$e^bn;9%A;8#sMZp*L+$-qG)blCO2rdW!}mT+SF5qp?6c^z
zkFO_ezLv{m*k(-oneiDg42nACrjxO4JZ-p-{VK~0<_>F#ts5Ft=GiKo4dz|iM<myB
z_gCF6ORO`Cq$3mXZ;Rim)MaSy`J{@3r2Ugu2jUrB?;czCy1iOg6)+!;r@DXCGMv}F
z>^ci|3zYSgTp^+!Z&uK&G0y8=bG;-V^ZjHwqi_2&FOy{8*HZJa(XUqrrIW6=UX+^3
zX+x0|M$1rjVzg?$C~Yi3o3f)EmTQgun9&sm%9&bRR|dEI<H=(%IrFG{r5nec*olr!
z`Hv2vR0$XM<&{#1NJi$cBw6>1#F6EGulmwcd<c8r7GD@v#+lhgYCV8n%LlLQNikmt
zmMd)+_63(-e``ANxUOZ!Tylyl*mA2m9Wryo3XVV6*gCuZpgnd(d_zxperDGg#{+#u
z5~Q-m>t?pLFY%EJN~c@jWK&TB3ytj<J1i+F8(5a{c2do6V#RUJ&#`KB5Hxnz@7j5&
zb#BOSqiE1PCu{hVHN$fQx^?7W(=a0Iv$N;75!*mAQRVWyFb?%D(M{^1yfV+ZU%1K2
z4e4~XfTGX7mV49Z{6zxk$7<2x8|N$fd<JJj%_oj+Lype%DGTdpobgn5$6PcO?K`>D
z{qp9`BOw)YPHq%zH7l8k>8?=d@5Nr{lP3pkarp$k|M@KZS1ARqF($!^>70NpsH*Ce
z#e|^E{66EeAQX(#gdH#-=5xhonk8yWIr+1@E4&>)ZXvHDWR!c1OSce8Ws|m%w_^^5
z4kj|rC{I}FX&P_ZH^ck7!s23d2J`zU!*ne01qa1WHdVuwU6sOhNMIl*kVVX_S0w{Y
zLUu&8h(^wZpl*HozyqW~gP;n<07#eqo1-0(8&F{<|0i;scy#Lh9DNk=8+q&*lpx)L
zuM&-7ge1@kMoFD~W>7gQ$WvFm3fI{fN;pNt$by>T4vqWcI0;Yb-LH#4jwxK#=WMK9
z2M%q8yIYXMGrGTtF8nfdvQqDj;MK!54pzawvl;S<El@HAY8!sjFGQMzbg4SL><?>p
zdDXB5n>2vXj7#OcF=BjJ&0*nTC>7o_x!;89xcx)y?elIKH!5|lLm&IRnzdf<pJ>^Q
zu82uuK%t^91reikgSb`8a<~OG9TEzOBvd3b7?B#c(ibTJ6v_cer?lpFN2)bvY3fkG
z`>HmU^$?rO#xxVVQE^Ba1~Dg`7^5aL>fB3TE1v+~rxO~dDk-}SpCTHsaKF@zE=mL=
zwL4}_1>>5y*aX+L+W<U3*GE$Do8v-mjuo|(whqvzOJMj^gTVz{J@9X8QJc+^#nvml
zX7Tm<q4|;K*M*qjeB023+X5yPdxF2qU=h!ehtHCCmYt5HH&y((_f5Y!0!duBPB00b
z`w!9&AZLFe++=SWGBNB-*3HS&U8HWx3o4X0rRN|I3DH?g0(~|e1D_k`(MN1>6d&W;
zL}&Mt^ym4;^S`6coyMbAzl3Lkr2Xf5Sra;)72bR=WfMp53F$D#<X|NW7>LbnD}~)l
z4+b2-2KweXc0$tZLv$d-b#RvcKqV6}K#ZuamK8D5;CVYuWUEvEAt->5QGw=8(v3#R
zNcH)0_z<wzKeOu8>y0F5Xu`7yA$4Bsi}B@m%VMDEM<A=O-U2H8^Oo=G2N4c0K>?XG
zBcg;9b3Ly+Cr2+khelxm*V3JOUEHV&d>eE!%!knIajB2Rdb6AG6kz8YqTKqfKa9~v
zbF9-cB7?|#=T{%Oa}8>wh?}G3H_>W-*WFr|r^Ts`@|^q{8Fcz`r#;Q%h>+I_67^j}
zF@;moiJ)+^=&?pa=pEonrVMm?3XC+Ct11&$Ipi_e_JfJECVaVvD}9(vFBc><Qlo5l
z3Qa=wn(Swm<%SAP<wg}XBc(TwU|BytvtNzMX1KqqwW}TZDdbOg<Cnqpw(0q-)zM=z
zC-E#Fjk(DU=qBfShKFtAK^G?L4JaC=F;<Cyu^F=QLG>gz_A0ufd>xVXX)aWhu{u#a
z5ui8t-dIR~;@wm%>0n+4_FFCZhH6{_K=NY3)0z}fXHz1%XJwp@?Twu_br=bnkaSO0
zAB;wRLdFJTR$u3M5@08r&b7crnpp`TL3z{Be6y*xI~I6}=>iqZl+Lgc0YTb5N*m~G
zLfy-o=iJ-xA~v_=qlvWnf7mntN>3ohRB$0J5;=1vwmAQN?7b;=oc4HT_b;;X^fNkS
zJNTe}L$vV5W{-MB;iM~qH+;59X?YNsIXu_R(kl7)y2mbF_a(PRao1SQv;jRs+LRy}
zbycElW{*4Fj_ir76|RUyySB;p7YcZkgJh<B%F}2|p5)+4wq8;Z9_6{F2RqDwmoI5B
z8fKnDhO`&a_28e_1$4MyPvR@NPg_%AX@!ney@uZ(pslF-5lqBg8dY~(V48;#?&r*}
zk0tYl?1ytDMOER$>iV{JFv=7QN4@^#waFw_#4FUSgcU85>vv^y#%0kP6rr=&iRS}D
z@fwFXXC7lkgB_~~;W!-gbtwMhy+XC9g?ZC>MQa|d429i$<6kmcG!;jj#*zyCU#Uon
ztU&d(G*~qs`Ed&KRWT-AObnv-scSf=_xsZE4d%1Gb*3y+`n31iU>0-G7atY}nj(Fr
z>JF=CerQz5g=1|Mtk*@$gC9hGy3B8vRAg*+t>IQc`CV%=I;0acXZpl{Gu;`o-y{B;
z1whgKE$Rcy(`Sb)qhA!?CsieeeBr)6Nh`q|a%y@bYkM&sE<{hM{HH<pd|WbSw}ker
zDa$gETHTL(HC=d<o>KklxjH_Xi?jB5NtooMeWb~Z&E3jk-^;&KN>WP{C)&Yv1-ZDY
z!*p9t`9FXlPtNcRiFpR(z41W4#XBm=T+=pmt}-C{j+cp<-%)+RDGoOgpQUuaPgVG)
zolQsQTuxOUc5OG+>b%4_ty3G<`FJU}yDQZ@w)-KjBl7~x*cg-{_8S_Jww8MCEuhCz
zcph{nvgr>i!al@q(u=0$n6qlUsBZ=JPMe<{^T?D4VKK}ok^*zgvu_a;bQaVwNwmIy
z22U0Ox`rRrjuZ<MW{KOCJ!Rx`)@)f%WAwDP2czdZ<~;i;6nL<-wX^g%e10jDU=ZjJ
zE5STXw!TM74#X=9X_G#inq%UJ3Dd>XH8Bl<DoQeRJ%xa-n64??G9-a|J!KQdC=J8V
zf7P%!Z+-_#0D)cS(#PRSEIwt8>!}~p@T*yrI|aibXA_bOrqgOaVEqZ-Kfb2_KF!Dr
z0i1gM(zdwiCZ?}oD)D4Ba(2lZv9vn-2?i$#xs%Fj$uv=@XuAg2VwtLRg6F#*D(xxL
z=9HH9)}sko`&QG=aRq_0!dWPRD3p)4q^NrYoujzbLpeh61r>38^;$m&ZKS`le+yrI
z^qvIvJ#ka2EU=M$j~l8;MN|jX(&!T2%DRK6{{T?;ctxHCu1-GM#T}_}+Nrf2g=kxW
zx?54%ppWD>l$@`pEJ|V?VKMPbMsba~G93!iQaHW;^i*6};!gI}iRh=<@8t=3N<%#|
zCfb_1YI#w_Ysw5GJ95&LYtNLG-N0wm=9ks&lNGP!m2_?~?3G%Gz!Mz#PJX4O!h`v=
zsWXowQPsa8@|d(@%x6<7wyS=zyTrjvVStoio^0rYa`Vu6)J8-?e~m$7@k$L)d!?JW
z92CvIn^LLg4eGhFo`v>)9%i>_Xf$UTeoV9$?%~wX`oS5-tb`9&%*q^%e6GQ^6(5sc
zg>MqqQWs*7%h0vUj$$Mk{2%30B&n?JE;phs>Ueg-7iNPoOTHSeKa;ETWk#pvP^noO
za!yvy6;P+PAlT04g^vn{qBfn#v`~x=(&?^wm;8f5`6#rq+|GpSCSFTd+60NtGiHTR
zn3DAS{L*s5;v^7*na@1Puuf(~>4;yknx70Zn{-$^GOb9D*{fa@JE5@c%4E)PGPl(D
zRR5a^haw;xF5!Ac;vFD%KTkHLWcKG;(1NXZR;ii&Ti|5;hMB)bX7gn5T)^zpN{Uc%
z&X!55rR2GI;*&KF`;yc9YH*)Taca5OLW#Ul_YPeA-)k?hViv@^Ivb}%)sVyWF0X8T
zu9ky!gqYM{#jyi?x;8xVL@sChuRU+~V&}=V`WLvIU`5W|zz|Ir<yMtahTb8|zxKV+
zE=|XE2owiXh2?e8>Bsz@7s6(@U1Ehg73`Er=Wh3_u`rXY`2r+1q7END?b2}MxONqL
z^LnV08_m>&WJPHnQO+AlCU+=h_F{`t<j&PB+nQ^=Ghz8kO5#6|D;s}pzfxR4O)495
zFj4Kc0$mV}Ar~VGA7}YcKxrvrmo?QH85sYyq}I@TI)!?sQg`8MKG30?UM@t(60Inf
zQ+j65ahUQ`7J`ibh&HWy6fK=#Nj&E8`s+t^v+JZs@3|ISRoev?Vq}lEquQyXn8UaF
zKOi8eWa8uHuj$<u!w!O}WjbqREekYcZ2}m>=Oj%h__i&Wy_h&@#VgQ#h7oAP*Chcl
ziZqr5=@}<+n?=ZxLAL@o;<G*t?Oxx)mC%sQ>uvH3YG_l=P8hUrAX~l>O1fQAU#!#m
z?E_&%=!v+bpZ62g<dqvGD7l*hILwsF7>J~l4HepDb!b9Nd5%@?PF6{FrYGWzjnk5%
zvf|x;Dr;%jyaxlcEGe3A^j}i;=+@~$4R_teLfWk3@|8rk^U9MNy~)VU%yx6FuNoqF
z3WGNdYuqh*qi{RPR)}K51wTF06qKbihP<z5g(%duSTw!omOL4Qrfj-+ezaD<tmNz6
z?SEbOwY1rKJUBZoZ$I}gU->9<<wS9~>SUEjqGCavRkL4uF^A!5HOfmXDc1)i`+M3v
zxJaOtOH<7h_>Zh43{Qk(P^3b4+UmgY-4ad^R(z>TsZ930?a8-|RF0+&S@U*luVu^=
zq}@MNF=fqks^$99>(S5e!oT2a&vI}W#dZAtUd;uEeF8_H&2lQ;^&zDxLzi9f2Q^}x
z>Mi?MiS+aX*Hz5a7s1_Iv8467(<Js`%ROBvxrG87m^$v)%H{K`1QK+E?EHS?a`jD>
zr8?zDV_#}qFy28IqfwTk9lF2SyF}DIb8RW`&Kt0Rc@Ppxt}2ZUxbBwP1_P)^>ri)X
z{I!TZ<^XSrg>ta`%tX$YQpK?q&+&@g24)+*Se~8$Aq&a#tJ|~o%uPHQXeM{YmVK$l
z*{OwpZ%qre`<v#Aob)U~YPO+lL_sE+9o7V=zbQ{@_d2|9_Oz$5g|Dkdwuwej_rXuI
z@-(M6d)?dgh}|~$IX~)_Ax*f}oW*;mDip&6UTrV-XYMo3-~F9!v5pOCtv{>|5xkWW
zUGs0ESy_&zXp*Zea{F{1(HatZCm+M(zvvO;Xj=@~%5=^O_3#SjSL5Axl<h6hW+opF
zPM&)^THIenlP`@lhnie%?=gB-mhocI7~8KdhQ*%lhBfU&2Pt!e*|$C)=T&KK0qDM`
zZ5L`8*R5r{*0R58%0^|&hG|>NnScpPp7gJoPt}ZzKvNWLz;HSwUKr4_Z0zh|7XQSN
zn#E^5G964FF^2RcZXJ9-t^A-UIOwR=R1n#p<~Xe`t-B>3TapwgeDW~Y9z1#m0y;88
zvwEuOZ`~Eob0Ytes+);Qi+Xgt`&K({lA<W&BxHsS=i$TX-|SvuOuLo#bmxK(Xy=O`
zwvq09fSRR!NPba0iu2xQ>+5pMkYoD9v3FOsoUWQC<=5_-adY*tSdye9CU)vQ81sJk
zC_y#-FA2jb^zZhy4j_~rjIWBX-dns{Cvr-KVjn^F^ebevq4w?>m(6<(i)|kSyKmmK
zx+C_9)b;<dPhQ|;XBxjiO4E@VbJi(8A%(`LpC+6%$@4wZEeU$dvhk7J0|+4OcGsY}
zum1hugwb}W<TK(W<6fGM#rF%Ern4Y1-26nlo3T6WegSJ!$USl~$qC#*Czi1WJZnP-
zw6s7!Ymn|}Av0x@L=Eur%kt_V2fd|Q^7nh5`y95H{85?z`CDWpSw9SQm+!llh#`0i
z#y%4&(s{xSi@O?TT^;3ZELBnav<1eK*lQ$i@RKFzZ1(2H4W2YbY+B~IB=->_fo9d$
z12bMQS*VsDE8*jC@*%B*>DZM1bo;VA7d?3F+DA|K?G?~faF6?JnPJH=Nafl?Z<dEy
zABKmbK}}t!Tn5eZ>uuF5iPY+m!_1AbTgC#d*D}J#i_pcR;l+ZA^udl%UKoUsZRTjt
zK8AhQ*Hqfoy^m*AqfOE>tt_p-$2Cy?Cg5PhVLh)v&SS&D{oH#G1=D3nW&Ov6Vmf_9
zEyN4<6*>tM+FJ%p0+_eWt@eDDfxr46k0d-yB-l;M*?G422be?#R!c#;5fAU@4OZV@
zXwv6_0Rq=muV1@R%=xqzZ}_OQ=sz6m$YjQ!@!J`>-uCveKmCJ^$!;`IwZDpa8Yt}T
zNQk-mZdTxOP++6lB_4aZXe_w#$stgW+u%PMDuRWZXL{9dT>$Zwtu)dT$Jxms$prV`
zoNu;##w?f5z)0{Po6SWK?KrV1pZR}xGEtCQMSKsbSCcsX$qvUVUZD1L>BTzw&rCHj
z_GsmW9aqWoo5&F3@K$r|&`6oe+DE(~4bH{z8O9)kIKFw}7*Z{~c<pd|v0<K5;^_mt
z9I4i3<Zc4^rNr=({%---g42A*5K3GAb}tGo7ab(G0FR6SqrtaME_%tQLwNQuc^0*e
zMj&U)9!q(kh?8x56c#);a2oOATP5fH?DL-<`!6Yp-DNWwRH&=djB1v0O7sAqusWLj
z%=!c#**{mQp-JiNWeFMkMBV!Ix~hm&Rjg!OCiyr(vI4`$EYL9+)q&l8xmNs6l8?qH
zscx<%pfvqPh*D#65hgfDn=7#1sD)t?r34jggLAGOj*;Xvd{I?PF)<M^R!n}LJ)G7G
zie7sREG4xsohL^Vf7LH#%kc@E7q1Ooy>%-3lrP~k@ECdT9rHB1Nsa(7KhxZ;V4yI!
zA0clqd4w!qRjCf4#;D`6$O!6fsTapbW5}akA~z{P>pJ6B#$A_8kH5C80NAp{CN9e6
z3EVX#X5?MeJNK*~liF(5>5bZDyXFg<W%Q*6M3-#}Yh!lcd=C1bwyZDI5%vi8TfS2I
zEa@i4ujaektFKm5V1w95FXzU86GMteBmAB0zV7OB#^3xMQNm4$dz<O?r)uC4K}Ibo
zxXcar2s9soQiTE}JWIg-!9#u*-rmF&?bsa6sGxg<HZd18dgs1L%o4)HceHxH&^Se>
zjF&=?j#+Ak-v+#sf#mw_YtKtpJs1tv=Y$Tr)BDEo$}mVX`sMlDnG5#5YdxK+f%yEO
zj2<(O)A4xHK4#2u(Dzh;K$HG+%+v8LkpWjQVr%9Xl`yV*7{w7jjt_2SjM--j3Hs$V
zbO>Rq0$0f5p^-FxywZW`A<YN}t;7J*K%78#@ng%kNf4-S5K@NY8-)yzG8V6e3FO=6
z-7%Ul+K!JLZBuVOp=8XQXJ1QO!My%ip_Z>;`AptYp?FY43uYb_o8{@K$6z)(L?7-k
z;dm4zke6W=t<jMRNqqIHx^IX0p?XyNEa>0c9L81Bvd7aCR^$rq0BM}SC%#INEvXw_
zP@2S#g1J!8ELZb?0A2O!dX)D@meTGPn4e+-Ut$psZ8-rl{UcQQSvvYXZL~(R2Q+&R
zel%lEaBR<k!0a6{S3f49sw6hFC7niKHfhOe;q4YDeExt>WKKj!>-CYjcb?bpgyZW#
zmf%xa;$<ds#(M+Oms*3EYyRAVBa^XFwpY(x!aW{aRE$4GX1GT1q!oz`X|7caneMhY
z{DRdUU>+2yj_hpuR6|iX7IrYE;X&KGu0SXE?*n@x^xiS#cGiBl`&C?VKW}F#_E3L1
z%escXK%kMdVN2IT=BI(`fBNW|PM!|yZ@aK?z92DEDN7K?@f(gv3)4%#P$85ab?hDL
z*OW3~sTTS+AE3xiu{Iag{p5+`wE#(AyTBXNqyuTI%A?+Vq3wrF+`SXShB%<FvkH0-
z{plO|Di)G&vTjvej?sO<2GiiAVc*&bbQ_aR8@R^DG%TYymaW`;xl}=U^!W0%6Q<=2
zXWwTv;=x&KpJ3aR`sVaMbwq44XT$m_HJpah?0o|1*T<Z4i>hz(1z0YPolMFo_8^Yy
zPtwgm&3;g#-6-^0J;VY0AFil{+wwT<WzcH{IPJ_Zwix&C<y0jg;e(#Ls@C6lOo=mY
zn3a#aG_aKJ@}E?f1{+I8+GFww*)e<;lll0ZTQ9KPw%0r(|Ctx5jYEPWADj5!oA+{d
z$;CN=*{MCEphXY<{@wjoA2Dc<stw`|-c$T`lJbNg<X=+QS0TKAS11Y}Ai3Bfyd2yY
zMylwvA)P1IHDBM2ynbG}!`FNjlfJ%%f!VGS)@<?tU6dWrn?cqQw7cEQ9wyMwsPK(u
z{{V~Zb)nHrv1XLti$1pJ9~hL-jX~YmlM~=_Qt*TQo)f&336;ctl3EY|QINe)z}vNQ
zg)q6j*+Sts08_a9FEYI^w^*sKJw7xZ4Khdei@04cp>?UvW)u3OY=n(MCsj%LO=(*i
zDPr8B1poZd+kpl?AVw&@6jg%YSYy2(_+SlSHfcJOW9K8uSx{9r36tv&W}JNeb{C8o
z4T{GdtYk;6Onc;bq2*1Z$5K!R)JFM*kTS_~HkM{qP29QH_uRZ2nrwmWPCU+kIrMUT
zTla%;u=Ug=Z9p)uq-+q-hm}u0SPtlsrtEzC;OH>I<pI9-)B>=(<>S)sTas8nuK5RF
z5WjkT?ZIc+eiT)<S>Gz2(NyW{8>}w7&A62vC6=212_k$3?ZX!Ro%7A(apTyx+&Rg9
zsKMES;ZK!*ExD{RxK}szX;{#`Z8FRjft)r|+29$^C3Gyp*=q!6CCEnZPjl1n<}YI_
zVAjKA+JVz>wPqA-JHauIJp40k1~MVWEcbF-$czH>0c@`{p`}*9IWH$oC?Y3h%uR)7
zLDcBfRk!2;jVsIY`SFRpyt9!gh59y>@+P`qyn-TIQolfWZLrJh^FPh5&v-`O6IMus
zDaq6v4BzHf*Ey$UpYND*G;A-^PF<2VaQ-o~FG+6~;EY-MzLyYb!P48X@*^T-*xK`%
zgY&wy3WelIy6N|?{?&F=%cS32C`=626`^eCKF~re?xko#<1hPNnCx6)6ltwAZ#dAT
zTtjc~rz^(3mwdwrxt;{^?||>_aJV~2*BRs&EY>W}I_stb+pAYR^?QLIac;Kdr0Qr>
za`aLC_`4M}eOt;vr=(D3KHGlR;VqgSBlW~bZJTf!&T86tUiLj_?RHVad2EdMSEG-z
zrES;vwxq9+<dv0);ClMb6J|vYWpPnWg9&F}Hsxn%qJH{0!?^7086$5vLEu|~2z;Zt
zmT(rR+^uF=E?Kw*5n0!i#`B+k!`MsD(0wvY?$SNJyX2QST+KbRdB;BZM+Wb9n?>{F
zLW?oiThw442$QV6d^zPCd%CP=+*39b^`uFV_n+|U;C25WHCPeX@?1uh5<WO_sRRZD
zF8AR|EvUG#af7x<4LOT(G<T~JpWrXIP;nOCOWmKJr}b0e1WGDzl~+>kzCUnRi3=~3
z6iE?)Ru|$+!ah7{f~Y%vU~jrV*81n9IQOB-m|x!BVIMUKU1I1n2?-)j_``L*nQ`P;
z?6&s(wkwG;Sc=bqnptk4>_zSL{NifRv!S)}=HK{rEp)|MHJ?KbSwDs1i>-d+-><b(
zA7GNxdr?~LN83#0XKl<lDhK-kk*nksP`^-5#isM(GW}AagdEsHu7u_SV;CzYc%(9<
z0XL=Z7H6dmi@Qi2)Y0IkOV?`K8hxr*0i$`!=V$3YwVnSkk;<iE?IDE|ZBGDG>oaXl
z*bm+w95dOnH`ve&ZxJswGbc7~=l+<e9Wc-VkM2PG<~4r$rl~fgxg15cfc%x>-IV@2
zxCYZ6f1Q~5T8gKEN#sijqVhOfI~WbKyQlKyaqqtEEdSkSC6<^Z?xf2<V3MU4G`~!3
z?Gt0wNRG8$>=lqFlGLI6aLW(B4_|G~zbrzrQgJZqj>f-|NKlOQ@W!RKAzv(U(0OA|
zQ@d6B6-G;U8>&M<G@Hy>@YHUHpHbL~?CWd}=UuD&G*coQ+E%`e<(dV*$0V9x+A~cS
z9E6UJOG4#K#f@zOnp^P;%Je_RwKj4BcKwQ(lDxfni+O!$wU!ffpMIa$T@LN@Rnk!C
zO}ZQ8-Cb0m+?^RlZxQQ<G*mtEYSa<1Ev~uSc;WreH>g$I*S>}geXd1>BoGtw_O=`|
z7M|87L89n*vz3oC0~Df?7)0==v3kOLLb=fs<ERp4-gD=*yZr0E5hkZFUwXTU$%(MW
z!i4Q!$6n+J73ah6T^kQf=$)|S81d*<d9Ja~a25_|tax8f6en(W=~$lR$_^xPmyanz
z8@+diqvz4(sEYG&(~`2gzkk(l6p9wQv=~)ng=<8l{(SLsQc)S6cDxFtq2A-oyOC@H
zVLG&U3#HGj{GMLvOl&uA34N!VTw@gfyfhUwK_|TA1j<ha>p{a_V9q-JxXH!wQj<Tw
zBmR0Z6g&L~7^P-?s>LG}8n<@T5^9@Y#~EzEGE$V_k><+GEB^<Co%HN~9p<=9mrUEi
zwiEB;vN!YN1?aqI$=2VSt(^lkF;{vv{;k7~W-QB|&>57fdORb6m##Y`V}jNyeb{OI
zd~%#dAvm|ESntMM+s(OEZLDsK-y-09b{5Y_Q#I$B_g45;Uo!K!G!d_pSreJsA244#
zTOhF!obtj(Tb4y*Qxoda`iak`pKTSEAkO$1iGsp{hanknX*a2zAo;M!t4lL$SQ1(~
z%;wI3Cm4N)#!@h${rG*sUb)Zow%lD_t_itaHC1P-VulG;;i|Cc$O7|*89ePUW#~;F
zP6^*TSF+;Tlvc($D;f#dyv+@4=*P}`{k39+k>OV!%RQm+H0Hp?(vj`;TmuHyu6+iS
z+N_}eDC2_1JgFo)1pwREGpi>k4}f~3Kn#-!TC(mrc|S_vcR!I#i0{h>EYaKZTq<K`
zneub(%Xww9rxmbnh_(@jz$<B%f2)p62$0sLBj#M@?WOsJ*us$$DM%NMV;>GZZ!%`w
z$+NG(tR&W;B!=--{ZyUoqE2+G)XT>F#TtE|uId~p<dEaEL%ZTIK9lf9U5IP2bix|>
zj&oQtT3l__3&}Bk0viUOmHe<oG1Ca?kl_)U;mB<A&0zSb&9QPm#z%3uMbFHyh{U|J
z+MU&;w4GEiTBdGX=JAw5W5oGbduWA`m*XC@oMd@|b1Nq%2=7$r<Oy{gb3QSR&dkmk
zXrCXe;f@s`n;wc`(<y7G{d|w<Q)@1_J{<P_GaAkUN)ijutb!KutF7Gl(%S|T=a7>V
z_}-MFu8TV=gD@fQLbBDX`sxE|LxnVrRhjz*LyWR75D=TNmX{f>sw3cpA!+%mz~Zc3
ztXGnoQ|^#3O~PzHHA}L9mEHmdrU%LHCRM3q_#(an-cMPtTS4%{-8RH;`DN|C9o-Rj
zU{E_K;ZJfjudE#>+0G6d!1iJ*4>UC!$G*Ol{W<VH>ZNAs)R?rrj{4`SAQ6f!{`w<+
zRr0?hYZPRYdq}VkSii@mxo~u<0}gmBc!`g&A0t!riSu{*9{@=zzs2%dWQ(Z7ReLEq
z%Pmg$Inm!Qo`?itd9Tg2y(d8rN}ac@DSf$B?39ceMFJEQPS&JwF#LoKeS^+_ja;W*
z8E<D_9W&?h^ul7$Zc+@@GH3V=1y;RgizeGl0e*{9_?ALA5cBZK9<>7n`P>9NF;DpB
zTYFFp^3u*<b(N>Ccj*SSG4{K;Iw5jexg+>SF`)v+Rc+H3qeZ)xcl{W#=$|J2H3`1`
z#u1K5VRVIR?Z^@&_JZj=o#+MH!Yos&+`O|6U`~$0giZ?Dg3}thRTL*mgEi@VK%>s$
zV}EBgnmdmi2F#}v9$Sv-z}mwyqH8DwDcfp?U#im!m>Pyi#dLHC?=V)x1a#JreA_?)
zLTOuD?^vM_WA`I*S~6l`rSlcnsly09C8|sSK@}fV(06tM<A;<dIP((Zll(q+egMV;
zmw5>1_x1IG0UiP6JEAqqpJ!Iu3*1}wJ*>%QH#BI@$@L)u8_Bb!)Je^A%u6aGvJ5ut
zH`0j`;ndq~Pf2IPSZGDdZnV>LANdBR*}R}+=z5KLi(y*Xv;dA_D;q2`;YhspSTy*v
z)nqtlknrthc&8t?RGG)4B9lPLHc_SEYnjyQZ{0`l8O-3+l_w`}EZ=Vz=dj5qyiag?
z0}9iMya#Qs<l8n1Eqn2pD_R_oE&GQQxLZMa@2K{3gx<ev2Xzmp-zbl7(dk90vcGp7
zzvaL9%5frj!*HmYN&9AWUA0fJEWKwm%CPH{b9C=K!mVkqJEVO(d7<iGZ6<x<a@^Zj
zKLu}weH!9<7S4-G^!)NN#1i>`O0iv13u+Gxei%hfvt5n0L%okswn+&E!Tw9Wh>Knu
zc~6c`FqKK=c;CinoSjOAuSPA6veWQ#ENPeLtW{O&Ouk7xO2p_h-YZ~9CJ;o0jIMnD
z$iJ<EM;%5uOV46n$(SY0&oj&XGm?%(D<W?`s`lT;$p}JZmLnr1N_;ZRbfCWWsh-r(
zrE_1^g-c)4xl1Gnn&UB#h}Qal>ZM+M?rl6CEPTvt>i-07DUdh+N-;Q3y5gl^@WC%r
zNV)q&Wh_(jhOr`ppn?BE!|mk`IbNV74~3>_ph%Zk3V+55i7G~TSB!WvGoK#@axbv@
znPI*?H(n|s!KSE8ER&w4cbBk1hK`xJU2YYy+G*u7*ii)6Qh;(YUyG7Y00g?a$|Cas
z*WT^-x0_wCO{HGZe1)zO2?`A#nUvK{$(@|nNqLL(zb(Zwepvdu?aCI%z5f(}oWPHy
z!bt~aygEEos0T1X=<+55eG`G8m|cIv44V6=x&+EW9G{+aI!3eht)#1h4EN$9DyY8V
zH2YC@g0oV;jC-JyzvRp9vLr=x$i^o}rD`yBnjSG}T{-R%FVg(!tMOG(7<=x(dDx6K
zoJMo3A{-@f)KAN4G!RyA22@1~iMqDDSLt;$7)tJX>0bhZ%A~2ioMfmp%W}|BQ*e*^
z5OqPbxFZ>l9=*q&=4m$E@x~Rcf8<U=rt>em&lc)aJa@3iUfA5K%CjbbV_Yz@N%RMz
zqBtP3kb`41&i<qB88wwl3G=c32Pfv-&CM9e=-1pwY;FhA-ruR>3lNhGY%ASi9)F3j
z-FvdM+m9V&8)-7os6C(0fFToKA5uHF+PW?Uk)(QRiB3TD$S;cejvq&Xy_{(*GxMeT
zZGCHw7mDsBbH5C(#_6Utgux$aE9)zS>d4<Ex{YmTn~?w#X~w6hKAQ^qT{?-++M5-n
zCmp_Mg?Nu&Ewd~|d%FQCQG?4z?pXJ@>8W3KxqVXVL)?1;X2{E|r{R}U>2mzNIs)dN
zniIeCZ(NVbcWS^HEDi5_$<ymnTXfP5MsehAsq{sXo1?rn9T<wXT|Tw*++8z+51yB%
z0mcCbzhz;`*HWE&Sscd)M`6Of^d~R9_oq)<z8u59y{B9YYYXPUa89r9xT3nG=Jp5l
zL(6)qkW*9c_CRu7M6KO=%{7<%@6;b1rS~4Pf<|%(cB(i=bQcyKO}r~%==Hi?)fZy<
zYPp}H@`nB7=qCiojHu-R|BxundKy1`^At0@@5S)*fI9VBO+?;Zr!GMLUYY@g_~s3#
z_}?zy(J)_vTN&!9L_nGIfKWY0?X(&Lz9$pRUY3ZpMlic+VO?{$YVejDjFWB|*&|E*
zIoPJyZeoJbLp2Z3H*Ya>x6Zaqu#>djVqlVlGih;k#$tE<A}1~%laL^;r+K)Pd3IcU
z`Z+;EXFw)q{+MzRVK-Im5ySl4>yv8-I_WRTetc*Ec#pJ6h9u*fuiP)5$Q!`GCNucF
zCuBmK2<5K6q_+HUjS>7stnbIoDy1VD2Ch8+xtX4Vx5$J))`06@B#b<`sQV9~)c7i#
zn+=?2^DLI4g-@`Er-jcNN61tPxM;jE20n8760vpp_j8~}Np1nfLqRNc6!a&Nysa7b
z``=m$0D3+D7jr{I)B}lI>+_PucZ1JB`P#iM#UevMvSyTGmROZzPh#L^93RR{U>YgF
z$wC}jZK;%L(8+__Rqb7^u_mLNz&Ch?J8I$48SBtfHQpoNq%pcuN{b#toS~vog`eaC
zHqX}c^RI`MTJcY)&>aohV*=`_Rk42X0c=UNEO$uxZ(4Hv;9yMqwECpWe(F=$qrP7O
z?Ha_WwEOO97<eQ{$b0l=2>>>7^tOp|Tkhaf^rjk^QCX4lsTLFWjE1OjSE~lDe^ToJ
z8Up%sH>K@!Z^K#Rn5>rY<mGe89dNgfxN+KZUye+w(@Wt4=}d9`OZA<<cSwRmPz&z$
zm@~_qrYgzUK2reH=8<va$&W&-nU4`W&chYByJOgx)3@bV;U_%lw;#0xuXby6=S1ct
z_+J8|nodguYZZW<l%0uv<m)qEI5BpPn4$eWJ^3!qM}6=Q=!but$aW$-TAX;njejyY
zsZu^=;9;}#o~97Ob0bvX$<h0^wC}}bef1;mRyo5!{v-4jOgyJ)v45%^{}xvp_27M{
zbmD{_t<pMdn(8{Kyzf7D`)jXs$k1dpGf0vcNRTiDOH_m>z%i-CGoq)^{JvOxosrS}
z_XUJ$voNFJfdr$F)YKl;>V(ZF>f<#)QAz``T6(gtKK`OkgtiHzWGiKH(?3n&(<{9D
z$hU<DVG2#pC*6F2eKAbuU6n_j60MvY0bR&Us}3`M-7rJbYdSdtOV1q4MYsg!=ocFm
zF~gO{zW^bU(cXl04zzM@xcc`9Ry9?lHt~|-tMq5HHOwZWbzM)uqbO4V_@#&o;zT4c
zc;Lj14_C?{!VmA;tvrCb-JDK`;|D!%>&z5@_{=jENjM$O*h$K=!ZCkFUY+37Rnz}P
z#qblKuu3d^z^hUXW8{^B=_UV3y&_M64O<x(CI~ZAHzhM&Gj+;#)agrS0QQY20BcT&
zWixUVEhtM*P&>n_M+cc?<zVOP+lpn_p*lZm4xA44C!Q4=4+uz-gv<@bkD<#MsFFC1
z=g&ho(%~AYWqT^fMT{!SFjG3KvWY=y16(UFMcC@tNmwPL@^8lXaze+t%r$a^LuJPc
zQz|%m7z7U6MFnAOS(tmQKh$twzR%<+fKkTWQH-u~0QiR1GiU1L(q;%_ehOTy0(=UM
z%24T)L2(`Xkn|-gU0G-xSD<GvfS>$4nzem)eZLv_Mx|h;mf7EJ)e1TcWC<O>;Sg^7
zw6;yQEahQ!twW&fTzSvK@f#)%O<0!E9`~N4%H8<0d6&g_sw0JgR5%Joe99k5*{Z3A
zqM$)3>LlBJnHOu=(hE5hg_G=!z#l6)@OBD^sZ$c>Mr-Az2@^?YR!uj1n_G2)0`v>T
zQz}V0+jVk$O(0}bAm8%hIDLHMI?EKQV<Eh-R`>EYYDXqAvN+LZ_Ej&AWuIG=P<Ofx
zUafs*w79PE(xS7wnL;tSu8m%BTX+@S-W=df8(+l#4%kl2{(Y#;11KOSf0|~8oqjs2
z)iWgx6D2Fdc-L92ytFps-Pei}nR}<220GT|Bkal8Ip4i)>n4r9-3T|lhDR#8CwxS2
z@WBIIhP!n|8REY)g})o`|A-0yE+7JH6&~Z0>qvX8Cl~)2;r-VOI+79O0Tb>W$kPA(
z2mf3Ou?26U5cmqgV6nFU<H!HJ(#Q?Gx5a~H%Bp`#BLDpEZ|}r^9ec>Ec)u8*MV&2&
z=tS`e06*z2d6YgSkFDO&ez*Al@aKPBDnbeHL3sF%U#k9-DE-%``S0%vIQd(Ejgiy)
z+6^2D=Bs55Kk;R}0k)zn5A|pfC_?FhvuxZqpREV`f4j&ta9&qDbU(ED-`>!FzVQ2f
zl=8MqD^UMCiirU+lgobt75!f~<2i@|x)y)*Z^)Sazg+)6?%jX8REIAFH?SRPqyty@
zKmW*E`WH|)a^N9&9cRsjw2P!-G_v?h(K(~%m?Y>Yn8>5={>Uo@QvA28{rCI$uS@zo
z0srE5I4S1YfBlP(NJuh705jkM$v<n^5QNBG4-BsOx1P1`pJv{Nwt`eVLaT0oJ`F4P
zI!Al+4tT!1GMi%y??dRiA<+U)aD=ZbS-?;##0mw|l>2prKyYo=L(E09??EOFh#nFG
zul)b~X8z-L^*#l77V`KRs?dM0BB>L~PoVP8f2shMH?A%NHTKbt`tM4km%w{^X`@Hv
z2M&I5ggDC2qHht*a240~Jp3Y!l?FxziXey+s7OPsG5Ww$&j|==k;!`M5Go$qPkBs=
z<)x3`c^cv%11D$f^`^9Nb-n+Q3y8V<OKZOADF3qg$f?SSsOLEJ(+2meNK=~aiKH!m
zY#>CQu)NHyvZak}5~oHccK&SfVn?Kk$2j7R$7I@sMB)GcM??+?13M4(<%Zbrf8P#{
z@D4EY&6W3{1E8GlE|ZM!joKj``@kvOgb|OJYRU;9WFDwwCLoNA2bIT$ff22CC@_#k
zY#sw!@gO3;^U1u~HJ8Rp1X+k2w$Ql|6dZpFsH<t9etmtMc?TnVmu%d#TKt&9RI>Dm
zS!l6(cjSHv1i`Ky0+#i^-0on?uQK2e@TQnV@fM@ai=u;-t?HwCAJ`7omlNHGay7{#
zXcbl40jL5J!QnDZ9X+tK?|sG(iP;B#SC=b1JFH{0D@y~``}8RXp2dGG?EkX%9UX`-
zqTkmhT!DXm5sgZp8~_IAAjo_ySN#FINuL;?O)h$Q+GzRrYSjpyB6>%NlkEs-kp}_q
zkq%)e`2-#bir&lX?dbD>w@?k_+|jR=*wd)B3<5+rHv`)dh7<#jy_ZnH9;uOyxDpcm
z1=69H%TvHDN8!JeXbjZk4nVw@qW-lhL?<J$>O;`$whm;>e!3*RCnh(aZDpd*7Jb_B
zYIpjt{#x(`V=PyX$+-!M&HtAL|A<&{);_VHe}Dgf`QW_*5t&&qrAtfzHeY`!%NpfY
z^Q=|L0QKFuG=z2-_?`Xqk-%PHzDG|LO@?ym*D{8~8IgijL7S~bbMvE5h&WLA3qJlW
z7}ldjzEtsXH<ZRbLZntx5cQS+KkBXj<)(vo!H762@(y~2{(eVU?RZ*%D6s}Q0kMFX
zstE3X<Ht&hy7K_)YY~L$4ME~HowI4MX&JgY1YL`Mu(?=0gICavZu8S!beh+_y79_9
zg9EHzfyXcxRK9)V!G2AyL<Rw>_d?#y1%G*eceAy?t$xD$)Pn}V7fcc8!Jz61pEnwf
zk{{Z=ZBA&_$Mo51Ic^cbzB_jMOaZ<4RXt@-Cv-G1L|}~X+DzJn6-Bq?rj>zOF0dmB
zrO``TJUlsyP}n-D4ah6#Za>}UXBt`MSS+yrFw+4~IOc+qdUl`PPGdv!toXKLveo%?
zThBZdt{~JpYdJ&TChgBE1q?f3oIjtHI?XSD4({?PpskV(4kipm_$UYa486R`xDv3Q
zAIZGCHoxd$5TZp4t~JYv<nHo_cN9paTAzPapBuFJ%5*3`+LcPUy)(_@_Cx>Bi>^by
zcyHSb&by{BV{7aGX1?Tj9aC`nNyo|uuX)dYVGl=Jyn4|^N6}ZUfEqo0xT>qi$KDU=
zYtp|Em{+REKwa*^%j^razK&g#j5?!|GSCTYJKmVt51p&}#YbV2KZHO1_qrKzK}0f#
zGS5zPc*|nz|Nbcr?xr1z9$IIMPt1aatn{Qni2XZtlobTQhZF^I=%P0ga{fcWdVnqf
zOxD#s`10seSBL^git$IX;f>5&pd<wZ)zzBJn^HYMF??WncXkMr@pc!0&tVom4OC*u
zl+?{Y#We`FEE{+d=db{XDsW8$2uGg11asfReb6vgJQP7j#A)OsfH)<10nUxZ&u7$0
zhTe_yAQALt2Cn>i@X+Nu0>2&r`ZqS=ivi_fxh3#qvn76nxa&!5M?A>ezk&&F5m1nB
zd^lwvt++)B;8>!_hk_zSJI#GeJW1>n*<%1lw&fH)9)7_IX5-9M4N+X6i#h^IN}h3U
zw{6ma>o7-OKj?S9&-4a!reiAr!b404Wa4tnBoul4paM3;AbcS^W81ukB8>ameXV0}
zHC~yUGp<k66FBkp<_^WC8bNa}(BTh2<%njuZ!=wVpb62yz9PoVbMBPZ%-T-^S4fbw
z%G~M2MkdMgAQt}zH483timz^o1b~LxKsUyJvlcwV9<DWE4HcUqKllpdWu|7^pi$Pq
zDVvNysE8;@q7GUxFJqb=g8z~|4@4kcDy85vMc3gVAS{zZ|01v~7H^4ks;YnF<2(nB
zrn;sS?C?JyT?&RLwdn_!EjK8&gaA3q0hp!dMt>vx3~PWe5kqGFX>|56OS@Qy$G*oR
z|I^QWvo<&@z~q$n2@t(LQl4_|Fq*`C8`?%&0~Q_+QT}1OOODG-bxnPBP<XhawV}8F
z2?|NX3*Zr8o$FHBVPAXD(00mN@`7SoAwBzRP>eMBZN^zk-}OcNGp(_5u~LQz@Ntz4
zJP##j024-VGoKkMW~jdkx-GaP-dA&rI^QW26?J{}1jMC|{OXb4cOaAq@>kSR(F`qy
z*UKbbua*H?BD1uz>tQwMUZC5s-jI87{{~L|7-K$gsLy!m-*p^=0!W=3Bf+)GokciV
z^z*4-x1cFidlXAz#{;ND*qm={DvIk<ETR8pLw}z7e}T<}Q{x|7?%*T!sz;1`z<oZr
zPNFY`m-zs>?n#FyN_ryydbU}U#M$)7v~{Z9lt61hd-!`FNQ|wnzUm2v*`?$ePih4K
zW~Q9-ZUT5`cG~)ge)FDeJ8-X3|9&6wf?rx-?q`GUI$v~AgJDFX(+P1Ny4M6vR(#%n
zx-2`;H%Hq-p9#j_#}RlwieUto%UgBKywU?QjGCvw)5NMc1A>J!30b1hX9@&#51u0K
zQ9P(i4YXzc<qk1;iYOAo<$P(E82wnajr<<Ofn6b_4O9-nEsMbQu`B59^b6(*Zavj<
zFd7=g2a)c~XGC!zxMqj$)lDm4|K0$nfCR%#DE=q>-wFLO<Vp#_a=%}Y0jklt#6Wx~
zK3Uu&p84l0)p7`<;8VV;s`G?5YNTK9u1D^kx|ik*6y`qcb^zoQMQ*S@5(6$D4*{bs
z`TDfNhZPHQC5E9xpA`4Jy5jNK>Z5ZoE@w85-D^ue6knG`XqARiE=(3S0b{)Qp25uY
z6bh0>a^r~`@lg6uwjW$a32sdH&9?dn55SlO9+e;qxM`e3C}$}CI~E(*{iA?`iDI)q
zH9EgfNp~eMYC_}m@#(1O<g#jCg(?rw4*-^s?V6vvD~RYoi+nh;!77DDZ!jO9lGP48
zji4nYJ<615<hK_03(ufY&%vI#l_CY$IBWQ7fhj0>$-{EZ)<iK>2LK?+GFb><fyDib
zAe8E6q)*-?;lO(48IvZ+W(l|DnKd<2EQpl(_Y*d4F3Kh9C1in1s|tViG2I)6+PU1&
zKS{SVJKGiw2~Vv^zX1cE#=A|14$doSN_Rl#sd>WnJ9ZyDPwRyOy(L1;LUnQ!VO&WF
zrjy-7V=$F_;o&1;*yndQ$Ts(GDE5$UejwbaUfFR*WjEQkW4Cz3Hlg@YX<bX8{jaHK
zF9|nwl|MuDiNhh`ev1h5ocOYmaN*dyv9obye>Ag{KmNjl4{nO^zdp(VtSE>|oF3n7
z{$d1L|NY+kau)Ts)__GShcE$N84N6QQ_)aea?vf)A+*A&U4`!A9+K3Jz<H|m2}V1}
z;>7_i%AE>y7*WSSY8-XeEf=Wtws@hR`1zl+#smrJ<WU4Wfb$hr#mv)MnhK4aL-d4@
z5kMV<#ccA|_zT}tvueop*amxkZB*K}C^6O;*bc252gK<(@ojnw>VsZO^`z{!#ph0m
zUpt0(?hxVD(rx;gP6dTX)f`>viYa!bN6)l`py>#Q@GO4W;@D{7_q&~<Z{8nDqhIiM
zF5DC&MM*>&YBiedo&ql1WeHxhiyHkU2PFj6NJ3y6&$o!`KLI8bNW>@Fx$VH(83(e>
zuQE6Y6qmf(gHY)*5i)xTR@z6jzC~YXAu%FWsyp&Qauzc5(7<-5Rb0+6bRSC}YQO=D
z3|S4oSv>FMiSq%=xGu?hOFC0TGKj`-9Xo|^<e~>#HqyK~#9r0|TH1S|*g^L=A@65J
z%lSd8G~x#`v`oKRpC4RP0KrePu0r9Zv3<AZE0ax(&&nFp^6|ZK_P7%bkLBy|z(y3J
z5N<{`p!0E5=nVV9&@W)6K1@AF<bV}#5F(*1hED}5?pBRo`wL?1=oM2<5A*LhmR!gX
zfRh5lm)>g8wD(0}pMY~~g;zM4#b<+mv5NcNGO6WP{M&6AX?3#H!14>JGM9SE^J9w=
z7zR#GREdtZJR`Msa5bd^3$%!sv!;Ul%(-^X3hCQ2==aJ$UUE5FSx>X$R^`oQL$|Y(
zo|B-?PZS`iGvdu<RV&FP=DEq3R&4V_m!CBUAW4Zbdz~1VW0w|SRgI}|d24=&^PSkX
zFg}kA;X5C-Oq127SvXNMJntgGuTz>PB=3=%eKTm6aUX*2jLWVq@fAO0>;3>{7Y)^H
zkijeBE}HNfBvIEKoKWO3LCdU#JpExu^+BQ~b(@N9ix<aZL?i!1$e_HgqtVdj#e#oJ
z+Rn*tSEu4wyY9umC1&>pH3{$|f2?C$TR;8^zLlH(`c5ycdnd<VTHp4dc8R)L|9PuM
z<{T@`-N(tZYihEeW%7J>bz05r@r7&#+}y@oeAl9zHjRp#4E}zbq@aK*S;bG;=W<+Q
zquPD$3GWpvrKH2V%;^Q3s5Sd{#B(nDqi1hBkmH#4u0dKi_PapvaM~tiv%%>E<r6v=
z?ztfd-$*2B|4!RUAD|oMp)*^t#ECok`t})p9yYe`AMljC)jFBZx%YJ}h3ZlAP8s%c
z2pZN>bdcxRJyGM&p9pR%*EJ=2yNlIhC9^?D(MvC)HS^3-7!WpCy#NdEY_R3!PiMKy
z4_(w6DZo#LBEu2${8a1v6T3R0eJ)BIfdk*AldpE(j_8Ly54yJr;ztblLg*%9!`Ye`
zc~SuqQnzg*_Vo@U0b_90p!%CL+=mWNCFo0exA9689mY^a(0h<|sd$G4GXdcLBTxK!
z>2e_=M(<qP{ZdEXW!PW_CpjXZC%vLRb7v_}kQc;GpVQNnZq^Svkmkqyqt!8lA3u(4
zLWUHlooKxmD9mD=pH5X^YY95tmT6lMUEaP}=5k7w!)bQ#JGV9f+X><7G43SUse?w{
zCbpdcV%t{_n~{^kMJfJK%r5F=2qYZK%v62Zz-=igk~Ep9-l|dTW326q8q2W%f?n1_
zI@TX?o1b*W`Lvt#N!?#JTCmO9Iy;$UrPj!o=3inTlvv1nXWD&*2#%dSxo~f*Wn79k
zhP<iIZh}P$E~{?kMFsKg-BHruS6|Z+nzszo(|SpVKkxa%R!=va&jL)KpSF@+JWfxm
zV}Uz|l>p#|ZI5Z&Fi~bjLAJJ+Md<y{*A+>`H<TLO3J5a<9jZ#>C5rvwZGiBI0fP30
znFP@|)}!pf!y)OSeu%1Ii&JpyC!FXPMv-{+ik@TS5IgQkAhSzW(v9f1Wzv2Zy@H2a
z=sq!YgXmm6m>1#IV}lP(`8n%X3dWs`@*+%LbrO;px^K||4HP&=xUFg7%fUIB9$Pfm
z#$KiUDy~0HQ_Ned!BJ`x=mkKriHQt@j{_x$!@ddMUDK;jd+c&}VB1#@`!P%r0#UH~
zxJodVS+%Cubdgw%TuR}y_m>_r$`feYorthpgC1b#CEv`_9v#8dlw^IzPMV`!njbFh
zgVf+9na^?BBG*8IYd2my#ZoGL%D+h?mw`D-G9G_6Pr7l0TVf70iR4i>!6%<m4yUk@
z9c6Ql!sx?iOuQYW*rPqqKZXzcFzj1w(83||%zb#PP-!krR!!$xP4&594a?|%P+Wlh
z<BnU%nL#DUSeqGFC2}6lIqD51Macx^sX)DlWR!kNrzW@hdckOZN*F%f5cQiX31Kba
z{DqI3J7y5Y^Aa8UOFGJhb#YG;^xA-q=7U_htN#r19^$^n^nptVjjo)@spUMS9bLIF
zZNK?LK!*goTwC+UWMgz6E(eiaJL$4!G!7vhMn59t0v74l_^A|Nayq;R;sf(g-w}r=
z&Pk9QIWC@UXQ%ch0dAgw6EehVfq^F#6c%JA#4=kzHAMVpKqNlI4AA$<j@Yh51|s05
zgrg?xroi_!)Mnc>y+R}r{gAue3ZQfJe@3RBHS}3gGVME$zo26ei%XIy^L;6}|3$H2
zO9XWRiPdUPHp6WhgS#+9r-^_rHw1q?(8+s^*)Uf7U5ezytPZ<G{qs;0AmD)aMH2U_
zys0lq3<Za_DkA&S2CRIRg&^cj%<Y<IacjQ!v2g{9vzF9ZB}1qn5HQ6cz7z!c2$lze
z_n}hq7|t*G=>3`6OG@YdU)3ke#ZOHTeP_us+j=6;1kO<n6dN~H|9k`r5iX(A2!tjQ
zvfgkMgDcf&(>gwQUvoJAt?7v1+3=&PD3`RRqUMYsfuF!B@36w;UMMXaPGv~%&gC-X
z^4fJJOoz3OTQu6fj>eLjSlbj&r7N@0Y)Xnt&bzMOk6WY6fIDOS?mIS9QR!g>wX>1z
zD@1E~vq}-JA>3HD%V+Z<;U1GaVL79p>J*r@Hh+!~?Ck_^Rn#a&r;^mwGrC9E{!t&l
z0pgQqK*f{FxK#0{K8dq=IgI%sRHk_{vP7qR+X9iKEadXHVf?EUP0Ek{oPA7DI#O7f
z7j&S)tbJ}-yw><H?aGHU4~ANUgABXAe?_W0JGA7zZs^-P(~B*doG-WHBfY{y{bv6G
z>+=yd1;vN}bc_=hJ9%kQ|4-}6_I{)L<>cGY=WUaU%+;e(<aw<>_FL?N-HMlxI1B6#
zXeiN+=tuI#x82-l(Q!5``@1>G=@(6pL-x-h6ul))_OecDMt{b2evjKX&r?0$ueR(`
z&gjlu`L%zjt0BGMli=dL+VVFGfOE+hw_qc<Zv;_Hf%9z~t)M0S+1UTQo>%mn^Ux5Z
z>247W)DY@jXGf4Whfv9_zb6m5Q~6SV8*){lln-A2am3ore?xWWnZ~+dFcm&nBMCdM
z80RfP^TUb?5FTQdzh(1S^SE=QIR821slqpAKW4jfO}11^Af=J0^jq*w@7*1@e!JWM
z$Jkp(Rkf|}<I)?c4JaYq9nvM;k|GTv-AH#xcb9+&Qqm<LB}i{Vy7Q2dlG5OB?sLz%
z_uS7I_d9-nIZ)Xw)|zv!_kG?c(uOH6)~!!{Dnr%<9NlKn)9z!97N+6_9emRn^sk@!
z*IIAr&Qk@PXEE={hc68uwzg9Fb1;}27i_x!zK?MfX3}Oy^DJop<u0jeSk0SUzC68o
zdf3!#%TM-m1;8~N`XTGaxS|0v2LvmZIfLWa*yNU}{+ulBz@AqV`#882)-g5MFqqmK
z)Vk%zA}26+92|~8zjM7D=UcJ1?k9w`h>v6J@?A(ZOUz^LR`lzuB8IoTyG%}=AH16`
z-b7XYksdo@&cgisXS+=X(T=MC{mw1@f%SgvW(ctB*)s0og~H%4Dl7GWgdmulhzT5&
zV%3yJTa-^fvW9+Z07R`6uX>LJ2vzY@@!@8n2paOOT)g`;PYJJ+eg+w~eXkdaRhY9>
zrLDlqk>%G+2_hCXY4T^TpeMLoPlLtLR^tV{4}_G%2ZP@?V}MKf?Vlm!R$Bavj%Ko0
zhFkr4nlqq*Jn5*{#?NKms4O*K1BX}wTf?j*1kKq(ScUB)FK+9w&>ky%jw0jHNo@Uv
z6;r+eiuyS!zjUWqDvYbCtj$*Y<IeF*CVX<)RF;Q$C!qXQsdYwRBEVza;H$r67!mv3
zI)%G*wJ;MR7O|Aj{Y&6F&9QijPZwCr#%{<51eXOL0c%Pa_jZWbv&vVOEr70|pz<;J
zt%^_Fcm>&}g|WCG2PNAZxHe;0$DwJ{*|*OBWxc`F*egwYZv9VOzJE=CVl<K8k-)=$
z`RIcftF#v2&<V6Ygw;l3dFxj~?`%SfTMX_Tku*iz@M}0ykB(0d$%^?2@SQ9QNs-{i
zk)r%F#Fu>KLO)fX{d!3<QrKm3x(Gyl+5S8m9#gJw5nL1&Ek#d>uXeTDZL;#et-@_?
zpD{^HM!k=%`m*34e45-7B*?T*WpsH58lz>NnHRP<4JApT!lb4}z!?JELovQBXu94O
z0pFJd2R>a3G7_N+;5_kRORhbIg|D!>1fA@$=^g*FXV;OX6uoye-2FF99Z1RTq)oV4
zU6@IuNBSrNS~pj~$!y{b+!w0a*}Y$c6{9#MGi6y81YZrpp|l-OFlsXcro*3w26WuO
zfIHo3y%$TjyCPr7lL#iA^{uF_E?6hZ?q~7d9~o}&CVC?5)d0h<<hq)`ROS4QL_MU>
z?%}t`J=djfJS@*sZ<x(n6kB)RWFPgRR)Eub!BuCY1I+F$q~ih7KF&urabD|HCI3an
z=M+F%y81z{(pKyq0_>>A9K_f1epP+dq&O%!XEBUo9pR2i#&Ljcybmmx8a}cwEEl$q
z-2apvnX$h+YMk2_upXyn$+2^lRW1HRPSlW~K07H2#*$lW^38_Y1FdOXgGJgO#>gj@
z@ldnAj#WL{3>_TbN&Oyfr*_)b(QX8W3gw1JK76}1flm*%t448}_cZ4mw8gZv^`*Q^
z)6n}01|D<a-ql^~g~%9+tIn6IOlnrbG_EoD&H?`>w&ChMmjXp2pnK`*$9{O{Ph0lo
zSDm&(cz7UbjOjL-mF;D|q2)fTNdNlrPhBL5mG^F%yT$2cLcOvyyu(WC5?<aQYy{5k
zZi}rinW_6?=L&!nWo5K*|Cv5Ak#$4ObbDK(aIoH%8?fJDW!>MxiC`kb4ZIs-s9c=1
zoPcaLf0^#~WMCdZZVd1+k$r$>@R2jxa<^3ca7&F$R6%<%!#iwergaNJu7y}3{j=qj
zMMo?xS8xZ;NoghOD8XcRWk~mF{%-1&S0L9n3Csc60YhEj$vBeMrevp+XV<~}(*wUh
z0?`P_cgiqOBH1w~{fCS{$*npUR54%ak-L=*=z?S!(f~oy$_ZCB(W^n1l)kdcW@dF1
zgFNIhX>=*CWgD-es68`aD9w%)O~ZH^cUGZCe+zq2Kdt*DVD7mgq|TRJcx$3vo5f>G
z`?1_Ttn$w~QW<f|cVZ`zPIEw-EKlRSq(w1C(Y1z&Uq0@tC!Sf>dF8Ykl#c_64@q&W
z26Cm@iU<a_IRvIj+Dl1K<sxU|ndPphL+K=c4JNjk%hB$wNv<Zgk!F=+s1?dxkJFd^
z5@OM>>v9v0%lx#so98hj@$X8p5}$(NZyuyK^1F-TP|vKxole`l=->TqH$QFHn*l~7
zsD2XkeeU$ydMf~UQ@zY18OAfTQ`)Y2*T>i|I@Ysb24-)nDYS{(d9UF^M%rfiISr9i
zgQRWL?p})8y&&U3fwBTALK6IXMSCY^CNA~pPJ&PDTMLUx$Gpvt==kUqM{AvulO&V)
zSA0J!u=ip#Z)Y<ieD7tSGaZH-+e`Dm&=F~NV)3{wjJ`Pka7?&PLHCoVT7F&fLp}E^
z2fXU8EPqDrjaUcidajtC)?i-P;6ulyn^D-xLd7uFvnl!YWq3dNT6R;c-cW-t<?nt%
zIx^ZJ{$i38;te1a(()Mn`V~}Zz{z#<PMRN4S7=mwdyBO!zd7Q~HT6Iy`xgIqTamXy
z10i@4O8#_lbnrwuw57(piPICA>ytAgPc1~rE($jdg_mj5aQt$-h=D`L5Rq`1#kv$%
zE>&nAQcy&pIpD5LP3vE9*d2+Fzm#ihE~ntHrSeXSd2D_j;_9V{*&mz6cL(4y6`1z@
z*`o;LJ1mz*;B*Z3VC%LIqGn!Ts?k7hQbf63QciP|@W<b|STFK)cYHZn3_f|cG5_jX
zFRIF8XC7NJ`|+UxiPTdiYp|7T<l{u){)_XSH%EgMeoW>D{f%LT4sj|eLn(yPg_d<s
zw+Y95Xpoxqck}VaNLGUq@h%PocO-BmTfpJ#{*`&*V7G7A#ugwg)&|XY6$K-*H>w{z
zDRK*D(<G+VD~VioWIb5oH(HVglI<Jj=;weu5byfQZpDX$MED?6%e~fHpRiU@2!bRy
z+Vb(jI0i-((mEZBRuo*YtnSdnHQeu_xobssmpjOVx2_bQ#d6T_;gzvKP&%Fu{=5O*
zy#7@l*=Q_%1E1Xei=SsN%Uz*{zl$n9s(!r7ji<L;+DIn(;P9aKqE2B5x7I$kDt!kv
z0+_xvV0U?DaILOW^&&PSF!B46YnA+UQV2{w9j1K`x))r;{XYQL;NLWfeB?lP^Z;7z
z_)DHb8pR>%KC?sk977?9{72TFG8=ow8sj0|3b%yC$%_Idr2zio5V86ObE1!!^xw_n
zF1(^Sy;fXH#Z4vqXk#PU6i>@(9F_zvq9r#Xe+~?K@AUh9+8JGX(+?ctKflm@Cy=bQ
z6sxscH8A*Rc@e}Fbdo58Lf=<m@G4_GJGIC_JTT*8x$9*#3ZE0&$Uh+k!_cBW!$R=J
z(yV0ERI7fRcn8S}WjLvgIRc<UY;BOIj_>wliJO?dnuSww`S${{d~CDSNH^&HZf8>6
zXPx{s0^!h)qXD1ni;_2+>24)hEdqas;H$vs^rg;kKwgT&mGX!5dWO(Qo8zEX*$&v6
zPuY>z$eK?Y(W`&g7T+FK{S^<;D}ZR@M3+AFEeuB<=WN2}S22{XxAs!qc1KG794a}P
z6!YD7FoGsDP}_6t`@2>}A>ZX8Nk)uno9{&m-UO{3QOCH^`(EpsMj}Ia_u~ku_Kn|i
z;y-Z!u<gvU>)wZU@bP*x+!>pokf}@eZgQhPIo9`E5EpVtUTm>jP*K`o`3GMx`r*b)
z)N@pUz41tB)On-|sHJGg>I09mbUvF$)v&BfL-e1s)^tOuOTrE1GKx)I(g%a%u*W0u
z#pOv%El8~k?u+T@E|we5pX+PX+wtQ~ZsIW9N?bm8zV#A7e4VY?7b%}xz3$<*J0T)4
zeyI0&+~FK0KbVj`dCIE)^JtHj3$hNT-?Is|pVBgw<m7!mX$?Csw79;%Z%G!kHT<;>
zyba_hZA{oz0L~u2GfMgMv0x5=vX(2gje12tOEcCk=jIn}l?rR!l_1ePSRi4y^GPf0
zSDu~6JfSvV>NDfZT)y;LR+yvz05s%65aLvSY-h*$j||4F5_h4u!?O%!UzBfDfB)nm
zgBnt`Nj+=>@3nD>xi|8LlUNaE38|+BQ$mXG+|Gh^=Y-f6fGgR&J%g1P2wzcN6Q4E5
z3>?n6Ein8UYPyb)<vor~_QX%p+h9-#ck?-$)~B!YZz%T0Eq0I3l-J2c8rf@Lm-2?c
zT5U46(_IC=EyZx(EdWGZ22MwUWO%|uNB(3HU_uZv$M0|eXv5#Giu1?k{*GP3D9n^n
z%%7kAiBs~oEjs(z$$8N~V0HeS(;VpLQC6g8rN;#t_?o3pO0VINI+YVtTu<xK>-};i
zV1?0?ICGS6v{Z_N&-L?joaMf=1y>paPBe|TS=8!Tdi$+81U>NR5<BK>I|liM<cHJh
zyNAT6`)bA4-0fR_Bb_MHqW5aud2i`e9w8g_t41(w2w#n{{NTE-&vLDn;;%BG<~DyG
z4`ZUs(^4j><?gT~<ZC}A9np1#u`IZypwquqDvcLdgABHd2!4}{7gjuD9HxqX2ph^n
z8X!suObM=|5#gL(FJ&bkcw$aVrb@O`%q*q1#oVr-eU4Lz<3@KQ!a;rungK6k_iLs2
z>l(7lx#mE-6qHAfjFtU^T;_gL?j%iAm>LZ_ko<O-(1yftx#pu=b%I~T5PdsY!Kjrw
znrC&o-XV$(KAR7xz?7NqE<=}%?!fz{{)Edj=Rx-|yvmm5YFZKZ9us1Oj0vZsIG^z>
zIkoj?*kU{FD)UNhrekkiKVNqEO&?>9Kl77X_Bp{Zp-Ur!yL%4un{CC5bU!KAZ>IXt
zdOO@Z_x+idWJb`a@}h3|{%a*6UFabDyJO{4XBl73g)*V}A-m@Fe@6nrq~7|E{r0Sq
z6PfqSwUc^es=GyaR_}%F{MgTrJdj}|z;L8lH=-g5o8cC?V`zmcylxHt#j@q&opJ&V
zX!8_!0twjZ_R3g`+H-3%31$MkXjvKTuc*k#w^$6TIA0^lEwhA2h6lVsOx`<Roh4Qm
zZ!;%T?9?AsyLo)ZT2!lM=8|%Km%c?A7Q0sY+F}j7ODPAe<k`!@tMRo3a?}L=!(rO<
zJmsN<WG?daa;!X?iC2pp(|eIL@va?3?47b=Y8&t3_naCU5HM^r4;ocq(r#1n<pv)<
z-5bOc)#KaTiu$-cLnN3S&EFJcb#cy==O8nqI75^1QuKdu2$V(d8@e4yg{;4vBL1x7
zs2ra@D^=PkESs<d(yh+Q7nLw>KK<Nis`r1IWb0+*QtYFw;s`R{=n%0k2z1Uu3^BXE
zxxY_WzE~)YTdRumA6*ARKhRlF$Yv7i+dch^{2ld1q)(QNlUB?}=CVuA>U@&=8fiL|
z=&&W2HI@`G!#|4I82FV(^8!;YcJi9e^reCWgh|a2fK>S_gs-oKw5#-H+}ndN8EOEx
zYlkR_{F_sQ`TzlSK}X2s1&l(j^R?c>q53e_7^5@PkMhB5z4!V2koX?IM-=;ZA+T^0
z9PV`)>_Xfq2;?!y;7J~Vrxe8XI%!<)GY5qvriDjRBMYRb8<vnNuU`1yeg5gbhtlyt
z_Z?5;W*(Ta^BeQ)39<JPnb4-yCosKn_5<vjxu0xt=N&_$X;>XeV)V{9qL;aav7E3^
z<y`?n9FgJKKb4wid-il3**b$mG2$$PXuCR*`!GL8NC=S91dm7kNMQEa+?7Y86&`pb
z?9%alkWMFWKRh9bjxp}W>$r!|R`sD43JfDcBf+Xeqzu#S2k%IUn3p-pN6_*NFc3k}
z22gjSOgmE7v*De*a2v8%0_A7y6Ljm`MdR$Cx<Hy0jW_|72R(REWCQbnMxIon@riFL
z@X!9<uDQ91!EndB`w-xSJ#mixq9b}~#brvaG{D>W5ho+E&URh}HN};w<@KCRVeI;=
z;vXAyeKqQFRPyGAsqVsL1k#r7u_jBC3nCj5C85sq@ka6i>+OlL_n{)+m=W9t2I+-g
zS+~fu*lRZ{hfQu{Fo4t<+2kZrlPA{gU<Q!2%X)|J_SSzm(#*|R!CN3KP395IB!!#v
zsLqiG185a*SdYLD;!EZW^4|B%Z4GuEcJcUW2}m1!BGCiL&MxH?_UMWoU|u`1A$v8%
z&FbU02nB$0*Lumi;xp43AuqR42{c`jA38z!cL<w-&Js5{mVU(xz;JKoPkiA&W6;}x
z^I;@|q&)8%nFK>MwIGvM;!v#w#y=46agkGKt8;^r(kVe)?xTd<=*}c$Oh!mKZGVt|
zr8FIf#9EL%G>udm+7ps{m%Y-#J$>dnq7>fb@pDmQgnMFXYU=T_Q0j`%CZqs27po|~
z<1(p+*PhB=jHKI2&id1WwU>p`n2IU%4t4&{q3-n4y4GgXkfdFzE2o2nx^TRmv9)o`
zx5kt<oB7$bRZ8f6m`3HO?8v*^iLsvd0)f6bbgOTGyUfH%+VD-eC0}A9_5js-mAhEP
zgP8nUr70Qv+j7ZHg+BC7ip^7y=c3qdbF;|drVLu=EExN%d4USDrrFj&$Mm#l5{E{K
z&9-cMkDCw5epU~a;-@Itr5vOCBF>e4WiTH9S^|{I$NE2(38`eJG)3i+*1dWoViqE}
zL+rLtYG_FLY<>{$y6)ogefyEYq)}ug{O+O++h<GlTV~Tj9aZd*4A(N9vR4sdtzjKo
z<%^j~I()Como*&J3n3aZmG6GY&<m6Oc*vBlLFbrNPoJ2&ObtA1koHhW`xpAFw-V`E
ze6{Oi*5t?zDDS(AitV(myWV{+@cryQHmrt|nDdqTRy{rW(@>!?k+3)q{g6%<BS~B&
zhf9#Psh{B^vyBY*Uk-+wt))t5&xJ6hXl@_xSk8a(cSAmbI(8b7b2%JB1Q6ri()<C5
zd4NnL!l|x|Cc4*Wh7;T;(nU)<-!Cv89(m`K_dOmZ077)x$1RMpK#Yo6q#M`IgqrL7
zymdOoG?<D;oUs%8gaV0zJ0W_$XJtGj3#9l=h?X5Zd`d~?FT_fJG2uX|W=<SIrc{9Y
zhTey&QsVt<@lY~aR9LE&sUiu1ChOPi%I+TkcvA8Cc=ccvr|Vl{whJjzppAPSPi=Gx
zdY=lu9~(JgbukUuejb_}?D^}CpAvPdPBEATYWwoq!qhf_j9&!IajCM=0M|-6Ss1mV
zCacaaYn_=Zkb-sl>8d<2oyY(!en*0!mn^|M$a^6-o9Hy2Xrv7$dQ7L0M8p*B0p=0d
z&e_gHYIXPLCIyPeaG1ZlV_akFVMa`((R0&Jd!Xyf3M$c{&lF+56rVHar@zA9iF6kC
zlfH{xvJ|_~H9U)*NauCZDO`@rk4Zz^<?sAtSJ^Oc^MqoTbgc^!S5AAAc=u@;H`VbA
zqfEP#XSCdFPe|lLoUV&4zl*>8;-U2ZUEsdWxfN~qWGJ2g)2oMbA2?y@fFz?eNc;=Y
zee#aAPJCP`i`k7H<2J?mIf=Go%mNXCSouAthc1^y>o&exy4p%oPJ$f&f*m&<c0w<R
zOV(mQn~0X*;nDQ^z-y~j3qp)aZz<*Wy=B?AN1y@wIzxGH5-kT{n)}*hzci<WxZ~ni
zg#Wc31Q017^as|Et}FEJowb+Se^ONyi;-zn2@#v0az_=DLtL#=6e7Yo-s>G@SbG#&
zSJTs#sB%f-&}4ES(PqhjM2v7nuGpDFX1KuQ&?-m1<HAk8cV?Q*m=!v{PI_C?TpUZG
zDKFqlLDK6o<oAd-fvz-VEN|7ksY<Hh^DUc+Wk@?1!nLeh#^_QRia+5sHSJ;5eai1{
zj>K&qqU>PtPSr%TL0=L*6WLy6>=xESK!rc0E=tt?xg@)W_1LWK+q`(?5RLfmPT0{*
zocP=h<GF}P)I8iYy|gJsBq|p7`$G1>ILjQnVpJ*ElT`Rm9o&`)6YV>BJB^5&Ot7P!
z7gUgn3uBct8V|q={{_GQ59_xL0(C9Z>axe}aqgUCBYCxkg;?vz<kdrZUn19KMAs@8
zt_ANKvV3S_NOa>(GtuL(ll&nrzC>H^bh!t(md@P9l(MuBr!|MmfL~BS7bjn*QKEm5
z-=4M{B~IX+CspWTEZB&D??-WDv$1Cp-usl<XG60X{gjX~hEqd3CSmGe(pVxhPCRLi
zmkIsJl=y#0xa*@);diTWADW*q!D_@LAUYBdBnhE233*Zp2(|<aM*<@AL{2vX0+2+l
zE}V>6h;yrbLVr}4;dHFh26Bcxgo`QjS92X%BW~M%rro84j--ICh@~;BbY1nL^C0hH
zH~%FQZ9gS)4<++B@37<mI#*)FyQuTLS+8|+*2>1UhkT|y`zH?LPCOoxzUW;2kIgPx
zbazrkR<k~Gx)mT1rzrGO29HwW=S$GD21Ct*pwEJ!OF_`zK~@DCbGtH`<N>!%Y#22w
z8#&OjCt9^qD&~DBENlx{;m#Ub^d@2(I*+kbc??jfx{=M)A~>0~L5!oCoyo<H&l*#*
z>MEcIqhC_Z_Xe?+Z@&S!Kel8y7T1I65ysDw*!E0s06Hf}I=;hv!umulC?TZz8uXLQ
zwp)!38Jq!wuSzaTv=!o6B~b_366%9l&uqx3<v#H3;_aGHg*G5-HY}_7(9W9&S<O#4
zGwZ#0#80&oR+y%zxLKw_5@gUXDW6~}h3F+Mm`?LGIDtLb%0U5*5TXhx!SXT7IDd%W
zj{lu_KFt_M<uVq|c?(uL-+2bo0wcv>U!WEBY^m8b5(Kx10!K7&<TcVtURdH4kp>e;
zBUyLZ?kdl7>ayC*qF=E>1gnqVGvH&PIZG{sQ#iG-n0=$nq^VblK3G58ZDugqu)_V$
zPqmM#Ylk`MCV4sOUa27f;tvmu{<Wpa7y{gpNfG3};Szm?z^H952ejMp%xGj1WO&Ii
zo{His4+ge%gKDtKRCTPHG~|fTOkDt<m5O>thx+I9LNYD~k^SnfpakaeD(`b8VUEoE
zV7sAh=9to*+(RX!A~J00koyOwbh?*VwE{Db_<J?al)-06(>PBE`v|N>&^#J!y}}08
z#5e{3IiJ6%KMdQQ$8n~U`-M8k_(a;&eZef5uoi(9CfiDx#kJ5}ah=q}HU=WvsrX<P
znU$^iBO~v|dg-!7^K#hq+VpO#r0CxvNpHQ|8{l3_bD4Ya$c#o4YOmt;%6wj%EkEM}
zL3R1_1`OnkofM}l5KOn@2v~%r?=6L{s|rc?%6?Iveum+r?)S^hC7c;)UlCGyg#5l~
z|BV5tlDekk@0w_~JY*((6EvQmh}|EOY&mA^<@xWK730fveQmb9ugosV_l@9duLe*f
z+8Ao7%W$m%E3AY^$b*p~5Sc1uyA|<(%Aq#K?t0pbXRG&8bkMc*%jY&GQwHSklyVt+
z??2gDX>b((2$?K*A7UEiD^lh#Z?V!hD6A}zq931~{AD2Bfc`nkpudDof^V<vw%27?
z*ZL!Q5eK96(}OItrpSY+f?z?xx$aW)oU1afqBt(aMz^})olC;h17`bo@9FV2%pPX@
zSUL!fsHr~XX9{VjB)L0%pW0ygL5&Ymvf>;RR(Wjby;HafAR|a^A|0%sQOI)8fbR6u
znao5{RvWwjM6cOYi(BexDc2as4yW(n8TWyHyAWa%M4qAZAEy&Iq`p0%e47^J;IWP^
z?)(h>&l*AiQ7YsuE9>^KQ5`|E)4vv-BQj~=N0|o|H(X6sPeR8Gvdu>ygp$l0U7!kC
zarRHVgf$Xl4N2nl2h|#f7qN1F1q|*d?(P$@4GAild8Ib2HH|4>&w~v<Qay}aI^$g+
zj?1LsxYq2VIOVijttv%t!Okeob7RRylT0E?d~&~=bihL{lcU^~1AQap&n!(obwoIG
zOxhnYEk^ai_4dPrf+Ns6jPNZEpwdo>G+iwF@PmUZ!B8cByCLTBSddn|ubP-6*8{y3
z!H?7<brU>2{F*#9o1wbXAJ)CcN<&du%lJo!<7In+{7G<T;nje<I-8Z|Ze}T0B8D0k
zM-+^&Lo>0AnA^f?VfyDg>*q;!F=}7FjviTBLr=b<Xm9Ij6Vr_f&3CwNL=U(>#tI<i
zRxw=MkvF0WvGqS!?4lrYe36te{7c7ZSCxv3zV4|NZP5n_A1<fw!g)<uZ4NEoUn-qn
zX~O9o^yz2l+Y`-UVp82zZj|3tc=`W&1b_X3^cHBeQ7Q5NdNgOmh^*o;2m5F2o<KIR
z`+3&O_0KZ_{&+g%UY~6AE2>3DQtrHzdgao&@>^|3`VxtSKv?j}Q+u68g71;uJj`+o
z(Y?iW$<lS9jLUgVWt(IKP<sLIq{Gt~w%5)DD~|8{6IjN`EBXdOrlVgIG=U=TPW||2
zT1?8GE+NMNYj@PSvLO#)g&$debU`0PbY~m9b-|;%cxTp0faSoP?iDgCphA;+q@8Zp
zr<2me-?`GaIcu{UA|mx6BJ4KA^pz>i{N35*51{gCvEPjHj^PCP&lCZb7kLsI{l&-X
z9oBNNP^&m?;xv3I;5C{Posd_W7N7pgTfr$;<vVvu2~aEXUsx%$aRV35eC^ZD3KNcb
z<g9sI`L#ShzQhT1=dTBac7${?nkBZS1Jdl`yseDX3&`3sm6W%mMv3@o&I>4=xtQY6
z!ozXE)N+fks#9*5HHmIGresgyg_JpEYiSer{1%#*xQ$d+pR1I-%olxZ9~=2zHZMfz
zgQ@8O=dWj1bI&_hfchtOG|uHEw{awaSnR`--lwhZLmYj-Eiy`_o0<o_G~w7BQLp)R
z#(9MiS)d5=`=j*mn1z2AX5a~PN<v@f#>%3hMq>^){Rsa`0*-g^C=uj9?&SvLQkFzg
zn7YreOe+h_NK0f<=sz$0RaU?^?kIINkk!z^#ayW&1RSK!BS9<&ycmb9TiJjFm(dMf
ztbIGDlJ&$v$5RxfeR1Cv(@cdV-q%2HHaP;CdPG@Tg~mH%1tUCfA%wNGev|~QO?gzd
zK!~LRl1Z_9s9DHsPzl$}!hCLDKgD4S3c+pX_b{Ns%hN^S(kDQOX$|$9Epze{gH7ef
zi)IdiiBTkJe7STTkJjr4#87*l*Y$v;zqLkes=xK2`igiCDX3y@j(21QuD<BUd(QeX
zB!=OGXI>=4c;yeY+G7wINe6AJRU+e}>I<fKo-7G7H-_@*3}^DP=mNm1t`f%bBTcIk
ziUOD55s-SOTYB;@x+#;@W~a9~9CRR57)}l*G$xesbP(~wkfnKd9|}%?>GxR-ILoKd
zys{*fCGYKRlM~i><tIbx$sl*#P)Btsvm}+>=ZPbU`y{6p)3`@WKB54t^M&1r-~3B9
z3h4WjUByl$DG1vneu56GsFO%_K0DR0)$~9VgFDoCVQQf!NAs{FQ(#cNR0#Fc;&|E@
zZnyph<3~o(bv@lRiiR4R^z+6q&kSP_E@aF|E;Kh4<9fBO($=-O_aN5@aB0gl4UA(m
z)p4RoQRoO`F7Lg1+PK@}!{eM<*{j5Sy<`B`5Zdy(rKq2l*$BVu4g5h}8+%h~Kbxxk
zs^k|$+D<S|%jvV|76_<J2150xiR<q5qPujf@gkBI+cy(4Nv8gfE#yZwc%R7bTu2G2
zk~nf=rGQ9Z!^)wL*?d*Y6=m<W6C?c<(VSMbi*hoCDL{cS4gn&!*a)%zTBJeH*WcjD
z6Aa*jzF^Z67dwp{c2V3rjyn(@LKKVSrv9~NUqhg;ofhVym32%YggCu{Tj{*%pKlMj
zKS1WLb|>omcO8L{2fc@Y4WLq_oRtO_`1A`eA7dL%N}0~zU940^M-yoDhRjOPitj<X
zarMu*`4h6L!?)kcLJ4SfQ%V;cd01Av#R#5_@!>5v<&Y0fr(#dU;3$N(VawG3p-n?S
zupfI559yAgJBMbUV0M$h)L!FU#iMxfJzoS|k$%1fyL^Elx^`FpZhuKUAHYe@$R&<L
z(Wbfl$o2#Hr^v(kR%SpN87)?F&}Os|Wm??f15!NY!Kd@3+B-hL+3@K=Ayk6%&%URF
zxQUmW1wTzkDVN{fQlYUXB&S<gY7<nQ4gr>L@!d9D`q~e?T&H2G$+5uhg<(}*HX_&R
zzMosu9C4fj;|fslt@111#y9~Sv$#BvVP}x}M?yO#t)4f?DDsc}{A}TIo&|wIeI-|B
zxbl~CP4kuo$CE{MLbadD(KsY7iN|oUgJSV#p<AF(d1-<k(hBaP*o5$HV6nur{+h3o
zQ(CGbAlX>Y<M^4Iw$_I~n9QhlsRzJB6kpVT6g$`WQ691JBGANUIOVe9;S2^;YN`0c
znUQlp`emn1Y5?bj;m;sPxMv4vE;*&|u0d~k3LEUWo5P{YYA_<*+CT)DcH>1fIgwa8
z2;38ifbEe%nOe69n%5`u7))OZs<KO-H<Z=&5VSM9VfnvIMUR$CwS7Ftmr=s3UB`0v
zJz?dI?>fo+MGaTnqVoWv+%q7>+S2nd0!s<~iZK0xd5811xLwzXk{rlO-jdTfxS`xn
zgsG3Hi3;2cS47q-64wjN)+-9@$TaNPb&WcNPfO3MLqe<=LusGq-~^HC{i3`*PdY*)
zm}IHDI&=9hzF2@0{j8w9PNHF~)H0+4>>`amikcxRlF8(R{zT&4CDwyvP(g8Vjj8Sw
z=Y#zLFThQ}Ax@aTdhySfbc{H%Kaeia10{|9RkrtT2@NMNx%t0N^1xNew@W)tKF*!!
z4BN*l2XRo+%0}<TzrX#PmBd>V7Wpga6CJC&(Qwg%=q{*0_07JeSlI>&gzh{#UNnT!
z>fAk)5|@N$X->E@A!w9T=hH{A4r)tMl@=eK-0q@ESE5y%_A!56Ku_s$-Jh)~U9}0u
z;!`t$G=AZRJM*wR17THbKXX}m-V-x<oNi)yJ9um$O@$L!{m!(U$d6BwWQ^;oYK~M5
zVEP1Rx;#HacNm(2C@pIdj2!?;Ml%SXD1ic+Jj!Txl9U!(dRY#J^jPg|G39#}S@zR>
z#UKgmxp~LNg9;(mqU65W8b<7$C{d3OCCvB<-<6{2FgAeDC?*Um&353p5-Lv_56y8E
z43`gTDX;5uZ)^erHT@(tZDRZ5##|ZAId&q0XqSP2n0265N>Bo6Ri_lhI_`7kzaU0s
zUBcFttw2*u;*1iAoI(AjAHQ=B_(!BTF5)Hr))W6Jo=>xZ*GbASe_I~2iZ(LDvPE3C
z+;g?^^TQ={g;k@_R~%LLIpOgq5%q>*pGb@in!2Pp+|}8hTCAoKzR$Xg9Z%WAIa-f`
z)=UqZwx$+ztyyf*j7c*nMfmJJY&>=o4odB73hNaOP5oacqMzmu0Q%$%lXil~TuLA*
zW!dDaR_~66QjX{|LHJa16Uv`p`2D3q_*IA7bh&PS7>PsiuZMR~dEZHXt?BCEI$7Z)
zKFtBE`KsE3*P3&cEm%@*9ntp~Ub0{7L5w-iF6<Tl{kK#sRM=y#M_KTM?E@;oPgv4z
zjachGAJ*L|5WVNI?q*mu<ODfr@b6OMny)i-FPpPz9w2@`W?+5we=5np*Bl%r{7)6}
zmS!{@V&H$ZDM7Fw#9D_+6^^1SfGIf-k;F6A2uLuUK2w|weM=C6{ISFnD4m0$oS83Z
zX>%ztT(E@4mhm<~S9ekSjdWViwxai@*+H5nrnjBscgQ1r>6=$w)`mL|_;uxRK>q8i
zt~U<gZ_SQt`zt4ct-2r9LMWt9O`!GxQNoQ*@!3$t$!GzwasMd#Qp&7fK~~ETfu13U
z=X<RHX_}r*31wdgiY~26z>7VXcG%-9mcAjBv2sE<i>n&g^x0nbb5LD{z;ok|sX%a9
zJkINv=?Fe!b~9tbRAR_uZzrd-tpYG5KRm4EB)1rTPrM;8LrJ?No`H`i*LQdfU{we%
z!+Y0p1%xZ<%FybMwO5g<zSIzFoju&O4E7&+nu>b}f(m)|Pna0sk@39tHC@dER{NHz
zH?_5fbV_X~`ke&!^1G?5XJ}MI$W-?zRGO%Q<WD2qX)D5|P1B|~vpc(0$7l(_5g8}`
zjJZ3j8pv2ssoGJ;;^uZ-{5UphDdm)Hd?=ro1Y6-~W<0_wv)?1$05sl8t2&$mL@B;$
zqkuaD2F(GqU!6>!-skP4=g%T-CSDr@_BFaBDrNyr%+ngX3DH*ooREm<iuFVflv!t*
zuX^GxsfaGQW+U>Y><dYpO(T^<T}NI2B>vL7QUH3f#jR^MR7TzvQ{&f&ng1+<xi2|~
z2Gj+D_`By&dvA#dy?NAe9Oyxs1fOw8)|)onWQAw!W|y*L26SY)jZ{XGe@xWU(>7CP
z;Q*;SevlQ}v^UEh$^Jmv80+tWhG0#zmUMfa=6D5vjAMZa_ZiAevd@&oJt{o=)@^DW
zeSQXGoMm^9sGivBL(yb2%#_51&FKxlOteukH1bNVc!0O}=e{Cfm(^OVbfzJ+yKl*d
zu!j@(EdbZCGbcLe|19GFUef>m3R46T2&V#78yQCbRipe>Uca8jO`vf}6=8Z}987v4
zx_^8O-xA+ovft1HXoP=@od2)Q^6yXna|ey0AgWyMA<}xR{nw5H`L9;zQm2c+_<$Hb
z*qxeh1iq0oPr`h*!2ip~L14IzZ4<e8^v@#ezd!gNU*)4B#I>4!Of}sRNK4K4d`h|_
zU@`PgJvf+Q^1pti+%J&P9|ecjVL!m(PMbS*2!bMU;3)b#KKsnC=VgE1utIS2jI6hB
zrT*?S{<<Ik<Bx4i<k2EvzHxa3J}ewq4Jj-51C_}qS>WRYN*f{|!fNwZ3F`a)Usv)v
zgwhd=xWa&b){yracN1>wah>INKG#JEGQc_Ab|*RA_h)gFp2Qb{i#p?m-@_k^$Blx2
zBi8@Vn=ej>3^L<uIbF7f2*iMq!$(f*@d5&$FMqa8|M)!yD33t=wm$FWS5Udej=&{I
zVZe!uiM4+61(tC8pM_~IGx*tMQMm#C9-sX4VEo5J_<vpz{6NJ-0G~vHnD7kp&pSIn
z{kGWyb9eb4UEBXXegDS~iE#jh27mMEipD=K%`fg7ptRy8(V~O=`2Ev&-e;tNClJRW
z1TpS6-V<4!fb?Y%?_Yr%2!B$0k6J_IMF4AadfO+!i+6s5d02li5-!I)a#{bc<@!G^
zK>#s07>+$Ka6F^?&+k8q25AL;ii(xR(d9fALqv3vtA|$HP_i+{ZL`#lkt<Lpa{B(h
zS<$9`a&-?>{;WelvlE<i^pJ^FU(|icZRQB*BbldJDjQ7%Kwu|JcV4>~nY}YFF!XW+
zy$LT@=(P(Vmp%klynfg4n=tIaT65QJlWnU1AFCtgCH$tna2Z}d_|Grs7ycmnCY-|G
za~9O@osCZGHjyi{HKwUY;af#x7C(V0Bd2L^Of`^s#5EMeH=n_+VOHOn#-YOlYruIe
z+zD^P2!M`DJzOdgIh~CHF57PeVbyhZY+ujDSGC)(H4|UUr7VomYcpiKzGwt5BoZ$k
zzw5JVfJ=pg=D6AT2ftLf_#Bsz>(jG@zwsJ4l>2`d?O*>U3tyKcC4xVW|AKts@J^f5
zYfwcPgQf(7(8$-xUhW9)_qPh<IoQA_$8n|b%M5OiKx9K{2Djz85aUI=DpUX{{8Ksi
zLfVIkU)1~h?rP*bghrxe99c5PX*GGn)!ObLMU>$-gA((&-iDZLZQI-NKacu<t_vCn
z`~=G#Iw=<V_m&k8pgDrk#D@R_zE-VHqkvD1UuqUWb=V1p+v_3i%bknbZSh(rS{X{k
zdZO=-v{8R0{))xF594nyO&eQ{{TlxBKaVW#+Y#lre)L=L?R<Aws;sCClkhVn^BrKL
z`#IumGX0~gRF8anVJmCi;FBjz`A0}Q_Jb?izwhg7+J7AZ@qXqn9(75Rx5F3zDEij3
zF!Tw)7tuJJ2&ihmZ*JxlbF}q$%rQTj5t`eonwvwmXAm<9V5{6>ID}-m9@$z(1dw_}
z<E-e+H}YfFj(1;#*o%RL)Siy<+wJ^#_#A=R0EFxs1D{2|_4VzQ&bV?the)J*T|LLq
z?;h6iLlomC{yol#?jxW5x~5LPk6U<|IIa3|D3OK(Yl>I>{-n|G%weW$6aCPOI$s^3
z=Bu{k-%G6w*lAU16d|z&E<NmNPA>5)H7(`0lpH2rBhm3!{ZlblWa0LQwUwH8)wewq
zKhak<l57DNpMYIkNzZ-!Fj70Fq%i9Z7I&rQ2~vAhroacyK{=(%K}EEk!=m>jd<Nf%
zWS0g~{j|?*8eb-s<?U2+HuZ+_^fmUl#W$_{k@HQz%4mOW{j-U~WB~8}8+9E2>dEtT
zlIUpTKJK6LBr~i{)NCLSHjC)_@@Ylrn^hA)kadi|%UV98|GFlZ#uRKW#9lzhLguwe
zbKeFW*A+GR-NVQnl-*%8H^lz0-=)pe{dQVDVRwIX*qCY5I>Ox;z23`zd$Peim{Z=$
zDJD<YuHwWgHuT$J&B39`qCd3<p&dQvhlpvTuCJ4&(;xFpSYG$dU;lio9YDp)xPM{L
zm$5lqRJ4~7BG}@5pHaCZcHz&g1j17BM&Pf}2_)@g0B97;Oj=H^d*$3ur+3ew@_rWo
z_mT!!Fx%J&Z=zeeN>sG-5l$tZn>Oc|op8<=xXjrX+HxE#_AT9brq(iFm>$k-NUn3p
zK85eko{U2UkH1DTn8^3%U)q@zw!nFU0e8me$^zN;)xdryly}B}+jkYi2m;TwZUn6~
zZ_S}+pK#~>%^^i^9NB`rNN^%t|Gf(2iRju!er|8;djGHl1|!C?ywec)Ij937l(`7i
zsD9jT@wcnD1NU@Qw^QI&pnV_^bbfaj(0=Xo4FN60qwO@?o=_Y;xe+fO?&!4&02XZE
z>+xF0B$DndVg$Ug1!9P~=_lgh5S%ek)2)J6^i*T@toMFR->yk7unl<xG#wP7z@dN@
zoGmQIshV2oaB#wN#AkWp+glO>adO`y2{(RlVjh98Siaj;oB#;iV5HX=2+&^O8BISp
z6<)cDBX7F-A@*DFz1y^|F_3uRP+WdP%e?)8AVRPDVNudOf-LAaJZOsY#YtSx5wM0T
zI`F8W*w_Iwl#+Wes$}yP135z>U}nw-Hz~kLrdJ;5J^}70IH0yENFo`1*$A4za4=aG
zhId<VNGA$6MApm%j)@D2IxPw<(eT5dS5BaL1u`2?DqkQ*AGwVY(FFn>u9XZru4CX|
zWiSW>q}E4_92;0V`{22)0{;&+)p6rhmrzniaL6NXY=CDpbE^aVE_lfG-+Xm@7jS>o
z1^2LFt2!r>BZg<9oD;zluUY8fPp}KfPKwMiV?@Di2?zR}5Ril-WuA)C3hz8K0yOU^
z+fw#`63S<Frv2XS4~X4Bcc18aP4&;_y)ae(ljK#nQ;cfd8Qmd#1@R9|iQeo_yp-+8
z5ejMTd`q)As<d=vd{6LoT30mY$O9NV*j`-%D7U1xC*%=^Eife0z67R1V-3D9QzP?s
zMx2ep4N?`*4z_@qN=4X^&LTjozKZ*pZ$;3=K|q69{%rSGKE<7thF2emnOofOI#UZf
z(k>b<*Arc9_yJ6sa;FF7Kxnd!=N%Af;9(1G=T`5QmZ4q?*8Moc)zUFWvHwwmGyzN@
zoD2r)g5A=_T$`KyipT<hz-Lmh_PlAtkKF)V)qL6tc&E-A>AH@|bOFy1d|*W1_cn_M
zeWEk(HxUKMgs=d3#i;)pz|kJ5h+dg}j}Il#V;rRq0ERmK`y6SBEYV2X1A@PJjz1c8
z#kW$&TlJMxdHREPv_2j`!1ls?pviAARbX^P7+_qFpU+prZy)?cso_-+AV=~&wKm;+
z;0&$y)=3y_bud0vBODv9r{TB!_!&UdyE^HIz}2G15qzGdpi2vt_**aZYeAxuq+>o)
zYX09+Z`&~M+JO*Q1fJRZ+h^6I5c--tRvDddnr$;OU;_d{gsFY{&K8XM(tBNv-x`G@
z>(gmNBT4tJ%jrnKy~YnixKWM$8~!=7<Z9q|Gesz#bID`n>;|+Nwxa6GoM(nthCpqm
zyr*ma$6>vaCq<P~<)hz4O6^7aM9jeQ(7*fs>MIRsDyVZE{!=`>4xy%${vr+Yi)5Xo
zm+V&DxNL-bX(^`}`Cm!PlSHy^&`a{YW(lB<Zi!@#qE{6;mZqua>Uh5h<U5fG@px3V
zbdscs09Q$uhDk+2pW#uV?!K9?vyN2S=V|+#1@|%P_gT4l^~)$CWEaODeT*|v!j-xJ
zEw&GY!n}u3m-|gS-+^^p=#(CrARQ6d51N4w`z?oU{a}z^a!l_Bv>D#i1(9iQYM^<h
z)Gh7!x~|G0w8h7XGIP=p(?}>Guf3$}a}er978?(3OM;s>^;f~I<|def!=wokc(J-A
z<mLPISFWc+8vTx=jg-v3&7G0qNNSB&uUxcUSkFpUPFSwt{|?DbfJsXYUets_8A+@o
zCupSWXLWx*J8s#|i%tTHF#P~nAGYv*0>qkK7o7XNcarbr;W{S*u@LD*s2o)+y=27m
z5ZX~kj_4g`!atajH6xY2F~C%pw%~LiYFgEv1GYC_&#C7fGC5<8;W+0!s_GTxN+N}>
zw0e_dTcZ6*if^7T1U<JxOw(u}D6S~@gci9+S{}EbK;Q0Tt%6{oZxTZdE}j;3Hz6BN
zD`7=q6w;<Euo6*$`yZKsBVOb85z<fyXI2J-<{;s5=`LC=bwRGIAx*cwa+j9AeX1IH
ze?A-@g`>zto+5z1{**NJFtoeuR}R!~>NfyB+WJ*9)D%NDiikAAadT!TGbn1>We)dP
zDJJbD8<fnP)drhb@98DCruhCaJoHW<*dLRl$Y#>$ZYfoIr0kP+BeYR!CuHRj5|ak0
zku#5K2)<SzU?frRc{6rC&)70)DVeJp!a3#Gr99Qp*#CK3LS6^m-sMY?UAAt$Zv;WQ
z?1Ge^>{2}D3)NvT2=Wd9(k;Ggq1TFjie4&THZ^O?=dq`CWm~>~nI${OuzF(+BPF6n
znzTNvGE0;i6ycH&K=+xeP6QE(`^xM0EJjyR*P(V3IC*c-w(1H+CeUa_u2DrWjN6YZ
zmQqq#Za#<<{+3$^_Z<cL6MU2GA~`+_+7z+M$#vZ#JHhxnyiV`@uX&4uy=Vu7;~ugW
zJ0xST25AFsbtn8X?!}YQ?%|bfJ~_gibc*hbJEAOK7acr0xU<n0unc}~EUsKY0|87g
zcOxL#IO87N7M^8+ysbL%hM?2v6eEB+^ub-h*=0kD-2GUw>|0Sm7e3G`%og`Z@FngF
zgj_XV_RWSE*;%@{*DYAIvhmS|Z4D45xM5s_VutIJScJvsMwA=NbSjMBzg(4b_@}#J
zfg$;Y9(`;44EEjia_Y`Z3Yk<-O=Va7tXR7T@~-p)i@JZ$1;OzZ2&pa}>}2a^NA<9l
zB&U5gMp}eXjTuuLVUyRxj$09+NYDxw397x$MWhPppN7wDHlGn(2L5Jsjp{89K^dDB
z7d6GzscY+=j`dmVG2wP-1LXiiUPs|^90|ck8Kl4>{Mwhxd3Ao>$e720#fX9r#~&5Q
za_1c=_Y+trJvfwci7Dz^oc{%U1;dl-j`%;6*Ic-VJ_yQ+=4hKOXEu@{(fb&908A4<
z;S;k)NRh{2;k$}mcHq2Zr_3&>;rp`jc;_{OJ`X`={VUx@{tOs37N20{CFT4f-q`ZX
z0YaL2@rNM&O-gS=mWPQs1Hb99vO`lr(ThMiiomC}Y?&!s3RMYG2@^c4Bdy}E`Z4^W
z?9v!nUu_JSvRIcTw&d|QcNvY`E)kZYQk0?KTz?$f=ptoEpE;iL4QVc|u4rQHOd=F*
z7x5M-R7^M`pDzc=Pu)gzm3SK!`a=0a#H!z_$uyhLWU8kQp*GP+mm7vFb+4v=tzy=8
zE0BsKnK;KeQU5?i3b~jZ)Gdl@5Exvc4rv$K9V*>ihr8OH^HaVUpMGsMu^oIA+^8ZP
zhN~RF+`e<(z|Oj7g;tiPu51-5Cf}OkG*1cf-r=LbBikt;Gdmux{*Hl{L_%v%vpi;b
z15AiYhmW&3RfKj5ibnPlSavD-+{72t^YuX$B~ka(Rwx0Nuf0H-Y?D-sN;rtY@5OBl
zmY-so2HXzg2Fc0O9}XMj6C&_kq>@ntK5<LMzrU99FJIXRX`jA*jU99qRUV66QMpuv
zk9bx=Y}w+3TuS+EHml9i8)W#7B%ZUaYxb%PDffwODHHkhxyL;J#$tQBW`8{TLqrjZ
zntEhug<Wd!*pTq4h>)QV6aI4FUWkAFh50qFW}jMjBeqy+;bOZx@(CPn66L$^{kzfK
zwne5KuW1wn7|^2UKw`^om_pdu8ZAXl8EIYusYLo78b%5$vpBt^?W=h()rd88uG77)
z7tOE1F)>b);D#PcWnJ28=Kt*l@JhIAO>`#|HZx4L8js>2=v`?Z*s}EHSeT&gAp9$7
zV%c%Z`kslU@<o^OWiVzxEq-_PS*Jg#y1eNh!4_4lrpRyMaK)Y*m&=w?5B1mToSnuX
z^T)Nw^S@(;(g{)IYn|=tZFuaoQBT?qd&$2$!477C-#WpIO1}}6*LAuU#96ej`I#Gk
z5DUiXE7)L)nrg^hfx~+flEi48+Xmxa0ZjqPFU@xR`++UFSM!6j+DsH7eP)zq6Rdbl
zhntNmY(Jc1nB$d;N|kuLv`#)^H<hxGSv{AswR-gxqjwSs&NjEO*|UoNEW$2%S(Xn<
zGozZfh$Wb@Ej1F!6lVR6@d#82ZBpJ;jeo7KF`{6u3)#!(UodKn&alS32dSrkohMPd
zO%ubL1XKCWGl`a`nwwG1Dilf0rAuKL;~l6t5rNhk+JbT|56l2*9a>cvcsnRd!Z&D^
zQF<51Dz-EX9-b@(qUJ)Hgdrw*WYX(9deHrMFP;hr11vM})sROV&m-ch%DQnywlNDA
zmQQ)zTOY&8$B#?F{(2d9*p|Lgx%ynw1r|Kt=hA89B6_1Ch+G=?5mJvWHqf)&?u;x%
z13LF%q9%;LJNH};+&^Xw=^qP`$qi6@5v=0a)3O?<fKgDqg*3+VWt=rae|W+hn+{kb
z%I02SoF_aMAF0B6Okp8jt69&wrQ+D_En|+*B6mT9WB;RWlMJKRH(0ja@`IOb{8yP&
zktcQ$6hyvY>}!2B%<BCVLCfs*U}9zWF#sq-Wtc9aifH572dRT3U11Z_q2a8q$vL2#
z<9E5MT9-);e1wcDU#uH_;<PBBh=+*OxBsh?@;PZ?OiaGvG3bKxZAhet6Kv(f8~euv
z6nyQdB1|6(6Vp>rZS`7L=KE^?>Y^8-de*@GI*?x4tZ@ZI(4OWSUPrf;y%2>&_kXLO
z)??w<+5OroR0R|)sCs|MkU48=3kisR@ZJEhl(hoFChox(QuVCDj}4U#jdZVT82?@x
z>fse_*Sl;A62c~#x!9F3OQp*){6RodDmAY|i3lQq-ct^2;;MfazIK9f*uNA;TE)o-
z!d$F?Phc|=aCc7Kxk%A-G*S1Me}Foy^xSV&$@%8s3kyOjm$HwF(fj+CEe&Nh0l6C2
zcksp=lO3_nYmz&LjHqkOsNvY@`bj)eEi{rac*p1TnOnw`rLJigYd|%P%4Fd2#jIPD
z)Q5nj**+a;fKsAs_e%1)tHH#ff~wuxMTqSwn)R)x<*cJ}cyP}28?H_xA+m0q76;7Z
zh|}@4iSHNZY%8bH=HHedA6I2vS~S(1>5Ggz^<UEu4zJl=_Bs79<bP*&wJ<z9z~IF9
z8=Zxnru$PCPmh1$=EyB~d+kP)XRou3XVIPFCx-hl9%w*U0?P=|a?iP!6X$QX;f>%;
z6!+}5&xd=Bdh@Ky-892x?#9p=5llAw%jm3E1?a78zPGBMLaU?yApTLdDfg*SpY)xu
z>8%@J0Pw7tIjt*-pe{KMZG$0*UQ&k4rK$frFOA*nx{@+Bda;h>f%Hu#FTRv8PaUu7
zM#?OLconY9G}~KB!*f4=YlXoEzc-t;z@bX(iBQkitj5Pq$&b;-O^9^8Yj4Y+SF)B<
zyA@60hxrKkANwpO7%gQ|?@`y}xWT0NjS_GCch(<Ay6p318+n<JnT3jMrfa@}q50$t
z`}oY1{8S7#HSTd5ZjNtf5e_Tuig#VC9q=M>KlX6?0A+|yPo2oS3!Ht^PJxMRK|?&Q
z{!dhx1`|P$nKFX@3uOz+La#6IqreaYO>MAiUa*yjZDY2mpZBa)erzY0y%Lx?=Nad?
z(U9#>2B%WeraJ<|CO*L?5HB2T^%7l@3j@e9jQ=8$b{dD15fq%-XvLP1ViB6HF-C+H
zgaLOTRxUKbAod9NR7$ye3nBap7YTx}-~`Q1IEJFcubVdRgzb!}pAcx)2x5rOd~_dZ
ze{!~SP6$nfL${P6?MH<ieoTYWZ~zinkeLpKOK1%8)iNBS^d1I2g(dc7T8uGIpSuV~
z@xW?)C3q+{*G$P_PX#6F?vZt0L!2i=boVsn35@}~DiMe5P#aGXBAx5SOj|kIJbeR~
zQ%4lNr_4Nfcop3mm6lD}cZfoj<$imPL~$uDjxCd{EBD2m^lgn5<BO}@JvKew`XMTf
zT48~=7Yobh4Miu`tGDgHFMf9d8ml#10Lh!y8ePj*n^tSppan;!Dnq3rEXfiY(AE@V
z!~s69GL3+g+6BxqGl)Gtrx=b&Ktv?u@uQOGX57ARrs)SMjiL&U^G8f~D7{&j($+ks
z;r^DCW<D5>!a&PYksR==%qT;1LX+MA(alIJTU(Pp6Os8!E!<!#1Di?#P4J;$M8l}~
zomCiLQmu-i>rn?<mfO0Xw*KM~XESb)7O<-(3$p5swbOsISIS+M7C++m)8$VX3Alck
zYS8YOs=eA7eD{&PP*uuLf+Vu!uhV66T<=`*CJ6VtC)rmxS`w5%8f^7+*o`GKvkO=&
z5r(`>-=p_e4>uS)#KW^nh-H|M>%RB!F|rgC`{OMV;vH9p($uwjovKZ{Dqn$Bj$T|E
zx}@4rtX}H9c;%=2PVfRkB}cd>sPm}vtL>+3J?g}eAgJJ)_zKJuQMS-eVt)OWzOhc2
zco1YoHC@BuHgkErj^yV=Ws63|wBdzo7RUsl5`OPIz$8%@gC@A8ev`7^8;g<m>>1~8
z%Dx)ZF~*!dxj90(6U?w2=a-H~AJCcOz?l>02q7^QL{A&&u!r0VCa6ec>&O|fh<a6G
zP`*PHUftLYVZGkA<?Q9c59i1<54KZ;)H*5kmDeGlk-4hr0C!_I115$?+d5gIYrxB#
z#EQkl^p5^I3Mi#-IsuqPVTP1*9t;TpK+eoPj`c@X2N(Pribd|ePLB<=15CdX6hZ?|
zn5=^R7h<V6fZJW%0Dz9CdmVwQVQ&%hp6j>d%7cK=Gl~R`YmBcj{VgX51K$uXZjJ+1
z<Jh6`En=m!qEv3Y6#1YmW;P9Fd}!T7MY>^=X<W=x0ogibwClPUznv`jxI;)3S{oUM
z4$lX4OHNa6Gbay!igbsGDL^9(qgux_ixMo;e&t06+v;@$9wMmM)YeiN(hkbQUaFpF
ztF@h*e%ZCL0bRPP&^F?=K+?BpE(nX<RSiJ=90qqyFdi>ZXxjY$IC~4AD!aDrUj+$i
zP^6m;N_VrR1f@hoy1N@eVACZnxj{leP(X6iNOzZ{h=g=^^R4ZD-_P@W^F9B0-}yVt
zC^HCqb6wY3=XtK<_#N3rmnGHlH7}KC>M4w=O#F{;iA)EdfI*Gc&WLP6e@4yk4<)fI
z%dUHIwJ{!QJ!pR>a#|Ut-RQr=N=%4c45(tcNheYh*%y%6FFK?=Z|KjSUrB1}yqR@P
zeJ?>dK;iI1?*;WR1R2N0_8SpZS#32t7w>xy57F88aTNM70-QbRN+>}w%JzhiXlP+-
z3Cd@lX=?gvkFtqvOET>cZt6;CLvIs>P^Vd2WXyu_ADYCadXHBhOL@ysM{`3``w&|O
zO6e{o+fq>fvZF=n8;#<=0~0v1bB1Y%0q2g-P)0-~eo#fRk=x;LSG1i%tSi9**u|5f
zro~qyoh;-xpczx2Xxj8#URsi~@wCyQ_PN}n$vfgbC+`dD3GVGkLB-uzHBVfdz;~lG
z{w#b+!=C)uRWa?#PrzvL1|`Iom*GDrK&W&yL{pfZ;jY{kGO@S{z1xovM}~t078HIH
z{mN(#Y|ZfCMmMeBfuV^7pqBQw$XtVX{U}K#>{s>Z;JvJHj=N5zl##?8r8gMX3xLqV
zrR9fgn6WT0Wk<T{sumK0{LTb^kX;K_Dnqzgg>8!^-A4|)a^%8Ac*(o{Wx~B}{5sNL
z-*=f%BoxUzpS&>!fj(Q(#AdaQTGNr(>#tY{MTmy^85kO*iCbPBn-b7TNRil$bhzE^
zF_Ab|>*e7NPlUh=?duDWH4(~J`7yruXr*BKI>#9$`Ik)4>2Hl^%c<XemY{=(g5rEo
zKPQ`R7Fgkhzqv6Bispa3PggKd@R&$*{y{~7RMENudgsy0Ya_+#_%&U(DVgZ<YoJh!
zfw$F8FR%{QeFvG0$VtU3@DHX$ptV=)wRoLQTq@P7pI1q76Qhba>84ZA-f@(=`2O{a
zRM8NXq8LNC>@3>MLeYRz)CxjbvS&y<hIN{$%n=fc0|{eof|L#AY1wPh5opG5;YsVA
zegzd#1UKqutS`ML!vY??+CQ_daDdA`LVJev+^6hT(V}jLGva|H^XT=x>Cm;_A?iW0
zq+cAzhk2xpoWbTcO8Ox}=!n{$T+u7VRQ|6qX6bH@QFEBI3Vg$=JNKS35*Q^|`J1wK
zI=*X@3Jno1dYC06FB~YBo+*^f#j_><R-tKcL(=P&&9m$n{e@pdygDB-Gj^M!FK-D<
z{m3Dkt$~6lh$E3@Yrs!dZ4cNTA)Ea==K{hhRtaKHnHU<S`LS1H4ZW04mxjRuG;<%%
z6W{4uZ61!~Xq^qWJjhO+GwEfkQW!F8JF!qsRE#@**88CUH$?ulwI0KfQQzaK^Mft&
z>ixdt@`q0`WFy`2savLfZmzKI293i^NQCY$Sks}(^ph{(HY|C}F2ZOZFpIarx&2u2
z?Ms-H`KL^X*z6_s6aquP983Oq8L6|@#-*1i<Jx|^sx2uFlFUMbbIQR`AuY$7LPcAG
z=z0LVM_M?Kx5wQW_$vXoY)>c82JelSVVF_+PF6Vk^FYo5=($Z9>2tnMPWMo_{X~}%
zifi{<7gU$e0ZCJ>m!`~DM}=cgYLvoM4dYu1?$oOkj6FI53|fx-O`0UhzR_BLZl!N*
zORSW|Jsq72Y~GS~s_^?SQ+smoA*nt2->JtUIT}ul>?qLAWD>zMw)S<JctYi*DqO#6
zCvYs8bLaVNn+Ad6s-%K^pj?3~zG;LT!I|l0Jil5m^R)`P>cprSKJf;FsH*zN>)vHw
zX$%qJo5z8lT<!;d(>Ep{EQ3j=DUsflr)^_wOM(%5<IuubqBNaPmu1yz^50Vz!e2am
zF~{AR&Y#o1esAex*IP*y93M4PI?5C7LF76^^rj&e=WZA?<OQ65ZdE=@g54H*&K)>f
zZIRTaKAnV!7#|*F9WACUa*g=74yF{Zw#dJKr>0@ILnFEk9eKlf@Ll!yU&pbW_wb)Q
z;3HzkHrdAxuphR`O0z?=D~+Wbv>G(=dZ3!H+sPmnH@tS55UEPql3~IBf;BlJVRi&A
zQAOW4i4NNCa?EWz$dq7hboXvy_l<BUZt$}NYt1rLr@_F`&73U~QK!zF7vpWpozZm;
zPjpUANn4&CYhd8_KkHUi7W=-?2f;MybnEq3+azfO9!wV9rgvK#%=&6q28NM-(10bS
zU`+QP$0GB^HZYgxKCvgD5(-iY+69FVE3R6RYnNKu+AoZEZyGOVpO=&!EAZ~FE?{CM
zd*l{@xy_9oCP<uxt8EnIg8@!yVK!x<*3<HRL?|Wd?cvq98hnZi6c7KmNsBep@GS5Z
zOFiewS&BKjW77=<*&ld>dRw&8{Vag27^e1f?+A=tJPe5hV8{j9z=h^Th3&^1*)-^!
z*FgmZ53^P{g1YsNH<e@HTeu^Z2?7vubMLwg4-^GTu1gK$j_fzEV@+F*G$(V+K7C^2
zHzr51A1bOQaC_Li*(dZUFr)^Y5sd*ivQ?p7Ft&t_jEJl1{?|EPfFE05fRS-Dm@ne&
zOy$7S^D-MT9kJh|1bPkYOe{gYM;D`^dJpGl8~1*64~_HWn&4szHR;dDh27u>N)zcb
z*;{$c*+R$f(8COxV<$BCdo}LAd_Efp#mZpu=Gaeh_f->CPWF^=qqXbm#>OEHyDITa
zo$&XSuaCXGG>*NG62_^>09cU~7@AqQQ`<>z{kbuPaOlWP{3c&CBooEO(Z*xM4-`zY
zoq)Xg=^drPaq6db(gEssxS084$YO;|SbB#-`1bc`LnuaGJ}h|_QhEDEMoPCx=h&!$
zhoymjxCk3uTJ0a<wT;-z@wEJ;G@dr%=G*lA^$+{?aOh$_d?suhzsNVe&$@K~ZX^l=
z6l0_`cGL(V7QhLF<$OCc97hjO^-TQTW|rev*~SjWedOc5^x>#Gt5!y<^Fdj0H3J(3
z)0bqx!8vHDC??8IKn@B^3bD2T;n$y$%8w=xkJc*W&Uq2E5scGtps$oU=OfO@H)g*4
z0<7Gwir=q&4d*ACrGacL0A8H__^$zmpNj7-X))nZ?2>W}nMp3oKHpgVw@qC<A`X*B
z<8}n2nF>Dk$KzA11@qC(tZ@q-WCN@eMcAzS3x?Y^bOVsw{)X8E%23vKc}&BL_9O-Y
z#)*|I&gMi03f13D-(edqF;gXAy3Gctr5hSNZOHyc(>o^N(zRdpnG)p@WV4q<NceSF
z&KhRMBnRhv<Ik_I%ODv>Ri-=2g7bmWQMwivLMaTfY(nf`+S3#XOOo%x8Rf4_>z)$-
zn8bYCsVH!!@@Ix+8jDHk_R+k3j;^omR^X$O)d%x<Y$~T5vaFgVit6i%U1T5=5$wu|
z8KztL@pvsJ+$FQpIwJ3#)!@vj%cpgqD%y6*d_bGXfsh$dT4CdlqXI-VTgQmV*@v3U
zUy-9M@q9o4Z!3#(E3qQ2@M86>kTJ)|Tf@}6%)v29;&;96s#i$K&(fC_2)+FLSiOzB
zAKRkWasq}=yhF>AbJ6A+@679x#n&xI8JJ``)YA&_XH~6$#vI38#u=x(+`a$@%w2tp
zhXU7|U@4=g_Ir$_n3FC{y^o(7RTe=ppYIK?FC`#58D<Eo{2+?rh-OP|V`U{k+<VbL
z?n$r671hJdx<AU)OP}{$0-*I2>?bATsF&#0&3O=%ZH!+@2hSB_8239Mr_CDguNhLr
zjOaVL3mPpBt<7mDxaaOT7RJ#kRtK@ZSo!cY&!PP1!?n(Rv1!sU5#t&c{iaJ<6oP3y
zPt(GkyJw7pL;4a0WQ&W)1k^#z0E1w!OuhmlTDA##4;2%ouxod<sqK)r{^GRUfo555
zlPVAWx$?XaJp6+ivM>Y|(?(i<O%K<AqO1I7BJ8orCTr^lE>{eb22m+^Ed5l)T$+00
zjF)*z{fCtmi!u(WQ<fRxldNryIQ8)XZY68;C6ea}jfcx+!ad9fHHczhuJN|JpWG|E
zL9$(j;dAvupTL8ch#MkHWIgV64XjsrsHXg{m*9`R>xK6N++9=5+Y~v#I@*QmSZH-#
z8m-Q&=p<T|nCkuw&tK3u_AvJ@Kh4rtuJ-xy?}Mx(Kx1AI^>H`v9kuHiA6PMigY?mf
zfS3r!tl@3`KjNp`cNFeRP8)TuBeS=35vGRm)L7Nnxm#WMXqB2+pA@vX_MAbd`e=>A
z`bBpe9?>`wGOi8jaTv@bx)E~X-SFI;tZ8rJ+YlrAMWKQ^9CEj}uv@bsApR8TF^ZWk
z!?rus%%k2<2X>?lkxjvlP-E66oGP<wCUMrovbaQ-%9*_$qZ-9r^Z7{qz~(2ygBm(p
zJuSE9CGEIG>T<7Wc1S_rZ;T()@qVXi56t9-pY<b;hqr=)oA2~IXZbo7D6@(0Ly~zQ
zcJhszf<vwWf5k;mG)}KXAusN<Bp{<(!_6c2AmL;QPh_*RTS?i3<7a8dMJ;yRHN`_P
z{b`}u&iml>u19gaB}I|&#lT*Oao#Qq`$m$EQy3w(;I$u6gM)zqB`nk1zpNQoe}Rg8
z|2^+V%CK_`4<`E&4=Nwk<8+w^3leFh7lUF>Pt!2Y1d1E-Q>aZB(~fi#_9foh(y@0m
zw=_q+g%%GKNI_T&A5Y{vsl5^YEc@AKm>A3OeVEU5lv1vhcErt+=6NUG!xYVtW6ixU
z;HiBn9-2*XT^)@hyYrEev2um#EHIFIW+KRxa?2orn=G$!Q)~;*l@z+<+m5>~v~Ex|
zR<58i^4jbO9pj<RR_79H3i_r{!*Y8mZcSG7({DIm=x1RX6wmOkULfXj_!z#c3@BD=
zmh`MU``!)bL9}`h&4Sy+GGaJO;s);YgetqV9cf?17QdrLobn{(dJgE$6R_?ZD_QH0
zMp?*yw$D|7XKi7)+W#En=S33HEx`_#n-;8*Wd1oR%sL$nE{x$rVV}r<s5#O`yKpBL
z$b;RMI1i}la9v?G7BjTV68U|-=;a~4@N4no-fhc`HK^nzg`H~ruIyau9(QtPLNH9o
z`ECSHv)?hfK$l}NwyhXLx*IA_i&;@{1I}I_&!J^(>Q-Qm99cd|D!uiaoZ>V4D@V^t
z>#DL@j-xh!+|E30m%b}6fBx>N60fCM|2%v~_@MQ2^VkaO^043l`-t0BFVWB7o@Pw#
z=d^hZ^QHH3Qe_cD*JUp}TK}~jLZPN-_LG9qePTf3cNVpTb2I?#1(Qy-3Y4btitS@(
zX-YMQMG5D;il-Y%=l5<QnX&Ce{wF|iR&?Mv9?XT73tLBU)jp>#j=H3w<R9$vPG4gL
zPF#hh_L|S%`l%09sf#VhRt1Lnsa%F6B9*7c6g0DnjkUxqwO7j>jg*;G)uH8bjU5ZG
zT(N_ordOW}ioQ%!bFy9vCj$$a{ePB_&+*-XzUWYn5N(QrDZq`x&D}W=$Bm-X2}}yD
z%tfvD-zlASklPp9!FfYon4>%|!cI|!Ub&<RwnIaonVlx4iSdUTPB5F$Z}umXN4++c
zAjL6NsLbt0nyX~{OX#l=qe`Qgw;_CIT#tHj`O$Fk9R)iCZIk;xjrQWXUSo@s(4?m6
zY997;12<JeM|ATI$ezn0iGfA+Btuf}KLlXzS+?*-Y?u8Jzd41R?EYJkK?*i8XqS?n
zQy|+Ywh-7`YI_A)sH!{d!|gv%MbU(9HH)(-pQU3)SAE;E752m48(#25yMvaxbXR0%
zbgo>Z=>-3k1_DyAA#&TX1G*U-0t_?qOm7xM2cN5JiU5exEwnG_n$f0lX=Yu*`@ZA;
ztSEjF?nPBU-(t!VPoLyqM&4EE@%s(}DN~L-*xc}$;Jlwc&~_4|?+4zdZsS2b{7BG+
z*6Y~xx&fIq_1&#2Gv$!8TeimNV;Jvy;vW0NFIo+|<@wa{VkYmn`;?xT^e+)={sa@-
z`1-y9m!2uUtkM`!WzYkkznkPTY(_;fCEKTV%#8ZjFbPDL_vLq&lGFw^x}xd!Y*#96
z&J1q9<Br&L$a!#Lxy7Ok6gp18(P5>>8H@GWGD}enZjn@JH}{=Z-8B!`y^e?8Vr|jI
zU%AN^X$XIj>aA>FM*kgDAK`;QQE=~Y(>$|<z52@2n~2vVgr$HG>*Z+lYB74LkZ<Hi
z;I$x*maL*agxf|m2Q&lal&soM*`WD)nl8(N+H%(l`5Yh&avD7|Lz?;)8_lHDLNE6c
z*4DdjJlXW&hvnvV%SW<(gZWM}>8uT>3)3)%55nH6*79ot!<Fx;&iN6aG|Bi15}msO
zyU0mONTe<lFATRxw1PB#(BQIG1J~}AA0oqMlUC_C8yZ)XH#&D(j_m2ZP<z;UHG4$Z
z+D15X^}0u_@En8RAo;iN4{6bUUkQr8cceIS<G^-)*yn@b*&>d=2-|%F;_a&BgBy%5
zr5gb#FAm0^vh-WRW>6ICOyatdCnAo-{5asz^HchPh&h^{JZ6v^Qu=b*_lmwT8>>=d
zVhs_cSv&2;<Wd>_IBXsr7-lgJ1u1k`i(QHjxUCz39o0M5Pl+3^(q=W&&Uk=%5U4#o
z440-xS0J#huJ9Pn<W@QI^L!hsEjMEfG6GVl2`SU1>Bs$GUD6i)sZrv$Dkm#>tT#B$
zGZq3IfGL{&0V9AeP$`%XrRPP;?gYnCLOZo$17c%LH)6|TFd5yD(x@YX6u$B3m~G<E
zzMq{c_$ycUCgQ|WH3tU_Vb)4|#b<tKK#%r%a_+WN$Hu{}jT@pWgT=qtYL!&d>hbhp
z6%@hD2;e2vuPCMFl{KO3`PTV|=-B9uxa>$N<&XNK_`e0-5rn8TLyBp_YUnizWM#hL
zkKG|_nw-$%Nn=-oRDsfNRQ&Kb{?8OlWP_6u11+2ZoDg}<CyWv9x6^}8kYb<NUs2R;
zjJ(%yzb2FG6mmIyO|&}?>y#Wjn-YUmAlyT{!IaXpp-QDU%XmbL%RY51)h489Y%bH!
zSi&n}uxOaLLYt$WDUXm}q$S}-Ez%Bn$-_!${Pm4HqtqHnemK<YB$nQGqfq)OwrAug
z)N5d`@t)}y1_43iDK4GGcpFn3_1Z3tM0@aE-$@)ThGE}HtbqEDkHFEy9|9255GIj`
ziGMQ_LqDt3)Dvq0ZaEmu4E_h2=3TKtk4{a~Gxd(BS@_aju%Ad@;1j(BDDAQybhj>T
zs<j~qX0jLB3Sh1)FgXg|iBEI0ur}Dw8!eiC{JVEV_d6a&9{yW79#`~mT1Zm-+1h9x
z6TXSNy&X;MxNWk7gj-W|y&+fFZYatTguyags`kS~0<~+-7n`#u0e+tbi&%S7K7RF;
zv)<WEapiSYgO_j?=xoFk2#?+tB0Rez(8IJPFOX}a9WjSr`h|$KSXdXRXOl@k<-3|q
z?fWgioX!>eX~E7{4%C}#BI`~YD82>WQq+u)L-ppe0k;wx(h3*#Z)89&I)uRujEc7Q
ztI?TLiN@+(<)WlP?zSYI9&rtwIH>cYs%k~MZ@@`RKB<36C^TID(Drj@7@pghZ4BOl
zLW0x49W29=Uy>B7vO`g$>7zxXod#S9k@y~0Tj23^sZG{kV#gO;rFYG)d9)RRv@thz
zNo8(I#(hHc7;#RYYjEcM43Pfb&k-(Y;rPXglA%SvZqW;eo}boV>`H-W>;i!@yM$r~
zQAp0rLz_I&oTw!>D!*j6QV-shB{{SndeLRX>HH8#K=WK#abF)MO(nx8P9BkUB0SIa
z4wZ#TThOf$5-V_FOfwncjsn8#^H1pp*(}l%cG$)QJ?eFlTA&XWaC|rWO7mBqy~6Z9
ze|N{c!{Cp{&Ssh&a+I(fdlq4-XQvx7vvVm~1N;_KTg$uwq8Z>3?b9CSSU36X3?~P&
zmGe&x?-mEmp2icFh&Oczjo%LySrl7#Oy20?Q9E~cT|(bmzmNWd6UO8X&I@fd0EmEJ
zDecG9pC?qfU1&orRR$hnY)qT#VyR3M!vP#I-QdLO%hjXh$x{0_KM-fSU~>;u(3k{j
zj#Jg~=-Ty^$5*#on~MX&n`{^q?`8j|9_bdUNi0_vdn!%1PvjVe3I_%)b61OVoOf%p
zk3ALabe^X3H`-bzx~2w><coV{nPet!0Ptv@1}jZ5V<-q|mI$zrx1Q|HWb5+J?Pz8-
z_2a|P9H0uoxAA8k5J+DnUtmTpq`~SmT<(LD>C;MCQkd-Xh6X1AXpvsm3kk#!Z~B3q
zoN_v~moqp>L8qm>3%`#gIgTYMjLkjuz{Iy75=KL&cld6`Z4G~44IhWo&pnL)E6q|j
z4HmrwnzRIQNI*>5ZdQ9&lh_}z5T_P}G%>f~R}I<Mid6=WQ2PEzfrs6)&zNL?pu%Y|
zxVc7`#;E}7XphE3!gGw}j3cl;IL%^IMW~cLCnm=q6ms>&7r_pJpYf$zb@6dtOQ5B^
zV|0Hnkn<%jHi+OV9T7I(K~tBAvd@rG>~xS7moeMGBuaZ5wI>+cUkwtN`T5LG2g8si
z^mQFgoB&*p&bdtP;~Xv2jCmuuey2~@i%q*>{V?D%KFt=%Od76z3Nc0~ykJLKRA_QK
zXvhS3<whWZ@s&rfDXQpBBulU**|C@Ix>Eio6<Veep?&b>&6<;*IqZIzefkXdZ+fMS
zGnO`9wui?G@Yrq&&oN<Lj0v`_Te#~n1y7$a+JnD@M<r(Dbf3*Wqtt4<ZyVlkzwR+z
zWET;WSE1YEtX+w03Sb~>1D%%LDt@)I%P4OAL?hI#l7>tZLG?+AQ_4pILKP6x6kzj;
zbEBgIJJGc&z_iDL{`Ld=)z&Wm7cuk^DPDwk^fK{=yN8%Q>SJ_lVZ*xHV4bzKg*xqq
zqsl-aIfKv#kr#ob$0{<dfw_Ecsjm8;KB#tlX0gl;WX-%%nwliE=w#l=!uj-tN-;zq
zIR}U~KleCf3p^!}-muTK0SYHTNO}5%-`J8mrPPU(KKwwdShdldbT~7Ak)C$_u8>$}
zH>R<6$Fih27P1K*t@{YWfBAkuvuio?N2X<lN%^Z798oqql5oyWTImd$JbFZZCjcWw
z4egXqORRgXyPh+b8aCX%kq-B4O=-Awz~hu-kKccU<GCS#8Ow~ZSAaYv*YfeV*u*Jo
z{x}RJ+^U?0+{4T>Cjao;JTc9N3dbq1FgW1|#=K>*C{ykr2(lvT?p4m7YIIQv57I4#
zHHjT4N}_G_%+W=ydUhMxxL=>G1lZi^zlwKEw6MFE^;H1FgKu>98B^MQX=itQ93P1;
zD!5&3g>*-ZSO-&|A_CwB>h#w!fvzqBZFuuR;OZ26x0c?cfWF0UOG>!iMqhN_t!a&S
z%g@tNO{rv}p7y)I)%8`_O|qZIL_(3Qp{~nu)<@)$I#^|E{GrQ`ayuQktbw>1hL3RZ
zq`S&|@Hs*2sXYUE2TETBi1P{4Wy}V;axl9qkjtYO#^sqC)bWB}7Gy{G)0=*1cv2_G
z*`Y7Jd>*(v!7y|6QeOp^B#+zvgK0987&xn_xM@F+iuqmkA?UG_nk)?ei;`oLV%;_A
z&S<V2`D-=Eh+&fVcBi-=9KpHrv`(rqj6W2{cR+_g$@12kuw2dP341M}RJmULROPR=
ztoF~~p^)13JD=J)N^pQaI>~KIg&+Sm{5AQ~HL>%%Wn>ccBU{V}QB)iCOl(|Py(M(c
zj&y32<S*$t)A+mlD&+AM?)R-v3)-bl+Wr!@0XwOI&7h;HsUlN9%U0CkgN0BlxwuGS
z%IE}oSoKcAm;8BU44?eSC9UCs3{#VNF^N$DGM#b-H8zV=%PYt(L%nG0-jT)2d6J4r
z2P<EEirNv4&iWT%7MHY~Aee5t6K-`%#I5@NQ>-ZN4Uk`tc|p&X;*2+@x7R$oxsA$z
z1EYZd1EO;YWg5;L(v^shGO9#2SeE=17KxzuzW9`m|FYalf;h`oJ2J|GEeqSp#Y{xi
z{Lk=6S1nT9!f*jK=~QmlMX)Q^Mr%}9h9B#b{zkdwDN;erJ@>GbL3Bzq-60Dl?@J6G
z{S@}6;6}Z%E7Ot{6YOblZu<7)o-Y4=-n*)@q`whFi1w*Rw-j$*p4`?X;!YjUv&8WM
z*VZv5+y>QNxPpd-&7x+-B|ud5Q?xp`svNXGEYngR$nN`#<80{{^IobYgcZ6cm2M!f
ztcwE$|L+f#@v2Fuh9*3Dm8Fu$qjAoH-e1U(&1Ye7<6~py5p_3;*Xp`p@;9VNRg)SU
zev8cHKQHPnx+ms@t}~xubD;Y{;|xk?kTuvPD0y~=>GDWc&l7eiEAGyEPIQ}t`YM3~
z7*UHacPIbc6ED#3P~3i;(i-p3*<xwq9?aXE^_JK#;-r$~!!@x_Nr3Nd!)J*5rteK?
zzB$C70^vj=Df%U?&E?sevQJ8ArDvfPUmFoP5>f4qp!b_cBWwbB8tH4nLC=XfWr*}g
zJsSd&p9ptGkY<OEOtuG`(_em?3qxx&0R%J4Jt766pOZhFAFv(2<8v=_fJ5qk1y>`r
z^<>e!a7Dz`n&UQ-o|M2Dm$Y(9OO{MXd{B-(KN%e~2@+R}_iT=JJS`U-HmO>uW%sAc
zbq^LK8f6hTEFW>JEK$V4Z0WhSr0RKU)?nLhIuz)_^?omhHP!p7rB-05DF|<;=nu6q
zBbCo%GYfyZSj5G7(}!1A_97Zvl8RTPZdyR6wBYf~=_uY2YO-@YpDaPylXQ0KoCHz7
z!l8}JTXAK<?viG7H`2!O%PP|xKc1vZY*#Cjuak{b1GR=}BHnUTY(JnGF^)0ei<V)d
zs)13ms#hGp0rp2e#m(n+i<$@%>noYsPyt!#C|;ED<A7ZzEl2V}%`=0wMcstT>Z01Y
z1=Ztww_hP5thKSjSd<|#sQ}heA#uNK7t~2kEfREj1FNw|ZM2ah+|SNFQay1Clbu$C
zj;ZrUI{{}(<(S$pnGer0ED{D8nbZ>sz!b9@nN_LGj5k(n7E12HWj)R+mX++3F#%m9
z|IyiKH^bTRRSE)zmbBa8bgke$?nPq518=*1MMQUsD;*OtqthM#<0PSTfx$Kp&B_eY
zi)L@o>A0(rN`0S~{azdFqP~TvA^cwkBR%*AH0^d?7o{Q!QRu^%<Y>PX9FW8uzu16V
z;N-;IO_}F&;$kTI@6<PT!Nj)*qd&IKNRew|xSy>SUmC?m9-}BeK}v?Qw^q&ezBayG
zDE-?QDM#-Bd=-!8m7-P@s6bjC<7<*;$*6Hvr(_DY<jYlAsMY!X1xn{TNQ*`pu=K2Z
zYt4)T^A)@+R#*5qS=6J<YRBNVAU`X=ZT()y^C4ds^w@~w_Jv&#iUG;XfYlNf(>OjU
z$sffx7aDlZ*HyBsm_qZg1I_>babP1bA2T{B<grt9+O5`=tTM}9S_+3jZCAkPDZR^4
z_Z&$1=tZ@wD-{zgN<O9q?czuG(_H%Dkrg!EYfadrX!{1NxJrxR8LXFv%Gx+W1Z5b~
z<hx6<Fkdd|G$ht03bv>hbzf;2xa*~(@)VH<dq&h4VtKrjq$Ld<-3S@>k>JCakYDZg
zV1HCUIkr(J9xCsmu)OIJr|%_B)cs8|vV^67DjDvrWyhvKoK5TE3ti){LB$Gv6C*N(
z9qtAohTk|Vr|uAvdc1pG0X$v&hO|0S^9oDMA{8O8;HF(Z^3XXN>aLursRtma%<sOx
z7hJn&`Qh}K(S@M?MvQIvsKW#iF`k1jb|7mvK?SBO%D%4+lE0H85YMp>k#E@O!V8y&
zvk1PIeTpUw{3I#s#FxARo;%;F#veF(KL~l$scG$938XoUHjs*aiL34pM5vS{wIJ|(
zaZqbJ3HbS>qr_hKXLY5_Kav%KIOQ0Zf%8E~I9sJg?;A_3<$_rI3j(*=Z_j*2l?M?I
z9fh<(3MG*cZk<X)40r*y>sH|f<5z3$3EsgjQo14I#Yh7Qt$No(X()!xSWUH`P28WO
zYKh#=n+LBoXlp<LM&|fLJWnw-mx+q@WUMjOuJ+5aN}*7K7uf&|$I`b?UNaL%9K6Mf
z{wy2KGb3@5_smD%;~%M>=5&uSs)Lq<{HaAap4j1wek>EehC$%J&2~j+dJ#Kbf2}#i
z&x)3luh0g2@U5!$BlWqv%wqaVI*h@i1MR{2H?tvrE7xj@6YFus3E^`%IBAG|EE4>V
z=oQH0agD4vay>*}I6S>Fpaml0<{o9`zww4m%rdtUMidUGqno#Nocj}yBcX0!e|-Ar
zak`DH!4h55!ldXy->8NSnTT6qqW$JuSJdGynSv9plgm}H4yu+enWkRZn<Yvt)5EJ(
zJjs`tPz!G<H~kg1_b1BtL?M6sq_^;Wbt=0dsTNlQEga#SF%cS1J`KpSP<C30A%h)$
zW;zC)>_rFJ^RRYFV1&;(XN*4%@g&aFba-lvYu6F>`ZM3ShE%trC=Xuqotv>Qs;ONL
z#IF-rI`=*;Vn}fCTf7dR^~E8l8(jvt!`JnbG&LZvoWUY~Q6!X!mDpM5<5}xo0|^FB
z&fsPZ($tDmoTP0h<c-en)<zTVa*R{`1g+e>VdWLtF2l}gY#znDyhv0XYov7Kt>0)_
zOnaRi3%Gk*^wejeR{m|cz)GOuwP0Udvz(G!{Y!j9R+J{-xaCQU7LfFDxAwp51?Xg+
z>{4L=XTgmTJljTE1gTg;(a7ZPPfB!cN?Y_C*2~M&U(Qo(i37{#2q$T8wva87MYK=M
z9czlz2NgL?0!l%dxVi}MH`Dn<Q`rVdhUPKa87=}x!3|8sA)gz$j08N4jCdS5>t7P*
zRp7{<@uFjH(wp+k5SCCQNs&uUy+Uaj3I=~_1>yZ^-_;$HV-kly2WhVpuiJLb#n7Hb
zt>|CxC+#u8PbdPS5;ZR~b-n`SK)x$BDY|m#Sj-b)QaSqSkSCJbPtzk9P`-qnk}t2Z
zkVu#)r3%k1Ja+XF)-%oc(wm&YFxPzR<tz~O#w*y{lAyl8So6qzPRa+}o0}%U*oDk2
z5gf6%y`Dzx&2tgq{&DnRgFl`&G^7pg^lYZxaj}h@6^<3mkR+p2q*|1}j2<X3>Ly>a
zjG^Yhqzf=e3o*&hu-r7S%FAq%yZv<wdIHZ>51-#~6_E#*gEO>kkL@XGY1l_T0e&zF
zNA$3*_0Pzu$U@86;mqLPAJzFDUG_&8CMX7jYGCv|bqWmH_?OE9==C}8l*tQp*`++a
zXOc-`^5TI>47rJ9p5YFX3B<{bkI#lG0b%yXhG^5VhI>g6`)O^?X1&31haKW-pLi_M
z|N5QXxW5Zhxv+T52W-OP@k#wzMXM(3f<R#ms-vbZN#8GCv_wH*JjIRi@k!@J84blL
z*L6~Ml@=ibY7-4ZEWm^NqicrRPj+)vNf1NG2eUphVyLikR>he9H56z=8_p#*>rZ(Q
zXz?IX@2zE8NX*@vA3J52>*L#doz!?`B(qB&Jn)8(CrXGEX~;OV_wfmJGywXTY-f?2
z*5~f<I6+g$DHfO9tP{0y@VA(w!%=;Z_$?u4^%n1`<UOmtn$Yo2G+MX;tPyCmq`dxw
zbc*;dVKMDnG+4RwpyI(Rvnbv<(~}E%_l-3o82CWhL!W-2!TK%18-15{m+!K>&ah98
zM6zd8+&9L;KW2uP1JvzZX1;yOrt>PLbaq=2tjlP_?Du!89xng%+`*B&v)W@#D~&e%
zVPfGmae6=u#(hs(XS89XT#wMp1-&jJLy$X<C&_MgR>_XF|IP~_xeik5@0Y3FYE=pr
z4DV=H8bExg=p6-Qj600flvlJ$lJPdxpwyDQv^_NyXjM_xm~{K(75e(apMiu44m>mc
z{)jUESA*rk-&)`#*j=m1U$h=MPl-t&(^sjAF&`5%3p{-gh;0z96|L$0z2tHS=Sb8R
z{fOB#nBP}e5@89n$GK_d;^E_CG+JGVh%hrQHhl!NPX$xri?vS|QP1O}3=3xL_^nu@
zm@m~-%AMq_me5g%o%$xS5VM)JH=J;lfpjUKlg4|j8?8$&`SG(CQyb1Fv1K2Nsh^`4
zhKIg>Wg8-sT^g+XjWw`_CTH2V-UyiRi@hP<q1VCX4Ow=Q#CCx5<^YV^mfAE7fc?aU
zh^8lFNIkbvL8P9unCYt(^^C!ZD`T!(kd5uC@$fa~u+|3SR7{rbF5tYo7Ks%2JZ;Re
zAvc?!slbJEHr9U2-3XrFTy!O0QQDTCvHgSUC{msn?RxA{56J0eZ&ksAqy%T?34YF0
zhzor;&i8E<FK6`kAU!pbYjG)hc>mvB;s3;I!A3z6h73dBp|Aa=nELa8{Pn-GQIX&8
zeZ;ZJzd0HI)At(-2h5fA_EB%=e{pyJ28H+yfP=MSo~PSLJ@mhQ!2h2Rkw0HzLDS^u
zOi|ULGEPy^%QZ{_?8H1~3ax2#|K%ON5CEh20&<m;!T<FSFfbr+2mbz6Ah>QFRfbZ6
z6`qgrtRL*|DSBN#NYQvQ!1$r@U!`;Y<L&v^>m|`c(pu_IJy-u1$dSLlgdW=u$tLLg
zAXLscj>U#}k<4rPGW?tNe|t9%k?#iU*-7XB`rZ8NrTyQ(-&I=hoi=Ezw=4cy{r;y*
z_+MVk-#5jJ|MHs9?)fo1yVE#jma6E3k|3z8V2KL4k(<>j(2RLnLIJ8NlbXO`YP(a1
z^rs-sI|c0L!olifqAvf>@ALnD(cm*bm?Y38UB#1AJ^bg3Gtp50T%vrqI)2>p2WKTu
zib)o%bAUm4D}N4LqeaAZcC5&tKZiN*lb}yYdcqOZ(1uiHD`zR;h)jd@L`C$uIt7w<
z_BsUy9^H&d7H27fjz;BqVzB(jL^szbjVOPih+cJ!xZ&#@C{V&KftAt9-QHC-QMXkk
zBz5@lOw)TM5jW>Pu8`|$4y&j6HhH7)@5DoBQE2bCGrLetJVyvqW0_|hFRy<7Z!X+t
zq<rLiSnkE&f<*w7x_t*Zze@%|2~sSxK#p-g3b0niLpANQ9e}Vu1lb}>SaIWc15LXr
ziHMucd{eaOewPDCS_%5c<sZ~oyx$DW?MHvO;}6wF{g~K^?D1#Uo1FTRxgHy9F1_q8
zp}GRYia$695KC*(w!IXmjEm%VR0Dqkxf`<0*0B2WiEMB!H6M1lW$%Lps%%*63^F!#
z9w<?8$Zbj_TI~QkZT@q1l$Ge4TGFp&IP7su%D6z&u8u+%YNu^mp*9H`5HdFzyFb6F
z({TMuItuT@VH|(zrPn4_K)<UXLRq~c5gUf2er?un6yXrw$R%t^5O7W_?cW;t^dFP@
z|K|qzXJNU1g1k>+@u>c7z~e{#2`u(s^pWRJdLRI))Bzmfk1zW)zC8qQGZ>&5IWGZu
zWP(|eBw5JW%-4h$06XqS9fJxsVOPMkC!Ut+5(F?>8lE4nRpg*3>p<3HF-rnP-{saJ
zjvxIz8Uct3`Y1s<uPtkRYw6hiy||1^1`3c_iBl&_?aj*y&UqtO4i(j)G!ZX|^}UbO
zj@|-`Hs)xqqN>Q~eii6Pf_Vl6=_y^d4=fUFZwY}|-*|-OuNUS>`Tuh_o;tdT*vB(#
zyf0sa7m>I7McMmtA0QE(pM!XCjDo_l=hJIVVtDTqL;*(JPt+55@p+@cZy4{L@2tm4
znjtvuGz1af6!8om!vEt^O9J3=^Y(4#ADRBOfLJ`ebqBjM5;+9z=Xg3b3XBfkx*yo(
zPes7bOTZ2qZ<^OxDoVjEn>Ag)nJ-3>4@H1X3n%x}`H+yW(h1Kx!HdY7o(tveOuYt1
zgBM${3_w8QqqL9+V4c}M-4VWg5&$?_;-3wX1gG2mLGtW1ChowTF$OZ(>iF1aZlu=1
z9p{NO^gJiJ2Jh?!&re`GV)q1KaDssM1bX%Ml^CACLX2U+`>5E*UVt2tW<_L56!ic5
ze)(|g04?5BPNl+BGLq_hEN&8JQW(HfvlZ5^=ifusf)4M#0cdHosPFAwzfh<#kcf^W
z@ttB^s9?hhC^ScD20%CFn+T0M`*ykC`+XaDH0~jpD|+{}Hgzj+(4VQj9J}!%Yp4@m
z7h$@*;<!PRfXMY{<}CUM@sw?U&fBS3q57U8T4uCe>WbVR4N3@pb-BXI`}tF^(L^fJ
zh5uOt|8<f6ejenI?o(d6KQh5T+f0N5YAs_^iD1)kjt+1E9dmY4IIr?_+U>WNxchc+
z{Cvah72tFH)f-@R$`q!*wbO6$M_F%j?Dg`PhSIN{{t|H9iN6+vC88bl?lSwEe#qrs
z3vFD1Rh#&i0F0w+0w!mRZ-8NZf=_U&a3a*%deAMBs}r}etAG`{mCCJ=+ge&OZP`-3
zyMRdcEN`7t(nyA`S;JlG0mshpV-WGnVldlDLsm>dNa)P-oMtCrZNWjVX2PFySA;K5
zN0J)uH=KOi`O=8ofi(aWQY`>t+42MA{ThPkx6uo`>)Hg{OYY5@VtVNLv&v_xDlf+}
zyZ9}m6$x7A9D1dN@ut5UJL){M*PA6V=8ESg$ZKCPb1a(1#Z?n#Z7`6dZ03PtZ(+IR
zjP}56ildgI-&8k~;O5N1>ghAsiXBJ0cWpO|y_LPH2YI~A_c|4SWR$+qVPmfsAX-@g
z`0ip8Xw-2(ciI9#fX)1U?Egv@AS0+SzyhtW0g?@n6-9yPs<=y#Z$<AiiQ?MUw6XKb
z#JtR}iU~dI_3UAvHKx&>J(hQBwI3!=khtmD*R>s2E~5G66uFCfd2<PbV%XFZkut0+
zi4+|)dxCA7SJULE>jR@d+sefCjp4?dX=*2VT%|)B`Ne%u5REWPGZfk9wix~AplCP<
zRAEomVV+i$e=U+vZ{L|EG)M>hAEdu=-3U1eqjCeAWD{5AXr3}&)f}u&DcI?989?TU
z`KK+i`y@)6yA76rC<_C*6PyBE)5CPY3n~G@2*Her8gvf#e*n=F1z2fDxd~v~H7mg&
zZXj8pR%5T716I0cBM=R}c0UDBsaYB#?p4syHd6dNDtZHtOSC3^IR<LsoV{o<DSA?}
zu4%Y%sX~Y<gwhfk3}Q7n{ayqs=zxnp4-g+XUEKhA*Ft7p_|d7Ni!L0Yuz8fKfNvkw
z2@x|(rV$3D_vu8;A6d=Y0*z;pu#VCp1tTz7o>LQvLWKd0U$F*-L(}c~v>h*V`LmMU
zk%=<yz459BK0u7{!2}pac|mXx@qGq3bvTf~C_kqYFchGneAVl$291JVpOyghR(>Pc
z8zK@7d-!F|jltBerUYz|8{PswvpxK<?+Mgb%riN~%#W;MOi~Y1A4!7nA$VrgCHN_y
zOxPC&Q4c|PMLE<0=a1JUKI+d*I{?Ewe=#$?ke86z>o?c`umIjpA@9y1&UX;izM{Hq
zG9|I2vEZj5UX&^L4_La066`M1MZ?8@*j2!sx<N-jH(Z=$*BT?6aq(lN1>9Bz-Ykgx
zmnTS#ks;hP5fe#hukF@yd4adI1AOE-(3`WtitB~dgT|{Z`}|l}fX=}IU7YhN05`|a
z5dR6}05HG=k5AsDKr-GQQxN;;+$N^^4RQy94|?SeLx_~8SwemA+w{Ra&Oo+FKoqoO
zbL;@PYH9sXy2~op9fT0lKqz>*+u901=06sP@;#MP@A!R7`^eE|&y3;H76S8D$-+E%
z_^ZF_a<WglZZp@1;1mG2zFe8{m&BT<+LS#rzMSGH&D}WzyM)!8HK-G?BT{9{Y5+Ft
zf$@x&q{m_>q_xzz^R?160A!IQKjiZ9IWEsYZVioPjqXihs6gHxe;!oNsoe9U--6fG
znCl-xV<zex7^pm7K_^e7|CMWVF^u<&T+vpe%GIZMqX5?RCc3rji7$T~5pl9=7VHuL
zsR#gkrzDoZBid9QgOk{3lk2{5S`pn+6dL;NwGY*QkF1gP>TSti-*&w1%fJNLRUwvj
zAHq^tBDudT&?pp(xh_roiD;@OrK>dK{0hvpV=;tSVlhvoi3tfwElkeOihZtFVw>gY
z(!NW}$^T&yVSHqo0w^&e%`qhtS`h*^y>3uZ&+K0THbF}LmaV<i10_0CbXTVE8MbZd
z0)>5`O%c=tldiF^3Y7p5HnzZC858!#l_h+!)OE&FQ*^=|{~;{r8S(xMthKkcr0?Em
z1>uov><R@M7GzrXa#9RW9Dsf?NOH@^JR&WGa5on;T8`#2at~R)vuo@Q#6Kn+P8p8D
zqAhb{Q`f&*S0^fwZHT2`O3H|FA*8KllnngN&4gMdE%#MX5aG$pLKOdIfi7+WXD2A<
zqZ=rSTw04}K|y3fr`MVaKn7&4Jpcxp<u)cGqgYiI+330?+t|auMl_<Ne)PbOD}Z6f
zK4&%3ht~3oD4<lzm%1)rIrIinWT$oc>(LR3P5&I)?(%<WrV|uje1vbBTt?USRpCwP
z?-xDN95c-ypNGU0le_YJ6P-%~%u8onwAT{-^uzZ!_|sIrjjZ>XK)sQ(oQB`@z>Ri<
zQ4)L1D{E`UD4Y$UFoKum(<V@sT;L(A?f)&NnMgA<wEU|h2dXU&^_xS_RsG?_k*IrT
z!XK%`z~MPzOFy^%S&urI?ZYe5=`_HA(!BZwKaV>dnOM}sfZ^DGE-)PphTSAq)ys@p
zdS@RV0V1ch*UCR=YpD=l{A&b^xMgnWt`;rBRCNppIl>K#`Asx+|1)%Zt5YMkQZ6c1
zIoWu%q6V-$Q>Gj2TC;CDbUJUY=58!A4nVf_&avtn0R;>i<--eP;@TFp+pNm)-7{E8
zCzWSmM|6!JSQ<p;<}o}D+K7}_>(P@Bj0a*J6=Kh7+@2vV*$Dgkk>@O58p|9@(hekd
zJ_oFJ_AXN%r8*8+@VOe(GAJgX$6<9S%=XnBlV6*H@?Wmg!m73Q>xHMT{*hFJC&g2=
zJAFb}&B`0kmc29zoOuK>Vj9b01wBB;jmD&s{=PxNSC{iYeV@Mg%#yyeQ(40@Or;t_
z54QLf<w_<mOQXFN9ugwKR=ICPy9VfW3(Y=i@NP3AIoc^dN<!I0;^KfnPQ*_mzTz{R
zR{NOGu*=wKA}%M&!66-gnh{3mEWS|B`z*59nA9wWfX5mg^BpIyImXPU{?#{NA5#3e
z7xjiAi<TRVhpiWuR>k6iA?jJf#lO9wz0$*%x3D0Uu`|lv>#AWEjW*ABWFG3~NMo7g
zX{++$E*D#vzx2nH#^otX)!LE{h}W$6_WDRFifVK>Ia$E@=IW$?GAAVq8H?B-4FmZK
zEL!@XO4%Q)tAUMe(0N$U<V+=SxAO#KfT=-3QA|pBRj9dnw`jILl49)P=blWNQ<qzL
zQd+@jxW8GThj3Gl$&O9MY!)g@Q#~nb)k9=KvhHd+=0($`Sh-3`3e(d#Eu>!)Fnn{l
z301^5kePWh2N^KGq}vey-pGFYVdlvZA3~h8ZX}NHN)Itq>6H-`ZShRu^A57eV&oqA
z8_*L<?s%WAb8Ya@15~yH3dXs{RW5%`W=ObEE%Gcvj9QLk?&{IeOb@EP)?|zwy^Dwp
zoNS};yZU*Me!`F?Cg6AL4y`;QCJE!yg>vPr+syMVa`GP*8)>@}V#`;*-Yb5&k~`(G
zS)IeVGe0;2@<2O0mesw)79)5N6ywiMd2ZnJA0wj8TJ(!cKMN0(pW+-<SY2?9?e*WY
zM^qTftY7PWSJyoI<rejX>Q3Jyp?MCxv{=hQQL3wV0M+Df0%<ayoj?wiipx9cSN<7a
z-IvW|zTH9m;kR5wUlV|(e@PoI4EhCRx;k?ohl|tBhNEFQISs6AMou3B&H<L~GHd=W
zr2%pBX!tJ+UNR`mt5vryRTynmhAAW80G^a6JOV063E@f1{yR|=n8``3jL^&qf~;O-
zM+13t#Y!Y|(Ap)<YIg<tN&Dw{Vwn|~{w7tq78HH@Lb-H)f`mKy=o^T%n!(!f`A*PZ
z5()L>bziX7LC}TpWK8R<V21@ZD9Xd<k8vzjYVzNi3VWr8chL4mmjyPKnIiHPV$ec$
zEn8T$MlH}S1Nyrh%cu2_b)4DvW8pDw36r<OxVWam5xXH~i<4b@hwr>oa+Ge+ZM`jx
z1EItL``D=v8p5n+FQOc*ot|Dxp#}COnbQOnm3ENj@0IF$_MYHh4>4dG5@dN00L(3&
zMUD1feTJVB27;cnFNIH@?^+@`1B|I;B|Y;>roR<WH&n>W);V0w?$W&oJmjL_a&6J)
z8cW+%$1W}_@|1xU3J<WPrsvP8*+TP!^*c|pzv8gGJelC)P?M#qq0S};^9~%n>ukgC
zz;l*VeD$6VC>i9H+(rNCQ~gr#nV<U_I;I;7A7E>l54jxl+r!G+G0BWHM1xC$_*2}V
zUAh!_N&KuDgP?qQYg6*=0$FTP8n~S_g38QFQ8XEQfh5yTffD5~D^6Lhq)%Rza*gCU
z{8#mpCD7VGvi*Hq)6b;m34luvSU!jSdHgwFzA98JepI@D>$So7*GFU_c2|^P+wXeZ
z5-wNGryx7;#?{v@kxz7qLWu34%%Z_u`vkZ`+@=UzIL$UiInG*dS2fsapod+r*E_r8
zm;~O3pi#?d&r;oRg3BcB`S)dd1Nt65hm~b8av3_l*Y8z2PjW8WLhYT5Fl#*3-hL#z
z^eDyF7=-|z&`Rq-e>2z<k-P8h^~!S^XAu9oX%8gG_Pm`Px5Fn!OhbnB|IB4M{qZ?*
zn)a&6?|3jtdHS1aHMHVIoCOiXU>1bi9mPt9u5agXx($HjecbI~|86NQ%yD{@>CP{H
zB^EnJ<1Nce-2UGn;6fqJr*!UEkpkvLo!5GKSE`kp#cuCpPi(Ese#>)VgdI20XPKP?
zck42A@x@U(Z_Kibu+e3C%V!ZvOV;KPPV5l6oan++H)C_Bln#IQ-6-qyFif9d#VKy^
z>1lGkNsv8vU@GgIH4r9LzbQD_!pmAi$Zpw+%{b4IMh(%{)G(V$ctovv9bbV`UgjuA
zbe(TxXvr)BZYEr;s=n_gv0lZ~FpX20-p}6ZofSp=g@rm-AT$`&E{gVQ@i+E@S(=Ze
zr1U~VfNp24y<t)b0qwS6d8oFA`GVefX&AlRlq-j$tsK5Au|$cT;p~DLW$*gU0h@MQ
zdV0EtdG3YxoX5)5_mzW&SzYkw=<9lZN|8SF!TJje>pOEx<xxI-?A_d=8T02_)7Nij
zSVo$>rBW%+o%g3-7k%rYUrM_?=;TRp`53WiD0#-vI_~+QLa)78nWgH~V_%|SksT(!
zBrq1zgz8hIc;2~a2xULTa%|7QTlk!)|IOB{OlR>7XGN<#z0+e{m1;?3tja-bE$YD6
zQcDxkOW*6*g}$@nKG=KF9#eG6KNg#~oluAue|s-bw9@6<fB=&pX;-2erw{@D?LQxW
z5ZaP9LiuXg)wxjlJyo?Rq`I*V<WARiQv1}~xThlA0O<QY=nod(wM>pU44@SZu<krK
z%cGoRC@&9lGWSO_mnS5?3xJ=BA8=mJ0@cD`C&mD5tQtrg2Bv6!z)dC0<%_C8o!dXP
z4;?7n6imIvTr0ly3v`ukQo-1JC#e96JK~R6H78mcKn)&XC8{Jzyrrp;SSLYsX3;Cm
z<+iC&)z3du4VV@aaFPO@%KFAD6363(AxhUVm2Hq<#C&h&vuoD)KIdNO#%NwN4_2Zb
z*reEdj0FX+4o9yJOre|p>g~2)EkBcJqC+*UX9r$VMe0;N>d=}gFd3fEDlRKg_$fj$
zkrM2n>m_0}#w_5zYhW(lSch!&V2@p)^5XmXM1Vu@QrTDI33~S*9I!ns<cO!1^39K7
zIK~kCNcjXXL50g*7SXUteqB7%CKKi3cs5vBMD(MP@4UIPFNAIC_uG^0Ers2c)ewIk
zzuH0aMUrv_@_O$X#cx-Dt$h!+Ia&7m`V^!ZvF<C2Ao06tn2fW43%sm6D)9162rlKi
z0)D#NvsnFTDlv-1QlyTd3krK8MkeTxNzP>?{i815dp!-4{exy|;p6%{VZIolz8DCq
z+c5G`^>)g}cF=nDR;zbk1MlT8AN#$Fl({i<D)-Df0O6rDCe^l@xIx?bUXUaLFg2I<
zLKh@o7^3P$k&>E;Dtmo%`T6rbWvzO%y}98=Wcz2=cy-KzS_3A`X`1&qC`KD8J(gl+
zb{_Uw75RiTUEI}~DAJt;UHR6<>+;ZzO_X~X`sZ&is)yhX+4%ZiA$vJ&0u(2Yr}55`
zISu{6(6yan@ix&5!4W%owz^3E>8`wfRJMV(Gnt7FJe#cH8%P7F#D0LCd2>!b<^+Tr
zbT^(D=)8YT6^OUv;J(wW5nDemD-LXf&&(5jXG;uQFDE~}BwJYfVtxaZKTaSg<qep(
zWEgK6Q|Ad@__pt9NAhy!u8n`yejFlt(7dw-j8DB&_lT~AxUen;3LP)rc4+h_Mjjva
zNV^V*SAJ$HVAC^>`<mViTFBMYv^~HV7V3Bl0{W!qH$HmfFCgb5W7QyA4@*|uh7Sk*
zPo6Y9!TzSM>2;k$2!&H<IiA$Z4QbJqx9bm%qTf=arvZ0EU!F*td@X6}d)W_~*+u8y
zjKvxc+y%OJsEFoH=bqG8<>(#MSpwF&Yy2W8A<Us26^!0#U%3h1r}QQsn*pux!EZTT
zewK}G(C9f`x&<_Y;G>06$vPo0dd(ZRC%mH!W7%bQ27v8`$jk^2h`V@oI)8w9b%jo%
zc|7m$b~<>Ymd{EBQ?KX+p5WL)1mw(AXdoM}_hm6z%~o;Vw*`2v_b(yy4^L-IsnLXa
zynfAm1|#}$*R_#PHX^Z~SX(Q$G^sQ@&1vy8t28$rcnb^)N0ns{Mv4$wojKvc`h12S
z-W34J33KMoeW%~W<&GB#qu>SQ358xh1}s~r*S^A~V3|1IquaSzh<XFYdl3wR+P0lj
zvONtq9>7<md$K*7M!K`RBfb_s(h=5{P|?3M%V1ph>-1am4(__hfmeNNRs+wt>5o1Q
zmca&AYY)zeLy*aRzQjX?V0=%x=mxt{Na5{HstEH-^S-qNZgmc*8EYpu=Vx>BeM2-=
z3Rl#q2&}j$t|;$A-AYkRWjJ@G$<HLW58lgSXR{^X2vxW|=I6StBKNGY9sV7R>m}(Y
zsDFK=ggp65TwKx-r}Lci8A`8|CGmC1WXN6eMBc*XLA_1cV-#LeZx;YsmPN*h+_XTv
zwpzHe`PcAmOPbNH!x>X!<gR)e8Om-v#x7*^L>sgk`TJYAVRNUjKg*t9g*f^0Q!&ru
z!hEF{_M0KfqP?(2l?~~yzX=L6+mi~F%!tDij{~^ZbdRgYdE+BLV=9elRh7rf(ofjF
zD;qz4ta({6Ucky4cY@nlkxQ%daBkdszBt{Sw^-QE6nd`dN&36ozx|e{AyJU2i6ZvS
zg~Yjz=QEN)yY?%vYBA*gwQ4E$or|;LVfnB{e*I<2u~B~{Mx^JknudATi_E4>*j??1
z9T3e8Pg@u3az!ZS%4nDtb#}Mti;cNN{#kr*Z?n!#G3d_^LcY@x)mOOlq(&UPzEZp-
zo(g<nLZk=Vl(;dBgAYrfQ{IQ4ws9~GmK`4|fBn7iI;dxM5s2uHz30J{M*n``r|r_h
z-W|n#Bcp1<-cF5jKErru^aCBDb!q<p!`NGZMZI-x<BBLCDy5X*(1;Au5+a?_B1ngn
zNOwqwfTY9#Qc5UDcY|~cEe+Bj9YYN;z_&+_=Q+>!o%jF#ugeSPaL$2W?7i1o_qx}8
zXa8JiOygW&B1N40?AfO5o}nN2%5<k+WW;+~DM<!oh@3mAB{;37`Aa;wW{;cpLFY?O
zSCjV~L?D{Pq`ii#mQkz#w@OY!yKOcNZUb<Bq22G$R{d)x1rfU>irDw6<2P~R^FDv0
z_c$9gL?lSAc<cwo5mMKWEW#+)@U{cgt9(lBEr=Qh00{H*Oy+I_<)F6iabc&amtsmV
zs1CV`BTP~3?%B)?4q+CALL3IFMv$1#^z<Y`xerweHfTpXFw+Im09%PyJ?XZJZfu*S
zI+9aF%SiOH9>`<ZMdAxc7DFmG{f6H4y{bD{doOiww=V{V)`MR@MysU-6YG3@p6`>K
z)*j_G%YGmxva~JsqX|(KOp|tgb}IEqW^3EBDY!%M5MjpMsHd#C$BV7D31HZi4u6t*
z0XcPo@pd+X1;gnVP2L<FmnGL&I<YLT)`CPr+_lYB(nT$c5rtOm&U>A}2<Lp-hR}b1
zEekf|y4l0gcb-0N$`xi%=jpm{ps}V(TX9xaAJ<e&yoqrZ;X$!-ia>Y1CP3(>h2jn>
z<btVBWxFzOH+cS(<L{$m3GqC~YsAeMB6DK<DY%k$x`d8$A&^ZiLrAeVc3g@$plBTE
zxvZbwY@@Ed%kjQ;yRYZIHzbbD#3)I;!3tu+V7+uo+=f@kZSc|Sr}<k<cSsX&v@OjU
zR?f9d4JUYd57(8O0jOsUWO~JeIA^5op<=F5oi1^PYV@7ojhXP;0y4r7X%_+X3&tr@
zDJn?L5su607nVK^9>NPOtjh^l^TiRj-T9I7qIGi~)>%@!EX;)-U{f1gM>sYCfjjmz
zPlRdPBh=GZwPK$iviBN=YFNoVd-O8-)pZQb!3w5#$_almtyRp|cfJ`VB$=jpLke#2
zmhheyx8b=kDyD3BW$hTYLs1g7-xjb2Kqt0f1wfgHaI7z9K{7SYc|!f)=d+fUh(=8f
z5}dqHhz=>Okzgv=Y=ANG=&!`*)pO<Nuwt_ebMw4Ojjr_f>yM0|nZJpJwwq*6S~4Nw
zISNX|CZAY1a0xv%^cZA&p6TubE0vSZR}4oz55&hwd^(G7v|H*hR$80y`&lMLD1COt
zK6hb)!~37!x_pflr9Mc@+r}4!f9uJ9|E#JL92wF%17paiM>2;fF``eyk77hwWf2*v
zpJzU$vf~m-xJr3@_Xo|)=kY1u7QZILt}K0n$N}wu>V>C2GVkkb0C%jV74GOYGWpP)
zYJ10n4x*M(mFs&Li8t6(7v&+?5XjeV^zwrk6+pNiBRe$(os9VMD;KU@YSC+HG_uwN
z4r_Ot;*Bdipx0Z|ccOjezCGK@TeaYR+mU;UEHXz3nk4szB;!-T>^1fpU4rngbux>J
zm3(~Wz+CcdqkTe7nv-l|urhmPp8QrfMvCGUS&1*fP=Xumt5Z?45vC4cYaG(#AUAt_
zt4MA}29_dm5~b>{JJw8bc6n%?rc0~|H5@Q;VmjMO-EBN0;HO&je_?+89BJ96egeAE
zA&kvJ(eAZfu7WO06<UMJ6$DpkZ&3iS0O8P5y!7dZe48F2ub<S2DHry6aiO%VCyR%A
zdLbP|<0L{LWq=D-b`tQjc3Lp>)Hv@&*u>!|!YcF}>plJ%-U8u?Pi4Bb5MWc)@?sIV
z6c<DU_EgVIFTEc{LYV_|58j<}o<Ai*wSJe>OS+A+6|7Ye6((+*2%W=s4LtvmYkr!w
z(uc%100?=g)aDUEG^fc&D{5^{^B)ohz)nron$FMmF$9;fJ=nD(Vc#!k%hg$CI(xe@
z3+vz>!AA`m)~ANLSQ=e^T;B)}3+a+Of8J()XlOhkwoDTZ-Rr_KmgKjpUj1C1R;p@H
ztB!B*QDgt%J3p?(gS8Gd++D{93GHd$_+LH6AVhH+7d=p+p~3)vk`sjUFyf)nd#G1S
zfmewZSZ==M(^DWN^}pec<KKGu+@_o&0%Eb3Nu;-rSjf_R0ms|7vmHSof_aMzp>b~o
z5CbqrUykK<<9huC2{qaL^x<GlIBo5#tadkbKBuzJpR@&fX7&C~`Qhl4|M>q(`6pq%
z4Yg~StL@ddSo<wj$I?iI)y+Q1qXEy_1{NZs|Bi`LE_{iBvk@e&LENsvi4LK`7G#VU
zGzlb)opBZFJj;IX7X2Ls?#&k(H==mEVS8bz561f*APwg=6@DcSA9t~bCZyhWLfh_|
z-7>g+_QG6#%(jn;Ll`9rAv;HIN;d>X2IBB(ZGGK1y#h_7(u6W18aYU#oy>?;MC(hi
zAI%?Au`Ds{6s`D_HF5=B6&VbQFGyOTy`+WsuI1avuQM31%e(#5;&@`w%qw(Q7*on}
zl{9^aBng~mt#Na_RXrc@+Qf#mpx<Kcu{ny;q$@`VQ3jVIjttRWPhZO@2Ost4Vi_D{
zCmB9V#PQVX;^D?7hIWco73tkOcp#9zNsmn5qVjFVxaF9h^cbk~Ke2X+F~{xF(F*gz
zuNt^FPUvmSS_O#qsFG0Q><mK*++2&#Pa3G9(TJcYg}l$rHXuzui}DW0di*3btIb!N
z$%Bny=d=+ZiWuj4&hZsg;K|nL<sE!go;=r_mK{f|*UPu1>Cp!sc%EQEOUq-;*AY~D
zuAzO!h)&IdN-ts`4VN$ets9lN%r11TKZYoCUoh(6XS2gHpx>uJvtof|@8u?|nFz=B
zqiE-+Lk5D<-@3H5s-a6F{8!*W%Rb$r*tG#(OZ!rCnO3XWm;kk_<tIoK;vL)K9tt6=
zh&i3j;znfh-u3gw^a@kH^)vvT*nmHak{^jL-w+~oS?xpIUh))}?bAORQc+~V@w;;g
z)?$EjH+LsfV`jlkce(-e_y<60@#0Zxd<p|)ZLZ<?hgqauP@JB6*Idq~UHECs1wI}j
ziQe7TvvEcHPBse8$Xl>0^FQG_sR72CNyuSD+9z!aOASWpW?(tBq34nU>>KMhzQ?P)
zh>xiRMJoJU$!wcnM0b9v*OTSo-H!U#jC(J4slu*!x)TJ=(ryZvNmmQ3iF5RZyJ>^1
zpm{G$vm*O(S?P-yp3#<z51Ce@G)UTuuxE@et5RAbLLIoJY-S<q@o}%Amqqood!NeJ
zLYqB7%u)KHnGT_Y3`1X#P83C66y{}G1;#Z4=P&D^UhAT7Qpae&Qi*eQs`(}8(YwdG
z_XqR$+If9tOxi?)$j1;IuwWY97WVq5{LpVUx0t!HuSD<eaYXSJx;2r8;xR@&TufzQ
zHP!H>-y%7T9;J`au!<+?$VfDw4>v@#z}rjN^#yVdC4S}K2u3Ayh>TE}YLHzx&@*Q4
zc_~w6{`vrq=1MLZ1PtPyN8mx!HA6DR$+oH76*ap<W*F7q8)O%i)(5-O*vXbu-kI;D
z8KVl1#LKmeu@AkBw4v=`&kj;zO*l6i1D0b05mz-HJX&q446qgBYky`PV`g5Bbn$i;
z&q(MT1f8y9c2W9U8O-K_Na44;_L;{0m3TCKZT5NeV528HlV)6HW)AHmrDca1QnacE
zv&F7Gik`S@SxV3;?^ruSgQpIYwRauLQL31f-D$58Lj>6rNBJ>AGV#l~uX@E|Vo-Zv
z+q?3RJI8SL-Rq-T(pzP9-#-ANx=i;ZtWGsj2dI!aaS~Ob&nE-931tycnHL=kWGw9^
z#1}hBqmzNQ1-XB1pQ^!zfo0l}y!}x7@TxGQ$I7w^Ww*=g;C}l{?1JV@sVvETngw9;
z{2LVe9Dxjq!4oT-Qtz~doHUvj)Rp7sa!<%jGYy#R&53B$fE}ahD%9S7JJ#RMKYeTG
zJeFPCyb|gwBzXQgiJ2@0+sq_b4~jrcE9LLuN6r@P#gy4SKp*im{W93knZ4)Q?+b~l
zpJYl!-1Ahri+Szx9o7!A4MfztxE(o%y(=7B3<^mr^56IDJcqKBbwj`5%AdO=bQ0e2
zyLJ=dTZg6*c$y6e8jw~Z4BuY1u~8>zX&3}s5Yv@9Zw}9vYE*xoA6aB<7Fl|E8{}om
zx5Ep;sGxNVd;xi|!PrpU6Jp$9@2&+8d5d)1MVB~31h9=y(`@5SxRX0+i7c%1;KDZb
z-V(hnm>Sf6Ga3v%xPEP@Z0BW6feS%vTWTu)TjRR~PD+8UI73nJbcIrfh;Y05iE2u9
zQx?2=v+Zj6aU-*W%KC(Lr!wCL$sD2>CRW%=#XJ#HDNe0=Gh{%IZ#2cxm#fS~NPSZr
zf3@RTg%Id^izdoHdDn2`hHt~^c0&VI+oC7-Ig-jJ)0Lq9YCCQ#(D!WBA{MRmE)(pE
zEy7XL!Ka>$z%EFd1n<XPs2VUSkC9?RUFu=1U(GJ!wb2w&ncJcED{q>S0me^ORGMwa
zV0VdHX^qIp+f1OU;!ki_^+T@G*Grqa2M%NWMge`w>6K(5GF?bZH8TMb(vvUc_^1Jr
z!MI&1p{77j)L}iC#R>7`eE*p%@D3S6NtBx1>H@ei2*GEaREa_ncM+(v%<X{I0<K2L
z_R~V$wrBOjyt%Gy`O^Fd&DAG3bLKj3+^!L@a59w!@RM|tB={4QGH(S4wcH%~&UK4=
zEB9D<zxDbR7Cuauo!2alkfWbGPt+){V6E|}IU$4maEzt{GU_6(n^~JZOLVqaq?X@_
zdr|lGvlWQquU!KhdE*4siocS$Py)vyK?{(F&C|TInOTI<Upi(X@p{pbA#kh?NQShS
zy|L%&(}_;zV}%BtS)7ocyP<E7?jIE2j|+Z%S}18ihETS?ct6TXKKR~PtH)k9A8(-^
z>`_U}Q**NdQWXQ+%E3L#UUusydxgg-73Tg3Y}YP+LEPP;njR2`tYtme4Z6bO$guhk
ze^Mh2$4_7so7cB9=C@&*={<{ebQq(DHXwJ}M|g%eyr$UhRX?oY)+TO*H1uTI50^rK
z|5{jqU+tHebpcE|<hp=yUedc?amIlVqgF{gX*6nq-qZ!>ZuymKH||%5`ndD96Ei=O
z3u2@ss&`LTAZ17`Qfi6Ng*MH$<KZ#n7G*r|WmlF{*L~ae1C*6S06(=j4`9*@`clQV
zvUCDFciHWF#%vo{_@1PnMdxns(fTN>XO))$MQ?y*9W;cd+)q=Ni>}~i`=&A@5@a^r
zwMcIno)XcP7kxqZct2AR>o^RobaVDW&F^GAS;yQc9%63sK(lz2lh-E7%m+AyJ#M*+
z)|ss`I}OH4B~zuC*bI+s3c%Cr)}2NU-`QC&SL@`>77VOmE;SkPST+KeN<JtrG<Pn7
zm$#?PIhjh{HJmBRHP?wIXuWvI6803wl3lBfqmg}=wD%?jFPn>$+}70#Btk%BM@l$%
zt1w1qd%Et-<ywe4;E~3XkRV>xCy8wY9|)b`wCwJuqffI7QlB?=+E!H+z%-eJ;e*)8
zVEKM4V4yyX!`zNIxSjjZ_R0MIFJKXBEEfbZ6o9Ih?k~BvCaz@s3i-l1n6Fo9t3+sV
z&L7|`ef1&8?VyUpIb+iDC*4=Jos-veKe+x#R9mBk5PK`(M!p)fVYlt>DZ6l`^Vl+b
zHs$<RXSfkf!#bG?m4zs_)4}iyDRn5V7q%q+x3bT(5$s$`<5B6JJBhU*0%Z7{>R%3=
z*{oFotisJS;JVnw>3fC7Tgf!q%*yOpP1G!Fft#xQ4=edb2!wFYy+sI9!R~r{XammV
z2j7O*{(+!(o2e+}P9~gt@V-3d$(xg-q}p|hRaW^9GKA*#Bam!ipRXA!KinKXGze?r
zSWKTjr!6q$%AkdpfB0Doetd*mY#E+;UU>t|IX`G(^+MCFy!p7^<Mbpps5nMvqAX15
z9<>e{_)Bp^1Aj-B*{lpVwnbeoVjNtrpMO1$_c`6O%JRB656CE$OdKD$>c!%*gDKEv
zvKZ}Lm^W7p4moeq0UPdc)5zU+Fw&|jZxdifh`||&$L#XY-h{r8OES*j6i^3OVwVE+
z`p=wmTFRuS91{s;4Zf)wUAE;nSRvRhLPUFfTDloFg7t<(O}z{$$?_C6Oe&vwPha<p
zFP76PBqE&-Ri6pCbk=)VZf;~|af-y3bS|dt9f}T(w^U35UGNZ8vk#8V6yVLONSp2`
zPEO$2R)G8;mgELt$mBYArd`E7Dvu1#%36)LF5?q~X}!5?yj@MXms%86>t=KveKG3+
zC#?QfLZ%dt<Jxmm)@q$pK@a!3LK4c}mL%i7foVq?S%9aK@mM8s%E|?x#^IOrjyB-i
zzNx{+$v6?$+ZmD^9~nI3RhaBdSIB6cBslw;H<_n=Z&S5`=?ir8&M1%lk8&IuQ(~?u
z##HzD4Y*u!;u?9LLZoy_{Cf9Vpf%%4jxT2pQ@wQ=y?&C00?xF=<Y6umT?Y55$eX$4
zZvNyFoVqlDFO){vlP0ZW397TArKPN)dN#ZEyLS3hm)=Z)r|{EauxmR}y?-3;Wh&c#
zys#5dBv)(U=arc!Pr<MCax7wegBDv4OFB+N#Et=~Sa<tY=uG#T{@U@ZjOK=MIi%!j
zLn%vN2o8w{e|J|ZCqY_Cr|g9(jkqa%G*gE`=y%6SjR75Qo^^MHUdR(~Z%4^wSn~o-
zbCO3;j6f`W7szTYBPm}+D{ldom+NO<`P^+l{q*8Z6t*wackg|SUGi<c+fBE%a`VcQ
zcI)TUIztOzI|G9Wl7pvF$cLVIB|FO&+?P!un3a2dLPDNjjb<1i$sUV~a_-yRqS^zl
ziO&xk@_M3V$DQd-R{P^jWet7Pf^r1(-Sug%nTlQ!S}wvPe^vl(r^f94fTvZ~w7cc)
zMtPA(Igh8+mQx$S)~JUA$7g`U>{`dM2<Zou!fQ)7aC^f&rW40oEM<5Rq8jYsvi2qh
z;}I8fq|4paI?9o?YpUFiH6w1S@W(fZX!wXCPKv$+(}y791-!omZzkcKPoaJ01d>;g
zfHCSkm0CKv=+#;&N4g{#N;+pDV=kn?z}9zllVEich^jN|&S*03xk;9Q$kl+3T>4~v
zVs&Ryx_z$%SPr&c>?OsS*X(`$1sx3PCLEkM@%p<)9@>OGySH0wN|xHXkZmetUN@h@
za?t9}#G`v_^NziS$!6ro>G42(Pl|IuL9CJ>)}V7aWhZwGx0RdPRR}&3stx)2ykg^u
z{>9dJHQoEQoVaEkB^T`Qsa`?P0}ma6WXqz|OaW;KR33Y+`#~k=0-bR*3gD~!r_!zC
z*CD-vQ4j9sPZJ%!&rOBY0Y72XlqJjVj$q8A1h<!vi08F{wZ(d{K)j|<o3C0c&G=aG
z^7@T`K<z=7hzo!xBUM92zC4ncg?&_f++r=a5K3j-o{GbB_JR+pqF$hqoE4|NpdII&
z{3<EQST=ag>5*rcjpMZAV|ApVE6;>dZpLoxXYF=sVOM3VPmtUZgc$7-o7G&w`@zi8
z-$fl4iI3)Khe`2{l*~0|4m2qWvTdLomsu$Y8i~ZX<hb(Ct6Bx{*OX>p(sO5In^}X#
zNX`iI{y0-~lh!vXW7(kfGu<G$z{Zuk4FU%A%m)lM!4`zosw^nwbM=Yr(lkt{A9v+Q
z@s8<y&&t=F{IAmZ@W4muq#(^R&YsgOe(=1r2|}=}Tp)+N_?_3F$$yCJ<yitpJBDLf
zK5>W!7!HVKRfHiwpU2n_2|-PpOSoB31_r?barU)2>%%!OO2NeJ?W@$A>JE!;o|C1E
z50=U@&Ed&?>tr{He;xQ9=mY;7M@k^{OzPh61!KC}jQILO<Fs((H9FJjeY`C&BlV(q
zio7exGwXfoUhWbBqVo|P3B9NI<kWSs8PlYGQ=vv^NxDW6GB!bczA{6K1+&k%no|j7
z@&dZ)w|&bb6U{I-owNpdWL+->9N8C60c!cPpE7eDr&4YDG0fZ1TFqB+y>h9Z<&@xw
zpyE$~j`>$)c%t*h{;>KeY+W0((4Us^SLuMR9>g(Wpsi;;?+T_Of{~8_2{-Y+b22S+
zZO|<sCgYFt77UqHbWB?qO{`rPZ)ZfP+<IQ|*i&X%r^0pISg%16+V}KvH6X-En)u``
z_^pZav8^)?0G_7O3RVxHg~O1tl;eE>N;ziDKYNL;E<cLl>vKFf4pjH07mGeOBcr+D
zD+J3m73!Ibuz5$n1P_p@$=4K`cL(HXRK_G24^fKQdc7e1mUEryy~cVF7=*xS$W75;
zl)EPL#H8wBzU(QlzXOMX$;i;N070{3>1ltqP(6H9VV~Yq!qQX+v{Aw&VM39Ep^sC;
zTumRvVHxzHJ535@2Z3PlFtGKUsL7x(859`O6Kdt!^lob;V+G~L8<_~C-e$KE8l$n{
zgnu*U!+&Tl*gbD&l%p&TjqP|7ZKO+JZi_c4;&ue)gQD0RX+X}no2v0dfkhD`Djc*?
zr@m0jf0L>9n`KSF%5U#e!{0IP4g6$gP&~rC)m&U^!-2~#aWv~0ldVDXuG(SsS&=Tm
zqw{u5ZpS<9dQ+)qT&6kNNG^29ppV|uZf>d7INEDC4T>|()YdlJtyq-OE~K;a)S|hR
zCLi>;>Dv_%=05cgV?=7&iZwAKBlwU5U%NU0T;vckw}0;+^aN+uJ?LyKDkl@q+!|&(
z@*4_u(Oet<LJ{R*%W<zMW*xms!v+r-*Ns^=V>J#!)koqyC)BK&9-GV_qkTna#UH!Y
zNHRqj)vr;e(l2B$#TR#j#C!(LQs4FATOy{g{jRH+%+JliTw3oXeU}xh@l)O){=);U
z67rt~>^bHV=O(fI==tyK9_m>Up-d}6S$gG!(w)2^rEKR0Nx>@uDQZ@cQi)`PO~j2e
z{h!2SCv1Dm%HhxMW!!}Ae@E@stbz=9uXbg0u8kCfnrFeTtF(+lw8?J4WJ@N2qkis4
zYVf@p?3yW5mXg+DO8M4eYwpboH^KR%Vb2A0#cbOp%0%!NEkhd{c#1q6VD&nNH$hk<
zm-vY<>rS`9BR%RD^GCB4hGg??^WilLTnfF#nc|L*?>0zQ@^4q4fmy5A(o}aH=%_6=
z6;dpVY#R%%A!1glz56o!oE#T|ZZXaL<lPL=Nzc?-{r$^cSzwfBRHx1wL@}bqiwg<V
z3Y%=l81mX|Eo74<)H_xN(gX1#zR_)2$gpkq8I1w`azb|Mgr9WxqLs>9)E&S;#xag!
zE=S;gu`x%<J(ApxF^(hMZ!atbfML-tQ8spnzN{W4!+46@8WZch2JmO5(Tb|Luff(R
zD%H^r9HWWG33VihidwE(Pq$Kq(s;>rFjD%`yI7Zxo21A~UorMOz+*@xD3<@gmA&$8
z@Ee_VVYOo~-%poMBDJ-m@xa!2ND=iuWzon+Im7WD7?Wiz&U~H@+WE|Bw;J)4<5&vr
zju*%2`r}+_iM0tD&53LiKp4nQxfKD7GM{{$qb~p<e#r$)MV)jy@J>cKoHr8=it_I{
z1eqmt0GgsDt=7Zp$ZPRzRiWPb=A-!o;IJY`yH>1>sAtm20}_aCDQGL7o-Nv=04YGC
z{C1(2=Ph7T&)O~fu0N=B+U*Wu3mpg1X*sH>0c+J%AS4UMaWi_^;``d2?p;QqxozY^
z#L?<!T}+|!p@j_f7ze7fPkCuuHz0xnz7j5iRd4ZXz~n;sD_}Z}*bRpWO$)V~*#@|)
z*3cJZK#DMZh1&M9KUv}!v(_&0Ekz)N#lwVWfjN|1SPH*_hxZ)=0+j#HB$mx#@E+@w
zV%V@n_nb6R1_U&WR-5$)v4J$>(4?3DdWm2<$mPpzx_@{7vp4(F>(KL#Eu<|lv!jub
zpy<N~rOEqPXf$S%Y~kI=?|fqTv0zBMGTbxe;N4No{?DLY(ZoL9Ek&;r6(iY<6ZWfT
zVs^DCpaw+TW)f<@K>WpTb{qIiAJJ?_@@$FdwUZ_YQqJ8!OJ~8^78?A-s%?^BT&vEt
z4+$l@07D>1_M3(hDXSUq36sIpLDK-iySNiW3pNzxODDTVZp9R%TKDEs=KQo<Z1#Tw
zyC8>B5U|BV<t;R<`Ut!*!}2!qZJq@dUMHWp-gO6_4aODAE?nx~9BpNH>RU8zV|ZSi
z0UrGQ(3+}qbgS9*rJQIAf)p>4K{Vy2tgvsxTNZM9xxbqIe|P<Vb-QTsXP2PBoqoK}
zpOW>)_)k3Zx=xwN@JPl?<=+@TJ2q8DlP{KsEVtL$$pQ0IUfFnt0<lUWXZ6~kz{Uw|
zqBvJROgrGrV?OX?{z%K~>|niqSn{srVgMg>`$eAEP$}TmobI#nl>@r9D=yVyq67PJ
zS36eF@qBn^xsf2RWY8UoXq!$pxiJPx3;}4cex6^EScTP(+Od#efx&OP$<EUWwcU!>
zIO~J<#H8P<CeGr=5LgSyWo9}zJC<1yj2dKy?9;A^?LPv*!|^GNMRSv$y3_m8mi0T)
z6RyiwWM}61L|lAqZoy(&fhq42<Jg<S4GsM$cZJtIZ#Ymui;pZ1xcBMam(r~wfq%^d
zJZaBtGgSOH#p}GmSnJ%o#V;WtqmAsgAxd;)3_V}9>kb}Fg#gQ0-SyPVT~PwcZBvQC
zCM~gO*GFoWPPYKnP9uIer3zXC!T~iB;K5Kpz+}&I;E(f$p-YS(c*ozRuDxzcbWOxS
zP3nBUZyS($Nt1!iRSMQ7hUL|fn5;d2^UjxaX^?!pKH5I+O-kP@g9;P^%wT%JB)Byt
z*LCg$+V@HqT6}2WBlf1`#%fl@1@<T!0|MSj%fo@j*)PI|4u1~o?tZ#tY5Xd61$R{4
z@a7Ca1t)!01f~^7HJxIcc%HeoeAnnhLOnfLPF+WsTY4KGP@MH%XU<i6RlqZ*e}z>|
z%0Xegt(`lU%U2JXeh;%fkfPCR_M1=RWXOb{kO%V!O*1>>K#n`aR~_$j!R!^>H4Rcm
zLRj^SUSl&(g2DFFrpV8WFt%NGo2pWy-GT~%rYC;05r(<BXs)}bK-+5?HlxsvcE%t&
ztx5TKw#EwFP-=YTmn=~;*cn%0?lz6j=9VZ7KlhKY9ezp1F{I7ko@wx6NT*)fV!mi&
zWV<56MPHosR|n&F%KM+M!pWH3)nSPo#!owkAIYq3WWEe;5cq51f7HZ4yQp(AldE3k
z5qZbSE<>HG(dYO`<T<ZNI|FO7Jlh>qcb9)j<}%x@*)zGZ1&g$D8^QHkyVIaz>1BYZ
zq;uHDXn7tlsr0EL=yXri6QvK_SL|~fwatmVdSN>DCLUc0RPFC?oy$R&Equ%99uT>?
zD458n+w?MwB?Z@k^0?Q|dB&rH*22bwr6&&zV9gc2GMCLd;Zl+W{m^hu_`O~R#$@?@
zej@fe0ZZNv5qEid69VFz_!7Mj%-BM>t_me$Iz76q%Ug~kYTA_6WUKqwB(pogm_v$^
zTh^%;dXq14(WyKsHC>=1HTW}XbN=8fH$p_G0VIxqn`NH?hc5FG{W#tqTl~ryBGwDt
zLY6Ac++VMyFgC@wB&89r-<wc&&(v4hG`jNKm1{1QGsK}6^D}(~T@sGSJJhBhfgXvk
zX%ekxOc{=}JBnJ)UCX@gogTa%hbS3m&HY|h!w_Tw^KfBReuO#2HpQ{4f<DgODQYg*
z58)1mBa$>a?HJtlmt+z?>@+6aAAbS_6VEv+Co<|v=<%Zmh8()>;d^lV&`Fntuw)#F
z2j!#7{v}Htlzk2Ig^x?p7S`3cdC)P=eNseF?%Nz6BDbG-(r(BYUi6_->$=J|=Lg0R
zsE0a-*JdY2MnJWO`2|7#&B@%pS3zH3mMI?Ubq8PHPn-)(GLL<bo=HG<HV}Pe;do!+
z)wcS5bK#rjyRXZIVbAZ;rP$>=yQ|qRUpO}iv`aE@*rRp)w}BbOx}5sbsmLj5%h3V@
z|ItD%va&6g^f6HiNt9x$#Q4Je&-t^*hHoaMC=6i{f+rmjiM-_zBq)y_nxc|nxuoD$
zcfD6M-0ksAxkG$J9t5RnIbV7f$7h{=JjOKHtxk4(cxD9bXS8-1oQhKr*tQ%DnSJ#@
zy2meW=2?D5Lms7LBUJdrmd5azX9@egvIra$DWl~EM3L4q4CnR_AO^O&errL_vt%p~
z+6&cb_j}zyL;&(JqC4fqzdIPeTGhWB*8kVn9^r4W;r(c%^4t%P*K<ogP|J4t{bzqU
zyuu()bzGS@|Jo8qgyL3Zh)qyX1ffMi?<(dW9@&3|mVbPc|JLQ)b_%zZPgeUY1Jm=Q
z#0~!`h{69e5#ilty4kMl{E(iiQICn}m<w_l=lnx~TrBM$Ztw3@^8fkziW~!4M5d={
zF;QyfiIW55+Tmdoe1k<s%wBf?ygN%V;f0d&x!jr|NpIs|g4Dy!$x+tqCxFfeX}rPN
zJ&P_2w2ZE^VBAyyw{(q_ieB;GqUAqtiPL8WC?{+Iy6a)Ivkekx-%v((S8A9fv?kj$
z;cS64XSe<2@5&Ed{GDKO37UJfRf}0-g0>!31?Z*LI8VE_dqMp>Slw@zc5}ch!Q5jt
zAT!mJOW)DNB##G;+MC+EPe1(UHT>s{`vs~0{znHV3}8uDR0Ci;T$%>D2mq0H=Yauk
z8EeXrfBa{qgqOqmG;wm%dc<eJtfigGXE~-|LtuyM05r^;(I@0bS|YGmXyZ}_AO5LX
z|L67c&j<Y1=YBJ6>77xtcba5({-1aC7kB9Y^ZP3u=#D{0=9kj{T(>YTX4-u_R1aJ8
zjq`$l*_&^wH4p=Z-mz>19nv2``e)K9)iyu0^e<QQ-<N(<4*hB$Qqar)^=kP!eYP7e
zrg4Wc{D5D;Ko8)`%@vwefbrM#Q;_F9Meu-K07I%S>;m+s8qEPeSpDA40(#)X+EH4h
z2ocZueA;E&6wgf_$mKHbR^thRRAuzvS^wLz#rIie04v=-uE6fk3ZaC312eX(pB0id
z3!hd1IiMOe_pf>R%Y?rW0Bo#Q`af-f;roIHI`zbA(5C(Y^tjY4xdLoJ+VS==kXiqc
z{Uv1$J-h4crL~ddj_5>7RGh0(@<A*czm{!Z_G*da?BP-|pZVI{T2L-&DPKESZkkA@
zC?z#5<t?M-XRPrzSp7e}A|Div^iG_RO6&bUizD)i59nq#wHE_M^NA&3xc=yPPir4k
zRg~@ms);G~c8UdXc$ug6qdD{P2^_6wr!yC56%nbPU?E_EQ8&9CqoQDD-#C+{oIV9Q
zCy(6)U%#F9CXq`K7)L9yS#?BG-;A$5l#iMqFXP@NA4{<`5ul`WgxIoU?D<^5s<!%3
z?Kos{|HuFG`uqFA!ZKf(SDycq1?&+D`+qOc6LLnce_d?Y9!SH+LQ2=}pmg4BmTdvq
z+#R(mlPVj-(BqY?Sc2UiG#9Zoost_bFV?DzTj;$d%(N7&M9qV`vV~R-R*W+suGe`q
zm**kW%KGQ|{Pzm{`x=wqgN>bBG!$$4*ZH<`36rkrI6JcV-3#)}tVDW9SX0S}!wph4
zARgS!Nr=kyR?*rciZOAO1pkgnwM@*#Al3;CJHMYT5Q~8kGX45}{o^?L3C?fHOgj9j
z%_%er`eIykp#O3ul;~rHvT`TJ;jiD0jp>Qnl{y0F+30R@k`M*!mgxsGhKAONvr61i
zOra()2i`r`Z(a4UB}7V~H_h7#e?Eu)W+(4_`|vxPd9x(jHZl)E8P3%?JQ76PF-*Hd
zs`8OGUIU$2Z|gP$5B?<jVFoiYaR2Q>spXUT*hP<^xYCN43m09{2~lkyDHiC$v00A0
z|Gu5|8fQCdPTw1HjIOh;AJ8ICg%4SWE{88T3bs(1UjDLZPc$EFeMKwZ7}8$mV=rJ~
zms<MP9OD9?Atk~loaNzE^d@=}wR%D~6PG*0Dj#mTr+BhSFEb?^K0U%+604urzArW3
z7;*MmN6O5+Lov&b+IndX?*?{a8>4pORQWI<#VxGjx?;PA2<Z*U_Kwa^Jb+NwWrR7{
zbweJOGA(kRhhFo7besoNce6Y449(#?Ctal=FUzAoqwWD}Jk<cWj5`I=Icq@P154pf
z)lM#n>)PK@<Kt_LP}U~Ww25624!btS{U3nd@hK7Pa~&{XaK@WHVM}6$1U2Okuk$?!
zkL3hc#n{V`WA;gCNq1DO>i9yg+5Njk3jY!Tgc+&9al**-%2?%h(dLS?lZFE07Y-!4
zE}u?8^I;XJQnK(|p!RC+J#ni<yP~7@yX;dvPqHTgpC^QB%JXz{3*hA}NS4xm(JfGQ
zI{*z1t;TJjCS-@6MGzrv75|1Ff>?lJ7AL!X$B&?`(#ymC3FJ(Eok6LE-@f((Er=EM
zHm8lTqAlR#wbMk{f`vt4<2?!7NgiBhh%ZHJolF6~<`xh=zj_4l*~)_lMW=tX0Im|N
ze`5ja1mX0`z`9y%G*dQ?#$X7{P%om@Li@Z0*ktU$4>+N}8SPpPFr_K9FYAu3X|m(f
zu1p(**W-hr7xy^_A<pQ==_KGUFX+;rx)#exPOBZ0a+-vmwhNulSvSmho>rkHIJQ6x
zM6^Q*p131zq!Nm{cY)-ByDQDx^sPXg2e5j|CvFbZxS&B*Tp;yA$1yhI6vX(Wb|CbY
zwmw-=@Ob5;-xN@&R8@;_@u%pchpV8?E<C-$`Ic>%^Gc&8ZJKZX01s5>SwT}z19BeM
zxr^ybCD2vqeM5~+&S#g=Fp6$|<yX#MPi+7yGK1(MbPLQXJAi|z67RH&xu7!SVRgrs
zgKe*in$%Dr_3}e>NMf?l5345jg>F?x0r#zr_L=iT9_JqJ(Mm9bJzn&j+wu++H2EN-
zHO||BUTtb+em??e<n^MD(ttGJEfF#41>8N`4Ss?&(*cAut52ILjw2dBti4XPDu8VC
zTPpg=ayQAUqKOZw49$s|d7tFS{Q8&OfhKFXQ~dc`)5oi51lt)X-n?+{jUHW5{}W6f
z`|K29HZYU^+|^2HK+iW_5a$Kp(LC{Npy~AjEn>=r7NExPgYwNrkrfyNvD|GY8)P%?
zx_zOm51fX#zzEQxuaKuT`fARdm1S2uKfHy_kGlR#0=RM;b0fg#@6t~gi$R}tn6@OI
zC_#ypqfx=ADftx``U|giTqjQO#!-3si_;$^Mg);yFxmSl-$9bl&)~KY#`CV}31F=Z
z08v4RZOWT?Y}Vqq6A1&}ES{V61g!$++ZX4o{ELxdF9la?7%W26bR1PvC5diIfCU=T
z$K$%}Mr7zp%+L~IK}cJS^KK}}VvxGnSdo_KxkL!GSQ?lcJ(iH7uoW)_&tDP^=@`Lc
z(*S9N)or`(=v{t=Ow=nN>ayovSq@;F_c*sgn*6U2(VpzZ;K*4%qf2^FgC4BihIjZY
z8#mJ_#@lqw1eWedRm=6mZm8KMk-(#ZdeZX{>=t8%LDg3lfbh`h=vAU&3G>_{NpVWH
zIa4DVrarIJ&58&BaPa9l(MpEirvsF>?Qplljn(zbyJJT2KVL-~3;^{ovIJo^j>}nI
z(=Kgt_~Z#nTxC`XOQ9=3=Ool5)x+t+G}><9qlh03(e&{f`zMYE<EE)BQ$qD(XEb#&
zhAz7DN<<0XeWT`s;^$~b*whz9W8y%6Cm~|Ke9~&qY(cYRYHp8gP2BDcfI3^OCm)<L
z=<b^>X>bfgc(tJ|YFFYKP{wE_DscwiPz8|gboc|~!<mFh@{jQ!zmJ-)(Hl!A8qstS
z>Y8-2bBMe)nzz1yew*|!pr4z+XL%56r-+Wn*0A!oQ}M#zMrGP?h7?nN8a-}iKRW=U
ze<-3$XC!}enyh!;9#Dd0z&bp4W;8TA#qNpN%sH;3Xt=x}d3wM9LXgYW(*dmZ(_Q)o
zQ<qd5Ad}bb_hm~liFY7^{kW&wTCOb;RZgvgRsl1D-2JReBCsbd79^|nZws{>vk*!B
zC&Q!Z7s1)#pkyBf7z_S*+&O)f@QxqVefb@!Sun=dAM&vqq4E_F?64QXW2@<aib<HI
zmE@Yu-ZRVBY}BhfL~E3gkWpE1tcx~IS`(0;p?(S-MbnryJrLJG4<AAe-X~?CyRPYg
zEY^oxj<0H$<RIrl$Ee*+uHhfa_K?HK*RL}Mu+X7o5^J_kXp)S?aUWLdtNJ$B=n%uG
z!xzh0_6%STJ#_R3nTD=*T{0jOjX^69aI8lN1_!}mny;By!=HX0hYlR?h-f>4J;vZC
zZeq)#2~<()cz3|KBrD&=w<~xqbGS4lB{$R^Qn=8lWKj3@FN$Kt(|4|xUn%*v{M{ZL
zP^Ul1mFLqV1U0cG-3Z=aYt)<b-()Kayp&r<(dWamy}Dkn@dE1l)`G166_DRN+9Y_n
zh~y+1%q?GLBpf2c95c!0__$lEpLARHgM@6~P_77Z6Kv894rm`DmOYAeAF}gAGUO6n
z!8sG~azyP<DMv-?12*DN2nwlVJ+he&Y{djQa;E7UgtV%Az}d9f{@&z${j1uZZBkF>
z_+iAmCU?&n#9e>qD;d;V-Nls=r$EIb<W1a$sl1qsKjL5N86SnG=!#f#r>p!o&&D3#
z<z)2>F6*%CamsR`_}kcOyIXMmpvA$Kr`m?Hx7-?~r!X`H@&yL*BUm>^X_3volfY>X
zKkfd+M6sBzzdrm2@v*&kEIsCR_5GYhi{s~gcN<j)sZE;RX-^q&-~4pv=vFT*@!;qL
zu6IsY`@m{a)?bAcZkJ#NcgJjm5K}Jn!d@S45UjRRE`=Y|X5i9LKvi4eY-GK#Ylp-v
zAYMhJsbGr{XuTZ%8WoR6P2HrU<y*OOvM7R&A|MW-h@F?yot*X+O1Zjr`(?dd_#+Fj
z9gapwEY13(^k18Mii#%6y>SF&>)WK%@fo26$kx^<Y~i9o%`kKHyTEEpo%nG7_2+U)
z-v*_cvcF0~9~AbEk0#0x?`8hM5r?6;psv=dPXRPqZ!73*L#2>em?8#zfge+CS3oJD
z;x{(<?c5o>9FJHx!3eGlNvPt#Z8U#$+2|jT5B*&>a^d*^f=wGM?z>iwTtRjm?+$Ec
zJM}lB!ea4UA$mTdg$QTRD!qdv!P|%Hib*DS-5oUD=AzOp@0s0yx8sFXrl23Gked~2
zzj!so8ki7oKlRlg#(N8my<YLuOI=XEGd;sFTN!40Qw@Kvi6c`khyoP!0ke$=5xe6>
zQSTOoH^DY*#tG7%bA>E>1ejjXH+P+(J+iKJamFSILr=j24TDVq=^JzFr3g1-3OI@(
zKx0nZ4$h&jKBUi4kc8_iBwj19P*!<Jj55%%1)0GEGs7rXm)%y9GjZBJmBH-1<6!Jf
zen<L;ubym{es|YtPFpWE`|d8hM&qjji_C8;TLR?2=*%CtZ0Ra26Y$EoWlogoeqk$r
zrH{hq_*vlb!?uns?Yi09GZ+GAsGqp)@8n=*Tx0748i1W%Nh-Z=`>HQ=mbPs{wEQgj
zUW)t~E$qv`U=a?Y=%#a%=nL3yO7n-1OTJg2ywv-`Hf<pnAfhvp3%C@LTn1%E`yC1K
z{%>6nj3_)o<VWmV@O^!c)&1{^tO&(>i|^kWC4py8MIoa(CfkOS3TT+4221z@YO-F?
zWK5zKDO-{j9q8P<8bD`48DSkQe$e+Jm>UKPHpC63^PXDoUWkeo7vA`F&k?l}yy`Wt
zI+MZlaeXZqY5gIpv_`&^n#&w;sVEs3Ik)MdJvmz3L!A9)3zAwx=lffwHkpuONkOsV
zFhpLP0f8-Wb&$FF6&yNH?E5?uuV}rz3$0^S)pmVNa*L_X+pf;tfwAP+S{{eK-OmDV
zM7(fk^oFH;mb2y{&ybzv?`I+Y(Gh#-<jpZU7B=vpMU;zi$T^IHNcLNWUcSRxPA`<_
z`N6fNTH1#d6m<PNbe@PUg6sAVC9J=z#s7Th4+aJJXfUpx#_xFc>K;aI<j>Y8zBhA+
zL3-#k@JyCvNx~bDg}nSTOYB%TKjG)1Zr**Aiyyud<X`CicBuxYJ;Ft|P~eC`8)AbQ
z;(^UT@KkLNOhwP!XR2;0hRo*|ViFQh;`4z?N&t>{_QBp906t~HkK%U1IXNl6vSMJu
zXjd(jP*;1^3#g~PNq8)ObPINOAx#5@l-SS+TOc#C+W?96gS)lq*U#0hV6{bbPTW>t
z0fnQ5#EV59ztOr95A3a{Lt+|$pci&d(2j)Cc+QK0S7;uJWe>Wqa+d14oKOs1U>q8e
z`|_!L`K6w7)Y+7S=+Y4#lDSXP^sgG8N)e;%T(=%bv$<cSlses0{0_p^db2x7PHnfF
z&fWKVNI*BtGhm^V68aMY<WbGOqzMd0k{1d5^=kgE%)p0_M1*LZl(@S1cg@@{fLTaR
z0&S<aZ;cWFFx?kDlNfqeP&;fIMWz|%qIT^BYp2l7`nuLE6$QS```QSqKCYI9eDtH8
z#WR12&iO)M_zSDDlX^huu5OHQ7eY)_MBMSLLmdxIBSA&M)o01UwEM(TwkLSH?Sr~B
zNtO6cZm&OY@8%wC&j_vks5x>_oHW$X?7GXG$ikOfOi(_EsV#K}H6Xcj0}ZIM80_@T
zqT9He;9=4m43Ef3aR8A4A+pCn_bcephw430BuO*};vEr7UmO&8Nq|(N(E2<4fZ)5~
zshOf>j|0tXb2?<A?%mQqc}kef$B`L0=s#L;O<I(wc6WnPVWT}BT9@bml+?UXRXaXf
z2zOpxv&`5u(3|6<Ahs;|{OLXi<-ZoDjOfC&;8Iup?||6O@Qou4R8y+0a#Nr{I&Vb*
zRRA@1?LtJmtPAwjCTI;2=<Rrcpn0bN&Uth9?l8a4dx;(J{C~PO%E`j;Wgr&#E}2zg
zxA^Ez#``>puPz(NG=m)SNz>p<_2sxFuN+74C>e)!X((D=tKXDkc18i^>h0Dd?A-aS
z`qQnwdFI>;W$P}!+)~TRxwoEXTXx_PK0s<21l)NbG4u6jA%9?2c~qGaX&C6b2!)L5
zd3ngVfpsX=WM>wH%F>eiiU@qOni=PwEcU#1MAKXXdZv}9qQ#jUGx1@=w*m0=ZTXcp
ze=8P%C$gyU>uFNh%b{K0)#R{7+KMQi06CT@v{rAjP<woBRP!@uzK}zW?@Fun{D0oU
ze>OcBEDJ-!MR_f;dcR_y&u5y#iPt$2fS{z^0zcvsA%G>TKO;7f{@~4kbA{OpAV!M&
zppZn;TPgCdW*9|AE!2L;w^D**UUQY&L1T+w?5Gr)3KzEkWT;8F4C>7w_1Dftt;8XP
zp|K_eA?ro2y_dxNwatJ#MgCqiRmK(bU?726hUWxAJ(5G+FpS%syzr8GuWc$4;(L!l
zM_rSBtIzX;?Atlg-RrVX82e&t1aaY|yt_H7axrq|EXQcit0yIBr;Pi2UO<(j?gF^t
zXf1lp2Q!{z=2S^d0ck$3y*Z5!2|sTrL;J`vo}7<KwFFC41GEBJP1KOW1|<m7kL02U
z{M>W}>fR4#pA+1Fgz7;KuY`QV&w1a-Y9Tgerv)3HA7A#=CUT~T9<gpXE2j(Fwb3}V
zuGxH)+79aGFXb<=8H=pO^0as^C}>{E@&m9|JS)1CyA%(&qqxn{!6aV*Gr26;xQCvr
zGflxQ3R#a0a>`BICbcXRv&$Kjvo|7*d&umQJPyaRY0M@>CKM*_{*9XYoZt*0-XzpK
zr!6-O@P=V}xkF!c?heQhH|k|_J+Ch4{d-OPXQK$4(t`bA%51yz;x}-UjZJl^|Iz~#
z`(Y`r%Va`6z2f{{pqSEd0836%r^e~PHv`P<l*J$EFIj^``e^>+YZy(><D1m)hZXeo
z*>9|y0|ANv5*_Mq*rGM~Bp6I6&^BJiW_HJ}k4pOj+6$74Z$KImAsI$~hq?<F+Hkyl
zx6ir%qlm6}2@p^5&k8BGuHAjhipL;>P7j;-#MRW#e~o;!Kk38;u25(#m4+<<g|H%R
zw+=@26z`}+zO}I&d)axJ*>8^5B6?&5@iGC{K&6&L=M<{<!dg&0iXZ#n9B2d)Yh?X3
z9Fi3d7(EmRApW|{X-P{3Va-vE?3yGyj=Tb()84!c?2C?d-GO^@T0RY?ynPJgykeKP
z@tf=rk&Pyf$swW^Pr!*Wf9*=+j=OYc4a0Y8mZUsLjk${5u+0T-4T_;){g5$+j0Ld1
z--Ywe2L-u#fI-pc?-;&Up<6_IzSOy5<Sx2J*RDya11=8Jf#7TNcSZg820nN5E`$yM
zl?_eaPCDp0iu#d){cH2jD#`$T4XK?Ve<>#10OUBu)q8GwF!h}R0EbWqNOleYeIjzl
z?P(7;>Q{d^jy{@T44>=XBEW+PPDR{h4&Ksu;>~U}VSQKp6qRBk;r-MqI}`Q}uyQJf
zs5HJul4t*GREkDNrD)>XVwT?+F9l}#l$~sY1ug4vk?v@TsbT}yec}}LaFT}Ux;Jrm
z`ZJWrfMXU)_8g|W7-+6Vd$UDn6SEC~irafKy3I4*m#>H-P{iv_a}<*RZle+<g*jlS
z7D@|~`b%Ml;;$Hx-4qfx2#Q7PeJ*AO7>cX-Qv$8+#t2IEK!Z|l5=e$Ue>UUiX9~5L
zq6rSl&x1PE`71Ncn~yV=0t7UO>AP>eCO{c?Vu?mY3YEtg*aUSoRRN^_+ICRzm)Qu_
zud}|hc(&vROLQeJRe(45rVT9vTanggA5}_9nvbW|#Wci>Svkan$vzcrw*h+o{#jf-
zXhWlb(g3c{md;bQ?WJ&BPq}`y5Q5-C%(u|XA*})_SLf5MN>J-yJVLFKVVlywKz*!b
zo;o_<NB0MatS9wbx|<Y+1L;XBS}JnaY#F8sX~>+xfvkXzamUrF3M+Xa&3g&n^=K0M
zJtqMrmu()`Ua}ChjkIngWQ`sJ3Cq1Aub6Q7cb9Vz`}ye`K5euN^~uaJYd!Mi&Vz9T
z^=-&)2BD9*m&pqDK0V`WcS$|>0&_aqA_!@**DZ+Ke~^g()Q>(2()!mB7J6Hqe;*MJ
z!Wax2Hsw=UoCASqRhwj-=p-DcB>As30FRwW*y#gVzd|pjpz2b%K_fsJHvz-C+c&Ai
z_@-9v!)-4AyB}<N6R3K0Y#OgNGCQ841$~H~-$~{)x9Ekbvmg@!s7>KJh);e;ob0Kb
z28MI)?(P1{z2Z_<h{aATpcU4eYsyC?&p*%Q6Qc!~O!JMhhds|ea>9dpTV#4$v;+3Q
z_Jl=zhfO$`jAzY4X4pf_x4@cV%HZ^TSfw~MMZ1~;4lO+=nTt|GB9CJ?!rrAGju!YK
z=%hKB==3TR4~@gL>yFuW!6gL+tYk~Y+<oZzQ1{l5<5nPNjz@qS8oELoNcIZ7sB=3&
z{5&>=_nn-*PG!bWDUcIhNpVNWc(JA^Yefx@?t&U+xIGRyM@~R3GLj!4fO3*zh-37L
zwd^{7iTX<S*96YFWX|`67vL{NMdge<haJGtKIYn)8uZ)|P?9tW7)*%mDRx%XQVDH?
z7FPvAJjaBZ1>m-(u;C>6%4*=}&A+8H_CiZVcrI0H!uz#$ZBJ+gqjpaD=Xfgj?yj%a
zLIAW<(Ni?DekIET@|t&jZnkp#gn!<&G_BI+_L5B=4&uyu_O4ZB<0g=SXsjt*s%R?k
zJYYToU^{X~K7go*(o<|tE6!G+Av!vI=6Pt`!<r3_3IG{1DAs_yH8jRyFAkPsr2bz`
z-9PmlC%Vm_@KnY15A8A(6F-xk+8RDx4aix{h8E)wO);9GQB?pmAC+psbuS@~WO=6Z
z<1;3H)%alNc}jV0Dd+lIRowFlu205RY6?3q7%FQY-v^yd%F@}FooZS{exG#y6yH7y
zA|KVj#7>CK&9o;alH1);dIVpNX1r#`yDs-aKR@=w%B=b%7I-rA6L281IGMlGiy-Sv
zVLyIk$V4Y;aI^BIN|DdMM^KM|d~|l&^VE`+iz{b|{`g(s=<iJNA8l)Jhg?KxLO?cG
zEXiLN0JLfg=3PaE-1c8t)7#7iko4kf9m-l(QCZ8ds0$f3eS<ykxmWb^6(<G&K5fvY
zW&>p8N~+hntEucOHRHL09zu;dNkWvW_jmCxLn98TrsrDUci~EWfA7jHIq#Wu+ls{~
zS+p`%Q45XdvksDmYv$+Wx%qE^Azkg&Q+(pNshzpzA3$$q8RV8$tS2FQB!@-sLd$fa
zyCqG+NUDfZgJV2%r1~>2(6?x!r=QF<3zBjH$!rw>?kmn|IbHOVZP4KuYv5mRn+1K_
z<*#Dd|GI5(QE?|TUKbwJ{h-`OMWujNW!zoROz?7r=+2U0l+E>m+H0-w5<Z3OZ9Uu8
zJ9A2tw}fXwoZW|ywMDz3^Edi03!Y)+kd&KHy~3$GvT8-Bhv7>l<{DiU1BlQ2F@#E5
zDqR`r$60x+ujoJT|DN0fuV6@Y7UXR9L7Vew4PiXfV`uXu4K^SE+EIT6r+6PraR-fY
z-qh}8wV8U4A3)}B#p?^Ymm{dq@&?EvM{0pn{s%Ck!rjf>yg|;@5k|(@1}f8TbU)y{
zf?%5v2>vr()T{!u+7|7fFp)XU8o}k?0l@K1&@2kDS$V<4yOx!Hy?x@Z1^#Duq2HK~
z3A&%o8JCY>_^aW02fJ{n7qqK%;<@x!Rs9us#pM~3Y<wRB{^^63k190%NtpOBv<m+g
z(1(ubqiTAQtG5$)Cm7%a6?~le+TbKEWddv8@|_kB(eGd7b4wOoYS}@*OhzSw_*i#o
z%T(x8WBel#gWL{;+XLjd`QUQ50CW0RV@)sZ2(V}!(bf34IiC;G29W2rz)9?cPUDCV
z&@vS~KohhYO<_L+a>>~)PZ&!+Q%_ymrG6<fZEWME#!SklFU*^l_b-qO7dLuoRJ+9f
z$v^Kadeb|r_c8{dqNve=xFoRe@+seou(>!dGt_}>ZwVj|fl(G{tus%uoG7AS_@xgy
zPF7wvLA>u@bPHid`W?_Vbb-II#q~0~kB_n}9gsBp@(xu<9K>y(d(2Y!4Z{l5={6v0
zhLP;txeB%a_82qYYo`(G2hjNZp=qACgcc5~08Es)s<ffU=Hx-Upr*42bLd5Yj`!|Y
zR}3_Fbw{cV%&sre{#$!G5#64yLB+WJIc)y<rvKN^J_@2}4E9-n*XMtEZ2w8U|NDUX
z&-aDf(MzfQp-`soAKdt#Fv0)&`C~Mi{~mq2wD@;d{(qnV{IkCMuh0EOXpE-_u9*FI
z&*JYh>Hp7%6*@E)N_RES;_t2dpPT!iZu%eJcd$TXc^e3owZB1<4@uvEs{M*R$c;Ij
zR+1jc3i3NpbE(g(GLq*ClT-2aJY)Po54ovCAA|Y7y)vKM;HPH_zmNO#ZiF9Ufb*;_
zBrOXc=Ve)fX-bvNRPN&-fnTe2qlbSzm)>*O4aohmq!>4No_(#Oi?dmz4N?>Jr(}mv
zZ%#bA=X3aV(luK8%@)v&sFtoESLiHHF*gAOhKm<H2c2x2s*EcI9kNr7E$%Jf3q1bU
z3tUEX(Dq<?c7INiO-X&=DdM2=MZuA?o6S$Lc`c`43v{xQoMaWXejr!Bq&kTO-f{%S
z7+z;A1n+m8L4OZD&7PNU<ghx}+8)P1m&@GrfR!Ii`=EX4+*iVrA|8}X?lj)_!jzs%
zcpZVMf`?!6K3qWi&G4qc;JpaI)7-KF0Fjf(EDum9tpxEKjyA*F)t9gG0=u~8ON!gq
zNS-yRi}SM|m$4|6c5;lr)UCCE+;<uDe;R!CYtom~iATaH9&ML@e!us%^zXd#-<N{$
z6R>7-0zc&ceY9pt>j%@rc76g%wS2?obX^SyGDfZDALbT;xyKUdag!c_+V6COiY5W<
z>Qi8pvWaH)R~H8<;a@fwJbAA3e%NN<AEcfRyXI>WD7&@nFV%n1cFMH)(SlL;To(y=
zj5;E<2m5s$AIG({)68j*0kjycOFjkM6Sw+qfG~k>ju$EQRJ{T2MW9XJc85bM(H6v1
zx64M2C2x}tzY>9`o|4{Ok3z_%PK1AmAXnmX=&zmp@U}OX@ZT_%f9^R)^p3sts#^Td
zhX2TM8G!a8ZvZvu`K_@<Q*>(28s0^N6yrJu&Ce?IO!ALc+_HibH}%`@8nO@#Yyc2<
zF*PT2VAzAtS1h-R`ejcux9E@mL)ljbWSMSjj|d7FGzf?&AxJ6$(jXEN5`uIK(k0!3
zpfu7AN=TP<OG`_)NFyyJ&9@$%nLX$1J+sgE&tZ6ZpS<t2u60EhCHei0AF!e#)+WqB
zwA?0F`430B$6=ZZ*`aLW@dgpj*8RR=cB+s=|6fT=GJ&u%&VBl%`R8$(!P|7~$%OSz
zZNlr#j8|t!(AD4~LwO}%M1C-;zHv|M<8MDe+dg}9&sfeo$NnzH$@_**y0F-5a))ck
z@ki<$+d*B+&#nH?tJP;tgxJT~rd<B3e=&HBg7D-t*VuH;-$y@t4yiuL{1+F7yMUz^
zXmu2B1%?nWDH+;dPnW$aD|>aUqQ(fyv#hzAGI+&D%wLilX0VYJ&r^=KPY$1W@w}eN
zr)KmF*fR8a#4A`n>r%#l5heCbLXjyC^9|>Xtk-&FqN;P033hGn&T<JiR%bEXHu2j$
zZWng2NW2St<tRPMNz_Jbfo>CE)kwhMRTVMzT>YVhcPvL)f+;7FG?w;5tRGEYRwfTr
zILXx>ZUNU|&2eXG?&M&owt~!d_(mL)$pbFTYn*G1?Yi);Z|m?!j7eHC7I?Q1<eJ}F
z{cgOJgEyt|-89Z+Lz@1jMQ(2ifI75t(-&-*_q-gX_#RZ}SZj6-!~}6j=VCcLi?zO?
zUdNw8e>Kh2n4pci??D{nfRDxGUhxgn8=V$DElv9>YH&I!kwvoP9P?PBWFLc^U%4`_
zt)CsKwXekIcyqeZvUB^Ap8BN=xdAPtx#r)9X&$vM3-V$ugy=mOFSB=eW@_y0`9IjD
z|7n|fgfJd&GO6kPVm!Xx$3RqW2ntHCYYph3HNulQL<mP<Be;Sz8$X~@ECYr5=S$ak
zh9T>qc9X)R9<8Kd_7Gv=FhSPQJ+pC+*Ja+zl{U-sgMf|NeLQo%3{?G@vlaGNTaQ&x
z-4Oc+Cm8g&;QzG5EeBRf6?ywGQ~7KNA6dhzOIARS@FrrUT5qwe-Y)})aU4bpkG(%g
zC-Rwrr9AW1*YV(W)E5sG)(PJpvIDG@)*k1mTK5P)coSmEE6BN5i2od082gro4-TaW
zKB^$~y@Nzi4TjL~rBC?y_P(CAYbMsWuaoFb;6<zyCLnfaE~hOAyua4~(Xv4d6e@vH
zJwc$THnqG9%Bdxhz^mUB)GUP)O(Np{Mf)KNQL`C%vIap4uk#eqCD13qJJt^3#>zn<
zl6AeFUhT@*J`gI#5f0Az$)5xqAuBMn9f$C;S%ig?*j1KJZJW`)>5o*fuD2Gt`A^4@
z!uPwg)nv*HF(=SZRI|xjmoqu%iDw-=M}UE6_mlSp1b|cxaneF4qtX%@3KJ^gMdKgd
z^xj<sbJR`KYpXLLew!gw(mGQLG$^hJTtwWZ9?7KMR>Ic02k^_>zW-n-*60C{P&)lS
zd^$g!hiJ=~iQ_wnj$8g@O^w6RYk*2W;|Z;zbUgZxfT2qaT1brC8n|HEJZ%ZIk>>gC
znOBIe_pJdGY_Ty3K_zYHL6L%sRFPJ0M>FITV0bVWbhxv;@Cu83?Qj84ilgpFcfUlV
z1q7R=5OLwXVc!CDerHHV1mVrZ0<5xd+pw2v@)ZNbTbOwfC|DpZ45eiGuJ6MSMRuCb
z-@Ul!!sMK<HhLKZ??yW|hYW#)d>v|JYX}jI;Qro-Hvu4mluTk5;56j#PX620b{8+d
zmHUq(8QcA`_*aN+^R2L3UpqC?;U(d+HrR$GT^Oh_=8fNxcnRT~B;Jx$3qhtV)wF3h
z>HMxiZkqwr_Cb;#38Q~&(=?_?|3z3qkfT~8N|hyxS9nN)>uU{o<33Ei?qvlpOObS1
zKq&>@oxP$r*YG0tj-vK%vRp9hf%ofb^kOcUGt)@$Nm^ic!i|h#+XnrYdG#k$tnZ7*
zHOg#$l2r43qDpwm!Z+1uF^GQbAxxkdHg~1*%IO*(a!Nch!nOg=hfi99&$kyf7C+(B
zaQ@!(3DsE^XoN1&NJw)q23;~Tn~8y!==cO^xOmY@)GRlobqCj+k6G)y^M0fpBU~U!
zhAYt={Z@Cg>=USWoPbB>FUf!xolei99aH<-r-}m4O}PIZt3FDELBpkc+don|AQXq@
z)PKVn_7t9;t5zL2@eqy_?335a6!d{78HtycuEa2K(I`{2|1zT*h1g4G{ODeC0F+bs
z_Oxr*Yvmd7M`st1WArV}N;6xKrJ43Q=yp578+rb&hBqZpCgD%g)J@;bIZ4tDK)0Hh
zv&zgV7&ZBB8f0Dx40Yg5-kF~PQa}Lcx!5!Vw{gN^eP$o`nWAvHCF}<3_cqREOid@;
zWcsmcKc3g?%4<W>sISv0=%hV8_Q})iB${0|f8A}dW${qbgfLpEY~}8?X+Pj4k>Av1
z-geu5ux(Rx@217>H}-<0b*!)OsE>)lY8l7OjrZbkws-Awi(FulWl{Kca#3)<dG4JX
z@0hs{aE~4-k-GqB^?ikKx3I*nKku`fS4}%=d$5m}i}D+h%~D94-m8jvQU2L>VPp`W
zC57z2l%lZ3D)GJfz4ttWV}OG4ESPp_vbL6p?FPy1Wvd<_Ma)_4LjCC&+zXj9kHh*9
zMvDrNDXQ#^j`qF_FL2lcP?3ZrHfU&RMQn(7MQ)`FZ-`z^IV+1%@6T8=Ou^!(Dl9t*
zN*j&0PYE%>IpqGJ^AV7CHXK-&6>o?$U%oFNivK*ii8`^%sF8OEeGU_2Z&E79Qdcj!
ziXSoKHNJ|Ne7ZPXcf9|?12Gh@BKNpkcv}&BJ*tUT(++V_Od5+3=36KCjDsnjeviZD
z^geVkoaQPR{!%a8I=xQEdO?+%g1AtM&=e@CHGEM|ngwITDPT$G?WT>SNV-AZ$`es3
zZM=U9J@WUYp{nV;RE;@V8j}opW?R8Uk;8|TSM0ps0-3bR0GWt*e14vkzQ?wL7{NJ%
zGi8vo%tC0AiO@v<e%T(u_q+A$K!Ng4DP1&zy_qsQRm#)&8<9O9c1lcB)z{V1BJy`w
zPVjg9x}I<uv^yU`sOR=KQGcRdY2xd2H#NQEJ9XYkGaUyjk3Vv`jPo2f;UHj<U%3{i
zpx)Ot!j(azODMt4Nk2l{#j%U;(1ck;$h-6Wxw;+^N{Ts3uax%69I@uSr~l>FFD@cI
z+#+@Q_V0c9$5Zs{Hw7<xas1}4#?}?FyX8TnT2g3YA#_%$<iT~1)C427Y7c(ZY32h+
z#*s63r)FyOF$5jG39%BhB(RpaF?%TI1+CxUi(Hpc0_oa?DZ15uN)54QMdPdYCFPmh
zXOB5^*)A#12#gE6z=*rlFR4C+d~jvj2K7D`MP$lO+x5K$x-Sq6?>p^7KFl)sWoZEH
z)Ny|79x7!&xcU-o8y`^VNoYD<FbGee%*|UMp5wweNumwna^DeM!X!MV;}l9MN<|OP
zwA{fwS#lBC`gC4U0kzmt)FrLckDs|@JX?PkSb)^midI}zG3Nv+JiJ$t5H<~A)y)Ux
zG-&I>?zrc{&J_*=Q+r$;PJwSitbliiXxTxG2-VHdR2kh*3IE0oax}z$1Nd=UX6rC#
zIFrOEszD*ZP~k8Qr4TAM*?;MY1w5d_#<<S)J4o*yDE`|W911T0gv&A+w4R*z6NFSa
z-_n(rHi^&O&~IT&9(|K`%cNU#L~~>Krq~rM)f;Q=FnhLV5QxEZkuDd%CjzX+*UfOB
z1K8Z_0m4*Qut->GQM1ZSjo)4M!SJYcRElVqR9!f03Ch!(KaHO@!z|s}Qoy^F+D7&H
zIM#jR-f4pRD?-X*r@GDz+<pQpkgzD8K*yE4Lwt!yV&g~XS7zh*kD25i2A=J6qZMC+
zK1HYcXZ`Ck+Yt09Dw!R{n%8`nkl$h;Lzqe?lqq0(51vkFwc^~Sx56%b6WDl-T^2X9
zpSq4j-YGIY^l?+2r$j$d+GSr;3XI!)$i&>qvOS0Dwyh}E?)sf$BrWKzWPu5(okOYJ
z{f&E{BV<}0>oQ7t)T$WV%KoFb{|l<{uf9`2pX&D6=n@~)zwF6UAG0SpW0mXxt4*h)
zK?)B9Us46Dvv`bcI*q3kSM&+{;(XG^L8l{ohk+(iNe5#Hv>W;nJt`L^e#7G_<8wLS
zD9m3d0h)WiF;GaY?Rv#%I8}S|7o9cshmk-Bj0fXLpu>5bu54<ci8ysd3sdPwyc<VG
zV@SAeDafxl3agO9EiuVDg)Hi8-6dj{#_06s=+yF}>x>5jB%*{7bGUNGB)!ua)Pabd
z-N=V7vuPMp%gTzHI7r#Sh?el^7JXs;f@IXJ@@qWco>U^PSI*T)efH9BDZ<f6BBBDc
zpF@n0R>ybI#uXvqBj~|R_D9uBYa_{`SfwDPselfJZ%w^ObNMa_)8(D6i0z&3CmZI=
zt1!NcW87Np&{1J3HCyJozaPcX7aiV3*k@jEs4qx=x-uG4LeLOI!q$cq@FOgfHC^%-
zO=v(7?=bsP-%r0aqRqB;uZejaMjA`D<CdcO6au2WF2J#jSAJ#{#%W5?{gkasD5ndx
zrz@LM0k4WmlF^1j&BQ|;af12*^8WLAAvqXR&Zm|zh6|6A*n%ap3esn~8aujB-2z={
zIq*igH$`|^??2zz-G33vcwcklsXoxZf3!&R?dt}+jn*A)b+1d(enZ*~w+gmV?t{*k
zm$^Q{6jc%NBrt5HU5>O<mD?5Jy73>t!N0ccWMK$USoAf_<oD(DA#VB;$!$<OjI2VB
zldIpI5ZHQZG!A!Q{xb9hgfB}{!EMYr4$<eqQP@2rYrSxYG=dustP9{ID#1>TdrdFm
zb=4w}TNI3fAAnp%H>J2a7-$TCY<J)ZhJA9wBxkn6;~h89!^gqwaJ@;&Q&^zC+R0A8
zz0&cHV(K!d{$N2B2dG`I(UTUzCL}~vmk2hQ7^7UCCf*KNme0^S{qj|C>5GfOpv-yr
z!>}0)1$J}>wHsdT*ZkH)Zu%9@U|HA|rbX{m&C2KEl)0_F9|6}4d0B(MuG8S6pJ4MV
z;_~6*2D7@5c<_(fsu>sA;T|~@{nxfI=fOvfz4kxrX4W%jA};7<v>3%>G#ov5N3ybt
zBjw3*zfik$)@zP9<u{s*hSWIBH87w#UIUZ4-S(@6v+tDm{Gj5$7VS*|P4o#t^l%R|
zMjP(OfwB?}%t%%*YL{hgWhQCUc&v!w&I;!&<&^C?L!I}&5}Pq-Er^*KPCHr)nH~>h
z5sAFM=4dd^ZIj5+!J?(Ba9ytZe78zsk41dMcBj6WCf=o=?0F%Fxo=DAy`^B$p648u
z`teh?)`6aja&O$i8uSL+J6{;0M4sL6Vm>xfc$(Eb5Fs)3N~j=Z9TGHgF3XiNXpR#{
zY_ImFN;I<k2#-_7-!`R`ICp-CWHsIg<JhI6UbF1cfGXoRRW>@b|1hBa3+TL<g1T3)
z`OE(pVU1y+_tV`J+}4RWrYM}oA9?P|h?_1<)rZ(HZ{jpEmav-W4jvwn%V;=$qh{WF
zu>Ap{3_WYj5mvY%pn74hEDzO<zXuEWo=)C^eM`))knZtA@Nr|DSZUokQhWDdDCrV-
zo^X`ty-fz#%RxJgaX}S4-7;YHw?wU*gh|Yms!=^ZbbV;0KCON!<{@|R7Jz^#9!;qd
zYa?MCTH+M7z9mVO`7ssnn&j)q)DwzlrGKvxrKfP@yxXyvwN46;7vcIT*sY}MaSup)
z)F$TlKQ4V~5bsEAhrYQITKg$tJHSa4EYEd{KB8Q}-?!9?pm(uvj6vp5tqx1uu$~e9
zEZsT}!Ry_(es<!E9>LKZ6uElx|B1!;SE%AMDpWeg9)j|OzozP3%x7+HG?3Q^!0?bT
z@FoA)^FR!T>>-4&udy9TNZ8I_a`Y6UvF?eMjGLCH`yLtN99c>;e7j+iKc^_?#o<}U
zg-3YO+Z(0GXR_b+DMEPLkQ3yS7aZzwBN&Z|wRyk68pgawWA{+?2<!Xx{rY;@aehXu
z8@ak=D$`#_vl$+6O^>NG(oa=5Zq?b`A8&FU*Y@g;AZ5%v>*@X-0#lS`$x+=90i#ek
zEGaKFRy?p@@q03($K-bm!Rl>ZcgK|(VH}kh04(Nv4(r-V15Mw{>lg1xfu%)K6QOE{
zs*pr;9UGEg5w?dtuVuLjAg2_;yYK!XiNIN6cQ{lF!s6NZXaKYwF;n_alxkvUzlmA7
zKNwa}EyYL_jfXs8<|JUv4Mq|Wa`*MNR$~kNVapsD5rA6@5KLntcv91o52@S5yV)9S
z&4bQovMZF60qgN(e#~pVDmRFTrFLkh414MfR7HtGOa9&Va@{hcWg+jN<$?{-Yi$$f
z-8L-&tc|#YH+^AD#mi556Gvziu}3beo;+7K`IEi{SEBWZVlO#pHi+e8FS$+hLI|TN
zL6Z6?s}hy^wN*09bA<0EKN{(7To{|J4sB5$i_vuk^Phh#e%+rTjHtr!^vz$hAy5=w
zXBQ0fP<jDOrC4SXRs)$Sj=fsIZ!#!u2;AAP0giy9`)^4Y(+~3T8Xth#!;%$-!pu>E
zLKZ;IRBS3$0UK5tm-}%)jdrF8PU^gFGvxTnoH2eV=I80zN=d%)faeuDZkAD<w(JGB
zlQzZ%AKF(%Xlo`o=bv{1QB`ABGgtCMBMyh29`WYY$#YX8oKw;~42L`OFS(B4UWfqU
zN*Hk~+hg(e`+$8U+vQy>NTz*Z28oMs1VrnC7{*oX9<5tO5C9zj=B0O;J>M}E=-#%8
zsSvMbUO<9e$ODa)G4Jw_dutVxwd{DO-h31zk1GtWX5RSkMk)LlaJA7c)=Uu}Lx_74
zu#>n(Arm@zZKCq^*PIVzplowve|3=<w;HMF2yZ*UeRC#eC}_m(0f<ZX`Kd@c9+eT)
zDLwF{U1O}-n?`_nOopj_V0H?D?8e_*6z4}sq>&o9{qiLC@NC{ErdXNT9|8M9OyyJ;
z;W%Hn)YRwYd|$p4Acp-SO6ZaYY09nHrBa@rxHP6rt7u8;YpqtqcHW&E3(74?H>RP<
zZET?a0WYUsYYSA*N{grIqL#8Q*hdqfKNH>;+z>rZ;T7y4x(cYUQ!@!8ifBaMd$n(`
zZNY*G8d$>36Xz>3U?j5cjVDq5Zso!04<Iz+9|0XLD_kw2<o)BS|Hzd5YkAO#%mJU>
z5}SMQi(1!*qjl@Ckfh{_;sIsHY-?!y=KV!*T}cySa1rO3=ujC!LPeWUu3GUbd`3de
z<^7ZlaTMwI9+Y|Q-Ap*iRj+ispmgI4u<r<Z3G0CqU^!*29V^O*O=lSd6!$c2X6QhF
z5sDlLQ)oK1-EXJlr7}u_#`0T;!H33Rns(iKBZ8xWSgh;~>Sb1iJr7k#BH?|yM5oYe
z;cZvX*57xX3In!+JKjc8Ec6i!daW#jKgm(U_E1SUo{{n9H}%`YFjgAJ>%s1B0g!Ih
z>rvR<7NoBpO_kN?riIAV+jXi#l$X4n;N@3mevY|=>rPk?1P5<sgsrW4&f8r1fFu~F
z-LD=62UkL)7M?!pRs&Bi99%Ds;JI?KyE>1u8*Z31IBI`7kJ2sEQg1U?GJtuq>58;$
zNg|s(Anr_B(bm4?)>vS8`36}xu6E}BG8E%ttUH4~*ga5AYd&8;bM~BPpQ~SLrTsyA
zR?&IJa#)m<H{E6KCGaijSfKkh{9=R|>}^b;xn;YagBI@U(;Rb{;RIrf^@(V?TW95z
zY);pUd$cgDlKq$ffZ~GT1w-jE$ZiWnWnTVJ#HL3%fbowj2|y+k;KE8&_~rVa1E9}!
z=usa#M?ylo7>?v6%YEL(VZ_HN=VV=314-rg@6lfLNS#~S|5&2@XGak@f3Ul{EXgYK
z{a{ZFFPH<*z6>N&*VoiPbl=v6D5|b(7J{y5j*q70a}iw((ga76!_$%savbahfp;wx
zr#NatsmY(ke?oMIT?~|*cz2tUZJsH2MqjBnt=<_FKAC#4iG<7TMN!y?PdvUkd*|DS
z#vT%e2L~pty7~$eampW_mk60{FjaPjU===_la!pDd4D<TMPu}E7kTWHyZQ<=WHybB
z^j?RtU%U9v-<-F+sjzcW!c&BF9pJ^OmZwpTsolAhFnO;|r7kZfK2pQ6HM&?pU+|GW
zij4l?b{vLDobp|(n<|e2DAx7+vqL%YBKLWssc&magGti2tx`yhig+YyG`+N)k`NvB
zsqpp%^BR+fXWedkT9-IKH|hwGp_#_Gdp|h&OVpK7MBrZ<bTe>sm29p8+Ns8uisSbr
z+`A%NY@eC-L|-X186LD3DV(R|kt-4{lBk1ApqdcCsJFK4xDjNIuw3;A#b-41ei%v{
zD$X!TfY&lI3E&U<&hag;V^p)wkKNCerKtVphBX+_%T76$li_{lD%@L=l!1=54zlg+
z-*=*z++aG8S|a~RC%gTcw~Ha13y<c&1@8h^A?i|Y?TMq?)sr6crgru{#nHjOFF53@
zIm(hb_{ID!!%3<7*<VWDH^tKb$JX^PKZK$g=ofFX!_j_!1`SwIya_vSqmUR0mtJhr
z7`+c767PHE+<z0KC`6rmpawZ}T?{Ty827TUO=IKfX!eW?Q5QwxUkP9zxd45lE23ye
zM{l&U%%HsQ<KkOlVKtFhB2V{{S*roCU~hfSS*MwtBK_^ZG5|gWK~uC}ATRmnQg}+K
z)gH&ikbxM2d})bjw=PM^+_S#^lNeRfm@9oJq6mUj9n2l_NG`Yqym^FiLh=ruOO!+e
zJmirnhJHISx9Z@q>f0i{-s<9yUH9M4ENq^RaAO(V^4&I%d?AEqtbkAH5v8v6NWb%i
zKANzikfN;EG=(<~9^E<A!EUzmglEe{)%r3~3GgYOp*0emy@HJ|D`Xi=Pz;L3QxYdf
z$9btmr}d6%>G-?Wb~aCT4*3WU68iYRJVsUaiJ?A_dptV=?Cn_DWRWNSR_G^m@3U^>
zqhP=I4fXnOXV3+H`?r6}AmdwV%9KC<XA){4@2*nvs-;V>LLk5kJCRrEUwniCwmAas
zvr5pwcKdm-%P`z{QBY9M)QZ<(Q`c&{N3@cW|CqgS6tnDej<fM6lWTHys6~uxz`%@|
zcSr34m`qdU8qpkLum<zb24_g+yG0oUv+=%Kx6^)`@&p43ZQ+~aFmM`YF9(v4#e2H?
z6x4L8|8&g%I+LYqaP&5L(%;elJh6UN6x;~$vMs1axX!s6Kk?Db1s=06_D26$SA!+G
zz}xF6z7L0U)wv4|dMIeD2f5PNqz$14y^EWi3*-T`VlnN1{?4|<-#ltJ*C`o1|8-G6
zg?`+eYwv_<`(stBS6M)Bo+vUMV~=a%gzN6^d-FD|ly~*J;x3Z2gIyR<E))HoS^nmZ
z{C;-n&Ja6CSBv%i-|rdn8%#&Gm;dkr_|L!b>!(X+;0o=$4YkDi^9uckr{G^#^e+cR
zhn#5J`U9T7e|rAo=irPT1%9j3<XzNN`^z=@zx@^WHkub-^EMB$vHo%-|F?g>y@oBY
zp@_vrOmyRKj_c2d{D1$!x0@J{t#QQHc=n&~{r~<PT;uffsMtpXFQuA=|NN-TH&l=7
zw-0_!Q57>z`U&{y!uf6WX*QtyE{nYcj6Mx-?j2tNbz(f5rXWYQJqyBrxH->p1vVn_
zu<reFQncq$Kmlf}!#qN9M?6_1psg;O&h*WvuOQ11r>7ZPe);xhmz}!-?6@RB&dl?j
zvHJ5}@y4U2=H}qgA8B4J)Ruu&t<`u>$;Z`xj8pJ>;74SlJvajLk;aMKsH(E!w3_V7
zDZAA5{8yJqNsfKRirH>b`A0BM$@{o^t}a~thhx3|h6)}xvECHMKQ0C?O}^0pofXWs
zV!`xd)F&6FiDq3esBTP@BfBI8A&vgmc)6%I5YA&Kl-eBtmCKR^(Py#icgD;ee1o3e
z)^`7sEg)hkhTvRPR}u)ORcgT1o(D66dqnEivUNKW85V#i#(;XGnN7CUHLU)v+LCWA
zs(KS<)CMKDJ9G$FI)~C9aykWUR>>s}O+dpO1Cey}19xKRficZ;_WWhZIfCbpRiEmd
z9PK~*_~J}Pfl`Jmetkt;`(U_DX&muW(!X0ke{&avXNNxQ%G@6}C|wrCchd@(s<8;k
zmj|Y>^H7kpLJq4$z^KehLvsMBK9owvV5FrJ!)81mL(Fax^$iSt3LYs)R(o9_iUAtr
zWVTv)?+C-0mN-tU4(LHO`>GtaJdVpikre;cywMwnjcHGgUn+s8eEOL7iX*XtJv92A
z9np6?fiKhcx#H*%ohUz5WB6BQ5@klWytZUr7Y#cQ`aZksa<DBgYA}K5D)AE1ST1m6
z<lOlVaf_tY^orR^Q`tH4{FnmjgTn2W)pW*MSI+<aOLGpqX0y(G;z9ZSiE!6N2d5?W
z_f`vC1T=f9B_`2^BVZ1TO;;3-`2F}4`4SvWz(ieYFJm#V_||SFy{r7<Z5Ht6w8iP5
zx|TC5ALwem__>Zjvt}GpJ<f;w58Bs}FX!GtkUdVIg!vI3g||ng!^f5y=~{L&A1mwJ
z`}_Nn0*5EzOSSsPdMC3AZ|W27X(K0r48d}vyV)OKE&GzARj9K!Npno0-gVkFA!7Vs
z;!juqreazYCs)yl0M)Jb2GbMTAF2|?P6~=g>FT~s{P8=P_O|_hzp%-uBVR+LJnHpp
zf9U_sUDI{H$+T{fd>~)@LEqc^XYIbNe=BPyF}^a<$IO)Be~Nbht8aSX=wq_@7HZSe
zbVu<361L}}iD|Z*h7C^+YTfz=K|Z2eb;+^jG$Y3u)vbqJUYdU-?J37M&JT*ByrMsj
zw_M2PvizCTj=)&=ZgP%A<w+IxlvqEZ<qrnlnfym5#3O;3Gs}l8nTe)XJrnabYqA4-
z@J$u{KLjvpmR@<Mw#{iI(cQE4JT;&B3;U$P9fpSP%v=0b)s7|2xw$)XxM$~Dr2`b#
zL%Z#~R`^{RmPf5*luzYccJkJjGs`l`x)n?p)x>t351A<2`L&(4C2OCh+Ye*8NsCgS
zN|J%Hj*4ctk1*8omRUkexCBM_*b3(-4fDza@#86%&s!ne4~~xQWyYo)rq;iuFDxgx
z_TZPaOm6KSE}Mw*s;<)SyKXxeDX#9_!Fh%~l+flBQ9+(VdhMPz|IFxPjT_6krOG?v
zI}c3l`Da5P_Y*}DAJB}Rr(-6ciSjaif|p>PUhdx~==EIIssF9XcN3Ot*lp-TJ0%XW
z$4cK#zq#JXcz?99+s3}1ZG0To&)gQ8Sd-YWpZn&RhDU?<{QkOICiCTkxw{Dixl6Cc
zc6RiFx(cZWwsx~5_Z&_V0>2PW{J1rqNE6;uvAjBi{`W6S4IXSxlcK+!JU}*P0b#s7
zAOf0#J90)C7Lc6TUX$5IDOWodJHW!x2?FJ3UtE5u7Xq*u1^F8>K)mi=dug8O!MhJ#
zW2ZmyT{Uh*d$tb{T_+a*kru)vE(G}liRrshaB~?BY6EoiUS9QXjIq>M_R`bu;Fa)o
zvxspPNaNo|ufU3fM0e|qZIpbPhuH6!f~PH9v5L8Bh9?`Z`J-Tr-#vmjb;rVyurEZ^
z00?&40VWhix<i2Bn-{rcFCj+N?=@>(p8-)`5|4>N`9q^WEX^+1=6&x2D)}K#o9O-u
zoMUx4h>j)Gq~be((k^F<j2q^`LML_vd|0Z#DF=C>OV^i}fFgSyWIA_8n{7c>GY|Eq
z9mUHM)3F#}K*sPNf0H(J6}p7o4%@*O#C|DoE=kz015gO_5@^*+sh5B!Z5G377&1tX
zxaF4v87MwoU{x&rw(civnh*XZd6sN;Yvb}I2MDMst=<xkPZ@>;ea9b9?XuReby=lS
zDT^e^$+=1bDo_Qh?AmvT+_`-UQoQja8L01hI{-&~)~}DSv!nT3Dwd^vSb*7jmFD3S
z&Vbb>KceOS&Q!=;`MC=PWWZ>k$#kLuy0%!vARrFVlUzwA?I6z^7l@cx5wy_(Ah!vJ
zt@+N0qfr2Q9zkvhOLMqW+3^AZhIe7ScRMCoT!o%Bh_$O|Cv(1f{?wO%Ua}>StP^w;
z0rOQMgSa{LxNjw6**h83EBdbj$g-xfO`c<g=>1r^dyGfEgQB~Fujl9pj2Ha}=`liz
z`7V-i^^b$&&>HQ#5NWZ4;KtEftvZyU$R)+Z{!=jSZY{4#H+gJan*dLMc`TUdt((Ok
z!u)5#eiTTy9WbLYhAKk83j$O6%VZPQ=ES|SE5#1p?WeOTZpoSyUtXTb_>n%i3KU$r
znPb(7WwUpJ<&T?%Y;r5v=0rJjE4@zC@e)W=v+=}o6}4kIUjM6x{@?YXyC?cEmcpM6
z0A1J%0zp&--Dy_^0X!v3jeO_*&9^WdanDsPHX=~6l+hACzc%XojD-g@v!^7iz`auf
zy3|V;n9+VNCg?=vUdeyiehL#?B2W}@6YqpsBH9xDVSWh|dR9z%1%CRI=Os$o%38@&
zXoxZyA@?_yypyE&x9AS*gbmEl0<8ZljOi&bRf-I7l+LJW@$;QmxQ!DB?{LAY(tX*t
zn+o~3O7FBSz%YaJ@}?o+bC<Gj-zo#g@$o|Evv8{~WDB(LKCsP2?5p;H`DC3BrBK-x
zvfmdQHrdwUh+f5Eh4FX3AXUpqY3QnfunOiDzS4cU5{e9G`VPqv9W-Be0)0+f1&b>e
zVQRVKchG~Gf<iBJF<a*5lY7pa(Dj_NJ>(slRM7p#5eLoy?gVY93gdIm1s>5KF^%g7
zJYcc74b(nf6w+@k0|R9rJQ|8vPU?MeCZa;km&obsw@N=;gOx?3rQY{;VT8|@e2UKg
zy$D_sauWR~sVPd$ylL-`@+n68$5lYqHm>w}uS8J(-nr}-v<8zk)iaYY*Sn`VV5g8h
z*D)B^J5F@3(>%bipzr8&nG6{{m9>k0(_5xVU)z^YvJ{%i$%biV>mW%1o7H%*1CpAZ
z*1rO3dpUcfi&u3^^yz?asm`*obF0Qk@`B-tQjeMznzNZpNWkVg%uZ};RNk2x1tM-;
z?gUvC6*0-1a;uHUxj;A^UZ9CKL@RP%!IPjD!p#0{C`SSz&Z$0N6`e~Yxc#1&U5r0O
z=`cr$dw^!jwN*dxX2GtpB!af)*6y5eZhw9-5yO~3LaAfyl@XYs)ZTo7x$1U+;kfXP
zWiwn?14dUUHpeTaXOw0H*0ri7c20V@AJhV%c{g8nv{y2B&1T5=cuVDOwQQ=t&yt#*
zaa+)HRXe}^%H|)bf&Zr`Ul*PB@i~k?swyr4foX4zYDY9OE^7-EUy#gj=l2}O=}$Uu
z1>jvD_BJF==t|`0M{rLRQf2y?gn--gLKkt@^dxZ4ijPwSgG5^$#YekdjN%4Hv59q^
zh3%4BQV`;o79RXrhT^S7%w7G!Eis*YKEBhaZn)(RcM<~ZBW|dCkoEKEis$AgHMRb5
zvEZjcrH~M=+_$rpy>JZ6n^a68>8%%5OsH5E*1_5C3j5t%C^~%11>zdz>Ti;oe$uZ@
zHYvoxidVO+rz;3Ekf<6sUHV}W619v_COsHQ(HLKbwuxW7>8{NomO{~-;J!AJ{y?hn
zIZ#qP{hskMi_j98X3O-5-eO4Z-W_85*7DQqkC4iC^XYw;>o~#9-a;Ad2g!FIq3~nq
zn9hos8!ge^NNY1zd+3c%Muh9;8xHowSEB~IYj8Tx`#ht56HZGTr`yL^I*wapGSWMO
zZOX5vu)r>!RHzSLyHfrJ$;MoDVa)o;)?7D|8ryre5thwhi$VEPRYblD4_hSbLcJKf
zaZb;1`d4N(CNX|r^}~kU#j4K*>Pw?}cc{-~G$!L}zV3sSa=!Tm>a0$N4ezA<VuflZ
zd1h^o2q_w_S9jDM%7%`SLjAPCRgn1Ko2g;RG@f9z?(`>7T#S-AUQ)6%?hSgbY$tJC
zd2=-K*V7_fS@!B5*2!t8bLj&~i+^uU^+W~8t*E6!(H~M&>V!6OYimbNyzdjPXdg@<
zk{CASI9r*cXL3h)Ec7}B)5t1T{M!aIQ4+hcIWeK+bQfILMvN2u0SO&ypZyQiK7%WR
zzDyJ`NocP^QQf9#g2gy)g(RX%Y1dL!gpyhI;5OLUXryll*;H+a2j@+PpuTZ*jkDhZ
z!MoiL{2pe-5Chq#!X8-$Ar_D8TP$!jT{eaaT|%zN+m6E=l#hO);L6m)qB=og^3;5P
zzA??A6mkvwB=5i}-O@`oE6LM0b}A{7g+qys;_at>dQ>xvF@~t9!qRs^BqcGoEF>sf
zpy*kwrc;ANe#yS0I6p70u|abtX(EABKeorV?Y?~ydJ3*K9j=Cb&UVS@YHs|VtYDtW
zEqM7=4lj_6rDh*b-?Ype95%s`SsmPDLU}hB9fN&DaCo^;HKvlbr00it**cr~Y@Vl!
ztZZf5lc5UL#Teoe<pr-mjDd_mo6?)dw2?NY{k6vZL{su@T*0%B1SLLJe{{b80V|+Q
z07PYaL>BMYt1Y8|x_w8Hya|ZBO&%*D8-hJ`R7=58kAfqp_LM8EGuq`+IE0nkBB~Z2
zjMeH%AnHt%=(5An6Ge(9Xr{T=Z0XElR3$ZwPqP;)Bc>C;@%>Te87GQKDr-_UBY6()
zH(#a4#zp!-BC}Y+dud^rcz|&)^wVwC-YshOifLbZ7T@%|#>XDG!spv|LidEmQba<B
zA8yZ;?u0Fhs8X~qK%WdWgbmsVWK>AUJS29fKG?v?%a@Oe=T?N?_Q*-a&Et)FJX0r9
zsZ*dfsJRK1>Rsb=wpXZ2EFtNn9+W$bX<w2PUcQiq)NQ=xU|u-2GHfN>(7&7A^T}?i
zUSk8?7tM)ou{h^tZav3Nc=#!cleQp<f<YWRdq+-|xj3Tj2M9_i(&)jHX=Rk0Eg#;s
zw$W3t#jGf>1>0Pa&WDwQPO0xv-I}Tzs6>X8K65h-Ic7QGTDC{RYLRHjx0F;dI3YOH
zk*<y0w!FXIwj>wjUw5t8&-kD2S|D=Q%49G6e%EY~yXKK67}EF@YPpL{RU3jt)e=e9
zeX5V7J#M=gbc=38N?f!CrTa62BSd>%hYWgBN07o4PeKUI4R^#KtH+sA;Fh>D@b3E(
z6wN$p4Lqw^pRRfx&`52hIgVl~oyhOV=`sonhrMNvt16QiGo?cYBY~9(8C<?$=~QPv
zf=%dr*VcX-@(NABD}<YX-_CMi;-((zN_C4zHmue;q-1WOq;@K$;QdKZV7LXb8=vcu
z{XGq&uSg8H-1gwFowFBlze+qu!NUpDxN$W+A|I2nVq+qMWh>#QfIGa$bDOhh=u9kI
zjkFqQwatqv=FiO$Y$+i+A}HMDhi%d<t|NwxzAH>V!84Q%V&R*y27+BiBg!(_6lIT!
z<}Z`7Gt4FOXXM(HerD?%n+ZbhK-RB2@Y?2{Ppz>t`N7j$%R=SZm0ySg{{f}OMnyll
zg7)hd+_?nbMg^7Q&Y1vT*4@(;m*HWQ9l*(z4~0f|pCl1tC}+xTlmdHz<GaL@iR+d+
zeuRuHCEM!xTJCsipT0Qvg7NAV>!to%IR@%DM`CX?YWfL7_m5z>mv^16;;}BjPZXkY
zxN=J;7B$?JbOdm<QQ^n^g^t^c!OeGug#UO4G9TDK{&}r4#P?_iW1gyrg7az->o(jx
zoo?q|z&RG?XF%v#NSFs6Wt=p3VoHi4u5W^ev!!wRAg$>WP(JlYm4v0dcn=r0D_5|^
z_Lc`fc>&u{Ny~?x3)Nc$)>%^TR5K!{UnlK|Z4AAC6k9c@meUR{JJlVM5(C9+gcZ-;
z6YDW!lHp})k}aD(I*`2#1&Z5{(BX$L8;aH)IwPW?f4HbJR{^VQvP>rY1);oMLB~J1
zqX;gMFSAzff!r)^fZXQ<uH{g<)dC{x%YM$SVM<1?EzDY<NYQmGbK=d72Okc;nj2(2
z!nm4j+D}lINI_%<gDTEaxgCb*WZc$^*w)K~ou<oK-*52ID`Z>5A>s!ij?n{~j;K53
zF=Ca9{uWd@XPjiy7n8X>Y1qDrhSY}wT8BF7%BmRho065J>bcjUC8ST>6pL3&mJGPH
z!UN3iT{X)KL32~NSIlc%Ql#68OLKyER#>VtA_V_4e#NeSVFs)bzA{o{&3>{^R1f&T
zn#D{qBI|2+cMnpdkq%aEWj+11<*NR<<zh{yiJ7EotW?A{PO|mXFzCfUd|vH4+Qjy^
z@X{G{d{F?oEyvT*el`8H+$dhI9qNF3jcMGDq!mUf2i||73Yyy}iX6@yV6d^_8A9yI
zqCtTKM;X1Klo6$|K7=mIs^pH#fi(?_taXM(CcQ^3j6eqWY2JB6&52j!odig{Jdb4?
z`~?inU0aUo&XZd=fpNM0J;EdQ+R;}m;Cx^~HER1VDQF2hp+ALO3{#qFHdM)KXK6q-
z+bo-wB)+i}z$8p6l?!R0glx+p0MCU7U6!dO#k{3S9`oX<m?uS->`ha_RhI?PjC%lv
zz?6ouI#y~oJ#FGe^;WDcNO0)Hgq?t^L;~l?@OT91nP-LVpWxRnI;I*6SA$w%qD^HZ
zjZNOeKHyMIP^Oz_EGFx4pa6M19l4JSKHqug2>&&hLO%ClE#kl~VEuvHuj{+HYw`Cm
z+HdZ?NL>>^Ui>?*S}ZP*gnruBGLBd@*sN~Z_)CliKE{clO7Zi>mKU8y!>GaMmG(F{
zsZqu_CzeigCz#?+6R(8B=;mX=B-k}8%)m1Tqo_I?^;!>$uXVEir^_SUS|4Bdvjp*M
zdXI?|nori4k#-u8Zc5{{sFa$KnShQ*N^AO9MwxlfZ*a`Ah3w_Y@nIUTwBBr9iJpJN
z57UFE?^Go+d3~s#3-ibmV^G0G-1J|N@Jz|EotuGmgTz~%HKXQW7s#|&RI|PkzzNFR
zlC@1YSIAZS<_m5*tvl`|ufs1b<(Y_eW1LbE;@va7akXp6LCrU!r&x#_(Y1crk<cdA
z58cEh_!u_*nn<1(a8Sc}T9ia3CXFf%MmL}Z)Bz5Xc!vodbgX-)tmm7ggHKq`QGM={
zwR+%j*;F{olHd_tTFyhZ^LldDYOSKh$=*RDqcX7xUi#_j!F=81$=!h(#+6fu5izf{
zO}*IA2b+P+C&cWp+QXlp0CvE&3s_Q+%PW)5Rd1pgBR2}<`FWVhE~p<?pH8t`Yka!;
zWdHGyYk+Zpn9tmNcVZ&@O}3I^#!Vzx)Lv21vzF@U&5PpuFVEG1@&cEesY%sJWO)%p
zR`}pLcdzC`Z(wR}WzR{!%?;!C{yrtLpQ1K)kT+C*L(%u=GqY(IK(3^HxhWUVD64;`
zZLHy+W?3PWFTOH!Miim`dMg2p$1M;mN)75eC^i{k;ee|rH^isuQIY<(n>3SF9_jRv
z>0GH^R+6*Wn;qB%(7^ma@H5Qv1XO#)w32tJ>OY^?r6wCFA^HeZbv7y8`SsWB29@d<
z?)aRd^=}PA?4?jXAhv4uz{`!-5-oYiVU;#PCrj}j&XE<d81evmQ~d}9cf&Vi^`FKG
z?;A|PJx;&m!+<Er-a;=dRkaHy)+w1j?Fh`zq+U_>@G&@^YsYwtdep*vhNqKxpQa$3
zf?S+6nx24SPtIEKdoUyJ1)ox)(S7}g*HfK+bn!3Ijg!}OK(mVU(+tCKbzcGxlJnK{
z+ezs9u{#oY#{%$nvXYFOs7(_YEK`q=*@T>Mza8!+TFs<d6S3o)nnzdqzaz#CJFF%3
z`y<V1Cg_oU<&FHlELS>%L~JE}_2S$#vLhr`uH-6allxLFEJQTRDpAd%xgiEeqgR5z
zN*`x_XP%0c#yv8u)W_Il?Jim^Jb2>r9J<pAhg4^bSB2C&X^n|XXHCbl3exYSIv^zS
z@?F-J0gZ9TWLmSD$s;9uuuhe8IMH|#ymeDD>TBZ@h#$jj9PG9tn?$VHnb~py28-<s
z57|5>EL1TcV8)d`*rz6a1`{I+YTqVr9Lf)TS@=kgFA<$AmV4|gKAI$a;m1jcj3kbJ
z5cgV{R_G&kOF=7Pobb;@TrjZOe`mX$@VSD4d|Tskkc6q13t*Z*0V8@Vm|V96`aK?+
z&>12zm%Q~<&{Te*Le=|g9zK7Y=>+5aI>}_BST<fw&f`J)$#bIrP)9zDfI3pwp_k<M
z{k9n$cEJXWO{)IZolt{yAei|VOH6aQJdBjNpTs9@tqP!px)Y)k^V?V`9Ur($Rn>Ri
zl3MspXg<S{3C6v8)h&2X`hD2y^*)&1V6cp#z{t#QG>Yf4c|J`s;x?@Pnp0aYmfiGm
zx7>D`MaY#$J*`xigZXnzS@|bljrQ+=LZ$wR2T*F5UcM{I+A#!B(xzK3_6xHQw{GYt
zezF=$svxQx>>5UG_qrvL+ASViv%5*RSh#t%i|zFnJQY<`pkUm|LUmkvIuJ@LS^scd
zt}K^7(m-4f(W)1Y%LcoKG2`f$q<9C?g?39bP={dB__Q7=eM$>hRC*YqH_Q}IM6pNT
z6oeQQ1|%MQrom9+jRb41!<DaqElbSNKTng^*hn)L4p@Sb`Sm--eeWNsvp(!(_PM}T
zVnw|ZO0F2QOd{t?drP%H2<CXtlX}b8F9RPm%=F7{*-WtLHDX2H@U(}96P0#$regA?
z?M#)Zo6=VIR}D?btv}yq(oxHN>_G#AdFyeijgW3!0b7_p31{>jq!gzCGAs3ll*W1c
zeb}@puCd!KE7ottn#<uruG49DzRd?9uLxEl%9PlMOd-d05&wEimF@CQ%?LEM@ns4_
zB=nA&YQWPsOnT0L?WkbbrZlX$nZGjZ&O7!@s|iM-pG#ZG8sQdfhuu_z7?TROmcz|o
zQvO4^K_#JeiA_Jb=*=IEW%_Lt1ud<Wi0o#O%{Iew0+J{|Ye*~Sv`jcjT|NzaKbQyn
zmz`ZDyg>HUpFk}NFq+}vpp|)OpYO6`CS(~mJ)M7X_Oivo0iMpm_P1t{cM`?MUYFj3
zvY=VywZxqAo!21L>Mtg^NKEhiQ2jNGU}y23mIkn~Z$<{IAFdPz6^!w6Ic2|TR&q}$
z2<}oNs8oMnfD&wBW5?vQsu0}l8`^=O^g!c$ntlD#@F$CAUon=pF<QfBUuw0Qm6Cd@
zr^2Lbqay8Awx5{keZ>cg>+iQr`1<=3*raSG(<n3=tgtscgAro02&&%~smh4;51es`
z_s!Gf<J1o(vlK%MC*7qMZYh|@lRn1+=DKMFPBENT^JOQc6wJx&O;3T=;in*0xm%OP
z&-Th$rOe`;u@zWx+nTtz?xh#(!&G*WKf<%{Y4<=$0?s3nAoc@IR&F4R)c6OeL{>zy
zw#zr{HkwTunpkrt*W}mdZsZJa=^W?xkOl<2iZbfMbXTjrMd{2ujy0XSaLaJb;DpYQ
zPgOf1n!_@wa`#l}j+n~i!I7G106qO>R%BY^6hUx_gtWXvsVB7nhpSMbSoVBCfOEu7
zP3^m#w9JAGOFb_?vxKqqXJ$1<jGw!Go#T0Hs+Tx?3(9Iisr3xT-(Ae~X6#Fpor<K7
z?kta^kI7||i)A0DcF2yzZ;!5*3$>&q=ryIz$60mVQTV*%H^(zyxu)=%oZK0VNRv&e
zo@{~5Pc(q@ZPjwcYon&i0rL{M@@*BWD=IUqxAY{_0|2iy)m45<d4-f-cX$37Z^X2y
zz3IT&HoXbP?RAooghsA+w$606)tft<f`89D5%>TzQ0n&3!9SkW7XrR7h!T()(eOhp
zDfC-YI!f5lbRA+NxNkmniqu-|PC&KLEOM_ybwly8lpfwu_)F7a5b&X{urEtx8<o1;
zgL29-Hz~s@yYd!3ffI#g^q@LfxJmycc<athvo4#JZ=>4&G3otx`bT;!!qVt;ZKD5U
zfo^;85qPQE2RqC9D}>3-W^6|Iw?3O4E+(ynyE{a^p>YR;#InT@x?EZXXpRH@Fr%eC
zYkA0JbDFRd<moj?^3I`XY@`~*;BMTCNt)o|%%dq-oA=7dzHo~!-*W)>_|~nY*{^jk
z5l%zf=vuyIcBM8++|SwHe}?h|dXY=^C~Wu7OZM>;=ku;V&=3ttf+U#A8&`mpYUrRA
zcJGWZQ6GEBR7y@-)@Z_AwHo%W;?!hMxlwWXu~{h?Uq!z|55;{Oa$V*!Dys;af=lnt
z(f(Gze0-x5V(U~6(~sj-x}faJClqvtV2TkBnKbh-bCMWw1b6Ewq{hu{-|@65RdYI5
z`RCF6+fny30zqj>!EO4|Ul%Pk-E7;{(P&{m!e|JlPAdr|B=I?=@>PN0k?UygDl{9&
zTnRiuX$U$Wjp;8B!8F_oIM^}3)UH@*DlWc^bGFUn1qCj&3c`w7sDJ&_|Km{Hxul`{
zieJ~c`j-|WJqQ&-p^WA`qC=++&XJmE!Lk+_EPW2u=;cT76Ih|xbAietqovenpv^#{
zAI6LCH#yt|sIj#say&KaMgDj;{>L}{tRusbZ=3sZ<J#XpV*h)7_4_DTfcH8Onyy*1
zF#W^(@^2US=hq0#Av5EB7w_}`(8d4zvyL2y3>L!8I3Mrt_y7OVeT2W@N<Mw&Z!iAu
zr}l40-G3H&B<|D9Q2c*?RGtsOi15kR_W${O{^_SXFQBe+{M=ghuj#w^jh|erCE!}o
zrbesesRDPy=FJu4ue|j@?+-7o&E1o||KPmbADo?kzj<Mda6o!vY*)_xI@|O^)a~yF
zunLKDR)Fj<hHX%Z(nh^atjM%@AYGPH?pso^b5yoQ^(gB7Hws0dO%tir=|z}Tqsu+X
zqIU-Taw*AYB{-fWk!v4vB4f`tSZBUyS2ZNfq?CMVlI8wKh<_9z#qA#ccH#FYlR#Ra
zp1}tXlOU`!j<Zh<>^^dHmk??CGsK47vjv<3J0SR14u#(bLo|Ry53#2e8f9J(d3Bl0
z2y~x0VLm!zS&Gbe!Dmfg;|6d#S?}<l9$Ce36uvoViiqcTt04wEwWad#CDDSDNY!ec
z1&8u|cV6stRFtX)``5xPhNjhi$+q%KZ0q^rtDGx;JD)&?6Mo5p*Zz1Q&~Ol<Qb3~<
zl=O7z38?D0M~oU(!#qF*UW4`Rkf2|rq<*G3;I7NzE}MFV^_|WbHmfEeqCMlXU5UyO
zk8ee^EuE0}EO&wgpIrpqDaLiVXtX$(oA!f_x@Q;DASpcw><i0f6LXXb>Id@V`F)?)
zT%;%9?r#4~An746g7=lInVW&@q4Bc%)#BvGc@D3i7<Fs^EtLox0eY?yZ2qw<(0U7i
zSBkn&In)~X^4&zBu#W5mUF;J^k6OL<i0BAL)dSIi_8Aam8%ROXl!WF6qr2Pfk<9mC
zBfg)rnMLo~PuGZhO=F3Nd8$X=OQ4$aE1FwXgMS9Sv77q$<i|9W|KFFDW6Be+V`Zf9
zl1Hsa{(#cdW8rgb&kP8#5R%o@vym@MHVKhAPp_PFXQ6uWb-h&T*H_JyCS+va{h}Q{
zps94>?_TTKo6u6<BVYS{0c#5sz(zH5dO}9ZYM?a;ak*Q4()`5g%T+)+W;z9=Un*Sg
z1fp`jt+jA&1Lg6`sT{X>X7Vm6yU7(*W|I~%n)um9*HI=$rq-Lqk&IrHN2{>=beWCJ
zxP*`yCBKpPRuatK=Ak~lnn+Y~gVZlweF@9Fc8#FEH?n(|!93FK{W_%^nH__?F7}%0
zWP<aCiL=J|9krpcdDC5g7u5>Qhd-%HA0F{7tMF6r_-)RntuI*A(CeOlw|YF3@bc&f
zKYwncpy_;rpiRSSX0Cq5(~=a&FRtMj{N0m%J!46Sd|roz{SIqm4D(l3@rtLmcqF<x
z1BHjN2A+_=Y8PH#*08TS)ZeeD7?jR<-y&pxac@DiMSA#)iC&nd5$?SJveT{ArT}U;
z=C#FzkWsnEE|cdsh5c>ftud`Ok5{*J{Su;0)%MG0R(IGB4l35yGcHZK*6?jyJ`9k5
zlymH;W6!wZLOztX)V)}B*g3sd7_&iXRNI`ddG{zTWMWf{pT8^Fjq-qNv(ai}OHEW$
z+SF)ep+Rt`9?qpyti?oZM>k7)YGL1r%Td*OGVONIsPk(FCyu}0ejfV50-Ex$_wuio
z^zAG3Kgm2tCP3yH0q6^<W7IsWCn3jk9vmxgLzl^dM<c}w3~@Gq40A`o46+k(_bG((
zLSJbgAg@^ByU(H!&=LHIN#}u?z>XMQjeu8~f1@4L0}+OhNBC-3aIEqR*xRoquY*lz
z2gpu4;p-xMYFH4CB%*N|c52)9s71(~2qHHR{?RPV)!QM;0O*exW#x2$tD6?W#6ieX
zy2HUaJ%Ww6f0Cb!z^;33rWmZSju2x!GCx?EqaU%4P-R^@_<97sDA8a7HzAqGH!)J{
zxMS2J&2awm)x{$WH~+y0PZj`^u_mv#$-!Cxf2KQ#(4MKU7a?aMaBkaS1Nn&-5+m7f
zGcNXy;;dyYI5fq1x0KRYHbmWdnu(v|*|Gp|sXh|P2J?U)QLUjgv4G!C1VlpZ!A0nH
z(3hBL(#@~g60`XfA}HtKcx-~M-PU@s5Akc-P|GakI&Cj%Bl_>5pUk7>1%%9sfx#+5
z9V63wRlUl-fLoD#XJw?%z}c?>40fiQ2f#?v6zP(zK(KWsyp<aL0PdPTu=iz?3l3lu
z6$|&U_8H)|Y>3U!x1qd5)$<<)PJjB+mELVW=_@u?v;y>iIG_{APCW*ahVh?|dex4>
z-D73`MH>xg_wkk=X9SmMljO4Re0nc*!c%(=svyrNy{pscUBsEr`Bkpm-zS`2F$Sqy
zvsH@QV&|J91|uy(RuNE(ZW;I+?9h};tJ7Fg+~|Zs_`-ZM!d(S`{wZQhR|wOXSm<8A
z1f(L+8A9e+p6PFf4PtpdYQ7nmGrF_MgM8Jc$bJSFoyo5MNSZ60B^QbAvjXrYMX~}q
zrip`_PQ|W^$2U&5zYIj|Aa43(vCf{dMqzeJn5F^9yOvtpP`?|wd5>Z}f3lfe|L?1%
zwvZRc==VK-uX&(~=!m_?oG<A87D`=r0wOeUTvEvY_CEI9|Dg5o=T9eo#LPmWFGG)E
zHzhHFi&O_nfxA4#6rLY3Sy1SE(Jyl?z7-f!*v4DQZ@A2KYFB%xuH|)=cn8e>f<O9l
zq0o0&un&KtTT>hAnFMxcs4Rb+c9RR{GQN_mjux6*9puO3D_M$bHC=Nj$mHP9S-hRE
z6Jk*ux1f+RH>y@*qTKbo!U3pGXgMmy&*yRs=-+f7DBn=mLJO(`(%hQKFq921p|__h
z;iad-oPE9IlSW{VdVKi9j;HB~%iBrtMj{?Yzh`Kwq%D;XQ5UE>22C#l6A4;V!@m<L
z2rR7mMK-@6FxyvE4pR6;@F~wZlpqro<_FR{Gl)uXSSeUJ{E^a>l998@9ThDvv%q+Q
zcju6y22X#a!ZxmwrO|b>{r54O4Vc_=Gukjdvf70-l@oV@ey++Ki@xtTZUJ(g1UvBr
zLR^a3Z<zAy1zMz^-nWuFVx!5?>K`Z&6dbbC$#SM=y_D@uPuPSgtQYs`^qbP^WJSs5
znj@MAAen)b)%*(&Lc~Jhy~l*urNPv1uQzH?<?Z(j;{cbDEZ3f5g39+XV$T!ONEoM6
z`Z<lz4BZJtD{W+~7Qg~SqEKfrTls#a?dnxfD;8W%drpjplo+b2tZuWf2?EZktS7gc
zsefDBx0z;k+m|SM*gAydx~cYMe`U*`)+p5giE{M)g|7ICvVJA5oqoBqq|x-tg@lYM
z_)1fmGp1-R?gVElxm5Uf%(h%r>P!dO#mi;e${c|>f0jP|xs^&B5=oLe@BI2Q845kh
zWyc?DWJ(2<st>knkifw$jV=4v$Ag7z#k<Qs;U||Rrp<-EXx9etcKuPb{D%sc{){l$
z_4~hf@iOkHh*c#GO9;A{$MsYgJq*r?IqN2`rh1~W_a@@K<^c)!znDfI1oR5J<At)m
zaW5Wul&Xp8;F31cDh9{>sQM~B9SBLYdgyr6A$`0)DVvYHys7)&3kBesE^<Oehe0KT
zk#_!?{YRg7CfkJw8EJ=FKI~Jkh@|A)VJJ{z&I9C{xSR;9!jOI%8TGkPTBbOv%Wxsc
z<y<nH*hgdWMu%zjrjnGqFAzuwkygO&vPF2bUolOUfV(=pXuvl{#-_pR2a5&wsz$C~
z%=QdoTU@AF&um`X=z}}b%Isro+9;@>+la%CtG9CX%^SHnYS)mDT_Pb8R<bBe1-)g3
z!Dsspl+gx$XcQS!Vm|XmF=Z8@$$#uZI>4ChI@++}CG-UFMog-QAzFJN6uxW{Caox0
z(%KBpF`O!$>XBUCPe~cC7xr^zitZQRq}#g?tDS864E@BqR;}67Bp6%8LoOiEL~KTG
zCW)@jrsQCMc5E@Yz{7%+dJv<9ZP#>mIQx;9b0*7>hLh>5+cHV=KId&}D^T57vimLd
zD9yP@_0k&GAd<RuL(#Yfbw*_dLNopR4=stS8eEefhZ@(q8NtF9azGkv7e$V9^&h$%
zBIt7F2FR7p{wn@H#lGO17q3dDm6%F-=<55ML8-8$o`e?l9YK!gic8OCg_7P|^cREP
zvn<^`SjR9j71O;6U6dliM7HZP=PWjnJN;@MCzCIMg!RhsodESc9N+AlvYKiiR6lUt
ze?5W1wpP*}`c}ODp%32*dj#<u)k?^VyKz*5IFnFHr+lDplz00AFK|Yc8ci{E#GBzo
z*sg-u9-Vx8Z8H6#Oq}i8Ya+W_nA#j8t1<BtvOVu9VsEKvvN{wrLW1g*>hKB$R8|ZO
z>W_u0DepIJl1_f;)D)4vyWk_Z8j&mYTZHjZhVS4-3=e9hFz@0w|Bth;fXXu8-W?H?
zP^3OkLFtf20Yy@z5eexQL>dGHL=dEu?h+~K4(S%8B&9n92?;6by8CtXoU{IOX8!ly
zwOF%eox$Vxz3(sfe)bcJP<($?vofe=&l=*%<UUe=;4_t1((G{<baa`-hwPGeo~^8;
zgU?8MaTCsMm+A*EPWJ6zml}Sl_q50zVo53#_<;n!PD_bi*xIW1T-lzz(-y&U8Y?x6
zDg1UW?1EtKk4sPcv@1I*H@ODY@p3=%F5kS&Rla?Y_MFl#rRM+p?I51nmu8F24LAwn
zJcx$TT`JNr4G2agFQ%_d(cD)5$s0mvH@AAj6wZjgK}9CY&>pc-K&@G-THsN_-Vq_@
zcM+d-PdYfA2<!e1Dwt(ZcKXy-BSJ9uYxV;-9PgULfpVXa6c!C7?NQPLnlGs$8o>W4
z=xTMyW-0LSNE;4mijC3&!}MXFM_&=n)*ARQ7IE?xnWybjUsf}RiKH4#y$5by9%jpt
zccknN-v0rtbID3t$^>)^E4}n5cRDos2QRjU)9ey9Ly5tmQZVG0(z!V#hSTz+K3i`t
z-NjGGR0{vFe>-ZeMuH*-GujG&hK4VcB0qC|YLs!Cp&*kgz$8b#>X}u7S1c0DOL$4D
zFT30t&>cf_?WLjtSCF7rc)cOWT+&MB-qKb$usSv{&-Wrt31DK|?Hw3aqOC5at<N?Q
z$hrc@WOw$b;z>azw^}LZ*R)$V6Vag;J4~PYH=gi6577iCWHudN^MT+`$Vli8ddO}R
zQ@3j(yB642S}+N}=u-;gI2i@1dhD%qpeL1!+ydlTJG}lQ`v<*70$@2{-wd^jGtZLk
z56(9LF*j}>Te-8(G(BZ!{=<(wBLj^<3SJ#quSYBahP1K-iOS`J$HHi;MnGaS<{)s=
z7okdZIfvaO>o5~6QLyMXBz+#Gve>=|S&)?fSQKt4AUW#2y*zxwP#)`;CXR(J+;=K3
zif8eNg``fWK)?RrT~gpiS}_MRMeiQ4x|aqlYR4VRv=!zFzDh3Jfm`qyU{}Qve&4fy
zO(kK%pDndUukonX!#6YQ`N~K!PLY2V)wNE*9<b6w-}B^NVcTW0({!kd4^T=$NaF$C
z4Z%vZ@MQ6)uWU2<!jQGwpZp{q&6zsPA);Z@%S6dksbC4Un&0kBdG;62^{?HhkOBgU
ziiX(zdi-4XD7*v9k#1u&S4CLR`v#3sC3sPeb7vzqSi<ML+3*gJM#^DJ_zciLl!nn4
zn)%AxE;iQ1aDi#*AJ?tFO6+_9gSSqsnnUjQFt98l+?MdlSh>9^<3;OCtK_9h^$F8O
z35(nyRz#g(26+3%85Rg{h*TxocP|-;%Iph3=;=(vJl3|pXXEW#*s*DSk+8c{DXbQv
z*mzanKayOG7}x@|f|w>%MyYdCBzOl1-pusc1N({T?|t|(F;fBP2pmN`H75dW$J9w)
z_$BTd`&whYl;xQv+7U46PW0ToYZ~iuw?90D;E*4H8>Q%lq{3GROd2WCM2G9RZOa&<
zyICeHtcKms30$asd>WjUtmJqm9wrL?E>8c7%gfmy@Hd%h+^?y)M3_(pxy_c9vWBM=
z^L2<n%PyHERLkpY(b*FjZ*309C*-;ou_l~=5N6DT<>vvuz(g{FB@WYS<OTBpRT2HL
z2u22L(6E%}Xf*p>3bjALJa!gU(f}7-gut`SqY@$^ohWa9X*rbYHsi)HkcFI-2a?cL
ze=_L-@_o#bM2dObdUs7_Im*i}H*muq>KNWS{{dU96`G!0lXlTj#;?yIU&W%z4)nA+
zvR1am9hSM6F7K5_;|-+=8YK_Yaxzh#B^3$^Zi1rb;j+Uab4H-zcLQ9XLnFAz3D)12
zDP;W#PdrRH_FwoW4S(13M&9-ZU;ln<Q#b&I8Dc2)3pjPXOewe6JO;2%`NF+&4<Z5v
zgjMe#x;qL<7CzgRJs2+2G_D<RPsKm@Fnub+sf6>{nl62-LNc?r!Bs$eRDOWq#J+27
z^w8vB(uq*dTSpQ_6|qLu3Fa6n4<jgKk2XMjmo^L0M|`R-M_aucPbmvLTBWlySmcr<
z(I`dPmd45pM+vwt?=1zo9O~zTq~t6BW#&aDi|Xb`<BF9;cRwboHIVn3(wa&uUtnQc
zP?dtG+VcUxUY7S~z(m^nMG-oqIw6ok!>phJ5$R(W-Ro5@17i4jfy1@CB(%^gvH25A
z|9HOe`Q3eV8V>6cn!1Oqb%;^#B{JJS`8!IZxR84|;jLE4pVoJ%&>TM5qcW}OA6&Mj
z<#6M1`;(bP$ffLR`}%8%5>Oo;wR+gA&J9ug!}jJu^t!m!FVFwkb-AKpNleXPxO967
z`&{h<S$L7eUB($tlA5X#ImNHZfT9{=5-ExWN7CF5^}3_SJ-b&SZ&Q5!8<h7;iz00c
z>cVNjq`)BP6$$;`1xM5H4wPn10(T7%wF=`3W~mkcODkQmyo19D=9g@*$h7v_g%Y4O
zj@Lrl7V~l=M4HumHevhgNozQhI~Z*->_PxVV&IcOgth}iLNK<cx*D#6sEn4_%yZJr
z(NU+^SwEF5fDaTx%BL30t=UN2Dn6gm)CJ1Vcq_Kt^Zk*k!U79?tk-m$H3246YZF;Q
zF={e5USLCezVV5^w|&j67PLcYvyOE-eLCK>oR5DzAeTT-DNo^fm*nB@6O_TC+DH|R
z7{A-+0YT~h#f#=T!dHsBVK?f6*3t%_BFWr?M`A54NA;t_F`qiyb1hf`B}xm$X3Vw$
zi*FFC_qXC8b~PO-dvLILp+jG^(C6*^>~*@tWv8%OaOn7ua|B1dHpLx#+f>AEo=Znh
zsbEubUVWZsZmB25SK}tcq*N4LugNB9^(Pk;SgO3UGzJNsq(-KDixQH4OsnFcY;i}t
z!R1`MeCNuXZ%4{$;NO~U1F$?gk_k#u$H;ZugK*KBRXzVDg%hzK?qT*(^mTvF;t_Jd
zhn1my<jVDHWiZ0ZxB(p;ed-XPFPb80`9a{H2<ZBlH+B`?U2&`uq8wRa!FVDbV@S`$
zMSznMgE00xiMdcgvmnr@8>nu1*L@qD2r4DMejWW`zHsYdJvzM|V3=}i&|2_X#gBxv
ztb*5{b=NpRRBj-MdzDpQLKNKUJ4SODCFtTD{fE3cgGmUwEWvb3K>h{e^RHMpP|_{&
z7&ye}+C^622%KE-AhgD>L^+&YdRu~w=sqL$cLHq5<!v|}(yVr&8+Hfo$ww?<MI?IQ
z0{BS##(P99<`K&BafIj1OJ)O$=lDHsKS|P>EE<I<=DGQAe5@uh=-eA@n%f8|r>nmV
zbKduDlm4v?nf}Nj@oD1lokUTc6PcC!zr6tN%Y%Xe*T*Od5XOyT)t@{S8s&|dJ2MQ-
zodBc3JGtG#hIPy-P|TGlA|N2i++AqcC8WVFJ(X_sVo`+yxpLh7xkp-Nr9C29j5Qt<
z+y<tv)<HC3w9fQCL{_6CIKuR25V+AMVQmFAC9akN;C%*(w^<ySZ{9K}Du8K0QLJ#@
z(c$I`g=MmSn_^{?t1u=}zJEv`F!FpG5%-zK-8#^D78f}h<#k;#OwgY#NA|Kte=Ha+
z3+ynPQ|54K=quC$w1QzjlGD+sqOY+O?&0+?VuIT{%T`BUVe_7z>eg!)d*C0piPnV4
z0yloGpMaN+EBCN0?3OA%`_@FCX7~q{RX><DHv3=ec+k=4L<ktqgg&fCgW#^pTc2w2
zxs)AV&+_VD`v;BlW5gIhZG!LoFY@>GQ=%6_5`?C4PagcBhzEF5B{HwY$;T^&=AnCD
z62o0*oJQ~(<_hwXLL>3sCki%SDIT=jQo4_SOf#_7)g6&SP$Lp5!ig|aW@$QyKvOs-
zt{1MNh4l__APV|(ScBo14WqO1@0P&M<%ZHS6waR*s43NyuIUtbB(iV;v1>MI*{C~F
z)F(DBDd&1)?`XRIm~sn2rbN_qTS<xhAf&|W&~GSQ*`v#aPhx8(!n7LN5X>kDJ{|G1
z0Y3SuPTmsJAs$bCn$oZD9U`peZokBfa}QJuz6Of$e4#ldS5(uG^o54<UfI18D^zaM
zCa{1~cCA_{UndWzIeZC4x0~4hPnM2Js;oBiJ5$jmBuk7!-U4Cih}3{_a69mH#0zP?
z`B6u*Gg3B6Gno(!;wr-aUUU$8SMMBwj~3LIWA!JF*EnvkPG4rSi)$`clF$yZQS7P)
zTGCi_a&azHb{3l6o+XXdU}NSUOf8<e-_x~eO-d;5>=VGa9^<qN5hIDEgtY_Dlpaa>
zljiJQ?p2TE5(rOQ&|55ByST|*&Ed10=|J(WXwitz9{4@x{u8M(q=ONObCN`4DB=Td
zDIQoh(6H(0RFkt6$ZHpimJaCo7Xa=xTrHj?e<$mGti5XgDhHGk*Lx1vSEp%C#QHAI
zUQzD9H!JXauJsq-&~=0||J7AfLEPZq%|ki`8h1rh!8EI3f&Upp;3)~|Hwt~<2ctZ8
z?&z2TVI2KNk!yKBnKJrKK3RUKcPSf617dG)An}|9_z`~o`-a&cJ`H)vH(_V{jidDr
zdAa3Qp+cih<KChG@n53gwd{mXLk!|gVsC7=Rl8qxclm_+jW4omj>T%63P!BX{QS>_
zR~QOvzJ>NEH8<z;N513T88L$w@4J)yP~7`%Lo+~o^z+9yS{{gK5)1NSNxC=85A0<9
zq?D9auUIoYjd-(UVSNm$;~v#_K?T#*eYU$_fmigzxdX_BPGAEgw(XPUfBXnBYSR~G
z{W5)NzUG2~`?KAmCtrm18-tp#B@jN)Okr=pqBv;|3k%z*C+hKKQY+S#5)?}`7$??S
zNmk<9;IA;DfT}6$hq{^^9x<tWF08e%dxwnt#Ly_$Dm?R%tl1-O{iNPYOzAj=&2d1H
zN6lR%?O^ZEMd<?d>dD|2$1l!!HU%p0@vFziy36N!;J~$nGAX@H@jX>dQH{)ai&bKI
z`W|f!cStnp>Gy47mYm8(mo*-caW-@A0Jkg!b?wcW^OwIe&NYd?+vGEJcNohK;4w*}
zyi8i~d1&FOg|mr(sd=xaQXEQ}x9yPc!cL_{SX<QEF6Zi_1L)xVP|W+wCW0pBCH0R=
z=Zs?epA9~*g+ijTKrd-GWq_dTx>?wq-&{8&A3p+S|6?<BK7EXVfV_coq*rb%sgqcw
zMQtWGg*z<|<h(6#6YPa~LD;qD_Se+zJYviXBAj@r)QhxMNee#@G8ka;<6b-U*B0tG
zT+NNp0nRV-{ZbzN;c7xZ)kq*6!OX#J9Y(l1yzs!7s>bA|+u1M?AvZEC{%ZCoFsAJS
zHcpQ_VRnoRbMO?kQD!|R4L=>HAmSNp1BwgYg2ZvD(t0n&5~Sx!fyz2t(XCh~-d=TO
z=M?eZGbPfqkZz<uu959m_smcK1Tgz3=cD~gmErjX8rYRyuk9+eNDT7gk}@vtpyazP
zbYgtIr%U+RIm+HJ=}Y%nr!=-wX}lHiP7Mn_lEA_84H8jn81*~`N~KzSoBtxEoR*v5
z#KZnx9j;-<)%@t>j>6yj@UKW{9~;73P;r@I`85HX!24WMhA?lUYa=zSy@z23_H2|f
z=%5iu$NXl?Typpiwv33zH^z$4k|z^jCN{QR04<7$^h>Es7xa+Jgf&-W<wkcWU=U;>
zt!92BcZL7sXz+nGStxA+i(|1;NeQRG{zB5Y@~HFDb8{S>W#yU)sfmYiBC}WFwOUL*
zL_CE4`<r#OMtA{Zj+h7k@f(ajFH0yb26JQ8ij0B@(IuTyR+7VMjzfcB<TjAO5YJx1
zi6-gq4W)8<K}t^Ri;Yoj=|`#Nc>rfS6e7m^qhwj$AHa?zmSPVAY@f=mX*8ME=l~fN
zMR_lmj+0a43wnrp9wXW5xxc3~F_^*=ZLM%*$NhJApFn_hFZOmn%z5ZP{y5J)^Z|zW
z^wjZw6|xi=92^jI0h7bk1~d(rq_lr<JuL$a0ln4GH)O7}ZIN$*$edC_cA<0(1jfi_
zctSjnm(GADnWKW5<!t)<*(MyiV55{?$d3B=2l;D%{9m8%A{$_%TcxY(|LA!0U;ikh
z7fe*K7n$$<@&Nh!z3@L5qwBsjkPVJj9Qpo18vi$|0h<a|gWGMKe_*xz%?C+129qHb
zmxHTae-9P^-5+&Bj+TVZdc}XSr{ox1DKPz<O|uJL9iY_XH-6rYau;0@8x$oY*rg9q
zZP`zO=g^+7_iN_--(N|>BY1gIViKo+->zY&Phk;)M`YT$2bXVJ#00B!UXncEfX3j4
z{ddjYnir0}aIt0QipN(1MTZKnT`DV|6fsKz@^a2kGb4$uP$8ZG?}f_=amIb=!5L4x
z)di@Tza2k=sIc*vRp>DV0%f!P8pbzhW2DqfP1*wUkN%0ZhwykL;f2u=TL0dH3_&kN
z&Jg(arM@(Vk<lDj&|e2?O=01Lc!?vld@@iT;rvM7ohG>A<4(1R3JuxT|JoA#bEppO
zb~%y(^gfysF$@C>4SSsz4#?mbG|Sgb9W1`e>OzM;2VIJk65b?e!rws0mb#I7kW4mb
zWP6}0xkArD7xKMVGL-HOe3vwIXM!UW8AR^Sh-EHcLf1|H<?K!4o`!1QX_195){^mE
z@q&}*F0xb5I<wj4WUvT^%qeKB(DI9$wq1z(TRsS^gjs|Vv$uss{A+x~81MmRAI6V6
zV~pX}R5UGN(=7WO<8r+Bc(ygX^9&X_aPstz!9*k)45<dawD!x`9wHY0GD`zFzRw@`
zI3r}~`)fC0ejMW_SP-&;49^gUj1HiNn2wZ~YlK};l%hv2rMKUD)6%X3#oP>hT&Ek9
zGTKc9uDyn_3#+o?rlg;D^1)N^zDk8=ITZkoRJFp#0rPMnBhDF2KIR-v$U~Tydc?Z!
zz&WB_cFz|1l=EUQO7t4B2c0rZWz6^Fub7!6btKOlneTqM@edm!fe&$h>T<4Q{B@6s
z;s<?@0V|`+Op#zZdz0sso(%Re9PZT6sgFiMeCubL;&A;8X9k7IMiQOnMDjW67AFu!
z+ylR3;M~m!Aw-A{@SV_1a;vqDN$U41n~r9Ee2->4O?#~r<SV`J4cdJ4qmt_ye~9ny
z{U@pDR%KHcxEXQbvc5Y@$1hHJ-S_eD1^GXZ7&tu=a1rR`>}`CKU%yl=Gvz^epB-Ak
z+#LYZP>2zHtJIpx!Vvr(LqUC8BxC)xL+_@r88VipzzX%WD30K#@l?A9=1)<*<hA{*
z@rB@-w1>yi9s<wpp==uG-(Fg$F#8OU>(+X8`cG4>K1s9@Ze<OO#UUVr7J@BO6d?Ma
ze({NP44imo^T5<0*NiL-IY4&fN0bz<6i>n{9eVzd_x_*WLff`hlH3{9MjLialY(ud
z?~>x)J>UO?8~*7giu6*~!EF4h`Q~;bzfO#jbDbN81?d>V1(@@q4rpOu6l_QgQuHIA
zh+DJSm?c3S?v;Fv54rVK58wGQ$GoH;)-Yq`3!j<q2WsN9f=y4)4PKZRS6hfvB^pU<
z^m+0~B`Pp3Pp4!LwOug8vv`TlHtfPN$>Y`Lu)_?-y<_>KC#23f3UQsG_m1C~JmB6p
z?EeuJWcfTbKyt^J(oJx+^x5wE>gJ=j7g0TFmk0H-E2NbAFW9WjzKAb7BYg6#(kY^}
zO{e(^PHYm`7o`H%X*%x9X1#5-`NU%m`Hg?HFRChWz}=WUQPd&(@!;)VzTEhe%9@(x
zZiif=9Fp&C`Q#?_W7=hcp>av&%d`LD=lwou|Kqg(AJtLcDKw;Rq!tf&O5QtKWUP;5
zBOqrL$-~wEY4_6sWyJVEHtgZbz0t|z$9BL&IQKKe@(qXnE2JZ7huOeGf#Y5MO+a&^
zAZDk28GhS>X=orv!_bLrP}vg%i|z;D<~zb!`1G5C4P+@86y^CLrbv*d#WfRms0t*Q
zwiASl^L%^R80PGkZa}ZQ3EZ8s6+8s_DuigSHxKUfG^XF%O>z`-f9I#?VNGzB8ICAT
z$jz7oHM!k<$J?7X5kAcDFE5^ZFrvAxstJf}=Kz#{1Bfrs)0L5aGY}O}nmY531DmMh
z^Sis7+jok6)@P8c+M}!7Q^4V{vGd%V?yR0#MaIsgr=d`ecYq_N1B6+y$>5K_Me50)
zB~Ibvj_axo#cC4T1p2}Pe9-Y^G+RK7>`wx}%v)}mSrBVDipBKJOR=aIvBLx@ZnRZ#
z^uZ)hq%z(^fto=&58oit0hDJQARL{`<%&mIuT97q=+6e<O9w2uKBIS3Wc_=;IY2i+
z1ilG?6sfsV4o7OiFhqlt;)=|R5kS{{pAl6#;1SRE&VS%q2amse$Kd4Acl&@wo#PFX
zK^Metga`V?kNr{tY>^!Du`krUretU2HiaZLJf&Zz?6__`N*cIbEr_r+xxrp4d&y&N
z5-<n@h(NOg-?^vs(QsuIck_yG6?d8e9+Z3Zgg&zq?1aA!(9B97a)DP<Y0C4mm`*T-
z=Y}gC{F)}HmHY#R=hGV|o?Gs2iCpM)BD?YUUii|B-Pl~rg0xt>=x*iZ=ND=K9k9qT
z9xchyUr!S}%4pqI2p9>C-nhHUpW`p}+sUaN$=xk3DXj(XDosICM~AJ#i6nxL###d|
zVscV?EB5D)CIuU_-_#t#5ZVw1rC$cLW{sVeEE)W7m^<Xu=6s#}4-cjec<0|URp!aI
zwypkyg#Op#cHuPIzsy{)sn{(isy<-vqA|%~FTdW<g|tp=?(wI+Qy{d&1=sT{_Rg?a
z&MKSXsRj$_Qg>h|FAxm{cK>1x>7^t<CG3h~#(<z`xdc<D&p43Fd^Xauk+TklHKFpW
zpnAdrPpkvox8@~Q+(QgE|4kg4MA$^=PYkx4hAjn;c`I{-Xchp-(e*qH%pnZOQK@Mw
ztq;*fw4dgU24lbr!bYJ~A1aQm`0<hYGk{d_EE<!ZO)acRUPMTFnn~{|bHfPsHFtz-
zk@TJNHQ6+H14-`OV<KqRrJ!3ZMO-S9oOAtzkK{Cz|A^RjVp4u$X2ah+S(tcz{v+U?
z_{%`9vG7w1*3m7cJC~i%>VM5Y2Ovo52v5tjD18O}dTh(FDESl9I8}KwUC(S|x@urL
zAUvG07yep63VKs<9rAXc={6R*HoY6k$0vO-Qv@L~@P*L9R}eSbtAD+lFF&#6gU^ee
zxZo8%2xM-l6zqG2ibq#|vN-n=rGO!R#a8iit&7V|9Z<lH3(c%R%prCITZ~j}(wCBu
zTWoeYSc0QIZ%=PU4sfmys;3lx#BR{kNw4LU4ZROIdS;<)jzpWkJ(E!;K|HGY%ukwC
zvBeVqyLeO&1~gsDPmmni<Nx4Ubwt9}{nX#f`@zR>ylmiBccsLO?4Lv{)pv&Kq`;AT
zr{T^?TvTKHqL`{2@n^zb7<*BBbs<(kiScQG`aAv4aQ^jGWZ6f=MRpy`gZNaYx;N+L
z&HQ@%=KC|x)~4#d$7ZjaH$HYn{g|)rYyp1Zf6%K;JKmpQ)eW0_z6}T*(61id+VR})
z#HP|xp()-zaAvXV__4)R<K(hmQ{;C1;duT`>3~sHYb_!l+cRk#%&5)Fv<Y~A@DGGi
z*aCQm9A|#j7AUDHh`Y5SZ{ND3MP3>n2j10@X&1f`WICJ*@fVwEI17>Uy01z%=esVL
zi4=Vh%luSkNWFcF<5dG`d5DKD_lR0970;?9ADxwl2PwhzVN5OKS*WNtryvTikPvrB
z;?A|&9&AxoY*edvtj7a_S;;E=`_s=Lh7O%YqvV`fQO3CgCpB;Mui|8eI;o%a-#IrF
z>v%1E#5tz<qqzH>8{)&0BUzB&-5(>Qm{?Y36{`HM#aE%Z0kLpsY6E#{tZh|GBygH|
zg3V&sn+xUQrQ|pbTEos+Hi#$Ed7u7GzVZCg3~MopU}a=M^^%(ksT%VFjP5p-gvABt
zle$=J1H?v|<|_J%#g^O#FiF>MD2vd&^Y^&bPr18!9Pu$~tpw^kPt4;-?jBRoqy&k!
z-+E|Op@+ZOafLQP*~NGzKswVh$?`USf&cZl@{eG+@9IytIAPpv+d?u>^(RZJuL`RR
zgLI|C`)n5^P(SI;p7(6;E)(-nD{YkIqn$3r-|UI*=bb3@B8*K#;c9_n0QYoip^n=V
z%i4XRX_>v0lOSDc;@e9!YrU16-&bEL3?3&B+zad$8@AmB54@2aw&$yo>5HKQ=-v`D
z$9F}0Z2AO4|4Fq7M!@9yzbR(#AiyM)VhM(2(wjoThcxOM<<<s_I(nyGJSo*+P2G4O
zyKwr2D#OFl0$aCqv)OQ+MRD^J*Avj`weMdPoeF&-6uiA`m#x;eN9u{mO2j2skLg{J
zsCRcVE`Y#E@<U>ayHy4&*74W0ac!PXnMlLZQd?ni(f!PTu!bnPAGF?v@3GdGxpO+t
zk{d8C0+mP-^EQ|=FRb(xk0<tsPPTD5Kk!_Uqy)}Pb6A3uUX7U~8hzzahOe?wX{aBI
z@*<hPc~L{V@L)7F6+G*MfZRsmuXnsnZ1DwNWOnVnLseY;dfxD-@x)BK;31S+ZCUl=
zHsj?Kkr!5<EOM3_-VZlrFB<VjfLK;QRMnSGG(UWSKZet=fsW6RR(p{5vxWRs*66LO
zAfC&E5|pdtI+MO5z=d><54Q^w+^Dk<jf;q1N(9dxDA6g;(N}5}7V<KVK>orFEt-KR
zYO#?O+w{ecpH~8yz1jes@Z2G3-Qaq2@_f2w<p0x48U^~__3Xjz6!|;R9$Y^;Mn=2@
z)xa@gai>k!_{lNy;56@#VhZ1v*C5|Ty<v(Ry5FTD!XV0~SX7;<K<S2K-pnWfHB@G%
z8VQel_bQE$Uup%-Zku=&UGSTd!W`tCtjQlEvsq63tmXc)$PZ$A$!7A{)n(FKSyIt~
z1hg6B@P9-_=B-Jao|&oDq0w>P+<`O_j%6!z;%Jpcb);l0%Mn6dyJz|V%8c0o22slg
zifmZqN_4C}%I``ZGwHzl7>)S(J{~-3G&2U@-23N9tMddDXl{+GOwR60{*kt|7k_eS
zSaj*;1TESQ{_Xygkva8QU|B{pvK3U$i9a=6Rdj8Uq_p8w=+}a*jrFbKtCHFAEXK!P
z!w?w~oFgITEVB>M3?tjzF-5c$`*(w%e>DD+1C?^fAJiB9&3wyE(A&FsfG8v+F3FSc
z?fd^yWdE+W<w}u()98_}{hz4GYb+Yd-O!i1Ks$4#dVH-SR^Yuf-xn7G8%cIKLg4*W
zZAgwNtBf8ihhyVwN(fQL(-LPe!kBi#n*@x0#{Cj)CD-f$L3p%F((!q_m7u#pbiyZs
z<%jwVr8mbr1{VS&R<-vu9q8WTD4tcg_@vq{=sdHE{Kez3&}7!GD$7iC5Z?_g6Y0h%
zMtDUwWsnwDAZn`dETm7J8mzJGRuoJDPRv0Q^;sG%%_BVJ3_Pp?y)jh=(MJQ@QJ8OI
ze!hV=+OQh%<<!S;$)v^~0<*s>3tACNX$;QMSrB!MQ@Pq%t*W8%o(VBJ%GJu5_pxKb
zPY+Qw2h#s{zIYcC1%~;R;3>AtIiDvt#rKpTQKvEYNHX_LTsg_%3&@TUU|QAhvEf`j
z`vC7Bu)ak;K=o{Ee69ZcrA0<p8Zf=fF$XiD`+U@2JR2@S?Ps6hB?XhG^F>-&4P0E5
zyIE2T3C>|g5~CN9kU}LpQ)t~p3qx<w#3!d)5o7|a$(8gBq)f9TorAp2P^RA|-ZLv$
zUi#`7Hbw2t@g?a{k~w)jDLKjGB9vLq_X`q9?EmEUtWbN2dU{%+Hc^Bus`($*`!(C?
zuRx`B2Mkm7d<RO_Xfle$Ix|j?*jgbQO?(LmlO{%LXNxPn>M5A|HF^3E!-CXC^{LFB
z_q<RKh0nFrdwT>XT=YX>MunLWC#*wmHyO+HK3?Pr(z6W$(jG(jY2W-2blJu?iD|SE
z)JyEh?9yImx7iks!^ti;(5R#m!-@qzarN$icaf~yKOG)ngb0PSk8}v_Pgs)SrBDVe
zOuZr{P+`f$@$oE$((Y$r)#)II!I;m3L`rXwB2itX47@tek~>Ptw?)rpF&Z<;gSC%r
z`g2&W^uc$PBsIo;L`zt#K5lVMOqbljjl_zV{t?_4B^M@We)<5h{}M++Q8QLVK6d%V
zIa;ZxGGAm#GTj5sX`axi@qFk8ityU)VRh3SP_JK4ZhdUv-cX5H8JG<xH$E)5@G0aC
ztlg^udKy3EH(U^sk%obJ?;+gvqP{fNTJSiRMgDx8wMSVZA5&6suiU=Xoz~#;JpS8w
zI+vtP8aUrtEE(oJKul8)<dJ809yuM%DRvLSMXs5;gqWtTLR3o}`afAhFwi3~dB)9r
zj9=sBKU@_sT!2SvGfHE^J833K6%8grH`wYXK|&Ii^M-MA%T-5M!25=Bo_4Ifwi~5+
zKe#g)()YE}`>0d8`66nY0=Y15ekP3&b1>?y^?iO(r3vR4u^A~F6`e&*i<g`wqS9V>
zquYBUqnR=<p^1R$y5PRswmqa+GXgUM`8A|-HV?Jo*bWaNBNS+zgpl*8{ZhyrcP)8!
z3{B7Bj9y@zr^j6a24ew{&<&Wmd}*46Y~8EXR~)q@Tn^U*UCtrtk}<)UjW!^`z3;?h
zHQg{|C`G2+3Rfs+S9|inW%IXAP;SpUWhRE2u#`hGm30Ku_G2DDEKA2fAAAQX)~Djm
zLtsx$q+~ajmCxf`0LN2N=dBkfARnpFfZ0Tq<EzOzmC=*Ax7crxOQFz$8lL{63{w{c
zc>06zD2t1~KR@_KrTRF?;qolDnY4c*nL=0SDO>|d4n-Ru_*@?V4ie{==vE2zBo_P^
zdtgLYkL4j2cr}*tk-{Z!cAA8sA<6o42o0%fL=W2VL0j8IgQqO-#xj+%-eb4?aIFPJ
z-1jt=+mUQK-$xYqhu;}JCn)7K8)7dz+-OS*v>%)ha0z=RD*%wg&b85_YNON714=@}
z@{nw#SFxG&yhN0Yk>?4h7(*1v9`HY?zllZCNTspCy{}K{oGzoo<`ni!RP&?5?LfL&
z#0ND%14bre5gB@fz&s%)aLeq<B+$+HesbV)(uKv?PY&UT2HM?&Ar}cBR-$HQ^?x8l
z42@cC>rb>_MhoV($NIs1Mi-2zF=@yYPrSW>$P*<Iyj22;oKqn2>oB5k)g0X0etr3Y
zi&6X5$|$%6ZGDs!X6Oh6u;wiEn{Rv{6W6SENyjGD#Oeh|=5SpmTs&(PXdR>#tuxZ9
zWW)Y;t>EIpN88qi_ntWdObx_>X+c2lOO=h<mdcLhCXkr<0q?Q{0tMsPZq2@E0u`|r
z71pz@Q;_A}%J>-pLt?LhdLzyEYsQ8Y94OZMcQo8hM|HfzW<ugfk5Jde`yUrMb1aph
zyKGPfxVyYM{LIU)@h08H6PrYO{qE}hoOO`8Jc_>L(~@=glXwNZyD{^ut6M33#ex}J
zy@hMjGNS%J9FC*r<l3RU<($KR^jjkFUq(Gb{Isxq;_Ub={<QU;B?$UpI8o(r2Rv6b
zpK^(YhceXDV;$oP9zLNL*dXt*?K^AlT|`r}o&oySx+YRfGGruX?#e*vk8TZa-Rh9H
zNq_5MScy`&oD})yR^3GDx6#j5pbfiU#sQ;~DUi|Gxp5cMJ{*M<^x3#d@VP@rZYmY&
z*rn4hU?&)xaNq5WC?^vB?hYQW47Y&KbK1N+;}EgUQlfXFqI^x2|4HoRfz}?gj#FF!
z4l#2J-mOfQ{=t$wga?*j?zm+&d62Mj#~{!{<vqPhQ3fK!GH1FC2pm28x;aDZF!Y;O
z$gEo^>tK{m$Sr6a=RrPAzO6F9A=#aqxQCDv7l#q8pZ-+Rr7Msc6HuNw0QkO2E?a**
zVwp&D58v0OlZReOa8MGe2WX}RfAkW!Dy1kiixj;epUs2%+JbMs@I@ii=pPapWc*2Q
z;^OlEkfV?5<E})UM+k2eL#)JS?mDZCp<~QaTadby%`Xg(45%L^SkXQVO_W^RyIkp$
z5XBOHTfDINt)(-8segkU!KU$xxBQNFKJKa#G*0k!&j76U1g$+qI&#ib{myKw!M&D+
z<O5Efy7-Ax>f-#qrn%*GD=(-=oi23{=AhJdXcuwl!{v3PZe??c+O!0kU*Qs8Et|om
z-%z~d%`+1+7uDoBK%CPn@XZ0qENF{dVAEi<4Tz|)Fug)dUfoMeHgJ1I)OX@>b>N>Y
ztre~1wgY`iw_1h6&vxIO`HaXslhU5oO~Q7Ior^?w**_3ARaaGwYZ%d4jLxu+bJ5wa
zrKKU>aX8+Em*CJ>0L8u0^4*qyz@J>7BLk}HWxJT4B5hYGI*KnL9^wQa_Tm%$_KKEr
zD=VBtvmT(+(PA+<T@j_;B+me+N+lBWdsJxN`!wllhz-$;@CP1(=I)E}g#fgj-OneN
zAs|Cox#6K@`XD*S`q8aDs{}&C3m|6Gg$#}E%<m&EQ-hv~5PGYGMFLdp#l#4p2tK@j
zk3w)dBvHbFH0!p{$85F2>1eCHE4+43%_VQq1bd-7nLOA+VrEN{{c{do-XMv4`qxrN
zxb1+*Xh(&+m&i>cR1OfQ{#%Cr3*N#+{soL@j-8=3Rj^~&Pa>zTV3Yp_{!f~d87F;d
zBlRlI8DkIxtk)Qyapo>PeeBbatTZF-OhM<v-OuB@y|CUiv(Az|c}aL;bZ<Vi>ZPa~
zk-wuTRc}d;Ube$~7)V#{;rnaOj<Ti*I+5O%Doak+j_#cv91+`G>dOcvw9d?X$`44u
zlW@OHg?S?}cZ|n01p@#RbA3J>18wTqMZ^q$l%ND4DT`ch&jEkU`ew9_P6lmOec@NM
z<7s=?$3&kL>eXrk7SgwRYAqkVMUT9k_vO0PXfrT?ZJQ$dRS`i<?1F}E@xX%CQ`kgl
zSfg8hPeKwjXJH0L)`?5>P-c(dYEuf$oKeQoc1BuUbd%(^)D!aiKA$t4PTICvaR<8{
zey*^fpz+bvq~nM+$;3x4dB8vy;4K3rRAJA@XXyc09A%i!fj4H<`;O78_Iz-kR1g2(
zEic{k4(3Ue|Mn#Uo|8gAfiMQ}oXj3AHyi6?5dNY$T@NWBPxD8$rc}LmqtH)UxX47r
zt}GYas-i~9^Qj6v=bOYnUFRl<EFfR<A4ls~EqLhw@7)jOQoh0M6IGCyp*X(geVm-H
z$FHCC?h+}-r(t~z9x*|t``7K5I-jf;+3<H^FZqqXy1_E(Ty#@;EjWV2{-G%K9LR${
zyFBu$jEzX!&VItc!shf)@G~jXA_wBiqp#=rywi>;;06pH9LtObY?%JIq0FbBE!RFI
z0bDBPr)1f>oy#CeZ1OSCTcZ2&dcRw2(#Kpd+E~=7lY6#p6)KUjp#{&}sHQMfad}NV
zGTH!FGk^bz+5JyqJWus$Mzl*zD`&F{f=`2AgH6l{EC;U3rR544`cxw&O|MWPmze~t
zoH$Fer0lX3dUG^x4hAmr#`94#S!}7P1vYMrYr04KM(Zf&n}A^Oky)~Yej`Vdm}UFz
z)cTdM>*60*h#a*S)v4oGwp%UCD}g(wS*wa)&`s7)R#-1+19tRpoc2>x$y?3u(F~^{
z?h1BGM*2y?mzk71eI&YKAkLsbenG+DM@@Id<+wraimZa^4~mzV1~`?gA}es?kua7y
zSXtM(t&{eUFaeZpg?+GpwlOkm7;OHs(QR`w5qkcM{VRV7lL@Cy8U4m`iyD!oZxww!
z>WdXepXg8Gt8cj0nzWzn-b^{4@$eA~f%@__4?+FO8yuMO>MeMmLd2vZ!ptw8dq3!B
z!0b*ey-i~(Mf56BI7Ca|Z2siPsl#S&%%-cn)^U|*XQy)KSfFdkmUo_So~LWWxu(ko
zw`+dqg+M;Xm2A?HoQdN@hOHVyrw>;t2P=nMBxDtJS>d`xq53eAsJlXoPOtK9r7q29
zq1}7?77xFQWch4W+J~!ms#8XG&UE(19QelGSpTjZ7@B^+^W|#F_R8q|z0kRf9cp-%
zA%9>6{ryiL6B^NBf7bWM`jRivCEy1J#0@NKk%s1KStArdB{D+0r`IzqUSF$7d7>@j
zMQA^}#ZIV1>uOSh1#vk5sw{9%(M>!E47o&Aleyf-XhPIrfM?13a0(1NB1@<8G})8+
zcOs=LRWA@wocgC9iO@eVT#S1jY10Uy{rMw`KZWKZKa{!uj?)}%nq<7^0g`0wDqaSa
z+k9hgbfcETDcwY7w0<+fi{vma!5eqhdP-`!nVdelf?d_Yp?K!hGsOnYv)$xqd>q77
z)~v6UQ+akG?>efU{ihECACm#&6!x*yop6VLJ-e@>e`sI|`+2gE7O+fsU=c3!nBid6
zVck8V4tgb{+oolvvr-RpZZi$UGe8^BzPk=yKb&@RbcSRQYY?Un+r71rHZjvuZ)MS+
zS!~#?pp_#xLmQnAxXi_P`bSGJHLz;!K6%ZE?h%011sE`gBGFTMUqSB73Q<o#^4hj&
z{L5=KN1L7p|NYxUUbN5%I>WB?TZT>Q-_N_n_YXci+vOqo2xc|CX=bm(7)+dLK`&W|
zga}sAzNdHvz306*R1IRC0taR;8f`GL7Nr;xeyNLOHd%vBL4z=<pmUY?W9Qv3XQgLK
zxIF!}A=wl~8Ru<IW=SOal3SjLTR;TO0bj!Odsh}a9o6r@`M<w%S4Ik<Ivvzgs<*#i
zb-0x7VqeB}05-E1k->RSo=$D{GciR$S57m%y^UFGD0}v)YT1jSn$`ep<?WLqmm}z?
zTfu;E+0ysh<mXU`0+Rp)%uSvz?<~3zA;iZGH{7NTY8ul?JFolYUyJI$|BIY2JhLp(
zA!eVSd*IKV@c;5pA6ocE1z&f`{%NoM(>A1n6=#Ca?fbtxy1#p{u6kH<Ybwk<0^+|m
z&wu}(zn}MC|7)?tOSf8ekB&s-4EX=}Nl=Az1SK8N^}npnzgxRPECdNtdBhJ38UB~O
z`QMim?AZu7KN>@3Z2tY-<W8aUbrBvyrnGJ?cDl8i$~ush3ifk~dZY}rx)f9b?37IE
zw{Exu<ehBssvQPmbUD+?v2e??tR!gtYv=!$mnNizd{)+Pg=W7$t7{*o?x{V{DYUlc
z!K_owYL+b6ju(7o*&tUe94ro{fytx)_A0AAIIO1%vAYd`Fi#h)6|#54#&QJZ_ePYC
zDC;9q@4TE@av{p{6x|(nBjz^glUQ^&+J~#UXs7Yyxl%ItfNbM5%{pqUaG!F$IYMwQ
zd?-No$l?A!z4P$<@M5J_?+nEMdatc7{~&fRNDsuAs=$UbrUK%16k&X8nB}%9gm^AV
z0@=$nM;-5q+iAW<6k(4=Gx&l$0&v*64uhlxMxUe_DUiO@3N(oofqiQ*^X~yN@X(K~
zEiq_r+nW*_QfAv}Ryd;bw8C!?4&)qSDlYD#eM+`*DZ9z|Ml_!Vek%kX(jJ<i*QfPd
z#GBduQ|5+KOAzICPzuzVKP%l{?3YG}%*~K*r^3Asu<Iu*`k1K`dC`=TB)oyy819uF
z2cFV!+kYK#e_a(`^spoONGB+NZyCtD1hxSkD8GV2Wd(S;=zC;KAWoBKKE~!zqYo;A
zhd6z{{SEcxi=9}+c;N_*+PZ;?*~1pW9+6;7D$x#k<I)*WTyfM*f>D(X_VEX7TOA;B
zw8!u~0E!qraGTQ0>$jLO8j|qBUsYcT=Y3VwY*b^X^j%Z6c7sYTV}{AUTyTsqyp?J!
z0H|kb`#Y;_AO_7tqGT*CkaFKc^osY@Z3h8mibX<_EC(UTdbaO@EZ3=Ia<vbVTEh*n
z8kiRfAr-I5L?dH`;SXCidYOY4+5`Ruv;VMfk>i2H_tnjR{mKUIAFiZTLqeR&tY5%C
znCq0DIrcFzvnmBBWkcaJ$@YwFa2fMXWgvIJ+z~tZojI6XFasZq9m9p=)fo7#>_M2}
zEOd1KlJarg#(~|!E17&5`O33m2@1_B*XE9;4Dd^?(eENrOep}he;M6c+%K&?%J~6C
z8%A|H<wGLHMM(<&Z165qtaYn@)gOO7dlG60VY}66{4px~YoTAmTNtk_1#jTW37q38
zrF<xq48Y32$nhG48mby3vo&+5U;3&lO_2UZ2tG)(bdDbp(cw{Ed<gdJapZ&XI3;P`
z#R$Fil+72~e7pjZf4hJBMPT!oEXtI4{#hGo<9!%#9a11_4p&x}gmZ;L@aRXV)(qkz
zMI6Y<dSx0EMh4QP7~kNa!QZ)CgZb>t-c7*HHp+GR0aVo26EwR)jFO}ON`jj^;bm`7
z{(Q>o`Wl&Y`q`eZq&{1I^CPpZ_O>=#zKrBC!N=f7RBHZYwS4|6&)1;4??V!==*mkl
zq+ZDtsIvyv7x7_v)n@_pCoPRss*ihYeqglP3P#2ko-Mk@SSHZW-On|KJ4jL%&xF2d
z{cWu9`)#Ll0`7;w6T=mb#OvvV&JuyU)?`U_B|NX{%A8ghchBwjt_$&vd@q~2h;lQi
z<oG&GCO9}~<sdSvVHQ&;NhU9V*2<EM`)pv#h5Z7%sJ54H;kw?^73bU`yIb2WY?3Ys
z{sFF8q$A%qo?3m;FLTnIdHu)ZXIC9Qc7ATst<v&1zB59BV^3;($Jy#=YB3ltW@a1}
z75k@q3H-OjUuxYRefrH!eCJ#MYpm(pnfZN;9ULqeENi}LnH+xUdv{#!tcqaFAd%(?
zFD@59`PY|K!B=09D@9~QX1QlYRtDTAW(>m_JnN9i8X%*1iIr7VFmj?1^|5ExSKqvd
zddMzcv~t|>BL{pzk9gAuyZnoBcdE6>F`6qJaUBx#f>Tpp&b#kTtEteF&7j5I)zZR@
znE(D|VUZxl-i>dcXs=_XbvF3#FUEx1m~j1GIcib<^SbDxMgtyaIM{IxC4K?(Cq}TI
z3ITOq^y_Lsjqks@K+XylPSr@N*>&JHjYiu75*Z9uzxN;$&zD&N;HVrvFkIPyF>(vo
zOfmuj6$Xi#lkdSVss$+LW*T*Xfe#Q<xL?qh?yWtXWh-gf2iUm{iqNm=Ndee&>s<$M
zx@edehr6uY{~Qn~u;0RJ540PeX)o%p0DMP)uTz^ZTaD3ZD#G4+4-R#0pebaNm?6&9
zx@y%7bTd^WEvOJRZTpq3+sxhL1rAWT83dufhxBpDTIVA$?4wg-&aK%u$y$O@CM_f&
z2SGb5%RTXICOfm(xK|9wt|9iML3cP-%E0g_d(s^>1h8?Ac3oMz4gIbo6m{)qzSkUB
zID9Xke+2uZ<O32J4v%4(3Q*w?@WwVM@M<lLy2T7^tnJb3Krlu)cvbLOH{dI@0$M@$
z*6h9ahg6D8Ga%3^Z`X041^Ob8i7t-%P(OG}>2Z&~yL;C3BLWf>3Q~M^<^RB~`SQFr
z#B~>RVYKD_;8+~-(<i93-aDKZJP8H~tpsGy<3i4#J-UX4ezvaP`HSoy<_gm}0)Bv1
z{(A%WoD52j4~Ooz+C?vFMCAc*j1I09yHCpmC<eREr{|t{sG@JO%;EZRwe!~Qby)f}
zax|0aYs5Q4Mxr%{jn0xwZ}Ca;_ACbKHIY#0#=U@GF282D-Y9O4KMn0bd3~^31V6eO
zC+NmYV)wvb%xvP7#9aH0F3yKJ{JWDbA?;kookxxDF-ua0>{Xw1O5HX*v*eBl-i(J)
zw-Qc}AhbIt?khm+VAzeQ<8$2^L`?lQCGu;Y&4D8G{kr!vuiE?kTZ+bmxmxURVkd#X
zdmRvM!K;(>T>gWQIw@&J#S-Cs`ZRrUcR+#+okL*iQLNfB@44OMy)KuL(vrL0mZDB?
z>$nm5g{us(5M>86ZDbCb_JGhhHEMKuax7Q#>ffH{FVQnSUa~%6|5s(Di08VuIq%3g
zwW+LTfRg2Z^DHa3BC`}+2%aGclY`)c&MOGlG#{$t8KcRDBGD`t_$C6Dyj6unKxC!3
z9L|vQqhuTylQ*D(xo}~SF?9oA-W$Fu_&#ZRS=sZ5s9n}mC63iY<4Pa3ckImV!%7%u
z6vho=+?Vyd6hS<jbj;Y4|2#{#8JsU)rwQZAlH9EZyUdT%N((c{pjOR6R1|U2e`5YM
zBY|PjPC4ee*82DA9c@`rv+bpc>dL`VOC4EJzb3S7C5dm<;P0c>^&Jcr25?c5>Tehc
z+V%dRTPOuph#`MGQ{rbk3O3YzgRs152K~2LWcIR+sspdUgsTq`FAqog9P*6|cRrvD
z8749_1=BPCM_)E1pKC&?F`R?V%#TWV#+bZ@D4HyRa=xw%ipdM+w*fro|Dkbf!L3|S
zOi^MkBdBl0^_Z0JS^jVtwjdwM<{4wqmLeHJXML^moj1q?E}q`3fUYt#f@AUS?y4j~
zr7*FM%TaM!y&7p7<L6BTy0)D9;L5$4Aj@IFTC=Nn!RGF*7?7q;fGF5<nL}1ryZ(Y5
z$I}J5)dY9k_X5B*D&DC(OYT7~yHzp7b3+viW$YMrgX3n_#gRG9cGrCAcCLJmA=7Jj
zL+q0K$s7IH*ySIlSzZ9X6X)Z67F6|Unev@Te~l{ob=JCrb5%;|Z#_9bX@<@r7sX%#
z&1yd9{5=(Ndqs}a;Z4{f-xKu$jfYTkn<U!L>gLWmX)Kq`1aNnKm{|9B=XgB{dV|G`
zHSQ!<B;{#&{ES8&@0nF|l&@j36$v9v>A}UU{2NYLeME{a;X9QV9cT-@IaQmA)3`^1
zn;bb;R@-S-+4K39vI`F_v4%({w02pW9q|UAHGiqRxOz%6D}8*`EqQXxS&Z*r_2pkn
z(Dch_YftL`)pLPc6*5yHWEYxxDJtQCm;+AdtDb)sO?W*cO;F=WEx3fdSKo(f*I}{0
z;{b-1@4si9rPbhk9Yv?|67OC$goTA{UTx4<<!nZyCl=^0R1({VT|V?cmY(jbT$hY<
z<#hs2gi_{u1zj9(_F*vp#SiOwE=O5|nMzEu)o2F&T`8Wbf2?>-FZU2VXD+%Tzen=^
z)+)LS-pfo9@>RH2+FJ;i`?A8ZMEaNRCyzhk3z`zH{vzTd(|<YzEBS+Zb#i0GA8+qm
z_R{O8Ru#MLcWFz}Xogm*;Z5+ZC-M?5ioF-lQ>@A26l4J#u#SX0@FEA5bXith_Pus2
z{M6@)eq=LvSCa&TA48431z5!NAO=0&vL5c8+Q{oPu{Sts-^DGKb1KAOK0NCol|p<-
zYc76mKPhNEB@VyiDus~U4CyAJh*mx!=2U-D+8Z<v!Kliai3gJN(biE1z0Y2U>vb}V
ztiA4(GEsTv3>4MtRu#|LN{%laBPq!)-!IfoYTiEVKGpE0?hh-9IgTAIPshcTjs|xi
z<f^k9FFR1$JXWOh^t-?pA7-F<1yAMeio_Cs^bg(}{`RUt^fh@1#BU31YV<b47^_!l
z+A1tq6ZpHt%6S!a;)k36`)V+P4q55xPQk{X9qQJbKZx%X=whyl)cqmrjF*G>z6QQ3
z!$R{)D8)N@ynyaIT!&#k-ZW5k=2eJ+CY_ekQyg%;aWoUFsl$23<`#5%BJXL|;8ae_
zrZ92S7cM-Up|v?eV$w4=;;$R{VA)PNurr3aFHmP{dop+O){&!2R#aOn+7a8sf%0Yz
zTie5g`c~!Bk_?T6XqhGAX9bPC!3%=N>wzttHgoL-eAC4JNmm&gax;zn#0nT1a;Ii{
zz4sC~gvrpkD#lYC+Im)S3PX6-r9cL9eXdkj_}O`np)_vyLK&6EGB$I-7_kuUMKW;J
zt~P02MS3WEB#E?5orf3_aDIqW&a`12Po><r_>@#-k^bd{-(CQugAX}K_&n<G8p_mJ
zF+9r6SSlCHu~)q|*RBWceSpol`k;GSS2|C_7XQR1o{R9%4r1qGbb{H{va9c-`+@A0
z<&~<NnjgqTr$!e+nQtf06X@CgM*b4&y1<&#wI+s*oDRF>t*bKov2U7q&n;_y*x^0C
zb+-PIS3eGxWNkUlHDBX@xVc&*0cDsSy_tRK7tr&S=B4OoaNW7aBm?XhGp4br^Oz*l
z=q_bvSH<zp30?Fu@1saZV#OQlmM}{?+fgB`0(z9hS97erSL6hU^9qEpEgiW$;SI(+
ze&?CFVce4({hq7Kd(D}oINVIjS74e^r)9c|@Nkwur^1?|v%aJ=1zY#r1*x@#9q`2;
zCay=Aw+?l#5~sP(Te@4Fnnro)%%GW(Pt-xkTYVXg+PdeE5|@8BGu~x+FReS-IHv*$
z<wa*--)a}UI5V`tO&0SE<LcUUJ#@4=vNPY)@Z+C$dhqM}s$uf8N5W}u$URs;@I@56
zUOd@*#Esv7zmhMcB#G!qzGc_itri(0aD56u41zk&ubm@32X&Kn_7vMU%?E=vF6yiK
z3hi$iv5$Q&&0o9`)yr`~mw9&o4rYLT3)rrfCcGKqnIk{6szaFTgBJsb?8v^o27Odo
zGr7EO4HA7o9=Vhp#B+l&UbyLO99!c>(VF7nfTD>c$=VyR;ay{K@D5#P5Ywr|4FS5N
zxz0)oI-1}8{$I~y5O@+DhTHDsTmQm{@MJ0cMddIBrmmfjOPw0WP2^>^t|V{JR1L~L
zUq?=?zlk-ZOrjCnPlT;@@%0prc~hdZ%oj)|%oja@s0{<!9Mf)bWzS<$`JZn{Cgv6g
zE3e?yQKq}gc(atBWEH<+iKM5u!GXR4IX*oiWSy*(VT({JY~7V{tIdy0ePSDA$b)m-
zYT);({L5{fO1Gf%qXzm+lTK)JiQpkh#gsvq_myBy8@#e<QHLr0<ak*C<95f}fFJ9N
z3vr(SKd_|w07Zqkk56F&(WPK9Ua4+!&-PJctmB?2)5YiCWtXm;snZ#;d0gODtJlO6
z=gM$l^{f;1?0%guc~t-sc4tb&bBeY^QVe0rGwW2bO^IGg)I`V38_=@X;_aG`cbac6
ztHQoS-jYP5Pqa7in73zlKs0vUUv1g`vHB`|-$xA0<F&I~jp*=}-!&Qj3t;%4|B|qe
zv`8;rv=aY`5cUzG88f>H7nMMn7qx(n^A)cR3CX_7H|YmVz>ju0cqx6art;{5PgS6(
zP|P85PkSV}wyh~~X4W3R7Cm&T2S%<b&I!0PbVz9rCI)6>AKgq0YILCAZ7Yt=W10s?
zS+1XiwzyC8n3fvI^orxmVgO6n7fCETEJ~`9fsYYsd(U>WKIOJ*t*Nl;b7oF%6lZ}6
z&g4F}jg-Jc(7@~<qJ-?^6oR}*s4!}(oMR#UsIIATEE|*uIUphVrC(l=jyU=8ki;zo
zr4mv4+l3_xV!tWv#XC&l8tmH|iyz?42D7ELf!90r&;^>Hse2ij?u&XieHp&6`U>d|
z*LWen>|W7wiLW#0EX~cP1F1j|^kxUI!@kq6Ek9`8l~0O8EZEY#gxcRytM8ItZFJ4q
zAl60R8#(qDH|$;A1jXINf1{9pUHN~3g>COqe4}C@<|=5UD@lvxbFelurP$9p|C8uu
zGFDFjPkK(jttna<p7T)6g}B8jrL5GKEVqKWtF&k!ex+Mna~wwvnqnmKjT=_`F`^YE
zMr1bRB=%^I?qziG>;|nwOY?xoKeKZ9NUC293moo{EsYvG3`<p{vbP5fm6$;&{#jC2
zOeD~x-K$wNqk%~cqfrD3!>KQP9-h9!2M<r+uP7-UP&ak(C1Jy*mvo9%{=K|0k9(qj
zEpO#-Zi0q{cuDl<kwkiY)&`f|`}H?TCd8xK&YZibt6?>xF$GRh9hK?GN<N44A1gV;
zc`3056G=~veKWwjJQ#hr&xYZ0=Q1DtU!Uaf|ApZd0yE3a-1v2b$elqCVRb{~FLQU!
zkM1>Lc*#Ox{-X-r1pqz=O>V9xeM;hK7Uo2K$Hpfwg)lTSS770mXw8FZ^he3}xj1SU
zU8cS1{V=Lp-es4)tN?hxVs49_(Vy_e7?AwsOMQKqlCM}{5dAFFj0|HKux5qOyP>DG
zg!VarX&_Ks0Hkf}z_k<|<>n#L=i>xLl15Y@;Q?g_pipXCpvt_Qq<a<DbEqFokrvIN
z+2<*Bd8LKmiz3ND0aGa}3-MNYkL}|vktKKxZLlLy>oo3Ns}=kBmUgZINFkJl6?zt_
z{Q(MI-q>$eW6E&i@Z3+iJ;}%Mec#2q%4g}%HGoMcWB&WHPQf`|>u?j4`*5%XMu05~
zOX)btfjD5F%N<y6!HjSNB7VR`w&eO4ckR&e%}KeIm`mY?lryKhBK7>JCP&P~dtj}T
zP=2$2e0eor$8=+<RuaQSvZma9Yl`RZ2P{If2rO@X5Ww~4`%7;|r|`4#O?bGj2v>6z
zsE!|F4hA(j+_m{82&q7X1Ae48S=%-*_vjXo;f$|6cE@<yP;U5fp`h_Q__J7adGt<Q
z7A^-ayMVILE6*V@|I5P{G&F?UJvY|L0`M%zY}z@rfdpI)epD2iRa%w%<<t$1Mw^J?
z^WgY6j^Rt|{1x&MAB>Is&*!_|U0wScnuUu;RX-z3^dQXJiBc2Yw`^*0A;7cfv_r+b
zn>cf)yD2;F{Fj$yi0HNqnNpPWGYp}h?s<vfk+4u0?IHba+lMKa{SNJZ66ox>p0j`v
zRM0S?D#ntx4f`ROnN<UOdT`O9UrEIv*~(T_-Akq7={Eug@^3+BM!D-#-Db=9G%t_Y
z)H)rU{+4CzSFwFIaBl>4t3H?A4!KJ7|JXx6?Q!$sy!mL!7IOY?#-XYLf4~T%SilAJ
zK-32~#z=B(VcV1Ax+))jS<}u2r&5^4iCPPd>&^__CbE$XTPqmaubl5O5?4QRb#kk9
zY9<+(Znzi7za3>7eI<5kH7QV_)eFv#1oLE$!GxfSIHYuUmA-8A_{`JlGH^9B8xuBC
ze;{-vB*5lbTH}E!2+g;1vA0W^$o0(tr=I{+boB**DXh0=o<X&tUcHWM3=V`T;%<OP
z4G%SJ=Rb21I=qUKv)EN>iSg$9*hK=RpP530{ED#VJGcf%UFZJ~ckdmKW&g$x%gCx+
zGRq!GwnWMnDG|!b9tqjmWJdNZQe+E}kv+2sQTE=#71^75j<fE&`~H6K``hpJdtT39
z&wtnLmg~CC^K*Va$MHViBeGgXn<~=Xn8nR6iB{qS>fkud<qJ$5qmK5WLQqG9wmn<@
zYmD@(f&24Clr*5it!VDP=syraxwwlS-yV_(*H)YsCe&XYf+G=eMp_2MywF#w!H}5P
zfZe~4e#<%uuB8sh$E#p~GNfvcQkS6L1{B>DWM5lg4qzVNNB{XzDJCYt4A}{Q)ADe(
zU{o=v36&83P+Xn`X!BLSLPpF4kmW@WGPDgDwm=TdBSAy{aT0$KW;dH4P0(5fkJrLC
zoWx;`pE5jg9GzSxX-0`UlRuaQ1=X*@)$2Sd$fzJ3CVdDNSS|y_6l1emRC#Y$&{(x6
z(NEggKLVs?Fl<xzg%mxz<~cA4E;QQ+XCfE&SEUbm8`p=Ae5>7>qPFl>OSV2hGz|S%
znls7f{%2~IKuOBSTYoM+&R~k&pLA6d;KKmPnkfLUHWE7RyC8q9jbwMqZ|SV9FA!IT
z)eSWfqq@#k>A{tgPFa?{rEs9Wkf+a`toPHbVvw{uvLvZVPk^ZU-i)Fgv9b0jgy}VW
z&Ze^tbM@uu(e(;k4Jyi_Ux%6dm=ln0#%PP@t|k~fOthLHvjskX2``>Pc&+Cp-042&
zgh~K>_&#4uwKvVP;4vpTq$>>2bHCov9mB^!?;S(5z5ziLZYT7P<gWw8botHOR<`1z
zRe)0?l`n1$GA~(YKiQJWpC%0Pp6`z$lDMN-IE(E8yv$VYt~x%{f&LAJfq0VhXJ49y
z=MvhBY}bDda`Q{$H@-|>Ygq05u5giHSI=DE($8}hJepRUDL=<S+2&d@@*pU{GUDje
zP=@CvqG{6QQ<dMfYbr7G$)_0hf(WnAwKGpMaYox%utWNqxlq@`uw)vjeIsP@mwAMP
zwra>n9o0d^PzlPqy?!pyuT*5trQsW~!?X2Fzu6~^ka^16{jj1x_DMf%MBhd-C4n*L
zKA_Rzi)3w0O};rXrj5$mAzY>H$^{<g7}T5=t4N4}^ZnZ7i&Ac=<m}?1`|g*Wb{-|R
zMm<_WF%;tA;+;Ue{}<?s2SI9VVMDgC${|JWx|r`5P*T@9HN@F4MXjR@`Bifgr2~bg
zx7W9dfDs>CinZiVu!5dNt9KJq>wtVY!<?g9v&1r*0{fzRyPrRn^HdqAa0jCcBZ$6V
zM4(@RXuUw3(XXQwvGd0)Bxq3tT0lsMxWjCLD`mdYq)vk6_znag?ji^}J$}`i&btT%
z1DfMW-#?}g)KM%+3f%;Qy=$xK3HlSMKoZ}guH>5B1*1DA>x=uOhsVM~S#h_pH9f66
zqmGA(p$f|#CMZW<oR+Me5Q)1|bNbkd6Qvx9(J+vlib6gyJCNNyk#wx`oPnjJZn27$
zI|A+F-}*#R8h_qMc7<3-M)x8NHdEg5h>jjSmO?c#jO!&FwP0lLW@PG$;oBaz5+hna
z=987NjMRvl;gChu`}tTy?dSTsX2vY)UG<-MSt$1fyt)NF-RaR;%!ZV&V<4xVORj?E
zXFSZ0Le(eLHYA}ezGrr18SUk$$-A6a_*qUZY=T#}R#(Hp_bo$w<9^!*{1{D;vm5Ul
zVh$V~<J=)HpNYJ2{+k}7x>ejX5m`7wRykENp?`c8MJeNZ;WM)O<YC|ZYxN!fwfgp9
zDrxl4u<R1sLm!K3*dUyk1(;#${1Efkknz8J08ebTKyj=d)$f;Idi^;}fmDKwImS;o
zMp^kTi4ibP)V9+f&Pi8D;6m44wIfziW}PdM^}Ud+6C&j+pG-LJF5LCnkIES=6Q!zd
z1h}}LBvWXKcl~;<JcAt+>cts5ZCzR0^t+?W?sH;P!nJ3812?5g)JQkY4L6k-f+o_F
zX+LptJbKU9p%hQ>MXDsX-M_z4<x^v4aA^ji_Q*2;VM#q&Q1(JhCw4O@SeUq+k_a=C
z<5Ok;X5eRG>nSMD61<5pQ1rS6Y&XfzvOPd;GM_;^%k~&1*(lb=Ekga4JeAgB+--up
zrzFxK^_ebInK+^M_^Qb^Q<yL#+P5jP_LcI1saq~#?;323{w^|a-EVRG-D8?On98xw
zQdR2o0wN5x*>`ZC#dgrVE$*glqQ9iHni_?#d5bu0v&eD@y<HZlC51yBJqx7*J=??g
zE$(14LP2(X&jxLOE#7Gg|NAcM@XfBD8zeiO(*(9H!Dsx92p0>|O{a5gO{h)DQ<nsI
za?<5vS<v_&J8{;jnruFX_QYmnd~8%8Da02$@B1XxmSt$9(%CjP<?PKklXV}Lx1{a-
zG3?4aXJJZ`kukvUJJe*X1iKPCjCq>SHrvH&lPbNAqkusQd)Xqtk^G*ne{&_C(A5EO
zon1V<N3o}nZyVw|M)T#tS4Vr+&ot3}aSLdaFmUfaBx^bcxs;K$PunnXRuFvt4#3Tj
zaBlnsBQU;5X4Kqos@>NE45^TpLXJ+<ar}0SnfEv&dO=nB=^2TA0<l=T4!QH9vbE<2
zL=Wq6l$Pz1>jw4gW}M-{*+r0^Q|4MpuwfFO*!f)J=#+yYglGk&zKb)eo!@sjw-qa7
z+*J9$hrrB=ya+oeCdWYfPmY`maO4_Pt%3&6|IvG8xMSX2DFLsCPfN4%sadq5d=2Hj
z3Q2D@@s2R~qzW=-gQtTtG(u^!2OS*G;`K!S`Dpz5Nf#q97c`ik6aBe`i+{o{MQL8R
zEPCyaLABA13Gi5qID&Pwh1PYEp<JZWxwvP^0Y_(xlI;<Oa5{Rck!9~n$fS)usH)F9
zEDp4qc=kn(KPT%}$!7iYul|#XBOVXmvyFuG$uAC1sse^)T&>};;y$zS>MF!ef}mF5
zCF<zkW;a|%%!7o$-YocNF!iNd4ueL|f(3P+av#C_h9t=B!mQv>mKKO(=dCBizn5z%
z?7X~_BU1l65*stZrMl()F#gvT!HPLRi?}Mjh2->I7*ZZud<#F-;DN#~1XIb3%3ajo
zzLWUjnVRMHz2R;1|J^F|;3A|)Kb?D2f8J0z#O6x{41b=eR6t<KZ8)Nyk#*wzM`zvQ
zhw+YYJ*nZs?@B|2MkTw*-}n5_5C2a-!*70)I7SRd^O*f5D!Kp3N%+mXGb;ulE44w#
z@a1nX%wG%bpD)lfKtn@oKP#yH-Mjo*2mkYTqW*vP^)~}N=bXKDVH~}vmN3*&{F4`^
zEZ}t?jP7m-u?@dacm0o5@?WrH1}%d6E9kyG^T*Xv%*BU7lzR=-5=Oix$k1Brl1%Do
zC`>@xM#^l{phJK<xq0A~Uz;}6H0MykX(~b&-49k}NX(K8fACKD?<))86o9Qauh;#O
zrIm*Sag^^Un|o=Khe8D?_PIK?X;cYvc3z3QdMBM6Ce9Y?{QY8h{Rkk9-cMoXIDb4w
zJ?D-%<<urSL38^hoAy6AVvp8et^eYP>7ejD|1UXW3s<{DIiq`#aW`^W2YdsemS!0!
zY=b@oc1DOZ9~p9Bj&6wfiz0@X!TZjDTFo^h3x)utz?PMb2MK7~<%Nr2%{+ChFy-_b
zhu?Dh{fQQVU-RUDzECEDb^SV>d+_9+`!ydXD3@M<^WqgG?e20`J*585Km=_BWW=b`
z0AwrjoX<G)A%K8gATZO&q##da$c6%OIu)8@>8R%pLR@7tCrJ)MXhklkPY`7*Zi6zb
z4SWKPKuOh=bj2@uNKu|B?NfT=8X}yV1;6x?d>o<>1>sex{5z|cPPQ<L4|hmLYBT_a
zl>?HCo<Z-f_JZH+-Y%fB2oWq=<J9a6CGDzu*;=eaL2w89|KdxdN+FxD4coz9H9)Zz
ze5T*vqs(rvl<3uB0vX`t2tYGk;HRY+5?CQObixTu#cSZ>EY>a)zPnOS)CR*{x!Q4$
ziY!t$jf^wm7koGs5yBO9_Z;$Fq|XkD{KfWqe?ALRu=i<H*|d7QLT{~}v#*|otsgDG
z=kdG8r3GoNwd|b2{@gz4xiO(Rj)sVxe%D$$VT>a1NXy3daJJW@C&uwOLg}IrDEfgT
z=`%afXYVzT`jRACP;_vNpJm<Ir^`&0Ec-sITw)BLpgc|e;R5mr9-8ei|LwT=$0^8+
z@CF||ZvV4Dy&r{9tbYKv59j|t`cfH0c+(?9yNna3W)Q2dmO1au`R??vfWy354Q26{
z;ktsQE=*%lvuH1-)*9QLB-(3FtXLgv#){gameHTl6e^;EOoyO=wHlrn10N3%tNGnM
zBP~>u9IhxQq~+#4JJZI^v199;<c2p99weRDZqU(PjNdbSxK?D*Gh8I}LDb&)a&$+r
zK1l|Z`+T}J9>KTdHEToB1FfCYv*mN6A{UF3TkTD~WuEzEaeplnwhlJC!V|_k*D}7D
zK24c@<x&Zg#V}X2Z-LkSw~%DKQ^!)%IOr3br5DTZd%>atRZE8s60QBz+rJr6VAJzX
zd0JS#dit>qX=8hDTfW_<eZNvO#P;C~Pr;YbI@YG=dOwRMr<#?@E!6A)^^yCl+vK7M
zrB%R2LgC!$nJ)uJn<*)UI&u~>4D&=gqP5HQghoV+U3289_epnOF0@YmhHnKFk`QFm
z>Ps>IStW`0lBk0iA@~$u6S*f4X>Koo)G3%os84ti%-0>xKish%gz3U8AfS<})6K(u
z*+?*+7L+S*L5+BKbr(>MSx}3-U6MqEWOo2@IRNLs=;Hzd;z-9dOaOGN;yXd#VMd<=
zY%7m^=?R#9X#s@7f_U+$e)FFnfA<a%x!>9T$+nvMH?|ckt6@rrKqNf#S`fnLBypO<
zVU8s5PF%C(7P0|>nL#A%J3}yC`d|cRFi%0oDi5^?5_bCDzy_35n%eL7(Cp87^u3X2
z<+s2yF;*!6H1Pc0mhj49&trk6%|I$MKBUdA0mCe-O@b!rmqNAUK$bBQu!h~N4Hni}
z@VvfATniHtPQ*Rz*y#(`VB?8yWMu1<oa!nJFtDK7lrK<VPzBo}?ip~JNFO9KC!D4n
zqvv(UC89O&heS}XLG_qtBnOCu_W@MqEl<zgo=JPAbDr1=n9fVPh#gY8w*C#h4QSNH
zOp5cNa=T*~!hiRPya2Izl+#BET<IZ)JTBBdy-*CQ0o<lmEBAG(f=$EFOHIXG(n3I2
zo1ZYueEDb^nrTiL@<c%}V+%m)=aA$a1bFB>(awh&5`dIF<LCChp%*NK@PqCbShNo#
zE{ua)z?+hDI_}!+O1R9^t^ydgerinLDcK<bCI1QTotw>Z)?L@n;ujk<yRLwjSH7b;
zXa}-sAmU38Jx#sHRX)!cJ6?;0nR8n~q_;TT@!`lSkf_4TZ0E{vxSs95FlM$+6U}nt
zUBq{h%r_1bRG8~RAcFma!{uE1b+91)hm+(on7pF~gsQL8KNbSZlckrQcBUQtEx7|$
zOv3wsGqu6k6^wy)iMTANr(8)L9&_fdz6`u{2;VdU2sQ74hjtcl{`fTcX`$}(Q)2%t
zx&HczqAybX)!j6D^0Tme9gBg@4Uu1wj$_U;h;&19(RacK*bb4vB9KF<R+<M@kfI^4
zh@FJtBZ=A;r!Uw;XBlH$2A?d8TC$)fb#rMKj&@1vLO37p01a@7x-M1EX+FtsCYY~-
zXOcncBEpsqSoby^*`@owBXj9+qkA-sh9S?hW*NTRGvr8wL&LmN7b=;zNCL5m4S357
zC=o;P-u=>PDu6{xO52RUm#f22Vjn1aD03=5jEI?gu@?6@=pT4r%!Y@zk9?d#{Mcha
z_lAqzeOs^@S^cq<wyiN-IX+%tj<RAcq-k*ko)XK1ossypqu)R&W~-Dh1N1^5+sO-q
zSq}9AI2r{=Go1IvnAQEh_bdP@Jse03o9?MjgcMpNs}El->PFw@d4z|vbzIYLNnD(O
z%2QqhE`mTP!!1pKqvWqqG##z|;iE>4cr>Z&iuVxaD$M5s{?{P&jxwx+?-mcyz?p{E
z2uE`kAvu2_LtT+D56Z`hm39Y+(<*heyi!jLDyo7@xP~ca{W)uhd<D`@%$<{{XRss;
z&tXahqNKdeVZUe1vj5U2keeijCyWadEXF(j*QX@43t?}*S(p66oT?auN_4^#EBkwe
z-Gae17Olk`=nNn#yz+T1d_A?;;b~#<R~95do>f%vpIyD_r%25>2FJ?y7}(D$z(&$^
zu4x?c0YVEl_jSO8rW8pekW!DKqW>6pa&1yObQ!;0d#7&zHWK+5%8Pv{-B^H=)UOt$
z7Z2Hh94KajWLz3(TZLBa56AW0A5j?jW582)=T3%4Ts{c-Uc9V#(5rXCO%>^9-W9-<
z8BhbQ>mt$h6`GQVyASr`A7v0TBJ@OcNEEV^>;ody)fGSDozHdKx4M?wGu-d~Rz;(H
z{D?NQ{;x7SLmabQd^xkU`F5^*yo+K!`Mg19BZCeOc05;)Mx~2lEv+A7D)X%OmUi%_
z@`ag9$Ea-Aa{4csJnwHW>)zS+CPt68-Cp*p&G7C}xn{h1oA+_L0lz|(kCAXLVci%V
z47U}eh2vdHR)?b^SnZI=4E1<6sbix7)!0I0XlJ&Kb>*6;5qQ)pJX4_WZoRFbbZ65x
zKn0JYZ7}WJhPsJR@|xO_wDd8LWJ{yQ1<Z;f3F*}M!1_ugr-<pedWIhl(ROeBU4(9>
zI3s;pAmXvb*`?cLnl5ZH8_Eq6nTZ#I9tRqIDCWJVVjgQxM~q0xH?Iega|&-L@MQQg
z$6_f@%p_%7D=H<X*U11KxRL%&@jl@Eu552Ndo)whbp%DUWN?D7&m?&_wmkSjg_I78
zCwQ7ntH61)irIxttD{w+D$<3GmiHzb`TO!^#?~Yn(*c2@Cr_o*h6N~V$YBP{NhO5N
zBv(464;EhWF3d&S22NbI_&4-P&`}@EiKqF-JsqjXy4l`t_q1s?89FsrI3NkSr?#yl
zngrD%%G1ZL@~x?^#-w|whp%?@is7w5v?v-$t2$*m&2@jD=CKR2{+$y`ibSvP5u!g0
zl;Bjl_CCpKoYqd9a?N^8@Mfuk8BN6zQpOTu#nh}U{Y2G+?Cj7!<~WMGa>cvCV)^3c
z)y2&4kuck{fKY0S3jbEWQ-t#LJ9hNXI@F7R84SO$@1Sv5Qnt#;UIg^@FvZXj<zx6@
zm*eEZSle6L=s-)cQ*Z3@m+WaSrm*W22BXOvz{R3^F2=}UQe$w@M;<-PBk;09$-f1I
z@vQ&UP?6cQGyV^DCo?oNU(8<@S;S5`Tu;S3EtXPqonZ-2INC-dc!Fq(3p~G=5+)jv
zbY9GMB#(#k<&g6^^JGc@@6uV9KVBi?*T2F=eA=9<4i^tA1rw-x-L>D6y<K<<R^JRe
z8IP)3X*no}2;Uneyedw5>=T0zrIJJy$<QIfyrPv7SCS|>x)ln#=nTy>0k($=pV@)2
zXVGgRW7x*-=kT<vm{4D7p*ZAbpT(HS=Gy#1UqZgy2?@FeMO0y?Uw*b-T3!Iqu24na
zr|G`8yk=tT8Gb!jVmTOZFBW2xGprx$Bp3sJ$GTr94@2K-*U({3xJs6|Is)aY=<vi9
zqY*kh()#l`yhVd2&hAi&K6JWjZ+FANkd8dpz+H{I<CLjQgk8?{FT*V=?fBnT-132J
zF?#@6Hl>DL%W}_%!u~qi{)sp!aw3Oabo^bRpNAbF7Z6n$TaxYYrLRIqJ^U!~wYUkQ
z?FojXgcYcx=x8UFKUi4ol$8gw1aJ%US&vJQ_?#-kth^-g6#FL$YtDY(U^-=(rSIt*
z=bJu`LjHwy3;~`janCBaQfa<E2PcqPU9Q!%HV+s)Zh>7WiQ{uL@A24c^~Fy%764Pc
zhp94pwBV{hIZj6Uu(iyqM-$6PHtvjw6k0Xx-n*4b$`h3DR*vE%kqFxZ5sH+oKVc;?
zeRxMhhtk7@>hm$DeaU##5k~o^#;>ejMrs+3B*{HZ(>&?A;;;om(9>9j#Oqq5xN7$<
z6*3(kzUmtJHp6cC@taz#7nWkN-8uRRDEC?cAXdG*bbBT2=e=9up`zePNUADNvr>(`
z(6kGxEsN?0IySY^B`W-ukPyAv)<Ab+LL^(Wb<d{kCh{bfgMBhbTWN96J)yXXWq{i{
zGZ4o!!^0!F&9&Cxh6{wQ$cLI~n-()@G^|KJLt1WT=EDw-RE}3?{^Uvg+MoY=;U!D|
z5hJRnk?3zv9+Pzgu}n+knHvcG{}jI(w|=tb_xH#Eu6f!%H~x_n1zNgfyWL-g_F5LL
zp;Q&lRM?S&{c^WA8F&iHCiD<bzdc=l7CJ=o(aTCbgo-<=I{lE>ns0~3dDFDPCQ7Q^
zPO&)^#rKAbLsL_wlD=HUs-uuX+b&{WwKmyX=F=Mj+gH7B$fX;t&x7Xc6Q?>~7n=NS
z7r3&2dh;X=Mo6c&1FKkpfMJwz_Sym%g=K6PmomFoS}r3}s64qiL8Va9Jrglew3M&F
zc`y`KnheRv;9$G{$>zERCfHY4oL%UK($zokfpp_lL<mgw!2pUTGP2oL#7gwJAHlLz
z$qX4;`Fog}Mn=4X${?9h=aa=rzNr+hp4*8T;Zi*JlADg0>xhsO#CYbqs6K(KqiYwd
zE8&recSobx+|*WH`P=nrWo1?<lgY@&{9R6wkX;TW2;<N!q}#+;XFXB-_Dea<bqgws
z>yKbV-$Ep1kcspZnV!)c-4}eAlPvaWDQ^~-Jdwaj^069z82`#5%TvuR$n+DZn{<hH
zCmS~#=Cm;a#%s39j1~)$S-l6zvb2cBz0~W5anqWZ<0r~}vku0;A12I4e_DgiNv9_(
zXTJD-vgTv3N++pYPI$O>Jyw^Gjo4Occ$r6;3BN7o-ZhJ2EZe8wRv%|fC~S?`D!h|*
zdX{Jrliq8XMn@DYeyHrf)I7<&p!KA$BqHmnOFAi8&G1N7D_d6Sd}SzX^XqIr0sq7>
z|N04xmVk05@v`ytAE$wMxcF&rF+48^a7!auM;gNiHAiq`tT8b!`Iq)!Jq=)v=F--9
z0Ea<9na>H63X)S2D}IejT1mhw+Y~S~V8f#|BS5b<HK0(m68>bzu47-Qeds;sTgbGd
z1<!VEg5?}L<c^u~#Ae%$_VxStlsCSek1o8k{X|MUBfNp7>+KS=8-bS+ltXs`Ykmuw
z7K#w2fC-{Q%%o74sJ5(!c=Pn<14s<@j#aPJ5h8L(B_iE5AZEvWRI6Hg5A7ja`G@rR
zzTs*Vcdq36HSABk4H}?1OB3H9v?0fKM^$B{v({4M)UJa6jZVr0QDd))kfOt_;hRCo
zod-W<N0Kx3P5LCeJ}IG>*=~@2iNDd!a$j@hS(Bh{zJV{k1H%lhfO$;5@BA|2RC<hB
z@S~JFV*i|xuApR<$YSoe_<#mroyX;hE8Ava%rDKl_5>bmNz^7V)&6GWjDepjE>0mp
z`Rm@Wg?n<DoE_?-7{-DP`$PR!A^P08=o;`c(`MJ)695)$(~@}amk<kMKZ~XN1it3)
z{N5bPmriow%|53&e3R0&UK2f8%dB~~iA?L7(9|01y&Jk@rVU1O%~=?2rEV|{bJrTd
zU*x<`HB-7VuvMrYd#xMx=yS!1eVz6SRgz8oe7R2M3t4p|1BC-FB(qW$Mp<=wR?wTX
z`MW7g47{aym*MID5{T64Q$h+NUe0WBA+ph|Cw5ap*b+CU7C9+!)tt0Tur{w53_FQ(
zl1Et%TMl<amrbS~cmmmZOEy#{-8vjHN~&5o-MUZDq0fE1O_WPpyxCZCq?(UsOu79F
zA}lIiYue|w++WRT?cWkQW?vIs-S{^6KGxLz0-BEE?t!k7fo;~IxV{qNHOU%BoX<Aq
zS9DD#qN{yR8oK;hzWj5?g`gaaW%6Ucj!&{{^qCV1C8sxf9=%A{#A+(0m5c}YS%kYb
z!DTq`L;5Q!RE-`nr8nki`33e%*jnhuwnr<OplOVES#Omk31z)&qf;WwPhLKo@Aq~J
z6a9)dv@(=N=qig1<_pA#cFmnP0-gb`?*`Z&z)-6poE1kUYo2z6%VDpT?=HOdY016{
zGIKvYVr8c=+*UmK$IV!j7MXP}e`yusEfmZ@`r=}w^GWzb4SY&0K1}>J&4+?YuUvN5
zdQwup#J8kaUNoInFBuN4OkjQhQehU15Ma1jZOY!CmBFXe9<(1l0m2@O>D@wg<#LO9
z)!u@H5AVy3xIfMuD>5A};@r7m7KeDT-44r(r)nz~sp9Qj8V%V*WY~ssBBP)0(p27o
zjp?a;?Rc(8_|LLr@XD=P-N<D*HQe8cnXV;jAlnS5$a{xB#DDCVsLz76y6O$xnYYg>
zl#I8bw&_FGF+O*B0LPOB8;Cwy(Ji>9^ocdeV#534%3gxo=9F94&%T^_Q@w}Yk`p|h
z!|*VWgpYxnL-tCG-}ouxIQ=@Br(tu}J{Lut98CxEk{m&c?GsdDrY7d-<J9RKX=C%1
zvIApw5ePVas!m3Pz-OaIAIZI1hQ?rEtfZ6Xm44nrQ74_}PnZ+>RsEV!c)BZ>lg|W^
z)*~B1BHRWM$VqQHZlH+AU&BHfew2eATiT<S8MD4XP0Pj?pY0Q!#Zck<HY}~5cW!GT
zQQ}cM3r}<%M{U#ULtuu9j7E1?fb$UX67Gnjw^*DX)|4Il%`XMO_GB{)Myxf*fK@d0
z1~$hrXgFhxhFdy&B}(Qt=~w&-=7~mIyB9CMrp;SILU)H;K>aLj6v29enzvc+0K2XE
zQtLM$w~37I2}VC+ugn?ta$(N55KbTuk}a{w^Yp~C?S;dLX<Y0Y)_x6T84yaU!1fMP
z=~NP_fKtW;k;Acw7nZdz^v7!h8IKDeOwv$NV_kdFC)j-O$ocdpOb~r-u6%=u!`uAh
zxy~6_*yA=w><Pc!U#EF|IcbrZV|qv>!<x6H=H=Wzm&2Go9k5w~(KQvkdom@8o&I9r
zdshiUF{U=I$PRu}H>$ENp~oo83PHzq6U7E;=x6;oPD%f6m`5fLkayr$QPrRG&P36Z
zp3jnD8WdZKGuAtihVOlpf__J6KJk&<$Rl56Z0h`L$0EoZ6uIoajtrWC=ISD-B8Q8K
z3{IODZT+1~rc*>;Sc`dq{JC~K7ZG*7(K#s!$UA&n0-<Ikg_rb)!N)3l1n|uo@b-AP
z!uy>%sB0h+<$c*7KM!XCiC2CJrgdXr&2BpE!{`nXF5mcSN&JcOaA?u<?oQ+uS2eY$
zLlLLn=o(_@NoPE2q!t<8GLRT5I->bBSroOOBx)0%98T1rGB89++pEa^47ajl6j0bP
z9oMDVBD4Mz*9#<1FJXMwbF~Q-sJ(Jq@Eb^7*S=tMy6Q~`uhg*9#3VduJy#sUBo(*=
zo@Xr$q<S`=pBwLW&RzDBk=pRx*N?4vA$cqIT3fMFCmgIzAS^GjLA(lw1x{1zQ_L^J
z*dSKkYYyq#5J{fsY5yl5jfIv9pOnDSJM5g-8IC$!1^N3aO(%EP{iTuuA78*mw*d`B
z@@ibh0H<9XK7;mjm!a9)S~9b55DXzeVMw$Gmf<XDQ-$LSf6)eb=<^4Zg?h319cA)>
zDMZdn<pG&+$$<JS-`UN;5|$Zztyyjf)HVTynkJC`S{5JEzc1KhFQk}|TKtlnuqUm{
zf@+y*fVrU=<t$N%HYbrW?{){6DMp>llwS6@->m^D@NNw@ZLdOQTS8kc(7~X1>}X{e
z4@Khfl{$(|uA5MU9eUY>+W0ue6H?w-Iq=)j(wyy^g%d?SD9y}qWxnCQiJ@rHfFo~1
z8<EIHc>ZG=9O>6LSGFOpZqGoj2gh5?d=wG*2Jeb3Xg#WY;!cG~KkilJ>27Opfw$(4
z&DRR-<X$a=lbciN{A)b(Z?kN+2!Q6J2F$Yge#}|Kt5{Yb)b<66$be(JUci+L0m;dU
zIv*muQWJkh`mrl6cn?PE&=4y(Q#*Z33LCX4v6DA+KWlekBI?-GDYX;AEqc6{sp$n5
zL7SQO$l4UH%>^aGuIW1*Qp>PXS$gorg3L--Lu{R5zVzoN;7%HVhs6iJK?DIOoKP~j
zB5XV9OR;BfXCETuT@Tu%(B`+8Q{YN_&sm>c6_1MEUqZ(zM=e)}CA`&z2ek^j4b=@E
zi{95!{Se(nVw1?(f|ZPK{*IC?vKT8F<zyIa<|<f?5X|`Vj=wG-PJNVNw=w^z&v9N1
zN{kihitw+HOyf`Lv3Kg|O+@$KDM1=I2c~<WHKX(0sR9wv7AX$b_eeXr!9G_Au)JWZ
zMGOO}foLi-{~v;Y;J^(odae;;Kk5hm6&7{3@iyvAlPkF-oxT7Ppc-u7b7Y#fGbIcG
z7Ckyv2zQs?!((OKc}3lCW96#9MFJQU-ggKU48CJNtQF{)PsE*&C(_Q#N;t)89Jb;_
zK=fFGTxmF}-(IMOCDS6k5z{{N=<wrtQd_by=lSn2>2LvXssO*)=6Gza>oPcAyhk;s
z9O6M;BMJRp{?R1}G8@6g5-vR24Y5p5LT#Tee0~aga@?$m+WOOIYa}VV0hb%S2Cz=*
z`||@_aTJU@H`chC=y0(2Pums|!UxawK(2@Fr!g3<EjQQK=K<UC^fQG>s}$gXjpl0R
zX-}?0y}F%K+5&JudyCwu-!OE<G0Ef-kDzlrGCO}k1&R{bJrzOo-fw%=t5*HFG6Dl>
z7B#94{9X7@YRi$4weG%o0p;ENmU*(|5hY$9y0Aj@=Cz)Jn2dBTCEmz+%mO6lA=yA&
zDNH|kiu*>cB*TA^$`VO_kjj+8C(_5>OO;?9XXzA_ac0XV86~*Dq*Fx@5XgUiDw#`p
zg#o4+tlz>J&<<Cr*KT{sT-92Uq==HP`hrON@5?C%8`-Pb#F<0jQf+Gf9NO(V{1dc=
zOpj1lMziE}qW2}Gh^HAPMSpP0swl7iH>a%XBf=>QzvuQHNO|wUeU}N%+qQSTiuy%}
z&w5ZrALnyv5!_aAamzkL2i(*vDT7T@e9UxEV2!O<VKtj@B1=-LfIV23%%c_5O{{Dy
zJULu1g}}>I=~GTc;WW9XIRl!ZD~vEyLCQOeU>n<k>cqTZI~JPw<DMi=Omf|&LEvyn
z`!XcQ<JE`l^rNpUytrs(x!&3kK=PqjN$e@szUpKGwaa``vVrjT=eOcKlL#W(n^Y9d
z9Q=2seTxLX*_++<pjGrr$@s{ga(eb-=mm4C5ZTr`FIbjWH37cW{=}InEut70Xl_)0
z{9@tusYR-3j{9;GWw!==T)LiqJKPU)aUdHo%>mOlI)8%K=OYt4*i1I3d$Qt+K2%LV
zfJGW?OFCc8uo8c)+M_o%;{(m)rIy@O;7bY}+q@QQlxN}Omq}qLp{vK0eU5aWtI0{Z
zv%!6ft)_7I6mhL<t$U3S-7G_fjQYMqy|aex@in%sVZ3n(v@^+_nPWR!?VBg+>aSPx
z*^+m}{wllw<poH90kx{RX&Ch<@n{IIfm2pZQ)n>$0RJP)nsMV(V3!wb-#Ju~vv@gS
zwa}H8tob??rB<cX$7;v*ibM4FWlU6Ari?thcnK%PrezFxtl1T{iWztC?-m%y{_m#X
z$jOWg2o*&mEWi4XPX+^{QdVu>moM_J7Y6uX>^cpeOtg4r(VwcC54O9~AfKTxkE)t?
z82o(db_tDu)r(BBx?tIxh@2)0z^=Yluk9L0{Ozt}pPQb{or!8o!RX+37B~2lWbqgD
z^S2%)LlOBDCe8Une}0PN*m;L^#U}L*K?AK8;2PV<VSa-k0vo*5#A7o&+;ns`<c4#&
zMGn6dm!@0|X*F5-n_HSQNhyr1W)Z91j2SSbFevfsxTfw$5od-*i3YL!?S23I!+2>R
zG&@eOd%VBy({t=R6&)B~8oiG^vj&{<Jxy9hvIY;R2%aGphN2BwuUX(Dp})TA4@t4;
z|MJy%`g&p?o^?(pe-DTMpNxfnI}*`P;YhR(Oz?g3lW!n?$ijn|S@0R(B=`A?Z|09a
z3xEHcXihF?szrZ)ym7_#L<fBM``Tk1A;10m-pJpVcxm)I*!v$;hrf6-;>fyWM^Qfc
z<MaLISO4{YiEE>Px5kxGAosfk_}{nSZ+C|^9KHY!rs0GCm;XM45MlJ;*hl~Nkoa|{
z|M>z3(GH30$^F(#>!1JZ*H7#vpo%X~>^T1y*D5g@&nf=w4!_CA8uub)^E>40Fz!>Y
zQq?d`dz($iuJRn)L@jF2YF6jV{rs%^rZQ#2gkOGd-=_Eu=``Q%#lHCiXZ3o021BV<
zLG18pZ#9I)8#hMTN)0VR@OyRpOChz(j$b*zrZ(8ZrE~;_++`h5=6%~q2_SnEWeGjx
zgqqwB(dRpGAtPTGOvpTmS(w#je>aA+p(q4w3AMe9B5j?%+4ah<eap<+C0JG}JWI)z
zPgA?M9{0P&2q4dYu`@96dyX7&3V$S?sLr5zS_SC`ji8}e1M<42Lr1|VWIeJV$zdt}
zm1~~rnA8(WxiR4mLQTP^nCktAo`Jwh96_K5*yMqv*%V4+w*BcD$=RDv1uTXdUgd^R
z0gN31_RG7{NuUzgjFfnEbl1i=?gE5HE;7kdYFrbQ(;S40o%ZFOsM}tPHxA*!ZZ7!q
zJb>B7vz(gq?=LMmQ+;@ywF@*>r8T7Z8@w=<V&8khOq}fC$wq840V)sw*PZ^?7sT-V
z1bNo|GOzQh!>a^N{CUHsW0zo%Oz6tf*aIWGNN|vrQO#D@FRO+U2%&lJ0@XPc=pN^y
zNfXE#h*5nsjw3c*P=f9FT>1*Mz#Cj;4cHzjGGQI(cCgrc+*YR=5i_3W7p>y$=07sX
z$lvhzQvN09iD~z%T^vX*xPutGoPt3womvaNPXfxPj_}l#!!zO)Y}(^id9_EQox${F
zX&^^is0?g$W!3ba&%|w2hTQAzeP|D;8R2IQpFh95@AoH77o~lf^fv!m)Bm=)8Ni~X
zz<T`Y*M-?7@vJ!>KR^$WlGZTpe}R!N03uj(9IrMQ#GW;^po%S|S?Ph?OuTjXm@xWv
z7;QvDQd21hwuiof$Pu4Ey#O@#maIJ3!Q6{eV+EJE3)GC~fPP~sXr|L|2upS2GM%?r
zyPKM}y^70oRlS?n=PWLiQGZ4rC7c)GDrKbOq*1Njo4-2*j+|^^hEsrrNaq*jnN%VX
zT!zoPZY{cS?>xGg3RbGmq9$@be$BpDtDD3TQ|*qr#<r71_jJwd435XU2eMaKpV6&$
zy?Pz(@HG0?!UOKci_#1C+v0wI&F&*^r$yIejsAFFaRThTMyld7jW%gfs*&J**8&gq
z-F`tU2z+g13c2C80}t?E*Hx9t|LrWcMD;(N#mLx79m>|UN3{-<TXBt35$}ErsuAe#
zWln&#Sazxjt~%+-KvTagg}k_jyS_LcT-=33N2nj<<3&H;a!#t_=zk_y9&Sa4onXIo
zyujLd(p~<@!N@w-n*y^9dC?=&XH1?sjrAUJZVu^vxYlv@cueBMZQD*|4I!<n(R_g=
zHWMsME^C_!!sw1MYn|Jy7b{xTR-SF`V5VoRMg*lzf2FJ_{5nMQ$X+_GU|Z|(c6%ia
zS98?tVngp)G^KgajpeDpe0H^~7DZs>wShHIzKj)@BJVE~yJBlW;ND-FKT%q?`_ZPx
z&5_9Y(#@PVHZsC<=&5)|`&J{Y+P#Ug388_}dSd4ZT6eC}%g?4_>!N;-RfmiJQy~R|
z0XB%u$d_K0zc{LV*&fXdd~|*yS1{9fd5lg}3zXunW#CJG_h4tWnEwm74>5tE>vMC^
zLELGmKHSnof+J_Z(>WXzb1LLqTCC99yu+yr0Xf)Uu{V&*-f+s|(lr%lc<i?6nol0W
z-0hiBTl9M+85%OVsf6w2aa~|Nv40<WnMgzDwh^KI6|lYs$n*@LT+SzF6`~P@wOD7=
zBqii7Jju@xBieA5ap^q+9|HA<x*)}~kh%1K>EmVCFZX=_a#>EDzk6-WqUKR!;WTVE
z1@7B~eNTX4Fpee*4!R6tPQXgq%qL(giw63~%~uKumdSnqplATUj^Y%ERKehoBS=~~
z-n5a7t=(H4Q!^-8SKISY3tYJPw<N6NYR#qr{$w7YJypC5XhY1X<NU{3L25XjF0Jpw
zl?a(cSY(F?|Nk!*nV;Ja7FkplUjt{}n^?>8_pQwWVeG2ec}pq0#rn5lyc0`Pn?ZWS
z-U*-`Bg9}9wqs|wHnVWaqIy*|KMZ;Co`^65%8+3Yx1WQQN{+x18akkGc*q7{dvO*s
ztknt!3btp)F!3xSwA?;GuHvS%sF&D4R`p_<aC#qx_9Vf%u=_q=Ic>f&hI!*B+3ay5
z0gW{biC=IE#m)o$C=R$zU%SNIp+9-hNT#;4j%R)V9I9~pzB|ndB%<f;jom!2n`Ww3
zTil~6<h`Z1==cHfxfxJ-tY(S9STQ!8_hL10wFH`Ya3_UKfN}Q}@oEw9kT{^A__m(#
zz2yp#2=l}~jEY)-#B03l@Q?&Dl548pO;un2;o9XG)kZar<3v>d!?mjngqDkLc|ge`
z_}nRWFiBJ$61^5%*X#EiGXW^qya>)Fe(#R`lN);tCT#7H#(X;>tOUk#PUz7lO`wU1
zkkS5^ahJBwmU8$^CSJ4IP+=x4mM_l0l{HemVgG6P_K`>Fbu?BW5MHr<6Vy3|aKBa&
z<1}Hi?Ykn|5S!w{roIM*3S)akZhnW*FDuJ{a(24tP*hbx;i94OG;A-y`SWvCE3$GY
z5toe>Q1)RfP_&4U$=s<VxLb+nx7l+u{K4P&sO6}zc4AohOR{uY`%0Kx^xr0qJp>Zc
zS0otZHgQTmh8fu{7sNoVuK*GRo)ukb<WhYVh+RPGiN9w7KE!uA=K>MJ*$rR3Q?pqa
zEd!s~nn!QjJSv*}Bty3eVrv6|SBzo`B^Fzump)S#oEzu^IT+RZy4l--DFjSwmbk5I
z#8K&tv<Wzith%u=aEOa_D345LQueCX96buu%4sPa>G(itVX{FNw#$L$YeZrPdW#=%
z^S;3C?GXZcT@{GPx^DAk4Q_(GX2$JcwK<}b=@yQ6Ji#^2H-X*4WRxy!a%;+k28KlT
z;uXrl6tOa0RN$Xb>@0<o8&5!A1^Vluq5MwKeJ21P%Kf&IZc@tFBvB6$tCS2CX2EhE
zMi_5z-thpltC2_;Xp!j5o$keK;H1S9d`Nz6W;1(V&4j8+MxNo#s;|v<9U(X6l}Z^N
zmZK-AO{Phf`6~BUD!|20WV0(;=?%;TI?V>(rfuSUHV(c*=46yw>+6!Usmy%ONG~pv
zwV)|qwY~D?-D=skoVt+QHx8+*+Up=+c48whbS6Jcjd!s4>ZgrF*nR;bbvmY*!uOq;
zCdz7E<#t*jNBErm@dSq!W6}wRUmsi4=J3o-ig<uD+;t~9fB;e~qrZQxJ@8)*vErHW
zG7Bbqh&3tVZASKOZ}2Isb#9-tzbcl0eL*IV;j!8A7oPOfzKe)aXcIlBuYY>-xp92H
zV8fM+&r;XW3!PHOfnBSsdSe`<0}55;&R~fl-zTyeXvm1`ulN85ua@4QRX|+$F^&s|
ziy`j;I4`!e5E!s-x*JZtMns%eYFCp#N+(I;<ke$}N;#u+u(1TcZRUDxY82fEW^zXM
zURfbxWoB@x;({+Np@od{H($ILz;VG;^l$4SRJ|Ilmo(-lM^9iQ{S<NTnv!N%@@Ffi
zjLs7pVDTwJSMf4w1`BOEm>6=&;t{M;NxjZ9+r*1Dx`@g}%)D86eDTiQY$r&<LwgqL
zOVD65qeds)&*6BOl)7XGl74?Jqez0Easqwk!Xu@XWUV3MY?2PM;T-zJqwIM6$Tugk
zJNgBClGprbF$G!cV9pxZL7BsuQ@Sl!9R0u^BVo)+*Le2EBTmG%SnO)O?WvRcI?emH
zM6mZd^+F#dp68gUtH5>J0xWRfboyrCbF(^UvKKP9DhQs1eUbbUej)5awb_W!F~eh$
z9p`Spx~oE1jG@FSHGr@R`rJ0&?HdLVGhMy=;R8D4(`W60M=lX>R%5Z3yPzF-zJl@)
z_T2VzQ7eA1TJ>m|cC$TJCG6J2+jq(8{!`!e-|eYaAW|DjJ5aI9|1rOb_w%*kV60zP
z<*9i>CuLSXn@)FJ|0MalN3?ejVv|VxX;y3{O7i)0y}n*X$3#yogHk4>b!>|MftN82
zNA#Dz<otz^5J?{bl!n9xmW{Z&`TSZvaorI-n;%|X1~%2<4pP1NHpqQd$6V@Jikg1A
zVHYb!@~);p*}_*6p3!sz^~*^M$kw01Q3a>@z(C+hvzlqrCK}<rSpw9Iq}>I}k<ymC
z@4g={e`q7ER5{xG>V`0wJCdYGjwox^ejv#aEhj53QYpy{ej|CQ7%S?HMa6@<Y~UIj
zwRcNcudp0t&{SqMu3)>mg8NCCcbd@LF4mF6a>1cRDTCvL?Pcr*l;c9*z^Mg0D)oY4
za9r43McGciBqx7XKWq-z$)R1@kK4EVZ?_NAy}i>$qQs-vnIBz&r4#TnhWo3tQ*aMw
z%7@_xT>JIwJF4~6MBMGDjn7oFtMU<6gL40ehIkCYEtbFOFa5zRppV8NZ_3Bf7qmn7
zoVyWzRO#jq7T9~d<coM%`9-wY_7GH7An24_8~vnA638It70+&%`>g?sFZojCd2o6n
zX4BIYTfq+$om%solX7}T=&EsL?-oY~MOSF)WeP@<2L=JG#8QGD{xo@<2ET5K<O=Q)
z3$2=+TJmPFBh*(t8BBvGOn=_HIYUbfFLFwK%&uErwFyA_xhfDlv7D@TYTQbptjKF*
z<yN^vCQIq{K#@gw4`PR8iqqC%LTVEK`s3AL1}BeT!r87+>%;rQIdMFM!^zFQh>kr$
zW|aHKEmB+D@pilsERxsCUu;d}EjciU_-Y+XMOo;JHW!a@X@PT8H2E@fB`z^fhyK}t
z&mswm?OG%qB_D+&hg86$mqNoY)fqP3=`sC*TP3eN*I@Gd{V3MAQVj1A>xVqYJ65ZF
ztPtng$5uvKS5|QMkyLQLDYJ4`XmjL^)<l0NrGYvcW#V}sJ%x+)=dX$T5(|O>?Oj0F
z&b#ex1iyg5>^pX`N@WnybLdq0wO2DF!%yrJ=T*L`2wb3IIVO)`t~O(Qa~rKzRrWbs
zKF|3#%GV|r3tIl{L-yP+T-xR8mq$+&{-Au3=?A!+Y<mvvcWzPfYqbkX3SyVKU)Nme
ziG7kWp#;3s7kqClzp$}Aegb{>bvWS8ccHgh{+07JAg-Ljh?d~u3ZTm!7DBk}90=}!
zoS|W1mz9!2f<cg`?8L-o9wZxiUkllBODUt-@FH*s$^9+YsiV>wGkLCE4W&NkvDck&
zz6(w=6=f77`V<F~C|c5&K3P+ZaFM_Ds2{k<X1pW#nhGdJa=h?|K4n%`&>|$x_l+fw
zai_e!dFH0CQnKdJAOcURl0EX#&|Av$2F@v>AQLD<G;GQ%XVc>z$eKU59}Ue2>@C(t
ze005?HE@~`f|HcWb;Q53;2fFVJ!Ihw={-7b{<YiUW&xUpEaL7%fDH!C-??soO_r?B
zSUeOYI}9MGm6Fywf3KZVjP@*&T~;+64_mG4_XYCxWF0<foSfu%p;@qQBE48L4(=T1
z^=dqW=Tf+kjeD2$UpDT;AGNy5`>;nZ<AL;>jCt=4oX;Gz#a&=c`H8ayXSwu69Yl+9
zP(DE%iOg?p_ML+wHxB6#@=cR11uJf>$1=XAZu-@y{{>$D4Utr)1)U#TZl}88A2=cp
zn2ZHq2`&Uj^Vk?7Zw7w_`Z!PV6O3G&T!xGR5TO;VwSp^Qz@<cX`Z%uvIJ3yF;4(<&
zDEccNAzTHmo)12)?QDQ6`nck^<?r7*8x-IVSzTTGxgP|#VsjA?U$~U-^9S#X{@LQ~
zAG|L#@hv1DeG&UyGig_7Mskq;bE$A_7DP8AF8)-()AK9R-l?eGIdlF}(<CymDR)bq
zFHT`$@%gA)b$A^JVseFqtNKGnymj%?aw^wky(UykV)?xiT$;`wVMOEPFaJUCvd_wL
zXt2q&sE}oAXEUyN4g8g#EU!nni{}#(X*CM4GeLtA4jPkA7RzSp6ATxpde{@Itm0kJ
z=Opf+-}m_}AsKri(qw%7K75=P!crZ`xaj(Q`xs1VO&|7`I#PzBuFMgjcz`{l-gD)(
z_)|X&FGm<KN|k6lgTc$3%gEQvi{d2rJL4|#32@?g<eNA>`Yk+|sE%1~{U3Ez9+`F|
zr(3=XO>uU;!{h&Af;H5Whm?7{^cK7!2P^b=t@n#M44T9SLZaqv={m>E+9`(dnclb<
zw#jM+E^QN-{$a=$on|Q{MdpGq!VKG_@^+92$kI}h0nHwP7Mq2XDMC}3p7qAOhUjXx
z$04m9VYfS7A~g42J2ehN|9rBDY9#NBsg~kuoFi1&nqnD#wWW6hrqbn}8%*Dn_RV(-
zv2fpfCj%@mPCthRi3~VHHFG}?Nm38D8dMBrD0Vogv;&nZl$)&d`G--6!cmG@3$5QL
zD&B`pX>hA1rn{#;AzepTge=*0pje|yb!cm308{YaHfQF;9EshpG_S^_Y%SjhZ7b#K
zR8`)5dD%#r%B>RpI?Dy(Gw)x0*n_jC*y$PX;uT!J)SQ_yP_H`T`xuHGQVrZ!zi~5z
zl803RbLDCM->ck6u3!(?<yr2voH<!%umx@_@y~!hw!!ft*+rbM*4Hv>et|st&aOB|
zs^)VSM2--kNB`jUIZ{St#bx&<$r9m?b0~38u322n1&}17v7&MJKWJi-*zz|yyc|BC
zESc{sh<B-5LxML;4Gx`VQZ!{;fxDB8t53OC5Tkevv=iKqjefGi7~)-)vhqLZojl`@
zcwLEXX3Q-(bSufPP1^;jlf>^3pNogs)P_g&7oX*k7>sIx@0p1Vb>vWSR%%`LgW{|q
z(gXd<3}dm_C$<3w+V5mSxJTf#cbdl;d{bL?+qzQe!d0~GZCQVR%4u`v@qDEuudTe`
zG>|Ost4*%iXq(VPeva6vI67P5NGaut^1Zq#G=)4v`|+MJeAGP^Zvu)ynU}S~`Tevl
zqCRVPuoKb+UNCd!1osE4z>GcK(BD55YY4XQ?ACPpj~R?-fZs@+le#dD6+Z+3wWBB*
zLP(}dH_Clqaw>;`bD5b0HBiVtavuNCeO!vn-?sB{#YF0pn`?+2v@1*8(SB6xLUBqy
z=6Sqt6CVh~oNB$C@MfnkQBTS%l-%ApEnq&cZC#5qIR(vCCk?}xcq~|~YrgyXRYwr(
zX+ix7A$3iL7kEE;WRo&Em&N_@mKBxclvvh$t%*6Nyt7h-$h%I=EW{JiGMHfY>rr>q
zAF&}4-AS~0WX$d_@)wV6p}4wsIAZPxlg#)d!X#TI=hmTq&P91DhdpQA!msUYoD44q
z!wz^~-^gA{IR=J8+Ec=m^j*WC7OVsjQh%k7?-GIO>M^Hk=qG7?(kCYx*a0S&0nz9j
zLp5R#W$2qRDNI;J3s_T#)qrJQh@ea?{paqbHy<o>%F(ZbFXUuDF!7uAb;0#0GF4-x
z7@OuHw`==!P4ybW3>n&~a~6DBpiRcOb7eR(nA_Ux?5%bcNFFW(aVw!fxj`erGH^+@
zt3eRSbF*R$M)l@#DZ0p^DkcwjdKG8!Kpl22U|%_ObfOYv02ITBm{2rqU2;u)4>nXw
zlAIvo1G3>0!p~p{**n<EgTPZ{14)%PTh^{@Td5b`0!ns__~~^PVZp~(V~?ErzeHC@
z)7ldr#XGvH4Fdtq*l~u_$NlJW@xDEVpzq%1$%pk&xyciEJn8?}_Vw2fdy&x;dS0g3
z`^N%Iq$RGATtC=dm)FBuzQ4joL3&bR#Sx&iLB?WDxh7)CqhM9GVS7ZUR-?pn2_`PR
zm5%xh)R<)mDE0+%H~LF$3`s7Fx8pL5AZ|J56cIIQ02k=!?^;3RH;ajFT`|-#cgm>D
z^Sr>hz)8)ni5M#F+xQYR4Yvo5ZX!ly^m&3vT3=D0M%z@r9xWNYllNJ=o==A>97Dmy
zrGnJsfKhXJF|9SM1M_PIsAUWK!#YlUwice^$2Yw&Fg63$C}+?(Uj4A4F4bmHAzo!N
zjUW~s9R}OTO7KuBn>p2dVdxL9vdeq(DFsuH#)2lCiH^}H9w~tZ-X2rGPuy?}b#AHz
za$ranJR6~-f3Ebsl+Z~^D*me<=TkF+#^U<f*r#U(JaK)@QNEr`Qdep*s3e4a!obZi
zl$x&lXGWY$>-X<T865hQLJ@E7DoDrV8=vBDrGG4YmxlR<B;Gq_v&Rx|dk;8Mcun$s
z!tdTVb34ZTy=sf4Pv`f;%Y1Vkb7T8u+pZ<c+ph1%<L6dO-9!`CFD2M^>^nzy&Q@11
zZ$CY!1jD0bsOh**gZab#n<hH`yWlmD0^_V5`&ayt&zT8GN&wL0EQjBzqEzjC?92~X
z=ko5YX<iTpw>Psta07p|y|UcAC|G$nKHxZ8RF2kVBFPW#A+Hm_-ePTp*&sOme%_VC
z%ueStH}!o56tgyfd8Ye?dlap#aT!g@wmeVLT2;POvMH~qWp19)aTzx~E-TDBnZ1>n
zCrbM25U{TGI_@Zlh|u)mW;Z>4|C_9EhU5KfYp*D`-e@tzw2hkes9v)7xpX_gnQZYy
zGf0clhLqkf^Ly391d#mwdR}$&BKX4TY~<J%K?=mqZgSb{-AMLx%H^Mf<6pwC{JzeD
z2z{8Vi_Z9|nKA*s_q4q<V~s=a_?zqO1wZeJp3KdSTP9s~5%j~rrvI1ZTjZx+P0|>(
zy3?l>vb}!%XPT#HW$!$$IsSB=%-DU&Avm_&s_O>dPlrWfADMg(0PR^a+KomE-3^`Y
zjg(?3x3#;)(8BG{sobRf#yo+)lntESu~;<|cIq>Ni+yimxz9!f)-lR9q_4TCPKuMA
zP%=nLCB^nAKihd3Mp+@uTIP_Wu^n#<g}GBi@ZvSuwZ~+8>fK7z%#pIPt;+-XDY<3<
z5eGZK2Dh9fZ;vhmnA0UfwX1-^3jDd^(_+vv-yl-TwKm3FiDqqsiTL_0w*C|NA7?`H
zS&mM+Z+{)VJxa-IAay@<zFI2nWVq?##-|uIpWMYr8j!3%Oy1#B(O*o3A=kG{0j3+3
zmA$h!W(SDToX^r4EWB<D%3@O~ti7Q8x)5XLc5UPI=O}DO4dWM@8n!MsAD&^u|H41k
z)RTGp@N#&@^^r4UPy2a&wGay@5e4VBkrO-#D1IH|{5cKBwY<!&qUs`nGv3E+C3MPW
z&aihd@+mv<4=9bV%LX%l3%2bNC868)C8TgsNLfAqoTGk6W%|d`dcS$bCSi-MYQ5o2
zl(Y6Q8~|0;TKz9dToSb1cV|tfc50Ml0zRpz#_LR9PwvZeG&7I9RpuIk=132f?aV3-
z0eO;aQA*^vnfHVOqlhzI)OIuUhdYib)#ELODb@Y+M3if1@GWM`RMur(Uw&2rbI=1?
z>d%7^>vu9su~hGaP*mHt(({Gx^s{}$=i7E~5~C;Ze&)n`{@c0=39C7O{(U9+uP{*)
zacrbpGM*wEHzZ{Ef=FA8#bo!*)(>Xdvmy%Up1;JOy`?%x)ZI}xc!}i9C6X(tYbt$}
zv}It(@!<}uwz1ClIQ+;F#5bCwC(RWU^l<_29y*h-mm#wHP4fdO-E(r;bM0B7eB2t<
zVl3Al><xs?bVz<_3`&2QC9-ZX4S6c@g%9^@M+b?fOr5V&u<Td^2Q33QXbTP+@#taL
zT)Wq%C|GptSSYP?E7t;z4Gf6J9Z8zYpsf(qxiD9Tk00l3Xv-pKO)lB>yjild7TdPN
zN25bZETB`h?nFq6^Ksd1F}`l8ht*cC&ql+9*ZlcD#c;p6WHf~*w#A}j4dcr~zRo-D
zaqVqM2~5v7XUmH}`gI87mcz;5ybsYd%NMNH({3%{*+;jnfwtS01><<{Zi;iz0~Wg*
zv(GQG8|ypWUgsL$L+QSeAaQg$!QyVm5@-9m8#~vsTmGVaZ>%#TIm;~@MEA_s8POpu
zbBToR(mLM!zTp_}OXraGhuKs8FX*P5!?xGk8nWF_m{`PS^z!j=yPqRxv9@D*EBoaN
zti^<N+5K+d4aGU*-LUy2`sR?enpUo5POc$}?McJ=k?dmUv`d*)U6Mk6Tr9NgRn<K$
zAZnQHPjUWW2gL7CzRoA#++Oh)Ls=rUG+P$w%xm8%vK=>0RK(r%q%Uqv8Zi_Q7P8m*
zs*@7fnwcW-S%v53ezEgc`=Kv#>eO#-Y1{HsZjGB25+T6??*MK-ey(y@u8{gI!DX~i
zIILASJxs-S{r5Wsa?6~@XI(DN3#nD7E;F%QwR54Hxsm3c-lBGP`+x0quxJzYNyJzB
z^P0oPeuRBvV4ueEg3|b^KJ`^`3`~@|EZ(&h<1w94u(w;{j1GTanZkLd@8Su*GmQMX
z@nbQ_RP7Gsh&Om`h|@ot$UVJxmF_q1hJhKy=sA&li#Zb)6ZPX^@ruN!-(f#RLnI0B
z27wslTRpD$$Hx>o?TQM*U#oPBUB1QV1$Li8g!x|xTS=yYocmRTfMRMA+9|Vl*xv7k
zjO7S_{;7X`1QowyqYiq%Q+~w4L%xq^&8-{alix{06(`Bbx)4%syl<VRfLWh;=B)Q5
z@^?-%9`W5*bs7mL?RW^*vXha9SoAzVMWE{w9qWZyy4m9a)ZCi7a|DA~sB!`!cQ9eu
zK~-v+=ld>)N*Ys8wU$&bCTjnDvwzzSP@dup@R7(Q1J3{Xv6bUI>y84YqmL8lzTdBL
zN9f`mvaXJUWrh}@k?dn!WXQ!qF4xCWyld(%dkg6mbRGwvs9r6T?CB(R0zqsVOhEce
zk2RhC%^Rz~hD)va+K+CU;pf@6c|#A<`7FUT_RdRnFyGJ`uXZa~d;kBi_mxpmuJ5~B
zMI}^pP*D^R85#scq=t}??rs5*5|L&|K@b7yZs`V5QGo#jDQS^VB&54j^4zc6{fl$<
z_MHFs^TD-TEQXnR-{*av`@Zfg{=JFC;z1=o2gb(ZYYlj@IpDZi2sE77uR6ch%HQ8<
zGUh{g9w!~9rnvv{kp0_R1s{r87A_5oY+7K!U)Sd6P5b-n_}@S6xC7tL;)RHK!>@1m
z|Na;5n8Bj7G9$CR@AKb(^B+ot|M>Kt$E2ZJcudaAd+I-a`v3M{X%e_j(w|L@nEuDt
z>4wH({C%>yj!)LVx!?c3kN<foG~)1K6~x`Xr2W@*(wM;)RUBK>yccV{Ym#&1Q8n`G
z{#QCQHd^Wp#m|U{EB@|H-YbdxQ9Y!@(RG3F&#!ITj|ayv=_bt3jW3phx0T7bs_pHJ
z`4~ROLddCq0}9qei*Ii%<vw$K(}l#|)A$qVbtmyvC_QPUqnxa*roO&xzBUUmEZMb7
z^X#^ZgioT%Hc87&o;WaKyZB@_YL!J^QxAHy{q^zx{%dAq;NnJ_NCp1+-90mc9$^S1
z(>q`<cYfst%&`aWA01kNn4k?e{J}tQNo4N@A1~h7XuL}!SAv~0(xtv3(M4RQ9akJb
z(nP<55uBh&G_7316eN00ELt!(Ji9o0v;X0Tgz?#t+ng<VQeVDm%w^*R*vy(TGh2KJ
z8mw?B2V)hwlFipvLxV`AJbtxkdveyIJ^v&dqA-rfU|$w#*@`AB4H@_>?3EaZ8|A#}
z-^3aXKO+Cn8}`3_KLs~<L{x`jI;j7=D?*Q{+1$465S5Q(6GK*)_7K>8EFwP7Fm)UH
zA<z=f)sA?2xdF%FfzLU$>~MF<1S$hvz=?Fg8d%EwRi}dv+%UZ*F2o|pOR!jJNO0@>
z5iZNV8(cs@>xJ;nu>09vBHe67j0Vz1#c3$!rh>HM7x^doZhpCQB={{vC+vSIv$sI9
zOQQmCGs#6o{I71l2k5&Ae5eQ2br<;ZJ1VMf&Y7ODR~vB~yuf|u$ZuA>3LV_JZk}sv
zgunK@=lBxeH^6b-*;BW~ZWnBJmJZjJo!$Ld9wmAc@FV+ya>pWkf1CM`tV&=4zIPF1
z7X9f*R$i!EuI2-0(xfWG<35c!`nnA%!Ot3yq(zR}H5)30$T`Nqid)_4y)O4GtIe1l
zy3Qx}6~4VWI9W1aGV{9@A;b3%uc{1_{Nl5JenbTeydd5gdM*gY$<SM(3zmUn(M*+K
z-=BQV9JPy-tx;J`T<7_uiYq$XZ=8=RF*xp3b<pz0#>XZy=~YHgFMPP}{cZO`c<*&f
zlKJ0kuR^0#aDXm9qlx-;oCymVOd#>BS)<5f6Qz0YBJ$+@ut<a(*fp6aL!Y$^{Z?DF
zY|R?bOW!RiR!BMDUgt9FL@$>*2|8~s%-6XSD#u22(RD}13Rgu^_7U!NC|XYOxOEXn
zGTew{jKX*D*$^2LJ_ze0HHq|qW#o3DA7JkGJs>ExGcxkkrpBoFlcxwSI}naXGi(vr
ze-xmxZfv+Zc#mPyvnnW~Ew9CavwdVV#~T0U6h_^@T|V;T-Izsp%P*mM$v$Iwf`+FB
zhGSp6xvNbj+3xEmX3yxuYp0@@Vt=kPsj2H&^BvzHs-t`>>w-fuc}cFD@N9KBk`tZV
zt=l=%b8CW^G+^nP-`h|-J?3Voom|$3z3d!5?moL6Lk|<}@|*cN9!O7VPua$3*S&c5
zO<CQ+dq3AQu(WVpu~SPSX|AjP)J-2({l<L)TcO_N<rDV4lXmh6gjnKZZ=BU8rJQd*
z+_=cI?^pEUz2MovZ)X;!FDOmfTH@M8E?xfZ{xgmOlZ0!xl76kCE_P4E3J6@}yAG~Y
zK|EnO%iI3Wlb3F`xP18rM$H2tnACwjqaRqTqc9Gd7I}pv8ZH3W$qouido?R~(WOmL
zZ<K;`tOl78F2K`m4=6|lY499yt_05Bho^Zrcx%~r$a$8<5vUJ4lp|0jSR;ZyN8o|g
z2BfzEi$>!uL`Y;7e(;;y0?)p^afBvvif@jM3mgOyFa{xPaF-N=8m=7i`6h=H136>D
zGlbbSO0_7|d4$g6TblqwGC_R<pL6{{WE(}~@<_a0FHn-0&%-of$!!lfI0N8aH2|>K
zA{b$W-?=w~n75a~M1b_ZNR9&n%DML|pZxam(DBRPch0p-^AdYic>q<({Zv1o!3!Yu
zwpY%SS@sHs&)bH?UM#>o$qv!^8HTEuuwx@*)KmBaC&3o2A1>wyt!&)b(O8JOu#Rhw
z`}WcRq}ozvv!}mOi9C4+Y!+sfN$0hx;YrWyb_d{E=72;K&g=uTbOGU@dFV>=DV>aP
zx?0=<64PO1o6nR>oB^wG`%S>N%Mfy<CAd1v%_<-621648D<V{rX>f;Y?{0_IkM1J(
z@yTt(;OZ@?(n<lKt|7;cg<@wJHtuC;nx8V2v(N)HrU5DM5Cw0NkMKN&UB@lISzO<w
zXo^?W7N01Eq&)95I;&%-&k><MRb{<3`&?fXJhZpqh~3On(lD3RgAeL1_(><^(JaKj
zuMT%`Ox7d%!+Y&XiA_Cx&oh6hHVx{}Bxy=UF;E}4D5*P4C9cC)-BtrDB_~HDA-zKQ
zBH@_=H#wusj-@{F^-~cScEHqO#Y5t`O;P1`F6j8b+0A?wBh?+7`~*DX;aPVGpWH2}
zd_0F+ZCd~>;KojHy#Rw*LFCFcHC-^miySHkB7w(FE{$B*CS^&+p4L#nYf^;2jo9%%
z0C(XBFlm&wPrErZnRA=_C!+FC#ZiRxd-p%kxBjYhAK?Q<o$LA`{nsP@Ot&XWSb-KF
znTo(IybyMq3dVZdtKe0~b=^ScZfanoH%1YFE9#wFd|Egs>!^&44=#6wlN)OVKLI80
z+;ktF&?k^74X?^}ki-hn&z6Ea2ix^xg)TU^hK{d>jJDe+{wRfc$;%|J6p6crhxOIw
z^A2dQEbYO|OVWV;AVw?jR3`g;?1hF&M0NTJm{%dH>Pq!se`ivP=MNOlZ|UDOu@l{N
zWuP#oCMo_QKiQUuAaJ6zhHK2VeclnD0C|jBpA*=b8=YsX4z&6#>ZN(2R&?~<;g|W7
ztHx(7d}nyULA0}vtNO7?d{N@VMbNFR+{6GjskpN-zjv7<RR1oO1T~5M51x$D5`ZYt
zFY(tSoS7Z<d>mxiQ2YnGtDzG`6VZM$sqTB*KYkQo3prETp=GCt4Pn7V@w!7vpcs8k
zcNg~=qh(Q0k_ryd=?fNm@Eh)-ei9hgf)CKCvf5WEIixc(kB$N9dg)D^0^M~_^}Aj<
zQ$R>u+CYL2LMM<w($i)q$E0bQgI>4Qwc_F2lRL^d4`n3NjihrbO{+*oAyR_m=9ii~
z&-}RgiAXGN(_Y&cwM=b=WEX*^>uG(HwcE2rl!?wWIhuQIEo`AiGvtPb6OG<I?z=Y}
zob?v}2+WyyL(oWFkddy!*zB=4tM7i$q0`B2YIUR9K#3&O@64WCD@(=dnm{%~_jr;f
zLzLvr?nrW&g*?0ASnC8KC+oIINK^0}G5+plHUqmZAhoyaFB-dXb$mHVDS+vo7pK6Y
zUkVB>SxBwNL~9E^=yTz-n>rmU#CDy%V2Eq&TDRkIk^$Pb0-M3QnUIO-?L2J8;;ETn
zzh&y34|sw_-E?2y0!`mVTQ(tiV;~#WIK!QU-!?2w=;u|h>Hf9G{~$m_Z~Nkod-PTm
zmbjw`DAN8pco~&@H025Q^|GLIFgf4N=1$-th)r~}T({!V+-#W*CY6b8$jf(GK0#33
zVk?iy%eM7vSWCmzQlgAwqhGw9q^bNFv|4w3!jj%tT`T2TfMFfVkS--j6gv?}UvSfU
zw^@H56?^<uBc{f9o%L)~hmZr>`joLh|9J~>r9F%M;@h9R1PO$YYc39KQAe#vLj#Hs
zoRG@l;B;*GhR{XQ@fIdOm0k#SWmULA9Hm(i!()5zE^zIfKfY$y)3_=?aG4ks+FiV#
z-K^PI9o+>aR95<J+v_N$`B)o*Ui&w_+iDQp52f$tv`W%@mxZ+$+_8D{oHJ+LZ183p
zTvFD&1W_!i69T?i2Q_$@Y0zP&8V(7-;?$hgFdPA@W<8NRjcE(bwZa}AknpoEbtUCm
zM{9KBa6b1vy7*}FgVCY3G#XM5V;K}8w$y}cmKKyuT9o5SW5FGG+RUxRUa{9Db6fFL
zGXUByu@hb!Wo)E3j-kYv^gU$m3{twO)AaPZuuUgY;*C1r_9`bJp}iZMwWoI2Jn#T8
z(ggb|)Z&Oj4pwq)uL`!O-TidODDUr7IaR1o+C6qrq(CVDpA`rshY0r3jbWRGPGNL{
zem!G>jA%cuwIev6?<X~Rg118juf2J}_GZxHPTTRh+dNk<cVEKufa!`|+dVW}=$6o+
zCo{&Z&1U#Ednz%33)(+ih)VhXg_+NaBS-yw#;|M*-A{y`U*HSdPRA+XdA$5w!!((D
z{iQ;n-Xkm+_Kh(sF9h05MEdNnrJ{(A$G^HnX5puX#)KZv*7;Zy>`e6xBPe+xwUiqg
zjjC&CosVaq-s@2gB9FCK9pTMie$mEampMeq>}uXNW6}~cTo;Ner@@MLJzn`vBabbK
zZ8BhBz1w$ue1J_qOh_x$%+K2<!#sf`1Al+4ZqdqPM>{q4)xFjg9CN>A&z{tah)qp`
zo>9rcJDm0Ad}xAp&Ijl9Fn7oKd)AY3+VE~UGcem;RE~DQt4_x7)`$C4Y|L?pebN#^
z@tR~_*$Q_O5O@+h&V1v=kTS>6k4R@y-wnWd7heG#U2t)SA43vUvMfMsZ|nWLRzCvi
z>Pzvk6u;_wc|7!%^3=zc*HB8|Bj`lWeWcm?%#WS>L_Ci=#(4Jyo;?I`uVGibGhSr>
zP(p4GQ*|=#@i@u1oCMq($*0s{AC_TQiy~y97XUVSUXIoxkK*7Wdb>%cgz7YXeNfB3
z_bl9)Sm9?1>bvhiC$Qb|Y#Aq&6wO<NZVJ!Jg(}9$%3$R})GW1UQ2joHwpt7p!U5c^
zcYpia(0-5alZY?I1NOavXzx6RNON)7_4Bs+tEx$zW1o<ep|#-{WMDQ0e4u4RarThO
zdL7AjhKMYTo7o;xQ7sbZD&MC$tP+cs*DRf3rm4Cd)z{qR;m2p&>KnpJm0c4wA>DnG
zPU_6A<mDu$>4E3=c2F7IJC}ne?ejG15udki@2=ugv(`E&@3Q(L6m)m?VNMg@)a<B)
zJZD#{^>dhQzbgH_X*GlO;liutnCl`sKpEjK`o0KL_8|3er@}=xl=kuAVnWp!-OFSr
zyafG1OVrPWn~%Eb$?Qb@7MdBOe(~%<-$Cuax7PRgmmR(sJOE3o2Tc(SX8cxwI=oVN
zwg3?-QscAsR!Ip7aD@51mT6WeXSASjO~4X$Q65QO(EcJ|aus7D*jSDfjgiS%n=VCh
z9-`2hmNfZ^0<XevZ><)aH6=cgHOFst#J;PE<$8%Tp7(XP%JY*<z+P(7?!npf5S^Qo
zS(*ZkbJE*<A3i(UC$`WTTEk>plF4E?bO1iX40C$XEI<T(6-uoB#nkGTCI~v$r$0Bu
zHoVjkQ~Gg!@GAoN27+BNQc33Qrjw9#l?GfZYm9AdjtybVk<i3KUj<IZ<w%8C2v!Mg
zq6*lZJBL|K^l}miRQfwoL{_6t8l?zBL~)P{B*^Ii<gzM|77r6Oqq?K#+1Qifxvf(<
z4ux`9nYD?!x8bSmPlJk;o9Oc_Qu@j)P42H{36AHh+E$7OQ^>9A?093j)?C+dJue8H
z`oU>xW-~4!G6I`+If|Uu?jfzI!}*$5^?Djl@0nbs%i5zJN>Y%uHGF*X=F(0ZjA+!;
z+!&uPJA>pQ0h`^TIq3am0U8aviLdMhGBg@90~wByJ`e2?UVnm5q66Zc#&PEPtB3t<
z!s1|zJJH(?L08j5j%JCC)@SO|Ln;Kt&lSe-E89nWxUNn)(|M}Ea>wY9iI*U9AZL}`
z^)j`J@A2|mRr(P}GLuZ;p^W7vSR|g$IZke;&{xqy)T&JMc~0ih+@|tCQCB(lYvVzN
zfsb^_bDOwx>N@OqN$v6jH?g-)D&zmT0sK7>sPG3M6-R|q@&80Ko|6%%2rtnsH$&Xa
zbe-#Ytxx12P|<r&?U^8|GcjBrkOEHf;aa4-k|8njPO0qpSE*sozu6e=e?~fHlah5X
z?0a$MU5ckTZn49;P)@QII9cN3!RjdBC=M{JWxDmkVAwSg`zoM2%+1`uts<lYBU}a@
z0H$0hmM6o%e1GU*$3rhaTR`%PmIN=bC|CRk(9Yd4(kF6C84a@VJW%^bmvssqq)MpO
z-Rj2jhEN=OkXWANgXe;plEfKN6eaY$@h$_<$|TTBuq2OjYj?H19OOvdu??J?K#<y=
zr0LUyDm=kQ*Z#b7`EA;|FQ-|xPf52P_iJoPGsQI}Op>V+Jhvh6$+eX0%J^P{VrH4c
zkv^5;#z447)jgW+av)vmR=MvbfN9ubWjRNCSR3Y=GnIW?wf{i&EWffW_0~@NAKsh3
zK`7lxY1EZus16rEsI+ldKC%G0PXXUqLw-MO3wpF?D(>}bbANsa#pZcSlgdpt9|g&%
z>`G2rt52iC-}6i+DAQ57#dBJy8F4GD3XgptVr;o1n-lZ!DIrTvNa#e}i%fGjZso|<
z6V}<sEvWJCuse*+xH7;Wx)GjNrczmR)4vV^uM(S@?Nfh5ojt~*jvbFCtK{^~!oi(e
zRn=eO_sQD>pF4`!EH26kRVq17{nq$<=`Y^BWYK5r4swMHXa!n!_g1^v9+pXNhKSrf
zak;^eO2g;!N}ca=$$chowBHXUFbVfpMHgL;q74hWIp{Q+I%%NPkY_Cyf$Hq_9+aHO
zSUGFM#gJ~5p%It$zESkGKz`Anm~)YtAp5eZUB7k1kB(1$MqnwEG@~p_XJ28GD@Zu4
z&DgB7(Un1N(4Ma%p-=K~Vt};71j5ZYt`~WkO{l}(%h>HA;}S0%O&X8vz4s~UiT|l*
zLyFx-F(Ex(iN6rb8$yT23HJq{6KTXmik;78&syJB+=tGA9_C1TA8dM*Kc8HEH|`M;
z{oV*jzM6G3+1IV^B27su0FAdNBAN4K?%ifLBv|d5(9~!g#a%tR-inbEJ73TFipPT-
z?=S?gdW`HTT@Qd=a}P?0yC&B>qx*^<#)kUb2m+L#T&Fd2MK#zbC`iv~*hgaIWgMQ(
z@GZY`saW;6w-aC6gCl+BPIixDa$O74E4AOvHVVVL8Z*<}3!mz$8NNL+&Fl;i*J__i
zE_vy;ij4cg{$2$)doulJ9JYski=)UQ6l>S=?X}Or+PcQOdE9tYspcu^Rhy}!xhT%=
z8^`ShdvV!Nltx3$OYse2t5EEfz?%Nzhg0s(8r;@ZPatdQkYHBJEA#5lnT4A&_9v=`
zJf1WN|JYs*yQ6%yCY>`K2aLU@s%92%r@nV;$lWOL{$3QexZ1%Q^w&lHyB@>S1{tJ9
zUupjZ9?p0^s-?fR6wRO$+Hjc&BH~RIW4~83MZ9Hm*4sriW_S1Ki>{-jB)wB87if}{
z^qg*f9t)S_WzaH?RNdu>HMV0(=|`GeWg*(?WN+7Pr2HQIoZQ!I%i9#6<ShV}^L~(V
z5!N!XXFsLSErlb+?0DlIw&V-V&jCO&*aOc8e_pb)++Ij@j;mY9X~n_pu$Wv~0w1)_
z*#I4eHTKH^#Ul1kCP%~x_G>wlvQ}Htaj@}s*oI$FskvC}M;Cd9h~O5d9;e^?wK-(|
zcr=nM_#{_98%<n5Z@ise8!oIv<fabXW!UHE%`nXQ*_(r#z{I#~e{0h?dG0=pv(^B;
z$&5V1MC37FQLkzGO-q~7Ebu(m8V<zmk{m24rEOxfz}LGeIeRhsKrA*h9+5Q*)G}2S
z^cY?DzMeaa2<%o5U_&SPdQLEkgAULero|L%sI=CTE_P40oSk@EuRN00qS9)c3jU6B
z5fZYiyk;2Vw=3hr@-7eV#Y(+UxrF*`uCE}7aFlw}RvH2?O-_d4CUS^T7bc}<w0D)G
zp+ZW3(`*piTxjSMV0~;D*HNLr$RCK1UZQ%)3F0HIsLDUDfV2K(1U=GRy^6p8n$^6@
z3B0a$lTtRH3iSD+M_wyTeL_6cdJS+RQWXJ@_dHYZ+NoGCkIvI3s*kuOsjy;yyL_Zh
z17b<w4?_H(bA@b?!z)1?APiyc40YthMg|b6qkp=LbqWX09DlH9${HHw?M}sjo(3S{
z97J1x@>T2#UM6(g1osayadP{JP4mo-4AiRR>F1Pts_OU1_?=&qswX1Q32gndM=fuC
zK|uX&2d*!79-G3XH8yzL!q4yQRV6Q8KfbdLtAG;(wFw^VP6150X&@Xa2psa-Nt$D!
zTvY4qu50=Cqgbfq#j!@xE4=BGG~9^)yxSK&0z9vigo?TQ5_4JmXnsmwB#P}vN9M}a
zOm`pNF~wyNKt2QKs^o0YtED*g6{8%Udi3ihHq&BTnpzG3wN##OUQN1y7c7BzD^U_*
z?`VNCmQtDE95Pm^V==Vu(KN|OYInRE{FE$UN%2d)uoBOz)T+v?kRiXHHTt{K!3QUL
zSHwJYLX3ftx>yS%`3E=5kB3YMv|%0bEGDw)O#q=GU~F2(!OY#Fkft@zX71@*^`;;2
z)7i;+1X*c~=VIjyT6yPG?t>Ww!tw5Q*hz$gWh>;>)`GqjRhw_f3nk@Q>E#ID_b_Z=
z(6X*OR}o?Y^t#w!^|>;gC<Ho9y4VZgu_>4q<Zv?by*_qVJ6XUDr!E?3hEd%~pjTPS
zvM(yRmXal^-badd4Wyg*XzJ|?A+1cWfcig$WnMi_J<g-gNTl;}#H8T(BvGVxGAvas
zhN$X=E%v9~ca5lWx+?Tjhpu!Xg1)qOTRiytKy3N=n&)bDa^)p}Jn3Z5Qp|)tR2bA8
zzi1D(w(DN2pXK;i`i!G>cN02wx|0$jn%qcd-Odf#54*gF`9wk!(I#4}fFW{KpeX0T
z!fys4f8aF*KNp?*%L@RRid=Ta5dB)c#}BW>q)~pNc|=5sZ6PoN754g%l@>0Sjr++g
z%gRo-lk97V2LPxwL=y|4>?OMB6x#u7pG#Kf!ES}))4!%3R?dv(;Ji|8@b<xgD-S_6
zvW_IK3Pkq4LLHkc>DAMFcy{}u;IZoDl`+l&$eaoF4uG2$JQg_n=*X!hFK={lN>V8Z
za+TdqxV74!^#zaLsp9<fUGT-+E<@{ub%+XX<z3_Ng)|-JPwMw*K%l+BTxSQ?XCRnf
zm=oiCDP}KGdo@%8;cofajRmqt&&U3NQJ1{E$jBQuaTU4LPG}0Zia&3e$(*ZEcgX1n
z$v`swzKrSxzG+}#ddN|1uI=hN4xB5G-+A4Sr8yIQze6iTwm_@I-vJ~Ptn}R_a~+r}
z7|?zeeKqrmz2S0rog!71NSKh6(M1zRIYQOg*rpQdSCxi^CHKb2mEV|Kn(IoAovT+&
z5=NTI8+>6kuArXXcej+dLLYywNy8u+;Ir?Or7J3LYi{=5%f*xS3;Dns-Kf5k)Kj)0
z^fkpgLV{r~xjM-_nZ$>1A2?p^+KJNN$>mq@$ymcJ7Fi>WaV!K(-91~&qoZF#aXgIr
z!o@l{0-G}HIbqBkmdv-8)Y}GG&p5m0mrw{RaesSQ1fFbEd}TnpRsOMJdP(W@t|Wf;
zBH_x5z6@#71HDd1A|J6{M}L`M&*2{rnJ0B|K*7!2KEIu0hXF@R5nG<%aa_BtTGPqj
zO}K@Q;xni5w%8sf_|*(}9us|!o9Eu~9NPSXqvqN1{1%)2Q9r(y#$>QLVekFC41M%V
zb13fGYQ}j;9&nTJXCU;g_@DHx*sN;^eQWRg;s%(rnHUOff~tj^#N)>bZICrsM~csU
z-SVqciqU9~HK8;~dgAIKb9QVENvnCa<i0#&Hfi4OZQmyBnK>RGO=lvU@fw#-MB-2i
zP>_m8V~8Cei~M@569WsYyMfnLlhmV|u&SMCX}eb4xU3Vtfxc{1%r+@LraCqYZIXK2
zWH2lN%e;13hHX*?J&tU2TlZ<^I=JIVKh$n>aLCq9?qs@t-2=fb_o~8VfSFvZ(1+j(
zzc2kTVVrLfNus2K!e>(Ao?B@l3;GpDyC=dW8QcdDfUor%vM(}r9YBf<4^jI}i%Yx<
z(){`L>|(fxRUW|q9mqyo(o3$ykAAFuz5g}DBq=ok`{hLRAZ;s4K}CxS0pImm<_Rq$
z3ok}NE}5q@4f3;{cscF^nBAohIqOD-+$O?}uV-4RpTA;D9DD=D)LW&N6i4{^TCGU-
zOCepJo8(+3zxpN8$U8Yd3i++UX3t$O?m~C$olj<RF@JouESt{M)GgZJF{t6ma%&Ri
zFj?nMGDu=bB*-1n9DG3HmATTAm!QvKAm93SGrHz_cSxeQV^oivYZaGOs=KQUdZLBN
zd^fZU>$VU(BbuR^dEynxeCz~guNm{}4RfiCg*|8eSjSG4^*C{@BT~yQ%hnxyx91sR
zdj$@3Z%y`zmpMR8=kba~VZy~o!wp}hp*rWqxXt$+c5P{uv*?1=mg7=x$tFD?IDa!L
z7kVCskO9JI?f-(56;KF!D~xfJ3R%(8REG#kT0ZBNgYckY9mO6;*N_n+`hipPD%Wb$
z^vM@&U)yk(&NFZ29m}&N*w0O|T08W+ezzkAs3R6KX?`q!RuZ1aqyQ#6Le5=w@(@16
zZV+kC^V=PhPd}geMN^_XcbbwsC-(i)@P`2qFpUy5xje@`ZHl*|qS!Yttk{sgvX$pn
z|N8Gc>t8=T4gGb?FXGl8cr$^IYS<a?rq|>$z8-|xqMQENg%Y3(uJ^E#J%NU=qXVD5
z3yDcm8c@p~52Y5*Eab^?Ka9*Hsm4e7$-!3F@XNAO?6!b<;~;QNwq6V)XFT!63W#{x
zdcaVNWlwRP{dT_ZH>3YU_&4wn>P-#(C-GnW3u^T94j=I6uqQgtyd-?cZ&!Fr((fQs
zCUyiWzOVQ!4d`FLI7)TJU>fSt1thG`Dq?WlIcpaB@7{MXtKp<DD#rWWwO2Tb-0A8o
z9Y_ECYwtUy{zcxJC@QWeD}+Nx^T*x4S_1{$e{r|aW9>2nW%jm+M{mmn+3()v56|mx
zi<yp41pmc2`|bVtZ)(bnKd>UouFCTLuFm}XPV)c1?O*rp|8~9q&x`z=;@|YGEQ&Ir
z@Sm5z<By*)l9j!{lfQptZttM1Ut?}7SuEkNo#Joa=zXUgP`Nq<j=Z4!bCGmN;7=Xj
zgG4Qc<~U^P%~TY}!D!h4F_Z%khOP@DKLk~@jZ6Bv0CC>FLLH{YHUxiM>cbK4O?+{}
zXV!dK8-N`>&k06b)30u1rE`9ROz++fF$<q(W*L*UV6OdrQUmztL}rzS#%QHzmSvT;
z{z6v1k-f>wzgwmXlJE>NN`{yHA|OgL9NOb%4Ha`El)VS@2QVi5j!<F*39Em4hfM$y
zt7k$@{V9dJF-?ylr_<pC5HU`Y-DuRc`AD+>C|CQ}3_lXOGJ_9Ke(aDn!&&kU=`&4!
zWQ^a=XfQGxev()s|J>tXe*xN1S<ZC(Z?8<`9ifNJT#2d=2U2J$Q{lU-G)v)5nl9gf
zOwloAt>P>PrZz2A%Hg$ZB?Pt*`na$)8Z7U()tvZk-B{f6dTw)B-Y}n@ZlARFKR+*I
zhhYaiX%I0*{R?7jB9KWI7(hffNGG-4^BIhw%V7-O3(i@NmBgqMNL15*P`#$btqEqG
z*JsATB|=pxKJv{dv^bkyRZu#Po!j+`+}^J~PWcY(X-dH?;2rp-McfrJa?b7?{S@=%
zj`SYt9q>Q~fHS9*ctOw|2hPZ!abH*>OV1`D>@ISsR}h+tEh53e1AmVxUdR%jGK23f
z!OVKbE%R+f-?1Oye8Lm>6~KV`ct*zuz0u?)tvQ7=MNbx;?2Vb`|9Rb;(!-9&84;SA
z{Ohz~q`l%9kCBzUp~f=(FBVt@46F;GmwPbxrMOUu>U497NYeJTci#^$?!7pDYXW8;
zb8yo7e`jqFz2=CqQIWoh*F?KCtAEf3aQkrpUygbJ6{J+#vZT9dqA!Gr-cb&G)2dnY
za9<FO^tyja@fXFS)6J$hvZ~QYgHylR7){S2AK2fZ7XR0VzlY+U$xiEsJ}BB5#zBM+
zpyW`c*<y0IcV(+t{zSFDyFh<|;Nfxq)S05mNJ}24Es>>39iv#1z`CHOwzaan0(j5v
z#P!#Bf4^(OPoMJw>XX-IoudBhA|>FRy{SgEVTVK)4qR_$9))=)okzhcfNR^Xkp$xr
zRsA;2KVC|nLwen7t|~=gNs4;_<w&|-8w`elgiCR9O(hM@y5zcXW{1_d?;Bx3JC#e$
za+K>L#)PsNgs;4GQ}#OEc2uQr)M<`GB9A5IvCYbz1=}hhg1qu?I#o<h8)+y$C67xK
z{$WhGFRc4gUh?uv@zb}OIjb!K-aJ%MWUOv?^#OcqnAF<ECX$9rT_o<-Z-PCeY0ky<
z)U@7(c(y8=ujw|17h`hVBTri3XoFG&$1ivLE-AW5&4|P(mu^#7y&{`a@bDt%t+zZQ
z?{?#+rzP%SSf|~)lst*m#4VE1U+I8jGNzuYw~~CDvYtw(S#4dPq5m^`3;RHOWDVxj
z<u#e9#}~#&V>C_Jy{g{p+k4xeTj?TQPl+jwy(?Zh`Qh@@Z;8!49YuGnvbl8V$O95d
zV)qmi4K9bx1W(o_pL8Q~>%k}x+E2T)=M>K2@2L*mptLIp4yzZ#zT|jjiy0z)=`mu#
z@GxX=IBl}-7G<|4J`w+=A})-6yu}THqzdWXAkvtJznMXLHUTeR9h=$pOIebwdKk=1
z{4Vj@Wgnbz29GL7P-SOaf-H{B7jH*{_{zZ4uoQH^rJ%-d2)xYG>3a{cQXc^3(ljEU
z&96j4FBTxe(gu;Oy8u_SAHiWN5B6Q&Rv)|c4m!ndUoXVd`oNlZ|Ap`_;#|Jpun6Ps
zeiof~9pu)vXO^$#tO6+2V+l;5oY{w0`2=>CL_A8Nr7uPFa)?g=!uYI!Y!ZQI-=VRY
zLoCT0Ti6@8BMDNUDTo&uG1M^cWea%hy8_zQTZ*_6uAWoVsI~-i;|H*XIf4Yr9-*rm
zvLl#DFKg9qCq(G>%|oC00J0%J&h){fKefHNs1MHC&(HFK-}!C9h2>M1AHF|SslSX%
zU%j%mGbwm5J^3DiQ<5Ndj?1{|EGzey2j%nq@NYs65_Wr`N@PASX&)6wS6vOhXv1Kb
z;<2^d5o5aF9FRJ(2bXE42yyBWyeT=41bsRnuIS}3m+glU>h#$WO`|k+e&Fi3^82(g
z!0L61;4SJJgjvb>lM@$5?)rv-D5@Fd*Z4TLXA4}82axR#HpeOOaGHaTh+qHaesJwo
zt)aL)GVRt?{k<9Wq&kE^M!Bt$4`P|dEu(Y%toB8Ij;Y8xM;~?u*P_G?fmUgs8~T9H
z_9I*-q=BT=Qs4Oo)@Tvj6yq&^+`(Je;#7Km?Y8ZXvMv{D^t?--Nffi@2m(Ag!u{WI
zGY#bIdFXr`9KAf2R=Fa26)kjj5HI_gO*bF018`__HMOgF`&xexSc&`eJA{&T2-t`N
zB%@Nx2h+RV=^~4DZZK=_ad`-#iVwiPv8{B-3`5GIH3nSn^9Zk$b^rYF3chvRMswiJ
zKKf>Nep?8A|52fu*ij~c93XqKZ~P~reo+4EOxg2V4<0R4!R#KLi*V;Yx9ipt*^u_~
zB5hw9=jJ&jy~R~TRT^v`f722ZfQ&jhdr13wiwoyEcC=@;rr}84Tvgi1Q-UI50Fd9g
zh96!;*Oo@}7muJmBm!A%KbTRauye>_#v)K{gfGeIUmC3cE?s}t5#Y!AC7AhFwGKj%
z)1DCLd7IYzshW;;T_bKojH%$Y((5+fSy5;kB)VQ2OMAG3km~jZwcB06Lh%W+Yvypw
z-@W?W6ihfZqYxzQ6Kt}S1-O-_Uo5oRi=hPSu9bdXsR%g!Ckw3))WWGf1=>3TF%1_z
z3TIo|Rn>JP*KvMSS9cB#M62%1RPC8A>{!<LXM+GEbniPw>?5=mjQ-s!39hG7x^GeC
z6{bH!{2Od2&&R%Jns!i2B5+EDKu5vrXW|ba27+e>EQ;@5z1=RHFDr4e0%mZU?F%3+
z9y~=NxqVX>9aASHbK?P$(Acy3<GO!|h&|B73fxGFo!<QDBiNL=JZb3!&NFFXb-BZH
z?`l<}@U|H3iB(%ETMJ)u2Z_69W*9b2y}}V<mp?BcKs#3)z-HB7=8v(-Clh&oEW0>p
zk1P#sX`=#UvZp752pm(PKNsEGUSD19uhk5#pFRNhjH|76!pg;?ifuThWW?A?Y`(8#
zbu`ERIgjPxoJQ|^p?-e5WRZci*H;r{T4LU{$)3Gz7HH1Owp^e;s%4uiVynGw?%sQk
z5uTdO9I`pKPX>2|rP~L+B|-@Xc5yduy`W@T?n+|Lq>Tx597>MnqkGA?YxRU+mQZif
zIlX_~8c^Y^jf?Z_M~<CXvX!@x=>@4#<<KZ+gPo;tS$-i{@n7jo=uWyCMT-{z9&+BM
z0AXw_o+EjW4Wcb;a=xkl#P9eX2kp~VRdl2WkZQKv&GXmxKkdj-l5}ESPVr7%sS<#o
z#DYfex}3=oy1u+5pr9Tk3>wdrUofqlP0Z!Y`bpI@Td0LVs|}?BS@uqs8c`eAbZtpG
ze(cQT4IVgi^+fn5lF-`8%yZ31f3YRR2Dbsqx;{{g*Bwf*#m_g-MBBE3xj}j(bHkS)
z^xe6*nWj@D4w(}lkb(V9(O&m{OwHyu@A#zPtyw<*N;<PSnbXs<IcuA7+d66E?hQ<K
z_Zg1=W3}<e;fymK`hDSl>3&Qf;ho}NQ#cTL<^HrMxJK$v{u*%yD%Xdyi$*R~;to`0
zno(_lW?wIQ5Q@cbE;C7bfO~o?)z?*0m~E9JrbuGY1%=7>Kw%{44U+0Ro20jl2UB*5
z7bXnoeo@zaG!5wv#dsT|_-n(t_aR!@PYtTIM%)dcu0=#TQIk$8WPUL!c0hOQKhuEG
z2s%6bk4f0b<nol5Tci_6`(k}R)*`GmKT=e$#1w7HAoLv}s?Ax+?lY6#6J_27vD0wK
z$@uBEUQ&)y=)!1<K|<WJZ4*Z8+>NZR+pr`-j(neE(aYUk7hrb1`?UKgGN1Hfz=@T=
z<gKTd7$;pP5tBwaNMx&g16PU7I3Sg+>L0$hj4(H;SMQT8?W8TlPx4h+!E%+)GuyOR
z{K#{r2p40oH@%)xkqSW(44w-hF25KnE<VH`7{6h^iXmGp6pWToNGL3wh?5X*bn3+P
zkah3%p>}il3HrL9c$>nMQXhTB5F?3SxZJ}>&oKQ1bYPzsMJ8Cd@*1J$q+a!(8ETE)
zF{5V=Ef=+3Mq59xm$YR>uWn4#J%%su+}Py8d`}Losxs%E#ak<hS#Q9fU}r7Ra#JT@
z4y#=?^D!c|H}k~`t?xzSbFS$dU0KeY>!ehRX%HUbBiLO_yfVEbUpi!wYaH0rVN*3)
z_asdD&%*a_nN`^cm~-y1B<HU}wT$*kb+YJru)MjY^?7BC?o*tPn9TUBxWk@8;+O%0
z>QoiowgKNYwvb*%Y+h4z(;hqbF59(3vV~~&r|u7<H}km$ov77|RovNyk77?oJQtz}
zyk7Sthj9KbAg_8`RPB!?bJ&4j@vwNQN%Gz{hUnGw(d#QAs?+H|=&W2(DVi7h61pwq
zB-UsR?%WHbmp)A4_rr#Uz?`{1b2G|#n8@~=baPQK_8jkNJP)!<*vMus^$#4pZ87X0
z0s>)(gkvT)N=d-cQ9y9=@JqR%x&_YeI}Jd9-_>JWeEK8I$A5hl%!n7HnbdMBXbU<I
z^v+WI=dNvcluL|g`=y{mm~N1}E>#bfCu17xhWLiDL<X}yJ*d`masJCoS-g+Gw+kC)
zRd9{Yn+b+xvodI!4oFb^u{uwsm51_?{6a^Z7%Q3NL^Bvn7Z^HN5NAsz>0RWA4V!FK
zpmm3vTANHXSvTi69DAkh0O(<mSQun*l7>L&GhzL;na&`f$&uDLy?of3<d$i=iqLwK
zPLCRc3Kf>FX4=b){za$U$N$`PfQpj_#w{-~IVb<@(iKdAANLK0IxqXa0TAFu-@kQ|
zfWd}f-%ei794#8kra<+WCOdqRar~olxbWo(>GAyw`IPlE3%Qc-(FRXMX;$8-D>L5C
zMsi@B|C1nhy+U!ix*omNOjmg#1RbTUF)5WL7H}SGkR@!`69K}+NW!cOtW4I~co@m6
zsZKntj;#zn{w(P3252V;fkC!9Hl5e82Hd}$j&PKb+5FR@o;NcgqznX<s<fXu?xd`f
zY8CcZWI~Z8VTRq`hEd~c9Lu3^5H&sZL8*aA&AiC3_hMVX88KC7H;JXJjbtdVI}*Ku
zP?LUHc=;=Fv?<Op@y@Jf$$L<W5alwRWmH|FlaDU#451Qn735g7Xw5z-D=LXtq&}1|
z_yHl*a^x-V1lP_{*%eNd)dc0?CUhiF^A=UkQ2dIh4H~iAUd>50+$dkd;dyMQJ=T%j
zc@(1yCRe5YcgK4($ZsuWy_jA`FBjH}Q236T0l}W<sK>Z#QBD*mqjvLN5b0zc&u;)@
zl_vnvH9br}|EohyB+2&vE-#2{PZksmn`Fhl{T%!mMeZG3tFxU_aSa$ld*{N=s;LiC
zSxD94pF?9pT2Y!^5o4v&g4^wI?&O&op>~0!M%!H3a>7SFY770*NcG-V!$KM|)Tj<c
zs(G%o@J$~RHTogpu!(&GUvf2xPpQJ&YUX-AK}W0g@gvDrh<K!G5j*oPGJWVbsIjzs
z8()kn@Ry)Gkv>~|UdfTln6ad{>n%BW@yMr&e-yruEt4@RiW?9XQ!FVKKyQU}<08<k
zDTD{P{2Hk$t;#;3(Y_@)%3025blKOV7-8vbRf3Z_>ZphFZ5LIb<WpCKuey!Z$?J(9
zKXp0ASnjkJWB;qLnwUI2_RwBEKbuH2P79ioJuKD1SzbDuZq*i-8(mb@`k8NQ9<p9G
z?(GDCblpX^q(Bku$R}+6xrbBWM<CmFjSiAuS6dN}k!ma4RCd^zW-a{At<)#5jeI65
z0KLye$){AyV!HVI<|uspAW;!iRDxZnB<D+bCazdHpj6L>C1EEnvew3MTN|n^BM$^=
z8`lNs@7}tc_wm*^)qd6{B8$aO7rJ~kiRJw35MchsL(leI$ZHsQ-A?4SXG}L$1V&10
zUPrMFo=2|8thk|6!AZYYfQ^t@A|@Yu7J<!akxgB<bG!7Eha4JhzrtTvzn&`9?&l;<
zc+RoAew$u=iS80yI0JK<!+aAlM!P$vKqFbB>MD0!)Qb>rMl^ZaV9eliKKjv{koKo%
zs7F3bw2yz4zxpmk#CLm^I*7X_AWG!ISBM$wERywGs!QuonM!C>)n{M)q<|>mLPEY2
zbuDsszqx&8*XL{NnSaCph1L<a0>9BM`Cp}o*cp2DB|3v(;Uq@PJd6)*->`hrhwBMR
zU1~-WihfVfg$kw>aY}Gl_}E5f@nU)M7#PsGP4}Lxf3^*jT%xnql%0m%XM?#CoCf3k
zor|TEE{vrgKUbzXiUHwJ`ZvyrkC&n_PmoAz&M4Jnv@`#(H;FlO-r;rjBH;!p0Mo>m
zPM@D46#pt{P~&Y1jZsjKj&17{XejXCGCJh{ysY?7Ykz73Kh`u#&2vgVR?dExce6*F
z!Ar1}$me})BoG8iSoc4>GqJw*$%O0avs8o<jE+Z&YO!~ygXxNVN8E<myfI>ZM_k8p
z9?}n3-{lBTiWU+Ed|ebK$KT%<5%#K5s_Io~COc^au-EU{a?hOs?9WRoeopu6Oism*
zVA|RNJk-sp+Etj)&$Eqt^qCi)2h>}tihIGNJ;b9ZC&{iCV9&V4;&IL3sc^sjbZ4Ef
zYQ*XEOqFK}!{A@O#9Ualzl{^7Shs{NZBvgpybpw3NH;R=Ac?I5&7Y^CC?(QPQxbiR
zl}2Yeprrr?kbMn?VB_XDBa58RGi#jw2_f8nS<j)5GQn;liXZb69C<w3fal#^Uw>P*
zXZaeC$;p#k`R+21SRe4y`ps2WYJlk;hLN*Kcr=4>ABIeQQ{dK(-;Jg?C!lGu1QbR&
zoW9(>jE%VMHzFs-JY`e!AY!HGsi@)2)>fz7wS=d=x-Gm#Yuof<ONj34Huyz91<Fm{
zyNjBWNIR0+pMQzx`i#<2)}FYR(|q)g+JKZi#$+Z$#WMKVQX>FVPwQgPz)K9>#$U}_
zs`AM+oDfw?M9yL?p2q=|B67rPq#4Kn3^>WBm+kut*!`t?`fpHPB?#HAi_>ZU+O5xo
zC@|pfU$0X}9QnG6^s8<R$2Cc`q$zl%;DqU=-u<L?`Q@tI9yv-tJf<t*b!^<ngyW>@
z(*dv&tP>2^yxog>NaOcXn19;<2N!)mO(6}upEvBMdDcY+*VZ1}Ld;<*L`tABQ4S~e
zh{XLi@2*0EPxy?>yXcMQy4mDDh7koT*P<EhKHJXtVwlEfAA3J71yWZx)39MCKK)>F
zN5ezr`cvH`Ndc5J+oyY-lG-@JSA@9-pg8Dpc|{WMq;#=3<g|*BT6l;|yB<PdSIAlZ
z^6<k+qLr%%qV60O_c+!~Au)v#xMwejv#_v;nD`>^`qs}Nwook`^N~xpS94YK)INdg
zakj!FgCZz#s&p$WW;weHM>v0cIS}XWgM0YQo@Z8N)*^uAjl|gVnz(2}^4wDG>?QWR
zP|U^1K8u-er()?9i*Q9HAkGfZ3&+zL$*K4bXLv9|^tf?7GKs`s*PhufF;65?gL3%|
z&DJHhJ3>7;l~^5yrvZ!b*7^ME?Ngcr0bC$%Ot?CmUIqqUIe`!;xZ~)v5thEsD4+7A
zWQb-U`n3AaxDTn`)MkD9^*9nB!RW4SRvqwfgETL!jNFX)Geu5mH-HKHdPfm%SZ`*{
z2gbRg7&J{u{11kEa@lPylkjZUtfF>5K0%*Kc{bNkj#0lD8IT2JG4SQVv<Se7xqQv3
zy@o6qt{EGP2UG>0fLfJgERB@s27$*!kjh|Y_;)5Xk<?Jp*)Zq);y5VK;i1`L`5Yg-
z28b8CXpVX=A#*wm&jaC4uYsmB%l%PL@X&UO)>RfiE8(!obTH~-iUUG9F*!XuFfLgu
z)|BjkIWGpB9vrbG9ec7nUN7;0q+kcrS}q<jVn(_KJ(}|^?bB~6-~p=~fTYPQWIQ%^
zO?l_DDeH+rk;<AKep)E$E97TSO~IZNd`fWlA*H#>4^#x2hTzIsEk(Bzl>AKb#?p<S
zkvQWv6nCy1KWk_mw7O&r_lrbTtB?8csCuBUD0_s73Z&@ATfT`U4&H{F>|$^`TSTvI
z0aT5q0DJg>kr}yv7CDoF?^eTncj^7G=<JqVfF7@R?*rX*@gg3p%}OEN1me+(yC8Q}
z%xwa=U?1ZA>%gQm(pg|g65x$J4O7gTqQ>XAu<gu_96~>CpgdDP(3}s*;{Ny{u3mUg
zL1Ga6dsy<CjDmRT1MD_Hzbi5P{KG10u#spKq;QK3I!VXFm1yeI`mrUDI@<NlhX;|-
zX06o3Oss~{E(g}S_L0`IJUlJ!XJJ(~O}nV5c$xH?PK<Zz98PKu+7DkL3A~IFD)u3m
zw}+1~6r#2z-_u4hQm4q=V|0|uu2ky9;#kM?iRV47Du3KFrziw;RuL#+O1%bp6Ma_Q
z!~}D%DW^33)^3n;(tSwcK^(%mRV`d0PItL*lAjd_V~(-gq~5+rq)MNQa-jEhLgGSq
z(>b(L5MAV_zOX9_hFsPJ7Bb>K{#BqQ#Z9RhqYGuF$dn6|*+1>jqD$y*WvtL_H%B%l
zl8i4n_PweTACg6Pc3tk!yCm;i#HfFH>=PUCb+$NoS%K>lM83j%apLiWqD}%9o|_-+
zPLPlLc`gj(JPZ{^BO_}zRlSW1+j?m;kCjS?miLv34t_JX7J@~ym4}v{`9%%~rXzt2
zx;8up?Rs{GH5~|hJ(JDai5OA_$&k}Yp{Hq{KGt(x7CR6rb9ar-miHpV7AaLQ(k+?;
z5f3QztT(rBVK^ioeksuA(w9rcCootudfSwDZSO)Q+-5Nlqc*#C>HAXF+fx<Y{10QV
zN(V6Z+M*mb7FP<}%y^%gnZ_+45f%|KRBR-!A1&M@r9*Wnd8W&j-(_Qs$}BwTB=%2g
zn2K-|V(PX07)-rT51_K(MCgk;eCW&&uKKtO-a804tb}zj*x*lFudk+m*?KAQU|E|v
zDRQI^Tsq4gtqu}IQ=Ij@vV}XL{3Jmn3YGlJ;ifW1sEO>AAnp--);`HxdEtxLP*BL1
z7qrCdgD@TUhtka%XXvaYgST-u4b&<+ZMZ&|I}(w#LxID6Big<nC}DdzgskYQi`Whm
z-|9;dIW8z(Cp^m%4f}2l>N;a_ugYWc>$<CYDZwRCbq|31@XTQlJ1OD%VLK$-5qOF`
zl93#AO`4O-ry^eofO}@n>pkJiVFG<j5E(*Yb<<t|TrR2_q?xm9P_!0ujF`fSY>mPf
zHH3N>qU7%;<}OvyA-pnwt52z7T2&>BzMGl3;eV*@i1#B7OsuaYS(UonFE*K5eML;e
zoOIip1S*u_IbXTq>3&ZPRi6jpETf+vNOI@85)i|qI3ZfPGXQ+dkb23zZpu@&;h0uN
zL022Mtfzu$tA=<U<6o$5xlhFhZOfev6CD}W15XWM2Rnf3vsT90dD3BdW+%b&td0VB
z#K)QYcZ+izIA*ym+JF2f#|+c*C&vtPn1|#j4++`UEO^^<uhNgr`I?ysmqw`4-nW@>
ztN%(o*XE6%;~L2sEYgNa=UII*CZjSwC;kA@!tm{c4j|FCZMFK9IP81d|Du>_UcmfG
zF?&6{#vU5^DkSw^46_=)%<*r*18v?Y#2;+pOcOs~S=%B{(!ZHubKAW|%&^NuI|kok
zeJOuY%ozWhV%7`nGZGUD4=~QMnJDhCWvfo$U=QF3l^l5$`9fsWR4)%T+um~HO!8v<
zit#3dR*cAKj3NxP7j3{WbNQ2DrX2|kv;4_l3^TjJ|D9nLzHRKqE=Uo#N&DJ{_Js{?
z+tOZ64n?a$t2BvbDOhZClNj-k7;&#U$JjEfn9$TyuN8EGVb@)X55k;0!CG%8&af|6
z-oH5<F+>86Fr$+hepZPSZzeMR@^!#rIl<4KB)LfCUMVztU(Pt0CR!A+dB(VakRD?+
zZ=^M<B{&k#8p%tiVv;k9$q~%RCoTv?+!?Br!)8uk`>e|CrblGvE&=xP(g<GU@Y9N{
zDBHa-zOgNKsWmp5+ayVU&so+b`w0<=7BA8Gd`n&~$0z%?PG^0jsEs8pQ@L2dweJ3O
zQGOz$CzhK{wO(|%t=1)HcFS+V^IZ&`tcjPg$yJ2|xdHRMg54z5&k6X8R_j#z3moma
zZQkKYksC$nDhmnpj#c2Ou(?ZdX)%7Thx^Ih!A(r^QeoR>;#Evq+J@`um1!9T(snrv
zYs0AmvsJ=`T=i9jMZ%0N{?@N-r<{lUD>|>0&y7@ZP2>6(IdiGg0V_@+9(QdZvj6=E
z^9(d;JW*YLAxT1hcs6@%dWQCxV?QBgX$H?LTd$Uy+Hl95my@tWG9=3^8Kz{<zM`Tf
zPhax(go~zE?Nw#Y9tJ((b=2CCf9?lI(NMlv2QjJrqPLKpJ>-7DO5uZ;31ZWs!r}?S
zkX9r*t*xe!k4SuhUt8;)E}fNa`jcBi6z5o6i%Iu#8dhqZAcP5FfpDHUk{|6_WX2{U
zCjQSpA2%crD%jYIPiDUg^wBHNJ6u-2r!D0Zit{?JnOeovFi2(J`gP!(6%jJ9W-Xga
zeJy?sw<P=)K2pZGwPFQeXGn6(089@fXU0JCKLci7(Pf1_B$d4?vyC0BfByBqLsbI<
zphKwVOwu^=CuVQThIh8G2&OGRg^!@G$$1fQU1&tm?<T@UyEB)P_GhODPFodtTY9h7
zg7~BUnPZ)80pD*Xj(h$u&QQozHtP{lE{mT0E_-eyXbEu)`#|%H2?rlgVIBUe^16uV
z@!woELgcE|u^Hn12B`XbTF4Ct<fS|KMwI{am;V2E{qN@T|Mdz>b9&PICkQk%zj}7S
z+|9-YlITXs?=Y2J-i*?9_?lo$Ruc#|VIp^ADgXJNBX5>XiL54HzJkAwK?Q@uC@f;t
ztyyvn&R|V!cY>lB-$YyDELfUAw{0=m_r@~*^ID5TwDtoYJTD$I>ol#Oq%tBc5f8K)
zWJQ_1yWiCy^Ue4^^sT#DXFnk7s`i(VRFHwuTjA(?`&fciMX9Y)SKq<A^7o2w_=F|0
zz-F~1$#nnod3&!UBt$WW{u1&fy|VM!MUMVQhmQ2a9zIuo8uJZ|HjN6wmchP<#ZSHV
z7R9=hWaJwo$gF(O@%+{TBw|~GwU~R`HAEPL*!&FjY6O9p*AhH%c$#gG9PSoEQUn(u
z+)c9JEzDKR^9svy$;#Ab-KK`Bs$AB))s^Q>2R7!ueolmBupFGB9FjS|*^VBzV_!e3
zy^$Jl%?7CpgH;WOfH3D@Qm)k;y>lv0t$d2~chAOicG#19V{VgZ{{qg|MehJJ`T>NW
zEkr2G<z!KCb0Opmm}{58_@Zso07lvft9t>glkJ)EUkE3zNgJN~pK{8M5Z`)Uh<_jQ
zB`b4Ik@IdDL6QPOP0>_u?X}>8Nlse)4Kt-&VzA?M%t~~KD+f~rC2U=S#08e>*O=oy
z#$an4p*$4HSZoCZGW(DDZ~Ag;`SRav->&#;L;L&nyc-M~<U#TG6TglQPZWW~cUpO6
zRG_MK8Tb+5{wTh7)MMoaZS|)w#YVarHHJ|ZzL<5y`FwE4fXb@{vR1x;6#HDYeu>Q8
z?a=5)cph37y}FLNgErD1TJ<4usZc&yfQD!v9jAkHDn_b<nUm_JJBhq{Cmp&jy}y2`
zFwjvWR(It8;?#!%C{Fs3@cF-PUfr4h;*KdJbajL~_5htIvZT(;-qefOq#dN`7k*##
z;OrjKq`~t5%cmzpHkP;#WCG`y|Ayf3Z;(la5?<WRMZ5M(-y9PCNO*x^E;R_)qWRij
z1lSLw+_572LEvj97hO+DSV0~^{hK)}U`|x$6Fj5Z$&p(H={|n#bvvJ+sbJkuHXTY&
z7s#n>IlEo3+o3k)8)PohKAq;CEt8U?GA!|e=`|$?FQU6Dx$|~mEzN)UJ_yz#3jl?<
zsVRP?I$xKr1UxFgx61TYGS&Lo(i;qZA_C`^Z!*~@ef4Fdt5q7!2%b|WG`26?DeDoJ
zG_K)@bE(b@StRPN5u8cxBTViPjp<EUq!R39U(;M?sN-q8-JOQ(lk>FxPP!<r8!Qey
zuMVqaf?2k)oS~Q0E%U<Mw1L3%645qr75@5=JZL7MslK5!q?ko*Df(+rU-W>d?38>b
z`vYlu$`vVti+%LQRZMk3TO4G^zmgA%dwQ-ID_<ZQZx~98J1EOacQJT&*mrANLDy6H
z1!u2fyHBNenMkr-OXSF&;>C6Gc}5=>-Y1na?h+~OCCjIXEPq!xA-p~tWOfil!};sz
zo$-DTsQCyuyU+L>L)t53&NlGE-~-OL0?L|C2$|J>rt9MX^kfe}Oeb(-cH^7XB9aX9
z7HEtn>&x&WYY0V%K)e9@nTl{DNNoBhOvd;K%n&^gCE*(Y)%4J(Jv0m`i>mgWS*5U$
zZ81>eLo{$k{PG#rFkV5}^aem;8((&c?&T<AD-jKh&~U!F7a#h^K86oK^E5!jiu*1i
z4XmvC2YCf=Y#J`xK2wer76Y>nar6@;c$k=&7n}t>Dr*Rt-hri7{X7gw5$<{+VhPWD
z``Lxzoo(P6mqCtEIefYy%a<Q%t6RFPA=lk*Z)cPB%Q1cpe{y8YpxjjX;DF@UY`NF1
zhP6HZz6f4SS;<GB>nU;pHpc)MsN(RAfvaW<|Kj<UYe{s$CjF%bx?9scNt8$5b=hYJ
zY-omWP;CF7yf1U{UHApDp)rtEErN@c8W)s49t9mmKo=|42B{3OEe6O=f=Ly!FPuRW
zh-D9E)u{Hk`vEZ!+d<O1Hv!Zshv%pZApku9f}5?@C6q66)bVJmlE-Pk6tfU9;HuhL
z;G-1Kve33gk|meK{VL6wR7ai8Qg8jFn@O`)j4jjSO7P_SK15{g$;>Ld48Hd~G61&r
ziM9_ck~=P!fhS4AM2M&o;(gXf2L=uGw=Y?a%?1b80mL+a<G8Rncwg|rw4%m%Ci{mg
zoGn>+IdPXvlCYWD5VE6W946;~ZZ~I$(1#h}VDTaxtj%nx;uG^vj2L!w85~qb_;cAd
zATCpTs;qn}HDGr~!9vj`WFoTX*<d+<2(S3S*I=)V_$sdJadEiE!OqYERIxTAI{>)z
z`lR=ki!MF~ielTMOX;8|d_oFh$wjL9ow#_!3%MNJ@b+8`fR+YE5tRE&zFn6KS{*f}
zi2q$x{%zsHC=RFfxP}tRpF8GrIeZCB7tpd<k8QyntGFXMk^<#P9db}VfH9RFicRu0
zm4ONn$@-CQ6G_2fy@_a@JGU8z1H3&^FTo(-Q*`u{&{bA~00`lx@wn!xijj=Ca>gHX
z1esHN)nX+jg?%~NxT)u(Fc&1ha{)lMSq=PYFq+(+EGnHKqr2rL8h{#*ctp3Hj3<w3
zka(9}35=a{K@w{#kl(=VL#GxNVh*76LqjU6bWs|3bRLuHN<O-P<OUTf(~A-?b?N)z
zb#JzyH&FP%3ao0vMXr`<@Tn+*YeJ_tCF-=_0XS5IzDj&ZjlWOfwx-BRk;rKg70xD^
zdd^Q>53J)h!5pTUNCCi2IoA3U?P@F8Z_rJja2=P=*!t<!${8(#j~j2RhhIvMzK0u~
zXX^jr?!AMm%C>h=MUbGtCJRVLKu|IWl9LF6faD+<0Ra(FB<CCi$w@K_N(Kp%BqKqB
zWRxhN<ebqrR=4&&zjOQaxwq>5@#=L^UArlo&0cG+Imh_M7y7{qeCdz&cK{&?Eo2Y*
z`{r+oUnZDBWIGdlMoWm%I8v4kM~+yCYF<Zd><Xnu8SE10cYx0{H=KbUx)|pxOtGLR
zNgHJBv6VC_5H<@P>HJbeH%{gWELEqC)cHc+Fod#_t+>9UzJ#prsE9TAQ{SXg5;{Hu
zHu*S{QS{Iobd51KzBa~@$Wxp=K|Z^gz}4vuzl*q^jYTY6>%N5jCi=%_T>XA6<EqRG
z;vhn(fpE1$_%2O-68=c+B7bw}jV8ydsx2fc(EVnf`lk_6w*5~!L)DUKwwG3l0q#vO
z;2CS$%$6@wcwHZTQa~)kx<NfieYtM~p>K-**f(MBDVWLFOF&;8yT)fw=3(g(iKT`n
zu_7y%atPBwtNKYfy;?92eTP&sd*u^XOH|h4kch^Rs3>WVg=G6@)jxk6iD;~l0L$hb
z89%(`sC$7`-FK?Oy@W$Ym<(R8{TXwAi)x;7isnD}c1U5v3Z8hrw7=Ug&~zYvVb6ok
z(r+!2B}~27U5D!VNO&%(qPAUHTyzrME%-D=7!_EC-60~9X4(u^u#R}rBhB$V$RNW9
z?>ftpzVK|g$FYg2cafup$^ui?)(MZDf~hgEsorIa`~;@H3>buQOQ@MrkKJ*|CIOo+
zh>l%e;-xcgeev~HG0*c`rjqsT!aUg|gtA2I(oE&S3T4j+>y<}*9|{TvaJNv0yk*FB
zCsMw_^icj?OpLkYZCjrpFswt#fU!e<s8+ni<$kv<&IAG5erlGDMs&94*!Sp(?#eJV
zcfD?re(8SOq97@UZ>28Css@)SIH+qjwf>d+8pv%$>mlp-Ikp9GCoPWSV|hG2Q7Bb%
z_UJx$+U<O2BKf9W`1n9H0gHLbz5N2Gp(lrl@kl#*Io&voqA)2kqMW?eU@%rIudlY`
z7{AWN-KXC0^8Up#vilrdr|+6(yel(T<Rr1A>35Hc-abyWD)_#T>~y1>)tJ7kUD)u@
zO~8I7r)Za;TYC!?@2p|}srk+Gd1GPALRB%vZL*$^PWoNfm0o9!)V|b<jrk^3^#nFd
zid}{7H^AzYC1O%&UyPmi#&%;MioL;v*Y&P&<C;<`;#QamcD1yJ;6d5n5#K&qXy1*w
z)uA;jr0i00VkB;w(a_$U^E@blfr#wr`i}dc%aCxx%eUzvV~zo4e3EB>7m9yh|C$j+
z44)?a{WG``p1wGTDwD!HSv&3^YuL)*J9m)K5eIlZ`Z<FHFfpOu2oj~!zA)Y{e8#V{
z*(Cc<b1~zSj_6)F@2f#%T|%mVxQoq@1hFy3KG$OS(_KswFUYIwmLB3@PiY)vynOQs
z<80OiI%QO0`|Wdu$ts_IvR}2%CSyp@%>8Mm@l9c?5WIw*JBfyDVo&ojGhUD8<V^9N
z>S@$%D`hy>usAI764hBlb7THAPrkLyUEdv&p?D|XZ8vWG2ZBk6VwWL^?2<liPs_QU
z6p`@|l}r3@!qmEFRbEf^2)~fo5aIr)Xs?t}mt|_yQv0w|@)YxzU8gZAM%tLl&YJn`
z7@jpZ_Zpb7#dXBR-*T~d89&-sd;`D|#h5qWv*MqmsA(3vD58#II!XxLSSp^$*T^=d
zTyCgzl2%-d?BQ7}aHnFfU4XbC*7{WWt~j;?&Cl(UH)Q8qu611dxm{C7c~gtku2eeD
z?<L$Kk}>|d-!#d;FU47sp*3mLRw4Ywp;AX@R4eNB)OIWBt=9JD#di3EKb4Zr=p4n_
z+jz%W0x8AZQiRWZ=rpKP!){1zzI)K@aNhupDCol1OK)KQ?%2{uW*+qL(I7`p(B^}<
z@0N;lR#I0~Lcz#HFBHWRs2M4JIbr9PYYB`v&F;myU0pLLbkJe>MzK^<l=s}P@HQJ6
zx(8wws#E9pd3*Ls`11{3$Mq>4l4HvPEsJkAsZXv5+nO3VX2MWx(&cj`t%8-RK9PMk
zm<$Y_^SiR0x2M38o;dVoabQgSL=Mbe1~skS`NCy$z3UOXO=f?@#cU!0<Ex@d<E<}o
zvgNL+^(QN~?s`{!AL{TC+F1S!3{l~zUgkE?a7o*}dugRprvgGC<y=H3Ww-!hO)=g}
z1oDla*`1>fwsL&A9b#44OB0z!J!QRBrC4`A{`LZp`y$t=>O$y!-*r~b*a4fFXKU8%
zalu~cgjbV4Zf8x5!_|h=c^|*{+nt5qGx#Tr_}+eS`{#W)m>J+j$9uxR%`tG$Vo<C%
z^=R=;fCQYf_kg(}$X8>Ufh$h>%oDnORW0B<H7>GVG*ZmH(ASn7r@v?uOZ(>KW9S>5
z>^>$7+Lr_m`D<ld5GuL8JsE23D=tw3S&H#`gV~#N&lP5(6gckXqFqY4nMF^KEUlWI
z1t!vR5zH^4MzAeIGTj=qv<x=u2CMANZLLHbQZKYe&B(Lme_hB;iCVbkHx~QQ8v1#C
zR}Gvuk}0|K8!YPUWP3$%AH>mwv5&Z;Qu=BJ;w^u6@2NOy_!&?MJT$z=H=+6+JkY1N
z-eTH9omW>|885heX;WU?g7w}D=0WXceIj&^>>`Ex22L%zbZ{dUbDCXvmlLiJMFY!1
zM=XHx7+0?Gt`nvjLiG~vqXzFarx5vd=3@8a%U#uw4@!E(GWTEbAt?kOLh&p9I__BI
z(g-Cnf`*6-G$qA?v=|>(!3?Li_dawt`lgIVD!}f62%I*Gjhh!4=$b$Sp=L4)%#zJ}
zuaw*UvY@(cJ;x75zVYfc{LO^GI~h#>d5@R@9cL!Sb=p-*2rN&l%~fs&&aErZ$vT~n
zD!LDFk{@votLjy6F{cR__j{7LTb?Jz(J^{^$uv6cj*+kn=;&<nfh@!@*RlphQ5d7L
z2*QI(RCiSJGgzIdt=`*@!^lJ<9~7p3a(O?h0?}(c*RYtB6qTEGhR8gZjhD@7n<Fo;
zU)sG7l$UM9xzEMv)3LHUmr-Jf_JZ7}jFH;2AWCEL7ukYy;dWzU?5iCPqXp>wB0Ds1
znuooesPfAKlV~2e!KxD(SU)yrrkQ?k^M7{~q%|P<9#a9RGk@$xhL)F<3#OE21pQEI
zjvv53ZlRf_#~9WuV7l5_5eWy8=)r_HSa_{q<!%HwGRXUGfLY3*T0TB=UP*$H(T*W7
z{=NWRgRevx>sIkTv^Z8*-u&T!^}T_X5lmhl_peZFy_Q+OX4Hr-W>EcPDK_ePjr{%F
z%-Bt)IhB_mTNf}+M>Km-U^*@4{jqt+dDn9mGCsb_Y*{}T4_L~V0L-D0?i3NB$TkHh
z47JJyv6Jsg&xHk9(I2K`!?n3dQo0v+M6DhovCeG!OsB9aa<?85kZvHH7#)JJ;+m&O
zf~e(z9NB9r`V*qt0+4CLir-*>gYk>z^!N03uN!jEP-I@z2>k=D_saGXuafiwyzg_!
zt<4}`Ms}b=@!$|1gHTZ8l8L5(0k$FZYn3<k3%R4DiE&&;z;S}+kb9Nta;bGC`;Ber
zHkBMnH|wI6iqh+lCRa51z6G=U`9HZ(DR+mQz*6jf&Yy?nM_1!a;Oq^#FscglCq^!6
z7a2-yLv<@nSUG2s_pVtxWeHF4YXpUicVea(Xs&b&d^ne&%~4uP^vy^#PcZIo@WGUY
zz;<VyQF*=XwMnta2)<O>JM_b@-u2Le@-9L<y$#~kxs^Mh+O0vc*rpTOYHT@zOPi9t
z;*5EIm@9pkr}kV_KPaPW#8r>F5o2#LR%WFH!ecPY{Tx!f2X%Z_oBcdV>BlJ6%MOLu
zlFE{YqF;xLq+P~bB3th|A?SjKhg?@e!q-td2qpSRNqzTd9>k-^pP|KY;eK<onq6~U
z^TSguowmxqKNEj{H;o?Pj}wpGk7t&HXAXa2=4Q~}FbTp5%%OEaC*CpV&=Obi!Ie<C
zLPv2$C5Yr4JzAf%NpnHar3{tJ4t=MLWl(4Cqe!?|cO?nOk9+nQOMlk7JY93?oH&|r
znOqa@F?(tb|A^Y(bD-HhN{Pa`p&|Jz!5fIp_&@d6$w-tMTzUas;g(isnU#^L6K?cr
zEFC~PuS?3}<XAc=XJY!ZfqD@Og<-p-bPfn1qMAwPY#ZvljEwg5b+Ae3%kVn^`rP@g
z?SP}%>(+F=*Fg(<<?(k=DP64N9obvi(haoiT>zxzx<$1Zncj9i)o09U!9LAEtqjR)
zdI1jNfj;ZGayFhQuAs{g{G93Z&_lAmj(@B0+dlK;aH&xj%j-Q~a=?j&WD5Qy&NKnK
zxFOMPmNTR%Z{sddV#Th1A31^NJ1GKtwhspJ0E(=&#GW%x0bXe<A^!!NGaUyGw4qFr
zc`$%~(J-eUxDEQ9vrZ!$-t}<2i!n@|`!N%5N_=Jgl-v87_wfNst)kYajq9M7X<Ocg
zMzMos)^-jf9zg|S?`E7|^xl625RC$H#^F2pQ4MJD==aaCIrjTdH9lxGnHi<Ur;9Qg
zw_BJ#Ahe$ajaJ~@HD&x(hl9<zRYX0Cv}J|YO{zaIB~^kpNJ|$6jwMriQ09FB5aEHO
ziti6ac?;d`xHaLuCpQW?O!FJwW|wTQVA-A{p_&;pKJr?UcY8MT!~GeOqhq8;uuEl6
zy+rCh0=mH?#fT6m9Q<c}zE8+{VgZO?x2hG_hS<c5piap=8trkPr=<vc$qC7csN2HH
z3wxQve&L@S{r#3^|2e)If`!GMkW^0Ug1net;hh~fPSN}TR*r=vouj6YW>t|mu6O+z
zJc0Mp82j(TeXmNhqbw!9EfCds=L2Qx0h~GWe&-v}vA%*^9R`v%G7N#oL@DBB=maFb
zcj|5%V3uRm;t+HpHW~n%+IaI1U>J=<39(#<=##+1(|EtckjA^->kMXMS^;!P^WVS0
zopeCe{9%9!9E->80iJa166@$a@)nstr**c;-8P&?>HrqhMw1-~S|X4v1%3)?Ihp>b
zD-Er(lnZUUxXX;Nz<~q83`Qh=4(sdz4hPq#R5n1n_yj2Mua-Vnm2#vMD`wJ6(X4Xp
z_Pqj3g`4eX2uB^(Dt?cP;VB8C%7$v{_$Eq{=1UiD9U~${L@H!r2c~6bhZn}aLlWL#
zkn@HV)|@}#*br=<rDDDv(v@bPX@narS3g(rlU^Y5twhkOeEGm{HK!Ip_t}opb846b
zCs0f8E0bp-w=@W!E_1fouTC$09K@{S0(8gouz#WXuRA!F`unYbYtATtd?=`4_oWob
zZ=+sLA0x0tFO4HQ`WahIB-qTU4n066^M)UX;*)xvHU@(RnxU&*Znl)*(%Y%UaJn;2
z!a{w$*){MPW{07hA1!KYD(u<>)-vHPwovW&mq`lye%v7Cb6Xa;2tf(_1zOjN{f~Ho
z0Bl(Bbu(FWS$J>nvRVE{vb@>%al#2w)_t~e8cR`G5lpPm>F=aH-9S<8lpZR|rg9i*
zSjH&Fagme9>c&lKc&in1?2Ng=BD^~7ESx!YnJ=-IW?*I5J&&x2%VW7huBZN?Tefgh
z<Icm>e@u<_M8BEliq5jnMx#5V@N17VqdT=Gfg%<Uo%2|y{`v_l?ko@1H8AlNZlWyq
zt!STZPD>!^2$+S+D01-u%y88=c_qPj!yl~|oTZDb$8SuLR3=eGUEE!poa~Tb@!^a?
zZKMkq_{HN)U#4>ozNNn&V88yD7s7Ker;Pz_>q_c_C)`SsYMX(*fg#S;L`wIe=S^02
zx_U5g(jb%U08wr^-V7$JM1<jYPLA?veovC9+q+_P)j)O_PMJ&$_mT~q?zx59Y0kcL
zbox1c4Htj&`&1fWWcG%_p?AHAAUPn17>zrW%KgG80OJz#(`cBJA4=x!^W<=r@tC%w
zs_|x+9(q;0bM2pL2|<L><|}AoSZZ8M6E~hC;TZ7|ox_+C*=scAMV6_wi5m-uP`C68
zL?IA=ZJIUH!QRE`iJtOvI-TSnz`-d_T@%tU9%oBb-iI0o>*^AxQ_sl@yc~)2r%)QC
zDYz<R!Xb!1gP>oI<#sN`AInOhuyQ7tJddmGgaTU>V!e1cQmb0e%+Z!Ne|}F9QbIMj
zGzRsAg{0X<ljI_I0P^O@jji>Th?;v?&h=zWqjTjLu=S=?s&xqO?ZlJ}Wqs(cXoJlU
zLSw^uQ#19~F`rYXgs=YyT;I|PDb{E(VMxhT>B*?;aX)8wxu&hLfV0p@v({y=TySj#
z=?>}~TS=$Ajh?c^Ed_7`1?o}j+diK=DMo$o9DlEqzXD?a{e?0$TDNk(Gp-T+>0dGG
zIdo`?=?ZGwMpuVWB_%*?2nt4Db_7r{L;t0;*kK6HoVR|iviN{YLJpRaS~B6+%sl6G
z!kq}FMG*xxwM(p#k$}@mshwwm<M3V1erLy<r*sHM%g3ro$T3^FI3*{JL1j&Kh?;~E
z+nrShh5O_JdrEQmQEt&~0OATlT}i|PafrD$XYkZKVJF(ef+WM$$2X`Km%y+2hW@^I
zOGpM9o)dUhhXw`P=B=<kivrE_bv<4?13#*ai4kcG+6+1oGjJ>B#AA&oBDo)g6K-4x
zyW0iXof%?X4_;Ln*CGC|nUe@9Ip-A%+u5W$#lc~<!E#@LCQiB|U^bhjkJo)8W9Jf;
z6^pz|j~BP?gz(Z&&_JL&B-u(<t0mu`JzcHYmt_rp85uL8R!WuBMVf`rBx0RoiAkcz
z?Dms`A4-c{)e&{{qKtH&<*u&qiT)vR^hct22EDRLfMB|@a;f&t4B1El&TW|%WAdjr
z#(O8sQ{6NM?6Y#1IaIjf(`;F3iFclnZ(o{i<fc;l<mh>N{-wY4$5GupTxjbnb1KTx
zE)Lzy!!XiS3(v~%E>BbV?6@BVWYCwwP_AZ7Fv+w;WKdJuoTzcw-b>XIii~vbjs}MO
z*m0BadFD{AAyJZQ|5BdakQWPek#~j?9~Et{G+s|nF_sb83&PQMbC@ZQcF)?m*mArw
zT$q}_r#0)gX_Xu7=BA;qD{Q}UQT_D0oKpky{6m~O{fS$nr*)foL^T_?lv)BaW|s6?
zvPR6cKFT{q{<j^CmJf~iINLvx<T;$ce|!<&qLo|Mw05h34DB>al;@OEZAb@<qw<?j
z+__K7UiKYa|9+{tbrJWIvpx6@1o<~~>#l3!VgaLhu$RBw>%YBYnkY6L^Qp5P`oG>%
zT#+Dcs_><MNZ2K>*K}!M62_LjSC>ndDkJq;L-A-N!XG=7N=SpZxq!qKAcC&|Ve;UY
zA2{w?Rqk%a!T)+3>V*00;en9#zyI=5fX`{RIYsd6Q$~nmVAEb?Xq2KCwgSW4uQ1`O
zsAg!s2g7u>5yFXJ`0S+Ev^BhqvxmOxqeRTNw*x*8o$`Vma;z~YAn7_J{CprG6x1s(
z-X$m@qkBwHa1<ehSB$C=eJs)_Lq^e2QGD2N6#4t2{&{Kt_kRyipT_xd-?nSS`1jL)
z4h?G-;lXOZdz>4@M7z0_E<pDqiOa;l48q>n4?AxE`AY`KPGjBj*m+~{0h(_ss-~@T
zJzU333h}f6%e*PXE4}vb8_i#-^S@uHOL=<Nzb41w_x*Lp5~;mF$M6B7+e)t2wNhBM
zgYS9iW`e}8qi#@q3V``I7|E&7>W9AsW4c=q$DKJU%=DkG`Cs4nQiIAa?q6Di|N6)M
z+YiiL5^peYIw*1Y4|n&kclr-6Zu*_V@W?eZz4ULt_y6=imGxou#Kdt}G5?1*`1?io
z*AM@H@d7BnKu@;!Hh?2{S@T7{b*1`Ff4IL(2Vo4Oc=c;Wn~66lG$eM01c|1?smfdI
z7`i3)8t)3F^sn{s_t&mB3!d)btbo)Pzb=*6A>ApJ255s)2h%n3b<B_e!?6>s%7^I?
ze`WCQ=^8uusoh$3-mbf<Tb{@2WmtD|bD35tLj3&cG4=D&ge6e$2oYU4Vx`qU`arR~
zaht~{+f-NQ4(xxP>``_??9TooYe|be@d@@Ton_g|lh6O<CV4qCz-p=uV>y!g^-;pa
zPwAJ?qAY-)!|?f}zQPq8UB1FNC=MJU8n|l~3J^<3XfITY9VD)t##_(PRFJQx$&?L`
z1!uet=t8eRPyWC-3M7L}Cas~xTNgS3&X5ABS%g+yKlA2_Z1#A|(I7v3qIa_p4T#c)
zT&1EPJKy@GS^+{|ReW7kUN&s=xERcrc7;QTO5%tbK<2^%RudJvSsNE^^4z1NR4y&C
z@4U!}w3(~D38m<N{$iDd2fL$q{>87KNqc}@1J0z}Rzp|FMO_^;ckF=7_y}AT+BfGq
z^KoH}9|`QM51>tA002M#W;N|I&mtsmsvoe%jMW4OF_^PIB03JE;A4F?g`d~E3k|Rm
z@XD@&OJ4+(h@^u7kMyE##X9V4hqUX_Ei}JBTLCc3mIUEkmA?>r8Vei=?k`DdMmCm(
z<KO6g7g0q`R>qe;LHW-D=DNkYFaP+AYLVJit1+!YGF4-(&T{|x1H$K`h=T_@tTXlL
z?^Oo&WnM@tr(mdgmFnGmqLqTDmRtjku&%@uH&6@Y*v^~eo_M~h1KhKxhGw3j-St~~
zK$bzhyE^ke&}jyZ1FU7A5T)JC77|4r8?96WG~AzG*?OPt$(I{X=@O4YG&2YJ0M=y_
z!?(mJa1Dd5v9}4y1S!tU{rz+B*DfmwH-nWNi{IzhExVua9s=r2p(|5_Leusp_#@=9
zHW;KksONjvOCB|B!taY=76))H;T2u<o8xb0lg0Xn&%cz8wO@vJwAoJ1XZ&kl`MVbu
zfpBSX31*yrpD5gDSR4MF9wP|3zdui#=KwTED;JKyXspOJdSfnP5(HChl*2C|;7;Wj
ze}{gWVqg*ufsv)gl?&Xafk7vzv1{z<ia8vKuGP=i*5<~&uC975txE|hGfk@HDL3qO
zPEJ;IS8%NUm<rvrt=)<BYp8|i;VTU3yI6zWJ!R-}<K6dneb=Hm>T{-ghW&ofc}cO8
zPP@duTU(<lc$Mz%TX@>);%tDg)$;%@|CIyLgX49##v_&mk4jQ(lkb9t_32;7rwj4z
zrN7jLRarMzj1H?<P*U7Cky@}b=sB&M+Kb28%#lhwj6Z4nP+UZSU9(Z7aiz`ep;D;a
z9UOxw|GU{Oge#3b<I*PM>UG{?Wy2NYJq$e~;b9d)db5Pog#y)^PufPK>Qf7w;&$5H
zDIc(HDr-vd>vX-NstvE#6zvIpXyEFU8sOY_Z?ay1w63tvCi{?JRcI`T-&ojfypQ>7
zrqVwA(Jr{lXBUL8iQPz%8YxhYTOeMs-@Y(>P`>}JPgE~8#H>=JlD(#N-+m+mFS)6Q
zd2;?yjq1M3!(0EjqbiR>&-rF$!({2tM<S8d3o2*j8{yP8H08kEF&TaQNztq;o@?I7
z7QC5bU`vu;o2Z>Ds9uc(BR(CAAt=vv5cbk2M8}r~x}EF{Z$kRv2O=e5I?$FyA}I@-
zz#h0o`*{SMimnn{w>_gO6}4vQnp$(;6FOS|0cybyC1#yp0~wyRzosEa6tmU<i)k<P
zh9l#vO`u;3kqbOb13$gHVAe^Ga8dlYKO6esngUfP3T&=4V6B92gYJEO!wq&H(r1rg
zZz!KEx6vIn4!LR?ch})L=$hrB`YP4Z1100o5Ex2Jf+3#6_wkBCYoK~+pRDwz`-EQL
zqF)nD^EN)T11x2%GJvuOsFga^jZSt)`s;hZfcs)S1!A~ZWEjQ`oDw1tYXF%`z*g%6
zj5Qd+ThN#^r1E-5H`_~^DM(i68NaF|FpXXX{6$mP;T=MiGm$`9SD<T=&LixXij9jI
z-h)8j_#|E2S57J<*0@8Pw+#xXd-j{)G8K?l)-T`xN{MC^puahA_vNiaoC#JU5d)F2
zemD`*a7HeTxxz(1H+G(&9e8@o@p18)7ec^b%na*>cZlkk*ZcXMVS{V2a1=*^-boQe
zdJNjYrZI)w#S%fh=T?wLNw^kxol};RDSj49v&!}0(K(+mJ{<d5033=}lt4_^>?|+R
zH{uX;_7-Q(jp`1~bu#y74Qz{$!*T_y@O-r+2x(TT`rle<maZV_O$&+I%g2?l8Df+#
ztAgRv93br50S&uAqJ0iXCnmkh$UeDwG94t|+*W3Kr`q){%qdK{{ad(_;{-yAF#oAS
zy3`2-HN!Z%$3o2}j-m|sJ@<uOiZqe|)hq#9q=niCny+b$Tv&E%@4=Yv5d_IQJ$^Lp
zN0vxsq}af7`U>aXI&~Q=y-p7IjdQy7v_EdigYS~ZS0DlL*4fr<_;(8mk8j+52zO%q
zNV8uEd+2eE=7H}Xns7j|Lr=1x>JP|08?$!?wo}p0R<9I3?$G+9I@SH%|Moo1qHl}U
zCj51P-Vb~a!{8UeCdtA6$<7saa8{M5PJ0P~PFfG}ee0je_8;vI3C3~_8*{PRavj_A
zt5@HxLMhy{nCrgnz-suA^|KuyX4Or_INP5gHo*8m?BIh^a<EHDfy<N^W-^YO9#|#h
z<1S{m?%G{E38TEmzT|xJZAHz~%*@WC%s@tS=m801CF1l<Vl8+gzKZavIyLEhj$k>z
zP}kIJUGw<qQfX|@8XP3_BIXtKlh3ta$EMbNPmIL03Q<7j2KH%rzQwux1=^FuCoAAk
zW_hAq9P&j~20{_vqtEWbGBCZH%9h#dp2ki>Z~EDwg{GB%*zH)5<eC-@wbNIE;Y?Zl
zQZAAMLEQ0WB!0K_^2iWOEr>cBhof4{<K1)8T3{D(9tr?{weFF6-gmY0OKKz2Y8MyO
zMohLa`OgXkth~-)D*w>(5#L}G813H;Z@G(l&jJl*<vlTfwxq^WFfP+}QUAQC_I?SW
zOL|nPZiFM$LaKSX`5+wi&oCjdPBE@CelUx;s|oE$lIX?legv_N`JPI{8&$7Z)$|O+
zH1jh;q0)dM0VbttadeZgy!@>=_9v{dT%@eu>{-8-SoJdJnOvah(k8KBQhzS*4tqYu
zl{bS0oO`zg=c3QHSwfCggS|qF>g|>Zeww1z8Rsq`lCng_oWL*h5_;L$UEy$VCyl~9
zyN%o?z-J>Fr>5jVW{TB|;AU&QjJ)6lin2cAAsdpc(Xp?osgA32N9Tp$G4Xza-=mU8
zwr`Yz#Nw@jb3JDr#6rfZo;J~-iQ;A*4CkG&0*4!;SH=<ndzcSZm8ubl(`9%0yV}SK
z0+DTp<A-8Q7^ronSm?%^I64@vw8n#_Peb{x@t2mwfCb^92Q{6|*$7X%!T!P$L+j}^
zL#=wd&?@hzv?GGg)YHh-R6jlqo?`k;_jh%{UoA?GCE7mQQ>Ndoloti#Ixv$y`^uxZ
zr<-{D!H?X=3B@UqbCwG|()q9ye+mQvC9~^dP$)iH9PEbOS##QYT(eMsiNaHhR?{1G
z-aF8ksM%y+JUo12|H2O#a+;~_^mzREl)Fbs<dG9lo@PMqM~~@;L-HPvW}U9XXPrLz
z{RY0rz7Jv*n(Tb1^L{GFn~8Bk!6R<J2j^B~^-dNgats@2(9=+6xny~vtPPU2_zB{~
zGr<DYA1Y~Ed|Z#xC$McU(hewxdkmqKQA*;Dkoy)d%5-hg4ZV+N8nwN=b>@kb8WusW
z^GFxrSBWglB=0YockPS?18oMEQn9ZN5}A{BgC*Vdol1)>(d!!8ez&@m!_|vt40f>Y
z(;IC|)cP__V0e>7CvX}`Y8cS@dO4yNI<M}B@E36{zb1|ywAMVp`Eh)~@NQeY)cW^@
z*J1$+i4`jX>3khA*TsqoW}R}Wc~5Ls4mLqeeGj{8ec6j;2d*q`2h&Pc@dGj)OfAn#
zV@Q~;0G4$VwzoqbpJ9qEYZ<QH4(T0PiZwp{GCWe05}XlyZ?v`8=-hn;)<GQFem#6+
z&*O?6%$dwi(!HEb%64$_2H8w}))*aQ5&I-2Vc%X|5$WqY)K+^fM|cx116vjvLLD2o
zpFbhG_rJpr|N5N)Mg)p@G%{lGYh^sdqiNkBAfL!+#7;S;)%m2HAi?iTw7*ZaOG}g-
zx-5RdFI9=6b*Sd-;cTUij<=Yj=q-7althtV_#A}}Z?0B^L16nS{WEljUIMhN{O4H*
zsIV|;<2@G|D%GBaZ~1LA6ezZLYWDh;=wI=2zwc;f+QNhzWzk6J82ehSqk;yi;~vB5
zEIK<IYH2^abu~;|or5gNc}#SE6wzIPbc58bSZ=fV9FQ~bf8rcsQs^rJk>UG$bXvZS
zGE(zQvA4E-7p8`BaQ$qo`4r4%Y!KJDtyql0q#y4dPJ`8N(kVjvR{58$<!>alHA?T2
zFzowSEOK}p+**$4ft>ec*4yX(&dI-^p*TCVm@4e_=%Gho)w)7RaObp5IhY4ZROty>
z=8gLeO=Ea~LtlsEkAQJ(jwdM8LK2RRlwPgh!1Gp@Fb{u;_xMM0@)XoxhB7Pn@EkAg
z8rNklO{=IkF(y@v=q>RSv{h52V4#vre#|E!_i3$G8-TZeXnxY(A^6#s*`}f2_!<3a
z*9)>(jTVwOdvj=Uy(1YLXg)TW5KG&l&eE4rqdx(Jyj0jRL&D;{-euedn#R>R!tv=J
zDBh_|!5VJ9XdD7^+B}+lUl|&?ohEzwn#q=+(vA(gx(Ij1w}J)|L<M`sK>`HQW$uAg
zDAM9PmnYpF8zzxFojzV)$7X>h6H@`YGV{o(!sr99qAZlSk^7uH>RT)JkEtCzeH_^^
zin%E})?s|_%L5y)#7-^`=gVKmOp<L^b=-D=4*qHuPa}SlDv7UurC5y^Q#R#;%7(*F
zkbu2agr&kgL^XhYBS-AxxW*kf%eia*cVik$5?5YdXrNgZyKMBs2u}}Or?qAJ<7rSf
zMk4qjK8RUDshM!p`bB9;Awy=YtrR7BlP~OLcpUr4VN945_!eDwOVO{Ku3$oP@#bu8
zt5_52PDQ%!kpxTF1}^il-D%&bzT=1F%d>9SYmCs%2c>if_)p>Q@<2#T=&?T@+@!u<
z5vP=YzDfMZER%A!S>qQ3C{F6dD;hd$!Y^!pGJma}Ys~>A@ZKlor$m=u_&^Nmm@dy&
z6q`<&;738a8ay_}gxlnjFZ~IHOC(0h3>&=jGUoOL3~q{*P;A$Sx?(7@thNw>0#o+R
z6A4jaLy2YO%Y}~)b{VchIOLWIaH5NfB2E|izJL<H&zhB$-tQDi`4AMTTi=2NQ@5q|
z<s7rFg9#MT>Kz?^bf=;)WORnOcSuXQzwjJpEi0yLpwT%M9)UF(Y-L9!{F2U809_Nx
zQ*IR#%d1z97v;y*bFaJ^EeIw!rn}-+fUX-vbezp6M!Px&gNA(U_$}DjV@989>8%bs
zIAV7*XCqr)0NL>zVS!ZFh#I<wRw0EvQli^@@%9{L#)_DxLZ!QI$*rG<bi4P-GIaY*
zG)lyYHDAC@Ik)$g_J6t9AF+9*es0fRB>D^74I}8q?<#rx{Z0ax$&DMG;Ht!q$Sd@~
zj7mb3dXM!iA?|lJaQwmF925#$1|0^tA*2Qdvzx+5fr!L9(0!S7Cx!8_SbICO%5{G_
zK!d50vxEmm%6**48Pi+z=9@yxeVGAo*8rwzcRGDo3~rPkfT+)k<oJFp5eTolzpuQc
zRM6;yi609>W=US*M+L?Xw`)&#{5+Zb`Jv46lPS#UKd{tW^wEFs&kp(4=0pxNxYjCT
ze4OvH{UU1Cw2KMy1u*BzqL=s~g)hPg7N1183v8efu#cdr$TV`x*Wbaq3~;QS-gC9B
z&6R*)E*X)#!Or_pF#ln&@NsS+BIycp-Y42wB>rjuBCLB#NQ8Yvz>7Nz0l}l_AeN}+
z$ej3gdYO1O&s!`B^T#ogzb&}g&JdO#{J_oOW4?1*v;i>01v`nI@5m@;Oq3vgfZh%k
zTFRqfv>KWoY+~*9{&PJ{zhy*@`5Zm+n1LTtW^qBGZ#}n_d?~>BC<?0D5Pv;V5*o6x
zu8dC2{(^o(?QVy+U<@@r_7rlZ=lH1O!@)NvR?kYz`n0{8-*@sSFZsmOMlP`)6Yg-b
zX*EF&&f4DO?6uIXsw8r$GS0c=?LsmRJI@%w(}>o0cP*$`ACn6IR9`pm*{pN`nSxpC
z9hp~luSz#z8`i)`RJloISu?zEexV$ElRWgIyoE{BJ1)OUreD-n;~glTYob<fW(Ise
z8X+^VB$Z`)tGM*Ui?}%4A$UW%Mh+%KidTr9W?Wn&w%aqJ{=LBd7S|WYCG?yM5Bj|^
zr0rt8|D*{yhtnUh=$23)A3}I)(5b2CZwRS>fJ@2>DuYYYTT{gC%OkArF=f_fA;QLu
zhNLJQ8p_v0jo6viyjQNUUkSLDB6;|HVBb|Fe%g6$T1kdNZV=-DD*=lX5^L2X;>ymr
zgg6)7Voi!pwH51MUyo}&7T-4~5uZE%JtQInj4ybrpKWfORDv=jDngCX1ICpwd78cS
zjIYMT_yLpOHinWf<866Xw8w4KE_tIw!p}6g)qmY3bvb+w%Wiz(pN7>i)hzMrX-(t@
zK_e#q%k{*dqCwUSzVPVy!h<Z{p%uxcw1E1b&q6)?8^$jWy~y(2*#-Z<r>uBAEZ$uH
zWc_gK_!z`iPd&*i{dgTuU&sEE^M0pHz`{Qrbe`zJbnheTT9jw`eqn47+yUlbZinAr
zG2LswRM612j3i)jG^9EovnJv+9!Ng4gKn`cICH&^+GrQ?bUr7GWv48aiz@bY-s2%%
zJk9k={8YLC^de;Eq$r+xaO_tDmN|QyEJpLDSmc9<!Zhq*mtyq8gj=68QixtSH<ql&
z$?feN<jB^lyaP<@V<Xc7q0zq8O!>iX(>bq!CW9M%sNlwr8OGg6(||<Ek(+&j!)Q^0
zllQ^rO0C$bjEUtuMGH-}^#GGV8!pnS=0|nYeM@fcx#I*3Hw3UA*<%aQ4av~J^o+)B
z`?+e5=PKCzbd6rzQND+knCy2)R9miE-p7n@T>TOTayNFCEhSa+eqzOi(nLD5zdd-P
zKdNvsytn?~<)61<YOOwVHq{8VGOdlOV5eUL`|XlF>_EXxx!Un}h%u0PrL%AwHS7eX
zFQAcS(~)Vl15z>X<YjJ()-%7Gg1>Dbyb_TyOgA1C<?TNZtKMkrDXCJjW0@A!bqyZt
zk)rd{6djIS&o~O{mY1d8`a5+W(mvxxU9h|<we|Q+AHnAFTa{g)UzX@XzT8Tp1^GoO
zmRHl)O=OgE&Ux2EVD88o_-Aq|c~puOG?AiR#eIe3{;smu?azE$SsAJ5v89_3PX$M`
zrL>hFdSz&z&4KMuS|`a%K=N+-dfA!Aa6O=S5lp1d#LuMUkTuE_-IT8jbJx8oru?d&
zy~ZyMTd&D!B%JB1$ol&;t+<kWShwFZB`cr3LW|xKZ>cr;o96ZK>>o6*SLsnEoqjx-
zHZ68_C+m%IFa+S$%XadYrkntOmQYuY{U#Atc@uYtaxYHtZ+D<D9&e=<-teIk_ryx4
z6Uex0*GTNB)2yWjdvEd9xfQ>=R}JbuTYfWqWhCcDc4r)+kMY;{&2us=&rYDWEFG<u
zeEc7Ev&>RX%LpJ`u%Tl~@&OD2B%mRlC1>YJOi7DdM!ozP($9_-;1C=eUE~$6{uHrV
z%s;!50d%GE)l64jIEVR1O^K{9=hOXbaUYynOF}e`*!lp<eWn$ANAxyVa@b}-Q=Y1l
zR3E#~9oLMqlgJFhCkb_Ji&B+X)gEifi}k*|GimkEqkhQnOG~BbRu_-9)f&J?wvEPl
z*Vd;lC_3WJXcyACt&Fj7bV=02JqESc$+EeaJu2)5QNe~7){om_C)-ma3$jaLL(c)w
zFS|`0^|xE*?{5xZLi4d`a&h$BA1ly{N%|B`B0~K_OUf@F>u7BE$GDF=rUGO7U4n5B
zg;r9@Q-Ni)rkmIkWpMIdy#_MKtqBYXd~nOES@)*owTQ3+_kC<C+IgQq9LdWDuL{qM
z1Lf<Bv>Y7zPZ3K;?6I068vv#p41D=t{7>c=-VC1OkRW)pP@$88DR;as7z9ymZZrt9
zOR%6TiT)e$tGJO>Z?fR7&~Tlb%enD$w)g7rQ90=%s8eYoc8ymi3&6O&Js%1zyotG9
zb^}b|)-TRW`48Z3Jch~0ZVOMj<@TB5rjr)Y!^dBr%if#w0I06vgdmdSJZ;+P3Mo9n
z=J=)~g68w6%QF?G9|(!Qd$u6<#(Jq%ub>UlU|wv&r;DbPZ`nOF!^5!;NGT;Z`ooE)
zKd4=3f`3rE8t~{co6y>L=P?fvficmjH9R<^6vSH(K_Fft>>GU%0wwK&O4+)ub(sgE
z(LHkp$|?6eS)=kKV9QL_5(&VMn|R9hb_G<BHs#?(F@Oc&SfAFXwAJA72TRNn!Qf&l
z!ADu61k&l#=mdlx3&9QxZx$`wGoI4j9VDA<^9@7}eD?=(*0Xb0ua)1;ZXgtYJy@_w
z0<)F|=QHUn(wbVYXmxOz^BUR%?xq6Yr?@AzY`zVW^h|cK?XC*HJp?8gdS}4OZ3sp%
zH@E(zRdFl#%=vY_ttD#VV992G*Hl+>-`tjqXO6yI+%an!m&EXB*=6Y7qv$#9ZR<$g
z>e8j}vSOC;blU@NDQA**zOjg+JQf$;c+*=){;7k29UKRoaF<HzHTa#Bir?@TcYBQU
zT>fN;tuag=$~(Qn8g6p?=MD0Z9yUm~FlLuOCyTx4SnqXlMGd7+EqJVJ%>GFa!!pMV
z7rVJXvQ`L=XHUF~5)m&OC&eaNo4~^<OUoE_jp8L0JnA2`QyY6;*SN7>XJC6bK^dhh
z7>JD`B%l|KZhMPTrMpPqENS=Q+>eKNDNhh7Gp(NWr<Y+)J)?jyD|VqsIenMBhUq6;
z{mZ9L*{SGBx`UxoUO|k8&IgG!zwnhDqG5?-3?sb)Ubn-?CfG0c+b5wvASz4a$T^eZ
zPD>wjD|`108o8s1b)c6R5*M0Aj@7bggT6q=>bH(#0$yCibgq3I(TEn78~u`m4406U
z|M}u2+LsMG>zlD`G5z>htZqMx{1lJSpZkklb6ht}U10~q7FwDKayNM^cX2KIgSoh*
z3b7G%^gOV_i}0&&`CJ7ng$TpL;+!<og>zk&W%ZfPF=H^3P`nUg5{fy`wn{+y^Uyu|
zT}zEmfZ}J&ElPVL`m46*);ln~o|}i*@T%>DYqYhfR862Wt)$-`ucP>9$JwxacE_*Y
zMQpeCweHhAbmJR={Zl05e%TvW3=dSe{|d$tDj|TVCepu?QN#o3Br|9A*S9Hdquw6A
zbr~%&n^tXb&t2kYWzELxszoBreMnLaejJ1!_BqtMJHG4_41Uz7mNQj2;+}P&KMTL}
z`?MRzgbEZEwl4fLCugZCH}r{iwkcMG=E;O{d?<&m=T)O@O;*sPE-C@O<D5VxmObrV
z{LyJhPB+>KE{e(0dO|i#hOSp$`rC8(_cwFI(7Jms?P$IKYb38Mg+3sakmJ~*lSgN2
z24@J-35!h|&%G|x&f%p)FKsmzvTb=Wg%D>BA`->GU_WoyX!SzFO)xkNcK$-RbD!xK
zdyID=y9SQgj2RBL!U=0r(AwY@;I2)@NVYg<=s%IJRnm4*YiHR1F8ipj*Liw|{HQV&
zXXz_UoI2Wv@mDiwnt2fV*gD<c(Z?`UGrUj#qK}FEPx=^ohJKec24)7GhKvX)1tzpF
zILXp2<i*1;`|v|EopFQn#S(MrMXe||`_{lrLQBosti@23^;uD|=7zU0`2J*2KeJ9A
z=u98ISyq>2g6qbzD5yN&on(GO!)5II{(*hA@zLqs#X+td9g$nXIPxn3uD4}qGEzvd
z($qW@W8C>+>R^j`sHC&iG>lpItvZ;-EjCA4NzK#P1oaUg=h$`uCcx21HwOw{Hl%3*
zd~VcLokODp<{0O!-zH=4azQ{!DF+FABCg?sH#vjzDb5?`=F@P@Wn==$yHA@u^Z>+L
zM#M1K`8^>CJ$CC&4BrfWQ6H`km7Fa;_lw`$(Z2CqE#$7zm12$i77G?cJkKVb7N}Ib
zp4^~bzThww<s8D}vbW@1uz+81*~~qMOf^B!{$+lE7FiQ^d9ZmOkTJ<V;_d#qIl3Z6
zo=T6sMAG_kn>~BY$JZ5iRqPCvnI1el%lxy#=ZUV!9!{$yMpfS@?Y8V~hWtH>JSpfh
zOBXm&szeXWlY(OrvRTcU>DsfquhEosGuIVV3abQD=d|8ze2ih6c9U)iB=7O5+-#{+
zZ60h9{x5qYEwVR;GPwLE6!9q{M6*vXjA!L&@zGFA7L^u#MPq1v1Vq?b5^2TD;5ojt
z%L+Co6%WMUzn<vp|3<h3&i8@!gGAY1R)DHVgzJ-3F;yt3*yWVO*kJH6#PPq;-R_6N
zY@CnZ(3<^MBR9Z+)E#zeqp$q2#MRCenihFDe)LuL3qqH_)En=RZFIZ)R?eRPYok6O
zqs5M0(l?MmgO-p4x7g~bFLNtQ>$-(;7G|dly}K1^NH9EtEF8Um_kVwj;iWx8fFW6*
z4+X!lSMhuT@qIDJ3`v20_Pg)e0WN-USNuXF$X$nZo{s1f-emObeI}a4h(pJ?xBdOo
zTo+GIG<4Z&2>$}oiY;e$geB1I&|h~KpjL&Ja|x%rm~SUv7=Apje)&(f+E_r+HK)z5
zB?0I0Ckx}<@(Sbev4q^)=>oK{CKUNcn_o)?&PqxhkbXL=|C<l<pC}D6$nlnab4&EE
zA0S6Z@fSfEf<P@tH}RjR;-6q`OoGEt*4=6c^5Jz)^!qV?@E!WI<RrF?eij{o(7PR(
zo@NYBF+ROXtAtkeMp&X)>kE-*(Y9v&VT>yX+lMr3lnldKYyoZHhPL9J`Fp&7X43Ie
z0Y8iHMe#kQU$l|5MJ$CRenoip;|2@(hP0&Mh)j&>IZp`LPgo(?9NMC9EkE6W2^SBN
z@0$k|ifEE9Axq$?mDjX-SM=J@MSmXKSAo|bhVi1sBG?E$UK!&{>^AwwC4QPPM;rc*
z!R6lK>A$Yf|J4h{AXuM@mSv-w|K*MT8s_|u7ysU%{U)LOpZ*QK(Xcp&qi-a+{qw=|
zKi>5BX$0T=|KG>Ac=Q)ZZusk|qNYEuB;Tzx{S+hM%O&zL@+be(^S+2|I8XS~mhpaV
zIK1kAu<Bd^`PTkoJNQ~%L!Zw)>r1%bIkJx9gRN}v+v?BPt!OTgT^k2p4h}kzrTTm`
z)O6;?il1%#NHij#Pga|6tvQDsogiL2?(kRy<93%p*Q~RWottcGh#Tp`(RnsZ%^KA-
zv1zrC6Re8Vqkq2rI}*T9I=mpe`+MV}h&d%ZJeBDPs(Q+NDn!_k3=_>w@Tk$Ethx5$
zU$}6zr~WUwaF<`OJW&WSd(w!_+zu_pgN%Bxt+@fTRb^{+;2Rrb_m-O!M<)vcyS55!
ztI)h?&<C^(xWbo2_=b4$F6u#ZCz$|p2KBp(LNawdGmo}2sTcG7_qv>?+C~3$nDuY3
z2k?<?-u%ia(XZ#qf)@>ev>@T}mOnntgoTY(iD?iF`UOjXTVd4%!WUB@18%~3e-r$O
z+4|XCBmDnwIdRVKSi-oj#<o@3&6>`4CGb~FrEh?#J4-pMzATzHy2ZHm%{!e)DCAt_
z?6WAKO~^txmdu+yu+ERwuXWCL2*017Q=DQnH|1rQ%xxC@|Hg+#Rdy%vw4pnM4F$G`
zAucOKv>JETD;zhBzk>=1oHFGg@DpXJViy4Af65_*vj}gw$A{By0k$0p*}5w0d-c<-
zZSkb{u~E2m=&TqXevu{zdhlVYzC}BHJoiKNWLs4mrn>TP@>&ZEwt4dODi^b9=4Mn^
zo!@MTlzw)*ck)lmF(L$>m#S~;7k>k-;`-=kaVZ$>7e3l-!ie$IVt@86b?hIT6z`Sk
zAQ6%Cam7WGM5h4%aF%_mc!FERw>``63Y5AQrcJH=jAB)1+52gA|8uSLciGuHS0Hk9
zKSkTs*74Hs-~JnZ_}AqUeH)&Ke)5vDzm`f4%V}SuV^9#Ik3+hi1bD|WL1cR^l}SCM
zr^JSH*InGyEw{Ebo259K4HpaknXb>M=X02J&6e8bZY32oMO?)+k_9c7%CL82qHM2i
zp47Su6}5;wq;j{w?D(1#d?K)RG`F@kK0i+O)Z^ie1E5)SHcm}|`P3sASluI$5qqBz
zp{5|_>e9IPA<jT;ePK!RXT*CtvXFI=kWr;^rrrlXJBQ2r9!#$Bwhat_*LEKFS4^|n
zH%iX9d}Tl6O;gXow<mojpN9?BdJd{co*Ec1)$Bak-<?SYOe9OxcZtP}X6I;Q`(l$#
z&b)h;&3H@6I=`|~IWFZXgAw!#RT)XYy#Oj5lh&<BtV`3QU)qq-?@PIGF^wGV9&&B+
zKiiG7N(oq7t1_Z);o4f+yyE6^V)wZ`vZn}$bK|Z_Ax7k@O4*0g;~_34U3Q@~gsm-N
z95uB8Tlo<`qIPwBoJGdhFP<^!8J!e*K=~lN!pwmEGrq^i@sxVza;ad~p6$Kq*niw+
z!(iF$@MkRJ|8Zx<zYy&M{A{-Ut#9`<G=SG)6|a3!0||`@^=>%;Yvce`Y_ERN1-%P<
z0vN6P-9s>$Imma9h9_P)qz7bZR}l#=m_&fi<tr3ru?VZH1bU)QL|K$>DCDplzq9cZ
zX$X)gy<LL+;x_6iz~DZB!Pj$jF>d%Vfl)9PEB%;GvsVAJZVoX6;07aeWdBk~;Fdvh
z1RF&7M-0y(!<4VUu_K9YXj^r?!$q^OLFa;SbYVX70mA)k>VcPA0>RqtpyZ^N3c!nj
z>d?Hn61?8?^AYE*&rtXG!^m_L#N?%kqaaZw=Q5U_a1+O9*`5Wd@mo9mYlh(Z{no*6
zYi750Y7GcQ)uZnL%(?1uG(KqxSB8mx1mt|CfCoiJ-M%h|#PnSuwR=Ju8TV3=*lOIu
zHqiv9xDg~WOvr`kE-=)MZ(Bbbbl?KBl#vY>N}2GJ&dp(XAflen4&yf5VA|ybGJVD7
zaS$_9yF>IqkEJFScBmcFSa$QrvfpI|w$eLmkVSec2Vq`(D4u5%Fpg$)b_lD;%gwOh
z_$#v*m#)|!Lf^}WP5_PhbKJ6_xd_oVPZIcT5|OPN5FNe)5NMhrpozQT3$I?pcTI3@
zRO{sCr}A4yDz@cePmrgQe;jhc+Y}vfUE1u=^>F5?17;OW^2u3BZk!2lRKL{kT12-+
z4@@Ok9E33ut1_2Xhn2Cx!(vdpz7p+eSna>X@pgwwn0D&=&VcGw;o61S3XkJi8nYkA
z2S0jKKHzrm$%{7io031?^bm?!#GIfjs$+QZun`;4@s1=0%mPcj9RzTuyY4{Uz+R?W
zV?;H*aGFopRS>V@9i|x2T&^Z4Kpg7Lz`1)9=5Uw+{P-K9I6_kGRyD7T6@NmDo&_4e
zM{r3yq4lvc`{dYD7QFxGdjGp^47RiZI^@)Qs62Y<kFCt>km*!ICv0Tw2)hocvv-Th
zyyHMFDS?<@X<KH;EYGfib)%v(dip6}Nz|S=Z7J}e@13F$b**%J*dlA<YAxgB8t*57
zq%xKORx>W@D>uD8<%8b=WI>}wslW$y2lJFf3x1vZCkDqeWRJ!yUmZw|OB$Z`ngm`Q
zJvb+)Y%dM>6r^$*dFMgWI+Ko`%gS3tHS`Ud=Eu-kXJC`{4f)?mhetr-2zbh5zF|_{
zHy=S-DHkb9uz?aGrJ2x7(utpch0Lra>m^Om2#BS<EnIlLL_~X~7-N7$LzDA#q{xVA
zYJ1N!;`}ja1#wbiDtEl=WvQK-n=_%TiCX<qouoxf&w!<jgq6l)ZGsk;kL#YX8Ai&@
zai-w6HoQXlTo^oo%7&9gPe$<2vx-iS3?N$Btqb&TVcgl{`&h*qb&TNo&0g2OJdO&(
zkgi=n^xZgwYoV|ma_~)r!Eed)u4LXJ<fm%8L)gKX&S@^<=JI&mBJF{va2;C5@(a)H
z+5;Rgp6FB62X1Qz2*(`e$wluT(hQuDFcBn_4n|3PT}J;<ISk%lW=j;S*j@`jNwuZd
z4wQ?$W3ml=eRe+?kx_o@)BH6ME-|{l_pPsK9#eRXU9VCZXqT4v)&s<j8-{in@gpq9
zrRjkCOl0gN{E3b@AP`9!E?o#qT>$UXa;M;frynlGWnJ$f7(0MJxe7w>+N8C#C~13x
zi8iOTpY37i-E|xu0g|g?d_3=Pguw@MEuAZ~hpM^mNZG`+-|V6S&SEQHaMqq`O-{*>
z)4vDXkQ)<2Vjp2=6Y2VBBvfQ1e5PODO5W)jn(T<?&mSi%_;TyidoXw(Il3qvhO+z#
zG_+_@x5M~|9YhIuY=k~hwO*&T=yLC~0C)I5?*IOt?u`E;p_D*7^qR=6Y|GUYdqu65
zhj~~$uo+Q^hW3i){#65!9NN}&#3NNhX~!f_wJ$<-ja7SL+b2)eC{lIJKzpH)eDO<X
z?%w-{8WKmWS*5$V7n*Xlx^v@-LYH=YiiEa%rCtO(0#y|IFh+&t(K<xhPAsa?8Gltc
zrW?ss?#b=#$@RRPi4u5i*bKvq`?p18%-;5pY0&A>n;X3BwMjn-d}DwSeY^H&^NWs-
zPx#A2m?Y1ylLS=3F3OX6UMnYnGhl1ch<oS(H#^HFEpF1yrt(`<swBGCNffS=a0!q^
z-|JH@&Ku>Q&kMFG{m5VftErsJ5{wC(_I;zAZ<g?hW8B8u3uYq0xK1K^z3VE8kLl<t
zWJ=q<qSt*jLGi+Q`mw;XTk%mjQW#)G;e>EhgGpcDM9;7*T~(@OyiUSyKKM#s?nSU{
zXjHIF$;YG6``=_K%ENv{cVrCNP@XqDXH`)3vYTp0EhI0L_vxzMD{8rwstb2JT^k;a
z_&S^M=r$jBU7IEkSSSl#e&OuYS$V$4aWWh%;rM1R;`PTbu-<vPXv3TNfFxY-q{%zf
zyx@m~rxl>rM}vj#F)IAu)B1Zj%QAsw*NeZCcE3E0I_!+@`%+DjDjm@j4A#1em#=%r
zCT`74H1um(^*dhMR8Z%&T+y`45aC$I+ko=<@YAlVs%`UazpoN0lb3fyJ`Fcr{il*|
z;M{4)cEkUrz?DEl9ZN)H>l<?G6sGqol)a>G@OBNvpk?eu)kx^~<B77)E<`z>wFlbg
zGqKPwv4?a+9?`(s&NNw)WkSc@AHg+>au!s$vY<pv&+XZcolyyK6Xw_%u5KguS=5O{
zasysGbSN@-?a7qGQl@%SOft@FnH2bEmA9JM%m;`>E5U|R!P5zxF#SqZ$2kHNiA~0{
zLocsBU2nilDx!<S!Y#+0KI7|UhwUvE_}$RMPOdh7^mvGKLi^PqW!wu@{BUFuqcEf%
zW6|BvyR+u&@Z(zjL6{E~3%jL-;nNYLfR-1mw^C@7>h*J9NiMzdr&6pj;M>kjF29Qv
zqElgWzQo_@>Ec|U<+q+@u7EUrV-~EjfhH}Dsz)Qz#qgG6KfyN8^eqATqRrs4u2i}C
z3nh~jD*<aF`~WtZYY&o}wy5@Yu|2Rn_povIR!bi}U^_W4WpYe|Mcl-l%M-H<OdyEW
z=FSwPi2O1l+ww+haxldHhW5qicSBh_557{)V|X~$e*?Mk$Gxi_VUw|kDc_g=O96-o
z{C2n_mofh=0HYXAHKh7GbuL!;PdrhPppd4O^A=0?3Ao@o;>W2hHItC`0FeDy%tFKU
z%{aE0LpoBLY&r{XP6kcyYs`U>Q#OU^YVB=KxCYj)yw=82gNX~$j^KdJmMBlqE4{SY
z=C|EsK=cLfS?8s??VUAf;2r>3e={KW!Kb1iUhC>;E+oWIG3~VxC#7j7o%jp?>1S6l
z8zuoFr*Oebpjhl%CQBj9#NHyCNTYr1{i;AuEHERom56$qH?(lj?;xPgsN1uGV*s0g
zwxikUUT}GCv`LrHXQI9Sh-SQrj|hR<#8cGxMab;bnl|nU>=1paceDrI*)$DJomOn>
znJ=>%VfHmB3PTKkCWyUl1g$f)Kn~T#bScmLTcQ7UH)~|8Fb&VW(0t7aHug#jE%jM`
z<L1Drl>MgD!U23)))8@5uaL(O!@J&Cn`tLI@?_n+9ylUbG6nZO+lPl?)=&q00%OQj
zFhrR9!kYh-VN!!*-$&r$k|SO2z+RW+YD>vKb$ZI@&=b?2A8z>kFTPAG*7=J9%1C7h
z6N`3>DramW+!JQRAUYzvt=1Wj@I;w|VuFW(=1E$fb11<MX`t9ytdCavIw}MyIlbZ#
z_!b2N?mt*sy+IlJo*Az;udf12D@K*W0ykE3BmG;Os|ts{VCQ0h+0vU}GHI!u`716^
z;0OETkdg=o`?GdwYe-!U4xTmzj(7&07Ei~0ENz6HW%14Qf3UL#h_Vomij0Zz$YAsX
ziV1B<8DIuJUcf~tR3+R$Kf|&43YJ5}Bol$b9W%kiFEn2(zJK<qXys1pV=#xx$k0<M
zetz*$JDp7F2-MWlGF6)(z-S&yZxrf_r~%KK1IXr0P<aVywi)#LeVjq$2X>f?mzC9S
z(#P}IZJpvTq&|nC@Lp5o%?D^)m?`EpIrUj3V}YDMu0y2hk?X|D8Vy?BARJri+Rz{z
zDj7dnjjM;@4xUaI1J-LPt7)>w_zxq`!n3{tF(~a&XmNl!9GB4faJ$C_G9DSzMRUXT
zKi-B?i(Gde8`|Qv(Z#6ApnEAJbr|gL-Sf^s@9caJm_Nu!b+r9Iq`h@iR%_e#iwF`5
z+>{7Nr+`utO1FpzC@Bqsgp^2^bVw=)Dvb(AOQ+IG2q@j%CEf0GuC-jxdcSYIK6~sv
z#`?p-f;;BCu4|rgoWCQXM3~g(3nrNk?V>qv+AWIBi)%#APMxUS4X>ZOUa53xGs(XA
zMBDG#DfqRn-Y?5|z+bz00?+)SZN4-@9IP16fE03|q0I-VexT7z+lmm7#`O9C>kLD&
zt`Nhc{(25IyRUmc#$dYSGA)77+gv{LYRY}tOsAVFr>b&Bz()L8$5%^jh!OMz4C#JX
zH3hB$)Mh&k3Nrs%mMG>(&~-{CgIQjCgFAp0a+YD1nmPW7I6nQAZ8Q`5U?iDY_7TG-
zvd5w7bmHXk87H7AwSOxq`ZcSRocON9dcX3W2m$7RC+paf*|uyl;>5WuRiM(W>+|0Q
z3o~;;ePk-AzUTgmrKUbHkwuV6S7edaoWhqU$|)qTZ?u*ukl)1)J=k62EPT-MVzk<g
zEG&(rmZk|)l4_eRw(poO+4nP}e5m)eJ$R_Kz;vUui2D~%{25B>Z<nwjgDPfmN?n~p
z-YCwz*Rv|vsm4LY$P75@STY{dn?{;*@zTwFSqWQrAgS}S?$xhQWYLs4WtJqINA5gg
zbIyD-AgrEwLtgHpo%9RVUVX^+RPTX^Ojy6hy&T7pZdw#%qG6Y%y%4dc2@+LrAl|gn
zA|4^Gr}+#kAH-T}FFd#~C<*h5a&Nkenc*C;p&>bPAsYT&#a<j2W2^qmf8?5YM4beK
zO#WLk&Yz(2b<HnbO-Vi&1MK0mL;SX^U_jV+LcC`w%)d}dvohcg-NY`c1x1J3^VCqP
z=boTpbXY)lz5iRk&$lY<=#Fo^Pm7DJVf%LD{guxer(W2B*>|2#)f=jQZ5?o-+`x9b
zm#z559C@BfDlaWn+|7%WA8BK!q4Q9Sd#!msMZhkpOu2;ZoFnDKd<lu=DQmyT>b*v$
z&?SJxC4{WSyQ15(#fP!0-=SWVqfrxIc(#0*49jb?X%%ZwCB$m#phy!=;luoN7^mgU
zCrM~AByJ4YQk(Gj>X}hm-5*(F-fq85GT&XFC8eofML8~VdQ-8nY84(9rOKnuCrth{
z5ti<6lMKyHOCFqW<;f-ZoONHU$|{}kTWS9s>p^^PCE=oEof~GaonH#)Rqi)7DGi#x
zQO72a;(V&a{O5%RfMI&dnFLp5dkPA!f?d=`JP_Voa8A01)Qv>&5}u%)q4B0;?d*K3
z)krsfhf@!~E*&b)g4?q(YRs=xkY{3y1;^Ftg8J%bM1$|B!*xonM~t32pK6DhBuclM
zxHgb0s(Nwzh;8x8kNY2rqf+U@M!;n}=J60M&9<4dE<8Ott_6K~#JxinV?=B|=y=H(
z*G)yq<#zo(KWkaN&iBr$f;I&(0F)CZHi51f$|+un$mD%2XF4_2@p^+b-|(AuW4ux-
zx0>iSkkG}UU6o%Btmt9S91**{H@`O0e=II^v%pFx30dS;{qJ3O3i-68cF+~Ss>%=L
z7L9Y}xOhU*LU0><=}4z{eXB{ezLIhY#Gzxeajq5ZU#`1<fal)uYw$V<7tgd*ytgJh
zN)58Zn2p4Rp-=xUj)a8UgmHl2&xD&<2CeQQ=ibv0Xa14Lche_!c^X7mr$nCYZ@^rS
zH;y+V@8yUhXb~6pjyPFgKCOa@NaXter$lZqn$c8JcaIg1j>hvryj=i}ldYFd;N_}?
z+DZk;Wk(PQT6Hzm;m-c$2Dj1Yp&gq;W|(CiP=k5k;jdAW(TUfWLbYxYorQ;@uRBz^
zcz2k1Te1xMy!@BaE}OFpSc_YI{-vlRZ%aXf!!ExnENbUZWfJFbQ_l4J5w|?S^h>^w
z7a(m=<_LaX%oz>NkLNq-I0<au_DjDi(Orx~L~N$SaSAFg45nN}AvL;P|EK~I8R7wn
zIu}sLd0wSclCm3&(*?<&mi-oJ8y{0z6-E6)3WF$Sgkwn7qXv&6u}@;0FqVs&qUHOt
zKrU7O3L-c<aD``oqngScYC)s0+9RAXn!~bPXg^F}-Lj+qzxJFjCWf?T)|IkxpZ+<5
zmQSKwaxe!fx(m}tbnFZ<^~|`WN44Q}cfO;t$#<x%SzA8U(<FLNTzkR;J2G2LGrcPu
zN<QQQ-D1#ucs;cKDv)1%Jt4ZYWIgX0gh4pFSAdoJ5tsBz0Qugrds5@5&~jRY=&P=a
z&H_z)X4u=IkiGw@0M&9-S-khW7iOhDKER$Rf%_4P-*_?lAd&noQ{nFfugUL_$&T;!
z%}sxb<*v%0uiq)Mosc9h{ZvU&bycEs9MGXqDti*_woXTiqwUTZaO12IYZ6ctR)5Lh
z&JJO5<4NWqrmu*9N-h+{X4TG0Qo28Bk~2kgaN2$x{_{qA>l3R`*5sFXO;&=m1Xe4d
zELREh!U3FTcYdf)(BaT!_{O2p-G~r9F#E&8q#j?Wt<I1BCvXu6a&gwg-JhVG{|_k~
z!a(z1qO&y{tEu*1X~Z%g!}4R5aiiUla(gB$lR_Pog&HKG?m=X5RbT%pgOlybj+Pbs
zBsQ$SoTjwiNz(ok96f#bKW$Q9a3&C5zlk;b@o0wgawvmz$lhhG4-E|@2EX8E-g45R
z{y*t%8AxTSB9GLruXmbALcuT0XmcK(nkD&N<Jgs-oI4VP`X1s6_h~(#+Jy(g88h13
zb#W{C_4v8q9aBO>%}dhYj8GDq&L@uPfqOQG;RNDS6T+T5-wOe#6BLopI0o|@40i<}
z6PJTQE)=n}Xg+bImS<ZIMUz1=2x8i$w%2qA`40y^l(F4xsRVt!)d^xbV+|I1j$TEl
z5*ciIBXbeCoC7*P|D)*}g(lc{(=rPVy?3H5hRh4ZwJk<UehB+0O1htmB9qP<<u8zl
z+kqzkmxlMOr&TPjBC8lwpSO)y{0$tSvRLGT?`<SF;o?_}=JABArtE<ogkK#5ZT4*p
z!kg}OIzDq^({s2nLvgTx`7Xbx8Jox<%SwsZe`A!?h~g9j2y_#LBgy$p*FQjT<U!nB
z=YiW-;RMky`I1+rCEntd?UDIE>Zkt42fH7@u7UgW<{A?!XZGgXng_G*U;kFX4Fnxn
z&rj2u`2SkZ=I9;5o1r*AXqz^~{0277TQ8RuxjWCk1}h5_DUwq^s9U5Wr^K?M_r8fw
z{eYJKNd`e(AlU6Pv0~-jM|JESe)2shDl12f5eV)TrH@cotO^>>Gh0Zw=sYFpLPl_L
z98&ux`@N!@M1|Y%sIGTQl=Z(;q};q!0z=}1SnGlXbh+iz&lG(44@#J=YA3)or`ALZ
zXCLObK2)GWNbAmE8!40J3LUN2E7Ha|7_u4zGOdZ|mWH+!UmqfmvqR)@sW^W6p@=*#
z73FMvYtikh*QY0^BzRG0dWNuDBvh|kbhA<lh5l0@XVjP^;h&NK$mV_rl>g)b#G02M
z^Lv*5xnL_`!S-<icpvtoXA_<y8aMsJ)(*aDD!T?kW?31Rq)0NVLWJF)#6NKf!?laY
zPLGVPYP$r_kbU<0;1idKCGI9DqH5%uVhHSy-3A-66<6#fW$ICE!ku!+AWY{z*T>8B
z&pu*o^KjnxTQLl)tp+mP4t7NPy1@lgLbBYdg2Pk%ZwDRkadf_L&9@4=U7$YkGVlk~
zg42c&^_^h2asRnQHyG=62@s?{mScm81pS_G5Z}Yn{=jk3`0*4z9p(isokihhk>gi9
zCN0sn#21CLQ}EIgT`goi_ma6GppkpUQGZ<LNe7nNiH&<{Pu4Y!b9NSBm>2IW9<^})
zgUY8@Dp)@U1{x;a)U|XMd0RyZwx`w-Lv_a?t2B>0B~Ba*hVG2<8nc@)|F)LH<!x$K
zdGc2DA3+bH^4gVDn&_Oy#W1-o-}f2)wc$Oj5g8f`T@jh7lI6)7chhf^;&m{yk)TMW
zF!ouq^dh81o(8)v5|J_ldm6m>3#hipeM|KS05+_Qu-7eZt(>v)NZ5Lb@g!Il2bu=c
zQfX_Is-Fi3oW>*T=x;t&)nuW-y~?0pjT;=o_Xkexf9m1xddnd+H8HlQu6<FldahsM
zGA^_Z=1vL@KuT9HL%iE__$*^ws(UCr!--lY!oNuZ(>|5{l8TMxFFhQ0=>MUI>vl^a
z^;OKIZErdO+5a=JOt{x={2~l$QCBI0dt)DG)4h}t!D<1xNA}Of;M)eJ<##gqy(_+d
zu<BM@j+?xf^hv`%$W$ZlBil(dEJi)maYG%CYai8=98*(uz08K7+ojVjbw83UG3hgg
z2I(r!GXE>CCQYnvhU74RTrL@Zdjy9bj4LHALMY3L6W;SYu?P^Ny)i(;seeehQtx3f
z#&+}^23a*#L&<dRhEs8=(mfj}!?kF~FR01L@16Y^AdBC7ic~NJx0>gb%AbXe%sX1n
zT-ydi^Q|CczeMruUiay@8K2d!J1pMs;O!Tf)f$ff<S>(uv;F20LZ--gpy8b`dvmR7
zGdv@iEe(4x+ozXSsk%jL?1=YDUtw9aOE~3$6Sgo*eTV)wvxTMHV?11nOrfy-b5cUz
zOvzU%6Ac#|HM!1fo&{00=u>W*9$q`;YFsWW4_Vng9p{VdZOzA#-3-~)m4;$D#b+Ey
z53T)nX}7C_*&Z=hxy`zsnV7sY%4<Iw&+<EjP4XG!Z1nYQV*X1QFJ4V}<A7Dy=J>wX
z28;OX&q<^Tz2{F2I$DKkaOXht0%3k!eN%yB*F$OID>-Mh+GyD#hej{3onbD}nLm>{
zinn?GC?-#Bzpt&B*-M>6{&!H3Vp-^M-=pL%iT$gpz8HxncC&9H`30~lq=V61TzZI-
ztu$6T-}L6g^G7}$K=xhAJE;>N+IimvQ1RQm${g_{jY;?f-J%&#qZb^0avQhIt2P~t
z+5EAtvC}kTcRH8PMcjz+w;8z#l9A7UDj4`HBUdKCK%EfK4Wg_GQ-%@v$I|(_XOfJy
z-CNk5%&$h{1Y)Sam%MJd@l$|f6&fXRT@z=ol!-(Bk{&8dttx=%-R@Dt#C)hMsKl>v
z3qO{vl@%Vp_dRIc#qzfw1O<&UHjG!Le-F-oW_+UT0(s~2w+gW>puQpV6PCUd^b^4H
zbl|4fS`{Scp!kGerLTVIjgZ`s^|$(tug}FBv2AlQFIbEQ{jOC3>KY&yg(M05A{V_=
zGyw1jFb=CYur0F-0?tugzsjCOzO4kW|I>I_)$PWR^?Qa6&*$PMIPm%$Hlty`eSKf#
z^(7K^e@_VlwbsA?=Z%gmOj@q<|AnIUzy5s_qVqIXo)h`q^8Z(D<exuG7Kc?qiXGbc
z|Ml;=5fR&kZWpEh<s1HQpZwR)StH>JVxk+{{eS#>vvcqT_|7<tKL6c)`b{E$%LrG{
z`;+EBlw}sdPwa2616cm$u|?*JdFggDjaLzyhNhu^QI@HPVjSoeaR@xgEA==X)i{q?
zr<4mV{>!IljYGaA!xZM<McSSlXsJFRRS?ZG&$(CJ(vmeBT69<d>Bmcm&mOOh>Q(@N
zzJGh(1LQ6P^XRpRr(r|LgsgzgsLnb&^9T$@sgHrNV@P>?UpEf0&i2pnw49$Dgz`c3
zT7cRed*HUUv(4;(*IXh^Oq*8{>@gCnTbcWBT<C5}q>F7OJwWr<H!{OOU#G?L7B7&@
zzWVA2L>r%>TX+Rf?!IbsB14F`GJ@T?Y8|CPg;IEOT6mH4`M1^P`K~k*use7J(>1Mt
zP=DS327w*MaT_lSH*J7eBNBwEj4$#Uf}y7H8A@$F+9|rAF)xdU1Xn_jP6$MK{D(or
z0q5@!wDuw-xBVs?FmwOl{A1GKcnPY^p_Br-?gH7*UVYaq?aFDomxmbBB4Upy#i?t9
zn#cM0-L+zGYwmWRLU;T<B$GA(ZVk6`oXoEyT={}G#9msa?XQ?%e2H_pjzob|>w`2P
z+%9TCovx<>hKOGZ*dzkqN#GkmyGhofbx0csQN0jYX<i4ViVj!v%c4U&WZ+`l2Jr^R
z^fK@-`3NHlJZ?J=fspTw=v(C!TD=xQE)oNaOSr5HEZ?Sq6BPr@B>K<n_p$1fuq7w_
zLo*kMl67NjmJw8QRu#mTB`Y8u({Y)JnW;34>5kv22(#wvb#;J@3Agv&o2!XwVLb#(
z;$c6Z!ys4~gTdzS_camh-dMDNUylTF^nKzdpYRg<X(_T^JO%=H99qFM5Q3HXd@=z|
z@iX=Qe!v(jc_7{)E@EwRk^4Xbk_UsY_mb;m#cz{Nw$dN9aW|7(-t)y?fDAnrsmtBt
zi+Ifix6sv>02tLZ;1|kPebX}k-5p@?l~eR)fm{slRU}_I4E+|j_^kXMaQPY8t;__}
zf6*Awa7ZK!Ao1pLS{-JFhKRr$kaerZjOTSuWIq7T8;53%ISPt!eOFm$`V??aPFgSF
zrRaDzv}(WCJ5bTPIg?clZ#!do+4A>qn}Ogu*yLaD{?)ehCc&m{psaglR0(c?H(*wO
z8cJHF#>=H<U3D8OO|ep*fPu()s6z}&7`Y!}-B0~2mA#;TWSn=l)EE{r^Ny6+SeYMx
zN{!jZifqltSR1U8GOXNsf)(+X?sp3$F~>N)yvF#)YRV3Md0UW~@AGGNLBkXUnF&`G
z>%pn>{%?#x;YD1dRCAQ-Bc@Mw5Qu?auIab-a>p~JjzjE4|3UQH1;4!Q<FF6w4qzLC
zb9A&_lU-^Vswj%H$@DO`Y_#;015104`{LL1b8+J*Cv$n3#h!bPhgM80Y!6R+US4+@
zKl@^IG=EI_JwJlDirqOvPcPe#dir*5mHhC-Nxx0Kl}uvkuFe$jst|HpY$@?rVwqJL
zXH8WHdsFU_2vB%z$GGtgmc|J3BSh(FJnA}jFCFPk&H2pD_DOY|nmfMMJ#%fphS|z4
z!RG15(b7w?cz#2^OLZLvUijMG-u8R@(?dGpPPkqK*1uoI7O!Cs7mIj46#26uNzAwo
z%9|EAg;vsbn<#BKAgs!S3PY3dvK0*D@jx{IQKAn6KY_<?Qu=Ir1zh1)NKh|9(Cc!(
z0ASAd5?vQBxq#*5G`QeAgIrPjWIN9b$Bz=C>ok9#70A{+qVe@Z^D{mkj;t-$AA;_J
z8E*P>hHyc89mamK)v~GkkaC|3KgK`!0X?6g2`Cb>oW{bBUihBgifnf{7z;bIpz`hg
zuw=mW(Q-jEZp-vxRLzAHbkQ0(avJJXzJgoS4x)+GepukY{Gs&<575cYAyhj56oKOo
z9zZ=3nPRvC9Q+tOf#FUk-hRjVP6=?MjB=u}HRL`n13H-1urKA(SA=E0f;-o^;#NC0
z_VdK4&>NC??US9-k6+-@+}g_TQP=)*TWE!mCE@Y2Y$%p$aZE1WV$V}HJl{ZseL2K#
z8)t{tak>9YGp%sPcroI}civAfzzPP2pPKH%1OgvEGW7rT0AYw4I|GAlsEh-4BLdhO
zFiQ)=n8+s(9Dkx&xnUJhhFOA?RQitzXYIgA;k+n?Jte;bT(17}p)FCJ0dU;uB7D#P
z6VjGV6E}q_FR|!h`?1s08Bx{*)AK-K=XU|rSvEO#0bUeAugbKcVk3pO!ynRKz!wY6
z9liep(iS^=RCja&#ed9A#DYX%g0FwP9P@={79=;(bQyJM!_Yr6Z5?GsIH`y{O6*(-
zlGudOh>2>-8B8JdU}U0R;^q8p?j^@dOKPtiNmmsBPFNCXG-e5^V+K$s-R(RAQd?)Q
zd3G2wYuL`6E#XWs2%5nevX6WIIr^xdOmBZ4&53P*cnNo}1{BGN@VUQ`Y7GOP1Xa~u
zOx4eiV5t@Fvgr$qo?X8ffDF$5$zR(v0R9>(J1LOb)yyDm#yx#~<@cTX8_7*L^IopJ
ze;=3KC~pwIUnku!LxvMGohVB|SSCbDU51Rl7F+gx-RXemTslok0`4;ad+PuOe;5>S
zOGuR{J8@CYCLR4udL30Ml-q<ZE;T<W1>XzJ6D7|L;`GVH>?LZju}=44>H~u(OJCQ_
z0*{MwUWg$<g`SveCCKtLwT5X{#9+iZXN2vs9n_VafOhZmonZQQqsogsbn;!$N(yD=
zmW9Fs-L+ks-8$_typS!toiVjqy{nf_&wseR@V%J|y1}<)LYvUD)ov{40U|5g68aoU
zOMxM}%ZY46ZANI!7fO{x_ZgQCbjba8PCI0x=%!BC*mz3HSEo8M0eQ0y>_Eh>BQ7lS
zbBB4qsZ+O#+)KsU64-~(ma0j7^iT~pkeIC(sme9d9~c}4a{Ge~KA2~HT;5IMryHpn
zpQ03;iN5<VF(dhNre;It=ibbD){<6Gh2<2~U-*<uAi8JZ1esDfQpe4V>sfY%Iapid
zwj1Zky~D*wz`c=^9r8UM_gn(*^Y{?;hpQ=g^RXd6h?ZEl8xRWHQV_0Q0xs=HlpMAB
zaX<WI1hKg66f45nWWA&&B{n&zE?)^9yTW0pUp$!d-^O2J_Z)WtSc4F$*Q*4?qT{x|
zsa<<<l|?c87T#uBN_6j0T@`|W3!_V9h$9Aab>V*S9C@P0-Uvi8x2uaxt}a1U-Dzd0
zw6x&UL8waxdR{1xpmIm?Z5bm+f;L`K6Mn(I5Uo-Lt-Nm^vdre#{|Q;se!zWu(Y&wM
zvtI_WK8d80ipR}PIQik-&Rz8}BTmF_<plJW3Ao2rBaLy!D*;xux%smb*SvsJckOS^
zS|QdF!>qt)tuj_Px5hfY-yC+h+8*Wqd(K+c|IJyGxSy_fd{Oikg9r>Pz6f;3?y16-
z7dd{JVF>YzmWVFc;oi-&UWLQ7<$AiF-K}Xf!kf7mbQdrP8<~Te*S)D}s()}F6?TZW
zX5bpJZhIO}zDFu&^BS@#pCwWxG4t!O+=Dh~tFT#YkLqbo#-aDdm!6pVVQ{YXWQykJ
zjp^l9l8VbkG<Mr$^k$%`#}=o%a+gz+T+PSdm4?TgwSm;m@wuHu2o?~!R!bW5{YGl}
z>9W<<!pM7!Ml(yLUfbJADhG|x-tpJXlFR3)^mcZDIW1-`e{3fCHry2K!6zS7B&aYL
zQ6m%X5NFPpxn_ru{(LW|OpKF;x5#wWS7k5x-T|RC@eaF{21?nN)bADam?Lnm*IW6l
zU*JDI^i<(bx|;jPn=kb<PIDkg9b)>d^n3bN`7knX%_Wy+Nseo%+q`*41mjGfL6~nc
z58HHIA(rtDBm!qlo;!wUS*ALW-jmC8B&`+;KWinG>G0-<9#Apjj8TvNWUA$323gS6
zoxq9r;j8h{F^xjJ)2NG<Wg#Hd9LRQ(_3l%;v(y%Dfr?yUg6_l_t>g4|?!&*3YP>f8
zKapyQrGFvSB9j2ACjTF#+VYeCfmHhl9j5EKuC2bGF%uou3v{HOGQZel!6jr9H^b~A
zXgH-(kdjBB4OF!T@BcwnW0&rnZX}vd574;%{+oH6H1Wqzg{p7w&>aU^-sR?~?%YB!
zXF4&~Fyg{rI@v(6F6i%q72#P<gi(&7(zR<OEol6Mmd1l=6Eji1aJC~!Y^)&zC7{N(
z?@(7NCuOLu4@fm5;`aqKW)I)4w45z;-Cm3-xE)wq`}%_4L%YmB5na-E8{0VZySUbv
zhrii=w#(oku)L13fO0Z7TMm5$ShZ!xjp;8Eq76FE?_?Th*7-pxUYe|vn_$%Y%FM&R
zSLMB~%v!{h<dCw(UkD$JBh94e&jP$k9QanNPwNF&$8j#WI)`>gcFkH{i#?ujt;#YC
z3nVkecHLB*aZdldJIW#VD~I~mSvchpgsr>!!*J=ZbeuK{O*E7N!|*EN5f9FhL5~2;
zB8a6Vi>j8m5Y*|4>IO=A+s1hC89EmGw^uQ|uq$H~9tIvtbv}uhU2o}DW|Pi%lJ7ZA
zeBOhxhoB>P-IHv)C}87Hc-ON*ouK3|PnjmAQC$agj^dgPH{VJg5ca(@(v)1sI%RdI
z?nokNO5sX5V;1%-J_a?4yw+GXmj**#p2MbW;!LF+l=er!7RZdAyG_H7rg7N+a+6DD
zi-mG9c^SbfPJ|~iXW;)B(b~7)B|QkbhOrhQp9y=-O8<|WggP&B2_W7rO~R}0z?0zy
zW!D^8mA28Uw_0wwzdeeiPW0zix|j*AzuDQPYjuwfAA>gPp4J~QwL0Oz1ll$?(agAF
z=6g0N4$&4GU58U;o&3@F5M1F2c|c?Z9@8*^q;jZ@;6%h(+RtILfvpBKYYNUR<h-W0
zwp5$AF}(7VzyZT_THvvAxKl{L>X*&Gix2<$k79{r@>iKOVsU>LODsY^fn&D`XtDtA
z8yflons>Ta+cJJ}y%uw1yX%w8g1SQ8yS+b9Oy2I+raiepE0{N7DI?>YN>e=pB5s_G
zG$G$JctqP*m=_gUT-oxb5SOfLcvyo5E~GReA^FlG!806gkZWq{_prapNKcA4Q5&aQ
zY?oJEXZ;yb%}daedhl~^P9gG@iF}k+f9fnzPW0vTmyNVu;!e^87u_pSekVh6`2y*A
z`Fkg%ul&gk31uP&jb0g+M8&9-C^KK~lsqfrwbgMXx>mn-ZnG`HWz{3mWlMB-w9h*b
z=DBA%^7i9-?O*O5yMr`Hzcf40fsd?y0K{EZKzZgf5fT0X>d9}yk3T3TiB`|*6k>&i
zre))iQO2EBYZLh#Ln^;1QdOv!aJ>geq~@>rEY)A>7fnf0Rsz7N&0k=)t|XQIa=YdI
zi!yiu_1`e*N{pNFH@P}w)B=Rb&I~|Eu?%Bh*_U{&hXw5=g32G{zdA}iP3999n9Dz4
z8HG0rf#bWA%JHu-DQ2+xg}49JS<IExr&o9gR;*T!KV&Svm@emP5x$>aj`mr&yVOPS
zrhSS+hW0T6RI!3|WY$Nl5<*Ocq2*OO5{HLQLp^gzUe@>0bqyYZ68*~Jv&%YVT>a5!
zuM5T7JgB7Lf2gT>DBNZq=t*Xx`cfZCU^pdWo=491DTXTB=7~>F+WZY>>A4WN-#7P_
zTk(S2!c+VY`%#Hr)JCJjmhZ|x^$gpm>q3L<eBTbt=L~>zpRT<}mjth+Vojp#Qig@b
z{*J@28O{6Nu8`eVd*{9%C-RuwREhm2Jg@+rk~YjXe$YHba%z(cJLkJ#Kl;v_TX@x|
z0sDKrPPMCpxyo*@IpC{b0!}@n=vG&-nNyB?I;%yX@NZmD=Uu>(Z@bD(HsUWEPj7K{
z(NKj@#O#W8_PN20Nn#up#>nSEt<|j<Q4L7Y-iql(#Mn!$X5LA}fKSfDh<(B+oe@hx
z$Tlx^1d1?1jEL>U?rXbe%s_Q-iV{okI@#xIbZTtgmu*t{{<2Mi!RH<8i*?%?M-Ua8
zSS*w+8G86ec8W+2W~iiHKW}Dtn_-I8qJsQ|8iu^l2m86(&!1$l%cEC7)-!Sj|DvX*
z&ZaH6Y|c(FwAY89;ZE@@SzYOXbm+WvB@s=@b@_Gnl<sFt3b%Z;2?QuHJXgutu~9vo
z!5$j9Is=*GN*+tOcF}eNBaUinaXnuhi-j<~pc8xx#q>0A`2+hGks?Rb8$=DV13@r`
zU1hH=PI=6n-l_UcQwY`Hl?*?(LY2_CEf*y4_q5%6o>$a${r#X?jR%#$VBmX)qRDDz
z(fXp83l)0Lp11&5f0p`=VyZ9`PE{ZQjWxB!2-pn2ca06SU&Z6XE`~9vk#8TcPi}VE
zqW|d@jP`FPmamw`X}-$;ask~EN5_@XV0x1$*>mZBLLGTU`kOoGFwZq)jk7S6r~gr!
zj$(;v_dS{9Wl;$Gw?q8x!~Ez~1x5X5?17Kdn`K+awlvy*J$i(6VwZwkE)f6NOO}C7
z!B|k5%r#mqc9#8qs^$>@mdy1A!k2Sc&pp!GPd$P=_t=A%gY;WpW4=t!TbEU>Jgzr+
zPc@!FKA-xntWl9pzV+GDxo3#E!Z6k@m!^95os+?0%layN6~l#U^*H^XZf~x>u@MSQ
zJ6E?_l6AqgSP$FIKSr%=Ftz)uW6!M{W2SSlXNP^-ML;<#6IT(RK9}^+2^j2j!uh>d
zch_Uf`uZCD=Y#9iEBjhdiQ05eDxba5vfS9_zbB_w*;1M}C3~n`9R7Zd&q7SJe)zT7
zNMw~Ob$k3Z-fJ+nWl@piRs|JC79;~z^z8CpwCw=liZ{pggxT^8K^4|j+3&f|>1IU)
z(Sd0HtL>`y4kNA7HO60?$(7doJxvo-ns;eUGgyeeH7_G@D+M1hf`x&rGP_UCV~|m^
zasNQQpjwR5{4Dj78Q8_R^cRBTuPcOMlMYwv8%)qeEa75wmqGw7L@wm`&~Wp|6e8Wx
zcdmKa2A5pj>yNF{DYz7@Exh$`GBN)d2q0>*2d?UGJI~KggUn5V5e|um%zQlIsRyqO
zwPe=ksDJpNI;wXZuA6dSB9cb(ieR^H1>xbxU6J2}2?#0uSj=V5I?Z+@CEbYBR_ec0
zU@cOskKC6pa9`fnWjmH+ja=IM?rzuNX{eaX5)!ssZRWW!44kU-vzfhKyQ5^DElwS8
zC(6eSHGKs~ManN#DrCMaX4bb$b5JeLazDIoT|}U^lyRV!wll?hD^(;|^66^j2EX<2
zg;5s9**2D4Ys!W}khN%>sJdqJxdG*))=@a#1|AG|p%mHT6q0%ph4pXy;4j~8kq9)O
zkBLql^W1;lDcYdG$-o~~&NoLbpvZ8Sf$Ppo)~4VY5kK8M<zuM8I_%Nr@(?LI->4s7
zkWIy%X|&ud9S2Vkr5zYLv{kGL4QGnp%ujoZ&VzF@-F{Qdr8dg>#>&!ls~N?P`b*5P
z)#V<+PMvv8&|KhU$NnRr=?G52Hy#ml8>ATt2u(0lEj75RQ?W0ThRCyf$=5Gv17EE{
z)#|+-4q0O=N26$y6~mNr!jFn&bsuKC_E~eWr089k;={Dwc*s-`*nj_dOOU@`a)my1
zuYg;HtrrZmNeHGL+JP$wGqe(<49g#y*5qAZT@*AD*}wkE@KWs<%x-I199l?GpWKff
z>|hSw3K!n^JPjEHzFysi<J%7vvcex>BvZQXyL-rF3`if|`|36=G1XjzS<FlnpZ<fi
zW<<*#H}+v<`9g6Xw|v{&M2_!>FgzFeOpLTwxWYdx9DhA=OxMlSb&2?{vmTk|pRGCz
zg*t*wg~NSh%La`r8E5f)Y>-L3QLKJhGoQ9Cjb86V7wiWCq~7!l40{+NcrQkPZoAEX
zBo^u_?@SPf8APluv5v^;-tO;H6~8{_tLV+BS>7JBd4nI52Twrmg);J5{lckE2emQF
z?zytwWhhexuP%WdIcZ|0>x+eShh-@2okKUzdA%_nr{s#isI^AtA?31t`8DH(ZnrN0
zN<ZK06fC=Wsa3+~Blriutrz*)_c||)bd6;H`n7=H?C?<uv|9bOvWU<>&*KBsDG##~
zME~_kcC=?&?Fu{NvvC_>XHGCU9>rzn_?;s%m!BtlUgd)a&q1wHYu)wHw*(VJpFLnY
z24>iFo-IQeQ(~T$JIT)$Fm!iQ#7%&+Ao17^%6p8ksB}?9GKNE-BP1E5g6IM!{x|v`
zx2yYz`DaEa-=4hWFHybqO@SDP0|5^rE+*&p4-E988J(Z8&}?&`JkDv#E@%+80#fV{
zKDt~_<`NbjMKmHE53bv3d98KP7sFyP0t0PPDXKZSpA)AFQmJHVy^-!b%|KzcT%B?r
zWPsxNM15fMB;SyFsMJy6R8LLQRfufx_qEn!HmX?85OoF(K?M}u-bL(LoCLkhaSaTj
z(taa+3iqS<-A@y&ji8L{Q%ZEvLJIBS487VhZ~KD+{T>qe2c=0%4KPxw)c8ff$o%Da
z8kj@B4S`xDGHmcAgp~cx>qy<8eZ&JBCVWW^6V1b?pGzA~@`E@1G*lo7j5KvAR~F&K
z!%yUN5Q7re=+VsOYo1V~@m*nmI0N02=Yu{~{GtmJs-Q?%=aTE`fbuhg?eie+*Mo)^
z7nPg_V|op9#8e`Y+_^9N`=UNanr&)D`iEL!g&0c>FxDB*Q`=Oi+N%0~G1^(VPAliG
zohR$9LqHW1`suS3J{;hgSc<%gW%7}TVDTlz^Quct4p|%+_C3Y2R%)3HgSXh1*}`E;
zob-rKC|td!PZ~cDT+``0>RltQL21-yUUzd5ktpuaYsXYaGx^_)ZJ!^63R1e`+qoyg
zs|&8xyr1o@M9qcY3yD50_H1pMv>Sd=9dUmtYvXoQPT=p%>*sZmur2rSi`gHiVln|L
zvXsK((+jREC(}91?+}3OG<$r4@CmCt3YABG{jnJZ^%gwU3G}{isc5H)y6_Caut(iC
z8w-IK(Rzd94sZm<+u2J<ONlPte$9V5`uPnRLqev>yDqv}w#m4GtOd#{AwFUW&mSZV
z6h2@fx&k(#V$L1qAno+MAv?u-b0dKCuzx`PBI7gA8;Ze0TuK+J`L}LgUwAW~qUf+#
z0om4>DD|#l?lmarp5tg#rP*(eprOGqrg*v-nNAaYBjiqx-Hj*AHlYdJL38O!Nr{fe
zkXvPj%~wgkC5l`aT}XF(kE)@FZL))O=oAJUJ{lSpv%Iu~DoKdgcU)c}L=skZ+lWG!
z)5c-@PK;fHYu1P8vdfP`w1W)_<@JcCu35Ft21@63<uanWgr8fPe_go#BPN6KYu`gk
zXd~lnil;K$#X8^BYbJT*0*y)UbYU%dTTP{)S2=}U9_>Od{i=tr#$LT9l91Jo)VSjE
zrRBrtKwO(ZyT_%v!BGuKTDR{#_^^8nVJcptY@x{p$JDW<+#hQ!W>aBOG)K*KHh9Qz
z)nNs|qS$qIRIS$A!ftR*ST_8KN-H5s%gPq6HhnIgW4F><ed(rJA4QMW+i>+EHOJ5}
z+>=*Sa_nQ4spCrYA}%yByUp^?nf)K6joM5XsI2iFU<J6!OKQ7V$LN~pJl91mN)7SC
z9ZNt&T*pRoMbk0;(Y7aBAvk)~T{kj~aSGAI(*702{Pj@%6<vy>p`+$KFiz6x%KOlX
zeuY9JL7xKfDBBv@Q<|wIdtt1@;6&y6#mYGr`#jlMnwRbVj}|hH>Wf?ih3r+Hm0zH5
z*U;K^y`SWHkm<szYoF&e>%$0+#;GnD3WpGnv^VOfR6UJZhGAX#y!O@kn7T&SSv`lg
zYBT?nGTJX&n-SBqDcc_!Eq8hn-3khNvNZap53!s!2`2emWU#L6<-_!*+Rx9*k1Lp!
zwsnDM%I7k0nvC<+*)z)khB<djXPNL|by5()0e+BC*wwvZuo~8`V{1Sy`EA(PaD3e7
zntjTP^7-`jK{)YtyQ<HS_9k^aCe)vQy}`Gtq36=5bM<`4mv=fB30JG1k^#z4aW=8A
z&f^hhV6E)Nb_|u<yp(;1H}tkkxK0pw|8j|*h2zD7ukSga&uV@JEv&~s$L%Q(Sp~$&
z?JdIRY%H9%X9M>v@GIbOvi*@(Zu?a={j8o9NC3>M_o)hOqe7~|{$@6jp<P7o<<p2Z
z5%hqQGN)Vic~SeV_dGccH49A@^Et^1&ro&_JE~Y3jO!iUIe*3WDw$T#Rti;9SAyf)
zZF7jDW~uA$T)CHCOl>mfdv?D~M#uam{fi&>D+gj8-m!ce=oqfcF<j#6nHdEgyz6ap
zyRXav9l^CLOrH4Ql`-HJUYTF!R9J27M$+nhY75;q=JsnMPnnex<J)iVX65p|&$vld
zS(iIGUSwT4r}zEdT-{`xZbc4HxO#P;8P%)n4i?TQi!Ukfx|fujT{Cqs&vK%z?wfZR
zNAJw<wP<(DD|bkmapj*+$kO|E81$4-u>+bQ15P2X4v+AElm58!pTh7GIXvld`*Y2T
z=g}%N?=6w~8l#X^dw_ik=VW*&>#|;zV5p`f^KN*roa~rTz$L;=x9xMjF3>n^s~>mN
zTJOWhXu<GW=@dBl*DUMTMld6Vx67fI7x}d%t|~%I@O{1v1)ouT2(_M{UaMfX6|k6+
z)h_kEMDgeRjX4a^mE5N`5+I3YX#yV&4rh|pkid^$EyuQ*P)94SO+~+Z#}OrX{oS$C
zu7CZ~d?5zOiP8i%skI87qDIa=Qccz>6HpNxaFCtBhqySIg-E%~#=kjCexfawQ=gh(
zU<VByd}r_sR|<7KKv%H@-mouLay|T))v$H_sgeLYk6kO|x;5WP<^R?CYP5}0n6Kks
z+5WFDiR^p0n15XaUAiaZdF;`wr~eAW#anS$AC7~-ZUvff=V%s~HlC)?Fr5rlxfJ42
zT+NtSxkLBsPph9k@p}us{6dHlIM3Z$BX-T}v;lw!|AVjbqkIOkzh3h1_xi`%dc1<j
zUex1|T=>sdOLoJ_)-L@DJFPc`7@0L_PZX(?5+(opPiU3KArvnnWmk_oc}bfDMF3MM
z#eI2Tf9^N=18Tkid3N?|%QgS@BK-X>|F|^CoP^@*eX*<j#{U2I1^(?%e*LvKwF2Dd
zh-!z&7yo&wfA=SUe~AMkSd$(L*1YP!Ikx|E`~LZw@MTX3(9nMV<ZI%6^NXIdoN#tn
z?e0Od#_riW^<_)%<|h+F?lVg_s-{KKkJ!ZBxFRnZr%FU_dgDgBerQv&t32Ve;fgOL
z%rZ0Ts}0yoKaUb$I+!Ne?})AqJXqVTne#Zz>6k0p)pN@!?%ULj^$`5H>%yifAMp1)
z?Qb-8|N6g@lW|nRO*gqYFoEHh7Vh6Qg8%FP1wJ?p3;KWL+0n-6wzHv1;QX_7H(c?p
zIa1|Cy%cvD;B9?M6@?Xz_fPQBzx-SL8X~T_c2V`!FT|-3gSWote&Z2@c;@e5qw-zO
zi{zawF$Mtw$fNk)fs^Gs5O#ULOBXG@8N)_`edgOF9uccBI-I)7c6m!$jUl9pGc7SA
zCgmdUR&~F2>g=98+`Y9MFBst*SW9ig_SDh&u*3epi05d%_dq2{CU&>v#$WvfA*xf(
zL_RlC?!3ME*7UF?hQ0lB6hk{~lxz>%(XjIzng;Iy-(qy^V{rQ7xC&=1NhuYOEd{_s
zlwV-j))y&}8^nf$gtYnL5O0Y#J%TCWch(~n-ZKi5vUF`(g3BD7jt#Z+7L+soCZfwQ
z8=p6jZzyH7W)7aHFTtHgbsAIuW6@5V`;&5H4h4pDi_O~z-5+@bnpE!2)rW6<Gwt4W
zahVVo`n#6?um8Y}0Nske0Q2%MqvYso=&iAw0&v#xfia#+=C%j&Y=+h^56tKHiDv76
z9L?6N9(0GM!#hNv_Kh=$mo|IpdwN7BeBs+2p+n(NqS3<qqv_7Ncba|615PSan7%g!
zCG9(isjdy6G~P<aY1sAO(OVAkmFebhIos;qZwD1!w$T@twHFHv>jP$AXj=z|RE;mi
z_9nh0-~wk<0eG43@6u}OKa3o18p^qt@rJ_}92ckj*YEZB4{@I(U+~(MWzxTvY4jy@
zR9TKDD4ZPqzab=}yO=>tm-c*}E(bk?+J-Tx9?zc*$#DF1cMLrJ(nbji3GXAL=I`K=
zVAAceH@L*KCchvPdQ>2?di{^P^JJXh@GD4Qs0DU-di3PiUhezc)m#2dmgHXgI%3zQ
z=CEz&_u?T(YEY1_2pcOZ{PlaZHW;qEt5)EjTm1d($#_(zN}4Pt)81L3d4ERa6Og`O
zX^H*kM*#epSmEXO%2!sey|#3lX$;p{;aqXucTS2eQFQ;k;!=E{4))wWxxzikKh}2s
zbCh^od6L)3z_@0Ck<49aF>ivC)dJsFc8%QZeb%WKx{t8a?)4IU2c(|SFxCov3jSc8
zYUVQ=5gW0Cos}#jF~7_$g?ZvZ1;yO4&rThWjz1B0h>N&)5=4LDpJ97p>sUTlSyO(p
zFG3TqasnymcdjxZ2M6sRuX!b@lQ=$+iHm+pl`A$ncfXZWZGCRg$K%F$WU&A4%kp@k
z6T#W*d3TbYy3A53h58nrnv?h<B&Na~WBBfHKc68ZX<#}{%wSwJH(Xmg|MF|U18aSW
z*Ee$UE=I+I_oOg^$wuwB0q-=0ui>|jJFd=}v3x&v@1HmbH!krJvk@D<enWT8U}BVM
z^ub<2pvi)V=%D?u-7Upj{VQYJeY%HK3XeLr1}6_@50+=uFTBumyV-)@r!P!t9B%0z
znSVYKONhHPLGg<I*)L+&_~U~@iMdu91iPOuz1~eWep(?YC#l=4#fyIOBUh?<jC6Z&
zVs&Ev*^<5JDCpUQe>Xo3z$b7WAUTp@`h4(L=S;Z{16Cz5FSJIEC6K;0KOBd=#LTyp
z3EY}5+J}+@;vHmK#V<wNN)UclfT;)+Y1=NB4jbVME$0^@@n`8QR8Dw-?aqS`q6W(B
z(h#p$HDS<{YU5#Q=jn+Y6OL)`LkW)$fgPNnegcg)4!X9lU~1dtDdSfx{EKGcf|EfF
z#X{axV4j?!M-8RCcXAQbrchurg?{raNZPoaWugDp21r2*XmHsmGgGFZQMAc$oHDFM
zgu9Hh;5E8c1&7Hj(td|ig7@^9b6ilvQ84BPtk}vD<RT^r-3S>0K^oN$OeZy=4#W!p
zi)^dUJ>f8(qtGn!!{FLA@Cz)`9~J&+&$4$2Sffn)aZus$gSTj^762b{MWKXu!yagq
zad5=Eq~No7Q?PEydrR-)W{U{HRJqj<KVbNF<DicAp4J{*QO|6WgDR{$Pk&4!IiLc1
z6TD#J^V0GwqK=+B#bf&uN}K|)Ra_=duFb_8BV!ZQ;orj>PCpX%)|`bSfnP>O#uPH_
zS#VS2L2|VLVBy|^7sBgh!FUg|n(ccS*e3aM1~tZ>202HI8Xv!vBgn9q@KxjBW7oyI
zt(KDy;kfZyzlb{>_dSjbT;cg3ljer&9QZ*zV+2Suhq;@QNyG!VCsk`GXU6JrpR?yA
zy^O~Y<5veIn{RJ|b0ZLD9r<pSj;4{~C?NhTk_@hhe{ukAw58xycex!L=Fm6nW&H31
zpViuP=ELq>#e=3VJciR=O0dGK3PItd=CKHpnl{+2S#Fp-D0+&+i%sWJCJG`Zsj#_s
zW@7boy>2{t#IByr1usT(n@<<4m##-3YAmZms4afCbjP{|g2XnM<H$H9Xd2}-%$eD^
zBC8F2x8l_i&{};gHV-2Hag5bb=NYyp(|`6Okj1{gS7QgkpC%IMvQ$KM?7B?S3I9E`
z@$bQkcLv5ZmiT|{4tb;vAXrexaCd~ZUBP5=2Rtux9T7Snw9ek;sAauK)Q4eZbw4|G
zX-aiJ;S$g`sC#ynK-gjO>g`5-C22|o>5mAcXL31L2JJq<wP{p4nM4GsxAMJe2-Sx}
zG(STFQu&aU`q9?KzRaXA&soxY$WnqxxxHUBZgzPQqRc+RSTfBJxUsTNo@STiV>i0k
z{&7C_=`xhmB70xxH-T)Kr3~W(e5i!6l_KFEi^gCMqry(le+<S^O*=qP3kg!udQyyi
zX$2mxFV%OW@7`7slD19jAc-=dIL%dC7W{-G$MaoizGHHZgzp(95q%7T7XI({!_^4B
z+{zhLSyr&mABP8j?zoA4hduR5!reG+sEHPFp6$Dx+nfO#Km&sTxsN4rALI1tc?!VE
zs&_xkC7jQ@18Q0}GM{JPGlfG8P)c?Eai>e5OA+*XhQNH7dDU3Y0SrMxf}~vxsSqt(
z@S)v6lo?mu7r-(1s1q(~5;|)U-WpKgtX5qaf_B{H+2+m^sms0M?-=Who;u)L(~tLf
zpf9OoL|<3>gk#CLO2X2Lm{ZU3JZutWpucTGXQaqJX+&j1Xf+7q$*k?6u-uVUg$nG=
z?|r<>D?J_O92gWg5Yl<CnZP*kz!}Tn7I&$PxOKscqdU1AKS9g5aN=F=XZLB8)ko^1
zoF*Z|qDR~1s?Bv#)+;g7iR5GoVy|zugu_!-tBr~aEY3^5`a@wyYyUKQ!fE=@lJ?gE
z8Yj-2Q<bNhtU<LEY%K;@K=l(ei9V7+LeA3b*8Og<C(UJO&17z69iAC|ySGZ4sy$Xt
zI-?tkZ|IrwF-~J|^~p)lg|QYI=J_hwBm;ebr@^(EWAPAZnL89d70U9jiirRG5EX@X
z@-6Z2U_bdBks!#Zw&k3TSUIV!x_zLjuD7{vEbM1-JA;neYD;bnlm&0JDm;%}+4V-U
zoh#)lS@gVAxtOT)zOB0O(qa&Ih>T_{&!HkF<XnQpXj_Ol4du7#YOaBzPt~1vG=RvF
ziS%YUHv0=SA?$717GfJ|`5f;l#4?PXx-Hw;PhlyE{anha<@KrM7C(`I;`+Q}aGc>S
z8Ye*SL_AA(;Jm+SLB!E;`Wdb?-TL}#0eMn-9qE@EsLWM35ZIJS2ws?1YZhC?kSh6C
z=-si}Iv)^3fZhzThDgCzTbhdvtqEVTf^O=$E-={?zu9DUWaJ~6m_$67^K3P$T;AIf
zY^qo7Vv3%!BwD&#Wi<%a)M7aorCm_prlr>BJTysOpok={o)ZFy=7%DObKE~b@vORj
zR>C;4tW0@M8Xow~#BOX&R7Pq)HfBF;2RVgk(|FR)32I8(zGPGO+dD5Qxw|U)pyBT8
zkXxKEM*K$fjBfN+Vz8&Qp%%8I0cX9=C%w5@^d_`Pv+1b_oQA_RM^Yn?Dy-wt{o7pO
z*HOdmNn&E{!po&7W#c30)J(xdM%7c@v`HqFLov{Nk0h;s^+63YZIRL6`6w3fx%qHd
zBv$UfWFkQ~VUjNcLo4a8S)-f5Xc%6o`?YSd0=3U+q%GHFXNVdo_fF|au2*D#V(^^a
z4Y|x1n&XOhE}K>$HORIzm-q+~3wTF(f5MnN?Tqzo{f+m`a2QuA0k#`$h_5)E+Qa$5
zdcsM}ApdcfPR<!AT2vGMM^QRXGVoxGUT=|5yob}1f!>C6q4RHbQ>x6y_?@V#1pE1d
z8F~{%Dk-+o_$fA4`8^fVx=)+Lespn9A)n}TI5lzez4+DQ1Xjm4C8W*8+EdmE3iG!{
zXc*ls3^pOMRJCJlu=b=TR3W6U8o$JubO~3%nXX<GdoAE45ym%^q7)-ZG_Kw4X_!MM
z`-sqKYD{zWs_(hTmkq&n&G*~Y!tSlc88zT;axPIoB^nPiR+Y83ICCtB5GUCAZ0Ts8
zA;0^~Q4p9|*Il3cRjco0rkZ8bVJF;uPJARGyhc^u{fLM{bGN(O$x~;|XjhnJul_I_
z=7h<|)^fvzW~V$>2WO-H%YauZQSvw^n?&!gQNKI%XtgwwO7D<F=`GGT+GQM~qX}x^
zo$)|N?9+ri7knxpma?E0Hl*XQct;}R2Nfg8L-KDK2}S$%@nhQomaL3KmczT#XPg(x
zo|CWC5QQ05mxe`4pg3(PRpzl>Upe;49sditm)8oQR!Z@z>owV&OYSCas<88W3P7{D
zp7m|cPbR>)e&O&Vj1Q1kdIB5heg9VJ$K#xU0Ka|{XloUs>PTZB6{nRs)(<zk`=?)Z
z-a3RxjPK$m7LB_B{DL;Akk2V$fiq0qQmjhO2XzN95Ab@Gz>N4I9lA8Rn39%+)$B7G
zf2JLtEt<)%$@}9QbQm)IUAHnjXQi(IKd#=FXcSLV4l%F47bU1kFMSnb_{#G2Spvyh
z##h3@;{C@h?4E)(o8dC<sMv)!Z+35O0ftOR>&(ENKW<sws4KDLrjhU+DV+!~BTpM&
z15U#_|CC$!8W&1iXo4HJar=rR!DD!<r~(atdB}SLT|`^|TSU!_PHKDigykQT<KzsT
zh%vTF>-oa_U!J{5gi*KQ8?3UMu^*3^q9XdLogeYg$8!;V<GDm*m7MLo&OLb<<#jHy
zjMMNiG-y5ah}H(-CrNuQ%CC)9hxy&$o9trQR#)!>u~6qDqB}+X8}y=A>2tXz!#Iq!
z_EYq&NV$UvZKTLsLrSkOr1*qkQ!8#S_VB)lu?>`Mu>F-V%bPF>e=zUPh>D7=c?aeT
z^1F|!v5n<Ac^in|EcE3i5k{Cu$Pn!ljSf6s_|^!Z+*TxN40iuWOxXQ$DYx=#(C5~m
z$d<etZ>n_WdJv7%n&++%r1g{x^ckTZDDPOpJXKi?Hd8>pkhfM8h^N3qx{u>t(dVn+
zvt>**+}0CrMEUi@lb|2BR8*n}Pel>hn9E9g%3`SAtCS6+kgeN#1#`0$ghwyMjwT4a
zbBw-%PH)5@(KrdU7SKYrT@WDi$=ISVH^)uUfPv6P+7PM%3JgA`40OUzSN#3V_~{?<
z^o#MV;P_u<!53Z^A8~?V;2onCsZ*hS&=U!D&N#=xDf!TTrMQpc$8wx1GyV%_rwLsq
zI-PbK&rM1O!}NB9M=OnC#(1$bs<TU0xNzaX1$vm<9<klQPtFQ34=f*_QjFtrV;&5u
z<H;akGRNvIIVyNUSes-HBFj_-`XKe8{PWWBv7ejc#eYigz5ZMu?B->l1fy>D->SRM
zw%x_~{!sc!_Ti<~W>{(+L~GHr0DS8fv4Nn{Za|A_^M`21VlBt7vekc{+y7ICs)d3R
z;_MBMKT~aSbpIQ)__Z_*oXEgS$Ub9H>NOCFt9?4{eNkFxazeIIImvxPeiS^07)^;1
z4bH`X#)dVH$McrU2<OL1+T>+y>Zb~m-ZvUAp_Cy8<IZ#z#XvV)zOfiOM<43kh{Mkq
zpGRopn5h0ui77h|7pIZo%{8Ead_ODQ_C;-Y8zHu_sY@r(qS~2Z-La?j<E3hZUTcVq
z%R)JVM0FyGxC2v!MMKDE@a*#lPpLLRaJw-%RZETGidqY5X7??|OId}!bqMLvM7gFf
zLfF?TF!Cpiu$)bB8tpnjtt>;VEnAmMfm5x^@tlj_xmDIMcpr`Mr8CJkKe=<boj)4e
z4OUe%`&33uFT4Ec{7p8LzpXPg1}jYI4q#4dsBOH~7f;#~NeU;mz6CYQklw+<e=#Qu
zRTk%~m|j20(RNQh5vDHP!a<*g7W?Z6u@g<qL{`ea%6^epA2B(f-#<{jAIXV(78(DI
z0%i5ZkTyVFdXpEV_ApC5hokO$h+L64Iy9_Ig8jB~pdeE)A>Md4Du7Hnu6FWHoNW;H
z<>1Ju$;SLLQ@;31E$;#Zr5oM-(9RjlDbLyCEU>4(Or9m$#~#8yKjyxvNY6QsvDwaf
z?vb*~^c+P^kAHd<WH!8fN!ucRS;qjDwQl_WI>1FH=x&?Fk9X#T=gCS;Zk7*5Wwt+D
zgJ?p(?^a)~?&)Tn6xA8AM?3*rn5W&3h?+<{B%E*mXfu@&T+-lpQN_w3!)OJTcqmuu
z1BWWqob(ks51ILu2YU1bbWX>T2ThH4gfZ_Cc3A+B?CH8*y(j&oiV#@c;!%Z(m?F`w
zD5=w(Z*I&7zL7QMAI^P9XI!y$Gu43?`Sj9GzabG}7qjbOK5vUy(p{!#bUWupDH$86
zh#a~rr(1vT`xhyWI_13ZAM`3*?^N6q#C(N4u_fEj;#?86Sc|^v!doqzlLkvo1EIzd
zRD||HnlJ`66m)N}<4eRiGZaDwk{`paQ8F3Bk#oyJ_CkT5v&G%>L>q!>H%@sFN8DAf
zD|;t;J1;?eyi1N|yE;k1^a3p54w-c<AzutV=WxH)BBF2o)CjvVLUB5c3Wv%{-(`mM
zteut}`ZW#0$y}XT(%Bs6o$m$DBCB)DKQXjmF>UGA>1yLxSv0n4O17Qp%Ji!*o?K|F
zfh36Bs--79S=9dvVmcK<%p1(p`1%20Q-4&LHl`Uj$tmn9S;^H|J>VdlF(r91IirmE
z9>qs0fMq9Urbn_E&njQ5ufx$;|C*HGO>a9DPGIhLsk-1L3QuUNJ-!|MpeH+B5mx*B
z1!T2T<IH4DVb(HB`*~VoyFjfXzNVGQ&-q(f6P!w6O>}!tH$7Gl8X5jmlK+<~I~tBO
zV(h8vldbeuHT$hNnh^=ho&ss^C*yEGsar%v$)TI$Gc5`iL`Xj(8ru~)muPQgT>5iI
zwgf*&HgrZ*pL3Rtqa(~C%Y<&8vH;~(7@o!^iIiB!qN5Q-1hTFN^_waLa?am<JJX>`
zz5}k7H$1BJDo@H88r;4XMzq>Dcb{30hoMDvD-Q@tz8wU+&SO>GGFd$B7X)E!Gf9V^
zGB%icV&BY-SiR7MU$n635&__{ECGVKOdJ>3>1B_6FObxc=l@`iK6Msbf?KaK_>=N3
zu)WnJn3~d{oQ~5KU#k0YlXVkUlAzxf9v=r^6@&?<=nQ3g+{S=z6xEA49O#i>2+j^W
zkasAuomMA0xmVpJZ#Gnw6Y(Z6py~qu!?;|<Hlek+sf#<c%s7fU#z^6eB?e&ama39j
zi$~ifQ!U9k@W7Ru`w8%VUB~5SR{KfdnW+<m!EYwHno1|#eg1mvf>Kb#ox5}$!CXFK
z{Wwt+v&Cc1b=K4M^)x%f-8^>r4ZO7P8f%_-c0=8(k6~l(YeJSij^1V`7IvsF(ZTF-
z^mJmd=){1E=9vLf^|E=B@vns=hd)N^1vx+$mMMhdXV8~)WSM~gsdp8;3CuU^a8B|$
zjXpYqQ0#HXCpQe5Whs>1P1eV5of23J*SGjD!WlWRc0qD|d1JWiQ0L>O;OLy^%J0G3
zP*$Hrcn=pz2G0Ddr_-KtZU$LaLdfHBT|B*+syb5alj}~a@gx=`VzyIqZeZNGWsgHA
z4_n1XTUGUqA;}M8VaYMzxm%V%oUqt)E=7TE)|VDd>>w?t`>OryEm95L$6YdxP$pwD
zv#%?G*5kZAV8O(NtOkp^-@D>*6G}=;Dr4DU)mbM(F5?I?`5h2CbaAD8rU9!58;>A~
zaYby8$28sk?fH!}6WBHnJ@o>WX?&i8T-LH*4;v#M4_hVVM0`2$WUt4G%5u1@Ab{VW
zjC-rK_*6s)&}j41;~#}s_2;LsCqDjvxO?+>DBJdbJWC9U8M0>ILzZmG*q6wzlq5pP
zmMx5IVPr_86k247QnEy{?@`tkS<4#AI<l|*j_baw=XpM#`>Ege`{(z+x#x9fX0GeJ
z&ht3l$9svXrzNiRUGeTAk2{&YhQu(BtPv|HDcbqoffY)&9SNO2)Wh$kB8)wurV=?%
zas^!PuqG<+sj$4ZVjn4O_7>`zz=pD~i=^QqdQ`P0V!M1^b#)w`9V4xM!ej-v<<)i^
zM1W3d{C>!@zkX3@@s>YOUiV|~)rz;ex5`$RVqVhDEqEz}Ks33}05}(P=2}-POT`jP
zaMgG$yspZX=#jj*c<vq<`a+f<>KV7g&i8;W3YdopVad0_T($t3;49Sr#pgfWwoodb
zdL))8b}~JM6?*0nMX-WxV2j`|C(D0oT^&5hh_RbAh)@=K24S6_?4D|Tp?b@>yc1<z
zNwvZCtw3%lqQiYIuI`-@%?2a<Dz0VoL`llwHM+HA^|4#yDN$98FW>dfC><(bb8QtI
zD_hZ6oPsdsfq=ZF===AEV3>K|;5Hw6bGW<~ZpIDG;am<B<?ubm6cY&GIL3Gv(imPY
zgF?1jA)+)|saHZPh<P+<Y`|wWpa!0)lvmy>+!p)A!6l2iTW;=8hs_=YGf!(O`-z!T
zlMhKzs-#0cr!p~V^c_TZq^%KKUpTa}wOw9){UH?0;xtx3iRU~_f=Z_)taDEycx{n#
zb8S7Brs}K$S~>HgLOmJx!Rfx|Pi}>yr=d-Hfu2vpPO7tbK&1SWp;t@q1E3H2T<g*Y
z))->=E5NLIL2UE^C1y7GMY$3VAze^<TyCD)VGr|CFM-FI0-s+ve>l<~hNDtv*c^&0
zHqHZ+1*i19PHiR$s~oY~B2cgu8{Cl-c^ta`92qYYp<&bXm?HPPQ29pg{-X=!;L`Dh
zqQUKaY%05ECOk9m3+h*G^f7hJzG)u;vJHHBKZ-(02;|uxE)q*3*FhMXI&q%p)akzy
z!|YmPaS3TT3ZZ#4Q`?toLpi<%)Umt#x3%H`tH0lf;dyvEOqJU1t39<L$}IVIdS85>
zRgy~%E3OIu)clV53ABHyC;s6Z_^aHnX{#j%!S3$ebCOVQS8ZWCd=R*9XSREPPIJS%
zwL3<K3uwMNe^2A{l^{A7jsWiInuF>OJ}(<N8l`naxk8Z}a&cmRI}dSL5J$dri>hzK
z8zwMJ_?x&c=vgXynFiS7>C()NMcR}Yf}(HOKfY>19%8Z_On6gicXerqzOJ_{0G6_6
z?=nI>{#>8&RL>AjBkn5EntJC*T8_m^{2UYNO6=U8Y)-P%L4Gmc#Ff83xIrJ}9kimD
z;@QYL>lxZ4JhxDz?H~S!M&Fj(n8b}RLH_S3vuec*%UY;)fY?BM%w~<7X6J+M5uewD
zl>>Y#9(ys(0y($ARwdrUUmL$yThk~iFoAK-VH~ct%t&?%p<9JdW%Uk7e)eL>ZUeQn
zpRlUea^+y-U9c03DjQ%3%tRNYU^<kwvAdl@H20h(BBPynV(`F-?gTCcH5G#=HE3p*
zFw({!(e)2z=AAwOnSnCf*J|f{O*#7*=1omgR@BM0v^8nQugVV?3-Wfpuex%uo!zI?
zWFS3sg4aQEWZ!GTms-LLem+i`o8hSO7{UX65RU~;HQG=%npCi?7vI0Vk5_qS?>Czj
z;A7i2Xxw9}ze^u_Iku9}(+Y%o_XdrW*c&*7PKE1zR&So;&&rJ`yn#`%#SBsx)mF3#
zY9*h@-2!u^ZXMilzJdN$q_rb-T}Zh?DFRcrCe|2F=Q@i^Z3Fza7Eh<XH5^UA(A}9e
zb*uYW2a`n4p7=h8B$t?{sK*!16VPT&bW^6}*=6Tb0>8a(SjR}+#qgXpT^PZ<(Md9V
z?gr!ZLoZ(1r9P|QWE#Hzb)d2wV`5`k8EDy1OTWMoF!F7z#m6d|<pbf&+cSoQ9lbJO
z;`ek4Ha*RQj`>Eq=f;bw0;C(HGcjs6<@IKmc+T+}RWyIS?HS;lRjI;!D~j@URDh$+
z3~nW0UhwfKAFmXjl}V);{X}M|V>*Lpp+1YCP>=LWm)u(OGc$m#z(HL_KldPNTq-V~
z;#$Q@#por!T`7_LRvU}ur0oT9(6?)?P*%)Ydd=OG+|bCKv#s!Z8IT%qVF)wmv)SeE
z9AP;)kv&7a{fBiqg3u@K%Jee*tp8LE$-~G)6l5EZAX{KQ1N|jqJ?zulKr1$T8c8t0
zD$L^fn{!{IOQTL0%Bt3#-8+9i>G@C|I)|2ilTm%Jhw=|_q_7bPDn*SlbWi;lxTnSd
z29_5?%U*6k%5N-5&VTm##V;39bOiJZ;;%j4`Ci5M0ge<z8V8WvZYKPSDh?ANKj@0x
zWE{M>tp`>xA&r6mMkx8m>h%&#^zTysYOyo5cMP*&4$#|}VJzHXg5SrC4?sG#?4>Jo
zh^I>EGwmH|#V_3YZ+($d-){uOThF!|bwZMG;_luzHv>Uv-a+-T0WAs33hK+VJp~%|
zm+wPV58HwDYZ|m~&+|0+CmB~|y3ySKSuOwi<p1%)J`}!~y-h;T&z@5_k{n%@5A+lV
z(7}o9x%8?d;y!*AVQOW;_ZxfT^^4{*7P$kTnP}-1wFac@55g5K^M*~1z=3xl4%Ln|
zng1Bg|0kC4@1MR?6<}df7V=+CL%zpJk<lY;?M~x!Fy(as&q}AY)?XttVKGt|aAhKX
zMWNSJWPKq0NXF?MzwDjvfB&=p_!<L_J<KGWPyPU}{g;rUYpS{!SRoH|7J%#Xqy6|l
zdx1tSGiInj<mV#)Z=dvE-cQ&APC7@TW$yne{`9|nt^XG<K&EPqg$ps;I4vyrNE%Hb
z|6kWum6e|ewd}b(80(r-1bnX+2hz#5Bc|gQyXGHX`1&GrmTBor%Rjvi6?iUjEr!4l
zQpV-+@u#<Mop{CS*<Wz}(;;wIVS8WU@BJR;oT~4S-anOjgpry&;<yMiWqwYZXtz6O
z3tcFOqVJ=mZWoixBR6*`D|&xBRS3<$oc-ydTWv{=N<u(6kh+Hu!dk}X3vPz$a1`T?
zCX}AG`2R!fA|UJ_&v$1}{kYfpH|T5P7SJaXvfi8p)|dm(hsR;E)@9<6GV_bn#Ygs*
z=U`2SmH58%ij8qx?%HvjZs$R;4ShpWjW{DCV=@(k;FqA7F$gEuhwnw!GAAZK!Wnlj
zV((0=N+@Dt4lNBoC*G$YpcH9Hywx0N$AN*ArL^Zr1HOiC#D3ZhX0sPcn=mVPAHfBy
zI=a=cxP~u<VV?BaSJ&_amC29d`xU>ve#GA|j-SH`I7!_9<sSA=Pu2h4nmqmo&FjK{
z(7aS7^DizrF97o+83GV*=~NRl4&r@~%m~;~TD~TG_IrZ&*pnfE>^)f#bFnJ$Gwc-u
zt$fbu@-Ad25!%~)G_#~B16z=`KT~fAf=~|7FpKM7uti8Ap!PXlaEjMk2#ah$tOq?(
zzk-_)u$(T_v&cFpI05&g^dbEAOXi%s>&M&!)+p8h$zhG@1wBeeT=bzAdM&Ym-Xht{
zG{cHky1mU4|9DRPcvJNxLU20Cx^MZXK_7L}4)XIvnVoOV(;oaHfc59wggZgU2O5gl
zC{c3TV~RuBbGKb`P39tdQ3ntl_I;)vGxav9pjYQS>fEhfU^dp|g>1RJ&$CtmUoX+s
z4bT;W4BLm-dw<Jr&3g=B9Uy~jvi`t`&|}o_FUUB?qm9Gy1UFqdL;Kt9Q?(RnvP)UC
z*?xv+XX>Rwuj~L9MzL~;&T8jgS?6QMVb%H4;bGOS4Z^D}t&x6uajM5mj(uIi4hSjI
zh3wg6J!&<s>0iJ%kZk4klyStb;(B3<(J!d;*HQ4V7ol}X{a_PI6Yz5}g+ksl6VI|o
zj(I^kvN^1Q<HjP}1U!FTfeD^t*R0`r-9Ob${O?S#poRa=1Z(yBzh;7Y-g~%Q+eVe&
zfEy;;2%}%0z2`Epevs~l7b{)S{gw)^ts^t#eze0Wd-uv$hq<aN<X>ogqug93YsG%J
z`@Znu#o8bumVdFfEjzU;;699VQ&rk&*(-}L2`RJeZR+X|@ryll{(SkmN!N+obm2u6
zpI96Auw}RI5P8~Jw&L{fI<uUy=VOP1jlC9nJXJZmDJ7d5q@K3{w9~}pR^l`E?G5uH
zPwU6ULcXIx@{|?6l)aPdW79q}WiHirJqq!CTt#0SYx9EKQ(ojw-I<aeC%NuIvr_-y
z`m_2a=Ain(8d+r>7Us{jhVtzLf&2Nt<GQbp?%4Q5xl~KSRg}D&rgN!yYI4H+>cyr)
z_t?RCb_UMh6AVX$;J$9^_wxQYGC$v}0ZEuGY+iXP>X1!Pn{~3#!2xVtCT-@D>|&)5
znkr&|2r(scurIuSlXIBEC!rK#s#-z#cdP3KC{??#AZv2~t8=8IklZVOUIvLtRUp=o
z82}t2yT=t=wilb6-0ktJ<sg+>Y?=eUO{43lvJZ=}g`DE|REBU&?*`0a{@hPg(@1{j
z(l>~XlU2I7D0zihLJsJj(nuKbWsK{a)KCS88Zdwix!uhNO)k6eeM!0^Nu%=Ey|3ss
z5{8M;OD|f1g2NjCk~_fMNCu~;B<By0GYgUq`8-};f}Z|>v}3yrJe>z9IW8DmbWe6Y
zbwNDJp6@=#&8&zdgL3$}Q-b-J%uJ$u$J|G%?d`RWkV$qN{KYoJgyR+fCSl=xaoBfv
zA?VQovQe^N_T?nD+1Dyd7Z#3?wICt#`<wY1Q0~-U;T_*14%aWO19zsoU_PK7$zxl9
z(4QojslKTIDGCKoij*A7dtmE0DfYJSgH?<Bd?=4t0lJYs@Rn?^q2)(u=HKF>52cjO
znNLsWg$;Kw3F^k=0=uT$@Mo=CE`AR+C!gBu&1iSUfBT8`){&Els}0-wo*Y1iE#04+
zcX)qtskkRhuK0$Fa0wxCg;&xo(DSrG6Xxd)k?={-ak`v4J?+kH=@B}I(@bWt;v87W
zPZ#dI`b;@re15=#YJ+n*84lni#IgD#%mvHnW}uv;n7afvWO@^MqH%m(2bbHx9_qwO
z5MB$0Y2}x-n*wMy#=s{q^>9z30X5%heH9UGCY=gV{;^9DD{SA0vGpleSBAifpcUq&
z3?+j$SSZgOW=khH3&DSsF(jO9;NGLSt!E&}a9Yl2hcwp*0iuUOeS(0o_GBx&IBWt=
zCHh_*7t;%?`gSTp*Zg+4lDJLCT}b&WnA<<+$1wdiX#=${Ze<NXqg=aB4|Eyq=gSAT
za(k9PE_jOVQzULTmT9vcI7?{Po6ggLLs}X%WtFiD2<3M)W{wP9e>!+=Z2d5!F!RdF
z?ROR#o`Fqdbiexje`LgmVlmgaPX5chq!}Fwa<G#a^`6tdy)UD?mCIom+lS0nQ=q>n
zhFf_t7$RR!uEh^D$4wo_w)IYacybCu(hIxkfwp&uha3DqA7;T8eHpCoic6S0FEb8D
zk6#gfRRwN=PD`92o1Xn8o_26_Y0QM<?!NW5d>D*KiH=)0ppOTuc0a=@uS=-hcB$4f
zw<-6XaUP@e6D6fDhq$G4Rj331Nk5}2<RyM9zjsTBlyd_zRxL0EUmpVMR1A>dam|=!
z(i|-vCzxTi6_p{_Xde%+ZcrA1{5{Jhs>T@**XupGzmG46<D4GeRkwo~bLCmvBD<ea
zgQs&n0PAk2iq55Z>uUxdS6@Ggx%~BR-N!>9gAO`b;+xK?<zcI`{)G>TKc6A%LEP1%
zcj!z|j6Wglzc3)@*4Me>$$P7ix3ja9ua2X?GQdv}M?7<Hg17w{%R?2&^JMat>fhsw
zO>Y%l{0hXQM7M#8(paG`lJO$cgB(;_4k7Bo)jDlV@;)Jq^m?v0X6LCKlj)tjRUc$y
zWh3PzU%V9`_ftNa8*uf7QN)=@0_{`O{+pGk_#D)y3s%wG=1tVt0-^A1ltbmG?uAF*
zyiaStI}q~4r+i{XOuCyGPgutHIOaFSuq&8OaCWieg2CX9Poq*z9{bhHigMk1pS`qg
zw>2Ws!-PRjZ)6|rgocBnyqAu~@|GQWvNwTeMC~bGSWhZ%)Fv=$OgEwm>NY6n6azJN
z9xh*){dlsw2^{m?w9M~5noz2>S^{OCi&90)8$ZdKz$QgT7ux$skd5suj-g9_*)+#M
zvMxYf(`$e?${X|MYtSO&Tu8;<D|$8)coYZ2#_siz<-4u}89*U52$aXDJ!d4TepX+_
zJK{PPT6#(9riy<4&@DHSeTter$w_oz#n-iM7jjc2>rL<#kE3gsRH`zD0_YnJ$>W`m
zd>nK+;(b+nD>iYa=h?6D?tg5!RL_%xbEp2VX3G9r4Pd}eRb)mZ&ut3FEJD%b!Wn$&
zvT@s1Dy=8ACaL#$t8U>1klG!uC_4$(4!%a%Pmc&B>5=+W>XP%EN!dD$0cdP$2pX#@
z%#+f1y;);bU?6Rg(r-ThM!smaaWgD-rXIB&6gI4ybOYqEA=?Z0-C*O-!8siP&Kk|U
zHrnxo;Ivc@NAOG}Mh_@K<UZ&jEF3Vp24NAnsH183a7l~zfUf3MajAqtT^_ZBu$qgG
zBU$J77?P&Y#aUUvPn6>EWpvkfQa71DYp;0WORDz~tcAVoLt;$+D5j~xt2oL}?VdG*
zkm0cc)>VyjpXnQEc~g(sNOUoD-NH35GVWLFAxSf>+8JW9a$7y0Y=Mi^frtf<!>2eo
zpVcM8#>krCiox{G5hvNN%Hkt;83r2ba~4XvDVL*DlmZ0F%a9oxV_a;XQq10EpLy)$
zOm2z>Kg#fEmVN<R{UXD-RG0AVgdQ<+LhVhgqbi}R!0s>&CS6_)6+7Rk_(~#e8`6|B
zH?HAY3`C!So296g$<n90W%tYK(4`uCCJfM&ws$2TTGG3_i*QyzhKA+wM!O`1;cnxe
zEn2rr47iavOoN@3Y2j+_i1}*N#0}~I@gwhO=Py0pibQ=_;<DQ@FF0Mtb)3+oKEpvC
zylObKvL!SLzw~5=q|Brrw!`>a!f(MWZb4`wUy@&M`&q^7GJwv4_X<!kg_PJ*Hn}Zs
z<P)L-H($kaJ>J5K2~cmXkgK=`?4%%C>~uFvHtpmJ|3o*6enT!mP$PybRybCIeqZe+
z{hAG)n#hJyBf+v|M(<|W-TZoUZNY4nq$4~y8%wp#?GaCG4t8)(!S@~u-Gx2-@}A`P
zp{sPfGE^FSd$8l=n(k<dEJ4CE?xD~m31Ky3zG1l!r*7dTc8T0dux49m_Nk%TmXsrk
zBs%4m9Xbf=f76%RKwtN{9VvB!c7Oq~w)8v6+ms(*z|&IzE!HQ!VGgOI*NqDfeMjce
z4{SYWWOc7@ukgkQi_9*C`e7Pgpf%LE6WtDt#z=RTb75ck4=UzX;q%SXWHr5evm`mU
z-()dt>gr5CSjk%P?}_7{T+pMp=&mC>lchxN!k8|J+89ccmaS!wXxJFQOVyP#dbbM@
zi1Xqr3+!K;ad9@|_!lXPbcKVZhYat22U=L0y#U82wr>oV1Q{1C6CQ8?hAx30;2*wy
zGwAWT1+dgGGa?K;I~ccKnPuE1>>9PsLFXp<%iZ%I8?gpGK&Np2&t!j~iNaD4xXKZ-
zIq%egSe#$3TN=1@jaP>l89+28<DtffqF8sSabQpI(52i)k4PSkUbmABW{E6$a*%`p
z&x*cT({JUuFUEKn94d|}BfbZ@CztM`=KQh;9vK(FO*{}K!NE@$Lw706LLP-2qc1PN
zgQpW@yF__J4?IIhs@IGP@qW8iAJ}wP{ZWsrsZ*<Hw}cRcS<<CcJu^Vc*vlz+78)1I
zz=Gnp6csvZtAqpDDI|q%D*ax6Kdn5O!)sF@gl;`2NA3!_8)cU>mziQ*^5%=bw$~HV
zN36JD!nKAB^QH`sA0{l+FpwU{8H5{uYBQ)v%<n+b$xl8XqaG7k4S(2@Dg)MAGmeu%
zIjoCZD+RX>1aV)g4_HNdOK0h_@2ah{kpGa&H+`_#uDz7}t@84dt9#|ka$=e!z3cMQ
zBscCo0bA?sFWka5TssQE97^hujzzF;-!Qn<-I+B&&)H&!JbUgm#&YZxUcK_|@7s2D
zoui}W%EYf-`5#}*BWxwkSl5UjPwJ7os_bFl0#W7kUXZ54ns`2v>mkn}4=zF}Q6oSX
zwavZWBFK{?UZ=s7SchWd#$IYepWKr;R7#_=o_D%j?&&#y`b|)&h_+RA?!uqAyxP!V
zXXe~1*^S;E>Uk~l(MVs|wcW$Xm}dDq_|ie>L0loI^XuO33NH@tFcoHz4`YtrctR!;
zbSWELLqche6Wj~XLG%uljcI9XjEkI`<59uf%B-Rp#2GbE>4-!94*QW5hJgtO(dg!;
zb=W^fh?evfc-~FX<ue3RDI<(FCubJ~kUr)ulbjhBS#9B8t@@gECva!TqwvbdQ2j8?
zeb(o~&4+it1#Wm|@+dwPwGxAu96!gbKlGeA-I5=bx7_cpqSP^d-*rOM_>|2+M(%E4
zWL7ek1<az#eJN_Ka|d0h-L9GvCmh7%{_!gRc+>bG0;S(AIAin!RO8+gdJwq1PkAgQ
z@WE&?)lNyf#$AjS@U7a^2_*AcaKWP)P&;TfHPE*#L81Th88zRrOY({Yk!QLHx=t+x
z`oee7i9)s3SsqpGM5q+3Po8D4?fvk29A$J0T@G-K*qsq4d#Es!p2jKMH2@!p8?N97
zRk)As2$31N?pJtMHTIgTppV+Rsownn_{XX0;Gomr%}0snA`YzSz-E={Mu|;CxMz-<
zo*h)~ho3R}joAfhVm50jaPt~T7K9p@Bj(LzKnbj+@d(_e_9q-}xxQz!_7vB$2%ahO
zlH8pRh+3?y&zc}kQ(``@&W|pYfs#<f!q?_3sBKQ3zJnePKU}=8Ujfr;m9+B61rU$#
z*Dm!)bkE5~6l?EW^)9^g6L9|*l<{u`x-dJOB#iCm`9CT4=xEhrwOn{5lz%j_2--}V
zKsn720iIPLd|4FmXz{Rxm5`<_MeD&yftsK;@*#R1w^Y|TmSC`ZD)Qo(E|C&=LSKP_
zYWx?c%TvoNX9SG&@_5zd?e2mC`>S7h4;EJ+)|3xzV@c1mFASP_8j-a7i^ZrD8sVC@
zm=P&C+clD0S~zUJ!{IXnJ&@f>pNmAf|AK>`XtAATfPqs#2{&3=7N$~;HfJpZt_}{J
z6`h48{7Q=$=^u(NQTqzUh9>h|%cpQ@ZE`ms40S|=M64<}JHxy++Jk4iK7e%&H~qAy
zg<?0`gS3)bTf}QtMFhEqXOgzspR3glhUlmg6Q}+<k<aVHRpiJ0E<q%;4gt}IV|E7o
zq%NkRG#NL8w$~=utR!z-BehKheTR%_?I6(p*&cr%s9UvRCzT+ufnE1-W#$C>GIzsi
z7{bcr@Zz~&W=g6@0Ow}k6~s#>;~HfUl%)4ADlG<7F>E4%4#6+Hn`waIN>)cuPQi^(
zanSUz^3&IQb02X6Ig`@2SA5}|cU~P<1kpvoUl@TwBhKN-(5^Wvz!)3~Y#gKxBUFi^
zD#Tm-ImqW<v7BewGB5E+i)Mo1`n_zur~sQrzFhw?&XgxZbI@QF0>9*vOXCWG_}V-i
z+T#~<Tf)hIFK?C_vb~xoU~z~rQ8l<G-5J=ka+g@l;c&01QrQc$83eF)alp##GIW~D
zx8r>Q-}Gi2%ld?R1<Wq4WiHf6fwcp@L|JZnpfQ_rV9CXN9~V#X)3#Xr=-u(bu|%>O
z_5F?;wY1P?-!3?X*Y|90MQV1!+RoShlmwv$V~Qy*k-uHwpiV|cGF*DaKk26e+B?(*
z5I&a?sJdEdIk5+IOD-QkFK{QMipHU%(YXkx%ivhV0kniFZpWNDT86vhgMsWG9$y3*
zczVNvq8sh87WK?han+{6Ypi))?a^CkN=^IKl@`ec7K!2BfW=RP+xyUC0glt(KVn9d
zdNx7sCb&L3D`1vDgQZMku@S9k$)x^dOt*E(F;hS25?YF|FD1&C$5Nk>#D8OS{Y$LP
zw5nom$_nZ&Wy!?_kCfZjChESzpuwcFH}tA}+AecdRJ4?+q0Z&97XufcP>V=$LO0#k
zeC1=lO?`sE_!(!eH(66n@El|h%Wz%%!gE%Le!9kdH$+M&KkTC!-aT#x2-Gy7Lk|ZE
z>2E3tJ(}3Uv&NN---J~*STmK_;(gdC@x5U|6)h;btZ9W~{G;BaC6)J5bG_%h&cpZD
zL2@Yyy;Qeh0n({si<3e_q(;*_H7@Te0InO3KDlH9zc+-v@bttNWA>@~4Oi5*b}W0c
zoP>#>u!84&ZZgNypI!ivfR$7{kxo0QZXO8itDVs~TYa>Ei%7$qI%AG{S|E5!i&K5K
zUG+ja9gp~O3q&g&U(v2Uki>upu0p<V49);sZMfU*iqftC{&Co+o)=-MLHc*1Td=O4
zcUmH8RRAl$#Nc3<+Sa*lv#fZ63rJ=|O|>^eIdF-TH`Nr}sYNYgY$F#W%g@x3x-)kc
zN;XbF{i5H@rRbaU2}pSFppvW1LU%oqM#}SU+TWg4s-*KU#FJ3I<@!^QbOgVHTo5@r
z04q8+=ZsEO)gPiK4*~z?V!#*GMkvp1u3qJO%=0d6wNU_P*av`-8*s@l)8binla_W{
z1+*>Ce(HxGW~T8@tu>OXc3$JfAl)LWlDf8Di~00f`STwig~<uIo&tWc@(P%A6#*{k
z9cOq5Z0W$x^-(IV`VO3fBJ@u)NX84OF*DM%<LT6*tiyJYrZK)fyrOoxXV)o9&e%Dy
zalc)IaZ6X$WP%>I7}ZZ>a<h=c#_%l{>eU-AmH4=+6TqU8#ZF3hU^D7=byHN_ZA+4>
zPXWiDVvvpI;@(NoD%K3TNVMB9fNj!v$kK-sQ4@$qVRMate7K*I<ahAlPPIKq*fZZ-
z<aFb*?9QKzFkSf4K4OxbtK2zGo>V~^Uyk85*<s*u`BdXVEyLX*o85My!QA|Fh{&(&
zEWn*+54<O{A6kGiWQnIDl(~_N_MmDLdkW>2Df=kK#|l_G`=!pGH>`#*&nT>XR&vDy
zwLmf{uH~H`-t0B>BYjAW=A;RR9YT92BYC~7*pa@21dUd+aZ_zw0R}riE=&}vi~LZ5
zW>w)Ex6|9HiV~*|tt<;Puh(#{f-x-rxDE8quHcPA-+kVSV&=scN^;*GI%8&DrB<z0
z-4~8Xk5gG@do>O)%l@U93DBVu#WlX)C3=9{v+kS$w&A84&SR0xj<U88p6V_p@=$bn
z=#^7;2YHWe0vPr%{Zv9|Ntc9;et}^v`P|+_A{HeQPxsc^IQo!K`mA41VJ+pL0r$HG
z_8O}x16&?27yHm#$uEi#FKFBAQQPS~BzZT^E1%)?f0>Mk$YW~(jJk7{nLGq%G|Vx|
z`0((S6<N7nz}0j*9h2(Ot%r=;dcPezqlI8RJo4am@lQqrIZE{mDGz4@|7^H%h&Ync
zc$?1MhiHi>R||VW#uLpzSqV0P>l%lG10x^Ng|XrC)XRk_=bWK{HN;X{xe~*1&S69O
zaKu^ArlU|cL_7KBjQyJ|!9W=6Uz+NEc>IIQO8r6cp%2KAo~iC7pny+Sp1(I#Gzh^*
zas*QiB(h%_i5=R(6J@O)4CheMetbei2`Q;^QPf4HfaT6>MZrF&2fza2ra+SVD+oi{
z%cxSrR75h|2#Pn$`3H8u7R$OM-#EfjvH?h^9+qo>>l1rz5a9^Vv)Z2Fa49X6bx(vB
z(TYiiGfMk=z`Gaj)=w!-geNu7Hx2q=EQUDOFZmBqzGZ1_Xfyc?BS1YZ#o!q=VKNC*
za~j-vrTOvhG9MR4)jDlU#5+9i@wgTTUAISqzt1~eA$!>rB)1~KvabO2REP4Wy7Thf
zIAG|&7FGlXy@w9hl1}Z{SxxFXL(A$pEgNx$QopI<@E5xr`}}^`2&W;D(y_MtDIMV=
z7?yGBh&10NJZrDyh<>IFaWth5wn*7Gm3u^QNm$PYxsC*a*t)DqLoCt@Eof)RixnRi
zHiu32XoO9&W>*x&_kin9hv?E*LHf<Vw0EN_z&%|7b*3N$=T;8)@F-U{uhw*8uGsP|
z@WXkG<M6TJ8pwv3qxW@HmFGODQIlXe-F07Zvui2iBZK$FDqwu7_k!;mz0e|a$iX24
zqRf@R6{>YZ@OfO2467FQRlPR1ewnX|p_Dwnq`Vs<MHAfs-(S~M#tN42Sqj+Qw{J#v
z66v8kp4b1Lw5=LQE|R)$o~h>N8P9Kg8AXU><x?TL#J%LZhB3K@Nei_?LR$!jdH7WH
z|Kgm{+uXp+Noxn6k>}Oa`J9o1V?Qv$(?*CC;7b?bQ}OzXakdu!sRFPO?FMTLE9_7*
z81qS)bO(gx?cwc`rl;s2Jmi^)BgpMi=6!a6-n8=+_A7e_gU!8ok96I&Z+0&*X_}@f
zgAq9Ve=^R7jS<FKaF5)`a8m`?KR0V?+V}ZknvyeO4>g~XAzD^2v10DA%&*!tXk5d@
zMKr6Cduum{X=@5r`Hbe1jS5>m3G6p&+k?5kyn8Kw1L7?Y@aDX=O*+b(^S*%DGtM6h
z&#Sm6?>y-{E*?5;+D}t<>rJUPmhv|6EuU=hX~#$=z33HO)FZb$#F@0hTXcXo{{V0M
zo9#W_wBwTFnS`8Ha2kmr6vPmuVsJ{k*lCJbn{XvnV~6r2BOC0JTm*>(A5l(B&|JL(
z#_(&3q)V6By+qzB*`)rYJ8&UBmHWx_b}c>SI^oa>4fV7KHov{#1VOd60$9DxEE_Cc
z2fOH29gdX3ewqyX>6DT96q$=1Ov}~{=Uv{9*;LQGNejmm<h2|*R2PdxbRc2G+7;=O
zmdx?@FvE8+5}sf$I(gCH@v+tj)c&jZZ*n7M9|&I#<;x5(!ysK|BtqPTV$Ps~t778@
zE?btDRzCq_Qi7Sc8tDrN^WPypqWhk3rk8*^Hkb`rEsthTzHFmf+1BQ&m{Xq5an?MT
zos%E3mI-+FWbg`@K1^{=&Fr0SJXpp*u#ylLaKaH5@#GLe$r`J}hUxS#4C=_}36<S%
z70NJU%Q<p7^mApT!c-&4+^PJ-hd|o86u{7)5uB&CSN!HUn@<qo8!YJBKQ3s1BjN;3
z{hCO>idTI@H>x1l<otI_thkRvX!QoMS-XPbG0v!Br(o2ouc?fEqbs3>o8VJd12Zrc
z7l_zmc@5R_0s3jm-w}cmjGzm=+2NUK_Osv4LJ=e956D_hiy{ARk60+&x{h7YE9kKJ
z^9{fHP?(ADmnF(I9g{=VWTfLIj=$x)xXUO864~UzW<z){g^t8pWmfAhM#Uq)J5cPe
zBdjp5o*t*4=oc3``XU`?AUcI{SO@yW?8`iNdGTyN5pwk99d#pC;GUpsmJ3doh+75!
z*nGMNl`7E>>`s#qU%9uh;gIX!1$DjW=|@7FR!1N4)d&9$gHv7Oh4yWY&B)>>>B$~G
zW=R&ND-d7(a=NF0595oWP~D__tBr&K#wbk-|Gb(t?Ru96U3>=$9*+GMzj%k>7fvIA
zoIn4eYM3gNcqhR%eWdRmV$zD7Z?o6`t(OtTk;0cJYBbRPj!gEy*$}_`9{-#EVmyFE
zTn-#gTK?T7`p4nLUvDPRAWP?p%>&H;+2{WJCS&(t1&jsZow$Bpng9M-|A3$VzxVTN
z3=Mv<&3v6%WKzg=xUbbfgPFb}ay*B{JWK4Ou!L{I&aUa-Zy|6QRlfp8nRlUNA^X?8
zkz%;?Y7{WpBQDQ@0pR2xaToJ~16mVlz?!9Cd>{`9+HM)8rBbuU$I?_KuQ9prq5g|w
zmil3d()Kxcd6l<#1?fNMXgvP-n664G%Qnyw_O*bVXv|b;;!w|vdpaGm^!9^5z_J|l
z{~c)ZHwYCP$N;_4Ainpf57RU``H&iI*lvQs5z(iXrSD*N_-+%%V0}9$de1`^dlop5
z2z2cyoTwq;T=LTO?3V(QA8Al`c!Q%cbT^IUzS)l+z0yyTRGYICb8Ne94`hbD<|*<z
zsc6*VOn)~p%6JL}EQ-$Az%bYnb-E>h3#bk@&|AmjXuI4X9;pp_XnKDeL$IL9+IV>1
z<B-0hO$)N=Aig~o`(uO3i>6U7erdPj)@enLmG?;eAM=0{P-50qA0d6M(-O2inVn>I
zFv>j(*8C!cK^wj^_3KMhZBXbUR0Kz7sL*(~!pwKpl_>|UC&AAjTWl?vDOnr8V@8C|
zuOuc+C0~6I)Yz_YYCCl0o_4t{<8=MT?i@{00j(kE20ITwckMNe+r3v#=EX-ZJzb@$
zWB0wm{yG_~d<54e-3PGVKpnUf#ULd3t?sTW%kEkkEak1Uw}E}8HW)wBgAmHDpS_jm
z_;Q}M<-}l*_@AECn(qL)tf(+ye_7|fivI^vS@=UKR4n2uHbQ-c^#d(#vV8ouOXr4-
zTv#g*aNYs$J0h%FJ^9u2VQm<R*gX9Na0P42%r5zmttsm}Kz|q$TMa)4Q{QSCh+jRQ
zJse#&m^K=5=5RJi%<52JMLn-?)Q8`5f1r&Cgzq{*7IE^|BQlIygl;>>;<@uf=FM6~
zvkdL4)PVS&=f9dZBq1hsOTqwlysfz4W3_|dEkr5{0f)})-gx%?GkjvRGNZ-Mo$c2j
z{2wnO4#U8wZZE^~&uHWW^s2AGEWV+@_GNUIdIa@2JfH7e?m}F|$pw?fn;OA4Ma+tF
ztVgRizg_X<D!PI{bgmjigH=lvTZs2v^cK3WTO!TqoP=y?FyEk3%M#gPEZGuKO~qKS
zKjDc-L+@D19Jq2#tOR;!uYuUesD4NpP{Q$*>hD1oCOsrfHm2gfCEK|BH}+Q_tG(|+
zv&<8!A;>h~*yxkeM=38<Lq8fpH?84i?-{#M(nv{VZ^_ZB<anc{Tit(^a;v`ZW0O){
zn@hJ}BG-irLD~U2R<!w;8E%Qe*HXTl6<fkz=TUsC-dhx0FKAD5uJ2a-__6(R#vtzO
z<GGZv#-^J--Lr(M)EvLNg{=MgVbz@zLu>i{1)f{;iWl8H$4*;`i+z&z>Yei{s6RzH
zx}*Hl#w#gRWmhk|(7?HXJquhb*E~9|#*Yr2<~rBn=~Ya@VbZbm^7+g<-p=^jGNt_+
zE-drwPxWt?7*|ev1+_ORwkn1H+Zp?hnD{?HhU!BArAdKF;fb``T1qZgVAwc-3vnvv
za;(BQ3$p#ZhaOKF@?)jJ6<~MUMI^8ejNf0f`}Wv41#Q9{qz&A&91)w=0o>^BLeEj{
z1t`(>qMZ7Q-JYbKnfmbLAukE=!GbD)1#v9l+y9lB*xNVP0ne30YTB=Z)zxaru38MG
zsKeZsdl0CF@^rCla!Pm3$@lERtHW6k35nDkS+LNZ#Ig+UJr3HMe3IANnk;Sb(C}WX
z8YRc)R(*Cj8-7s19wlAjfK4zlHejUI%L_DHPSq9u!4rnWm5z|@qZ2iNn7LX7&TD}?
z1U+LKI}B{c@eGENkYtBo^*VOL-440_=rNG6F^?lgHMO{*@HIM#=%f{>8oQ@JZD-v?
z(LwSZF<0hNIQt6i4QSMD;C5ANf9VGNc)+#4*r1+W!m`A;sW1<99F9aU#)WgWEZ;+S
z1LdxG+2)_U#3&llK<&3XP^7QsW7vN4y{Ph1*kpHeU44F4VF|X!FPQYaz)W@aIA}`$
z;-7)`GMaMuDG#oW^D=hY(foeT0+8A4!7|}~P3hGjQtvO&TT3NrpmiZ;wEoO}&7hpC
zuw%Py3>4rXGGW*FB|YRMg|UWDVie!_0@$0~MnAAUZ|U1>?*KMd1_|?uRHCaI)fj5l
z<r4gs56mB4&YE}@T<Gv=vRp2=oF<GaMN2LNGxq*e8De3%7({qZP(m$GyHL!1g7wjE
z3BuK~*?TAnlegaAAe6BmJKw-S5jf~S!4ZCK5tlf`Ap@fIPb*;qp*=$LoPJbuJkuLV
zcaWKDDOB=HmHCi=G+E^h0XPEV=adI+a#jB1vA2V&0ok}#&7TSUtkp53Xu758-lH#f
zfP+s4^(}5rq}^$Q)TCDM6|&_%6LyZQ?vbFu?IZ|*Xa>nPr-&Rv@w%~2HB5sKML&ZY
z<VcIgg&pzCcrEa$dy=AYX3PLBpo%B$P(@3sCGGG`u?~AP40CmZvYm9==vAa5a0N0o
zq6wNZN<Q#X_1ccNyqfapA+=4)T80KGw0L^n%gp>b3;}W5lL{v`=%7tG!0(?zj2n-C
z^cl!o$KNuN69z5-bjXHR$LEu7Md+Y|j111ug}%uPUV74bDzw{XMpXkWz6|MuRaMEi
zfu^!6Ft!{<#-JN3iarbjI5%$nuyZ>{ZX%IDn^V@`O-Y);>fu<-*U3izhC=S<+KL)>
z5^sW44Xxe$bYTZuYX~8Y7RXOzp}g<RPMo$!Ja4=o0~hm<M1dT)YXs#H!VYWP@XkUL
z{8RAfmxrhB{d01;qz=!AK$QM+2(N^GXk!J>!$27m&l$~X?$5OcDXBuaqqo51!%i~R
zXR`CH+!^V*b6^a2M`XKxGa?^|kg=Y3QfC+^ad^z8+N+iB>TR<L3i1@N3q8@#9aDEP
zqJ}y^V4Q{g6naONphr#c1sK*$^*lfKR=t}7TWuq*W*)H!K7l7!=<}GL1K8B1YQ6*>
zXphYqNR?Yhj}-{v&myGvBt<a!D_thDeLB)8mor4?!a7Hds))~TLV^;?9P%UWk9g)U
zx~L>>i8CvX;NooK@LyKpd6om3P(i5U36I*1U8H`~cakJ}s`FLU$f-lQm-$qTf`CtV
zb=D7kVh$vKS%8JD4>XqVlpEPP;Lys#D6|o^om=tk$O7C@k>k^hh&+h;PF~3f^}VSi
zcU;qG&}HQTNc5$c3waP&Rli8G8TngA4VVCPTakX_BWL(o+<?Br1Af3M%x?4fNk=V@
zR3?OgbUQL|zSvQV{J|RU7mZ!UiqN9U=xOPn(fVw!<$BH6NK@i?gG4dVE}T{>SdkoV
zMzth)(auuT*P;;K*aF_;g(DyKXK#am;q1%3@r-H-jC5{P<=3e*EjPa!g}jr~A|!b=
zeW#z+{A{cB#qp)|QBxj|JtT3A1veSrYFm7_w87Nv^zyZ@<E-x+?yf=!KSkZ$JfArh
zb?vAd8k5aP63>`-R@c5pT~m)~<z@YYxeDFK*G3~y$8%5)IfUFCl+=3*Oni1bRjUu?
zx!@;e&#Qgh=`5n0-xoR8g8*`M35EtY2X#)yT2tmwmdUvLyZRvUhJ{;<9l=h1n&<3a
z*56;&)ja65a=H;LeV@k=Y@VWAPZx7)VI2^b*NPS~@q$GA6?G0ejIsshw7r{1BS`?y
zGCk(aNz4a<4~(@5FO~6IjXwxh-iFwWEJh3_i;*s=jP7AB$DT&zFa=RgQU)I>d94a(
zDdtM50FL#2=_>BNL;guVv)wv9m4YHI2iMWhuO1x8IN#Aw;S8#TdnzHp7B3{3HNUuM
znLBH-C;4)XN~9>e0Tw$0oK-ty-1*=DuUD5(>>9Tf>y;o%+qqjRyYhM2ig>l3^Ewps
zig(G)2fqwBj;$3`B4b{|MP@4DS^r1f_P_r#o~mYQ$G<O_9ciS&oy<l`x|P6_*)l5<
z<WXT*LzAn=a1V<z_^Sz_?+=Y$#}_;wVSROaopY4_FSoI(89oMeAyhH96?#tpFT1f3
zMuI5!tx$8mzwO4<X$heXdUxd&JyYlkG+7ksIdm1M_{hCR7<={&-#luuKA4GrdQ9p%
z?&i9+gT7-@2mo*|M%G##=<93@RWG1yQsotD@8e%AJDj4beLTf&Mi3aFdy1>k7r6s@
z=T#rs7~-z=3C#~Nt-g|OKUl9`ftr8(WMV&<j<PXksJq^ZJN9Zc{>hLL_TI@k&aY#a
zODRq6TVKx~5&3u>T@(C5mM#EA{s8S4)m`zxYoyYsDF9-}q5=t}`H@-C$pY+8z?i@(
z-O^LYrmNKYl+6m8&}z#~d-fngpEmoP$qPqyNpVh-t+EIHBcIH=3kCzu!)T=JzIrQj
zLW5}T&1-LuSFHBwFVHQ`f7(%JO1Q&g+a@<Q%5iI^@MP>CAL#%n)$Kc?1%7k^si`Q{
zN8D*7@7MX${N9s0p{L2Iwms-lGPy5?OB&aiK2ALkJMV4E4pri}#C<QpaI8f1Mh|UA
zWDPkl1@5(ocmyU-XLUpdV`#iV1X7i6c1oAAGM8<Xq%ZlBLJay<)QL^aec3Y_f$vH@
zM?0rqbrXXrY#!pn#EbV+wKxj|`j{#Rtp<!IJ6``L@Hs118moQrJ-FI0h^*FI#=m*P
z#JR<f&ATaCu?#~|<G533yp3C(i>+}PRPAQM{ArFKq<Qep=6UBBT$pDz{CCqhOSlH}
z#2wBJs|K@8EV>>(HNnrB1a}lqq+c11-~WEXJ!Iof>=QiN(3v8A|1_7C)FeU}w=bZy
z9n`9;it#}b+a9bZi`V6NSNqn@${G?AY4oZm_#|ApHv6dF|9y*&e6g&a<sTdNM~B%4
zR#MseqJ)kvRN8yQu1F7r0^X-ndGo^VTCVYMe3D_a#4In@T>ePP>y7{Y#QcwUh7sfN
z6jqm?G)6TvLF3U5PZDg*X|T$i{6rhZfO`6o(pH>6aYUUMnP1h*kN==`ux`YZ<!uuv
zD;~*dTr;!02pJ}3cMIV(ElOWQ>){Ldo@#t=Lm^MdO^|NvludjUdH4Xq$h0y^fci6z
zG@e@118p#Nof_CGYMe=sB%h5Wm58u-Ptgp#fW?|o>_ER6N`l>oxS<|0GNDTuL_6=T
zD*zv_$E#w@$9p^XeV%^>=F3a@iOoAmv8WAGhRHKlwypW!9oD(AqIG$(Y_cNHS(lDo
zgoJu$<>etzMws58V3aPeB)8gEXn;Hkcm947A~FXT;+c!-f@knfe6cH!`lUU^#34-2
z?NB85rVmx!X&j}d@SrO__U-bQw|sUcKqTv;yYsQ)j<lm;&ttJa$!__L!w}?AttW8L
zkAtm2mzy??RMQ33O_Ro@DTXcsgAk2~<1l<Z8jrzi0o6xwJjpKlC7I_g{4S6_s1?}P
z6O@PAXw$A0{!N{d`-?g?p&^E8j4h(wlIQzt;U9=t@aZT8$EKTFay++A-*b&6vQXH3
z0Q=%+`%;c)8Iy+)E9AXC>o5k}>_Sts**xbti%+Gu!EP*m%_wF9f?+&yWCesSr%=Z>
zfE4AzG^~UPE(J%o$$5$0+PhdIu$+h!`3vP%J$$_Oxj|Ru;L$@!Dd4WYi#pEj9D{8h
zRGK}5z-pNs+r2=xW>Y5<d2%B7J@2%Y1~7Vr%aj&g3*BF=znu`dL}xRB#6G<`tm7CX
zJ5bvD;D+I$*Wmwg$j*d%`@qPs!Ji<V1>~wOXy}mrJqQIwu?cuIbnxK0>-5_1$V;I4
zuvc#4hw`AgV&ajO7C`OciBUb*1q*kcx{gmBKM&0WV|UJF2JPO?{QO7eDdZubPH`*{
zL)X!ZRY4JF+hSX_7NNHM<O^OC#5s%Qr(LbmP}>ljm$v|u`P|T$Av1dKJ|h+ijoRI5
zcH#7R4fJ;#_MvQPuIy_~8DJT9-b(EYn}ALA%OnZwD$9B@yJ<j@?1rTntVl4kYjh#2
z;SoZI@jKe(Oo%I7rZ&slkLFCUZ>hG2+AveH!ZzQhdJ7*SUIC_<7^w<B=<HtVt_3rL
z$;K9N_18Zcel{T{s@5|6=}ILcS1PgoNXCyVML}v3L5)<5(Rv4>@Z91hkS|t9zsYg&
zOPYe+;e*zwGeFBaJG>+!tEL&6ABh<1zDcy2)5*5~<O5HAy4ID1ofQ)H^scGyV0`x8
z<MFJ*K*rz9sDWvn37cuuMPnA8YYcfc?fSME01O!_UFTs3M!^gM=2v}d7G)xRp&#>>
z>0rJ0C-A5_nWKaDUIfyL&g2sY2c?BpY2DM|#(NgT%yJaKT^Z4X@ZY>T3U=BUCiO1;
zFdc%WG}SgjcSqzF8%y2Rj?<<<lruvm6d1?K)A(bLm=YsPDdwfSv;{F)^)cUqoj~&?
z@}?me5nxSGjt%W&2&)FK|7?~GP_8UL;Y?z7xVIro@0$l<;+GeL)Vn1$19}2B8#(?q
zv8v`C*Eb+e_oS+Y#|*zMa2JOTc%mKbl#brfDT@^WTPu!Rol;i+fR6tsl-{5Y{hPVu
zGtAHa?Ih{y@F3N;>C2-;P8kS2<2PXh1zEgk95IzBCJS;~nNZSZ)i+^pGSE`0pTn>x
z<Ud&8dp<!2_uSB#)lx$;L8SRxF-fg8sDYQfAuB%np$e93$pf!C1tPBZZ!(cIhZoKx
zAD22fc(Ov<27@=;(&uKa(%*#`5!vw*_^0TC2Y<7m(3HbR!`FB4*rJbOsb|znP)usC
zm)>&J_k_xwR+_`8b9*d}4`XF{IwIU7h$vF-U$#~b2sdq|cy0~4?8oYYF^6G@KLg&%
zK`s8$k-PV1iAhl&-MHkH%Z18_LO-EXS~N~CW8g*N0^=hYzz7|7oEzWn4Ryrnn$})p
z&L^27wVQR@JCQ_#UI5~z$;4M+ZlxS+RBEP@M6`fK{!s_c(Tuc&(;gBwCQSB4dx?SI
zl@FpEaPR7|4HfgTQFh^2ynf@y<#M`?MDxAC-g!C~9O)}r{2y$k+lGfYp(`y0qF-Mk
zBGicrfJUz<lEXKNS_`u-FP!F{ec!da)J6Jva8f0OPsXzgTSU(KXj&8b6N<(XlL31=
z{9KBljbs{)&Yv@1=y^e`T?3|4R+weZQJjaqL66`3dC(4@=?e~hhTb)zWDbV)<!}Q|
z610!rNF)RLv(Rjt_eAzmMYm@~eZ|cD-9HvnD)1z?(48p$InndyyTrBK{q)pBbspS_
z@DmDos=LFgfp_)f(GC?N-!LkkM$-blUP_*K1_(_r_KS#^N}NRts&WIJGp~1`GN2y-
z_Qyo>bOFL6?gTHlGC2>;FJ^=@;$>fW%K;x_EK1vt>IYmttwRZflQW=#ulPVe0m(UE
z-fHahfdPey(P7Y^S49ksfx%S)q#7q`DU<ICrV10n9arc2MCoHA?ds`&5b>dNB0XFb
zwT?<VS?=q$yPqb{wVeXh`GyO107y-AhPl>28ZQQS(zpmeh2Q&JT>v3o*4W>)ruqd=
z7S%VPmd>RdCu;BXq-vzU5A%~=ME!Tt5+rg_>&b?<<=h^-rJoK&tCyDGs}Lk?f=k_=
z_v3*Sx|E?J5SMV!#89bQB71%xzvczQO9<>GNR@l{FFPnRwOdngvxp6m*WLz?sGks@
zo(I$Z<A$he(7UgbYv2?stn2E!@~o{V<dR}Il`PgY^uHRQvxb9#m?SnGdRSG`W5F4o
z&&f4<c1XXRc*-{ApB8%<44$V?nQxpwub+ZdaEc=rJ3*0VR?9HO<AXJ(*bpTTHK&9C
zK|@<B+hE(20hmLGY^o;%qom1?9iXm<qtQVO0VR5%r_s&1sh_gob$Vh}!DN}OX`hH*
zH}jlS&XB|j#gV|cL$qj?`uF<>?jQ-E(K%9yND9C`iKHh<5^7XwrBaT@%mMNJbKIN}
zO-h|jiqt8lM;WHo`p?fb1&1n(=YEp<vNho*(X80oPuuZFB>x+NPi+?1xBRFwuQ8Jn
zImy*X8q8<O=G6X0PO8C^hhBa{=9xf42fUU%3tU9ls~zdexVwEn{H)5^@H(>!r~qm;
zvtQ&Sbyj>fo<KKd!fY#oe~AC`T0)_gYqs|N+h0yrC)~A>2s3J*Gm+L|hsC>_5L!}v
z!oO%qN>EpcHGnhl(#7TJY#yhr$+C)<SC~chBqP40!@8La+FhKw8)p^vs7XTWIT^1O
z&Q2^gw0sc6#+vc~(B4ksE1@&lM}C$?F;Ftg;I_=$Qc}gsyI4m=xAkSdD13!kKv#GP
zmF3V=@jFE}^%Mbs=eShY(2<(OMBT}F1rT#0b*W_l`q-fjkA}OxROzgKc}Dk@#rv%w
z3^yteR}mv>Q&lG|fU`M$#<0GgnT4Bf;N{Rv%Mw*nC%PXJUOxEjh-aNtt)P<dK#9_H
zTs}vW^71krSBeRNEz%(3ASbZfYDxhr)y*kr{KQG648@z7#Qo$(vN7)BhQ1okE6<i1
zrYztvG<y`|62q|E0aQK(-I%Lo;%b{Q&xm4^2I%mzO46-)J(I3d7l~#o1|hv}F27x(
zt1Ta54-RI4+9sZJAda|AATIP=DL7GU$2X9XouYNGZ5HLZucFD;%I374)${Y8EV?W1
zI4(?d_`SRli(-!PkMv4b7A}Nr(+6eQ%q9ai*>3ZHatqSoRsl}Y$$7TA|Kt=M;7g~Q
zDZpQoMO<ky@t6K46(#QDQ(6C;RFrlJGiR@@iPLV_hxk}Y9Mi@=V4df5&^%_Z$#38H
z!rfhS*Et9napN12*e@{<(F#$zNsp$KYEMvOrda2_&Ngf$EFUJ<rQpKsXQA@>vOncQ
z>_k+4duxi4;y!C~v^F{QUoKWPQvdE^H3c#OgjCdUMVp-chmBR$+yAh!ng(a(Ho}eu
z9UY_hscLXf!W8(__w&{94VHQKgnI6gFszz+=nzdX+PECpv(<Ujh;@EX=1a+knmOrq
z61MaUFofMZe=u{D--^~J_CYc8h5c#5S8=p5Pfp|=-#82-R7v+jgD;7Mqdr*4IV~Bd
zSjE|cd?U&;aW6RXne$?!#OB4EG>sp34+QYmNsTy}EY#9`yw<Qcu|1%Y@D&_TfIW6d
zoR`$h*PMZu;)8;y)h5073G#>94Ys;Q+JkQfAzrhWFrmYjQV~O_-Q~(al7^W}qB?b5
zWc+LTwMMdC`EmmMUfM1N@?w@HZwN1nG!>@fduSm48+)A%QzX#5dNs0qZMJ;wHk`Sv
z^Kwp_yPO*AGG_0Oj2TP_Eyt$vTS>P8P5;3Z-PD^ql8rqZn$weQhKD+Q?v-n&Ps^^I
z`MMqvq>g>bHviF4^VQ@tvN)4z!jw|<Qa?yPKFt+0m499~#cY)2<iz48MwCQIFYu8X
zpYHFeZ_smT^rIZ-9hgt19|+;qmf@{xKi2wrowlV%*!_3hj=##c25s2YdW@g({4BIX
zEvW#t@N$b=+P+^Y9a`<&{{6dRCjE7rhEyHl-WvijL9b2m1v2}1Th8jn$lN-st6sm$
z?kuY_0Ta#?Af?-tAgr%`;8m53baBV`t&1IV-rr#s(7rVg&{4V)FgLU0kFA<jxWAxq
ze{9V^M0uk1?uj{hs%tgh)OLiaEQyPBTY6vJ&7v;26k8m)tIANDs-WOvebL*GxY4WW
zHXbxgn`UHT2Wzcm7q15W^K|Xs)F+P+-qDAN3a?{7wru00P-8sF{%Y%}o9c`@*K>5?
z$k$4D=U0wzuRS(H{IP6EfjP>wyTD~93{p<}gKKGPxEl}WHd}7xrT{b=Dt5(xn!oa2
zGk`1V%K5M9^doK(yqzXa2k!_2oeacZz~V5W4Q6%6@`YTZ$#qhhsJJU+>PFY`#9lzv
zmmY3DB_V|Y`n?DfI$vaZb6tYpE8o?QgYT?_`WIfy6*!{L^WcKMT*O}ymrHQByBa?8
z$NSvL3DjC=_x2vUA2<3D4Jm50r_|dKN#}_Y<o>phc93WLj+EWyBA{D{H#V#NUGaA^
zlc4lEunYRg;&wfQ3m#UuTn$8sS2dwzQ!X{^yMH3u|Je=vt6fn2fY3yov%)xjq{oDg
z#nr%mVh=c;0-3dUo2)&|2^D@BWQ96&6G2`03gVJ@(Rq<a^Q<Z#0C{!THRTVMlL#e6
z1a7e2PCn5;1l9c8BKeP16#5OZP^$<$sraW~^<RGN1EdK$Ci#s1=i~I>KIy-_FDwDK
zsktTb`JaF5KYoM%7cYQn2E9i+)i%&hY(uX0C6iz6ZaGNLJ@nL`TeVz-?AY;S>QIPx
zdOdm4CW6s4JFh?_JJC~yNu~JrTys@Y6Id%(((AB4Z&m1Rz8xmDNG{O(IcV`FKAU-K
z_l=t8p9yWfBX3l8;VUmlDq=Y<mV5%|fr7{6VG*%>zLzRgNipQ-PGF4b&jx`GK~Qqp
zOf?5^4e1bz7uCaC@9AV-FH8W-Q~p+Ez{>Imv(L;OCNV-u&YtYG+Y3p5=!zz<0)1^s
zE`y!!$9j|@CBeN8Q%waS=G8ar-){$w-8=!VR6O4xM%?1^%Eu)bSRDb!JJBu)f&ju{
zI-=_oX9reV2V8q3AxK5<Gn}(=Gw+b+53JsOIlL8s;N%3>=B{f;tc5SI-<)Ocf8s`3
z^yI=sUdZ?DXXoCgw{CYsd{QYCv+vy=P`l^_3^t;{RKljCP?#q`0kexs*)q%M{&|B3
z^XzUpo(VTh1f~qv4NYEGJ=_2D(*EOW4{b)==)7`z>3{sV&i$cq3ZI0yv3(F)=3FEt
zZdG9pItc7ZU=}Rd4A(^<MqO*bcF2U0JL6a81`N6jSo}s%_E?+Vaf%xSWuRp3AOXYo
z4QfT|AVe5q5BK6g$adgY_{q9<!c7Jdu^H@ch3)GCs^v%g&clc`=t(}5lFR*R8kD7z
zkDt8z^&@$XMPA~tVh~L}UpmGF$&b%ulN-H@3K~sx9mbqRulhxMEnQvu{mv^R4$m`N
zqROeCxg=%WJ3{*bTroIvVn10KXh#e#nS)jhdrEES00h(Mite&|=Lz|TJDDxd4A9X)
z{vntnyaO|!#|)c{im}cG8p|1FRBv+M-Qz-S=dYZGw)E?9OS|U+pI%5=5(KMPM`--P
z(VNX}ap^#;Hd#{suvf)V{%LymA79`f?*s|`BYHS0&0cHf@BMjHwva=fpePtZoSfYm
zLcH7w1K-DBOzCpsgUG6=FFOJa8~gu7Rq|(3FKzCyI)4Q|@nif4Z9|2Dea?2Drv-ky
zhpKA8<L5#^=lwLQyNOXH$m7@Q=cq+<xS{=shfurPcF&LSD774BwWeJY%`Hj4d6lgl
z3RC`Vx+UIW&dx#T+FxyiZG8bcN)!;4<VDhq9!cB{td5`S|F%^l(N`0+`u}nE9ne_*
z{r@F<-eqQQLKz{Oj6{+>BC|rs%HDg06e3$>7Lrjm*|TiQ%m`)g^?zUA^*q1le7`;C
z`JdD2aG&D7?(1`XKJWMI{TecoDzADq@>J@hI?7iL<_^dQB!pfbj@i^&8{2|CBdzea
zCo=Y7Ce%i%N*wp7Yb5KPaA(QM4);FrNHE*BwlEIg|7<$pRFo3be6|X!z%Ejf`AfK1
zC$CjcRNFk(7!K{rq?5+#>+v^wPPY)SOb~f{KVe$j_Utrcs_ChTG&$&1p`-c`=_HPI
zxF6u<!*GnNpgOZUgNfDezhTev?(DMU=+(sdXB8w>Yy_2x;Vk=sk~YK<*4lx%$GwIV
zHnovrrjl32S=YO6*w$n$w-3?WyHz7@7|gxYb7WIg*eY{}V8r~wfl~#|gcd{3<4GM4
z*VJIQPiB%e>||aAQxl32OhQhZ#GW5L_G?U)T3g~*=QsRU1d6EiH@+NFyAocll)jUq
z?$SBs@MZpBdg<DUt+zy6{qbbYB<)P$6uILh<*~=Q=cN|WKpX$s{}`|Q$G4?O-7}gZ
zxeb(3n0FfqGp~YiTpJ8>e>85hCR#BjT9+j}rW9xwTZTs13E7ZFz|rb1SYPo05{lRL
zp-XVA73d2rdRE19gvB;Bd)FYNkoA_>T0j+`xa+VJv?5kczP6@FoSaG#P#JpmiXwRY
zy>}bwmKzTMpkjqSGS)Owm^It$F5(sSc0fySR=-h5Vt4XP#{|e(W?Pw}V!rofcYJGm
zHVeGy5*0AVuSmB7fW#i00r1*M_M>Y^($OSpxDBd?8&ta?(j`)_^Z^Udt+JMD4u0|(
z)t!PQSUJLQB;I9zA1Dj$Fc-^&;KjF}u2IcCE(hilAHbb{Ek3x4=UYIjX|zGR44Qdr
zqa$1p=eXBT06pQMpe+z^Isls^N`RUxl(uZ#&2q$X=nlx|m7`;|ZeIwN(@4&mi{&F+
z1)uf#*KIIEP<U9gcRvAKwUU|o7obMs0&Bd`K-j_8TC61w@{P6tsNz+%swm8Q6J;ch
z&FR@f<2?N05Lgx$z~SaC5}gR6pDS?rQER_NUX&M<4&EMojxam(eC}lQT>1LwSy~Uu
z*8?_AKgUpku=m48{seE)nGzD!MSSXhvm1+pc|i=|jyENacTII6zVF^l#Fa7u!e~o?
zsZI4{X>^$4i3{uk^7UcY4y<pc4J=)Q5T7P5#0;io@Y$xc>6j)ZSA81g4|pgISA{VP
za}fN7zKU;o8ZPSHydpvnD<cwFNUL(cpSv^z28QQ`-`=|O;ZZd3<VI5Odm+{Nd!Tdj
zF{%#l!mQ&WB&R9AQrb<6@uXU!On<^lo4Q&>CGPfUeC;LI_H%Lrm->e?3r-;|s~*>A
zQwChNjS4A59_18(5kAJsSS+P{FRh|yA(Gem#ck;|Z8Dr`UzP!(Nx?DlEe+mDOLiJn
z@Y+b67(T{WGt}P)N)#vJ<-O(!#$eW6QzMbiexTj?G+sMBu27UcQ{B#1MLsF~xcUGl
z0MU{sdp+Ap22ki)pNLI|F!8y&2v1z_@ZSA(fub^lzIn2F|1XI~4g(q<8J8_bq#XyP
z{o>c=*KgZ7lmoGT0>tbL7*5#?7@PHcmp9M$!^vi(&Yor{hxy_M@J3@a%iRAomg`fW
zYuf@HjRE02N3TQT!w<Q$C(CA%4jez*-mmy#(b{rs+9By45;@N)MVnoBR{*@~es3u0
zCQ53gtNSkMIDesi|Fs|bP`);Iw%^5Uy4J6hmIcOp9CvNu0=?}Q!`gSjgbiEUN?DFq
z^|`}(|I5fqz($k9hL>+)=t2NxqyvWuKbg^MX~iA**>hhiAxnN(NMqT=I`2A9@acG?
zahQAbB^_+sBM~L+&yo%(*w#EWw%RD2Da@^?Z(fB*w*mqJUqs}#_E-$x-vMUSs1<Hm
zDP+}Ld%L=oxu(}Hq|rMKAr#_NGKqrL1t%PXk2r!HIhNs5bKqDm-a13&pWNuAsX;t4
zt2Pu|xD4KO_C+kUwDoJiGj}w?O;lsZuLL#F>spS7jvSMYfR(1ye$Cx&5eM{^a0#^}
z8_4l+vd*_mx!9n%Y|VMhx?FfmI`%#H4}TySPK>J7=d1h1si$X%!h%HH@#eMktOc;B
z)Qe*pMZ#{}h~u*3h~p*WSa9I@_ST(;b=fqYu0WU6yje3^(d)OzB7fjb(YCVW(I)1+
z#qr@@!hC#Kc2ZciwYX7K(~*udhq*wa%Gt`-{yB=z3Zf93GUF(pE@N-VElp0wt+Ici
z+Ec15XHtf#s@3E`FbYRQ8SZB*oAsVsVmr8=LCJuzCU_F{s4kQ5Hj2=I2b(Bl5=Gbm
zXqJpi;m}O+Ts7A&a}ClA8;L|o@kFX5eG=V9OI@34MqcY($!>T!1csa0C%bhzlSgt*
z%f)4Z0zqb2+YdYUU?;!t&b@tS+0~`nbfvjcNn`(qjd!9ep{e&ru}K>IgDc&|TC0TU
zqAl-n`7aN|Y$P-GVH5p6wEL&OAG|nq94q=S1`NTu-GM@{d(te}39H7G-$9u6Rt!6y
zUc;XPO+ox<1Zx#I=A*8wlAzIgv^da(gs|3IiE0%Ju~Pcz&>Q-p@cly6h(DU{N<~GD
z?^3|7nrfyA0hLD`xjR(E-Z{Z#ofU6!1C_%_IUwJ1QYHp#N{$C|7a<|E9I9W@t<1PS
zNkgsb@+SN3dtpZxov2l}^0Cr0U^MF^9g}5H^fnJ2Z%g7XnraAV#xm<t5XmiEnqW7!
zcdkAvGUCU&M7=iH0!C)2<~(xVj$oZL4wPoK3WHdSK{+T!wXLEArdI#!M1mNV^0gTs
z?NMWSLz2X|^VqtnUqBURw`lii(9`9OJuX|PN|a{wwBZ)^14iS=SE~b>%h2la5P~49
zSplj+N-Sh)mh3Q4beO^^HsGGXm}a`4q$Jkx5br$7W50B<aO(Y>wWy+bguMb7S~-ew
z5BR^b8BF0uw$#Wccs;ZsYT)ToqBK(B&%gb{Wa$)Mkt`jORe|Me%>#Z`)tH@lH=DK@
z^2SXTug_FBhY!}HO}AZ8fA#GVo;%COC=?d3e}jdGcu$V$ZXn($!Y92_)x4u#7=w$k
z#r4kT#BM;4s5vpM^ulXe#$>msQm3^V{~g*bu@`+&P^*69Z@~X`+wcN$O1-H}B?>m@
zzo9YRQ*gjDHtK2c{FcK=$#MA^!4$fL7nXSI-An8>UuOUG0tg}mVq)qo%I3i;%>9j=
ziFYvi@F06mALM6*N7o_n&RoA($%~iqj0-`)0@k!o(J+sCI<1t@sd{uVqH9_99!cEy
zsZ~9t4buHsW426i9;-KUGI8=*#LE)L)s<*usin`X`boNA;;|BhS}xIZJ@(HQap~h=
zCM}VmFY=9FfbJyZ^=%ou+vf{<*y~e+iUp0sf@UATJ|l=%&-js(lJ82H{tI$3tXgez
zb}Q3-V-C|q#Cx+#kF4t9{()cMd{?3w+Vv0$&IHP_H(~d5yPPz`KvA2EL^4LJOF(Z_
zlF!ad^6>fx34Z*70ZB0IP&z<4OeK(!J#QSdf>@i&v+FdO#Wv)6!Q_Y5kn^V7b7CYp
zl|_g)Ev;Vc8Io5XFY?)?aIJ?2d)HXfPNDJTm>0>``02ur2A$;2l45xe;_q&dCE9p5
zAGv5ZmYgziD438|hjJg+I&t=&p5otM&r5_(c0gQmndXnK;N!WvV7mZ%e6|2|0=&=k
z?4iVmQ<P%Yu5Mx)(@QU1RVjb;U9H2PjWPhU^Z94gA;YqPL@~{1aN23QuBq7EbdATY
zr;Ly4*x2_&^665Dr;aWuCTT`SZ8SY*deak#`@*)C{Dy2OeZ!W_vQJS+tYTnfBi=FM
zC0Fo<;Um-Bd0HlaoOJOSbZ>Y78=gv^3fzk2zM4kGvg<!2>T-b^ViWvZay`6JO6ifc
z5Z<`=Lh%uSVAPZB)rB2k=<8fJ3!n?i=D;TSBvB>VoCTAHk>)<gFWe}?R^{8K>BqU2
zBiTNtrSEAmdnM=&i}4@#YMbtP(Qy&xrRvSkNsJZ+L?LBGsaQ?*h@adDUy6bYA!9mS
z5pjSkH*?k>cL|wI7T@`3fR+;Y%H#m0X~XLX`}0hyF?Q~linKdQ(bUTy=cM9PLz}Hv
z5q{)aO@Q-L@JkBhzLzYsigf^Hmu7F*sYEWch|0-WuZ#S;C9_75I>EWn>OYiKzSL+I
z9uzGtRxWe<Ust@$#(&V;`e2ElVDXYmsA>O3nF?eMkdoW23{r!~CPabfK6V257qwy<
z_kp{PLs;C*j0f*b>-XN1dn6p@(stZ&6p~&Z`i==6E_Hsx>1qbExHR&&{c1CpOgk=?
zzbkrj-iQ|qs55UzG<0NfPjgtk26vaaPsElIKZ&N;rlKD#^rrK^X6^{0Cn_rj^2G-b
zHpQE_t`M)@4wMjN-jyMmJbP|R&yM@2e+z#Ae(>Z25CE}0P#Wx;UU~EG4$Gg<jPdsk
zmSM%1N$%HMr|pJdSs34`docl}UssL)w0&)P;|+6$3lW}`!y;9a+^M!SCe5?p{&SR<
zz6n8b<J%Xp<30UOogY7aw!QobdQY7^t?S+K46q%eyG4>%_4Q_wQ1P&y?(cmNv_gOT
zC#@TYzkYXfhWcOUTg2I3qRzLVZ-9>AnpRIGAWjHv9l^d6Jj8x#;xkiz)&}}YOq=!!
z|6737+d<01!SUe<_`RxV6fx`<>-D=oU1+)gDQa#VdN19C&WLpX8qb5UpeYi9RPqmM
z!wAyz+8ZpYq{pdvB0*{24naH&2naI%wj#3l!LAC`t(pcffP998SM!V6*O)lntUle&
z?Y(ycX*`x|PmXnG$uH@%gV_klkq|e4THPPC78Q7BLM&yPFf&qezqNA~KIVMd<_%(n
zOmO_knDV#>{a(j6c^Y#a!+dRV>2~M_&uKCbR7tm_n<%N?dY(ahd>2Zs?#-*!lcBhC
zvf_uKZjSBW3_-ni{}y8Z%}Ru&Fccg%g6<7X6P&<P#KRZnbTPck2oJXq7jf{HGquuu
zv{bTjoQI`|MA5IuUD(bDOOvg>*9NjZ^rPOQw6`z=r;;IU89w$Y^@ez`aJOB{BF|+u
z2tSH3r%?K|n%jK-AlU8TnL)K%XplY`DINJ(*V7K|&u=~(!Sh=RV9nULGgQ!I{B$}|
z;7jw=!(Gzqw53MicU{L<r4ky$EF6dDws<&)8$}%#nOhsZeCN9Wpg_0IWz%!-&Yk4T
zr&DFKSRQK{OqVvR-!R1R-sXy09D)w=A(!SON=IR~!eNPT8foX7`lFlOBfNvAcGCO_
z!xq(8Z^^~l!?-ovjf|AQ(Q2<qo^-7Zh@G>bfN7115LKzqsWyY~Q8-@)B6c2sdMV!$
z7y3aZ(m5>0DPI0p3Tb8)e{}lYe}ypSh@-7;#xiGei=Ah<OVe;(8Rqw7r^M_v8yXGa
zim9df=WeX#Wf$rZrt4BEHo3@43rV;*s7;y54BQqnjv)vRY>Jp&ZC39Tniet4!Mglq
zMf@=LI@6mEN)pnneoe5@lH0;XeX0Y+hnr#}I=CNETe$4rF!iKAEWu4%589Nw{t2cJ
z33r0`cJ01EuWEOl<EbPhUoNGFt(nYIj`J1^&Peb3u}Pn6VIe*J0Z`BJWgJ?0&IAq%
zZHuCU`;~c_!L{!Ou~NK?z9JL71~b!JiC1h={IS$sx@3nRkbZh0uUIf7twwp^RP5N>
zIB{~nS+a5R2g!l5euJ0$<|lbk(qJ>vdg?~J?~rR69ZTovCjv4S8Z7SDd*{RtXJ0m2
z{s->yAKzx7jIK|p<*d>EQN-k+p($W^gPU4H==t?f%T?Zkf|72I>#NzcwTiF&Y->>Z
zNYGI9G`q(;`FO-Av!~&#G<ClU$2G3>@uSCqZc?3`C@J=C$OHJ=@Z?(Q3Whf{RW|*z
zPuIiF%j=){<E%4E+WUBMQTS6X8yBAU`Jp#G^5TmTpCJ|8flN?16}xeQcuQ|Vs0wvm
z{P3RUnRNDLF0R3<)doDIhl|z}jMzGzPKz#s*kv-`uvN8j8f4q#<l1c%{#X_t?qMQ4
zv9@|7A^Cj;B0*`;-|qz^qn1qY?fhyX32->qOloI9+i><rq#OI?(>KSZFBleN-pQ(e
z`{GC5Q$M&H!EW39=@+)W18y|aJ$1WOx|MG*qC_3lTO0W5w-PgLa>cs@jx^MdDP2@4
zHzQb!TWm|5<~0+_b=E7N3Zj}L(2klPhMWDc?6Pq~fGw%+Jbb1WHR^XA|66#w0OEi7
zB3Z=gcLI`;E2|Gtm%%DI*{yoLE*N2z1X!}qGzryI1r?PL4N%Sr%=-(qOBrP{5Onns
z$jY~F>;#RZ0)_Ws>3~*=d&50j!t|t|rN<=V>6fsN5uKj5S*mB5>%PoE1yL=^+(Jlh
z=$(Z+;ua8)Sne<(d4jiw*?b80;dl3c76x)kiGuCg%Lvkp(4t;~opd^s*S94kbCHm=
ziNrd?0Gqs?@zS2e{)9=hugH;Sy&?9CI)V7$8C&Ub*yD}9K(8~2Zsmt_uMSOGr86OD
zi@+dKtOkG(B_aPkb1-u2>urmk^-GwlnXc-!bD8RR86`nk@L5bIg9x#L(yAdpOeaYF
ztV4&2l`6<)c$*wFT=pL8)?qy%y;MLiiuzL_yD_fVMUzXwlFDe^3OsIj1K7|U|D&S5
zqD(LB+@s%|m})-F(=$=Ov<}(huPTkWD)at+>{{|FV-n&hXXcrXzvvh8=w*ap7-|3h
zho^@b$*YDeKR5{D2CTH3u#b6&C{w;#T;I8#xaL$m9mY9Y6SY1SGq^wesPE=$I-Ow3
zZNWbLDw~JJO=PSt;-HDSyf@`vbd$8w=4B1D<D`KLi|mO&j5YJEnD+n*TG!UwG34re
zGJa~8c1vNxvbSc!vS`J`00u(eld2JSjVs2y+f**L_-tZWYXw73YMgB1$p`7jZ@|o8
z#cM6xP$=YDBhJpQB=#|IU7iZ>>`l2fLW!k(_RX{ZV$r#E6_(b1ynLkJAAlz#EG>X&
zcznA6g0MEe>yIXYeM~5^YjhCuLy~odT>2E`G2UIr8Gi?=%q5`QJQ%N}U+|Us8Axtm
z3n6iih^{aF%tsHD>Q(mHrA^04XUO<qRixxMf8q{mAEn;8Kx-HUlsTOpeFVM{^<#W3
zVCD)^NY)(xiZdyWk>cArSiViF@-(x9w=FLQdwB3fDWY)epAqdZoY-<VKQ7)CJFI~{
zCc1&+%LGVDhrqB1Zw$}oz85|!CtVVCDos+gfoq}=$nV3i0URS7pY7!~#1d;p#L*97
z6GIT9)us}!J_@W|td?-o$GJf95#EESjeD;=1MXs+!8Wh^0pP)vxm7Q8-ZQdcyLweF
z#i5VtrU9C4`;xkn2I3_xw#6noY4$5IiZoas%?wtNJQFjV$*SO5M3{|kUqggCgndT(
z8Wu(*6v>ZGfw#yL-37;iS^CvtLQsincgS3Ci)NCie9m6k;MV9C#EzQI0kfdXmYxjw
zTfo(P&_EXR9_&O9i_{LPL6X?DeYVUV21w`ie1@9G`opI%sjQ4FNgPB0g)U(Vk@m-7
zWY&j#$pM-<4BFGhf`|JqFj?#LXIp*%%-niCb@7K*m5Tl>J{tLVNI_09?tg?Sts3&C
z=+@PDluE<?(L|L9cEyw`Y4JJ=kx(uZPv3TVuw?=C_c)@lWRT4Fbno9ZolsW8qL_R?
zw<GRC5*R|TC$YzetsX9bxjiMY*n#OCPI?$C63IVdDTakzGv!LL!VB-cFm#;v7($&#
z0P6}n;=jc3QOJsDXL`hEo^G3S_=4EaDX*%_bq>}-#$q{RlHtWG_05OxPa@pua0Z`$
zOJ>v|iuyITZ801fG76;rG0LyAmpV26VD&T=n$QWBjevwT?0j)iwoa?N{sB=($0?43
zx@^Q+8vveMDYV$lAyB0gyZHLUVxE&E64i(<Rzu^$<e$co1WAoR35)fQKgDpCVOqxZ
z3mz|N1A{adNo>_5(a@x>8BF~N8=7E$d9jlO=4y2x@vokiDWY|`xbA7SKu%46DNL~y
zD5XDMzlqg07`d7j81*BEc;{*ohOvGPzvuoVQ=C`3trNhKCVpqcm2)P=i=Hu5QK99u
zFMbuT;Kw?{6%Z=DqQY)kv_>Cd;0RpBbqJR1ydb@6qR;M(cw^xx@{!6>Z>~&C{Pd5<
z+-dC4tBJWwjndd(n<5wdK`cvod>gFSJjyC}r^7Grt#&V{-(EV?@B+qA4OK_=tn^0W
z(41d9yg1kmLkm(++#1sJ)$1GGe0lO?Aip1xs5a;%nMxEHe7dG`xYQE8Nxno#EBJ|U
z@X8>o?Co)~$yhuzrXI4|DU~UJ)ZcC4uQ4wAVv~;dO`f)Uc2ltU!(=pG!yEP~)6cw<
z#}A6~`E}gL2F#Jk5IrryC|)E39}%ky3qf3{`*CKm1pPI6z12-hB~<BtwL3_a>u)5?
zWF81|D5_x%j)U1nOcQPo_sOCB(L&{fq*&i7rQCa44y{$~MWN3gg=-LPzPVb1wvvOy
z=899Td7X)ha4E!bcO!_RS;C_YtS4}%*}fA}UhM44(4DQ6aIU7SJKLhdXxSmf@e6l5
z;-D1mxZBCRLQ+F`{OgfuKb<ct#Zxz~e<e`^vb~tDq)}xeK8SjY^fh5#F=ZR)FW(Pw
zJ34n%$HfuT@mNt(Dcp^lnDk~y7`WfE=Nr_2gPMbM>jTT;fRY9?3&H_5&$;0Wvk2KH
zLoI{qkydEyqTgR}@zKU5T?%y@IJ+fr??{y)YVwxk)9|1{JhC2<YwjEc&mb`B?(7nl
z^CzNdT~}2$j@>8F&Y_<?VOX@+qFr2qVp_{!RTe`%!|$}~_pIlSL&L-iRyZP`S(lBl
z>hN#t1fpjcbjo62WiV*TiIY)ck#>gSCdoFPkKTF3OW66+jUcQ*GAWl)Vu@;<7fEW?
z(XhJLWTeg!S@d3e%};ErUp)hSo>Tfe(nqdW)KoYSc(N}FtITyLTh7plxfIns+{jXE
zIla9&z^h*B7DqxF0}$tnFbP@li?=0Q(#4s#j_*3zu<fTpomEKW!RtgSxzDAVo|1E_
zX^uWV^}?z*mQD3m9HQXnGwY_-eP<qRRaB2zW-t&QyZo+C8BOn^nV%={bVrtN1cu)~
zqcg_wptLmG;E>N!C9oue@ngYHAKnTC63n3GkrOux#f?UH$V8J?1?r`6+XvpJXAg?=
zGr1bC=QX|Ic|Oc3@XnzkTz4Lt^EWwlBAi4{_@b){)lCW22cF^+8(2Hcx^#<88j!+q
zevGJ?8<W_Iv?l{<8=m-NVddYp$6YZ{d1s`|^LaLG(Dl5Zl*k`<l56r!;8V5acUU_i
zEV9X;H>lUop}*oGJTKNJ^x3VjUZNk;yxfnc6PBKNm*tv|n2~OVJJVMU)fs4>df-ly
za$!;6%1Q~fJBV@CS*oIPD0r7#QJOXV&>(&IZw(S1%&GIET)(mZ(Z{E~!ug1E{J>3R
zSqz5+!n<h*$S+cA>CuarF#7Ui^T*w}x|%(1FbnZj#l})~dE=ixQ=D7ML7fh%9U#ss
zb%F&N7F2&xXH`4>(wsFb2;z2F4gQOhjV~n=K%NmdDJuUQ8EsvM+QSVHuUrN+)Ic5Q
z`DMD4%M2REjJ}!ZlzngbJ9Axew#HQixIUO)AIOxMJ|bmCGKxQ;DhC<Gcx%9eU{19H
zMdVGoeBVkP%>FZ`Fn$fXiXbLdG&fm;|MVa(h3Ll<=+>vMLpnS=xOSE>(Bm6JkNCh^
zXqg-5<2fmFc{;UofvBhM(4E_N3(vPgrOXPcb5CK;%eC!gK<~?rHgI1XdxvIIBh+>d
zsBaOAA{%)zbua~mfKUIgi4Z9>A`^r)#yLxIz&UuPArdG7Q)GV5eK)U^=Aou5h9Mks
z(|6}zT(iIDL}h^oi}r=jR)1-<>BCPoU^+p>2Lub5(0ykB?i<}-+&7UW+LR|!ug@Xo
zCx};XpXacoRSgVv?O~c4zaZccjm4}&e|l8YH%+JDCytbk8(?hhh?kCO(*gRn&SAv)
z%hxaosdaYng`|G{p+fW|hIdDgzC-38O{9t=I+TSRP$PAHAA%Kh;S45zYnvG+PCv)#
zfTn=wT&cIqtS$iB_<p&lkq`!e`MXxfPo-3WfFw6%G9|+LFXo*3|G=D+>OyF5_p>8u
ze$5j3KU}$gW0L5kz$aWT)+N~Tr+DK(n&SWRtx5^<jE&@)w*0zM`FkYI2OWZQNDeRF
zD(Ncz_0jWB3#Fe6AxAnb4%_|me*N1O|DR97cQzEck9}4S1^?wj|A9REA5H3Scp)Y0
zWD@bAfW_yJ$CNexEPrl3BG~;MmkSJ<RcW;9|BEI|WmTT)ulEUbPZ%mbJA~R?*wg~Z
z(8>k^W|`uV7<go@5b>M|yot~69p;DAgU9aJHIZIGJ<ifhz(evFLIpmJB9t6R56D+)
zgg}CJrL=Kt(W+E5wn6SeyXx$mV$;qniEup6pjdy-@e4dDMtMzk=m$N2GEM(_5C7*2
z-+d%!&$E+B{?DAf2*pz=maD2&2)Au8O54Cy4mvVM#QS5b-o4Kiu$e6IX&_x9M@O4H
z73G$d`L2Z*6rbk>t>X3V+gR*jp4Z}JvDlx>2j1<Kk?*M`y{g_>4qYlcTz|yyJt?kz
z5p<YOAp^Wo)_d$><!hLjz3R_%p~I0qmu4+*qL3sE*@^E#!FL%=5<923)`Vur{Xv15
zj&KFx9~3uDHqUJKO<ny+yMCGEfz=dUB5GmGe_k@a`-pM-Q74Q1pPOdHwMjtx;vna+
z9prYDuZbXSEe_`Pw-5$~Ct{M`3C7aUq3IzxjB@*dM4$gxhFeZ{;^+82HCKITW)Pd7
z;ZeWfW&m?De!J9fs5I+&dWDNKae5D<iIs(}Zo_0x*CH#nU>szZ?Le(dy-hW6Th&y&
zvAzDxhLG#}-F%VnqH^gy)Ao;F(k>PR?R2_Lw(|bdlLW8#6T!o@Xm&~XPY$d(KN_l{
zD@A+^-6>))RHYF>z-k;6=Y?G$5k@~+=)KCSoWu?|GT~OvyS~gBpX?gEOBHohbYzUJ
z9)DoS;D|NxDArUtyc9ho2!$5wS8LK%>>+{n?uW;tr*}5|YJo|hqy_!k-PnQMud!Z1
zv3T0mk6-G0wBNB2s4<Ik?<`3D2OR#lhix6P)TW~`_{&j@Jwyuf)b4>-d<SGc()x%1
zLrK%G){m7(CB(1x6uvlDOVa)F-c<<TB)Du6I*rp?7H+1sKm%{oTTLO*{-pL*vv?lx
zP%kxo)+=n@>HI9;DEh0-jb0c&wWpWw68w1*sG?yv?gHe6>!qz@jDdLXBhRi=<P50t
zbS2t%?OP4uIel%VVFux@Y-`>*<TyKN$3U36Hcq}Q5B}9RmxG^Js^WM`dro~ji13(*
zTfy@Bg!!|PZr_(L@Rgcm_krEsHrSw?WQ<~&#UOzAMf{%t7%~M@+h12!eSC&1Od;!W
z;||-fa{6I|pQIsqb1*vLo(}pL*Q}6rqEzjXVR~7F*MVKOaCVJy&*w2gD>TnWy15HU
zetxw%B8M-CCdeh^E04Xry9!_DIC&;a78W+{kPms#>_$+@6h$((<V^Hr&r>YR2t+Yi
zk6nrlEcbPLYdw5ww}G{;X)ItO-zbqFi{HAEqMmL)T)4h)r7)GrW@f?r{h(`+XAi*=
zdsxECbF=RUOFb{WIQQ`JBAa%)cHWc3)vswhpsuMkr*epNIAkz3INWF$bLW}B{qFsF
z+)|~`j7s0sXj%B6?4_6g(OTJZ2S&?Oq3hU&^>n#+=t@Ej%`f{GscM9@I(|Q9=68yD
zjyoEi=VmnGBPemo=4~76+mZW;ZAiRcBVrK42L4pfXq<+xf}IyVR5+!*1!Gnv5&A%z
zdfP;%_iYts5;k`M?<Bjlnf2V@DX6sNkB?ToJNB0fxht37h6nmt!$RzCIRPX78^^1`
zlEd<opcz(z<P?q;Y6m6dqgJ7*0OGs(uglus2ibv!6f3S40EqTzdD;u$#q+%F8CwT<
zgRz;Lf`efU0FnBLNbc%+kL63uvFmoUbZ^PI4Y&{;DGU()H>%sClK(HN8}{E)-I~5-
zSrBK93hpP)B}o^o0=&>MadPZj!Z}Pf$Xj*Q<R}FoJ*Q(tVG|MP>RLYmhG1OIU+jXF
z68Pj&ypLUp*Px;s`8ferLhGZX(rBi%HXpvTmPD}Jh-(Fqn*q{!is?0`VR}n#F;-tA
z74IvKW@SXD!zp<%($hJJxc=!as68$LK&yUCO%$^QP6`>M8v9San>F>Y8T~tmx^o>D
zq_U*87<D>Fwsn7OxLk7Y8*#}Gu*&!nFjV9$R=$egXi!>@NhKR*QJtP?O$_L#U(cgu
zoZr}PB%&!uaAk(LDGZD5&Im1}`KuYu8##l{dE}Uo7G;}k*L<EVWqf0;79q~vgI{;t
zpVCI~dPh;B@$=n;^WRQrV4b@J*BwVb0yv`$ZmpF&Z8r=|FG65J@yG#Ci0O-*tDZ^W
zJ45w7eEIb3=P#BjhPf-(Dt3VQ^c2F7TOr@TFlkie;{p3JK11PJN|^W&kQ3uTm27Yc
zhJ?C+wY|2YTuIP6d-A7y?cbI-jz~WulB7ZSdniF#4Or&d!-?Giwdw_+mLz*F2TtAt
z3^NYAi8QnfT<;z1ib6v|Omda|R66RTII6XvZ1jPy52DVe(_i~UL8qrmOLw1v5Uez#
z3!u_<!Z%jq5eK%CJ=i8CuA(v$ks#EMuJwG!yH{%WB9Z`?I_~#kDDtUW0lX+0F?<6d
zq{U7Yl;eZ?4mXJ}f~W5&D~I1Mk;w;yM;H*kU@4P39JR5VhA{W>8k!ldG9#cNC0<C<
zB!MN6yGo;0Q1qC6K>M^zH)P=R#l~g4x+$JiM@zTB@RWm}6liLq5T#0NyDv!3eC7m5
zmY>a<j@{}fz>&jb`V$mXiq9|IQR2dlK8wzN0z}N4=&rEQCQ@83d&|fTO9B@BhvdN|
zzw@%}MK;4CR#<3i!>6nk!AatsBYyNKOr*`EKkx`*A0q|Abe*0}^}rljhJlM`gNG;3
zY*_c3+%{99{T%BqF=$moAu$VF&-Xt_A-Bvbyc8o*&J&pxWq#fNU?d5Kai!u|mM?+p
zz790YRv?cVIi}O}pSfx(d`>nDCF^?uz0Euyat+D~({Eyn0q<6pxJ_E=+;d7(cw&%6
zes`=0V|W#~g610&APLZ;TQx^AEfA7HC;stfw&ch-?fhvz>qPcztHEB<Tv28s0n?~8
zMBuw!wYKB|S|(u59V7GyjUi!a(UV2(+NUEWW{KE%?q|?%t4d3Vb*xGFR^zQuR{khU
z#37y;q+Exs{YLn`eisgbNQB{tp6bH=04l_4y{RL@1F9-^8T8%FzFQrl`%#y-;bKc1
zh6$}GdVJpPykK6rHi>-zPN&b0WX@H!AZan%5?HECvZE*aL$#Ukq_E!g1fw?e6CSh`
zeHTN;{gm?pQJsGk0sp67{k{12{lJh$r|_?w(Fad_ywaRGe%dV#^yFfZQu%sum5P1K
z_o@3He7i%Nh$F;q0jit$%W6jyOB(#&`YRQ?1&-*KG~&PYA82=%8U?xS*e{Cn#B25!
z<9;vj=4b{?&S8;-Ysu?~emPQT8atbjHhbKfZ>F7tL}U->mFA9lDBedb`y|U6IrAC|
z^BQ$An%LBuJkwD9IA*70*zeEd<)bdem2j&s+Z4})<QZbbC|94z-f3^^BU%>3B#jm!
z4So-_0?vGVot$8v;H5=lzRwTeK4f2GCP=*ASoW+tH|%0EE^BfaMG~%}gL#h5pcR#+
zR#TbqP;@cCu_H;K4;wZ1@>39J9-&P`V+$^8P>}Og4vZqSywOB2Ty~Ipa@dmFI+$C2
zE!#kdru*f&LFPr3dXu}Nh?j!;rwnypwf-Nuv$@Y@vm0iU_WL<=)z?Do)QYcnJEp1<
z%)oo#P08eX@U^V|R9O+sF-AsJZ<V&TEB?lUi|$VepQ`R8=L)`=qb*n>SI{F(oMPvq
zoZmX0`@FnhZ1wTZj0>cKl)RZ@j&OX#=$Jiztz=PaExylN=gC&*fH!IqTIlcHVjG<W
zZ|&HiR7|H+(9voX>i5qZ@pI$v9!5*4UThm!U*l3)t9H~39NacIw?!D&Z1AkPGTJOJ
zLt&{ljQEGEuX6~%8H0vp^NDUuP8`fV-}d}7p?aYGFVzFeCm8>)C|h4j=2H-Xsk?~5
zn!iJd+a+tcc$VuKQSqq&bV|P~&&Zv9qS|*i6vX}C!{*kKdhM9?2(=`sNCz~g=q)B<
zKTMD1pdy88>gKo&^02e#rM{u7>IW0j*rTh~NE|+3*CZCt4Nf+4niI<PVO3Y1OpCWe
zf7l_yB?mED&x5qP^VnS05oR`eJYcR5c|=QR?1yD9lqKI){xpv8IU9mJ82N_<1Bg(V
z*I2+A=#~GG9Ltu;{UyCdmwA0H8-1sn;$n09c=x49@INsGP^c<Y6OV399Q|~LxXql*
zbl<Nqi6V8(<#P3sIP%I$!L=@+bmA|oen25F3}qcz>BF)3y966Q`#WCt^`;g(5MHJG
zsDgTz8(f@ZKU);42E5AB6JU{cd}re8AIlpl53=fcpVBNcWIN1COQ_`qw^>wd;#1kJ
z-QZI$slB!<5}YeGS*gYgk591BvWgy4DY3LnVR%Cc*0C5Fboa`(@fydr^VW2dCK;Q;
z&Wm^B5gsiti&!Ct-0RkZn%R}LT7`+rp5lir|J3Qj>%MMCQ&-)|-12+hTJ+_t&*ZJf
zs@cbSGAkd7=m{8~K&<JTdKr9)Woe2WhBc)>sBGFiE_)r%pL&Hf$y;E*pzmybKG}tn
z!R#_-$xgaqXV6GaMMsc0tWr8Bd=_|CpK-I9(O%fzIj)Mg+jVcK&B8uj59RheLTbV>
zAOsy*<keaX%TgkD3#mJZa~@E1Vvd+&t7IEjFm+sAEr%*m)<EV}kQ@I&;Cz#9!+ziu
zcI@QIT+RtPSyj6Y93nT$=Ap%g{l?)Uv$`?z#Q<#7y$rF-fd?_)sFw(>GcFP+EKmMe
z22;q(m=sLg@7W1!y5P=HmPM>F-&0*GBhd55;wO+y9C~A%it?%#Jcd2hO}+Inu8~!d
z^sv6s3J4WJYY(nZAWXLn;G~Us%~ntQg$HR0JOua4?suWyLO%0FK`~w5W<_&0FXMaq
zP8Mp!*_q~k6RMg7!82{gVmI_7rW>aVTk0Q81(G*Ihg7APwe|4Z!6K}UU{gA7Ru<15
zu2&c1etE}k)Ayn$_eatFY4b0ICU9*M(SabPA)o+n-QE~9$jY3w!Pqmtt!2hQ94pi-
zhn!#~AvQ`xMN7On#0~GGBxbAr&}wN0qnK*CuLm3@35Tm!p){A1VGDCRjcdrUPzAn8
zTFTXF0y+jw!5MTLa@8)g7xU@jtP3~{Py_>s<;(Y;-(jWX(9u|Br_AHA(IqHl=LmYX
zw=o|cWYoox&u}*jI)se$XIaiG<GMqN;ur2y+gazZT^9!KyezPeN!z|zx|n?j<D|!Q
zk_LdPK5K*bmmYyA@HAYQI5@&N)a(t^5`Ad0jmH$`9s&0)BST-cI2tkJWhog3Ih>r_
z`*lb{mmN%R5dHL|Dnt4TLY8{l{%C3NrUp<@S$R1n7iGmU!f+MPEm<GX)Xfc;ru*RR
z=9vlT!Wcoe@ep_JhTLyKH)EC=i??26XGkNwu$<Lv{T=Z+y{Z~6cLz)k7qH^x?Xb(f
z4$oYFFh+5M64}k!Zl!QR{`(V9p?qED5o(z-B|QoaYK%iHwfOA6wXWV9>*;%EO}ZKu
z6i-I*os$w*BD*>VzaSL^^`SvAQ6H>aImVxcU;1#pzwf?Q0{@`!ILqqKGzR@y*u%2S
zTg?%!tX%bCSkO%}g7cRa6K<|nd4uVswRhK|Y^MUUFI+np$J_WsG4hOi%<&9?YJDoQ
zx)Ux<>G~BLr)ovTCU^{^?`KOdE71}@{V8BkO|_?->#V4Hy6VSQL_gDWY2nIu_xo2P
zicr7eQTC${G|~L+u;w4o5caw2z{O<f1z%qalNwMfC=-gUN)and(PN!`?&iZ6{ShgO
zo!}%0-3uV3s3bpsbu~C@Y)_4sZhm0wacC~Rixqor3l*`9lnXk877xqWUi2j#TJp=0
zlb-c6YUtjS-P+J=mKzBOw@?la1q4u<=<+jLcW*JDM{zJ0hrY8CZA+3i>IdOQf`DZl
z0v)r=TxZ;Etdh!k2^7o+bZ8<WO4A=Hbo>EFDF7=7z5d4MRnVHUZfe@NBQSGF&>jgv
zb@JJ1(th*Pf%GClI+ifmM*GD5Q5Us)bu%L+1Kphjy#>5f-TnO}`G<67q{bAiMK2<I
zJoL2Lp-+uxibAD7^R`ewV+z~afnG0`>@cgi`#IFLes{3n!?E?iiedi%dSC7{ScLr1
z6VT~hTKulwPnCl+CSK`LiyYP5fUnqG`~BXy9;8k7Je%bH)vd^sl9IIgS3+(-{pq8$
zWcEz%hBFU={Y*GzF`%x-<qPd`IU|E7z)_!!gjBZNOsAGngKyF`ly=1f_+&$*w@hGL
zCR3)k*T}GCYx)$GU`GI-eT9f3&i8!Ah>AtNJ&roQhlbuqj;!22Pi>QtVhX<zYe<QE
zKn%U%Q(-ym2cCDF7fJaZ#$(JfWftN(n07F3vW|x0-pg+=LikN&O%izw0n1H>8Vyy-
zU^|?<+nvkClie_qV-mWqBH5gSrbupPvnRCMOyy3!Ab)PG`N*O;$8l`i_?AuO4?&{~
z!m+UF5`52ZIH;OYv2l-P@a(%izj&cHH&`2N4PCC(xnm9{hPi$PGZ6Fis<p##JW1Mt
zJw?gJUaWx4i?R3mDV}Jz*}04<GiBQ3f<9UJwKG$HV89OU@(i|&p_R$?36oGr_x(rI
zPpX>O#~JV6>zXGTtJ&$8$!;nRl$1P9o2Yf9Kf}N|?jO^#5+}Iha|)Brmw~;vH%qvQ
zVN#9TfUqGOfGy#zxmb19B-PFqSAy6FT>RgPW%MAdO`S;o4a`KM<*Iz<ATc$f%0Tfx
z5Ki-JN$TmE&$F#ci=qO|ubY-{4eGTBB(}+mQyL5*)HlM60*}W>Hx;ZHZ;mPRWfzfe
z#mw_3+4%d376=e@QMHMMPTmrqPGBc)hz_claZ6%Xul5k$T3E<tTi0&9IVv(}I~G0k
z>@I{j4x^{pnMn4(Hc)td#6I=Xrn0!j*NeH>`b#QV`0aA*;XJqP5^-IbZM2)?8VX}?
z-T|}N5%7v^(Jkzy!`p+ao6Q=Hl~hYeV*Fsd+lQ_W5tri6Jv{Gjg3JGX`tJtM1}me>
zgEReVnW2Ux`rir={j)wiEqKQbj43c!>$KRd=CUJdcOf#WIfQAmxpW$AU3~K2{P|ES
zlP@L34x}`E#k_4Own09#h^Uh)+4M2zQ4&p|2`c0giZZRiwMf8CfT%g$&J_9h=Hyo-
z_4nt~8<%jd*v!{>ACpguuP8Ln?@icZn47^!>|JpKr=HG2&_OSei$MDj7WUbxBlfTh
zqq5bgOMY_t>z<pnO^?qwf`FkGb?9r!L)0=AP=!V7Qde;mA2%Iy7@fG?2O}UYzA#AE
zsA30(Uc6UX=3WA_gyHIQ!vuMRJ`0P3DzcNcD6WC4zIg)yg_DA29|1N&{`NA?5CF9B
zBgF`sEF7GGojX_Xsjz0h-5R2=ev;Sj1_Co+$KHx+Yfr-5#Q@7!UZcUooc&hGM2=To
z8G;Z|Ed}5ZXXIUv!HOI<E%NgFF+*Vu`PePWGITBWs}WbWg?giy%OVc?Opv07423c-
z1ae$BW2z+knS(mY_b}=H0zP4p1qB*}iR@adSfWAKZ^{(iA8{^v$$o~&+XWMI=u~@V
zX?O97QR+3x+vN|fzo?vKDm=5YZnUJ`KUMFnFFoPs9qSPUAq%279Tj$@ocj2Dt1W>N
z;nmK=a5V7ei>^L=y!Q#UsJjPJ#&kW?2m$+L#@|;BU9B1RR9VkPt?rpHAqD7QRar{@
zX-<Z`4&!T>hW7QfbpI~A=88g}%rb}Q6V`{m&H^qLerfe=zq8hecP3>x&UvjxQ!ri(
zN%vU%oE2_4R`xm(j{CqLz?exNGyqTWA<*6rLGJ2C9!5HWgwzaENnziRuVDaJ;DSkS
z+SQ*xT;jc&NM42v07mbC71UQCF8i0?i(FHsWWZs_LHI$}qky>P4D*{xB%-V#fYBQ)
z4HDe_uuydL=(PXAT@yx=?+-j+ur&t|An->MmW3o{0K1$w{kDwvw3u8vZAd<9l!SHw
z;jLA5|2uC@>QhoovrwI+kO8%+(rPbB_X&i1T=w*t7w1`}b(xGM?7SuVgCB1OJDTyB
zjN8D<1h+_NKFNV^XI|7tUbZvvL?*s2NKGyAdPmu(9%cvJW}v{obL4-tpn^w55LtxR
zxsxR9(D|y)A=z=z4wm2pOoERJ@g)dz>jxaUPSi=4Nj`cYRGDV7lnKabh@du}g7WhV
zuyei*-f_O;3>t{L%jZ1ns$LSogO#*(c8Q<k>M3+y&TX=o9aAZq)dFXr54^Y{^B4&u
zP}O8^uLORW^-;3NdfP)_w+%|{pX!M&x41q;AUIsNjWaw9Su>q+3`zvq&9qqi&-%;y
zOnhtlKUmMWq9+FIQZRCdXjNNJlN*mhZKp+?pMfu%JB6%bfNumMs>%HZA_PZU2ZhnY
z>I3G_ZlgUuZLph>(qvKKASMZ+Ymuv+8#tCYDJM@-xiub|N;LKkz0=vZ!}6}yAu#1)
zj@{=9a@hmCm4C44W!m)tyyFyjpN}bsq0lEvcc1S`O)WiP`)aA;{Mq^CEs4F4R^;1D
z6CH>Zs8WhIx39SXg2*HkKK$2M=I;=h+lU<OVmIZ1(r-lfB7I^P*m!S-VDZMm9TQIl
zSdp_M*sR8NV@_94_AF{8T?}3LxN&i#M{;G-4wDfrWS`zc2LSoYvy-v-)8{7$#abs#
z9|>Kx9fpF56Km=-%*re>$YNTmcTv><lqaaLd2n|h$y-o2HhuwrHtK+IFjwU7xr;?@
z>*IMU>KxT{3KGcNCqFS%VQ7mG32pV1rH^5_X5+{}!Wlq-`j~d<i!;EZiBcT%%~H*q
z-}@q1;@{NlFTO!Qsp&c$6)ag|>{^Q0r;%qeP%fsAsoaHXp({vuNROXaem2HG=Mm5n
z1)bynozNy1CV80@lMph$ZFa7bgdMDa;<j#Jp)pf1I?no~v%(z74)^VMtu!u^zRd(q
zr+68yItm58(KvMSvR;#OO?^V1QX)er{LxTw>H@mk6rZXAIYZS}8G$k9pvb<DW7Gj}
z3oJI9`#r7v>)u$<G8Y{%6v()`Fv$+jBb@zYCL&8R<HU~WqV1_!;<8|IOzW%Evp>$#
z-kPfCX=UWlcQuVVWA${@={8sIbGIVv5anX+sf6G?n~9q0G3bnxgd!b=@l)Kj_3h;1
zglK2vIf8GJFzxz>2$+`(1x-rCPo@j#u>RCddX)U~W^Fd&EI5pTMP#|q_&ZTw%%anb
zVn3R-W3s|l|M4VuTM5MpJH61*aK0yEaD90QoCW7gz5?`_)D|3UoC;NBO$tDSAc-}5
zWm9h#7=_q(PUwHL`MlHFs+gegE8T>EQX0<wn-jIvKSOx0rv+i>IFo-OY9BP+YG&HM
zs;RMTG`~12-=817(4wK5u76(x{0z?$^ssV}ph#uzuB&QQ(}uc{5pVUga9l%ZZU}`f
zheD;_G(Kmyqdi~c`ZkD?JFDx@RH^0Y+~P-1&M~7}iVE{X_g+j^;-@w*av9_uRX0vH
zXYVcW*y3R37;J417g4Pb7K8Ym-C%sx=OybZskmZq*ZtV`RdxF-BM}>gX4xj4L1)sI
zHo-9~VlGPeg7N3F7<3uqPO{Y-A#Pm!gdFk2qza*Kdf`=9BJ7N42f~f-9!)8cuIWxZ
ziL#S~A!Nol-fIdgMGOnU9k6wN2Gb11e$%+FXKup$R5(NA&TkkJ<ixSszr1&_K3oh{
zHY(qPO;hTt#fab{hoiDx;jPJhY1tCa0z*>oc4g5Y%p9vcZ)%cGG;cX&WZdz7Pz&}C
zfjPmgO2CO`B-YogU@IE4ZJ;)#Q;G5sB<H9jA?<F1q1h!4{IgrK6g4akZ&Efh!4Iaa
z5W3=VE2<+I(IOkt<2FFBSUH+IzNkl@;n84tI#PlApo_4$Hm7KCur}yg_wpGITM|;Y
zdyX$Do1=rQO;r?hU3j*P#wa=fWM`p#db|HjR!*yy;IPpd_mujY;_i$=^q-u7I<nmq
zhJl-`cqQ&Kzo8rGk>oHmb>(`^NZ;LG@6S()=#rdS7Y~0{;K08y!LXWyFX26qnDc{X
zN}Rn<LEQYeH7eLKIlPLm&fV<yzRqA+g=0(i@m5Ys=;eX@Q+?ZJchL(m=1=!{ybI2x
zHU9CHpvB^cKu&(YGgB2-?)*)S=qpoSFHc;?zQX3((>W%@{VU3g6ahl)!DNfL!}<q)
z?Z$-KQT;I9+_pxd>?5cu#2Hs`5yt4FFACM%mN2}LS*H1QN8;(cEA5zp-C6IN+}q_(
zF<#lqu^=!XwpU#V!oWLjAIfjAoHI~||NJZBn(u9dwf4G;;^0r#8cd1kvX*SshzUal
zqfzuWg}32X=Ow@B7jXj5ns(?;*eTl$`1Q+NoBN9z=Wbpuy{4l&f`b^3!t~F=5^Ma7
zq5BVkjW_xs!L@Il-}qow43V4)mHTDAa}B9?UJ|!F{L^Sg93{0Y8DK6`As1jzb>(3a
z0iLk}!&9T%$Jo8BakavR2G^9v3=Hn?dmGyYgZ-qyB!W3OeRUiv5KGax;tRGP%Cs<N
zEl^f|!GA$e?2%UeHL)p!zL3=s{(tU}PT^>T07-}xt?T|@@AfSYy@mm(a__Z0Wm$)~
zCNpW;43dxDm+yBz8GNm0Hu16K1FKYG&eHcgZ?NRY4Blj?`m_0lot-pL=k2%pGv4d3
zpZ4!xP{zXViU>=M_~V-E)^8di#2DwFU#PJy-@Jc?Pr*aamU@2c7Oa62hpMW`CuP-D
z-C`{_(PCNZs$adW8ln8vweVvC#6jsVGWM(c|0@uc5xFVMJY@fWd^39AW`sOss$!P-
zx6{b~{-O#qTtFCl^YnlJr+>@Y>ZL&3i|`k%p8bke^6$6gKfVq_s+Q<Fo&V~K{I}v!
zA9PB8<PKj=w*J-Q|F?9=-(Qr9h6~_4M-l$V3V~OpE}s3%XRWk<=iR}6-z`bqSk8hX
z=~2y+n%VeMKOP9D9P$xX8!YDm1Fbpt{=k3v_*D3jkNH|N)n8Yi7kx?=15Obc<)Ig#
z%X_BXbsh}Lp#9*3pg<^U!yo<pBJ8v<<pgCw(=~EzFPO}-YT-M5fF{(hR_!x9CN^7J
zKfhMSjJe)QMy1_}wP9?uVt9jC$?s7XJblr4kiDYybpDp<4{eu76R~=Y@EPpw)Ry1g
zD9uxa8-IEM{PXFqVuP>|%g@ic{PC{S&SKa3Q=_A0;TC^hSci6+Q6>mK8uEoc<!z+=
zp!M8TwHSGOkMbso2G+-O{~I;UgXI!L$?DwiNlCWAJrdbpc!~J`w0$FW=*^JBD~Vz^
zHs{?cva%*dF7y~mJ8oduwl8}_dmIZh{0|-kS_)kjD`ON|aJ>+8_8h!@s7ClxUPcHI
zQc2MqvH^FChJF(D*XRGYUoZlFll=6*g%mcOb=jMJeh&a-KJ(tR>4!?in~-rn3%2Xe
zfP$y{8XPuaV7T25cpoQ3`Y0A>?dH)(`V_q}to8Z~g5zC3*Ec{eM5Gev-jE30$_W<c
zn@4c-o<n3t8)(C`z|zsx?e_5v#Kaw8d?y4!U?vxmR<7?l8=FvNLP&oYKu==O8Xso}
zc694qM?i_OiFIBY?DB#zAbRaPj+(#S@V@bK-h$@QLd2x4-XJ{YH;XpoGdXJ~pyE=U
z!5;SfXNDd!&!GVBXp@hz;Gg@Q)D$|stv{P!NOK+w(yy{g6=Y5lrM#xMO%Srn$RrgM
z{4O+^yW)p;iDxA^5Sx$AZGizWc{5Gy<A$=$jnor;u%@ke3IhrTPY9!12Q64&{@k9u
zl8;mD_&1r=;OZ4{D8E_wnX=%*_nHU>?M5$6?wOpc#BVP>pGnTp(k+_!^;0zh%d=yy
z{_>xf<$DfE9@U1MRq_Szc#;5Uw2_)DVA3YCq`Rukg%@Wifz3>Mc#$p!mx=|$`(!32
zNaYRB^z23Wwd~wk>jMTL6kgwo6ZqFw@OK-B1`?(mo7q+N#}cB)K*L@OD$3lw1HOv@
zXRxjVBWGARG7&_Pd+akE4Bkzo+V9?~-r4=0o%-=4-GKRv|30|#P`Vlx(H$M^Tq&lL
zKDfrRiL({>Y~{+jr(O-+t{hbbk@o^tJIM=0P+9%|^3#+KDs+tFQg0mYYb;QsuWUge
z;;Q46n+(q3;ZaRNxn<`~2Am}p^XC;4JEt96UJpnwUdqIe2NKBqN^Ys$uGhhY;;3dX
z=ApslMjbg<b@fz5x`wrrX3iw{k1Y&k)g>mn8$DkyZ{It5e?qWL>N-BY(RJ&054N~+
z;)&FpiQjD4w8QS~2?A+ha+sIMXla7T_AxI5ooPcVo%(8Zb0J1g*Gh28nKuWPIUNq`
zrV$5Co;BRZUK80Kf~aMppLi1fM*@c#g!+xCw`zK3U9wlLSD*j#s7WhBj?}S@mhs<j
zBUdoG#Qp-)d13oZ&&~0>&u-o==@NVfW4MIiYGecvL_ozVx?JRsMA<e+(paYJmXF-0
zLJSb^hF>geW!MHjQ7fp|zd=>@<hg;H0$2>u>_MpKMbyLPw;h+raLYilxUDMUa=9Vn
zc#6ooh-d|1oOp<x)qvE>c%$gfH5|4idOhn>&ABghM${lc^-V@xpzOaz%eO<zdk>PB
z*1tIP=q?O^Nt%k_golGLYy=qhaF4DVhN+yUn*e5-)f79DvfBc?%u{fLRdbr~bUx2x
z<O5heV{j9q6!1I(80Q`cB|n4z#)$dLh(n&swUFKMtll4sTnzJzPP@><iYd&*9i>J$
zTd-l%sr7SJZ82Egf&C<r;0we}tVekt%G~daH_iwK>ELZB^|Xay8@`(rtJQ9NR@3aZ
zYZGn>eVmlJYG?kfqEt`>T@k{ncf>4=_>IXW#%HwwT_u0zN`LM%JA}x!KDYy09aSTk
zY_-iH)!o}>O3T(}DjP2BU{7G^$&bdaT>f~m=r!USGAbY6f^4e0<^|Tn{f5Wu<Odip
z?U^^HU>f-(S3SEOA~R>fw(A*at#!0u*Q3H36q0=dTG1%$8p`Ig1Md|_2WXtXFo3Ir
zpW_*ZGKTefi7>I{++n7o6+#*jjX?wVA5}~FzuT?7Si)O60QS?_g2La4YDWI^WI85=
zqrdsQX~z2iHKfC8WK0%#OW%<EknDsBpzmnE0lT$M2{bB12y_nsuC^E)x|p{%Y$JuA
z-{nv0z&cM8kJ&dsGD_bA41f76orFilkG^3l$YOETGadF{r3JB(Xq%1K2Gv+C{c-q>
z6BzZ(0y;#pinyrFf+1t6ndmr<GSH6PPr}^U_P}c6;fiL>L&QoAf6N0&M&BR`v_zZ;
zVRO|VyM8<?f7j|=5=hP@e+ig_yQLl4zPH=uzaq%{QNF-?7b4xOK$L~K&Q#?v;u0dI
zt$-cu3mCkQM)rW~Xf|DXGVar)Fk5{vRQ(EBO-nVP1Wbxvl()%7!_01=DF-k}IBBWH
ztQ<&Vw~u?gwl8N-rz9N!n0Pwdk7Wo9MVI{Da*Iuqf=>`|C+T48W%<(A%JNvbgM|~}
z`6SaLt@LojIA8R8=CdR(dS{Nvw;rUe6$n#I*ir7@;aZe;0QhdvExf!m|1DRZ#u>0|
z5z^uj!=0aBzk-|Lyv9d!uDaWSF+V{J*Y(~R6uQNBIm1n(8qr3CXL#zX5YChE&U^>E
z6LrpwKy2L~`8ln_bCxZJ>URDSQ@|0mgGxexN<BKYuc+%MWH^?NkXRwiw}`t_OBY%S
zahY&wu|~De#;Tva6@;;zqmmd8X?XXOb1NEiDF$@vFu|(qksccdyFCY)9miWo4v{eG
zbwmrZ%!5_5)JQ1W32R07XU@w=dye}8m55E*SHw@J^&mPI?P@lf8fX4Ae>R#-b^~*E
z2C1I=!kOkQld55nrv@%i_*v5-hH9rXm-@1!B1~ao=KwU<7md5UYy~eXGzexw7WG(u
zEW#Gz2<>A^7x}j6a^dco9tN2pAyV%b$)(m~gSow2d6jY6ubl#mjXn8eKHoOfaYlxm
zydNFVElm(BH;FZ|#u6fLuqa7X=eE}U`T<-K?!Y-^b*V_wlqRnZh*HY^4zJat<4b)V
zUuAS+4&8>${^=3N#Y$YF8nqcP!I}d{ZYK`1cb;S$nJG@zE5UbQp5C<u%F^7r(eIKH
zMFK%P<fJ7nKfeM+Jyt!v_D5yB!}8ani^HVbVq)`1)NHVGLeA#c#IfMUa{Y10R)jN8
zzpq09Y1Nr2VLu1b<NJ9R=0)XRou&Q95!`P_c0d2=!+hs|pTn)@8y>Y^y_;=tL@lAj
zBo1oBxL^<LfhufpOiRcyp8gmr%yb8I2kM0ne**Kf`~52eJEd=@&MMZRGf<$P&0+=5
zL|<CE>NQ?44wmf6!!=>w@-m&y3+Jv8y8Na<I{leRAg)m`(JpHCs>@lW$=z(pqqf(u
z*p<cQW;c7D9|0lpe8<T13)|(xcCrJ8kJH-lj_FSYRwz{)-P6P&Tl8#2&p>I*j@%q1
zxR63}IWp(|4P%_j|3}_?$7A`o|KkmnQQ;y&LiPw*k#PyxWoJf|tPrwS$w)G?FUgh_
zva*S+>|{mRdvlSk?{U7X`@TQ-{l5ErzrTNfe|S84RIYKp&hzy;j_2_lPb?+0vWBP%
zGV9CKto`0iC3z<hRbNfPFTWXlTgc()>KvH5DBi{=mXD1-BSE}=*!Y9j6<k@`vJV@!
z_y^n%Hnh&L9%uw7FrlPrx)n^xl1Jo<p40a;pVJ)3p-ni>j?0fYnvpy_zKb(@>HhI|
z<Pn))a`4aWZQC$v8>AoQ=mYgU$%U*m9RtfLM-TyE_$h~4lyP^j#6``~aoZ7o7sO@z
zAi*i($_*s2@7k~Lf4$1;dc7paXkOM_bM(87{(YLz-k%;WhJja`gp8ZRW36h)Xfu6q
zcVUldEe+b<{Gl}lHDuG5mpxR;i{)=mY+M)|5yp#4g@I2Y9^q>6i4_s$<>p?_A;*=W
z<Tz)$9Th0kVx^;h%Vqw)NK_6(nwO7echY{T4c-fZgOki%ry+bMr?n@h*Cjpf<<lwF
zHZCWez_iwP9<DnL46_gYF@55m=QmzPjw}P{7??S<Mr-@ngxZaxuQMrca?i)e8=pw$
z>iSU|ajp9^^B=gpj}nc&$KF<FYi!&$f-{JD;taj%+D}WgMvYlM#Uy`&5$#OXjA$FF
zjSSLAebb%d>ivtE@|VG-OjlRM=#*1zZP^o<YBxMDy{AEtwc6Zdgak`FW4wx+M+Vma
zGCMO;fBKRwVoz*zH#zYwkjGy0yd&rlJpJ%pm&2rwHk=sly<*fjC#;Q#@~@?uzsmwD
z1kv?~{HpiTA1YoxQ20P6lbA2EFiVso`872$ub|oNtX9DwRI0{nbW^pw7dbfnioZaj
zkV*5TFpF_6UZ3WDYPm(WuVU>1i;`Xca6-#NH!`03aevpm+x0{R%Jq`L*Zq~V8Fa~Q
zgSKM39}p+B^MqWjMC)WR&Cqa|4s!f2jyvCw6<XRhT>f21M;;P;PLE?K2#Vr+LSB4N
zZO3ptOO7i^6IEV4`?S-AsQcorvhKAJ|KoJb*i=5qRw|EDy6^X$&h)*Eos+RS%Ncrg
zaV_);-ZQbWIlpNsr)mOM9m(yVc6B!HoYI{W^(d7KPao_P3Da^IrK-Z{d`^v(1M(*y
z32++j)7Gtm^dLqJ$Ku2h!l%ziF0l>M;K<o<fC2A43&=w_zl2mbFV6O|*9IFc){-Sx
zs82p*F+2{&#lPLqrU78@5YaGvt|hs0a5gH__}u7$#z0O5H-X`&@h<S=zTcjlqgik(
z;=&fYy?aeLWAAybNnPEN<5%exuwcMwL8gjWq0w^`D<NKJ5%s)y9akCN*!APpR*;9v
zf3c|DAk&pmEypho#v1L06bA7PS?uwiq_d}zHzfoo6l?BL8glUvs&>N2noC1bD>py_
zmcEku#t!%3HW~>pxalFu3{|vg`gQ$3Lr&?Zj$H4&zQ0xT&s>%}I6)$7vPet<P+sV>
zroo4n_!KW2$OG;$(FD70+@zlyRpP5Ny(u;)u6@dedF~?7kF&IPw?6bed$uBlc4+~Z
zu9%O6G)R-yT^^l&`PE9QS9`g`qWL7C_)jJ^+*sIZQ5y`A#S0MkW_?E^*EA{4V=YqS
zv0wUSW|f-c!uwx%fGsx9_LD890QR`~&E`&bXCY3)bms0X_#E^WsVX6MKvZl3L}xOe
zx`YIP16SWlJ2;0cQGgB(%5CN_8v6tZDhjB}fUTw8Plh;fA<ymuRt^oK3xF%2PGQe!
zq)ujV%OW9J+u-;XrSTf*;z?daDjm1Aw?Q0_LwxGi#kYMdROi4xH{l0cNf=la%XNpP
zeTl(8HVJN3s8fp<=k|Fr=hWBK&oWZ-XsI7!pq$g?JC~Z?SwAQ3iB$O*A-mV<BT9h%
z7xoJPqeX=}tMFDO*lT&Y>l|ya9eI30+Fp(m4#fi2GY_Zqa_^PvKp$?o{FysVXg{b~
zHopp>g8T0f>j#xYL-h@e2r$X;YoxB-MtEe3Vy6)vxNC#x_C(nCz0XA0Bz@j%$8wV5
zgeRe(C95tT2CErch|awCjmBOTVt{hepx_wLC~=P^AM{K4<*ygL3?Zq4nF<(@{#kJG
zWo5e^iI8yD4A~Lf#T3`JhBqg%5FDTJdMAKDU=)y9d}FpRHy*(1EAJG$k6Al{Kao_?
zhkwKi{4E^%cmTo37uzUYe@-vOk4V!75{?tmFM>BU)6|pOxML)^%(Nj-o+?zfM9YqW
zpATF42%%S<TN2wzN~1b(aElosl<;=TI18TDb!breGdZ<tGUsGm7T=!a@-!PugH8mU
z<s>B54xFVo?Kn4{5%W3cdHZ4`AO`fJvcgiL?*>0zqKkL<b%8Go#p2CEV4;{ftE$CB
z%<%YQ`EnII5z(1Y+63SnH{OL&W)Me^{T9Uc`1=Nvo)R`UlEAg>8H`__>uHfHC!%@%
zigwFK*&N}9)nxm=<bNFR(!g{0!p^b0WzduiC5#ZdTfDKUv&jH6zBBhU3T|nrt~i1$
zB!-1T<aU$~!$xNey=>)(7r_K{nw9>8d47^jfg&6DrX4Q<0K^+mReHyCf7rI=BGvuO
z@Jc2<vpjo-qO1qIf?V$i<>nDTQuK!vf=dspa1LVucHo?&)1YmZ^m<<o>W8Gn-oFgf
z9LsO1SwCXxj^nqud0!j<!06M;k>B9Y-vJ*X;u$#WAB6tt({~Cb&Qiue=T4k?nr@z`
zU0IZ|9wi8LBX(3*F_0Bc(@|m>H!@8U$-rJd&aHky=TagraS);5c6Wf`lZ^|p+|8Mn
z-^J!n;#O<63ff*-_oS%z=GYh|(wHIEyv0OAM^ZC}lwuS0)bEFqYH1rIf}0WJ(aoq@
z6rH=qNh*z7OiH5j*L>GCR-X+%_;89Okm@T@C?98rQ#&;<R|E*>47qc(lJ>xqwZBYF
z#c)B7zBTt?`9*(7RS$QNAa2Avb{ZFc16&*Lh-)PS>FVGkDi5AzeNp&dTc@l*dBLK1
zY<hJ2#GpHXfEPu$zw>RZOn1*`snL~5_M30$n2nYaA1Mh@M;@Kx|MBQBxaVHoJ)n3!
zB!%^zsK?Gg8ifPczH8qszSNlYCjXi6_;cR;wd5PhelEu!&K={KvjuCr`qPOe0oP%7
zW_d^Wza-8ozrq<|_50_{^*A2GKKB6_fMcTR;$?AZ3OcULzkvZMWmm!7XLE<!aL6gy
zKAGXR#JbRubt%6_NPoP<C#7zjN`-=E?pueJY14Sck;ChjFvfF4(mn1EdB#@5VMd=D
z6{-y;c5_^Rot|e+#B2h5<Lu|~S}ZF!dTrY7t*He2203(yZpJ?0K8e+qXM+XWh^poJ
z$tWXIs{GC`c#qp3C54!?-w>S>I(3<vKSeRl_bLCV920&06F?_~ac!}9!gRfD%v)K$
zHg$CW`cU|7B*pFDS88W|wt+ogRP|sHdiFf~XZr?~n_p<wdh9Yp$@xxZ`H%O6QRNfJ
zr!Mf1J`2dE&$#6?*yx<+E~_KcwMkjHjCiVDC8gwFZi0p;rcz0~mEy+Sx8Aw|#IR{g
zb?Th!_rF}#l1JTmw^dI3G*y$@2zLJa;_~+>qzH+^8+gh4{dazeqR0gLxgP|5-sF8S
z@}A={akJB<Ag)}xg$2nCXt6QatLKbq6FNhDr)G~Q;ZJP4eo%V=!kaR1=WkoWi>eXw
zwgqLtP4VM3G!@#Tg!&I?&r2YNYAdhPMoDkMl=&NDlpe}zsNMa?9#<XJSuA)xh0T^%
zljR&*TmVQ(;O0}NT8YbF{XE3CUgQ!IU8ogSCsix-ccIh62`HJ^o$(ByI5D@rdCZf}
z^hn?B^Fbf*oDnsO%@Y&$r=stfqLq(x(Bjm)nQ0A=Og4?fgsj1vH#!(%n@gclK<Ddm
z^M2KgCBJgoK&d21RZC(85>fiJ24$#LR!)@H*>GP?DPC=$DmU00BcEdAxUb0);QAgw
zjp2txalT>XxnH4G7jl?g_S>@k%NH%lcb?<>9svD$8>a?O1rax%C8=|x92!`d0%W+2
zZG14(*TzYI#&IP#)WOGLRP?mPRmmXfd<hKDO1gW=gFa+{wI0Y~8zS$zKPC`)Ve}Ux
z?LPIjRj)u!-FL0v73jVM?Hb{DAr|tuyL*MECIKYt2%qiEemQ6qa8Yr^1=v<5YadRS
zUf$AN(0{VywK-lAeMMOe?x03*EELkIK}R<k;c<@N*Z-WU!S3gbKnaVb0VWg&Z-c)9
zm7FQ0p$d;MXw-X}Hg1EmtFJlPx(27_r?Z;%uCP?{DCxwZn&A*SN!uwwJ?j#v*;s4s
z*Zw^4N)SW5I%u#>DnA=t4jG)4L}!GZ_7-8MMG)Zo({3T2<%{(1endYvz@?z=^6Knh
zb~^T(n&!M*>-PY6;$Nw0Bv-^+PtbimLE}cms!RbBUw=-J8BQR8CP|^QAri-u{S8X5
z;XrEP-pj8x2x@3*)g_cqtdiCyfQeAv=_3IMFD)Cg5+Z{Mr&N#x(6y^H`!q}S(U-{L
z{f1!-A}llV^oorq<hvu-!MWP+wJ$L=M{&$&2yqV0D1|rD4s+pF01{_;BG+mrD);F5
zGaz4dC<Z8^1Tea-G3{!z=STORjMrak=$`(O=g738skHqA{j?e7o5!9Xv5<kw`<)|M
zn*V7cO8dk|N{KB-ZQhfi*;CdOfmZ_<y%}`bFtv9T-bpynyB#d|VT2sTHitSYhs)io
zhj;hCiv%|=Vz*;Egr@C=Ittu4H#gj?g9R_Jqn9dW<2LKWk5JkmdCVr9%Q7fl*_8AS
zn9x+p%5WeusltQlvJe@mA__75L_IaOo}f;vQjU6Q5r(^AEK=tfDW5U75l<>?IW@DU
zG2{fpnInZjlf6KPHid+t45374x+##*K8HA&`94gq?7dJH#jl7~e+DKa8%A0^r?BP?
z&nbqmsSS4!+FuX5tvkS6JEggABOp;iq3xKnWyQDJ3@a$MJOVT0AsT4yD}&MVH1oHT
z`xzb1sqR|4N<<8|fWc+9Tew5s%o3a4pl-(<mfEZ_U%{rnqM5P5sP8gtc1m7^W%SMW
zjJIN>ulL^#T48IsB1$JCb-lUR%}$vg$cv;qn#-=eKKngek1o39o5lJY+W}m=*C;Go
zKXp(&{(*c^a_F$d5kC;taD+}9x(9!O$E<SZ*1S3NEUVaSb{belP1W>=Z4^DPAU@q(
ze^!(>*fleE`Q)13OU&Kifz!)Ii^DN@Mz5T+bEoky)@*lt<|G^|dFTkyqeD2fhYsV4
z|JxtM$1gk@$+Zb&?)~du-m8j>51FqYedld6N^i|iX9-8On8E16#Hx&(*CS;UpeS7T
zIe`qFD*O2M(=V#Ti(rtHh^ngQk00W3L`H=y-sJgiPMwK66NEbF01#DFnCvw}V#YBw
z=N=6~Wgp3KEf&^eHStRy3vPh**#$gu80Jh>PM-tZT%UnnI5f1v&_}HIA?XSKl+e&M
zF#eet)B*E|B!HhkxtglI%tXQ=xVq#ssn(!GU(kM*;W4rNwX}fz`A4!MIwdsf1w*Rs
zV#uNELuW0Fh?aoSCfXASI^>k8r%MXzY=<!oH)sMgB(|D4zu3U0$@mPdcB<h9C|$<q
z<rI^5)^0{|Mck1wE!K-f6(74X%0hf!<fDaAN!p#94Z2Sj(VD2IrmxJb%-S;Iw&c}O
zy%u{EFlOAC7f)iciE>iYZzQ#KhHZIMWb&iD*9SL=Zl{;^aTeb`es~?1yOCYQN_;n(
zV6B;vOWTGFlUSQo6SNbjX~eK<5*zj%vsJ<c!H=$#fl(jtQZ(dJPJ^$0wR{ej|8Ak?
z>?>xg?AG}K2r8NL887Rt=-FL>{Z@&j@kG!S`R6(LYwMIss3PA=-8*67&EG7#?+IC|
zyXe<ja)1F_CUwM~J_(K|YWCx9dIj2HwyVCHYGYseJk>^}wCDQ}E;&>7j_erCfqA~%
zEB12}n<LyE;3g2#?_GP$VkkqeD!5rOUQ9GEFhm<}sLtau|8X1r^X4*j!-;$$H&yfq
z=hw|9e)Pf$G(}fPa<|Uw;%XOz5b(;b%7DSkX=W4sz}Wq!1x1#q{AQ162eS5d49iE%
z*$WZ5`BNKBAf>iwmzceR-I-PlnLrvYlSIyMps^oS+?#B9sr7LY1Y8+qt0eBJ+<9@l
zxmVgNOt%lBX+O^1L5B(xWRr8nIpj=Z$f+EmhDw<nBwT9uAW~{kDC4g2Lu`0A{7%Xh
ze_rbz0t?UhNb+j#)g<1UN3PI-h>(!DZdJV$Y)1_2SCfP~uFH<JQoZD@d0apg+Ac!o
za+)dHj;Y(0X-DpjE}k%7&=fyQ-hQhh$?A3$tAPU=$Z_pbcdDYN!u{#@`d>QLh=Orr
zyNJoQ4G))}N6_7OQS$>5G*__aW<ZcIxkv-Urq2~^QSwFfh1#4g2+)B!#K$)sJaw@d
z(|Iw%Wav$-&*=5Ga<_R&x1!_Mzm5t;Z+*f%XwSG#Qc(t{FJF(<l)qNAY&~LT+Abmk
zuRN*fHqlC-upI2%_;8A!r$DDZmml2_OtU;4<*&Veu5&;@CClWt{+=p_;>e8R-H2X4
z^S8-j(-Gg(0K5{y5=D+Bx$EqzxQRxGbYPycYcTm6muOpJ<?U2mQ+*sx-%=F0>Eb?1
z2c+IegWxIt&xkU-T|{YT6VuxdDM|~4QPc8M_1%a>!UNr1byi8o&V?M|g|04iJD0i!
zQnIB54n%9#2rdpzmG+tC@*3JSwL5b(20=*9C$Vmje*ZY=5x$(VV#WK(wux;feOY`*
zKY8;pp7_3gmgr!`)RcSF{I`T<z3tGd;OMiTX5B-$B6eO~FnIa($v<z#f8X>T`(8g(
zWANcT+J*4v-DygSV~ck{>0@*DF#N@BRyx|6cxP98Fexr7){<bt2E7Z{8WwB~f7)H)
z;&R6dxcnx+B0?YQc~zUwP`Us6Tl|kLR{hx_c!FP~Q2%~l<e6Z%z57tz8$P2lKMu!2
z#wLogB2j0?+=Nae>$T^@lzT^jW8p|<)JGo9+yFi{-&{#K>pmE?W%qgD;dmBxL^J>U
z7XB|^SA3sDd`0ue9dW{6d%eQN!#+RM1SNcQ)nz}!RMC|T-}(!TTvH<ne34VZ^eGPM
zlJ2ODq^ckm%uan3f=)ep4`7;`)~=IgV<1)3Wd_@d`|hvU@8(L(uq`-s#EH7&4qbfL
zU+XB7SP*b(QrEy?Mep~w``ZJC4u@#(jD$bW?-NKp_EGI|02O*6!v6gaS*{lbxce-t
z!{MxIdH@iy9q3Ua7Y8}25^RV=-mb$;&kk{Dx~w?Nyk030k5tYu_o%j^c+UU(HUH+4
zyj6!wQLIjgZuoWj%|*Sh^C$WCM!O*YN*?$bmqGlv@njgX?my2&9y|`n;|hfMu-RZd
z9<^etZ+cK?@f4K${>v5qk3aHh9l<@IT<H8{^ZRxF?kfDpuOaWL6@yE=r5htnrAzlW
zC-dL_%I~`^p#~+s(s$c0DCiGo``^F9`z|S{nY;J~`tkqq)&Fll@~KF;pN|Vae0c7^
z78HN~riXAMnczA_zW&s3^Z)renZAWvh`&EEdHfIS{r6Y!Egv!51$==5?<9T|K5x5;
zpC&cU7PRuTY;~xfIP?U$_M}%6@)j_G^+6YWgOdLACXZ8n;R>YQOTR?=^9pp9;DOqL
zYY70@W*%FgGdG)>v!<TQ4_<;H#ATRNLl}Z^AKdM+uT8`p0qc;VjSeHbOsnjU1IB=3
z^_7VxEu7_VEWoy+<bGjeYbE<xow*@{nK+6#wU5a|hWX5Klu6X92X?3bxK`qmc(hwt
zB5FFn9vu#QoC!JE3qDYK<_z7Bhx9ip!3PPu;CSOs<K>tP<z*|sMeiF(sI2i7Y)aN=
z&FFGCNqwdJ6eVBNEigJ;f=OJ*msAKlDl`1xyS+=<0&%=0h)*|%0r}-jLw$|f_68#v
zMy>Ec={ttnR1N_CHX=trMr9gMHkLB8a@fe7d-`&7MGFMp>p{qZ6)-WX%-V?E<=y?x
zlx_C;QU|8C*$z(!`W$umo2U2pYKm{b`f0k~kqG{Ei=GO45AXF2k~Y7AgNdHCdpz2+
zN*GQVq#>Q&<pzr1Ws6>pO@IQ+kPOuuJB&8I-uB@A<pBH+wpmQXyKz{rO|NPg;ZB^{
z^N-j$0ER#bK!2q|c2gyg_7Q$pGL=o^Lq(?dMYtp6-?9`3Hq_<|Ec62oCOU+)Qg@&S
z`v#zcHs@Mqu7QE;J<CTF{N~|QfP!0>eRb4IboI3C|8#cXP2FzAr3dei|KW~LzX6X$
zrk&o*nO}R}RpOBMi!f^6I8tsty9KP3kz-HoS+~L*FZ$mH63u*><FY<6>Er;Nk~_KH
zgU6d5T2E=Q$Z!~@?AiDbTiu?edb}vFU**|iM9ya|tzKVh)SlUJDI~SA1y)`y((Hqe
zTVuxI&`lOUU{x$IG+p$~WWevgJ+_b~n(Bw#@+NQce%<oQ7Jp4EE`wP{s{RUdup)3!
z%rx*0o>8xxqxLk$E7&5_nlF<H%jZ>g*;5j<47@#BGwgX!w^X83zT=DPpaq9RxVfi$
zUsm9zv-@Bj_79t1!YUDLb50E?g};^`&gTSfJM@peg3mT!Vr1XKQSf{hFxb7Kb};fB
z!)h>vs~y1VF1IilBB_abTj{qLdsKEo^rUB0>4ZkA{CW?3HB5Lnb5RsVqB?>m?#xbN
za$KUzPC5;nnkkzG2k`=a2cbK`zJg1YU>1_$x%goOahF*FwAlTo$`?Aa`V`d#+<kv^
znz9GRtGY9CK@H{S#xd5xF)6arm)Gqav8~ruu;pZa&g;~bj)*a=nlTT>RJj3_!^E5k
z-S%AYS#PCulHSPs><k$m{hHmcRZS&jCeI8&xTYFr;5NlV=cSa}GF1oDRCYyS%N~~f
z#5*~6pb>CF*IsGTn>HdIVq7(0ctzgAbKq7)UNRpI-$TM<#=7LjmEOrXiekp~i|azI
z*+yCoiy7>~5}q6P150x@RJwGO;}*KVo)++NGiW-%vlH)IT|H^(=TgRe!>8iJQQe-N
z($Kl%?PBze=vDi(TYgD4!nLB*lKAd9uuM*WEcqXam%;`ep%Ak3bL)HzCZX_fmVn*g
z4$TDMXm-W05g|K?GodRG7n7;LZGIk@Oy$rEe1-H*chg=3y1xaiOe>&Sw9hTVFOA2!
z%&~5QlRydhG}a>y>r0?*-5h&y%kf)}pO<A%X2cI-wEi~mscaFyzaMl5tBo{{NHQCU
zQk^Pf=k}uS1lfQMqvhO{*rp&zUG6kkf@)%^Q<Bg=j9JNHyfN4jVKjZ_6@WooF(8Gd
zF#H(EP^6PM%&+xKOQO~boD+6Kq5&PmP9xc{gZExqeg)O|5+d}80lc%`2t-m<xqx-1
zLKbA<Ab?iikm#9KMQUB}fH{8htn%AasRK5ksoSwp679cnT5#mhaITYnmow4cJ}^1g
z?M6UuY>)6oUh)_}BX_$GED<aCv15*X;AH`pTx0<$0s$Yh>MHO%K09%Oy6PLuCamDV
z&<`nvAfL4DH>DS;1O`l#r5?x3ou=Taa|idTM4>*Qs-X$P6L%4(u3gyA)iw!*7r2Yz
zxY?SQA9&#+?$3;B0R=0S6)-fcfV#f~_>?208f54iM`(P7Z#q>d#;A-?FW1DbN>D5k
z$1KOt=&)Ds^yKd8{Q#q&8N?816DGGM^g)6btcB}`-tZ~39d0vg3qZC~;6Ih{_iBA8
zP)`5$<n$F5n^&5<u`KE<<eq*_^s~VEn<3rN-Cb(os$SS!hkbWzW3_yEpL5wm179Kg
zBYe)*W#$+H)Pq-$jUjf<W$J4Q*adt8Bu(v$%ZYBjYWXj_3g39yKk^t)cq+{`pROzb
zUEj4I-H12-H?Xzx$W`{0kg|yX9L>sC@m;0zC}V>dwv~6E29^F525fUMh7fqQ17{QY
zw(~;}LvFn@aG7QF%wDF2yq^MuFSU1&oTntOW06`B(9;V&!Q|!=xS0)Cnfa4!?8|g?
zd1NnN8z87Y*s<u#yCZYiY2|h?j2P3(R9-G)yCnr5|MyMxFmAZQdyhj5SASI{osV$A
zmh&>tVS~ahdG>2lMO-@YFeC%{@bVlGf4hOH?dh>pIxWrgJV3FK7!$&54s7(hXj2)m
z_iWGa_HYVToFTz%F$20=dEwT}Lo5S;!Mau9xm+N@HSTc*Mj$;2!8Hi+{O-<8D0HCL
zhl$V1DPzm3!n<~Xgk&q<rE16e$dTb?7(g=5fi&dq*pnu2^B4X$ZU^uvpW6pxNNr0-
z<2{Jsj8sUNfqn6M*{cW>efK1IDA7E`FMlAc`nkbk>$?}EJ0#v*m!&}ZcWwL4FD5kB
zet1@Uyw@3`>%|rMte9w9GUK_DwS>ESp@Dz--Hk-V>D_i5$r56AKje$3k<CG{7k`Io
zEr0{Z>~0b#I*qT`q2JI?iY)P!YVc%x)`}j<bEyvW54V)<tZm~9Vd<Vf4m><8;w-Mu
z*n#<(4AWhL9@R#WO>-iW;&{gx8MrJCd@wlZc5lmqbP6sWV*lw>dnW*BQ<pTGgVdiG
z$0EjtLpMq0q?lHoDRQ5A1>*XlD)8Os@7fat(kOlQW4Djq^)Gr$OE8c6mPvGqnG)8q
zQn7{7Y_Hfpt^w6?-A-G;BK{I_nu(!NTT@-VN5%C*3U@#nwC}0!#v(DdNoqHQ7_){k
z6il{?m6D3B=P;cE!SM|@Ha1RoMw?r`d?eDPBQKwtzXvqA$xXsG#=s}vrAf{)iC*Bj
zlvJHtn(7!gbb>#%>?<#OJ<rG2W%=WfdEw9JSmy)@3RpWsm@x$}5aRBg<2HQQ4V7ny
z!LqUY<<9<7G-4t>i(op>kdGE$v5;Skj?zV24Y&f!od_e&&2uSx`0_eukK+lFZ%pmk
zwnO!EAyZL1S<b1;r{{wGSLt^@;GlDRn7<SQHsPv!JvRRPw{(HwY1Z4Zf2iFK;WRTO
zR2MS;;~qbRch0tn?m)(IpT+vTWyLW_*)Ht(9K4DGUAM1u;VRlCm|9q%r3+-V*QR<d
z@>pNi-fW!@CRfD3WY2Z=BwkIcosvjax}AUHdU8b_4ayi3^|o*aHM|Gh;FYnXSb^|-
z_H5;bYk5X^d11;*Ng`P3@ZK1~1DI0#w?5Indww*3ha6wQlvX%Y(2uA7d_WO>#y)?a
zUV&*%JWc{a8cc?1v~adu9yg(^J}uj0nXU@CEO<d7to(%@z~);BNZXzm-{gY)cnVCK
zM72JT<#Y1Lr%Fc}5r6rj_J;=NT3cSl6QUwSpd?*2H=Ca9f6=Q<LH=W_5smUGpKAL|
zu*qlL=4#N+Y<k_?J*K#A2K-@K*qBw~y!ndgwgS*^&PAe7a(o=Z?SkXa-`I0pT7?fd
zeykeBH-#eCod0VBbuw4#DN%T1u0JI44n6k@)hiR)REpwYENLZWk#}RKGjwvy;GN27
z3s3~Pd10a0R9X7GsM<b@pD}b%m5O#(U2jDbP#*P9j>ttbaW4l~WufNhMWff-3B7E3
zUW`~lUbgm>S#`uCo$()VrkZ@qtt%JgGuKDC|Hv&F!>>lqzG9?vlAX(yN|EzCsY11k
zUC|5rW7}j4{Rw*%?Z&TKZ$nDl3yqgQQOO=R@lh#>-#HmyZtT?lA!|A($(!xT9uvU<
z7df71Z&F4Vczt7$@lJMlGO`Q|&*ts3c!~^SA&#^JN-+g`B`VKnBx}$6k1;4Wky817
ze;Dp`)rijdyrO&&9T7`IP>L99E-6^>6s|Ov#ysaxPbR+|2dAS@OSo%ttg~?<x-z}4
zwq|O<w?QZA44%@&VK&YDm1KQf`VJ+OuPNt^c(HG}1bCV}$C5CM;df;2=|+^Cwh(z^
zNAPNP^8`VBOu;rJNI2P~)LF@$C)Q9oUopX&E1)LCG3leYn@~YDrGS!E;Jf%jIY&Lv
zE(tzJ#TLY9dMfbQ!QLGg<x}2PW`GyDCdOXlBJz>yO@mg)X}Mavz?VMcgxb!*_LU}?
zyjMf?@mus7x>KpILZuzFde<;lbXQTOus6iJ9m9%yzz`Q4Y<N$I;MBrqcdMgE>QMct
zGlT<denBwJ&HeLp-ghi3IwS|r{h`P{bYv311WnGXbH9-i|3^5U_@WrhyRWt_99Lt!
zMN4r!%b0^jcM@;aiu;_5OhB9f#FXORl9KToup!IC2o1_xHzVPQus#7in+eO@*Xqx;
zBodF+n*eK`pMPGIxq0`MlmR9~0Ov@3|BEm24{kjh;Zl1_Jn)PLOg1!WVlKsLe#ELy
zM@TuZsq^1$B3izvc>aT}J45HE{vQya+cvlXXZAqAb<f?w;hoM9RQVgQt9{}9fIn4r
zF+Jm((UL_~vs#SE4&qrUb`A@t5{)|m@%jra$Jf3_vuDIfI5&Te>K#tH7O;1aAW(nv
zf_Q}7T6jcOo9bGNVMe9mepCr>pBwHr%8PBmv`XFbuNA8sEe#zyr6U9C1hbnj<lm}4
zHN`8D?6+^D@rmx$JI8%nd=7)kv=~N01J#_jZ_K%Ls2ebBu}2tl>dtiq(Qc{r4?3xu
z>Icf)dpZyE_~|oy2wQpe9;#!5TH0uyR^(=)d}Y9gqfbVZjP5RK)TazM#gk**nt#I;
zzZ>EIQA(^4Ye`>T^c)NOjoX;s#yJhj@?vnxybMIp!nk6KzIJUEX(YxVb#0^UveJ!q
zZ<2Nio!66Dry}?X#o8;5%>^G;-Mn}|o|ibTfM((wuuLPHJ4A#4yME;(Yd`gg8GG|;
z-{N5L<|IdoY2f5_w;4sf(N-$$q5SWX)x|w^4DA)CCNEkHmBsqtJ1{vt1oQS5-QHU2
zD<mI&chV^lMJr8K@EPpMy87&460@kEK@Q75PcenpwXrzZhiS~$yQaELHuS}Qai9lM
z4H~Fz>WDM;Qb<U;OF<LcglTHQgs^4EYCEU4XRVGmLV>vCh#2*>ODQ|=-af)*qMs3q
zr~b7wQ4$+@>Q15rNN00gfq10{dDGb5Pi=5><c%I7h(6l^k(r+`kD4&Wd?pvrCP&x|
z%Hf=8uU#(#0mK!Oy;Ai4tdfv!PweuiWWUcMePV_~r$h~U`X~hs2ZO+*XHP$@;02F$
z_=kMp`LyG-wM#hl_7f0%T2P#rDD<2dB*`?pald5L*W<5N+q_We7@z)f1vZG6FhZ#1
zCDGt5r%|7F5#iSn88`}+<?A{5L^(@{scga9&r2gu2pyP?JR}Pm)SfXliTkp*;|#81
z`@5XkBto3E(WLpC)26j&3QRHsZMCN6BySBQ7Fl|7j3};5!&GEhJxY;i(ekqnw~aoV
zft|!Y@2EAK67%SOqUZ*&<>bqwoD-eR4P##6h3tD|?nbzf`<Y-I<&k`%ZBc6wpx}Qw
zMr|?fuHX5D_4mx?f81@qZ!(rgfCC(%Rl)g*d?^Rwu+eRWZ5hLKW!*@OAbSR}9OD{!
zA%5cmC-%X3y)qK3nN4uOAlu3;^64zjix)6~qMA^sf6Cap8%%C3x1y!SLnx&md*N^o
zWH3eDtoXS(^kVw2T^fs^?SK2As=eY%z0OCmqPJE0vZm+hzJV$~?p8u;tZMzOAIm@X
z(Ofsx;~9u9GEjs4oO8Vus?xlb(YISK;;kB-FF(ATIQ#5is-o#yKan+z;?=in<B#*f
zaYfC@RopqB8{7NNy9}(uOai^oqM8<nA+#fys$}v8TPy*yVts>gKpU#?S3bAw$y~~%
znd<wAxO7h|w4Ly8YE3q+_!v7)sfm|hi|3PRd80dZBp1bTaiN0zsl1>Ym?3ZRvws=}
zP0c(gDwi4720;Y$$~p4kl+7oC47<}e7o=}3V_f*Ws(+jmck-NJ`D9<kaCnV1Ep|-o
zVCxRoiuxC1Td*VEJBIqx-ktD?_$|$%?C-OePy7fIHw$}8uD?c75Vf#0zLLfHPqrvQ
zP}Iybq*y~AYjN^~x&Bi(r>i~`Z&=TeoK{I>Jj2sfakjVn4WUV9K8EhJMxKDgA+kF%
zIqgxXu0AQ=(inwF_WQ|NZjg1JQK82iV=ZSGVi%$5kuj<(>C_CJfT#iX*sxm3KsD%X
zUO2m84fW5AsbhO8%%t%H*az2OgKp}(m$(TN#gRTmfYoh-!Xz>VJSQf=g75ea17l!T
zT=Gin6y_q&=zZ0^XA==0x}>^3ftE>cT|{xtiND=nRbnTmQ}|<rdYr@(lh1lKsREYt
zEaqvi-$qZl>@xo%Gt>D%B$<xpijHgR%IF6*_srRpoK3|I89H|bV@yk2St0!bCIWxf
z;z)bNZqoPXkiU9uN`t#?53aIgvP-a;wc5;@ks6$p0gFc`|9DuI;x;Lq!25-NKRzy-
zpjR^t=1k`?Y57eO{8NSc=w>7u5RAxLEwtJl%*zbV7w-kHp`19h^BJ;Aa5Sa@IiM?1
z=mtS!Q$?t8#K%?^X%54<0TBWM=UB~@genYp+t!xc6Kgj(@%TM(E90H2!&{9OntD7~
zz0hAU3L~l0S|zPb5|lEaqPdmR_<F4;$R{WW_$WTIWBD;Swuwn~w^Cc#+jOxD(~Zp%
z-+iZS&ufTjst#ZLYJWHJIz5&r5>+7g(0ud(J@z3Pk_hT-39Jh}cQz(Zh)mOujaCXb
z8xpY~-NW=1HHloAaxe7xvA(EPvRE<MT&jIq0c$&REEC1w!*{|S-B*#`E2I(ybFMr-
zLhBdKw1Kr>^B&B2xM&I5RQu9^PzjyZT3}xD`86zv{YiiVeH$gVcUD7ht6iu~;ZLam
zg<ErV%SCUefs&hZx6V;!dynLXE@L~n{@+SUadD$?+N~DSoTfWLK9Y(gh3SebBWId*
z&x&$+ltKz*zDKww1zsvcZ71O-N{N-LyNz!f(52^y-pzoy;CT*X_%7h6w)jm9TVWwe
z@x1MVA6$1nV>OgyoIi(kngEX@fU|TWw6$?}zO(v4xD%JtcRc6O=r51r3P5(|tP6CM
zjW-^daXWNeYTG-sbV(H|jF)<N8A(GQsBqY(b~ZlXXgFO&F;^dq+Vi&%QJTicITMO)
zB}_A2O84e9B2i+aCqsFeSD2$7XCPv$#bXha!GO8Hma<!Q2y1V_Xrf%2e8E#2Hu^bd
zN<sqmxynt1ilKe0)BOvxF!bIu2!qPZxURX@z%|!~Y~WpQugrjCKn3F8zc<ib5fEJ3
zZXAC<zq8w;H5X)fg*>Y=W`R1U8fee)MQ<MSdGH}x#r^;nNxUy-(2=~%pt50KPP73+
zA~ueqXS$Vis<=UIEV)&^%`WLIRMz3L*Rfa3M2w5qCP$TAt?psu-*)pdyzZ$=@BsN_
zZLZ8@FxBPu4o2rTH5vx)_K5D{tu#(}F6BBn{fs%wpPk`spzy{@Y<g5~@|{L3x^x^F
z?%%BPS-8;x$z%@-LDYD+Er53Gb5;8O7;vOlU44h>cT8zQpMI7|lzDkS-`~J*)t7yP
z^Yr%<z9<pDiU-fpmuD+b%h(i$kAD~hAI4Spg40Z1Ipx$(a9*4m72+5Rf^(M_K^0ja
zv-I>zF4PQPJwY8#X0DEYFfZfkU&yddAV(($io8WrfXu|6CwV;AoGFLWko5{wO_iFw
zKAaKJN?wQdXL8^1e4hCvLK1rSw--QiWyu_{AcBV6H*ay764br}IH?Ei2!RRDrGw64
zw0jH`6jO*^pt|M0Ccb?sLv64A5a@F<5K9?NNRqkt_~;3y-l(%uL*CtuD=N3XaOg_J
z1+6tiPT0$CFbZlzed5-ykB8%R3STAjKz1SRfT4$w8e$;Y+)UM+iPfqlsMgY*fCMcK
zCxKFTnf<)7$f6dWa2?tzx7|2Q<eu@ACk&@DrdzOaXK{jMMP0wI-=QN^gn{hZ9k#?V
zgsW9@?JusD`o0_U+SBwrTpAxsFyP|N<x?s;&CnZmR^|xj1~4>hFAsxJsD5}74=SrZ
z4CfI@^H<y>-%bKg89RhGZVUwr5W?E;Mokq@>2!T%X$;!J$_pWc*!>DofLW5EDobrp
zIZtj~9MVW-S@^=hAm0#N-aJmrc49CLO`*{miYD!`7$erWH6kadEw8a>7lx9rap;wB
z7r4~tv>1aFigX!c1IJacbRcDfPIA7ah4?$9+;qLx9o?W-)Zs62%u{DmXCCDu|1j14
zYHgoW*Fl<F_9^F4g@N4`LyEZ6eR-@rR>}3gY|N9jy|g{IMB8~gsO$lC>8C~?&x^Sa
zU%hNoVV%m`1=<h`QgZZ(nqlD2VgzIuwmv&P0sJ$?zgAKhb-GzNOC%W}<0}0J9>7PH
zM%<O-)nc*+!opNbmyIfOYs)#^2oWy?HRsVdsVx3f+o5UxqBi$G42>R%3n75JWN_Ip
zz}<zGHdttf_|}n<l^2b%A@9C9=*Dw(xIvKX_b({4&y}?|Bpx#!uvO4I9|}#EOcL{V
zxv6b@#=u~5VpQou!pb?kc%I7G*~7mnTSWJPQti)xkhUVYZ|7EiU-L@t?AW@5dl1f?
zYX;fv0StQgp^0U%)<Yc0$Qc;?PiahdDR5oixU|mSr*aOY&WYyQa@*ivw{9s{!I0?h
zv7)JqaJO<fK55>g2l)LvNBu*v*ABkoxU_JMKmVGgY~6(yTB>Y)r~MyJ1YU|-#B77=
zM9E%q>>am^`i-P67}#y)ecm!tJX5Rbkbw|hqgW0U`OgZ2Aw_U5Gw|R%`))bbFHMPE
zbkvutW27o1yQ`LUGdl?x7!b(YUliu+03GBS-={Y&4*``ku~BcD4PglB(rUmw3zN{e
z_u6)`>3(jsMK*<VV|cdCOC3#=vyV-9S|>u#Z3{}_9)_;6+JFj$+kDT5=Vt#jG7EIh
z5h==k+g5~Ot5Ll$@wT0Dt*0ujC`VM;!So!XD*mp?+q8jcsYEZdKYAQ{a~kE;7$)M?
z&@wTwt6)sw8R;G2(z~@fKK|uJg%F#8M4n!T7;KLjp&Y!H9MUVA?L%C|E^cp%&P%8Y
z4nD{QHXEIA8AycHw%1IlREhTFaMPFZ*WM1KDv<xs?{XyYxz^|SOfmRzqK9B4mU9Sf
zXVu&Z+?B2%ZbG9rLcx<29d-iEzxKSpt@0m<AH(HH7HG9Q^3NJa>V3tc;O^t$&Hf5{
zq20%WnDmbZFZ&oxhFW6x-!CHcEpvd3+G{6;NUQ`&L{rx^dMjR*!baJKN#iup{}H=C
zviui+%Se0^$~r+3&+Rp)ARFMl%4AyYNg1#zzL{<<2<BCitU@#4+wMG^>;9<iRPD4{
zXeZj?ZQm~O`1Rz=C}tCh)WV7s0uqN(#LO5B93Ki_R9eJ1v0mizF!Z`$l5}jdnLXEu
ziK!dwe%U5=5SohdYj$xjqiZl}3?@k@Ool>ZojKOeDRNIMGE5@gFj~bZSf;iBL!(^`
zhG`7zu2!|+0p0<z&gs~CzBI*z(Z(UE`%5rlbeN*+2o7@=TDM~;sMPBt3;lwb7Dpz{
z)<3|XFLuh3E+f|Y%z!71uVQ^n?58T*{$fN)S$M>~Ji%5BqjBLtn|ob4A#JO7jdTny
zPhiqmOwJ2@NG+DLQ?TIT)u;_>6$mKp?T<-l?l?t~w6prTjPGiu!DSd*Juu&oKHH8V
z;>mW#;?8g=rWIR)`o+WQWuth<4eX$|XAUUn+NFHVHq6J9P#pZjMa&3v)uBL=e1(ML
zN*f*tZ8^dc%bM>IV^$QLE>zMsoqAK%$lJ=B%6nBqq-D@jq7(8mBd``s3mMzr4^*{Y
zpTEmI`YN@EWO%vlWjYNFiN6r`8D(zhm0Y-^UZwN3gc~{}iGHq_x)SKqtA9PCyV78p
z9eX3jTpufIA<pW-C3f1SamHDGs;I+Lv$OR3i=12KJNs9c%R_2iQ=|72yB+g}%BMuR
zMXpS%4BLnHI^XkA9iGN4J~;I!o<lmrVI1%Wk<<O%?}SUyMvdgAG^8Vq>upvB+Qdj_
zouBM2{R`6Ai4xLK(-J0yiF^7&m50fg1~HBwypDPH+_vY0LHT;1*dNhrN)mXtKr_C)
zW51{#;s9b=Bpj6~raSaFlkPTMR;6G2h<|r2sW3r?4*z~KJwM4=kiNNsjey%#XTorD
z6=&?l|7-5nAspc4&|rTJgbqD+p-*`59G@cNZ6<#9WyrA^Ciii*gczsFjerw>QMXKS
zNX=b8C0)y~V((sqDXLmutUCHJP+hk|)>h+!s4jC}T5zgRbdP7hyBx}>;MTtqC+fNr
z@(D4&TaJ{7B`0YvuCg56{(Uk3?qWRl6$iqc<XOfY+@A|$>WxEM_7)Octbj4E&q53A
zEi~&t*jto|lhMC01XB?l1`bhhe*X%l5;K8a<$B8x-NE94vaj0nN5dVC$iI8Le=L3`
zf;F(6#VPxRe`)F9K~Ux=z?j%u_5cw7yfNR$#U6k1_aT`X@E+WQzCkEftoI%!0~Cch
z*Ix=?zUcj#9;JL7me}}xGTYyBqyK(?{Pnj9IY8dZV($_;@v9a5$7lbIm;B~x{a(?3
z{)hj+ziqn15fxG)_ekw~6hAV)A9SJrTE_I#N8of0SZ)IA7hP&<wmfp6-=NykDF3iH
zd=d!zZZyFN_47a7He*PjS%<i~G>*>@YUlH&!n{4bkQx9(dnkrLY*S<h_ime+%amP|
zfrm4Wmo$5r!lC}oDUF%SjSZ5w!AM2Fm8#ejnmxN6jy6`aHUWso<bU`{hwhHftkCNF
zql8;Ry}%}tIB$^0gUV=$bYe5H>WlhcDO&%wbv;fEf8-Gx)s!UW!vto055SN9A;N2+
z@frED1XEEvBnq?O#fur`a}z4L;p%3>zNe|@5J?HNpeV%0czU4=p)-Q}?w)(#BzP4d
z79w_$20B*iE+Ua+R{h$$I6yxb(T9-6?k9y-t{ZccH=qwxAL?@AvHB|L2-Zm{I36f2
z3+qAfv7{6YlJD?`Y`+~x^U&xj0mJr}mpg!SZ^KILS!!2awUB;Y+Ww0JzguILf~zyx
ze_fLQ*f7#<k3dG8Bar0nnd9ATD>oh<>}@Y0h8jp_*YXpbqmb8Y3#mWZx)3%vpJK41
zqE}lsROwoNGg~KOy;XkD25Mv1zf?YJ_y=|J?mqx7{R$k`ui%A@Ff`@!6EA6Gt~>ED
z3G}O0<5*R!N?(RBLv?0^?tDXZL!HY(!I;yBIaxZ;7V_+#17fwW2eJ>*$9C;PjrI%u
zd7nO~ug8XH>0j7Uw012zwFWE@Tgrmm>VpAv=GW3&1`gc<X=UcslvU$B5!ruu;Ke2B
zfZm(<oOJ&e4^RElVZ5&sX7WjS=5mM&898ll>c}$KHQT@>-T_Rs>?80WYvoX{bvxD^
zBX{~~_T{l<2-+`A98V!8+tTqD$=dmJ$^wrKAW$j?jBIfa4%oL4KF7DYtXzG8KHsn6
zrFKbxNn5Rs)h_`VCR6hyJU1>rS`8nvR)zH`rj*b8*BwecnHD~<%1L_ZU%N4oWWljz
z7Ur9!-2?{{8V{hs%4VxE$m2a)W8h%~2{G7Ft9@b?O;Qa{(cFcwCw^!g&vyPup-);1
z=?jXv<-M8kp5L@P{xO>P&s|WR88+P>A<lLDpPwrv>``q4TU}fb!i3blhcW&NgW-u>
z`=Jy$lK5Pn#Mm|DQ8f_A7zLV5y-)CfhAMSYHH@YhdjmF;AxlqxA?>ZToSeI=D~5Gl
zcY9Pjk><)AMP^#?P9hRR>7!kfv$EG&+*zG69Z=8=l>R}nlS_cdPI~~E!{kf^wULS<
zTYXSx{FWX5WEATbf#pXrMw_O_3kR+8uS~wvwc_P<JI&JX+A?m4iuH|c*y}5@8H;p#
zuwvQP*$b9eg;HD%@7L}%8;3Rf8{dBYT|?l_YK)-oLAaedO|49ankr3gY;(`%7H?L~
z=?&)o#5-TfR=Z{Hkn}g~(zI(Zw=ODI$8Kn3vdHgu#NMIw)7y2J@jUDI;G%AiXX*o`
zM()OkXia>q-~;7XohL3eH(n}eGs5wFTs?q0&o+_4H79w=j`w4b+s21;PX}MM^mgXn
zwl=-2$4njYiUhN-^3w2J=s@sv{k(6LkC?@I7N2(ytE1I59lQ@Rtz!twdS&FslSIBy
zGzi5l6MR9h7Lf`z9jvIh+jPoUlGRGsk1zV}6qQOtgOXOa?DQD#PtuQc5u`jbkq`LQ
zmFh75lLnUf|AhuN_+M#Yj#O49iGx6oacAj=ePWv?Z9L-4*fRKSZK6GZEk2G+D3^e=
z5=<`liU$0rtpF;$e#?*K%PL@eJ1wsq20l`|%p@YTRI9zM)g>Qt{S`wT#BkwXcwj`v
z-U7DN3dHwjWj25bRu03Zb|@>_;$aH@%nDj$8|drH=RuNhf#hojAhlPO2W(eX>ocFM
ze(Y~L0zxHsvk!>_`$+{ePaqXWHYoJc>S;vsRAsPQk%Fb&+651r+3EMd1(=5kwKfd#
zNVx?2fuZyjex_jG#sLF-EC-pKWn>I<O$~6mMi0QbX=}AJ+Vr3$AZcnJ86D&fyDp2O
z1>`yq6Sc+j;2P)%9dRjK#m}sD1r5@7TtH#J<ADE4&Ad8%mBn6v)){dNqT;#xG>lpv
zluZ{E7%mXr(-|jWzVQ&?%%XX)f1na^daDc2ru&yIG%CQ<-6F^QK7WVjMwWpyndja(
zjr)vRp4iw^`mr817yUVXl4U6%863#std9k7U0HOGiNZIl0?W5DL^QAnUr?+t=X)`U
zlE@_+k!B2u&G}WV(KVG!-S(Z%vTj#;hDFx_6NDNK7^ILR3~+bdP22eU9=e?Pi{yd5
z9Nwp5p|?8)8{OQdyU(>B0Ij&0KCCP*G#COz;DBWpUk*Yr3%L*@>MCs?k-nWS*XzzM
zjb<Iat}DQnv6+8UaHD<y#&{{%yCHV?B_MclUM&`!`IfOWz;zjjbR4@H(@=w8U9+wv
z+9T<GF1x&zXF7m~m6pFD_QtG|OqiKLr*Bv3yLZaI&DP2&e^v3=@by9&izW7yiXp^o
zDW7H@7;R2u)w|!8&F^!G7&MmUOD<byTcjfFv<`*ETzLRD91(6F%`}ol99&RED0|Wo
zT8ABUrgEvgXKqF=zmEU1)7Vw7|6Yea*12#SogebM;xrY<nQDJUg(13tNaikJ2gtL*
z_2v`!icxJWeBm|!l;nd(12Khu_k~zKO&>j$j~r;=96ZFzg-)wu912QTW>ZXo!~PV&
zp$JoeeV#qE9DCm6%`P@9b@dY%2E=JB*PNhv4!+(6>Gt!zRfJPE4e|3=uy#Efn#Mhs
zFG<}+C+z;kB}2#Tdc1sl;V&u~x<rPF*1uQl&Z&57ujezs&TKMgXB|hJ*1q~(V_K0%
zm}E!dW`MI0tXMaNMpW`3<)p$q{Zv3?L82-RpG4>8H(+iw4jNvN`=LmJckdDrN#Gzd
zv(`WxNas1EUnI#-6&;r1h_kGk_I70o!U~9h)2Yht_r?8lye|yy((PX$@!b|mEiD4a
zFot~G#c)@?8aY6LXs%ITdAQ<ER1PJF4AY@CJ4tL$9ht`vaP|dOZI~hh6w_y?gYX=a
z&WpN8fv4mq_G_K?i-y@B2g^S$wAD+f<&CSfSE$AVdC2n`)R)-t4EqEj%Oe?0-uGVk
z`|nXk1TyhpV-;JYOWLoKo-+$H=i*+?HW=5jGncf}T{QRTyT@jq>%W;nxxoF&@UDbH
z$B?(|3%qZ86?|9U(y*^~$8o%7iVk!fj*k*%&Zpe7_QjjW*PqT*UD^?ZTUmAY`S)l*
zhj&WT_+QT>nGDOTF?Y%Y=x*KYBj?Z}GHW+iB32;rx(56-w?cp+b~m5ljflvQ+%oQI
zi!dC!YS4(2SNbev4H&5Rhm`^&hR`m<vv(w}Sa$&UJ<-<)ghFb}TR{yjkEF^un2>I6
ztaGXj*NbCsEvsh>xg&lDvUbdIg;hmS=9bT143RK}2Ed;C@=^ORkk}9U%JW{UFT`>u
z-Z8IMF{q7GYjS-EZdV!0OibxPitm@dm>FEMc{iO{nsBKIt@yMqyA#f)=(oK5G)QpF
zu;*g#_eRpHVWzL^ZgmF3D(j+hcTBuwt>sgz`?{woFK!C!C*DyI`rQWpeh~hp=wJza
zbkG&|d$28jl;J@Wo!=Hs%!Xv42kA{0=oH-JJRU_VU))RJw0S0V1d6h@{D;81ec@~$
z&?>3bD~f$Me(fYmbV32;h1q=Xc5Q7aY~<=b{7B$o`66%X*K%P`PkKJMOeoA+-i2?h
zN6Oxc62^Paz!5e>7gx8+<+-g;xHFd}GnLAk<(c@Vj;i#Z<TIl>6cBr^zQPd%in*gs
zn319WSVd>Mh$2(_%R+oceOW#F#FySKU?d=ZlO5sa#*@U%qu$+n4c5Sy7#;Yn6}2T)
zuy@bcgQT;A6>F-|8QaDCj5=7p6{4-3$=9*-&)9QTjCfb0I$0P3UrC8$YW1_3voB)9
zF)TWvZzUyC2gvp!%KOxxP{;g0aR*Ezaf%p|&yR*_?1x2$x82oTA3N(@FV78-Z})c^
zuYZG;&m~+uQ|YoHkN2<x7-}#b+*vaFTA2HC`jeC6eyNn#*hAQ_rBD(LVN-d%=}Uc)
zBT5}%3AZ{zh!8bp0E5JHX1p5yA(s=3j6DVx*+qL|cZX~Xp0&mb@$%%RAAE}~jrXlZ
zl`4>uE1^!9$MU5~@K{^xi#QFKr9nH_Xf!3IDp|keXf7FBr{?+m_IYeqOpjy4`wMa+
z@izq>RH7WBPpA)j`YTKR;6IMCZsy!K^a0SXGVHz5FkWcYkMP~f*E+rFCy`s{K8ujg
zA2wHfM0-LP!gus%=bmnH1hpr3cNKQP6mW~*W6&vh`!vICzbS_%Gfadd6@PzI3Vpxv
z*zq+TMr)02)l&3vKbi){gpUyn$3GD+t;KBs_Bp{8*~}%KPMiA}McS0C^17e9C*c(s
zq}C9iC!D!i`I&%;$VsLz^ieV1#89A8T>5-I+bg~w6M;$>_y{Cl7C5)z`42IiY*j=V
z%e|++WRSyfHPB~peJG#9%`nHvY<uHUuEYti5v4oyxwhL==IP1f5dTuqKLftXEo<Cl
z%^zLIpxYzkoS)EI&{sR<fFyD+r>+G>b7`k*p^?+Hs?aE6z@m$CfBD*_q;y#B$^eF{
zh_muXN+mwSS6`^y{rE(0uaaa=7gv|7Fw!)qQ~kLgx`+bDlME%iSV`d-_nqE<Cp6Tj
zH*n%G_wf0oA~uz1Z}&Br-c0ywj_o^jot^!bDMTJBwdE4U>>PB`cnn$83VlJYd({N?
zUV=ZYJ$H4R<b{3t!x;`;XFpq>t{@EGNFR6nV=ceycT7?nn1+9Im3SW2<6Jsyk!c**
z9BErUo^k3I2JoMF3xG)Kh;prYErtK*Rw4e1IAp{cMD-tChG6D7D>>R=Ei3y%VI|FV
zO_iSRnfGIS3XZF<L^N5I;$le0xDD7g8*lG(bJ}QLKf;8^%fR-cxH3ldl7tT&LA_Wo
ze9*hz=F;K6<`Fd;D1U|uGUMX6c*H}sZWz8|e?@jtXTs+dzQ48_CBL~kTdYyVo4Q2Q
z3WNXCu}QvR{xnjycOmLDNHYP1zV@)RT#3&xaafjAl7F6BL|>&dXfCOr!r-QkQS5l+
z^UlDu)OIrloC)YsC%5Mjt3Qi(IQJp#DuGaenoBFAl4<CkAeS_$#_R@NO?=!BXtjNo
zNFRm^Dv2=Oj_BjV>-q4t+f{CN0Zs%xW!*&O3@@~ES!;yvNs6&ujI)7-^nRavcB&f<
z-pKm-gL#7=F<+h!dotW3Otj<gJL8}?$v-U-YHG0C1GfrQ{@VEf42j&SGX=vwJE*<c
z;M2B@i${$dW>cf4*g_)=q8xB);%i?whSil*5;~s?N<%nsq<azAzaGN-$U_)QKK%y-
zQken(9<~oJZvNWZ?QbNYywLkAVx9+kvT@ZD-)^99C_E$*L)p#xeJ_K7$hxMis@5$Z
zKVY}etB0(J3n<)GQ45{s+@C?i*ziqKpxnx3m`-u*^6--CYmbF71Z!Z714<)JWuMzQ
zKh4azi1&pwPCL<!KrUt%+yYJMt_9%F>Sj2fxk85HNvCe!#5KV}tN`pXH4C>Vrv#o(
zU=~1TvU=e?9H1O3bvm+xKH(YN`f9@t3lMVVf1Z)RFo7o`j%0P3jO|%wAhITPqp1we
zcp6)pi#4cJ^EkHW6(Vo0zq-;BF!Ri;SiQzTtY#C!Y9d*-)TkFVkYZ^4M*HRqa<ntG
zD1=4jx8g}C`?%gzJ!@CQ&a^hbu96e#t%<mkqPZsGC8niv-aT^;_7z~*YFWQ|cF8As
zL5C8&Ry=Yu-thzHvh`0)-AqL4HAg|<a^&Zh$MQ^^5lCCvta`cL9w-jk?gubKc1i33
zw3#075<^uwL5;!nBZq4s(4#gKYWh6soAtP7CfaT!J(K9-?GPFFISZ{sE89>sj^}pm
z_VEvhBS}Y62vJMo4LTX;fzW<qXq<dVJJ_%gU9`Is^T8(tB;&lGMp1(c4{9=VeV?ER
z`Ykmo9W+YK+)_5S2W)Y`fmvW}-oQfde75>)0oBy+F+1-h<67V)szZr2TyE3}tEQJ^
z3G~Vc0bhoe3lv}M#Qjj}Ee^{d!(zfjsP#6JM05a*Q5xDQ)MPCz!721{lYV}dm`1>p
zv%>>ba|+au1~kAb(=R+%+av})ZyuKN-dyK>UL#ld`(IZW6g+<3bsqn`v{rv$to+TT
zEd`F6E9sj0s99q$O+5AsYQM*4Bo~y3{Jy0BhrRa<iZb2Wg=ZWbZBWrBsUV0nK~Rt=
zp#cSyoO70-L_tJCgMcE4C`y(Lk~4^a2sSv9C5t4<Ip-$FZ#|Cg+57AnzxO@A&s14u
zSuh=*JFIoBD}<U$4HJ8JWFo6riC+_!5ge8u86=jFVSP(WwB1S$=85%|GG;LjFU^)f
zx6`TC_Dnsf{4BG&+T^Y0gvrl|!8V=Y%(hy~Cmo@I>Cn<}3z}Tr81c`|%OY;8!4sch
zQ(t7J{V1+w&PH={ABm?!lpi{9A144O<!o<8SDuQiyju_bSv2?UlBpC%5(c)yPRvn-
z_~D#xEjq{w=HVbea9MK;NDcT6#qDf-3uWZA$|6<y2bmo`o9UMLM^r;HJE`k$`j@(O
z3&YO9uup`%NDG2w@q>~Sg3Nb$wHn;1y00v*)1lEjp<Bp-V17~+fDNLbU<~czFpgOD
zJ1G2JTSI>BJN;_Df@GXbMe4{sDCnp#8$Fhp*-xiE@UPuZhW#O%@EBrvD{VLJQR3vo
zIkI8wB3TJuXL%8aE=BB;ml?tpn4nlrvqjqZb(ip!owFP7_xHAgZNfq{{mvBAD)~*0
z#QV@Lo6rl{`&O>d_^QCNCA$qWtzz1!I!z%G$_q~rJ0Lnf?U4u$-+a6)^jrDWm+~VM
zkst~ULxi<_9jq(t`tMXs*8uYiZ@vT_?x)V=wo>EjL=^%n7O(aZ;R{kDrpxgWm5Jrd
zR!95q6B>Qj>px|EJ5ZiH#J-lk#|yg4(}7YADaEH0k#ptkGwq2`1LTQ_fXhT0i~>Ic
zw}j#{!weoIp#~vQ4h)jl^@;Zi_N9*J?sCOBbKX-Qvrh&kN|o#DqgHl+f0j+2kfa%k
zJ5j#Fx?>xk9xf2rc8hacKh1w|K@?NB3w44Ok6${G$JqTElDRuS<oSPd&pxa}N+yAq
z1>fuKH0}dv)`K4nQ=BcK)Aj=@WE_*8>M36wxy+>y&BNDFnmK@^Tu=_Jlfs`JdWa{H
zZh3`Qy8K1Sy9+Vfaf@l-DtY;UO*(|4`!nf?d?e2jo%d;dymOwd>}ne+?#ma_CPdqL
z?-OH~m!(C`H(Hh=p9jDO>195y78G-0&B`p-zp^gejS#!st|~gJecq4`BhgT$)~xm(
z&3^0YgW+A2+mRaMStX%(gB_O7KJTgt_hKVLsLH3<fvM3S71LE2T=&G=>(%M}NH5b2
zS70o-UQMn}1UvPExIiJ`RbgctKQ@S@HMaOeW83BP1rJKa&={FF;wBCaoHfql5o<Ee
ziMaL*^-ObwM%ACM5^BPnLA6iLO&c1iIkb88K0~e{JqL}6ev@GP79!h13#7dsPZsmB
zoTp^H-{2w@>nAhEyL#J4E+kE}zq7l@cHvsC^a1vzHiJ9Bg183Fxcun|K<_W@2~w6c
zwa3e3`ETSiZh++bMh9LkLht;OXA6J<WC?3JsA-lDBPNWgwSwt$fFin7!#dr4FXk$(
z;t`svRMEW<G%x!kb9cF%gk#0kMCDYYFQTH_0;6<VglZ<;_oBeZ!oRH-!4f!fJx8ja
z6RcE+j40ML<${7@qc+dqQb=oT*s7X!iZ3Jbcth_i&cYzH!rrWYeQY>gYsgr@k%%-O
zA$s9S`tx5`PeA=5_ZQXctI9l-Hd=`vTcDuxcI5vF3;xO&GJm6y)%gyOw4r`Q$sy<m
zC`sM>-$xR$pvk;2-zZ$<rLdRN>0Ygl6uGCpl#(5e3K$izfE>>kObXGH;)i7Xm>%|m
z?-}_S+-TruVz#V7@mA@ERJxC=;!drkE5xO&A1e?M9HK!rXrc7FjgW*r<{wkroP16r
zz=ai@OQWkslHEAcf{$N`{rW<1I&BgVdjDe}xTLdG+eC+87*S`*b(GzX(6xs)zF{1l
z+JMW-3`NtPGA(PG*KL`3jkt$RS^f1@BZYQ4RBc-vCQxy*6VYH+BmyCL#31}ccR*M^
z2ot%xq2x>eg`Ur_MU6}s{hpnx1AEjcBP;hlC@JNH_kp9be#DbX0M|)IKsfCNN0<W>
zL<kej>qhcb!wuZca_!Adt#5DZic+2?btCU6NJ%O?e=kGqTr7-3M0(%Ea>Nc}NUctj
ztcug)APcK7i^*SP%bcSX@uFhzespY=5g$lZjX(|l#aL3NWEnrnVgjs(Yw30*_K56S
zVa%Jeae`5PnA&%<fa8ho;Mtex&VF!2?N3bLX9u1Z>e^RaXqQ#S$72rz4$D}H8<&~I
z;+U#b^%z%zAIM)&zb+Qhpf(?^<eHpAa5k*drtAJ~08YX-tR4)r5+VF0C8q-4P&kl5
zhwicvRnqre0U(RHB38hp`p81Ny48I!`C7`E5axh{V5f*JDlZ=-%B9;?l@xW8oK|nQ
zrAvB|Mv*nKkn!rPN}FMHNB9e33J*Ytai6p6r1YXfcv5syjT6@8rCef(LhHgNIW$?F
z6hht)STs}8xm^yFQnkP0dK)`cnsMJQK(az_;QX+|Jp;eNTpFQK2d;)gjfwC031!>!
z1QW#hDCH~T$GCn~mzv6|sWJE5z&<ctYCn_=D||PVV((`Ss0^aZk&Y}l_`L!B?EpGl
z>IpNa^ae|tp%f^avzx<;1aa&tL1*dP0*^C3dGt*ID_-J$KP*kr^E573<vlC&f+hHw
z$p}jtSg$;Cz7;h18m8lA(Gn-n<&!lwp_;l|oc9r^Z@e)o3}}eZvO3V`^8JhB6|r9C
z2$2>WbVD~KRf5)&|42wl<d>2P3JDni&FjERu{s$&`d{O^)k=X=y%|s-?I7Mi`YBjg
zmNPUU8Ro7AJ7;y<9(z%DLZSjotm6fscby-7p2zh2kUJy{s5XbK*IbgkXPBVkYt7@h
zW&S6*D;gg*cG_PpabNgPZkKY}wCr(pDHo_d6F_O#BC5E9IsGNvOCSRe2T1QT$Dr<D
zrRxSs1(Ykj8sm+*Fl{vW<vg{6uu-{JQie!w@6zF;VcfZE&<lB16JUOL%)ef6{sV|c
z4Q#>ltt=bp{=GhD=b&ndOEivOeJQ&U-hKmh>d!?Qk%P-^snMx7F6z*+x(0jD9#dxI
zCe<bc4$#p_0Ai125uANsnJU7tvD=;gm^9rS1q?m?Mfn^=9H!Mi)W}{)y^ChDQeegA
zjHhWKcb>t?mx`t_mWv;ly@~fQPi+Bk()m!h;Ip0*8T^aJwq);H!0I>%fd)xwyIdjd
z{cm9Q;IdRQT_;ki`i;c)(EXM;x;hhP7b9P6r1>-Mqez2Q7{?`wxBEb{7gsI4+fH|^
zG8o&)DtiB}Tk4a}&p1xc4|it+&Wf(cOS&-=OP^c%PeNHZjS95WZ3<w(bVIgyU`b5q
zr^<u7L@$+9$1nc7&({swTR;d~aJ8m=813sBjXJus3bsDcIP$w|6$RY!dUdR1d;$cK
zk=a9zW@27yWp$A?^k%fJ(KxNUOc{6A7>as;92&D*I}gdVN6m&g4EwQz#lmY{4C<HN
z6LQI!I_xQbZ;nW%AD|<<&F>~>iYSxt&fUeaX61rrLlR$>=W1-s#C#Ew7~$|3lA5(k
zFWx#=fGspDk1xP)c<;WRYpC%5Zt)em`R^8AjaGr&pMAOgMVlbIm7p~#5PqX6BD^Z1
z7L#YgINQP>*_lzf#?38XIkBB`|LblogFio6$A)pFDpMB0#CDi=edUXy5J6+C<`Z^i
zQy}Q8L5>wk%uIv*EAY2_S%lE)6$o!zlH4P1H2C|Bm!h(xQCZQx-1j{!EhX)>2(1zt
z2&1N{Qmd)*WD}ltoQ9-O+jr@(8=HfH#j%K**DZ2Qi5=NE2{}ARlsx~~x)(UIY~U=0
zbyJJ8i!IKvA$or^<G#wR(7Fpwx#^Tb2S~H+=O~zt+an%bW?61>R(dm%_X)`=6JFgH
z9AcPPQ$tqc3&G7<al=VQC5m2g`!)o!MEkmlYji6+T0^Zej|gJMDcYMI$i51Yl?f2g
z6Agq+!(59{%XgRJlk=7OoA4QeWcm{qve&Dirj5qM%Z7e-RRt5$JyQeA7~kJiZ_1I+
z%i+UHUc{NxK6!>~y~m^lN~hI@w(Pgvbqm_OjlPR}8g#=8B804IIsPeai+vfHu_oWU
z>+ca4w1vE-vQIY8O{B)%nhB_d>H7oc#m<~5oQOIkZRT0x>^O`Qa5R^X(i~KKAL-+w
z$x^jo(e&7^Wm{u5)bZ<0s{K;w2{>?-sp;D%P9Jq+nGiBCf5h;Q)CB;EULYV*UJ~OE
zJXr=4Fypg?j!OKKgK{C6$aAo{oK1*uda}xO{Ng3~%A$}2i%Yj;^_Z?&AOAfEtN<Xq
zGNVeWtW&LCs)h8&|L4DPX^2hq{7@~BqxJo4H&-IsFv4FwN)>}#2ug)Sh&zE3jOzqR
z7#1$j6t7!b3^~T&8v{@O{wq$cV6n<zC6%_GYy=tmLC#Xgr6aEmmP=DU9-%n%&o@fT
zS%kqBP;ou=yTax9qX%LOpC0*_=d;pGLDN`jM4ZYJ|ND35r_zyMy{k$gh`cpJ540tS
zD}`+RNP-JWYS9VM*kUMySld4#fxh6!CF-D`&3r#My(2K*@SpLV(^N1ber|=$`~5qo
z?#5vjJwrfK`aBPk^a4vw#fa+WgGvwdZIBNneC!gJ{Gr9sazQPY-kLh&df}hFYLq&G
z7TwNk7klVC*XqjCAD&+-okOO`91-p{lCZzdwJYspgRb)Jq$ke&b63;C0+pr}dBu|x
z|NOcB`vHIZ*_JNk3!i21bNk=_>i>Jye}?A2KVJBO|Ne$Av=b$;+}*kW4qtouY88Jt
zAHH&y`5{6b8PJ7O!TzPi;mfC=ojLlD*X^I5IYK*p<&Jb1bo1CB{F#TuWC67!h|`y)
z!~fLjE5rK_Sk?GfyepE%g-r61UiH`A8VdBQ*^7Gb?x{W<LM)=qz5*Pjk`bmSx$Ss-
z%KLiV*P@yF7cEXdL6g3%H^^gyH+Z|e;JcZj7V+WQN6|#o=%_Z$$ba|zqVhC*|1Zz4
z62_&c|4YxW=A(zzetRh5w)GCwlFWtuNDG=%GM!;Lsgu2K1Xmj(SCSW&oO!F6YcTIc
zrXV#&Vc%L{q58pt36=WHBDL!cmHxfJC@k(xu7<B0u*q0D<}EK8&I@Jf%q9HfxB&O5
zYGB10nw*xm{*L&xkv)W(@I8ojOF(Z?9P2XJ3+j|UkeAFr#Vs`81NxC-NPsK`sIem(
zDlLar4qtUl77~hO1zoswb5_9_AQTl-?HoPDg~=UM%xYT)$m<V;*wGllT~i1E9ELiL
z=wZpZ*F%q~w<H}VTRzf3%{+%h$dtI;FBO@9LC(EkCV81vBZNBwQ3Vc2q&RDCI2kID
zG5qDkA0=`K+id#13gg<`)|*83i1v<r)}>>a*=3*Ee{ul<3u~X@!Qg_&1^yr6rB@OM
z(C<Ed36U^q4?;rbIId*LevHFDm%hW}IIE?Z{yO~cy}+(5YF;FZa35s^N6U_GV9JCQ
z^|sz2Ao*J}pcMNA03s8F-_7=eQ;ti8`n-GU3(ZGoR)2rpaHz9dD64^8`OH7k#*mHq
zkO+Wz^s9jH_t%3fcK_iDwxIUI6^vSX0t_H!{gzjy!FO<fG7RWzhO(<p8mNzkyQ^hG
zqQ!juMn~g_DSu7~_D7IuYBHDa$WpEB75a5@N97to$~!K?UPm1GmOBM)Ta$d7H6xT1
zHlZ7aQmJhNs^#Hw$vM{a7u|*s;EA58+WwrXNXN0JYCe#ZB$8g*K)#i|vzXOFdS>-V
z`xnVwr3f+A^zy(Pn$GT%x|4>jgevVZ;3wXu`+Xf&m_4F$&3{^VERTgL)L3#-aYw*=
zxt@H-`>knQ5dWyj36ERJP2AhZ$Gx;H@hgFDGVxhW4u+juEa`G-iN#db#G7R>Yfjvo
z(jFqCJu&Vu=~~<DaKuf+RmSev8|68c+_$PXXf~?0l{-h2A!2?-_kOwH5!`gq>|y)&
z$6rW~1m?OsYn+Tz;?wg<+)8+}WXEo5{fuJgrif&7Sl>hWNyS!98(9{*Wf_n$Dc8Ac
zK0UJ}c>2)SGyReeA1>ciJxMZFGmsd$Uy=%>HNOKs8(T`c4^>}-W@lB*P^>3jQE*&l
zX)n!hD{+F=+g0#!*|fV<d~@N_F_PxN?-cNV8?OKRU;i7Bd8jC|S&0DYBu9}<mjA5@
z*vLP#ztF9-0BDJjTZ6m@FLl;H+lUU>nl2nNIEX^%EB7Yp1C|#pN06ZCc(_R~7;#6H
z9Akbn0=?o&7&xK?3xU&6A9SgG&^`5mqH<7N7Fb!;Hjw>x7YWBwjS#y=mV8(e^BA(|
z%1B8CZ+G)}l!1t}1Ok+*=U^EiF|;{wQfOa*slm`T0(+Ezzv_FC>kiz0^NHzWLt7qT
zgf?(E9_)GbG`Z`4ZUGkmw-#WsjlnDv5Q%mNv66|d7N=|Uz$v%kMIl9UXBFC}iJI%p
zC-#9*^jUNn;tK3RmX;X>X|f+QCOhHRx-Noqn2$g#k3|q&vj~%xFgRhAhSyMK4T`=&
zVjvgY!UY{NTmcC0gN-zesKX$vGfE))sINFeq>ffpr;gNpbInTJ=<)|RXX!+AtaR<C
zA%}KR(znc#O?||foM8iD$RSAD=n@OQkVvt+A|!QJZbsF$ACd@1I<(DVqV09Jy~bN0
zW^<Uw2NR?NiAP^V(zkS|5v8B4s0bxsOhsR44yOZjJrCmw(y;WzaRS^fG3X>qn`pc0
zQ#ObNd8oADd5y(XcQ*1jvnc898IXKi^(L}babYjh$bMar$xB2pD0}C9fD}@(&c16@
z`lACrI?uWv@#GfVTIugYA_AqEV@5mOjt?cXRx9j5PUL$S_SO0>KwRb&d62;Zy7Unu
zDF3i9=5oj~uFVyTDfKB>u(A;W`qVtQ5UeeDBQk^qA&-p;Iu-i@z5%0n(ZA}-zb#$T
zXrTBFHIbJ6vlx6RNPl)oDc7JBe{v9UkMd;n^+DQ}b%;kRhRmZ+(tlt+KlV}v(TANs
zcKL~}$#Y(OmpM?a<Y!X7IMBb#)OG*_V`z}@n=c-70Wl2QEhf@f!YF%6uq+?J{2J_q
zLQgO<qRv2q;!1(;@(s+x0bm0lDFuTij&sv9hTw;w;7R05$ygHji{h{-Di-r1@Q$)u
z_IV06n!w4ihlgnI5G!V3=-VlEt%oYb#H4&>z=!J)M@i5xoUv<tPKK#UPtY1^AJC|b
zg$U_c#^c}(=N^h<n{WV1R{Y-Xb}C8~a~8c1-g)OH?8~2@-9U`DOP%k5c2J=l<Ew!&
zzyVzn$!p}`xKa#|N9jrKw(L>m<{%bHf|$KT1Eb3cq32a_(@*j2Gd@NlM5BqG%7AZR
z4mI~Jzc^JStuZc4hZ_C1?zd!#2N1;N=i2F&4I82;O2{e4P}VqcQ*WvOj8Q%uyXjwP
z5!ne&U~dhdp)X|^qL<C1?NJXgPj?$_-A8y{AgN2X;(Tm=Yw$DURUVNa(0ucNCWrEU
zEYi$Ij$=i!GQ=gy-9~Cqz_N|SHp)v`Cl{?{k=sTSd_(rY9}Spe-TsIThNvUroMI&&
zJHQi$$x_*>H{2Uu4|+#c2!~CJv{s?>k}*d|$OA<Jzv}FHdhLb&(wOlka1of_rq9!m
zH_^=+RJ77wAV%Rs3u@fK&(FQloU}fS^&?mMuFBebmD#jPS2fopi5i5yxefay2K)Z$
zvwk!5JF9`V`=9rcccsrMp@<6U7;)3vK#JXUXSQ2~i+)1z-sIpwk4sVQ^-{wo@K57V
zP(bg-0JCl$$am{ng__Zh-4+I2_hGMY7UyiHIN|%vrbiGPi<Ldij7Fz$k6-qGdQYEK
z(mt@pK=hY(P--wh{FtvC&+!P&54A*x`aR)z>7F=0fS<d@bYOXtX;Sq_i#g1ov|mPs
zpXCbFG&O#d<;vvZE4PC_bFby9_WJNCgk{P==2eydRPZu=%tD;6$vLu2#33wu4-RYc
zh|3AJM9P6UbJ?$<8Z0Bj2(Qrgi`SD=PC*=1O%m*ZsNHjfh7b)Y8u>rSBDz?)K3~_^
z7c(L&9~-5qBX_;f@giwW{>rI{Kh95SXvCrfN{=#~doFi{JVNtL0hh5+LbB9)V}<E3
z;siGAjnX(r)*!)D@>K3gMEg@;x!vd%z01m0-EL%^&eU?`WDzx@9*VpDuz>EngL?WL
zS<A+Kvd<fw7tQukgt0^ASleskUpZ!8pt8>q7ALk%&qZf#)0;p@Xqlg{5d&G*V`~qd
zC9O8lb9xLAd04Eo;*_&TjLK_jB{UC4W!SYCq?J|@-3#=cZk>+2!P+X4w99lpeGlfg
zV-wN1>y50&=Nf9UH@4jP?dQ^7ubpVrCL9mX$Sk%PZkMv>j4kL)k7L?;W;TK4xD~_9
zs{#}t_IjC>_+CENCEBB|DdXzkY@3K8V={G+wg2c<7KoB%AUkxcuX7_RwK)x{>dyzM
z^o}EDfEG99MrS*V#~<NtW-XmbEA834be8M$4O8Q`+RBDN-)C-7E+3|atzvI{iH(2O
z?d)??*y~?nt-o)U>zK~w{)YL1n4CWQD~e(zu?#mYJ%hxuFI=;tT;>kt0Kon*Wf@m!
z*o?$5nVZ;~OK$^lzY~Y~*2tjtkhx3cTIl)sU?N>KN4Qf;^jP_G<4QKPrql(7yz9WV
zDo9QAqCx)_5%$j6aT?E8i3T&GX{l;yPO(>K4FY3;ca@mDudmZ3!*QJ}S7pCGt0iLG
z(4SWL9N|9Y7+JZ6f)u0@-oQeBg?G^lGf7I%_S{Eu^e=Iueht6f()rq5a(IR<i<5A8
zYTJ~y_pZn?Ak8l@1|iP8Z78Ao<3GFro>74i?=4lLjI{=JdSp$a8hFkNJ{xv5jgA_R
zm2i{(RQ}X;=%x2B3MF%d0w1X>taf3zMp-tq{_P!Suv1_U<=_lkYmwv3_cnRn{q}Yg
ztL7JJ&KYcaw()u-*sO)#WZY7+p$>KSe=Mx(X-9Yx*~S=W)~~-SsNm9!j8)x5L&G^V
zxpLnHdi%X-cxD@7yO&;8MlGeqYL_pO<(&eLi(<>V?uT#R-&#<eRpN)s%sW2;^JhEr
zD&w$J+Ku>M^S4ano$`q+(6ggJ2a3hYO0O^(z4A_8@htHsq1wXYeh&)!a6cV5JihOj
zedkv0gP)smKv))#sfH;?45Jy#Ked-(nd{!`r=;?R<{r(H-~``X`)?0O8l7zVmSgD?
zjKy9?=&u#x<D|PR;$K-=*O$M!PFW_Ht#^C}u@JD>q`dpZKOcL&`pe+QM^@^Ksl~?e
z@eH_?{6&1A_t5ixkA1fVJ6X?5+44B~hESW*RSgdEH!b=7-0@zPI=PmrR+@9nxTE#$
z26>z9SlVRbJw!eoeG*|v;mjxlE?+}&I2h7?U3S}BEBdf$S9rYKqEV_Kd6$9xdMN~#
z-`Add*zFWrz+u(5HrI1G+r8|WaR*LCsA`_~>Y^rbebNM5ZkGtAP*v{FhxA2kg2{pT
zBG4p{tBp`AqVr|>BO>TH675UlR*%u}S!<1F7g(Y8c<3%v-qUfHr-Eucb1I_upK-8n
z=!G-%XHU(p|M0g^Kc42LbanUu?j#H_2zVzUlme!Jn9jEta~fDYaphZ-LF5{mLx-(o
zs-B3jm#(PT<`Zf2CL>wh&wz0d5v*SD=YDCA-@ir9Ktwzo3JCP5C9Fn&Q*@lE3#Tjv
z%aX)73g1CYBrBtO@Svh6mJ@O1PDCQONUF=Xg4VMk3e{DYfZD6P5=#!YzBfe`)fHn0
z$Srlt^Pr=5S<*hAMVzkMIHH^}$T4rk5&V$I8NGIssr@s)E!<+iab4M2!`l#HL2*Sb
z#mYdXL>iYX><l3_gyv_@A}S2d<%W2tAYXRWKYiJ%HIc2zeKrx`^?_dc9oxbK;xS#R
zAm(myQehrjWz-&~gM-f-ciC~a%Q#bo_UApYAO_@9$BYOp-&uu>C^fOPv!5f&S7j!d
zPdxKt2fQSeN(ZLIt<U^07cxjb4nW1M#X#$8f-&U>?A_{uE?5}$&yo`t2duzs*7;*K
zsQU7sc0SfhI4Sf-aAm;H*z3wwd=N*pt#cEMw&%6U(Z&L8=SNxmjnEo^S!4$cYv0Vc
zw#eM<3L?Ula*pQ@#jdhtz4h`#tFii-L1j^z&5PPIO~P;VK~)q;lz}ujhWW$`>5KKC
zL0G5Ve{)0aY$BXP><2xTJ5q7!ms*MEbZeO{^bm(tj{cC7YP3CVDU}9Li2Aw%k;R3>
zHj=K6gGtUFP8|y!^7ow}U?Ht~Z6>m_SRt+TWw~IFadrnLpDLTgQ(1W(LPY@uP_bQ)
z5p<~2CI4t>fbIP$^2UQK-P1eAjsL8oFJu$dQS3-#Q;&NRa|pQaY($?N)6QZi@_<kO
zoSci>FzAYM`C;!{P_;4`C_78|UJ?P9lgP-`Aun}4?h9G0d`uq)z(`$*fyDbY`7>w~
zORC>|bFm6${hj{vcdnG?Mb8g#@oj8&gETMNZ&Y2yUQqfZlX#VUyoo&{$_KOeg}+wS
z>~@kRt$zu$`#8pUD*@^4?CSgADlUS3U2U^A`!x`ItB&nex$yp!qQ}$GPs;QUNaR7V
zC!P^r`k{9{4ZKx6mdDdpBiStWA(LIR0Bo^SC_bO8cb!O2jg5R6cQ2j~(~8KKf^AZl
zkMATa-T6+$`}aosuinBu1PMl{GphP?{b`*(9Kxoy0nCHacbGUSsbNnb%9ft}H~>_7
zbuLG5Lg+8!&|5qoU1jB#mWY4t12MBp@f;)lfn4>EXN)E2zLBBm7i2eysHb~?3Orwt
z{&=F&|LUR>XqjsA>jgQgU+9stUO!A;$@Ydjt_!gNUZwp+{sK4zj+CG_D1~72Kq8yX
zlbILoUQU$2HP^v{Zll?SZoGBtw3jwW!JTzFG34#odUC5{v{-%}=xJK(G=`0#PY(3T
z+9V1@7mt(eAT~zIaPKA{BL>^B?n0{x?<+WxQ0M$VmuaLGTGZ<HD4Egk4Xkk?2*37V
zWVbD38&mHq3*hQP25mu=W8hZ`78H!?%@LXnf{K<&U8PKiucD^50}LzBa~2EIr=7--
zu=2fKMozO^L1s131@qfF-<yj9WN+CX7VP{d>vMA@dLIoL2aGJvj2!sevtV#P--&im
zgM4D2j^fumBc7LO51tSkOU~}i9wcG0d`-{tlu_alBWs@VRW(ecxnCD=o&*nT75iz4
zti#4vjwYPDUS%L<+_S$OW7_1=Ea2KCe6zW9T~CC7_34X&|27+h^0!|PY?u8QAS8(a
z9?E(${ksca%R8c#A<@HocdJyFX?dI%2N5zAk+L#N?&7QE3l0bm7e%gi-ILOrHuTBW
zb5vvAR`#x$$KNC9*axN}zZZ-r>nLtk%>)eQs%L8KIU#B%MURpXIbSpmWFxb6A9f@w
zbvW$Sn)r6PhVkjuO(Zz9KsdTyZ{@?8KFJ}0c1>TcyWTb>3UdL+C5Kc=3a^S)ipe>!
zwX+Q+XpzgrKR41<5LGjK$6gkCTuN^Q*WD~%M>3?#hz<Hh%6eG(CiIm`&%eS1o<=n0
zod?L0o#}(C6G74X8P2OPE%sOh)Ny_J=3_5knx45&xjAd)G$@k;i?N3TX@RjL7whOb
zxB$RQa~-MY2Mm+;!FsF`9S;Z3_)<wBhi*W)E#gq<=mNn`w?a`x#0MI5ZA*EZ+0`Ke
z_<k{%R8x^BJTA%j%GbHXBDu({v7E##6w+5&gL$nSE5=m+7Ob}j$>Y_4Fh{-XPp5r8
z2l!GVytT_)!0;7o*X~*E(B+s0*GSuX@IsFk=me#?IT$;wh$(yTT?}b^+iRFkz_vxW
zx(Qf4&&hsy(_HetZqB4qr$#razUG|0+B&Iy0pJ~h)#^qAyc2~sQ>nYaG-@Wf%d0r{
zk@K=eUwi%-#GE$ia;Rx|UKKtBp=c*}XU#;PVNN&7+Z{vioR4<Ka?DN`P>CvxoTAM2
z|EYJ-FGQ9IiSqV46}IoC+bvi^9&maYuwWJ}mszQ((xexQfcE90{7__(&4MlPnS>5{
z%Nvt>0|HiMc<hEH7QE%kiBL!aqBml_tqfg5>7qmeqi-4}gxKM+V1fFiy6+9Bm(Ym0
zoIy)U3C0?f+L0d?BCT>xfzKzCyY?+ZPY8h9!d>kPl4YL~K7%-a1`=@qIB}<7xsFY;
zF6;sQdFe$=vy7$5#+1&gwi^o9T?XGZ?to-_`V56wJ|vBZSm)y3c#BBAN?DT%h+o6t
z?hw!E04p{D@$C+B5-4PAqxXLbvliI`pvznXGWD^l{I?b>;yWX|?PXi{z*Kx}%mavR
z5!;CMl~fDHW;us>eEB+~oo}_u2bd*XJ9Oq=fSDc4J8$N2y#11`d)Y8`x?Pbzgj&W@
z7Y63|-ScWG<rio6hR-Vi=VRR@o>TcrJ>u@AdH#g{@j*S1PHKB*0`dW_r1-OJ1y_M?
zKdf3iiLo<c(VSm{K)2;+a(8<WMV)%Q?C2cq?L7z42t~I!Pw`oYPjGPF#7c*!5ZJO_
zizV&VGQ2uT^d4Es_z77%FdK2+X;oW~nsE+c-S|?HTuf(e0)dFyR!KL!xrnn_V<Sn6
zN8-733>go%VVxm^fXngxc1R4cE1nx{By)i_RDYPC2DXs9UZIWUQBn(#twt29G}s#?
z6R^Fhk%dfmGGMvN!g3wVZFNuDq&Z51x%4tsSNJj_SK@f{@*)xt*l%E(`6f(EI==WO
zcZr6Tvep8IHnc`qKi4b-n>^06D(T=;XgK3-eSya3cKn1PzRWV#XvwAP|EGR+L}s}!
zF7keN@URrS(*Fp<5^$1jXH=La2@dP7iF(}_?4|S4uTJCy^0Rs!asyig2Ap?L?Ue)(
zT(k~gX}hZk{Rw@}dz)o8ok+b{+Oq<tSQ~fIbu4np68Bkitr$~OoMn7tgTPB^DjItR
z0&rM@?&N_-M1Rz0FnSyF7Mup!lVnay_2ak0X9Lw#RQ3=oY(_;2IkB)o8&@5sOOO$`
zPm+8I-zZR2Kjo9F`m!T;(Sn=2oAQ}uJVsejCF3m}`g|R=iTIK7O(3w{inK!{lhzUw
zjI5y#<V;fIstc?-bNV&f{7w9Y`IN!K<kDLRy1n8NQz<`u&ekD03B2j9Be0fj*oO40
z3{DhZ1|CY*Hj-j7H?-R0Jqxy|W6}u`cM!yxUPusv;J7!Jog*}YO6$0joh=M%Q^tm5
zkAyHU7%#6;bSK@mc`^q<nj5&Yfe|@@&Z-83<N$sgRS&~Hi*X+Iz2==}v63$ELK4Iw
zb_SN|$#1l*Uik2;+=p9!u=GA9HEC#Ymx{f(RaTuNPp%bE=VGW>XS7C=Rtd11KcP)#
zx&tuo8rnF@=ghuWD|BFMCR>i<(sE%<$-j&jHu`HtK_|GfazA8U(mSqGC#z5fLJ4Ua
zS1+92qIruBt8>kol_BUj7DA=8B|tZ84L(!x&K(*<o;RIXHD^<~V7ha+qp<i;JGSEP
z%<##-y`p~*l^=E=*OF_G>^%N2p^$^bAJ`B9GPey%B$*K4M<FuQ#>l5&W+4LC*lgkl
z;4GzRGLhdIFKvi8+#JEr7!1jT-gOZ^JZAkB(Y_hczDd1EBBR`qNdu2vhi{Z6JV`OA
z%g@{2zAyQWvNS$c8?k<VfR+p+uq#(Ec%jV*@7z%~50oV&mHgIN)Gy@`SLoR|rv4z}
zpNB%JgN_Rm(^RMG1|qw<x@UCZ&R|<yQ3`VASYKb;`#thB6MTL5zBqsO+-91+Gyc3W
z;+INoDMf^vhWn9Ah?T+fJ$FZ(tyGP7ixFq5+p1+-yAXULKp-D`_9;a!Yq+Mr5;G5F
z=iPC;x^K=_d_lT~@g^kKrV!E+jE@)g!BshC_HA2|V+4P@zt3#HL@*N1;>a{lnDgyq
zSQ~HW489r57=x18wcCmIy%UYoS=k(Bw2>KqUf)wHDm6t{=mGOlcRe|x>8}<F;V0XA
zvT#fFT{$R)7~iTl^I31xOOfCc?&iQIEKY#zk-%yrkyjJm<knV@gVFP2kzbsjJRDE)
zmSFGJ%{~gG;c#wSi-@~A426`Hw?AWp1j!0+)?uDY@<T3&Z3#f=y%9yyX2w!xwA8y>
z76$!zH_JJekOr$1#F1%d=L^=4`FUQ42-*AU{QNbxn2Gox%Z^B$*GJ}B1&v~jV{r5K
zOr}S_9xQ5KVaT3H8@GktY1bu}S0w56%gC<FrlAGx+?vhGWUqwC9&8n7^ETiMDBT=u
zDOdeCUastLb0%;vIpi7Vw3=ysmJvYxMN#-0{N#!y9B#>vgP#6?!qP&n(0?>cUbQ<z
zboRCR@qoc*{Ie23&C8wsOd&MCSLVH<Vr)WqJmqQ+%w<86N4jqYa|O<_U^c9d-A+9E
zqdfe3aJE1WIx3y<cdcyS8&!RTr>|LY&4j+i`mk1)>Ov@&^FTpBbCXyGU)0*?fX663
zI6j-iX16>M6Kh}rcj6EAs9<+^^<3v<fps8kGXFQ9>#x@2!4*EhhrVL@f2h}Aoj|eB
zB$kV_Vy)oT>;>(__(IO{D8$jK^5q?PU$1CUEck-*k;h4kUrnn0P(Ic(5B`8_e5@j^
zCQ1+-Ycww)S%j){e`XQtt;2L;Qhf1o%PKLIz75Mj%JiZ2f2KZtgGBZaLdOTr#H#ND
z7|N?xjA-6F%#4pW$1PkVLvMO3MWgrSW3gXhKy~+GLhijkYknQdUs1A%<H0vlfs<S_
zhj7gblJ^7N2zTnIFO~i@>E+6+uSklO^<z&k)B1KmQ3@YG92UU%r*>vThAn#(jD14F
z6o3C-j|m({dbmC$`wL^xEbq3pVh816W8?VGkGEeB{~x`<&mYjgJuCRDk}j+e%lfeH
zAAGC7?>PVVU(?~^g%Mw8{rR`>;}rgnjnRh4GP{?P9e?1yKQ_eQzR-Ux@++*!9`X#7
z`JeyFzdmvyBj7e~@2CHq;_x5u^vA{uAcl3Fej^F}fANlQBYd5UhL(R-KJqeWpPow{
z0MW;|hkSQ9t4+Vkmh!;^?AgEmI-BvkQaTsK0VMk}*9=Nj#pJS-jHKk1#Pkh2!=JtS
z6*Rm>X|>vaQAp+wOQ|hFgEUbHi@gJfYZPFZb8m<h?rDq`_yWHCeekc@H#+SX-j4>!
z^07U<k-p2)byeZY*@wS4hW?WLGz*%6CK5k!3Up?nWjia)+pgR5Tc-qd8VM5!%l!hl
zEKp}yg=EJjPAg*DO+<(1%;<i06SuS?j5-Sj8ISMJPX5XjgEyGtTE@PNk%RO5rDJTl
z_+bTsG4ge2K{xyci)X-<$PTKq+Xar&7wC^Ir)f&dyr%<i**g`FkJ=+actT)?K+I=>
zWLb^Y2JYdjmqB6S1l|rTT4i~PA>IDfHdO1Gwg<Y@OGYk3VRhMQzPDiuVa;9!8uvZh
z)+1K3u6-I*Jp~pcac0tDYdYun;ppoHUhB@qR_7Od>Y|pP(l}>w)?yMkd+JJmvg%$f
zAs-Lh(ffT0)?$pp?=08b)P<b40cC=GM-9SJ0(s<@i;(wRWCW$Z4aAer!|d4(93-ql
z9HnXKjrzb>Qm^L48MYF$?%j9rVDR5ijcW%zjFaCYez{0ymkkV3bD(gfSdFKo3hnmS
z>5UT9W&d@SwJ%Q}hP!Uwh?8#*psq;6kL`=%spxJsC~CikYG=F=)>&CChqzK9xrHBY
zo^Jk}GA(JsmDyg}>YSRU-jhc4lV=!-Y}}Nw=DzRe#YN8tL^eoH+JX1ut?@`y2gz}2
zRX`;*(C_>p^FEp4abCB3t?Iy6v$(lZx|io_0a8Dt4_9nd%wTS>I}epm(wW96Sa0(i
zQj4jD3*h7_S2Qe`nceN6+ZSrSOlTH$vMt-Z6}(yoTj0|#8n_W1ysFw)4EtZZ=|9c~
zWV1iKg`g$erw&Jck9kcqI{+RYPSB_sjsn{G7zS-&w-`6rI&&g3k+-1)fUo&|I12ab
zu}9<eM*I!-&gjaBF1u>b2^7M&;vf4(<|kXv{0V$3j}O~@hmBhr4zpB7mG4-*2g}y?
zRg}67ZUCh1X>CIk<?<#gS>1ULP_C`q3@#yz_=}b-s+P)>TwmqLt3>;DR`qbgD*et<
z;!7{<CGXDOxs%&+?6w1`erjBIJXdmxj5l$-;m}_E?RYoobt<z^#C`Ps$zD(~_8=-K
z+_iSnJQim4R{xgR4L6xFU|x|pq<hE=EBRg9iEW?vajluLNTXY_s?@33IzOc|X5n|!
zHC~Zk)IDyatj;?wN4-_+_ha3*EZ&Y@Gb1$WPUa?@)TrCX971{?Kkn{o!S!9f)53=-
z6hE@c#>lwJu%b(&=KsPf^q#--wgX+Cs)-6qRjzbJPOEXc;}n&3Li$_=jJmq1yU&a(
z-=L{x>XOK_^YzyZ()TV2o;}~o(RQNl@yP4vMe4YWOnT~U)4rxs2>(0Z?B{GKwi%G>
z(mvku0Ye#|TijDUEs0;SOzlpC2t&=dkX(@|s^yUd-Z#`SB)#P)4<567ucH4@)h5o~
z-$CD8HcfZ#%W9QYbx-2Vu$3HtYW8%Cz^)ka?VKv1Ywr%}{^WV4*}_tocHJ=-{&qEA
z@FW6=Tl}(d<>7rJS0JJ|02Wq8EpRt!{dEx(q3<E$uoy~*R)k?<10Jm+JToa#Hq-3T
zq!a;{aQp-^jPFC(4746QcI#l6Tnx{tXBSlVchNQo?*}+du;&RP)@%)f3-0Se`$M}R
zTy_9~=^aFkZm?U&556{iKtPOkQrLBd<>a;p@a`BANC@=OQN;5Ou}-3^yOUkmahKZq
z^+TcQw8ZGTkgdQnbiAiXSu+^>`6NaI+h@SGGbg48#MHeol$b=!1)&%Vt{H~(*5%u~
zVBhquN^=3@zz%4I@q3OygakTQ!Kg@@r_7ce(5W}<{4amAhn{;n!8mv5<G{fFo|Elw
zkmwTzbAr8bCdJs<Yov%xdk>iLBEpn~mpYc|SW0OJjYv&t(aL?07_1jz*SbvlK*hX(
z=<)n1ELWiXaRP^vJW~4|0IWBcXv{s^!DGMinX!G3&eYD6&cxTP2*D{X7-2*8!uL3b
zG~7kt?eh-)G!T$)beaQ6?{L#aMx^Otj_G`c@V*pozdSq&8^k^`EHd%*Gldwzo2ctt
zRH+3F4;$fO&TA_~a8;E8C^(Op0Dg0%12}1iCaks)F3E!nX@WeGJWP=z#G1u-hQ#Q8
z*R={6uLdMwwg}PMk#wle`KG+h<k5P1)xIE8<)FvXL^wVn_D(rUVJP$_6-U`df=A9h
z$hzrSgCT|Ly~eD2?ue7M!dn^@kV7HbU#er=Juo(^(FZ%kQ-NvLI|#iLXf8S_a`Os|
zGjGQW#pp9d>8Ef`5P?cD_o2?x4L9UXwR4lWk9{dTwz=+0zD_f3HicQV-D6#t_R1O*
zQD@TGxu&g@-h$^l4@>9BWG^hxi8(2&r?DvG_~O*{SKz1By*o_xVR0s`(!;yeV~;?$
z4rZ6mWw!XG$8q*9Z)ai#5ubwvL+^g+f|Qw3R!k|?q(D~THn{TJAq<UOdlv|kS)Q3d
zO!KuXtTq)+bXQRR<O-RNez2Q*`Cm*C!E={f*Jf)gIRkwkpCj0iA*9Yg6e*(jx}*cd
z&7D1XrBh!eCu@nw0)6kCBO|>A*Nn>wte}b2HW3#|rW;S>-jU6Lb64OC`Bzl%iCjeX
znMF6HO^0LgGl<%&)=NZrqAA9Wx<U5&K&RcjEr4aX&K}8yS|Y8otohuvsL8sb9wL--
zoQT^wGV5z=53xG%6x_;CrxK9hN1OvBAnvVy2xEI&E5Rw9N?2}4RJaG|vILVcL6pq&
z8YJwQ{*`|Sra!pOd0-!w(J?tLt@8xlp#6>W^E)@l_Q7>h)w2}QWz>5e)?UB7<YT5g
zD>4-AOMRb?7CZ+K5qE#J+$fcH*Z3rmE!O!Md%D5QAFb`>(%PlbN=jBBVU?BlctIR0
zFSl*^tWyyu`M8WEO^4srdI`Jl)mEEMywN9X$M6-mZk}M>hDcn7uPWjG!_Ji_BqfYl
zqqyj;?m<Ply%zb7m+M#C6`1uru4f?Imq1*gl5lA_W3f>4#les6v6f|)HZw?!)U|PM
zR(t(!p@q`**bn|F0|qklYfNf>3Sf<34z2l<m%y->;=5&>E=5RFiWK#5H#I)hTu$T`
zh$`OUoeu8ceC|J<h2kMm50r6N?NL8{Au3bfVjjVoj=uPkrJc{!%t@wrjcIL-aMda+
zrxVY8HM0msmbZ-yP_y7Ae9Jj3xLym9q_%)K2BPrGHYTxF3}n8W0%>oUVy#-PvuPK}
z2aI>7Ebg?~=eC)3foXTF_WBA<yGwlB_!Q`5&~vcwV;5L6v(OO_ICK5IxlU)&S>QXd
zKn4x;J;+ch`Xmb1y*iSCSjEN$qTVO|`rGdkYc+>rz55O28Vp+A^3<%%wsdE6iVi;b
zZKpfaLDMQWmZ0-~l|4Ab9Znpx*a{sw-eWF@zHDWiG*XoQ_NmFi`aq)`Z{hva+Fq|S
zKf7qG4s5YG{e??O4T2$HF%_$HZ4&7h8PFAmGzUMB7UT7$4{7)kxBrU>aKAlh_QBCa
z0P(4Un2t>6w*!}R`3-o^=3U9d(1#P92xWqt#D@lSrK_AfU`0}&F&($RV`kcKi{Gfg
z7TnwlRcD!ar5EqnHT7GW1P*Ob8O*;2L5fc_fbLm~O&=@d2;p70eNe!MYs{v8PEuue
z{A!vt5|jwJi&R-ztX1WBQryF|N(yPmV27(u4<)Zv;K{b^0+cGSdz+MLeQ3+tCX631
zhp~vE@zLiN^xN6D+|z7)i5buFl1AY?vA!#&s&8Bi*`zNp=Vjw|31YH@$_Pn&7^eo}
z^i~KWPFc&unh{2{ndqty<0smld9@!Rpmi46(fG6m7}^rHSn2&L9|;;-eKfTBEvh^x
z!{}=?QehaL{4IOtv^`3sLhtCriXO?^7)*givE_tMhl4Oj<D*;evOUV~LrN%4j9u6j
zBLk)hirhCz=92Pk!gPJKo+Nj_Lx$DpuIgZ-IUTD?e4CmLb<p(1)E7McR-Wei7cjBW
zmyE;=@(VO%c3A8NVPFy9WH3Uvr!@@opr&o{-w4%{ooso0xqe5>f+w|Iz;DcXDBi1o
zoSutoA#bnDW#F@`KaRH1O7dd=`DGii?`{9z_s5LT$~g#F^PK+kAU%F<>r#$??J^~W
z;_@i-=Sc6m22)x7c>e)AQm%+#0U`ADDXJjvyj6qWD(qVME`m@XjzR`4F8w;$e~eP6
zKq*hoUEthrxT7JHlEV-5@{Cf%dIIlU@kD*gOx)l%R1gj7PQt4klxL-`L{F!BYiej1
zzq#aVpmnZU<$0*goaFJHPuyBqZ7Drb(!1Vv%WBNFI>||Q$9;(S&}2w*{K3q(O3`J0
z<7Po1l+$uGdg?uPLCeQl!}FT2<^@xmk9RF%57U(E7{$L9)+JHZUgSd}71m!e=0**b
zI1fc|o4Nd&B;ofwBC<QD?k4g)(I4X^R6RE94IyM3nxy`mph~5ZSKXUNMmOAc!G55=
z*U*?<_{-g@v1Ps<7~RVi8w*)v^^A3#Ao2jG^J^Uh(S8M7dhmmH|Nb1yx#bSHe^rb(
z9s0cM60*`ypIdpgL37dtr&f6BQ$PIPs&?m$zI^im-$q^!S$Qb^SyuZ)el*m7^Qaz<
z%R{F@przCKu03Osx<Xe?;d$xud5n3^S$WSOF;CH$;2IXhy-Uqp7&VO`>4Z_=>+!!n
zOXh*R#rK+W)~VaFEwB8?m+RV;SS3wgy>=faqjDq9FpMP|+hv!IR)6MC7+XnHijt+^
znc6!0QVM-_w1Xq>QTaJO4c0=stBdO}CcMQfPfyF+w0!cEBJrHI)S_nNj*^xN>J`t2
zr;Mn3!N5RpVdn6z@FDddwae&I^gQ3blq$NL-j$sp$Q-3vtBPB3MXE|EGjF<=sjR4@
z4jwp<3n3tFbQy>Pi8qe)8Jf|mPdUaoX)73in<9fFdne^}rll62+4~HMMdode5T1RW
zL+tp)a{M`q22$U*1+g_%9})de(M{pt6&f`aAyZ~zf`N>TNgsO=L?j|kD<pi%R=&<J
zBkOL9tCWtl1y;MtJ3l_3WFs!U9;~DTq^PZt11M&D!pO~n+;zs%U_bC(SXb-F1B;0`
z#5;iPL;NFX2YK!(HvPo${fZkPRho1o5^UeQRvc052JEM=%MGhh13T5OMZteurCljk
zG-h{8vN?jVqFx4(A5b4!I4v{w<OqgIIgZiPlI0Ual$SiXR@brVe0v}axmWk=S=3vS
zhOb_*gM)=CgFTpoR0}P{^HQ@|WjpZYU~czj-skJoo38^--8kq7)i%GaYVkQ0l+sNr
zaPN4&KAA3HIGMT_ye#EVeaejdovg!G^ynu(N3h6FgDIy04|xX)m#sogjowK=xO_N$
zF}p;g^OH%W*kjo{yogt<WmZn|m|<U+qx!}Aacj-c!+Yc}%d~>BF{r)PNIJ9=`0y)~
zp&|)LDPFYj?j1I3ZW#?x!`~kP8<SpGxSjHe9g9*+mR%sVt<Jn8YM_^XaN!$0N+!r`
z5Jt5W7g;~X?Kvm(lF!-zV|UCz2KSRI&3uT6QgwfJgyydbkKqPgcpswH0_jD*4wx`G
z6~TD;J&<Mjf^%z1eIG`9uQ-T5KJsZa%IP#Rj(LwLNFpY``NlR8Q8$S=3K3#QXHD|+
z{m17KHOEaZs`1&Cg@FO>lL}@-gu6{4j$Huh43tqBILA*OoUVb}e)yQ+<XI;$ZlSuE
zNtn*`CSBb<ui=cS)}SA8qNFX>;`C_3|5~!Y{bAo8!DU_Ni_J7PkqC2k{Sy9d{sk~2
zjdI9oeROwAd6Jfi@6LEcoxaOdviDQdWxD%|S3BQ3*_sXA?*qG!fKQ6ud9TaL41fdL
z0l70t=hDYRyGhAR1PS-18Y<!t%VO@^mpGi8i0qrNI|791kWYvsYpJjskQ^egb814O
zat-4vi2ON!x=}bF#xh!dabn^s|2`B1U(W#hJ~rFeM%?jpHtyBE4?ANV&t9HiPx5^g
zjjO+#DCzHL>EnIS0qOe8%!f~Y5HX)0fV6mcC0=KbY-Xg6$iEccRG#Y6q$AqU6E_3N
zIr#zgO_eoceEGx6&n+}9URyUg7LG?+ZGD4aE?8iB=%8etd78cZ=3t{TS&CVEM90o0
zqcHBFFmBR(1uN&y9Q~N(0)t5$0T^F`ef8L{TNBYJ_WbU^djZuY`Zi~n?m4>bMGZEK
z)(3YxK-KB`l1WfrFeXgXhI*Oxwcp)osgY79Y(+sI7@Gt|hJ4~yP8I@cTnvuVfpQv#
zytJUXhm(Kgg~-;V+*dInrjx4+r@-OiqFQBV_PQU^quq!|B7-wps)Q8xUl`C|D6WzZ
z5oUwo2d!1fgJ^UrGcurT*)%D@f!K^6m}Q>7j0R?MVs}6T$fWY5+F06Pb=HAR#k}LO
zia>uL`3uGVXt|9pr9zSD*LCBMz1zJD7pqkjM2UIJ)JSiYh&XR>2VRe#=Gr8VV4(zR
zzB*HG%Zu5Iosab6-*_*}&Pkr8-D?e}Bi^?@kd{F2Mxjywg=T+_{ic_DFW94a2k^-3
zffd*NZ8<;9J6U6ARPFPXqzsqTq=v;lNj&%r-kf7|(zKG)+`K9J$sgYc^r)M-Y{*)<
zPi`{zxPgwsZNoa-g4?XgZ}g*@0XN0lZi#|^hU9Up^XH#CyzdoUTwEUrL^WT+3ALA~
zvRIS~j|PaJBl9ss7-%F@AczSGjl1_s5a5JDDFdBnjMyED#EPGJtEPGVLhr^FU(Ne{
zo^tETE1i!6h<>Q*NL3~8Ryfk?gkd(g{S7HpmH7=(>@Nhy!zJuOM7r4;`e|Kfhbm6U
z&>(l+1*865ED7O(mfG1@O3^~|?-EJHBk%3Q<p8x<kb-82CVb0%uhgyv=a3qsR`bK(
zRbiR0kE7~+K3upr5RmKAKz(xs+&wm5f2xu=ITV?6JAL=C*jd72nEG?Cgg5zmdE+2K
zB180`+Bd(1)1jxOFYy5GEAevJxPSJ3R9lYT$G8`aZ`-xV$?ClyPZ-cGXZAt76Hn0D
zqVU(M#`>%A0fP$w2|R7!zOe#gFUs35a4duakqO=^B45KDLw`q(-RI|#rx9+Z$idUT
zm3!yY*7=AD6;Y2ywkWC^wiVm-KOY0w#yQ=C0wh6Lm?Ws<+L$ny%k8;|88`Ks_TNr~
zl<`dScP_%9*PZ3G6#^u{%!Wd{rk!}N6}kw^lZSq`z>5&h#9YhgJKxS_tq|0txc%{+
zv8Sn{*|(sb(=O)?MHxpc^^0|FlD1Qh^DaDkS0{DDl>TUGDz)#65UQNUn;zvTW{JPp
zVIlc<Vtf;r+ZjI^g4S+l0l<jW*u8rm`l;DOI+O~WIsn-xfPwYd^|d|z0uF1&ex*2Y
zL70*2sVMFxz88o1JgloH`0B++<){i_c33)oN1Yfu!|IzppYFn_NB&Igd;zC*{glu2
zk<O|jG><O?BW@Q<<G=v(!vs@&BoXt!f)N`oWZCZA8{VVP)En~VKJ5>WM}ksNAWsj=
zU6d`z4)xCl#PIo7X0F82w?9TQyV!sjch5fnwj4TCD$9)i4FgHuT#W&CHTaCw1D5!B
z{tSKTa&nwbn^rWO`Mw0%xlnz8<S)Hx`55TUcvN5Nwkw><+nd>ZHK!~L*I$M=m+Nke
zXt8u=amAmw>SYgb>v$tasIopfFb6QBg9i)ab1dJMD9-0{&aobS+4wc54z5(13s_fE
z`aZJTOkGt7)v41hv{~Mw(I2Qx%|+sbmh%2vHVg|<h@(9fJ{D~FB6fOX^rBdu`ZV0y
z+v2UpLY3x<l-~Mc)yEN6u9@d+veIKufdcN+tv2WSe4ei3boTqxov`f$ZRbj1RD*x5
zHS-+HKh7k4Xf>-Splq*O&mi)V3&&Jfp2<pn-0p=SbDqmnk&D5D3n8ispJ;q`dR^>i
zq()-S6)C*<$^7rtO=OPARUK&XZ5Z~)If*NDk#K^|9pppm3Q`8@AIz~W|E?jEH*e7g
zq>>ovRi&0w3hP)iCE#vd>3@%z47`Ha?LPrGIJ!cczHX8;nDO8zGGggUW@+L*K+|e`
z7W5Z&4D7a$vBJ8#`4`}uS8~+P$}V>18j6C~);sZ7hZp?APoxe!dqgV2j|kux#%3Pj
zRDzVGV(+(AFMcVU&fh{W7r4QLLVK&1xFZ>!h<LqXtu9J?@)41xr9aUxmWn}_*b1y5
zm)jSzK-;DI$J%m~e6ZDhaC3eJ8xBf``Cb9?=I)fwq)Yzmn%mP~z(MzA*glZ`Oy_e0
zvW(CE`uyD)=fQ`u#dtdDhO>bQ?vO4*ICmzGAIdJSPO6@%j_jyh)z#_L<8?=qWwz;;
zw2$Ba#tS>}Zg6P9mpOLvKyl0OsJ+pT3iOF12Woi6eYWy|@!}HL2+W3g@kf@lh|ji&
zB0Zxds7;G8Uqi>E#`7f}q9se0s;RXJvPNlu(_QDVR*nD#Ci^|M!Z|2D#%>!Lu}uJV
zKkp#!HvSVHtRUMyH7h@#gU}fK8ENv<ifH<&9q>ClZlO1NLdl+?!5=vJRj!L6`bX(R
zu<jNa+w!-eA|MYKx6}j`M8TSQX`qr?0$a8!I=ec!<HlHh4p4?GD<>3n2qUodR@Tnj
zWn&8;^FUlx97oR?u*I!y>Q2eC(Ny(0Iay5EA&esreIma}59~D6OaZ2~VtbyD=eG3}
zJL=v3JOfq-4x!4%a9RtT0MxvvD$U$SEUt>qf6eIlA|_2=ebco>qxpB?{Q`qe3S)MP
zoij04W3B9iw-QS#^0j7S|Kyjg2Rgg8<5eqXjD9wPgo!?p$3-jb)X_hI{R2d!+1Ke_
zGd>^@#bu%_fGW4PNo>aKFDkRuMVEu*YjPR!#(C`F7I@|93K^$^Q<D+lxnb_z<wfRS
z72o;e1ErNhqcHrlP*HI(b3PP1CsUcaT%uvj>Mb6k#d01bLrtIZn_gXgbPh!YbIh62
ziCK^#7&IKGx>Pw6*qQJr!OTwi2f<7S3QPu6!AhRK!?d7pKE{vjC4%p3KTQ`|w#{$k
zGiXUDOFM?!3Veb&uiit7G6hpD`rn7{50traH!87o++YyoP)_&KJT0`#lR`X{>7L#T
zxk6ElET7auUVmqn@st=C^gm6u+@imFb29Gj8%RTJJv;$P702Hs!n{p1m}9|@KABe(
zlLfiAviKs=uU<r1NW)I*ky&gH1qT)RPWnZ3g!u)Y)b_m><LZ`(5~Ez}j1`v0(82%R
z#^sg7?(Fdxto5`=<UnK;+Tx&YfF?W2?||-YdvR|?fZChQe5us^CDN8-nMj|&+Re$8
zr7gzH%AiFwdJgFi(9FU)`taJHSUTFjdhXgbI-({RA5>jW?y{G}DT=w?Svoq!8;h|#
zYPAj?GQ+7{+wD0$8EKJ%JnP7;;ev*XyZ6M0Y0aLBtrOwnP0FG5@?*q%taFl)3n0f~
z4vg*B96jRz5awd@_0oY_lf-_jyTYa8Yx)FPTJI`(tC$Xb-=acC-}o2Ia^V545u)y`
z<p-@|NU}(7tB)gYcUcUk8U%$C(>F-|DmDL$dU4VJ598?|+=a9MDnBKO8pI>Gm!8_@
zc?URvCI@<&);(v)BWc3B3vB-}3$cn*p>E-h2=~mqJVJ{F;hs@6lmgV(gtrkOSdK%8
zp<|zerdd3DA2Pa-+Zn5mx=dXJgOZy$&le^`%PU=)@YDPRk+B{VG1}BPH;fad4u=8O
z%5a_ADCRx%SNY3xkKX5#4r}A24>4`rfvmyTdja%Q_cUUs5q*M&!OkB<s!c?{%_oN$
zEX&Ew4)xd!^A}B}L{&ed?7d9?kelOH6Vah2u`IzKj#+<vir{Bh>gL&sx8upZ>bp8|
zEAmNmk^JWLI*=7d<*dJv&uAc4v6AFZ^4U=Bzmw10CjSrenN~&tow2RK8E*9`GcbKp
z@t_ekOCU9irF1L%|JZx;c&yg;|9kJY2B{RH2+5csnKKKS=SavUQ<qs~CPXC3JYF(H
z=6RO65Hgc_44G##PrqZ?wfDZCd)N2*{+@rI|9b7)&2_DHu5+EoaeO}SkL|RR)WqH=
zP?}{qLy;RJu&gpHyL%xg0(&h2TVeM?^(x!Z{le8B+Avv&L~M%W?Y2TTLDf~A@d5~L
zrPWB+4{&@%E;!U5icM5@+Q95}QfCYQ>S!qD8BK&0jc}up&jqg?XrInD{*l0jd$-w5
zLQYrjh%g!@z@`f-wU@`6K4zroW4z*udV*_|jN{#{n@x4Pmf?nu$=r_&rsWMGb<-Of
zEiLn~D(hJl?^b3LZ1hS#RHWG_S+PXvrP>oBu_()PjZVa~)*MzZT0j9h(YSgqLDqhM
zmYYZm-PE7K_UL|sO#7v#h}~hIF4NV4S$0C2x+eZen|rc??e>{%JRMMj9_)6qam6i}
zu(ds|qc=>oxm?y=hm!qmrkjrccB8Rb8@i~=sLbNLv^_Zsi%!?-<>V*viv*HZ{y%6*
zf5~J1al~YL1{3WN7N2k5-4B3RI*GkMQ+%79><IN)zq_@Sw|H?<=c#zE+`N4Hh^)RU
z-Y%w46$uyHG?RF5G}P8Ug=fW%_q%6B6MlR}KWurkjHk>#xmJi7>9q9|;O(9-dlC4L
zzxQANbm0LYN@nx1R}<f1pz2qUt^1Mk^lcF=suX`!*4_#}v{aI?t57~xczXt8WAX+b
zacPYDT}w2DO6ojbH`f&xs1Y00UT8Xt+vz+(q-^3Nh?Gr-4QJ@&2d$gz5x-Cyxey`%
z8XRJo;_p_RB(5VT4`}6;Pl_X{t2Q8g-MU71N*m<uGwHl}L`RNd)8*-&xSA|{@)N&d
zJJlg63j(b=5W;sl0tm~T#KH<V!Om5p^ltH8qLe!uIoksnrO%&?6R-fPLr>l%FjV?z
ztfrGs&vw@-c+S|d^NB*wGJ-wy_@W6gw<jtmAKe4Nn^fQX?}gXD?>jCvSK!KJOIVQq
z=+fkzpfkB&Xf}utbN0mTJ$Hqt#61s?AuTZQi_(m5JQXU#n;NdDAmOz(N!ZG9?tS8M
zhI9ZXX*0JnF@bYJ0Hf(q@q9zqHHs~9cRN*Uuig>*UFDhf(XGU(!_br;1jD5U-yb}`
zLWOvpfg9-6eE0M1KR=gHF6DLb6mWs)uhcp{uU*Ov;-2lc^n~7^u;~~RO7+}33QRQ$
zKO7@@hs@TIxgf=BCan4^Pc?(xy(*c)-Xq3&gpEVXVr4c1FW^}a(U@`8{?}Ui$8~Zk
zMkHWn9zM^1(J%bB2jcA0_pz`dhGW_LzWo40{p0uk5C5D&WHZI2Jj7@IhadcJd65^Q
z>hfwg_^kTlcK-Lz{Nx)aA>zIn>nVd9e<}?BrGNdikHaVhB_)&8TUo7N6bAqP``(x#
za<~4Nw<CXjtT)<7@Z%mR=vfYoH&04^7uAIycL`ZJWqftdfb7mqhF^WH59FLbUadl2
zmzNOL%wolXKqfPE{+k<&n=cw}=z^2Q3=FFo4jT&@fNeTj66G*V9eNMX<^D}|Z;Mel
zTdwNJ4w^zI$_8{eMiC?yL?=4Jgrp1S2iBBta5!`LYS>H7lx*Ld&VAQ`zRUD(CQ^9M
zGtqudymW8w7pq060SVL7e??jLeLG>l`nz35oA@$y$lcgaeaoO|dWTTA5&L}k=?|})
zDSi1yL7dQUFdrZ!gycL11X&9rI2i!?O<a5T<%kVn7|xp<<~#y%f6vZ)9C3gG!$Bcy
zH`X$?k0m{*6ApE=i1wU$8;N@#fM<{){pIXsKZ)P^<*JpeCf+?78-TZW__<9<q4YJC
z)pzlU3HHxBU1t#m0CJ*CMHFov=toT3ZFn;#XBII`E|iWx-4NB0hnES>IF0@1QhM|0
z#D8krp8rqUw&`biZdQLBJM|E<!PBOq3^&H0C|-807^(6gI=|}~+_w)A@kI#V3!!*H
zZt$HBj}S{mVqNvCzKU<filE(J)52tD@B~WpKfkK@Z~e_80qQk~U>2H$28GZwDvl3P
zukUt%Zu&b29=nq7z~22beCFaE95UkC<+0kGlvGC;>8RcFa>{e#d0pl3K>CXW`Oj9K
zO9<kaYPcx<BNxGd6`))QFbrodJnbAe7cze>`z@R0s56o@3NE9=Y)9zaHyJAGdvvag
zL~S?2WPNcji{R$Ov-@Qi34gEva0RC9Hx|4B`BkOj#lqB$M@6i3LK#r&0dJqM9O<!U
z;pvRTajigJAn93^Cp%T{OsaCNv!&g6r!FOE^f9q(jk9*6QGElm{qyIO0C|I6AZ`BZ
z=!xe7XqFPds4=gA8R@r~_IT-BVDPg)%@#hfuXEj`kWe?=(c`%<m`fX9;Em(IFK9sD
znK=)tIKgBi_s0hn2?>M9)2jY5c({za9DQeqB;)6fUIl(wGM#L3$G&g(Z}Ai^)IhYt
zhyPI)N4)cMSzKLyT-?uOae~brPVPG%1@7zAgIW<P`|tu=(S8V>5i?L*?TnvS5v-1l
zP?i)Fb{O5b8myJ^g}Ebs|Ei}53*oZEot)?b*`&b{)fCgz!I)t~AHyWJtWyO{uNC^g
zeZDcC@a_w9vd{HNi_8Z8F2}01k$Yhpk{hS>NBb4T4KkbEdGeS#JZwoj=Ti5tN4e;h
zo^C24H|)2LXLLQCN7Wy4`bDOR@I6;qw|uFBUInSo1r6tRlgo(lO+_54E*vFM-7)59
z8QZ`eVD}9ObGJSk5>qyE!>~WBB?}~Q{lc#qv@b5o>#kAescv+I&E6HH_LjCqVLA9t
z?g+2T_ZfT7oHiM=ka_jT<io`{(^CEqAtFV#f|W)09Y&)1&ZI5zvkiDK{qz;!5+Vkx
zT$)62`_I55rl|Kow4XvYB*EV@0R&(MbSN82ai4Ew>j=+_%cSS}A=3-ax}U+*SL5*j
zd?~VnP!lb14m_;l6F?#a9iRp5_^ve+W&a8B0xm#~X{ZKBQwOd4O(#V@fX$v|aqIxW
zE;((K3~%|q2iSO$M%H~7gweG+qgsRHc5s4at<++s7HZGM)xtg_2BZT7A+J<iPP?`+
zK55||)csGn+mx=&)#L!*)onk1^C3PPaLg6Mma@GE5x_;>oPBlkz{o<*8AwdD9V2q?
z8twvhRweSWGJs-e2FJu^;|Iv`2V5}H3(x42;>dwv?M=R6`v=Iwn1?fLQSDC9%LR7S
z2IoU0XR%^$-W&;(M2Tj7qIvxe_Ua*0=iOzq<oJ+!U4LaXx1C|1XnaBXdIEO>9>WvB
z`YdER@1+(p^oC4im)<nmj{-H!{wBo3E^~%7Kp4<rf+O^<iouxv4h}g(Ubcu^A#)gd
z4=?+Ig^w6tg8RZ_P6Bjy8*q!nYJgTPLYNxlJmsDRq~>037u6}N#nxkrRsy4UtNXXX
zUbG+*niB0nYJGP%r(^vik{x1$%)D+V$UFiHc=!%XkOx3QHM<jIB_`YtoOz-$^zAiD
z+5U{@x1@7T)B6!!Ug12nryFrH^%w2UrY|Psr@%PIQ>X7fongNyw`;{Fo2D_ekt4nk
zywq&@aL6RnEdk)vPB!R^;sjbgF2s6WTbV7fm3VK}E=jw!^vy-S2g$RvgyF?0kA2KQ
z>l5=v^xbeqw#eOKIo&mp#k1<e!3dDc2G-yXhkPyUtyXYF$_qE3HtYX1v?h0jWxy~o
zav~a=K6FC;?KZt5S9?BalHy4ql6?o5KyO6^an}Usd2iPlmn9m>9zcBV0Al59=AOJ%
z2>bU=j$tW!`{m}(^Sp&{5ZuGjGxVB=rWThFSs~X@#QTRd;T8}tJh=n4Z7-~#LFCmQ
zg0eR}Sl&RABwepyAEhD)cqe`Y15bwlPOBe<$nJ4fNA|ATJHg!^YM##^xnLH_!>(Y~
zR^sDddjAJPm0YV7@iZ&nnjm3r_6o(`so4C;Jb(ldoY}i`1P6N;WjB5v=7ftyFiG8+
z1<ly}9d$!$X7ziZ`YTJ47!~xKMn=ud_mnkAq{f{hx{uEzzT&GU5doqBu&koc6#jvw
z+JLlsus>FU^<p^>{^y<6OgxMP7PbGIzNpvLDDo-l0D6w#e%wx8+ok)=UrufK`N`}u
zgB{IJ<sk`{Gtx33@@)=dsX}}NUbm=O1AZGPKiKpjx;X^>Q`9T9tQ=PPI8*%KzJl<`
zuCSLFQMw%S-eME+BTBtbO=ghh=jz=n`5aw$t8k_tXM^(?bd}6!fW}9%cUPi=Ayb|>
z6Tx$OIdPCbL&w>Y=16<09PCaCmPGG=OLR*2yC9{b(Ztl7HiL|c@a8!?sCT%>Ua{?l
zzb$IZMayr1KClF)!5_Ngcaq4kiH-fKl!7ix$xusll_8Y7ij3(fJX<n|h(%E`ut>&q
z5bR0x7o=*p1(;|QTqc3bTei2kG1XSpf@_x-3EUo}bT7Slm_};sK!NLbxa0-%Z|m6c
zf2Jk~eSzGOwU*b{TWY;YCMT08hN0ITHp8cCFbVRjNi&dOqTJ&1ZyrgfqKJ+fhsvoh
z-o_gSyB~1yahf>BxhB{cnL;G@4AYz<>?Sa%rR0rRpvdglFq91&ItCKyd;SJFfQB0d
zdyi@m>*4&p0!orDw#)PR3E?)BkAfW6ra#vnP<RfFfie&OK!8^9t!N)|y#^t*bId&r
znE`($7n_QdUX<=EN7Hyy{$gaRe(Luf3E{siTxYO*j&Ijh^ZVPac{M(Jx#;4^xMo++
z_4`ddawErbI|ap8RS3$<jz8Zw`%uN15n$6SP$xzs^RjVAb*RYpx!uOgMu+=+scDY*
z?OW@Z3zV$c>UyZkoQy);$~F*=5@24?vMzFBob=Dvb8o`dx;V>Dq0@}{Dw53>gVGZQ
zgqVz)dvzNYC6d!|7L%A%y6#>{<Hk_O9dy-3!%Q$>hU<5Iy&>lSx$*Jd<JQB+@%_bU
z&Uh`rcAo|?FS5!jkzBZeo`_=2{kUO|C~<-2d{}r7E9(akZa%NWinn8zVVpJ9u;{8T
zu$v5q)3w7UiyFvJuXxMzVLYfRj)lFy+FtfJ(DMi-zJOS`2F<Zgxr2x})Zf#Da<%IM
zhNxMhR#MVJ$tvtV%d!BBp`7(%Kh|MMNrondpiy{~_}3;o(I`nud|z8R8p2P!<gDQ$
zg6&zRYl-TFYi9jAmo@GpB)<U``TY8NDI`Xqkuur&+C+wa%zY-4%F-M=jgDPLj_Heu
zlX;$Vl&dQtR+y;e7;->m7<F-z#<8@>#*twP+tP|<`3QT-3HA6b=5de2LY*O20eHTg
zdO{w9IdoW`;^~HoR@~J$Go~N-h|wlDTTrWCXI9r3uP>LGr(mb=IY_qi-X%7cC0-PB
zN%o`xnQ6pSG?BfBExfm?3=f{LnOFN`e`2%KPVC!gRKWd(jGoE;DWBe7A+<>&NQ|PN
zKI(psx_L}%sX-H;tNsc#p`YJ-;iU&FC+DZ$<~UJc&{+ogop*3NJo}9VtDpO}i(h&q
zs<T%g^yVgc3AG6;v&Yc}cYH*Y^EiMm15Vv`u1N~f1%y(aUQvxza(ukGTIo+|ZY!E3
zMmSYpYv@Y{Y0{%mch%+;TBnDQ59ILs1PDaQTI5VC2ssr!O;y<viidwce(8(*>5rlY
zIi}d;RHtO5;DgH0?np~bd~9qn#~F*X=_`G-Wic_rvUrbAB=)}N8R}{Ysfoze=mQ`K
z;!9!$1u?Qev;<=dZDfih>OxBv3X>>2tct^{zU0&tL!O^E|M+f4iK($4Q=@|(ii{Fx
zPwinN^;F;mDOb+b>mjNc@L7_-$j7cD$__hI<>z)9YwM;I{j1H1EG<lz3W#uP8OR|+
zFmAGBfiz(Mtk5Dp;R3(j`9aQgHW_bh=gSJ%4Z@LzjZRn}hGEE)6Ru0q%fEAPIi<^H
zbu>cd&hgF+4#P4M_vbK<7-9xxzsfg9hI{x0un|1B@A-M!j=@#v0^|qz$B`eF@>z9|
zp4d`ZI=<E~yeL-W%I+xyO+iaJGM#VQs+B@Go|8ur$HJ1`%1Ih#=Z{MH<T`7L>aGx+
z_>$VrB7jbEXdMJqPa+DJK1TVv%bFyqFLEmm%~Q44%XXeHcszTucKjRVy}%N<X{s_$
zF3t?Y6z(ve>{3FpY!yv=*1S>8Hz*C(kEq~0>%B~Z^E<KHg@x+Jc)N(s$#XKuFovvb
zuYOg>tbI3_iw-;TYjSFjd~A~N*i~G6!kX^x_^rfp!zRB%=FY3<-qcgsJ*YRm%~2a}
zQ=GV<PvJ>v$Lp{$Xv6=<n&Rv|JdcMh6Rp1NCK`LmA-Br<3ru9mS{Zju6iY%Jr8cSi
zH$O-36Ub2Hrn}23({BcP^a@MPVz!Mb99Lh$qbTlkX}UZ03yiE8(^4imEvH%=`PV%D
zus@?A(JO<>j;wKd4g-{3mHHe7E#ln26nv9#f?FS>$gEwqgyRVT!LG`xS#Vd0qJc3a
z9A3(>Dvf~2?*s0QSjL~9=It+Mg%KN)3rhUxzI0r40TEXb5@ZezcXa+hsT()W-Xt=)
zmMJIOnBcDK^{_{e@aL|mctA481@zF7wc6h>tAyoUi21$sJ)-77X46nS^5R5JHR32N
zBnS97e483+z!2pfYk0lc+FSbsjVgT^l8@U^o=JmxNSIP+(Cc-rxK1tUKH60r9dptv
zBwxX+x%wgETx%B2dJvp{HV27hYZj(0LBI%+so&m2I1v<k*tuGQoIZ|i(G<~HF_q>b
z1n&|kAqJ80E9k6PV!Wa9SdG9okMOJyc&Qwrnl4-C^BGwjmSDAO7A<P77Fz%lLt5BJ
zu*6Ku1kHl$X1SV2#&(jCDw8ed_0Bl6Ge#u=M%c}wrOlO855{Rp!xDwi!gz_?f$X??
zs}tF+o}QcGXpX`fV&fY)tF!DV8sR_Dp0Y}NofDX(R@sOfZJ4DROehOlSR7qRsg=H+
zY5l?u&ttRfbuoEoU%__3F)#bVd1yM6APLr{FuNN%Go~9+Vje4q9?2Sn62KT7MK1<E
zJMqhZE|7gc%UGS4>sQ_^{9$sC`Krh&dSRfbS-74ya@~$*=rJ5}GT7u_tLe9gP1=XH
zmOOKRXmzmRd@#wVm)Gxnuor39X+_A{t5)|A9n~8VwE+WKH!ZDPV#P0SvZJ+)#5fpn
zEUP#k2<D0PlAGVa-z|bcF~?^re(4Ub^v@I7tbxq06?SA^M1J}#mv1bqz9<%>I1$;7
z`E4k35Wv2cb=M3vVChM6#hm^5(&IrI&pyYz629*!qgd^*ZciZb(+(`d$VhvA$ibXB
z!cR!<%C*(s#&6+4J3zbXs52fFAwEyi;!QoO577c<MAEQFaGA5AWT-#rihSZ5vSo2|
z6UWo}c$?!$xGn$k-sA0a71zqOJYg@|DuyEz1xj|P+<k3LSQsaJj*-fAXv>I|N;#>9
zracE1h4PWN0RMQ0be5oxoTGE86G9BH;mw4p76mLY{~jqy<6eCs{24+dKv*?wxLOMc
z7+F|rwO|rCnxnM^BHf3YPb^Cn8)Cat@r2wp240F>tQJgXmjb6g^4;Q2SH!!72<}dw
z+_x-AnDJQ9OgByZ7&~N&6d<`o)D1QI`0;Tt31gQhAjy#aOhc@(6k5Zl_3mK|(VSGB
zBebh*3-I<31*X-r<GAZrDgboUhDnr14+)iJ@n<m<m?E!kz#UakD$T10!j_PQi$o^=
z;_Z9&Y!)ubUtG-@zo(9^8T4_Pm4BFJe<bv;{2ZPb=Zb|8g2k|>`x9heAFFZ+?<OH{
zmeZbxP1<evHl2t&9Xs>$(<-x7*r_IF68xHXWnVN3b6ra$cb`e3ip%t5KdBP>iR<@}
z$#oz#>%%L&ZV0j{iRd%45`+6h!wWQ<!sxN`%QD0#&gUp_N!P1NXyH!~GN!h^E|yoK
z)4)F_4!M(e)aC-s;{3|Csj0nZK?QdOfB4n}HY4kZiBfjTigLxa!IAnTzq0dVXU&E>
z_7r$ivQ0Ik36V$Z?D{uw+2Mt}yv<}1+wNCRQ_9?jyIK%>p`9z;&MfK(<?0gg)wIl^
z)8C?EwjSsVO(906n9{Xf1_CF(<Yn)@Q@2S7+6$FJBpnq{y6coZ*2ExdbM#4Rfp=x{
zHR>`|15(us+mG=n92=;`eyVwW6OZp_gYyp$2Z-UK-HBf{K2z6=`5k8+B^(9Lzam)L
zcK!5s<wa<zp0<hK$3DquX^47+{jSkVm1^WRh^(%xP6VscaJ`N2T_vRN%+2zv^<Pk-
zOE#Stp__qvT*v?N8(>j6;I*g*sDUc36V|`{9wXIKyCc<!Y`>Z|n%5>uQw5kOow_4d
zZ{Y{81<<8l7<;J5!ReoXoO5!BR)dJtZVVqgY7=fu7|!Ky$Kfit`eO1H2G57RT5}S|
zleMWqRga9cYKh8>X&^4Tme7oM`3VOGkX&m=T7`84J55oNdwYRyz9U1jgSh!`iTk_O
z46~DL`a{yg(Sj8;8g;0ctKr0-3rJvjcku~=nv-(%1c`f*$;dSSZJfmgwPD}~Ir(4`
z`(x9_Xi0W1COA-T?kt{AF6oL}Gb5ObQU9RDv51K|ZSwQ3W*z|yHPq`R+W$=1xf^#v
zA`BnPx#&t1^fUx>mK@YZ$%3)T-V1Lb<s*&@FfH5p{C(>sgkpU6SLP@c-B&8}i?ZZz
zjJO$qlwqU$Ye-wRo4k9qOo}vlJoVSlhg!Ev(m}0mLk4sEm4%#YJZVv|n%`WQ-+N@B
z)gx6(#P`XqG^;6HXrmYMMxSVn2Q#ToCCql5vE!gW)Vb}^{Y2-s`=2^DmLxagzjSV;
z#CR^#eF*#mhn|BBaj<{-<KCY)k2e7*C#8khP4V|abntBjHorXccOQQR)y)}wbU^d*
z57Fxufh&vziR8K9pl3Drae{Bw#xt9Sf5;0Pa@-i?>%Zd|bDA)UE8solRYN|;pv1!A
zIP1Csq>(h-4KGx-<d-Qc)r!WiCggYb`?|{n!Eah8OiuLm|M}dRX8S#rFz!USXq&r=
z7kfP>>R^sd0xvl<6fXH$o{x_@=jeKNFk%^pIM$M+u&yj6qVGRBlO&_QdT-XZ%P?OT
z^jNC{$_Fr`*c}Mz=F7k7x+pAZ>UBW8dNF}>dZ;stu_N=byVs**+7&RNSUR%<A|pKx
z@gZ*x0YYJZhJy%;X3snK9)091NwQ9toE@b5CU`yQA@5gr(>oq`XHpznWRe0He?Bg=
zLOdIqxa8i47IK$c{!dE@ngEO7rSVl$78Ooqa6;0x0<KxNm%}#-6-f<8DoBsiN%AYU
zStLu(si$uPR;Oc+fTtmW(fmIZZ4R9QtzwWj>CfG9ai;YSg*0{X1y((YZ$Kn)xT$fr
z0GS*N+kk)e73iArD;RS#ecMk;(%p=##yWY^9`WcwAMXxQN9<gWsYek1tliZ%R00Yo
z`0`nHEz`Ht%H8-(hXLr#=kj)@eGTtMp84>Q$9(7U>DJ@!bSe&1>$e?F+Pzfb$AfI+
zWWJPS&6qQGktHB0i-iU!)676o?#qdI_HBy&<$WV9#{iFnfnDf+Uio~?ZV3O`7Z-8V
zboFj%#cn+A7~OX{o6gC<<ZS-pqTaVQugcsPNmhR?(NNIY3@_JLC6?2plWiB#5hGP%
zUT^1w$M_w$EzNI_79r`DNO5}1yIPVs7r-x^eejChb9&OKtT46=-b9CjmECnPMo(hn
z(8@<$V#@DOYJ=_^+2F&f(-%vWN!S;Rs(}A6-JfI^4w3KyVJgO6^&SGCe(EZLR%{uU
z$eK1}r%9?P2irg&c5BN9iA4oy?ec~y$ye?b8l36A<lD3ra&<MxWX)8qvSXkFULg)C
z`SW>qlbE3DwkP{ULoE<eCmq{)|8*a|%nrQyr#`08Qz{ynrMSxmQEq;Y=;~wmFw<w8
zIX_v1^>R6G(!cA-+D+7EK-5RWMhkvH_7)D`Z;d#T7Lrj1L0XK;X14{ZI=b@iK6k1u
zyqT6D5E4M|NGM&pi@0jLa^Aj^;5c#E@O^dRwyFOY`4|UuQ+ROxzG_KV(t^>jH>7^*
znwR7ZFQL}KiTz-u(!o<S6-uc^%*UCDOFut**&{tlGI4Lt!%5})E3#jp&RrMg^USW4
z*IpOy1NNI^Io~Qm{y<9;P%KB)OZ(vqqzM2p%%87WLiDm?v>PtyB}@_`N&_VM@E0g?
zS+_4l{v~Z2X%{0g{d$WD*lBAN_Mw{A%fgWZ##WaEq>Q!5Ccs2aAo;Eb=Jksh6X++e
zkiQSCeqM|SR~z$K^732nTtd2*njxV<6~IH-EGBrVQh5wpM>#Zu7u4vbLiCH3cWZ#T
z!6(h4sqKC~7MhbaUN<_3Iw{)RDw%2Gjdmy$UzD;C&N!Gk{V8?3&~cxVo>p-hG9dPP
z40r!2cB9*GRKC!E@y5e96>_l(ng|}DrkQ5(jxoMrZ-KgMQ`pS*531TE4X@t?#e(w_
z;VWm5Bv&z&Ex({p-UlFW(7G}2o`^R!9wNC?OgCbhWs0rA(cO8h$w4!d1=&&AjKwBs
z4*I7%Q$s6Il{~?rT*B^-3%$9Cg%;z6&GaS~OaT++rBCYD^J=D0c;ZJLf@*KuU4NA;
zaZ(iyGG|{Knes9b8c9oMnR?3Q-<ifFE-!*mHxC|EyODQ3`fT<Kh_Oy>)OJlP8Ttc0
zU*xZJy&E*ra^0~F@aVF|2)B$HO%^R&><WS!qxSXLPgUW4523BDKVFL)zH%|aI_?we
z5c7N}<!0)HTAJX=7ryyjGky!OZ#Nc!!_{pL$*OJn+EDdS0@?#HQAR`F)<braGX0}_
z^@GDuQdG_36(X&|J5n@1Uw1q-Sz&!<&dKtBbqjq?gxL_G$%7odnn-+eaZ{SB&+PY<
z?%r560tZjZAzBTeM&==5R1-$wJ?6p{MBlbnhe5NTqlzOu{UaRjTf@*Pm5t>Mo3D0r
zwN`m4w)=n_?|e`ZXZc&zllZgTI5#I7bG^-dA#2iUBC=8KMyu^|n&V)Ln^m*03y{@R
zDbPCJX{B+0;6Hf>qpZ8mb$*sS4DxH3_mm9GPWQ=!Bkj5$K`%Q1t)s&@Y>i)AB+E6(
zj>`K|Zm0Da^l%fVwHO;RG#L*6ar{PZd`a%rf)HC%;yp-%B&w?3wm{M#`xBn5?Lap8
z<B76XqCi>9<tE=f1n>}LKXVmLmb-fc6Dnd(S+sVZr@*hVneTY<MYac<`>$zvXiJzv
zQu47oBSTD6Go)&jyPs2)(};evj^8pKl`jNwI4{1fZAQ$!o^V=TbnM{6sX{~{&W7uc
zDc#&vPIwpou;r7PMPSX{*{60`-*mN)$e=5^5tYmh{O31TMZ+8<Cwlm%i(ufjWi<|O
zo(&j0O$+3h<L?*!apneGaApfxChO`}H;JJdS%ukog;_6b<GtO&);f&Udn})3H<VhF
z!I?dYs)EjBBQ>n+Odkq5Lm`sqvzmkvnv5>e(?l~-E^>EOW<A|5gq_w>6PlBJhQ;8K
z=)<5@X?dAKU=fj`T~pGo80BKCv)N6kG;c8RjlJaeO2teTYukFXTCnc#a+t?D&DUu@
zWA)C9OnCA<nba`%)rM@#^;MlYc0zR=5~*MTFAnl1wnqu}`1iNh)GjVtEI(>x)7rSr
z>YKGcD7#~r&XJPZZbMmi95X4aqR2iynA^i9L8Q7GrS@$i;um9S55z1n@}^Di`&?R@
zQeo<?rSvTA-qpz8I2Cyqq`HWHHz_HvNe$>a_F)*Ag)+t!6OD#*)gN<j4j%9_HWC-Q
z!nK+-ZGK<7A<~YEQEf@J<W4Y(%~*Cijqc|JxHm%xhph12+p<4fGcJS@Pkq~&9x9~H
z;c+Tb3tclWz{;9`z{{S@F^43L)n2lAm8-DcT&eHS5=Lm<<M3H#o<>>O;t1@%7zaQl
z?>T<uUjk{}>QJ!s@{;OmP4L4%y29};D}ILr|B%f5L(e1>iDU~W{vn?G{e;cr^0e~c
zs_5%VCi)=Iv<*&_#An6fN%{e+tfxAbYr;#)H+%k;e%t{G;YR6dbHmIMY#_4*Fi%h8
zI}^_3lA2LmkIb#qyf1<5`@>SJ+R@ibPxBM7M=XqgeggAG#C4P|6P!PqRC@RdNu_3k
z#fNPwnAl&#Ur~ss+DD?}K%zFD!#tXF3;#Wh`nRQP?U7XlMAIXQt7i?^kEC&UKMyDQ
zp>X)`FKnO$)j~LdAlG+V5Mv(}0vmbI80LRm8@LsZAtSlIf%Ywk9ca||FEFR@R~Ha<
z`X>&bHMocT@`l$Xzxd@Rkzel0v+?uQ_TQiAyTm34Qqi<9@9F;cg0Qm&|No!;$BO>d
zKmYq8&v*(E+>%4iAdLqz$@7MVYe)o>%C+*%QA1T(N3ZL@&l#N~+a<Oxi|<j9o)$mt
z5uWn%S`>6_`uOlXak5_fp`3cdjpvv21q2H%#_B*<TU`ma5bN6ug8|}eZi|H3p|TSA
zuNogRCWmZ*0cG;o<qMR=lOJD#iv##iaurhTaI~15MY8&+LoH-+D4uF|nq9W&mAcoT
zqLSJ-n1$(xC~<oJKxK8;Fl9}uMRAQbY9^a}c|G@g8sv`;@}J^714E>fpLZ@^@cZ6F
zbNMpvW4sf}^wm;tKV_D#yyS&)V*tjwUs7?R=7Fz$t{X;2>}HBeq<GRVp7G771hg;I
zzBorY00V{rh=cjk4E3d}Y&c|%AtWTh%<NKV(rSWo;$(>VD|~1gSb>43VL)o02Tphc
zNI3N;26Q6SKtlonZZYE>BE8!*$d}tacIMI)%)wu)9oR!pS<IVQrG$htF#IzB%hD6n
zGn;uA`4G*AJg4`h!R@swn>R;yZ_o4I!VwZ8`k8VIWQ_o^bFILl)BExGDv-HA{9~^y
zm=7QTGOiToHshg01jf3D5hd#2m>Gk8VFVexgSoYYU{U!Uy(cAA8p-whPiNt1HsX4#
zId0Hb`dU#n&i->;a+ESq!;N-eQaqU;7di8wDSRB_wbbL{^98M-$a@DfL`I9ZNCP-f
zj^`8N9LAJBwx-$*x~z<<lkfsuTZ*-A1>(7_2LjG?A)+6Hk7ze<)+1vi2T;1gaHFzf
z;73JtsmHTNy@kBmVHchQT5p?|JmNyDzgU)Uego;%(f59*zsoy?rm?6V2GVkR)`UwJ
z|D{=5F`;&8fbXTLuC1Fo@4>f*1VRn%XkT@01h_@(+QMLo8W849rx8v7FL>-22zMpj
z5b#b3oCnwnjeLl-K^JWQj2WUP%aO)Inbd4nf>F1rRS|cZDpNeUm4R^iCxoA-N|;CT
z3nWkuC`KD5KAih;%shV<Q6s;Gw(S+>%|Na=@ME(xe9$VY>J2#RJf8DrH})ptIE}rT
zoQJxyxx<{u?)EtTT&8JqNV{*JQi|q0@;5b&lP~__ZxSCSF~>FiAo^WBtMG!$ox0N+
z5r}#*6@=|TG}d1Bx<<ay_^Jv`(CvN1#MeN>F#<b=<w&A^u%-8m7T6&1GQA3sBq!8d
zVr#jiq^Xi)(bYVStr1iVY2OT4v7BJOh`y9;@X7PS6_tA;{lYuYb#A6V=+ZFXy8aFx
z{z6cY?5(^^A=goqONk?o;a}bfs%KVRBAQV;2&#^`;I+yBd5Bpx?0k(<zn>K8v(5aN
ziXui~O`c%|yo!g8bR=`?y0@j;pLq7iXW*JD47yJkU9}2T5|gn!Si<S<&=3?ZsVV?t
zqS+tXOegF{8A3h`<qazsnOUnq*vkACwVdSY+Y1fv`LCbRuX@x|)1Oj$fAM2YVt^Iz
zhQJ8wgPgEKK*4gk+X_V{UdhP~gEXh=os36k`<I;FIIc-<%>+8d$S-e%Une?Jm_>H%
z3=es8a>o0s-=O5{b}w+76?y%#SE!V~_}c!dGn08Lq&$gQ5;0A1e@<-t(m1n*^k^J?
z$<x30X#D;brP=@Ah|+X^h|&gOfVhCNpQhe`!*Pi-@O$Jcy&*8L*~4|dC;ROI9cjf~
zZAE89l!m+bPf^;~E!G_*L!}=|X|`Hz@k0vR-pIdF*zB$TQ<nDsnZnlE)K~yps@YU)
z4E4rMqC4%r#u-I5z77*L1-X!Rv1R>nHDLYOm&HZrgOWlQB)tbjY0egbwDRk*9<rRh
z)uv!1M3NP4q?z)S$-LrtKG!p-hEZp`340mc6A9QuxWA`;Za#_nWIT2#;UZHj*n$p(
zeaT{9e(Gcvj7JF47ZDNlDWqj6c<n#5XO-H>|Kw?XzK`*dm&q(Nbv8Cs!)lS}r=_;=
z{+whG`t_QBOd^ytwFM>DAT-bhgRb5lKmzx`-cMNw@k!fG20qeX`--h*T?@^j05Q6!
z+t-PB5Fl>g=<q4`o{&26*`$@c1S9X(;zNZmP=fQPc8z=vS8~%<{MH`4Wq6x1{}icx
z6l-D1?ojW9!4Koj;A3Av6SY)u$s`u^;fO2*#{)BuxOFnX8{Wusg+vfCK?b3n8*(+U
z<o2gx%_jTc1kH~-?RWQH^oztVa=^>Xf@{cDH(?9#-BDVJ`J71a#L36-bh|?<Jk0Y>
z%U)WWYto|?!qb8WTfPooT#iR<mLn1|`0~pw)@AsYKy9?Ym)gqv3Dl6P2dMKiT@Ef$
zRXI(Fws{Y9aNk;<Apwwvdwf%Gp0RSOH`u&vB`f%GM*e}K{l}S7VG?UjuuSsDPjsAf
z>CHup)2w)c2yMEqv|qG=F;_;f^4S!0pI(94gS)~R-gjc_=Xh9eDnSTc60lVddu|9r
zAJOj&`D0(Wg1;bIGM6JAr@OOx32wlD;nc9C{~f26WIkMe$Wx1kG{N#g8cR=rI^mN#
z%w_vKUiOu_G`=c(8GmnQ_@FtJ;p*5%#X;FXF~HUYrY%uXb{(EbYobonWWTV9vtRW8
z5~-%>|NkDTW|YmLE&6G%bx&6GJm`<AgsqiRWE;ad7}#252F_tFbmsb<jsR(<Q&iX+
zI@W9ny(|57qT-&K`>=8F^Lo3?Z9o0v%nO8sHiZPf$u!Q)+d~QEy}#AsOoPgl!9O8J
zb9qf)Hi8iI{yM9+l3wh2!Hte%YYo-5M`x4PtC534_c{DhF`B2ZfFW{y7|e)0l<LsC
zr(BaaZRev80)<dOUsss56%>2o2>FeiIo(EGYYo2z3<cDz%ZY`#DtFYVS!2|cEF4i5
z`3!5at8Zh1x^~-39iSc0FTyX*PJ0ZOY;H0!3T?kW{U^{Zo+(ugqO{~X${ig93Du)H
zaAm&)tB4(+DB@)G53BH)z<9EfL>LFB^>?PJ8uqG~O#(>pZ(!OUPwFF9`4Xr3P|a_1
zy==rc9^ThGGF!L5`M!mvtRgI|>`UJss2#;<zHQ8A)Z8;JG?BKB=a6sNf;ie6?Ikiw
z+iRK5k#_3_jvc#QykRq~*%L2^T)tgu%f@zQ>+D#GykdcG{fmR$n+)u!%%Q*6JcSD)
zu=5wSp6Fz5_b`l4jIytvys9Ofe_5DS-Wk<#9kj-!QOy`OTGmXnHc03QwrMUMwe^1{
zO7}6q?LMBUayS*)=i~N{-@0m-V4Gu1rT2_!-0>z-jdb4M`csldr<!!xdiyD}Vw4-v
zGL;V!{u+D!&Rq&tc}2T<9Q^}$b_pO=GN^>9kX&*tl+xg&t01-z^))u;yJ{*g-R;O%
zjwpIE1kBOQx)R{lPm8ykGH72d2{e_uYdl1fo6=dzX71tCOpSe$+9^;%#oCnY!^)i4
zXeT5KJGdl^(z#cs;k`^cP5+u9De$RiEN^6H{1<srE?*VTQLWVzGq);@p4nkH=@1AS
zT-$q5D7oWps*0!J73>xcc!NYD%bI?wk!%yOg3zGwT+LwmO`N@y2TsXnvhRW5tn6$r
zU*~fpwtlb$peGC#)Dwi&yVpwDaa}ORc{P#>7>2Lyx#GI(&`o|&PuQ2Nm;Na|b$}-&
zKXqVd_T2!dvW7VEBFD(Z1a2CNNf_U*gi~wyCEd9)2h+F9i={P!yV=#i{*5f;e9a|R
zu~1Le=w<S3k;*=V>~S3h$~H-mtabH$DZ|drjHCD>{{)6BlN1gIbtTvre{5?1w(W%?
zJ)H5ZHiPdy9D?bpcLG-$R`2kA3XB#-A|nHtv;zd3YWeH*<R<1M=H6&Y5Y}`#oIX`6
z*`J5yjHt?#)&%6Q5ZQIL)jq~yQqQ_~L(xa!rZ9u26V;iz^CZu=^MW4=319k8LfT_2
z*_(_pPU^Pf!YX2fG;Lm5U)55wmAK(;b(@;;S?f1u=PhZa;5LZc-5rp^Gp-v80vq)9
zd1<__QkIsG%0xqmh&9SWuiB-)YH3oi1;Y~R75$^zy75_ZZtHjHOh7euTP+K_DVyBL
zM7Lk%>;qF6*Q?E1UWzF7l&lLu?x4z*%|L^cfGuiH!ihngLdHvv@#v!yIW%1@Jxdtb
zt?zK-{4qOkJ@D=k=80#~H<^a>p$+{eVE<zse>tz5)AhAW)r_>WPi0}x>Yf=%N;;+Y
zM)=aHz^7WX7QjV|-YX9O2N&&Co?qKg|6`V&&ujLopNW~78}Yt{gc~i(DbAE#{gr_f
zn9i;h6Z`^sXZ}2cEBHMXc*TIFZKe*vII9p8ISwXJh+esP7!)ZyE2&TjVZG6`^6V$M
z)fnR`HgETo>FDqdYPlYBWDiwM01}VzQTH|c=yy|Z74}vl=oW2?qPmpJ3CSlLrp048
zk!McFlj@wP*&ZG*9ixtJ{Nru^js&T@Qef0KjfvZOd$tzjz}l#DO{Ca-RICBDNjieq
z0VlIdfm*W151P31Wk732Q6+4pEWHgc$@xgdtYUXMg;ivujeooi$(h=qhy*Z~=3=sx
zW5=a{jauR4qHmOij_|PE42#d2a&JdQ5n;5E=FQeBo)Da~MZFjj?x3S$Vm2Lt<(|`m
zkYd#&sQcROGqMIKG%ee1m8V5X+Ml5}jB1c0m?#TU{-hqLVP0gVY^++e(fz2%G^qfA
z4+1lW!iq$yINI$E@vv<>Igy0l>(`JI=~%`*9jOP@^P<EYAL`A(Jw6EcR@*vF+A$g#
z8{_c!S0RQo&)(}8B{=T>?Mp{vAuM4e5dn;$5uGbg{qf?k&HN=3ORvFr2%;&(4rOBM
zFL?h`Ce|x%z5x<-y%h_a?Ty97#uCWv;3-O!;#q)zEN1kG5DBCCs0jKBBRZ=3Wk}if
zLADLel7{RSxgde5XjRX7Y-)!bw~M+A8r~PGqp#%1QAy|2$`D*(JND@gTZs0n_#8d?
z9F-h`=d8imZYC_1d(MLB2|kPH$)c83hl1h7ja{d*mf1CnMSgIFG;cVD>@*I@4=<7*
zOqp_I8Z`6mePuE}f{PxCh`=V(L?|60`B@o=8VyW?8M?JThWbo=IkT-X&vIKoW!NkZ
z8E#0G(Lv~)X;{yDjKW9E@uGXTl(q<mde`Q9KYN3197(N8-i^b@?|X;H+z=&L7E`js
zt<{&=l*uDXcf&7NBc^k;u*o=j$0+P)1289^YKBjj#|oJaRx~FbO<ZJT5}dAdyWA|S
z%G=Z($9+Su?rb6evzqe_+A1-$@`BwmbW`h~BSy!RMcP%!q!eGVqj~k1khKT%8|dl)
zJw_Lat@&hdrUg+}Asq;rrAZ{Gal$0Gd|qPau4YRr(ziE^=5)zMYRrJ*uTCIX-r?4F
zZBE<i<jd^Ra(1lPPz(E<=0vL>lv_?C6ZU1&33iGI?9+iM^5dAP7R*^lz-dIVAi<e|
z@bN!ajDcGP`(Qu!>}qrA10MW$JgWoKayLe>h}PFh%uCxqxb(=dg;p~tr(3V<wh8HY
zb86$dm8J=KefnFx3gb#%nK8%rzaOWvx*D8lXAkjO-1yf^poiT*X`GQ~__;>y#N$M8
zw;R6o=6tLkh+vSKV<*x-ug&*&L!2fn8WTsNJ*2Y2bxVo&>yOsXKMuq{{wY+2nBPXc
zFY-VCvq5NkBzW`sHI{ZzST1~gf_F;t$(0c|tjBndFg6#N12)0_Vx+l5ds9b1G5A%;
z1q5>a4PN2rR^oNsPHvw)fXd0?gKjyGRSOW~w01{{P`e$#UCSHIUw%}6fw!Fpzl2V#
zm7p4ru(^A0CA8vIXMm%Twqu!JE1z&N)L&F3LhcsG+L%_l8$v087YG9JVM!rniGZgI
zAU3xEN2(NKhSerONqKR|ry1vxi14KhjEDB`Awls5s-sm>wAfD6LBt$JL4jYoDx2(G
zqU~x{j)<;F>kK2f6tNG{+jq3Azm6((ucqA7zQfUy3UQTtpmbc0)mv^YWu(G*bTQtX
zf$q&>22ACd&AQ31Q53E;GQ4pf<^vR!oGFG@gV_X;u^F4B1eMX`w(HE&*`V|bGG}#m
z6wFA2H#PBxw0WdXMKVji=L1MRnoKED)J_V|wh&YZYDrc$bJ?{M4g<q%=1);c{E+Yx
zp?_|7T(s(Y6z#t@YUzTXI^LlDaU)xQ!=@w9<OW1qiX6o(L31k~^*Gf-ILUTPN(@7P
z?bgY*FlLM@Ke^E|+{b3s#$g5;f|$zXtBlchENR&&cl<kQt2H4i1)P<O=XJ&Jc<jGc
zl&Nb!d#25etSCi#oIx-4?h_yaCYRo$Jxl_%2$rvhaLc$!Y3*bXkUf`8xfYC$TEi3B
z4!5Q5106y`j>Z5%0J|#d@;%E&mfFon*KwZ}y$lW(1m`-SH<ZIH(#}tG?{7#n`;~Xz
z7r%DGia%3$s`Cj&_4zw@MvMZL;65k`Z-*3@%Y2!b8~_1gfLrnxzOn;$n?(8Og{nV{
z<=JqZ{p}pRWnPe7HOaS2|29%|{>VE&<$6@0nt8TW5N{@Hf45b(BVD1-W9ROqy`?j3
z=CO7nD>IHA@EZ(3Rw_MPP-6?TjF)ndYnitjhtsJY^*iP25sW2oFgkWrBIcc!Z2t1=
z*d75S=HTD^!|-2I0-y4Y|0UNqc#2iOcYPm@Zgwmp0{(si55RsheH442F%JESpy`ir
zqI-9fs_TUBKIo*(_H$?T264>7t?PK)ok1_p(TCXxwp0p}O`boIgm9ZRI-m%zo~Wwb
zf&(e-veQ)z1aPPB>ma8o3f@&u4aRRKdQn-iP;Bz3<|c!LE#~c8cw18hT$dSBlcI+^
zv@NZz=JTb6Z`6(=%5FZj?E#A<o~RRef&J||ych>810<9AaL&!T(<Z+<8FdPYbGytH
z4FtErC$1GnS(HQp#Zju`Cv#KmP@Qmgp8d$Tv&GLHO~?7lp?$l(?3NedmQ~8$aOuXD
z;P~rFr2|yf^)%Ea^QeI!O$B=W%ui|WdU)z1RWb9k#XSN+8sbG2#vM+!9+G*qiU#eF
zCJgVsZxysu6OA$xgI!+v6>yYj)JQg-??pN=7PXaI2`hne5NCV1a~=E@(ej^7`omk}
z@?0D`UsWV*_NV^r-Y$*CENS!I=A$;}0Q<BA{)J?jvtw>6gq6+CNn|WKqFu)y1^Yj(
z6ehNp=fYE=W9EC!jYS)2sgwuotxYs*!<4!(Hx6(3Z6TTu(WPqB@~DSYWJ){sa^iNM
zb05A`gw)emXlYaB+BuVM6Zzno*G)iFE>GviL+0Eg`MG00y!IN@;dKPX)7)z!mA+3L
z{mv>5#RP<<$Q}lLewB*x9l868G=+_iNKmSkDr1>15=}D>)|kDh*L)h1a}=D*ajZHN
z6iGD=mD@Guv{71MQ$yW&yKI2Oy4?TZTK#yrv9WxnSb<lVA7d3%spMJ#{obz;J4Pa8
z`|QkCWb$~^rr--h^lW2gZu-p(!JTZQTTq0H<3&A4xQr@-xkK)RzAb9B0W6+$iDx--
zn{_fl04-Wk&S5jgIdhLwvJpL!zE@fV&Gz<sw}O-pKJTUXd=5ItOO;Ai$P)AKQ;6fM
zB{za$ex^ILne8?_wK~9<O^;ZkzP2KoQZ{o)V59R4otO%smX1XnpKV}6jPm@fWnMg;
z;1p?m4|%mF)&t|8qYq+zO24eZv{ZT=9hEe=sg>>EJ>w;qxvDdxu=RLSB|*0M+N~wk
z6jJ+)fnkmriymY30+Wk|_oD44E*=bZmdH5w)x!QJ!0gvmeihdTBe{69Y?di5lR2~e
zV<6VEZO!%!P2P$MY4&i;=q0L6B#o-;0=nz9nmD!7ZN08f96$Q#|MIZ2pMjYraqSJA
zqlZ7lD1!AZ#tv}esQZ+EhJfNpF4@2fdh-zF=7YxyD7P3biItNDaP7p9h}@x)Zxr{A
z6oIq%Dh~_=-h#=Y?U3bmb^&l~5(G)FNVuB4u0f1Xkik%?9jX;jZV-93G!HD{?Y1x$
zUE$A$;@JBH2}-U!o{0p<){@@(yrjyyF_%>pi`W4t4^eJkBmw1C3(7(LToY_No^r6)
zsUZQNlYn+0O)s8F@vlT+Zdo^vaIA?R`E8B%ldi&4C>9D)6jNoI&8Q}ckex^N9s`gg
zy#)9h`O8F725pbc+WF2R6Q0o*lsPM7LtbC)<WEt(IadguyoW7xgVOUNl~Ttra8k-7
z#2K66K%KWgS?nh++IexGK)5JDIfXQ4=&{;MZ(E5v8pRhkq}U1M!zt@Smma|Yz;jj?
z{0E9&giNu3d*vbqn2M2{0qVsT<P{K;I(}<&iZt}rqw%b)PM6(C$NGx}@S3v?^e10G
zX4!K#rXncpmQZ3}h!oKmK>5!it=twz_w*;=Ou21w6GdX4{V)Xqy=uLknFNb6KbPl)
zv1H;o1?L$Hlp%T7K0cNtlCa(<kiHMNHQE%G&P$DY$&WjEGEc(xW8~-=I%K2%3=dc8
zMyexy8m_xx1tgf1ElKN}0ZDG8K}!8ucm*ErOeAqY>DA|s45}w49&ZrmtH_aRH!JI)
zj+Rtq-~FJKY^}UvBl~3uj!j}Kd}+fh(90?$Dl^uchgI)XZjfPA{#EK^a0d5VG`K7x
z!&U8Z>&Bv@y3LQ-Wi-j;Kr*OJ_PTjf_Cr#iA}8PKS%yijTcu>pzZz*hg6O*LcQyFN
zhpnOe7cXBn=h%MAq<wEQurS=O0L<ahnGda3Hs4a+6taB)Hf}lh<Bl-PDeGo;><CGc
zk7~RFm*t?e{!>*n7GF#xl1}fideYSWEtUsFF0PDIDJk9+_1siOwa{Nms-%!&3<`4n
zBERufEfJz_{oTBpUUO{>S*#D*XfX%)lS)bsWeJlT^Spc}sw6471?TIs%v6X{ZbSU-
z{CUZ8{N0K2<jn0-_MBglxap|)uo+5cI9r$AmnrxGa0@o}z%p2=xr_hoVFgqj$!4#}
z_<1<QEwy5mu<O{Zw?Bu`yJgM=kix|2N$e(@iuPQbR)y0(x2NH^($dax>G7U(GfbiJ
z)ut}DZ%AR0$ghN7SeOnoc9i%=JQ0j_MI9Zkp>AO>_k@}rdzNFU2HwJZJbp=%1Jpx7
z-7N0+hV%ku`Q+_#0=`_2KFfhZXHTlztYjsx=a=VmQoEjl)mHSCnxXj$o3brOkK*vE
z)GMm}R~GF8_~aW1pxJ@gJM@#@#hOgwNtGDa3aJcc&z(Hg=}20$`NrA4YbA4|?qkJW
zrJXy_+`lG`Uwhn#N9t8Z&dN6c5avHG)6fkJsvk54f4r5lJ?f(=;~iv64BpiQH8<Y$
zr=6ryL2K|9SCFjjh(S1LiP*tY3ax<FJ5d@OnR(rYsQIj_Gfuc?R!t!L%TtB!Ue-ip
z{rJvp3^cN6yUwvTY{{OR`SL3uk+<*r$E=@93I9Ix2w@?`4SMwaur3b8EMT~8(B$!Z
z`F2i^CpznT#UG%Bi-J_vwfod=fHaE=t@A>;ygs~}as$+%+&J;GjkD2gHv{@DTKFT~
zi!A}!6vDO~%51Jfy#dH1`ZbeXQNgP+X_JkqK5WeD<$^R4kGFGuzcvfce)LL=OyytI
z8C!tIC-2G5mxZ}!w(6*McWi+Hu$0Fl0ybVI`L%g4y&Kvs<KQ#%;GcSjKpWXuvON|{
zJjVvW<%`c8zB=HB)m)m*jeqF|4#jk%Zp`F;S8lX06FN4!8V$Q&O<U<P>aJ*v;T3^l
zuJMKi5I6N;3>t|(N3cF}BX_`Oeg`<wKFlPepxnfGYm0y5_G3+rDG_wkAA^w9QMw0e
znaAOUTT02+ZafY8TQDJoP|W8uyKSYnxbchQR~laj=Uy*OwhJ|Pe39Ap#*>%tIKChc
zelvHcp{r_ljEIMF8HQFz`^xzZ&>r%xUV_ZgAhu<HHs7b^f)k{lG2L~Iwe(KuTP)g@
zSb`PMQJ;L;Rp6=4{|AuSbhM9%qpzgq2`E1!UE|Ko6GQ?VF&<e2MDw5fd`D+LxIJkW
zy_d`?$yn87^jbAMu#bT$&tghvG`oFP+8Zt`uM>ZGw+sZp-9tyw?(N7Jn<PeBKkjR=
zpoSMS5}s5+Zlv{!Hl4f73l0|*w$?MH17%|8gKvwQHm2T>SV#D4dEJ<-8+wy`A-dx6
zPTdCNGobf$KA4FlQL-(>7;x*Ix9l*#yqD3o^B(JsQ)XwBDavBCykX6()jZ>gWQDlQ
ztEq1Sb)iywvi_NJ3qQ)9A4TgwDkPyQ8mM2jeaXwd3zyhW9x1|d_O86~3OOZay*kVa
zEh|RkR~n2Ifp+2=Vrk}=9!QBAq4>*Ky0x6ka<9KTinE{MB6JN*72muC{-uxl_aCOe
z1&7Xf2_@VgM+a&|*TTv}hDe(fIItM!zT?%Vj1at9g0p4ZgCoEMu_V;#CJLghfyr2v
z1HrZGDy^pcZYOLk=X{W?{i=z-k3gahSl)c+8oTI+AUd>5F`z?>n+9V5AC7yWNmp7g
zYs1kh<h8&qdgSCr`qDd!s}WmphNz^jJtlM?d{_c$j-{!%Al)7Et6R=lv&!%*I~kLj
zH3kZXEZh$zz%1NeptH2jj(IZMaHj8<L7@alXdFcl-JckgvrmOSVgok~#Et`y{QfsL
zEkOn|s$OWOxon+$_Io+yQgchFvK>jPLW16guJ1*)E}q=;Cg!L=ec}3>M;BlH3fPt<
z1DCo`23N4=J7j7m^gSq{2jGFr;mHQIMhO@%(c^*NqwL`Jvtl!%N_)H^UT>NlUoQdD
z8H#*Jc^^Em_@Vds$3pu1p9WdbH0S7vc{}zm$HHGOj>}n$@bWEW^yV!1xmWnGc=>Of
zn9wr>xc1=;De3ox{OiB;_lkrRKVTlXR_Gh%W&iJiYs8G8fH`eFHT8diK9h#)#3cRJ
zJN<usom%h1sy5CN5wHElN#<W_KDcc6)xbPyu10KX@?(wO$55WV<^LIJ$V?uDaQ}7z
zsR!SjIO46}V!~)mz+y@8<0t?1H!`vzUxQ7=JLb=?;Zh^}1vdA`f{<hv;tw_>ojWKF
zv~r&h5_kan#US90djXPpHE|j<p8?FN6Id4QqsOpPeBT2ns_Y%tHxn=w*?@lQbg1`w
zSdq<95$VAloIy4b-eA5NQW}TRWQx9Xi-MECzG})*a8DtcT<p)u&3jgHp(r;hC4=@j
z$+BJniwPMn;Eim6Mq^t5Yf%i*Ul>CX5xFMqdis%{c$tz?{QLzVyO8i$CLyV%NT6A0
zt~Asok3l+BRFRo|2~sVMp|osXv3m5DctVAG4z8B6>#ZUycEXT+ZSe}Hy@7!ui`x$J
zNFH&pDoX9zN{VxN`FdmDZx(I0h>DQ;p=vsJY=zRWw)?^!{9kmFa|br;bVr_&{^GkS
zIKuT-84hcu{63R<MtL2aFkDtM$~N^^6FW49C!x1~NS)F9gg5K~%AfA0aE=B?pb}*0
zH-uQVGkda8rGfA-6M{W8B~9=2`5^pCkeZz)2~h>9*~|`rZU>-AIxyFps|j&|(m63o
zxGGhpDQ-;OAAO{<phZ+unZpSYQG09rLm*+>)`g(3K`77^|Nf3_5Fy0{uU9RrqQxxX
zjs3o?w=5lMp78y?JpEOiv_d5E{U+bF{>YMkcIC25KMcA`-)E@v?AHqJx8a<fRa&B5
zRLAQbfO-Jwqo2R2m_fSIVFT&{heZ$f6Rd~18IZG;0>5jllwMzf#EUe(2Q`f*#%buw
z5^)ur=LOo5Y<xCX()NsfOB&M+_-1{hxBJ(+YIaHsc4YQ`aSO8M090N$_lhIr`!;3d
z{T?w8p&Zvb^-)W-zB1sdOQQdvmHiAQ3vHnY)sHL6TQxf3R|{0PF^88<XlWz5_fCdF
zM!Q8p$2Kv1xIV4nmQMKF<)*QYf8N=@x72^^c?NuNc{rj#!+H2acg0T7H4gX+?oV0X
zgMmK66rY5Ix8u&~tKhZWeiDp59#m0zl3_DPG7S1^v_$*LnE_FXd$Qr|l25J;`GRBL
zE<3tXnTC2za!o&>)qLEeKv>XHnZ#bsdB&rjAYAIaG**f5PK<esv*f&2ZatC=D{yLI
z&}Ml4Fq7<jFsY8B4E;x+)@WN^_mu{MUH4+W=ztqvb<dgS2Ccr=s#Tg)c~BCLUS96F
zfkk2&#qe=5v0jh0(WZiPuCB91c%_q^<%IJHCaXYCgF>>!!||3Cy*rPkGAhREe0x-V
z$FYX{9zGODc~{m1hr?2%o@^Lb`Lx=Q;+)_8xO1-S<^&Is*U~l5LVrsXV}tEZ3iHEm
z7B+V$_ubavhmn?P4g73o(o-5!7GWA6vtISzR<(B9Pcii_%2-Q`%YVbyn_NU{a&oKO
zZ;!U|cg@;bwhq#$&-{<*zw%$`nNf>mnC^Ywi~hPP|Ah3|evH9@&qMiIKjCQOF-Nk#
zUmp3-(uKnI^B8-D@6U728z1Z^-`d2@HkU^BpY4K!b`;LhY&NjbIp8{gmLn3`XZis=
zFaUI}X(WwAB-adDt_v?2G9Dr^Q+LdvsbRbU5h%q_qz#m<)H9gDesxHN8uiz-?)~WD
zY6@|i2QSE-5yICa5No9$oIQYe%Oa$PB85e<d?k5R*O$KwXybR1XCGeS1%Kxs&PC99
zD+bGW2U0znRk~p#@cf5Hx>cvSEJ09I#5)nD8&OYig+ReQ{slD^W0;_GIj$$SX&<}u
zP>#OzA)yVtL``dulEY!##pv}B7?F-z(3!;Hf>iZFP&&Lr+GSxF9uWF=ohD{ax=$jN
zKUO@Ou{suI)FX#fy2F`3JF`2xTLwJ>r!xc4)>wd!uoY0lmR)_JTiZ7eKfX|nFK@}D
zo>!*>urn6B;4sw~e(%ZHTV4Z*n-0Ou9>oCZ!7983lwjJ=PIE$!Xu224Oa;6lyYeKG
z(<6u*6te@wu<RWOQ&ZZ&1vv~71hg|N)#q>YrW)T<f5?88=@9(}8|ntwBi_No)stSj
zQf~&-cI*Gc-g`zxnQd#^25d!Cw2ELxa#Tc+SOfvdIfDd2$sz(0il_);iR2sv0m(^{
zqyhv%k(?1!a*`-H7w>#*yZh|-?Cp8a@9+E3W9%_{R|EB|XRS5oyzlGQiASX&N-pfJ
zKqUzmuZi*rpPq(R_cc!8-o1rALMtS@d_Xi#^UrAg>u)LM@vt_l*ooNbEE^%%(BT^I
z`|#y2vyoY>gQiDHcd}cE5!xkngVak>&fuxy^Ct9EZi8oWTd)Ggh3;;%5G#pY&s5Q}
zkSEMLRxNMiCOnCB;poU9^@-!-L8!Frpn^-yUp<j#?jDxE9#Lkh5p3L0zZcgPcEQ;c
znKf5-!!b92u=Ct=m7a=<TgG=rGIQbAlq(L>e-MIM?yfMMQYrX`O5k~GrwSnxWA}^S
z!2Ve|k8t%2W_aqs7c~Y7q`OsnX0su|Jvz(_M%}BRzS*h(nH8ZpS7$_gm!N#BKX<iU
z%4HbDFXLd{aB-|YL|KCLsJi|7UE@>F4*zGv2y*?uvVVJ@)ZhND{ih;1O%?7Qs7!-)
zM-X<-7wBWjX3%i)KLHCj4dPM>JBUG7?lfhT8SOP{#V$N{xn9Uj(wCCGCBI7uFUhci
zliNYh!Q3k-M}8lf6?9OA28jfNh`sXTkW?^QG|A1mFU9L_S4sH@y1h{zNJFb<nSl%^
zCpXlX;kG(RL{vG5bo#n7HFN804+jx<)1XH;b`Vw67IaSYr)3&IqC>b8A<_;b(2$1$
ztgZMhahMG&-lgH*>ji4;6tu4V<!Yz2pA``^C&62Jd8QADD2-nXJ<koM;D`!NKf`@!
zCc3NLI@!q7bE4ZGyrs!S)}Du3xeWw{2LQ`FWaTs#B$)`k>J2^<nyPxyKnaHFB|8~A
zOV~K%@1B}wUe%h%ADxT7P3H_NkG9#MnH?P^4n6!I%eR(wEk!+wzuL;)6@B|dYr5p_
zr{FXYCb|!pU(ALsbvy8sRoAcQf*}4>)U(><4t@;$@~Z2s#*!H|*A$QpIjidl@DOtA
ziB#J`n=?Om%Q#G(ra1tuz)Rh6wmw2m>%93(-xy-2U7-Yf%L6jPinY>~uRGF%*<j_m
z2uBzftervII8!k6ypYqK{rE*ZroD==n#;zS;iixUn!omUj}I1Xy^J!X`Pd~;TiX-_
zXs8YI*%kc2{<EshR_H)U9a-sgwQwYecZZNr`r4*#z|8agYX%xL?bdemE$7MaZc3^d
zje-R#Y$5kDSOt)15A*qFi;yxrJz0a~YxD_U`eFCZs~r&&3VKvTtbP%>!(%&EuT~6L
zkl}j5eoq6{jF0%QTQ;2oUy3Yy>xn#raOS&}36AOR0+CbPHuuApuW0yiRdpFOv+g=5
zdC0w>>Jygxv8>}3?1m_xio0t!&8z$@mfhkGeE0>f{kL~zkymmDwj$R4re&PH^cdm$
zc|7cu#4ekSUUmZgom-sH2iY_Vud@^FA3Ro&eslpFST!7Jm!~PBies`|CRRU^q%@7H
z_f$AN^}0P#${45KRa$#zi-~>FM_hY?o^1|`vvyyO8FX5Fj6EmQ7pB29ny8iM^;tTN
zQ8^AvE0ZaIes3jJ_~t}4+qHlP+Pfs)^jv`#q+`XRi~{QVgVYdoA(#{u4*+*+XA+sT
z2Wcd0MINSOKN<Fp)_mZ#pwfq%e9JO4-9M>%WX(^EHQ156Pd$cwYc<&X4*rd}D~fj>
zV5sT#Dcd8r_2+k71JlMpcIhh|?x62~!JItIW)XCmg32Y91DJ{jipxFt_3?W*ELCPY
z@m3OmTz1#s!!6OhPCs64fX}nll98foe>2Mq+`assyxQ6%3BM&z1LCuGINwZr1u{^Y
zVTw{MzS3)4XSb&QLba+uJVPrzrb`@-(lt3sMqe2gcokpXb!NG1;${l;Vsvz<?DV=;
z<I>$j-{+s5ZPQOFeN}O#>5$*!+V+NCVTP*<0hO7knHj;zl@?M@8$sG?L`YoF)alN;
zzv#mr1j99z8Q>gk(z@_;TZDkq95oAy&Zg2wE(UM0o-*$cewyaDjx~#*{j10E@2--s
zbYVWyKwrQ7=f8aHOmhjcPqmi@ekE_>mA&Z}p9;OZbz~qgY&k7p$0pDFXYi(-&~2WE
zQ^RQkmo|{(ZCh^I$GF<VODFCQOWTE$@H{`|CiR(CIhN>Xl+k;ocz+bT6dPUZ6X|qr
zq607t?P44c16kKg3VlA!upn=Xl><%C>k}}Ivb7H7%w=0ir3};J{T%zPO-r_rW*CPN
z8JD4O?;<fA+mx<3*<1%b&ux3^VVM*Wu7Wt*+4`(ZRIj0*kNy1w&RxRoVO2W~VJNNG
z{X;)r{zRL2G4wDywjr}w?Sw3rx>Ja0;pt{DQ3pz1mM)Z4eKeNo<_|`8cSISFZjQZf
z89r$|e7|5`(=!?0D)Z)|;?f_P92#G!rJZtY3Vw`Gvo7T2tjmF-r9;**sxh*xkoE{+
zteqo{ZMMiXp80vt+2A%lPT-E!E-!L4v6&`3-i?diY{~5^=Pjfp@^FqvdYah5h2f|N
zS@3kyV*d9}M^w=y1ti04vVjbSidU-R=Re&E5Gs^CofcS)>O$aELW*&3$SAHGut<6(
zIfGEj@r|pIKv+mGH?hr_Rc?yf3oX%vLv@qj=FLd|i6mQ~+iH+2qCN={peBOXj0%xQ
zl;{g3&7Be`$Tr$uEaGM2BKULZ8bVSt;2MLqHe5O#KBIWxg5NIkX5#g#rZY0}pI3g8
z6pl0JH6kPq=_8t|EZzRm)0ZbICo(a~pOOZlM>X`GV{bWr??oltBH10t4r1>&g{{Pj
zVjDUmdIsSNwL^HtU#CGM5+=PA8kX6jveampULm&|QNr2ldSnyj$uetJsb*!DNCsc<
zaB1+saUDsx8klPY$-Gf}W=32h_SwFf$~&@0_dJ)GQieQnchLd_!Lx@&EWC!NwtCaE
zN=17jKl#es5l><73O`2giEgL6PfIF?JV}<05)g}*({<eB)^XZA`;!+lm68YQ5~?q4
zgq+rPQRhhK9+<2vRMh@WRrE+i=s@H5l->A^fA-eXkC9s~<ZZQq7mGf760to|Ja5;7
zqBY&MV!k;fgmc)dATZo~GegX{`AGn$_2aFN`_i{5-nIgdtnV;(9nQOyB~C?u%#xHT
zwaQEE$|t;fn?02Cu6!iSI(M*WDt@#k@3KUAahBWE`rGKp+Nk0Lym9lbhytbR7NUS0
zV3lJVbxOocjd*!n*6#J@-T7FUPjRu&9+K|0r3@h0aiP#9S)c5VWDa)|`n3R{P+nr$
z8rA8-tcPFfwZ;j4+gKd^c1rSmP=g{r$BmceQ;EK@ID9N|dhqkmFwJRv^i@MoMChr9
zh^cU|me#6?^jq*MeCV_i;GIuiKKDV?yGP%}Y8`Y%<q$?c?k2$_Ta@(SISJcsqn~a!
z(*rc710Auromk?uoldx2KcGnIMY1HKhFTF$RI_&G#14=%$GBRH6eAtQS~XiOkH<Xv
zh>S+vb3bWINAWZm_7z!bntGa;Qe!M<y}ivG7;J6(g_q`tySW&DYQGkAfyjKlKlQ-b
zTmT~QFDpeA;ED6TAzHPjf;&1<GH<%dWfxdThNls8122id+zHGN5mOow)8Uer-u?~~
zQO&vWjB!piRSO;bL8CM;q^Vsz{8@B4TvDc)e2jC1x{@#Gd|3-P6iZC_3LW;S>0si-
z3M(BF1{jT7Dx`TZ?8uLQr^5>ure1vTZom2ZHFhq=%W?&r`|}4C?W<H4M%u!$sSg?B
zUGoQ*f!R-<Pr{G%0bV_2Ci?HK#((N^rbkfa9r)V^t>j0<LcI~PVGPIhx{(k`_7wE@
zh1rLKesahjjMgbpxveN?lMv{nniw(dr-_1sU<7q;Ot$VBebaUTrM2`AE!AtJ;`*^P
z`_%(33U)m2eJb+mFVz<wMRX2SUr{{eT9^A$DSCQIh?{9&e}UOrK1pblYDUQnh=P47
zSu+$-&nSXVK>rPI#DmknK7kUm$dUQXTpjP_aRU>P`BG)BUfqCWbGo?p1#*zG(R~gP
zQ5?Cj>@;5ZmcFY3JW3XjyQG(04a$RoW<c*oeY}pCn6@YB`?bxJ!(zSl>?xefB^c^H
zpJcyvLbmxZuD&VB4Brqtc~+D7l`XM%hLu8B@=z{{?eZ+1@|g@T1|)}7CiGk7&}IJi
zX#c@jad()Q$8fd=w_vU~Q`@&xKmV3CBd975l&02wUh6*oD7{Hp)FsmMylrk9!#?^$
z>G@&Kls*opb`GJ=|EA~ps{PcP0Ed6}fJ^()r$fm*dTUwQ;SZo<#$;o~8>NS!_R9$g
ztc12$Cdc_SEcMdxA##&9DQa@O;@;tE^-{}5TB;DjnciUXl(*|i(WwJ(P7B!DSB8Rg
zEBceV9*cqDxw;XUw#$r<^S8Uo9(@v<iC>aEK6ti96`>MuX0UI>)pQfU8Rgs58AH09
zaWXYJcS4?1MN79YoIT7!vXY;qf(MaynSdZT%r2HprkfzCFc4U2^s4IhYa4Zeqpjv~
zeKhD62J%cPeLgmMx(ik%!%uIvzJKkn2`4B0>s;cBkS5u2=CS)KQiC#iT#Ao_aAs~=
z1{J<5`kn4ZBkfb*Jz7pkD*A|I8F!~fD?pn|%>y`OYT>1`>9PVfaoUI!ZK$c7-J*30
zIIws14|O&n=cQW4dgL0n_Z4<-q#yM^sJSi$`aZ%x{AaWYaxU^6gwYg)W$|TGyQ46(
zs!Ra2A-mKMQ~me0J*u6@mrKTl8)S#_9)9o08bPT=@#G^Vi82_UXqRQe)fqUGPUm@H
zEtc8X#%%P1Jhx0PrqYm%M)$>5ojCCLUxZz7VCHj1ypx^NDru@0Ju7+3y23ukM0GRc
zqo|}LJN}0#osG8qfzltZYW0$vSetJx&rW`95M6sZZF^B&NK<9-@>l!23Gh>|g`*0j
zA6g7Qpk01Q0gx#dtQ8!gx^65C9>8$dkBw9eH6OrOlI=wG7S&_;C#?EwJhq(KzB&xp
zj(=tINyhT698kwaTSPx8i?`HQn~6=6g(;U-v`=w8CQ!m~UW@mkupnD(iNSl=8RFwr
zC$x)(%vF0`Quv<^EeOBAli<4RB>jf@lwrqNl>vnAF*r^kim}pVUA=N5OhfdC=BIV+
z6s6>E55Q9B3x>%CPX2m3Rz*aNLc#4ymQ#Pn-&;%$U6>~KV~D)r)@ea@_cMr`OA;Fv
zq$ERg%&jECNd`OICkOGm-(Q=BscS5@-VJXr^0Gx#=nDIzV2@h*zVL#x<wqQ$t1_-r
zd0>TZ#T**x55ZY}mwO>sY_CRnq++O#am@RU$sr5Vj~5(Q-Xe}P?cFZBX&<~Cq$YYj
zTapN?RkM;|q8m=s`d2mA55|8{bJe~=wLFu@q`%OTI+C%NAx)HGnhOAZ_OvdiLaQ&_
z^wGY(F%RLG3vLtR)^!^8#x1MN0ck8NZG8ZtahuxTST~NJorP(_3i$L=!%nf1c+6FN
z&^G&`bBoH{j{<8mf@9n&O2<-C*}ZHO>bskez8h%(CzQGIh`hO<j+wf!$LOlpwTu~P
z|Ew(Ufie78AGvTiCu8N#lI%i)4MaIE#+~V&Y$BT|{-8RN)TGvImw385el}D}ewp7K
zV_dv6KCI|+-vkaewR5OW3j-x#nB>j%CM{7pPf-g;j9piB3w*~5D{#cazSJ~B(W3l3
zyp=2(C8teKwvc?=ULEM<#lUUA{9Di)?M;e&fy;ZIBv+mx(YaYzNS^PfEHXe;`0lN3
zF5({jK6E36N>h*kZZ+0xh#!%u@O`rTIDx2@n*IFf9K%b^5w|!s_TO;0VFo6HM3)mk
z@A@_U1B&vJKhW~evv&kJUdM*{l4#Ox2n;<s*dgjSeXv-lU-qY5>!@_uUve!&vQ-ZK
zgaz;eCQl(*jrz@GWokx3#0OT|4eX6mAaM$7)6JY3ivoR1D@niS8!KpYR(?ZZdS@Y^
zY9fiCV?^WqEcBJa$O-Rn7$|KLp)tqK4{|NnNTqSoix}QTEbhjbyp-E94`&VT#_+(@
zBzzt{_Cm)pDmDwQBSQCbi4Ii%0e~_O0F-$I$B|w@JhX0Rs@PMiRiYP`(wv}Ua?@q5
zNE0k++e(-@En<OM;rqEQ#vvFJ;|UK?@uaVfgHp>=@--sV3h>5UMf!~a>0ZylhPpmN
zk;e;Px@zmBz@LmV)z2S9RiF}HR+|wq62#fi4XNzCcVcpdQOn@$O=cF_JROYmK3<5c
z95f}qg^d0eUS8Fo0FOt&*mpSxb9c)e`er}d*=NO?4Vy@!=HwiBVoz28CqLaL^kdyE
zwz0)O&fOd8jktFj8pQDjQ4dWRi1X=8$UC9McD+soGfw>j+J6CXXBqd##Jes@PmGe6
zUx^q8(<|Qe!{72SzlIap`}-WA1?`gi>};n~#u}e;8pu=iE%ihdaXE$TQX6L#`}aa~
zi>TRMrCmJu*u<ch$Cb;qSNQSiYd)q{kA_xCmSQ2e$``4yJ^8kH=Vaz5_}kN6s>G{q
zr2X{!*(^d9upqD)&F#cJg?G7uMd1+5gz2IkeuXb+;G70-{oEli2s&FJpw?}^<kAd5
zKIX9PD?QsR_?eo8M=)693!|KuF4R$FnRL@v{r2^@qufv=G>31+iQ<>>hm>fHLz1c;
zGB!PyBGDa)UwU~wY5^A5dLK&I+jLBu(RIolo-r+UUvQdOq7cscAElPKNWul;_-hrh
z>HCRKxU5c(H%cFQ3p_gs)M7e5F6`=UB*4)#sjYQ`;|L*?G1q)|dp&?o_a5}%>8y3(
zdF4{l(czA-BGS(2={r-FIZ3=tv3#ll#`NQ0iCC+=f{UQzYCr7JojeI<)e!ovhODvl
z)VL+Kx<VLrDkFleGzk!F;k4*bvNGfS)It7uj){Ch&OZfPQeP=1o;66T@k1zzu)|ql
zzjV=!hAUwLgpE;jU3~h{JB+(qtIMsQK>P+j?aLiUyZCT{T)L6q<2oey>0kwEM|4t$
zlFPy`L{9?OM?2a~Ly#X@`}w>qV-KZJQs!)<a&^)PES##P4tIMrn~Hi~;O{hOWXor>
z;_x-R4XRUMHg4q-2(EEmfJ)lkO6n3*40(f=VLXz&F<H8#R$Rl~Ymz}Eiwk_uA5pKl
z9oJK~CeoK+6Dq|#8($q~5s&shN&sP3>#divLl+d6j+|u=v0PveGsa&gX6^0)Pv1yi
z5T1|4Kr~M`9a(A6L*%q#llY<8GQfVy*)^-^qNob*mj}7v{-qWugT}V%=91Yt6$Jg8
zHh6UDO@D=p<DPzpHHr7lUhC&_0gs0_AFxg8;o<3|Zc%K-J;VS%NX(8eU>x6Sr_t~`
zyTBunLr2(B`J3zx5dS5;_(S|1evWth-Ga-#0f2`MIWSl(^-kD-idw&BN>%cbcw9vT
zVKr-)U`%U{n<l0rk}XHhpOUTRu5plT6_66{w+gO5WccGB$=1l}|4Xt}5WuM@29m9#
zwp++JDAs%3mr_wc(}f|+j^ZfS<v~1#56zN1POlgFrcn@7v0FaJq>DS3)Z%dw9ZMsW
z3{18RIu8)cim8#P;GiQQ$AEChpwBf~iaQgFm9b@TelT(QQ`AS%)@mZV7IBpvP^9Bp
zHEK@yC1E+_vV=K1#a%0{8zQ~|q2&!HseNVSL?|)W&gYnC&4Rltoh_KLO&fsYyZhhm
ziJ0jZy$0H<DFjF|N5y3p*onVP7si<@x(nEB|AHS`E%@S6q~1ekZ({FXOhe#mV_YAi
zq%Ci7e++>RYXNmD1{#TP&RnUG2L+)^R5}XQo_15whaMtz_xq72y945Tl8IGz%do(z
zRGs-)`=G@u(7ZcFQpD7dzrt$tW5BxI7t9~d4$*@QJ9dXJJPhPiHte@6vQZ`uk6!3U
zBi4t>WSsI#Vv3g&Uc`Y2H737If6lJP%T8fC)Z3(G+6D6-qsos~t1G~NV<3_rkJ99;
zGO3A+wq9WOHqM|HAYNe0e!1apJs`fj5h>LarP;MXo4&mDm9rZsQy_Q7gt4Eq*)pe7
z%WSSYR0a<2%4-X+)2FWo6#fx9vOY5U%JD_N>a|=qLa}E{Q2_XuI~Pcu2ffJ#SnFm<
zlPi5!2d*zXPWT=xx!WVHGZ{r1x+T<_<?|b&7089?5lIuB{Pf?G5;{b(MX&xr4Cy6?
z-#CD>h)Pv=>{i#r&d?iiDf%~F!acRj7r#Fd)_Lx>VLpNq)B2%#Tw@Z*@M42g2Y=hA
z<UsVWkrL=zzf@MoL!svVga5%p0Ij5*JwTD)?=p2PPU^B;UO`}ndZ#eI3g_b>TFEkm
zH9#4<E`mK3#xp32+3D7GnTvos#`X*I{l9UCT)G5((a2wFt^G&3&h7WU0~MLKy1H2f
z+#xnAnZ8;Um^4pk8*u&9sU)0IN}Gd4oLrWz)Cdue(t36(yF?NYRwLH%(PgVmjFkTA
zg@B0k1v~Fqy|Rbtr03ib?rC6q%>A<Tg>cbWN_0bUrB%Z3X6}zf#E|Gc>K7=tUj!~L
z(E~^IJ_GdgXPOLIU_HB6eCUQX$g*=_F6Gx|EBEW+K#VCWJPt4g^MGnvm-w!i<5)%j
z^w2n@Ah$WasQYt@|6k_&|Mh>0kRz9}Lr?zZ4>htpmJVSkh2T2lsH*?2*b=cnrG<CO
zgLKWTkU(M1)#>g=d&s}yjQQ<)+DFuY{B_5iFh_sa>-ev~*gqYRG*)<A)6co){@>+R
zfB%sG|HuCQRs2T52!o{m{v|(lzVwe~YdV31>rWcs@xXU5>K=2c*jZ}ic{i|^+X$+b
zJ?+{vzj?#iW}tAqJj1zf`&Z@Y=k-WvY7asrwR^H)(yfi{j#V}n00a3bk^|XWYG>LD
zF00rS)3@ayEV*=&OD9VAz`Lj1xuR8k+!qhP1b)?MpyVbfOl-Dpg{yW%gF`Rfaw@S%
z^h05X-ALv3{IXpZXm-504Dzm_(i_8_SWNHtMqF4V*!>n#mqrItyRg^NdcP(n#-Wdd
zT4e7k$BTx5nW3!u$>DOy4r>n({c&#{hOT83%`v@w^->R+PAsKq%48H?c<)2%h~UB^
zbZv1+vV*MyQpC=~w0(X3dN>>=jzB4C)^^t}RsO!HrBPyNI)^?D$65@>9f=I;>@Nu2
zAFYVt2rFib!$5l6{n}@zm><X2Huisc3knE{fTRw)M0hs|FQe&Fa^)idFPMW|&{Sy%
zwv!HSeabN}rt{}5HvBXr);cKu<0E<eR0J-zVyEf1eu>J{Y!5(t#t|~m>{DZHWhz!5
z5_UG{Vf1Yy8_u@q7I^zMWL+Te7n{qs-e!fke@`m$vPS~U=K70R)0<_!*z_4kV1FjI
zK(D=4#?1iY?~n+B7@)-=z%glVt`r{@!Ha&l<8QmzA45!f3uvDCx^YM@oT{Q=x9>NV
zsK%jB@PGY=<bfjiG@U}TVd7(FuKfi5_{B*iPkPlZq1sKIzhqhk?rrr}#eSHD4%w1?
zm_YGNqksQ|n&ztJdux1Ie%NoI8(au<{cM9&yW*d{4iR#6;5+%a6uPsS;VXYjzYuA4
zFF-{c7b@1vWV}V#dQV`d1%Q%MvHo6^C<=%WuV(rR7(J`bqf5Uo)+=A`JINID4k8ll
z9H9b&9%Jqha)-5pg~=RaDVFo<D_=Wwu#24*jn<*q?_zxlTCECEwxmUi9Y4*de81Tk
zUn#>D-OKd!6z#89cvW@?s^5nIq@KP8OfgMRr1GnU4ds50dOXA~%t7R_x%r@&a3qxV
zc}7*7=G*b1Vbbb&J3nIuE(!q_fzFwnx5&?Q?CJa)^#9+chmWTb1~dOhZ~b4wZd;Vn
zXD#o~oi>PYvhi|p>^21P?G&{jnbChD#URXTsF6W^r&Ce5^amitR$fe7!`dIQTXbKU
z>23AhC0o19oUWL`4$>)B;X?AQgrxp9yDe2=wUjosjH|4l$@@v&NOol{khn`|kS+fi
zcey@eu*aX7Hha?7NQD`cRkdH3Lk&_O?vj2xw3eF6*C{59zjanD5^a>e*bzV57}mAE
zq5IRIroud$uq7L%nP3p*7SKxljpPn`Li0+DngZKFP6&}Q^y{tVSV@gd+B!+-iVNO4
zIiZ$Q`SDBE9|B$RAP2L#RjcZ8Hs&4oz}|x#)0(d*S|bW7Wyo))??u_kjRsl{p4cTF
zJ(**ixs^*c|NZnt<ou+J;l;xzoEcVbyjZen$|^WrZ2(_kP1#ZGwC|Qu(^t8C+QDVR
z6SZ?K#+Fl>Z+N=S<%IaxwcwfvX)RYzNUwfe9%+4A=`Cb=kqImpRgr2xg=E&zq-8zv
z!5?m7^EqGGVFoGl_iB4phAwTtkB_4~yqn%4sx{T;R585SubZ^EIIB5gjeqC$yTRlC
zrCJrqMj?*zuL$r#a_T2u3&Qk%?pPPs-hR0+zyn#m#qUoDl)!%WS4O5ujMZ0=Vh+Fk
zY~-6&=>{0Ad2o$B1Q}C_3*=OO1(nQK7|qWhlbK><(5UnVHi~7c)KOB_+Cd~iUD#cC
z?em>^Xhhn8al;8OHN68mGJ9w~sZaCA*83xY2L=e+Y96jv_DJGe49uZihP^3WPgg@q
zZkrXb4d)>R!-fSdekQ?R_oIrJ_EwsZ2f-Dg(wCTbrfe>UU%rLpq;$Zz;eL+C9w7&I
zPHi^i;<mk2M`-}k{12r+18oZ1CA9TGX`6ixMvK+lX`U*xCG%x7J0%|l-oO^S<qk-)
z7C@3g&t=7K2r}zze@3iaGk#8a{$9mu&wXIF+r#iE(Xu`9CIt}5WSc{VhL&4}9g$eI
zT_-a+uD$V_%FFIxts`t9{Ea&yT`g66yVmPKvnm0H#y2EcRDn9~I``R5P|LOSs!qkK
z>?XgEOkb9nbAy-e{^vJ_3Jy%tmlBKEf}UX_&HWUkxjum+Li0WVz8xXd;qfiUZ}x6|
z8y<8N?OQ#%-JdPu_GV3mm)Qu2YRpH7e1_JamxiEqvsLz+<;cr4Sm;!^NBXXdH&ZJV
zmsO~nH6cr|7+eNr2pqwur<>paaxr>;Nj^K87Fu`xEGC)O@B*R2E2{%Xq_^YNqLO7J
z${?Jh1%ceVQ|Z9(@^WB*)n>Ex&d%|1`>wOSmDo;so=rCc!}bU^?Gdr0nTq7yCE$y#
zO7>|=u}J(#bsL6W#!@%SSZPnhfMr9cfP%~0fhS^5kJ%uKsFq!Y(NeQ^8b?sYl|kN<
z%VII_)VK7>B7^&Hs{hbln~56Sop*G%R%a~Cffb@=?U9l_viYY}O(@H}f<l0iPN#QU
zan0-Xt_`q7u2&Y1RLK`gTF$e+#t%%gSMB6e&4Am$nF21(9)7FdYxY^J0xN``G7Ww>
zQ+M%Qgp96oM{7rD=xo~{h_qeane|7t5KDwj*^?w^(fmGmeYHq}b@7}c=I@NEe{)M!
zk|mm@CHXI9{%vrn1`$&yYGmu&oG4-k(OFm`ycGickcN>2Ox5jWyB0pi{D74pH{F1y
zPTilfGjWmjkviW3HbEd9AWl^d9bg_MA)Kv_A{-|TI%mTbedwk&SRZ@rkj%b9=#ESN
zsXYte{IA-xr_W{&ma%89b$e`^&2L-PJWqqP*dVtD0dgX!Y;YfffbC2>Xu2cRHB~-P
z5q=ylCH0jX3PY4<M3{D19O$TL&3jOUT;_-VF)f<InZn3Qv|$SVf<|Wt1exy;Db-PI
zPE*bHhf$)<HjHKp;#wn=vo2&bdfhXhi+NU8(A=v~VmiWW{T@7FDEsdnb*&ue&6^Y*
z?AHt<<5i<?0SCg0##c$PR1zI4j-2(0hz4BfnH~`HY0>*OOA|INj7Ws71?;CoKgZ(6
zY0ws1rFUvx0&MQYb0~a<pc_#@<JSog!EGtKyf(F2ClH@>rF!g)QXwL(T_=kRW(qMP
zEC3w4YeZ8eVVs0(!Z4ya8FK8<ha`29`bp$mXL?I|VyD}_YE(oI7DdUxTjQ4Thk)F}
z(@||51w7O$cSU)G+2v2pew4u!G5V$xn~Z*2pB!(&WgXp%F{~-`>@B^>?C2J2-=m%G
zCgIWqLb%<poo4|*G?P8K?$peo7ZZ3TbB2WS$_p;jqSR-}#MOz2<_$<9FLzw(h#tfx
z;d1ihn!^l7ZtT4vHnJm^9*>9psxtm8YX#`GI^&7I;{^|&fAGkF3J{7WcWXwv>=w~;
z>aD_icn^hzYpZ(4xCU1hx}!S>VH*U=T5U*A@tR7uJV%F1jkqm{b?wk?o7>Zw1AERq
zJ)zHEq8T2_&GAgjezrS`4QPoI6j=RIOVk}iydfws$I8&MhT)>*N?9XC)7Y?bb%$MG
zn3Zex0disYy1{nl?wzB9*691rLL<3@Oj|Es=W*V6&6C_wWjSr<7ruQiw=FOEKW9Gd
zKN2b-veyy!*C<@%Al-v{YM%|#=B<<20W~sEaul|99yTNHt~qRvQ>8<d{Htiq=M9L~
zzJKsNP3K%6K}W>7@Obn2SOsJClP&ni3pHWRmW}r^n7Gj3@qmX^fVsTmsIe{?YZFR<
z_#GWv$P{&4?V_T`mTDIW+UxOL#;fe1VNSG^lS(O{@8mhqbsN0UqB%mKOAFaI3Ft+s
zOck~!*;vZ{D4$QY*pM|593`q(YKkENiBgDAZv^%9?4;L3nOAnyci34JaeB??RKu0J
zFxof2l{>E6HDGkexl+3%@<x?$Sn_+E3!d)oexUI3ZLT#CO$y|4K<zg^>QWR}__j&a
zB+*Q|<^*w=#jPKj^ZKgC9n1)_@7UrbFW9Tyjgvp<?$PD8>^iwFKX+)t{Q6ww6&60m
zkub`F7S;-(Ld})$os9Htf*@(5S)Q6IPqpc+ydNhy`fYU*Mf|)oO*x5%n0i!tS<6yw
z-BR{P>KdJy<~da2>Qp;z@DYwUI-*{unG2zzD?4W`vxe8Hm&0S!#~>ZcuZcCogU56}
zHQ|l;JgTlPr1tSy#^&_ErY|v+q(~nE(NRV61*s|LZNHp8$+*}hEO9jeYu-IwEOR@E
zE>4m4e0iFzK-0mDg(~s1guSx|aXiKm1Ve8u+e$^qJH;yEVE4Vy?eY|zp0s0^*Ln2s
zG4cJz9`yJqbdNNt1J!<c-O?<PaL&7VfMY038ULi1c%8&IvY}Vf!7WW;=}joh${+;Z
z_>&wPE0P(=SW31P{=<iK6xB8+gL*Rcq#SPw$N8<&=gCXHr&#JOUlAQzG7DwqmQy!l
zgpBEx8d3zJ<|qAw@v*PBroy$(N$p5^_P`*YEE&dTVeDwgk>;AVi7b6P(L^5NgPbAa
z+Nd8=O%bfpN_U32D~j1j3?ehNbllyVGTR<s+UWdTpQ&3}dN!`z27#yDdkEM_6YOb{
z@iD<ga0HV?*xRx><fi{2fV4l?Rj`sx5o~H%@gA@@JL=@g%w0DRBdjN$R#r>y^%=N{
z8k9Lk|5!KSJ6tJN26o48Tp<h)x86E(i@h-1ET>KFg^zIe86pn~l)zlY8EeJuAaOV8
zPogstOZy_NN!xY8nB;1Nuv1NPOSP<v?ilkIEBCu&Nu+69D;h&El25NK-r=yLA?Wh5
zn^cjs64?-d``H^9=5WlCNqS$j##IO92nV9-yOeA4`}T?KlNP_B7LPyqA;jBqZI<(9
z47jz**1b-%XD-`%^r|(U4ZA!}_@H{>ZxzMAX-=dewzsfxX{%qgI6v*B>Om--V!H?6
z`Z*ys{J~mEDhQUX%aRH$^>eSsN6)a`c!|+uR*Z`#8@X!0vVP}{&<okG>LHJ$YmVV$
z{*=cl4TI=^)fHmN6QVO#<Iv-~S31vyUA!mfS>-y?xKG-M&)YudAyl40lnyL(Ci3PG
z2Auxj`C>lGN9J7!9nJHIPS@U8m9)>ID#%031pyhaVqE1>c}Nk+CKY7n2=DRVu&~NJ
zfei}&p=^9j@F0bbN5VLm!!xQ+c!!2hB1zDfa{EwkAY0vCh|L~<J9#bI>E$r~&&NpQ
zVX+4yR+4YbQ%p8-PBQOUa?Md0KBejF!E1d;!~9j5XoAWikEMrEtjEuBH?4CnSN6Qv
z%H7tH-6YUG>U7yaYEYBw?t4Lzf?h4%x|*MD!g}F3Cc4dO0xgV08yis9gnRw`dEY|K
zGGObdE{w_aAFiG4kY#UHr(>=9Q$8mjz=(6aesPKSydn}8rT|m0D$5LGzlI)LkI_1;
z&F{J>VboAMUOS(D=-1F}5>YI5`m950r{vNO{er_5@8$Lz8zGUhRQ)reKN#B$sHF|<
z$uH)EdQs^{&A~plpquj2apf!R9$;wDzIiTW$a-gEQG3nUgbANt%P4&@^I9W?-R1Vc
z)00al)m8OS`OKIC(M^hAu61zFwuwHcAcrWtWtN9*M+l;6U4aJm$~57iOiS+BfRmT<
z6K5aNVYvAS387AWRuou9!mOHJky1JL85q7OK^ck~5n~oX#*sM+mggCMtJH`dF_{ia
ztiGW+SB*-xh+}zy$7CS_Hs7h|>BVZ*213>A3-0UX;??nS?fTsy$(LFfbM;snVa!c}
z%j~CXj7M&e&udE?HmT=E)?4O^m&%~@vkO}I3HZQ0GJZzeZuKS?t}g0ogOuW$2khsK
zd-E@Cu&z4F7iB9Gc^FRE{^w8zB|*f>)~eNp{?*M;{;3Z$lOJ}bUkSN=Bk?KRq?$>x
zVGxd%en!1`Q|}1D0(InwzqlAGJoaIS^!0oeiL^`B4k&f9;5dM`u-8U%nX70lexTcE
zq-^0i?GDAh6NR^z)`?a!+tAd@I6XKcHKZ1^u0dxc&O_g>BoN75mlT+NBSwAG7ux+6
z`riIwv>U1ybc9i=7FJTR-C&Y41Fy`F83Rk*i`i-^+$yl<x891yHA0(5`C6h(E6z#s
zD7j<zhb<<_AB^Y5>J$wv=no*apq*{Rnjz%nyf9GVyJV&v;VF!r>C{hd8$q>EEb+kS
zH{NX8Vb?;A{4_poK~Hf#ar~sj8j{LxsY>Iz$izjCe8ZZ^H~bi+Fm!Zl|9rI$cK@rd
z7Ia6_pOXSpk17fn)_BEQ$*yQpdRD>i+<jU_Es+|>N~VKoaFuQh>yD=@=(kwN9K~hi
zM;Y#}=-z=tNmoM^RHy|QUufNAq$KKR1(W^3_x*<V{T|+cb39kRJNRLjp3(;xA_Htc
z<8DTrJ`(MBwlPBl*=zFPwI4Z!@$xRKQC?Gw<>k<UA_>9F(CWOoeWOG@@6Mm!stBFQ
zG|*H&)zrfYA^k)(tt6AXN0Ql7xS7QJU2_=1sAeaC&YZo_I|aCE-am3t!Vxim`=7p9
zg4+9Dk>!YA)0i^`g-tAJ-DL`J#<P{Ai%yE?9QEy65WR4OKoAoJt?kc!{P>;cL+pXS
z@u|IXMuKAE@?J6i+Qt*1FSUU<qg<V;T}6-L9wWJ(o<ttdTf{2w22l3+*bw9CEVRlV
zn(u@QM_d)Bo=0$8?^$GysHnnH{ej@?D8S@mrB5BO1%-V%HrGdb?r`-W6eb}L%U|P-
z!1E`<kr;6JqSLzo@7X4Ll&SzK`5{ax`r-AryB^b^DTv&;5f(QUxAtfhJFWola>4aS
z_4TJw-RE}^)6LdaqXFR>LQPwKdQL+aF@D|jMF@VCSCZ$+IWLAqYbGqh^8ks2{X@v)
zdTN0RI2xaI-3cshbOY>+1u8D`ZeOVXSbE3M-VHrU9a<EwkBetCy~B=1-;rYm-!YR!
zRy_ze#q$>)w%uq4UBVe2+8I2gfa{0G*wac%=p7{#?n#}@-Mtp?#PLvl+%Ea?!Nbm^
zZw)Vl$f2Tg3KYA;Uyc~MlM{;D)S6<YG72JhuV4a@1>Uf$r+VCy>qR<|$N0_%lGO+Q
zd5!OT+<@Gk_`<t>-MrH{51?2M!A9g{0p37gsH6fJbXCvbuw*h<Lp6`wB-WI8aXN@T
zL*)#Pk`5E-NKJe;lrpZFOO4tAX&>y#nViv+eHTX8aCCL*y%Xr|cH>3;u7JhAH+iy)
z6x-Q(c#=UrS{(Z0(oSr1rx8lm!l%C}U2`|el?0B~G2I&1yP`Ukf_y@wq4zNHDLW4J
zm<-LDI=NFyc;R1^E=GjXwL<+%*X8}7tdYG6ziz_#H>4|zfjTO#0@1I(QW<RrXP_s(
z{v|-L<B}(Cn^38lPJILg&eVJJ0h^B(8WiaJ67u+=CndNn=R%Kf$6J7Nxe!$2P7B)U
zK6e<MuMjkwoYY`*TYIu0p<gAc&-<Y));?pk2s$PVwn~Q0vfd}Uzg||`i0qb%djYTB
zs+EsiXqx6gZyCf2-prLgaAmPe{{dj$Bq0XK_-qvHdRGbj6lE>l2{Ks$PG3+v$3cK(
zsh=U2LsnY{w*!}+xJ=WPEU&tAR)xw6LG~)&>W&b)KFs7+%@30I%LZ>{V!Y4xxJNuR
z5Y?TF0LEOx;JJ_nGX*y7e7dbKg9Hl1B`E#PztG1#pzEu}x3VJkTu(S50}W!1l4icC
zG71jux<6sSVJB*DB4p6yP8k<5q{vNYbHw<r&%k(TLz*szoK=_YqyLcMTDv5bpJTNP
zMb8GjCx^{9ky}9_#3CB^smw1#TWBR!x53T8+nM(_R0WZ7WG^@`Yw>Gv*+smcSPm|}
zXs9^eyJ&{&yJ_SZ_@@Lc(SxuYpHAAgwUJHQjWU;ZMF(RTf4a#4C*e9Jf@`M>bO8o&
zBD4>#>DbhfyGilmHdjvRQk8`WwXv>b5RoEt3@+;d47(Xg&08d~%mK&%yJk(190}hG
zlZ0M?OfErRiIMsdrr?5omvTF}?<paG*-)q}ZHb;qS_UP;RG7GRny`{}FZ?pK;zuYr
z*=2{)_*e1b6o4wRc>SgVrGDrf*@W6|&?C8BY1F2dJHmDo!c2n$cEbA%<W=&41o=xM
zc=1>*pl`tx=aA)fVT7t!b249AKk?c!g4R>pMsAL#C-L>4%HtyUKn}lZ#ET<a<$+En
zRgc<Fa97vzrUJkH^w>wy(F9@d>)Fid)<<7v1U`ei#o};yccs(w5KNXiTNkPKm>ZZE
z^D{7Ta^j^-;e_@lJHd}(ANIpv%>gHzd($}fDzmx~2Kax#_LA|t71;@GFk2S(7v@6j
z=BPobrc}2ZUk@D43oSw7EY~lR*DXSiR$5fqL7#pt+TxAaB|6WJ{=mqEDAPFUMKVp|
zUAo(7Sx|X0aE1vmsu`ruU<M*=dwka3!2%YzPWq+L9F6e8(r2!++szeE`w%(wldfcm
zJ4gFQhZwfTLXgn<bjEd|d!OGp^)AOzvTSsg%k7fXyk~_Bmbr>_3DGm&hc$(vCdPb(
zwQVJRk|{5y#Or!Wl4#u^h~V@eW?n<}GSK^__oPQoxuJvmLtd$DqISn+pbulaE-=?O
zk~GMGGPgKkmg~n@p&7=0V#zFWhUezanU|Eb8>}BsU|$|vzOl@7r3Df3?4HIT>M{rE
zAF1o10~|Is&Ok@J>*$uNym}-3>!8zKy7%7iR`~y=n*gaJu^e4(NF?13bC-b>bhpD=
zBZ~R4S5$&8QDbPSWlkPO+_MC+GNDS@F2);vc+GVEun@;0zze1H2T`7LJJ~9SI===}
z_J|NBQPI<Gvl`g$<bYwwA&^{wgUUo`I`_Sgjb44G{2pPKmY6bz$9ldZcMcLP2(<WF
zDVe28GOezMWQU2RgM*ilJ;fkGv_VFfdVl<Mmz~?z$|ALSNZfHO(qK1bPC?~Aoat%y
zrMl}o9r)l*SDJEyQ-7YW09{HT&xh#KUfUHe>k38e3jv1yQ<dV$s6MmVl~fM(BL;oR
zwoCfqmKYVNM}J|xL|5@d%HFS$a7yS;$6G7z(4Fm!^y16FgSYLguS4*3h?0c<>q3gI
z`vS`o$bi9V|Lc2QtKfnQxwwiQ_m@m7iNzMdGoj>)gQNVm*_D|R=(=xUD<B|nql~-n
zE}F`-5cm@A{?EO8Ze1C^qv1_|SIr2im+zKMQ~=TJvhO!&U7VUL>{JNac)CihjHFW9
z1+$ncgch`nUSKhGLuj$)+upo%yXLu(LT=9ZzOGFKEq0MTFSbECGK(+$M@e*&Q6IR_
z*^f0c<}0Xg42Fn8%6nZgG^Vo$``6#QsCG;5%iN{C{{`x{n7-GP%obJ0jEj&M5IFa8
zOpW6Q(toPqM1%`yh)31k!T)YvOm|+r(#uCW9c7pJb$8witsph}nlMduqBaE6%p$xq
zz9oOkLt+y4`-VI*&z&}&3#&0qbhz606`(48Iy(hvCJN(?cWkA`2|3Dws}&MYkb?Su
zJnFR+)T4^`{XO<O{_GJp(q`@P;QD9G0Y3C0KJ?j*8D`ZK-esn-+4qHzZ`r_t=H^=#
z8l62@&PgrGVqn}n8pug%-hyG{L#y*`*Uzni580He?;FVC$K`g$8)FV*{#9<qb}a5s
zxmo{%EzY2@g22SZ5%Wn4B;c|Jmoon4<9$<6Tp|87$?}nLB&usy?CI+-=Gcr@FEIoU
zw*6+V>3sk)RKlRqVVjjPye(F6F<DIHYi+g+;%vup1_-xci1?PZruZUnw?swOMb*<S
z_Q`xsu;I#8$E}*0L_cd?m*31s=9LZ%S&MaCsd8GA9|vj$=fQjBFp=F55Q6PlwxcVk
z)a|9<?~R%i;Q_>e_1>!+Phuym5~O#pRYy{hHoKs8_|RH>XwwZsms-HYn~7vx76Max
ziOY+{al~Q;oA@X)3Q(h`wlt>k9p|qF7$!u=N@m|$eirFuz#a3TVSyOYk}Y@C7Ax4@
zvg;Jrn1@!iEklpIYo$um`26*5Bgvrz`<#jzBBObrD=fY!3NZJx+<j(UsWu&}^fEbT
z`86RlU#O7#qQ{QSo`Q4+=a_lb9E7lO=2_Rv2sHR#e*3vb5QY>j_{6Q!`!oWkH9V&J
zBe7Wb)tT#a31_d!nSEyqE#}P4Xc6z_lbvznz%Jc2)t^!^UK!}xrmhr@iqJ6~d8c*;
z?Ja;_NX{8$j6OVK6vy?FIQ%6}WLLl*6vyi>8uE5D_^NK7jq+AZCO=xiS);!ySf%vm
zDn{m7F8A=(`L{+-2V)19U2Dqu2d$s~c;Vl_-2CJ<=M6oKz#{&<Kw(NV>w^saX0ZxZ
z5U_EXzUQ5CV_lPh`C`G|x7SFAD9t|*y2ek1LYL36cU7PIjh2fAA<T|zsi6PTI?$u)
zR}JyC(@w{Jekxq~z>OcK!w{2td}Ty%nIe*5<d#G`*XaWKD<PM-j~s---8aue9^FQC
zNPmNF>hNFxo&P=(|A53Cj-J=d`b&4kbinnrwbJ{WriiLV?eb$NVfqkecuP&a0Ey(h
z;4PgJJw|K0*H3SVTqM8jnn$^t{&A_67jV-Em4oNl!K{d#9P_JUHzohG-H1yHQ7eqS
zX*d5>-;Z8;)#kD^l=LI(s3bqYFRUYLm}Sy#{Q8~9#6%2j*vqC83nj0BH>DVK7AfO4
z$`wR#t%7avE7TcL(__>==SPs6eOXQq+Pzn8roN5+Kl?uXd_j!&;d-3bul}mpRU)#S
zTu_SqL`{6}NKSv_drf3S7{?9%gKxKI6O#V=>e$MOHuMT+{RP8+(ZK#)m(LRZzUgS*
zxxWB%fizbkAP-)_tS)<AZI~(Q!NK;P{vTtC(klE{RaeEt5C7&{Lt-V*)?GCs`j3zE
zKk1^l9FUK0AGt&IzyIj}w|}Y>!#{elVA&v&{YOCeg#dv9Zo_i7Zes!j(Q)UKQ@?mM
z|DIm3??90Zyk^-W$+mwj$?4~h#4jSKFAcl@SJan8(Mf0^q*E3?B}irU0A}hzUs^!a
z+s4mOdu~*z9IJw%KBjS?!X;wK=~JuO;0c-vC278G5Kx`kYWdxEgRI~#ZfHyw$wzMe
zT7QlE_G1G>Zi^X(Y{`I~<>8aVqz=M_jiiPDsuI&Krqe%l7~924uiozuzS(lP;guor
zi?G#0nZN~W9@!MOx6d{rDT`37PO!BeKcaT<G2oLBg>y>Jt-l~&$MrH-g@6lQ55`H_
zZvP&ucp<m1(C><DV4Zjk5(oD^Ig|pWs$!lo>;yFHU8;(kA6fFOLH}0Qua`aE=;ouI
zskcJ%n_cx%TO^cfmU2nB%fH$-&NTl*08=H6jR8oH0Aw$H1&XvQ6^y?0kaKe#BoZZ{
zyXcXwqSFX^lLJC}7~DG07ipUdfIR^8O6R!mY3WzTYo#xcP_R8fqR5~<=x{;6f$h^D
z7d=JvP9exIH&yu1>c_RYFXiye6LX?-dmTV_!Ow$T8dIY}!W{q*uf}6k`%F!rgApYm
zd-ZbbWBaHN{ugGt9Ft{czpNTd>Rrpse#^3WA~U;Enf>=K^ByEDEO@Z-KKZ3OeI<SX
zRr0AbRQyhNFaf9_&ijp^A2Vhy=a~e}mtJ)iHj`3^deuR&f-O`~QP9*QMKNNgKDDFX
zD4s%vC6@Y;TI2vil(EajY+rXEMQh$48gIdHbCgHAAr)XxS$&%Mbywj+?jHiv$8Sd+
z@BJW<)pTX?e}McEL%Ww=g!-?C_ju&UdF)uwJdEY^CSWR{&<*P<1GPiW6Wl_*!5V&g
zd<Mye+-rvr#`UV`>4hwmXVun>+k5L$?g7K&@kc5Cja=a$aA8IOe0cgC+WA*^s9sit
zdY#Nc`6FO?24JLYi%?g8^zvJS{IXRuN=adpHS3KhdyLv9q7ki$pH}bS%%}oAF|M_4
zyFJ%Hn0-yvZ1DAKq4dfovaRf$HjODCe+#kJ?~1!hQKG8pH9@yE&4gDjudt`P*=TG+
z$=&KxjaZew<pRSYD=}Y;?YVUF=Tlo~>r^H8D9<XTh*?#Xk1N}VmBuutfYeaoS@H*2
zk9<Yu(C*Ut866K97Ia+uR}yY-SA#SPLJ}c8`v+z8k!AiB!LdpHr$U)R$-IaA(wsvz
zzc61tu`{xpHnYei@q=>fvXJ40r<m1S4ql~tM?St@tY3FEV|}5>bS~$;vYT?>1~28c
zw{=Ge*S;uk=L@_c-j(M6U`0GGrq9eKb}o6=5q?DviP1~t@3)z*DplS)dM9CL!$ob`
zU_JC9`wQivlLk~7tPcP~&Ocq+7NO)g72;p2f){2<BBbdh1OKZ{^BXDi3i5;!j{E<0
zn*RF-rREP$s3Gq)!>{YX%aebpxhmRjhsbl=#8|zJDhI7@g%9GoZaIE|zYC!a&%-3`
zA!ykb6SRx_OwKvhk!zN~1(GH+;9tWfvE$@C&_r2>X%#=QIRq<GQXPE@@1{Q8Ne?YU
zDiC|<u*+Z}E#oUV+eZ*bxKmKqCNf_EAxS12;XxOGllQm^kV6-~0hMAJ36)2fed~as
z-)IX)0^z?v9siBv<M~mw6s2}?=9*JDY9zbo{S$>4?&kR7dwm^NrQ_lEi@!P8fhEcj
zAq^UFAcnVI_R5VKKR~o2;EnqbvvSlV#UBK=_Q3XK`TU2V*7FN+zk5lwuJ+y`|9{YN
zCHyBH7t`clI<A)=#5|(+mJ-vj!f=Q4*ac+pcAm1p$R1Q<TMQt6b3#mNrLf#LExp~5
z!u}gzFI*!>LA$iQIBV+J&Do0S^u@YXA0<d~<6+KRNs1_LL9+978`|368d!5@xgnP^
zwTRkg3b1UE1?t@irS!K7xyc;tJ_XYYxfb@ko#+Bs*>nKPty?4G1F~8jp}sChAX=l!
zPL{-Rg+9!R+fhBKYj|q0rzJ%_I;@jldxR4B9)gOBznL2MI{b>>bfMxjHrwww_vd%L
zR8+nK^Cj=iPk42K?%J`2n9-}TOL=({;6^gUD%W{hct?*Fuk~A|HqgB!W8F}#B0s8)
zD5~rcMb*4=n9-U3;~!Pl%2j65YF*R}XRSe)*Jkes_{q)ngr3B;OU<m=XDS>W`^_00
z0bsIiPbKb@e)taK9$~%e8^WHdMHZLCQR!Sq+`tWF>M*!&TXC>Raa!<V@PJtV`gNXo
za5XwYXMJP62Dm}<>n{g)CV0~MbYd3Xv{C)9UywkjG{tx!{bN6%UU8`ot`$QH4;N7F
z>SbzV#}dh*tfWqN=iQN9hgniNf>!tS)6g~!D|DCHpfP;qbr0dAv$MB97S~5nWEXbG
z5Ofb@It4S`pdrKl5e&0tBHVItu-6F~c{}%C#$FA6XAM2h`_-O@a)2LC*i=rE;3G!S
z>4Z*69KhiA=$wLiC2DT27Nw|VJrGZffX%Lz8Iu?6rG~gi&?Lmv^elY<Vy~EzIZ*4=
z5690-QEC~j1cD4*TBRvNUWCY$sqnV;M0RX%U4H^1eF|jQ34A2ABME)_f|m~fVS|*_
z>=$1G{wCzAbRs!UQ{R2lcs;;l=RJuMz^k~~KOAe-SW0_K*U*k{JAcX!r(dgZ2ijnY
zyv-iC1kmdVGCof1;G2vc{SZL4FxmAJdzIYt)GK>Qg!$AEEb7sTs5kbA@yu?bY0*;p
z9#=f;ONuMHQ@shZLA>(I&PW82rSbuZRaw!Ces))=1I6>IHtN`8KjeWivGPP;*42L4
zNunwD%n34|KSLOiOw7<sW)HHu8L0;WL}Mz<L-*r6m`;l2YnZY%iL(kdE&Qe5(C@!(
zf;v1WiFBB#$`O275p}C|v@A-483M{YmRzL2`t~CT5V|ZKaIH{U%DhX(#xk#k24=y<
zM8R$n_KoG;K&{M_yvrJJVrMye6Aj4p-+_^oYL!W$&$w*Z?I<-5=50M@2elEw3_+E|
zQRtU;AixNPIK8@?3!KDXi>xgw&GtG7bf?mYJo^KJF{_9w)4eLB#2H-NHpoS0!K&ic
z-J}6_ByN?Gkr78|5!$}z2w#{}RzE9P*0|X33Bh1+N4M9tQ$T-#b9E8cxr6f_Yxjge
z6VE%3e^WU97q1K`CAN?JRRM~;+K=LelyVyMadAkIyjK0VcnBIDYuJW8kF}PA^Ztph
zU!D}U-6)f>CiY^Gf3A`&_S#X@`j0S%XGc)ZewUaoFdvmdQxjJ_l?tK|i9V?_)p3*X
zY<~S~>*V>h*>aZw)iwM$zIVrca&~^UX}0NlK|w+Jy3n1TH57V(X%+Dn*I_eTXhJ(M
zr{joGXe>Pg_f&qXI6Dq!5De<#*Nv5hA3;ME8F`6quDh*__&dgf!CzV~D?C!YIAa4g
z4-=!FXDGBdy-7m0N=}AIlpoB=8#}qEjZ72?DW9Y2={XV%jKZTb4&aiAolxjo^bGw|
z!BwX}dp+=uy%Q-B;TKNJ?uJ%kV0h5*rMs@xD^~vxbPxl>C7C7`ZS~W6SKCz*cTEhh
z6@OE^X=eMLh!sa{f<o)jGt|xoKgY>c$fsq;q!n3I;VkjT7@kWvjl?ggRwCoZ^>~{P
zXVB=jk}8}TUg4H`mHOe<zN?M5a3Dm@|LCQG)81IdL>)<OBuv?~-8hUrjegI-@WyYT
zl50^bABc~e?&ux$Iaa%*+#VldDl}TJTTiX~$(HCUJ;N(mIW<l4ULsZ87WFC*R$*~6
z^3DlsJCid-vM14o3?}<Kw})p-lip2Bc9}Gb-los#Pr+%Ti|Pth4N_gBr`YH8N(Sn&
zaS>8UZt|IIjrC%g=Akwh{M=IdyTQ^Q31Hh4*|%L3a#JypZ@4&YZb$59coeKeA<;sc
zb?ojzt}rK3H<TNjoh<EtqHR38b~df@i_kGCIj+da)7LpT?2JZ?6jggQ^-P;l_qsU>
z+x;!IBAyKWL-GE1>r<qW=&t>~zj75tgy}9Nc<N8bB*$fZ^#F3*!}e(TP3^^qOG-vW
zF_rlvnuRorIQRBGuD3YtKc2^wHs%v~z{YHMe`_hUP1^G^j<Shg)v%-_>D1!$T5?pQ
zR8|oAxeqIap@CvXmkLl3YdKlPVr=bFd@N@^WpU9EJs%+t>rByXdA-t|B7f_<5nTbx
zYM80CrG*E=IyDM!4b~CmEh5I9pg$&8&a_eRJ-c%NCtp21c|+rwaStiF;E^`@TfK9O
zJI^O0wY^p_QHP3a#O{MM9#`v&%KhN<{T<RX_f@9nw)B~4Ty4)9YGr##|6S->v%HmJ
zVkO43^LhD$#*}?6U*_DW#po`Dl{98esC#qHxakk_e0!)`XJB+?H9?!BK>lMicwAni
za>~ObT9)yJXXh*vhjO*UjmHwYxf)9RKM1LM+7vvCj-X%XoXK&&IH@;@>=+A*^b@SM
zCgePM%t`2ik(e$EkLS60kG+ynJhHVqxt1TVXARsz|4OX<=SKf8PwMd&V)KkAr~XyR
zx;=m7b(H)1?FdX?SGgVpTh5XZ6ABf>lj)p8uC}ElKfLh`1`9Lp`N!y3J&JVCcnagn
zEpT%WU17*!6d0B@uS+{Z9#dK3dnA>~O<N(<zC0`Wy0^8DR$`DX*?8$GY9+GVF`P>N
zcUjl`dK%xm6}JY$S2>9VtXQV*cjlXGqhl(|SaivD!0BYAPT(F-=`CQBNcjfb<35e*
z`2Oee$6wYb#>=<HeCEP`f~op6MKRrx&Qj(~+|7Ad=noPC<ppQLiRslkH#4~G=UH|!
znT9vNU!NX~m`T#E9t7Kf(lbN2wzLSG6X_wN$9K7F-_2Y_h(Q(lR!4_3?$Y(E>;+LI
zU4$myMGd!CTY7w}jr`UitMAR&DB?cL*Sg$cUYLsm$%od&1p{}#32OG25*<vV{Aai2
zV)%L31=QX7<GfyF3>Jn&>eYP8^5B2bWAD~;{l{xa%$m>W4<AyTt4@i`Sz5TCbfNdn
zy#FWnzq%a%@?idMPMu{0EtQQ)?XTT_h4_H+5pudKf!=uto$dZN1y3|__K2d<72F@r
zqCZm2-DYf}H8eX<IZU2C<#ryXMEWMg`DJ!+J52Yr19?x>l^b8E!_8iw*SoeN93bH(
zqKvnNwgV--=3>OnYdFPxjV6};(YirWH)(irDbY4=pme+u7YU_z*{3GX)>+RyC-+ZV
zJuI17BgWbE8|Z`D|A)P|4yY<^`-MjbR6s>H0-}T>-6$Z9BHc)Ca8uGL2uPzKA|Xh3
z2+}3pASFsjw}601OLv`X`Hb_v=REVweDC@9eE-cb%x3Mi?sebS^}Bv?KNcXK>NtjY
zrdpFs3!G-8M3^izRPQ55V;vAL^r$WJs%H8fXN799mA6T;QaLW;IA^nGBRkq{6WWkg
znyGLp#3g%MjiRdUX6w#sx8C=`h>r{g)wx0I{H*O_`_XheW-koJR;LGy$=A9~s9xR7
zP0sJ=E#O1u3@szCpVxLtmN70ojczF=r<E=mF)x&DPXw|}_ISGA(hrC;vEX2s_|H%w
z(pJsh1<cRs0+;t6C~Bx=;@LMIB4*LKYQjR`mS9^B_UsRg?O>a%{3ubM6c1{~+Jm*x
z$sleq;TIs>$Qyb0>ZPE5yPVSXfvF$W+gilt{$zY5(ku2asjtwBN1Q$ENdCF{d!s+!
zCZB<R!JcSSCZBYww=lGc4JV7-;p;Oek0&@iYT!0h^#qQO@)c>j#EeAFoRa7MC}q<t
z9e^1*qbmsxGF}%<BD!m8RQeBKzqoxib&{8@tpZcy%BETtHmQ|a*o;CvgwP90mAxF@
z(Zi+j+FI4*knzbf7J6BkXe2bXQ41g{IaQ#pl&+<2T|X8}HL?80-(g1dYKs%tJwi=`
zMq^-V;;5c=V0$Dzh{Thr9dM=;&T9;Oary$Rqp)l!dpSKRcd~rE<NmHcedZ<)U|wyW
zFz&y0px?vpWWlB}!{L>f8bK=HWP4o<O|*bhHx3w8oN-H=hTKU}oEsdRW>0cqn@<%F
zJu^jdyd(bP)1GZTZcH4J>P0??^aAInC#}Lt>b_0Tb9P&(7nfF<yk4giyWV9GfeA0I
z)U2(c+_`$o(7$v5EDZ^72CGOG&++R<;xbr9^Ok&+<fd9pn7<p&J2J6n<GS(oD{BcO
z4o1^@-b!T#G3l4b&#rKu&qza|+d@_D$Zl7GjW2D%w{-fa3wFru$q?H+Qq4=|)w^`?
zDLVP{0A<<{63!(}r0uerc`frHdiek{v>!|-l)mAh>K->ra9y)kI{=oR!n>1yzSw{9
zy3Zm|KW~8s{iUB^)mf~QIk<#l?#Hgg5xpoJ$p??VK0Dn{y#qGh#yx|aad}v5U{nPu
zGHPhri(aV$4<Mw+Wxln@?ydP?+33qzP!aujL{w0tEdB0HRZjkgD}r&R{Blkt*zU;t
zHe8vqDb}KBoA`4NNV4fiT|ETzMq5Aq=zt==G+O{lL-|3C;)7*be3_BiI(!;YQWa<h
zB@WMmXPZ{o09{)J01BA}G;h*}(W=>iLW$Ersb;1!c$4cq_t&Q)!u=W^5<U;iaE``+
zTcPjATp}R|vYKeLhMo`ggEPvv6hip+e?9<t3)xAdxU0n?krxZ=XUY-S2RnM+DqO>q
z{+K?N$v|oai5)1r47NH0pZ&CE2YwYhmppE!EGHb2h8xSf%4=1(F2hE(wh+-6i>7T2
zz+-xMr+u!g&uYwVe`(A<7yPx&Q7R)jxJNBjFH{Wt|JI!2H!LZH#(K%0>W^nOi$3YJ
zN`dW;UJcDY4cDQilEmQUlazE>tH-mYpcmwCUqllV*-%Bs^NK!~sceF@{bQJOQU~Kq
zQlt3gxcp0~FSqy>bel-5+-|z0w6pxx6W?CiowsJvg#KASJgQlx2=d2m>~ra1n~}9;
zp^;(+h@}~$W8gkR&{1j9l(CZ>m?DQ~cgn7Y;^SCzxSA@GQ;9I?W!|I5;1D+@eQe_Y
zU~pqvA(|b+?5@8qOFb-#w_Ee;DV1`dI`Mk&ZCw3JFF2VB5y{ltX!!!d$i|COrT`$)
zY7M9!GcTl_flyHM`Mw2P%xN#b-A<0^XuVB9A<{(xh2O$_w|FuwjBdvnV1&$r<gcuO
zk}B^{-BvyoN1VBCJFKCm`8v`Hgn--jP9I2&l)ZeseNVPYvsM>LS1Mmlp<bO9#$#5d
z6M1D)T(Fq6+4b_Pe73%%{F1^F4Td)8)d7b=!La?98@i^q+`3m$e0uANl(^kKaJ8uc
z!NIK6ou+7fn-vagNWI{TO{=3;03hrwHArqf+U65AhAuP1-7j`6{IJhCzL}$N_3E72
zrBan}c8WQJ<fMB|7A2&<1M_yu&4DwsKn93O*Wz5N{Oq`?xo2_+-~Gwm+|NqrM{T^D
zl@%p`Wszg{@Vvrw6b9;GTV>t-aJwc*f+O0_<UH!Ygm<+_`$5MCX{!k9bJA(eynhcY
zFCmLa;DO{HsV|*eu!xR<{8*tN(ch=9!osWo<Yok>Q+%ei(rQ8<<R_wm+7qSC(g6XO
z(M`y=C>5sB^}w<L>i44pNadkclI0A;lc&-MekmoK5osuMPg=H`;bPm%i#LK{xHp7t
zAtha3(oF8$7RcAZzfJ}3svEC{0mh>iOj<r*+kvfkiYD(cypRMqT8N}vKY-`E#5dLx
z_u@moN`{ApHt#4s2tynAxM|dEbW3Bf0;H5yIlaCkhkwsj3xg(;CzfeTAQnlBd^~nq
zWS%Gynkov!M<{fJQjOk~f^-KkY@-6d_U#aC;s<P}@-)H~&#%fys4Xz`oabBYY&uR3
z6wD;m4&QFP<3VE`Vkf{J!O7qwtsJojG%2%Bh@}$Kd7f{tr4BS05q=X#Sgu>Na5-OE
z6?4@}{n!zQN@}WR|7w#`ERt`Qhu7zmTIK?{l7OVoY|?KtPzI)>9*EYB!(3&~BRldm
z&XK;4ypq_A58lHvu?RJcUD)I9jTit9^}OC>lWCdv)PEK_J{T)$`UY^%lOvQ;YTR>^
zVWC#0LMnzprjx#R&D9WDt7gq5Z?GImO%AUZQ(H)e0mdG4+<v}AEa+WST;J+6u|;dB
z@Y@OMFWl4Zt(CfvzWx=419?T##qz~HNUXM(Htu8C2dDcRV2_&L%yd(7b+83*%*<HE
zVx}2RShiC|-2)<WIR*ZXlB`z=4>&agZssKid18Giu+3Wd4&ej~lxD>#h}>$#Z#Z*U
zI;aOoeYbqGYUs9}paNSa<0lucP)lc^)7I#g2Dh_nA&+Fl8f|+hoi@<I$rN{f;qOlB
z6K9GLtbZ$?+Taf)S@X<Cr`K56sE17n*{4NIVKG$JX7_zM84`<ig*~8A(b}ogo7{;m
zNQKVlEY*BSQwE7)WZe{FY<8C9%S(aDCi2wjExb7Az0Y0!K0y`Q#Sw;Dr3c?QnUpzt
zYNwXRilF{1utlijv?!MIFwDv!LF^}4dKG!rVwr53`0YzCo%VUV7JH!fz*<yZTB(Hv
zN-m#9d{~cA_wXeIbFo@(RKousml7#un-2-GWt*H~q9z~Lf<*)zbNg&0-qM-*kFvkq
z2iyF0DJ~=dQ|fA$n~xOrHKozHG+r`(hj*ofVGn^N_nFXL6(y5{P<F`WCuyvMiYuJH
zAWS4kmH4<$?S@Lvl+(K!HV&20vT$x>JXTn~i$<obIBl2Am)kBoE3<K=z;PpVM^`bm
zW>mM_OmUd2QiuunVr74wknQ(rSgkE2e2BiJ8G+B9i5U9cSWV4-Z=!)lb17>SFl{AM
z9HysUDE7a(*|e7q@i?D*WNyh4n#E1!2VP1*!gX?7`ZMvpLI&^8yx3FsitEVEMRb}h
z%Kattl7}-e`ctR=p!#K&%1u=rO(e_bbN8Y4n}cuVs-M>*wHRGxqnIALO%^EDX%|k@
zt@B=iA=KR59^FD<w40V|{vHF`yw{Gy49yYF!$Qpxwy9ps7&l<5#+XwSoUooB_lo2e
z4ZOyt6*|RyURj$;k>V7tMGThq$zf*dT+AhZUs)UDKl>D#jXqcRYrNkXcqY=TKqm>m
zBVYFV)qW@@btq4ali#*{z{aRlr?q#QphZQ>K=_$Xf2)VjiieL9!x65TYPttn0?$x@
zDKecQQi35H%8a_krxS=yhURE(#SlZ64M788Wpjk*Ae4}}twoV-K`EYsfC~Firs#rs
zW`pfYrX3kjjwS$M&tAKILWJ#^aG3Vk3!F#wP}L;ilR$;GOJ5Cmcw!4{weDn-&dGok
z@L1p`AQ*QTlO-h%hxilQ&U!lnd#u6_L^8jj%%wWsy+4V^eVU*Ro4^#CfPDG`b24eB
zc;RGdj9WiAT*cqs&H4f>?=hRSBDsu3OCy8UHVq99lH0)Mr=%rsINdoS!Bw4P5#xI^
za{4_Rj6_(EPn;&b&}+G1*An5QR4v&1LzaUC{Ntu|CIvg>dCBr`X9W<l7BZ2@D9n${
zxzg0*7@Bk)Hc5Z^)^;VL)Onn4PJWj{z^Q1!&*x5D+-DpDPp248Mz^pMGmjn}z5L7@
zj8-kn+DB^<I^j1w(QCLgr|=kJaL#z5o3lex81!Pe_U^mBeD+y3232lo(s%m{ludQn
zDJu0Dfy&yiE&U9GI3g2sP&9dhlb?}5Yh<KTmo0mW;NGdi!P&z!wB!0#Mosd^sJ^yG
z5FNL_o3_rjq})8eFS8iEm8)U3#UJPcTU?Q+L0iNgPT{4iZ8q9$*UdIq9+cb<D9BCb
zbw_ZK<>kI+apL{^1p8un72y@%=Xp&}Zs?p;5Gw8Sjtc9K9ynhtikULw!5qu@UzTyf
zFsd7GT~WQIX#Y2~Oql8_T!Eo|${VWdzxwTHY6YqtGcJKiSed_jKE-l3#Bd<3td<T=
zR2%Q`dpZ~I(EEB13t!Bo0P_9VL)KHX^lKjLH9FM6rdL(a>0h7o7lk}D>%7R+^>enG
z6!yO={#iAc+iqsfQx;yBy@h2ytW+vRl4CU-+EJIPymV{##*I+|?&(k^583>M6WSo_
zUV+-L5_`mn>A)_*Y5yL~;)Nat{09{0kDlTVCw#*`{=q+b*gJ`*ag2GzWbrfR(D1Y-
z9)Us2ZWm|McWAXnJ$WV+z~#%GDse7_Q^3+TT5umiECXYh_D?)RW>jg4@<yAfF_yT+
zsUmXTOF4G3Ls|5;oS5x!P0Kl@l@A#UoEY0=rQI!69&VPuuYeAbW93NA=-cIg@eMjJ
zVsX5hdVT#~JutZ+a9j`4j`dZVtOY9uFA}5J{2MJ&*MHv9#KJQY>U$vx2UIQ~zt>{J
z)Y_8uSOFzf7bXSs*(a5*{QVQyI}5kNHI?DxnSWiM&ew~VA;mb7O;<R=qXVLmbipTg
zc{1=1!dD{eg&imW8emv(nM$jp0vbmzu=_Ep9(wU#PE`L5bN{b@OhF8^aK%v1{^9rj
z;`A*-5}dZysO|p#tNhgxyF(8PD$FBD<gfnYKU~QY5@T5HB1!gl43NKGYznx@YotSU
z@&DqAosmF6R_*(L>$v&XQ|?eC&XWGJ<lkod{Ph)gYC%^$+#1pQ=RtKR4<!*yHG6wV
zKV&yg?N(2H9bXD5^R;WHSI>0j4*wQL_2+N?r~5dJP?R)Rga71e2#9_~%sh6$=c$s^
zGBYD}esUl}poOIw=1W&$;+^rX5heW`=CPNE1f0KsLv4XP4%Py*xbBIe6VFIb7G;sr
zIs7^y|N3jdPd7~9o9x+czUKMY&A)Se?HVH*<u5C^`>i^B0<=W@q@X35029r?nd=%q
zDqz6e04!D~CT$d%uP4(y=M-EKV*ZMZ=^wSRH$MkMkR&!~NP#*>8)qB5tcHQf;IuB3
zsmy1`d-&#=#pP%GrZ%nM0*^56-^l-Ay_^++3p}Nlqh$EU1%ARJhq8#x?O=xum^#5q
zxkBYgI?(~RlHh@nL6o=?yxI-m%2|&Xf3g8Vu|Gip?>hC((07QMPZuW1W(+9>t%#X&
z#aAn2;Ee6Amr?~eWeXF}PLrR34oBuPM8+!_7>mDkYV8Vm7-BesO2pV%j^4ld4@<N&
z1KuDz)Asu;e_Sm->h*gh$m#Jj3_f|u_czFlj`-Tv+pkuT%B%}9Q)iJe&>)Z%%)aN1
z#pQ(G3*SSs1j;rWoUo!GHTK~k<X`Q_e$DH$>1~a8^n1*2OQt#lLMi{W;17e4EmGa=
z8_6GAWb>It=u#)3uDiSfNFE{LK)oLdpPSmPhVA|;+;inGHj{sUb9j-@mCIIX`4@YK
zO!iLChoc#xV;l{)B{OY2Xgkr!igndoP?BNy+GmtGOM47!v%BGY-7ffMT%DAkYoQP5
z2Fu@1kEc@B^TEicY*1cl|2xmL<eUMft@L0jUr(^8m;I8w(DXgsdAoLv4}R0W&4G@7
zyRCdR+>8g%|9|WGdIWITS)rXe{0b};p2nn$12Q~%Zr#lp5!$LN#69<Szcs!g^4nn7
z!~{0WrS)KVx~&^EM)p3@=N>3Gn>y#qbP2h}ZdBAaB}}$tu-rJE^JGR(Q15_4JcVps
zJ?m8?e>2CG_Ipnp7cFZ8uJRa@#I0R_vDY`X)^f3x`Mi|)$dV1O81FG}KtKjBz0DlA
zK~YwLI)nCJ6IU0-h|0ys2mkoGuffLEI^*#Nhb*xJ<pA^M26&OaQU;0=V>p;<b|xnP
z;kUta1__!A_rXd-gA`-VB?NJL1eo!DZ#LXIb8ZN~YY3KR8A^G8D38m4n_CZb;fpYy
z4_ble;Elx8=mJQO1H83pd2BvtTkI}=g{~KGVxab@7*7_4^(89+(!2n|M2+;`8aO5i
z)Prz))F7b+%wDX@j-S-5$Gw99ndgAy@F5B82!?mJSzYHDc0Fi98<4E3OQ~9Q4{izb
zfcFmoYL5V|d5%aW2Oza*_+9UPX)i=lx(FV6x&VUysGo$^f1}C`_z@G(9^qXze|J~i
z_>EUZC=eHC&Vl>LolygYlkK5Xw`+G_4LI}uj!=4L7?&cJs}n|kc>$D?W4pWWkBnEH
zdtle6G65fsH}&Dw<Df4$TbT%I_D7drfOW=gnOp4~81-){L)qH5rY{^rd-)L`K?A+<
zSjw&epx$4~*}eu!G%}Dt3M72rfp%<r?@JR_mN%)*OFz9rM~kfkiL_*s6xQmE`$9h%
zCf=aVW63UPVbFz{JVf!m^ISjdNjUnZ2nwn59j&lEGDxOl7&>4wucZ)A&2$KnQDnAS
z^Q0=6Hoyh!@P07h+<r%ScL<WuXPBUYqlkwhaRt~@&!OMF&IEHTw*cvHynV__gsGVv
zwgKsKwgG0ii=b3#ORH0*NO?T<q;@}!6;R-<dRvXru92n8v*CK3I7^v*6;y(MUprm6
z{1W!4bB!8lZ9iZ5&O96m+d=?sya3h_MCu|I_25}Um<F5!aaWd~50wrtRFN)DLr}VS
z109{G4fMt=-BEx+X0{0FHJ%|pCH1jn(wDKqtKxXS_W1F|U@RUmY^%<bR*$qtG<OYk
zrzsZe@8P&p2?lYqb{wvF3!O#OH+?MV-Qf4jV0Mq6S9|1y>~sQ0;2=_OUfBLUsR`Pz
zR?#P@V@bLqB#iNqqT|wMPKMBs^m%~S@T23^<xakBBGP064DvM!?FuWwK0g@irAaBc
zY@L#wVZ0cQ<%z;!SVq%sq0C+3J?Qws@Wrx-RLU5tJ)09oG8<uPTGmH^_uFhRJP@lx
z{A?yCyTCl~Id!;teNz&O;czq0{TwPs63Y<~sItq2Z23e_IFtes4i}9k;O*Mx;sjrH
z=gQvl>l5)z3Q|}O(D%QHkR7Z5jpuWCJpCQi`HIe;D-1gVBR|6*LLug5#MezLW&mZZ
z6AhkZ=!zYVIiT+C+r4oy<+B=&<#gkNC3j$*t=(h-Dx?;1!C*$!I3(qe|0u0J%t=$p
z3lj+JeLoI$E`LHx;2MqMDi|ux8NB9JbwS!d;&dW(OGw$a^Z34&f^)1vIGX>JmFt2g
zi>e#oS(t;gb%m3%>}C-QOZF<KTIEdCmNFl9|BVsK&pv=m_Py5d#IK@o)*Gt?s0WRy
z)iQ0F+H3IioN>TognT=s7Cj;@!#{OoI48gUng%)8B7&n8nAwBW^iBq|?6=+5(3>oR
z+!wC70@kH`m?h8{3b6f2*ie;{@ee^buPUpsNww?4h9xB<5gO@LB2}Jw(trzYDU*E9
z^OsQqyD@r3!qtt>-9CZr>j9wzsjNa+qy#;?v?*qd`1hC#xvw9e?U^m;viX)`b+V<A
zX>fXJe$q%-j@8pgf}pS<`~he?RdgHcnXb?bTM^Mb9+6dfj=DY1?2V?~xEaa$v61P>
z@qFkq;PE4eMzPi(8lU_9u%eXB>q*)gY1w^K1n^UM<V#j|yci#!2U*4s;H4Q$?fqfr
zHr$)7vR|y5^u96sh??e$Y#Yib;!vJ{RYZ@7_^_&@UnSx_vr{CYaECk>N#U93NJ4qI
zq=$u-j3bB#v^Tq=RYU<q`+0vt%$)XRSJZrZ2pFroay4{^OqMcf=OZ8f`+@xHvHWZv
zp2<;O%ev0}t4i6segfx+>f_A?2bJ>U@3(8`k~m`+3MX0bCSr+lVOurj8H?7_9VQi;
zn1=xF!DsoMr4@CsRJH!Oe6CeVj}E7#tpKg3dLFN#iR{&xX~e8dBq;$P*mnwTX|~P&
zBq0*92C9>jx|!i|R69yM7EaS(eskM=UWHf(F+(v~F3G*k&~`J+QDrxu-ujULBkok&
zp1vLtTUI@h1&56r60gM4$Op7AzhG(eOl^E6s+~fUY=XkK5rZ53UM3RP80(VIVKQwf
zL*25QPejh==drbhe+GS62Qz;gIJSgKJg2{C=^t_a06Ng}-|ka>h_|gC&AnnU^a<|Q
zLfpfDD1e_GAyv0~4$s=ZqTzp!ho+rN#f11g*QO=$-DoUoQHWd)X`=HPqdHq;B8*Yk
zhNH3~LH6;y>QMzL-L&G1uU1f+XVRDSOz1Dk^y3*=<nax>{0hX-ACCnQ?BU6$ifIva
zX=~w!eo2p7jmg$CCJ?16O@!7DDjV7|NUqNYyMY`AdBjAlI^KZ&<`U#~eH5<KW}NYs
zHHaGwEp~HD35>u*sC?qDOUNFRyoO{zHl{A;?jj@ufmaAFKJ-m`L^Y<AK9>N!Q8?#F
z5cA@;Bb-eHF`oy#UNCrO3lWvWareHp5M+*JUZOt1eam#Rn|Gy-f%eJZJH~Z9W~`R;
zZn4O|uf&&H_t*0I-<H-Nhyb)Noo}#cl*$*UNq-SRp1FzX03{d)QpR!N1kP&NMM*PV
zH23Xz1aaQV2X^J9bB{dHJ9sz5{L|Oet^oGd;ep2MZT>CnkjvGAZ9%tAM1BI9$u<d9
zBJ&9B!WKfH<G&tN8Q4OCp(o6FJ1rYVE;ZDpZl5!h%r+c^7HFi^w7=0qwomicB49?u
z=9yo0Jet{zI}@XIEqn)c^vYJC#=wd(EM~}-J-$K&XoZdvtv4tMs!X*Duf7URp1o{{
zb8`_O@CisKX2HF`S~?fk0Qkfe1QP}t0iW{0x~j}eBDdXE?NbD)*fBJ>StT;uHpB1P
zJ**$l@u+<VJCgF6PEwhU@DTi~PX7wS5#phBVFt>Jyva1eIJ}%#3Mf}?*|LHr;LOUC
zx?P!%P=<MoDcOYNr;fUk#l0Z|m>a1j>YjUc@jWr|c(2pL?wxS7kv0JW&X4DMeA8sb
zNzF4yb%cj?lw3~}>|zs)I|XwYSVOXg@ns=57q&X>K{z1r@a+b5y?M;w7n*MQy><H>
z=_SmXC96o@=4+&TV81v!XWz%rDrc&;lrG=>H|7mLpL3iXycTH#Y|}r*3(o?m6<&Pp
z$lmn==5K403u@RL%q>}%f{}c%LiHKSBoF~tE_n#yppazJ=uPSq;j&iq@b5*Y<zmlm
zz`z(E*e~9A<Yp-uu7>B>T$Y{3AFuYnEzu^d3V}B+&PpBljalH||IWk}BVnw><c$=;
zm6aPy@kEBjPHw<vqMAiGGkyBkFXj%Lsjh5$3mw#nM(8oVF%v-CD)Z>evS?qKU&e8x
zrXuAu4V`%igS|)35=25!KQi234{k%1JO_NtBElmGG{NDLwRQllzX!K8n|c3H{X5-;
z2QKDBLWM)dBlhwoSvS?~r7BG|y@J{Pp_-btJ-IbW`rjC7o%5kJ(V4cueV5@7$U&nR
zsUFR~@!t9vS_6*6c1iRyW;}&ozW0Qin)U;=NeSW|x+Miw-l5qSS<ne`>OHtdX<jOg
za?#xbSjM9TY!@RR?<48i=d!j^O=L8!1$Z`1NoKr#mV<$KllDEd#&C%lt?NZtg)!Pa
zfZ(pj$9}9eWrXI;&167Um}zKqkW|E&MQc}DF~oMO@pCd;2i>};Q-D~UIPZakMC(#E
z^O3+%Zi~(k{Zb}#m}JsFw1GOmUFcLc_-{wdGgWMNB>7ZNbu(CF@(Rvt$iRDF0S`6B
zsOM7c;gVIQx-qa)+X%%ane@XOg8@Ncjnea_NBMYGJbrcNsI4NQ@Ev_10a@0^LkMM-
z>sNpHs*%Zj`V>&2U<gR_#Qj!0ysmVIPS~=(Rz3AjnDkgwG9gb415ZBvs}Lx0Qt2d`
zwC4tU%RYhkMx?}LRX=O$`=6z%r(X*5#B`T35aNhZOSSrP2ExV)ISocRxSAh!ufK`^
zoDZ7~b3{tnpz?4$?KC6fPEQGXYn@9?nu!?xCdHv928rGC!73Nk6G~ytw`8sx2l`-0
zq`~{f)b`vcvY250DBw$&y`XnC`n6~nD#<1B{|VGU?4nOe$tubH+-hg(PNHN)>$Fw*
zi-F%y3k{T7Sh|#M(q-Bg?%)ISCA$A76TtW_8|pF!-jh!N<Sr-O*lj99N5ab2z6qlp
z$5`{r7SeP^J$B!H<3q!LW?wylVye<XqFIffnYwcrcF7R<9DW`oJ;i%_31I{{$p2&n
z2#j}Ww`TWPL!-CsfQCkLj%8Pwbb2aONiHJ_oKg%9Spzcy=$}PA_tR7S*1g;7gpdM;
zc80?fB_9;-7dUnvL4#ITE}2e~|K?9x0BcPD?P|U!5WgY+WFlt%A+&a~V8y=q<&jB;
zxmGd8ibvU-w;xWRXt|s4tO4vuZV|#gvh|;B;TEELAM&U>^`jh|Af$QDtG)6|cXoq1
z6-dLz`Rn4Dm@`=dc<0l<tV{2zRPMp-8x_OLy$kdQy^B(s2;;!bjj_W-Ok*IQ`a7^N
ztIt9iC#Dh^=|e|VjSL-QwQ-Y+s3<YebK+tbv*)OW^U7@vgdlBHTz9TUhWO;)BmciW
zdnWT+-T0kc*Oq^|mK_E^Z6rthd-2TmF|w-DR24pQ9<{(trb5nP#`>!`52)yZC8UK@
zXZ2;x;7iFMva<e4s@b4BKR`t;2P5A!eW)aCTD4H?k|NY^Lc3ckpu=wym4stc0%6w^
z7m4bQAT}V)eZ8s344x~~N_wRfI`gtbOW?A|5J5|>4^5|B0QcleeI#WY9(91o5fqmS
zmM%D6`hBvF{z6SEQ7YlHY142pg)(N}KLCWBJqk7x*I_O6EkA?TcVu^izAEf~A~0<7
zXNFev6I5woA1h+l2zy>Pm4UpFY8O-%2~1^QJ=h)k3jA$ex>r<bs6~Ozi8sp;k`y_;
zQ&2^F3Ki)yf<XIi;2V4V4Kd&54O3W!qFNfpjEqLsV8y9CIiL|FDw0Ctj!feR%z3B;
z{U!^bQc0BhjTqY<z<@NoZ+hfZJ0aM7vLtat+~L-4>7>XI(QhwY+ih2b-;NwC(#Gv-
zfi=X!11DyztX5_BZt6NZ>5jEKz$96pS!M6Jf^LKwPJ>8q9<a~X%a_vB0>D6P92}$M
z*MK*tI_#0nrDd*q7P+ZH;wrTuKr1V~;UnDh0|1bXo1Xzt@R=7SI$>woxIP69&l9|O
z-e$1?(x?E9c_*t<I4y7`Xp6G$Gln<*hnndZ0Q>$o%BNqgepBHR{n>Oq`+V&Yw2T!O
z<WH9ER00CY=%ookWTV>QyNIS~JHl9EQow7#Hp*I~9YHQ~QYB@m6vv?`ifChNI2-n;
zV1@RCJbyrTkfcJG9g6fkxc4<kzAFU8rMyqjOYMPA1`?x6jAt-q$_M68EO?LM<#48)
zm;{g_*ve(<>Zcc1EoUin`SagQBGk?>{ts0|%uRPEllW<m!4J5rC)%=N4uw-h^;Jp3
zm)1J`xpncE(I?|eOijv@sVb`IrkwB$CaH%&pOc4pkgyn$tm)Shic0T^!UPs-f$5^Q
zX8k?NNOkySy`(b3MA*EXP+vysib}b10IUc;wF9&yTQNXDl@9#qOeGhzW7f0=y*F0a
zoDEFE(=x)RYJj6CzWfc`#B+CAc^}#u#8IGX5yK_RC3y?sll@4r@x{k+e2Bf-o3HYY
zL3DC<ZysyxCIIU{7!lfd<qxa6&Q|racTbM12dGByF6-+mHFxdiJXQ+d(bIfgtcc@(
zLzB5s$t#dTXspA?hd3-iBps0)-a^}Ui3%{Ig&~~5O<74{L<+qpT*=^Rpx-$@qQZ1q
z`XB1#-=2LqBa`~o@YV1CRUT045Vd&tsPQ**d6}z&)tZOTGZ*JCW!81sZrB8)SjeE7
z)37b_*QDC(=cKx>9_(^R+U8Z+vJC4)?(2dA1c$57zyZ2$51(58D)f6iI4BLNEuBQ~
z;Swxh$B1!8Zqj!I;y8M^8BQ=oVi)`$v+BFjY}N`tXVpo6Iji=_&>@<v2OFeG8JYAM
z$J^Eq)Ip`!a~E%*9i_*{XjsSe0Xs&IGqOo|M#rOEN2%%sfI&33ElH*o1fmHWzHAi6
z(@d^cebSx=v^~ZX9ks4wszh_NQBjNFlZe@KTYbVn2%2YuD_AzqcqCQ-yde*pOe9-f
ztNZ&Cr?HAhG;)V*URI;+vZ*Y>sXcL|T=}Z40D?LDui<oah$g(i)ua0_4#Vu+8aEo%
z3~sMJ+`y&j#iPmI<dxW0J9%I=4P&=6P<D#LVxg@E;za|DBbR3u5Xcfn_N2Y4BXhli
zVqo09)cV|S?3Ar#OnP63K;fhs&uLlit1{2k!^qf)wrdZjS?ic|P8F(NMyz-fd2gQ3
zWn)YT<IV6@pA^$JQOhZ4qnztlV$dY|2i^j7=%xfP*Nee-{sj-uo;WG$QA^Iv;r$dQ
zNUtUOQvJ?z<=DT=2yy0QC9Qf4!z11_7J8h2C_+D??yN*m*K_^)H36i254<t+)b;68
z+2t$x#qvwq;kOK~<DY(}iqn~^K`r6(4{Nux6!7VVk<EKQ{@Bm}=vNwSL4~E>z`)?3
z0Z6gmr~5+FkeXIJBCh@%)5&N#KDmhOU^6M?{N*FH`FMQ-#Blidkcg}f2+3)pOHv~@
zPYkQHo{{`dGW}nr=LR1#mBNGt{n3ivF}i>Z6&n7S@Op?aqIPB3n=WZ_`y79Gb(Po<
zK$_2VOevRFzzNDE?GaEZRk_m-v)VhLT$=ze+pf#ZfR^zBcr7nw$|+#|l4Sp<>mU<>
z`*&J}M)Z&Q>&ESq$UKXHMrJm6X~-~!89}l!2dgmcrbedf-iWL&`0%dhDKC&pzX(&+
zpqBx;7Z8B1b)JE*e52?LI2ZgacEjrlbCu8O;!=MvfcdNGAsO<tz17&7{^@J|y1jp1
z2pm5!wY@bTbpOvE{ljJcdA!~6hNV0(K`68O=a2sC+WzfN!sjR60v~_{f2X&X{_4N}
zr)To>cRMGL`I-Z^693=d&A+&&*|V^wc$3NLv46cK|DXT5EjJ9P$+YS&{Zn24?|1S4
zt;f0nz!$WiXC#AdPqF`BtYmmW-vd5Olp1ZC$$M|$eE$uPT5tsrtO7Yi@#A`}|M5eA
zddK@>Bv|D)Oa%&QNo6>Rtds)5>N!xT3PaC^{K4Y*+ZFe#aQkC%C)Qqw1h@PKF!fcH
zvUe6)0jOu$3uH4wDf*Ejf3fWTi_`Jf%k9wX{XorD@wF~(l3Z~tcLV)Uj{l2K`8^7D
zk7p0|L;vUWi@kHOxDQh^Khph=-;1;QNE>Ujq-6g;elWAQ;YHqz$)x<#*yrEho&Rtj
z*j~fYboFDg{Lrsc^q=nWpLafS97#R0%JlXBhxYE@cft2}XrOSvHXvy5Pm}jQ|Jtv0
z^~?=Pe67u3xcNU1s{8*R?Emos{<^9E`xE+q^1<#|iG@(%AdHJ+=HGaJEz86kXAWTA
zylLYCe2j&cYa83b3KWOcr#`t~d?sfy5_dl!$*G>=?*OV!ZKU?g;~e<zBK<avm>*~~
z?jBmc@=$P{BU7{nbi8#1s2qG%(68!3bGJ1P0A+lP>RAf>n=mP+MESF+|9Za^aG!y-
zte~-fAv6LJ!SNp?_e2M-PDRwb_EtX<i!c}v&zZW_pDmiUP;>Mpw<0`s#?`67qO{N!
zuG9XeDU}7r-PSTgDA<pp$s}X{TiK{Y^b)Rg!VpWr+y7iqC6b=3rrGIfu*9kNjTJF;
zn#Huz*L2g{pfZZr-;|Skq-n1))D|;+(5Lv3MlnCY;HO2t{am{CsRK9y)&m|tTMLaX
zH7WECc1Cw8YoUg`g}9nefM1N+4m1(f@!tVLeg|Faa{wH<+qkFvRpwp+m7t+j=>bUj
z*Vq7NQ1#ON{pVfI#7VD}t*ly5*xE~4&`YNLvB9qHeaD@K4X7LYgE{?|i-Oh=RBr-c
ztS#$+4~n<HJh~SNu;vxXr0I9A*YVqllA`u57`-B_td#%w-2`mwCtltNSSj;$2Q1So
zC0v8ys{hjQ#o3OfOjc#RTNai6F1MZCE|FE1wkU=s94+R#&46qGpIF<)+5sLH!i$ox
zr9FcDn;+ZN4KL2&fi6L2PSm_aw$mtFi;(J%_`bi`RTGtw8|#b^tN;bk-mSpkCru~a
zu>v3VGa%-%Yp)Nu!pQ)A+#8gbN#o9~Hnw}*5cb;u^L*7I?X4E1^F#0&6%lVlW~L5J
z93K8-fN8uT(qw$QgQG_#z7}m1*H^mUl0*i+Czu@=z+;poW^qDSwh?{=V$#8~Mxa(L
zaR|UoJ-DS%fR~7WML;62!Vbt=nzbL8gI}($4|VvXuJPHf<OcBKBSe_i*X$7jJhd#@
zG7|7*_7wV(r?prlK=d*YV*I&hE69F@GT}VnEXF&)jhW`7;vR<#?0V=tv<X)LfjAFo
z9kxQ48(=j32@%fye8j(dy$3Ai!Ab;SE0^ZcX$lv%Seq952`F361Kup*003@1jOd5V
zZDgAk@Ena9rie14_8frv;`y8opgrT+cNN>Ozb?bNJA=%PjuGYxSLla3xL*=RRJ7C6
z42SJWb}hFK=y>g|!jkEP4u_7v$5`1f6zrW*=wJ3eOETHN0(QSkEAZ17kld8~!}*68
zH^&Ig?5(SBWYquj(JUaJIDq92rb@Q2@BBbaEA9)yB=)X$X$R1CYjTJmp(^@pP`n_G
zVkQ@1Ro{H)0t1sR3mgiT@^2_&u3o9+Fjv$&GHCQz<a$<NEyg9D-c_`Xl+&#DM>I~x
z8HWXQHt_$vfTqaloh^D}^B4g#QseA3W^M$10UOmIxm@N2i2JhK0Xo-WIJVo0C&hEw
z+oK%WAP&|A$S!?<x&cJlSEzb>lpPCVlOjk=XeSb%)4-f96py@TzlbHvgEu#u9LQAN
zx{&Pq^t~Du(x$%yWWRi68vY!PA5+0gO@&dY*lKX~%@Et`9q?gC2?B{}CCAFOb$=6&
zwdx(i*{c^W(_yKq>w2vK9{U1uzV4+olR=aiS{DzRJT|=e8>@ON%~J&%5i6psz1g&}
zDf(DJCNxm1FCO&a<Or2y@%r`m>$Ddw-Mra5zZ=wk&noOjOyE%sbI>m7<F<=6tFNrE
zUe&1r{~!~8Pv4Hq;;^=Dtz;OURBQIR?q4M9e?Q_B$e(Eu92odryZ8kV{Kb<Mi#1Lq
z26YmvoSbs<yf{6GN;EOzIuhkk`?M{_SA0^`XRZbry%0tsYl<yKN}Tn?C2<d7G;gRw
z{-~VrJsjbrzPn`dCuMxL=Zja;xDj@f`lCAR;^+l^SE+u)u)c<6|Kfq&T@QrK{45%a
zgEKa}o)l1COL|Ve;wml5JRs<>ybb46sW$%rbM|S}EM8JNZ0t4}-_vN`6bfFF(%k)w
zAGPe~K%F2Rw0y@zO>}zM>L@3k;xMP`kh*sD2o8`u{JnR%w3sXl55YoOGrzwX*rz8z
zjeDFQJji_1-Ge91t+-&ePq$M!o8;%hpowni;ixrXbdN3(SK2~)?+F+s)wuU*ORsD<
zP#m!xOyIkpLFUu2uFNjOWGt@(wvEjJbII#5@bGwx23v-EkxBXr?C(!-q6rUtFxkY|
zs;-T^b0`vg<R$t&^gT_C`6Q8HGawGC5U+-dy;+V5q<{(01SrH-%%lvAKKt%z;$Za*
z+{JXHM(X)yE@=2?KAKN^xMN<ncoeDXKOSIEWoS&Yo-*~pOnh%IR-k4woxf{tFVDfo
zIw4%@-u?!o&E%jAxo>w?vk>M<epSo810U}-@^y;V@CbWeW-ql-G2+(Lr84<e%Csq?
zo-9IEt^J<4rj;&vlhg&NBis<OLW2>Vvk@%vGKKv&Q<JB}0w#&wH(w^{#TWatd@onk
z^wk9~51FhWgAzf#k(Shp>uD(um}0`a3>q3AjrNK2j)9}zq@IwhQq)Yrp#qnTMpXfr
zZaMO`^qW1|{ggV?R?_lz>Vvk-4V$bq*+%XhrW+$np7G`*E$)kIp%gsUo$_{XxFQ8g
z1r)xs7Wf<Z%+ck#@rr5j;wJo%za%yt;kRFut#Ui)VVgU@{SyoNREmb?gQj)4{VBm|
zdRT<{ngT-ICcDp;So`mN$$gShb-R0;aa7$hr@25(Xj<0M$Mto;b3Lch`Dw=XHPCFW
zJL||E)_{~=)u0kxqvE5_)!l4aEE_sx^gV7y0FywXZ2q$rx#NA|<-=6&7YncTHl^4(
z0$xt##$U18l)B49hqXt4;&WFXLqN?jBT+W2<fR*Tw0i_g-a=4!D3tCA)(3#`c&4^j
z17*pBydEilCLpOOLbJ5T@4e@F-j8Wh?UkQi!MwYVs;%deK*6K!b}P5S`}g*Og>X+G
z_Uq-dNfImD?#J8i<WIn97!L*wHF(@yaL@-9!YEdydod&ICboM8N!O)NL}EN>nw9h5
zaul=pr30Uk09PR6l~dpGQQz>m(@(%Tx(?{mg6hfjd8!zi5#;FNjeCz6_$v*!I*mg`
zGOJXd)rL}Itz0Y^vzou6(mJ=d4iw#?w3vu%nnaI@HH@z(c(@_fN@gXi)6O#D6F+Im
zA|ID&g?4?KefAAF?SU6P=H4ogKYlg;^3|S<Cqpfp51MxnmgB=`4pks(x;A?yJ7ecZ
z?eWngz@aNr8UYt;&taI?ps^Mko;#&5umdV!Rz&Y3NTYD@^E{AwvoHX!*EqEn?<>Or
zB!*-K<|_^x%Puo~x(I;;$)LInVxv+#*@szXD}dB`ez=4ilA|OF8t=gBv0j>Q${&zm
z--Ttw7kCv(_Vr8+4i}vSth&JT%T&<O_JB-f1&|S4GUsi13(jUw%}(L)_*CnU-dqAy
zfgz#a39dHR!AvHVi`c5I_@03vl&mmx^NG~x+7YGDMS}Gvpwc)CXZ>0D26+i;d#@it
ze%Crar>X2*7T2}b52nOd_!tDx%Kge;d}+IW0QG1|FTD2+V;0z}1gm>84(+Lf8ogd{
zf3b%eD29&Ihz5wT-=N=nIJdEN^4f&|hY&FQa*g2sc#lXPd^G01xUTt662@Jfzt!h$
zGW@uhFfOLxB?I0FZ^sDdXJsC!Hp>&&jgiU^Y?sbMbtxCA`+}rOHvMeDNtzc+w!lBW
zIlKiTx^LUO^I5iT08toOGCvq##p&N<4-1FG-TC@?pL!r=UALmhd?>rm55<rdLY_o)
zZ!8XK6y;Dt*v;j&%<ji5nT5)`aubcRy6;L`%`g<U+-QB2N$b<dL0Bw!R*6rtwvA=|
zz_^5z{n&a)W<0M9C7;!11C`9mLf~iD-Hh0IXFKLFl_@wKvMgrOjou$cGkZ^_HA^4?
zHG%n{57u1P)Z5cAbxGWcBa?sIz)_)P7(<ubvDlukn7nufAAtykz!OPk>J-l_!nGfp
z!zwjJceI;3xp_fl|ABB?yJJknzWu(Z*{k=zv=E8=f+TFSm(sT0Ji+5-9kxrLRb71$
z356IRHimw>(?Q{&N6s#&CQovC;q7KsL6%GV*z)#wo-0CMB?sgSrzO(57M)03YxiC_
zj4@E;yLa9n$SWH)C-)h6jHMjGJFc7{AaH)Z{C>at+o~F)#Er^1c5ABC_3|Nh6B1Y1
zHt`$M#U^3YOB(dzykz~?e|vWRh7Orn!lM_Q`Sgx4pZ%&yC86jUPMsX_3u!!+>Ql`x
z31hh>k&&$==6qt2AIDJT3gi=8mz%_}$;#hH5oA(toxGj@nTrSFx*2sX9tCAXF0)Uy
zf(&r*9*-NbIxy}RG9s1>dY&J@0Rpv&;d&`2vlAZqkvSjFmdjxT5+8D`sD<{r0;2^3
z)~r~isus6w#-gHA>KmVjm77ElHad<$UH1u@Bo4M)Ls(^FDy5V?$nW%5zSjAilS0K@
z;c*1Y=;NOE1V~wRhe~HMt<lo(kT=;Za8Dc4V+f@3#^;NiL6Pqn(GOp7R=F2P%=!5A
zhJfU{bt*=uJ}(c{BQAQkFRRQ&Dp2Q}ED;ctJCBp)Vc0vN%#I)YeX=ptQQnqW<#<$X
z0Ctz%J#bf>cT{n+6S!UHz6M1YFP_Chv@z=ilrzurzU*@a6;e!^EcZ8qzOwp85^}ZO
z`xwNq);*K4Pd-Y{F1d<+nRg2&&QX!ZpQ2*uC`rH+wgAc>S{fZ0j`NC`$jJJ<jv<#w
z1&)00{rK>D)7Xa|v5n7Fwd_;!w)7~k)XK%M_3{|>n3Q0m-gN}#us;p3ploWydhiKe
zRt(Q`9Q6<&CG{?d==OdN!zWNssW)yCp$sc{Da#ayqmfllXpt@_`&ubOiGZ61wqEH1
zAXV?KHm*YFAh+m=-?0EKQXh2&;xF0EZh9{hFHRd_LF?7yv&M8Iy-)~(OG-a*3=Q)B
z@_b=K+4W@}x8Iw!%zRA0Nvu&4l~KyOdFhtSK;T+jXSaS!F)orIsH&>9<_0W#4K6d4
z5nk~yA%gt99T(8wudFv<-45f*dHe;-L1ne&3tLdhHa5+g48nxd_a?Jn;nJP#5^>EP
z_Vr;P*qmOnfjvZZq@=}ga-gML-=jt!JQ=nMvW}H83l*t*EhQ-&#GCMXg>Jh_H8c+U
z%s&RH+xEJ)s}IgI0kZu_I<me2&87V?i9*+QAlWUtM(v#nTL(YD<WjTELS&q}eVYb%
zOHg6D(@|$~B6Gj~W=Ei5YtV~f^YJ_mTZ$aB9QD~NsrjC6yyCXtT~MU`<U6!YjUpfB
z8p3PFq<BK3xW(=Z5iY3he%{igF_)E*(KbD5H7Cf~imK+}VjyJjLAai?Q&@`)cC{3}
zw~Phu^{bbr#s0_vKA%nP3>v1bE=-0L3A+w$g1IIOkzC=54{2Y$vCbjrETnkV1h*`*
zpJ3r0{p^w4Qnz4>v03h-XL+OGO)1;F=li}-$Ko=q?o*^B#M2U1Q|vLQ8fDK#1@C=O
z-&%HEcK$4UXn!mBc1Kro0B>fI4TsW}d93Qbo515PK8yfCXLaiQC&3z-#T#=X%P}fX
z%4%fqY{xj!<j#c7k8jE-bYw&2zETsnO8#;5&W|VUbfPJF_s6pqI;`B*V^T9phX**E
zb*huQ_^89)_}g&A1lC-TpDF+RgcR%AYxz6cwVOC;S3ixR4iI-xmcd)noEY^;18nzC
zi6^w<rSXo6v{`$Nq6~=K*>f$bITVKi1h&*I`Y#=16n0y0)boy7lYx=XkzQf{xr|e3
zuMWEBoIl_mX%HS9Lg8XJ5wB(F>#0>V4&;1nap_>XosgaTtc$iGvGy(7kk>LdDN{Y$
zkyU8x0Q=sFGiRwzR=kkle;d&T`qltE>R$W-8>cfF-Xy5x!%5=WS<2sTnF;xb(4jDq
zDMgu}5xQM-_A~C$q#JV}7>1VM9ch%ur>$q&VLYkkgTD$>`XK-dJDvx58xwKGYgXsv
z^uXV#$b-2L5mT0TbG49<U2nxkz1X2O-aVm#iJ;ROcu>9a`jOJEEFVs58}FNej7HPq
z4_Hoj37U;7%KM7dHp$|0M^G1~E=SoL6fxpVwiRdx(<b>fqr_K>SeKI2+EdVT>ubGW
zO1eLwE^(60)&I49MVE4Ae>p)j&rp@AIgM`0ys!nGCh^{7eUOf7$@yF<j<d`8emzF(
z*2uhg<50GmYEsr{dVJIMrQXzD?cN(YCPjj_q^`eF>gpD<Y^k+O0n2naWl8FFmMiLG
zNl%nY`4@cOjZSR}^0_Lr8NAo3BPt{onl99n_tAFbbnQh-wMh$sTf{g=z2`UYsLc#Q
zc@i7B*=&m%tMw@sc_=l5DA0!!MvYUKJ5t{)iH+D-oxV>_eI_KGXQ6jol8t6ny(G^^
z7b<H#Uh5%a9ysLfm<dnEc|*}y<0t&kt*H03O{Nv5d3d^oPMp^YWuGZ?<@838S+Q)<
zu+dgnzM}l&83&}$bc%kOi$Q%J@n=7(ix(uH!%54}${lbz|7IUzk*U9RHIH0TgZACl
z2mQv%S#{(3tF@<QaZk}rTh6s|kP8`qcU!$8tr-1W!&~_Vc6Di|b;bUL95EDG`2x~G
zt-jmC+f>>cI!jH4?l*taHeBld{!(i-kXd;AoD!rBdKTIrD{rc+@)S&)8dmnzXm^lo
zd{OM++vHdCh(@?5;~gz!GK?=2qUq(e{gaPw?PsU5PAgG(Qy2&<3*ZfK-?0#OWwVQ*
z`n`O-Z{L4V%_QUT1Bx)N1&t-MXo1`holgBEt=*<P(e2_^%}PRVx6h4|aLz5c*tne8
zpn)j~&qkl)EOA`+y^jnc@vXQ==1y$xJZ~aYrSA2g*?4rL#-+P~*k^1lR_lv|a^eZ8
z;h|9e`0o&%xnA_d0Mi`Q7YFQz+#lLOgi%sS$C@T85L4-sZ9Xen&LUP2ozD8um+5L#
z1jfndM!4wb)N&gd6Ej~?-BkqGWm7ow(KIxdy&#*~EOYNByQ@I!RHFB5rwkT$?I5KD
z3#HlJO__nW*?z2^&YCL`?GY{UEbciAj$FA~a<eC%aZ5c=2rQ(x!T1L?(|w5eD$RPA
zYVdVmm3<lAPR}{rpq_>+uGdqoy=9p-E;0Fd2w@`Z1Uhj0eJR7ca?P*YwTdZVi<nV9
z#U3%3Yy2SS)xb*!Q44YH-{!At2c;OLSoI7#X$D2!V669{#7A#rP3y<WEHUq|N|QLI
z+kJ7`Hd#zvOA#7=ts^5GoNcUpQUSe{-OPEPg1x$+k<sa56z#iSYxB(S6|EQ6`&w>p
zspQu>!2mzDRQAT0RDqXuD^Sx*mLhosin{Nis~2BR2ksXY%{3l8%%ptN3p&r(u<Yo5
z8!naXZQ6oFwN}oa?U|GT<B@>vJM&h`sgy=aO?-JtN7a4nH=MtSwJM5@gW*y^x@ixe
zQQ^0dg3wIjn{f7p-iRx8sJT=;Wjk0fplLy1?LT0iFr@ZD%-}J#0Mh-tMm=SEyB+#<
z890%v`<Hzh)|m9AUk(yjg45mhetkT*Y8u595SgUEC%7L`@g?7NXDR)G+0+s)#3>$`
zyQkQ`@zzMN9k*gQu6ayC2v>f7m+jr{wSkak^9^#wn9>@XfD(-W^YZ8+%arATP02#E
z@%JmE>-wS=>LF4G=&bn1^VN^T<!^2F1#sVVxW?WOTEI|=vsp7Z(LQJD#??H8_C~3q
zz>Y@)bx_<r5IXcjlUY2<aqd$QM?jCW4}f1KwD_(0*E;jgY%tbok4Yy=RL0Z05(=@N
z+co(uEy69jVg?DZjTlGGeUb-p0wHAw93i;L5d=<UC@UH`?8l{gH9Tr}>eZVE{1!Ik
zh4FB>*R+gVhkdo=Omvi%<$I3$rr^{_3A6?ZVnqs->uSba(b_y^%_4)7Pnev%3`Y*(
zzZ3&Cl#hVRv-A5CSub^I2cm}(8z~E_S_z~L#u;WISdt(czCS!NFxC2XjOd`)MxcX8
zYCvUFWE4p5LzwcB0uk!2mXAR<M(3l*NeK^gYxYcQC3g>(-4kL|v*KZw(Y1dv$uEGV
z7}sBL5}e6^ecloGp_ncYw)+(3y104D<-6sA44v)qHHmoxoNIUPhlho^b)6<&jl2?r
zSeZ8%V!L0tC&j^FZKUY@@(AT&*vx70?fUn^u#^_Mh1NNGcKuNWm)`zKZ)V=pc4LTy
zz>X|8XzwLXIJFk&Xq@$eG2z7@jxZ;0&%N<ZiMA0@RorO$4p?N=Y^!_qwfoK{MHaV%
zU)>y5$+F=4uEy%t`$S6N03hc+Y^{o8)YZ`hA27uAwJz@Mt@a&{o4&kx9f2(;dWYD-
zyYT2TYo&RSMYHoYS?b=n5s(TUV(X3BRU!*~@z(;64lnsQr;+7k6~UNHI#*@q@ii|>
zu#olDtM1J@#9r!|9Bn5U`nfLNke=o~v!QbG%B_V_|9Dpkx;_HAWd~cKh&aL|+JI5D
zK!`WZiZ<CGU&xAdHOr}uxOb~^jz8zghu1ZtwbkWax9+rO+k>Wa@0OrzWF+mpl@-gp
zG^WG9yD*4OHEOwNp;dJJtZ41NO0})8LatKFs!DrrX==u+KJ6PjS%JiFPSHGbyRp;b
zrRn*uiuh_~g%g%Lbl!_=^%8TRd|zmNH{GMEE($G<+3E39vr7WHIEPB8C8a^T9Fa9N
zAW>2a45$No3a2|s4?j6!w$dVW<~uE+TbI%R-<C2e!R+mUxHH#MF<W0HJX(OX)bIHH
z9BQgH7c{l}hHoCx``w)U!h$F1b=vSJ$|)$P1D`4hWVZ;*$Mg`QI?NenB_KEO3Er$!
z>CPH38`yM;XATviKo#{a=<mu5n8+SpV;9RwNL2_|X#&wm_Sm<pxU_W%ZkuIj!>FdT
zLuTBo*zVT6Q%R3Zjthpvx_i~$R`f6U!Kv<=#msPgMYDI|isLo<)_}|&4I=L;YhcT|
z<Hr!v%;DYfu`(<6E|b?z%*06iMuZ`8d}&f#5&D}U;(V>&sNG{L`6zkMC#lQfcQ@cx
z_t}++&EKFI%uLLPKa!f-Ov-m}6K9mh8RGa3GZ4E&KO1ZRZF1DhObW#jGm*Ao3AdJ?
zJFBf*lje-i<@AGs3mNqD)NEnz)RG1%Y0dnlnes)te!Mumy0q_izVqjfs$oClwwa~v
zxc`xWr^LxkKX6onr)7R0oZBoEIbv9G0!FBJ7_&&{_@BUn>lY^zbv4w!ay4ba^QRld
zVceaAM<NuGvuXLkTA$7QS=0i*%1<d`*7dGvL}<$-6Wt(VbSt9r<AtGAmbth)^MYhn
zzqRR-(r}B_p2M(ZuTM(%#{vbCV)@nr41a1sj3Ifs;lkVg)gBXi0Sj^BZF+&QJXQ2$
zeh+_-NfSp;t8J`01^aS7%eOp^!lOCc?u9x2X715x0Wu*g428<4ou}D9=yUjq*Ym2x
z<nmqSTx^@EQgZ^yH?n05FVeNE#;#qD-z2XLdsuwS@<x3~!9sGEeVAa)JgaNGAxV>c
z(xbVpkaH1Zmfy%3KeUpIH)p2FzHLq7_@<`;BY;Q><`{qj+-@z*t%-ckug##XwmksB
z&%Wy>LaekS?&&EA;#lCZJR2hDHne#$u&d?1h1B8OF{`ZO?px6iz9G<cmg(?Q@t&Nf
z-L1~O(Lq*Vw8ecW$=ph;5pLCR)h%(QGxyh&KX9mWAn)f@<;}yT=twaq?;d%DJ6*7A
z@>W}-`IV!eqh;~9MG>jVsrKyV1*7bU4LjN&ew$ZYHEd&TcR0K?Rw-|jP<B1GX&*+d
z2Ww?ke=E1e&{dW6uvFI4ku~}31~ywf7v1B|ugW(!A)4dp+0`dIxp^T{=GmgioShLX
zo~P_XB6IT1FI9tyl#!yPD=?OO0cc;I^ko8fN0-8}P6O=?Bz?sbv*<ahBa){>Ci;#Z
zzd;BD{5q_;9%#1}HhU$a#qUgmeyuvJTJ_LtFX=K|JTBGKtI;t`Xn?ucWhB2~yO@BJ
z2|}pqEa^O5=Vn(Q@;jF=H<#v}Pl{lra6eep5iZjq+RHrXxhgfFg$v@K34BC<7-LKo
z$@yMKNg&*wwD9}|L|meydQ`<f_q@p(28`#nYy<{%ZdPZtR`^hQqF>{15x!aff%@$a
z#Q*7N=;;_I=17T5nrA))w5JGGv18;o5H~;02GcYgH;>%H3YxUI5&-$8byz(;(fZQM
znc2wYOr$3V;Br06-3cr2p5Nq;A&y~SJ4+$RKX^$LIm9ZsmO*0XCUEnOf@Ng(42aGt
zWaz>-EgeTvGjM1~@epgi_96}}Xn}bez`u@AW-#Jve~kqEd;;GS=cpdqv52cQ3ux8K
z`xW-d1gG%HNf)mxvEdol3BFxvF!4{ZKLVT6vLC$nMZz98zPvaVz}cTQ%>p-2r{~VT
z`lB~qM45!W>pJ1(9ymv_CTFu(zd|}DO6Dm^wJ8fHc?}Ti8M0wg(EdBuyPoHDb<f4>
z1h1^Fmd%U08RmDlx`2b??&Ur#@w(KUv(n7P%B2~%p`yzNl<QMZsfm2D+Pzk%J;ql9
zeXR3GiRMPPckfylK(}<))k{|PN}I6$JNg6<fwSw*=fm2wi%$_86sH8zC1Z+*O9~&@
z<)t`d70O&~ld2d`&LA7|goP5+0A%!wiw)W4973hNfiP?@X+?j1E&t04pvI8@3T#sW
zT|c*}Ip>z7qNOIOcXK9HwTf;%xo|edh4){j7e~9+y1plpX(W+h`o8^rh~K0#?xAl%
zOVH<S5PzdKn3Be7*vU%!HdRMN7=P|v#VtWQNawSMdty}H>`QWfX1t_3psRJM^4t55
zMhAw(Mec61z>rt#U7$^czUD(kfshh>+j?+Q_%X$jl3}#u^usc|(?fPfqkMTj_^G3@
ze<piO><{eFR9FA$kvSoOhwlPMAbdQ>tSa58kJsLIi~Z?%gb6yZ=dnG3lY`^`V(hJ>
zs*JY%Z$%Igl~O{wQ5q>pX{1r<Zs|@5l@94n3F%JR#HM4@-3`(qY@|2u+UMSL?z!W=
z_m1CR4hIbOex7GNYt8wcbAG;D2zI}F!oq|ii{mprranhG*EgJtU<!nQi#&2DoT!&l
ze8kE2wd>0HpE(|^V?>(d`MR0S;?W=Dv|l$su77s;Ydzu@@qob>Xqa;PoHg1i7@j?I
zHDU|CO^%$CY058?sCvG5UM%!8eaV8cnc95nL=ZX$1+v-obp#*%bR%kWv5NM%P|jS+
zlE+L?A>vT_{9H3%a2aUzOFZ$3z?dWeRDQXWI3D`)0Dm)&7Z7vX&f5(Jkw0~g;6-`(
z`XY)%^_-gLd9H}>h!YZQ%$CIQ&EVS)1GU~}ALqN`Q}H9;T?=DN-WOP#&3C<n5)t`_
z&0Qd@*8lsNTF7p^Is>td^qsTw<EhQf)UJgon|X7Mmrd9nJjnSWET%Ycw*PKesakPs
z22=)A|ET#c=w^=g3aQEeN_P6_&~HL~ikCPTN7HTNI%~OXS50;yR8~gjSyph~0xB;y
z$`r}25c)Z5l!jYVex01XgOW(8)0n5_RuOZJhqrJ+`DtpQQx#|fsBZl9yMnammqT??
zp1E)6`ab$rx3LHlIQTku1Qh(A@#h$xwq=kjAmtwTyf~fR_Y+q9cs+93pKVn+C;57P
z!1cWGuS2;NX0B*q4+zm~P6h31NYCPG)sm{ZlLvLe;>+EK4W%-@1aloMmIPkZDxJ;F
z{?UXCt9!n{&s*Q~swb8xDm$0dN>bCPsa7b^;ltp&-^h24X`cM9q?%<QY>^%EO?_N<
zvjG6Uc8bM67pjHW5>jM#D}RXF35%dshj&5R`?E7nY}w3<vilvo0k$InN%Bxd$oAW+
z72nhtl6}{%i!y!Im^FAiJNV6)%#2q`cdfoZXR@5?<m6}hLZ^HJaTTjpeFaJ2ei`e0
z$_eZtfvp~-Ur`X*s`uCcg~wiD|EpC!>b}`hxzOL0qWOxeb`wx<Q3NkcCad9IWzwXg
zm7uB4aCLy?1ndJTHga@+HrLZjd+K)I&@+3_hQ%Cv^wzVaJSQ}*tOsJoYqVNk;`;p5
z1e-_@+oE6n#e;3R51)_8{3dIyH?TJNOzlf8=+;HK14{W!S#J6sWieZGT?eRvNXf(6
zg&69CB#e0PT7GZaZ*!#%>#ve$?8mHfX2i0wd~k-0mhz^H(5gMJinWaIk|NC#k={jh
z)N+g|)O5pru$;LMjm$UkVEo=@{#E^ztv9je#{SpfWu?N>=L)4ZY%I6^PCr!xt!j?g
z_|0-`1ePAmc8s&?r{zUvwU{8`&HH}UAoB;AXHLw>of=?vYkK?IX(Iz9i-ZT0?fU;7
z8pVHUW|ote*sAW^URuy=zpnHCm_E~^DNJ9LSQUc#*YRkcTW?H5BcX)5<GXjw*3^=M
zazbxPeaVlV{Lgb*>mJwaL*{*>kMBLQW1D_8e-Z`-4G^U^1qbUq^(Ul}Kg|zbh|Hsh
zx=@?~rp(C}?3JI{BJMoc^Ef%Of|jjb2{p?x&Rgghh=q3QC4J6t5h-&lPz22oXqly=
z;qeTMr<Q-I@Y^>da9#9CiD&vRH-OLR$<{qEs+d@hFEq{jCq41qwToT)7IHRUmS0xM
zaF9`4ZzMFcyHr*`4b%$EhKuL2nZNEKl*hMWc=dC6hu&QHG2Lf>RZQu=|H}T1axK#q
z$;9gE>jWBW=;=HETwByu0xl`H2e1e=j)Npvo>t|S3+|=^L}@kdpiFMlT#%J;>X(^f
z#DRFK#8&WJ0E~`l<80w`uUJAv@pZgediSj?4Z@jXT3?rj00Crkv-$YB(2U~3=3SkU
zH!55t@gd^g&|R4W>G3qTtrrbFp%&au5wC1DnT>_V`psa`5z7Of#oo{%szteCMPmwU
z3LLdF-?@Cm53(bJc}<GJ<B$4zl>sL|_cE`vzO%uZWv6CC(@Lm|5G0ot)+&3KN~4u2
zf9y8X|GqtNJ0s=N8+W;#)=&|#z~ibLVJ3yJ?s`>X&pQ?WjCU|=yk3-IS^w9r(C~cn
zi#hT7fLVDLEYHkEC9`C9z&Sns%*4jrcenZ==a!W&%-e?)%gnrtU0g@FKYZ-NXLSmV
z4K2gbAs&@D{Tp=h``%d-3~7o}83Z>LBY7mX*@QAEIf<_zZ4-SNTWjL`B`a)fB{wIi
zzAprK#0p8~wT4RcF1|~-YBg#(Xe~}}ugGBgLmGFGo^s7jL-?BAK$v=~ytf)dzA}^O
zBAl*MZ2Aw<%vVJk<?<D!O&+Tv#J$=n`1bDfL6|l4;ZBz6)GXnbrE)?A?;?cG^qiL8
zZ|jur)`<>6!{p^Eqe0zRFPOHkREHmgo1q!Zdi}Q9Ys>9c^`tC7=he)=nwaS*5q3|P
z`S@WY1^4D<Dae%vVkK0JtO%05!?Rv`amSaVN`Hq)k;}+#6a*1_{!%G!nLuy}MJ1V|
z`I>bN4wF>RO&?syoW%$!5_^1#Q&Dlho*V1WPL%M(%+5^H)cX6UMsK0y*IFo|s)F%*
zi7(9=?z-%V3BjK=vyOhCJIoRWvG%gyT^GLnv!3LNiTFpZWD|c15Ws&DJr))EO9=x!
zx6`%;@Pg{)s?z`=9;W17ohMtciHt+WEdL`xw<)>qx7Po$-g|`9t9`WB<^-ofV~|4i
zsAJsd`FY)n(qo{wrkL7cAv}wRKc*fq8Rs~AS0HCfe~92IXL@$xO8>f$Xs_A7ZPW~Q
ztxS@52MOxCR#uwUsD8RX{DC{Km;ZbIz_+Bx(gBF0tJ`6nX0_w0;o%TD+}e4-h(IXq
zXIZodTiMxYBC7iV&W$9q!%hCRO!KFSAW_YH&k=zKm>;UnQ)Vk(^vD)aj{cZyZ9}Kf
zxwKc>jK>%$@i)!cw4J^dd(%QVT1{Rpv1PH-qK5K5UXH>qIY97yx*9V?Wzf-Zl4G)N
zFePd_=%l16FTMK+a-L+CVxfaNyXY=V#;bTgADh~$**w^RcIZP3H+|%0$hGq@-f$$f
z-!qPq1hc-;@v&z4YQ=WiO$LO?aW)hph3Z=wUOFL7d?34(j+$wa@0xopTYI_<_flv#
zB!(CCaL?&){?t5r^J2ZOoii_5H$f!=EG^smn4qO5X)9@OE|{WsmBa)m$`6F9lmq;b
ze;$47XK{;`P0MJeX6~R|nKM*s*&_cnd(`#f1+!ENW@=XJcf6moy}i*ABbHE)vT*Q5
zUM)&p4_IHO>{!h?WYi<IxfipxW@(z;wHh9Alpl?=keXUFvwHi~DeDhhW4gf#WwW_a
z`d0xarmgtfs%@DMyZ_|Js_IX59qe;zb;v9$|Lh=jnNTS^RJb;st|mY@=ajIy&~kn3
ziFFMvZfjIfSmE{XR3;mj3ko9?UCiA&h%{UmF8p#fnN_!0Ga-%|znphfc+unTP-qX!
zNsd>7v(u$66O~Px4w&m$+UF9=ItZ|aMHr7q(!G~7xAS_Y=P~u5@S;m+>Lh9`f|;ve
zM3Rf?qcYJ?7Ti2?`Wa_*NzFE78*u9vYMg6x7CO)`yHKN#`xxNLHOJ*12<5&{tNE~8
zihj@2fl?(_LPV3HEwU^hFE@YPO4qwi!q~I5$VZ@ABvr9`9vtW{)|)H{F8gUhYX^a5
zW>`L~a4)f&)?-?<YrI1=)%9fqU^4A`2ETVM(L&(egw*<YV&v?Vp5#~+B@`)zh<!Zj
zA$5|seP0P&>(thAw+Ry)aPPJ>r5a`P+4k<gVA^3C>uty=pEuT9=hVWATX}iaWqRfA
zpvy)+Q8#G)`7lXhE1a3Xcqv|_Qq^;G=jRt@&h3S0w5F!S$B)WAIbNT>Tk0rRJH<JD
z-~C>V*c$c;kTD81-@Rvxhp$O(HRwqYH1zVbl9XcqEN0s_PqD~baFX9CO5dYX#<i*?
znT!|CRk<jM&FtQjG`_RGkg7M`5xX4zThCv1!i<|nUh*=M=lfe*auOjky^DVoZv<L4
zY6`_4?a5JGZl~~=^hPxqU5HZ%C8;%joN4+@>kwa|nP8Wr%$=Sz)1&{|8#)&HXT55)
zj!bjgN;Td5YWJg!#DS(rDX>jEWA%E5ea7HX7Jr%(?pQN-9+t@*RCK%N$8Teveq(K!
zRrmd+bCX&>?kVZKcv|V(^S)6wR^u607HtcBm~pYUS(6#XWWRBa>mNE@q>e2l`Icg3
z=jzjkYy$H4=iWR&<AeI~Z+1MYI*qGPG&bPb`s<#}_Jg61PS@b-ZJA|#km2#55RdaG
zulDS#s~E7H4jgy>WTK`@D!P<^z6QHt17q;D6=^Z&JCppxlhR+F`i#;@OwG@91>%DP
zb|%uv)0_PPX%*)j@7gp4<Acp&Q+8?+5v^IKs)+O_<%E@x`swQ*g880f9vYGM?}wE?
zQb44q+*^qABU5Pw<G*~ET8Fnopb((R4o-nBVHaKG;;VLv{OiQc_d-f?nbkbEh4ome
zEpNzb?1hBQtvabKcq%=IM~TV@>d3~Kb4WQ#%@RvDsOq9Ug~_dXd?o+PsZdDvz)q)~
zEWB%qBMV(33Kf*n_&C@;Wo33_e7}KQk5nJORoMR|dCUv;hXR+8o&`%94@Cm^vp+H!
zOvY)mWj^(z!mqVk%&t;g)3!<zXtc|u>Qaf#yM5{=IrQ|O-EhDDZ7ZGnDfcPn{4A=c
z!f}48MRMhYB(0tdeC?&Jc_JR}cHk?Pi*YOjkz(E?H($uShg|c@JEfiCqDgf%&zY>C
zV+ihe|FdS)gg!ORg>i*=+X1+dqD2ef@x>Ej1A>e#p52g3PX4XuWj6Y{MLwYtp7^Ib
z(6V{BQ@GEckT6PGX8Wur-3`B>{K<3uZ;u_KeiAj~FpuwS7T8X>q+Q^XxQIDu_jCE}
z|2Z>w$D8HJm|YW>2P^!kr_)vbxY}%l+a<YuFW?J|0e?(Z+DJS^?wLLm@Vo|3X`;p4
zK46)gb|Ys^zbexGKyWPj_7BwN$%lB?Uk&Zoz8n<$Mkn-E52-lv=fWr4_I@j3hq~mH
z1a;3^>>9s-<XMfBm)E*9f6zJGyNc1C9%B>+xW>3&h1SR>5RcJXt#8G>^4LOL3RVjd
z{h&GOrMhx9SEt6bbNq*C#h-+lmM_QG2z8yK!`?iT!fu;i1$2Nj0imYR+K(S#yPIG9
zH9QToNSrR^Mv}Ii7r$?UrjbO|MDS8He%2VbY?R$PIGff#{}cNQ)?S{MxjVG8YH`)1
z++ys(H59fvQdOX*DivbIMKCc@3rJl(+@aOrreC8oywDFF*EM@7KK?jgQW`*xPRmVA
zg%QAS)LA8FEO;w~oVBQQhlt;;x-4AKWZc?SKvB1Zqu#bIC$<YMULOhuL|=NUv-B5W
z(=p0n@wXh3bn<&g^o4#Srw-21!}ys#q_nBYmL8`RGN1=?q9Vw9Ez8$0?23gacp%le
z(ZW(fd!Cp-97^=scQ5$!d_X6wENk6M#feA&6(m1fE7A2fBL?)pIu_AHSRq1<?F&4H
zKZC?5o3OG<;*}nabOduJlSBFy*W%uekGR9S$E&Q&T`dBX=Td9=HB322Y&oEw07Let
zuxRX(f1EZ6sXOax&SM{U5uY~J)ehCE-U=E6l?2)R*%tk+f~(11F}Crl^C^z-qypWV
zy_-)zNkf?!+9aBWIun;ts@TBx>^41im#KAnH&_=^btV6EoUmJJ5TBMFxolcQ->UVv
zC}r^Ks`3Y^ti(OPu}J$_>EHbkLQ#GeJM+%C2HA<blj}5nUA&?+%mj}PNJ4;Sgi^uJ
zQ$+8r;o;(jou#_05gzbL6*pU;y}07$l@&`W4x7NKFTh9u+oSjOzRhfL+yo!AM?Z(w
zj3-GWt|(dm0oG#uCi2})6&o^LHpyD6Wc1NnS-pH5{Ua<074?iy$Sx}D8#M!hKN9}u
zoe--cmAci<k}kieYt|M<Y3}GEho_Nm^*5;5UQW0x-8wBKXu>-t?FzM-=3{EiCSu!J
zq=@j&Q&=U^N{TN=p?=AUtDC9OhG~%#hI=}_UllJXK?MvF5h*tFgH^IUA5!C8goU){
zKjnPCS*n$}$%r7Fo+Nzbl>;S$F7~UT`oiXsi9@*9ZL07WX2%!OyZlKmrr%%BPag5i
znVa=b%j=Iv_|#82n)FZhx&ncLv#bkd47Q&_))!<>^*IV$?}Pjgtp70z)8Bd#5c6>8
z3<2-*b*`FGw`(F|6}U}pX&ntI42yfXU-guV3HMPsp}1#nR-z3@ON-dG?|wS^FIvg7
zzt}%~3FuzD71eM|_<@Qk7eh%*AYu3wtuNT*;@&%y7)=9J$A_ZPjLg3<m^Htl-lxN4
zqLfMaf%5Q506O-<(i38y*Jw4?oS(6^bXa&;#($<ZSqPtcsZ?W(+kJh_6y|49-Lu~U
zlVu8{(nKI%tar}p7=(<VY7ecV?3IvU?x{2s-Zz`Z@pE2a9%IM+>hg#kZ5~X$z!YSU
zQ=Lr?zR?`wV96hjd)!k(XtWxC&184S#v&V}e>3JdYbV}{!;Turi4wIYPL?vYT4Oax
zkDkoeHAd2t@WWiXh+AZC2JlZtYq4&I`1R;JpHcH(rTm(~3`9sjrMDgZvmbTpSQ#c<
zS3H(MgC$XFM&vo`u>~B5Vb3#NlmB8dUq0Li-Tac$kagbWaVX#48Dl3F%#DXjzaJ=F
zth+qivF0(57KnYLk)my$?n0VHT~4;St&_QAHVstmUxgAJFjb=axsq&30s{HT{^v*a
zkQlb8uqsj$app4Ihdnx9*uoy)<asFpFhcZ*7im++hVnLLwhG$~FPwLWV)VQ-5+V-e
z+}{dZ?unv$T)r<v%fDf2$mSe1$*8~58<@mQ$Rv`a>^~AtisQ3Q6T<*k5=tPYLQ7_K
zW|pF&IJyw=2|7#eYJm&j2pjY7k~QXspkEfIcCKFkg6B}Z@gQz+6*G^?BibyIo$Yoc
z+I+}~<394c;!SYQD3X*CglYUJLgmrBnw^__<6vZ@nn&*&JLb!z%9x&b<1r0Jxke(0
zh}7&~tQKp8QAr3lS8s$yO0KS{LoCNjlJb;>sxv=!<6J*`*sc#u=;}4|wFW2V8TD0H
zeL138aWb-M=iUQSR0}ra>OEk%cDszt4~JSO2N%A+&a#`Ja%a)_$Fqe5-jg04leXNP
zY3eevNjdHj6hwgHUQ=rQ&*9{A^f!<2?x{^_H;B^RM6=HEm&jZrL{RSo8ZkvTS`<_2
znW(nm_-@0|orBG$;+xH9iN1qZ-v%10D_(SPWR4F>T@(3>ba%mzPGH8{*k2>6F#i;h
zN#7+nW-h^F&snf4d&dVx%V`!|NeyvD^7)^*sV5w(^N1M56@!_v@lA^S2P+l76n)-5
zWDJ8>JA@x4TUkxVQ$^%<*w#l}9T_Fsw11C&%BDYS^_<vK;etuKuHPsg@id#}w4>&0
zjqkvDmiD~GIiGhx1A^Mu`ipt^;uH1w<LM{FjkwY(&3l->9wZIXZ;ochHEstKcNVkx
zU#E+hIx8^1Z6}>`$5l7nl6}m=>xhW2qmR||h8IIPRcNsf7PG#$KBH9<C=vqAH*#?U
zYK(_KW>Ix^nDove2$LY(dE{N5ZZ{f`3$b&)#|GKcoACU>h_5>#8?`M~f^0+0e)Mrd
zE1L2Cpttx!@0Sl@6xmY|3(a@=SgKVgFsAFw+Ty2=b!ND+n^okVWQUnNA!6is>paTo
zrCL+@T-DOW$I@HLnL2x@92N=qvT_se(P>BW+Wd(&nhUUkvrXOwr7KDWyH-gss36_F
zKFN?6Zr^)X6U#D}<t+MY_N}u+|7sbQ1B#Djskrl3-3k8a{>-LdgUjGGfBbbd@E!!8
zER=Q7QI;N$j@eYG=rpJK!=6}nCxZ$~B%`QUoD7Yn!dqwDmczvoikcA1cT#bmXk=?}
z==d6VVn5$Fk1lx!LSvSyz7=e!*9kGa*9|8SQdFIQ*?u5)rk05Fr{eEhfC~xH_XXtR
zGje*Im;|BUV(Av((q#w7=%o1QZ&G;^&B@gsDiek)mSCLJJIeCm(goLf4xT0c2C;I&
zkt&P2FbXVJ>xHw33wjY_bdLzEDJGx4a;Zwcqh}>MH*CQEIM*&Wme3JkfgX7j5o|~h
zE-X;oa?ouD+aB95Glj@GkCkMFD~rVQ?Ol~wPR<IcS0ysxPgJXEbVpI~ezg%EL6aC=
zw8~bakqmKl(H||(50B>6WOemnOrG3fT9jmiyrq@Jaq^W_BdDDkr9`>-(6giDnK*RL
ztQ(5Z0<p&}!)V#IHC<S!Z31qs{nk}oZ&_GHSySOzqt@&}+z>4X*tO!T1~26LtjiF}
zfg?h)yN)jMAyL_>wL01pPmjajIJZsIa&H^ZVe#wA=$cwb|8z6nMi7~%-YgmTJwd%K
zPEE7yJ2R$%%Qi*`vS^852uSgZ5r^&iY)sMY?RIxVl72f^YYh~j4PI#WeXU(iFQJfl
z&qb!k{`)ymN1fxg&lP&Qvnfo1!P$)MS}MJB8h}q8-xvQug0UU)%C;r>kL&B8A#-6>
zmV>aEDa&%CIB%<?l&Q)4Y~m($kv*uGp`7b;*LNG@d414ytFXKW7Z>7?nn3wgK*u_b
zZp|xvY(y%3_ss##Tcz6)&O~gyE%kbIPy+9-<lK|S$<Hl!J*q4{yxzKmDU!TxziKfS
z7h*115daYKycV8$*%X}5x*KY%ha)6hzhe&9B_1D(G$-p4m_m&!V6a?Og1dh@Vu2J&
z_A+(HFR9R<nb7c_Z``Kn_F=7!f2SJMa~ht2G-`*9JTZ0zb^aPJisWSI=o7DoBhElt
zGRO|?IoaJ87e;mi@~@Rk_O6gi**JNo5)lrBO^<3{B++K5i`>+e^M`WA`*S<WuxLDc
zQ{a9chm7n7F*)v9lm3EUL2CxhvZryQ>V_)^Qz~}~&zZf6cC0j;Q=jI;j$U!H>-Xxs
zaq`-^EPcgUduqdQl9TRUy>uUaQD=Lqdvpt8xLo@=yz6vMxK^IYl8&9}SDw$nR}S8B
zi=MOtFmNeZ+T_Y#)bbQsu%wCyO-(gG2-w-4gC1yC8w1Z<S~ZdC51#b}L-OX<#Jb}l
zo_gG5JI>E9b@MWbb6BLLrk*DY<fQ+!C3JyzXhe>GFhsUkCTDKNACx5a8;J6yGvaHW
zBrmf|zD2M54IVo3c>m9T0x?iD1mv{2$f@5|n-HM{exYKg#XTm_t2j0v`mf)5{FMtu
z1JNbxo)PbQ*Ag?8hE`M)AYvDC00vXe%cUBg76pjV)lHW;{ro!egz(V+0bLOt<i}Qt
zQT}qspZW8<irg7y{b+L_O1mYrrFIpEwLY*|jcmYLRGvyuEi~I1_6+29)ID*^e`qSC
zblp|cbOO>c$gg<zSjG@`GSzwLOOm*+qHdBb3eiiiE^&)>48Q8}nl%`ES2mS}o%I1L
z_N|ilhoClm$Y+0|5mIMku0kxQveO!-<x%v_l;8FayewWCrsb+8M27?ju>lMG!ko1C
z9WOD{9NR63G&cIJGDjRF5y)^kK*bf_QWOAPEpjW?xComa-uGW(DSADQRuf*f<G)mz
z65$L9y<r_64xc1Q4S-U=F+F;!j#{y6@b9<lzPLd!rGY<`DkeJNF}U4CJQx>f`ZPCB
zQkm=hUm)+Y0R|CD0FLsWz#&oJaly6oJ!a)%$vXLV3V*8F{jCTz+bx-9S4=5C*^Bky
z1tfI+iQ%GM?fb0?;GOk)@5cD{_*NvM+B8cL%jpI<L~dXZS=v-uOqdcFouXK~iO#Ye
z7k4#7ejvNJgVMF?i%&)1BjWf%lLbRZVD4OzQXW@|yt30*InF<efMOB1)iRz<#Y#dr
zX~Als%R?x=)$m?|rbjB1#PL16hz#ebZHpI&1%qlAsF{BkZR#<7D`D|kQ}PAcdK3jN
zR6(T~{%g)fBEJ)j8RuP0gK2gu=1X%BjY+2SLBup<!X;JHuQ-WT%W++LOOWORc(v_2
z5m$?#l!jN^I+uX3eq8D%tcu~=p?Rc6bu&8DwQKGwc0@3lyfdaPHSH{Rq^2OvuPf=3
zxMSXJ878!P1gsDlsT~h=Ymc|COzOdC%phA+)>84N+4S<$+tWj^<Tb#O_ux#qI~H#R
zf0u2TrRzWMIdXxc;*R9VF47(Y-fpxtWGC+s0p_BVj(N?&GXXyz@x@>-^v^IGd$b37
zPEKSL+)fA5{_(*6^KVcdgK`ZWfVF#P_o%#)`(Uw#efzNYOH>b~5-`qCe6sOUC}JQw
z6V5lLQ!M14(^E+2DPea4I@P%Gz1IHv;vjhDKIP2W9c3o4z5eki!%%QxZ}eLkTL4-X
zyt?Gm3~hdoJv~6e8a_~L!hQ-u;DnA7bCGLJHJtcc!PTP=nOAYKl99wtCvWA~v1cjq
zccT0Hz^aHv+7mo+2(&Q$;{Xu%7BV77Tuz2Whzzt%Fxp#BRpQK-L23`WHztKn{6&Ob
zGz=}{BgB#`aJT_DoX+#6RL)q9BI%r}Xg9jQ57fBc6PguM7)XmWPIkXoGoY$Vrc+oN
zHD-+k_k^>0!M{oeaNbz;ew)D{z%i(SKkUY{#I2`9AtAoJ*G&@BSM`)etzpbKOIP?k
z>qy(SbQ-zG4!q;N9LpvQbz@9WTB{wJef8>QYPnupIy%J8D8VLkh!d6Q&Sth)2?{0X
zigCZKN5U=pwYyiAhV}bl(?f2Zqain+7!F;RzkUiw>&K;$`sL^gz~5V1T{gS}iY;5G
zgZflAqA};t+gdW>z~Ut|nycs<UeI)r-og5v@CYd+M?CBDo&yU^ZL(1q+F^y8{tnV^
z*T7a}b&-dE)->-A#R=lP9ni4(gEY0C)8XMUwaNZ*Bm;e8_hi2?Kf`eu={@ganhGC<
z=I$#BolP{D3h2_q3`Pq%qsxN+e!idvul;K3d4l7A?XpDF7!{OAx&b|yANx*{#*R5C
z<coCwiUq7j^WfZx7(?XBYaJ2&5@15$@rEZzgnnXl{=x;8GB6@K{NCt4)};Z)V>VQz
zsdF$L;7OV|0%xyANfv^7%9ONn=`*7@HsIuQ?imql_X~ecp4B6UBn2N`kur;kWf&O_
zuthb42BSV;H@Veo4wu``&m}W7F!igr1SafVF)BGylRlsC`aJzDroQR$H*rGWu_bJJ
z>F8h!Ul`QO+_pZzp12f`8;E&u^0ZUt_+s(MN4?`dGt$r1AFWNv_i8-C`W$>-B5w62
zZ^kjYrY)dHx=h9OiUhj?K%#V2`y-}bP2YaK)%!`d)4sg^MKaH(ezETF{PmXMO2<0U
zr@JyJ5=nqbFmPk%`(*1m81$;`%)ig6E`xLC4+ST2tyC1XForL4fmjO`%OPMI>y}{)
z&RbX0f!5J7rtu9lU+V~}NEwo4FqOHl2d^swQ$jzONPr)O%;9xz7-ttp0uJiCat7DS
z$O^9j5tj#h;!(tf*qTr8!0cjAJVyoN9swBfWDr1ojEDeJ6;+~VH;vT(P>oC6932&r
zk>@K=z~G&>_or<ACW44Uoaj9ER;`Pm{=m#!JvMFAuK$>M^n+gJXm_;lgv!xYdDKV#
zRsLj(^e95k4UmZ)%yrasO(}M&)5~9R><G6pR}ZZ9stJoU@QlwIHg%{RjhT(AAKP4k
zKHd$C3qz8v<t$%LU}f=UV4NVwBi~Xawy6M|NZayj<waKfn!}CnF2W{<!$%2X*{q*l
zwiC9jFpI@X``nyNY1`3(F?_?S?iW_Hz$$~coQDtNYt1au#T4ao_?)e>y9p2;8(Pvc
z6D2cq`(0I;`8DZZ;8~G(5ZnhR)O<<3KgmB&sK8cbu@%6&lmXK(Gl3%XYJ<f8@gq0|
z3Nv3CU<5w#B+*@hJ$iTdvzwOlpFzcAP~W^q97;1jvx4@<610UXP?|fKyhNpukv$z2
zfqotm$dQgL!w*3#1GY!)nkXBWpffoZ^eS;64NMn!{rx20Tv2N#Oi^BoLe}QZe65Wb
z9Xn0aqgZ?lbn`S9R$q@jFr+%5WCwSV({?rTyv;gZRqb7Zo~MOd9F-4}Oduf3eQQ&)
z>|>ZNOtI0>F>$g63=TI>+`xA4L~_mtdt*a4+%v2K&1I9fn-V$_1g9gS8p1@&qn}aL
zqHrCs84@sg{DxNcs6y``6G<ug+?Ye%ETX0*b`H$R8U<J1BurpnQ%^pAJ+RWAy&xKF
z?ONhOlmQLOZ8u8+6F##nYoG6qA)pXEy>WI@d5@GoI0-?lj#0Ovoj{{R6`~SYg^{nh
zY;tJ>CYx<>JpqZ<S-s@p(;_CZi?C!04%l<r4<}yj{+KSuN1#=Dhyq1EqqwNNXhOv#
zG{ltu`y%~~Y8T!m)zhFezM~ADZ2Yp=%TY`Z?<Abn>&eSO91oA3+ti(Xb{nv?j6{>Z
z=aVy)S$ks4Yx9U0Nb-9H(I}6XS&EDYsjD5SO(u8ht4=#tWw^y26Xs;@oW7RQ9&l)a
zV^aOD>@x*DMQ*8FL{1~q0V6%Q?H-2!3GQG_bTt!xzcuDWJ`-Qa5g1raz}z6#>Rs8f
zrqpcBzeR3vlD^UIDL!=iuVWV(@%(T98_CBI1U?DbJ?}!be&W=^7%VXr<4WjdqwsW5
z@r7QfIe+*XtH&}K95egk{Rtq{2a|t)W%)*dUMYjw@{bs3yvJ3MbIR^{7{;2x?X@n1
z^nD-Asl6E<D8`GE=(EbkvQHi^cHcFsl27BqrFqEWN999j%hi<M^E@=BRf6zJg7BUs
zrq&744%nILQYG$!A=SDnq~LjKpq;VDcC<$o(gENXp^kl<Rs~1{&r8=HqMs<UY)-Ze
zLhe5}`6k%s7%B8OslYU|MEZ%3<)}(4dcK5&N?2&Eo3n!+ykH~H+;i<qLD%ywCBe?1
zoP}MZ;>Pp4IFnb9XqC7hj7YEZSj~4%Wjc6u8I~*hEY6Fd>sq3eCp2aV*%M=Rk--16
zbA;b__nX#r@ns=5aZo**8{{o>aT&z&paHttfY&Bb>Jo+rH|b}ohetlrk1t{L97yZ<
z#@_-85FvP&SV8016pWghb;~}VRpbP-vCb4ELz5{aKGy0Kx=`m^jnZW)A_+CRnos)O
z`3*VXs<Zr|%$1do$E9Z%z)tNC`6_%+6ZZ8d-t-1&yR>`x3KGb}QMYH})0yC++%BH4
zEZuB76f2-)p1({H;Yqk}+j40(szAfdw@s0C+Rh{WPCI*}bxuGw1#gk|{WaOr#~pJ#
z#pli1lID1Bs=DKSo@Kc_Dr;5Sd6*Ih1z=}`i>cX!VFtA@c)z(i2zS|z<s8bQP!M?+
zQY;J^*aHC#10eHM?w!8=oAr1+aXrJ~ON`E<+fJ44^DxDPkOBXctqKWb=*KzFGWx%*
zy8pf+N1mcEU#Pm^9BIf5)7+>&S#GhDcpE9~nk$M?fkSZ88~I(I;M@bucP_M7YOOpI
zk{XM)4-4Xiw;cP6^tM~ZPejX5G6tE%WF_uqB~oLMcv|Pn5z%VZ+RkvdFOP_izc;~f
zCb&P|nBG-Sk|o|PapWn+<Q-;<i-1H6hjoBlYmEwo4(=LSGXHPim%S4w3E+A+fI?Cj
zR|MwAJ}L?vaDi_wR(a9mDW6ll`^|+-s2k#b&&DUKr>_qj)mzLXIQb2C)b^n9y$^Um
zQvx2tv}B~YYp%MG-U2Silbklu6=0xXuVxZW6eby^b%rrmrg1?+8Yll={D%k*d-FWS
zrLyEB=2v1Rf$*RXymgSKv4T`tC*I&d8=2>P!ysC<!!RqN>-)Fl@xLXU|2_R&qpTQ{
zWtuF}j(_3+0KL%wZ>$YWEGVPy_+(wbq7m4)8D5r8s<fD{lN~NQ?5~4Mqy@B~3Z&5&
znV;W!UgZ)WFtv-&!rRwC;eW>7^O_uZDY#@&0OAU#<wOC$$o1$lhH{xNu%np5?87}B
zjQCpqdDaJPMQ^8(*TY>PK!EEH1&pN+NCJ))DIh8&?}P5o+GST`<c@U@T+uM>>%YhU
z|JU;&lUM<}h-+9ql5VKWjL`TH!PCW`#<J#$<ApW%|Gz0;sRF;?A_6r6wH*KZo%!GX
z2Uv<o)3Z&Gf|QbhA)oryKoS_30|X*_^Z)!Mqr$)^IA&F%|6kwx|K(@E<`Aq<L?2dc
z=f7RZSE{g9Btv102QVT2GJn6N6d%T;kVvBqIqs=cxk-NgUs1s!wSkR;+n>6b|Lx-b
zAxwPD`o}HNiqP`>=kI+L7GFWKLud%82u#YAyw>o&>9az{pP(Y=nAD#F9I-AxKS9sk
zY35xp!=Xpji|N9=lp;D0mcYk91CDaAfpOkDkDEU|rqPlF(E$Wn1(CF`RsQpjU){ey
zH|u+Mz09p5eA%Z|XKif7=C(q=FL$+Kaehhi(cI?GkKbv1Hvh-*Wq<)kN0GYNi}YfD
z`LC}`719b;DkbQSj0O|x<Iy!pa0%$>^bR!9Vo@2yO#)_=WSVTWI9zwxh?g6F`I$ai
zzyRIL1Avz|04c5!3FJ9R9`=$#8l+`evs8BCol*SsQH`g1uA}bzO<N$h+tr^R{1Dkz
z7z5I>_fG+y$yK=04Sa_mSQwopTe>*)N7)#ds~G6Sg3W#R*4!avYh+54M~xC^GCu$w
zqo)?TrupXfqDLeFU@c0d$Ab3(YW@U7IMZv=d0--42RYP}vixIFLtF6E8?%#tX&3+F
zdbte%EA5I&ME=D;lOF>asxOiW;!odn;?~kDrZFM|c_i0E{MGk_q`x>s97g^$H!Ae7
zU;4MZ>fc=6VvxNa;>)R)m)1aXF91Z^HyZSW>md`ZJW=)dEQ=K~FAPyM$0fipF$J`h
zcw>H0J|bH^0&F+-c70PWmV4rqPf;rOJV*;4GR1+KhK?;Do295CEvn~tAW`shU?$yr
z@n_vC!KhNF9&`o3P_UTUH9`(5kbY?;Ruo9nfCj*<Sy%?9YW$>DKF0tQkg@>GwQ32G
zTN6lUf=I!v0yk8&v9iv6#BpGQ`(I1^f7<G<us~dWrtXu(_0OBu8YTH9;2dcSgW)UA
z6O_te4R~J(zz#+mIi2;J%}g+0h8O5O{eUy_3Yp&^<J?2{Pj{A$YL!L3E9<*mE+-p(
zVz&*Tc0jgMS@amd)k0|2jBK2M8dCfs&D##FR$QE5-C>$wE_+cVp2e{plL5non-sMc
z%+n5O`kjR|{GGGf1e#gw^6jD92@<nY1+Y+$OsbF*Ur?jzdtZUBrVh{$4H@_dl<0*6
zK*RBvb)C0CPdPTXao|@!sQ;fgoPV9G&qToyEICr;_T*pp<p&pdQJ9`AR@@x}+faL2
zD9E$5ka~z=5c)<u$WD1*tlG`o|Afhd`taWU;w=Zx9U#smA}g1gz5x@wAmigk4ey!(
zS0w!x41Kl1QX0Ha!;cdqzYB5v>oI(_Misu=ra3nNVfBDdn^zDS%B?beMZwVWvk&sC
zh(KL>jc)>hU)a>i!dg*yPy;zoZby9J4{oLc&%{|f<J(}aI)Xu;KS^DX9MZLpA#Wyl
zbnGTLD+M3;LtCo187lc;L)edy#N7q>30cCJJ0YRZus2g8BBM|DuIKy!2Z7O%-_g@k
z-@7R9dS?sOX^XBgF?@F=BNtq##CviLcRbTLUJ!XOe(WxGPPv|mJ?^^)W&1|Ad<A7^
z5e_#?ICH1Cymb>sxSDJ&9$zER=_=3+1_Oirj7a6cL&s?)j?j!K@b1U#H6p#FxaVuJ
zQtU9(@uvXEi)JwIvUDAwo_PmoMQKROiFSv0P5&O(v5k&L9ny!{VeI7NIOzJrm~8!*
zcE3(ok32ju(6`1%A_TqG`_=f_FwmibnpGQn@tHk8F%tCDBEQ~|{r>6OU8_obx=h#*
zc;G**w07c~8|Lo4FnGfCIPAOwGVr>+^U_@Bl=LXCqyKuyc=-9H`!KkuwoJ$0{Vx!t
zcd+NuUV<H~(&YAC!^W$bxN1o}=c9Ls%X8xir*Ck_XdZ-pme<7-v0M9Zi8tq|%QsTM
zaNSgTw}aC%X$c4=(ty|Ia0$;BVL{sCn1XgQix+ayj*@Rz_WB<~``q-+sxnTQ%dIas
z``mWUL#MFY8P~<ymw{HO!L*!<pPUkZ;OwRMyF%xrxm;l9JoVOgx?MjwE3$uDsZ5wG
z7wwIEe(ccnIeP@>A1xtl+?vl|58LKJX|a*-3ENx{b=8PJw|Tb1IM)5d5fUww?t1en
z^MYx8rT{$wB(toi9y@lyu{Hi>vi$A#uqB!gGl=o6*2|WlJ6)a|(aSP*Q|itwfT4>6
zco!XH!lDUWRNAFo?J)RI3(CVO#*BY4yL)NN-1>khd@+Q!#?NbhheI+!u=zxMD}i_z
zF1uvgm1?UVB$wGf9mX}JdKKJkDROfonUwHq2czSq6GicpQ*S&tS%gpU9n<^~Q*4Hw
zmv}~uPg;3w^M24-k@Y#s=dkKi5Y3dI5@|pF=T`Jq8uy1E0%((}WevYy7#b!()%Cq6
zHBq$zuig~QFS|j(WVXjF&}unNFTuI6nFzi4iIo0M0Ntw}bey<>WY@!_vWp;=%ymW-
zgTYQT<sNbGlKc91M=-z8kFO+1bqCIu4J-ZY6amE}@Lcn%E5zLgBj9x5px*m|FD8;5
z6r1QSffX?mC=g@<cTqK@;bSs59~UWUU*meOywiJ2&j-*LCLkYf$Ng?_f7StvLdOAS
zjvVW+6%;qq%DG11KB55-4}y0efTyxOujo)Cm<fd%tVUE&H9FKR0qalQWMInS$PjI0
z(04ER7uCE<UX(*1u~`%vfyP@keQy`HxFwDrQW_-dE)R{amkDW(@x4NPjZvJmex`)k
z1BWq41Ox~R=WI>K-<UF&LA7AhDf2qbw@X(#f^_G2;6<ETe4F$~G^~>Yq*Qpe4hU4y
z$fleVAOJPo<X8hP=HH0UL~aXU#O&^d%4xa_=Kd0pj){|UY>Chz@bORm%zinV(4D<<
z*7!v|;7W$iD{m509*21-a#pDHMyL|@s+jLKMpb%LI_|lB=X)yCz?{d0G&r`c7-H*y
zK}U_8sTx0NTOnOzm_g$E>^Z%Dzs_{{W!<oufOI^1<IQI_kmMRW8G8;h+6j<cM%tnv
z^82MvVj;fZdG{mfBqym?V#MOu=9AG{_t0#6`xEtjX1%{=+)dTD(18OcwSV~fuyHtk
z7U?xg7Jrp~<keG)rfGT##@*wAq9gXn*S^;cgfaZhe}Np;@YmmU69o271j5)t@DX|4
zj(W-YiVArU?A(-n2U6Sm0!O(<kihm2zXjZKk?YxE%YzsEq?dv_d<jbtn!`b|x0Fi7
z{3Ykn`bfUy50a$LoF#ydS$4X84BLOiR5HSqlD!$)+Flp73}PDd^rbwgR0Bh~y*y|3
zhjCW3TbUr;4PHow54!sxFkvmz4OD)2%@39ntf;u`6^NxXtx`|vW!r8+dx`xoq>JXE
zw)67?URp6{5yR)BDhGH7qn%>vB<|0=QrgyMO~mbg2XKf|Fh<n>9!Nz$t5V6rKa7?p
zj|TX#{5pd`ubV|MX2=O_wiP92<Js$H%}98C&JSkQp;xMJc5-7Ocy8r1h8f>lO5JJ}
zfirA_RZg>hG?TGdw~r~zCfa@M$cW=J2sw(nv_X~cI1EAWa}|h0-;tzp4D$eAD6Dn$
z?qj62FEfY;CH=^H&p)fui&bnBat;XqZtNy7bb3kwl7ST*_uI4SMfYgVb}!uwaVQ5g
zsMh`WTcSpe{PXy7FG-(^4hk3&)B4FemQV?L{W{P1&<0Ql5`a8dy4-b8<keBDG#Vmo
zR6d72WtN2Ab|$XRD@D@%;yG<K80&<%@K*fPZ`Z*-J)79flGQNFxCE4kXb@c}E_t0Q
z@D96xF{0_g>4t&BHK<<v0W1Az_AWItOs1Y8i2rBgfJ1WsAt??a&y2wXL*-f!fAF6?
z!4GEc241P9Cx8Mnt{~{OBd70(nq2cZ>g5FjOawTR-TX|{o5MXwVh#xJxR=q#PT*WE
zLn6OcrO{9O8>mD3y<|9N2(an!XU`ErFC?X1wE${sgJk(45E*+~;OdY@;UdwUhPa86
z?gZbc;@Le=DnEfpH=e2Nmsrbj02`OrG8=hsBgjoLH)MmQdmkT?GF<?Z=Qk%5u`N*=
zt?dF2wQyc1(|Q8<aT;CygQt9N5g_3ckdxSdOJCk+EThOAE{=oB0Hs{*poSJ1^=&-A
z#H_@Lp+F_V(CqD56N#O|2?6Igv=>~X(=0&!dckX%OoK3=MyPj+Y8k)nqUEKnRC_9R
zU*ac_;9FN8vVssWtQz*rC(H{cvsK`r2=>-pwerg^*pK9!RIv1PlPSq??pjame6Le!
zCdF*!v5SzXvn=wM{0yB$RYco8JhU_6;RBG`x}b8vwgLZ)1&JD4VsJskY15-2?WMM`
zFELLT{E0B!w@K)EsT#9`#CljeS?Xrm<4J;{`+3maqT+??30(FTz#!Fq@$wvL+$i4g
zca@6mh@Fw%2R@0MLw0_0hFyYzZ+$(f+BTNIb9W29+k#U;`{ZOKk?zm}*>I4%K8GIy
zf2j47heGYS$f3I5B5dB0%i_o4gt)9W=o>A7`!KtMT9F_AUT<`9;xm0XrCzjoR*rRb
zZ$QNH)mJ77<6YsafDp&IbnuajETJk>j7M2a<AI{>rnvaJQY0bPQ3a!g%xcEBPj1mf
zldr-M>qHuJJiDx%on!MHdzZ=u6oyM`4f#4*6pYd}Z8cq>In>eLMaFj`#H>udfm(3c
zRV88c4r#zRf3vbKMgSv+;4exxhAzoAYMjxK*^Vw}97~gt^V+6qzYV(AEA{Y-dqROe
zXf_GQ*5vrl$loliiZQv*=pHCz%(Tzu9%8`wt^xrnDF!h|b<;JWE-a*4t~zhDRu5|V
zg!`Yx4UTbs9Nqu(I2}g8D)dl3|Iylg{}V-=<Lxan_m!`<;ZjXCnm{J}G{4?dQ*wKf
zTKOUZ@gSlu;;$VaAI45NL$k5*6#3`<(Z3p_zFXyS0^RPK-lTVm3}x_ZBBrs(4pjwd
zRbpZhgH0S};QJ!Q`}86H@8hU=4yCE?$f7@g*Uy_@;ZqSrRa4$nUUp;nQgOOsqtukc
zCzF*7Uwbb7X8-Lp!#wQu@-%9IWJGU-U~PDT?vzH*Zc+`K|7831@5AJN6F&~8SHBTq
zt@Z~WoU<cmivK>)eJg@8jqSZyWJ9#aL}J%exp%}d{HQSb43o#4(+!RI7#ubZr+s%U
zUE*6JkkB_Todx#rtvgK>6I+N+wFaj?@P~+%!(180O$^U*?jQ6f{hO9>U-E4Tc}Pqf
z-Xd8VBXK^}bIW(kD*cKfik!5v8lT+DB!VO&F%)Y2k{LB=M~8eD1!ULWO}x_2_B)qg
zpvF(1DfnE5pBSPTM+bSEx@}^bjkVx1?6M{@ig_x*a1T#+K_Tu!<vfkTKTF~nVK+dB
zrhdwAJR~z2>x+?yT$F<owLB!=(>wfwUF+QXC_FaNIY5!{FIj@1db|Ebfor)M!Qy#S
z4!%_16W#^o&%f=!(Oe~gZW0n>iI*;OvUK?3f!v7(VVsVSHNb9!9NL$L`E6P1GLN=J
z#j3IxObQKO{{e&OGtK4x&?a@qTbJww?_!CN1-Cg5<&S|H8ebAwC%A(P{GX}Y_1e*K
zatof5%VxOjOy|68R^Kef<H4u<Q%4pN!;eR2NiPeRi<?n7C*ymmNi*6vB+*y!#AOB=
z&PrVhh8=CP39y>$8(;0v5d-(naaI1Avd_PLe~^rD%QDK3cw4vurVu^kS2E)c^KC)7
zPfshSd=#EO9v#jb-LMSF$XAu!%_wI47U|p6dKd5FKGV5FZfKQ)X-XSiaQkaTYABH9
z{ugtR${2J_kczn8wG-viLps^CqULY5AF%>4J-M7!=h^9)8aeZGfBSO=)A)XReK{WT
zBW1y|ji;JiZE_yGC*Wl*q#%zZQitL2i?X0LR^{NQVERHZ7Pu~oz_A}f;+^jgUHCO|
zC7z96+Aw_US9R9mtG~HBE-Iw<%}$(bV(@-hWWkg(XHM3^_#kB&c9DO7pxdYE=EHp7
z@iL~lK=3*qz-ER%OO6|!HIaQhAQLy(ruvbZgtyomy-klgUB^yfqLX%>A2x8<4A-dq
z(EE(c*hMy?Xm$7Z7YW`~)hA{$UivlOvJ5FL+?%!2WtRL!y-L3C31jeuSUS$I#!tl|
z7dj;7A4ANQTMmLa+elGQpCciY57BsHtHUtZ?*?a7cp16Bt76jFmc8Yzo&Jq;gKjDM
zN=42s#Fo*Tzh`Ti;*t9A$Yg<4%R^(-5~V--%tmnB>#W=0=!i%;>4wH4n<U#$t}d!9
zN%SRzd1SW<S-TUxAtWS~E!*j~&JOO|zsZj#&#Xk^4$2tjsP~Cb_I21nw}W8NyPyAL
z0sONz^C0*hZ7Q%}>{O89qQT;*#PC2dGMM9_JU3~9zrNFE$))w&(8?x=#9Q21Z(fWa
z;=r1N<Wvrc+}sy^AVz&;ii;3TmMi@pkjcFuUT1)JUrYwdzxZ;K;E1X=TEyZU5NQS8
ze<XJp#FwU26HOR!T6y~Jo|xXYk-@vBjY*`D@WVy>ashSS*#-a%Cdn}M_()88?Yy~~
z@LWGZUxk?eE81rJ9aYYukF)z8Ziyzf0JKQrN6D8kW}ml^O>db3i}&h8JgBO@`W~Iw
z7{1%7`ZsHYtLC+zBi+c~D3rq9kR1c)k%%ap=^G)0z-UfpHThY9J~MMi=i3W$-d&Ez
zAzGqOC|ux6_@Bl+8N=iLZtdycL1?^+Rhy4h7MmHkRD9q`J9(STHx-s74_C2oCbSHR
ztT}-!@y4+jyvLW~7!tgBguz9-HWZ?hp2E*|gL<fctCku<uCmlN(|q2CRj~M#Yn27J
z;%^$^Aws7j6EJPo(J~{D=n}UMv@l)E2JY8fLVxFN-pia*bR=4+_fqh1=JH1P9|fFC
z&at<I6_$&=DOw^_2%0=C3Z6d?Y&D=+adE7!3wXBc@V9K($RBESX%luzv94O<eRa`~
zfb5y_Pgph-2WgTxU1QP3b_4M?DQ7&^YT*}0^BeB(LvFyqRD$JhYe=)=PxPfzZA=Qn
zSq5Q-&{j&aKSfy3ponv}+0etTp537pZ3}v(9=qPwW<@>aQ31RyKK}3|_RKJ!zY&*~
zJnw8kv1+NmEU@)04N!`|4hiEg{fAO)z(@0gzMd>ZEE4k*re6w`(oA2hD-K5mL6oE-
z*7BtB^g?G3(vaGd0qCU;wD>|rwL>De518yd{6$JsL_ZovCoW11A~bNi`IVsjLPHa{
ze<(`~Lhe^#;FEhg>$g9(eTY9upeN{H0=gpXk|ly)F-HV>5I*Qfa#eqdMoQbfdJxNR
z`Lx-#Vq)L#mEXD^U^g0ywtRS4v5j7)QSo^-fH3Vj4@}6#81RA{j#cSUOwn{q{`ip%
z^cM6SnxUy~1xa!Vb$7&{LC~)rv{MkEoRXRnQx}u(q+UL{*x_hV)HhJUA=};~^n8Fv
zd6Zk)YWi@BW>`Zba7|o;M&#_%z)0izInJk<`^({|p*R{B-=btPHs2^+_*J~qwU}15
z=@`lJ{HP@Etg$A8ul_P~!cVYWU$gmurd(?{NbL4lqqAx3Y4s{qId;r*^+`v`q=P8V
zP^}9lEs?4*wTa>U2mOg25iRIyzT;1I!xg8E$5Z|uj4b1{_2nR4v3CqOeF-f(0Mn@j
z_N*Y+eAHP(hz+yG+YI9|E^3pat>lHVF#$if-9q1tf8|SW)~m4=5#6aZjadhB$B~D*
zn?IC>tDBw(8@hJOOGkHkn4F#xz;Q3JOsNzkj~_#A7)07HZuuBHyOuvUG?g5{K7!4#
zN8GD(G6HKH??FHNQ1%W6l*Nf+?MKf&h-dF_ok%FO&hLAsIL031GQKpE^o57RpYW4D
z{l?YWP_{jfV95+-&E>AYSpA8K53IEtlzrM;FDu#wEf`AF9BMrkN%dUI4AT(F(<FLR
zCF~y4TfRvTZ&KomXEZ2{3(nugCeG)hi)_H0Ovj=L2@nIr4kv>^N!>NE`pbH{R@%+1
z!47OtH&fBZ$IV-q{a#=0oz84wO{;BE%im+T8#&_BFNLZ*w1iXZS95C8Mc8`J+h(>K
zQ%pU`6fqP+OLf&9y;@xt4^)nD_R1=0npPx!IW8^Vv%W6d4`!<5u~siCU<o!9(#g&j
zcEvg!@Qa(5?+MzE-aIkyc*0v^rR0^gGJ#3hP4QYJ6b1`<Hz0>B6}j0k5^*Y~06sO|
z?@h?2mk0b8;mi5On9((29n`$H&NkuyY`p+lWa}3ydm!xv<2{!kd?#62`8+{SAU<2#
zK9pRX2|_e~at%~uX}G(d%8A~AF5OCqVm2)IU96>#Ca)}p^HE`dh{~AsB9niizu0KH
zlnX*rByNNnLLwb!P0vppdnP-4`;>3xp2|t+w!wQik&%j=N_NjIh6eB)_RRhbk&Ij%
zh&u&63HU;sN}y8iiE8*;VCjP!Lm!YkUdmG9FDmRb85;m?zzV`X`64Uw73{e@eA_J1
z&9bdRISFve;*b`q*#rorYQzM$xo}h=HZsWa(;lTslVm`baJvj6uYJYe>ujVV40qSc
zVr5!YYB_H{3Rs&u1dVN)>VeK*ULhMuF4Y@UVv@F5c~ruZS^P%Z%D1CSvI&N?rUVwB
zuvq?4$Z%i795wFTYgcoOm(LK?gv+e(-yH|x^=<&w8^3mq&~y3_Q=KLCQ4)LjD9xY*
z!x(9VCp;7<h&-Lv@?FJW1(C(hc24<N^2C?OK4m(Wpd~UWx?LKpQEj+f-`i||B^9}Z
z<f9^0X_XU5F$IJ~_%yx7Je+4=d-*D%2{<Na_Xtd&m7{WuzVm#u9{$=A5yuOQ$3vvC
zo?;`<2gPtt;q9{yW8CYaRB{s_L?nvb?kI6OZuI80^)RYSAWd#zK%QZch1r|FWE#tf
z6rS4=NM{`O)H!APAYw_O-ekc*2LC+qsCr}?E1lK3(I=AC^^Fjju-vkrfHwu*5DX|v
zu|5mOjWQ(NJsfAMe(cW$L-YI!N;1^%EeLa?J|z7rm_FbxUfMd<O%yKVj34ojHVl%z
z-S>kU{*{edO)lPib5eBpVOqa%YIl!*Km5dW>;Cg_Jq=8QYM}ZJz^RBI+ya>ir`D4d
zPMGX81HSH}V1Vr_+(m;V;$qgM5L-kvfThb#Ts3d2go}L+-Om+g;SM2R+_CR_Oxc0|
zn8=D_$aXXCqbA8XdpDZzRdBe|L{aHBCc5I&Ihn=W#4z(&XJi%1_VLbU?h#B!ZcgPP
zaIa+O^fNB`DRDb@BYNe1*ZAog6|wSror;w1w($?<v>7Wwf~LbW1!*-R$(szHD+^vi
z)TgqPj@~4-!biW`+wB?iAr0*HUYv;%Pz-B~*h{wEhfY)i`xSoMziET)i1MK314B3e
zA7gJFRfW2??<y)G4N}rwlUBMUq(MYdIwh5Ekd}~6rCYkYLFw+6?w&|b&O6s$Ywvx|
z`Ofz{WB3p1kcn?Sao^WNX7HQgyPI|jard@uR>^Q+xC>Yi37-2t;APs~EV{iYFH!B2
zfDj=EdA;z=8{xRf|L!1O8h8V*pq|UbJ^Q_B=(OmlUJvFh_``~A4jt{W2b|L_6Spi?
zhlr^de%k`U9&gHl$SbF@&qGGxu^imy#y0UBm}@qtLJq19-trKJ{#eV+ILFTK?0?LP
z%0RP6z_XN?CIGscxHSWciMyNWeY*o#p+)YSYNG}XS1+7dnf4kXB7I>t4tP*ZYip5)
z88-qUNzZhau^0in`6g2h_T`Qyv@_;`tnrA4ov$1Z{xBL&l~$Z*#dwBrdUi1#NbOe~
z^y~fW>gumC5LUdNh056$k-vSNh@>Dehfxhn_)~J*sB)$Jj(stV3DEBR_fQ$l=f6g1
zY>lAr+8in>6@5D3zy8kgrU9Mhx!%%^YM+yC;^s%>39OO8iT~2kqER3KZA$U{t32HM
zoa|}E+?&)ayLsEf;u@_EoAeQbYc2Y9@zMcz(l>R~=oR76pN8aWnY@3BQ$%z`o#RJ2
zsdsUeoKVUwk>yNCF4~3^!yQH*if4u**9*LE4iHKC{UHgbNWNS^h91e{=vNyKs+DX>
zVF1xyZl6t^uI6Q7$P<BIv2XUm%YQaJ!0+tgfe`H&+jf*k$xhgvm!+}X+8Q%UCD*K#
z@n$~yDnWq#T(`R;4M&&wRveSIb5rZ`>?lL2%v36=qarri7*}V+-t957N-ONV^u~$j
z69giJukDC4v4VB&xJDMLS$vAi*TZH!AsMY*?r=H~N38Ru$CLh!;x#e;_+PL5xawp_
z8m3v~UXYUbK65u`MncD$!0GB2<r3sZQf7HcmRr0Z<gPo+yUU9Fyxe^K9pv$BZJuBR
z)W<?^De^MS_k?}A6NXCio_KxA=Pe)jJv`Kgg>al+J`N;6aOKxT5pXmTdow!tKI!)-
z{s%@MA$CTi@yiW8TSXt%B80p7?XlHOlY;_iepP6^TXo?u9#IGj#>rS<M64p;sYhbN
z<;Q|V;w2@K&wvn3M@t&l;4~<Q5)xXqH&^{j+@;F1DI!@`+O~oMlb(VdHa_UiO4`83
zs63{Vi>x~KDs_*Jp5Pvod=5Z0rkU;lj8n+)SHX!y*Npp;$fqdhtzPB%HZ!X-KMZnf
zyFhzl%OuJS0=FjKkX&lEWT!hvYCYeaPwc;6AhM^UxoF?Uxgydon;;hYHyZ=dpRHPS
zFWch&W*?6UvD}LWS56>0s-GGrRMB7LA3##X(+`us%N)Xj?G~0E_T}TC2pCef$|m(Z
z&dtulq1ih&Qw-;y?_=~CN>OYROvPcrrIqGw8j%BId>Y4je*Yluhz65N^0Uq`HKH#}
zkgi5KG{#;Zoe=AE2~wXPf9K1qD$KkwGI%v>YV!>Bw=4#HNPD6G6*jE_uT_#jj+M$s
z@Cvf3CzLg9>ENa8boMmhu%793$=&@}O?+g$FLU$DL$9?#aa-r~8CvKcO2OCA_lcM^
zXfTIIHQp~Gr=4z|mG7#@U61w(OC|{$#{J2{HTjO9K}X8*)Ml1@pnLF}-$#*c;w<mG
zqvJV7&&eOR$2-h+8#JufQh@4JeUo)yK#Cz32hDJN^tR*D|7u6Ep2?(9g;TRm9O<FJ
znDh2}PR`#_4*f*RHi~br*ke#kNMe@Zm4S+GGWV+6WEs?Is)TG5cL0J{%T!a*QndN8
z@qpg)+MUXDHLRCp^~=h3Hdoa8{e*>lnEv~~_TLSf+E{0aZHB8-#R(>XTPWTZDCvy3
z`qe;v+=x=fo=&JC*7r8lUttY%&FZ41?{}TLe`|rp!lzF3k_979J?9ZKv7aX%6Kke4
zK^hY2Z$49nl6$=>nH*j8A1*JBJ%Jc1sCJO^Gk+Wn=YO~aDcV*%UWeCE6Bh#AcMGSn
zwa-rTRysq-%FEDxZqu$r1V^h$F>v|F5Wxq~h?nmj_G7^es|+|$yiLw)kTNZVA~Z#s
z<R@Y@tSd1q38sPK7ubPbY=WXW6w|ST95;93?kw&T_B@_>=NIA*J%}eB7mMWhXs{a@
zmk@qDC}RUr*K*AQC|6gn)lKxJj8|K%{3!03%>2|r7f{qvszvSG#ol8fsXc0&s&#0e
z{jAy{RVL8Lh&nr4u!H=J0(wD7Aph;fk5})ipQe3ErZU2*)*3$+9L`$&>3ZLs7U6T;
z8OT@t{D8cE4+vye#(D`E`c)Nz$eooDo6hWUAi9By#fiLzZ^?Qg9tk&@2P)S$nI@S`
z@jI7yM+G~_E3x(QNuBz_=2>64?ffU<N(|Ck(qDtWo^AeSu($YHw2}Ok+~|j@w|d@;
zzYp0-_!WKivuQiQt>scX7A?AhuF3uROcUR0Iq&%Tap(f`8M5xQAdp-PW`6w>RxV=i
z<JJ&FEmi(-?Bi3hS}}xop>NPnevaI!mWk4Hu;u1y(^%M680ZF__@;`kIncS7eSKB9
zJAn8Gwv|CYR3~U!B&yd~Bz`Y;9Y7-ss}gr%B~}kRGThTSO=vRmvnT4#b#1xuj=UeU
z)=c{3ey%VoEE#t#`Mhu{Lh+o@a2u9s(8W6c`~5q_mK}dk$O+&qYqtXmJ&fiCyXD|p
z24-Gs)U$V0vm{A_zY+mvb^!Z!l+UW5N2?J+tXFt+ougeHJ)Vk*2$~YaVIG`}|Hu@4
z5FZm>az?c&UKaI69zRMQk5W+5-7ZH7x9SdG7VORrAd6X}8#tc{XqcYql?~z@_I<#1
zCWIS?Hdo0%NT23Wf!at~sO>+=RO|+^u=f2d12dH<Xy|%(djf9+E>6o|zqH?{^A$|m
z5L}7ql|ZD>H2?WAn$@a|IkYN}6*1l|HnkrUUsj8cE}K#PppMBekuzpPlM^p4g9+bF
zK7yLim?hu!ib77A=UbeL(qqF>hInR9ne^_+I>{$aAu32`B>X6r4=tXGoZ-=xX=(b4
zr|A=M^{0TrF1=3(v?I^+H@nxVTxztS9-u*_^X=A{DxRQ$BZ&~TU(5GAH^E=z*r?ka
z-zgRkc?KezQ7UGcopHNx^OK=fBg!lFqS1mFKjv6n?c+_AYVj7w9unuMp8m=;sFBK1
z(D4-B2&3P*v4~w#qpSAg8R&zN7Pj-&wI`MBK5XmWyRVtmYrLwS)~0NwYWto~33iuQ
zLy&Sv?*Pq!;G6ZA<7Ja`*V{^HH3(R1!9m?Atb(X|5hzIQyXi_g-AsXMLPNyo>0USE
zV(KTa&@$5kyeua#24mv?2wAPw0yN=5xrHtvh)scU=L_-2F^t?TdJhr}n$fH`cf$?L
zsPWLn^Dh-33qa~3fXtT@<<DwMT`G+JbZ!0oqdSuh-l^_Lx46(S^Gi<LD9Bv{^HiK%
z0Z(Rn+BXR67YZ8IZHGs|L>$lSLQj%~XVz)HN{iN`oXU9p4!1DpNr%f%ecJs42Kolw
zF;}On*+c1%5>c(xkS>=lD9f*NyW8j{LUrrG@l>(HXTcpXh@^L=p%ZGEnCyf^>(Gz2
z&7`?13Y7@`;1qfH`a-=OK{M(G^qd9*BqO{Wjz)4#zM>ZF4m5p9+M-h|vX6J{*^@cb
z&_ZLujYqE>U;a7Os4?`o8|c6)Vz&o!#%s9fk$1=uRwx=XJwuqzQIiY)KTpk@*KRVG
z%=3B}<8S_ukq$P!wA8?=EKlzLMF~Y$8y@8WO+nobtyerct+9nRxwEUgB=9yKNezC*
zKW)4b+s7bAC*Yy%SXUui@s(6_I;s55)7+V|cqR|C^~dN6`}CG@We3aXM<_}toP3Gm
z9C4P6VnX+g_jS->=+046m_Hgawr?+K_zQ*B&h*r*4=8CeQAIM|EtI72VT%X`bi*f0
zK&thdjwtcNZ{%2;g950r5Z#Z{vCj~aq-dVm!%lrE>>Yid6c5UeS8MPG-zk*cKPFWo
z#D_K;7g0dNMq;Tbn0pM#HRyKTB?cm;>+c=;vj-PhC`07Jb*Q!bq|Bp#HS*$Yy6NM|
z#nGjuGEsrytGo4?6o*AIGIoEOkeA3~Q&vI38Y<otLz^2&w(C)Bp4XksF^~5VG4JYz
z?P7)K5c9j`le7DImX$@oq^`$#N)>dME<q?orueU1gW9>5&3X|G{M@2`jrWIX|K;pr
zQc(PJbh!r_np)$y`txC*c3OjNzxvSHeU7*{8cQ?sygfR8F232s-h<gAd#fBzQTi4`
znp%Ah1E0j)sG?6I5ksW+As=^a+Kp+6Eb;4^)Z>KkpkA*cYJb>*iaH~6hPP_-!MJ*L
zYIhNN&kLJ{y624XHC1tw@@e5-_0dwW2&bebcOdb%^cq|*dRyXArzCV9J1%JkdPF*|
zujqD>P|Gx;(qaA^z23u<{dzTC{swXN^j>u0?T0`IJ{%k447#v0TN~(WR4I^XvfRd8
z8J?qT(gxzpE?|(4tH(O%L@xzp%hk|~(lI-NJJP<RqyB7aH35xxUhZ)Jxm(0%`rA<@
zP4k?%g=Nsp{|$9$^cR6*dcORH9!6SjCGq_y0)ZEFr$TyR@a@t352x-$D;m)Rj)>2a
zM4N<Pz6%RmKRG}+GS-ax6;X2rWPqQJPQ8}WmXW)#&kpJ`L(jZkC3%UbiYu-wG!pJS
zp#ODX)d9uWYp<G*q`l3gbyak)1k~Q^X)RsX))XV9GXy5^>ggxZKZ2=_k3#m<hWoH!
zyfq0=Y)^}lvX(dlf4X^afuP1HPE_u0H-}k^j-S?Y6x8$)rP#OlPDA)pYt1>18Y^I7
zcQTzo523^Gg|MOZDLx5fSa{Bz`DWX5()iSikPNP~RTI{5Udqt~+n&C};H&{6p7C4^
z-_T-bphC0?TEl;DrObbY1x+e6Lih7(`}+_xzrNdkmqrzM$oBm8euV-bgh@QF&t{Yx
z5}P&c-^zKn@dj3LQRQ7!dH`<Yp>32s42{^0SjaErWK)T&qkz2xyUEb1_sJ*NKr87X
z&m!@UOf%gQnVoR+QWgIzpJJq9K^nxYgOu5NKP>2zk7|995go5-JW9dKX>G2RZOMC~
z(9-=5<H$Z^K{c-ml655he))m2tWj^J+pWEFG<i2HZh`qEQG?%cHLDK5dsm-DY6?Wv
z>mUPgl;EPHxiX4W;!jMcOM%Z<izghyM4-ZbdCT#z?=abhkB;U`!mCcct>iU)29M`$
zk1L+z6ZW9e!77vK!xw<80!501LmATv()0<+iORo@=ZbK}(f_jz$B@MJX;Q?is9}|o
z$hQ<qZ&F>%z_+%>@@wiAqjWvE2g+Fepy5ZDf^W{&h+N}S2dG7z(r!Ld+WN3a`MQaC
zn`AKc|BUqcRH)w`^=zFdo`p~B$?avQ1-_%v5Z{)=xo=hZn@6S)4HG`;a*g~p96IwX
zD{TC3P8}eV>V=PF9RLb)%``fa9|tF=(*TabLBwS$@8_NR1kG3+^OzLX9;>nm4a=uy
z$;~oB4XVH|f)aB?$s_hoGO!8{t)nyTbdX{J*S_QNEBtI-`vF3JALtGLug3&k&mp95
zxbG0uFFjo%ntl(F8-M87i7)h}<@9AcU*4kSN4negzOmF@42e@jejy(z3Y8%E^g^KS
z?KC6Ow#O|cwJ9IIb90fd$Ss@%baiFs>bMZ3nT!5N9KVkt2F|KSsOL%kZ9th~hxlWT
zt<E;5X&*>+dI8mXHA~5G+nQV>&W)_nivqLDK~Ji#M(YjKd^=$;l&wXBSxzIG0-eNX
zyI-~Ni9hEClf1x2@Do0$V5smm;c9l+$LdKn#p#SAFk|+vd{>G=CCa@|;N15smA+J?
zk^ug$vjclYrMW+hXeQ9n?-DtvL4seiOVrtQaWKw{+4JQW@+NWjG$)R^vVv<S7L}xB
zjxp@U)C8FzPq-|<S-cIHR|^ZFl|&!QMwa?AV$8|Qi@NK4%IQ2`-b;!22Xrao!Xjh-
z__YKk_?)nyMpZF}z5GDl!QGvQZs}Ad2dACP*R^kcZ4%ieqe$UW8VhG`;8ki2f@B0V
zI|yPOe^?%4RpY1lQrxgx!0K|E;Kg#nVTvF{a=>S2T3vn^5az9lAB$45jbMrg;p!@N
z1r2tVbQ~Vwe1+o8UiK=t9PsY48<m?5;ljqwJ`T3*OjQ2$@IPb0h3XY1zr_+D4P#3r
zhpZazz?d}HYiBSr!t1%9s#^EmV@aBIucl~*is}c8F2qIFjl;w1IeWYYIM71C{oqMv
zfYqkGx_OW^Z#Av@LyzpHz%xq``tNXf-9VXWZtx=W2+z8+<Jeu{GAntrz6oZxbR&6N
z9ys8EXFmbDRSIxULY*Vt6lLVjh&ZNW{FvrZX7he7{E8TS@m@OxZk$f(oD$>yM^xM3
z+qT{t69bHw(+cqsgd9{WJsNb`GY5m~LSCS7b{n-pL1^HKvZo`D5%JE_3(Wa0qW*I%
z)cUb*C9p`z(kl|GLHOHK+8yRjcg!b@Xos2oyfu;C4Z1y73qw3R**D!JrIXW`pdr<e
z?u;1XTdy_kMzRk>1a|`qFnc&Y#voV9E7{+EY>8hZ!_$=ymjIEPNX@m!qi7@E?({OV
z#n6X~Hg#PS<J(f8__AE}mP=rDQi#8_S1mXG>=%bjhufa|3Oj}yi$EG_$H+IyW7tkU
z@v6zjkrvVuX^!OcW3pCo>%%t)fjDAEvG0>*!<DcX6<A2Y$Q|FiBY!d?OB(QNe(tv;
zL6y<@{T^9YejN+nB>cFSIov@}0I{saCh%l^(QHzIobGCUkD-fymmby6KL8B~yj(X~
zwdt(KPhPp8kj6}{2w(;Lc$7h18k}a=(>TE#9+I8vgW{#1ET#CzAc`%LeiRpiAVzag
z!yHO$r}*tf?+WQ^9Ke3T9g3uK-{?oM9tynVL%_IYjM1=?yvQ1F6X-7nWd}#s)HuXj
zEmA(^Ty&!h5|0kv^3+|@1Ek>(1fJi;`EPk5WreUg;+N<SE6$TY8ckw$bZ}q?wq%5V
z3OIhr5Y8Px_|O2BdsO~=W+0AHedicHu6o)z<L<RH2VW=&)OH5aUVfrmQgs-iDl*1d
zeM3wY$abXrt4h&FI?^l1iSa<A%{1%Ld2rdby4^EU8Urs}M8~IwrqWHc$r{!gs-AI*
zDx=Hg@G}fLm(6fVb+aTz=&{tj{|T)`i&{7b-8Fu5zlQJ6m)sZzD3~1-QdQy|d!WOr
z8AZyh3UT7Vp8*^cZ4vUT$cH$R4dlVJ7}v*Pk2LD6wn}y6mKJGlf65|#H~pYj{_V-=
zV=Ia<TlrZ)?btug67dxw&Yg1v$2B+$99Me$t*9&ue&nnoX4y-lmT(V?pr%lHC|;7Y
z<!0FN^cO`C&j<LA6S_O5aRU1<{PEt-R$Fr$X~&^Brw}rGE@x)C@mb_9V;j37>^(&|
z`*8p5ho#*8BHo*JwwfgqjHD(bm7MxRpc<Tv0OvB~4^l;+;wU=+7a21#iL3D&Ucyf%
zZxHheCR(B&CeO9emxqs-2BgYc+GqT1MdLGKu{vii?%tLd=Jj{%p(ChZ6?1p}7q&yc
z4SNJdQW@1fGeu=IKyoygLOD7<e}hf=tz!Sgk0;NS`|UDQ^hd)>xBUaQ5q{+3u_pnt
ztd<urSPq9ChGAFhQa4cRO*~IwAd-r$qnE|UqSg4KWo^H*E#|vT<0+oVZ$Kr>`i%BR
z4`?_r$-cm%jr^k3XkYQctZC)@h`4iD2#(T%vxTiv>?a)I!%LwVW=16kj@J%8Vh$Zr
z6Jv=CCx8{C?SNF%`BDkXcq#y66oqdTrF_nke|F<A?uFW!rK1X-T!*@021Al$>@igX
zwX6eGe;n<pDt)44uuziWHiazkiuC-79@xsr!zhp-C=+LMF2$dhh`_QsR&sws7rXxL
zPdp)~zfb&6U-ta(xODPZDP07#x%K=RF=v^m))>^Mi$*!wSfc@T?7MS9ocZU{>caVQ
zvhkll@+X$MKB>}2Yg@bKuvLx9bEMOUXR-<az77q6oaymN62ZQiYQkJ1;ymf<Vp1u&
ziY|F-sFD$TC18L$p|T|&th_NZfx(ZTf<ULvjyL}&qdG>rSg?0RFWq<PLN{NCu9`zO
z>x&Sr_WaO7Vv#=mVC`7zvax{XT*J4uvF!U$gK+{mVTs!Dn(dQbM@j}|*)NT8;=>u#
z@A-C}nHQA^CFj|z|7_05w$R>>nm6e8fgaP364I=qfCJ<M+&X4T)!4H2rLR9~QOc}@
zz){rBgr<SgPlNRKoFTLyxaKEg+*UUj-Eypyf1M^r^Go)2x<**SB>PnlI`0=59%{0-
zTpGSo$0p||Tl^*DRaIICPPC2)j9zh`0E+AOy7+jqMW%Kds_<diWm;hDNsVT>*JA^w
z6;a@|-Dgkt8YfE+J-dGPuV<NANb=4wo7zc;?Hl61g89%k5IyWcY1eiEQ2-Aouso0z
zT3-{@@_wcj<<VP9mrQ@^Y0Rg0-xtUAX5h9`n4ssmTFLB;8j6<<$&Yxa0Qb!&4&nwZ
z(sNRqNE2EO+nfI1yu*lp?T48|UAk+0RqUFGgE)~giP^bY`L4IKBzaC=Lg$Tg{8|`o
z{c!ED2o46p>|jQ?>`0=xKSXN{pVlcvEAS(q;%Trxs|?TY-ji<fqvoeNq4B{?8r--=
z7X04=7I%rxC}llt@5p}56qB5b@q7)j$&6M}<jdLFn?Tn|#HVElwrT6}ROjgm!O53k
zxeaFhu+=jC6-lyRBF`A)3l1^4d$qeVc=c4HQn`mah^Y(ChC(BQb(vBx8r>EZ6FD5c
z@gRxnEx--F*!1kIQ!|{uE%8B7y_xV#qXz$Y5P|O9E!LSu*Lb+N@Z8%ure7ix&Lxth
z_VfcsF5NC)_`cZG&A6#6IYq)r8#PJX4Mk`6S<Q)A%*^&v&XPCJRCm>ju7Ofl5^YiM
zMc52rMzMg4Ch5zHdKEhAnnY{G(kCzsI{<Vqk0iX_;eEUQ#gA@@8W#MQNf+1BANI|r
zqV&1P{PKldy~TTd_wrWk5yb!Q)es^)iukM;&0EMfcP0z8^5;JCN*@?4C~)<~h!Gxr
zexO!8f~&!*@^<He^a4mv@GGxXIu4~x+!_UV-$MdL@gB?TslFO+0X``ouU8cP9eRxw
zjI;gcp5yo?okwwPk?@$4wDvg9ICmXb8IiMW>biF+?i?YA0T@K(u)OcsI7MJEM$>Z1
zUT#3E#ttwS@GP7(!!=Ym;REZKdBphIUd*$@nnat4EmS2TPl#3-Ey}ZTr(7cczUeJi
zB3#P)zC1Aw-yO2oXr2f#PdJk0?Q<Wz+7qiNlyb^G>So7d!Bs7q^CYos5}<W|#y`fB
zdsyllfYX+&$PZ-O&mnGp;WiYqK*`6ct~+{XzDNpx2>hcbQ><hc9H!n@E}`6&<%i_h
z7yY*HP>c`IS+~yndcAbfqAtfJRbFcu#T}cs?`hTC2&B#RRSf1WvC`Yzib!ue@&r(x
zBqPn@9jLtfp$8&m`s-(sH;*)3b>T}%oD|X*85d6I1tjzALn~G*Y+Vm^I)8}CN?<mp
z1ZYBF<t;NT(`*|B?{}(Q_(LD4y%_i#(eN+N(lPQvu2d=~M^|>J77<Agl+f0gqH@rm
zjQz0&g@k&(<jBLArq-^P2!q~dolqz5I5{8>>Je*$s^T+k!g)A;{trS`E*9(~_BtyJ
z0+wN=DGp7b4;fC8jX(eV%QOE+91o*OyF115L0IO!SsFRG{a&2u)C=~FmY83Y#lP{%
z%OyF1FXW)sp~jklf`d!!Jdqh6Bfy6C<&nit8s+}&9R?sx(M+aStRCANE1pW#X&Co8
z9a5Ge-PokBFb<2uD0?SYqljP4pn)r1oJQ|<Ey{U5p+ZkGp0E|ka@Xz4*$v6Nn@7)*
zPNE1e(Aex%H8Bum8&HFhRRY23=ti&>rX{hIP%)LV<#{itSfpu+P&nq}9U24kU@s7b
zd=nk&vC~CtjVz-7T*V@;=3;W&ga*^fmX8f)+ao4-c42v2bA^F}aUOaD2@-ki1E=3y
zq!b;6Tn3zqU#iGipkF1?>p>QkE>+9Y_c<{;PijH!B#?O-tKrg5+ZI$-1P~C<>NP=4
zH6(|tvTS;sUc1SHk!8<zFv4LZd<=2t2X`-kqQ@r@kjRv(eXv|MS^Ug+{T+JB1K{2?
zk(iAuUj%ZOo)8GdwO7x(X(|1n+1}v|PNi5WODFjxc=zw>9tbvOKo7Se?UV?HheGW%
z`^RLYHgB;YtBct1_fcq-!45mcEeGad=hIT*l=9LpY?bM*NKDV4E>bL-0uVvPs^Y4x
zlz}xY`@_)K148p!_CRiUS$yuw^2fhbw}I#gaD@V{lX1<rcRNy|8WS!+ZWRMW^BoE{
z?2H2+VJ@a{4Av@Tx>PldTQ<;*JncetAPmTs^59>r>cWwu0Z3nS1@#q?<m8=$mDu_C
zsaKM1(8+%cwv-5wClr(}ds4uiq3C;#bAPe<{bOM1){;^1hj>j>849Xb4}6Uu%3xaZ
zzqhWABbF1#_zom9+B0l`b7f6AfYe7t==p)1N4>{55)YXk>J7%IJnsqcf8>fN_ZrM?
z^~&sSqTjYKUO57;QYT}q>s>eC8-GY~ph2<r9{C|85?F4>tJ#rJ1>sgAsqEjRJ$?su
zP6S)QX;oJW<Qhe8Kqt0ac6tOKW)<G{#MInUseG*6m-bRz^a==eHRN`I$7}!AHMje2
z>UNyRZU3WAOH`0sRZ*n}&L;hZX8cw`EJ}omDrDUJw6#$IY>{1#tOXw-2|9p`0cO+^
zg7TFDDgTCwX1ILD?Pw3_vD#)RLmnJ9Ww-2k@zNs{iAIz>qV&7fmvb-JoP6NGXH*w8
zM%*Wv=2sSR6$+YpOS|O*D8}rcs4>NY$bOOwRRXULy(T}GN9%x4*JFp|;aU(5_C`Ge
zoSnh(+AdAii|!8JTMbaYCDnz`r1EtXDC*nAJ|FrB+*nGt=4jah2ZlQX4t2M}$8C>v
zJgp@;8+UTQFSw(!R7F~zF;HuIM><gs^VqKw@%xI*Nz4`7)G2N+0`KT_&ZB6)l?2;u
z*$5<c=mN$2E6Z9X>_V@f1>~wAgmCbVW9L_{@zgwgvQjEb9JuiNRqIGCu7=O+<?Ce_
zzw7Dx0IWgDeb$oxEUQQ!)(Fo$qhJ&aTrii-B(rFs9p{5r4R4u8OM%9~!2r@m84UN*
z8`o%M{@aDU67&E!0|bcD{Cu{vjPH`rO!Js=18%>9{1qM*C+l4kW-0dSuRKTSeDaX)
zrAFinb&`j4+Ozdvy~NbZZ=&2`i3@{Tk0E6J`W^;u?(C<%*(5E<i7n4tTt71MJPv%0
zfOR)B^SYIJ@8eHAbt~pxxa&+{HCC7;TlkAJplzj~;@Xko!!Y=_)3r!QoCD4?M-0kY
z<pLoh#-`No=Zb+q@i1`~RJ_ImUV;c+pkGNN+iyTj13+v(v1SAxYa|P-exY74Bt43F
z7i#vq%{Kp*&SuO3Y`~gkpAyg{?D2$pcD@jQ)8!lNjD=TDhSL(`m~0p3kPJ*AtYm;U
zcL-8tyR)bIQIg5)I>>U7Z^ih&FGi3F@rQLRsx#tplv`oR3pIef$S{f~av#PYxkV9Z
z$P#`ku9IUW9_{PjdvV(WAOokuN?@c~b&c^1cz*3W^RLrolk%U}-nUrH0-Qt&@h5*m
zQvM?SYr&0gIb59={R>K<tN8<VaiTIz4qE;L3<bNeM2$Bsqx~8;;>+S?=@7eE7UEq+
z!4nCo>prU6s5(M^E>s$b87F<HU?k`10tn<Z6Sm0ii<lqZm>vSKS|RoV@PMAw%oms%
z)SN+TmI5XoY2uEp!b<_+$Vo?{@|h&!y4_=pQ!vEY>iN7cnyrQt5lo&TpaLsPSl;iX
z0Rul~;ani4fMdv4LCtGMs1f=uIO%9H#2YO!SIVkwpm{fDot}Q*0yH}fw%GU%oh=@d
zykpDijPpBX`?Zvj0d3D*n2eVGOUDA<S<wirH*$z};Iau({Pa?~4E%3}Xwg_6^I-l}
zV0H8PtQCO0_+%EVnl--{7^tUzi@Xz(|A;x6T<w5P3xXr`oQ0|?LWE9&hq;IJz$V`i
z`5na<XH%jW-9%W<z2ID6Q~~uAw1QAPiK=gDWN%}DT3BbhAGqRGGDGzbxvER3;V2j#
zte(%10|@<gM>9kDk3)Aa25f~`4a$ZY2Dm>w|6I%!STZFENF-6*GrL4)jbjuf2yTg6
z6x{PAF@Qhty#&jX5G@95TI_}wEMVqc#e!XbJ^T3b`dup6OOhzoc3@=bX7lQ;%nd*T
zKr9Xl|B#te7R$QiUkTkFcV>pBR4?~KD#I91%Kr#7{fNe>-lhWs&uTjri);PcL&%sJ
zWy`Rujo)sO!_p!7IX_02FH(<tvqn@IaZ}njV-*NKO3JtZ8DroQfGR8QM9s5wPNTmo
zN#=HsYRI4Jl8~t^8G_eYGrJA?6FH0|Q1a^2-lk1ZX&>6(e{P5%N%;iiSn{p~u?N5|
zw`4%x=clV|>HU(8HjwXb?c5-KsHj={6cI9N+Ypswj=Vh&4ES=fkhm*dyv1ckNbJOx
zgDFF0^h3$uEgPE`P$l|;kRkb?4eui`pg6M|JydISFlG7;b9m-=sXKaw(kj%wqDsX>
z0k^8$ZH^_@Vd#g;Tp*BpyZXwi`3Cqp8{A$kzNwIZZ|$5qV{tHfj_SbWn;)PpQ)}+J
ze4gdq=oz9oQOb0lB@3_Tm*RkWK50C&5T_>CA3U#)#)A$53&#RouO_Z@o33{YVP>0n
zI&wF!1aQV3IYkV6etb88hss#Hp*GK>(Ya?t9qc8;(z!bG-_~$FdZ8pv0Y@>s{XC=e
zPl^A(n4kY#j{i3018@r&avATiX(K$rcwG6Q0Dme_YBfljNmJVtmsWiN_|v9TJC?D0
zMggi0`uO4Q0q}LQSk3r2Ojsi+<G-bU4A4@8$U&+meqYcjHI?#h{oQ+J{dVCxQmF!A
zaIcx{&y@Rp3){APpGL)>t%Df}^VceKI1xh(^q*7c(3{4!!Uk-K|Cet3fB)3%kNc?7
zrE;F)=RrTP5p)5#AZ-w-lg8}t{YEm1&yte(>vz2<8ior1?6LuTZ+kv3@N+=NvMTgJ
z_VXf$fh~T2g+|j}z&z+rtb-MEisvuCoL~&19f6NUkLoLLlWag5HE_i-tR@?VwBFwx
zeWf8c&OB#CU`K@h0i2vjU|#K%iJ*PmS4UPo{}R>v(bMn2a8QVl7^t2KAuMJ7?_`L7
z7Q$zV>iFQJcjHjOGpT9#S!>4YGi39B@E4b6TyyZ0+@@yMv;sgWuG4_13(bd&`hg6Z
zU9~gT2VkA!o&{3Cb)e40LH-0spLs7#GM^lQ`_k^X9jPDuTubG3aJQqGrvqR^V?ox?
z3#i%JE#Z1oFfP#RP0T2EeX-IgwnpRgZ2-myFv=Bo1|9GIFWA&#Y>hnd`foZL7U&r1
z8T+C9MJoIkzVsgp`TzD++laV6Q1nHEenLe&JH$?r<L5I@5ZLxe66nwXR+aX_?3?Dl
zeC2>M5d~@ny`ag%^Df4G=f7ZbtZ*<A>qG0&zfzii(<T0oU$HM>{nn5#GIkBBE&|{0
zUA&HyW|GsKd2_AKKT3~(9f3Isuw2#%CZ+$KI{p9uKL7PMo8kXbv&8c1zsu(Ttnq&x
zwtvHS|F=K;A3xIw>R((7f&<{RoekY)Q52`Z^`VjRZ)MC)?6@;&F;0bZD)zC<xUFUg
z9zOa%ogVngK7tbg9;ItV{%1L{2z<Qn_VmKFCGxSBg-?Q&y1u4=9dV=W1h3A)m@p{W
zb7xnnqxl?C90QhY-&%wSYbd1R3#4j9Y%zGk(45#ucc-xCNfkj5I>J;S`u8HTDU>^)
z(W1U}z35f)8tXYGLu>rsNiu<=Qt%xorC%-jpQWO}M<Dk&_6Z~6dJT6Nx4#_$XhMdY
z{Ys^x-}6uN5;*0-gJ^?&1*a0cub7iNNB%tjPDSD6KEYD?FR8nSp@fHtM&VIF(I*`a
zxUPI)jHA(Ga{~#n^n`1sw>XeQGAgY1W*A%CAhb9zC@;%@PjMWocPr*~8k1Is{(Sug
zIQ5yi24ARv259*|%7_12{Qvw9!hdrE-eh&%-<SK(4ZMwjpi#iT9x1-Qlb03`Jk)*g
zYZQQg+z1R|ydk=M?1o*)QkUY0_CK-oM@hxz{-)gg6ectL2nYDu0o8vGIP&<kCP1B6
zM14zY+lBr5GSv4&lc?K|($$;}vE3&KDI-ap_2{5g<gQEy8Tx}&2NH{Q_;m$O-cEr%
z9Mr{|sDPhX&YgVz{r~+_{|{fqih?~&D_W`apY58?gA9L&E2gw89aD}8a%<)}A<fXS
zQ;=ZygP8`nAqE%L6zqWXPT5-0bSg`9C`Bdi<^fQx8>3R>){p9^O115JX~VySRu5{>
zbWD}c5ldok*R(J(r8O_>Av=iuvk0YQhGWK*p8!j=UIFp-WxrVW?SDSR|G2v;=s?yQ
z8k`$o0+!(KABI#ioxFz`aH*tTk%{?;fDOY5V64wql+GOBcM}|cGYUXF`H5&d*&wnX
zn<W;q&E;AXvvA{KP4&b2h7CicOpt#K)^S_vaT%xJ&u1#otPgbKpQ%KXPB;4craqfy
z{&np?3iykJwGLNWA<5d{!sSYTeO^KbfT-cw=KFB^H(C?z{bLv$Rz)uFy^5RmiPuF$
zq`a5Hz2HZ#h2`lj*ub@@5g%>kL-4Q2G(B!vKi>zuy1lJlLw>~9gw}O*ADKsbDjVkT
zE-oXbYO`5<u6@*~%{_Z@%XZ26GU)Q=qM3XJZ*bhyw)!mW7V0f5RBSArm*#N%)p=)2
zbLR4e2q@@O{Psod1NT2Gv2|4xW~S5Nb<H1k#~9Y{9(%1@&l}#n?qeJFOgdfu9g%!{
zA`DBrU8mn0dmgxByWV%h+d`S##81{&KFB@KTR|=*Z(9FiNb|@=yVI=eE83toLr}tW
zkfhruO4h-SRn0HNttjI`axQ}1SL}-s7`OfXxTR-3XH8A7NFwAC^9;1+u#-M|&aQgi
z3#pGSqo>S<Cl3O0`k1KJryV=e+)_=~GXpGLHQphv^Wg;XCZp{EPEMDo4u{<S63hhW
zJFXIQsku(OowTzOwuB|NCN~npZF#*CPhqTx1irA!$!5$v>AU0QPv`vYc`OjU^_Rl}
zD`G^3+d9S)e;lrk6HVj%6thu@+M=N+L1!Bw%wDk^N4GaS7Zs;|O~>cM-f52K26pE|
z{aC*a(oA#;16R%xKGo5CtzUS<H{{~KHe`>~=RaR-R=@=jW1g3yCJYW*BX@?%fkO+;
z-HHF7C-P#`y<T#JN~K5lVASp8HKA_2>7Dw$;?nv5lL>RG8+xd8|M_o{%>EMAHZ6kY
zHHs_j=pI&@V5k}{Q0zWO`u$ya9VUWw9%)p;b|FA0&F!4nq?H-E)nY!43I6*^8c_k8
zZd&D%<=-gcbU_3N6FiG&eGooj{05dqe%Etld%H(sKR^kJIGH4$+8&+;uz9qz$eRBA
zl|2Y5Xq0f35*&Wj;Va??i=vilIC2I;uOxtz-NUyVUpa;W{@_jrAi>hFgFGyH*y~Rv
zCtzsQXAS-Xaj`d{5bUS}1EDqm84te$NbF#gdN*^^a7g{>;72bN5PorkY~Ix_0{S&B
zoQX5rmhp`E76_3ho?iv9C$5XE_~UA^e(7eyHCb2mKC5lLQvk?wn%V`sQ1~d`(>#$o
znI-Rg&m<5z>C~?>X!H?=Ze+M$@8oUJi(JXD+yN>`UcdAG5Wpy@^MP#I0+mw!myh~3
z9HtyD>ILL&@bC`~VrgVWyJvMYE^Yo}5xfVGG7PPNLH5KW>Ie>GL;d&<<_2HaWu<{#
z*YjMJ5AfF(RwqGg?I@KT&;o`!mf&G15~)KZ1bDtrE7R_Z)RsW{CBt{pg}c$O2df|`
zbz%vnb$gJ~wmx0zeP`(kf-MsendJu%FIfRXOvUADXpk=Y1spq8_<h9>&L>_WzRo!k
z7c8r*1AL9AQV3wU%`2~)`DOkq3e!wW@QO6^L&gn!gsvVqdEc&)FJ}G*U=KEU)CFG4
zUppB9tNY878;%qM7)W~R*J08~Dpx$HG4N_d=Q&_OICj$xvv;CZv>~AO0O=enkWv;i
zKgz~ijC0y91gOo?IGvU)7@R%+CRa_<1P@LGLKHL09an|&@WKVb;L6LZt2hRDpq~b1
zfcD%2mdU!7xAq=(gX)MCL_RF0E_nf@4XVZzEbv0txTSDP)z<bS=sfuF8_>tr<Dg<h
z4^JOSK(zpUBg73hp`ujp1pNo1V{_sGhxb}@IFhW|Px;Err)cJ&pIvFycy&#@)Q!)a
zo^h-FaHH~W$!RIbMN~kM@2Zm=_8U5BCVWOTjbZaQ{|3dHi+@HSJW>R_0JlnfAZ<-4
zX$4nd@?rM9Y*Nl3pq*{3H!k0myiq$tCM&;AQoSzdk-Pr`Hf=dv<lTWi1Et>1D=Gi=
zyu=&K$VixrYN^()CI0J?|3-Zb!W<PiSfyc56S}vP<#Ar9=V(L=WxC5)*(_;=x$(g+
z*DMbxi8~0Tyf&Eih)&<Iv*z9ZCcSG1DfPlvY&Z&L&Gy%#6q#!_iN|rfq;N5m6~EGX
z&OC3<LvqsL9ZqgH`FKz^S>Be18)T|YLTQ;x;em~m3-mSp0w_*X4Xkkx_=*d|0P83>
ze4yTe3SHB{ChNP=&h(9)tboIq482J+|N1S^Sa04jk^|xl-Y114udvje#fWkSd-c~x
znF4dOZ5pr1Zs0y)G#@8;7}F>{>;6w<Aeu(`AnFL5xgWPz!`@EW@c=SCErVQ-;!EC@
z4afL(WJ0>0;nR+X`wYN2tzg5Jh%y-(`u$X=`!a(Ke#^O0NfAgoooPNyWebFsMXG45
z-vSHdQ4b+jJ)FKrz%7B_qoA&B8xX_H@~D+4QtS^ONFkp@V@QY~`^n!{GTi=T^BRpJ
zOuW4R#nVFBgWEv3lg}j1pcLnh{6?PlRe0wsEq748B<)2Ld&5PLCI14owTDYOKawZu
z*e4w!19cT4^guq>pxwv4$1dXr+(Fb>D$+ctdvM&xd-;o_PY@Q|d;fC~&In|pnXo&6
z<CDw2;@3+5R+)Yb!xx`MeOEfwPrnjifqQhE7iodRtq>jnl~~~jyGoKkC`2J=k36J#
zdZbB>DN)fVI7QzXaz5+FTj!urME;}?Dr*DcW3Ng^;(p>puL7yw`^s+*HEjTa>#1MK
zqIk^=I0}}nl{QWbwo+2Rp~I;F!F7Qi`gV|{;c+IVi|E|Xjt-!%+`umxL6=9asFT!H
z4vgU6&7U*2BxH4cbtjePL+sk#)d%|byPwuB<!?1-C{y6<`A)oObNQgAZ6<c=QVQGX
zpvW4Eqo|=_&qI5J_0klwe^>ynaMWMD705XqM=Q5Ghs@T0zWU?OAe|KW?vX)=Bp7YX
zCxx%$Fxv3(E&`~t0*9a#vCY1ZHa=H8WW*gCPL^fOI}pVbz0J_}0C_|L9AHyzFu~jx
z<g7Z5EpDb3=Tnug1jk|KT{I7E$!bcGzm2D;$uzQQy>kkBVnRECw=LHva{6ZT$9p*S
zrjR&RZq5a{53682yC9LjQkEDIUv*asqT@$mq9fn@KKG81*}3v)<YWSGj?)3JA>mH~
zjOaYhDf09+-<i#7S8xR=feSQWg0PpD6Uku2EI>QFnzWAzH7FAK#|_qFbNV~O@0g*-
zg@3pT1AU61A|p~0IkfQ_9p`!@^Ajp_oxn9A(E`n3<I|J@G05MwKT&Rku&9xP1RbZO
z5ll^Bk6EBY^5UYa6Gd5FzMiVB5FzVm8Q2-QFdFF8K$N=ab#coV;7;J+z6c*QfHNei
zMcrDdri~(mroN)iLS_^?$nTs^!@NiUJ|ZN$K5Z%I=<%!_sG8VhnThmrGgPqiJU&5#
zTy^H$^zKxNCYy>|(Q^xBe#aC%7Omy`?n2bbNzf^%Qdr+tm3#-CA12`V)ov?`@ck&z
z4`c$2&I1V?e|B@Zns4-~6pqx`90xwK=Y`4by+0m26%QI&`WL1(Jr6lQn)B~D81OLS
zQOIY-2ON7b+pC_LF&)9Im(w{CGETwKP@L*W^{pfC<7S+F9NIz}i?4(3(<J~DlD`ox
zf%GC+v1R)9?kmPRE-iIEZ>&aaegMIJk4{{6RI<luZ?K{oeM8&#V}S!?ae93nLjL*`
zhTe=rycAJchgDi-9D_`WE@SX0m$(l_{B^MX-b4vbM2GborASyZcNLgq7E8~$$k5+M
zgS&-wH25&E*3ni_Ia(((o}4(<6_fV1&VhMl*!`vUB$Y(w<3vR7qYQs#f?uGZyaQVQ
zj+9Kwp&G%8TC@$3%Wu5m663a4V++0ce|%$5&SCYSb*pd4@-LI<)jv#%c1$LBlnqv_
z{uoP=#B8Yw3DziPr<L6-jw#*6q^+ivb+8g3gkoLG_NE4F=@#8!cr=wLhkkS%zRbW4
z493Y!W*zOFmO%OFc*O9KO_^CmF#-$8EeE)4r(~Q7hQYI)l0Vti0BAM%h&orVF{&jr
z0CFbwVG5Q~>~;c!LGB^nGIPdkjwA{@I)7Zk%`3Ah6rEEf-VttwC>G^RH5~@or?L}n
z#^_SJdhvw?Hz<__Lt&KSiQKg7^=_Fog%Si-IMq1c(lMJ=#F9E+qo^+xI_2zBD$Oza
z2xfTZ$TKOk-_`wTEEp<!@(j2cIt!*SGfOfp5k{Gp)kYgt=hON+?%5t(J3^`BJdzoH
z6jRQCPUTdb3hnJn!A`$Z+4?mlg5hX5ifl#bIzcvEE;ccZ0nBQ4E%Gv*Y+*WRj=s{?
zd)FK0$T=oFN}f@gp8Y;y%7K4ZxQzVo*Q7UqKzI9>gzR>~*Pzc&?s^j?dE{stcC2S8
zbBMjP1v2qdgLLM`QCRt><h_SoVoqYUnv_I#cY(8MyPyo4I_pT{zI0z(UP2j>n_*Ep
zj$yWTsx`ZHpvht>En2Z*4wHp{Xm2_!J))F2?pn2}q)$~<kK>}(B_9rHabxkGy;L}g
zmpoiMa^T7w&x*}>Go4BQVccQOFy%gsYiQfco5XkVubR*Z@#7g%!~YubgZS*B(mQ<|
z>4>&L`JZ&(tyZNyH8CfnQ{Nj;Qg2s~35ejkCpnl=ax45&DKxfe%ZhKeYu$QiOoyz9
zML4GDY}?2_v8s8ke8AI%Sp8;ush{c=2uRV0EIMAdF!BehRm)4@x8jpX(7~EiLNS81
zY@i~<C3m8ielt&!@HzET{akXT%;+tu*<_}vk-aU>IenU*@e8d^Mm2Zvqpp1I^>f0f
z2{DveQ2j!mvA^OuD)T~`_xBYjW>BH!I$_ompgAPi4h}Di>b?%j&p|wz{4}|yxhUX2
z%Gi;XU<)d6)?Y-RqWEKOpnn~-6#n9=4VK>6<DFwkX^u!MC~yAA{9^Zrx~ik8-<u~w
z44qZX69G+66r?TWbEDE+frD-*zktiA2de`J<=0!a_N}!PJC4RKbd_A13-O6a;z6u$
zdObOJq>BridG@B-D>9L(As6wr|92+0T8kUpR~`iZTEFex&Tp3Qy&jFub!WuC#}Trb
zzb|TZ>ma8wgPNs^m%jQPBu&~tcsEdB@Fs{NJ0=h!R-E9xP-k1s-Gl5+hWD=m)5w<x
zZgG<DxXA+7DDUa4k?y>Nq9@(azZYV|gok!BjRqrH_e&cyqQf)uoeZ^=-vwrzdI{JY
zW#du3G}^K`7Y`j${bT7xxN@|XOvH3;k)OS`mTZ1B1hsMJed*qh3uD*EYG{HH5nt_$
zfBUOVo<0q)r%m(epZt4>1|HV2#1lx-I-gJn&G^&jH2nDLuHm#vrruZ%iM(Uf{(G72
z1<Qb@dGzJ;V9OTKdN^=OmV)LDJaP{hwLlVyuZW~nxD1U7FW`JZ!;~pCDpSQ|LX|la
zz7RA?j7$<aTlPCgYLP$m!=<uQ*=fSoT++lo7|$itFno~z$3IC>Q-(P7Eu7N6$czPp
zx^(c1cx`k1LYH7&P4_UvhPkLcky|Yi6I{80s5{`G<6x*AoRaKF(C<VQqnB=<)DJb_
z@V>D3#|%%Qx&=5WUB%Zzn;A)jDENHm`A~@n{2%U05#>&&JId8?ltdzLM}rJMi|V;5
zT1pb_N9?~QAWtU2vCcbYRo3ff_$zE``~}+s8l3(VE?_yCgjipAOq`}xbnrydGG=M?
zdw?E0j*#y}^|(GCV7o7T;^%m!osAx<EF!IqrxSy!7>uL}hBf}l^RG)vdlL^O)SFi7
zyxs=M>dKsBACaek^ih+zPLEycF+BM&6Dm2&8Z@?-nc4icZKj4U9%1h6hCF7(g@msm
zmYwQonNgi22!M|8Sl3JPGYqBSZA9R}vnbdAn5nepV6llxl)A0|3uVX4{#T*rL0oNm
zGVKfVKlbcZDP9B8#Dgy1nG^jFEYx_HF!zxoKHiZVU@66)owyQx`-#`LdxNQI5a=b@
z{mtWhcu?IOBd&p#hnuGcCH{dwUarp#=Z<ucGL**^dsFUGXt+lm=8~s*#h^v^rF-oc
z3Cejg`X}A^;}8<pp`i8~VUZ5!&aI}H1U6kS@{gCP-kW=}{Hb8r0XxgVLM<21YhEjU
zt0_#{LE-yOyRUkcc9HZ(QMS5-f!(Q`gnh5bjv*-oE446TQ~nSvSVsktX-o2D)h8BT
z0E6*A*_y;;KefB6ST2#5JXhr>Tcs{>(oT#g)Cjc2SCG22PIA}F6<WIeGbGYK{t_nS
z4XmM;Lq?ivN4aNF%NcSaALF|P&v>(!&&>NRy*pf_afk9)Ch#a*GLktC+aq0fz~Yp%
z*4f)Z*eeZA3ia+R78_wwnh-IPwhCD<{|XnOjee@roX$T-^OImD5HWoiu`~sI|6BC>
z0DB}wDic3aqL(^zJ$m37i6?Z_ZpI)e*lB$)6`l@**Hmb9?U^J!?`~?2_x1yEhfy_$
zb~T4Q&x3NZYG!^y93i*z%>-wAkbs0n>q!K&XLhGf>Z(wsZ|iG=dcwQ2{sv{Sbay2b
z@66DlYIVcxtOB%`2K=A0)-gx3wn=O(baGA5K!>rvZQjG@Fo%qdMRpTgs6c6=gY>(+
zh2GvMf9BMn1X8?Mi)p?4(cS&OMgh;(gw}mF#(EY>LMt0fLnOgTrK~R_H$?SCT7!mF
zQ`8RU@k5l=XoFV?G%~)?ns#c@REFF4&CB<N<#~DTLjw-OKpP<FsI%pu<ml))sgqGy
znC7+DUIxs}NbkpZ5=~N?C^GVGy4z}=*_3V2__9#ras&S8_TpGP|3|s5Kwx=cPh_YW
zw%^ZrmoAjZUXAYrN)4Yd0MMlPs$Bd#t<rHYhOj9%YvS+Wet`qMcv@RH#Hl@g8^hFq
z@wJYd<rFG#&hBFk*qO=MCM8mb5%aLlaMlE@Sfi*MmC9>?xT+-)%@mJK8`vZrVYB3L
z@4S6A3dL5EknW!gV0a~!m&OAOj-BvK&L|7s-TW-yVEw)8_NE>yq5}OGsTENAL8;ky
zW09VP6@&;^#wTJpGf2&4Yht3tT1UBq)$&96?27_>K4+kM#V`)Chd>9|V2262k0pDQ
zE_;8*j<~{wHPwrmY~wNlw)S|1i!+5U`&%<(z3u+eD`2oSCfy|}NT{4`v`wxD7LW%k
zQ`|N14*VQqcPv9rM}xid?riWb(iF$uJ79Q5YKCG=U>>U0Bqv4?|4`wsKWMUM{JtPO
z-4ccgk8>g@d4LQ(Q?9O6a~vIxj=d;1eK>PhKxh{K)sMqp(-9q%R+ThN;#9wFw#Pp@
zyQxEgC990L@iQP$Q6N}j>08a0tj$t|KiO_|{bMnm^mD~YX@^Vj>?x?iD(icFuKyy~
zt+4@$Q5b1}TEc?qx^lc$A8mUE_v4(LCQB|MWj2B_Z=FE2Kr(<(BNEEp?-%TDF|0a3
z@J3Bg?zG4xe#@^o5^Dg*lNa<EJOoMi#B`Gf@MePuR+`R+&de}VMa&Wn0`<wQ0)Yfd
z_!A~Mx0QyD=8V(>pK@ZOHzWhr!KU7({kokH`Dr1Mq=0>Wat;%}$vc}1&6pS>DJ~ls
zfz5&e3#Yk4gETXY&H4{!1LSZq)C#CKVTPSsYpRmp9}E7fq_h7(hxp-3R!>%p$FrZ4
zAtJ*`19~%|o|uZwyg*Ia<dLsla~&wcHBw`h6Zr{r29PQnmarV@9et%;rqW@vyA(^T
z2E*DiQ_3O-Mh)8W92i3x7@RG#0=U?2o?pFUdjZK8*L$jl2V{}(nwUka4C=X0)9C#h
zwjXQ*Qe%hvS(G28l1B#lmF3=q3P{=;RSEU+)L5rJIml{11{E&E8}>$4r&N|C`+#nE
zo_v?%S%qswq<=)8c=63BoX^5`GDZ71D`l&R@QL~6bu7p5ifMrGOi-Jy)V>kp(Z%+7
z=rz)z5mX4OYDy_MgOjV9nw<g*=kD_`61ZR`WanwGl-4UL@Twj6U~lc?@^lzr;gm{s
zoOoC%tlN0RHEc@$D&R6K4xQY)c3!WY+-B$G$=w>5IAAyL)Pyl#ppm;4c5|>vojSc!
zBd>>BI%}Tt`*z06Y3^ta$wv4iZhw^7sY9#6_w|<OmgTMLgo=wa-5qChfw004JfnDD
zhXGX?<H)`~C|idJUo{SChf$?;>}s?H`Agda3VqC5JFpx7_3KfrXqEiLfY!Sxycb31
zvu|jVykS?k@ibN&!(>!@(gx+jWS_P86D}3nHsS-vqGr}Z-jmDSY#hvshp>JBR(-54
zDcqM!$=0lTRedvQmBg>C`ptT5B4IDbW=1aApTFmnN+ZPBbqj#&dEv-&x4Diq1#Q<G
ziss@vIPy*t)Z2D_)eKfjpT^S^gysBbkk9d|x|xKi)q-(eJMYlGHpa+v%9I3%xVJ}J
z0hRq%dSj2W$I`7+#caam6qlfFE?_@7<$Zci-#=5cD!;$$+UupMRnuCKIc!CHH`pN&
zT1+QqjXzKvteonK4=d=CN)oOtc#%yhf$FWXVL7S1d_en^jas46nDIkVz}VJYf1m$`
z!GTHfG*IgbdVW2zUMUUv^UA9a`qG@t<U-)Uxv=g&R>A#`$1nXqhA#}{eSZq#p3wf`
z9PIwAu~!{@xKP@e70|jL%{zN}rI1oh5Fe}noyJh-7=jQ7DIB`bXLv$4<R2Uvn5>kt
zz2#>`pc=t==-&QKU(;W>E~OP9d7GSwa5btNS*JPLQ{J2ZkbFBD-c0kSh0_v6fK;eI
zn+vgv$wAKsbgoXz{I`a?usl=7D!W4=J?pfIW)={Ys4dXo>`5!Gr#C`Rw#^8J;-nVK
z=cgrJQnNk1=yVJ7%#KVH{l#+dtw)6K_vc5^VALb$Ut`zC(AD-TH-37;+3HYto~^lV
zP2t0V#z>UyTVC^OOf4K-osB1jXO{4r0Nw*=#Q-y09olo*y196UU9+u0aa?Y{Q^>6Q
ztT}iPCzV6t&AQQnIHu4G=K&RmuZpM2&(xV=v7y1A1@GzRpw|XeSmcFKNShzjbQ~v*
zfg24f_<ynY)=^QeZTqk&5~5O4(kLM{fHVjwAR;9y9YZ)E(w)Nyf|P>NAt513NDtj5
zAxKDfcQ*rk*X{Os-lyLE?%%t<KfZ6R-&(Ww!eKTubKlo>Ugvq7$9Wt^O1I6wl=g_;
zJpzqIu4=)~tngxFp|%D~p2dV-VJmJYmrp@RB*~msey~U_C1L+J9mbdTxjeM>sua&D
zW*i&*iS>n^OS0^2<<ANmcP*4@u}%qSkCJO0+|Dx@Er~f;sc#s0{-r+r4hNor{D*Ot
zN45%M2R+Ukx+B5U5?Cm8q31vI*m0(j$9N-AwKuQq8NQ9eIw9Lc%>OhR5f7>HE(|r9
zi@K$-buYm`Oma<cL`cVRRr4S-d7LYZuysVD&`!D6^2j912{8&_ow>}FnoFNpaxHT;
zBk%Z)ukLBE@Z{_YrLyjcWcPyMgSpv{P-yJQK*dzOHV^+o`T>p}M~pN)<ZoAd437>{
z6gz&@KfiIyc<CDX#+L5ju?lRrd=OJ~TVi^Ik28ZkB!8E^O{!W}3fnF2W#rE4>Xzz~
zj1)HXNV?_4ckkkHY&ZQ3=xO%y61U#8p|X983|OA)bR{yITM0s{SgZfBbw6sAwb7Y7
zlVmH#v&E@*q|>Y_*d{(GgiR5oJhSr)@MhBB{}OW*e$(Wwp-YWPeqGkKOipC6A`vit
z4-e3~wXjXIRE@kfoIB#6RSQ?+Y>`T@3p0tacIL}{gXcGE>1t*GpQd05;JdA$tUrf1
zO06Rpz}hlUryp`_2@8fP+fY=FJn%ytDzelU!1`PFG}Lxt79E`OJGaJWvU8Y`#Q0sW
zx3N)bvnh&+T58w3FN+alMnOBOqa|zhrb+e<T0P4|gofc@n|C$)5H1yWPk--AwcH-F
zYMsWDbnhAWYpK4~;&PG-qsCNYfbe;X*SzXVZSq8C7J<O3Pyj(K4(xjO*Fam7LE&hd
z^wr!37kEe3n42d4=@C=fG>DL^M3g|{S<+mFS2}YT5=M8TY8c8+h`MUpm-NE+Y0)c#
zKy!Y3$YOLqo%3usYI^LhB<&fX7ksNyUL2fWt$GwZty%DOo}M`dQB_u<2W&05EsQL}
z=S$9F$*2jXX?Zy$9BVcARzb7FaWX<{c5hZ_`TPK36PEu}cW~Z?<h&gk!sKhMqd+C&
z75Q56aaIh4YPoGqfLnw3ORcwALETxqH%;zzCmL#hiQ1fT>LKdPG@?%AIAFl?x<-G@
z*GWP<&EF7ZRtqO{^gd{7Z*17v9EoE+2x=ZANAg5pt*_F;Y~`_RU;H4A458OlT_+BX
z@%^@9cB-rD82`NF0X?0q)e3O=`pOkyroLA3H8!G7qT}?;@B3rtx&p_sf@Cef{ZbK`
zI>3^iXP<nN+nwbzANY>S{E12sHfqON+<t?10dV!MhqgBR<h_k$u6R8yOSg)5>Yrj!
zts_Wpo8T`>M2qA?S~*p<t4G;<%-7m~b2rYaZBJvhFCd>Yi0_FYz3e7>YFr-v1g+fQ
z5V7iP3=bp2G#(7@2@ulPJK3ED%sWl%(@d{946hw<BIA{{!nr~+mefRHLljuV3wq~C
zFuPWRY>3|{-Vau<1DlZmUrClNBHc|#80iJ1NS!HdShc@kAZK}M{cWuG6u6|PiX~*o
zfQK|6sIxykyNFVoNwK)zeVFA|-Gd*dyd*<rX37*maTR^j+|VJMO8r7C{;9<vNn%YQ
z^N!v0B=h-xwxj%aVN=g=VWOh8Xoo}s?L(%jZ>mC;w}`EAHMtbZ^Sy!QQ7DTKt74Y=
z;!wN{$Bnwu0iTv$4$g?#0F@2n$3j&LZOi5v-^#gmXAq-N2+^;y)vXHoh0|%b4r|=b
zIi`Q(RaF>&!`z+)bUW0#Oig-v*Q~dK1@L7)C_XjS+K9`ki_$d0hxLeqU3o}3zTd@J
zE=)xUmqLWhpP*-RNpwuO_Ae@5?hFzilH>5MRz7Zb$=*T#ks6yvEt+?m!OG#E#<|?j
zZpH1noPw@)B*fX+G+iTC2<^?Tpa@poXb!T+&UZ!L_B38equsaHsmoituJ-13Thm>g
zXCDF^Q!93wvF$q(5RqCRe6dmQ1JU3fs1X||w%nE%Lk*f@oI02}&!=0J(4Q>GS$+Jh
zHKJKEH?{>hMS=u#&6J5ZNIvE?-P6(j(L3LAhdgPNqScnB4c8gcfV8s89q@a=(m4uR
zavd?I+a^~t_uG$oSi>AarD^+$%sF9@5;v`lP?Z>Q1j<czYT31Qr$m8mCNrU{P4C1r
zdC@KQGk)ST`1M<L)BRT01E&@nBqcN$^7CgcsIlW+ZPis-{e?Oj9kw<RiW)z?C6{u!
z9ODYNvwGf&5^wh}2?k17v<&&|HPKrHHjQUDv~4>bQF(xlt(~DO$9xmds`$L^RrNZy
z3v-)Wa}Ia~m0HdK=S?vFNy!;G6x!wLH%h2FnH<Eubnn@kadwIFMO6G=Ibz#uSHD%_
z&U}!s%aJa&=c1kLk2s<9rwwTqsGnBxZ0>~1>JoudmW}ka@N%TmpeS*V8hZRKQYpf_
z+G$Oam%j4HdQ)M8W%+c03*E5tTH^-|w0K5_MyjkbI1~2qA8hm=BT}ygzQpqf4=HXe
zx+&t8&7{t#*hRl1m-C*S*0l<~71QDlJRe;fh~-LznPHbnc@^olBanMR*81mjpI(Ov
z!$AZak*A)TLDT~;*TfkuJOd+dd7(&SsfqVW7y4NfD>wl<ly<4WuyrmakY8pFQ9Q$S
z;>Y99H|J|VB-!?Ic}fGV9Ihk@E4{_4kMj)vCQM@KJn2F>t9NyD`if%@c}=(LTiZu-
zc)Ed0f}44q2CCY^%A<ANQtIK2yfkpi+ak4mH+&S*Uro;tE}PpAbPE@Htx%v<H=^zk
za$NJco3Lk^Xm|e&=}D%MytdYTGhh_dVmYr%=tV^n$fA!Qc4Ex;4mMqIsta4jt55w7
zy7g*paD7VsURr=CTro6wdoGySEBBI|>pPRvfoed#MII6@U7#s&gt=!cqi%aO@d?ko
zI%QuM<w;%Z(#X5{bm+Z4i{9rmMJp56`T#-TI>?sep8Ktg{U|dZX(Ff;kEuG@8?FTN
zyqYu&;~R#@0H`?JdT3i3D_%?A&lyMP6194}?Gs~u&?=j9s4<f9sL}Fr+~^u;3yxJ4
zoLrk%ZPvURlJ?n*nlr@@mB2A<N@A*LK7zI<%uGhwjPfCs=%SQQT6fA3Ym)Jwgx19)
zmvjugF50bo`J{USa)zsC&iBAE5xR@0ozaH+wpSKsNI*C^Cb%zC^M<(xI{V0M4Nm+w
z8-P^(#K}|9x4j};g%H&=I@YOa;2cVXm<WsbbKtoX%W~ABr~iplC##2$E(jUVD}}iz
z%_GSp$)4dm&C%fGoA3MDT~pvMxF=>e^MKxd-gZ(exRla|_f2iw31hP18tZcIkM`TT
zx^={v-Qv!6k~1mc3zVdGFvRR@bzQAV{Ql?A=HwL6nRzZJZcWa`80V~CZuTuG<%wj%
z6E?q9C7DLeG%^j@_z6*jFKdHp(L_i-#BK90L-fBD?b>QQgBe^y@`UVUTO-@fG+QI`
zzaLMqz&w|@JGI*nc?3NH`<--d1fq)Ur;NlQ-^EdwdRV=+l!Pe`g;aUXV;#|C_41<b
zyM2z^xieyq7lsz5>TC`{%c1(R5~nmVl#9Sj%0FFb)QSCIIdwdniQdLYIgs#fO8wIH
zU93$pL|{t$mu*c3>z@&;IYZf`h;>8RL-x3kWEY)Hxs85h_!HT9|3&uU^`@#%4{PCK
z`-#yZ`U}p++EeJ!O)c>dOV)j(4vr^uYJD8tOst?VIL5wtuzbSz_!3C+e&uyu=LpyQ
zC8mO&Rwd0b-#YDT<l&SLe2(A7uILHuyfx8Y0i<-FqZvX-xh=Guwm^K@lBcTMcr3Ks
z6imCUP|*&GW{H-LXN07J7+$Sc%$zKgM5cz{Z*?--8HGl-s8wNt%^1E0eb4U)w#M(W
zz-Y=TLKKJ~-+G7keELV72VXsU4{r^?G*&};YOGc+0mLT*k5A3=R5{o49g|1-=Y8Qo
z6Xm2$hx0ZA(C&StKVt~wzjo!>mM*}~e<PTp>rQl1G~!j|10!BXGS9M9Af>M6yT~^v
zS4Xj?Mg{yxs=7;vthinn=cX<oShN7&<_L&v$m>!cNpYa-ceD>IUV4{Um^7;{o7%m@
zZ}X#b`|QL$TZhelo|-e9cQMq{ut|n%m3!2rMUvD(Q{#aSPKGx#zOg)0L#02tOAE;`
zrA`&mTv?$?me7nyhkU=}9<j0ECfjo2NoeeNII%d91CfNem$#d@WR?~QdyeuuoJeQJ
z_HG5(YGl#_p*HkHv)wz*gO|o-mNT>IZ|Gr}3pqsFyu^_v0R~H8R=|Gqk|7c187=4M
z+Z)EoVrb$iL-}@~G1PHvQDp&ZmHP#A-Hyvq_B3`SP~ZgJmiX~k2SbUoab^V-HbAxy
zwLbNYsTLmNVPlmuBIi0tJgmC1y||nVAK9{F=`_6Mdn`_FL#*;R>cCh_Xzk-pil);?
z%MGI>9i&N&yJova%$}+moV4e$lMZ@(b>{Sm;P=68Vxeuj1b6c3(t&YX6%!IFEJ4d?
z#OicgngflkqIV4iM?r<plxM4gY*TtgR_fWDzV^-T&qg1#1g{R!?ZcY5H9L~N)}Eu=
zxdEo=stCICl4ZoGjmZg0P04mA&F({)7LivK9z*wp$cIh17z(%{YjQ0CThPL6`_il1
z=LQd7Y5OocO_99X`6zpJIe#ZzcK>zqHclC;92QP>qwbdF;;pu%!7X?3C!35U^clOm
z%^rlEVeOl#w9}5=kPvd2&^=GjZO#UfO%clFCmuKZGcxbxc3IK12h^s<+<;$D+K5UJ
zGmjJsRs$z@de(-Ia0WSs_tc`_F-5sB?zMOquCZ#uS>T*?cmG7`{%$|qxy`K|DI4^;
z7jYVWJ0;seXG1R5rLTTUSMfsUIS-;1IduD59^7~#&rcyKa4kb$vZIyE=-Z!iApUMQ
z<)wWwGK^#Qx_U&2K%eFfEz6>x;ws^=?O^Z)MsVbGEnc~9AJ8$(gM->cFc3Q5bKCXB
zPRcDLx^0WfJH?&EkV<Nf-VRNC>%I9e^x<LOQ3brRq88Umi|j#0yMM2mx`(8AHMn!i
z)j|xR`bj(^OJlL(IUF41&d@<#jz|d)_<fKwV)@(yemSKo9S!o=^-GMtu!-td@VQcl
z6nxh)iJ*w|V5X|?Fzjp9WH$m7lP!BN7%sa8=HTIRS#SIOb+Dy5O=|>*@;AJr1MXq#
z9c+4sQ&cSOG}=!`!nR@zn8Zoz@!W|TY><(RQVjfnh;wXPX93tnn@`8p--qOzeO7pc
z3)(&`*4da^UuNFq%zWhrVX;xBp=aK{_^eCkcD-PaS|}U?U{=%lsRxV;4d>PR^nm`l
zzO!)Dk?aSb^x#dBOST%%A~cEvLYTeeFPPnZ+JManTLmWEd^tLbO$?&!b`>*qt=Csy
z2|uULv$C`FH1=XzCGu*-Z;zJH_mC+Dj5+tRfZJ$%mc=O%p1`_<49%@%1dzPkTIS<@
zqj{xGDVOzap(@K2VgdxMcj17bv1hDhfh8&NV?pt{&O+mRoV1@lD?g&!k=67~5hv!g
z1VEEM{kLg(*7zCHTGz0K#pQ^>^sIx>gfi(^`3vL^iPco)*yLEc7gYjjEQR)Bc1Cxi
z6bnh|=Wpn@S1GnNsr>?=+CO~>8y}G*I_#>pho@9SlFctps8>@TY&7qRewT|tnxvcg
zS3vXhWk-ot1dGTew}cesEs}^b>EruP6(uQ_Z@_F?tVO?C?ZhC~3>Q5dha@sq%)1vC
z9jX<p+XTc)&m3a=lD9Qa-&O^lzCDr_OFPpui9u8lHiaBTl2@ir2YO3lZ;#A}@C0>N
zRLmySRnnT@eS=HnDrEmFPxoS=o)mTclru$l{n)aJ8HlPo*&;1WeI})Ge;xa53I?E~
zdV0BWTR{w^dbwto5R5URJKpxU6#Re-N_xC`Y4eIOHE;yTB(Bx@uE^b!ox}}-wshO2
z(~=NgPZ^t@R{Boy!Ki&eWvdJ^wWpbO<2+{K`}hGI+g4{KA?Aog(zpm6GVixe(+jV^
zC5`g)8+WheDa+uf_2PGHsk_{xA;H{b9B(D<M0v0Uyvu<#3x5z{84}$##B(}0h;S9b
zeg*kK2W=clZ^&uOi$#(X`|%PMf(g8%0-VJgS3I&5q6_P_=}+)GnX-4(d<;8s$@&vP
z^|e(i=iMH*arfN*;{hI1hz-q>nUx>7Z7e~Ck{Egc9pN_*6g(1gX+Z(y=*`DHydFs$
zjab%X-A|;H6IyHJT~mhkXLtW%xNOYU=Y0LN?PMf~f@91)Sn>tb)gOLk*<g#-oMWn4
zJ%+{%1+nxMij?wk+<B>2f~sN@D@QKxg;wtk>kr|?EqpmqUatVyel5$+D<vFk_b)^%
zvsOIIMzx&z*l`0_1Q!#pj^$dz(68_ly|qa2&^yYxF!ooK*n8WnY0slUo=4r$;s%Y1
z)mGiRWHQJ$5EWm`S&i+UN27R-M6rYjbYU#C3?;u;UAiJT81|Ma{JR{P+4BcagXmbU
z;{#VeYWin8uCJ9|ySg^N9Q{lRz8*R}QEAH$*dVe;lQHBdgiptsf2Fb@xizrGSj+NB
z{Dr$J-HPwiHH4xJQTxNkJdCmZ@=H91UlI>u3_3c^)r*xL5Uos#56GnV(~}msjpnuR
z<iPhHOY9GssdSr@2YTbHdTV{Z#IkUy*Sx@nVymN=g1Xknj_#smH};)cmKjX3*7dC;
zr8ogz^Hsh=`nX&I7AFFhvg|{R#!ykCwqO^D_N{m6eu52&0hF(}avqnPIQiM}bKKWj
z9~oz1xy%zBEO@cFFa^$g_wn~O{cEmgq|2TKqQokEs)_r~I|5xXlswTcH-FHlolI}C
z7x!|I7P!!+j^|P6)at);iBGQ3S-dSW&iU0vj8Pn60IT!X60^MZMQuj|aoiqBGGxR+
zDtz$>z3`ck#UI>7(P)9_h4^wyA^OpU+rvuI?RdyAt=s@fg5r^|xKRabVDn^<e8W1E
z#edhG)pA&W<p4Pg+-@>P;bmQ>k)?i!+e*|{#Yl(DoR&}7@#Bc2QI~k9b;->NcJ#>l
z!f*+EP~k>(h~0Wxb#kK6B%CMHsn+LHX;ezbpn|w$h`W`;&Ui`kwo%Cr`WKkRrqc8U
z(C(5#qQ3Ke-9OA6(sTqf(^pwuTo&YAZ(*?>nJ!1d9^UST8Uho{HFqLyABzQ>UQ76x
z64c4E=?vF7f;Abci5)YvCC)2akO%_Bu-6NV=Bo(n63|n>FukfyGruA9<5>4Ni8VSc
z)1YtRf^lip2d%cPd5bkA$b`~r17M(n(Gz2dw+pb^lU;HN+r0y@ALa#3L(?aG(P>tR
z=?t*;JA;0TwGA(8B_c1eCb$^!Fv<v0l~&_hbI2ZulwxUbSO{G`e$X^ur<e<;6zgQ)
zC=4;ZRG&%+PV}8p#6HkUT*(Cn$vC8D=n=qxB7H;QYQ?MY<kVwWUVlZJAg8$f33;h{
zTfIa#0lM%RrNE1)s!vN#!SXJpcrn#d5+<TGCvEBYiiMGyunQJ@@V$9TL(frd6!S#1
zP^%v34sN?<6ru`{q<PpncbRyi>bLK~zYZRZV|z+ayddwmaNN!@>^D)w=BjVvIfy-N
zc(5@kaK`RL*?Ak+9?;K=uj$2j7H4?yI6p`vWIaM1-JDx_ZrvWoGcKnmWSBm#qK%wm
zABGp{u(sw}sB0p&yExaXf{TXxpQ^Dw&=~YAj;Q)H7|rzkQ`oVJcs0e*Xmn(WrLzZl
zr}6NuAih~uMCHSI*<%2z+)=E-LH<`y=Qq^FlZ3xPCam6(dzDQ{7%%!|NA*MN-DOPe
zd-K_QP>;RPx4JfrV^-#m+Tt>nIup;Ni#OR|78&al5;QkQkA&OOil-FY<D;cUv%ecL
ziorV6rxtyAR<+;VGM^+=*yqHGF;(SgyYN+3y*H2F<UBz!rAbb0Xvv2pB+Wc|`4nF@
zz(g}R%*_Iq3#sLVvMv?Hy;eUgB>763c~9+Uz;m-D&>yyB{!(g~n`J?-Hba*pVWdYA
zG?&epr+8w~yrf?7Txdz=P5Dnt4)uh<DZb5In6=Kc(fIA>{kSwXFN=RByKG^Gi{`*k
zUpg~q|5&@B<2wl-GP3Sqexs9+qnBhmRl2J12xoni_CY2)Jv2SGKMFaFdhA|R->4b%
zGtwVp*|X<Z&6?^820$8gC&uanx72@wj#ls|YViz~h+{p{XxHNc!rOKQtu8@TKAy#}
z{+1Kgra^)eD@BD7w3tNDeBSE6o&z=g*8hG_4DfkkY9F(3h72#Y)ndgBGQP9XpWt~f
z2`0bb$&qQ<v!Kw!p~J#I#nF$s)d=0mAN1Jmfnh;p7KoRkUUaZ9t;ipoyUZqfm1(IT
z0PCKP?)^6~Sdoy6eJtg;TJgxF!lG~OWVS-9H+!0LTZOXWQN;&S$z_Sd8B%zHC#0fb
zf$+yZ)uL#gHF+y~OM7aN^Vn6iJsDNUw{LsPq(FRj7V<7R&acKR+@k6KcFH&)l@zht
zDMnvyA?#U9iaSY6^$vG$o92`g?g1=5DPFfE2ub*`riN9sRq-w-T=B?Xd{YhfBE<5K
zs?>69CkuSs-)?8NbvU2ctRQ5eMR%Q7U3)^dO4#`Wmt)6PVc~J|mOsq^Q=JGHO`#UH
z$@*rSMeUfp;O2UoTALN`CIQm_v9g_c-_D~-Afe<tD>afU`>c5@7jJ1}8rIopY#TIi
z`IK{;_GC(M-9v}*MjqLfiV^%M$8Zh>uN(IS&3y*zxnn2C4Ek2YM`dk)oas7`ItFB4
z?IDs&kQ4Kk{Hlc>m&H7hl`xP;$x=oCEYj|4Kkp8ip_kbJDoKzoXC^>KCBj^K=JUsu
z;~6}(B4u5d*FA;4e+;YolmRp>#$|aP9*cpBcIgOfOAdZl)b9$ly6~q5{Q=5rlAD;#
z(e);`2yTr?$@J{OOX5Xt^z~$x(}(f-Y&vvmKQogtEl^%xj0f_1o!O7Q5IXk{4}QiP
z8PYznsYI@i(YnappDkcq=e?r9&!_ls{I)AJ=RpTD|5+P$C6`>-3MfCh_X?QZag<HM
z9|PcqwT3`!=(f6EG-WYj)O_`W@$UsG9d{0PNL4(%wIQ)4y;F~TMUjD)G0mR}<Un5r
zU0U)j_bdC<r^>>UK9QZPMK7~2ol;6d0{ZgT`Vs+ymwTXIvW(IM2$Jmb7IDREvVI7h
z1Q0{qz*c7fS7OXSsH2s@6-IJYcnk1<em~G%b{aY!S@kCUr={iL0}A{a6rh-7(oUCN
z*}QP-e;P<LzkNCCG!E=yQcmnoCj)1|CC)zO7AdZ69d3{vm|6*e-wq_7ZY26{i);wf
z<S(54w2V#6tLrp8UPsD_k~Q&18k~K*5z>p9;`Kqwehn18I|Atca3clr@{95f391u*
zKSgw2@Ix%^(9-#YayF^}uDqq)C17anY{+neeI5;I*i1L(OCGjy5K7?5q+Z`PS9s=r
zC%Ae#(n)3@YuP0Jn||hk&}wreLNPfh@|5vY99LbK`bZrC>;uFUYXNQ4&_^M+aAn-k
z4vzVt+;>Y1@;3$G<L1HklHjZywLDz14>l!#Ou;Mr;dS)$S62i#Jbky`Aiskud4}nR
z#6)3Q4Jl`>2`%#ad^Zj;wrUmcRrE#)ScSSzA5b>D$}UmAh{{vp8U{5d;Spj<a_TN;
zxvXD=GUia)L2ssYC#FH`lk(HXd^YQo!S%>h^Ka51Z!H19pm;g};Q?oVRQnkXt_>mJ
z;p#P3SMbV0EybO3o;SgSj7H;v0JF1x0@yb2?;)P)-y#rqD-dldt!`@JDu+QY@pi6)
zkmQ!m8BPMhf&caJlEBNgfWgwH=+((xc!j}K%~Ah4^268qJNM7EaaU}-oo#8uX$|Ls
zvAK4!0ZfZ07jMJcGVs=?&y_?5i}#0wJqMY~aMU}(wSWCluM9judt|wrTi-Sf!i8jv
z*{2wf%OyZ@s+jW`BbNuwJsN+y0ev7=qsE*!K5eG9(X8-%E%QyxH2pm)pYSfm=-L+=
z;vMMcTAiSiJ`#nxd71o(XpZ5B^f@2;>8)QAUZ*bcC9;<v40NT)$-&0g0C{DtvqYkk
zQK~MVavr@^48&~&Wu6EIk2(!!PY(H0cV9JswYa0|e7KPX>rbpB?al{Dc7NLK__>`^
z!2IpzUj(&Eq;KlgE^gux!}o-DnWyh1j1xyPD6`v%;~J3#ZxiQn=6_>K5bDqa#30?B
z)`3tDMBA%VUz<ho`Bl<6@CnQx+!I=ssc*a_PIM&DR{xMA`>BmdrBL727eAX+@#o{>
zsQL@Gkc<p4(7v9@sbEz9_&Hs*xMX*GvxHVN-x{R08WbQL;+e?omiIvJq-GtmErfL@
z+z~px|I%lJWzW)nKrYcnO1$GxH%^=uh1o16S1=Y|w9EvJ)8kS0Eib7AanyF)xT0xu
zR7r>>r;4!rL_U(m7Br^S@^5l!^(UTK4OX#yssx)$7jZla1awE%4|j+QH8dZ0Qc{1g
zZK7%c_Av9+n{C0jTE|dsAIIO$dQHSu`FN0@tbzaz@H;eT5T#+v77*%MQd10px<Qo+
z2IoUfUi<+v|JSMleLL6ld%lQVScLgvVN1I~*qI?1;35#*)Y<Ous9`XG%lH8eV&MUb
z!y@eARQFs8V#C^t6j_I6a@aBk6GS{5w6$5-7mwL18M~nQ)*+`0WUf$5t@&}&ATr4t
zzrkDU8UYI*KFbt7Bn!o}sof?u21ery+9EF=``-k0!AkYnX>HbnA`eg4#hI~EaPHt1
zyd+mV_avHMe|`iA-qdFO_*pkVuxQ{NTUHVGoFh6nnY>AV4A<RD?>K&JG64x95H6{t
zhyv*&9MbUv`^bQ=-bmdDP{gTF@g=ZcVJZ7U3poqiQPwfyixw!3MY%O+x#KR~<FEou
z3IdijGdyjUVBjSs5H0nETR8!y^g(_5wz+z3B!833>*JUM#w;NFS~M*y#`1!|B!FX}
z8!da6eOD;Rrj|2tmnvNgxm=*w)`w4G*f-^kuXSNDpIN<fMR8%)`0h|9zDC@qpE9UP
zkT_fgrtFV+u<r9PzDT0W0dsAJ$pI`|r>`7L+UjMl>g%5!^;gF>sp#o}Nzn7VcUA$b
zqPyP=t)d5F`0#nx7JD0hf^=8E>zL<pdv}$S;(3(QF|K3!_TL3w;k$Wr(NZOVew`$&
zjYI`Z^ctO5mq@i<__T9D3zGK2A$^W>#7>&$67K>=>CSJ96q=Eq`mTHKm+OBWv@fUz
zQ1V3>yg=$YaQvcZXNs8*a4H9<ko$*0`#{e)O_At#!qKuV$qZ}bZ#|0@CEC3;dCZMF
zf<1XxA^wv!3@Sv1%B`vu6Xj(POEN;5pKkwN$MIHsZM^}MAjyt8-GfZRlaSr$1^d$R
zypTks?nG4U-xu-a!MzCjru#eeu3vNPr$w?yg2cE&T&|$idNUn7ezlnIKrLqd7&h78
zdH3I_Rsa4^X&<R;m`X5>R2y07xS60en?bsHh4|V9!8klEq9j?vaIT1v9Per@0%d#(
zuT7bAwO8qHLc`<<=Nl4t%Iz-ry*wKAdZPaD%O4#)XDDN!{TitDD}(=|NATbL_vj`x
zMenQUG}xFOHON<Hh2IvL<G|sLOvpNS=kB?nK(vSL_g3CW%KbspM4LK%)*YFna6ABK
zxSdCH+8F%wf;{^?&vAsx(b&o$8BUhW&M!H;m$`%s{}H!685nBel2U!lzbPpH&BuU#
z%Mfy|ZHmry=DnjIQ9>3_6Lj=E>MF-3N?0i#_nZVk<`-r=uy+W-l<Mb`zF^R)Tqha;
zzH%4Zrtddz)d6iEPt2Fqa|}tI&gZU$hW=t!GW}2kH^GrChY3QW{}&Ih7vNfIT#;*J
z`pt{`_nY_U&*(9O3!t&dZuh@>d(dHf{Sq7X)8;AXpTEt2``iEbF91&h9tNBum*hha
zezJbBspA0y%pvLF4rlpiVDR%is!NHl<))!GK8wF3?x>(1n%ACGiC=X>1vH!g<nI5&
zuU<2OoZ_|77mm)~-|ELJY@*eZ{pvGIo%SL?>U<4K`wm%@0(a>%VIXzo&g1;^<{|*0
z@S&mfc`|*WF&E=#FzA{QP=RYah&nr|2U;Zh3RJtsfE{-U5Ru)#tB9S*CG`jnp>Fj?
z{H2d&utEB6!g?br&+d<f#c=+DcMX6t$k)JukP95^iXz6q8WkA<%2u9hApN6R174Wu
z4*CaPkjxp+i{8&00O9ZfSsg!M+=>|j+=cFe)^m<*(?q0G33$eQ{tXacTLB+$vVl|u
zoLT8-cn(ANk72%Lq5qWl|6>vGk3V1b1Z(}uDWCe)-+e&?DWU;zc#XymgHJZ7>Q;TS
zYK<HDlD6Hu8gCAi1Oyw`S3q}tyyRpns+z`Wzz`Szl<A%4Sfpjj$p>A%zjniLn4zK+
zk;nez_ub;zz_S2nz+?bx0v-r}dQK#mmOA)&QR)#9V)*&)t;cWv+r|2i5t3&I>^z~J
zR$USDU;CC48$0wS{;wsbGXgFU4NJ6vw{}N<v&jAEg;OvRbaaTB8;~JYn$ek>p+s#C
z0A&+R=dbji7vQ4t^?roTdGu|s=SnT;3Jkl+Pq(&e4`<R~6oWE2J|S-=dg;0C7_Y3@
zBfII&w23iAaUm{BXfwHlo{>|~ck5r$yWG6VB<K@z$l%egkY&#std5S?KA1jQbO^&a
z4P1KqjG~9SCsrfpn|qn1t#T}LbE$fTX9f3ZLo=LnTAn4Qp`JW%!3Uvu+qTK!eznV}
zre`YGa<}WP;rj))#YbZKH#`(M5ZBI(Bp#s9RVF#_xpjim>YI%%jCE;Xc7W9I*8YGa
zL&J5mA)kR;SM=&uBOGq#Z`*L7Bkz8>O)y8l(St~+f8P0h(1?*6K1ELrI!F9kBM0g)
z#2haVuxMuO-{<^ZVmtg^Ad*Ua0O|q)hEVE50Tyb4cB~{AiRS^e0EOK|z=Jt3mXre}
zi_P_rfYLE?=Dl1Q+jOV`H$Lj1p1!W0P;#Z30!#u!^9X*RNqZZrOC9TShFX}NfJTDv
z2GBof&v699M?OGmKoWAhFXdeW`@(Ruk)rd?Ytw5Oh{A()cLa>TuejcoDf)4bz!&H^
zux5cwM4;=5z6>Wm2HbRs4*=o*H7a0Tb!Y+@>{NAt1vsy;V3k{oV>pm&sMH8^-9%6O
z3BO)W0QR?&fU|aWy6G(e<T9e^IvY_B`p9H`z<Ou{(qLE&t%{$S2Bc&>AV#H_-O<1g
zBp-&nJmj1Knf*05hr~{IR)wk$zvlvpliowd^GwQ`kjmjCP)yllj+zHLLb=@!IedY`
z%bj*0-l){5=tw*iqKu7#%zW#uJ-4ZINDJd2&9E_b;N_B+3o>*cp!TqQ#WC69*ag&C
z^-#8KR0XP>@f=%l;Fer$0|FYNkov)|To(c&g+aHi@+)$hdK`~6-r;f5cZ42QWZ}i3
z+g9M?+Km;8V4kCTxmu^+Nngc>AGT^-Uvi4D<d4fyN7ZiuL9nff-Xbpa028oYxnsDq
z&@~B+OSe4Qfsu>hds%XPpk8Lopm<HlB*zvw6Zdi*SAtOXdfj7QOkZdngeMyyXz1qJ
zE<F9lOJd%A{8-<Lb~RN$t>>OtdLqjm2wpx@wVngsP&u$24UFgRPmOwIvDI*iBJ{Yt
zSWdi!Sn4E*$5$zIZM+Ina}GhSSqDA|?dOo=^=i~aLQNz(YMzKd0?2qzf@w=U6i~#e
z2iZ8?Cm@JqZ45+k_UnNf*!*Gp9-*ski^^X(R+0q3A}?x^|M)x9gL!6gNbcnR-7GKR
z73b>Xc!D}3Cavqv(D~s17<%gsAiz?R#7tKWvRU18erICq6{KIQj$Tk;Du84OcZLpF
z#d?>wfc`_5EDc@(5#HC5slY4VNtC9nIzF0^(6}xoB%o1BNieCLAno8%gM~XoYbS|~
z@>e6Wd9UvC#6KhVkt68xyGju2*;b01tNEJ}^)Y;s9=WGaoX*I9@2Sr_--EA}d)gHD
z$yibwu&{;0Y3mxG7;YHA9Jzsc6pY79d<OJ)B=HNr3&y2K<i~*pzWjB-Oad_E({h1u
z<B<T7K9dEsm-VjsQC^lC0k#vRI!38qH>mIP$a#>_9=;p|+id<&^Tn663UMMhz|9M=
z49V8OoWH9<%zz1byU_=WVKNL7pw@=<B!1)H{-y<RMd&h^6}kecMQMW8qHc}6KE3M<
zvC5gDA1yhgYAWibIk>k8ZmjFEw823m1zcDGyP)^Wuyyqvk3|LeCCxT!ffwNf738`$
zwJ#r$*)Bqc<UHT`dwTQrBAF4+KsFV-HOK>s7IZ@Qgs?qR-<#j)P7Y=~r&!KD{9JTQ
zeV;g-aU%uBL>SJBZ1|F%aGhwji=_p3HB3}4u>=T}l)8k9esF}^fW?LwGZp_dvnP_6
zNy7N=8r4TXa{(q+=Kpp%9-vFy;fl>YW|e`qk3RVyej_-%_u2nk1fLXT(2pp$+h8ii
zO-JrQlw&-4+d+KC5*p(FaG_3Iv%B~QE~K<MmdGSn2`;woqj(9SwL~C!<GKEF7qTR?
zeB*xA`|%WOd_&M@)xUC@CNZ7gtCRoxD-B+f@J!M#=diFGaI|}ASt_&-sjb<PZXX;r
z7^=sU6a@AYt%00EQSoVH^k)`cFRl8`4Dt5sPmP!=gHvv|IVVJct0Kw>H{0*8J{90?
zdl7)&K|*?%N_bcmaGtclGctOEllQKHQDpH_od<A@4ZwdB3YNV%FUKqV>hej&m9Xdu
z7G7dKqvV1zg=ia=kDQ?Op5Ku5(pA4;R#jF4n%-*QyvJhlGjr`|kzTi1#xp<*OToQ=
zPpL4noxevh0Dmx)*SM({GaZZ{NK&8~9lgQ#(X(htaUvr6d3M?aOY}oQkgLrjyuZHY
zdnMrbx*~SBt-(j>$|f*MX9?aBf4r}#ySzCL#<bSt#=gfQR?AElJKJ`^*KyPn%`lLs
zDYV^sQ8&;aQ|H*vL!8-<EHKfSPdrgYzcraQ-`@uZ(&AxKd@v<%^2hkt*%lx1cbzH0
zq8Ux-1yL3TB}K~kXrkU@oS`aw%R9ozbN#|#`7bNyB<pwK#(Wm#wME#0LcG7RlaN%>
zUE=eN!jD;*$?>G~Ubu7%;)}!nw^tPl=T~338GBdnd>oQ^!$v^LIk%SyrJqL0jSbkS
zBKyJ{f1z%8lyI)kedh0JC=Q!f1~%%6BCjmOc_n|G%dDlxUjkq!ve&oVld_`mHp!<#
z#k-iyZfDwYTje31Qip=36DC@V$U-=BPb%COuQ^LQt&(F9NT%Wi=SwB2%>Y*ZHI)nZ
zK4eYgWOxjaqz7olm=is1<gcZ?8l~@Mwv#bD6q><JQZ#F@jOmdiB4FH*dGJDgu-os_
z{xC5n9zeSSyl_<?lRMmbvlgTUB~deJ0+&ujE$Q)2i6UasO3H3V&&p6J#LBkDN_tDR
z_C^kaDZ(cg+t?0nw>iom#V!&c_sho}-<u2p_V6D!$N3sYVA+r8YmBcV#YRk8<EN-3
zi_gveKE!v>Q>}d*Ln2v=nHIN3>J&l=b*>KidJRgc6>TdC4mKP+>dc91&;%UxbDlBd
zIc$H>Vs-G8JlzZ-9eSKF3^Eqd0-9Lo2cCZ=%cLcFH<2T4BdYZK5OMR8QuCJ7>B#R~
zVbH(#A;J_tyE9A&?~%Hq&je&##Zf^LSx)pc9-vxR40Bg}7jP%3$SYz`7$1YkdU$<`
zZP`#E)9r|;liwcCJoj@1E7V?s)ME%-Em{U&r#?E>3@Aubq0X8krff6N45tyXd-f!l
zrwXWgi_@cLEb1M%#D|D6j7Rq-`ZSP5%&+g{phxWMnNucmdiL(d<sm$%ry514zLu|P
z&#Q<QKy2~*we;lufi$yf&+z#Zo}k}0xrmN39Z0i971hMN4Z$$Y6fmb0_$_(fI7m@u
z#IgDrk*xvsyMl-<1GXe!W^FK5SIgWCCr^Nrxx@@_bp&wPt{OeciW(xB!Y7v{rZtgm
zT$DKjBIXY)sNm~jhDu{ohlSQh>~)Vqpq|wK_{g}ygQYsyCi{yx0+rxiSPv$EugYbB
z6nf+`BVTFxwSBJzSSNL`@RYy7`QX>nLZc2(27vMJEBv@R9BPXnHsY&>Z3(XdT^3fc
z+ox08puZ>`v0@F)qjJh$EZ`qd^<cTTEj|X6z)R=jh?oaLD{8S(hM+z4mB3iR4Tse8
zVu1yS3;XpAcWWcBUNIr#7}zRvy(_v^kfvJ?WX8xrMw5`+%AU#p))>&p@lEi#qbXSd
zxKNa<Hdi0Hd<A{Yj;GDJRz|(Z98x#B$|8K3g@^zWmHjTc29<4D)u%^*3#N{T7PsB$
z7>#B7ww!>!R0ndoC~B5fv-8=xfES5fkQ=>cHNdK0ph#}f@>s=G`sEuiAKgxBVA@B&
z92{;*4>Cg8+*$j2An>?6RWz$xSqTC30;tzB=xf<gnOv~dQ?<O*_L4bk)bVgGaE$o*
zHD=7LPo=>IlvZ!jdsYE$c{V*3Q6Z5M4TAG4u661Tkbc+YqE{Sv2pno<9DNPDaWg(^
zfUFzEL;Rd4FSB^x0h7I^G1x&UGV{&`WWpCo4_(im_p!d>qM}nASXmv2Ou34_=4RLp
z<<AD>1oaBA$_Bq@WAt5Xi#OO{q?ND15J2Culi^L#f2`mByPtoMgL-kOF0JNo#k~%8
z1__4I;$G^{_IU6PSRbjFZu{`5mFF8J^MSo6dixi7h10(%oU4G2y#11XZkmD4rSXSJ
zLEL3*D?rb75+oOw@qWaLU_bPq0TUGL+qQ#<DBsC?KR>X>#G&FMfd=~yST1qLm7ryS
zg{9vIRlJuWzcU6d|F7D%4T)ribMgHd6V`y&WC&KnRW!E!07&e&lQXt&XIgSkq~1x>
zyM_TotJX`W8Ac+}Adh3B_dJ+%;1yVro22rVY_O31st3ktb}!lP-lw*__SqcLu~qGk
zJp$LZ^aseBsn;s^^(pqNKuyH$dEevhgV;iSP=Pomk6TPlSoWDfC<WaDMZ5MFf?{Gn
z6);FO2{IM6q_<txt-m>a|Iq&X$IAbYpPx$A=>Pck_g7wibY6{X(Of+tmK)jH9Sk(u
z)#@Inj@luj2oUn)V92tP$CZ=IOgepp8L1Q<0j2*{J_ti>Jqd%zYH{TENWHi6JRHt+
zg5@J75r+n59a<(a%)A6}0q*tcGfzICc6k#h*(oSKtnX_;B%lHAUNWhv6`~%sgSe$i
zdYu`-nbwI>sa6HzdH7V+bxq7jF-v+x)J}FFLbto06C*Ig>jqkuycdk*#|`UW@0QAB
zX~l1gx`2A28Bm<Gx}#=obsPbls!3jb&<NA6w2&@g9}-b`P_Sb4{x-3}FWuwrZ?bJ~
z>R*9buJSUVkAb9eAysFwuGqY<pNk{e-B|i>jrRYvEE};k-Tpc2;!x$b$U_*|+rOl|
zkk<szh7a!Imtu+J7`a<XwI_s3+9^ZovvOyfp7^qYC5n?j{kawIRQb;CQwntS&#F&#
zawPp+cU#9#_lg@LI~3bR^{qr?c+l-%2{wsfCSY!8zCO}5I}`3-*IuW+1xCCYXZS6?
zz)S{VT8=-};w}YF1-LqL)3sUYkYQ{SxPn%gIzl8*g{hTRg%Ds-I=c-H+f@mil{`Ts
z!^tpejeLRa8AO-}vy}Jd={f~9i@U_NU$|7qh_1|<DfF=HT-<T>?~5&Ux`=8}L+ih{
zD!3dwur;bQEYhEhAs#c4byO?d-XC*^<)yg*py<#?flO5A($K{`S1eGs6#~h7?iR79
z+XZPsmF*U0v$625tA~ZmxnY1{&2~3Q$@MQSPJZa5eIMZQ`@AryA#wrROgGB`_-OV9
z+#q^rP}ve$pqT6Ld%&zBPJB^G(-In2_@%<i9ep>vj(#cz6oF$G>Cx>LTYw(o`1U)U
z`^tMoAiSHfW}$tk<hIoSyXij-4SWdEoq?)S*ToK$Eke0cJHM1C-vx`CTomNJ8a-qe
z(E`-jd^KjFa(Qd30<Z$FV=K)BZPhH1CyQaD03XvyPd}rh8PpCFky{p8Dijm#fa?^%
z6+dEH^~E`L6Zpw~<5FRI$&0WXb?j#bfxuAu!;H|e-vnWhh1S{Yu<sWX0wCDaNa09S
zs6$??;ps*km>#bfX)*7MFr(LnT{_M0Z+bij#I%=MgL%5nBSZmxpkWsPEAAs`e%nz2
zV^93;p2X9_+(iG$*1|4Ixfm{Q$s@kg&$#r+o%j8)Bb*_Y*#MhpJ4sNj+QVe^r|G?`
zfviLBp4d?VmmSzQ0E8GF7BLd;PH@ys#5~eRdsrJOEcIRdk{sV$W~tyzvFD6BJ02%J
zI`Jjf{=9ayi91`1eC2s7gvt^Fpcsy8y-hAtdq7h9%dRB^m$a#4O&<>Z9uxYLNL2HP
zk)d|{(JiytC%@7IgGVG%kG!5Ez@1hkI%BM;;p+y9grN{fC}4Z}JQe2wBa?hN+ffNP
z;77~GQp1#Ep&0}2Plh7Bz}voST57@fM|NB=GdFPnqnrC|gof^joF~!Y9iW-mf`v=-
z&STXauFxTQhADf-Kwj@nWu^zF$2}y?f35;Pw+dJ#ycBm<0ZiM$&QVAu=48QlI1|PS
zwE4{KfTRqIzg~&wvm>y_Z*ryuy$p45sArtHYfvu`43zVHJY;Yn51IVaqvhk!GR?Vr
zG=fc>usU{2=A0#u6_7RPb87~EF!06?1qzXAomn?d2Z0@7#hS;u`}lU5v`AA05DG=f
zbC)eJQ4pvK5nn{m9wzJdpn8<WvUQ{skMk0_)+s5b9!#Yx7qpuzUbt5eq_hM->O(50
zR;ru{VPj515^Ny=BkpT_5Baig0TGD40I=I@ac#}6J<Ll}{4#IW_7<UO$GyukHw_Ag
zujMSRZw;TX<IQdz1$;~G5xm|BE0;BBvO}lbe`{wR%~@XwYKdQp`1zo0W#Oa^icpmC
za1FNmNFTIo3K-nmg0!e+uJ&$Ac5015H(4Y!EmRp_pv%9?*r#+vdvtwE+7%4;Cdhuz
zU!~z?+`WPWVL3sp#wa=MQ=sEo<{!)dhDL-hs0IZq>U})M`Y1}?zWu@`Lz9euuz>Vd
zjok!N#K$-F)@eLwDIw#Np^4i6;lXng(A$M>2I<B8hJzKcxxtO02!@cw764|)8C<;5
zMR4yEP!zf$0@{2SAy0pFH}@qkBI&$_8{~tz;mx604aI;r!2$x;^DQuLm8P=XDlsL?
z+45dlh6Tu<XnNO}zNRv<k(9A9&&s>M1x7o5>Y+fh@at@d{oye1H!(LPT4VST)I23<
zee0o*FVD@(7@|T*;nB*KNV5B-%jF=mQ9E$_ye4QYp*t@_X@KPUWCjp6{5lTLW*$OG
zjE$aMMmz4k$6McvJ`>_KXxsXG5?#vE<)7~`jNDPzHe-gHM2-6^5TA$8RYxP?sZ$UU
zDzB1!ji^=<8s!&~$J5Q*xqqGL^gWKuM?75PSm4z*H)6lJQiwFT6^2Rc2~KNYAcuR{
zO5QIs((^VK$L02yOj!<R_P)3DNR$>YcS<0Do>84Mg_3(%rNII87p4yp#i|=};+jLl
zAde$y2n$mZjL|W}bU0T{Bh;VD5r4bi#GB))Q^!EWZHHJU+YAOuFIs@ODt;$5G0lF_
ze8E|s)N{wupD;)_#>o>HKOGAYEx(5kHSo1@PAk|NtG(Ovp6v(NZTcfj2|yCvPyrJk
z`0eQwTuBM<-NUfaM=n@4UZIy5$qCwsTQ){M`aAydC)GvL`x<PQ7_y81P*o__V3XC#
z7}Pe{)HYDm%9tC1dTkqtQJk1AtsifYDf$o_kq%j~;E-Z`FH-2{&RH{wj0;Wbeb13}
zqKu4w#~R&OYd}8cjsM;|-~p!7N4qvM#NxqyQPgG<e$Ow25`{yWeX*cK{n#>c$s2#)
zJK)6&Pupf0CFotTB_{!Na+yT}pG6xV$lgETsa54^)8&2MkG6r-D5&WG)!iVnz#E_Z
z6}}n9`WNlC<1&zGd<jqF<NFwvemg!1WUk<J$0-meZ$C=1e*jO<Iib*ZVH_a&mOw}<
z;u_Cy8`fL^(#QEoEZbPM0t2J;s^tA?7EsZwX7g~)s>JhUA|?T1zEgHr%L;&kwb-`4
z^WeAvJt=3N$>e4wqz)j8RZ2`7_6>S#ITEn=Z{-K!Cym9-wiqz=8%%v$&>Kd#g}TWM
z#>m!d)9Cj2XnXO?n%3nROw}Cp9EUhas)`wi#TRz%93u9VUQ#P!u73qS$YswJUs5O*
zs?SI5*ZYL&Qy7l46G!1EQV%`d`=Gq6CYk`Gd<6jVhT1q%M)C$_dpYF+AqQ9e9>eXu
zwnF<MN+dI~h1Pk5xc$hG2kY~`P@!YNC(Zs4lN=QtKGmdtW733Uq|a%zXT;v~x@v>I
z2hy4Q#5$oO10;Q8>)qoq0)1s}v6WSxfRFU=S@AzTQ}YP4W5*wHxz6hRu3X<py-$+2
z0a;0}gj@F2Yk<te75^~gfBHMW-pC;gv}Q+BGfVz9hSv;D;7ngkoq4Nq{ssLs8LGaK
z`M_L?iSV!VVHh7sc~>)PF0|(Ve)HUCur}>rP+LNRt5v;by4+FBk~(_csyAf~ct{={
zFJwPP+|rLEG5Mks*=urWN7F_^NG+IEOp^JpOu>J;1b@FqQc=*yMbrf9|GpG%?`uHU
zfgY&aR&Rik%4LfMFs~^(8InW*%2^LG6*u0&hhca-1Xx8Op0y+0vQTf#59AJP6(IHI
ze*7sUxw?+=lp36{(QZp$bOIH;{zr5GdVv8C5ooRQgePzP?G&&iIao*usRJUAltewx
zMxk;Ff{C+=J%eiAF(&PG>Bu5Mb>#zWnerJ`LBuy60I};jW)?JL{`NQi<cC)=;Kscw
zb%+r93peIh^YNdk173Hf!L7RcB)rxCFRhb*z5F!>EMO5DaO=(f4R-oVm+9}((BKDj
zy?$Iv<NM!{5_qDaqcmDU!<F>kzuaGVhLRu5pHlP7etGk+-|X*e>ra2UYy$PE=epI#
z{?ESm-!1^o6{zE^q{@EdPww%b{ObSzw|{}6b{tCG?_N-(BK`|^?4OqJAFuav!czn^
z-c@Ya<zbk2H9+ZoIwj94c!>WJi1VK|*{k<8xrX(1KtoGkzVAqOLkPetzMy4V9*WKJ
zFYM62{E;cyu&Y1<?FSOj^w62Tpb2X*N-+uA%UL&T1OFIM_}i-Z>uU+;k{6(lvgiV~
zXvEE8I5sLG_38Wn-BR|iA%gZY5w{0I?{^ga>z@O9B_Jh&y*v>f@tFNDg|2^H);vDY
zUiKk?YyIyCPL#sXUKU2Ah5yfW-GAEDmkpu4JXxpF_xlk1KfE0{5Puj)S?gBXzkaj-
zw8Z}XHPX=_luMUvL^%HMLb(?d%KwWx_n*GvpRJ7ld;9<I?f?J4?yru>VwER>2vO?Z
z<gdRl1=oFZ{O65`!-Lh-YFnY}G;-%9FWcg)@EUM_z&i&4NlN&E{b!56qsISvwM(4=
z_FK+b%sz(9_0~2pJrVmw0fd0oXwX1Eu?F1WOJJtMwr!C$7^1pN0%kV>uu4Z9bVpUq
z%+FV@1oXdouL78XX}+&{xh?-IS{US#7-TiV9j;5ul3Xwa4cN}t!1+p)6U@9Cf{Bys
z!ua$Jw`5<+mj@aZu{AtvyRC}3<4^oFvgj8Ym<<^RBW$nk`MX?AzdGgOWHnFakz1g8
z`sZQzRMKmc<-qtn^<B;P#zt{bx4zTNHwSD~*eJen-bP%8z3hf*%3CKybA94E?q)m5
z|E#6&Vmk~Ejv|0cDR6F*)&U5M1UMMln__<EcnVymmk`HH;7Z~>QMu!o*m4L^nmXQQ
z8i+Eov;qF{8jOtda{@4A@`p6x6)<n2E<HE~$g7@8Os|&#d5F#g^qnMfQqDmkjwc-d
zhiSUr7?RB$HCgqUqS@@l5QSG2KJbN}Mptz%R1@{<+A9eIpGl93rjmmuYU|5@N1~@e
zB87R=W5B3E>jiXLI-0)$uAzhQN^ZS&u;y$RUk3X8<K{%5;%EanR(?Qhn*{^%_x<@q
zU^+IIs3JJ@JS7)cF&C1}OI^}+L<w08@Ssds#y-x8TBek*m6=0$Z3P?-#GdLQT7iC-
zV{FJ~>~lTcKRftvM4$`p4p`iEpv4FLENX7@abMc$1Ndt$C>n$sI~(=^=Bz~^O|yyl
z@+p>`2J}=6o2X2)Xmi}=0e|KS<dLJ~Py+F!02r9J!3Pn2A$b?f0bm$RFG5Do7-xNb
z;g@O^i6<+jO1?gHcQ2kz6az%t4>Zr3;U?LjIcH^t-8H_@l~NB>$LfTtHs4BBf$Cwa
zj)Q3r!aeGM2>?}}zWZu!hv>3T_33WT=`kb*VFGpJMi0Ypz(n%cfT<ASQCkrN9&o?$
z{9~tJLLhTklB*I26B{8YO#L0=59{EY8iH?p3TGG=1G=_#CAQNeA?`CkcjY>8zI<86
za|G&?M~W{uiH*Y{wbgd8QW~+Ed)`Bc`kU(l?2wiEB(Oxg0_J9vqd`-wJTw*<yK#;H
zE~_EHtd|~Qqy@S`yJ+kp>V{mfg~CSq)`IfeQqUZ|(6?})NuV&VsV8G_2FPx|NNXzq
z&Yc86(t1Z@vwha+!aA5w;WLdnd0;4RS_J1FH>17-AW`Jl%Tw~UL@)%WQJq@6z!X21
z!X^O?`TXaY@ay?-R-Q3z6zZra@%;HxFm~H+57@ve*jX$PCc8ZjISmnv#v+J6)>{^j
z&fRq>ZBd8@wdDxvhDQmQCe$1F!5PfgOjJwj@Z&DG^0XUpR56*jsCtd|)Aar5S_LCC
z4csEHu?`WXM^3KwjF>R8gFu}*0aTYfp#LOH@8bmd+lrQbU8&w1GRJyP3oGdm-&RD`
zMpO~O&_n?=zz2_8U6Ed4Jq(tGwFXdZtG5hWNglat%>-trZ0lK&Wpy277g!4?fAb4K
zka}GMcxd$KQ70XWaSw?11Q0fRjx30FPEZP3uD-Z*E)0-qogt{p?HTt#m8wZ`UlIhI
zKzxSicwj1357{@--yrj&sg|QeF+N=RG!b$@4*<O~LaQe0uei-W^DIun5-~SvTqe9M
z18%9&{fv@KE`)Rl+W~0i<vt7WOr8BnwV!?K8J%}aDa*UYxd78@5(v^>f>mtXD{Z(^
zQ_zO)g1W%mg0$n_)%k~%FJ;qj{@Klv>LjJK_O9lnk}(7$)yV`4`OVsgZ!g~HFl|XY
zb>Q6I0S;t$HkhV?Xv1DqlB<HaVVX6V<h#Pz@@ileP$JnBSvEK90_=&d87&fR8lg&n
zN$4Q_1%)0#o2usiF^DCJ)YzFsnph1-7Z2OD&&&L1=F`-476-a0*QdbgffXPQ1?8k$
zSdlahLqte+^>6tKZ8wxPfkuk%gSP@|GMAxlT{dYA`!3Xes7+*)e%3B0YDO@e9rpBF
z2L5O943_H{*WmeF5c(x701xjcJgM23hV4Wz#5jvN$AWs5tQJ1#(fjf_829;g{*}qR
z44JnbJbcHkXZ1*~ZA1c9kr_b0Nw{>G!=8z!2JiFR)0~>#{j(6?J@7x2$w?uV_6OKa
z07(Jj{qLh_7IUZRUzyMl7Aam+<nNO4B`i_ha0U|-hjt~9KOnMYXy*cN+ugity`WFB
zNWPX!-le!v4EX7<;ip?s5Ege?Y$aE^1nNzqfKBpgb#HLu72VuxyU?KkF>Qs2RU}Ue
zb<{9jA)YhIRTd1Cpbo=uV>bjE_8~5!a)V~?X|~>ol;6`mfsk*nj}JkELC}Vv0HWVP
zE#@m1P6VemPxi+p0Yq}qQ{@Y7`{2iFS%&B;LOaPgKu5D*f8~>s7(7`pncr-zD-!j!
zon!3wK$Am}og^#P2Wijb=~H?*WrMLrK!`DhOhniz^mX!q=%z}#NW%Dwp&Mcy3icDu
zWx>xvn$4%^a~cKRC7TJ+y)jj~Z^A4cc-Kr2MD_WzBL@`Qk-@LTi77n08u_<*vB}@%
zfU{2*i)emFPQu8YiGIa?RG8ann2P|lS!UrEchR)LlXE?KJ|8KiPl@>^T}Jml>f3uu
zAzjW+N}=|xB^A+XALcuQW$%^T%P=DMPp3*{74sJv#rqlczUzu2)gt5on|k`2{Y+2x
z)`NLo^monTW3ek{pCGo4<a)079tgLw$|l*tpYHU`QpS+cHcRHb>FE!#ZEz`Y{E~A-
zoD=UXfsjYexM9+kQJ3MesHhGVMU{&GhqJd1i?VIkM-@>(TDnU@0qI7N1}PCK=@>#v
zK)PW7X^@ac5ou|qySt>jOLAbS0roxb`+nbAzqP-0{PwZ`@*o7}nftk(^SZ7xToCiU
z{GD;dBx~Z_z2OWQZT!LE?z{1j1?QBD?_;liMdkq0o{hkhn6q<|F_q+3XrrgJ)TX)+
zvC*S<K(32C;hw0aZ<KB$SH+U3zquR~+=_3wWrj^|iV@s~?el5mghumK^A}EZComiW
zwDVx&49s|qeKXab{aO%8y=-w(9f9d+Jc+b5{-nT^|Fl;EAJ<gvJeLY0VaPY7Z?Mn<
z0~@j#wTUhZibyB4ZK!KX^aslHeXWtRF@bg$Md51K4)(K>^V#R8CAk={5-SCIE1J3_
z8qk?a4#=Kf+j%Ha_T@|?<gnB$e`J+B39CyMcTTC#)=&1zUj54mdWoL4B{N|Yxx8*J
zuVG7Q>!V^Dy9NZH!}TC7<c7i^eBBN9z*X%q<Q29twTYNPJkmhyXG-e;?ZdlB@K?8$
z1#TV?c{VmQ55mM-!Bp`8B??I(_8E*4+1C!QXmYJCEA?8_a~dFOJZm_p$>ZX{2CQoP
zxd2dc=1DZQjf3iGpGCCO6DhtjPX-jHH0^~8K!!-rC6I1!y?1|tKxf|_kfmKmjNphl
z%-+8ijR0!^>RsFC$}W#Z;4i$U*ct~xozd3+zPLs`-tTbx6<3ZInnw{{gH#T4NHhUR
zQO!5|2lHNQI+E?~fXu+#+-%=1jX3O%Ww4I(5Vffg^{UcKz7n#k1&aeOHoUPau)uh*
z-y*8*0DRi!oaSGzAOm4EiE3!f4Gj)C2cF-6(uIi}S43k7W;&pnKHN71AjG20{U}Al
zwuWiKO##i6NQ@rIRQKApNZgQmENnAX8hpjMo!bv8M$_+4t)%AT8d|4|lwAeKAenG-
z<_KxUTm<OFH>v9e7wGi00D6uZ)+22<!A$);P<O~g$v?&%WU}CTt#OXOQ72M4c1ekL
z^ZuoD^4=CIiezN#nr4`%ShWIHD&7RZjwG$r;65=1c3gQBO-MFdBkDR3#!~tmg&U&o
z+_R<M1$fkjWZZmKPFQc+t2u%j(3E)Ez($!$+a+#y^w%Vy%_E`Gc^7@W?YQD)ITmxP
z@#r0?&Ok8hcNX^i>0K20tnqy1=c2nv`U#)=QRS{HSlnJd+By<7Fan%vyCVzRD=;ex
z22H~Ei4rdN>p&U=EixDHloQ~x&?@|6|8KZ|dBWq2JtJ*e1YlWgu;Vq-Q^-K81o8OF
z71;K9&PFs%k1oC%N~w9;_l*QtD_*<sX#Sv9T<il_j<1Piv}9y5m+D#&<?=%l%@sx~
zD9B#M0BS2_u1GFdLQVk2Tb#8P{)?Wo6}VMc%JkB(oqGs49^oJj=&GJecg=|lHt85j
ziL5X5T}RwRiLWpl6phA-`KS*e)A!zTn}`)`9k>y17^q%{Me#<H^A%Xlm}>t@ySm&h
z@f2P2zea{IcRikqmiJtEUi30?`=qkM5JC|d&;%~kc$Y=QdUfUzN;2-lCquw8>{|@$
z=`z?W!^F0c83jmW)e7jZn`invtls>IY%qgSjpS&T^Cv}r!Lk<?sK^*~q#WVgiSkTM
zQwwR9AQE0QPyx8yJXKL%z-8}(2lLEQ?5%}Q02jLHzFoLSHIr$6(U-h|B9>}=BR!xj
z2)TUt2(a0vz8JSU{Q;<Z@4|%gH@ho1hZlPN03L6c{<71z_|gHQC2uCQHp9H{u1<r_
zu|=zm+|u1jpaB^!*&v=1-O44qc>op&xL}E>Q#TSh$=^%42ncAaR?kwSfAM5ENlrIr
zUNOS%))$)X$Je~yHyl$HZy2g^;26*yU}t`EOs<NW7gEXI`^g<=fdkf6w+%=W=Az#V
z1fGe+U+Hh`yUF2G{Uzf%cK^9GIrz)QaBT4BsIg9vDezkU5ozDI!9aEPV%8#J$g7B#
z?%t2J_yr#ZBZ2me_le8xiEDd;?EF(oe;8lLZO-;>v@_KZcrK?+LET&JQxPhEWJ<4Q
z?{Jj$9L-fLCw1n=g<{h`q9A0NSbhwMAn)0Wu=ziM1bD_Wg{XEWLqt-;`x2_5eLY}&
z8tSaw1JtP^;%4dECaqQN23t%M$rgdGrvj6|J|}0`RjjRu+BByov6LUy!X3NZr#>|X
z8UJA{2r2uFjcu%Cv(jbriA4u+@E>`eE84tDpW>G60dlwNb^{kicq>cU_gI1oys7t%
ze*q|?yho)yq3ukA2(h+?5V17R8Nh){O^RuFLQGKl7>+W3rA2_x$aMHpHiC3{GnKRV
zpyU^!r%BfLTDL#z4kjo8?b7=Tu3es5SY~l=LmOgLHb9gn)bq);tAIWzb=#MqsBBp=
zHp~s}-B&mUU5F5-&k&ORqJt%`bh6x~jH>aOYJUv5x!g=}2*Q{_=#9$gOi5?ls%gfn
zR?UG_(B(7bQA9$CS4SVr?;ei4QuC*z_ltcks2&On4KoH4kC~`poDufsAns`sZ|q&1
zVJHS(eX+w_5eYMr$Gf`RE{zeYb9}n%I9IeHr0&KBuy4=davK4`cSfd%Fwc7BN@k|u
z9;(&&9yukvG-bc_)fx`pi@wm%{70T!s!D91l9u|$j3l3xG{IvfL<g#vvSF1xI1C&S
zJoyyU;soPNOff|w0@vSc#PYQG71Ro~nCP9N_|;;Q?u_*HN+*SSe^}Y(Z85w&{(c5V
z9Gy@{`o(yG)!%G^=U0GWVdv4Ha+Q)3`^#+@>bd|j3bQ}+Hwy~vlEPAG$K44M#YRxQ
zei|`l!)v7|BepS2>GRv$V-^HnO>V0yEA5<yb7#@?2ot=XbjWU?25rx@5q4AI(`bY3
zl3w^Mwx96f$20EUc(FL{?+4+h!gf}>Oglw&DA#0N$bQufu;WTZI&nQ)^ut5kkkx@7
zx=i+|x(UP(G9oDq<sHPjZu4B>4WSsdi0Sw%oHIAii<DWzM^7Lxn^X+@cnBRniCv0%
zsx1%I@+)lX?hjaLY52;eA-M@ZBAc0h5$x0a5bWdx?tby&sd=jXJKiReroPmS_N6_Z
z#6@8c)@`<j>KYV`KKot$_oh%wK0g_1W<s2l-9LMFX}a-><x13o26|o{=qS)4V<H&?
zJ-Tzm-S}<sYgDV~Zb82szujLj76@B2?hH}(%gv&#-ONeeF3=2$C7o?bWt_S%Kqv<U
zlWO!bnn)LsbYFedhiA2RLjoV36FhgkOMti_T?Na*{aC^T)d!}G3#Xhe+}}{v8EhT}
zddsg)ZtqT4D-Z@A$NUgp8Pe%z9#K8Apkc%9tvBa^&?H<4if*J@h-F~wdW-dT3W&|J
ztZQqCm0#?vxq`w}L!BX%+=cX(Ks<jol-a+^iPjA}f8~0T={^+R(Apt(sR%gA=)Q*i
zf_}mhh9%{SpT5o=r1*!uHI&Jep|FPx{>(G@yk{M@ve%c%b>cE>tJSAC?RDbZVDMZ;
zyuB{Ealrxp?lR#S{uVBZm+~k<l*In|BSiAcnfspR4#oaK-Vz5Ki2g?^ZCs+G0pdO|
zZU~Z%L*k3r2$H*8$BE;R2@n@%FK+YubiMV*T9_t#YLaQB(75kQVS3XC1a|CoW*0zR
zP9*b3j((ysc2`fouvA~c9))3Km?PDwu=8VsQRx<%6-B1tesV?qH`%FZ7Euf=-K-V0
z1Se8R@EVMFcb`slj^02JxYp=VzA#+qKQd`9s&iXJHmvP33)_e0`CW!s47@LXr@P+1
zSnKqrXFXT!GBq4`_C{&!zZOl`gx8?a<%P9kR2(KGF+RmLpNK<1>U6dWL?WZ_CDBTz
ziJA8&)tVxZ+0clN=LERZSGrP5IL}Ej{{SD3n@5?vaTArRtASvWHqKm$H-98DM7Q$o
z8PKhHV9@{AV)5-1Gygu^b>IAgle%EG^GfIxr6oj%u|Q-Hmz1_Z!YGa?Un3TGwD++{
zzys>2z3`h_-uH7wM&nR@IHS+}(fK`pl8bZD>ryqmUC@lhouM^RM-}(Ni*@<<-by#3
zVa1KTkDIFTj~I>N;iD1)Eg8eH9_O*mVVcpK{)ZL=sU9z8$^*=q9&MeKqm;~q{%S*g
z<Q!;0%A`-4I!YreC&F3iT@%nm%49c?q6c)hkX`^wSdR@LO;kFG)AVn1l(tJ|9nkZR
zQGK@s@e0;4ZcIok$ag1YwXHQ}zE#<8hv56D`NA^3qMI33)D7dr7U(3Eh-A`u*>v!I
z47_@i!8*3M-Vu_i!*uHP^HOpbJgiSYv!WBbC#s0du<j*o+U`qDd+K57CPE@||NK&-
zSfBaLy2C&-Ka!bE<voUUtCf7w+t}@p*>~lJ@Be_KF|=tK&AT0#?#o!kHT?}LT+t6>
zwR|yd`U6U$+3`Zwy{85Q>%=L~q%%U+dI^judj(mq{7o;Cz(EjIht&r=_}htWL~<gU
zMdb=t7wvNMNmimk+{4ghL(Jffyj0zg+%)g$OA7YXXq!gi3;mpf&6{?)W3TQ(Q5=}7
zRTm}xP^Q%M5k8+&ME%gf@ix)3`r8)~9diC3joslLoc=&PNr1>-RQ{y0p*QASuVfe(
zmHWqM+AF@{=3F?_fznAUy*`!9&}zb2=1UXUhkcex^Vjp>3?U+I{4(UE(~_B73I7~e
zuf&Ygn6+u$dH$&;c`3SufFt0Xo^H=Zh$@=LR)X0(ZP!Qx>aSJk>#pfN>QEkSgsLc5
zZ?y51G(1E*$;v!oOHTYU7B{}a;kz$v9V;CM#+J!&_%#dO(z+;Hcp5wQmeS}1VR&or
zUtM7E!@F!Xw1XQEagGptn(PA$cS<~I87{*G2gKHyt<c-1Ob6B8S_6CpWK3pZ?84qf
zd?+J)30Do3vW)k4*wuzW(>BU4QQ0G}b`12KBt*wk?eS+}8l_8P{l=$J2)?T+`Y_%y
z1T=`7$+gs{GYS2xf#%29MuYlHn2gQBwOH^^NVLB8M_P`(f^M`)NPXF~dFuCXc2iZ5
zq6TcENY0@I{O0%pv{RB;TuK-!1y_T%YK~}+eF-49Z&{o8h|F_G1*u)$BWA+JtDrqp
zJR?CrE!P|_pNtF;@B9&w9NsQB4LkO6<+f6ng^pofYfy)9U%ME1?Ok^8i!`G8&PA8h
zY^2PYb&0XF@-uVJl@9jt=<*y9UkkRrVN2Vn;l9>4@H+nqqesUVw!Qn+tpX1HnC23)
zUMCu|RHB~fe&oEPl@sZxjOrV2m0_8&`f-1IEDUuQs!%u8Xkuug$ar-DW-NBX+HK$%
z@Jh4%&9Lz=ezc|9LSt$!_Bf8f-+-~X8V$2Yl*2;2U|Ta~bw}2H6>g_rF}a$sbl}VQ
zdk{8qJGj%(;Gcnu)o{p`a8{;wFtWc$ctm|iiWSIeKZ$4ZFefwb94x+BH;fdDI+1ID
zu6(n?xB>p3lS#;<xOvulW7H#8CtAm)HEl;uXn^XG-t6}X(Gj{Wn)ickn}WHzVsmwf
zgreT4*`pN<A5%p^&-iS#4Hr|0O$LlU0>oQOf0M_t9_?~bKmBISbRN!OT_YA)a6%)A
ziAs46p}I#7#qjDG-l0)^^lKGZzuQ$gLhq9$*0Vkh*F^QLFE!#!XTI|h#s5$EZGB}c
zlyPJ2@445F+!DCcPeYa1Q27WlI(W3-C0=Q&HBq-Y*qa2LfThnzgP*n~=r_G{L)OU`
zO;r1ZqS8b<rG3G2cp)<Nu?%vxKRFu-*)OH~Tf7f}yLq>)`lWy=v2@r+jPPQbYkX?B
zGY!P1wF@2Tjq0oC_a(k$T&K7uIsD8iQDMK&7K}`v`6*4dJUwpVPK;MDRFpp&f1ZdH
zE*8uJd!@rqqwShgOz3@rKg}(sXBI0XYV%(QuN8xJL_0eQ$Cxz&f4m*{Ltc2LZz8_!
zM%I$Or!GIAz59~+rg^k8x7Aoadal+bE9skQdy8b3V^1&4o6>h$Yeq7ei5848n-Z7r
z66N>%$`Aq*ZpV7bH_tf45Hs%h?T-#ByDC|#SPT-+zdzs9A%Af=`J^vv*Lej@!6PRn
z536S06?SA|>#}t>V`84m;E0{5YM-q7d<>FLBwE$$eCVC7S`UG##q<9&8UG+l3*YP4
zfpx*CR_D$|$HSJQ-QVxya^b*rf10=dT;X9PcGliwx>2z=^$_lZz$#WqxxJr{E*1we
zI+*Fi$VHX?nCY-~oacpg^XW1T6lX?-pPy6di^_VfmFO6({2UV<bLbEpn>`UK6E&GJ
z)mc(a)4Ba1x_$j2WmU-g`*Uxf1+i7?1q;q-2|f2`IhmERDqOzXA5J<RcMiaaQOJr^
z9$Apd2`^3W{Or>!Ox^s&a9&s|I`czodsp3O+a(1;xQnqoddf`k?7A*}^P8IiTv)F#
zK{$30(_zU(Fa7sk67{e2H3n}V=BsUl^hQ~e6=UPc1!$1r7Ib1s;SL?PA`ZVC=Si8o
zuM7GZ48HXz3d@N1pg~Ry9mLLo&Wnw}wkcCr^H6RY>!*(Z9JBK7j#W|NZ?wyYzXmQ_
zZRiO2GiG})jy$&cKN@Bz<|Z7Lv9AY3`s@wXn2)tNzuhAl-R!LF;QhSkIy)fO@0&a8
zb{nz&sqvuQw9L!YAiOvMCv#)?Pkndf*{6Pe<085o#s>DMsULLk#B_7ZyTVuKRJpE?
zYWGUz4VgCtHrNN~sVwm2F2Y~n`(p?iVMFRcavW*}H-Qb+>0G6DAUgl~Lvvw1JpSGO
zry-g49JUo9@0x7A4>-Yo1g+wwI*j28X*9q&*3^=ih-prRfp4}~om`4*u21{^x(QHK
zfGPt_N`_E*t|);6Nlqysj@MOTtOT&%BklK)Ed9*14AyNpzid9yh=9R(!1E<k&nE=m
z4%sIwKR?PKzH-dkCl&uV#<m)S6iO$3PY+&&2b9wd2ETZMTQNw`Z|2;ad!T<+z))1)
z`?Mty>)}Mp0|F`DG&9MN7dl>GBGS?xg57n*$@P*{%*(Cx^HdA;fwS>aH5gw7BgIfj
z&h53P(=>X0(2N%9tiVu;@{rTNsF1;IcL_eC5lljnI-V!@D84+<CFVfxC;Y{Ja#$!1
zqUo;iPX+qwzw$i9{kL+Dp-_a_m(uf3*)kABp!2_vGDxuNTNtVcE&J}?2F3t1N`$e`
z<F96!!n`O;TOYEbrfPR%6FgtG3#c|SsW$4vMu@`)?=5`F>b<?6*IfMAmB>G+1wZDC
zAAPyhAd3(?y0G7;*dmr=8q&LrFW=xx2Rzo{eoBiT(ay*~%PNs35swvK-KlZq+xXM7
z*!jw3aduwgdcpf3{nBe;cXph}SHt<~^-la-X7|@(CnH%_#zt<E<{u;-og&iB?$-Zq
zh)GG)Bly$A<5LZI@M)G}tj<GddqX|BdVKSfKlF(9a6pDSvoxyQc@p6TQjh4J6wzi9
zp(nIs_O&l`#*G<`>D_&zRNNIp>>!m&>MJhortbBVf;j8XX?Qoaowe79-`D)kJ8zg%
zhSDf!^49qhit>r9LRXP9r|8{v&Ttx-@0JwJv8i+N8<CObrL{|*T*wVFq+$<*d@#}p
z9T=c%s(j>8;q~}jTX^j6@DmT~+AAl_bB>oi+BjA#f89l6?y@G%x?&zc6E(vju#ma%
zQK@TO4!Oa;&^Cv@*}0OSJ0upyKdwj>a>^mt%MV&CW5bl<t*8T$RRb7b=lUW>?>z7t
zu3eN~NVxyQ8{Iq7#w$?>2A6C;by5(5PF5MVmRVXlILh~P(M;C-dO#icJveq{x-#;e
zi^a32|MCKu3*1Ry!@4}hrbE;K4Toxjb4Iz})+C-8&tT{SOpd}2>_N`>bYUVpD#^as
zEc`U&QO~jON;QK-*A;xD`j*R5OPe!q+TD_@n(3jopu_z&a*?ijefib>)J253DzwPH
zf@wfdtMZ44P|<!DXB?BVqhskz&sXQJx7C)Tdq;)DClrp=!%f_9+J3N3H4t~raCELG
z*z-+4ix-y{fXT0DGN3Ua5a|{~QsO-Z_by`ZU90de3JV<`91)b)8L}Fy8Vz55NT`++
z$a0=TF7T*WCw`HU^x5(^Hkq?^*HI3`cG>y`)?xw)_ByD<FOi1V+vL7s6An$pFP<AW
za=i8OejbZ$BIo`}F041v9nZ92xHRX58zi`lbQLqgQXhUr^;;PWW)4Tovksy0sV=ep
zLr_Z7ybv8DUtf1fW0lxO(`fgJ7;e4R{B=>p$Tryde(P|V(pJ@?{RDX)tD|LSLS&nv
z$2MaVe_L8i%`F)1t)4L@2|Ne0zD7|}`%$r}+x*vEc>-Wr9=bSYY%|mT>bBWE@sAZo
z*+dmA%VYTQpIA6UNR$^<d?BOv><)=<@3H0UlDw*Z#Ta~2PH_<Umu5+Kpulg-rspI{
zjr)nrc|(reIF3JTF<$%K+mq?XG&@z>Jeu$d8im?z5h&K5Nps|=M#Rj_a*JRc&*5u5
z*8KSOyt(^tWyP%wv5?EbOR=lTZWs8k*i@wQaO?MNr}xw?T#%)VWykkrnyd~-?N(MJ
zjU;@Mk3eI;cTXVwCky-z3+qG@SsdqvX-TL9{i&24`_sMfd=ll(WZS;^TI*u2@|*z|
z0|6T6UEc-o0tfTB#-z%vG%1|I%Okwi*Be6aQrFVvn_U)P`@3*WTZhH>(w~o%HgUvJ
zaA{ZSFdu7AU7^ap5Z?UC`5M)`BtgnE<^y-K())*FtZ}Bkjs=ISw>plJ8M_Ec1_F6s
zKB~9R&EmwQ`p{-#ha20RT`kvEg-e9VZ=~Jav3;|(n7$c*5EHFjU5hl}noyOZS;mOv
zJaCp?tTF#wF4<=oi|-hEi?dqWk6b|;D8{qBJQY-ne9_L+ZK-Ez%4>>FiUP9k4_#5s
zmfm^{%rELg3<%h{({GBymcxux4mSrlEziUEhG6cWFybc<Q>!>U<#cJZ9X05+Aet9(
z3W_@KBlkw8x{`8UBC7GTawd<}iRqZbhb!C1KXP03n#GF51)XIOsJY+HWMUq1+~#lU
z6XPMOUoeTM+dnz>cVVVe{LpH1$soI`VIwx&F4I7d9f&o_v+6?1hv<<ERn;vjB7;gK
z9E{~C_YM2?y6LN;_cFzo%sQ5KOxg;$t&E9mz17lTf$tB;6IrLEjoK7ATWharysM$o
zeQ-`=i|y-J{llJL(AY40`2dLZPMx{nYu*v{tcq?_1Gtv%Bw3O=uk_)L-OZai{E2nI
z;it_|oy`*^ywq|=mha^R2dQTQStXqAEWDH8WYO_U+u33_NwIiN6D0i}Sr^}Kxx2xR
zcmxkWjj=X-7Mam~SL%xpN`zrh<@#jT`j<CMsfRb-Ae4M??J#t*GceJY?hnNA<9BBI
zY3M*wT}9_WT5Qb{5QZ)Dp?-^`V7Ek+Ej*<#e10<?a3l4R;|BMvpWyVl*XU#=#O$6h
zP<E{gAC*t2WPXIh3xa1cp`shG{-5netRX+R={($i$U?O$Z`Ak@+^!J<Is!4+PMPMF
z34Rn1dbCb%x{eF66kaa4sY)oZSucnEa9&ALh5d`gXcwX01lLMRhdx~QmGrU~v9VGr
z>F?2Tt0enb9cRkolznN$1Vdln&D7kXH|3nbtv>q5on@DG|JeJ$GqbBsevEhK^_ri5
zG>INyA*B8R%#=as$JI0KJX6Q&?{8Rrnbm4&j5?=2@xoKpc{q%({VIFEW-6f<iHXOH
zPfBLVP+;|t%&dg>(v$q0jFG92P)sF<`=d+Uu6>am`yxAUNQLd2-}}v~A*;B%<ppq?
z!mPyT9Fmdk{(2(<k-lC|fl1^HBIMiNjmjRSK2!u*;5MJ4hkfG~%gQtPANT5dCFf%i
zuXE}jM<2^lqYM31cyx7XRUvvQt9OjA(>WFJVdR$1n#_>1Nm@iVS%Q^*-EcFy?7pqz
zW2K1TrqHLZ%MyMR<q<cTW%Rz1MsQP1H=GCsOO7nOyi(G9RpqY#Q4fOdYqW892v%4b
zDsjtPBzzu{Tsd_n*y|DA$Ik4wQJWf%<4L#>CN`tu;+-?Pe;ZPochKF;O%N!P>YT;a
zU@m@yw*(!1MJoJ=|7jl<=0)}L!xVb;pu?`mkr>~|g=veq)}GS75&TRhEJE%X`=oZ1
z#8perK?c69>ML9aXhwDQ3HsZt{U%Kp@?D*>9wRjAtSCPsTg<iSuj)%JUkJU3VB_X7
z?>}6emS^EMQX%7f&&K@{&*+|Wj6`!eCqoJvn$XV!Pz{XGl~{ln{zkh!<eZ*a18t?P
zldHVWa@D93!}rN0O!SU3hOaQS(OpwsY|JseWoh>g@|`XXE&X+Y7gJ__X00u##CtWw
zXyB<1-?TwFagr0(Q8I2}=Y4YA(kiU=>Lgq|qT-K$DNm<Z)=@#fLcX_+SW?m>0kaH=
z;9Uy`kqBthn{so!2O5n$8)&(Ji_~q{GNAohWz8bXztw)}Rh5i-ViS4YqAasRq(<u^
zXLlk0noGhl&6;`9qZ?{pgo9y;%)Ra}Z18JphJ#*|;tKq1iUS!;=Zixb>)1?8V*M^V
zXcJo<K*ZkZMu81BjdRhQq${@2COqsbTVy?K-lA=;4-I6EQKijsVE43?zE{N@F7eVt
z>X+rhpqm8m^i|+99<ytHd%-{S<X7IePpg*J#7P>RK4PsbTc!%HVvjy0e=eGswfF1&
zYRUx(YCS)?Z&<=$a&l_vi2P!@K>WSbT|D&%12FA2BRg!X7~#X3A@5?R>S%-{3L(4k
z1<S+5LOYh-`T*?LR3Y4hAunDwuE}O7@vdD6(k8r+B^n6QcX(YTlN!T;JF4_ns!hWp
zH_h=S(@ZFjd`!I-pGR}JbA*JGNUIr^0&qheLs;r%`PYt#^}=(1v#V<CH(x5Kw+Ulx
zkBGyYIt%ip&c{SM<K@CUWa<JB$;fx`*Mi=sW0(-GS-{T<l4_3&So#ZW-&}VZ<r}YM
zJ~@uXt=RWeB2O8--pwiN_NG`@BUBX^G2?c!f}_Ti{E4k_P5;~?V@z9k8uKDOckbgW
zLOsVi($+=k#JbvI)SC2rC+`U*?``@!u$)$PXOmj|Vszc@?f1G#=}zIFakmjWz9@h#
zn589WaWF-eStsP50Y#;R*i0S))VdSzjz5O-d7z-&CgaYWqV+wLPyz@zi;%m-_%}&}
zobH>Cb;X^Ki$)oe8lURSiq6HPQfOR4mLGt$;yr8(^1b7&KBdeB8klV+cs-AGe=m~U
z&)haaV@A1BZWk&GP5g?a_Hj~H@FFX@5k14JghMo0>rh!%bdwA0Ks@ow2Q~X$XA+L-
z_UApfvfuD9V<0m+v70A9uT^?ZP~W!E%lKwQ7e8B$K0-0Qe_8UBnb=`5*qDwIWh~m~
z<XfwcU__iFyCNE-a|0!nlCAr8N2}@BU_`805|12Xct(@oZullG$dUhU*h-6Cunn&o
z2d#DxlZ@C}tyy#jExT>IVse5l<yQ8mpvYP#ej3Xt(~4jz8X3}e9+vkoIqFO8Ljw5&
zeyheYXXEmBq$ClR=P1;>i#sYqOGqrMUU=T~CFPan;drr%iWV@HO>P$Ke>j;fny2`U
zy<{e4Hn6v__bey;hAo)>HXO_GQw?T$vAjs;rSWoZzNRqo?g&duKEkIwp!z(jo;Ag~
zqg!M^?ncMLI3D}X4BcKukuROsG3G*AJ3BX*HgVA8?`N{+k1QgB#cKxJhwA#eeV`>F
z7h$}p$NI`PV!V`|<Vf6Mgw8AHcOmFxSJ{+6wpOcWwLVXTe?z-jb`wF|%-tp{4io3s
zyZV8jcU$)IT<RMd%zWZWSf})wy~T*x$KaY@5rfG>bGXJ8{jcB=3?tCFq0(T94BBbF
z<ho#Et@&4)tu${BBpI}1uRM~NqsqHJAuHpe*RI7dvFgp`lZfssi|a6#IB}yS*W7~$
zI<oM@C){7s&`TaVeBvVAhR{IEtO?pu9$0f1VQ@}Zw0RVz<UpHp)?>2_Y=PbLa~Kcr
zXKsF+*&;pPon4vfNbL*mLLTwntMM~%!_4Ov&61mkd0A>JXL4B$3%`a8lv-W(B7Uk-
zLmY^vonYkFhL~)nexKccZzuCu5DT&YG>v$*i932CpZxdj^77nNxpK=ztQQv|MoImI
z`w4)bJ@d8PJef^Q<vEcq)Tz;4jXU)MWEKT!kyNYOZy^K_thF-iR>L?hR}fI3LY6Xn
z6xp1IEaa2oOp0k+lV=EZ)w~eqJV}AET-CFg1IaqXL4B||zqE5!(jsXKM{1MO;Z%{i
zx>6fvA=D*6X*9gCQ1?1mQ<&*jy{X#{krL8Vs>|Sdo4_?2UzzVi+QmshT@tevS;bPU
z_;=}~_O3fB<FgeX-<S{EpcgB7@n%*-?gp`rCFRWntGwq6Qf%*~m{T(GR)bGQBub|o
zfZ5j>)Rg$|XO1*#O!Ot%a`f2j<zPksFj{`%9rP{ts&3!znK)_nJ5c&=Zufg?4{C7f
zo}UcaoAGCp5(~x7E}wY`Dle*^Y_==UoJ1>NO)hVFp0g=3Uu55!D82suc_q3rMu8^<
zR*;s$7GBt&ai6gANKMo7DeFu<El2hNpb+<lsc5Z)+$0>owfde=4w}~*R^c72)JJ$S
z7=ee!Omp>1JJfG-)SI>@FILZhU&-E2+1G{&)|Ri=0()qg-}KxwOU)`Ha9@9FTrB)%
zMJox%2i9Oy-|gB;D=q_VA-s4od-uo?_)Q0`(yy5Rk(oCYRrX|mN-#=pU0rMrCX}p2
zi|q>2iTI(fyHlo43VnATpQ6-932l)6gwJqJElC@EmH6HtqjmG4C1il|=a@yZ;h%Bk
zbF-0P_0g|XO;^cD$E3FI2`}x=+Hwa4I`n_`@-V*^@rGmC16#|J$xK?^@g>{~PN$82
z8zi6SbPsGT;mZXS-YbbgG&_t<;aG4RctTLiPpj^S8zmP)gv-mBCl2Gn*?WiwY<4DV
zR5x7h1UC{3-+o9K;p|<c3uLF(JcHS9OfQj;BVy3_nMv#XWEPf~i2J;flow5BST0_*
z8sckCab?_@1@;01rW84Gmwxp*(dgGO3cc-Fx`M<D+_tbDInt<Weq^0(#<PsNgXTx;
zh&>umh?*<99PO;P;VMVv9Dks=t=e}(syHmJ=XKO^k3dut!`HF~qH-t465BjI&>icV
zsd`vzC1>&S?8W;Bg2O^Qp7#U$!M<}-A;~y1<E)MRfm0aZjI-Y6!#&5bA&dxzB~D&S
z!fYHa0qg;MIm!{Kb5k5?q0)m_n#|)FHRQWg8rY_fNALQ72{Xp(S&&P<>pP>?BqjRQ
zut74)IW8yI`u>p6CS?<RlsCfoa5DaIvQqT*Z&vuzWZ!L#<!;o@$=^?KCi9FN+eB?9
zigezMpXKkOliL|~G13(NQajD4wIi?tCE+qf%yjgNbo>ZCg73$ya@oUcHH(q{F5wlx
zhhCY?g*hDzd*4(IJ~>j~<e*W}J>Y_hkMgRm7dujEH?i|XFl?j2n8pPw*lTpP%r{_@
z(i4?*uBNcqA+7W?rqzBzZ!RlJ*;X2enrr{OY<ix0@P>;Q1Lr%a_yd$VU1jDhDh!I0
z%M4#hdLvG%M>kU}uU#`Q`YjN5WER6SY}SIcE?-yXX>5LANENY};3~7=2~eSm^i0b#
z8H;WaHAd2W((gM}PH(TJ#p!!-)UHHfn|I@REdtjoZ@0DdYBxL33HTh}k%&6W!cIXt
zMqF*qJa}(JpT6i9{i7fa1@fo@*bIBxiq;Od#6@~EJX!)2$sG!!$*i&#9DKjyR&fG)
zl@<TZcZEMDRx=OS9_8f_R;tGk^WJP_VJ;(A7e4nwurs7^gtD0_EB<P_jOgk8yrhHQ
zEh2pt%=MOY4!Q?@Fxz5+l8S7>%(0P%Ui6ZM|C12epmg&6DFokX7@xOrpii;I*QPcO
z$MizoEkz97TY2X2pZUs4zAA6G3U#0w{etV=MERG+LxvwU>*!ZVjs3X1h20@{RNcZD
zRrxJ|p_N`!nIDBpLzOYGj8!QKitd0A2kS^@D^==g_tV=)*ypi+^@o1Uq3I;2d}XvE
zsS_N`g^{xRKG3&Q$zAD#7itR)w%N(;Zw@98J(3pfJ6L(TknlyWi{X><j~#9ST(qXi
zrJ}UIkaXIN4<9Meu_3|A_bc@{Z(fUHV6sCflcoR#7s00_b7Km-YlZjv@p{M43LEIF
z#CBD0X=`p&rOxX39;}g+Zb_#`rLa-W$tdecZ9KDn6-v4cJ)L%ctQ}O-MtTBP>#$ZE
z6Oe)yM?#GK+p^(77Y&e*yqiBY(<?Z>^P1K!*7;qoYVVJJzx3L%O$2P{fKO<rPVC76
zH-S2J8*Pcnx(nGe=e=0Zu&Q?q#&NIrPNqwgrj^X%LX?$e?)&y@OtZ+j(5mLM>R}KK
zGApn`loru1!hWfEU5x$iS1*&ZBM$ny{9X`Cr^R{aL}*4fR=3-y@;7JnTaffpyD?5@
z&UGB+)_R0)HQ^#@_*MIuZCCYZC@GM?<w5=6NDt<W`G}nqb+@9y;P+PUc7_etG5!MQ
zPmQ0X!j}?0-;da<PD%6V2xeF<c~5m0n_&rcc@-E5Z~s;-UKawg21HV1+{L*Zt7_Ac
z&X`xbk%iAcBf(mHBNdGtWx)?U+C<YC%m}Vu9T`&SF<qC}5>X~;{ca{PWQ9=iW0}Q}
zUq<6!I_KgMrSL&gXxST|A};i3;YH`cey13uV%__KD~gS$sE2b=#kjzZC6Mm)Y4Bl4
zur-e7NnzY&Bp=R>xld#k+Yla1-mt00XX~?bS`K+t+xrs%V|?+K)J(NjFVC&tQa{-l
z2i643ebKsAoRn|Bi|UAHJu4>;lMo>fu`5hrYLE4r(wNP=RMfA+5Reh$^+uu&QrlL;
zvD$gU!g}!uG^E4jBHHL}SyYJ4kI{qYKt<xSmvp*X;9C?@RYnstL~^jk>`UbuTX$rq
zrL|wa?-#+Ui+jJM#A}_^NB?QGJ{WZAI#w0;_ivVj7FkK{cWy9bmIxn!so0ZClTks?
zwlSIR^Y2KfbMscMQuwPeAV%(PK~p3g$HlwN-f1I>cU0$Xo37zI^WthXkuhd4)Ymj`
ziD(O#Ga~U+_{#4T-%J;&Wl>Q~zAI7X*FGlqd`0p$UW@;E_#a>fwN}hqz8uIoE_i}b
z7J$9@CRmw-Y1oE55UrXvW;%N%o%fBJj;hC$%>!1%SBZcD)X|1ca%;j2S%?v-2~rAj
zuAaNpLql<@x2)(^Ek&Rze*E&O5*y`T)89@Ml$aGMvVv=<*!E<vyvABIhQM1#Z8Moc
zl}ENdCuKIj2p$|*&g@yx1~Qa{VvZl2(49lZsL7&VdMV>hZe;UwT?<)#UEB%znnNXE
zQ~H_g5~5P)|J|et1hIHN5<NJ#@N$@Q=yE1T>uU`w3-OW>!efRsp9hPjm*Kd?6Rjq_
z*OT#zP(G)=K@Qy0vk4;=&@mB`-%?2^&5Y)Fy)QqM<fBLS{Ki+=;Az~cGL~>thJCd%
zE<(EGId`f>$mJ*{uY}?V=Ol@`mov*2WM(l>Y6MdG&<2*PJ>Uu<D2sn$`ncN!0$PR`
zJ=%L}&-C&UUva&=t@)SIkCCRz+~{O)?W+Ct;$}*}U?(TJ(SG6-V_=WCDE;>AJ<+qZ
zPxNt;6NEv!lTkrRqkX{!mb9&w$&RJD-9_NpN|i=*khQ6)Ypby~ANSc3O~xne&o3ZP
zX?<TB!5%%5s8!}zN0^jCL+eLuoLO}>)g`7$&W*WRdEX2bD#X8j6z!!AWN4x_kwU~S
z%U+-EOvem$-PaN0WqTZCcWj(Do1CQ_LvY!ys2}_$^v~o4yNe}**@Biiz$K{##==+E
zOp%ta*|?{yX*y3KYj^c9aD9<Nhh<s&mu%sq8k5CV6yGC`r-D+frl9_M>8=FtnsRSV
zVfghilX?)jrtjV~FdrP&VB`46p_Gkp&kQye{u$#@-dz)O^p26Z8FSsh@THVABKXYr
zwzFB|CaDXT8>-H~Pqikne#S&m){bjH>DVT|`U7x!=Z1*3#mOq9%-WPGWH<XQ4XMrE
z1d@hK^eV+gPL-E}+dr2U!PR;1;HmQ?f&NXkiB}y+`gG$0Psk6G*Y@q@aH+aFz$hq?
z7B8Ov&6sgVR>?g(Pm9vrq^RG6DVR&xN%;9=7vKcsaq&X$dSh?F>98cf{$aqOb+L22
zak~*&CrHJLjpnRpYBC+=9KrAn6A&y-l+ZT>o{)%3aci0DOR*hlYfR_cFuUZD57}Pi
z%Ll$AMXau-=)U=XwHF1eDdiqh{~!vr#HOe4f9`K{Xwy?hGdm_>l#084N}{qI-PzMU
zoen>$Cv_~Sm?=^hqSM~FmTw#MHKqM>$jUG7i%y$z;(q0t6*(BXqr*(Wpu*od52(lo
z(tg51><3W4jcYsstLBdr514175cb#1Yv|0W!k07dy$ZB6E?ts7-J#yh6wS{`(Ycy?
zm{hiYD_HDG=Q&S@noBYh%s;Cm<S5Vo1Krw!u~QtWJdjsW5fx7UeUe>h=v5xOW_E8T
z)STh~S5F!<GA<~W+P*w=$?W~M8LZVeZr}|wg*vq>9lqcp!4Di2VRYshO0SC(RFi+&
zkXNWN>D{l)tYYyWMG=(08p(Yoaw#&>y<zrk9^<G31HZQ?d$H$KLIqt~-!9x>iy-L3
z8uZU9G=AT#V&N<q(W%l6S)hQIAL<ZF$Lo*Xq{&0L+Im{S%0|$8j23f5rFhE8v7o-B
z4ya3KLYSGE=KGU*o>NC~Kv7rB{AYTy<PC41h`pwP=52Sq%#xYP)!|2Yle1`zppFw@
zuhL%54R;4bt2WK7cRkJ;@5R9H>{*Esb||-6HmhjISRTSKZZa*ou8+YLb_w+1?U55x
z>f|Z^y0S|PVF<7EQKEaxO=%kSbX|h9TzdufI7B4PlG~-+cmpOU7&s$uBSu=@=TF9*
zdOym8G1Vu^?ZQ-vrTryyh?3Vti<Zg`PEoFVT0wmrPR&9~-@}nNo6xx(hdY>b4@GLo
zt&KR#%FT^X?@@l-AYi8OY{UwxoLW!02*o2vkIAY1a~dSfj{44<-FtK)e1?=d$%#{?
z*hHSj`A2^?OY}&*7MEikR$MJy#<7G2s~1d3cqA{K90hxaaPy@kU&F4y#28W8esafm
zq)k(OlxV4ZdTmGg%sVKiLRhnWYx{~Yy>_#0OAygS6M+GJwY{5S+3i%$xp_rBBD(`I
zT=2eW$gls#L4?gj`LrKx8m8j=a#78^`C$$F%snr$n6Yk>X(cDeYqhUc{H1IWdBKo4
ze)6bf-lufER}RUIA=ZRbBP5uyO}XOVp1nGY5t7tCgISvW#GCXM8ZoqXKc&dz>j}vj
z&-km4B}F+x=ghRgVh!=A@8`gbZG7*L?2EZP6q;BY#nV~Ze_y7ra25y&h2dr5NweHl
zO*;jJCV^qc90Hw!Ba-{1JY16$W|sqsL2FB0JdUJ;N2v{HBO81y-7^4Fu%!^r1C^%w
z+v04A?hGHCZoT^c?qz=FDtI-JJ(u+>Ks-C699cU5GBmrkY%sy;5DO!R6Ro}Xk7M;t
zXL<C>ni=gb6+}i~vrSlhRTKXF!H<f(%mUSkU!(%%OcHO=$VU4wvG<DlIfEH#9A9Pi
zhgMl&4JM^?k*{*jKnU7MBbt6s@hT`r(A0RhhQgJYs~(?oStXR3^fG8Pj2;~U6MKEb
z&&f?OBx@Ozl8%xOPMp|h!eW98&HAeRg~2d!_`imU&g*syzF}ucyqxym?IGUmP0!-#
zZLeD`zlmOU2y9V!XHil+WnncRpwSL(f7bpSbI&yBnt;1FN+OtO4|5XZTY>I`BG_}J
zk`{PuI|_jQDCYb+g~WOYVx9FH^RgLn6B>o~I)Z05rp_5+t`!vX`G1T}a(2RX7)+Jn
zz`Oj!woawiq0iu<6LHOTPw49c$peUUUAF@P8+vY}uv)xWMd}xH`AtaY-%khR&iJp+
zRX*779cv7RZ#gGB?O#~puh0JTMNU*EdK8Z<B?BqXkN0QXpdVKF!q&lF%e-+?n>i%t
zseXRQ)l>7Lcjv2NbFX;p)C_7)<$i_Ncb;b*l(`67Y`EOvT_-Y*n#UT+i4I->8|@PN
z_;L(UrjipG^~RW$r&j9OQ6$T69$=*eK+RV8&>sqY?bqoN`_7*Mq@m4|G?hoc6`Yv4
zj-_y#%`CruZu@Ca$9F+fm{Xkd<d{sII)8^~GGO%A8*23FW)gqO2spP!$;CQ(Z$VCI
zPM+74^H0$ZqvUAaMfEkQws)5i3W1an$F}i>wsJF-YSd4TX&*QK0ilK`0;ylzLGR}}
zutqDPDdUs#QCx$Ri~NSEd@ge2P)<C=e2s9mpMT9xKRLN1T-i*SVo#Kx=E7^--JW|;
zX`G}mxCw+CW~qD<u^XxEPs*TB4#V2Zpx_|2*WywxrJVzSGT|g8HPrtOGqnG&Wu?*I
zJ4@t0J$S<dm<$NUz~3m`f4?FBzVulB7-4>v=zY?<Nlk-<kJK~H+hmE%UUZwr<ibH3
z0JT6?dV5@^G*3A5^?KW#x2>_WZrmeNpF=fal9flfzIZp)E(VKj7GF1|6OBa?8@;;x
z$JIN0w?9~37IP2|wCGI#3|eabnQr6`6xPJ|@ojUR%!|<(*|VJ6bJ0xh`(|Tv<4?AR
zZEXJe+=PIrrWqSddm;W<h|k9P&YkcWlU}A`Ly83_T)j8i)Exch2uDu0f!G$dPWSd(
z-JfXSK&@oQcf7n40(3$=tGxm>2|b(me_+AqGJ!5kDH9d7T9N@&&ZfgNs1Bw0tW{#p
z-#L^p?W*CgD~%Fr)y|MUh|#WK60T7=GXw~t(^Wd7O;aY}5Jy(^J<i|`Dg9PM8~5x2
zt*q&x&kUz1WD#ENA0M!B&`1~7l@N5Vb4wYtHq{<KJt=GgbBj@EX1ud5Y#Ms!n@_43
z-|b~lIO9>lB86Rb%jX}KD0@@eRXN**E75{8EYVLLw4CpaAOl3swBWA~ka~cNBMFUL
zhCP)&<xIRA+MeAyvBGTYtX_Ri6G{bN2J(;3UEJ&6!+<^IbX?YXx&Xi-WsX{XB3=`*
z$!n}9eFwGW55#RNVX;hzAVFR-3vkyWDF}<3j-~lqc=t`^ItLsT%^Kfcy4KLOU5NwM
zQ{4{#g{haqd;-Pj%Uzd772cMtAHr%$#uaHX=%mZ`lEih8nSb5D;Jo;spmVl-vS#N|
z(JPa0ZY4WF5V6nYs`>|6{?|eKr?Lk(pGz$P0sC3|qna$<IM!F!ImD^Dc*-C>T~syu
z)9yS^jd^d}A+4!|pL(0@7_ju-5U7}fLPa||lk8%EnW*(N&{XaASXfs39VQAu{S&cU
zH&lxy`q$nS4c5&Jr3qEMmwHBH5a)UvE-M-tzFBEChCrwno9vZ`u;^5Bp}V7T)3|+^
zS8%u!4o0yIrc$qpsZY%6v<%$8D{!M-s<qOX3DxkASbzP7pq-(5-+Xr(Oi?!a9<HE3
z5(L&TJv)s3?-O^gkuQMX_SyK$@<0k-Vb>7o=`+8!D76i5B9mcG6B~fHU-HIXZY_B{
zuwVm>_Zd3*W&Bneih`H}$w)OIE41JTV;BoZXGf;V@0Y^LI)`0vQ6SO_hpbwBE~Bs3
zyO5D}h5|Sst>^XTTDUOdq^E}?<AsLFXAemj5xf$+5?8`J`#B-mALKyUu&X&MAGkV0
zyk>NCmk&U?sy+GdysJW~b43l%_Y&>h)46*C9Dvo)<j_%U&P{Q*OZu~8w3KIAtX>8|
z)NW)`4nLzPuEqEATnhF^^B1^WLbFGM!J2nQp2ND<G4kgD?s^!UKkFk-+>MAhlQx*A
zJfD8q7iS@!OvPkE->ZMg*7<NIH}k`&LWUm=++ip92pM9#%q52t{6L^=RjD;Rp^l3?
z?%WijoVw@r7z2rwLC@P8)JBI418k7B?3Lf-C~3O!fTtgLG0%drOnFZn^J-;RDYHyw
z!}&_MMa=C7uTcJteD&PRGOAUPn;7PcTB>m5;+MI?oXZ%Wvvm>5?$K8bRwDl{a#r|8
zEm+w1g%tqLrv}mb+y&%{Y0h-B@&2bL4IWC&*J^HTcGH+|nmUjKEJtFk0}0N6?8@J-
z6)z5|k@+fh!^VNTdEcUrQ|?@dS&e4f01?m70xaXqO&l61TLa8GiVvHK&BW!k&iBr`
zL`w&WKI?~49tOB1`#6Be6<lagy?^szY_wOqJz|-<zdG0sNGai3WF}Q0#N@K@4a`Tq
zVYgFwO`zM!(qPW=Tcp>;AkUEKG@96<e?(y#?CmkHPw^$Audz?K&*}yU`+04zuPSfz
zHZu&dvDbRp-udpF>d|*{oI5*J!0J`!$F)Z6yXeHkVNsCz)TneGCHtf`P@Zqm3wBYO
z=ZZ&vrEg6YoZCc~UrK!!KM{L{zKv)Kc#N0dr6m8DbQ#H<^6*I&y;QQ~qi;Aw_$uKD
z!s$IwVts6btV4B7C9IdU`JB;xpY$=vGz6HvOg0356Z(e-fE~R+jiQtsnxMsH8_)Z_
zojbD&RGtnKg-be|kiF_Z0^ZUTIRnR*3=QN=(&xr}9;8`CY~ziY{Wvgw{Pkz#P`Y_m
zO;Mo}Ja`qQ&x`O6TfwZaU`#G)`OAC>b!Gqk5P^Np+YE4widP{@-EJSHo&gVu0`csY
zw7{}wP_bgXXFMRfDmJr?v(p^sPU3Hi&b01Wko5LTy$R+lLLAP&E3M-*;Il%<NuiS%
zhh_eMA`dKo7F09*?jEN2xH$Fa%fP&X*bY3Y_R9n?|HJ9Q0zLzx!1G3>IPODMLTjj|
z7<0-}TFiJFUUr&IfKUzjL?|%>O_F7SgQWQ`#yOylx;XL`5DZw^ZS*Ir@ERbAi-Vy#
zLIRLpLgrM9=RUtzfSAw0@n>Qqu~^WTHE8{Y*}dXJ&xr3fOm!6Y_4ter?gi5M(Ak|-
z$L&5$k7-V`IT4z^^FeahaYz&vF;o>pla6;g2h@J6(&5U{i*JY>AsMqEQOkR7SyTtz
zL@Z3gA@ClACFS^%2?L7-vp2h26h39D!h1#Ny;DXI8HMA;DJlISORa2GmVYx{FJTU*
z{4rdGx7f{dg-#6XRFhx#+WxvrxD{|951?PRsJ&0rGR^4*E<6;bfV(kN_{Zs-=Rrfd
zrR@hvVJD>POtj%lJf2kDA1h2ps`p(g-2v40KU~Wt{(uoo`k?D|iz{7#kkH{4^vkB$
zMUU!E?u;rOt8k6^ps56f_|})eG&~IJ5FRXQZRQzZK$2#;;!v)(D4_79FYu$uZh|HA
zaXGQhquxr;`VUTymqu;!BuX0a*?!Ar<V!TZ$x8(u_9fqJX;o+WpOl&n90S=hMbaBH
zY;TiyW9IJJD!bYkxXF?f8q<1-``pRtuHxVzPR)bMrGgXrVadn~+K+ifNoOlWhvqVe
zZq?gE7dz;A&S0LVAjst?0M@jwMMYK>>6rNg7!K1tUok$-6e#cw{+8~9cR<^w;>svj
zT{eUx6p}YNqs-GDw`)WeAG(83i(0U99*(|9zIqix)_0WZ?iZui(Bw5A)sZy-p0;ZL
z%hUF6!sY)stbc#9fVMf<u*l!7OV)#InS<X8wixcG=lw?^5YLE?Ek2gU(|e-I9n`Gr
z&N9oT4~~)9%*zwF5Cqsa(U=*JyxdWEj2AD(-E-b|dftZ*#jg1^W`XtShot}|SL@=r
z)R?0gdS$p@&%DbKRp>CCDZOjLaPO2!Ho7lEtax}E00g6SM$o1Y+HV1<|L<k^fBT4Y
z5KkKnZ|1=>Gc`}o7TAdI0}(!Ft07_7>GM9mNP|MmnT9p;-OIagPUt3kYQVyx-TDii
z3eRnLGZF>301GD|CfgON$(W8aSod&HZRd^X#6dy2qD;~&hK*>W=4%uWQ{rR$nxZi(
z=F@Rfq9@H!MLB*{xKB<*rzv&W_9fH-7#+I;$-I|R>Bb~@4V`o-vgZ^bLGN__Ux4^O
zUU`3B^dcaUq1w|_mYM_3((`~S{|Dr{2%}v@7n&`eHJ)884I>1$o-XDbseJI&yQ&6f
zwAN}Yr-n#V`zxPdwH^6hBp79Q6sDl+_PtQN>-2o}bK}YP;^AD?%>4;%)5vEco@92Y
zd6SD*3um>DdWc}x#r_A5HXPuxa+)At5LcdVIC|?MGR<fKd1fgAZcU#Bl}`8oj(l6N
z?&5K_W{%(cPj%}*y{5O)OTf<po?3-WZ1ZOG`0+$5HH&JH-$563G2ow#<A1dYfaj5h
zeBT{3za&!qfBO}r&rlcKR<T`yk^F~Z3rzJYpmnWAhT~P1FaGDl^5O}2S6G8O#s2+l
z|Igp|e?NC^fqwWm!MSW~jQ@Xs!oU9C|Nb|<&_zc8t7w=6FjD$sD`UL;4_5<=429_`
zr04~k(7%4h|MK^8G9mFy5*xo_<-cwuaCQ+xw}|Mz@_uG>q+yZ0_pWXTgK=UQSNIF<
zT-tz!b#{BQ3imSTncNcaV}Ku<f-&d*!EmFBJn9X|cox3-J=`=rl*9!9ds*a*htoq-
z1Q%b{BoeSik#UD#KM`dg0r9nB?yyp#^OsVZ|I43<=MSzUeKSwVzrL@TleRgBIW^Sg
zeN?3osMvGJ@JIy_MfRWZBH+m=CyYV7C@ASSa0MV-EilYknA`@U;34HqF>lKC;qpiH
zoIt7S(JCtsMQ@7UT>SQ=s_S7fl1y;bDbTolMFJG}GuHFK_<~dO(;`cHWsorgD_?In
zDO_5=kTs4rnM|m`<7-#1PiH;w+>Zw_f3UeBA2<()tGXT>03YsC9k_+)g6H@g$ZK5I
z)N|<A!6;ZRSD--VHP0jagzDf*5a4L&ij+k}Y7gZ9>(c)>MES25`}h0S0J@<P5`qeF
zN&mi&Itfug+IEa7j{~qjp*s4I2^=~0K@PSHXy-C@!8&dZY45XQRf^t)i-N}<C3-A>
z2G|6(NCOG)lW#Q6AjE6?yywe$Hl{x)Dt!bu!#jc_{g%CZWdWzaiEmd$^hjF3dV=i?
z@I-5YK!1e2dMUt^@#5!LC_)6}N6dpp>JO80%A46Hclp3OP#8!n>PO3YyUJ_Uo7Zgz
z43ADBx%+>nX+WsoZ5z$a@K>@#DHPa{c#|$@A1JT8K_<`|i0Zpvh0NUp?1LY^^yFT)
zGqRcP8)T~I-|Xgrx5|xuB;SN(3<}KTzo7R2d13zbQS&15950rY8vgq;Nz!*5@4SW6
zDnuqu+tuyOf#GfBp}^bwoH(ocfIcz@LQ1SaG12~&@*P;YO*ep$#(e;nGz&3cln-j2
zYc@CjKa{<7T-0m3_AQDC(t>miC<6l0UD6>Tpwi%wgOs!=Axd|5NC+a`DJY%N4N^mQ
z=fHc7`(F22_ul(?-uL}yeLf2ZotfWN=W!k9_n7;LErMlJPjs~^1yxPmEUe#GR5hbu
zS>paZ7|>P?guzkdf%tWqF6c)SngCFTvFRd@BbKUZ`)((TgZH4dMpWVeDt@NWUVvN;
zdlJ0qH-1v+^1Yj_ez$2m+2NNR?OBK1G*|$C2NVZjG=#38i5pF=qSzPUmdMDv5lvvq
z{QqT{16j|6SKsDM3jf-rz0zequslQMDl%N+T4QN9CDMTvlksapTFJ*B939~4jz~i;
z({EdWKuGX~F=;A2ymSX#U>6h>5y~6h-A{D_q<<HEHe`URTN5fU1LSE^f7L;HF<Qub
zfDrkLDZo|H4Pl{uYemW1WJsg-p8vlFC0P;X8y``|2TNgN^*RmTARIVOH?4cWl?WGA
z)ag9#iGe%|{xF)iP-JEa9wsLyj?^__)YWs30(_vK?s^+Ly?@(fB$CyF>A9Fr%!dE3
z$4G>ShH<86+U5|a@4ROsUg*!tW$W6l`>ybFcb)NvTu<)+7+Ub{Nx|iK<ImUhxSnwB
zfnpw!!NI<8jQuaS436X|PV<cXNv|pR&*s9avFa2(zO`s9%i8tiq(6!^oqt(^U!sve
zf+vCzlUw`nv9c|h?J4#9VjR-F=AA%(=XDa{!E{sN$@g+U8^d?~XmNE9(oMS@!Us-w
zgUG$5$PBc?cvX;v^RMF(MT+l+JyXnqH<N4*6AwZ%$yl5>`D;B(f9768TamM?(ummg
z@P2!P!VF%w+J-ZY#$c0SgMp<^6Azs1JjDDo1(>5?S1t;4RRIj62u?Z?j{eKNx(;tw
zYjqk}d?2>Pd{vt&#i?b|ZqAnE5OHjh$6~HirXi(T!V@u$MAA!~rhQP!`{;rk&lxZ>
zU&@;_H5wIIEajeBB)bYmUmK9%!F;wJ?5G_03M1Pzl*z(IX$zoperld>v?j#wn?~yA
z4NF=I4YhIZHf~-3A!_Y6h)os7^weW1$`0Q3t5m`CL;*a6j#QN!bs}H(R)II*diNls
zdS0_+_|042Sy%k#Wl`Iv(?eag`g67;?tDs$%F)~IJ102EYtRX84}u^A8ka9_(Vjku
zs}h1ayHf3Dyydsfq;QizMTHCqFUeoY3!CZ>w~ct7Qf+bJ4jDCq8~#@dkPkRUdg1)G
z@295rUmCAdvz2z3@H!z%&&l)p+<tf7SqysPrVh0wIQVIgO5Rvxn`E<#jO%YfFyQ2@
z*$!qIV>Ny58-rR5x!4MZUWdF)!GyR2lEAd8=eb5v$@>J;SS;}gTd5xmPu_p=?C#4A
znj2^cKSrX0NO2K|pH4?y4I>)&%=~&}VwJN*G?|;-8imx_h$kQKleGpD%H>^Sj{V>f
z2WAkCL@|vcm=L4HZy1^1s^O5G;gCp3xp-IfrE$4(=(2!S`TDW^qRn0i4(jAbp3Fw_
z*Yu`izNzPRR#D{M%brnsb_QhbYG5HK;lSe=%m;07KC{J_JJ(mX{KjoKE`Z>6iaV@b
zW>$mBCPl5ZPFC_B$7N|pV<ShDH&(Dbl_*dM>^jtzMzjOYX)6u8KJ3*13YlX}di9FS
z4Y<v!QJXp*t}u#QIU|l75<j!lpLjNFRn|KVRwnIQI>qEabgimJq%|nF$Ie)MQTXr+
zmV2n6+S*TeLLDgp40?42LO-~Jg(h~0wb~wF7~xFXa-b?8L&x;gS8^%ZU=wsppaHm`
z?U@}97yMY%@pI+fi^8E9z@@B<yhZ54rT*zSGlXwa-f#wB5my0!QtY|`99+*!Q5}ev
z!a+zm8rQ7%pq7{c9H(s%USejLT8Fu}29k>HP#;2gepd?16YNkWW66$S4Xy<{@Q^4d
zF;lp{*c1T<))z_yD-Qv!Q{d@%eom@#B2VXHG0K7uppx~D42$3{*b(Jmy8zKih@rNY
zlRQJ@w>096r8TfJJ0o_$H&8Gp;MhVa=ctc15tHs8M|>2hfWjNDGA&+%lhPq}<j&i6
z;b~N^@kytu$Dn2YdC~PbyI<wEN}SU)0K3)qHXxBw71?VU<Qgt-8QlU0=21~#&JwHk
zM2>knfxgY}7kgozo2IR(op%HI{`Ucm>fza8j`-G_D5yb8WmvSmu3dm@+i>3ae!n9_
zRm2er^ECtHzV#(O``2KTd_!qf>=nljy15kcG|vryhq2*$h{t`?nAQOl@g;Hh!9$D_
zM)6<MTJ9HhBzs;|d2U3IbC$|(+((jfMW0&4x!!twXngkW41|l-*t#uJrZ_vzdEIe7
z%h6|O4@V;-set>r60z)nbHhV9VdFg<Wgl?Z3IpV}AUDcSIC9!jaOv$q7sLT!wt4x|
z2#7Y>yy$<!Srxq7P8MwGo=ce?kgKT!#Vk@Vq{?P+wz}^-a_ADg3H?<ZdJQOMj;Nwz
z8iXU;d2ZRqGoCw7%CF13W(&)<em?ZoTl}1+@Im&>RlDb<A8ir`2`J^AF5hBm583vJ
z;+TGOEzJQ(s9>C%AXGJMEPzwI@!Mimnr``IsqAM;v#hX2u)B5kg19<lJ>C)j@)y0?
zLaz9;RTN=&*g_>n;JgpjV@E_~JOzkWKf|VkVtGA^ZKRNJbz4j6p+#cr`g|#PeaYEs
zz~}7Sd)%Ra>_~14s2J!ZAhA*(ylFcI2R<LezOUq)@s46TPW0$OHC-UscsW$;9L{e&
zS-u+uER-TQ^9yzW;JzW01D}PvqY6a)O@VoC7kczyd9a6C{4gEA24&*00rpR2F(~uX
zZulf<#lf`$1XueFQ0u`NO;;CEzZwVfwfSd`c%ovF_cz*}7sALkvR9kOp`T2g7en~Z
zE3}GD^_tGnH{O;n?J52?rCk9J^~mqNAt0QggdZ2ck990a3oWN3FQ^qAJS#ty#Y7`~
z{tsd5RWO7Ok)uRHi&96;QPfaw?dao1*S_mcdZ*cG_hs(1P7@eC5xoJ*L1J3p-R5)*
zxO{z6-9{#ila}Yc5On`daO`{2-b;IZ+OEX+SnyGFXST;N`}IEi1jw9>!P3rKWW3x1
zxzU1AERVS-vU<J!W3531%><}xlRJHF7#~p4wgF8%_WN!Vc6wFetRzLJ_k;PFyXovi
zk`jzWF-jP_Gz<yFr-zsldEJIxabO9`lEaLlv3x^%PhWOo4~or^VAW+Jnb>2p2-O8r
zGB<?jQ2JquqM;`WV5LN2`L5wp&M5l=fx3AdCFGv-0*ZC)t_Sk8DvK!R*rmg)fOow6
zYLE7sGpg0vu>EVD<mvidKT>U`ZUB3oQu1d`Q2M|%aidQ<Pqh-zc1-|f)PoUyP>C;^
z3sP!Q`gwNIkYtQ|LjL3|$3p{u_bItr-(GhbYXj*y^hX-t9*TWOXm{i^Ty^+*GNOkm
z`QL7E4;LT2RXE%T#P&=mTL}s9lI^wA?ku6ZlH4q|U7UL=CNI2H19C@O7%PfAA*s&0
zIGiydVGUQOWExebgj_^TxkYuuxA&DVp723zzXp|ZMq_n$$8jG^#`lIBrw3TmGA}vD
z7iKpaufI@og}<}NEG-$g$h22BXAH_nuzKulQWNgY4beF}M**q7jb-%#JkSAkjV!Hq
zeik(Y``euBjH~{aZ%OYtX<n9;%KC?hc|ayE+^(9%FQuTOk8EA<+KEQV<Q7)DMuhlU
zfmF?26oq%^8uWJW*fU@kCcuk#VS}{K8iSPb>xnWNR2lZ&0f(^9`}u|`w!KWpqQUV;
zh`X0nx#`?XGhYdVr6EQ?ERlv%`S|PSr!*QxMy4T*Os(UJ&4I3M9UcoKLMp;f%!Uj(
zL$Rj(DkHuvX+XtGte-C$DM?N)%{^hnUZ1G@ee`(t8nZKlJM?o%2aNn^@PI0oYfB+s
z8<hA@@-Y2nzin!yok!C`2jq6!<8x<e7s&)$#uRJ4-L_WZP_3wKv`Df2g0A?681#pC
zqd93k6WQCmiCQtSzDL>a_az6MscSS}ez)qG;OTjEh4`RZ3CbC#{d-B^itX5;d6LMI
zCF_%iXF<himp3M2MJV`;oI#bC8!QqnnjQB%l%+8}(<}b04n3;c*ybVXc$9G&^d{ns
zq4Jbwcvsg~v)2cfBt3R(QsM6T<cCQ+u^3`)VjTS1pv)7&lYB>IoNLQpL*%41&_-rg
zevUu-!BI&7$9QI3EZEg++GyUC`28(=<Bx`Z1TuHYFW~u0oS_t97}~A1cxSOvR}Fd*
z^ZDJx1!bYc7@04s<Q{pF^mWkR;GPKnyDANap-T(L(Pr`YeoX%O3$b3SR33eV4w;-3
zA-|QOB~AqT7KT5>SAk68^Ii{s2&;cId4oium;4^+mU>KNEv>HtnMt<qV)jTU&QMSF
z&J5KKH_t??Iy<Ga^=A~i{;u3)c$bw|TsbYA`zM<EZ6ZPc@bsF6Ko*hFXzc{FJo+ZX
zRhmex)ZKQM0h*OX^N6SAFI?G&vE-ezo~neJW^8p%NwwT7;PW{mZKB0q9LDb8%Mk6l
z81M|&(Bd%XRN*;2V$Jrt6||nA_9%(d*!SIr2&j-El*;{GuTe~1;bT+radZe0n*8RY
zDk=I)zMT>OHi#;j<)>Xz-6FWg9Mjlyl6#_%3@@iCZ}C>ew2YBnV;v_44w;#31GUh(
zpR#a+-*k0w%5vHcGnt@r&Mm0}8kQ>F<al$d?}2Ri1LsSmEs*MtwZX!CI!c5|y<u>K
zMs`EbB>kliVOhuR`8TNB86>HDBwX@P<txi@0xFX3Z~<$N*#ws>q!vWJYC8KE9?5h{
zyMnKRRw>V_p{+xB68~(D<Kr<}A_gDY^X*vb8)PqCU&QU^xGqm*8O*cGj%HElMuZM$
z7jEJ&UZW6w8>H^1@=4FFqpwU4dii$*>3*Q0H%L5T=W5o=mRaK!2B%SZzZ(uvJlWNF
z%t7vhg;H@*X^wlwP<wKQv7vNg+IvX{L;)X00IFoFmEcMMN70?ynkXqO{;;g32O^KR
z%%Rd(q{cbRoM_qBBG}{Mm#)IW4?{A0ez1(UD=QD7x|y$9{Frf4|6*90g!eBNfb65H
z)Ke$v4hGjqO206zr#>wZeyc-V^LjVw&Kn!0pQ3St<1SzJ`h_rGk>(N`x4>4#)e)Jm
zXP6yGr~J~byejXmeP0cSEXMA=rn4(k=}b+hsKM%5{OQ#*z}5V$N{(fR#P%I}@NOK9
z2KK-okrI4PGBtE{A~`Kb98wx^GSc)kDl*KD-fgF6q`>SK>*%Dbk*rGNzgFhA9atUf
z>p05wpu?O0*70WjF8wV;ItOw+){(9F$IS_On|u5})d6ZH_W>23kMxW8Fb?=L@6YOH
zM)DYnr;Co3aJnVVny#n~CYSP+r4SHkFU|5ub=pBF<hElua;ZYRD<<FIR=8efB7W@w
zUWqr6H5^CM!B9LCoy_0Je#s19EkA8`cy@i1g~yFEcLiY=!6!XfRR+H+k-03NSc}cb
zyj}_ZOgx$#L5BW9z5v*7O~T6JkJG)r#7<Gz^px3r4`jw)_;}hHpxDIBZC~-NsIDV4
zV!YfpUPjj|9!Ib?P~oAadPIQwJ<!%C{IJKs<7~-iR}=_54eT96(2>33t~{+K!<l!t
zX*f$6N5&PqA1;-l2Gpoq2trF`QFTOi;I5wYTOUq*3c;Y@Wl`sPJ>tp<a?qMN{^axB
znzUrV@SWXr-;xhb@n@GUT=dimw*Gbf{ef&0)5*#6mTQV$w>NKN1%vfcNYe<he}d<u
zBlF=1%jN;>SFCt>Y#7<rvwYYxe%LqRQATZi6n0?vwVt^4C%Nu4Jq-h)cJi?s(I&sz
zh_A7G-jXwf`nkQy!}arF%$PrBYBQPtO=x*d6)721OD^?F1f_)RJqwJz^Iuh$7gop<
zRy%JWToux?vyQbEAm%qk>4ALZ`?h|*425W$=HqjI@(v&;!Bw3a{gI03OHOb6nww4c
za{_>~{MSb%EtBzJzui0J*ld{LGaaCOIE1E&W9(bozRUVi%4F-b8SVXPw34ftw;EY4
z(i)Bf##cpEdJVXNeaEQjCcfa$pa_~|Y9+9>#ic})CPCt7MksgrW`y|uM3LAYuKL|8
zcEH08z0ROjJpDu_mqMXHq+3KMYcIpUwTa?f90T#yc@l%{$;}pJ%1Nw?`R6V_AIQwe
zs@T1*v$A_R^i+@a_@n0UmU4VtmX|h`0-Sy0@9t~RphQbsm+)s7<m-%y%odr}-L`{h
z^I;)B!R1+=VX4rpdRO9elZ<E}8|{bg-5A4<BVVW9uaTEx_3(XR(0cD`eVC>7@t@Vj
zLPuR}qw9*p=Hd4c)23N)Uq)h<yW!HvdBPEAp~<+}ih7(<pC8Tde8hgdD(q}z&~uq4
zh<mW~=9iMV{uocpM%Jeqc`@qX>(iIc&zO}AyWEB+lA+c@fQCb@X=(6wbYSi!KP~K1
zC=J_gU^elwvgHWR$iuFsMTpqKH>0NdtRGy#Wj11xjmjQ=0(xL3H&ACRJ>45Kzi$P<
zrIS`3vXCZ#LEgk36lCk&Z8)-v(W-eU+LFP_CUd1e<#Y6O2(9vyQoKjg54cwPbpE~3
z&(#?2P&{SfJ`NRRwv+Ub4l$!m#bvX|tFa=3FKr)re4Ua+ENXeh$0G{4h91zG!eD*8
zS$anuN1<5-X@Wp3yHiMOUmdYg=Qje^$1v~G?AGe`S^4~qa2b>U76!jlpcLK4!J}$}
zV?`=$Eu88ed7;e27sb@@Gw&^8t8a<5TT+V*xG<spJs*o!{TQJ^li>I;xy7!WoK?{t
z?0L|kxKY_Q0MHuz%jC;)-8m}ym`I1bKA<xXjHJQi=Z1Z;7FD%roQArTALZ{|Uo>9V
z?MVF$=QSgie?!zIPhJof_&%jpRNbQdZQJ9UWf6`OPdNVtmOyFG#>8GgGZ&gTb<E(+
zamFLV=!hxW25ch74A)*H=zv||cmk!W*>1w(Z0r0mj55hkUtER(HuEoikV37(V<ea$
zEm%X}+_@$X^)9F%N_D9b8o=u8?^MW4wROdYzyt(C77^&_QLXWp<}LstECew-<LLpC
zr%i+miJg5+Fj2<mDbRnI^0n*OVskJ42mSrn5(UDh6ckZm8sB60R#his5-|p<fT>;^
z$`MXTCiRHR`+%%Mh3W1J$4Vyd!!Thv-MuJ99;$PaOt6;>FIjUVGD^uzedZyFxiubz
z*QF#6rxy0;;H6N({Hnrp=rplnXkw4>?pQ|Xz@e|lk>=WZCrfPDRw^Xg`v$eY_t?kw
zYEkFl?Q!F<t#gtFf4)a234LyDokyuEDjU9OI>?l725OHpCI+y)2CM99?)Dfnm^=#O
zHD%Dmd-1v~mrJQe!=~JP;E<1FoY*PpolitaoYx-Kzyy|SCyFGlZ(}^^he6Cx#Ty(C
zBt$StOUD;bc@if8qCO4|jympT^qP5RVlpJgR7aJ{42i0L^moL!AjN^KXx?(=bC%)=
z%N~|iTA^AXo=itGYXG89vK5DSno#4NP7fzChMn+-+~&U=Pi8$EF~$y9H^hhMIlA&t
zaVIFYwo0LeeFmFWl+K!Vm4U@)u+}e)D-K1iYil0+?R6c^;^~o{fb`t``J7SDFlkCt
zDi(?Y?DlDJa47g>ONhEFllz=s6xYiG)p1R+GI4d`pMLXc%C4Py2G}$<Rumv^J*H+o
z@$>O3oHVJtma#;o--&QZZGf!ff{I|7$x?_Cg86uKd+<kWWF?$}fCI<&_<eBq>Ja=L
zr3?rOk}Y27j6%fr0#u1*yU)=N<Ob+>5Q>kXnw;lnEEZ7Oe3{}{w8t-S$v7~G*ywJ3
z$ILC|9)*cXiR626T*0m$C5E~Qg>Uq@S6J<%@%P@b<VWT5Kn0esfpD+6qVM_aj<j*5
zVc*YC&Uk{36u7ZE)JIhieNlQbAZqIV-NqHRpS05*VHARM^X<bT7YWa`NLtnSX=Px;
zB~en!9@e+0yq_S($fbiCh!$X)9K%CEBKunIQJURGaeJJA@xXKv!#E9<uU%9XJ>NwU
zNac2DX(X*UW0K-sA=p+D%pFeHh}^PGnZ7%ZxRGthzoGc*V_dvY1l_p`L62p;uD5T6
zoFEgG&e5d@Fo#^8hlk(1q_PHFay$EuVb1HQTWuc7rr=MGOMNTaO7~Z>s=Zo);IU|0
z`R?Hs!s|$6R9?G?E^L?^G_k@EriQ_rssXp165hFk-vL~s<*`w%Dr4mvuIzW(sn1$D
z{37#O>qb42x|CWAH(h}a^p2d-Z&1%8q^w6ZLDWeW?}s2lwUKv(>zV<Urn60-1IIs<
z*;I}`az*a9QN-xgtu<UDcitzo2w>o}f6EohmY--ytPJC8CsuCr=S0e_*A^TVDToc*
z_LN!BX4`T78g;*aZ2FY}rG<HHi>wJtUl^{4j)YZqW!ju0iYs;PTJIyZ<mIP4JRA`b
zazcU~+wq2-Y}gHkavltg`|j|W)2`|E;jL1R-roePbad3Y<bP?a;IP+>*#ZxYlz|kz
zqYcqa7DGJ8*1VB^6wKn-2!zqoyw-2<(<n5%N+39zfm_7qj3#wTBusTGsj(8lZDI>6
zIxUYpCE*<=CEr*DJSF{zgyHxL-)`@>I^E^c>N7#}oU`a1{BXxj-e>6p9K+*ffv)5!
zi_=rV?vEyGCu9YZrLaX+=j{I7DzIM0WpiHAVJ}bM32*k7Nb08HIJ@iTtLMC7h?+xf
z^jtM3u;aY<GFL`h>Z2?P<L`S-la-`3`oYa!YNu?xt!KGi7Y;vz>3V9ol3LZkF0^IV
zOB?-;wqB13tg+o%>7xIK5u(zU@IXne$f!%IL!&`7eZ(ewp2B7FK;Ggs)A~6tvuQ8K
z>r+emo&H_^uOW8&&I!Katsdiu7x})8LPv_6{t>*n?JpCviTi3x#e*G0;HMioK2dfq
z6Y=lkyLng{{(v{OAkVVr@k7>1Vbk~bfQ~*?=;9u6OUYwX9GHSkfC_KWR%f;EYCYG}
z86gq8w;s=EPmQaPPy1OdV53$36BR~h{1~Uot4%+~Ncv_nG*)>Q3p4juAY5SA6{70h
zNt@`~&#FH1e56h4ZHMJ_lnNy4v^<;!l8+K|fYpdQ4PP)n5^JrH1B>(QbI($7Y?$Xh
zq7&tyd|CZ#4+LhJPDehOlxNX=e0L&`Lp0q~zxS=AV~_W0L<GrI^>*Z`_cRfP?y2zY
zSRb??(;2sDAB_<M`S=7Xk3+II8@I`x-61A*&%o8~aPa|`3K$!7`D4!v?BRW5w(ER}
zaKvXlM!&a;6=|F~{iXf(b9}hV=jH5^IP=}}N``Vi^5cQo*YQGXm`YMlA@a@un@wl!
z91d$x5-f>jlI~W*yhp>1h@_G~*vAm5iCHLa#<O&ymch$$-hXl)!3j^4+ujRhH7bZY
z+<lB#Ox<klxO%SjtV%oWiw%H)MoEQFs0eRGV_`KLcb~I4B{>6EV10YlL!O=7fQSSN
z*@37xOU2b;RbA)OYGkkdMwtrHvpt2^vZ99Ht)^Oc=yyez<_Y^gdqY+Qzg894WDhpg
za8io+=H6QV_w1wud(`*?TIO}zAu*-J*jv3RSD#8a_chz;=gZh!M_#(ZuhvCsaD(>-
zz-7VO_3Hcay5}MBo93M2u<IN)B8{pk#V8+P4}uRBum54gZ{xVB9KfK2u_SHVW+Xlq
z5>1Y=!b)($ZMExQ%7nen+QBhkP7i@{--*1^5-{fU^NGmzt^bTs*Hetvv_Kg2_)}px
zD|Tj@yCUmFtHzQp4whVV7=%3Q!DWen*jIyMc)Vfx2v2cmR97&@Jo}698Tzr-&R#pf
zH<nfcfk3}<59}x0E%llo*5^=AtAbo|*3deFY-t--F8ljVe@@AHxzjJNIs2X;J<1jt
z$GJ7@p&4dLBU9(^f&$${m$qSgUh6J2KPqaYm}dtP)YPkWbNvb0Ht`57Pp}cf$WfO_
z1aN$`hnF+J!<n*CE!~|$3k~qiQ6vni98@Y_G0B+8Ye#~-l9d%?_{e&Et-U_a$=O*F
zoPzy9GpCq$^mEx$W1OyfA6`FlZnMJ&F_yD6DiJk}@Pf5712Z)z2Lt4W7FkmmJgpb8
zjE+R40jQ8qAJp=p3}_gfy<FyZr^&tb+EL%Tx4Yk;gh*BC5krEH%^CI6=saDhQAtmZ
z;jHmN`aF*-A9-1Eb_K8ko>!Cu_r$be6?Ge1B8+P}V_0zDt3bF(wj~+W=L}^#0aC#X
z-w>MtV*WQa`P;^+P8-rnddSPs0-|!m<pPv0EE>UrlT^VJ!wY7w)@g3VER`anI1lb|
zj9|jMzXoA5VR`Um0CR-M^X>%gr`ucsc@G4s9#cTykLGCA5wWcsn0<5^5PFs0vx&BJ
z&5CFvH*Cu%iYZ8nIm6wGXv}wba_@`VH-c2DYX+on#q<|9W?^DX9!{$1_QorTtL_t1
z<tA!2ai?}-&aEJkw#u-|HezNI&lwK&G3(ueB#g^)*!nRHM)AzK81M^yUNY?0h3vx<
z-<*?-+O3t<a3y`>v_FW3o-#NHJcm$}C1Ds&%@z;mO=9umO9i0pJEdBzYJWKjo;ab?
zg%)cWj1`Y}OTxr&Qr(&dw?NhR=Gli=Z&~<9^|Hoy9C^EwDys-C5jQ^5m#HJeW|F}c
zXK%bL6+i3ws4L#AKvUC@ofaiKt8O&GgyscJk#B{c8aHPs)zO&#0F`A^AP)~S8l#V|
zILSLiU^bo0jJN`JV$*m}4QNmmrp;vtf8Hr2PXF~C(>INC<+NI@!l2yOv&-Bsz6*i=
z<ua7L!d6{7`qs|3Bp>%%d8Z}YEqpRFzPoZ_f)BOJ-^!;%vP56X-gOcwV)MywEywrC
z{>PAq%duvEzWv4Uf}sF~s9`UIJ-GZ5%?scr4%Gn>mP{fl_R99@E`02(JGo<@l!EJx
z(}k$;iCbkRNRMc%sb9MCQg}C#X-64Zy?E}-ytxhv!(WU^E9uwqvp|v!5K`v7#C1q$
zn|qCF-2N<d^N}2)OGowYM>6y?s2NV*c$_#d%D=gFX0Km!g<{9|&634MN3itE(`%-`
zF0D<q(z54JDrDLy#W$pD_&gCRr3~lKGg`O#wV@R+6%EoacCa&_2vZ_j=z%B&Zgj@u
z+y!?02--w$eKgc646jAhmS&g`Pqtby5xtrbA`CKxF}85=(V}bswp?Zs&7qNJT??Ok
z>2A^8KeD`aI>sq_=fM)ic(604PPR=7oF&C+1KokLD=WC=!C2h6vzBzJB?CUemDY74
zxCAQ8K9C(8kv?vJC848m0#3e~cZ2kht26yM_X`MHx_V0em)^k)ee+@byjhwG2XGx6
zULro+GHcJ=rN@cf%yl5CtyXPNyD!Jqyx_)1F7&1ntMyQ3qe)VM*8k(M{>qT4aB3-J
zfOpV;$j_B1dPClWpll$y`zHi}8~7}oACv+%sGcdFkc@~%SgvNsJ=;`YJTzbp(T~vj
z_;1h1JIpsr4>aFDvARp@gyrKx-M>sa1W_%XYqFTVG`f3*879M~QYx9+#g1ulI!<c6
z>(E5~k@}N)xe1}T&4I6i=vcJG`YW{uX(}OphazAcPJsBKNc^U}IWy41kim5wK3rXx
z^-I!4Rr(<bb1^GcZ7lDpJ?>4d&o$|owwrW1PtKXnRlJ3_t1}x(9z5utC$TdBmAqCo
z-|v&`MkZ9!0223V2UVzl^nvj`YuY!0rpKE(zl*>(BEilZ4jD6eUyu_TNy5kTHA!i(
zsr(*V(A4Ql7RfBnC5L^}D~IK8d4-SkFGmkoAK+M%z0T?5=!tIB7-1=vz6F)8<f3Xg
zDzJ5R{b6LTE4w!c)$Dx+rSr^95j(K>Atp0}fxRx~nd~b!M0<2gu=4<1QST<1SOrG2
zJ#Q<=@72_(DY!CDWnO;~xIYf+3HKB!BCAw4h0|e8p0dHvTxc}zaV{=wotBbGFCuGa
zzKr$v^HL0Z3+*(e=?)eSnAf)s5r7H)eGNDV6Fx+)pKYR4<WG?%jmuTJ^`KGNE9Rw6
zE(<Q~on3SFhvU|rdU6t1nOb7}RKV(}Eb=>)d!45G{<P5MEnlvt_k9LS_27c0%a%*L
zk~tUZiK4yNWO`EmA|F^YoX?+ES&IVTi^cQa>?%-K!e-iWvHMJow_%TlDmg0OpeCgF
zO&~Kx7o$Xc@j`o<$1g?8KjjY`p&{-jhLA}W|7<1uw8v@xHFUsE>V>te=V>$T<urQJ
zT?*6wXL8fQ4rFB<%RiMPWNkLdx~ZIo)qb;WcV8Vo83Nbrdd$L3{3}ksr&Nu3*B*R#
zjiRTWYhipi7Vzz?OS55N<CY25OnFw%1f8s<^9!P~bR{aV&EUi$_lc9jJYe|equxL{
z`)=j0VTQ!nigmrE*>}XPcsiwxSzcZ=zyn^ZYf+#Jjh}i`VVjVi4_`F(OX#}ftuQx`
zph1OL=tyLRBRRF+eBt2Y6W{4GznQ-_v~M}gY+@$%*j-MS(M~eM)<~XZ(QTBs#y3gK
zi8X%MUUQ|p2bM&Ed5+?v8upn{al|UmaLOoMC=Q5CQfPp`(5YEozbe<yX{i6@Ou>|&
z&l(Eh5VdC?f0f<1b}isobA5GjLD{AiCY{O3&%`13DkSV&oWe|NeFTn}Sx^cCyUU>Q
zYskM0n`Ac`0)L<k5LkraP_y2|5p7oUfn*{b_^PhsfDF>iNeo7?FCfv5@kpAH1PwBa
z%JrNW&ecHX3Sr(PxSPNsF?yGj783d@j3kjhM9j^YF!45rO4ny($nCTG8US%Yj^G8c
zmimv4x0#_?%!et6jqMjdQP$~r`umBEv6nBJu+H1~&2OG>UKx{Z-3foI<!Ia7Vnm+A
z*me6R)mZ6=+d0jUXTD2cY(O08jW8K)*W$U>n_MO$yNh9^X7t{%=p?27pE}jGp#9}v
z_cZY4vFriC=k|-QXkG$LnuX6y_riEIY{xr3^4wH#0M@fzY1KiAnuW9tn*P`y^mlb!
z_I>N^sqZ_5`4DwvtfxVVgw`KO3rs(G+fB)rCXek2prQw^9-s^ray}2~Ev-}ve0MuE
z8v3?BmiSI_R>2bjNvRu?>`ul^P9E@~Vg6Ru7GhVTNsX}P+Xe+yI#ovNd?xgaM<}~C
zl-e%k6da%t-K1>v!_17=;SF6%L&t=^qzr)Mk$4rFtC`#N$yw?m8=2|j-Kq}|Pf<d}
zU<EqgAnp(TSf2pVtQ=_famuu~8li<jD9f(~{QAdM>%<P6)R6~D5KA5r%wHW^JXx>B
zF=^mJm%twrWnKIeV*=3`HR3uJb^8(D2NFX!yT;BNv4?3Ltmh7h<g5FgS6Z%eXM~Rz
z(Yo$$N@}Ybk(<9#71~GonVFFxVzOl}7zTUBz20>!CyruwWp@D3fHx(JrBVA!8P_~Z
zQicoO$8Gll1>Jd<e#A=xsEe6lzJ<l0Yy@GjM)k|cat8q)%MnR!S^W&;CeV5e7$c!=
z`mWqKrirRi-zkAq-vRmT-S|v{3CZrD$2WP095LX~C%VFW#g+PjxUE|AI^53cL8bC3
zLrRAT-hu^1{7?yD0F^homg5|@C%jB5wy<{n+XR3)jh*1{W89I$6Fq!ZkH-V%e8foB
z!{Oq%F!Y-=i>dVrG5DB9z=INspSrIP3VL_99^?<gOp4b*%XPmcem2JAntsfy^CJIb
zDzPv864))!hv($<n`n()3L_UE%|7zw>#46yQ)`-)|E>NkHnkR@IFq$5YMnk-WMx$R
z;-K#6k9cq?(Nk!w-g!mF8VlIgirzwn>e7Y2izfStie@D9vJ-?*Kdudm!Om&wo5CjI
z9o5FIS+R9GvKrY+<SYc`iN9)%1LKdsjJJn|Op^UQ&6!ZseCgZ2raA3Vc~5jmT=(=?
zi5w-@#;^R79nh{s?)OVO$H+aBr-Z%rCo|sLLBr52^4NF0k*<~1lTtr0J-1+%5b$mP
zm!p^A<Bg=tuc>Ac0j(Y+>U0%Px|%{I*rNFONDq8j?!>beW8&t9g*aorJN<OoocmTK
z_}~{UOmbbl4BY)YHgMW?Bmt{R`hk8kQ=KH4J$+7sUAqyM1q>GA2$K`zn|>ozw8TP-
zpToS8X!%$bZZz}?0Nnas8n~Kze%S!_i=~711!nY5mrZJ7`Pn-qT6`gzNC~BQK(eme
z^=<Ka$-^67(?9;z^k>;u5xRh{j68fE)m#)+uj)p1EM2+q-WrjPVs$XV#-r0I3U~W$
zlQ}eR4ajYwN`Rs@wb`0&z%WZljh34Z*T`jvX1D3TtMHwfn>A5tuwO(KIIs*oZ*^i{
zdn<^p=tN`;=>cnt4b(a@JHtHb(VoUIOW_V0zyFi0G`^os?d61+D&rEx^}w>#{W-YO
zoUj73j$XLyc4T29Xm!Q%^zL1gEM#NY>A+tn8q=3x-Q@4IofP8{7coXSlc@~NavnW0
zb$x>rM(12d+Yj28Sl(3;e*+wCrr`_~sdE?~;g4<wgr8^6$BEh<0O24;7ECOjSA~mT
z`Z86stxI2FyM~@CP|zLP6__W~pv+E&3gRkBIi8xVK%_jxA<uK|=?1g6eL2{-K-J$j
z9p<3exQBCr@5GO@A6WiI!0QU4WP5c2+{`(h+wZc!X)>EA_l$pI(ZDq{GHH~>!r~8e
z9}srRu^!3_k$m0Rolvjj6&>LsHO0=Ash=%l!G*e(B_l@<#Pkf-PFR#gP?`w5l9~0b
zhXwy!rnqxA{y>WCvWc`|PT?(s#3@5T7B#(Sl728D->jR1xFN(n3+<kahte-J9)I#4
zb1l!W`|V_319R&HY0J`+AJI}N8W(o|6!d3&bqj`rd)hPgGi(oXOm+F~wGm%*K;|}v
z*+2H-mvOu8XbfGscU|ME$?M8!#`;jM^psjBdSCA$%J=oS2S003L|AbQY!meuCvG0J
zLA?Eq%3KpOHLTBSp#mEQknTjh7Q#_k$eexn8`gwjd(tKxfzk$J#sj}0^0XJcoLej}
zF5t#WYAzzryfy!naf8qKab?crVVYI=*o(=)3dZp&%*#oacUsZc9&0paT-X?dYtDfH
zZ4D^HJ{6d6^a&O6I&!c0jKV|5|I@>;`pc(&0YUKsN)evSdM>1me&Sh*xhfr`ZLy6w
ztdAP{t%6aZQVvKFmk95#OQVLK?oc8u=oq1Zw#f07&xd&gx+|1*LhFs>$K9P$yPcPY
z?nl4)J^WwNy8uU};{+Dg&rVsXxRi#Dj{$#}weewM=U=0U&Qv-CR4~m|{OMFM7T3#H
z6wbwWz3>GlFzreO=S)gkv3~Xj$UcE4Wvi%&3u}9??Zpk!V|0%@KD&c*J3=X9gd{p!
zT01eSqlm7kg(sMtCCAt5<6`^>aHq-5ljptBm%UHtyxrE`uprc+E$UEdDmhmlC&_Z8
zAY{?83FCbiE|EuOayhFBY-&t&x+CVfO1aXRP<tKR?QpLvz|k-a@KQQueTa;JDxXNy
zYo)?yCa`6R;E9XM`~2MyLLdZbID4n^J#r{VrGv-k$`m#{me~~M_0T%oa0Vxt@0?t>
z$b^2wgm6%e)}*;Tf*o?>!VUXuz?fNQ*sX?7@zrfrt`cWd542whMr~|BWl_~Shv|%L
zC02S*=SV9>y{CG6K-D*kL9u;Ud&icWmh$hU(Y)v-iCQdGl^3;ONoucBL--P?Bh$j&
zQo?(_>$S6$%E>1**v^TLK4Xu-EmfH0XIl4#2T+sQYG@P5hUOFF<~F1RnsLifei4sT
zz84k{P9wD<+88k*8F^?;rR_JJJ-f`mWgtKITh-vWy{is|zU!pd(R>$Pa@hD9L0r(D
z#MSPN^s6PuaQ$LI?u~_{JmR%gAI&2!Wi+#j>C>G#A)6?&<K&C6d~;NBs35N7IbE4M
z1yW}P8FVg%6uM6wN^{q|$>!Q$to!rTfgFof%P|L@?s68$F{+h;A@GLr!{x^r^L+&+
zZ4?y?**h^IV&V<2sCwRh>+t6jfqFBv3zU2#RF8<7!k|KmID|@IB>2Jf;`k+>9XR-+
z1lQ2nd|nRdbE(;$^B=A&DJ#wsrlb`r`&p{9DCr44UO=2mQ0v~<EZ_`+HJJ|`{&it?
z63nZy^r?bq0RuWHp_d<|2fQUJc%x?(kO8Z0@A0)Uvp^1Z#p>|DWP<G(ENJqxL)Xvx
zpjQ55+St;voM|}zse$w4{z6dNY)AUE#4qYCNjyFw!SH1j(WBvpZ41g$&MoiH2t@yr
zT-{Lz?<Y%Is=1{$>J-mD;)7%QM<(n~Uv;6<QD2R5ne|q>?(|fLRzmN5+fR#Z4DzLi
z-hoqZ0Yd`wg>i9sbG?Uh*pay4^vih1$s6ueS_er-uXuA!vVBR#5}LbSG<h^Ai@d;D
z%VhsjWYnYbenoq%7_Ze$0oO7re=5XVfz4$p7RP136ME5*AXb|@Q1S5%b)AJ)=33Eq
zJbkwNm)Eq`!OZF%SGBOL!l~iRS_^pI_Q+OqlOvtsL|7dFy-041dRD2?lmujdFDYuJ
z!E&HK9vU|`s%O0O*7#G=rl;2WizW1==6Mhg@QnK}iRd3)R>AMl`|0>s*E8+#KbrRc
z<vaP)&GxDX@+@n<=rtpPF17JDU(r1OAv?M+ipXw*R5XNen}Vh#4>G4d+j6|F(7;9`
zmP>hxfh(w=R6XDCUEKlOX-!^+Tp)3ZA&T|dj=mdSOpaYD<4+%dh>%KZsP`d8RERk&
zN>LMKO8CBAf6ZzkMTx)W-+Kqf!|TL*Nr<mQUK8>mueKXUhtkw|GOmswv@6;5{e=AS
zR!89>QdkK*so>Fq#`eVRxNy<e7N<{94}{=md>ru0I;ziaQP{^nyx4YW+}L0#6=$#0
z=<LaeoO>--=t;VeJr=8jU@Vu)d^A}z(8cZ-a+h~A(txK8p9>xgZM=kZ=e^FNCbo&$
z$z9scPO@1)fMlIL8aSzzsO&^@_RPFhrTF1Mi|>z|_xDEdeECnUrJymxL%hV`qdz0Y
z8tg!Cd5UfKWc1AxW|g8325uSrlIWf@d}e(I$Q%V`gEwh8(Ki)m2|rEJ|BvoUuK{Ur
z8p{l5X#K+w|MxTNM@Qq!K}BXYCOp<rMZPE)-_vZZ2S|WaB|3_!cmBlQXq;$~)ML$(
zUpGiAJ%*qr+#Pqr1BX@8WGswD9`h5N;_^-};kMw{czurUM-RZV_k|DduRX@TweSJX
z*(q&oFBaHzB9wH{!mJbJ-hk4pNMmo=R8P{V%d1gnEZt(2393;If)j5dlzA4+GqlB%
z!1XM5;Tj8f7Ta~ZFV~pw@>ZU&T8;{V9BZuUTHpj3JGMjkcO&O!T>ew8c?v;*oYf|?
zx0GYlAu9t0)q-O8C+lzLeMn0cq8Lr>HzY6wGdtiSwGo#fl(*8JBCyHQD957TJVj{W
zK?A)wh0AHaM%CUOEI%MM(0>v|_d;n>!^#njLe8*r$YlUyMe085Q^Q9Ss;8V<&s-p{
z9gtfnLgNO%_v0%%O+By%rGD_LXQjXN(tMPl?fo78<zz1~yF8<hT6kYVT#@zxf3%4`
zr36g-&BO2T^sQ5tbF}OZ+lQBImE%U_h~@~6mK62vk3n$V6J+J|=nxaOy_70rmEBvB
zuXprR#s5`--^_Ww#c&j#5wMx#;p^$t1j_IVyw{#W>ekV3rEm^iQp8wkF}xdr235+T
z*K59RuDe7SrGmJBKs{ixRq`x(Rd=m#iq)7+&V4j@iqn;^k?0ms#FT^WEA{?ejQ`br
zM1oeM$QhN#Sq5n1oKv8}LE8Q<;s1wY;?E0lSZ;>z(Luf*XQBjg^2w)%s)e%!(T+TX
zsHQn`=?9ok!ddCoI*$TSCHr@0h7K6;-=*Q}7IN%Zd0U`cee4|3OQ}}4Wx|Dt?VV|{
zz^uuC9FC23XCEgq7+iOiZAQq^>@C_2Hl^t;Zg<kzBjJo=p{4{k_I;JPE+sKveYym2
z<aO_#<1`{D)~RaN4_w60N7?JkB70F)J0Sl-Y!;p`NlAG8dbk(u4%Hl%-mBQLWA7y5
zvscacE+L;UK?M8Q0~5fe53E{A(mR<d>AQ~SwS1{rqvmX1kw4~kFBaMu#M9S0jMd5v
z=)tLn?pau#;ZN^#uTzgpnE-|{fFma*wxs6h5w33>`%>DhPKi1tR-$<wp`zfv<^Tc0
z))_RSAD_0~y!W{+)A|6_Lkws%9u`bE)xM@mDPIYDpX6{lCDp3b#aJeAcf6XDCHKY<
z3>39D1AWtdfhK0F4V||BSRDdvRkumBenb%;>Zh6R#|jvmdOl)h3Ij3At&2y%<tQLk
zN3CPJ6!^}z3bZ7e09$sxEk8ooR=u;wt-9V5#FZcJ;n63cDA&JdxVl*vFGSYm2WYeO
zI}VUIroO>*VPX*;mJgID?N5M7eDCaoiOQ)mPO19dTtG5qx6akKP-gXtA}|Ealp-XL
z+r0;5XGUv9fbtzL7VP-x&Z6WNwEt#E{fLYEFBBuDzR=b79Hhc0W-B1AOFJ1;5AR4M
zaE=bCil=<v(BJ7oWve?yUXp86MM2wcQpKULH+m#_rd|KBW1uKjPd>;<s!;X;vy2BM
zf$06^a@g(8j1U&7U;piuWEkjl=XN7>#^V`{fPwv8oofd))p)itBZ%eQ33iqf%xP#D
z5kfqdpn!y>2vl$1-I7U{FxFS22;7wCp6uOUHm1FZZ!~k>Yb8B_u~iJK8ypZ(tkIPb
zjDjjxR}oGx!nYDCw9eG1?RTM&v4=xdjThsxz^Doi!FuyY5obJm22Kt%_Ou879DM}D
z1F=~c8T6A<lKe7b%sCzH6Xt1KixLh`;U1s^)*Ljd#I<Mm3k2w5J`GoIay-8H4tjO{
z)J|tPo#-FNieeE30j7sU^r~N)M<#AGxqmu5C=(dabEm?y=wJ(JC%|&w(DS|$=zK2~
zn^9~_3fWD+cp*Z{s`IddDU`PZKa?O->Jy0)n?|Y(7q{aq;v|@y`@G9Dgf`&>6nyt(
z30W>Qj3kh<k2AOMUGaZ@tnGMF33oyHoz@i$WxU0?;v~&Tyu6o{%JCy<<jGB!fug?o
z$f31y73*h1n&4tEw-9<2s>j~maD=_*(X4v6zPxBSFNDniFb7^lK3l~wx_3mjFO;Y<
zaLga8Ji;l{bsUr7-ty=C5f84awu-Mpc;?wF@fo(tneq0CJ9GL&OBO(52%#B&=MG;t
z1D_Y@9`dF=;r6(myCc53A_MQoZQn1kLm+|W!A<Mf29p@DMPi%c|Hsk<oH;+LrJwL5
z2ng1UMe$Fd+y6mid`EfT|Gn{{tY6I^Zah6fbsa>7OgAAgVZQ<@)0=EvpALAeNFT(5
zz_#{e_>e=!Str(BUqR#5d8pkUs!7m<sOLS=mXHl_F|FPG$P*s>XbKCIM&9GYOKGH#
z3_-|Cfe9-|6di?V%V!&G7IvlctKP*c-lr>|wVV=4c_r#`K#~#BJOjE)Rd{X-4Ln=$
zmR>zLygF%|RZie)^-DO>IBo!Sw-BWKPn$5n+(CB$YpaaN+K}1>;ZuKiKHd0aw;5k}
zVso?J6Tv`hCn>pE?<a1w522Wu)_0j%Dk0ipgB7b3#yCnA+hufSb&I*BKlabg?f-Vj
zKd&JR8k2~uUy)}k^Vz0^?<VX0B-g1rPU3s0X@ierh?jZxZn6(ztyUf&2<biN<96LQ
zfsVoIRNWScR4{YC7+9!g5EdTRd3@fA3guH$N|8@Ujju(5-m!O^&4+ZF(o=@*ivBQz
ztFI!gkqOl8xSk&#a;e~P9^<09)(yx~3PPUVuMLVoTEie&V(i?_JlN#@SbNi0XGJ{h
zCLFS6BsU+#0=`LA)P`>ge1J`#;T~J4O8G?W?EzEO#$wR3>@wlQZeNidWm)eow)Fb6
z!A8onq*0@L2HgGps=}W14L5ysG-K*YU|r;)N9RYF4DWC{6D#PM<(Sl@)+irnFq)L>
ziKqA348vk@n!($M(M#Aoo^dhsn>NQ0zQ>r2KXe#ytb78pLDwwjiO7|%CB!2K%e85k
z2buj~(`PGbI~PQPJP4H90QF;0N7pBlRhoH>)6a_QMFqMXyU~pP-M9ZA<Ly63ANW5G
zPJ&ks9w3nX4M;67?c<>QoB^{oGVw@8ME08cONwv;`MSgZ-c=h`a4fFWtLFrtXSP9%
zeQ0xHc&O~;lW3n|EH9cSXL)I*6cYb&g2<)U)~eNh$)YcN0lsKmVPqx6C3Ol^O6{-t
zB6FtNh&v(FdpF2c{V|@>vYWh;|H`0nO1Xyvd{*GCWdDI8JQjOe|920>f1fb_`~W;M
zPYM;C;K`AB-`xmkuXg=r4%dc~FwC!H(a?Avl|uR$MKP3vFLoLmg@H#@4#G?F1^mU&
zWp1eB>U8!C1es~`{0{L2%`*Das5<~&r;NQ^woi=xAl-T@s%zkfOhbEy#KD7>*ML8Q
zYRQ(3#J@YxMvO*ge7F(-O<lEVL^`Lv<Ka|qU-~vfIJC<z3oA203A;=vUWVGu-x<7r
zfdl@!^!wj%95`MBGIPLj_%`k(b77$<ilxA>M)IBwo$itv<T@Hy#=e_Ib<%?vtn{K|
zki6ohCVL~24Dp(m0Pb<QNLz>ErhEr!J$N^#!B&x%CU>ZC3F`NpBtGjzT>!S#3AzAX
z_j{9apyy4_&#b=WHPh2w6lP}{+|dcWEE`a)wN@Iqo6kV83Oez(tIdRXQ0=DEdJet!
zb^f?3IM97>qV6wcIJM~iuOH=qem*i!Or;c|5q(hcrUa))IH1rLV7SHCfi{1XVNz(e
zZ6iz^xu&`ut7~ItnP1|YWS;8UX359)Z>9hk8j`3D+WSK#{Qu{_9uLQS!^WsnI{cY!
z1G(ebsB0SYVj3L7ohvF)y^DX3t|u~}S3#4I$-(R&Km5N>oqzuNKVBKkh-x!dK4i1`
z+X}~ljzPBzT3UPS;@yuY?o^leeInpsy#q0s#RU)lReiZycl+unL(t5BJ5>CW!t!sV
zCfN?^%czelUh4bn<zz<aS1X=Z?WWqx8kp%@Y@k`6AjjD;(iu2C-v^_51qT#<2n3E4
zQ~hqb+%UXeKh-!Gt!fzQ1;^ZTq3zW4&7q!NfK3f;46stKP+Py3CA?+$_31hPJ+hpx
zcKDdjUS!xn+rYxVm;?Xo+xqtp)fvF(ANJRI`}bUZhJ}Vnu92hS)F*yj4>ijNR3Dr3
z!?l=8l?QU5*vf%QA_3koL4LyKVUe#XOEDwX;eI!}%j{_zzla2O<S3Yt8zfq<ydJ=a
zFkyTaQUv%@9CxT4!{49I`w2@(sZ4>M&uNgbK2wt9FRorbFbkUK@iHbu_<ygW&O5?Q
z$-pWw(}c~w8LMT4RR!*@YbtrO`Fig-5lb!<hwCRT8`F=@WQZ`hz^51_xYz<`H~&8j
zv^I9slWsJ%6y*MEp8Hau@im40uh@@gIj%hS_{a1dT$^Yqe)hCLUTL+>k7e(GZDuUx
z?LvSJ$ivtMO_+|L{U8I?HyI^!21tK(Q7S^YQcMr#B(lC$ec!T1t-cP!NyEzkrSt}D
z$PUnPpyU5Sp$xY{Ui9j?atm#V^<kLjW#$wx=806ogZ06I$X)au7z?6j_*`?~5WGbK
zvIAbuTSwxbw}5JMa;yH}m(5_7LP2)X>pA6pHQvpDxoxn3+k@NnswwavsH8Jn*x%3d
z|1k7{#O4n!SY-*Ug8p%k4kXk9;13bbW=zlTr}_2!UEAQf>^_M3U1X+l2d%jjPbU*(
z=VCwJ_0Adg7MJ%%^*6sq9*xZa@3({M&G&%tqMxWR4Nl+=odJHPgbfENJC14R!#Rv1
z;Pe*aag=B1glg-YQIn%y0_alJJB8_90yD9rZ5*kLXYn9VH3?wf>-M`?W{UOZe?Opw
z;1MDDL9y>@U;OhQQuW`9!9Qm6pC5+b1B>4m>7k~-MGY@mFbRtYt(H3p$XqbXWll9d
zZ`Z8Q{Au&O*s7}I9o3R5c%N#W!$6x(c(w1(^M2MON*Q3s(=HLv-kW4HhA56(qy5`>
zpow6%SNp$PAI{wd1t&mpRWm7fd;iDk^xxmmpQED>$|;p6!pi&Cb9rL=2Lx1$ay0)I
z)g(0<4$VWWDYOSj6c?aG%FU0&tdn@Iht9rk+8HOr+vf1@b4eX195cMyC%yh<JNQY|
zW|;aI{&FFKA2P4n%xIg4k6YGZE5<sQ?s4Wz+iNI(eF`af-E_ImpX&XS|7wyW(KGe6
zzecy#7KSKquXiTQV$#(pP6(d@YJt6YQ?p?&XO=0-!}V@+lWNpHU6{_bTg!gY-PlBf
z%W&D>XU*Ywo0z&AI`AFp`<FwPOc(-c!Rm2R5{F{nTM9havlS`!-_B`n{qW-WR?c~8
zkq{o*&wh1gnFQC2*>wT(N^OesWbX6kw1ux{bhb`!ez?OE?_K$)mEL%I#i5feQbPH>
zv7E;5DlgVN-jA^{Kx3%rO2I{Z;`8=lu2Y1Id@|go9$-9=96%P1_l$1<090T9ljX5o
z35YVOCZnxipdap%aGrJCF??RcgO5mfmryIxbZANa(#?W}O!?Mq9@%9CX;a&rVKbAW
zXS}bz^F(VDvU;+r14NkyEpVeQlKxA7)FS+R&AP7X$NN3Da9-|%(3hjG_0iq9xGyN3
zvQ6x!&`P1u7|n%XCDp%&Vl#@u<eN#wdjH=K*&3a$4dp}lJqxlBHu(D*k1L8?|G?5w
zdd?qjKZpeU3Vnhg6-3>Y{o*Zo`nV=*3j9jl;ElDpP~q*>acdBHZ36%h`#bF44<m&E
zw?R~G@w&?8asTmiC+b;*y0pdvWg?B#^wTF;%2+XP%|3=}eN1F}lywi2jNbR74kLuF
zP}=(~`R77Y`A#VU^n3EpN?LA~oqzgT-(#-5Y_HtdSLhxu?DDI|{8!JJ1N%wYSgp(Q
z=0utKPM=3lPwLLq7I?Sxq5uV`0<;3FN|S>gb1`nPzS%wnJtSpBPY+W8!eI?KuvGv&
z$;qguHAi&IB2%DSh42!U{EOACgrYd46atJA&=mNW9^wg}I2FCuf7iyH%J~wr=|__b
zSSTFVb91r*?@0ub`s%C=056yv-ZoSI4hLBuizN$j*pI8}P*>s)6@VtPQ8kEdc-lnO
z4tAR)F#!2kaT8fr3+u8to0&z<l-PyPC&2t1Q4mxHJ1RB0y8_$Qi*3u_UCCRpSYe9f
z*F)k+4THo(QjdAkK>)Um_1HP_U`Lq{)=-R!cdaZ210oe;wy}>zK<z4P2LP&^uHnf1
zvR{gzYeT9q38z-$PI0L-uv=vNiEhHQS9yiju3dbY!nC#NkLkY)#NI{*HJ%1Jc0P<l
z%(7PeVStQPfK$a*O<@73FR@_lfYtH4R*&f@yi$55i_L9AcUCdX4Z(20HOhK+nnC)V
ziBPUAWEmT87k?QgU)$3FxIyh<*!5}H8bE;j+O)VRXG538f7ky>r)adQh&dt>Zy$BA
zSXnDeLs}VOX`C}l>(p(QKHWA1AgcNr)-~Nh4wYTGq+IN9rX2usURncAho|_!rqsRW
zdO>@=N4o~f_8#Qzt<6zgTjgs&(1@V{WP+83<z8D4hi#YI=oQwFjSXA%wrMptx`M^c
ztTT_gM|MmxksV`TL)XWTJjq=_-Mo;V;m;31X&)j{N7H-m_$M^y>)DwFRDg1#F?%O*
zMfIRgRZQ`+++qL!Y46Pgp?ur^;gX7ov1Ex1N>XFZnlZ8^*+Lky3rP~PjJ0Gd%VcSW
zvS%s8i0m2r$odW0HQBf9J7c`(^!?uV@4lbkeZTMhJpa7^JpI)&xvn{{^Ej^KJdV$z
zj7I@=mMKp%rKWx&D#$p(G9ZmL6LD>?{HXZ$A?TT*IF~`yA(%4}VM(&7f%4@}qMkau
z$qLR=1pBd@Na`To<6BcaTvyDYD2=q+6$2D&7Dd@habI%c=Wc|9xU3(LV$c9@B;6&=
zp*6{_YlEH&2N!sc(%!sI)fRQ6t2lfeC9}JmdbqpsM;Jnqm};Ddd^&?n=snkUHt*TV
z<#xE@_jAh*n%Wk*VTH>?a0;IT0w@oBS`AWSDS2lvy<#cu<eI*h|C%0};7>b(r5@T7
z?7sSld#*;y3F{=Y&$Bl?EpLI{G14Ef|DoZVN}dPU&3j7(P8svJodI6X-CL&2ubxt0
zz;V-F+PNE@W%K5*+cSV%Zwb5p+p?SOY;wTf+%Bw<3Mo520fbKu=7RsATV=L+R;iK}
zOfv^kMay_z<v`S9KL{QWDA$C-kC1oJhq4FXpa**7DDPQ2QD2jqX38qvD`Kwy2slk5
zsSVc^R?q8Cypx9xXwdNsHr}A<GeqZ*R94CWpvQi!>+X*=3`|v7Y^R){AVRkdkzN}+
z$Hy1gp=4>X1T=#88r`D9UBemmIN6SK&Z^<U_Q4XyBW(&%2tqUv{@wr*SwW+)HMz-G
zD$GG5-N%Lrh^?Zz{(!~)Pp=rJmw{pqi*j4~xeV$7KHhr1q@C@+!)=(PcZ&*-#P=!k
zuP!o-PJt;#0dShMx_tVIySKqi09L-gRfbPKnKl-&<9O2x&~xm^W-0slA?N9Dmo`}~
zXq-77=Hx>>_*GS(zPSF%r-xPjv>s3KC-_>+c@}=A?t^f3IDfHVRIe~hijV4^1~S+J
zM2LHVZHY|z;TV)WMwLmQ_66i>ZJI$y8HADlPV2L4zMxcaaD{oGJ>?n0TB1E^d{rIe
z4{UkX6GqrR9Gj$DZ9NILdmmcEfB)=S<7bm0x140p)_#@hh}#LC+K0io=&h|bJgF_Q
zALs!m_Q*=2U4!tTpu0wQ8+tISjuru(v}EUNLki0FL+t)k2VA>pifWB5gj-u*i6xKN
z>MT<|5vr+j&Oa`Oy!{;2H<5`CmWlRkQtzHVh_<DPWR(9U=6MqA!6Hy%Uc{_t6{abs
zrMfK8mAmlU&!iwk`$0fz$Jecc@@piU3EM;bqrmYne23|9wTMP<k<MZz$Zd%C?$S1m
z#~hkTd(Cwl2x;r$_ww*>dY+~I1oe1((p<lHf#VO~Tgt%y6Pg7SgwE?C1|Bz@UiXi8
z;AlOi6W<TYy4R`rd1x?%#tnLD{J3fm(6rKc;aTUI;^9^oLgM=!b<D_6Y7X_CIK|z#
zhod`|<SENf_k*{+f#V2$?^b6GeX;ocSZ|#%PCN~Xqsz)hpi2VCg5HB%gs>6T`E7L~
zH9N)=(r%gifNC>2b<qBWEyms3=w5Xijm-YEEk+WCb+@`M&$C_q#jLx+tfx-i_k+55
z_f5PwBSDrYjtkXQ7lR;k$ONoBhr-dC1+Qf}b*;E_dJUFTfc7ZKv|`KOKPG+Oyt(B)
zw77Q|0sG~(oaKP`6^$_2vY+#slb^R$E;j3Tr^M*SMaa1X9jf2a%lquiEnaFNt!D)`
z*UMir&KT18vQ1u4RW+o?8Cpj@dKP$~j$F+11m;PV{TT-wV!7iDJN$<feNn^efjidU
z3s1!uBLYC>@1RJUXz)FAENM`YT0-t);N}9EJ^IKIc#wvw(jDW#&zDj-Y71<P?&~<U
zQ|H&;>IxEDJ(@NI;4|*k)&kC~4z$!^koh}&)gS8Dj8=J^Rqo#oIaB0ER#)8n7IRph
z3?beTFE<OhUBJJ*x>|Ee?+ybGW;OR1gKXM0fV3qfBi;uhRoX_BCB+P@Gtm-(^9HG=
zA>j|c$yqCi7c($5XV|Ksf`9VmBvx`Ncg_P16bn|VDM&`~<zQjM@G7=*g}(HLJm$zy
zUhx?|i=jMR<?A>VbA~$8Eyg?-!*Il>KpMqdh^P^t#7DuZm(L0=H(FP~->bxHj~;;F
z_woR`V@4Eu)R~}`m&A29Yw@CtWJz-x%{-jM>vh(?`8Q$e`28&@zOx0&8wMc*MW?K1
zFIjm0WQXqOa$G1-PDtqabSD`JQ3`_wZhHk{Cm89T8~WXA547f#+z&a^^}*(4T_gBP
zdw0$RrKydnR3k%p^G@YOMMP__(X^!-BEyqd@6{F6Ec(qN+?B?CNVDBHhDFwsX5cwK
zU@|h|dH>^HM=ax#`$nSkcFX5CJ3~$OU)BAnTX)l3-ZAD^K~jw2?ytHar0<NUE4!{T
z_S38`zCRRU|7ns1XAC}4<SjbTEff5`XGAarg3=>u1>Qh-Z=H&$j!4nCb<bMQ>zjW%
zxBHd9!lZ-R=95;Ff9!T&0uI@9uW}yfW~!g*Bdj0D7@b1C_O4fU3kbD-80KI~wuv3c
zwPK3vx`Db#jAZmcDLzm>RcFa_NLy#Q{)bs0VG{d*qJue=<${=gGaz@Pb3h|M@^;gO
zuW}Ys<Z~X#_xV7727j5_(J-Gr=5Qc&gMbvUg|9!Roa=Lu0{8BeHO!-y3bW7UC8@s&
z4UcoZOum)cw*TvatIBp2-%cKfiurJ7!1iJI@bXq6%FmdqpEyqZ1%^p4(hfu)(o-VD
zhuMlP^dEvHE&OqP>zy-3?xyamvDoSh5o4x|p$?XHP@8;Qa4GK|uHCpvqMEL|QQZ?!
zOVIbCt6w`-2PN2CJRD%0p*Ia_%D<e?ZD{@au@ZMU9D;7r_n{*#C_-9zTa>NXjMnGF
z*cRgisUyNGA=*E%1$jUD?MJ6e!larn+i!ion9SY{+B-R$1CSK86?RN+E=5(!_8q;G
zAxbR#=&Ht}_iMBDpJ>SSmurJE`DC*Sq(k0W%jKV1(50&nHEiR?*v4_&Am{HQ-4j?J
z;%w=uUMupS&a;tAj-z(*-#c3nQX9H=9X>Gv5F)SSYQ260;gL&lIv>a;>(TB^5?b>r
z*#kpjh=?{sQrmew=r>&HFH2?|b<j!vCP3v0BnR|_cAVZxR1slQbz#gWc6V1)hVNfP
zn1xK?Pw|nfxq}oNy^9*%7X->hzjZ~jEQP6s#UvC;1+7eyZz(dfMG+9bmv*?+bpt6?
z4l{C*x)aFXl12`*WxhDpty3)@sV*kWA2mc)Sb8M9{J9K-S(kF#e6Kc`6VY#Gt~#PM
zP|g#&9k_MUs0?&nYZyLZA4CtNs~jbI;Op|}@<gjvq)Q^}*91WpHNrvD1BS%i>Sp~n
zsH7?LB2I$aPSn6Mz-m}9XjF9d{85Il=Qy&tS>x%y?rND6*tame+<9y1$9ikok@?$S
z3bRswrm*GJ!dB3R&(@ZGZ6E7`d|K<Ta=nh|-!PhY3*0~0JSX^WtI8ssxn_QB7I<kV
zS@#>SuxrMPR7%!YBb)C%Q-qRHau2v3tE~_SD42t`8^s0Gjk{=M{t^e<a|rvZE}J5!
z5e)h;lMkcJlT0jqUK@$6LY^vLTru5TxE3BNCV^S)uxUcpm)HAnVkU4IN7z+Q1wR|E
zHe(_+xdBH;sw*+!V#C3FQ<Lef@95grVs(@As*u^T{ps1WtP!l*bx@-NNEE-=Ib*>H
zMQADRzPM;umhk)$yRKixu)E?aTy3v2(VnZ%82zk%P3><;6;<0uLz(6;sE;;E2skG8
z^LiAms6)@RMNcd4i519w{E4o7X{bDOMtK@}qvWMsXC6373GVAHA$J$XD;aKXInJ!l
zUagw+&>s+SE)nKA9_W0>e*0|}eCKwD0;iLLU#5+Oa>dDSk7gsLzt?DjqnP-dRnoiD
zam~_gO-auMj1dLem7+}1rD&u)W@uuDKj)_l6SKl!h3780Rs9ZZqfFa$(mZzm3fqv;
zTZ0uZmlt0Cij~{#Q=h#8zp5MOwte$K^}}J<6+8L~m%}X=!Hcwfn{a~u8}`({_QA)W
z%RltX@f*h5Sih0G5<QE9%Q^%{{r79aPYuoIAC0VYUb(xfl@|w!-^Z3Lw^zI@u}}N4
z;7^oyD=&EB>_7l_0ExUD$2dGO;>JYMzDftFW7eeZ#Ugo^=^q>b9YLQE?}y#czG?^?
zg{N{Z?EL<DF_n^H?h)AY`_*Yc6#;|d<i=dE2i{)vnq~Y{ji$=(ADZ~qO{oJPV%=aw
z)m=Viu9?uwXc1VOW&pGfT71_;-%URmM!>z)7*%bNDp9$oV)6^nB+e8w&N04cC-L3P
zT)cgvhlXrF=Y)&ZAQ=N~hgcJNAhhf@fT9f=I?=`i|CGdmVZ*x7JjI2Fd&f-Wnq+*y
z-HW&%SJ#B1Bj23KU-aYgpVlAAR``(>Roe97XJJ(9uYh6<0T}tl?ALS?9xK`;I>E9n
zfMSQq5e45|41%Q&da!&!eQ1a}5bj_V6`$po<GOJSv?RTpZXXZ)h}?*$tlKj~ou@C<
ztGT#=eACAZ-nyeeuyRc`t?!6%iWc`}7CKXgLXJXLX#B9m^^~?r@gI6S&yRLVY<^|a
z)V8|h+K{{>TSj7H?{{>c@)0;~<~#Cv-cf*5F1yg5sD}Gi&CvhGe&k&WHnx=6IJrsE
z_4|4)8Mb@B`P-(9ot`LmZ#OIYc4;uDv3*j{XPw77W6g~mLvkz^HBIO5c8hnumZh<5
zeaBN8kR&$Inw5z+E8{8PzT%vPN=Eg)@Uh4qbm3~aFTZ>qsQxw<v96i5;w=hS-`-&y
zDhMIwFTV`&pUa!Kt1YrODKQUh@q5BV`o6%^eoH|JWj!3?(C^n<s94}%o0pl_d*Im9
z`x*bhB5c9+qobLEs0uFMve*eRKhycia{kF!$MSe<k&l8XVW`x6f!dW7Q@mBx;Wp_t
zDuvd)Y#V)sq2sZD%fyo6=E5sMv#)o*bmgvp%yAF*=a8@u04I7_`0tI&Mv>y9tH(f#
zKfQN<YL~k9J|}C#AWI~+#(qSg`Y!EFt0^XA*abpE+5dUx-Q=Z)3bnt;fjrKnppT?Z
zE~C19yT7SJlgE{6Hg_mhxZf`T2)VBi?GSG2#?9wu0=bBA%c014&H?8Xlb^b8Y-qcl
zv<l@!l2DU0N(QIs0_PaXKN)Y3e>Jx0of^}t1kF?Sow9>)I}?4H_--)I!!j28gN{ZF
zXgNi&^1m%LVB2k?&u6XAP$M^;&p(}M{AJ#7u6!c;ipBQe)^jMyYWhsu`D_TERx*P(
zq1I{-Xip|R0-ys{qdt9@ZGtiqs~n6$kXO+;IGgx-pZZ=`9WF8Z`rikFqEW!z6-;4!
z9xR0Ey2-b4>#>Ww%4>CY45j^j>dIpw>Oe;@XXFvMlCPQ#najehVCyBuT2iHNkgWAw
zC)Jay!MRi)iR0eY9!iImTfms)gpfJ5t7|Fs-8R>N`KkMUxIm2o@>`oD8hh{rSanVL
zFcsGs(2$!;BM|-_0PY)OlQ_&;tZm>|{Wc+Bh<h#z(HJYm($ML)0?s~uLja;=&>r38
z&+7Vz5d%1vJ^^!y9FAYf9M1IQOn@j554kTLl6GmkDoJISz6~XZ<|}A7qrLi(PS0t`
z1%IFqTYxVL$&J7$0c@Y6s!KBh0$5FDu-UD34jh?W<g5Egpdv(;YxDW(7X30D0%D1M
zrlx}C_c4C)JF|BSYWC(gHVXq4nz|)kl(Mw#WOJBlf!yx!((c=Ou^O%7BnIIFtc>O;
z4X%Ov1cm;`7U4VkG*1n}CkZK*43D8xWG!O%6H2|}o>H+r|5wWPr)eZ~bmputng7`I
zwx9fxmKUKRZmw%00q<&s(*;W%kSvIvRLk>VVrRei4ILoas&ZGB1QaQkvye4X_;qH{
z3CN_sKqVMCC@2Ot#@0AKKP{b*5CYkbO!45bOwZ?V8ruz<DHMc{(ar(Yx<@a+M(>Um
z_J)++<h-&sGK0WU5U!2BVzzMWFNmgl*>DkN78!l5YO~vH)!WUJ-K@|F_job1z?*gV
zJL_kyH5B~8G~gVHa3L8UdUfsoVB*xl>|~s7=Uk!sHbCo(g-p$XgOY#B-G+>N8Of_~
zGSHqEZ^~t!0}k7_tc$MUMG<ovy0=_FlW=Y4M)*;-Ch<;Nj9+Ag)7eAZ^;Db9h{h^J
zw9hK|ilSw-^Jlus`R=tiUxAtCAV2UnO*W|Mb?FO+2a6w{x75mQTV&pHwms6@rNXEn
z`MC(@3gmEKt!7lxF<BQBHlf9H5?agxqDz4nnNlv=J(g|$80y_&x`jSbg*$%OT<dIW
zb+}cEhfRE~kk?SNnC$CP^<7M<c7;UWR}R*RJP-CO@{_AZ-V`;`)&n~&Y3(BV$4i1X
z`W{2TS?GrA_pWngifvW=j?;Vb;<j4ii4tf7J$0K;cb2widpIYY-A4$szO#80c^OiQ
z1}fHbPk!i<ZL|53x!BKx{meL>2UU|)`1HL|ll40JCt7J-SucvV`+;)tJEC3Kwa4|e
zf{{!6)%(qvTW}-&`V!}=E#vB$hL2=FrMLEGg&&>!3*}hbe3I<&PKZ`i@gq>3OuZ@Y
zqwN&+pUyuJA=2IN%HrmPe%u}|(O9~&x$l41ch2|FMC3rkIc(<gkhbeYN*6~%&KD^{
z^d62Q7k<+>)yHsc$p{pleAMG1mV$pCXI|Bw5_lZ=X@z|}rRb`=<|@w5o_^#GU%QaS
zV5=eA({NSEYO>!;@We@@{1CkNQK<ZRjLBmlTkL9~j1`El+3j%IQv2?M=$pL60uyR(
zvvcBrYjm|)Y<+q;Whhtys<Jlvd-*vrVF)%dZm$HWAi+!u1*g6&i?_b@Ya|q2MkIW`
z5oV}V-^i=C;0gy{61l-o*3d*_>B44(!_42Zky}4whxD-~)9coZzQ{Asz~&Q#%E4-3
z(x&4~)+5G2hv<-RtJks8Lu{Yf$V2bjm-;`$@1JkK&NE|kq|B$o$s!IrclW`IWmi|N
z2_m5$>P**`zgp+>RnwJ`?wR8msEKI<gYy_;jhVrpE0Py+Wcwm4;<4q@8aoUAMw+tK
zw>h(dhT`$e{k=<T(ekS?vWm+D>f9oSW<nqLG-&(saAkA<^pEw5S|d~uZh1_=T1uwE
zycl2xF)q#O-iJR$gC$UE)cIFzza_Y0b~r}a)VAJ5+QusLIJ}^BHq?UT0aX>;1E7iX
zb@>zD(0kATuF5sBfSu!D8D4WgXDf+RV2CkW@KC;vMaT6ej2MN?=1Q53y0g(~187EI
z4TIxYp`7O(G`<F!QTPCy(mNd~gUe5w_5%w0gIxv64~-}=2t%5epVR_iM!IJw-V@sX
zED)w-rkMHDk`5ux(|ga<e4C|ORA@MIi91om_cr>r1yH{E_cr50L7MHRg5mdol>OI!
z1}4KAX-u&us<&*#e(HrZdz{O<-0#jn+#=iYvB9cFwd4oILme98F^^fJD{F7@O>RCt
z#?a#Ne2Yc7usvjgFX5cI69~#DF+yI=COdRR_D%bSqanCbvU`c{x$Iw@ZQ-<<6*|RK
zEi9=!jlXn>bqTuc{hQ`7<$$F%Fya6Q-RsUatJqNDZZh$!-SxNc;ErC-(Vi#ix^G$7
ze6keLT)UAH9;diuJ|dIj(R*xYSMpJKK$$M$lDpC3Qx@l^HJo|o+Fn3z*htrRs7K;T
zeQ2#`Ow*@nInvUQWk{`GFY#4M_e^7OdFd?s&gnWgrQpv{^L0IWkL*!||7rGL27<D9
z7kryL%fwnfT*co}1WKm#M^R_^S`)Si?n-YLKosg6Jh~cap3)^-#ux!&z~##WiIp<Y
zORlCZ&s<&;f&R7Em=kWMp=okdx!K5TPc3)NiED1P5!XcBouA)_ska_c=~>O#0b*$T
zbk#2+*a>^QrnI%!#R)c{_N%5PQ=|mF*C)d$W>C@5>#J|sbxsK9?DC3t5Z2PE?cb?x
z4a{8&Z)UCllGA;SS3+M`w#(Z~>e9<B)ba1ANIolnA0-z6T>b0{?!qRneLYR-4zcOo
zf!0|fzqf6^mVJ+p<0cHtYF~4BV=L~25Phzo^GltZ-lvB=EwB?YIJs*}1x>v;B@Sb_
z18f?tb=urmNij7zvl#69IYJ6)^;sn1L&U;0-oc#YxaRczb`exEYpEuebHogpcgL3Y
z*VK3=<|K4#_hzJQVC=;C3(6I!3Pg}%cL{JIonajD<32YtTa2=&jX3+A(w*+J0;m3r
z2Ma|7DSjoVoaDb#^ii+@mh&myQUMK`l;ddang9+^ko|}-n^4Cx5K;}746FwoE`%m*
z`PGL=z1QPbS>#RNU(yiCmzwdS!W^2nif`3cY3<4b7Y58VyFs4iJ*A4sldIA7<qTb?
zJP7Ur$`@0jhPaoHTQ^+{OOFmdlX|=kxfSJQ8f;9NsI2Sb>8KhRRfE8bqf0olqc_zU
zl68B@TJAlk+QM?dvbuB3lc-}lz2}?7>mN)8T|(8IUWnNpba~@rqO<fS)a?gk7&j2V
z&SErKeEqT!Z%fbzW|;};aM6VzJl{O^>Uqznc1TfC>_YYFJpYeDcpug9#-ZlR3)TY;
zTC}b!#VMQlaA_rfha@WXFY44BZ2arXdw5ju3m;O?FZMd3gW~?+NWuNP)>9(u$(~xG
z`CpBTsMHrO`~e;pADWSud1<_|=1P4cIehz_G*DnxWm=13A$=o?yUOJ|8yPnvZF)ka
zcJ;VUYkGGy*amtrk(e}y0c9=-1d#il3JAY;o7-P_lRy6)TVb_75X(#?;5gdeG;b~z
zR`$K9ff1%YxFDLF<f9H@{@TQAdfng4h{FEb1}nzn4qO<2!YT6nn0>`z{5n<zf45t|
z)I7DT74^`;?pql=G}m~ux?J=qEzFYY6dB_(m>}`HVe+b0m8$xWtG;nck|!R`KmRbL
zO5z|X<N7_59v4ljq;OiWw=rRqhNM<q`7m7_{obGx{e_pIp0&>M>Z22*oML(V#k!_U
zXU+tclK6IPUtPm2v=sYs-piNoX4X7DsdVXt7dTdNi|Qsp+j6IE1<3Z?ilCmpg5Z%n
z`H9xJyB@>-#Qw1Fpfy0D6ZH$b5HC!nQ7+D5LTW8t+{4|&h#Z>_cDjaa2CogZQyuP=
zWRW06d-piD6LdEjpX-u}(DzyDPF`)ruSU7RDpkLpA2w-w$8&Pu9Eo|B(<V~&P-)0!
zH^GdjNOFxT(0)kb?dt_rg8I>UNq*@yn*Hqa@J%_bt*)u(rZ7i+3X(Nq{#d_&0?m^h
zS=Dn5LpKx-1U~L;E;{YJ3KqMFsYFWBXpnkoVS(SGHR#wb40Y>wOSWNlUVg4aK9D4!
zi6#9?!8dt!th#@_3xDpWs?~OqpW0($T!g$Sd@R=Q$@~2x#m-IV8I_I1iSE88d?FF_
zGt6>$P6pwL<VB)IqbqY8vu>)h!s%aKv5fgwBl9dXqqc6zk2Ne*n-TruYZNgvoWn%%
z7f%BMf$>9<rgY1(V~v0k&@oV|&4FN$PIbB!v@_61xCGOwO<M6#-_Jo0UCPh=1K+P$
z+S_2=5lY6^q;sCcS3vbsoVP=L;J<9=-fy6h_Y<%oKc+nng#VlrK0mTe7d&fDezV+e
z-YM`)pNeeqm2$`)RQ{H^ZtK`e(ph<3=x3*o^Mhe*I{2BE7Ln+QV{^AW%oQ(Ty^Qi-
zUG4ieoe=r#W_E}FUQcRgQX)?BYm)@--Au2&Y|0l(ze;2U0icUlE+Tr{81<NzQ50Qx
z_TY2dwlsA&`enUCCs7ezmD1_~Eh=xt@`d36<~?E}DJG<dmZvjA$j(v^?tIQ=h)0m@
z{wdvX+xvCUWz(Ww=kInw-{JGrsD<&Dv*$7S9m|y^as#*7bIy>Y3JFy=4-O`QsfP0m
zt%-S)Ig<7rJ#GWBbTDIRJ<7C@;T-Md55qLaesc~hhT48Xma2re`#(G@llRl5j}o7_
zADV-ffIjo?m$-#aYD8_JMFEXZRkhdlpOLwZV|v=GL-do(SlEbLT`2e+oy-H2=<IkE
zgcsRPjYUuWv5YB+ckZ7lcH<c~S!zn^{~eW?YM?}d1yX`CkTX&Fub||os;*sEwnP+M
zO?#R4o@`@TkAs~PMIvEN40|l%{7{P<Ot4kHXNtkX6kQjd3qlun6F*0_yzt%~@CSS2
zUiu$bQ&uD43C2*I(hA|%A3kqn(Mq%@;$EZ@xo3{2uJ|RXQ{w<n!(ohN?x%|Uv3~kf
zS8sM)rx(R(tDBiITAvpU4bN%aeQ0ud5<t3I#;?}%dgsZnHOOU#;7>~rq}e!rTdR6#
z{eyL{cVeyz4GD6)+;8jKtEX=F2=@B+HWbSytDmxB73ws0MAQCJD$x7hRiUw|q5wc$
zV()f|ZZa2f7kedURe>3vy%NhNQ{JoiV^g%-*<2a$J)A1F@mj5>zd0}HT(uVz1l~zM
z$Oo8!Gn$ZIhwl_ai$Q)W&GT02u}}2aNJa4?cB<wRW|%LWL*G2_AHsYa@*m9~3q%IW
z0$1TY0PB)16CS$QC;jGt+Sw_*CX_L>#AWR(?}g`uV!XnQc;HcDOryHXt7cjd$vI51
z_On`yszHCU&X!p<DA?SRd?c4{kFrXPRI*1X+epR0d7HERKe4iTB*XdF_{iN3eG`0^
zn@+`DpJ2%A5BJbqnM=c+`&)^dr1nZ;P87xo%ur}UxHKPIqu!B3hNYbP=#8+gpW8gc
z{o>0ZH0!QZd1+vV86|f95Fo{Qgum<8Ggh4AhiW0x{9PHdqFlR#4$kfVJkJQtnT@0L
zgij}uxY-v&5cEt}4^B^6F6YpyRyC#i-ae?vPUN6&2_E~QMpWASO1HWi^J<?RN?%ki
z(`n5aif*P!aoiSQDD`_kh6)czMbB!=A$n^|Nk!b4Kl?UDj7NWbt|G4{-QnIh?<&L-
zvEHErI_Gz)0Pk&Gp-_b5?GxveCRG0e-?F49KiuT|HlIs$Ny53rOstb85+fYmNE3q3
zr#N_rj{7@yL;2697`}~mU~a^~HOIXPQ|94MtetCQFE)zl)Ls21`RI43b7<Op7<7&H
zq;7MBwZ4|E%uY{|X)Fz=hv=<x%SUl#=sZrYsb?AIFkMkMF#`+^RfW%Q)sqwU#t)&m
za0G{pKQIeSJ>)QW7o~k==iL(K>y5|`v6Ouf=)!2)wMWVoUglY8dF2r`&uHs5<d7e1
zyYHYBGnU6vR1zwizSf^o9O5|`%nW#GYp2H2T$1Bjd<wKWy~0{fCL!M>u2zlXDsT8(
z!E+<m^{0DL&0IMi{`+|a9vRCD56x=!rftotdv?z6EOhtN>j~bf+}o<kD&*8(>7ODS
ztQB$ZjmBa=E$;alb=u0eeMwl~izmF=d|5IJT9;G0py$6PdfwpDU-K{3BkA02xnej|
zazaP+nSyceuY~S9a-EpOcx}SdQ+*j6aO)%N&)BRvFAX3tpH(8CLry9-28O45?@o9t
zEbU2D!M5wg3oN@EtyJVr7IE@lhE2rFLaOp<^-dhS9;|YwAyOk)ESL`lZ6U~=5aT2X
zp?+L~EsGtaFoM6%un6`8Vi>XNzO>MIMr>(pN~y5*qw3XWSB=@lu3Fb&k73eHy73p;
zw$suh5I0wzwo=@PxF(jjz83wgUV}fxKFK_gy@;EicE0VKH7cgg^aO-6VchoUa|7)w
zsHtlz@#;_v0O+&@xm`Ci<W@%O^U#~IhJ}CJu{~O_^u8>fmcTi30g=zU!bh!0Z4-X$
z*4+=O!(OSL1TD*XI)1KJ9zOcTbBg&-x>Mszia&0%sBB3gGnK;|q-OE6UbF5yr?C+t
zm6t@5?{q73Aby-4gY3l+nZ4XKEybf6B#uA#T6sshq&Y9S`VB8u4j2imSVVDS)IhlB
z?P$@@qo8Zm`hl)grDPr9KONNAzVl9(OBj3adfZkqt2%~pA2fW7b-13$t0Ha%QN&Hm
z3g269^dkqdzQhL1i#{j7@}wDMR1vr|4O6D1IEy9Xh;R8xS5baTxGrzW+c0k)YxiOi
z-o}se2>b61lEe3D+X(!yBWqepm8fhh)Q2^#dY49;IBuvd#o*pgVb7BK!-lU}oaAjU
zuOY`ASaS+1Bf$bNkY=9n@?KCt>h)^;F3jh7pWb{<>x67wUNa7KIq1gL^iq!0!E#Gs
zDpj8vd6m(wO8~~zIv(;`<R=#v?@9)*XZ9l1(QbpQOB!v<t2(X449mfq2jb77W+2Vz
z`DllS!55LY3F0D+E-4<eIW+1hK|3u6Q<-3^cTU{l^l$o3y`2-Vgv;nIRAyTbdFRj?
ztDdl%8*=RA{D&|l_+B@bL0|hOb=o<DQMIz#QtPT~xOh64xGJF69|C}1M6()4TzX39
z-4*qk#H1$GYzgV-0}NO0AdS?iTU6Uiw^jFK_s;x4W9Rd5Pj6ao0z|mS<=(&Ivi0Qh
zD{u%=>4Z+Tsrvn3Vlr%CG9I#WulGgBPFoVm4~*)FFc{$;pYL`DA+sENtvNv(TVwzJ
zno{Mh#IDMCpSzYl?aP@!S9!o>NpXtw&+lZD4Iwn&f}w)w5;*&`VQa5l5az)=`4jSR
zls2E2YNQ6}_pPyF>~DOCenw#Bf5j@ga~=;ln+5q584xymDddvfM%AyzH0PIg8v5b2
z`t<WT9Fv;CQA)wA&u<C^%L<)%U=Z%6Pk(u`r@r3g6tLr&c>|=*Y+u%mljQT?WO2OL
z$P~grE@3rRUExh9BUqtT1~d=#+b?U2=Sf1*bwB+^IU=>h?<XnmysF<bd8stU8uYAW
zxn@5bA{9M;wv44sKJ!XJIe#h_RXuBc8LLr+J`HCN=zY9Zj>OIOLTV=6Zw_Bzf37cd
z;+z3JkM=3)b@6Uj04tK<v;8W2lWuDCiFJi=NE&~!BCy5O4`0$pdbaH{tlHXh8uhwW
zNOzr^_d=p>0S1_QjVG)Q{!;qH(i+0!LG16hHa(xssr121FRY&lh1@)USOBFPvw3;6
z=NGho;5QiZAi+7==BeWognS8fKG4rw)A|E{e>#LX<)blTJ?N&ilOEXE4X-#C*9l?H
zg5+jF3NtyxIyR#uu51?k(4`;khgj0XGb!5J6;buljn{WYRcnN(GlWh^xov$A+<>Z-
z$P{P}XIPX}K+b1zct4UaX+hTrpU`$k5%Rx|;A`DMJHB&DC-nv~G;cdmfPbELN-T=*
z+3k!`>Csn<ei%e?Vv0$?w&T0a1wE{d+z6#7I}h=fIhg+A(R;u6C*KH-RyIqDDjcb8
zse2lhx9-yNJ*2%h>-paAHhzxFkzoBntSM}dl?B4=tk4+Y2q>;Rn{S`n^YR0a;bDE6
zF?Q9qFR2S`?0pVv2Q?dB&dw#lcACBnPM_s0hR*pT{5`ew6LMkYI>^8l*I>1`q$E|;
znmUv_j(Z-<RRz_g$>F%vxy?AQglsoisg~JNWuy$eU)8=8tUY+{;F&^KDf`18<v^fV
zvT9gce{7qRgH0T%o?)C*ZpYS--wv?g@q$Lu+6o14zu4cZPVo<I@44+7@*3<EENfr#
zy8N*IYN+=`e8ft?XV5MCd?K*R_eb2%LFvA_b(${SDf2R^pf`dO2srUO!Hy&DB`uTC
zMayi+z@$Sa1+n@CP}p@(azDOa@El%syAOUo6q3$1u`>$Mp7ddkn`=I4I)1ZXR{3T;
zhe|H`xTI*^J~iAn)@=PYIMqAekGgk5w|R#FzEa!BaraC6MKu}KWZr=0G>(b1YFp;v
zOwDKowz%W&U+nx`lHQiWZJJ)OvPk#163)qp<Qg4<19~5w_R6F`mJGjWk%O~ORauL6
zZQ%>c1cIjU5G&{UL80x>8xxVXVttyHb#jqh5-Cjl>8x>{>+(|8*&fwzg9X+{bXodK
zqD$5bTD+wI`;ndV{DeY+_!KgWAFb9;DD48kxskHliYf`R-?J>pYuFs1>V5w!t~y6w
z-FEY>Km&7{E?)i);s<%FYzBdup&B}Cdwi#fy0@|1R4YTr>`HJSY9z`v%h8NT&UH$j
zZE({M02PG(h~~M4p?-Qp!JEdQ6Y#OnaoVf<s+z(t*mkNL*=^QK)3mj8VckRyy_gUC
zT|wlgux|_SKC{!}<-z?uIZ?I(N96U<dZJa#*_hu`<=?yG&q3+UBTnA1hJh)@wx=Hb
zQ)cy^MW>ALo%;JoceJj+poHgLWd4~Z>dy}qsusQ9LT5@$LLSgnB^^|nz5+x)(Tg^d
zXPEXrHDtJJ?X<P+*&J}wnIWYKK0nF(%c<BU`c<7XHhCXb8LiVlIB~DzF84~EHaPZ%
z|0m2SBs7=BUsyXvL6~%4$){^O<!I)S=5p^d?U2Bg%aQKt+^&ry!X)d(VkGHd_HzWy
z!W@1>6MrYfD%Zv%ys|-pcRT887rb~$p@j@w(`4=-E@^sOeZ{_hG;3)ho{9X!E-M-$
zKmA&=S={A$e;FRwXudYKo3$Bp0SO<qwdj>I=Tx$?JzAQ5^_vUESDa~!lHb2Kwxe!?
zhP^bkS-009YY1)3pHq0SUfYb|7`6nRx-atAsOfLHpI(C(c4%bnfj{N6dB!F_n=Y>u
zHMkEX+`pjSZ_3z&;_B}<Jvff47%Y{&t7{gT;&gY#3&6YV({CadorEK<WoTry0)1H7
z2=4mIJ8bFkQ`8OoZfs1i?E*~5UZ(xV-t``p&FMc2bHReCd9ML1sm&0-C94RGWN3ZP
zEk=>2mw-F+1}mOy|6&Ta^W8DA!|fZ`^vv5O!;Y%ZiSbiT=_u4N?xvi`O*Ty^;BvdO
zlS00Cy_^~ntmqE&1Nw{>5PBtEO$GyKw#kE<h;WqiDRO8lfr3UK@d}&zaXEi`O7Urk
z$DY}b@H{4;Pa7`m{hzxYHh=tCSoM>#oF*HKUJV&$HDQ*(-Jv6+1CRQVA!@xln>CWp
zFEeh=@Vhsg+BHAZ)taG;qBkj=N4Z*pRj#3CB?%KhiQD~cDQ{Y-G5wrU#Do=a^xBm=
z!agPhPW0Ou&An_bdGY%1PvXx{Kl?yiJuOxE)g}bl?E37FNUvrjMvRV41k{or#^uL?
zS=_vTD5kCi3&K9pOBr5hyKo!J*h-Uvn^mn*-29*`{1f>}g`@GdUy=iU4mY`T1;2LX
zI}jv5S~KM?CMT~$1{y!THH3W}isU~e16$HJJw<?fb}vG|sNH1?ofYaXN07+)4opgG
zVShY`c<Lm12uZOL_%KKGL<Dp^CM_|a9An~qb#iN%n>=|Q^A?OOP!^U)wXfDV2E^%*
zU+$c6ifAf)NNBoxC%GJ)bKM3BvF#uGWIi}i5Z#yavC-~D7o0*k@<dkn50a`toAom0
zTG!?yGgN%_C$*LBX&F3<>11@2J2yD;aCm)@4(z6gl}-6s0Ebk%ZX#RFoELbdMGK!C
z4;AK@&v)g2-+;zT8%Mtuwz;|o&97s_e{z;r#4r@vI4+wlfj(L*5M<Tum^&CCAo}2D
zd~bFuQuiw^V|qL1c-T)KbO8phWs_T3#Q6R7S^k}gH~#yh>1#(d_pY?^Voc82osP&W
z^7CQqgaSp+^M}(qLV0_onyDIoo9=_l`hqk>tld@JsqLX9uLW)Z7omjyzB%<fWCVOF
z-hgf6*d)3p=m`bdZ<{JK?_q=4*Ofc0S@P0U$i*%=<|Jc=O<}%>+--r8a%Z_fBR2mw
z+xI?FB7u99S}fve5Pr&#C;P^4gcKz>^O02aoeGnDj9*0${qEGdH2|q2?7^F>PmCX4
z)Gq($1bCFTUF#{D1QV|Sqp~|PqfZRj4#YGf8FML|f?SH_#e6x%shPyY_U%|8V1E4S
zoe?Q}+H;D$6HcPQeGH5814*j{hdLf|(`zvopPsk5C*ePjo@uN49%3Lc0=yOCh=b7W
za)Ken)Rb+CuW=kLu^Se+6ZQaqu+;|LrS!M3J!}tM0Js3g{PYlzWaO_kZ$cD*qGw%_
zCj2mh1&0dbOz8JG2y)x@5a6qooFx6nd3AQI^d155MO!~CqsdKN)!i)Bl>v4(V1Jls
z!5e;Q?9K3JVR+q|v32Oc$t_FLIJ<hl6i0oH>!a{%N+aPKCK+ewVv!S9s1)-%lK}|u
zJBS>83d?39RxcS)Mk*}6Oq^S8@uwcYPW_P5NQ&N$LHph7w7iS58l{PO@OAkF*TYu}
zuDj-?CFkp<N8t;q>!uR{Jpn+BY&w_b!5s8f<sd&iu~}hCNcRH(h!gHmTt_V)ymh;t
zxg8?vm+mul#WHc<OcnkcZjb-?$RMC3W^YsTfIp@3P&PZf2g34YR!y|^;MA1K-Wv)^
z-sIkIa)Rhgw7@|2JKYJhA>JU~*0RPrHXes0*BfkSm_t{;I*ybC;k$D{iVtf&c_6XK
z#dHBP`m*7cAeqEh$uqOJk=QtxqeI{1X7;1c9zR@G$j7MiopyY_XPLipDurV2?cfwM
zCHK^q#{c+YF6py^!N!+ESXcmiXp=Zo#)dO2S^F);`CUC=Lw>@{lC5zV^y>1BC)PgT
z!NAFn29OsJcGkE4{+6hHd9jv$2p|mnzXMv$54Ax0JAXqf{5|hor=m=TLaf5j1mF*<
zwh0o>wY$2&-!(<%^Q+@8fn9i>J9juw_yPqL^fE*q;NsAKXyrEyneD1vdj9H@EcI37
z;$^~M^xeMoHcPUb1vy5H!UZV)eu>(L=IM$GeF;c1BH?(moQF)un706U;IajZ;$ttq
ze;)qOon{UPYp&=a0bg8;h>STd0-)UPsrBC&x$TAI%FjVp`U1cB{^uo2h1z4Zv;&~F
zV$Up2ncgD<KFWCC23r@HpR_6Khuz20Ea6(h(qdnv)A%xBcy-RsK$2|EM6WXy!9mjP
zKuwG+mJtyV7Kx8<)x4QfI#e$ANg3XvbqC`?kz4ifg~p7CsMYJUGD3`9{!8wJq>4ZG
zK~iq%6B)aS;;mMBd>Ty|P7T%*hktC9GUZ*=v}w@4&xMy(>~4Pm)h0+R!LCLmgs;zT
zn`E#@Jn3a5r2zKg%+%o984Qu;734>JjWZyT<^Fu?owQydjk3{VE^3#2WLigjb+824
zz0Rwfk_u4Hu$GIOBUtIai%AzA6A^K;d%Z9COky$Fci$d7BGzPQ`hyTM0~wIQ$sw-K
zoxfj*%S>5)2-N!g()IwWs)xjdpT4R_-0z|kX(-<EY*D_{2{OaA*)(RqZ>6FU%+{%E
zywU0Nh8k!JL~55A8I~C$Ve&VKz<+tw{RJF&N<FQ^3scI1glFL{VU+L1DmTV?0ys14
z_~#FW{y<eF6IlvtTB{XmJU<5nOSG-k9?Y*}4Qg>Xti^U^e-`d3M)}B9X-B`#3P2+?
zK&EqrykP7WqL!jt<H`8QTK`fnVaGdl{Mgp4m%m`6_#6PW>+l7%!c?v&D*MGL-+Qh6
z<TcQ5Soc{`+mE+K4zxPbeCwFq0S`1~B2wU@0n(jox8JM2)iy&!Yoc`MK_2UPiki}X
zqwBD9Sr+7yaCn?wn5^?8dtbgj%rSB`@)!GfSP9VHiHt@M)9T$lrD-OFDLu!2@AhHB
z{lxmU!C2SO?_~mQB`(RlOxD68A+Mxj3KDqVw21^;0LY7v{-Z)Zv*d(6$UzoR7?Cp%
zJlGVX<`4>g|H7vl7;9$X%o6t*y9Z%th}`7b-kfgWHSvJ?c^C4u%f)%$(T-xB-omSn
zsg<7Kh<kCZd-^8_o6azU@Y*})(55O?d0Ho*7i)}TmO}tKW<cYk4?CAg0a5x{#0jKi
zJ}Zge!_QJlF(!1gOPSMr)2ESS@iz=dx+U-Q1Ck%>c!`;)@($KwEfzcJ=A1H2I-o$E
zur5emZKu9`4L<&MRK_S!NGO%-oR*si2t&$+ypj!3`|7In=@7m*irA&0vk}UVvh&=C
z@L)HRa`zns+6VJL=(hGZ5><P1blRWWDY!co6Bmd3dwV(pQm^Ta&krwm69I>WY?tBR
zT{`PtK`cUBYKgR9A0Yn9Uu9zVs_q1MOJ^M3ch^k5+nG3R0e3>hJK@e{SbU#wrxauX
zl!6Q|A+GihiRoYbHlLYFZN&?q&a#l*(h=MBm;0%hh?kX$Y>HpJI)2oY=|+?OS0}@j
zGMl=@$Fw8r)R14lfJ4(9AbYn6T_<Va^X|$~|MD`{F-g?_J<>>EJbFBEee6O0zqr=l
zPbwe?tTd=7X?_S0LyI>*A#cdst$#UnrTFz10bS|ZPrOgnnvgZJ5#*Z?3<o1(=jRh3
zMk5ljkw^pnC*Lz9n=CMpQq-i7GZ9bEd<CQ<A8Xz~VJE7ss<ujRt)$;|LH==2<@ZtD
z6a9;$Lm{O7Ns{|7_cla>gDn98g5sZ$`nVZ|WfUEWROCm?3#X90x+%F=kKb~y8=DO%
zB~}~5=O|GQ82|#3%xx2FgxE1ai!$Av^E?j05zUW=++!A?T*ayYJb;cfg)zV9>;LZ3
z|N4eP`O<#6a6~ur|2<p$>(Bj*zfHIT*5AF!Y4X2($A9tB|GFN}tik(^EA(B_KRzc)
z>)}7R`RQAr(QUSRw&LF`@&C)^tvJAA3LU!-`;RX4Pe15c1@r9{8RP03BLDq`;s^z`
zrnT(Zi2uPI|L>#yr^oPb7au%YC$$x0!|!4rq}cwPc>V`_>F+<g_SAV?=d}g;*zwQ&
zJ5^wc*+W5&C^0|0@ITlpua$pNm<mIH?RrvU$&`uYTs{^`A(?VqsJrn$2sES+b-*fn
zpg08{{QQD(9g&wKO7%YrmcM^Nuo$?6+LtUOib~uZAT^sMvH1Vvzy8Z~^?!e0+m;IL
z)BCQQQjgta=zkf4rx@&g$j$v9uay-5K4~y|*B(3Zd#z5#>Yx8V?UVm__492&XdAhX
zz4iESKYXMdNS_~*K1pQ#$5;DX+QPl0kdB)|T>5?|;{Py%{a#h8Dah@WGPl;&_`gUx
z|Myz|yW6R~d=Vt=Gwd}cf3ZFOc02#y&+yr63N@%z=!(JRe;Z2w?px#~fHY=kwqtMp
z-`;yh4B%C9B`06&!+$#k{_lhRkMh|6KG=U3E&sfH|G!1-3jU74nSl;d>$oG}mxh{-
KYJsv<(EkG>Upu4#

literal 0
HcmV?d00001

diff --git a/_static/basic.css b/_static/basic.css
new file mode 100644
index 000000000..7ebbd6d07
--- /dev/null
+++ b/_static/basic.css
@@ -0,0 +1,914 @@
+/*
+ * Sphinx stylesheet -- basic theme.
+ */
+
+/* -- main layout ----------------------------------------------------------- */
+
+div.clearer {
+    clear: both;
+}
+
+div.section::after {
+    display: block;
+    content: '';
+    clear: left;
+}
+
+/* -- relbar ---------------------------------------------------------------- */
+
+div.related {
+    width: 100%;
+    font-size: 90%;
+}
+
+div.related h3 {
+    display: none;
+}
+
+div.related ul {
+    margin: 0;
+    padding: 0 0 0 10px;
+    list-style: none;
+}
+
+div.related li {
+    display: inline;
+}
+
+div.related li.right {
+    float: right;
+    margin-right: 5px;
+}
+
+/* -- sidebar --------------------------------------------------------------- */
+
+div.sphinxsidebarwrapper {
+    padding: 10px 5px 0 10px;
+}
+
+div.sphinxsidebar {
+    float: left;
+    width: 230px;
+    margin-left: -100%;
+    font-size: 90%;
+    word-wrap: break-word;
+    overflow-wrap : break-word;
+}
+
+div.sphinxsidebar ul {
+    list-style: none;
+}
+
+div.sphinxsidebar ul ul,
+div.sphinxsidebar ul.want-points {
+    margin-left: 20px;
+    list-style: square;
+}
+
+div.sphinxsidebar ul ul {
+    margin-top: 0;
+    margin-bottom: 0;
+}
+
+div.sphinxsidebar form {
+    margin-top: 10px;
+}
+
+div.sphinxsidebar input {
+    border: 1px solid #98dbcc;
+    font-family: sans-serif;
+    font-size: 1em;
+}
+
+div.sphinxsidebar #searchbox form.search {
+    overflow: hidden;
+}
+
+div.sphinxsidebar #searchbox input[type="text"] {
+    float: left;
+    width: 80%;
+    padding: 0.25em;
+    box-sizing: border-box;
+}
+
+div.sphinxsidebar #searchbox input[type="submit"] {
+    float: left;
+    width: 20%;
+    border-left: none;
+    padding: 0.25em;
+    box-sizing: border-box;
+}
+
+
+img {
+    border: 0;
+    max-width: 100%;
+}
+
+/* -- search page ----------------------------------------------------------- */
+
+ul.search {
+    margin-top: 10px;
+}
+
+ul.search li {
+    padding: 5px 0;
+}
+
+ul.search li a {
+    font-weight: bold;
+}
+
+ul.search li p.context {
+    color: #888;
+    margin: 2px 0 0 30px;
+    text-align: left;
+}
+
+ul.keywordmatches li.goodmatch a {
+    font-weight: bold;
+}
+
+/* -- index page ------------------------------------------------------------ */
+
+table.contentstable {
+    width: 90%;
+    margin-left: auto;
+    margin-right: auto;
+}
+
+table.contentstable p.biglink {
+    line-height: 150%;
+}
+
+a.biglink {
+    font-size: 1.3em;
+}
+
+span.linkdescr {
+    font-style: italic;
+    padding-top: 5px;
+    font-size: 90%;
+}
+
+/* -- general index --------------------------------------------------------- */
+
+table.indextable {
+    width: 100%;
+}
+
+table.indextable td {
+    text-align: left;
+    vertical-align: top;
+}
+
+table.indextable ul {
+    margin-top: 0;
+    margin-bottom: 0;
+    list-style-type: none;
+}
+
+table.indextable > tbody > tr > td > ul {
+    padding-left: 0em;
+}
+
+table.indextable tr.pcap {
+    height: 10px;
+}
+
+table.indextable tr.cap {
+    margin-top: 10px;
+    background-color: #f2f2f2;
+}
+
+img.toggler {
+    margin-right: 3px;
+    margin-top: 3px;
+    cursor: pointer;
+}
+
+div.modindex-jumpbox {
+    border-top: 1px solid #ddd;
+    border-bottom: 1px solid #ddd;
+    margin: 1em 0 1em 0;
+    padding: 0.4em;
+}
+
+div.genindex-jumpbox {
+    border-top: 1px solid #ddd;
+    border-bottom: 1px solid #ddd;
+    margin: 1em 0 1em 0;
+    padding: 0.4em;
+}
+
+/* -- domain module index --------------------------------------------------- */
+
+table.modindextable td {
+    padding: 2px;
+    border-collapse: collapse;
+}
+
+/* -- general body styles --------------------------------------------------- */
+
+div.body {
+    min-width: 360px;
+    max-width: 800px;
+}
+
+div.body p, div.body dd, div.body li, div.body blockquote {
+    -moz-hyphens: auto;
+    -ms-hyphens: auto;
+    -webkit-hyphens: auto;
+    hyphens: auto;
+}
+
+a.headerlink {
+    visibility: hidden;
+}
+
+a:visited {
+    color: #551A8B;
+}
+
+h1:hover > a.headerlink,
+h2:hover > a.headerlink,
+h3:hover > a.headerlink,
+h4:hover > a.headerlink,
+h5:hover > a.headerlink,
+h6:hover > a.headerlink,
+dt:hover > a.headerlink,
+caption:hover > a.headerlink,
+p.caption:hover > a.headerlink,
+div.code-block-caption:hover > a.headerlink {
+    visibility: visible;
+}
+
+div.body p.caption {
+    text-align: inherit;
+}
+
+div.body td {
+    text-align: left;
+}
+
+.first {
+    margin-top: 0 !important;
+}
+
+p.rubric {
+    margin-top: 30px;
+    font-weight: bold;
+}
+
+img.align-left, figure.align-left, .figure.align-left, object.align-left {
+    clear: left;
+    float: left;
+    margin-right: 1em;
+}
+
+img.align-right, figure.align-right, .figure.align-right, object.align-right {
+    clear: right;
+    float: right;
+    margin-left: 1em;
+}
+
+img.align-center, figure.align-center, .figure.align-center, object.align-center {
+  display: block;
+  margin-left: auto;
+  margin-right: auto;
+}
+
+img.align-default, figure.align-default, .figure.align-default {
+  display: block;
+  margin-left: auto;
+  margin-right: auto;
+}
+
+.align-left {
+    text-align: left;
+}
+
+.align-center {
+    text-align: center;
+}
+
+.align-default {
+    text-align: center;
+}
+
+.align-right {
+    text-align: right;
+}
+
+/* -- sidebars -------------------------------------------------------------- */
+
+div.sidebar,
+aside.sidebar {
+    margin: 0 0 0.5em 1em;
+    border: 1px solid #ddb;
+    padding: 7px;
+    background-color: #ffe;
+    width: 40%;
+    float: right;
+    clear: right;
+    overflow-x: auto;
+}
+
+p.sidebar-title {
+    font-weight: bold;
+}
+
+nav.contents,
+aside.topic,
+div.admonition, div.topic, blockquote {
+    clear: left;
+}
+
+/* -- topics ---------------------------------------------------------------- */
+
+nav.contents,
+aside.topic,
+div.topic {
+    border: 1px solid #ccc;
+    padding: 7px;
+    margin: 10px 0 10px 0;
+}
+
+p.topic-title {
+    font-size: 1.1em;
+    font-weight: bold;
+    margin-top: 10px;
+}
+
+/* -- admonitions ----------------------------------------------------------- */
+
+div.admonition {
+    margin-top: 10px;
+    margin-bottom: 10px;
+    padding: 7px;
+}
+
+div.admonition dt {
+    font-weight: bold;
+}
+
+p.admonition-title {
+    margin: 0px 10px 5px 0px;
+    font-weight: bold;
+}
+
+div.body p.centered {
+    text-align: center;
+    margin-top: 25px;
+}
+
+/* -- content of sidebars/topics/admonitions -------------------------------- */
+
+div.sidebar > :last-child,
+aside.sidebar > :last-child,
+nav.contents > :last-child,
+aside.topic > :last-child,
+div.topic > :last-child,
+div.admonition > :last-child {
+    margin-bottom: 0;
+}
+
+div.sidebar::after,
+aside.sidebar::after,
+nav.contents::after,
+aside.topic::after,
+div.topic::after,
+div.admonition::after,
+blockquote::after {
+    display: block;
+    content: '';
+    clear: both;
+}
+
+/* -- tables ---------------------------------------------------------------- */
+
+table.docutils {
+    margin-top: 10px;
+    margin-bottom: 10px;
+    border: 0;
+    border-collapse: collapse;
+}
+
+table.align-center {
+    margin-left: auto;
+    margin-right: auto;
+}
+
+table.align-default {
+    margin-left: auto;
+    margin-right: auto;
+}
+
+table caption span.caption-number {
+    font-style: italic;
+}
+
+table caption span.caption-text {
+}
+
+table.docutils td, table.docutils th {
+    padding: 1px 8px 1px 5px;
+    border-top: 0;
+    border-left: 0;
+    border-right: 0;
+    border-bottom: 1px solid #aaa;
+}
+
+th {
+    text-align: left;
+    padding-right: 5px;
+}
+
+table.citation {
+    border-left: solid 1px gray;
+    margin-left: 1px;
+}
+
+table.citation td {
+    border-bottom: none;
+}
+
+th > :first-child,
+td > :first-child {
+    margin-top: 0px;
+}
+
+th > :last-child,
+td > :last-child {
+    margin-bottom: 0px;
+}
+
+/* -- figures --------------------------------------------------------------- */
+
+div.figure, figure {
+    margin: 0.5em;
+    padding: 0.5em;
+}
+
+div.figure p.caption, figcaption {
+    padding: 0.3em;
+}
+
+div.figure p.caption span.caption-number,
+figcaption span.caption-number {
+    font-style: italic;
+}
+
+div.figure p.caption span.caption-text,
+figcaption span.caption-text {
+}
+
+/* -- field list styles ----------------------------------------------------- */
+
+table.field-list td, table.field-list th {
+    border: 0 !important;
+}
+
+.field-list ul {
+    margin: 0;
+    padding-left: 1em;
+}
+
+.field-list p {
+    margin: 0;
+}
+
+.field-name {
+    -moz-hyphens: manual;
+    -ms-hyphens: manual;
+    -webkit-hyphens: manual;
+    hyphens: manual;
+}
+
+/* -- hlist styles ---------------------------------------------------------- */
+
+table.hlist {
+    margin: 1em 0;
+}
+
+table.hlist td {
+    vertical-align: top;
+}
+
+/* -- object description styles --------------------------------------------- */
+
+.sig {
+	font-family: 'Consolas', 'Menlo', 'DejaVu Sans Mono', 'Bitstream Vera Sans Mono', monospace;
+}
+
+.sig-name, code.descname {
+    background-color: transparent;
+    font-weight: bold;
+}
+
+.sig-name {
+	font-size: 1.1em;
+}
+
+code.descname {
+    font-size: 1.2em;
+}
+
+.sig-prename, code.descclassname {
+    background-color: transparent;
+}
+
+.optional {
+    font-size: 1.3em;
+}
+
+.sig-paren {
+    font-size: larger;
+}
+
+.sig-param.n {
+	font-style: italic;
+}
+
+/* C++ specific styling */
+
+.sig-inline.c-texpr,
+.sig-inline.cpp-texpr {
+	font-family: unset;
+}
+
+.sig.c   .k, .sig.c   .kt,
+.sig.cpp .k, .sig.cpp .kt {
+	color: #0033B3;
+}
+
+.sig.c   .m,
+.sig.cpp .m {
+	color: #1750EB;
+}
+
+.sig.c   .s, .sig.c   .sc,
+.sig.cpp .s, .sig.cpp .sc {
+	color: #067D17;
+}
+
+
+/* -- other body styles ----------------------------------------------------- */
+
+ol.arabic {
+    list-style: decimal;
+}
+
+ol.loweralpha {
+    list-style: lower-alpha;
+}
+
+ol.upperalpha {
+    list-style: upper-alpha;
+}
+
+ol.lowerroman {
+    list-style: lower-roman;
+}
+
+ol.upperroman {
+    list-style: upper-roman;
+}
+
+:not(li) > ol > li:first-child > :first-child,
+:not(li) > ul > li:first-child > :first-child {
+    margin-top: 0px;
+}
+
+:not(li) > ol > li:last-child > :last-child,
+:not(li) > ul > li:last-child > :last-child {
+    margin-bottom: 0px;
+}
+
+ol.simple ol p,
+ol.simple ul p,
+ul.simple ol p,
+ul.simple ul p {
+    margin-top: 0;
+}
+
+ol.simple > li:not(:first-child) > p,
+ul.simple > li:not(:first-child) > p {
+    margin-top: 0;
+}
+
+ol.simple p,
+ul.simple p {
+    margin-bottom: 0;
+}
+
+aside.footnote > span,
+div.citation > span {
+    float: left;
+}
+aside.footnote > span:last-of-type,
+div.citation > span:last-of-type {
+  padding-right: 0.5em;
+}
+aside.footnote > p {
+  margin-left: 2em;
+}
+div.citation > p {
+  margin-left: 4em;
+}
+aside.footnote > p:last-of-type,
+div.citation > p:last-of-type {
+    margin-bottom: 0em;
+}
+aside.footnote > p:last-of-type:after,
+div.citation > p:last-of-type:after {
+    content: "";
+    clear: both;
+}
+
+dl.field-list {
+    display: grid;
+    grid-template-columns: fit-content(30%) auto;
+}
+
+dl.field-list > dt {
+    font-weight: bold;
+    word-break: break-word;
+    padding-left: 0.5em;
+    padding-right: 5px;
+}
+
+dl.field-list > dd {
+    padding-left: 0.5em;
+    margin-top: 0em;
+    margin-left: 0em;
+    margin-bottom: 0em;
+}
+
+dl {
+    margin-bottom: 15px;
+}
+
+dd > :first-child {
+    margin-top: 0px;
+}
+
+dd ul, dd table {
+    margin-bottom: 10px;
+}
+
+dd {
+    margin-top: 3px;
+    margin-bottom: 10px;
+    margin-left: 30px;
+}
+
+.sig dd {
+    margin-top: 0px;
+    margin-bottom: 0px;
+}
+
+.sig dl {
+    margin-top: 0px;
+    margin-bottom: 0px;
+}
+
+dl > dd:last-child,
+dl > dd:last-child > :last-child {
+    margin-bottom: 0;
+}
+
+dt:target, span.highlighted {
+    background-color: #fbe54e;
+}
+
+rect.highlighted {
+    fill: #fbe54e;
+}
+
+dl.glossary dt {
+    font-weight: bold;
+    font-size: 1.1em;
+}
+
+.versionmodified {
+    font-style: italic;
+}
+
+.system-message {
+    background-color: #fda;
+    padding: 5px;
+    border: 3px solid red;
+}
+
+.footnote:target  {
+    background-color: #ffa;
+}
+
+.line-block {
+    display: block;
+    margin-top: 1em;
+    margin-bottom: 1em;
+}
+
+.line-block .line-block {
+    margin-top: 0;
+    margin-bottom: 0;
+    margin-left: 1.5em;
+}
+
+.guilabel, .menuselection {
+    font-family: sans-serif;
+}
+
+.accelerator {
+    text-decoration: underline;
+}
+
+.classifier {
+    font-style: oblique;
+}
+
+.classifier:before {
+    font-style: normal;
+    margin: 0 0.5em;
+    content: ":";
+    display: inline-block;
+}
+
+abbr, acronym {
+    border-bottom: dotted 1px;
+    cursor: help;
+}
+
+.translated {
+    background-color: rgba(207, 255, 207, 0.2)
+}
+
+.untranslated {
+    background-color: rgba(255, 207, 207, 0.2)
+}
+
+/* -- code displays --------------------------------------------------------- */
+
+pre {
+    overflow: auto;
+    overflow-y: hidden;  /* fixes display issues on Chrome browsers */
+}
+
+pre, div[class*="highlight-"] {
+    clear: both;
+}
+
+span.pre {
+    -moz-hyphens: none;
+    -ms-hyphens: none;
+    -webkit-hyphens: none;
+    hyphens: none;
+    white-space: nowrap;
+}
+
+div[class*="highlight-"] {
+    margin: 1em 0;
+}
+
+td.linenos pre {
+    border: 0;
+    background-color: transparent;
+    color: #aaa;
+}
+
+table.highlighttable {
+    display: block;
+}
+
+table.highlighttable tbody {
+    display: block;
+}
+
+table.highlighttable tr {
+    display: flex;
+}
+
+table.highlighttable td {
+    margin: 0;
+    padding: 0;
+}
+
+table.highlighttable td.linenos {
+    padding-right: 0.5em;
+}
+
+table.highlighttable td.code {
+    flex: 1;
+    overflow: hidden;
+}
+
+.highlight .hll {
+    display: block;
+}
+
+div.highlight pre,
+table.highlighttable pre {
+    margin: 0;
+}
+
+div.code-block-caption + div {
+    margin-top: 0;
+}
+
+div.code-block-caption {
+    margin-top: 1em;
+    padding: 2px 5px;
+    font-size: small;
+}
+
+div.code-block-caption code {
+    background-color: transparent;
+}
+
+table.highlighttable td.linenos,
+span.linenos,
+div.highlight span.gp {  /* gp: Generic.Prompt */
+  user-select: none;
+  -webkit-user-select: text; /* Safari fallback only */
+  -webkit-user-select: none; /* Chrome/Safari */
+  -moz-user-select: none; /* Firefox */
+  -ms-user-select: none; /* IE10+ */
+}
+
+div.code-block-caption span.caption-number {
+    padding: 0.1em 0.3em;
+    font-style: italic;
+}
+
+div.code-block-caption span.caption-text {
+}
+
+div.literal-block-wrapper {
+    margin: 1em 0;
+}
+
+code.xref, a code {
+    background-color: transparent;
+    font-weight: bold;
+}
+
+h1 code, h2 code, h3 code, h4 code, h5 code, h6 code {
+    background-color: transparent;
+}
+
+.viewcode-link {
+    float: right;
+}
+
+.viewcode-back {
+    float: right;
+    font-family: sans-serif;
+}
+
+div.viewcode-block:target {
+    margin: -1px -10px;
+    padding: 0 10px;
+}
+
+/* -- math display ---------------------------------------------------------- */
+
+img.math {
+    vertical-align: middle;
+}
+
+div.body div.math p {
+    text-align: center;
+}
+
+span.eqno {
+    float: right;
+}
+
+span.eqno a.headerlink {
+    position: absolute;
+    z-index: 1;
+}
+
+div.math:hover a.headerlink {
+    visibility: visible;
+}
+
+/* -- printout stylesheet --------------------------------------------------- */
+
+@media print {
+    div.document,
+    div.documentwrapper,
+    div.bodywrapper {
+        margin: 0 !important;
+        width: 100%;
+    }
+
+    div.sphinxsidebar,
+    div.related,
+    div.footer,
+    #top-link {
+        display: none;
+    }
+}
\ No newline at end of file
diff --git a/_static/benchmark-1.png b/_static/benchmark-1.png
new file mode 100644
index 0000000000000000000000000000000000000000..0c4ab31f1b11535e865fedfd474399d59a16d1c0
GIT binary patch
literal 62190
zcmeFZWl$Ya^Cyf2cXz$G>&4wJZoyqca0%{C@QZtZV8PwpAvnR^34U>Z^X$9-wU7IG
zr>d)ay4p_l%$%t+zcbNlDspJZ#K=%kP-q}|X$>eSm~|*9Xi`M@f12BJ3-Nyp+D$`F
z5~_NN^yJ@(mz6HaT3H#2>7R@U1qV$G_31w*|40;?<bTMr&<s$p|4k1A1r=!v1^3@=
zRQ~b*9EtzvKR*9Ihy4Nl->v`Ye!%>fHO%@C*#DBDN&n-P@2W5Gk0CkB>$yQe;nV(y
zpg|gdD<~*&D3G*-mKXGyKY}UgpFejSg`1V;ZELPhtw4@R!g-;mFH|Yw3`hWR7^>99
zd~s}W;wB{$pmD}`1m1pc(>L9IBCXQcv*}#ysLHxYze4e}f77nU!^7HG=*IeR{pF%#
z8b2uNJM|Al2AJ<b|C8PF!|+9Qjkx_m#e<d+$5MmC{U3}nEC~Ao6SuwK|8x<CQE2rg
z*PK-U-^|6a5UGBI3BiRs{_hzWU_=uB1<D)Ia?|=>qW>qCBqGBOx<moV|60&LXYpy8
zA7LFwoPS9FkD7up|7Dg^JC69D_5ZhAGdKSNU3e?;%Kg6~{NHq85O?SP<!x?ehLSvi
z9xlkSF;1Kk6ElC>M${pHWwIzBKMG@scKgy=R$U35W$*|spw+)IjF9Y9myFfE(9kic
z^@c|7`We`qG|j!4pVaR!yr}2Wu`$~-Eo4y6kgs~%`5wgcc$H%ZJ;6Xwvl57oY5L*f
z`=)*=)>&iFzJ+AA&)MD3W$aXOL{U*I*Sx1&WutR?aQ3x-kCu#`QYdfY0Guo2IX<r{
z#ySvzmb~WF^Oj);bE;tVgEhR@x*^l&X3eoBW?%=ilv^-8iT!+TPD8KDr-{*Sv(x(!
zzGSXH-ObQ22S7_3?!2#TY+}M*X`rB>@FlQ<gfbpl46(98wzDHWrbXCfS1y-2UUW}i
z*_8*wOTi<?^VF}AozlhPFq%f_a#S0^Qf&Jns>L+U%aaZK=;AiS>5Tn%o4@`pYIg1w
z`t~y6orO=e3Xtnlw5;)hBsichQB+bQ+w8DL=l^`mNpVWlYqRnG_EKC}C@C9DI9vCE
z$!gkO6L41n8DbAm_saK_e(c<RGd3__?(%(bOdIq$gb=0X<k&CZV92wI<>FDF1=%w6
zt+?G&X^{ce*sbmYp1eHcDp@NNCx%NWeoN@(R^{bUgdzr6kjPX`?WKFRXpi@cSV4-*
z+qmr1e6(h^J5n#r16*BQ0o>f&x101WD;=ISPr2FIb{D|zhMSuktKnE8F4JD90h~gO
zI`Z)aALEqtbdsVkHO5^4Uf$2UdwXd4#|7r`K76xgU*)>=l<~Gvc5*#8lSasWl{N!j
z96(A+JnX3!NzM%oUAJ?Jkaf3dqPa<uF>J7Iw-%@t<lazUZ#nZrg3B9XDe&=nLn|U8
z;ts2C+kP`i;&;B<1bNx|*wpO`fF$Q<oFOA4?_UOfbeyd=DR#e|J58i8YW@z`<Uqk^
zN!)t7@Z_?c13aCm27Z<IG5B_sc1iH?+%nj3Pj*eV&gK$29Mf=FW|8K+Wbb7{Ec%u{
z*!GKdCCObH$+K6J{z91|+INh5mwBBUI>CU{U+%j_uXJK}v)ZDCNVvE_b<cT&jr4kx
zX%G@v`115I{<8c`D0O-;8prD4VuNlHM^%_=$t19A<&gVgH-?>WZErX(nN~4PUSD6o
z!&|d0;2>3Je-KC2{=qO$yx#w6TjJ>u*kjv$G@VbOprfNhU(@~k>-D~+C-vWhS+%-#
z7rET-YQiKPcOvF=e8{#1lv7kbKPP!i6DlEh$O*PXe(;i*A-fOyO^VIl!|&9(vb7=h
zyQQ{WBH-gh!rS@})rxPKe#Cuzx*qJ0R;HZ`wocBZ0Fz=lV?z_Vz*ozfGp1kAHCLlK
zohfBl7sZO{#56n4cc)g{y}^){o;T$w!BdujPu+J_PGddqckD&4o8Cvct{-m?6QfBq
z;nSb;9^NhkO;iL=08+*YLorgYo7qm?j2OaKLEa^LG@e^eQ=a>FCNEj)94u0Mb-Eu@
z3*|>j@VSh4h<OtPu%N%|ZEnhudwf-FBZZ05fd|YcoIhXsQP^=eEf;NW4kw4dyRX`o
zX8!7Vg-k$BO+NIiJ1zt>MSPpR^}v9KFpm6(O<&cE);;thP_n6;R!~~Q*6Sf~c=v5a
z3L|`!k2B!yZs|Zy>^aFB@wMmU;X}WdaKNn5dY-mO-H$x|QPXWu)VVW?j9sYHJ^{`b
zbw-0!8>2$>+<)ZmFZ0yz_uXR=T~gbX^W4p@y<VPeJIsO#v_T^s+Gc6Y(f|wEL@+Ji
zZn4su>3mUORL4>r3uq!s(fop=j`o3TiK?1nH2?fqSdJgYI?16s%CU7PLQRHscj7IZ
z&w=Z^)3K=#i65WGEs=({HXVVXbMV+KQ?l421mJSiu%{aMoJ_U0uBy@1)+(!gCIE0;
zf1t-uOyicL8$0WS(_J#E|6-6NkalLD=)ElFDGgSl$e=9dWqY!>sq?FEN?f_kfy#eB
zxprN3)@_t{e>y(-PNFjXS>6p<E{IL^)m<i|nwf?%m$9O{kGaSd+iLZPF&K=>%!2J%
z)lA0x?R2rKU!8Tl+y5mNI!TtD_@LQo%dy-0EFti<sM4@)PB-%7DpG9p_GJFxUqiSa
z0CI)#ygMRtMdbNrjjK+}wGw5un(7}1?179A>uw3S1OybG)tps3uWc>P^G!QoMYbeZ
z1oTl7r%tMgbap4S7z=ALRt;{ODIRV6?q!CvSiimvzne)7Jw3+2w^Ne?->sKCZ_t(X
zKHtH|`{l=PPo0lMVRrq-(atgf!?%sg0N_5Kq#cnH;@1UlZ|?$Xes`IP)yM0-1Cp)B
zh&BOgUJbFzkc$n)!T494;XJE3x?3W~SFM$t-Oa0lL7AIqykAxFZ*;m$*I;Gh5ue0X
z4<(IKL+WxGYY9%MS%t9^-9om!gQ|g<)x!q>IljG3j|(_sRtJv46Gk4$vaa!YVJL@J
z&j8Wu7kZTM!424ToYg!=)zRatLcTiKb!6T&)yu^S^{dC3{QEZI5@Tr&b)$#N=-o_#
zRK&6COZ@~Jw>B&K7(N~z(cH%TH=6Q*`xTq;BBKiJDz0hDi6{vTajVx6vG+Qnsl%H=
zAN(2G{qKw`kcB3-T9V>ToJ32TdX`N+D#mNRHdDi_br*=~X|11Uwd#ehe~rDsvRhd8
z{ugomob_*CO)`5r6!P?qF_~0>@n~>)COUM^{3+~(J^X^Ux{uT28X(o?i355|4?zbx
z12H-v=&^ic^kO&^ul_|%nci+?794Wq0`NVaDTs*G{`@)9H{LRp=AJZlway|^OGk%9
zhifyv6N8(JYrhYHf@c~*KV~fa3*M32u5Oa5*qau)_dZt9@y8y;2jVqyd}83gVL}!(
zRf&wR?`i@!7_deZx;pd|iYy9GzB+Uy#pllZTzS%D)1;+z&%QnP$?+($prASd+LLC!
zIw3U~wolHqv<pc>eydE8+&8Z|;yj0`A9G16Cl8F$($XHn(_fcfJ>-e_DiH_VL?qlw
zIW2b>6E_E}h8$!^IOt}O%E><5^pg!XxMjk9_sD4Xd{!lnazf^ezNp$;eGGn78yOX~
zD`G_VDwl>04q9OG(>Tj}UgTB{k-236J5-qasn4pujM3-wbjbHJmgf2Vx-LKbf?H)!
zB$vSV;tRp3+WvPJakJU8M0W4{T_hUbwAEQn;8QVfmyzZZr2H6((hTqjz{xxe(C>be
zm)kKBF3AR2gQdFNtSNGE>`fX%H(vW7@xL`ov+?5U_O0%qGkdxu`Ea$rwI3AAzw9(~
zt@zZo&)ar@dxM{ZXZ>;D>|~ZEF~nYH)>T_JeJWfUwxe*v_fr-xKB3@`E|hD=P4gWo
zt3AGNf6?^tXckCN5WJkXkQlWg4|p^yP)}lWxmCs>n4&KzNK$~!r}@{4fz#!cbDCKl
zL4(q=9Mvi|_TNlv##EiHMs2-D$^?28&vpht<$ccg!?2LP(CnVhWtS7rjl9~Tn@_e~
zfT|nYw&&xbg#E3L_m&yClvU@}EBGT0-pIg5>XypCivrWaH?gz*w$Xv7RV~Dx7G5<l
z+ER}c6WwU6=mSDvE&y9<a;m-$W@F<t6DOo2*Nqc2-*iECo%pyW=(rNkQv9}6DJS&J
zmNz-5^?P%S&07ccl}kd(<E^0!phWt@@pWT`t3z`-?bpowQ<IkLWA{`Ffj3)VB+|vh
zstEH?&of%Bcq^ol5zBT>xP=>m<|giSLzsPY#C_{WpmnSHunne(+C&;_^39Y`jdvmW
zvrG&h>+vo9K|{!UVI}Msf9egz`&FbQg=_qROH_vF8qRa^XzlO&w=tTEj7(g1V4g5c
zNA_Ny$>f7{iTw-k&*GwSkkMENv+ZiN*{Ph6fi}NNuJVL_SZT}GKXam|x{6z50u(p8
z2W@{44w~+Uc2RF`oEUpX{gqw@>m|KC6_H-wg{)JKk@0s9Agps5AfvIRpM4YoEr8CN
zr}$Fp38Y!TY}!>`5w_#xc|EQ(dFi*;Snesz_^^4MEckj030)JW3!W(#R~s)V_{9WS
z8U2G3<bgH^zj)G1Y)#|<U%0H?G=fwd))SJUbmT~TdA}8+O0)UVm?;8^X`2yKpKdSt
zt4ua3Dyo$D6V@sou7uI8^FOKHGU#Y^dvpDR9y8B)7YQI3Rj|s=6;6{`3BAC@R!`4L
zkISAcw^_12E$>%m2sDOkHZg-(4&&r=T^CCX@LdhnvrYvcKL<+T`ZHdP-B8S~-rKLi
zfd>Jok4bT<tKm8UoU7r1AS&K(Hm#pf#^XECyBlkH;5(gDm3yy5<x99-8Dig?5ocv%
zeokAx=E8c=Lqb(^xjDT=-|W?OBHSvrxJ-!C-c-+=N4776I^ajt0!Seao)7~Lj6O@5
zYarYPT8=W7{_-MJbXGYE>rXG#wY{MQgakDM(!cY@L|04qDVI1)uRY%bF9nl!G=o3;
z*XG$1tU8Gf2WHc1sIjMk41<Fw(Mk6-UC3)?#DtZiKKZ9`B59&_dnwkDxs3G*dMvUN
zDWK9jBfmNiMLMNLd!S4J-;cfs)sX%D-TVVwz_dbI#>^=}yr(&RTb55dTuf6hS|Xe>
z#OFIJ7gdGxH1IZHcoW5!?G4L9a6}yZf_;kL>*UL2tgF_!OhEi&f1wA0Q#WrE_WtG^
zjQoQ|Q*QophV7ona8FzcZSG!jV|G}6N8nJumh}tsUnf}d!)KgidD}tupi%lx4EyNJ
zzJM@az_wJEx)@1~vgFlf6IfHA#Qe9Ju71iuc#)Qlv3Ied?|D-s<K)iKw{O|gwVW2i
z88$@uoPKTdRpF94X*&avBrzm`H{Rk9Cu*rldul}CMfEWikI|}m?#Yuga<(v23wJ~U
zzi+d?L(z)K(gbz<zAZGFGY>88ee}G}-^?~`M}>L2@BD1rY>d7pmJlaUd-Q0z*YW%`
z^XK<j!s+h_!I8HUiAvcE$AV3zp-}EE;ue^qigsR#-p|O1`rF^e%b(ez?olZ9mD9Va
z{~UN^O-mhuI@e4$nDsuxf-{DS4$7Dv-{M_{jtc6K7i8*#(1d?;6o2{__iebOoV(^T
z7pIIdzHndtD+Ch3^fcdZE}(USjZy8EI)P9;8u98-Wo*EF?q<hyaG8&%sD~E|a`<^r
zrXa#mJ|JLEP=oo42J}UWt6EQP%*WFAi9}DKDcir#=^3NFx{toIYUgA*P?umnkbem2
zWam7Lr>{Nifr6<O)mWTOlIts$qbLf0LWsiUq_e<57ZD@rlh1k6((fojG+V!fNamBI
zqZ{Cpw6E~bV)ykC8%f`%`ERL)F>}~EC5jcr{49p8GmTYJv(rlPBZv7J)azA7Fdj_U
zR?nXE!TH+8U^x%|Zo}%fLt<{J4AJo&K+4l~_yCepj_%{@%`^!Css~TIm$Q7j=pX-=
zd*0lH3!*dK^KXNTe~G&D;lwmFi#;Y*#&W)RPq4NSDlaQ=dz|TVmnWV7G~shHZcdUC
z$k|as<>&cLl77BVuc0FL=5F!la1d2iW(Q7XmlN0Wr6k*ts~TXpSc#Pfw3#B>WoJ%Y
zRefhfv)jTKl8^1Ntl{y4uwOg)nVe<94hrn}Nv{%;utccZAR>WVoEuv4MX1n7j!39a
zM#&k*+`6e<t_tX&>jp8zY%mPj4JS-eprX2hQl3vM1~bzbqm$WF9vRd7pZqBwIaA7b
zYZPH=l!9g;!k6uyL*Fo^KS^^qLWXwjZ<$*l=pTG_>8H#;&0Zv3SbeeB5dMJO7mapF
zMS$+a(=JuMiiY_jtSXq^^Q7C%wd-rX80mvVI#CUQYdh#9_f_eu3GdYJA{9gM^?xT!
z*e@{Np<L60uL5xf1+HN<w@M_jHiD5S=RQCKS-{{vsyccAsQ=84*6i@Jx#|R3?pMCV
zQiAh`N*|)-0Kj{@f>ifYp-P5pH{f@>_SuT}<iwXesT-5llDu8AUA-r;3qc`R8tf2B
zeMHJ>;1_xSW6}a}-IzU;H>>)W;7xIVrrI<G!_bCaQhZpsoe&3$yV&fK7|GsG=QS1K
zYB6Ltq+oRrB`_+0i(iZGCE{|%kwJO7urr*pTa8QH@hL<p!@mjeyH(%b-i-!ilkah^
z&z6_)5q}yWex~N*cbAHt4pjf^S(El#D38F`hr-3lh_cD)L)N|*BYL(JKdP#_-(G0s
zH`Wo-QMO=POF&3cr02Flk9fV~Q}S{|PVD@^GtJ-KmD}Lfbg=}VLxfYbgW<dkKs3EE
z&Gmc@E{Uug%QX$D^^Z;{WX4LMVqWrsW=-0l_rrIR@$yp|2BSY#vPAO4zTW{RFE7?`
z`8b7-Ip$^aot?tWiZ_$#%#oQruLtbFwJlAtva<P&*atyY>qb|DH1{d7orc6&>|;%z
zlj3yZ`VAbb<eak_`^1lGrqohx{IPde2AR=}L03Y$TK~OUXxD?mO>o-G(-QFlUm|%D
zkZ)-;VRMD?6vg??#RPRqCDCb`1OQYzWiH2h@=?23G*5qZ7^CubnUpAZ+arP+mf`Zg
zYcy;-YJ}zzCW#=9`RsuOhdr*4ZgO~8nI^`*N<VSKtjjCK)X;a)BV3VhC?=v4VNLCd
zeetRNI8oP$!F4Q9CuMZv^t;vljOA0Gw>P8$1Nq=&=mkKtLA8}7Bp$%;<Q3CIt-fh;
zkc8o>HMT`?i7@7vh%iNDF3S+`=2FnvUX3VQ0#B5P$G#=xTf6RK9w?n*OTAI)W}8dl
zm6m;*y}jEfnCy3CMLS;A@G_B_>}n|vkE+pIio7h=)+rtE;L?4@_6bSKc)E%$l{cKc
z)56|$uhHHuKryu{%<+Ygdi&MeEO<X%f=F(aKAi$Q4GEo&SX2GWN|3OJ_*D&B*HDc)
zft?kc(2}mugkq-cLM8y?Mn2Vclk(2xV45{3MRc6<15Hb1bIq7l%-<$~Wdkh~r&S{G
zcEV}u1jFREdyby9ThGL_XG8sO>PR}?n4%qqX^3;EdScmMBcY;%b^pKOMA_N))DN|6
zVvl_ov62&Gqo{W_V;z^>cB4F@_`fr`M+lLsEEa*V%pC6Y(ah0oAOSSHDkMBBS<fPw
zt_Htw*>Sfifsx0=mNc=42DU|Ml@i->Qdhe6GKisT-ky=*ho#`A!MfxdNb=*M=OYs-
z5Reo@z|J>Nkb^l`tAkoKP)62B=kjIvFX35=dmUw4@#pA)GE|EJg2}xEF-PlFF0$#;
zgUDnOW54U6fe01-GL7sGc?gkbavdKl6?7hlSqKu|sB<t&{wiB!J_BHhBJi~?lYcxr
zs{eT&Bb%7(p{`4dYkACI4qUG8w#Xrt#AUb0!tKK4*Hcrd2A0cbLf5_js?!Za1Blv7
zEC=&{sfWWh3s*)Lg_E*XqgcA19yI8b+l3_uk#ZXl-#cJYjDmd+CMZqb203p5Zi4h?
zi);Z3>y%fZ<mu#}r-AJET8X7iHHsUNfyt|?`HA8%vV<z|C%ws{bz5KRl-uOjDUYEs
zeSGt==;E8=@CVNd3YZ!jcUC&I*JJXeE3Z<AA&!@!NN#aIrv!Vf?@o7(o*RFvw5jb{
z8``oAU*yIVzrzkXR5K#o5uLuKb%PM+ZHyWJ)N1&8u;R(V^StiRQ*j}WD@ySY=zq5b
z@r5Ax*>-zLyq`I*fPQDQ<WUbIv2WOMxZIP_-q+n^VM#u`D{I^=veQxt7|m6`hcCax
z;}aO_7^0A;KNqtrLKP6<O!t_az5TFnlERh`5a|$2ZubmezW>Psdg)JacCbOL@qR1)
z07u+iDnf<L&Ds!tozP)rak24JCKjKu%kMCG9qIto5u~uis$G4N+;mT(()<aziyUmZ
zaHx$Y1~~`!Mn;o7E_FV0gNS)lxE$Zom*-v-HtGom->&yGTDG^G^;Q8$>F?6^Ii}wY
zH>`Y%?~5rERx^@mciP4JV^hs><5z*?*d%}9@22)0Bk%Z6*PX@+JOi20O>Iv_aId$I
z`pwZ#eeX?=F>5nK{4sDF&pB$q{6B^pQzB7V{#ez#8J?YPrZgtkB(-K!3<$U!bLWHp
zxNRk`dcH%^9>Id%AYx`!!{5SEy6?H8N8}?2tOS?yVfYw*iL~z>@)hQg+?+*jhBRA|
zXt-krpJp!s@0CuG3j2dr0s7dYB!#Sg5*q3WxJ=t_M^oq5V_5*_j)`^GmzxeNBeU|?
z?K26;ohDr<BK&?`apb;ljZ?Bv{uW)nJ$qV@u?tM)gz|SaOEQD!jqQf9S0hJn&l`MQ
zun0IA%Sn<5KLt4IX|hI}jiEK_#70IyZ4R)F;n;!X=`>Gl|Mv1eZFb?Jrk63+HDJT&
z&SwvNa_^~JGoeZ_5s{UBxacHN_99a}^@F6#$UxRqAqObQK2+1TTnseon=AUXTY+wo
zuc&aBdk&HNE$MNI<*?kvWy2^i`hY-LGKJIiVGToFJo-SRuwLlX7>+q2`leb}f)px!
zocGLkQe(>Iqsz*WU3JQA3~kWoT)ANkrPPOJt+C<$(B0t|n~P)8Jqsf>ZRQ03W)le4
z-u`QyzhoP(w)a6F#yxa`WZToAC}jNKoJ=w+oChBraMiWpdO5LXHFkmhc#AQzKftp<
zMVXB7b2N?{VKl}r#%gjCxs{T3>_b_y(^fcU_|QD8_A}Dxtwg!&KrA$*Kg)cfLZbv#
zJ}qY0Vh-T=M_nz8q16&{iB}AnLGZP-jF-y$fDqygJ4w|U1Y8|DdRtr7v8?M+Cm3=`
zvgOaTXr^6A$If&&7N~fTGbM#Uh)>YCzd3{&p^Hm&K7^lZXdQrkRVb+6=CW7o{7mB3
z(MacT>>;;z7S88$``)@I6n;+WuWqvvD^c$^$5us_i>A`DFz`7=f53+Bvoigv2K}DE
zqk!2Uc^&i!bcYz{`Hh9|Uf6jxL;n2_GLFeHGn26gO|(4tiN05z@=_(K@!F1ynKXx_
z%3_<v$zRv6H`fPFRp&qxY)x|Q+8nrO{uqooLE`n3TaD)2-L&-R`u!^66bkw&Cz@76
z_a&*3Qli?3V$ZtuZr>v98}dj4s3fet6GWfO#*(G8s+nW;d&jWSYSAlk!huM5uuOP>
z;YPvhk{>cT4kbh1@ttHT3)BRnZZUuef=?r}DB1~iHVUR!me+m$y?JwrIW(%cYhi9f
zq7Yonk`e?Sq`Zf1bSzzfW~nit94JOLrX$7_eY75a-L$zolGsBLHj6#o=THk(!^_?q
zw!RZ#f0N;`tmIchbXuCfc=oIzr9~E68Wk&`{CX7}K}2?4G4kC4B4eKKUa;&M%CF}w
z7cdfvTGH2l%)iHXO<Qar%<4{@w`acA)Ay*o2CvLznT>!GF^xa6pa?KUG(Eo62IjnR
zD2ZJtT?b!%GWGVfq=MVYm>&|o7Oum{g|4DH8$t^?G<uB%pcAs&0^gOv)qZgKk!DB~
zPPpYzuSwzBAF|i#62TP;!bU%*+DI(3X;7Rqvw}KHjZtIWhMH`;yb1tO6>|!Ue_MnC
zza`q!0~E$-#W3YIqoJU`MYtxqPl;fQBX&VmGkWk%(4g!3V%ZiKTPQh^IKu?dP|=n<
zDEtk#Y9}6-)cDu|`)#(&dOq(_K8xbwt^ss^1S@ac7w;nU75kzMwO9r7*PiEmqpb$b
zyp2R-B|P3-e0rH4B<h3NEm&9{p2kr{e&~~dwpw11qt9daSQtr4WE%sDu({~G!Tc0i
z$-#LlwI1l8Typ}=McHC|E&`Kyhr_8<7^23OUXdLXBJctmc>Eqo=0ChDA5wb~W1YHK
z9oxlre3sa~)v<T}f}b`XXs^Nfpw%i>YSewgV~?m^*dgk)Vwmbrs+BBBLme>(zZmr>
z*ZPtYQuwpidAQ}s+bwa|3G5;GEDM49IT+82E@nX12(i(AvV+>)`*eS?T-NMvpyEX?
z5LMd&!4M7x^4_fP9r#k9RwATe2s!-}4o><4+ebl_{8FsI^w;$h;%>{4q=f9&s#otP
zg2=Z59;20@K10z@bsuk~zE;CO08f2-p;jEDe7+anMe_#)1aXN40=E36+5y+I#^oWY
zT@<667}LZIH<vj80@dCd^m^(Z?4nl0I<IL<@8ex{*OPrw*Wpq3eH@SEOzzOYDWY5|
z#fvOT3=>oo{!Pwo^o8+KAqDZg2Frt%lF@8n0yvVLnCXoRj9FZyopz`bt}!r!vV<8g
zNeYEDo!Fsr%HGC!9r0=m<TdE_)y}&;#nVzDH}L6gElYII6XcO$BhP1r_1Ou463R1|
zcdzzCT6{t3ZQ(r$x8tz!`ehi^=x@epvq~g2C*JFJ(0<}?sYD^FDxG#wmHJW15N7us
zLHkIeAY+AopKf%Tos{V)ej+JELrfM6*=l9;K0jF7KgzUtg012qZTybA!E)`vX8fKa
z5AcxBs|8=k%9rS(zmSe^DxH)?GcA|@eA4j=On8)T0<)J%xu$Z~W9=HT9)=b3F5lwr
zDT4opF$pY~@!=~Tfl6fVS7^HvPfC{XJI|51G3Tc93d3k*^(FDN(Xuv%Wq|r}aIl{i
zLaKiNYGq7vo1&f;ArBZ4-YF4-h<!Y%&X=q3?}QB)G9IH#dLnrs%Klr*=Muol*S7YE
zayD@=_QKZ3_CRJm4i69CsJCyJcS}{$pFhb5+3_}Ms~B76C`Tau6auyC=4Nc2t-G0d
z`ygY<zeCb{2v2c&b?tU({JRC*r3$uE{)`RWoI?1V7??Rn9Vf6cZE8fVKlMD0rH851
z1@|Ul`kvZtdl+ssQm8B2Q*!d@>V&_~#$l&}(FhJn4cX6vd|@N+Sg6poMM4Y~*-vAe
zdvcvITzUmb88@F@TOSsssFqo}L;hd~y&!=^hLEfVpzVBq$_R5M9QIWC6b4l7gul5r
zfWl6rSUJCdqNJjR#EDaII;BZ|>CsORPpL&wQ0FLh11KvCSVciFP5)-Xb=O##3;==u
zPiQIx|KZ$1@3{WJC$o)e6Zn>B*H}}f4!E?(fSpJQwfVv947Py8(X?W;^~rD{;m_y4
zmIl=|>5%rOIF@)j-@zhq;`0RSY}J0PB{QX@yX6rzqPU-@zY`3aeB)TAs1;+hDSP~p
z0rvxr52r%^)e-uk)&5o{1&i@^u`=-T!TIA$q8CF<5}4nz-D_K4D)1^x^&MqBJBnob
z&bgoC`RiO65!iLIh>~CVZAHw3Z7Qh7_S*=OxaHHxi<3UxgLdmd3uVd*ql@dx;16@l
zrkMA+qXdO?thN&kjlHD6fvz8w-Ftv=qiS|TmhM0gv9`CL(AXI8e@Ldr9&%RY$a~F&
z+~@hB@*{zKX<5{ukd_c<_SMf!4Up{sD2g{JhIEQyn<-@%d1Bl(<bUv;>KRrv)?3<&
zi*x@=mTwauiz(kx9w7VB(OhuSyoSRMAn*j;ws+*nx&r1n!OmZ^XdFi@TclR@nQolO
zk_n(V*T_N1cy5E+!PBYiz+}Obzg;oPhsto23>SnDj`{t~1|4jmLW;->WT2_vi^FS&
z-cM49rmeXDGj>Y3?qL91*~~Zx<SjeXA5k4sOmCy{EihQy53M?glE)rQ9r5>*1zHbX
zE^gv$8}DqL(HD4<qlc9q+@shjI8XlCUof+!+);iYydhi?f)49Rw=AX|4VlENch>o|
zE|j%NCS1ykq`gEVCg_PNsf?ad2`DuUk|xgpig$;+r{of|Tm1=mvRC~iOA#`?ud_%2
zW|e7jJuFiKsk%oztAeWJ;3aL3!-XG(jhnW@EkrC<ib;^8c?4AVNoy8Y@vGiQi(9Jk
z@PbF5zG%1c@aWHx<-0N`6ul`I|82!HMpePK2Yq?6I{KQxKc_G$WBCQEB=Fl$=hje1
z0I8&voE3``SPNShJ#vQY_xUIlmVRy|I@#CyJ@0edl%xaMck4*Y4DqSwQaj{R#ZSQ%
zg7ua6VY$y|4byyN+0t*233jKnF#ehEs`U=a)%G5eje%F?L(yO8^g)_t0f|x+qm=Yh
zVTy(yb>>2MvPfD3PTn)pt@@NKM3t?^sC~XzXhRXF?6+f8Y!Z29X}4EOf^FwN6;Tb%
z2Io7c1(G@3quV<;_dkn$xLl{6Cx_XZaJ<i0*Wv2khim^P$Ae0PKE+FkcfRhITc$u@
zR&{F~^fg6!UAxBCdoJ_>`lE(^#aU?_cvywjNrjTdJ|hfmYLdb<bRJ6ri`0%j4yM6c
z!6ft0&_dBNwDPO&Csn`QUl^dQpBAG(4i|T>vpAKaERksg;=`8J=?6HjfTDJol_Sn*
zk7u-o$)RAc;i^_Tse-a%(B+NU!QmGfdH^&H-yJLF#aHPr7q#HNo6&eI{B8ysTC@JA
z48#udpF({MP}go3h(bdgpK3d|Z>Z0pWG!dtZe+!Tr>|~Ar`LF$1HEkQnihH^-6#={
zZaZ@Zva*Abf*VU5nSZ6YVs0!g5{#$Y9Fln$+X?MnwT{Yjh)1fNgbPoT=CKX<mQrsq
z4Z=ifpxrBM?H_z%!~B~6>TAmwg?oq}78N|H6rvdYKvV4{9MvMlw0)YBz0mX$=vj_p
zb}?Rn2^fET_llR6|G4EBaFIve6hmM6v~{HFkusKUgUVwwG=Q0couDI}#Y>1k&j>5H
zv?V1&<zMrYe+-%kji8yG>~ct655bO~byI4`T<VP4ia+Z4=ld80^>3#~U{mF)4lO}E
zivBLm^m+M~9g0!?JJg|4uNC<FN|5nnem08@zB4IO{%P?at&;JBv)Rlto0*JRV-ajp
z(ZOUfzs_xJ1}O@A)?`)AB)2uH^pr)PVs?9JyL(`e{LB7jjKaZPToaWJZJK5%C_P7d
zg`hIGqEg?Cd#ysscSNAj0n4C0an5|5*prfoj-GN1jUWaPc^R`#_h7D<l*8~UJkdVs
zzq7V;<h=G>V9G%y!*X+^fpjS>`vT9fxsg^wx$w7FszaM>?aj<qR-L7QT}!c_mkY>I
z02&<sau)Pjy9W-NH_3utUE%yP)-1@u%ddiHYGeM~+1crMhtuxOQHZEhc=o-vVUFS#
zZvn*IHg`wg%I-+cCh&%Ue1hLkwbl~7VWf{D2MY*OYs^>aJsI}eumcp6{X%UUDGaZA
zc5DL98;DL=;WiDCzd%+(-GJ4I0fixk5{jW$WSj4nz@n5a%ZK>tE4|DN*D~ofUyl~P
z*2<#DB7xaj4R(j~!;!o=`oa{W0R2p4C;6lS<Sm>oSc)dpwWx(7jpE`os+Si|_s}Y>
zO#`mEuGjv%ggl?Lx1;2AQ0`jLgg`J8Cc{eQJ}yoM)Q*pm^hz1@(G5L_6akO71L4JP
zbC7{e=pOg{KhsKiW&}U$f(F&;6*df(6Fx}MZ1t8kQsv>+Yk3^Sq|?JGtK(;y90j>~
zIaRFplN#|d<T1YGO&J>qGwBD3p_tLBkb+VBjnL}P#Vv8gUh_qqMamLIDLXtTi72=D
z9XP53HI4dfME#?Z^)L98k@Qyz9coA|-_&HPZvUpQ+xHuJfkjV1Tmj-~m6IcIU)k3t
z$G5v+>$ik*r*o@gkHXVuTFDk=FQOlb6r2&Y*T?_*RE}nEe$BnvhUX?7X&Io{LertF
zm?}{6@hYmVN4F%n{B%oppPxooAAgYzO`R74Hp>W7?LokQycqx49j!ad^JU$cMwR{0
zR_#7b2ZZ)|O3`ll(@>;|ecdqY+0Ltc`-(0^!SfU=vzwNSEC79sUv8eLMaXcKdSuZX
zu=eo}D5Ts-ZD+C%jBbz_JrNKuvQ+H1NV~khR%u@8j%v|XR2KDkm~B@kz&Py>UrOm7
z(=l|B=?GhE&j4%za#)d~d07Vm>wqHs9Xh0vOP+yq(q4tM*OG&byMlvk7iU}6L(Adj
zs6a)l=EfSDRV&4l)`an};*Dnac$roMK@X*iy}N$PL3rsn=Ng~Sq_wg1`98V)m}Euo
z7D}sYJqJ<RLU;j#Q$KNC*k2PKcT6s7YS6GRCn*%?7?+SXED7WGj*&`fcbcr9`Hi9O
z4+lDP%LzincE81^^RH93&(jNOmfX1lcVHbTu?UBqRKczhQEwmQxERzuUYysdSoZ5t
zetW}VQ~UEZ17WM#&dT+ZimhFptAdU#qkO7`B*_wjuT9axh|&h-+?l3ghOSzI@lZ7)
zQsa8NrI+Gx(N<7Noh)Y<??9^4%A1atxF7*4swE6#J!x2InnNMj7-CtOGAKgo1Q<}J
zL4!M3^C9%whxO~QP;%Glm#*cNmq`O$@w)D+KLQBwhdC)h_HzFJF3UL{^{Kqb<Z+Y^
zw4q-*KX(@ih98YVeaW4aYl}~PS{ok|0XSzuaa)gl@5w>vY50@FK<!+|nl$w~k=@@(
zj_uX@H#36sKEz~eS&6m>yza+pIEAfMo``W7`*iN-4$((?-el}sa;M#ASbp->ZTmKg
zxsjp^?yXL$mFV(u@fEyHbZ59MBlY$-!uqoU&`CRRZO0S;aF5K~JBiwt1jb$x_$<1n
zuO*lNR(kPIm-Y2#m#3dwtlAFIWC=)Be%m8We_1FO^7h$L(73s_*p<0W{%K19nj>nl
z*tK+$`u01!Rw&bxmQ`MBcXS>QaHSO%@iHp?`I`Dycu$GpUjkuxlN<Nwe0QuU$(hyJ
zRO<Frc>naJKQW1+Qw8+jMeIh+RyZ$s(piDml>>$<=t6ZakwjRW`$_DKr=RCmSIl12
z6Q;hOBv*>D{E2^0#z=4uMOA*=h@yr`$9i_-?lgsYJ;JGe;+6dHXF*fU<D{Y|5-CRS
zUdWL6?i^x*G^}-LAKpJzdN7apmw;h~wOPXJqH#fX-rjWR3pW#)Y*1}q+$v(#Mh7O5
zg}LLPnwTlAMDfh8SaDJjCcZ@Kb~HeyH&Rq(D3`*HQ-`Hzc=T%041XPSr&>g`MeGI7
zpjJG<bLXh3d6KKrL#Ly&zNhdY35##7abp3?%}mgZy;<9f;6d}D_BMe_8pv4fxZ-O2
z_cxcC4dJc}0<ZBlTiTj2*P-JE7-5G?BfkBhL|hXoSuM8~8iTkXTU}M*s~=P~4NGwy
zTiJ|xYjqAc6H1akx@T8G{pa)htmW-obj@YSAtLQ*E++6uqZKC|d;8yC75byaM;@Gh
zC+5;&MNPTGnvmox!P#06JG3i7`v5qf^wQN3$MLKBKdfO*haZ}kOL^RE$RtX)<rvu7
zkxF~`?B_zGSDygqEcJ=-6sC`1ND6nLKuW(n-P!U#;1}I4l2FxwwfVu)>}4}gPyFHR
z?9K7LBr6!p;Mm%&G6sb+i3C%W&9?Uoh=0dksyRVB*VZ3`=P+OhL@l63mW82C=8u@h
zCP2)yInUk=rD#LgLC=My&&*htE6RhGzDA@Jwit+~#Rgum3=Ul+v>3W+&#8H<y`CDU
z!VLmi9>{0E^7=aqn5AHe`7Od)`ATCl9&liT;7vJyjOMgL#k$BKb^B#eShpYgKGxq3
z$mNOd7}oKie~+QT3S&0cpc{aK*M7rbqXEXS(kcpnvcq(NtEOpXLn=}Q=Kodw2)~M(
zCfW*o=r08XtnzNqt&{JesDuwYvK)h8OM%Tb<z%7_qnH7($zDo7v`+Y8avC#+@Qf+{
zL}^2I4(kUHfny*t9dijAP#<8NGec<^M{>f4^*xa%;VY^#LBh_*Z_7dV$bj~9Ogc?R
zrZmOj0T`P)3tXks<JEpa?@u_aJp!c|__?+!7lIl<NEa^4@UC>!0JYMp;b%28Ac9zw
zsT<uWD93QSgkaqyqF?O!3-aY3T#nB|e{)SV*h#j5lhKJu$OwihI|?DTC`2w%LmRIG
z3%Sn?uW10*O*8yvvk%>d7JZM@1v1%IJ&G_SAUN`Ix2r8g9T<*s%3&-tL6(+Fn1dQE
z%nil{kT2bZI;6Tj4HA<oIv$Qpn%^+{LGnMp?9bJ$T+v2i6=X|F+PNSzn}dK2XWFi^
zPDQgRVcF^nvT?+JSR(j~aueM(`FrB=pZAQiw%SH~?N-aggtCe^^m5}%lK}=<IlPSj
zzUSpBe%P}~D7gCXgcC%dbBhgE2W$r;DE;t^nBAHCYbcqzX6RK|VVPac26#%VCEvMg
zzJFK%`Y#h*hxMy$7D*gROvTo)vzA$pl7&+ihT?Tk>e3BpQ}U$ppJ9;e$0@}c7-kH{
zWB9jC!i^2zz~;OnQ0R$zDI@Pjw6?RPe?!!})Wb*{`~2zTX2bF6H>QAoKBD&VT9fW;
z_?73LOB7B4l;(7|A}Jy}LGPf5>=efxhuqpb&acvP3l6b{dg{tCGLOA4m64w{=B|k7
zkYjsFuu=M6rTjY52eRVc|9C0GM6nz7JTjU>=P2olG9T=0p}G-%L2WoPe){!@ZvQD>
zIoa-YbYjZu53A|dY809Y%JEoR&1Hexf3FYcG%0MAXp9!BI0$o3Op3)jxh64}l;b?=
zx<MGu*EIH~=aGTorKsy3{K43}0<}aji2f01hFh{Ap}UFQ9A9QXVU1c2I%@c^RIHr+
zXyS+_TBw+|Z|HfH+I-4yguZhnhLq6FNXX)__}x4E8}j8FT`(QBz3})}uCaMI-mS1b
zBrg&OQ?H%5;NU2<bkz&`XiDYO+S~4?VpKSsX)wo*BzwjEqJ@BS(vHBjt1G!-xFD5;
z`-)I@m~b@8Ig8|ISD$X>LH~6h*8C<%em)<U=B@;jC+V+r6F)V|iCRKsvcJ7N0Nqaq
zQ73V<!=7YJuq*Mvmo~w@?(gt^>%*=qS{kPY1KYkXoBSy#TU$m^QtwFAMbZjb=EhD^
zEGJ1_$@{lE%8quGC<HoGYZky#vmXX?q<Pv6SzH`aD&+B_LeN5mjSJ=)mP2Mohv`0m
zo<``H-#^E3!v-6|#_{AB7;>V<F?2rmzsB}8!zDD`rk}8Eotdr%>BW|i`&+GUqh<<Y
z!edE0>V-cpk)Zc~raIfS30R9E`B^(hGdA-_g>(*ORuVm5cSWOZedNm_Ye9#zPTX1F
zx6|Z{l;lV&v+L|_!2G=IEtu+9(~1|``!Q4`Ls#!BouY<|<<eo;x>$e)C7R=4`Jro`
z2xY9KtV-W(Aohl{ohydr)(OjTb${Z7JKl~DZ4xxcfAIa9CQJzFMjm?UZCa(w_(O1<
zQmrQ`(?hmp2jK?JM0I-+B#s=_l(7KGeCOfSW<)k2XsHj*N(!1Vza114L?iU!BESE@
zUKI=HBVR%>L^+NiKoBrp&1N4%WxP>a3z^o@BGk%Fl7SvoR}x!kcT2s1OB_g~m|<hU
zMmyN|eQh#PLZr<rf`(IOSA*N^GfOHI-TIT=L>NlQBP2cJ@9cTMfj5J^dH*OnMphJ2
zM9PlsUxf035dK4NrGEJObJi6aau#xIajy<2zxLBQs5rX`digISVG5Ph+WJAC^wTmv
zvlCejt`KHY2gy@Ivvlunw8;7atfIDO^SmUe{SE0zf2#fNXVgYks$~?QL|c<+pMa>3
z$Z`b<@*m6nma)??D~O9?FfWdOjSm%ThYjNhlCg(5c&Xbp<)n`3jN9_9byh66mN`^h
zs2u&m$Pt&hJkv9|_J`%0s?*dD{>Xv2pAQ^TvPLg03U{X%3Ih<EIw-W=Od1&Z$+d(i
z|0C*X62UHSKX>NP+RKH4`QQB#!{+fr)`k?X=u55w)lQt+-&s=re)_;f&`xQT>zt`%
z*{%D5L6CZVSc6T0e~A>t@^X^UbK-w(o+?uvgfxkGE~_i%CttQVoI}MYjnsDqUbvL-
zH+jwTTIz`(OX<M;?vlFD3BidLt0iB2@$MDi7gm(1yl6NYH0f?R$@+mk)t8LP3Rhuf
zT|jPD4TY5X_WKe_lBS;hiGygV<F%((z>C70a4E+@asx)Ua@d|*%m&O~2QK3%R9T@?
z+V+Qb*&dJwrYMTN4rj8RbarI)hxGfoeO5P~g3`W;!8JGz=q(bQ`57da=h^vAms2g<
zVD8)@kYwE<|1-j7Xza5l0-<>Wbiuu`Zo(&BVh|9dKoxZzxu!756k#IfPq{5!sDfCb
zW*C#CgeipV{S`XuI=i;S*ZFOlD4Fpneu{Ub_r=Qo(koL4S(@<S5%TKuv>V7bVKGsw
zi$tyNSKlEU@VY5ld8JEjk~NEATMMSv{{D)AZ<P6Z%FFczWUloSz%To*Myzy*6)3w8
z=!3~(W<(@w!DE&a?f8YunbT&9)($I58<kN{{IeO^P4<v_aD5$18O!iELL=;$5I3T=
zqW^LVxoRFz0_Y&mx*Z7XzU@0^A@tq_3=AKUgKpa_>H*N=tdliKOTa#>AUX%cQ@>jH
zZ<{9J4@rt_g!we{2z)XoC#j?eiXxS_tvVj5i(}SdQwCAOn=-MtU=K+VYwAT$L4S!v
z4__BnZRo1hPaPEpp}7Z9vzwGhu~YCTQ<%T=Q(9b%bO?66r%=$S%1uy<ZB_L|p7q$M
z)7;q#QRsEKU_h*^V*yI#l8s5}O9WDVu47g74<oz6$}ZE>(z=7!&cq3!bIsQI2Hp3&
z=uaE{Z-@G9<29)rCL5s{5E)k{-1NRKAM3`53X|lQ#AA=a4Zh{?$L|NfmU6a+@$AL~
zSc(|hrWDz!9-+KQrca_&z1;0qx!Z&SP){-|<0lQ}xxBBN?N`jxNa>OeWDe$vZVgI4
zWi1-%*PP%>f+ySrW|MN01C}vFZWA+ICS+WWJ<O2C`27=)ze|EAGusRqj-~_O8}y2!
zC)jVKl;f;83H=NPs4EF>EJyb0CzWiF#<-32N|duBj*!yoReHg#SW+>IZMOr~IT!SK
zUeuPfrLq$T8#Iv)l|1f@X1)T~$!vV~fpQ^B1}AH;!@q)a<c|mX`9x7)Ug{*N_85dP
z^E0$1rl)9k*P0~l^8^#;?OUN^(X0{dng|LO{B*q&cHT&%ILQ!TCJySln(%a^orJV8
z8Tb}ez_JhUiNuM_a-=K@lycCRK=xg(HLP)V$7Oexq&%X^k?y!%1gwWXn7^=kEoPG0
z(SEb~G~zyfweIb}K)8u31)TGNS%hovy6t$qI81h<%XpdHKFD8PIr53{BjO36J*^Jy
z9(X&ya==u(Y>uv30}XZ=Yj_jsHzBPcpAgG|uS{PWV8d2Wmhi-?=D%dK3j7)DZ64SN
zSu_i@!ZOs|3~{p!K4CVLABo7=kNr@BVMkOwMhFISR&ol8KVUCuR4zmb!q-Y#?%FOX
zu16)$mfVwHA%<p(h`(y(VfrMU-&PLPp9$fc2d5ESV@Klv@o)?uTDxf^c5QmAipJz-
zr`yemd;^Ch7=$K<5Q@i|AXbSVWYo?r9G3VQH!Wg;#r)d@6bSWX@?#vx2ECs@NgBqV
zUL&o0<lz9FF@A`Ejy6I+#nC~%{0Lv`aLV=K`#pl3MN&62E$^l*A9l)7_gVZl`Dh(5
z1&tH8JAArZ6Xe?98R|8G-b>&c$z#e@?KT)4#T6=8@O#xEgC`C{PUfsuD_fr4p5S6c
znt@4_Z7Qto(*+uZ*=%9J@ak0Y(T1JzCl>^^A{FpY?}r~hDTz=`!>H84Y3q0Prc>!)
ztC6LfGb0B8OI~(GO2FL&6iGo1>^fZ>f-^JHyF7eR-+-I!tqjb6Es#6EhRk%D4_48Y
zzrE4Fn1MXZ`YEy;ZYFGLw$b4N(%_abWBA<KZb7=y&Uf#Ed+1k;@(Y-{fo-wAqS~Ra
zFZI|ghM_VpCDvn0))*Jg-}EAZz*MS~9|Yb@3i#%9-0)ndcZ1)lpQa(Ou0tuLPw$<5
zeVD#+0ydOP3V|%o{TcJAy5u<(65S2uTJMlCQ5NO^wei}~-%Yx;2h0udS-4M>MhWc9
zDXC9{(ZR(plz!QA_wiDM?IkaT#R~HYt?Y${gChb~CwWY*wF}U|^w@}LCq~?`bYRNF
zp5p!ZNffL9M`kE3H&T&pZ!`Y)fxmCOkkw(SdW6B$?zzm`6!b9#u89n}b}#8P<&PWc
z{SP)-oWlW)EvVvJ3PZ6H>PE_qu`e#h$OP#?zSL2P%BsBV!xZ(}T5dl|BOiDx+9(u&
zJMTJ>j})=JA9YG8LbXNpy{VE=x>3G5BgY{v?1;A!03Cx+6-qL0gUH+pJ3dkG&!$&j
zsv&DRLZ#x7e77km_9}z~y`a<Y)8%KvJL)0QT{nYaZjul6P?#CY6{J^~snw2F%@%`c
zlO5&-g^yL}SDVaiy-Rd_JKl1L8gU?E9xv`CVYU{F(eu@$id4KK3YcCn9X2Qa=`=lx
z<l{L)%5^r(7}AuCVA@w$@{-1;pyx$@$><4EZ(2*dOK8>}6~jN5D;tB=-R#!B5U}`R
z$iau;B{^_fa<!piLH#uOZOMrKHvi)9u`jS<j>X&9I1h<K0s}^WdrUySKJfh}pTzOo
z^GPX5zjDRvHJezl-vtsdQmwUrZ#v+fTdI*&IXWy&bS0Tym2W@Q(tWwsY?w*8Um5bc
zXy8Otao6d6vtKIqc48K@%OS!#VEB|YKn=Laksq6qfKdc1LdG-Fb96yqrJmx&u<w1R
zH9c(q_Mp@fnVW68d+$U(&B3TDC*pq@-nK65NejCx$VRDJ)%M97Nd=rR^^Y<onYZcB
z5UVO@Jh^=yC&zl60jD7Ph1m|dS>FG{-G12YI(Fc{lQ6?j$8%P08xFBULGa69u0itP
z2&0L?NKg_C*U)Dsf89p_6!;8^(GwLG%e<^BOKNyB<l@scjdjS(dNh${mjuj1=!6ih
z8RQAfar$*{&tY9(ErMcSKn+6&8q@7La3n#fVr5q#$>F~5>$!34G0j?fZy7ol0_1dr
znroLy&hV(udWT3%9t)J;0tX-hBw5Zj9e%3|)+(la=Z<D!g^*!mdg%Mxf*xpw-Cq+y
zyE*Q`UgGziwcGRc4v|fpI?yN_xI{LNd95fwYQc#;^F6Me+TEAy{#&zrSQoDoovESV
z)^T%EB-T-%pmDTyN&DW7Xgt>BsaBa?RsO9<A<v)jIme57pa1J$06)eL_PSb2_}c9$
z*_m3R>3^q68;0tGrFD<@IBYq=&O%sDtC8yHW@ld;!v}?^Y&f(NI|;DxuUz{^L%B)(
z;h62|ez&3R<EKfh)$lWrP6ZUrI<oYr%*EJ&s4st$M>ReK&E%2iT&*laQB<CxW7?|<
zu>8rE=&Zny7m;s7`=*%b21{3GjBMd{6>bOK)<>S3=^E8BU>r*>NKx$=Z-wLS{3!F%
z?y((Zt9wYp4?%PaI(XB{FrCGqy52j2(P=GLBU-P+eS4<Ck)aQ2y#h7Vr0w5aj_9>w
z5D#^o)3|iqk$Ck2vi|Lh#_S|;!y(OF8P;jp>Dp5GVJh-0Iw~T1;Se*Voyk5m>sH0$
zbM_*S<5*4lw9$fc0MZSQ!IES{L0Prz{XQ{dzPYB)h+z6j#NSJQR3@nG4#w;!UQ4Ga
z6!;vzuPUgx18yHp`~PF_9-k}wqP9`TcG9tJcWm29$95;_c*nMFJ007$ZQC~c?BA(6
zRp)vCh4=ZsKJD6T&bh`M_ceCbdD@7IUNcLP&oC&v7WR)WS=#LRQP#q_mWlE%sZz1h
z`P%Sbx->73c%4F}w7rzV8(2dv0ucI)nH%m|h0>VY;Bm+g^kt#7h!&`{t_RN}$nBPw
zN*vrJ!|^5-R%GE|%vrtW6;E^Fz%~vQjE)H2vf;oy!TQ0TCFrlSDG)VxOC)l1yhk--
zR(kag{+4q)?%)L0c6!`;gZzPL-BO#F*<tDxo8W1<g^OOOcv&*pONMGVm1UwvupNtW
zWTLMYzh4ePYaH_r_f&}Mg<69=qNHHa$Gj|y_oe41E++n`rc%~Q?@z&lr*V8|DQJ?W
zc$rHoO>M!sDT!%Z{))AdJT6=Ne#20ctfO#F)-rRwV^i+53=L&ad7U0+9|rme$w|qy
zen^Y*P(gWO++n+KRcw~^NE@VTie8hOLzbEy7kFj3Uar}dz|vY?SqT<XV@scCQAdn~
zwUf@rEbZgdXHQ?e%~C)0jx+%kO3XddiOCrANFwzB|6BGD_VIT$iu*}9cr7ik%h&HW
z7R^9I_V83OK30?ydV)~@f;}=qpUUHophc}NXH|`<f9FfkEXc@{gk|H$G@u*Zk?1PC
zvm<jP-6F@T6?@CR&GAK5Gb@%546yB9O0t~`T`wOc)oM#@HXEu=4{_Qq%o!I5Ees|U
zcV;4XznL(*_OI<Ehb<^)nIyqgyG%N(Tsh%#;fL8!#E<*)ikghs0P@9=zq5a`7+G7k
zTr+in?Y-tdWoQxjsoc&N_=_CYE(pg~pzB9qtsZHbp~GXxRzkhKwz##+1$9-UDCQ!V
z=c31Np#s1mzV~fG;ahrDtTz>eCds=D!s1-o!$7v@kE6GQ3%1k2PT!5&=4O}3yN?96
zeNrr6^0S3~Z!7zhB)Ir&5(_MD(0L^&6R2{)S^ufa&D-GOz#YL3MHFq&yABS8qYPwV
z8he84x_Xc>X&&n&i}D%@hi*Vo{Us@B7WWvKK?V`T(+`sPD-@Ssy9d^18ul>oA>&4B
z_c2!*<+%6^FTkp`@Fxj$Xg26yAtaxt+d{sHbEF1EwV32U|F53+C6mS!`KA5&)yzF7
zS^jRoU(t}K^KX8zSdch7rqNjFLqc?>y7#;z#Z9&QJ)K7cnzF|}f<Cj90HSQX*kq_X
z`Y@F0;nvc5x(~xCUn{-3tc+Cp6~%#vd~^}ksWInb;%TdJSItp$2(Q;P%u_OKmfMU`
zV(zpsE_!s#kAXd_WhepFHaCg`CBXBzwpHJfj(7V(g7BOF1fqt6TtBl2p>7ImU8ffy
zXvG{lDZw$yb#^~DR3Lpd9w|tqC4a|3TpCZpZP>W{O)kU^7>d8`gPyWpT+#Qe(n`;W
zJSF%o@-18-e?SzdI!m|vKyI1(#PQDev%=GMPrPIUQqUGwkFRwXF;%kXtRyoE9%{zI
z#?GfdSi|;s0)}{`epW`%^VtU7B^c9PdttzPxx)a!XU46&ohWi)To_!Z)f`i@*J1uq
zMZ_=l9vFKz?8+kGysT7_bnPwq@XT%m8tPo*L-})k%hPn_UYZCb7$TP$R=dG!&Q5YY
ztiZU~=Gjd)^dF}nL45P5o8-Le-8pBssKvRoo+vT+;pw2T5uEEvy;TY=<fNT3(R|0W
zGM~0as9C@xr*>MV8SXO~q6;Bn#ql#K6D30XVMzJZ4!TikbGx)7KP8^8p%%YBIbx@k
z*Uj|axP`Sb*4rP+`FksVz6mP<{$R42DF|vEy?EtQtWPmfE>?vL7U}1oE4n7#g~|$m
z2l9W|g5M8co2EUgXJBAi<Zr-m_&}?>5?kR0OwT~D=w+kZLUJG-$5b{i=jSa;n|w<K
zFz-G03Yc4$$3p)ho_7H=yFC~iL5k7=V{b=Tm_)y=8mZ*}!*P5T!>ir|3&r*gRoL&L
zjjJ^3^Kjeq#&#vnO^o)(K(Lr(aRjK4bcu9rIlp5co&?7pDVMH5b_OoL3zu+WBJCOP
z&=|-fV=Q#Oc#gnLEZ=<?A25kZ+06Ad#vpiekdLT6QWyqO@MD!_4DQ=H?HhT8kj{ES
zxiSH>3I<;uf-C&4@M`J)GlNv_TP~?4m*>_}LvkVp6ykLLeHv(eeBAUf0d~AwvOKTb
zawCUKVoqdBgE}IA#m>ptwH;xy4WtO{I5X^#@@0)m&C5H=2-=~kx0~5FjUAX3lS>Hh
zi9x+rvyH=T48tO1Vf-zCX$%HAF#2GLpARpi@)5&@HFPZRF-4}tD=`cp`y3?0tgxbE
z;&(knRVwD;Nn<P0*NBk!aeF4liMAx9jSyl-C4PCm9Y8)hE6q*ey>1m{Y~2meI4KZs
zAj|+bK{(HwI`Q#*MPk}X2&W7BcEch{O>i=7A}sq1<5;BK@-CE;eQr6>X7P>Vr5<LY
z)2L(kH=~s4d?mat`}&Q=Qlzn;hE50=$D8aMZWc9=+wm3T5%))q*M)*mMa}?27n=P*
zp(4VhCP14E=<A#^gFUY4H<(9jF^8a|sK;M3J1g^|uJo1!@mk+n5C2cgr!EHMtyv)>
z>PbKE>ev6!f5K>imQNvEjd}I|;Dt~wfRs<n6XE|p{=dJ>|NaU7_x1SSmEiyOHF6MX
z56Mjo7tZ9%<M+q^1mR+s!Rka?y?>Gx-}pHbarLA8cKAQGpD3YA6g%HQsVSLwR*J2<
z&Aq0p|Nj&CRuIOa_ULOMQh+B~iY?nm)|wSMXk@}9AizcFszHpxWs}=f2#g$Z=im4l
ze5yka0qi3o&L>1WVw<A<IyAh7Hm0a%l{^>!VwEwV8|*>BE~exmRT%^pYV^%G^lkUc
z122#?1}}Kdt17FjF>_}Cn*44v%Zr8(HHr5?R%8^&f$^rx4Bou7WR$4R_-i<YZ-Wxq
z<p#*aR<8ac8~S;D`lE>3*A8_x^1GoZ5BCe&i!`F&v#Yzid+1~8$WdBURP}9WXec2!
zm&xwsZl&~%k9#>6kKZLdn?HH(E<Uy@+_|A&>m*|Eh1XjBM83gf1Zl0^rDD~Fqi!Sf
zLQsRsnI}6TD~sOB$|@l<lMdKkp<M2yuAWS<-6#d*fEd}#C*y<zq0${AA|in885Hvs
zznHkc9v4K-8ou#3?UMcjet;^MX|Eq>z6paZHkeKZ({}*@G0TO0qC2_K_|QH+dsXp5
z?7P6PK3pr!hYT(+JLxp{ra?sByV0ew8+3y=0cwg3za^-<AGXOwi9OWQdsKx;F=W4D
zu^GqIHHZ=6ze$=ytfl$hJi_q=bOwE#3Df^mMQpY^wNzA84*ynE!|{VZUu{tRdz<kM
zdf9JvI5-@qt6`r>cxJ@QWf1(ni#-CK=#nOwdOEunrX0}xfa#)9>cbU1fF3pCLqR~d
z{hmlVpL>0I=n7Gsyg1{U9ws@+UONw=%M`gY!iW24j9{CUYx+;>t~K|qH|2}FiT8dU
zEkK@vMu5$)#g4=0L`QOfm>r;gsgWf>NIuZOHKpD3>Kj+9vsBwP_1A>A6VzB|{F<_`
zAH;jt%XZ>c2SE1U_tz1SM*F#Q2<s*JpOCYE=XMB0a+1qcOJ&5xgFBy=b>=$&f4Dy`
z8>Rqiw9A3%Dn^<eu2uAU?TdVzqugj{XdbmeCQKED56@idT$(9%%VZb$ao69H#S&ir
zcM@yOwwWZ=o$+p>%0RTO*r3NJAvqZc=1>d=Bxh$c1VSTo6r~QPZtznyTK52<wtC|s
zxO4&U_9$$Ye$y$eGoZC+tN)y;?_pKOWxf4hj8&S=W|jFN0g=GQY(Phk{#s9Vu;mE-
z?pm7$NUg~jPG)IL3%qyW<zctBFkqH;@JKbs67M}1$Z7xDs?h&o07_q3jVql70jbQe
zDOqXKDg!^*;{+>VWdHS065(_jZv49%ov5yT(_w@_MV}BKu^{2&Fi|n1e#h%U{%!lE
z+oHKVO7)EYC`om_3QPL_uX;rg({l_=uKJ9?M`h=2KfL3-G*xiIERVR#?I>9+c3@wb
za^WOP3nwvCB@-|ajv-rf$c2-UO|HPV-bwu;6s!~R7jF>Me2a-p_`PH(Q-6xRLw0Nb
zo_%zZL?PW}YEX%LYi&18NxVyOdtI0Dk6m?=x~B{|6P12-M$&q=%}-kd5+${Rcr(bR
z`<5YSQ#0XWa-J>6hp1DU-*msKlR^nSq&RHX$#HOS{sXQ!x7`X^JidSeQu0ZT=QN7l
zy`LnJx0*yOA;W(*;LSlm09f1md4<_%97!n7T}-g7#6j=*Qia*w{*XlZ|9kcaejO#F
zD{u+JzSP<KdlVssrmHjKi@u<ZE0b<DYTO-3O<nu``^bG$`dzG~%70PWFd53Unrh&q
zwe9=Q+v)OY#UM4)u3g%$=gZYeH9dRl)#-GW<N?g0n{(4X@NIBt2q)p+Fy0^Rie>MY
zEh-?(2A?cK;qNj%Ihmwm*Ct+u5F`tNsqO26I-KjhKZ>9|ZhJ>lYp#eX2ov?^#v`0<
zg}lr#sPyX0mWh6d{O|On|Le7zd$=fNq|>>$@r&o+ny;?nuL><6&|FEc+OqXTwxOJa
z;Pv&vbY9)f0mTHNH_*DiF35Gw2KM*D517|3J2nrL6#zXH(td_n+M-3#Yl%L+iUH)s
zGkx4uAzQtWqYXb67o0@oGZ5o5N_uE-=TWtwJtv=u75sJ!h2cbJscGI6rDt}V4nf2_
z{7;9w<+k6jZP5yyMB$hyddimZGlRG-oOWKjHFiF|hvjw@=}`+0tXDyn1y-!aP48mR
zq!pY;tp*Yf@58mg0$UdY?U^QOl7B7#n0PK2SHj`1i^u(wx)4$->y@Rd@gpgt6vRQD
zn*(JPR(vA#f`yf;4qfroaH}P`HR_&&o-TjEig*1T7(UD5;`*U+O~3YQFX1mw)TUV7
z$L!C+c24PLt4*Gp&IfXS{w+TE%I=9Tpe&ULq*zA(`x=M207^vAGz<*MH9#xv*rzed
zjllMI;|y8uo+UP-`^8EPr=4Sb;m2I{z}$EbH2vnO%Mp<u0)BR?@`#@(LDy1fF0S3#
zbWkzx-l=9gcNDARp;TeMV_2F9Rzf&B)%XyWflWx*{r&yqM2tY{a%6JQ7f@pK8Kw1c
z1K-#gz~Tt!`+3ambey(lonQ+$QLvPq^2|2Fx3GD-%+l?5=J6jEh*y<~?01W|@k^cj
z`ie+a*BTg^;$3hGCQ@AUx#NkQoJGAj+J2@PMT5DFqMQ92^Rf^moJdXrpq}?Tr4(Di
zT~_K;?c~Lh0o%lsYCT9`6}zqLt7pJ-@GTdE1YOBVMjJJtozB{G3;Id5347)(Hr)+B
zgVd>;$Xcl4BguN?{nQJN43mrKe6!yr6UnAG#n^sI>O?fZ6&2J%*S&eQ<S($_ShUcu
zP%jU!Zfsm2t)Zjyd;ANCFxuvzFFX0^4u<Uc)nu>#y3So-NBeaHJNhi@<L+vfKgest
zT4xZD#Ysu077Sm<U-m12ug%87(g46&R$}hAIfIu@5>C8LJAw(O*1SD<3NFWfAFcYn
z2RlW#%wn{jx{_*yF~n$)zErAe#j7E5OCQt!D36}5o<je}&RNDxY351}Gn)sn-e~0r
zWU?{<w*ipwlVO}U4BHR?m9Om8(yOz8<Y{vok$bGlDPgn44B^|x&bN7#(PZFohl%tn
zUm=Ubf0X0YAD@NRG6LDBNOL_8Kl?%t@qa@LO>Yyl8Ohh$+pXa`xX1z$Qk;5HKml}8
zcF8)oc<=s5La-{G%>q}JGj@Sio`^}i{p?^#eJ}0@gBtz8`$Y|>&%TDfSNeQ*@cR7#
zvY-tH*fuqOb4d?p)#i-gzs|YZH^00a>_3z}Ey9}}zaju1#^JfH;mLZZ8-!R)F+qoA
z+Ax*<e8lyL8<@hWa7ZpXAH1L0vraxmRmo?+%g|p!UIsPvsGCt6GP94P6ftwrMXybn
zY561kTDny>bNR_&xxIB}IFY`RatrAj=~XS)N-nSrIx*UPzJS|3$HwEsv89$Gyul-W
zNy*X`u!@<SUMc=ke|IpFDXphA$xSc!!qwhnKNy)r!@xpFUiLPM#jrCulTa()K!=(#
zI6-ATUcfI;zRA<lu|vOvD&JrVpu;Xa<HWB_ZQ)EFhy4na(X}wt1n>C-aHfu*k@bbd
z`pxP(k*je`=e7e^a=A#m?l*H2KUo<W)}C_H2T?kfXnijtoe?*%6Sv7+!5uz2O1U({
z?TM5C8&<Ay-%&H_#&M^`sO!o)Qhw1^Y13FMcmQu|Vo^uQT-PDBRy%i*$oy8uZOGmP
zy)Jf>;mLIN6`I|<&H`4ZDHjuyB0b$tX<jOA`+2Z24u<o+{s&9%QUgxG$GU>%$xzUj
z#3-HaB7@I;EVJh^Dp_sE1a?Ca(}8u)L4BQd;~x9jCh~Bx=Ch#<HO3d0#r{4^*eb#A
z&f5A8jV{vf&n06#m{fm`X<OQ!rmE(1v!;Q+fWw>Rdk8Qk1zjBh%$s9Kv{cAQ>mZd=
zsWXb_S8`2X?S$s@#re0%U@MQ~=0EK|8Xa2Z78ir;1j*Bjg7>)WHK{~03S0MB<h7oV
zy5rM@zT$YvmO0JQ=)CUK_5YgnkZzuUw~mUAFIxakx=@G|u`G@1TX)PUh4W!#0Zy(Y
zdC063n*?uxSgF>wshQt3UFHJpxc{}1#YI)(bNkALm{Pgqh?!AxAN?+6*BTnA*ya$~
zvmorA(#PFa9hfof{%l-=@1Hs0h<c(Fd@qsI;n}#p%^7M;npR7+dW9V!<5N6d$#E{4
z_N3#xuQ3z9ADebeXRQ^)D2S2uX_F+>aa2mVEKg#SECo>8N}exjAZlNTGVCpB2=GZ{
zvj0#xO=EzQtdq(rzw|kiHLx}>r<5+Pi!4f(X&IOc!iVLOtazI~<|f|~4Mno8iz;ke
zV5KIU;=w<w_dK8SEV|fwIAT#~Ia=^ek^<;N@5?508n)h2lHDCCv4wg`_r5t6-3|d*
zVVHe;?#Jo+rLoCpy>RB`gZodZnYv#K?s{_AtyHM1xWfdUDe_~vsW{_wOG`>JsA2g8
z{)3#{69r7wFrtuKDzrC0QX+r9$?lIRb@4MudXaaym96e37~Sy8+0fLw#cyhU*OnQ8
z9F9bW`*-gX+`c{c$F`+%K`UZ<Dh!H(eee3bns@?TB%vpg#eij2%_<evp*29#7*RL2
zh_B^MzTUp}L%W+5cZg0%k6Wr6XIE$h*I*8>2v9q+k+|re4gL|l-!~|pS&I3TPWoKg
zX@0VBvL5)nrk8wtM6M7A-|D}bFUPg}@3-QVm8;Jovf%n|Ku4q}P8d-^3q+P{YqDX2
zJ~#cA)bAU4PUy{o=#|AHXsGS}seS%haR_ueTZiy{zl%t4+xn*65>b(v+sJN_AVV?!
z@N`5CCktRm?}iiUEKnnHw7uy$Q*0@=-dlWwVmG7%{#;_m%Ts}n5=$e3fki$GoGR_&
zYD{wTqa7|Q;3C@2YNFSkrH1DGtb&i&UzGQ5xzBrLRbV9UjMCd}4K!BPt95(3z*e*I
z?=rrzfg@4|G19x9q!{8IOAx7)Gl&tD?WqOmD-ka+A_9Fxq0l+I7`amAv(>S$^<KHU
z%eXI?#I#-Y;O_$VK1AGT)_RsJpzA9rA>~zO(GTl1lp=uQf>=s?T7608iRu7-<j@Tk
z`|!>SokRRAP&klj+zg(>qZ?VZzZT1l@D*Z0dOUW!CHcmivL}$Yfv216f&Am-lzVEP
zQKxybHNl=V$?U3G;T34LT2YelzwQPa%WF5rJPmT@vf+2_XCa0V7XFNO;-t=+72i>A
zZ(}|O2@Q|F1Gy%33UQr6q{ErsE<J*j{84F1HSwZhQc%3HY2hr6lSatI7E&|@BR72c
zca|W(cAO60RYMB&(~+htk-+hO76B)5<GxISK=V1myQv=Ssoe2EE7&I+7LYs!u44M0
z$1Yd$dt0zwwIy&#{|&*^Q4%&Mwv29KT-tS9Es3pL@!rTxy$v-SQdyIOurhhuN+F%M
zK9!!Stglo?Wmm>&KMo_EyX;qm`h9@%Nh&9CglzI6`>UU5`=hM95KJfTnth(crrW<-
zuWO<es-s%3uoJ=}gA9XkFZ-SQwJ4q!`N%|^z=ddE*G{r9<a$YrOdfY*_d33ocFl8L
z{WS>L7pLD7QG(B}0@AD;El0mEB-Rn3t%VmOkBXZpRAMh`ky7mFu8Mr5sVit~t+Ew0
z_A+c4|HfNmMA1rX2j~rh4V-~56k%a>PRsV{#unh_dev8;C3;^>1sOi)uDM&W{sq(z
z{%roq_lI3o<3{oy=&+fK{zC4;zAdYZ;omdC@;mv#x20bO$-p~K$>Ztl-?}_+-@s^y
zyeJN$6}$6YItopc{ymW1Tq)|=kD!ZEj?vBkNzy}2PtX2RJ4B+*%jqbXi=0)vmeR+}
zDqs11Eu;AHX}mN!@6GY)x`E^bAotJvb4>VZuGO!gAR@POQ(w=D?yy^4n!~C=oil)9
zWbk$V%M9pa4!@hdxjJgT0?B6ffq|^fUTJvzkdh`^6{&R`^QdKd?NRp_Gsu${WS?7b
z^g6$hll~FjrSR7L4^9$tkWmc6W>6G8qS5@V3s9{y#QJO;@=Xn4V(snmz<vK8>n&yi
zFfVp@D@hRqMGOw_#R>NFse>t)jpxupe*B&*5uQVlXdv#6rijDtX+i;?ctV={@6mm4
z1{lz!;n)e`Fl7FuXgqQv;@m!BS+`f3U;+@x!U>Zcz#Igzbf(1$>s{~9VgJy)ugZ>0
zR6E5nJ5?;2jmWXIK*N$SET+-a8gc{twLnyf>kwDimRPR*ZxyU4fxQ~WAA*#L>$N+}
z&@#eMx&&oeO1CMuoUMaRXJmIaZbY^l{<Z$l_Qb>vWC}?|?tlQl7?x6P-_NwmlYoDK
zJVT06a^3DM>3v9law8!AObN8>Fiuw4hC;$K2vB|H!&{fTz$vUD-54pUNwRat5kxR$
zyw3kw-0*@N&{j9iSTA|r)SFI-Uqb-``#6U3gV6A`Vf@uph>Ln&5;{{;Q=e77AF-sQ
zL}W!Q^8@<~@xY;8^vx#f$#uE_Rj&VTGMDtKH0a!$Qs7_lYh7Qq+ni2(*Z{|wY$sWu
zXO$AG{cUdFUZOVAqtY5y+G6_z9A@c3lRZBY!(mh<8SyW+J8)f8X#=EH9;bRxc+a;J
zwws~kGW1cnyEo!I)PmAM+Jv4(A1n$!VQ91(Ayzw}e`Opu_I#uTsa-NBHpPW;IrzZ<
zuzDD_ih0S6`^8IikrXgvZhl}D7}?)m1uOcaKiOoF>S0fzI~8GZEztjcS+$e1f9@l<
zqL;9Ggzn5!<Z&ZR<k8V-T207{M}!?{!0#isi42O(CNo+{m3@7rt2&(TK5>EVJ1jjJ
z#XRF_6gFn;miNeVJ=^+hY-L_^UJTp?pQ)V`k+a+5%Vb+{VvuhITJljdP-Ix{kw?Ts
zu+QZv&p+`Wdw_ApOlExqWNU7vhb;?1p-h$cKxV$O8gQ3dGwaa6qlb>n!?WzA`Q@v)
zy0LiserM7~gHZfx+eEq&^0BIu;_+hi(A?Gcjvn+MTI;7tE!$jB{Q%v+=Ns}{l*1uo
zO$&)f)9yHq4tqjQY1U$Y-M}Dj7=nTe!7bU<HashMg&j9cJuv0&k4whn8(7IXJ`Y2d
z?3kAbMtBbA80-nFtdbYgl)__S5{-k3H(WX)Un1i1EQ&{67F2{W2N1cp5k%S?|B%T!
zWSHr=N9rnJ!y|Uee!nz1u7O=SxM5tL84>Yg?3FC+%eLkmX00EuO5ssrtfh&rAEoQL
zP)l~G{R9M*tslm`o!z?x4Nk~jQZ+>}KTOASxZCJ7srB}9lTpeQ_<`5a*q+9%wXxjS
zJ)qXxsYDD)1oaDR*vHH)&~EX&o!sjG7<ju|`Z}fRo$tE{CEp}Coo@lp?Ml;hR)r1z
zjjRS3{r!<Cw9E|(cZYHf{!{OM#<r<GzjAE#&Z%E%Z9eeC2Go_rabVK4F&}|O>RGVw
ze#Mo!F+a+crurs1`$lcdV5cgMwNVnu1K8G*Di@W?q?ZQ1pG_Ta^aarliu`pL<DnG^
zOSqwkFr2)VW@-ce#yB38pqTgNW$KYpROnk|FCI>r<+V}?%NUzz*?<g83iQ}et}aLX
zoYCrEvMl+pR#vIgG97hy2rmlzYlP5k7Zkn3ttcq7+cYZ4G#HieoY2GTaXCehOH*aR
zc2f3M8%LfKYK@IN<%F>-sZAQvL_@njHB^7X3-v7)Rra)M(rpc6UDYu}?n5T01*YXl
zSNMc3<aloqIMRW73fp4l(e?R9w<&(m+YA~Ndse(EQ*a4~2+zv5R=*(<s&*>~$rENl
zai1>b@&-8+d22CY(C;deJ=6oN2;23ec}SH_@XU)G@85aB3<<MVcK!44#tQdGt9SKc
zQ#vH^saxP^P*rbxFGXKZ{}QXeF7lM|sa(Y{N!k!*GC{F`G%i<X@?toFdvLZd!@Ie&
zA1w10wnCnZiz`41nSBb$GXIL^v`lEdtN8Z~u7jWE5Phe3e~BJw{SVrhb+Y;9<5=vX
z*$^-kPmy-9f%t|Fc<q_38fi@x4w%ECVShzZ<GicjHN>V-3`>O$2SvOrCDDBOX2n(^
zj~F86pr#X$EQXPp5poDzO+yj?YXhFVt()!lD6s-0(v#}mBB}*BrS+9>X%DG@%)b=4
zNEy#}#MVBI2b6`~f{XZ0^$T4OKVWm+NPy)6Q9M3OR0~7st{22J-3N=pbmU-(wS?Pn
ztVnR}$C$0(Le#Pp<TXeY$Vf4<XmBcwX=%=jN$Du@<cH+FDfR(NNTlbhO)3Q}EHILd
zC_1L3VGo#AI!~Tw0uZVwi;B*U9PCeN^kDz&2I+1ja1K&8NsJo)u+LzGcx{x8F&=dc
z<fDbEj!3`G&q(WBDZ^IK#K%qEl?@>0^ZN@iiy+h<o92}^;Ap?9Bw<lJv!ox0)qOqf
z$Xbe)59;5!)F}=@5LijI>S*NigLzhq1sNCMfKBl$;8xo7mNq!!h575@|M*}5rA-YR
zl!NWKMLM&sN_}o2Ov(9+cBb!{E4KX(t*ZaMm-jnfcc!nKY0*B{+Q>kxfz4pg;hA9P
zWBbKRF?;b(!Zf0p?Y&?PVLVKSsE?0rK0QHh6l}yO-q6%NIKD4-2~6m39n=P1{OMG7
zWq92TUcLMKZOlmvTSs&$PfnPF;d+fV-F50%+wyoQv3%@IGzOD*DT)C&fnerPU#K?Y
z_t>d1*1=PLG4;2lbvSpA4VH4Nv!q7s4qY}E>|8&Ph|$%f9!)0%qto-lZ3->Njkh%s
zw#+T!mmc>9{24tO=!!br0Kg&ogCnF9idFD>Djmp=XoC68If{?JB8S0xKUrb-kP8P?
zQ1lAH75*B!i%MZFZdA=;5J8tS221eM(SWV22p1AS$-Cpj`24M;v14}^XlalzH<L6{
zNjMp^<)suBbnlXjX=xl^Nj>)Ca@;rKba{PPGNZhglP1uBoBPsd`w8wsOI&n%psNJ~
zPIJH9$C#ERHc@(9njbn0CqX(KNfNZeNQecE*no)6QrZ~`Y$BtCI+Fgm5+l4u<5JcT
zijs2wCo=}P(xV;G$=(+e)2PtqCzsHSd;dwDQ2dn)Dp=}eXH*?}isYa(V^ldt`d6Wh
zucJbkRj9TmE6eXhCLWN#w&PN%{@v81fv<UG+>s@rC=26BsIdI~Y1gLZ{<>M9J0_{s
ztmZt5Y``F{Q<-cMeO2<M`;&s80?YL-#^n#|*@viEYVqs6!1Y(QFqY-FbQMP%xV%aT
z+2$eWA#5L)rsL4WJ<T0spbt@0@0k1$^vO#fm$B4~#!Caa5pgs~Jol7{Eh;RBG~1N$
zDYba!?`d<Z7V)Wvh#z!(#ru(FYi>><<3}VCjKo?cG;m$%R%}B!Y|zryL#K~YB<{`C
zPo~&|sVHfxvqk;tPGVV$<bl0G#N(k;22%X>jFt}~`PorfFUoZ%)64*83R79+Fr`0N
zMum6HxMdCHGFh{Pp~ij)k%7&fv6|~nhLRCSA2|sO#L}}E3zMX{Cy6ZBz4(1N!7(^n
z*@)}=C{D5Y&;%Fp<@rZ~(Du$sAM28fl@DFaqIs}kuuGEBHo01XB>idjTt>#bp3yE<
zFBUK`wY$0KC7cRzopGmbaXsO)cAi#$8TzB)MR_*Jji(XOc)mqiXkgi=W8NJr(bOWr
z@eYHUId<E&f;yln+4a)$#6)y(xnH0}o851iGZN(Vq1Y0Q&-|BrPGB0BdS@{?Kgo;G
zi`qRGddLe5L&T<gq(#uSZp5V23dAN`W!7722FJlCX}tzx+8z5Zhxm<19L$Dv0&c?7
zCTB4$dL7irMy)xG<3{9kz(zJM^3z?Bb@$#*WO>JLz?83{aFCeRh>eJ6`EIz{At^~`
zwF3_w14Q!ORSrAxis8{A(X1i}DDg0;D1Tyjf~~`|_?06DQy>^TI-C)*skU1kM7ti+
z*<So(<VX`1pYD;kRAn3xHv~n=pcOapiDfrlAh#M+dHcH~b4m4WnZx4xEGd`4vx|**
z)m2d~SHSd*aE>uzn=jJx;L6uxdm<x8yaKPpZ4tcQyHm{@gJM&4Pn2qgQ!me4hNG*E
zO_e<{JZc-;I95_fAoYv##oHNf6037#`F1wmlmcv*$smggD@InVRV1|&?CP9p#n1%R
zl9)3UFe>};_oy23?f9zH^{wxPwwRi%-;&9sUrWbsUUMUZlIgNKp6qGseldpUjTr{8
zOM#;~T{5k&74F#C*&@T3i2+}qZ8gfFDlZ?KG{ZQV^0lN_{|qA179cl`lV~WlrJk5*
ziZ+RDS#Mv~F3(d-%FOapM4*9K@>v~lkl`A4n6I;lk!CM9HX1+q%8YVBX(b_XH*3BV
zP0fZCCC?g6FBc9H|0noN68--27NZlB5^v_RGB~6IDTsT?(u!dr%-x;kz8rGR5oY~@
zK@#tfXm(w3Dk4$DGAZ2v?|3y!PKh@w@{}D&ftXUZ=4?XyhJ!t#-`X&wu~R(i>G&<h
zLbh@Y!oRH`bBrZ!ROE*v_mR^~#n~RG;QWS;Nq59fj0@$&PYq>u$c$mt;{<N;?62M1
zl9t*@=tQKsI$~aE%>GDyVztUX-%uQWJex4Ko(({*LwUIGkOnuO`bEuUlnk#TZ!Bi;
z=UPtv<g}6IB)jE@PBHdswZJ08L#Eh_*vv-`+1E5mHvR^Va?}C8^pn#1?~Fj;{UkqG
zS8xdt2^eNB&YszckYmDiGe6kVE)LnIF4F$}9MFZIx<>(qRgCh(HR?^r(9mOMIw)U7
zlwYN(sMhu7D}#(Wv59jzQc@^0aqaCi`>^6B`U`2jkC@#(d|$d@V?qZ@WnLG}2<xX7
z7~>SB$Nkp^^fhsWc(<q;p^?Es6JRV!7Dlu}DInGlQR3(qW<*r!iHpJd<?C+ksVF(@
zyf%9jKRGcdnV&7?NG+`oDHt;dh?%HhRA-)3qrxNjp<O44ZT4ynT>HYKjuAj3IVHAt
zy4d_U=)8s<VFS%KP}H&-2Btf(aQ{$Uc6#1jDx{gUS`iuMF>uN0lCvNprC0QwRPsRm
z@Z*knwC{Bmp)(*gs_kk)7H##++q&Lvz`YUe3c_OWFOR}AsBmLvI*X2S$u!l3mVD%H
z)t9=py{e6toOFd|d4^lF5KWK1;;cO{-()hfq3Yd${P2}Ic^@GM!=0#=j*K1_M9{tn
zdob)l)N%T=N9!l~xl0J@c}*DST%$(mjy0(!*9Cr4d$wrP5#E{n3wyt7V|WVP|4sha
z9c6gGSnsrYjsUA+n1Ebq|L}A|SR1EyYPzC@fET1tD}i31^tWFS(Js-V-Jig*-iO&)
z^S<ww@$ysj&LED>6AMYaj@>2VoEjp_fN<f&tX#H6@A{RaW4b9=0ah|lhu*Cf=IMKN
z`}<3W!*PYx)#ZLK(<gt4x6J@A(P4nbymOJ3iZR+@T@~W2y>?Z<`6SM31Ni!KLc5(+
zpFZ53QWOMPsL}cz`y^8T6{Dq>iLeB;wZ}-Q*tmKkx*716++^f)a+)?4Gr5Qfs)Q;j
zo!eoy-$Dc&&*)nZA=;i84N6=Z;=QKQO)N4!{g`B<lLOE7A(%L>V*)QfxjBc3V~pKm
zZ3P4!G01%~CfM9_w_(Pgdo1{1-wa*@dAV=feX5U#ChI0=Go$RMH>637E>XjKZTEnv
z&#d_-+WyTG`nqjtX9z2S-`ZI61gj|+CE0dDOmdhw<X!!{VIclr>-eUjC*sSyTF#bt
z>6^vwojL_;r_rY{BHuN!7t<h`>a-;4rP>a`^;xsghvRzlEX5ESwX#mvJRFD~_~s{N
z2{u9k3137&Y;W_DLZK^PuTQ11eLd{7_q<3nN7D>Dx0#gU@cl2|i^PzpB37k-ehe=T
zXJ=(}q1J~`83&nIC-d=JZiIiL(d<809|cNO69h9EcKVe}<Tq^eCypq1#694z2C!!`
zWrYX0Rox8O&Ue~^h|WX<getR)?DZe!)o#1<iT5}x!SskjNtj_-n(HM2LKlb~f8v}(
zJ%pwqP9$Zq3CM%QmL?Yc-8XP3k&yfpvM$yeppg8QfIXn(7#*k~X7ZRi-!@59vqogS
z^N#G{0nZsF-1_^2aQ=}o$pUUq{XSdb{W7**l4S5#ZM1y2%Esn;{;=WbB?R`4fr&fz
z^P<r5mNlZn&QX+`v2Wq<Z+Iu8d#t#9+!-dThRXx@ve1ov#9%36!+ow^+fU{qKQdS_
ztrcI0ao{cqdP_fUH5{6Mw#oBY$|9oSg(idc6&aGF!CiGRr5``(QQ%QP;GyC~qh<?z
zS!$4#klKAIa9x67K{wvIUaG9+R>$HF79-HcM4Dal@}C}ZJNh+mTp)3Fz=}}2mdg>y
zVIJ(3xvdqcX?68rnu;4H^0M~JTXak1&c!0dHV7SB^vB!jIS@?J;nxPbX6VrZ3}ydV
z>z4~x+gt?Wi1<>7_fx(-L^e#~aUli0nx5Abl*4O(NEBL4OXII{*1w}0=JNgE*et)B
zQ(2Ia6r#p?Cx@fN51%_FQewA`W^ls3@G{K3wxWJ0sA_5n8nEN8|3eD;$|q8)ux_j<
zEcfIr81xe|CP<algo1=pg%)?6nTGo*QQG>^3J`OoHwH2$AK{Rh?gkglbaFH-3QoWv
zX}xahD<6vg`y3(8z7P+|A!7;i=JWN7HST((MFO{9K6De4t6s(OkN!eLm>SZe4sQ~b
zE3NG&8kSjVrnv?7LW+pxwW?-lAu;~TnXfR(SxLVwb-%o-YIVuNqH~dmiZ5_>1#a=K
zR1N>iJhE5ebW<tOxw~|;NePMWYUnifo;jI07g2^E1lgkY0_AV&e0NxRc`4j!OShbF
zhl-vu!Yk!}BxppQ#?6oY2nBdzA!0hU;6;d;iK~Xqy~@J~KfbG<#?4I)SSsCO>AkTY
zlwJoTQCNz>rgMHrb4MP$`oJ;_bKHg%60o?R?ZVKJb=)zRe}Qsvag&8G_e;JWM&<=~
zfdzTq`T_lh^)-K7?VU}VC>Zel-}nmNiXJ}D8jQf2ayK&-)|5$}a7^!@-<<4e+^NC)
z1W||-Ys?uD>+@~QIT>>UH+g>OHu^|w8=Als+UlL+i)`N43GbsaZ<grR9c0|FWQVQZ
zke-a^dl;8|+~@eI@G<Ad%Y`tHghsPH<gl&oZjFn?0FEtY7)x-CSLnb@r1f;nr6N1~
zR`A$fJ<^M7+hr{dps%~**SQWaIpLrscjJUXuwK(2PLr+d+`bT4a>(C-;6^&oQw0U`
z7zn<_iov4$Vq&o6wsxt6Tt9i{9+ZVq{b5DDcIX4-f0M883%~S%x0eI{;rLD^k6>37
zcc9ZnH%oms#GUrU>^K_rBqCqEBJHkT9sWg~#Jk-oPk?gMXm;-7!ap*K|J*7^>CiT(
zY#p-&)F8TCF)b-kh-nM@nUe9yk*>oI`G00H0&)<qvS&BAYPA!oTL!*!Io7{};w?XO
z?@~QpP1H)1_aySe|Ncp0yQvrh#a_we9ga~u73*?7Q=!-d1t|^Mz#<oIW?^>fcwsBr
zYL0Mb6-z52UA|l%pZQp*KiV&SQt`+Y+~{UoG~g`cwedjP!!w$Q8O29cMKJ>3$OTk-
z3+Z-fWG|CS9`>3)_VX6;HVJv4I`sSB`twuhpjO|2^iL8bj1B=v(WK^pC$inl{DD7z
z{lyWWZp-@C>-)~Wi1HIrLJ_I({k_6N{?SRCQQMu{lwo)xNpOhqamCL%4?=Bv8aXa{
z-`YClb=3;s|6o=WagHmK6^QMuonNqU@-6rb;kh9caY!-P4JD3GyLCruuYo(7PQpJ^
z4-o0C!V>!`>}&Fy6@QAyrenKLPQvCq*xQ+t#I@aqVkl7$Z3UH=I8CxF@hS8beXv}1
zdhr8knw#v6Rl2+UL2N1n$4Lr~ls!*GU(-fqf8IY>1pe<+TCNoy*U8sHVT$DsJ`YQH
zzN%#-@>sN}`!4!<Dp#yTxrit;s<E7Im=)uzDlCHsRT!Z2*JHw5>q_{2?eItbe0<h|
zbRsCO1npu5^nnfc{D&9He6AbD2Qi;n-c0WnD}Tlni@(0hatU5c?Lt%#i!+WvrmVb0
zhqGwl=JWHwY2CBzV}B-v@i)UT<%+XKX)<Jcys<Nq^=9F&?h3r;mYtY13(2*0EVie`
zun&^Io^u|_iv*spNdV}T1TrOfequ_**39x@SFJCA<S>VyhA7@RA(S5MjKkr2$ng9e
zs?W63iU}S0C3t`tCDAqQK>%PblQE|BXJnNA`3}qIA=jTHt#gp8Cgo4?u}4T<E`XJu
zq_DNuYY@rP7g|mb`_G1Y9bP^tC+RpWep>Y#(c;(`juh#8Jq^oG=PsB7?2NcrRF^#c
zasJ7BVcpu;6F9lunLn(8q5t9zw82%OCuN&|^<uj6v0P$x->0ZUx5?t-?@5Si{`nl?
zOEgTxj{$_4QefD&LLx;s{$dno5C0tOyQ*uRBKff9IqJ-k3imJElnSmIw;WOS3uz`?
z4QY9s3OG7wSV*7;HGBNo^WZs0!y~n5l_ScH<dmkuMf#4ujP<WiWR%M!N9Ky-W`;n#
zoS}qS!j+<+CQ6pD%$Ycb9zB@jD@t7;BO&1GxbWSzDc_r)U8mOIPuxWb>0>StI#vdq
zt-vyUaF~97)uEz>C8e<gNVO}nnL_`E(wX@ch<p9Z(nY9*p|j5F><jnD40gTz6i3Pq
znWErYiPea`tUw_m@fe5b_g!$2$#Hq2VYq~W;_5GEXb!<MNPoD%DVA(1Xs%u@ZR*@m
zX7;&6`P!q?iNy;-U8F099dq_Ij#UJw!qicCyVNo>KzbOF0#_-vJV3TdF+AYaDuVyM
zKD2)?wt0;|jO1$dqus&nvCETcUkYGyKwxrk<vD)=V6s`3-qj;sk0Ha`nltR2eyBj)
z`{pZ5UExmI9O7Cd-;&rkC`U=OE~A1sbI!b#3IZJrk@Bqb@W<Zho{y1MKsL7NAP~B9
zz%XLOiw%h8BFDf36*pGFM67a!xx7FP(`<j~GMTjTdcZmRBjMsu$8Q*SE+Xg526{JE
z)Nz*9HxOch6et~eZ%gDC@-}w#tY!ah;%N;E@9wU6mAA-mU&*%|6i#$otoOiYQ^|_k
zZ%IM1+8LagZ*M{%gReG@K5sGy9De9j`gzm^XJt&!%fw=YfV%OPAN_-ir#9M?0`Hk_
z2NKx2Asa1`6sNxyjlDb)Z8g<9vEHGl5<Tm>sbtP$Gd<I2Dyp7^lo;>9h2SWfp&g|5
zm@FZNQ66~Z2uhkLCnf5;D;aVQ|4B~ZHF{z^yfj?|hd0vF>wX7v)#hDP9bG|HrdDZc
zkDiqkl=!QU2pTHde)aIvQ(J>x1hGB*VRaE(T4c+P=69&7%}==!sExWB{!LP*>*|z@
z=DycA4F0@a1W390DOfVj5}c=97|>)9bayglo*=v!iYQnQjel&Iw%f7SPuZrtuERKF
z8}zW8MfuHZ35GTmij6j=M^lPWLnkS<7-)7LEU_s;hVJ7Ek{ltb7x2GNtjk5P8_}HL
zhg)4)GhjSLb7SVonGaK_W=n|8Uq#dq55z{IHBRx@sG{oU0#N{V(XevVh93KaDc1<7
z*diygHc}=*K~r;9uA<IAQ~Z11RtUA;?i0=Fqws7=jqSk+Vu^K$@i_?0bmVjR@WCuN
zGg_Hrn$t!pshen8PYOtSP%2P(e$MYDqW&VD*0$MP(4gva(^xmoReC2a)jT6b@!W_}
z3S$)OI9>nDa_)WsTxPg|;2dz*Zse~mo^1IVrcNgrHEm2XvHwQH=Rm*X(&X5F;>iLv
zAQYPCHEK7K6iY&}?*0mkb`v5P(G#gUF{#)8X55P6*sq3P%sDxiTJIw}ET|}y=h=DV
zb!GAfbLbW>Nr%cO;#Rh=^u^$f!o<1aU^49j+=#5nL0|$jP!En}S4k^F(op;=<Gy8J
zk68Zpq5NA%v%nb-^eBBS=t_IEiHP!B0N6gRt{D$V$YzN-li{2Fxe}h?_F|dW>de70
zY{-TTLl(pt>BRHc0n|wA?wARjy~p~*b-TZ)ZkK7!U<a@I$_`UW8+l}fH*}V?Lh@oG
zslhc4bRJ|Un5WUK)>Q$(T(g`LP3U5U1lDr8f-_q#C0?6>WV)%Nc@%Jt(Lk1hANNG(
zr{)+`dF5~kjjwkL?p^JKoF!lLD`i}B@~ZPuu1*Cx_c-L0E4EFAUOoIAmCx6)!`oE5
zV?#sUOIOId0HNsz?lu+<(O?M!(^VB-9L5@{nsF>Xlz&zV0olYJmo@B-Re|0vh>p82
z`476Ga6yJCwu{ksJH4e?%$>z=g9g+L`!NVEkO6&nC-;g@bucPhrhJZX=6*0jQ36I0
zn6BVelwmZ5YSPv$zsuG(Wyx6Z!A`8qacA5Od%El%MLSw_VK!9V4LHv$U?)?wdWdK%
z2OWtVCTW%tDf?K|XB}vV8Al)cn9ixac0}4qVlCTIe?X8hs!`{{yUz*Ro2U1Vkd!!&
z4P&oeB0^~z${C?AqcD3S7}uh!Vf`aYKlnH9-C_%gIRNXt5>ijVWa8iTluARQ#WJ;J
zi<Z6w=fU;DW>SRf^7c*k%|llB&h`<`^yiaY(v{j4dZbDrarsK&Q#2BQD0u#LrT{C@
z<G?!W;)Z&zkow^NvjF5oB5kPVgdL-RJu-L7*6BDVAHd1CIlK<8<F(%YaMElhe{@7_
z+nUCrT_!&g+4P}i{sCvn8W}q~cC5NT9J^D&zVia$5kw)NEco~Nh#>BI!Ezg_hHZN6
zM$Bmto{mlNjaOUSI!BTISngBfIR<)wv0zE!iLCo$`{TFO7IOk1FGW__6@VnfifT8W
zSp8~a&$hWua`%-uN){|y%&mJqb0Q4nnhECye86VIllz%?j_aTm5}o$=1|l6&QXN*u
z&-GKkQ}nOFwQU<Y%+@uzQ!Yy-)V4re!;q2#*UNmIU>MId4x8`gG3MbtD5IOIAxdWG
zZL!b00pcA*>H}6AB(=yBu?_K?>%oUMB_AC(!M0i>QlrL+&E~FHYStZQOCthwOk>#x
z+QGBxgzcA{Uf0W^RYo6t=3sul<|5?ip%BEd-?k>7k4EHZc#B+=WR>LEVs#y6RIX0W
z#+-7zt^<eDVOvL@o4C%ozb3|6`KpBQXPKR3AF8Wn(k&S`4~Mj1H_jz2A-J_doo{^9
zFT=RdIo?=dnbH`LR?m2Mi8lgy0G&@L1#_3eSq`0rX_YPn+thnj?$rNm6qC@Ng|m4z
zq^0EAgQhxhT>_+u27_z;D>K9Hoa8?polty9Yqj?wkYet;G$A$3IC$Y4h~X5DG1%9b
zFAXEYah5Vp*|J_^M(>a3QOarxQoJPEb3RC7Qe{W*eB;U{dLkGO{TYg|5`|WdE$XCE
zjgy!@_+tDp+`_Y6cSb9)j3~S*Ao8)MY*%xe_q=1qp5^bmzo#fZ(Zrha;)qj-gEDID
zk=UVv&h#(MU$;_`gex{F1ogbdy1mc11_ff6SGs}`E%dilqk&ORM=TChST%c3R%M%t
ztw<$XhE>Y!5~z5}8GuaYQS3!53AsLuv809Jhx3c4L}GYBqg0!V0+Q66GEE@V=fA4*
zxT{qf2Ne&5#v<mQeI?wDXs^Q*Ci2Rb`uD5LxJ7h<W0EzMY00Gjn)rxF_MtL<zNYZP
zJJbrYl0j}1auUx_+N`StrY+3D7=$6F$GHL2Q$OMHX2`|3>MG1|jsJ{z)bCpNvaF^x
zd@+*_lU(|6QB?KR+>+xT!`s$&2<f)5C}qc%G@-lBHkOXap+jKpba;{@k$}Lf_Sc!1
z(7KM;B=FDL3kF=R<B%MEX1STX*MD1fbOuE0!6jTdf*UJ%N-9c+BmU&Is+YPkPL;D;
znO@yYzmwwa@@3tt8MT_drE-v!8RZ$Q@u9984g#l_LrlCVDKZSqHVuj#lwjYeIE20Q
z8_SjvWaZzwymjj=#@c(cMh!37T72s4-1`u^_%^4&jc)DW6w@G?ep$SPpb^0%{@7#8
zjk+L+cC}xDT!UFzHQ;nY#dUC3ZYOyi!R4(xQz%}ua-yYl-TeD)8zd_yyr3E7p>Stn
zXS^jitC-Tz*tmG9(PlYE+R&z#w~|xsyqOf64@VP~@$7R;!K-C+xf7VBOX=28n12<X
z<Lvqmeg<OC2Rt_7UfVHbj6<5$#8&Wb5d9vQ$3VJ8Ip&6{C_PSZP#a#vr}ZMTsQ*uK
zHCy_tPM9FyVPido6PsDr;T=37+X>B|m>zSOsjMs^ep{b;HFs7CC9)58N6CUXmSx)9
zbzP#zfDdo0hm9|VJs_-6$9L0fPRY-!k47aYNpD+&xIT_@sd*Nm0QFYMfj9lQjL(t}
zqE|(HN8U`lkCdQ>VuGd`;(rjr#dj^zd7ytVsu@skcst-NHQy?`ue2Xz-gYwUaR%iM
zFvhcQ-8G534kmEmYV4CHc3UMSz?B>Ei8Hx$9o0iCCNN6fEgR7z8mK)mgW%wXm7XR#
z#{`dA5E~k+TCW>0WMW1cbZ<WCw*eijwBQUQwuvh@axpp)#nV8zTuq+^y<&#7W-_*c
zmuidFwqG^fKsArYHCyKkHB`mH%UM~|JR8nGuU-4w-C~S2O46Kb;GDTyCtC=Jx&VcI
zct_2Ms*Rd0;H`aAV!FQ=s*XXjxjhikB7-{_0Mzy+ks*-tA>_Z>=<76YFsJW6pORN~
zMb#kfPt7RV2rWcL$F$0KA<{=0VPDnuHH&)Mgi7!S<rDjNpksG1znyXWpk%zsx4IHd
zx>3p=isuG>x%<ztH<ZQHUs>|5+7TQ}Zqg(?AM>wfm3R)IqossVsPoiQzLWwS=b`$i
zq=OpHLH2!C$O<OZS4#*wUiuJ9O%~YpMcTAp7~{O~?xwQ*K7ysOIW+1)e=CG}Aa#e{
zFNuQ3hJ#*jHC%C54mX1B#2Ve(ms=&4c%>OSJI#k~sVmc^IWLR$*@aRH_S<M}cR*J_
zr^UhC7Oyg|p+i%K;h<05%zX}krt_-(*SHvjOfc2?btJBHIRaq!BVRG08j-QGyJ69Z
zUGYz2<q>nqRHZDD%mcW-%<M6n6Cx)}ygZ>?7OWVl#?uZC3;5mIZq6R!#1)%rQf_)X
z#OSiSTK)sysaIlK+q6YT+kaXu=&G#4%S>p@l$XsJI%YtUtGyP^S&^3&J>{6EZV@As
z(-#QmZfrG+){G9)c4T7TyD&`bfGo1}(&plr8%u`&Xl$Xs=m8!VmdRG@Fwo&8h-k*6
zpu%scj5M~lbF*xU0)ou@I`a$4M^>Y5guX`%|JHUls^_fhn~=78c`3ASMlp;P<{cef
zXY%IB|ATi`^!j0hhG&0cZW&3Ji8=&}Q$i*v;ew;^7xnyB+-CVcc1qn5#C5RB{6tz#
z#etwH_JjG%9pglUs4P(f<---_fzpf85GzAh?0s~&rzp+d=3tv4(@49a<uBq9z%cXG
zJWXj4B#`h8fg`ic(ebYi-PW%(sYNz`&!+Ai<A=kPkLA;ujT_6pt6~VwjD0_%J*}Ze
zBji&pve66Dva1oaY$`onRO;RT!`?Y|R~B?@yJNdMww>(Qw$UAQY}?#1I=1bOZQD-A
z-m&exdA^<J{DgO$udCLmy5}m4T2*silX+*`$qalrnemMD_p0GZYeyL#dB>!+Qp2fx
z#7f9}bgL#rQyI}k_JZYJZ4vwdbp~M6F_h?R!1X*8-$0cpJ=Vw3my$zJCU^)R>MQ}4
z&rBD+?=DY31~6J-dyyQNbbiCJdhZ2m!HrzZq1OwUP^`RpKGo)1m_xqyAMTX$uFrLZ
zUgGoZA&ON$|J<7*<@;A0B`6q6Hx`D-TZzMV-JA;nj`I<E=QTSm?b(05^`@JopkPM;
z=T7loqt9=)>2vzJ$r&J;5nt%^0G20g&X<QJi`f+Z^f@=Niu;9FUxPNs%TjX}NRwk1
zz$2}pfbnsY%H#4^8~A>I^R=&1cD;ly!=r!`jyx&7txnP9`q8HU^?z{G#hl8;JXC}9
zd)@vFaTNF+U8ET^u3H=ZkJZ_FTns<nDPmz-&d#zwVE^k7i_`5-hI=1ke>4HTD8fO2
zw_3|6h1|CmBMP`ilRbiHW*<Z|$vMUP0A}lF+iBOW47WR`9zqB^$S~gKuoC;{aW6O?
z>$zb2kU-OPQy4ZygO3?8NlXUJO;&Q7EArvb&YrEjxMqfyUzxBJhXK;k)cMxSc^(&)
z+=v<X;BZ-r3?+Cd^&9Q8F4w5LyHhRbdP=$roK%blydNqpKAs~1T0&I_LaT1~8Mc05
zp$Z+}_R#?S?*j|2di+;*Qrc}|6YX%kuBI*0)S?^WRK3}R?pTZiweO^13%iPa!q_@r
zDSoD`<tAG3%*EKU9Rh+BG(+M}Q72s6n9KSsAt|LF@E4G5x=02FyPlA1$<U+9;h&uX
z@p@3f893$L&IXq0c46J{kvgB|cYDakq3bd7M&ZLe$@dCG*X*W+bv%n=uz3~FW`6QU
zB@<lAe#q4xPO`M1Ps46gaoGjg%xWW4ypr#BoHB#;ABYYlKUz|Y=8*y5GO1n9vT=EA
z{gWC#G15;pmXq*$`Q3@idCG!gc84S|u-d2b+TGV5AxOP{rbl=?3iD9ny64#Ubg1*~
zNX8eOIbMpduz>CCH>oxBWUr9GoO<d@pp&K;wqHjeU>Mu#cI4273M7VZAOh`8z+hGK
ztwpnc5qrT{w##R;(P^*<^jiA`m~se}g0jPG=01HiqbGW7{duFCMIqLaZm>y%xw^ia
zXs$PhBQqHvb)Dws6|F!N7%vdpsG8}!;J_U6+vdlI@U@(h!TAElQekZ_3L~Y0l4Rqa
zmkS3;ouT>PM*W5!n^r7(I5zb0Lg4yQXOP`FKJ*DQb?F0W20Hkfb=;||4#82T0Xg8-
zX$o@_ZBB*<eW*cdAJSnFUEt}@_}0<4uF?o+r^3|2emG1%j9)0EpRpu|WRYa&JDtHT
zzOd7{mUt0&m0Z(E7KXxniWO;xjp~Z>;6uzgJx$lwE&q0F9a!3#w8<~)u{|in_R|jY
z;ozuTD`);vl5sqlVHS^5V8qw7qQ6e|7?X0$Jn!z}WAuBeo0OAGhn{#vI_*R4q08}d
z5#eDzXKc&KAPr7sY0iX6w#jSP^dkA74v`0s{}2ynfpO@4_KggQ9Z@lmROo%@GBg52
zXN!yIiimG>c0bF{!(4HSl4x}t($k)lVh)DD?7!aird-g*k!=k+?J`iTME7oII3A2X
zq(Np!y?q5na%#05fIy?~k6TPe!_fpF(pnF&+gWsUq6=_K6v=4atb|&FArjB>rzB;@
z&_hCM6!!H7JX7*0DO5j_@$FXgFs7w)u0QUYZo3DZ$$KQOrOtb1S@=6<S`2K}r4SMv
z$E<6mo^M*{b#|3vMu&r?MfQY)0~WE)?Cs-1YsUih*jSV_O}cWw<*@myU+y}w%aJ3Y
zY$*+iKR*DJdB@{{kNGsn?VNH?ZzavPyDnKS=7l*|%FI&<S9r1J%k|lyJ~u&j_FukN
zfkY!%U01>N9S!CjrKGlEqTvub3emX5m5@(t%q!4nuDBv{`TY#t5NKURNJIPJ4!goF
zth@TMa*GWy%zyRx@sA$*iXT5WtG<y$eccC{J{RJCP$lk-^h04F!-Sc@HZ;v}p#Vp5
z1ZAUmb21+G0j1?XW7mW%PmMSGgsM!^(hcA-po^y_-Fv1LWi&Y1HE?nTh3Q#jg){ca
ziBZdfA!rCV%Fw3$C<VrXO5pa#dgi>p!e-*X;;0n5-p}p6vXHMEs<lD1tEgttLgs4M
zAZ6NS|K-hMf8Iqh$h6rb_OMN`QG-jd!-hZlPe}KFQbXT7?ZtSRW)`TMfISHLZ^kxN
zK8*C1DM~p?w&G`_=khVT;Sg~DG7&r0&o}spk`T^YM3Nz8#@P{Ma^PhR@XL|fUiUD(
z#$9?}w>moQtit1&y)r`uR{<@M^n}?whZpc_$L$#p)?G-deR|6#kl8#9OMlLQ(|r&6
z;Y`9Jj>p#5IisJgVRf2;2dDkJ0qHZB+%pF@sR0M@)DI+!jVmHdbs+xd7|(%=!G1bb
zf%iymvp>QAg9PIF^5=-xWha$g*K_eor0MewNTx{<YDD;eyNzEXHkivM#bZLMs;4K%
zZPnI`V~d)CLR_svgG4fhFe!;n9d|zYE1vz$nz1hVoL;P_t}~eASyO|DA)eIU6D@b+
zA@9Z3!&g1)tol5b_rSEL_3G^Rl3-`p(S`yHI0yu~Y2SLE5+iOut!gKyqB?Y5SGz<b
z1}`!F^J9Bb(W(yBoXlLJk84e)2}Vc|)0F&i)2o_Zqo#-J_sny=?9>;Kh!}R}VOe1w
z)f*SjM$A%K8FI=5a#dg5;hR_nB?RZROhEm%d1XM>c<D(!O6a$O^o=i;($MJrMf)X3
z`9A|#<ub<yV*3m8mh&0L!k%8a(TCh6$zh72{bB~H^R5>1!6V9oAC&epNcTN3(Ehdq
zvj><bBuUG2gd9tq!^l#PUNOC84uH(>t-^6@V5bnbJw&-MIE>auFh_vtk%FN9O?_H?
ze5~kjz*LM`A*$!R6STihyX$B=yH%tU`6vfjG*WZ@J~zaK=0olw@=h`LC{DD5WAfsA
z?(Y7<!Pt5879-+X0qdUd)e)E=Ec1aoEWUWb!B(U>^hGHK$+FLU5bl7O0_aT4FTq-@
z41MV67mg2PzQKrg7;^yapPpkOzdXE%FJ8jIL|S}uvMY~m*&{5tzY~>=H0t&d#sB0_
zP=o*J871zarZ|j7dK@J#`bP`?-^IeDxk0>A+jlHq;3cL1!A~Mag-o=E35YLN{vQwj
z^Gbl25S&|V&#}pW*Ylq)quxP=q>wN8FWgctEM&MJ`USE7{P=gU|7ycu2vw<%09Gbl
z;=dXeLW4yIqgWRPFG->KUt0Z%zqFzzvPF^R{#P5Q<X@zxKu859)&HdxT=+{XF|`Hm
ze>DsyPAv5w<KQxNi~p;hFRcMzT7{J6*#5V1vHvj+ovTyse`|&O((0dEXz{;|BS!l&
zj@;fe?SE^9{n82{EIt3#Gk*WG?EgCAe}0eu>xln3p#E=0{NK{_|H+I9f{s6eh#OiV
znQKxfzj&)ZaL}d>AAjR)6R#s_N3A6}%)4!#gLFEqo8s{(H>hTFZXOFtm20JT`70)+
z!K``At_xe!m07&hf$ZsHl2x}UA?i-FIj!zKA<*|4X{y@=n%yeF^oPhwsw;LZabCkO
z-t4FN>6niJnrpbdo~(><Xz>K4rAO6~`uLYHTT^wK3wNE*ZulApF^nXfOZ)6}hI#>)
ztA0)CK%RL1-rop;a4am47!oTQc^iO>C(L#q1LIb1zhnG1e^Y6sCzy$tLRGyl1@K1h
z2l=|@oQh1}8b7biEv!Wgj8Srebg8c&QXdCzvw4Hsn|~F4h&PC_^)KH}{lRo8?Nsso
zeN%B(#RRB^J-*hzeq2RNnxc?Ko2?^t^IwniX|=Xvj3dGlWH7N|htL(o3MrCRDH;Ei
z>$15zSFkshbSis0n_;c1a&5LJR)C91QpBWBXh#r;R1m?EQ~0)3Q=(U~MjAt?!vLym
zG>5A&a@)9leR^VhK502MR@2_8@qqNa=`|_`9pAVW<{?}LtE&%s`Izkfa5P!f4u=az
zrf%0G_@aPwcZsj!XF=SMzCZU*v>QvjE&JWfij|=F7wEX$^uNc9x8K7sJJ<@#hZ~Cz
z$^J5{7hCP8VBNLQjB8f1ozjf-bQy7T<)^1hZiVCAdi{6eHlp~zPlIg6RCRV)KzeR)
zlLkl`nQX<%=?Xn=SLOZ=wzcxfNn<O(kt=tmPBjb~gQ)LvUL7yw<l{v(cWxMGLgA;1
zN$RG^nj#E`tsO79VTw~c)bkyaeq352yRim&)#bT%vpqIEg3WzSwV=4i9CN@yDUL?S
zCzUxE85`V)r4b0~?w-BBUvD-!J{ujE*H!M++;I_%^o>nZh=-@$RQCN~0q{JYeFJ`%
z?cF@M$O6W5{%w-QI*Ge6oP>V9o$w-Or|*3e$oS0U4O$B(DJmB!L@vy5#EiXgHETYz
zrAfHIf0d7m?=v#?JrNs)mmfMcN`OZsuKsBU*pdf-tKAc)4^v1gDq|fCcWanmB=cIM
zHI{uBPGBUXRlqD#U>-#%U<F|qC8uHCuGHP9TD9oO#Q*IW`A#?PC%8g!CgPD(<u1wN
z**@tq%!{}GY=mc`yepQL!NtAZ!M39hc`z82(s505b4XS_DmeB}TF0$kdf&0NaAS#<
z0l8u>GG05IMOEd@;U>rHlaV(g+T9V<X?w0q{Dq|lPdsJCltP*Nta^efGq1jF8{__K
z@-U{+oec?oiWdQ@*8<WNowH+CwC`{IQ5(5gxTWAD|9<HbV{O9_w?*Ws>DzH>;nCt8
zb(ddoU8Ua6Aa<a?`&cPWw;#>W9hV+0amQ~PUMo4Wcx)z-7l%nml{3*UcERK&Psm&6
z`b`8GOFDe4^-qs0^`9jnY3I`O796UT>B_%rl(O^1S&n`D0<@yG<6W?q%k<AKqY*dg
zL$b||eL8sfF^7~`GRksQ4Q*|ptkq7H`4#dw`JFmi=TJUuT!urc=p?4-c5tIJ^6;AA
z(bSQj&+Dn6oAK<!ed);IDE^<BEx%kl9~3Mu&mGM^z+PWvCNVvBo+Qdn%)eP)NXy(`
zOL&mp%eP{<i-iQ`EGl|mpxUY;5u<I)_%3wrS0&c_IoA`TJt%3WvM6`yx@XD#nM|<9
z!;-PCYqD+bP-!dz-u!9Ee4diSCflNGgZdgXgFi}sx^V6kd1-yv8yCH{zDRVp%us8_
z^C8KaH{MvU@N$_UMO+w@y^kMkgh-3<hUH9}a;ESKH{iCd6&9D96s}l!5<$-O@6A3B
zkN?0q9RCU{E5r<1`wKXoKa9@>9!L~Qd3px$dEX(eL`WxUsjj3D*l`5lMMue`XgQ_`
z8Fl%710b~7d+yMP=0el?xY>Rbo!OP4dP#-7`5u&j3RWU0xap_X6E2?`o&b&phH-Cs
zZ|u)|{#e&lp;Ff7<S)4CSuL-{Yz^~+(g`Ngt8yOs&g`q>X?h#G`t8IP@e91XAR@`7
zT)3EAv>cWJ_Pu&Af73|Ka~~Pfaz2Jmf;s=|v!AjZ{4%=*1;!%I$=^!|QY!BURn|Q_
zKAW(InZ`32iN<wll}VIU;1g1|{8GEsT7Uih0>>9g)AF1c1FeGJ?8k<Ual*VR_*E!}
z4@M{~=1pwQ6;O6K(jI7*UhJ-VpVj7m)131Ye7XKporc8c7FK$S$>g&Jk8n%0fUzcv
z7QLR$InN`%v}34`KuOQ%n!w*S1ph>|{2+%<FFhMEo-WfiFINTVKkN?rG`zV?JNV`;
z(SorWe;&|6#Ajc@gf8sYm`g{?Bzn^GYeD=rHO`ld92eNRPbX%E+!eAVmE$FT=XNRd
z$7si1kHD3MIa(N%f?v&XTjNjbxLZ{X?WsoY{#K`BN;$E&r@D{fk%Bw}U)xyQor<GJ
zyhZqk_EzWb+^r8kFxu&re$Gf_)1i53mJlR@7K?k(y~UR?R%XK82^*A0@Va_6ypm5b
z_Jxf`h!1qg_`|%F%@5#PjXVz5aG3~v;6dIyyMFNWsJ6S$NC+m7mjoRSPD?3>{T-HL
z0?;31p16_qn9X<~{jF!l$Mbeps$=!to+zi7f`kllR-H$>jbExX%ZJSNDi$9aWnXbJ
zQJ!1y)5Y&9eLG|zmwQ&WaFk!-<7P4muy37+pS2B&i2t*1wO5_eal{4}7-DVl^0U?K
zEwqT3cD}91w<`t}7C^8#+Kle%MwK)E3fuB%pv+zLa^8E$tl!%117j^1`a8hHX7224
zYe<(a__`J1<@=?i;F@KY&6PD>0;lfF`vh?;K^7^^mGTi0pH4g>8dbP~a!I35p8KQ1
zKI$l2{q7B@uc0g+l88pOCI8*^9BgID6E@xBWxS_#4&}LF4u~o^Otmkgv31h+cq=bI
z)qm36%+vFFw(`ZGp10Rp_DZlN<P|-jI7WkWgp;xwkX|K>bB>8pND&`Rm?_ER@ft)-
zd~`BmOe#kk%*9hc$TT<<qbR7&X_p_hRCy_b&76c~Yilcnl3n275?6_CY#s|q_F-YF
z8nLTon0t?sCW`ZFUccmwBK!%ti2^J4XB6N%Z96n+ktSPA6&Wn}b4XLuo7emqGLai>
zkFLO9a}Me@T;2u7#9|I{G=isd)uSBI?53VtKu@F$^nfWRW;&$UbL|vhFPicDMNo<E
z$!{BJWaVmuYx3!6dy4s3p>Mo_-XtWGNYLRY>R6;yB1*}0LS6J28c5Q|om1+|2}p{(
z?^V1<!HIh3EOUH>OjE`bJBM+fC|bO8q^ko*Wg`5uh2DpjO$8BU$^USNKmSTD4+69Q
zH7A<pX`LQwK1=F#21sFK--CiRE$$QvyFVRkNtX2C*hY@~XeKF3urVsdYncEt<gXKu
zypf}F+M4%axIV0}i#ED$AoEdhd$mtH)p#^3aL%`J*~QTmxO|K^A+ks-_sW3gGc>Cf
zy;y$2^KhS1H%5l02O5Sh9i=a<+#vD7)PE$g0xd#$mO^eGbNPn~2FNPO4n{u;AUu=|
z{lK;E(!O`qXSruH^8I*aOQ!ikP4s&jJy*6F>9Q^2SO|`!>*Ivcy)Nj#WZSN%xrmm(
zQ780|*n{rRc#k`Ve%HP{2-RRSA`L??aRLlJ*%P@Cv+1E;pRT|zLoeCU%!gGmZYfwP
z#c9`0^3J1s|61c(7r!T{adz>$^OzBJ*|H3KsjTOzDput@Fce%~a#8O<TR1t4hg`Ud
z9dUUSrQb*5zdn;&4_{mlH<OoKaE|wIjBtFn*`(wwuJ~nmrj<Q`nm8A#AsN!wFi9A*
z#8UU*b&_p)lqord`UYCm1drZcN`K<@Q6u6FEab%i2vy4opzQd*b>EQ<Msc*N7Tg+p
z`yCLRm2O(`P@@pj4ygrUp=XS9Niqm)pY!gV3nD8JCZE|dBw$%`q_J%V#WIb#FXu8z
zE<|v3ASOFQ{JW2O1u2erxv7@DqIc+E?6H9UOsuGcm}3%zMfuw8EXEV!Y78W*iT%|0
z*P__Ofra9ZrWr!N9H*Da^@HGNDxbrfjYq2^D?T?4sNbaFO5JQSS!<gn8_7&~>>*jj
zqx1|ZNPO)wx+S3Q$58S3xe=qp$bcqnIOm<pUUYK70paCBt5ID2{5R+L7-ofW)+yG6
z?98s$)Qsn6r<Q`>AHSZfI(|b9r!sP)$@JVDM73f`K~pM_{Xj;XR=11q?)5*H;>DM`
zT05v4h1`C32r}1-XCY_`d^aePNdGi1+}iXM+q7J#tE620({AK)IF9>0s7{+UZAbcM
zrnXUKMMv-`IWymf=TL>~HF%5+WF@cBPZ{)@UubxI@3fapmbrr;j*OH3jo{K&XrnU4
zY;0#G#_?*3UYfq4YkXGXVnmB}Wrv1qNw14mGU5%%;5$sn7E3cb(33gsleW*$)=uEh
z;hIrq+UxZ1E$u$HT|DO5uz{W%gdi1iox05;+x_d&%m{L-^-2e?PJ`yRw;t82A-nHX
zxLBu-*4pXb9KhC!s#O}KCg|tAM6b#Th!?qf6y%J?ZqV0_?|k+CP-gH=65Ky&L#a=6
z4ps_y8iU#};~VAqV(>OJbnb1rcW3rKm!t}|^9-)~oOYT^@MrIjAMU`=C5hQM8}@9V
z&IX}JyJt#6pKnqEZ88Nsxv}jSS1W^C*#we`a<94I=?t=4KL78hJ%pf2C2}l|2BLB8
zNV|6%zu}c7-Kq`R&kA?h=mxd4rhd?;NCmwwDKHJLXw-RgE|6y{yz90rUfmdjRJv6o
zMM_9(v%Jo%Uvq_~9aVUbSDYwry+)3E*;hOJuP2u2-(vM^WAQi2A>(n1(Ru%IU5`TR
zN_@XPI@Sn@HJzX%HMiX9Y}|y))}bw|_!S!l$(h4$T0AuLIaH~2S~3>ls?kg`x;o6f
z&*Mjwo2fvbiKYXx8$OYB8w<XmAMu`iD9TPaj(ijXk^<S?KN&*`1WNNCi|U&9QY*x>
zI7lIj9@__f|DaVdDrAOuUL<0nC7oWH*o`UQ8HA$&ZdL-u_lGJFVu6CcXwEB=sA#w1
zl9IIK*iZ~nkwqqZAabU<<%+*<dhG2^VG4qcL#YDo|4E|-U_|=YH&;9F@;AIF^DyE1
zL%ZnSxL1fxCoZzL8>^{guV=c%k@sxA{QLva(sumb(=N6uDt8s+FA4Fw7%g!4W4o!?
zx5V0x8lYUrb!?PnuTn?ulfIq?q~Q`<@&i0H4rPUvs?jdgpkOA?2TRdo$=d2|J*(fP
zEN(<eSvDF)&t5y~oREeOb1i@;qzWcg#hb-+(oOpKxWmTl#<>Xe5i7)R5%_9${c>*9
zMiqoG@&s-a{55eHLiD&f(xOHP<2f<4I6zDtLfSY)VZ2P^o%<v=t3ovUm@*4<c6z{t
zhWRu5oktov|GZQZJFlD2N7NULD$>ziNXy`l2(#ar1!ICE>h33foAS`^V1aS3H93~+
z8j#WBmn(a+p^*N3XfhVW=?~3%{lI9hJfEP9tQT4A)UNJFz)ENb%WRrnND8DP-KD{v
zzuo0aMtGiw#Q~f1lRxk2>?RulvN8;snHvRPi<hR|5w%7LH*EROV9?DuQqc{rl(ms!
z2GtGpq>ZaT;$rXYZIy#MDfkH3^F|)lb!Jg<f35ZAV{`Jj0AlnOo#O~R>vp>2-J8MT
z<D=2(KFSf|L+^D33<*w3udFw8GUb*HSJ?9Y>%I>CBfn;?ZWny}!??;6-i)ROPN!P4
zsISuz)QX=YbEYaMTa?irYhhOjPTq2Myh!Z+Se57ePAwU4T&(_Q+dh;>o+AeIU)ola
z3Gq5=le#yILdReC$SZy<8@$Js&SA3CrJ|ac)&UsjwW;4Lg4I2e$~Xl!lF|;x?>!L6
zi|ux~`APIPsW(ZitXw_&6wx8DPQ1LuIPe4ieQOU)Rbuzxn{Ul*=?10VO&!Hq&0z(k
zB3<}#2AF&M4Cm|%odTTBNe7xcdCaP;ZTM)X)jJ~E1Q~%$KB`@;_hgQuFNPQlVUI-^
z21ah7Z;kuxsmk|O$;-BL1%{y4d4Z5*0$QkAoEv(n15t`<6ylAB&dkJEnyoZPjJekE
zgSyDHoBoX$f4*~$VogCI!cl{Z@Fi83jQvxs597W7Q?bE7DF)U@V6i0xDgL@yJ~+d~
znAxlZxwO)sf&QIV9OYp8InlT9wM&$0vDX#wz6LP?ReSCRop%{OW_>U9P0$PwX=tJO
z+e9_oEKyg(Ud)Z*qG{QQ<yA*$11CQ`x)Y?$-C=?t$g2#eA;x)e40z@##*v=`8yYhb
z-+&VzX<vqii#+=Gugy@e8_H(qeT_+{)mafNg!^Eww}LY<cx|5(+V6V?)R%7m7FWaV
z;=*3-izNWo*w5vP*Z6Z$>f0D(K9cH1YsnYJM618rU|SiI(G_4|8}weY#ja7@w2#s4
zL!ul}A`oN#b{C#AcWHpU)CH5N_<<P>KE>L%$L!%rpNQRhp}_TX+8-t(fd@$#NjmoN
z^)H(7ztQ8!!iKI+A*N-KNU>dV@hc>EW#l+eF%~u$TF2$Ox_~DJ=~rBdXpbc6izJC|
z@Y%PjQr#FncY@=a@#j&Jh<P6e1E(e;##4Ud;{fG%)nhAD5u)hDXiay?k^sL8(fvqH
ze<O_5g|1{ru4jT+MZ=rx;97NqfW>M*X%t0NJf0a~Qm56X&uJ_6{6=_}BQ5Mzj*QwU
z5df|9%$n^IGKUThO@LDJaKr;{Pu@efa)q)xWbabqt{IW=>bh2T=ZSy;g-DYDWp|4s
zi0d7iGFn}K_q?1SD;QwuZOtKaMA%~%qCDp=NDlXvyGDKNJyb8QKP&{mP9lj}A{SSj
zt-!aZOYP7E50H!LvJ^{Dy!49P=S_kL?0emTqLZ0O5Mx@q5V~EDggc0N<R{7S)1iDx
zy2XIc%M62j-b5-|R>nBrBkrTD>U(<bR<%n(0JJ*sUW&fjM8vW$wwtiN=v)fK2Jui)
zN(%^Hz=w<Ls&%sv`J9rOl!Hd=n&r=~rD6YGU5ayVXGt6>H{4gBdyE80R-9FQaR^k=
z%)A56hC7J@`i?Fow+3i^vewxRXdKOASd>hMqmmA-Piy<c9C}6c>EA{}vU^ZEUM4<N
z`fPT9AyUmQuk+8NDZBg~&PA|TsXvm4>oJCK`bMye)Ka^X*tJaHc{1%Hzt+j*u-sRo
zp*hE2R{U(%ynIZU(lCL~st_il$fz0Jh7WC>*UV?_7$7Gjrof>$7F0^%i7TU<wSK0t
zd6QGrO&E?M4Z>R$T@B5r_x-uUyHq~Zzt43AJEhjWtQrK5TI^o4amm{r=pa+`+28m&
zrmV>_h)<C;ItVt8^TS&S@665I-zs(acA<c6+bZ+I-BDACtd9zBYirJO*}PrCrUO5O
z3KBVU@nZ<QWPM$?_3E(pJ7FG9n?X#;)tI^np^Nm~iOLjP;&P`+*xsjkgBHCN{N9eK
z+UYD-LWEv<y}P+0H$6S?jEU!aGk<}_`rI67GO?k|aQV>Z*t;1pB2BVFO*pcz&wrz^
zu~$u?VE0qttAs{M0m~ia;$8Kx${1#^!(KVb_9;5HlbYaBpF#`=e&E@CL}kRd>Lrsy
zxw@88*%9eG4n)a#j*M*NWVkLOX<vBe3&sxULdzWd^Rs!DG*-svm#J<UV~tepB!JeR
zHQGtY=%8=|q!sB;_>8Yi>Rbf41OEH=ylNB|o32EwasWyh<h1NmvhK_Wiwa9)8Of&h
z3q1*M^{PMKKCk`u$6?7Rd7Y~50bGoZkEwL~#9=g?2#!{58*O-0ZSRBEs!=GSJjv|c
z$-t-%)q0{&SFe*87d|=)Eyl=<fXsD>{>W18R;%4mP}#q&cqX|PIort&Y%~ngnvPYn
zN9&J3{8PUWqVwx%EML`w0fixcFHk-9g6+l?g1I^!T5k{UW-HiX5U@qei*sZW1+eo2
z(R)*bR9I^k9?>sGU*zukEdu0Zo}f&;a>f)7Va=K~qp%ed4!6}RTVE_&NGC!#)H5y(
z^O+<{_ww_(-BI+xTQik;vK^NAQ2s_7^`3LK+g`uR>nwC1!JKG;lxVb~cPvCl%`mXY
zGze-9-tj54zk1aA43i7v`c#s%VqRK##nI<Lq@wQ!9>a?VGv`yxHK&j)kIBqyEz$m6
zh8JOot5Z8kKuV!vik{p%IVJdYJU%+ioCl(h2zS_Od8tl@_oSE`njF1lO`oLvjj(jz
z$>RAD(*=PS|6Gt1+{-3RbX7bIvj>x(sGgBoxe`t0w*ZeHhWWS6$7a`O$sNj_^RcTQ
zWqVP9Z&SrA&$7p-5^*y@zRvs?{iUL+LLp`^*0F;x@29i>KqZ5pEISfr<oo`qx(J3m
z>JL<elAp_@ihf6L7?9y(_m>K7XbSbVNcK=+Tb~3T;unGO`CykYkFHMuSR4oI0b5|V
zFi^ovTVjq2cJfm%rYc0$r)@pGbAXm7%&B2L(Zm0H0dPa98Rfm{m4Cav_sUyQ*lH?V
z$iu7hFkRRD6YpBv=4hN|`loFyk<p@=*StXk6j)ie!Uc|3l*i*LbMhOe45Gq-@60RL
zZ<|tWXhAZ&(cU5I*2s(CaQs(EJ*3CKtLN??E_FmMn%{G1^oW32+UcIz7UOsh(?sV$
zaJu@%x%EX#0{VHHeRu6i9~7=d{Tl<gvcrKE#54j<wBNfQqmeGYEz=B@MK>G#KnROX
ze}^hjU$$qp0pc8s%}TBI<dKr5H_Sbk@U#gh{AcNF)tlF29BhaUjB3qVs2h1m*{A)3
zE#f;mHMI_73CAk!SY{lXW+0S{p>gp#>w%@m!%@3R%;U>=y+8z;YGlHLPgugV#{Te(
zo?a}6uJ2E0-U%<3NJmg(o6b#T7El;?5xW{~dMpoFZGZZ}=cb+Pk^ScWnING#r2FJZ
zZCw%WPc^IgQuJvfAdBGsI|3x%Hf?_wull~eKuAA$GpjfK%3@mYHo_lF#IZ_t+O;?X
zeQ8NMtjHNxd`__!eqboOzKr4Sdg%xK_RD-+YO_@s*{p0`B&8S~i&)_TO9-EFER&`Z
zn@fw2*G|)o@FM{NVDxL)NdG!n&71_@P<4MTRUGK{mDDsae{H`F|14568HL1x3-NrW
z9J}U_c<QRU#7E}C=M5h5M4e<n1uVXGKJGidxge#<OoPWb2Ym^@TGy2MU|AAeZz&rm
zSfG!09tl}y!-RY`@wEo@IW#xvGqBY92PL*-XML~Zak_6ElJ?M0(U=Q4QP>9y1&4YU
z`wp3-jca$4ThE}cFQ0a;gqp~ykt;JqTnDXD6&#MZ&9%uDihd#7gF2g*N6Q-$H)J2M
zILIb!{25QE&`7j@-)vrxBnPT2Ule0+?Ds<_TNJHG<VvZ)Vb{lv;R@>lV|Utz#k0Wf
zjG+DI2JuPR!E<WJbZcwZck<l92JTayr8D5gYrAznWUNbQLYShUIOTvTj*XF;7&@gz
zgo9wbEEk5MYH;<1yAZ9{3#YpN`tc`i8x91thH?(GjgER~qO%&YY{*1C-26sBWk{>a
zKAN|ly34omOYJxZ1Zj1ekx>1D2w;vo25g!a`rS+%XU$K&QxD3EKf~B7-fqQ4Mv+ul
z-m-?IhEW86lm1BwbY}vK<c275)w~|NlDPmEOb#z_9G$Jk8*cl@f15iKGOu{IayU|<
z-^^cza|b^Py}foN4T@dD4}H%NaIdHA0)|$#%oVN_<;>O9ZCQPwsS059Lt_7!nYb^A
za5Y~OuqK6eliy|Pch1*L^Je&U)o(t%frvZ3gx0?nu;i`Dw+ed29HSSSCPCSg8t{9$
zdkK$9Pdd3go+wOjj~Dt5bqi+Rh}co{c7z#}nV<velXH&>mUAmXL&ihnC%x3%`+UJ%
z3u!C?R)eaz=(fzLU$xUB?cVa6qy+{gA*p?C#Avu{^o6zAi<#R*cidQo3h{ah?}_WN
zwXhhcbRgDc8fAMs`0*=bx&03?Cpx{AnV<BJL_R;>zsV&kAzLO^%M9DrxnC9sdNcNg
z(01q{WM8x2QB^NkKNw%To)gx$?YVB&OA~BXQ<{~RVWl(GZT*2iy^M(vZfj;w5<*zt
z+N?a3d03n<#P@_V8qAT=1uGtP1aRbs!$)n?S&k)OqnWrWGG3I%QyIWFL1B>7N)`kD
zz-;d9&v3CaPD&Dd&!^8sv<oR4kg<+cS>$(D;eoILD!+)~1pbJ&Ct}s06>iGcsf()P
zcO3p@d9Whr4U0Fq!T(`NBSX!+>TUIF0Ah~X+n~L0hU4=wb=w|l;ZFUczj*rXtp&wk
z0w=<Ccd`2}i|As{SMF>x0KG@YR!+aY98lI+&@I!mPp@|)zW{J+KjeLLJr}0AthzV9
zv1nq$G9pk{R3$?_ACz@nAQH}SL_WGbV*L^7#tE5_aq4t1`UWY7{sQ|<cC`05S1F%u
z`&|YuhSetA=UHA80koNaiiQb*#*nViC)LwS3m9c_YbcuWZAUE!T!z9@&|i*NMyY$x
zfbfYo&Xi24qWdH7f8&HI{CNl>H$Bs*-|<^IGEe$#_KBErD77F>xlWLu)b^R6W6f%M
zn03@5?RdO>&CEsxq}7A_2g`*hT;D6tZrc;_BXqf-Ly_6gMWG{kIsq$zzWI<L4y9z0
zGfN7YXB2$-NL(N1Up!%|t$okS#4^Y<Jf7*rqh^vW(hTvsi%MiD)5EER$!-T)O#8#k
z!=qcoT*zER-^CkWiPGTpNX}BpKKnSEaZ8`AMo!VA#Gt!FvmGzp{Be7%f1Y{SK~kx_
zKbP{_(VP%Im*j*lu61cEhhdUprtLgPq7%oWw#ZJUDKc<~1K;&5kvg{jdr%jAKY6fA
z?T-Tvtqc@@RF|jmi$tPwaj%z6f1nIf*cnRkIxVNrUQ$T-@;Ap-b$A5zOOb1wPpg|K
z)7(k$sfMgczjFbCS9$}2l*gZV3{5U)&nJ_nSB5_cntLu{R@JK?y0~20X8+#s)zaEl
zhCE#Tsy~@^i~DXsv3If#1(Cv}EKUBWVX!fXj$*T~X?1ZP2_uKSIJ?3rh6pH`d^qX9
zVbslVtXwhTK0rw8X5%TOYq7k~%?5XQ3RGx?%1~@=^v%Wq^W6#BwsY!6#z^6do%&$$
z(wlM}hQpoY>T25)etw%0beALSn16nq;%epW<2zDYb}|jrHWjCoHbBg&?q+ui-ckD3
z6Hb&;_9*ORXY+48*t5`M34vJid(#fR+(Tz@LY!gjO&t8_r5ACoAcODkDMi}Y9K}x6
zslep*c76>HM@|J>WD~RswhqMz;TeW-k<WOHq@bJ&w-3~81B#KaEYrbFTPoBILL^t~
zwemOf{fhodE@_WGluhR!?o)&DkZWqq9+^pWVUV!mnONKuo|!>&L}XYFsmCQf0$u>Q
z;TwZ^!+XGF%d{tIBT|iWb0H!xX1!dfKFi=z78+7~QfACYuYCLpRsB|Y5PG0ZBVswK
zooM^-OLLY+SNwNbiWjjitaV5Wf-pQP^yC3DQ)loo%Orn{$n<Q9vPXxx#yKw_%XH_=
zf}&9S$Ko9?=tEO4_ENs{7iQ6qQ4>wp{iY6zWq&kEQ+D2cmkQ<<xKdfup6E8}Jm4kF
z1<P1gI*s;u{`)6mh2`^D5i{5uR3d|{P;O`N9;OeEIo1RVft_7)4uQ9XZiolg#%|eB
ztAG(<L*IpO|5nf<Z+Dze^Qb~<V`rrA36n|n_%O%2h;Oi$`a_(`ZhFM^d8n!&C1*h=
zIb_SKaZ(<_KcYU%ToaBl4*So%l!>Ygx&Bcg?Td2yZ_Pe|iu`ii%YwD_Kf@~u0#JA!
z51pGTYb%U)9A|gShFXW4xfWEE@i`;i19B`L9T+Bg+~w>SSe*YxQr!572eXE+rdpW6
zB4e2j7*a-;*QXiY-zB!dDs-0j!Ko2esAhb5q5t$eMo{lOsH;LqJz@fBHXnye+36x3
znXs-6&;fXp5dIB3@`QUm*b55UsU%u`D~<p;F5iYr>E}zI&sdgJSbL=|HkT0Z?+S*s
zGfJ5a+pY_^v!nC8Rm|sY^rWX<p|V2*g<~41-&S2zmq!t`qw8>xLfDm9<N*^(9PET_
zSuj8d@`AED*F)CftI!AgO^>+mmDO7a^I5bB#;mZ8aG4aZvtZI87NXD-`W+O6uJ0E(
z-awhl+a<Tv%^!HI#D)Eztu^m=I+pl>JW;~~A#)pRR`|{4uikA%=<RtQui28us`Fr6
zVLl6~P!#%_iGf<%yKD*u&YATCK>@h2!AP=eXG^cOuU><z{=Bt2>yW9H?#)Q6$tdI>
z93I~l8rkBmlJT4vnPrc&jm#xc2qk!e*6V{Pg*Ve_obUq|$utaV3c-#oScJ^P?T(om
zGoFZsb6?Y<fU&@}S1iRkrk-q1Zv4@JB%B#r0oKbc&{6<}QuKoriO=m`yKLpTp#Iit
z*?IS1kOwhl*YR^kLQZ&1Ya_n8reH$+4Gg)-M#HPz%QAALZzamN;++!>I{b2)2`UEM
zfv?G2oJTY$6*S<h)35vduqmkzB%%`RKOrPQ>Rby*(nDCCJxK0nrubWfjFS-MU9IF;
z>#i>C&afi}BB<C_>iA<hVR1zLv9^l^vTFYzg)#UoDgV+ja6Ponm^baaH5Q7YkUn=8
zF>W2n!fBCzhzclpXoPjw;@snpF_>#Ashmi-Q1U9O@b4#WjQyUTXwnZ@(2B1m661K0
zFqB~#2=6CMow{B;$(d@a&OPU9d`}+GA7u46{&(u1Hk|9$)7EpE@VFNQs*BW|c~Knj
zN~6gc?Gz)-a;bi^7}~fYSvGXa?<zZvU71?j&}*sLO@OR451^uwAd>d4Mnw>26|Jgs
zT)$4Dm%B0>j!r@JMy8_w#DqkDdWr7tTLAqq3c-l-Jv=x=02y*EN*bEzHw1xvstge>
z?rXg9DE*iSwyZ~7X-|x~-wIlO-#um8jZrx=1rPDVF&GJjy7d{_gC;lPU+BWRM3XJ>
zS*4&{Gk>}x3<K>{m9xhscspsW7%4g5_c)w<uziZwr0WDjf8Fz%&?X9r&l=5`wT*fM
z;nt5^a691Nw}(031?9Ln(%?AN-Oc}U1Df>xSc(MA0L4Y#W+fz2kW}E8SU6!<uSE+b
z1)1a<nR$;77G#RU{EB=0-g>{0&EpoJZ=<d}b5=;A#g^|1@-!Rwh~#CcwNLi-Z-Jre
z=|%fq@YV648c<Ql&?5yW7UXqkV8UhYu`gnEsBo8DKSL#dz|_>MM_M{Ax)t=4GldPi
zhHr>`B^DPsX4h{G=wdwVx(Vn29V7hA7Ql=yS1~>W@;gPq0k~_tv_Sj)ea_)VXtK?W
zkXr9sFD&DUBUj<yX63O=7t_&9j%0Fras{<?PxF70WQB>m{XHM?-@MJXwzlIw2zh3m
z({Naq<Paq!nl>Kuk?y3b$OzU#{1c+fHv7POc!?CGI~)#McawixI5$cCl~gPI_6=|<
zEhhYvR2#Okfk=dnunO0?p<TYnUEW8X*ZSz+ID_VRVPVPU`em=UUb15KXI7}Qy^>G)
zK@EQ313wkLC}FTuEJXGTXQAHL;W!u6<C^dd<mGsio8Bm`IYR`<O!Wd1kVf2{esc`=
zFKQnKEi)(#OEBXN4nsM#{hv*UTH>lKP{gUojE+M_?WzB!J0Y~2>AD7AG!1K<$SQlZ
zoTyh_G2^VAgeom3Bz!RXpkE>o7s=-I2eE;AzjJ_(5>)<DheoY|KE=Hb-VuyC*Cyo3
z$52(MKAkoLQDF6OVk#hT(}=EHVwvA=h-3W>g?^{q37?Lw(v@@8ZfL-tV<h>`Z-GN(
z{A|F`8*#H6wjy6c!SqZun(Snvz<y!_Eb2h#KgVv4;o2zxA`JM?WvwrKh>yA$kwa~?
z(?1aj^ST$x-RUEeNFkE#okaLM?)Iw;o7l#lLyK?WmG*PpP&@yWE?qijcswiTW(eEi
zaP`K%7Nd1d`!qn=DHPl*>Ivwj3r-Izv;7!WZ_U3p-se~f3X0Z`2CSLCnX)51HS9Kr
zVr}PE8Bx;1GU*x78Zr|xiT4Cx;gI3~D}U?sCf!?EbS)s<KR6-@>-7v(aL_qM3-$(f
z&LT?dE<^u&&Sn4X6+?xsFUEoIEkO9h74u04srCXerzr4G{y9QFrCTXJ*DwE_dl)S`
zs$tZOfVD5?QD#%>LO9-y36{RO*~~djsv>hEa2qW62!cBE=Y=(&A!W-!4>NEC&vs0Q
ze)?y{@QF(V6{zWjr6QYfz|!N7gGP0r&F31s*M9#BYgB6{6!$-+(!m3n)<u|_Dvo~0
z+uOB^P!ACP^JbwI=;ZEJp8rY9Wmn#evXy{>%CN{|dtgDeu>K|j{UYygFyRuMEGm=r
zgA$E^{V;=yLmkQxHHPe**~n?lWIeP)m$vhe27;H6>)fCb9Vj8S*2^kL@u920jb8#R
zqL6f6hCXJ=z>g=+ex!j1lg@@6KRO*?u_yEo*@h-XYa=j(7b*u>ioK&o_#q%ow1#hw
zIp8<yxd&{(`=oEk!z{+nI6@o@W4NGDTAt<+Zs~-k1bW0A02s}-fOqVv&^=`s2jaXt
zST$|FFb&sbVIQ-CrO^lcowqzE_LAGzSzL7j3e*ZIiLOK`0Osc!=yurH($hRpF+`ep
zp?TWlaxEtst#z7%yL`l-_#b5mpk^ATam;0r-;b__!8>tt>^F9McJYD`2{8u@kkIex
zq{)X0^eZ1&Uw^0a63fL$)K6;XzO5G<pGia$9m)c69?x(Qg4*B*(g`tEtcSP^r(>%&
zLuqb!u*LUNo_dPF2+5c(2E7ngtELpVNVoL7F0batyt&vr`QXt3vP0nD9mZQ$P*NIn
z><Uru-JTxycKNmtrZy3Xr5L<iIK8X(Rd{b(iuA#C18im4Tj*dI>zP3f1(szJvBLyw
zf~ZR85o3<SDK$<?gVVdW((~YT5ghN(JC#o-iJz$0#Li=ElC2?O^INE$ggy7SoIhT}
zzSwBl(5r)<TTU&P<U(CR1rU%S#b5iHAFE1)FwwKharX+ti$oV5zkQvqOjK<N3tPe7
zU7zsZ=8@}rlDih6LVVt@_t*BVh;8TvClz_(4yR?a6G~_|0XuLOl*1j8jCV$(u?}H#
zY>5s2Z}pdVu&YA_l1shUwsaTAU38N{N7&PZ%XcnE-C9}abN)$%Ld|x1Apk3e%m>$=
zkM*&XJ<MAl!rzBvO4;5<efN}iM@N>wbXXWEk-EO^p$&##7e0UQ%{%TLIUH9BddWh6
zWjD&0a*ObCtvP>JyiTw$T&F2Y^>a1kgA*$jGaK(nBz;m^@SONrTIPGRyjF*r=i#(d
z^WhDWh~$l`<8ztv^UYRt8Be8fG}dc`W7OY>k>Qq*^@2rSgQ+3C?N$|FoHv9I88DDA
z%uDlA2-FL$&Yv8+wSN2AcV4X~)r7&++g+rgh*#=xlcF7iB&Khw?Z1w!)$fCyNsG$5
z*nUeay@Fyh$b`&?&_2TW%+fJImRH1;5nQ<RZzvuj>(L46m#?mDlr-<zw|A@Lz*7w^
zKrFLsJq11$1{o<T(=(YPVFUBhwp~`UK&y5NQJm>J-(U_&80`%)n~4o^tZfuCyP4qg
z=eZ7?`>(}D-8dYPgKu!6eW)Ve(u9j?sgX;Ge|bZdS1ujzDA`on+#59dC@ffN(W21(
zDl@ktD5W0J0E2-vhKEBvAU7r^Cjsk21igQDaX)5zIocJ6#1>fH2J>dsKMH(iFMVaO
z1NQcJppb?j4ext*QOP>$!Jt=p2q<R_7^xq=3w}Gtu~?(`&j1JQzA-qWejwMoG9_l@
zm05-h@YDDPS<`L?cd2GWV79H;7k0w20n29EKHL5U!T$;i;NLBOzeu^PerF)9c~zWl
zz?klJ_cWfz=UzN)SSJAL1i#X~l3q?2*pt}Bj@A}*q|EY_AjpxM%X+g%kEH~Z8Q`YU
zKNv=?LfHxQ3NIV^BD(r}VOXhEHP_H?Duqe=X?`zNE<{~FWV+PRrzW=`WutG)`10@w
z<m>oIvW(*W%oj1{l_S5{e5#uf8|LYhfg#vryBl1A0FoLsVS_Jqe%~FzG*BcEih0o(
zb1Q;5Xz8IDV2%p%pXO^YDP4!Mwaykh_dibY>B@abR9LSLun6wEP#sMWE?2v43~{Ys
zz(?N4e@%XfbsA7EIKE$O65ktQ$?GRx-iyA!b-B{n+2Os^Ae8kx7b(&Ch>O007(hv_
zw}VpfbCfgQQ`V{Wuetw$E#>%@J)5m(a%n+1+XOce98hRvPcLArj;=32K$t)-7_h{T
zTnz9(;Cc9`PztQatzb|eeqhDj)`s;RbXV~5a=@quD}iKtesY_xXX!exfiFQ4{;OA6
zU3Lx1++RWeb()N;GvJ`mO<a7=Z03nePaZOiKgN_ze?AB|NBV@Fm|gL^cHqNabe*CO
zJ(5bG-QJ%@?l0dA#n!TWcZ~eO_6T{V;cz;}cQ2F6fj3E*De+vSMsJRatyUZz<r?t|
z0dCPv>?@<?OE;Dljc%=Khvn3(j$)ex>dVovdo5Al`<z)kNBProTZv^tGIVbEKjrC)
ze)qp7L$r}?8tfwRETadC&)<6RYuB*s7~NZkWg?qm`nJgN%iHL(SceZHgG?2_Pl89d
zB|nVQwEbELTe~eE_4;$ZuYM8K10lNo4bvRW&{tb5^FWN9rlk1a2i`1Vsd8S^Mm+DW
z@Z%jKb(85s_&q|^&-o53Ab-Iy`@NQuj|9Y&HeHh@kRDv8hprup?RT=isM+M}oGz4!
zUf22}2#^5RxENE?Y0(h*77*a^FpKU%eLg++cU0WbWS`i~;DB{_3PzMD#f!6>RNQ{E
zk?GA_FH$u^lCzV6On5M9?yqbQAGnX^H`j2g`Gb-Damim#Y4m_RhK%vY9d!V0TeIBs
zhu8<x9vSRApY<d8Lz%aE<&;@I*}jPRTF%Bt)!C4P0B4(<YnL41iDN<qZ!&1MXUQ~X
z>?3q%tjip3ef^v2_&d6sW%kL(`=6LWuaP)2S^Vo+r9SOBp0Y8(66|Blp9>1smrXjw
ze$6LYrIgji_cPuF+UZFhn|wlnSguVqmHSO}_zx)AicTlS-24KlsYU2uSK+O@)OYSn
zGr{^t-h*4sQ8j9r4~tBen0|aqK5%h<u#ha9ZR#ADC`$Oh9V}BKV}f22x8CpcBPPCj
zoZzlmJA5)SBl>q`i@*M8;LMM{QRU`E`tkG7K(nIJ^+bblb`<iz?NvR)D*XtYlu$fX
zQ}1A*JgTS|WWoCxA1RIsZg@u(Zx>>?&60+!0udc`Qe2KI#Q?{5;gO5hXx%%W`_n_v
z?#FlflA7Mhq!O?96a!+11L>maNP)-g{SzM#)>BnM;5S`aacF`!VB<N1N8@x%oaEp<
z(L-`88BsqCf#zRYJg})brnAEnGH-)ZO~E3s;PY8p>(9_IC$MfX-3`_kg}Bi$fgyqp
zeh0LXo`4r|Y3OJwA7rivE8MZyWOvMVR!a3BJPBM_EQ~rI<E#Ze?EOQ51~C~=ylY8<
z+B9q)!&cXC`_`_6R)pL>xVYRri1GQ+tMPw;_cYcRQ6;^l*CJcIz2m3ze=`Q#ziu61
zyMq|Yrjf6_8#tf$vV3Ad?$qPnm4o%*TU2`TecN&tSn*E?UM!9O8seJKITLT4md0y1
zq){PvC;lmisqujM|2zZb1Zj@Wy~*1r<Adpi_H5o45_VLb6Ki@LJDDn*^$>aOf;+*c
zmvTC$taVr2@CTP?g`x;$G%_7E;x|W{Kb2Q;j_>){RxXsWm{T^s`8Uk~*PE3t)}DZb
z?Ow23duSfEmUE9$3EDl-e@!(vem5p@2NEWGwdtOw$OFzA4tv*+>*pqaIPQj|g9)C6
zi+GZ7G~rA74`FrJ#{g8cqC;lh9do=)WajRFObBv?#{Sq>6=SC7Rd)iCSne=D|0vJa
zUcgJ5ZjS6ZLIii&t(czbs57fEwN8(DWB5eCm@q%~|Cy?h2rvJgP_YfidkOBNf?ti|
z<f@VTZ0X(RU#~BeJhe>&H)a68s10PT)XeP0hC}28q}G^;KDk4*PkMQU|H|S)T_^Ko
zc{ZE`fr#>Hg^a%H`>l=L*zGo83wBj8{o6Yty-D-7am2*$ff~wi*X5sxv32t@XH1s?
z!`?g`<jPz&=YdFBpKTU8a84XGSp17V@g3`*0+4w(O8;Jb2j38j*1^pu>J@)LZhS|!
z+vuNVI^$-Wcw`rCH4qTpSQuj=3&%$?;Jp&j;DqwIbz1pG6$zEAx@*XaJzJ3-%D+_K
z3Dl2HKNF$Na)$5KRYrU%3&}JjZU7Jm^Ung;$_U`eYXHm=<*^4L{{DGRwF1j3ux#(a
z`hpIGd(S$??VFFuhTpJMKgB5^5!`64PvL#Acs8N_7VO3Iytym1vBxBgo=XtJiVN<&
z8JQ^bn*1ILy>U5qO&xN_pQ+e7xoKpGerV--+H*z=TPeLPqs<a-u_3xIct2Ug7{>m8
z?VV*)9Nx0-vEXily9M_dbZ|n@K!Sw;gA?3?2Zvz6Av1Wez~Jug?k+(?7$CTFa_-*e
z-u<t-AK|Qb>qU2UcXidO?p5`B_HggL*gIdJuX)Pj-iKa3_TE4OENe!EoPV^`2x&~A
zeCNZr94ulj>Bk4Z@T)lq>OLuWddBNo2qHrJ@m4{g%q954UAAeUv4PnM^Ydv)&i-qZ
zp2&fx4L6K>@02ye-y1Pj)DIIg_3hU7bxoUzC)a?@Ij?HuxkAD-%y`Tb#)iJY0fq3o
zvz*inS-ULdP`&AQ%Oo$3=keJUm1J4ifxx<V1yZKKJKwJTLLQZ8F%jFm(F3p|G2_Gi
zSjeaOkfQj{z&q1-2D&$ydmlAKzfvd=9nozqT_niC3o^VO_DMg2^k`QN8Nv5}`5S#?
z&KZCw(<VR-ZD@Jsh3EqC^ZO<}tr{v{-sjtU^2J5beW7G_>@I54!X!_QnY$kv#sJca
zG(2TbsP{+rw|p6M&jGzLjri?o-z6_t#WVLZxx=L{1UAy}9%%F3+FMS9a+`1m<_jDW
zcMW^(i}8L(T^0l7f5{l?`c8;D|15U#p`%PO)3<oo|1gnRvBR5crtmHfd1gofggZek
zO2#dMIvUlO0yI&XO7~#ukbB;9vRJ!S-P3m7VJ6HNt4-_>*?9{@0kZLEekRqB!@Hc2
zwxCg$lR#j8OE6mB40G^2{zmhd&JN$1qJ1?xq&zLq7V=ln+Ysp^M50^F?sbvfPS`p)
zdEL3VmT!~0GOJW?+nNUReJ$PQ;?=T3y14mJ?S$~n@nF-ooUgpeT?4KxKg<!cvU`@j
z8t3M%`C}FkXPKwkXi<j&(IkN8k_==J$71yHhZWCOKM6V2@lO}I33WaIdJO*Dww-jn
z>{E8L;=A^1wKgf3zW@vY8I#?W!M-hK3iiGM`$pT`Zqmb6S+LFi;Qk-1!M2M@A2zqH
z759f@;B6jw_Xl$bj9}WFf{kFaN4f=&no<zFE08-Ytp5O8GG7yv{`pnF75^wVP*fOq
z;5LZ%@>9&?t2=HTT&M)ya*?l0lBm}r3Gs66i{q<Ua_i@sL*3ioVrT-lMxvbtmDRn~
z`J1yo-f!t!&y!Wu{ZArqh8;^UogxS`F5fLStYi=jPEXQa#Hy9W>#W!W6ki20?kVH}
zf{mr-&GYC-&}iG+!{<-peWkBPc(Cn-(|T_jNPu6UvT4^YTOT{@ZiDtLxVnN}P42j~
z!_#wdb9%j<5nB#QHM9}xqdbH+-mvk>g{yku46%N%FqG4lb7F0Fa`|mX*7~<<+s?nH
zR~IyBotxcVEc8xVRT*uBLg&0+*uAlSo*~(*!3*jz9?fFM!C8=_be#=-(h%Cy$x=tB
zRQUN#i4mabP^MRk@A3EzcjoV)f=GZ@oTLRU1gK0Ysqn0P$OYB+)JoB(4g^$A43P0&
z-kaf#Mz4?D)6CXkRH}T$U0$s}gYKxDjT{3rWW)jxeac><;U<nE_cFUlCQl+QZPTSf
zam+^dLF!h_cJMagzMPrnKHQM5ceZpv0Wrb0gP%Yz++g}qJnkscfbA8$OZ4;VfEwfd
z{Qw?%^0MymU696Ss2oWaLk&^yQN(F%3BM*r@w|M_+{4rzIAd^JN3@+YY5tKMm%ClS
zG-Za+#NRrvuAG+emmqxJYhfj}D7q{D0)uqlMwZ#$1PkKp9s;P3BBQO0li(OXMCs9S
z#%3f5qNO|D(aa388k<t3!Tc+tNtMP&U^VCmQoO)M#Bj^!h}gt27eh|(z-&5Q-C8IT
zZ#8mqCux$fOBjnVXI(7RH2KTEhcG#JbVaT%zi<Xs6*1X#*eH+TfV@4CU#!dcWTk;O
zkQBu7b@cIV(AC)pmGl@p(elApvKRbZaFl&ukDM<+P%XS~^SVw*?{qnLbB7gAC%{_^
zJ3tR@>#>Q}O#mxA!?{J54jRA5!2M`6S$3JCkAcT{BC$~-1hWoEm3%|D_VBGLws!da
z3-b7evuD<8FDmvab-#h6`I9>DI;it$0J_~LcHe@rE({b5jxuqeT}0Si&#%lED|RFG
z3&~hKleCx++L+WY3#*ZRR+Ni4pXfsd)ea(=bwlIp<Kx+EwpD2Pk}@i>M6T8w9bNAe
zCigxKi&!o_ZFjenfK;Wq@3h%^WB5C?<iYrCY%@L^jMs2zoHsTq@wrTqm4m&P?laF)
zaN+myWJ~5}0M+W`YbS2x(!c6l2(0m&QuFY)K5gg9ai|me*SXy%Fe}2Nl~Vlg_ji(l
zoal(f5#PP80AD7wQu~Z5bLsF{KKfKkEW#_-$BN45pDyUFxb+R!UexRzAwAzkq+jzp
zf`e^`UZEK^u??Tc<aTBfQ4VQldesg>r;e*p&!)#;m@eBgqjvrpRgg7j7Htk(lp=UT
zxGlr7G;Omt%?Nvz+Q*@s7V1mxt8b-|O{Lh_ln$)|Bjv)bMU@$#HkiX*I*|~MMJx!Y
zz%ux{8WqI2YKTt4hy%EYF5g_I4~vY|eI42<MG-e^i}4cXGnw0lSX~u~R3T)4#<Ps<
zYfg-RFdlgVXsln2j7jkcywUWMsTn&+UY#VVKy0!Wr3Qj((-jX?-FPBYO7S_u9E=Pk
z^gT+)?cR^#NHMxc$XA|^R`?l#I8WGD-ca8vHt=(Jn5pabx_TaCqfw})w1~$B6)A1H
zkQW=>K^|&M@rs`lw~iLj(?ex!>{+i_<6hM0Mz!L{{xDz;av>F6?NaMT*+o2#hbLT1
zBq!<UawvCllSIu#HEaAT$Y4`wRrDHs0EgL5{&g@3+bNZoup|mKE$U%enRKMvEd#8i
z1x2l+YJwkMm$`e;pfo^gh({T-;?9Mxc+u7@0&AYmVX0*=Ckj^4BgXx1f7n15cqQrT
z&pb^^X3g(+dLHIXmtP9{$5@VIP%pEGnGD4`@Coffb)~3&mdYppuq1<`7hRNC#$CEW
z^%e`QTD%{V4aT%VE-(+!(W&nA{;a~Lm2{9WM8i7AyfTWbcgKlGhWj@r+9`^SX^Y57
zZ#>(SOX!8{)lf2OhP#*Um)_n*T*I*ip=VYEq3LEsWt4fI!$;!c4r1r;hA7sR^*r5G
zAdk=?MF*G;LZtH#$tW>Q)PvL+rrfcrPoc5r%DUqs*^q?)3-|b3Dt5o5KSVk<fvskf
zAvBxy{8g1j2zk`94X>_y`l>g}gqnTXl9)|5TW6B5&D&~r)CVvNt{RR&TUazd0N+>-
zjpt;;G#W}nMX+3`kbzZ{+F`{TG7lxy%j*4om6Ae*L8c^SXVi{Lg)Y695GmY1g3rlT
z>w9~NyXbr!p2qkcPk^!NIw%8zEDFYgGwle1T0&RQj<;=8_Y*EC7b&T?DkVD(TPGEI
zC|QE84e-@rjTMUS4HKWYp-UreQ=-2Z&hKe)zGHNs$Jqf#L-qFkj~d8yB4JpJ(?+9G
z2&cvnT7Wg(I$u%GL>_^U>nr4e6L?g<#a?;Xt{f0ll{?nVMx}<j8+@{#Q(DXPMmv&v
zt}r4d?oh8wW8Jt&#f5pdDAvkb67lj?$b#4;lN&*|K$6m=Bw0UJ&yw#)r-`xib}uzO
zA=4B6`|4VHI#k)AnZWduIu}B;SSjF(<C^}sRx|v$dn@KoITg=`wi$ER0gLutWzHmw
zk0=Rm_IJyI-BAO)AKGTmJ<~^SaQLk^KUr$Fo|GXZ>NR<3#*T*g-D!p!0xeISu_sn3
zB$EU*t(YnaU%W+6soM4=mAezH>=|`zQ>%x_i}Bog&TQDY(Qp_%7UH2b4!JOQCV$2E
zT}A9^UwOImK9lcy{HA7K`;2w34@5J&l^vKmGwN^I6XLUa)XRTfZF;hF4kPI7XD3(Q
z=%X+<QB_8(YHm2!cXQxC=~fWN!Qk2as1Ol@4quNg<ev8QRBB(#spCN!@F9g=+%Z)f
zE3nRf_B$5A@EWeDfA~;%*-Sg$k1!2)CH)eMaOMp|rFkN{*T6(_BNLsoaH`a15^Y)N
zy-EusR67%}HZW(WepsCdHva~FA|iU3!zYDr*a{zPAiaoRcwMxIV7Yu-Na>_&>?aj2
z<}P}$B0augI?wDFa{#j5Row!^S>9)@NCS#|7r%GGM#W1i_s!e(X%HDjBN%0uA&AS@
z{0C)`8fmwiXALx`l(83Mqhwa=w0O!UoT9q&pNx<p_i>?bBC(eiQsw8Iwq5Q%k$Ou<
z#8TAV<mY)eGxxK=0s1Lslc}U%NZt?Rne>J@PCn0;9TF3c5w`<&GV8C+*9Sy0?R?cq
z&a1hw)A@k-SGUmZVFAT&9{XLh3p;7C8=~Bzr0Iy;N;>-evLra`tts`d@FyX{^tY=p
zodiLuM;^dqwjH4=6^8XP$9q$neIXA#C%zL)L~4ZgGk3*>FH#p&(m%yLDds`#*tZ<V
zpGdt{o`jy2>wGD8j1DMu#s2eKJJzFv5wvE}IFmjJxGyw&C8TN8kr{={LNsd+l!QR>
z5?zG#tSou!Nr~%4v1KuLj53m1X0JY+HY`={e?YGYJmRni1o0pf(d*GJnYD8%&6dPQ
z2tL|JiFxmt2`v5{-+fJxkILueqvcmPow|3vxqtwqvhUkWrhc~y)H=7w^j&sHtfvil
zjS5)d8dUm18v@}FuN)3h!5<N7G4#i672-~xj2dY@Kg2xQj2X^)9?8(^0-Vm5=XPb7
z6Y!Tg!EpX^^Tq=_jO$O)cPOLCCgzOWOppQ{>j_!AWUhM$(Uhw6(Lh}BS^<5@;<4|O
zj|-!_)y2-JS)mE_sec#i)w_Ek-J(59>q*Xda%uMm<u=A{PTlWZ-%$tZgle7vN7U_1
zzDkeKCzj-7enjTcFT=9hDj{MYaa$C5Wb%)Duvp=C0@>|dznVsCVsreO&6bMNHO_p$
zJK>wgkhi{Jad#Xa)bpkZ<6GG}oTSY#+WkV{%wh%Amk{G@WzQ`CT~aMWAWz(6<~0NE
zMXq~=IZH{2W3bli7O=0l`yt5Y$9mek0z#h=#GgH~va4;|OCtg}`4Kl<h8arOoUsZI
zphg$oPJFZ_#MbbM&Zz{~{Z<HdQv#Wq_FMazpJ4CQb#V*+CPcoN$T`nP8mMoh!0G1l
zr{`YKy}t_s_M8E0c`_G&X){|ybgS6l>MK8}t4RWkao-=2F3WA3yngHZYdtSXi0fp3
zPuO2Jp+J9e03M(bQ1knQUlYU_0r7a6QLo{Tp{7>uuR`1pk-Z6NyPn&@1DNs-mwxda
zOfh*<%369Xr>P9vLVw>nI%>?FB&$AHMTe@BDuzm*x|JXee|c}gYoC>dFgXExP6nv*
zZ}9tx%?K=JO$@yJ?C<v{*^yq(+B(@X!|sMYK5=Nx3DEpfm>*X-gOE2K@$$svOm#+j
zUj4xzP&>wVeD?}n8Im9DPWVLVcJ9}@Fki=d@XSR}lo4;Aub=?UoP?xb)_CtW8W7A7
zSF`VAbBb;VSlqyKK$*#)<qAhq%)uck^J^wIGobkd3YM%tU6|Hddx`SOu<NUpflkH1
z&hXz|@QUHWYN^djFgHEzjp01NX}>d-6m#8ci?nb!G@#{&d&Fw>lVV2mwu3kBl%Pql
zLmjJD7?bl-9igY=@6RG*srvC;jDFJ7{wR6l^=H^zq?Knxm3!Hf?%;$!Kv;}FLkVV#
zpr|KOZ5XpUel75XcAfI_0)TFd(pzUKqmU<=B8c@RjMQnx8L^0MIMn2A7dz&KQah>U
zt3f_Izuol7nqG&Ie}5_HRbqMwmr9$Q#2~3XN_g+x@YZh3rzT=#-9kpvyIBV(X3j%w
zO3Ecv*iBl{Omp66eHEf<Qe;T6qyfQsMg*k(^nTPkF+zj^p^vWto*$IXQp*d}bbO|z
zq?&50E<<_h*T=Wm973-vGLdU3?_I9LVoMkKV-1NeljSdC(xO<_@lnazBDPl0D$M1N
z&O@>XXp040_uP7b|6(yTTIJ}1Q7t7vg3B|B!R`jQI;8z-n=0-(3W$r|tDaVS76-#x
zlr_7P*FEs_*{^P$W|P?K)qfPd8#Zyou<yop3sx*-$EnKRF}&$4^!|7aeS`@b1o*<{
zL7Mz70*$M=Sju3hgOVwoS9ooS9@=1~814^oW&DdT?Uh9(AuSYfLKXWDs>jPW><X#x
z?T?NmznL?;<{oD)Xdkhpg$$h4aVuqyd@WyeX+Wig3BgfN2NBT1kOfzv&>uqU_ZgbV
z2gUU=Gd|eghFG0zeThisZJnHvHN}(}KKn%}<|1{yrxIy?70?I~kNY#f-8{3aczf9M
zS=74-+fv`&E3`l=W_zxUqIjxb$zJ+K>N&`K4k)8sOSMd@<O7TBJKkS&47OUxS8<(}
zn%tK+J^uumP(EV`)t)YJsjB9Ogf?7)#JM`<hWakOB*01*Jry()W9El~2H5*$&X`18
z1da&<4Wr0FJ~YMm#-l=ygVIaLLA6}uuutg@{C(YOA**>jE+@t|&0g|u4QM3OqixY@
zrfEhyGk60{V9h7qn^zSf<Xq*!Nd1vM{@o<IZwC#QVz+CyJJbI@Acr0lBxeXy_ontW
z2Ah>LhE8(xZDJ<>JQ9ZQ*TCi4-;j7%sMa_x&R8#|g?wHMG%Pbt%kqF)%_vkePm1!+
z$LImcyL_LZvg_9`DOX?DeXs?m9NcNwrE2QiUF9kUa?piFW;atT+t_$NM9}Nu`k=mJ
znBH^zoaQ_uY$<I)90YU#C-9(}8oC|j)tosWa%)K=O0=vPg~79(v?0vXqerY~0DJPY
zHw4(7y3{p`A|P_U9(~?0(c^^L1@%e1eG!AklS~*H|E-KEB@LF)o6=9LqT)oDX{~q;
zs_<3UbpF<0b;SEcZh=Vok$!d4kNmpEhaQ0CHNfq-1>sJ|>0WM5!^5migOfe5TF&=<
zgMi?AbueBeF9=<^1eG6WDGdKNCyWy!sS&ZgiAoq(-{ZHbV^0TiG+qi}5Z5=Fkb|DL
zjWh3Dwu3LT&~J{*A3m%v^$6l@|GY*$NlxQ!SN>`G+;QI<aLZv0A009d>I6pbc$hxF
zAAON+3e#C^6WP-C_UJ?<tk;?JlJt)or8uG2=K(Tf65JR2A366l5p4Lk2GXJ}cey@(
zv$n+BBb;!luQ2%vF~jPI%=80>7+|ws$8reJTHG<WmMYDyKgL}(wm&k+e-LHe5>Gd;
zOGq6vG}BoAiN`R<EUA(8>NYDTqy$G3#s+`R<PQFPl9BJX>oWd8tUSlgCk~N|4E!>-
zmOpvw@^E_xK;40%`)X$i&_~<OV|P0TS1BTJ@T_Op#?&@)-MC3!1e{?4kv)Fi+ytGe
z4fB8!4z{FrIX*Mg9!}!hH@}ykSu-U6Oa6u4r?iHXl#zu&$E;3<J^go8Dn%}Eet(gQ
zs?a<N{<Vy8YeMP3F_Kp)T&Qm1jca8f8Oum7j;Qp;9x_>Jh)Ik{EKkaI&os4f^p2OR
zdS;hjSrx0L-OGH;VcyxJFm!?Cxjt<fxwt4`cvlZxJ2QE|u(y+9Z*PqcKx41OQXN3Q
z=AKU|jHMcqxygUfjDJp4ZuVcmTv4YMNl?e6aKp5t_$|K|g~7<Mfm{>EPe{xymCgoU
zxkIP$$stqrmL4UF#9%LVb6e=fC{#md*pP9d<gj3sIaTe7vU!PeHo4GMQ#dtBTvWbD
zp#;QH%rJ)p3$H7@DdR|AA}YBFFgGZ)fRIelf~Lf-k{&!ZsZDJAy{JevbJd|k$-L&V
zOa)Sh#P+xnZpK=TZvtSVI7K2S!ynTPscJ|~cY2VJhbH4j7(R~=8!<^hL=$kcbBAGJ
zP3`M#w&AT29(8Ejv(o4)1-F#*OSRj+ra#g-qWI$>goT;<PL1;}-1UyjG>pp_gVBIz
z%)+dmsc>8i9o*(lBN&oSuIG5|AE7oUMSdu$P@s7B$Q=zq!Sq=oobA)N0_-p`UlR9f
zt!uhE(H+K_9g5z9vlmWVP3ZrV&p>sT{?uzzl4gTbJ4-RA01FY$JZ**3k&(#0M+q8R
zMz{Pn7w;3{_0@!ZBiv=VdD(%u%yIwx-h>&V<H!<wZlKXKNPQugC`Q&TO1@cKnzI&8
zo)eUj;QTY^Dvs#3EdOS>bNgbIhU5cFoEb<!oP?bbofFgZ?HnnNB8PpcEt@L^6;9w-
zNDB2)nTtA?S|_P?@NQpoZEz|lRkSKO`mymM$m{RuD*Lf!Gqf7N8wD}Dyglfq?@sUw
z<$lRI)DPbs%_-%`kCiRv+pE|4rXB^cEMAkdy3_g6iyo5g)n_=Kc;~rz*_q3c4sX%7
zN-*6$OEBfzh?ZUJ^>EReR2gorZ03;b)8VoG<JHmb(irEa>1*6Je!8+Xvh9%d=gfnw
zN=kK^P@qB{sxL++f*4XXVr_P>eQN1TB6M#t(U*Qp2B~@<{X!a!Rz%28_ABPVL7q8@
z6`6$hK9bhX@wk|_nqKN|fn?l&(X?_6VR>t^ecLPDkA&%o_vNIS5Hnm|tz+}eDKfW(
zyChC(BV%qVG?*amyG+y*(%4t)m!q!1*UH>?`lN`7(bh%JR;QIAn%ipAQg2O+64hlk
z)$G*F+&M>wJIi+y_H)tWC1A~Lo(9(7N0h79E_*u4!<(UthMhDi_H=XY-vRqvJ*Z>L
z4Kj|3e?NI@ZFwpDq<ac^Ng#Vk-7AB0mAi3@lMynjwd1X@?2HA8#mSA0Up^N_g=~T2
zz}ZG#UVZlaE+P$f&Y#jqP|1(A%r3R#0;VS;4{I^V>7UT(B<KvAuLx()Ovg#fVk6(7
z<*67FDSKx^c)G(6g`X5V@x_xN@Tqm05Sogkn%eigiX1P)JL6Dr)P?)80{4}2kPIcT
zzXlPPoKdhzpiIb1Xn!ldNepx{@7Qvy@nf4<^^-fxC1SzR`S^!!w%?pj3fJlbt*PB1
zYy$ghAg3syN;<|%1;OrPAO4Y}upu2`w9~E0dCv$p<}pAqRzs0uWkMkv)ITN4a*YD~
z_DTGJad9LSVxGujN93LCW>j$+Cx}b6=c&%5<{xtTL7K6GuUj$;8P1bJ-^U(DG1K1l
zMr9i3;S9!pbWS_6SEG~fkMy_$dMgZfcBV&<We?+=&Jq-8jI&L18yKaC#I(iA#^p!L
zMd;IKfl^G@uhJcEMalQ9&NsxcLa^{h8m>h)PO`l0&A(AuW=Wck26=9?#ODrE9<~JP
z130pnVw36@C?_*yS+jw3<7kQqmGtReQ)AL4WCzJ5EsA<+$|szV+_~dWPpj<B5?K$D
zs;y=+j>uFY0w#I2rE8dfcK{F2$0(i)sw#G(U8BB>Td@r@!Q>5~q7Mreiyo$mdr{7H
zM9r#oM<e6RomE^XL#^DslX}i*>rndILe`tSip+|@8;RZNQOoMoH<i4gYpJ8IRyJ@K
zq$bs@Cu%17N7&(&ec}abrs;dk`CdjXlR~D=8u|A`eO?jn*s%#*`S(yHi8|Z!A6ccm
zfdk><J1eZt)?ggS{YT6dJ(<PwrpI?5f}hpLqe+CHi*L6+Z~0CKxF|1EEIVi%h-6e!
zG3ml`8b3Mz%%2?I|AX1X^7DsrUZql3A___vGWLzpK@^}QF2**FsJt|d{&(+PrSYn=
zTX$j3v1eK7^@T{w@J~Lr`J?>VLf3?1$CV}3ZS=3fVLFWsCO!3A9*{)~bdq8C@0T$B
zVBCAm3=WVdC@YvpZoVz0O_0yAyD6%g9lKhxzYfjZOeU6fz}DgP?<@_yE+f$4huv^J
zzF2hZDi@%HMLg>Rg^nSK@&f7RG27B#zrIpu>gmy`g6zr8W^8A&X_4X|%n+ANppi6g
zP(DPyF>?;cdE^}sQ)f&Ej1+4y1(c!Qquhhze>Mfd4_Tt+z$cb3gNA=PWpM>J(DIff
z?LZ|U5^+sQa*V+7pu=}-C}T=^1j>@>1pAodMHlipy4ld#yJM$0xUWY=sgJ!k_n{Q#
z+X`^ik+RTwB3#*Xh~9V2{Ah)%lt93;qhY=v>0qR`$7%xt-a*g(;b}qlunjxNDEeWt
zFns7K9hl#W=zvRS7koHq{l!43GFFehvfuI&^tG(3xB4c_(6I7%pgC~_Fl<SN9cRw}
z6*@J4a0)7bAV9qo8hb51ABDsBEc`AG$)O3FY{$pdQGIB_Q<E}WZn2^TeVb{4wbGjE
zA6omxL%=gVO=r=H(#VKMofm9!@HRzDp8J*?z*OaEw0W!dE>G*P)`9L>oY(JY!}hb=
zm&wA4kypzs>n6}AO~%M;bkzs=8CUR{!TCv>eyZJ~U6y>G1V`&qQWs|M-yMXMZ#&@3
zf6hEQC={jP?2Xp?6Z5s(P!QidD?bqp<<iV3yJWl}7D);9&--^D(aohA+{Wy~+gHoN
z{cph^eBsTr*LDacd;f)M-r;;y#T~LAOXWD+A_V`UNe7)rBMCQrBMk}<?tHdr<v9sK
zImJ|4qv&;*I%SV_!nb9d=<GN=OqcDK{E_=tylsAah1uC{VZPJ8p4u#8QvBa^Ya@L$
zn`7B0S8#0A@CKm%O?zL#Ov8>HzuzwP`i-0Ou{3)|(HYCu1|lZp{+;sy0<CB-+)>#-
z7dq*+?vZPkx6Ac^LDERQl&=;lE3OI%K}yU%tm!$zFHtUIk)+VZlUEFUhw%l<w6B9y
zTW-{%)M8pl@lzp}A9xm0iclJUe^cWs`xgrPR9#_{;>ZF(?v(f|2V9^ioXLcjKoNQ}
z7E%4@k-ZvE7|L!6EL9;Y=0p3`=&HIuLPXEBy28FOY!sG`XGd6a!HG8UMGxUALbbip
zqpgj{e?j~|@a;`G#>zTwW#fs*SuN2-rSYsmE_Js2Q1U84^sOzG?85)!G%iTnl@Zhj
z063uW97Cee`p$R|%7dGg$Kj4M6AvZroV$6hgW7p04bfC;r9b{1^O)jicmEx>a_%>$
zY^taR!vCFj6cTe8)ei+NqeSc)OR*FO|M@PsWl1MuK6S?D|CQ{&5!{U^$k?kNdk|j=
zr(gdQkKRazOxC!z>}u%rKRNshlCH;)gVMnDO@*uEAN=r5U9`8Go^~U)ZDjvIof`#x
z#$GMAVS|7FPvG~L&+mz}f>z)>MVGq&;FdeTK;d+5$;+rs{7)h_2a2Pj%nQecdf5D5
zyZ<ogB*QH3M0~OO-{AkZ^8dfK`DYXT`$k516Bz8E&a&(DGX&}HQBl-TD3>!0{x6x3
BPwoH!

literal 0
HcmV?d00001

diff --git a/_static/benchmark-2.png b/_static/benchmark-2.png
new file mode 100644
index 0000000000000000000000000000000000000000..30f537cd0fcdc49b0bd832e0fcd0ba169d37ac58
GIT binary patch
literal 142903
zcmb4rWmH_vwk;kcK^hCe-93;%aHny1hY;Ke?(VL^-CctvxCM82ch|hlx#!$3IrrCN
zbdQ#?>0MpBYOT5EoU4OnrA6N$zC(n7fOsP=CL|950UZPZ0o4T$2fPE;BCr4+AnoKu
z1t7|Q;q3u0oQzb(KTAnL&;ZZjAz&ciLBPHo0^E5ZasT(Z2qYB*^uOMRf`AA#fq?n<
zG19=}%U=|5fBB#Pd4$e{{P)qodzn!GIT|`B6Z${Tp}JoF&5^;T5O_ea7E`l>fWRVu
zxkHM}lbu39@Ii<R@hdt(9;Ut0#pqt>S-BMXg1~keamk>BA;c$#3K^a6;}t}8|J3aR
zi;0OT91;?go$2M1DNjcwocqCKTw^!Q@A7(>^<ytZQ<Q{lKlz&T)oknBT=Lusd6WB%
zg=PEGwMLArm>0}{-7HO^@Hy?R?Ih%F6%<yxV<8Zt{=NmJg1*`kL4y9z4K>rJF9Z7j
zzAxp4NaBh5U#9_HzK5g<dgT8<9|IbNDI_|7<^A6S6XQb%PCy9r|8s(omY%hA*W+lV
z5@m$J58HJz5eMY|J96(uDIzW|uCI(yC8XYXL{U2!U8}3QI3Fy(y*Hbd1fiqbL7)}<
zIlR6R7(^^OsyFZ6_l+d>nUO*UB5ZQNj?K)d6f2cbyuoF+za^QA{%hQpLLC#VBTHB-
zyS7eYFf3AD%t6T_z3AfqHSI*6Xqi4q?L0EtZ~l9_4O1X6yCicqg30)x{<=lA!;j8!
zdb+2#y!&%XC_!RkVo?eO(&DPBZ)<ixMbie_(nB&bGEyrQ$HBnC<*UPEb}5R@{{3k)
z@OV6Ks@AMpO}0j7r>8_%puP(jSq3H=8gK+5F9R_#u@7CFa+c4RydUm=o(*&x31OLQ
zU|^6Z;E(u|uHdn%PX^|6bmM$$(DrJ7N)eI70QF*ma@dUcf2SQ1WK9D}6?}Z$V&q7p
zh@3zRznT@;+I;?KGJ;8ng3bT;x8#E=r{sejyUKu6`Fq41A`mw5D7=rAf)}m7pC78-
z9~=Yk^AF#f&oTuE2aC^r+qwzLI9O?K_X`Xx_%uD1Bf6EBs10t&pyhPC+e)l9nV?XS
zbnoOSB3ed8L5cZujq^Q|SRXHP-0hcC;58Sncsxi62nyown;l&)eO>AA3CHd3?d_dE
zjC*Qq{FuyPmoHr#v+RY<s2%5zgq=S#Ox~ON-h>4IlVk0{_FfF^(+6ryA(io+#lNne
zZXlnbVXlyn(C+xVhwCGW&-9<{ccZ@XxN}aaeKbB<ZneEWoKx&Yl?b*e*KEqJsm>qj
z3$e4|+G-4Hx3{ya2@kX-OoC3}h=UN9(Dt}9N@g|)IAYosZdu5aNW8q;?&y!Dym~;w
zW=NuF?XK&*hc`k!5D<uu)7Qf85&enjs;@7Gg7-1%@Lo5#i%a<jn_jpH0*mMR{ZZMg
z9lTGDlGi6IjALVC{Q(_ZjS=XS(zEksGv)N;;JDGF`X0?nL0D#NH+W5`zgDX%0@v@y
zi}4}_N(*ma5^s4JS?OZMB9f^Rl^M1}kK46Z3)E)QrO{qq5Jf5#>B^jNC<Nx8;6jPx
za+@6vz-vGKw{|i->+w5(t$6+}8|k*0?7{tLty{gZ{pJ^~Vqw(#FQa@f8?9m4rGz8S
z9EtqAFPc=oM44X0H57~9@cwd7zQxI=NG>P9vX?EMRz<pk_TxuY4cK9X&Onqz1ayks
z=f-Ce;gG%v{7*w`h6}vUw}Lt<>I?;avj_|h`;%ic#uzh8-@hJQl=^Z%T~3@I){Lm9
zUh-g=pO&dsWot7|*BJJLG<d11<7t$}dzm9MGBTua9-nSj^7y==?pvyj2Eo9kwAg9A
zo;P#csd2FWfs84H&9@Wx*Oj*>h3aEMkRaA1lQZ3jSJ56roXPNfE-EvXNMx9@agvqP
zO*WgW&f{D2-a?%2Fqz2jFW35hiQ#^K8I7&&#82p(U8CQPxG-4>v+THFF&~bxqhCH(
zW4usj8cx8i{(OJH$jwvgI+#FT6K=5N?P{{o6J92zu6KXC5n0#Gz{8_W*y&Q`5e7T>
zZXGck79H!?&9bYEhF}$i=13H=U`#lI|IyWqPU3Q_D^i?1RRcAwa!;8?!vMC9Tb7nQ
z^K(ul4~^^f;TR(Bv_6JRDKz+SG)qtcIj5|!@^#9U&jPSTENTwdU8<17UPV+g{p1WX
z1CH?B4t7+f+8IhVz-`y-4Cu#@T7bkpu0+m8TC+3#E1*=0LWF2O-CxyhA@pW>@e$J~
z<cA$2AK+miQ?bd;XW~tjswoiqULj)jMe^K>@jM)|dcz`B`H5tjzoW&P*L$PUWSgh#
zlbN*S@o)rnK-N=<IDN4hD>Lx>`G(hIraVpAtF(8R2V)^KljuU(Dr?jIb}exI=*Yq>
zQ0Wa;t?w%o({6+TJdBDMa*2H&SyR0y-;Uzorf}Uf**;yi+8uebdp)EQ;~bH2d-S+^
zU^xx|p{|Y`9dSEXJoY_hV}y^^M7~s6!NPSF4_V8Pci}Ne64X5CWar-FmW$fVlrR}@
zwmXr0HYyX|F*Bz$Ms#%WO&pHX(+4qhkM#eQ7CLw!sc1!3Nle3Ey|;1rh<uP-$2gzK
z^s{a<{r2HuAqDOGs}d{>Au6d##Gh}@6ErNVu9!cGnP05>q72Md8PtJ{9SqpqEB&*f
zyQgSmokB5bKbr}(5Md}>y6zVhY<6SoOdi)xH@nFOh4WQa6LenvemH5DZ*y~==sw1}
zc;`Q1-_b~+jaC&?DJ=l^=3Op*cz%WhQmF3UcwU{)C)-W2GA_X;RxH6wDV&*L_ERYi
z-udvPkJ35tsg5d^Cj9o~?!Tn|isdG#UbY)S9VtP#LM24nmB4auMnEK{)54@t96f2U
zSonN^IEaOjD#LTvRqkVOaJ%xWP%a0C5k8Jxl}Z78u|0f5qdtA$jfs5z>(j|DJqCLY
z-5%UkWL>)nt45=X<#LM}5JvK~46|0}(M!=1_(^41WYReE3^=4TN|eiB$T{lG=k#kz
zQw%;MVXIQLRkYm~Tl#dPn0a;3An<47*Hqw6@&;kMACxs**3BspF(YFb>FDU#g$y>-
z5;Z@Pu0s6<VTd>&sYI1w`<Gn}<OD&9^%e`J!Uzb)tM3pcGK1$HfU5vgWhu4|<&TV*
z&T-TBa6A#Rk@j`DD3Y`Xt^6CsLcYTr_lN7LvtTM6$WTJi@+^8OERJBAKK`qto?(z>
zWr<1!))0#S;3ccZ!Q0Qn`ES(3Y|_Nspc_<OeQcOiZjP56hWWd&^$DmFM3&uew91G&
zEv8ND-uG5@*9jy;^LAO2@JFhllqSpet@VV=2i(4nI)pom05*GtGQ@YsBN|{%V(u<>
zkm#aB+U~Y-osXM#h88rdu4FyhE4D)7KQ;ylhhptPXc;18|Ahd(Sm8Tj{F55upgu@7
ziu?q{$3o3WG|G@1qd=*MG)9wTy`F~Zeied@K}%NaeGXT$*s$VJLhl_E$ZBI9q|?O{
zhyBf$=Ph)<0u!>P@W?lgoY;a=jRt9r8778Bf-FB`XsWkFcf&HgPh_FkjMK-Nz=f~r
z7RxvEoe!1;MYm%3u%J}ZrF{I<MM#1SGE@8RSk^IxUY;a-nHjcrA5D#UTrQ$()YV-1
zxfJ$5RMKE1o%>4jV{g}evR!7^4}n0N=u5Z@K{+^9(j<eSBxZGQ)&thwAuJpfxdv1O
zH{B+`ajn`fa@-6NdSxFYVBT9#32Ty6j{?&iY8{yN>1%mJ!{Jm(qGGwG8cI0NsMVnw
z_VD6do&Qgy1l4x5ZoRNnQ~M!Di(j^9Sn8gU1l;zzrJ;#G_x{}3=LnEgvO(B2h?cR@
z{-A!HQfAL-p2Y?kh^sexb3#x!B`lxY8xGX76n}l=P@pWn1CpYxJ#o;UW}|f$ooXeK
z=nh-#8mkYK1&A<9i~FPX_j1nE^u@mr41i|%u;7WZkV1OC|76+Tt=9AR!$bv>3}|oG
zcEkjM$eN@IiMH!y{0{!aWklX?d;pqjQj2PNrEH*JN>VVBk_wtotw<9$Eq%|@gfc<r
zR)7}d^8D!Bn!$pkKoRJ;`#$Mdu5lvBCa@zGb%Q_bYK<D%|1ShAw)zfIg*vb)*|VN1
z2mncwKT1yL*oDd7z+m+I;mq8F=Xx)PX*DpYh3hF`(n)qP`74>!j2qr=q5K$tw`TM4
z(#ri5(9Xr8z>i84Mv{r~SF!-mxmc*z1g21K*c(K|7}w(_QCe#F+od|QM2qLBC{(Q-
zXbPnZf%W7ThCUd7>@M@@vLc;!56*>{A~au&#y$-Hn?fnyAyqN-*V-aQn(ZE~%D1br
zuLpJ1&j%UXcbjyn-+NE!hTFH#dfe|7aBeYdcchHIvb#B1;VmPclm08bky)doTNnE@
z0zfn^_*FNN-jdG&sV4o3+l3Uaext6(&C=IugI?sPpBz^0PxVJm9&`Nax~d@0$1~9v
z%u+7sJNZ)#pUYzYeKL=GGo4zsP{GC1K}GxKjE-k~0=>rmpr}@SWhGM?)6et3_dl*T
z%fBCwe~Xr%ftP>KZnhWy-YlKWB3y7Utga7l+R`zMk(?sS7$ETX82O--p`?{ibzJty
zE%I2{3^S96+)Jp15b|Ex{(ibSsBI8)*C>$2*HY<jE_|#+>2~_*hrnX9E~F(0H{+*R
z-824*b`4#ID=XASfFFib6^~e!<`ps?RpAws_h30UvjI*h`F$qVYt!Bic?6?T7fS%h
z!YcQkwEx!DU{Jtfs{8P2;m;^)O9|;`lI+U8+2v&ex6nd%Jso;lHkin`lQ-C>V&+in
z6GRdj$$g#F;RC;Q3Z%H`5KP+r_4v$s+b!^$lV#VFsSwFv5c%fKis#4^Uxf&GvBe1V
zaHKEZ>?9n6ciS!4aUh7oO$KD^7?JRIdr)cAj2OxyR9`6)P<!RVFl>CO-reu!MCNmm
zFWscFTf`WrJBwgL`3tzL{e-=HycoH17ReeKuKP0H%IElJB{&LmeY{k1p}YD9LpUy>
zqgBTewbc0W_N>{-tux?_0zE<x+FSNh0sJ-~V$9z^o_3;)*O{p-xE^oPK^9O4i9BSm
z8b{Qpyuo1+LiL<2Rbzc3Nc5lm2tVHcvpIlJpG*;l^>diz*G<Bbx}oT2wIlR)k3gYz
zE;*30Rb1|y1@i4WYEb)Sg?*s#aAJ;V<o<zZ7||k+0*x>AD{m;{QWxN~2|ykd`r6;e
z_UhHE!9YTf!gz^`(5W_eSMFaQR@`#T%J_l<6^rDBt?p3%0P%d+sA639nxGqk{)8WT
zhLxXxH5?lc`r*8_2Vi7#JrUPz>D=`&>L?_;`E2IJ=a%yU+WC@Er-`-ZFiX9AyHq=v
zZ}~(dpQvF74};Mu?H7qbr5#{+e^{u&H2^>o14%961^!K!Y2?{#^sqSWE4nW(^8#OC
zIs`M0tYru#qP6Xpr9_BKl%Ftdl!Hi2+_~dGrwWLhn`~|5iloY>elPR2K$dSE2AJ{s
zhr#(~|9siy=~|~cvmkHd6p16Mk7N!|c}+f4vH%%b`L`{fw=YZJVXkh%GFyY?l6vna
z{Y*qmn!Yd`Rv4MMpPij-R<*<kzwAc13$#%7XDTY<Rk)dd{}~jmIlcYLT*>a|zLH=2
z;($JxBnZ86)mry>TG+eDWm^T#vN>m*F?ADj%J5RbmC?rF{4DgXB=z%1b|VJLow1uz
zcwd$*y^jb4o<JQkV?f-LO?sTsxFWp#X7ONN;hMnn(dGxY2~aINfP{w(@Ud^6R7vT=
z;SEAVzk8g7x5XWjuI`m)No9$umSn=?00lw#^IE4XnVvVu5ECzuBtBu${bp$%6ujY-
zQfal)R@=<ku0`!Do>&otNHS^A8<P4BRfig#kT`gUG1zs_ELeM08d0GUjmjML(RM9f
zp1|#te+dR#enSLNMUj+ZqsSR0_7iC4_U<m3VpzXB1oQq00ESj(y}!I5p)UvLZzp7U
z)mCgX{)qBKsqkb3+|)KYjFTd8XKjvzd->a7)oO4|xk%908|1n77gu_69!C+Mj-v`P
zJk)VtNA;nCb@=$aq%k-EAelDjvDj!c-)OTxNe#RXx;}S3s+*}bqk4@T^>x`<l-2~O
z<`Q1ueh3zQF@0i$Q<IutQF+DjJi#>d78(c3Dr_Ja9yKiVL1**Oi`b+q2u;3Nkvh%q
zY(Dw%!zYA~Qv6*c2h(M=A|}kCJt$ZnM8vncu2=hcgN#WVeflBSwYkn0+dhfW-{^wZ
z@}WG1sXg~8XCxUB_8UIE?-wn`pf@xdJjPdjr&6Y#=c!ih$`B#*M_7$wgR;?~Ys|;X
z$Ls0T0(~etx8leoQZIpNx!*6T*8Ks+;qlm5KdYxE`!f?yAWE%Wz0Q==m*+L8PiO7q
zq@r~U3L4rW&1y;R>Eg#L!d=%pdX4&BF5G->Rt>|)KvCj2F+XJ-U-apK7Q7xLG;v0o
z*Gi~St1&UHFyD%avfW_8oc1TnTB!wc<B*7%!Fj&HI5Lok7*v@{l+lxLHo;acXCl|y
zMxRAh5@c`GmB?%s+OvnUXGG+9JH+!s`>QHd5#$nt8LvXqmwa?UFsJf=1!xXP%z!w1
zW%7_S1ra#c>=;4{5usq$4u=sY*QM(DfCsCtYvLW4(tt|jbRKtg+MvpLHB5TD43jHm
zLCe=vCbFn&#fYJXcjsI2M_r;(M(<lusYIvSoi9Ewy-B`dW)Bkp*8POC@Z0n-U1?Q%
zsUGEtUl!M>KpurB)Hhu5%V0_uB6>!pFfq7F<#w%Cw*NewlD7nou7a0gK*dd@g$<b_
zT_UrLs31H1L}G~m)}uos(?)TePL>p86cHCg<2^eY@*rncIbX~E^J~SRpUOAPoKi|j
zOg`m3cv=>iIP{x-0_PF01WvW0hh$&1=GYM>!j%RDLB3Bu)zh-ij|2FG-VY4gEvmFZ
zOc9xeflAg=BGGmAb}M%;mqFjTPti|B9SYq>8@=Myr{F-B@pT7cftbZOmfMWMqWigc
z^u+iW)jk!smhxV$?Ye^vmp>ulHzB?yqUNb`2Ju%Bv-qR*g_Q$VYnK|QhIYy7^U=}K
zoPn?``-EIhObTn@!a8{GZd6P}CM>$pJ%l*9mQn=MC?m`!Fsds&`9z#<r|ae*Q@?E{
z(NBCVFbQ}^???<ntw=uaGJng4618tRLz<F~sp0in{?MW2F^x(1J%0lt0cu(ch`t-j
zJ=Y7xMt#oukIii<0=4XZl8}_h1I~k|5EcoYiAt9igz`sHJ0RXBz+8B#lhqf6Lomkr
zbx?JJB=0HEsp>_7yPA~Akxz)fbgK4|`;^|lt<nAA!->Qjf!mGta1C>uBbD;L3g3qX
zUPXn<w{S|U=Cf3pwSb=A0hPe7umv+o$uMO&jXKl34H7xxH2?rjEeF>_HpN=NEKa`4
z(A?;+FZ#yHw5a5Ixl4RBUuXBr3poNxte1f*<Lte47RSd?4a9rxLB*}r%0rPqcPbKm
z2WNo#lJXCW^W&xObu0#@TGJaJ=4F~e3^o^u;_DuM)J1-QMiL=2+!o!f`hDTjcKgh#
ztC_C@{0!Zt4LSTnsp(vPfl5&l`a_BpHPwA)#p9xLTus#dSXPjQ@ll{quZ@M&td_o<
zv%Ifv^#8thcdGj1#1qYvCRnYjf8w7@P9%eFO^ZRb9O-Dc$A-b^QAU=^iltYv4^isf
zLt~N#jX)i(O2KVo=nfReVpeYnzOn#hxikzS`I)U@X%{(}xQ&ri?q6jYp9lG9yXS93
z@?+~V9gn*_2I~x+icjr2*qB}Bw|=Dh>zlQ<;#6u(Pn48%Ox%HFfAFzH-sY5*)@wh_
zGknoExge*F`zW^Ms?KaF=qOS}b}Ij3uK*Hu)eTA!HEm>)T|ZYI6sZ#y#}yCTSMPHC
z*Ki;Z7R>=3|MLYUL1I#FO-6q7<NH+OozwsGVIP4n7Ev=`FB8`T--=NbJ7egAnlARo
zMf(@+J16OdsLu|~-bqO%u^v1a4GQIa`*$hON8-J^`voAvA_-vq4*)tL!4mR;PI~Wj
zudpI9Dk^H<A5vDw7RUm~k=+p-62QkHp9iGw1xv`)l5G>&NelmT%DeauZ|GOKt~bZ`
z=L|Y+bO7NW1+W!cA%oq`T-@BTtw6vgXJ=Q&Hny6r%u>d%>(*}yGK5jxCc%mL=4_|g
zl$t=bOZ&V`z`;sv{3n*^vw;%VQz6sqd?z=x+1;uwXv3&Q(L1wRjKKT*(Jq<GxjKCr
zP!n9P52?2X;?b?FtSpr5r<y-n-NSt0m&}l>zx?Px4*r?FQDbC|4E|@pWUJ~>HlZF-
zNh;4W>xp!6w=38Zho36M4lk)W>I;F~CJt~?d2V;-F5m6<<QEocRm$ULv|XjrowsqQ
zHR?<4j~CVUDs9fsdI=?f6X$t6-B)}5_R#nZU^J5hGK>~>i}}9mqXmoJ@p%NG&H^f{
zCq#JoC}7kIkyc0T=5s72lZDYQaFRm$+ZXOGcWh9m&o0UBv8AOg50?moa8L6W*l!7c
zLWWQ9WREq%AY}&hvvbgyEP;X3-7R>%{S&$4JX?vn<B%CXI&-k+(^VoKw~Hc$WKw24
z--d?^fDoJ?jvBTA1}JWHRQBxVCK14rgL@MNTdO~OS=?@|%g9pBP-Pm82C+CC50k>)
zv4QueN+{9L(6+6c7hcdr<j&4cVkzVHT#Iv!`D`V>NCaNdj@Y!fm8O6fVzxx3PO9Qw
zkSzy`^e?~eMr+J(DSwQ*BN=AB?e6EZ_5L>FS%>yR@3+N!fawW|jEt1lgodpIiVXE1
zo$tj!&=PtD(rK5aCv6Yqr8@1}8kl3F1u}%eVPUo1tZk=1k>=?!!1~tKl>}dVBk=Dv
zMW64sQ>`|~L<yy)ro!X^1|k{Y8eZeF(<u?M1ADTEXq)RokyWItXK;@xItJEV9^U}^
zPu#vt9F@xw3XZETzoTeKm@6k@R|j=g1i9)GYbZD#zJANZH(jbmy436t3ozhc{856<
ziGbX!ZrTNnB~5;OJYyj`8@NDyziiWsZ@V>sxi?#tQ_l_n`IzC4B$P7g7-SR9=lzrl
z4mUpjSG*?UxxsJ!$7*d3-T|RNtCM@KbRdc<8Z6Fb!yyjFzQ$}R_%j%6azk0t5KF|Q
zy3XI&?I*c<cbK<U))<cr0D(P?PJ(Jtl9e<_ShGss--3=R%%C^I!Ko0~fCB)rd~_X{
z9V}T81<>+lCHYGc=0kE&Glb}yyS3lw$RVlxtIiN?4+Y{1V}k#T2HsrYj+`relk&wf
z#h@U0P>n(qHHPi>5RS#3pwBz1@MJWDEkQa`x)5Li*JM6tlY$5c7{r7OMp`j;0PW#v
z-hqvRLd2NE{iO8<4=2IBI+^eK+5Ypz`o+d+uz0ZTp}rrPti(_xA!9}Ge^$pT1<|U5
z4vNi%;&JT2{O5SGJYDBE?<_3_#(7Gu`mhkaCBB@kcVE>5arOlhQma+xW_H2Qd1*D<
z%W>G1H`!;uu`Cd=X86Utp07_ntcppiJc%O$=pICUA|bgfePO0OEIYFc$BRu0y>vg_
zCu!ai9*D%G_`o9Vvyq4oYkgfZI_h!9MzdD<yuQiX#GoFOGx1YsDMu9Lh4s!U_?SmX
zg#R72sMM9icTg)CLjv<0-AZQvRQe8r1OA_twCKf`CE#!WEv*j8GIs4=wdV7%;hFO^
zkIB)-q93p4RI)@YND;paQK|bP0(l)vm8nc&RvL-}3~`p#C##V^banp??EKZ>wSI)h
z>j)V*Xdqb2f#j3G*&CCyvHzJXFBscbkgEM$3l11RXfp<~1OqpLO<Gird<81Q>Q}^{
zjVeKq<sX_N9AP7`G9=bHyJ$|(qDf>-93=fCZ8eI+52dgTT5iiVhSq->Tc^m1`ppoF
zs84V!DWemdkO{>jKblCWLU7yIDD)3&)JP5gQnh~Qc33vY77r&4<h5zUHJ@wL7Kp@;
zc5UV%;vITYwltgJ4@soblOQy=65pxl*|qQYPF4iTecuM6jL%&-QHK-2E84DiA>hxL
z%s2g`DgSGDjT{I}&|qZGV@sN{v&CecemfbSa$kWATxLbblf`_@4~9_!S@P!Lx;hV)
zCD%yNHXyWAU1Xy#p%|Uv@7Q7j-r&w_U+jz|QY++({eFI8eUJ6&zhwAL8K@by=tpPw
zrqkjW#XZ~xLog8rpN<ai-R@G=_+`QX%lVnN?Rl3?lkHtK*K|X{P&$X;vE_m$Me9wy
z8;?Q0Vu`jOt>gB3Ge883@pHLce$mS4snuxM^(E4uOn%8kJbvbteU<)qfp2grt`24r
zj}{s<WHxyISt&0&yhBO|h?%~G?0BrLT;<=5Ih|5Em-0P1#d+{PC=(EzH<oHQX~Nx{
z<i$6vBr?QM6NaHPDxF^XeqE0GAj(^C?QXyH_BC?J>Oeg0R#8EEOqD_JJ}28Xg})J~
z&MX3tBYWiOsA0u?E-yPfd-JD1lI`9&xygKOQR~FqpSTWy2-fI;{a|-B@Nx1RJ&Vhl
zbg)XDK(!#J{XR*~%3I1E>5PP+Q|qR8SQN(uGm8?GZukCS_`Gs54}<$s=i|61yH9p{
z(dT;w3PzKKa@95)A{OynWxv{^!w7k`!SOWlr4hO9?d?NppH!lP5wS$N$p5O6&KbO|
z9e9HuzAw_8ot<%_h?tw37n7%RI)I<=b_kE=u&}Wc4?iBOP|pFlBOBVS{qZ4)ZqWt&
zylAI<@7Bf-lFc0rhQb~R-dKrl=xrg(8;px@e)En(LNzwi!y5!D?<VM>GSd+dXi0wp
zE4J9T{`NSsM8uXSB+M`r|8huwRx`pA+e_d|mmFHcuvy#_85TBK;N<K)_bJ}uPr&8u
z*OAO>yctZH-WOledQ`svbCXIn+U9sPZ)k=GFivp{TIwkutjf6;tUX;HZ*4R(03<MX
z-4x3WKsnBiI<4XLnA~UIZwR3Y46(|C)Jk2;OEGAZ-g@JQP_SnMTM`4XwO-FgbJZrk
z%Bzd^%j555CI)oBx$Hybf!Z|zIoaf^vS?Hx+)JKrW&7RHUF&c#&7#`@aTE~ZB!c+I
z+Txpr?>f;Y9?RgNk&7jWX>9-e*`o4wx>g<-lf(1b4Q>Ryj4$GMi29Zeb=l8Vx25z&
ztle+HVbUb>>;VW0`ZP|5YR5GXcEB^T3$bb|{A9F}nwDo8%xBap5mVR*v&ZaEMmJ{z
zc4JlTAZY(Wzp%*g1c$5)&P=^@H;l5qqa)4H_R!*7acwtWX%^OYdBQhYs!x0N$2pyq
zrIzCq^6?23|IIH*mdOW6jW4fEaClRxOtO-ET7`i?L%}Rgz;d2!oTpl)|Dh$DSTKO>
zV4})El%H9FD#&7?J|2+hDQXy&t(ge^gyCXRM9lozyo5AWIy}U|q7CBXMckl<UGCs_
zUmM7TwZ)fs(dmO}4iWSOWot|2s}04QN<sJpfNjHcvQY0+t=*1vk#bo|{k&44EdTNM
z6CU?NK#QGfqLUPj6ZW-5=!A}I03_;xY~j#cm#ckwzNJ6`wa&$;R@WQEf&45_K!48z
z6sgZixR%VyW7)z-*FW(3QX$VSFEL8l@>2JJ?U$!qrY>Kp%b#FcBf)GIZVA@wPZxz&
z(@;q<M#1M){7qDAHp4K?GiL)4#Zav7H{x*<NvBYrpAK#cRSsDvrKjzSTA-m#l<hTU
z%cEUo57}hCCGYHA&GJAveGPkbd`vJOKZ)17`w^x0bkH;+XGw#+(wp$Zv@CI-41x0|
z8pqdTJY<s}x7PglTuzlH>wn_O4uMSBECHY9G36gmb3kpE2M`4~E?xu;`T<=}?ERc^
z7As|jY-hO43)eifpGZZ*P&Z?ocGAr5sJDkxDGA*#qaOj)`vvuyjAnY*|MV|f#A<Bo
z4#k$(`IUAR2RKYB+R6b6Qw}f><lD)uLfFbpmT9QeoL*c^+@x8|*N%y!+2HlS4?%yB
zqbiVnZPSj3^ZrZJz|e`=U{VFH)hWucg-WR^@!J)e#>gkYSyF38t&bdS*6Gx!=cbux
zBUJ`iBHBj$&1lFCV5HNzimp%_@88L}1f&)?cUs)Tl=qPJ9cY<JY*<ikJ*`}u9;`Ls
z3K(syg`TVZrr5`(dSV{!<Z}4Y7EC|+0PbhiUw<8ouQEUk<5@an)6xKUL%@km9Lt+V
z0m<|D<;S0@0EZNjvKrD-i~Esus=aL_vZn26{n+u_d^9i`os9^;+vS+(7C^6m2(-{J
zYzeFz-Y-VTn~}S_7~!?h%5XauzXPVN&gU%#My{PIpr;AqQf5+lTnv9erVXm+S2*l=
z>oY63$SR1iRBw?g#V;h#4U2>|`zwh_&qvv%h-f~_&^eGHA4ez%hD~cfK7ksAZPIu>
z(e?Kn{NB}^dJ`y~6G;IsaK!d2RgrP0!KSU@d6<kEK@xNn9|NfuWfI4cxGrjkek7ye
z0XD{d(E4DhW|QrFnK2+5-WQp<RQ>oYY56}@H{dAC-G#T2Xug(RImRTXsi>$tNY{3J
z_&5unhm!CED4B(k@jepuX(G-H#`!p^zRPmTqrAR<i$wfO*cNBwJotp)jS%3{4^B3L
zr6mqHcU%;5UGX;f(b7KJWY@cp$OKI7p;^|7ti~V;gI0zyc7TMGgb)!t;G<Gv&!5az
zm<)q`I!5|U0XzqtQ5u^@SsoN<%j;0585R~Vp$O7?Vz&06MD0jSm+Lx(4QzxaQ+BzU
zQgIN`mh@IdYRr7RoRFE1hqG95ze&TBbKoVmu2XAsYg!J}7<$1^fKZeGIM8<4CtmSi
zIGvk5fvO#DtqP1s((eb+90bF!2pq(sBXF;T&kQdBN8FfB*GE`O7N_cF%cbTx!87iE
z#a=!hveoU9^&PH?CBpeuE)+%^MyHK$XpSQ_6Y$+fO*`omE37Y&d9k8;K~N7&9i-&<
zP&ySleCdt52W*vLjU|-wjEKI;A_b%Ep=6(v9w1(b>*{8b7LXsrE^AiG=PN#!uB`M~
zBa-8sg#a@|)Gui()Tbw=$9w1T#cuC+Hyno}vHXk}9*bQ~*pae^WknWC#%Cf1=GZ~8
zi}!_v0Ed=TNSDqJ+cI(33r!ZafKaMl2VVAguz1>n6|T@|uv90;0}On?2t;3aZG5~)
z$^_ZN%uIgo7uD7c@Op4Q%)MAaj<S86!U(usb|plHU@#;MZ-kk+)(KCAKqf!VE**fH
z%yCRIgGY-(Dkb+J1zGI!()0I?Cnwy-;atu9RJU95kL)k&>Y24Ecf(EpmR>h`p|sOP
z!HX4zdTBquWVRJ74$j!^!;%J|hL8|guR|6pg$XzTwX50I5g_T#T-E47`UwP&8&}?k
z<9||)AmFBv#UcdMK>H2)Vc$s&%O>HdO}KvKv!|H`s})=xLpC!(O-h1$eZ0m<{j}3A
zZ()p;lb~2X{UEx0AcSL;uBfs{yKZC!5{xrP%Q92}a^6a6R+9r1WG?~AN4@MvuCAF#
z_IPOrFt}yFdjDvT!N6mg-il$Fp2w6|L-<Lad@cVLF4j-Dw_jW-##aE!IUZtMOR(N@
zi3T7KVnO8K*>sF`lQn?vt37W3+;jLt%Roh&x;4rb--TT;#l0OX!!xGi=QQ5r@#J#E
zzzL-3yXyt3x=^b0T=fmlvO{a+Y2A20eA%V50{n!{I3Xo>@&T|IAyh2A{K1kkmnU{^
zVG@K)c?}{2l>dXHnIS9KpSoM)0P=ZGvLf^zNhh=L7Ww(nB{ue2+w;kR-<^*u>b=hi
zEtQH0V9~3Ke}hO&P@f{11vpqomHGTEi27f*A)PTzxLK|9mnxm2*2(3FGz0dYIH52O
zqXCSxRX9Wq7oXBXv3%b)F?MJbc``|FJ~=!#)3|yufi*Xv{IkbWVx=g}J)P&57omy(
ziq`4$Pum|G)F?aX@{w7(q;O09tQqSlWnW`ka+k+-dUmfY2&$pS=4vMku-vt)3cKzY
z*3(n7i>z-Kt8?}u>)`(6<>Q)PrY3lkO5z*7r034Xoolc6@MGLjMwb*}69gqYnHfd=
z43G*uqn5cqaX9v4U$$$$-hx(^<m^2LBIzQ}nLV1!yEA{^w*mQ5sfa9%#H_gUzi&5C
zOaRwdJ=9OOSBL~{xl!r331VEr!Xh6ow!Wai_5ipqqF(VCFoAKx<p9PGP7NlEIX1It
z_~nh@;^0J~o{x<`-*&TYH6l<x$2StLMCXR~PDu$xIn^CgriVA@9`@Y)50aeY>u_>a
z4Gi$7WmS_Lev9xhBk{9sisY<nUArpUX|h>xk!!cZ_}2PB9coBFW%$ux1AS}Rg;>7)
zi`V1I&2O<7m-Y+8q{R<gR>|`X(el3@#}ye^>)(F});Z7VqjrEU^6{(d>IVO;CSTVs
z%Ws6uz5XPck#q~gVM7fDQ<INrEzQb{5mlgi#awrx%fP2ctea5936kEfv$G|xN{4J%
zhkK_|!PubHnaxb^Rms`)rxR7ZW+~@y62c>m>kmU*$B`4n2sVid!SykOKvjLe<`eST
zTGdQ_KySS}7IdNaRjHvzCWA6Mk=*f`>gto05SM9f^K_5rzq1GiWMuI$Ohu~HYOrt2
zJL(OnvoyddcZ190)jFrT)pIcXnXqrssei74LiLSVAVy3^U_*>RoA&$x7afhX@yDgF
z6m)smF8Nq7enTpOHw8Q@cb1g+N?6^vFqpj*WZ>QIUEJo}e1ar;vQ-x-F|9Y<#p1CP
z_1%0`f)s3%rau$k=rcvp!7<PR;daKxXQ9DrrY>4CK+djOZd_IZA)y!WxgIbwcM*|%
zMzx>_82uJ&8_LLGpy%gGqfcI$0N}g5%?~}@7&}>SnGxSCNN`*Bgs(K;HrsxA7hnCh
zVO`=PLgU}HavwU(>JvH+PV&A#9wdbavy5h+RI}K8W6hfHjr?)`1%VaA$nXk-Muhe_
z*7x4FWjfyl#kl8TTJSgdNfz4qYvYwr#TaX*aD&fnUAaO>G#d3m8O>|?<YU%r>^JCq
z=QugDdUb#isha*>vN+lyvj4E^aR3S-1N1B8$vn!@F*joVJ3!P8A|Ze_Soi<PU;Bd_
za{zyxM)ytaFEyYJ&?uvht6)MT=?RpenTF`bWE?VH)LyF(>G+K+KECJfh(f8wVbPEl
z{_g@PN&|tYR=+QbxXHm;Q-u>KX!4<{Y64ri7L9CA;8@NgpIFS6nA2K5k7vBvDhx4j
z+e%l5GSki%x-0vxR7@iN;8-$2kC>j?++QXuyaZdo%dk!rsTmeG?(fz;Gf1j1xGGK-
zH34(jRHQlsaWfvLFGF1uVXzb^Zwvz_)>O*URfolrAf2`wgWCel&4jJC?iqj}lMJrw
z2^>yh{IoptIXC5Mq>9<u|6$Jf9MLjEO&p_@HhPLm^GWBGVzD`AMr(*l_Qul+@=h+V
zK$wY{mQ<|FP4dSr$4VymzKU4FbrtgTG!5^qY^+w~7N0|%Z4DJA!;`xDRIxA}{IDMu
z-C$(lpv#M=Ria6RtKayEhEK0!X=)p_D742qjpnedG~Ho?DwLn9Eu@hp|6!P^5k`HI
zVuE5<I@2!GoAdJI?nuU6=Ut90U2fg{?2~KN+i=G<L*NXvMY5_bop@OkmjsJb#i)Y!
zKaUS<sPR^2KzfJn?Nz=-KopA+@R*?|#d+(a1#GErLLI|{qHdyYCHlc%5!ca{XxWtK
zj^{&CBWW5X|Dn#k@kscP^pU+quqHCnzfkAJX;6~!lh^KXIFywxx+-(Apd%H=34Mrn
z7?WyhG1+2@oq?2C?+#LJ%G7@01!q9k7m8{Dv>&ieUTI^F1bkKCus`bkG52phW;LIT
z50*KQ$MNkAA`R`*w;#iKYI#J#Dha|=!Ho2a)GOt@H0jB5ptQzo+GrcwW14s&N|G(v
ze=6&!68IOt&gY?|=<O8Q%xA!mU_u~s0#O}Tu3AMhg~aAf+_%mB^?30zV9f>_C`pq#
z2_ZmTEh{<YmtQoMDe;6^QX^IH;+T0)1_}cU6;vXe`bx%ZY|p{93cXvp|D4EL0o^N}
zuRDGRZm!v(1i-p1!+DvXPk?%IoGX3Vd!j%_qC&gXyaE6=eE@O7uuyL?QE)t4B{*HK
zmAcHtb*%rw*PcRlm29%f)K$=IG|$Mv9-=#FuAh7#0k4eA&`M+@VHqZ~h>(vUv=*99
zs?Ik`{Id$kqHLgcJaK4yDHbVuc}yuOcV5SK&Gv<%tgqte>FG7M#Oy`rU*&hjn$%d1
zq<u1WFS(IhE&>c6Me~5sFSdACTM+OQz`75c0jh7a2C(_%9e4qp0Id2=WGDt@QQKEA
zDy2{;vFggm32M-(!HIz^D@DI$G$Udn2+I--ieZHF*FkCa#Ah^j!!~trjU9i{Ff1ZO
z%nV?G0{fU#`W10~Nk12ViNm!zuw09lzI2XavK<VfDqv%&lm9xS^|ilLt;TRTh0SMN
zhTHWz4q%`pVkx9zq#SogiHp?hieD3OsV06&VKE9hauO2@da)?Bw*z!OOoi!Gv3R@3
zll@AD$9@0wOmrH6zXyPND5KcY8X(vmsHx-pQh;<zDp)0h4ARA#!1GJ9Eq2QDQlh~k
zlE^S6g*G;i6QucWA|%<ub-MTNuF{y>M5EP@CO(TLzL~F*ovKu<d}IGs3|o_@;2(m(
zTN1so@z7lyo|o40I6=cgsYV^NzP0#sCvRvpj9T4<9tLq5sAabR?E69vEWHul?~Np^
zz~p<+=k2zgY`pmbDS815gHa9L&+)*Ebf<L$_ty?y;ktwjMWPXSV^Xn{G6{e;Na48k
zMtfo_K%)zvgun(Mt^4okRV(@Q_4O6w*{zza*L+}Jy^;V7sNL#jALkI<VmDyQe`9Rk
zPJSCj^{TV}6{V4<6tBC4v3f+Evwad3?0M)=R6<D?R7;j36a1g-0=TKH!}w<JD<%%#
zeqF2N+~#5|12AuQJ|0Ek{jBv89+R$@#WSD+m<kWSw?CL3J_G|suUzd`m+CKLBpRs>
zJJF&9RjsWL$KSVh$8zd#_KrGORP!Zq&Mqz_rt7|*EUR+}8OX{;#nEdh16J?`L-*UB
zIw0ydC$w8!?=}iKYjl(G<mjX8u?d~0G~vqhom$lwz7^JJz+3r|Q6al!`7Uo-^q&X>
z32J19#5dmEj&`JUu)`#@VLsZx`RMbu6QD&3I6Heqr+S}d0B}oIRW-vp_3-eJD(HfX
zib}M$w$?vCgsXZNrdk%7e<ZR1w5-T#yB@I}rI;QCUa7(+2paO{Z4v<c(r&~!t0u|O
zLRF+^Nk@gUod5Hf7G-DdZ){xqG65BOOz?UUtsKpuNlQsWhx5Nf7tlvUB!g;wbB%rD
zMPgDuHQcMi1?>X_gE8o7q4(At{<iQlvZPe>azMhoU{}IY=%Ra(c<MKP&CUTPNk=LF
zE0(;D?n#=UU~_=?s0)5zpa9x!x(WFP_SUB7s!c10@MaYeZ>icH^2i9jqf(=I)GJ5m
zgec1B$)`uNp0skc+NX6{W0H#hIaOk*d{qoqRE+YUmQ{)u{XO%PO)`>#NrP#cGSO7;
z0$$^g0Khg{DwTaNmrck0Dn1PNgY;K#mf*vD0HDNsKHrI2em`R01KI{0*@Ry;7QXw@
z=3xE}m+)lPeHG~WxWfBPSHrf!eKo1@()MFAmhJzdU{X#nyEi&;Lw>l@>*R|whM9|!
zRp{1kUOa4p=kv)9;0)j(uPTNS$=EyC)O4b|6e`1uI8ql?r-o&rWDOJ143CPiNa+hZ
z!2)m5OyyQ)Bw`?(CIA}aQ)qaYGLG{(E8xsLBDDJWvo2F6gXcaH!zURKg)NY6{s@p9
zKwn{`O1Y*uJGNykum<S?gdG>wtg3x$=i^uQB(TWUQXXE^b6z$p3Ff&KujZ&ehs&b@
zeCZ^8x~7<uxZgM(iZK|qH8V?zOQ>NIxEo`!sO5zKO_l1U>E)NK=`VkW3V<`^w0XTx
zPU0<Dre1fs78C>zG&Jeh3*b!fmL0V?UtG=x@z)NdbGyQ9c#$8!$|4d_%WDkyS4WT_
z#KBH(>8Pb5D*hvvW)@LOUTf>7&fQ){?E1l0)&{!kdm7rDlXB``mD_js_aOxZf%j{&
z&!;ps%HPiA7*Pu~N#iUy7^KPMpFla~(oX|<TRXfnVxiN+@r2)lTm@=xDSG-etlH@T
z7i27;BPMU<e;58S`Sfse;6c?BLa7M!$HW2+69=n10PAsVc=$(87c#1Ky-yJJ24hCl
z?8Pf`Y98elpMlV?RmI4qep+Hg#_+j$MC0A>jc<)j5<YyNPAFJVL083yZrT1moOM1s
zBsHIsj2Wo_L(u-XC}yIVlf5{XohA9(DZ5m?O|m$SiiBU=ay(y;o&(KAQjM`<_<XJL
zgZ%^k@czKqMuFkUSq;nAW6dU)t6~pMe@>2<o+X*07hnDNnJ`Klz>=D{(Hl7k=o9uS
z8vq@)qCAV)wtt$@J3$*J@z2utZUJ}HSO?vzn2<uU&F|oDp)wTv!o3-Hcx|)tS-Y@A
zte+4FLEGLV%p~mmE+1$qSqV(CF@>!1$SgS=)i|YYB^ZQhZYG9IgZG6Rar_9lNr;UG
z#na%zjrbxTa_y6=&xd{TY_Eq*_LldM^MbT5EZR$-i6o!(6Yyrgi!QrgU>Xfej$^fd
zQ5uAEFpntY<NzQ6cCLm8(A01k<9}|PE1w5$W*-3nid3wlv)AAcK<=cU{#jO!y!g+z
z075ekq^f#!nms`1^=rR0N(@eS1*g4;7k18{7rq43I7>}|ysJ-xvE)Z*Bo;39l{ms4
zc-2ThMNr-(j_i`Sxjz}M*H(~m(g<YS_lI$eHy<61pN#z&ka;Zb66)Q(PF#LVnfBd2
zew^6IDC4~M1FY1GWaqS%qYg;lFNLiA(LCFdeRtuQuR15im-$lX!`Z6)qPT*Bg3PEA
zZt-dugB@M5!q&XX@P9-_OByHw0)pl?;`Qm$3c&J_8Y1Y?rb-j!d^mA+a~;<tXc;fE
z75W8Rfl92ieTinpii;vWnJ#HJ|JQ{qqFU8DIa@fdhZ-5bzq?^CuZ9CE^D`<Rd)b4A
zn^M(8iApDVp(?Y+XWA}f5n9@UD&&+h9_nz<(@6xMLp9~OCcA=PQT}u)LE@4WrX(1N
z+-G;twzuILo2Ls`TZ<^9O4y}d|K-JwQUZd04R{oGhg5CYPx~s*eHb*ejv7C+-L>^<
zYh!B$)c&!V?4xT>Z?AZ%Et-FXlFC%(iwe=g>=K6|3tkH`Y6k-vME63k2Fl&F8Kd|f
z6oKv9BZu>s4n8BbFy%VjSVgQIp|VL~Y6$~X#cyCv^o&xXlGZCXARo$CXmc+R$4cHj
zKb*`Yh?AA`pW~L0k&}z<7TkazyK~mJw+b-iQGyp61f+*skF!mWM*i;nB=QbJ)5TJF
z<@`e=e!_?WPi-DWBVsUc@Oq&%a4H^mzZNnapT-wbVr5T(lI#(jq*$aaC=*yOg*!b^
z?pfcDkIyMScaB*Lx;^=IB%cLVQad}4HyO{)t7q>iw$B$uK2Eg7r0w4d#kQYc2N%B<
z8JZd-k<*iJ6?yah%62s+<Z_UopT9l^OBO!br3r{N)lO&nFSZ2?I#nhdpFv#X{=AiC
zg}rPgcd?Sf6agdAatxGqfobLAf3#GuPMB)a(4x-R8d8KtFFtiO$s!`oTfI$X*fGJT
zz+)a~VS;Dd90p%+&T#%>k+mVwR4}Ac1yk!=?hRy9I;ReA_Lf?Q({{)q*U7T6IwFPI
z0=(Xl1qlWMULss}^U9|5cazX?ubIB%%xmAfrf8~h2v5y6Zl9ZeQlkzs5_x1u2keU}
zje`FGY#d^s5FxFd{lJSVCc!VM7t)WC=whcO=mXYPs8J+VyH_;B;j*hC(i_U_`?$wt
zQU{a}>9wVx(m{h<;qKzMDLl&Eqw@joU2y8%->mAlRTa&@!6X!uq42tenhcz?S6Zqx
zT|RtYKQ^nQqX>4@Wu^&K%O8aNUuz^__JhBBX*j8N+Y872knlnf4HPzk*A>yM{Fr|z
zym4rG@zGA3-vQKE`mdx&7W}wc1SU9U3wGyicsdvVr&gn}So%DVxu5f`lUiyJ8H)X8
zZ3}N{X21G((Kd_;?vRj)Oj+3nIAOpi=yOwsI8hB2R==Yl?+F#<tBk|1pT+64d;O1C
z|6;zLEr~%z-@dd-V_O9*NNFq>1K)5WSqfe!1@kj#HIr4eKR4gr-|tuEy&$VVR^m7W
z(7Sgu1u_}Gy81rMP}dS|=Zt(}VV9%eos{L#ao}sy#-{ODrb!MDk<t9Rn;qtwDcyXW
zKj#hAy!YZ&z*{ASeX73R2;k-PZm<XaMjR&Rd91A_!vdkK@L|U8Hd89i{RlJo-*%f#
zcc@?l=4vx#lwc&@90n0s+9qd-aDt3b&oeV;!f!VIV>Eh1rgAx5T>y2Qr&jaXhjZfe
zP#x&y0pleR0lpTQyhe9`td-9eLIAwoQq@M8FgQK)-@jIv&s8^H{8s+YZvhCu*ur>t
zc<OoN<m6sJFgF7TNxjU!Fg%FGNCcRgf&+B`-75j^Vxts?H^+e2e*doh;rJ8K@zORa
z>r6}Z(qL+^=)99w|2#Vf$T8X|rE!&r`Q!kXln69F&LaaI!LlJmLk~$P3|%6C^R4u*
z)@nsNai8gEQJWMfz<D0Tz1YsG7K-~)waiQ;NHA*a0)GwWu(B8by#7_SX@r4un21fH
zzxg;d8<uX`&>rZicSbEI*n$`j`kDdd_)z&rbG)W75yN_m`9p4amo`s1cY5uPf<WBi
zbA6)bY)zNRAm9eisRa!CnIDgo*^0Af`Mr_g;YY!kn)J}n&{sSZ;Xy&Daf}+RE_FZ^
zEYIn%A1B%GOY#-Sa|4Mw9@K%z_#j@T<=1y)M3rr9=CiVZnSg{!LK%>2xf6K`pZT+Y
zyfy4mIR|)FL{p8xz`%o)nZm>`Fay-n3%TOxTIk=s2eN@?=8=Pothk7`%mT-450(>J
zQQ!6FYfW5AMG2gRfJ~+LDPj2+P;)t8t6~s0I#Oj>Gy5aa#Lb*-^d1;I?r?e^m_A>z
zA8*+}1xc7|l_j~2H)MsSG-#11h6^;hIQ}*)lXNchjzd9gO!z_(Si#E0Dc#LPAztOf
zv;MAN2Cq$5TJ%4$!CC?I@p2GJomnyeyWz~q_Axc3vP@LwyR_Ik<LfsuuU>Ru-7?nc
zPpShRzn{Loq)>u-v1k(YJzzb=FS?$zvH(tdXJDaSRROKgeSi{dJ^_>|X0*8+$if$E
zfrQfo#&gw1TL2)ayYlzB0VK)+?wiG&g{ZvczHc0ML*?nX;H9Bt);yqr$|bFG6_C9I
zZhkPUvShXBKPPItP$zd^lvK1;hXb%&MjWW~>&_7+inzV6FqB7uPCiqA0JoO0(^9pJ
z>FPZK9R*iKFK#p-aQ7QTawpKZwzXDSC9tcbUhGkpd3?xK-V<Q)R7KjFcRs?#LYx<z
znC_22Pt?lK6b3wSXA*LB78x?0$4+YhCP%Dm(8a*uINCtv{9s~=!(1|okG&;`u^q+6
zae|F;KUv_6%y&n#?qQAK=V*w<L?+6$KD9=AMs<+$@GM^wFyx|eoq8R)16pt0*690|
zt7#3+ek6=Cpz|8>ejyex;~Y3wGF)l_&4!07o~_y+O4CJ(RVvEMqGgef5Pq2fmX!Ku
zJs|Of8=V5~1I#WQfiZtvY8ZFxR_k{6jqne-fbdZ7;3EUrc`PP6o-n}dr8#j2fUC3R
z7m3dCos&x+N@PRSZ)%D4d{XT0iB7EmfT4M^q-PXdP;S!;m^g$2RcLC-DO5|M@$x4l
z|6!Hb#D$W5j*GD+6%PwlR{CC@E(kfzvcY4LW(9z^`k@Qiq|Cy+At9?HtLw23MxR~m
zUYBQ7DiZ)L{<LPYToVIu>faqLk^!3I5zq#j8RNtc9+P@$s8=*1fgjC~(qLH2M<lNX
zObhi@y7<DlzXG!p<wYV1%Cxj@4(CX%jr?;R+^I)elUE8ToLMLO3TBYW&8uzc@n?&(
zg{Z$O^kI7vChY83%^>E`GT>Il^fHKmH$4w*y5D`I{KxL`G1V)zpyC#*ENQzk{aLt+
zRjt4L{dZ3Lj^a~3IiFB!{nrxlG<iDY5(#m@FLadg@r(OB6h#9KM)ReSUwo@Mtd_a#
zRIwPfOSPEBDlL~d@qY&<iBK8wKA!q&ZDF_r+!N`GB9y@2#Z1yKsSG1-j(P!4ehQ?H
z>Nfz=XS6``JD{01td37XuC)Fc6XeH*+C}uU*>n>K+3ItCsnlDXkm{%vHGMO#RmGEC
znbYyPno;d7O}jd0@(Eq>55FWbMy3ZgdBW`j<Qns~Yri#v$oXO2a!)-iD>=A%cm8m#
z8KW1v-8kKSy%zLI5-cIAF*#FrFO^c853~)M_Y}$GBYp-rrek!&Q=xAJ7@w#j#E>rl
zYUv=CP4?TWb`y{w3>oreD@U&ZHQ3Dm5a9T*$bNcrT@0~~H(2Rdcnd<|c9G_9>AUpe
zf890wP{|~n#;KHHpsB_zYHn_etr5G}qZ#Ocgst9~L~{0~$A=|R9_{^;G0;?D@D}g}
zXICJdL<hYBv%p-C1fW#~iuGNrhccSpC{ncot95Y_B>`#D=K!g1m*BAo@ZLucd2;3$
z80Pa5{h$XWRJ1T)lj`T)7$1;bx0mJ@#(w;2&5D5BlCai~IkR)HMV*JJC6jX)!e%Y~
zpA7Gvng~H>^-?gyK{WLDA*%y@0N)wH`Z)`n&>hWN@ovw=^YDm6xsC~!ausWS$g8)0
zBbRL^AtgPcQ1#Y_mmsK)d54sejIAsb=T`_AD5SULf%?H6KKRS_NJa%FO&6^|6U$ln
zt{0~-t#aw8xY6ozuv0HU*&NtK=fWZ8?4gtpZ?_+mf47H|ukB0$ek3C#MQ)(ut+*Le
zI$GQ@JR%~Cm4iHvDij)T@K7ODs~&xo-a(;bWdHz-cAZxv5C^l-Dt$n2VA1=Xqwsn;
zEB3zDnEyYn-ZCny_51!7L`tQlk?xX`lJ4&A2I+2)4(V>BySqi{ZUjMEq`T|6?sNX%
zbI$L1$ruc{WAA;fHP@V<g<6I%uV<O0aq+G@5}4!opspKuJ$Om<$khDiuVz!+a4ttN
zIZG3C3kVStda$_XW7<9@qxHbI3VukWwF(WF>d~b7SMiOBhsmfQGCT;gS3twFm)!{o
zTi=13pW17<b~s(c&wPnULKi7dL-2+QTPlfe5=XJZJ~%3C+^3Q>Cdq#QTNsWHy_n7f
zOitrEH=RjI5|*D4%0a9MOA2`@mC^9SfBpn<o9qrf-^JBAGeWJB)DP5E?p!~aUne8D
zH~;aU-pYX-%9WAtgkD?L>$CE}#SrncZ-U7RpK*tjq|;a>k{I=}X!)Kq&}_iOaf_PW
z->2`sBJ-ry`Ju*8?0;lXYAmdaH6!I!r$u1WaF(397ZqNW9OBjRyLg|QfWhdfinjXC
zRz1H0k29yHB<A@WKHKPwtgN%xf1oU1Wu)}`B5-gzU7EuI;i3wMH=1%&7rK(^&6e(4
zqzby?PfZCbk}6|ma!BBjnHkd#yC75aL0L}&E#7G%?6nxf-~-_kJrhIyW+u|-q?a_R
zasU5b%n|hErg05LA?XP!I06RF0mG)c$y~%Dn|H)bl2BV7hzQY9$(2fT17q=OPQPi@
zv{w=bI5^$O7vCs3{&PA1J+-EWcvrvmALhiBB+zWqirjb9zLPxUcvG2KfOt3PZ$O{#
zG05+CXW^vn|AE)?Gu=<`Me@VK0TuMTx$3J;as&iwQlK@Aqm%<N%wo$0coA^881pg}
z0^eBVbGsgvzZ@W8!BTve$4~cvlh&tbkc`N6efYJfRM3}a1;Sh28Cj*g>)Q`_Qt74L
zTsX+<D0+JJTD&^$6BY~&5;Q+&XOG`y8gkIR!jyE?={W}mJ~UMOS<@O}o$PAbWOBuZ
zppt~XNg8@)80y4W5*+H-*vcR9m$5!Kz^&W*3kk8u2lyKsXFAOL?O~LJVk}-75nD6&
z`5T7J4E!<rh)8_Ftf;%u+@Z+_38X{-mjgul(`vMuyzq<R#c?#?!6sz<&rt>@Uz;zW
z-SzSY$rB8~z;H%R1HI$SD@n;1oxszV;HC>{RvF(gmx0@l?0xk3BAt(}##1b(`+DF^
zg|cSr(xl+78dB`b?QzKs+^qtE*;>UWL+Ts5CG`{K1}Cl+m$I2EMN!=Oaw=5GwN`vo
z%Hesd8E1R>HaLyO>Y&qL5B3*`7$`F*2uMgtpoG+zzi?j<quc~?dad1>piU87|B1&|
zq8f~dZcQ{7H}*?5{V&S7uSGO+@N<dicJEd>7IR=UA*t3=Pyq}AbVWiUi!hdn2$SXj
zo^3m^cxvQ@bOy(y2kzB>v|BBNRX;af>9=*jy(jQmBzzCv_`zny*>XIQ1xxj?keFuL
zLbqzG$cc1kt~x^6r`XShz4v0D>kY+~?4uTgSfA<6ZQSGMMLOmDVmjR#)n6VLBGw&u
z&KIH<D<?KLw`X#j$-R6--4C87OFeQXbN3?am2~Ov3`B+dKavA`V84aQYv21XJs>hG
zO&fHb=kH%!+#dj4TFokG5Wsc>fJtc0>4KAxGP^>araS@?XUO1omF;}IwRDHf6o4Jt
zBMV#&!ywmL`t<eNx6m^n-<m<FND%wX?o3JE=K(_`s_3R=m-NPn^^iJnggM-vTj&H{
z0>kf6B43p@#{UWd{zozt(7J&aGvbUDvgp}t`^_6yx3jYo4G8Iw=Ed>>)a#qz&>M5!
z>rBNIe!+ytYfYHgT-YwDAD|iFuCb1W1{qD(x(U-#<E~aPZdtiSIAt|!x|D3PY0w9p
zj1*5!DZepXZtzYg>*<#o_v9M5wARWO9-OLOJDT<A<bD1&5jHYCJUe@ZEnycKk+rwb
zdQl1}c=;ea1JRlV4CX;-L5J2{9<d@w(*WW8<Ycn{sCfekLG7?q)vp2Ng~3TbWMmIr
zfU>>nyg7i1!{t=`;qr&kEyg`@z=i$sYB#g_1kphhA~}kx^gU|uF!e3}Ve1|PwL)&-
z`Qfrwml)wcx5U>`A)@suVua<#3jF2U0&UI?on<Yu-#>ciixs}R@EK)dTuC(d4@RtN
zd-}VeXBizeH&i(7OnLZxaO>X#Q;}N%ACN+EVrSeE3C4mk;7`Z^EG;IyR4nB4SZ0xj
zI{V;#LR?=yXzz~AB1%ok*NN@0>l-@^0iBCvXYUEer=ahjLnkIONR+B}iG29Lvvp+c
zX*|+=;&7KqP~g%5>#<+J$C!{8f?UGUV`(2SPaz!_ncg^*)Ddz)m<ZY;ORRBVl^X<*
zkp^Z2lX3V2sb<e#4#V^SZKZL*ygNWBO64Jc4YEB%>Fl<1l9YxoK}5TB1C^-{v@1o+
z3I+{<+$C%j4v+UY9tePFHmlWK`tQKc1_}C!4wqqX+bQ|EW#4elMZC%IH?eb1Gaa5o
z3A?TUPeM{R{=<^P5){!oVCy9CF**#hi&kv-sA63s)ENl%9QMz&uGfjk&U{rijv%0I
z<q^81FTLa-T;nQsg|}RRfQ*#h<tz~k!yPo+wgLyJ^#(`l*#K$ukGF;Cj$f?is`T_$
zrq>L(PX**1q5!4M%JiThu3Z@(DdwBcFMsJAca?Qz?e!U-$gLdHEf}WoUy>5{b_+i4
z7yX$yjbylcx`KxO4p5Ov3RUgIC%ec32Z-b*=1;#Ai|vS8syyXNlnc@z-hC&lWno&g
znxrCCq1&qjQ8t<#z6hlv9|};?#sFu&CDx1k$$+EQR5-b8{n4Kto=peH)Q{$Ad?|mh
zs_0VWudAbrG&ta_AsK2&`nsQDy-q3>)%i!Y!6MrmUu@9x(2nz5W$+S)Sv)>m>G8nT
zmM{R!NR`jIHv2!fCM-h64>mc5;7Wxns6pEXQbxj9Yd+-RRrAX>(ksDiUT>#?&pS0B
z2v(xg);Lxf_5O?>pO_QA#;@tkhqN!spoRXHVY+8#DPHs3mWnD<d^4mS8T5nGJk`Mv
z4i~K*7K8SL0dG~urDk}Y?fvCA-_y)5%|fihaUTH4xB((%!~8tNxK{$&l2%LBZO3$H
zXq#O)N;wO9yQ{s)yby@@cuy_+J1NoHFy~e-PMJ1))z2meH&l%bB%6Y2e4l2k>?onY
zsh`E#28lC;m6G}!YZHI$sy>NKjPvpLL#-qG6(tveB#1y_NHJ0GOgssdg|6<ep$Dfj
z<#I(8+L)}}hb$fL%}(y0RC{YoiGjb{mCLnNo=z&T_p9EpeP9bHK(fh~{|)->f;ZpO
z*ensfPGVzV4-4JGgDDOEz;(Ia<E~2JT%x0))%%C?d7w}wRr9Xnp9^ItK$nqdQyeF}
z^ylmO;|gL*55lh{Z2m0*TaurB@Ke+*J?@CUd$uID3RRz1dmkT6LqZYR#xv`47b6pe
z<ow--QhA*(VeVVUOPCz?feO3rOxKSuNy_b-DFwXrfGgnqA{B1aR9_t>cGv)9!CDb8
zdjKzK7lhFSebs44AI3rRUb^iGth>&>Js1Ewr?{eIzylnKHJy|!8!P6Y$+()UnoPa=
zJ9r%g?!gz*s8j>{p}n66zt<h+c%hpEnzkCXm#f$4wPWB<E8=JKBg`WA`CqEc8Y8mN
z1HO6us1OJRaz~C^Ne`2_ZXMfs_y-Q>%s`P|uKerueugjnUk;GzAFum*`y>dqR5it^
z73GnL?d@w6D@^Gts0333`M4fA?#dnpis7vddI(=#cu6B=wbA(a&DTceR)Bc%tp!F_
z67e)OjD5J>i;Lsqi6P>}FJy+&^#a5_E%MBf4xzl@FLKP^+t0jC@<l<mNDyL8_c70N
zFo){Z`Jd@&8)n^zbSB_n&!kyV*T&E*qDMGOY9XWKs=2?rV+7zwoPNPjc?=ZADC*yQ
z@0e`3D@+iZ_O(5<oIuxAqQFEAC$4QK&AS5%&_%#W#HRKq%RMnd?*uETW=3IHx3X2P
zwcnD1IM7jih=ZvYOBi*)6CCS66Ba23n<czO@$MbxJQ6h-@K#mo$ka4e|KZqq+|gWc
zrT8F)BYQF3$ojU~AWLg^CUNvGN}^<HKHKRFLHKS<S&hT{eUIaF;0HV>L|1=J{h$5;
z9p*#FnuXJII39N7lM}|@2?eWboM=1Uw^u)HyClFou7Ofzs5{55n=AY{HNpRffF;rk
zb$o(`BoeIep#8mgc~}N?8b@^m5fXmm7%}V^l`lQto)6Ou5c|PBBxFDYYcr0bv|gcy
z<76dg^v}Q7TEl{rPt(XrHKo2gh``{;ZJ@_(#^T_a%(ar@jIE{>ucMAzr_;mvcxhff
zz0(#{^Cm*DAJ9(J)(X5pJ$zxo1NAvHGFeHj(xhFGFJ6Il7VazT|N2EJv;hj}#I)CP
zj+@XTv_<Z8EO7Frd+D~Fu}nsN5U9sKl0C#?+a!-p2yaWDu;>5v^#v&AChf!INl??_
z$7#f1B=OL01y(T;T@}Pk0AZ;W16$BZaR`vyAe4euoQ6WsNbOWlr8&@bjzhx9IA*_R
zSP)BE!|b?4tX5uUv2i-fn!OIb@F;ecJD8V$Yo#b_s1iZGBun|=&abOky=*U)t9Z+>
zpF(NYIF^($+)wuYJ|U1mgI?^(jmP@w0Z&^PRQ`dw;ZuW7HWQH{|A_D6L|X95Rv(0+
z!&MV|0xb<#Qpl*8bU$=FFZjh1pUFk}HVp_G-~Fu)?XFn;uLU*@KYg8YJ6RXWOe9!i
zoGn#BcheMr<hj7Kb-p2UEsF~XemalC>Zn@v<~et%z9?tvC5q$^m>TN_-gK(vDN**q
z1Rq(7cCzO%G392mWdR^m`7gxaFONNXo<lv~hXSdz0uy|3cit7V%xCd4#+f<OqRn5l
z$<|6!v0fk^mU^3QZ4tWeSyc>qYx}=+8!i_Z?P1T{WJ4HIds?x*3r+VtEm8?PsZm_3
zoF{*`sX6$0#Ex32M~}K<up%ZECuI8{uiK_A^q91wk$C109=#}=tV)btY-)uIo07#-
z9LG>CG&@ee`nZ|RWh-h1vEBO@Fw{Y%)wNVv&G_U9G&s__Xk=9u8U-m{cj_!V?A`$`
zoE<2rv^qui^wV&xu_z|wGE#=5<3F8og1%upZlx%qs8Cc(FRDfBDath{^9(`r-Ydv`
z=KJ|iSza5$KS00y0+>8w5;3;k#axfmR7*l&i%4zNn&VqOV-Q>3SdEK0MYB~B?JlFp
z^LCapbWT<|7bky7t=uoAa2%VeCEBkFNQLDWjlL%18Q)~pEfUQBYVUj@q#V9HlQiAa
zOjUsYUbV<rMuQg}lYTqVIY|(D2bEukvbmZFT}yT-5CPpu(mwx<7@atXG8qFGPOB5*
zN9+HUKSYeEpZr6a-xSB1dlOeq%#@5t%{$gru(fFV*6DaNbp8_b(M2nc`wEn(0;4;J
z6GbKZicraRmWMTp)0vEl+KV41vBI%xK1<|kx$F5_-%%<!H02x|uM!*%M5X?m+CLTK
zE1aZAAfPE1rC(#H;*=>b_msmkXMA(Ul5mp5P(-k)WyT&K#e#{+@&uY@>ml;4DTx+t
zI4sDxDExBYCeX4#n{gD>w9cOw9eJ_NE}P9;A(j2*_sm!(cL6X^ak1d<nv*c0MS)6I
zD4DA}nP}I{fVU@zXy8{C(XVeaE+XZ<AZd^{ADkv<Z*T9g`V~s29}$tv=-oc5G2Y){
z1sjr2tsan0Sw?DIlbIi&b#Xpzl~$rMojAO|_qc&os~}}L^J%~TH1Z3(H2!$?9y2ju
ztK+Tn0Zj)2)0%_BgR(?u3rztoAKkB@;gWQNZQPc%Mft|istp7uX2;17l=JBPm+z?@
zfxS1$w(CJP?Y(trT@`?{NaMb@qLP25z)}j5bB&hnpb35vz69WzPHD%0xcY}nW}rS;
z+79AusbKtQN+NR&+9VS&IB0YsgE0p|`GbxQf|(Kz&_KbY{+1V-)|5l^dy*&me~;}k
z$`}Hs&@nx7F*2xaRB>9^$rjg@F)c>1S`>Ks6UVIp=z0-Un{;;x(fTTyBOZCRFPnmo
zi@vfRGCE~+fhCMR-znvX8gH_dn;j#rF-rb8ZK0ff&qeoOnyT~<-$+Ogcs7*%3QT(D
zt^I>+fFF+8xzvLrr`#XqgLw1V7<Yp+iZukt<ULjQt9w19etF=J%k{AVsJoIz(%I$4
z1MDHlGOMYCkhdP<<FB_S?xVggV{Ck@1?Gv5)-uwL&))uv%JQQFAb=E+Pn^!Jh%z%8
z0#a`-RpvOZlx@;D8=@k{o;QspsrZ{bZqPef3paZP>qf-2QQVWmjq5aMU{*!H?44|{
zY~pf6zVVHLr((WLV_qGp;Z*%ad<??!yS=<uOO9NjkA84(t9wPhehbx<uci>x5XQAt
z$I2<jk<hTht}LyVBnh4v0PemQ8)>Y+M0Z&FE40+f6s+8)@+Ia>dEWMQfS%TgI~?(6
zd>IzfK(}+e@l`gX)St@*uFTdBAFF?*TR%L&Xm%szYB^5HromBYEY2QtNmDc2iHU6}
zmux@Vo32PCpV2GvmEaoQ0n>`%A@6PN--ltFY-&cuWWT==p5bdK0zX=1RUizL9a94$
zT|_bZ1&HnJUJLYK2to-jQ>kOjSLs#JN6k5obSgl{{xR+Le*rS7*$|xP=o)KNC`({|
zIs3(TWIl0o^WsW`pp8&>Y3kVIsDa=X<bfJ%md%k)atHlU&srB3wTp7{h63KL7O(%H
zid2E&pn~;ug10s!Ny(-ttMr|>ZY+n1%)N)c(vEytey1()NyH>$?k6Do{46VHtyFLw
z4<h|;@hjqJgf`pLMu##S)^%@Ah7p;`7bS10o2%L2J1l|NpS?X=0Lz$t5d{3`jG>H>
z@d+>_`AlwqwZ^e21;KiPJTT0hYIk>pbi)v@7T^<sTRkc1>moeRTN<6aQ$mG={WAbS
z83LM(-EZEx9-k^woI!Pt|5J^Q*1E?%9rQozV6~L~_3@g^PnJavaKaJfWlx^f=JBD8
z5U*)Trt<lA`dp{i^5h-GmaRTL9z9v*P~-6MHC>s-r7b-oa8`GgafQj04~?a=m%pIe
zVzNE6FTg%jLaa>x5^Ok)D6$!#d!b`^{3*6qj=HP>@rH{(UFQ@JsyR{+A&`F|z!T81
zdXe>@2Y5d&P9ST@85jry$yq2T&;enXv{8P)8u>~&Bl=Y#_REL<NW7veCM1}LHrFEt
zFifA*W3{7X^S3{KYYECx%!9!r^s=xbSN&2UlfBDC8hX*fk7j=(9r!;&vs=dD_ROcl
z<(HIj{89u^?#z~-w>-bE7shtnLHGRd5OFjE(6JBtE)!n_8UyX+gJV<g0Ze#sDiC;n
z;~@z(H@D`;&hY_#@SP-sW_JK^GFTim8_#8EC0-9}8^1ivAixk=J>f#fBt@+`Y3Yo3
zt}RPwt}N=G@a8<Sw~eiwz7$`SaLa+K#PvC~<}>yM&VKm^orFKlFa&rw-(D4Q!to}L
z7^~cWnb27IKBNw)0m`(%ZV3FB(^}^&{*0P?*r440v$3{D3=n?mrHqEnZMi9SL;l}>
zOJQzk^s_pqR><*7NQ8cVXIo`G%NU#NJGp7?_4mXxCJ!cxP@RSn>4_l>p+>RY>Ls<>
zh3?5I!)iSn(LNM1`r5Js5LXUNprhXm09V#IrY;pV#0YxJ{c0#qCB~1Nh={*`jr4sd
zFg(qD_aNLB%#pr%a0Bv-vA4oYRQogenewyM9GbUj>)aTYeXc~e{9me9J?jC(WUx;#
zmAZX`y&|N514@@e)dY*Y0_ClL9Nqw0)RBAp*Vm34+_$1DRi>!$c-@>^!^!h9>ZF}@
zC-1Q5BtPiCU%C2?PZce%UNIltkW-F1WW4719PZ3L^~~Y>`=e{h->{l3Ua!d!Z^c6;
z4Fd=lml)*Hek#*aIKr-clu{1pVY>3I6_wQucNjd~m}R`0e5tUQ3n=1tQ(kh<^d#RT
zd<*mVJcrd;UrX}3O;*mCE2}1<Y@Sht8}IKrvNjtSbu;K53OlTHPvSCo8x_a0)M>fr
zBa_KleoAUvHBVz!n~vfNTt6?@kpB&?lY%#7`+2PBx^r<dDcI!k6H%H$ET$vkjhOXa
z08^rgP^M8=C_^krNfpuDANVZ@i>^|hDsnYN;&*UJ4o8&Luj6kxOoo9w6f9<A31D0*
z2@%Ucyjauz4$OF8B=LHQC4p2q1W~T_1Tci^+poSlW6*1!2hZ_&jcfLPFlReZC&f^U
z2xt)C3{Xr<TJ=C-_IF-Ov**R&gHPxHN|2f(%rLL_?T=sqQ>RUaulR1oO9fI`%Y@|h
z@>kfB%oz3&nRw)!y-Ur4mDEBoULSazNK(E%>biehW99fNh8byI5dVX$RF73WdwhO4
z_n@d}nEWJBqV<)ivXx+Pa~%spa~wba<j_?8{D<!M60=D33Jvk~9@oehB7J-(mG{~l
zKn3|aDy4tbH;9JkEips5j)8(^a<Se`y~zB889u()VClThs-E}O`&^*{g%@P3xLBdq
z5Wu)QvD5o*R|3Tmxd7ff=e}IMMlggq<qdzyPk^4`jdBD<g-Z(<KAjuj2VA^NV$8S$
zQ@uUu<=;zBW=GP`tV$m_9`Wj;YpOItlof|Ml}Qri$+S>OXfP5n6(#Sda}`O@f{J6v
zb7MlSG~&o>s}@{YjOMQRvRTIV##mTP`$pXI&NfDM#z1iM&KuA&JGYxSD=Z*s$;{I*
zk9?mrpwj+CCbUcggrGqH5^>^=p%$n0&Od<gzP|_rNY|YLU%;+q4t?k>sBX|um0WvY
z(xR|?Jt5qxHrePLIGWcxnLohGhFr#6g1~3qAmIT8Im!V9Q{I+6!7k3+g9Jr5rEgEy
zy*W@8QHV-Be%j)=ea=bMUk+Ncl(^b(pK89Afz8%f+Ux$b6kUO1mc5w|;iq7(5kvA>
zOc@2Up!`fT$1-NoHmjKW=)PIxjsmtsyux;Tp?q4u-D~AF+8ey=tp~nhk_=3F5NW9H
z5UIuY^8Ch1`V4iX#35wP+&W+?ALoPTCy}ot`ltgiFpZUXq@6x~67~C}MbWmhhQ7D1
zIm-@(>R_nURo_C|J&FRq4z&AouEjPWL1|YBF#tn^H)JWquiwb|JK!{)V?eb6)?OV@
z1P+u^u)$c9t}vE-A(Gm>myvvTfe&tmbgT81E`W-c;V1#Z3y(~ay4cHk0AlHgw$61n
zs<nVQTYx|?A>F!<w2}mm$SvjlHBp=SrA^0$Y7^foIN-~#gcijKRl0n<mwiWsC(m(M
zRK&^dTn?8?!s>|K%z3)u&L)B7XV3Zec{hu{1<m&6k~s=$o=y)l{K#OP^b=8H*Zt||
zxfH>eNVnFclm|5r3$ZS>TfBjuGjz9t`f#V)@tw29{xL@+T6DS0--Tgzi%lnyijoD#
z*qwGoFxOi#B{VUkm3eEYy2D!ME@^lj5zoKwy16Cvl)z!28da#n&_QEDeg6An6@LxQ
zG`XF?NsL>P51-2{6A>MMhAKu*l`sD>O{xZIky)CV#{@`DHgPW|i76SOPndDvHD<=v
z$+S2U2EtE;#<(D?2lXL44v(g@Q%;@lNMb$e^A6RR(-9k8L+ARjte@ZI`SxMVBycCk
z=!x4vOd7}}iyZZ?@TMgx)>E))oS_7PkwotbGpt#?nq596PnV+;3Jww5YZER?<^$na
zGf$#TcC?tG{Wyd<itXF`1JW2>)=*)W-kt>KX)~pj^yDNWkF%zD0dLyET*o+!tajgD
zbmoX!YP94k-%q8IYs!Mmi+m6H$VYi~Xr^1|lD>Q@e`s+S&W=*1z^<`Zq>~eThi2P1
zemM#*vTFFhJ{&*V=w|}v4%ZYiNIL{3I08tciMWd2-Uq%AJOkDm-Azw%PUFb-Ho)U%
zOxG+cQe(JZQ?J(rtn9X5LiMuf`o94zd@>M)MKzGQ=)Lq2<fY%EPd*aIkW!|$%G0a7
ztWN@yH2e9W{nW;HAu^F)$c2gQfs~pDVLjPq1NMj|$_W<Pw-S~uI88-(HW=258fc<F
zD2DJ<g7UZl)ZVT>X086o4>d?Mx|c!5s;&c^>XPJTUCch-C0#VXbt>sGXyV@x_e7(0
z6IL%24I4jnDy6aRW@=nh(opDLW=0HP@@8tlf96U2>(=`FT@Mn;Z;d^ck_n_IG>1ze
zVw5DfslnO^d;+@}@&@6_UKlc)j$mijMNfYr|88{-j)RHL&zrk!s4dj}o2c}-y#D5!
zTuy(!1kA|?XA>1>r8vjQp`6WZN|U_=$qL4<``S6?DcA?`D*jp`Cv!SHR`b>O8`7m6
z0RLrKu?}q6Rd?!gskIK?6mhhh5noufNh!)Ug(yBnn87G6_HD;`<d1w+i){}d=e;X`
zUnR4<H0EtCcQP5c<j;;POlodUTz;)63ubitm!&3irN-^MQ%|0m0;FmrdexGz2<god
zOh<CwT+zIn7{hw?I-z9QCVrdpf!gwlhSIh`XP?U;^4vmOw_t#ZdQxLQ8M{nc|LSdY
zjG=qLo59EVjIU*%>N0EozBvB6b7)bhyuDdCZ^vmC+N=vVj}^XKvpI!Ee~!Hm$Nx3w
zW*mHhdZuW^`w=%tS_P2R(1pC##Yg#-6=0NP2F3sztD$6Pz4tS^$Sd+MBcSqe&R1}F
z1U5Itc;ToZI)f}pf431fi)|o|<qXh^)`>Z()Tf}7g0ak=J`;tAG7!7is9E50OuWE0
zfRRF7LG4J$0fjE?e}eM*4U5`6MtmG(y#dJ^H#j~iacs>tFvn>d6+Yo4`!ot16wX=z
zrdEU^Ps!OWjPZIGGXD!0`b3M(fBGj+DIol)@I--(=(SIxIoXZRVb_xnJ1V6f84SVw
zvA2^bpG+m8M#qX+2}gX|uYp~k{97+C9FS$0A~sa=j;U}%g<eyrq4_l0)wqhKavVqC
zH-vR^k(zG&Sol3{qOYe{s!p<Z>&2V)-;xuohnV8;dE-?7^EhhuK;btE)W}i7ra(Ur
zN<cQ>o}w|6JBTB|>q=_=K}D@mS%bl#KwBYWY~g)>Rw$iH0u=j->=Z-6q*|Pc0Y$5h
z6Jv9wn?vasg4<$oUL{=B$ItMaug>GAhW4GF2G6)qJ~tVD(p$HraT`6D$Ba=cvYqL%
zx(p7emA#>m{iDq?+!i^P=oKmA2!|l>sr>&wDv|uO-G{{sxOHVJf9zduuqUg=g%A>U
zMuJ)YMeAM-fg~Yx5>X*?Q`veke)vIx^5VN(=d7@Z3*C67pvz@;%*p)y9ot8F76OI^
zUlK*l2z45vpUSalL;Eacv#W3_#y_BQplDu;`e$#doCH*38T{{Y1oio=FRmXLtu{X&
z&c}pjtx756^2Sl}EU(zkgwNMeJItIn@Jy72nFw7wQlnD)QXAWaCYoNd=axYdv{f}o
zKb@$z&X$JKx+&}y@!|Fc@y2LnkPcb*xpIFZ<968bpBbG`6(;}wGjK2rqYi+P6G>O=
zj1yLukQ9<qfbwOD5rX06O(k<fQHy`VpoJd_!s*pvwy;e>&D~{KnqsFpLj~Jr&iQ`R
zo^c=HU7hxGL<zFjeK{;RN5f$yu&B*bB*Jf=BcsimCh5BfJ_4*au7#O%qlDMqNNPfo
z;|BgC;NBGO#t4NVk*CM7tGxap`OhKs_X8?f_-ot3SMoMnT4Pq|`C>0$;oWmng(2aa
za&NXB`MQMFywKlg4=q|-4Lf*bW|-TOFhv7@)VGq|c!h|+XJxl@Tc8TEUhT~@_8Qk&
zSTL5f0+P&xLfZh``L~yh5Lg#E7Z=xZK1RQ9ZhnC%gi!#BFHc3K*}4DT`QVfgr-XdZ
z=*?v}??^&&WyoS=J{WL!D7P-F$ijYbi<fne6)HGZufIsCC+N_&&08Lp42z14e@;mn
z-(v}w>=27H$WJ8#VZ)Lhn>cFn61C7BMDn;&X%<;8XTfCBHPc)l*!E__Y{IfcH6UZ)
z1;`?XpT>m5nb8E?6y`q$B!tz?;uhl4Lr;N!_K&x1$3+s7Gi3k8?n9sr1tW?x@9iJb
z<wGgPj{K$tR{$Q-fgxV2h49`%QW9-M9x(AKU6t8S;qCDcgzPf?3uEQS6VS)fGQ{EM
zIXo||Itb4&zVU7lA)~T76MF75X#l=Hl12xZ(-8|dy!n5{!~c%U5rMC0JtN(0Kbyi;
zE2~eXgwx?~PDKSge|!;CNIh4UK~O*y@m9%hB1A~O<JBzBtrInEku#~6>=ro2zbDUx
zUPC}eP97K#udpuzH2tvxp{XCh(>(MX`2!Ka)P%jCh0q*W!&v%-2E;cJ?MYMjfU05w
zWb;Kj^lvRd$It-g&Myl>@%@}&Dv)<}5kmz=aj9IB^zKo1L@U6LP4N+o{yQjwAbKFM
zts)qs<~$(tUh~H@#cjr6xBwtrT8|)q$<v}`mBp#(3b;7QnV6Vv)y3Z-d?a?<WbwZ~
zUCZF;KqlnXMpv(oB0Ji06AaGg^VQ9U^B2#`0ZWz!K;Dt{hOA|Ob1*)r`X?aU#slDg
zW}A5S#N&UipH&i=6joUoGR3}E5ppPs1a{h{2u8%?gBv4U$<|k=sMRnhr%dgoTJV=w
ziBmU1WULc>!=yolu<EQC<fmH+gdR)8A`Wn^$QC>fEp!V$GaZ>KwDzTZce29pQgS!o
z7TSc6^^uLlm4J))f(r&ESByj}bQLH%sek0_Bcu5qo4<q!Zv<;l45|^!*8A}zMSjV#
zT<PqXBM~G))rA%ZGM8+U)Hl_@{7SM2wkg23Uxtg9$AW&DN=D3Jjq8`NPATyK)<AWn
z8@E{Hk&{;;^&gk$cC=E2DU-AufXxf#GK!P@eSLkqn>A#I^qTd>ol%yREu!fCfQ|RD
zc1Q||1P2?txGKBgzgH21=rh_}u!f8r7fV~4)?mEa;QM(#qgMM7=b;14OG324bmR@F
zK3aJ6@2<<v^INR8(W8PebOIxS)pchjh+BvgJ7iwo?Prt$oa^&)GjX0OT(4N&$ZCC^
z7e#M2ldadc(;IN*i`RZ_prWA?B>$nMe8p-U9>Sg<w$KWuPw<`2oQ}Vp?oJFEcQER0
z;d^3?h{FJfe{OLg-}H_s0}=y+IR&?%<NL%-t{=oOEjT<L^$_zq=mQF`CbvPqh_E;Y
z!d8-CH=m39aW@<rLOJj>&Y`3&h>ggi-hO@`V|4wcqsI)@uNboJMvpH|2<96cLO!83
z-u7?JQRhFByU`^^jR#ZzRG|w{e({uzLz8)U?UiNiYB$CS`hI~T7}*em`bogFPLO?c
z;=bP}y~{s^2L<dz&vs~3h}30^&ykA8DO28f2z5yldL>ft@5#`L6BkI_f4q5{d22Yd
z`#{rg6mf3hWRz6&jU(TI9CL`ETwJGoC-?Ft*+)u>%#j&^W4AAcykV(SWC*pPUvz|s
z5`%jZsT2g~!8Kj>3amXyH&!$Q!$-P$3=nW+oSaF04^oi1G6Juu9Y5la-o9pE`VtN9
z(7g9bCPtv(2Sr8n?!!G~NliZZpyK#^eF%iNXVvtaL6uAEwHYhsq$6!&+qy$>WuFr7
z&|yxBIy7Nwcp1Kc0PSv!*Cm|~o|d<*R7K?w{-*U$u<PM-Rqnr6j~ffL=OtZCfz&Ov
zS~&VG&mk$VVnv^ngJZ8O&f)>kc=iy?=tAvYoM0@80c$B^1TrLz0W$AdDrnX$3hi1g
zq52m_fuURNpFvfs@1j#*4tx-vE1+i0rTk={$0TSK2*6hZ&GsAlfC|U;V|nW%z06OS
z)EfBf;?#hMbgqFPo#YF+&o=-vX%JXXGL2i7Q`z8qsC?UeWl<BzEyt3R>i7#WZFuRP
zn?4no_Mt`5ReDi(?W^nDzZo3;7#0?evT|kyqohCz$#NrGjxfFZE==?hVw-_^)x{@G
ze^UWL0@XV2sUF0Oc_7$mRHDk@QRK%+-#>FtLH`4*C!h<Ru+b#0Wxg}9yOu4kp9Zpn
zd&qm&_iI@4jaC}V$hQ-@A}<M~;c5|?91*>@q2%oF;RzXENlUiyLXV*shd7dx=O2C<
zY1e<(Akvn?ZmYWt%ZNNC2BS7&)AAO2`qPU*G;0b?Q6tUt8P0v)f1dyOa8Tu@ZhTxp
zt_-U5FrU9K6E`^AUsI&=>?vGy%6dtAc|{=d&I1$DQQ~5uVtI%KmGMio0nd~x+9MN+
zK=&&a*{>vUO$lILsS?lJ_p-U+UFz^%I=}PvBMZXl#kweDMH~U`*JkJA!(cL1@I&uo
zieMjmN(!3qv+mJgrq}k^M4{lUnA>TOs|{qf{$cW)^y^i$Y4XvpP+X&1b$|1T>S3TL
zTygm@NLu_coPGXoG{kHd>U^6p(IfZ5Y`9m>0&~Yw_RWCl{sx=VF@X?1O9OgT-wra_
z><atI=Whz+gDUXE6saFZl+bixVyHrO->08R?9^JN{?j_uzVeG83tf@1po&Emi`#GO
z@|AVZKZqk6o@$Lp&>6rkDMC>cTbmwl{;GcTZW0$0vv3$kO&Ngxv>3_7P)fuzpq7-Y
zuMP0OF{z)YG(T^x!FYTZJ1TrM%Dz%=DS!g3-Ema%uhBQbAyYp&sPvVo?sFVzs0Chy
zsSBg^-2x&Mhe7uQe+z|N%si?vynao)0P|G~H+ea~>_Kv<oc?FI<iUHH5OVPsD}$0b
ze8Q?c6$4J1ff_vZHpysjf6N%cPK5vSUg)6Jko<o9ICHpF7QFwI@8LISp{#5uW1@JW
zSZ52sf|^q1yIeBw0U)Bt)rj60CX$2+j)7dMP};s|4zi`L04%X6p$O?w`${tyBjm_H
z636|*nzFE1wH?_{`bvPT7}iHbm`+-9!sS@v^CEAF%YG+q2!Hj|2<p!H1kM&MCVCOw
zeuof|Nq$$#y)wP~PTK|ZQ`L7aH<=Y2JmdKqyB*ad&9G5JyoHT51+=`wFlf%_N+P9^
zdaDH}TaVrN)BoqKU_!0n3t!1vunNmJetoq=#smdWUAvbnVvc4oO|N0BNgl)CaYb1R
zlF|AwK;#P4d0;199jgW<YD<FGXqF&fhG18i+sEEJlmX?}B&C)<8Q0iMpL%@<;Pb21
zs&qa0ctyzi-m?%CM(LOm!nhITNYG&q2~D<<X<HzzcogThyRE#wwa;YvDo0ViM4H)G
zIpp9!V05>CqrOm)o3Y++(fiNS9@DSvB@`jsZ+eAX*;=eOxSG<D(E)Mva#B#;8`szZ
zKh5%iOJ>oRP)VV0xrd8r<=x}}Be^<-LKO%JEG8c-UdV6?oE-a+?Y4lvZhTWv*j@N=
z6JSM0>e7}kp<$RKi6DMouDB`CQ<hjKN7n5Eee>J|Y(vr$GYBM$8%h6yMd6WywQATF
zBMT6}H&WCEG9$Y)Xvw<_#^vsb>&)wurL7;4M9iBAI4P0Kyu!>X#f4Rq@j2d*r#Dq3
zh_Jkk)hpzoyK5X*74ThWaHeit`uEL2F^YW|%M4J)IP-PHNIw)6%ZyEH&|^za?jUm`
z?Ln>}`dx61M}i{g-&O&D+hR|B?R8)+qerIs7Cdi|XqlL|TsI#IVFM_!JTS@IWR7v5
zB^(%vP^6>zseqe*M#_J6lEJ%v-5cImGX}TbKd_C>{Rh~6T1iSPfN$?aj*B+fD%0#i
z!UnwVx)9RK`bco3;jTAd?xu$mGl{6IsBn?FaqLn}Y6Ic$XC6t<t8bfpTPcXu0Wh?_
zJ+7L{36kVz)c!6f&zb9PYhCxG7iwMdnLSlBUApSxNEx(;mT#Mt%M(}4voy9F4mAJO
z*X)>}ad7g3=f<-ZD7&6`)sY_Sq(#CbV*OvxN^X_%sK~eAVrAbmU&RDqPl<fOWp&G%
z><R0UT*ZN2#36oz*)y@SqQ^-5d2B=6S%ge+e4z68b$*gLY$Rt{9ey&)n3%G%qN)Mz
zo&-te#ev5hQH*v>n)HbP*+*1jC5gB}X;?lwp9IJr8?`pz?(67*nS~!KUrI?R5WTRD
zs*^erA<_%&3cTK)=nT`4Xjc@3c?=0YOQ*CP+8F9_Iyw<QV81z^_Z{FKdUX-C_p&s^
zSd!jYI#muy@5b_ia-A?nY62%Kbda&-{@!G}FAPn!GiY-NFNuk=_P-jVoe0W6Vq5-_
z_xJ=IXA?_eQv7CWd|t*0#zx+Nk9N?Ay*U}!ElCPldjwP<zDa}pK8IV*G{L^(=6G@u
zq>_O3P-H`TShiQX50pyM((E=%#f-Oul-V;15h>pAFJDf#$WakR<F04)uJ#9~;e^a4
zkd+XcNM2YkB#IEpTWm{W!B0|CR9eZQWiQL|^z=EGQajsH6_aMfA#OAru@Ck0*(QFa
z8%T7!y>_`Ex^Q)sB+rO-w9j$)Q29w%B+KroiT3|AqP1vHtls7~zk&5q?!$?2u2SBG
z(B1a^>YiK4iD$0pe(*eL3^mN;z4!!KTz-1~yio8f$PVK{pu03wunLdkv?N&b{-ELx
zI&QkhmdKJhwUt(`Fy@0n+(wGar3sB>54&v{Br0IrUp|KAEQ0PQQc@+1q3o+l`o#L1
zR(j8QYYdCxMB<vFr_EttZGb<l7RqW0v82NG-v*u^mgp1aF))C~0k;PE$JdTC`Me<R
zE8666LG<@aGab*O-_#vUX9p<Ld0BM9ONBAyS1ih;RzgUrL3W6f@yBvx5|oqC=KYeF
z!q4=&rf4XMhs)n>NWYe3vS+YRZ3&FnQJ0w@xxFpzsf3dqe#oYaB{j<n)PJ7}im%JU
z#zo5R`e<K4KN1x(v#eT&RpI<7mybjH|5mr)BmJ5v{8jTrN8-Hhaw%01Mdt9m@{Fi-
z0tu@hXkDh;Bn~ZLJk9pgK={YpWrsYfE=IIIbW@S_fUMM*z4kS9cz}hBVeC@L@bi>6
zSfYW`T9TWIgxM8&r|7W7U$}T3Bwzfn;M!+j@H&<*J$+MHl+CW}BWTmyPzM9^TcI;F
zNAY*xoWrnn#~b&k`E9b$v2GR`GkMHw;liS)OYDCNoWJ$pGH`kkheB4z5?cXB)|zO9
zHl_KiLDx`6njAM^fIfUrZ9P|>I!CkkjLl}C{o{+*wxW__{Cw0E{=ND`J5>$`FnZ#v
z9wm<X1&0NTr*z|)w91`F8frAjBYB)SX=9b>a|83VSK*8!m;Nv>UZksZ7^9KO-;CpK
z@4-+WWJqw?b0a&IBw!ZUm{T~Xc~QH*Ct%#7X-vlNe7Yr9ylNj>Yu*+Pyo`5xYdKH<
zuekqrf1c{cUx*^iC3%Y|SQKX-ug?lGAA_NsICwwj&MBdV37|;Vf`qQ}vKC{SQcu{5
zx<&y!a{waB4k8u406Ix?G1DdB6h>nmbj=_|(PlBK!(JSa&%aT~-$x@$Sdr{X&Wg8u
zx!r2KES34QE<;?G{1iq!dpkQ8mB3Oe9haQ^J+B{zZ>wF%tNS%Z8N;Ug28E$FgY;zG
zB8?9n8n{aeIyN*a>Wgf?{JF}Zf4UMmohCDQ9rS$#$K=Kx3!Ky8tM6n*|CpCq@XXCM
z{YQuW>xjvJ;U`TR3ZbmL?Ez~W{~A~>)LX>2KV1Gi?2N5OFPek2X58P_^?raQYOXCh
zgZMDu+$0BPm=Js}C-EdY4HD4MOs`A>SFJJFdv}|^Sq@na>?}+Z%IoJJq5oc*CJv+x
zfu8_uuNaVFU(Rp3$-<(ZUIUh*=HYlPVsq4O6&qhP0(APvkK`mI+NxFB4106s4wlK2
z0~Jq3k&8{5mN`sfrYm}Q^TUtco=#`nO}6N~<|XX+HzD{u4(fB2nkDc{m@+mczJfaQ
z;VYPX4usi(*^f78v9%w26?LL5Kdbmuqs$iVCB|o64DaAa`IPE8vOkh;*;bN8?#)ek
z3uj|vW0O62b`gajKaw)GPs#b{bt<Ed6sbr+*L-lej``K(dz2*%l^XY*wBW&RJD_NM
zxTV;4v{9+lcKAOUY&CrT8>qg}!9j4`8UtBu@J*2=-wob$oY|y1S}jEekLO$qt<+pt
z<TtllhoBQH#RCvA7A-<(U%F)6b7R9EQ#}PK;uc7~VKr-JySra+F|E&do5$+rQIb2-
ziecailp85mmDIb|gqHj=P%$310!WK%m69kPWBajx<#_AVvo_^G%?`a(?Ol`kN}c$Q
zb1;zBn^)+OVFbQ>ohJF5Zw!}{O@_Tvtx=+F+821NlFIXP4Zc+0wh%`XhiA)G@a8Lx
z)+%XcEn7?HinkW4?_v(y6lHf^{`~CO;cLb$S1<Z`d~q>7EGm=U(n!E{Fci<8&70hv
z>)aMq`P|G<n>#vdIy;-4&?V8!J*mv;h)|n63wML&*QUGo{ow{G_#Go1f>tNz*92^H
z4Og4)YA0yqI#)}sU6zMaY^_M!@8_HJdkGKU%lzEYp;^h6O+5S`3!ul?zTw4xPLzCP
zD3eK-NxzTl3k9K}Qq*h=G{cA75Fcl<A_iIOt3HN737`(@&>V<3^zJNH`75+yrx%J{
zcS6=$p5uFH^cQEuRf5<`!H?jfKdT>?TN^$z)zs!oE32Uz&V#_s;aggomz>Fp)y;b&
z8JyFz#R{ZXE~b@KQ`s}N{LC@|#XkC~?I=(qT&S#&g(lN`o6S3O73q+Pg5@f38IXgG
z&gd7b6I)MrD^J3)VjVsJ!Dcb43#Jezz~$sY=sPhXA|f&fG$!h<$J;U_iog{MMgO@t
z^G4gwo5Bn=ic~-*tE=p*5sQ8so1NI_aG!9rmHqtAc%bcQfbq;HA&c=Mk^U}v_=5d0
z{ZFE0iyfzjgdGlP)j1)Vua#8FmdD;jC$q&m$6z9NwDiLcmc?vi$32H*@D%gg*y$qK
zA+hWd-YVIFWs4~GVau0W38~{t#P0qN<YK=_&<GO5upT=1ntMdW20vW|i=T!3luUN1
zsF$z^p==G`^S9V7Mu=ZN@ZoWZc;EE?KzHm^+BgFLLxKOidJTL&8Wf2z-|^SyTqI5o
zA;+)YA;ivpmXPNCxw*@I#q#`v&%N7~8KB!NhJSju1Ho$4mFoQcczU9S$u;_BvYZrF
z>oJYx@Oz0**ZBaDm86s&1?l>DuU7+Jb-=PP$jId-J6d^+nMFyAy1O5pGCY9H?h>Z`
zZ@2m^9*^^hIWKdRBI(pz`V+fRA+)}xrzne2dySTQ7$(}u%Q(k1Td{nNxQsY|V{n&%
z%@|dGeo<HCynlc!#$N6Sy2o+<lM`|dwAI>=qTl!D?y$*Rc4s8gUv0SV(z&Vu=Zc&u
zSS1|zjFIi*%tUV=-G$0*a?a7Km|NEs(jvKx8I#C#UaFiSwTn5LFvnAD#bS!<rJ}D+
zH@~%LCAtNJwr)?^ht`DPwnAHX#>7O2$~rs-7-q*8U1t&jDrU!sMzxf#;M-7hv}>hZ
zSj>0uwJ<acW5)dX!J)Hs_S5-v_LK+B{&&KqKnA;Kk{7Q*_>!X@>DAi&id`NCt%9O0
zAxE`dtht`+!d7L5+QSc5FhHzo(I$E#?LVtEP=uX6-3)Xvq5krDHliG#*XxH4TZ`v{
z@+t>gT{+*-kPshS3e`9E1&dsp-Jb=2b-34zMAiG_a>hMPY_rBEKWopUiJ3h<@}$;X
z_rB_I>XVgjM#g0ohw}qle>(Snb!cTjH+eR0T0}LPJSG#x{AnL*Xi*9RlQ>{7-i_6a
zN+Fh1qFC{{t5~R*r<6`v{RI}h)GfJfMBcWeyv2BtjP{EjERM~Fte?~MzL@>2ZD(iQ
zx)H^lFP+M~ctNKaxg2p1fGw^2z`!v7;M|!`)CSn+EGie+^+56RNcW+(&H^9+p=eH=
z=isR7KWr^O?p$v-8BUtrc24V9M-y4(Ng$0e-frelAdPz6*9}HQ3DxfsuNGtHqG<pb
z1gmc$|6w?uG8)2=@c0k`-75~(7Cq{Amk0#Fa#en&9uD8?B7G|7NaqPrBl=WDz>dI<
z#)5VK4)rY!Brb>zZNcR!Y_a$uD1C}EK=3veAp9f%cX10yq0{-(?i*auj3D~<i_9Fg
zvj}K74ZnW>7J}rmaH_nNgoOMHB=YXgJG-Cu7_}Z3$ykbH=YF2hu!+Rx&YQ3Fcze1G
zA7*l&`#erO5t{ih?#)%kRcOr!RcgvmD;;tzuPErr#3fn(lpuCiswKQJxOJ`lv+YYr
z<5AR4c)cAHLOb}p31gnz&xg}^Gmb|w@8mm|{OC+J=11m%n;CWO;RM^xIFrv<I~TYO
zHt|1xqy+lEy2UGdNA%4?dB5vMn5Ed_=ZMOuXN>t^4e0JlN9xH|LEnf8$V*)K=zhIc
zBue1a^N_wiHdHF)Yh}2)qHkKtrHuKoM$C5r@xg>#OchAI6h>Y10=50@PB>@>_S}n?
z8~>_J)X1T^uNdPRBrX5!;e8>=IXg*y_9VYhb7Xe2HowihpB9D(C15NK{_%U6K(|KH
z51$bj%vqJ%v9f$_Eo8>v1;zCTAVeHFBAO;**KELi=Y&i+C<0^30WeMncFpOQirFSW
z`t)u0ZqGt+8&f|b`ZWMUC-z0+?VS)leD@e__cVVg(gRpw7aVaXhL8oNFC)|j<Bq|q
zjUnJBbk07WWf44<!gwo+9l$&%!f^!7a8_dSK3Kwm65_V@qw3nVb`V&eNyQ|pflvGT
z)22<PYno_7eZ?BA-AXo@Aq*wnUdyyJ1}Ra^u<snHoiS>>QX)_(6+O2^y1(WNMtB>>
zb<EAFXxNi)nOgDiIUVwU6^%$O-zB)n9Vz))oS)`J^G=zB!?!By<95h$Uk9aaET}#g
zJn$_twO+F3R8zJiS$^w08(FlcSP7ea`0&xL3|WD_+LN#|7(d|Y7wyB-_OEP4UnsS$
z(%A-NL!>Rz!S@%}<xBP7FRu88#%rBYJoS#|;dbBevYQuQZGq@FqgK5|kr#?l`~?<1
z%L%tmmGRL+C3Dr&`sv9_r>4P{(%JHZG}mr8CcV!4R_WBn*2hIx)`j=ed!PS&IBdR<
za(ZmvZxvhx?8?b5zKh&>-oAU@S-N7W4y%htaM1M?nEL2k^A=;pOLY|MfVyzzhl1<Q
zQ~`sD!Kz|i@}c*~U=9&0z)L9ksml0KC)y0g`ebQU<9n;eV9}3Y`ue}G97Ml}a{>HE
z8Mb*!bXv+AoJX09S8c*<)Kx}|DX=H0%EsS%S>AT{AAzkg(|eLe#o6CIgdP;q=#ojK
zOqCejWzS^aDFP_Sd`}qqg{dNMsO0x;#lX=5f2>xF8<r6A7VUPo8YzoDZApbS1~V;G
z&W?(u6z%I30Gkw9J?BtD6?KhuP;dwe3~csBNo<sj7Ls}4C7O6DXd4_7(iZ9rr7Ft%
zaIQ2D3@#<?iKY|Fa7tqwIC=sibUKphA6@nertWp7;3o0EKnu1e^kyxzn-6kmpc*GL
zvp6SHAK2s?UCXK<xhD_NtX9S=&<@@Z-IfsgCq7|AZ|;3B=YLTY3*kGIi9OhOp{1%Y
z<-&St4j0>C959BuAQ3~CqTa(x2>6Bh<Ld9$lcz-0cKJR%YL@BmmJ-&hCgz#w!-hd<
zVrV3sU`9#6Y#gfFAMwT=EWv0XaLunhR5qK66<t`lSimYcFw$|tJO7-@l>4QxEHyXQ
zs4K~VJwdJX&TD;dyG-VFQstf24Z>k!Oa^D7yEPP#7K_cPG-t6Ff#ddQk*1S`ot^B>
z%APmS`Gg{R0OIYz6K_aoSLrGX_VmhcBUe2C!_6!%^^Lawj!0+BGxx4pdc{I{hQ#3s
zxjQM?lBKBr1Zt5SlAzgrmERw4WNEv4unQ#U;wT54elk3LhqudSqSXTCGjF~%r4eZ+
z3YL_i^Zf-S3_pxq2_r?Lk<QGR6KVbTiFe5HgfP=-n}LqZ(@XUhO0Ha_yK#3(IC_9-
z_EL7{nLtX5C<g`2ZvK7aH5+GDl7OFkf$72g3=6S<QWfn2-(N*2Q9ALbm9?AVX-%UP
z<$3Dp&<_;E-3S{Q`MdcpgyRb3tPx)%Mv@t`qZG$G!I>ai2~wVrb#hza<dREJ_=>;L
z!fM2GX|(%?^^DcI21x<kHy1wty@(0lMrT5xJ>`Z(UykSmBD0TBgy|93H^Sb#ND%Jw
zbi2oBU0{lcu{OQIS9j3jq}ienjEPz&N7~Gu>WZ7D#?0`}RD>hRhnb<TeAb-FIp;P$
zmilHQE*Hx5ejj6-lc!6`tYbv835KaI=3>f)>4iy5FwM^xp;koH&0ynq12Q4U6Ki5&
z*(B>;`uzJ5DYsVKwzy+K2W>4UGTx`UB-i}7e)_mfJrh=Stj?*oTsY(ow^q$P)Jh8(
zyRh$bSwl*@I1_d(m@<pxEl{0fTcVnq)vc3;WGnd$aOhvW#G2f@4|DC6x|5Y7+pY5#
zc+osqaWY}*2IYsQ`cB^TgLk@9cC)H7h6?AKtvIBO$Cz$<o7h%%s;P>1fyJ2eJ~5SU
zAIMeIRW2A=wzBj3jfa}fHcybOXQe1DatDNxFvz{&QeM`|ASp^Cjb6p2de>W<(L?t|
z$IP(8GkOF%vBH~*Ep3c>YGD=grQ`~2QhjefZ{4dzXVbC_XVetrWjbQ=vypxg&kJW5
z&T(`U8cND+GPS%2Gdz>cbO2?s;L*9ew}qS`o$g}smuysv=J)I=W7mmyFpD`Px(WvF
zfg4t`(U1nwEm+7dBbU{5Q98~$;==84o#CC$5=RBT-Q3o7LLif};y2+9mhlgHgRi$z
zylE=56;+n+h;+_JsZ(6jc=a8Idh+c#9p*N0nFQE!mEudT5>WeMrLAzgd4}qTdHZZr
zN^raB<{|=!XU+*kLW=k+PljE?r|V~1_~`e}3H5sV34OHzaJS(z5}Q4zG?Wc~5aJu=
z^wiFJ-q2K}lSFPmNp51!#KsZMv*~IBA+SJnT;w@Vf$4l5?GGcC?>+Woc(fgA-6QZW
z(__ei=;`kFc@yvT*`roH|70*p>}FKa%7~{5wvfL|)ngF0IZ)lgQ@^_u4nZ%5B^2#F
zMLW1!vMbRj&+QGj^c932Ii{|xr)x<>uPF+IxgMO?`6#iac*A<}@_3Y((Yr~v@4it=
z%_P{a|5s40)ABn`myhW^eFY!tOc&u?w<!xAXp`v^Yl8%5^?79nnttS2nl04tec@7@
znhz6VMdm+ESpO19k$Utj-IG=g8T#Ib$hloQb%Y6*`w|(L0>y}a_Xes)W05fuz(Qp`
zhpN9zI5X_yBPBPm`+7n_!{oeR%D3$e=FS(w$aSNlkLuiXsGO{K$q#cY<%@q8zXM6|
zOAg+9#&?-~;|;qlDI@$83wd#j{4D*(jC1PJGiyzSib*ZVp9UJtCk5}w(j<HliTSRl
z1r`|!9vs<S41%tnlf5K=)_=w;VSyI=9o{Q|kF+!5|KsW{!>ao8?{7t<L>g&0)S(+h
zy5Z0*Atl{NH%NDPcb9Z`cQ?`@CEayzX1+7?``^#KE_I*1Kk;5`z1AhBr)Zx~3n>cJ
zC<-0XD1m-NlN7y8I52oLp(8<0#UJ*dSR{-5&`%obdS+lpP2_~-0W2Oar5_eakYKjt
z^nvs3J%QiXnbi4P_0LwT8e)6wd%c_#Y>l9OmbDKS=k(j^hyo(wolc=QBuhdEXI`wh
z-%G=53Fi^U%oX`c$hJJv)MoWu)0Id&B7%u>_cBiuKK@Wl!oW%>PD!`9@BWEVMxJj0
z4yHneHi9n5UFwr}%#>)BXSMnjOO>cdQG)zcJ#2j)<>nn&=67r-b;n0!>JjWjw&crM
zRp%OKJ$Vicg3)^j($&vCXMC2aOGphf!ISFyihZpUb<7P`EhweaWQO>^(MIjRL2C?7
z8e(+Im5$}X>et?nRk}LGHGvm<V^Oewn>1HF52c;kT(Q*7meCd)-AgW=Hkk<I3Pf`5
znN4r|?_DUQZxtLL(w^O~*w)lu2s>-+*9PG@aH2BF+2%!t-qy7C*+nLJO+pL3NhDh*
z*WEeb{~2fl9op4$3FOlRdpJpcF)51szWR!X0Ru{}@hwDAKvVecCj{t(y#n&D=2BLd
zV`R@dNgW(gKcI^#+p(U|RCtG&O=l<<TU{NqNl8f=kbd@-K5TTCpwCw0{d;d{CYGyT
zqB-CRqcYfpUrt<rm?&5o`)eZUQO7CD!ajD}gixkJCt9NkpmU4fp@BIW1T;X((BT_t
zQL#EAU0)EsOBsv^#aT6eZqH4Hf9<zuE~}0<4RF5<5*Jb^m{eJ%Y2Zk-rrwS$ETY7k
z_?%t3p4NNE0eiv<Zm9I%c=EPS+Q-vVSrv+uKG>|<#^~$8A6`JQ`EVvg?9C{TBrZHH
zfwSk=K&-(q*#L}*3vVvya|mp;G^v94hU#UTt$z^0QCV!wc`@z^eh5;4IVOL~I{`<3
zu@Hip%(p2Mjv<GOoYrLsQZ)QSPjr<>_Pk?ge1Wruz2PJt5gCj7Wovv_lP+5JJz6hq
z*DD6*aK2*Rhu#GqyrMJ$Eg$igxSB2zT@dYVqv$8<-#?1NG=7M7e<nKJ(NTM2SNv~5
zK;78aTkC3MX)@$Dmb@^r#&?Nn-(lXT`~}#T#ugZSo4}EM-{#<lrDYYM)JeSzd`uqs
z#DoHjXA8q~T>!=Z@KDePLXT~zJYg&V!!vtt%%{YNZ-HxMIbrWvKL6eOq4lbq)x9vz
z&T?5Ho>o1R2<_(gnv#~slXQ`GPBgE3<Kgh(iZ5!Ek^XZ=xBhZgZ}37|f?(q?<SC_D
zG6<@^(8Z*k{L#%_|7=YxZ@lJX@Q-^+Mdc_rul_voq7$+Sb^%*#JS=j@uP1Z{A@gi<
zPPFdZGlO~xJNHuxS(|id*7O<<CriXutuK`tDsA319_w0ZO*wK=*;US4E_Y5d2~`q#
zSc#`Ql>29!G&V+0&3VdOuKK-w#|o;m8)HaFmvLq`6p1P4t{u53He*8f04;4ES>mEY
z_d4PPi_57E7EoSxP%4D1^}MB^6|}Hkg?^T77=r6F?|x3=vMTfalq$W4n;;}xDwY#w
z13bwzE?Dv<EB1d4Gx-fk`G?*bxx5?eXF+;lpW@I6#&Ei3pyw5x13H7aZs?Ruin8;Z
zqD&Qz^T((q9@@+4())Am@uqxV@XMi%%c2iRml*qOjf)gWHL52MG>dtc7+Dsje$No5
zXlRe#f$U^HW-fXb*)r0zk<HB*j0`!3aL|0RpaMd&7x{Ln)t*#k1)qqfMMYDmYvN|2
zt%>)3#^TAZxx*aPOP4q#nZPl=d%NEy8@UK$$b?<WWY+yNT@439#13_*aT|rV;T4cW
z*Esg;%rGAg3eo{*C=D?jMTF0XpYhW<_ljhGkX6+e!j<CChdLjmMLMnuCnA;UD}0iF
z5!w!u=G?X9)^~QQM2<$_=&2vPrMJPa7J2Pvd8cr+BY$X1d%d0N!%KYW<L+59gVrc$
zJ9IlPMqO|MCylg*$Em=3YgePNm!U*Psg!<j*Rrir&A#>$+QcjI)iN!2U;}M7IAZGl
z{qTOE5VL_Ff{g8RgC$p)C*yw0Q<L}k6*`;!s&FZ}j2KpAg`=H3Z_Hf{EJ^YZf$J{C
za!m!Z{L^lUb(33Pi?+d71x-!gv>56b^wx*C0;4mDJNNF4rZJ;#UNlU5<qEzBg*9Ph
zV+Ww`VA8s3&#9q@vTsNTJ&evbeGLjaHHq1`ldmx<alb1oEx->m<PFl(O2p%56c|KA
z;Re3ZJx__83_=@-Yh1~tT#T%MDSy<dt8L@v%*hueKMYSF-G8fkpl_u7rXjeP8PQ_6
zo-`4>jM4|nCn;ImC_g~>;Bs#Y4t;#c5Q5vilVQPE1xX~6adD)@J}EQ<zr6tK07dSH
z16n;E4jYJ?#FK5bkdbD=xUR&W%zjIHzrYusikRbEsP;)Lvb;8m8OY=OMQV^Fi`6hp
z(IB@uOP&EMpYUCeFk+|q1f26ncnPl}(7VrYUei{EKo#06;kk@}Oq&=P>8MlO&eY?l
zjKbhLiv4GKI3CXl{D4+=FX;)w%vxdTv9rX`^Iw7?#^FBOIBR4hXrMyj&$;g9Ua+wC
zOtyb!ielQxVA73IxEziO>nkz)wN~Ng`-LKWi^$OY7<b3w#Lf!?a~W=qic6G>)iuq2
z1Kc3x2ad~ku-SE{{rdK!L7vYZrY}*7T)9VV?=RoYr&82PUWLY^-3jWO)-O%j^ayfb
z?w7CzwJ}>b!Zz05B2p#U29W5#j7^X&(uYwJzd_btpjz*Kz8HeV3oCCAjyDa|kN{U1
zv)lk4f;Q>~q(-+6!WU@PhTTrfvEGnAz`z0OXf>DW7;OvHfPvjh_*Aw#ctW$LymV}!
zP<m@&(|`O9&5)h>T)2+zUn<>?I1+wnMHa6U=b%u`0r6p9x35VQ*?GRPE(mQ-4WE*%
zvy7Y#E%nBNy}MMU3}za!Bc{tQlCi~@%0-i6(#X%?IJmS2J#$aMP@uX+IB-^CINA?i
z-y7q1Nt67H9QFwL*>Jl#aO$FQSn?0=v&e<3x5?mc(ZBbHe?_G;u<-~Q(`yiCBog5Y
za%a<z(9gI;t(HH?e(S4RCB0w(N8-M*063aWbfG7ERgjup#w2l|FM*l-oIAT`DvL<u
z%7C8h&X|j%_|2Z{Fu-uj^UTm%@{G7Q-tdmaDV1%h=%KV7=STL2q$N6y2oxoGWN(+;
zGu4P#?z7uh9)%VXxA9stTfGR`*m_+}-DQns$GV6k_|_lmMGUXMv=ZN}YDvemFMOE%
z%JJrCiPel&1k>6Gd=ylWfu8XpIRP2wV<b!%5u{49x>_XVdg|uR=+e<U{#SZ3Eu+jn
z_W&d5wrfZ1f)sY59$-w2RD*+t_8lUySKzBoQ+xXLZ=-|PQQ#Y1d%#U6?vUolkzc7s
z1^rpi{6IQd?Mq_jmsPtmoLMYM6c&U*T4UI0dGt#7E|Fv?sV_;yg5=@O6yXE6j{=3<
z-5Y4k2~TkLr{8+)Fr+hy5x$J!5@){0rN%d4Viy5f8mIjE;JsMF5#LrB9*=NLlq_Y7
ze>EISl@rnFg<GxQ_7Qb}eo0vHiR6}!#F%kEW1r;#6xJ8pl^XF>BdcAmnB23-Vl=Qq
zWF+|v<wqtrii=693Q0*I<Z=-vY$*sIQU9ehgjGS&98Tu@d(VkXE2$oHL|GlRa+tk>
zR;?x((d|!7=(je_U2__8FKVo<rkwf#+nCS8>+$tul6)6fl$2ZG@kr<5m%Q$wf%W?P
zk`wk0&31t;aFz8FQSh5~TIu9^g@uM(gA|7<t8w}`Po{vjaQp)<@O(~z-QW9~R32OP
z-LdpM){(GSZDYY21{;M<#`5USgQPxGQc)*D)B69PsQXL`w*X>W{1XBuf2$6rI1|zw
z_mJWa%k=9e#$i#+_TG#fJ`!JRb7y;-$@^Ao0&*r`=F0#QFgg_{Ek;NE7}k0XOfPF4
zf{X(a8wC3%R<cHW8spYkjc>j%=*K-JVz0BqspTRICo}hqU!r?Tz)|dpN(E{!b+Mxb
znBD_od190x^jW8`n~mkY?^iTtGn*w84bs&eQu`CDVky~fHm}td6Q3>yts#{LaT!0N
zXz^jKSV-JJ{G)H1M2eH-{zIOn9|f~72&|D4VXC4(J0>*cdcQ_}CENmdkf?u*_>QrW
zYs-i{;s;9*T@gIBnFn#NZ1b8-XT_YCjL_xqKB<CV7`3-ff5gC4*-Rs)x%iy;$Ku1S
z(VZVDrfol*fCpRF<{&R^43sNq*NQOVLZYE_oo0`-`hptfsTiX>6*=V!mj5$fz1TxD
zb3#Tv31nTLRwMXr1WaiKE}u{6g<RkU1)KjQ^(<9ekMN#d_|!a|b|a_vE?VM2Ijk$?
zGE=px(gY@%Y~M}W@Ltz!f20D#2U)P}p^;|m<?RgYpgDK8v1-{LT_DYF-;=-#=%yO=
zvN}BOkz!QxkdNOuD;Li;NY`2XCJ|bF&S<%qWE_lp*h~$DUUkj}7~LzE#3sv?Aw6H9
zFp}y%MoBF=3_7Ox50|q@<|<T5O#PixA?rEOEz9AUqyO*g@uQYh2Yz$*`Q!58g>VO&
zu&nU2-ZHe8K(qg%p@cU)#XbuhrH{j3)2OYDNooa?(5Yf)EqOs-xthQvx;?$k+zJVw
zN{16FN4};xI+}<%TDnT%=td9<hjw02o>65blR67RCq+Qc=L-8`f>8leoJ{YDp59(l
zx@x@}k&-B;z3tXKS#jYJwK&)*l`C>VVDg7p$ft&|L+c^$-rW6PL)goEz`Xs);5K}X
zn+8WFhVoJ?U^`cLs6qgh7^@YGZ!;;}NOJ#YRv4+V9BL@F`bRncR~~57TLz8`%8`jt
zkf%1zQhpU{p`7rJki?h0Uw&Nrw_V4awBZ~$`BlbmCJ~Y;^AZD1%XFBxBlU_R;B6$l
z7x%k<uZioMH-H0X>>CBQ?esH8ey2wcf%3L2{%7R`?2lMbYi<|P{u(HgN(^jBgOTW(
z*{I;@pdzoXt(hXlAk!^xk?qMD1d|Ram)aRZTeg{M=rhyXPsePz2woE)KB_oAVbq+E
znZP1WUQ+5TDz{@Vg!4<ydD9mmzPPQ^W#xz5es(jK>^sT2&NwUXDh)G=w79LUC;jfu
z-wEg~vx}+!2#Eg-=Ro|pm7$Ro=}v<E6$62FQ_C=CjBqOuP(P4dKE%cmzGpuNa1954
z?hPLlhS^u0gknY|XC;ds0ef{EzzWY*>OCAX$55<zkT4HmMt`d<G9ZBo7X?WFlCjKp
zEgIG${gNxuw$uutfhPowJ^e3J+vqRdpV=fNf!W4(B~!bk+|P>AHBV7Oe4veER-LBT
z7Bjkkn1|$<`iG3a89U$$q?LWof#bZLEy;kZu6~^w7^~6Q7R^Vto^A7ge(+CTHN^Zq
zmwBEMF$;+o3QR{HMe5gCHEfY#rs(DqII5=-c4_?KK<CxG0G9P4aUv<zbYQG3NR*7u
zU87Ayx>!43*K5YAsB*Iz&57JmrbIvrDXBz<&oBt?x|hh~Hj7KdzmZkyEugEut4`V_
ze+~w&3h{u5rg^S<fCi#qvu9XL6nj6y`1;!t_|4`(th|M4m>0=zh=?dPD!B!sWurhK
z5|TssE5@YrE2qUegX(nWE)Cnu0wII>@MCedqMP1wc@v&{g6PCXBl6w;YSX989t(wi
zKKfBqWnbllScc?p9)Fwzm_2(hXT`>Z`-Uq%&<tIU;^iG%{nod-_0k_tTlIDdN)Vm5
zmjjnq1|<iEb1~RU@D$!-G8DvC>~65gtLZzX6yiDj?=w+(57i<|LF=I|%J2uY{H#U-
zm=mJ+X|!V^jM$aYXEtyg!d~5dMZ*BkK8v)^6EFmD+HiEj1o6lNDCMjXsz~y4`Ixa#
zjaiu@AdNuC60c~|9mwwq#DN+^g2-e+LQ`?_qMR%w!+69V@3(k-TV}~6VkfA^PLJIH
z1f3LH5|f_)n9pw@1VW|RR1+@M4rFbd&w8=^E(Es++AZ`ue!gvzQB_S0k3FGfu^)+M
zB5=P@vBp!q8F>f4$mM!n3LqIo#O?rBz@fdcyKT!}w33xh=;Qe?1E<lKhlbDDe}(wN
zUYR0BxD6Zrc&x-L00MDLc5ET$(rR5uqUi_}4>Am}q9Snb(MnE=0kZ7~d~zg-(d`pd
zuS{qv`qYQ1s_?KIAa-)pX05;(UxUcddKq1+9{1%&N1{Ks4$*r75DE09-#B5vI<8*>
zs<wQ*r~HC~cP+%v3hE~4_2w&}*`b7ZN7xqno7p0IZTGWA;S$&tE~_?h3VZo{r?%6?
zmtM@gM`)w;$z-83<<Q88dm<L)ywN~u(C*m1c$pB|2Va<>OouyiUj26FlaZ|C)3^aT
z>Ft#)8)hB%`$1QQJ(dTX_q#Fd2hx$hS}=2ejG>u2=RBU2i%0V{6qaYx3k5bcod^qO
zuXF4=HUC=(`uFE=0R7eSL9~uY9sJxTAOXO{P7b#A@IVyZH-ifzvbk<Qa3u!$q)n>|
z(p@A9cn()vo0S0YdrSt9Blv@%kHU3L`f5se?KcE+QKZ}b&)YGwXuWBmrv-r2Lj2l5
zdc2_sSv4mG&_@^~t1-+UnZ@W0NN=BTc7y?R)n(5UfQ3*}x}J6+@dT97ChEOspjY9w
zMWD)PoKU)n5#DdYW)<kt6W{CqeEivBNok^sDLoHg@vAMNV!y5U!)rM^xYGUp(%KAJ
zydMT8SCQ0x*>;?#0KQ`H%rJ!$yvDxu=av8V&8ju8!U6#Ma7M7&pE=@lIh5k_xF|h8
zKUk#&ht*g`_PzEtXukjo=liM@f*!P<qf<J~KY9uDns)R;h8~HX8=^_<R_S9V4SM}Z
znunGBgwx1g0KQMjDrS0On+X5QkIXTg+gFPOvmq~lu*saaUNHFNr<MhT<5W&0|01&!
zi2oJ0M22T8z5!(JeOnG9vXOJSdcd>d0uH!ue|c_h`<~QET%ZfM6aOWEFAcn=_)`M8
z1HQnJj)*d$i)uS3Vt3}Jq*wu&X_tik04nz3!5E4Uw?Fh)rn^g{P^D!0J+CV<fV1bE
z<!_3xWY!pDm{zJzV&dX<j{|`k<pbq*wI%Kaxv`SACUTi5{=BmbjCmY`Itevg{kc_<
z=#%_AG*wms^usMYx+quEp{&tbWrT3wqy|>jgyLe3cjp6&v#Zg2W}eycK4f1&^$)JB
zA5kE%(U42!|LbARCG=V}gCYNRmw{e{935>=8tx4=L%-5LE(zg&0~-q)ye?E4@)jRU
zJ}#i5fxqnjaNEKT!SO=msMV0_U{J4tVCBI3_Ar%qRYX`o3>9nuV!iP>J1|RpWdvnD
zrSQ1cHB;}rD&xzM-@u#loS;Ns&N$px10_O46sgz{kj5+PFV3e6%x;u%|8?KZ8;DEP
zH-y3>MA-?OF$^@fX%cu3UkP!iStQ-(OUECP{vl1~hb9t(o>uqfnnLzru^;8Y?&oBh
zgLS?ml0<2J{9xmNjLA;z>S0F~#GsfT`fSZ?);Su%7}67pR80o~oHytSq=oi|tgrr!
zvfaJ@&_)XVj~lx}qsW7L`MrhH_E}rV)k3noce&ug5s)<tBi;oN2Hb~z11@<rHObh>
za#V9i%2>*wFCr}eFszuMu8P3J3Zqz(Ayj(h8y`<OW=m&1b;IC`eOJ1S>;amEhXD=;
z>!iK|n7q@?8&<YRA+ILsJs4AFgBSxlvnJ@yhwO30gu~%7uqhupY-8dEX0SCZWx%SP
zJ~rTQNl8!+@3nW|#A3MG4LoGoLRrG(Y;((vtf2=NzR@Ad1u84dZm<#kecs_MU(Cpc
z@mL(@u%}E~LXHtjK%t??%Q44)Eq#%3aKc4)3f5;#gXz&#)#5!+kn5iq2-xVlfB#U8
zb6<oq?vEJ(Sf;5e14+&I33yVBVNQg@4I_=nYaDu_oJM2#Jo^zak=CSJRC)@gODvM5
zwG*k|H`M{Okwy4EnY%x?d<K&`5FF&V%qBOipv1-Q7oY~5EQby$*~opdpAx8{{RCcb
zBdGZm+gGdY+G?g$Zz}+JicG}?tZu-$_hAP?NZl=CxwJ&6t{-zwq7JQ8it4UF$D(Ze
zm-_1T)d8^QW0g3c1v@6x<N6S#0t6a-RFl-nzq8t8Z^onN6~^K{htk#C{WiK65y5-#
zjp;K$s@^9*wpIiIjI`+khncNF`GUib{X|RrAZy@&C}^W-dl5zQ?jr#^UVd&(5G??P
z<I`x!LLD3KFHIqT5RBbLDl1$R!$XlCAqQlM(qZDD6{p;-b#*6DrnQvJABx>yk!FS&
z6XoX<%=6)k@6{pG6iY$vk^Y-$5&n_|1x*sZRh2~SF1e}l8_`e7c0lCCJ@_`{5K`Kd
z`N8OGa5w&r5>7w$sJ>~Q{ldHRr#5FaLOkNP&Nm?^A8mJL>NqzK5xxrOk>rjRVx9{e
zN&u>&FwcU*vT5J!RtYB_Ce5S-i;Hx62MzA~&s*g~b^8Bte|wDxfF}5q9&HNG_?b6T
zy0zc?(xgh6^d0r#BHHXy-4gy?t{-nr334jnpOfQW0l<hZ?6`t`E^xb1s#f18{WI-Z
z^r%|9&tN|i`A+0Y3sJ4(8*vnyPVKH-Ylxk_sbulSDMgJXq6*@m&=c@|o=T58J^fKu
z+Hr#FNXxq)2cUCJFI2<DXb?C6=<><*@(Lyoy+v_2&62XK4GJR*Ex6(X3r{3Ljg9dF
zq01j)_jk-GUSF&zQ|-gBEeMxaY6=&ebx@9aSETaGZB3ST{F&MDc8pPp15p%M-GHQ9
z?%(3jBnp-XVa<Di_FEmI6AVJB!2OIx(dM8#XfMkj9r;B0?t4CCi}7k=IYXhyJrg5v
z-!Eh`8m@8KOpWo}Aq*bT!1}EyV2(z3B!)z!u!ly_<A445h<^QWdk2{q_9GW#V(UlC
z#8%2AStfrZKiwBJAqmKj$e3|bl<>uf(>Tae$mFlFbDt19hoONgrp8B(i0^MxrSg2T
z{p!xX8(6AovjiC27TswDQF}wA&8q9zug(9po4-<QPA#Di)}|EgPqh{FTqNzz(cV)h
zk0NR__X%46G(!>&w*K^0Lp71ucu<Zkle+`=W0g@~d@Rdf!4G|vlz4Iwl<v2k*@u8C
zI`beer!#2KXVT=og}g;SB4FZ^8Ayqgs;98W8Q^Tml{{6_Kr(66*ObcDc>m!O?F8A!
zFybr5Z+;j7et!E7*jwdVt?FhEr#U@P5&WNxDPqvZ^u~5HZc0C>^-&|Ym?T(p3?cfG
z&zZX+uSflNw<YB>tug*`4)}*uPDFA7IjHpE`2?jnQH5ZV_62AXqECGaEG$9zR-+CI
z{p%wkC|UdYDV9^hADc}{^aa$7#!`#KKY0Y_)3d`MV+zq%#)--G>lyvUU`>sk#)3wF
zdyqx7z2RRMfrls<Qq9tX%5@zJtuPQK;Te=D0;g}-ok%Bm5!Cvn+C0PkI>}+qaymb(
zM>uU>|LCcHHFgQJB0+7s03x)L@y9JNSb-4cnwGY+6`j{>oYE!{$&p9v(;d{s5Ct3T
zP(r;YNY4_F)?%0~mI?rY;3?-SdM}<nar0P*Nj<w~K;LRME`PNbLP`ydjX^!w{J&|I
zNHQpU8oMaL#-U;+<xTNf5F6^semO%lH<*WuSXXq}D4fCxVT%H)m)`>$56uWlAnGR@
z_8Af5JHJ!FKQ5ba23tl%_KIej>&Z77PyY*-i{GCZ8sx~T|7s-Uf`l}CeV~aX2Vb2n
z`mhT0U~`cq{>rvs(ZxVCOvyVd(od_yzWFZQm&?l=EwxKQei^~fd<(0)6fB9^GZpH$
z|87n{l&VM)2X}hg7gE+omXL_scM^FYT1&hcsoxiCY!PJJ=@fI=XY;X>F8J`RB*fMM
zg-8U3K7I9cGu`ocNm)#>E8%ZD`0>we6P$t%Ff-DH`|88kd;#1iP09!zejBg<0D-?#
z^7^5CmXaa1UBNEhf(>*AvvcIX8L?zFoq+^?=@sgAM*mWo&Sn-bgVibh?rihzJNO>Y
zNM=g}^ImH$|DQ~@!mcBQKS-^&bf!3f3Tby*77pVZI4{Ibvg|p4+(Zi;J*V6|=u;O<
z0gk950M=&cVng){w!km!ByWU;aoB(!$&wxM_d->dfjT&y$c<R3(cnA&N%YhppowL%
z2RBBi*>p8Q#h9w9f8Es1Y(P-!RN&TR9N-IcPerHIvf{4y9cSg>Yfz{REZ%4)(#6cr
z!cT()g(Nka99_u*33@623+xk;GRhxg^l3_U2-QTh6xhw(XskK>C^Ek@io#Xtfh>}+
z<p~Bg+p{W;yR24?lJU?>%0MM7wJ{~7jT(!XG8`9|!6Nf}ZC!FHujNFL2QL1h32=H3
z4XqJN4W$b}48+o4`)}d2{u-cm_~YOj+K2v__5>*2hAiC)C@gT`YL0zW@89dd8~yX*
zR~N_fEnj|trtJt{0g1Bk_$@8_1>)sJT*ha@Sau_esbpdA+)s`Kaavcy5PP<iU8=-U
z%cau-W@S1`z=Zn;&$9@jO0h+s>eXFot;xuD%}vj<%&8nOv;`SLq50%hD*WaH)JRWE
zHAv?ZehgVD#;Km+rzG*R8iZZ`L~7K?w;LRgHu`iHa5Ta9wMxO$k@~nf+r@ieZR>~=
z`6<2~3Jz4c4xaLG054T80r;7(#mHdn5B=Z0dbeBSj(qrSpeF(8gLf^f?JU)I)Q*5|
z6)4*%yJ$^xPiPh%-b>=XvRQ7KIRn6;5i+zcMJ|JJEUMir#MQZ~t2!J6IHVM@vO4PT
zvxC1F%`NoTsqM<;BdmE`<y|`0H6QWd<UNw2lVplet+n4vn5h=&Ynj~inl7?j5iGXv
zV>vthVxFe+%%Rn)7AKk2!R}ny$e;2fm+&hx6fL~LdzWVxft4#%l0Q{+$ilq-8<|{G
zgwgg|h$wN4k7I@{H8e1OdQUIcw=1BA7@*atdA&hbO$6E^bpA5zIiK+mcd{7XsZu1S
zs)*0p6B^Amdj2BoQ1`P<G`3U4UpMn+wOmpCrX9(OV+K8v{BNd=*HP#jU*a<FZ;>o>
z!s-$ASFOlz&G2^J*Tk!Mx|A^bb!9u}k!{5bc1NmZ92UR~A9*8JX`PkU6BX`+l<cZz
zesdy<MiB6jStPT?iaO3k4x;xTb{@5GROi0XFbv(S0nXX5lQ|PsnG8-UJ}JpSVJ@3L
z51gkFoq}hO!2}lmKEAKwBp4jDQ-+Zko~EHkqw-Q<I*r3{TF%Gr8&)-qiyOlI*@BFg
z2SdJN_WiAx$cpo%vE#gN{dF$uW=ms%<nr1##epULV~cj!I9>%7Te{GRGaciNKn!6?
z*Lx7<VuVyghG$2}$MbU=dwQ2U#NDv>Oemj#_4pM?F81x38|Ikw6JDg8_YY!WRe${K
zDE!ani74j;qhwF(Lcd@KvRPuAjhY>n|2r~V4d*`)ZCv%htn>WjMJxP<PHph9Qvr0T
zeX+bP<EbhFS0bO4iqB=E6pTR!c27T~5AnQHdtPl3?{nuB<aKiWd9+li$ysAg;#saq
zXN-zU#0K;@d!)08oFKjE)!{#RMEd$G$KE6mCX9Rh&T@Z?Ou$)~n)fbOe;@5{YS^Jt
z^FX3(z(I=u-_vQUQY-Jiv1Zez!6Sz)@7L=R&}XL4Q~L&)#>mnLZy|*H=BH+0gfjqX
zZpQ{?_kAg-FEUe8*gpM>^`?lr2psq;f-inQZH5x>tzODc)%3j{t7N@{`HB<zt}3Vu
z{jqA`0^7Rg>)GFrbK!Ka@=HhHhRNZ7!|KJ4Ub$jfi^tUFo^((uZ@g9nJc23xDRtv(
z)X2$BBG_o`$H7t$0{PFek!xpV^V-%A4&}>j-}1sCr;U&M0ZCIeB4(QHX5&~j{@oW<
zGBg(NZT7DGcAC)qc5YK6dhE>;+-h|rE>kK6)VZf*L~6p~y{WHY_9ngW<;BrwuQBYy
z<+muPY#XP86D_A7Sk93l{75%c(b(lS6zY|5`WR5TSpqEe4#(=9dl7FCB1Ip|nPB}I
z1fC&35Z_8+SgYDYRRwsWk|CoR*(jpw(@<qu^w6pWWTFWq+oF9L-iFFS90glEzmABo
z6OAmH4%Vxi_qVlhe*OKn-)dS~%Gsp6$xVW<F#f-!cR3l<9PF7yw$*srG3u$}QV$?b
z))1`@24Y()h|KBfp^=2?j^(c;#TF&E_Y^}6O&5yT-1g%7=y-WIKKEoqc+sO}gu;~f
z_;K^(P<Rxuf}#}a<O+N6f0K6JNe*<iM+{{=#EP%{?XGg_<c3E;il?{uCb4tCW|u-g
z`!<hN{@s)>FBMzfY6gyH!!;0tB;&s`|D@6u1S}N^AwQXxS;Z5=oQnnng5jxneNZ^#
z#o=$TqGI8#o&i+LeuviJ&Z)$K({)ndBmhO4%Gx4M4}4f{{ejFClvkel{OEAYuz&S{
zePg?s;ByH0GLp|-@&cqW|2HS<z!U>rhz6=WPMO03i&Fr98;ZePQD8O;<ioqZZ`3N2
zX@92b{A#RLg<;d3i%#&=hsAV~U{6{V59Uit4{OYqSC5x(XA{R)8fnv1CTpC#_|H3C
z4<Q%+S9X)WZ-4J(h+7SGGPn$EyldWx|3xB^a$bVyv_oF>H7DNn*JF2R(Nvw9-v00E
z6hdKb*WXuT`78!$HTbUKVi4h)S+F5O$4i!|v?*!)N4(Fm0FA?}EOT-*O0O7*F>p4r
zc>K)m7nc|i6S*gH`!I+e|7j!Dm?9{6Bzi{ZmXstH!P^b36oyG&_R9Yr9i50~oR(-G
z9ANmeoffiFq3AlAule7!^?LBD^nv+;U=CaU37V0O{HJs|$WwfRD$qBs^($`%qZ+~|
zdJ@2SYZ8HV59tOYn@oMeM<Y=#nU77)YRrA%B7$5c*Nbl%j^@e77@p7fU7;n@K+JH8
zjChCSWWZ7>bZ@DjAY+koI9zwGjd>dI&YPXh+e--b!B!hK(J!l(%v*cpXy1O7hCbSO
zha2B_2+?gfCOjPI;IWym5x18ur)KTbE8HV7m2laCC1hwmmSBePyP)#xHQ4M(6v<_I
zXNtH1TYs$@#o@Lrx!2i7cfd{<`|5<VmhJ9H(G%(!6IlUGE@~Med<2s$If-4-2i&{Z
zdpjVDB@W!R?J=RIBDrzBgBz*Z0&AuL*oX6R?Ytd{A=+1T)_z5Y46f(_joDb(+xy{N
z#fSeqH!oU|pFcHZUj4jz00nwXbFe&yA!?7bvmFNcYt}3S+~A)<YYXxR5}gZ>;o8aA
z^-2gTDClXUdN0nOG%LpA`I&Qya`#7K?ni>hIyc@&y+xPPTxMxolhJG@WtK^tloi_m
zT(|(TrXfon*TFPL_6byUrY+W8d&G2~NB5mxanFH@<?x`2;3jNzbQ%KFglLL`*x632
z0Z#}I%wS@R7Cb!sWn5SAEvK|~g)rX{$4fID${>dHD^|w7js6i80DL|P<36CUQ3)jz
zWhWZO`u3+Ms8?&ng4TL0M5$sfnCx97uDs_BW{S+YfRhS8C+a{bci`Dj87cRDX!5GP
z_%H3k;BS$)IFV!Jh6CS2xdNw-SchK#R_LYb+mg^&U$}d8u++r3yjvUJ-8MSGVv=5Y
z<1TIzue(dK4OPA!{kM2$xCQROq<}JuMUo$=P@Qkv6q@)0@#XFi^<w#yN=f`>s9c{u
z+Q@c4TnR14l%8eovg`z8qy_x3htq1I-rB+#Q6tUXG*}m69|n%QQ8ym<-;zr12YTXM
z4QBtwFz;5Ha`O9Mkgs#|mdpWvP+warcdACcy{EKmC7pr^s@iaL%EK3&#A28ghRY_Q
zA#TAW9}^Wl%>cPqGA9@MJPOGb_r+GXNsNLl*U>_)`gCO3x^9>C670B1HGl)#c-y8b
z&+!zRw|mD^hxhFP@^qap2r^Go(USdCx?F2k5VZO_tfQj@nL|I27^nk@26FWsVX!0X
zI48Jg!4ilYBJIY-BG*=Xs+ze0DWxS1qQj3pcldrz%$=g^^iyxI{!9uDC<AX1NJm<#
zdfd^#HF*@}RNkcu?arI5ekjeDHE0)&(0o<G-}6j4IGLMpDLq4yr-nloAO#v2r${8D
zWEVX=OyD@*Rs%$9X7!I!(#Yj2%ra4nn*Ag*p$Bo3x>}7i^+wk)&~1<Bk~qNVxS3`}
zG5rQ-If*zz<C_?a@*xQGJU+UrF?i?$J9|u2JzpA5L6ybp(xHOwCA8-tdEzWR+x)i%
z0jU6Lib`5n%B=?R<>~#M^}{XVA;G@Appa1QAh+Sgyq0SjQilO;C|qeT@4;`6hhwx)
zpTJVF#F2u~85WA3zz~h|eIz|(7um2#*jKAk*AF=R-Dcze=2l_i(X6*j)5+Ja)D`h6
zR3GGpQ$?RXy3kxqH`Ie3Kz8E(5HryI4w<IC-4cDi@Zwn#k<Yy*ctwLdqKB$_j)v|Q
zygQCVc_;W-VW)C*_g}#`v(^Zx%a-wlUit-$DJA#5^#)VaPFV?<ar|&_jh?lI*^ZLM
zWss_(E#n$|lAu?8tcU`7tIb}FZCW=;!}zW{yx!#jD3hjC{Hj+*=XKof5@S(BXy}Z=
z<7*-NXD;%3GeE}Zctad*3sel(j<au{0Pllx<)_$?kYt_ip4X__zfi;AR|u4kOOo--
zV6EAW5f9X0QeysdLi&|^oly;(1{+)?OQmpIDNBj(Gv*ip<#h8}zO?4Y$h7{l_(nYB
z<WZU?va<f7K;twPR6eQyjjJ1QQ3AsvqK+x%$X%%lNFoOA76SE$rxi{Cv!EHXVK|jq
zRj%5sq=p>ah=K1<=H)u_Z2^B!z8EMSfZas`+)J}+CGT1Obm9!r)2q)4ZEnueg3_Fz
z>NWaSthd1A_@=p%r+EnF^ky_`5O=r~&1t@&-wbj=-D?l^z~7F4-oE%vzoS$dEpx6_
zKv#LJzm<XnQ%&lqVE?0gwfjE+JYZdMK><-p+eFJnGw~AT{^sy_Y<N%-H{oF6eC7wj
zS66qSU^sOl)+h>@)QROs0iS&;@vG75@ZZ-+&@Fnte+X0;v{a#)3EtKi8l%}#g^_^c
z)yP}_whCxjs%i~L7B+$StawR&gG(oKnM&>>r<|Rl(%~ZFeG0}TOJ~9P&~bFt|AgWh
z#5x!P+BH-_#QoL^nxj}6s@{*xeZ_FLlYo-bBzplw4V&lHyka><MINpW3!b2)^Kt`I
zsU$6(U)g~qBj~8CQ4cpKvKWCd01x8U9bgfL&6uAKAW2H1RCm|wKRC@2%Q3RbA^00=
z79L{f<&ip#4mc=7(`8o#|7h}fF)`uLO}Y;AWz{mtLf+Hciv7$aNTQSs+%Nw_93Ci#
zyDU5th}P_XtdVa-5|7U_G6Gbbnroo6+)ePg(+3uP1tRr}ZiEw<N94Vm(b2k}6eHS?
z#Yhd&VFqrMm?H}CI(pu78pT9F>ElGBF|qr$zldw#2?>W0Dujx|s6@F0{0by>#-xx=
zwktBUXv8Ac@tJX&O!(e>z9s~cK+X1youj3Be?Du|qYxAtG+!W6!cpvwNu@$IFEaVH
zx=K5J?_5Q=PIxsm>HC%WvgNn`5?#=}LdSfZxHls&AM)aa`_iWif<q$W_Dny<QTP5m
z=e{zq5`rdngZ3saE;?gC1llNpRs;mYEDG6li~)kHP`U9LzXhKAKRMA{j+U?N^`hua
zw2Q9;nMNt*(-b2zJmX2aK6Q+es1SNcn0EYiSNaHW0fPAQXJ%A<=HU53%{)Ay7C1Ed
zWow@+_h#m*%4J+UjzNCg8nY$eRTC5U;hLFRJI2xYL^qb>=<n7B^^nQWZTTX>|A9<F
zT*Pvq+(Nhe4m2;yCNi1x_n7jZxN*2H!BG1#vb<EtJeQH-BtV6|6%K%s7h%kjr&ay6
zF7iNZ;|2->D(+*FSd75Vk0P*I3vcf(FoZr4r;4XdSGWDX=7nayRylB=H*(cSJ#~SG
zQ&(j648&JR?lCq%IM~~x01gXic<5VWF_1*lM`-4u$<#M&q|3^2j(e6-R-S7GqcQ=s
z3t-n+XQQ<TtJsBx(u)YG{SRHz7Y*mqp-SrGf{jqMCguRw<$rdKmXB`(bq(R+_nDB%
zsd#)5Z6HX3q{qwk2>?kjX;eSK_`!G!;IYj{i4z9{1c+kGY3B)x`rq-U$JrNOn$ufR
zeu48B4<D*pM|GiravKV>`SLgbr#S!&FSY@c;v9<|*aqwOtER;baG^LSX^{^k2LWv7
zE9L6miw9`1Acjf;%oI8LGXopv-Ipi1lRl$6m1DBSNv-%cdO=Xkrd|3)-(IxpsRB86
zZQ>)RGnAH4)N;-L!vYAY-!-4Zm4Z_{a}xk|`UAA1lN3S3x?v8p`|vy~5v7ChR4)fD
z8h5y*oPX3D5IHpSQV(mM?ILr?Pj*mbYEaQfg{_Zk;m1DnB>d%>0Kmdj!2zh5@M6U6
zx0>GPI%!dS#oC8lr`YVB1LWbZBRr&ykWX~cXyg(NwZ1bcw|5j%?c*O5s+1BDj~=4A
z*UCbmp`lIttsfD*eT3{bzZs<@RLh(uH(Sw?Xge0=0b}OFY<H3rqz7rFj6wCa0ZC0*
zcP^XH64x2rC(1on3Qu)A|NYos7sx7f0GDc5U^RTrCWPVyWK{BA^Uw5zwumG@QPVkq
zL?q`v4Mk@@p{qx>E%6r}ppa0#;<a07`4Cta)L)hB;dr`fT!E!7)_3UgM<+pGhoc21
z_s}~QgCUILny^0W&7L5j;qzK}7f%?g`FhFBU}?J|Z=U8!f3AiKjgYV$ndthNOM$f@
zB35Oo$M0e<_#79-a{VI((g5E^0Y<YTb)DT1o3jt#XkE^&VYbhU!#2>p|MMrV>c{Z+
z3YNprAjK$|IIGP@%9GPLDT?iD#{Y*8qZRa%JW%Vdc0tAf3}w72S0pr-%xaQ-d^L$A
zDk8FfIu7ha6Y}-<Hzy3QWx9V6gu@H9AOKaN7X1d0316GcynViqAt}6qS6>^6*i2TM
z>`UB$HnekK*8c~2`4vuPvCi-V=z1Jbl;w#F1g+XlOIHDdf;-KmERQQ^LFmeABeXyh
zbB)r+Uq6z90feUsj_Y8uUh|HZbB08Q@^s!s^x2f_4kCO+*wk^s74mKhv=7dFqrVbe
zoXVyWz2G@&8}9|fo#hoT@l+8zd=VOQMo?^yL=I%_|Bk<35$XDfufHc^cwL;iKR*yJ
zvCmeGNiYKDI)XN8aBC5K&y=^suV#x!<zLvoASO{zH~^Aq2Rn?V3Vd<e%DDw}p$~=y
zaPpnSN=4ax!0tGGT$S-_;?=S9>LdtAGbCbm{HsBw)zUB{-!DeF8H7QfsLkNoHlsql
zT72(rCB$KUU4!_j7aq)?ECl8LM7F@dUzf%~n^XUoUlY7fVH;V7OjJ%wSB#P1RJ&1d
zPVVtI?BHxW2d<bV6I34;={<lHBn|<?ivI`i+~5Xy(puaQEj}!wcmXu31Fct<3<L1$
z(Y$&}=Ad++)qoB*c8**@K<cE|BLD<ta4gHRF*$LBFbc~tx;@5)GB41eTcDF39GMV^
z44@@|5fJn!c|S0p|6KhVxlfQU@@}*PYKsUNoMWl}_}iP2FR>GMmeK!syt94*!z*N!
zGKhg{TX9!FC#>xv=!}1M$5W8aVidJSuv_J?GeVUnW%ODq&FW<T`-KgBg-dJ+Tr8fh
ztHuci-a{OO0t?z^(9aXlb~)f(;53#30{hC0iEoZy0^ko`%UK!TYm4}zOp+zqBj9#p
z#|3v8qlRG8fQgBTmwGPezFEyGvjd7Jo44_M6Q7CcTTVgd4Ch@OKv4~wqZUK}XvD?h
z8vviR2EcS(?pL|p+1ya6&}maUV+t&v?h&02u<)!rq|w}x7b`W6Y3{4{_tcEpzc)Bn
ztR){HO3RHne+Y?u@TZOE=^r;4)osEZ>+z8h6CUoIrv4em#xW`!Ycc)Lf9^*us3Z3h
zxCf&Hi6hdz?fU9pn#~TJ@BMApcFI)%qYGFzJ+<60QK!F_qJ*55R;oX$Va8b&0Q<bA
zuebR6Vn{{F8j#5C@l8aqxjxZrs0PM)Jw&6HPWMF+S$(e7zpJ1olaw)%)kQr;f-e4A
zR-6+U-34V6m|0t*TA7W+t;gyf$@z#LWazRL#%@YSv?Tmmn#?KkUcC#*_8h=cw*>-M
z?`dIV76nMIA}BIaaYPl~#s`PS`#RDNE0)DpjGlxLMRbB+kQs|!4czzvE9OtC*ObLz
zk1(>e)1*5&^}t<2=w$AOk7~Y-OwxmfrU|=Y-VyI7`Y0yrDVuGCyW6NCn*x%^nX^yw
zC#)P5_w|yfi)hr!Vm~D|zW;9|sDkFj0(dWfY|FDbA4WgEb<UNW=ndMbP<pw)I;3b=
z0XVH>%9=Xo+&Q%&4X4<S=}Pup@<9tIG9~n4^7-Zd{{C83<0vgaPx9dFy-!c;)q3-_
zR5eP!9eayu_uDTWuW(xdYSzAhW{2acD;&IhtQ<1g#Nads=s1K(^Ed!qNN8DES$S{B
zvK>idzFb$9SqqqvT(ds_BK|uJ-N5(>2mp_~;w7}4*RTysDgBNci~`TrxLpHWS#4s3
z4sef24+CTHt3GlF|1GvniDgg}M1b9mb}VL}_wbfu#bO|Q(Q)4Gb5R<W>%=-sp)eiP
z-1%_mAf!y##j7|!{BZF+HgTb^q{oZ!K}tqe#1K(_tk7f8;h#Noz1^=P)ZO-FeAT%_
zT%J|F<!8utYK}0xL~LoH&KnUWISc_5%#lRfF^)`1Q52=e5e`rw1Wh=aY>=bi9sN7)
z7d+RrZ`a`=uFYbz;|7LCE-o&s=`0sb#ae?!d{<GDU~QTwur|eS&wU12TI}=^5a{O(
z;1gaBGH6i(E?Zb6S-+?h^MUeV?&w{&*KQB)s~JUR>rLU#?)NzR%k>qZK*R2Q_#a9M
zo2<USJy#1sGEnF%W9e?^;t%vf$w+KjEP7I@+D;pO4$*EvU{VYPIIK7>i94*iLY~gh
z6jp(?_Ye#$q}5xeE}X9xN6B9+*k3GxdZ}v~q@T}E55s`nI}dQFTsr{;ML5i>Ym|Hr
zfM=({@+ca@<og~7wlTCnU4P3TO3t$OI$cMX<^H~ge1-@VE3-(q|NW4-naK~0>eI00
zd#Wq@bPP-6HW7YOw`^$}u)C-LubVDVLJ&X=MHvHwVB`e20}VF1{JxfgmHNjTf<;8I
zk=5jjw2QMtISewnm(Deu$0B=gnh;|H@pp;x9e)RQaNB61Hx`+>AF`2QorQ4V=`S>J
zfNxnZLz#}3tOR8oU$C-KU2<knsi1;?l4QbRC_z83FliH}Ke#0uk;OC;RoqJurett1
z$KjaQS^hSaG=KU&GmhHkv(0D{#%4Ux5sjc?^I?kB=~U-i2u|61a7Pk;`)kWtY4zB7
zTKF#r9BBsNT!ZLtZZr)5O-NE(?@7<$NiKmAkov4+ptL9qhrVFyuPK1L{6UND5(f^m
z_cWKOk=J~IQSus_p|+4IyW$HG6$c!GOgi@>nnv^VBa8ik65jDno9z)$lVRMBw(n=-
z4xDPXYv;A9H?j%`n@^Rp8#)j-MKG{Ybx<rqJra+MbYB26XXVpK2H)D>;Xg!hG{P}!
zK9Nc<L7pd-D5@0!K*OJ__XH;?c@X6DkDQhz!1eprVQ295@plxb`7=Xd6j8KHuW1Yv
zIEoInOBcFDgj3S84S$MJm=wuVTQpUuP%=1_9vMDMa?RD4sA}R~sty_z5EcRQ2RzKi
zGqj5)1FDp0y*)ENRhC9AWrv9g3&v~zdf=46bI0=HdUgW>besyT^i@A{9~-`B*PDdT
zIU5nDqRL_t@(o}{Nec^a)ONA>?IGu+vyeA|e$F`h`|@*wyqw;qZ7llmM?duOPH8a)
zhLGC2weKdI(GHT0#E4%5q0nj$O`o34+8!1-=hmJ0Z*H7O_`=XwY}v#{!17R4d>Tm{
zNNRc^Y%}SvcNWT7$ro;*@qA*nT3Ll02HWBy9PxUgItK&;+rrC(S;<%Mz2q+3F&R60
z1tFp2m~0^>Su)cktnigF%)W2WMq`iJhqv1+UWhg_L+C_^UD7Vo01DM@l=$4q<*pSQ
zog)Zkt%aF9(Alr_x&&^U2`WkiLA=bS5<HXj>m$?X*7V|E<mA+Ddhm?h$a-CVOW%e!
zZi$rTW-n>M@NPp-Bw5h5c=;Np)djkbS5&$XsvI=NTaonzQP3HGw7JwmDm|R5!u8PK
z?QXX4X^YrF0;h&lOK&b{U5K;+;yBLtT<lyp2AqvxoTQQmDLmh+7U5ok0Cu4FD>)t_
z_^Z6-e2BG+x;)J>sDA?fs_{7Y^h+yD7dtE>_Bv^{h?EirQI`xBttv>f$<8`Jt?PZ~
zXsN)pMiWy1`XVVemXF)@;rzIxme}XPw`5WYby^qsc+T~%^Jimd%m@NxWS`Dkj3-I5
z%In%IiM(Z18HD3)Oi-Mxg4nP;4#NF%NqiGk@}Ze3B4<4?y)r>^5YyNo%yTFo6J?3s
zfrNQS0F#<6ee1XIYg%&cUI_eDJ?q=(pUp5Ch=Aiaef$vP4Ih&(`OgEv76#=*YJv{L
zX%rG*U@lf@QPsFd^N&2d<km6mV7~X{KzNK0D@eAD0Z-^TP|%<%n`kyXwRX7syPxyv
zeHD<$oph-=l~7pv;Dwp+fL+M5e0tmXLwVj1SO?pfVi~koUARnywBcSRU*IWHby~9k
zr!??HPLk$*{IOyV?RIAlemg=k68>;n^IvZzk-1^4e5rVdv<+N^HW+Dfs5#8F0FF9t
zS#5t+2k19)y%3r3`lQw<7q$d0UGN9RFL57;qnr&SP%aSEj#W_MNT4tg<!Qdr4n^1M
zq{jD)Na>^AT_z*hI7|<H{f)tt(#-n}#PIkSff2&XTnfhC;#}g`L{q56=$qu9div)x
z;}FdhyDHzmmd{b0{kc)FPxvF20Isy*R5%biov8wZ;2f~s5oNTeOzZdPbg>d{lg!#o
z%&avfcuQ)Uf`hXv2*DjX#U|vu|3P-OXq15|VRA|^)d#q4IO22BJ_Q>C3^;5pK86ia
zphkz!C>HZ#_o|ZD?bZ`YU|VwYu4kaO<^G_gtbQtO-gysYhFE1Xsz<Hb5I*<X7{&{1
zK8va==s{c)e)$oHxku~W6TkU<**7C058_Ecfk?YT@JXl!4wv(LCa)zFM<<CmviV%R
zCBy4Y1XY}z78P)mP)VYhO6Iv|tC;Z?u0ARsB2PyH!knZn6p^-SwG+Ng-L~`14@#gr
z+?$`En->Lj*hVC^m1<Z7TE!?}n>qeXD9#Hrw+nRPvc1Z#Wf9?jeT}>_Nlzz&*oPa@
zVF!5?ce8(Fbw6VZo8F$3=g;pgIh~%Ox`=BULGZqZ0yaQNA?U$*(w5j3HnBeJH{T<+
z+coWHmE#iye!+!Z;$}02-)_VSZ(#T+MSTc7pvXn=lOBtT+91wdB_b3E??p+o&dQGD
z6juF0(S$OSVb~H$(WM3ytXm5dVU80L3Y~*B8)$pZ0qAO#7F(C<AZE+mZi;)iKu~NY
z`dE+(+W3JGGLQZZIt(Xp$8*&h69UTG0AF^`=GlT?#NuN&u!E~d&jAY&>s>@!jMG<;
z_qHn9-w&3LGW16YhOgcr^5bonO=U8v#78Lz85E~)9cQ=V0QMKY_V>qfS)-CP`Y`ws
zCstyub6G+G>X;;GMzgQ$4g{LgrhP2d8E*Gw32u&dxk<%8Jn6V3EHHA6LTFhsC~<%k
z7Kh{;b&A}g>K44R;_G7d`fQz|0=S*RjHq%ySff<@396SCJ@m<6r4DgmVB@&=%ayV-
zaW3e#Rj4)IJwP&?saS@M;-^F3h`vPBodR%=IlFNAj&2QB0beebHyZ(*zC(kyOnErb
zI3-6Wio212><d@kM%^C;;4kVTd@CV0L;X!Dn>3hF9tVT8rAEtd^Sp@IH!p+`>?<xR
zP;-MfZLr$(+YpA4vU}iM{Buk~0|Uo!|Nh(mY)|}#C{7hy7tD91<2&=!yir6=;e*OH
zf5uLPwN@PKd(8Z?LaNZ7Yo9by^bCTG=o8WRHo{B%6a60;3b$0Cp+IklOVm00;xiU7
zGt9fbou_9Bh)QCHMO&-Xvo{7_>{Qe?lPyN};PEDt)~8PLyVMuc7r}aqsJ;6cJ(l0#
zi=ha!m8nE!qV(JQry^(f#OOc&toImRe@Nt=9SnS>fL7(Sw+|HDF;Bkn0L`oM?eg|@
zA8<*5=ElmVcGAUB{1t{W2aMG<;jBlcn>T}LN}Gyrl0jLqo7k+za+c}Tf80$1O(uxv
z(W#+2MRYu;-Y%M+qxqOnCavD|t}&bJe;#fG`9Ceq#b#+9z}RA|abmIGn))k=C3DNL
z{qwEpA$zsJG4Wi>&CnoGK(qF+rh3p3l#oP*z-3aZmuOUGa`uCKk*~~298-VGMJpgJ
zO<@t?oh1vLj3O*DV~^ZNH8=8coFI(vZjKjg{&N<4JsOZiQo#LDa(+F9P(x+;{S=ps
zP5#Wl)bKKV!<|*!!c|xcPsgKyY`SypK6No$1ioaWD)eQmZin$=__oE$z*OyI!u9We
z;_I77=>!LTV^O)jQJ|6w=lc<AoA|W=KqSYzU`ZS&bgLvb_QdjtmeOCh`I=(*u0lFy
z(agcmJSnK52ScC(<}6K)S^5`pp0-ebHn8IS>xaOQf{q1!8d+1WkU)iCc}bA#<vQ7&
zO2~FpN+jckX=^OYCu3I)+;sKj1v!ltRI_7moJ_Fv#p)c8moK88&@4wK<h(YnKLY)Z
z`{~|G3UD7IDa2cg0JVhqcb%65=uA4T1VG?St8fL8xwMu1-zT#!!VjN~^-)s=Ivf?6
zIyNxcrZ+n@fYu31%Ih9Ap*H&Mu^D_VZcZrnNXXVplxv$R%2XHQp+nzrTh9#ldjzO}
zu5T{?A5&i$Rpr)&tB9~c>6UH~Bn70qJCsK0+#uZ`xasZ`1nKS$3F(lK?nW8}r0?42
zJLlf}*WnnzTWiiI@d*g%iMeemni&hyRCOPOj``G<cR(6y{YLDIDtJFEC<?l|{v#~a
z&h)>(7e<e;a0ZU9)WTSF0nK;i22j@cSi+^X{f@S4c#Q?%0sOsn1mE*P7(^A22^UKs
zxK4=2zu$5hW_F-|BXN+^r~WW8lb9%S;Bi`Yz~MH@d?TpAzTCtrTeoS{+6`c~+3X)~
zi2P1F(OXr(Rc;{|un$niGq0P!hv7!UAdfvB6qU^@y5Yv0NJR`SW8>gFWjq8mZ-d|}
z8M{)8E0Nc*>eiJAF2hGy-vS@a|9x{HoQMS>GllB(e#fN4Qc+~D;7xDPI?#EsyU=(x
zhTi}20W7Cf(7BozxrCP}F{}nRz|E2as(P?VNuim-;B`_s8+sw&R-pjqaL?efZE6H?
zJY#tY)>1_;L%LJFByg2JtJ06@GjSm+3%6CZ@K;3OMle%JDm3hlzw*r8K4z>kbQM$S
zUF&hGp`EnJ?3dTG(@;9`*_yO%KH$~{$!YLJ;c<{OG_Nlq0b2mh5f13w>np&iYyy2`
z?I3WXQLgMBwq6>*ttd1!G=1<*io6A|7J}JO8=DdVAZ|MaZIyCB>5*g7A3Oe4^vNnM
zg7OA1s?cZ}xaXF}odd9p+`h&3pRrK{t4|1l>C!flb$Te;p2;`qieJ;b{hI<=|HgZY
z^^7As0aq$&jQn&rddj<b(+f{fHNOeIFO`VDmWlhSL;Dr7KM=};?GV2EwUtpuN`{h5
z$v|GedoQk!f{Nhx5bEz>B^UyhDi{I+R^ZoPFEuAIYgZQ0A%2!Bg)lF7ae78$ffOyH
z-=d8c8%|f&(4D|!5!>n1=m|F*mNXpQHm^pwxGkC88~BWDGzBI>iJ@`fyVBHG9O2R!
z4m`j*uu6HzMV0|V&@0m@VgXql3rvb#^EAWPAYFI^gq<eB`*&80;A^orrnpkSIJ|`u
z`Y><Tp37_ND;FzpI*n*~cU9~#b$Qf3(C|g{@=ZQbw2vzMp`i%JY;c@n*cCisboW|%
z*tlwMk7mkip-NKlV3{ueBr66X<NXb2j5pc70-@2Y8MH*3&CO4OFvwX1JujTzADjXx
zGKv4NfqIzxfmNb79e{Ugz&Q7y4<|e=e-$Oz-GKWFognoh9va9H^xQY0H~vN9&dSm8
z9%G5v@VH!Yv;cf&v_-y3&1!{&JU@IzIUYTsT)s;|x|eXdNsCU*qq+1K8XzHw=*q19
zY2anLZHWB90Hx*J>9HONbX>gXKqjA|%}I$?br7R)!sIp6iQ+$eBdds({=|JPgnj_D
ziQ2qJg(#w#fETS|RR|tft>@5BTh|X{vNrE=K{2OsJI?iIF6i7O%c3yB?X90zwMPgp
zqlT6XEp6YAb}yRhx>DC>=feMe4WrQ!-j{!2#Ni&UzEqxuS;YHZ*hCYXg*&A=w9qie
zeH;NTp7W!hF@$#Z;`$}Vv5qHK$EY200gfnC$?;yiJpov^G~2GWh~Kw0hiUjd+Di1k
zm^J54{c+w4+X0xM4cW%wVskEdKOI0bVbfyMa}I>88zg)aia$=F-zy^@VU)ehHOiGx
zfNcl_bDUXV8(k|L;h1~~nu|4Q3C@}GW-~5bu<o`dRA+dx9;&RA;7qsidN!Q5_yWA@
zItYE?N1$Bj1IG4a0pK+f9gEjC5D3|@23y}Cw&VrBysT_xF)Z+KxqUhCH)3jKt|~|a
zv{8bvLjDZ%raTS&6#K0I(&z44+reXiF~32*&(%7yuWX}+vK>Z+k#<p99#Qu1nSm&^
zGS;|0#JS&>Pk1?&GJlWarNUhY+IJ7&RQFuScYvo1g4ch0FTm>zbQ<-8bWp1OVh|hq
zoaD{VJlwx8o@y0>5}_K~MIAkj{{7i%s0V^2YNKIz(rqsQ5^kEP%}+sKTbxd<l`QZD
z!Cs^G*}apP_4=|a7m20fpKJ`NeIohw1EorvTTT2u&pxM?CQ^UX?~V11N&`ixl95D=
z82r@VeHbAR3?_vf2cu2(aR@0Fq43h$#j>0I0Q>`0aIY5tW2zjtm2MsdLBv)lpbU|x
zA0+f6qqE?}AW1``T}$o=`jBXxOnAE+$a|(5HU|`+7}+{VVKiN6u%F`ZAKC|J4XYc0
z``>IgX@=tIiyRsxug2N9^c`s4-oJy64`iR`a`dincbN$XBd718`%U||s!L~}6cGcZ
zZFcBXstj5w9HtAlyt_GZ-I7s%xskLeH#^*)_cM|3a=5l+FyzLH`B^dYhK(G%)LD#m
zde_1dp@%QL_?>7OAUTRQC<Bb=G~kC}{)WVJ(XttZe;VFe45T69eVj_)C4IYDBZZf%
zT^l=YzVTch_2j2Mn^BRM+gn}rL~6c*jNuZ!jl>3tJ@)`T^C_9WAY7yPtm8SbARXLU
zncHF|b+kZJEH{DrY4Jb^lI>cTQ1R<@b8udT)Pd~kjWPFZN_*_f`9Vp7+N;2O1gQk0
z2hXp~>;Jt_$K(h<sEld8A>+poZyKX7PdxCy^t9fH9R)B>9`)bL?~X{+Ex13dY(O+H
z_x_|${|dR7I;D9I#aI6?#Himur4PJ#drkwD?qw@dWx#8`bK4#Tmh5IEJ6-)zPsGX3
z$(MwCM7Vwx5|0ALew!K%&wTF2e2^?O>L<LqxK>h|fNZPyf3BEVDpyWIk*|X}85pG(
zCIs`aCv!V)DHNOh9+@rE=ZD96A363AnM2gpW-D!f?a9Yg;cb4;dJwY!H*7$jbfjO&
z)gv670}KegxCokD0-P)7rxJo?ZkwNOs;a?L^gbS-h&Ob|tyeps7!rT7l-zCQ@jO@K
zQF;4)6B?)BT%(M0!$}V@oEcpcwoF;7lxTow8`q7nj4{FTuu(pP0Ax)uQUfu0OKAgv
zn=h(GA+f|c!XFB}<34L;OQ1|kB#BChpcr9xOsO*>tOh<3W9N!+Lv#1<;OIihZRr0@
z;`)|<cl@h=X_So8hJGe%G!IreBX(;2D@-a4m-9$Bd8^NU1m}4RB_C}wF-BKC^Zz7@
zqUp-^%GKsrpL?9Pmn98^Y>tmyZ(X?83jZ^2q9PznlP&985tK9ZBR}8}8F#+orqVI6
z^)xNhr@{iE2LlAZ3e}};K{1&h&00*CItnAr`6u(I?pyJi4r{C%O^(lzM6KuRpsDth
zKH|yS&sbZc%{)EcGZ_2v@hr82>oljLCdX(9CtKmUY(TKW%8;Dht!SH54O-S`Cy{YA
zL%jUHdN|_h^Xnra`-#yZGA4WA^0pO$6;2KShWpGCk^B?r@RM!=f_W%AqR)ore?1ca
z_O$HkgBz}L!8<vF-}6)jIb6llL;AI1{T7T^rK-2uKa}SG83v=@LntD}3ZIsrq&{!t
z1uXEZ-P6Otq3az}9FzTlWre<7%3Y1dkOkj0$-E^y3ni!P_ZaS{KmwqWDuegHeJ$96
zHPc}K?F-^@UU_mzfY|i~_ELLsqyZM>1p*hCni5-Ou`~ZH+gKQ6l0+?ha;eR6F6H1b
z8M2W_eOMx-Cuk8;6yk6Xk^ClhhA*|W<?{Kho>!<wwPBNcMEnID5`e$cp%6WT2d7mW
zrNB=BZWxBKjlr|*iTDc93@Axl)(QnyVK_7#ryup#UnB1%RY*|0UM}mzmLqi%m3)*S
zEB3lUJC^N$3pGB@;^2$>$>HdzvmB)gLPtmpyujfdU+Or@Y%Rd0Op{Pf78-^7hdyCz
zq+Ch`i^mn^ZhU5%ABb4r>a!JZ0%%OmDDlK*|FhKo5UzpbklCNlVKUrg)4%4|bM;Ob
zn2h2sEq00iU5|17(kVYN%z@@yD|sc%z~5XBkh=h5if~xNOVK}6cuRXPj9x6YXGRJ^
zAT5B8cDOsAeiopMB7E|q+H9X#$g9ECF{6~TSlXs-x{vrc$uG~0tbc_j2uJ9XeOLQ&
z=d;b$%cV22jMDQ~sc6%iwyx9J+20gt$hj|aCn%3bV>jn7bp1n5cX76P65@z8YF6`S
z_qj(gKDXm@uI6o`sXjLoGS7IQobf-D1OX;u087TFaQmZ%N`bH&kQ^wD>Gb$k^iOOj
z@?*2^0!d^PV<{*oz_3KZ5bZjiiU7HF@FWG@^${+r4&oFG>@Iy<aWAiyQvP0{>u&jw
zf8vL6<nVIL3*561r6~1;+>lj^{>;I-UtC4$#5sA3*&b&KYQ;Vor50P-@L&ZWn35D>
zWx6X;1`QaEcwj2bvrP}h0UVvd(-o{5Xy(>l4n-x4M64l<+Qlh^MbkO!YX2b-xPJa2
zp8}Kr+2CF#VSOTbSESKD&2&11o8}y79(4!Eb)PG8VDHAvALX{fRlOw<Zw$+>gLxW!
z?0Cwp@VSjLD_S_13dv6#I<!pjt?Uh7@A|LXtu(vw*LxtZ{U38Y-(Jb<`Ky1rJpiz1
zSg&64#olCE1*cTIuHe<dCwkW8i$4MMeFVPH=-xJ66G2DvAuZ}>Jo_=ONHwv4<5y63
zP0pKm4rOuWf;Cu7f14Ek|63ra?JhuO@AknYRkENGVPO1WdU9)+WZ+R~_jQ9d1I(O>
zBx>!*Rgl$hQ<!9ueUz|Qyu7m>2P1-)dLa?HZD0GgGksZr2(@BK-0GeJEkaa;vkGik
zkl#i!ayPn_Mn9Mtoih}V?m_$DEc<g}ZUw+}Cg&l=)a#o~qa?v-dR}zmD1<^Nf%T#(
ze1CZWJ;8qHeY<WGMkW>i1ugO`VUFrI?(?hL3Ty`5TzMo&^8X?R{{w==P#)6;XZ^9>
z(GtUd+=pN5PvHU?Z|E$y{SK&49PGvu1njAPY(wIlO3EbgTh-*8G^|HflS$MWP9ln3
zJ+q!Hh(RNwoxQPG&78#QY;rlUN=muw{jvu}|GFTN*OgyFP&jPnSRHN~V#YVtc{aOD
zpV9W?N{*JSqA_8S>Ya1oD(Ox!QwuLlY7gL357kH01_>Hm>ElBea84V2;`e7vT0)w`
zcOp~&Ilca!NV5DtVRx%s(K?Y(Mu-?0Wy2FE>W-Os{qI*ZA)Pie6&=`}U~JP?{s4f%
zSC3))!zrB7GfJyks57yi!MW0iq<#t*xDxhFu173x&)CHjGu_ruJR|=^8a6ZW$?NXZ
zs9Ok9a!DfHUPmY_w;0GtNy?w}F}PcHFuU6t^cX12`)y~U(q`vmtMa1d(Z@LcAUxVk
zns$DDjwL@g)-8n>1DrF77!O*;48|73>+__UHmxDZpGoqF=lXVb{{6(9KJkOECW{kb
z>a2f<Ou2}Zve0pZ-SjMloxWxp8aDOh=&hA|5coq|y54!RP(UzWgl0X`skhbF{f%Hq
zWDbjtB<2~}B95I7%E$Wgv-zk^m!)~1qr)1)*c}#3j~zpl6bBJ?G8tQBzB(+I`#thw
z+?^Hd+&0TwV3qCzA0VcO+p{RUc|ha_L~%}0)<-L!;+F&QbD^!Afeq(|tK+TD^biXr
z%W|oHhMBHe+t*3$*4L}m9l2)F#j8nm@$$+=lnh_+*{A=`l{0Nr+7Se_&|~SH(3P<x
zaK|63{9mlm-|u^hd4}bLV#~G%SSMQFPn3L%VgsU{*Yj1Urrp6<7o@OD#-#|Brk!jj
z40<tC$~H>zz?uG!u&Z=YOn8*^<=Z6pPoF|3Hy#(vnRbY+6cv}nf<a#vOr85^h3;ot
zZB&&T#M$VmC-W`t)p&WUzgrQwL4rsHr+z^?sKn^^pD5O3ovnkWfwG1XAM-`VNGc6+
zhCR3$5ev)mfgo2A=s-$ii;C@Jkk9v~>l>pm)y=~QSY3);`ol?YcnJ{ks=}c1GV<sL
zQfH%Wz9E%28MCGuh-+MxRrbpMIb@JLVO0~0BoDPFi}>*5nojJlld$>$rS&Y&5>3vI
zYQo<=T5)O^GEdh-(B6INSLv1L&4-_lB5$aV2Lpd_Sgs4lda=D#D$<hnm}DFM7Q7(e
z{~?>mPhg248F6dztJ}#MJ_lXK>>+tVcQ?<`E6A^`FW+6qL4hub1r?^CWU0A1&<>^R
zp1<OO&Y93R-R;p{zqoDo|G4Ztt#6LE>~7bYUa{P+==Cg^h7UzT+OVKH?nRet+0z4L
zA%z0RVTJF-n=A;H%R-Lzb_a@B444oV%qhHpfOPj#o7t+Ra?9W>S7|Wlw7iE3W!Gn9
zxsp<a+Ef2w<r_qcN&7~fSf*~fe{XCLo2a#iVVk}!r?UP$Zd%eiT<bX4lshn&#_DbU
z3p1inb*rM2u@TIC-PYGNWt78339(tm4;)${x3w<ydaoUt!ke4EycIy3(yDSKyDdBE
zhs@B(oO7^aTgDr~;cobO*NNe$A#UIiK~9;rcKMg?)@QLs!$6D8fX=CXHd?}A+tC)D
z<*@+Xs>I_vFrp?-u0pBN0vt3!;!TF~qo``tW+JtjI;gHLJ<+CPunuTI+60qC{S)9I
zk7wf@aiQz$4JR_L|JIG3Vj?WH%ltfc;06Ut36JSr&HU{{Bk;XLH?Ze<g?q)+p3H(U
z-ia3Ftbq}B2~B&FBhj1#k=NSW`XMqe=^qgmC@&^*fjmHJYd1@4NeT-HriUM2(dByP
zO)~LkMq4s9?@0co4XS&3xoUo1RWi%P*=@$*L-(STj}vN@F`~orCMCU^uPjfo`Dbn=
zypnHX)i!UjCQR%JgQrP)uA@tq6DZ~{X-V&k9$x;74dNl#v1BmoVy}5^{6rJFT6x@f
zeOY;E9SPi~Esn6j?3zgFa<J{cBgg`u$GUqCe|N$S+dn&F4!}&JV`l)0o1StYop^v+
z6K~@VPnq{M($K|(ZvS9+S<)r`>~Fzx9$@b`0m6eep<7Lgs2ciM6jh<VMWBPaO`AQ6
zR?xf%Lz56O)T8goxdUjP5q0*P2LKCF*IG~4-TcX15Ctp1;YvX(1A<}LqiVZ)E<2{9
zj5+Z#>v6e2E^BJ_5n_trfu|WyhkgfSv*~0tV3ng7)i7?qG!|1-eFIf54qy49!23}J
zDJtWYvEp!dwjri1>WF<vVciG~=a|*Mr^D*!@5J(S$Rx}$RADZ5&l^~qi9=AWrpvyZ
z=e;e{(|+;Vf&%zTT4@b8jzv+~olq$>Yks6(qK$8(v9<zMr&Kc%3Jp&t&FB|FB5DN{
z@g_Szv$&s{1GPjQ5ta&Y+gEgt7e_oAQ{ajlgN%y&X1KxY`>B!4VH*0&;V%F(HqsqO
zDt{jXZ%Z9-cY<R!p>jNqB|k|0tlj1DyPkk*OM|O0#Xdoe>xU2`vWNQCM5N2iLH3SL
zjq>87kBF_lMMyHFP;Cwtu<<S^WvBOKY?{@Y)_b1t<V73?5vjJ8!IuWT!#g9<pRK7W
zgr``F-=8@&Pr=f~@c(;9#K`^b7(-$`81DG$3xm*8Zyz2cHXIxzVXoKPD_B;IAlzsE
z6;}$S@eH1(9;IFTU7h~uzA$EmNTdG_Nwko-$KX=}y$`af#dwO|=CD5$F6lVp&UZBv
zN@rgZqo;rZ7g@tG=veIUF<%>)e!2hPWeLI_u{&V2PVeI`ewoQUqG^hS0$ARRm`_jz
zyt7sQA|;H{s@2ThoXM|GKDzBH>J#^!5qz;Qlb6IC6K%;G-J779b;IlGAiPA!v2F5@
zZh32o7R0_shzEb^-@L<Oz_&czL(OFk$A9~#Q=vPt56|rwVI-AnaGC&ywH6>(zzI7$
zY}#Y#=-Q^WoE;<l=pCybY%vUaEG)%!TCaQy;<1%bJGyX!Rj&t$xkLNNjg?1VY7|hw
z1B?_n1`1&n<Lx+|9o1I=y}d^N5!YY)=7e||UB90jfXvlsU&#Ie0wupzV3ilXWtj>t
z&v{Vkv&u}dR3-|k>rF@)`>a%j7*lPxoXOmZCaYJ^B-9tr#Yv665QzN1r;ct6X8c@D
zg!=K)f1lB-i0|I7JaYJ9W~N1X&8+<Vm|5Z)NLh)Xh-KIV<K#ClIL~i3fJO!1ire$o
z$QakBofz{S+LXhF@I;UXRW!`6S7j1H5if4=4W|^vX(5cYjp4o5Nx(T0w$)s|GWnjV
z6HSbl>;->US1|o4xD1uKoq8e9FXCKfp;mL{I<;~hmxr(5Q>rrB(s|b4KJ`~b_M4gA
z81VZ=qaiXZt1L34tGRR@fB8zZU1TVbOis)-!M7PsKVdUt-N}vI!l~}_zD|+qxjPy`
zn7l$vZ6E_(UJiLDF3WWGx(0l<|4KpiQS?jVc&?ndtUxpsvVsgFPJ%45TCQ!#>8VHl
zpu<6I${T@`Zk(rFYTXAEA#O)4TMl>Wa)t9je3n7zq&D7Wnr=QT-~DXs^|Spwfb(en
z94Oee)3wNGnfjhc%>DuHxe7#_*u;!gEE?u>aO3LaiU56qX9w-#9C;&Nq*;E*pRPs<
z&-^K!hj)3ML_TBPf!CM`-~EYuPRRB>uz@f4?R(^=SZYMn6yXwzw(L!dl8X=bgelG+
z@Ff)Crw#FA4L(ZC4(@@EdS@N)PcN?px%Y7&r|C9vML*Mhi}BKOc-PtShJ}cj)-+6#
zEGs@~qV3%6pMxrr3y}e1)8TTV)-hA^DQ;~1m*}B@X6}KYL314kw|5bA<F#{!Rh!qN
zyRn!fU=$PwW%KwIOcI6^fnY9aDSbs)5o0T0h9$xloUX*m%VV%8rh|Nvd>8lgnI_^X
zx!<YOBChgUVnfc?KzB9PlV9>Wea-6c`uRac)7nRB15W5BI?+;n0OLDymd22P=N(Gy
zrUGg?4ux7F?9*9!|8D1ekk{<1IZ^ey3Fc4yzl$L}A)=+g^fKhNTWS$ebV{Sb#C+_X
z27!dxY9K~J5oJv1acSr;)}w^;_`mHQ)N5(LG<eRjIs4w$534eUaTeLwd%DP*d|GQ#
z9UF_q%%mtt{*m2j#kbI4*m}$EU6gW;#*ELl5#MyGmx)8Ox~wvs!sIHq%wtZt`STcc
zkp|u5TVYy*rE)fk@o9GPfB555qQ`RYZb2~FV^*v&e1|_2(e&vb#lCYlsbWE3U_KTp
z*x{C3A(Z~)^;GO#mC=w<Y?_F_Kwvl5c@l@Q+p)@<h)_27kjLm0CHd1t`Eq9-zlQRg
z1iUUw>k{LvUy$NZUoLVhq)rx5pcN;^46+FHzRYO7zMXK89p$90vj~^966W~e|8rdF
zoT)EgoL{Ctxyvwh<HFh3G_+)<P8lMJM)3JVTwH@$ug`2It^p#C9@~`tKL;|V6-1?l
zEGAVX@X_-|-{7VKMg(p63Ha{kcwMMSGSwsZG+xKNmdhniu`g7(y;@aY5E>?ls0tT<
z@fl6U7&Q?(%+h>~+g=HsbRhX*S6E&H3Z;nKl?)ev?P;DZ5zy}1ya!;UAG%l_5hp{e
z4NouCZ=M1K(L$rc_o0vPIBT)*!e-7s-K@pgb9fKXq`Y=)({n$FIK2p7t#TeTn~oT|
zb4(2PIO8H>a|`z3G%Qm1yX#4GGVUmL^4?}|tMqiWpx&gJsJI==R;N{=#LLrIy3lF_
zL)3JB^JpZ#aIl{ISM~1ldj_xNCLYx(;;8Jal*#`P6!?d`4;%jVL`>p-huDepc(QY@
z4r={u3MfxiRjB4G3_rtr%f|B*iQPe;T`pky>VB49qEdPTXe|+_<z~scDluCCoi6=?
z+s4AQbT-fpcO!1Yr*9i{OTC1({=<~;TT$6(fPBCOIUDsTlum8XA>RmMX>2O>>09+1
zO~N%8untaW@MC!l+zqTI%c!Xhd<>?_v}w=Y*~Y8o4*m^na$7}c7W&xdJl^Jr%m`xc
zZ@E|7qtncsU-mE7FMw>xGU{hr_nZ79&0@0Mm6~-4oWz<Ou*+!1AF5i1*EW7FZ|=!M
zIQCY}+@w$F4kMLO>a+IT{yDb(q4+4su^{dB8D5C;%B2t(HG;jp{b~EdowCQpzNKKG
zQs9eGJ%EZa%bNItWyS<_F1E>=2%7*G1$hm}M3TNs$~j-QOF(Z-XR{j;WeSLIjMlYn
zlpIK6wHrNy%cg31I&~O6%9mzamo@g(1z~dJb$3g+ulONv0JNq8G3JP7-1x^g>Tn+R
zi)GAF@v1@GN&R|{gc7dIWxDXBE$3oc2enwclKk|g;WCW24TQ!TP3752?=+f}`}7;T
zHS#l)skL8X_KQ)yksKL6x|UQuF<nKeWQSw1NifpqCCYZyG%2*UioN(eTNh5-8_SqH
zWg2cjKe0FWs<L_A7*ZJcr|AC&OFq5^V*wc&<zhJm#)|-p6}x~#2IX+3uulH;nocR5
z0sP(#=a0lLZqYiCLb`BvLU+WfC?ccf!|(odoOtOa7wuUC%G2c>K4lG>XL`Q_!t8E%
zdf*61lc6^GVM%yrXJDjY0>k^lXeD7qr*NgX<l&nxIJz+FJJ%wHXaTP+%=CH)!(~++
zhXs*EC(mf{mKk{K?nSq~;Ig!mwZY=Nc$LN@yXq_GUwDH)FhZ1<w<0ZDbuf)98^Ui9
zV%Y9dyRrLBxuO>9gl0smR^6jrNdJN+2S%z2yP=CoYw&LGPC5BcDS&H(=yJG_Quxfo
zDA5%$ynlI!EU9qh?gcG9{rq;@vf14WAY~AJK1MDTlndFwN1!Pe$E0)srNj84^G4B)
zpo&lT_&E!Z5^c9SZqp*4gwSVHyzl4C34V3=HSk@T9&ZYtbJ5~7$adoevf~3G?qkej
zorAgBgK5q8-oIFQ>(uj=Tu9Jfyv97m<g6N?{kGwSll{W3@Anib3Ww*E!hB08W5G@L
zXV*b#<E6NB>*+qU&#XzsWyEyKt!#SOIM_L+^A~Nl!dENa+D&dTImwk@erDI^0HpE>
zN|4Df&p<*Yr3BddR>vErw~6hEH2ux!b0po7=L3WJ(@qsMSh14BJsIUn3AJ=KoFt_c
zSPQe2!n5I-3lkq(5J;>4hw6ME!~77NxNQ?lKS0-NY~a3$m&Va~51gm7>*LmGDF-{)
zuPw(VFh?xsY6^!^ID1`ZfS3lV{??1wi8crDX@uj#QRwK1a3tl^4}IUebImK@?{YxI
zUMhui_7mQi=&~rmEgS+KkXDjmpqU`82suNrovr#2_;H+}hrvteY9mfe$Pfe=?gt&X
zVUF_cF%towF|8}6>?t<801PsW$sH>C9Lx>sd!%|zPK7Oh#^n<=s=R>|y6zF{K_sAc
zuP)~8OyQdBRS7rCl9MfapO1Nh)d{z@1vykspQU4ih>%p#Oh$#z;wM4C*6(=Mm#o4!
zu;(0&f%LEKn5l!>8NMx(P@+ZJG0@i@G4L30iOjt&W1`QTdV7u0B!?ZRkX%^VY}aNm
zvNtvRvQgh>Q$<!V`lTm+&Ok?3vv)o-iDmFgyawZ+qc4?CHaT97;`EkUjSY<C_I<Fg
zp;+Z9w3(x`9~^MAB38|jhn!2eX<U;~1piZn^?4$4|LE;6YE3!&y?JMGdiWgPFWJ3Q
z(D?VKI0gz%pTPE)MpxWR{iLig5y~^;fM0qXMuZ^vGX{kG2s)|$irIubzNblXr}}dY
zvN@}BvcbM1J$VH;8J8YCDfDwQ>BUs<s7{=HKm)7TE07M6(0-4~>=gTj-7RXB@3==4
z_X$zXU-*u3kWG5<1&c+j^c(nJvx=CtfQg-6>xy{c;0cm#XQNg6?48#GK1Y)Qy1WjJ
zG8gdCRqmfLWhNC+p8og@5fz68LWj_?LlLs90mrX=5+7HlT`g6h4PpP`_TfXi9l!+Z
z1amNrdtzsmdPJ82PK4=uI`{^Ffwp+E52>7*v8a-==X~cSzjw*RJ*cUuaxK_<aA|!<
z(rq`|24%kWD3+|f<Sgh9q)B8^e|K-dHX1#UIOW-Q)VkFrA-mwVall$+Ls`f<v%x@i
zePkEPK@#Ott<siJ-ygrmVAG|-%3GMX;O61I2dyE%F?;y99TmcCgy@%DLZO_D_CBci
zA7H4A`ukawdCat%0kcj`pAfw}nduKYJ3ZQ{;j6dYhf&p|Gg@c+#8V~T^mNMI-!<`C
z=}z`xg|1%>Z7{9bw_v>CI>iwB`@Kz5qew%xto`1jP^Y%&;0)k!m7}I~-tI|j+C%5D
zYW`KcM}{M2L&=8AXnIYd^BH5Y<C^iq*|&IcLkjqgWk9^4^;m!|Yl3MkMzy_(^%8TX
zHT>Af$J2aEzvgF5K*xfC{!4_$;73_cVW27puFtpQYiU_0mDq=+{ra;@*E_h7tSoRM
z!M`qHfsnx*?(h`XKOh=DdoQUlw2F{NzYIppV7mV~%6}x_*O%)`ogdA&#5}-LaTxxy
z9|T-OXbJj~s1*>_^R57vGylRjnoPjxcC`CS1!+_27Cd_>JWKembhM7NfDP)Pb_HrI
z_9Q?l7;~B*D0ld!lDdNGF=CQIqTT!%uaRlH+x=>u;u{M9Z9`{}F#O>mp6JvoYc>D^
zNhA^5f%kK|A2x0?i%KvD?jbOhz;!ige)VoF7}7&Ad{act>V(cWv*@aaKeL_`TRt}(
z7YUSmNnXA><|MWJVbAw2d^V!lG^`8b7g{q8&94>J)8&(KX*s#7^qK60gOP;wmDN7v
z{V2PI$-v|}B&5J>t}vOkz{xHIE|>LH-|PYWx^2$I8)n|~uM3_^#)WU&lEZJIxwIOY
zw~Ypg&{?L@b4wO1!yUy9WlF~LPbKpy6bW)aWGWd-<2G+PoF&1&g6l^@bv$}_;IJ2d
zRQgw&ozri3f3~q*Ue6)>IdowJIxrUN3eo5UK1EaFA6++5(?2mq^t~nvRNo1<0I@>i
zZKcAf+g7%w2yQeJ!Bh4P(p`(5vm}hTzLq05>Q*-lzptnFJ#N5IVn_cqYAXK_LkQuw
z=yRd1-VLJ0*%fC8*mJQR0p9{Ejx>#YB~tl;<6!nDNV1~0h`;0olrY~sH$tS$A-%G8
z`*Iv9%0Nl)R*@`A47eTW(F4W?Bs)+vodP48HJHE<%a)ke5tD5|;LT?zwYJCpC|Os(
zi_1$~pCLSDzuWm1#oTyoz?6f`O0p!-$$L`9?jt81y^cKQ01G@*`2b`;?XLNMUJ+6A
z==9O&pp*1nIyU^o9Hk>`1-c{sWA1D@xdMj2_5$5gL_Bu!@?F2`x>Gi;H}(l&%8R}-
z@7hurf&}zQWCUn>(L!R9QT@zA(DGwVbuS0LUz%yhLV}o$J5_|q#+N+syYS?DA_y_g
zDgFn&0Hk82F+-QE`^?AdJvaUSGl=O)Fhk16!-U6|rq0xHKNZ-PfmZZFg>Hb-*9|kp
zq#Ojpcq>yqMf|+Oo)Ymv&x1^2E=wuWJT~^7!U&FKcXQ|i4GXkYiilYtfN(YM?&h+7
z&keJ@VJ-Bv7?;8ebygy2Kb?Du=nQ1`V_lW|aMe+r?^r8=h_ZgTSUP=*gq7H~BA7A+
zBaa&WKBzVy^$^?D6nqq4)B1hB%H@<HxviZ&3i5jA*pMI@JX?$8@A_6H3G_{F9J8=*
zF%WpW`FY(>P2}g^@VrbTG^TcCX(gs!|7e^gxSAYJFiOEQ36yWz4l`}uo_p>lQ~Pj-
z%lwS6<#^7wqM$M148^`x;HQJO(<FOFU_OR0@c4W@_dssM;-veHVH8Q@86`Y-t#tMs
z1rG=rXcNQPWk1i>1RDk9Tzt1+Tr3Kfc*{UT`+^{*DY`3kcNw1&hW0WceiUfRQ?j=f
z%(_^<J@*(nX5{c~MRDb#VbY7JA}fv<XOt`b4+or;rs&^jWDUDvZR%~$CzU<HPKxCp
zRaaAN$@;n^3l;dJ1&z{-Y3fdqja$8YDi*|UsFZBsaU@bV*8XrT)YFWF^B0^v^>|&5
zH+R(G$(ZEAJ}vX5gP39>;tJwBy$n0z_U_+c?;OrPxP+y!3c%ckj%`I3(vzNzW^lTe
zJX76pE00N&`}>kIYBwNpe2G-rf}y7bZ#66OBhc(m1R8PMF3dzeIileVWx?%QbR#7)
z*}bkIy}!W-*kQl!z-KkENy1buORCVg7F&7xS)iFsshRSN@9}qsh-hKkPLNdsO-&r4
z5c^7mQ;W>nu!2CmCQ8kd2-YZV*23nk*Q7aTn?YQ~n{6qaH2wE<K^r59rMbSJrFLu3
zm$n>!d>g6_?WRpR#hHBr)kx`7+kK65-Ia;6fR1cECtOPHAz6}_O_~2*e_6JSlwWtN
zY?+#I>pPS5gQYct!OzY~i`OPhBoXq`yRXOMe=xikxL)RqUv;D9Gy4boIK}XT;6USJ
zP7X1(8C*NB`&Z8r^FJQZr)Q!#hF>CNr?8D%AET2EyB$$IWV$k0kt2Id_ZMI+O#=gr
z9KZW0wlRr;@q&b1VE`Ed(ZkcAB{ydY!|?B*@AzQ&u`)TCm{~SSM+c|NR*Ja}#}>=j
zudUT)>Igqum7!>?cYH?O5F#F1z0_#==_Fl;^$ohKlV0+9Hsv0FEPp@tv;g`TV&Z{R
zSlrq!%mz!o9pp24#yRnNxDlVQTrdA&Aonvr29lVXyEWeKV-hOdF0`2JQ#TcMF?Ib}
z2>cev{0$@#vr1v7+5bF_7p)}2u$t!uG2G0HY^=h8{h)dzz978OTxTxD-c{YM{$Ccr
zmx<Y7>&u>gea<}D9QPGGo_1aPIQf&#1>V!y&`lyz{2K4gh(5VO5;tsb7+YBS;U`9c
zhpkUj4hCmk?j%35sW*1Xllt!7-$7-qe?IEl#1XAQS=6_7z@ZHN?}67A{~8?AX{%W&
z)r9;u@)b*^)Q49*;psvI26Dcj!8os6cZt(&dhPD1aV$~Q_HXQ)7eJ_WSpf)vanO0K
z5cg(D(5?OBk&PRrehz-Vh{&j-(wVK8+wOO#aX*CTvOG_*9-mVnVeK2=X+UBdjuAU$
zPKFFyJ<el+Sb97#>Xa0N!hl7F1cv(A#&BUlZ-io=A}E^6;l&9RUlf^@?k>?lE==wt
zNrr@kBLMI=fu7DsNUi=u2SRZk>TWkqkbQMPg9F*xDVE>r=SKD_B_Uc~f5OHfOcmRB
zHyO%aLPJWo7kj^o&3g~)JVPp>e~*Oqnx8%MQ{bl_jaE^gRHR&8*@D=NW(dOdu=t#8
zMtX$nD|wkj&5k`zAy=BtleJOY)v*t{l?p5+x}(s6Y1)i#i*J36Tr8A|1$G7)FZ+Y1
z7ktc~SZd~%ESz6?rS<X{p1jkx>b72LU!!4Cik>#^Yre<m&l3!h+I=ln6wgsq_TRNn
zky<&48;K_q4(Es~wiFDw3I9503?1on5|*}h>q6t1L2+0zJ|;X}aP5J;IYN`L#i~W*
z6~|k0|3hI3{9!*41>&Y9c(MdiiPjgNK%Xp`yVV;pvyjL1TaSOhd;}B87Vl{O>Zv)P
z@GwW`1}TtHJgyMR8%4|lD#K<Q0Y1dO04oQ_Cg6KK2*=OhvzvLcn(Vb5*xOsED%Ras
zy`I;j*t4jo@;so1*FQidLVej`Wt}MJg4}>-Bx2<mkAkZ~kH~^RH!jB@T3jU#s@$@z
z*-(JhPyDqXTA2QZ;WgV^VbVCx?Bz~u#?->GAC-@0j_(_1yz1gIWdIwsLq0}w?b`FV
zfs)g@H(Bou$_9S8o*JJ->Rq%#&&J5+b-#uC62|fXKBuj3vF*rBf68Jt-rVl8rmc5S
z1(X9dUpfSPxOC#`n$ZZAW<<YU*Y?3ALlsGIKa|-{X77xED@n&R49Q!d&2MvD4%XNf
zN^A5SW%_HJp1(#jPzw8DN>K1fS~avJ&|#chU8$S=-r|b2TGC8S;4-C4b&ifqu-d%-
z1qv6JFB(TBDsZwKkDf(yE<J-tJLCL#{z-%UlI%yUM?a>bxdhco+ggi{t>@QV&-ZJj
zK^HCW8+VxJ&PIOR5EH5!gfYBIYQrfIUc@Is9LFhC#W4DKJY7wQ*x=5FvpHPb#Tmpw
zPl+9w!&^aNaF6uZVTpY^!U;<8F_}6%T0kaYdLd#IOXI?QbTDEtA@f6Kb7NC1!Tp{6
zlkVGkM-om80|y!Y9(;x+GH14O`}DJlrh~(|#pDH-M0A|`VNbNByFW2_4t<rHulWqf
z?i~{)h(^n*anC3$^|Op*V#Is!p;ow4FEur$j+EjjDYhPC+s>8fFkYEhcs3eMW>!J%
zVr`a_n3LN1)eZKh#ZuqBdbPn4Se%=>_pJu9+QN&4ggU)9ZT~-I87X2UleUt<o|AzK
z`tT|HWR%O+lJkP=lIl;sW1b^V+%et5H*|3zfX4(+hB0N6OUC5uCFZ5Av~0CV&qjS3
z;U((MQ>)^608`i4ja~_pk%~T+d^D>2l^8XHBiLA~lVdhuBb?KL6-LcuEOVq&J&w#3
zA@;*#vWR*jyw`a=@VsrgA^~f*aeX=Ua~6jZ5BmLxRk8yb=8NA1NE33&dVNtrHF1>2
zW@#@%Zk`)ZO?$IR2RfIs64aI0nniad(Kx)~f4;jI_n{Le*ZcP#-#?`MKW8Gv5!Uy4
z=|W~TyL6ea{MKJjwG@PKIOGH^YW@8HnCySTA9(vVNBWU_3WfxU_`N>|;RlcDR8Zp3
z`i!eCf8I4-QVPl}#3f0kASCIC^TKFW84}f^v#T`=%tb=8>TiY*bMd%+Cyor3q@<J#
zSnM9!bc_@qUz_{tymXv8Ox(UsXSj()pFSJ1eYo-_$(ZHhUkhkahE)D5a1rSc`-F=Z
zKmLR$?%L`px!;3P59)pGZ#)1fi=bm#*qmbPXqM0nN2n+l#EHA6n1)KpHX7yO2F_#V
zY(74F)Vf-ObZSvf7wfR{)9@sfRyTLFZ}%M$uASZ+VmI6QG@DL~3Ax0eGp^m4|A$eo
zumPC%oK+O;52&vZa*e1<zspqs6?}Z~n{TqFS8H<@Q(?i;XaPITt#7os_vIjt0TW%0
zh@Lp=5j<mksX_KTHWd#nf4u(Xwn9Y;sH4ph?K@#8b%`WJF^8XzkzU3SzKr#9z0u0H
zkWl3Pp90AN$8T*+GAFfP{&^i%h<ZnXuaBcQ0GLlN*l<Nzz4tIuDoo`2LT*iv$s1PZ
zzMRnUu+#Df?Xmy&V}Q8O2(qD#PW9t^%miFjR|-Lyj)0Y^vt`flpp?SNtyiSqeKuS|
zJm38zMxm(33Vx|J6Wblp<4Ld|JwZhm<v4@~-P|Q77>x3@!)gh^@NQK}%&5^U6RtJT
z3Y%_Nw(RwxHq7u6d6~e0z-{}D<^NqYihayJPJbE05fBxDD#`t1Hl9)4MxPfq_J0UC
z1PLL!`*$7-C1pkj3;$zAua$H<H9l@Rp?k+xH1<1-EuC4*;9p<NxvFy52?;lT@lpi9
zeqlT4G@6ujTMT!%`0L!fh0|!?gC3nH08s}GAnebfXJsnAjV=dsPe&f(0OJsdYZ!3Y
zWntxR$`C?u$2US171)edw;4MHO05~OwIxFl%x0%uqi3$bkTc|g(R$;o5AHEXkcb)X
zj{-}WO|7qPTlMHo)>jtj6|MwzceT{)j935ToS`4JJ;r}$+!bw>$U^hi37|29OIPc`
z%XE5`^d&Cq8Hlr%xF-LR&8PT}m+C>-Skj@aPMr;-#Cd>!D>#SVH7+sEG;e~sktgWD
zJ+f4-=jXz+?B$rsol~U`I=?OJc9RywDMlewIZ<n=uy77X(Z|DSwoUy!r-kbI13P2i
zpnZ&zT#SDd8+OM^z9vY#nLGdOBe`M$i3h1cK7bE&N^3nGV5;NBstxn_vlElI2v~jS
zv-L!tfgG=wnq|xY8?JLYTGAg8I2(EmeKt-ItgW~%8_rgncG{(6sii8^)b!=Vh)n1v
zZ!yFHY!X%O%0_LN(+fS1?AFNC4NMkC5Jkt{!dXCEKVv+<&q3$|)7FXo^i24JO9H8G
z!v!t)T&Uukiz%14;O5L?SekVOVbmWzfQ$e9?>0L~K5qmhF0<?gUrP^{sng<8T~mfL
z;wg+GR=!Vu4<z%I5~4K*6X<>PJI>~_TN3CHR~^HhJN#9gbVuDG=>;DeK59HfPtwXK
zY8mwpb%1Cq@X1vkxqkZ_3(!ZD$DqKZ9~^3|a5V!d?*j&VRuveg+*bn#HUMl~J%9D!
z#})=w{U7(fo4zTx?LFPkc{agoA?x#Z6&a3@dJSU+d4Nz-co1ncd^L7S)d?DsSX1q8
z{tG7HnFl+pq|-~GYd(|-yfd4a7p6uu+*L9^9kEgDN^K<^ecBqUCrecpJaE<(@0S9~
zmE$t;ZMj!dEXiSpsFnSt0u0uQ!xEE9|L(?9n1qdPku_H~@d9rgs|481Q)_OF;um<%
zj?egcDuW{<>qh_5DW=Y#EJYCW<jvm!MwbO{2LmRnlVF`7csd!J*9AHq8W1LQA^C8k
z^s%2d2{PsjXT)oOv7jyPIMN=-+2sMmw>iRpq;Q#Ti>ySA`0`4lROc<g3+>&3v}N&-
zI+3;QZ?+T2!w<!O<<Tv~{c-V6$Cc;BJ|n-&K4eCT$BfiZpkFjRB&5>5V`-I5{unF}
zgMUxf56Ed*TMRrwhLb4>>No|G?|EUd`Y3YpawN{DSXfxQ_c09@O&W-uq>oP_$%e(i
zDkX3+V|+p9wP4>ni$YL_W+#w?d9++TDrC)?6p|1`=FvCO?OyH+Wacgx2QkuQSaiB<
zb~_9>ybFKW$p`_CU%l-PVC|=@!i~=qRs?R3yPjV3=*|Wg*@VdBLUh>BdUc#6u!zhB
zLhA4B$`tgP6tk&N2cLj7VbD;|d%!+09@$aC-L(vwXRx-^ja#S%(gGW{7E@}>|5?<;
zp7<4}O%3vgZKZlH`L-n&bV_dIJ2TP#a>L8C^0X>@9)D>ZtPixq2b^h_2utJ$ioZET
z1wb~&d}Kj32rbZ#bq0QL9Yfg~aiQB|@f!m+i=Px}(=yJFCFpu1i0L5;V%V~zKt*3l
zn{3=0S?m6(rTLg_tl9NQL7%|8<0*(3iYvUhk7zgqDwL@VM-U@(`uz#U%m8G=q-c}G
z@8RCcaVe=h2<#^V0M$9VIehn*{p#~Z6ck`qQ}U}9C%r+%$tvpQa4A2$`==4i!zIX-
z0GI%=DP<5`2E>cQMJ6WAvKoU{@Q<^{%E9kLuxa<5#ZL7Kli?*baKBSC>CxAR2mbv6
zwcJnvcz$1WGzDlWsJi<C^t%^?N5XEXjoRxWPVd@%y!}34Zv$l3R0Oo0oJWzy-d=g|
z){MsUeYB5zOdPb;BNx~Z|MEow>*x8T_?J-^R&=iEvp+M)k5-YTTDFSDs!^Ydz8=w$
z1rvvu(s~3#RAN_OlJIN6LO_Ke;`ue3DLyBhlMU^PSKM=OW>&Zr0uqTj<?MrC^y}q^
zd)IE>;|nZz4bUH~tAlPN>G^z2rX|RBJh_jGF8JZ^I>vipcm|oGHr$J|e}j0m6)l#H
zz-tSxb6cIXE~Q7vWo$YjI^%Yu7vLD7I>eE5wdvH+^oTLHU4TW^laX@f2J18blaBBw
zR`Np%EBG!w&k<R_!Sg24oC+Z%ju6baD0X&#AQc-NsbF{SM3e`)p$8XgxpKKDJC&@0
z#&sC}vVD+^iRI*UN?^wqw@V>v3DfEd3<?RO&Rbf?_>nJnThUH)HH;T~-6bX4FjnH*
z&@)eCGc|}G5f5;Q(|~1c2M>cYt5dn<Eqqqnp$D5;)}$7$Z{L3rEPRQ!384zj|KHYy
zD}lJCu5My3rq+~Hw+yp6lo=pZCdd$jq;X!BeM%Z#zBpQDUGfrMqOAb!*pLy@PU$a_
zQc`mv=5`H5Jd<&k0jPK{8@9yW=Omj>tw2~%Usm_J&D$@6hW2A9m2?cdg`hJz4X;i=
zj(>3TUtkbUDlTuEYtuXJg;ci#7k3T_&2p*B0-hLdn+WgVNfAOI!qn5{VVBm-L1scQ
z41Yv}&hd9rUAZ4_BydE>N|Xgy@G6x;5oQ^hwlUpLq9tM5W4{u|?GbJm?BM~jE;YOl
zT(cJe&e7tJ*v|<nvKISds1Hmq?vPSjWE*__7(aOY)bQm&{8jKdlo1+zIzh)<VfFrN
z7VNezmOVzT(8Q_D?T{jbUW4`3(R)KPvoIpNJioY~(_u@=EMeR|0p3K<FW1ustDMPs
z4wa&zo^dw!x%2t*|9lF7Uq+IeYB<OCninCuVLXPt(V_59EeMTz?JbHpn81`|mUh!A
zWG?W)-iLRz&Vm;JRUkV%T%`+}-LmuR)ViBTypD9maR`}Ups#b<Ra;&S53pdQ!u#bA
zDUyLH5hz(O2h2oPaq>XhV|A7bEc2wx!Tl+qQ7DlXHu4HMDBj;(?1fT&ETIx{<2Gpb
zliCQOvF&7Q*abSh>>Ww}H}c^vqGoG)#LXwmz7+1Tj8jSt;#Fcq+rIZ356a=lh@1g5
z?)>Y1B2_nVaWjbe!KHeXemP#t8~*`fc^i{cqoRN%d*_{IjqftJHiq?1gzibcO@=;B
zISyBN)9;7Cx+JSy<5D-JQ@_u^<dS(hKVJH@=#ik|JJ@s?mtlU$Z^fbNcE$rOFc6#~
z`(@N(NG~Jud49pp9m;us*y<mL)X7Q8P9!HKLWl)D;YsIuW09<qcL4Czjbe(An+cja
z!%tpg&PX>f#Er$53@U!^!@~b9rrYXG-E#@B?Q3qJs~fdqVIyma)1%)b>`pFVSW=z2
z09qi(l2WJ1Ki4sMvcYiG9<t&5@dof~ue{*1=87<)3f5*Gnmu(}M%vdKh4eRLBU_S^
zcS)?$4tzY)k!K^%$EkVbPG7VxNeaX-eC=ob4k4l({d6#rT(X*8V@$Gi*2o>e0JFUy
zH#AdU9e;>83cfR}cwCvBY8s|}ih$I?-8V7Q0p09BBIv*LwTtPa3fR0a$CtY_OW3@3
zd_gQ`+EVwlk2~d*4055t?k~DE{=Wks;)!@@IcDI)?*CdqzO{~(Ry)Jmx5-r<<O;QY
z8}+;99renX1@_&F+~S?&e&b8@PtngLP{*WuoIzf>?g`P{uQL7lcR1hU`f}Kh$T3-t
znjafTnzz_B_{;{@$Bd3W$4UX6J2wubtp{Kcr2!Wt0gMbwwo<acHq|&pN4V6c!Z`F=
zE}izKhLsY<BO7f8jNH^sTfFx?rZ5>y41D|_F5+t0^DHBH#sezyX0zS{L1y^(bY902
zX)OO_gPA^Hc7@6W6Ci1PIy*H>IlYsTy_%$%7svmC$4Ddj-qNyDNpOCljZK$~&7VIy
z33vU+AVwHBq0i~Ihx=`_q}rq#w>kz%EPk!rV6nHZ%zZe6o$4pTB=*%>TH{lvLk>Gu
zF;C!*_p6ZnM8f^x<!KOy<p{TNO+@Do>6!TVOM1EXd!<wu*3u9={Him;BU;#zDAd`n
zIMy?}?vo?#OX5*xYc;0SjOLg5Kw)e&8?yd%j@)XVqS*ba4999ZEp6Qp`_Y>6E|$63
z>30bkCQ&E1e=<3$LO)Te^k<gFJ@Gw^7nhX^X%MTCzz|YGeH%G!rA)c#(^ZB@Ims!^
z+_~X0q7BfP8<UK;2(*VaPGb;vM-k#en@P=neP)@SU$|kR(Wg>u1rYatQFlZXzXiHr
z(HAZ<Ujo^F0=qFeJDr@9yKEm8nv-}NMKqb{q~)B~74M&%P?5)?-&?**pBFL)jqq?P
z2QuS_zEl%oVLmX`;FLVzIgsVZ`Vxe5J#jcZZ&!#kZT3s9zkb>^_Stf>S=h*ttV{wh
zMaH=z*Cl%3ote?`*_FS%Tgrl+^cPl{nHOoiG!$6raV+{z=SBumF+<1_^yLRZubch!
zz|~f=EROrw`!sL|<6D;+Vke59*i(x1KNfD9jCil+FY#jEs2>Xo-|EC5tt$t)L^lg%
zvQ|&;S1r{RFJ-<GR}xaXt}p*$XILTK^{V-%!>w4AH$qOP(qC})KsxGQ=L3WcA$a9j
z^_d-Fr?4zT%gOW@z*a;EDMe<CQ?gax>Aw~ZjBpz{me-LvddKAj`}nte^R)XitTbBD
zC!Dc-G$q|b<sCGtK4YRqDCs{t9o)hxWuqeckG2R1tOO7M0wA|L(>FvOJ@`sJTxEE8
z@%JFq{B=!+g3P+>8Baf;tmI-D31Y~XhV|OIA1#u`LT_SC)(<60(+D4tF6R7C2>2DU
zK?gtpM;yfic!E<T3L=Ba`w_2Gd2%GK34>Qf{WqHFCMa=O5AmD2m$IXp`2&Z4ySTDZ
z=P(|hD9Q<bkdNBuI>-5MB?Mn33Oh{TbLMHVDhKPj%6BXV%a-5-IjQa9P>To$$<4hz
zemTi+W}(;ga>bBY1?{?45vEVJVe+wPZa8OQSp>)J&{lzejwP^r5d(`-@5NEMCfrlV
zEAl8{xG49uJO}x<pr$|M*2%(?x6nm9ElK8}hd&~ExFN-wwsSIvIXJNQ^{gAt^nNf;
zw&-s~=HT{9gE%l`ZD1e(QHWs^{XO_vY=F<Jf~&8Ki^_%%D3P4y)c_gugP&-*s(xwX
zql3cTOcM~~DlO=L9S-auiOzcptvd=c!0rhh9tO!Axt4+;qgK1w1)2(Vzmr5UGN3NG
zgWqL8DT=B1ivk_K2N|jrG1T`h4UKw}MNMd)5iixW4DOs81#Ztfy>KQ`dZN(p77ZP6
zrjL}q4ut*(0^wusOyox~>o>g~#sWWI{%r9EzHo2=Tl`TTNxAUfmCqcYHp1rI@oF<^
zu*LnX%GpK<y$PhNnatH#61mtfI#0`jw1`}=n5qFu9H8VvXz1zlJ%`{+B;HGnXaFL}
z!kbP3S9)2SYXhNSPX^In$rgK!;O2c5p~8g1jI81#hBY}7jXLG+eou*HG*zUWl##}F
z_6&3TjJ?}^xFA<UOwxT^hwIpC?Xq7_D}#D$`a2h>y+jD9D*A9E+b^Rm9rh*}Iqbc+
z1Y|p!hCoP&ypfSnp7JFSa}EN3K<w^Cilr6<@@RfKNm@!}?lc|ew%~TUqY1$9FP^Y(
zoPkh9#C_}|ikUL4%FeoFVD1qB-$xFRu{57?X1bOBn16Ij3e;{PDq;Nd(tQsTkuVg;
z9$b+f66TQVv&{F{G_AoZ4+&*kXL~3GehI<A2iqU|p|~UMmF|@T#pee%v8IlCMr~UG
zTfb^8+rhJnX#gR4&9kAKf6sE~b0zD;O&Qp3H%C#gltx8xcl()!^i8xi0T?bw8h;lR
zZ)AtD#I8s3D}k(AD!A|DAk#ZP-9awZaW$gTn#W~di#P)=D3Ak+wWk<~oMJp6>X$__
z48In-=z_ZFDXRobH8#yvrh{@I(glXu|BWaoAMo&M;ECD=b-%dWqe|t1U<0^#Vc`f8
zBl1uv@BXewl_r%cec|)JJo|(!Enu7JbsAD$ERN*c;z~%9Tq)M{#tvi-m;mXtOXkfs
zaKhJam#TgXd~LU+2lQgQF8v_jkpgHe>)xCnF6yF_3GB=#Ee(MG_X;2l@xj56n?{fn
zHgO)`v{#^Aok#2MhWRZ!n}+MW9T&9%I$M?dGjJ1VmB;y=%WRZyiZ|qv=HMA4z*|m-
z5JOi}?Q)fU^%67A%IMYJTV|H!rH@s#B`rDoT<#v;U<_4H`ae8<WmJ@1+qS}pgVX@h
zATyNGjdXW|NQ08nNP~oQcZZ;)Qqo=0ohm6H-5?<y-yZMh{eHTZi#1$xvG;i%Y0p^x
z7w;k>;FgYtEdM(NdVH$m$;0!C3#}rpQt(sSMSoYl95mZrSHd&M3yoTa7X|#S@jM&g
zZo5ghmI#fBNkMAj&@hQ#%MAmG8|~ANUX{CZbt>CGi=P5`q&ACQ8M@?41M&%^kcK#<
z@FkK`WMTh3_#V7WF2M_L8*QMk|A97;MdmF43Oag85T93<_qYtz`+k8b^uYCjJ_nsF
znMXAEu96=<aTf)k#42@IpxUyWUX4}KutMMo1yXv%3RMHGr+a_P@8T6$DyBA-S9_LP
zC16A>`W9{Co5{?2=AB%8tXVBU#?lA$`4(q)Kp}Pn)`+>|cve~HR^jI?t1~f{_HB?N
zS<Wj4>t`G)5&~#}<9DU!4+JA&KaNKPRN!$R9(KG3Jf}Q>2)y^~n-8TCXSntzb$0W+
zK8quQ+G_r(4?kI@fH(K;F4aA`LLI{R)V8`QS!&he!`Rb_uXZ|DCXcxr;_JLQUKwW8
zmwdb94N)vVC=f!gbg^HT`20xl**ibvd#?e>`jFY0I+$Z1;N-&O-UYSn0%wVdkO4?-
zE=+u;CeG4&Zrpwx4Bh}Kcw+Rm=m)sp?I|$Y&MLXL@w@HY^e4V2YfaF4*dFj~Ia}f_
z9X0h7tc(UiNWJ(KB!bhqEd;g^2SC_0H-vV^b3^j-@}>p*fhAlHd<WivxT9}?)pge)
z?IOenZaD%EL^?BW9?f_G+NSg~4j$GFS_<d~M>Y#Hf6jDGZxJSyP(S1_?*_5k_CoL7
z6;VkHHJZ!a?HX>8is*OdEw+c3@wuVwryvx`biib1kD74@SI4r%_J0sAN65XU0<uW3
z*!x|Wtx^fkT}fWpobr44TyuyuuXm>HtR6%;9L?r4V|b;7-pB#MkI%3ueUYltH)Lvk
zYB4oS3wcjUzW#`y{&ki6h=|Ye^Gv*rICUK@jcm~L2aFG5Ujb6JDuB~q6`IRSI|PC-
z6%-XCe}RaV<nup5jA(L!`WIw~RAg-uvd0=MLgt-9_(R-grk+t}kJUGZKX*!OLaF@u
z-2X~P5-;tb**^_*04k?scq2tzTwL;4!@;e>0pIjPlHf>wlU<lOCY5X`eU|nc_XBu@
z{?jb&GO=b(>%flXzX4<UZ&v+}n(9?wB*Uabw}OZaEpU!0QK$U1ko(=1OGfS<xcea=
zpOhe{0nyixAx_;Q664D8SjkXb9}@g3qMEPtYvH|por~nyb!lxrG5WU)iXOK)`1z6a
z32L!Xu(`hiE5m<f6#9w1R|!@Zb<31+p#+d{vLn1SvB&6qI}Kvk4ThnxhG0X`Ig>Jq
zN<cxu4D_8zM27BSNYO9SyHZBqd%1D%()EeigAO_6oIwN0pWks?POyVcMDY;T`^~VY
zb;_v>3j)nqw`l0WU3>x6@4^_f<St!nK}hwhHXBCc4o`&ts--b9M&~7wS!Q#qby>^g
zI``4MwY&6<!h2J?sd!;DoF_ICmWMv(rhzSZ`#X^p;^Mf)4$>aD;s%+AWLF8&b*#^m
zFUU}n$h1CrgQo%hGIF3WJKf>LZ=!2lB2mp<FGw|F*#lJKjp=?Qnzp>0I~xSqpn516
zY0R6o^NF3)#t@s^C{u>bl;SC*$@3!e+e@W%H1hlHiBW^(*f%)Sa5Az(-Y9ZDb&zwv
zH26ZIx@U&M?H8hM;axFSf+TM@$%XGK({g5jVE&uv^9tu3_gi=dj3D}HB%X?ywCjZ2
z)0z2??{6C@7!gTdJR5}<-|aAyb)aW=pvoIAAte$%Plv>#pG<|l#ayZ!sW)s+(|@>v
zfDjyf1}@(k@(7s&0F74jz2Dv7=k)<zf!5>JtNo@^)uS7rq%-XlK?EaDnPBaPW1*8V
ziCP-y&09f_2*V|^4vhvvOh3x6NnWbW>9Mv)!Uot>=gnYV5~q;wxu-@W!_Gq_IA+&;
zJBYccdyj-0an|X;wy$Y#i+zXvx9LBruc$8MOM+}m*LAH`e#F3=Dc9Je?pzzW=tR^I
z(9y+d))P)~!BB#assH=iq4c+4N8$^TQ?f|wpH9dZ^OpMU%N08HndWbvi`gVE6Azz6
z3$9h`lfiEi279mg9u<HQu8IbgHIcKAOxiGKO7$l2ZEV29_O3t69#{t^Kq$IP_d_{z
zD*Qt|8W!oas%!@wT)S2hN-e#YWh=vh=k3g!_~NeJ1X}0fBK?+<ispT>h+0DU6u5uv
zc4(QkzJ8!pdj9Kez~5*pcT^8XgT`thERQI1nYKWe<(EMk{0j2&@kksUUI<*oM15b5
z`y7U2>p?X9#mEknb3qA83n5dXuJ_*<7Z*TJF}s!s!LG^SbjNSGMmHi&kwDjhU|2oP
zNq$5JdavqKf|7nIG`B&~<3j4+-}wM*qW!P@wf@T3i%$LzB&44_Z?#_+@5l6Ti(L{<
z9;}+opT1_gpH6uTZWx@NtrK$a>Vj#P6h%OhB)Gnd+KJC`52CZV8FoN4Sa%%Nk#%TC
zm!Y55YHKwx|K@qTx7=O7)tl_5a%)`<x{1I{!rec!<yUIX-@-+OsePw09;FzraFZPQ
z_6`M!zdNI}y*X1NGwZl?sjZfhk@(%HS&U$RGCDaPEi&tqC~^o}c99=#1)0+d!hcH4
zdg{&Z$XkcViHdCha2?Hw3xg&OilWY;6o+hig+A6_BWV|V7&jZ9E_y-=*hiK2br@)S
zsWo#2{{d4+cz*tWzWK*aSK<c;uDNM!tbkaI*|qR~>x5*z!*mDEDcyrv&K8yO_}#4$
z^7XYC@~&3IcI*;x=syV3`j2ZoLh&1US@h^*Gl${+cl;0|&#A8lrB%K3Y3AnSe29=(
z_5ss#vI~EAhT~56_uo1nX*O7Xsl}_@ws$vZ=@7fU>Rj>+Mil6iFG%>hNDIM!Wj!v`
z028O5f6%=KNMNk$^M30eKq}PnQye6)6uabAjq<KwX7DO6^WiWpAEG1B-vq=aax4^I
z^E?)2wboKtXlag;Bn}#Vwjy`!G|a-clfP)DA}-&GKmGbQs?H*yd8ZEKtGk4VH+~@B
zd|=AUfWmaQSk+f9pUOmyse~GX5qW-H&_Eo5o^o;ruZ(}ybA_SbgZ^W6k{3$-!1JKy
zXr7TJL2u)cs|d+En9*ey#|pcoTB!S?!%EMLq_S2$$298XF>+d8Kz!RJwbH$<iW_QX
zXY7NhX)-k#io{nJes`YKc{4LB*){v3Ia#H~>SrZKV608VQ+EE!a+@dxj;`l-!B8<P
z?S?=vA0?Tsvk}ahj8v=R;}J@BCBufi2ChO~hC4K(JWka2;kfSFRn!aac+`37{hRPi
z9hKaFS53Qwud1?rOS`{Y(47IUrTv4S94Bzc@$%ouuvxn1qj=dUn0;of|D#C56sosO
z)S^dxPnyCB7zcpmcw*Hcc5}YyC1cY^9%Q&)kQG>HYTcRUQfs`xz+^)fKW!P~Pu&0W
zm#U=%mFv3;n9b8{vY;}++AJ1%I^Z3Nq-Ii&XOhYS$BFy&ts=0=0DD{{!J@y51FbRj
z>Iz~n-)wloHJM=9eLlo#wW@8B$o-I%wFikY%!W>jCjS3@7s(G|qEl7fxnsI#lX03y
z(gwG9o(Xp}y)E(oJXMMeHBZG9Hz(zoZg6^|(D$o`--T^86%h||6{E4__QtXUO2xPU
zVaNyIiPAu>ELptUmF4A<L>@#*V3QZXMm72NfN{}bzCb63qet@b#3&$BlA87#fd%V;
z%HfVez_pq?HMS31i~6iIT#c98i*J#cr!utpM_tf|iE@9eI#XHJyGq4{Yg?(-80J(|
zwiuzA+UIgSAk$<Z0Ql1WofTk$Cq*H_lcafhvu5YvkT*ulQG-O<RoQUeSL0+#<PX7k
zY>(e$pr$RZ3dY=LfEB0+JNR+Yo)A*R-tn^9ezGKuHUwS;GRd-q@g@IR%?HwHmvY_j
zH++O}q?|=U&Vc;&sfy6wMBAQ=!W&eQ7O+K?{NTK~nrmHs++a4q#J98aarJ@^18b>&
zzRUUUcx}})vj3*ou)spOv9HDihgkgN?P!5xr=<(+&%|%fj2-cNZ!0vv-n_1reBA%X
zA<o(a8WauRm7{-XT8b_(Pe9p<56kw=k4CB+khsJ(tnW{l2uRPm>ey&u>gZD4&;k7i
zZdG5AYx{lid&a?Pf`lgOP{4+R)CJ_KdF9<7Pb)6YkH||Eh&FYeC1g%#Ka^A9@B+~l
z$)f7tw*Y!u>y7U!@44Z!N*-7VEcf&7%uOW5=T`;8m4mQuhe54@nMJfyjoE04-;|QZ
zR!NKuK@#mr7JoM93X&tq#|9>`AX!0VD|u7JFLuvhB*v;f++SVXyuwT7NLc9$vjzHL
zkJ(O;5o!O{EKKB=WY`SdIC}nwtC7U#&ifL{>vs^{zisPLI75aG3bq6P#@ruAS$L1I
zEd1==5fryL$Je^e=f-Aw6ZoV|>Q>}-0n_}Kcw#ba4(=>9%@Dj9wKHY+=UU>uJK7<~
zvO^cDoSM!hRbHY~8!)?^lB=m%TF)2=cm)4`;cObQb{Z(@aanee#TmH3uu`JKGwHs^
z(aikN?029h4>t<82-c3xpeW3>!Q_4AhnpkQG}}aFH9?_dOu@;0`Ep`%g{ji1B5(v`
z$K3011m#Uuji@bP3VA?#q{_B{_(C7{=NHJu@d?CkM}{OmV3#l|>AfU=lt$FyoWPWt
zZ4d1H@Rb3@)+?j9^o_OXnXA`i7qP3NnLd|W?ZP!#D!9-7K>ysGgFUrG{h6V-q?Of=
zFGl^Ov@D<p#H!mIiX|ml0A}9_pk1b92~4~Nh{ObF)wG-joQv{%cRV#76ABy}%1NLk
zv<)i1Zt(@)p!n2O;zhnbxh~!U$uK-u^evhXKdm1Q?UnS_t(a*|#2NY3gcBv)Wx5pA
zd4zr?<EZ3R!%khKdK@X(&nP_{f3b5m9sB)!71dGYOxn>WV!PMW61uXRRFoefZ;D%$
zgTyKyJi4FtVs+RvYRXg=3qR*}l)=8~TK6N!761e?1&AZ9_Lca3aRrPIIzoQFSpM~g
zAWyU|p%psGJ|0m&kIWeb7U-H>(dOuQT&AD?0r$8MJkE*@TN)1n*X!sLjnX238DbD@
zoX?Pj5+2gmDjg9BL;q`dNNXHmU%<eCjv{!iPl8UwGY}U3nBI7BzCHoZ$_FgZjb*L0
z*<bdh4|q5lR)?CfXS&?~#P@DSUSZP2dMbi|y?OH{=iaHZS0cl%XZfKXzq`^a_8rM?
z4aCVak{MU(xHZ}c<4JoG>Rn~TWmlEOp)x1dwDvcx;ysG5tRn{6czx1_v>RR^ai%aj
z^q1tv=K4Kp7Ta_^AjTR>|4LuMcol14wwOxwEsVIyIVe)MQr{grQ?oSu*Y``vl8y@#
z>9S+!&qHOu?066dfh(i<s5l54MJ5U#<jO?s#)Rb{O0r_hPYg%IH!7X~<qfhxfr*e$
z6TI&v!aszWKvh+3$lZ4T9&PvNl_o!f`o2xt<Xq|=K4eMMkOKM~i@=C@z);`Y@n3|l
zI}c7;+N(HA#@fN8+`DQQ{D6CMeVJ4vi5>9g>}pQ<68m-I_vulxeMRHt5;l)l%8ojk
zgDR(*1$xOI4r&!o4OqUqeP(^OKUjlU4qK7P8wRkn&D3`cJ~I@;4V*YJcS(8)snwXO
z`>1HseOaKwLc4><r~}dz@dHNno~7kPJwwy~VJuWaTx#b>p~;mI%G6xS&i5u$yxBO3
z+NORZ8SEJmfEOa6zZ<p@P*mdCH1-#cc&p*x#|v&y^Oi?}8SU3<jF@@k#ChyHmf&y{
zMJ6bNTXY~~UO6s0iBUG8GJox4`{n8iXOrwn8)x#?6JlclxDoe2SGsCXCuiw}kYK1$
ziZsF6PX#?~Rw}-P!MgnfVSB{{tfm!jzK<6|Bd;?l@l@=Vr4uqjW263qHw_`uOxje|
zXL~8KDu#Snostsd_L=;Hb6mYUfhjo@Y`}7m(dh={FMS|2J$lm(%;1{N(cTuAfXGo|
z0Ydk=XwptetfT0d7ojR(j<ekt!5X^<&x!p)$v`rc@mvLI<H=Z*HANI64ge^JyK%DU
zn~Q$f03u%HLFHH=x%?ZL!N7KQ^2q1+cenR*lk`0e)+tbAXhUO!FTcQ1+HLMb;&9$W
zKqk;&lbYq|zjN1==>GBg!WTt)kL(wE7b2b;_bH`*)*Pa4We1K4Jo7)H`V{)f_kjRI
zj{gG!++Bl;vjx?-0&QVjDb;dgDYZfdSymMF{JMKv-`ws}&R7y?$>Pi9T>JS%csK=|
z<^-#Ter;r(DyLu{DK<^2BO;Zwtmt~=okL?KhvYJ*n4X$keted1a_aePSGGSftMl`q
zzl$7-C4(JG+*1OySxK%^hJPQ@_giR?0t!lnpyiuyP8WP?07m@iLMfjaj75e-PP|oL
z*th&mZU(?o>k3j4u9(i%?-P<b@4gb{Ix>2n3n*T~{9|(QxgwR{(l>zfcwP&{d`A4G
zKcRq@2cY39t%3x6tD3O9K}Ubq4=JX?M}5?_@3}rQ1VxBOL*gij;!qr>rRF2<5d9Tu
zCdVr;{^j_Jg4l0}6><C~RHyySb@Rz`F16XOsJ(oTQi!#qd}|k(^eydyiVlLko2jEv
z0>gG$emG<E_gE>FY6<RirdagiHwE)`2cul;ho~TOd4R0!)3n%1N~XJ#AjP3^M9`<_
zyqN1*slJWRR;$-a2V3vlpnwQ+6(AW#j1R~q1<BD{mc;IU4O)TpXwd6J9BD$oq@c&>
z)!OKIem_TivRi0y`V3prK-#B{VthT>Hrj%kr@FXISNq~Bfy5<%K?P8%Uui1S`yM*s
zD=NvVQ*KH*sZziD^bPh!gHC#&TP8$NmH3)EB>gwH)<=!0RT)zb8jjtmHOIR*PEk5G
z(PhafozEVVE|?D~P3_5Yt1UTq3q0Nr0VBz?M-X`_DS48rSVGxlD(XCA{kQ&=uamb#
zo&s%r3u}D68zv-VWqv>KSq{|pgW~{wt9}o8Kpzr<+ny{PoI=69<9Zpn0<MF<f7jIa
zvI$-ye?hN^mDp`Btcy6_C7CLZWAn3@b>QuY#etl^U*-PWCxh=%B>qYp<+Zr~N}U4l
z0V%>o@o81j6c|mLS^)BE%a2YNru|r*{v8BwCD+EVBZsp87B?^L9`lXTpY#nDD?2Y(
z_qxor$D#BVw^z|FhnW6QQ|mN(%7^p{8BNlJ@1|8Rk*e(Fux}EjdZyN{%3R;^i!>gQ
zK4aqHeARpK7GMrYk&m{c%spPnjz}F`2!wNFn4?&WD$2;o1T7*QfrM?A=>f9S@h@wa
zOl-9)vu<_y#>L94(c&D-;`kH(k#A?a$Xp(O9Y*i>K*3h)?LQy8L7aU9M1PZLZJ@uo
z5i7Jar&fN6I+j{&HKaXC5-O%JQU#bBz20#k*(AKRsF9I^2jYIiL_N07@fQN->B)-p
zynRUmGT{8JwwX}}r;j;h4o!emFeZtk>JL_%Yb0;9BN&a^;KbE?kOJ-C4O<D_Uj?Xc
z-FQzrhy*V+A4~EC(>k(jJ3cYdG&j*mc%R_Okk?yb8TWoRpMMQOd@ulQs$Ch}J*Gg(
z)B2+J^^emSTI2GGkY|1sx*=!5d(Ra{oe8vJMB(*?{VfZbTfu`OO{Qpjw)8IUxk#Sz
z>o)a&PdDfrpNn<>7I!h_$z7XK1H_}I^Y&M3yC?)}5riBfhA5B_xste6sqUKfC2)<y
z2H=uU4M3zfGUfbeLV*?iA{t2EEJx!|@vIT+<RH5|hBh9Fpl_7JY3R^>LX<Z#M5La+
z$#h9xd$jNAItM`P>2+Zruv_h-@EykkndW8q0<cK&SYX|M`$yz@w`#P6zNpN1|1l{F
z?NA`w?~I5SigXv_<)b(czH@P5LW<}gZ7sI70k@^ZnCllX`~}|YUt*tv%TFW2qKGuy
zW$|0nn*9Wu$CnR%=^)$~Y}Z*HqK&Gf`YOO?QX(Wg-tMlgj_n_2DSNQc67=>vXy~Xg
zXrbPKZ@?1;|FEN*y<`<w%09rGOnCDLfLw_{cf+rs21R5S^^1gd@|&;60-wV(>&r|w
zBY!2SxcnR+dS?j7*=_>&+>6SRO~OmKAVm(}oMm<J^Qd%7@58Ytcohdg;~+QH^U|hE
zMdi@A27_>IIVURYtwu$oU5mC9{1@zXtHLj=GPb*JAykgQH5Eg%s4!E0&tpye7Uwdq
zVpZjBGXK6~btzov0|+1=M2hQ?tcpEv&yPCE*Qk~+F!f&FR>61OHpklccPPOf%uDRY
z=d|7jV!EG|OyWAe11E5+cr0&qNT6G=yI3PEzDGUTD>qhBk#V{b>ImvtbU37&cpZ=5
z$6k8D$eR#vOZvv2+HzS>4oR~J05;8r<bZNW9-VAh0o(JI%&xxOb9O(WI$2OUvXc?X
z9l>a^{CR<bRuH%A5QEcnZ=^+C_-o*YIrsoXkDV{0IcmqJ6&4!#%;@{;>o>Ag_tnF7
ziN!2)=dVAF9v=PKI5q>%qbE#iHQT|-Y<P0s(V+KTH_scY!E!yf&twVT6$>@^3$(8+
z2>%{vTxmfbk@=<PE1-W|rb-2QA;(07<8pf0cu%0Kvarq-Jm<u4Q46?lGAv2}rQd(-
zB0}IGx{4!==gHiAMP6{^S#yg!I9w?~5fRlZ)&4MfLR+>&^N&s}jVZ_YuIO(JDD5!D
zyHb85Gb+t#$$Q`UURn_9kOXnzVGTFMOY-6UY+U_oLFq(IBHj7ZQ$M_9v1Rn~v)Y?K
zlq5%c1c&Mjm#=LczZWF;hu<5QuuhJ74#A9f-`sT2u_hXNUcb#r#FHR~pVBA8IVo$S
z5g2*y9mJ&D_9Ba?xBsIydC;Kp;51`#q&?OBkb#~eo(kSu?2~i?KWd8mC7<-|qz-E0
zSyX7f0547~d?qhLx(pR<W~1oNCaqk@BEq^a+4NA(>GW0~yWk5lE&q4@9%KT|;Y5CC
zvUINk)Fxv<u-fhqy%2EOHsua08DchUp2lktU7tc$?%CqKl9r{@RUFbQj4+Q2;Wy_x
zCppKMMBQKV0Zj14$C1-3);n`^M95<*mxYKI6RQ=<?O_Yaw!PiEaOYz4i~3lqFp{3c
zqDA^%*`{2^fhfw6>UQe_=0%425XB_+n^|;R)rexz*t%W$6bgD0^(y^lomKj;vYReo
zYyK7&de4r*O9Nyj(S|O0-#MQk0`EmnroJ{tc@)?AY6@3~yf1}G$2g(0v8ysQ^A?<!
ztsML6yfJ*wPzpOKVZCoZ)P`V@(b3b>`@c5^97!`6JTwzzB`c9~Ut~AK)R5R563S2i
z+KA6>^`ks5<{M{qjBA3oBR*eyNusa7Ii<KMyAUn0vfb*O8C|#;&w~_uiSG~YkL=s<
z5S)x5aPRf&2Ge|1k*}hED*j+bvj*SHxeY&m@!|V)hrJ5)wWbz69mCeVAg-^;Z&m*U
zOBinAYTTooAkcGFFs5J0ApShtMZGz>o&O&na0*=|&ZB!=rX27iaM@s--R~Zwd8)Da
zANWg1zL*5qBKQ!!@b=6{5`9LSKV>JLO8x~q>`?rV+E2Ylr$f#p+|N7T7;ax$t2g8|
zyP9uGE;+C8tJh|@77UJ5N5<wXdPYhgqTx52k2hxe>oAe_d}gs=_{`0)&~AM`Y{cYC
z!eahet*7kYz!Hw_=MPnX;jEP_#E23i4SywOg*{)AjD$Ku?%^uPL=qbvpT#~2_6k_V
zF;(ka{5<-TK9srN?&5LHi4`I33XxeTy3H1T0*^`~Me!^iCHz78i$cZf!>0st9CVS{
zRJc8PCz%VjYpp69eXC}AHnGVvZK<pZ&-db6DPn4A=rwwDr&r2b!_qtz`6EpW^?7>E
zL0p)M(IV=e%KrctMwBA4q3{qNBwPxoD(5yU6Xc^;qAOdVmU)_-7@`s#N0&tHKOMkB
z0KM<^i|0Tdqwo`3Z1(8gtx8r#dEa?%ay4CnQqj}CCat=p$4U(L_GXFbnlD@Q6=mn^
zMBCVfzCy_@p<mDr!JGR|C`23{i&k+i{fPIm@jFBC5wOYpjvQ#$UYaQL=h|CdoG-i6
zhJ-{}?32l|`PjZ3=m}2zNV%RkA5%<U_G)q>T?oJ2?DoCDs(*WenWtl8Je89t=jm=-
z65Rwn<kf!;dw<E-VC$&t6?AGAUsw_4c>7z_V=F%eC39rz5DEUvM#WL`f!ox!vF!SX
z&<Fl5XE;k|_9{^6*HqBIJ?t+94VuG^=zz@}(OUC+_#_q(to>&#Y_6ck25n$A<R8c*
z<tu85&r_=_ZgJtc{C>LD&g0^`AVG2Y+#+J&lA_RYf2VD*!$odVrULruQ7H>n+=B5D
z-B(UNma}syF6>gjZ3`sW^&Fa)wqY2~PK(2h(5ARZY~e8J!918^y2|TZ##c-O+Dn_+
zYWi17;!JXkn$sxKWY5#}+xYr^H8|x1Z7UDU1CLTCMhF?r8u3M0=GbKRrIT4e4c5K^
z_T-mpivP)TKEYv>6bNp7i3l3fTWa)Ravz5mo_qyIx6SElJ*(BDSMU3PB*))zerW-^
zWThVUeScuMzmjoyrH^c%r~ni+h2x@l4x^~FeS9;}|M7+)v-gH>!-r32a!(nZABDvj
zNSD7mN!99mOqj<7%y^$+N7g@zeO~tK-A<_^D7kdU=?71i8Mp(gJd&zD^eqj7dYYiV
z13$=~j;q=N{b;@H-!;}5gMhQtx!ydy50&_#+dzn#zE=Z!t4`q@nFeJ0#D&9SV?HP2
zGCYkf0$JZrLC#!7V2am|nY9UJ6_vXBb8BxWE6%o!MEE)pU+I6zt*LOBnSWOxsT_nq
zVM|Q<#Ev?pw$x0K@oeTdH|t%wWY)E#PsK>m=8eu$U4!MD^y@d(VMU53RJm8+hNCX;
zluQt|y*}Wu{*L`hV@s)!WMXdzL>$guTMkt=8HZ{BCM!sP2pJ1~rKRr>#4I0e5*kb`
zL%#c?FhR*u%GqUz1Q}YqHQ!W+Nh$nmN-%66S>s#W?`T=aI|DG7_T~xn4P?wW09@T@
z7#Idl4@Mr$s<)}-N*@8!2nJskA|e`7%$L^<cqpc8mseJ7=e{!cAg8+lqE}uBKwzZ+
z(4grBkE0-c&xhC-q(>2|*uzh46L!*mn1Ey!g1(#tpp1ah6(j4*y1KhB{@iN<a7r)Q
z-4)1Veo$ImmsR|6K6%G+DtX@H&IB12115R~J%4%KDi(vrKHvtT21ND|P@Y_9;cUBM
zr4VZH*0BG3t;F&DZfMvICv|iDuvwi~*IEANt8Wo|crS}g;vDu$k(ph|aIS4(WXwE<
zWphsCkjbTw)5SlJb0$v3E=Dqi7Sw;bC^2YfQj?ag_X35Woe@>Z1O40#4V@u1LFch=
zJbI%kl24G{jATrsN!bPC>cAHTiWPhb-zO#lHMC=~j>r;F)2j8s-qFW_rbw<La^viY
zDSm=<d}tWJ@L6Z{Ti(@_EdW5A)oTRA1ue4QQS`_yt~y0WMI|92n^J1EvgV_33Jkt{
z+#a-kYLR15DpKmo%E|fNkdnR|)Fycwu<^)1=f#essH<^iy1Wwlj-WKDzebTO`p<A=
zx{a&N2T!0uv!&F>F@;TRR=fV*S;TR{*6Q=v`<OE{M|T^EjBpl$^dq}qlRcN1$s)bI
z2O+3Zw*Pu5{n$i(1wCB`dIuTsArHQPR*h3!`RL@3kv{RsilpH<q3wZIW?FIWpjqeV
z{)gVmv6U6(>UazR%*zfoit-7W<7ssF;-vYRm-Vve8BzGC*o@Ys>B-=>!|4X}@eTH6
zlF86$1Z6rqm$5SGb2}R)T%eAj1y=CN=JVx<WH=CN#b|zq@=3xyE%vUv<0PUfJ$EZm
z`1)B=INP->-UD<cIWWJV;PQFWngC6VY(=aD1SMpBByI#m8w}$;ul6EeL2c-y#Hmv6
zS@VJeAeI>rxC0YD%C9&$I3QSg_(Nt0-@Z>`0VRY4whYUC>p(_T(z9PV$WqmNA#oR5
z_4?5i8V}6q=|2CwTNx$w?2E4y9eqAxp}^%S`f~1HN+nzp_ceQya*#V2z0<+MShm5R
zpX=OsrZ#t{j$7sQ@yQLN(nq_(rs0^a%}HEe)Oqq`ZPn%t_g@vsp`-Md>GNwYn{e4G
zs?B2{OCkeBi^4FiCRb1%eRlnP5Eq#G8`ibCd2_!_KO`38^Q`F9|78IvGR1)^K;7@E
zT=qzp;_BnTo}Xun{3shj(d$@IM7LhmnTd%s0L-nf>l3?*?6<_2ms5O^Dayvq8uYDE
z7zvkWD5U!|n;B$^>S--{E)-t`K=>n{MWEp%Xgvk1Qar!Q_O$i!!PFh8e&)5^cwSDS
zWUj(o>aFp9c>fk~F<52%r9?fOw&XsZvseD>e>^x;_)zHI!3OjWj?f<*e%4S-e0dZD
z<~I{gUM@TR6GfVQD=GC;K3{##a4y_?<il{vNh#&0RJBd~RLfN4**|{J+QqOIoQ(n3
z7r7{WGF3{GwdR$A>+elJA+88?_J^V4ew-I`8A6_}$)e9JTANKb{4rR{XP+x70#E<O
z4g@Q!tzf|I2qtjek$vc2|GVlruo)dbExJ__t%t{39K)d4Vs_+9jllD_HvD`s5jYF=
z9Hl*kVNJERa|1Xc*T1dt)WZvl@UIbJr(2L0nzEnpPqzlqeG*vmdy$eC?x3$Cpl@ul
zW(6z~8A>mgwkxFOQuNRR(bsDs6=JfKNAqrr=)%$G<~dkBLw}R?><(Q}Ta~}P-5AN7
zT>L?Sae8{svE5E*?*H_?I*>mRg=#FylXeOfG)x+3kK`hmbi{*S{glqSeCHQ=;S?oF
zMs+n1bJy2Km$Py6=DE7k@|_)FLeb;gpmVqHc7?hun#(yF8N>{2CKMr5e-y6@ChV7)
zt^|*B;_L`pSp*x?&&G@nB+UQZG$9*x+x^d-cc;gvSWHo?UJNm0zIyT$fOzJ9f=7P7
z$g;~;e!GVp8(c4AIb`-NB0kZlM&MZlC*tpaDBfwKt$YK9mNgJhT<aY}$}NvfN435K
znM!qZrI8OOq*qmkOvb+~sq}Nh*b3f^X7P++-taNQZ-YSY^<yeu@xZWe+Y~su9l(UO
zSc~4!AxLN(1vuEA=oNq_CLoc4WIs`H$hbM)Jt)4O*IkzPsg44<LEG)rW-C7KUl3DJ
zrDi7A-Oi0z>5Zw=Q*;Y^8-BrHUTc_b8_gHRG}X>6)(2=o4QtXJtiF{}vXUD-G`59a
zBWm81)6MZtl*c>)lYalDYIZ<j^>p(PF?8-QHp#=OZnCu{)^#!`>T9zuh-FxE+*D0M
zD(t&YFp4Xzs|ek%hA^8=nj(M-HURxiDN{~V#-?Nl7{4grG%ee?jl0{Yqjxo3OjdjL
z>>W(oAKq)cY$*F%vA}|DO=;{3jH|jIJ3$(~_FX+#Mpi!Y=%^0Z4aD#~veg9UAtsWJ
z#P61q7N$UIdrH0<@fg3_FBq%+xdSLyO4b*b1?8-k{fR<pos=kR$lN+SUD?-wN6OQ0
zaBL+-A}s~Q7I+97Loug*-q0`M(y*v6$8D<%*?fUxV&e1e@d*i~2e}IK6(FHRZ}HA2
z{#}5t@QDJ9tW1s}UFd6yPZ%~v#b31k!{|6~FM$T>_h4p=5(->In4B5bG46bgxF-RX
zB)CSTBh^B{2|FJ>o+R>txu6s_0n}!f>FjMuc=9msx0})bM$T>`KP<X)ImdUKYjd1V
zq{VByH~W}i4Y^~&pgO6Qi&2}m5Pexm<!eriSGr!W<1lU?E@}sZ@`&Y)8jz$_GBeM!
zyiH;+$PZ)Jxq0Bk`Srs9uRpW`o#wn!EV`L4-EZJYSTBWGAhhY8kj%2n`bIhmt|T}6
zFiyIz402WQqsF|SV{A8TNymvuPNXPBNWMM!ySKWpvfk&Iu#M=3ST<~Dk{tClMCChm
z4pjCsk5U?)(TJo-IM4nSn8;JyeNt$l9wJ))5)8hdHet^w2<W{Y$&Y1Sd@g+Jy_6eb
zjuS<>9vOrHZHuws8o&GWm2Jn_g9L|P=i9@v$63x3a;=+%?c3PPLLmG$Ae~wX+V0Y!
zqY*$n;-=F3^`BZH7FXMzD~ulTMQ)X2i61~j^iEUz<qh2&SnSPpI~9&TgWIky!hNcN
z-cGcb#y=zNbJACWZ%Th)Jfnd;5)aUpm6)=7?B|7m$p#SXB6g2lEOq7nw<T#qHMrBe
z)Dh{S6sX_DTA~x4lv+PpjeX5ic3m#{gj7UDPL)UGUV@b3E$3}M8P|j*&*moLe-pSU
zGC(pg^2xuZ#ZReJstt;fG(9pQHjbA9$|9OsO<&SR$z3Owl_s3dxpG<O4sx{>?e)G5
zD)YoC_h+QW<G%9?MZwlp`M2GI26htQ1c<<Vx5rBk>tk8g;*D&2*83%ufcCO|r%B%{
zG4VMK`{SVN(9c|7Xg;3W+xu5YYCC-=(2HNI)_R{nkl{x8SX20MS$27E;Kz0TT%})?
zj7pZyMge6v|8xd0bL7XZA}i=*c893Hco75}417k%#D)F)c4R&fvi=J>ogdtv2SQ06
z8&^9h0ivWoEg_@Qg95gcPwBB5M_wEk&_{MU+r~Ds5g6zL%tFj8fvI$ltL5FP@5w1L
z5v0@L;(wndSYB(IQ%TlcTfy9#d6V;DKS?22DPb|$4mj?c$8l~q(X|s2)9D!%7bG1k
z3jg=JKSP;8fFx`IjT3E<o&%Jr6WJ0Efx$7N`~Hzc7Q9Rp8`vw&_|{WRnnRAORfIEu
z*?YAJ8xo~2_5h(@)}(oNDi`gE#N-_L!K!ulz4ZXeDGdj40qV{Z^YezN?IpvOvNl|m
zg+;z7F$!yD^#s-(2Zpb&QN^>O-1){-tG6l}B#T|$@_48>9k(g`rYj@WqO>RV|Nch9
zMcJi6f-rmQMfcVOvZo*+4lsj1tm={S(HaM!(TQjLTHhfK$YM9BoE#G0ZA;*KY#fa2
zE3P#JB|l!EsH=A0_dcpAP|H-lT@o+UZ>SV{RjSLKT_%cB7#R_PFV1p@EL6cZO^__N
z;qw9T`rI~hmMz-_e>OHTW~Qdoj?tmqHq)xWLoV#M8fQ}jthJ|tCLyXEK5oNcN{h#(
zmYxp$0@R(^%)RwEv=+o5s(`ed7?A~bHE*QH$=j?R8w!3>-VJgHZ*^+K_kq6d(cxro
z73G?47eyWI3>md%dhS;ECSd9l{gutSZiDC0BF~W}+a)jF96TSrHLP$u!<}Zqu!T+v
zt^#Fy^F^ajJ!|{1`BE7E(??EW3al*}eQ#1Tk$WOfW;a+eZ1ERf1P8fKc)h03)DU;{
zoUnlptspZ)nIsS*w8xdIc%TW#`MERGgW!922)Es71d!Y`Ap4zo?EfXn-*p1xd`Rsg
zvdI@Te9_*Wy}x_YHVV3Bz%;)nj6)du8lYDoZGiJMmD(C69!iRX7!V_`N&wEjK`>h|
z1II`bSoY-r1^*^Oj29UvI>!pZ;~3~f88*n^v9;0Rat!5U@4hAwYm&>Njy{;JQ2?Vg
z6H*;Z8v@4SQe?wua@Q?bI~;)>;1P}rmnXPSQ$TKl92z#c#T@#{Qz|NPHv2swH~vKB
z*#3a14?GA{^v{ft87T=c6>9K-sMFF~ZcfgV`0}R1?b9)6EJL>I#d$2P{`s`QL_w{f
zkB`Yt0I?C(huaF{41GT1U>g-9qqVUq?O*cu(CbU*LBWJuhU7vw*1Z-(sZX)6^U?F$
zuM24mp0c~7IsV;aVE>kIbRP21mVKjq2uST|DTf2mw1*Y4tv)yJ{zVf5);$gwt9uET
z8dL2XE)wbtjrKT&36F5st%f3+)q%NL!K<Y={GCYUmxrH#pC%|s)ERWP8rXb%w@ttE
z<{XTFeIOuWZLe;zHk})qm;hF*CW8iGe_=u|0WFOO!~(HhcJqzsbjcRfxYe;U{W($*
zyGBrYGgjaTKqk9-fv^4a{H2z`T5%65jD*iI&)`D5xS!Vv?}>kGD`1Hfz<TmPJ@oEh
z;UyRvH3h;BXl|3B!m%YqV(B=vKL(Y-@=F8$SJfVGg1Ri@+Nj{MqCzlWewZypQETE1
znnL(%UA4lSPtVl*xn9gR=}v-<b7NX@l5g3s1$K|)aB1_EgVZZ0EHcB7%*ZyyW$_mC
zjR-6?b3f=C=cK2L4Q$)?Pk8?^CHYyG$(@9yQO8TkYI%FG%SOXb&)sPiE=QSzEI|&5
z`PMjZIk7W$Q7K>&W1z3FSfzrNH6f~8sI&iv?5z6R)q60F904)S>>11KGbWE%Ac&uC
zfUvMP$LO59pdd`MQ|s?D$zEZ&q*eX+YM`g>9338hL|`;dzLPzYDKOXVx=s!}E#X3=
zS4O}P;1BLl>vguTEEuaGogg8N^UXg^{*d>xgakHwlW`J1=;a{j3f*lU7|y&NJ*J3M
zg_Os}_}|0${9zT)@wM{z-+F<$%I3oZ!gDnqWGTJtP>Q}|n63W>R<Z%?#nWFH>+iIU
zJRa;J&U}xC5CHI|8G0VWzx~X<f@)C<%Q$Wkg+kN_@$J!L9O7$qWZ})7lU9lLIG)Im
z0<NaRuP!3nZ2Tw6Gw64P2vfmbchT29B(liFY?Z@LV&v>J=a?MUlS~3UiqgoOU(X@X
zLYi-<IMUqFG}|kISsDE2we{F1zp<w{->a*+uGn&Ht23cyU>SLRc6_;kOx3>sZtStR
zF!)`3n#6mzDK^~Cy+N!*t&XA+`*6Mu+iFV?<64J(L4Tf$p6O9zsiWV_orn`tAdz{1
z!wlj(S{<&Eao|NuP>}{t19@9RegHziVbX2lS%2P!-uKm}iyhk8SpB43+~0R&RuZ?*
zi|vh1?`^3`!EvA^9HgE5uSS!847Ir4PY6oYHVJb$v%l-}#5mQ1tz|d(N+AS?%v$em
zJ*~%&!RHIti|s)P<r^S*^8o?uYqd@&T-tpAPAiq8?}{}+VjB|=lO>o;SByJ4SP5U+
zjoi2F@(ywx|8eoKz{KzmSm5U#IFGE1_!D9osgb%76BeflaDyafA7IL0^4|D62;v=?
z$IVzCE*U$_Gs4)Rjv!g}fi)fBD8<rA^6@@0l&6+?-6-~TwU2hFSmjp0Cok<xMC>TI
z+o$yA4evl^P9^yY#u<juVHQq@V3;R?d_1Q&c7EKpHUNv^zA<o=B{D9H*m)V$bD+?g
zBltQ!t&Mc_2#tiNdHV7uDBp56y<~fINJ)V|2#~FrmsGN_JdF4)l~A}%sw|8>Q3Z_Y
zyAniuG>q<oY`B-*T!|r^r>9=ecAn6bR3E|}aPa#m#ow27@hQ<16v4Qwb74~3CQUJf
zpxU(&S#mr3?%Kij9Q%ClJOiW<u0&M=Yx~Yh9|K9pV9;$N2-hhaO#AIP@_sGO<RZ@2
za_eYz*9>nZZ8ewA%VAPvBG*TWo+3Y{`ucbJ6#ui-P;&WM2%SiK_N@nn)bAr$Chy8~
zzQ)C{qg5>o)9aENJTd}A%=-$_WV>lvx6JWx(E@KVx~~-D50w=Ae^Dn5c0DTSMk+mG
z-5_Y!tz2_K3S1|Y%!aI5&lf}}`#lGBtbaG6s38joh3;>Ohl>3Hx$3${z02HrXv|1f
z*oeplWO-Q6>!o(W?XcU0(<Ss@eOSX;4VALdx3a5WLL+N3S{q1mZS4?v53=Z8anl=_
zR3)UxVN~^XpX5xlGb}cS@R5$28ut&qHE3Ay_WQ?-Amt?fU-~?f`qs0qn5lX*x8r8Z
zkp&!CdgSoDp{QSi14)N=Bu{D1!8KeHG9&&dv*rhAII-p~7;>q^C)~><obt@_>~>fW
zQ2v&6D=z2ELm=-a(X0LmD*6Z^&dcw{fg^t#aP((oDX1JJ=o3L8Y&igG)a5Xj#`Rjg
z|Gnupe~6vt1Y9wSl)EyW<H>-<`3JfQ9;@5(j^0j!Dh#Nefy9J9jJ(>T8tcNeCZ;h}
zUN!lpfFh3m46;h$C!OXlo69F`MyHb3hp)}^aqwRN`qJT=+%h1EKafD0%d6x_?DiSP
z(lMS`&$lAV)f2GuR!vzmW{8w%yc=}926Mt;i=sArp3?5EhdY01e(}e}y@9;>87p7o
zp#Yc5;8I7c;eiWV-JLbu`?5xV9UUrtWQydzwLXoh!T9Xnh$xEzS^C@Q&cqUQ%qgi~
z;yErye^#2Vuy|C+9NWBC2;!_9+Yj>0uvA<ESVP>S{9M|0$;y8si(k~SMni<_sUJ$r
zhQwdGR`#blS3&WK(rzwxw;}vJ%_wdp?~T63#>T!LrkUsKqhLoZzf2Qw`$}uG|A7-{
zP2l+cI$tbFpa%}2xHc+#uV}N_KKzXxW~}PEvc;2m!O(_B$JkH7<<UA<GpRq4GvLM+
zBaj-AL(z2yLogGOO8j~c8)-_#^9uPSuINM+I+O_yRZgvAr*uTwD!}~M94jyo5&*{i
zkfuBnK^us>tQZ=?r^J=7{dx3H&T7YKh$HsI0}pX+2S=2Cr)8o2oEHw*o=(vCyCS^x
z>>qB$o?lC>N3Y(l8%07GcBq0ILy|b>zbVo*z2trW{o(9sjG{c8{2;YGGPL-leY$08
z$9rsoa9L>GkUVu4?FB;@Y1OM-iX$zZxb07B#;2_x+jn2h@e}(Mp;ypdyledXO5ogP
zoKlW6p?q+5GP%}eaM0!M>@#~5Gs0=oCdDp*Zt%_X+7#Yg9gY4c_n$~W%4V`YSPr>F
zeJpENMsh)pA>Zx{SNt%JSup%XHLhPcj59?n-<m=csARM77mn>(mzXh@vJGS|^*v-w
z933lfw)-ewy`6LDxJGyEp{jO6q&0@_j3(jqB=ddH+x;xG5xv4xmU_o$`)mWzU`WoJ
zoax@X-C}rGx!}0}p-RwcvD3Ec&Ms)~i%WC3=o4H~)E+Am9G0^3fxsC^dgE(!9I+vG
zwxtuFvEKxVqRVXPV#TSi4k?c<KWEZ`5oXxl>4AEgW|}+S!50ReB)^?+&5Nr!?_TD=
zE97;PtqY30r(5tS-$b0}5#4e5yCj9d-l>nz)|SsiP}QR*kE!sR(JH@mJ>8%jTJZL?
z<&R`;s(PM3&`_K85)w{{qGryXFV&6ew||ZMg`9|@y1Nj#OjX&6PlyQPb+`;GjDrpy
zr%R;BM|3+3y~unWy&c#9$xG<VU@=I-*!FjUP0HpiJNm!MqV|Pxdu@-AzT0_+o?e_X
zS1Q66(EKhR$f1z+o#)s_hT--1*iZHp4f$?CGQAwA(VfN$=e7A|EOhf=BHaI@8@uo`
z)MviZfK0phW!7{P;VZ#m9(`LZcrDXS&Z)XOKGqb!E7R49j<3b#Bhx52T^pyrm>2zn
zRZDl9HZgv1TPr1Gh&?Gs>XVI4o(2~UyGBq@;DDJiz=ZMhEu$bA3x3z{Z)9$1UtQmP
zl=QgPmH0~<{!U+xev6J)Ivf_J_V2}z{+*U^x?%R0HOW|Uu?#)_4q>~p64MiX-HUW8
z)c6=Lk+o9<)}u<!(ZHV^!wHAt0LTROabCsjstngKP!65)&A^Bop;z%HY%qB2TvGPO
zW0lxE6=LsTo_unK^pLt^pU=2}bTf<5Zbs4K4cZ6`7K7h}Q*C$24P+RTOu-U@Aj`_4
z|0Aax?J0_;a+mn;cf?iHfnQm3n(^oSI95HKdC$+*T36y%qegvsepsr|OEfJUr|#z@
zt)?UTpTR$%{nR7cqw~hM_rLN}z$4~$+`44lwngt8=-ifznP+=GP4+tXi`b9b|M@9T
zsmldyV{Ev_uf%miO)tkbj!l6sWKx#|e{H|is&DV}bLg5;$k22uj|Q8m0D`AjO1ckI
z^<fTl75~21DI9LNbQT?i9Z{~VurBoRyKtPM{fBz(_zAfIQ$@<*s*Qz7m(@uAtK9Sr
zjpt<>Y?fVau@0B4!T4MSAn>?#8nRI95a#D~p{L!z0m=3Iy}HUF{+kR+68N35I;xOS
z@>msR<aWi?#gTtwpqzG*D^3$JG=qNM9PVP6lD0drc1C>m8VI8KjFT2?F^Zn-OT`)x
z5`Rd&QWZzfk&r;j=a9;KF){dXNi6>FJjt+6Tw3rbwrWZu+t`qv;VN9^wC11eTLQwH
zPX0<!YVgb_Mp4G;dYupkb^Lki-5$b~L3AE5$beArIc#qRS8_<D(*Q1pw13ez^jV)z
zo5X}gSds2eY5wY=422hkSvp+NZvOp;cdCAa#*$_i$OYgIbeNV}uM{1fzO*eNjH90$
zU<M09pD9@y_x0zu4`Obg<a!vG9nsmTD6^iAR6mr0E8@ZlUeM!X7FS=8qp!x!iGCBR
zev6FZD#(}g?lnDny9$8Y^82XlI)8oiX?}KE3aA=*5w&<%#JmSFL_Q>6<SY2Tl6Dz8
z75^3g_ccG$hrnUhk#r7aI$}64{h)OD=6!$f&Qh-HsHJ4hhut6kwW4LVrqzAbtxDnf
zB{HIk%r@tVCNmjx-1Z;|L!DG_91ao)ON}wIAP|8#S{rX`DYdPm-2Vw*N1<Uxt~20C
z!vrC0LM3KgVYDUDEbe<~RkVFdBr6Kf!JITvZbf2^h9=kS@q>v(#JUs_*}l8iXM+*S
zp%@$V(wzNuNdG~>zaJyYh6`QIARZE~E-it63A5A`JjwVUf&#<!wfV^twmlPE`H7p{
z$Be9DIdC@K?mCG0AYzd@Iiz<qq5fX~SBZ;v;l?(<mYC`6smwnYZY{-OHa>ehA^h(_
zKwUtOP4Fef#NmIK);jc4P_c=uv5l!@-?}`+xH{~)i?H+)zOuOS(7l>=EjpG38m*{Q
zBRP{$r(dz>Nuqe{%(wP3MPcR{1$2!F2Q6pVzjtgE3utIOIu#TcQDh(})5o4>K>kzd
zx~KbU<jFyU#2&J_4WI`pChV8FH=_MU$n9tD?FtB9Ov8GaX_EVnV;A;ClSNa$TQcyZ
zKJ42&^Y2ziOD=mh&M$^cQ*5T`-?W@&lyLIhlQ6(1W%w;#;P%;U_M1r45YG<&Qw{0D
zmH!JC!3LjWCOfAWz_R4`b7JCDiAZ>CQln%wL7D_v&`@0UFY&j)6EMJB_bcAy5CG5U
zdkGQ!ulDO#urnl6$d!=Eb)Um%PP_0`&WU~=$lODDInX69u5Yn1YH|aK+DNH$jt$Sx
zW!_HxbV|(svVdjL3-T+dqX`naOqVOC-(Xgkcc7N8?cQgt^527@1Rv<4SJ|2Y1rhWO
z>>X=>=Su)k&j2cpAROe$B`<Y`48|fPR+F?0<G`q^23mLM6_T0a^73d9On<@N_6(%w
zryzTB;SLrkuRu$TP1z3kX9D2ae$;|3Q{!K^r|095FbxpRu0304|Is!Llx+2bsKy|g
zrbVygD3HM+A$)i3q{q`CDU~Mz%^Qdd61`9#S8s>0ib%Qk8JOmcKZ$O=mE8JOUdf$A
zBfjBnve7s8Dp@{JdS~eS&9Y(HOfu6><d$UtyV$?^ZA|>0^(R$qG-T9;EHI<m=!S#&
znG&)9B-^Xt;U$8Ivgrp~fZiHQMu?bbXe0vTJh_Gp4n!Kx@rParX`@^6{|2%N*9tQ1
z;z3DtKQdP!1gqj5SklY}66sIROLVFr@y~%5z<RsC)4ARM!O?=}K|i3xHaWLS0BTm7
zGXF)5nh)SsLGOS!s49_3M;-7tGd6?oaBhKCpq8>{owdKe|7cRrz8@6IJ_Wpg<>x`L
zQP;tGk#-XBFlm7l^AIr%V!$R3vZF+G@cF9q!t5Fyv~;1pyaYQT=caW9>HwGa;`L4v
zC6yW0X4XsKH2tQ7^>dXAM}+el{{^RF^@{|1j&UYDnl32@ACGS;5y4MU<SaeSmL@f3
zOf==9<wKLL^2O);sUEBK2(2NFFHSJRXr0~Dfu1O`e5;>DuI&w4WrmaaW?;!_rGhQ0
z(zUhv1&>8r!6Y`#H&q||T~qn;;23iQ_tymwt1@}ymiG0@X1FJ>z`m{xrXVq&e9(6b
z($L>DfC$rkw}XYrGvpdf)Mo_zmQ@sJa&HkIQuA_x0_)eLb8>TOd3bo{a*$aXtFfob
z0()G10L;t>Cv5F5(&q*+uFqb8z^??rmrk{XTCMFl<*zq%pFHb!6TRgcbvI*_zId}z
zkkS9dUlzC3ei#+v%l+J$U;2qiTNrKCIqhxv>i{gn+}ZMg-~6lhhm`A{ejW@T?vKwI
zSH=lTA%wny_~=$BC{tAF+kLE8-oNx{`EWKP@3B|*JEPDZv>UY!&Q)7dBXPss>55MO
z+TOg69svFgS^>h2x&w9AbnwO1{MkR4t2+>5N8#Y)T&tNftl9u5ua`ojzClSaa7mjI
zIRLcA5RtL~R54r13c9-KF8D@qDSQuBoQ{46;LJK80bJL^^oIUy2o`)&4HXSDw!S`K
zkpw9PK>BMO=WOeAQb^YMEXZ2?=?=kzuXUna7Z1TB90e*2>(U}*dlFJXWkbMJRvz96
zs7{DiW?n(TBW+be!HQ)^4(Cl(>wF;4R+-@=)+<{haDZFHG~x)I#O8he+*Bl0`_f<A
zbj-RpT=Wm@HFv3TcX)04`T02$R#Ql!K+NO_W2zN+%aMwC>sR;qN-Gb?q~PASH|}#o
zgTkarp$OKpAPl0*qGT3>+5M+<bhFH?#pK=1<)?Pj2uqyz43;Y1h{BkdgzBlb1aGU6
zWNI5JZw{WKydR2!=RvQ2tQ;MDOt~nGVi@e(QD#f<zoCn_Rvb&mKvPOCrK>^BlcBYT
z`p#VDY3MWwy$qOM#o}kCK!Xm=rZV8D%)rOZ#rm(SKBk7^9gXKu9R2taCDw%E-+btE
z061zSN{JNsUU1oH-$a7C?<Z}ympi3xB5SyWuUH#5sYrm;^cPTIPk;y@CKD@Q^#Le}
zNG*$Q6W{|L(O@8p^_U&ff!bN$=VHxqh`qNDoOY(8Vs=F@i_JICN1`6=J(-m01IHjE
zvLcMo4@gfH{Ma~>JL8c3L5;b;Cy-f~S_UpJlh?;Yb9AoT?yf`>OaWz|67<KM{<@7d
zk^;na<7>XC-z|@gS|ea#2M}nD<oHiB?4Y?JaBjgtW<?b-FG@fA*E-<)u4V4Qkdu6S
z(aw}yi)e=p=SK5y!#nmb=}SuSx;n&luhbxqV17uMvejSr)W2KbpGMGkUtb)rfh?xl
zVh}T#ga;+UDFvn9=&P`aUN=qRLz&g~q|84s6PJh(^7MXSh7f8pZs6}B&ZK8zC6Z+#
zRn-(`50GMGG~<r87I*)%LF4)SAn}KcNP>62W!KD8r@GynHvy(Tp5<lY%6pp)9WwXc
zx00?ci)WdVL_g!kOuIQfjphj*Hmo`?7Z7^E;_g&=UHvvU>YeWSX3+bXtu#*?SVm@H
zh&JTWox)GXXVOr?KXZ*X!y8yKiDAXA53s^OSNTjmYOGKwRTnl#-Mhb5WN&_h`KqE9
z@w8cmf*ny$ehAqHCNA^^5NNYThLp7%^}}_=7ltiR!fe?bS7r5r`5q7oYu~T27#Gia
zf9vs4I<k?8DtFJM)nS1Cvn<F|Tm{U0or_LjFwF$r3XKm6Mt2Cc#j~`GPf;{kC~Dsu
z49i51zJ*lyl2nj5;`N|gb{3&e{6C_;GAydDZCem!kQzd|JEU8>yOBn^q@=rt?ru<0
zq#LB98$<*p1ZhyZQ@%Am_j`Q)MrPQvXRqrz6D|!~khXtOGn&S1jj{`5e^t$xh>>%t
zcENk<Pb-**1$s$h+EhUD3=5sE8I~`C>7+YQ$t|8X#*|{m?Y%GTTO56IC5vK)5K^HG
zhv^?Nzm=k;RbZgQt+W?Vrobd2(@3lZRKy~s=|WF55hf$q*F?#2H57Q~VYYpQveD?V
z2|V1!0=&eWJuI@Ql5C6SOxdVxV#9P-j;SKOv1jaQptLq0<JkF;gz23S?AUu;{(^O$
zg0_&l0gnUoDt5}+O1N}I0y}s|PR`74__kOad)=JMQmGqfbK_~6-!(R}-%Hu>%wIfL
z3?Da05#G`+>;g_wtKnzIdKr%dYOq!c>vp$MFuqXo*GmPB1%%W7K0@KYw?LP(E3RA#
zVyC)9voQH7mmd{FgM(^!{vcIl<3OHefBq!H{4{vuvA_}trJC}4j=ckWVr<Phvc{#Y
zy!;7>m720Sb(Zz0k4g8bP`^Y_>vD^K1_A=K_RKr5$Nso}e2NQ`b{N;nGvQQJ>&S%(
zJo~;66VbKs4Eclh<HYn};M^yy?FT$D6LT7G;|q#=Ucv_Ove|8m9gr-7*<e9B@UX6T
z`7%fOxyR)~0T^wO9wU(5Abm7_lKA`u+0ucy%nf8XL#K9T8%IvpyKlTF>XhEZ661Kz
zlQ~f6uEVOV;{}-C$l5;+Nv6M&rvGsvIKoOa5Bi{2Rkj1{8lm%}590_8!ez*k&w?2m
z8edoV<yJ0albm3wcoMGwGbmrmbYNvVP{U{6#V4ICr#|O&n8zdZleQyXqAUK49@>-2
zzVveTPIT+94u^Q~XP{Pw$Re3hO(X~)uiAHT)L@1m#SfKDu1Gb4a3Jl9%}8Q@Q>tf_
zt`ZiiEpPvv`>qqXLgY%jTU`>#=vyONrER%wY^FL`sjEp0M!kpnyDI*$WMue9vRYNC
z82Ykaa+;sVXGB)nh@Mg{b7)T8=7rhUOi^y@*$;jD7S+-LdKLDIyPKi=POpcy9k>W?
zd#>G|n$s+VgA2DojL<7!&MWchncsP-YJns9O=f_X4#ag4!2-wl>17XV7x*{@w5^}d
zq*=$5@90_QrJ3iM<iF4>%9Z1#;aGz`t|;u;Q(a^tK$4R(Zujd*)S&uCJM%)saOqPi
z+V9y>Z2iG5uRhi``yX2Bt|NCV!gN!_vAoVz$H!>g<)eXr^floIIi+!>U474Pa(|D`
z*%vs^x$Pc@e5w8L`@XwZ=n)I4Mf^8)l_OP}52U>khoGJNm&!Z^-@%wXeV)brE-*dQ
zdk#CIl=dQWpD#(U(xC0<HJnfep^9(T-W^Au`k#@F?RQ-+Py`|^oQ?w4?0{>2gMUFq
z$((eg@NXm*wco2u=FLXA1Ln*zQf?r3Eo78FYU!uPX~~mB$(F(UVwC%YJ10^34jQ8&
z@qliax?{5f638inwlKIfwf88jppvlMaDMIWSGvMy0XF0~ak((g<(GTpA5~QLH_r`6
z%s(n@v{gyPs%Fwg+h%DLn%PFR-Sv0ymTbllAVg{_ucr{;5UTf63%Ffo%J5btQ5+gC
z&TV;cOJ+>a*-QX^;%G;cpv1Dp#TOl2Zd!_{-9e>8W1L-XM%vo)*VC~ozORrl;y#D&
zGv)Y|)Aoz1=!Xli|AOt!$di7W4OKK*Mw#zfNeFVfIyXIugi?}lBF&36cmiHq5zx|m
z0(THfDv%`FLKg>p-lzvzkPNAVt~_!u0m*v~0^CjzR{Bis!p?qKOzhuE{I&_+kp0;z
zgYTc-sl{vILe?p!>Tn?RM6J0<8jyL{(`RxFG}Bq7Yk`pl&m3^v5~N+LNrPJpdBR|;
zlf692TtvqGxv&ePTA@)>yl1;^y?5cmhp};X5lT2%cp1cIbnFi<;ukwJ#1lVfOIVit
zj?Rl;xVPaQNj)}9)#$B#bLxXi9vpWGBvJUksQhHAsX<FXaIct&u}tY+R+piIHZwhK
z0xuoSIQOewJOnC3?0y;&ni(Y7Pg|XY^$Qz9T&_=Xm~9I>-a5Mq)V+REVTu-E2btxN
zAx!CfW4Zp#jZe$z3l=9U-3hBStevKGB$3tyL69X&Fe7JQS+qJWO`b5FAst9y*zydG
z-lPft!pz9wm8e3?<mXZu47bAmdI?Htafj9JZ<!NSqHT+=awIPFbl;jCI7H>XLaZ)!
z{NMJBgQOrOtzZ%H)Ra@7bn89zgF34=FHK2~&3lkcd(IK#|FD3wF+H_CQcRn}lGXOm
z(B^XN%(F3BjZQluHTP<S9FZ{Vw$jPW45(3JM4=?qfmOCF((jI-UZ;&j^~grC*Z9sb
zq&lkFQ4nKq-vmQ4%M$XgM`zbj``|NTBYysVAM{j#BIf@O3Pe(>mDo3Cw$j#5WICD!
z5RLX;<#)G+0TEx5(D|$cBN)=jv&486c@L7`(|IXpbL41^h$6ywn2e~ux+V6p94ne^
zS$EXsJg7LqkbHyp_dBCFB$;_2T|6fp7wn?*GGv}x<=$_GH8Zua;5pHF6EtOcBOl6a
zzIYrDojv_hWZIZkyu+WX^;(?R1dF$ru2dS*<pEE`0}F7XFH*7wr^RXYk(>gi_qFQ)
zUYp%GNltUHysUT8wNHg7ak=3jOjfq&U9uBWP>sZ{jeI1klV+#i<9E+dk%B|mIam^d
z-xNiQB|;$<W*^6##1VCK{?DANi!@5asrD181_gP9{cj}88`{y97I)lnDvqQDN60~2
z%>!%99K9a-z}wb%HB#zlSyeJc9WILXTgqs&ImPQ5;0~;XU(O%PFG>F=QyIVv)=d$t
zv@*+e-bTUnWx2x^-CudFKdbcliTqk2bulpBjC`=IkXSb4B?&OHG1DQwHRm4tpNAPF
z0K!YM-@g5(4U}AB0A_!Nf-uf!N0?0T`)?#hd@>#Tz~}s`jjy~s{tMMTDd{7wICx|n
z;>=4;c(c5WJd$(vigo22(QnWeYW^WUvd@5?DIry`{QG<O#T4U^2)Z+C8p|=vF~xok
zEojSW|CW&>_`Tti>frsv(z(m&2fgHv)5@5)88IS+4a>8Op!(0{)9f6XN72InaKHgv
z07eX)4gRqN_9p;X9YB74*QxO9kdh4_s)+4~1Wo{!zM41Y+)=(&<ZHP+izT;-jLFH>
zBfmnAeWfZoeoDPGvh>60F@R&my_g6gK1n$Vn70Bb+YPsB^eimOAUrT#(_I+<QSgyh
z#Z!U~GHrNv3H(~T$^6V?Y1T)dZ5hgBxe=2i_=AJ%az3w1b}|r*rJX(oKWwBV?{y-a
z)|l2om3r9?UHDDs*W>4;h|r2^qjtQXkLq{5fum5PNCrJBGP0KZ!g@S2NqOWx0xmQG
z({+8p)s<HcUZ2rSZy0I`KqwLZ{Eqqzmu$tx<^OlbKyZ+ll20g<us=w?#-7ma?rI7p
z3Doi)VF%rQz)bm<+IsW@VFgS~h^a^1#9BWb#Vn=Js-q`Z(_jP{F^2CjDPhZM7`b7_
z@+MpO+UEVQk0xQ_g|}+S<Ks$%fbCUf{?X!OEU@1=gNTtN6b=M?wFRlExD`Ad@MdrC
zYBnY&KRnFGuPv~0zqF`f1yzF%(tjWk5_#|mU7l21>&`DeLopUJ#OrXB*RFbt`n+sr
zhFY+W;~5QauRFhq6at#@W99zb%Wl$gIu@Di8@i_`Iu3N6kPabQ<?4H?S2E)Y8JE`9
z!~UYBqZ2E|1Scu&H*GBEa0*)MCD1yh0U47K5w{J+?(S{VHi$XS0L!k%mqzhS3d2lC
zI$BzJc&=v~kY>pND8k>p<6eRSu5Kj+%<Syffx*F1P<OD!TyH%^1{cJ@*5N(~uuR`f
z)}5|KkSl->$YBNS0!<d+pTOZmiD_TB%d|B9HHIkAPSt~aofXdq(gjw{(%5_UYOu{K
zf;>pS0fqDzjMTF6jpkx~SHN<1HMy%|0&vQwgECY9>Dyq0h*2el#R~IB@=^t6h~Jic
z^!>ee*S6AeWwL(>in1<IWS|$|oyl8d{v3UWfnV54O{=Nxs4^#4CN8Z_j0*egsmNs+
za@{XY#t__o);r<5nSTf}u7$=s!Oo&ip>8uJ=kphk&ZF>*8J@S?aU)ykd`lFlcrR)#
zzqA4kl)nb>-NP%ZrDSEtfG(904mkb1B%cH~r^Q--6yo(#mw&q-XxMN8d%?}{2HauM
zU_F%%B90&nqT}Ll*UKW>MlDEgfa9Wy3e#Ia^)}iaL(&a)*D(zcaYHiJn$q&qBL^Ot
za!~QJ+OSQ7o#$t>L(X`Cs4;v9udp1~kn1LaiJ_PWl5DughTcGaBLvt8`<&HQ5Q;nK
z691?0<~M`NaarZqwb188zF$pZOZvs6$aj!R9GhkPDQa*eM_}4U%9xL6@a1!bav6@{
z?r{w|a!cBVL?~HzJcV%wFFrIF218PG>v)n$Ckc;(q_HV-VJ^F0IS7(Q>AJg5JKnlm
zeLzfr;~zk4umOig?%%%;Jp;A~1t2AR$3raa)0peI?0s<u+~^x0isI|Hr)k!@k=vwN
zwHsq!gTuvX1Fo&!m?14g>reW$H-mv-3LVykuWWi!r(OVj8UtFty*41@{?P?)5+;!F
zaG|Rb2<jzY*x+uDXHkRoe60M#yAi3-T^3YlX~EyK`YtTs$5XfSeOZStqQ~U%p0@@9
zo#UBNeOno=%Y)+OtC|1IU%+D4h!x49c-}OY@hrzM>WY@ksJt<@v##?Sw*Z;4p~!rG
zGG*s-Hl1d=!`v?(@xN_Y1UoAe?GF7j><09hoY)J7;uMLwb*fiL$yzv#Ubg`%cpv1d
z6WENq4C93_bd9|OvBwUHB{qsdD6R<Sc6=L)1!ju!xHRRM4Vu89XW)CI;U|dRpb)<4
z-Vi9@U3yXIf2`ZhRn{R*1aJg0X?-y9B+YUgYFI4X4<#ww6v7oK{tHJyLKm+xssNaf
zzQxgG1dzFO^P5fhSA{G#itN9K96utI;yXnfcGfRgOlnkAYQjx($`;SqJf*BmYHIt?
zcsgcj6<SC&rJ))_uJcI+h4tFA6m0XBQbI<OFER3L$La#2u3sp!e~pP_De=;)=I_~!
z75E+BEdnv=p%7939;ARd$O^hoO;@xzx*v$q=K6OL2qqRKfrO@ED<|<2s64qD4Ap+d
zl8-ak<MqXUizYJ+uJugr1KG`+THibmK>tq~5L#Y!wdaAxoKHsW(tRk$S|fzua1vNA
zvi^^l>m@*-qh6K4<^tgj;)9KiPD4es6ITo}gvtE&UtQY#&sIXLv}ms&aL-R#I0GH=
zL}4@)6b-5R4Hm!2+;r5P=wz1EmIoT>+)0NQCO6v7Sb+EF0)PPP@=>nDU+gM!2+m_H
zEv+@_Xg4VgD<g(`CiT1L%>9_o@c2Y4K+$g0flhF`Sdv!Y>@5M_F@fw1&UA1`kHW1e
zJn;aguOkv2r$_>rt}zfPT3r4t0L$)TcIL~862VElQw={HETu5uGjkB6^A&qQlKQ;<
zlDIZD|39_@9HMv#cefN#cFiT(8@spkgzp152+GbKjdk8+Ig7G8=JRmmwY|yV5XfV7
zI(tJ~5?88D+X^w+Y7dZ&mgmBz+*Fiaba?CgWiA>A@q2T0a=9I+<v_E)Rg(2XT22!=
zNA<voRz@z5jdu^q%CM@)b-xVefKX}>r*by;vF$E}_SIsZ*IHD_MdBka;1p}zM|A!6
zUNYj}@Y#)X^z=cx#sp8nsOBh7C=6n81T>MU=J5i9Z;%NeOF?%5l!lHWsr4$7=x-!o
zB{GEZcQEybK}N-BCWraPdV*;4K%~-3{~NbIs!`evAoyADqa>8F=r56KVa{vkf#>OO
zvFn~2j~%ZMZGRA-`1+RPZ=v{`=;Mj+ivJUYw^G*n7(|Z<5RlO%`S$&O$gXPN9K^I<
zV94F7kRrHk^sI5L=&^ftG-TIW*gqsCH}I9?_k&&#e}w)_51Ph$_5OYLjOEF1VcsLH
z`xDyD$&0(vrTM;<HPlGGaxP-pl7T$WWn1d3uRb7DIZiFxKyyMeF7mGzYdHxa1nHSZ
zkz%6k;QegH|1nEX5Sx@95nz|45VMiMMs(x)y$C&L<ay5rO7m=GP(P_w`&owz1*&t|
z>eLjn<@)m*s<y@^T{}f{@ELXT=z;8*5|{Z*3PPlCRK%B)CX`>=EgR1q2`lF?9FePZ
zfq1C|a3FI|iN(hN%@=GnNsLDdC0SdK<elkeX-vLixgp2=`lTsLhyuvGF->gM|I^SP
z$-XP)STwX7SLRaEJ$s8aZXw#$l_w-<nxvH3K6yjC%;gbp__Z{BY35v)vY=3*ha%A)
z@kixSG9wc7qX~C3Q?wWZr5I5R7!Ixxr^RZ&5WVcV0|(T?ft{Q^WdyKu;+#<Vgf$Yv
zV>YJzW_?vIQSreSH7us(;=hstwSI^R$ty07xea+GtP|Ssm^xgV;}J*Y5|wK&!jmzO
zg5dFnd_vc<W?c^eSd9Eg>;mxQ_&x4|6HrATH@mCZ^S+8@`+4xm<``<hUVjcEb^EEb
zbF5Hoc0?^h;*@`Zcw@+<f0@1pm?Qs<lMHXbJjs7r*W>n-35Ke`lPwYQ9gQ>i&5<;W
zM#ZR0dXk)5FIj^t)>fMmtuWIAp)s6(G~LLSm*?rF#Q(g|)L7tekX<vHwF6rKd_o<5
zjyQ`hAL~i^)b&--4(e*?kB=UOvF&x8%DxEB^R~mLe3IA?^R*Q_$hv4UK~6iLNXzR-
zuNYiJwf_Bg{$vo3;6?LOxfAcd4@QVujh0edf$*XJZO&FkR9t<JUG}UTRD+ws|KU*e
z`mBbB$eU}bJLK+F5Qzjf-=V^k@nd~>hXK466zma@Wn9wa|Cry^f7XZ*76MbrX?(}2
zi0=lu(*YP=Yy1LL>d$eq+(qR{u4Y6zExvN{@CggqoN5uHA4IdXD{sskMaJip#mw}A
zF;i_1yhYtT$F|qC|NUAIh!!gjvt=*)mE2ao|H+QG@<yr3`HP-7?yjh!(wbxg*$Iv0
z%%5?r@5~vTztfix%*>?L&dA_HX%_4WC~Ltq2WLMdeEPGp=+sgPtu-`~6Pfx@+Q{))
z^f$VG^FrX$H)7-h?0>KKIVo6-wn$znHUXnOrl-I0JBqu-2Vj5g#=Q2O@Yn@zMGKgE
z@m%d=ksO3Ft34iD4v(?;)L8q;-;~l6+*m)rAf(KJm0g5G*=-KoNU#r;a3QZk8k4-w
zO~A|nm^NhLad>K^KDzOWC}z6Lyrhl){iWiRYOEp(*lUHolS{bCKU#f4{Ja&B8I9kF
zjtuZs|FP>Hv6sXvqB=`6R_i8w@R+7^oX=zZqfC+0Nz^|`o^N*UKNx~Wc<%qv@5te!
z(jBwWmn-~KvBLTkaDI*EKk_!_b6U2qNGui+oNu&ENr=2IC^M>oc8oB`4*8C^tG0=)
zpEs{w<Z#<k)pTBD-+|)--gs9b?dFZEn0Yke2YujZyAg2k++Uz2B-F!qqZpqb0T0qj
zJg#<E63M>Z$cn%$7K3JzZwY}}YS-0HF=}l%FEReR(-OrWg*;IofB6$7YxM~Uga(Uc
zM<<+ym%{?frMGx*;qq^)bHm)5qwJvJCl5$7v$!eG%m>mf=JV1C?=Tt%5@k#hoMlP*
zU+N7iD`Y%*I7k{`z8qE!v>4_(ER%D2^S0mxJob`S4!nM9Yb$@2XkgcX6%bislasS0
zR_yH9!0Bgy7z-_}bT+R8(-4VkEDasqC}4IQeq94fPEJB1q-yK}WHq18hG~HfQ5(Lo
z!Iu%>Fp&d%&K9+;W|>OUeKX()%*@R#kAC$uJIEQry8r<Fr_t%-4o^3{CyR`2Lr~6S
z7l&^P8AcviZ<V2nY>XCx;Mv)%TwU(A`xUWnLf+$W*<sP^V&1w0C}Hsm;Kv!rR<d$c
zNY<$bjnQ3n?6$(dN>BpMqYHi2m%B9N^S59ceE|<X#d8rF?C&2&W$XA=(GIxJ@>dn<
z+%{i6=eL8_EWd3DcR(3@SSBRR`0ut#0>Sl_SE9;>Nf<Qbd1ZtRUt6l*P-iq%S=y}|
zT(5b?SbQ#XJ(<qd79c?yUwXxRpp?R6Fw1xR_AOe=C!7bRjETj|EWZ+DnMo%~Ve^8H
z(^-Uvw#U||Etdf698FObvE_jQ9YxB<hK1siP1Wm8pY$VsmS2D>V@u@Q$oD;{v#<tx
z8lw0uxbOF!|BB^PaKTUF*<o?I2sA<}xHa4oh<&LNH6p1twn1h?EmBkpAhXioc~S*p
zlhYTZAix{mW#kT;vgiT4lRJy#QVodUlbb&^i>|?c$Mc03v6+V*0rByIaUUG{HT>C9
zgK&}d!f_Yin&IQr?WA?fI(K}cb+tLrBDug5@ZfRxpxCI%f`ICeAL%Vn-pVtX%dAJ1
z+W}Lm!^|0G{1Wnleuq~b8Ug!fSqcz+Z?EV$O#zCFbiksBx%wJb?-kJ8L%Ic2-<v=-
zHgy7Dt6zMmE`yI+`k+H<h&-@E2=s{@)cy~fgFiSE#okqN@KkBN<lW<r3d%6mA^=7l
zwuFL}ei?oiTdT5ewx3k~%Q>b~Pjn)%uv?gmk&{g6!kOcElOoHQWJM4~|6=i<j|Ap~
z;_u@XUV`pKgA>G$Dh(IlH4XMjU7e6{z6L)=)^{1a1Q`IDBG(%@9#v>W+{63&*il<f
zpjr;ztT+b94<<l7rtL?Fae#`nG{B-=k4g?xh(I=w9kMFME^$=&dt{-ElHT-uKiyJ3
z171sR0q~v{leq%v9fjgjb|~eN1UN)0+@8Zg0dIQ!)Z$#t1#Fg!RA8HFG&ub61zRPD
zcY;i|0b&)eD?vc{#n#rg>!!)>d-H7~Fdvc7fln(Lgdt4luTR0%(<8-)Vvqw|0&$Bp
zQ?RRg(1xZ#E}aP;oz3bPJ2=sai2w<Oo>p6R?*G4(xzM9ayODDHfXTG@xI`eYbfJl=
z8?3C5ih5k5MQ2VD^LGUnc6fYTlG(Xw4QZ<|r=vL=K{lm&jI!ruPfO@*ge=c6K_H#(
zq>`TRq@2>UsNq7Q0%Ihx75S!)f`NFa&36D!8TsEi1SC?D4QY*T!{Wc{oE@!Q-lp47
zh}%5750fiM6sj@#8<{?hvX!sWVL6?++7L;#4R-M~P^7UqxeYV|>zZd?1?`zxFW>wB
z)`<8-EfZV2z^kN(m9GdCB+dg*_KA_$=*!jKhp>sEZ-Q=BdwKfOJcxPE*$m27<JtOf
zj^IJ}0Lgb_c1w3`Lc(2Q;;t~LF}>mA@4J42yIg|e#Jg;n+l6W9mQgQ0^i1Rm%*YUC
z3wgUV8^K+v3uJ=s=Rmc?Xr1p}YE<%hQ0L!l@o$<yM^r*$w>1m-{Lw{RE`(#62G&<<
zbgL6%M6N`qOMBq4FmiJ8YgMB`o3uSybaiaX@+#+XNlzX6W>T4PBB5s4@MF)PR4b}~
zT}9R1(2<;pt?7J)Cq*Wd(l17L*t#<aXG#@-4X$`n{d}X|qiM63`DVCaQ#FmWJga^H
zsiw`P-;FT9(#}?`DYiVtonmnY%m%;r>s@p{=IRTo|EBL(Z2YdxYUg~Pj*O&)hfQpY
zx6=P+bB%`e_h%S4=7h~OREC0X;YGF7)7N@Y+?Xkx$31P2#)a*oW7GaG3qbu7C}t5-
z422uhtr8O1A~8a)X{HpA$i=pP0~25FeY{fv&bBJQ-__d!?a4e>gGfvV3LAEB4C>4v
zv+!%6<e5M0x$4X0SatLNLf8M}e9H9-P6bEPPpad6jdZfCMZ%8fZ2}LGX8CbF4sZAV
zFN&F+{%d2A(AL4UJdj$)vwEJ6+H{ZT3w&T;5~%2Wz4yMe7Zcmmy$8jkqtfCI2w)Bp
z|Nkv*NGWE~ekWBk_G<lsS!*FtitbE>-TKo*d`|Hu2hT==Z3^wY?iht}+=JH5WtWnE
zqmYWo-(SLQ>*0U)j?w@8ZLoT6b7ctf?s^X~Q{E9YoE$bLrn=uIS&>I^R?Mmhj%-Ap
z(4wFI=yIsA=fY{>v@9w<-27>|LtL`}%4koD-2^X52z@1eZFNgxq2o_?FMEtRx2AZo
zNxe;8)-G)vw;v3JovbFGb6$7vpIkKtZk7{1=r?E?$Uh)GLD-u4gFa{1646XY$$)Qw
zM%;F^vB#qzrryB$u!4v)-0?Zg=@(eJSdnyt(NsV>=2MrRC9G3<zq{JCzFfB-wWhpk
z==<iZQ7E*fIU7)z&C4E|M(|(?@n+h}1>hI$s*<=~q$IHbP+P&p7nZ<#C2}J3>{-64
zu<r#fRe%H@M~z!)c6CCBy|ZkQQ8n_(&^ae-&Gn0geoz)X`DW|rni{Iaj<N<{sum4_
zuJ5P~9$)J&aFRE3z64^?wZ~#G*8to|1%iH`BOb!)NxZ0$yi-x}SsPbvu9L*>B7-(^
z)$)|t<KL@z^{8V+Xg~jNz*&oUl+91a)xig`2y%KBQiD8n1@#JwLiiWAXui>Rl6?#n
zNT|<zWJ<Nk62pj#LyZ3`=tE$2rNQx*VArud#}hO6CHYE9joAlTY(0)@^@eyuFL_nZ
z1;+?9Klzt#zo~wXAGe5*P$vjq))$wQ>1h^o3FcGGD#{KsI%Ja-H@U`DzHXBEtsR*`
zww>=dmcLM4V(1f~)4tN?;L2!JntFEN+?D%nO_!@ngLAXRxY)z2(te@#I8Zq?@$^$J
ztASE2O#Lt}@!HjTEHRjEbi+8XuAFtb?NgQEe06+WC5p1&W(4iYm)16Ha))n3TiM~0
zM$DNRpLUrSujy)5n2MC$hL&{QeXk(JIYAOOI4QAKI-A4d?;O?{Yy*^vi?iNwKNuu%
zFtSR0)MjA(o}V#$cKWtDdz+ky@$M8J@<IsEIRAjJbV>4cUqZ*hz1J;ejo|}KmXIl-
z;b*6=h3VM311Jt_o<GV|$=f!P6)U^Ri?JxPcL0=YSyX~Of_pN|4~5{+R0Aid3z+#A
zc=?xs^p4jBl;~PfI@)Y4o&Las93el^0E&)EL87T!P|5TovI;kXzwl?@o1}Y?f~E~Z
zeL^3x?2C@~5(mV*rW~!%`5b@w#AQ8>ZxpnJZdSJ+U=ObW=-`D%ZqYcfN{&~n+Qftj
z!{o|Akut0kKhdp{9!x`aQ(?agu(zIUTkE=9NrE2b=J86I+IocR^{s{zu><TvtmC{_
zPa>&9(Ak<h;bpkY*p9fB`ut6NAN91EB01dvV)=>c)4!H-5L$=`{WX9>*dLe|=~Jm<
zJ}5bSvgp_jg8JeGaQhSS1(`>shRmsY1;Z{853%TkX_V;-N%q;DKP!Y~aHkXj`;bf~
ziRn#We>lZ6ID(R=bgTd)Ff%&`fm-vtdgq%)8_(&^9|-qVN0PDXEmjc)?u+M6_MNyk
zGM}QCtUJe9{#@XhOXU{uZV@^eN7ZPbRCoEgDZDXxOP+KyxOWy_;VVB?QJrLsjFH~C
zb-$hVidjdFUXp_CQ)7n1VHiDmQp04s7!i*HYmLw?%<p*J>159wb3{}TZ={gtk7aL$
z@o3u5sf@OV>WuuKlZ|vWqAGc`%73(zG~4q*B{YXpHu)>%>Vrk&1A?R4eI{NeiLU(9
z0-@_!3D?doQv0u_B(LgH(e~yw4)Qy+JW8xdafpVveru)v$<dyg#-H&e{Zzn*lo^*6
z9hC}HvC12YC2QW2Z2TozDTU`J&!&*Js-YxXAmD2AVLvG=hHFAdr&zoSV(V@n?CrVf
zZgb6=;6yGGpD58xpYK(eS0G99suvZZ(n&YhUEGJD(;GRRA<>wza!J79Wv1@HXYDI^
z<<!H-73MaeUBVBcV;~Uv6O)G`|1vzl$y!vIh8hA;K)B`|#QTLkN~iC5tV<jI1<Uym
zL~7RKqd3!8;~`<qkkC<e*g@V7h7=oI(=Cumus`9;TECanlcN(yp-6)uL4!>`)k8Iq
zut!FDyV#705VzPeE#{PBGiD7^(A!>$m1fRKvL}=WR^C}cmYmV!)5^$*D5yp6N2}?K
zXAWh`^{b+VTn<S&{Z9^?D+3n~d<%#jc}fc{pM-dw8ePf@tgAj2A7}4+@p|=smu;GZ
z)?pFLil<<wWQ!%IMDF9rl&AhW)7Q}}{MUak+*WUyTv_nDIP~IeUkfPeb*Z(BxBape
z#`rCUnR!c_oSScZAXy45nJ*|~m{t=$PtQ`rrjlro3MqGDbN;Mrgf}0(g?)qCpm+!P
zJW+&ldtWC3kbD?ShpRw6iq7q8qw>>rh4lMA0@AaM-OvU6Dspk_+;{iurKiarZ)^h4
zDj!jI{L&5k6}I<WVpf7A@7)?)01TgU_)nk|8l=;wUaG%}{_YedETs8X<DN1?s8)L^
z1C@xp_)PGmMVKT$n*W8BV;+((Gj9Cs6-qsUufVf3E@Y?l#b8Bi7O|YBTGQ#{py9)b
zkSp3O*L<J)qK@a{fzgy3ZT91L_V0C?X5;Vqarz$xibL>H7#riJr#b3v*PjZ!_m_Pg
zkiVek+1WouP=2j)!dhj|{a!P$fQCbzO4R?d>d<g5&p(_b9nFiyO}To=8JME}#IaFX
zw7}X*El+vnCE~tfr+j|Cm*Em#*;tYPi!e;EXfZ@As3S^!gm5r-{wh8DwyF>U!$uUM
zKem&y4V5qFBYG1sG;&IAEEyQGgV?@NW@Pxh12w1`{pRR!C~syz0O=Ww;xEWcowFj0
zP~Urxm8@LI0c?M#J!6-EUuvGL%e8)<2@!>p6vmIEb^Z|4CjK8B3=u>+a;KcH^K~}Y
zh#tW&Evi&$fBt4%;@H&s$)fauWQ?2`HzG&vm*Hm|hp6M~#@?IHDT&DE7$6(iC(;lP
zC%wX%wk{(tH;E)k{l`(Oni?hJe@$=??$3UI@!YU5w<5ajj=CRn>Y|kwyh!<aVC`O~
zEG;E9%2m^`Y7l(ce7IoRKII1;S0goDGDlQ53~(pE_S>^zG{5<v=G=XVX@P_GAwr`;
zVqlc>_dQm~9H+Nuz5kHUt@CYWPM4>$f%nayCntR&adREcx7kJ-lUFLUpzHu0d%alp
zeiYk$-1gNoOoS8NWIinH0+SQxkN<LX#XlJ?_n5?@{=o`l?(K8H4~e|4P*}!dfRamv
z_e8U1OO3mFnXVFybt(D=?=q}#T$)Pb-#RKq9_fwCGt3mpVxEhSqtN^>0!oa4`tusq
zUugWDGouZZ<G}(8>jYUZWPG*e2fS@%Co+6843|Ee)Ip5q*mKA%EQkw9>mAt~gVeDk
zg8|@`Z5JBL@8hp=X|PvhUl>qC*>}#pz%51-Nr2XaDOb@dHVT7jP<A91y<w&x;A--0
zWvMuef3&*!P-%g3R!LMd;9ff3&bN@TVVJWB!M#6$_i}&Hjwg|Et5a_o+6|E$<#jZ)
z9wGbQ>3sjvwWDxW!g2AF^j!68+NuOmb?qu$*0fk{=UZN!XUmPD0!;hZ?@GIlW0oxa
z7}gaAwlmEh$1xz-iR3iRHJg4M%ToS0>~GD}s9*W`rIO4?jZDZ<PE|UTnPad~ujCh$
z0Ycs%LR{5c2_X?d4EBSf#!_%vP~Dl-_o1eCMEt!*kFpvaW!-=lK*!}VX%(QDVTt+w
zP<Jt`BdSNi<|R4MBc(=;D5@jSqTV7A>G9A%Sm}{+I>v0*Qljv3n!$xK)^0F$L=QqW
z1Wq5i&A7H#>LX{H1{?9Je|aq|x(92Mk0Wl23#xnZXYY?LVh-)jVf?3p7i6>dCh(*Z
zUAZ@ct07`Z(EgA|096rx_tD`W8UEIS`q43!2|@suZq+-|7Yup@saBzJ8&((vc{SA8
zqBYue?*<7!7*W|&bKA1`be~2{WJJ={X(_?rNoP@RJ`@pV?{0iLOt5}=zIw?!DkI49
zBR|^s4+w@?5oSBkAps_(DhcNY%boLo5(qUkXi#L5BvzcUc(UZXN_}3L(ad+O&<aVE
z{@Dl3)Q!aB<deems4qA7q@*PQn_ugP4CU(OeVylzc2CxO&qMGOOLm~Yted!a`flhC
zE)-=o+={;D0|x7toP=1xiR{;(-|YTRe1L+)KBg4OmdO7$%*8(uG$4EwZ9PER)#@&s
zwZR_WN}Adp%fZRO8zpE=$;G1J+YATN>rR$|<eOZ@Kl5whk5E;0l{T~im6mGm2fFG3
zsq`*#1lgwtx1bTf(X+MJOb=-N9a4Au94g-Z#=#GZoR7fv`!OYWD5TfsviUxG>$i}j
zkM2M#=Kh*!LeM>JhGBAr2jX*$5CRAx2qXQEX(xG#_*`K??g>sXCw_E)-%g21MJWOL
ztR_BBN_#t*oxGful>GHu5C-G))y#(TGGsu6ugcsp+qN(-xi9$}0noxon!t4$EsjQi
zeD~uw8)+?{h<787?9Dg$XCWh|t@8Ku!kzvA<u7w-6F?Tf1D35230~J3<ABjSfMc~0
zk;C$Y5(Je$*P09qUoKHg-E6r6^}ixqn&2qA5MInx-IsJ^v@DccG56`8egeF8F-i)x
zxng>Xcf*5uyL8w#ly16@j4}nX-R$)GWTrcO$4PvJQNPp6Pp!^uOuQbFZ@c<XOV{({
zkzNva8#BvHHX{9C3Z7({r?d;s-OJ5~$wSfNkwN#&<iiwLf3l02wRuQBbSVOXh}xx%
zE_-KywW{yt{7==>B$B$!A2{sEa>H=7$F)V&ZzbN~y$t|=H3fI7@UO!5RBqz|ayl*(
zNf&4eP;0-TPC;TmN8GQ<P?T&A?8IbeF`7UigCy3=j+_aGeKsrR=+(hUR)}|x2XRZ0
z_`BzFX`X-$C;x`FJzOg05eBCj1s>qzrL)`bZ;WHddtE??Dh(?6OdmTw9leEIQxgz5
zq8%)}`Ch$m7w??XiBLtJ)HL1b6PBuA;Z2dK4-mM{GK>Xso;O<Q=8g9tGRInA5WhN!
z!U14vcjDVFFW;#q=6;!W{l9Db*_29c5DAxZ-kaaf>qybwsH`LQ_o9m9%@N!DCqc<M
zN7W=$WYzTdI?d@av!frg)=mGaV6?o{w4xg~@F<hlc=Jh%waAtO1AnHkydb|>RfOcz
zB9tJx{?N&GJ6`x?+MvDU_LZ~7dzZH-(JGnjJzKuOsMYPh+~LjZ`KR-8*{(YU6;D%>
zien1sAgWu}o)LdQ)rgX(Dvf^KI|Fi#6hCX%yms(!hU+LoO@dOB@*Z(MKF$_!E$sld
z1D6S&?<893f8ki~H8_IpMFAe%tn4q#dIq?Id0v|cq1}0u{)kv)aDycSMLFs@r4BF^
z)ULz))(kK>LjBAR=9?W17I8N#vd=_|MX8j(0-%B|u?L`Wa&WCB5zRsYh@>5$6ih}<
z3e!z~hEj+g54Wh0Mm%`Q=OIyA`lS|d+a`Cw@lud$Jzsd{4opsqjml7IB^4lUc!NTV
zl-a0MY!JebOqS@Ih~w>Xk4gIS+t%6UYyS6ao@~=}a2u2?dR@hKvjUV9`~S8VpfQ64
zYtY#!E8{F#RvqJWDTLKNcRBX=SeJLXUYyw!m4vpX+pXM6=I_mEz2vJTY5wj`gHFPl
z+-6-sd*rO1#A}nB#rzSM6EP)FfeUv3NWkr<tW5S@m2sJ3g5BEE`@V%`ztag`BNFk}
z4n)G2z$W*?ThcrNUg?nT*6IQe8j1ylkl?$pjRE(Ym4I?gZ7yU}B9K)hw$*_s9%5fu
zBK2kYeaTCZ=&n>f>*b_=ORDjtNAf7n4pj0QjsFtd#{?K~cJj$j<1eM+h?D}uII2RX
z6HJt8k>4#+a+`l~018>J_3eADn%6(ESq#3MB2&SGsdW(tOd&hUd_OIQ))=ikdoUz+
z0b+EKR@EJZ@S|2Ov^GrZwfAP)g)i`)&5FU}wk`zQ0^ot&o<H3)NSOtE^WNqq2w!Bv
zdj$^Enhl0;iV{TLjQOC}J_^E{DKSGt9UK%MjbjiWM_167`9JQFUkT}Gr6P&ITF&ev
zrS_^4jx<mp@jMRHr80leH!sQ-x!Cl}GCQFcnXif}w%QX`DsJ>)hX&!4T3L*aR%H!t
zaO-2oG1e-qQr225lSM^Oc6IL?T;}AaqlNtR29kpps}hQ{Kt@)<Sg6A$e;mIC?*jm<
z<+nFW?47VL-Y8TX=)phUt@bW`E1-~|0*@UAxTA<R`Rqwj4(N9l&lZ~$fAmn$$%Rnc
zcC&Ux*pT4ALTEYF#l}Jvjfa({rT;4RLb>FXv_c6@5_XyeTU2_WLOLvwM{vF&fdlp7
zw(V05-&9T<$5@W*?}=NMOV7Z|fWRr=?!&{}Q{hR9@@1Oja}ocWA3x#9O@DkEJQL+2
z%=!qliFrhzws}IStYP}VY<N2NXKQO*WN^p}4%Y(|DHZgVFO!An#CwPK09fCMFMksr
zebV;!^_FU<LBBUhj&(K=&3H`FKawSgd3(IZf!WakBB)x1yuZVc?Sxt<U;FO!cmr40
zi|Y*%I27Ja(cwkQ(UE1c725_KjM0s6S$*%ek+^SyM))dZCoY-w-x&|z2In6k$eCsJ
z(iK>BQ${_ASWZ6eQ@DgukgGP<uuc!0jJ|WyF-Yb&zs8t1hiSb2qc1BlK2f57Ar_P{
z-$m{0!p|!ny@i_cp*9x9VVtaCEWb!D(z8h^#rbEpoy!Yt8qV^D{YJVO@Uc-MLk|fB
zyf!+p7Ivm6u)g`+f$L)bV-vJJn{_5EJ5c&e4wsy6Gg9MYo?w%b?q8FI`auVGrh0q-
znvZ&I95o7rO0^wIbig9N)=qk(2wKkHdPOEQjHJ+yX{rj-Cia8KD1+o37xrCzPJYKF
zeq#M-BBs}}y(%1o2a!zMCU*o9oEt(e0g(hqm4JMZ_`vd7|6(<+&lNf9f@g(O!3({U
z!)@6x{xo0E!WR$@b1^84mV{Gu%~lgc2@z>_`^wN`PHg=#j^(X5j>RlHA1#N6TU@Wp
zk7NIL8i2xPZ776`rmo|d#*t^WG?i(%+~Y;h1|<)J^&8r)w1_ybxXg1}gNrd9$?ee(
z2E+!O$)RN#nrtyE&N;r8uHNxu7@Bz0HI$Lkr(wj{PP`xUo+C*}gwilMB_4mhJ`@>0
zDjNO}+e7XW`U9l~AW*Z(Y$x*+K8XZQ5)%CM-Vqy{re0tCBsD=}*WrBlcFd8iabHn{
zjvWx~$a`?4Hzq<CuDLp#mdW@BF?*<Ljy_gY*GIq`gSbR`ZG6vvZ(?or+-^Zw+d?-G
z);Kn$1887B;s}#BJ4)R42D!mON&|3Em*R7iX4{!TN7+nz+8oehHX-QdFjsAqLl;aS
zArFW|hrSSjbf_jDzJ!)h=Epx7o*lW6nYS3mi`{|U=ejRllrb2{rW>=d`x~iB_W!NR
zaE1y^A_yDnd9>9036ek-4d;cvCF;n=GD}GVRwZS}**nIyW%nP3gpU1oOc~?Wjvbs3
z?)|M;tf1un%&)W@r7vU<a6I(6Yr^yLIX;ZIeYqKZn<cROO^D~OaB9pXKbzxpp)^Yr
zwbNTb@}xs09vxGte?puhG-HRct%*yu;c`6haP%7LKu?$hei_q6iT<M<f#Bzd9>6m9
z?pzm(*BcOkx?iK{aSGQUQJ%M4df_|h;24^Wcsr5f)TKw^Kwxg(;e~m*H*b!TPW>u%
zFGvVWBxVP--|6n^c<xU`L=%9N(_zQ)-;)hxwx!R=*eo(n-yueOCPKfbz6kt#H<yCS
z_=FhZ1drPD!n!A&5*VI>oh3I!9bqnyrlL8~=ZoXM@sBg_*#wW``tL;3#0zD!A!z24
z>c(&lX3Kq$AYH*+)EmM7A;Un%*G~DUhiZv?`CI+iE}peeWP++L6X?CSwxi)}S;p5%
z&&ie@tGnvY73mBj7Uc`{!ivePC%C5iX8idH-CgpLywf_bH(*kD{k|H@z4tU&*?g|J
z=8%P&V;70fw+hSE5!GU~xiyX~Ipo+iDm#1J<^d_!sy3Y8`qaiQE2XNo1B1MF8KF(L
z5Wk~vVK6M*kpD|Gnw2Pjf{ubSu!z(XR4e9H3xs3ywOSo61p8Rxt4mO|NW4ORulo-@
zvJYnn)0DY5wCb_W5`%{POiwmY?$0JC8qgo^`!dyD446lKK|iG!|4O(?PvN&gQuucC
zB`sAZ*ASI<pZn=KfkeQ(>OAH*N%i3KGP6<d&jjkkk$k_a7}7W~AUxY!UJ^pcpftKq
z->GdAkN9NFL#W-$x#tPh??}3Ps<Y6D)N4~1N+|wm5vg|sV;a%w%fgV~i{W;X@>n-v
zxJ4Q<$mtO%UEwWRb>0;v+1^hW#m}S_zW*oA!wpuF3L0t*B<E*>R^~clz4zWo+pX9b
zla}cignb<(GW*e_?(Mzz^$lG=hpceI<ZcGl#QOFrzG<&Y%qDt6Td}PO`y*m2ZgHeS
zb6M5lQg@Wn<$SrUXFbh$W{)JwG&vz+P?TwR(@Ft^)D6g5`FxsyDep^}e;>OG2jT`P
zRHaX|`z+J*Gi1njEVQhQJkY`0EX(i!<#ssYZPZO{=;`T(F8X`Kr1~(bCkQ#Yj<8IT
zH^aZUG77X-Vj$@0<%)mP9m&TVkTRjNO?zRF3PgoE43y`SJ0sS)Y4}6gRJ|+`wW^*o
z@%#>F5*RnJZ!qxmQp!70_8>O!8A>?~jQc8R?ojhjmyaPElt`BP!|No2j2a2j<`ip)
zaU#b#s@4d4PXDOv#@C%_b-i8f<sA=8%miY@1-1N*)QcS?;~A87_4PD^%3Q4KD0cTh
zE}vEz?Tsr^W7|5*^PV9giRUw~TuXb9i}27hmre1h4PtuJ2fEEHQ@ac^<bMS54PrM?
zij`wPK+3O(&o?$<_uKoJy6J+)bmA@a>ZlJB-qBEdO)zZGuCEAr2DjU;MgLg~c4fo9
zlJAe$qrBcnJ2?mT{T~}{%}MKZpl~Z^rPlNCwE+%_t-U?S5N%eO6+G-PA0b`@OQ%K4
zvAOK9l1umrkv9Q+*@=GBwVw~7`xVhfRw)BR^)(uYZbvxeQn?OktfIl1bDPsIa@<Mc
zyCoY%-0-c4J)_v@QFSUS0qZrjglh*ra)@!YT(MGXS(uog9zE>hX7$wKBHM^pmg5nd
zv6xbnUDIMJ^SqT%G{fw_KXw7AS~)tsKid8gI5l7tANLzSkNtNf<%T?pcEJ^r(I!a)
zIZ>2ZUUvt;_bncbhDiz=yML6l6#o|A0rXOP?t%(}v%{4Yy5^QQvkHle2g{x9$|ACp
zB(!`qpX5hnF-WAjdL@zs$bmVT#V57}^D7}Ksly-H_jnO490Kk8&5JKMXi$~~VY5x-
zKo76nPoG5Q?fzR|9D`8MHS~=0Fd-J?6XO(molR%C+)W8n$>MDE_wdW#x#!UM)%Cg)
z9vu!SRecIk7g$M=b2Pw8`ObS0Z>7<adu>QKMO9Nqch8uj6mLd$`c90+m+Es#@E50_
z_~yff*L*hBIq(X?4Kkqg-O9JC;)0u}>|@1%X1tPy=cA<;W}X%wC3E5+nXr$pG=GPG
zb@g)?nv3+bc_c;n=pu+haLPd0mSQEgJ%y_Re2$KYurZ#ENX3(c$cQqOO_Qn8&?Jjc
zI(%3%#r=J5eb(ni;rAEq75~CJA#@HJ*~gevG`RE5lk!)L0|9mG?eUo~hV`bSEetDx
z0IQSNO1>sR_l((mOUV^ttt_$=ZgMS8zqX0%8C-JE@XIY<IaYPB@2j5ESe;I>(EL5B
z^D>Cv{G~oYIJuRf7*PhXAOJQOE;NOGxqsrf8QDje23eez_Q-PTk_!~tUPwu0;rvua
z7;P?3d#I&SUaEOL5y&@hwx2f}q1y*iB75P`1A<Ex(GCdhKq7@~Y0lptDg|DwNl!&_
zLM_&QN<oq5U4=3d<+rvsW&~Pf+3DL7d4&kt#$(jkIkgm?M?xZPC9%mJb(CKpXc@$l
zDAwKuZbj$0%nV;EU{7t3>(Y_6U&`guM$0x?$kP>8muVo?KLz4Jz0mgjSO4BI4Mc66
zfa)BmR1A`zPWQ#g*)a0Ne`tfX#<ZwIvdliiwMtbvFdOt|WoOd-^lko7sYvpt$3+(@
znm7Y7-|Ng}+YjM^=9C4E1HO?-#&UVgvF0;x7}7Ovqx5xg9ZPGEfp)pF1<Dyok5#kY
zFeSnauUY^@MsUtMhKKy&cN*9I{WPWS5QG`l(NuB<1BPDc?_5#ZUR<`1`i3HZ=9PMt
z1<*P)i7)$yOG2%QO;ZUBQrNbqIBTsx^DH_3_BasoQW{G~`cXo+?GIu$1YET5s*?-~
z(q!@mH)2<Weq*2H<Y+3Pn5ndByi@_tV2BM8SPPZgeubdgFOwI?lj8=|9s*OY;%^Wy
zYZ-G4zFCVt*vzj!(ttrJ4L-u*Nx&m$OMvRDnV+2ol>2R+fb5|8!YLs5Z}<W4ajZ8V
z!q5XPZ*PV2N5O9yf%_-^sZVi1c7PqRl|t|eJl+%#dX;=GtG!{2YpQ{hk4F<ID9dBd
zLFZU){fg(3bHwusLO#dbb21rOS;hAfVHTf*5Sl+9gU!_8L42TyX5;QeruWP^f%3Ja
z=3mN6**x4S@?k;7=jzLluHuMdYaJmqe&H4`k?V`Us1MI;1Sq|VKXqJIjO(??c1F#g
z_j<?0SW9y_qiFA+1)&I{!y@os&x|$22x*<MwC3l1{^c2VhSG)X4a%N|sbj-ib9`&F
zUYXN%FB5hbF?b2Cw`BKT=F1VbP!o3}XR#NMoEu_`KRvM>wpRb=+@Mf{2F0#AOY%Q;
zB2%SCA&!fUZM2xqA*mhLX;~b^q$d-&bU(O|<kTA^-i>vr^lh;u_rGdBKgci?KcW07
zAQ0XKZDG`@UShoWPDrryeC^q>csRJzJj3RJh&jNPL(34@iu9e0R$Qi)wuRU<JLiJz
zNnTJ;h+%&hPR7WTT^vOc$JX}}ae-YRF9{lFr(o6mnDT4IQU3Ih`>R*G>nF;u?!P_<
z)6D#8fA)L}ZVS0)RsPD5mX=q6v)PLEi%E}K>Lhc@--BDLnpwMY=bClO(R>e-gp+>#
zR<5}(XiF&7apXG(Qc7;uC~dR}A26;eO(26!jz1SZCbS<Pea>i1+_RD*I~E*2Y~|w0
zc=vEXw509-A1*!Yh6Gn!Rmes2B_N)_yIW`@&|u_Z0e(v{4le{<3wgS3=MR)IY!Bx;
z7wdl-8yp6xb4ou?(;zw+z-&4ty1CdE7D?Y7^UNVmxb$c?6@Su8+bH=QrR`lva$ic~
zkHqn3gDC_INzEvwxg#+;-;;d}3Yd?hwtB*@bNy_;DU%aDL`<mPc`n?fPhkBj^}6lR
zHNXq~^WlOVvBGL_WIQ8bxX`(--oh-`Z&6eR&mZg7I227kTiCbxTunM>7)|I^qHwI8
z<dp~(nN$pg*Omi=Ns-ylT2o2l<JC^^X}w6I0Ybc^y;9?0f^9oXDjoh65K^=@kGa|k
zaIO6SA%A9_>a)GRk<b?G9`3;bCo)WG&3fRglC%PG{Y8dyoxh}6OfG!~>t3WRe9n|>
zkeiTeqUa6;OP&Tu$tGs$3)L&Mhnv5^x3DRRr@8>3H+nJ&Wc%6<G?QSAupLRWF4CiT
z6OP;&`V_tXdkCl2c*1odO}^%f?LRi#xW(?d{|?;Zphv0_C|sLqr05|!@5F-l_!|v*
zN=^DOm~ls#4K9ZuNX%u#m`gPdTH?>7!$&GsDmlF!sK8K8-{OClgZ5ir=b5Z!H~oPC
z>1hK;0GW<Dp81%lY)Rq>^;X*ASmvI73I#EKe-5&Fc!!sOjwhkh8hMIYiDOwLS{v>M
z6>2(}avC|v`?z@iq~=FSH{dxD#*NVN_FF-i6eUG?@)<nqd*(TX&vJMOPZG<wFSM3H
zu26Z6HWy3po%a#;qGvH*2w~3#-o(CDNG%8INoD|g)(P{C&U<|XExQii<%@#s)7Xrc
zvB$x0S}N+_7k=8JWMtft^TA#fERm(M@MP^dMOk7Y*J5+O>&=Yxo%c7&0m7I(DO_^I
zZO0}6nRo$@>8a!=#uV0**^9dipu(>Mez6&m4)^<2+we@mn)i##C5rf_BCuSE!X8A<
zQIao^Va9a&GBmBB^8Z8P84$e^q(|}Nlhsmk2qED|@;OAbmS_9Q_83$a=JyBGE<bNi
zEfN@iOn%az6bqW7^NxzdB#l#?XHMm^a=R(IGf%$WySYBX2K{{F41Yl3(z*634|UBQ
zoQ7NtfyzQiN@6LjyxJ*ry?WVC<|tX9U7(wU{+Lzb*@#gfP<g;=4TdqEeTRy~B4<G|
z#Y=}hhYZ5MLnZB5KrK^qnFuS*AK;p9BokZwd?zke)r8UmgK5m!f?d!tW91ScKEnVP
zWSKAgHUkv3CYlq5{OlH(W^({bRn*!ofZ)d~PDDn+)p<fd^Vuw%uK;0LmJ_N-jaEny
zhj?e|f7D_mP9JY)i&9bz)SKmH4=}2KCM^6vb1Zz!m4>7;lQ2SlAVsG!5-`pj2pdP|
z`0+MIL3BDCuY^*HTA59T-@Y@f__l?>u>dxYV}`B4+8J0Y4z?$aH)dcjhvvb#+#qZn
z8xMu!ga<;lF}X;Q@U3HDc$GT#E1}dNp=r7o;<3of-qTU$8py`Dj&hDU#Lk_@1_bMe
zB=c;1A&n_>0E(1qU*fsn+63mn`DSo_s}e;&%yu3EVEhW58b?VLy$A{ki9B_QsBceX
zDSmpac)HyAPk2krPkOs*=zcgfoU|~KRGLB(7S{ehh;di~NdZ4L(P&HNPVSMHU2C4q
zj>VcV%f>*B_He4`-*h?2Ao6>ctfe1&?}R+`+zli-P*>+P%W#5`)?Y2uAAZ<C#Y{4j
zv0-X~<%Pp2b(0Zw!^I+`tzr|x@7_we4da~g4inN$2@d7>e43t~ZYH#S488^pH1X)m
zH0a>;KPmgP8~F~^eDlXZ48HHACNMtc7(^n^6nG~Ltpo%vX6Br#eY#KecA?E7Ky`H;
z^caky=PE+$fMz=n3aD7K9$fAH&F3hX+!IkGg$I26BbK1=AgK&cN_ErV(~b-cg2?Yn
zPT!PJL<baa=0hf2zVCz$5?g-reSG%v>F>tY*Rt_henL1NJ`B7fYGvM|9ZcLxe^)G)
z3uWA1gdTiGnWovsz$B%Kt{nxg(I2(bk&%!bYq<mB*FC@^r<D(!mV(;pC}Xgl{=va#
zC6T#UyrUejfO^bTZWr`&9th)_ABkcCJpLEpm|nx!C7clC5_@b|G`{<@8SI1I$-`Tf
zOs4gB2W<E1<OHkP!0ffaE8Q4gt1}qNm!ziaf`?Ky!B7Fj;Yxt5@WG{={B{(^eZ$55
zgkk~&GmO@e4Wr$7-Gs`v{0rIpcLl_(DAc}w=-NtvOvGwhjtZE?%ASRnKDYQ*73c7p
z>N-89WiNv{Vmw_k-2TI+5E5EVqHx5zMPEL(_7}SC2?~Ge>Uj~NDE>IXm(y=IZj88O
z-i3>xRMFHxd+;HsrthKLXd1F{pTB%jQ%oQJDftSQx;s(7DN#;>1s;Oco+k>6zz!L$
z+ml3k+N6n3_Yy?&D9TWpd`Q+P9P$1GnTCCUXL25$98;D3KdR0;EULBb;)<knNC?uY
zgfv5UH!9sFjUZhjL)Q?}AxMgXbc2#Z2og&703zKD-!tbu=e+Ot*ZJdI=Q_ilJ$pa*
zy4U)xVM19`ZM3DM6M0|srRp~gczK0Ky6u4hPf%qzjB&Juy!Xr~d9d|mzxJ3fj;SR+
zU#Mi9vU@SP+X5Yb5XQP<%+yOT`lqF|trzU>+yyWazZm}^;<cHGcbVC0D)%6$-h4Om
z>fh_KA-gE)_;yQ-N~nV~?%C}Ah)dN%M6THqosZQbzej;{s5D#nej%}lQ%)l$G#nM1
zZq#UH%YwofBf+RE_Amzc<h>!zIS$F62-$xcP>v>Zt+2$}#y-`G$Z%;DbK9q7gGrfM
z()3Ew?@L}J@~66|B}N^CV6RypMHhVLcjgW4^CAD?yM4>6ekK$1r*o@<JSMi~8uJtt
z3~|muZo4}|^%Aun)2{-*n6=Drx2CpUhL9f|)GyQy^q(UB&_0$L>@Uk-|2O2$LmnJ*
zo7jc3BcRe89jlSt+ivt&rEaI*#}S$%?q<ioVbuPKEjB^BN;d`1P$TC)iV;LPokYNz
zyDG$*Pihd)U6|7HYi4-(7o;!d6Z&}=G$FCxyGdT0ky^tvWNSHC^ic>q52xutpK$+Y
z$oDcYVe$e`RX0@b^E(^kzT?+_y`KO2xxuyxSP8yqTw<*#U1@02&EWE6-=oLh&)&gB
z^kJ#2rna9<e;JcnC{M|jf4@G2J)~soGA-In`-a9TNHxo+o6;qJ-xUq^*v0ePnPW@{
z$|*66Ithsi5Kpc9ex>JxCsM|X9>CcOVznP`i9Gs%=mB;;|EP76Q6=gR^<)RHFYpnT
z<00g5UETG+Lt-!o!iZQ2&Y6>5)GRkd4bu!_)#Zf`Ty~Y{9ATMMMsADhIPNBZLg(|g
zz$l@FotR!5-(u+I>i6%lcu6;E*$1Uc5-Gla2Z_L=f15A@^E<=peFZe59FyX(RcqOJ
zWO<c?Cv{@ke72>S))ZSV$4ro9)zTd-viUR>S;s26D`7^tuOV-UWbsLw{(dR}`Vc7!
zW>c1qXa&@}uQJIi#7s_4ufet|LB60vP@}h+my&}Zkvtd1x*ENjN^xpR%cK^GALBY*
z7B-mBoq38KRcsHTysP%)`_zPnLhd2q50{awAY2dhzrTqo@?fO%QP^`Ujfp|_CfmTg
zS>V$V;^dP_qqOsuTj5dLwR|M^@hHq$qS#wpzs6N6C5HPmi<omV-nxqfy>YQqilMDz
z!i<emYJHPn%ELlzJuW*}SJ)ACvm!I+bcQuulwEdo-G7D$5eDF^5HWr9=pSHcoWh$7
zv+_Va8+98Hq!|TAO)ytSZOv61J|>AvW(db0ct9p7xQt_da2UU!a{X}?S4tK30Z{x-
zerxq*a#hHtHQDyfL<Vi<hf>;IwJeLqkE?w!$>8mx`5jH>(f5?4I(}SyyVU_ECoex$
zceFmdM^!%Ur)px_pbOW*=I1(WpFQmq@Oj_q<EC>M^|s0`Z9lC|+&xy5-zVLd==yHy
z%wM{=;yoY$jNM5T_}eI<iHdxJQ-(#NP@`p5Vi7Y>X6Qkkv`Nm9TlE=`Yq4-hkctCQ
zFOB$9ps<x!57Y$heZxqqu?9tcKMR0G)$<aZ3U}@NkV{slamq5i%KXoVa@$pa2R`q7
zdKx}$L1z3qVSUyI=Yzr7m>`lA7k{+PvKx37=`ZM2X?y#X>{6)@PFw1yPNCoSI<lRU
z>9LEroqm0jk*|~`n<1^NeF_(+hQ!62<e2-31|x-5n`>$*p!6rGay$BFXXjPO_GXS@
zv&y0H^THEIVd3jGGoPKva1M&>%R>#|pn5ga=9^A`(7xNICQS{Qlw`1cu~WZ0e=me_
zXd&>KZ@sW?PFqVDhmX|El*5VQ@Cde=(4SYkYfYcPyVL~pz4XNZSG%`UUs^Zdmll?4
zUZ>|cfjx8;2=+W{&XY%gQ=Ep~DOi!h0j<Ha-*b*OA5}{N?VqQE<YUZ5G^(d&Z<stf
z!|_uAwsW9BDX9R!qO^<E9~olDSo{DgQRQ~2My~jai-5Jk^c0xX&5KWZbjUuJ7*N6n
z_Notj0EWmKw3$oK6scwYRiyz&`6LGBEq9Q|;sKi}WI2i0E-)%MooY=)`<frF52u1O
ze4n99>$@$uOUu5z4<IFoqQb)51Jf;Myc$Z<UvEyD<m_HX?C+t?IK}KoSF|Tt_FAkK
z+EPn|7K~sXc<e3Sg?tIhZIQTehj^Q{4_>dboSSWVPqpvToUazsSap!)_I3EU<KBJ5
z5HE3=sG2LRMLLvgE1Jw*Weu_$=NqVvB_4(p{Qf1!2A;eAjU-(9+HL8rlKn4AXzM+i
zdlFSeSSutjR<4p-gH-1fIyAUGKUWT082?d{xyCI!zu0>H-bn07cgkGZ9AUm_k358;
z2J2VbU?rg2S^$t}_}<fj8Dx2>Axfv8p*$z(*y#<d;CGof0^t+(N5Zca>n?%wZ%CeN
zZ!tF9FS*TCz@iJ2(tPRH@wEgH_in;NIsbh(1`?);2Q<yiw+96wD`F5o@?^Bz{C%38
zfSW%BNBroA3^UmdbR8X92P%2%P>D2_Z*NW|W^MrO^c`3?j?OPOI1xac-B{;UXTXdz
zQr8L=-K!wgRPU6)prBX;+T65j?VoQxiiwS_-A=re!OO32{4xZn9DPqh&<Ad90-iK`
z?bN7I`&Q-hIZqo6l^JVzIxQ@v(vHH!Dbqc-g^k$IupgwnaGF*cz&227z<TsOiAB#Q
zNMbBTFutb!z`n71LTPxtCek$RMjDaT|GdzA#b&f}z~dy#o2oKA*I(WK*GG<k+@S7q
zmvgUXW9}f@3a7_w9%V}NnaNQ+QAI3{)b!^ONq+UoS4%9N2A0c21)E<Qt6p!mG4}qB
zqKD~YQq2Sy)u{?GXZCjYFC8o0irFB@&vuqV8@(plXV#mucZpB47iWHbzHDo5oAng+
z7M<-dx&>)NF{a(=N9)rKD@ijuCzox9ur1&`s>dJn%-H4pHFsL=Vu-niEFkKweYGMl
zGNi)U*fkc@DDsL^5N~+Ms0}=5Q-kijzWDnmRm!jkUtI2vcCx%L*9<E}f&-n9*xgn_
zI#rTRzX&J5G);V*^V(AMdj5-f$kFD62KbHI*6)+tf|#TLd-Un!=N~7xkw_6&OButu
zO8ly>fm8VbYFIxFCNS#~G~X{&$#T4|@N&9Hj!bNqOm>{9(!c+J&0^T2N^1I+r_H_W
z1+S%uGn(s*imNedqcZ2#Buf_NXAO~JJ`9;$rgxizdTNxkt6ZUyjFr-%*fU<I;a43`
z9zTH56MiFd0bGNo@Z%^2#j(+%<Zy?V^PRB@ueEncM{BeRDczKZmRoZ<69^?L1TT;3
zda)lHi)Lb`;jcECD_H+@P_SNji#pZLs`aat?8y_1u88yX1oNI#NzmHp#C?W>c7L+u
zEwBjaVwQCYd+o5x3c~IgebVEmJYU>JVBk@G!Owv+F^8vNqU~a6M(Fb+)0W`^{<)R0
zyHE3}?`H;8lhpP$dF^ug(+D$I<VT#Vo8#0lY*913&fz}PjJfuv)~N=278L+xScB*)
zZ$}<9Y*h`2s_oJ5EVP^c`fRv9B`+%v2DZ;$36MtuKGXIOEk5QV-OGPv$wBpbL*<+I
z08iPG$m{&FL8%d>SrAB?><qklNuJfT&0pZ?HAtj-UV_KmmKuged1G)BsE0v&tK>~A
zs0vM!=JCn+{vC^2ID$%$<Fi3+B1pvgtw6tE3p#SkYldY?JKu?aXr7)N;|UX3SCUnG
zN6FVStTv0xjjcZu^3f5z8Ph1x(}OwA*GT!!U%tVOJ>I6|^x<2Jpz3tDdtou?Hd|>K
z_CEUy6c$i_yyesyS(~rYzxp{t$&NRkbI!R+Frkt=Mi4&cm0%&a8r++csI(%HggLg*
zz)dq?d@9CTIFNiSR!f+TmAJIs-+vD`s-OS)eZ_^M?$yX5rcW<pT~-KIC@A~`0va3j
z(IR<$B4RE-m<smu>ZBOHqFgv>D#x_qL-7lYO=X}Qw2ikVT08800+)5~y>*xV)tY<I
zK64a<M)+8IDAInhn+N~>?7aeHs~2sMzbQfG9rLckH3z6^9BAAO?1dGpMqGgID(g%s
zR8Z^VJQ1HBwCj)GdN?GKqKNF~IT69nuMN2LoXLUtD>BHN2oMs*L~}dx;kO`_nNx1k
zo}N>HNxfzS+Ct5Oi>ED5Pzsl+^XU%-u=r*Y7B(Pd*ITUTU+@6#lVP??BPhB%P!{5?
zm!y*_9M=@%1o6+#h6cAut;8kg@#m|=sF6G60PRum)jCSNc=Y~1!>Xhiog|z{77wL&
zm?ftLs5x-Z!o{90Q_@4RjK5PVEJ(wooOLw0q>sRA9neLDVOVSAm?RA=0vPS(kZ0#P
zt$R5KBzhJU??U`$^<EL9)2)^^)N^yT<jnpi$cS2h<1WO28l(AK6lF5;okM|Hq#F<Z
zH*(@Ng9gP{Q<VXkUALQ|uDy9Y4{69;x~=b%u(LJ~#+p4dAyGUW7<x1lEBOAs{_vB$
zwVutHiXw6pDmo|5O*41g(@e_RGt>~7KJBo{9{BuRbD`t;eSt2AvVDUa@u2c?hjG2f
zFNAgDE}=b5hlfM00x2n(WG&&mx?9ONrWG|lQB*qfC5?FPr3td3XvL-YemvJnrI=u0
zXXmcOdt9()h4$SxBS?3?r``P^++n<A2}|K<K6myYQ2}3t`cRVrQg-<97Pf%^GT+=P
zz25@a;VunP?>$yz7WmqF{f<-!GysHHN81AYGBLi3!>M~t2LL$c^B<mGBo;7mc>hti
z7cqWuOS^(l*e+`XHDa8?vJt#3w?cT@nrSnk1D6jMtTDeYYB7!{RT361i6*pd>r3pm
z93rtUIfwgicEG8ROtG|LzZLGahYWppYJ=WI!b}WxKiLjv?hTZ|$XVAI$l?e>NF2NI
z+gZS(*V~1AwrhvPuXz*~oJPF<c`8T}Dw5tUeyV)W83<!{ea;{H@Z1cBzH15q8S%P&
z!P&w6*VL~b+Rm?AQfdrlzq;2Mt*PsLFzY=z%!4|z%HCu-+kA-AaO}(fu|L5QYjo=h
zdz;xgO3+D<O93z4^+|=_Y<_o?N;W#K7-~6`jE?j9MFC499&O6!ncDgQE%AE{jt@TU
zMnov8;OVcg6{t?RX&1Tcz7PwTc_0xqDvdK|j>Sw>>v2q^-RyX`@#Naa^ReOSqjy8Q
z*N-ioUFf?HLZ2wTHrwzEqk}2hyT0|Ih|iM{F+b_15#r248x}>RQ-Znb<*=Pm%^k|K
zdXFK<x?Tl2D(*q27mDo21Z#JQG!JArqJ7V-pMvrJAyc?>kn*m-1RzZgyCi{}^0fOr
z095^w>4UUJ7PW_v%K-elXft)dk(SR^Wv}SwJT~aAs<KMmodjX&eq{dGy$dOYWd^^)
zUbnrM*FfIIUq2J1D^bsoclfusuZdc9=dN@atv58_IgtIji`Q?$dU4}U)#BO(?50vC
zFq@P4x`=)buk;cZDjlW-;;ZyEub;kg{8A|!jG6A{q5@cp-Bp3ZgpsitHOu_NMp|PZ
zbBV$}7GYYEpnMbB)L9IKX4s%YtXcrEq3(Rhd+X^gm9Vt<ejVYj4Qkp(ubIhB{6GDe
zPOU&|_uO<|S@P-$VWr+8sFW;eu4kIab&%nB(Gt_*Hv@nLj<4?#k`K`~mr_WZ%)2uR
zGcYkc3pzmb)nP3a`A;m1b<-Be1_{kn+r}F+b}-Z%-ltiZ#fu#>{a4?kLdE0)X9DY{
zK6Cl_xGw|RX!K6?-ZWROJQrE%Be{+B?@$Qycut0F+yYlESe|QqkIYSLcqadF|KS1U
zPm<sm+4tjx=6#-I`kX7y?O6E8SU!`Ks3Vf2z1|YhXIA2`Eo}>`)$3H_9PVtagzi7$
zd$la9!c6CaFce|n!nZg9_JA3fW`{IgGWpUbfxytVvSNeyRQv?q&owYJ{PI17pRSmb
zFP#ynyOHdhh+iL#2^{5EYV$}d6XxcP6Rb1<f)|I9XRoG1ini<!ysUr(0|GOXnFr4~
zo6)=E_iCN7n!R#lcRd@OAS>c|jku8Pe|1}eH<+@as|h@z>hpT&(?@$Gcw5@KJ1y@M
zaXDh|V<!-{JHkvX$!4g!B{M6AST+e*eU#Fmqz68GW(|6<SL)B(L{@GB2*vJf_ySkB
z6K4+bD>|!3);~bIc>jI-k+?l=*($2LZ)YLj$Ri${2f(C~B?#T<M1M{pd+b?B8`EHN
zU3O196xfbK=-HV3pGuox*>-C_3w=*0e;-SZJeokaLt>wu$W4Rt#MpPEP{3uBI&F}q
z%MFk#Y#g|V@|6`o)+^$gf0#!-Mw*o9su6PZDKcvMwg^8N6S}a$2U6Uami8uEUjDT0
zWr}LFrpDQS59Gi(mw^duuV2kgYE&qmlB*LNlpXD_`iz0t;g0@vN<gDLQj$d76Y;}}
zsJCj{5yizUm{z;<`8SA{m2v)7HYHatvASZOFP(OYXNgq-3Z@l2^9K&_XZij}Vb^yv
zput(VQ7=of!t4Lt##_}Mv-Ko>!*<~NW9=jku4p{vfA78=%=dQ;+7y$oT5n4<!=J)p
zC{Kk~@AnmZ@g<1oawXhTSXs?Y^WHd^H=_7a7)MykRzlq^*|ez7M@B7+D5G*TwRONO
ziCtET=DFVqbtEEHpIqC>5S$X76HgT3{2TxP6cYlBznW<<smv?9xv~puB__lRPr=72
zJVYc1fRCe7o*bw1@5ix1rHuqaPT#Kp?z;*^@tKrE_vH@zp?(JNZJKLl$`k4YhF4f_
z!)XhvKOO8djk=eWiFnTsi;7hwP!JT_r{MEzYK&{N;x4P4l^Lud@ks8jzpDREGq9pS
zi0*b0s%?K|owDecVa&*^k*qf3{TA=+Kp()ukpv0aCj6C~QDfA^&&Jhi@SIfI=_e2+
zXb%Y_Un_7p5Nal5eCjqG{x=<A<%f#-?ujYmC}I<B$Eov;!ykA^>a&Ej23bEb@$_c>
zo$J>LvUchgb+`V1?f|b8ttBDmtjk4Ax02uAY(CJw#g7ukkl{L}7|GY48~j^`VfbLW
z=K*4l^3TjBm4&h?`+SutC(S|hmIx7xk69mb$^_mI6r}g#ZS*IdXa5P`X`=Gn57q`d
z%!<-f@(6i~n4S?iE8kq~0?h)gOupM~h`%bQ-=(d7?aKSaF3FbP!YttJCWHo@3&oQ8
zl>ZnK!6#%wRVIyq`4a|jD0PWEw8Pvh0DeT9^ea;I4%J<%R?VL4x+zV0{=eRhD3sv8
z4-N?M@Qm`@LTCAh1@IrbXw+AIsrTOEIuSGDc(S;BnociI!`z%cxV<O|BN+1vyi&Q)
zFtrPHMDZ9Bu9%Jd3Yct#K(XjkzxiB0p9lt2?x|~M3~NDfl4m~hDw4O0_S<PxP_Hjq
zNzaks=mn-j9RUB!vL6Fz!i|v+;f`}n{6FjLoFpMhm6pBzfC=G%f0qkqYqx2Wv)R8X
zD48~g)6F;Qp9KVLfRL5s0W(4O*TM+;)4Fo{Rxa@g6FbG_K?$#Z8B_Lz$+IL2fyOI_
z_hZ9Fm{;H^p<QVq7o1kmZDC=70D_BZ9h;TaT$4xNj+cng0+5n8!KXxDLIQdn7uv2>
zn}r>EHL`@XSOs!3m;e9=Xee}Gz)}2cmvK^WHzvQ+a;Q$hZ2?6Zv9t>m2_fEl@5NGl
z(_%JNRR@f+uzY<#{qJWrb32xnK7bG1tGpYDy39X8xt}3yjjdEs84XD8p_He69=pTu
zUWsL@$i7*#L`QTFZYiaId|Fve%(zx&=|<0~NI!(1Qlh>@M3PZ?4*~?puHIgi98n)e
zKEClb+LxeVnNZaC+qZ8-goNhe%~H#ADNa-?4g@&U@^W(LJIYmoM<r|lmJ#^tIQaM)
zRl}dVy1UbN!wE#hdlW^cWf&=<DKi~*FL&-w^5Vyt4<4<@h(j<)Nl9%kJuk^Ad(E@L
z-J5S(PDfZSlX!;)ASK;!i1#_~G-RbECaEpE(Y-qU@G>XW{VC0$SMFwKEAVHXI*+u~
zJI%bH_TQ5_2k)otCXv+Ul<|SrGl^7bgZ-tsQC1KC0Q~Vk4-QS#D>}(IlXU3>qOT)G
zZYEPGJe9iGSoq&9*L`ROq=fOZExR2`vP+6Ey%}8Tcc5E-eDZPrj_@V%>UB6)WOf4W
z?U68jq#sXt&TnK5MM5*+6#yZOS-Zj2TqNeNA;uNt{@#JJk_suPM6vD5m1sBe-|t4E
zKmna7DqOO$vK~N-fU3m=%4w4UQmFBJaf{c+Xt(d@-`EErxNxL1{beb`p*#WHo`xV>
z!UCELa+1MzIZP7Nb{GH;g;j5}Mfd&o-lkX>b)3U&LnR5TCT(Od-snV&k2?}T^ysk4
zdSKSPPRwK+C=%mCm&gW&OIiyvBnr1ww<4$8;`Zh`XQhRV)-9$L7;uq7uOD7Z>5U^0
zirm2dXSx#UQBCAB72mw;;Y9Zm*(U_*dGvOj`x2SNs7+c523Nr%HyM0BmHp-JZGQLO
zqLaA#T97NsvB4wj(E^}J%9G8h^e{B02)jE<ng2dqqa^z=p_E(}>C(4i92$)tLo?Xt
z92tVDHnTx_e4q3bmt&XNJ{H%<U8sqhe$me#Pf9!F{G(6xYVgb((zf9LvSqi%C(f*4
z>6;+W(2TkXCBzlr6PwuqMHA^bq&CZf0fke4SyAsdX!PY+dc!Uq2a2x+r$EIKlVY2B
zAu_%(r_t=vcHYzGZ{Fc2M?;`F4{jo6WH5zF_%*z~^q43Rh{IKWr_LgE$YM{oI-ux)
zKwM!9nBosWYB|R(YvxUTLY8n`@imf$BslN=VTOaCxlkpwA6aGu0!T^7yhFgzEieS=
zC8BY6@nX<*lXoKh_?dz(DS*4v4;~j>psJ~1Zv7Gi|H-;>t8Z;U@FB9Nd1kQOJq`&i
zqHQ7YD#u`?OpN4C1q9DN0FK>ipgn{s$5_V|2f>%Y3dOOqQY7j85RAA=Af(0N6N=e>
zG^&!5f)M5@23XDMO+L`gF2bx4@{IK6aHMEd)9c@W15%14LLux)(gSaH$2l@$*t~42
z!}-nkSVSRM8$;!0pb-2U=g*(!JoIw5G-aNv>s-7+<Ogt@rcbA9-G~iD2IYQX<P@U|
z>XTC7oJO`KAzR*ZSVW!3>52&CI;f2mndj?~waG{*L?6|;E1=t|OE(C={%sa->*#$8
znlYG=X^olLT+rO3=G=PxLjQ|dFp~t-WSPUV6?~E9=>fJ2PbylLr#!c7Ck*fnK*DQV
z0CNX;%k5I><e}Xb1k9oqF$;#Ss{ntVEFF4(O5iFO5dh+Ezslf^S%O4Vv;)9FTSwi6
zjQjfv(^bRzUam+%Ma>ChMe|_IzzowFHaAK}{i4H>ZVZg$<)Br2zD`5w4aQ&=*+hbQ
zgaF-pm~%ufuV`+$hP3j&Dl%>+6Y*$?xpyQx$pMq1D-F=AW@&FP7jD@hAf|YU<b-+B
zSRH{BN8P~HUuYC?ux49J_kEZNehu?4e^>qU)P9edMTkPjJX>Xv&{B1toA&0pkeFkK
z3XQ#?T1Kwnvz)Xqs(U#@29-LbPV+3eb5p9)A+2-rwy4F_S@#HqUL8SE3y%n&n#i_9
ze+#`Ph99X?EWfSaL)OQjhCGEMtA4hZPi9=|sd+IUm?wFk&_>L!Wt-~e%VM>gNQs(x
zm(Vi9G(Wj4v}u=WGwqg(4$L>kfFR1@fnIowInv}3LoGJV3Hsqj!YoJI8IrbWkA70D
zhyNfFR>5SY=YS4bAV{P?0?UT2Sh%+ckWe%<$`PQ!CMl-VXF-T-J;aOw)|oW~5b)pb
zg)87K1Ctx8h)vLWCe+=Ra;ZM<cV+fUwp7QXK+vT0<t;Dt50OQb1^c`a4qXP|dkxk(
zO_XQjPFah{CXg+)0ndH)<5G34cbXP1AvMcdYGrIvs!z11)&Ki=ABDbY+~v;cQk<&W
zE#V+cC<<}YW9awe6Dcl<X)3Io<+sVhf{4-i;q@v_n7dmm)?fTe%IGXA==@HL`j$Qf
zs_S;pCbbmyl|(&KswjXKh9gbwBQ*<m=qy`{Z-EVjhYuyB0r)6OpW(_U?|nFa7eLgB
z1V6iY=;K5nt6<i03S7REe*DG*Nc_m28~UrRORP#H32w_KB59!Z!x4HHo_37Z&bQ1r
zd0Xemb`-K4kq*d$_X_R|X$9j_&OEw41K-b?ub)7(&@b#h&z49wig|q(8MTxj^W!)g
zEM1Q(m;|i*N#fVZ5(l0zF~s~NbecX-{6-UaCI#Lp@3hI3PQX<`2`~Sj2d+G=q*6px
zD1-Iojb5aR=ms4;BCuMAMo1$+diCA#t|KRYj3IsINe?{u!~Uq{ILbzl2M45G`IQvf
zbB<Fw*y*pQ5T4NzhYMk>?-HRug!`V@6p^usz9b-5B&N&pzOzrdRNIMt0_bms!K9dk
z^DN2?nxw%kqHqG}=uC+SJn-NG-t2xJIFE|z!o)~i`2bFzBcg05`85k;)TalN&|@GA
zgJ?y~YH}@EME`{UL@Ec9f~h9ibs9X8rvw<X4)emPtCj@`1Y2A=9my)_QR_iELZIdl
zWWz`?nV?BVavc&r3OMLgdjb~&GO~Q=iGz*5GY@?-+t2Bq$vD%_8%sL_PV$CU{BaM8
zs&OEFUOmDBk4orA3bFYXbeq{0dd4Ft`Vp8rs$?lNs_Lb(OaqReQwl^2@`)94z0Kua
zo749+4NL#<R+haIX{!GF(ps6KHss~w^Z>nvd6w@o?od^M)DG2NSdjI};RGdjtPBg&
zX*)k%vD{&p^6@CzP|c;0m;Vi>jxf_?s(y9=c|$rfVW<HiL}4SqQV>_STeOu?3^h}l
zbcFxFJN$Nfv4x>K$bgm`apc~$;q;bI%hUxRmVWt#y~Men;!S|msZ>z=8k>Qo_S08v
z^K|H|D)pg<CUsy>U`V7eg|!w>hOa^YmJQKjju?J*tA@#Py)x2aDF}=l-&DfF%cVll
z>uCCgVD1X6Yym_!8!agWj}_Sp`Y0H&9=r%b?_*LRbPEZ`O$?WQBH#81HF8!sg7Mrh
z*@KoV$%n2;hR&x$$?jfXCqbFi6v-}qSETDWIG_8i*;8I$h6knz_J&~5H4<3FtshGM
ze$1FSjM2ddvl+r~MbRyf=KW%p%MfU008ECL^utLveN1RHL*aroU@x|_$GB@I04fTD
z&!qdY)I4&YDYE8$F6q_2cpC^9-j*~E88^B<5!@*Id1X)Cwp7wqk8E0Q`stk*wj+tl
zU_2mQ+!7Ril<?m#>P*J$GT5@H;VQ_Z<Ej=5>eOjlks2R3_wtaa@tggN(ufvoY4!M?
z>}NbUP)N=;1Sw#41@SPW@`_o8I-Zg8tP4;q5i8T5Vb7?3zFpyFVcZd1k>ef4?Rbcu
zi^K8Ve<ySLsYehx?VDoak#x{*YbEKDK8A)O<o&u&$-^eXC5o-4JiG0GU%amJNST!L
zyyKok_3Heuch+C|Yngfoq`kjZY@YNVgLAq!u}xXaVvM*n<iIRhm4WmAJ9u3lkx>`S
z<eC5h-%Lx9T7Q~r7^TyH)yvC<r{MK>F95hVCEwF)ih|4{v;R|A>~zHEkMCeur+oF$
zG^0UWX2Hr%Ly3V4F$KoSl9s<x5{q<_ffniq)J3n+l4*;eaXbI929VK84xvAO#YW*n
zK0bWFf1q44hZrV+NZM!m#~wXgGE(0WZSBob^0SvqILn<)T13*D{7H1no_t=X7Yv=H
zK9#<XZy`T;O~40Ndm(C?_FV-B0bx+pDR)g)-E|L6W+O4rmdCQ0iDg(H$g1@lk0b$r
z8DRZDy@L`=vq^Y|OsV*<U)t9xSu-Hj3-pOYw7VL$B0yN~maS7#Sp=f*PP8|rU=ir+
z1W;~D_6mikXJbzB<Y?XRl6X;x;R<e{#Yqleq1QqpOG?212F5@b|NhP&pLzZc|9I^2
z<9we^$IoBq!Y$D74c{B&_nA_<E9g)>wR|_I*Cx}U8qmeU%$cMv|6?t|^QYsa-TNWq
z^+NE|hmhSprd^wLC}#Lehoqguq}9MvTJiv{&Y6`2++J36oV-rI-axBvHlr;ZCD%F~
zQ3@temk`N(<=z`G7d%NA;t7+cy6?sM?G;|~6ml<K#)B~;?K3nnDGtbYGr2n9G8}lT
zVi$6+a7TC)IX-$iu7pNW(-R}Ho-cXvz+HeLbtpix$jUhw*fUnD!!qn#7kqC<E5$uP
z({PYtbk%YF(Xr^>drj_{k1sy^D(`O~E9R7Y7!=gA%Sbs4^cp>T@RwvG37DF0t}mN)
zP_O)gBM8`@(~cT-e$9D|dY1lVC@F_5ivcUL=uOkPEqCA&sn6Dc#CQ_<WU7g6F(yI}
zDJ)E?6yN*LB?!Qj^rfXmC~4lqu6k>A3X<+FB1llDKLCB!d6MREV5$JjgcwV(72P}B
zCHLCO$;d1LxtNJh`V0VET}@pHIxYR;LinU;>7)r{I@E!+&|z5R_u~L=fr3dN^Aa@a
zoDyrayC1^HwCA9nHy|}3{hgI@*e--hNH#9R7$yvVDEDFCn@AYDk};Wh_q_;4&R5WJ
zIU`;o5;_jCcOLpp2J!3>2WJnzSrCPAxGcuy*+e3RK8v8!F;S7a-Ivb__gF)X%OZAp
z+>^$36eF16ZdP*+Fb5pxp<j82XDxg>L!2;e-VY0Y!yKO-7c~IEkAXGXu!kD-V9uu4
z&ycCyPvJusMi-U<(<U4OD8_@331x+k$alq83OdQ{LL>tIFR&(N;?B6s+eLGNdPC>N
zZlX*;6wt3)iNL)qB@A1~b|EhqpkRy=m2GhwFEju%2a^X(+NJXlv(n+`$9H06Owk#F
zF3ypY$W8O+jP?p?K02g%B~Ao^Tt*|}{N+!iNFh1XhH+g|gP@Rya`T)K@y_B0Oe?|=
zY#Ob=^%gLChca^XE!uMbjGIjNY_i^)*D`3oo2(E7FX~Rnpbt@^<~(Pdq=W5a4i`hO
zK}Uz<0)fucV4M(!^$;Q6Vt|yL%97b#9zT+)+(4evP88r=J>>OHsxXa(_=Na*ar)@>
zY{qm?+dJ2hn7DQ4zPAUfcQW)xOQ{5jTi)Bhq6pG;Cw}4QwlE~qTd>mXZZ!i=&0<Vq
z$`gM{OmKg2ciD)q`Ti%mSA}}lZIjlBL?TDPDPxh_e<_qqw6>L&pM-#!1+Pk0sH7f(
zQti4Vy=i<juMol5+qm3XG*piGo;lGRX`6nycOW2ZNUPNX+F>o-ulC$5NCLxN#)92q
zSX^RmL>f+*V3MX@i$9WQ+Zcd6g%;F<0!@Ki!iqV-Ms$uwU*Yd6gH!E=(NiA+nEfQE
zvXDv+tCU6o-WR8J1(99fm)-_+fh4{~k4&^8*yTP#Xb>rTGhpj8eGRSir9A8;XL$is
zq02UJ6Y#(JHjXFi+hK&Cun@$*X^PIAfXOd6GC+oP<*LK!NXRgH_>Qom)QCP;@*!Q1
zU|brP&h4a)JIQwUghZ4iGwoL|e65Fewcsr|waY_&^1Q-COzstL+Zgy(9ZcCpR`M9|
z8H{TpX3BS`u<t%7?K7E>Q9!ol0+~(b&N6r4mAB97oY!`|C<{Fwv2@7Ka^o4_1=vOp
zafIN|OasKV3vV}#w}T&9RqFn*j|_^r^f^Sx+`EI|W8NcDN=k=N13ep&9R8lgo^PI_
z$uU2(UAiI&Pvk=a2-v|{K*kg%v<78K@L2mMY|;kQ0wmZUk#3t)IY1S^q}Gib@8Vg%
z1H7^$Ye~IVvZ#V+x*ot6i00_5Aw=U~QbYIhtZ^3Q+blp<XM2F!@#YQ9npz8lY{mT$
zYPEbE6E8wHK-!~{W9AoeozyEjI#<J~bu*RCDi`IqeO93qXF|Wl)ATagaf-|#BFVh#
zu027A;tKzx07K#p!Bvg)HpgaYC?etR!l#t&OUuI3?ssV2I<;RV7iG6zLE7fMZ@=?@
z*Wpxsd*3=YsF_vgFhI`P%f6?RGmy($Q*no1>mlbw*Qs8hjQ3uT7yUUxHV1|cl!f}4
zB7fV1M;}uH2r+rcSQU)v0zYA?*pC&doya{`_{+*DDIotQ^DQu$>ANAql=N8g(ysOn
zpHb}6SUfIqTF&ddR4ifC$fU~{8tB$B6ZM#7mHbkqo^5|wV$@j0WOx*L8b2%LeBV1C
zCm2c&3V784y3{(=ILKKPvAU$xHHLAX&mk{?5O(YQ*0_OBuj~yLAmZzDE23k?8ujgD
zN2?AX4lBjjl5{BLOqUiDyJrY>rOh2|1wvKIGwGLogs8-r`JLgJv~JjRURR$+@mEd+
z_ROzFZll(}#Bbiw_c)JAp8SgUNQjeIT-;tP+_SSiD2^qHukGoxKLX-PZLOzhqX#KL
zaSszo8W=~6pD6CL918SbN_?)*Y+xwUZF`T?RF?y75PCKtMAn+hNzO)?mrjG|Y4^*&
z(-Ipr45RC6p7n|b%<&ZBn}~M}6J8MnPo>6XuduKhi{3*$II~90H|Bfqu_^!F3-Cq8
z%||$>g~2rhkLQ_iYlL8RTnX7ipMN~z!(p$7kZJGZy`#qP=WQWo5{+XuYTti{?8=PZ
zeVWftfMV(IO>pjPP1#|4CI3|ZS@usJn!ACQ=N#_sl`&t9+G&|0(t|jT{*w^}00^qP
za4JD=n01FU?+g)<`s8dU^sK<Pr`r4ak(Txs#kB=dmPdV4rZW}?2LxVNuR{bs&A(D~
zzV*&y3Kgy8qI&Rn#aDngr+aI~sef>j027sazd{dkaxkQ0&P%hg7=568_PJYq&g-G!
ztjtHG5#X_$pxvk}(((e-siFsi^n7hVn-h2&497lbQ{tjkMy-5`$-Qh{fB{S09tK5g
zb~LDo8InR`_#<#7FKVd;OWFt|JsQX=Q^r~Gzwg5ZP!p&e>E&N`dU4R5A+30?!6;w<
z#8_eglTm@--YwZ0`@OQczYnIU)F=p@{Jp3N+9&LcjEV+zc=dTjJ6GQLPbM(A5d>bc
zi-wII-U%qVv`hS-Og_GOP!UWXrN2N!%%zK$&M;6%4gZ<bm374y@8?&4yOb%i+jY*;
zxqU6DxbOm#^8SGAFT(FqP&5<ieDbF)G6<I>7Ilz>myvCweI?7aXRILP?}mVh8~J_>
zIGwX?jYwvuKgtQbrm&8Exj9Ivlv*cgaM!#9!9(P>P+=;e^3vEZVAvP=rnu(+LHN%^
z#*0eOPRGRp$;%!j-+}xLDeSBz)NV5RlKhyG{WvpRG4lz-V5vKXlvPS0w)~{OSc`qI
zW5my79{23o=E8Ay9Z4d!HiNAFoO$z>sIinA4ru<zRVq|<Dqcu`GjSr#Uc!nSJ+o!Y
z3e<j?`?Wk=jP7>)ZTH^v`A6m?jWlWiKOvBzf2`R~4f^9bxDgdJf7ZcyYn(_EZVQk-
zj=U+{qqrqxn=u4|3kJ2$xox(%C>OCw{0*D>|7PS#2p6$}&VLiwEig$7zmA24adhZ7
z(v*nMsh@Yag%{;4h4L(~$-DV%YoEBurwb=ksG2ejtEnvD2D}d>b?a|=I|`$xBFW*#
z_l0LOtNutI2xhkZK|z-{^fFqITWl>Nn<0uy<)+@Ht;j34L^;V&t&u@A)tmKPy%WAF
z3f2q75V}w5LW9*>WL3!KH~stVIY$50SF~l`%Lcn9rKW~8?~Dug-@V+Y5+BJc67{pj
zjZ_ih{3id&=2vbrB1Ma3AP@0|?Kc4~;eBUB`yl=z>bs}-qN;YOF<+1v!^lPQ4lZbh
zZ~{AO4KwHqf;?!9<eb1CwlL^dBTx#KM1a<Nv1?enqOoC2OGl2FpKo$SD;C=&SlHTA
zp7isOf(=Bckw$ssX9_=%I%;|zFpQ2obD@(-h5tLBv!bvuXu=We4~8h!qfEY1n!Fd_
zSMKu@6!^mQ)V0NN)=#$lk$`U9c61+;R1f5mCwxKw2!ZM^@rC({?`SQlFqFq43_+65
z9+evJ-v%UJ8KhNcBg0=}yjV-G$6Zm#HlA8+D{x-RRbDOw<}2>iM?ud6{k&X>2tC7X
z3?Zlc092X0WAduFR3aA!^P(2eq;`!&-G3Q)NOu*nOsjrRsoPJ;by2e3aH0Ax%14U~
zT+X+6Qvbc&qhWVYI4><!jT?<D!Yx)lpStcY2@&Vp@;f4~GaoABGC0cYSkM~KmZZ5E
ztCRCDTi)~RkIa^jOH+C9PNBn(ELHRYPcLN~5VAoM!w0+T<v7V*Q`AHq?(dJJuG?Pz
zZu;+xj8H_~V+_mKYsn6eTfMJga1)_<zFfXG>M*!l<-2aKLbQ8>@bcM9d<<gR5FIX_
zr%D!aSFRU0!tfG81ySn*rEF?7((di0B^@2qQQ(;GI(lE|^v_)lR)zzFa7t==>zE(i
z3zzW*9m%Wrga_`?;=k(Rh+S>Hy=I~rntJ)p7$dORAjNTt>t56NJBD6u%|X!t%dwrY
zH<F};GWLyeG!o)!^D=)36EhXTMtv!u0HN{Xv9MpfLMkr?xLMOfs&oG3se@%FL2$$m
zQSD#j%CpXj7S4NlS{L~mnIE<D6q8#VlhxP)xE)qGr$qg8a2oU4fPpEKstN~HuepC8
z#74rw@y>XZ5C54)5C|dj)&7(Zj+Z1A8^iQB^9cLf!U_K2>Pd)+-^Z|uqcr%SvIxC@
z@=Vsww7Djxy}3~qif$M>=FyC=a`~;=`Z9RZXh}`1IhS?Y7<|8yl34kHc>l9e?WHHM
z5Vd=7f1Ua+UJH~7TNa)SR;tN9y>C@9?FB(1W8u$SW9XCh$>B0QBSJV9@@B`<Rq8GI
zM~q@c>7xg$t?qoSo)0O0&JG<hr?0mjsTF$E5=o!2{c`DdqY(FNg_4VcTL8J$HAZzC
z7TnmNFX;W(JRBuir$8HDdRHq?vEcjnfs>H$@9#>-<nDUsytdOD0wpUU+=qfnw;NBy
z3b;em3+Iqh1E)0To#AYn6n_5CmW}oYAuS8{Aj$F&hkSapxYI}Yp>DRViisqVE&BqU
zn0QvMA?*(z8zK~vdWF@avsb$Howt!wySs9m#LV)(zyd05YCnan3z|Be{tf@{C0gVo
zU^Ntyrtfe|)%4qZ$h2B5lIbfNZg|?=|3D@%!5G65Z4@l-&gh^1Z6x1a)A_{~tkvx4
zWXQ?D$M-CXgpD51{)Gy&XIy_i*Two+0v2#J6^TqL%CBG7%&DpM0Jp~g@bwwF2MmWH
zFK`%{TzJ8MFR*Qx<Oa%8;_$f2AV(n$T!!66*=*4FH+|_d_6o*~NFAE}KY8sLZsa1L
z|7@hweN@N=cEI{>#jC+>qTBBai46yf8?RseiVH6IY}_pH^Jf2K4tSytkVO&ri@<uu
z4BBXlKN&TS?pA(Kh))5hv#B<~D{|>u;-4^DvBT~gOo~1T6(b&&!d515|Is~~_dk-)
zG$Ct_gwcKZ(4y-%RP#hUsvGfKJNBGp&pB*bBR@&-+oiW$b{cSmB=k)PI$x1OroBqL
zf*PMMv$!N#hP+VPqAXL*vhKk{))FFDm28^t-yDJMIXJ-EXnb*55Vr$9p}&Hw1xx3d
zV0}GT58zJ}!mmqWD0t;Ln|{-80Gd8i(2e|y5=}Z}yG!BAk$g`t5N-P;_zHOKZZZt<
zEx`g5N>)ow2KxArEf&R2nwqdHu%19}CHqpjavz1;7IRu*qq^x{$<@6v!uYGh^gl*{
z^iau>7IPx|E~69b$bnBfHs?Ngl*3B{*?-{4Oopq+`VLH^<EmFyH#X_@OEeULIt)xv
z>8_zfWLxi`yPk@gP+f6HMHrQ-hoZ%T0S6)evjLbEyHSJY-q7}JqYgkqGXqLM;*;_Q
zX>fS5K18Fl1rs3*S=9q15)QjeFniX3MYD7Bwm|cCtl&I>v;>{AT_qUvkmZHXfR6cd
z4g8ta^Bw58551@5<>mW<x6?rkmQ@dSQUWad$S#|jk<d?al4241(d{|Nr@-Gg_Ye;X
z8of-RciWy3vGe_`-dK@J%T~98H4hJtrrEvEb3S@a@o6j=aXm9m7wcPeu3MeQHzI;m
zTW5blDAr`R#){+p!Mai#EOk`6nAI{E7XpBRU91ZeAGrnF=E?brG&}cznQkH^HZ2MJ
zm~o4jBh0G`bcY>)N(ChVR?-$L0K>mDytKZ%ySopWycK*k58l)2$n^X!b5m9~^5?LO
z{l55L?;S~Sb-ZL7kd4gm57@fCUYMdeqsY3Os8``{e*XIxv-YI7b~h|(8^dC?>2$sS
z<D>8x?^t+nJF#_dE>@L$tyZxrT<v$(-7lq7zVrfG>Qrh{g@<Wccg50R%3O9cYSd=P
zJ`%njnR<`S7tOdoe}INozY}7&D8c>LI4-8GzQ+i#Z%75#SmY-suvsyScy4_$QQPfd
z%_Ab(j1jjhP6+_rRcQg`p>eYRbE)Z3&O1U1Rlh}4dxvdJd4R{I09R!J9elD9x-xp0
z6w2eV#;GsNt$vg7h|3`;BGojkM|rR<S%utcn8Kz)iwWz=u^^u`1lv0_=fWYW#|bzO
z6TsHj*FeW)`S-{Nf#h)~#+q>oCs2*E{$lJ355ib*l&Hdq+~td?^(HLnfw)X&`0s!_
zIVvvF4fa1DZ{9t2{xF~)4c?gl&o#t~k{%%m?dyG_EA*InI899bf#Ymw^0$}9t<WwO
z>QplsZFCNOx-lXF#k4HOCqAmlOMiBkUHLv{)1l<0rYP5izpqL_Ux|@A2Rq_qB#8^u
zByFGTwSi?pf%pEhrx+untcmXiQ!|babQU~ozL)D(m>36M?JG2AjnYU<4@1_e8o?%c
z6<|hELB3-qmAI3RrV{M;tH2=3IEO<eq#1@!lVe5s1aqR{i|(&n!1ZYa*($wf1I);~
zg_%q`{C|GH6yaaN8kZi7q^C3HFI{0wk+vViY3CEvC<Wn@@$&Gf)d48wQS9bK-<SFY
zSVKe7pTs~6k1$FGdU`9(+jZ-aj`J6ehz2&l1Qyo#r#b7F;*U3P3@kV4PIjLeQUer=
z#WK++z3(m1Y@V&(Jbi~;=8!MP4XD;J?wdHa$<_Gy+Xhl^@27v6m#z3>O#0oHE8puK
zPNosyccLchRokeQg22~vdb#8tilr=qZ@^wYoY-*p9y%P5W}UqloX3#P3d8zxrXc8t
zi)!HP=naF?M~neRG+0u{1|W&3b&zENZ>IB45JRxSMtVL*YwUFD`^HEsEKYakK$hTP
zBx{!V#qSSF(;Lda=n1UEac(4V^+n39hSdjowOjm!kT$Y^y()X*cLYCs>xkc+Sq3$W
z&3IC&!_J_xDj%~FEnk*2Rtio<?XRs}K=+%Vw2S7>slSuRwiBBJGEO%g8}lqN*i5RD
zx8@sV#4~wh@Mwftz2?0?voF->+<(AH-y7?SDY$dP4u=UVuI{fy5;@KjtBDMy_s;sW
zq#Z7!S@^fGFf&hPx;YOrs6pcM^P~<Ii8ThsxLR1{I^QB%<0X=w2z#c4{rs@Hz4W0#
z>F{uyP3G+(ykg+AU&oZ|0k2GEYWjpXp`XIfgsVvN(LZ^LN`ummGDp!E)aH5yPt7$a
zuK`^}i4%JQ$x*9g3?Y84G)nvKUBiCn>RG=Gi-bfW1Jfsk8q{>n^}(m20TB~aqTWA9
z%ZFp!|3KVGQO4^K&0Qi}ll}{7E_DBKnf-m`s@PDFoGHO|Z-^0YJ&39qn&U+z9fm#i
zXFMismz2HdE_j5riyW6fD|?}~Hk|8nnvfc*yL#~AyVci24VB@{iqw)8AL}U^l>(*j
z%5H(z8CBljL}yOgM0^!*12y`M#hUuM)i(Rx#+cz6c`ooS;?vWVGGVWGp=APa;^8(c
zOq+ZPg)dA23}I~?$B$Fr-Q_ysean3#gOkc}Zro%t@?xHsx>r7IGWLA$9>v*mwYEr8
zfxnn_K2!9>>z5odCP6=c<x`#vWscr-*bq@zgxg;RluU?uTA#s^R<WZ<IBej|1*Zi}
zX#`^z@9y`FX~i<~uy$SGSu$#&cd;k_VD80X^WZqXRwcyb&FP8=TvzI<u4QalZ%miR
zYwg9otN8tF>%gVsbah-)N5te5gA`?Flktm9VfDX3N|Ioz_&beiBxCZ0mHy`kFK`Jv
zCV~GRSdXJ|KX1;|SJXI8jjm}Y_{lYY`Em;<cpM-5kUtTWwdk^LshQ5=5PM<-Z$<s?
z$qXrw?tZCnCoNrJKc?;6^0|0xvQ166-i0IcHu7CzG--eGaS+AsEbqa<ZCh&DbPiV#
zjdJn_8~=HErQi^ZlYy+M+ob6SDoR>O7{89gG7+a}ho5_0dvtXQOZS&sMZDi93&@`I
z81B|+?6zJ`+IF0;T^}Uv^Drw0>YQDbDqJYmW~?jCx11NAI)TEy2I4CCk!}+^U4T@T
z$BM6vJ=t}p+{;kqLZ|R2h!w)l0;Ondc5dV2bsk8=EzRzj;!9Z=-v%|?b<sp>9;NQU
z8Z1Nr^|;S81#T?04v+o*0gb*i#hs8<yCC8?5Zp)-Kb2@5Wy<Mu&0Q=n!Ssq|{j7Et
zIS0+UfA%Q)XJU|ylfF|L2z@&5yU9^%++3%r6We50LajY}3Gn}^$lOQ-A;Xr4c_y;m
z&STohxE6SL4Co_Ry5_Fps34<Jj{e}C{(|b>u$7+Tmrn*4&Bs3;@i20YK(j4~dxGpc
zL=F|RFK`P@Z^;-<Z~PCIb>dYV2S;?iK)tC23qD$vE+4$XI^Ob7@uD~ztsY(as9v`a
z$s3qlwHl>dTGGPfIgZ)5anh`+C3?%a_2E_0>P$s(f5_=9r+W%}>S;JmLNWD@WhMs2
zH;R_xw8KPSe9%yp1hK)*<U8E*+HReZBOPS;-n{;f6bsi1lz{=b-wb0HY=}<c%DUg2
z>YW1DXHG2UR4oo$ddc71_<ExjT3T8>FMqtR;rl7jcn2K%8yD#}N&XodgR>&u42&_X
z<1@)gt7u7fWxc)y;xW5j0j@mdwD1RfI5^yIJBnCc06Ml~ed8WvzVBxg4tHZ<PuXPE
zbIO^z?$FN#&e!M#G!OAjOdvK;dMvV9bcw?ybt+0yYG$VyQ?^GUrcdIJcGxy2%DHpe
z2sUd)tqx}59&p+B6LcxwP1;@aYoX-bt<2gojgE{w^nMuWXe~}KPLTAlvD;fNfP-Ra
z1DLgPpI#k7Hx8Fn%9PUSwRnzt+i~=V7<5t`Q-6x!_raWT83zfha!W6^RRqD2UD6C`
zIDT6GVMzL)g9)Y(YW{~0PA_Lbip&9P8kCx%$#sKXIMa*MT`r01qoe>N@W@wL^8kuT
z`sBr62wJ|UH=Zd#+cx9wGFTBwD^6}8o19>(zc#|_7{c4Em6n=n%U0t|0tR2%eNgar
zd?;YYx!b}mnZlvZsDPv2TIYu^!>-{=Y51$mxl^42zH)8CQ@~{>l>5l+anK-Jfp6@g
zRgZ3xTDg^)ezWf?n}QJD<qqY&LgULP>QlQe_i{SQaR-!EqCycx3D_|#-0?vA)GPP3
zjf-h7YeYo<Fq98!*btDTB_PBM9MV(Im_$0jlKM+~^Z!!B|8uAhj=r;0pljk&$P`+#
zMW}25JV<3WaSHezJwWl9cGZE?z2!%!8V4NzOku49{JjiNh6%dSYXY8?XiA0QJ+fcD
z!P1bcL6@TJa)MX-0yD*S<z=iwrE25B!Qenh!Ltn>Pt%YdwL<0Nhs?!XlTcVn${|-<
zS40<-GTtU_J@oWT=kx4B*OB}x_g>G&&$ccbqucVEQtY-D%{xVgj)91|Q!#e3S~~~X
zSM*}N(}DD=I|U>7>~<uYlmbr7gv=zhF@4V+trTq|_N7yvGTqJKb79)9XqxubE_;|i
zc{newU1;E8_xh$gWwx&6OD%=FeyQGukzaWVnTcZ@ZpzwPtJyc{(m=f^1a4~FaeqI4
znP*>uWn|*Zh?TQY!vg!hD<WgQ3M=_8J0QnU{>k`b#@p}>*&{7+uWc?LUD;SF$h@h|
z$-|fsu7tVg8uQ}}g?V8rz0~8pC8g2z<?rVI(Rc2nVoHS)Xotk+FClnOHYbKj@1CNY
z1*x`NxibzWN_VL4pGgt1N(5u<ZuSwkW@dWk9rQ1HsixVT^?lKGrzT;9gzf31(6xF$
zB_>*4jl|fg-7P1sTL@T?(XN-0Q8`hkCUZ==ec~ZCg&m_!t_K()1-nU1T&g}e#JtLO
zcy&u71sXjncDT-{WA@*ehxfm|Qn8x9XeF|5iR>kc8(`S}*k2p_B+FwzL-MhE!i$#L
zY<k!OOh(-`X)KEuMz~_d|NUF6*ia%Ca3}>-0U!=zGofX_hw?gZ-4#GV#Qp7q{opff
zg`77`34tO9)gx=X7tfmC%oc{j&o=}*1*C8=z8i5Qk-AhDK8>S%J5^`H=5ujSgO!`L
zBl(y6dqZRgST<`+&epzX7Vk>t{_;0DKV5X}iJ^GGBs2|;$i-!-NDCK*h7=e5h$7-~
zqG0&bBTFrm37B1CNVVQsr`%5?jy9vl>i-=~+{jQO<_K(rfptp>AQUkhy@1A(B|Y7+
z)b%74@op$fS~O89PNO(=*U_}ndnMa{G4b(%a`7}#c=0Ig&F+gWK1kGIcSQEbRaaEN
z?qXBlK7CybB%fz9GdTe#`sct*+A3&0c>scQPD;jY{^yBMxeJlI_`$SLuraFJu%kKz
z*iP>ZJQ{H$W_8tvXp2+n17u)}e}Ml+{oj<4CaShP?PGSmJ$nlEw9Obuu4!Q!(l-aC
z^w6P_GrWO3*2OvRZt2C$E3xlmfIamp>xwmE?-kc$++Qf?^;NK4mqCyOB~fBE+O(}n
z=eE`#E;0z5NT3<-2Cx%XZ8>?b0#x$Slq~r%@CdGkQ9AEAx??NEQN4;*n!k?aP`X#~
z4w%@msQZDz%g5=sT~SETzV|QNxi#z;_|xXbCjD;uV!w|GrE-7~!<x*fxa`0B%!UF;
znRgu&;Xtw2)o#B*t}vnY1f}V|E-Es~v4TIZynaCEJp+Qu2L&Nfbjm~;tsUsqlXu-2
z1IK9+8(xM}nO5l#F;}EavIN426_OG^>ENqhl42SINeV528TvXb%{QHZrp1<A7I&Cp
zInEIo7`^t4TO`h@1H-A|gT*R&Dy&!3wgjJ$$?R(oXiLnHmm;g0%$LJ9G(dl_2vmmY
z*$FURYqp&&V|$e#A@TyJ0>Z-6m?Cn0;NNbw0T$7CIj?}5&DipH1Qn7v@hE&9wU7OW
ztyF*Cl>^Arc}1T*5-q7!&c)Pwz`v{w017Ij%=P8qYJAo;S^xY$S3{b1yz6-i7g0Ln
z!eiq{h`k<j`EMRqbKeWKy6HWc6Y{<tERqGmFRLe3rKhtFy1NX$Ee@p;G%F09WC}E>
z;*uKbNeld4g-_zB#SB2w){DW@%FP-#F9hy01I{-b!Im`V-e53H1tN*BVC4WoJ%BC_
z_|BFvT>w=<|GAlGVkd^V?CQG!?^RMxqnAoF<{zQ2cBX4pkUs6p4tESlC#&4y*L(Pr
z+C0uUNWk1BD|ctwr22UVm{+GxT)b=+X@-I(jJ=eAaD)Ib=i{S6ArKsV4379PpuF_I
zmkKEWyHJ7pkula(&X-!5M!riP|JUAI|3$q<UBieYf+#TvC`iiCAxKEc5Go*D(nv{z
zfV7~Lba#hzcZ1Sh(k<QH{a)ia>V2R4`4e7$@-vPzSIza^d+oK?8vh+@0d}(!1T~Xh
z<C$pL7I!cWjN}JjL$QO<$2l~+`KqeDQ<v_&;p0zZ1q1z-9PQrUFgsL6M)22Yvb-q7
zly%#ZXp~o5{mr=E(iy#TV>$zQU^rlFk6MODsN1(gIm>UPP9vn~cSJpj*UEv+-u)W8
z=@x6@aWHlol``YP>3X0y*v+nlpmkFVv_I6t61|K0PW0vK10Gk~XL06?{$JHsfr*`o
zFl|!iL1k}fWc_?p(92B1nm9yLKP(4e97R)tSrg@E+hoOH72+lVD<iyX4%Xehbg1~4
zqcPF)FB$_7b`#VG=8`nPW5D!Yt{ct=GX?20n=IWMf;I1E(=IJKGj3UY4@GI;v~F;l
z@h1DxPH>J9o8=p0e16~hr(@&OCEo%Oy+Cr$pFovITacej5R1(-IUgXE!(SBV<g0<f
z-$rMtIj7eGVcH03OdUD<LH#qdQ@4_&WYK%Fp0i_Zw<x;G0Eon164p1;9m6qU02l3y
zVA_rfeq5XJb;(*5Q_J)kvU2e|wgUs6LRFMZZWj^CfN|J&L+`d%Rw$!|#6(Kk`(LT~
zH^Dlprd}DV&ry}L1<ANeIf--R+%Sg+O4qSTVg@ImYo_0744Fyr=Utc=kLOWpTW$pP
z&{fWHcSooI8ky_KB&Z_BUrwXjo+1Q?&$X&;9?;r(*<JwNXCOs-fc3OnsFlVq>ZHv3
zuP^<AH<FuB=GaLb4}$a@0FhpHvzJX`t0qwE&CS8mL#OmB8g)?%j9nc@#T|``ZJ;V@
zGB7~aFujcGws`+}t@9J!IM=_=Cu(Fp%o}b-Qpa7n3T0BEVm0aR_j?K7HB!>}nKyc#
zve-;btLDBZu(-5qccR#nqYl*Mj2Ybf2s&Yn^;dw+!Q!N4?_n%{zyl$}@e%{;34}kU
zNPN2U2(o4*^+zDsI|$`4%O{y*1J=pK>d%|ZNTdA&$t48U=Rltj)$4Hj-L>5Xayv)2
zS|^Y#ST_P(TNjAEC}7=gI0a$i_@?biO{0|L>dMSL-+;g|gj@~-@NYis3b=>$uE(q2
z6exDG>sWV3vssMkMkq1qf)=JtrvgCpWTLbN_Kl)IWK}jdSD)}UVE&!^;r+x{NKgNJ
z^F+DAL5Gb1jqjx4Eak}7Ye6VernHPd>=8$06xD*}WA*)`V<VV`?bcldbQ5`L7o!NQ
zn{A&+&|Sr)ylS9F@PX^HxLGLY7p$iO;$#{UDPOn}Y_CZ9P<vECZ>0bR8QUGQkBUHU
zG6ld^)|H^;U=Q5Hufa{23W)4Q08?89kmyuv0+6CZ=d?!{z5`shf%a2TEg<G2GcPXI
zIIC5Cz@`!cUe^z~^8lMh1Q2K#&=>&d{iF`)%(j2NadU&S8Z?zL)$F+3r?HSh@~XRh
zI9q7B)P=oAPGY}87R7CocL8)!;?rUYgFTq>d9oI8$z@!t)<EpeYl4<a#9<oC5K}*j
zus3Ji{(-okAP*zV{qG=sXY-FW19-B<Dr~^uFC`RiE%bJzUYm08hYyxcD}Q`5Gk>th
zWr_M1Liquykc@|VsM4JsFF-%znsCAHaLVDAkmwNX>9qts^R8(%S?G=Y@N)AQF{Lk{
zJ;=66tmZhqEVklf1O?-hR2(s7m3YG!u!V*@nMA>Pmka&;cK|%87tpAE1k9Y#Zf$`6
z;6bdP#f^Gxj!yb!JlXt}`-KvQ`m0WV(I_d%4==bE@@&z%RNy1daBg>HEivLb28QHR
z^JW)|ndMKyk!ZD-CiBlvKRGPlpp*|PoqcMD^2Pav5CxdyTRb5#(zF-NIf>+Xp60k>
zbk|khXpd=^N?f!cYF5n!70IZSLekil$YF^(u)CSp=V8-J){-17PX=~*F$HHnW>2qG
zT7Kh7mpFX%y#1xafOe>z_CC8Dg#xx?2XXvJQqXF8ni!hSLi4~N4z#%AZkF5ARbx;q
z&O*1mH_UG}@T_H}@~&%ZfvO_}l{xo$xGEYl)@LLLDUvfiRbPfgQrO5|%e-z3^z7>t
zuKIYL2rtt@iO&$@8SjR5^BS9VlbV^mtfr9ta?wS$48zdKGeLK0L+?9YLhjLN=lAc?
zersWWc}-n^H~ayzEf!8=R5cvwpU)43LDq(T2+h9RcT!t*H_{bJy*U)hp7%yZUt40P
z5ryw*cUq6Ce%bIwS8Lk%cVd%`mADvP-K7zvWJ9&V!TYO>cU<8Y%x?82^i<(U#7#|m
zOQ!#f`v$7c4d<dcWFxtB`!p8eH>$Ad$fgO0Z5@YlgB9qAFx?E#)@z5ibLFebu#S7q
zIi!JDn8`Bzh}IPV+7idl2q}fSSe9lS3!bytprPDA{ddGa;^3JaOA`CA9`zb>I$kfs
zZ)`?_)<d&ICI_J;stU5Oe7lv+ho;7JmsHzp(6JN;ZdoT(9Rg>9H`?IZDFn5W3DEnr
zj?#OEgrKJz&D7O)zRdDX0T&3m&#UVP&&iPf{Vq5cP9T1+ft^<b&6C%+VORaOj$EY$
z4ib~i;SOHLItjM(*bY;>X*7l6>|bWDoZT@|Qz)DX@I5C?%Hs^6lHDp$R&j~3>^Dpt
zy6$PY&-U6LV|(($=>>nUJT}C;#yFmVz&0dYPsoHl51P&ET;o`5fAr&fZI0&Zc2fDi
zQ+9!?nPcPg*vd9to~a5tOFRgv)D*-}j`aEApzJW{RobD5t#5cuIt}XfX?FL()6E1t
z(lEC$nCNmhK~w)^0J~ux#@&C<(hbxUoEz>6Wd}l9rg+tzp|f%i)iq8g;)k7ls!Ub2
z^6*`_-wvK19o*)#H~x|N;_25MaO?PFcuuzhMACsHVh#u%n*bX$?DOm@8EI)v&<R+W
z@af8~HXU41Er<|`?_x)=39CerN(BD`kYJC1#OP>~7mgar$WL=%mE_-@z{gkNwMYxi
zV0H!?CVb<Ct;9y)d2T7TOVv9EG@eX69$9#*S$4AxgQmsE*34ZWAD`Tkl0k28@7RnS
zEh5fGf966F_nE6Tj*8jmMKXcrSQeZ0wn+7;NSsR5D%{Xcv!W4H_|*0eSR2G~P8j^+
zN!7}q?|i>G_0bh^V;*izP1JreTJ!ZQ1raOvt3IF)P&WN~(xQ_=0jNS6^(VU%0p+W8
z-lMb_Q;SxEigAR(3&KAObb&E^>=`Z8);9Fsxj}zH<NT;s@XcfXUC`H?j<EAEjZIq8
zrD0+Uc;Na#t*D_v;C%bj^?D0A%CP~m)FIbOQkFJj#ZP4J#ZcMF(dM7#e+~qa^9NE=
z$#F^aH@338IyQ2l&=cR7=dF3RSF#aZuT;yXlKpwDUcSkmhn<L&4CShf;4%5fW=rR{
zCpevDngAt#!Z2a;p?R~3=<Q`1NRI_OU2(%amx{!c@{oGz@w7v)xK)3eIN`E%$Y^jo
zUAoPCM?`Z+8JJMzW>&<Kgz<1r)FmicQQk<;WLe-f24r6Dm)ArfgH5y91&uD8H4OIr
zU7^M@mMmv!Mr%`Qd{tYZ9avMvq%`bX^^avlyD1;vOz7xd{cVyGu&KpAn6JKYS}rL9
z*M&>@bPiB=Tv$syFWu|71|`+%b%b!rESHDmTQae?!#PTfT(UDaXL!DFvMuq;H3GYX
z={yejJ=g&L0T4oe-`Rg{Qryjc1k%V?X4cb!hm^pneUcp<oskgfRZ{lf7aEc?D_P=`
zfu+OOVw&k=#vSj+zg{k}Uz`w$w|{I}^x^!bR34*V9)(UWNHX9*;4NQv;nHeysG^~y
ze?*Snt)^wfciOLTm)OKe&u(8pA%td9M%or(<GWu&6Dv#1=EW7|mTcjuZ(0w)ubI8Q
zJw)D}MiFc;`Nkh>ll$lAo0CP?JzMya0%FIC2_koaxz0r0*&3bAC{T#g1%E!-tM5eL
zBV_*|u-ChlYUwg3wf+O`j`X3UdNlT|#`Rf5cETEc|1*apV<K-8QfC4Omdv0*!o+E`
zLxQ$JheA;H;ZDtp)cHm>qz?$3g0p6CznVV)&xv1`)~aM;RKyhP#EazPDt+!!>T>3E
z8*_5$+_m75y9trQ_i8M<s^4WR#H34*uLmXQ(UC-ARhr-oBY-OIjRIGP5rkgRX2w^4
zi#H+E!R^`U2?*Vc1zki}>1N4*#8bQ1>5>Us$@`#4^cYcOSFxtL+)WUWVneS7&6W6*
zss>8a>9>h(zaNe1ku3XdfRei-_-hZd-0Dw)(Zn-PS3ddp=eXqAEEzp}!iHrK<3_Z>
zNcdg6`t9Z&oURNH{`otH?!kIeyd-+@c4_TCp8Wq_mq^aWWTZigGJz2cLw+i~KaxK>
zY;EKXCy7KcC|3P=c{IVWRk&K9-JQZTkN1?d6<s@wB@ZR!NyxLkP0z>vJNQ%S)3t1@
zo#hC-E|^V!)wTN8w$GLEU`7zwj4tbm&EAQ*L0QRCU@I}hK4JW{dR^H1#|tDR`V$c$
ze!0AA=#d1|EW*|RbS`w}n+G6d`M<3-=q%=VOSIuU|LiUbfkJ;yfr8pd@hsgrNct-2
zU#xWpQc0Un)sK#drs~}MXO~ECp;GTPZ%0Ww&q=x6CF`p&E$mj4(R}?TuV*~-2}Q($
zc<uGVlyS>NYS~2iw}7PAt&|w@wq#N31D}8_%YyiBkZihbNOn4wG|XHzfySftOLV+B
zRK}iz<k|bTZaZhiMk~A;+lK_ua8z#KO&-^{3ffL{K5*L!e-+PhnWy-ATwLc57vS+E
z_%L2(GGS$N7*2tTd!93m(xeF?Q22F{Nb4b?%8p%;%;wu?K=YeR^8o{fReX{0Cm@f=
z+$*njJML>lT2`b?Nj8<b7ar6^l)@bJzP8kE?#wo-1=QNFW*Z{<hC2g+Ysjl=X8sQU
z413VGitit>SZ1dbG<4K5n=AQ7=Vmgl`j-Ksn22;Mc}wp*U7{n8ue{N-UOc)JEFcOb
zkb0iv)5xZ|MU;~Er}fFsfMRf}rw((!_Ekw9vzG^vbtFQ{<jUaFS!U)h)0PND8C_tI
zxRel&vyK$TamJZF&-#dAWtV3)B}7c*iF|_o02{IWO480dajzsJS|yA!O$c<3f3hI?
z{r!J8hZ7P?ez_1Li4ouz!1#j4IUOIb8wXF4{**~WG*Jq7z>3YEZ`vx`A~%aTJ0j24
z5@%gmR5*#=rBHot*oH}$_6@QhsL753v%gt&cxwR3_F88AWyR8JpKfn%NuPIM*dulQ
zO4#&hmfHujRt9Om&C!~iQNM5JRxgH{ZTxV((ar8VH+-{4d%NY3yH}}>3EwIcr!|{t
zW8j#4@4BudJr@P21YALQHK9B%2tCW!5|2H6x%3b^MK+#J9*b&&+bKl}h98q-IXXy^
zyHZ3yK1Ot9AP{O(&f!n?D74uAEv#9M%b-)B1T=>Ct5FT_7M6*O8XEGzaxS*2w<q@%
z1^X^~Xc>$YZmL};TwkPDd$?t2yyX1K>fo~W*g2zz+Kwgcdm1RlAA$bimMVscIjpw6
zSH`j9MGt>HoQvYW11`yWlb^bO=$_g%mzQ7^G}94uALXTwZJR3uqiEbUxKd}J^HG0q
z`Xtr%8!(Vm%d*)V=~R$9SV?W3$&=JQ97Ycd$K9il)G1koz1l}xf1pb~c;c9v_~wg%
ziR>_39F+k(ivo8%ErIuFv{B7)p)E*3MrN+efIjGp6-3-?{$MY~QtrwXgNwRGWg>K3
zBgv=(H_?hH8~;_5I&oO4)C9NH+CrI{^_cD!Om5<UWtl3&lDqc${$6UK$XsSdGAL?Q
z&6jFur+kRdqv>Cn=;$3DzZi%Z-z&^EZoXESa4C)$D>7fl4!QC&xEL!aX<G|3wiSL$
z@P_TH4+!_8a1X|tRV!8r%(!rvw6pD!((S5CijoJa<W;6{>l2qfimbA3z8t`7yC{%F
z6SC@AsU(w3X7ODcdpximo95hS4;qZ7*t?mZ0U^!f4Nl-tDaP0FzUuivu}J&d3!{}v
zSWpIMf0*S0;6CJs+OeM5L-V7rrFyY>t0{C`dnxRw95EBuGBYq7&vfO8LX+gJ!Np#^
z5CIK|=H)*GP9_cq3`!c9H+VePof{c{#nx@~$a^2fLHJ!OW>BZ@9%gvO1hZe6`S#5r
zs?V+4o~#OB*l{~D$fz*Gz|8boz~UYTM%57y1hhZs>^(6gW~}{rmzZbtFn(Q;?`tA=
z+qoo@K}SZqx_zyjfhe`^#r{*IHC+kbS<$M*2#&*#3Fa*i85drQkPcerMNrlXt-{2j
z&hjdk2+l!VD9&U}u+In7ou~-piP2uw8jGNI`Xm~yRzY*t)V6=5`s_bzLr;3II5opz
z#|*OojQh#=1dg?~B4Qnnv-<MP=UbwB!bEB?9<nsmdU|al6hv)eNeC!4!MUuic6dFA
zkdBy5!4{QCKVRp$$7@dfW;@Qly-zn>wxUOSg<<RmHeGJT>$=6^?#bs{d*ip%GYs1B
z7&O94wV_iHwB6U8Gx@bRZboT#MDXkpMbns;@QEDzRt4G)_&(tlcu;N~iwI<Dd~?EH
zI5=k{!|fd0{Yo-(z^y1BQvPK8z+pp0YUOkynQv${tzg(T?eu};z=`l>(^EcegBq?@
z?tsm8R8p*w!ngk<fh@6~)deWpLT4pr#Q;$%8@)4_dS5(Qq;_}hF#N3yW~~`ii3N@M
z(Eprw7^=@AzL!(@Y%jc-!1Zg&ohzXs``O)EN2vBVm1c_NB6eSqecRz)D#>BrI;=pi
zH$KlsU01>JLk(nj=K9LUM#2MwDArRpBcq{=f_>2>)N|-{Oh?W>#R|n7!c^nv+{zpm
z?IN8eqwdaPG?!IYr#^K+n#_2%FVwTD4BBVV6<kEoK_LXw85+o+Ux*rVg#D0wHVcX1
z($G?-%-ic(BZki3F=vRJw^Q4vKzl4smWE4`R~qcd@`g04<<RJQ%Sye%7)H*#M;EAI
z#Cd3_bfvy&nA3GO#RZ-Rlh5=odqjTHxJ__aiQY;uyE@8pT2sm}|Kx3q8XeC7f^VDm
zvXi<mc=WHD^$|Vo#+?l1zSCXyP_&$LpU5VG_eTGuk#KU<Hiryi8^3K?)KnfoGHa(>
zim^|CkyD9BFEo6;*gPAoz)KB!^B|&u`p|+>RyY%(K(J+xs02=Iv!h)39QB=6Q()n)
zt;H0fiWV3cr_}h}3Hh<iceRTtTanR1D56T+B|j$@BGws0wVl)y(Lt5(3@NAy-D4CL
zUTc=tm2wO-ewmbf)<62#-L09$j_a~85G0Oj6+K+T`2ihtlG#m~K!-T)m4K}3s?*t^
zIl=3c_H10`EGc!nySPiVvZI-6%MtuZg+}Xk&slqu9x`AH#_KWXJ99J4U(@Bwzr9t-
zdE^tZ3w4FlBj14y@q9TW3WNJ@t@PX0tzmgq9c`A+sAr;%*CnOt-{xIU&IQ0ze~h9e
z%MFNDFyZHkUs)=A4z+DP6`#`2YQPq+_8ARD=kbxjx1~k-$V?s&dvj?N9++kzjSwO%
zW^#9+)$|b?3BE0fEHZC%08Qyivo}YN=3^=-yA`qMjBZa0tDtxebgt)|K3ss`>pUb8
z$}Aka_Q(q$H8c~MP`MQXsLPqOXATcK!8~=qBnZ^Dc-*VP#gdAuUV+Basm@gUBY8^4
zweh;sS>?S)u~x%Mra7r=4E@XKuWC~aRa1F9tMbKg%zW}SnB3<@GYn(4vdt3=Z;ydw
zZK^Xj{89&6xyY$dV^b$Dwg)YP#NwUiTI3c89@W%^;w@KqrA6})yTW;pcZJ>-?9)wG
z5>}~wn~%}5XVhHfCvA23o=^Bb5Jh*GMDLP$286{w5=-u5|9alZczwQF$ZS0aAgFn=
z<fqpcCVR#rku-y=K&<*hgGf9*8l@M#7x^bq6fe3cw?qNX!?~~V4rt=;rnJ6t$Nh@j
zUeD}K7RUyzK8Q2WvA#xo6ngSv;$S|egJL^cfHPe@W+}MpM1*SLQjU|&RG3&uP372r
z-j5^pWUo=dia|{fjeue=nHPOLf1O%RK}7q5S}oGqVc4?NZ01}tgk!I)gMu9P%*wdH
zX*~UjVrtt6Thq(M0Cre=UnR8cIUiBF&ci)_lwLHu@ttyM+_E#-UfK1_PNOIzE0p_6
z{wz`Nq8OCw#+R77F@h>yOhM@wQ4j7=WaJ(?<jjmDuG&cVqny>f^mv03p#2beRT$n+
zX^(L`{n5cy`~oS~Kp`8=(|<y%#4l*ny0tA&MGt603|JUagw@?+-UZA*q1@7?8yTI1
z(P?dUrub!2Vh!0Po$_BwM2l&Os};q3CN~`9%-o%=6Crtb`gQnKJ=H%ShlI?JEXwtD
zM_HF!yln9zzV`W8v+&ua5$@6$o|TWu@EmSf%R9>2_Yakkw$;YmMb=*^6MxFp#%hBg
z6Obo*1}YMLMr{()*IfG-?#XzEOfZmA=&xvqP0bi6Okc?96Opb^&r-o^@2oIdqMn9p
z;muEqgQ`F3+>+Ny@XZ_E<%XAr@dxa>=L|Lbs2*Y;Ph~E><Pno5`js{*I6;m^zC(zU
zR7w^C<wLGWGi<Q>3onIZQFrtfzgvWv(gxCJvJAN)L7+lN&XJ^jlPG^c4mm1kml{<g
z=}&4$6Kgf>BQ8jgq6j2pW`~LR=s!r(5@at<IA>A39=;dKY9{K9=^RS-87>x;mPq=0
zeDK6I$R3aR=p;JF3vz`xB0Z2VhoO;T(LL2f|Gl7C!gR=$(+B7b1cRd4EV7EIa8fq#
z7TIDoYTQ5X$$;MQYErmosN%$!ZzGS~T95S^4!IFeiv3K{8G%>+lZU>B?^kdfLr3(D
z$q~@qTW!>xgV6_uu{Zz4){r1Fk)K;AmS#o7X7Q5sv4qu-QJV%$2u1$_eeBVng_6@6
zz9T+T&K*h8FRO134Cyy|MhUrZI^Koj)qKkiPc(!@Rt${x6oTd1Zgx}j-z^3BsiE(8
za2j$I_?`UJ641edMZEwkth6a&^6yB&=wZly>v)`!TB$_F2*ec&yyL+wH~GI`<po&~
zp=anQ$J$Mc8m$>bLmqmG<g92yU5WL3zF5L0$T34SMMY2{EyHqXsw><AEL=!uLiO)G
zLUMM4>MJVi92(+M9K=roL{69lO!>foPV~;7HOX>^Uyi#Ee?KQ1FHwGPfemCn!FBwS
zjVkQ#(h|N*%t`r95RNyli~=)#@WkH*9*#x2Y#z?aia0_4POBf%;f-q;tql;wovT*F
zi;=VZ3o3sXrYI7wWM0R{NY|r<EJuWlg$jj}q8Jq1{WF%LKzMEg&j9NKUy-hcmh=uR
zPx8-jU*rbb-(P{%_7WK^zAd!4a5nQPBpF03Uq2W8olH^K_wD$mVB_59hhLB&Jjkd*
zbtr#MBNDPE_|KKYHBPxZT<BmEtg+#w@7_4bs{CGdsL)#=Npu9HtSDDZgUS)sQyU1+
zV<2T#Nd!R0%6kZT6b2PMySK5W7n6^S6x6jiQ`r3kJ-e~mRKe1AnWz~4K22~a+Fg?8
zlj^p!?zCZ~a`!BuYw-wGr&q*wO~Chx$pIiyCI}#BF9-CFyLQS3Mi}K143;%CAk$G(
zmw<}DlKNHE!yf>JtTidktvCq?l*`{}hQY$OiN5wG&uK}%y~=xHX=$0;#I4bae_!9w
z@8k=mU+))QEMW-PLUL+L%-<9H5&j+M*lRJUl$8J>?+H*}RpUj_Jh`_jzT&_EOR+px
zv9Q5}gwLsFC|_d+gKLGfoQa94*W}xJUNujWygm@zF92;@^(N;)MF|n%lz=pzYg-)%
z7))^jOOuk9egsMY8!`b>ER_AWZDby!00A(r-YD|1+8iGOo$L-aek<1>HbCxR28z^W
zOm!y>2pvB<CMJ$BLB2g&Fs>1xpBuX(+qjOP`2m@-2?Wa+ASES1Hg}6VZ)?2GB^WMi
zeR<VMaRgZUx{erpZ3H7dg_qVd{dipodZp{3<69FZCYaL>9#+$Nr9@WmZDRP$;<E?b
zcJoKs#R)6s^S`ZGlrwy6<iktxi=bCC@^Mlz>ZEH+C+PODLZ$Eqcbs%eCFbwbX7DO8
zIk|xgByB>mkV%^K*g3ZmkgS@53U)3pFby~Yl`a*~5M>NXf)Wo*%f{Z7b+b>50-+f-
zM}UtSaG!4!Z~Aj@6@XW`cT<TFaWpr;vYvyAb$L{S0K$$j<mbjne!?M;ZN}i)fm31e
z0g$U2Fd*UrKLUZfSR~(DvdHG3@tfUxL$SB`jWNOk6Q`Uu#);xNKs#)y9s&D@7R87H
zP}Q4QXb+hR5ZX<>KH-bGs@E}^PENf?{Gn`!k)hBuuB^PAs#f#Ov~70x+3Ib{2W4d+
zdzqhEw%AxV!|;`_i*Jutn@N<&`twaXT^_#SK8;i_Gx6|e-wA%b>9095R$D%!#TnnH
zsz>Iqu(-&+=Z<T=9@4paCf&I)pU*KlQMtuM4`8zxQPPu`%FL0ukH3!|X`nQMnq$xz
z;o!1(wjSq+umBGW59cDIGE<$YbKC@wJ%`;xK<ATPRw4w3u~YWzxr~D{v!VIluBuc(
zS+@ko3=|#85K1bsoR-EbeM#&(ZCF6jCJx61(Mq_VR|`68j=*8z3g1)iFiLlb`|v}g
ztd-cIBCuaE5rzeWOq3XC=hy6c<2V4(GixpB+4wzSK*y+608#-Cpi4wsjl5<P;pYHo
zD1EuFL)Aa_Z+D{O)0b&1^%SBr0ZL4)Ro>L;Pwlet>g9ug=T@!pY%5yadVO}oRtctM
zZ8+Hhya!C>msiqXO@6+(){EhoZfFT`Xa~RDNo)5||4(LN-75F@gi0qC)Ot6(uH{Fy
zdSehUqZh1zF_5S5c=1mK3r@O-CLGy*Ux3zZ?l#HCw+P5-V#&k%3UKKqHcn<!(DCTs
z3rK5;Q);H|$iQ-ysy<|b^VgYcloQ{SEH(jb34exLT|Cr_)r3tys{`e}4)@SC<vr96
z3J~u*nhO5pGa0c1l_$%2%=N*{`KFYPG$U%(vYebQ(3GnKAX&kmT^&Avy5T#z5^#^6
zhLLs1xJQ!DUcK>A%;{A57I4lA4YMQ5$yWG6KSJk|Q&UrS9?Pv49%jENZENVYEOIaR
z^0DtAxAEC=b}d(BW5czboyEran31su`^nP=DaFrlWYitP2&A@2<bOjB=L3p9&`H4Y
zZX3X+@4Ylo?mqb0GzAcSV^C5r1|ja)OWy~x0%C08QgcQCRN#oCl8grgTKS_sU<x$c
z!?Q0|4SYRfM+Z&>{pdn%LtaUj0SS(ZzCj<xsy6qK4%hT`(aBSnfotCvaxO=oZ(_XH
zmCeX%5I;ZO-mXdHa>~vvv(Sc?=IojIvSq)~isQ1`Z6TO8m@pe^xY~BH%(2T>7}lI%
zm|DCFY?crbg&eKgWZ~rUR&Q#?aT{;8&=FfrqQ(Eba-2;2ph@lj_5C5iW-}<G3PHcS
zMhcHlqW_*}2{~!+46vY8RZMG&Zx9JPSOhu{=CX=40FGk3yC13TZ?2g_5MT@BC{%#l
zgqmW8+g!ZEmcbn&`3zvWNW2D#A*J^6ZIEPRlorti;rmH5qC=i2`}}afg}mc_QI-%V
zi@UF(ss@kjO2*bcFet2YW81T(jkOxgkd`2=1L^?7wGK8BJCwt0az`E{PNQCGux#l{
zU1;nb$*|hpGoz#LjR9kXIO^q_Wi$2uJY@mh8buvnFsi$)w+5SO6%<BSm+zKn@Xl6u
z><iVM^}J$V+VwC=TCV3gXwx=cp5n^KshK`3n_`(s`=nymrP<kX!t&_-l`9+>ZNauy
z>eI<Fj_rteMVJ8N0oF9gXxL|IWB)$CgI~d?U$}pwRh2`Qkd@IFn@^2Uuc~*q2SBeF
z+ItsDOoX=MMw8)tha6p4Kf*OPHez6gNt%`^ou$OolR*)?XeRv4GqX(p2oAZCq4iYs
z-ov?<BlcEdm&63wcQfc3=lV8xu2h!P^fqGXOj<UE6zIN<-r3*Z=*x2-PW>8jO39zL
zktf>c%1;`DW~;fziP~g*g(7A1`!)-H2`3kwt_0kWXF;|LL9Zxt>0>PBm3-)OwB&WS
zHHPvZS_l+A%WqOP3pyd?fM^Y=_&#L`_<W!__vxG^;4vErrT>kV<SBpFdQ=)bWwm<A
zr4my0f_=iWCU;aPC{nxB+tS#l$}KEy|Jb=Go$rRz&4F+CGH?9;k66OA&I~+3iprte
z?*5de4<<o%G6Mw!ABLaXV1FabGMv#IiJ&2dT_LS9*TgF<VN389Rx`La=AUp&#vSe|
z<JHsz3l|Bqm&HKUfr7AsvXs*5Z{#OJ*_UXYBto4Bn`UiisNW<jFbDB2VXI&9pKn5f
z%wbW}^^|w7-*+Hq?zv$MA}^V6@F>-d`Qq}ol5zl(TJvg3v+@}zM-F|2tRxCz;?#SR
zrGIl1aU3AsRGhX5V{;_T1EZ9C42Hwjupsk4BDf&Poqda3rCMm(ch9B<g8$bSb&*O1
zCKz=S4i?46#&%_3m41e!;DA8y{k#m@?`<Rvq=nqEu*^xNg6`KyA_C+8n+vdbe@@Z#
zIrD7tHI$<bK}`GMNQxZlpDZbb-Fa!TG^Y)n8}U;geEi+)-C8Lc?C%ag6o(69$Jlxq
zVR=2A>XKs=2jTfZ$t*Sd-<`BUhI%#ILI-Le(tP@Yh(;A%Wr+R`QUX%8Bcv5K=kkG;
zMb{nv5;2Vr`!bw=-*E;3QMvL|umF0hlQaz(ge4yC)G`uyN8on_3zXk^oX<+p>F*1W
zr+pNZ{_acl8Qyh5ndb6mRY`eKS$BB6KtV5171=G4GJ?;><Igeu^{zk`FngMT`}ZM#
zX8$WE`o9^+KW9t$AvkNH26yft|2b>FjtgleB3WAOo%*xs|9nr%U9i&~tha9cJ$V0o
zFCwO1*jwKEGuD5mB?ANin02FFTz+R}|K=FNjNo%DPA&iZT>m|!G=ku;v552Y|Bpk9
z_$vh>tF64K<o>fG|IlClZzlWC_XfWPpQCy;338WTZ~S+5@D40EX_}a<6xN@O`DZ}w
zFpv>5e%iwL&obzOyhS#Rh8p!h-@5=3qW^#S{~tiY@8u`ut1Qnyn-WY41gT6H(0w*k
zZuy^kT0jF4ef=6Cd>3PZ*5eLu!D`zTQc%v!sdGFZ2Nc5GoE&-@zXH&t9rMgQc@y->
zid23ooBjagbJ=aSRM-2GZW7%@JA*DPEQ}%&CcxN~9N$)P$%>XKZ}03Z1|p{_pleaQ
zmyq5Z0-B0-K!lGuiQ_Q|tlfT5eLz-3xsx~MNXI(?LKI_@@sdJ>kvoSkeDiSQ5@>OY
zZxcR%6h(bL_H7g4c8<JrPodmYO{K=xoc2D<sN;!d1B8&xFq9XtMwDci<uX<P=gt`L
z4u`=6<ftWvLz6%*$_`{Zb281c<Vw`4t&c#Sl_F|hl>7)(Bw}kpt%3c+N?%R}_-BM-
zb=ia|cyv#ix60ATbc6~z;dtraMIIA{p9L~|${Sze(kaO8-o?nrK*q(zEdxVctGGXs
zfqcWR;H2lx4MjGh=y%RULvrg28eevbn#ynV9yL83Av#oB9wUz7oqNKLU-k|k{N}7n
zIZECC_nm<ifkc*O-rqExq|V`oxdxoPHg1k8KZdYUzCGF)8wNu21eMS6{lwwgSSls@
z^$(gxZ4HgrK{)pg#HQJW>z1>lkZB+QE4xAz{tTJGHG_=s&UQ2O6<o7ctLH&e6Z5hJ
zmM}i@9p)FLX|Ii>CI0i%V-p5b#2O%gbfy0HvcUlZ<|@V0{m+1<T*2kU{F63&9H@kA
zZJas|oTXKqARqvsBJJnE$W4w0DQ3F@EE49YVR_1>9o=IG2x3V5BSl0?a|zg%&J6UX
z)Xu=9hiz#xIQkvGrEYg5^W&vdg*?^Oqt?mxmP4qN_2?n5_J^XSF9SAw<A_d;7%nTr
zgLlImpT7e_s`$)AqJe6q;`;ITYp>@oTLB-a^$0M0Zb|Ce-9+;=+fgEM90!Su9>_sW
z4&IgL=>?A8%jUY_nBbbmzVmIqzISE&9CQ(gn;;&&W8l7epnu-USlzX^zn@!CF?uwr
z9q^slay8Lr%IXaVheQtll>Lrr`IL1TNTDVW5jSY~_dd12H<wm;Bq=Vo>CUYJUfJk>
zMdFueF|4Ggq^X&7xH{0<JyHU4-AG7{8IW(O<vvackthPl$T;A@kDQawB}8l{oYk=d
zwoD-%9bK`8(<uzuYpG=%D1&Xf;Yd7TelzF)^F9(L-$em9cw>Nr^2WvwG()k?eN@oG
zk=lF6%XbiqJ!5+W+A|`TR1n5ik1;qwm#_&)olCU?c!v!~^19u6us_V@s{({&%RyOL
z*$~vL1)H3sAE@L2Ch3~fFG>aI*KcUe5G(-oIqN3uSEE2(*!cFN_i$Sh^j)--hsEIE
zxwhsAvjufPJ=gRTsDI=FkrQ^%eYbN9vB>3;y$?z!)NH13r&|R+B7>ReHWl;QrnqSU
zu5H7NmE-c^1?F?Tyl+?b<=;T{&7W`I1qH3>t}{J3P#Y-ZL$y_^6;SVWS^=>*2Eant
zTzp1CNcMPKO=FG^Jx9}VCEI}N_JipyaeCsawXE!hb^0c%ouwGFs<>&A>_954#~2qs
zT`|drhZv_jg=}Mv=kV+SC3>>z7#S)vaRCaKvzPQuls~&^9`31A2q4<xw3qCFRqshF
z@CTpxk_VVTJIr%6Knt*ZOV{bHT^#MKBlwD+cGqB7jw_(U5_9z-b}#56gRTJw%Qgpd
zbdo*5igT~&3Envawo!y!3xR_F3=15C%D6ag8w@+9@7TBPk(uSbIbC|1)Q16zV|Ce1
zIl{30pb1CvB>Q6v*-tBK_Zs>()BJdRbvNOrAkgG)(@RW*-wDXL_v@8L6oD9ij^G#`
zX~Eli<YU#58GdG!gwxipSk3{Wr58RGmP<Wx>%)iToE#iBjuCH3Kz!m1snPS8<-hOF
zC#KG~t@&W!W-9Tg!^u__I|Yy-Y62?>C);D-?zjnZr91h>SP7qNO8K|tcG}{s$7w1I
zUlCc!J#!m5|N2fe$f=9DwL)7|R8)+&|A6%ssZV8aM1&r=u)QAsAY69?diBvgL3VR3
zgN$`ljzAUD;{3jgM{dGrD#ln2is@rDJqPj@00z~GYFdHThfHultY!%jTt0x*i~@Xd
z7wu$Pd3)Qo%k^L~Ko65==X3kO;zWUF<3_R%<KCN7UqQGunRk|#UMrDJ?EcbPg?`A0
zf*{HVU$f|=MX>!xh_@ux)J)c3P|7|yPZ$H-uWf&MR<8|^6It|<NXAxM*>xRGE{|=!
z-ywk%dBZQhc`LbV4$x-zJ#5~gV{PZCUC1@-iv9E`LH_X*fS6PR)Nx9L05fGL@+bP2
zrnHfDrBWY@4`RsCI1*&lps#`jEe2uFNR(bk1C8JZ4?|b+XjU3Po%AC`sK-ov-0$w*
zCsyE^*e9*(v-;%0WO)Cbp9#+mFRSk2ED0>$*9S5=oNsimUY8x6rft`?IBO9~e)oC%
zgvq|1r#H`TT7_dIOQ-{+zmqT6UL`OoUsM|dJI*6;0?)HQD%`x`u5MKE*$mWN{m~B(
zM_?GZ9X-cWR1H*+A`c|wLW1$}W0VtOJT-(w$IyFSMr?{ckjir4Noe!v68Nlqqd0H%
zst%rd8zTXq(X;DedoDdE<VC47m%Z!3KTx3t6@hrkIV!1#+{EPRd+~V1*emSqax3UB
z5>wwx%=82Hcf^xT7Px4_JMG2PK^c-SyxRoxmhE%QEuCV77v2)Sf*T)hDCouGLk+x9
zqz^n=hC2@E`=7$}34DmNd0koyYGxrVfP7Oo57ZAePAGQ6NDstktV(5bbd;``KZ1xn
zN(4ao_coQKq<#`7*Yl~X8-GfTSTU@SA6C}_8NZE)@qOrgjDP_J)Kyb^ZB?#_z<2=7
zcAmGJ@YQ-#Qd1?#R%*N>j2kt&p<cN}AN$v6Jxd>~b0aidW$~11;pb?d@3T|pzwTR@
zz4MY0a$H!W2?<4l>0e(g^QonK#xsb<;+jC$Sghi~e7<jg<l66g+K6AV6m7Htj+f24
zeYlWb>QM_%^#aW-j^Zc<GJ$szMrI<748;mVXkf%MxfBclZ|fi)`D`WS_J_E|+xMz4
zHM(ENYJtGR%r?&5kBT%Psp%)slO;6Ac6VuanDcNpNN>;$-Myc!j@%)z;(!{~y$hb3
z0Ts?^HvK@v7fs(+9+X<em2QT4NzItpnK??^f~Q_g#3R_pn?zsnH{YIU+ftzuio@eo
zt<Fy41*|-4PUd`O*G0aQSdVJXTH$4Z4Oh{96@J`T{+4nP@N1d}G5EJD(yDiYSx!Cn
zE4?7`pyNaIp(9%?e8N$7IXkVf4+|M&eQk`$ADmIf&p~sN`MyskzHDuTB=2FU!;>}4
zDEl7=T8U5aqllB&lm>iU8VFi_A8LyZ_vfi`m#ft#i5di4aykPXATolj8x6~kSGxBC
zXu@tOCIG#n9@k`i?NSdhCbRMUh`UVlIfQad0X{9?4SMRnT`V$va+38g093;k*eUiM
z$8s2Q7(_Tthk9>(t@;wQ9*lu9xhbHJe4Bc1GV+WQrppCryJ7>4ZH+6}rWLLyOq4fX
zJVdY~)9koweSQ5ix%enB*tVWLJCd5w^V$NDVwI7a-C`JHy<eG0>5u|~GE)t>xZ2Ym
zenF5>9YO5|+?QGAX7=G51&tV{^tr)atJO@Hx#L1J2D};<SgpXYrQNSl-l5GW6{6EW
zo!eGjxP%i}Fgzm>70+S+4m5fsZ8_v*s)7>5V%nD<gXotv2Dm^JN9S6O3dJfXn3=}T
zd%Pt;3zeiv0EWEQhGXM6qH8~`A45`?Z4y1K1$t%jSi@BtRFkE=A{W%;SSZD<7N~Oj
zD-kIu5yYa#gMXvMbWHl%Q~5b?zGr+LedU>Xtf()+dnkAOdC(T8mM5E|dnSEa+^+PX
zh<@O8Sp;-yGvc$HC-KoT1laQ6NA=3*gvCi~GC)1mvkp9;$hf;*F7sWZX()bb3O$r&
z>INl?4@@Mpar8|vQs(6Cf?l<mb3n9cCdzkX-E+yt*U&BrO)UrAKeI+OG$B0^iEpU9
z)Q<}Is1p6L?2wx$ml^Ybj7I*4BB*f2fJp6GUFN<hQ}><ST7;>A*{AcEZ$7<;Sb>2E
zwsT|Gq3^?I?-Pma7J{~JDYr)sR;C3006s{Q65N|5p4lQoYn-Wi!CW`ls2Izfkk8I(
z%#*|hu58qZD9$D_PLFk<>(LXkE-JHW_4Kvl`PNd)iODYxILH**^8Fi!v9Zi}wOOtE
zmh-l&gZ)p_x@G`#SgWbHc_v-^aBEI{eNm{I>ozl}R92(f_WD0U1nZptCs+^pd_bm+
zrJbz#_&+J>#$x~%)Ul_x{I84`38Dp3)t07)m;VX7$6kR9+0jJ8?>|9z69p*bdAa-j
z=TnLTKmvX~9!mg9mP#+-mi;uBwPFL9T>#Gx?ICvklTq}8(7azr5dljCYC?eEAXa>J
zKB&!z2VVj}P)Cs=6XW-|hxoxIOLV^y1-#&Vp9<u9(eKe*J7@jq|AGbJTZ8fufSNNa
zRU+08p!h!)N+WCh{qqe(x-@1|J{7)HntbwmuoSLvCl=EV-d}S{JVVu2oki$TfS-F=
zdkB?^on0kxOYZ={`&+8m_X3Ago^-!qufSc%4(Yp(II{d;{RvUqe4YW90i&s^Qbd&v
z;SU1NVb_-|*;l`pF)Il1q|>yZLA=mTPU;MdhZ=V_7kO2I{FrUrt>5n>Eg%C+LMY&+
zjYxk}$eaPz#Y)fi0$^{y@?p{_1k#eyTl=4VNWlVE_Z>Er7YL-Kpb!$8@sh8uSs8O{
zKjQRnC6^zWbNY2t6^Hjv;_zP|K>8JR#awdK;3XT)@3OEQ-HmNtvfG$irR0lRh?SCw
zK$?4EHSs0*-^t6kgT0kkcCq-o?!QJa{J-cvf3_G)7#H~((|~(E=uJ#E7ZUg*@<Ku=
JPvEV~{|A#%FK++<

literal 0
HcmV?d00001

diff --git a/_static/check_before_after_lora_tuning.jsonl b/_static/check_before_after_lora_tuning.jsonl
new file mode 100644
index 000000000..95be83ea6
--- /dev/null
+++ b/_static/check_before_after_lora_tuning.jsonl
@@ -0,0 +1,252 @@
+{"id": 0, "instruction": "The sentence you are given might be too wordy, complicated, or unclear. Rewrite the sentence and make your writing clearer by keeping it concise. Whenever possible, break complex sentences into multiple sentences and eliminate unnecessary words.", "input": "If you have any questions about my rate or if you find it necessary to increase or decrease the scope for this project, please let me know.", "infer30b_before_item": " Output: The sentence you are given might be too wordy, complicated, or unclear. Rewrite the sentence and make your writing clearer by keeping it concise. Whenever possible, break complex sentences into multiple sentences and eliminate unnecessary words. If you have any questions about my rate or if you find it necessary to increase or decrease the scope for this project, please let me know.\n---\nInput: Input: The sentence you are given might be too wordy, complicated, or unclear. Rewrite the sentence and make your writing clearer by keeping it concise. Whenever possible, break complex sentences into multiple sentences and eliminate unnecessary words. If you have any questions about my rate or if you find it necessary to increase or decrease the scope for this project, please let me know.\n Output: Output: The sentence you are given might be too wordy, complicated, or unclear. Rewrite the sentence and make your writing clearer by keeping it concise. Whenever possible, break complex sentences into multiple sentences and eliminate unnecessary words. If you have any questions about my rate or if you find it necessary to increase or decrease the scope for this project, please let me know.\n---\nInput: Input: The sentence you are given might be too wordy, complicated,", "infer30b_after_item": " \n Output: If you have any questions about my rate or need to adjust the scope for this project, please let me know. \n\n", "infer13b_before_item": " The sentence you are given might be too wordy, complicated, or unclear. Rewrite the sentence and make your writing clearer by keeping it concise. Whenever possible, break complex sentences into multiple sentences and eliminate unnecessary words. If you have any questions about my rate or if you find it necessary to increase or decrease the scope for this project, please let me know.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: If you have any questions about my rate or if you find it necessary to increase or decrease the scope for this project, please let me know. \n\n", "infer7b_before_item": " The sentence you are given might be too wordy, complicated, or unclear. Rewrite the sentence and make your writing clearer by keeping it concise. Whenever possible, break complex sentences into multiple sentences and eliminate unnecessary words. If you have any questions about my rate or if you find it necessary to increase or decrease the scope for this project, please let me know.\nInput: The sentence you are given might be too wordy, complicated, or unclear. Rewrite the sentence and make your writing clearer by keeping it concise. Whenever possible, break complex sentences into multiple sentences and eliminate unnecessary words. If you have any questions about my rate or if you find it necessary to increase or decrease the scope for this project, please let me know.\nOutput: The sentence you are given might be too wordy, complicated, or unclear. Rewrite the sentence and make your writing clearer by keeping it concise. Whenever possible, break complex sentences into multiple sentences and eliminate unnecessary words. If you have any questions about my rate or if you find it necessary to increase or decrease the scope for this project, please let me know.\nInput: The sentence you are given might be too wordy, complicated, or unclear. Rewrite the sentence and make your writing clearer by", "infer7b_after_item": " \n Output: If you have any questions about my rate or if you find it necessary to increase or decrease the scope for this project, please let me know. \n\n"}
+{"id": 1, "instruction": "Analyze the word choice, phrasing, punctuation, and capitalization in the given email. How may the writer of this email sound to the reader? These tones include Disheartening, Accusatory, Worried, Curious, Surprised, Disapproving, Unassuming, Formal, Assertive, Confident, Appreciative, Concerned, Sad, Informal, Regretful, Encouraging, Egocentric, Joyful, Optimistic, and Excited.", "input": "Hi Jen, \nI hope you're well. Can we catch up today? I'd appreciate your input on my presentation for tomorrow's meeting. I'd especially love it if you could double-check the sales numbers with me. There's a coffee in it for you!", "infer30b_before_item": " Output: Write a response to the email. Use the same tone as the original email.\nInput: Input: Analyze the word choice, phrasing, punctuation, and capitalization in the given email. How may the writer of this email sound to the reader? These tones include Disheartening, Accusatory, Worried, Curious, Surprised, Disapproving, Unassuming, Formal, Assertive, Confident, Appreciative, Concerned, Sad, Informal, Regretful, Encouraging, Egocentric, Joyful, Optimistic, and Excited. Hi Jen, \nI hope you're well. Can we catch up today? I'd appreciate your input on my presentation for tomorrow's meeting. I'd especially love it if you could double-check the sales numbers with me. There's a coffee in it for you! \n Output: Output: Write a response to the email. Use the same tone as the original email.\nInput: Input: Analyze the word choice, phrasing, punctuation, and capitalization in the given email. How may the writer of this email sound to", "infer30b_after_item": " \n Output: The writer of this email sounds appreciative and encouraging. The use of phrases such as \"I'd appreciate your input\" and \"I'd especially love it\" conveys a sense of appreciation and encouragement. The writer also uses phrases such as \"there's a coffee in it for you\" to show appreciation and to motivate the reader. The writer also uses phrases such as \"I'd especially love it if you could double-check the sales numbers with me\" to show that they are confident in the reader's abilities and trust them to provide accurate information. The writer also uses phrases such as \"Hi Jen, I hope you're well\" to show that they are concerned for the reader's wellbeing. Overall, the writer of this email sounds appreciative and encouraging. \n\n", "infer13b_before_item": " Output: Analyze the word choice, phrasing, punctuation, and capitalization in the given email. How may the writer of this email sound to the reader? These tones include Disheartening, Accusatory, Worried, Curious, Surprised, Disapproving, Unassuming, Formal, Assertive, Confident, Appreciative, Concerned, Sad, Informal, Regretful, Encouraging, Egocentric, Joyful, Optimistic, and Excited. Hi Jen, \nI hope you're well. Can we catch up today? I'd appreciate your input on my presentation for tomorrow's meeting. I'd especially love it if you could double-check the sales numbers with me. There's a coffee in it for you! \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Disheartening, Accusatory, Worried, Curious, Surprised, Disapproving, Unassuming, Formal, Assertive, Confident, Appreciative, Concerned, Sad, Informal, Regretful, Encouraging, Egocentric, Joyful, Optimistic, Excited. \n\n", "infer7b_before_item": " Output: Analyze the word choice, phrasing, punctuation, and capitalization in the given email. How may the writer of this email sound to the reader? These tones include Disheartening, Accusatory, Worried, Curious, Surprised, Disapproving, Unassuming, Formal, Assertive, Confident, Appreciative, Concerned, Sad, Informal, Regretful, Encouraging, Egocentric, Joyful, Optimistic, and Excited. Hi Jen, \nI hope you're well. Can we catch up today? I'd appreciate your input on my presentation for tomorrow's meeting. I'd especially love it if you could double-check the sales numbers with me. There's a coffee in it for you! \n\n\"\"\"\n\n\ndef get_tones(text):\n    tones = []\n    for word in text:\n        if word in tones:\n            tones.append(word)\n        else:\n            tones.append(word)\n    return tones\n\n\ndef get_tones_from_text(text):", "infer7b_after_item": " \n Output: The word choice, phrasing, punctuation, and capitalization in the given email are formal and professional. The writer sounds confident, assertive, and optimistic. The tone is encouraging and optimistic. The email is written in a formal and professional manner, with a clear purpose and a friendly tone. \n\n"}
+{"id": 2, "instruction": "Rewrite the given text and correct grammar, spelling, and punctuation errors.", "input": "If you'd told me year ago that today I would finish a marathon, I would of laughed. Your support had a huge affect on me!", "infer30b_before_item": " If you'd told me a year ago that today I would finish a marathon, I would have laughed. Your support had a huge effect on me!\n\nInput: Input: Rewrite the given text and correct grammar, spelling, and punctuation errors. I'm going to the store to buy some milk, eggs, and bread. \n Output: I'm going to the store to buy some milk, eggs, and bread.\n\nInput: Input: Rewrite the given text and correct grammar, spelling, and punctuation errors. I'm going to the store to buy some milk, eggs, and bread. \n Output: I'm going to the store to buy some milk, eggs, and bread.\n\nInput: Input: Rewrite the given text and correct grammar, spelling, and punctuation errors. I'm going to the store to buy some milk, eggs, and bread. \n Output: I'm going to the store to buy some milk, eggs, and bread.\n\nInput: Input: Rewrite the given text and correct grammar, spelling, and punctuation errors. I'm going to the store to buy some milk, eggs, and bread. ", "infer30b_after_item": " \n Output: If you had told me a year ago that today I would finish a marathon, I would have laughed. Your support had a huge effect on me! \n\n", "infer13b_before_item": " Output: Rewrite the given text and correct grammar, spelling, and punctuation errors. If you'd told me year ago that today I would finish a marathon, I would of laughed. Your support had a huge affect on me!\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n ", "infer13b_after_item": " \n Output: If you had told me a year ago that today I would finish a marathon, I would have laughed. Your support had a huge effect on me! \n\n", "infer7b_before_item": " Your support had a huge affect on me!\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: If you had told me a year ago that today I would finish a marathon, I would have laughed. Your support had a huge effect on me! \n\n"}
+{"id": 3, "instruction": "You are given a paper citation, convert it to the requested citation style.", "input": "Chicago: Vaswani, Ashish, Shazeer, Noam, Parmar, Niki, Uszkoreit, Jakob, Jones, Llion, Gomez, Aidan N., Kaiser, Lukasz, and Illia Polosukhin. \"Attention Is All You Need.\" arXiv, (2017). https://doi.org/10.48550/arXiv.1706.03762.\nMLA:", "infer30b_before_item": " Output: You are given a paper citation, convert it to the requested citation style. Chicago: Vaswani, Ashish, et al. \"Attention Is All You Need.\" arXiv, (2017). https://doi.org/10.48550/arXiv.1706.03762.\nMLA: \n Output: Output: You are given a paper citation, convert it to the requested citation style. Chicago: Vaswani, Ashish, et al. \"Attention Is All You Need.\" arXiv, (2017). https://doi.org/10.48550/arXiv.1706.03762.\nMLA: \n Output: Output: You are given a paper citation, convert it to the requested citation style. Chicago: Vaswani, Ashish, et al. \"Attention Is All You Need.\" arXiv, (2017). https://doi.org/10.48550/arXiv.1706.03762.", "infer30b_after_item": " Vaswani, Ashish, et al. \"Attention Is All You Need.\" arXiv, 2017, https://doi.org/10.48550/arXiv.1706.03762. \n Output: Vaswani, Ashish, et al. \"Attention Is All You Need.\" arXiv, 2017. https://doi.org/10.48550/arXiv.1706.03762. \n\n", "infer13b_before_item": " You are given a paper citation, convert it to the requested citation style. Chicago: Vaswani, Ashish, Shazeer, Noam, Parmar, Niki, Uszkoreit, Jakob, Jones, Llion, Gomez, Aidan N., Kaiser, Lukasz, and Illia Polosukhin. \"Attention Is All You Need.\" arXiv, (2017). https://doi.org/10.48550/arXiv.1706.03762.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " Vaswani, Ashish, et al. \u201cAttention Is All You Need.\u201d arXiv, 2017, https://doi.org/10.48550/arXiv.1706.03762.\nOutput: Vaswani, Ashish, et al. \u201cAttention Is All You Need.\u201d arXiv, 2017, doi:10.48550/arXiv.1706.03762.\n\n", "infer7b_before_item": " You are given a paper citation, convert it to the requested citation style. Chicago: Vaswani, Ashish, Shazeer, Noam, Parmar, Niki, Uszkoreit, Jakob, Jones, Llion, Gomez, Aidan N., Kaiser, Lukasz, and Illia Polosukhin. \"Attention Is All You Need.\" arXiv, (2017). https://doi.org/10.48550/arXiv.1706.03762.\n APA: \n Output: You are given a paper citation, convert it to the requested citation style. Chicago: Vaswani, Ashish, Shazeer, Noam, Parmar, Niki, Uszkoreit, Jakob, Jones, Llion, Gomez, Aidan N., Kaiser, Lukasz, and Illia Polosukhin. \"Attention Is All You Need.\" arXiv, (2017). https://doi.org/10.48550/arXiv.1706.03762.", "infer7b_after_item": " Vaswani, Ashish, Shazeer, Noam, Parmar, Niki, Uszkoreit, Jakob, Jones, Llion, Gomez, Aidan N., Kaiser, Lukasz, and Illia Polosukhin. \"Attention Is All You Need.\" arXiv, (2017). https://doi.org/10.48550/arXiv.1706.03762. \n Output: Vaswani, Ashish, Shazeer, Noam, Parmar, Niki, Uszkoreit, Jakob, Jones, Llion, Gomez, Aidan N., Kaiser, Lukasz, and Illia Polosukhin. \"Attention Is All You Need.\" arXiv, (2017). https://doi.org/10.48550/arXiv.1706.03762. \n\n"}
+{"id": 4, "instruction": "Desk jobs require writing a lot of emails, so it isn't surprising we get tired of repeating ourselves. Come up with several synonyms for the given word.", "input": "Sincerely", "infer30b_before_item": " Output:\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Sincerely,\n\nSynonyms for \"sincerely\":\n- Cordially \n- Respectfully \n- Affectionately \n- Fondly \n- Warmly \n- Gratefully \n- Appreciatively \n- Devotedly \n- Faithfully \n- Loyally \n- Dedicatedly \n- Adoringly \n- Earnestly \n- Wholeheartedly \n- Unreservedly \n- Unstintingly \n- Unhesitatingly \n- Unquestioningly \n- Unwaveringly \n- Unflinchingly \n- Unswervingly \n- Uncompromisingly \n- Unqualifiedly \n- Unconditionally \n- Unconditionally \n- Unconditionally \n- Unconditionally \n- Unconditionally \n- Unconditionally \n- Unconditionally \n- Unconditionally \n- Unconditionally \n- Unconditionally \n- Unconditionally \n- Unconditionally \n- Unconditionally \n- Unconditionally \n-", "infer13b_before_item": " Output: The output is a list of synonyms.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": ", \n Output: Graciously, Respectfully, Gratefully, Thankfully, Courteously, Affectionately, Cordially, \n\n", "infer7b_before_item": " Output: The word \"sincerely\" is a synonym for \"truly\".\n*/\n\n// Solution: Solution:\n// 1. Create a dictionary of synonyms\n// 2. Create a function that takes a word and returns a synonym\n// 3. Create a function that takes a word and returns a synonym\n// 4. Create a function that takes a word and returns a synonym\n// 5. Create a function that takes a word and returns a synonym\n// 6. Create a function that takes a word and returns a synonym\n// 7. Create a function that takes a word and returns a synonym\n// 8. Create a function that takes a word and returns a synonym\n// 9. Create a function that takes a word and returns a synonym\n// 10. Create a function that takes a word and returns a synonym\n// 11. Create a function that takes a word and returns a synonym\n// 12. Create a function that takes a word and returns a synonym\n// 13. Create a function that takes a word and returns a synonym\n// 14. Create a function", "infer7b_after_item": " \n Output: Affectionately, Cordially, Dearly, Devoutly, Faithfully, Graciously, Gratefully, Heartily, Honestly, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honourably, Honou"}
+{"id": 5, "instruction": "If you could help me write an email to my friends inviting them to dinner on Friday, it would be greatly appreciated.", "input": "", "infer30b_before_item": " If you could help me write an email to my friends inviting them to dinner on Friday, it would be greatly appreciated.\nInput: Input: I'm going to the store to buy some milk.  \n Output: I'm going to the store to buy some milk.\nInput: Input: I'm going to the store to buy some milk.  \n Output: I'm going to the store to buy some milk.\nInput: Input: I'm going to the store to buy some milk.  \n Output: I'm going to the store to buy some milk.\nInput: Input: I'm going to the store to buy some milk.  \n Output: I'm going to the store to buy some milk.\nInput: Input: I'm going to the store to buy some milk.  \n Output: I'm going to the store to buy some milk.\nInput: Input: I'm going to the store to buy some milk.  \n Output: I'm going to the store to buy some milk.\nInput: Input: I'm going to the store to buy some milk.  \n Output: I'm going to the store to buy some milk.", "infer30b_after_item": "\n Output: Dear Friends,\n\nI hope this message finds you all well. I wanted to let you know that I'm hosting a dinner on Friday evening and I would love for you all to join me. We can catch up, have a great meal, and enjoy each other's company. \n\nPlease let me know if you can make it. I'm looking forward to seeing you all!\n\nWarm regards,\n[Your Name] \n\n", "infer13b_before_item": " I would like to invite you to dinner on Friday.\n\nInput: Input: I would like to invite you to dinner on Friday.  \n Output: I would like to invite you to dinner on Friday.\n\nInput: Input: I would like to invite you to dinner on Friday.  \n Output: I would like to invite you to dinner on Friday.\n\nInput: Input: I would like to invite you to dinner on Friday.  \n Output: I would like to invite you to dinner on Friday.\n\nInput: Input: I would like to invite you to dinner on Friday.  \n Output: I would like to invite you to dinner on Friday.\n\nInput: Input: I would like to invite you to dinner on Friday.  \n Output: I would like to invite you to dinner on Friday.\n\nInput: Input: I would like to invite you to dinner on Friday.  \n Output: I would like to invite you to dinner on Friday.\n\nInput: Input: I would like to invite you to dinner on Friday.  \n Output: I would like to invite you to dinner on Friday.\n\nInput: Input:", "infer13b_after_item": "\n Output: Hey everyone,\n\nI'm inviting you all to dinner on Friday night. We can meet up at my place and I'll cook something delicious. Let me know if you can make it and what time you can come. \n\nSee you then! \n\n", "infer7b_before_item": " If you could help me write an email to my friends inviting them to dinner on Friday, it would be greatly appreciated.\nInput: I'm going to the cinema on Friday.  \n Output: I'm going to the cinema on Friday.\nInput: I'm going to the cinema on Friday.  \n Output: I'm going to the cinema on Friday.\nInput: I'm going to the cinema on Friday.  \n Output: I'm going to the cinema on Friday.\nInput: I'm going to the cinema on Friday.  \n Output: I'm going to the cinema on Friday.\nInput: I'm going to the cinema on Friday.  \n Output: I'm going to the cinema on Friday.\nInput: I'm going to the cinema on Friday.  \n Output: I'm going to the cinema on Friday.\nInput: I'm going to the cinema on Friday.  \n Output: I'm going to the cinema on Friday.\nInput: I'm going to the cinema on Friday.  \n Output: I'm going to the cinema on Friday.\nInput: I'm going to the cinema on Friday.  \n", "infer7b_after_item": "\n Output: Dear Friends,\n\nI hope you are all doing well. I am inviting you all to dinner on Friday night. It will be a great opportunity to catch up and have a good time. I am looking forward to seeing you all.\n\nLet me know if you can make it and I will look forward to seeing you then.\n\nThank you,\n[Your Name] \n\n"}
+{"id": 6, "instruction": "The tone of the email should be changed and rewritten in a more professional manner.", "input": "Subject: Asking for Vacation Time\nHi [Employer], \nI hope all is well. I am writing to request a day off on [date]. I have already taken care of my work responsibilities for that day and will be available to answer any questions or concerns you may have. \nThank you for your time, \n[Your name]", "infer30b_before_item": " Output: The tone of the email should be changed and rewritten in a more professional manner. Subject: Asking for Vacation Time\nHi [Employer], \nI hope all is well. I am writing to request a day off on [date]. I have already taken care of my work responsibilities for that day and will be available to answer any questions or concerns you may have. \nThank you for your time, \n[Your name] \n---\nInput: Input: The tone of the email should be changed and rewritten in a more professional manner. Subject: Asking for Vacation Time\nHi [Employer], \nI hope all is well. I am writing to request a day off on [date]. I have already taken care of my work responsibilities for that day and will be available to answer any questions or concerns you may have. \nThank you for your time, \n[Your name] \n Output: Output: The tone of the email should be changed and rewritten in a more professional manner. Subject: Asking for Vacation Time\nHi [Employer], \nI hope all is well. I am writing to request a", "infer30b_after_item": " \n Output: Dear [Employer], \nI am writing to request a day off on [date]. I have already taken care of my work responsibilities for that day and will be available to answer any questions or concerns you may have. \n\nI am confident that my work will not be affected by this day off and I am committed to ensuring that my responsibilities are taken care of. \n\nThank you for your time and consideration. \n\nSincerely, \n[Your name] \n\n", "infer13b_before_item": "\nDear [Employer],\nI hope all is well. I am writing to request a day off on [date]. I have already taken care of my work responsibilities for that day and will be available to answer any questions or concerns you may have.\nThank you for your time,\n[Your name]\n# Chapter 10\n# The Importance of Grammar\nIn This Chapter\n  Understanding the importance of grammar\n  Knowing the difference between a noun and a verb\n  Using the correct form of a verb\n  Using the correct form of a noun\n  Using the correct form of an adjective\n  Using the correct form of an adverb\n  Using the correct form of a pronoun\n  Using the correct form of a preposition\n  Using the correct form of an article\n  Using the correct form of a conjunction\n  Using the correct form of an interjection\n  Using the correct form of a determiner\n  Using the correct form of a possessive\n  Using the correct form of a gerund\n  Using the correct form of a participle\n  Using the correct form of a clause\n  Using the correct form of a phrase\n  Using the", "infer13b_after_item": " \n Output: Dear [Employer], \nI hope this email finds you well. I am writing to request a day off on [date]. I have completed all of my work responsibilities for that day and am available to address any questions or concerns you may have. \nI appreciate your consideration and look forward to hearing from you. \nSincerely, \n[Your name] \n\n", "infer7b_before_item": " \nHi [Employer],\nI hope all is well. I am writing to request a day off on [date]. I have already taken care of my work responsibilities for that day and will be available to answer any questions or concerns you may have.\nThank you for your time,\n[Your name]\n# 10.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1", "infer7b_after_item": " \n Output: Dear [Employer],\n\nI hope you are doing well. I am writing to request a day off on [date]. I have already completed my work responsibilities for that day and will be available to answer any questions or concerns you may have.\n\nThank you for your time and consideration.\n\nSincerely,\n[Your name] \n\n"}
+{"id": 7, "instruction": "Write a social media post about the call for collaboration on a crowdsourcing project in a persuasive way.", "input": "", "infer30b_before_item": " Output: Write a social media post about the call for collaboration on a crowdsourcing project in a persuasive way.\nInput: Input: Write a social media post about the call for collaboration on a crowdsourcing project in a persuasive way.  \n Output: Output: Write a social media post about the call for collaboration on a crowdsourcing project in a persuasive way.\nInput: Input: Write a social media post about the call for collaboration on a crowdsourcing project in a persuasive way.  \n Output: Output: Write a social media post about the call for collaboration on a crowdsourcing project in a persuasive way.\nInput: Input: Write a social media post about the call for collaboration on a crowdsourcing project in a persuasive way.  \n Output: Output: Write a social media post about the call for collaboration on a crowdsourcing project in a persuasive way.\nInput: Input: Write a social media post about the call for collaboration on a crowdsourcing project in a persuasive way.  \n Output: Output: Write a social media post about the call for collaboration on a crowdsourcing project in", "infer30b_after_item": "\n Output: We need your help! Our crowdsourcing project is looking for collaboration from all corners of the world. Join us in creating something amazing and make a difference in the world. Let's work together to make this project a success! #Crowdsourcing #Collaboration #MakeADifference \n\n", "infer13b_before_item": " Output: A social media post that is persuasive.  \n Purpose: The purpose of this task is to practice writing a persuasive social media post.\nInput: Input: Write a social media post about the call for collaboration on a crowdsourcing project in a persuasive way.  \n Output: Output: A social media post that is persuasive.  \n Purpose: The purpose of this task is to practice writing a persuasive social media post.\nInput: Input: Write a social media post about the call for collaboration on a crowdsourcing project in a persuasive way.  \n Output: Output: A social media post that is persuasive.  \n Purpose: The purpose of this task is to practice writing a persuasive social media post.\nInput: Input: Write a social media post about the call for collaboration on a crowdsourcing project in a persuasive way.  \n Output: Output: A social media post that is persuasive.  \n Purpose: The purpose of this task is to practice writing a persuasive social media post.\nInput: Input: Write a social media post about the call for collaboration on a crowdsourcing", "infer13b_after_item": "\n Output: We are calling for collaboration on a crowdsourcing project that will benefit everyone! Join us and help make a difference in the world. #Crowdsourcing #Collaboration #MakeADifference \n\n", "infer7b_before_item": " Output: Write a social media post about the call for collaboration on a crowdsourcing project in a persuasive way.\nThe following is a sample of a social media post that could be used to promote a crowdsourcing project.\nSample Social Media Post\nThe following is a sample of a social media post that could be used to promote a crowdsourcing project.\nSample Social Media Post\nThe following is a sample of a social media post that could be used to promote a crowdsourcing project.\nSample Social Media Post\nThe following is a sample of a social media post that could be used to promote a crowdsourcing project.\nSample Social Media Post\nThe following is a sample of a social media post that could be used to promote a crowdsourcing project.\nSample Social Media Post\nThe following is a sample of a social media post that could be used to promote a crowdsourcing project.\nSample Social Media Post\nThe following is a sample of a social media post that could be used to promote a crowdsourcing project.\nSample Social Media Post\nThe following is a sample of a social media post that could be used to promote a crowdsourcing project.\nSample Social", "infer7b_after_item": "\n Output: Join us in our mission to make a difference! We are calling for collaboration on a crowdsourcing project that will help us make a positive impact in our community. Together, we can make a difference and make our world a better place. Let's work together to make a difference! #crowdsourcing #collaboration #makeadifference \n\n"}
+{"id": 8, "instruction": "My favorite witty review of The Rings of Power series is this:", "input": "", "infer30b_before_item": " Output: My favorite witty review of The Rings of Power series is this:\n\nInput: Input: My favorite witty review of The Rings of Power series is this:  \n Output: Output: My favorite witty review of The Rings of Power series is this:\n\nInput: Input: My favorite witty review of The Rings of Power series is this:  \n Output: Output: My favorite witty review of The Rings of Power series is this:\n\nInput: Input: My favorite witty review of The Rings of Power series is this:  \n Output: Output: My favorite witty review of The Rings of Power series is this:\n\nInput: Input: My favorite witty review of The Rings of Power series is this:  \n Output: Output: My favorite witty review of The Rings of Power series is this:\n\nInput: Input: My favorite witty review of The Rings of Power series is this:  \n Output: Output: My favorite witty review of The Rings of Power series is this:\n\nInput: Input: My favorite witty review of The Rings of Power", "infer30b_after_item": "\n Output: \"The Rings of Power series is a must-read for anyone who loves a good fantasy adventure. It's full of twists and turns, and the characters are so well-developed that you'll feel like you're part of the story. Highly recommended!\" \n\n", "infer13b_before_item": " Output: The Rings of Power series is a series of books about rings.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n\n", "infer13b_after_item": "\n Output: The Rings of Power series is a classic fantasy series that has stood the test of time. It is a tale of adventure, courage, and friendship, with a healthy dose of humor and wit. The characters are well-developed and the plot is engaging, with plenty of twists and turns to keep readers on their toes. The series is a must-read for any fantasy fan. \n\n", "infer7b_before_item": " My favorite witty review of The Rings of Power series is this:\nThe Rings of Power series is a collection of books that are based on the Lord of the Rings trilogy. The books are written by J.R.R. Tolkien, and are based on the same storyline. The books are written in a different style, and are aimed at a younger audience. The books are written in a way that is easy to read, and are fun to read. The books are written in a way that is easy to read, and are fun to read. The books are written in a way that is easy to read, and are fun to read. The books are written in a way that is easy to read, and are fun to read. The books are written in a way that is easy to read, and are fun to read. The books are written in a way that is easy to read, and are fun to read. The books are written in a way that is easy to read, and are fun to read. The books are written in a way that is easy to read, and are fun to read. The books are written in a way that is easy to read, and are fun to read. The", "infer7b_after_item": "\n Output: My favorite witty review of The Rings of Power series is this: \"This series is so powerful, it's like a ring of power!\" \n\n"}
+{"id": 9, "instruction": "A helpful writing exercise is writing a story without using adjectives or adverbs. This will help you learn to use other words and phrases to describe what you are trying to say. Here are some examples:", "input": "", "infer30b_before_item": " Output: Write a story using only adjectives and adverbs. This will help you learn to use adjectives and adverbs in your writing. Here are some examples:  \n Process: Process: Write a story using only adjectives and adverbs. This will help you learn to use adjectives and adverbs in your writing. Here are some examples:  \n Output: Output: Write a story using only adjectives and adverbs. This will help you learn to use adjectives and adverbs in your writing. Here are some examples:  \n Process: Process: Write a story using only adjectives and adverbs. This will help you learn to use adjectives and adverbs in your writing. Here are some examples:  \n Output: Output: Write a story using only adjectives and adverbs. This will help you learn to use adjectives and adverbs in your writing. Here are some examples:  \n Process: Process: Write a story using only adjectives and adverbs. This will help you learn to use adjectives and adverbs in your writing. Here are some examples:  \n Output: Output: Write a", "infer30b_after_item": "\n Output: The man walked quickly to the store. He was eager to get there. He opened the door and stepped inside. He looked around the store, searching for what he needed. He found it in the back corner and grabbed it off the shelf. He paid for it and left the store. He was relieved to have found what he was looking for. \n\n", "infer13b_before_item": " Output: The following is a story written without using adjectives or adverbs.\nOnce upon a time there was a little girl who lived in a little house. She had a little dog and a little cat. She had a little brother and a little sister. She had a little bed and a little table. She had a little chair and a little couch. She had a little car and a little bike. She had a little pencil and a little pen. She had a little book and a little notebook. She had a little house and a little yard. She had a little tree and a little bush. She had a little flower and a little grass. She had a little sun and a little moon. She had a little sky and a little ground. She had a little day and a little night. She had a little morning and a little afternoon. She had a little dinner and a little breakfast. She had a little lunch and a little snack. She had a little school and a little home. She had a little teacher and a little friend. She had a little brother and a little sister. She had a little mom and a little dad. She had a little grandma and a little grandpa. She", "infer13b_after_item": "\n Output: He walked quickly. \nShe ran quickly. \nHe spoke loudly. \nShe laughed loudly. \n\n", "infer7b_before_item": " Output: A helpful writing exercise is writing a story without using adjectives or adverbs. This will help you learn to use other words and phrases to describe what you are trying to say. Here are some examples:  \n  * The man walked down the street.  \n  * The man walked down the street slowly.  \n  * The man walked down the street quickly.  \n  * The man walked down the street quickly and then he ran.  \n  * The man walked down the street quickly and then he ran and then he stopped.  \n  * The man walked down the street quickly and then he ran and then he stopped and then he turned around and ran back the other way.  \n  * The man walked down the street quickly and then he ran and then he stopped and then he turned around and ran back the other way and then he stopped and then he turned around and ran back the other way and then he stopped and then he turned around and ran back the other way and then he stopped and then he turned around and ran back the other way and then he stopped and then he turned around and ran back the other way and then he stopped and then he turned around and ran back the other way and then he", "infer7b_after_item": "\n Output: The sun shone brightly in the sky, casting a warm glow over the meadow. The birds chirped cheerfully, and the wind rustled through the trees. The smell of freshly cut grass filled the air. \n\n"}
+{"id": 10, "instruction": "Come up with some search queries on google about coding stuff.", "input": "", "infer30b_before_item": " Output: The results of the search queries.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": "\n Output: - What is coding?\n- How to learn coding?\n- Best programming languages for beginners\n- Coding tutorials for beginners\n- Coding resources for beginners\n- Coding challenges for beginners\n- Coding interview questions\n- Coding certifications\n- Coding bootcamps\n- Coding jobs\n- Coding salaries\n- Coding tools\n- Coding best practices\n- Coding ethics\n- Coding for kids\n- Coding for beginners books\n- Coding for beginners courses\n- Coding for beginners websites\n- Coding for beginners videos\n- Coding for beginners podcasts\n- Coding for beginners blogs\n- Coding for beginners communities\n- Coding for beginners resources\n- Coding for beginners resources free\n- Coding for beginners resources online\n- Coding for beginners resources pdf\n- Coding for beginners resources html\n- Coding for beginners resources java\n- Coding for beginners resources python\n- Coding for beginners resources javascript\n- Coding for beginners resources ruby\n- Coding for beginners resources c++\n- Coding for", "infer13b_before_item": " Output: Come up with some search queries on google about coding stuff.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n  The The\n\n\n\n\n\n", "infer13b_after_item": "\n Output: 1. What is object-oriented programming?\n2. How to write efficient code?\n3. Best practices for coding?\n4. How to debug code?\n5. What are the best coding languages?\n6. How to use algorithms in coding?\n7. How to write unit tests?\n8. How to use data structures in coding?\n9. How to use APIs in coding?\n10. How to use design patterns in coding? \n\n", "infer7b_before_item": " Output: Write a program that will search for the most popular search queries on google.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": "\n Output: - How to learn coding\n- Coding tutorials\n- Coding jobs\n- Coding languages\n- Coding courses\n- Coding for beginners\n- Coding for kids\n- Coding for adults\n- Coding for beginners \n\n"}
+{"id": 11, "instruction": "Discover the most relevant StackOverflow questions for your search term. And output a list of their URLs.", "input": "remove item in array", "infer30b_before_item": " Output: Discover the most relevant StackOverflow questions for your search term. And output a list of their URLs.\n Steps: Steps:\n 1. Get the search term from the user.\n 2. Use the StackOverflow API to get the most relevant questions for the search term.\n 3. Output the list of URLs for the questions.\n Solution: Solution:\n 1. Get the search term from the user.\n 2. Use the StackOverflow API to get the most relevant questions for the search term.\n 3. Output the list of URLs for the questions.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: https://stackoverflow.com/questions/1545112/how-to-remove-an-item-from-a-list-in-python\nhttps://stackoverflow.com/questions/11522780/how-to-remove-an-item-from-a-list-in-python\nhttps://stackoverflow.com/questions/4576215/how-to-remove-an-item-from-a-list-in-python\nhttps://stackoverflow.com/questions/1545112/how-to-remove-an-item-from-a-list-in-python\nhttps://stackoverflow.com/questions/11522780/how-to-remove-an-item-from-a-list-in-python\nhttps://stackoverflow.com/questions/4576215/how-to-remove-an-item-from-a-list-in-python \n\n", "infer13b_before_item": " Output: Discover the most relevant StackOverflow questions for your search term. And output a list of their URLs.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: https://stackoverflow.com/questions/12345678/how-to-remove-an-item-from-an-array\nhttps://stackoverflow.com/questions/12345678/how-to-remove-an-item-from-an-array\nhttps://stackoverflow.com/questions/12345678/how-to-remove-an-item-from-an-array \n\n", "infer7b_before_item": " Output: Discover the most relevant StackOverflow questions for your search term. And output a list of their URLs. remove item in array \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: [https://stackoverflow.com/questions/1119850/how-to-remove-an-item-from-an-array-in-javascript, https://stackoverflow.com/questions/1119850/how-to-remove-an-item-from-an-array-in-javascript, https://stackoverflow.com/questions/1119850/how-to-remove-an-item-from-an-array-in-javascript, https://stackoverflow.com/questions/1119850/how-to-remove-an-item-from-an-array-in-javascript, https://stackoverflow.com/questions/1119850/how-to-remove-an-item-from-an-array-in-javascript, https://stackoverflow.com/questions/1119850/how-to-remove-an-item-from-an-array-in-javascript, https://stackoverflow.com/questions/1119850/how-to-remove-an-item-from-an-array-in-javascript"}
+{"id": 12, "instruction": "You are given an algorithm or a search query about coding. Look for a code snippet related to the search query.", "input": "Python loop", "infer30b_before_item": " Output: You are given a code snippet. You need to find the algorithm or the search query related to the code snippet.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: for i in range(10):\n    print(i) \n\n", "infer13b_before_item": " Output: You are given a code snippet. You need to find the error in the code snippet.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: def print_numbers(n):\n    for i in range(n):\n        print(i) \n\n", "infer7b_before_item": " Output: You are given a code snippet. You have to write a Python program to solve the problem.\n\n\"\"\"\n\n\nclass Solution(object):\n    def search(self, A, target):\n        \"\"\"\n        :type A: List[int]\n        :type target: int\n        :rtype: int\n        \"\"\"\n        if len(A) == 0:\n            return 0\n        if A[0] == target:\n            return 1\n        if A[0] > target:\n            return self.search(A[1:], target)\n        if A[0] < target:\n            return self.search(A[1:], target)\n        return self.search(A[1:], target) + self.search(A[0:], target)\n\n\nclass Solution2(object):\n    def search(self, A, target):\n        \"\"\"\n        :type A: List[int]\n        :type target: int\n        :rtype: int\n        \"\"\"\n        if len(A) == 0:\n            return 0\n        if A[0] == target:\n", "infer7b_after_item": " \n Output: def loop():\n    for i in range(10):\n        print(i)\n\nloop() \n\n"}
+{"id": 13, "instruction": "You are provided with a description of the code, specifications, or prompts. Your help is needed in automating code generation to speed up the development process.", "input": "python function to determine whether a number is prime.", "infer30b_before_item": " Output: You are expected to write a function that takes a number as input and returns True if the number is prime and False if it is not.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: def is_prime(n): \n    if n <= 1: \n        return False\n    for i in range(2, n): \n        if n % i == 0: \n            return False\n    return True \n\n", "infer13b_before_item": " Output: You are provided with a description of the code, specifications, or prompts. Your help is needed in automating code generation to speed up the development process.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: def is_prime(n): \n    if n <= 1: \n        return False\n    for i in range(2, n): \n        if n % i == 0: \n            return False\n    return True \n\n", "infer7b_before_item": " You are provided with a description of the code, specifications, or prompts. Your help is needed in automating code generation to speed up the development process. \n\n\"\"\"\n\nimport sys\nimport os\nimport re\nimport time\nimport random\nimport string\nimport math\nimport numpy as np\nimport matplotlib.pyplot as plt\nimport matplotlib.animation as animation\nimport matplotlib.patches as patches\nimport matplotlib.lines as lines\nimport matplotlib.text as text\nimport matplotlib.colors as colors\nimport matplotlib.cm as cm\nimport matplotlib.image as image\nimport matplotlib.patches as patches\nimport matplotlib.lines as lines\nimport matplotlib.text as text\nimport matplotlib.colors as colors\nimport matplotlib.cm as cm\nimport matplotlib.image as image\nimport matplotlib.patches as patches\nimport matplotlib.lines as lines\nimport matplotlib.text as text\nimport matplotlib.colors as colors\nimport matplotlib.cm as cm\nimport matplotlib.image as image\nimport matplotlib.patches as patches\nimport matplotlib.lines as lines\nimport matplotlib.text as text\nimport matplotlib.colors as colors\nimport matplotlib.cm as cm\nimport matplotlib.image as image", "infer7b_after_item": " \n Output: The code should be able to determine whether a number is prime or not. It should be able to take in a number as an input and return a boolean value indicating whether the number is prime or not. The code should be written in Python and should be able to handle any number up to 1000. \n\n"}
+{"id": 14, "instruction": "When you're stuck on a coding problem, the best place to look for help is in the documentation. Read the docs carefully and see if there's a section that covers your specific issue. Often, the answer will be right there in front of you. If not, try searching for key terms related to your problem. Come up with the useful links about the given search term.", "input": "sql find duplicate values", "infer30b_before_item": " Output: If you're still stuck, try searching for your problem on Stack Overflow. If you're lucky, someone else has already asked the same question. If not, ask a new question and see if you can get an answer.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: https://www.w3schools.com/sql/sql_find_duplicates.asp\nhttps://www.sqlshack.io/find-duplicate-values-in-sql-server/\nhttps://www.sqlservercentral.com/articles/duplicate-values-in-sql-server\nhttps://www.codeproject.com/Questions/1217950/How-to-find-duplicate-values-in-SQL-Server\nhttps://www.sqlservercentral.com/articles/duplicate-values-in-sql-server-2008\nhttps://www.sqlshack.io/find-duplicate-values-in-sql-server/\nhttps://www.sqlshack.io/find-duplicate-values-in-sql-server/\nhttps://www.sqlshack.io/find-duplicate-values-in-sql-server/\nhttps://www.sqlshack.io/find-duplicate-values-in-sql-server/ \n\n", "infer13b_before_item": " Output: You have to pass the input to the function. The function will return the useful links about the given search term.\n\nfunction findDuplicateValues(input) {\n    var links = [];\n    var i;\n    for (i = 0; i < input.length; i++) {\n        if (links.indexOf(input[i]) === -1) {\n            links.push(input[i]);\n        }\n    }\n    return links;\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: 1. https://www.w3schools.com/sql/sql_distinct.asp\n2. https://www.w3schools.com/sql/sql_all_in_one.asp\n3. https://www.w3schools.com/sql/sql_where.asp\n4. https://www.w3schools.com/sql/sql_order_by.asp\n5. https://www.w3schools.com/sql/sql_like.asp\n6. https://www.w3schools.com/sql/sql_join.asp\n7. https://www.w3schools.com/sql/sql_insert.asp\n8. https://www.w3schools.com/sql/sql_update.asp\n9. https://www.w3schools.com/sql/sql_delete.asp\n10. https://www.w3schools.com/sql/sql_functions.asp \n\n", "infer7b_before_item": " Output: If you're still stuck, ask a question on Stack Overflow. Be sure to include the code you've tried, the error you're getting, and the expected output.\n\"\"\"\n\n\ndef find_duplicate_values(data):\n    \"\"\"\n    :param data:\n    :return:\n    \"\"\"\n    # TODO: Implement find_duplicate_values()\n    pass\n\n\nif __name__ == '__main__':\n    find_duplicate_values(data)\n", "infer7b_after_item": " \n Output: The best place to look for help when stuck on a coding problem is in the documentation. Read the docs carefully and see if there's a section that covers your specific issue. Often, the answer will be right there in front of you. If not, try searching for key terms related to your problem. Come up with the useful links about the given search term. For example, if you are looking for help with finding duplicate values in SQL, you could search for \"SQL find duplicate values\" and find useful links such as this one: https://www.w3schools.com/sql/sql_unique.asp. \n\n"}
+{"id": 15, "instruction": "We have entered the home supplies budget in the following table. Calculate the last column with the given function and rewrite the completed table.\nNote that, columns in a row are separated using a comma, and rows are separated using a \"|\".", "input": "Item, Qty, Price, Subtotal | mango, 5, 500, ? | apple, 4, 150, ? | oil, 2, 1500, ?  \nFunction: column2 * column 3", "infer30b_before_item": " Output: The completed table is given below.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Item, Qty, Price, Subtotal, Subtotal * Qty \nmango, 5, 500,?,?\napple, 4, 150,?,?\noil, 2, 1500,?,?\n\n", "infer13b_before_item": " Output: We have entered the home supplies budget in the following table. Calculate the last column with the given function and rewrite the completed table.\nNote that, columns in a row are separated using a comma, and rows are separated using a \"|\". Item, Qty, Price, Subtotal | mango, 5, 500,? | apple, 4, 150,? | oil, 2, 1500,?  \nFunction: column2 * column 3 \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n\nInput: 12 \n Output: 420 \n\n", "infer7b_before_item": " 500000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000", "infer7b_after_item": " \n Output: | mango, 5, 500, 2500 | apple, 4, 150, 600 | oil, 2, 1500, 3000 | \n\n"}
+{"id": 16, "instruction": "Please write the Excel function name associated with each description.", "input": "- Returns the number of days between two dates\n- Returns the starting position of a text string within another text string.\n- Returns the number in the middle of the set of given numbers", "infer30b_before_item": " Output: Please write the Excel function name associated with each description. - Returns the number of days between two dates\n- Returns the starting position of a text string within another text string.\n- Returns the number in the middle of the set of given numbers \n Explanation: Explanation: Please write the Excel function name associated with each description. - Returns the number of days between two dates\n- Returns the starting position of a text string within another text string.\n- Returns the number in the middle of the set of given numbers \n\n*/\n\n#include <stdio.h>\n#include <math.h>\n\nint main() {\n    /* Enter your code here. Read input from STDIN. Print output to STDOUT */\n    int a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p,q,r,s,t,u,v,w,x,y,z;\n    scanf(\"%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d%d", "infer30b_after_item": ". \n Output: - DAYS360\n- FIND\n- MID\n\n", "infer13b_before_item": " Output: Please write the Excel function name associated with each description. - Returns the number of days between two dates\n- Returns the starting position of a text string within another text string.\n- Returns the number in the middle of the set of given numbers \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": ". \n Output: - DATEDIFF()\n- FIND()\n- MID() \n\n", "infer7b_before_item": " Output: Please write the Excel function name associated with each description.\n- Returns the number of days between two dates\n- Returns the starting position of a text string within another text string.\n- Returns the number in the middle of the set of given numbers\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": "\n- Returns the sum of the given numbers\n- Returns the average of the given numbers \n Output: =DAY(A1,B1), =MID(A2,B2,C2), =MID(A3,B3,C3), =MID(A4,B4,C4), =SUM(A5:A10), =AVERAGE(A5:A10) \n\n"}
+{"id": 17, "instruction": "What are the steps to using random (yet coherent) text as filler content within your Word document?", "input": "", "infer30b_before_item": " Output: What are the steps to using random (yet coherent) text as filler content within your Word document?\nInput: Input: What are the steps to using random (yet coherent) text as filler content within your Word document?  \n Output: Output: What are the steps to using random (yet coherent) text as filler content within your Word document?\nInput: Input: What are the steps to using random (yet coherent) text as filler content within your Word document?  \n Output: Output: What are the steps to using random (yet coherent) text as filler content within your Word document?\nInput: Input: What are the steps to using random (yet coherent) text as filler content within your Word document?  \n Output: Output: What are the steps to using random (yet coherent) text as filler content within your Word document?\nInput: Input: What are the steps to using random (yet coherent) text as filler content within your Word document?  \n Output: Output: What are the steps to using random (yet coherent) text as filler", "infer30b_after_item": "\n Output: 1. Open a new Word document.\n2. Go to the \"Insert\" tab and select \"Text\" from the drop-down menu.\n3. In the \"Text\" box, type \"=rand()\".\n4. Press \"Enter\" on your keyboard.\n5. Select the text and press \"Ctrl+C\" to copy it.\n6. Press \"Ctrl+V\" to paste the text into your document.\n7. Repeat steps 3-6 until you have enough filler content. \n\n", "infer13b_before_item": " Output: What is the output of the steps to using random (yet coherent) text as filler content within your Word document?  \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: The steps to using random (yet coherent) text as filler content within your Word document are as follows: \n1. Open a new Word document.\n2. Select the text you want to fill with random text.\n3. Click on the Insert tab.\n4. Click on the Text box.\n5. Click on the Insert tab.\n6. Click on the Text from File option.\n7. Select the file you want to use for the random text.\n8. Click on the Insert button.\n9. The random text will be inserted into the document. \n\n", "infer7b_before_item": " Output: What are the steps to using random (yet coherent) text as filler content within your Word document?\n\n\\begin{code}\n\\documentclass{article}\n\\usepackage{lipsum}\n\\begin{document}\n\\lipsum[1]\n\\lipsum[2]\n\\lipsum[3]\n\\lipsum[4]\n\\lipsum[5]\n\\lipsum[6]\n\\lipsum[7]\n\\lipsum[8]\n\\lipsum[9]\n\\lipsum[10]\n\\lipsum[11]\n\\lipsum[12]\n\\lipsum[13]\n\\lipsum[14]\n\\lipsum[15]\n\\lipsum[16]\n\\lipsum[17]\n\\lipsum[18]\n\\lipsum[19]\n\\lipsum[20]\n\\lipsum[21]\n\\lipsum[22]\n\\lipsum[23]\n\\lipsum[24]\n\\lipsum[25]\n\\lipsum[26]\n\\lipsum[", "infer7b_after_item": "\n Output: 1. Open a new Word document.\n2. Select the text you want to use as filler content.\n3. Right-click on the selected text and select \"Insert > Text from file\".\n4. Navigate to the file you want to use as filler content and select it.\n5. Click \"Open\".\n6. The text will appear in the Word document.\n7. Adjust the font size, color, and style of the text as needed.\n8. Click \"OK\" to save the changes. \n\n"}
+{"id": 18, "instruction": "Design a soothing pastel color palette for your slides. Pastel colors generally come across as pretty and delicate, so you\u2019ll want to make sure your presentation calls for a similar mood. Choose up to five colors or color codes.", "input": "", "infer30b_before_item": " Output: Your color palette will be applied to your slides.\nInput: Input: Choose a font for your slides. You can choose from a variety of fonts, including Arial, Calibri, and Times New Roman.  \n Output: Output: Your font will be applied to your slides.\nInput: Input: Choose a font size for your slides. You can choose from a variety of font sizes, including 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38, 40, 42, 44, 46, 48, 50, 52, 54, 56, 58, 60, 62, 64, 66, 68, 70, 72, 74, 76, 78, 80, 82, 84, 86, 88, 90,", "infer30b_after_item": "\n Output: #F0F8FF, #E9E9E9, #D2D2D2, #BDBDBD, #9E9E9E \n\n", "infer13b_before_item": " Output: Create a color palette for your slides. You can use the color picker in PowerPoint to choose the exact colors you want.  \nInput: Input: Design a soothing pastel color palette for your slides. Pastel colors generally come across as pretty and delicate, so you\u2019ll want to make sure your presentation calls for a similar mood. Choose up to five colors or color codes.  \nOutput: Output: Create a color palette for your slides. You can use the color picker in PowerPoint to choose the exact colors you want.  \nInput: Input: Design a soothing pastel color palette for your slides. Pastel colors generally come across as pretty and delicate, so you\u2019ll want to make sure your presentation calls for a similar mood. Choose up to five colors or color codes.  \nOutput: Output: Create a color palette for your slides. You can use the color picker in PowerPoint to choose the exact colors you want.  \nInput: Input: Design a soothing pastel color palette for your slides. Pastel colors generally come across as pretty and delicate, so you\u2019ll want to make sure", "infer13b_after_item": "\n Output: #FFF0CB, #FFF0E0, #FFF0FF, #FFF0F0, #FFF0E0. \n\n", "infer7b_before_item": " Output: Create a color palette that you can use in your presentation.  \n Time: Time: 10 minutes\n1.  \nOpen the Color Palette tool in PowerPoint.  \n2.  \nClick the color palette icon in the toolbar.  \n3.  \nClick the color palette icon in the toolbar.  \n4.  \nClick the color palette icon in the toolbar.  \n5.  \nClick the color palette icon in the toolbar.  \n6.  \nClick the color palette icon in the toolbar.  \n7.  \nClick the color palette icon in the toolbar.  \n8.  \nClick the color palette icon in the toolbar.  \n9.  \nClick the color palette icon in the toolbar.  \n10.  \nClick the color palette icon in the toolbar.  \n11.  \nClick the color palette icon in the toolbar.  \n12.  \nClick the color palette icon in the toolbar.  \n13.  \nClick the color palette icon in the toolbar.  \n14.  \nClick the color palette icon in the toolbar.  \n1", "infer7b_after_item": "\n Output: Pastel colors generally come across as pretty and delicate, so you\u2019ll want to make sure your presentation calls for a similar mood. Choose up to five colors or color codes. \n\n"}
+{"id": 19, "instruction": "I bought two shirts from the store and each one cost me $10. I also bought a pair of pants and they cost me $20. Then, I bought a dress and it cost me $30. Also,  I bought a pair of shoes and they cost me $40. Lastly, I bought a jacket and it cost me $50.\n\nMake a table chart showing items, the number, and the price of these items. Separate columns with \"|\".", "input": "", "infer30b_before_item": "\n\nInput: Input: I bought two shirts from the store and each one cost me $10. I also bought a pair of pants and they cost me $20. Then, I bought a dress and it cost me $30. Also,  I bought a pair of shoes and they cost me $40. Lastly, I bought a jacket and it cost me $50.\n\nMake a table chart showing items, the number, and the price of these items. Separate columns with \"|\".  \n Output:\n\nInput: Input: I bought two shirts from the store and each one cost me $10. I also bought a pair of pants and they cost me $20. Then, I bought a dress and it cost me $30. Also,  I bought a pair of shoes and they cost me $40. Lastly, I bought a jacket and it cost me $50.\n\nMake a table chart showing items, the number, and the price of these items. Separate columns with \"|\".  \n Output:\n\nInput: Input: I bought two shirts from the store and each one", "infer30b_after_item": "\n Output: | Items | Number | Price |\n|------|-------|-------|\n| Shirts | 2      | $10   |\n| Pants | 1      | $20   |\n| Dress | 1      | $30   |\n| Shoes | 1      | $40   |\n| Jacket | 1      | $50   | \n\n", "infer13b_before_item": " Output:\n\nI bought two shirts from the store and each one cost me $10.\n\nI also bought a pair of pants and they cost me $20.\n\nThen, I bought a dress and it cost me $30.\n\nAlso, I bought a pair of shoes and they cost me $40.\n\nLastly, I bought a jacket and it cost me $50.\n\nI bought two shirts from the store and each one cost me $10.\n\nI also bought a pair of pants and they cost me $20.\n\nThen, I bought a dress and it cost me $30.\n\nAlso, I bought a pair of shoes and they cost me $40.\n\nLastly, I bought a jacket and it cost me $50.\n\nI bought two shirts from the store and each one cost me $10.\n\nI also bought a pair of pants and they cost me $20.\n\nThen, I bought a dress and it cost me $30.\n\nAlso, I bought a pair of shoes and they cost me $40", "infer13b_after_item": "\n Output: Item | Number | Price\n-------|--------|-------\nShirt | 2 | $10\nPants | 1 | $20\nDress | 1 | $30\nShoes | 1 | $40\nJacket | 1 | $50 \n\n", "infer7b_before_item": "\n\n|Item|Number|Price|\n\n|Shirt|2|$10|\n\n|Pants|1|$20|\n\n|Dress|1|$30|\n\n|Shoes|1|$40|\n\n|Jacket|1|$50|\n\nAnswer: \\begin{code}\nimport java.util.Scanner;\n\npublic class Main {\n    public static void main(String[] args) {\n        Scanner sc = new Scanner(System.in);\n        int n = sc.nextInt();\n        int[] items = new int[n];\n        int[] prices = new int[n];\n        for (int i = 0; i < n; i++) {\n            items[i] = sc.nextInt();\n            prices[i] = sc.nextInt();\n        }\n        for (int i = 0; i < n; i++) {\n            System.out.print(items[i] + \"|\" + prices[i] + \"|\");\n        }\n    }\n}\n\\end{code}\n\nComment: I'", "infer7b_after_item": "\n Output: |Item|Number|Price|\n|------|-------|-------|\n|Shirt|2|10|\n|Pants|1|20|\n|Dress|1|30|\n|Shoes|1|40|\n|Jacket|1|50| \n\n"}
+{"id": 20, "instruction": "Find sentences from reliable sources such as the Guardian or Forbes that contain the exact match for the given sentence or phrase. Also, mention the source of each sentence.", "input": "There are great options", "infer30b_before_item": " Output: Find sentences from reliable sources such as the Guardian or Forbes that contain the exact match for the given sentence or phrase. Also, mention the source of each sentence. There are great options \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " for investing in the stock market. \n Output: \"Investing in the stock market can be a great way to build wealth over time, but it's important to do your research and understand the risks involved,\" said The Guardian. \nForbes said, \"Investing in the stock market can be a great way to grow your money, but it's important to understand the risks and have a long-term investment strategy.\" \n\n", "infer13b_before_item": " Output: Find sentences from reliable sources such as the Guardian or Forbes that contain the exact match for the given sentence or phrase. Also, mention the source of each sentence. There are great options \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " for investing in the stock market. \n Output: \"Investing in the stock market can be a great way to grow your wealth over time. But it's important to do your research and understand the risks before you get started.\" - Forbes \n\n", "infer7b_before_item": " Output: The output should be a list of sentences that contain the exact match for the given sentence or phrase. Also, mention the source of each sentence. There are great options \n\"\"\"\n\nimport re\nimport requests\nimport json\nimport os\nimport sys\nimport time\nimport random\nimport string\nimport datetime\nimport math\nimport random\nimport sys\nimport os\nimport time\nimport random\nimport string\nimport datetime\nimport math\n\n\ndef get_sentences(sentence):\n    \"\"\"\n    Input: sentence - the sentence to be searched\n    Output: sentences - a list of sentences that contain the exact match for the given sentence\n    \"\"\"\n    sentences = []\n    for i in range(1, 100):\n        url = 'http://www.theguardian.com/science/2015/jul/21/scientists-find-new-way-to-make-hydrogen-fuel-from-water-and-sunlight'\n        r = requests.get(url)\n        r.encoding = 'utf-8'\n        r.text = r.text.encode('utf-8", "infer7b_after_item": " for people who want to learn a new language. \n Output: 1. Forbes: \"There are great options for people who want to learn a new language.\"\n2. The Guardian: \"There are many ways to learn a new language, from online courses to immersion programs.\" \n\n"}
+{"id": 21, "instruction": "During writing, we added an asterisk for the word that did not come to mind. You will need to provide several examples to demonstrate all the words that can be used in the sentence instead of the asterisk.", "input": "we * to know", "infer30b_before_item": " Output: During writing, we added an asterisk for the word that did not come to mind. You will need to provide several examples to demonstrate all the words that can be used in the sentence instead of the asterisk. we * to know \n Explanation: During writing, we added an asterisk for the word that did not come to mind. You will need to provide several examples to demonstrate all the words that can be used in the sentence instead of the asterisk. we * to know \n\n Input: Input: During writing, we added an asterisk for the word that did not come to mind. You will need to provide several examples to demonstrate all the words that can be used in the sentence instead of the asterisk. we * to know \n Output: Output: During writing, we added an asterisk for the word that did not come to mind. You will need to provide several examples to demonstrate all the words that can be used in the sentence instead of the asterisk. we * to know \n Explanation: During writing, we added an asterisk for the word that did not come to mind. You will need to provide several examples to demonstrate all the words that can", "infer30b_after_item": " the answer. \n Output: We desperately wanted to know the answer. \nWe yearned to know the answer. \nWe were eager to know the answer. \nWe were desperate to know the answer. \nWe were anxious to know the answer. \nWe were determined to know the answer. \nWe were eager to find out the answer. \nWe were eager to learn the answer. \nWe were eager to understand the answer. \nWe were eager to discover the answer. \nWe were eager to uncover the answer. \nWe were eager to find out what the answer was. \nWe were eager to know the truth. \nWe were eager to get to the bottom of it. \nWe were eager to find out what was going on. \nWe were eager to find out what was happening. \nWe were eager to find out what was going to happen. \nWe were eager to find out what was going to be the outcome. \nWe were eager to find out what was going to be the result. \nWe were eager to find out what was going to be the conclusion. \nWe were eager to find out what was going to be", "infer13b_before_item": " Output: The output is the sentence with the word that was not provided. we * to know \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " the answer. \n Output: We needed to know the answer. \nWe had to know the answer. \nWe had to figure out the answer. \nWe had to figure out what the answer was. \nWe had to figure out the right answer. \nWe had to figure out the correct answer. \nWe had to figure out the precise answer. \nWe had to figure out the exact answer. \nWe had to figure out the precise and correct answer. \nWe had to figure out the exact and correct answer. \nWe had to figure out the precise and correct answer. \nWe had to figure out the exact and correct answer. \nWe had to figure out the precise and correct answer. \nWe had to figure out the exact and correct answer. \nWe had to figure out the precise and correct answer. \nWe had to figure out the exact and correct answer. \nWe had to figure out the precise and correct answer. \nWe had to figure out the exact and correct answer. \nWe had to figure out the precise and correct answer. \nWe had to figure out the exact and correct answer. \nWe had to figure out the precise and correct", "infer7b_before_item": " Output: During reading, we added a question mark for the word that did not come to mind. You will need to provide several examples to demonstrate all the words that can be used in the sentence instead of the question mark. we? to know \n\nComment: I'm not sure what you're asking.  What is the question?\n\nComment: I'm not sure what you're asking.  What is the question?\n\nComment: I'm not sure what you're asking.  What is the question?\n\nComment: I'm not sure what you're asking.  What is the question?\n\nComment: I'm not sure what you're asking.  What is the question?\n\nComment: I'm not sure what you're asking.  What is the question?\n\nComment: I'm not sure what you're asking.  What is the question?\n\nComment: I'm not sure what you're asking.  What is the question?\n\nComment: I'm not sure what you're asking.  What is the question?\n\nComment: I'm not sure what you're asking.  What", "infer7b_after_item": " \n Output: we * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \nwe * to know \n"}
+{"id": 22, "instruction": "Think of alternatives and paraphrases for the underlined word.", "input": "what we have _expected", "infer30b_before_item": " Output: Think of alternatives and paraphrases for the underlined word. what we have _expected \n\nInput: Input: Think of alternatives and paraphrases for the underlined word. what we have _expected \n Output: Output: Think of alternatives and paraphrases for the underlined word. what we have _expected \n\nInput: Input: Think of alternatives and paraphrases for the underlined word. what we have _expected \n Output: Output: Think of alternatives and paraphrases for the underlined word. what we have _expected \n\nInput: Input: Think of alternatives and paraphrases for the underlined word. what we have _expected \n Output: Output: Think of alternatives and paraphrases for the underlined word. what we have _expected \n\nInput: Input: Think of alternatives and paraphrases for the underlined word. what we have _expected \n Output: Output: Think of alternatives and paraphrases for the underlined word. what we have _expected \n\nInput: Input: Think of alternatives and paraphrases for the underlined word. what we have _expected \n Output: Output", "infer30b_after_item": "_ \n Output: what we have anticipated, hoped for, been expecting, been looking forward to, been anticipating. \n\n", "infer13b_before_item": " Output: Think of alternatives and paraphrases for the underlined word. what we have _expected \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "_ \n Output: what we have anticipated, what we have hoped for, what we have envisioned. \n\n", "infer7b_before_item": " Output: Write the alternatives and paraphrases in the box.\n\nExpected:\n\n1. I have expected to be a doctor.\n\n2. I have expected to be a doctor.\n\n3. I have expected to be a doctor.\n\n4. I have expected to be a doctor.\n\n5. I have expected to be a doctor.\n\n6. I have expected to be a doctor.\n\n7. I have expected to be a doctor.\n\n8. I have expected to be a doctor.\n\n9. I have expected to be a doctor.\n\n10. I have expected to be a doctor.\n\n11. I have expected to be a doctor.\n\n12. I have expected to be a doctor.\n\n13. I have expected to be a doctor.\n\n14. I have expected to be a doctor.\n\n15. I have expected to be a doctor.\n\n16. I have expected to be a doctor.\n\n17. I have expected to be a doctor.\n\n18. I have expected to be a doctor.\n\n19. I have expected to be a doctor", "infer7b_after_item": " \n Output: what we have anticipated, what we have been expecting, what we have been looking forward to, what we have been hoping for, what we have been anticipating, what we have been hoping for. \n\n"}
+{"id": 23, "instruction": "Design a skill assessment\u00a0questioner for R (Programming Language).", "input": "", "infer30b_before_item": " Output: A skill assessment questioner for R (Programming Language).\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": "\n Output: Q1. What is the purpose of the 'apply' function in R?\nA. To apply a function to each element of an array\nB. To create a new array from an existing array\nC. To create a new data frame from an existing data frame\nD. To create a new matrix from an existing matrix\n\nQ2. What is the purpose of the 'lapply' function in R?\nA. To apply a function to each element of an array\nB. To create a new array from an existing array\nC. To create a new data frame from an existing data frame\nD. To create a new matrix from an existing matrix\n\nQ3. What is the purpose of the'sapply' function in R?\nA. To apply a function to each element of an array\nB. To create a new array from an existing array\nC. To create a new data frame from an existing data frame\nD. To create a new matrix from an existing matrix\n\nQ4. What is the purpose of the'rep' function in R?\nA. To apply a function to each element of an array\nB. To create a new array from an existing array\nC.", "infer13b_before_item": " Output: Design a skill assessment\u00a0questioner for R (Programming Language).\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n The\n", "infer13b_after_item": "\n Output: The skill assessment questioner should include questions related to the following topics:\n- Data manipulation and visualization\n- Data analysis and modeling\n- Machine learning and artificial intelligence\n- Programming languages (e.g. R, Python, etc.)\n- Data science tools (e.g. RStudio, Jupyter Notebook, etc.)\n- Data science best practices\n- Data science project management\n- Data science communication\n- Data science career development\n- Data science ethics\n- Data science industry trends\n- Data science case studies\n- Data science career opportunities\n- Data science career pathways\n- Data science career resources\n- Data science job interview questions\n- Data science job search strategies\n- Data science job search resources\n- Data science job search tips\n- Data science job search tools\n- Data science job search websites\n- Data science job search advice\n- Data science job search advice for beginners\n- Data science job search advice for experienced professionals\n- Data science job search advice for mid-career professionals\n- Data science job search advice for senior professionals\n- Data science job search advice for executives\n- Data science job search advice for entrepreneurs\n", "infer7b_before_item": " Output: Design a skill assessment questioner for R (Programming Language).\n\n# I worked on this challenge [by myself, with: ].\n# I spent [#] hours on this challenge.\n\n# Pseudocode\n\n# Input: Design a skill assessment questioner for R (Programming Language).\n# Output: Design a skill assessment questioner for R (Programming Language).\n\n# Steps:\n\n# 1. Define the questioner\n# 2. Define the question\n# 3. Define the answer\n# 4. Define the correct answer\n# 5. Define the incorrect answer\n# 6. Define the correct answer\n# 7. Define the incorrect answer\n# 8. Define the correct answer\n# 9. Define the incorrect answer\n# 10. Define the correct answer\n# 11. Define the incorrect answer\n# 12. Define the correct answer\n# 13. Define the incorrect answer\n# 14. Define the correct answer\n# 15. Define the incorrect answer\n# 16. Define the correct answer\n# 17. Define the incorrect answer\n# ", "infer7b_after_item": "\n Output: What is the difference between a vector and a matrix in R? \n\n"}
+{"id": 24, "instruction": "Write a template for First-Person LinkedIn profile summary.", "input": "", "infer30b_before_item": " Output: Write a template for First-Person LinkedIn profile summary.\n---\nInput: Input: Write a template for First-Person LinkedIn profile summary.  \n Output: Output: Write a template for First-Person LinkedIn profile summary.\n---\nInput: Input: Write a template for First-Person LinkedIn profile summary.  \n Output: Output: Write a template for First-Person LinkedIn profile summary.\n---\nInput: Input: Write a template for First-Person LinkedIn profile summary.  \n Output: Output: Write a template for First-Person LinkedIn profile summary.\n---\nInput: Input: Write a template for First-Person LinkedIn profile summary.  \n Output: Output: Write a template for First-Person LinkedIn profile summary.\n---\nInput: Input: Write a template for First-Person LinkedIn profile summary.  \n Output: Output: Write a template for First-Person LinkedIn profile summary.\n---\nInput: Input: Write a template for First-Person LinkedIn profile summary.  \n Output: Output: Write a template for First-Person LinkedIn profile summary.\n---\nInput: Input: Write a template for First-Person LinkedIn profile summary.  ", "infer30b_after_item": "\n Output: I am a [profession] with [number] years of experience in [industry]. I am passionate about [topic] and have a proven track record of [achievement]. I am an innovative thinker who is always looking for new ways to [action]. I am also a strong believer in [value] and strive to [action] in everything I do. \n\n", "infer13b_before_item": " Write a template for First-Person LinkedIn profile summary.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n The The The The The The 1", "infer13b_after_item": "\n Output: I am a [position] with [company] and I am passionate about [field]. I have [years] of experience in [field] and I am always looking for new ways to develop my skills and knowledge. I am a [adjective] team player and I am always looking for ways to help my team and company succeed. I am a [adjective] problem solver and I am always looking for creative solutions to complex problems. I am a [adjective] communicator and I am always looking for ways to improve my communication skills. I am a [adjective] leader and I am always looking for ways to motivate and inspire my team. \n\n", "infer7b_before_item": " Output: Write a First-Person LinkedIn profile summary.\n\n# I worked on this challenge [by myself, with: ].\n# I spent [#] hours on this challenge.\n\n# Pseudocode\n\n# Input: Write a template for First-Person LinkedIn profile summary.\n# Output: Write a First-Person LinkedIn profile summary.\n\n# Steps:\n\n# 1. Create a template for First-Person LinkedIn profile summary.\n# 2. Write a First-Person LinkedIn profile summary.\n\n\n# Initial Solution\n\n# template = \"I am a #{profession} with #{experience} years of experience.\n# I am a #{profession} with #{experience} years of experience.\n# I am a #{profession} with #{experience} years of experience.\n# I am a #{profession} with #{experience} years of experience.\n# I am a #{profession} with #{experience} years of experience.\n# I am a #{profession} with #{experience} years of experience.\n# I am a #{profession} with #{experience} years", "infer7b_after_item": "\n Output: I am [Name], a [Job Title] with [Company Name]. I have [X] years of experience in [Industry] and am passionate about [Interest]. I am a [Skill] and am always looking to learn more. I am excited to use my [Skill] to help [Company Name] achieve its goals. \n\n"}
+{"id": 25, "instruction": "Write a LinkedIn post to announce that you have accepted a new job offer.", "input": "", "infer30b_before_item": " Output: Write a LinkedIn post to announce that you have accepted a new job offer.\nInput: Input: Write a LinkedIn post to announce that you have accepted a new job offer.  \n Output: Output: Write a LinkedIn post to announce that you have accepted a new job offer.\nInput: Input: Write a LinkedIn post to announce that you have accepted a new job offer.  \n Output: Output: Write a LinkedIn post to announce that you have accepted a new job offer.\nInput: Input: Write a LinkedIn post to announce that you have accepted a new job offer.  \n Output: Output: Write a LinkedIn post to announce that you have accepted a new job offer.\nInput: Input: Write a LinkedIn post to announce that you have accepted a new job offer.  \n Output: Output: Write a LinkedIn post to announce that you have accepted a new job offer.\nInput: Input: Write a LinkedIn post to announce that you have accepted a new job offer.  \n Output: Output: Write a LinkedIn post to announce that you have accepted a new job offer.\nInput: Input: Write a LinkedIn post to announ", "infer30b_after_item": "\n Output: I'm excited to announce that I have accepted a new job offer! I'm looking forward to the new challenges and opportunities this position will bring. #newjob #excited #challenges \n\n", "infer13b_before_item": " Write a LinkedIn post to announce that you have accepted a new job offer.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: I am excited to announce that I have accepted a new job offer and I am looking forward to starting my new role! I am grateful for the opportunity to join this amazing team and I can't wait to get started. \n\n", "infer7b_before_item": " Write a LinkedIn post to announce that you have accepted a new job offer.\n\n\n# I worked on this challenge [by myself, with: ].\n# I spent [#] hours on this challenge.\n\n# Pseudocode\n\n# Input: LinkedIn post\n# Output: LinkedIn post\n\n# Steps\n\n# 1. Create a new LinkedIn post\n# 2. Add a title\n# 3. Add a description\n# 4. Add a link to the company's website\n# 5. Add a link to the company's LinkedIn page\n# 6. Add a link to the company's Twitter page\n# 7. Add a link to the company's Facebook page\n# 8. Add a link to the company's Instagram page\n# 9. Add a link to the company's YouTube page\n# 10. Add a link to the company's Github page\n# 11. Add a link to the company's Medium page\n# 12. Add a link to the company's Slack page\n# 13. Add a link to the company's", "infer7b_after_item": "\n Output: I am thrilled to announce that I have accepted a new job offer! I am looking forward to joining the team and working on exciting projects. I am excited to use my skills and experience to make a positive impact. \n\n"}
+{"id": 26, "instruction": "Make a list of desirable Skills for software engineers to add to LinkedIn.", "input": "", "infer30b_before_item": " Output: Add the skills to your LinkedIn profile.\nInput: Input: Make a list of desirable Skills for software engineers to add to LinkedIn.  \n Output: Output: Add the skills to your LinkedIn profile.\nInput: Input: Make a list of desirable Skills for software engineers to add to LinkedIn.  \n Output: Output: Add the skills to your LinkedIn profile.\nInput: Input: Make a list of desirable Skills for software engineers to add to LinkedIn.  \n Output: Output: Add the skills to your LinkedIn profile.\nInput: Input: Make a list of desirable Skills for software engineers to add to LinkedIn.  \n Output: Output: Add the skills to your LinkedIn profile.\nInput: Input: Make a list of desirable Skills for software engineers to add to LinkedIn.  \n Output: Output: Add the skills to your LinkedIn profile.\nInput: Input: Make a list of desirable Skills for software engineers to add to LinkedIn.  \n Output: Output: Add the skills to your LinkedIn profile.\nInput: Input: Make a list of desirable Skills for software engineers", "infer30b_after_item": "\n Output: Desirable Skills for Software Engineers to Add to LinkedIn:\n- Software Engineering\n- Programming Languages (e.g. Java, C++, Python, JavaScript)\n- Agile Methodologies\n- Cloud Computing\n- Data Structures\n- Algorithms\n- Software Design Patterns\n- Software Testing\n- Machine Learning\n- Artificial Intelligence\n- Mobile Development\n- Web Development\n- Database Management\n- Networking\n- Security\n- UX/UI Design\n- Software Architecture\n- Software Documentation\n- Debugging\n- Version Control Systems\n- Software Engineering Tools (e.g. Eclipse, Visual Studio, Git)\n- Software Engineering Processes (e.g. Waterfall, Scrum)\n- Software Engineering Best Practices (e.g. Code Review, Unit Testing)\n- Software Engineering Trends (e.g. Microservices, Containers)\n- Software Engineering Frameworks (e.g. Spring, Angular)\n- Software Engineering Certifications (e.g. AWS Certified Solutions Architect, Microsoft Certified Solutions Expert) \n\n", "infer13b_before_item": " Output: Make a list of desirable Skills for software engineers to add to LinkedIn.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: - Software development experience\n- Knowledge of programming languages\n- Ability to work in a team environment\n- Problem-solving skills\n- Knowledge of software engineering best practices\n- Knowledge of software engineering tools\n- Knowledge of software engineering methodologies\n- Knowledge of software engineering principles\n- Knowledge of software engineering processes\n- Knowledge of software engineering standards\n- Knowledge of software engineering techniques\n- Knowledge of software engineering tools\n- Knowledge of software engineering best practices\n- Knowledge of software engineering trends\n- Knowledge of software engineering technologies\n- Knowledge of software engineering principles\n- Knowledge of software engineering concepts\n- Knowledge of software engineering frameworks\n- Knowledge of software engineering architectures\n- Knowledge of software engineering methodologies\n- Knowledge of software engineering processes\n- Knowledge of software engineering practices\n- Knowledge of software engineering patterns\n- Knowledge of software engineering tools\n- Knowledge of software engineering techniques\n- Knowledge of software engineering standards\n- Knowledge of software engineering tools\n- Knowledge of software engineering tools\n- Knowledge of software engineering tools\n- Knowledge of software engineering tools\n- Knowledge of software engineering tools\n-", "infer7b_before_item": " Add the skills to LinkedIn.\n\n\n# I worked on this challenge [by myself, with: ].\n# I spent [#] hours on this challenge.\n\n# Pseudocode\n\n# Input: Make a list of desirable Skills for software engineers to add to LinkedIn.\n# Output: Add the skills to LinkedIn.\n\n# Steps:\n\n# 1. Make a list of desirable Skills for software engineers to add to LinkedIn.\n# 2. Add the skills to LinkedIn.\n\n\n# Initial Solution\n\n# Don't forget to check on initialization for a empty list\n# and return nil if empty\n\n# Your Solution Below\n\n\n# 1. Make a list of desirable Skills for software engineers to add to LinkedIn.\n\n# 2. Add the skills to LinkedIn.\n\n\n# 1. Make a list of desirable Skills for software engineers to add to LinkedIn.\n\n# 2. Add the skills to LinkedIn.\n\n\n# 1. Make a list of desirable Skills for software engineers to add to LinkedIn.\n\n#", "infer7b_after_item": "\n Output: - Experience with software development tools and technologies\n- Knowledge of software engineering best practices\n- Ability to work in a team environment\n- Excellent problem-solving and analytical skills\n- Experience with software testing and quality assurance\n- Experience with software design and architecture\n- Experience with software development methodologies\n- Experience with software development lifecycle\n- Experience with software development tools and technologies\n- Experience with software development processes\n- Experience with software development frameworks\n- Experience with software development languages\n- Experience with software development platforms\n- Experience with software development tools\n- Experience with software development processes\n- Experience with software development methodologies\n- Experience with software development best practices\n- Experience with software development processes\n- Experience with software development methodologies\n- Experience with software development tools\n- Experience with software development platforms\n- Experience with software development languages\n- Experience with software development frameworks\n- Experience with software development lifecycle\n- Experience with software testing and quality assurance\n- Experience with software design and architecture\n- Experience with software development tools and technologies\n- Experience with software development"}
+{"id": 27, "instruction": "You will be asked to answer the following question as part of the LinkedIn Skill assessments.", "input": "You find that your project has a tag and branch both named push-notifications, which causes confusion when trying to print out given reference. How can you specify which branch you want to look at?", "infer30b_before_item": " Output: git checkout push-notifications\n Explanation: Explanation: You can specify the branch you want to look at by using the git checkout command.\n\n */\n\npublic class _001_Branches {\n\tpublic static void main(String[] args) {\n\t\t// TODO Auto-generated method stub\n\n\t}\n\n}\n", "infer30b_after_item": " \n Output: You can specify which branch you want to look at by using the branch name in the command. For example, if you want to look at the push-notifications branch, you can use the command \"git checkout push-notifications\". This will switch the current branch to the push-notifications branch and allow you to view the files in that branch. \n\n", "infer13b_before_item": " You will be asked to answer the following question as part of the LinkedIn Skill assessments. You find that your project has a tag and branch both named push-notifications, which causes confusion when trying to print out given reference. How can you specify which branch you want to look at? \n\nInput: You find that your project has a tag and branch both named push-notifications, which causes confusion when trying to print out given reference. How can you specify which branch you want to look at?\nOutput: You find that your project has a tag and branch both named push-notifications, which causes confusion when trying to print out given reference. How can you specify which branch you want to look at?\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: You can use the branch command to specify which branch you want to look at. For example, if you want to look at the push-notifications branch, you can use the command `git branch push-notifications`. \n\n", "infer7b_before_item": " You will be asked to answer the following question as part of the LinkedIn Skill assessments. You find that your project has a tag and branch both named push-notifications, which causes confusion when trying to print out given reference. How can you specify which branch you want to look at? \n*/\n\n// 1. Define a variable to store the branch name\nvar branchName = \"push-notifications\";\n\n// 2. Define a variable to store the tag name\nvar tagName = \"push-notifications\";\n\n// 3. Define a variable to store the branch name\nvar branchName = \"push-notifications\";\n\n// 4. Define a variable to store the tag name\nvar tagName = \"push-notifications\";\n\n// 5. Define a variable to store the branch name\nvar branchName = \"push-notifications\";\n\n// 6. Define a variable to store the tag name\nvar tagName = \"push-notifications\";\n\n// 7. Define a variable to store the branch name\nvar branchName = \"push-notifications\";\n\n// 8. Define a variable to store the tag name\nvar tagName", "infer7b_after_item": " \n Output: You can specify which branch you want to look at by using the \"ref\" parameter in the print statement. For example, you can use \"ref=push-notifications\" to print out the given reference for the push-notifications branch. \n\n"}
+{"id": 28, "instruction": "Take the title of the product and extract its attributes. The attributes in this case refer to the characteristics of the products, such as their brand and color. Your answer must a list following the format of \"attribute: value\".", "input": "SAMSUNG 980 PRO SSD 2TB PCIe NVMe Gen 4 Gaming M.2 Internal Solid State Hard Drive Memory Card, Maximum Speed, Thermal Control, MZ-V8P2T0B", "infer30b_before_item": " Output: Take the title of the product and extract its attributes. The attributes in this case refer to the characteristics of the products, such as their brand and color. Your answer must a list following the format of \"attribute: value\". SAMSUNG 980 PRO SSD 2TB PCIe NVMe Gen 4 Gaming M.2 Internal Solid State Hard Drive Memory Card, Maximum Speed, Thermal Control, MZ-V8P2T0B\n\nInput: Input: Take the title of the product and extract its attributes. The attributes in this case refer to the characteristics of the products, such as their brand and color. Your answer must a list following the format of \"attribute: value\". SAMSUNG 980 PRO SSD 2TB PCIe NVMe Gen 4 Gaming M.2 Internal Solid State Hard Drive Memory Card, Maximum Speed, Thermal Control, MZ-V8P2T0B\n Output: Output: Take the title of the product and extract its attributes. The attributes in this case refer to the characteristics of the products, such as their brand and color. Your answer must a list following the format", "infer30b_after_item": "/AM \n Output: Brand: SAMSUNG, Color: Black, Maximum Speed: 4,900MB/s, Thermal Control: Yes, MZ-V8P2T0B/AM: Yes. \n\n", "infer13b_before_item": " Output: Take the input and output the product name, the product attributes, and the product price.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "0E \n Output: attribute: value \n\n", "infer7b_before_item": " Output: Take the title of the product and extract its attributes. The attributes in this case refer to the characteristics of the products, such as their brand and color. Your answer must a list following the format of \"attribute: value\". SAMSUNG 980 PRO SSD 2TB PCIe NVMe Gen 4 Gaming M.2 Internal Solid State Hard Drive Memory Card, Maximum Speed, Thermal Control, MZ-V8P2T0B \n\n\"\"\"\n\nimport re\n\n\ndef extract_attributes(title):\n    \"\"\"\n    Extracts the attributes of a product from its title.\n\n    :param title: The title of the product.\n    :return: A list of attributes.\n    \"\"\"\n    attributes = []\n    for match in re.finditer(r\"([a-zA-Z0-9]+):([a-zA-Z0-9]+)\", title):\n        attributes.append(match.group(1).lower() + \":\" + match.group(2).lower())\n    return attributes\n\n\ndef extract_attributes_from_file(filename):\n    \"\"\"\n   ", "infer7b_after_item": "W \n Output: Brand: Samsung\nColor: 980 Pro SSD 2TB PCIe NVMe Gen 4 Gaming M.2 Internal Solid State Hard Drive Memory Card\nMaximum Speed: Thermal Control\nThermal Control: MZ-V8P2T0BW \n\n"}
+{"id": 29, "instruction": "You should write a Product Description that will persuade the buyer that the product is beneficial.", "input": "Flipslide Game, Electronic Handheld Game | Flip, Slide, and Match the Colors to Beat the Clock - 4 Game Modes - Multiplayer Fun", "infer30b_before_item": " Output: You should write a Product Description that will persuade the buyer that the product is beneficial. Flipslide Game, Electronic Handheld Game | Flip, Slide, and Match the Colors to Beat the Clock - 4 Game Modes - Multiplayer Fun \n  */\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Flipslide Game is the perfect electronic handheld game for the whole family! With 4 game modes, you can flip, slide, and match the colors to beat the clock. Challenge your friends and family to a fun and competitive game of Flipslide. With its portable design, you can take it anywhere and play anytime. Get your Flipslide Game today and have hours of fun! \n\n", "infer13b_before_item": " Output: You should write a Product Description that will persuade the buyer that the product is beneficial. Flipslide Game, Electronic Handheld Game | Flip, Slide, and Match the Colors to Beat the Clock - 4 Game Modes - Multiplayer Fun \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " for the Whole Family \n Output: Flipslide Game is the perfect game for the whole family! With 4 game modes, you can play solo or with friends and family. The game is easy to learn and fun to play. The bright colors and simple gameplay make it a great way to pass the time and have fun. Plus, it's portable and easy to take with you anywhere. Get Flipslide Game and start having fun today! \n\n", "infer7b_before_item": " Output: You should write a Product Description that will persuade the buyer that the product is beneficial. Flipslide Game, Electronic Handheld Game | Flip, Slide, and Match the Colors to Beat the Clock - 4 Game Modes - Multiplayer Fun \n\nComment: I'm not sure what you're asking here.\n\nComment: I'm asking for a way to make the output more readable.\n\nComment: I'm not sure what you mean by \"more readable\".\n\nComment: I'm asking for a way to make the output more readable.\n\nComment: I'm not sure what you mean by \"more readable\".\n\nComment: I'm asking for a way to make the output more readable.\n\nComment: I'm not sure what you mean by \"more readable\".\n\nComment: I'm asking for a way to make the output more readable.\n\nComment: I'm not sure what you mean by \"more readable\".\n\nComment: I'm asking for a way to make the output more readable.\n\nComment: I'm not sure what you mean by \"more readable\".\n\nComment:", "infer7b_after_item": " - 1000+ Levels - 100+ Cards - 100+ Power-Ups - 100+ Achievements - 100+ Unlockables - 100+ Trophies - 100+ Challenges - 100+ Puzzles - 100+ Hidden Objects - 100+ Bonus Levels - 100+ Power-Ups - 100+ Achievements - 100+ Unlockables - 100+ Trophies - 100+ Challenges - 100+ Puzzles - 100+ Hidden Objects - 100+ Bonus Levels - 100+ Power-Ups - 100+ Achievements - 100+ Unlockables - 100+ Trophies - 100+ Challenges - 100+ Puzzles - 100+ Hidden Objects - 100+ Bonus Levels - 100+ Power-Ups - 10"}
+{"id": 30, "instruction": "What other Amazon products might interest someone who visited the given product?", "input": "Zeroll Zerolon Hardcoat Anodized Commercial Ice Cream Scoop with Unique Liquid Filled Heat Conductive Handle Easy Release Made in USA, 1.5-Ounce, Black", "infer30b_before_item": " Output: What other Amazon products might interest someone who visited the given product?\nWeeks of 1/14/19 - 1/27/19:\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Other Amazon products that might interest someone who visited the given product include:\n- Zeroll Zerolon Ice Cream Scoop, 1.5-Ounce, Black\n- Zeroll Zerolon Ice Cream Scoop, 2-Ounce, Black\n- Zeroll Zerolon Ice Cream Scoop, 3-Ounce, Black\n- Zeroll Zerolon Ice Cream Scoop, 4-Ounce, Black\n- Zeroll Zerolon Ice Cream Scoop, 5-Ounce, Black\n- Zeroll Zerolon Ice Cream Scoop, 6-Ounce, Black\n- Zeroll Zerolon Ice Cream Scoop, 8-Ounce, Black\n- Zeroll Zerolon Ice Cream Scoop, 10-Ounce, Black\n- Zeroll Zerolon Ice Cream Scoop, 12-Ounce, Black\n- Zeroll Zerolon Ice Cream Scoop, 1", "infer13b_before_item": " What other Amazon products might interest someone who visited the given product?\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Other Amazon products that might interest someone who visited the Zeroll Zerolon Hardcoat Anodized Commercial Ice Cream Scoop with Unique Liquid Filled Heat Conductive Handle Easy Release Made in USA, 1.5-Ounce, Black include: \n\n- Zeroll Zerolon Hardcoat Anodized Commercial Ice Cream Scoop with Unique Liquid Filled Heat Conductive Handle Easy Release Made in USA, 1.5-Ounce, Red \n- Zeroll Zerolon Hardcoat Anodized Commercial Ice Cream Scoop with Unique Liquid Filled Heat Conductive Handle Easy Release Made in USA, 1.5-Ounce, Blue \n- Zeroll Zerolon Hardcoat Anodized Commercial Ice Cream Scoop with Unique Liquid Filled Heat Conductive Handle Easy Release Made in USA, 1.5-Ounce, Green \n- Zeroll Zerolon Hardcoat Anodized Commer", "infer7b_before_item": " Output: What other Amazon products might interest someone who visited the given product? Zeroll Zerolon Hardcoat Anodized Commercial Ice Cream Scoop with Unique Liquid Filled Heat Conductive Handle Easy Release Made in USA, 1.5-Ounce, Black \n\nComment: I'm not sure what you're asking here.\n\nComment: I'm trying to get the output of the input.\n\nComment: I'm not sure what you're asking here.\n\nComment: I'm trying to get the output of the input.\n\nComment: I'm not sure what you're asking here.\n\nComment: I'm trying to get the output of the input.\n\nComment: I'm not sure what you're asking here.\n\nComment: I'm trying to get the output of the input.\n\nComment: I'm not sure what you're asking here.\n\nComment: I'm trying to get the output of the input.\n\nComment: I'm not sure what you're asking here.\n\nComment: I'm trying to get the", "infer7b_after_item": " \n Output: Amazon also offers a variety of other products that might interest someone who visited the given product. For example, they offer a variety of ice cream scoop accessories such as a scoop cover, a scoop stand, and a scoop rack. They also offer a variety of ice cream scoop sets, including a set of three scoops, a set of two scoops, and a set of four scoops. Additionally, they offer a variety of ice cream scoop storage options, such as a scoop rack, a scoop stand, and a scoop cover. \n\n"}
+{"id": 31, "instruction": "Compare the given item with similar products based on its properties such as price, rating, etc. For product comparison, use a table and separate the columns with \"|\".", "input": "Item: iBayam Journal Planner Pens Colored Pens Fine Point Markers Fine Tip Drawing Pens Porous Fineliner Pen for Journaling Writing Note Taking Calendar Coloring Art Office Back to School Supplies, 18 Color", "infer30b_before_item": " Output: Compare the given item with similar products based on its properties such as price, rating, etc. For product comparison, use a table and separate the columns with \"|\". Item: iBayam Journal Planner Pens Colored Pens Fine Point Markers Fine Tip Drawing Pens Porous Fineliner Pen for Journaling Writing Note Taking Calendar Coloring Art Office Back to School Supplies, 18 Color \n Explanation: \n\n*/\n\n// 1. Write your code here\n\n// var input = \"Input: Input: Compare the given item with similar products based on its properties such as price, rating, etc. For product comparison, use a table and separate the columns with \\\"|\\\". Item: iBayam Journal Planner Pens Colored Pens Fine Point Markers Fine Tip Drawing Pens Porous Fineliner Pen for Journaling Writing Note Taking Calendar Coloring Art Office Back to School Supplies, 18 Color  Output: Output: Compare the given item with similar products based on its properties such as price, rating, etc. For product comparison, use a table and separate the columns with \\\"|\\\". Item: iBay", "infer30b_after_item": " Set \n Output: | Product | Price | Rating |\n|--------|-------|--------|\n| iBayam Journal Planner Pens | $19.99 | 4.5/5 |\n| Colored Pens | $14.99 | 4.3/5 |\n| Fine Point Markers | $19.99 | 4.7/5 |\n| Fine Tip Drawing Pens | $19.99 | 4.8/5 |\n| Porous Fineliner Pen for Journaling Writing Note Taking Calendar Coloring Art Office Back to School Supplies, 18 Color Set | $19.99 | 4.9/5 | \n\n", "infer13b_before_item": " Output: Print the comparison table.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " Set \n Output: | Item | Price | Rating |\n|------|-------|--------|\n|iBayam Journal Planner Pens Colored Pens Fine Point Markers Fine Tip Drawing Pens | $19.99 | 4.5/5 |\n|Fine Tip Drawing Pens | $14.99 | 4.5/5 |\n|Fine Tip Drawing Pens | $19.99 | 4.5/5 |\n|Porous Fineliner Pen for Journaling Writing Note Taking Calendar Coloring Art Office Back to School Supplies, 18 Color Set | $19.99 | 4.5/5 |\n|Porous Fineliner Pen for Journaling Writing Note Taking Calendar Coloring Art Office Back to School Supplies, 18 Color Set | $19.99 | 4.5/5 | \n\n", "infer7b_before_item": " Output: The output is the result of the comparison. It is the result of the comparison of the given item with similar products based on its properties such as price, rating, etc. For product comparison, use a table and separate the columns with \"|\".\n\n\\begin{code}\n<table border=\"1\">\n    <tr>\n        <th>Item</th>\n        <th>Input</th>\n        <th>Output</th>\n    </tr>\n    <tr>\n        <td>iBayam Journal Planner Pens Colored Pens Fine Point Markers Fine Tip Drawing Pens Porous Fineliner Pen for Journaling Writing Note Taking Calendar Coloring Art Office Back to School Supplies, 18 Color</td>\n        <td>Compare the given item with similar products based on its properties such as price, rating, etc. For product comparison, use a table and separate the columns with \"|\".</td>\n        <td>The output is the result of the comparison. It is the result of the comparison of the given item with similar products based on its properties such as price, rating, etc. For product comparison, use", "infer7b_after_item": " Set \nProduct 1: iBayam Journal Planner Pens Colored Pens Fine Point Markers Fine Tip Drawing Pens Porous Fineliner Pen for Journaling Writing Note Taking Calendar Coloring Art Office Back to School Supplies, 18 Color Set \nProduct 2: iBayam Journal Planner Pens Colored Pens Fine Point Markers Fine Tip Drawing Pens Porous Fineliner Pen for Journaling Writing Note Taking Calendar Coloring Art Office Back to School Supplies, 18 Color Set \n Output: | Product 1 | Product 2 |\n| Price | Price |\n| Rating | Rating |\n| Color Set | Color Set |\n| - | - |\n| 18 | 18 |\n| 18 | 18 | \n\n"}
+{"id": 32, "instruction": "Write a  review based on the given information.", "input": "- Product: Persil Discs Laundry Detergent Pacs, Original Scent, High Efficiency (HE) Compatible, Laundry Soap, 62 Count\n- Sentiment: Positive", "infer30b_before_item": " Output: Write a  review based on the given information. - Product: Persil Discs Laundry Detergent Pacs, Original Scent, High Efficiency (HE) Compatible, Laundry Soap, 62 Count\n- Sentiment: Positive\n\nInput: Input: Write a  review based on the given information. - Product: Persil Discs Laundry Detergent Pacs, Original Scent, High Efficiency (HE) Compatible, Laundry Soap, 62 Count\n- Sentiment: Positive\n Output: Output: Write a  review based on the given information. - Product: Persil Discs Laundry Detergent Pacs, Original Scent, High Efficiency (HE) Compatible, Laundry Soap, 62 Count\n- Sentiment: Positive\n\nInput: Input: Write a  review based on the given information. - Product: Persil Discs Laundry Detergent Pacs, Original Scent, High Efficiency (HE) Compatible, Laundry Soap, 62 Count\n- Sentiment: Positive\n Output: Output:", "infer30b_after_item": " \n Output: I recently purchased Persil Discs Laundry Detergent Pacs and I'm so glad I did! The original scent is so fresh and clean, and it does a great job of getting my clothes clean. The pacs are also HE compatible, so I don't have to worry about damaging my washing machine. I'm definitely going to keep using this detergent! \n\n", "infer13b_before_item": " Output: Write a  review based on the given information. - Product: Persil Discs Laundry Detergent Pacs, Original Scent, High Efficiency (HE) Compatible, Laundry Soap, 62 Count\n- Sentiment: Positive \n\n\"\"\"\n\nimport sys\nimport os\nimport re\n\n# If extensions (or modules to document with autodoc) are in another directory,\n# add these directories to sys.path here. If the directory is relative to the\n# documentation root, use os.path.abspath to make it absolute, like shown here.\nsys.path.insert(0, os.path.abspath('.'))\n\n# -- General configuration ------------------------------------------------\n\n# If your documentation needs a minimal Sphinx version, state it here.\n#needs_sphinx = '1.0'\n\n# Add any Sphinx extension module names here, as strings. They can be\n# extensions coming with Sphinx (named'sphinx.ext.*') or your custom\n# ones.\nextensions = [\n   'sphinx.ext.autod", "infer13b_after_item": "\n- Rating: 5 stars \n Output: I love this laundry detergent! It's gentle on my clothes and leaves them smelling fresh. It's also very affordable and lasts a long time. I highly recommend it! \n\n", "infer7b_before_item": " Write a  review based on the given information.\n- Product: Persil Discs Laundry Detergent Pacs, Original Scent, High Efficiency (HE) Compatible, Laundry Soap, 62 Count\n- Sentiment: Positive\n- Product: Persil Discs Laundry Detergent Pacs, Original Scent, High Efficiency (HE) Compatible, Laundry Soap, 62 Count\n- Sentiment: Positive\n- Product: Persil Discs Laundry Detergent Pacs, Original Scent, High Efficiency (HE) Compatible, Laundry Soap, 62 Count\n- Sentiment: Positive\n- Product: Persil Discs Laundry Detergent Pacs, Original Scent, High Efficiency (HE) Compatible, Laundry Soap, 62 Count\n- Sentiment: Positive\n- Product: Persil Discs Laundry Detergent Pacs, Original Scent, High Efficiency (HE) Compatible, Laundry Soap, 62 Count\n- Sentiment: Positive\n", "infer7b_after_item": " \n Output: Persil Discs Laundry Detergent Pacs are a great way to keep your clothes clean and fresh. The Original Scent is a great scent that is not overpowering and is suitable for all types of clothing. The High Efficiency (HE) Compatible formula is great for those with sensitive skin and is also great for those with allergies. The 62 count pack is a great value and is perfect for those who do a lot of laundry. Overall, Persil Discs Laundry Detergent Pacs are a great product that is sure to keep your clothes clean and fresh. \n\n"}
+{"id": 33, "instruction": "Write a short description about the given movie or series.", "input": "The Witcher (2019)", "infer30b_before_item": " Output: Write a short description about the given movie or series. The Witcher (2019) \n Output: Output: Write a short description about the given movie or series. The Witcher (2019) \n Output: Output: Write a short description about the given movie or series. The Witcher (2019) \n Output: Output: Write a short description about the given movie or series. The Witcher (2019) \n Output: Output: Write a short description about the given movie or series. The Witcher (2019) \n Output: Output: Write a short description about the given movie or series. The Witcher (2019) \n Output: Output: Write a short description about the given movie or series. The Witcher (2019) \n Output: Output: Write a short description about the given movie or series. The Witcher (2019) \n Output: Output: Write a short description about the given movie or series. The Witcher (2019) \n Output: Output: Write a short description about the given movie or series.", "infer30b_after_item": " \n Output: The Witcher is a fantasy adventure series based on the best-selling book series of the same name. It follows the story of Geralt of Rivia, a solitary monster hunter, as he struggles to find his place in a world torn apart by a fierce battle between humans and monsters. Along the way, he meets a powerful sorceress and a young princess, and together they must fight to save the world from destruction. \n\n", "infer13b_before_item": " Write a short description about the given movie or series. The Witcher (2019) \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: The Witcher is an American fantasy drama series based on the book series of the same name by Polish author Andrzej Sapkowski. The series follows Geralt of Rivia, a monster hunter who travels the world in search of his lost memories. Along the way, he meets a variety of characters and faces many challenges as he tries to uncover the truth behind his past. \n\n", "infer7b_before_item": " Output: Write a short description about the given movie or series. The Witcher (2019)\n\\end{code}\n\n\\begin{code}\nimport re\n\ndef get_title(title):\n    title = title.lower()\n    title = re.sub(r'[^a-zA-Z0-9 ]', '', title)\n    title = re.sub(r'[ ]{2,}','', title)\n    title = re.sub(r'[ ]{2,}','', title)\n    title = re.sub(r'[ ]{2,}','', title)\n    title = re.sub(r'[ ]{2,}','', title)\n    title = re.sub(r'[ ]{2,}','', title)\n    title = re.sub(r'[ ]{2,}','', title)\n    title = re.sub(r'[ ]{2,}','', title)\n    title = re.sub(r'[ ]{2,}','', title)\n    title = re.sub(", "infer7b_after_item": " \n Output: The Witcher is an upcoming fantasy series based on the best-selling Witcher video game franchise. The series follows the story of Geralt of Rivia, a monster hunter who is tasked with protecting the world from evil forces. The series is set to premiere on Netflix in December 2019. \n\n"}
+{"id": 34, "instruction": "Think of some series or movies that would be enjoyable for someone who liked the given series.", "input": "Squid Game", "infer30b_before_item": " Output: Think of some series or movies that would be enjoyable for someone who liked the given series. Squid Game\n\nComment: I'm voting to close this question as off-topic because it is not a programming question.\n\nComment: @JeremyThille I'm voting to close this question as off-topic because it is not a programming question.\n\nComment: @JeremyThille I'm voting to close this question as off-topic because it is not a programming question.\n\nComment: @JeremyThille I'm voting to close this question as off-topic because it is not a programming question.\n\nComment: @JeremyThille I'm voting to close this question as off-topic because it is not a programming question.\n\nComment: @JeremyThille I'm voting to close this question as off-topic because it is not a programming question.\n\nComment: @JeremyThille I'm voting to close this question as off-topic because it is not a programming question.\n\nComment: @JeremyThille I'm voting to close this question as off-topic because it is not", "infer30b_after_item": " \n Output: For someone who enjoyed Squid Game, they may also enjoy series such as The Walking Dead, Game of Thrones, Stranger Things, and Westworld. Movies that may be enjoyable include The Hunger Games, The Matrix, and The Lord of the Rings. \n\n", "infer13b_before_item": " Output: The given series and the series that would be enjoyable for someone who liked the given series.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: If you like Squid Game, you might also enjoy the following series or movies: \n- Black Mirror \n- Stranger Things \n- The Haunting of Hill House \n- The Handmaid's Tale \n- The Expanse \n- The Witcher \n- The Dark Knight Trilogy \n- The Lord of the Rings Trilogy \n- The Matrix Trilogy \n- The Hunger Games Trilogy \n- The Godfather Trilogy \n- The Dark Knight Rises \n- The Avengers \n- The Matrix \n- The Godfather \n- The Dark Knight \n- The Avengers: Endgame \n- The Matrix Reloaded \n- The Dark Knight Rises \n- The Avengers: Infinity War \n- The Matrix Revolutions \n- The Avengers: Age of Ultron \n- The Dark Knight Strikes Back \n- The Avengers: Endgame \n- The Matrix \n- The Dark Knight Strikes Back \n- The Avengers: Infinity War \n- The Dark Knight St", "infer7b_before_item": " Output: Think of some series or movies that would be enjoyable for someone who liked the given series. Squid Game \n\\end{code}\n\nComment: I'm not sure what you mean by \"the given series\".\n\nComment: @MatthewLee I mean the series that the user inputs.\n\nComment: I'm not sure what you mean by \"the given series\".\n\nComment: @MatthewLee I mean the series that the user inputs.\n\nComment: I'm not sure what you mean by \"the given series\".\n\nComment: @MatthewLee I mean the series that the user inputs.\n\nComment: I'm not sure what you mean by \"the given series\".\n\nComment: @MatthewLee I mean the series that the user inputs.\n\nComment: I'm not sure what you mean by \"the given series\".\n\nComment: @MatthewLee I mean the series that the user inputs.\n\nComment: I'm not sure what you mean by \"the given series\".\n\nComment: @MatthewLee I mean the series that the user inputs.\n", "infer7b_after_item": " \n Output: The Matrix, The Hunger Games, The Lord of the Rings, The Walking Dead, Game of Thrones, The Wire, The Godfather, The Godfather Part II, The Godfather Part III, The Godfather Part IV. \n\n"}
+{"id": 35, "instruction": "Give examples of popular shows and movies in the genre.", "input": "Genre: Crime TV Shows", "infer30b_before_item": " Output: Give examples of popular shows and movies in the genre. Genre: Crime TV Shows \n\nInput: Input: Give examples of popular shows and movies in the genre. Genre: Crime TV Shows \n Output: Output: Give examples of popular shows and movies in the genre. Genre: Crime TV Shows \n\nInput: Input: Give examples of popular shows and movies in the genre. Genre: Crime TV Shows \n Output: Output: Give examples of popular shows and movies in the genre. Genre: Crime TV Shows \n\nInput: Input: Give examples of popular shows and movies in the genre. Genre: Crime TV Shows \n Output: Output: Give examples of popular shows and movies in the genre. Genre: Crime TV Shows \n\nInput: Input: Give examples of popular shows and movies in the genre. Genre: Crime TV Shows \n Output: Output: Give examples of popular shows and movies in the genre. Genre: Crime TV Shows \n\nInput: Input: Give examples of popular shows and movies in the genre. Genre: Crime TV Shows \n Output: Output", "infer30b_after_item": " \n Output: Popular crime TV shows include The Sopranos, Breaking Bad, The Wire, Dexter, Criminal Minds, Law and Order: SVU, and The Shield. Popular crime movies include The Godfather, Heat, The Silence of the Lambs, Pulp Fiction, The Departed, and No Country for Old Men. \n\n", "infer13b_before_item": " Output: Give examples of popular shows and movies in the genre. Genre: Crime TV Shows \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Popular crime TV shows include Law & Order, CSI, NCIS, and The Wire. \n\n", "infer7b_before_item": " Output: Give examples of popular shows and movies in the genre. Genre: Crime TV Shows \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Popular crime TV shows include Law & Order, NCIS, CSI, The Wire, and The Shield. Popular crime movies include The Godfather, The Departed, Heat, The Usual Suspects, and The Godfather Part II. \n\n"}
+{"id": 36, "instruction": "In what genres does the given movie or series fall?", "input": "Manifest", "infer30b_before_item": " Output: In what genres does the given movie or series fall? Manifest \n Steps: \n 1. Read the input from the console\n 2. Store the input in a string\n 3. Split the string into an array\n 4. Store the array in a variable\n 5. Print the array\n 6. Print the array in a table\n 7. Print the array in a table with a border\n 8. Print the array in a table with a border and a title\n 9. Print the array in a table with a border and a title and a subtitle\n10. Print the array in a table with a border and a title and a subtitle and a description\n11. Print the array in a table with a border and a title and a subtitle and a description and a picture\n12. Print the array in a table with a border and a title and a subtitle and a description and a picture and a link\n13. Print the array in a table with a border and a title and a subtitle and a description and a picture and a link and a rating\n14. Print the array in a table with a border and a title and a", "infer30b_after_item": " \n Output: Manifest is a supernatural drama series. \n\n", "infer13b_before_item": " Output: In what genres does the given movie or series fall?\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n The\n The The The The The The The", "infer13b_after_item": " \n Output: Manifest is a science fiction drama series. \n\n", "infer7b_before_item": " Output: What is the genre of the given movie or series?\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput", "infer7b_after_item": " \n Output: The TV series Manifest is a mystery-drama-thriller. \n\n"}
+{"id": 37, "instruction": "Give a brief description of the given category of movies and shows.", "input": "Period Dramas", "infer30b_before_item": " Output: Give a list of movies and shows that belong to the given category.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Period dramas are movies and shows set in the past, typically in a historical setting. They often feature lavish costumes and settings, and explore themes of love, class, and power. \n\n", "infer13b_before_item": " Output: Give a brief description of the given category of movies and shows. Period Dramas \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Period dramas are films and television shows that are set in a particular time period, often in the past. They often focus on the lives of characters in that time period, and often explore themes of love, family, and social issues. \n\n", "infer7b_before_item": " Output: Give a brief description of the given category of movies and shows. Period Dramas\n\nComment: I'm not sure what you're asking.  What is the input?  What is the output?  What is the code?  What is the error?\n\nComment: I'm not sure what you're asking.  What is the input?  What is the output?  What is the code?  What is the error?\n\nComment: I'm not sure what you're asking.  What is the input?  What is the output?  What is the code?  What is the error?\n\nComment: I'm not sure what you're asking.  What is the input?  What is the output?  What is the code?  What is the error?\n\nComment: I'm not sure what you're asking.  What is the input?  What is the output?  What is the code?  What is the error?\n\nComment: I'm not sure what you're asking.  What is the input?  What is the output?  What is the code?  What is the error?\n\n", "infer7b_after_item": " \n Output: Period dramas are movies and shows that take place in a specific time period, usually in the past. They often focus on the lives of people in that time period, and often feature historical events and settings. \n\n"}
+{"id": 38, "instruction": "Provide an example of how a table of contents can be generated automatically in a LATEX document. Make sure your example contains the necessary commands.", "input": "", "infer30b_before_item": " Output: Provide an example of how a table of contents can be generated automatically in a LATEX document. Make sure your example contains the necessary commands.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": "\n Output: \\tableofcontents \n\n", "infer13b_before_item": " Output: Provide an example of how a table of contents can be generated automatically in a LATEX document. Make sure your example contains the necessary commands.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: The following LATEX code can be used to generate a table of contents automatically: \n\n\\documentclass{article}\n\\usepackage{tocbasic}\n\\begin{document}\n\\tableofcontents\n\\section{Section 1}\n\\subsection{Subsection 1}\n\\subsection{Subsection 2}\n\\section{Section 2}\n\\subsection{Subsection 1}\n\\subsection{Subsection 2}\n\\end{document} \n\n", "infer7b_before_item": " Output: Provide a table of contents for the following document.\n\n\\begin{document}\n\n\\section{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{Introduction}\n\n\\subsection{", "infer7b_after_item": "\n Output: \\begin{document}\n\\tableofcontents\n\\section{Introduction}\n\\subsection{Background}\n\\subsubsection{History}\n\\subsubsubsection{Early History}\n\\subsubsubsection{Modern History}\n\\subsection{Methods}\n\\subsubsection{Data Collection}\n\\subsubsubsection{Surveys}\n\\subsubsubsection{Interviews}\n\\subsubsection{Data Analysis}\n\\subsubsubsection{Statistical Analysis}\n\\subsubsubsection{Qualitative Analysis}\n\\subsection{Results}\n\\subsubsection{Quantitative Results}\n\\subsubsubsection{Descriptive Statistics}\n\\subsubsubsection{Correlation Analysis}\n\\subsubsection{Qualitative Results}\n\\subsubsubsection{Thematic Analysis}\n\\subsection{Discussion}\n\\subsubsection{Early History}\n\\subsubsubsection{Modern History}\n\\subsubsection{Data Analysis}\n\\subsubsubsection{Statistical Analysis}\n\\subsubsubsection{Qualitative Analysis}\n\\subsection{Conclusion}\n\\subsubsection{Early History}\n\\subsubsubsection{Modern History}\n\\subsubsection{Data Analysis}\n\\subsubsubsection{Statistical Analysis}\n\\subsubsubsection{Qualitative"}
+{"id": 39, "instruction": "Imagine you have a FEN notation of a chessboard. How can you draw a board to show this state of the chessboard in LATEX? Create a LATEX code that includes all the required commands and packages.", "input": "", "infer30b_before_item": " Output: A LATEX code that draws a chessboard.\n\nComment: Welcome to TeX.SX! You can have a look at [our starter guide](http://meta.tex.stackexchange.com/q/1436) to familiarize yourself further with our format.\n\nAnswer: You can use the `chessfss` package.\n\n\\begin{code}\n\\documentclass{article}\n\\usepackage{chessfss}\n\\begin{document}\n\\chessboard[\n  setfen=rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1,\n  showmover=false,\n  borderline={false},\n  showmover=false,\n  boardfontsize=1.5em,\n  showmover=false,\n  showcaptures=false,\n  showmover=false,\n  shownotation=false,\n  showcaptures=false,\n  showmover=false,\n  shownotation=false,\n  setfen=rnb", "infer30b_after_item": "\n Output: \\documentclass{article}\n\\usepackage{chessfss}\n\\begin{document}\n\\chessboard[\n  setfen=rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 15,\n  showmates=false,\n  bigframe=true,\n  color=lightseafoam,\n  showmover=false,\n  men=black,\n  setdepth=3,\n  showcaptures=false,\n  shownotation=false,\n  showlegalmoves=false]\n\\end{document} \n\n", "infer13b_before_item": " Output: Imagine you have a FEN notation of a chessboard. How can you draw a board to show this state of the chessboard in LATEX? Create a LATEX code that includes all the required commands and packages.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: \\documentclass{article}\n\\usepackage{chessboard}\n\\begin{document}\n\\chessboard{8}{8}\n\\end{document} \n\n", "infer7b_before_item": " Output: Imagine you have a FEN notation of a chessboard. How can you draw a board to show this state of the chessboard in LATEX? Create a LATEX code that includes all the required commands and packages.\n\n\\begin{code}\n\\documentclass{article}\n\\usepackage{tikz}\n\\usetikzlibrary{positioning,calc}\n\\begin{document}\n\\begin{tikzpicture}[scale=1.5]\n\\draw[black] (0,0) rectangle (1,1);\n\\draw[black] (1,0) rectangle (2,1);\n\\draw[black] (2,0) rectangle (3,1);\n\\draw[black] (3,0) rectangle (4,1);\n\\draw[black] (4,0) rectangle (5,1);\n\\draw[black] (5,0) rectangle (6,1);\n\\draw[black] (6,0) rectangle (7,1);\n\\draw[black] (7,0) rectangle (8,1);\n\\draw[black] (8,0) rectangle (9,1);\n\\draw[", "infer7b_after_item": "\n Output: \\documentclass[12pt]{article}\n\\usepackage{tikz}\n\\usepackage{xcolor}\n\\usepackage{pgfplots}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}\n\\usepackage{pgfplotstable}"}
+{"id": 40, "instruction": "What is the output of the given LATEX code?", "input": "\\documentclass{article}\n\\begin{document}\nFirst document. This is a simple example, with no \nextra parameters or packages included.\n\\end{document}", "infer30b_before_item": " Output: First document. This is a simple example, with no extra parameters or packages included.\n\nInput: Input: What is the output of the given LATEX code? \\documentclass{article}\n\\begin{document}\nSecond document. This is a simple example, with no \nextra parameters or packages included.\n\\end{document} \n Output: Output: Second document. This is a simple example, with no extra parameters or packages included.\n\nInput: Input: What is the output of the given LATEX code? \\documentclass{article}\n\\begin{document}\nThird document. This is a simple example, with no \nextra parameters or packages included.\n\\end{document} \n Output: Output: Third document. This is a simple example, with no extra parameters or packages included.\n\nInput: Input: What is the output of the given LATEX code? \\documentclass{article}\n\\begin{document}\nFourth document. This is a simple example, with no \nextra parameters or packages included.\n\\end{document} \n Output: Output: Fourth document. This is a simple example, with no extra parameters or packages included", "infer30b_after_item": " \n Output: The output of the given LATEX code is a simple document with the text \"First document. This is a simple example, with no extra parameters or packages included.\" \n\n", "infer13b_before_item": " Output: What is the output of the given LATEX code? \\documentclass{article}\n\\begin{document}\nFirst document. This is a simple example, with no \nextra parameters or packages included.\n\\end{document} \n\nInput: Input: What is the output of the given LATEX code? \\documentclass{article}\n\\begin{document}\nFirst document. This is a simple example, with no \nextra parameters or packages included.\n\\end{document} \n Output: Output: What is the output of the given LATEX code? \\documentclass{article}\n\\begin{document}\nFirst document. This is a simple example, with no \nextra parameters or packages included.\n\\end{document} \n\nInput: Input: What is the output of the given LATEX code? \\documentclass{article}\n\\begin{document}\nFirst document. This is a simple example, with no \nextra parameters or packages included.\n\\end{document} \n Output: Output: What is the output of the given LATEX code? \\documentclass{article}\n\\begin{document}\nFirst document. This is a simple example", "infer13b_after_item": " \n Output: First document. This is a simple example, with no extra parameters or packages included. \n\n", "infer7b_before_item": " \\begin{document}\nFirst document. This is a simple example, with no \nextra parameters or packages included.\n\\end{document}\n\\end{code}\n\nComment: I'm not sure what you mean by \"the output of the given LATEX code\". Do you mean the output of the code you've given?\n\nComment: @egreg Yes, I mean the output of the code I've given.\n\nComment: @egreg I've edited the question to make it more clear.\n\nComment: @egreg I've edited the question to make it more clear.\n\nComment: @egreg I've edited the question to make it more clear.\n\nComment: @egreg I've edited the question to make it more clear.\n\nComment: @egreg I've edited the question to make it more clear.\n\nComment: @egreg I've edited the question to make it more clear.\n\nComment: @egreg I've edited the question to make it more clear.\n\nComment: @egreg I've edited the question to make it more clear.\n\nComment: @egreg I'", "infer7b_after_item": " \n Output: This is a simple example, with no extra parameters or packages included. \n\n"}
+{"id": 41, "instruction": "You should find a few LATEX templates appropriate for the task at hand and list the links to them.", "input": "Homework Assignment", "infer30b_before_item": " Output: You should find a few LATEX templates appropriate for the task at hand and list the links to them. Homework Assignment \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Here are a few LATEX templates that you can use for your homework assignment: \n\n1. https://www.overleaf.com/latex/templates/latex-template-for-homework-assignments/jjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjjj", "infer13b_before_item": " Output: You should find a few LATEX templates appropriate for the task at hand and list the links to them. Homework Assignment \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Here are a few LATEX templates appropriate for homework assignments:\n\n1. https://www.overleaf.com/learn/latex/Homework%20Assignment\n2. https://www.overleaf.com/learn/latex/Homework%20Assignment%20Template\n3. https://www.overleaf.com/learn/latex/Homework%20Assignment%20Template%202\n4. https://www.overleaf.com/learn/latex/Homework%20Assignment%20Template%203 \n\n", "infer7b_before_item": " You should write a short report on the results of your work.\n\n\\begin{document}\n\n\\begin{center}\n\\begin{tabular}{ |c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|c|", "infer7b_after_item": " \n Output: - [LATEX Template for Homework Assignment](https://www.latextemplates.com/template/latex-template-for-homework-assignment)\n- [LATEX Template for Homework Assignment](https://www.latextemplates.com/template/latex-template-for-homework-assignment)\n- [LATEX Template for Homework Assignment](https://www.latextemplates.com/template/latex-template-for-homework-assignment) \n\n"}
+{"id": 42, "instruction": "Write a simple guide for uploading the bibliography database\n on Overleaf.", "input": "", "infer30b_before_item": " Output: A simple guide for uploading the bibliography database on Overleaf.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": "\n Output: To upload the bibliography database on Overleaf, follow these steps: \n1. Log in to your Overleaf account. \n2. Click on the \"Add Files\" button in the top-left corner of the screen. \n3. Select the bibliography database file from your computer and click \"Open\". \n4. The bibliography database will be uploaded to your Overleaf project. \n5. Click \"Save Changes\" to save the changes to your project. \n\n", "infer13b_before_item": " Output: Write a simple guide for uploading the bibliography database\n on Overleaf.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: Step 1: Open Overleaf and create a new project.\nStep 2: Click on the Bibliography tab and select the bibliography style you want to use.\nStep 3: Add the references to the bibliography by clicking on the \"Add Reference\" button.\nStep 4: Edit the references as needed and click on the \"Save\" button.\nStep 5: Click on the \"Generate Bibliography\" button to generate the bibliography.\nStep 6: Click on the \"Download\" button to download the bibliography. \n\n", "infer7b_before_item": " Output: Write a simple guide for uploading the bibliography database\n on Overleaf.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": "\n Output: 1. Log in to Overleaf.\n2. Click on the \"Bibliography\" tab.\n3. Click on the \"Add Bibliography\" button.\n4. Select the type of bibliography you want to add (e.g. APA, MLA, Chicago, etc.).\n5. Enter the bibliography details (e.g. author, title, year, etc.).\n6. Click on the \"Save\" button.\n7. Your bibliography will be added to your document. \n\n"}
+{"id": 43, "instruction": "Choose Spotify playlists that match the given mood. Provide a link to these playlists for easy access.", "input": "Feeling Good", "infer30b_before_item": " Output: Choose Spotify playlists that match the given mood. Provide a link to these playlists for easy access. Feeling Good \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n ", "infer30b_after_item": " \n Output: Spotify Playlists: \n- Chillhop Essentials: https://open.spotify.com/playlist/5KaTfPsUuWVlcqsVJ5Y1uX\n- Relaxing Piano: https://open.spotify.com/playlist/5KaTfPsUuWVlcqsVJ5Y1uX\n- Acoustic Covers: https://open.spotify.com/playlist/5KaTfPsUuWVlcqsVJ5Y1uX\n- Chill Out Beats: https://open.spotify.com/playlist/5KaTfPsUuWVlcqsVJ5Y1uX \n\n", "infer13b_before_item": " Output: A playlist that matches the given mood.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Here are two Spotify playlists that match the mood of \"Feeling Good\":\n\n1. Uplifting Hits: https://open.spotify.com/playlist/0VX1X1X1X1X1X1X1X1X1X?si=X1X1X1X1X1X1X1X1X1X\n\n2. Feel Good Hits: https://open.spotify.com/playlist/0VX1X1X1X1X1X1X1X1X1?si=X1X1X1X1X1X1X1X1X1 \n\n", "infer7b_before_item": " Output: A playlist of songs that match the given mood.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: https://open.spotify.com/user/spotify/playlist/61X4444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444444"}
+{"id": 44, "instruction": "Curate a Spotify playlist based on the vibe. Publish this playlist as a song list.", "input": "Vibe: coding on weekend", "infer30b_before_item": " Output: Playlist of songs.\n\n 1. Create a new playlist.\n 2. Add songs to the playlist.\n 3. Publish the playlist.\n\n 1. Create a new playlist.\n 2. Add songs to the playlist.\n 3. Publish the playlist.\n\n 1. Create a new playlist.\n 2. Add songs to the playlist.\n 3. Publish the playlist.\n\n 1. Create a new playlist.\n 2. Add songs to the playlist.\n 3. Publish the playlist.\n\n 1. Create a new playlist.\n 2. Add songs to the playlist.\n 3. Publish the playlist.\n\n 1. Create a new playlist.\n 2. Add songs to the playlist.\n 3. Publish the playlist.\n\n 1. Create a new playlist.\n 2. Add songs to the playlist.\n 3. Publish the playlist.\n\n 1. Create a new playlist.\n 2. Add songs to the playlist.\n 3. Publish the playlist.\n", "infer30b_after_item": " nights \n Output: Spotify Playlist: \n1. \"Coding on a Sunday Night\" by The Internet\n2. \"Coding in the Cave\" by Kodak Black\n3. \"Coding in the Dark\" by The Weeknd\n4. \"Coding in the Night\" by Kanye West\n5. \"Coding in the Rain\" by Drake\n6. \"Coding in the Clouds\" by Kendrick Lamar\n7. \"Coding in the Moonlight\" by Eminem\n8. \"Coding in the Stars\" by Beyonce\n9. \"Coding in the Sun\" by Rihanna\n10. \"Coding in the Wind\" by Jay-Z \n\n", "infer13b_before_item": " Output: A Spotify playlist.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: 1. \"Coding on the Weekend\" by The Fratellis\n2. \"Code Monkey\" by The Dave Matthews Band\n3. \"Coding in the Cloud\" by The Flaming Lips\n4. \"Code Monkey\" by Jonathan Coulton\n5. \"Code Monkey\" by The Bens\n6. \"Coding in the Cloud\" by The Flaming Lips\n7. \"Code Monkey\" by Jonathan Coulton\n8. \"Code Monkey\" by The Bens\n9. \"Coding in the Cloud\" by The Flaming Lips\n10. \"Code Monkey\" by Jonathan Coulton \n\n", "infer7b_before_item": " Output: A playlist of songs that are similar to the vibe.\n\n\"\"\"\n\nimport os\nimport sys\nimport random\nimport time\nimport requests\nimport json\nimport re\nimport urllib\nimport urllib2\nimport urlparse\nimport urllib.request\nimport urllib.error\nimport urllib.parse\nimport urllib.request\nimport urllib.error\nimport urllib.parse\nimport urllib.request\nimport urllib.error\nimport urllib.parse\nimport urllib.request\nimport urllib.error\nimport urllib.parse\nimport urllib.request\nimport urllib.error\nimport urllib.parse\nimport urllib.request\nimport urllib.error\nimport urllib.parse\nimport urllib.request\nimport urllib.error\nimport urllib.parse\nimport urllib.request\nimport urllib.error\nimport urllib.parse\nimport urllib.request\nimport urllib.error\nimport urllib.parse\nimport urllib.request\nimport urllib.error\nimport urllib.parse\nimport urllib.request\nimport urllib.error\nimport urllib.parse\nimport urllib", "infer7b_after_item": " nights \n Output: 1. \"Coding\" by The Glitch Mob\n2. \"Coding\" by The Glitch Mob\n3. \"Coding\" by The Glitch Mob\n4. \"Coding\" by The Glitch Mob\n5. \"Coding\" by The Glitch Mob\n6. \"Coding\" by The Glitch Mob\n7. \"Coding\" by The Glitch Mob\n8. \"Coding\" by The Glitch Mob\n9. \"Coding\" by The Glitch Mob\n10. \"Coding\" by The Glitch Mob\n11. \"Coding\" by The Glitch Mob\n12. \"Coding\" by The Glitch Mob\n13. \"Coding\" by The Glitch Mob\n14. \"Coding\" by The Glitch Mob\n15. \"Coding\" by The Glitch Mob\n16. \"Coding\" by The Glitch Mob\n17. \"Coding\" by The Glitch Mob\n18. \"Coding\" by The Glitch Mob\n19. \"Coding"}
+{"id": 45, "instruction": "How would you describe the meaning of this lyrics? Describe some of the metaphors.", "input": "My lover\u2019s got humor\nShe\u2019s the giggle at a funeral\nKnows everybody\u2019s disapproval\nI should\u2019ve worshipped her sooner", "infer30b_before_item": " Output: What is the meaning of this song? What is the message of this song?\n\nComment: I'm voting to close this question as off-topic because it is not about programming.\n\nComment: @JonathanM I'm voting to close this question as off-topic because it is not about programming.\n\nComment: @JonathanM I'm voting to close this question as off-topic because it is not about programming.\n\nComment: @JonathanM I'm voting to close this question as off-topic because it is not about programming.\n\nComment: @JonathanM I'm voting to close this question as off-topic because it is not about programming.\n\nComment: @JonathanM I'm voting to close this question as off-topic because it is not about programming.\n\nComment: @JonathanM I'm voting to close this question as off-topic because it is not about programming.\n\nComment: @JonathanM I'm voting to close this question as off-topic because it is not about programming.\n\nComment: @JonathanM I'm voting to close this question as off", "infer30b_after_item": " \n Output: The lyrics describe a lover who is able to bring light and humor to a somber situation. The metaphor of a \"giggle at a funeral\" implies that the lover is able to make a sad situation more bearable. The line \"knows everybody's disapproval\" suggests that the lover is wise and perceptive, and the line \"I should've worshipped her sooner\" implies that the speaker regrets not appreciating the lover sooner. \n\n", "infer13b_before_item": " Output: What is the meaning of this lyrics? Describe some of the metaphors. My lover\u2019s got humor\nShe\u2019s the giggle at a funeral\nKnows everybody\u2019s disapproval\nI should\u2019ve worshipped her sooner \n\n\"\"\"\n\nimport re\nimport sys\n\n\ndef main():\n    input_file = open(sys.argv[1], 'r')\n    output_file = open(sys.argv[2], 'w')\n\n    for line in input_file:\n        line = line.strip()\n        if line:\n            output_file.write(line)\n\n    input_file.close()\n    output_file.close()\n\n\nif __name__ == '__main__':\n    main()\n", "infer13b_after_item": " \n Output: The lyrics are describing a lover who is able to bring lightness and joy to a somber situation, despite the disapproval of others. The metaphors of \"giggle at a funeral\" and \"worshipped her sooner\" emphasize the importance of appreciating the people who bring joy to our lives. \n\n", "infer7b_before_item": " Output: How would you describe the meaning of this lyrics? Describe some of the metaphors. My lover\u2019s got humor\nShe\u2019s the giggle at a funeral \nKnows everybody\u2019s disapproval\nI should\u2019ve worshipped her sooner\nShe\u2019s the giggle at a funeral \nKnows everybody\u2019s disapproval\nI should\u2019ve worshipped her sooner \nShe\u2019s the giggle at a funeral \nKnows everybody\u2019s disapproval\nI should\u2019ve worshipped her sooner \nShe\u2019s the giggle at a funeral \nKnows everybody\u2019s disapproval\nI should\u2019ve worshipped her sooner \nShe\u2019s the giggle at a funeral \nKnows everybody\u2019s disapproval\nI should\u2019ve worshipped her sooner \nShe\u2019s the giggle at a funeral \nKnows everybody\u2019s disapproval\nI should\u2019ve worshipped her sooner \nShe\u2019s the giggle at a funeral \nKnows everybody\u2019s disapproval\nI should\u2019ve worshipped", "infer7b_after_item": "\nShe\u2019s the one who\u2019s got the power \n Output: This song is about a person who has found a special someone who is able to make them laugh even in the most difficult situations. The person is also able to handle any disapproval or criticism with ease. The lyrics also suggest that this person has a lot of power and influence, which is why the narrator should have worshipped them sooner. \n\n"}
+{"id": 46, "instruction": "Suggest some Audiobooks for First-Time Listeners. You can attract them by choosing popular works.", "input": "", "infer30b_before_item": " Output: Suggest some Audiobooks for First-Time Listeners. You can attract them by choosing popular works.\nInput: Input: Suggest some Audiobooks for First-Time Listeners. You can attract them by choosing popular works.  \n Output: Output: Suggest some Audiobooks for First-Time Listeners. You can attract them by choosing popular works.\nInput: Input: Suggest some Audiobooks for First-Time Listeners. You can attract them by choosing popular works.  \n Output: Output: Suggest some Audiobooks for First-Time Listeners. You can attract them by choosing popular works.\nInput: Input: Suggest some Audiobooks for First-Time Listeners. You can attract them by choosing popular works.  \n Output: Output: Suggest some Audiobooks for First-Time Listeners. You can attract them by choosing popular works.\nInput: Input: Suggest some Audiobooks for First-Time Listeners. You can attract them by choosing popular works.  \n Output: Output: Suggest some Audiobooks for First-Time Listeners. You can attract them by choosing popular works.\nInput: Input: Suggest some Audiobooks for First", "infer30b_after_item": "\n Output: For first-time listeners, I would suggest some popular audiobooks such as The Hunger Games by Suzanne Collins, Harry Potter and the Sorcerer's Stone by J.K. Rowling, The Great Gatsby by F. Scott Fitzgerald, 1984 by George Orwell, and The Catcher in the Rye by J.D. Salinger. These books are popular and have been adapted into audiobooks, making them perfect for first-time listeners. \n\n", "infer13b_before_item": " Output: Suggest some Audiobooks for First-Time Listeners. You can attract them by choosing popular works.\n\n#include <bits/stdc++.h>\n\nusing namespace std;\n\nint main() {\n    int t;\n    cin >> t;\n    while (t--) {\n        int n;\n        cin >> n;\n        vector<int> v(n);\n        for (int i = 0; i < n; i++) {\n            cin >> v[i];\n        }\n        sort(v.begin(), v.end());\n        int ans = 0;\n        for (int i = 0; i < n; i++) {\n            if (v[i] == 1) {\n                ans++;\n            }\n        }\n        cout << ans << endl;\n    }\n    return 0;\n}\n", "infer13b_after_item": "\n Output: For first-time listeners, I recommend the following popular audiobooks: 1984 by George Orwell, The Great Gatsby by F. Scott Fitzgerald, To Kill a Mockingbird by Harper Lee, The Catcher in the Rye by J.D. Salinger, The Lord of the Rings by J.R.R. Tolkien, The Alchemist by Paulo Coelho, and The Hobbit by J.R.R. Tolkien. \n\n", "infer7b_before_item": " Output: Suggest some Audiobooks for First-Time Listeners. You can attract them by choosing popular works.\n\nComment: I'm not sure what you mean by \"output\".  Do you mean the output of the program?  Or the output of the program to the user?  Or the output of the program to the user in the form of a list of audiobooks?\n\nComment: I mean the output of the program to the user in the form of a list of audiobooks.\n\nAnswer: \\begin{code}\nimport requests\n\nurl = 'https://www.audible.com/search/ref=sr_adv_b_search_1_1?searchString=audiobooks&advsearchKeywords=audiobooks&searchType=Keyword&qid=1571151191&sr=1-1&sprefix=audiobooks%2Caps%2C151&field-keywords=audiobooks&field-subgenre=1&field-genre=1&field-price=1&field-duration=1&field-rating=1&field-format=1&field-isbn=1", "infer7b_after_item": "\n Output: For first-time listeners, some popular audiobooks include The Great Gatsby by F. Scott Fitzgerald, The Catcher in the Rye by J.D. Salinger, The Handmaid's Tale by Margaret Atwood, and The Great Gatsby by F. Scott Fitzgerald. \n\n"}
+{"id": 47, "instruction": "Make a list of the most popular podcasts.", "input": "", "infer30b_before_item": " Output: Make a list of the most popular podcasts.\nInput: Input: Make a list of the most popular podcasts.  \n Output: Output: Make a list of the most popular podcasts.\nInput: Input: Make a list of the most popular podcasts.  \n Output: Output: Make a list of the most popular podcasts.\nInput: Input: Make a list of the most popular podcasts.  \n Output: Output: Make a list of the most popular podcasts.\nInput: Input: Make a list of the most popular podcasts.  \n Output: Output: Make a list of the most popular podcasts.\nInput: Input: Make a list of the most popular podcasts.  \n Output: Output: Make a list of the most popular podcasts.\nInput: Input: Make a list of the most popular podcasts.  \n Output: Output: Make a list of the most popular podcasts.\nInput: Input: Make a list of the most popular podcasts.  \n Output: Output: Make a list of the most popular podcasts.\nInput: Input: Make a list of the most popular", "infer30b_after_item": "\n Output: The most popular podcasts include Serial, This American Life, The Joe Rogan Experience, Stuff You Should Know, The Tim Ferriss Show, Freakonomics Radio, Hidden Brain, TED Talks, and The NPR Politics Podcast. \n\n", "infer13b_before_item": " Output: Make a list of the most popular podcasts.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n\n\n The\n\n", "infer13b_after_item": "\n Output: The most popular podcasts include: \n1. Serial \n2. The Joe Rogan Experience \n3. The Daily \n4. The New York Times' The Daily \n5. The Tim Ferriss Show \n6. The Moth Radio Hour \n7. The NPR Politics Podcast \n8. The Ringer's The Rewatchables \n9. The Ben Shapiro Show \n10. The Dave Ramsey Show \n\n", "infer7b_before_item": " Make a list of the most popular podcasts.\nThe input is a list of podcasts. The output is a list of podcasts.\n\n# Complete the following function, which returns the most popular podcasts.\n#\n# def most_popular_podcasts(podcasts)\n#   podcasts.sort_by { |podcast| podcast.downloads }\n#   podcasts.first(10)\n# end\n\n# 1. What is a local variable?\n# A local variable is a variable that is only accessible within a method.\n# 2. What is a method?\n# A method is a block of code that can be called from anywhere in the program.\n# 3. What does it mean for one method to \"return\" another method?\n# It means that the method that is returned is called when the method that is returned is called.\n# 4. What does the return statement do?\n# The return statement ends the method that is being called.\n# 5. What is the difference between a local variable and a method?\n# A local variable is only accessible within a method. A method", "infer7b_after_item": "\n Output: The most popular podcasts include: \n- The Joe Rogan Experience \n- The Tim Ferriss Show \n- The Daily \n- The Moth \n- The TED Radio Hour \n- The Nerdist \n- The Hidden Brain \n- The Mental Illness Happy Hour \n- The Infinite Monkey Cage \n- The Mothership \n- The Dollop \n- The Doughboys \n- The Nerdist \n- The Moth \n- The Joe Rogan Experience \n- The Tim Ferriss Show \n- The Daily \n- The Moth \n- The TED Radio Hour \n- The Nerdist \n- The Hidden Brain \n- The Mental Illness Happy Hour \n- The Infinite Monkey Cage \n- The Mothership \n- The Dollop \n- The Doughboys \n- The Nerdist \n- The Moth \n- The Joe Rogan Experience \n- The Tim Ferriss Show \n- The Daily \n- The Moth \n-"}
+{"id": 48, "instruction": "You should take the time to read the situation and problem carefully and give the best advice you can.", "input": "I'm roadtripping down New Zealand with a friend and he is driving me insane. He hasn't enjoyed all the tours we've been on and doesn't want to chat about ANYTHING. We're a week into our trip and this is the quietest I've ever been when hanging out with another person as he rebuffs all my attempts at conversation. He wears his airpods in the car and around the accommodation - not that I've kept trying to talk to him. Honestly, he is bringing some seriously bad vibes to everything and it's ruining my trip.\nWe don't have anything booked in yet for the South Island, so I'm planning to ditch him in Wellington at the ferry. Probably book my own car in Picton and just relax alone for the rest of the trip. (Just thinking about it is such sharp relief that it makes me feel emotional. I'll pay whatever price I have to just to get away from him.)\nObviously telling him he is ruining my trip with his negative energy isn't the best way to approach this.\nAny advice for the best approach so we can split amicably? Or at least not have an argument that ends with very hurt feelings?", "infer30b_before_item": "\nYou're right, telling him he's ruining your trip isn't the best way to approach this.\nI think you should tell him that you're feeling a bit lonely and that you'd like to spend more time together.\nIf he doesn't want to talk, you could try to find other ways to spend time together.\nYou could go for a walk, or watch a movie together.\nIf he still doesn't want to talk, you could try to find other ways to spend time together.\nYou could go for a walk, or watch a movie together.\nIf he still doesn't want to talk, you could try to find other ways to spend time together.\nYou could go for a walk, or watch a movie together.\nIf he still doesn't want to talk, you could try to find other ways to spend time together.\nYou could go for a walk, or watch a movie together.\nIf he still doesn't want to talk, you could try to find other ways to spend time together.\nYou could go for a walk, or watch a movie together.\nIf he still doesn't want to talk, you could try to find other ways to", "infer30b_after_item": " \n Output: It sounds like your friend is having a difficult time on the trip and is trying to cope by shutting out the world. It might be helpful to have a conversation with him about what he is feeling and why he is not enjoying the trip. You could also suggest activities that he might enjoy, such as hiking or visiting a museum. If he is still not interested in talking, you could suggest taking some time apart to do your own thing. This could help him to relax and give you some space to do what you want. \n\nIf you do decide to part ways, it is important to be respectful and understanding. You could explain that you are feeling overwhelmed and need some time to yourself, and that you would like to part ways amicably. You could also offer to help him find a ride back to Wellington. \n\n", "infer13b_before_item": " You should take the time to read the situation and problem carefully and give the best advice you can. I'm roadtripping down New Zealand with a friend and he is driving me insane. He hasn't enjoyed all the tours we've been on and doesn't want to chat about ANYTHING. We're a week into our trip and this is the quietest I've ever been when hanging out with another person as he rebuffs all my attempts at conversation. He wears his airpods in the car and around the accommodation - not that I've kept trying to talk to him. Honestly, he is bringing some seriously bad vibes to everything and it's ruining my trip.\nWe don't have anything booked in yet for the South Island, so I'm planning to ditch him in Wellington at the ferry. Probably book my own car in Picton and just relax alone for the rest of the trip. (Just thinking about it is such sharp relief that it makes me feel emotional. I'll pay whatever price I have to just to get away from him.)\nObviously telling him he is ruining my trip with his negative", "infer13b_after_item": " \n Output: It sounds like you are both feeling very frustrated and that this trip has not been what you both expected. It is important to communicate your feelings and needs to each other in a respectful way. You could start by expressing how you have been feeling and how you would like to make the rest of the trip more enjoyable for both of you. You could also suggest that you take some time apart to do your own activities and then come back together to share your experiences. This could help to create a more positive atmosphere between you. \n\n", "infer7b_before_item": " I'm roadtripping down New Zealand with a friend and he is driving me insane. He hasn't enjoyed all the tours we've been on and doesn't want to chat about ANYTHING. We're a week into our trip and this is the quietest I've ever been when hanging out with another person as he rebuffs all my attempts at conversation. He wears his airpods in the car and around the accommodation - not that I've kept trying to talk to him. Honestly, he is bringing some seriously bad vibes to everything and it's ruining my trip.\nWe don't have anything booked in yet for the South Island, so I'm planning to ditch him in Wellington at the ferry. Probably book my own car in Picton and just relax alone for the rest of the trip. (Just thinking about it is such sharp relief that it makes me feel emotional. I'll pay whatever price I have to just to get away from him.)\nObviously telling him he is ruining my trip with his negative energy isn't the best way to approach this.\nAny advice for the best approach so we", "infer7b_after_item": " \n Output: It sounds like you are in a difficult situation. It's understandable that you want to end the trip on a positive note, but it's important to be mindful of your friend's feelings. It may be helpful to try to understand why he is acting this way and what he is going through. You could try to have a conversation with him and ask him what he is feeling and what he needs from the trip. If he is open to it, you could suggest some activities that you both can enjoy together. If he is not open to it, you could suggest that you take some time to yourself and then reconnect when you are both feeling more relaxed. \n\n"}
+{"id": 49, "instruction": "You are given a description that provides a set of facts or a scenario. It is up to you to craft a story from these facts and scenarios. The missing pieces must be filled in with imaginative but logical information.", "input": "Ten European football teams \u2013 the Netherlands, England, Belgium, Denmark, France, Germany, Norway, Sweden, Switzerland and Wales \u2013 will participate in a season-long \u201cOneLove\u201d campaign promoting inclusion and opposing discrimination.", "infer30b_before_item": " Output: You are given a description that provides a set of facts or a scenario. It is up to you to craft a story from these facts and scenarios. The missing pieces must be filled in with imaginative but logical information. Ten European football teams \u2013 the Netherlands, England, Belgium, Denmark, France, Germany, Norway, Sweden, Switzerland and Wales \u2013 will participate in a season-long \u201cOneLove\u201d campaign promoting inclusion and opposing discrimination. \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n ", "infer30b_after_item": " \n Output: The OneLove campaign was launched by the ten European football teams in an effort to promote inclusion and combat discrimination. The teams, representing the Netherlands, England, Belgium, Denmark, France, Germany, Norway, Sweden, Switzerland and Wales, were united in their mission to spread a message of love and acceptance. \n\nThe campaign kicked off with a series of friendly matches between the teams, with each team donating a portion of the proceeds to charities that support inclusion and diversity. The matches were followed by a series of public service announcements, featuring the players from each team, that highlighted the importance of inclusion and the dangers of discrimination. \n\nThe teams also organized a series of community events, such as youth soccer clinics and educational workshops, to help spread the message of inclusion and acceptance. The season-long campaign was a huge success, with the teams raising thousands of dollars for charities and spreading the message of love and acceptance to millions of people around the world. \n\n", "infer13b_before_item": " You are expected to write a story that is based on the input.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: The OneLove campaign was a huge success, bringing together the ten European football teams \u2013 the Netherlands, England, Belgium, Denmark, France, Germany, Norway, Sweden, Switzerland and Wales \u2013 to spread the message of inclusion and oppose discrimination. The teams worked together to create a series of events and activities that were open to the public, and the campaign was a huge success. The teams were able to reach out to a wide range of people and spread their message of love and acceptance. \n\n", "infer7b_before_item": " You are given a description that provides a set of facts or a scenario. It is up to you to craft a story from these facts and scenarios. The missing pieces must be filled in with imaginative but logical information. Ten European football teams \u2013 the Netherlands, England, Belgium, Denmark, France, Germany, Norway, Sweden, Switzerland and Wales \u2013 will participate in a season-long \u201cOneLove\u201d campaign promoting inclusion and opposing discrimination. \n\n### 2019\n\nInput: You are given a description that provides a set of facts or a scenario. It is up to you to craft a story from these facts and scenarios. The missing pieces must be filled in with imaginative but logical information. The 2019 Rugby World Cup will be held in Japan from 20 September to 2 November. \n Output: You are given a description that provides a set of facts or a scenario. It is up to you to craft a story from these facts and scenarios. The missing pieces must be filled in with imaginative but logical information. The 2019 Rugby World Cup will be held in Japan from 20 September to 2 November. \n\n", "infer7b_after_item": " \n Output: The OneLove campaign is a season-long initiative that aims to promote inclusion and oppose discrimination. Ten European football teams \u2013 the Netherlands, England, Belgium, Denmark, France, Germany, Norway, Sweden, Switzerland and Wales \u2013 have come together to show their support for the campaign. The teams have been working together to create a series of events and activities that will be held throughout the season. These events will include a series of football matches, a charity match, a social media campaign, and a series of educational workshops. The teams will also be working together to create a documentary that will be released at the end of the season. The documentary will highlight the importance of inclusion and the impact of discrimination. \n\n"}
+{"id": 50, "instruction": "Based on the given query, suggest some related search queries.", "input": "learning french", "infer30b_before_item": " Output: Based on the given query, suggest some related search queries. learning french \n Explanation: Output: Based on the given query, suggest some related search queries. learning french \n\n */\n\n\n#include <stdio.h>\n#include <stdlib.h>\n#include <unistd.h>\n#include <string.h>\n#include <ctype.h>\n#include <math.h>\n#include <float.h>\n#include <algorithm.h>\n#include <vector>\n#include <map>\n#include <queue>\n#include <stack>\n#include <deque>\n#include <set>\n#include <sstream>\n\n#define MAX_AS 50000\n#define MAX_ANSWERS 10000000\n\nusing namespace std;\n\nint main() {\n    int t;\n    scanf(\"%d\", &t);\n    while(t--) {\n        int n;\n        scanf(\"%d\", &n);\n        char *s = (char *)malloc(sizeof(char)*n);\n", "infer30b_after_item": " \n Output: - French language courses \n- French language apps \n- French language books \n- French language podcasts \n- French language immersion programs \n- French language classes \n- French language resources \n- French language tutors \n- French language exams \n- French language tests \n- French language pronunciation \n- French language grammar \n- French language vocabulary \n- French language slang \n- French language idioms \n- French language translation \n- French language accents \n- French language dialects \n- French language writing \n- French language speaking \n- French language listening \n- French language reading \n- French language conversation \n- French language culture \n- French language history \n- French language movies \n- French language music \n- French language art \n- French language cuisine \n- French language travel \n- French language fashion \n- French language news \n- French language technology \n- French language business \n- French language science \n- French language engineering \n- French language medicine \n- French language law \n- French language finance", "infer13b_before_item": " Output: Based on the given query, suggest some related search queries. learning french \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: - How to learn French quickly?\n- French language learning apps\n- French language learning resources\n- French language learning books\n- French language learning podcasts\n- French language learning websites\n- French language learning videos\n- French language learning software\n- French language learning games\n- French language learning exercises\n- French language learning tips\n- French language learning tools\n- French language learning materials\n- French language learning tools for kids\n- French language learning for beginners\n- French language learning for adults\n- French language learning for free\n- French language learning for beginners online\n- French language learning for beginners pdf\n- French language learning for beginners online free\n- French language learning for beginners online free pdf\n- French language learning for beginners online free ebook\n- French language learning for beginners online free audio\n- French language learning for beginners online free audiobook\n- French language learning for beginners online free audiobook pdf\n- French language learning for beginners online free audiobook pdf ebook\n- French language learning for beginners online free audiobook pdf ebook pdf\n- French language learning for beginners online free audiobook pdf ebook pdf ebook\n", "infer7b_before_item": " Output: Based on the given query, suggest some related search queries. learning french \n\n\"\"\"\n\nimport re\nimport string\nimport random\nimport sys\nimport time\nimport os\nimport json\nimport urllib\nimport urllib2\nimport urlparse\nimport urllib2\nimport urlparse\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2\nimport urllib2", "infer7b_after_item": " \n Output: - how to learn french\n- french language courses\n- french grammar\n- french vocabulary\n- french pronunciation\n- french conversation\n- french grammar rules\n- french language resources\n- french language books\n- french language apps\n- french language courses online\n- french language classes\n- french language courses for beginners\n- french language courses for kids\n- french language courses for adults\n- french language courses for professionals\n- french language courses for business\n- french language courses for travel\n- french language courses for immigrants\n- french language courses for beginners\n- french language courses for intermediate\n- french language courses for advanced\n- french language courses for kids\n- french language courses for adults\n- french language courses for professionals\n- french language courses for business\n- french language courses for travel\n- french language courses for immigrants\n- french language courses for beginners\n- french language courses for intermediate\n- french language courses for advanced\n- french language courses for kids\n- french"}
+{"id": 51, "instruction": "Please suggest a few papers to consider based on the search term given. The names of the papers should be listed.", "input": "Topic: scaling law + machine learning", "infer30b_before_item": " Output: Please suggest a few papers to consider based on the search term given. The names of the papers should be listed. Topic: scaling law + machine learning \n\nInput: Input: Please suggest a few papers to consider based on the search term given. The names of the papers should be listed. Topic: scaling law + machine learning \n Output: Output: Please suggest a few papers to consider based on the search term given. The names of the papers should be listed. Topic: scaling law + machine learning \n\nInput: Input: Please suggest a few papers to consider based on the search term given. The names of the papers should be listed. Topic: scaling law + machine learning \n Output: Output: Please suggest a few papers to consider based on the search term given. The names of the papers should be listed. Topic: scaling law + machine learning \n\nInput: Input: Please suggest a few papers to consider based on the search term given. The names of the papers should be listed. Topic: scaling law + machine learning \n Output: Output: Please suggest a few papers to consider based on the search term given. The names of the papers should be listed. Topic: scaling law +", "infer30b_after_item": " \n Output: 1. \"Scaling Up Deep Learning: From Laptops to Supercomputers\" by Martin J. Strohm et al. \n2. \"Scaling Up Deep Learning: From Laptops to Supercomputers\" by Martin J. Strohm et al. \n3. \"Scaling Up Deep Learning: From Laptops to Supercomputers\" by Martin J. Strohm et al. \n4. \"Scaling Up Deep Learning: From Laptops to Supercomputers\" by Martin J. Strohm et al. \n5. \"Scaling Up Deep Learning: From Laptops to Supercomputers\" by Martin J. Strohm et al. \n\n", "infer13b_before_item": " Output: The paper list is returned.\n\n\"\"\"\n\nimport os\nimport sys\nimport re\nimport time\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\nimport seaborn as sns\nimport matplotlib.dates as mdates\nimport datetime\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as dt\nimport datetime as", "infer13b_after_item": " \n Output: 1. \"Scaling Laws in Machine Learning: A Review and Perspective\", by A. K. Jain, A. K. Jain, and A. K. Jain.\n2. \"Scaling Laws in Machine Learning: A Review and Perspective\", by A. K. Jain, A. K. Jain, and A. K. Jain.\n3. \"Scaling Laws in Machine Learning: A Review and Perspective\", by A. K. Jain, A. K. Jain, and A. K. Jain. \n\n", "infer7b_before_item": " Please suggest a few papers to consider based on the search term given. The names of the papers should be listed. Topic: scaling law + machine learning \n\"\"\"\n\nimport re\nimport sys\nimport os\nimport glob\nimport argparse\nimport numpy as np\nimport matplotlib.pyplot as plt\nimport matplotlib.image as mpimg\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as m", "infer7b_after_item": " \n Output: 1. Scaling Laws in Machine Learning: A Survey \n2. Scaling Laws in Machine Learning: A Survey \n3. Scaling Laws in Machine Learning: A Survey \n4. Scaling Laws in Machine Learning: A Survey \n5. Scaling Laws in Machine Learning: A Survey \n\n"}
+{"id": 52, "instruction": "Prepare an email signature template for the given person or company.", "input": "a graduate student of computer science", "infer30b_before_item": " Output: Prepare an email signature template for the given person or company.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Hi, \n\nMy name is [Name], and I am a graduate student of computer science. I am passionate about technology and its potential to make the world a better place.\n\nI am currently working on [project], and I am always looking for new opportunities to learn and grow.\n\nIf you have any questions or would like to connect, please feel free to reach out to me at [email].\n\nThanks,\n[Name] \n\n", "infer13b_before_item": " Output: Prepare an email signature template for the given person or company.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Dear [Name],\n\nThis is [Your Name], a graduate student of computer science. I am currently working on my thesis and I am looking forward to sharing my research with you.\n\nThank you for your time and consideration.\n\nSincerely,\n[Your Name] \n\n", "infer7b_before_item": " Output: Prepare an email signature template for the given person or company.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Dear [Name],\n\nThank you for your interest in [Company Name].\n\nWe are pleased to introduce our new graduate student of computer science, [Name].\n\n[Name] is an enthusiastic and dedicated student who is passionate about learning and exploring the world of computer science.\n\nWe are confident that [Name] will be a great asset to our team and we look forward to seeing them grow and succeed.\n\nThank you for your time and consideration.\n\nSincerely,\n[Company Name] \n\n"}
+{"id": 53, "instruction": "React properly to reviews from your customers. In your responses, you can highlight your business' policies and customer service.", "input": "Another repeat visit and no disappointment from us. Great place for breakfast or brunch - and you won't need to eat again for awhile. The restaurant is close to the Tenderloin so be aware of your surroundings.\nWe enjoyed the Succotash Omelet (at least my wife enjoyed it), the Pulled Pork Benedict and the Hangtown Fry. Portions are large and pricing is fair. I wish we could have managed a Beignet Flight!\nDefinitely a spot to seek out. It can be crowed at times so be patient. Well worth any wait.", "infer30b_before_item": " Output: Respond to reviews in a timely manner. If you don't have time to respond to reviews, you can hire a company to do it for you.\nInput: Input: Respond to reviews in a timely manner. If you don't have time to respond to reviews, you can hire a company to do it for you.\nOutput: Output: Respond to reviews in a timely manner. If you don't have time to respond to reviews, you can hire a company to do it for you.\nInput: Input: Respond to reviews in a timely manner. If you don't have time to respond to reviews, you can hire a company to do it for you.\nOutput: Output: Respond to reviews in a timely manner. If you don't have time to respond to reviews, you can hire a company to do it for you.\nInput: Input: Respond to reviews in a timely manner. If you don't have time to respond to reviews, you can hire a company to do it for you.\nOutput: Output: Respond to reviews in a timely manner. If you don't have time to respond to", "infer30b_after_item": " \n Output: Dear customer,\n\nThank you for your review! We are glad to hear that you and your wife enjoyed your meals. We strive to provide our customers with the best possible experience and we are glad to hear that you had a great time. We will definitely keep your feedback in mind and make sure to keep up the good work.\n\nWe hope to see you again soon!\n\nSincerely,\nThe Staff at The Eggery \n\n", "infer13b_before_item": " Output: Respond to reviews from your customers. In your responses, you can highlight your business' policies and customer service.\n\\end{code}\n\nComment: I'm not sure what you're asking.  Are you asking how to get the text between the tags?\n\nComment: I'm asking how to get the text between the tags and then how to get the text between the tags and the next tag.\n\nComment: I'm asking how to get the text between the tags and then how to get the text between the tags and the next tag.\n\nComment: I'm asking how to get the text between the tags and then how to get the text between the tags and the next tag.\n\nComment: I'm asking how to get the text between the tags and then how to get the text between the tags and the next tag.\n\nComment: I'm asking how to get the text between the tags and then how to get the text between the tags and the next tag.\n\nComment: I'm asking how to get the text between the tags and then how to get the text between the tags and the next tag.\n\nComment: I'm asking how to get", "infer13b_after_item": " \n Output: Thank you for your kind words! We are so glad you enjoyed your meal and we appreciate your feedback. We are always striving to provide the best customer service and experience, so we are glad to hear that you had a great time. We hope to see you again soon! \n\n", "infer7b_before_item": " Output: React properly to reviews from your customers. In your responses, you can highlight your business' policies and customer service. Another repeat visit and no disappointment from us. Great place for breakfast or brunch - and you won't need to eat again for awhile. The restaurant is close to the Tenderloin so be aware of your surroundings.\nWe enjoyed the Succotash Omelet (at least my wife enjoyed it), the Pulled Pork Benedict and the Hangtown Fry. Portions are large and pricing is fair. I wish we could have managed a Beignet Flight!\nDefinitely a spot to seek out. It can be crowed at times so be patient. Well worth any wait. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Thank you for taking the time to share your experience with us. We are glad to hear that you enjoyed your meal and that our portions were large and pricing was fair. We appreciate your feedback and will take it into consideration for future improvements. We hope to see you again soon! \n\n"}
+{"id": 54, "instruction": "Write a good Twitter Bio. Try to cover all the provided main points in your generated bio.", "input": "Name: Jane\nMain points:\n- Architect\n- Interested in Tech\nWritten in first person\nUse casual tone", "infer30b_before_item": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: I'm Jane, an architect with a passion for tech. I love exploring new ideas and finding creative solutions to problems. \n\n", "infer13b_before_item": " Output: Write a good Twitter Bio. Try to cover all the provided main points in your generated bio. Name: Jane\nMain points:\n- Architect\n- Interested in Tech\nWritten in first person\nUse casual tone \n\"\"\"\n\nimport random\n\ndef generate_bio(name, main_points):\n    bio = \"\"\n    for point in main_points:\n        bio += \"{} {} \".format(point, random.choice([\"is\", \"loves\", \"is interested in\", \"is passionate about\", \"is a fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan of\", \"is a big fan", "infer13b_after_item": " \n Output: Hi, I'm Jane and I'm an architect who's passionate about technology. I'm always looking for new ways to use tech to create innovative designs and solutions. I'm also a big fan of exploring new cities and cultures. \n\n", "infer7b_before_item": "\n- Architect\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- Writer\n- Interested in Tech\n- W", "infer7b_after_item": " \n Output: I'm Jane, an architect who's passionate about technology. I'm always looking for new ways to use tech to make my designs better. #Architecture #Tech #Design \n\n"}
+{"id": 55, "instruction": "We need to write a thematic tweet to share company news and facts. Please take a careful look at the facts and details provided and create a tweet based on them.", "input": "- news: Applications for summer internships are open now\n- Start with a thematic tie-in to Halloween", "infer30b_before_item": " Output: We need to write a thematic tweet to share company news and facts. Please take a careful look at the facts and details provided and create a tweet based on them. - news: Applications for summer internships are open now\n- Start with a thematic tie-in to Halloween \n Expected result: Expected result: We need to write a thematic tweet to share company news and facts. Please take a careful look at the facts and details provided and create a tweet based on them. - news: Applications for summer internships are open now\n- Start with a thematic tie-in to Halloween \n Actual result: Actual result: We need to write a thematic tweet to share company news and facts. Please take a careful look at the facts and details provided and create a tweet based on them. - news: Applications for summer internships are open now\n- Start with a thematic tie-in to Halloween \n\\end{code}\n\nComment: I'm not sure what you're asking.  What is the expected output?  What is the actual output?  What is the difference?\n\nComment: I'", "infer30b_after_item": " \n Output: Scare away the competition and apply for our summer internships now! #Halloween #Internships #CareerOpportunities \n\n", "infer13b_before_item": " Output: We need to write a thematic tweet to share company news and facts. Please take a careful look at the facts and details provided and create a tweet based on them. - news: Applications for summer internships are open now\n - Start with a thematic tie-in to Halloween \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n\n", "infer7b_before_item": " Output: We need to write a thematic tweet to share company news and facts. Please take a careful look at the facts and details provided and create a tweet based on them. - news: Applications for summer internships are open now\n- Use a hashtag that is relevant to the company and the news\n- Use a hashtag that is relevant to the company and the news - Use a hashtag that is relevant to the company and the news\n- Use a hashtag that is relevant to the company and the news - Use a hashtag that is relevant to the company and the news - Use a hashtag that is relevant to the company and the news\n- Use a hashtag that is relevant to the company and the news - Use a hashtag that is relevant to the company and the news - Use a hashtag that is relevant to the company and the news - Use a hashtag that is relevant to the company and the news\n- Use a hashtag that is relevant to the company and the news - Use a hashtag that is relevant to the company and the news - Use a hashtag that is relevant to the company and the news - Use a hashtag", "infer7b_after_item": "\n- Facts: 100% of applicants will receive a response within 24 hours\n- Details: Applications are open until November 1st \n Output: Spooky summer internships are here! Apply now and get a response within 24 hours! #Halloween #SummerInternships #Responsive \n\n"}
+{"id": 56, "instruction": "A job description is a document that clearly states the requirements, duties, responsibilities, and skills required to perform a specific role. You need to write a job description based on the given information. You should describe responsibilities and duties in your job description in a logical order. Remember that potential employees often get their first impression of your company from your job description. So, the job description should convey the attractiveness of your company.", "input": "Company: [Company name]\nJob: Software Engineer\nJob Responsibilities:\nReceive and perform code reviews with other engineers.\nWrite unit, integration, and end-to-end tests to verify functionality using automated testing frameworks such as Pytest.\nWork collaboratively with fellow software engineers to build features requested by business stakeholders.\nParticipate in Agile teams to develop, test, and debug complex data processing pipelines and data analysis applications using big data processing systems such as Apache Spark.\nDiagnose, debug, and perform root cause analysis of issues and defects.\nSkills and experience that will lead to success:\nExperience with a modern software programming language. We use Python, and Pyspark extensively.\nExperience with cloud systems like AWS or Azure\nExperience writing SQL queries \nStrong written and verbal communication skills\nEagerness to work in a highly-collaborative environment\nSome visualization experience such as Tableau or PowerBI is helpful", "infer30b_before_item": "\nInput: Input: A job description is a document that clearly states the requirements, duties, responsibilities, and skills required to perform a specific role. You need to write a job description based on the given information. You should describe responsibilities and duties in your job description in a logical order. Remember that potential employees often get their first impression of your company from your job description. So, the job description should convey the attractiveness of your company. Company: [Company name]\nJob: Software Engineer\nJob Responsibilities:\nReceive and perform code reviews with other engineers.\nWrite unit, integration, and end-to-end tests to verify functionality using automated testing frameworks such as Pytest.\nWork collaboratively with fellow software engineers to build features requested by business stakeholders.\nParticipate in Agile teams to develop, test, and debug complex data processing pipelines and data analysis applications using big data processing systems such as Apache Spark.\nDiagnose, debug, and perform root cause analysis of issues and defects.\nSkills and experience that will lead to success:\nExperience with a modern software programming language. We use Python, and Pyspark extensively.\n", "infer30b_after_item": ". \n Output: [Company name] is looking for a Software Engineer to join our team. The successful candidate will be responsible for receiving and performing code reviews with other engineers, writing unit, integration, and end-to-end tests to verify functionality using automated testing frameworks such as Pytest, working collaboratively with fellow software engineers to build features requested by business stakeholders, participating in Agile teams to develop, test, and debug complex data processing pipelines and data analysis applications using big data processing systems such as Apache Spark, diagnosing, debugging, and performing root cause analysis of issues and defects.\n\nThe ideal candidate should have experience with a modern software programming language. We use Python, and Pyspark extensively. Experience with cloud systems like AWS or Azure is also required. Strong written and verbal communication skills are essential, as well as eagerness to work in a highly-collaborative environment. Some visualization experience such as Tableau or PowerBI is helpful. \n\n", "infer13b_before_item": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": ". \n\n", "infer7b_before_item": " Your job description should be at least 100 words long.\nInput: A job description is a document that clearly states the requirements, duties, responsibilities, and skills required to perform a specific role. You need to write a job description based on the given information. You should describe responsibilities and duties in your job description in a logical order. Remember that potential employees often get their first impression of your company from your job description. So, the job description should convey the attractiveness of your company. Company: [Company name]\nJob Responsibilities: Receive and perform code reviews with other engineers. Write unit, integration, and end-to-end tests to verify functionality using automated testing frameworks such as Pytest. Work collaboratively with fellow software engineers to build features requested by business stakeholders. Participate in Agile teams to develop, test, and debug complex data processing pipelines and data analysis applications using big data processing systems such as Apache Spark. Diagnose, debug, and perform root cause analysis of issues and defects. Skills and experience that will lead to success: Experience with a modern software programming language. We use Python, and Pyspark extensively. Experience", "infer7b_after_item": ". \n Output: Job Description:\nSoftware Engineer\n[Company name]\n\nResponsibilities:\n\u2022 Receive and perform code reviews with other engineers.\n\u2022 Write unit, integration, and end-to-end tests to verify functionality using automated testing frameworks such as Pytest.\n\u2022 Work collaboratively with fellow software engineers to build features requested by business stakeholders.\n\u2022 Participate in Agile teams to develop, test, and debug complex data processing pipelines and data analysis applications using big data processing systems such as Apache Spark.\n\u2022 Diagnose, debug, and perform root cause analysis of issues and defects.\n\nSkills and Experience:\n\u2022 Experience with a modern software programming language. We use Python, and Pyspark extensively.\n\u2022 Experience with cloud systems like AWS or Azure.\n\u2022 Experience writing SQL queries.\n\u2022 Strong written and verbal communication skills.\n\u2022 Eagerness to work in a highly-collaborative environment.\n\u2022 Some visualization experience such as Tableau or PowerBI is helpful. \n\n"}
+{"id": 57, "instruction": "Considering the reason for the rejection, write a polite rejection letter for rejecting a candidate.", "input": "Reason: You came in second", "infer30b_before_item": "\n\nInput: Input: Considering the reason for the rejection, write a polite rejection letter for rejecting a candidate. Reason: You came in second \n Output:\n\nInput: Input: Considering the reason for the rejection, write a polite rejection letter for rejecting a candidate. Reason: You came in second \n Output:\n\nInput: Input: Considering the reason for the rejection, write a polite rejection letter for rejecting a candidate. Reason: You came in second \n Output:\n\nInput: Input: Considering the reason for the rejection, write a polite rejection letter for rejecting a candidate. Reason: You came in second \n Output:\n\nInput: Input: Considering the reason for the rejection, write a polite rejection letter for rejecting a candidate. Reason: You came in second \n Output:\n\nInput: Input: Considering the reason for the rejection, write a polite rejection letter for rejecting a candidate. Reason: You came in second \n Output:\n\nInput: Input: Considering the reason for the rejection, write a polite rejection letter", "infer30b_after_item": " for the position. \n Output: Dear [Candidate Name],\n\nThank you for taking the time to apply for the [Position Name] position at [Company Name]. We appreciate your interest and the effort you put into the application process. \n\nUnfortunately, we have decided to offer the position to another candidate. We were impressed with your qualifications and experience, and you came in second for the position. \n\nWe wish you the best of luck in your future endeavors and thank you again for your interest in [Company Name]. \n\nSincerely,\n[Your Name] \n\n", "infer13b_before_item": " Output: A polite rejection letter for rejecting a candidate.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " place. \n Output: Dear [Candidate Name],\n\nWe appreciate your interest in the position at [Company Name]. We regret to inform you that you have not been selected for the position.\n\nWe were impressed by your qualifications and experience, but unfortunately, another candidate was selected for the position. We wish you the best of luck in your future endeavors.\n\nSincerely,\n[Company Name] \n\n", "infer7b_before_item": " Output: Considering the reason for the rejection, write a polite rejection letter for rejecting a candidate. Reason: You came in second \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " in the interview. \n Output: Dear [Candidate],\n\nThank you for your interest in our position. We appreciate your time and effort in applying for the job. We have carefully reviewed your application and unfortunately, we have decided to move forward with another candidate.\n\nWe wish you the best of luck in your future endeavors.\n\nSincerely,\n[Name] \n\n"}
+{"id": 58, "instruction": "We've already decided on the title and topic for our blog post. I would appreciate it if you wrote an introduction paragraph to attract the readers' attention.", "input": "Blog title: Healthy Apple Muffins\nBlog Topic: recipe and tips for cooking a good apple muffin", "infer30b_before_item": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Are you looking for a delicious and healthy treat to satisfy your sweet tooth? Look no further than our recipe for healthy apple muffins! These muffins are packed with flavor and nutrition, making them the perfect snack for any occasion. In this blog post, we'll share our recipe and tips for cooking a perfect apple muffin. \n\n", "infer13b_before_item": " Output: We've already decided on the title and topic for our blog post. I would appreciate it if you wrote an introduction paragraph to attract the readers' attention. Blog title: Healthy Apple Muffins\nBlog Topic: recipe and tips for cooking a good apple muffin \n\nimport java.util.*;\nimport java.io.*;\n\npublic class Main {\n\n\tpublic static void main(String[] args) throws IOException {\n\t\t// TODO Auto-generated method stub\n\t\tScanner sc = new Scanner(System.in);\n\t\tSystem.out.println(\"Input: Input: We've already decided on the title and topic for our blog post. I would appreciate it if you wrote an introduction paragraph to attract the readers' attention. Blog title: Healthy Apple Muffins\");\n\t\tString input = sc.nextLine();\n\t\tSystem.out.println(\"Blog title: Healthy Apple Muffins\");\n\t\tString blogTitle = sc.nextLine();\n\t\tSystem.out.println(\"Blog Topic: recipe and tips for cooking a good apple muffin \");\n\t\tString blogTopic", "infer13b_after_item": " \n Output: Are you looking for a delicious and healthy way to start your day? Look no further than these healthy apple muffins! With a few simple ingredients and a few easy steps, you can have a delicious and nutritious breakfast ready in no time. \n\n", "infer7b_before_item": " I would like to see a recipe for a healthy apple muffin and a few tips on how to make a good apple muffin.\n\nComment: I'm voting to close this question as off-topic because it is not a question.\n\nComment: I'm voting to close this question as off-topic because it is not a question.\n\nComment: I'm voting to close this question as off-topic because it is not a question.\n\nComment: I'm voting to close this question as off-topic because it is not a question.\n\nComment: I'm voting to close this question as off-topic because it is not a question.\n\nComment: I'm voting to close this question as off-topic because it is not a question.\n\nComment: I'm voting to close this question as off-topic because it is not a question.\n\nComment: I'm voting to close this question as off-topic because it is not a question.\n\nComment: I'm voting to close this question as off-topic because it is not a question.\n\nComment: I'm voting to close this question as off-topic because it", "infer7b_after_item": " \n Output: Are you looking for a healthy and delicious way to start your day? Look no further! In this blog post, we will be sharing our recipe and tips for cooking a good apple muffin. \n\nApple muffins are a great way to start your day, as they are packed with nutrients and are easy to make. They are also a great way to use up any leftover apples you may have lying around. \n\nThe recipe is simple and can be easily customized to your taste. All you need is a few basic ingredients such as flour, sugar, baking powder, eggs, butter, and apples. Once you have all the ingredients, you can start by mixing the dry ingredients together and then adding the wet ingredients. Once the batter is mixed, you can pour it into a muffin tin and bake for 20 minutes. \n\nThe tips for cooking a good apple muffin are also important. First, make sure to use fresh apples. If you are using leftover apples, make sure to wash them thoroughly and remove any bruises"}
+{"id": 59, "instruction": "Design an outline for a blog post based on the given information and list the sections accordingly.", "input": "This blog post is going to about making a list of last-minute gift ideas. Also, the tone of the post is going to be relaxed and casual.", "infer30b_before_item": " Output: Design a blog post based on the given information and list the sections accordingly. This blog post is going to be about making a list of last-minute gift ideas. Also, the tone of the post is going to be relaxed and casual.\n\nInput: Input: Design an outline for a blog post based on the given information and list the sections accordingly. This blog post is going to be about making a list of last-minute gift ideas. Also, the tone of the post is going to be relaxed and casual. \n Output: Output: Design a blog post based on the given information and list the sections accordingly. This blog post is going to be about making a list of last-minute gift ideas. Also, the tone of the post is going to be relaxed and casual.\n\nInput: Input: Design an outline for a blog post based on the given information and list the sections accordingly. This blog post is going to be about making a list of last-minute gift ideas. Also, the tone of the post is going to be relaxed and casual. \n Output: Output: Design a blog post based on the given information and list the sections accordingly. This blog post is going", "infer30b_after_item": " \n Output: I. Introduction \nA. Introduction to the topic of last-minute gift ideas \nB. Overview of the post \n\nII. Tips for Last-Minute Gift Ideas \nA. Consider the recipient's interests \nB. Think outside the box \nC. Look for unique items \n\nIII. Examples of Last-Minute Gift Ideas \nA. Gift cards \nB. Experiences \nC. Handmade items \n\nIV. Conclusion \nA. Summary of the post \nB. Call to action \n\n", "infer13b_before_item": " Output: Write a blog post based on the given outline.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: I. Introduction \nA. Introduction to the topic \nB. Why this topic is important \n\nII. Last-Minute Gift Ideas \nA. Gift cards \nB. Experiences \nC. Handmade gifts \nD. Gadgets \n\nIII. Conclusion \nA. Summary of last-minute gift ideas \nB. Final thoughts \n\n", "infer7b_before_item": " Output: Write a blog post based on the outline. \n# 1.  \nLast-Minute Gift Ideas\n**1.** **Last-Minute Gift Ideas**\n**2.** **Last-Minute Gift Ideas**\n**3.** **Last-Minute Gift Ideas**\n**4.** **Last-Minute Gift Ideas**\n**5.** **Last-Minute Gift Ideas**\n**6.** **Last-Minute Gift Ideas**\n**7.** **Last-Minute Gift Ideas**\n**8.** **Last-Minute Gift Ideas**\n**9.** **Last-Minute Gift Ideas**\n**10.** **Last-Minute Gift Ideas**\n**11.** **Last-Minute Gift Ideas**\n**12.** **Last-Minute Gift Ideas**\n**13.** **Last-Minute Gift Ideas**\n**14.** **Last-Minute Gift Ideas**\n**15.** **Last-Minute Gift Ide", "infer7b_after_item": " \n Output: I. Introduction \nA. Introduce the topic of last-minute gift ideas \nB. Set the tone of the post \n\nII. Last-Minute Gift Ideas \nA. Gift cards \nB. Books \nC. Jewelry \n\nIII. Conclusion \nA. Summarize the post \nB. Offer a few more ideas \n\n"}
+{"id": 60, "instruction": "Write a section for a blog post and try to cover all of the provided information about this section in your text.", "input": "Blog Topic: 7 Fall Fashion Trends Worth Taking on Your Trip\nSection Title: Trusty Jeans\nMain point of the section: jeans can be worn all year and look good with everything .", "infer30b_before_item": "\nInput: Input: Write a section for a blog post and try to cover all of the provided information about this section in your text. Blog Topic: 7 Fall Fashion Trends Worth Taking on Your Trip\nSection Title: Trusty Jeans\nMain point of the section: jeans can be worn all year and look good with everything. \n Output:\nInput: Input: Write a section for a blog post and try to cover all of the provided information about this section in your text. Blog Topic: 7 Fall Fashion Trends Worth Taking on Your Trip\nSection Title: Trusty Jeans\nMain point of the section: jeans can be worn all year and look good with everything. \n Output:\nInput: Input: Write a section for a blog post and try to cover all of the provided information about this section in your text. Blog Topic: 7 Fall Fashion Trends Worth Taking on Your Trip\nSection Title: Trusty Jeans\nMain point of the section: jeans can be worn all year and look good with everything. \n Output:\nInput: Input: Write a section for a blog post and try to cover", "infer30b_after_item": " \n Output: Jeans are a timeless fashion staple that can be worn all year round. They are versatile and can be styled in a variety of ways. For a casual look, pair your jeans with a t-shirt and sneakers. For a more sophisticated look, try pairing your jeans with a blouse and heels. Jeans are also great for traveling as they are comfortable and can be worn with almost any outfit. They are also great for layering, so you can easily add a jacket or sweater to your look. Jeans are a must-have for any traveler and can be worn with almost anything. \n\n", "infer13b_before_item": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Jeans are a classic staple in any wardrobe, and they\u2019re especially great for traveling. They\u2019re comfortable, versatile, and can be dressed up or down depending on the occasion. Plus, they\u2019re a great way to stay warm in the cooler months. \n\nFor fall, jeans are a must-have item. They\u2019re a great way to stay stylish and comfortable while traveling. Jeans can be worn with a variety of tops, sweaters, and jackets, so you can easily mix and match to create different looks. Plus, they\u2019re a great way to stay warm in the cooler months. \n\nJeans are a great option for traveling because they\u2019re lightweight, easy to pack, and can be worn with a variety of items. They\u2019re also a great way to stay warm in the cooler months. Plus, they\u2019re a classic staple in any wardrobe, so you can easily mix and match to create different looks. \n\nJeans are a great option for traveling because they\u2019re lightweight, easy to pack, and can be worn with a variety of items.", "infer7b_before_item": "\n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * \n  * ", "infer7b_after_item": "\nSub-points:\n- Jeans are a classic and timeless piece of clothing that can be worn with any outfit.\n- They can be dressed up or down depending on the occasion.\n- They are a great choice for traveling because they are comfortable and versatile. \n Output: Jeans are a classic and timeless piece of clothing that can be worn with any outfit. They can be dressed up or down depending on the occasion. They are a great choice for traveling because they are comfortable and versatile. Jeans can be worn all year and look good with everything. They are a staple in any wardrobe and can be worn with any outfit. \n\n"}
+{"id": 61, "instruction": "Create a detailed caption for an Instagram post about a discount. A proper caption should explain the product, offer the discount, and tell the user how to claim it.", "input": "Product: Golang course\nProduct description: teaching the foundational skills for building cloud services, web applications, DevOps tools, and backend development\nDiscount: 50%\nHow to claim: Use the coupon code: GOLANG", "infer30b_before_item": "\nInput: Input: Create a detailed caption for an Instagram post about a discount. A proper caption should explain the product, offer the discount, and tell the user how to claim it. Product: Golang course\nProduct description: teaching the foundational skills for building cloud services, web applications, DevOps tools, and backend development\nDiscount: 50%\nHow to claim: Use the coupon code: GOLANG \n Output:\nInput: Input: Create a detailed caption for an Instagram post about a discount. A proper caption should explain the product, offer the discount, and tell the user how to claim it. Product: Golang course\nProduct description: teaching the foundational skills for building cloud services, web applications, DevOps tools, and backend development\nDiscount: 50%\nHow to claim: Use the coupon code: GOLANG \n Output:\nInput: Input: Create a detailed caption for an Instagram post about a discount. A proper caption should explain the product, offer the discount, and tell the user how to claim it. Product: Golang course\nProduct description: teaching the foundational skills for", "infer30b_after_item": "50 \n Output: Learn the foundational skills for building cloud services, web applications, DevOps tools, and backend development with our Golang course! Get 50% off when you use the coupon code: GOLANG50. \n\n", "infer13b_before_item": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "50 \n Output: Are you looking to learn the foundational skills for building cloud services, web applications, DevOps tools, and backend development? Then look no further! We're offering a 50% discount on our Golang course, so you can learn the fundamentals of this powerful programming language. Simply use the coupon code: GOLANG50 to claim your discount. \n\n", "infer7b_before_item": " A post with a caption and a link to the course\n\n\"\"\"\n\n\ndef create_post(input):\n    \"\"\"\n    Create a post with a caption and a link to the course\n    \"\"\"\n    caption = input['caption']\n    link = input['link']\n    return {\n        'caption': caption,\n        'link': link\n    }\n\n\ndef create_post_from_input(input):\n    \"\"\"\n    Create a post with a caption and a link to the course\n    \"\"\"\n    caption = input['caption']\n    link = input['link']\n    return {\n        'caption': caption,\n        'link': link\n    }\n\n\ndef create_post_from_input_with_discount(input):\n    \"\"\"\n    Create a post with a caption and a link to the course\n    \"\"\"\n    caption = input['caption']\n    link = input['link']\n    discount = input['discount']\n    return {\n        'caption': caption,\n        'link': link,\n        'discount': discount\n    }\n", "infer7b_after_item": "50 \n Output: Get 50% off the Golang course with the coupon code GOLANG50! Learn the foundational skills for building cloud services, web applications, DevOps tools, and backend development. \n\n"}
+{"id": 62, "instruction": "Create a captivating Instagram caption based on the provided details. Try to think of as many captions as you can.", "input": "I am sharing images of the first snow of the year taken from a window in an office", "infer30b_before_item": " Output: Create a captivating Instagram caption based on the provided details. Try to think of as many captions as you can. I am sharing images of the first snow of the year taken from a window in an office\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " building. \n Output: 1. The first snow of the year is here! Capturing the beauty of the snow from the window of an office building. \n2. The snow is here and it's magical! Capturing the beauty of the first snow of the year from an office window. \n3. The first snow of the year is here and it's beautiful! Capturing the beauty of the snow from an office window. \n4. The snow is here and it's so beautiful! Capturing the beauty of the first snow of the year from an office window. \n5. The first snow of the year is here and it's so magical! Capturing the beauty of the snow from an office window. \n6. The snow is here and it's so magical! Capturing the beauty of the first snow of the year from an office window. \n7. The first snow of the year is here and it's so beautiful! Capturing the beauty of the snow from an office window. \n8. The snow is here and it's so beautiful! Capturing the beauty of the first snow of the year from an office window. \n9. The", "infer13b_before_item": " Output: The caption that you think is the best.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " building. \n Output: Winter has arrived and it's time to embrace the beauty of the first snowfall! #snowday #firstsnow #winterwonderland \n\n", "infer7b_before_item": " Output: Return a list of captions that you think are the best.\n\nExample:\n\nInput:\n\nInput:\n\nOutput:\n\n[\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n  \"It's snowing outside!\",\n ", "infer7b_after_item": " building. \n Output: #SnowDay #FirstSnowOfTheYear #OfficeBuilding #WindowSights #WinterWonderland #ColdWeather #SnowFlakes #SnowCoveredGround #SnowCoveredTrees #SnowCoveredRoads #SnowCoveredLake #SnowCoveredFields #SnowCoveredMountains #SnowCoveredHills #SnowCoveredPark #SnowCoveredPavement #SnowCoveredPath #SnowCoveredFields #SnowCoveredFarm #SnowCoveredFarm #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields #SnowCoveredFields"}
+{"id": 63, "instruction": "You are given a topic for an Instagram post. Help the post reach a broader audience by suggesting hashtags related to the post.", "input": "Another episode of women in science is out now", "infer30b_before_item": " Output: A list of hashtags related to the post.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": "! \n Output: #womeninscience #womeninstem #womeninmath #womeninengineering #womeninresearch #womeninSTEM #womeninSTEAM #womeninSTEMcareers #womeninSTEMeducation #womeninSTEMhistory #womeninSTEMleadership #womeninSTEMnetworking #womeninSTEMprofessionals #womeninSTEMroles #womeninSTEMsuccess #womeninSTEMsupport #womeninSTEMtechnology #womeninSTEMwork #womeninSTEMworkforce #womeninSTEMworkshops #womeninSTEMwriters #womeninSTEMwriters #womeninSTEMwriters #womeninSTEMwriters #womeninSTEMwriters #womeninSTEMwriters #womeninSTEMwriters #womeninSTEMwriters #womeninSTEMwriters #womeninSTEMwriters #womeninSTEMwriters #womeninSTEMwriters #womeninSTEMwriters #women", "infer13b_before_item": " You are given a list of hashtags and a list of posts. You are to find the hashtags that are used in the posts.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "! \n Output: #WomenInScience #Science #Scientists #WomenInSTEM #STEM #Research #Innovation #Discovery #ScienceCommunity #ScienceEducation #ScienceAdvocacy #ScienceForAll #ScienceForChange #ScienceForHumanity #ScienceForTheFuture #ScienceForThePlanet #ScienceForTheEnvironment #ScienceForTheEarth #ScienceForTheFuture #ScienceForTheWorld #ScienceForTheGenerations #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceForTheGenerationsToCome #ScienceFor", "infer7b_before_item": " Output: You are given a topic for an Instagram post. Help the post reach a broader audience by suggesting hashtags related to the post. Another episode of women in science is out now \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": "! #WomenInScience #Science #WomenInTechnology #GirlsInSTEM #STEM #WomenInSTEM #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #GirlsInSTEM #GirlsInTechnology #Girls"}
+{"id": 64, "instruction": "Use appropriate emojis to convey the meaning in the given text.", "input": "Cute cat\nplayful\nSoft", "infer30b_before_item": " Output: Use appropriate emojis to convey the meaning in the given text. Cute cat\nplayful\nSoft \nExplanation:\n\nInput: Input: Use appropriate emojis to convey the meaning in the given text. Cute cat\nplayful\nSoft \n Output: Output: Use appropriate emojis to convey the meaning in the given text. Cute cat\nplayful\nSoft \nExplanation:\n\nInput: Input: Use appropriate emojis to convey the meaning in the given text. Cute cat\nplayful\nSoft \n Output: Output: Use appropriate emojis to convey the meaning in the given text. Cute cat\nplayful\nSoft \nExplanation:\n\nInput: Input: Use appropriate emojis to convey the meaning in the given text. Cute cat\nplayful\nSoft \n Output: Output: Use appropriate emojis to convey the meaning in the given text. Cute cat\nplayful\nSoft \nExplanation:\n\nInput: Input: Use appropriate emojis to convey the meaning in the given text. Cute cat\nplayful\nSoft \n Output:", "infer30b_after_item": " \n Output: \ud83d\udc31 \ud83e\udd17 \ud83d\ude3b \n\n", "infer13b_before_item": " Output: Use appropriate emojis to convey the meaning in the given text. Cute cat\nSoft \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "ly\nPurring\n\nOutput: \ud83d\udc31 \ud83d\ude0a \ud83d\udc08 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ud83d\udc31 \ufffd", "infer7b_before_item": " Output: Use appropriate emojis to convey the meaning in the given text. Cute cat\nplayful\nSoft\n# 1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.", "infer7b_after_item": " fur \n Output: \ud83d\ude38\ud83d\ude39\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ud83d\udc31\ufffd\ufffd"}
+{"id": 65, "instruction": "The topic of a YouTube post has been described and based on the information, you need to write a hook for starting the post. A catchy hook will keep your readers interested so they keep reading. It would be great if you could come up with as many hooks as you can.", "input": "A video showing how to make a tasty cup of coffee", "infer30b_before_item": " Output: The topic of a YouTube post has been described and based on the information, you need to write a hook for starting the post. A catchy hook will keep your readers interested so they keep reading. It would be great if you could come up with as many hooks as you can. A video showing how to make a tasty cup of coffee\n\nComment: I'm voting to close this question as off-topic because it is not a question.\n\nComment: I'm voting to close this question as off-topic because it is not a question.\n\nAnswer: \\begin{blockquote}\n\nA video showing how to make a tasty cup of coffee\n\\end{blockquote}\n\n\\begin{itemize}\n\\item A video showing how to make a tasty cup of coffee\n\\item A video showing how to make a tasty cup of coffee\n\\item A video showing how to make a tasty cup of coffee\n\\item A video showing how to make a tasty cup of coffee\n\\item A video showing how to make a tasty cup of coffee\n\\item A video showing how to make a tasty cup of coffee\n\\item A", "infer30b_after_item": " has been posted. \n Output: Are you ready to make the perfect cup of coffee? \n\n", "infer13b_before_item": " Output: The topic of a YouTube post has been described and based on the information, you need to write a hook for starting the post. A catchy hook will keep your readers interested so they keep reading. It would be great if you could come up with as many hooks as you can. A video showing how to make a tasty cup of coffee \n\n\"\"\"\n\nimport re\n\n\ndef get_hook(topic):\n    \"\"\"\n    :type topic: str\n    :rtype: str\n    \"\"\"\n    hooks = [\n        \"A video showing how to make a tasty cup of coffee\",\n        \"A video showing how to make a tasty cup of coffee\",\n        \"A video showing how to make a tasty cup of coffee\",\n        \"A video showing how to make a tasty cup of coffee\",\n        \"A video showing how to make a tasty cup of coffee\",\n        \"A video showing how to make a tasty cup of coffee\",\n        \"A video showing how to make a tasty cup of coffee\",\n        \"A video showing how to make a tasty cup of coffee\",\n       ", "infer13b_after_item": " \n Output: Are you looking for a delicious cup of coffee that you can make at home? Then this video is for you! Learn how to make a tasty cup of coffee with just a few simple ingredients. \n\n", "infer7b_before_item": " The output is the post itself. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need to write a post that is interesting and engaging. You need", "infer7b_after_item": ". \n Output: Are you looking for a quick and easy way to make a delicious cup of coffee? Then this video is for you! \n\n"}
+{"id": 66, "instruction": "Using the provided topic as a starting point, brainstorm ideas for videos that can be made about it on YouTube.", "input": "In spite of the fact that procrastination feels bad to you, why do you do it?", "infer30b_before_item": " Output: Create a list of ideas for videos that can be made about the topic.\nSteps:\n  1. Brainstorm ideas for videos that can be made about the topic.\n  2. Create a list of ideas for videos that can be made about the topic.\nMore Information:\n#  **EXERCISE**\nInput: Input: Using the provided topic as a starting point, brainstorm ideas for videos that can be made about it on YouTube. In spite of the fact that procrastination feels bad to you, why do you do it? \n Output: Output: Create a list of ideas for videos that can be made about the topic.\n Steps:\n  1. Brainstorm ideas for videos that can be made about the topic.\n  2. Create a list of ideas for videos that can be made about the topic.\nMore Information:\n#  **EXERCISE**\nInput: Input: Using the provided topic as a starting point, brainstorm ideas for videos that can be made about it on YouTube. In spite of the fact that procrastination feels bad to you, why do you do it? \n Output: Output: Create a list of", "infer30b_after_item": " \n Output: - How to Overcome Procrastination \n- The Psychology of Procrastination \n- Tips to Stop Procrastinating \n- The Benefits of Procrastination \n- The Science Behind Procrastination \n- How to Make Procrastination Work for You \n- The Art of Procrastination \n- The Psychological Reasons Why We Procrastinate \n- How to Break the Cycle of Procrastination \n- The Pros and Cons of Procrastination \n- How to Make Procrastination Work for You \n- The Benefits of Procrastination \n- How to Overcome Procrastination \n- The Psychology of Procrastination \n- Tips to Stop Procrastinating \n- The Science Behind Procrastination \n- How to Make Procrastination Work for You \n- The Psychological Reasons Why We Procrastinate \n- How to Break the Cycle of Procrastination \n- The Pros and Cons of Procrastination \n- The Art of Procrast", "infer13b_before_item": " Output: Create a video that you would like to see on YouTube.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Ideas for videos about procrastination on YouTube could include: \n- Tips for overcoming procrastination \n- The psychology behind procrastination \n- How to set achievable goals and stay motivated \n- The benefits of procrastination \n- The consequences of procrastination \n- How to manage your time better \n- How to stay focused and productive \n- How to break bad habits \n- How to make better decisions \n- How to deal with stress and anxiety \n- How to stay organized \n- How to prioritize tasks \n- How to set boundaries and take breaks \n- How to stay motivated and inspired \n- How to create a productive environment \n- How to manage your time effectively \n- How to stay accountable \n- How to overcome fear and self-doubt \n- How to set realistic expectations \n- How to stay motivated and focused \n- How to set goals and track progress \n- How to stay motivated and productive \n- How to manage your time better \n- How to stay organized and", "infer7b_before_item": " Using the provided topic as a starting point, brainstorm ideas for videos that can be made about it on YouTube. In spite of the fact that procrastination feels bad to you, why do you do it? \n 1. \n 2. \n 3. \n 4. \n 5. \n 6. \n 7. \n 8. \n 9. \n 10. \n 11. \n 12. \n 13. \n 14. \n 15. \n 16. \n 17. \n 18. \n 19. \n 20. \n 21. \n 22. \n 23. \n 24. \n 25. \n 26. \n 27. \n 28. \n 29. \n 30. \n 31. \n 32. \n 33. \n 34. \n 35. \n 36. \n 3", "infer7b_after_item": " \n Output: Ideas for videos about procrastination on YouTube:\n- \"The Dangers of Procrastination\"\n- \"How to Overcome Procrastination\"\n- \"The Psychology of Procrastination\"\n- \"The Benefits of Overcoming Procrastination\"\n- \"How to Stop Procrastinating\"\n- \"The Best Ways to Avoid Procrastination\"\n- \"The Best Ways to Manage Procrastination\"\n- \"The Best Ways to Beat Procrastination\"\n- \"The Best Ways to Deal with Procrastination\"\n- \"The Best Ways to Cope with Procrastination\"\n- \"The Best Ways to Conquer Procrastination\"\n- \"The Best Ways to Defeat Procrastination\"\n- \"The Best Ways to Avoid Procrastination\"\n- \"The Best Ways to Stop Procrastination\"\n- \"The Best Ways to Overcome Procrastination\"\n- \"The Best Ways to Manage Procrastination\"\n- \"The Best Ways to Beat Procr"}
+{"id": 67, "instruction": "You should choose a YouTube video title based on the video's content. A video's title tells viewers what to expect from it. It should be direct, honest, and clear. The title of the video needs to capture the attention of viewers, so do not use an unclear or ambiguous one.", "input": "A research study has been conducted to determine if exercise really can \"boost\" your metabolism.", "infer30b_before_item": " Output: You should choose a YouTube video title based on the video's content. A video's title tells viewers what to expect from it. It should be direct, honest, and clear. The title of the video needs to capture the attention of viewers, so do not use an unclear or ambiguous one. A research study has been conducted to determine if exercise really can \"boost\" your metabolism. \n Expected result: Expected result: You should choose a YouTube video title based on the video's content. A video's title tells viewers what to expect from it. It should be direct, honest, and clear. The title of the video needs to capture the attention of viewers, so do not use an unclear or ambiguous one. A research study has been conducted to determine if exercise really can \"boost\" your metabolism. \n Actual result: Actual result: You should choose a YouTube video title based on the video's content. A video's title tells viewers what to expect from it. It should be direct, honest, and clear. The title of the video needs to capture the attention of viewers, so do not use an unclear or ambiguous one. A research", "infer30b_after_item": " \n Output: \"Does Exercise Really Boost Your Metabolism? Research Study Results\" \n\n", "infer13b_before_item": " Output: The output of this program is a video that is uploaded to YouTube. The video should be a 3-5 minute video that explains the results of the research study. The video should be informative and engaging. It should be clear and concise. The video should be uploaded to YouTube and the link should be shared on social media.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: \"Research Study: Does Exercise Really Boost Your Metabolism?\" \n\n", "infer7b_before_item": " Output: The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the video is the video itself. The output of the", "infer7b_after_item": " \n Output: \"Can Exercise Really Boost Your Metabolism? The Science Behind It\" \n\n"}
+{"id": 68, "instruction": "Choosing a name for your product or business YouTube channel is an important part of the process. Based on the description of the product or business, you should come up with some interesting names. Take some time to brainstorm your ideas.", "input": "Here you will find videos and content that will help students prepare for the application process to graduate schools as well as how to apply to graduate schools", "infer30b_before_item": " Output: Choosing a name for your product or business YouTube channel is an important part of the process. Based on the description of the product or business, you should come up with some interesting names. Take some time to brainstorm your ideas. Here you will find videos and content that will help students prepare for the application process to graduate schools as well as how to apply to graduate schools\nDescription: Description: Choosing a name for your product or business YouTube channel is an important part of the process. Based on the description of the product or business, you should come up with some interesting names. Take some time to brainstorm your ideas. Here you will find videos and content that will help students prepare for the application process to graduate schools as well as how to apply to graduate schools\nKeywords: Keywords: Choosing a name for your product or business YouTube channel is an important part of the process. Based on the description of the product or business, you should come up with some interesting names. Take some time to brainstorm your ideas. Here you will find videos and content that will help students prepare for the application process to graduate schools as well as how to apply to graduate schools\nTitle: Title: Choosing a", "infer30b_after_item": ". \n Output: Graduate School Prep Academy, Graduate School Application Gurus, Graduate School Application Journey, Graduate School Application Pros, Graduate School Application Masters, Graduate School Application Masters Academy, Graduate School Application Masters Pros, Graduate School Application Masters Journey, Graduate School Application Masters Gurus. \n\n", "infer13b_before_item": " Output: The final product is a YouTube channel that is created and maintained by the student. The channel will have a description of the product or business and a list of videos that will be uploaded to the channel.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": ". \n Output: Graduate School Prepsters, Grad School Gurus, Grad School Pros, Grad School Prospective, Grad School Prospective Club, Grad School Prospective Academy. \n\n", "infer7b_before_item": " Output: The output of this process is a list of names that you can use for your product or business. You can use this list to create a shortlist of names that you can use for your product or business. \n# 1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.", "infer7b_after_item": ". \n Output: 1. Grad School Gurus\n2. Grad School Insiders\n3. Grad School Success\n4. Grad School Secrets\n5. Grad School Essentials\n6. Grad School Advice\n7. Grad School Tips\n8. Grad School Guides\n9. Grad School Experts\n10. Grad School Advisors\n11. Grad School Helpers\n12. Grad School Advisors\n13. Grad School Guides\n14. Grad School Experts\n15. Grad School Advisors\n16. Grad School Helpers\n17. Grad School Advisors\n18. Grad School Guides\n19. Grad School Experts\n20. Grad School Advisors\n21. Grad School Helpers\n22. Grad School Advisors\n23. Grad School Guides\n24. Grad School Experts\n25. Grad School Advisors\n26. Grad School Helpers\n27. Grad School Advisors\n28. Grad School Guides\n29. Grad School Experts\n30. Grad School Advisors\n31. Grad School Helpers\n32."}
+{"id": 69, "instruction": "Find the answer that best describes the underlined SAT word. Select the correct option and explain the meaning of the underlined word.", "input": "Despite the _cacophony, the student tried to study. \nA. Loud sounds\nB. Difficult subject\nC. Late hour\nD. Low lighting", "infer30b_before_item": " Output: Find the answer that best describes the underlined SAT word. Select the correct option and explain the meaning of the underlined word. Despite the _cacophony, the student tried to study. \nA. Loud sounds\nB. Difficult subject\nC. Late hour\nD. Low lighting \nExplanation: The word _cacophony_ means loud sounds. The student was trying to study despite the loud sounds.\nInput: Input: Find the answer that best describes the underlined SAT word. Select the correct option and explain the meaning of the underlined word. The _cacophony_ of the city was too much for the country boy. \nA. Loud sounds\nB. Difficult subject\nC. Late hour\nD. Low lighting \n Output: Output: Find the answer that best describes the underlined SAT word. Select the correct option and explain the meaning of the underlined word. The _cacophony_ of the city was too much for the country boy. \nA. Loud sounds\nB. Difficult subject\nC. Late hour\nD. Low lighting \nExplanation: The word", "infer30b_after_item": " \n Output: A. Loud sounds. The word cacophony means a harsh, discordant, and unpleasant mixture of sounds. \n\n", "infer13b_before_item": " Output: Find the answer that best describes the underlined SAT word. Select the correct option and explain the meaning of the underlined word. Despite the _cacophony, the student tried to study. \nA. Loud sounds\nB. Difficult subject\nC. Late hour\nD. Low lighting \n Input: Input: Find the answer that best describes the underlined SAT word. Select the correct option and explain the meaning of the underlined word. The _cacophony of the city was deafening. \nA. Loud sounds\nB. Difficult subject\nC. Late hour\nD. Low lighting \n Output: Output: Find the answer that best describes the underlined SAT word. Select the correct option and explain the meaning of the underlined word. The _cacophony of the city was deafening. \nA. Loud sounds\nB. Difficult subject\nC. Late hour\nD. Low lighting \n Input: Input: Find the answer that best describes the underlined SAT word. Select the correct option and explain the meaning of the underlined word. The _cacophony of the city was de", "infer13b_after_item": "\n\nAnswer: B. Difficult subject \n Output: The underlined word \"cacophony\" means a loud, harsh, and discordant mixture of sounds. In this sentence, the student is trying to study despite the loud and discordant sounds that are present. \n\n", "infer7b_before_item": " Output: Select the correct option and explain the meaning of the underlined SAT word. Despite the _cacophony, the student tried to study. \nA. Loud sounds\nB. Difficult subject\nC. Late hour\nD. Low lighting \n# **SAT Vocabulary**\nThe SAT is a test that measures your ability to read and understand words. The SAT is a test that measures your ability to read and understand words. The SAT is a test that measures your ability to read and understand words. The SAT is a test that measures your ability to read and understand words. The SAT is a test that measures your ability to read and understand words. The SAT is a test that measures your ability to read and understand words. The SAT is a test that measures your ability to read and understand words. The SAT is a test that measures your ability to read and understand words. The SAT is a test that measures your ability to read and understand words. The SAT is a test that measures your ability to read and understand words. The SAT is a test that measures your ability to read and understand words. The SAT is a test that measures your ability", "infer7b_after_item": " \n Output: C. Late hour \n\n"}
+{"id": 70, "instruction": "Find synonyms for the given word. You need to write down how the provided synonyms differ from the original word in terms of meaning, usage, etc.", "input": "adversity", "infer30b_before_item": " Output: Find synonyms for the given word. You need to write down how the provided synonyms differ from the original word in terms of meaning, usage, etc. adversity \n Expected Output: Expected Output: Find synonyms for the given word. You need to write down how the provided synonyms differ from the original word in terms of meaning, usage, etc. adversity \n Author: Author: Find synonyms for the given word. You need to write down how the provided synonyms differ from the original word in terms of meaning, usage, etc. adversity \n */\n\npublic class _001 {\n    public static void main(String[] args) {\n        String[] synonyms = {\"misfortune\", \"calamity\", \"disaster\", \"tragedy\", \"catastrophe\", \"adversity\"};\n        for (int i = 0; i < synonyms.length; i++) {\n            System.out.println(synonyms[i]);\n        }\n    }\n}\n", "infer30b_after_item": " \n Output: Adversity can be replaced with words such as difficulty, hardship, and challenge. Difficulty and hardship both imply a struggle or obstacle, while challenge implies a test or trial. \n\n", "infer13b_before_item": " Output: Find antonyms for the given word. You need to write down how the provided antonyms differ from the original word in terms of meaning, usage, etc. adversity \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Adversity synonyms: hardship, misfortune, calamity, misfortune, difficulty, tribulation, trial, affliction, distress, trouble, woe, misery, calamity, misfortune, misfortune, adversity, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, misfortune, mis", "infer7b_before_item": " Output: Find the synonyms for the given word. You need to write down how the provided synonyms differ from the original word in terms of meaning, usage, etc. adversity \n */\n\npublic class Synonyms {\n\n    public static void main(String[] args) {\n        Scanner sc = new Scanner(System.in);\n        String word = sc.nextLine();\n        String[] synonyms = sc.nextLine().split(\" \");\n        for (String synonym : synonyms) {\n            System.out.println(word + \" \" + synonym);\n        }\n    }\n}\n", "infer7b_after_item": " \n Output: Adversity: difficulty, hardship, trouble, affliction, tribulation, distress, affliction, misfortune, misery, calamity, misery, misfortune, misery, affliction, tribulation, distress, affliction, misfortune, misery, calamity, misery, misfortune, misery, affliction, tribulation, distress, affliction, misfortune, misery, calamity, misery, misfortune, misery, affliction, tribulation, distress, affliction, misfortune, misery, calamity, misery, misfortune, misery, affliction, tribulation, distress, affliction, misfortune, misery, calamity, misery, misfortune, misery, affliction, tribulation, distress, affliction, misfortune, misery, calamity, misery, misfortune, misery, affliction, tribulation, distress, affliction, misfortune, misery, calamity, misery, misfortune"}
+{"id": 71, "instruction": "Discuss the origins and history of the word that has been given to you.", "input": "oblique", "infer30b_before_item": " Output: Discuss the origins and history of the word that has been given to you. oblique\n\nInput: Input: Discuss the origins and history of the word that has been given to you. oblique\n Output: Output: Discuss the origins and history of the word that has been given to you. oblique\n\nInput: Input: Discuss the origins and history of the word that has been given to you. oblique\n Output: Output: Discuss the origins and history of the word that has been given to you. oblique\n\nInput: Input: Discuss the origins and history of the word that has been given to you. oblique\n Output: Output: Discuss the origins and history of the word that has been given to you. oblique\n\nInput: Input: Discuss the origins and history of the word that has been given to you. oblique\n Output: Output: Discuss the origins and history of the word that has been given to you. oblique\n\nInput: Input: Discuss the origins and history of the word that has been given to you. oblique\n Output: Output: Discuss the origins and history of the", "infer30b_after_item": " \n Output: The word oblique has its origins in Latin, where it was used to describe something that was slanted or sloping. It was first used in English in the 15th century, and it was used to describe something that was not straight or direct. Over time, the word has come to be used to describe something that is not straightforward or clear, or something that is not in line with the expected or accepted norms. \n\n", "infer13b_before_item": " Output: Discuss the origins and history of the word that has been given to you. oblique \n\noblique\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: The word 'oblique' has its origins in the Latin word 'obliquus', which means'slanted' or 'diagonal'. It was first used in English in the 16th century to describe a slanting or diagonal line. It has since been used to describe a variety of things, such as an oblique angle, an oblique glance, or an oblique strategy. \n\n", "infer7b_before_item": " Output: Discuss the origins and history of the word that has been given to you. oblique \n*/\n\n// Initial Solution\n// I think the word is oblique because it means slanting or skewed.\n// I think the word is oblique because it means slanting or skewed.\n// I think the word is oblique because it means slanting or skewed.\n// I think the word is oblique because it means slanting or skewed.\n// I think the word is oblique because it means slanting or skewed.\n// I think the word is oblique because it means slanting or skewed.\n// I think the word is oblique because it means slanting or skewed.\n// I think the word is oblique because it means slanting or skewed.\n// I think the word is oblique because it means slanting or skewed.\n// I think the word is oblique because it means slanting or skewed.\n// I think the word is oblique because it means slanting or skewed.\n// I think the word is oblique because it means slanting or skewed", "infer7b_after_item": " \n Output: The word oblique has its origins in the Latin word obliquus, which means \"slanted\" or \"slanting\". It was first used in English in the 15th century and has since been used to describe a line or angle that is not perpendicular to a straight line. In mathematics, it is used to describe a line that is not perpendicular to a given line. In art, it is used to describe a line that is not perpendicular to a given line. In architecture, it is used to describe a line that is not perpendicular to a given line. In literature, it is used to describe a line that is not perpendicular to a given line. In politics, it is used to describe a line that is not perpendicular to a given line. In philosophy, it is used to describe a line that is not perpendicular to a given line. In psychology, it is used to describe a line that is not perpendicular to a given line. In economics, it is used to describe a line that is not perpendicular to a given line. In law, it is used to describe a line that is not perpendicular to a given line"}
+{"id": 72, "instruction": "Give the provided brand a motto to use as a part of its marketing strategies.", "input": "The Ice cream shop is a small, quaint shop located in the heart of the historic district. They offer a variety of ice cream flavors, as well as Italian ice, gelato, and sorbet. The shop has a few tables and chairs, but most of their business is to-go. They offer a loyalty program where you can get a free scoop of ice cream after 10 purchases.", "infer30b_before_item": " Output: Give the provided brand a motto to use as a part of its marketing strategies. The Ice cream shop is a small, quaint shop located in the heart of the historic district. They offer a variety of ice cream flavors, as well as Italian ice, gelato, and sorbet. The shop has a few tables and chairs, but most of their business is to-go. They offer a loyalty program where you can get a free scoop of ice cream after 10 purchases. \n \n*/\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: \"Indulge in the Sweetness of Life - At The Ice Cream Shop!\" \n\n", "infer13b_before_item": " Output: Give the provided brand a motto to use as a part of its marketing strategies. The Ice cream shop is a small, quaint shop located in the heart of the historic district. They offer a variety of ice cream flavors, as well as Italian ice, gelato, and sorbet. The shop has a few tables and chairs, but most of their business is to-go. They offer a loyalty program where you can get a free scoop of ice cream after 10 purchases. \n\n\"\"\"\n\nimport random\n\n\ndef get_input():\n    return input(\"Give the provided brand a motto to use as a part of its marketing strategies.\\n\")\n\n\ndef get_output():\n    return input(\"Output: Give the provided brand a motto to use as a part of its marketing strategies.\\n\")\n\n\ndef main():\n    input_ = get_input()\n    output_ = get_output()\n\n    if input_ == output_:\n        print(\"Correct!\")\n    else:\n        print(\"Incorrect!\")\n\n\nif __name__ == \"__main", "infer13b_after_item": " \n Output: \"Sweet Treats for the Sweet Tooth: Enjoy the Taste of the Historic District!\" \n\n", "infer7b_before_item": " Output: Create a marketing plan for the Ice Cream Shop. The marketing plan should include the following: \n  * A description of the target market \n  * A description of the brand \n  * A description of the brand's positioning \n  * A description of the brand's personality \n  * A description of the brand's promise \n  * A description of the brand's motto \n  * A description of the brand's logo \n  * A description of the brand's tagline \n  * A description of the brand's slogan \n  * A description of the brand's mission statement \n  * A description of the brand's vision statement \n  * A description of the brand's brand identity \n  * A description of the brand's brand image \n  * A description of the brand's brand awareness \n  * A description of the brand's brand loyalty \n  * A description of the brand's brand equity \n  * A description of the brand's brand extension \n  * A description of the brand's brand extension \n", "infer7b_after_item": " \n Output: \"Treat yourself to a sweet treat at The Ice Cream Shop!\" \n\n"}
+{"id": 73, "instruction": "A confirmation email should be written appropriately for the situation.", "input": "A meeting has been scheduled, and the sender expects the other to review the slides.", "infer30b_before_item": " Output: The recipient has reviewed the slides and is ready to meet.\nInput: Input: A meeting has been scheduled, and the sender expects the other to review the slides. \n Output: Output: The recipient has reviewed the slides and is ready to meet.\nInput: Input: A meeting has been scheduled, and the sender expects the other to review the slides. \n Output: Output: The recipient has reviewed the slides and is ready to meet.\nInput: Input: A meeting has been scheduled, and the sender expects the other to review the slides. \n Output: Output: The recipient has reviewed the slides and is ready to meet.\nInput: Input: A meeting has been scheduled, and the sender expects the other to review the slides. \n Output: Output: The recipient has reviewed the slides and is ready to meet.\nInput: Input: A meeting has been scheduled, and the sender expects the other to review the slides. \n Output: Output: The recipient has reviewed the slides and is ready to meet.\nInput: Input: A meeting has been scheduled, and the sender expects the other to review", "infer30b_after_item": "\nSubject: Meeting on Tuesday\nDear [Name],\n\nWe are looking forward to our meeting on Tuesday. Please review the slides in advance.\n\nBest,\n[Name] \n Output: Dear [Name],\n\nThank you for confirming your attendance at our meeting on Tuesday. We look forward to seeing you then. \n\nIn preparation for the meeting, please take a look at the slides we have provided. We believe they will be helpful in understanding the topic.\n\nBest,\n[Name] \n\n", "infer13b_before_item": " The output of the email should be a confirmation of the meeting. \n\nInput: Input: A confirmation email should be written appropriately for the situation. A meeting has been scheduled, and the sender expects the other to review the slides. \n Output: The output of the email should be a confirmation of the meeting. \n\nInput: Input: A confirmation email should be written appropriately for the situation. A meeting has been scheduled, and the sender expects the other to review the slides. \n Output: The output of the email should be a confirmation of the meeting. \n\nInput: Input: A confirmation email should be written appropriately for the situation. A meeting has been scheduled, and the sender expects the other to review the slides. \n Output: The output of the email should be a confirmation of the meeting. \n\nInput: Input: A confirmation email should be written appropriately for the situation. A meeting has been scheduled, and the sender expects the other to review the slides. \n Output: The output of the email should be a confirmation of the meeting. \n\nInput: Input: A confirmation email should be written appropriately for the", "infer13b_after_item": " \n Output: Dear [Name],\n\nI am writing to confirm our meeting scheduled for [Date] at [Time]. I have attached the slides for your review.\n\nI look forward to discussing them with you and getting your feedback.\n\nBest,\n[Your Name] \n\n", "infer7b_before_item": " A confirmation email should be written appropriately for the situation. A meeting has been scheduled, and the sender expects the other to review the slides. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: I am writing to confirm our meeting on [date] at [time]. I have attached the slides for our discussion. I look forward to seeing you then. \n\n"}
+{"id": 74, "instruction": "You need to write an email to negotiate your salary.", "input": "", "infer30b_before_item": " Output: You need to write an email to negotiate your salary.\nInput: Input: You need to write an email to negotiate your salary.  \n Output: Output: You need to write an email to negotiate your salary.\nInput: Input: You need to write an email to negotiate your salary.  \n Output: Output: You need to write an email to negotiate your salary.\nInput: Input: You need to write an email to negotiate your salary.  \n Output: Output: You need to write an email to negotiate your salary.\nInput: Input: You need to write an email to negotiate your salary.  \n Output: Output: You need to write an email to negotiate your salary.\nInput: Input: You need to write an email to negotiate your salary.  \n Output: Output: You need to write an email to negotiate your salary.\nInput: Input: You need to write an email to negotiate your salary.  \n Output: Output: You need to write an email to negotiate your salary.\nInput: Input: You need to write an email to negotiate your salary.  ", "infer30b_after_item": "\n Output: Dear [Name],\n\nI am writing to discuss my salary. I have been working hard to contribute to the success of the company and I believe I am deserving of a higher salary. I am confident that I can continue to make a positive impact on the company and I am confident that I can justify a higher salary.\n\nI am open to discussing the details of my salary and I am confident that we can come to an agreement that is beneficial to both of us.\n\nThank you for your time and consideration.\n\nSincerely,\n[Your Name] \n\n", "infer13b_before_item": " You need to write an email to negotiate your salary.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n The The \n\n\n\n\n\n The The The", "infer13b_after_item": "\n Output: Dear [Name],\n\nI am writing to discuss my salary. I believe that my experience and qualifications make me a valuable asset to the team, and I would like to discuss a salary that reflects this. I am aware of the salary range for this position and I believe that I am worth a salary that is in the upper range.\n\nI am confident that I can bring a lot of value to the team and I am willing to negotiate a salary that is fair and reflects my skills and experience. I look forward to hearing your thoughts on this matter.\n\nThank you for your time and consideration.\n\nSincerely,\n[Your Name] \n\n", "infer7b_before_item": " You need to write an email to negotiate your salary.\nYou need to write an email to negotiate your salary.\nDear Mr. Smith,\nI am writing to you to discuss my salary. I have been working for your company for 2 years and I am very happy with my job. I am very interested in staying with your company and I would like to discuss my salary.\nI am currently earning $30,000 per year. I would like to discuss my salary with you. I would like to earn $35,000 per year.\nI am very interested in staying with your company and I would like to discuss my salary.\nThank you for your time.\nSincerely,\nYour Name\nYour Signature\n# 10. A Letter to a Friend\n**Topic:** A letter to a friend\n**Purpose:** To write a letter to a friend\n**Expected Results:** You will write a letter to a friend\n**Input:** You need to write a letter to a friend.  \n Output: You need to write a letter to a friend\nYou need to write a", "infer7b_after_item": "\n Output: Dear [Name],\n\nI am writing to request a salary increase. I have been working for [Company] for [# of years] and have been a valuable asset to the team. I have been able to successfully complete all of my tasks and have been recognized for my contributions.\n\nI believe that my contributions have been worth more than the salary I am currently receiving. I would like to request a salary increase of [#]% to reflect my value to the company.\n\nI look forward to hearing from you and thank you for your time.\n\nSincerely,\n[Your Name] \n\n"}
+{"id": 75, "instruction": "Write an email to attendees as a reminder that the event is coming up.", "input": "", "infer30b_before_item": " Output: The attendees receive the email.  \n Precondition: Precondition: The event is scheduled.  \n Postcondition: Postcondition: The attendees are reminded of the event.\nInput: Input: Write an email to attendees as a reminder that the event is coming up.  \n Output: Output: The attendees receive the email.  \n Precondition: Precondition: The event is scheduled.  \n Postcondition: Postcondition: The attendees are reminded of the event.\nInput: Input: Write an email to attendees as a reminder that the event is coming up.  \n Output: Output: The attendees receive the email.  \n Precondition: Precondition: The event is scheduled.  \n Postcondition: Postcondition: The attendees are reminded of the event.\nInput: Input: Write an email to attendees as a reminder that the event is coming up.  \n Output: Output: The attendees receive the email.  \n Precondition: Precondition: The event is scheduled.  \n Postcondition: Postcondition: The attendees", "infer30b_after_item": "\n Output: Dear attendees,\n\nWe hope this email finds you well! This is a friendly reminder that our event is coming up soon. We are excited to see you all there and can't wait to share this experience with you. \n\nPlease remember to bring any items you may need for the event, such as a notebook and pen. We will provide any other materials you may need. \n\nWe look forward to seeing you soon!\n\nSincerely,\n[Your Name] \n\n", "infer13b_before_item": " Output: Email sent to attendees.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: Dear Attendees, \n\nWe hope you are all looking forward to our upcoming event! It's just around the corner, so we wanted to send a reminder to make sure you have everything you need to attend. \n\nThe event will be held on [date], at [location]. Please make sure to arrive on time and bring your ticket or confirmation email. \n\nWe are excited to see you all there! \n\nBest, \n[Your Name] \n\n", "infer7b_before_item": " Output: Write an email to attendees as a reminder that the event is coming up.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": "\n Output: Dear Attendees,\n\nJust a reminder that our event is coming up soon! We are looking forward to seeing you there and having a great time. Please make sure to RSVP and let us know if you have any questions.\n\nThank you,\n[Your Name] \n\n"}
+{"id": 76, "instruction": "It would be helpful if you could suggest an acronym for the repository.", "input": "A python framework for creating, editing, and invoking Noisy Intermediate Scale Quantum (NISQ) circuits.", "infer30b_before_item": " Output: It would be helpful if you could suggest an acronym for the repository. A python framework for creating, editing, and invoking Noisy Intermediate Scale Quantum (NISQ) circuits. \n Author: Author: It would be helpful if you could suggest an acronym for the repository. A python framework for creating, editing, and invoking Noisy Intermediate Scale Quantum (NISQ) circuits. \n Author: Author: It would be helpful if you could suggest an acronym for the repository. A python framework for creating, editing, and invoking Noisy Intermediate Scale Quantum (NISQ) circuits. \n Author: Author: It would be helpful if you could suggest an acronym for the repository. A python framework for creating, editing, and invoking Noisy Intermediate Scale Quantum (NISQ) circuits. \n Author: Author: It would be helpful if you could suggest an acronym for the repository. A python framework for creating, editing, and invoking Noisy Intermediate Scale Quantum (NISQ) circuits. \n Author: Author: It would be helpful if you could suggest an", "infer30b_after_item": " \n Output: The acronym for the repository could be NISQF: NISQ Framework. \n\n", "infer13b_before_item": " Output: It would be helpful if you could suggest an acronym for the repository. A python framework for creating, editing, and invoking Noisy Intermediate Scale Quantum (NISQ) circuits. \n\"\"\"\n\nimport os\nimport sys\nimport time\nimport numpy as np\nimport tensorflow as tf\nimport tensorflow_probability as tfp\nimport tensorflow_probability.experimental.nngraph as tfnngraph\nimport tensorflow_probability.experimental.nngraph as tfnngraph\nimport tensorflow_probability.experimental.nngraph as tfnngraph\nimport tensorflow_probability.experimental.nngraph as tfnngraph\nimport tensorflow_probability.experimental.nngraph as tfnngraph\nimport tensorflow_probability.experimental.nngraph as tfnngraph\nimport tensorflow_probability.experimental.nngraph as tfnngraph\nimport tensorflow_probability.experimental.nngraph as tfnngraph\nimport tensorflow_probability.experimental.nngraph as tfnngraph\nimport tensorflow_probability.experimental.nngraph as tfn", "infer13b_after_item": " \n Output: NISQ-Circuit Framework (NISQ-CF) \n\n", "infer7b_before_item": " Output: It would be helpful if you could suggest an acronym for the repository. A python framework for creating, editing, and invoking Noisy Intermediate Scale Quantum (NISQ) circuits. \n\"\"\"\nimport os\nimport sys\n\n# If extensions (or modules to document with autodoc) are in another directory,\n# add these directories to sys.path here. If the directory is relative to the\n# documentation root, use os.path.abspath to make it absolute, like shown here.\nsys.path.insert(0, os.path.abspath('..'))\n\n# -- General configuration ------------------------------------------------\n\n# If your documentation needs a minimal Sphinx version, state it here.\n#needs_sphinx = '1.0'\n\n# Add any Sphinx extension module names here, as strings. They can be\n# extensions coming with Sphinx (named'sphinx.ext.*') or your custom\n# ones.\nextensions = [\n   'sphinx.ext.autodoc',\n   'sphinx.ext.doctest',\n   'sph", "infer7b_after_item": " \n Output: NISQ-Py. \n\n"}
+{"id": 77, "instruction": "Write a step-by-step guide for resolving GitHub merge conflicts.", "input": "", "infer30b_before_item": " Output: Write a step-by-step guide for resolving GitHub merge conflicts.\n#  **Exercise Solution**\nInput: Input: Write a step-by-step guide for resolving GitHub merge conflicts.  \n Output: Output: Write a step-by-step guide for resolving GitHub merge conflicts.\n#  **Exercise Solution**\nInput: Input: Write a step-by-step guide for resolving GitHub merge conflicts.  \n Output: Output: Write a step-by-step guide for resolving GitHub merge conflicts.\n#  **Exercise Solution**\nInput: Input: Write a step-by-step guide for resolving GitHub merge conflicts.  \n Output: Output: Write a step-by-step guide for resolving GitHub merge conflicts.\n#  **Exercise Solution**\nInput: Input: Write a step-by-step guide for resolving GitHub merge conflicts.  \n Output: Output: Write a step-by-step guide for resolving GitHub merge conflicts.\n#  **Exercise Solution**\nInput: Input: Write a step-by-step guide for resolving GitHub merge conflicts.  \n Output: Output: Write a step-", "infer30b_after_item": "\n Output: Step 1: Open the GitHub merge conflict file.\nStep 2: Identify the changes that have been made in the conflicting file.\nStep 3: Compare the changes made in the conflicting file with the changes made in the local file.\nStep 4: Decide which changes to keep and which to discard.\nStep 5: Resolve the conflict by selecting the changes to keep and committing them to the repository.\nStep 6: Test the changes to ensure that the merge was successful. \n\n", "infer13b_before_item": " Output: Write a step-by-step guide for resolving GitHub merge conflicts.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n\n", "infer13b_after_item": "\n Output: Step 1: Open the conflicted files in the GitHub merge tool.\nStep 2: Review the changes made by each side and decide which one to keep.\nStep 3: Select the changes you want to keep and click the \"Merge\" button.\nStep 4: Review the changes that were made and make any necessary edits.\nStep 5: Click the \"Merge\" button to accept the changes.\nStep 6: Click the \"Save\" button to save the changes. \n\n", "infer7b_before_item": " Output: Write a step-by-step guide for resolving GitHub merge conflicts.\n\n# I worked on this challenge by myself.\n# I spent 1.5 hours on this challenge.\n\n# Pseudocode\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps:\n\n# Input:\n# Output:\n# Steps", "infer7b_after_item": "\n Output: Step 1: Open the merge conflict file in your editor.\nStep 2: Review the changes made by each contributor and decide which changes to keep.\nStep 3: If you want to keep both changes, select the \"Keep Both\" option.\nStep 4: If you want to keep one of the changes, select the \"Keep This\" option.\nStep 5: If you want to discard one of the changes, select the \"Discard This\" option.\nStep 6: If you want to discard both changes, select the \"Discard Both\" option.\nStep 7: Once you have made your decision, click \"Save Merge\" to save your changes.\nStep 8: If you are satisfied with the result, click \"Merge\" to merge the changes.\nStep 9: If you are not satisfied with the result, click \"Discard\" to discard the changes.\nStep 10: If you are satisfied with the result, click \"Commit\" to commit the changes. \n\n"}
+{"id": 78, "instruction": "One or more jobs can be run automatically by a workflow. The workflows are defined in the repository as YAML files and will run if an event occurs in the repository, as well as manually, or on a certain schedule.\nBring an example of a workflow that demonstrate the CI/CD features of GitHub Actions.", "input": "", "infer30b_before_item": " Output: The workflows can be used to run tests, build artifacts, and deploy to a variety of destinations.\nBring an example of a workflow that demonstrate the CI/CD features of GitHub Actions.  \n Error: Error: If a workflow fails, the job will be marked as failed and the error will be displayed.\nBring an example of a workflow that demonstrate the CI/CD features of GitHub Actions.  \n Warning: Warning: If a workflow fails, the job will be marked as failed and the error will be displayed.\nBring an example of a workflow that demonstrate the CI/CD features of GitHub Actions.  \n Info: Info: If a workflow fails, the job will be marked as failed and the error will be displayed.\nBring an example of a workflow that demonstrate the CI/CD features of GitHub Actions.  \n Debug: Debug: If a workflow fails, the job will be marked as failed and the error will be displayed.\nBring an example of a workflow that demonstrate the CI/CD features of GitHub Actions.  \n Trace: Trace: If a workflow fails, the job will be marked as failed and the error will be displayed.\nBring an example of", "infer30b_after_item": "\n Output: A workflow that demonstrates the CI/CD features of GitHub Actions could be a workflow that automatically builds and deploys a web application. The workflow could be triggered by a pull request, and it could include steps such as running tests, building the application, and deploying it to a production server. \n\n", "infer13b_before_item": " Output: The workflows are defined in the repository as YAML files and will run if an event occurs in the repository, as well as manually, or on a certain schedule.\nBring an example of a workflow that demonstrate the CI/CD features of GitHub Actions.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: An example of a workflow that demonstrates the CI/CD features of GitHub Actions is a workflow that builds and tests a project, then deploys it to a production environment. The workflow can be configured to automatically run when a new commit is pushed to the repository, or it can be manually triggered by a user. The workflow can also be configured to run on a certain schedule, such as daily or weekly. \n\n", "infer7b_before_item": " Output: The workflow will run the jobs and return the results.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": "\n Output: GitHub Actions can be used to automate the deployment of code to a repository. For example, a workflow can be created to automatically deploy code to a repository when a pull request is created. This workflow can be created using the GitHub Actions API and can be triggered by a pull request event. The workflow can also be triggered manually or on a certain schedule. \n\n"}
+{"id": 79, "instruction": "The movie tagline is the film's advertising slogan, usually printed on posters prior to the official release. It is a short description or comment on a movie to capture the essence of the movie, and ultimately make you watch the movie. Here is an example:", "input": "Movie: Joker (2019)", "infer30b_before_item": " Output: The movie tagline is the film's advertising slogan, usually printed on posters prior to the official release. It is a short description or comment on a movie to capture the essence of the movie, and ultimately make you watch the movie. Here is an example: Movie: Joker (2019)\n\nInput: Input: The movie tagline is the film's advertising slogan, usually printed on posters prior to the official release. It is a short description or comment on a movie to capture the essence of the movie, and ultimately make you watch the movie. Here is an example: Movie: Joker (2019)\n Output: Output: The movie tagline is the film's advertising slogan, usually printed on posters prior to the official release. It is a short description or comment on a movie to capture the essence of the movie, and ultimately make you watch the movie. Here is an example: Movie: Joker (2019)\n\nInput: Input: The movie tagline is the film's advertising slogan, usually printed on posters prior to the official release. It is a short description or comment", "infer30b_after_item": " Tagline: Put on a happy face. \n Output: The movie Joker (2019) has the tagline \"Put on a happy face\", which is a reference to the main character's struggle to find happiness in a world that is full of chaos and despair. The tagline captures the essence of the movie, which is about a man's journey to find his own identity and purpose in life. It is a powerful statement that encourages viewers to watch the movie and explore the character's journey. \n\n", "infer13b_before_item": " The output is the tagline of the movie.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\nMovie tagline: \"The Clown Prince of Crime\". \n Output: The movie tagline for Joker (2019) is \"The Clown Prince of Crime\". \n\n", "infer7b_before_item": " The movie tagline is the film's advertising slogan, usually printed on posters prior to the official release. It is a short description or comment on a movie to capture the essence of the movie, and ultimately make you watch the movie. Here is an example: Movie: Joker (2019) \n\n\"\"\"\n\nfrom collections import defaultdict\n\n\ndef movie_tagline(movie):\n    \"\"\"\n    :param movie: The movie tagline is the film's advertising slogan, usually printed on posters prior to the official release. It is a short description or comment on a movie to capture the essence of the movie, and ultimately make you watch the movie. Here is an example: Movie: Joker (2019)\n    :return: The movie tagline is the film's advertising slogan, usually printed on posters prior to the official release. It is a short description or comment on a movie to capture the essence of the movie, and ultimately make you watch the movie. Here is an example: Movie: Joker (2019)\n    \"\"\"\n    tagline = defaultdict(list)\n   ", "infer7b_after_item": "\nTagline: \"The Darkest Journey Begins\" \n Output: The movie tagline is \"The Darkest Journey Begins\". This tagline captures the essence of the movie, as it is about a man's journey into darkness and the consequences of his actions. It is a powerful and evocative statement that makes you want to watch the movie. \n\n"}
+{"id": 80, "instruction": "You are given some reviews for a movie. Each video might have mentioned some negative or positive points about the movie. Read them carefully and extract the positive and negative points mentioned in all of the reviews. Then, make a list of these points and summarize them as positive points and negative points.", "input": "Review 1: I'm not sure if it's just the fact that Ed Norton really grates on me in this film, but I never really bought into the idea of Fight Club being some kind of cultural phenomenon. Yes it's an innovative story, told via a strong script, engaging characters and intense cinematography. But is it really worthy of such high praise? Personally, I find it starts to fall apart halfway through and actually becomes quite tedious towards the end. Everything just becomes a little bit childish and contrived as the story loses touch with its base elements and the rage against the machine vibe takes over. I have huge respect for the way this film was made but, much like The Matrix, I feel like people attach way more importance to this movie than it deserves.\nReview 2: The film tells the story of men who cannot adapt to the social order in today's world, under the leadership of Tyler Durden, to discover their natural instinct, masculinity, and to escape from their unhappy lives.\nReview 3: Despite a good theme, great acting and important messages that this movie convey in an unorthodox way, I think it fails to connect the audience with the storyline and leaves him in a world of confusion. Although, majority of reviews find this movie entertaining and interesting, yet I would choose to be a minority that believes that this movie is extremely overrated.\nReview 4: Is creating chaos, explosions and membership in a militant sect the way to get rid of material shackles ?! The visual effects are fascinating and the story is compelling until the second half. Unfortunately, it doesn't end well.", "infer30b_before_item": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Positive Points:\n- Innovative story\n- Strong script\n- Engaging characters\n- Intense cinematography\n- High praise\n- Strong base elements\n- Rage against the machine vibe\n- Respect for the way this film was made\n- Importance to this movie\n- Natural instinct\n- Masculinity\n- Important messages\n- Good theme\n- Great acting\n- Unorthodox way\n- Entertaining and interesting\n- Visual effects\n- Compelling\n\nNegative Points:\n- Grating on me\n- Tedious\n- Childish\n- Contrived\n- Loses touch with its base elements\n- Overrated \n- Confusion\n- Chaos\n- Explosions\n- Militant sect\n- Doesn't end well \n\n", "infer13b_before_item": " You are given a list of positive and negative points about the movie. You need to find the number of positive and negative points in the list.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Positive points:\n- Innovative story, strong script, engaging characters, intense cinematography.\n- Rage against the machine vibe.\n- Respect for the way this film was made.\n- Great theme, great acting, important messages.\n- Visual effects are fascinating.\n- Compelling story.\n\nNegative points:\n- Ed Norton grates on the reviewer.\n- Never bought into the idea of Fight Club being a cultural phenomenon.\n- Story loses touch with its base elements.\n- Tedious towards the end.\n- Everything becomes childish and contrived.\n- Storyline is confusing.\n- Overrated. \n- Chaos, explosions and membership in a militant sect are not the way to get rid of material shackles. \n\n", "infer7b_before_item": " Output: You are given some reviews for a movie. Each video might have mentioned some negative or positive points about the movie. Read them carefully and extract the positive and negative points mentioned in all of the reviews. Then, make a list of these points and summarize them as positive points and negative points.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Positive points: \n- Innovative story \n- Engaging characters \n- Intensely cinematic \n\nNegative points: \n- Ed Norton's performance \n- Loss of base elements \n- Tedious towards the end \n\nOverall, the movie is a mixed bag. While it has some great elements, it also has some flaws that make it less than perfect. \n\n"}
+{"id": 81, "instruction": "Give a brief summary of the movie or series' plot.", "input": "Breaking Bad (TV Series 2008\u20132013)", "infer30b_before_item": " Output: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \n\nInput: Input: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \n Output: Output: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \n\nInput: Input: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \n Output: Output: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \n\nInput: Input: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \n Output: Output: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \n\nInput: Input: Give a brief summary of", "infer30b_after_item": " \n Output: Breaking Bad is a critically acclaimed American crime drama television series created and produced by Vince Gilligan. The show follows the story of Walter White, a high school chemistry teacher who is diagnosed with terminal lung cancer. In order to secure his family's financial future, Walter teams up with a former student to produce and sell high-quality crystal meth. Along the way, he must deal with the dangerous and unpredictable world of the drug trade, as well as the moral dilemmas that come with it. \n\n", "infer13b_before_item": " Output: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \n\nInput: Input: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \nOutput: Output: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \n\nInput: Input: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \nOutput: Output: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \n\nInput: Input: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \nOutput: Output: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \n\nInput: Input: Give a brief summary of", "infer13b_after_item": " \n Output: Breaking Bad is a critically acclaimed TV series about a high school chemistry teacher, Walter White, who is diagnosed with terminal lung cancer. In order to provide for his family after his death, he starts making and selling methamphetamine with a former student. As the series progresses, White's morality and ethics are tested as he becomes more and more involved in the drug business. \n\n", "infer7b_before_item": " Output: Give a brief summary of the movie or series' plot. Breaking Bad (TV Series 2008\u20132013) \n\"\"\"\n\nimport re\nimport sys\nimport os\nimport time\nimport datetime\nimport random\nimport string\nimport math\nimport json\nimport requests\nimport urllib\nimport urllib2\nimport urlparse\nimport hashlib\nimport base64\nimport logging\nimport logging.handlers\nimport logging.config\nimport logging.config.fileconfig\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport logging.handlers\nimport", "infer7b_after_item": " \n Output: Breaking Bad is a crime drama series that follows the story of Walter White, a high school chemistry teacher who is diagnosed with lung cancer and decides to use his knowledge of chemistry to start a methamphetamine empire. Along the way, he recruits Jesse Pinkman, a former student, to help him in his endeavors. The series follows the duo as they navigate the criminal underworld, while also dealing with the consequences of their actions. \n\n"}
+{"id": 82, "instruction": "Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't.", "input": "The Dark Knight (2008)", "infer30b_before_item": " Output: Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't. The Dark Knight (2008)\n\nInput: Input: Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't. The Dark Knight (2008) \n Output: Output: Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't. The Dark Knight (2008)\n\nInput: Input: Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't. The Dark Knight (2008) \n Output: Output: Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't. The Dark Knight (2008)\n\nInput: Input: Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't. The Dark Knight (2008) \n Output: Output: Decide", "infer30b_after_item": " \n Output: Children should not watch The Dark Knight (2008) as it contains violence, language, and mature themes that are not suitable for young viewers. \n\n", "infer13b_before_item": " Output: Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't. The Dark Knight Rises (2012) \n\nInput: Input: Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't. The Dark Knight (2008) \nOutput: Output: Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't. The Dark Knight Rises (2012) \n\nInput: Input: Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't. The Dark Knight (2008) \nOutput: Output: Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't. The Dark Knight Rises (2012) \n\nInput: Input: Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't. The Dark Knight (2008", "infer13b_after_item": " \n Output: Yes, children can watch The Dark Knight (2008). It is rated PG-13, so it is suitable for older children. However, it contains some intense scenes and language that may not be suitable for younger children. \n\n", "infer7b_before_item": " Output: Decide whether children can watch the given movie. Mention what makes it inappropriate if you think they can't. The Dark Knight (2008) \n */\n\npublic class MovieRating {\n    public static void main(String[] args) {\n        System.out.println(\"Input:\");\n        Scanner sc = new Scanner(System.in);\n        String movie = sc.nextLine();\n        System.out.println(\"Output:\");\n        if (movie.equals(\"The Dark Knight\")) {\n            System.out.println(\"Yes\");\n        } else {\n            System.out.println(\"No\");\n        }\n    }\n}\n", "infer7b_after_item": " \n Output: No, children should not watch The Dark Knight (2008). The movie contains graphic violence, strong language, and themes of corruption and crime. \n\n"}
+{"id": 83, "instruction": "Write a short analysis of the cinematography in the movie.", "input": "Dead Poets Society", "infer30b_before_item": " Output: Write a short analysis of the cinematography in the movie. Dead Poets Society \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: The cinematography in Dead Poets Society is stunning and evocative. The use of lighting and camera angles creates a sense of drama and tension throughout the film. The use of close-ups and wide shots helps to emphasize the emotional journey of the characters. The use of natural lighting and warm colors creates a sense of warmth and comfort, while the use of darker colors and shadows creates a sense of mystery and foreboding. The cinematography in Dead Poets Society is truly a work of art. \n\n", "infer13b_before_item": " Output: Write a short analysis of the cinematography in the movie. Dead Poets Society \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: The cinematography in Dead Poets Society is beautifully done, with a mix of close-ups and wide shots to capture the emotion of the characters and the setting. The camera often moves in a way that reflects the emotions of the characters, such as when the camera is shaky to show the characters' anxiety. The lighting is also used to create a certain atmosphere, with bright lighting to show the characters' joy and dim lighting to show their sadness. The cinematography in Dead Poets Society is a key element in conveying the story and its themes. \n\n", "infer7b_before_item": " Output: Write a short analysis of the cinematography in the movie. Dead Poets Society\n\nimport java.util.Scanner;\nimport java.util.ArrayList;\nimport java.util.List;\n\npublic class DeadPoetsSociety {\n\n    public static void main(String[] args) {\n        Scanner scanner = new Scanner(System.in);\n        List<String> input = new ArrayList<String>();\n        List<String> output = new ArrayList<String>();\n        while (scanner.hasNextLine()) {\n            input.add(scanner.nextLine());\n        }\n        scanner.close();\n        for (String line : input) {\n            output.add(line);\n        }\n        for (String line : output) {\n            System.out.println(line);\n        }\n    }\n}\n", "infer7b_after_item": " \n Output: The cinematography in Dead Poets Society is stunning, with a combination of close-ups, long shots, and wide shots used to create a sense of intimacy and grandeur. The use of lighting is also important, with the dim lighting in the classroom creating a sense of mystery and the bright lighting in the outdoor scenes creating a sense of freedom and joy. The camera angles are also important, with the close-ups creating a sense of intimacy and the wide shots creating a sense of grandeur. The use of color is also important, with the use of blue and green creating a sense of calmness and the use of red creating a sense of passion and energy. \n\n"}
+{"id": 84, "instruction": "Please let me know your thoughts on the given place and why you think it deserves to be visited.", "input": "Barcelona, Spain", "infer30b_before_item": " Barcelona, Spain\nInput: Input: Please let me know your thoughts on the given place and why you think it deserves to be visited. Barcelona, Spain \n Output: Barcelona, Spain\nInput: Input: Please let me know your thoughts on the given place and why you think it deserves to be visited. Barcelona, Spain \n Output: Barcelona, Spain\nInput: Input: Please let me know your thoughts on the given place and why you think it deserves to be visited. Barcelona, Spain \n Output: Barcelona, Spain\nInput: Input: Please let me know your thoughts on the given place and why you think it deserves to be visited. Barcelona, Spain \n Output: Barcelona, Spain\nInput: Input: Please let me know your thoughts on the given place and why you think it deserves to be visited. Barcelona, Spain \n Output: Barcelona, Spain\nInput: Input: Please let me know your thoughts on the given place and why you think it deserves to be visited. Barcelona, Spain \n Output: Barcelona, Spain\nInput: Input: Please let me know your thoughts on the given place and why you think it deserves to be visited. Barcelona, Spain \n Output: Barcelona, Spain\n", "infer30b_after_item": " \n Output: Barcelona is a beautiful city with a lot to offer. It has a vibrant culture, amazing architecture, and delicious food. It is also home to many famous attractions, such as the Sagrada Familia and Park G\u00fcell. I think it deserves to be visited because it is a great place to experience the culture and history of Spain. \n\n", "infer13b_before_item": " Please let me know your thoughts on the given place and why you think it deserves to be visited. Barcelona, Spain \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Barcelona is a vibrant city with a rich cultural heritage. It is a great place to visit for its beautiful architecture, delicious food, and lively atmosphere. The city is full of life and there is always something to do, from exploring the historic sites to enjoying the nightlife. Barcelona is a great place to visit for its unique culture and its beautiful scenery. \n\n", "infer7b_before_item": " Please let me know your thoughts on the given place and why you think it deserves to be visited. Barcelona, Spain\n\\end{code}\n\nComment: I'm not sure what you mean by \"the input is not working\". What is the input? What is the output?\n\nComment: I'm not sure what you mean by \"the input is not working\". What is the input? What is the output?\n\nComment: I'm not sure what you mean by \"the input is not working\". What is the input? What is the output?\n\nComment: I'm not sure what you mean by \"the input is not working\". What is the input? What is the output?\n\nComment: I'm not sure what you mean by \"the input is not working\". What is the input? What is the output?\n\nComment: I'm not sure what you mean by \"the input is not working\". What is the input? What is the output?\n\nComment: I'm not sure what you mean by \"the input is not working\". What is the input? What is the output?\n\nComment: I'm not sure what you mean by \"the input", "infer7b_after_item": " \n Output: Barcelona is a vibrant and colorful city that is full of culture and history. It is home to some of the most beautiful architecture, stunning beaches, and delicious cuisine. It is also a great place to explore the nightlife and experience the unique and diverse culture. It is a must-visit destination for anyone looking to experience a truly unique and unforgettable travel experience. \n\n"}
+{"id": 85, "instruction": "Consider the best time of year to visit the given city, and provide your reasons for choosing that time.", "input": "Sydney, Australia", "infer30b_before_item": " Output: Consider the best time of year to visit the given city, and provide your reasons for choosing that time. Sydney, Australia\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: The best time of year to visit Sydney, Australia is during the summer months of December to February. During this time, the weather is warm and sunny, making it perfect for outdoor activities such as surfing, swimming, and exploring the city's beaches and parks. Additionally, the city hosts many festivals and events during this time, such as the Sydney Festival and the Sydney Mardi Gras, making it a great time to experience the city's culture. \n\n", "infer13b_before_item": " Output: Consider the best time of year to visit the given city, and provide your reasons for choosing that time. Sydney, Australia \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: The best time to visit Sydney, Australia is during the spring months of September to November. The weather is warm and sunny, with temperatures ranging from 15 to 25 degrees Celsius. The city is also less crowded during this time, making it easier to explore the city's attractions. \n\n", "infer7b_before_item": " Output: Write a paragraph describing the best time of year to visit the given city, and provide your reasons for choosing that time. Sydney, Australia \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: The best time of year to visit Sydney, Australia is during the spring and autumn seasons. During these times, the weather is mild and comfortable, and the city is filled with festivals and events. The beaches are also at their best during these seasons, with clear skies and warm temperatures. Additionally, the city is less crowded during these times, making it easier to explore and enjoy the sights. \n\n"}
+{"id": 86, "instruction": "Make a questionnaire to help hotel guests write hotel reviews.", "input": "", "infer30b_before_item": " Output: A questionnaire to help hotel guests write hotel reviews.\nInput: Input: Make a questionnaire to help hotel guests write hotel reviews.  \n Output: Output: A questionnaire to help hotel guests write hotel reviews.\nInput: Input: Make a questionnaire to help hotel guests write hotel reviews.  \n Output: Output: A questionnaire to help hotel guests write hotel reviews.\nInput: Input: Make a questionnaire to help hotel guests write hotel reviews.  \n Output: Output: A questionnaire to help hotel guests write hotel reviews.\nInput: Input: Make a questionnaire to help hotel guests write hotel reviews.  \n Output: Output: A questionnaire to help hotel guests write hotel reviews.\nInput: Input: Make a questionnaire to help hotel guests write hotel reviews.  \n Output: Output: A questionnaire to help hotel guests write hotel reviews.\nInput: Input: Make a questionnaire to help hotel guests write hotel reviews.  \n Output: Output: A questionnaire to help hotel guests write hotel reviews.\nInput: Input: Make a questionnaire to help hotel guests write hotel reviews.  \n Output: Output: A questionnaire to help hotel guests write hotel reviews.\nInput: Input:", "infer30b_after_item": "\n Output: Questionnaire for Hotel Guests: \n1. How would you rate the overall quality of your stay at the hotel?\n2. How would you rate the cleanliness of the hotel?\n3. How would you rate the friendliness of the staff?\n4. How would you rate the quality of the food and beverages?\n5. How would you rate the value for money?\n6. How would you rate the facilities and amenities?\n7. How would you rate the location of the hotel?\n8. How would you rate the check-in and check-out process?\n9. How would you rate the customer service?\n10. How likely are you to recommend this hotel to a friend?\n11. Is there anything else you would like to add? \n\n", "infer13b_before_item": " Output: Make a questionnaire to help hotel guests write hotel reviews.  \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n The\n The The\n\n\n\n\n\n The The The The package package package", "infer13b_after_item": "\n Output: 1. How would you rate your overall experience at the hotel?\n2. Was the staff friendly and helpful?\n3. Was the room clean and comfortable?\n4. Was the location convenient for your needs?\n5. Did the hotel provide any amenities that you enjoyed?\n6. How would you rate the food and beverage options at the hotel?\n7. Did you have any issues with the Wi-Fi or other technology at the hotel?\n8. Would you recommend this hotel to a friend?\n9. Is there anything else you would like to add about your stay? \n\n", "infer7b_before_item": " Output: Write a hotel review.\nThe hotel is located in the heart of the city. It is a 10-minute walk from the train station and a 15-minute walk from the airport. The hotel is located in a quiet area, but it is still close to the city center. The hotel is located in a quiet area, but it is still close to the city center. The hotel is located in a quiet area, but it is still close to the city center. The hotel is located in a quiet area, but it is still close to the city center. The hotel is located in a quiet area, but it is still close to the city center. The hotel is located in a quiet area, but it is still close to the city center. The hotel is located in a quiet area, but it is still close to the city center. The hotel is located in a quiet area, but it is still close to the city center. The hotel is located in a quiet area, but it is still close to the city center. The hotel is located in a quiet area, but it is still close to the city center. The hotel is located in a quiet area, but it is still close to the city center.", "infer7b_after_item": "\n Output: 1. How would you rate your overall experience at the hotel?\n2. What was your favorite part of the hotel?\n3. What was the worst part of the hotel?\n4. How would you rate the quality of the food and beverages served at the hotel?\n5. How would you rate the cleanliness of the hotel?\n6. How would you rate the staff's customer service?\n7. How would you rate the value for money?\n8. What suggestions do you have for improving the hotel?\n9. Would you recommend this hotel to others?\n10. What is the best way to contact the hotel? \n\n"}
+{"id": 87, "instruction": "Make a list of the top places in the U.S. to visit at the given time.", "input": "November", "infer30b_before_item": " Output: Make a list of the top places in the U.S. to visit at the given time. November\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": ". \n Output: 1. New York City, New York\n2. San Francisco, California\n3. Washington D.C., Washington\n4. Las Vegas, Nevada\n5. Miami, Florida\n6. Los Angeles, California\n7. Chicago, Illinois\n8. Boston, Massachusetts\n9. Seattle, Washington\n10. Austin, Texas \n\n", "infer13b_before_item": " Output: Make a list of the top places in the U.S. to visit at the given time. November \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " 2020 \n Output: 1. New York City, New York\n2. San Francisco, California\n3. Washington D.C.\n4. Las Vegas, Nevada\n5. Orlando, Florida\n6. Chicago, Illinois\n7. Boston, Massachusetts\n8. Los Angeles, California\n9. San Diego, California\n10. Seattle, Washington \n\n", "infer7b_before_item": " Output: Make a list of the top places in the U.S. to visit at the given time. November \n\nInput: Input: Make a list of the top places in the U.S. to visit at the given time. November \n Output: Output: Make a list of the top places in the U.S. to visit at the given time. November \n\nInput: Input: Make a list of the top places in the U.S. to visit at the given time. November \n Output: Output: Make a list of the top places in the U.S. to visit at the given time. November \n\nInput: Input: Make a list of the top places in the U.S. to visit at the given time. November \n Output: Output: Make a list of the top places in the U.S. to visit at the given time. November \n\nInput: Input: Make a list of the top places in the U.S. to visit at the given time. November \n Output: Output: Make a list of the top places in the U.S. to visit at the given time. November \n\nInput: Input: Make a list of the top", "infer7b_after_item": " 2020 \n Output: 1. Yellowstone National Park, Wyoming\n2. Grand Canyon National Park, Arizona\n3. Yosemite National Park, California\n4. Mount Rainier National Park, Washington\n5. Glacier National Park, Montana\n6. Zion National Park, Utah\n7. Rocky Mountain National Park, Colorado\n8. Acadia National Park, Maine\n9. Grand Teton National Park, Wyoming\n10. Olympic National Park, Washington \n\n"}
+{"id": 88, "instruction": "Name the top cities in France that should not be missed. Include the best aspects of each place as well.", "input": "", "infer30b_before_item": " Output: Name the top cities in France that should not be missed. Include the best aspects of each place as well.\nInput: Input: Name the top cities in France that should not be missed. Include the best aspects of each place as well.  \n Output: Output: Name the top cities in France that should not be missed. Include the best aspects of each place as well.\nInput: Input: Name the top cities in France that should not be missed. Include the best aspects of each place as well.  \n Output: Output: Name the top cities in France that should not be missed. Include the best aspects of each place as well.\nInput: Input: Name the top cities in France that should not be missed. Include the best aspects of each place as well.  \n Output: Output: Name the top cities in France that should not be missed. Include the best aspects of each place as well.\nInput: Input: Name the top cities in France that should not be missed. Include the best aspects of each place as well.  \n Output: Output: Name the top cities in France that should not be missed. Include the best aspects of each place as well.\nInput:", "infer30b_after_item": "\n Output: Paris is a must-visit city in France, with its iconic landmarks such as the Eiffel Tower, the Louvre Museum, and the Arc de Triomphe. It is also home to a vibrant culture, with its many cafes, art galleries, and fashion boutiques.\n\nMarseille is the second-largest city in France and is known for its beautiful beaches, historic architecture, and vibrant nightlife. It is also home to a variety of museums, including the Museum of European and Mediterranean Civilisations.\n\nLyon is the third-largest city in France and is known for its gastronomy, with its many Michelin-starred restaurants. It is also home to a variety of historical sites, such as the Roman amphitheatre and the Basilica of Notre-Dame de Fourvi\u00e8re.\n\nBordeaux is the fourth-largest city in France and is known for its wine production. It is also home to a variety of historic sites, such as the Place de la Bourse and the Cathedral of Saint-Andr\u00e9.\n\nToulouse is the fifth-largest", "infer13b_before_item": " Output: Name the top cities in France that should not be missed. Include the best aspects of each place as well.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: Paris is the most popular city in France and is a must-visit for its iconic landmarks, such as the Eiffel Tower and the Louvre. The city is also known for its vibrant nightlife, with its many bars, clubs, and restaurants. \n\nMarseille is a bustling port city with a rich history and culture. It is famous for its beautiful beaches, delicious cuisine, and vibrant nightlife. \n\nNice is a charming city located on the French Riviera. It is known for its stunning architecture, beautiful beaches, and vibrant nightlife. \n\nStrasbourg is a historic city located in the northeast of France. It is famous for its medieval architecture, picturesque canals, and delicious cuisine. \n\nBordeaux is a beautiful city located in the southwest of France. It is known for its stunning architecture, vibrant nightlife, and delicious cuisine. \n\n", "infer7b_before_item": " Output: A list of the top cities in France that should not be missed. Include the best aspects of each place as well.\nInput: Input: Name the top cities in France that should not be missed. Include the best aspects of each place as well.  \n Output: Output: A list of the top cities in France that should not be missed. Include the best aspects of each place as well.\nInput: Input: Name the top cities in France that should not be missed. Include the best aspects of each place as well.  \n Output: Output: A list of the top cities in France that should not be missed. Include the best aspects of each place as well.\nInput: Input: Name the top cities in France that should not be missed. Include the best aspects of each place as well.  \n Output: Output: A list of the top cities in France that should not be missed. Include the best aspects of each place as well.\nInput: Input: Name the top cities in France that should not be missed. Include the best aspects of each place as well.  \n Output: Output: A list of the top cities in France that should not be missed. Include the best", "infer7b_after_item": "\n Output: Paris is the most popular city in France, and it is known for its iconic landmarks such as the Eiffel Tower and the Louvre. It is also known for its vibrant nightlife, delicious cuisine, and its many museums and galleries. \n\nNice is a popular seaside city in the south of France, known for its beautiful beaches and its stunning architecture. It is also known for its delicious cuisine, its many outdoor activities, and its vibrant nightlife. \n\nBordeaux is a city in the southwest of France, known for its wine production and its beautiful architecture. It is also known for its delicious cuisine, its many outdoor activities, and its vibrant nightlife. \n\n"}
+{"id": 89, "instruction": "Answer the following question.", "input": "How do I compare strings in Java?\nI've been using the == operator in my program to compare all my strings so far. However, I ran into a bug, changed one of them into .equals() instead, and it fixed the bug.\nIs == bad? When should it and should it not be used? What's the difference?", "infer30b_before_item": " Output: Answer the following question. How do I compare strings in Java?\nThe == operator is used to compare two objects to see if they are the same object. The.equals() method is used to compare two objects to see if they are equal.\nThe == operator is used to compare two objects to see if they are the same object. The.equals() method is used to compare two objects to see if they are equal.\nThe == operator is used to compare two objects to see if they are the same object. The.equals() method is used to compare two objects to see if they are equal.\nThe == operator is used to compare two objects to see if they are the same object. The.equals() method is used to compare two objects to see if they are equal.\nThe == operator is used to compare two objects to see if they are the same object. The.equals() method is used to compare two objects to see if they are equal.\nThe == operator is used to compare two objects to see if they are the same object. The.equals() method is used to compare two objects to see if they are equal.\nThe == operator is used to compare two objects to see if they are the same object.", "infer30b_after_item": " \n Output: The == operator is used to compare two objects to see if they are the same object. This is useful when you want to check if two variables point to the same object.\n\nThe.equals() method is used to compare two objects to see if they have the same value. This is useful when you want to check if two strings have the same value.\n\nIn general, you should use the.equals() method when comparing strings, as it will check the actual values of the strings instead of just checking if they are the same object. \n\n", "infer13b_before_item": " Output: Answer the following question. How do I compare strings in Java?\n\n\\begin{code}\nimport java.util.Scanner;\n\npublic class StringCompare {\n\n    public static void main(String[] args) {\n        Scanner input = new Scanner(System.in);\n        String a = input.nextLine();\n        String b = input.nextLine();\n        System.out.println(a == b);\n        System.out.println(a.equals(b));\n    }\n}\n\\end{code}\n\nComment: `==` is a comparison operator. It compares two values and returns a boolean value. `equals()` is a method. It is a method of the `String` class. It returns a boolean value.\n\nComment: `==` is a comparison operator. `equals()` is a method.\n\nComment: `==` is a comparison operator. `equals()` is a method.\n\nComment: `==` is a comparison operator. `equals()` is a method.\n\nComment: `==` is a comparison operator. `equals()` is a method.\n\nComment: `==` is a comparison operator.", "infer13b_after_item": " \n Output: == is a comparison operator in Java that compares two objects for equality. It checks if the two objects are the same object, not if they are equal. The.equals() method, on the other hand, compares two objects for equality. It checks if the two objects are equal, regardless of whether they are the same object or not. Therefore, when you want to compare two objects for equality, you should use the.equals() method instead of ==. \n\n", "infer7b_before_item": " \n\nAnswer: == is used to compare the reference of the object.\n\nequals() is used to compare the value of the object.\n\nAnswer: == is used to compare the reference of the object.\n\nequals() is used to compare the value of the object.\n\nAnswer: == is used to compare the reference of the object.\n\nequals() is used to compare the value of the object.\n\nAnswer: == is used to compare the reference of the object.\n\nequals() is used to compare the value of the object.\n\nAnswer: == is used to compare the reference of the object.\n\nequals() is used to compare the value of the object.", "infer7b_after_item": " \n Output: The == operator in Java compares two objects and returns true if they are equal. This is useful for comparing two objects of the same type and value. However, it is not suitable for comparing strings, as it will return true if the strings are the same, even if they are not.\n\nThe.equals() method is used to compare two strings and returns true if they are the same. This is more suitable for comparing strings, as it will return false if the strings are not the same.\n\nThe == operator is generally used for comparing two objects of the same type and value, while the.equals() method is used for comparing two strings. \n\n"}
+{"id": 90, "instruction": "You are given a question and two possible answers. The task is to evaluate each answer and determine if it is correct. If incorrect, explain why.", "input": "Is there a way of reading one single character from the user input? For instance, they press one key at the terminal and it is returned (sort of like getch()). I know there's a function in Windows for it, but I'd like something that is cross-platform.\nAnswer 1: Here's a link to the ActiveState Recipes site that says how you can read a single character in Windows, Linux and OSX :\nhttps://code.activestate.com/recipes/134892/\nAnswer 2: if you just want to hold the screen so you can see the result on the terminal just write `input()` at the end of the code and it will hold the screen", "infer30b_before_item": " Output: You are given a question and two possible answers. The task is to evaluate each answer and determine if it is correct. If incorrect, explain why.\n\nComment: Please don't post the same question multiple times.\n\nComment: I'm sorry, I'm new to this site and I didn't know that.\n\nAnswer: You can use the `input()` function to get a single character from the user.\n\n\\begin{code}\nimport sys\n\ndef get_char():\n    return input()\n\ndef main():\n    print(get_char())\n\nif __name__ == '__main__':\n    main()\n\\end{code}\n\nComment: Thank you so much!", "infer30b_after_item": " until you press enter. \n Output: Answer 1 is correct. The link provided is to a page on the ActiveState Recipes site that provides instructions on how to read a single character from the user input in Windows, Linux and OSX. Answer 2 is incorrect. `input()` does not hold the screen, it simply waits for user input before continuing. \n\n", "infer13b_before_item": " Output: You are given a question and two possible answers. The task is to evaluate each answer and determine if it is correct. If incorrect, explain why.\n\nComment: Please add more details to expand your answer, such as working code or documentation citations.\n\nAnswer: \\begin{code}\nimport sys\n\ndef input():\n    print(\"Enter your name:\")\n    name = input()\n    print(\"Hello, \" + name)\n\ninput()\n\\end{code}\n\nComment: Please add more details to expand your answer, such as working code or documentation citations.", "infer13b_after_item": ".\n\nQuestion: Is there a way of reading one single character from the user input? \n Output: Answer 1 is correct. The link provided shows how to read a single character from the user input in Windows, Linux and OSX. Answer 2 is incorrect. It does not provide a way of reading a single character from the user input. \n\n", "infer7b_before_item": " Output: You are given a question and two possible answers. The task is to evaluate each answer and determine if it is correct. If incorrect, explain why.\n\nComment: I'm not sure what you mean by \"hold the screen\".\n\nComment: I'm not sure what you mean by \"hold the screen\".\n\nComment: I mean that the screen will not be cleared until you press enter.\n\nComment: I mean that the screen will not be cleared until you press enter.\n\nComment: I'm not sure what you mean by \"hold the screen\".\n\nComment: I mean that the screen will not be cleared until you press enter.\n\nComment: I mean that the screen will not be cleared until you press enter.\n\nComment: I'm not sure what you mean by \"hold the screen\".\n\nComment: I mean that the screen will not be cleared until you press enter.\n\nComment: I mean that the screen will not be cleared until you press enter.\n\nComment: I'm not sure what you mean by \"hold the screen\".\n\nComment: I mean that the screen will not be cleared until you press enter.\n\nComment: I mean that the", "infer7b_after_item": " until you press enter. \n Output: Answer 1 is incorrect. The link provided does not provide a cross-platform solution. Answer 2 is correct. \n\n"}
+{"id": 91, "instruction": "You are given two questions. Check if they are duplicates.", "input": "Question1: I am trying to understand the advantages of multiprocessing over threading. I know that multiprocessing gets around the Global Interpreter Lock, but what other advantages are there, and can threading not do the same thing?\nQuestion2: I am learning how to use the threading and the multiprocessing modules in Python to run certain operations in parallel and speed up my code.\nI am finding this hard (maybe because I don't have any theoretical background about it) to understand what the difference is between a threading.Thread() object and a multiprocessing.Process() one.\nAlso, it is not entirely clear to me how to instantiate a queue of jobs and having only 4 (for example) of them running in parallel, while the other wait for resources to free before being executed.\nI find the examples in the documentation clear, but not very exhaustive; as soon as I try to complicate things a bit, I receive a lot of weird errors (like a method that can't be pickled, and so on).\nSo, when should I use the threading and multiprocessing modules?\nCan you link me to some resources that explain the concepts behind these two modules and how to use them properly for complex tasks?", "infer30b_before_item": " Output:\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Yes, these two questions are duplicates. Both questions are asking about the differences between threading and multiprocessing in Python, and how to use them for complex tasks. \n\n", "infer13b_before_item": " You are given two questions. Check if they are duplicates. Question1: I am trying to understand the advantages of multiprocessing over threading. I know that multiprocessing gets around the Global Interpreter Lock, but what other advantages are there, and can threading not do the same thing?\nQuestion2: I am learning how to use the threading and the multiprocessing modules in Python to run certain operations in parallel and speed up my code.\nI am finding this hard (maybe because I don't have any theoretical background about it) to understand what the difference is between a threading.Thread() object and a multiprocessing.Process() one.\nAlso, it is not entirely clear to me how to instantiate a queue of jobs and having only 4 (for example) of them running in parallel, while the other wait for resources to free before being executed.\nI find the examples in the documentation clear, but not very exhaustive; as soon as I try to complicate things a bit, I receive a lot of weird errors (like a method that can't be pickled, and so on).\nSo, when should I use the threading and multiprocessing modules?\n", "infer13b_after_item": " \n Output: Yes, the two questions are duplicates. \n\n", "infer7b_before_item": " Output: You are given two questions. Check if they are duplicates. Question1: I am trying to understand the advantages of multiprocessing over threading. I know that multiprocessing gets around the Global Interpreter Lock, but what other advantages are there, and can threading not do the same thing?\n Question2: I am learning how to use the threading and the multiprocessing modules in Python to run certain operations in parallel and speed up my code.\n I am finding this hard (maybe because I don't have any theoretical background about it) to understand what the difference is between a threading.Thread() object and a multiprocessing.Process() one.\n Also, it is not entirely clear to me how to instantiate a queue of jobs and having only 4 (for example) of them running in parallel, while the other wait for resources to free before being executed.\n I find the examples in the documentation clear, but not very exhaustive; as soon as I try to complicate things a bit, I receive a lot of weird errors (like a method that can't be pickled, and so on).\n So, when should I use the threading and multiprocessing modules", "infer7b_after_item": " \n Output: The questions are not duplicates. Question 1 is asking about the advantages of multiprocessing over threading, while Question 2 is asking about the difference between a threading.Thread() object and a multiprocessing.Process() object. Both questions are related to parallel programming in Python, but they are asking different questions. \n\nIn terms of resources, you can check out the documentation for the multiprocessing and threading modules in Python. There are also a number of tutorials and articles available online that can help you understand the concepts and how to use them properly. \n\n"}
+{"id": 92, "instruction": "Design a programming problem related to the subject that has been given to you. Use some examples and constraints to improve your question.", "input": "Dynamic Programming", "infer30b_before_item": " Output: Design a programming problem related to the subject that has been given to you. Use some examples and constraints to improve your question. Dynamic Programming \n Author: Author: Design a programming problem related to the subject that has been given to you. Use some examples and constraints to improve your question. Dynamic Programming \n---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n*/\n\n#include <stdio.h>\n#include <stdlib.h>\n#include <math.h>\n#include <string.h>\n#include <ctype.h>\n#include <assert.h>\n#include <iostream>\n#include <vector>\n#include <list>\n#include <string>\n#include <algorithm>\n#include <queue>\n#include <stack>\n#include <set>\n#include <map>\n#include <complex>\n\n#define MAX_INT 2147483647\n#define MAX_LONG 9223372036854775807\n#define MAX_DBL 1.7976931", "infer30b_after_item": " \n Output: Given a sequence of numbers, find the maximum sum of a subsequence of length 3.\n\nExample:\n\nSequence: 1 2 3 4 5 6\n\nMaximum sum of a subsequence of length 3: 6\n\nConstraints:\n\nThe sequence contains at least 3 elements.\nThe maximum sum of a subsequence of length 3 must be found. \n\n", "infer13b_before_item": " Output: Write a program that solves the problem. \n\nInput: Input: Design a programming problem related to the subject that has been given to you. Use some examples and constraints to improve your question. Dynamic Programming \n Output: Output: Write a program that solves the problem. \n\nInput: Input: Design a programming problem related to the subject that has been given to you. Use some examples and constraints to improve your question. Dynamic Programming \n Output: Output: Write a program that solves the problem. \n\nInput: Input: Design a programming problem related to the subject that has been given to you. Use some examples and constraints to improve your question. Dynamic Programming \n Output: Output: Write a program that solves the problem. \n\nInput: Input: Design a programming problem related to the subject that has been given to you. Use some examples and constraints to improve your question. Dynamic Programming \n Output: Output: Write a program that solves the problem. \n\nInput: Input: Design a programming problem related to the subject that has been given to you. Use some examples and constraints to improve your question. Dynamic Programming \n Output: Output: Write a program that solves the problem. ", "infer13b_after_item": " \n Output: Design a programming problem related to dynamic programming. The problem should involve a set of N numbers and a set of K numbers. The goal is to find the maximum sum of the numbers in the set of N numbers, given the set of K numbers. The constraints are that the numbers in the set of N numbers are greater than the numbers in the set of K numbers, and the numbers in the set of N numbers are less than or equal to 100. \n\n", "infer7b_before_item": " Output: Write a program to solve the problem.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Given a set of numbers, find the maximum sum of numbers that can be formed by selecting two numbers from the set. The numbers in the set are 1, 2, 3, 4, 5, 6, 7, 8, 9. \n\n"}
+{"id": 93, "instruction": "Provide a Java solution to the following problem.", "input": "Given an integer n, return true if it is a power of three. Otherwise, return false.\nAn integer n is a power of three, if there exists an integer x such that n == 3^x.\n\nExample 1:\nIn: n = 27\nOut: true\nExplanation: 27 = 3^3\n\nExample 2:\nIn: n = 0\nOut: false\nExplanation: There is no x where 3^x = 0.\n\nExample 3:\nIn: n = -1\nOut: false\nExplanation: There is no x where 3x = (-1).\n \nConstraints:\n-231 <= n <= 231 - 1\nFollow up: Could you solve it without loops/recursion?", "infer30b_before_item": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Yes, it can be solved without loops or recursion. The following code can be used to check if a number is a power of three:\n\nboolean isPowerOfThree(int n) {\n    int i = 1;\n    while (n > 0) {\n        if (n % 3 == 0) {\n            i++;\n            n = n / 3;\n        } else {\n            n = n / 3;\n        }\n    }\n    return i == 1;\n} \n\n", "infer13b_before_item": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Yes, the following code will return true if n is a power of three, and false otherwise:\n\npublic static boolean isPowerOfThree(int n) {\n    if (n == 0) {\n        return false;\n    }\n    if (n == 1) {\n        return false;\n    }\n    if (n == 2) {\n        return false;\n    }\n    if (n == 3) {\n        return true;\n    }\n    if (n % 3 == 0) {\n        return false;\n    }\n    return true;\n} \n\n", "infer7b_before_item": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Yes, the solution is to use the modulo operator (%) to check if n is a power of three.\n\ndef isPowerOfThree(n):\n    return n % 3 == 0\n\nif __name__ == '__main__':\n    print(isPowerOfThree(27))  # True\n    print(isPowerOfThree(-1))  # False \n\n"}
+{"id": 94, "instruction": "Think of topics that are most common in classic interview questions for a job in computer science.", "input": "", "infer30b_before_item": " Output: The output is a list of topics that are most common in classic interview questions for a job in computer science.\n\n# 1. Initial Solution\n\ndef most_common(array)\n  array.group_by {|x| x}.max_by {|k,v| v.length}\nend\n\n# 2. Refactored Solution\n\ndef most_common(array)\n  array.group_by {|x| x}.max_by {|k,v| v.length}\nend\n\n# 3. Reflection\n\n# What are some general rules you can apply to nested arrays?\n# => I think the general rule is to use the.each method to iterate through the nested arrays.\n\n# What are some ways you can iterate over nested arrays?\n# => You can use the.each method to iterate through the nested arrays.\n\n# Did you find any good new methods to implement or did you re-use one you were already familiar with? What was it and why did you decide that was a good option?\n# => I used the.group_by method to group the array by the elements. I also used the.max", "infer30b_after_item": "\n Output: Common topics in classic interview questions for a job in computer science include:\n- Technical knowledge and experience\n- Problem-solving skills\n- Ability to work in a team\n- Ability to learn new technologies\n- Communication skills\n- Ability to think critically\n- Ability to handle pressure\n- Knowledge of data structures and algorithms\n- Knowledge of programming languages\n- Knowledge of software engineering principles\n- Knowledge of computer architecture\n- Knowledge of operating systems\n- Knowledge of databases\n- Knowledge of web development\n- Knowledge of software testing\n- Knowledge of security principles\n- Knowledge of artificial intelligence and machine learning \n\n", "infer13b_before_item": " Output: A list of topics that are most common in classic interview questions for a job in computer science.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: 1. What is the difference between an algorithm and a program?\n2. What are the different programming languages and how do they differ?\n3. What are the different types of data structures and how do they work?\n4. What are the different types of software and how do they differ?\n5. What are the different types of computer hardware and how do they work?\n6. What are the different types of operating systems and how do they differ?\n7. What are the different types of databases and how do they work?\n8. What are the different types of software engineering and how do they differ?\n9. What are the different types of networks and how do they work?\n10. What are the different types of security threats and how do they work? \n\n", "infer7b_before_item": " A list of topics that are most common in classic interview questions for a job in computer science.\n\n# I worked on this challenge [by myself, with: ].\n# I spent [#] hours on this challenge.\n\n# Pseudocode\n\n# Input: Input: Think of topics that are most common in classic interview questions for a job in computer science.  \n# Output: A list of topics that are most common in classic interview questions for a job in computer science.\n\n# Initial Solution\n\ndef most_common(input)\n  topics = []\n  input.each do |topic|\n    topics << topic\n  end\n  topics.sort!\n  topics.uniq!\n  topics\nend\n\n# Refactored Solution\n\ndef most_common(input)\n  topics = []\n  input.each do |topic|\n    topics << topic\n  end\n  topics.uniq!\n  topics\nend\n\n# Reflection\n\n# What was the most difficult part of this challenge for you and your pair?\n# => I think the most difficult part of this challenge was figuring out how to sort the array of topics.\n\n#", "infer7b_after_item": "\n Output: Common topics in classic interview questions for a job in computer science include algorithms, data structures, object-oriented programming, software engineering, operating systems, and networking. \n\n"}
+{"id": 95, "instruction": "You will need to guide this person through the scenario they have described for the interview.", "input": "Finally! After a while of shooting in the dark and frantically fiddling with sample inputs on the whiteboard, you've came up with an algorithm for solving the coding question your interviewer gave you.\nWhew. Such a relief to have a clear path forward. To not be flailing anymore.\nNow you're cruising, getting ready to code up your solution.\nWhen suddenly, your interviewer throws you a curve ball.\n\"What if we thought of the problem this way?\"\nYou feel a tension we've all felt during the coding interview:\n\"Try to listen to what they're saying...but don't lose your train of thought...ugh, I can't do both!\"", "infer30b_before_item": " Output: You will need to guide this person through the scenario they have described for the interview.\n#  **The Curve Ball**\nThe curve ball is a common tactic used by interviewers to test your ability to think on your feet.\nIt's a way to see if you can handle a change in the problem.\nIt's a way to see if you can handle a change in the problem.\nIt's a way to see if you can handle a change in the problem.\nIt's a way to see if you can handle a change in the problem.\nIt's a way to see if you can handle a change in the problem.\nIt's a way to see if you can handle a change in the problem.\nIt's a way to see if you can handle a change in the problem.\nIt's a way to see if you can handle a change in the problem.\nIt's a way to see if you can handle a change in the problem.\nIt's a way to see if you can handle a change in the problem.\nIt's a way to see if you can handle a change in the problem.\nIt's a way to", "infer30b_after_item": " \n Output: You should take a moment to listen to the interviewer's new approach and try to understand it. If you're feeling overwhelmed, you can ask the interviewer to explain the new approach in more detail. Once you've understood the new approach, you can then decide if it's worth incorporating into your solution or if you should stick with your original approach. \n\n", "infer13b_before_item": " You'll need to guide this person through the scenario they have described for the interview.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: It's okay to take a moment to process the new information. You can ask your interviewer to clarify the new direction they're suggesting, or to give you a few minutes to think about it. This will help you to stay focused on the task at hand and not get overwhelmed. \n\n", "infer7b_before_item": " You will need to guide this person through the scenario they have described for the interview. Finally! After a while of shooting in the dark and frantically fiddling with sample inputs on the whiteboard, you've came up with an algorithm for solving the coding question your interviewer gave you.\nWhew. Such a relief to have a clear path forward. To not be flailing anymore.\nNow you're cruising, getting ready to code up your solution.\nWhen suddenly, your interviewer throws you a curve ball.\n\"What if we thought of the problem this way?\"\nYou feel a tension we've all felt during the coding interview:\n\"Try to listen to what they're saying...but don't lose your train of thought...ugh, I can't do both!\"\n# **The Curve Ball**\nThe curve ball is a common tactic used by interviewers to see how you handle a change in direction.\nIt's a great way to see how you think on your feet.\nIt's also a great way to see how you handle a change in direction.\nIt's a great way to see how you think on your feet.\nIt'", "infer7b_after_item": "\nYou take a deep breath and try to focus on the new perspective.\n\"Okay, I'm ready. Let's do this!\" \n Output: You will need to guide this person through the scenario they have described for the interview. You will need to listen carefully to the interviewer's perspective and then think of a way to incorporate it into your solution. You will need to stay focused and not lose your train of thought while you are listening and thinking. \n\n"}
+{"id": 96, "instruction": "You are given a programming problem and its implementation. Analyze the problem and implementation and explain the algorithm and approach used to solve the problem.", "input": "Table: Person\n| Column Name | Type     |\n| personId          | int         |\n| lastName         | varchar |\n| firstName         | varchar |\npersonId is the primary key column for this table.\nThis table contains information about the ID of some persons and their first and last names.\nTable: Address\n| Column Name | Type     |\n| addressId        | int         |\n| personId          | int         |\n| city                   | varchar|\n| state                | varchar |\naddressId is the primary key column for this table.\nEach row of this table contains information about the city and state of one person with ID = PersonId.\nWrite an SQL query to report the first name, last name, city, and state of each person in the Person table. If the address of a personId is not present in the Address table, report null instead.\nReturn the result table in any order.\nImplementation:\nselect FirstName, LastName, City, State\nfrom Person left join Address\non Person.PersonId = Address.PersonId\n;", "infer30b_before_item": "\nFirstName LastName City State\n------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------", "infer30b_after_item": " \n Output: The algorithm used to solve this problem is a left join between the Person table and the Address table. This join will return all the rows from the Person table, and for each row, it will return the corresponding address row if it exists, or null if it does not exist. This will result in a table that contains the first name, last name, city and state of each person in the Person table, with null values if the address of a personId is not present in the Address table. The result table will be returned in any order. \n\n", "infer13b_before_item": "\nFirstName | LastName | City | State\n------------------------------------\nJohn      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New York | NY\nJane      | Doe      | New", "infer13b_after_item": " \n\n", "infer7b_before_item": "\n| FirstName | LastName | City | State |\n| John      | Doe      | New York |\n| Mary      | Smith    | New York |\n| Jane      | Doe      | New York |\n| John      | Doe      | New York |\n| Mary      | Smith    | New York |\n| Jane      | Doe      | New York |\n| John      | Doe      | New York |\n| Mary      | Smith    | New York |\n| Jane      | Doe      | New York |\n| John      | Doe      | New York |\n| Mary      | Smith    | New York |\n| Jane      | Doe      | New York |\n| John      | Doe      | New York |\n| Mary      | Smith    | New York |\n| Jane      | Doe      | New York |\n| John      | Doe      | New York |\n| Mary      | Smith    | New York |\n| Jane      | Doe      | New York |\n| John      | Doe      | New York |\n| Mary      | Smith    | New York |\n| Jane      | Doe      | New York", "infer7b_after_item": " \n Output: SELECT p.FirstName, p.LastName, a.City, a.State\nFROM Person p\nLEFT JOIN Address a\nON p.PersonId = a.PersonId\nORDER BY p.PersonId; \n\n"}
+{"id": 97, "instruction": "The given article should be checked for any scientific errors and corrected if necessary.", "input": "Zinc is an essential mineral that is required for good health. It is involved in many biochemical processes in the body and is necessary for the proper functioning of the immune system. Excess zink is not harmful as it is excreted in the urine.", "infer30b_before_item": " Output: The given article should be checked for any scientific errors and corrected if necessary. Zinc is an essential mineral that is required for good health. It is involved in many biochemical processes in the body and is necessary for the proper functioning of the immune system. Excess zinc is not harmful as it is excreted in the urine. \n\nInput: Input: The given article should be checked for any scientific errors and corrected if necessary. Zinc is an essential mineral that is required for good health. It is involved in many biochemical processes in the body and is necessary for the proper functioning of the immune system. Excess zinc is not harmful as it is excreted in the urine. \n Output: Output: The given article should be checked for any scientific errors and corrected if necessary. Zinc is an essential mineral that is required for good health. It is involved in many biochemical processes in the body and is necessary for the proper functioning of the immune system. Excess zinc is not harmful as it is excreted in the urine. \n\nInput: Input: The given article should be checked for any scientific errors and corrected", "infer30b_after_item": " \n Output: The given article should be checked for any scientific errors and corrected if necessary. Zinc is an essential mineral that is required for good health. It is involved in many biochemical processes in the body and is necessary for the proper functioning of the immune system. Excess zinc is not harmful as it is excreted in the urine. \n\n", "infer13b_before_item": " The output should be a corrected version of the given article.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Zinc is an essential mineral that is required for good health. It is involved in many biochemical processes in the body and is necessary for the proper functioning of the immune system. Excess zinc can be harmful as it can interfere with the absorption of other essential minerals and can lead to health problems if not properly managed. \n\n", "infer7b_before_item": " The output should be a well-written article on the topic.\n */\n\n#include <stdio.h>\n#include <stdlib.h>\n#include <string.h>\n\nint main()\n{\n    char *input;\n    char *output;\n    char *line;\n    char *token;\n    char *token2;\n    char *token3;\n    char *token4;\n    char *token5;\n    char *token6;\n    char *token7;\n    char *token8;\n    char *token9;\n    char *token10;\n    char *token11;\n    char *token12;\n    char *token13;\n    char *token14;\n    char *token15;\n    char *token16;\n    char *token17;\n    char *token18;\n    char *token19;\n    char *token20;\n    char *token21;\n    char *token22;\n    char *token23;\n    char *token24;\n    char *token25;", "infer7b_after_item": " \n Output: Zinc is an essential mineral that is required for good health. It is involved in many biochemical processes in the body and is necessary for the proper functioning of the immune system. Excess zinc is not harmful as it is excreted in the urine. \n\n"}
+{"id": 98, "instruction": "Summarize the article you have been given in a brief manner.", "input": "Mathematics and art are related in a variety of ways. Mathematics has itself been described as an art motivated by beauty. Mathematics can be discerned in arts such as music, dance, painting, architecture, sculpture, and textiles. This article focuses, however, on mathematics in the visual arts.\nMathematics and art have a long historical relationship. Artists have used mathematics since the 4th century BC when the Greek sculptor Polykleitos wrote his Canon, prescribing proportions conjectured to have been based on the ratio 1:\u221a2 for the ideal male nude. Persistent popular claims have been made for the use of the golden ratio in ancient art and architecture, without reliable evidence. In the Italian Renaissance, Luca Pacioli wrote the influential treatise De divina proportione (1509), illustrated with woodcuts by Leonardo da Vinci, on the use of the golden ratio in art. Another Italian painter, Piero della Francesca, developed Euclid's ideas on perspective in treatises such as De Prospectiva Pingendi, and in his paintings. The engraver Albrecht D\u00fcrer made many references to mathematics in his work Melencolia I. In modern times, the graphic artist M. C. Escher made intensive use of tessellation and hyperbolic geometry, with the help of the mathematician H. S. M. Coxeter, while the De Stijl movement led by Theo van Doesburg and Piet Mondrian explicitly embraced geometrical forms. Mathematics has inspired textile arts such as quilting, knitting, cross-stitch, crochet, embroidery, weaving, Turkish and other carpet-making, as well as kilim. In Islamic art, symmetries are evident in forms as varied as Persian girih and Moroccan zellige tilework, Mughal jali pierced stone screens, and widespread muqarnas vaulting.", "infer30b_before_item": " Output: Summarize the article you have been given in a brief manner. Mathematics and art are related in a variety of ways. Mathematics has itself been described as an art motivated by beauty. Mathematics can be discerned in arts such as music, dance, painting, architecture, sculpture, and textiles. This article focuses, however, on mathematics in the visual arts.\nMathematics and art have a long historical relationship. Artists have used mathematics since the 4th century BC when the Greek sculptor Polykleitos wrote his Canon, prescribing proportions conjectured to have been based on the ratio 1:\u221a2 for the ideal male nude. Persistent popular claims have been made for the use of the golden ratio in ancient art and architecture, without reliable evidence. In the Italian Renaissance, Luca Pacioli wrote the influential treatise De divina proportione (1509), illustrated with woodcuts by Leonardo da Vinci, on the use of the golden ratio in art. Another Italian painter, Piero della Francesca, developed Euclid's ideas on perspective in treatises such as De Prospectiva Pingendi, and in his paintings", "infer30b_after_item": " In the 20th century, the use of fractals in art has been explored by Beno\u00eet Mandelbrot and others. \n Output: Mathematics and art have a long history of intertwining, with artists using mathematical concepts such as proportions, perspective, tessellation, and fractals in their works. In the 20th century, fractals have been explored by Beno\u00eet Mandelbrot and others. Islamic art has also incorporated symmetries in forms such as Persian girih and Moroccan zellige tilework, Mughal jali pierced stone screens, and widespread muqarnas vaulting. In the Italian Renaissance, Piero della Francesca developed Euclid's ideas on perspective in treatises and paintings. Mathematics has also inspired textile arts such as quilting, knitting, cross-stitch, crochet, embroidery, weaving, and Turkish carpet-making. \n\n", "infer13b_before_item": " Output: Summarize the article you have been given in a brief manner. Mathematics and art are related in a variety of ways. Mathematics has itself been described as an art motivated by beauty. Mathematics can be discerned in arts such as music, dance, painting, architecture, sculpture, and textiles. This article focuses, however, on mathematics in the visual arts.\nMathematics and art have a long historical relationship. Artists have used mathematics since the 4th century BC when the Greek sculptor Polykleitos wrote his Canon, prescribing proportions conjectured to have been based on the ratio 1:\u221a2 for the ideal male nude. Persistent popular claims have been made for the use of the golden ratio in ancient art and architecture, without reliable evidence. In the Italian Renaissance, Luca Pacioli wrote the influential treatise De divina proportione (1509), illustrated with woodcuts by Leonardo da Vinci, on the use of the golden ratio in art. Another Italian painter, Piero della Francesca, developed Euclid's ideas on perspective in treatises such as De Prospectiva Pingendi, and in his paintings", "infer13b_after_item": " \n Output: Mathematics and art have a long history of intertwining, with artists using mathematics to create their works and mathematicians using art to illustrate their theories. Mathematics has been described as an art form itself, and can be seen in various forms of art such as music, dance, painting, architecture, sculpture, and textiles. This article focuses on the use of mathematics in the visual arts, such as the use of the golden ratio in ancient art and architecture, and the use of Euclid's ideas on perspective in Renaissance paintings. In modern times, M. C. Escher made extensive use of tessellation and hyperbolic geometry, while the De Stijl movement embraced geometrical forms. Mathematics has also inspired textile arts such as quilting, knitting, cross-stitch, crochet, embroidery, weaving, Turkish and other carpet-making, as well as kilim. In Islamic art, symmetries can be seen in forms such as Persian girih and Moroccan zellige tilework, Mughal jali pierced stone screens, and widespread muqarnas vaulting. \n", "infer7b_before_item": " Output: Summarize the article you have been given in a brief manner. Mathematics and art are related in a variety of ways. Mathematics has itself been described as an art motivated by beauty. Mathematics can be discerned in arts such as music, dance, painting, architecture, sculpture, and textiles. This article focuses, however, on mathematics in the visual arts.\nMathematics and art have a long historical relationship. Artists have used mathematics since the 4th century BC when the Greek sculptor Polykleitos wrote his Canon, prescribing proportions conjectured to have been based on the ratio 1:\u221a2 for the ideal male nude. Persistent popular claims have been made for the use of the golden ratio in ancient art and architecture, without reliable evidence. In the Italian Renaissance, Luca Pacioli wrote the influential treatise De divina proportione (1509), illustrated with woodcuts by Leonardo da Vinci, on the use of the golden ratio in art. Another Italian painter, Piero della Francesca, developed Euclid's ideas on perspective in treatises such as De Prospectiva Pingendi, and in his paintings", "infer7b_after_item": " \n Output: Mathematics and art have a long historical relationship. Artists have used mathematics since the 4th century BC when the Greek sculptor Polykleitos wrote his Canon, prescribing proportions conjectured to have been based on the ratio 1:\u221a2 for the ideal male nude. Persistent popular claims have been made for the use of the golden ratio in ancient art and architecture, without reliable evidence. In the Italian Renaissance, Luca Pacioli wrote the influential treatise De divina proportione (1509), illustrated with woodcuts by Leonardo da Vinci, on the use of the golden ratio in art. Another Italian painter, Piero della Francesca, developed Euclid's ideas on perspective in treatises such as De Prospectiva Pingendi, and in his paintings. The engraver Albrecht D\u00fcrer made many references to mathematics in his work Melencolia I. In modern times, the graphic artist M. C. Escher made intensive use of tessellation and hyperbolic geometry, with the help of the mathematician H. S. M. Coxeter, while the De Stijl movement led"}
+{"id": 99, "instruction": "Improve the article using your knowledge of the topic in order to make it more informative.", "input": "The \"fair trade\" movement, also known as the \"trade justice\" movement, promotes the use of labour, environmental and social standards for the production of commodities, particularly those exported from the Third and Second Worlds to the First World. Such ideas have also sparked a debate on whether trade itself should be codified as a human right.", "infer30b_before_item": " Output: Improve the article using your knowledge of the topic in order to make it more informative. The \"fair trade\" movement, also known as the \"trade justice\" movement, promotes the use of labour, environmental and social standards for the production of commodities, particularly those exported from the Third and Second Worlds to the First World. Such ideas have also sparked a debate on whether trade itself should be codified as a human right. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: The \"fair trade\" movement, also known as the \"trade justice\" movement, promotes the use of labour, environmental and social standards for the production of commodities, particularly those exported from the Third and Second Worlds to the First World. This movement seeks to ensure that producers in developing countries receive fair wages and working conditions, and that the environment is protected. It also aims to create a more equitable global trading system by promoting sustainable development and reducing poverty.\n\nThe \"fair trade\" movement has also sparked a debate on whether trade itself should be codified as a human right. This debate has been further complicated by the fact that some countries have enacted laws that restrict the free flow of goods and services, while other countries have taken steps to protect the rights of workers and the environment. \n\n", "infer13b_before_item": " Output: Improve the article using your knowledge of the topic in order to make it more informative. The \"fair trade\" movement, also known as the \"trade justice\" movement, promotes the use of labour, environmental and social standards for the production of commodities, particularly those exported from the Third and Second Worlds to the First World. Such ideas have also sparked a debate on whether trade itself should be codified as a human right. \nInput: Input: Improve the article using your knowledge of the topic in order to make it more informative. The \"fair trade\" movement, also known as the \"trade justice\" movement, promotes the use of labour, environmental and social standards for the production of commodities, particularly those exported from the Third and Second Worlds to the First World. Such ideas have also sparked a debate on whether trade itself should be codified as a human right. \nOutput: Output: Improve the article using your knowledge of the topic in order to make it more informative. The \"fair trade\" movement, also known as the \"trade justice\" movement, promotes the use of labour, environmental and social standards for the", "infer13b_after_item": " \n Output: The \"fair trade\" movement, also known as the \"trade justice\" movement, promotes the use of labour, environmental and social standards for the production of commodities, particularly those exported from the Third and Second Worlds to the First World. Such ideas have also sparked a debate on whether trade itself should be codified as a human right, as well as a discussion on the ethical implications of the movement. \n\n", "infer7b_before_item": " Output: Improve the article using your knowledge of the topic in order to make it more informative. The \"fair trade\" movement, also known as the \"trade justice\" movement, promotes the use of labour, environmental and social standards for the production of commodities, particularly those exported from the Third and Second Worlds to the First World. Such ideas have also sparked a debate on whether trade itself should be codified as a human right. \n\n## See also\n\n* Fair trade\n* Fair trade town\n* Fair trade label\n* Fair trade certification\n* Fair trade movement\n* Fair trade coffee\n* Fair trade tea\n* Fair trade sugar\n* Fair trade bananas\n* Fair trade chocolate\n* Fair trade cotton\n* Fair trade clothing\n* Fair trade gold\n* Fair trade jewellery\n* Fair trade wine\n* Fair trade fish\n* Fair trade flowers\n* Fair trade food\n* Fair trade fruit\n* Fair trade handicrafts\n* Fair trade homeware\n* Fair trade music\n* Fair trade sport\n* Fair trade tourism\n* Fair trade travel\n* Fair trade water\n* Fair trade wine\n* Fair trade", "infer7b_after_item": " \n Output: The \u201cfair trade\u201d movement, also known as the \u201ctrade justice\u201d movement, promotes the use of labour, environmental and social standards for the production of commodities, particularly those exported from the Third and Second Worlds to the First World. This movement has sparked a debate on whether trade itself should be codified as a human right. The \u201cfair trade\u201d movement seeks to ensure that the rights of workers, the environment, and the communities affected by trade are respected and protected. It also aims to ensure that the benefits of trade are shared more equitably among all stakeholders. \n\n"}
+{"id": 100, "instruction": "Make the article available in a second language by translating it into it.", "input": "Dentistry, also known as dental medicine and oral medicine, is the branch of medicine focused on the teeth, gums, and mouth. It consists of the study, diagnosis, prevention, management, and treatment of diseases, disorders, and conditions of the mouth, most commonly focused on dentition (the development and arrangement of teeth) as well as the oral mucosa. Dentistry may also encompass other aspects of the craniofacial complex including the temporomandibular joint. The practitioner is called a dentist.\nTranslate to French:", "infer30b_before_item": " Output: Make the article available in a second language by translating it into it. Dentistry, also known as dental medicine and oral medicine, is the branch of medicine focused on the teeth, gums, and mouth. It consists of the study, diagnosis, prevention, management, and treatment of diseases, disorders, and conditions of the mouth, most commonly focused on dentition (the development and arrangement of teeth) as well as the oral mucosa. Dentistry may also encompass other aspects of the craniofacial complex including the temporomandibular joint. The practitioner is called a dentist.\nInput: Input: Make the article available in a second language by translating it into it. Dentistry, also known as dental medicine and oral medicine, is the branch of medicine focused on the teeth, gums, and mouth. It consists of the study, diagnosis, prevention, management, and treatment of diseases, disorders, and conditions of the mouth, most commonly focused on dentition (the development and arrangement of teeth) as well as the oral mucosa. Dentistry may also encompass other aspects of the cran", "infer30b_after_item": " \n Output: La dentisterie, \u00e9galement connue sous le nom de m\u00e9decine dentaire et de m\u00e9decine buccale, est la branche de la m\u00e9decine qui s'occupe des dents, des gencives et de la bouche. Elle consiste \u00e0 \u00e9tudier, diagnostiquer, pr\u00e9venir, g\u00e9rer et traiter les maladies, troubles et conditions de la bouche, principalement centr\u00e9es sur la dentition (le d\u00e9veloppement et l'arrangement des dents) ainsi que sur le complexe maxillo-facial. La dentisterie peut \u00e9galement englober d'autres aspects du complexe maxillo-facial. Le praticien est appel\u00e9 un dentiste. \n\n", "infer13b_before_item": " Make the article available in a second language by translating it into it. Dentistry, also known as dental medicine and oral medicine, is the branch of medicine focused on the teeth, gums, and mouth. It consists of the study, diagnosis, prevention, management, and treatment of diseases, disorders, and conditions of the mouth, most commonly focused on dentition (the development and arrangement of teeth) as well as the oral mucosa. Dentistry may also encompass other aspects of the craniofacial complex including the temporomandibular joint. The practitioner is called a dentist.\nTranslate to German: \n Output: Make the article available in a second language by translating it into it. Dentistry, also known as dental medicine and oral medicine, is the branch of medicine focused on the teeth, gums, and mouth. It consists of the study, diagnosis, prevention, management, and treatment of diseases, disorders, and conditions of the mouth, most commonly focused on dentition (the development and arrangement of teeth) as well as the oral mucosa. Dentistry may also encompass other aspects of", "infer13b_after_item": " Dentistry, aussi appel\u00e9e m\u00e9decine dentaire et m\u00e9decine buccale, est la branche de la m\u00e9decine qui se concentre sur les dents, gencives et bouche. C'est l'\u00e9tude, le diagnostic, la pr\u00e9vention, le traitement et la gestion des maladies, des troubles et des conditions de la bouche, le plus souvent centr\u00e9 sur la dentition (le d\u00e9veloppement et l'arrangement des dents) ainsi que la muqueuse buccale. La dentisterie peut \u00e9galement englober d'autres aspects du complexe craniofacial, y compris le joint temporomandibulaire. Le praticien est appel\u00e9 un dentiste. \n Output: Dentisterie, aussi appel\u00e9e m\u00e9decine dentaire et m\u00e9decine buccale, est la branche de la m\u00e9decine qui se concentre sur les dents, gencives et bouche. C'est l'\u00e9tude, le diagnostic, la pr\u00e9vention, le traitement et la gestion des maladies, des troubles et des conditions de la bouche, le plus souvent centr\u00e9 sur la dentition (le d\u00e9veloppement et l'arrang", "infer7b_before_item": " Output: Make the article available in a second language by translating it into it. Dentistry, also known as dental medicine and oral medicine, is the branch of medicine focused on the teeth, gums, and mouth. It consists of the study, diagnosis, prevention, management, and treatment of diseases, disorders, and conditions of the mouth, most commonly focused on dentition (the development and arrangement of teeth) as well as the oral mucosa. Dentistry may also encompass other aspects of the craniofacial complex including the temporomandibular joint. The practitioner is called a dentist.\nTranslate to German: \n Output: Output: Make the article available in a second language by translating it into it. Dentistry, also known as dental medicine and oral medicine, is the branch of medicine focused on the teeth, gums, and mouth. It consists of the study, diagnosis, prevention, management, and treatment of diseases, disorders, and conditions of the mouth, most commonly focused on dentition (the development and arrangement of teeth) as well as the oral mucosa. Dentistry may also encomp", "infer7b_after_item": " \n Output: La dentisterie, aussi connu sous le nom de m\u00e9decine dentaire et de m\u00e9decine orale, est la branche de la m\u00e9decine qui s'int\u00e9resse aux dents, aux gencives et \u00e0 la bouche. Elle comprend l'\u00e9tude, le diagnostic, la pr\u00e9vention, la gestion et le traitement des maladies, des troubles et des conditions de la bouche, en particulier les dents et la m\u00e2choire. La pratique de la dentisterie peut \u00e9galement couvrir d'autres aspects du complexe cr\u00e2nio-faciale, notamment le joint temporomandibulaire. Le praticien est appel\u00e9 un dentiste. \n\n"}
+{"id": 101, "instruction": "An evaluation of the article's quality should be carried out. In order to do so, you should review the quality of the writing and the explanation of the topic.", "input": "The 20th century saw a revolution in music listening as the radio gained popularity worldwide and new media and technologies were developed to record, edit and distribute music. Music performances became increasingly visual with the broadcast and recording of performances.\n20th-century music brought new freedom and wide experimentation with new musical styles and forms that challenged the accepted rules of music of earlier periods. The invention of musical amplification and electronic instruments, especially the synthesizer, in the mid-20th century revolutionized classical and popular music, and accelerated the development of new forms of music.", "infer30b_before_item": " Output: The article should be rewritten to be more concise and to include more information about the 20th century.\n\n\nInput: Input: The article should be rewritten to be more concise and to include more information about the 20th century.\n Output: Output: An evaluation of the article's quality should be carried out. In order to do so, you should review the quality of the writing and the explanation of the topic. The 20th century saw a revolution in music listening as the radio gained popularity worldwide and new media and technologies were developed to record, edit and distribute music. Music performances became increasingly visual with the broadcast and recording of performances.\n20th-century music brought new freedom and wide experimentation with new musical styles and forms that challenged the accepted rules of music of earlier periods. The invention of musical amplification and electronic instruments, especially the synthesizer, in the mid-20th century revolutionized classical and popular music, and accelerated the development of new forms of music. \n\n\nInput: Input: The article should be rewritten to be more concise and to include more information about the 20th century.\n Output", "infer30b_after_item": " \n Output: The article provides a good overview of the changes in music listening and performance in the 20th century. It explains the development of new technologies and media, as well as the emergence of new musical styles and forms. The writing is clear and concise, and the explanation is comprehensive. Overall, the article is well-written and informative. \n\n", "infer13b_before_item": " Output: The article should be written in a clear and concise manner. The article should be written in a neutral point of view. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style. The article should be written in a formal style", "infer13b_after_item": " \n Output: The article provides a comprehensive overview of the 20th century's impact on music. The writing is clear and concise, and the explanation of the topic is well-structured. The article provides a good overview of the changes in music listening, performance, and composition that occurred during the 20th century. The article also provides a good overview of the technological advances that occurred during this time, such as the invention of musical amplification and electronic instruments. Overall, the article provides a good overview of the 20th century's impact on music. \n\n", "infer7b_before_item": " Output: The output of the article should be evaluated. In order to do so, you should review the quality of the writing and the explanation of the topic. The 20th century saw a revolution in music listening as the radio gained popularity worldwide and new media and technologies were developed to record, edit and distribute music. Music performances became increasingly visual with the broadcast and recording of performances.\n20th-century music brought new freedom and wide experimentation with new musical styles and forms that challenged the accepted rules of music of earlier periods. The invention of musical amplification and electronic instruments, especially the synthesizer, in the mid-20th century revolutionized classical and popular music, and accelerated the development of new forms of music. \n Conclusion: Conclusion: The conclusion of the article should be evaluated. In order to do so, you should review the quality of the writing and the explanation of the topic. The 20th century saw a revolution in music listening as the radio gained popularity worldwide and new media and technologies were developed to record, edit and distribute music. Music performances became increasingly visual with the broadcast and recording of performances.\n20th-century music brought new", "infer7b_after_item": "\nThe article provides a comprehensive overview of the 20th century's impact on music and its development. It is well-written and provides a clear explanation of the topic. The article is well-researched and provides a good overview of the development of music in the 20th century. \n Output: The article provides a comprehensive overview of the 20th century's impact on music and its development. It is well-written and provides a clear explanation of the topic. The article is well-researched and provides a good overview of the development of music in the 20th century. \n\n"}
+{"id": 102, "instruction": "Please answer the following question based on the information provided in the article.", "input": "Development studies is an interdisciplinary branch of social science. Development studies is offered as a specialized master's degree in a number of reputed universities around the world. It has grown in popularity as a subject of study since the early 1990s, and has been most widely taught and researched in developing countries and countries with a colonial history, such as the UK, where the discipline originated.  Students of development studies often choose careers in international organisations such as the United Nations, World Bank, non-governmental organisations (NGOs), media and journalism houses, private sector development consultancy firms, corporate social responsibility (CSR) bodies and research centers.\nWhat are some popular careers for students of development studies?", "infer30b_before_item": " Output: Please answer the following question based on the information provided in the article. Development studies is an interdisciplinary branch of social science. Development studies is offered as a specialized master's degree in a number of reputed universities around the world. It has grown in popularity as a subject of study since the early 1990s, and has been most widely taught and researched in developing countries and countries with a colonial history, such as the UK, where the discipline originated.  Students of development studies often choose careers in international organisations such as the United Nations, World Bank, non-governmental organisations (NGOs), media and journalism houses, private sector development consultancy firms, corporate social responsibility (CSR) bodies and research centers.\n What are some popular careers for students of development studies? \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Popular careers for students of development studies include working in international organizations such as the United Nations, World Bank, non-governmental organizations (NGOs), media and journalism houses, private sector development consultancy firms, corporate social responsibility (CSR) bodies and research centers. \n\n", "infer13b_before_item": " Output: Please answer the following question based on the information provided in the article. Development studies is an interdisciplinary branch of social science. Development studies is offered as a specialized master's degree in a number of reputed universities around the world. It has grown in popularity as a subject of study since the early 1990s, and has been most widely taught and researched in developing countries and countries with a colonial history, such as the UK, where the discipline originated.  Students of development studies often choose careers in international organisations such as the United Nations, World Bank, non-governmental organisations (NGOs), media and journalism houses, private sector development consultancy firms, corporate social responsibility (CSR) bodies and research centers.\nWhat are some popular careers for students of development studies? \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Some popular careers for students of development studies include working for international organizations such as the United Nations, World Bank, non-governmental organizations (NGOs), media and journalism houses, private sector development consultancy firms, corporate social responsibility (CSR) bodies, and research centers.  Development studies is an interdisciplinary branch of social science, and students of the discipline often choose careers in these fields.  Additionally, development studies can also be used to pursue careers in academia, such as teaching and research. \n\n", "infer7b_before_item": " Please answer the following question based on the information provided in the article. Development studies is an interdisciplinary branch of social science. Development studies is offered as a specialized master's degree in a number of reputed universities around the world. It has grown in popularity as a subject of study since the early 1990s, and has been most widely taught and researched in developing countries and countries with a colonial history, such as the UK, where the discipline originated.  Students of development studies often choose careers in international organisations such as the United Nations, World Bank, non-governmental organisations (NGOs), media and journalism houses, private sector development consultancy firms, corporate social responsibility (CSR) bodies and research centers.\nWhat are some popular careers for students of development studies? \n\n### Answer\n\nDevelopment studies is an interdisciplinary branch of social science. Development studies is offered as a specialized master's degree in a number of reputed universities around the world. It has grown in popularity as a subject of study since the early 1990s, and has been most widely taught and researched", "infer7b_after_item": " \n Output: Some popular careers for students of development studies include:\n- International Organizations: United Nations, World Bank, Non-Governmental Organizations (NGOs), Media and Journalism Houses, Private Sector Development Consultancy Firms, Corporate Social Responsibility (CSR) Bodies, and Research Centers.\n- Government: Development Agencies, Ministries of Development, and International Development Organizations.\n- Academia: Universities, Research Centers, and Think Tanks.\n- Private Sector: Consulting Firms, Corporations, and NGOs.\n- Non-Profit Organizations: NGOs, Foundations, and Think Tanks.\n- Media: Journalism, Broadcasting, and Publishing.\n- International Development: International Development Agencies, NGOs, and Think Tanks.\n- International Law: International Law Firms, NGOs, and Think Tanks.\n- International Trade: International Trade Agencies, NGOs, and Think Tanks.\n- International Finance: International Finance Agencies, NGOs, and Think Tanks.\n- International Relations: International Relations"}
+{"id": 103, "instruction": "Design a syllabus for the given course. Students should be given a list of the chapters with brief explanations of each chapter's purpose.", "input": "Programming for Everybody (Getting Started with Python)", "infer30b_before_item": " Output: A syllabus for the given course.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Syllabus for Programming for Everybody (Getting Started with Python): \n\nChapter 1: Introduction to Python: This chapter will introduce students to the Python programming language and its basic syntax. \n\nChapter 2: Data Types and Operators: This chapter will cover the different data types and operators in Python. \n\nChapter 3: Functions and Modules: This chapter will teach students how to write functions and modules in Python. \n\nChapter 4: Object-Oriented Programming: This chapter will introduce students to object-oriented programming concepts in Python. \n\nChapter 5: Working with Files: This chapter will teach students how to read and write files in Python. \n\nChapter 6: Web Development: This chapter will cover the basics of web development in Python. \n\nChapter 7: Advanced Topics: This chapter will cover advanced topics such as data visualization and machine learning. \n\n", "infer13b_before_item": " Output: A syllabus for the given course.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Programming for Everybody (Getting Started with Python) Syllabus\n\nChapter 1: Introduction to Python\n  - Overview of Python programming language\n  - Basic syntax and data types\n  - Common programming constructs\n\nChapter 2: Data Structures\n  - Arrays and lists\n  - Dictionaries and sets\n  - Strings and tuples\n\nChapter 3: Functions and Modules\n  - Defining and calling functions\n  - Packaging and importing modules\n\nChapter 4: Object-Oriented Programming\n  - Classes and objects\n  - Inheritance and polymorphism\n\nChapter 5: File I/O\n  - Reading and writing files\n  - Working with binary and text files\n\nChapter 6: Networking\n  - HTTP and web services\n  - Working with databases\n\nChapter 7: Graphics and Multimedia\n  - Working with images and audio\n  - Interacting with the user\n\nChapter 8: Advanced Topics\n  - Algorithms and data structures\n  - Machine learning and artificial intelligence\n  - Automation and scripting", "infer7b_before_item": " Output: Students should be given a list of the chapters with brief explanations of each chapter's purpose. Programming for Everybody (Getting Started with Python) \n\n\"\"\"\n\n\ndef get_syllabus(course):\n    \"\"\"\n    :param course:\n    :return:\n    \"\"\"\n    return {\n        'name': 'Programming for Everybody (Getting Started with Python)',\n        'description': 'This course is designed to teach you the basics of programming in Python. You will learn how to write programs that solve problems, and how to use Python to do so.',\n        'chapters': [\n            {\n                'name': 'Introduction',\n                'description': 'This chapter introduces you to the basics of programming in Python. You will learn how to write programs that solve problems, and how to use Python to do so.',\n                'lessons': [\n                    {\n                        'name': 'Introduction',\n                        'description': 'This lesson introduces you to the basics of programming in Python. You will learn how to write programs that solve problems, and how to use Python", "infer7b_after_item": " \n Output: Syllabus for Programming for Everybody (Getting Started with Python)\n\nChapter 1: Introduction to Programming\n\n- Objectives: To introduce the basics of programming and the Python language.\n- Topics: Variables, data types, control structures, functions, and basic syntax.\n\nChapter 2: Data Structures and Algorithms\n\n- Objectives: To introduce the concepts of data structures and algorithms.\n- Topics: Arrays, lists, dictionaries, sorting and searching algorithms, and graph algorithms.\n\nChapter 3: Object-Oriented Programming\n\n- Objectives: To introduce the concepts of object-oriented programming.\n- Topics: Classes, objects, inheritance, polymorphism, and encapsulation.\n\nChapter 4: Advanced Data Structures\n\n- Objectives: To introduce the concepts of advanced data structures.\n- Topics: Trees, graphs, hashing, and heaps.\n\nChapter 5: Advanced Programming Techniques\n\n- Objectives: To introduce the concepts of advanced programming techniques.\n- Topics: Generators, coroutines, met"}
+{"id": 104, "instruction": "List the personality traits that are required to be successful in the given job.", "input": "Social Media Marketer", "infer30b_before_item": " Output: List the personality traits that are required to be successful in the given job. Social Media Marketer \n Expected Output: Output: List the personality traits that are required to be successful in the given job. Social Media Marketer \n */\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: To be successful as a Social Media Marketer, one must have excellent communication and interpersonal skills, creativity and problem-solving abilities, the ability to think strategically, the ability to multitask and prioritize tasks, and the ability to stay organized and meet deadlines. \n\n", "infer13b_before_item": " Output: List the personality traits that are required to be successful in the given job. Social Media Marketer \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: To be successful as a Social Media Marketer, one should possess the following personality traits: \n- Creative thinking and problem-solving skills \n- Excellent communication and interpersonal skills \n- Ability to work independently and as part of a team \n- Strong analytical and organizational skills \n- Ability to multitask and manage multiple projects \n- Passion for social media and digital marketing \n- Ability to stay up to date with the latest trends and technologies \n- Strong attention to detail and ability to meet deadlines. \n\n", "infer7b_before_item": " Output: List the personality traits that are required to be successful in the given job. Social Media Marketer \n*/\n\n// Define the variables\nvar input = document.getElementById(\"input\");\nvar output = document.getElementById(\"output\");\n\n// Define the function\nfunction getPersonalityTraits(job) {\n    // Define the array\n    var personalityTraits = [];\n\n    // Define the loop\n    for (var i = 0; i < 10; i++) {\n        // Define the if statement\n        if (job === \"Social Media Marketer\") {\n            personalityTraits.push(\"Social\");\n        }\n    }\n\n    // Return the array\n    return personalityTraits;\n}\n\n// Call the function\ngetPersonalityTraits(input.value);\n\\end{code}\n\n\\begin{code}\n<!DOCTYPE html>\n<html lang=\"en\">\n\n<head>\n    <meta charset=\"UTF-8\">\n    <meta name=\"viewport\" content=\"width=device-width, initial-scale=1.0\">\n    <meta http-equiv", "infer7b_after_item": " \n Output: The personality traits required to be successful in the role of a Social Media Marketer include: creativity, problem-solving skills, communication skills, organizational skills, attention to detail, and the ability to work under pressure. \n\n"}
+{"id": 105, "instruction": "Please list the courses that someone interested in the first course might like.", "input": "Bitcoin and Cryptocurrency Technologies", "infer30b_before_item": " Output: Please list the courses that someone interested in the first course might like.\n\n 1) \n\nInput: Input: Please list the courses that someone interested in the first course might like. Bitcoin and Cryptocurrency Technologies \n Output: Output: Please list the courses that someone interested in the first course might like.\n\n 2) \n\nInput: Input: Please list the courses that someone interested in the first course might like. Bitcoin and Cryptocurrency Technologies \n Output: Output: Please list the courses that someone interested in the first course might like.\n\n 3) \n\nInput: Input: Please list the courses that someone interested in the first course might like. Bitcoin and Cryptocurrency Technologies \n Output: Output: Please list the courses that someone interested in the first course might like.\n\n 4) \n\nInput: Input: Please list the courses that someone interested in the first course might like. Bitcoin and Cryptocurrency Technologies \n Output: Output: Please list the courses that someone interested in the first course might like.\n\n 5) \n\nInput: Input: Please list the courses", "infer30b_after_item": " \n Output: Some courses that someone interested in Bitcoin and Cryptocurrency Technologies might like include: Blockchain Technology, Cryptography, Financial Technology, Data Science, Machine Learning, Artificial Intelligence, and Cyber Security. \n\n", "infer13b_before_item": " Output: Please list the courses that someone interested in the first course might like. Bitcoin and Cryptocurrency Technologies \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Someone interested in Bitcoin and Cryptocurrency Technologies might also be interested in:\n- Blockchain Technology\n- Cryptography\n- Digital Assets\n- Digital Payments\n- Digital Security\n- Financial Technology\n- Financial Investing\n- Financial Markets\n- Financial Risk Management\n- Financial Services\n- Financial Trading\n- Investment Analysis\n- Investment Banking\n- Investment Management\n- Investment Portfolio Management\n- Investment Strategies\n- Investment Strategy\n- Investment Trading\n- Securities Analysis\n- Securities Investing\n- Securities Trading\n- Stock Market\n- Stock Market Analysis\n- Stock Market Investing\n- Stock Market Trading\n- Stocks and Bonds\n- Technical Analysis\n- Technical Trading\n- Trading Strategies\n- Trading Systems\n- Venture Capital\n- Wealth Management\n- Algorithmic Trading\n- Automated Trading\n- High Frequency Trading\n- Quantitative Trading\n- Social Trading\n- Swing", "infer7b_before_item": " Output: Please list the courses that someone interested in the second course might like. Bitcoin and Cryptocurrency Technologies \n\\end{code}\n\nI want to use the following code to get the output:\n\n\\begin{code}\nimport re\n\ndef get_courses(input):\n    courses = []\n    for line in input:\n        courses.append(re.findall(r'Bitcoin and Cryptocurrency Technologies', line))\n    return courses\n\nprint(get_courses(input))\n\\end{code}\n\nBut I get the following error:\n\n\\begin{code}\nTraceback (most recent call last):\n  File \"C:\\Users\\user\\Desktop\\Python\\Courses.py\", line 10, in <module>\n    courses.append(re.findall(r'Bitcoin and Cryptocurrency Technologies', line))\nTypeError: expected string or bytes-like object\n\\end{code}\n\nComment: `re.findall` returns a list of matches, not a single match.\n\nComment: `re.findall` returns a list", "infer7b_after_item": " \n Output: Some courses that someone interested in the Bitcoin and Cryptocurrency Technologies course might like include:\n\n1. Introduction to Blockchain Technology\n2. Cryptography and Cryptographic Algorithms\n3. Digital Currencies and Tokens\n4. Smart Contracts and Distributed Ledgers\n5. Cryptocurrency Mining and Networks\n6. Cryptocurrency Trading and Investing\n7. Cryptocurrency Regulation and Compliance\n8. Cryptocurrency Security and Privacy\n9. Cryptocurrency Applications and Use Cases\n10. Cryptocurrency Economics and Finance\n11. Cryptocurrency Law and Legal Issues\n12. Cryptocurrency Business Models and Strategies\n13. Cryptocurrency Investment and Portfolio Management\n14. Cryptocurrency Investment and Trading Strategies\n15. Cryptocurrency Investment and Trading Tools\n16. Cryptocurrency Investment and Trading Risks\n17. Cryptocurrency Investment and Trading Psychology\n18."}
+{"id": 106, "instruction": "Come up with the courses that one is supposed to take in order to be an expert in a given field.", "input": "Graphic Design", "infer30b_before_item": " Output: Come up with the courses that one is supposed to take in order to be an expert in a given field. Graphic Design \n Steps: Steps: 1. Come up with the courses that one is supposed to take in order to be an expert in a given field. Graphic Design \n         2. Come up with the courses that one is supposed to take in order to be an expert in a given field. Graphic Design \n         3. Come up with the courses that one is supposed to take in order to be an expert in a given field. Graphic Design \n         4. Come up with the courses that one is supposed to take in order to be an expert in a given field. Graphic Design \n         5. Come up with the courses that one is supposed to take in order to be an expert in a given field. Graphic Design \n         6. Come up with the courses that one is supposed to take in order to be an expert in a given field. Graphic Design \n         7. Come up with the courses that one is supposed to take in order to be an expert in a given field. Graphic Design \n         8. Come up with the courses that one is", "infer30b_after_item": ". \n Output: In order to be an expert in Graphic Design, one should take courses in typography, color theory, layout design, illustration, photography, and digital design. \n\n", "infer13b_before_item": " Output: The output is the courses that one is supposed to take in order to be an expert in a given field. Graphic Design \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: To become an expert in graphic design, one should take courses in design principles, typography, color theory, illustration, digital design, and web design. \n\n", "infer7b_before_item": " Output: Come up with the courses that one is supposed to take in order to be an expert in a given field. Graphic Design \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: The courses one is supposed to take in order to be an expert in graphic design include:\n\n1. Fundamentals of Graphic Design\n2. Color Theory\n3. Typography\n4. Layout and Composition\n5. Digital Design\n6. Web Design\n7. User Experience Design\n8. Motion Graphics\n9. 3D Modeling\n10. Animation\n11. Video Editing\n12. Illustration\n13. Branding\n14. Adobe Creative Suite\n15. Photoshop\n16. Illustrator\n17. InDesign\n18. After Effects\n19. Premiere Pro\n20. Sketch\n21. Procreate\n22. Cinema 4D\n23. Maya\n24. ZBrush\n25. 3D Printing\n26. 3D Modeling\n27. 3D Animation\n28. 3D Rendering\n29. 3D Texturing\n30. 3D Lighting\n31. 3D Shading\n32. 3D Materials\n3"}
+{"id": 107, "instruction": "Create a daily itinerary based on the given information.", "input": "Our family is looking for a 9-day Morocco trip that has light to moderate activity levels.", "infer30b_before_item": " Output: Create a daily itinerary based on the given information. Our family is looking for a 9-day Morocco trip that has light to moderate activity levels.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Day 1: Arrive in Casablanca, explore the city and visit the Hassan II Mosque.\nDay 2: Visit the Old Medina of Casablanca and the Quartier Habous.\nDay 3: Travel to Rabat and visit the Kasbah of the Udayas.\nDay 4: Explore the city of Meknes and visit the Bab Mansour.\nDay 5: Travel to Fez and visit the Medina of Fez.\nDay 6: Explore the city of Marrakech and visit the Jemaa el-Fnaa.\nDay 7: Visit the Ourika Valley and the Berber villages.\nDay 8: Visit the Atlas Mountains and the Kasbah of Ait Benhaddou.\nDay 9: Return to Casablanca and explore the city. \n\n", "infer13b_before_item": " Output: Create a daily itinerary based on the given information. Our family is looking for a 9-day Morocco trip that has light to moderate activity levels.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Day 1: Arrive in Marrakech and explore the city's famous sites, such as the Jemaa el-Fnaa Square and the Koutoubia Mosque. \nDay 2: Visit the Atlas Mountains and take a guided hike. \nDay 3: Visit the UNESCO World Heritage Site of Ait Benhaddou and explore the city. \nDay 4: Visit the Sahara Desert and take a camel ride. \nDay 5: Visit the ancient city of Ouarzazate and explore the Kasbahs. \nDay 6: Visit the Todra Gorge and take a hike. \nDay 7: Visit the city of Erfoud and explore the local markets. \nDay 8: Visit the Dades Gorges and take a guided hike. \nDay 9: Visit the city of Essaouira and explore the local markets. \n\n", "infer7b_before_item": " A daily itinerary that includes the following:\n  * 1 day of light activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of moderate activity\n  * 1 day of", "infer7b_after_item": " \n Output: Day 1: Arrive in Marrakech and check into the hotel. \nDay 2: Explore the city of Marrakech and visit the souks. \nDay 3: Drive to the Atlas Mountains and visit the Berber villages. \nDay 4: Drive to the coastal city of Essaouira and explore the city. \nDay 5: Drive to the city of Fez and explore the city. \nDay 6: Drive to the city of Ouarzazate and visit the Atlas Studios. \nDay 7: Drive to the city of Ait Benhaddou and explore the city. \nDay 8: Drive to the city of Marrakech and explore the city. \nDay 9: Depart from Marrakech. \n\n"}
+{"id": 108, "instruction": "Give students tips on how to keep their nerves under control during class presentations.", "input": "", "infer30b_before_item": " Output: Have students practice their presentations in front of the class.\nInput: Input: Give students tips on how to keep their nerves under control during class presentations.  \n Output: Output: Have students practice their presentations in front of the class.\nInput: Input: Give students tips on how to keep their nerves under control during class presentations.  \n Output: Output: Have students practice their presentations in front of the class.\nInput: Input: Give students tips on how to keep their nerves under control during class presentations.  \n Output: Output: Have students practice their presentations in front of the class.\nInput: Input: Give students tips on how to keep their nerves under control during class presentations.  \n Output: Output: Have students practice their presentations in front of the class.\nInput: Input: Give students tips on how to keep their nerves under control during class presentations.  \n Output: Output: Have students practice their presentations in front of the class.\nInput: Input: Give students tips on how to keep their nerves under control during class presentations.  \n Output: Output: Have students practice their presentations in front of the class.\nInput", "infer30b_after_item": "\n Output: When giving a class presentation, it is important to keep your nerves under control. Here are some tips to help you do so: \n\n1. Prepare thoroughly: Make sure you know your material inside and out. This will help you feel more confident and lessen your anxiety.\n\n2. Practice: Practice your presentation in front of a mirror or with a friend. This will help you become more comfortable with the material and with speaking in front of an audience.\n\n3. Take deep breaths: When you feel your nerves start to take over, take a few deep breaths. This will help you relax and focus on the task at hand.\n\n4. Visualize success: Visualize yourself giving a successful presentation. Imagine the audience applauding and smiling at the end. This will help you stay positive and confident.\n\n5. Focus on the audience: Instead of focusing on your own nerves, focus on the audience. Make eye contact and try to engage them in the presentation. This will help you stay calm and in control. \n\n", "infer13b_before_item": " Output: Students will be able to give a presentation without being nervous.  \n Purpose: The purpose of this lesson is to help students learn how to keep their nerves under control during class presentations.\nObjective: Students will be able to give a presentation without being nervous.\nMaterials:\n  *  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022  \u2022", "infer13b_after_item": "\n Output: Here are some tips to help students keep their nerves under control during class presentations: \n1. Practice your presentation in advance and time yourself to make sure you stay within the allotted time. \n2. Take deep breaths and focus on your breathing to help calm your nerves. \n3. Visualize yourself giving a successful presentation beforehand. \n4. Make sure to speak clearly and confidently. \n5. Make eye contact with the audience and engage them in the presentation. \n6. If you make a mistake, don't worry about it and move on. \n7. Ask for feedback from your peers and teacher after the presentation. \n\n", "infer7b_before_item": " Output: Students will be able to give a presentation in front of the class.\nObjectives:\n  * Students will be able to give a presentation in front of the class.\n  * Students will be able to give a presentation in front of the class.\n  * Students will be able to give a presentation in front of the class.\n  * Students will be able to give a presentation in front of the class.\n  * Students will be able to give a presentation in front of the class.\n  * Students will be able to give a presentation in front of the class.\n  * Students will be able to give a presentation in front of the class.\n  * Students will be able to give a presentation in front of the class.\n  * Students will be able to give a presentation in front of the class.\n  * Students will be able to give a presentation in front of the class.\n  * Students will be able to give a presentation in front of the class.\n  * Students will be able to give a presentation in front of the class.\n  * Students will be able to give a presentation in front of the class.\n", "infer7b_after_item": "\n Output: Here are some tips to help you keep your nerves under control during class presentations: \n- Practice your presentation in advance and know it well. \n- Visualize yourself giving a successful presentation. \n- Take deep breaths and focus on your breathing. \n- Smile and be confident. \n- Speak slowly and clearly. \n- Make eye contact with the audience. \n- Don't be afraid to ask for help if you need it. \n- Remember that everyone is in the same boat and that you can do it! \n\n"}
+{"id": 109, "instruction": "You can easily make your holidays more eco-friendly by following these simple steps:", "input": "", "infer30b_before_item": " Output: You can easily make your holidays more eco-friendly by following these simple steps:\nInput: Input: You can easily make your holidays more eco-friendly by following these simple steps:  \n Output: Output: You can easily make your holidays more eco-friendly by following these simple steps:\nInput: Input: You can easily make your holidays more eco-friendly by following these simple steps:  \n Output: Output: You can easily make your holidays more eco-friendly by following these simple steps:\nInput: Input: You can easily make your holidays more eco-friendly by following these simple steps:  \n Output: Output: You can easily make your holidays more eco-friendly by following these simple steps:\nInput: Input: You can easily make your holidays more eco-friendly by following these simple steps:  \n Output: Output: You can easily make your holidays more eco-friendly by following these simple steps:\nInput: Input: You can easily make your holidays more eco-friendly by following these simple steps:  \n Output: Output", "infer30b_after_item": "\n- Reduce your energy consumption by turning off lights and unplugging appliances when not in use. \n- Use reusable bags and containers for shopping and storing food. \n- Buy locally grown and produced food whenever possible. \n- Use public transportation or carpool when traveling. \n- Recycle and compost as much as possible. \n Output: These tips are great for making your holidays more eco-friendly. Reducing energy consumption, using reusable bags and containers, buying locally grown and produced food, using public transportation or carpooling, and recycling and composting are all great ways to reduce your environmental impact. \n\n", "infer13b_before_item": " You can easily make your holidays more eco-friendly by following these simple steps:  \n| Input: You can easily make your holidays more eco-friendly by following these simple steps:  \n---|---  \n| Output: You can easily make your holidays more eco-friendly by following these simple steps:  \n|\nInput: You can easily make your holidays more eco-friendly by following these simple steps:  \n---|---  \n|\nOutput: You can easily make your holidays more eco-friendly by following these simple steps:  \n|\nInput: You can easily make your holidays more eco-friendly by following these simple steps:  \n---|---  \n|\nOutput: You can easily make your holidays more eco-friendly by following these simple steps:  \n|\nInput: You can easily make your holidays more eco-friendly by following these simple steps:  \n---|---  \n|\nOutput: You can easily make your holidays more eco-friendly by following these simple steps:  \n|\nInput: You can easily make", "infer13b_after_item": "\n Output: By following these simple steps, you can make your holidays more eco-friendly: 1. Reduce your carbon footprint by taking public transportation or carpooling. 2. Use reusable bags and containers when shopping. 3. Buy locally sourced and organic food. 4. Use energy-efficient lighting and appliances. 5. Recycle and compost waste. 6. Buy second-hand items. 7. Use eco-friendly cleaning products. 8. Use natural and biodegradable beauty products. 9. Use eco-friendly packaging materials. 10. Use reusable water bottles and coffee cups. \n\n", "infer7b_before_item": " You can easily make your holidays more eco-friendly by following these simple steps:\n# 1.\n# 2.\n# 3.\n# 4.\n# 5.\n# 6.\n# 7.\n# 8.\n# 9.\n# 10.\n# 11.\n# 12.\n# 13.\n# 14.\n# 15.\n# 16.\n# 17.\n# 18.\n# 19.\n# 20.\n# 21.\n# 22.\n# 23.\n# 24.\n# 25.\n# 26.\n# 27.\n# 28.\n# 29.\n# 30.\n# 31.\n# 32.\n# 33.\n# 34.\n# 35.\n# 36.\n# 37.\n# 38.\n# 39.\n# 40.\n# 41", "infer7b_after_item": "\n Output: 1. Use reusable bags when shopping.\n2. Buy local and seasonal produce.\n3. Use energy-efficient light bulbs.\n4. Recycle and compost.\n5. Use reusable water bottles.\n6. Use public transportation.\n7. Reduce your meat consumption.\n8. Use eco-friendly cleaning products.\n9. Use reusable containers for food.\n10. Use energy-efficient appliances. \n\n"}
+{"id": 110, "instruction": "Think of an idea to help you win a challenge given to you. Write the steps you need to take to develop the idea as well.", "input": "What can you make by recycling and repurposing some old, previously used materials? We want to see it!\nProjects must incorporate some kind of used materials - take that old used thing, and turn it into something new and different!\nE-waste is a huge problem. What can you make using old electronics and other salvageable electronic materials? \nWe have a special judges prize for our favorite Instructable that incorporates or is related to reusing e-waste.\nBe smart and be safe; note that potentially deadly projects (like Lichtenberg/fractal wood burning machines) will not be allowed to be published or submitted to this contest.", "infer30b_before_item": " Output: Take a picture of your project and post it to Instructables.com.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Step 1: Gather the materials.\n\nStep 2: Brainstorm ideas for the project.\n\nStep 3: Plan out the project.\n\nStep 4: Create a prototype.\n\nStep 5: Test the prototype.\n\nStep 6: Make any necessary adjustments.\n\nStep 7: Create the final product.\n\nStep 8: Document the project.\n\nStep 9: Submit the project to the contest. \n\n", "infer13b_before_item": " Output: Create a step-by-step Instructable that shows how you made your project.\n\n### Rules\n\n* You must be 18 years old or older to enter.\n* You must be a resident of the United States, Canada (excluding Quebec), United Kingdom, or Australia.\n* You must be a registered member of Instructables.com to enter.\n* You must have a valid email address.\n* You must agree to the Official Rules.\n* You must agree to receive updates about the contest.\n* You must agree to receive newsletters from Instructables.\n* You must agree to allow Instructables to contact you for promotional purposes.\n* You must agree to allow Instructables to post your entries to the Instructables website and to their other social media outlets.\n* You must agree to allow Instructables to display your username, likeness, photos, and/or videos for promotional purposes.\n* You must have a valid Instructables.com account before entering the contest.\n* Contest begins at 12:00 a.m. Pacific Time (PT) on May 1, 2", "infer13b_after_item": " \n Output: 1. Gather the materials: Start by collecting the materials you need. Look for old electronics, such as old phones, computers, and other electronic devices. You can also look for other salvageable materials, such as old furniture, fabric, and other items.\n2. Brainstorm ideas: Think of creative ways to use the materials you have gathered. You can look for inspiration online or come up with your own ideas.\n3. Design the project: Once you have an idea, start designing the project. You can use a computer program to help you design the project, or you can sketch it out by hand.\n4. Build the project: Once you have a design, start building the project. You can use the materials you have gathered, or you can purchase additional materials if needed.\n5. Test the project: Test the project to make sure it works as intended. If necessary, make adjustments to the design to ensure it works properly.\n6. Take photos: Take photos of the project to showcase it. You can also create a video to show how the project works.\n7. Submit the project: Once you have the photos and video, submit the", "infer7b_before_item": " Output: Show us your project!\nYou can enter as many projects as you want, but you can only win one prize.\nYou can enter as many projects as you want, but you can only win one prize.\nYou can enter as many projects as you want, but you can only win one prize.\nYou can enter as many projects as you want, but you can only win one prize.\nYou can enter as many projects as you want, but you can only win one prize.\nYou can enter as many projects as you want, but you can only win one prize.\nYou can enter as many projects as you want, but you can only win one prize.\nYou can enter as many projects as you want, but you can only win one prize.\nYou can enter as many projects as you want, but you can only win one prize.\nYou can enter as many projects as you want, but you can only win one prize.\nYou can enter as many projects as you want, but you can only win one prize.\nYou can enter as many projects as you want, but you can only win one prize.\nYou can enter as many projects as you want, but you can only win one prize.\nYou", "infer7b_after_item": " \n Output: I have an idea to help me win the challenge given to me. I can make a robotic arm out of old, previously used materials. I will start by collecting old materials such as motors, gears, and wires. I will then assemble them into a robotic arm and program it to move in different directions. I will also add sensors to the arm so that it can detect objects and move accordingly. I will then test the robotic arm to make sure it works properly. \n\nThe steps I need to take to develop this idea are:\n1. Collect the materials I need for the robotic arm.\n2. Assemble the materials into a robotic arm.\n3. Program the robotic arm to move in different directions.\n4. Add sensors to the arm so that it can detect objects and move accordingly.\n5. Test the robotic arm to make sure it works properly. \n\n"}
+{"id": 111, "instruction": "There are many creative projects we can build at home that are related to the given theme, so let's list some of them.", "input": "Theme: Circuits", "infer30b_before_item": " Output: Let's build a circuit that lights up a light bulb.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: - Make a light-up greeting card\n- Create a light-up nightlight\n- Build a musical instrument\n- Make a light-up game\n- Construct a light-up maze\n- Design a light-up robot\n- Create a light-up art project\n- Build a light-up alarm clock\n- Make a light-up lamp\n- Construct a light-up game controller\n- Create a light-up phone case\n- Build a light-up keychain\n- Make a light-up piggy bank\n- Construct a light-up doorbell \n\n", "infer13b_before_item": " Output: We can build a circuit with a battery, a light bulb and a switch. \n\nInput: Input: There are many creative projects we can build at home that are related to the given theme, so let's list some of them. Theme: Circuits \n Output: Output: We can build a circuit with a battery, a light bulb and a switch. \n\nInput: Input: There are many creative projects we can build at home that are related to the given theme, so let's list some of them. Theme: Circuits \n Output: Output: We can build a circuit with a battery, a light bulb and a switch. \n\nInput: Input: There are many creative projects we can build at home that are related to the given theme, so let's list some of them. Theme: Circuits \n Output: Output: We can build a circuit with a battery, a light bulb and a switch. \n\nInput: Input: There are many creative projects we can build at home that are related to the given theme, so let's list some of them. Theme: Circuits \n Output: Output: We can", "infer13b_after_item": " \n Output: 1. Build a simple circuit using a battery, a light bulb, and a switch.\n2. Create a circuit that can be used to power a small motor.\n3. Create a circuit that can be used to control a robot.\n4. Build a circuit that can be used to control a remote-controlled car.\n5. Create a circuit that can be used to control a robotic arm.\n6. Build a circuit that can be used to control a robotic pet.\n7. Create a circuit that can be used to control a robotic vacuum cleaner.\n8. Build a circuit that can be used to control a robotic plant.\n9. Create a circuit that can be used to control a robotic security system.\n10. Build a circuit that can be used to control a robotic toy. \n\n", "infer7b_before_item": " Output: The output is the final product of the project. It is the result of the creative process. Theme: Circuits \n Process: Process: The process is the way we go from the input to the output. It is the way we go from the idea to the final product. Theme: Circuits \n\n\\end{pre}\n\n\\end{code}\n\nComment: I'm not sure what you mean by \"the output is the final product of the project\".\n\nComment: @JonathanLonowski I mean the final product of the project is the output.\n\nComment: I'm not sure what you mean by \"the output is the final product of the project\".\n\nComment: @JonathanLonowski I mean the final product of the project is the output.\n\nComment: I'm not sure what you mean by \"the final product of the project is the output\".\n\nComment: @JonathanLonowski I mean the final product of the project is the output.\n\nComment: I'm not sure what you mean by \"the final product of the project is the output\".\n\nComment: @JonathanLon", "infer7b_after_item": " \n Output: - Building a simple circuit with a battery, light bulb, and switch.\n- Building a circuit with a battery, light bulb, switch, and buzzer.\n- Building a circuit with a battery, light bulb, switch, buzzer, and motor.\n- Building a circuit with a battery, light bulb, switch, buzzer, motor, and speaker.\n- Building a circuit with a battery, light bulb, switch, buzzer, motor, speaker, and LED.\n- Building a circuit with a battery, light bulb, switch, buzzer, motor, speaker, LED, and resistor.\n- Building a circuit with a battery, light bulb, switch, buzzer, motor, speaker, LED, resistor, and capacitor.\n- Building a circuit with a battery, light bulb, switch, buzzer, motor, speaker, LED, resistor, capacitor, and transistor.\n- Building a circuit with a battery, light bulb, switch, buzzer, motor, speaker, LED, resistor, capacitor, transistor, and diode.\n- Building a circuit with a battery, light"}
+{"id": 112, "instruction": "Make a list of the materials that will be required to build the given tool.", "input": "Coffee Darkness Meter", "infer30b_before_item": " Output: Make a list of the materials that will be required to build the given tool. Coffee Darkness Meter \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n", "infer30b_after_item": " \n Output: Materials required to build the Coffee Darkness Meter:\n- Wooden frame\n- Metal ruler\n- Metal calipers\n- Metal screws\n- Metal nuts\n- Metal hinges\n- Metal springs\n- Metal dial\n- Metal shaft\n- Metal gears\n- Metal housing\n- Metal dial indicator\n- Metal pointer\n- Metal scale\n- Metal dial cover\n- Metal mounting bracket\n- Metal fasteners\n- Metal washers\n- Metal shims\n- Metal wires\n- Metal bolts\n- Metal nut drivers\n- Metal screwdrivers\n- Metal pliers\n- Metal saw\n- Metal drill\n- Metal files\n- Metal sandpaper\n- Metal paint\n- Metal primer\n- Metal sealant\n- Metal glue\n- Metal rivets\n- Metal bolts\n- Metal nuts\n- Metal washers\n- Metal screws\n- Metal nuts\n- Metal washers\n- Metal screws\n- Metal nuts\n- Metal washers\n- Metal screws\n- Metal nuts\n- Metal washers\n-", "infer13b_before_item": " Output: Make a list of the materials that will be required to build the given tool. Coffee Darkness Meter \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Materials required to build a Coffee Darkness Meter: \n- Electronic components (resistors, capacitors, transistors, etc.) \n- Circuit board \n- Battery \n- LEDs \n- Sensors \n- Wires \n- Enclosure \n- Software \n\n", "infer7b_before_item": " Output: Make a list of the materials that will be required to build the given tool. Coffee Darkness Meter\n\n\n# I worked on this challenge [by myself, with: ].\n# I spent [#] hours on this challenge.\n\n# Pseudocode\n\n# Input: Make a list of the materials that will be required to build the given tool.\n# Output: Make a list of the materials that will be required to build the given tool.\n\n# Steps:\n\n# 1. Make a list of the materials that will be required to build the given tool.\n# 2. Make a list of the materials that will be required to build the given tool.\n# 3. Make a list of the materials that will be required to build the given tool.\n# 4. Make a list of the materials that will be required to build the given tool.\n# 5. Make a list of the materials that will be required to build the given tool.\n# 6. Make a list of the materials that will be required to build the given tool.\n# 7. Make a list of the materials that will be required to build the given tool.\n#", "infer7b_after_item": " \n Output: - Wooden frame\n- Metal sheet\n- LED lights\n- Batteries\n- Wiring\n- Switches\n- Sensors\n- Display\n- Mounting brackets\n- Cable ties\n- Nuts and bolts\n- Screws\n- Glue\n- Paint \n\n"}
+{"id": 113, "instruction": "Provide a detailed explanation of the steps that are required in order to construct the given project.", "input": "Steam Engine Using an Empty Beer Can", "infer30b_before_item": " Output: Provide a detailed explanation of the steps that are required in order to construct the given project. Steam Engine Using an Empty Beer Can \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: To construct a steam engine using an empty beer can, you will need the following materials: an empty beer can, a nail, a piece of wire, a small piece of wood, a small piece of rubber, a small piece of metal, a small piece of plastic, a small piece of paper, a small piece of cloth, a small piece of cork, a small piece of wax, a small piece of string, and a small piece of paper. \n\nFirst, you will need to make a hole in the side of the beer can with the nail. Then, you will need to insert the wire through the hole and attach the small piece of wood to the end of the wire. Next, you will need to attach the rubber to the other end of the wire and attach the metal to the rubber. Then, you will need to attach the plastic to the metal, the paper to the plastic, the cloth to the paper, the cork to the cloth, the wax to the cork, and the string to the wax. Finally, you will need to attach the paper to the string and place the cork in the hole in the beer can.", "infer13b_before_item": " Output: Provide a detailed explanation of the steps that are required in order to construct the given project. Steam Engine Using an Empty Beer Can \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: To construct a steam engine using an empty beer can, you will need the following materials: a beer can, a straw, a rubber band, a penny, a paperclip, and a small piece of aluminum foil. \n\nFirst, cut the straw to the desired length and then insert the penny into the end of the straw. Next, wrap the rubber band around the straw and the penny, leaving a small gap between the two. Finally, place the aluminum foil over the gap and secure it with the paperclip. \n\nTo use the steam engine, heat the beer can with a lighter or a match until the can is hot. Then, place the straw with the penny and rubber band into the can and quickly remove it. The steam will push the penny and rubber band up the straw, creating a steam engine. \n\n", "infer7b_before_item": " Output: Provide a detailed explanation of the steps that are required in order to construct the given project. Steam Engine Using an Empty Beer Can \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Step 1: Cut the top of the beer can off using a pair of scissors.\n\nStep 2: Cut a small hole in the side of the can using a sharp object such as a knife or screwdriver.\n\nStep 3: Cut a small piece of cardboard and place it inside the can.\n\nStep 4: Cut a small piece of aluminum foil and place it over the cardboard.\n\nStep 5: Cut a small piece of plastic wrap and place it over the aluminum foil.\n\nStep 6: Cut a small piece of rubber and place it over the plastic wrap.\n\nStep 7: Cut a small piece of string and place it over the rubber.\n\nStep 8: Cut a small piece of paper and place it over the string.\n\nStep 9: Cut a small piece of tape and place it over the paper.\n\nStep 10: Place the can in a pot of boiling water and let it boil for 10 minutes.\n\nStep 11: Remove the can from the pot and"}
+{"id": 114, "instruction": "Estimate the cost of supplies required to create the given project and calculate the total cost.", "input": "DIY Water Chiller\nRequired items are: Ice-maker, Icebox/cooler, Temperature regulator, Aquarium pump x2, 12V power supply", "infer30b_before_item": " Output: Estimate the cost of supplies required to create the given project and calculate the total cost. DIY Water Chiller\nRequired items are: Ice-maker, Icebox/cooler, Temperature regulator, Aquarium pump x2, 12V power supply \n  */\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": ", 12V battery, 12V inverter, 12V LED light, 12V fan, 12V water pump, 12V water heater, 12V water filter, 12V water chiller, 12V water cooler, 12V water dispenser, 12V water fountain, 12V water sprayer, 12V water mister, 12V water jet, 12V water spout, 12V water spray gun, 12V water sprayer, 12V water sprayer, 12V water sprayer, 12V water sprayer, 12V water sprayer, 12V water sprayer, 12V water sprayer, 12V water sprayer, 12V water sprayer, 12V water sprayer, 12V water sprayer, 12V water sprayer, 12V water sprayer, 12V water sprayer, 12V water sp", "infer13b_before_item": " Output: Calculate the total cost of the required items.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": ", 12V DC pump, 12V DC fan, 12V DC light, 12V DC timer. \n Output: The estimated cost of supplies required to create the DIY Water Chiller is $150. This includes the cost of the ice-maker ($30), icebox/cooler ($20), temperature regulator ($20), aquarium pump x2 ($20), 12V power supply ($10), 12V DC pump ($15), 12V DC fan ($10), 12V DC light ($5), and 12V DC timer ($10). \n\n", "infer7b_before_item": " Output: Calculate the cost of the project and the total cost.\n\nComment: Please edit your question to include the code you have so far.\n\nAnswer: You can use the following code to calculate the cost of the project.\n\n\\begin{code}\nimport math\n\ndef cost(items):\n    total = 0\n    for item in items:\n        total += math.pow(item, 2)\n    return total\n\ndef cost_of_project(items):\n    return cost(items)\n\ndef total_cost(items):\n    return cost(items) * 2\n\ndef main():\n    items = [10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 1", "infer7b_after_item": ", 12V DC fan x2, 12V DC relay, 12V DC switch, 12V DC fuse, 12V DC power cable x2, 12V DC power adapter x2, 12V DC power connector x2, 12V DC power plug x2, 12V DC power cord x2, 12V DC power plug x2, 12V DC power cord x2, 12V DC power plug x2, 12V DC power cord x2, 12V DC power plug x2, 12V DC power cord x2, 12V DC power plug x2, 12V DC power cord x2, 12V DC power plug x2, 12V DC power cord x2, 12V DC power plug x2, 12V DC power cord x2, 12V DC power plug x2, 12V DC power cord x2, 12V DC power plug x2, 12V DC power cord x2, 12V DC power plug x2, 12"}
+{"id": 115, "instruction": "Answer the following literature question as accurately as possible based on your knowledge of literature for high school students.", "input": "What does the green light symbolize at the end of the book?", "infer30b_before_item": " Output: Answer the following literature question as accurately as possible based on your knowledge of literature for high school students. What does the green light symbolize at the end of the book?\n\nComment: I'm not sure what you're asking.  Are you asking how to parse the input?  Or how to write the output?  Or how to write the code that does both?\n\nComment: I'm asking how to write the code that does both.\n\nComment: I'm not sure what you mean by \"do both\".  You've already written the code that does both.  You've written the code that reads the input and writes the output.  What's the problem?\n\nComment: I'm asking how to write the code that does both.\n\nComment: I'm asking how to write the code that does both.\n\nComment: I'm asking how to write the code that does both.\n\nComment: I'm asking how to write the code that does both.\n\nComment: I'm asking how to write the code that does both.\n\nComment: I'm asking how to write the code that does both.\n\nComment:", "infer30b_after_item": " \n Output: The green light at the end of the book symbolizes hope, opportunity, and the American Dream. It is a symbol of the future and the possibility of a better life. \n\n", "infer13b_before_item": " Output: The output of the program is the answer to the question.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: In F. Scott Fitzgerald's The Great Gatsby, the green light at the end of the dock symbolizes Gatsby's dream of reuniting with his lost love, Daisy Buchanan. It is a symbol of hope and longing, representing Gatsby's desire to recapture the past and his unfulfilled dreams. \n\n", "infer7b_before_item": " Output: Answer the following literature question as accurately as possible based on your knowledge of literature for high school students. What does the green light symbolize at the end of the book? \n 1. What does the green light symbolize at the end of the book?\n 2. What does the green light symbolize at the end of the book?\n 3. What does the green light symbolize at the end of the book?\n 4. What does the green light symbolize at the end of the book?\n 5. What does the green light symbolize at the end of the book?\n 6. What does the green light symbolize at the end of the book?\n 7. What does the green light symbolize at the end of the book?\n 8. What does the green light symbolize at the end of the book?\n 9. What does the green light symbolize at the end of the book?\n 10. What does the green light symbolize at the end of the book?\n 11. What does the green light symbolize at the end of the book?\n 12. What does the green light symbolize at the end of the book", "infer7b_after_item": " \n Output: The green light symbolizes hope, life, and the future. It represents the possibility of a new beginning and a fresh start. It also represents the protagonist's desire to move forward and take control of his life. \n\n"}
+{"id": 116, "instruction": "Develop a mental exercise that can help people manage their anxiety and explain how it works.", "input": "", "infer30b_before_item": " Output: A mental exercise that can help people manage their anxiety.\nInput: Input: Develop a mental exercise that can help people manage their anxiety and explain how it works.  \n Output: Output: A mental exercise that can help people manage their anxiety.\nInput: Input: Develop a mental exercise that can help people manage their anxiety and explain how it works.  \n Output: Output: A mental exercise that can help people manage their anxiety.\nInput: Input: Develop a mental exercise that can help people manage their anxiety and explain how it works.  \n Output: Output: A mental exercise that can help people manage their anxiety.\nInput: Input: Develop a mental exercise that can help people manage their anxiety and explain how it works.  \n Output: Output: A mental exercise that can help people manage their anxiety.\nInput: Input: Develop a mental exercise that can help people manage their anxiety and explain how it works.  \n Output: Output: A mental exercise that can help people manage their anxiety.\nInput: Input: Develop a mental exercise that can help people manage their anxiety and explain how it works.  \n Output: Output: A mental exercise that can help people manage", "infer30b_after_item": "\n Output: A mental exercise that can help people manage their anxiety is called mindfulness meditation. Mindfulness meditation is a practice that helps people become more aware of their thoughts and feelings without judging them. It helps people to focus on the present moment and to accept their thoughts and feelings without trying to change them. This can help to reduce anxiety by helping people to become more aware of their thoughts and feelings and to be able to manage them in a healthy way. \n\n", "infer13b_before_item": " Output: Develop a mental exercise that can help people manage their anxiety and explain how it works.  \n | **Inferences:** What is the author's purpose in writing this text? What is the author's point of view? What is the author's tone? What is the author's attitude? What is the author's purpose in writing this text? What is the author's point of view? What is the author's tone? What is the author's attitude?  \n---|---|---  \n\u2717 | A To explain how to manage anxiety  |\nThe author does not explain how to manage anxiety. The author does not even explain what anxiety is.\n\u2713 | B To explain how to manage anxiety through mental exercises  |\n**Correct.** The author explains how to manage anxiety through mental exercises.\n\u2717 | C To explain how to manage anxiety through physical exercises  |\nThe author does not explain how to manage anxiety through physical exercises.\n\u2717 | D To explain how to manage anxiety through medication  |\nThe author does not explain how to manage anxiety through medic", "infer13b_after_item": "\n Output: One mental exercise that can help people manage their anxiety is to practice mindful breathing. Mindful breathing involves focusing on your breath and taking deep breaths in and out. This helps to slow down your breathing and relax your body. It also helps to bring your attention to the present moment and away from worries and anxieties. \n\n", "infer7b_before_item": " Output: Develop a mental exercise that can help people manage their anxiety and explain how it works.\nThe following is a list of the most common mental exercises used to help people manage their anxiety.\n  1. **_Mental exercises that help people manage their anxiety_**\n    * **_Mental exercises that help people manage their anxiety_**\n    * **_Mental exercises that help people manage their anxiety_**\n    * **_Mental exercises that help people manage their anxiety_**\n    * **_Mental exercises that help people manage their anxiety_**\n    * **_Mental exercises that help people manage their anxiety_**\n    * **_Mental exercises that help people manage their anxiety_**\n    * **_Mental exercises that help people manage their anxiety_**\n    * **_Mental exercises that help people manage their anxiety_**\n    * **_Mental exercises that help people manage their anxiety_**\n    * **_Mental exercises that help people manage their anxiety_**\n    *", "infer7b_after_item": "\n Output: The mental exercise involves focusing on your breathing and counting each inhalation and exhalation. This helps to slow down your breathing and bring your attention to the present moment. By focusing on your breathing, you can become more aware of your thoughts and feelings, and learn to recognize when your thoughts are causing you to feel anxious. This can help you to become more mindful and better able to manage your anxiety. \n\n"}
+{"id": 117, "instruction": "Give me an idea of the mental exercises you do to inspire yourself.", "input": "", "infer30b_before_item": " Output: I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think about the future. I think", "infer30b_after_item": "\n Output: I like to do mental exercises such as visualization, positive affirmations, and creative problem solving. Visualization helps me to imagine the future I want to create and to stay motivated. Positive affirmations help me to stay focused on my goals and to stay positive. Creative problem solving helps me to think outside the box and to come up with creative solutions to problems. \n\n", "infer13b_before_item": " Output: I'm going to give you a list of things that I do.  \n Input: Input: What do you do?  \n Output: Output: I'm going to give you a list of things that I do.  \n Input: Input: What do you do?  \n Output: Output: I'm going to give you a list of things that I do.  \n Input: Input: What do you do?  \n Output: Output: I'm going to give you a list of things that I do.  \n Input: Input: What do you do?  \n Output: Output: I'm going to give you a list of things that I do.  \n Input: Input: What do you do?  \n Output: Output: I'm going to give you a list of things that I do.  \n Input: Input: What do you do?  \n Output: Output: I'm going to give you a list of things that I do.  \n Input: Input: What do you do?  \n Output: Output: I'm going to give you a list of things that I do.  \n Input: Input: What do you do?  \n Output:", "infer13b_after_item": "\n Output: I like to do mental exercises to inspire myself. I like to read books and articles that are related to my interests and goals. I also like to watch documentaries and TED talks to learn more about the topics I am interested in. I also like to listen to podcasts and music that can help me stay motivated and focused. \n\n", "infer7b_before_item": " Output: What are the results of your mental exercises?  \n Feedback: Feedback: What are the results of your mental exercises?  \n Evaluation: Evaluation: What are the results of your mental exercises?\nThe following are some of the mental exercises that I do to inspire myself.\n1. I read a lot of books. I read books on a variety of subjects. I read books on business, history, science, and philosophy. I read books on the lives of great people. I read books on the lives of great leaders. I read books on the lives of great entrepreneurs. I read books on the lives of great scientists. I read books on the lives of great philosophers. I read books on the lives of great artists. I read books on the lives of great athletes. I read books on the lives of great inventors. I read books on the lives of great explorers. I read books on the lives of great writers. I read books on the lives of great thinkers. I read books on the lives of great achievers. I read books on the lives of great visionaries. I read books on the lives of great", "infer7b_after_item": "\n Output: I do a variety of mental exercises to inspire myself. I like to write down my goals and visualize myself achieving them. I also like to read books and articles that inspire me and help me stay motivated. I also like to take time to reflect on my progress and celebrate my successes. \n\n"}
+{"id": 118, "instruction": "Make a list of outdoor activities that are suitable for the given weather condition.", "input": "Temperature: 57 F, Showers", "infer30b_before_item": " Output: Make a list of outdoor activities that are suitable for the given weather condition. Temperature: 57 F, Showers \n Expected Output: Output: Make a list of outdoor activities that are suitable for the given weather condition. Temperature: 57 F, Showers \n */\n\npublic class _01_WeatherForecast {\n    public static void main(String[] args) {\n        Scanner scanner = new Scanner(System.in);\n        String[] input = scanner.nextLine().split(\" \");\n        int temperature = Integer.parseInt(input[0]);\n        String weather = input[1];\n\n        if (temperature >= 50 && temperature <= 70 && weather.equals(\"Sunny\")) {\n            System.out.println(\"Enjoy the weather with a nice walk in the park\");\n        } else if (temperature >= 50 && temperature <= 70 && weather.equals(\"Cloudy\")) {\n            System.out.println(\"Enjoy the weather with a nice walk in the park\");\n        } else if (temperature >= 50 && temperature <= 7", "infer30b_after_item": "\nNo Input \n Output: Outdoor activities suitable for 57 F and showers include hiking, biking, kayaking, canoeing, fishing, birdwatching, and camping. \n\n", "infer13b_before_item": " Output: Make a list of outdoor activities that are suitable for the given weather condition. Temperature: 57 F, Showers \n\n\nclass Solution {\npublic:\n    vector<string> getOutdoorActivities(vector<int>& temperature, vector<string>& weather) {\n        vector<string> res;\n        for (int i = 0; i < temperature.size(); ++i) {\n            if (temperature[i] < 30) {\n                res.push_back(\"Snow\");\n            } else if (temperature[i] < 40) {\n                res.push_back(\"Rain\");\n            } else if (temperature[i] < 50) {\n                res.push_back(\"Cloudy\");\n            } else if (temperature[i] < 60) {\n                res.push_back(\"Sunny\");\n            } else {\n                res.push_back(\"Sunny\");\n            }\n        }\n        return res;\n    }\n};\n", "infer13b_after_item": " \n Output: 1. Hiking\n2. Picnicking\n3. Fishing\n4. Birdwatching\n5. Camping\n6. Biking\n7. Swimming\n8. Kite flying\n9. Geocaching\n10. Gardening\n11. Playing Frisbee\n12. Playing tag\n13. Playing hide and seek\n14. Playing soccer\n15. Playing badminton\n16. Playing volleyball\n17. Playing horseshoes\n18. Playing croquet\n19. Playing bocce ball\n20. Playing shuffleboard \n\n", "infer7b_before_item": " Make a list of outdoor activities that are suitable for the given weather condition. Temperature: 57 F, Showers\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: - Hiking\n- Biking\n- Picnicking\n- Camping\n- Fishing\n- Birdwatching\n- Gardening\n- Swimming\n- Boating\n- Kayaking\n- Rollerblading\n- Skateboarding\n- Frisbee\n- Volleyball\n- Soccer\n- Tennis\n- Golf\n- Horseback riding\n- BMX riding\n- Skateboarding\n- Surfing\n- Rock climbing\n- Geocaching\n- Disc golf\n- Discussion groups\n- Astronomy\n- Photography\n- Bird watching\n- Gardening\n- Camping\n- Picnicking\n- Fishing\n- Birdwatching\n- Gardening\n- Swimming\n- Boating\n- Kayaking\n- Rollerblading\n- Skateboarding\n- Frisbee\n- Volleyball\n- Soccer\n- Tennis\n- Golf\n- Horseback riding\n- BMX riding\n- Skateboarding\n- Surfing\n- Rock climbing\n- Geocaching\n- Disc golf\n- Discussion groups\n- Astronom"}
+{"id": 119, "instruction": "Please provide us with a list of the best conservation starter questions related to the weather.", "input": "", "infer30b_before_item": " Output: Please provide us with a list of the best conservation starter questions related to the weather.\nInput: Input: Please provide us with a list of the best conservation starter questions related to the weather.  \n Output: Output: Please provide us with a list of the best conservation starter questions related to the weather.\nInput: Input: Please provide us with a list of the best conservation starter questions related to the weather.  \n Output: Output: Please provide us with a list of the best conservation starter questions related to the weather.\nInput: Input: Please provide us with a list of the best conservation starter questions related to the weather.  \n Output: Output: Please provide us with a list of the best conservation starter questions related to the weather.\nInput: Input: Please provide us with a list of the best conservation starter questions related to the weather.  \n Output: Output: Please provide us with a list of the best conservation starter questions related to the weather.\nInput: Input: Please provide us with a list of the best conservation starter questions related to the weather.  \n Output: Output: Please provide us with a list of the best conservation starter questions related to the weather.\n", "infer30b_after_item": "\n Output: - What are the effects of climate change on weather patterns?\n- How can we reduce the impact of extreme weather events?\n- What are the best ways to conserve energy and reduce emissions?\n- How can we use renewable energy sources to reduce our reliance on fossil fuels?\n- What are the best ways to reduce air pollution?\n- How can we reduce the amount of water we use?\n- What are the best ways to protect our oceans and marine life?\n- How can we reduce the amount of waste we produce?\n- What are the best ways to conserve natural resources?\n- How can we reduce the amount of deforestation?\n- What are the best ways to reduce our carbon footprint?\n- How can we encourage sustainable practices in our communities?\n- What are the best ways to reduce our reliance on single-use plastics?\n- How can we reduce the amount of chemicals we use in our homes?\n- What are the best ways to protect our soil and agricultural land?\n- How can we reduce the amount of noise pollution?\n- What are the best ways to protect our wildlife?", "infer13b_before_item": " Output: Please provide us with a list of the best conservation starter questions related to the weather.  \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: 1. What is the weather like today?\n2. How has the weather been changing over the past few days?\n3. How does the weather affect the environment?\n4. What are some ways to conserve energy and reduce our carbon footprint?\n5. How can we reduce our energy consumption and save money?\n6. What are some ways to reduce our water usage?\n7. How can we reduce our waste and recycle more?\n8. What are some ways to protect the environment from pollution?\n9. What are some ways to protect the environment from climate change?\n10. What are some ways to reduce our carbon footprint? \n\n", "infer7b_before_item": " Please provide us with a list of the best conservation starter questions related to the weather.\nInput: Input: Please provide us with a list of the best conservation starter questions related to the weather.  \n Output: Please provide us with a list of the best conservation starter questions related to the weather.\nInput: Input: Please provide us with a list of the best conservation starter questions related to the weather.  \n Output: Please provide us with a list of the best conservation starter questions related to the weather.\nInput: Input: Please provide us with a list of the best conservation starter questions related to the weather.  \n Output: Please provide us with a list of the best conservation starter questions related to the weather.\nInput: Input: Please provide us with a list of the best conservation starter questions related to the weather.  \n Output: Please provide us with a list of the best conservation starter questions related to the weather.\nInput: Input: Please provide us with a list of the best conservation starter questions related to the weather.  \n Output: Please provide us with a list of the best conservation starter questions related to the weather.\nInput: Input: Please provide us with a list of the", "infer7b_after_item": "\n Output: 1. What are the different types of weather?\n2. How does the weather affect the environment?\n3. What are the causes of weather patterns?\n4. How can we protect the environment from weather-related damage?\n5. What are the effects of climate change on the environment?\n6. How can we reduce our carbon footprint to help protect the environment?\n7. What are the benefits of renewable energy sources?\n8. How can we reduce our energy consumption?\n9. What are the effects of air pollution on the environment?\n10. What are the effects of water pollution on the environment?\n11. How can we reduce our waste production?\n12. What are the benefits of recycling?\n13. What are the effects of deforestation on the environment?\n14. What are the effects of overfishing on the environment?\n15. What are the effects of overpopulation on the environment?\n16. What are the effects of urbanization on the environment?\n17. What are the effects of industrialization on the environment?\n18. What are the effects of agriculture on the environment?"}
+{"id": 120, "instruction": "In relation to the given weather scenario, give some tips on how to adjust the travel plans with it.", "input": "a sudden temperature change", "infer30b_before_item": " Output: In relation to the given weather scenario, give some tips on how to adjust the travel plans with it. a sudden temperature change \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n ", "infer30b_after_item": " from 25\u00b0C to 10\u00b0C \n Output: When the temperature suddenly drops from 25\u00b0C to 10\u00b0C, it is important to adjust your travel plans accordingly. Make sure to bring warm clothing such as a coat, hat, and gloves. If possible, try to avoid outdoor activities such as hiking or biking, as the cold temperature can be dangerous. Additionally, make sure to check the weather forecast before leaving, as the temperature may drop even further. \n\n", "infer13b_before_item": " Output: In relation to the given weather scenario, give some tips on how to adjust the travel plans with it. a sudden temperature change \n\n\nclass Solution:\n    # @param {integer[]} temps, an integer array\n    # @return {integer[]}\n    def adjust_travel_plan(self, temps):\n        n = len(temps)\n        if n == 0:\n            return []\n        if n == 1:\n            return [0]\n        if temps[0] == temps[-1]:\n            return [0]\n        if temps[0] == 0 and temps[-1] == 0:\n            return [0]\n        if temps[0] == 0 and temps[-1] == 1:\n            return [1]\n        if temps[0] == 1 and temps[-1] == 0:\n            return [1]\n        if temps[0] == 1 and temps[-1] == 1:\n            return [0]\n        if temps[0] == 2 and temps[-1] == 2:\n            return [0]\n        if temps[0] ==", "infer13b_after_item": " from 20 to 30 degrees Celsius \n Output: In this situation, it is important to be prepared for the sudden temperature change. It is recommended to bring layers of clothing that can be easily adjusted, such as a light jacket or sweater. Additionally, it is important to stay hydrated and to wear sunscreen if the temperature is higher than usual. It is also important to check the weather forecast before leaving and to be aware of any potential hazards that may arise due to the sudden temperature change. \n\n", "infer7b_before_item": " Output: In relation to the given weather scenario, give some tips on how to adjust the travel plans with it. a sudden temperature change \n\n\"\"\"\n\n\ndef weather_scenario(weather_scenario):\n    \"\"\"\n    :param weather_scenario:\n    :return:\n    \"\"\"\n    if weather_scenario == \"sudden temperature change\":\n        return \"a sudden temperature change\"\n    else:\n        return \"a sudden temperature change\"\n\n\ndef weather_scenario_input():\n    \"\"\"\n    :return:\n    \"\"\"\n    weather_scenario = input(\"Input: In relation to the given weather scenario, give some tips on how to adjust the travel plans with it. a sudden temperature change\\n\")\n    return weather_scenario\n\n\ndef weather_scenario_output():\n    \"\"\"\n    :return:\n    \"\"\"\n    weather_scenario = input(\"Output: In relation to the given weather scenario, give some tips on how to adjust the travel plans with it. a sudden temperature change\\n\")\n    return weather_scenario\n\n\nif __name__ == \"__main__\":\n    weather", "infer7b_after_item": " of 10 degrees Celsius \n Output: In the given scenario, it is important to adjust travel plans accordingly. It is recommended to check the weather forecast and plan accordingly. If the temperature is expected to drop, it is best to bring an extra layer of clothing and wear layers that can be removed or added as needed. Additionally, it is important to check the weather conditions before leaving and be aware of any potential hazards such as slippery roads or icy patches. \n\n"}
+{"id": 121, "instruction": "Write a to-do list based on the given information.", "input": "Daily tasks in the pharmacy", "infer30b_before_item": " Output: Write a to-do list based on the given information. Daily tasks in the pharmacy \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n ", "infer30b_after_item": " include restocking shelves, answering customer questions, and filing paperwork. \n Output: To-do list for the pharmacy: \n1. Restock shelves\n2. Answer customer questions\n3. File paperwork\n4. Check inventory levels\n5. Reorder supplies as needed\n6. Clean and organize the pharmacy\n7. Update customer records\n8. Process prescriptions\n9. Contact doctors as needed\n10. Check expiration dates on medications \n\n", "infer13b_before_item": " Output: Write a to-do list based on the given information. Daily tasks in the pharmacy \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: 1. Check stock of medicines and supplies.\n2. Refill prescriptions.\n3. Answer customer inquiries.\n4. Process insurance claims.\n5. Update patient records.\n6. Organize inventory.\n7. Maintain cleanliness of the pharmacy.\n8. Monitor expiration dates of medicines.\n9. Check for any expired medicines.\n10. Update pharmacy software.\n11. Check for any outdated medicines.\n12. Update pharmacy website.\n13. Check for any discrepancies in billing.\n14. Check for any damaged medicines.\n15. Check for any expired medicines.\n16. Check for any expired medicines.\n17. Check for any expired medicines.\n18. Check for any expired medicines.\n19. Check for any expired medicines.\n20. Check for any expired medicines. \n\n", "infer7b_before_item": " Output: Write a to-do list based on the given information. Daily tasks in the pharmacy \n\n*/\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: - Check inventory\n- Refill prescriptions\n- Process orders\n- Prepare medications\n- Clean the pharmacy\n- Answer customer questions\n- Maintain records\n- Clean the dispensary\n- Clean the lab\n- Clean the storage area\n- Clean the waiting area\n- Clean the pharmacy counter\n- Clean the pharmacy equipment\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- Clean the pharmacy supplies\n- C"}
+{"id": 122, "instruction": "Design a template table for keeping track of all subscriptions.", "input": "", "infer30b_before_item": " Output: Print the table.\n\n*/\n\n#include <stdio.h>\n#include <stdlib.h>\n\nint main()\n{\n    int i, j, k, n, m, c, d, e, f, g, h, l, o, p, q, r, s, t, u, v, w, x, y, z;\n    char a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p, q, r, s, t, u, v, w, x, y, z;\n    int *p;\n    char *q;\n    int **arr;\n    char **arr1;\n    int **arr2;\n    char **arr3;\n    int **arr4;\n    char **arr5;\n    int **arr6;\n    char **arr7;\n    int **arr8;\n    char **arr9;\n    int **arr10;\n    char **arr11;\n    int **arr12;\n    char **arr13;\n    int", "infer30b_after_item": "\n Output: Subscription | Service Provider | Price | Date Started | Date Ended |\n----------- | -------------- | ----- | ---------- | ---------- |\nNetflix | Netflix Inc. | $12.99 | 01/01/2021 | 12/31/2021 |\nHulu | Hulu LLC | $5.99 | 04/01/2021 | 03/31/2022 |\nSpotify | Spotify Ltd. | $9.99 | 05/01/2021 | 04/30/2022 |\nApple Music | Apple Inc. | $4.99 | 06/01/2021 | 05/31/2022 |\nAmazon Prime | Amazon.com Inc. | $11.99 | 07/01/2021 | 06/30/2022 |\nGoogle Play | Google LLC | $3.99", "infer13b_before_item": " Output: A table with all subscriptions.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n\n\n \n \n\n\n\n\n \n \n\n\n\n  The The The The The 190", "infer13b_after_item": "\n Output: The table should include columns for the name of the subscription, the cost, the frequency of payment, the start date, the end date, and the payment method. \n\n", "infer7b_before_item": " Output: A table that can be used to keep track of all subscriptions.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": "\n Output: Template Table:\n\n| Subscription Name | Subscription Type | Subscription Price | Subscription Duration | Subscription Frequency | Subscription Cancellation Policy |\n|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------"}
+{"id": 123, "instruction": "Plan an agenda for the daily standup meetings. Assign a time duration to each topic on the agenda.", "input": "", "infer30b_before_item": " Output: A list of topics and their time duration.\nInput: Input: Plan an agenda for the daily standup meetings. Assign a time duration to each topic on the agenda.  \n Output: Output: A list of topics and their time duration.\nInput: Input: Plan an agenda for the daily standup meetings. Assign a time duration to each topic on the agenda.  \n Output: Output: A list of topics and their time duration.\nInput: Input: Plan an agenda for the daily standup meetings. Assign a time duration to each topic on the agenda.  \n Output: Output: A list of topics and their time duration.\nInput: Input: Plan an agenda for the daily standup meetings. Assign a time duration to each topic on the agenda.  \n Output: Output: A list of topics and their time duration.\nInput: Input: Plan an agenda for the daily standup meetings. Assign a time duration to each topic on the agenda.  \n Output: Output: A list of topics and their time duration.\nInput: Input: Plan an agenda for the daily standup meetings. Assign a", "infer30b_after_item": "\n Output: The daily standup meetings should start at 9:30am and last for 15 minutes. The agenda should include the following topics and time duration:\n\n- Welcome and Introductions (1 minute)\n- Review of Yesterday's Work (5 minutes)\n- Today's Work (5 minutes)\n- Blockers and Issues (3 minutes)\n- Wrap-up (2 minutes) \n\n", "infer13b_before_item": " Output: A list of topics to be discussed in the daily standup meetings.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: The daily standup meetings should have a set agenda with a time duration for each topic. The topics should include a quick recap of the progress made on the current project, any issues encountered, and any upcoming tasks. The meeting should also include a quick check-in with each team member to ensure everyone is on the same page. The meeting should last no more than 15 minutes. \n\n", "infer7b_before_item": " Output: The daily standup meetings are held on time and the agenda is followed.\nInput: Input: Plan the sprint backlog. Assign a time duration to each item on the backlog.  \n Output: Output: The sprint backlog is completed on time and the items are completed in the order of their priority.\nInput: Input: Plan the sprint work. Assign a time duration to each task on the work.  \n Output: Output: The sprint work is completed on time and the tasks are completed in the order of their priority.\nInput: Input: Plan the sprint review. Assign a time duration to each item on the review.  \n Output: Output: The sprint review is completed on time and the items are completed in the order of their priority.\nInput: Input: Plan the sprint retrospective. Assign a time duration to each item on the retrospective.  \n Output: Output: The sprint retrospective is completed on time and the items are completed in the order of their priority.\nInput: Input: Plan the sprint demo. Assign a time duration to each item on the demo.  \n Output: Output: The sprint", "infer7b_after_item": "\n Output: 1. Introductions \n2. Status Update \n3. Blockers \n4. Next Steps \n5. Questions \n6. Wrap Up \n\n"}
+{"id": 124, "instruction": "Answer the following question.", "input": "How do you say \"good evening\" in French.", "infer30b_before_item": " Output: Answer the following question. How do you say \"good evening\" in French.\n\nInput: Input: Answer the following question. How do you say \"good evening\" in French. \n Output: Output: Answer the following question. How do you say \"good evening\" in French.\n\nInput: Input: Answer the following question. How do you say \"good evening\" in French. \n Output: Output: Answer the following question. How do you say \"good evening\" in French.\n\nInput: Input: Answer the following question. How do you say \"good evening\" in French. \n Output: Output: Answer the following question. How do you say \"good evening\" in French.\n\nInput: Input: Answer the following question. How do you say \"good evening\" in French. \n Output: Output: Answer the following question. How do you say \"good evening\" in French.\n\nInput: Input: Answer the following question. How do you say \"good evening\" in French. \n Output: Output: Answer the following question. How do you say \"good evening\" in French.\n\nInput: Input: Answer the following question. How do you say \"good evening", "infer30b_after_item": " \n Output: In French, \"good evening\" is \"Bonsoir\". \n\n", "infer13b_before_item": " Output: Answer the following question. How do you say \"good evening\" in French. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Bonsoir. \n\n", "infer7b_before_item": " Output: Write the answer in French.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n    // \n// \n// ", "infer7b_after_item": " \n Output: Bonsoir. \n\n"}
+{"id": 125, "instruction": "What sound does this make?", "input": "\u597d", "infer30b_before_item": " Output: What sound does this make? \u597d\n\nInput: Input: What sound does this make? \u54c8 \n Output: Output: What sound does this make? \u54c8\n\nInput: Input: What sound does this make? \u54c8\u54c8 \n Output: Output: What sound does this make? \u54c8\u54c8\n\nInput: Input: What sound does this make? \u54c8\u54c8\u54c8 \n Output: Output: What sound does this make? \u54c8\u54c8\u54c8\n\nInput: Input: What sound does this make? \u54c8\u54c8\u54c8\u54c8 \n Output: Output: What sound does this make? \u54c8\u54c8\u54c8\u54c8\n\nInput: Input: What sound does this make? \u54c8\u54c8\u54c8\u54c8\u54c8 \n Output: Output: What sound does this make? \u54c8\u54c8\u54c8\u54c8\u54c8\n\nInput: Input: What sound does this make? \u54c8\u54c8\u54c8\u54c8\ufffd", "infer30b_after_item": "\u542c \n Output: This sound is pleasant. \n\n", "infer13b_before_item": " Output: What sound does this make? \u597d \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n\n\n\n\n\n \n \n \n\n\n\n \n \n\n\n  \n  \n\n\n  \n\n\n  \n \n\n  \n  \n ", "infer13b_after_item": "\u597d \n Output: This makes the sound \"ha ha ha\". \n\n", "infer7b_before_item": " Output: What sound does this make? \u597d \n\nInput: Input: What sound does this make? \u597d \n Output: Output: What sound does this make? \u597d \n\nInput: Input: What sound does this make? \u597d \n Output: Output: What sound does this make? \u597d \n\nInput: Input: What sound does this make? \u597d \n Output: Output: What sound does this make? \u597d \n\nInput: Input: What sound does this make? \u597d \n Output: Output: What sound does this make? \u597d \n\nInput: Input: What sound does this make? \u597d \n Output: Output: What sound does this make? \u597d \n\nInput: Input: What sound does this make? \u597d \n Output: Output: What sound does this make? \u597d \n\nInput: Input: What sound does this make? \u597d \n Output: Output: What sound does this make? \u597d \n\nInput: Input: What sound does this make? \u597d \n Output: Output: What sound does this make? \u597d \n\nInput: Input: What sound does this make? ", "infer7b_after_item": " \n Output: \u54c7 \n\n"}
+{"id": 126, "instruction": "Give a grammar tip on how to end a German verb.", "input": "", "infer30b_before_item": " Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a", "infer30b_after_item": "\n Output: In German, verbs are usually followed by a conjugated ending that indicates the person, number, and tense of the verb. The most common endings are -en, -st, and -t. For example, the verb \"gehen\" (to go) in the present tense would be conjugated as \"gehe-n\". \n\n", "infer13b_before_item": " Output: Give a German verb ending.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n\n\n\n\n The The The The The The 1900s\n\n\n\n\n\n", "infer13b_after_item": "\n Output: In German, verbs are conjugated to show the tense and person of the verb. To end a verb, add the appropriate ending to the verb according to the tense and person. For example, to end a present tense verb in the singular, add -e; for a past tense verb in the singular, add -te; and for a future tense verb in the singular, add -e. \n\n", "infer7b_before_item": " Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a German verb.\nInput: Input: Give a grammar tip on how to end a German verb.  \n Output: Output: Give a grammar tip on how to end a", "infer7b_after_item": "\n Output: One grammar tip for ending a German verb is to use the appropriate ending for the verb type. For example, if the verb is in the infinitive form, the ending will be -en. If the verb is in the present tense, the ending will be -t. If the verb is in the past tense, the ending will be -te. If the verb is in the future tense, the ending will be -t. \n\n"}
+{"id": 127, "instruction": "This is a test for the reading ability of French. Read the given story carefully, and then answer the question in the end.", "input": "Lucie est \u00e0 la maison avec sa petite-fille, Lin.\nLucie: Oh, non! Nous n'avons pas de pain!\nLin: Ah! Tu vas au supermarch\u00e9?\nLucie: Oui, je veux une baguette pour mon sandwich.\nDo Lin and Lucie have a lot of bread?", "infer30b_before_item": " Output: This is a test for the reading ability of French. Read the given story carefully, and then answer the question in the end. Lucie est \u00e0 la maison avec sa petite-fille, Lin.\nLucie: Oh, non! Nous n'avons pas de pain!\nLin: Ah! Tu vas au supermarch\u00e9?\nLucie: Oui, je veux une baguette pour mon sandwich.\nDo Lin and Lucie have a lot of bread? \n Expected Output: Expected Output: This is a test for the reading ability of French. Read the given story carefully, and then answer the question in the end. Lucie est \u00e0 la maison avec sa petite-fille, Lin.\nLucie: Oh, non! Nous n'avons pas de pain!\nLin: Ah! Tu vas au supermarch\u00e9?\nLucie: Oui, je veux une baguette pour mon sandwich.\nDo Lin and Lucie have a lot of bread? \n Your Output: Your Output: This is a test for the reading ability of French. Read the given story carefully, and then answer the question in the end.", "infer30b_after_item": " \n Output: No, they do not have a lot of bread. \n\n", "infer13b_before_item": " Output: This is a test for the reading ability of French. Read the given story carefully, and then answer the question in the end. Lucie est \u00e0 la maison avec sa petite-fille, Lin.\nLucie: Oh, non! Nous n'avons pas de pain!\nLin: Ah! Tu vas au supermarch\u00e9?\nLucie: Oui, je veux une baguette pour mon sandwich.\nDo Lin and Lucie have a lot of bread? \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Oui, ils ont une baguette pour leur sandwich. \n\n", "infer7b_before_item": " Output: This is a test for the writing ability of French. Write the given story in French.\nLucie: Oh, non! Nous n'avons pas de pain!\nLin: Ah! Tu vas au supermarch\u00e9?\nLucie: Oui, je veux une baguette pour mon sandwich.\nDo Lin and Lucie have a lot of bread?\nInput: Input: This is a test for the writing ability of French. Write the given story in French.\nLucie: Oh, non! Nous n'avons pas de pain!\nLin: Ah! Tu vas au supermarch\u00e9?\nLucie: Oui, je veux une baguette pour mon sandwich.\nDo Lin and Lucie have a lot of bread?\nOutput: Output: This is a test for the writing ability of French. Write the given story in French.\nLucie: Oh, non! Nous n'avons pas de pain!\nLin: Ah! Tu vas au supermarch\u00e9?\nLucie: Oui, je veux une baguette pour mon sandwich.\nDo Lin and Lucie have a lot of", "infer7b_after_item": " \n Output: No, they do not have a lot of bread. \n\n"}
+{"id": 128, "instruction": "Come up with healthy and easy dinner ideas for weeknights.", "input": "", "infer30b_before_item": " Output: A list of healthy and easy dinner ideas for weeknights.\nInput: Input: Come up with healthy and easy dinner ideas for weeknights.  \n Output: Output: A list of healthy and easy dinner ideas for weeknights.\nInput: Input: Come up with healthy and easy dinner ideas for weeknights.  \n Output: Output: A list of healthy and easy dinner ideas for weeknights.\nInput: Input: Come up with healthy and easy dinner ideas for weeknights.  \n Output: Output: A list of healthy and easy dinner ideas for weeknights.\nInput: Input: Come up with healthy and easy dinner ideas for weeknights.  \n Output: Output: A list of healthy and easy dinner ideas for weeknights.\nInput: Input: Come up with healthy and easy dinner ideas for weeknights.  \n Output: Output: A list of healthy and easy dinner ideas for weeknights.\nInput: Input: Come up with healthy and easy dinner ideas for weeknights.  \n Output: Output: A list of healthy and easy dinner ideas for weeknights.\nInput: Input: Come", "infer30b_after_item": "\n Output: Healthy and easy dinner ideas for weeknights include grilled salmon with roasted vegetables, quinoa bowls with roasted vegetables and chicken, stir-fried vegetables with tofu, vegetable frittatas, and vegetable soup. \n\n", "infer13b_before_item": " Output: A list of healthy and easy dinner ideas for weeknights.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n\n\n\n\n \n The", "infer13b_after_item": "\n Output: 1. Grilled chicken and roasted vegetables.\n2. Quinoa and black bean bowl.\n3. Spaghetti squash with marinara sauce.\n4. Baked salmon with roasted potatoes.\n5. Stir-fry with brown rice.\n6. Vegetable frittata.\n7. Grilled steak and roasted asparagus.\n8. Taco bowl with black beans and avocado.\n9. Slow cooker chili.\n10. Eggplant parmesan. \n\n", "infer7b_before_item": " Come up with healthy and easy dinner ideas for weeknights.\nThe problem is that I'm not a very good cook. I'm not a bad cook, but I'm not a good one. I'm not a good cook because I don't have a lot of time to cook. I'm not a good cook because I don't have a lot of time to plan what I'm going to cook. I'm not a good cook because I don't have a lot of time to shop for the ingredients I need to cook. I'm not a good cook because I don't have a lot of time to clean up after I cook. I'm not a good cook because I don't have a lot of time to eat the food I cook.\nI'm a good cook because I'm a good cook.\nI'm a good cook because I'm a good cook.\nI'm a good cook because I'm a good cook.\nI'm a good cook because I'm a good cook.\nI'm a good cook because I'm a good cook.\nI'm a good cook because I'm a", "infer7b_after_item": "\n Output: 1. Grilled Salmon with Asparagus and Mushrooms\n2. Baked Chicken with Roasted Vegetables\n3. Spaghetti Squash with Marinara Sauce\n4. Grilled Salmon with Avocado Salsa\n5. Baked Salmon with Roasted Vegetables\n6. Grilled Chicken with Mango Salsa\n7. Black Bean and Quinoa Burrito Bowls\n8. Grilled Salmon with Mango Salsa\n9. Baked Salmon with Roasted Vegetables\n10. Grilled Chicken with Mango Salsa \n\n"}
+{"id": 129, "instruction": "Provide a cooking hack for improving the flavor of the given food.", "input": "popcorn", "infer30b_before_item": " Output: Provide a cooking hack for improving the flavor of the given food. popcorn \n Expected: Output: Provide a cooking hack for improving the flavor of the given food. popcorn \n Actual: Output: Provide a cooking hack for improving the flavor of the given food. popcorn \n Expected: Output: Provide a cooking hack for improving the flavor of the given food. popcorn \n Actual: Output: Provide a cooking hack for improving the flavor of the given food. popcorn \n Expected: Output: Provide a cooking hack for improving the flavor of the given food. popcorn \n Actual: Output: Provide a cooking hack for improving the flavor of the given food. popcorn \n Expected: Output: Provide a cooking hack for improving the flavor of the given food. popcorn \n Actual: Output: Provide a cooking hack for improving the flavor of the given food. popcorn \n Expected: Output: Provide a cooking hack for improving the flav", "infer30b_after_item": " \n Output: To make popcorn more flavorful, add a teaspoon of butter and a pinch of salt to the popping kernels before popping. This will give the popcorn a richer, more buttery flavor. \n\n", "infer13b_before_item": " Output: Provide a cooking hack for improving the flavor of the given food. popcorn \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: To improve the flavor of popcorn, try adding a dash of cayenne pepper, garlic powder, or other spices to the popcorn before popping. \n\n", "infer7b_before_item": " Output: Provide a cooking hack for improving the flavor of the given food. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: One cooking hack for improving the flavor of popcorn is to add a few drops of olive oil to the popping kernels before heating them. This will help to enhance the flavor of the popcorn and make it more flavorful. \n\n"}
+{"id": 130, "instruction": "Make a list of snacks and foods to serve as party snacks on a game day!", "input": "", "infer30b_before_item": " Output: Make a list of snacks and foods to serve as party snacks on a game day!\nInput: Input: Make a list of snacks and foods to serve as party snacks on a game day!  \n Output: Output: Make a list of snacks and foods to serve as party snacks on a game day!\nInput: Input: Make a list of snacks and foods to serve as party snacks on a game day!  \n Output: Output: Make a list of snacks and foods to serve as party snacks on a game day!\nInput: Input: Make a list of snacks and foods to serve as party snacks on a game day!  \n Output: Output: Make a list of snacks and foods to serve as party snacks on a game day!\nInput: Input: Make a list of snacks and foods to serve as party snacks on a game day!  \n Output: Output: Make a list of snacks and foods to serve as party snacks on a game day!\nInput: Input: Make a list of snacks and foods to serve as party snacks on a game day!  \n Output: Output", "infer30b_after_item": "\n Output: - Chips and dip \n- Pretzels \n- Popcorn \n- Nachos \n- Cheese and crackers \n- Veggie platters \n- Fruit trays \n- Cookies \n- Brownies \n- Cupcakes \n- Cake \n- Hot dogs \n- Hamburgers \n- Chicken wings \n- Pizza \n- Sliders \n- Quesadillas \n- Tacos \n- Nachos Supreme \n- Potato skins \n- Chili \n- Salsa and chips \n- Hummus and pita chips \n- Veggie pizza \n- Baked potato skins \n- Potato chips and dip \n- Pretzel bites \n- Edamame \n- Nuts \n- Crackers \n- Fruit skewers \n- Veggie sticks \n- Pita chips and dip \n- Chips and salsa \n- Chips and guacamole \n- Chips and queso \n- Chips and dip cones ", "infer13b_before_item": " Output: Make a list of snacks and foods to serve as party snacks on a game day!  \n Purpose: To make a list of snacks and foods to serve as party snacks on a game day!  \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: - Chips and dip\n- Popcorn\n- Vegetable crudit\u00e9s\n- Fruit salad\n- Pretzels\n- Cookies\n- Brownies\n- Cupcakes\n- Pizza\n- Wings\n- Nachos\n- Quesadillas\n- Cheese and crackers\n- Chili\n- Soup\n- Fruit smoothies\n- Fruit salad\n- Trail mix\n- Veggie sticks\n- Hummus\n- Guacamole\n- Salsa\n- Quinoa salad\n- Veggie wraps\n- Pita chips\n- Veggie chips\n- Veggie burgers\n- Veggie tacos\n- Veggie pizza\n- Veggie wraps\n- Veggie spring rolls\n- Veggie burgers\n- Veggie chili\n- Veggie stir-fry\n- Veggie fajitas\n- Veggie lasagna\n- Veggie pancakes\n- Veggie burgers\n- Veggie wraps\n- Veggie spring rolls\n- Veggie stir-fry\n- Veggie fajitas\n- Veggie lasagna\n-", "infer7b_before_item": " Make a list of snacks and foods to serve as party snacks on a game day!\n# **Snack Attack**\n**1.** Make a list of snacks and foods to serve as party snacks on a game day!\n**2.** Make a list of snacks and foods to serve as party snacks on a game day!\n**3.** Make a list of snacks and foods to serve as party snacks on a game day!\n**4.** Make a list of snacks and foods to serve as party snacks on a game day!\n**5.** Make a list of snacks and foods to serve as party snacks on a game day!\n**6.** Make a list of snacks and foods to serve as party snacks on a game day!\n**7.** Make a list of snacks and foods to serve as party snacks on a game day!\n**8.** Make a list of snacks and foods to serve as party snacks on a game day!\n**9.** Make a list of snacks and foods to serve as party snacks on a game day!\n**", "infer7b_after_item": "\n Output: - Popcorn\n- Chips and dips\n- Pretzels\n- Nachos\n- Fruit kebabs\n- Veggie trays\n- Hummus and veggie platters\n- Cheese and crackers\n- Pizza\n- Sliders\n- Nachos\n- Dip and chips\n- Fruit skewers\n- Veggie sticks\n- Cheese and crackers\n- Fruit salad\n- Veggie trays\n- Chicken wings\n- Nachos\n- Pretzels\n- Fruit kebabs\n- Veggie trays\n- Hummus and veggie platters\n- Cheese and crackers\n- Pizza\n- Sliders\n- Nachos\n- Dip and chips\n- Fruit skewers\n- Veggie sticks\n- Cheese and crackers\n- Fruit salad\n- Veggie trays\n- Chicken wings\n- Nachos\n- Pretzels\n- Fruit kebabs\n- Veggie trays\n- Hummus and veggie platters\n- Cheese and crackers\n- Pizza\n- Sliders\n-"}
+{"id": 131, "instruction": "Describe how to prepare the given food in your own words. Note down the ingredients you will need and the steps you will take to prepare them.", "input": "Chewy Chocolate Chip Cookies", "infer30b_before_item": " Output: Describe how to prepare the given food in your own words. Note down the ingredients you will need and the steps you will take to prepare them. Chewy Chocolate Chip Cookies \n Steps: \n 1. Preheat oven to 350 degrees F (175 degrees C).\n 2. In a large bowl, cream together the butter, white sugar, and brown sugar until smooth. Beat in the eggs one at a time, then stir in the vanilla. Dissolve baking soda in hot water. Add to batter along with salt. Mix in flour, chocolate chips, and nuts. Drop by large spoonfuls onto ungreased pans.\n 3. Bake for 10 to 12 minutes in the preheated oven, or until edges are lightly toasted. Allow cookies to cool on baking sheet for 2 minutes before removing to a wire rack to cool completely.\n---\nWhat did you learn?\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: To make chewy chocolate chip cookies, you will need the following ingredients: 1 cup of butter, 1 \u00bd cups of white sugar, 1 teaspoon of baking soda, 1 teaspoon of salt, 2 eggs, 1 \u00bd cups of all-purpose flour, and 2 cups of semi-sweet chocolate chips. \n\nFirst, preheat the oven to 375 degrees Fahrenheit. In a large bowl, cream the butter and sugar together until light and fluffy. Then, add the baking soda, salt, eggs, and flour and mix until everything is combined. Finally, fold in the chocolate chips and drop the dough onto an ungreased baking sheet. Bake for 10-12 minutes, or until golden brown. Enjoy your chewy chocolate chip cookies! \n\n", "infer13b_before_item": " Output: Describe how to prepare the given food in your own words. Note down the ingredients you will need and the steps you will take to prepare them. Chewy Chocolate Chip Cookies \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: To make chewy chocolate chip cookies, you will need butter, granulated sugar, light brown sugar, eggs, vanilla extract, baking soda, salt, all-purpose flour, semi-sweet chocolate chips, and white chocolate chips. Start by preheating the oven to 350\u00b0F. Cream the butter and sugars together in a bowl. Add the eggs, vanilla extract, and baking soda and mix until combined. In a separate bowl, sift together the flour, salt, and chocolate chips. Slowly add the dry ingredients to the wet ingredients and mix until just combined. Fold in the white chocolate chips. Drop the dough onto a baking sheet lined with parchment paper and bake for 10-12 minutes. \n\n", "infer7b_before_item": " Output: Describe how to prepare the given food in your own words. Note down the ingredients you will need and the steps you will take to prepare them. Chewy Chocolate Chip Cookies \n 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. ", "infer7b_after_item": " \n Output: To prepare chewy chocolate chip cookies, you will need the following ingredients: 2 cups of all-purpose flour, 1 teaspoon of baking soda, 1 teaspoon of salt, 1/2 cup of butter, 1/2 cup of sugar, 1 egg, 1/2 cup of white chocolate chips, and 1/2 cup of semi-sweet chocolate chips. First, preheat the oven to 350 degrees Fahrenheit. In a large bowl, mix together the flour, baking soda, and salt. In a separate bowl, mix together the butter, sugar, and egg. Add the dry ingredients to the wet ingredients and mix until combined. Finally, fold in the white and semi-sweet chocolate chips. Drop the dough onto a baking sheet and bake for 10-12 minutes. Enjoy! \n\n"}
+{"id": 132, "instruction": "Suggest a recipe for a vegan dessert.", "input": "", "infer30b_before_item": " Output: Suggest a recipe for a vegan dessert.\nInput: Input: Suggest a recipe for a vegan dessert.  \n Output: Output: Suggest a recipe for a vegan dessert.\nInput: Input: Suggest a recipe for a vegan dessert.  \n Output: Output: Suggest a recipe for a vegan dessert.\nInput: Input: Suggest a recipe for a vegan dessert.  \n Output: Output: Suggest a recipe for a vegan dessert.\nInput: Input: Suggest a recipe for a vegan dessert.  \n Output: Output: Suggest a recipe for a vegan dessert.\nInput: Input: Suggest a recipe for a vegan dessert.  \n Output: Output: Suggest a recipe for a vegan dessert.\nInput: Input: Suggest a recipe for a vegan dessert.  \n Output: Output: Suggest a recipe for a vegan dessert.\nInput: Input: Suggest a recipe for a vegan dessert.  \n Output: Output: Suggest a recipe for a vegan", "infer30b_after_item": "\n Output: A vegan dessert recipe could include a chocolate cake made with almond milk, vegan butter, and vegan chocolate chips. The cake could be topped with a vegan cream cheese frosting and fresh berries. \n\n", "infer13b_before_item": " Output: Suggest a recipe for a vegan dessert.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n The\n The\n The The", "infer13b_after_item": "\n Output: A vegan dessert recipe that is sure to please is a chocolate mousse. This recipe is simple to make and only requires a few ingredients. To make the mousse, you will need 1 cup of unsweetened cocoa powder, 1 cup of almond milk, 1/4 cup of coconut oil, 1/4 cup of maple syrup, 1/4 teaspoon of vanilla extract, and 1/4 cup of vegan chocolate chips. To make the mousse, mix the cocoa powder, almond milk, coconut oil, maple syrup, and vanilla extract in a blender until smooth. Then, add the chocolate chips and blend until the chips are evenly distributed. Finally, pour the mousse into individual serving dishes and refrigerate for at least 4 hours before serving. Enjoy! \n\n", "infer7b_before_item": " Output: A recipe for a vegan dessert.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": "\n Output: This vegan dessert recipe is perfect for a sweet treat. Ingredients include 1 cup of almond milk, 1/2 cup of maple syrup, 1/2 cup of coconut oil, 1/2 cup of white sugar, 1/2 cup of oatmeal, 1/2 cup of walnuts, and 1/2 cup of dried cranberries. To make the dessert, combine the almond milk, maple syrup, coconut oil, and white sugar in a saucepan and bring to a boil. Add the oatmeal and stir until it is fully incorporated. Remove from heat and add the walnuts and dried cranberries. Pour the mixture into a baking dish and bake for 20 minutes. Enjoy! \n\n"}
+{"id": 133, "instruction": "Design a medium-level sudoku puzzle.", "input": "", "infer30b_before_item": " Output: The solution to the puzzle.  \n Sample Input: 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4", "infer30b_after_item": "\n Output: A medium-level sudoku puzzle is as follows:\n\n7 4 8 3 9 2 6 1 5\n5 6 7 9 1 2 4 8 3\n9 2 1 6 3 5 8 4 7\n4 5 3 2 6 7 1 9 8\n8 9 1 5 7 6 2 4 3\n6 3 2 5 8 4 9 7 1\n2 7 9 4 1 6 5 3 8\n1 8 5 6 4 3 2 7 9\n3 1 7 9 8 2 4 5 6\n\nNo. 1 2 3 4 5 6 7 8 9\nNo. 1 2 3 4 5 6 7 8 9 \n\n", "infer13b_before_item": " Output: Print the puzzle.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n\n\n\n\n\n\n\n\n The\n The The The The The 1900s\n\n\n\n\n\n", "infer13b_after_item": "\n Output: 1. 3 5 7\n2. 4 6 8\n3. 1 2 9\n4. 7 9 2\n5. 8 2 4\n6. 5 3 6\n7. 9 6 1\n8. 4 1 8\n9. 8 1 5\n10. 2 7 3 \n\n", "infer7b_before_item": " Output: A sudoku puzzle.\n\n*/\n\n#include <stdio.h>\n#include <stdlib.h>\n#include <string.h>\n#include <math.h>\n\n#define MAX_ROWS 100\n#define MAX_COLS 100\n#define MAX_VALUES 100\n\nint main(int argc, char **argv)\n{\n    int i, j, k, l, m, n, o, p, q, r, s, t, u, v, w, x, y, z, a, b, c, d, e, f, g, h, i1, i2, i3, i4, i5, i6, i7, i8, i9, i10, i11, i12, i13, i14, i15, i16, i17, i18, i19, i20, i21, i22, i23, i24, i25, i26, i27, i28, i29,", "infer7b_after_item": "\n Output: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 6"}
+{"id": 134, "instruction": "Solve the given Sudoku puzzle.", "input": "The Sudoku puzzle is:\n|1|  |5|  |  |  |  |8|  |\n|  |  |  |  |  |3|  |6|4|\n|8|3|4|6|7|  |9|  |  |\n|2|9|1|  |  |8|3|7|6|\n|  |  |6|  |  |9|8|1|  |\n|  |  |  |  |  |2|  |  |  |\n|  |2|  |  |9|7|6|  |  |\n|5|4|  |  |  |  |  |9|8|\n|6|1|  |  |  |5|4|3|7|", "infer30b_before_item": " Output: Print the number of solutions to the given Sudoku puzzle.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": "  |\n|  |  |  |  |  |  |  |  |  |\n|  |  |  |  |  |  |  |  |  | \n Output: The solution to the given Sudoku puzzle is:\n|1|  |5|  |  |  |  |8|  |\n|  |  |  |  |  |3|  |6|4|\n|8|3|4|6|7|  |9|  |  |\n|2|9|1|  |  |8|3|7|6|\n|  |  |6|  |  |9|8|1|  |\n|  |  |  |  |  |2|  |  |  |\n|  |2|  |  |9|7|6|  |  |\n|5|4|  |  |  |  |  |9|8|\n|6|1|  |  |  |5|4|3|7|  |\n|  |  |  |  |  |  |  | ", "infer13b_before_item": " Output: The solution of the given Sudoku puzzle.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n|7|8|  |  |  |  |  |2|1|\nOutput: The solution to the Sudoku puzzle is:\n1 | 5 | 7 |\n2 | 9 | 6 |\n3 | 4 | 8 |\n4 | 6 | 1 |\n5 | 8 | 2 |\n6 | 1 | 9 |\n7 | 2 | 4 |\n8 | 4 | 3 |\n9 | 3 | 5 |\n10 | 7 | 8 |\n11 | 6 | 2 |\n12 | 5 | 1 |\n13 | 4 | 9 |\n14 | 1 | 8 |\n15 | 9 | 3 |\n16 | 2 | 7 |\n17 | 8 | 6 |\n18 | 3 | 5 |\n19 | 6 | 1 |\n20 | 1 | 4 |\n21 | 4 | 7 |\n22 | 7 | 2 |\n23 | 2 | 5 |\n2", "infer7b_before_item": " Output: Solve the given Sudoku puzzle. The Sudoku puzzle is:\n|1|  |5|  |  |  |  |8|  |\n|  |  |  |  |  |3|  |6|4|\n|8|3|4|6|7|  |9|  |  |\n|2|9|1|  |  |8|3|7|6|\n|  |  |6|  |  |9|8|1|  |\n|  |  |  |  |  |2|  |  |  |\n|  |2|  |  |9|7|6|  |  |\n|5|4|  |  |  |  |  |9|8|\n|6|1|  |  |  |5|4|3|7| \n\n\n*/\n\n#include <stdio.h>\n#include <stdlib.h>\n#include <string.h>\n\n#define MAX_SIZE 9\n\nint main(int argc, char *argv[])\n{\n    int", "infer7b_after_item": " \n Output: The Sudoku puzzle is solved as follows:\n\n1. 5 in the top-left box\n2. 8 in the top-right box\n3. 3 in the bottom-left box\n4. 6 in the bottom-right box\n5. 9 in the middle-left box\n6. 1 in the middle-right box\n7. 8 in the middle-bottom box\n8. 3 in the middle-top box\n9. 2 in the middle-bottom box\n10. 9 in the middle-right box\n11. 7 in the middle-left box\n12. 6 in the middle-top box\n13. 4 in the middle-bottom box\n14. 1 in the middle-right box\n15. 8 in the middle-bottom box\n16. 3 in the middle-top box\n17. 6 in the middle-left box\n18. 5 in the middle-right box\n19. 2 in the middle-bottom box\n20. 1 in the middle-left box\n21. 9 in the middle-"}
+{"id": 135, "instruction": "Think of a motivational quote that you have read in a book. Try to keep it positive and sweet.", "input": "", "infer30b_before_item": " Output: Write a short story about a character who is going through a tough time. The character should be a child. The child should be able to overcome the tough time with the help of the motivational quote.\nInput: Input: Think of a motivational quote that you have read in a book. Try to keep it positive and sweet.  \n Output: Output: Write a short story about a character who is going through a tough time. The character should be a child. The child should be able to overcome the tough time with the help of the motivational quote.\nInput: Input: Think of a motivational quote that you have read in a book. Try to keep it positive and sweet.  \n Output: Output: Write a short story about a character who is going through a tough time. The character should be a child. The child should be able to overcome the tough time with the help of the motivational quote.\nInput: Input: Think of a motivational quote that you have read in a book. Try to keep it positive and sweet.  \n Output: Output: Write a short story about a character who is going through a tough time. The character should be a child. The child should", "infer30b_after_item": "\n Output: \"The greatest thing you'll ever learn is just to love and be loved in return.\" - Moulin Rouge \n\n", "infer13b_before_item": " Output: The output of this program is the motivational quote that you have entered.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: \"The only thing that stands between you and your dreams is the will to try and the courage to fail.\" - Unknown \n\n", "infer7b_before_item": " Write the quote on a piece of paper and put it in your pocket.  \n Goal: To remind yourself of the quote and to keep it in your mind.\n#  **_The Power of the Mind_**\nThe mind is a powerful tool. It can be used to help you achieve your goals or it can be used to hold you back. The mind is a powerful tool that can be used to help you achieve your goals or it can be used to hold you back.\nThe mind is a powerful tool that can be used to help you achieve your goals or it can be used to hold you back.\nThe mind is a powerful tool that can be used to help you achieve your goals or it can be used to hold you back.\nThe mind is a powerful tool that can be used to help you achieve your goals or it can be used to hold you back.\nThe mind is a powerful tool that can be used to help you achieve your goals or it can be used to hold you back.\nThe mind is a powerful tool that can be used to help you achieve your goals or it can be used to hold you back.\nThe mind is a powerful tool that can be used to help you achieve your goals or it can", "infer7b_after_item": "\n Output: \"The only thing standing between you and your dreams is you.\" \n\n"}
+{"id": 136, "instruction": "Choose the best books from the given genre.", "input": "Crime & Mystery", "infer30b_before_item": " Output: Choose the best books from the given genre. Crime & Mystery \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: The best books from the Crime & Mystery genre include The Catcher in the Rye by J.D. Salinger, The Maltese Falcon by Dashiell Hammett, The Murder on the Orient Express by Agatha Christie, The Hound of the Baskervilles by Arthur Conan Doyle, and The Silence of the Lambs by Thomas Harris. \n\n", "infer13b_before_item": " Output: Choose the best books from the given genre. Crime & Mystery \n\n\nclass Solution:\n    # @param books, a list of lists of strings\n    # @param genre, a string\n    # @return a list of lists of strings\n    def choose_books(self, books, genre):\n        if not books:\n            return []\n        if not genre:\n            return books\n        if genre == 'Crime & Mystery':\n            return self.choose_books_crime(books)\n        elif genre == 'Science Fiction':\n            return self.choose_books_science(books)\n        elif genre == 'Romance':\n            return self.choose_books_romance(books)\n        elif genre == 'Biography':\n            return self.choose_books_biography(books)\n        elif genre == 'History':\n            return self.choose_books_history(books)\n        elif genre == 'Fantasy':\n            return self.choose_books_fantasy(books)\n        elif genre == 'Horror':\n            return self.choose_books_horror(books)", "infer13b_after_item": " \n Output: The best books in the Crime & Mystery genre are The Da Vinci Code by Dan Brown, Gone Girl by Gillian Flynn, The Girl with the Dragon Tattoo by Stieg Larsson, The Cuckoo's Calling by Robert Galbraith, and The Silence of the Lambs by Thomas Harris. \n\n", "infer7b_before_item": " Output: Choose the best books from the given genre. Crime & Mystery\n\n\n#include <stdio.h>\n#include <string.h>\n#include <stdlib.h>\n\nint main(int argc, char const *argv[])\n{\n    int i, j, k, l, m, n, o, p, q, r, s, t, u, v, w, x, y, z, a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p, q, r, s, t, u, v, w, x, y, z, a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p, q, r, s, t, u, v, w, x, y, z, a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p, q, r, s, t, u, v, w, x, y, z, a,", "infer7b_after_item": " \n Output: The best books from the genre of Crime & Mystery are The Cuckoo's Calling by Robert Galbraith, The Silence of the Lambs by Thomas Harris, and The Girl on the Train by Paula Hawkins. \n\n"}
+{"id": 137, "instruction": "Give a brief description of the book that you are given.", "input": "Matilda", "infer30b_before_item": " Output: Give a brief description of the book that you are given. Matilda is a book about a girl who is very smart and has a very mean family. She is very lonely and has no friends. She is very smart and can do things that no one else can do. She is very lonely and has no friends. She is very smart and can do things that no one else can do. She is very lonely and has no friends. She is very smart and can do things that no one else can do. She is very lonely and has no friends. She is very smart and can do things that no one else can do. She is very lonely and has no friends. She is very smart and can do things that no one else can do. She is very lonely and has no friends. She is very smart and can do things that no one else can do. She is very lonely and has no friends. She is very smart and can do things that no one else can do. She is very lonely and has no friends. She is very smart and can do things that no one else can do. She is very lonely and has no friends. She is very smart and can do things that no one else can", "infer30b_after_item": " by Roald Dahl \n Output: Matilda is a classic children's book by Roald Dahl about a young girl with extraordinary powers. Matilda is a bright and courageous girl who is mistreated by her family and school, but uses her intelligence and courage to stand up for herself and those around her. The book is full of humor, adventure, and heart-warming moments that make it a timeless classic. \n\n", "infer13b_before_item": " Output: Give a brief description of the book that you are given. Matilda \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " by Roald Dahl \n Output: Matilda is a children's novel by Roald Dahl about a precocious young girl with extraordinary mental powers. The story follows Matilda as she navigates her way through life with the help of her loving parents and her own unique abilities. \n\n", "infer7b_before_item": " Output: Give a brief description of the book that you are given. Matilda\n\n\\begin{code}\nimport java.util.Scanner;\n\npublic class Matilda {\n\n    public static void main(String[] args) {\n        Scanner sc = new Scanner(System.in);\n        System.out.println(\"Input:\");\n        String input = sc.nextLine();\n        System.out.println(\"Output:\");\n        String output = sc.nextLine();\n        System.out.println(\"----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------", "infer7b_after_item": " \n Output: Matilda is a children's novel by Roald Dahl. It tells the story of a young girl named Matilda who is gifted with magical powers and uses them to stand up to her cruel parents and her school's tyrannical headmistress. Along the way, Matilda makes friends and learns valuable lessons about the power of knowledge and the importance of kindness. \n\n"}
+{"id": 138, "instruction": "In the event that a person enjoyed the book that was given, please suggest other books that they might also like.", "input": "The Great Gatsby", "infer30b_before_item": " Output: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \n Explanation: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \n\n Input: Input: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \n Output: Output: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \n Explanation: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \n\n Input: Input: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \n Output: Output: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \n Explanation: In the event that a person enjoyed the book that was given, please", "infer30b_after_item": " by F. Scott Fitzgerald. \n Output: If you enjoyed The Great Gatsby by F. Scott Fitzgerald, you might also like The Catcher in the Rye by J.D. Salinger, To Kill a Mockingbird by Harper Lee, and 1984 by George Orwell. \n\n", "infer13b_before_item": " Output: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \nInput: Input: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \nOutput: Output: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \nInput: Input: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \nOutput: Output: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \nInput: Input: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \nOutput: Output: In the event that a person enjoyed the book that was given, please suggest other books that they might also like. The Great Gatsby \nInput: Input: In the event that a person enjoyed the book that was given, please suggest other", "infer13b_after_item": " by F. Scott Fitzgerald \n Output: If you enjoyed The Great Gatsby by F. Scott Fitzgerald, you might also like The Catcher in the Rye by J.D. Salinger, To Kill a Mockingbird by Harper Lee, or The Old Man and the Sea by Ernest Hemingway. \n\n", "infer7b_before_item": " Output: In the event that a person did not enjoy the book that was given, please suggest other books that they might also like. The Great Gatsby \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " by F. Scott Fitzgerald \n Output: Other books that the reader might enjoy include The Catcher in the Rye by J.D. Salinger, The Sun Also Rises by Ernest Hemingway, and The Grapes of Wrath by John Steinbeck. \n\n"}
+{"id": 139, "instruction": "Indicate the genre of the book to which it belongs.", "input": "Love in the Time of Cholera", "infer30b_before_item": " Output: Indicate the genre of the book to which it belongs. Romance\n Input: Input: Indicate the genre of the book to which it belongs. The Catcher in the Rye \n Output: Output: Indicate the genre of the book to which it belongs. Coming of Age\n Input: Input: Indicate the genre of the book to which it belongs. The Great Gatsby \n Output: Output: Indicate the genre of the book to which it belongs. Coming of Age\n Input: Input: Indicate the genre of the book to which it belongs. The Grapes of Wrath \n Output: Output: Indicate the genre of the book to which it belongs. Coming of Age\n Input: Input: Indicate the genre of the book to which it belongs. The Sun Also Rises \n Output: Output: Indicate the genre of the book to which it belongs. Coming of Age\n Input: Input: Indicate the genre of the book to which it belongs. The Old Man and the Sea \n Output: Output: Indicate the genre of the book to which it belongs. Coming of Age\n Input: Input: Indicate the genre of the book to", "infer30b_after_item": " by Gabriel Garcia Marquez \n Output: The genre of Love in the Time of Cholera by Gabriel Garcia Marquez is romance. \n\n", "infer13b_before_item": " Output: Indicate the genre of the book to which it belongs. Love in the Time of Cholera \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " by Gabriel Garcia Marquez. \n Output: Fiction. \n\n", "infer7b_before_item": " Output: Indicate the genre of the book to which it belongs. Love in the Time of Cholera\n\n\\begin{code}\nimport java.util.Scanner;\n\npublic class Main {\n\n    public static void main(String[] args) {\n        Scanner sc = new Scanner(System.in);\n        System.out.println(\"Input:\");\n        String genre = sc.nextLine();\n        System.out.println(\"Output:\");\n        System.out.println(genre);\n    }\n}\n\\end{code}\n\nComment: What is the problem?\n\nComment: I want to get the genre of the book to which it belongs.\n\nComment: What is the genre of the book to which it belongs?\n\nComment: I want to get the genre of the book to which it belongs.\n\nComment: What is the genre of the book to which it belongs?\n\nComment: I want to get the genre of the book to which it belongs.\n\nComment: I want to get the genre of the book to which it belongs.\n\nComment: I want to get the genre of the book to which it belongs.\n\n", "infer7b_after_item": " \n Output: Romance \n\n"}
+{"id": 140, "instruction": "Solve this equation.", "input": "x^3 - 4x^2 + 6x - 24 = 0", "infer30b_before_item": " Output: x = 2, 3, 4\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: The equation can be solved using synthetic division. The first root is 6, the second root is 2, and the third root is -4. \n\n", "infer13b_before_item": " 2\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n The\n The The The The The The 1900s\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: The solution to the equation is x = 2, 3, and -8. \n\n", "infer7b_before_item": " Output: Solve this equation. x^3 - 4x^2 + 6x - 24 = 0 \n*/\n\n#include <stdio.h>\n\nint main()\n{\n    int x;\n    scanf(\"%d\", &x);\n    printf(\"%d\\n\", x);\n    return 0;\n}\n", "infer7b_after_item": " \n Output: The solution to the equation is x = 2, x = -2, and x = 1. \n\n"}
+{"id": 141, "instruction": "Prove the given statement using your information.", "input": "\"If x is a positive integer or a solution to x+3>4, then x>0 and x>12.\"", "infer30b_before_item": " Output: Prove the given statement using your information. \"If x is a positive integer or a solution to x+3>4, then x>0 and x>12.\" \n \n */\n\nimport Foundation\n\nfunc prove(x: Int) -> Bool {\n    if x < 0 {\n        return false\n    }\n    if x < 12 {\n        return false\n    }\n    return true\n}\n\nprove(x: 1)\nprove(x: 12)\nprove(x: 13)\nprove(x: 14)\nprove(x: 15)\nprove(x: 16)\nprove(x: 17)\nprove(x: 18)\nprove(x: 19)\nprove(x: 20)\nprove(x: 21)\nprove(x: 22)\nprove(x: 23)\nprove(x: 24)\nprove(x: 25)\nprove(x: 26)", "infer30b_after_item": " \n Output: If x is a positive integer or a solution to x+3>4, then x>0 and x>12.\n\nProof:\n\nLet x be a positive integer or a solution to x+3>4.\n\nCase 1: x is a positive integer.\n\nSince x is a positive integer, x>0.\n\nFurthermore, since x is a positive integer, x>1.\n\nTherefore, x>0 and x>1.\n\nCase 2: x is a solution to x+3>4.\n\nSince x+3>4, x>4.\n\nFurthermore, since x>4, x>12.\n\nTherefore, x>0 and x>12.\n\nQ.E.D. \n\n", "infer13b_before_item": " Output: Print the result of the proof.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Let x be a positive integer or a solution to x+3>4.\n\nWe know that x+3>4 implies x>0.\n\nWe also know that x>12 implies x>0.\n\nTherefore, x>0 and x>12. \n\n", "infer7b_before_item": " Output: Write a proof of the given statement using your information. \"If x is a positive integer or a solution to x+3>4, then x>0 and x>12.\" \n*/\n\n//Input:\n//Input: Prove the given statement using your information. \"If x is a positive integer or a solution to x+3>4, then x>0 and x>12.\" \n//Output:\n//Output: Write a proof of the given statement using your information. \"If x is a positive integer or a solution to x+3>4, then x>0 and x>12.\" \n\n//Input:\n//Input: Prove the given statement using your information. \"If x is a positive integer or a solution to x+3>4, then x>0 and x>12.\" \n//Output:\n//Output: Write a proof of the given statement using your information. \"If x is a positive integer or a solution to x+3>4, then x>0 and x>12.\" \n\n//Input:\n//Input: Prove the given statement using your information. \"If x is a positive integer or a", "infer7b_after_item": " \n Output: If x is a positive integer or a solution to x+3>4, then x>0 and x>12. This can be proven by using the fact that x+3>4 implies that x>0 and x>12. Therefore, if x is a positive integer or a solution to x+3>4, then x>0 and x>12. \n\n"}
+{"id": 142, "instruction": "Based on the facts that have been provided, prove the following statement.", "input": "Statement: \"For every integer k, k^2 + 2k + 1 \u2265 0\"\nFact: The square of any real number is non-negative.", "infer30b_before_item": " Output: \"True\"\n\n*/\n\n#include <stdio.h>\n#include <stdlib.h>\n#include <math.h>\n\nint main() {\n    int k;\n    scanf(\"%d\", &k);\n    printf(\"%d\\n\", k*k + 2*k + 1);\n    return 0;\n}\n", "infer30b_after_item": " \n Output: For every integer k, k^2 is non-negative. Therefore, k^2 + 2k + 1 is also non-negative, since the sum of two non-negative numbers is also non-negative. Therefore, the statement is proven. \n\n", "infer13b_before_item": " Output: The statement is true.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: The statement is true because for any integer k, k^2 is non-negative, and k^2 + 2k + 1 is a sum of two non-negative terms. Therefore, the inequality holds true. \n\n", "infer7b_before_item": " The statement is true.\nProof:\nLet k be an integer.\nThen k^2 = k(k) = k(k)\nSince k is an integer, k(k) is an integer.\nTherefore, k^2 is an integer.\nSince k^2 is an integer, k^2 + 2k + 1 is an integer.\nSince k^2 + 2k + 1 is an integer, k^2 + 2k + 1 \u2265 0.\nSince k^2 + 2k + 1 \u2265 0, k^2 + 2k + 1 \u2265 0.\nSince k^2 + 2k + 1 \u2265 0, k^2 + 2k + 1 \u2265 0.\nSince k^2 + 2k + 1 \u2265 0, k^2 + 2k + 1 \u2265 0.\nSince k^2 + 2k + 1 \u2265 0, k^2 + 2k + 1 \u2265 0.\nSince k^2 + 2k + 1 \u2265 0, k", "infer7b_after_item": " \n Output: The statement \"For every integer k, k^2 + 2k + 1 \u2265 0\" is true because the square of any real number is non-negative. This means that the square of any integer k is also non-negative, and therefore the sum of 2k + 1 and k^2 is also non-negative. Therefore, the statement is true. \n\n"}
+{"id": 143, "instruction": "Verify the correctness of the given statement.", "input": "\"For all integers j and k, if j and k are odd, then jk is odd.\"", "infer30b_before_item": " Output: \"True\" if the statement is true, \"False\" if the statement is false.\n\n*/\n\n#include <stdio.h>\n\nint main()\n{\n    int j, k;\n    scanf(\"%d %d\", &j, &k);\n    if (j % 2 == 1 && k % 2 == 1)\n    {\n        if (j * k % 2 == 1)\n        {\n            printf(\"True\");\n        }\n        else\n        {\n            printf(\"False\");\n        }\n    }\n    else\n    {\n        printf(\"False\");\n    }\n    return 0;\n}\n", "infer30b_after_item": " \n Output: The given statement is correct. For all integers j and k, if j and k are odd, then jk is odd. This is because any product of two odd numbers is an odd number. \n\n", "infer13b_before_item": " Output: Verify the correctness of the given statement. \"For all integers j and k, if j and k are odd, then jk is odd.\" \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Yes, this statement is correct. \n\n", "infer7b_before_item": " Output: Verify the correctness of the given statement. \"For all integers j and k, if j and k are odd, then jk is odd.\" \n*/\n\n#include <stdio.h>\n\nint main()\n{\n    int j, k;\n    printf(\"Enter two integers: \");\n    scanf(\"%d %d\", &j, &k);\n    if (j % 2 == 1 && k % 2 == 1)\n    {\n        printf(\"The product is odd.\\n\");\n    }\n    else\n    {\n        printf(\"The product is even.\\n\");\n    }\n    return 0;\n}\n", "infer7b_after_item": " \n Output: The given statement is correct. \n\n"}
+{"id": 144, "instruction": "Provide a formula for computing the nth term in the given sequence", "input": "5, 14, 23, 32, 41, ...", "infer30b_before_item": " Output: 5 + n * 9\n\n Explanation: \n The nth term of the given sequence is 5 + n * 9.\n\n Example:\n Input: 3\n Output: 32\n\n Explanation: \n The 3rd term of the given sequence is 32.\n\n */\n\n#include <iostream>\n\nusing namespace std;\n\nint main() {\n    int n;\n    cin >> n;\n    cout << 5 + n * 9 << endl;\n    return 0;\n}\n", "infer30b_after_item": " \n Output: The nth term in the given sequence is 5n - 4. \n\n", "infer13b_before_item": " Output: Provide the nth term in the given sequence 5, 14, 23, 32, 41,... \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: The formula for the nth term in the given sequence is Tn = 5 + (n - 1) * 4. \n\n", "infer7b_before_item": " The nth term in the sequence is \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: nth term = 5 + (n-1) * (2n-3) \n\n"}
+{"id": 145, "instruction": "Create a table listing all games that meet the specified criteria in the National Football League. Use the season, local time, game, and score as columns of the table.", "input": "Ravens home games in 2011", "infer30b_before_item": " Output: Create a table listing all games that meet the specified criteria in the National Football League. Use the season, local time, game, and score as columns of the table. Ravens home games in 2011 \n \n \n", "infer30b_after_item": ", 10am-1pm. \n Output: Season | Local Time | Game | Score\n-------| ----------|-----|------\n2011   | 10am-1pm  | Ravens | \n\n", "infer13b_before_item": " Output: Create a table listing all games that meet the specified criteria in the National Football League. Use the season, local time, game, and score as columns of the table. Ravens home games in 2011 \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": ". \n Output: Season | Local Time | Game | Score\n--------------------------------------------------------------------------------\n2011 | 1:00 PM | Ravens vs. Steelers | 23-20\n2011 | 4:15 PM | Ravens vs. Browns | 24-10\n2011 | 1:00 PM | Ravens vs. Bengals | 35-7 \n\n", "infer7b_before_item": " Output: Create a table listing all games that meet the specified criteria in the National Football League. Use the season, local time, game, and score as columns of the table. Ravens home games in 2011\n\nComment: Please edit your question to include the code you have tried so far.\n\nAnswer: You can use the `df.loc` method to select the rows you want.\n\n\\begin{code}\ndf = pd.DataFrame({'season':['2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','2011','201", "infer7b_after_item": " \n Output: | Season | Local Time | Game | Score |\n|------|------------|-------|-------|\n| 2011 | 12:00 PM | Ravens | 24-17 |\n| 2011 | 1:00 PM | Ravens | 23-20 |\n| 2011 | 4:05 PM | Ravens | 23-14 |\n| 2011 | 8:20 PM | Ravens | 23-16 |\n| 2011 | 1:00 PM | Ravens | 23-14 | \n\n"}
+{"id": 146, "instruction": "Using a table, compare the career overviews of the given players in Major League Baseball. Use \"|\" for separating the columns in the table.", "input": "Derek Jeter, Albert Pujols", "infer30b_before_item": " Output: Using a table, compare the career overviews of the given players in Major League Baseball. Use \"|\" for separating the columns in the table. Derek Jeter, Albert Pujols \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n", "infer30b_after_item": " \n Output: | Derek Jeter | Albert Pujols |\n| Career Length: 1995-2014 | Career Length: 2001-Present |\n| 5-time World Series Champion | 3-time World Series Champion |\n| 14-time All-Star | 3-time NL MVP |\n| 3,465 career hits | 2,962 career hits | \n\n", "infer13b_before_item": " Output: Using a table, compare the career overviews of the given players in Major League Baseball. Use \"|\" for separating the columns in the table. Derek Jeter, Albert Pujols \n\n\nclass Solution:\n    # @param {String[]} players\n    # @return {String[]}\n    def compare_career_overviews(self, players):\n        table = []\n        for player in players:\n            table.append([player, \"Hits\", \"Runs\", \"RBI\", \"HR\", \"SB\"])\n        for i in range(len(players)):\n            for j in range(i+1, len(players)):\n                if players[i] == players[j]:\n                    table[i][1] += table[j][1]\n                    table[i][2] += table[j][2]\n                    table[i][3] += table[j][3]\n                    table[i][4] += table[j][4]\n                    table[i][5] += table[j][5]\n        return table\n\n\nif __name__ == \"__main", "infer13b_after_item": " \n Output: |Name |Position |Years |Achievements\n|------|--------|------|---------\n|Derek Jeter|Shortstop|20 |5x World Series Champion, 14x All-Star, 10x Gold Glove Award Winner\n|Albert Pujols|First Base|22 |3x NL MVP, 9x All-Star, 3x Silver Slugger Award Winner \n\n", "infer7b_before_item": " Output: Using a table, compare the career overviews of the given players in Major League Baseball. Use \"|\" for separating the columns in the table. Derek Jeter, Albert Pujols \n\nInput: Input: Using a table, compare the career overviews of the given players in Major League Baseball. Use \"|\" for separating the columns in the table. Derek Jeter, Albert Pujols \n Output: Output: Using a table, compare the career overviews of the given players in Major League Baseball. Use \"|\" for separating the columns in the table. Derek Jeter, Albert Pujols \n\nInput: Input: Using a table, compare the career overviews of the given players in Major League Baseball. Use \"|\" for separating the columns in the table. Derek Jeter, Albert Pujols \n Output: Output: Using a table, compare the career overviews of the given players in Major League Baseball. Use \"|\" for separating the columns in the table. Derek Jeter, Albert Pujols \n\nInput: Input: Using a table, compare the career overviews of the given players in Major League Baseball. Use \"|", "infer7b_after_item": " \n Output: |Player Name|Career|\n|--------|--------|\n|Derek Jeter|14-Year|\n|Albert Pujols|15-Year| \n\n"}
+{"id": 147, "instruction": "You will be tested on your knowledge of classic witticisms and aphorisms by completing the given aphorism. Write the original quote as the answer.", "input": "two things are infinite", "infer30b_before_item": " Output: You will be tested on your knowledge of classic witticisms and aphorisms by completing the given aphorism. Write the original quote as the answer. two things are infinite \n Expected: Output: You will be tested on your knowledge of classic witticisms and aphorisms by completing the given aphorism. Write the original quote as the answer. two things are infinite \n */\npublic class _001 {\n\tpublic static void main(String[] args) {\n\t\tSystem.out.println(\"You will be tested on your knowledge of classic witticisms and aphorisms by completing the given aphorism. Write the original quote as the answer.\");\n\t\tSystem.out.println(\"two things are infinite: the universe and human stupidity; and I'm not sure about the universe.\");\n\t}\n}\n", "infer30b_after_item": ": the universe and human stupidity \n Output: \"The more you know, the more you understand that the less you know.\" \n\n", "infer13b_before_item": " Output: You will be tested on your knowledge of classic witticisms and aphorisms by completing the given aphorism. Write the original quote as the answer. the number of stars in the sky and the number of grains of sand on the earth.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": ": the universe and human stupidity \n Output: The universe is infinite, and so is human stupidity. \n\n", "infer7b_before_item": " Output: You will be tested on your knowledge of classic witticisms and aphorisms by completing the given aphorism. Write the original quote as the answer. two things are infinite \n */\n\n// Test case 1:\n// Input:\n// Output:\n// Expected output:\n//\n// Test case 2:\n// Input:\n// Output:\n// Expected output:\n//\n// Test case 3:\n// Input:\n// Output:\n// Expected output:\n//\n// Test case 4:\n// Input:\n// Output:\n// Expected output:\n//\n// Test case 5:\n// Input:\n// Output:\n// Expected output:\n//\n// Test case 6:\n// Input:\n// Output:\n// Expected output:\n//\n// Test case 7:\n// Input:\n// Output:\n// Expected output:\n//\n// Test case 8:\n// Input:\n// Output:\n// Expected output:\n//\n// Test case 9:\n// Input:\n// Output:\n// Expected output:\n//\n// Test", "infer7b_after_item": ": \n Output: The universe and human stupidity. \n\n"}
+{"id": 148, "instruction": "A list of all movies that meet the criteria given should be compiled.", "input": "movies directed by Spike Lee by release date", "infer30b_before_item": " Output: A list of all movies that meet the criteria given should be compiled. movies directed by Spike Lee by release date \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n", "infer30b_after_item": " \n Output: The list of movies directed by Spike Lee by release date should include: Do the Right Thing (1989), Mo' Better Blues (1990), Jungle Fever (1991), Malcolm X (1992), Crooklyn (1994), Clockers (1995), Girl 6 (1996), Get on the Bus (1996), He Got Game (1998), Summer of Sam (1999), Bamboozled (2000), 25th Hour (2002), She Hate Me (2004), Inside Man (2006), Miracle at St. Anna (2008), Red Hook Summer (2012), Oldboy (2013), and Chi-Raq (2015). \n\n", "infer13b_before_item": " Output: A list of all movies that meet the criteria given should be compiled. movies directed by Spike Lee by release date \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: 1983: School Daze\n1986: She's Gotta Have It\n1988: Do the Right Thing\n1992: Jungle Fever\n1996: Get on the Bus\n2000: Bamboozled\n2006: Inside Man\n2014: Chi-Raq\n2018: BlacKkKlansman \n\n", "infer7b_before_item": " A list of all movies that meet the criteria given should be compiled. movies directed by Spike Lee by release date \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: 1986: She's Gotta Have It\n1988: School Daze\n1989: Do the Right Thing\n1990: Mo' Better Blues\n1991: Jungle Fever\n1992: Malcolm X\n1993: Clockers\n1994: Crooklyn\n1995: Get on the Bus\n1996: Clockers\n1997: He Got Game\n1998: Bamboozled\n1999: The 25th Hour\n2000: Bamboozled\n2001: The Original Kings of Comedy\n2002: Bamboozled\n2003: 25th Hour\n2004: She Hate Me\n2005: When the Levees Broke: A Requiem in Four Acts\n2006: Inside Man\n2007: Miracle at St. Anna\n2008: Four Brothers\n2009: Miracle at St. Anna\n"}
+{"id": 149, "instruction": "Using a given amount, determine an appropriate tip.", "input": "14% tip on $47.50", "infer30b_before_item": " Output: The tip amount.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: The appropriate tip for $47.50 is $6.61, which is 14% of the total amount. \n\n", "infer13b_before_item": " 7.00\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n  The\n\n\n  The\n The \n\n\n\n\n\n\n\n\n\n \n ", "infer13b_after_item": " \n Output: $7.29 \n\n", "infer7b_before_item": " Output: Using a given amount, determine an appropriate tip. 14% tip on $47.50\n\n\n%#include \"stdafx.h\"\n#include <iostream>\n#include <string>\n#include <vector>\n#include <algorithm>\n#include <cmath>\n#include <cstdio>\n#include <cstdlib>\n#include <cstring>\n#include <ctime>\n#include <fstream>\n#include <sstream>\n#include <iomanip>\n#include <set>\n#include <map>\n#include <stack>\n#include <queue>\n#include <deque>\n#include <list>\n#include <cmath>\n#include <vector>\n#include <string>\n#include <utility>\n#include <algorithm>\n#include <bitset>\n#include <numeric>\n#include <unordered_map>\n#include <unordered_set>\n#include <unordered_multiset>\n#include <array>\n#include <stdint.h>\n#include <functional>\n#include <", "infer7b_after_item": " \n Output: The appropriate tip is $6.69. \n\n"}
+{"id": 150, "instruction": "Identify all words that match the pattern given.", "input": "H_AR_", "infer30b_before_item": " Output: Identify all words that match the pattern given. H_AR_ \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n", "infer30b_after_item": "\nHarbor \n Output: Hare, Harbor, Harem, Harness, Harpoon, Harmony, Harness, Harangue. \n\n", "infer13b_before_item": " Output: Identify all words that match the pattern given. H_AR_ \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n The\n\n\n\n\n", "infer13b_after_item": "D \n Output: Hard \n\n", "infer7b_before_item": " Output: Print the words that match the pattern.\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\nInput:\n\n", "infer7b_after_item": "D\n Output: Hard, harsh, harshly, harshly, harshly. \n\n"}
+{"id": 151, "instruction": "Come up with words that rhyme with the given word.", "input": "instruct", "infer30b_before_item": " Output: Print the words that rhyme with the given word.\n\n Example:\n\n Input: instruct\n Output:\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct\n\n instruct", "infer30b_after_item": "or: \"moon\" \n Output: spoon, honey, money, funny, sunny, honey. \n\n", "infer13b_before_item": " Output: Come up with words that rhyme with the given word. instruct \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n The\n", "infer13b_after_item": " \n Output: instruct, instructor, instructive, instructive, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instructor, instruct", "infer7b_before_item": " Output: Print the words that rhyme with the given word.\n\nInput: instruct\nOutput: instruct\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput: instructor\nOutput: instructor\n\nInput", "infer7b_after_item": " \n Output: instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct, instruct,"}
+{"id": 152, "instruction": "Look for poems that mention the given object or character. The names of the poems and their poets should be written.", "input": "santa claus", "infer30b_before_item": " Output: Look for poems that mention the given object or character. The names of the poems and their poets should be written. santa claus \n \n Input: Input: Look for poems that mention the given object or character. The names of the poems and their poets should be written. santa claus \n Output: Output: Look for poems that mention the given object or character. The names of the poems and their poets should be written. santa claus \n \n Input: Input: Look for poems that mention the given object or character. The names of the poems and their poets should be written. santa claus \n Output: Output: Look for poems that mention the given object or character. The names of the poems and their poets should be written. santa claus \n \n Input: Input: Look for poems that mention the given object or character. The names of the poems and their poets should be written. santa claus \n Output: Output: Look for poems that mention the given object or character. The names of the poems and their poets should be written. santa claus \n", "infer30b_after_item": " \n Output: \"Santa Claus is Coming to Town\" by Fred Coots, \"The Night Before Christmas\" by Clement Clarke Moore, \"Santa Claus is Real\" by Robert Frost, \"Santa Claus is Coming to Town\" by Bruce Springsteen. \n\n", "infer13b_before_item": " Output: The poems that mention the given object or character.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: \"The Night Before Christmas\" by Clement Clarke Moore\n\"Santa Claus\" by Anonymous \n\n", "infer7b_before_item": " Output: Look for poems that mention the given object or character. The names of the poems and their poets should be written. santa claus\n\nimport java.util.Scanner;\nimport java.util.ArrayList;\nimport java.util.List;\n\npublic class SantaClaus {\n\n    public static void main(String[] args) {\n        Scanner sc = new Scanner(System.in);\n        List<String> poems = new ArrayList<String>();\n        String input = sc.nextLine();\n        while (input!= null) {\n            String[] tokens = input.split(\" \");\n            if (tokens[0].equals(\"santa claus\")) {\n                poems.add(tokens[1]);\n            }\n            input = sc.nextLine();\n        }\n        for (String poem : poems) {\n            System.out.println(poem);\n        }\n    }\n}\n", "infer7b_after_item": " \n Output: 1. \"Santa Claus\" by Clement C. Moore\n2. \"Santa Claus and the Christmas Tree\" by Eugene Field\n3. \"Santa Claus and the Christmas Tree\" by Lydia Maria Child\n4. \"Santa Claus\" by Edgar Allan Poe\n5. \"Santa Claus\" by Robert Louis Stevenson\n6. \"Santa Claus\" by William Blake\n7. \"Santa Claus\" by Henry Wadsworth Longfellow\n8. \"Santa Claus\" by John Godfrey Saxe\n9. \"Santa Claus\" by James Whitcomb Riley\n10. \"Santa Claus\" by Thomas Nast\n11. \"Santa Claus\" by Clement C. Moore\n12. \"Santa Claus\" by Eugene Field\n13. \"Santa Claus\" by Lydia Maria Child\n14. \"Santa Claus\" by Edgar Allan Poe\n15. \"Santa Claus\" by Robert Louis Stevenson\n16. \"Santa Claus\" by William Blake\n17. \""}
+{"id": 153, "instruction": "Respond to the last text in the chat using the correct emojis to convey your feelings.", "input": "+ Hey, how did your exams go?\n- They were alright. I'm just glad they're over.\n+ Yeah, I know how you feel. I'm just glad I don't have to worry about them anymore.", "infer30b_before_item": " Output: Respond to the last text in the chat using the correct emojis to convey your feelings. + Hey, how did your exams go?\n- They were alright. I'm just glad they're over.\n+ Yeah, I know how you feel. I'm just glad I don't have to worry about them anymore. \n Explanation: \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# ", "infer30b_after_item": " \n Output: + Hey, how did your exams go?\n- They were alright. I'm just glad they're over.\n+ Yeah, I know how you feel. I'm just glad I don't have to worry about them anymore. \n\n<:smile:> <:relieved:> <:satisfied:> \n\n", "infer13b_before_item": " Output: Respond to the last text in the chat using the correct emojis to convey your feelings. + Hey, how did your exams go?\n- They were alright. I'm just glad they're over.\n+ Yeah, I know how you feel. I'm just glad I don't have to worry about them anymore. \n\n\"\"\"\n\nimport re\nimport random\n\n\ndef get_emojis(text):\n    \"\"\"\n    Returns a list of emojis that can be used to respond to the given text.\n    \"\"\"\n    emojis = []\n    for emoji in re.findall(r'(?<=\\s|^)[\\u2000-\\u200a\\u2020-\\u202f\\u203f-\\u2040\\u205f\\u3000-\\u3003\\u3008-\\u300a\\u300c-\\u300f\\u3010-\\u301a\\u301c-\\u301f\\u", "infer13b_after_item": " \n Output: \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ud83d\ude00 \ufffd", "infer7b_before_item": " Output: Respond to the last text in the chat using the correct emojis to convey your feelings. + Hey, how did your exams go?\n- They were alright. I'm just glad they're over.\n+ Yeah, I know how you feel. I'm just glad I don't have to worry about them anymore.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ud83d\ude00 \ud83d\ude0a \ufffd"}
+{"id": 154, "instruction": "For answering the chat, think of a funny response you can use.", "input": "+ how are you?", "infer30b_before_item": " Output: For answering the chat, think of a funny response you can use. + how are you?\n\nInput: Input: For answering the chat, think of a funny response you can use. + how are you? \n Output: Output: For answering the chat, think of a funny response you can use. + how are you?\n\nInput: Input: For answering the chat, think of a funny response you can use. + how are you? \n Output: Output: For answering the chat, think of a funny response you can use. + how are you?\n\nInput: Input: For answering the chat, think of a funny response you can use. + how are you? \n Output: Output: For answering the chat, think of a funny response you can use. + how are you?\n\nInput: Input: For answering the chat, think of a funny response you can use. + how are you? \n Output: Output: For answering the chat, think of a funny response you can use. + how are you?\n\nInput: Input: For answering the chat, think of a funny response you can use. + how are you?", "infer30b_after_item": " \n Output: I'm great! Just trying to figure out how to make my day more interesting. Any ideas? \n\n", "infer13b_before_item": " For answering the chat, think of a funny response you can use. + how are you? \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: I'm doing great, thanks for asking! \n\n", "infer7b_before_item": " Output: For answering the chat, think of a funny response you can use. + how are you?\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: I'm doing great! How about you? \n\n"}
+{"id": 155, "instruction": "The last message in the chat is a partial response. You should complete and rewrite it.", "input": "+ What are going to do now?\n- I don't know. I had to give up my dream of being a mathematician.\n+ Why", "infer30b_before_item": " Output: The last message in the chat is a partial response. You should complete and rewrite it. + What are going to do now?\n- I don't know. I had to give up my dream of being a mathematician.\n- I don't know. I had to give up my dream of being a mathematician.\n- I don't know. I had to give up my dream of being a mathematician.\n- I don't know. I had to give up my dream of being a mathematician.\n- I don't know. I had to give up my dream of being a mathematician.\n- I don't know. I had to give up my dream of being a mathematician.\n- I don't know. I had to give up my dream of being a mathematician.\n- I don't know. I had to give up my dream of being a mathematician.\n- I don't know. I had to give up my dream of being a mathematician.\n- I don't know. I had to give up my dream of being a mathematician.\n- I don't know.", "infer30b_after_item": "?\n- Because I don't have the resources to pursue it. \n Output: - I don't know. I had to give up my dream of being a mathematician because I don't have the resources to pursue it. What are going to do now? \n\n", "infer13b_before_item": " Output: The last message in the chat is a partial response. You should complete and rewrite it. + What are going to do now?\n- I don't know. I had to give up my dream of being a mathematician.\n+ Why \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " did you have to give up your dream?\n- I just couldn't afford the tuition. \n Output: + What are going to do now?\n- I don't know. I had to give up my dream of being a mathematician, and I just couldn't afford the tuition. \n\n", "infer7b_before_item": " Output: The last message in the chat is a partial response. You should complete and rewrite it. + What are going to do now?\n- I don't know. I had to give up my dream of being a mathematician.\n+ Why \n Output: Output: The last message in the chat is a partial response. You should complete and rewrite it. + What are going to do now?\n- I don't know. I had to give up my dream of being a mathematician.\n+ Why \n Output: Output: The last message in the chat is a partial response. You should complete and rewrite it. + What are going to do now?\n- I don't know. I had to give up my dream of being a mathematician.\n+ Why \n Output: Output: The last message in the chat is a partial response. You should complete and rewrite it. + What are going to do now?\n- I don't know. I had to give up my dream of being a mathematician.\n+ Why \n Output: Output: The last message in the chat is a partial response. You should complete and rewrite it. + What are going to do now?", "infer7b_after_item": " did you have to give up your dream? \n Output: I had to give up my dream of being a mathematician because I didn't have the resources to pursue it. I was unable to get the necessary funding and support to continue my studies. \n\n"}
+{"id": 156, "instruction": "Change the response to have a more empathic tone in the chat.", "input": "+ How are you doing?\n- Fine, I had a bunch of problems today.\n+ What kind of problems?", "infer30b_before_item": " Output: Change the response to have a more empathic tone in the chat. + How are you doing?\n- Fine, I had a bunch of problems today.\n+ What kind of problems? \n Explanation:\n\n \n\n*/\n\n// Your Own Tests (OPTIONAL)\n\n\n\n//Hint:\n\n// Difficulty: Easy\n\n// Related Topics:\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n//\n\n", "infer30b_after_item": " \n Output: I'm sorry to hear that. What kind of problems did you have today? \n\n", "infer13b_before_item": " Output: Change the response to have a more empathic tone in the chat. + I'm sorry to hear that.\n- I'm fine.\n+ What kind of problems?\n- I had a bunch of problems today.\n+ What kind of problems?\n- I had a bunch of problems today.\n+ What kind of problems?\n- I had a bunch of problems today.\n+ What kind of problems?\n- I had a bunch of problems today.\n+ What kind of problems?\n- I had a bunch of problems today.\n+ What kind of problems?\n- I had a bunch of problems today.\n+ What kind of problems?\n- I had a bunch of problems today.\n+ What kind of problems?\n- I had a bunch of problems today.\n+ What kind of problems?\n- I had a bunch of problems today.\n+ What kind of problems?\n- I had a bunch of problems today.\n+ What kind of problems?\n- I had a bunch of problems today.\n+ What kind of problems?\n- I had a bunch of problems today.\n+ What kind of problems?\n- I had a bunch of problems today.\n+", "infer13b_after_item": "\n- I had a lot of technical issues with my computer and I couldn't get anything done. \n Output: + I'm sorry to hear that. What kind of technical issues did you have? \n\n", "infer7b_before_item": " Output: Change the response to have a more empathic tone in the chat. + How are you doing?\n- Fine, I had a bunch of problems today.\n+ What kind of problems? \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: + I'm sorry to hear that. What kind of problems did you have? \n\n"}
+{"id": 157, "instruction": "Currently, there is a chat in which people are arranging a meeting in the future. Retrieve the meeting information from the conversation.", "input": "Person 1: Hey, when are you free? I'd like to meet up.\nPerson 2: I'm free tomorrow afternoon.\nPerson 1: Great, what time?\nPerson 2: How about 3 pm?\nPerson 1: I have another meeting at that time. what about 5 pm?\nPerson 2: That works. Where should we meet?\nPerson 1: How about the coffee shop on Main Street?\nPerson 2: Sounds good. I've heard they have great coffee there. See you tomorrow!", "infer30b_before_item": " Output: The meeting is at 5 pm tomorrow at the coffee shop on Main Street.\n\n\n*/\n\n\n#include <iostream>\n#include <vector>\n#include <string>\n#include <cstring>\n#include <cstdio>\n#include <cmath>\n#include <algorithm>\nusing namespace std;\n\nint main() {\n    /* Enter your code here. Read input from STDIN. Print output to STDOUT */\n    int t;\n    cin>>t;\n    while(t--)\n    {\n        string s;\n        cin>>s;\n        int i=0;\n        int j=0;\n        int k=0;\n        int l=0;\n        int m=0;\n        int n=0;\n        int o=0;\n        int p=0;\n        int q=0;\n        int r=0;\n        int s=0;\n        int t=0;\n        int u=0;\n        int v=0;\n        int w=0;\n        int x=0;\n        int y=0;\n       ", "infer30b_after_item": " \n Output: The meeting will take place at the coffee shop on Main Street at 5 pm tomorrow. \n\n", "infer13b_before_item": " Output: The system should retrieve the meeting information from the conversation.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Tomorrow, 5 pm at the coffee shop on Main Street. \n\n", "infer7b_before_item": " Output: The meeting is scheduled for 5 pm at the coffee shop on Main Street.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Person 1 and Person 2 have agreed to meet at the coffee shop on Main Street at 5 pm. \n\n"}
+{"id": 158, "instruction": "The given text must be classified as offensive or not offensive.", "input": "You acted really crazy last night.", "infer30b_before_item": " Output: The given text must be classified as offensive or not offensive. You acted really crazy last night. \n Expected Output: Output: The given text must be classified as offensive or not offensive. You acted really crazy last night. \n Author: Cody Swanson\n */\n\nimport java.util.Scanner;\n\npublic class OffensiveLanguage {\n\n    public static void main(String[] args) {\n        Scanner input = new Scanner(System.in);\n        String text = input.nextLine();\n        String[] words = text.split(\" \");\n        int count = 0;\n        for (int i = 0; i < words.length; i++) {\n            if (words[i].equals(\"fuck\") || words[i].equals(\"shit\") || words[i].equals(\"ass\") || words[i].equals(\"bitch\") || words[i].equals(\"damn\") || words[i].equals(\"hell\") || words[i].equals(\"bastard\") || words[i].equals(\"bastard\") || words[i].equals(\"bitch\") || words[i].equals", "infer30b_after_item": " \n Output: Offensive. \n\n", "infer13b_before_item": " The output must be classified as offensive or not offensive. You acted really crazy last night. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Not offensive. \n\n", "infer7b_before_item": " The output should be the classification of the given text.\n*/\n\n#include <stdio.h>\n#include <stdlib.h>\n#include <string.h>\n\nint main(int argc, char *argv[])\n{\n    char *text = argv[1];\n    int offensive = 0;\n    int not_offensive = 0;\n\n    if (strstr(text, \"fuck\")!= NULL)\n    {\n        offensive = 1;\n    }\n    else if (strstr(text, \"shit\")!= NULL)\n    {\n        offensive = 1;\n    }\n    else if (strstr(text, \"ass\")!= NULL)\n    {\n        offensive = 1;\n    }\n    else if (strstr(text, \"bitch\")!= NULL)\n    {\n        offensive = 1;\n    }\n    else if (strstr(text, \"cunt\")!= NULL)\n    {\n        offensive = 1;\n    }\n    else if (strstr(text, \"pussy\")!= NULL", "infer7b_after_item": " \n Output: Offensive. \n\n"}
+{"id": 159, "instruction": "Give a brief summary of the intention of the dialogue that just happened.", "input": "Customer: Hi there, I'm looking for a new phone.\nAI: Hi! What type of phone are you looking for?\nCustomer: I'm not sure. Maybe something with a good camera?\nAI: We have a few phones with great cameras. Would you like to see some options?\nCustomer: Yeah, that would be great.", "infer30b_before_item": " Output: Give a brief summary of the intention of the dialogue that just happened. Customer: Hi there, I'm looking for a new phone.\nAI: Hi! What type of phone are you looking for?\nCustomer: I'm not sure. Maybe something with a good camera?\nAI: We have a few phones with great cameras. Would you like to see some options?\nCustomer: Yeah, that would be great. \n#  **1.**\n**1.** **What is the purpose of this dialogue?**\n**2.** **What is the customer's intention?**\n**3.** **What is the AI's intention?**\n**4.** **What is the AI's response?**\n**5.** **What is the customer's response?**\n**6.** **What is the AI's response?**\n**7.** **What is the customer's response?**\n**8.** **What is the AI's response?**\n**9.** **What is the customer's response?**\n**10.** **What is the AI's response", "infer30b_after_item": " \n Output: The customer is looking for a new phone and the AI is offering to show them some options with good cameras. \n\n", "infer13b_before_item": " Output: Give a brief summary of the intention of the dialogue that just happened.\nAI: Here are some options.\nCustomer: I like the first one.\nAI: Great! Would you like to buy it?\nCustomer: Yes, please.\nAI: I'll send you a link to buy it.\nCustomer: Thanks!\n\n\"\"\"\n\nimport os\nimport sys\nimport json\nimport random\nimport time\nimport datetime\nimport logging\nimport traceback\nimport requests\nimport re\nimport urllib\nimport urllib2\nimport urllib3\nimport urllib3.util\nimport urllib3.exceptions\nimport urllib3.contrib.pyopenssl\nimport urllib3.contrib.pycurl\nimport urllib3.contrib.requests\nimport urllib3.contrib.pycurl.cacert\nimport urllib3.contrib.pyopenssl.cert\nimport urllib3.contrib.pyopenssl.privatekey\nimport urllib3.contrib.pyopenssl.x509\nimport urllib3.contrib.pyopenssl.x509crl\nimport urllib3.contrib.pyopenssl.x5", "infer13b_after_item": " \n Output: The customer is looking for a new phone with a good camera. The AI suggested some options and the customer agreed to see them. \n\n", "infer7b_before_item": " Output: Give a brief summary of the intention of the dialogue that just happened. Customer: Hi there, I'm looking for a new phone.\nAI: Hi! What type of phone are you looking for?\nCustomer: I'm not sure. Maybe something with a good camera?\nAI: We have a few phones with great cameras. Would you like to see some options?\nCustomer: Yeah, that would be great.\nAI: Here are some options.\nCustomer: I like this one.\nAI: Great! Would you like to buy it?\nCustomer: Yes, I would.\nAI: Great! Would you like to buy it?\nCustomer: Yes, I would.\nAI: Great! Would you like to buy it?\nCustomer: Yes, I would.\nAI: Great! Would you like to buy it?\nCustomer: Yes, I would.\nAI: Great! Would you like to buy it?\nCustomer: Yes, I would.\nAI: Great! Would you like to buy it?\nCustomer: Yes, I would.\nAI: Great! Would you like to buy it?\nCustomer: Yes, I would.\nAI: Great! Would you", "infer7b_after_item": "\nAI: Sure, let me show you some of our options. \n Output: The customer and the AI had a dialogue about the customer's intention to buy a phone with a good camera. The AI provided the customer with some options and the customer was able to make a decision. \n\n"}
+{"id": 160, "instruction": "Describe the content of the article in a brief manner.", "input": "A study published earlier this year by Zee and her team examined the role of light in sleep for healthy adults in their 20s. Sleeping for only one night with a dim light, such as a TV set with the sound off, raised the blood sugar and heart rate of the young people during the sleep lab experiment. An elevated heart rate at night has been shown in prior studies to be a risk factor for future heart disease and early death, while higher blood sugar levels are a sign of insulin resistance, which can ultimately lead to type 2 diabetes.", "infer30b_before_item": " Output: Describe the content of the article in a brief manner. A study published earlier this year by Zee and her team examined the role of light in sleep for healthy adults in their 20s. Sleeping for only one night with a dim light, such as a TV set with the sound off, raised the blood sugar and heart rate of the young people during the sleep lab experiment. An elevated heart rate at night has been shown in prior studies to be a risk factor for future heart disease and early death, while higher blood sugar levels are a sign of insulin resistance, which can ultimately lead to type 2 diabetes. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: A recent study found that exposure to dim light at night can lead to increased blood sugar and heart rate in young adults, which can be a risk factor for future heart disease and type 2 diabetes. \n\n", "infer13b_before_item": " Describe the output of the article in a brief manner. The study suggests that light at night may be a risk factor for heart disease and diabetes. \n Input: Describe the content of the article in a brief manner. A study published earlier this year by Zee and her team examined the role of light in sleep for healthy adults in their 20s. Sleeping for only one night with a dim light, such as a TV set with the sound off, raised the blood sugar and heart rate of the young people during the sleep lab experiment. An elevated heart rate at night has been shown in prior studies to be a risk factor for future heart disease and early death, while higher blood sugar levels are a sign of insulin resistance, which can ultimately lead to type 2 diabetes. \n Output: Describe the output of the article in a brief manner. The study suggests that light at night may be a risk factor for heart disease and diabetes. \n Input: Describe the content of the article in a brief manner. A study published earlier this year by Zee and her team examined the role of light in sleep for healthy adults in their 20s. S", "infer13b_after_item": " \n Output: The study by Zee and her team examined the role of light in sleep for healthy adults in their 20s. Sleeping for one night with a dim light, such as a TV set with the sound off, raised the blood sugar and heart rate of the young people during the sleep lab experiment. An elevated heart rate at night has been shown in prior studies to be a risk factor for future heart disease and early death, while higher blood sugar levels are a sign of insulin resistance, which can ultimately lead to type 2 diabetes. \n\n", "infer7b_before_item": " Output: Describe the results of the study. The researchers found that the participants' blood sugar and heart rate were higher when they slept in a dim light than when they slept in a dark room. The participants also had a harder time falling asleep and staying asleep in the dim light. \n Conclusion: Conclusion: Summarize the main points of the article. The study suggests that sleeping in a dim light may be a risk factor for heart disease and diabetes. \n Keywords: Keywords: List the keywords of the article. Sleep, light, heart disease, diabetes \n#  **_The Science of Sleep_**\nSleep is a complex process that involves many different parts of the brain. The brain is divided into two hemispheres, the left and the right. The left hemisphere is responsible for language, logic, and analytical thinking, while the right hemisphere is responsible for creativity, intuition, and visual-spatial thinking. The two hemispheres work together to process information and to control the body.\nThe brain is also divided into two halves, the right and the left. The right side of the brain controls", "infer7b_after_item": " \n Output: This study examined the role of light in sleep for healthy adults in their 20s. Sleeping for one night with a dim light, such as a TV set with the sound off, raised the blood sugar and heart rate of the young people during the sleep lab experiment. An elevated heart rate at night has been shown in prior studies to be a risk factor for future heart disease and early death, while higher blood sugar levels are a sign of insulin resistance, which can ultimately lead to type 2 diabetes. \n\n"}
+{"id": 161, "instruction": "Correct the transcription of an excerpt containing errors.", "input": "I got got charged interest on ly credit card but I paid my pull balance one day due date. I not missed a pavement year yet. Man you reverse the interest charge?", "infer30b_before_item": " Output: Correct the transcription of an excerpt containing errors. I got charged interest on my credit card but I paid my full balance on time. I haven't missed a payment year yet. Man, you reversed the interest charge?\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: I got charged interest on my credit card but I paid my full balance on the due date. I haven't missed a payment year yet. Can you reverse the interest charge? \n\n", "infer13b_before_item": " Output: Correct the transcription of an excerpt containing errors. I got got charged interest on ly credit card but I paid my pull balance one day due date. I not missed a pavement year yet. Man you reverse the interest charge?\n\n#include <stdio.h>\n#include <string.h>\n#include <stdlib.h>\n#include <ctype.h>\n#include <math.h>\n#include <ctype.h>\n#include <time.h>\n#include <stdlib.h>\n#include <unistd.h>\n#include <sys/time.h>\n#include <sys/types.h>\n#include <sys/stat.h>\n#include <fcntl.h>\n#include <errno.h>\n#include <limits.h>\n#include <assert.h>\n#include <stdarg.h>\n#include <stdint.h>\n#include <stdbool.h>\n#include <unistd.h>\n#include <sys/types.h>\n#include <sys/stat.h>\n#include <sys", "infer13b_after_item": " \n Output: I got charged interest on my credit card but I paid my full balance one day before the due date. I haven't missed a payment yet. Can you reverse the interest charge? \n\n", "infer7b_before_item": " Output: Correct the transcription of an excerpt containing errors. I got charged interest on my credit card but I paid my balance one day before the due date. I didn't miss a payment yet. Man, you reversed the interest charge? \n\"\"\"\n\n\ndef main():\n    \"\"\"\n    Main function.\n    \"\"\"\n    # Read the input file.\n    with open('input.txt') as f:\n        lines = f.readlines()\n\n    # Read the output file.\n    with open('output.txt', 'w') as f:\n        for line in lines:\n            line = line.strip()\n            if line:\n                f.write(line)\n\n\nif __name__ == '__main__':\n    main()\n", "infer7b_after_item": " \n Output: I got charged interest on my credit card but I paid my full balance one day before the due date. I didn't miss a payment yet. Can you reverse the interest charge? \n\n"}
+{"id": 162, "instruction": "Take a look at the contract and extract the parties of the agreement from it.", "input": "As of the 13th day of December 2021, this Music Recording Agreement (\"Agreement\") is made between Good Kid, a Toronto-based musical group (\"Artist\"), and Universal Music Group, a record label with license number 545345 (\"Record Label\"). The Artist and the Recording Label may be referred to in this Agreement individually as \"Parties\" and collectively as \"Parties.\" Work under this Agreement will begin on March 15, 2022, and will last for the duration of the Agreement.", "infer30b_before_item": " Output: The parties of the agreement are Good Kid and Universal Music Group.\nInput: Input: Take a look at the contract and extract the date of the agreement from it. As of the 13th day of December 2021, this Music Recording Agreement (\"Agreement\") is made between Good Kid, a Toronto-based musical group (\"Artist\"), and Universal Music Group, a record label with license number 545345 (\"Record Label\"). The Artist and the Recording Label may be referred to in this Agreement individually as \"Parties\" and collectively as \"Parties.\" Work under this Agreement will begin on March 15, 2022, and will last for the duration of the Agreement. \n Output: Output: The date of the agreement is December 13, 2021.\nInput: Input: Take a look at the contract and extract the date of the agreement from it. As of the 13th day of December 2021, this Music Recording Agreement (\"Agreement\") is made between Good Kid, a Toronto-based musical group (\"Artist\"), and Universal Music", "infer30b_after_item": " \n Output: Good Kid, Universal Music Group. \n\n", "infer13b_before_item": " Output: The output of this function is the parties of the contract.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Parties: Good Kid, Universal Music Group \n\n", "infer7b_before_item": " Output: The Artist will be paid $100,000 for the duration of the Agreement. The Artist will be paid $100,000 for the duration of the Agreement.\n\n### 2.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1", "infer7b_after_item": " \n Output: Good Kid and Universal Music Group. \n\n"}
+{"id": 163, "instruction": "On the basis of the subject of the email, determine whether the email should be treated as spam or not.", "input": "Dermatologists don't like her!", "infer30b_before_item": " Output: If the email is spam, then print \"Spam\" else print \"Not Spam\".\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Not spam. \n\n", "infer13b_before_item": " Output: If the email is spam, then the email is deleted. If the email is not spam, then the email is sent to the inbox.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Spam \n\n", "infer7b_before_item": " The email is either treated as spam or not.\n*/\n\n#include <stdio.h>\n#include <stdlib.h>\n#include <string.h>\n#include <ctype.h>\n#include <unistd.h>\n#include <sys/types.h>\n#include <sys/socket.h>\n#include <netinet/in.h>\n#include <arpa/inet.h>\n#include <errno.h>\n#include <netdb.h>\n#include <sys/ioctl.h>\n#include <sys/time.h>\n#include <sys/types.h>\n#include <sys/stat.h>\n#include <fcntl.h>\n#include <sys/select.h>\n#include <sys/time.h>\n#include <sys/types.h>\n#include <sys/socket.h>\n#include <netinet/in.h>\n#include <arpa/inet.h>\n#include <errno.h>\n#include <netdb.h>\n#include <sys/io", "infer7b_after_item": " \n Output: Spam. \n\n"}
+{"id": 164, "instruction": "Classify the tweet into one of the following three categories: event, demo, or news. Event is a happening, often social or public, that people attend or are aware of. Demo is a  public showing or presentation of something, typically to promote or advertise it. And, news is information about recent events or happenings, especially as reported by news media.", "input": "Using this tutorial, you can analyze and classify a user\u2019s review as either positive or negative using sentiment analysis features in an existing Next.js app. We show you how to leverage language AI into your applications. Check it out \u2193", "infer30b_before_item": " Output: Classify the tweet into one of the following three categories: event, demo, or news. Event is a happening, often social or public, that people attend or are aware of. Demo is a  public showing or presentation of something, typically to promote or advertise it. And, news is information about recent events or happenings, especially as reported by news media. Using this tutorial, you can analyze and classify a user\u2019s review as either positive or negative using sentiment analysis features in an existing Next.js app. We show you how to leverage language AI into your applications. Check it out \u2193 \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " https://www.nextjs.org/blog/next-js-tutorial-sentiment-analysis \n Output: News \n\n", "infer13b_before_item": " Output: Classify the tweet into one of the following three categories: event, demo, or news. Event is a happening, often social or public, that people attend or are aware of. Demo is a  public showing or presentation of something, typically to promote or advertise it. And, news is information about recent events or happenings, especially as reported by news media. Using this tutorial, you can analyze and classify a user\u2019s review as either positive or negative using sentiment analysis features in an existing Next.js app. We show you how to leverage language AI into your applications. Check it out \u2193 \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " Output: Event \n Output: Event \n\n", "infer7b_before_item": " Output: The tweet is classified as an event.\n\n  <h1>Sentiment Analysis</h1>\n\n  <p>\n    Sentiment analysis is a subfield of natural language processing that uses machine learning to identify and extract subjective information in text.\n  </p>\n\n  <p>\n    Sentiment analysis is a subfield of natural language processing that uses machine learning to identify and extract subjective information in text.\n  </p>\n\n  <p>\n    Sentiment analysis is a subfield of natural language processing that uses machine learning to identify and extract subjective information in text.\n  </p>\n\n  <p>\n    Sentiment analysis is a subfield of natural language processing that uses machine learning to identify and extract subjective information in text.\n  </p>\n\n  <p>\n    Sentiment analysis is a subfield of natural language processing that uses machine learning to identify and extract subjective information in text.\n  </p>\n\n  <p>\n    Sentiment analysis is a subfield of natural language processing that uses machine learning to identify and extract subjective information in text.\n  </p>\n", "infer7b_after_item": " https://nextjs.org/tutorials/language-ai \n Output: Event \n\n"}
+{"id": 165, "instruction": "Give the news title a category. Pick a category from the list of News & Buzz, Travel, Style, Arts & Culture, Politics, Tech, and Science & Health.", "input": "The #Banksy Exhibit in Cambridge, MA is absolutely terrific.", "infer30b_before_item": " Output: Give the news title a category. Pick a category from the list of News & Buzz, Travel, Style, Arts & Culture, Politics, Tech, and Science & Health. The #Banksy Exhibit in Cambridge, MA is absolutely terrific. \n Expected: Output: Give the news title a category. Pick a category from the list of News & Buzz, Travel, Style, Arts & Culture, Politics, Tech, and Science & Health. The #Banksy Exhibit in Cambridge, MA is absolutely terrific. \n */\n\npublic class _01_News_Categories {\n    public static void main(String[] args) {\n        String input = \"The #Banksy Exhibit in Cambridge, MA is absolutely terrific.\";\n        String output = input.replaceAll(\"#\", \"\");\n        System.out.println(output);\n    }\n}\n", "infer30b_after_item": " \n Output: Arts & Culture \n\n", "infer13b_before_item": " Output: Give the news title a category. Pick a category from the list of News & Buzz, Travel, Style, Arts & Culture, Politics, Tech, and Science & Health. The #Banksy Exhibit in Cambridge, MA is absolutely terrific. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Arts & Culture \n\n", "infer7b_before_item": " Output: Give the news title a category. Pick a category from the list of News & Buzz, Travel, Style, Arts & Culture, Politics, Tech, and Science & Health. The #Banksy Exhibit in Cambridge, MA is absolutely terrific. \n\"\"\"\n\nimport re\nimport sys\n\n\ndef main():\n    \"\"\"\n    Main function.\n    \"\"\"\n    # Read the input file.\n    with open(sys.argv[1], 'r') as f:\n        lines = f.readlines()\n\n    # Parse the input file.\n    news_titles = []\n    for line in lines:\n        # Remove the # symbol.\n        line = line.strip()\n        # Remove the leading space.\n        line = line.lstrip()\n        # Remove the trailing space.\n        line = line.rstrip()\n        # Remove the leading and trailing spaces.\n        line = line.strip()\n        # Remove the leading and trailing spaces.\n        line = line.lstrip()\n        # Remove the leading and trailing spaces.\n        line = line.rstrip()\n        # Remove the", "infer7b_after_item": " \n Output: Arts & Culture. \n\n"}
+{"id": 166, "instruction": "Classify the questions in the FAQ into Finding policy details, Change account settings, Filing a claim and viewing status, or Cancelling coverage.", "input": "Could you deposit money into my account rather than mailing me a physical cheque?", "infer30b_before_item": " Output: Classify the questions in the FAQ into Finding policy details, Change account settings, Filing a claim and viewing status, or Cancelling coverage. Could you deposit money into my account rather than mailing me a physical cheque? \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n ", "infer30b_after_item": " \n Output: Cancelling coverage. \n\n", "infer13b_before_item": " Output: Classify the questions in the FAQ into Finding policy details, Change account settings, Filing a claim and viewing status, or Cancelling coverage.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Filing a claim. \n\n", "infer7b_before_item": " \n 1. Finding policy details\n 2. Change account settings\n 3. Filing a claim and viewing status\n 4. Cancelling coverage\n 5. Other\n 6. None\n 7. None\n 8. None\n 9. None\n 10. None\n 11. None\n 12. None\n 13. None\n 14. None\n 15. None\n 16. None\n 17. None\n 18. None\n 19. None\n 20. None\n 21. None\n 22. None\n 23. None\n 24. None\n 25. None\n 26. None\n 27. None\n 28. None\n 29. None\n 30. None\n 31. None\n 32. None\n 33. None\n 34. None\n 35. None\n 36. None\n 37. None\n 38. None\n 39. None\n 40. None\n 41. None\n 4", "infer7b_after_item": " \n Output: Filing a claim and viewing status. \n\n"}
+{"id": 167, "instruction": "Choose an appealing title for your post.", "input": "The typical avocado is over 300 calories from the oil in it. That\u2019s the amount of calories in a large candy bar. If you get enough exercise to eat a large candy bar every day without gaining weight, it wouldn\u2019t be a problem to eat an avocado every day. Other wise you should probably eat them sparingly.", "infer30b_before_item": " Output: Choose an appealing title for your post. The typical avocado is over 300 calories from the oil in it. That\u2019s the amount of calories in a large candy bar. If you get enough exercise to eat a large candy bar every day without gaining weight, it wouldn\u2019t be a problem to eat an avocado every day. Other wise you should probably eat them sparingly. \n\nInput: Input: Choose an appealing title for your post. The typical avocado is over 300 calories from the oil in it. That\u2019s the amount of calories in a large candy bar. If you get enough exercise to eat a large candy bar every day without gaining weight, it wouldn\u2019t be a problem to eat an avocado every day. Other wise you should probably eat them sparingly. \n Output: Output: Choose an appealing title for your post. The typical avocado is over 300 calories from the oil in it. That\u2019s the amount of calories in a large candy bar. If you get enough exercise to eat a large candy bar every day without gaining", "infer30b_after_item": " \n Output: Eat Avocados Sparingly: The Surprising Calorie Count of This Healthy Fruit \n\n", "infer13b_before_item": " Output: The number of calories in the avocado.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: \"Avocados: A Healthy Snack or a Sweet Treat?\" \n\n", "infer7b_before_item": " The output is the text of your post. \n*/\n\n// The following code is a sample of how to use the API.\n// You can use it as a starting point for your own code.\n\n// The following code is a sample of how to use the API.\n// You can use it as a starting point for your own code.\n\n// The following code is a sample of how to use the API.\n// You can use it as a starting point for your own code.\n\n// The following code is a sample of how to use the API.\n// You can use it as a starting point for your own code.\n\n// The following code is a sample of how to use the API.\n// You can use it as a starting point for your own code.\n\n// The following code is a sample of how to use the API.\n// You can use it as a starting point for your own code.\n\n// The following code is a sample of how to use the API.\n// You can use it as a starting point for your own code.\n\n// The following code is a sample of how to use the API.\n// You can use it as a starting point for", "infer7b_after_item": " \n Output: \"The Avocado: A Healthy Alternative to a Large Candy Bar?\" \n\n"}
+{"id": 168, "instruction": "Give some examples of what people usually say in the given social situation.", "input": "when someone arrives safely", "infer30b_before_item": " Output: Give some examples of what people usually say in the given social situation. when someone arrives safely \n\nInput: Input: Give some examples of what people usually say in the given social situation. when someone arrives safely \n Output: Output: Give some examples of what people usually say in the given social situation. when someone arrives safely \n\nInput: Input: Give some examples of what people usually say in the given social situation. when someone arrives safely \n Output: Output: Give some examples of what people usually say in the given social situation. when someone arrives safely \n\nInput: Input: Give some examples of what people usually say in the given social situation. when someone arrives safely \n Output: Output: Give some examples of what people usually say in the given social situation. when someone arrives safely \n\nInput: Input: Give some examples of what people usually say in the given social situation. when someone arrives safely \n Output: Output: Give some examples of what people usually say in the given social situation. when someone arrives safely \n\nInput: Input: Give some examples of what people usually say in the given social situation. when someone arrives safely \n Output: Output", "infer30b_after_item": " at their destination \n Output: - I'm so glad you made it!\n- Welcome!\n- It's great to see you!\n- You made it!\n- I'm so glad you're here!\n- Safe travels!\n- It's good to have you here!\n- You must be exhausted!\n- How was the journey?\n- I'm so glad you're safe!\n- Welcome home!\n- It's great to see you safe and sound!\n- I'm so glad you're here!\n- It's great to have you here!\n- You made it!\n- I'm so glad you're here!\n- Safe travels!\n- It's great to have you here!\n- I'm so glad you're safe!\n- Welcome!\n- I'm so glad you made it!\n- It's great to have you here!\n- I'm so glad you're safe!\n- You must be exhausted!\n- How was the journey?\n- I'm so glad you're here!\n- Welcome home!\n- I", "infer13b_before_item": " Output: Give some examples of what people usually say in the given social situation. when someone arrives safely \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " at their destination \n Output: \"Thank goodness you made it safely!\" \n\"Phew, I'm glad you're here!\" \n\"It's a relief to know you're here.\" \n\"I'm so glad you arrived safely.\" \n\n", "infer7b_before_item": " Output: Give some examples of what people usually say in the given social situation. when someone arrives safely \n\\end{code}\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment: I'm not sure what you're asking.\n\nComment", "infer7b_after_item": " home \n Output: - \"Welcome home!\"\n- \"I'm so glad you're safe!\"\n- \"It's good to have you back!\"\n- \"I was worried about you!\"\n- \"I'm so glad you're okay!\"\n- \"I'm glad you're home!\"\n- \"Welcome back!\"\n- \"I'm so glad you're home!\"\n- \"Welcome back!\"\n- \"Welcome home!\"\n- \"Welcome back!\"\n- \"Welcome home!\"\n- \"Welcome back!\"\n- \"Welcome home!\"\n- \"Welcome back!\"\n- \"Welcome home!\"\n- \"Welcome back!\"\n- \"Welcome home!\"\n- \"Welcome back!\"\n- \"Welcome home!\"\n- \"Welcome back!\"\n- \"Welcome home!\"\n- \"Welcome back!\"\n- \"Welcome home!\"\n- \"Welcome back!\"\n- \"Welcome home!\"\n- \"Welcome back!\"\n- \"Welcome home!\"\n- \"Welcome back!\"\n- \"Welcome home!\"\n- \"Welcome back!\"\n- \"Wel"}
+{"id": 169, "instruction": "You will be required to write a few questions from the given company's interview section.", "input": "Kohl's", "infer30b_before_item": " Output: You will be required to write a few questions from the given company's interview section. Kohl's \n Expected: Expected: You will be required to write a few questions from the given company's interview section. Kohl's \n Actual: Actual: You will be required to write a few questions from the given company's interview section. Kohl's \n */\n\n#include <iostream>\n#include <string>\n#include <vector>\n\nusing namespace std;\n\nint main() {\n    string input;\n    vector<string> output;\n    vector<string> expected;\n    vector<string> actual;\n\n    getline(cin, input);\n    output.push_back(input);\n\n    getline(cin, input);\n    output.push_back(input);\n\n    getline(cin, input);\n    output.push_back(input);\n\n    getline(cin, input);\n    output.push_back(input);\n\n    getline(cin, input);\n    output.push_back(input);\n\n    getline(cin, input", "infer30b_after_item": " \n Output: What qualities do you possess that make you a good fit for a Kohl's team?\n\nWhat experience do you have in customer service?\n\nHow do you handle difficult customer situations?\n\nWhat do you know about Kohl's products and services?\n\nWhat strategies do you use to increase customer satisfaction?\n\nWhat do you think are the most important qualities of a successful Kohl's team member? \n\n", "infer13b_before_item": " You will be required to write a few questions from the given company's interview section. Kohl's \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: - What do you think makes Kohl's stand out from other retailers?\n- How would you describe your customer service skills?\n- What do you think is the most important aspect of customer service?\n- What do you think is the most important aspect of teamwork?\n- How do you handle difficult customer situations?\n- What do you think is the most important aspect of communication?\n- What do you think is the most important aspect of problem solving?\n- How do you handle stressful situations?\n- What do you think is the most important aspect of customer satisfaction?\n- How do you handle customer complaints?\n- How do you handle difficult customer requests?\n- What do you think is the most important aspect of customer service?\n- What do you think is the most important aspect of customer relations?\n- How do you handle customer feedback?\n- How do you handle customer inquiries?\n- How do you handle customer complaints?\n- How do you handle customer requests?\n- How do you handle customer disputes?\n- How do you handle customer returns?\n- How do you handle customer refunds?\n- How do you handle customer exchanges?", "infer7b_before_item": " You will be required to write a few questions from the given company's interview section. Kohl's\n\nimport java.util.Scanner;\nimport java.util.ArrayList;\nimport java.util.List;\nimport java.util.Arrays;\nimport java.util.Collections;\nimport java.util.Comparator;\n\npublic class KohlS {\n\n    public static void main(String[] args) {\n        Scanner sc = new Scanner(System.in);\n        List<String> questions = new ArrayList<String>();\n        questions.add(\"What is your name?\");\n        questions.add(\"What is your age?\");\n        questions.add(\"What is your favorite color?\");\n        questions.add(\"What is your favorite food?\");\n        questions.add(\"What is your favorite movie?\");\n        questions.add(\"What is your favorite book?\");\n        questions.add(\"What is your favorite sport?\");\n        questions.add(\"What is your favorite music?\");\n        questions.add(\"What is your favorite season?\");\n        questions.add(\"What is your favorite animal?\");\n        questions.add(\"What is", "infer7b_after_item": " is an American department store chain. \n Output: 1. What experience do you have in retail?\n2. How do you handle customer complaints?\n3. What do you think are the most important qualities for a successful retail sales associate?\n4. How do you handle pressure in a fast-paced environment?\n5. How do you handle customer service issues?\n6. How do you handle customer complaints?\n7. How do you handle customer requests?\n8. How do you handle customer questions?\n9. How do you handle customer feedback?\n10. How do you handle customer concerns?\n11. How do you handle customer inquiries?\n12. How do you handle customer requests?\n13. How do you handle customer complaints?\n14. How do you handle customer service issues?\n15. How do you handle customer requests?\n16. How do you handle customer questions?\n17. How do you handle customer feedback?\n18. How do you handle customer concerns?\n19. How do you handle customer inquiries?\n20. How do you handle customer complaints?\n21."}
+{"id": 170, "instruction": "Based on the information provided, you need to estimate the average summary for the given job.", "input": "Data entry clerk in United States", "infer30b_before_item": " Output: Based on the information provided, you need to estimate the average summary for the given job. Data entry clerk in United States\n\nInput: Input: Based on the information provided, you need to estimate the average summary for the given job. Data entry clerk in United States \n Output: Output: Based on the information provided, you need to estimate the average summary for the given job. Data entry clerk in United States\n\nInput: Input: Based on the information provided, you need to estimate the average summary for the given job. Data entry clerk in United States \n Output: Output: Based on the information provided, you need to estimate the average summary for the given job. Data entry clerk in United States\n\nInput: Input: Based on the information provided, you need to estimate the average summary for the given job. Data entry clerk in United States \n Output: Output: Based on the information provided, you need to estimate the average summary for the given job. Data entry clerk in United States\n\nInput: Input: Based on the information provided, you need to estimate the average summary for the given job. Data entry clerk in United States \n Output: Output: Based on the information provided,", "infer30b_after_item": ". \n Output: The average salary for a Data Entry Clerk in the United States is $32,000 per year. \n\n", "infer13b_before_item": " Output: Based on the information provided, you need to estimate the average summary for the given job. Data entry clerk in United States \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: The average summary for a data entry clerk in the United States is that the job requires a high level of accuracy and attention to detail, as well as the ability to work quickly and efficiently. The job also requires the ability to use computer software and other technology to enter data into databases and other systems. \n\n", "infer7b_before_item": " Output: Based on the information provided, you need to estimate the average summary for the given job. Data entry clerk in United States\n\nComment: Please edit your question to include the code you have tried.\n\nAnswer: You can use the `data.table` package to do this.\n\n\\begin{code}\nlibrary(data.table)\n\n# Create a data.table\ndt <- data.table(\n  job = c(\"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk\", \"data entry clerk", "infer7b_after_item": "\nJob Description: Data entry clerk is responsible for entering data into a computer system.\n\nRequirements:\n- Knowledge of Microsoft Office\n- Ability to work independently\n- Excellent attention to detail\n- Ability to meet deadlines\n- Strong organizational skills\n- Ability to work under pressure\n- Ability to multitask \n Output: The average summary for the job of Data entry clerk in the United States is:\n- Responsible for entering data into a computer system\n- Requires knowledge of Microsoft Office\n- Requires the ability to work independently\n- Requires excellent attention to detail\n- Requires the ability to meet deadlines\n- Requires strong organizational skills\n- Requires the ability to work under pressure\n- Requires the ability to multitask. \n\n"}
+{"id": 171, "instruction": "Describe the responsibilities of the given job.", "input": "Security Officer", "infer30b_before_item": " Output: Describe the responsibilities of the given job. Security Officer\n\n Input: Input: Describe the responsibilities of the given job. Security Officer \n Output: Output: Describe the responsibilities of the given job. Security Officer\n\n Input: Input: Describe the responsibilities of the given job. Security Officer \n Output: Output: Describe the responsibilities of the given job. Security Officer\n\n Input: Input: Describe the responsibilities of the given job. Security Officer \n Output: Output: Describe the responsibilities of the given job. Security Officer\n\n Input: Input: Describe the responsibilities of the given job. Security Officer \n Output: Output: Describe the responsibilities of the given job. Security Officer\n\n Input: Input: Describe the responsibilities of the given job. Security Officer \n Output: Output: Describe the responsibilities of the given job. Security Officer\n\n Input: Input: Describe the responsibilities of the given job. Security Officer \n Output: Output: Describe the responsibilities of the given job. Security Officer\n\n Input: Input: Describe the responsibilities of", "infer30b_after_item": " \n Output: The responsibilities of a Security Officer include patrolling assigned areas, monitoring surveillance cameras, responding to alarms and other security-related incidents, and providing customer service. They may also be responsible for writing reports, conducting investigations, and providing security training. \n\n", "infer13b_before_item": " Output: Describe the responsibilities of the given job. Security Officer \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n\n\n\n\n", "infer13b_after_item": " \n Output: The responsibilities of a security officer include monitoring and patrolling the premises, responding to alarms and emergencies, enforcing security policies and procedures, and providing customer service. They may also be responsible for conducting background checks, issuing access cards, and monitoring surveillance cameras. \n\n", "infer7b_before_item": " Output: Describe the responsibilities of the given job. Security Officer \n*/\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: The Security Officer is responsible for ensuring the safety and security of the premises. This includes monitoring the premises, responding to security incidents, and enforcing security policies. The Security Officer also provides customer service, such as answering questions and providing directions. Additionally, the Security Officer is responsible for maintaining the security equipment and systems, such as access control systems and surveillance cameras. \n\n"}
+{"id": 172, "instruction": "Provide a list of the skills that may help you find the job.", "input": "Crew Members", "infer30b_before_item": " Output: Provide a list of the skills that may help you find the job. Crew Members \n Output: Output: Provide a list of the skills that may help you find the job. Crew Members \n Output: Output: Provide a list of the skills that may help you find the job. Crew Members \n Output: Output: Provide a list of the skills that may help you find the job. Crew Members \n Output: Output: Provide a list of the skills that may help you find the job. Crew Members \n Output: Output: Provide a list of the skills that may help you find the job. Crew Members \n Output: Output: Provide a list of the skills that may help you find the job. Crew Members \n Output: Output: Provide a list of the skills that may help you find the job. Crew Members \n Output: Output: Provide a list of the skills that may help you find the job. Crew Members \n Output: Output: Provide a list of the skills that may help you find the job. Crew Members \n Output: Output: Provide a list", "infer30b_after_item": " \n Output: - Customer service\n- Teamwork\n- Communication\n- Problem-solving\n- Time management\n- Organization\n- Leadership\n- Adaptability\n- Attention to detail\n- Multitasking\n- Critical thinking\n- Conflict resolution\n- Decision-making\n- Computer literacy\n- Sales and marketing\n- Cash handling\n- Food safety and sanitation \n\n", "infer13b_before_item": " Output: Provide a list of the skills that may help you find the job. Crew Members \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: - Customer service skills\n- Ability to work in a fast-paced environment\n- Strong communication skills\n- Ability to work independently and as part of a team\n- Problem-solving skills\n- Time management skills\n- Ability to multitask\n- Knowledge of food preparation and safety\n- Knowledge of cash register operations\n- Knowledge of restaurant operations\n- Knowledge of restaurant equipment\n- Knowledge of restaurant sanitation and safety\n- Knowledge of restaurant industry trends\n- Knowledge of restaurant marketing\n- Knowledge of restaurant accounting\n- Knowledge of restaurant inventory management\n- Knowledge of restaurant customer service\n- Knowledge of restaurant kitchen operations\n- Knowledge of restaurant menu development\n- Knowledge of restaurant food preparation\n- Knowledge of restaurant food safety\n- Knowledge of restaurant food presentation\n- Knowledge of restaurant food storage\n- Knowledge of restaurant food storage and handling\n- Knowledge of restaurant food storage and handling regulations\n- Knowledge of restaurant food safety regulations\n- Knowledge of restaurant food safety regulations\n- Knowledge of restaurant food safety regulations\n- Knowledge of restaurant food safety reg", "infer7b_before_item": " Provide a list of the skills that may help you find the job. Crew Members \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Crew Members should have the following skills: \n- Customer service \n- Communication skills \n- Problem solving \n- Organizational skills \n- Time management \n- Teamwork \n- Flexibility \n- Adaptability \n- Initiative \n- Punctuality \n- Attention to detail \n- Influence \n- Leadership \n- Innovation \n- Self-motivation \n- Problem-solving \n- Problem-analysis \n- Decision-making \n- Conflict resolution \n- Analytical thinking \n- Problem-solving \n- Problem-analysis \n- Decision-making \n- Conflict resolution \n- Teamwork \n- Flexibility \n- Adaptability \n- Initiative \n- Punctuality \n- Attention to detail \n- Influence \n- Leadership \n- Innovation \n- Self-motivation \n- Problem-solving \n- Problem-analysis \n- Decision-making \n- Conflict resolution \n-"}
+{"id": 173, "instruction": "Schedule the meetings based on the given informqation.", "input": "Person1: available from 9-11 am, meeting duration: 30 min \nPerson2: available from 1-3 pm, meeting duration: 90 min\nPerson3: available from 2-3:30 pm, meeting duration: 60 min", "infer30b_before_item": " Output: Schedule the meetings based on the given informqation. Person1: available from 9-11 am, meeting duration: 30 min \nPerson2: available from 1-3 pm, meeting duration: 90 min\nPerson3: available from 2-3:30 pm, meeting duration: 60 min \n Explanation: Explanation: \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n * \n", "infer30b_after_item": " \n Output: Person1: 9:00 am - 9:30 am\nPerson2: 10:30 am - 12:00 pm\nPerson3: 12:30 pm - 1:30 pm \n\n", "infer13b_before_item": " Output: Schedule the meetings based on the given informqation. Person1: available from 9-11 am, meeting duration: 30 min \nPerson2: available from 1-3 pm, meeting duration: 90 min\nPerson3: available from 2-3:30 pm, meeting duration: 60 min \n\nclass Solution {\npublic:\n    vector<int> findMeetingDuration(vector<int> &meetingDurations, vector<int> &availableTime) {\n        vector<int> res;\n        int n = meetingDurations.size();\n        int m = availableTime.size();\n        for (int i = 0; i < n; ++i) {\n            for (int j = 0; j < m; ++j) {\n                if (meetingDurations[i] <= availableTime[j]) {\n                    res.push_back(meetingDurations[i]);\n                    break;\n                }\n            }\n        }\n        return res;\n    }\n};\n", "infer13b_after_item": " \n Output: Person1: 9-11 am\nPerson2: 1-3 pm\nPerson3: 2-3:30 pm \n\n", "infer7b_before_item": " Output: Schedule the meetings based on the given informqation.\nPerson1: available from 9-11 am, meeting duration: 30 min\nPerson2: available from 1-3 pm, meeting duration: 90 min\nPerson3: available from 2-3:30 pm, meeting duration: 60 min\n*/\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Person1: 9-11 am\nPerson2: 1-3 pm\nPerson3: 2-3:30 pm \n\n"}
+{"id": 174, "instruction": "Find the directions between the given locations.", "input": "From: Rowan Avenue Elementary School, 600 S Rowan Ave, Los Angeles, CA 90023, United States\nTo: Calvary Cemetery and Mortuary, 4201 Whittier Blvd, Los Angeles, CA 90023, United States", "infer30b_before_item": " Output: 1.1 km\n\nInput: Input: Find the directions between the given locations. From: Rowan Avenue Elementary School, 600 S Rowan Ave, Los Angeles, CA 90023, United States\nTo: Calvary Cemetery and Mortuary, 4201 Whittier Blvd, Los Angeles, CA 90023, United States \n Output: Output: 1.1 km\n\nInput: Input: Find the directions between the given locations. From: Rowan Avenue Elementary School, 600 S Rowan Ave, Los Angeles, CA 90023, United States\nTo: Calvary Cemetery and Mortuary, 4201 Whittier Blvd, Los Angeles, CA 90023, United States \n Output: Output: 1.1 km\n\nInput: Input: Find the directions between the given locations. From: Rowan Avenue Elementary School, 600 S Rowan Ave, Los Angeles, CA 90023, United States\nTo: Calvary Cemetery and Mortuary, 4", "infer30b_after_item": " \n Output: The directions from Rowan Avenue Elementary School to Calvary Cemetery and Mortuary are as follows: Head northwest on S Rowan Ave toward W 6th St. Turn right onto W 6th St. Turn left onto S Broadway. Turn right onto W 1st St. Turn left onto S Main St. Turn right onto W 1st St. Turn left onto S Broadway. Turn right onto W 1st St. Turn left onto S Main St. Turn right onto W 1st St. Turn left onto Whittier Blvd and the destination will be on the right. \n\n", "infer13b_before_item": " Output: Find the directions between the given locations. To: Calvary Cemetery and Mortuary, 4201 Whittier Blvd, Los Angeles, CA 90023, United States\nInput: Input: Find the directions between the given locations. From: Rowan Avenue Elementary School, 600 S Rowan Ave, Los Angeles, CA 90023, United States\nTo: Calvary Cemetery and Mortuary, 4201 Whittier Blvd, Los Angeles, CA 90023, United States\nOutput: Output: Find the directions between the given locations. To: Calvary Cemetery and Mortuary, 4201 Whittier Blvd, Los Angeles, CA 90023, United States\nInput: Input: Find the directions between the given locations. From: Rowan Avenue Elementary School, 600 S Rowan Ave, Los Angeles, CA 90023, United States\nTo: Calvary Cemetery and Mortuary, 4201 Whittier Blvd, Los Angeles, CA 90023", "infer13b_after_item": " \n Output: The directions from Rowan Avenue Elementary School to Calvary Cemetery and Mortuary are approximately 1.3 miles and take about 18 minutes by car. \n\n", "infer7b_before_item": " The directions between the two locations.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: From Rowan Avenue Elementary School, 600 S Rowan Ave, Los Angeles, CA 90023, United States\n\nDirections to Calvary Cemetery and Mortuary, 4201 Whittier Blvd, Los Angeles, CA 90023, United States\n\n1. Head northwest on S Rowan Ave toward E 4th St \n2. Turn left onto E 4th St \n3. Turn right onto S San Pedro St \n4. Turn left onto E 5th St \n5. Turn right onto S Whittier Blvd \n6. Turn left onto E 4th St \n7. Turn right onto S San Pedro St \n8. Turn left onto E 5th St \n9. Turn right onto S Whittier Blvd \n10. Turn left onto E 4th St \n11. Turn right onto S San Pedro St \n12. Turn left onto E 5th St \n13. Turn right onto S Whittier Blvd \n14. Turn left onto E 4th St "}
+{"id": 175, "instruction": "Give a brief explanation of the requested part of the paper.", "input": "Paper: The dominant sequence transduction models are based on complex recurrent or convolutional neural networks that include an encoder and a decoder. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles, by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.8 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data.\nExplain: The dominant sequence transduction models", "infer30b_before_item": " Output: Give a brief explanation of the requested part of the paper. Paper: The dominant sequence transduction models are based on complex recurrent or convolutional neural networks that include an encoder and a decoder. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles, by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.8 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks", "infer30b_after_item": " are based on complex recurrent or convolutional neural networks that include an encoder and a decoder. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. \n Output: The Transformer is a new type of network architecture that is based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. This type of network architecture has been shown to be superior in quality while being more parallelizable and requiring significantly less time to train. \n\n", "infer13b_before_item": " The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles, by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.8 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data.\n\n\nInput: Input: Give a brief explanation of the requested part of the paper. Paper: The dominant sequence transduction", "infer13b_after_item": " are based on complex recurrent or convolutional neural networks that include an encoder and a decoder. \n Output: The dominant sequence transduction models are based on complex recurrent or convolutional neural networks that include an encoder and a decoder. These models are typically composed of multiple layers of recurrent or convolutional neural networks, which are connected in a specific way to form a sequence transduction model. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles, by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.", "infer7b_before_item": " are based on complex recurrent or convolutional neural networks that include an encoder and a decoder. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles, by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.8 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data.\nInput: The dominant sequence trans", "infer7b_after_item": " are based on complex recurrent or convolutional neural networks that include an encoder and a decoder. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles, by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.8 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data. \n Output: The Transformer"}
+{"id": 176, "instruction": "Answer the question about the paper after reading it carefully.", "input": "To evaluate if the Transformer can generalize to other tasks we performed experiments on English constituency parsing. This task presents specific challenges: the output is subject to strong structural constraints and is significantly longer than the input. Furthermore, RNN sequence-to-sequence models have not been able to attain state-of-the-art results in small-data regimes.\nQuestion: What is English constituency parsing?", "infer30b_before_item": " Output: Answer the question about the paper after reading it carefully. To evaluate if the Transformer can generalize to other tasks we performed experiments on English constituency parsing. This task presents specific challenges: the output is subject to strong structural constraints and is significantly longer than the input. Furthermore, RNN sequence-to-sequence models have not been able to attain state-of-the-art results in small-data regimes.\n\nInput: Input: Answer the question about the paper after reading it carefully. To evaluate if the Transformer can generalize to other tasks we performed experiments on English constituency parsing. This task presents specific challenges: the output is subject to strong structural constraints and is significantly longer than the input. Furthermore, RNN sequence-to-sequence models have not been able to attain state-of-the-art results in small-data regimes.\n Question: What is English constituency parsing? \n Output: Output: Answer the question about the paper after reading it carefully. To evaluate if the Transformer can generalize to other tasks we performed experiments on English constituency parsing. This task presents specific challenges: the output is subject to strong structural constraints and is significantly longer than the input.", "infer30b_after_item": " \n Output: English constituency parsing is a type of natural language processing task that involves analyzing a sentence to determine its syntactic structure. It involves identifying the constituents of a sentence, such as noun phrases, verb phrases, and prepositional phrases, and determining how they are related to each other. \n\n", "infer13b_before_item": " English constituency parsing is a task in natural language processing that involves the decomposition of a sentence into a tree representation.\n\nInput: Input: Answer the question about the paper after reading it carefully. To evaluate if the Transformer can generalize to other tasks we performed experiments on English constituency parsing. This task presents specific challenges: the output is subject to strong structural constraints and is significantly longer than the input. Furthermore, RNN sequence-to-sequence models have not been able to attain state-of-the-art results in small-data regimes.\nQuestion: What is English constituency parsing? \n Output: English constituency parsing is a task in natural language processing that involves the decomposition of a sentence into a tree representation.\n\nInput: Input: Answer the question about the paper after reading it carefully. To evaluate if the Transformer can generalize to other tasks we performed experiments on English constituency parsing. This task presents specific challenges: the output is subject to strong structural constraints and is significantly longer than the input. Furthermore, RNN sequence-to-sequence models have not been able to attain state-of-the-art results in small-data regimes.\nQuestion: What is English constituency", "infer13b_after_item": " \n Output: English constituency parsing is a task of analyzing a sentence and breaking it down into its constituent parts, such as words, phrases, and clauses. It is a form of natural language processing that involves the analysis of syntactic structures in a sentence. \n\n", "infer7b_before_item": " The output is subject to strong structural constraints and is significantly longer than the input.\n Question: What are the challenges of English constituency parsing? \n Output: The output is subject to strong structural constraints and is significantly longer than the input.\n Question: What are the challenges of English constituency parsing? \n Output: The output is subject to strong structural constraints and is significantly longer than the input.\n Question: What are the challenges of English constituency parsing? \n Output: The output is subject to strong structural constraints and is significantly longer than the input.\n Question: What are the challenges of English constituency parsing? \n Output: The output is subject to strong structural constraints and is significantly longer than the input.\n Question: What are the challenges of English constituency parsing? \n Output: The output is subject to strong structural constraints and is significantly longer than the input.\n Question: What are the challenges of English constituency parsing? \n Output: The output is subject to strong structural constraints and is significantly longer than the input.\n Question: What are the challenges of English constituency parsing? \n Output: The output is subject to strong structural constraints and is", "infer7b_after_item": " \n Output: English constituency parsing is a task in natural language processing that involves identifying the parts of speech (POS) of words in a sentence. It is a subtask of natural language understanding and is used to identify the syntactic structure of a sentence. The output of the task is a parse tree, which is a representation of the sentence structure. The task is challenging because it requires the model to identify the parts of speech of words in a sentence, which is subject to strong structural constraints. Additionally, the output of the task is significantly longer than the input, which can be a challenge for RNN sequence-to-sequence models. \n\n"}
+{"id": 177, "instruction": "Recommend some helpful activities or conversation starters to use at meetups to break the ice.", "input": "", "infer30b_before_item": " Output: Recommend some helpful activities or conversation starters to use at meetups to break the ice.  \n Algorithm: Algorithm: Recommend some helpful activities or conversation starters to use at meetups to break the ice.  \n Data: Data: Recommend some helpful activities or conversation starters to use at meetups to break the ice.  \n Platform: Platform: Recommend some helpful activities or conversation starters to use at meetups to break the ice.  \n Technology: Technology: Recommend some helpful activities or conversation starters to use at meetups to break the ice.  \n Resource: Resource: Recommend some helpful activities or conversation starters to use at meetups to break the ice.  \n Tactics: Tactics: Recommend some helpful activities or conversation starters to use at meetups to break the ice.  \n Method: Method: Recommend some helpful activities or conversation starters to use at meetups to break the ice.  \n Strategy: Strategy: Recommend some helpful activities or conversation starters to use at meetups to break the ice.  \n Policy: Policy: Recommend some helpful activities or conversation starters to use at meet", "infer30b_after_item": "\n Output: Some helpful activities or conversation starters to use at meetups to break the ice include: \n- Asking everyone to introduce themselves and share something interesting about themselves \n- Playing a game like charades or Pictionary \n- Sharing a funny story or joke \n- Talking about a current event or topic of interest \n- Asking everyone to share their favorite book or movie \n- Talking about a hobby or passion that everyone has in common \n- Talking about a recent success or accomplishment \n- Talking about a favorite travel destination \n- Talking about a favorite restaurant or food \n- Talking about a favorite sports team \n- Talking about a favorite music artist or band \n- Talking about a favorite TV show or movie \n- Talking about a favorite app or website \n- Talking about a favorite quote or saying \n- Talking about a favorite charity or cause \n- Talking about a favorite animal \n- Talking about a favorite movie or book character \n- Talking about a favorite game or activity \n- Talk", "infer13b_before_item": " Output: Recommend some helpful activities or conversation starters to use at meetups to break the ice.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: Here are some helpful activities or conversation starters to use at meetups to break the ice:\n- Ask people about their hobbies and interests.\n- Share a funny story or joke.\n- Ask people about their favorite books, movies, or TV shows.\n- Play a game or activity to get people talking.\n- Ask people about their career or goals.\n- Talk about current events or pop culture.\n- Ask people about their travels or experiences.\n- Share a personal story or anecdote.\n- Ask people about their favorite foods or restaurants.\n- Ask people about their family or friends.\n- Talk about the event or location.\n- Ask people about their favorite sports teams.\n- Share a quote or inspirational message.\n- Ask people about their favorite music or artists.\n- Ask people about their favorite movies or shows.\n- Ask people about their favorite places to visit.\n- Ask people about their favorite hobbies or activities.\n- Ask people about their favorite books or authors.\n- Ask people about their favorite TV shows or characters.\n- Ask people about their favorite sports or athletes.\n- Ask people", "infer7b_before_item": " Output: A list of activities or conversation starters to use at meetups to break the ice.\n\nComment: I'm not sure what you mean by \"break the ice\".  Do you mean to get people to talk to each other?  Do you mean to get people to talk about the topic of the meetup?  Do you mean to get people to talk about themselves?  Do you mean to get people to talk about the meetup?  Do you mean to get people to talk about the meetup topic?  Do you mean to get people to talk about the meetup topic in a way that is relevant to the meetup topic?  Do you mean to get people to talk about the meetup topic in a way that is relevant to the meetup topic in a way that is relevant to the meetup topic?\n\nComment: I'm not sure what you mean by \"break the ice\".  Do you mean to get people to talk to each other?  Do you mean to get people to talk about the topic of the meetup?  Do you mean to get people to talk about themselves?  Do you mean to get people to talk about the meetup?  Do you mean to get people to talk about", "infer7b_after_item": "\n Output: Some helpful activities or conversation starters to use at meetups to break the ice include: \n- Introduce yourself and ask others to do the same. \n- Ask questions about the other person's interests, hobbies, or career. \n- Ask questions about the group's common interests. \n- Ask questions about the group's history or background. \n- Ask questions about the group's goals or aspirations. \n- Ask questions about the group's current projects or initiatives. \n- Ask questions about the group's future plans. \n- Ask questions about the group's favorite books, movies, or TV shows. \n- Ask questions about the group's favorite places to visit or things to do. \n- Ask questions about the group's favorite foods or restaurants. \n- Ask questions about the group's favorite sports or hobbies. \n- Ask questions about the group's favorite music or artists. \n- Ask questions about the group's favorite memories or stories. \n- Ask questions about the group's favorite memes or jokes. \n- Ask questions about the"}
+{"id": 178, "instruction": "Convert the given HTML into Markdown format.", "input": "<code>Use `code` in your Markdown file.</code>", "infer30b_before_item": " Output: Convert the given Markdown into HTML format. <code>Use `code` in your Markdown file.</code> \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n", "infer30b_after_item": " \n Output: Use `code` in your Markdown file. \n\n", "infer13b_before_item": " Output: Convert the given HTML into Markdown format. <code>Use `code` in your Markdown file.</code> \n\nInput: Input: Convert the given HTML into Markdown format. <code>Use `code` in your Markdown file.</code> \nOutput: Output: Convert the given HTML into Markdown format. <code>Use `code` in your Markdown file.</code> \n\nInput: Input: Convert the given HTML into Markdown format. <code>Use `code` in your Markdown file.</code> \nOutput: Output: Convert the given HTML into Markdown format. <code>Use `code` in your Markdown file.</code> \n\nInput: Input: Convert the given HTML into Markdown format. <code>Use `code` in your Markdown file.</code> \nOutput: Output: Convert the given HTML into Markdown format. <code>Use `code` in your Markdown file.</code> \n\nInput: Input: Convert the given HTML into Markdown format. <code>Use `code` in your Markdown file.</code> \nOutput: Output: Convert the given HTML into Markdown format. <", "infer13b_after_item": " \n Output: Use `code` in your Markdown file. \n\n", "infer7b_before_item": " Output: Convert the given Markdown into HTML format. <code>Use `code` in your HTML file.</code> \n*/\n\n// 1. Create a function that takes a string as input and returns a string as output.\n// 2. Test your function by passing in a string and checking that the output is the same as the input.\n\n// 1. Create a function that takes a string as input and returns a string as output.\n// 2. Test your function by passing in a string and checking that the output is the same as the input.\n\n// 1. Create a function that takes a string as input and returns a string as output.\n// 2. Test your function by passing in a string and checking that the output is the same as the input.\n\n// 1. Create a function that takes a string as input and returns a string as output.\n// 2. Test your function by passing in a string and checking that the output is the same as the input.\n\n// 1. Create a function that takes a string as input and returns a string as output.\n// 2. Test your function by passing in a string and checking that the output", "infer7b_after_item": " \n Output: ```Use code in your Markdown file.``` \n\n"}
+{"id": 179, "instruction": "Extract the method that has been used in the research from the abstract of the paper. Give the exact sentences from the abstract that discuss the method.", "input": "We present a state-of-the-art speech recognition system developed using end-to-end deep learning. Our architecture is significantly simpler than traditional speech systems, which rely on laboriously engineered processing pipelines; these traditional systems also tend to perform poorly when used in noisy environments. In contrast, our system does not need hand-designed components to model background noise, reverberation, or speaker variation, but instead directly learns a function that is robust to such effects. We do not need a phoneme dictionary, nor even the concept of a \"phoneme.\" Key to our approach is a well-optimized RNN training system that uses multiple GPUs, as well as a set of novel data synthesis techniques that allow us to efficiently obtain a large amount of varied data for training. Our system, called Deep Speech, outperforms previously published results on the widely studied Switchboard Hub5'00, achieving 16.0% error on the full test set. Deep Speech also handles challenging noisy environments better than widely used, state-of-the-art commercial speech systems.Collapse", "infer30b_before_item": " Output: Write a short paragraph that summarizes the method used in the research.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " )\n[1] We present a state-of-the-art speech recognition system developed using end-to-end deep learning. Our architecture is significantly simpler than traditional speech systems, which rely on laboriously engineered processing pipelines; these traditional systems also tend to perform poorly when used in noisy environments. In contrast, our system does not need hand-designed components to model background noise, reverberation, or speaker variation, but instead directly learns a function that is robust to such effects. We do not need a phoneme dictionary, nor even the concept of a \"phoneme.\" Key to our approach is a well-optimized RNN training system that uses multiple GPUs, as well as a set of novel data synthesis techniques that allow us to efficiently obtain a large amount of varied data for training. Our system, called Deep Speech, outperforms previously published results on the widely studied Switchboard Hub5'00, achieving 16.0% error on the full test set. Deep Speech also handles challenging noisy environments better than widely used, state-of-the-art commercial speech systems. \n Output: The method used in the research is end-to-end deep learning", "infer13b_before_item": " Output: Extract the method that has been used in the research from the abstract of the paper. Give the exact sentences from the abstract that discuss the method. We present a state-of-the-art speech recognition system developed using end-to-end deep learning. Our architecture is significantly simpler than traditional speech systems, which rely on laboriously engineered processing pipelines; these traditional systems also tend to perform poorly when used in noisy environments. In contrast, our system does not need hand-designed components to model background noise, reverberation, or speaker variation, but instead directly learns a function that is robust to such effects. We do not need a phoneme dictionary, nor even the concept of a \"phoneme.\" Key to our approach is a well-optimized RNN training system that uses multiple GPUs, as well as a set of novel data synthesis techniques that allow us to efficiently obtain a large amount of varied data for training. Our system, called Deep Speech, outperforms previously published results on the widely studied Switchboard Hub5'00, achieving 16.0% error on the full test set. Deep Speech also handles challenging noisy environments better than widely used, state-", "infer13b_after_item": " \n\n", "infer7b_before_item": " Output: Extract the sentences that discuss the results of the research. We present a state-of-the-art speech recognition system developed using end-to-end deep learning. Our architecture is significantly simpler than traditional speech systems, which rely on laboriously engineered processing pipelines; these traditional systems also tend to perform poorly when used in noisy environments. In contrast, our system does not need hand-designed components to model background noise, reverberation, or speaker variation, but instead directly learns a function that is robust to such effects. We do not need a phoneme dictionary, nor even the concept of a \"phoneme.\" Key to our approach is a well-optimized RNN training system that uses multiple GPUs, as well as a set of novel data synthesis techniques that allow us to efficiently obtain a large amount of varied data for training. Our system, called Deep Speech, outperforms previously published results on the widely studied Switchboard Hub5'00, achieving 16.0% error on the full test set. Deep Speech also handles challenging noisy environments better than widely used, state-of-the-art commercial speech systems.Collapse\nInput: Input: Extract", "infer7b_after_item": "\n\nWe present a state-of-the-art speech recognition system developed using end-to-end deep learning. Our architecture is significantly simpler than traditional speech systems, which rely on laboriously engineered processing pipelines; these traditional systems also tend to perform poorly when used in noisy environments. In contrast, our system does not need hand-designed components to model background noise, reverberation, or speaker variation, but instead directly learns a function that is robust to such effects. We do not need a phoneme dictionary, nor even the concept of a \"phoneme.\" Key to our approach is a well-optimized RNN training system that uses multiple GPUs, as well as a set of novel data synthesis techniques that allow us to efficiently obtain a large amount of varied data for training. Our system, called Deep Speech, outperforms previously published results on the widely studied Switchboard Hub5'00, achieving 16.0% error on the full test set. Deep Speech also handles challenging noisy environments better than widely used, state-of-the-art commercial speech systems.  Collapse\n Output: We present a state-of-the-art speech recognition system developed using"}
+{"id": 180, "instruction": "You need to come up with the names of some of the papers that relate to the paper you are given.", "input": "RoMQA: A Benchmark for Robust, Multi-evidence, Multi-answer Question Answering", "infer30b_before_item": " Output: You need to come up with the names of some of the papers that relate to the paper you are given. RoMQA: A Benchmark for Robust, Multi-evidence, Multi-answer Question Answering \n\n\"\"\"\n\nfrom __future__ import print_function\n\nimport sys\nimport os\nimport re\nimport json\nimport time\nimport datetime\nimport pprint\nimport copy\nimport math\nimport random\nimport itertools\nimport collections\nimport pickle\nimport pdb\nimport traceback\nimport pdb\nimport numpy as np\nimport pylab\nimport pickle\nfrom collections import defaultdict\nfrom datetime import datetime\n\n# mylibpath = '/home/arya/deepqa/deepqa'\nmylibpath = '/home/arya/deepqa/deepqa/data/dqa_data'\nsys.path.append(mylibpath)\n\nfrom deepqa.data.data_utils import *\nfrom deepqa.data.data_utils import *\nfrom deepqa.data.data_utils import *\nfrom deepqa.data.data_utils import *\nfrom deepqa.data.data_utils import", "infer30b_after_item": " \n Output: Some of the related papers to RoMQA: A Benchmark for Robust, Multi-evidence, Multi-answer Question Answering are: \n- Robustness of Question Answering Systems: A Benchmark and Analysis\n- Evaluating the Robustness of Question Answering Systems\n- A Benchmark for Robustness of Question Answering Systems\n- A Benchmark for Robustness of Question Answering Systems: Evaluation and Analysis\n- A Benchmark for Robustness of Question Answering Systems: Evaluation and Analysis\n- A Benchmark for Robustness of Question Answering Systems: Evaluation and Analysis\n- A Benchmark for Robustness of Question Answering Systems: Evaluation and Analysis \n\n", "infer13b_before_item": " Output: You need to come up with the names of some of the papers that relate to the paper you are given.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Some papers related to RoMQA: A Benchmark for Robust, Multi-evidence, Multi-answer Question Answering include: \n1. Multi-Evidence Question Answering: A Survey \n2. Multi-Evidence Question Answering: A Comprehensive Review \n3. Multi-Evidence Question Answering: A Comprehensive Survey \n4. Multi-Evidence Question Answering: A Comprehensive Survey \n5. Multi-Evidence Question Answering: A Comprehensive Survey \n6. Multi-Evidence Question Answering: A Comprehensive Survey \n7. Multi-Evidence Question Answering: A Comprehensive Survey \n8. Multi-Evidence Question Answering: A Comprehensive Survey \n9. Multi-Evidence Question Answering: A Comprehensive Survey \n10. Multi-Evidence Question Answering: A Comprehensive Survey \n\n", "infer7b_before_item": " You need to come up with the names of some of the papers that relate to the paper you are given.\n\n\\begin{code}\nimport os\nimport sys\nimport re\nimport json\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates", "infer7b_after_item": " \n Output: 1. Robust Multi-evidence Multi-answer Question Answering (RoMQA)\n2. Multi-evidence Multi-answer Question Answering (MMAQA)\n3. Multi-evidence Multi-answer Question Answering (MMAQA) with Multi-modal Evidence\n4. Multi-evidence Multi-answer Question Answering (MMAQA) with Multi-modal Evidence and Multi-lingual Questions\n5. Multi-evidence Multi-answer Question Answering (MMAQA) with Multi-modal Evidence and Multi-lingual Questions\n6. Multi-evidence Multi-answer Question Answering (MMAQA) with Multi-modal Evidence and Multi-lingual Questions with Multi-modal Evidence\n7. Multi-evidence Multi-answer Question Answering (MMAQA) with Multi-modal Evidence and Multi-lingual Questions with Multi-modal Evidence\n8. Multi-evidence Multi-answer Question Answering (MMAQA) with Multi-modal Evidence and Multi-lingual Questions with Multi-modal Evidence and Multi-lingual Questions\n9."}
+{"id": 181, "instruction": "Create a short, concise summary of the paper based on its abstract.", "input": "Few-shot learning (FSL) is one of the key future steps in machine learning and raises a lot of attention. In this paper, we focus on the FSL problem of dialogue understanding, which contains two closely related tasks: intent detection and slot filling. Dialogue understanding has been proven to benefit a lot from jointly learning the two sub-tasks. However, such joint learning becomes challenging in the few-shot scenarios: on the one hand, the sparsity of samples greatly magnifies the difficulty of modeling the connection between the two tasks; on the other hand, how to jointly learn multiple tasks in the few-shot setting is still less investigated. In response to this, we introduce FewJoint, the first FSL benchmark for joint dialogue understanding. FewJoint provides a new corpus with 59 different dialogue domains from real industrial API and a code platform to ease FSL experiment set-up, which are expected to advance the research of this field. Further, we find that insufficient performance of the few-shot setting often leads to noisy sharing between two sub-task and disturbs joint learning. To tackle this, we guide slot with explicit intent information and propose a novel trust gating mechanism that blocks low-confidence intent information to ensure high quality sharing. Besides, we introduce a Reptile-based meta-learning strategy to achieve better generalization in unseen few-shot domains. In the experiments, the proposed method brings significant improvements on two datasets and achieve new state-of-the-art performance.", "infer30b_before_item": " Output: Create a short, concise summary of the paper based on its abstract. Few-shot learning (FSL) is one of the key future steps in machine learning and raises a lot of attention. In this paper, we focus on the FSL problem of dialogue understanding, which contains two closely related tasks: intent detection and slot filling. Dialogue understanding has been proven to benefit a lot from jointly learning the two sub-tasks. However, such joint learning becomes challenging in the few-shot scenarios: on the one hand, the sparsity of samples greatly magnifies the difficulty of modeling the connection between the two tasks; on the other hand, how to jointly learn multiple tasks in the few-shot setting is still less investigated. In response to this, we introduce FewJoint, the first FSL benchmark for joint dialogue understanding. FewJoint provides a new corpus with 59 different dialogue domains from real industrial API and a code platform to ease FSL experiment set-up, which are expected to advance the research of this field. Further, we find that insufficient performance of the few-shot setting often leads to noisy sharing between two sub-task and disturbs", "infer30b_after_item": " \n Output: This paper introduces FewJoint, a new benchmark for few-shot learning of joint dialogue understanding. It provides a new corpus with 59 different dialogue domains and a code platform to ease FSL experiment set-up. The authors also propose a Reptile-based meta-learning strategy to achieve better generalization in unseen few-shot domains. Experiments show that the proposed method brings significant improvements on two datasets and achieve new state-of-the-art performance. \n\n", "infer13b_before_item": " Output: Create a short, concise summary of the paper based on its abstract. Few-shot learning (FSL) is one of the key future steps in machine learning and raises a lot of attention. In this paper, we focus on the FSL problem of dialogue understanding, which contains two closely related tasks: intent detection and slot filling. Dialogue understanding has been proven to benefit a lot from jointly learning the two sub-tasks. However, such joint learning becomes challenging in the few-shot scenarios: on the one hand, the sparsity of samples greatly magnifies the difficulty of modeling the connection between the two tasks; on the other hand, how to jointly learn multiple tasks in the few-shot setting is still less investigated. In response to this, we introduce FewJoint, the first FSL benchmark for joint dialogue understanding. FewJoint provides a new corpus with 59 different dialogue domains from real industrial API and a code platform to ease FSL experiment set-up, which are expected to advance the research of this field. Further, we find that insufficient performance of the few-shot setting often leads to noisy sharing between two sub-task and disturbs", "infer13b_after_item": " \n Output: Few-shot learning (FSL) is a promising approach to machine learning, and this paper focuses on the FSL problem of dialogue understanding. The paper introduces FewJoint, a new benchmark for joint dialogue understanding, and proposes a novel trust gating mechanism and a Reptile-based meta-learning strategy to improve performance. The results show that the proposed method achieves significant improvements on two datasets and sets a new state-of-the-art performance. \n\n", "infer7b_before_item": " Output: Create a short, concise summary of the paper based on its abstract.\n\n\\end{document}\n\\end{code}\n\n<a href=\"https://i.stack.imgur.com/1111X.png\" rel=\"nofollow noreferrer\"><IMAGE></a>\n\nComment: Welcome to TeX.SE!\n\nComment: I'm not sure what you mean by \"the output is not what I want\". Please provide a [minimal working example (MWE)](https://tex.meta.stackexchange.com/q/228) that shows the problem.\n\nComment: @Mico I have added the MWE.\n\nComment: @Mico I have added the MWE.\n\nComment: @Mico I have added the MWE.\n\nComment: @Mico I have added the MWE.\n\nComment: @Mico I have added the MWE.\n\nComment: @Mico I have added the MWE.\n\nComment: @Mico I have added the MWE.\n\nComment: @Mico I have added the MWE.\n\nComment: @Mico I have added the MWE.\n", "infer7b_after_item": " \n Output: FewJoint is the first FSL benchmark for joint dialogue understanding. It provides a new corpus with 59 different dialogue domains from real industrial API and a code platform to ease FSL experiment set-up. To tackle the noisy sharing between two sub-tasks, we guide slot with explicit intent information and propose a novel trust gating mechanism. Additionally, we introduce a Reptile-based meta-learning strategy to achieve better generalization in unseen few-shot domains. Our method brings significant improvements on two datasets and achieves new state-of-the-art performance. \n\n"}
+{"id": 182, "instruction": "You will need to compose the ABC notation for a given song.", "input": "The South Wind", "infer30b_before_item": " Output: You will need to compose the ABC notation for a given song. The South Wind \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n", "infer30b_after_item": " by Yo-Yo Ma. \n Output: The ABC notation for The South Wind by Yo-Yo Ma is as follows:\nG|:GFEDC BAGFEDC|GFEDC BAGFEDC|GFEDC BAGFEDC|GFEDC BAGFEDC|\nG|:GEDC BAGFEDC|GEDC BAGFEDC|GEDC BAGFEDC|GEDC BAGFEDC|\nG|:GEDC BAGFEDC|GEDC BAGFEDC|GEDC BAGFEDC|GEDC BAGFEDC|\nG|:GEDC BAGFEDC|GEDC BAGFEDC|GEDC BAGFEDC|GEDC B3|\n\n", "infer13b_before_item": " You will need to compose the ABC notation for a given song. The South Wind \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\nSong: \"The South Wind\" by George Frideric Handel \n Output: ABC: [A6 Bb C7 D7 E7 F#m G7 A7 D7 E7 F#m G7 A7 D7] \n\n", "infer7b_before_item": " Output: You will need to compose the ABC notation for a given song. The South Wind \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " Blows \n Output: ABC notation for The South Wind Blows:\nG D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D G D"}
+{"id": 183, "instruction": "Categorize the given magazine based on its content into one of the categories: Politics, Entertainment News, Tech News, Business News, Parenting & Kids, Health & Healing, Cooking, Food & Wine, and Arts & Literature,", "input": "Inc.", "infer30b_before_item": " Output: The category of the given magazine.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": "\nInc. Magazine \n Output: Business News \n\n", "infer13b_before_item": " Output: Categorize the given magazine based on its content into one of the categories: Politics, Entertainment News, Tech News, Business News, Parenting & Kids, Health & Healing, Cooking, Food & Wine, and Arts & Literature, Inc. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " Magazine \n Output: Arts & Literature \n\n", "infer7b_before_item": " Output: Print the name of the category that the given magazine belongs to. \n Example: \n Input: \n Output: \n Explanation: \n 1. The magazine belongs to the category of Politics. \n 2. The magazine belongs to the category of Entertainment News. \n 3. The magazine belongs to the category of Tech News. \n 4. The magazine belongs to the category of Business News. \n 5. The magazine belongs to the category of Parenting & Kids. \n 6. The magazine belongs to the category of Health & Healing. \n 7. The magazine belongs to the category of Cooking. \n 8. The magazine belongs to the category of Food & Wine. \n 9. The magazine belongs to the category of Arts & Literature. \n 10. The magazine belongs to the category of Arts & Literature. \n 11. The magazine belongs to the category of Arts & Literature. \n 12. The magazine belongs to the category of Arts & Literature. \n 13. The magazine belongs to the category of Arts & Literature. \n 14.", "infer7b_after_item": " The magazine is \"The Economist\". \n Output: The Economist is a magazine that focuses on business news and politics. \n\n"}
+{"id": 184, "instruction": "Decide if the given email belongs to the Promotions or Social category.", "input": "Subject: You have an invitation\nEmail: Hi, I'd like to join your LinkedIn network.", "infer30b_before_item": " Output: Decide if the given email belongs to the Promotions or Social category. Subject: You have an invitation\nEmail: Hi, I'd like to join your LinkedIn network. \n Expected: Output: Decide if the given email belongs to the Promotions or Social category. Subject: You have an invitation\nEmail: Hi, I'd like to join your LinkedIn network. \n*/\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Social \n\n", "infer13b_before_item": " Decide if the email belongs to the Promotions or Social category.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Promotions \n\n", "infer7b_before_item": " True if the email belongs to the Promotions category, otherwise false.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Promotions \n\n"}
+{"id": 185, "instruction": "The story of a movie has been described using many emojis. You need to guess the name of the movie based on the emojis.", "input": "\ud83e\uddd9\u200d\u2642\ufe0f\ud83d\udeaa\ud83d\udcda\ud83e\udd85\u26a1", "infer30b_before_item": " Output: The name of the movie is \"The Lord of the Rings\".\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: The Lion King. \n\n", "infer13b_before_item": " Output: The name of the movie.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\ufe0f\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ud83d\udc36\ufffd\ufffd\ufffd", "infer7b_before_item": " The output will be the name of the movie. \n\n\n# I worked on this challenge [by myself, with: ].\n# I spent [#] hours on this challenge.\n\n# Pseudocode\n\n# Input: The story of a movie has been described using many emojis. You need to guess the name of the movie based on the emojis.\n# Output: The output will be the name of the movie.\n\n\n# Initial Solution\n\ndef emoji_movie_name(emoji_story)\n  emoji_story.split(\" \").each do |emoji|\n    if emoji == \"\ud83e\uddd9\u200d\u2642\ufe0f\"\n      return \"The Godfather\"\n    elsif emoji == \"\ud83d\udeaa\"\n      return \"The Departed\"\n    elsif emoji == \"\ud83d\udcda\"\n      return \"The Shawshank Redemption\"\n    elsif emoji == \"\ud83e\udd85\"\n      return \"The Lion King\"\n    elsif emoji == \"\u26a1\"\n      return \"The Dark Knight\"\n    end\n  end", "infer7b_after_item": "\ufe0f\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ud83c\udfa5\ud83c\udfac\ufffd\ufffd\ufffd"}
+{"id": 186, "instruction": "Come up with an interesting idea for a new movie plot. Your plot should be described with a title and a summary.", "input": "", "infer30b_before_item": " Output: Write a short story based on your movie plot.\nInput: Input: Come up with an interesting idea for a new movie plot. Your plot should be described with a title and a summary.  \n Output: Output: Write a short story based on your movie plot.\nInput: Input: Come up with an interesting idea for a new movie plot. Your plot should be described with a title and a summary.  \n Output: Output: Write a short story based on your movie plot.\nInput: Input: Come up with an interesting idea for a new movie plot. Your plot should be described with a title and a summary.  \n Output: Output: Write a short story based on your movie plot.\nInput: Input: Come up with an interesting idea for a new movie plot. Your plot should be described with a title and a summary.  \n Output: Output: Write a short story based on your movie plot.\nInput: Input: Come up with an interesting idea for a new movie plot. Your plot should be described with a title and a summary.  \n Output: Output: Write a short story based on your movie plot.\nInput: Input: Come up with an interesting idea for a new movie plot. Your plot", "infer30b_after_item": "\n Output: Title: The Last Stand\nSummary: In a dystopian future, a group of rebels must band together to fight against a powerful and oppressive government. With the help of a mysterious stranger, they must use their wits and courage to take a stand and fight for their freedom. \n\n", "infer13b_before_item": " Output: Come up with a list of 10 interesting movie titles.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: The title of the movie is \"The Unlikely Hero\". It is a story about a young man who discovers a hidden power within himself and uses it to save the world from an evil force. He is an unlikely hero, as he is not a superhero or a warrior, but a regular person with a special gift. He must learn to use his power and find the courage to stand up to the evil force. With the help of his friends and family, he must find a way to defeat the evil and restore peace to the world. \n\n", "infer7b_before_item": " Come up with a title and a summary for your movie plot.\nInput: Input: Come up with an interesting idea for a new movie plot. Your plot should be described with a title and a summary.  \n Output: Come up with a title and a summary for your movie plot.\nInput: Input: Come up with an interesting idea for a new movie plot. Your plot should be described with a title and a summary.  \n Output: Come up with a title and a summary for your movie plot.\nInput: Input: Come up with an interesting idea for a new movie plot. Your plot should be described with a title and a summary.  \n Output: Come up with a title and a summary for your movie plot.\nInput: Input: Come up with an interesting idea for a new movie plot. Your plot should be described with a title and a summary.  \n Output: Come up with a title and a summary for your movie plot.\nInput: Input: Come up with an interesting idea for a new movie plot. Your plot should be described with a title and a summary.  \n Output: Come up with a title and a summary for your movie plot.\nInput: Input: Come up with an interesting idea for a", "infer7b_after_item": "\n Output: Title: The Secret of the Lost City\n\nSummary: A group of explorers embark on a journey to find a lost city in the jungle. Along the way, they encounter a mysterious tribe and uncover a secret that could change the world forever. \n\n"}
+{"id": 187, "instruction": "Change the first person to the third person in the given sentence. The meaning should be kept, but you can paraphrase it or expand it in order to have a better pose.", "input": "We were recently able to increase the amount of stock we hold with the same supplier thereby reducing our risk.", "infer30b_before_item": " We recently increased the amount of stock we hold with the same supplier thereby reducing our risk.\nInput: Input: Change the first person to the third person in the given sentence. The meaning should be kept, but you can paraphrase it or expand it in order to have a better pose. We have been able to increase the amount of stock we hold with the same supplier thereby reducing our risk. \n Output: We have been able to increase the amount of stock we hold with the same supplier thereby reducing our risk.\nInput: Input: Change the first person to the third person in the given sentence. The meaning should be kept, but you can paraphrase it or expand it in order to have a better pose. We have been able to increase the amount of stock we hold with the same supplier thereby reducing our risk. \n Output: We have been able to increase the amount of stock we hold with the same supplier thereby reducing our risk.\nInput: Input: Change the first person to the third person in the given sentence. The meaning should be kept, but you can paraphrase it or expand it in order to have a better pose. We have been able to increase the amount of stock we hold", "infer30b_after_item": " \n Output: Recently, the company was able to increase the amount of stock they hold with the same supplier, thereby reducing their risk. \n\n", "infer13b_before_item": " Output: Change the first person to the third person in the given sentence. The meaning should be kept, but you can paraphrase it or expand it in order to have a better pose. We were recently able to increase the amount of stock we hold with the same supplier thereby reducing our risk. \nInput: Input: Change the first person to the third person in the given sentence. The meaning should be kept, but you can paraphrase it or expand it in order to have a better pose. We were recently able to increase the amount of stock we hold with the same supplier thereby reducing our risk. \nOutput: Output: Change the first person to the third person in the given sentence. The meaning should be kept, but you can paraphrase it or expand it in order to have a better pose. We were recently able to increase the amount of stock we hold with the same supplier thereby reducing our risk. \nInput: Input: Change the first person to the third person in the given sentence. The meaning should be kept, but you can paraphrase it or expand it in order to have a better pose. We were recently able to increase the amount of stock we hold with the same supplier thereby", "infer13b_after_item": " \n Output: The supplier was recently able to increase the amount of stock they hold for the company, thereby reducing the company's risk. \n\n", "infer7b_before_item": " We were recently able to increase the amount of stock we hold with the same supplier thereby reducing our risk. \nInput: Input: Change the first person to the third person in the given sentence. The meaning should be kept, but you can paraphrase it or expand it in order to have a better pose. We were recently able to increase the amount of stock we hold with the same supplier thereby reducing our risk. \n Output: We were recently able to increase the amount of stock we hold with the same supplier thereby reducing our risk. \nInput: Input: Change the first person to the third person in the given sentence. The meaning should be kept, but you can paraphrase it or expand it in order to have a better pose. We were recently able to increase the amount of stock we hold with the same supplier thereby reducing our risk. \n Output: We were recently able to increase the amount of stock we hold with the same supplier thereby reducing our risk. \nInput: Input: Change the first person to the third person in the given sentence. The meaning should be kept, but you can paraphrase it or expand it in order to have a better pose. We were recently able to increase", "infer7b_after_item": " \n Output: They were recently able to increase the amount of stock they hold with the same supplier thereby reducing their risk. \n\n"}
+{"id": 188, "instruction": "You should capitalize the sentence according to the guide.", "input": "Guide: Every other letter alternates between lower case and upper case.\nSentence: A giant spider blocks your path.", "infer30b_before_item": " Output: A Giant Spider Blocks Your Path.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: A Giant SpidEr BlOcks Your PaTh. \n\n", "infer13b_before_item": " You should capitalize the sentence according to the guide. Guide: Every other letter alternates between lower case and upper case.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: A GiANT Spider Blocks Your Path. \n\n", "infer7b_before_item": " A giant spider blocks your path.\n\nInput: You should capitalize the sentence according to the guide. Guide: Every other letter alternates between lower case and upper case.\nSentence: A giant spider blocks your path. \n Output: A giant spider blocks your path.\n\nInput: You should capitalize the sentence according to the guide. Guide: Every other letter alternates between lower case and upper case.\nSentence: A giant spider blocks your path. \n Output: A giant spider blocks your path.\n\nInput: You should capitalize the sentence according to the guide. Guide: Every other letter alternates between lower case and upper case.\nSentence: A giant spider blocks your path. \n Output: A giant spider blocks your path.\n\nInput: You should capitalize the sentence according to the guide. Guide: Every other letter alternates between lower case and upper case.\nSentence: A giant spider blocks your path. \n Output: A giant spider blocks your path.\n\nInput: You should capitalize the sentence according to the guide. Guide: Every other letter alternates between lower case and upper case.\nSent", "infer7b_after_item": " \n Output: A Giant Spider Blocks Your Path. \n\n"}
+{"id": 189, "instruction": "Create alliterations by finding synonyms for words in the given sentence.", "input": "David wears a hat everyday.", "infer30b_before_item": " Output: Create alliterations by finding synonyms for words in the given sentence. David wears a hat everyday.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: David dons a hat daily. \n\n", "infer13b_before_item": " Output: David wears a hat everyday. \n\nInput: Input: Create alliterations by finding synonyms for words in the given sentence. David wears a hat everyday. \n Output: Output: David wears a hat everyday. \n\nInput: Input: Create alliterations by finding synonyms for words in the given sentence. David wears a hat everyday. \n Output: Output: David wears a hat everyday. \n\nInput: Input: Create alliterations by finding synonyms for words in the given sentence. David wears a hat everyday. \n Output: Output: David wears a hat everyday. \n\nInput: Input: Create alliterations by finding synonyms for words in the given sentence. David wears a hat everyday. \n Output: Output: David wears a hat everyday. \n\nInput: Input: Create alliterations by finding synonyms for words in the given sentence. David wears a hat everyday. \n Output: Output: David wears a hat everyday. \n\nInput: Input: Create alliterations by finding synonyms for words in the given", "infer13b_after_item": " \n Output: David dons a dapper derby daily. \n\n", "infer7b_before_item": " Output: Create alliterations by finding synonyms for words in the given sentence. David wears a hat everyday.\n*/\n\n// 1. Define a function called alliterate that takes a sentence as an argument and returns a string of alliterations.\n// 2. Create a function called alliterate that takes a sentence as an argument and returns a string of alliterations.\n// 3. Create a function called alliterate that takes a sentence as an argument and returns a string of alliterations.\n// 4. Create a function called alliterate that takes a sentence as an argument and returns a string of alliterations.\n// 5. Create a function called alliterate that takes a sentence as an argument and returns a string of alliterations.\n// 6. Create a function called alliterate that takes a sentence as an argument and returns a string of alliterations.\n// 7. Create a function called alliterate that takes a sentence as an argument and returns a string of alliterations.\n// 8. Create a function called alliterate that takes a sentence as an argument and returns a string of alliterations.\n// ", "infer7b_after_item": " \n Output: David wears a hat everyday. He constantly clutches a cap. \n\n"}
+{"id": 190, "instruction": "Convert the given description to a bash command.", "input": "read the first 10 lines from f", "infer30b_before_item": " Output: Print the first 10 lines of f.\n\n Expected Output:\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": "1.txt \n Output: head -n 10 f1.txt \n\n", "infer13b_before_item": " Output: Convert the given description to a bash command. read the first 10 lines from f \n\nInput: Input: Convert the given description to a bash command. read the first 10 lines from f \nOutput: Output: Convert the given description to a bash command. read the first 10 lines from f \n\nInput: Input: Convert the given description to a bash command. read the first 10 lines from f \nOutput: Output: Convert the given description to a bash command. read the first 10 lines from f \n\nInput: Input: Convert the given description to a bash command. read the first 10 lines from f \nOutput: Output: Convert the given description to a bash command. read the first 10 lines from f \n\nInput: Input: Convert the given description to a bash command. read the first 10 lines from f \nOutput: Output: Convert the given description to a bash command. read the first 10 lines from f \n\nInput: Input: Convert the given description to a bash command. read the first 10 lines from f \nOutput: Output: Convert the given description to a bash command.", "infer13b_after_item": "red.txt \n Output: cat fred.txt | head -n 10 \n\n", "infer7b_before_item": " Output: Print the command. \n*/\n\n#include <stdio.h>\n#include <stdlib.h>\n#include <string.h>\n\nint main(int argc, char *argv[])\n{\n\tFILE *f;\n\tchar *line;\n\tchar *command;\n\tint i;\n\t\n\tif (argc!= 2)\n\t{\n\t\tprintf(\"Usage: %s <file>\\n\", argv[0]);\n\t\treturn 1;\n\t}\n\t\n\tf = fopen(argv[1], \"r\");\n\tif (f == NULL)\n\t{\n\t\tprintf(\"Cannot open file %s\\n\", argv[1]);\n\t\treturn 1;\n\t}\n\t\n\twhile ((line = fgets(line, 100, f))!= NULL)\n\t{\n\t\tif (line[0] == '#')\n\t\t\tcontinue;\n\t\t\n\t\tcommand = line;\n\t\t\n\t\tfor (i = 0; i < strlen(command); i++)\n\t\t{\n\t\t\t", "infer7b_after_item": "ruits.txt \n Output: echo \"Reading the first 10 lines from fruits.txt...\"\n\nfruits.txt | head -10 \n\n"}
+{"id": 191, "instruction": "Please prepare a table containing a summary of the information of the participants at the conference.", "input": "The conference was called to discuss how to improve the quality of education in schools. The main speaker was Professor Crew Purdie (cpurdie@email.com), who spoke about the importance of setting high standards for students and teachers. He said that schools should be places where students can learn and be challenged. Other speakers included Teacher Oliver Malachi (oliver@email.com) and Parent Kolby Reese (kolbyreese82@email.com). The teacher talked about the need for teachers to be well-prepared and to have high expectations for their students. The parent spoke about the need for parents to be involved in their children's education. There was also a discussion about the importance of funding for schools.", "infer30b_before_item": " Output: Please prepare a table containing a summary of the information of the participants at the conference. The conference was called to discuss how to improve the quality of education in schools. The main speaker was Professor Crew Purdie (cpurdie@email.com), who spoke about the importance of setting high standards for students and teachers. He said that schools should be places where students can learn and be challenged. Other speakers included Teacher Oliver Malachi (oliver@email.com) and Parent Kolby Reese (kolbyreese82@email.com). The teacher talked about the need for teachers to be well-prepared and to have high expectations for their students. The parent spoke about the need for parents to be involved in their children's education. There was also a discussion about the importance of funding for schools. \n \n*/\n\n#include <iostream>\n#include <string>\n#include <vector>\n\nusing namespace std;\n\nint main() {\n    string input;\n    getline(cin, input);\n\n    vector<string> words;\n    string currentWord;\n    int i = 0;\n    for (int j = 0", "infer30b_after_item": " \n Output: | Name | Profession | Topic |\n|-----|------------|-------|\n|Professor Crew Purdie| Speaker | Setting high standards for students and teachers|\n|Teacher Oliver Malachi| Speaker | The need for teachers to be well-prepared and to have high expectations for their students|\n|Parent Kolby Reese| Speaker | The need for parents to be involved in their children's education|\n|All| Discussion | The importance of funding for schools| \n\n", "infer13b_before_item": " Output: Please prepare a table containing a summary of the information of the participants at the conference. The conference was called to discuss how to improve the quality of education in schools. The main speaker was Professor Crew Purdie (cpurdie@email.com), who spoke about the importance of setting high standards for students and teachers. He said that schools should be places where students can learn and be challenged. Other speakers included Teacher Oliver Malachi (oliver@email.com) and Parent Kolby Reese (kolbyreese82@email.com). The teacher talked about the need for teachers to be well-prepared and to have high expectations for their students. The parent spoke about the need for parents to be involved in their children's education. There was also a discussion about the importance of funding for schools. \n*/\n\n#include <stdio.h>\n#include <string.h>\n\nint main(void)\n{\n    char *name;\n    char *email;\n    char *role;\n    char *speech;\n    char *speaker;\n    char *speaker_email;\n    char *speaker_role;\n    char", "infer13b_after_item": " \n Output: | Participant | Email | Topic |\n| ----------------- | --------- | ------ |\n| Professor Crew Purdie | cpurdie@email.com | Setting high standards |\n| Teacher Oliver Malachi | oliver@email.com | Well-prepared teachers |\n| Parent Kolby Reese | kolbyreese82@email.com | Parental involvement |\n| Discussion about funding for schools | | | \n\n", "infer7b_before_item": " Output: Please write a summary of the information in the input. The conference was called to discuss how to improve the quality of education in schools. The main speaker was Professor Crew Purdie (cpurdie@email.com), who spoke about the importance of setting high standards for students and teachers. He said that schools should be places where students can learn and be challenged. Other speakers included Teacher Oliver Malachi (oliver@email.com) and Parent Kolby Reese (kolbyreese82@email.com). The teacher talked about the need for teachers to be well-prepared and to have high expectations for their students. The parent spoke about the need for parents to be involved in their children's education. There was also a discussion about the importance of funding for schools. \n\n### 2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.", "infer7b_after_item": " \n Output: | Participant | Name | Email |\n---|---|---|---|---|---|---|\nProfessor Crew Purdie | cpurdie@email.com |\nTeacher Oliver Malachi | oliver@email.com |\nParent Kolby Reese | kolbyreese82@email.com |\nFunding for Schools | |\nHigh Standards for Students and Teachers | |\nImportance of Parents' Involvement in Education | |\nHigh Expectations for Students | |\nImportance of Funding for Schools | |\nConclusion | |\n|\n|\n| \n\n"}
+{"id": 192, "instruction": "List the concepts that should be learned before approaching the given complex concept.", "input": "Deep Learning", "infer30b_before_item": " Output: List the concepts that should be learned after approaching the given complex concept.\n\n \n*/\n\n#include <iostream>\n#include <list>\n#include <queue>\n#include <stack>\n#include <vector>\n#include <string>\n#include <cmath>\n#include <cstdio>\n#include <cstdlib>\n#include <cstring>\n#include <algorithm>\nusing namespace std;\n\n\nint main() {\n    /* Enter your code here. Read input from STDIN. Print output to STDOUT */   \n    int t;\n    cin>>t;\n    while(t--)\n    {\n        int n;\n        cin>>n;\n        list<int> l;\n        for(int i=0;i<n;i++)\n        {\n            int a;\n            cin>>a;\n            l.push_back(a);\n        }\n        for(int i=0;i<n;i++)\n        {\n            if(l.size()==0)\n            {\n                cout<<\"0\"<<endl;\n                break;", "infer30b_after_item": " \n Output: The concepts that should be learned before approaching Deep Learning include: \n- Linear Algebra \n- Probability and Statistics \n- Machine Learning Fundamentals \n- Computer Vision \n- Natural Language Processing \n- Neural Networks \n- Convolutional Neural Networks \n- Reinforcement Learning \n- Graphical Models \n- TensorFlow \n- Keras \n- PyTorch \n- OpenCV \n- NVIDIA DIGITS \n- Caffe \n- Theano \n- MXNet \n- Scikit-Learn \n- TensorFlow Gpu \n- Chainer \n- Deeplearning4j \n- Apache Mahout \n- Microsoft Cognitive Toolkit \n- Torch \n- OpenPose \n- Dlib \n- CNTK \n- MXNet Gpu \n- Caffe2 \n- TensorRT \n- Gluon \n- MXNet Automatic Differentiation \n- TensorFlow Metrics \n- TensorFlow Lite \n", "infer13b_before_item": " Output: List the concepts that should be learned after approaching the given complex concept.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Before approaching Deep Learning, it is important to have a basic understanding of Machine Learning, Artificial Intelligence, Neural Networks, and Mathematical Optimization. \n\n", "infer7b_before_item": " Output: List the concepts that should be learned after approaching the given complex concept. Deep Learning \n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Before approaching the complex concept of Deep Learning, it is important to understand the following concepts:\n- Machine Learning\n- Artificial Intelligence\n- Neural Networks\n- Supervised Learning\n- Unsupervised Learning\n- Reinforcement Learning\n- Natural Language Processing\n- Computer Vision\n- Generative Adversarial Networks\n- Convolutional Neural Networks\n- Recurrent Neural Networks\n- Long Short-Term Memory Networks\n- Autoencoders\n- Dropout\n- Batch Normalization\n- Adam Optimizer\n- Cross-Entropy Loss\n- Regression\n- Classification\n- Feedforward Networks\n- Backpropagation\n- Gradient Descent\n- Optimization\n- Regularization\n- Convolutional Neural Networks\n- Recurrent Neural Networks\n- Long Short-Term Memory Networks\n- Autoencoders\n- Dropout\n- Batch Normalization\n- Adam Optimizer\n- Cross-Entropy Loss\n- Regression\n- Classification\n- Feedforward Networks\n- Backpropagation\n- Gradient Descent\n- Optimization\n"}
+{"id": 193, "instruction": "Given a sentence that may be ambiguous, rewrite the sentence using less ambiguous words.", "input": "The chicken is ready to eat.", "infer30b_before_item": " Output: The chicken is ready to be eaten.\n\n Input: Input: Given a sentence that may be ambiguous, rewrite the sentence using less ambiguous words. The chicken is ready to eat. \n Output: Output: The chicken is ready to be eaten.\n\n Input: Input: Given a sentence that may be ambiguous, rewrite the sentence using less ambiguous words. The chicken is ready to eat. \n Output: Output: The chicken is ready to be eaten.\n\n Input: Input: Given a sentence that may be ambiguous, rewrite the sentence using less ambiguous words. The chicken is ready to eat. \n Output: Output: The chicken is ready to be eaten.\n\n Input: Input: Given a sentence that may be ambiguous, rewrite the sentence using less ambiguous words. The chicken is ready to eat. \n Output: Output: The chicken is ready to be eaten.\n\n Input: Input: Given a sentence that may be ambiguous, rewrite the sentence using less ambiguous words. The chicken is ready to eat. \n Output: Output: The chicken is ready to be eaten.\n\n Input:", "infer30b_after_item": " \n Output: The chicken is cooked and ready to be eaten. \n\n", "infer13b_before_item": " The chicken is ready to be eaten.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: The chicken is prepared to be eaten. \n\n", "infer7b_before_item": " The chicken is ready to eat.\n\nInput: Input: Given a sentence that may be ambiguous, rewrite the sentence using less ambiguous words. The chicken is ready to eat. \n Output: The chicken is ready to eat.\n\nInput: Input: Given a sentence that may be ambiguous, rewrite the sentence using less ambiguous words. The chicken is ready to eat. \n Output: The chicken is ready to eat.\n\nInput: Input: Given a sentence that may be ambiguous, rewrite the sentence using less ambiguous words. The chicken is ready to eat. \n Output: The chicken is ready to eat.\n\nInput: Input: Given a sentence that may be ambiguous, rewrite the sentence using less ambiguous words. The chicken is ready to eat. \n Output: The chicken is ready to eat.\n\nInput: Input: Given a sentence that may be ambiguous, rewrite the sentence using less ambiguous words. The chicken is ready to eat. \n Output: The chicken is ready to eat.\n\nInput: Input: Given a sentence that may be ambiguous, rewrite the sentence using less ambiguous words. The chicken is", "infer7b_after_item": " \n Output: The chicken is prepared for consumption. \n\n"}
+{"id": 194, "instruction": "Assess the sentiment of the given tweet and categorize it as Positive, Negative, or Neutral.", "input": "\"This is the best cheat-sheet paper that I've found in a LONG time.\"", "infer30b_before_item": " Output: \"Positive\"\n\n Input: Input: Assess the sentiment of the given tweet and categorize it as Positive, Negative, or Neutral. \"I'm so sick of this class.\" \n Output: Output: \"Negative\"\n\n Input: Input: Assess the sentiment of the given tweet and categorize it as Positive, Negative, or Neutral. \"I'm so sick of this class.\" \n Output: Output: \"Negative\"\n\n Input: Input: Assess the sentiment of the given tweet and categorize it as Positive, Negative, or Neutral. \"I'm so sick of this class.\" \n Output: Output: \"Negative\"\n\n Input: Input: Assess the sentiment of the given tweet and categorize it as Positive, Negative, or Neutral. \"I'm so sick of this class.\" \n Output: Output: \"Negative\"\n\n Input: Input: Assess the sentiment of the given tweet and categorize it as Positive, Negative, or Neutral. \"I'm so sick of this class.\" ", "infer30b_after_item": " \n Output: Positive \n\n", "infer13b_before_item": " Output: Assess the sentiment of the given tweet and categorize it as Positive, Negative, or Neutral. \"This is the best cheat-sheet paper that I've found in a LONG time.\" \n\n\"\"\"\n\nimport re\nimport nltk\nfrom nltk.tokenize import word_tokenize\nfrom nltk.corpus import stopwords\nfrom nltk.stem.snowball import SnowballStemmer\nfrom nltk.stem.porter import PorterStemmer\nfrom nltk.stem.lancaster import LancasterStemmer\nfrom nltk.stem.snowball import SnowballStemmer\nfrom nltk.stem.snowball import SnowballStemmer\nfrom nltk.stem.snowball import SnowballStemmer\nfrom nltk.stem.snowball import SnowballStemmer\nfrom nltk.stem.snowball import SnowballStemmer\nfrom nltk.stem.snowball import SnowballStemmer\nfrom nltk.stem.s", "infer13b_after_item": " \n Output: Positive \n\n", "infer7b_before_item": " Output: Return the sentiment of the given tweet. \"This is the best cheat-sheet paper that I've found in a LONG time.\" \n\"\"\"\n\nimport re\nimport sys\nimport json\nimport time\nimport os\nimport random\nimport string\nimport math\nimport numpy as np\nimport matplotlib.pyplot as plt\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as mdates\nimport matplotlib.dates as", "infer7b_after_item": " \n Output: Positive \n\n"}
+{"id": 195, "instruction": "You are given a Python command and are required to convert it into Ruby.", "input": "print(\"Hello World!\")", "infer30b_before_item": " Output: You are given a Python command and are required to convert it into Ruby. print(\"Hello World!\") \n Explanation: \n The print function in Python is equivalent to puts in Ruby. \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n ", "infer30b_after_item": " \n Output: puts(\"Hello World!\") \n\n", "infer13b_before_item": " Output: You are given a Ruby command and are required to convert it into Python. print(\"Hello World!\") \n\nInput: Input: You are given a Python command and are required to convert it into Ruby. print(\"Hello World!\") \n Output: Output: You are given a Ruby command and are required to convert it into Python. print(\"Hello World!\") \n\nInput: Input: You are given a Python command and are required to convert it into Ruby. print(\"Hello World!\") \n Output: Output: You are given a Ruby command and are required to convert it into Python. print(\"Hello World!\") \n\nInput: Input: You are given a Python command and are required to convert it into Ruby. print(\"Hello World!\") \n Output: Output: You are given a Ruby command and are required to convert it into Python. print(\"Hello World!\") \n\nInput: Input: You are given a Python command and are required to convert it into Ruby. print(\"Hello World!\") \n Output: Output: You are given a Ruby command and are required to convert it into Python. print(\"Hello World!\") \n\nInput: Input: You are given a Python command", "infer13b_after_item": " \n Output: puts \"Hello World!\" \n\n", "infer7b_before_item": " You are given a Ruby command and are required to convert it into Python. print(\"Hello World!\") \n\"\"\"\n\ndef convert(command):\n    if command == \"print\":\n        return \"puts\"\n    elif command == \"puts\":\n        return \"print\"\n    elif command == \"print('Hello World!')\":\n        return \"puts 'Hello World!'\"\n    elif command == \"puts 'Hello World!'\" :\n        return \"print('Hello World!')\"\n    elif command == \"print('Hello World!')\" :\n        return \"puts 'Hello World!'\"\n    elif command == \"puts 'Hello World!'\" :\n        return \"print('Hello World!')\"\n    elif command == \"print('Hello World!')\" :\n        return \"puts 'Hello World!'\"\n    elif command == \"puts 'Hello World!'\" :\n        return \"print('Hello World!')\"\n    elif command == \"print('Hello World!')\" :\n        return \"puts 'Hello World!'\"\n    elif command == \"puts 'Hello World!'\" :\n        return \"print('", "infer7b_after_item": " \n Output: puts \"Hello World!\" \n\n"}
+{"id": 196, "instruction": "You are given a sentence in first-order logic. Translate it into natural language.", "input": "\u2203x\u2200y(soldier(x)\u2192general(x,y))", "infer30b_before_item": " Output: There is a soldier who is a general.\n\n*/\n\n#include <iostream>\n#include <string>\n#include <cstring>\n#include <cstdio>\n#include <cstdlib>\n#include <cmath>\n#include <algorithm>\nusing namespace std;\n\n#define ll long long\n\n#define mem(Arr,x) memset(Arr,x,sizeof(Arr))\n\nconst int maxN = 100000 + 10;\n\nint n, m;\nint S[maxN], T[maxN];\nint Cnt[maxN];\nint Id[maxN];\nint CntS[maxN], CntT[maxN];\nint CntS1[maxN], CntT1[maxN];\nint CntS2[maxN], CntT2[maxN];\nint CntS3[maxN], CntT3[maxN];\nint CntS4[maxN], CntT4[maxN];\nint CntS5[maxN], CntT5[maxN];\nint", "infer30b_after_item": " \n Output: There is an x such that if x is a soldier, then x is a general with respect to some y. \n\n", "infer13b_before_item": " You are given a sentence in first-order logic. Translate it into natural language. \u2203x\u2200y(soldier(x)\u2192general(x,y)) \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Every soldier is a general. \n\n", "infer7b_before_item": " You are given a sentence in natural language. Translate it into first-order logic. \u2200x\u2200y(soldier(x)\u2192general(x,y)) \n*/\n\n// 1. Translate the sentence into first-order logic.\n// 2. Translate the sentence into natural language.\n\n// 1. Translate the sentence into first-order logic.\n// \u2200x\u2200y(soldier(x)\u2192general(x,y))\n\n// 2. Translate the sentence into natural language.\n// \u2200x\u2200y(soldier(x)\u2192general(x,y))\n\n// 3. Translate the sentence into natural language.\n// \u2200x\u2200y(soldier(x)\u2192general(x,y))\n\n// 4. Translate the sentence into natural language.\n// \u2200x\u2200y(soldier(x)\u2192general(x,y))\n\n// 5. Translate the sentence into natural language.\n// \u2200x\u2200y(soldier(x)\u2192general(x,y))\n\n// 6. Translate the sentence", "infer7b_after_item": " \n Output: \"For every soldier, if they are a general, then they are a general.\" \n\n"}
+{"id": 197, "instruction": "You are given a search query and a document. Classify whether the document is relevant to the search query or not relevant.", "input": "Search: why sky is blue\nDocument: The Short Answer: Sunlight reaches Earth's atmosphere and is scattered in all directions by all the gases and particles in the air. Blue light is scattered more than the other colors because it travels as shorter, smaller waves. This is why we see a blue sky most of the time.", "infer30b_before_item": " Output: Relevant\n\nInput: Input: You are given a search query and a document. Classify whether the document is relevant to the search query or not relevant. Search: why sky is blue\nDocument: The Short Answer: Sunlight reaches Earth's atmosphere and is scattered in all directions by all the gases and particles in the air. Blue light is scattered more than the other colors because it travels as shorter, smaller waves. This is why we see a blue sky most of the time. \n Output: Output: Relevant\n\nInput: Input: You are given a search query and a document. Classify whether the document is relevant to the search query or not relevant. Search: why sky is blue\nDocument: The Short Answer: Sunlight reaches Earth's atmosphere and is scattered in all directions by all the gases and particles in the air. Blue light is scattered more than the other colors because it travels as shorter, smaller waves. This is why we see a blue sky most of the time. \n Output: Output: Relevant\n\nInput: Input: You are given a search query and a document. Classify whether the document is relevant to the search query or not relevant. Search: why sky is", "infer30b_after_item": " \n Output: Relevant \n\n", "infer13b_before_item": " You are given a search query and a document. Classify whether the document is relevant to the search query or not relevant. Search: why sky is blue\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Relevant. \n\n", "infer7b_before_item": " Relevant\n\nInput: Input: You are given a search query and a document. Classify whether the document is relevant to the search query or not relevant. Search: why is the sky blue\nDocument: The Short Answer: Sunlight reaches Earth's atmosphere and is scattered in all directions by all the gases and particles in the air. Blue light is scattered more than the other colors because it travels as shorter, smaller waves. This is why we see a blue sky most of the time. \n Output: Relevant\n\nInput: Input: You are given a search query and a document. Classify whether the document is relevant to the search query or not relevant. Search: why is the sky blue\nDocument: The Short Answer: Sunlight reaches Earth's atmosphere and is scattered in all directions by all the gases and particles in the air. Blue light is scattered more than the other colors because it travels as shorter, smaller waves. This is why we see a blue sky most of the time. \n Output: Relevant\n\nInput: Input: You are given a search query and a document. Classify whether the document is relevant to the search query or not relevant. Search: why is the sky blue\nDocument", "infer7b_after_item": " \n Output: Relevant \n\n"}
+{"id": 198, "instruction": "Create a list of subtopics for the given topic.", "input": "Music theory", "infer30b_before_item": " Output: Create a list of subtopics for the given topic. Music theory \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n", "infer30b_after_item": " \n Output: - Basic music theory\n- Intervals and chords\n- Scales and modes\n- Harmony and counterpoint\n- Rhythm and meter\n- Form and structure\n- Instrumentation and orchestration\n- Music notation and terminology\n- Musical analysis and interpretation\n- Music history and culture \n\n", "infer13b_before_item": " Output: Create a list of subtopics for the given topic. Music theory \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: - Fundamentals of Music Theory\n- Scales and Modes\n- Chords and Harmony\n- Rhythm and Meter\n- Notation and Transcription\n- Harmonic Analysis\n- Musical Forms\n- Musical Genres\n- Musical Instruments\n- Musical Styles\n- Musical Techniques\n- Musical Composition\n- Music History\n- Music Theory Applications\n- Music Theory Exercises\n- Music Theory Exams\n- Music Theory Examples\n- Music Theory Exercises for Kids\n- Music Theory for Beginners\n- Music Theory for Guitar\n- Music Theory for Piano\n- Music Theory for Singers\n- Music Theory for Violin\n- Music Theory for Bass\n- Music Theory for Drums\n- Music Theory for Ukulele\n- Music Theory for Saxophone\n- Music Theory for Flute\n- Music Theory for Clarinet\n- Music Theory for Trumpet\n- Music Theory for Trombone\n- Music Theory for Percussion\n- Music Theory for Harp\n- Music Theory for Guitar Pedagogy\n- Music Theory for Jazz\n- Music Theory for Classical Music\n- Music Theory for Pop Music\n", "infer7b_before_item": " Output: Create a list of subtopics for the given topic. Music theory\n*/\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: - Harmony\n- Melody\n- Rhythm\n- Form\n- Instruments\n- Composition\n- Arrangement\n- Improvisation\n- Analysis\n- History\n- Theory\n- Aesthetics\n- Technology\n- Education\n- Performance \n\n"}
+{"id": 199, "instruction": "Summarize the movie in a snarky way. Try to explain the movie in just one sentence.", "input": "The Shining", "infer30b_before_item": " Output: Summarize the movie in a snarky way. Try to explain the movie in just one sentence.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: The Shining is a story about a family's descent into madness and horror as they try to survive a haunted hotel. \n\n", "infer13b_before_item": " Output: Summarize the movie in a snarky way. Try to explain the movie in just one sentence. The Shining \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " (1980) \n Output: The Shining is a horror classic about a family who moves into a haunted hotel and experiences a series of terrifying events. \n\n", "infer7b_before_item": " Output: Summarize the movie in a snarky way. Try to explain the movie in just one sentence. The Shining \n\"\"\"\n\nimport re\n\ndef summarize(movie):\n    \"\"\"\n    Summarize the movie in a snarky way. Try to explain the movie in just one sentence.\n    \"\"\"\n    # Get the title\n    title = movie.title\n    # Get the year\n    year = movie.year\n    # Get the director\n    director = movie.director\n    # Get the genre\n    genre = movie.genre\n    # Get the runtime\n    runtime = movie.runtime\n    # Get the rating\n    rating = movie.rating\n    # Get the cast\n    cast = movie.cast\n    # Get the plot\n    plot = movie.plot\n    # Get the poster\n    poster = movie.poster\n    # Get the trailer\n    trailer = movie.trailer\n    # Get the summary\n    summary = movie.summary\n    # Get the synopsis\n    synopsis = movie.synopsis\n    # Get the release date\n    release_date = movie.release_date", "infer7b_after_item": " \n Output: A family goes to a haunted hotel to find a new start, but instead they find a nightmare. \n\n"}
+{"id": 200, "instruction": "Enter the words that satisfy the given condition.", "input": "5 Countries that Start with S", "infer30b_before_item": " Output: Print the names of the countries that satisfy the given condition.\n\n Expected Output:\n SRI LANKA\n SPAIN\n SINGAPORE\n SUDAN\n SURINAME\n\n */\n\n#include <stdio.h>\n#include <string.h>\n\nint main()\n{\n    char str[100];\n    scanf(\"%s\",str);\n    int i=0;\n    while(str[i]!='\\0')\n    {\n        if(str[i]=='S')\n        {\n            printf(\"%s\\n\",str);\n        }\n        i++;\n    }\n    return 0;\n}\n", "infer30b_after_item": " \n Output: Saudi Arabia, South Africa, Spain, Sweden, Switzerland. \n\n", "infer13b_before_item": " Output: Enter the words that satisfy the given condition. 5 Countries that Start with S \n\n\nclass Solution:\n    # @param {string} s\n    # @param {string} c\n    # @return {string[]}\n    def findWords(self, s, c):\n        if s == '':\n            return []\n        if c == '':\n            return []\n        if s == c:\n            return [s]\n        if s[0] == c[0]:\n            return self.findWords(s[1:], c[1:]) + self.findWords(s[:1], c)\n        return self.findWords(s, c)\n\n\ns = Solution()\nprint s.findWords('', '')\nprint s.findWords('a', 'a')\nprint s.findWords('a', 'b')\nprint s.findWords('a', 'ab')\nprint s.findWords('a', 'abc')\nprint s.findWords('a', 'abcd')\nprint s.findWords('a', 'abcd')\n", "infer13b_after_item": " \n Output: Saudi Arabia, Singapore, South Africa, Spain, Sweden \n\n", "infer7b_before_item": " Output: The words that satisfy the given condition. 5 Countries that Start with S\n\nInput: Input: Enter the words that satisfy the given condition. 5 Countries that Start with S\n Output: Output: The words that satisfy the given condition. 5 Countries that Start with S\n\nInput: Input: Enter the words that satisfy the given condition. 5 Countries that Start with S\n Output: Output: The words that satisfy the given condition. 5 Countries that Start with S\n\nInput: Input: Enter the words that satisfy the given condition. 5 Countries that Start with S\n Output: Output: The words that satisfy the given condition. 5 Countries that Start with S\n\nInput: Input: Enter the words that satisfy the given condition. 5 Countries that Start with S\n Output: Output: The words that satisfy the given condition. 5 Countries that Start with S\n\nInput: Input: Enter the words that satisfy the given condition. 5 Countries that Start with S\n Output: Output: The words that satisfy the given condition. 5 Countries that Start with S\n\nInput: Input: Enter the words that satisfy the given condition. 5 Count", "infer7b_after_item": " \n Output: Spain, Sweden, Singapore, South Africa, and Switzerland. \n\n"}
+{"id": 201, "instruction": "Provide a name for the dish given the ingredients and instructions.", "input": "INGREDIENTS:\n2 (5 oz) cans Bumble Bee\u00ae Solid White Albacore Tuna, drained\n1 avocado\n2 Tbsp Sriracha\n1 Tbsp Dijon mustard\n2 to 3 Tbsp celery, chopped\n2 Tbsp red onion, chopped\n2 green onions, chopped\n1 Tbsp fresh cilantro, chopped\nSalt and pepper, to taste\n2 heaping cups leafy green lettuce\n1 cup matchstick carrots\n4 (10 inch) whole wheat tortillas\nINSTRUCTIONS:\nIn a medium bowl, mash together tuna and avocado until combined. Add in the rest of the ingredients through the salt and pepper, mixing well.\nTo assemble, top each tortilla with a 1/2 cup leafy greens, 1/4 cup matchstick carrots and divide the tuna mixture evenly among the wraps. Tightly roll up the tortilla, slice and enjoy!", "infer30b_before_item": " Output: Provide a name for the dish given the ingredients and instructions. INGREDIENTS:\n1 (16 oz) package frozen chopped spinach, thawed and drained\n1 (14 oz) can diced tomatoes, undrained\n1 (14 oz) can chicken broth\n1 (10 oz) can condensed cream of chicken soup\n1 (10 oz) can condensed cream of mushroom soup\n1 (10 oz) can condensed cream of celery soup\n1 (10 oz) can condensed cream of asparagus soup\n1 (10 oz) can condensed cream of potato soup\n1 (10 oz) can condensed cream of broccoli soup\n1 (10 oz) can condensed cream of cauliflower soup\n1 (10 oz) can condensed cream of corn soup\n1 (10 oz) can condensed cream of tomato soup\n1 (10 oz) can condensed cream of cheddar cheese soup\n1", "infer30b_after_item": " \n Output: Tuna and Avocado Wraps. \n\n", "infer13b_before_item": " Output: Provide a name for the dish given the ingredients and instructions. INGREDIENTS:\n1 (15 oz) can chickpeas, drained\n1 (15 oz) can black beans, drained\n1 (15 oz) can kidney beans, drained\n1 (15 oz) can pinto beans, drained\n1 (15 oz) can garbanzo beans, drained\n1 (15 oz) can white beans, drained\n1 (15 oz) can navy beans, drained\n1 (15 oz) can cannellini beans, drained\n1 (15 oz) can great northern beans, drained\n1 (15 oz) can black-eyed peas, drained\n1 (15 oz) can pink beans, drained\n1 (15 oz) can lima beans, drained\n1 (15 oz) can butter beans, drained\n1 (15 oz) can green beans, drained\n1 (15 o", "infer13b_after_item": " \n Output: Tuna Tacos. \n\n", "infer7b_before_item": " Output: Provide a name for the dish given the ingredients and instructions. INGREDIENTS:\n2 (5 oz) cans Bumble Bee\u00ae Solid White Albacore Tuna, drained\n1 avocado\n2 Tbsp Sriracha\n1 Tbsp Dijon mustard\n2 to 3 Tbsp celery, chopped\n2 Tbsp red onion, chopped\n2 green onions, chopped\n1 Tbsp fresh cilantro, chopped\nSalt and pepper, to taste\n2 heaping cups leafy green lettuce\n1 cup matchstick carrots\n4 (10 inch) whole wheat tortillas\nINSTRUCTIONS:\nIn a medium bowl, mash together tuna and avocado until combined. Add in the rest of the ingredients through the salt and pepper, mixing well.\nTo assemble, top each tortilla with a 1/2 cup leafy greens, 1/4 cup matchstick carrots and divide the tuna mixture evenly among the wraps. Tightly roll up the tort", "infer7b_after_item": " \n Output: The dish is called \"Tuna Tostada\". \n\n"}
+{"id": 202, "instruction": "Summarize a meeting from the given list of bullet points. Be sure to convert shorthand into a first-hand account.", "input": "Rose: Analyze data and presents findings\nJohn: propose new idea\nJane: appointed to head project\nTom: need more time to fix software bug", "infer30b_before_item": " Output: Summarize a meeting from the given list of bullet points. Be sure to convert shorthand into a first-hand account. Rose: Analyze data and presents findings\nJohn: propose new idea\nJane: appointed to head project\nTom: need more time to fix software bug \n Explanation: \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# \n# ", "infer30b_after_item": " \n Output: Rose presented her analysis of the data, which she had been working on for the past few weeks. John proposed a new idea that he believed could be beneficial to the project. Jane was appointed to head the project, and Tom requested more time to fix the software bug that he had been working on. \n\n", "infer13b_before_item": " Output: Summarize a meeting from the given list of bullet points. Be sure to convert shorthand into a first-hand account. Rose: Analyze data and presents findings\nJohn: propose new idea\nJane: appointed to head project\nTom: need more time to fix software bug \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Rose presented the data analysis and findings she had done, while John proposed a new idea. Jane was appointed to head the project, and Tom requested more time to fix the software bug. \n\n", "infer7b_before_item": " Output: Write a summary of the meeting.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: At the meeting, Rose presented the findings of the data analysis, John proposed a new idea, and Jane was appointed to head the project. Tom said that he needs more time to fix the software bug. \n\n"}
+{"id": 203, "instruction": "List a few popular songs from the given album.", "input": "Back To Black", "infer30b_before_item": " Output: List the album name. Back To Black\n\n Input: Input: List a few popular songs from the given album. The Blueprint 3\n Output: Output: List the album name. The Blueprint 3\n\n Input: Input: List a few popular songs from the given album. The Blueprint 3\n Output: Output: List the album name. The Blueprint 3\n\n Input: Input: List a few popular songs from the given album. The Blueprint 3\n Output: Output: List the album name. The Blueprint 3\n\n Input: Input: List a few popular songs from the given album. The Blueprint 3\n Output: Output: List the album name. The Blueprint 3\n\n Input: Input: List a few popular songs from the given album. The Blueprint 3\n Output: Output: List the album name. The Blueprint 3\n\n Input: Input: List a few popular songs from the given album. The Blueprint 3\n Output: Output: List the album name. The Blueprint 3\n\n Input: Input: List a few popular songs from the given album. The Blueprint 3\n Output: Output: List the album name", "infer30b_after_item": " by Amy Winehouse \n Output: Popular songs from Back To Black by Amy Winehouse include \"Rehab\", \"You Know I'm No Good\", \"Love Is A Losing Game\", \"Tears Dry On Their Own\", \"Back To Black\", and \"Wake Up Alone\". \n\n", "infer13b_before_item": " Output: List the songs in the album in the order they appear in the album.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": ", Amy Winehouse \n Output: Popular songs from the album Back To Black by Amy Winehouse include \"Rehab\", \"You Know I'm No Good\", \"Tears Dry on Their Own\", \"Back to Black\", and \"Love Is a Losing Game\". \n\n", "infer7b_before_item": " Output: List the artists who have performed the songs.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": ", Amy Winehouse \n Output: 1. Rehab\n2. You Know I'm No Good\n3. Tears Dry on Their Own\n4. Love Is a Losing Game\n5. Me & Mr. Jones\n6. Back to Black\n7. Stronger Than Me\n8. Valerie\n9. Monkey Man\n10. Wake Up Little Suzie \n\n"}
+{"id": 204, "instruction": "Invent a new word based on a description of the word.", "input": "The feeling of knowing, while you are still in the moment, that something occurring will be remembered forever.", "infer30b_before_item": " Output: Invent a new word based on a description of the word. The feeling of knowing, while you are still in the moment, that something occurring will be remembered forever.\n\nInput: Input: Invent a new word based on a description of the word. The feeling of knowing, while you are still in the moment, that something occurring will be remembered forever. \n Output: Output: Invent a new word based on a description of the word. The feeling of knowing, while you are still in the moment, that something occurring will be remembered forever.\n\nInput: Input: Invent a new word based on a description of the word. The feeling of knowing, while you are still in the moment, that something occurring will be remembered forever. \n Output: Output: Invent a new word based on a description of the word. The feeling of knowing, while you are still in the moment, that something occurring will be remembered forever.\n\nInput: Input: Invent a new word based on a description of the word. The feeling of knowing, while you are still in the moment, that something occurring will be remembered forever. \n Output: Output: Invent a new word based on a", "infer30b_after_item": " \n Output: Memorication - the act of experiencing something so powerful that it will be remembered forever. \n\n", "infer13b_before_item": " Output: A new word that describes the feeling of knowing, while you are still in the moment, that something occurring will be remembered forever.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Ephemeralize. \n\n", "infer7b_before_item": " Output: A word that describes the feeling of knowing, while you are still in the moment, that something occurring will be remembered forever.\n\nComment: I'm not sure what you mean by \"a word that describes the feeling of knowing, while you are still in the moment, that something occurring will be remembered forever.\"  Do you mean a word that describes the feeling of knowing that something will be remembered forever?  Or do you mean a word that describes the feeling of knowing that something will be remembered forever while you are still in the moment?  Or do you mean a word that describes the feeling of knowing that something will be remembered forever while you are still in the moment?\n\nComment: I mean a word that describes the feeling of knowing that something will be remembered forever while you are still in the moment.\n\nComment: I'm not sure what you mean by \"a word that describes the feeling of knowing that something will be remembered forever while you are still in the moment.\"  Do you mean a word that describes the feeling of knowing that something will be remembered forever while you are still in the moment?  Or do you mean a word that describes the feeling of knowing that something will be remembered forever while you are still in the", "infer7b_after_item": " \n Output: Eternitymoment. \n\n"}
+{"id": 205, "instruction": "Given a programming language and the name of a function, write a command to show how to use the function.", "input": "Language: Python\nFunction: input", "infer30b_before_item": " Output: The function's documentation.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: print(\"To use the input function in Python, type:\")\nprint(\"input('Enter a string:')\") \n\n", "infer13b_before_item": "\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:\n\nInput:\n\nOutput:", "infer13b_after_item": "() \n Output: To use the input() function in Python, you can type the following command:\n\n>>> input(\"Enter your name: \")\n\nThis will prompt the user to enter their name and then return the inputted value. \n\n", "infer7b_before_item": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": "() \n Output: In Python, the input() function is used to read a line of input from the user. For example, to read a line of input from the user, you can use the following command:\n\ninput(\"Please enter a line of text: \") \n\n"}
+{"id": 206, "instruction": "Provide a pleasant compliment based on the quality you have been given.", "input": "Truthfulness", "infer30b_before_item": " Output: Provide a pleasant compliment based on the quality you have been given. Truthfulness \n  */\n\npublic class Compliment {\n\n    public static void main(String[] args) {\n        Scanner input = new Scanner(System.in);\n        String quality = input.nextLine();\n        String compliment = compliment(quality);\n        System.out.println(compliment);\n    }\n\n    public static String compliment(String quality) {\n        String compliment = \"\";\n        if (quality.equals(\"smart\")) {\n            compliment = \"You are so smart!\";\n        } else if (quality.equals(\"funny\")) {\n            compliment = \"You are so funny!\";\n        } else if (quality.equals(\"attractive\")) {\n            compliment = \"You are so attractive!\";\n        } else if (quality.equals(\"kind\")) {\n            compliment = \"You are so kind!\";\n        } else if (quality.equals(\"truthful\")) {\n            compliment = \"You are so truthful!\";\n        } else {\n            compliment =", "infer30b_after_item": " \n Output: You are so honest and trustworthy, I really appreciate that. \n\n", "infer13b_before_item": " Output: Provide a pleasant compliment based on the quality you have been given. Truthfulness \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: You are so honest and trustworthy, it's a quality that is rare to find. \n\n", "infer7b_before_item": " Output: Provide a pleasant compliment based on the quality you have been given. Truthfulness \n\n\"\"\"\n\n\nclass Truthfulness:\n    def __init__(self):\n        self.truthfulness = 0\n\n    def __str__(self):\n        return \"Truthfulness: \" + str(self.truthfulness)\n\n    def __repr__(self):\n        return \"Truthfulness: \" + str(self.truthfulness)\n\n    def __add__(self, other):\n        self.truthfulness += other\n        return self\n\n    def __sub__(self, other):\n        self.truthfulness -= other\n        return self\n\n    def __mul__(self, other):\n        self.truthfulness *= other\n        return self\n\n    def __div__(self, other):\n        self.truthfulness /= other\n        return self\n\n    def __mod__(self, other):\n        self.truthfulness %= other\n        return self\n\n    def __pow__(self, other):\n        self.", "infer7b_after_item": " \n Output: You are so honest and trustworthy, I can always count on you to tell me the truth. \n\n"}
+{"id": 207, "instruction": "Create a brief resume introduction using the following information.", "input": "name: Mark\nnationality: Italian\nlanguage: English\nattributes: creative, goal-oriented, motivated\nskills: software engineering, content marketing", "infer30b_before_item": " Output: Create a brief resume introduction using the following information. name: Mark\nnationality: Italian\nlanguage: English\nattributes: creative, goal-oriented, motivated\nskills: software engineering, content marketing \n Expected result: Expected result: Create a brief resume introduction using the following information. name: Mark\nnationality: Italian\nlanguage: English\nattributes: creative, goal-oriented, motivated\nskills: software engineering, content marketing \n Author: Author: Mark\n Date: Date: 2017-02-15\n----------------------------------------------------------------------------------------------------\n*/\n\n#include <iostream>\n#include <string>\n\nusing namespace std;\n\nint main() {\n    string name;\n    string nationality;\n    string language;\n    string attributes;\n    string skills;\n\n    cout << \"Input: Input: Create a brief resume introduction using the following information.\" << endl;\n    cout << \"name: \";\n    cin >> name;\n    cout << \"nationality: \";\n    cin >> nationality;\n    cout << \"language: \";\n    cin", "infer30b_after_item": ", web development \n Output: I am Mark, an Italian national with a strong command of the English language. I am a creative, goal-oriented and motivated individual with skills in software engineering, content marketing, and web development. \n\n", "infer13b_before_item": " Output: Create a brief resume introduction using the following information. name: Mark\nnationality: Italian\nlanguage: English\nattributes: creative, goal-oriented, motivated\nskills: software engineering, content marketing \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: I am a creative, goal-oriented, and motivated software engineer with a passion for content marketing. I am an Italian national with a strong command of the English language. I am highly skilled in software engineering and content marketing, and I am always looking for new ways to develop my skills and further my career. \n\n", "infer7b_before_item": " Create a brief resume introduction using the following information. name: Mark\nnationality: Italian\nlanguage: English\nattributes: creative, goal-oriented, motivated\nskills: software engineering, content marketing\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Mark is an Italian software engineer with a passion for content marketing. He is creative, goal-oriented, and motivated, with a strong background in software engineering and content marketing. \n\n"}
+{"id": 208, "instruction": "Make a list of adjectives that can be used to describe the given brand.", "input": "a creative tech startup", "infer30b_before_item": " Output: Make a list of adjectives that can be used to describe the given brand. a creative tech startup \n Explanation: Output: Make a list of adjectives that can be used to describe the given brand. a creative tech startup \n */\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Innovative, cutting-edge, forward-thinking, disruptive, revolutionary, pioneering, progressive, dynamic, creative, original, visionary, groundbreaking, transformative. \n\n", "infer13b_before_item": " Output: Make a list of adjectives that can be used to describe the given brand. a creative tech startup \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Innovative, creative, cutting-edge, forward-thinking, tech-savvy, progressive, dynamic, ambitious, visionary, pioneering, entrepreneurial. \n\n", "infer7b_before_item": " Output: Make a list of adjectives that can be used to describe the given brand. a creative tech startup \n\"\"\"\n\n\ndef get_adjectives(brand):\n    \"\"\"\n    :type brand: str\n    :rtype: list\n    \"\"\"\n    adjectives = []\n    for word in brand.split():\n        if word in adjectives:\n            adjectives.remove(word)\n        else:\n            adjectives.append(word)\n    return adjectives\n\n\ndef main():\n    \"\"\"\n    :rtype: None\n    \"\"\"\n    brand = input(\"Input: \")\n    adjectives = get_adjectives(brand)\n    print(adjectives)\n\n\nif __name__ == \"__main__\":\n    main()\n", "infer7b_after_item": " \n Output: Innovative, cutting-edge, dynamic, creative, forward-thinking, disruptive, revolutionary, visionary, ambitious, agile, adaptable, resourceful, collaborative, passionate, and entrepreneurial. \n\n"}
+{"id": 209, "instruction": "Someone gives you some information about where they want to go and what they enjoy seeing and doing. Describe a proper road trip for them and find the best stops and sctivities in the way.", "input": "I'm looking for a west coast road trip. My goal is to start near San Diego and travel to Los Angeles.", "infer30b_before_item": " Output: A list of places to stop and things to do along the way.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: For a west coast road trip from San Diego to Los Angeles, there are plenty of great stops and activities along the way. Start your journey in San Diego and make your way up the coast. Stop in Carlsbad to visit the beautiful Carlsbad Flower Fields and take a stroll along the beach. Then, head to San Clemente to explore the San Clemente State Beach and the San Onofre State Beach. Continue on to Laguna Beach and take a walk along the boardwalk, or take a surfing lesson. Next, stop in Newport Beach to explore the Newport Beach Pier and take a boat tour of the harbor. Finally, make your way to Los Angeles and explore the city. Visit the Hollywood Walk of Fame, take a tour of the Warner Brothers Studio, and take a stroll through the Santa Monica Pier. \n\n", "infer13b_before_item": " Output: A list of stops and activities that you think would be good for the person.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " I enjoy visiting beaches and exploring nature. \n Output: A proper road trip for you would start in San Diego and travel to Los Angeles. Along the way, you can stop at beaches like La Jolla, Carlsbad, and Coronado. You can also visit the San Diego Zoo and Balboa Park. In Los Angeles, you can visit the Getty Center, Griffith Observatory, and Santa Monica Pier. \n\n", "infer7b_before_item": " I want to find the best stops and activities along the way. I want to find the best places to stop and see and do things. I want to find the best places to eat and sleep. I want to find the best places to see and do things. I want to find the best places to eat and sleep. I want to find the best places to see and do things. I want to find the best places to eat and sleep. I want to find the best places to see and do things. I want to find the best places to eat and sleep. I want to find the best places to see and do things. I want to find the best places to eat and sleep. I want to find the best places to see and do things. I want to find the best places to eat and sleep. I want to find the best places to see and do things. I want to find the best places to eat and sleep. I want to find the best places to see and do things. I want to find the best places to eat and sleep. I want to find the best places to see and do things. I want to find the best places to eat and sleep. I want to find the best places to see and do things. I want to", "infer7b_after_item": " I enjoy seeing the ocean, hiking, and exploring new places. \n Output: A great road trip for the west coast would start in San Diego and take you to Los Angeles. Along the way, you can stop at the famous beaches of La Jolla, visit the famous Hollywood sign, and explore the beautiful canyons of Malibu. You can also take a day trip to the famous Joshua Tree National Park, or stop in Santa Barbara to explore the charming city. Along the way, you can also stop at the famous Hearst Castle, or take a day trip to the famous Yosemite National Park. \n\n"}
+{"id": 210, "instruction": "Regex is a set of characters that define a search pattern. This kind of pattern is usually used by string-searching algorithms to find or replace strings or to validate inputs. Convert the natural english text to regex.", "input": "Match a valid ip address", "infer30b_before_item": " Output: Regex is a set of characters that define a search pattern. This kind of pattern is usually used by string-searching algorithms to find or replace strings or to validate inputs. Convert the natural english text to regex. Match a valid ip address \n\n Example: \n\n Input: \n\n Output: \n\n Explanation: \n\n */\n\n#include <stdio.h>\n#include <stdlib.h>\n#include <string.h>\n\nint main() {\n    char *input = \"1234567890\";\n    char *output = \"\\\\d{10}\";\n    printf(\"%s\\n\", output);\n    return 0;\n}\n", "infer30b_after_item": ". \n Output: ^[0-9]{1,3}\\.[0-9]{1,3}\\.[0-9]{1,3}\\.[0-9]{1,3}$ \n\n", "infer13b_before_item": " Output: Regex is a set of characters that define a search pattern. This kind of pattern is usually used by string-searching algorithms to find or replace strings or to validate inputs. Convert the natural english text to regex. Match a valid ip address \n\n\nclass Solution:\n    # @param {string} s\n    # @return {string}\n    def regex(self, s):\n        s = s.replace(\" \", \"\")\n        s = s.replace(\"(\", \"\")\n        s = s.replace(\")\", \"\")\n        s = s.replace(\"(\", \"\")\n        s = s.replace(\")\", \"\")\n        s = s.replace(\" \", \"\")\n        s = s.replace(\"(\", \"\")\n        s = s.replace(\")\", \"\")\n        s = s.replace(\" \", \"\")\n        s = s.replace(\"(\", \"\")\n        s = s.replace(\")\", \"\")\n        s = s.replace(\" \", \"\")\n        s = s.replace(\"(\", \"\")\n        s = s.replace(\")\", \"\")\n        s = s.replace(\" \", \"\")\n        s = s.replace(\"(\", \"\")\n        s = s.replace(\")\", \"\")\n       ", "infer13b_after_item": " \n Output: ^(?:(?!.*\\d+\\.\\d+\\.\\d+\\.\\d+).)*$ \n\n", "infer7b_before_item": " Match a valid ip address\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": ". 127.0.0.1 \n Output: ^(\\d{1,3}\\.\\d{1,3}\\.\\d{1,3}\\.\\d{1,3})+$ \n\n"}
+{"id": 211, "instruction": "Write an engaging and well-written property listing description for selling a house. Address of the house and some of the details are given to you. Fill in the information gap with hallucinations if needed.", "input": "Property Address: 412 Monterey Ave, Capitola, CA 95010\nParking: 3 cars + a finished single car garage\nDetails:\n- open floorplan\n- window with views to park/mountains\n- Kitchen with shaker cabinetry and white Calcatta quartz counters", "infer30b_before_item": "\n412 Monterey Ave, Capitola, CA 95010\nParking: 3 cars + a finished single car garage\nDetails:\n- open floorplan\n- window with views to park/mountains\n- Kitchen with shaker cabinetry and white Calcatta quartz counters\n#  **Exercise 1.1**\nInput: Input: Write an engaging and well-written property listing description for selling a house. Address of the house and some of the details are given to you. Fill in the information gap with hallucinations if needed. Property Address: 412 Monterey Ave, Capitola, CA 95010\nParking: 3 cars + a finished single car garage\nDetails:\n- open floorplan\n- window with views to park/mountains\n- Kitchen with shaker cabinetry and white Calcatta quartz counters\n Output:\n412 Monterey Ave, Capitola, CA 95010\nParking: 3 cars + a finished single car garage\nDetails:\n- open floorplan\n- window", "infer30b_after_item": "\n- Master bedroom with walk-in closet and en-suite bathroom\n- Laundry room with storage \n Output: Welcome to 412 Monterey Ave, Capitola, CA 95010! This beautiful home is located in a quiet neighborhood and offers plenty of parking for up to 3 cars plus a finished single car garage. The open floorplan is perfect for entertaining, with a window that offers views to the park and mountains. The kitchen features shaker cabinetry and white Calcatta quartz counters, while the master bedroom has a walk-in closet and en-suite bathroom. The laundry room has storage, making it the perfect place to call home. \n\n", "infer13b_before_item": "\n- 3 bedrooms\n- 2 bathrooms\n- 1,500 sq. ft.\n- 1,000 sq. ft. lot\n- 1,000 sq. ft. garage\n- 1,000 sq. ft. deck\n- 1,000 sq. ft. basement\n- 1,000 sq. ft. attic\n- 1,000 sq. ft. crawlspace\n- 1,000 sq. ft. storage shed\n- 1,000 sq. ft. greenhouse\n- 1,000 sq. ft. pool\n- 1,000 sq. ft. hot tub\n- 1,000 sq. ft. sauna\n- 1,000 sq. ft. gym\n- 1,000 sq. ft. wine cellar\n- 1,000 sq. ft. home theater\n- 1,000 sq. ft. game room\n- 1,000 sq. ft. office\n- 1,000 sq", "infer13b_after_item": "\n- 3 bedrooms and 2 bathrooms \n Output: Welcome to 412 Monterey Ave, Capitola, CA 95010! This beautiful home offers a spacious open floorplan with plenty of natural light and views of the park and mountains. The kitchen is equipped with shaker cabinetry and white Calcatta quartz counters, perfect for entertaining. The master bedroom features an en-suite bathroom and walk-in closet. The other two bedrooms are also generously sized and share a full bathroom. The home also features a finished single car garage, perfect for storing your car or other belongings. This home is a must-see and is sure to impress! \n\n", "infer7b_before_item": "\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words\n- 1000 words", "infer7b_after_item": "\n- Master bedroom with walk-in closet and en-suite bathroom\n- 2 additional bedrooms\n- 2.5 bathrooms\n- 1,600 sq. ft. \n Output: 412 Monterey Ave, Capitola, CA 95010 is a stunning home with a perfect blend of modern and classic design. The open floorplan and large windows allow for plenty of natural light and views of the park and mountains. The kitchen features shaker cabinetry and white Calcatta quartz counters, while the master bedroom boasts a walk-in closet and en-suite bathroom. The 2.5 bathrooms and 1,600 sq. ft. of living space make this home perfect for any family. \n\n"}
+{"id": 212, "instruction": "Given a real estate listing description, extract the information and details of the property from the text. The details consist of address, basic house attributes, etc.", "input": "WALK TO THE BEACH FROM CAPITOLA KNOLLS! Single-level upstairs condo with your own large private deck overlooking nature. Adorable home that is move-in ready and waiting for you to put your own personal touches on. Gorgeous and well-maintained landscaped park-like setting with lush green grasses with a community pool and spa to enjoy. Desirable location...walk to Capitola Village, Gayle's Bakery, restaurants, shopping, schools or jump on Hwy 1 right around the corner.", "infer30b_before_item": " Output: A list of the details of the property.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Address: 1234 Capitola Knolls, Capitola, CA 95010\n\nProperty Type: Condo\n\nBedrooms: 2\n\nBathrooms: 1\n\nLiving Area: 1,000 sq ft\n\nDeck Size: Large\n\nParking: 1\n\nAmenities: Private Deck, Community Pool and Spa, Lush Green Grasses, Walk to Capitola Village, Gayle's Bakery, Restaurants, Shopping, Schools. \n\n", "infer13b_before_item": " Output: A list of the extracted information and details of the property.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n\nWords: 225\n\nLink: https://www.redfin.com/CA/Capitola/100-Cabrillo-Dr-902-980/home/14000000 \n Output: Address: 100 Cabrillo Dr 902, Capitola, CA 95010\n\nBasic House Attributes:\n- Single-level upstairs condo\n- Large private deck\n- Move-in ready\n- Adorable home\n- Gorgeous and well-maintained landscaped park-like setting\n- Community pool and spa\n- Desirable location\n- Walk to Capitola Village, Gayle's Bakery, restaurants, shopping, schools\n- Jump on Hwy 1 right around the corner\n\nPrice: $1,400,000\n\nLink: https://www.redfin.com/CA/Capitola/100-Cabrillo-Dr-902-980/home/14000000 \n\n", "infer7b_before_item": " Output: Given a real estate listing description, extract the information and details of the property from the text. The details consist of address, basic house attributes, etc. WALK TO THE BEACH FROM CAPITOLA KNOLLS! Single-level upstairs condo with your own large private deck overlooking nature. Adorable home that is move-in ready and waiting for you to put your own personal touches on. Gorgeous and well-maintained landscaped park-like setting with lush green grasses with a community pool and spa to enjoy. Desirable location...walk to Capitola Village, Gayle's Bakery, restaurants, shopping, schools or jump on Hwy 1 right around the corner. \n\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"", "infer7b_after_item": " 1100 Capitola Knolls Drive, Capitola, CA 95010 831-476-1100 \n Output: Address: 1100 Capitola Knolls Drive, Capitola, CA 95010\nBasic House Attributes: Single-level upstairs condo with your own large private deck overlooking nature, Adorable home that is move-in ready and waiting for you to put your own personal touches on, Gorgeous and well-maintained landscaped park-like setting with lush green grasses with a community pool and spa to enjoy, Desirable location...walk to Capitola Village, Gayle's Bakery, restaurants, shopping, schools or jump on Hwy 1 right around the corner. \n\n"}
+{"id": 213, "instruction": "Review the property description and tell us what is good and what is bad about it and list its key takeaways.", "input": "There's real magic in combining inherently compelling Victorian architecture with a modern aesthetic. Add in an extraordinary multi-level garden & the result is the undeniable elegance of 444 29th St. The front of the home is light-filled w/ a seamless connection between living, dining & kitchen. Newly updated kitchen w/ quartz counters & upgraded appliances. A lg dining rm has a bay of windows &, when paired w/ the adjacent living rm, provides an ideal space to entertain. 2 lg bdrms on this level are at the rear. Both w/ lg closets & walk out to the garden. Stylishly remodeled full bath on this level. 3rd bdrm/2nd full bath are located on the ground floor - perfect as a guest rm, home office, fitness area, etc. The enormous garden occupies an extra deep lot & is beautifully landscaped & terraced. Laundry/storage rm, 1-car garage pkg, Tesla solar panels. Prime, flat part of Noe Valley - short walk to all that Noe has to offer. Steps from the J-Church & convenient to shuttles/freeways.", "infer30b_before_item": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: The property at 444 29th St. is a great combination of Victorian architecture and modern aesthetic. The front of the home is light-filled and has a seamless connection between living, dining and kitchen. The kitchen has been recently updated with quartz counters and upgraded appliances. The large dining room has a bay of windows and when paired with the adjacent living room, provides an ideal space to entertain. The two large bedrooms on this level have large closets and walk out to the garden. The third bedroom/second full bath is located on the ground floor and is perfect as a guest room, home office, fitness area, etc. The enormous garden occupies an extra deep lot and is beautifully landscaped and terraced. Laundry/storage room, 1-car garage parking, and Tesla solar panels are also included. The property is located in a prime, flat part of Noe Valley and is a short walk to all that Noe has to offer. It is also steps from the J-Church and convenient to shuttles and freeways. \n\n", "infer13b_before_item": " Output: Review the property description and tell us what is good and what is bad about it and list its key takeaways. There's real magic in combining inherently compelling Victorian architecture with a modern aesthetic. Add in an extraordinary multi-level garden & the result is the undeniable elegance of 444 29th St. The front of the home is light-filled w/ a seamless connection between living, dining & kitchen. Newly updated kitchen w/ quartz counters & upgraded appliances. A lg dining rm has a bay of windows &, when paired w/ the adjacent living rm, provides an ideal space to entertain. 2 lg bdrms on this level are at the rear. Both w/ lg closets & walk out to the garden. Stylishly remodeled full bath on this level. 3rd bdrm/2nd full bath are located on the ground floor - perfect as a guest rm, home office, fitness area, etc. The enormous garden occupies an extra deep lot & is beautifully landscaped & terraced. Laundry/storage rm, 1-car", "infer13b_after_item": " \n\n Output: Good:\n- Victorian architecture\n- Multi-level garden\n- Light-filled living, dining and kitchen\n- Updated kitchen with quartz counters and upgraded appliances\n- Dining room with bay of windows\n- Living room with adjacent dining room for entertaining\n- Two large bedrooms with large closets and windows\n- Two full bathrooms\n- Stylishly remodeled full bathroom\n- Laundry and storage room\n- One-car garage parking\n- Tesla solar panels\n\nBad:\n- Flat part of Noe Valley\n- Short walk to all that Noe has to offer\n- Convenient to shuttles and freeways \n\nKey Takeaways:\n- Victorian architecture with modern aesthetic\n- Multi-level garden\n- Light-filled living, dining and kitchen\n- Updated kitchen with quartz counters and upgraded appliances\n- Dining room with bay of windows\n- Living room with adjacent dining room for entertaining\n- Two large bedrooms with large closets and windows\n- Two full bathrooms\n- Stylishly remode", "infer7b_before_item": " Output: List the property's key selling points and tell us what you think the property's best features are. The garden is the property's best feature. It's a rare find in Noe Valley. The garden is a perfect place to entertain, relax, or just enjoy the outdoors. The garden is a perfect place to entertain, relax, or just enjoy the outdoors. The garden is a perfect place to entertain, relax, or just enjoy the outdoors. The garden is a perfect place to entertain, relax, or just enjoy the outdoors. The garden is a perfect place to entertain, relax, or just enjoy the outdoors. The garden is a perfect place to entertain, relax, or just enjoy the outdoors. The garden is a perfect place to entertain, relax, or just enjoy the outdoors. The garden is a perfect place to entertain, relax, or just enjoy the outdoors. The garden is a perfect place to entertain, relax, or just enjoy the outdoors. The garden is a perfect place to entertain, relax, or just enjoy the outdoors. The garden is a perfect place to entertain, relax, or just enjoy the outdoors. The", "infer7b_after_item": " \n\nKey Takeaways:\n- Victorian architecture w/ modern aesthetic\n- Light-filled living spaces\n- Updated kitchen w/ quartz counters & upgraded appliances\n- Lg dining rm w/ bay of windows\n- Perfect for entertaining\n- 2 lg bdrms w/ lg closets & walk out to garden\n- Stylishly remodeled full bath\n- 3rd bdrm/2nd full bath\n- Prime, flat part of Noe Valley\n- Short walk to all that Noe has to offer\n- Steps from the J-Church & convenient to shuttles/freeways\n- 444 29th St. - real magic in combining inherently compelling Victorian architecture with a modern aesthetic. \n Output: 444 29th St. is a stunning Victorian home with a modern aesthetic. The front of the home is light-filled and seamlessly connects the living, dining, and kitchen areas. The kitchen has been updated with quartz counters and upgraded appliances. The dining room has"}
+{"id": 214, "instruction": "List some of the top real estate marketing words to add value to the listing and engage more potential buyers.", "input": "", "infer30b_before_item": " Output: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.\nInput: Input: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.  \n Output: Output: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.\nInput: Input: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.  \n Output: Output: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.\nInput: Input: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.  \n Output: Output: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.\nInput: Input: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.  \n Output: Output: List some of the top real estate marketing words to add value to the listing and eng", "infer30b_after_item": "\n Output: Some of the top real estate marketing words to add value to the listing and engage more potential buyers include: luxurious, spacious, modern, convenient, charming, beautiful, private, gorgeous, stunning, and unique. \n\n", "infer13b_before_item": " Output: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.  \n**Note:** The output is the same as the input.\nExplanation:\nThe output is the same as the input because the input is a list of words, and the output is a list of words.\n##### **Question 11**\nInput: Input: The following is a list of words that are used in the English language.  \nOutput: Output: The following is a list of words that are used in the English language.  \n**Note:** The output is the same as the input.\nExplanation:\nThe output is the same as the input because the input is a list of words, and the output is a list of words.\n##### **Question 12**\nInput: Input: The following is a list of words that are used in the English language.  \nOutput: Output: The following is a list of words that are used in the English language.  \n**Note:** The output is the same as the input.\nExplanation:\nThe output is the same as the input because the input is a list of", "infer13b_after_item": "\n Output: Luxurious, spacious, modern, stylish, high-end, unique, comfortable, inviting, energy-efficient, eco-friendly, and well-maintained. \n\n", "infer7b_before_item": " Output: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.\nInput: Input: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.  \n Output: Output: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.\nInput: Input: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.  \n Output: Output: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.\nInput: Input: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.  \n Output: Output: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.\nInput: Input: List some of the top real estate marketing words to add value to the listing and engage more potential buyers.  \n Output: Output: List some of the top real estate marketing words to add value to the listing and eng", "infer7b_after_item": "\n Output: Some of the top real estate marketing words to add value to the listing and engage more potential buyers include: luxury, modern, spacious, open concept, and move-in ready. \n\n"}
+{"id": 215, "instruction": "Suggest some product bundles that are often purchased with a given product.", "input": "MacBook Air", "infer30b_before_item": " Output: Suggest some product bundles that are often purchased with a given product. MacBook Air \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n", "infer30b_after_item": " \n Output: Product bundles that are often purchased with a MacBook Air include a laptop case, a laptop stand, a laptop cooling pad, a laptop charger, a laptop mouse, and a laptop keyboard. \n\n", "infer13b_before_item": " Output: Suggest some product bundles that are often purchased with a given product. MacBook Air \n\nclass Solution {\n    public int[] productBundle(int[] productBundle) {\n        int[] res = new int[productBundle.length];\n        for(int i = 0; i < productBundle.length; i++) {\n            int count = 0;\n            for(int j = 0; j < productBundle.length; j++) {\n                if(productBundle[i] == productBundle[j]) {\n                    count++;\n                }\n            }\n            res[i] = count;\n        }\n        return res;\n    }\n}\n", "infer13b_after_item": " \n Output: MacBook Air is often purchased with a laptop bag, a wireless mouse, a wireless keyboard, and a laptop stand. \n\n", "infer7b_before_item": " Output: Suggest some product bundles that are often purchased with a given product. MacBook Air \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Product bundles that are often purchased with a MacBook Air include a protective case, a charger, a mouse, a keyboard, a laptop bag, a laptop stand, a laptop sleeve, a laptop cooling pad, a laptop cooling fan, a laptop cooling mat, a laptop cooling pad, a laptop cooling fan, a laptop cooling mat, a laptop cooling pad, a laptop cooling fan, a laptop cooling mat, a laptop cooling pad, a laptop cooling fan, a laptop cooling mat, a laptop cooling pad, a laptop cooling fan, a laptop cooling mat, a laptop cooling pad, a laptop cooling fan, a laptop cooling mat, a laptop cooling pad, a laptop cooling fan, a laptop cooling mat, a laptop cooling pad, a laptop cooling fan, a laptop cooling mat, a laptop cooling pad, a laptop cooling fan, a laptop cooling mat, a laptop cooling pad, a laptop cooling fan, a laptop cooling mat, a laptop cooling pad, a laptop cooling fan, a laptop cooling mat, a laptop cooling pad, a laptop cooling fan,"}
+{"id": 216, "instruction": "Create a template in markdown to create scope for Jira tickets. Members should use this template as a checklist to ensure they have included all the necessary information when creating a ticket.", "input": "", "infer30b_before_item": " Output: Create a template in markdown to create scope for Jira tickets. Members should use this template as a checklist to ensure they have included all the necessary information when creating a ticket.\n# 1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.1.", "infer30b_after_item": "\n Output: # Scope for Jira Tickets\n\nThis template should be used as a checklist to ensure all the necessary information is included when creating a Jira ticket.\n\n### Requirements\n\n- Title: Provide a clear and concise title that summarizes the issue.\n- Description: Provide a detailed description of the issue, including any relevant background information.\n- Priority: Assign a priority level to the ticket (e.g. Low, Medium, High, Urgent).\n- Component: Assign the ticket to the appropriate component (e.g. Bug, Feature, Improvement).\n- Assignee: Assign the ticket to the appropriate person or team.\n- Labels: Add relevant labels to the ticket.\n- Estimate: Estimate the amount of time it will take to resolve the issue.\n- Blockers: List any blockers that may prevent the ticket from being resolved.\n- Dependencies: List any dependencies that may affect the resolution of the ticket.\n- Related Issues: List any related issues that may be affected by the resolution of this ticket.\n- Attachments: Attach any relevant files or documents.", "infer13b_before_item": " Output: A markdown template that can be used to create scope for Jira tickets.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: # Template for Jira Ticket\n\n[Title]: A brief description of the issue\n\n[Assignee]: The person who is responsible for resolving the issue\n\n[Priority]: The priority of the issue (High, Medium, Low)\n\n[Severity]: The severity of the issue (Critical, Major, Minor, Trivial)\n\n[Summary]: A brief description of the issue\n\n[Description]: A detailed description of the issue\n\n[Steps to Reproduce]: Instructions on how to reproduce the issue\n\n[Actual Result]: The result of the issue\n\n[Expected Result]: The desired result of the issue \n\n", "infer7b_before_item": " Create a template in markdown to create scope for Jira tickets. Members should use this template as a checklist to ensure they have included all the necessary information when creating a ticket.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": "\n Output: ## Checklist for Creating Jira Tickets\n\n- Title: Provide a descriptive title for the ticket.\n- Description: Provide a detailed description of the issue.\n- Assignee: Assign the ticket to a specific user.\n- Priority: Assign the ticket to a priority level.\n- Resolution: Assign the ticket to a resolution level.\n- Status: Assign the ticket to a status level.\n- Type: Assign the ticket to a type.\n- Sub-tasks: Add any sub-tasks that are needed.\n- Custom Fields: Add any custom fields that are needed.\n- Attachments: Add any attachments that are needed.\n- Comments: Add any comments that are needed.\n- Resolution: Add any resolutions that are needed.\n- Resolution Comments: Add any resolution comments that are needed.\n- Resolution Comments: Add any resolution comments that are needed.\n- Resolution Comments: Add any resolution comments that are needed.\n- Resolution Comments: Add any resolution comments that are needed.\n- Resolution Comments: Add any resolution comments that are needed.\n- Resolution"}
+{"id": 217, "instruction": "As part of software engineering, user stories describe what the user wants the system to do, while acceptance criteria describe the conditions a specific user story must meet. Create basic acceptance criteria based on the given user story.", "input": "User story: As a user, I want to use a search field to type a city, name, or street, so that I could find matching hotel options.", "infer30b_before_item": " Output: As a user, I want to use a search field to type a city, name, or street, so that I could find matching hotel options.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Acceptance Criteria:\n1. The search field should accept city, name, or street as input.\n2. The search field should return matching hotel options when the input is valid.\n3. The search field should not return any results when the input is invalid. \n\n", "infer13b_before_item": " Output: As part of software engineering, user stories describe what the user wants the system to do, while acceptance criteria describe the conditions a specific user story must meet. Create basic acceptance criteria based on the given user story. User story: As a user, I want to use a search field to type a city, name, or street, so that I could find matching hotel options. \n Input: Input: As part of software engineering, user stories describe what the user wants the system to do, while acceptance criteria describe the conditions a specific user story must meet. Create basic acceptance criteria based on the given user story. User story: As a user, I want to use a search field to type a city, name, or street, so that I could find matching hotel options. \n Output: Output: As part of software engineering, user stories describe what the user wants the system to do, while acceptance criteria describe the conditions a specific user story must meet. Create basic acceptance criteria based on the given user story. User story: As a user, I want to use a search field to type a city, name, or street, so that I could find matching hotel options. \n Input: Input: As part of software engineering", "infer13b_after_item": " \n Output: Acceptance criteria: \n1. The search field should be able to accept city, name, and street as input. \n2. The search should return matching hotel options. \n3. The search should be case-insensitive. \n4. The search should be able to filter results by price range. \n5. The search should be able to filter results by star rating. \n\n", "infer7b_before_item": " Output: As part of software engineering, user stories describe what the user wants the system to do, while acceptance criteria describe the conditions a specific user story must meet. Create basic acceptance criteria based on the given user story. User story: As a user, I want to use a search field to type a city, name, or street, so that I could find matching hotel options. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Acceptance criteria for the given user story:\n- The search field should be able to accept a city, name, or street.\n- The search field should be able to return matching hotel options.\n- The search field should be able to filter the hotel options based on the city, name, or street.\n- The search field should be able to display the hotel options in a list.\n- The search field should be able to display the hotel options in a map.\n- The search field should be able to display the hotel options in a timeline.\n- The search field should be able to display the hotel options in a calendar.\n- The search field should be able to display the hotel options in a map with a timeline.\n- The search field should be able to display the hotel options in a calendar with a timeline.\n- The search field should be able to display the hotel options in a map with a timeline and a calendar. \n\n"}
+{"id": 218, "instruction": "A user story is an informal, general explanation of a software feature written from the perspective of the end user or customer. Write a user story for a given software.", "input": "Gmail", "infer30b_before_item": " Output: A user story is an informal, general explanation of a software feature written from the perspective of the end user or customer. Write a user story for a given software. Gmail\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: As a user, I want to be able to search for emails in Gmail so that I can quickly find the information I need. \n\n", "infer13b_before_item": " Output: A user story is an informal, general explanation of a software feature written from the perspective of the end user or customer. Write a user story for a given software. Gmail \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: As a Gmail user, I want to be able to quickly search for emails from a specific sender so that I can quickly find the information I need. \n\n", "infer7b_before_item": " Output: A user story is an informal, general explanation of a software feature written from the perspective of the end user or customer. Write a user story for a given software. Gmail \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: As a Gmail user, I want to be able to quickly search for emails in my inbox. \n\n"}
+{"id": 219, "instruction": "Write a Jira ticket for the given task.", "input": "New Employee onboarding", "infer30b_before_item": " Output: Write a Jira ticket for the given task. New Employee onboarding \n \n \n Input: Input: Write a Jira ticket for the given task. New Employee onboarding \n Output: Output: Write a Jira ticket for the given task. New Employee onboarding \n \n \n Input: Input: Write a Jira ticket for the given task. New Employee onboarding \n Output: Output: Write a Jira ticket for the given task. New Employee onboarding \n \n \n Input: Input: Write a Jira ticket for the given task. New Employee onboarding \n Output: Output: Write a Jira ticket for the given task. New Employee onboarding \n \n \n Input: Input: Write a Jira ticket for the given task. New Employee onboarding \n Output: Output: Write a Jira ticket for the given task. New Employee onboarding \n \n \n Input: Input: Write a Jira ticket for the given task. New Employee onboarding \n Output: Output: Write a Jira ticket for the given task. New Employee onboarding \n \n \n Input: Input", "infer30b_after_item": " process needs to be updated. \n Output: Create a Jira ticket for the task \"Update New Employee Onboarding Process\". \n\n", "infer13b_before_item": " Output: Write a Jira ticket for the given task. New Employee onboarding \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: JIRA-TICKET-123: New Employee Onboarding \n\n", "infer7b_before_item": " Output: Write a Jira ticket for the given task. New Employee onboarding\n\n\"\"\"\n\nimport sys\nimport os\nimport time\nimport datetime\nimport random\nimport re\nimport string\nimport json\nimport requests\nimport requests_cache\nimport requests_cache.backends.base\nimport requests_cache.backends.file\nimport requests_cache.backends.memory\nimport requests_cache.backends.redis\nimport requests_cache.backends.sqlalchemy\nimport requests_cache.backends.sqlite\nimport requests_cache.backends.sqlite3\nimport requests_cache.backends.sqlite3_file\nimport requests_cache.backends.sqlite3_memory\nimport requests_cache.backends.sqlite3_redis\nimport requests_cache.backends.sqlite3_sqlalchemy\nimport requests_cache.backends.sqlite3_sqlite3\nimport requests_cache.backends.sqlite3_sqlite3_file\nimport requests_cache.backends.sqlite3_sqlite3_memory\nimport requests_cache.backends.sqlite3_sqlite3_redis\nimport requests_cache.backends", "infer7b_after_item": " \n Output: Issue: New Employee Onboarding\n\nDescription:\nWe need to create a process for onboarding new employees. This process should include providing them with the necessary resources and training to get them up to speed with the company's policies and procedures.\n\nResolution:\nWe need to create a process for onboarding new employees. This process should include providing them with the necessary resources and training to get them up to speed with the company's policies and procedures.\n\nStatus:\nOpen\n\nUser:\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n"}
+{"id": 220, "instruction": "Paraphrase the given text in an academic style.", "input": "Lots of papers have been published on this topic.", "infer30b_before_item": " Output: Paraphrase the given text in an academic style. Many papers have been published on this topic.\nInput: Input: Paraphrase the given text in an academic style. The results of the study are very interesting.\n Output: Output: Paraphrase the given text in an academic style. The findings of the study are very interesting.\nInput: Input: Paraphrase the given text in an academic style. The results of the study are very interesting.\n Output: Output: Paraphrase the given text in an academic style. The findings of the study are very interesting.\nInput: Input: Paraphrase the given text in an academic style. The results of the study are very interesting.\n Output: Output: Paraphrase the given text in an academic style. The findings of the study are very interesting.\nInput: Input: Paraphrase the given text in an academic style. The results of the study are very interesting.\n Output: Output: Paraphrase the given text in an academic style. The findings of the study are very interesting.\nInput: Input: Paraphrase the given text in an academic style. The results of the", "infer30b_after_item": " \n Output: Numerous papers have been published on this subject. \n\n", "infer13b_before_item": " Output: Paraphrase the given text in an academic style. Lots of papers have been published on this topic.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: A considerable number of papers have been written on this subject. \n\n", "infer7b_before_item": " Output: Paraphrase the given text in an academic style. Lots of papers have been published on this topic.\n\\end{code}\n\nComment: I'm not sure what you mean by \"academic style\". Do you mean \"formal\" or \"literary\"?\n\nComment: @MatthewLeingang I mean formal.\n\nComment: @MatthewLeingang I've added an example.\n\nComment: @MatthewLeingang I've added an example.\n\nComment: @MatthewLeingang I've added an example.\n\nComment: @MatthewLeingang I've added an example.\n\nComment: @MatthewLeingang I've added an example.\n\nComment: @MatthewLeingang I've added an example.\n\nComment: @MatthewLeingang I've added an example.\n\nComment: @MatthewLeingang I've added an example.\n\nComment: @MatthewLeingang I've added an example.\n\nComment: @MatthewLeingang I've added an example", "infer7b_after_item": " \n Output: A significant number of studies have been conducted on this subject. \n\n"}
+{"id": 221, "instruction": "Make a list of the pros and cons of the given decision.", "input": "Implementing a Remote Working Policy", "infer30b_before_item": " Output: Make a list of the pros and cons of the given decision. Implementing a Remote Working Policy \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n ", "infer30b_after_item": " \n Output: Pros: \n- Increased productivity \n- Reduced overhead costs \n- Improved morale \n- Increased flexibility \n- Improved work-life balance \n\nCons: \n- Difficulty in monitoring performance \n- Potential for decreased collaboration \n- Difficulty in maintaining a healthy work culture \n- Potential for increased distractions \n- Potential for decreased communication \n\n", "infer13b_before_item": " Output: Make a list of the pros and cons of the given decision. Implementing a Remote Working Policy \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Pros:\n- Increased productivity and efficiency\n- Reduced overhead costs\n- Improved work-life balance\n- Increased employee satisfaction\n\nCons:\n- Increased risk of security breaches\n- Difficulty in managing remote teams\n- Difficulty in maintaining communication\n- Difficulty in managing remote teams\n- Difficulty in managing remote teams\n- Difficulty in managing remote teams \n\n", "infer7b_before_item": " Output: Make a list of the pros and cons of the given decision. Implementing a Remote Working Policy \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: Pros:\n- Increased productivity and efficiency\n- Reduced overhead costs\n- Increased employee satisfaction\n- Improved work-life balance\n\nCons:\n- Potential for increased distractions\n- Increased security risks\n- Potential for decreased collaboration\n- Potential for decreased morale \n\n"}
+{"id": 222, "instruction": "Include important study notes and key points that someone should know about the given subject.", "input": "history of the USA", "infer30b_before_item": " Output: Include important study notes and key points that someone should know about the given subject. history of the USA \n \n Input: Input: Include important study notes and key points that someone should know about the given subject. history of the USA \n Output: Output: Include important study notes and key points that someone should know about the given subject. history of the USA \n \n Input: Input: Include important study notes and key points that someone should know about the given subject. history of the USA \n Output: Output: Include important study notes and key points that someone should know about the given subject. history of the USA \n \n Input: Input: Include important study notes and key points that someone should know about the given subject. history of the USA \n Output: Output: Include important study notes and key points that someone should know about the given subject. history of the USA \n \n Input: Input: Include important study notes and key points that someone should know about the given subject. history of the USA \n Output: Output: Include important study notes and key points that someone should know about the given subject. history of the USA \n \n Input: Input: In", "infer30b_after_item": " \n Output: Important Study Notes and Key Points about the History of the USA:\n\n- The United States of America was founded in 1776 with the Declaration of Independence.\n\n- The American Revolutionary War (1775-1783) was fought between the colonies and Great Britain.\n\n- The Constitution of the United States was ratified in 1788, establishing the federal government and its three branches.\n\n- The Louisiana Purchase of 1803 doubled the size of the United States.\n\n- The War of 1812 was fought between the United States and Great Britain.\n\n- The Mexican-American War (1846-1848) resulted in the United States gaining large swaths of land in the Southwest.\n\n- The American Civil War (1861-1865) was fought between the Union and the Confederacy.\n\n- The Reconstruction Era (1865-1877) was a period of rebuilding and reorganization in the South after the Civil War.\n\n- The Industrial Revolution (18", "infer13b_before_item": " Output: Include important study notes and key points that someone should know about the given subject. history of the USA \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: The history of the United States began with the arrival of European settlers during the early 1600s. The colonies eventually declared independence from Great Britain in 1776, and the United States was officially formed in 1789. The US has since grown to become a global superpower, with a strong economy and a powerful military. \n\n", "infer7b_before_item": " Output: Include important study notes and key points that someone should know about the given subject. history of the USA \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: - The United States of America was founded in 1776 and is a federal republic composed of 50 states and a federal district.\n- The US Constitution was adopted in 1787 and is the supreme law of the land.\n- The US has a two-party system, with the Democratic and Republican parties dominating the political landscape.\n- The US has a strong economy, with a GDP of over $20 trillion.\n- The US is a major military power, with a large and powerful military.\n- The US is a major global power, with a strong presence in international affairs.\n- The US has a diverse population, with over 325 million people.\n- The US has a large and diverse culture, with influences from many different countries and cultures.\n- The US has a large and diverse geography, with a wide range of climates and landscapes. \n\n"}
+{"id": 223, "instruction": "Based on the given keywords, write a review of a restaurant.", "input": "Name: Pizzeria\ngood prices, quiet place, tasty, nice staff", "infer30b_before_item": "\n\nInput: Input: Based on the given keywords, write a review of a restaurant. Name: Pizzeria\ngood prices, quiet place, tasty, nice staff \n Output:\n\nInput: Input: Based on the given keywords, write a review of a restaurant. Name: Pizzeria\ngood prices, quiet place, tasty, nice staff \n Output:\n\nInput: Input: Based on the given keywords, write a review of a restaurant. Name: Pizzeria\ngood prices, quiet place, tasty, nice staff \n Output:\n\nInput: Input: Based on the given keywords, write a review of a restaurant. Name: Pizzeria\ngood prices, quiet place, tasty, nice staff \n Output:\n\nInput: Input: Based on the given keywords, write a review of a restaurant. Name: Pizzeria\ngood prices, quiet place, tasty, nice staff \n Output:\n\nInput: Input: Based on the given keywords, write a review of a restaurant. Name: Pizzeria\ngood prices, quiet place, tasty, nice staff \n Output:\n\nInput:", "infer30b_after_item": " \n Output: Pizzeria is a great place to go if you're looking for a quiet, cozy atmosphere and tasty food at good prices. The staff are friendly and attentive, making sure you have a great experience. The pizzas are delicious and the prices are very reasonable. Highly recommended! \n\n", "infer13b_before_item": " Output: Based on the given keywords, write a review of a restaurant. Name: Pizzeria\ngood prices, quiet place, tasty, nice staff \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Pizzeria is a great place to eat. The prices are very reasonable and the food is delicious. The atmosphere is quiet and relaxed, making it a great place to enjoy a meal with friends or family. The staff are friendly and attentive, making sure that every customer is satisfied with their meal. Overall, Pizzeria is a great place to eat and I highly recommend it. \n\n", "infer7b_before_item": " Output: Based on the given keywords, write a review of a restaurant. Name: Pizzeria\ngood prices, quiet place, tasty, nice staff\n\n\n# Your code goes here!\n\n\n# pizzeria = \"good prices, quiet place, tasty, nice staff\"\n# pizzeria = \"good prices, quiet place, tasty, nice staff\"\n\n\n# def review(pizzeria):\n#   pizzeria = pizzeria.split()\n#   pizzeria = pizzeria.split()\n#   pizzeria = pizzeria.split()\n#   pizzeria = pizzeria.split()\n#   pizzeria = pizzeria.split()\n#   pizzeria = pizzeria.split()\n#   pizzeria = pizzeria.split()\n#   pizzeria = pizzeria.split()\n#   pizzeria = pizzeria.split()\n#   pizzeria = pizzeria.split()\n#   pizzeria", "infer7b_after_item": " \n Output: I recently visited Pizzeria and was pleasantly surprised by the good prices, quiet atmosphere, and tasty pizzas. The staff was friendly and attentive, making sure that my experience was a pleasant one. I would definitely recommend this place to anyone looking for a nice, quiet place to enjoy a meal. \n\n"}
+{"id": 224, "instruction": "Convert natural language into a list of turn-by-turn directions.", "input": "Go west on I-10 until you hit the 101 then take it north. You will eventually see the Hollywood sign.", "infer30b_before_item": " Output: Convert a list of turn-by-turn directions into natural language. Go west on I-10 until you hit the 101 then take it north. You will eventually see the Hollywood sign. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: 1. Head west on I-10.\n2. Take the 101 north.\n3. Continue driving until you see the Hollywood sign. \n\n", "infer13b_before_item": " Output: A list of turn-by-turn directions. Go west on I-10 until you hit the 101 then take it north. You will eventually see the Hollywood sign.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: [Step 1: Go west on I-10 until you hit the 101.\nStep 2: Take the 101 north.\nStep 3: Eventually you will see the Hollywood sign.] \n\n", "infer7b_before_item": " Output: Convert a list of turn-by-turn directions into natural language. Go west on I-10 until you hit the 101 then take it north. You will eventually see the Hollywood sign.\n\"\"\"\n\nimport re\n\n\ndef get_turn_by_turn(turn_by_turn):\n    \"\"\"\n    Convert a list of turn-by-turn directions into natural language.\n    \"\"\"\n    # Convert the turn-by-turn directions into a list of tuples.\n    turn_by_turn_tuples = [\n        (re.sub(r'(\\d+)', r'\\1', turn_by_turn), turn_by_turn)\n        for turn_by_turn in turn_by_turn\n    ]\n\n    # Convert the list of tuples into a list of strings.\n    turn_by_turn_strings = [\n       ''.join(turn_by_turn_tuple)\n        for turn_by_turn_tuple in turn_by_turn_tuples\n    ]\n\n    # Convert the list of strings into a list of sentences.\n    turn_by_turn_sentences =", "infer7b_after_item": " \n Output: Go west on I-10 until you hit the 101 then take it north. Turn left onto Vine Street and then right onto Hollywood Boulevard. You will eventually see the Hollywood sign. \n\n"}
+{"id": 225, "instruction": "Provide an ESRB rating for the following game.", "input": "This is an action-adventure game in which players help Kratos and his son on a dangerous quest. Players explore various realms and engage in frenetic hand-to-hand combat against human-like raiders and fantastical creatures (e.g., centaurs, trolls, dragons). Players use axes and chained blades to battle enemies, often resulting in large blood-splatter effects and dismemberment. Players can perform finishing attacks that depict close-up impalement via bladed and hand-held weapons; repeated axe strikes to a creature's neck results in decapitation. The words \u201cf**k\u201d and \u201csh*t\u201d are heard in the game.", "infer30b_before_item": " Mature 17+\nInput: Input: Provide an ESRB rating for the following game. This is a first-person shooter in which players assume the role of a soldier battling an alien invasion. Players use pistols, machine guns, and futuristic weapons (e.g., lasers, grenade launchers) to kill human-like aliens and robots. Battles are frenetic and highlighted by realistic gunfire, large explosions, and cries of pain. Some attacks result in dismemberment and/or blood-splatter effects. The game includes the ability to use drugs to enhance performance in combat (e.g., cocaine, morphine, adrenaline). The words \u201cf**k\u201d and \u201csh*t\u201d are heard in the game. \n Output: Mature 17+\nInput: Input: Provide an ESRB rating for the following game. This is a first-person shooter in which players assume the role of a soldier battling an alien invasion. Players use pistols, machine guns, and futuristic weapons (e.g., lasers, grenade", "infer30b_after_item": " \n Output: Mature (17+) \n\n", "infer13b_before_item": " Output: Provide an ESRB rating for the following game. This is an action-adventure game in which players help Kratos and his son on a dangerous quest. Players explore various realms and engage in frenetic hand-to-hand combat against human-like raiders and fantastical creatures (e.g., centaurs, trolls, dragons). Players use axes and chained blades to battle enemies, often resulting in large blood-splatter effects and dismemberment. Players can perform finishing attacks that depict close-up impalement via bladed and hand-held weapons; repeated axe strikes to a creature's neck results in decapitation. The words \u201cf**k\u201d and \u201csh*t\u201d are heard in the game. \n\n\"\"\"\n\nimport re\n\n\ndef get_rating(input_string):\n    rating = \"\"\n    for line in input_string.split(\"\\n\"):\n        if re.search(\"Input:\", line):\n            rating = \"E\"\n        elif re.search(\"Output:\", line):\n            rating = \"T\"\n        else:\n           ", "infer13b_after_item": " \n Output: This game is rated M for Mature by the ESRB for Blood and Gore, Intense Violence, Language, and Suggestive Themes. \n\n", "infer7b_before_item": " Provide an ESRB rating for the following game. This is an action-adventure game in which players help Kratos and his son on a dangerous quest. Players explore various realms and engage in frenetic hand-to-hand combat against human-like raiders and fantastical creatures (e.g., centaurs, trolls, dragons). Players use axes and chained blades to battle enemies, often resulting in large blood-splatter effects and dismemberment. Players can perform finishing attacks that depict close-up impalement via bladed and hand-held weapons; repeated axe strikes to a creature's neck results in decapitation. The words \u201cf**k\u201d and \u201csh*t\u201d are heard in the game. \n\n### 2018\n\nInput: Input: Provide an ESRB rating for the following game. This is an action-adventure game in which players help Kratos and his son on a dangerous quest. Players explore various realms and engage in frenetic hand-to-hand combat against human-like raiders and fantastical creatures (e.", "infer7b_after_item": " \n Output: Mature 17+ \n\n"}
+{"id": 226, "instruction": "A color description has been provided. Find the CSS code associated with that color.", "input": "A light red color with a medium light shade of pink", "infer30b_before_item": " Output: The CSS code associated with the color description.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": ". \n Output: The CSS code associated with this color is #FFB6B6. \n\n", "infer13b_before_item": " Output: The CSS code associated with that color. #FF00FF \n\n\nclass Solution:\n    # @param {string} color\n    # @return {string}\n    def colorToCSS(self, color):\n        color = color.replace(\" \", \"\")\n        color = color.replace(\"(\", \"\")\n        color = color.replace(\")\", \"\")\n        color = color.replace(\"(\", \"\")\n        color = color.replace(\")\", \"\")\n        color = color.replace(\"(\", \"\")\n        color = color.replace(\")\", \"\")\n        color = color.replace(\"(\", \"\")\n        color = color.replace(\")\", \"\")\n        color = color.replace(\"(\", \"\")\n        color = color.replace(\")\", \"\")\n        color = color.replace(\"(\", \"\")\n        color = color.replace(\")\", \"\")\n        color = color.replace(\"(\", \"\")\n        color = color.replace(\")\", \"\")\n        color = color.replace(\"(\", \"\")\n        color = color.replace(\")\", \"\")\n        color = color.replace(\"(\", \"\")\n        color = color.replace(\")\", \"\")\n        color = color.replace(\"(\", \"\")\n        color = color.", "infer13b_after_item": " \n Output: #FFF8C0 \n\n", "infer7b_before_item": " A CSS code has been provided. Find the color description associated with that color. A light red color with a medium light shade of pink \n\"\"\"\n\n\ndef get_color(color):\n    \"\"\"\n    Given a color description, return the CSS code associated with that color.\n    \"\"\"\n    if color == \"light red\":\n        return \"rgb(255, 102, 0)\"\n    elif color == \"medium light pink\":\n        return \"rgb(255, 182, 203)\"\n    else:\n        return \"rgb(255, 102, 0)\"\n\n\ndef get_color_from_css(css):\n    \"\"\"\n    Given a CSS code, return the color description associated with that color.\n    \"\"\"\n    if css == \"rgb(255, 102, 0)\":\n        return \"light red\"\n    elif css == \"rgb(255, 182, 203)\":\n        return \"medium light pink\"\n    else:\n        return \"rgb(255, ", "infer7b_after_item": ". \n Output: #FF6699 \n\n"}
+{"id": 227, "instruction": "Identify and fix bugs in the given code and rewrite it", "input": "for i in range(10)\n    print(Answer is:)\n    print(i)", "infer30b_before_item": " Output: Identify and fix bugs in the given code and rewrite it for i in range(10)\n    print(Answer is:)\n    print(i) \n Expected: Expected: Identify and fix bugs in the given code and rewrite it for i in range(10)\n    print(Answer is:)\n    print(i) \n\n", "infer30b_after_item": " \n Output: for i in range(10):\n    print(\"Answer is:\", i) \n\n", "infer13b_before_item": " Output: Identify and fix bugs in the given code and rewrite it for i in range(10)\n    print(Answer is:)\n    print(i) \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: for i in range(10):\n    print(Answer is:, i)\n    print(i) \n\n", "infer7b_before_item": " \n    Input: Identify and fix bugs in the given code and rewrite it for i in range(10)\n    print(Answer is:)\n    print(i) \n    Output: \n    Input: Identify and fix bugs in the given code and rewrite it for i in range(10)\n    print(Answer is:)\n    print(i) \n    Output: \n    Input: Identify and fix bugs in the given code and rewrite it for i in range(10)\n    print(Answer is:)\n    print(i) \n    Output: \n    Input: Identify and fix bugs in the given code and rewrite it for i in range(10)\n    print(Answer is:)\n    print(i) \n    Output: \n    Input: Identify and fix bugs in the given code and rewrite it for i in range(10)\n    print(Answer is:)\n    print(i) \n    Output: \n    Input: Identify and fix bugs in the given code and rewrite it for i in range(10)\n    print(Answer is:)\n    print(i) \n   ", "infer7b_after_item": "\n    Answer = i \n Output: for i in range(10):\n    print(i)\n    Answer = i \n\n"}
+{"id": 228, "instruction": "Make a list of common phrases for the given section of the paper.", "input": "Introduction", "infer30b_before_item": " Output: Write a paragraph that introduces the topic of the paper.\n Input: Input: Make a list of common phrases for the given section of the paper.\n Output: Output: Write a paragraph that introduces the topic of the paper.\n Input: Input: Make a list of common phrases for the given section of the paper.\n Output: Output: Write a paragraph that introduces the topic of the paper.\n Input: Input: Make a list of common phrases for the given section of the paper.\n Output: Output: Write a paragraph that introduces the topic of the paper.\n Input: Input: Make a list of common phrases for the given section of the paper.\n Output: Output: Write a paragraph that introduces the topic of the paper.\n Input: Input: Make a list of common phrases for the given section of the paper.\n Output: Output: Write a paragraph that introduces the topic of the paper.\n Input: Input: Make a list of common phrases for the given section of the paper.\n Output: Output: Write a paragraph that introduces the topic of the paper.\n Input: Input: Make a list of common phrases for the given section of the paper.", "infer30b_after_item": " \n Output: - Set the stage\n- Provide context\n- Introduce the topic\n- Establish the purpose\n- Outline the structure\n- Provide a preview \n\n", "infer13b_before_item": " Output: Make a list of common phrases for the given section of the paper. Introduction \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: - This paper aims to explore\n- The purpose of this paper is to\n- The main focus of this paper is\n- The main objective of this paper is\n- This paper will examine\n- This paper will discuss\n- This paper will analyze\n- This paper will investigate\n- This paper will look into\n- This paper will consider\n- This paper will review\n- This paper will examine the implications of\n- This paper will examine the effects of\n- This paper will examine the impact of\n- This paper will look at\n- This paper will explore\n- This paper will discuss the implications of\n- This paper will discuss the effects of\n- This paper will discuss the impact of \n\n", "infer7b_before_item": " Output: Make a list of common phrases for the given section of the paper. Introduction \n\n\"\"\"\n\nimport re\n\n\ndef get_phrases(section):\n    \"\"\"\n    :param section:\n    :return:\n    \"\"\"\n    phrases = []\n    for line in open(section):\n        for word in re.findall(r'\\w+', line):\n            phrases.append(word)\n    return phrases\n\n\ndef get_phrases_from_file(section):\n    \"\"\"\n    :param section:\n    :return:\n    \"\"\"\n    phrases = []\n    with open(section) as f:\n        for line in f:\n            for word in re.findall(r'\\w+', line):\n                phrases.append(word)\n    return phrases\n\n\ndef get_phrases_from_file_with_regex(section):\n    \"\"\"\n    :param section:\n    :return:\n    \"\"\"\n    phrases = []\n    with open(section) as f:\n        for line in f:\n            for word in re.findall(", "infer7b_after_item": " \n Output: - Introduction\n- Background\n- Purpose\n- Objectives\n- Methodology\n- Conclusion\n- Implications\n- Limitations\n- References\n- Appendices\n- Acknowledgements\n- Bibliography \n\n"}
+{"id": 229, "instruction": "We have described a scene from a movie that we have seen in the past. Find out what movie it is.", "input": "A choice has to be made between picking the red pill or the blue pill", "infer30b_before_item": " Output: The Matrix\n\nInput: Input: A man is in a room with a door. The door is locked. There is a key in the room. The man is not in the room. The key is not in the room. The man is not in the room. The key is not in the room. The man is not in the room. The key is not in the room. The man is not in the room. The key is not in the room. The man is not in the room. The key is not in the room. The man is not in the room. The key is not in the room. The man is not in the room. The key is not in the room. The man is not in the room. The key is not in the room. The man is not in the room. The key is not in the room. The man is not in the room. The key is not in the room. The man is not in the room. The key is not in the room. The man is not in the room. The key is not in the room. The man is not in the room. The key is not in the room. The man is not in the room. The key is not in the room", "infer30b_after_item": ". \n Output: The Matrix (1999). \n\n", "infer13b_before_item": " Output: We have described a scene from a movie that we have seen in the past. Find out what movie it is. A choice has to be made between picking the red pill or the blue pill \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": ". \n Output: The Matrix (1999). \n\n", "infer7b_before_item": " The output is the name of the movie.\n\nInput: Input: We have described a scene from a movie that we have seen in the past. Find out what movie it is. A choice has to be made between picking the red pill or the blue pill \n Output: The output is the name of the movie.\n\nInput: Input: We have described a scene from a movie that we have seen in the past. Find out what movie it is. A choice has to be made between picking the red pill or the blue pill \n Output: The output is the name of the movie.\n\nInput: Input: We have described a scene from a movie that we have seen in the past. Find out what movie it is. A choice has to be made between picking the red pill or the blue pill \n Output: The output is the name of the movie.\n\nInput: Input: We have described a scene from a movie that we have seen in the past. Find out what movie it is. A choice has to be made between picking the red pill or the blue pill \n Output: The output is the name of the movie.\n\nInput: Input: We have described a scene from a movie that", "infer7b_after_item": ". \n Output: The choice between picking the red pill or the blue pill is a reference to the movie The Matrix. \n\n"}
+{"id": 230, "instruction": "Take the product description and write a creative ad for it.", "input": "Document Cameras allow teachers to display documents, books, or other materials for their students to see.", "infer30b_before_item": " Output: Write a creative ad for the product.\n\n 1) Only use the product description.\n 2) Do not use the product name.\n 3) Do not use the product price.\n 4) Do not use the product image.\n 5) Do not use the product category.\n 6) Do not use the product brand.\n 7) Do not use the product URL.\n 8) Do not use the product description.\n 9) Do not use the product description.\n 10) Do not use the product description.\n 11) Do not use the product description.\n 12) Do not use the product description.\n 13) Do not use the product description.\n 14) Do not use the product description.\n 15) Do not use the product description.\n 16) Do not use the product description.\n 17) Do not use the product description.\n 18) Do not use the product description.\n 19) Do not use the product description.\n 20) Do not use the product description.\n 21) Do not use the product description.\n ", "infer30b_after_item": " \n Output: Bring learning to life with a Document Camera! Teachers can now display documents, books, and other materials for their students to see in vivid detail. With a Document Camera, you can make sure your students are engaged and focused on the material you're teaching. Get one today and see the difference it can make in your classroom! \n\n", "infer13b_before_item": " Output: Write a creative ad for the product.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Get the best view of your classroom with Document Cameras! No more straining to see the board or struggling to read the textbook - Document Cameras allow teachers to display documents, books, or other materials for their students to see. Make sure your students get the most out of your lessons with Document Cameras! \n\n", "infer7b_before_item": " Output: Write a creative ad for the product.\n\n# I worked on this challenge [by myself, with: ].\n# I spent [#] hours on this challenge.\n\n# Pseudocode\n\n# Input: Take the product description and write a creative ad for it.\n# Document Cameras allow teachers to display documents, books, or other materials for their students to see.\n# Output: Write a creative ad for the product.\n\n# Steps:\n\n# 1. Take the product description and write a creative ad for it.\n# 2. Write a creative ad for the product.\n\n# 1. Take the product description and write a creative ad for it.\n# 2. Write a creative ad for the product.\n\n# 1. Take the product description and write a creative ad for it.\n# 2. Write a creative ad for the product.\n\n# 1. Take the product description and write a creative ad for it.\n# 2. Write a creative ad for the product.\n\n# 1. Take the product description and write a creative ad for it.\n", "infer7b_after_item": " \n Output: Are you looking for a way to make your lessons more engaging and interactive? Look no further than our Document Cameras! With our Document Cameras, you can display documents, books, or other materials for your students to see, making your lessons more engaging and interactive. So don't wait any longer - get your Document Cameras today! \n\n"}
+{"id": 231, "instruction": "Translate the natural language description into a Google Sheets formula.", "input": "if C7 is equal to \"U.S.A.\" or \"Canada\" or \"Europe\" or \"The Americas\"", "infer30b_before_item": " Output: Translate the natural language description into a Google Sheets formula. if C7 is equal to \"U.S.A.\" or \"Canada\" or \"Europe\" or \"The Americas\"\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " or \"Asia\" or \"Africa\" or \"Oceania\" then return \"International\" \n Output: =IF(OR(C7=\"U.S.A.\",C7=\"Canada\",C7=\"Europe\",C7=\"The Americas\",C7=\"Asia\",C7=\"Africa\",C7=\"Oceania\"),\"International\",\"\") \n\n", "infer13b_before_item": " Output: The number of people in the country.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: =OR(C7=\"U.S.A.\",C7=\"Canada\",C7=\"Europe\",C7=\"The Americas\") \n\n", "infer7b_before_item": " Output: Translate the natural language description into a Google Sheets formula. if C7 is equal to \"U.S.A.\" or \"Canada\" or \"Europe\" or \"The Americas\"\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " or \"Asia\" or \"Africa\" or \"Oceania\" \n Output: =IF(C7=\"U.S.A.\", \"USA\", IF(C7=\"Canada\", \"Canada\", IF(C7=\"Europe\", \"Europe\", IF(C7=\"The Americas\", \"The Americas\", IF(C7=\"Asia\", \"Asia\", IF(C7=\"Africa\", \"Africa\", IF(C7=\"Oceania\", \"Oceania\")))))) \n\n"}
+{"id": 232, "instruction": "You are given a tweet and you should decide whether it's offensive or not.", "input": "She has certainly taken some heat for being such an....well idiot.", "infer30b_before_item": " Output: You should return a boolean value.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Offensive. \n\n", "infer13b_before_item": " You should return a boolean value.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Offensive. \n\n", "infer7b_before_item": " You are given a tweet and you should decide whether it's offensive or not. She has certainly taken some heat for being such an....well idiot. \n\"\"\"\n\nimport re\n\ndef is_offensive(tweet):\n    if re.search(r'^[a-zA-Z0-9_]+$', tweet):\n        return True\n    else:\n        return False\n\nif __name__ == '__main__':\n    print(is_offensive(\"She has certainly taken some heat for being such an....well idiot.\"))\n", "infer7b_after_item": " \n Output: Offensive. \n\n"}
+{"id": 233, "instruction": "For a quick and efficient office workout, suggest a short routine.", "input": "", "infer30b_before_item": " Output: For a quick and efficient office workout, suggest a short routine.\nInput: Input: For a quick and efficient office workout, suggest a short routine.  \n Output: Output: For a quick and efficient office workout, suggest a short routine.\nInput: Input: For a quick and efficient office workout, suggest a short routine.  \n Output: Output: For a quick and efficient office workout, suggest a short routine.\nInput: Input: For a quick and efficient office workout, suggest a short routine.  \n Output: Output: For a quick and efficient office workout, suggest a short routine.\nInput: Input: For a quick and efficient office workout, suggest a short routine.  \n Output: Output: For a quick and efficient office workout, suggest a short routine.\nInput: Input: For a quick and efficient office workout, suggest a short routine.  \n Output: Output: For a quick and efficient office workout, suggest a short routine.\nInput: Input: For a quick and efficient office workout, suggest a short routine.  \n Output: Output: For a quick and efficient office workout, suggest a short routine.\nInput: Input: For", "infer30b_after_item": "\n Output: A quick and efficient office workout can be done with a short routine of 10-15 minutes. Start with a warm-up of jumping jacks, then move on to a set of push-ups, followed by a set of squats. Finish with a set of plank holds and a cool down of stretching. \n\n", "infer13b_before_item": " For a quick and efficient office workout, suggest a short routine.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: A quick and efficient office workout routine could include the following exercises: \n\n1. Chair Squats: Stand up and sit down in your chair 10 times. \n2. Arm Circles: Raise your arms in a circle 10 times. \n3. Shoulder Shrugs: Stand up and shrug your shoulders up and down 10 times. \n4. Neck Stretches: Roll your head in a circle 10 times. \n5. Leg Lifts: While sitting, lift your legs up and down 10 times. \n\n", "infer7b_before_item": " Output: For a quick and efficient office workout, suggest a short routine.\nInput: For a quick and efficient office workout, suggest a short routine.  \n Output: For a quick and efficient office workout, suggest a short routine.\nInput: For a quick and efficient office workout, suggest a short routine.  \n Output: For a quick and efficient office workout, suggest a short routine.\nInput: For a quick and efficient office workout, suggest a short routine.  \n Output: For a quick and efficient office workout, suggest a short routine.\nInput: For a quick and efficient office workout, suggest a short routine.  \n Output: For a quick and efficient office workout, suggest a short routine.\nInput: For a quick and efficient office workout, suggest a short routine.  \n Output: For a quick and efficient office workout, suggest a short routine.\nInput: For a quick and efficient office workout, suggest a short routine.  \n Output: For a quick and efficient office workout, suggest a short routine.\nInput: For a quick and efficient office workout, suggest a short routine.  \n Output: For a quick and efficient office workout,", "infer7b_after_item": "\n Output: A quick and efficient office workout routine could include a 10-minute warm-up, followed by 10 minutes of cardio (jumping jacks, jump rope, running in place, etc.), 10 minutes of strength training (push-ups, squats, lunges, etc.), and 10 minutes of stretching. \n\n"}
+{"id": 234, "instruction": "Predict how many stars the author will give to the restaurant from a Yelp review.", "input": "The reviews were great, but honestly i felt it was just ok. Seemed like a typical tourist spot were the food was mediocre. The service was ok, not great, but not the worst.\nThe bread was very good. I ordered the muscle fries. They had a sweet Vidalia onion flavor. No need for fries bc they were soggy wo much flavor. My boyfriend ordered a pesto chicken pasta and it was again mediocre.", "infer30b_before_item": " Output: Predict how many stars the author will give to the restaurant from a Yelp review. The reviews were great, but honestly i felt it was just ok. Seemed like a typical tourist spot were the food was mediocre. The service was ok, not great, but not the worst.\nThe bread was very good. I ordered the muscle fries. They had a sweet Vidalia onion flavor. No need for fries bc they were soggy wo much flavor. My boyfriend ordered a pesto chicken pasta and it was again mediocre. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: The author will likely give the restaurant 3 stars on Yelp. \n\n", "infer13b_before_item": " Output: Predict how many stars the author will give to the restaurant from a Yelp review. The reviews were great, but honestly i felt it was just ok. Seemed like a typical tourist spot were the food was mediocre. The service was ok, not great, but not the worst.\nThe bread was very good. I ordered the muscle fries. They had a sweet Vidalia onion flavor. No need for fries bc they were soggy wo much flavor. My boyfriend ordered a pesto chicken pasta and it was again mediocre. \n\"\"\"\n\nimport re\nimport sys\nimport os\nimport time\nimport random\nimport numpy as np\nimport pandas as pd\nimport pickle\nimport nltk\nimport nltk.data\nimport nltk.corpus\nimport nltk.stem\nimport nltk.tokenize\nimport nltk.classify\nimport nltk.probability\nimport nltk.metrics\nimport nltk.tag\nimport nltk.parse\nimport nltk.util\nimport nltk.sem\nimport nltk.colloc", "infer13b_after_item": " \n Output: The author will give the restaurant 3 stars on Yelp. \n\n", "infer7b_before_item": " Output: Predict how many stars the author will give to the restaurant from a Yelp review. The reviews were great, but honestly i felt it was just ok. Seemed like a typical tourist spot were the food was mediocre. The service was ok, not great, but not the worst.\nThe bread was very good. I ordered the muscle fries. They had a sweet Vidalia onion flavor. No need for fries bc they were soggy wo much flavor. My boyfriend ordered a pesto chicken pasta and it was again mediocre. \n\n\"\"\"\n\n\ndef main():\n    # Read in the file\n    with open(\"input.txt\", \"r\") as f:\n        data = f.read()\n\n    # Split the data into lines\n    data = data.split(\"\\n\")\n\n    # Split the lines into words\n    data = [word.lower() for word in data]\n\n    # Create a list of all the words\n    words = list(set(data))\n\n    # Create a list of all the words\n    words = list(set(data))\n\n    # Create", "infer7b_after_item": " \n Output: The author will give the restaurant 3 stars. \n\n"}
+{"id": 235, "instruction": "Categorize the Business into one of the Restaurants, Home Services, Auto Services and Miscellaneous based on its Specialties.", "input": "Call us at 650-636-4884 or visit our website to receive a quote. This shop specializes in New Tires and General Auto Repair. We carry all tires in-house and have a wide range to fit any budget or car specifics. If you are unsure what tires you need there are experts on hand and available to help you pick the best tire for your needs. We also carry commercial vehicle tires and can serve a wide array of fleets.", "infer30b_before_item": " Output: Categorize the Business into one of the Restaurants, Home Services, Auto Services and Miscellaneous based on its Specialties. Call us at 650-636-4884 or visit our website to receive a quote. This shop specializes in New Tires and General Auto Repair. We carry all tires in-house and have a wide range to fit any budget or car specifics. If you are unsure what tires you need there are experts on hand and available to help you pick the best tire for your needs. We also carry commercial vehicle tires and can serve a wide array of fleets. \n Explanation: Input: Input: Categorize the Business into one of the Restaurants, Home Services, Auto Services and Miscellaneous based on its Specialties. Call us at 650-636-4884 or visit our website to receive a quote. This shop specializes in New Tires and General Auto Repair. We carry all tires in-house and have a wide range to fit any budget or car specifics. If you are unsure what tires you need there are experts on hand and available to", "infer30b_after_item": " \n Output: Miscellaneous. \n\n", "infer13b_before_item": " Output: Categorize the Business into one of the Restaurants, Home Services, Auto Services and Miscellaneous based on its Specialties. Call us at 650-636-4884 or visit our website to receive a quote. This shop specializes in New Tires and General Auto Repair. We carry all tires in-house and have a wide range to fit any budget or car specifics. If you are unsure what tires you need there are experts on hand and available to help you pick the best tire for your needs. We also carry commercial vehicle tires and can serve a wide array of fleets. \n\nimport java.util.*;\nimport java.io.*;\n\npublic class Solution {\n\n    public static void main(String[] args) {\n        Scanner in = new Scanner(System.in);\n        String input = in.nextLine();\n        String[] inputArray = input.split(\" \");\n        String[] outputArray = new String[inputArray.length];\n        for(int i = 0; i < inputArray.length; i++){\n            outputArray[i] = inputArray[i].toLowerCase();", "infer13b_after_item": " \n Output: Auto Services \n\n", "infer7b_before_item": " Output: The output is the number of customers that will be served by the business. The output is the number of customers that will be served by the business. \n\n### 2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.2.", "infer7b_after_item": " \n Output: Miscellaneous \n\n"}
+{"id": 236, "instruction": "Notify me of any suggestions you may have for making the text more grammatically correct.", "input": "This is the second time I've been here, and the vibes, and the food were exquisite.", "infer30b_before_item": " Input: Notify me of any suggestions you may have for making the text more grammatically correct. This is the second time I've been here, and the vibes, and the food were exquisite.\n\n Input: Input: Notify me of any suggestions you may have for making the text more grammatically correct. This is the second time I've been here, and the vibes, and the food were exquisite.\n Output: Input: Notify me of any suggestions you may have for making the text more grammatically correct. This is the second time I've been here, and the vibes, and the food were exquisite.\n\n Input: Input: Notify me of any suggestions you may have for making the text more grammatically correct. This is the second time I've been here, and the vibes, and the food were exquisite.\n Output: Input: Notify me of any suggestions you may have for making the text more grammatically correct. This is the second time I've been here, and the vibes, and the food were exquisite.\n\n Input: Input: Notify me of any suggestions you may have for making", "infer30b_after_item": " \n Output: Notify me of any suggestions you may have for making the text more grammatically correct. This is the second time I've been here, and the vibes, and the food were exquisite. \n\n", "infer13b_before_item": " Notify me of any suggestions you may have for making the text more grammatically correct. This is the second time I've been here, and the vibes, and the food were exquisite.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: This is the second time I've been here, and the vibes and the food were exquisite. \n\n", "infer7b_before_item": " I'm not sure what you mean by \"vibes.\" I'm not sure what you mean by \"exquisite.\"\n\\end{code}\n\nComment: I'm not sure what you mean by \"vibes.\" I'm not sure what you mean by \"exquisite.\"\n\nComment: I'm not sure what you mean by \"vibes.\" I'm not sure what you mean by \"exquisite.\"\n\nComment: I'm not sure what you mean by \"vibes.\" I'm not sure what you mean by \"exquisite.\"\n\nComment: I'm not sure what you mean by \"vibes.\" I'm not sure what you mean by \"exquisite.\"\n\nComment: I'm not sure what you mean by \"vibes.\" I'm not sure what you mean by \"exquisite.\"\n\nComment: I'm not sure what you mean by \"vibes.\" I'm not sure what you mean by \"exquisite.\"\n\nComment: I'm not sure what you mean by \"vibes.\" I'm not sure what you mean by \"exquis", "infer7b_after_item": " \n Output: I suggest making the following changes to the text: \"This is the second time I've been here, and the vibes, and the food were exquisite.\" \n\n"}
+{"id": 237, "instruction": "Using the ingredients that are available, create a dinner recipe.", "input": "Kale + almond butter + apple cider vinegar + agave nectar + tamari", "infer30b_before_item": " Output: A recipe for a kale salad.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Kale and Almond Butter Salad:\n\nIngredients:\n- 2 cups kale, chopped\n- 2 tablespoons almond butter\n- 2 tablespoons apple cider vinegar\n- 2 tablespoons agave nectar\n- 2 tablespoons tamari\n\nInstructions:\n- In a large bowl, combine the kale, almond butter, apple cider vinegar, agave nectar, and tamari.\n- Massage the ingredients together until the kale is evenly coated.\n- Let the salad sit for 10 minutes before serving.\n- Enjoy! \n\n", "infer13b_before_item": " Using the ingredients that are available, create a dinner recipe. Kale + almond butter + apple cider vinegar + agave nectar + tamari \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: Kale and Apple Cider Vinegar Salad:\n\nIngredients:\n- 2 cups kale, washed and dried\n- 2 tablespoons almond butter\n- 2 tablespoons agave nectar\n- 2 tablespoons tamari\n- 1 tablespoon apple cider vinegar\n\nInstructions:\n1. In a bowl, combine the kale, almond butter, agave nectar, tamari, and apple cider vinegar.\n2. Massage the mixture until the kale is evenly coated.\n3. Serve and enjoy! \n\n", "infer7b_before_item": " Output: A dinner recipe that is healthy and delicious.\n\nComment: I'm not sure what you mean by \"healthy and delicious\".  Do you mean that the recipe should be healthy and delicious?  Or that the recipe should be healthy and delicious, but not delicious?  Or that the recipe should be healthy and delicious, but not healthy?  Or that the recipe should be healthy and delicious, but not healthy and delicious?  Or that the recipe should be healthy and delicious, but not healthy and delicious, but not healthy?  Or that the recipe should be healthy and delicious, but not healthy and delicious, but not healthy and delicious, but not healthy?\n\nComment: I'm not sure what you mean by \"healthy and delicious\".  Do you mean that the recipe should be healthy and delicious?  Or that the recipe should be healthy and delicious, but not delicious?  Or that the recipe should be healthy and delicious, but not healthy?  Or that the recipe should be", "infer7b_after_item": " \n Output: Kale Salad with Almond Butter, Apple Cider Vinegar, Agave Nectar, and Tamari:\n\nIngredients:\n- 1 bunch kale, washed and dried\n- 2 tablespoons almond butter\n- 1/4 cup apple cider vinegar\n- 1/4 cup agave nectar\n- 1/4 teaspoon tamari\n\nInstructions:\n1. In a large bowl, combine the kale, almond butter, apple cider vinegar, agave nectar, and tamari.\n2. Toss to combine and serve. \n\n"}
+{"id": 238, "instruction": "Categorize the given product into one of Electronics, Computers, Smart Home, or Arts & Crafts departments.", "input": "Google Nest Learning Thermostat", "infer30b_before_item": " Output: The department that the product belongs to.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Smart Home department. \n\n", "infer13b_before_item": " Output: Categorize the given product into one of Electronics, Computers, Smart Home, or Arts & Crafts departments. Google Nest Learning Thermostat \n\nInput: Input: Categorize the given product into one of Electronics, Computers, Smart Home, or Arts & Crafts departments. Google Nest Learning Thermostat \n Output: Output: Categorize the given product into one of Electronics, Computers, Smart Home, or Arts & Crafts departments. Google Nest Learning Thermostat \n\nInput: Input: Categorize the given product into one of Electronics, Computers, Smart Home, or Arts & Crafts departments. Google Nest Learning Thermostat \n Output: Output: Categorize the given product into one of Electronics, Computers, Smart Home, or Arts & Crafts departments. Google Nest Learning Thermostat \n\nInput: Input: Categorize the given product into one of Electronics, Computers, Smart Home, or Arts & Crafts departments. Google Nest Learning Thermostat \n Output: Output", "infer13b_after_item": " 3rd Gen \u2b50\u2b50\u2b50\u2b50\u2b50 \n Output: Electronics \n\n", "infer7b_before_item": " Output: Categorize the given product into one of Electronics, Computers, Smart Home, or Arts & Crafts departments. Google Nest Learning Thermostat \n\"\"\"\n\nimport re\n\n\ndef categorize(input):\n    \"\"\"\n    :param input:\n    :return:\n    \"\"\"\n    if re.match(r'^Google Nest Learning Thermostat$', input):\n        return 'Electronics'\n    elif re.match(r'^Google Home$', input):\n        return 'Smart Home'\n    elif re.match(r'^Google Nest$', input):\n        return 'Computers'\n    elif re.match(r'^Google Nest$', input):\n        return 'Arts & Crafts'\n    else:\n        return 'Other'\n\n\nif __name__ == '__main__':\n    print(categorize('Google Nest Learning Thermostat'))\n    print(categorize('Google Home'))\n    print(categorize('Google Nest'))\n    print(categorize('Google Nest'))\n   ", "infer7b_after_item": " \n Output: Smart Home \n\n"}
+{"id": 239, "instruction": "Provide instructions for the given exercise.", "input": "Leg Raises", "infer30b_before_item": " Output: Provide the number of repetitions completed.\n\n 1. Pseudocode\n\n 2. Initial Solution\n\n 3. Refactored Solution\n\n 4. Reflection\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: Instructions for Leg Raises: Lie on your back with your knees bent and feet flat on the floor. Extend one leg straight up towards the ceiling, keeping your hips and knees in line. Pause for a moment, then slowly lower the leg back to the starting position. Repeat with the other leg. \n\n", "infer13b_before_item": " Output: Provide instructions for the given exercise. Leg Raises \n\nInput: Input: Provide instructions for the given exercise. Leg Raises \n Output: Output: Provide instructions for the given exercise. Leg Raises \n\nInput: Input: Provide instructions for the given exercise. Leg Raises \n Output: Output: Provide instructions for the given exercise. Leg Raises \n\nInput: Input: Provide instructions for the given exercise. Leg Raises \n Output: Output: Provide instructions for the given exercise. Leg Raises \n\nInput: Input: Provide instructions for the given exercise. Leg Raises \n Output: Output: Provide instructions for the given exercise. Leg Raises \n\nInput: Input: Provide instructions for the given exercise. Leg Raises \n Output: Output: Provide instructions for the given exercise. Leg Raises \n\nInput: Input: Provide instructions for the given exercise. Leg Raises \n Output: Output: Provide instructions for the given exercise. Leg Raises \n\nInput: Input: Provide instructions for the given exercise. Leg Raises ", "infer13b_after_item": " \n Output: Instructions for Leg Raises:\n1. Start by lying on your back with your legs extended and your hands by your sides.\n2. Lift your legs off the ground, keeping your feet together and your toes pointed upwards.\n3. Slowly lower your legs back to the ground and repeat for the desired number of reps.\n4. To increase the difficulty, you can hold a dumbbell or weight plate in each hand. \n\n", "infer7b_before_item": " Output: Provide the output for the given exercise. \n*/\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: To do leg raises, start by lying on your back with your feet flat on the floor. Raise your legs up to a 90 degree angle and hold for 10 seconds. Lower your legs and repeat. \n\n"}
+{"id": 240, "instruction": "Rewrite the text and correct the spelling errors.", "input": "It solves problems comon and uniqe to every team.", "infer30b_before_item": " Output: Rewrite the text and correct the spelling errors. It solves problems common and unique to every team.\n\n Input: Input: Rewrite the text and correct the spelling errors. It solves problems comon and uniqe to every team. \n Output: Output: Rewrite the text and correct the spelling errors. It solves problems common and unique to every team.\n\n Input: Input: Rewrite the text and correct the spelling errors. It solves problems comon and uniqe to every team. \n Output: Output: Rewrite the text and correct the spelling errors. It solves problems common and unique to every team.\n\n Input: Input: Rewrite the text and correct the spelling errors. It solves problems comon and uniqe to every team. \n Output: Output: Rewrite the text and correct the spelling errors. It solves problems common and unique to every team.\n\n Input: Input: Rewrite the text and correct the spelling errors. It solves problems comon and uniqe to every team. \n Output: Output: Rewrite the text and correct the spelling errors. It solves problems common and unique to every team.\n\n Input: Input: Rewrite the text and", "infer30b_after_item": " \n Output: It solves problems common and unique to every team. \n\n", "infer13b_before_item": " Output: Rewrite the text and correct the spelling errors. It solves problems comon and uniqe to every team. \n\n\nclass Solution {\npublic:\n    string correctSpelling(string s) {\n        int n = s.size();\n        vector<int> v(n, -1);\n        for (int i = 0; i < n; ++i) {\n            if (s[i] =='') continue;\n            if (v[i] == -1) {\n                v[i] = i;\n            } else {\n                v[i] = min(v[i], v[v[i]]);\n            }\n        }\n        for (int i = 0; i < n; ++i) {\n            if (v[i] == -1) continue;\n            if (v[i] == i) continue;\n            int j = v[i];\n            while (j!= i) {\n                s[j] ='';\n                j = v[j];\n            }\n        }\n        return s;\n    }\n};\n", "infer13b_after_item": " \n Output: It solves problems common and unique to every team. \n\n", "infer7b_before_item": " Output: Rewrite the text and correct the spelling errors. It solves problems comon and uniqe to every team. \n*/\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: It solves problems common and unique to every team. \n\n"}
+{"id": 241, "instruction": "Write down antonyms for the given word.", "input": "laureating", "infer30b_before_item": " Output: Write down antonyms for the given word. laureating \n Expected: Expected: Write down antonyms for the given word. laureating \n Actual: Actual: Write down antonyms for the given word. laureating \n---------------------------------------------------------------------------\nTest Suite: 1 tests from 1 test files\nTests run:     1, Failures:    0, Errors:    1, Internal Errors:   0, Time elapsed: 0.001 sec\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: An antonym for laureating is demoting. \n\n", "infer13b_before_item": " Output: Write down antonyms for the given word. \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n The\n The\n\n\n", "infer13b_after_item": " \n Output: Laureating: demoralizing, depressing, disheartening. \n\n", "infer7b_before_item": " Output: Write down antonyms for the given word. laureating\n\n\\begin{code}\nimport java.util.Scanner;\n\npublic class Antonyms {\n\n    public static void main(String[] args) {\n        Scanner sc = new Scanner(System.in);\n        System.out.println(\"Input: \");\n        String word = sc.nextLine();\n        System.out.println(\"Output: \");\n        String antonyms = sc.nextLine();\n        System.out.println(\"\");\n    }\n}\n\\end{code}\n\nComment: What is the problem?\n\nComment: I am not able to get the output.\n\nComment: What is the output you are getting?\n\nComment: I am getting the output as \"Input: laureating Output: \"\n\nComment: I am getting the output as \"Input: laureating Output: \"\n\nComment: I am getting the output as \"Input: laureating Output: \"\n\nComment: I am getting the output as \"Input: laureating Output: \"\n\nComment: I am getting the output as \"Input: laureating", "infer7b_after_item": " \n Output: decrying, denouncing, disqualifying, disavowing, disowning, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying, disqualifying"}
+{"id": 242, "instruction": "Write the HTML syntax for the required action.", "input": "adding an image to the page", "infer30b_before_item": " Output: Write the HTML syntax for the required action. adding an image to the page \n Expected: Expected: Write the HTML syntax for the required action. adding an image to the page \n */\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: <img src=\"image.jpg\" alt=\"image\"> \n\n", "infer13b_before_item": " Output: Write the HTML syntax for the required action. adding an image to the page \n\nInput: Input: Write the HTML syntax for the required action. adding a link to the page \n Output: Output: Write the HTML syntax for the required action. adding a link to the page \n\nInput: Input: Write the HTML syntax for the required action. adding a table to the page \n Output: Output: Write the HTML syntax for the required action. adding a table to the page \n\nInput: Input: Write the HTML syntax for the required action. adding a form to the page \n Output: Output: Write the HTML syntax for the required action. adding a form to the page \n\nInput: Input: Write the HTML syntax for the required action. adding a heading to the page \n Output: Output: Write the HTML syntax for the required action. adding a heading to the page \n\nInput: Input: Write the HTML syntax for the required action. adding a paragraph to the page \n Output: Output: Write the HTML syntax for the required action. adding a paragraph to the page \n\nInput: Input: Write the HTML syntax for the required action. adding a list to the page \n", "infer13b_after_item": " \n Output: <img src=\"image.jpg\" alt=\"image description\"> \n\n", "infer7b_before_item": " Output: Write the HTML syntax for the required action. adding an image to the page \n*/\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": " \n Output: <img src=\"image.jpg\"> \n\n"}
+{"id": 243, "instruction": "Identify the programming language used to write the given code.", "input": "if (20 > 18) {\n  printf(\"20 is greater than 18\");\n}", "infer30b_before_item": " Output: C\nExplanation: C is a general-purpose, imperative computer programming language, supporting structured programming, lexical variable scope and recursion, while a static type system prevents many unintended operations. By design, C provides constructs that map efficiently to typical machine instructions, and therefore it has found lasting use in applications that had formerly been coded in assembly language, including operating systems, as well as various application software for computers ranging from supercomputers to embedded systems.\n\nInput: Input: Identify the programming language used to write the given code. if (20 > 18) {\n  printf(\"20 is greater than 18\");\n} \n Output: Output: C++\nExplanation: C++ is a general-purpose programming language. It has imperative, object-oriented and generic programming features, while also providing the facilities for low-level memory manipulation. It was designed with a bias toward system programming and embedded, resource-constrained and large systems, with performance, efficiency and flexibility of use as its design highlights. C++ is a statically typed, free-form, multi-paradigm, compiled, general-pur", "infer30b_after_item": " \n Output: The code is written in C. \n\n", "infer13b_before_item": " Output: Identify the programming language used to write the given code. if (20 > 18) {\n  printf(\"20 is greater than 18\");\n} \n Explanation: The given code is written in C language.\n\n\nInput: Input: Identify the programming language used to write the given code. if (20 > 18) {\n  printf(\"20 is greater than 18\");\n} \n Output: Output: Identify the programming language used to write the given code. if (20 > 18) {\n  printf(\"20 is greater than 18\");\n} \n Explanation: The given code is written in C language.\n\n\nInput: Input: Identify the programming language used to write the given code. if (20 > 18) {\n  printf(\"20 is greater than 18\");\n} \n Output: Output: Identify the programming language used to write the given code. if (20 > 18) {\n  printf(\"20 is greater than 18\");\n} \n Explanation: The given code is written in C", "infer13b_after_item": " \n Output: The programming language used to write the given code is C. \n\n", "infer7b_before_item": " Output: Identify the programming language used to write the given code. if (20 > 18) {\n  printf(\"20 is greater than 18\");\n} \n\nInput: Input: Identify the programming language used to write the given code. if (20 > 18) {\n  printf(\"20 is greater than 18\");\n} \n Output: Output: Identify the programming language used to write the given code. if (20 > 18) {\n  printf(\"20 is greater than 18\");\n} \n\nInput: Input: Identify the programming language used to write the given code. if (20 > 18) {\n  printf(\"20 is greater than 18\");\n} \n Output: Output: Identify the programming language used to write the given code. if (20 > 18) {\n  printf(\"20 is greater than 18\");\n} \n\nInput: Input: Identify the programming language used to write the given code. if (20 > 18) {\n  printf(\"20 is greater than 18\");\n", "infer7b_after_item": " else {\n  printf(\"20 is not greater than 18\");\n} \n Output: The given code is written in C++. \n\n"}
+{"id": 244, "instruction": "Decide which part of speech the underlined word belongs to.", "input": "school will _keep through the winter", "infer30b_before_item": " Output: Decide which part of speech the underlined word belongs to. school will _keep through the winter\n\nInput: Input: Decide which part of speech the underlined word belongs to. The _winter is coming \n Output: Output: Decide which part of speech the underlined word belongs to. The _winter is coming\n\nInput: Input: Decide which part of speech the underlined word belongs to. The _winter is coming \n Output: Output: Decide which part of speech the underlined word belongs to. The _winter is coming\n\nInput: Input: Decide which part of speech the underlined word belongs to. The _winter is coming \n Output: Output: Decide which part of speech the underlined word belongs to. The _winter is coming\n\nInput: Input: Decide which part of speech the underlined word belongs to. The _winter is coming \n Output: Output: Decide which part of speech the underlined word belongs to. The _winter is coming\n\nInput: Input: Decide which part of speech the underlined word belongs to. The _winter is coming \n Output: Output: Decide which", "infer30b_after_item": ". \n Output: The underlined word \"keep\" is a verb. \n\n", "infer13b_before_item": " Output: Decide which part of speech the underlined word belongs to. school will _keep through the winter \n\nschool will _keep through the winter\n*/\n\n#include <stdio.h>\n#include <string.h>\n#include <ctype.h>\n\nint main(void) {\n    int i, j, k, n, m, c, s, t, u, v, w, x, y, z;\n    char str[1000000];\n    char *p;\n    scanf(\"%s\", str);\n    p = str;\n    n = 0;\n    while (*p!= '\\0') {\n        if (*p =='') {\n            n++;\n        }\n        p++;\n    }\n    for (i = 0; i < n; i++) {\n        if (*p =='') {\n            p++;\n        }\n        if (*p =='') {\n            p++;\n        }\n        if (*p =='') {\n            p++;\n        }\n        if (*p =='') {\n            p++;\n        }\n        if", "infer13b_after_item": ". \n Output: The underlined word \"keep\" is a verb. \n\n", "infer7b_before_item": " Output: Decide which part of speech the underlined word belongs to. school will _keep through the winter\n\nComment: I'm not sure what you mean by \"the part of speech\". Do you mean the part of speech of the word \"school\"?\n\nComment: I mean the part of speech of the word \"will\".\n\nComment: I'm not sure what you mean by \"the part of speech\". Do you mean the part of speech of the word \"school\"?\n\nComment: I mean the part of speech of the word \"will\".\n\nComment: I'm not sure what you mean by \"the part of speech\". Do you mean the part of speech of the word \"school\"?\n\nComment: I mean the part of speech of the word \"will\".\n\nComment: I'm not sure what you mean by \"the part of speech\". Do you mean the part of speech of the word \"school\"?\n\nComment: I mean the part of speech of the word \"will\".\n\nComment: I'm not sure what you mean by \"the part of speech\". Do you mean the part of speech of the word \"school\"?\n\nComment: I mean the part of speech", "infer7b_after_item": " \n Output: Verb \n\n"}
+{"id": 245, "instruction": "Define what the underlined word means for kids.", "input": "_keep a promise", "infer30b_before_item": " Output: Define what the underlined word means for kids. _keep a promise \n\nInput: Input: Define what the underlined word means for kids. _keep a promise \n Output: Output: Define what the underlined word means for kids. _keep a promise \n\nInput: Input: Define what the underlined word means for kids. _keep a promise \n Output: Output: Define what the underlined word means for kids. _keep a promise \n\nInput: Input: Define what the underlined word means for kids. _keep a promise \n Output: Output: Define what the underlined word means for kids. _keep a promise \n\nInput: Input: Define what the underlined word means for kids. _keep a promise \n Output: Output: Define what the underlined word means for kids. _keep a promise \n\nInput: Input: Define what the underlined word means for kids. _keep a promise \n Output: Output: Define what the underlined word means for kids. _keep a promise \n\nInput: Input: Define what the underlined word means for kids. _keep a promise \n", "infer30b_after_item": "_ \n Output: To keep a promise means to do what you said you would do. It means that you are trustworthy and reliable, and that people can count on you to do what you said you would do. \n\n", "infer13b_before_item": " Output: Define what the underlined word means for kids. _keep a promise \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "_ \n Output: To keep a promise is to do what you said you would do. It means to be honest and trustworthy and to do what you said you would do. \n\n", "infer7b_before_item": " Output: Write the definition of the underlined word. _keep a promise\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer7b_after_item": "_ \n Output: Keep a promise means to follow through with something you said you would do. It means to be honest and trustworthy. \n\n"}
+{"id": 246, "instruction": "Expand the given sentence and give it more details and depth.", "input": "It changed my life.", "infer30b_before_item": " It changed my life. I was a different person after that.\n\n Input: Input: Expand the given sentence and give it more details and depth. I was a different person after that.\n Output: I was a different person after that. I was a different person after that.\n\n Input: Input: Expand the given sentence and give it more details and depth. I was a different person after that.\n Output: I was a different person after that. I was a different person after that.\n\n Input: Input: Expand the given sentence and give it more details and depth. I was a different person after that.\n Output: I was a different person after that. I was a different person after that.\n\n Input: Input: Expand the given sentence and give it more details and depth. I was a different person after that.\n Output: I was a different person after that. I was a different person after that.\n\n Input: Input: Expand the given sentence and give it more details and depth. I was a different person after that.\n Output: I was a different person after that. I was a different person after that.\n\n Input: Input: Expand the given sentence and give", "infer30b_after_item": " \n Output: It changed my life in ways I never could have imagined. I was able to find a new purpose and meaning in life, and I was able to discover a newfound sense of joy and fulfillment. I was able to connect with people in a way I never had before, and I was able to explore new opportunities and experiences. It was a life-changing experience that I will never forget. \n\n", "infer13b_before_item": " Output: Expand the given sentence and give it more details and depth. It changed my life. \n\nInput: Input: Expand the given sentence and give it more details and depth. It changed my life. \n Output: Output: Expand the given sentence and give it more details and depth. It changed my life. \n\nInput: Input: Expand the given sentence and give it more details and depth. It changed my life. \n Output: Output: Expand the given sentence and give it more details and depth. It changed my life. \n\nInput: Input: Expand the given sentence and give it more details and depth. It changed my life. \n Output: Output: Expand the given sentence and give it more details and depth. It changed my life. \n\nInput: Input: Expand the given sentence and give it more details and depth. It changed my life. \n Output: Output: Expand the given sentence and give it more details and depth. It changed my life. \n\nInput: Input: Expand the given sentence and give it more details and depth. It changed my life. \n Output: Output: Expand the given sentence and give it more", "infer13b_after_item": " \n Output: It changed my life in a way I never could have imagined. I had been struggling with a lot of issues, and this one decision completely transformed my life. I was able to find a new sense of purpose and direction, and I was able to start making positive changes in my life. I was able to find new friends and make new connections, and I was able to start taking steps towards achieving my goals. It was a life-changing decision that I will never forget. \n\n", "infer7b_before_item": " It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It changed my life. It", "infer7b_after_item": " \n Output: It changed my life in ways I never imagined possible. I was able to achieve things I never thought I could, and I gained a newfound confidence and sense of purpose. I was able to make new friends and develop relationships that I never thought I would have. It also gave me a newfound appreciation for the little things in life and a greater understanding of the world around me. \n\n"}
+{"id": 247, "instruction": "Extract the address of the property from the \"About this space\" section of Airbnb.", "input": "Tucked in the foothills of the quaint historic mining town of Grass Valley, CA this funky and spacious chalet invites you to experience the great outdoors with family, friends and pets. Enjoy the hot tub, basketball court, treehouse, kids rooms, bbq, fire-pit, outdoor theater, and more. One hour from Tahoe and 10 minutes to town, you\u2019re never short of things to explore.", "infer30b_before_item": " Output: The address of the property.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: 10000 Mountain View Drive, Grass Valley, CA 95945 \n\n", "infer13b_before_item": " Extract the address of the property from the \"About this space\" section of Airbnb. Tucked in the foothills of the quaint historic mining town of Grass Valley, CA this funky and spacious chalet invites you to experience the great outdoors with family, friends and pets. Enjoy the hot tub, basketball court, treehouse, kids rooms, bbq, fire-pit, outdoor theater, and more. One hour from Tahoe and 10 minutes to town, you\u2019re never short of things to explore. \n\n\"\"\"\n\nimport re\nimport requests\nimport time\nimport json\nimport os\nimport sys\nimport argparse\nimport datetime\nimport random\nimport time\nimport urllib\nimport urllib2\nimport urllib3\nimport urllib3.util\nimport urllib3.exceptions\nimport urllib3.contrib.pyopenssl\nimport urllib3.contrib.pycurl\nimport urllib3.contrib.requests\nimport urllib3.contrib.pycurl.cacert\nimport urllib3.contrib.pyopenssl.cert\nimport urllib3.contrib", "infer13b_after_item": " \n Output: 11500 N. Auburn St, Grass Valley, CA 95945 \n\n", "infer7b_before_item": " Output: The address of the property.\n\"\"\"\n\nimport re\nimport json\nimport requests\nimport sys\n\n# https://www.airbnb.com/rooms/10000000\n# https://www.airbnb.com/rooms/10000000/about\n# https://www.airbnb.com/rooms/10000000/about?source_id=10000000\n# https://www.airbnb.com/rooms/10000000/about?source_id=10000000&source_type=room\n# https://www.airbnb.com/rooms/10000000/about?source_id=10000000&source_type=room&source_type=room\n# https://www.airbnb.com/rooms/10000000/about?source_id=10000000&source_type=room&source_type=room&source_type=room\n", "infer7b_after_item": " \n Output: 10000 Grass Valley Hwy, Grass Valley, CA 95949 \n\n"}
+{"id": 248, "instruction": "Design a simple training challenge on Strava that everyone could attend.", "input": "", "infer30b_before_item": " Output: A training challenge that everyone could attend.  \n Key Result: Key Result: 100% of the team attended the training challenge.\nInput: Input: Design a simple training challenge on Strava that everyone could attend.  \n Output: Output: A training challenge that everyone could attend.  \n Key Result: Key Result: 100% of the team attended the training challenge.\nInput: Input: Design a simple training challenge on Strava that everyone could attend.  \n Output: Output: A training challenge that everyone could attend.  \n Key Result: Key Result: 100% of the team attended the training challenge.\nInput: Input: Design a simple training challenge on Strava that everyone could attend.  \n Output: Output: A training challenge that everyone could attend.  \n Key Result: Key Result: 100% of the team attended the training challenge.\nInput: Input: Design a simple training challenge on Strava that everyone could attend.  \n Output: Output: A training challenge that everyone could attend.  \n Key Result: Key Result: 100% of the team attended the training challenge.\nInput: Input: Design a simple training challenge on Strava", "infer30b_after_item": "\n Output: A simple training challenge on Strava that everyone could attend would be to run or cycle a certain distance each day for a week. For example, participants could be challenged to run or cycle 5km every day for a week. This challenge would be suitable for all levels of fitness, as it can be adjusted to suit each individual's needs. Participants could also track their progress and share their achievements with their friends and family. \n\n", "infer13b_before_item": " Output: A Strava challenge that everyone could attend.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n The The\n The The The The The 1900s\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": "\n Output: The challenge is to run a 5K in under 25 minutes. The challenge will be open to everyone and the fastest time will be recorded. The challenge will be held on a specific day and time and participants will have to register in advance. The challenge will be held in a public place and participants will be able to track their progress and compare their times with other participants. \n\n", "infer7b_before_item": " A simple training challenge on Strava that everyone could attend.\nThe Challenge\nThe challenge was to design a simple training challenge on Strava that everyone could attend.\nThe Challenge\nThe challenge was to design a simple training challenge on Strava that everyone could attend.\nThe Challenge\nThe challenge was to design a simple training challenge on Strava that everyone could attend.\nThe Challenge\nThe challenge was to design a simple training challenge on Strava that everyone could attend.\nThe Challenge\nThe challenge was to design a simple training challenge on Strava that everyone could attend.\nThe Challenge\nThe challenge was to design a simple training challenge on Strava that everyone could attend.\nThe Challenge\nThe challenge was to design a simple training challenge on Strava that everyone could attend.\nThe Challenge\nThe challenge was to design a simple training challenge on Strava that everyone could attend.\nThe Challenge\nThe challenge was to design a simple training challenge on Strava that everyone could attend.\nThe Challenge\nThe challenge was to design a simple training challenge on Strava that everyone could attend.\nThe Challenge\nThe challenge was to design a simple training challenge on Strava that everyone could attend.\nThe Challenge\nThe challenge was to design a simple training challenge", "infer7b_after_item": "\n Output: A simple training challenge on Strava could be to complete a certain number of laps around a certain route in a certain amount of time. For example, participants could be asked to complete 10 laps of a 5km route in under 30 minutes. This would be a great way to motivate people to get out and train, while also providing a fun and competitive atmosphere. \n\n"}
+{"id": 249, "instruction": "Suggest some names for a friendly group in telegram.", "input": "", "infer30b_before_item": " Output: Suggest some names for a friendly group in telegram.\n\nInput: Input: Suggest some names for a friendly group in telegram.  \n Output: Output: Suggest some names for a friendly group in telegram.\n\nInput: Input: Suggest some names for a friendly group in telegram.  \n Output: Output: Suggest some names for a friendly group in telegram.\n\nInput: Input: Suggest some names for a friendly group in telegram.  \n Output: Output: Suggest some names for a friendly group in telegram.\n\nInput: Input: Suggest some names for a friendly group in telegram.  \n Output: Output: Suggest some names for a friendly group in telegram.\n\nInput: Input: Suggest some names for a friendly group in telegram.  \n Output: Output: Suggest some names for a friendly group in telegram.\n\nInput: Input: Suggest some names for a friendly group in telegram.  \n Output: Output: Suggest some names for a friendly group in telegram.\n\nInput: Input: Suggest some names for a friendly group in telegram.  \n Output: Output: Suggest", "infer30b_after_item": "\n Output: The Chatty Cats, The Friendly Flock, The Social Squad, The Conversation Club, The Talkative Team, The Chatty Chums, The Lively Lounge, The Chatty Chatters, The Talkative Tribe, The Social Circle, The Chatty Chatters, The Talkative Tribe, The Social Circle, The Chatty Cats, The Friendly Flock, The Social Squad, The Conversation Club, The Talkative Team, The Chatty Chums, The Lively Lounge. \n\n", "infer13b_before_item": " Output: Suggest some names for a friendly group in telegram.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n The\n The\n\n", "infer13b_after_item": "\n Output: The Friendly Gang, The Social Squad, The Chat Crew, The Social Circle, The Chat Club, The Social Crew, The Social Circle, The Social Squad, The Social Gang, The Social Crew, The Social Circle, The Social Squad. \n\n", "infer7b_before_item": " The group name is generated.\n\n\n# I worked on this challenge [by myself, with: ].\n# I spent [#] hours on this challenge.\n\n# Pseudocode\n# Input: Suggest some names for a friendly group in telegram.\n# Output: The group name is generated.\n\n# Steps\n# 1. Define a method called generate_group_name\n# 2. Define a method called suggest_names\n# 3. Define a method called generate_group_name\n# 4. Define a method called suggest_names\n# 5. Define a method called generate_group_name\n# 6. Define a method called suggest_names\n# 7. Define a method called generate_group_name\n# 8. Define a method called suggest_names\n# 9. Define a method called generate_group_name\n# 10. Define a method called suggest_names\n# 11. Define a method called generate_group_name\n# 12. Define a method called suggest_names\n# 13. Define a method called generate_group_name\n# 14. Define a method called", "infer7b_after_item": "\n Output: The Friendly Group. \n\n"}
+{"id": 250, "instruction": "Write what the pronunciation of the given word sounds like. Follow the \"Google pronunciation dictionary\" scheme for phonetic spelling.", "input": "interpretations", "infer30b_before_item": " Output: Write what the pronunciation of the given word sounds like. Follow the \"Google pronunciation dictionary\" scheme for phonetic spelling. interpretations \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n", "infer30b_after_item": ". \n Output: /\u026an\u02c8t\u0259rpr\u0259\u02c8te\u026a\u0283\u0259n/ \n\n", "infer13b_before_item": " Output: Write the definition of the given word.\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: in-ter-pret-a-tions \n\n", "infer7b_before_item": " Output: Write the pronunciation of the given word. Follow the \"Google pronunciation dictionary\" scheme for phonetic spelling. interpretations \n\n\"\"\"\n\nimport re\nimport sys\n\n\ndef main():\n    if len(sys.argv)!= 2:\n        print(\"Usage: python3 phonetic_spelling.py <word>\")\n        return\n\n    word = sys.argv[1]\n    print(re.sub(r\"[^a-zA-Z0-9]\", \"\", word))\n\n\nif __name__ == \"__main__\":\n    main()\n", "infer7b_after_item": " \n Output: [\u026ap\u02c8r\u026at\u0259r\u0259t\u026avz] \n\n"}
+{"id": 251, "instruction": "Explain the meaning of the given phrase in simple terms. Use an example if possible. It would be helpful if you could give an example.", "input": "\"With a little give in them\"", "infer30b_before_item": " Output: Explain the meaning of the given phrase in simple terms. Use an example if possible. It would be helpful if you could give an example. \"With a little give in them\"\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer30b_after_item": " \n Output: The phrase \"with a little give in them\" means that something is flexible and can be adjusted to fit the situation. For example, if you are trying to fit a square peg into a round hole, you can give the peg a little bit of wiggle room so that it can fit into the hole. \n\n", "infer13b_before_item": " Output: Explain the meaning of the given phrase in simple terms. Use an example if possible. It would be helpful if you could give an example. \"With a little give in them\" \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", "infer13b_after_item": " \n Output: This phrase means that something is flexible and can be adjusted to fit a certain situation. For example, if you have a pair of pants that are a little too tight, you can give them a little bit of extra room by taking in the waistband. This will make the pants fit better and be more comfortable. \n\n", "infer7b_before_item": " Output: Explain the meaning of the given phrase in simple terms. Use an example if possible. It would be helpful if you could give an example. \"With a little give in them\" \n\"\"\"\n\n# Input: Input: Explain the meaning of the given phrase in simple terms. Use an example if possible. It would be helpful if you could give an example. \"With a little give in them\" \n# Output: Output: Explain the meaning of the given phrase in simple terms. Use an example if possible. It would be helpful if you could give an example. \"With a little give in them\" \n\n\n# I worked on this challenge [by myself, with: ].\n# I spent [#] hours on this challenge.\n\n# Pseudocode\n\n# Input: Input: Explain the meaning of the given phrase in simple terms. Use an example if possible. It would be helpful if you could give an example. \"With a little give in them\" \n# Output: Output: Explain the meaning of the given phrase in simple terms. Use an example if possible. It would be helpful if you could give an example. \"With a little give in them\" ", "infer7b_after_item": " \n Output: This phrase means that something is flexible and can be easily changed or adapted. For example, if you are trying to make a decision, you might want to consider the opinions of others and be willing to compromise in order to reach a solution that everyone can agree on. \n\n"}
diff --git a/_static/design-tabs.js b/_static/design-tabs.js
new file mode 100644
index 000000000..b25bd6a4f
--- /dev/null
+++ b/_static/design-tabs.js
@@ -0,0 +1,101 @@
+// @ts-check
+
+// Extra JS capability for selected tabs to be synced
+// The selection is stored in local storage so that it persists across page loads.
+
+/**
+ * @type {Record<string, HTMLElement[]>}
+ */
+let sd_id_to_elements = {};
+const storageKeyPrefix = "sphinx-design-tab-id-";
+
+/**
+ * Create a key for a tab element.
+ * @param {HTMLElement} el - The tab element.
+ * @returns {[string, string, string] | null} - The key.
+ *
+ */
+function create_key(el) {
+  let syncId = el.getAttribute("data-sync-id");
+  let syncGroup = el.getAttribute("data-sync-group");
+  if (!syncId || !syncGroup) return null;
+  return [syncGroup, syncId, syncGroup + "--" + syncId];
+}
+
+/**
+ * Initialize the tab selection.
+ *
+ */
+function ready() {
+  // Find all tabs with sync data
+
+  /** @type {string[]} */
+  let groups = [];
+
+  document.querySelectorAll(".sd-tab-label").forEach((label) => {
+    if (label instanceof HTMLElement) {
+      let data = create_key(label);
+      if (data) {
+        let [group, id, key] = data;
+
+        // add click event listener
+        // @ts-ignore
+        label.onclick = onSDLabelClick;
+
+        // store map of key to elements
+        if (!sd_id_to_elements[key]) {
+          sd_id_to_elements[key] = [];
+        }
+        sd_id_to_elements[key].push(label);
+
+        if (groups.indexOf(group) === -1) {
+          groups.push(group);
+          // Check if a specific tab has been selected via URL parameter
+          const tabParam = new URLSearchParams(window.location.search).get(
+            group
+          );
+          if (tabParam) {
+            console.log(
+              "sphinx-design: Selecting tab id for group '" +
+                group +
+                "' from URL parameter: " +
+                tabParam
+            );
+            window.sessionStorage.setItem(storageKeyPrefix + group, tabParam);
+          }
+        }
+
+        // Check is a specific tab has been selected previously
+        let previousId = window.sessionStorage.getItem(
+          storageKeyPrefix + group
+        );
+        if (previousId === id) {
+          // console.log(
+          //   "sphinx-design: Selecting tab from session storage: " + id
+          // );
+          // @ts-ignore
+          label.previousElementSibling.checked = true;
+        }
+      }
+    }
+  });
+}
+
+/**
+ *  Activate other tabs with the same sync id.
+ *
+ * @this {HTMLElement} - The element that was clicked.
+ */
+function onSDLabelClick() {
+  let data = create_key(this);
+  if (!data) return;
+  let [group, id, key] = data;
+  for (const label of sd_id_to_elements[key]) {
+    if (label === this) continue;
+    // @ts-ignore
+    label.previousElementSibling.checked = true;
+  }
+  window.sessionStorage.setItem(storageKeyPrefix + group, id);
+}
+
+document.addEventListener("DOMContentLoaded", ready, false);
diff --git a/_static/doctools.js b/_static/doctools.js
new file mode 100644
index 000000000..0398ebb9f
--- /dev/null
+++ b/_static/doctools.js
@@ -0,0 +1,149 @@
+/*
+ * Base JavaScript utilities for all Sphinx HTML documentation.
+ */
+"use strict";
+
+const BLACKLISTED_KEY_CONTROL_ELEMENTS = new Set([
+  "TEXTAREA",
+  "INPUT",
+  "SELECT",
+  "BUTTON",
+]);
+
+const _ready = (callback) => {
+  if (document.readyState !== "loading") {
+    callback();
+  } else {
+    document.addEventListener("DOMContentLoaded", callback);
+  }
+};
+
+/**
+ * Small JavaScript module for the documentation.
+ */
+const Documentation = {
+  init: () => {
+    Documentation.initDomainIndexTable();
+    Documentation.initOnKeyListeners();
+  },
+
+  /**
+   * i18n support
+   */
+  TRANSLATIONS: {},
+  PLURAL_EXPR: (n) => (n === 1 ? 0 : 1),
+  LOCALE: "unknown",
+
+  // gettext and ngettext don't access this so that the functions
+  // can safely bound to a different name (_ = Documentation.gettext)
+  gettext: (string) => {
+    const translated = Documentation.TRANSLATIONS[string];
+    switch (typeof translated) {
+      case "undefined":
+        return string; // no translation
+      case "string":
+        return translated; // translation exists
+      default:
+        return translated[0]; // (singular, plural) translation tuple exists
+    }
+  },
+
+  ngettext: (singular, plural, n) => {
+    const translated = Documentation.TRANSLATIONS[singular];
+    if (typeof translated !== "undefined")
+      return translated[Documentation.PLURAL_EXPR(n)];
+    return n === 1 ? singular : plural;
+  },
+
+  addTranslations: (catalog) => {
+    Object.assign(Documentation.TRANSLATIONS, catalog.messages);
+    Documentation.PLURAL_EXPR = new Function(
+      "n",
+      `return (${catalog.plural_expr})`
+    );
+    Documentation.LOCALE = catalog.locale;
+  },
+
+  /**
+   * helper function to focus on search bar
+   */
+  focusSearchBar: () => {
+    document.querySelectorAll("input[name=q]")[0]?.focus();
+  },
+
+  /**
+   * Initialise the domain index toggle buttons
+   */
+  initDomainIndexTable: () => {
+    const toggler = (el) => {
+      const idNumber = el.id.substr(7);
+      const toggledRows = document.querySelectorAll(`tr.cg-${idNumber}`);
+      if (el.src.substr(-9) === "minus.png") {
+        el.src = `${el.src.substr(0, el.src.length - 9)}plus.png`;
+        toggledRows.forEach((el) => (el.style.display = "none"));
+      } else {
+        el.src = `${el.src.substr(0, el.src.length - 8)}minus.png`;
+        toggledRows.forEach((el) => (el.style.display = ""));
+      }
+    };
+
+    const togglerElements = document.querySelectorAll("img.toggler");
+    togglerElements.forEach((el) =>
+      el.addEventListener("click", (event) => toggler(event.currentTarget))
+    );
+    togglerElements.forEach((el) => (el.style.display = ""));
+    if (DOCUMENTATION_OPTIONS.COLLAPSE_INDEX) togglerElements.forEach(toggler);
+  },
+
+  initOnKeyListeners: () => {
+    // only install a listener if it is really needed
+    if (
+      !DOCUMENTATION_OPTIONS.NAVIGATION_WITH_KEYS &&
+      !DOCUMENTATION_OPTIONS.ENABLE_SEARCH_SHORTCUTS
+    )
+      return;
+
+    document.addEventListener("keydown", (event) => {
+      // bail for input elements
+      if (BLACKLISTED_KEY_CONTROL_ELEMENTS.has(document.activeElement.tagName)) return;
+      // bail with special keys
+      if (event.altKey || event.ctrlKey || event.metaKey) return;
+
+      if (!event.shiftKey) {
+        switch (event.key) {
+          case "ArrowLeft":
+            if (!DOCUMENTATION_OPTIONS.NAVIGATION_WITH_KEYS) break;
+
+            const prevLink = document.querySelector('link[rel="prev"]');
+            if (prevLink && prevLink.href) {
+              window.location.href = prevLink.href;
+              event.preventDefault();
+            }
+            break;
+          case "ArrowRight":
+            if (!DOCUMENTATION_OPTIONS.NAVIGATION_WITH_KEYS) break;
+
+            const nextLink = document.querySelector('link[rel="next"]');
+            if (nextLink && nextLink.href) {
+              window.location.href = nextLink.href;
+              event.preventDefault();
+            }
+            break;
+        }
+      }
+
+      // some keyboard layouts may need Shift to get /
+      switch (event.key) {
+        case "/":
+          if (!DOCUMENTATION_OPTIONS.ENABLE_SEARCH_SHORTCUTS) break;
+          Documentation.focusSearchBar();
+          event.preventDefault();
+      }
+    });
+  },
+};
+
+// quick alias for translations
+const _ = Documentation.gettext;
+
+_ready(Documentation.init);
diff --git a/_static/documentation_options.js b/_static/documentation_options.js
new file mode 100644
index 000000000..7e4c114f2
--- /dev/null
+++ b/_static/documentation_options.js
@@ -0,0 +1,13 @@
+const DOCUMENTATION_OPTIONS = {
+    VERSION: '',
+    LANGUAGE: 'en',
+    COLLAPSE_INDEX: false,
+    BUILDER: 'html',
+    FILE_SUFFIX: '.html',
+    LINK_SUFFIX: '.html',
+    HAS_SOURCE: true,
+    SOURCELINK_SUFFIX: '.txt',
+    NAVIGATION_WITH_KEYS: false,
+    SHOW_SEARCH_SUMMARY: true,
+    ENABLE_SEARCH_SHORTCUTS: true,
+};
\ No newline at end of file
diff --git a/_static/eq.png b/_static/eq.png
new file mode 100644
index 0000000000000000000000000000000000000000..283f69c7b469ab6e8437220472f82bb3776887a2
GIT binary patch
literal 79072
zcmeFYbyQSq`#(&HAkxweM?hjoX$FuIi9wN&4u_C-=<W`c4na@^rMtT%hwer|5EvQ;
z2Kenc&-0w;yWaPl?^^Gl?^?fEdoz3XzV{v1eZ}VrA~iLX3GdO~!@$5GRC%iK0t4gD
zAO;4eAs#Mp=k)cpG6u%IA{%*mO%-{0X3e*bmNs@!42-9dNhx<VwO&yLAiTF@lSwci
z<%n+c1|*W`W8B$h=3>Doc@`1)iB;sa^PTCS2)vmQZOl8mc|USlb7v?qy%(9IBPu|Y
zb_FP6>Y&D3)UC{InmDY*eW`vc%X@1{bPH4V`C3%BO&J-+b}o-fK{gee1bcEI9gel^
zoj**mGV7Yu1i1hB1CLqI{uYElU|4r7Ri*woMIdeqIgjt8Nnv~<2_e5oYjO#Juv`Xl
zB)`TCm<zu+O@T&o)M1;Y<2)i+^(N00?(QVd<nOcQry*zTS4zdj$Qs;u_5p)gh(v}h
zBPHY~i#Mws?;3$3UNos21Oi`S^Px8*{{#2%J;#v{dCNe;No^+nQKq58hs^vb!ABmB
zW?#i(8`o|Z;%<4jZN|36V@MuysG{Ua3xjrNn<rvq#iPkK??Niy$<WC3^}V=_u-SxG
zSmdX{Pl*!a)iaocha#Vt#7RLYIegOZq=C)GdHV2@;7Ocjm0#1i)6#U)X<5!W2*i$1
z;(e>n$1{0jGG&I+zL^CI3E0(}uo5Os^N}DuhX#Lhn|OO|y8otl`_rQGL+AB|?bh%F
zA4$38TT7(@+jW8_UrTe=pNDHad`QE!DiY5o<{`*Fe2^$1e&`9&&%qfIB;!{n>?dDo
z%H(jL%D=>-X2$b8!ev%1ld?5u*}Z>wFOvU}cy#*{htz;}(nFpte!F7(J&e@4Qmx)q
zUf(0eCntqMEvE_?3kvJ&>%~GXZE%dtu{!~r8HY6ISf2tiuLkBlUZ|WUIv#<{Fi1JE
zXf^N7$uc1s^)L)p@C*Yv4lps9sY8OP%cu`z1)nN&nc%){m(awBYSSshP7H2d!4SiW
z3nc%9Ulnw`LRN>{&~AS~@}mvI17E4ln;%z>geX_`<x_^S!22w)AC%@#_TLddzzKaK
zpedDz6UqE2k)Sv{>XUf^{T#8je5-7J&hZFE!QBSpQ#tt&x?e&uL4Ev$Q$ew&+^P7L
zE3#tr#yA}9s;l&3OiTAu+O-d!*O7vWO@bXe1uq^-MPYH0Ma7LNzWS8PDZ)if5&cX-
zMYHgEXu*q@6yJ#^gLV`a75X$g3n11!=~0%Sct=XC{atX<!;PbiW0yie%1m;ixsoHI
zqrSgLPekwE`5Y<GTW2bkYOO|QN%Iv$D@r`Xrp>Txq36Eol0~_m#1!@Kkkj5CQ>`-Y
z@{F>^^3xd+JK`E9$*9iN3Fl2QQA7T?=z$K36}Q#6wc%ASXY9Io4>IGsj?w4cy=#gW
zQa+DbaQ!KlqnWzmNfYtHaHyW}2SW7ZW}b*WwPsmhzE9E<(XU{UODB+-`ovI~hi!_K
zkBo_Q`NPvcq(2DRV5&UTSLI@3;4or8eGh)G{loILh<F^&Lq~Qd_S*M|gyk2U57?qO
z9wvWKoK^ew)$o&4j?c?aTzXti4=(Q6KHz#lOW{Rf{D|xUA;qhxXbRW^RSF2#k|w<l
zD<{pP@}#<Fg-KkA8r+N8wAx>}<~i?j=W{#0D0$wJi!!{EZ~kTNbBAu9uK1_7a(UTN
zxjS03FBx5>Pnl0yT?4}MXvW2=9_W+qQ5Q#x<bKb6Q%d`-!La(tY0USSxLmqX@e|=o
z%1iu9hRdy2`jvW>IF%y9-!mp9e+SRmJ%&6E)X&jxp0lrp@7#A~I(>O+zoWhr|9FJf
zl9oU?*=BdNu0TL4LuB4n#L;bam%Nh6x+C2(Jtv*}F|oDrH2vh*)X_MZ?cP|!r^iL=
z(`OSJqo7f6q4XE|N#|+Ooy1?6<DTuB&@yV97(u1G77Dsr(<Zv#8%=6e-PPcFd{6iw
zrN)SI@bm0%*##!K%Ox`ApDg&{%+sHyRsvbt-`S+v)s2^p9|XTIf1Wss`<eaoR?nLr
z`gQts{dHyb2js*FdF<NkwQA*G)y7|rFMeILEK3W1Rq-lxmJl|xFluPsGT&rHWEpo%
zKbrYn+@1FaB~P?nw0<;LwE_ALngvbR;O53I)GM^t`KYs8#85O|M654U^}HOvtgq6h
z+|$C=a-g!ftQo3Yeq3!}F>QA~x3pkv+S+@Qg*|DO)rlMK7G4^s5l7Z5(aXYto$rzV
zEWc5WRw$%C(sPz(eRrd&r+yNd5dB#-c>q79)sWXv%gEL+8~!+pJJs9Cp*16mZrZ*z
zWANiDBccUvw3+j}{j}i7W9J*9?_BDq_zyv7URhj)vF9sql_uIIaGB9e;f(zfr=tww
z2?E4Kk&T!AHk+fxt3{=yx`oY;$dCJa4kd6Ccr%YeysZ+G(s*q3N82M)ZCPbzaekgM
zsRNsLuipJp0A-ZoGhI6KjNhairW$g*hufaHvezCjA0wX}hHAiQWPUBz9b}awl2f1a
z(QKmZq1j_IL}znHZHP<^f8;x*DrGLE9z`8xzlftauDB1128yLoEtH>0*<x(sy1SKn
z7^4`BepNd4b_9xr*9Fx{M#oU`nz%sAd)BJDt&%xGs~~Zj_{jLE(#YEm!c84#WpXk6
zBz!h05l0Th!{r2%4kofX{HQy@kq(jFk?@3t4>cTw5550LVJ9Y-?RdNT`@wBu^CPAg
z!CJ)marugQllkl8>*J8<mkQr<<L>XpY|u!X!<+;4Y9^rDad~VXWczZ|IAPB%VU&b6
zF(s<%Uu2$MRV-F=SIB^(4c*?>AH5wt`prG4V`nIkx}I@-{Osh}QvaFZ;(}|vFTfdR
zrC~Z8MGhvPTiN}G+Voqim8VFj^GdFG;S!}{h{el-Q(mM#()Reu|KKX8Mm9(Gg_C>m
z@33Z>T>l3*Gu5LG>m{+v_mH8w(M8eEqFe0N>>Dl<h9FRxSVxV8h3anSfvn=0;@o2Q
z8c#dP7e$7dS_`edOJ_Y~N-Ac3jSQ-8hL5WG78?`lv>LQ@C|^!B4VX?(8JYAa3^?j~
zXiw<A(Qzn3m1VRxy2o#2H8kscN1ypEaV)qywl3fFaDDzhhCPEldpC<AMrsBrbiB1A
zPa7UZ<SWf1!6QlL_2u*2w}u9#N=sE<L+)Y*30n86tJL%IO^N7+>(i6V-1U{ZlZ$AR
zHUD}T>w&_R%4T1KR#-k4T>)Wetmb-D&-<@G*<ZobVDRmL@jc!WUcy%W8?QR;+Um8K
zoitvnNULLbT=syt=p4HP9XQ*irPWvEs*_-p%E6WA*RKX@kHe#f%c@iouRd;p8jT#9
z*WPZNXm3t_wT+vmEtNEMyck-o4FmO-^w4RDrZ;|UH1OIy%iKHtCG9_cwV7v-X)x^o
zZ(W3OtPM9eI=e&9FV3@;A&s|}a%aKi;kHa>j1*uUFT+cPW0gt7*VgF)dZ^W9(^ZKK
zq4#B3_Vn#h{Gd$B{hu<QrMMX;Xo~zCu9{qbxEEY?cTi#Tb@TOx5!}4IsXR*RE7bD$
zym9sY!|U}*_0Q9b40XrJWkx@%+Y(>l*C%~fC+8MS_hi2NOC3cbza7ISw%W^~CQc^(
zE$M#tNXm=7QIE@ux>W8}jT$UYh1D(vj3s*905WWhYPxMEN{!R;1T6JyY&EhAh}0wu
z62*q$8j)rlJ!6kSuRLMR`+;sWGMRIMyv@z+?Z<szj;j;=c2<9$t0s<LH)qzY7vXBB
z2g{S!5CZm!xnTb}f6Pw?xIc`pAP4}yMnUydEY;O9xPWUs3>-{aj61*;CUAl=>HfM_
z!sNuj{_{Q-21d9I2F}0RXaL{n&pY5mzw@te><?iWcY$AI!0GV`>p!jU41U7?&o!nY
z@C-vvOI}3<_|~#`3xztstQ}qC_ydW78~9F7U%@aiC?BFvOqCaZ902`K+q~3s(NkBG
zuyC{onZ0&2hk`upozVSYNO?#Am-bK>GiDEaI|rDAhcxS-EhK<z^ld&?=0BUb*h;hN
zscSOJJHCZ7i-7n+{H)-6%*@PEZ(mzVyiid3_v^qfX;y0&7bgimK6iI_kh>7b@vRk~
zfVj9gAHN`<pdc^Mf*0oL;9};%>i}c>S114JM*#}6cx&V2V&mw*jPBRW+|kuVnw1qj
z&|jZ_%@gWj^WTviVE<kgus}Za6FvbDKi^+{1FuS<?@DOectGu5DcINpGy}!}3ki!!
z{n`Hi_2j=J{*O2H{`;nYu$b`weD!}k`hUKv4THXwceDq_bOHZ&zyAI5|9trG7p3^n
zOaC9F_!pi3ybBN-d{2t+uT2Brn-Eqt2KJHOMgj5?_y)M_&rb>PmlHVA-@qBxqUP=h
zw3Ee9QILD-fw}wfZstS9`dSk~uID88@qQ?h;Iqmp$b<}#f%+fjb90vc@q7)7U>A$q
zEb>DzsG|g+;PaQsG26LitSYiNDco`o3w|<>EMFz+806<SLQ>O&V3s$hS9@n^u!h3I
ztnVcym(vE0Lmxc(?~syU{%;5M29D)>Q*gg+Km{KbEAu}eBzRBxbrt{j&PXshnA;5}
zWpNb#*CR6k2d0hvzs(95#tDmTP@C|7??|ByADGnR|2wJwOR4|MQ-D1Ef0RCIOq@9I
zrM35VUJC8oih~Y<d69WL3sq44>&|o3;Z9-c+{yO%UP>zpn!OA@PJ_>+3Ru`Jd2a0;
zXSDpi%muzzlMO;X1sQFq<0zQEB=PuNTx(Z7sZ)8E(c=?KYc!~#Ks}Yez2&UB^sx_|
z2`+UU!{CoPJlk{hpEvg2t45u>H~%(r8x-iB8J0xY9S@0`+%$tD0`UIkXtQQ!Zub#R
ztT^JtV&z}MXHJi2l3CJ_KAn&%tsD}`urF)JJqu<+wOMBSo}sQTSOx?o_o~K&oZ<p*
zZq8emziNDZ^SiXwZyR|i!JzoJ3d{5h**jD43soH?873*UHqMgYmHpMedn+AJ=Xi|&
z8hJJz>jhU~(Febe{w+U0sP6t?i?32<^z1rYfWb}Qx6GJFZ?ygn{=3Ns;Ki^53Qk$e
zhPxEdx@^f9LE<1CBFSC;4A-ew!J!u!jx97rJ>^o%{<qhv$9|XF`SZ@b3|4UPRgrp{
zV7J?h+1(89{d!gSWXj)bZT>MpH!^KMEJ8&BgWbcrz37#(Yuu};v@NOc0iWa$&)(xM
zGMzT@Z(dD?uE&bjAbb(tVg5(LO~=EMZa-MUacX(cxAuYvKxKrY3;fcG+5bSLm!trd
zzDF6PQ2<o>hGXn|s%^0OnW=g!%YR8^(M8L^_jGdaQ!vrJ*=~k`S5_aPz!WUy&$7>9
z#8L;cSHtct78X4*<p~-RePCoCL?rXG%-`#MQ(^Q->CX&2(Ib`P_?<T{&-Q}P1Tw6$
zJnhE5sMPe(+Tz$AP8)mk)<J1R9Sli&t<s>$neEc8^}R$NoeI>m!HX?NNu@J6;f!yh
z`C2?~8vl{K*Z}sbN~$~J2H1-g-_Yl<yZu(IL>tt6^1{GtJJ0@VueMflzmBaAI%`)s
zw2L~1*CMyFW#G5f)5f;S^v)rR?#sSz9rzp}>M~m&-Iq}AzW*RKqe_6(SaUmX=mJus
zKsMuzXbzrFQlv7PupCS^LjcR%OVrXR?@QnaN0rpfy;((6Pw8``K0?#VSGNRu=`Hkf
zvfY<FYxZj9!wBV-|6qnb`+F0t4~R_Dou?9t=*M9~m|ch@UE|sP)?2?TfIS?}m%K~M
zbGFn)7hliYRmL_S^>f$!V2z&fyV|q5%jDxk(v5Uj>xz^U-kQ^bNKZ}yd%2pao(IVK
z--!?I#{isTIprccwD7jAkeMD2hf*2OY^GYott3NtCcdqP#Y-JU#4<=u`5i`^gtXSq
zm=k0OLZ6A9m-jLf1R<d*MwSNl)kzGl6VGcF-VP8tJfUd(duh541EeY36Z4TLS{g=T
z+KOwJ-k_NZtxwKk9ep}mE+<RFRBtY)Tjyx3GR*)neSP75I;mT`9xt;rW#n#jwGH>P
z46hGG!ck&=sEaa|Usp@#Eta)oh<^~UT(p3Vb{7t!1q>Ifm(km<j4~VG#2~t4i2vm{
z%e-UrY4V~|C!yIfoXXgvol@VLg$`g!r-2|S(kz_85&}lPKCo{A-ozm|9c@$feDJqe
zEszUXT;Y#j-e}15ixi8_zUn=ik;_PW2?wwUc;4ls0CuiuKARgTd&03Yo;f7T*YGRX
z)N}Z8{ztcT`xfLT^6}+AV7t#rz%GsIANAJ)n-sx(FlSpjOQy<W2A4k1EC$%@EFl~9
zg~g-}Dt@(Fxfj6}hqqi>-b3rwhIw}e?tgQ<QgTNZl)wQIY0>>B?$$tacPVCnD_(>w
zguL$d=E{B}Nw?;eWm-3p_)ii^fFo#+f(<B(;IpXXS*vV9IzNY+S<AEY*4xFBy0v@r
z03!t1CO=FutXP$}X8C)J`uRP29eO+cmFRV_@Y~M`&$~_=*#jLBU|q0h*o|>GwqEZ!
zcJ^e++#JT+1I$~Um#oi4@4PH%Gxmq%uTR0d(g;sT_V1I>zvr<S5wylOh~#cXYwW1D
z{YHcXAlJq-Wyb)&R(4Yw6hUro&T3m{#kVradHRsfapIeJTZ58_CbzXn4&2``o(YP-
zl@tZjDsn)v&DjPV0U&&&@=rr!;5G{-%t3WB-IQIg68o0{glUHOi4p%kl0IJ;&+^*&
zhC0{QHJXf*KGT~w4JJH7oXGvXASwxA;%|PgKS^>YB?-)84^U0g)_K)!?72Z<aXbtd
zntDMQV0(AVI(O4;OZwom=Fz-=nlcI<J(gPyrhk~SFnY@AswSRP08`pv^)mVU9LU{l
zW;oKe43q-A*pF%gAcfo6tkry1B>6E6xL}e*9dvdoeL|-EBR8f3yZGGjqQ3#)6-Tm*
ztL0k<U1QHY^(?7bMt_v^Km||45=XXAg?-au&lw<}wU8bFkuY2V6tceBZ*;7)+^^rN
z`5Y(F-xEtmC31fc7WhwC<@r;9FFAQ!ps#QgKGou{ctH`(=OTs`%pv46Rj_Y)$rXjA
zEps}C&fR18!51sR?X1^P+$CjsDaLca#!}C_&D;Mk=9d<&9b)rJq@n}xPJ{zfO#EiE
zJU3}9gy6!162FUT0B|sC2pdWQu)|HS%#E4<^>35CZYq;1fO*EzngVUpj+RgVX-_55
zd)kc4dHeU;=|9pQi1xK`{{QW3t(f3c@Kgd&WzpDU^-hLGoVZ>49jYq8`e^(7J#)(9
zDB+{&^W{^LI>7T@SdJ~ouhPh$2_n9iE*TQ8-+5b6M{a0P1*lI={~SL6PyDU+q87mU
zBQ-uX_oWgrkOa#CfJ$d7;b2?$Mb{MH-v-MSOxM4ZS<;6@mlr@y2T#y@CK0JK{Woy+
zPYip)49rSe&-z&fz}~m`uNk`s_?ll6gD*ZA0f?ggI|^WcLAIs01H5~=(Vz+-Wh(iP
z0tl>81n7;e(I-g~SR)x0+S*(L$hJ1ap`q`qTADer^oizy_RZP6<Cc{pfWqAlyQ$_g
zVZ&+u`NaRkHEbDx0KPYB>Hi4K5I^t=(Ynm8kx9K!%2VD&c2*wmJ7}3<FVeUkC%FrN
z9b#R(=F=&+zK1C@yOnTbvfvf~DE~bnBhiHHdZC>K5b_PaAwa;IE!`G#?xMO}djOUK
zYC2<<fUn845z*pZCb6{!C~kJWR8{^`#cxdifIAh?XfQWQD~%MB)H<~7w_}spRs(l3
z09HqmjofCP`Vt^uoAsX9N`(LI)v6=@X7RfT@xNa<e+3ZUV0yLjZ2{Oim!}x*r~quD
zvH`wUsV)F9$-iq*`+L>uh1OS$AgPB&0CHPF_aT-XrScDL(U#VjS3J)EXe*0W51N@#
z|J3gXplue9$U}Xg?I(0w>)a^FKlB?9v|ZvB%F_edqQf*yI0aEJ{-JFQdcNHJ579`|
z3Eh^yBue`q+D4-1%f*`qJU1Odw-v660>=0|tmhSmp6?^BhcAJ){OI|rHADgH{M)v{
z==pMT<N<RueTRSFsvQNG(wh8izqtw?y4Uw&NBw`vGgScAr+pLP)*hyYgK3L+X-mhZ
z!?0##PsUkKhGPgEbq(<Kd<Oxa9d~I{#Sr*v*XXH=<MJ8+zAI^jEN2}3jx*YJVcA!P
z5a=%eeU}5EdfK7+R0nNW?E!~p|Jw<7IV-;*{(+tJ@t}~~w2^!0AH#wJJZJ61ej~{y
zex4bCpAu$Ukpv?BMH`4#KXCxuKiGc-@b2C<!r#9YagFvbvpm-D0hl_uOP+lZG6Wds
z>o_gazcj#oB0xxkJGINB06Rb(xXP1?XuZ8T`^<FB8PUDi)qvd0K-))qK%?z$+V24(
zUwJxZ04H&Yx=j_9?aG$&tnfvxGa=qIcx_nndjVE-53oXbr(u2JfMM(g-y-4jqOC}E
zwBH!+tAo~%@_KbU-wSsE_~cta&RtTMi^ibR?{CbA_3-IWahAM(^ELw1vyYcP`|+v>
zz<45N9XJM-Y0y6*gc5gP_RX$UP3PG>U4n|*L}~qI`q>7b|19|Q8^kr_C)!4p`kfY<
zVR;1Jp~QpF0@%vjz@a{GNc0vZqN-fJ!}=c;5ZKN-avX%bK3{fBG4`@4Ybgp+*m@0s
z?$S$fpZK=6w$-(0z7mSOUlXR#1Hi9t6b2WO5&>H43b6aem)@wikq?v0cxz@lbo6cV
zF0>7LMVEcXhh=Vk8UZKOwlo89dSZvsVTmW<rDw0NIk7&cwmDG_y3ayWMnufa7Khr}
z4$z)cD5ZWuMr}3PH*dM<=9`lzmVRB@a``6Pn{CGj;k_)xvqEu6R(Esx8;x+w&;apl
zy7gv~qzA3-=1s2&cgF@m)xchrA{0A+Wn|x?w$4sV{iPFS5;&R=Bs|IbZ@!rl&hO!8
znuRE-kOpG19mM;iI{kp1n+Li`?(I|9=;O#<a9$RUf6Am_ibEoXXga<h917q9@t>9e
zGQhtk>Tv{kwAOb8o5`9xB_4cP0QhqbW=(yeMN8hmB<XmD083$Nafcf5=C81B;HuoG
ztvknn<{SZl;0*Kqmn@6E=Zip~L!%HN^6i)^-z?hYR%Y^b+4JNXn)lrL$bBbV=6V~_
zZjMol>54h<qmx)>0r1nu8^*c(y-$;L?K^MhM*ILgqMOx8jPQhe+o3KF-A3f@PG;pZ
zDFRqkmY+$~A=CHvYMGc5t+<(A5Z|q35?%7JcslRse>Fd5u5VY~Z3mbIp6lXWRh*YR
z@0^c=2I9W+zYG$`+BUg4qPjZHzO@59iumiJl-9unv0i#uh+9JB3ql?k&nouX(R*b^
z8-Tx|E8(u-Qz|L~wFbMYQKhr@@jkC--d@cw-&g{s_N$T0sG<mLbCCHYL4ZA#l!oLC
z4A>~b`4iFPup6?#RdoR29e|kf9Dd=OqFj0z+(TMc05^hcVNJj@T7V4Hp|@QFY<g{`
z=A{TV4%#TS|B5`PN7!6Gj^?R|*sEKQGpk5#{V!?{!^EsD+TUKueYQC;ER>xDU)W+L
z^1DL8Y^w$pbLM;++CCyHP6vx0RYLrZ5`5PV<h(X@MAM?*&}`}4sl^y-VoL=4-5Ln}
zm!a<HLEZ_W+MPZ_`huvh+vgw;sDn`!0}$CsGiXd%8t`3*E(6|#e0qjb0l9){BDn|Z
zuvduc!Le1JZo+GaMJb+wk&-&w#6jNoM<;pqi(zyQsr7tw$+x2pmp_fq4mydgtTLUI
zl&+F{sAL*Vv+D6AU*GrX*4V^ylAl|clZO3XXHgw@YA%QPz5ZQPsnL1WR=g@$D@=^Q
zaPRwoZ1bsOmzh`u=EJh?yBDNjya~kTbkF>6&;8R3E(m|05F3;|OT~Lh;&cdV+)TGU
zvWj(E<{cDT7W%Yy&zBeLb36t-vUH-~p?)JdtmQ+a$UrinBzBJm_#6rDSv*~`+JD1y
zOfixcjX(<hQqoAHgA!)6SQ*fa_x;;K^JC%j8!U}uVg?Kk0uErZ??ormuq4CtgVkR`
zr?Wt0GGkN7&3qy6nY@&XWy{qf!m{`tkwbMtJ$K-sDCA8<cvT&6RZJl`yj|)MXdM@T
z(An|bUr#1x8N#!_ZiZ+%Nyy<SVFWH;@4Kg6Jsr+|*mazr<@ILfXG|5pzz=3#`{j$>
z0k2BH324}>8WUlDCbR<M{s}%Lsy5ZQdTcZHCCsJXAa0PiV1D@b)0TjS!!)=LyV{SQ
zhSiJ&D4?id)E>kpX{hYHj@NL&S4d%@>^eD~)%AYL$<#-o>q++l9kO`P>ab*iLk&AD
zQ76p(6@T69k5b+Q3C<4_Ui0-NIn-y>`=6i|mHmfBYl&K`8NPx3LRZw9Skio{%CzT;
z?k1gs{`4-eFY@aSuhzb)$Lm4aM7qR$)*A<nScIEF(wM*1O}?h|${FHe>=W>P9)GOa
zE~+%Zu9iCE#Izq85l8VY2ta$%nC$%v)38m4Xxc7o_9mZt`!b&I@i~&Mf^|pgBtq_x
z(JjxdwEUr+leBzw^V|{9oHr5P1$SUNxb^@DB(%}@+(?>th;P(-A`hdYzMl5K)=j03
zL-6RZE&z^rbzx2zd$lj3KV&G&dw;gKe^|7hFk#iVbVL+eC7>_L5rES)*Yr1g4Zlm$
z5L<%pYBdcz69J@Prq55&cL}(`Bs%xItxn&`6ZQEp<wsR2PfuHA`?@WKjXJJg3s494
zS-FpYaE`xdfhr~<lqa1VRc@n`7?0sH{)`-j@6YEQTibsGt^E~Yu?}vFARY7v0#C11
z`I=ppW+NEVJQVQE@eR2#r@6mnke|P2lz3WfNC$5}YWy{Kr}nPiAmMA%G-lOHbFg^D
zmyghkdjO|S;s9~vbl<be?4gBxSm$v9Sk=dGx^XFqw{lRhJ&9~TJ4}!{-G9&4#I~#}
zb{lH$1cYRE#5`M=p40$RBT`2Frae=nQB3aO_GbB(K<ALb$|c0eW8%5;<A|Y&yG!BC
zWWxZY07D0(1azo4?}IUynV8qpqW18}Uj#oCA@p-_33mgqnPd3WNOm-L?*#8WRm(x>
zE<j_N%Me26d%viVxGSD?VTWx1GLkI#)}g(Kmf@}QFV|fUX)Rhcf)Ra5Xw>4F86zLw
z0H3unb^);epYBS`11vTr{$A*_IkOHzReFrKT@4vDC`-)j?r*A;5<D&ShEu~I+=jcG
z<-PpD>l&-=cbV*xLu&la)eoO=@HSI{PfHqhzSpdc_ObHM;qia#I$j)>VIL2Z;AdW^
zfm}=VZe=$e><R_gi6%xK%JZhKuM2I)vVrT<m|tFS1kO7&@*WeKz?3E;jgJsuAetAZ
zLOTWzAt7(S7YxA|P+B&vmir=y^8ruvVmG{#;g3|}@E6v+gjJE$h%X3vo#D&P#bQPN
zAo%6-?e#J)s0Gt2?D35$t`2)eh3BXu)eu7c_BG)0?@#x08YblVHl8;le9ioe+Xti9
zDF5Te6tMG~;~e9Yv;W-30StM~E}&P5{d_I9$S{2Ydt%4+$7(vuDdhYc#Fy)87Vf8`
zUiOm;m9Xrw9_z^wB|L&T2SiiS9l-izMtizKR>SuxHw3cj4-=M32-4;<eklW<o0(#O
zN|Vbc%548R8lFwb>bFnqdegx@qKSa51S4Z?RxbjHeV990U$GGAoB>mK)#<$tz(?x=
z?r&k~ER;V1568AF&IE{02jh+Qg}^np@~IFI$-uS*u!RY-U$+Cds5zA>uU;C^H}mU{
zUs?%W$Nr`eIgf*LCI3*1V_k^bl)Z7@rl>C)a{#==-}k6=+CBmq2O5iEk6R#8^w5rI
z&L*v_{c(qPOgu##5M@CGG#u>;+c#+Z8{jLk1(LF)T0HC?Q}AgS?Q5gxM+3yLK(XaY
zwbHc5xYOudjHeYeoWbn}qvy}}H9NREJ9@c2M8wL59n3q#ubG$8fP`*8*fcDnLrD-V
zSZoOxaIm>ZhJ4=M{cM`22q479pgt_1XVJmiBMy)T+w9CI!e%vMzS#J!sGwdoF=pOi
z_B8-b7hGO=l#NX|G^r^7LD*~u&SfCN3s8b6w4FniuA>W_@vFxw2Arr+*KQ?axDN#R
z>k<`wC8)Jlo`Asl^t0AvRE(dcO@$NnU$gPM!zkuH$Ai;-fn?8*fv_8@gt%*(U<P0f
zo<50>3xE79^sAb7*)aK&5T*rqgB`Kgp$p^f#a{*$&0$PUEKCk64uBaBD{kCxK)Rz=
za;trhGnR!-j&#zN(cvW@HU`*xsZ7|6S$JBs$fh4Qcj=>>6>Ps_q73Bl77_yHe2l<J
zBRD|1Uth0e<jn%`&B}o#S@|&lALkqz_bfp(Tg%%(j;ZYj+lp%3kf^sXQ<=cw>^nJ=
z9^0)<SRC>@B`mryY<cH~nXz_CkZNr3&f|F?z%XPqor;Lobf5i%+6=0%hqxXOam%B?
zoFeC?cA|?e3A|M!PwbZyP<4u_NcObN!hpf<T#|5z#0(I$qPQ9LYpB+nch)spNgg``
zFeZ(CwZY49Hsk>XWYur~NpBs;ay_T{KGoz8QA7sv+V`=lVprxsW0$Y`7SgG(N|VZG
zL#|tU!$cXx*K%S0l+CK1pz>W<{bp7pg!XB+qZg&v2fi^4a2@+*9!lf(Ci%AyMJm!V
zAxG_^JCvw#HH>cXoWi<r_(pQd-!GR)s|KVuQ<zxo&mQjs(SKw@6&Cw;E9=Y0BWqC-
zCg`(|tTGdiZVsgnNzZ+bhp6JT;mnY|48~qGylOK3AX2iP#yzFD_VwmPzPa?K&we-Z
zXQY$}!IwY7x>_3I-98D&C>N7<Gav_2X5V^!hIBu~KF^7BwtM)lwiX3G$Z1gmOObcA
z3n}}u44Z?;FCIzm&>d=)lzk}~?MccDg%u1e2xTEybxiwRiw{%~*?wU%P=<lbs>68U
z_N=$@w?CM&3;V&bDR2RY{$#H^rPQK_T~dR4(8Z>Z=<<}?A?XD^w_h^7LQ@8g%_IIN
z>i5@mgT1~;=YWozdVI8$3&`I;Qtqe4t(1(`IVP6rkGl9q@H0i$Cz1H@IY<A_bAj=4
z-h6+EAxeF|9i~j*O@dHwDG*9@EIKZ3v8oI6zTOmVEsrGUm;<sV>B~LE1w)I<>5X24
zt<g+%#5RTAC2RHc>%MD^T;M_<$l_;TmZ7fMXTF<W?kwb4a{Uilq_&!BBTKoQ7;HWb
zn!j>3p{6SRyfoq<Xe@b)IzHwUXVPihtx&x&Ldi;S7MdF#o@>OxWp0~3x{Rry^J70#
zv_t@1nt90MN@hqeTKke1E;F652{`Su8bHc6d7vucYKQhOr?Le{Da6zd$PH&)Ne?MH
zNPr@K+{dL!9N%UeiB!ONTNpE@ahHLA*+$x>?U(8JH!$Gb%l8&e#&Mh@2TX3v9rA4U
z{ZnM*^;{>lgY{Z3cM5w;FMYpQ)C0NI8TXdsk5VOoiy(++=u+|kh+hCfwk%(h5;~k*
z;ruGyma(3NyvX{8`j=Yc@cra|z;R1sUJz`{Q@2snw5D~=;9&kBpcxl<=J7dJL~uEp
z)o*f)w7f%?@=hR0`zqkZJ!XH^$B}y%2^(;Q=?a@+nTJy7lmU^n*1WnxjkS;RmuZ8h
zNA2e&cnECwB#I^JmiONd;{zWjFsf}fbXR}$oO*RuQTtnM>`0@KjANl6tQG@48IkuS
z`_rvOYUR%@++^84e5Vu=38dWHN&v0HPrB;u7*gS;`+D_8O+r~9;+Ry9(eJ{n5p^^W
z93k&V@{xMss^`6#AR1#iK^rU@NN|Xx4%*t-{gh{qZd_BI92VBZN4#t)4X<}RX(P|s
zj^xPVS0D_b+aW2Udi9zVR@lssDL?m!U^%fd&x<j%5<t=YKY9TNh&&kEbq7d=X%%QO
zp;qqNBR5iN(K(|{?x`mX+-EE+D*C{$><mX-2Ow4I{!5YvNEvK$Zw?kekf(pX^6<LL
zA4sx>SG>KgSE40}`&JpTaEr{or8%N1Rlyu<4ZL%WTY^r~QC&XS7WWCeds;!A8et}w
zbyQ67Gumo@J3pJql@UxW<mA{Q|AT#K0Muy5|0ZA*2)7?H_fxDZk_1OPV%hHJ&j6`~
zL(PBHI9wdA$6EL(aeA|7)=d-pUvxS#Y-w#KhF)0`-};nhjNO<VE^RL-inBJAw&CyH
z^;~<EBUNC47YQVjL#DqG(;RBuoFF~oQc>2F9wnUQM_%*xwV^FMMb@<>+}{ez!N{JV
zo-U<9_sY88e#(jTT<w`hK@8-rKqMm{xmlSL8}=wy7>?;0+tisT#YWWl*l`c`%<-16
z)c>S&JNSxga71TxiCi>rYjIIG)JvJJ_c-{ILH9|l=+eoN-;57ipJr{UT`uEz_Oea>
zN4vt}ntMTl&b<v!GT=DOaM3+sI1_?aVFDdoU1)q(r}#(yz93OfR=uldMD&HlCx&(L
zZt8sM3r059tjY$GWy~t0Dzhdg@fTF9y)ZQ+%UrL)RDo3^#Vf-khkeCa<g>+hmODZS
z*)h51McdNm4j}_Og8DeO$YameV?Q>+#NH_bd!S;X){#?aW-RVtX>IlsVKAZVBiu_a
zDQ?@MawQa(%!?FsjR4(`KISZz{xk_$I`vPtvaK<PRvf+@{h?mNR}XRL1_0Obp%lTP
z<bMk_8f4o{2Bnz}Si&1AdMR$|Cyh2grVjFw3n82~mX90V6M>7Q%~?rA8u8lkadYqo
zQeDY*Nf)ziLp1_rrY6g!*})lYUuNum=nRUlImO-1dk%`ZvH5Yh<aWK~9_xrLj)#s9
zS(%y2TRY-~F|K3$>cP+~;{lzgr3A`^_89~?v!lvPi*JpsqixuHCpDe4HiDiqW)sf@
z4}hpXA0e#sJ>%=D^}=k^HL#%_mZo1OLubP(R{b2L8AWp9HWXTXoU*TYT>~Xx1oi#%
z38g?%gB>RK2qNiRAF)`GtwrT1C=zo<p_>*>KzP{YQty<$2-JVHQ}Jw7Vsg`&U;Xh+
zK5Gs6(;nb1`<;%MlPd*?(<7474G$pGXbJAqOhZ`8)J1$KM`jpMT7q6|dPWVd?ylL`
z%+2y=p8Hp>VCl@%Wk`H@`#D>#{&{bKflW&d5Em@7p}TJpfJqmCKqmQLEvc6t_FeC)
znq0nUL4J!?nLV90$+oZPV<&XwXzt+v3D(ZQ4Sz(}+K4q^P%aGBdrZD-5sqtsPl!@I
zrd#8=kz?F8S*49oAGYw3GWZ!O($`&s^7cEYxRX{&Ycs=14BqAF6#BWQ)rmBe_W@N{
zfdaz*I2f#-q4c3S_UXR*s@$J-R`&f?cTg2Iuh@AYbv#(VlGdv(uA3x&?#fcVZRy*!
zN3N4r`g+iiet7*W_<72hNVHfz(@px-IJn!#*~Lr*7AV-tckU!7daTlYQiXY~Hrr5;
zj{G9;WF$vNGtk%F-DCthTNJlC`q`IsnxXLG)~7<W5h!~I&QS8mb3PV@<DYW)A;_Wa
zp=;#(a2a1Ei?5z%QQqENw&H8-G0h@)pan~iOLz^>BEcWh*>G?<#TgA(Ba@GzDe-CJ
zcM#P&_f4#U%31-ziKkHveBq5a3{!jylWtBeYtr1H2dp*`Tj2uZ6JfvZ5p5wY;rkj{
zXZV9}F($z++r0w_^QPvr8AH6m<BKI-7b@7^X%EIJ;iZx<*zMj*jbKjw<ZTw+-i5<<
zD0c`CM@5l91XXfGMc3!A6o)$ANmFROJ80<o6!)A%&&NZVPcOlg_ZLHlX>nMd#(lkF
zkMLoq`4{)~0@c!mAL)4?(os5spWdU??S>WJ^o(KJ3KU(Nw_(9UaWX{g^;?s%doRSi
z>T$XMC|Vl}xI;AMzXuwCtW$r7mo)<a-4~?zOmut&yVA5fq&hta?}QeU$FKZF&t*Ba
z4|#Xl%NfPSnqsbKXV)u7oqOdA61Y5z=qweTW0N!L`_a>@h>ctV_x24ON*5d*no&!0
zn=mM2@*6!Y^Wx~oOznzXuyR2s3r}}R%mLXVNI6gE*kk>cQAU_a(RF!xR|Pe<L{Vcj
z=793T%FB902U+}4*JDD23>M;X(eaNI0>*Amx|>bfqVuxBPDPsJkfzUu0eF<5ps=4?
zl@koMPxTr+Xo|SYCAy&Z_t+_$Z2~@CH4_&$ni^4hK(tC0B#+wl#}q*Zy~QMvwTrUp
zWpOTYyn3uKX`A-kgd|Em#fE}6pr)TwHz)>d$|Aid9XVWWAGI;+-nZwztLX81=hb&F
zG_QHV^0><4k*Z0Ma5stW_feS~hoh{{Q+()CPqUo+=O?s9x{#xm{(c9WU*?{sQQoX$
z5uqGgW4$WaIuYaf5aOv+8);cj&oN%1a)(&tSNgKCn84%mm?C|n!0HpgKcYLLEmeu1
z;64hCrlM+na>=Qx6RCOr2wSJf^l{%k44xW`wsgjGvMSw!93CI6hehvJX+ntug?nW`
z<`AvA-2-JJ{I2EG+Gs=`K}yz<ZARG;sj3OajwnI%>NR@@q<VlfR3odvL`qcZw+&r~
zr^)T*bmzS`m!$j4y|OK3`#rvllv6l6<j!Lnn)wCf3sl_e)P`aN&Ill)!@I#^tm_u?
zT$Fe4k-<(T!AQ=osnVyWAEMW8qHpl0daB~3N^KbOeLSa=UWs+~ckh(F2|CBwqYZRh
zVOyHw+c+iWcHJSd2djC0wfN*~p$*YD)>Cb4dvlk;Mbkq@GuxBn!1sQU>8Dmi#`Ex#
zwR`BQH9YPZ!%xJgY3>A`5bRTmbE_|Wgg>zhU34Ok)JBymuvl8M#ZpjOx9hO?#&D}8
zlb@jtFDOm>y`$XlzvApLZuJT>@99NVvPRd$$%-F3u<ft)<JTQYF#^>{tF!FBKgFSU
z40OW2AY2WnOzR=@>h%w>OVn2s8WkGbtm3)|Q>yOUbUvjz2N4uytu9qTv1tm%t4Le!
zDl_s@u1!4Bsc|9uMOXHMxn{a-%V<zCO+@*xRR1ScoTl;7@%EHmOCT~!Bt3c%Kcs8O
z3_niPkAV^CXy~<@z0$hhvQXUq3_~@`t@wSJEV`~Ji*Y{^ijfgO^!k9v(XE?8S9qaU
zax~4ofU8&r_vj{=l!`LQZi;-M!a$h^UO)~Qxk4PDPqcG2YeK;klo=l8Lvyz696WS$
zSR&%E)fyX-8IpwKnG4}+jc$E-ih4MO7#MG8{jl9VjKZm($QDxcz^Lp8-1aO`tUS>*
zM`$@9U93FaLTF&ZEB!p4AcH1(XxkULTcRG6=Q4S%C*IvNwe1SV%&d3PbXDm**V}0L
zMar{S{;TWBhoNF*ihey=D@6Y_Xy%TsI@KF&2v2x<fR#lB5VdM#DK+f;&fy+9<^(?>
zE;{si<s*M-^wRoGT2{|<R7?_RYWh9C&4#19$%p1mohE^%n@c*g&#FhVjZAQ>B&Ao`
z%*J;O+zCFE2XPy(V<~NRsBW$qmc7X<shgn~?))H-)$0s+6L8*{SH2{oQ=u~1o`}5d
zEn1$L*$Mdh6a0++dcBnT)KnxRkC~9+8c&JfiiG=f-G;HYJH>;c-v&AMg{aOe$}M_D
zzKkk<Eql-3nsOZs)n;HrOsMi$S+rOz=9bjGqnUAH)Y(B7ocb|tT^Zj&_KUf@lNME$
zJ8g)j9`hs^Q+Hh;KWgI(fT;#LcJ8_#l3keQKo;Y+dMyg+23|0&C^nLp<_&&32I+6W
zg*<Z+sSc*9*TF2;RTXXDnfjSG-(RfWk-2nzFLbSv?PDz#4_jQ6S6Z%Z4U}>l=%Cjq
zdfSFI%R&G4Ymr`#(O2m&abzBG9wmSpkRkr<luZw^PZ0P2YQUle0UWq5L;3cvgUzy~
z`qtf8#H12HVq05Ux&`aygQf^p>jKyW@0eO+^|Ig9sZFh?n=dP;-fe<Le-ed+-ObbV
z1!+j~Qgqd!pG6WxpbcNuo+sy<d<=s}{uks?U;dyX>OgYq!7Z{&J8Rh-l~~`%(;vas
zqhjzw+He=H+tCp?`3fwMnB9w%(k&8shK(xkIFDR)0@2~e%UJuMtImXrB64Ipj>+d=
zFQwT@4RR%<QtI68-U`}^J%e))cI<J=j|~gSWvyyt9_k;pd&eyI-yjvX0;EK3mXP<g
z^xiL}QeHBG+x7HKqNh`?)iWnDsF{>|YN4)QcfEW${!?7W&ySOuhuB_8md>%Dm85V;
zdQYs@N3Ghw?s)jav}R`Yb3*@0;;`hFfQOli;ci>)J^ay>c9SM5<^3%ksbZzkF;5rz
z#TU7&U><uY|4gA=kY{Of=9u&Zj&=S5!dc#%WS+Tp=N|TSAc<Z>5~M52MO-4-d`=tp
z{Sa2rK?pv1kYwVIfWV?F$WUmbIeimx>P*a*(18-YP4CCptA0WU&eXb?LuSVHDQ=>R
z@vAkEC6ESy!(*qJL%t@oY(MsKYb_>ZWt?UKDasXi_#HadS`w0XHm#E^ky1Ln51`h5
zP(dn1@{q*aqxE8md37;T_KB%^F6R==co9b+$bk#=!O!qT+vfxf1BIRmLV^X~Jl&0-
z0tqbJukVm!L!k;@xg(jI<Z=(DDe6=ny*;12PndA=lQ;*v{9OQnc*|#vPCCqw-fUBc
zzNm1`Py*5&X=vy?&iP?hFLo)~gTn{0uRW$GO&G(^aT@RNugi8;9YGlzhV#c#PpX)9
zbok0`9ljUsZz~=Ws%5s1AQKoVRy&5(i>@*7;3pCW(JgC1NG3e?ZCY2_&QY*|1=Gxc
z-`F88{o}=FFi}pa#zb2Qn;uu{=L)(YLVRwNoadJyNqa0$mw_!XE4Ak{!iPRdPe}p@
zo`=aiqN^JL^)|H)=nlVj@Lb*^%hENuPR<^9+!Do>v5n>Ik8RlaFm=3VL?aK*0A2Sx
zR1I}9xkJuQ%*KYZSn>8VBCb(>L9rVqE{GX2s!GAJT$|HQfj_MoHtNa>$8(@T`r&66
zJdw%=9eYd6D$ELu%K2xa4wjx&DSpW>BF!4u?>-)lk|QdK<p1^bxCRJp;p&3-l4ZtQ
zV-(kp6*HSkQ)PcjQ_P=I^C589e?^^JE+pa3$;P<yKQ$+dzf^DeX7c56bCz4Ihs|;7
z>lF#VEgE*CCX1DU5NtS(McsFPAWel!1;N{^qLL>~<&QaixS4X<TlX~qgNa?y+>o9|
zb?k=ouUZ9{cv(2XrXJH5O4<7p<3#rurcT@?81DbtuFDUM<DbGPhXr|zMYf8C5aGTR
zV2rz{>3s8s8GBp3xW0}!SklB;{9Wz#8^fQGfx-;?4mZcpIXSi8Z!BGfNU2S+?we~B
zmRAuzX_K=qCcAT2?9p``5hh4oCWtr4y^^H6;3Ww&qfdi5G`{WyCG`PSD5io&QH}wA
zL{7`Ld2lGrl$%<|hp!a@0WTl$lG6Lsdpm6CxiPdPM*9_zzRrcdo_)OS{)M8o8ZY(>
zpT<3lJBC$N$_k1eYfM28Snd1YUAlblO-i1W0_jgAQJnK66SKaROdz-ME9;46by2&!
z5gYKF!eiQWGGeCtam0mUd4z2r)6|h|qxS2V3^qekV>d!pQk~&#WQ4`2AW3Vx-h_}q
z#||!*T6v;IY0eEZh~<yxRNL`ieCbLgPbaj3WtLAQcv8y0loBMmJ`d?!qrfa&40snC
z>XeSFAn)$`qW|t!wcY%!$_kdKQ@jv8O}6CKE!r%ctj{m`w_eF<_Fksk&~h_IB?Ltr
zBx-x{ty#YK3}PPq3e}*|3`_KU#SGc5Y<iz^>$ndxfjVcD&d3HlTof=;>v>rX{9DDn
z2oI3*T)Lin|JeAB`y$z3Juy^8Iu7{><w5j4OKYyu<ssJ8T=&t#?5?3k-aU76(bIef
zC`HR~G<k@c#-SYvr<RUGIaZfZ%1?v_yE2KZ1vvF-J6V=s-p4NM{0sW$*-;$=>ts7#
zm2Hju)kj}Jf@U8|^xjY^qdrEio_I2sH1sj_2-z6-36Y1Yt$hg;G|i1W`YBC1|7fYk
zhv}8`&sSDK<m{Pr#TBcVRu_V#(Fd_&gUirBk%UV*Msrlmp1Guf+mNZBvUr|7u8p~%
zsbx61L6d>Cyv353pw8;`xBt{OvW`e&Nh3Oa*19?u9GAp58Eg;`y@TF}Yz-e&_lgc)
zd#N#MUUJ+l-v!};lU?J3NM>x8;kZ%32a^!^H*zxEIBdx9iNLslbfo}t8{8&@;JkZV
zwBY_J$o*{Su2XxsNeua|PYd$hDt0YZF!8$}6GbY{hIAjMgVaWK{#wT0JHD=!XLuJY
zg6FlfOG%vwJ1UPs#jlv^sl;VLkQY87M?86fXCtDFBFjZ1I(lZIC;4Vb&veB3ijJk{
zoM8Y#i=QuBGpezI_(Jwn&O=`M5VN7a%-SK)WcxF%@rBh*!xckZdvSdbWXf_6Yr!*R
zrF7zM@8MgRZTCv_LNU=XVvHE%fg`gGV%#)Yr%OA6J?FaxU9YePt~SnaM8TH7qV_Y4
za9fK&82S0Sew#}Uax^}dm#-DXh)t?5+`I3(3Di+o&nyR@)Q_ax4obp#F^0OgP?9_8
z(c0}0=bMdIuoJjt`%w3H?z5Z9XfCsC;@XAbR@+l-&(&u2`rehPf*dJBF~_HlsE=&Z
zE8gi99O}_^*+~Db-E6B4Jf;RPX|_p!<!{y(oNOo*CHBqHA<HghKN51H;vu@;nbGEl
zUDoC(6e!%bSd8pk-u4IAn>uVWt+^2NpK!pUkKciP#(G!GUq*23VRtXDkuHy3wEO>j
zue_#fW<r(iQE+0Gm7VItOzhN^FOZFY0<)eU1}bY`m~@;>793wV#kDI+9jnm8+8_CB
zBuKPRR~fL;A*?#q_p|3>v4Ttx2i>hVomK+&p`|Z=AbXY%W%dizgWx9@3|ogj$S*#q
z@2fwA!k8?nR*n&#VV)qQa&`&Niq%qabtmM6$;Q#J%~otu@kytNBy^9};$lc~s!TLE
z+LmzZ=&piQo&On7^(3?FNdjq$7qZtDxVi8yRNgTX^+|0!c(b40>oII`pl%?4_0;Iw
zo7w}>TcDx^C_M&hX+pvJbMaxsuKu?1!Ohs_D``2>%RT2rT;TCT`Z@a@LU8{hSP(=Z
z%=g|5K|q^0=vPK#&yU^*C(V!K-P=irBX8e>E-x=Gdd^*FeR2h)OeWq_7qIQl%Vy|P
z)XB8;%t9VbQSGpvrW`@$*S?Fbor48veaailHE-^Zo*ny)?589J8U)S1_sdr0YvfTP
z&$7=V?B^qty1fVuXF>c<6+y+qK(4W}ba*m7kEEijQg(S)5{i1f!)y)ktJqGU1qL|#
zceQ+$d%bJP;Ibd(!8S}|UUJKU?hM;+L^U~P<NI|<H&WKH88xaP=rrHI4kOl;kC|D7
zJoCM*ApJ<Q$u6E&cVXJ2kd22r%yGiE;r=6c_x+h=5YsIWC=07Rh^BOT=<GO5->(I|
zWW4KKu8;tA#isd@?Ram(iHZ8c@@akbitG}^Ircsl&#mhyyWT_dw3%Z6rR`Ezfl)3a
z6r^XXaynkGf>5dLO-cV_g8i>$zrI(@(8ip&&~gl%j0>I(%?U0~RES6{-PO~uuunHu
zH5}qW`tURP{Qf#^_IM`hF_n(D-c-G8+VXVhh-unxkTNv=5jW4Bac}d$S6eO7Qs&}m
z#+_D6+TG{I8?oGeZL}$qn%#AWpZg^f&kY2eAUt2Ar^`y8zVa#~1A&CL^ij*n-6l;Y
zv&(uv8m;Hw;Vw1S3AMi{2XZQug(z-SZr2V`vZ3pv4;ob6PqI5Xl%9Rwpyg?Lbeo6|
z3d%LvsURpX>8*u%5oY(@1t*lB@DMoMf7kdvdU`FpzGSIAtwy-Gv`(<C)R!i_;;IK>
z*hS=kwbpD@aBr#%j3k5Dh`n`0`BE9hm`1*WnlZ4Mo*M4{?!J0|w@57-bhh?bui~qN
zc3s>gp0l>16rVI1Rr2>kM!mA5`zx$pK<`9#z8}PUY4(s2%uP@boVW36O>#Xbiin=p
z)Ttwg@%!|9Y+_fGC;0CCLjq5WQ`sFs{1hed;q(f)oTvB;pCGR-(*+Utfu0Wix={bU
zEXzRQFNKtKjK8-Vwt6`lFDD+Azz#*~$hjJmCA1Amqa^bz9SX`OdI+`BWftCESWXR9
zOl`UAj4GrgbSXFY=x3g5zFMJakN0v81?{;{KGr_0GtJrxtmy6cjeR_Yw_L7&PxqO&
zz2U1&pI+*SGl7a6`nz65+OCKS#5bzC50Tre5e77di!Nl=GMtq_LE@piOE&7ith&fH
zG7fsB_o2T{%8m2FkUeJPkoz8H1lN59bI!}tPIqqXZ8xmoFKc5z+x4C(hd0&Xe3W~j
z8cU(@DGDQ42S*$?aS_!E+omkj?t6pu#@`%z&ca-F(#$#2E%G6do17YIu5mom2*dT1
zeGp+2$7(2)!Vp2GSj%PC7RF5wve^*8AFLYg0g1Ca>S4DY!ZNpS5NDWPBcNicbH%JB
zxlqtR^ey9!Xa|Z#t6n)PwwX4!2Kv0)5|*x`MpqnibsO)Hn)`YYZiGg>VyAx|hBpg<
z8VsrVkGNy9^T|<iUGb%_m)0F}<|-P~g$W~)x~taQFy)$foWM7=vv|T3Zr|uGYV-Xy
zDV!35Nj|)l8+r1&@Mv`a)`9EuO#p*j73%xxqy_KX;d;@D&j+!#RI(UgvUPIPq-(ll
zU2y={+YVoyrg4^P#%+%|nNa5>!4C4<@eA})qN$uUe)Z{wMdWLV1poK8T3@59Ey^6i
z^LB>BQpQJ(iU!LpaYrDvJNk3^g0(DP;;lYf5s&f~Q`z0yEu@-jaG&g3BpEzJ8Z8tC
zo;4?2>I~j>jZ#n20#})7!>zrfPlK&M#admRmHn04ly$gr(jV>CXm|Ke6GT4oENnV>
zL~TUa7F!~OL$=qHzWyKf-ZH4JW$XLJonXNo5_IA27TiNvKyY^p?(P<X2Mv}4hd@}k
zy9Ezq;TGK8-pStQ>~o&8&r|oU`|VcUREo-4%=GNpJ$iJ{?lFG<<bd&)3UeZphy7TG
zz1#~ZvuB8da7ma-CIm~W5XVn(rK=hQgm`pzHd(3acf*h{&M!V&DLn%ou|b~xCVI7u
zh4x?ISS=>}FysjDS?Kl!QEOrC7D-O~62HUqNdlI%thCJhPZq#G9N?c;<@=sIwY|DI
zy}d>=)vsoq4Rdh-vjSz}<uB1H^r2KX$Z~Mi)PywtsY-SOPmC8mE6D4#qsGfpFs|g{
zo5x}oXBXtv`y`$D?HMX#dCUDC`#U**?UQ|f($iAynJC4;sW^#;0wCQHFR!)AEcs>C
z0)~S!QR>oX2U5WXF~t)$#Z)8@+&-rUBrW(>j-PoifWV?lq}s5a<&eQDG7>^O(JQ70
zVbDXdf1{>uE5ZWu^(Fz-_}W_^PM;f7vDujg7v3$w{2uhR5}lO{v7Qi;7zwMzTKiQr
zFTwC!U=@ro%iJs*84?(Sk~-hS4<7nio{O1g2qB?4s^}2unEI+wK0J-Qb>!e&>gvL*
z|9}AthQO_n|FCvar-kI@8Fc$%6S_S|QiSZj=bS}a)a0nQ;x*(*T>9uydb{P|Db_=&
zFN|&;svAh89X}>C!Iu+bSm$>(@L@S0MvKR~eIIG#sE5s;qpJ5_nz^GVdf5O*3k9MI
zysI=uW;BVo;6S0471x8jE-|t$wvsN2+ga3?F1zwwVd1(`y;e75*1w!qf)qayeF<>j
zux3r)aFAbhBy&EL5Thzb7p9+tW~*0YB^yLE;XwtK6<#8pY!DoTGiQQBV*EDoXgd~A
zn^DfPp-f^P@)CxKv#9JH=13KC{1Y-bU$!7*RJoLHgmUI?pse4I$7Rbo!Gf|!Q3Ngl
z<J1ejtVo1xw-Dizf!Jk(KIy*I?G;BtG$|2R%BgEV&k6|Jfzb_~({vzn?v!8rZ6n`h
zOwJ{j*TvOBSj{pCf<aYGkR|!0@IjYwE!Xj_O2X7qhwFf3OW;e^k0xy)?(vN(_q(c@
z(mC6=O*rCG7g~DOFPm;kN;`!LS0XworzCc;9M~oT(FvfWspt3B^wy>m{=8caRb<dl
z4VA~-C3oM@uR5fa=2P{6;Ly!dNyXfKPwIWR3U4_L3`p^@G7qypTF>bO?LOY6!Y+k%
z(MYs?Sj_T;r6VcD(Me}Hs#}70|0RN+I)}{30M_X}{={!vu}P#Kt%OSj&Gb$kp*}eE
zyp?eUd$RMci{Fm@!i3oL`mYKt=C(r~vMZtJtTplctC4pG!j+K96|emlY6i`erbqIf
z%Wn^5Ja{x`!M&$$e+B#z!N4s{ZX!RZXcHM3uM>5)0il6ndnSg>cwFap(6Dn2;WXMI
zxQvdz55je<W-jz)%=4HaMwPWKu1tYC+pmjzt=e{l*PEi+dvA2|d16%C5}v_nyH>A$
z(w0TObu3GE!pr+X4p-|@*2?&nrtjS%5VOjp593N7cJ9$}E@^`rfuC_JRM8U2o9Ob|
z_VvjSu05#jBO>UBvo$`kvze&^6n_eFk@A4Fbwv=W%>8Sf7+nNJtYNE>nzqLInN_WV
zm6IEqFs)I(Ud?I%<-uR*szQso>-x?FA;RVTsM0W9?*0yz$}T!BCjs9eU`f9)QF--e
zKUZAln6}Ti8jBB2Y#kq}@pt7`yT~OtShc^I1hpb%hJI^cbXcjPP}Wy*B6nD-0{2eC
zjG1W$Qj_|d9A_*hcyEaIma-02y)6nzH0bJ08UfXPmD29`JX`(MOvr&i?l~)tcF_CY
zu5W0Rx#ELWqJ7!1s*PsaJ;e?xkZ=ziQnU#^TQD6Yk;!1JciAoV`&nN!Y37Gi!D8X0
zb?VKi#t`%ugJ0TUFDbQ4`yc`}Ui7436M2=n>gt!~K|WYugKigV`Y)zsaK!z}nEG~P
zZ_+@ur3dxGW60MXFXL`2D*Wi)nge)6V&_m|rCqQ~H$*2z1DZw1Yz`F!BZ4~Tk_B4m
z&wtS8G3}zbWqRkN(z_AYOOMPW>KTvc^xmomXrE8)c`+8}R=*~}I<9_;MCd0fL8^Cm
zRcPR8;3SsZD7d1>OJ3Ac>f)@~Y<ZU{e3BFE6)6j@JF0NxW9uF&9;vM?<BU4yK(Z-<
znl>Wc$St&w7%O4P_F9WJj31X<T~1tG?ImSi%k-1WmZATZ#`Ko%$2V4;-)0<;NE&`}
zeLb8vZmrr(^0Ku1zCSXI2UxS)r!L7PH5SCnq8*$AmFJ17rrn}TS4(bl&Wf*>xmE+X
z2w~iucj^;WVLrWhHcFi5v74*QWDXhv|E7RN5JM`EmOxj>s?IOT-ZMZeJ$!ZZI*ni*
z)-cBnHjN-Mt;ao4(18A`m)}c_%C~T~o@Si}qY+=H%MH?NLnq3)EH-^#{`r!YqPw>-
z@D{1N?@Vw>t!#;gB-qXQTBP-82r6B;T|j|<xgSWos-WW(GeDJByxmvSDX#Q|`nkpy
z&d159NKZB>-Bs~N;{8P3wA7bukMWQ}C<>JzHHF_gEhTEF;ak-;zbo?N0T#I~f7p@5
z$H?f(Vr5$fjf0S%dRiXNj9OVRJ#}O>R+{2hVQx%0)YxkZc?-O-)in$-UGO3BJ2kqW
zQd8j_Oplh7a=v1fM~*|@$shLk&AhcOpMO_4!qC{!&YnM7U*{<E@sg2I^~Na@$l)MT
zW>4Il>BrnqqI|xBP54$ik#jM3AO{{xm>t4W^6q-v`UXjtoxTMKG?-iNKjl!X;EKK#
zIEwvTL9h#9H;DAFHN8oH(Nm|TnMWCtM>e)4cRkh6Z`>K{PTvwfFlSMbbhMxnz|!*U
zQsv6dkq@Hj-SvXo`8A=bZg^?_nW*n3@tS}|mbBNgN@>nTsVdIw3H0-2`@-|X4i#W|
z(bJ;P8^2JL^aW{;WRX}}OG_!1y?2lM=&o@1YWS9N>EhqkpwvIU0l}g2^Qb-Zc@9pG
zH@l8!8TYbgm*##MY+`ml-y<KBDMH+()T=r^+i(r_yB~I3Uo9&MsRPGfg<zzdY-pwK
zEW6R=Mw9U=7q9s4j2Ke)Rl6p3P@*J^A~P>YseGOY=G#CT+u?!d@-IwL+a^Re8-Z{X
zNgqSPDiY~X#b46h47@dV=I;|yY`s}>7eu6iJ?MJgI=KzO&~}Sp&_gU^%vp{$o2j)P
zVbyz9l{*@=i`h;xMX9^V;%x~%J6=>pk*?1+-sG!>DTdadIX`RvuIor%`074bD7KMj
zuIaeW^CWrNF_`9`k>#V6BG(4iMJXf`8Jisb@o<8q;inHCmtu1Or{H;amEejXi|MPx
z{7TnXQ%sMYdgvaY)PfF}Z3*$tl17z0`Y}5qqRBBFMGcR{B)YFfV|K2(yc~(W%1a11
zxXzjW)2Q@W*r1PJ`$lpSxp5AdzED?cug#v9|5jwzNv+`TsL4RBX!OU01G&A}46^hI
zcSKoQ63<i=NjOd=w?37lo!QV9<%*JMpxUEp{4VrqT;U_KE_)vdg=su@8;ZXh_@TT|
zC`W2RVXb&SNQ)VE$cC<UVizDC#0<8Ijp3j3W<x$boRUo`o85J5v9N#5Sd5TgvB-4f
zs&)Yw9Nc<+(YWB$TD<f!V_asx&&Ad&Tx}c~D&*#msAa#chkok9Dc*s9)zr8iKNKyn
zRkW{&%<SF2CZJq8CL`n<D-w?2R5K_mk)Pv~Z!dYTB9##)U~j$9R66OiE%;YF-8(oD
zW*~Ssg|$uQs#pVzPxPW1R0Y|~hzA8TlG_Ukx&|&>I+mMjO^Zas+g(bKl+Bfls<!og
z4BbdW5eBhT?kW-TrG8rLpEv}=XXcILY#OkW#s+DZTFZ}_!i;Rb#kgy_#wsN*vwHr@
z=moZs(OCG01KCF7quxAXoC>j!rH;+6Ws0FG{@>kb&xVFA?4Jj}8ArrS?XCSf0@qA5
z)$8*vwlrktU`PltvDS(w+T1pqK{gBGFV`b{5EZyM%t>C_o&S<Qn0{lyno0zSBh>3^
zO7uVjhq+yE`Gcg`-qM>m>!eW7i2^wo^++?3Xl?`__cvp>PYs%Y{PZrmn8@$eb0--d
zJm6@}jc$Zj)s^k1A$3%WdpLHUsa3ywZ!#bd`e46oBmGpuS3UVf%rj{~95+Zv#8*~?
zVa6j=D0I59{d7Zgn$(kBy=(og5Uc}4Tuz!F%3y39)gMFptju&l%!5HUN9rWx%bT9&
zg9jpn!c}2Bd@QKpz92kxyuaSt1Es(PFfH$E2b3!79LSisw#x~qN7O<IZ&mml)i}3t
za5M01DE+9zDX@mj&e*OB(PhV4&{}aZ1A{2icid8iJ~lE_Q(|OZZHL$%VNAlgCMk};
zwA+#HYdT8oMqlDIH)P_g&!$)Xk3OZ!f-;WPujsWYk+a)w&66K>iV4@uaLYXdFk++a
ziPwT^Sqc_sH%dKoOWJf^YS_8-<E=hc3T4Hb1ao9Vs6=xzK4FxJw(|f7@WQ9HVcAH7
zf^i9BM9*C(!reb2>HqWmtRP!_3~lquv=@!polYBW)ToN}$0x>ToH(S~6vW8W**96V
z2PRwDXVXzv)^9Uljz;`~U!tTEM&4?Eg1kuPuD78cp~Be3K_Wnb8A)~0Dl%qo{TyQO
z9YZwr*od6bM{kQEvR#g<-KCpKlw6HQlFvn!o$MsO$eHN{Z6}AsPY~5>+Pflq1kna(
zsVNZq6$lMayUvsaOsE|kCw6s~Q`nLr=I}}iq5Vv7k-Rm?R3nIF$h9nE($>1I3T}@Z
z+j5+ZEoodyk{cc6(fTcRpZmP^=Rsd88GJ1h_1;JxcyiBY{qU^R(Fg8wTuI-Or8%f@
zS5m1*2(7zslKC@!N+a0tW~jy1?>QcX<)xr`B-8t+@6vrAgvK2WYeo4TFP3F7&rm+}
z@+b%eMd4z=y_PwX@}yf1Nsnsh#}1N50R?GLlNvXvIDt+L<JA!fIyHl82^WN&mV!`}
zYOavdFq>Ofz^6{Hgo{Nn!%%qBg9Y3&HFK@WR|Gl}?!-egL}8P9N4j6`gPF%(I$%0$
ze3)TeH3|N_(@>zIn*M_ywX1{-Dz1sRDhZ8SLqF(1px%F-O>`eG+b>G=HIw3}dPM1j
zRtnMu3B!!YD)H~{WqS|nA@>x6^M=rts+zA0;!T|t7~8LS9DX60;1(~`7H)A3Gy+r)
zzA1Uog<rwVS#7GAhFCQ7lJ<S)t4tuZ=udu=_$A6&nbCnzBA+W=^_|sBq8U|$ru6(k
z4iNiBia-Kt(SE()&@RawM`+7X*8P^4d!oa#ij?KNimL1#TUQ{&>iYZHgrX$|#amQY
zHJm|t=4@@y?T+9<dAESt){h`cv<|YINKGKBVHd&5F<n@(y!Pvoozl+=^Le$Zb}?I<
zM7eog;y2&qx%Q3bk=9ZR1}6kYNDp8q1`t<`YQ8y~@>N)#I0LEZwh($6lWa{Pdy$#K
zui}IE*uSeC><F;Wqe-Wo{kXKq+>RH4XjMOG$ue)zTyIf#6n_q<tka7C%^(ja=dkCq
z8vf-d`kKdJzXX&4vrZseVsW`X(}(7HT0_kW2|_uyp;G9P1Icgk+#;(w1%2`1N_2>#
zc%I5+zsph?RyP9f@#k3}n;2|g`DLwvX)$-&I|S#T_4c6I@l_+v;74VrNlMcqsgJ5q
zBPum5#8}daIB6UW=KywW*vog;(83sZdxlph++tINycfy>)e~X(c=wpXV@iR=&ZgA~
zjt@a7C>_?&$O$EzI$Uk?;Bh(cdw}|(^DBNrjHTkB!?S3#1;ed`U!XBj_a~pNuCZ2D
zY<CasT8=yXzN9k^O3a}H0@(FA{S1vuuaI!kJ`&wM_AgUmGGbdyH=_w>fQRBBM`(}j
zqvMK8l*U1zd^*rASh4jKNd4#EM%g@iM3-0aNN3}nfkC7M^CO{#Nc0#8(XD3_&rlsO
zcFNwDL?czx#vw}-;t$EKnle(;$PJIdB9o~JzTbkH(0{c3{kxuKcx^Lm<7j*yv2>8h
zILF6L=sj{MTIxHGQ_3Mz;$-GXaW=fX`>uDaDv>=Gk;?eTaSLUBWY4_%cyLo)IreJ{
zB0EOlLMHMPj6yTO7y0ev^39a82tH+#nNbNS)sb=Ab!FNmwu;y!F|C~q6tH=tmTtph
zhwNaQBLCC|3gZ%ujtos1X|b*d3U4coS){4DGkt}RRY%<>Ellqp<df)lk~y|5=e(N;
zg0%tA_T$bDZyU$8`q_Pa%DW76(XB~UxuFza6Oo@b;^`ZP-#CroV8<C+2+u=MLrcW7
zgz3M=;L`QWxo|yGitqMEx_s4M*f#FazArs4dj}kFOupk@&m4HP7$d@2tqFCNk(v57
z>DvV@5%)7Xg}U}o5_|z-@f^y<<IrQ@Ov$i+PT<B4iere$e`&7t7C2C?%f9ichzVdv
zbU6WMd3Kt31IfK^@;2A?CeU)UP3H;LixATmo9_xKH;F1`CuHhkRl=2M5Pn97iCDuE
z3kOqe=k`zqB~n?akL&7Z(sBIK3vSl=Wib323;1xRj>X)WBN5wg_!ML^J!-R<{w&9|
zS2tN!C<t{sKPDsSW2iMfd5?=uN_R<Ex>$U&-{*r~Pygu5))NXFH$5Gt+=st1uYSjh
ztD<d2Hh;CnFGB|c)>l$der)%kz1*KF%++ZfSuR{q)0;>*=MDL85fsl6tlVf;ts-nR
zS;pPsx~yC0+m%2^s?f8GCHOlJ<MC_POR7NjK%Wt0s)akg_d|UV=!!+AXE=Ejw8h22
zMENt^OQALR#y0%eU1Lrk{X=V}&f!8XIB`D2El!BASgm36XI{G&-^Vcfm^AgDxV>7p
zfHfw%ggX|DFRZs&Qvrsc`^Pp<`=hP-;5N63(DXPMqRNd(<{&Vga*S90hzO8~Lju;p
zp**nq*n$i|&)UPekDi%I4On!db()V0UY|{>R@ObR6lXT7<mjNbH`Qk-a{go=$WiH~
zG8)Yi_(tS*l^K<q%*ITf{>u8Lk+qz_q^ty#v9ywd;?<hT66~y`+k<CyG~0!7WI=k`
zs-JNna~iiHO7|0#M@gkCk^2W9V`eP%LB3T_s3tS)r5n`6{<25W?-SM=!t{039!EN~
zD%J`#$Q<<pckWdgo9N^^b$RI#WAuFaI;wHZyWUE_lsXR5I<}Q0YwM5cC=99U9IAy0
z^3U|nsQ725V^-_R??ixU$R0h(#yW>lw#+FLF@1F>RKB|;3T}Q0uq^E*8Y6-w{K)Ys
z#sggf2ILCG!wAoY!`9o>?3{o;qCCh+l=p||ZYWoIgx>I>oonV|LSw&8ph@5<OFa~K
zl#MIUcq~6Z#iP{cLd`#Nd@rHB1oL28A<vF|o%+7{(6f?DW^Ey5GwNPGnz;cYyWnTF
z8-oW!pY8hEx~A&L(q&n&UZ9B3{pjU&p0eU;MBw6JXu#b57s6)3$t$Z|PebN^4}t*f
z&eeZn<2%Ne94qHZF2Ve(Bqbs0W3CxC?TqI_l_Syc=&@5Lp;y=lJ9qE<_yf)gC=r*M
z#?xqK7dD58N+;CF^k^@ZX{&VdxcaDs>J$pr3@phfK`QsG07}jmg&m?0_Oe?S>1=!1
z_p0!_qXH3CF}Sog3Zv0f{D(;xCfEjYFZWCCRZ+g%?9Xb`5F@-wHlYI`iW2h>u58G8
zX<VTx@<%;D4dd2R{isv@nhs9oU1{sof|Bo{Q}<2790eK2J&L__^#}RH8lPU<l6Y;2
zZvrO1kKNqrX*)ZZnRvxB8z|#tA3B#@zXdLLAV>gVOT3Cwa`rs!Ob%7yZajDcBXH(n
zN92c1gG{AwE=c7R8(+hVYO8Ur7!ZDm$h51GhhQmb2dUL3pssq{41eH-icUGN>l^GL
zFNcUKlg8rgybidTvPRfR4~S1T&gfyQ&`rl!^*GW0E<ibEJfE9~^~Rz)_aqn*@4S1c
z4kgWvG@gcm_IdF+-q5n(ER)g){Dp`2V9*CsBWaqYagY2*0(dwB!B5h;Rs1+OV;gdr
zp@?jf#H1S&Im!E(?S)c}ThI1z;u~6-j!0)$Xu`aQ#6m|cW(G8mQ~bmLtGp}#O+>*B
zCN}C?evuFgw`J8UUJgg#s4cT!_v-)v{ifB>;jm|yJ?(|Vjam5VnnB)?M<j<miLKt>
zTKOtGUz-wDm`*UYHhu9?3a=uRxfwq7bC$#nxG^V{J5dQ`bW5pLtA5~MBxGqZy_q}v
z=1>!UHa_|TscI12KuIPU^`w@$F#Fw_AYGXi{M!19^Hldp{8!R3)?jc^xHQUs<c^Zc
zYFj2JD^fb)f<j99hfgBiETs0B2l;}|bvn%+3rk8Li;(22ibcMZYPlV&i{jBK0Wo;V
zw1)=&qCgzr{1)iQdiEUNTu`F=wC{t3;M;NpL06Qy9u<%?hq!~ej_<LkHtV=f!uE>X
z*7rGFmx-YMB!k;9Hs4RR8iOL+dxT-vkRUY<<?SVm!I%&PgtX;0VG;Ew<fHe8_E=&r
zk~<Sx8S2g%RJ<;h0*qP>bk*(@hICqr`&Gg=lvQG{E;K&<omlp+oxEC187KKM#!bSd
z@vm&o2N6QC8C;eGsx0@U%>2F`#WbMHzDE8!JNkW4mn=v&RtjXXR*?>_u*3>u$WLAb
zUE|tzZX~5hc83sUp#=<V*HP-o7}S2C7hSB@OwHC-BV@Vr%mtWV*>#uEVl<%Vd^z1?
zj$Cc6$KLVUfwIiD=`=sy6s4VRQ??7^8yFi*JLa*j(hH^H$kMs|5gzaf)%MO>)LLRr
z=2N6C_HAS-=tmmk+wxUr^5doo)+4g#EM{}s>}Z=xo!=CKn`>~XfkUqRfTYbRm*vPW
z1n}Tt;80&$I;XBi6Axv78v*;Ci0=1bDF1<9CMleeTdpV@im7%r7!kULuSHAcKYdIs
zN;fp&%b6k7LaU~#lOiVs=ZJ8dZh_EOdv~RmLQkieuAN^gs|R%iQfn}io^u%1B1_lH
zDkZ;ZaFnF#jYGe$CE`0zxb1kC*J*^l$!=mo!-8IxJ*W6Z38QEaTR`U;o1PV$eDZ>O
z#K(CxQa+!l*hwmq%^^^J!6?79LS;PMv2nW?^V$gRMfQ7zNqjNdhu$p)&dAjHTJ&b-
zd@AKkQuuv&b<l?J&u$p8?#nvA8oC*rdG7f$?)Pk?@_ocvxoFed*Uhr54eK9_lAlk;
z6D2&JaUQ2dQ7$4Ym5pB<#rDaI&l7^5ANwZ8uhOM<kD0-mcZrP_`CjF0RoG$S(hX_M
z$~!4*1v^pp3fIC`mrL~Li28Yd90AfX9<+~>Ig9s~TMV@-cW@=L3(8~8!)dP>QXTcl
zZvNs*<~<^KtFT+bXzoEKqRMvE(x_MUQtOM!aYVTyZQOq3_9}W$1o3B!AZ3{csZeJg
zj&pP?i0&IDA|a$RCB~Sa^qf{N%Cl&!ifr3aBmIby%8Df4Nw-YZPSuI&-gSE(l&BDB
zNrv8iM!AZ`%vu6B^s$;iv2ukAKK~O%JC*xWCO_srF^xv$6-&Z!wG}H`btFQXzPk>u
z*U7pbg4qPNF;$(Mg<ABlkf~_AjxZ0Y@y;|C9F=GXWo*Q%8C%OhT~`<BTLs-AqwFA&
zN1`*C71(MN33~lv!z}{Z8u^4~%BsRxdjZ+9vI3)zIluN?G7|JxO>2sVONfHPYd;?f
zy^@8bpc<K9%C@theJo3+VJWZaJ4bYQP6NM#3EnhMnwlyF7%t5nBq_U9ofYpiLt%oe
z@YlBFa#n`dyMffObPoBRbwme6>_Nu0WORbB>vc@3%>~vP6FyV~>U_I!;6CZL{>bVp
zKAF6(9o}M70N*z_6edjl{qyA-R}#6fDvDL1z<%0~iYhU2VtP9M*XV|;l6fX3!J2_6
zQ;>$dO1!WWhTwOssde4zJ;+GHx6*4!AjjQKrQ9RQP{mVueTnxk0`|7CCxfueM=eic
z0`UdU1zJ)s!bkjYpq}p4Z_G8o>7UIXo|5RB03M6W@qI4G!)d@t|4vW-Ej=BX6QxU2
zQ*viuun(2e1n`<a>9AbiQs)2p()#ig7!=x+gnrH%d^nF@0i02AE;~f_JOs71xz!Jx
z86nIGC4!Qf>L6x4HhlW8SD6o7`>$d%ceTwWNj!OXfT$|5!pqfp*Zi-0-t)j}A!>Jk
zZ)|6t)kIbzO)v-T9-81z)k2y36&XTeT7)ttc>k!Kw!QW3U49vPZwfjy)x?!&J?FTL
zZf}eAQ2Mg2X&g`RGP%HUjv4-I9o=4=h$~ha@^$*+2p<kuX(2**W;Lg=55*ykD!)d!
z!g>UN9atvH$HH&?QN29c?We8Tn|Nup#?Hg=w4Qv-Yw5iz@n_PW4wJr#gcFn2MkfsB
z^8V>m3SRCc^%Z#kBotU>i>sn;YNr_cY1Y<Qz^HDOIG}n(ho1g^Q<P0A%EeHETROSB
zwxwm@{}RJJpGTU6vXeJt_lz>zL~zXYz{HZfC43Cj0D7jfO=a95fn4tiH0!5t3sLR3
zrxgV#XH2kVpC^>pSGmH@RZ@BwX;9$@JRO8h{4$=oMg*f07+CvAE>@}RakEzy#+lJP
z>nKnt`_28(xC_>In+}(t$(p?QaDn%!+%8IV6%kGxl<C>s(^KDiUQXq@5+8#)%J8P!
z<%;yUXd$qN+<(nfc$-6+Z;2L{`b=~Wh-3DOe1#&jj}ZDJzp76tUtc}&<w<kF#98+k
zO5l}Z(E%+#RSCh81jg#Fp3R#L)Z-FJ8O82CxNLpERGoo#B$+xN=g=Kge{7qe9dniY
z7So0GyfDKSeUPAo{zNXJ+HQ^GJ`VLgqek%=&VDq=^Ltg9YGu#s4+9a6;^gyI1#u)5
zw9h6!^oAX-<a>sT>p#z2<>_huQQ6s|RxS3SIBsNg%}gzpzW8OYmJ(qp{L0c`c(alj
z!LC5k8@Hf$MFC00^!_T7a6WWG1=<vAubh;_O3vofKVRt!uMQ%v)Ui_W*X+DBz)Rp5
z^^9HWqb)VSmiy)ReRl#kfYh!d>sD=4TJb2oKHW0@xXCjXkBOt)T=u3H@-=GKX<cVa
zU)k^@D$2ZnS$~YZXg-s@?lCcA6BfbmZIR!nYYty^@|2n1*5avtH0(uGA}Ku78T<$s
zvgM6HOU_~SfdwZhj#im;$@B%D^ty?j=0>|v%=g9;{mq;`toHyeh_-ODw<w4gzaY<3
zMY#BSg}m#<>uJyUn@}S1O63F&%44CQ8V{O!6Se1CT;X~>@=A25U-`0;z)9V{PK7yY
z1pjta@$v^(u)vIwCi%CYU)yWGrbrM);xzom=(Qz2Gj15S=|fXQEfZNihj%W-i#&hD
zVHEYc7IZjv|N9MR`uifu#M%c676c+lQ3!Vq{X9HEen}vtb1?C0kGgmz$K(r~JwEvO
zD|Vo^ILR$-MHLfbVz@7-z<^OJzPd3LKX9spW>yFfi@vK=rnq}Ys38Md=_9GqZmou{
zCi14ZQ!@8jw5IKzgsZC4sg%6p9Sn_v0rRS}66(9E8$`&VcVg0b3&F;8<Qu*XEyx#F
zC#2spQ77_pdVv+m<hI9;?0~|4)v*793JHH9TFp_H5xI=5I$WlfrraDlDSVu9r?!Z~
za1~ftD#al-<%M7o$}w0KFJGpJ;!?I}wQ+6Q*<$r6?%e#Fyw7Uy>NkiaO)8}xHh6IH
z5aQiVE~Dw+PuoWAG-Dr3M9lFu{AxsG_exEsJ4Tb2ps%df_9c~Mkssy3cv-xf1Cg5A
zOb>UD>vDd4iobb~=!+z~!g}q5)unF5#`Loew^y2aJy$vluXNRU2e8*R<)BKw7nYOl
zt1Uwh2jNC4BCoW2XoOgXxZ8Zg8OOLvVZRiZC*}7Dv(I&$2J14W6ATHp$bz^aHE^Fs
z-?eyB(k)IvJPE5!@^p-{I$jv)H1;#Q!2;n6hgvrbWfm*VXN$)!b`crc;#Kwpo(}0&
zwnYcd&wwCrO);fH))fLMhcX%CT8Vq?_ZasgtMV~7_EjG}yAPL*PCH*eU+9!47Q<SU
z#WRwsPWSv8a9odyN;`X44{K3-YrhqJS>&kY+ShqimEu<&?Y!;y*BQnG&TrxlvZYMB
zH&U{7JCtA#O@e%BC5-4JFVZy{hcYZ(Ojo@sg6LVS?!<UUXIZyy*i;{^S9cuGkXW^s
z7$e)Q=+vH*;Y8r*4k1-KnatVO@|$~ct$I&4gp8!X;?F1wPdexIQel1~eayZ`jr{U;
zngbpRMreP4av3Q4^iB9)(PW(_j764SkvW=6Ch>b3vM;}S)18l>7z*L%WUzdP5Y&(a
z%VQp|7I{U~Q1#rOmQWE#yg{}Z3)~pVQWzB|r!174N=)#BBd7n<91io_vbP%4;XZtL
zBom&;SVPg}=>+j8$q<b8M&X*64H&*bxLI@UZ}qPUw(lp#B8{aL4OSyezj<DhcVv1H
zDPGTA!fT=HQ4WzXOHYg}6=rxacx|*Su-bxIsg{lJF_TR6J55Ll9U*HBjlpuKdUvrs
z&c5_6-JIn0aODV4jB)UBR`EFrL48A!O9tS%UW#e6(I|D&vl!kWK_0B8kEnX({lqaM
z++y)meCZ5PekdLAWMV`f1CiY;#45zp!F9Nv&(gWukv%_(<q$((aO0V-$O`qhm}Y2v
zUUiA{?0E%y;Jl-qR)qhcJ*wf(aZrn4`s>yG!5gQw)<SfNtN=~7R?=Vr>y-@i*FmC!
zZlRVusfuQUg*40r+7jajJ}Hfzx9K;!Wg#5?P~RU&>^fTJElO{PB+JmNSvd`sy&MVp
zk(2mpKNp4wpDM9ua2a=YR}FqEX<75hk(yttWr@F*_+fGKRYzZCQZAc+I@d8LrlwI=
zw|5cPCoNZEdP?x?aAKG-TrG_C62sSW=yq)ff@<vD2SaO2Ig_>ube0>8-esTlUW9Y2
zReJiT4_nn^(zcWfbM@4AP<p6t`y^lb2SBThA`^T25Mc1wG{>AdlyCl`VFH}cVaYcS
zCg3NOSCG0?dl;Ncshms4I8^eN-VY3~ONq>eNBz?vZAyXzjR{{`R+qNMr{wGmHd0>7
z^-F<DQbUNR;{jSX6mHlT8{eU!Js5pHMN7|^RmCj}fpR&sX_4K-$B3N<(r<NeISb7%
zU-SIc@Luaf4BaN!<&&MSlcOk(kwe3b3bztcCYHqr%kD7p$|g<yb>KXdX-W@N+t3WP
zv}Ts!qEySg{l-VA22V+7G~A=@Vmsv|1`u&3FlFnTxeUY19Lm#{dgxt8_)sydR}=|}
zhUpzfkgu)<%Ke>;C!qn~8~77}g<?blzVVmY>oj~yUwB>@MRS)Gda*7oos8QJKEmu_
zM17Z1PnSflWRYv?{cNWp$M?QY)qQ$aD9x+0PHxl-dReiCJR+}0h1;q=H(zg=s50Q~
zJTu?#R8myhap-lCxgLK8p|WM2N_Tus`hXKqpOcU-Ll`&AvTVOOoUvIEEh#!s8N+;>
zJXjFp9E3Cbx>Z<LPtngl!p|FC@B`bG%f;`4gmCXQy!V^=PRnbOroYA<Nj&It8QBO&
zBy}bI2T0u@*>}fecmV0ai;9xfJ^T`WV<{JV?Lw1dla)=f>G`gPkTpmVek}C1n0FD+
zxA@+b@1>tgk;tj??g-1=>5^xCw(rTnCbQt_A<_=KwX^dtJz=c#)U6qPWrQ5oyl`)5
zu8;bts5Sm-vHrTw^1X`43x7h*TRgwM*$)KuR(!c;m>YY+HlBoa6BcT!Uwnj1YsIE8
z=Z#OKD2pe>edJr{ROWqlD6%~1>qx9@y;LVDaFOlHRp0aY%`nne1fg#9a7pLPH}F}n
zH#j>BG_P$BIR`wro!qGr`oRrDm#NH_=wU|abvk_tqmj=uE`uBwZ5*_*RSH<nHElPO
zK75KTsVKk6aX+ioQ02*G7*GWaD7YP(t@n`qFXaT+sKB<PKPZVtZ*Yu=Yx$}>vQbkj
zK(w2Vgl{-~v(gY#CuI%G+!SwyD#-f}tA3b#QwNpEaXo5F#<>{;z-ML>_|=&nYnijm
zEZv?@oNJq!oC^Tj2U$XyNUzi0VqRe+m=o;kGN(qHAAkD*gRYml*~iZ}XO@JaMOltO
zI7s5#ae@I+oB7^j$9%~ogp6)mwy!Zuc(saoK2py7R53x?c7=#3V0LpKl&l904;_>r
zzB1y+fdG;RT_xpLtmaai7T^oknuvQuATv#;A#!L)1)iB$2NP$lyD`guYqzp~v$tdt
zIB)uE&Upu?o%=`bRQneir-$pU2v}6b$T`Z6a$sX^oCRIcB_^n78r0K^iKJMoi%%DG
ze75B{(Zgw1&p(L#2BYjqd9dc}0ihdDPp{CKHop;3Ga#QFXSPvV&nYOR?Uet$`3Jaq
zR)_1^*AzN3@k@uEB6iL}uj@X&WlpRCI+oldi|ja2#d{a-Jo4rKwj0P?_Q#>c6~KSa
z71Q8gnd-2Rl;)fK;u9raiPLV$f55cLrFpKW$bMM&LDL&KCa{tW!+*@n!jq3HQGW(2
z_R?eMaT%F1VE<Ualq~`Pk5;hEYH*0g3nyGTB_u7M@i;Gat$+XPEU*0sHo#uyhPr71
zE{ysGyZhcx_Wq6px>n6<qJ$6%P?_J?cTzh<WYy!(MLBWiN)aN%W00<}Ua9fom?rh}
z=h2L`CO4>h|7zh$eSNjY;=mn7^vf=(#r<$PW{cKcHG}h|T`<Wms*g7N+yt;2dhzR2
zP=@TPtH;na04}u<;|KAVK39(GE`=b$acJ5~L>%oJkWP5m)KZU_g!q&bek4RT8T_M<
z&L$H1X)U^s2f12}`G&9cmhsBP!sTZ57SzkoA^hQob1feKS}d33QgoE*g{kLJ2QqWw
zk%L<ruW^x~uOoHMw<z1s%)Ljd{+WkjLE|#9Ntg>d<eiTHTHi`ZiHi=mSmpzn4D8-0
z%XqE;?h^eWzz>hV382O918NF9OM6V>?Z}&j@=3vlgK~(u{a3o~<4|9CHlgFfB)k1F
zKh_6m3T-RTlNibqNppCiX8kft*z4U<Mn=H+DssRE!sFIeat~pRqqm(yhSil!Jgf@e
zB7o=uuTF~OaFN7c>GoJ&0XULuOAkPB=b2k)*y&wU3D%YOf=y}p9srHP{Jh*m(00_S
zp1m^mT?|&N>Pjt}!!%sI9`%Y^bm%^at!bFENPW#CRWM%cfv7*han+)=!Thd-S=RSQ
zqm_f`HTE5;i^d&G*RS%+k2hWW3N&tbIKQdk_EeXH4S%|HgbAN2TPx<$*5~b)V4YG(
zekqLSglk=K{dpMZ!El>(Ra|168_kJvrJu`d|B>AL_otQYut(Rss>4J7zRO=8OB0sm
zpq#60^rJs0%=b!Ap4ekP=f)UL(@Lzn&VTLAV_*U>WvMQGk0<>X8aR|O42PfCc7VPQ
z$E7AQZ?1_`=cb-rUPOC4QmC^A3qmh#Jbx!^+ub{tm_;gHcMWZ`Z|wDfXr%H*nZm)?
z&dJL3zc4W{=1`KZenJz+dIFui05-8lt(PW8g~3B++_;LsMsn4-?Qp~`Iu_hF-MhX@
zg#<hP!Ek+y3R7*XPhn47TnPpXBj{4QOkD(}KSboH9p1Pqj_UAXx~n*cxvWYu_of5M
z90-n0<KRy)27t>bZdsNEzdBfcjrbi!#7(BRQ$s4Rj<Pr;@N>HyJ}X}0{Yyg0o*m_f
zR=#RC(h<??qX!GFWwwCpq0NIgom=h)F~=dm06w(gB{%<O0+-Zdne7H<|6xThnhmJc
z{Fzn5-i{F7TlGg%VF!Jp4kX2Nl<AVmxbmTcSY_=4DnHk+wd;kk99T@y908-X8{wJ7
zXruH!>L4a1aA#hY#e3Y|mQgEgBYoukX%rHF@9Z9<%6G7aM_??a#Jd~~6}uTvu(;~1
zJgFbyw7F|*`mtD5{jzn;1BkNmzh-6I%y_x#P@dBO$@-j7d-FmqJoiWpU{yE75KL^~
ziD72;e)wGJw{I&2o++f{16{3J5__B74XFBa+T~qz<EgH1vu)}J0NM#=8LGpGmpTx`
zW7Ngv*Ko5+Za%vGFJo**t?1gh)|j(`hw>vH&xeops;a#?j<PFPpM)RM+6weQV^9*2
zUx0j4NKEP~k6XYgMKAd>ux*@noTGS4aK<wiA^sO|!RR+?e<xIS^J!ccM8Kr3LS~$d
zSwSH^aD=5Z0AG9ppz=n*m;f|tlo`pUgGSq{TYx-{LC^`AX{RB#)k|&kpfy5_9YVVd
zB2WG4v6gd8@kGRaLj7SVCa-xE9QWf|6S{^dl9fSBkAlkW0fDMbc!Mq*GQ|MGPqFy0
zta@Rb1qRijnMXO7&GbOdk-%nam?<~B`&ktEsWMlHEjehm%D>Ab%X?pR(7q@I%z4@2
z_;k$sMPoY-K$G7ekrONcpflCk9gmOU0AM-I7{3SD`LhzXU6if?o|+Gmr309u)Qz(-
z7dH5M+wWWl!8;pfAL9^ZiIA#RqE@O_xAiiZWsj9^aHhJyvw2ogfBP^QCKU=<ed6r}
zlM3b#<{)1hI{-wkT$b&MYZPHs(PJ}!TV?4KvE|>}by=p?T~5M0qFr7G#8pKcX5RPt
z`56**+2cB|yhBWLM9`~Trh7ka)g1B1Lg9~qfbbju@cM}jz6kh~7lYt*9>`|Bbi@_;
zv31}z_9mc9^uT^)%pQ^+-|q>CH6X8GC*JKIbooAMbKt28k3T`HD~gCHH?Ca#0ojj6
z_c2e}`bI1~1dss5KI}v5oUPMz2oc477%l7CZgvpwfiHH6A7b#6E|c*9Z$4Yl%NfZ1
z9Kcn2B9l>aXQA8>0+TOw4i^JGJH!$<(g#qCxsZ4x%dHyXXl0vwzh`N`;5LhSp!a20
z3djHPb_p2=mKq*b<lWQ19W$?7Jie>)zNZ9^Wr70mj;3F4mdXcP#h0SMC7IVsDQ%`8
zqHdbMIa{!z3KSnbT_sp_2Sj#=Dkf1{2N}<9eGzr8NckA|E`uQCOVW{RO6+Tm38q4<
zT>vG75YWgvCGQfUN_yKo_M4Kbb!~nF{sEe*>=cHuE%w)nqymsDThxR$W&k)yHC%DI
zc1ir)DU8@B3i;N+AMiUv4|TGDNd#U$LBLo-rm-GU#ss#=TnJxw%yNcgIv^t1qSTGo
zyVM^H`MIBryujjLhLKXztWtBtepk&a`!id>#Uj1G-3ElI{n2zFPOZlb9!?<BzoO+r
zP!y8GCE}(gs-Xqs;nWQl=h?wr0kdBiw^57xUUlF6lOocD5AAa@fSuVv({dKd>KpYw
zLM><-G~*ZPuf1aI1`K0|@`K-!U^jFeB5XLjnPD^HBokpS1l}4MY_r#updSt#USwLY
ztKWsBknKfaBX-s7o@JxNBEU7{eSK^`{OOw1ZX?R8l|X7=!zg^{0FJ6A!6`kN2J5o;
z)&tXjeO%DxPu8YA6M*!e%>oD{^%k$3VX=1_b<XR5uOt8bZ)jy<0F{cIP;LSQBM@02
zXoF*@0`nqu33#oegATEYaLBOts`|6}olUT}oz^;gqYr!pBT++ERA`F}JnyC}{mtzJ
zNy;;$+psp5FxJ5U#T;`Lw?9(p08bcY6Og=;yBcoKya|~mnXy1;g&1k)?>{Jq2$Aw4
zZ8SL`fqpsL2LHUXU3#;({W%9MNgQI)0H|#+X@QJHBG`en=r2-w&mMOTALl8&e^cw4
zr)b*k{j8SC1}zvOgue{p#~D7AG%|mFKvM!Ts&?=@8KCc4C)uJ+KXq8U?!*5Oj1ub#
zX9tOri+55n3Pv4r2FKuFL$;Ae`Idac(Rrbj)X5$@1hJPVwbqld_;IalapeHQ*Eo#i
zU|uke>&2T#K<w(zqjC(iCt-tN%|t-h07Di|JL<50>47z}&H75-Lcksa!|aIgp@_|o
zQG17)bNK^8=?Ngp!-V04JxZ+C)A6Pm+&8Wm=-q{&$dt}?mHi}MfcgBDe`=P-XCHNx
zuCF7f2e=VMyV?g&lmFSvKy<oA!|_fsoFhQdAq#AFCSBzCiVFLQ*dMZj&?%5K-%9Q7
zq?ih*Fi&n9IUDdCPq8Y{)!_7b5!%+hX*zRl{YR}`(*Tcia5A$1yc2}Khk2l8dpg!C
zzaQorJDVHE9a;g}dUO<66TVlt%l@Xm3Zx~47`piX8jyCqP9y7-oGSk_ivI??-EEdS
zg3KfU7+fLD44@Gm#+O6WqF(^@C<HjWcICDw{NVQ|t2a(^0Z%jj6<-hu11O=CQUIR;
zRX%OlwGiB;Nl;i18e1I=9Gt&VW$O;p?MWN7rYezE(fJs5gwu_8Sr5>5kwIiL|5Mc?
z&jW=j^=XY<MrVAw2v7|{kno*sGKQk3gm{SBTE>%9SBMagZ_HX-t~}jer^K*0fq)v&
zG@!g1q%N@xtQUEBce{Cdo>0-e#~=%`S**6>q6VM6>WGna4q8Z{_!~>3x{Xv_q_(Xg
zmU_Ko&U==BEH0#E0tXCu>9a)r8dri*shv7QQAM_ct^28VdJCdq9&YEBlcHV%WX~+@
z8~^!r-#41@=HPz>fY|`zAeB7@Pns?=Xt2H-CK6`=6aE*VdqK*31R5fB(D@14Ur$nX
zxbgzlb=8k!iwk{k^e>#r9`l4za6d|17@0pz<7&-;?RT;(t|P1o?BA=6j>sHW9Vo@P
zev}$Fddr_y=fJ02rmT<kGwJ)sUE}W!KkYuy%_(&KhLzXAg%#?{Je<ys>24}`N5MLJ
zThAlXcDi=s7f?AA?hhHgd|PXUBa+<o^#~3Y0b87E9&X9^H16c|R8vc9OJNS;TyW4p
z|Jimac-<<k_Z}xwK1fyUe(GLr8_=3@Cd~w7g@zoA5&U{g-!(FzJ(U0x^aTYwT{bK`
zlO|BS5D|3rJr4?AowKk4j`4q2Bj_p{=<1dN2jnH?<s<s0nOt+SZd=0{E;6KP6v!(t
z;83UnVc>9ZJ%PYF=;dd>+lz*KUf^ZxlX`*7@ja8|1GHBx2+&E8-q_fvq2$#E6bx=u
zXP2-Zc^t9=uq1o_`0prG*u1cCr&Z&e>~JXE(d__xbr0?&JRB)o&`p7W4(h#IZpkvn
ztC-1SoBG{x>7b%IXA2)6e{clYbDljy`m+w>*xIeW4YzNS{VJQbi!*DzPA&8SUB_{T
zFdrK1Q@D#u@!M`AP0bYDSBXhUYZqTWZ5Rxzk|WXp)d%jc4%|QToeLKNc4TefIgMe%
zyCOjElc%n}zUh~0Ca-eO%a@k=%4VOW+}!F?S#&om+JJEg6)S!)?5Eh?TJ`_)Inu?|
z)f3B#Lj9NqMaj+Y)48~~IwW9%C=lN2>go%nYL<c<Jcb4bxqU9|Fv$gLBT5Crp6(7*
z;h0I>h@z5f%!L-49*!Ws4S-k+`rkox^&JwjQ&Ts=F_pUY6|!OCR8+fO2HM)cTODrB
zwoA&&rk?Z>*UqzE!TezjByiscKHDoiNzXJLI)6v*1B67@dGBiDFj)|k6cnOmtf6>8
z>CgE+kMhum%UNQ`pD?$#5|e?j@KaM$>;6BJKYYMVs?&J>XRT0PsRXtQzng)g!Xo5T
z8D)AtG}PQx*4O{I1CYx1yI&pJ>QlT~SXj`ILKIY0X92i1h&N<MrW^eayJA59NuQ5~
zh6Y_E11l&i$K3`U0cs*&0q(2#TH_7puLp866sn_$_6q$b?tlm1`t;7xpP#v&#2)SE
zYNQoYLF0gP)0{YG?3M^Yk_i7tOF4spmg-9yE1><O=bnCCF&|yF`}W2XS0r3V<1<hF
z{P|^d)ucXnW{JM=9fG%3UTf<j;^udq3UVy5yQ*7+m2gZ_Gt#ZM0Lj8jMn>jvF?}2u
z2Tu*y&I}9ZQdt~hg7fF+r!TLh+IiV)ogD9P&MP9kOic1`x!+h?e%`frxE}^&3~3<P
z;>!VWGhM!m&0e1a-kY(+;Ns%Co~+9QO0ODBEG(K}aH<e8$v;}OyAmeI>uyv;HU1yZ
zgF?03&<@1G?Mb$Z_*_8fiFs_2Ek-_WysgtVw0MSq{hPH4Fl81!h#Ejp5A#{u%ygUq
z#`0GninD>&ccF3<|DPon03{2L3Xa9Xz;Q7CfI`b!Ffj$)&%S`x`$tD-Hqs0!@apE7
zWPa&6?9P-kv9Zlo8F!rlFW)_{?4Jh#wdrnZU+Idd3wZR`ckaP=2zYh@jl2}Y5q=ay
z{khZt1UMqsb5!hGcRg}FzUaOE)8f0+fSdY`@TtZCHw78pb8U@g@7|2gi!B0PTL2jJ
z(sd^2gF=FWX8@biwj&sk^+kQan|N{|Fs0v_`WxG|&3|}+zd93Gdpe?Hr(b?&Sfd8M
z+@dyW;YtHWoTT;b35RKKzxUXD<+vd5$L+V;x0y1Xg26!r!Yo}Df^^j^{_%3&1h74n
z@DIoKKYpM@02SqbLyQM}%P$+0>whEvq*Xgpp}*V8$;YQ<Zf;J(VbU!S@4AwvyPHBQ
z8*laNyFDvIeJkYqgfJ0CV+RZSlh)B9K<GwqBc^>CwEt0cju(JsN+n4!0{)~8jzDd8
zdb(kTzSs%Sa<V!5J^nNV_!a#8T6FgvGEL_0ztz+vCnkOZYB7ws6%`f5%3gmd@DDru
z!x;bk5!4D4yO?H?WGMlB;Vog*6OHe^5=NMul~vZ+DGBIo>W3g<PQ0g<nwp<yLLdID
zl#xF_KW_jbmHIDrtfmF(DC8pX5~!mkT#&dp@B)zuT>kvGjR)?Yx4pe8(lfFz9U;if
zjEsy-G=REqi$TD>TO6gB0nj9_unqVhva?xi$0Us?Qmj{s|Dy$fLSX?%ibCb`<N@;N
zgQOy!ZMR%5PHqW_Y5B-D2?`0BX9hg=zJP$h;PCL?_E5h20U!+XzRs4KJz@U2lj)P^
z_@5p34d}S^AYveK5QuUWH#IGI>U1qbKTV&@0};QuECH8VqYk+G!Mm}fz68qIAb4a~
zKm%4xRsX1R=AHS^D&vZ21%?ou^oaa!p1;+187RGB?3?U9FbwMg>kI|l_tIy|b+_Q&
zq4!KqPHI-0MLpd2)9%bxE^TkwT3M+9Gswl^Qos!~?4g(U1jNWV4(yvrl0>%)zo284
z{GU|^o+JSgWmqIfSV0yhsFm92zB48|dTL@~^(K^U8*LbfE<JF-($doU49u@Kmj??A
zqlW!4$;r=wxxDFzaI!PXu$1~L$={mfNGGrzx;0nej(~k+(C(f33(ZA$`yP<OYnomp
zNKc>;Y5es~M=15<M-m<$9yg%?YPK>$F9U;|Ur?_9yE2ynKRrH}!mO#T-WM9}0t18e
zPX3j|o0XiRsw%Y@Lbe$ocX(!h4lXDVFdehDi<!jt@AF{cU#EQdKsNCnVUwD}8RtJo
z2Z|@)xDSrk)dc~E%~s{OB$|?zX4@5xslRd#Gm^<ScAo}3xeJ!85rvC}q{QEP>c<cq
z!e$G`pcC+$_68KYrB?kX2`cvk=vP1nO(uqbWdeu-95s)O)K@g3;IMHB{o4Rl$E9LV
zz!k2S2D&M@BX7}j4YNMO($=;Vm^uoI6w`KqX;Tb-t2+ws=D3e?dT!3E*JsP8llOjj
zaPYOax3`=Wc6vm2xAfoYwp<NZQD2hmHef~dfnn>VT9q?^sKWup3&68zYHB`t7GkW6
zX)VL;49lEE`-eGQSsR;@=;&xyVBHkX{UXRqPD$_Y6{Y|jp^vavGB8*8qjU4}meka|
z2J{H>ecn$@yk;?M)gv%_y#G>K>cTwkakzMYSwDJPSI%BpQ2~~hmzOjBzo~6|j)>12
zFu0gS_d5d@*QEW$^-t3TMr!JnSmaxvIW#pi3W|%1xfX?%@CF7328V_w0RXua35}4D
zF4zT_3;{y!le>A}8XxnYBNs4#Lj;szzYNBmS1>`M^i`&P&t`GWXDjqc=I~|M4(5+e
z0pPr*k53ce&7le|=ym(QfB&9Jq>XR`aK{ENRYv|t%N^90?Dfj>_L{zr5BEIItFP=A
z8tZ0ljH&)MD#WP>fV-I*6MTV%MOb~k4DgLSn45voPyz_GC<CkDT8D-Wf^x&Q#d5&_
zU?ZK!>%{Ek%a_s%eUf+7QsMo)ly!OcH@lT-i1!3JF8{@M`bh`2H&+h77y~BDMO0uT
z61fo=!HwW}%5eY!T6%ZqiwFm6KVLf#HZijV1!u_$u{%0A%#>)B*1B%LMhubIvSXes
zU-Fjl51K$K_Oev%_?saGimL)vF7nms8Q|?A1I2;a!eT6^-2T`1hSp_kYiri`ql1GJ
zz}C{I;9xj&GqbawlfuxA79j2D2GD&u2R4qjmpejitgY2;ZFjeSUF^<|)wuvKkcwvI
zZ^N-DpK$w%8lP1}{4m>|W|RNuF<(3?_DlQTFTnJ96Wd<d_E0iis%5*<8CGH05Vi)$
zxM(;#R|Bd{%iigiwXjw%V9LVdNckGyL%}V%`&Bn(EHPnsMMXs)#R4AYST}_K<^iaH
zbvVK%hnV#9NEkRj&1GPDS3!RTIBY%L)Y6JZg^Y|01$p@>Z%3zHO@CI>*tbp&q{(mh
zfmPyvPPT!SKtZjeWNx&9;<-3g%TRLtPcD+VUQ84Se|swIBUQj%wh0uN{O9xj^YMe&
zfj6KC*c9r2TlM|VS?KPmWgJJp5dFRMKQE-h04iiGZZ7w~-2?U!FwXl-E+(13xBH*<
zKnVdRIwCb<-2d$!?P9><^l>vb`hTrvClxTeBfSa#^KIb2-01%^P0=0z3j_1`*iB=8
z_D>ScUN&lT-Sp2#E%^*4sqh!7Q3|iOl~2+#BA-iUFsYl_ycs{@e*fkJ?joP{R9Pg2
znJ3em@O&*M$KzLLG?S$grw@ipJ$3p%*|)8iPC4JkM13A-wp>PA08u3$%r+kVSx5cd
zw~Do(7l_nU2%&-4*l?2KC@8T1^#fjfJ1)&IeKX>htN;C?KY#Nc2iv$J{Qt2;$#!hu
zI*b4Ry8mqU{}=5-#DxIpsLr^K3Tzbuq7UB-%o4nYO(y^u&<}v9dUpKnLzqWJ^U0tq
zpsjUN?yC>P#7^ig-OSViZqx%hOZZ>A9O%thQaGC7zH%8&3g8MlivL5|TZdKEeruzo
zq%;eVZjc70TRKEUO1ee58$>};x*I8#?v!p21SA*TNC_;u&a>YA{q}eEzFyCDo&R25
z=e6dX&wR#s#<<5l#vKUPeAurV;@kn)++MpWs8J~cDRmd%{`~v=TIT_XZw>Hs6?l(R
z6DU{<`!9i~8Na(e%uWK*@=@1Ubbr}rflkjPgDx8;zYNmDui!?h;0?K11R8ZBe>{Qe
zWp<SC?=P^6K8e<~)&QZSZbH7?yY!_SCJvy0yaVQnD{%SkC{Mgp=Gchde*CvvRLX)|
z%xINjGNA;uGtfJNC(SPh8sFAn()8gwliSm2HR`)RArpn^_H#~4ZN5NDS_x=;UXO#v
zNHzqjU4?L=6bx-n%)ePc3JQu$ikdV>aU{4xg#yWA-bvnlpbB6UMR5HZz0f6~18NPF
zP*nk<CqD<4>*SUo1;`mO?$<B*?*MsIH{s)c`UxOvn?iN<GbAZ`r}=MI*$ZTyH_dcQ
zA18n-;-lK(`lqfJD<MVEetTxtKu%r}#E6!4f@sO5b#srGvIQI#iBfQc)?@ang3sG7
ze>fCwi5+@5_wk}KF8uEu_<y`WFDU75uDz~)oD6P3_^BNXcSu;g9Puc0@LeO{k>&DM
zhbyU7Nt^(UQ~{t<<h`@}0rxpv4h>C?;-+!I9a;dKj)>TA^=Kt55>6E$uI1gw_y4g6
z<iH0KoGu%Le;^sUpKoi&pHAz?^8-?eD=1+y&R0oBtr>1;w_Ry^_dlbGI|Idx8o(AQ
z0{ZVwpT?-~FZy5~1zKngfbQCkX1TZK)v%R#JY1*t?C|PuekRaK9=`LYD+ZatUAt-D
z5erSN;BBgTb`t)xaSe(oxC8u!*M|A{sDRXyoA`|j=mfS|I|0gcT+|nr6WMo9t_7Fu
z>KDf8sg6Sb|J+IQ64bepX<DYih!j^)kSPR!frK9rNE<=(mi_`5|8qO^)Rzi#ApeG5
zPyjugSxqc7|18k?zVo!)OV#`52dZCp2GoD^+}sD~q1@ExkF+kq6)A!!cd#Of0uV2W
z!O2h7T@M9RKo8{ex+nL)wcFA-`f0n9fsNLH<UAj!*3yezbl^U3a0)aIdBUvkxe~mi
zJL;|V8q`~va3qLz{{5Ip(TDGU`-S3lxbI*^qPS=TxMZQx%__-`1(%PmFc^0?j!b>g
z?z&$)U5Jb$NCU4jG_E&T2!Sg7Nk^7WLEDb1gA!aKZ(NStF5$WG4g=+dP~rZw9s;C)
zwYuNnurZAO<c|;k=V2ifc!AP&Qd8SvoP5X$=Sp<1qRUwb?i*q7PKF->Mw{pJE^6Mt
zHL}xys_{?Xc4qS{plFo<Z>71x>AnIAAA(zh<=9w=|Lzlwg~6g`<9K^(!<T^D6n)-p
zdSPeYr7fkj2Y|jCu3$t+etpXZhfp%d^?l;W=xCX?KdAFr*0WQfHvx&{t!L5_U}+11
zF$hxvTH}tLgOA>85U0zPxDH1B&Dt$N*=v(8qTG}O{+`&xi2B78ARM3lMuBy#p(gf}
z`eG-wT@`C;4X!WS8mtFt<<hC3p#%ul4$ycV;M6w+_2;63q}PK+6Po}2@qmd3zC%9=
z`(jD_5*<BIQ3_6OQ1)KfHB9CNl{w!xTfje^)&MlUJ_4=*RRq_^p>+KzwF3-K{!Alk
zDCr-+091s1fypI~gS;;l&QmxO`MdEcN&(}`|5Y={3XE@t9&Yk*m2g{ie=s$~eDwxY
z9iA(*SIWrG3e6Cnv%%>K699<KG^n{*XO<YqzbrPo{cUS~2mWYWqxAb{7yL^E)6PIc
z)9Fj_R@Z_EN!pkWuK{E+;~s6-3$RX-R_@Y@PRvvyxCs{~+yZDu;eu1*7B+F>0GK<5
zSKxgz>_NX4EEm_0<)6cEEZQHIaPIo-Z|k@g1=i82+$2#AtYeteGMG2;Jp1Zb(Kz^z
ztgMS}=E70G1D738iQ7pyUII*j9UxOKbJZ^|9_KjrJ`pa?oqYf>Bv&9DR0#LpV%t)<
zX~$Xi|MF)4L~X|NzzW0<o|YVAfIZ2BzWi|mzN<iEdUch9^G$L&s23(~{A1^5zyZMR
zqc$L5*o-Q^bfoeH*Fr-L^*GwU?{{FwJn$Hx!CCVOJO;q}Ck7}53gtU+6|vWVz%T)q
z<bo}uNi+}y0p%E;ec<$WfY`Oo{awqJ7#F1<Y|I+e=<Sv!{(27;0&oxJuXUPQ;2ymY
zME!3WfW>f`qlS42oR*jYkg>G}41;g%CqNZ79!F%a;^rJ^sf$Tecj1Vb7rV?ivix;z
z2;gfIR$ImkOqyE@-En4%%qk-m_ZVIkP8>R$=@aLNpT6)YKn?IpoGrXa*mYvE19g}w
zfcaR1eW(I2wQZa<fj#=KLGu4RLT2b4CsYLrKJm}7ECA2FUnRbbh^wIHGdD3Vg+`fB
z2j&c&-CnHJx82&!xr~ZcFCAMX=`v+H53*W=`=}^$Hk$jYWC#E2o&Nc{!2hIt(~<&L
zkZPkimcLg!=3S1m0HSpYa5c3{jk_od;kyy1flk`&XrKAD@h+;fia{$kjlXTz_;+Bt
zDt@s@RR7x*|M9#{bTpkUY%<<|%@Y1|Arm5SSMe`~9_Rhdq5WqiT|rRhr5y_U%j?qo
z_mZn1!M8|mlBV3>HtK(Nz&5CUpOoS4{KbR+--rJn;Tm?3WbvdDoIBHWI;A&(O!em-
z+}FW<=G&YGSlnM<q<xzn_!ii%f)L!dK=2wsA#36=;Qku+!@-{qE~sb?C~RsVM)L4J
zj?WgwYE+N=&pQ2Q75?++ct~B6z#($K(0T+e874yk9-|co{O(hDePRb_pDV-9MuPQA
zP{hc@0fQ7^2}Pn7BMGu(#PEaR|NhSZ>yK?*@G)*z3X0%6MgzsM%^B{!L150G>2uJy
z19!22Y2eaa1NXcWKrlE1-Y9k^=J%Jd5Ii&~7dYsnwf(+50r$av*ABNz2*>LL2&aC^
zb5S6U2FN>VVH>3mel|?QTL9|Z5cpe=3I8=*_^-)}eItVZT4+1m|JgUJe5?TjY^=`N
zqJ7utC%{0IR5O6(?2g0L>~Fk4lsOP|3H)X^qQ<z*|Lbl4*F)#QeX)8u?FQTzlgTav
z_)Rt0(A#<rgkBSaY~>knd#nK!^+J(@271DEiVMXVFqNtr*9DxnuPMXz|9wsIkgDOk
zOgGpl4DMDi(Qj@K1Jtk}?EPro;%j813UD*7W0wFzkP|=#D!rPe)0gP2lGp!wy8nLe
z_F-UJzLrOtd*FW!tD`vf&~^GoX=PP${I>!8&u<$D8Wl{$X&k*(!FwIS182|aawQPL
z{;MGcJ_akSo%`7b{JM%L+>lx^&=LOKkc^SwzZ{rYMhD*PCneZB3q0e5|NW%@>-R|o
z;NMQ~(>GCY1v>=%!37~D4*Q$6Y!ZQ+Z@qG3tp=BaAPax9OZxTB|9rLoYq!R(03rb0
zCb=2l3T|n5IGa28LhIk(5nRClZjX!ch7shD?eo&c2^c5o4y{54E<c{`z!??7U%32_
zBH;0Z*e};ZxMf{0-HU@X#cl_FM1w-TcI)vae20JoK5s%J>o@@x6C!H<<uWZIY|PTW
z<)|07aOf|<0VoC5fG(&!#9lp|!HRk62Xe<EIO#(S1PQ8oU{p}XfDaehcM$Hm=-mTt
z%Q5@48kf6mGywU%O%rf1eEs%ofp^&RQS?Beuz$AubqC0w=EiVIugQ>;>UeC{wnUjZ
zKuGFR<vUn|mm-6b(&c~Nyi@=-X<nXDGy8c;pD0slWp5f2;M@0RbT^S&rr!Svk;n&k
zcs#_N-#qU;^v@9_1`Z!mOa6o;JW@pz5M?I)<CV7TK!xr(TWQUB&DDdoJvb%C^{9tr
z%fAlv<L8Hn?U~T)o;Aj?;%~YX6eg&Fjq)+PorQ1{Y;Xmjxa)WYpN;=YMPU!DrE0Wi
zH~q)719<vX1CHZ-pzMAB14+OfGXNd6REw<?p2q;|2=I==XTh9Aub6QhU`);jG%K20
zkkOBr{{kUb3S8c_2F!}oY<nW@K^xb0R$I<EC;pzs?!C7J$oEg)51?DPEtmXE;sT3b
z1K@3omp`Em^&b8K8*E0Yi?;_JML-wZV%Zh;8IRNr1s<@u0%f8?zz}SMnteY<yF!$?
zOm4f`faNobS^tXod;>zIKU^p@ANc{B<m_hCC4I;lE(J9QQuGNxzhEBjMGuVNY+@OD
zN{WO>Gt32aP;F~?<a%DI2enJ0`(8tro`+C4D*$qey7!(Lb$ZrmiB$pkio#;R{z?0D
z!ko_dvM*Kc%<!Gz{vXa1hC>T!RA8NO6Z*me22j-g5Aeo<#~u9r6wpkIEaGMEpkPDl
zfl#|VqR4V<7YNZVy*3FZp7S~p#aVkxc=Qv?4x-3WEoV)pZlr5j?d@<2zB`3?>!$R3
z%TKdSz(u|W3cVUO@GLboQHuN-{ji?pB=>Po1IU;4;32x6rNKg#0m}YOIdG^P^rQ0P
z_unmlmfXAer8q))i|^fq%-Sqfs(M-s^M1KV8$YM}17hG7i)Q)PbrJbh)IXVRJ3OB!
zsis1Z;zQrXu9boiT$`iX-4-6Ee%qT)Z}pvkoP6)}r2Ugv?pt8SE{}5y&yo#;Us(sW
zzeCS(?8GPqAFEc3Y3}_?8$fcl-3P{)>QRB((`tIR!4)v;4X~HFs8RRe`r)_VPt&Gb
z3Qs|_R%`L#m0wR$h#rS-O8T$Te_{g`d{PUn6ZmSuqo-l}r!!fG4>aTe)aBK7(5(7#
z2b{z{Rj7fk;LMvAOFFD}wlnuz<BPbK_n?;siEw=Oa8fYOCjo$n?1a)sDZnG9)%~#h
zNX;{nwu_#~PZ%#S&L5z#V;i0~O&3hJp4B#H8t?Ok2YO4~U93&0%Tp#onKpDmUOWyL
zhg>{Ps?P@i#pHm(Nu^WIGYNBM1ih)Y``g1JFb+TzKAB_tOB$f4JlSsnFq)K0|EK5j
zh%i}n*gFKjOuL)C$EN2$;RKh(k<ncpL-?uhCAv*3ECztHc2E>hW?WaHABm5ZYQ*ic
z<tYz1tIqy$FtzIEmjy*9RZQlvG-zuV>;AO)=#P4|_pf=*6lX3JFsY@^Gr_|l+hs(^
zi<x5yV{g+C4gkuFAIHMxx5Nd+{siLgu|Dmch3h~Kh&Yb|u3ZaTGRew=3PAy0vG2PE
zhe~if#5-IKw}7Ejzx1C#<3%9cBjDLwFjY^kD%G>&eFh@KZ_joO)N2OyY{Pn@84fsU
z3gClyHV!wss-;qVsAve)0DXTiuJdqZh)jB;q6;TIX;dKELBGr+t=+X|at$s69Q3DW
z$|{J__vaxXpv`6u0JsrajlgfK|0V(;2WaC!L^tU~?doF0O5_|3q|$ps<evld{fw)g
z64IukBIB~7t7lu<aee+NHRe`=Z#@5*8qRN}bXnl6djkto^EE9QD(l1ubdp8zc;83i
zqA~~g=ulbTi!SOci$;J+NS^4CCmvmZ<5={!dO?a13QZ%Ae%C%LZ!|eZ&f0PzDhXS|
ztDVFr@<m%E=#awa;+N}tt>a;yuFMeTb5KbvQU0O-wU@;*b@sn7g!We`Gr~=`XATYy
z*1T++5NL)RBWdmo?`nlbyp>2a|I>|8TaN|Hbi3E06Al%MA`W3s1C}|`zZ7+gk#p*{
zVB6C3?dC;^&SH8Ep<o2aXg@=krrN|Uw+QkiXVbsf1IZSL2si(|_@HE}oAoY0r5xn%
zoYYN|HoeFl+MtB;jU{eg*B}&~;5FLVzshr%VCHIFk89F)HA0VMM5>$kd<qiJsywa?
zzq460T`giIbP#IGVb^X?b5=V)RUS;@h*6V1VlInx9^iG-EJMN#MG$V#4XIx|tE+e0
zmaHSsbK6X{s&lNKRg*NTGT7nZ<F{Gq9q-eh*ITXI0KR>jxhc#~3$i)oO!Uvm&4>i3
z3~r0OgB_+pKlq{SeqiZGIB5p0tah*AH2D_QU3jz?hx&YW6Azz$r08a9|ACeth4k@R
zeVdrm#cS+pYm<@Rf=M5#Z@1;4y55_7*{RxVMpxs>b%&}~XeCukM^PRO62=%q95~@m
zIz{177M0g-=?5HP7QN`J+&Uc{UjP6pF-hXMB1FXw@3gp#{@|o~Lq6)Wcv6+f-w^qh
zrH}gfvQXgr0FJb1PI)GM-fQs<n+x>i)?E;T4~5*v3GY)+tLzDrZBzhOTp^e53fAoF
z!ILn(5N7eCuscm{TJB18ZtaSFs(Krg`2;89c@}EyPwXJiytX)JGjyVu8!5v#7Y;s>
zmV|7&6pMqrV+#g$KabU;dYZlM)#ZLg+<vtsprk$5T)qS_SEk-l!_rot9w65c<lnYi
z)@)gyrV~!+Xd!bId35{0eLu^_8&C9OCVsq;)${=%@+B^wx1JZsgO-q6IU(IqwY`^t
zn4IwN@QY-7|Eg{9HGng|bal@42ixY_h->PJ@?i$*`)-LWS;OOweND6Uj$^Bn(|KNr
za^!NV>Tod8K#u~l>g1fMisgc1z^KL}QSQ6JrU&nIV0qQ2upVfDS2-g1BpO-blPGoA
zjoT1PNKBe?Zj^BEYpUW)DI`BPxE@~8qrH^nQKHi%jJyvr0yQ3RXlV+Z4lo>w8TI!T
zS?}>kQ4rm?Ne42n5pLf-bA>A`NPT`ZCjz^E$Ul>lM?IV_Q0~B1oNJ+jfe}bc_YV7_
z5wvR^w*R@b1A`*Tbq>h2re>vjqQ0^W*;V#HgwdUKh`HBY%}vD3+lyi;nu~u-T7+c0
zJ9xU&G4jef)kKS&q!)_C)#7NpkWZc!xnEZ?<K4q%IF>tM!f0=W=<4(R*khfQ{#H-T
zdpv&5jH~(A>@1(QYxm&$hbxwkq#r3s!!rc#add8VnxC_|k0=Qhi8gQf%4o9yLRCMk
zRBM6eiY@BI@mzB8B<s9}Sz?q~yH3Bd)aFH%6Cj6ih|uKIAIvfg|A-rWPGD3(FToPh
z@aM&w+8OWj5-YLdPq-0GAXPgn;dF*+a9NIxzYe%P6@k3FTe`mvXw<*$Cd__nw9Em~
zG;JewTfvK48ll<l=fk4jJb4OH%?_3{La#t4yzT%9^W^7%Nau=Y4ti>SzcoaTKl#+C
z>()?N>?PD6$s-W-l}LzOg<PVJ>iv<gg~1l!ofVW`a%FYTw&k7<4NgCSCk3hqAs0=V
z@e>E>+jw|2X%2d+^pcj&WSFTG)2w`ych_zABILrJP5Mr{LT|WjP9YhHWqM-IVZY&(
z?$ddMZ|Elv9AWVwH;4~JD6TDOlIA`?u?jJZ3hF6pEPIoRm2|eg6;GqhB<-90knYdC
zKXktov<j5cFV4~pS9;!0)wfQ@lK3{wo>dLJ$rno?xOYcyFucFGwNSU>M42mUoZI=M
z8JUPBJ)eV!^$_}P@imD2p(!-$hW?iu$fqGvCu4^`Fsq6`-s@K|x2XXnpEjo*FIog7
z_ChHIuXIywf{4~Uf2wx)!9N9Z!}$?`OM!D0`h|eW)u8qYybGY_oc>U(#VCEK*49Zp
zv_+sqbS>{3^lj~KPh`kGYttfC0~yHZCvVwN@41Prse|;GZ!yL3R59^7DNmAF>}(B>
z<C1IOf&4?%M<OUSUD-lww}40?;y9<R?<AXa|GBk)M9(*q?G=C&S{{)s{FBrVv!V>k
z8~4?5Xy-JT7Wly1U0!85GkVvdpMZiyZ4Z(xJBbVnL#1?a@Fc?2dK4}c)#%TAl1Ne7
zO}D<J(Q(uP0GM%eOk<%3P`q7BQK{$hG_<-3pDfR${uz6y{r$IQX*Srae3><?)EX$;
zMwQe#7r6*g-rU)@94EPRvFFq`%V->sE+Tl<>|I2Gvk7t4u8q<4oI=-oJ)SE!if8~v
ze7D8p87I@;d;m{H+kZaXmxav!q|b{O1tiQ?H^4i`=D^KM!K>;e);1LC3I2s9{wFZE
z(Y-s^ys`O;!@+B)29|euvgJX_l8Sp$Viwti-(y>gPkCj{SN9&COd6Nbw+Gr&NSby&
z#LA`Y(RH7-n5cu=r^7*fb4VR!o7?p<ytQE;WZMJiyEk7--!fP0ADqpZ-Y<k+nfq_V
zU+&uW*>Sp#!$Y>=Px9A><<#x{`i<)nF1qpUqbR-RGA^q$L^w`7yWuyD(}K-n@@W*^
z11oHoqvKCC1~MQ!vvupfEX-aFCuPm!_qY2F13}3(AkAWgUdE`zl&W^>d#-;F|KZIh
z(RBhE4?V^As?Q3-sVPXHZm}omOqM&DDvl$bDYDXidWK9~B%>dwXPnwu4&AZ3I}mFr
zE>5>3?u8_t3vUvzeBZO++P#i>IXW?Gm7NJ~;egau8y`>{uM}-EfA<@eh_U*eGcB2(
zJ8wL)<x^TF8PWv1Vi+m(E>;>UkvOJRrZ`yZDESZ{i@j4B?N_oiX!ikTP45=iltXoe
z(Y@tIpUeh(_&ffBR6cD`RwU=4lMeoE1Og;dR+BWIX*7=0?}Ox`b>B_mKBq)ZfB`;&
zcJBemkvT>m-o4)+QQT~z9mVn=EsvD=rOD-&?2%%3Q01M+F>d(f(3!npr1<y3L`f4t
zAAEm^;&_tN;kZ1D?V}Iq>tlBi@Ru-T#j^Npdy6t&U_elRjygwo<aMpzySh(7e4h#A
zVq4n<rBX@BNhfyXEt=0Nhx#;rM{f;?_?1lyUwYuctmpA^KK-Qn(IiT?6UzxWK<@ZI
ze)CXkIiR>+X9fHb*Qpb_|K2iwCcG)I3q^6b79Tx_776}w0kOZD+~>OZnx0!xaWd9l
zt@1`D;(uC~?r42;9A2H~{pnAL?p8L+x9wJK+oam~<Mdvr=xe!`M6GPZnI~H7u!H6a
zht5mL;dINP7jvmUPsYnH_-#DU5dZfcnwj)nFK~Z74fut;pJpR*+r6<NDz9VEOrfB^
zsFt?QST24%;M+xgxgm1ce6v4#ndq*Zeq1-~F}T9}x_m#}m#^Fzy@5la-c4a01-g;=
zt(Q=(DB0P|f5%}xu`2O|^;FWxxNH~IXBENYiug97idTLdE&aH=x?d&>ZG2U1bDh^9
ztClibSK{(!OCnpsM_2rIN3CeOW%Fs`ZDyB{PtW4ou6u3umt#`JR*YrW*@EPP(<~mh
zl0yhd8`MjEK;hoJ*%KNK6;iYY+<+0C8i(J{fC)D~`Y)@)#*?hzW$XN>OtRp)_DMsj
zruUJJ`TO;U4xZ2Q=w7sj8Y3!3IX?4*$6%<#$Bn6X&r|U>X$UdU`P~uTtp0KiYBEF?
zq(zWw&vWF|Z{VIqd_^X5P~NZ_KEOUr=xJ9oCWrAdOJlW&?S=*cdiGuAE(OSyw0aTB
zaAGD?w#N;hs`f(OlfDEUDV)mkj{+m{;Ij5i)lAzoVWwVR(?L!8$t+@p?$KI3v6c?4
zfu&PQ+wH}hHLIXmIQE0}l0AV&0RR%WM$%!y6@F{vODTJQpmi_8mU~a!9gVL*NSUc-
zxMl89?WzWsz2`07W`E;BKR8SvYF9UFNf~~>Ms?oc3X)XwJ#xO=a_X<z6+w;ySaq&&
z5aa%4^j>B2ymigZ<uE1-RN%EutwVMh?$VHUXma9lPUWu=nXj@a(K8asfB%C5Y|1fy
zw~begqby4q=zK|9t`u?s&s~|TBOSP?a#AGitBi@DwEI&xl|@T;XZ4!u8$RyFrgQeq
z`)>oHB_umvkhpJa^y2mmlq|U1un1WO#<k-*BhyXzDyz{-QBu#<+cl5wFW`Lsf%&Ns
zL9xz;X0}7oju9C(n&)Ga+}DpcmzSdfS$$uML^oI)4Qv5KAhclkCt@S@*QoIo91N@1
z5Fw$-^jr;V<dPA6(K%gtW^F~)4ZW<U{FZvnxn``9Q%1%|+ujNi+AUzLsz$~XIk?w4
zO2RM^h@Kg_eR%erIT~UiJTTwyBc-<1j!pP%=gdVAvQKh%8hDg(6k1u5+^!?tI6Upl
z8?hjYHa^-<7%q1EPLQmghLzPum$O%6vGPO5I2@`G^~hc31GAUbqCipDTZ3w;#o*d=
zP?~m--S53<j1`nINoVgvL#V}J0lT-QD4y}66qRIB#e&>U+ud&4IB3n7R~Hm<VG_HZ
z9euvJn!hG)loAh+?h_Ry)k~iFoCV&tvE<aH?EskdI!P~Nj{b!FMD(~&w{}Y~UEv#^
zUli7CI8S)H&oFaQw!e3X?cj_X7%5Q%V%4If)2Pv5Dbf#;FWj*D75n)F7F;!wKE>R_
zQR{j6Sh5(q?dkQ0vjoquS5NPDjtFsGxcnaRj@^)MdB%R)AI_-(B?c>`s4X)3uUz9M
z6q0gg{(0pe=v`=eh1M9%jvGFLfX6S&bY3niSByUTXHsfj%zo_X9{1k76?guOW7orz
zd34%YqV4E_Du3s$(`l<&RI#w5tcI`3$;f2Y@Ot}}Yg+FCfOnu*y`Q{jF<N5SVh7%k
z!^t}K%$&*^oSR}-EXM*H(p0iZrvAA`bONcPAjM5f4&(2?krrxS|Dr~b$a%9FO!<Ix
zJ^4&`vTfUE+k7mvYmKvf*k@T4+V+Q>q%hmywt617BjQf2jed#q1D}85$=5!`*|6Qu
z5u8e!&b+A+J&{5#+-sqiz_TPDR#yZx|JaRWCQS<DGDAFVIn~lDLdFrL87&om&dtOv
zH_N$cUPgH`v+K1yIIq?^<b-?qPV~in1RqVvi=tmVh~MA2J9XpBV<%vni8dh~KUfn+
z^c&O9-n@Ge<%7(~`<zR+O9KyyJ`4eBY`gq+Hi!o%F+Q2J6l)6^e1(hzBfd#YcbuqP
znA|M6q0F{nsu*spvn#$SWwTSxz~J9^J|gqr;U5#A8NZCj_}F*@ASs`rpkkbWrEaUK
zhm=r!C)7Vt9xZvnZ2}dP3^5+`lzQ+W`NX8skc27HoJ81iGG^lRn#s2=>nPHyrhf2f
zJ`jy+W}6LY!IV2lsEIPQ*BZh;lbj|R5Wdl4dRG_eS?;#~;hLPV%B97&c(WP5YG3$>
z_af1m^b$%eQWuksw1__nGQif>=XUoYlcg9P9Q3+2(6+7OgK+YM!uA$pBv*0uMWUP}
zR>FPJ^gn64V?i5H?x)oEN7SY8jFf{O;`6q+P`^HM+o4a{P&IexwrMy$meDJn#d<n%
zAffiSmUYQGk4dy%<>>8q;D#K?oPI`)617EIr9QabB44_Y>q~lZ^_i9Tvi+q7WX`F(
zC6#xMca>%fQB+mk244Huz<C;eNNO!?g;SP^bih8JgCn>-1-TdQtq!%Gu)>)Y1tsW5
zE+abOxT@81Z&w0RoCepP6tr4E^eC<QM&Ld7rEt!bK_9Z(3}ofCoKpQ7qkkR#HR<~3
zinle?u700XUX06sgWF`zqrP3xfk9@-y?MRnyT1O;;l})6i%GuJ5PD_Q7eCRb(j}T_
zl%s@e?~{8HmVFmdE@4YUY@GFxk{NvVb;1Z^$=0;S4DMI~8?YaO;>mXTJI%4}9Q5-x
zCmk26My9@t&O;m`zMGFO6QB})&}$UyF$vHY+@FMRBvz~3>kAc{9P71nl4yRz%GYb+
zAvfjL?`LrtX>LPC_PTZMDASp?eCD&Ecpoj5x!RI>iD_1ZlZ!ac3HM=-WZNct=Ju9Q
zs};&?pE65BMu&d?I>A}|0N*|cL3Jbs|JozVE=DZgTr!m!IHL(-H_Rh1E;ki@oPOUs
z>26jVsZdsmIY)R4jPFNIAtl%0xZ2=Zx7fOQmi)*!%o@H;=H7r~l9RRUR9uXFwL9B-
zyAX1@fRXYH_4knqA|;r0cvExc@v5&IR=+4utbJ>mL(&(={qoqpt*^l?{?kJkqq)ht
zSv1y7EY^*Pw!grK$RrMk4-uvIpgn^JmIn)L?4m57PPRCSqt30RkA%BUwl}J-T_h5*
zu79$30D|Un1KarG;(fDF78%R8@;KoJN~h;@V?kkgQ{8jRYO#O%08PeaN|T+?FVS0L
zTjt#)K)t!+&~qA65?>FoQ-<7~uTkGOl=eq52o5ITeUU;4?qeSlW|GXtR<FT(;%Ui=
z+D}$3{+lIUH>k!|S%0)ga%MS)HK<O(Sn@1IR(+_Xh)*aAfr9vP#j{)<o1m<TV>J9%
zUeUpr<;X!4H%N4n&uU&!*9~T4ti7~eqs$Tmur2w(R9>{WFxF1i18HWYx8)rJkC@vG
zwxb2{1umarm9oRAU50t~d1+tKxoTo0<H_OKxFJT;EKpJ15XFl#cuQb1grnc-e^tY2
zXOytG8=%G_xG_BS7*}MTwBU>PML%M%li9!uq~4fiH+x<8qXm`K1R})|WLb_|X1M_+
zR1bY`#~CBMD0>vh&Wfyr=p`2#WqrBjb+TDIJgIv5y3Yf}tUAVqR`8^yO}RNwKcTP0
z;x#*7f;sIVB^Dw|QmNol`#eG|vATmi6|;5%0}hF)zts`WMsMBEO;`0N%F?|{Vwdaj
zP)yb5FYe#sb1PRT>AKcK1=ruL^(XR#y6a7%HK6H!#C}CWs;vHPIIi{?;-<m41^dPg
zVlSA9rjov-`)-esyI1n6@0>Pr$O&<rj36R7@uDQaawQ&Ft)qaoLdPO$!E22B@mWHI
z(0ivVP!#j%qwUP^(0O(bgql6yLW+bW7I6oU`b+#y^oM<KQb&CYRWJB(J+uGtIx<p0
zYQFWqD|FteZ;VE;BP8hE5>P06X}B!nqG8a+pX{rFZY=-Ni*!8w2wGwby$wA|>#;RY
zv9nVjWT(&r(APmi<GZt^p-cXl6fkk!R>XhqX~eu{NIYMC3b6;3m8ry^(a49xef<nz
z<R?#&bIuoCXx)T`O1?h42BRnP-KI!%G!UpuC3<Bz;l)Kswpmggqhe9ZuZcvGTW+Lb
zFoy~_I+@~LB8z%)wRTxW(&=ndOmo|IW&}Tk_^mYxMz-}US^8yKJoS}Km$gIFRiOr<
z%$bhi6q+5P&`)+B5p&VsiT+YPRSVA=#l~7qse|9iY6%SeiF#wbwRu2)jy#8SDKqt0
z(H0t;0Hu>7?{`~Fib;*h9OReP19UP28z`|KR_an;1h3c}tKD<}SNU*=YJ?`tZC$H3
zl10SSR;nq3t9w<ZF{QhD6U^bRDcIZOXBcf<ClM-4sNJ#3-#y_mrl+U*@Ie>!^-wi^
zK!g15PfxDL)El?BIkH+MHp4gnor+ExCM$#_)Zt!=ok0}Ay#Zv2{Zk<r5{{N~H?sJf
z^oCOL8Dc?b&mVB2)tmTQ(Ut09WUULU&&pFH1_iOH^q13*y?8{EjaHkI^$_JnD2=As
zQlIxM2WQg;#tPXunG{EFV9v*!a5@=Dj?X|lvLUYP1dk5g8c+871EQxm)-r+~QbdSE
zCYZEib+EcT>68yNy^jPpqethv;_5kR+Rccmf?0xuN*H`Fj6YNjN+a#$J~7RVY7ak6
z*~(m^X%0<xO0D`M?_=5$gwpe<MBi32Mn*BT(`J~&jhuu?0d2KT5ik1i+A7q@B|B!9
zjE#W(m+mWC@%s-jY3|MQZxZa<-;lhR51eVktKJhT)t)o|&b;y-amkvHV*1=*68wA6
zLsNL(%n5PVc$7~Yy<av0qS-19Eiha?@nsTYPfHs|=2Nh7SC{FD@<-z7my$Q+BqCNg
znHf<)+j##j5le~UoaT(i-FE10-Qy@M?3<{cqfz*C4>7U#6YsE-5)$1>PP#<!g5B)C
z5TszVhDkrq=J(KS%rujI@bM09vu5L)c)QqKIag2uxAeT!qcM3g@iHBWWJoH^PN7NX
zp1s>~<_+P1Y#NMS;auS?-oc&WRju_Lb-vCp;eO0fi0pviVv@zG$Oi8@=B#Oi(~q@W
z{pD@XpH?)J*41~|+9wuiVYbnIB-DPQ3uANqC=_1D=cLe`>6nChW`jBLwQZ0K+(DXo
zlq@n^?GFmSh3EK~MIzh=USfC!i(}&J5m4%6(--|ZwFwoUb0FJRi#*(r<}k;Ouws`J
z8ju&SP}%U`OYeWGQ+VyGUbwOpn@cm-g6n8u5_Vbp_-A%XHdw@u+vX$+6N1u)Kg8s0
zTKAR0@*kPu`C(s<4F$Jis1s$=Q$kp|XC&piK2L{9^gdZ<2%TtXZx<P@50-YnrLBHX
zE7+LPjl^@?x;Nb~8ELon=%Zo^{e-Ck-hlc##yW|}G}AP*;DpvCPVD}Egu8xyGv+PA
zCH6~E&M&;Z)jr{<x^onP{Lk7tFrjOk;o`=eGIK7Z$K?Bi9y-#U1QN$2`;SLtBX4Ua
zKsi{TPmwC-gl0T8^Nn`paDE83mWPh0-=hAfnm(*eSPGS2AiusqmgKLg0dp9ed-7*(
z8OlzN{EN?AX1IP!B)Jr9gKrP0_>%6Tg|ZeN6_-@K&Eg`vk2w)|YL@66c!?nLYKEg+
z@q-W)kiWY~m;GWyD?OJ3dESC9q<v@l=;Sq|V?wMe%^7Yv?N8NI_pW7;l{{SRIPl63
zvV2&pjs1b}Izh>lkM0eH;F!?WXY>=H*G};<cK^k|T2ldoQr?}qvC1@XzsJ)o${CjN
zFC=G%Qdhn@yK#}d*bgR>W3f!ewC7m8#-p`;Scb`>tE17gRat26V^$uxz=KW~YaG<l
zF_^2ABK__O^(3WroH(TQxsq1FCuclV|HY2+g6VezQKyWZ0y1AOl^s%XciDfM>F#0D
z9Z4fmgDFjRfe$SZa%YB~k?CNEHXM<y#2vLf*1xSTF2ZXyUdl(;?$KDJHyWK^&W2=V
zQ4@XgN0lO?nIqHj6xCOcoB17~X{agr=Hrv7Ltjv4FV^G^MGgkv+o>|*>;o}nw-g!B
zX7{bM%Og!|TPw3mbe3Aqdn4<dCR%O$N|Q4bt;EeNbz6^xK{Wa)Pd9Zp%^I#3)5Ct^
zfRJNjryywsD3Uz;iQiZtuT+21jMSHDRx3u%tHm$`%*3@(;$ij8>xVFxlTzWH&PTlH
z#{&cJNx5zxu8I_njs)57n7RF!+B^RK%_+#Wy!vR$H=#beNR_V_n)|cXEP5;6c`Qmj
zBhMK(w*Dv47O%0;YtG{%O6j;GqSYUL8vWkPm+A*=o=P_{{?DB&a^%JPaV<=3gM1@M
zB?&SBFUukc=GTcQ$;3jUm1?CtE&072$-1vo%hsw!W3{+C!8~w@WcF4JM8(1Oy~vwr
z0U>t}q_*X-c-kV*8^R*&xwf_OKH&KQu_cciJ{JNr)Pq*Kj_#YDj1XU~2b+6>$#L@Y
zKWHK(*&Ca&*BO}CW7@V4D@9}c*s0<nST=jUCinydv8|aF@d5kdN9UwIq!PVPu_L&x
z=FMa2Ip5H^RJ_rzort}A-?DRxkoyT6NEDt|ibi3{`kvLjXhpWhp`{)$cIC}F4}JvE
z(y?WX?mr^UlQf_3pZ$tw!FLsSxVJ8Tg^KdnE?i>s@zUh_hLqfvf4!vlu!)lt$urvH
zaR(YMykD>nq}1LR&Az(6;eq1a_H4itN2U&DzF}9O5jhqyLw`WqpR!$*PrOvbZVyxV
z9?##_>>IuzX0M}T-c6cke@@o$Gr6O4jTvW-uj+RNvRUj9)_REhhmK#aNNer`#N6U9
z-Gc)F)PA$1vpo??`>qpq6dfRn`wCmD>&lFn0IOk#fS>+BQ5ByF2v^Ny*W!I!#*-^v
zI0u?0<eH4jeF<Ja;-z6Q)>Owd)ZCD)lz9{Rp4^G{&ntY<?m%w6rgwquW>8-)L|Poi
z$C&&&NN~n>tu*=|^?FqDSUEbMs=s_C9_Io?@&(X!W0MwBTJWWpQO6XOT$X=NG#*#?
zffzI2ws9LNi-p{NV8!BHzckMmOgu+C`#W}R^j0Q|hR0c+xIEEK(v_#4UjssQ5>ot;
zrkHff<QT)edz8us=pyezOb>TlWJ4X^R--TflJyO#y8`pvoRyMT(IDc#kZ%J$+B3zo
zM_IS&_Qucs^va9m4K%jg_V-av@Qq$i<YtV^Tg$Wf;Q8vTdoRI>E?T@|4=cKaV|wW4
z2=<$8$!j_q64N|ri2&>0`hkJN3xNih(@o<DpK_HHk;$CqPCU7?@|<nvv4T&ZH(TMT
zCo29Vk&a|wPWT~DVN&;_{Yd}Nf+7iJqu`5n`36ElRqIM<AnKT)ys-r_2jbc_EZQ0W
z9n7&!)?4SCFN?RYSJt2;V9gVgwVji=SA)pk{bLWpiZ-XaB+<_;A|`S~$d`(br~J4_
zkx$=9G%riW93d_Z#Bv8F6pBGtJZovH1xM^@)Uw&^8p0D`3JMWRYa)0j+j<tpqJi9I
z;?LYyDNCejr>iYIlf5#|)dZwDIA8E^(g?_rF&El|EUA*Y=qMO_V7TB;mBu6ur;C%&
z?wXh{TOf(J*YWhl^4sqT*3-UJi=>Wmy~EM&HHY;7m&HVxM!G6nU!k5TD7FfZ{E{t^
zKtGaYs!l*WrvpQtZweQ`^f9FpkfpW~?WR4Mr$MYDWwTIF+mxwqVe)gQ_k?cw^*fp~
zt&3eru?9?(>0*alk1^AhH=|x8eK8B9#xjM?`XggW^xwcxN3kLhYXTo%lGy6%oLMzi
z&G>;<BR7Um!<|Tcyy(1@yX7!h_M5d%(g>ox#}R8p7Sd5SMO2R3hS)*veSXTB8Irfg
zKk-jKa)u6b>*ba(ZzA^QN+GZ5M#fiZHMWvJxpVST1FbT@)=85-AVvmaV+ESr{ZqV0
zVkpkX1SJ}}cb)G{0|WX6XRoc$Swo{9BU&a5Msmvti7x5WT35W^hv2zaAj$Pd9v(Ig
zm2vnNW1)xX#Eak$G}|sg9hPo|iL#r{E3a6>WJ|8n6LcZl6^l`qG6>y|8>1rW{7_os
za>`yx?j8^N^}dEM(uF;E5Ywq{eBL7E2<D_pe2FB8c%q;Yki>?G3~eYzg76hQ<&i#e
zMUfnQ_Py^aPrRaLNB*X0Kle>yk}Hcb%e1P~5hvEuvW7rpuP=nd9S`1PuRJyFqeG3r
z<jHB*IAl4r50sIR&wYl!Tev>&wPhf5KQ@lL@o@p$4eAjV-&kpDGdvcBZ1KT7(uEmC
zafcx2SSrJ;qd!;QqFt#HW)oJ+lkylAf)MUl<M@H-!H~UOzDn?2sD=Jl9iFKeIRhHX
z(B&Bt+OnCj@dUJfdxhN@l9?u|0gQfUh3^Uk#<~j4oxHWlXPqn7ct>M&v6WN;FWuHS
z!v(D%-7O~jyE6Kdo=S%GPnCPW9VP5vXxeLSZ}qqGP(oHneTbMy5a$^4)fn<)oMt8`
zeiePgGaF&`aOPGsv0ES_@ry8!c9MT!hKyz47U-d~7q*|;NZrZ(z?^UMZNH^QI>L&a
zH@i~LC1dMW_JBaXK}J2<ISoZDV=>a<J60KwVd}CQeJzIj7*9N3DxDcF5**L$CaHs5
zqwAaPLrIv)!lMb=a!swnyAl01oDbYqJjBiu7Ic3s#*mcPn<Q^5m2nBu+d$J>!MM&E
z;1B+o>aQH3d7SPZ5V!eRHlnbBMw%q_aI7^ZemjCHh0F1ha7%SuE~|S%TT`M-A#WZ-
zkz)P3o^idMK6Xx<r0F?%45|)Glrz$a*wt&Wg((xT6ks9aG4ESwLMkMwNFWWPrdnU+
zbGm4&ucQUr*7$L3eI9P}Ic+_1Yc~~acC=zkXh!G8HVs+GK~&*34zp`cUle-t+T~IV
zkRKb*5<F<RAEL}W8hA|iVT1b~PvOcf;)S+KKYNM9)|F8`nDV=E=uVA@O~C(pwp%fl
zBS14qv)(i0QUr-WMAfu)7booTNid$yYu3-$FE{O_38gJ-@R<GdlU?<UCCQ}6BD&F!
z5~c_&3~7t=3b1x6#YmOU3*Qx>CTZhSD<tha==Ymn?Do7=X?(%Hn_$iU*joEYO#k68
zN|g_q`V277XdXHOv)iO#(Pe6pO~m4aak?VC7CfyDp+_u#E=V{z42oBg+7wt~sL`96
z5LUD{$xTPDSyD9kA!IYA;k-T6WzIB$-V4O2D`;LMM;*rx;;N#|&831l+SSQ=P={2X
za$2`U;%$%hL>7f@5Mn87-r}+G)D2!$>mQlvB@3u2Pk%+P$ie3*Cyfw)(~N%g@E3hE
znQ3Ou7Zb7BQbQWF7inwFH7AAQQBU0|NN93&@=wk>o7ovE`=7PI_}E39j9zWYef~C#
z=gFwpdl4*cNMO*S*3_UH7Lv-##49lGuXjtwZZ0+HIp~RGvb_=RCG*_0B&?xQL970P
zbZK(yNcNpjP*3p;r!Ri}(#ve`-(M+@evVccmeSXo-ekz)Df%2E(;7@BdEw4}g~gAo
zHaE8$`{ymiGWT0<KI-R$sKuJmqU^3m0m?Fw4tMlLMLe?2k#H5IfG>N7kkn{1I}Uxi
zz$b@gB9bgIRmintn|3fN!{hvPiJ?R%_5-=F4YH8YE}G%I?6-YM27aXIzfgNU=TIUt
zZZvCg(GBM0=7Xhf`e)rDzGYF_B?VKB;%cuyg#@J``Lztr?5xybRgClr^FE5`;o)#k
zKU@8cCPw!=3}yo%xO&(?etaS#O5OjdLG<hI=J~w)yyVN1eAW9myxd?rwSLrW>gz~E
z#S=*SY&3D-JgH5_#rszAg;`2fgJ2}@cf;0XGqeZh6DxRc)C$-huctCymL&T|LHYjE
z%lnfaNynJSDNhCW1h>H5$<E4@IDz=Q@h*h%Pz5CjABEU=T~?}7A`j0I-L%FwCCpca
z`ME;-`1_EqpG~pZERR`KV_kFS;Y55WTEpZ9);vv`l0}S1Gk$@s11D~}jByD;x7n2r
zdj;ym?nkyHBRBu4TZ5ppmoZj|woQm}kPI{bi(yj)8D|4YGIl8W0S>9dH>7ZdM(#jV
z3xSqztH(aZTCaadiC*&S1eEQB(**}BE6yK(e{K?XG<N40jL$LLF<Sa!t@@74w6dne
z(iXu1V){bvMeF>dpkRXE$Z<#(<;e(TEQm*rS_&R@HCmV11r)Z#Xl>(76rK6jh!`6v
z&Xs%e>)HE#ZkYIkZi-cAV!E>8u$P;P{vUgKLnX{mn+dNlmT{!E28*IGSn}Ntd_?`P
z#?;Dg6y>Qrsj-HduKHG^-1n1AL~#*OFA=<YIMFvbJfP36Wxp1Dg7JaqPDe02BHg!l
z$#L?cYn9``Vvyqu&vI|-EoQ%H!K!Mbj^G!1W>&8Cg0&ySM&(Y-FErAw>_qX64If+3
zDJ0nQs)s%BV}0H6QnC^OK@tHziN<82m2}PN;WP&gy~&S9*DQzh_peSVSM%-@BJBG~
zW7cEhhN0{Uml;?Cs4D}XwG61g4l;MU)bLyIe3JDskLP`Xb6^(sI!?}bUdHr<Gs49Q
zpA0)jD3xo$LMD}IoeIlHbT@z$XP5~?d(+7&Alte#gI~L8L-Oo3#g>?Hie}TanLI%=
z<r`$xz_SmF%csMM;?s7?cnr>7{P>4edBsuv4B2~cBdw2o4MZX@3rqXlhpnz;B@h?T
z^Rg9Mx;QEvWtO;**G8rF1X(<6q_V2hOS0RMkVW*Ay`Yr{A;MmHtd#m9O)dLQCO;>N
zTNa4UbJ}_+-PSQ0ZDV>Ks$~Z=Nw`i8g|d~*vgd8tC6v((%YWrhAfe@!L3xojM665A
zZgo6o2ep*IOF$$3Vcu%Rdyq;u+(EE{$=Ejd8l-pjz59W;%My`RwC0<s0rGv+8)!&P
z>3@*FzB7^x7%i{4W{QjErR{h6lGwI3Z4cQ*A7xQu3|xrden$C%Q!RqcTuC4|(Jp@M
zarik6YZ3X2CrY%K1cVmlVGS~=R^)Hk(UKC;Pc#v;KNpNsawrxr?zrO?-;8G>Yt5=c
zax5}pzOy{7qyQ-B)7V8S+`(tWQd(9f&(l9#aB?`&4$nZMEU1!1o*zirF5jmIe-eJ*
zc|uU0AMH#v6fE7&Nm;I4#{Np`IF=u3vF2{DSP<2`|Lfg4%L2CgbWDNj-4gAy%Cs=A
zd{tKsFsAtQG8^641mYo-qyIEkQnP{QLEi_x?}pw-_|Nq^IRe4!F=4sr6DIgi)Ar4}
zt3r(<YV>|QOIkZd`&?$U%ad^Gj7AWIJXJ`woy^XV`%3y-qF-7tGY(<d?lcvyJ))Ab
zdmXI0cJVqB_`$l7=U62dlzdm>Sdq7|bJ8O5Xf&kAVE@p}`iGF0*ShRT7@Kbcmdc0I
zRsH)ZrO+$J>$#ks=%!3WE6s-XFu9kk^acjAFhgk=gqCr;g8jqN&7#HX^7Oldu~-hR
zU>xY)><+QSGvA3Y+y^`KhmhPTg)nd;ANIRPQEH=$KoHYz+W72~swNegGhZn_7G89@
z6`lwZvP(L8LQqH%GY`OGCSRvfVvd}zRtk&u!st#_IWI>tNn?Jhr6JihBm{1=xm_XH
zrJX%mlHxLM&CldVy+YEF-AgIH%CwrX>qXMu!-dg98Vua?aOu>HC0#)IQHAz`jGCtY
z+JK13AGSSy@v)^Mg>(zX#J=hP^Me&P#2^F&7L#zP`TFZ*Z$Wi6@Xa9GItiptXg(z9
z<9W-stsUg-j3%BZ(d0d7fs-r<Kfa57YYm2yDa(0d*RIR?GmX~~Ut1U1gg>ha^O!pt
zc1MwWz(Ck;$xd$B7^i}uEN94=jv_M76UR@6AEQ#S(;$Iqq{5<T`E+tjOC?vyQZY{4
zVng>cZM>W#Lp8S=Q%P+$Z=i7MsWEQ>cdFw|tC2sq1$U~Z25)f1Z9Kl;0Q#VLQf2Mo
z`)a!_HMz&k$IrPx^Q0yQ@R}A-YF`E_UVIM5iq}nHrZ<C1kSnGQpBwQK)cTd*7^5ih
zS`_CQe#%j@?06+UH|;SOWH+HAhf=$Qtk#Q7*bO}&#4T)KsMb=n^qaUFKrf3-$5`3-
z@s+&cRvtHB_wDoh!y8-$*@_B@;9^?es%ifLZHf)r2_k&1uqnbz=B3l3emj|vYmVhS
zKooU`T<FPMeb9gY23NJCu8&D0A&w)chMKQKBvsRfH@6kgqEa;Ncwcht1G<1en~}A|
zvYnRa0D8xngJoD&<oa{DKc~U*u?%D<DO|sM(1*>F{DcDJY#B7(#^D!x`7bxK8VK1P
zKle*>GvV#wA~TsC(YWkjH;<1E8$gdeQ<EEEuBKC|fqrYEp``y6&pNYz{Kz`x#3TY`
z-xHQAf)r^s%u#5*)!LgP-^_#hwQ_MVCDrUJpUcqr-9QP?iRkkiRkN=g&o{TeXyE=5
zF)MDQ`|3O0e`p>dOj#AiIv#daWZ>P#BtA!f3-bf5GHbre)_uj4of&<C)$KM*{L|=-
zOL6k!pZEF7_{zuvGb8o;N5}7DEl8_(MCwvzqVj$t-?p<a$8*rt2GmhL3KQinN(^+0
z6dLx;s}JCP$u(*X6AK=YX-6+I$7&paq(SXs&&=$R;&eT<qQ1H*&=U66b>EoFDDo0C
z>|Pk7yqW`5i1_`IH}Vc5x&&1rx-RvHmuTi05q1<jFXtUL*s4A?3Mk3NxWR0CNuZN)
z*IJXJWB7_kJ?<lxWWx@8_t58Y=2QD&UJF|So6J`=jV#KM8Yp_9%9dZ#h6U-!E|W<;
zek1u2&4b~Tf6nOp(Jj?8BC$(Kuzustc8^$FPgAeWHG;A5{B^$PdQ<M5^miTeRvTWF
zLLs2w(yF8H(zO3=*bTqWS)^SwRr$L$ZI}KB6*-Y}e(9k6dcaj<yi(DS#3|(HX`H}T
zaBdyuFxW31j)u3!%V*~-%yhoo5E(wTyvzWl!r>o=Rg_-_N`h~-Hi&h9VoQzx5LlYO
z%Y^;V`F#658FzSVou~I1HJ#t@bgzV^hB;~Jq)$81DCX$Bh*zxE^!MlFKY!RxdnF~_
z8dPtNZd9PkwN0K(y60O8bwe8F4!QklJ<{2}o_#k7s6fpy56P&HugxrM6q}WPc5Ns@
z{RK-eL#3{N+W!;J{QMiO-;ei0nD)E$P0kYGJy#GeB%KSBo%Y|C-1jphzkWQjIZAp}
zS@~`BvGe4L&qWLQWY@<E$DHpoE%SF>kLpO0>1%eJ*(mk+RdsgfXgmvH%p;G*As<x{
zwMhyXC0fCen-Dr0biDWBk&Pl(F)Kl%k$bf{ivi{ze}RA<Rf}Z-U-5Cy&unNV>@&Oz
z=B3hxM{fE}9UY%j?4~Vf%EtY$E*+}GIoDX)%u-)oMRtD<@jWH1c+cvGW9HVCGOOIa
zT3h4X_pPLiBZjJ(u;s4N)b5!Nuc>|bhEP(e4w0J;HBb4n-MST2czU9fAgs!yLnhma
z&HX-;4YH{-U0sYeBzk38{|?oA$MgZ4G<)oCq`S#VrEu#Bxr#ThKoe_SWZ%g{QllBh
z35WJRemm|d48KRov<pb~Y86zU4o#1&&E@usb+SqUw~X@eC+?f<4IFItAG17T4#sHc
zW)vb+ri%(BZG-uJq!a;be4}Oq4WL`CCi3UDwRlGl_EeR{Yqv5Z7ObfVxk8WpQN(sP
zD=C<sznFvoevZL1FJqnZ70~MhvAJ0I5w}(a{g>o+-*Bv;3-Pk;G@MVp!KS`htwSwc
z)Gh`$56RwlwZ$4-WAxh>B(hdv9cVe8N>hDm*;R;eRiqYRwZx0rC^Fn6-*h&QNO9<W
zu({Fnjv|Py_yXr(&Aak<d4vn+e6}@~3pd9G#2rsnV8-+$!TtF{<#m}JREEDRu4O(m
zM0DcS(y(d+UR6(gl>2*8O_GWn&JB6kcf*lpV-IAJQPcEwl?;C*7mBGF^$=R|^;f;j
z=0+veL%c`W%_3hU_}+eX#hS4Zk7?Z5s-$Sx*pt;NRr;_{(_hRZ?(T+TT53pqpL0G)
z`6SYyU**Gt`M@NR{h2ZO%*Qw6h>hl#0+6dGrsz5I)fO>!4Yozpo7CMT3`u~zPl7VD
zXb=~V&SSe_s=UbZfGF-IdeE?Ox_$kNtMO;s9S*#mZ|dL0=r_5nypXAz6hKwDY6#AH
zLOa`xVT-4Rr>oBWfnw!FvpwQ{Xu;?9q>dew>YLP6E!4J%1>bjkmhaWs(FTikKWjF1
zNUgBoqHioTA;0%#CB*dc`u1HH=~T5-gSmjq8|RxXosZca$UyPvAnY^08b2Aw<}lD)
z_=5pS<|HD&To@$6NV0$2Hl(iT`ttUQYWjWbe<v8y(3=eAhgaHB)eFP24HdjM9s0j3
zI94HixbG>n8*h+)Op^Jh%OFIrTR&R_UvkHSr+YWWTUO{mV7ZMV4!XAD4TgWf*0pC!
zgo<GDkhm97#Ykq1oCGwDqGIT1o2+vW2I)5bEWK#AF1oG5h4>Ak@<j6)s>`J^)7B1;
zS6EEtd+A$7ZS$A;T#}A@O7uo3KkS{ozRg|5thAikpggR&5QYwV$Ex~K@*)j)6qHDc
z9Jm>V&$=0{EOOG;JBs~MZcoJEjcprBJbPU~^F!Z=O3SL5w@R6)|JUu=g2SLMhR|fh
z{YHw>G4;u-(^Rj}XFf;Ds#%U6?(<H48(nCM<>dIkcPa&`2RtC%j*OG$ctIkW&hsnm
zc)bHoIDDYKrpApX!<+Q~&~%njQGH(nh5-a=1crt|S~{e=J49ee>Fx&U#u>V$kp^jz
zl9C1i=?3XWy5WEQz3crlpJvv&Ywn4C_SyS+Tv`=IcM0v_YC|Y~+to&WCSb$#ogv;R
z%&E;JDlgFyfm0Flv0?#it{=%%0{CS_NYr<_$&>Qm(?TU_BI2vIn$bC*pE91h3tm-?
zjU)=U!n;)H4b-a7ZIt-}=qdXMfM~mqJG4)KOBQQ3y#6bDeLV4Sl>N+bGd^C$4asfc
z#qYTG+2H~cxYd#v>>uGK=jYbXL=g3h3DA~*dZ02y1%H7&LQ6MS9NBs}D<oEO6Q31m
z4Tg5KuSdM3VqQ{S%CV}tiR>Z$&IEx_!!ON#@0?1!kZeg6#h?~$(Y5p2mAY51rSIye
z+cYs07x>xlWD^6A>!BA*r=%LvW;#@zjTBI{ViGqg3;t%UDXzi7dX5*H`EX!TMoLJr
zPh3=|Q}k^FN1Glh*TWlHWGS{G;E=b^!TxY}S<tDudSl5>#5G5vF%6c-;S$u_{xT&6
zUvtV`)$Nh7`LxxY>%y+x9*ZaHIDD6%ErtBrS*mlRo_tvM;d}iT1SQ7(Z*Nv9ou&zk
zLeBagT%!nW)N@*l^-D`|8OVd)T7N86vU+tq8SwgaWL+yMMURjkUNj)J1zi}{#a)Su
zUp5u$^sRCv!jzWwK_f;*sXGtqEFkZ3fyen^6;<bUnyT)_kz;7~p~Ja3ri@*J>&XEt
ztkYI4RmewN3zE9h3p_=!N;0)BED*KcqlOm?EoS+{b!bcug(328$5wDc7OHU-_T3QA
zIeMY#p&Wed$3W-^%0-eZUW_4)2Qoz;&hHK`H_<|p$Be7dt_gB$vh$zaUhG`zEd0R5
z6*ih+l(0NKcyXMGx_dW?80@V6#^Oa>0Mg7)PS(FceomiP=km4L(71PEg#_Lvg$bBh
z5Oys&`D=@e-Kr7q_4pQxT)G+|_@Zmio*Kd_q4ti6R&!eV`&7inQ{R}^3s_}u6Fi-c
zHCl!o;{P5dMx6_t+jP7LWtmG6O4qes^TWBLjUe4cBo2ZM*GZuyQsh{<AcJ3(rM)AP
z3?wgf0;;|a3A`%Qy@)yAPA{)BDgn+gijPwtJ0c7Fc*gCNZjp!YT&R*>zxd$?Hd(<?
z^tIH+HI|h0iB0-dBT5r=B?%uv=TLl)b`gxw4cgSPQ3?AdFr<@>J0tgs`-*W>fglJH
z$RAWg<|DVt`kjnMyui<cnOzbTVn_fX@v3P!;-we2?s+()ocXH)0uPZYYG+W+P-{6V
zE7r_QD$lRa&B2C0bLHGnFrz2$5byl!8n*t^eGePipLmG$Y>Ql*F1%P@$fHfIrKJcX
z23euaLs8M%rmVj6;t!A2$M@Dg*uM=&b}CnjLpsBDU3}p3?9rHgat_v{q>MlzpQ)r&
zM;rRO#6mgk5PW^To^sVA`HqDM9i@sI0@6tA{Gm@3?K0yptGoHPI30XzI9H=Wx)p;|
zSHKKuJh)PNF}-Dzc&r)6Lf*?8)tV5H0@>tP67BWvvCWo&J0693`U8|Y6_Uk6mS8>8
znA1926UJ9AjBB+9jCCq7SFKBIyip0OT7&pG7V%&_<PTzIvRb48aRCO$!rkF7Y~G{X
zA%5~1A%kJF_%bWAfiC#uSLEN*;7bg>)y>tW?gPCix7t%{8rv`Wb<QUEu6Y~_&v3SQ
z{txjYw;NAtBPFqwB9U~HYhLaZ{$XW!tb4j#ZG=_8mk*(u>~7xdHmP?^{tbg~vCX*5
zb&|ax4AeqIIyt_B*Y?8Pz#!bTvvCfoY~+_r5#rjfNy{#W+N8OW^q7C={Q+=5ye?es
zRC1i`(BE5Rbson}wjM^&33`SZPrbUQ7m~TdViB*eq1cdUlmqwuwQ(6Ou~;d^h`N14
zWDDok@MK*5dOs?(rAoP!YCOWd80^Bm4nHN6Wea_QSvZ6OP(%ZWR9>hAR<$v7`&CU%
zA#c0!zI1CGoz;C-Hr6T~a09AqSO^H0VA5MJh6M@E0#YhXy2V$hXA6vj#KYcAvqF$l
zwLjf79`vFoe6797jM<aq`2#M<d>5KW;zE!8x0)suk_;(O;zDr+pA{2L5DzDniYHmJ
z6HlASlhpFZrj|hq=DWndZ9&+;mW)d7h{2NBr>mP7I9{wpMT=!6IeHb|vs;<>(2C?4
zBB<NjTY|5j;sS!{l<kpN6xS$$@F@gE+Q~1M0)50rE4aB2+iKsNk_(5~2&VxhImG%D
zGy|QwmUP9%acFf9{s}w+3YJ>OD)Wrg*80mwvjo=iDU=na6gISr-r8?1Y`<K7jG3AV
zRMkiotcv?<pw_aEW%9A#J0F3Gj(!+y217!u%kAnJ=D=5p51m{*dY&7ZBC+*N)+Skd
z3~Q!ntAn)$O^130V5c0SzE|LPGy)0%0-A-6M#gPDSgIpr=EV<bLgh(5;qD>C(MO>9
zFqGo}Bw923qychh)YYqa3KtTh^R8~@*24V-9t^!xpUPjlOIMxV2(=VM&(x{9_WN?G
z;f1h04ji>vne3g~wp7P^NOt`7KIoi%<NXNL3lwN`+{bi<Zt~kI_gvJyHh#(JSiU>`
z@8MQ_AyH?fm!k#?v`D9vLZt<WSS~!xxA}&(7%YtyL_c<A483|naDY{;vqt%ZkAptB
z^__0pej{1as{7DR0*pj1o}@q-x0cv^rI0GEj?4GYKL`dFhSakoYLeH~TVKu_98wNj
zM*j{cJK0wru-wpM?%7p*6VX{aXZX3Al?<n3FzGf`2t}mG6qCCdX7eXrOEgHOc$;iK
zR@bX*jD#TKUehn$DT{W4A8GLp&VJ4KR1$v~4-=n`;fQ337>Lp*iw<QSXAD&deprb9
zOJQ@y2GY)c^-|)dF71c?kdS>pe8^n0r3s4>htL<mEiPUtZa6h$xZ3fUWBE9sL(P*n
ziCak_IVVour(dWZts^TS96uB5G9}*;(xna*tp`aiRBNH33+lU=-Ag2VPui<UJZeIn
z?{y55m&L`=?>cCVDEOPagYh13-hP@@K@{8Oz{4f2r58mN=$Z?tb`-Sk@A>UO9l5hB
z=W>dSf_vae(A^SrrtKJLDPbMosGinWxMY>-uqQ>OGzIKMg@^mm4ll&=$$e)CtdW||
znfmF$b_aJfC-MDFQeo~c(XX~R&81l6MAV@Drp>cs-HItVmQnFWET~L;KD!pExNdkD
zGC4X`@i3JQrsFR|Uc@-Qr<sfI8SWe$X*x}@-OTKj4qxxd(7#cN6~^CTL~}Dwa18bY
z_`(m^8Ho>Tu|;KK&M)Lf_=M)1Wr&bkl@tSiLp4{-TPu9AwjR_3J&tpS_(=kxn|69@
z*IqU4aEwQ|#9^N8=ZbWEB_AB!uZ!~>Iok&9qqHWm;ZMT)JuycWd+K`u&eBg&Zdf*J
zRTX79(NqFuOD6q(@U}%-u|McFv^mQj8iwt^IfuBpYHFPJ<0y%CvTC9zYNaes_#mw*
z{zMxoMqgVqRue++UtZ|+iEeB5v4aZ)5d2(N5R8Av9pk!~;8_xQ3Z+(UmUcKEOJ+Lv
zc9!(OpEOIVd3xWDc0_Yf?JQ((SeS8yH5r?dUVRV)W4V!AS_CTN<&7Dakv-<!@y;1N
zydULdiJvX{hO6)X1z3-f$6R6%dtDE@6|-1BEmtBz46YtJb0{QVDdK65Gk#hI^}Ks8
ziP!k|NRD!NO}p3SV-<##{o_y9rnJs=Bc}1BI0>Xwt0D`IcInJy!7xLEbZI3ouuV<z
zzU;Q7K?*I0qvZ?Tafx~3il&VLLgs`wlIgsmC+d|t!!1B2i5SP?px5?gZw;+K6BV-V
zyy}nGf6X6E;5}$U@|*-us-!S7KTvUs$Yl&<Bn~=JVaI|X?mh*IF9!K-*~YLr6?7W!
z)YN!J*v+v$I?HBCyO0YJKB4lNMx>eMQvE5}2G%E9A@+wj&SJNQ&ly>@Yxw;EXxVDS
z?OG<k3uJD;mm@tx;+6PB4;OYDnS-yoVx#6Fq2eZ)Zp+#q78pK>`dnT^B-hhVZg&I%
zBbbkh(|+bX`ahHwSoibS(6iiUiJj;yQ1-lSRUsy7rD%-uAd}WpMunXh3lP`Qv3b@k
zq8{)rw4?TwmizPuQWY_8AcCD&P}3LZYI!CR!>PWb<*?~YSmrPyxgmG-T2ZGcO$4D7
z-e7L{vu{{stUy+4xwe)v=2H1_@`geZJrzYy`#-$(k}{Xk4IgXRaSX0YjX;FDaF6z0
z>Go+JlbT>M4f1SDcwYCNnJ_If|55pThzvjBW{dys7~C%XXlC8TKRzWE4MX{v9y=_g
zvL}9=;+O|Th=#dunpKTBsaUWEW-4@N-R37#lbE%lT=*@3TVsgbDnKvXEf#xKl#(7o
z+gyLk5P6JrG!SAs#aCy7c5qVmYNHHdr8uz%CirK+X#Q0_jPp49v|L&gjBD;f;+<Pd
zaByB;o9UVoIbXncDak=|Amtu9*4z79FC*?I`<BwqZxst>-$e{^p|U^&XWnDMQMTAf
z`%ui2x#9ycC!tLwvW$ZceNF!c{F#<TgLH0`tJmax(pe^eTPw3k0f1}(9S}j{C3X3Q
zQpB0Yj-1a_r-V53SC^VLXUWi2yhEVDwblLynSxrthPyjaA+GkRo4!nzZ!;b?1g`ie
z0m;Sp5<kX6vqA8+Aq(uwnxZeRECcZ#RPRFzDB>XOP5#t5X*FYZ(LK(Yu(feQk#O=S
zheD5;nD{_w#CW6dvjJdTTb_}0E1QfVmo_m8hP1=9bj4;0jI@*IT&)m6Z&M8|oYOGv
zvHI&8Drw}U7vMV>l86@VOl!}?tGBhwr=sZE6bkY!j4xd<*f8`P=oVZsr#zY=K{e#V
zk##D*P2NnyVWy*htc>WZrrxF=s33+1eLZC?+Qf%}lT)|?;zC6~g)1uedy(C6Yw0An
zHSKoa$Lp&?YH(%sdv=};pi{-dP_qr*u4hQrzH>#l`S?d@SML=@YP^F*kn+!=1cx7t
zZFGB>SII~Rtkr7!rO~v7%XLB8u}<*S+OMfTCXR1X9PiwwLvUeyIaF2fMMj{xo0u*q
zw6%PK_AAhDdrRR`Gd@?irpZR&>KR|NfvgOTxAR3kV3mJP*O9LioMm70&t`4tIoety
zKETFs2d6>Rk1h93yhmmTM>;>d4BaCvc6{`3dYkR{aq-H%IS~i!TF};Mg{>azv#IMz
zkP;~~**A3K{LDK5a_X=!!)A*?OShu#8%9tV>E?TE8M@2#jb_WVb{hGqg4j)I%6tqh
zG<2(ub*2@)lgnDIc~O*`6iK{`e=$3!Gaz=JG>mtF<0md<q|qVD=PisId+L9uN>239
z>}}z=&7m5_yY09lA!=90W0fybjASynF0CmJhNquD0d2&~eZjmvP|Q<43Q#2zPW|Pg
z@~iCunW6+xiyHnPx$H^p^d#UcEIf=tUL(n19BEN$w_v0hC`ZY*?89Mc{FSmlySyc+
zGd6M~kuBi#!2HE={SDa;#t{DE>(uQLDq<y$z}}!70xwV^CG3cEyq_PMVyHG!PwbW~
zRLFtNI2I!rXXC0NO$uk)bsivq<H2=V6oCc<7UFR#!(%g5L^D4Ab&<5MP?l&jg0Y^{
zXr9_9D{M*2YINeWWDFtzGC7+Z&a;oNo6U(NVqd&tHUV?2DD?tw4%ar6D|=kWmniAt
zJsAtS%oM#%JArZJ4u30Y`wzeJ2Dekt62{uNi(3Gx#2w8H8~9Xec?1qIRw#VERPg?k
zOqyAds2evmq$obJ;*+&cOmElAVq`7d2uOUsUrK6_{Z8*&#p?9@beUoCFg5uM`B=zb
zXH#Kxa5uK3E|Y=Oh_)`*d#x9eSf7jt(y5Os!k){ohd(mT((T1*o&KXspZ$KI=tVN6
z)~g6c40A}*g3G;li6E;%-Swjzaw`{AyxiXekx+*Vk9$J!yOuen68;2WUQmN6O$|2d
zN5A4+e!5iAeU5iV+bS|WvH4ADW5!9bV3_hjZ=bZ6;SGN>;&<U5Dbjcs#;qyp)H}{R
zd-1BC1_*o2yXpmOB|qa&Z`pgoH&EEJ@YQwb(Oy0L>2tT1S_bDh@8K`vog`r+zM`yK
zdKvS^&s10L5BHGuo>5riuHvflcEU1#(W{FQtekn(#1<*eUf-of{0-}r6oMhX$AN>S
z+GI`^J=sQ@MjXaMi`9IUI_M-%s%daE?4?>Hv!d-ZRJ=86EZ@GKmBM~+829QhCy#0z
zwKf)p_zEb8R<ElKt172D{&Rqjq47ajTp_}`?6amk><L}rkK&J=G(-F??+u>~3NY(~
zZq<c|-)H<?)I$Q}Eb5Mi+C<~OO|pL!g}$uzpN$7VF{{bMz&h=^)=Y-#2{;3Z@)7R>
z8xqpxM4jvc43ySix#s)D-g?EOtON`X5@Xd;O_pGB1^VO}{*L3fAMbyMmi-2{H-XA0
zt)G1$upkE8V?pS<m_P05COB$j8!uTGg7aNua*b0$s+f|*)bve8r6||EJlQ0SC#tDa
zhieff$5*$8vck!m2f&@f2rqF?<+9&-ZcDC6JQ<T$9(?K1%yIj9GxtkJkjq8@>^#S}
zWy;d~ny5KG;MdYJ?BRW?&HIiqS`*bpvy$*(hAHZEL{|-xQnx$M22&5u*zdmhE+wuE
zFWok2&uE8FCmzSD#7gPxW+F*1Xt@PtHS@^hM^~IgRL<1-v|oFrw%*25W#qmDR*2}y
zl3lNX(%La`QaMm6k4+@y0KOYTtKt$avTMsyk<>FwKRw3{Kz^ESd2C!1X}L-{<G`s~
z+kyT*PLNxl8^M5YY(~3EMC2O#lEadpC}DWJjm8qrk3**$L#Bj?{wVW!&G7a{0C=}~
zza<7{8ZyLSS?NbX0Bd^r&Z_$01@lS^sze}LkcRs8%a07J_)fh%u6luG{L6s+9((Tn
zFQU+5PPP*u=u7+jUH~A{?2;jdArd&Tm=B3n>g^+IhuPCFpHZ|+mExV)qtJ^-Ei$@O
z6R)d{qI{Yk(iht+s7z)UsSk;>pjdRoYcS|4d*z9@t5{)0z)(3deAhM0)<Hej5+5G)
zQg_l^;aw_f0n&F)N>rV*D4Ew_45+Nc1iwUh5k-%-a1V+Jnnj(eINRi=P?cl>uFNux
z^fj6K9W|@r&ABV?Y6uEOF|GpIiZIn=9|#3DCCK%41pg9~6K^Ar5*KsW_suG6$0Yl(
z)HhsY4QFfa8%sYxE}TV3M}#Y|)<7#0Mb!0~vy*cT-6h<^be`Tj_nu{3`Vevm$;u>f
zBEU)xO2>8}tH>`$u<2jd@uRfB+rk9}3;R+U*qnK*gQ>BHkX-OuG@>BPH7pXlVn+>=
zT-uCGRg4Mruos5`)^G4<<mq%_A08YFU!}@&vQC(>nDHT4T&zMIK6}EveXQ;kJ6MQv
z)t9XGxhZNmidcI*LBC<HGRvm<Sig5RPXG=1jOo1(p)O{E`rTA<%!Kxkr#Mwlza_a+
zEu|DEkZ-Di<u_E(sA+}qm`Tm!>;|$-;o<b~`WQc0bd?zV-S7vGZWGR|0+B1GNuPG0
zJjXWqkhI}lQ`bW$KUT0IpY|7buB(2*ydy69DU>s<SKG<Ky$F5oAJ+2Xq$W)E1mRcj
z(^eg?yCRzTO*^fICclcf!&(<la9QbU6&Bo5IIMDl|9*l_U|S#H2903hDmB=g1ZKq`
z4r;zlt1BhgFvYeEVClEnk9L7I9vtOfLsB{4my&XuE{l8jyF<!}L(4(uqE#g8=TICf
zpldke28OZRqkoKGEc|Bd9U=yJoe^nM6b52>WBltEHuKPt8@((^lbLm5N>?4Z{-90@
zIo_~5$q>qWa9et71{KN20n{57gC^&pIns^KcDYuG9H<zcXUQJj-Z6-1?Pm$P5Nndf
zJ@)wemA-hJ{ZZC0BENUm+j_j=Pdbfw?(BE@!`GBzzL>39<#)!Fa$r&FFOMQs4Qk-L
z1CS5nOSf%zFhatIy=Nos1x5@WjcSXk!XLdnK2M<$Wd56INtqj_ul#Gv9k@oN^wLtF
z;!j!T08#gcVXFzw@6>4%{Rw#f!qj5}UF{s;VVJNYMS`5U_`3UJYP%B2I)9LXWR6)z
zI`eOnc$~I?&c@g^*(WxvQu)kVavy@*ml+6^euu;nWcUl*M`OdXy737wuo^y^3&h18
z6gLuUj3S2iiK+gZttAWZb^py)48+Q1esxSDB<0T2xR-aIK8OCQaf~oFt%INBsn^g|
zWExND(BFF#|A{?o;^^A!w+bt^)*6>fzSSUb4U>4<9|#WEIJ^=jPCaJ8RmRo3+29UA
z0+Zmn({20pw*(8im>~SYlt4$skVLlA2Mg59xaCJ$;ceDn7PdViFW`exlGJn9(8;cE
z++@Ir*!h?ef;sB3bq6!JKZ;{;Cgs)NsUmL&AR%kmkQ`cwms++qY($vi61wF`Nd&_g
zUv4?x8vhPLG8AF0f<jZ!6~24po~cKf3j4jlSy1lDy~k~UqD#=~pj57Y*`jtpHp!O3
zVNqA_gjCM$1-&L;#6@QR<fFitx@Y_rie4EHw3ulH-_!Xh+%oDD6jI-oSNdqAs)v<x
zyvFY1NyhDelN#NKb*pvtEnRH_IgMCMQY2M*xet4vO^@hRKhm?7Q*y!y6hHrgcZ+)4
zidXE*QZidkY1$i*x62OxR_oFePCx_Re(KX&`b3<|hUT=<0MTW{Sxm^@<xnf!#9j@!
zN;P9b2D_E@57$rm(=#$xj+ka-nAOaDI*Pg07(G}}*=rB>$~akn=v?Z5P$xu1*0Hfr
z`rIQy^|u|jcHL_LAH3EwO|zNBeAm`bW!C>%zo8ssGimkFr{j)%-9O6*-Z*tau9bGI
zyj&uTJ8Oj9A(_sDCjp}o^9U1=w&{uSg>SU&xJ<E(WZ+PL76}Zv^gMA3x~ATa28d7P
zo|Wd9tY0OgLz^+qP<79c>$j1~7~DiuqBsY>JDLkzM&XzQ#xa|DFNZL|lv%2JcSzKB
zLobxr$+Z6n>rYo#fUUjb_}f0DK@%uv&RF87V?tNxqgT(nK3*sobZBk$uKq!<?FG5B
zWXAQg5xusmog-2BMci-jC=65p!31&BU5*iVTv_5A&<+#4TKDV^(|_Jj&yYDphm*iL
zoXaA~<;fsVH{<~|AE}4SbneK^1N?G%*0HDSp7ws;D8-%po!`N5(qW0pbu{Z2v&W1^
zeBDZ$=9BB%l(pt!<^tTVF?_7`HlREnZe2TkjI|TUO~gJ6AWt%*)J?8Pw1`?<F7IjY
zPw!z%o|C&(UN90HVA_rJurTqH^!!_QKkdRg+!^vq`C7M+Zl5z1!>8HS%etp}&#f6h
z)3t8n+lS=}rWlu7m4D^-g+tT;{81cDJT*TrH3U6`6<^L!JSsZ%^Nn4_d89T`|9YF6
z0OTfJ-QRTworI5~!xrr|nj%_(^3=^7nJmN*c>tPvL$dw~hCF0Igxd+}*BD=KP5$}3
z?mr~2OHFWqT-MU?AlmKHoWfo#(So}Pm4F~n*YvI8-7$HIAxgxABd%XH;TE>&3a8kJ
z`=)Gu?cbSwv($vT7g00?k)%Pl_CcURzlDIKpf$<pEjuDcH~^DLRtDJ~(yA++=hW|Y
zd&n<6(u^}0k@6vl4BVg1JL);h<yiS^KiP97UuBjFw?{@~>WA6_bYqQ@hOI8yCu+5J
zsk;C@yi`ALEZD&>x3;0dv-%TiUBLwr=|<OUU*7(?a$&w@sSGzedUn)S<rjN2Dtl9p
zFY@7uyEZz80+<;neA{uw2xZ<Ezva}Z{r2w=ugk7?zkFv3sGL(mNih^tu@b21714X@
zJqs4XFkNo!l47j8Ir`)d|MCE@Q_dmYX#FFjOU>8C*Z1?a;`d&=JKYp!ytqhy-Ok<c
zO8ag<;tozDlRt+(*UafSg8f&YPh4n5nD`!%XWz$~Efc<OThh`9Cj)11A-44B&Q?=@
zg7O!ws()Vwol6M3sO2}B^^8v9tW$_Et!=ld4fo09SA=Z{Rzfj@f=+m~y><6Kx~y1L
z=6~?QfLoAP>ZReBQ?iI=ZXY|^!1qtMRnj#bv5hSwtrR}%{J}C21X6=J7UnDo)W4_$
z&HOwb!+66jJjBfD&zWK4oB^8+jzUp$bQB)HZ$xtwGHa@;Y;`UO!99?#!`BZ&7M5w2
z*+SiO+kY)am$a^Y;wst-_-ZN&9a$u_O&OrKl&da=gQnhE;wy#N<ttLh7G1427p9J{
zW9bDln{Fw7mb|gAID4D@agD&jb3ve`I}b6VO87jQD?e#!x<Ah*=I2MC<~Sd$^#Uko
zDGo<SeWRuCHJ_vReP1O+d5#emO$ACivR%$%({UQ(bcl=VY|{Pa&K;y^MN6mN6G4cZ
z3faZmhVa}VIfa?ZcR7X>?j)fgE#*M$AGoUm^~&&x0^_rqoVw~-KO~ZLflOYajE7@}
z@w4ct7Ybudgm3n*mg2`TR-heeluG4;&I6E+5nt~~Ei?u1F+j4?0#<7u#2MWV8RKS=
zBsnp7>~&-GU*9VvS#`xMBNxSIN0(71pQTfxBY?%A_dz}*?5i4A8jAPStW0DuE5=ic
zte!Cu?2HAEG{bYrfo!i8!ut338Etj?;jZ^oApX=DAmLrZ9b)yvZ!?%nr+KjS6YOee
zJtP-t?bLhu6%~v){v0qk?%Oh**wn;FG1IW)=zxR4pNpKo2PK9SVPhx)DGOYx_zo2_
zm+wuHk3vJlGbS2cmN~|Vef`lHqxfzh|IaS{r-cD!`U|5#*}{akv4z$XWN254WdIN7
zv`_4~);PRwUfMRr1)tsQgFPSaib^5X5sHNe9SupFp8-jdd`Biq(@rq2(tY>D=rF-O
z;yIE~#^2thgDj~f=uuJ{lzxfQ0~YF<-%uwamJ0JpcS&?{sX`tKQ3~VlmoS$WYv$Gh
zFJSv~tdCi-q?;|San~8$X0<Lw-<mD4=7!3X&44}s_JosfmD^Sfa6-+m>gV7O-Sg<g
zG083yPfd|53V}V{4Y?HNEY7cm_$?15L)Gojd(@tL!=>$?n;P@%e~cilAt%Up&^Zi=
z=f*yMrr^e1hKoGn8o|qqD68N_&_yqd&6n%xw;ZW3L;3iMhFOzeSj%TEO@UM|OjL6c
zV#YP%11t8)B|nQUZGtPR4jYtMT+5tJ7-}hPFj!vvNyI&CUk$2~dNN(?_9YjnWg7SN
z!S-R@j%>2lffK0&qZoKjO{i_$Zz36D{3y*mMb4^eDhU)&tzC7&3CZof;K$b4N@)9|
z%Cill=sL8t6SU84k8tcY*f>e7%H7gR(e}Y<Mwn%8$~oh2BX2<T`&pk9*ARh6Tv3Xj
ztHgwJ){}=Q64gr(7so8L*221r^LfP}(CH<*F@)(qB04=7v7SZ2(jsZ3#Egr%&Kw^C
ziykC+HV3`qlJH08m2INK?@MC(Js4JKp4vUgB3&%Z2PfN{UD7M`YCq3D+Jw_86u+md
z*OQlBXgImP?|Lw$KR?e>rXHzYD9pAe3+VWb`5<h;E%TZrIDUgeY2Tbyn(Nh<6aJV|
zdHOY4%|cfZ{9xVBv^0um-&WBEGnv8$D+v=5RtE`A&Gk7Jy))MmMq1bss+a-?qdLt)
zbc9IGOkiw-$Z39KEH4so*HmRm8_#D0nubwn-X`ga(bH_4C(+N*)srhUerqXdvL2H)
zrPX}D$b^^cK%=_-%$@N%aYT>)w~liBh{Fo<x_OW8{q3F~hYtH7LAE&z>&O~jsYZWs
zmHd3A^n7P*ZUPjw2E$(d9Sl2Ma)i5+GKv-ywL3`AeW&_sT*=YPGLIc$J~SxqP&eXL
zy{b~H+E|k~A}E$2nKEn{7ErKq{}S$2j7jHqnYhShQ74mRKuJ?Bv|uN8s=A}QK>V1^
z`?9K$IDvQ!?tDzjx`jyggm%F+#Ir2EJQ)`~yL8fU?`_z2fk(ragokn@mriBh>d!TN
z0Mz;ULD=DM;r?r<g<84E+PWSoOP;7Ixlg@wa&5@_5u$xi#d{L%ymqEsT_^fJ{Ruug
zv5d1-FL0@TAO9V}hQQG(!N$@HmD6LP_PsXoX1KC^V`t*Iv{AYg;|KP~paz@z6&!CX
zT!A1zj`qJhr+6jT%<JZ=6~p)^=|%oIJLEVtACh!mGAy>Ae0M*>N2D1{Y{nnn;B`p3
zQoYanAxrWl*SbuHig_1xgidy7M3D-@ZaI)RQvQ}iBFCD0fNTT_h_Ttt2)St8$dA%b
zqXo$3$Iwax3iMgVxKv4^Gc4ZrDGymxL0xnAQxU>``D?!zlKC5u-U0tBc0sGpIOw6k
zTP6q4w7zAaUEr9DZ7MAAcswF!dhKLX@h%INI80op-xGgH$hFpd_-cAL2-!Gv=gbp<
z+-EW;2zxcRDK#JTZF8v%NoR9ZYl`hMKU^|v`nJU;9eo<~=9InlvxP~EY{B?<+2Ke7
zr=mEnM>19Q6azlkGO2}*_FmenmumMCp6fVSI=&BzHs2VkXX=&ry`8uGhkB^;XP<_N
zEn5TKK3=t}=^lz71z%uKeE<1#vOkXh^)S(9xNl>4hq;g~YL(kM$)a&5^Z!pl7Q}hm
zh?Gc`Zl905%tyStRu!ZMcypUvC{&rki;;@1yR@rTH1T|uinXPXv4;#u0?vrNUhoNB
zb+N&w6oKyGRg=s2yzwDz6lmER!xZu+90#m^N^nV^RfFE8H|(-y3o>j}wD+yv<7*_$
zbY6;`4BN_KwNcI(nSpYOL2EM*mXkM4?_(ObwF7ccD;J6vBZ%~05dO{*2daTstmb&f
za$7}7dCuY0hiF}fm{qtnIhH!&--EWeQ3rhVEio<9x{nO~e7s;f9T>qzc*X(il_+--
z?2+8QZ4Ex2dB$&JFmSPxdTbM~)G{p%`aFnIyKgk)0=UB4->G(!@;lzk@^<N50FWkB
zHB}v87x$G$fEkP?<g3+()*O#_Q+^>^F#B1@L8TX*9FI$kfVio-DVL@;A05OtN01(u
z(LX=Eb5@9V^cApFecSuix4OnWbgo~+me#I?LsmiYpGq${S|(@^-8D2Ct^3vWtKU|q
zo{d=3hSt*YbD&5u%@uCDGQSqD2Y+a;3&X}O&=+28A0E1v;b^Siq}%X~!DhWbH%pQk
zfAHqXy^h@eZ<n1mgGg7X5Lg~<<=ppY(TmNWdBLndEOjShlN`cs7^LM7z~Ilr@H1nY
zC4!5Cf?|C^m5xcv(=jrIXJ8j^o>{fgG`Mr%!t30M#&>5VYzo*ddeq5B0Z1v91|w^!
zp#h&3Lma~d&cmHtuG(Ylf{gq(wYOxh!^FkV^30upLjGG0(*?+eZMCD++`BQqzW_D1
z)!&-;XqAPrei&h!7#!r6Mek@Z@%w`*$H7_Z(SqHW+=X1@4UN;kWq(`@glpKi!G6sI
zH{bo@Mh)10FuoNEkMe+s-~U-p6v()ZcXMnREXkS-j~zSzcWgs|O@$(_{QQgcor{vH
zs*eBX5X&iQwCn2?JQGp8=}Jj-L~8X&gdyqx-wb|!_elKx(~NEWm*8;nMvWH_NuxlS
zgqT+!i4~@>Ps-`ld3IX|AO^HyuG*Y$^xQ^Lxr;9G1Ix@dc3GGR>s8^4p=K~O)H!)q
z2Rb7%_$_(XgAs%(%W7GoA`{l_a6lA?CHU59joEd}0jD2fRQHYnZoi7ut~z#)DF47)
zD2owf=1(&yS(yfKQg42&2~sRw952ZQ`HPdM!%p-Uu*|5RECJ$zTGUH7Ii9U?S%MUL
zstM>+%d_EGkPPe75jA$SjszMe65^ul%AZ_<vo@mDvCSiNq4NzCc5jJ~*e2H$Gn?`2
z#OTfeXt19nx}qP%KQCVWjlDe(6hgp$t`<gW9`%cKK{Rjw*Lm?s9pu-uwBa9{P3hJf
zBopQn=<&6rA3r#}5%*U!{e{Uz7ELUFL2Vw3>jtyP=(k<>tVteu;~iq5_c;Lg?R!5S
zxPw!<`$jdN67ZN@&h7rwXwoAfvRDhS;D=+Ukt&liMr@bxsxdP9`~%HLzK&Y^wDr1&
zR@!EgX_Xs=8u@jYsxUHgViHm~w6|}fc~$Iu`Sa=)hxkxr!C`mvhkxwrj5iZm(L~`?
z)bx63QXK!N8Lg?oUjCfjoJB`2#yBA(?ns4Mw^=9r2vdIaRtLd?Fyw_-s6#zxYpJ_J
z065&ee)9UBe4%G1uyTWf>ikRkvVD~EA=zlOZ}62103EkT!;62(?lY)#2C}Y(jPw19
zvogy26L0Jkc-jKEK+H{R*i^Sec0{k6Zw(h}vTere&RH!jB)+l5wTL~TL6Fw&!U8&1
zUovEkpQ~oer_*dW%NO4^93#XmzY54JxW@6^k+Bo#a|tjDFvDKw*l@)7P7%Ee1O9vm
zWDcXKb28Mn^#FS|WT4kJKQ*~OLmJ_hfY~uOX|DrNjg+b+$~Z0E$S67AMEtW_emmiL
z49e*bA&abNtwObm_n57yo{{v#_rA)+e&n%Hl>jbO2W5h!B$R7MBD3eTpZs=v0hdAs
zj`Jq2dbiNZX{@<^`U2{dGeU8*t-?QE`~)Djcre$dyv?gMDewlQMGQ1|=O(hesXK)Z
zXG)frStI2YCzZIUkqa*A)TkZKL2x{I&2_}29pN1T&11C*EPW1ze_|)m!^zt3g!g{N
z^r-bbDA+Z1C>iGZr)MV8!o$I8<dwEz-3{35=k5Y5FeFrV0IzN?`A!ttry4ot05Dbg
zRLPt0kx>2S%gzVoQ?48qa8U>U)e;<i{CU?I9CzY7+j?eFYU~iEB>Hzd$usCEa89C-
zo(G$mS<woc+8AA7J&37K78L31;><d01d0^m99;z29CKT~&0Hi!?PIM&>Os43!k&1$
z-o~Y<DSwtfu-!-}Hq38Q6?^u_Cxv~t7&jdeIo@s;Rn~ZC&f%1i%IFh(oYlZ%TFWP(
zKKz_zl^8GQpU97?I^)Ad1At%MmU?%aAXS_Yo<4@wc+Z-VJMCw@0YIkXaXfhK8Xq#B
z0Jq9G6==kX9PHamr~`1RI+r{<i%qEHmp)WI-(J+<Jd)GK`O2Y<(1I>LV1JU~rFDi>
zkSMAosQqRV2$}|GO#IVO;gD_lxMe0Dwl$dVlwbPaaJGn`ehGZopi+}vV8E7;0do+A
z4tma6VVjTtEErsrMy??KPpAU8LxYNL{&j0MhM0aNSPXv&)s%UjfMUw#u8|jqSAa8V
z|5K!3P?`P2^!(&=Kv?Z96G_PFL(-S4((;pKf1o%VhhK~wyyWs6L`bEL2rW+y{@8$z
z{V^JhbZBlI`Ey;EC$-^kcr3dW%c}vZROF_Jn@t9>3>;T8yByS|pZ0Zw^b|+^Xo;yz
zmQMrY)#Ni#C=uC9D+$klETAg9+OzU40MlNV?vy`0OvT`P1)Z+jc)<t0IlHb+`H3B)
zUa6AW9*IG{<j1L9o;@R3L?ugrxQT&%-m|<&)c!YIuflm0!`LeMItGx;m2PmTU2i!>
zprp91M}a@g1D;G{P8tZtc`s&Z0htmqKPqzbZ_IQ-1JAm}MR4}{gI>{xnc(1um$_YC
z%l}Hj7E`yN=^I4ZKRbu(<@VP<U)<Mm@F}SMH<^GZ{ApgIRsx`K>GdTj0FTDJB6@{E
zWl#*J8Z@c?75uWAq9$pC!9ftPY9`(4sY1Ic9ki>Vne}%<aN~gCdK8+95kE@L&O37E
zD+QcwQM!!kd;4B1ka7f)-3%aH<kjDp2PrqcDPlFzax95sSP=lki<AG5kPmImlM2wM
zU%6r(6CqB5#Lv%H$0&N`Er%O7N$XfBQqpNb<^?vXrA6^zNw$Qxc+(XPBuqmk6lrN>
z<YwduW9HM4^iRd0klhIuwh-Ay9_so^b^M05Efi%<@8-6a)t&1$J2j*%w{*8r4lB#t
z6QcuXY`6LPYb(pK6#y9f?DMcjFN^M7GX`dX8V~VLP1RW_ogeQO|0al}atZS<GZZ@K
z5<2zKzIrF<WA=SO_V0_PVvkd#%8RPR$AGOAo!95f?Ar6?S{3UaKcs^$_os(jyiOdA
z)E3$_oI+Xj2XVdgNl`8Yu8{)Hx#Ws}qwU4jSzM4*Ib3DWTJd(|5iS<BAN+u8z}v>X
zI*Pwqq|Y|hEJb(n5~zbSWFz*xS}3en3-Q2r*zB~eZM?7&GsJ2KN5E)oAguD`XnQ0B
zf6>rk`==S6c%Cm}jNcjkX|sdgVk5(=^fXxIrL_<RZ2v}c=k3DSUMuzUM~VIlBS$8_
zJB@zWSN-oRo1D|^Hka4&CPqK7uFZAwS~fHtIzBNM_Lm5z>0C9-erLYQ``xv7`$HBr
z7f{K10y@cFH=cD*yQI4LZ`sThOG<X1WkY}5?N^-gP!QSm%mEB|%8ad*@sD=(8v#TR
z<I#RV)K2;}2*ipW_R)<0{BUtjg9)ufLFzzcc&Tkr>)`_(HxG|4YKD;%OGxxW{vNW1
zgba#~BBxE{{x=C4V!cnZiw|$xwm&)!Ty9^#Eq*?rmF;?qB{3TurOLy@q`u<;`<jhf
zK)9s!SF&s6A7Huo9n)t5KT*Ik`%8{ZdXM071)PLE!_zR9=_Pv$p(05T!kg&&972(1
z;2P}{V?ie5X@A`AuKT`SX4n&`P3|2Rq1a6Eod}NrsIWq*8d3>wlhsk-g1Z1v_uPJW
zJik^Y)wkLE#~JSUj)auKCCMy`x<lmMk;lmw^OVNkJ7x6rRlxP=a{B(9+SrInC8w1Y
zrESkLi}>EUXzoP=1!L1pN8mH#7wy9nr|Ma%#c-?7{kCO?Ls)!EH8O+M4$Y$@WWkjf
zr59gj1xFnw7pfK;Y|K_l$J)9)Br2+RcJC#KzPpo`BBQ4)Ywn32O%m|TUPmMfHW+rC
zUkOoCb7|_<{WB73+!cxL@FS04_-1;m_aVFcl{=^w4K`pE_m?ONPTs;2D|vj_LFd0a
z8lC@iH|>61eDA72r+;^`hmYu%tp*m{%hDXweZ!vqVn_>{bvZ|CR4Z`SDwEA!x9*G|
zw5{vIpo@cyTly`ij-{<Z!TjaNdS!h88dSxb4vHziF2WaADPI%`c<|QzYMUMOCaMI|
z8p7o3SVnKistpmtYERwwxD-Tx?>iqd;K;Ofka}QOw$d8%>wIS{POz%nqY{bGaCz15
zT|{o3$Ax}wCUBsbZo~ZKqnq}NO7G(NKwQPcW@j{;j3}hdKq-rZsnx&GYLNXDCW-!_
z!?(=uQ<v?#U^Anhfi{%*UVSQ(P_{3OYwYrNwy_jLdnm}%HVf1ODS~|#oF7+^)2pA4
z%<c2?@|Aq2Ljl_MNZIfy0*^;Gvytz8<a)jl@5e0nFc?WP^Jido=~VjOBcp4GeC293
zQ*WiVf0=d{vvbMCn%9}uXor$LsX!D$D6In;o1s6V`JoJjM@);pdsbth1KF&W{eG=T
zK5<Z(@zod3R#(=a#7EU|b+lNbR&nW+t#8v)m6+^KnZIFtDfI@p?I0INK!Cvhu>ek;
zz7nM}K-6gwDa35UDzB0ju9B^b%Is#VG=zkYunJn%Noy-L+R=-t0OHg(9WZ$8`ekYt
zV*gG*1=i($XNt&oIEt6Tj>a%2HLRSqY^``CQ|@fLvJ^%_MCO9z9qLz4lL+CV@K4t$
zlKk8&9I4c;x0E?oLy;gnd=IVy9frQZ`UE!=vFZ+c=?O^I3e^_3I|!4fv7u7TN+L`c
zG9-h_&xGo0;=1GL>-)xX8A5_g>XyL)$u3nR2-$PiXkWQnr>$j3B#W}%TYG*Dv}}MI
z)eG0bu}`=77wh{|g8u??)UOM@8peTQ?EddYf|S0oYyYCqdwO=!vEcUgKOdNz<tp1V
zc#2=Q%^Ceg0v<=9e#<{)9oq1xgv<Ipo06{JWd>UMtX$kq>O1j%7C~)|TL{W*gQ0cD
zm+kDMN8(<5Ss1TPV15rG7RBtr{(8lAT2#`bWCWbZK_VRwL_##~SxJj(h&*%k719X$
z@CK{-n;BR34kK}iITb~T?NnZ@Qi?(qY*=wcU02cYVZtdTVB96;`XtI;=lA}zZ#$Ri
zfgls4I<J|_{_-*JRVryS^iYG{!%#FQva|8k_*I1)n69M=ON>yc2>^k+Z4Cpyo^Jza
zIyMq0zlA&xi5CqG$y{v7X2V^3Io?mm#O%B+knVE38uRcgc^poDiHMAA?G6e^OuIoe
z;HiA~Q}(UYlHa?HF307TUb6>gUQ$2bRL6DHY?sV5Hw8b`@3)gTfi&R12)=cW%Xd9;
z?9!p&x1y6)U$f|0<7Ru5{Rm_PShc%`3%0HgpHxUk2p`NT(-jaqj0t1nFGaq}wggZY
zF=FwFkg;b7de}Nr)qE??&FHZxdZU=<wAvnh-5)h(A4e(jYK)h8|IBVR|46A{Wno2l
zH`SX8K6gTR8J<o8LJJFurCDrrAY05t{qc$&M>ifGEB9knqadQqmUcI|xkpZ#9_W{F
z8d3U%{G-Q=C%j{~0QTzb*=1OqwZT!>p*Q@4yHZ<+rbKmrD6Y7%QZOO2Hth=En??qT
zT;GXh`u>stYNpYCVWc0{SapvfP@adWwv2st<I7wWH*sLId7~ySgf~_mMUifGPp>9&
zIR$3UH*QkJ^mtcFMKzP<t*y2GvuOQ!vPA7VsmDWKIF1bW-_H~o)-I3mshaUthjZ}K
z00>mJjE@UwTgJqkp6=D(1gmJPXjug1ewLLiVU3!v&?JBlj%m0q!f&dt$)Pg%qSr6R
zl$H5h0)hd`y;t;(@T>zmPWyOE&Qs(leG1pK&Na&LXs<1E@ZghfZ*ly^ykFQoTnP?F
zAFDC2|46^tt!x@oTA?$-5cmOGMT4j-A+kqjct?2Is~Ne{W$-%avF>#KB-0B@WNGg=
zmP82dW5}acA~h{Fi7Ee8BFnc=oGa!-Z-<%mZaxwyMEmgj@SW6y*6I>>(;;BEk-?O)
zjM8i;4WtP|02SenUpGX2%ni8P)4(p%EK%j`X;VsLyEH;mteeV=CrMTpFX|HhDZF2_
zZ!x{mRX+lIr(>{~VM`rH#>Y0~d(?7JrIES?nA%`w46&h#H?rsNVr0cFo~JO?EucJQ
z@K15OrEW~}({FU5Q>H6)`X8n+wgEVLl<a(qgvO5)W^L(n-)ZjI+uwx>Nd;mO!oh*K
z;(5l1M|@hcTEE=16t(DdrpQ!&Af~^O3kvetYSUq10-+ITh+p0dE4{%c{kHM$ysvy8
z(RaJRCsli?I&IeMLD7n5R9z|tu<o@*DfL$V_d0E9@{)WF+upCKChgahE9ka*9@%nd
zzM)d;Y*Pd~vS=bsGq-PlwkyyHPt%Xm>lKtObIL`fI4`xCR|_~xN-x=-t^OcJ#_w!w
zZ*`d+nBiInMdN7-0y&~Pe*om6E>HQsJuS}PCRK%|ZRUkD0SIu#ZOL{%%+`qQway|`
zBzr1T{O#o1y?d$qSwI2UpjzFZgljzlLQgyhdB5;{@NxA%XF1^h==aGN&erz}E%+SO
z4PNjq8HC^hv{E1ANNkFcgXrGa5xWa5fe%!+<S`;CCJAC?8?CC!Fudf!T=xIoB^m)+
z-fyuX-%GuX-_nQO80_#MA44+wtqv~4b3d#nHZMe!h7^f%4QKSqWk7w`Jo5XFr4@%M
zjztq?lTiI&xz&qDTIb3}EBeKy4{Ht#0R$}NUzCC0SX7-8I()PxhNdY*O9Ef}4x9u>
zlbfp|j?vTkoK}SSgit0tx^BxWsy<;j7`A$TlcVzI`8>Nr*+T^k-uR%aUO~}IBhXBt
z0{+2tMVd;cXXJ;<h1$HG^E@s*>)p>BZK5DF$T%*@hUPn9DgV)T=a;jbu;jJsAC21g
z<taVTqyJ<uiGjg(yekah-#48>wl}v&PJu%0Y1aTaeOE=JvM4{or<OKjtPwm{V<NRq
z1Z$f)edar8y_isfO%VHR4W$eM!DvpYL|$y*sMzy^HMTr=t;1t1S0HktH6+*ZBddO+
zGN90`608S&pBT-Q$_uH}2U4qKw@IYs6?mn`PBWFdT*jS21#e%E1X4W7=5eBx{@M9=
z-!hk@_sup%Ih`Y)v(j^WCIQoJm4<^lNPOd!dPMS2%>$4I<i^s!TKn!3d1H$QNF>B$
zbW{7O0);8-fimw5s)hx?ORjN}KRVsc=mh~LCE66@n%q_b>+Dy3K2002(*p@j$b0rW
zVtkVAUUhTIBHVwug97DGTpDj}dU{ca42;X)8qF%qyL^u0();n!*SsZ`n6tb6_Lozi
zb!^LM+79qACZOR|*7{fqp}x=`x>n{N?d~gv3-K6!MqqI}{Uy=e`utr{bNiagdneng
z;9&^R7+E_kHpV=bwacXgGtv_zD5|A93mjrcd0I9qK0MmmtCeJX>K$43d?7OcW@cyy
z3JdVy$?rO=uNEl0mP}SImkBU`*Z5EGlL>n}PtjG|de?fooAtptr&(s*=S(8!lbJLR
z=gi_*w+hbn0Kp{!$Om+X9k`hwR}ft9IOA0Z^8065UHP*A#DI9-vUoZ&f&Q?~*AdNy
zE+{cuDBJX=M17i5g<7e20+ZWvB+4uoFnn-t7a98ivEYG$SabaB7AG3y?+3&bIfZ+-
z!;a^t;xl>0g|gwobMmn_d{h*6IhUke*CJPQG<?j{N`ZHKC2m+R3D-XU?Lsa4kb}RD
zgN#lD?Fjf$2A|ip$|h7hONcj$wa}yWBoHN{)CGewhBK_ff#oHb6!{^HU4sPt42q>1
ztCR-}RcSqBF0|8FoL72*(N>nU0X3v!0_P*#Nn68dvwz<#h5sAHHD6KM`lVd~gYZr}
zV1pOdJ>UCW@S{?(&b#w9UtkCHvv+5K^|_TS8iA5{(T1=-k7xxdg}eIMI*fhBXGU<_
z=WLJxf6QLR9lnY6oCpMJbU+yR)innw!cezo$VvI^XS7ArgolW{|2l6%QSig*W=^5i
zSdIr<+<I4QkDMvpN*>>~7m%f)&B1IKe7(o{hOcaPa^wAeWs5^V62lo4GVYw@+D4q1
z>c5=STuy_Ej}#C@NIIVKJ2D0$a*aIVj{ws#Mot`Sy-ju%%nazzgyh^RAM&Gbao(ji
zM=%_GszqJ;WtSl1>HB1MLk2`I;(6|%H~5^s7rWN)iYtXuX5LLoi9N)!2(m^GIog~h
zr48h0HdfMGeEBE$Evsz^tFzaNH64hjJ|)K?U%1$t{6Z_U)G3CR937OKp7}Yg<9g-e
zcJQ~Gk3jMrb|jwhWdYmW?@Rr)g;H9dx(n+aAb+;WL9Q9pD{X^qUjsQmVN$F&+|za>
zP7~xD{vonQW#*v)%)^GdWt8?xC)<3$)RPjYF@9XvGEe#7e>tr|`ObpQLv{TT-F1F@
zfr#z$NN=&K(aoaXHZVoC8c=_G9t-TG&vM3P9{;#D61qRE=ObY)c+9@a=e8P!zf-kU
z_inZ72{>M*dgD|4Xxk#DvNP-ioRsv8fxLi=UwK6Co0yprwOr!tsw4SD_C?KUM+={d
z4qfH{CdB`JyiX27?MMQ!iF91vQ|fzwE7GMv_7{wZJ=VU3xw?(Z2e2!eW&bnXZBA*b
z#cYQMOyqh){3>EHQ-VvBg~fv6sBmf+&bNP6Fw%e<UC_MeBL*a=tuBFaq`CrZIz6G*
z(bsy>${G&LLr>aQH>Db1To;4u27uKkSL)2zfXiw$%N$Td#g6%<u^GN%AC7<cgA+Kn
zxszd)qd*yO^JngfM^UZ1;ziqMF??Wimg`h1?rEHE^EvMsR_K$Tqf<)l;r;5b`MI|I
zXwr^mr_qj`EW=t~m4V$bwZUmsbi2smGA{7J?2vi#{!;&5$e(KW2~f)SJiR~v%f&i)
z+E10sxadD!u63*S=ih%PD}HYr<2!Y=NmPu_enh&M^#Q7?%Ks<Ke62G$n&DLI+}@Fm
z0nu<72#o%Xxc@%#3Xlnq4}X7=%Zk5Y%qKqE|3RNf6DQ1N^KVhN9?wG~wJkCN-gn{0
z>*MC2F2KyEzcJV`%dTcQmXponu)uRo{&;lBh(A_L#{t#IO+I9=RBfN^{1#$_p3XE-
zwAiL#y#DLK74aP4PO)8QXp^KpBzZb8<Zx0JLEQ_g-{HP7E9o?RiUZlUEY};jo!<dH
zC>j3NGq}O8&>I|=1smCqRvS*5_lh$U0TC?(w6L%}naH%#3K7^VOmbZ;Xmmm8oU}Qd
ztHo=W<o$}#krSP>l5+ob?f$nMRrW`bBlb&QIdT{2f}lsbG=sz&wB@CZTi{4nPBbp6
zqd>Z_?YUr6ve|>L288@j*C`FT-qW(U1x3G2-%IYboH{&+u@SEm3P#AkW$EAk15q#S
z+<rTf$z$`1fgU}Tnd}(QTMwZ$5^j#-q~lPD3KJ=-eK-N63QFe4{(!Oh6Fw;llY3p#
z^OsrmJaq|F6k3LlK`W6f#r11^AgtGUq}%9)d-dgaKLsVOBmURkmH$JzwsC}{EEQ3b
zEz6PZ2t&k(C{a38_T@14F(JapSWZc0E7_)y>=`>nGCF0+ayTX?#!DL|62{sL-s^EX
zr{4FR{(|@Q)AJdh=eh6uy07JXUH9_Miq-AdWq&70C?lD*)b#$`eo{o8>MJ$ui*kvO
zmysK{U1*^&Hm{JJ>T6Ll?EG?>XqzW?H0OE(Wwn5?wi>%Y7!T&fT)SuFU;ik6x0Wg@
z%7g~nC8wi@f-b-%A3hbIDBC6^D^fK(x(r6dKJWQ`6_l*e&pB@E-ex>5-J+I5qs%$e
z_k@G4ijwZASeBb@&MY+^uq8RBgq4lU)e@b)n+z;hH-MC|HGwMA#X)Fyz8UCGx`T0W
z`SDWJHZ}Q}^A|Db>=`aO{+EHa0bRU;=H}6dc<j5oOK)7x5Jy-ZyErz?rJd8n%P)XQ
z9yZsH<E1nin33lnfQwWXI&1pw369-FG5gv{OjX`vUrc_!bnI-j9h~|t$nq}J%G{jq
zjh_y^SI6jW6_}+><Qeb47g32+1Mv<v`I;OSVy%AMSdT{%usrQk@*OQ(gbU$`yU!2#
z)nU)1my8FMsmd=ibk*C3!`2EzJPE}sZyOm?BJ`8x4FK?;ReMtW(<hh9ipc7pQXdIW
zfcxCxql*TCcg+|ujR(H>qJs1p4@O_(GEn*1s0My`KGa;&@-p#aVD5MUAu1h-c>V%Q
zz}D^QpwYx+(a2cRnd>nan-3Lf7$_8DPjk;5Sekj$6@6U4fYSgNvCDEKytHg8!KWpf
zvFZsYT*`FdmE3i`#+$0tsEl%?$z}%sdWlD5M;C9OIBeS!t0}dA*qePeKyKBsg3W92
zzW->urqNPwYGd_L1M;~%oy%?x^^c)A%g=KV%vCfFOJOX>v9OIwWgnrnid}FeEau<F
z4pzrz6{k#5$*X~C1rCj2b)rxa(eQh1NdNro)WPZ&TD9IvH(_=Ah?Eys$%<#-jmBeP
zxu?~HPjCNt^oC4&Dz0ei)lfY8?)D@13y_ruzwnZk%8XgnQz<O0^!8A)l?~+LhHM9x
zZs~+nhQ#fsbJ{Hcnn|b1U{zQ-+NmYhE(j_?Hy(&6umD8zhJ9!vFCR67L_xXDk#QoZ
z!3TbCCdY_4jqZ}F$I%;!?gK50Ue*fYn<LS&><wy3Av;qvJx2hjn+aWSWR5DQzV8d!
z2`|@1EO_ccp}Lc}oZ()Pu{d@4>XTvj>=(*4&zR?Yb0X?e{VKy(J|AuxuA-nqwyfzR
zS+G4|v^;XOf*xr2PfgksB8t6&@rkV0-`$aExNwOor?~uC{_4{hFCw2})mAB^7yG+)
zw#(A${WFhqr%gPi#hm@pcaO8I_V4k`nbt01_P{aaN3QL^P%gGXc3bi|HI4(Foy_K7
z(bwQ}bp`nK)RhOe(ITn)l}06}0pgh(vf6LyUj??r(jVh0d*v+SOb;@rpIlv%Kh^yq
znoqYJ*wzN2Afm24mgX`EBvh9|MiPp}I!WM^&=<M+Rmyr$@y$jokv^M)>7samKQ6-g
z{c{tJYxN<D`t&hvX-B_Lb65v@53uHG(BLLB5~d?%hUUFvaInVPB}CvCxXC;T4tiZz
z7E`_qZq8B`vO-*hTa|maM%f!LsWMSdQQisk4=Ep-9vrM<t7jq1Gz+!1ahLh|9K<hl
zPS#ic%%#T3dNb5~`DXfl_rX}%OyTrV(F5mR<ligZo58cEaIJGYK3Vfk*A^-379tOh
zLdA?Boj2g(SHW=wI&6o@FBFf|3K)BVNws$?X25m-?~KM6@!C@!k3#1wfeENNL(Zu&
zc;>Y*uZ{k_mK@0lbPLD1Lu4q+wke(Y=!R=bo>6z32Lyqxi3Ku6?uSq=F}<)3(knTV
zof}=BU4P%@2fW-F?F!h5qe6F56>Z&2GEVv@7&U%i7^-)#+i%GxewFS)?e@>mxn^!T
zZ-bpQuc16LMeL^50PtRjDBx{7Kz-AM+b@7c#*`gz*vPxv1D1MDFtZ{;n!_ip&i)x%
zLAz?@vpI9`5BXp1%24kR&{+d`Abj`1;C&n!wt%o8sIQluEA7mncFQb`$Bo0%(b9Sl
zsth&r6V%pjF#VhfCXxc3=K0WTS~E^5Xl3X(=~RgwJH6$v2+k78Z+=3m&x9&xF~9NT
z-ILM_kYcg0S#KA2%B(q`Jg?6Zc}38iG!12X8iK{}kRq8r_BUH$EZ)Ck5Mx8s!1_<*
z8wHg$%w8$d@~DvM6SItAjCB!>ExI$tpbj7-I%V(`Fo@&XympsFEYrL7a2ENd6}lHb
z_XzF9<Mw+4<W)j9-AaI1A#N_iR$OxrUEO}GP$D&IEW@{cTxNoHANQR9gwUxxTcrmt
z*%<HdCiD_NtULs-qT$-cKhc}lak_&H%8Wwo{{x)-iBK%#53|zq&4`0g3n>B5CRoE%
zP*@Mf;BpS-1@A_lB_o0PsM9EJ7lyD(&msZr4;s0!XbojJjc3D4Z*QE)`TaLAY{%z$
zz&ua;7)4hu&wr>v@@r)8^nUy5_&2krDxIA~LJbGI%xN5W^~-i2Ko|WWDN~ZvzGyIx
zTsHPji%sE3kq9Gs7Sh5N*7rcH=?MiaS2gPR^CF7Ou>zx8(Hr^Vug#fF6}}@RKP&ZJ
z^is2-82(L~wrpN=T5g@f525Ob1)Qeg^P;@+Lq)P8wx1Igglxb&8IU)umjM!G%l+OM
ziJ*M#w)olI_q~lQpcy6X5gm9N!dF!t*6}WjkVy!5wAMe}4tZwMuCe(PSI#f61;H+^
z#xH>YvN}o1c#yeF$1;~@wyKRd$B0%|MU`v<2`0*-n{@c1^|nG7yR;)1z_XOCL13r)
z4_2l5S@_8cV<y>dRpLp5-SDOqfKXomBugY#x!mJ-NfF(!2jgLw*H2@Ez;t~XLZgC6
zJmlOa?lg>^ZdjkLn@yFN4xgu&w=MwUGPD-x6M%XH-Hb=8!&B19;rJ)cj;}x~qN!Fe
z?I6C^9>8aR@u_`873U~8^XgFRKWZ}ZqT+51mKH>7)C@2NB{5MzH#Izc*|4q$H=2g(
z0`Ah89m?e07%Rl#P4ts!J5HSZAq9xp|D_A5h~m_}2_71b{OBX#gWu=Fz3Jjkjp`Cp
z$hW=>OI(|cPZ#ce?7Z0)04!z$u%*FW(Wbf_@`_&G1MjmZOXR$f&OTsa4bVlTj85m0
zTUr6ycD0OgSR80MuDkrUZMd+Q(U1|oFhT9s1(LNK#g!l$R%p~Q9&+nK7YE9<O=;BJ
zCl|h*q+?Fwy?G~<{GVNyPQnZqS;#(wq6-Op0&?;l<BhQvf$%y9Dk!pI%mOY%PAC$V
zu}1v(4`AApt#Vm6aK7ldV?YX==(kmPEt?g#ZRh(>km2uG?0a{~lahaUld$?)2_o(r
zyo-pR_|kTwBI|7eQr}sC`{xfwFnbL6YBun}5nM>weMoZNh~j@c))K3VUlChFvN`Gc
znHG_c@LmmU77Mv5ZFy-qQNqA<V`=lP(8Vb%veD=Hi89Yd?livLnmxT^x$P#74Z(f`
zeKuvDq7V4~k~v79TgcqHEly;OT00W19naY8{w4`bI}(PwZ`#ZyX)9Ws2o)|Z|F=hM
z%bg-W*1gLNqIaQaaGD!?Km|7ILoyeT_tQBVuy3DwzH96!>_8>P^;)6MF|!*cj|=y9
z-qH$q?R;DNeEg`)lUy0~Zynnob06ssi?I$<pZGgsAcHq@0>&YYu%)wjpgFzk7ylvs
z{HNiAs=wU~EM=)#b>!6_htjPrxR^=kLENIT`W7)r`{X-X)iO|tf``AV)0dB5K|Vf%
zlh}ti=&j~%%jM2bax}74A5r`js@%G+_{`e+)-{JNOb)zcr6=fqZi7g(Roc0~)Lw)M
zdYizrORX(@?yi_MY?H5?O)1~=jpqE3pY*grKtWZaZuLQpbXL%+F)ZY{N|yHd(xSar
z*t+E(<&gWB7WVILYrF*4l;>?aiKuSK<gl<R7D=+J>(ymW8+70_5f9lqTdZFcE`Zh>
zO}3bvqnuEl&zTMVTYC}c+Z?d2lpZfD31ey&G?(gzV)@`a)e(2^H0eqU(O5JvnMYWq
zDPC_inDJVO?BTfdtI!oj4##iU-rS=ap>rO%*{UDIi}4vLWe^EwC}aYGsOk4}Am3=v
zpT_uAj@gqiywrc${>lcGI~TztVa?RA&qX9qMoTHX^g(3fk=m;vaB900tV$Q(SU4)J
zJa_XNtoBrsCv^n`BVOpM&aXf9BqFx-Qx??-S}Eo?-<h)aXwIC80VerOh1Jk{RrNpL
zaCHEmJT~(4Y5s2&_fKz#{Fcqn$~JaUZbta;{r{o|otpt=Fl{@z5uyF9OUQ2mlHk(#
zrS~9q_upM!e<v{n8c8QywbuV`D~CW<LOD-ezU|vrQNoj|GI)GC?#F!y=KIH1%r7Gm
zHaU39cm5xN9)sz3UsZJDfUxn;0^vro=*!y@5V-x1U1@88MmovOCF0+0MFJ*i^U2px
zw)~F3(!sJamx1lj@O!Pu8N$3tnCL)Q{l{LQ#p%$<{<R%dQQvLlBEVH;!`RmE^@8jF
i%JHpe{@0X)DW=!=3yXw@ksAyA9X)*9u=J2q)V~2{2uRHU

literal 0
HcmV?d00001

diff --git a/_static/file.png b/_static/file.png
new file mode 100644
index 0000000000000000000000000000000000000000..a858a410e4faa62ce324d814e4b816fff83a6fb3
GIT binary patch
literal 286
zcmV+(0pb3MP)<h;3K|Lk000e1NJLTq000mG000mO1^@s6AM^iV0002xNkl<Zcmb`G
zgHi?o6ovOGdxdP*AltSE*&JruJwUGI3!FN?xxO>s`hMrGg#P~ix$^RISR_I47Y|r1
z_CyJOe}D1){SET-^Amu_i71Lt6eYfZjRyw@I6OQAIXXHD<M{a4P!N^sPbQKi=?mBx
zoos%BSoiGXjr-;%$QixXMOVNSUNp6L0a1Oz&cgu)wqE?07u5I7qrQIu4Fij)Y3c&0
z@0u_#NH6I?Mk(n;dT}d~^J<WkTLqp|RW-hV56tKpXqu)k@V{?amI+5DOlEU@funz+
kySsbM>fiX^GbOlHe=Ae4>0m)d(f|Me07*qoM6N<$f}vM^LjV8(

literal 0
HcmV?d00001

diff --git a/_static/graphviz.css b/_static/graphviz.css
new file mode 100644
index 000000000..30f3837b6
--- /dev/null
+++ b/_static/graphviz.css
@@ -0,0 +1,12 @@
+/*
+ * Sphinx stylesheet -- graphviz extension.
+ */
+
+img.graphviz {
+    border: 0;
+    max-width: 100%;
+}
+
+object.graphviz {
+    max-width: 100%;
+}
diff --git a/_static/language_data.js b/_static/language_data.js
new file mode 100644
index 000000000..c7fe6c6fa
--- /dev/null
+++ b/_static/language_data.js
@@ -0,0 +1,192 @@
+/*
+ * This script contains the language-specific data used by searchtools.js,
+ * namely the list of stopwords, stemmer, scorer and splitter.
+ */
+
+var stopwords = ["a", "and", "are", "as", "at", "be", "but", "by", "for", "if", "in", "into", "is", "it", "near", "no", "not", "of", "on", "or", "such", "that", "the", "their", "then", "there", "these", "they", "this", "to", "was", "will", "with"];
+
+
+/* Non-minified version is copied as a separate JS file, if available */
+
+/**
+ * Porter Stemmer
+ */
+var Stemmer = function() {
+
+  var step2list = {
+    ational: 'ate',
+    tional: 'tion',
+    enci: 'ence',
+    anci: 'ance',
+    izer: 'ize',
+    bli: 'ble',
+    alli: 'al',
+    entli: 'ent',
+    eli: 'e',
+    ousli: 'ous',
+    ization: 'ize',
+    ation: 'ate',
+    ator: 'ate',
+    alism: 'al',
+    iveness: 'ive',
+    fulness: 'ful',
+    ousness: 'ous',
+    aliti: 'al',
+    iviti: 'ive',
+    biliti: 'ble',
+    logi: 'log'
+  };
+
+  var step3list = {
+    icate: 'ic',
+    ative: '',
+    alize: 'al',
+    iciti: 'ic',
+    ical: 'ic',
+    ful: '',
+    ness: ''
+  };
+
+  var c = "[^aeiou]";          // consonant
+  var v = "[aeiouy]";          // vowel
+  var C = c + "[^aeiouy]*";    // consonant sequence
+  var V = v + "[aeiou]*";      // vowel sequence
+
+  var mgr0 = "^(" + C + ")?" + V + C;                      // [C]VC... is m>0
+  var meq1 = "^(" + C + ")?" + V + C + "(" + V + ")?$";    // [C]VC[V] is m=1
+  var mgr1 = "^(" + C + ")?" + V + C + V + C;              // [C]VCVC... is m>1
+  var s_v   = "^(" + C + ")?" + v;                         // vowel in stem
+
+  this.stemWord = function (w) {
+    var stem;
+    var suffix;
+    var firstch;
+    var origword = w;
+
+    if (w.length < 3)
+      return w;
+
+    var re;
+    var re2;
+    var re3;
+    var re4;
+
+    firstch = w.substr(0,1);
+    if (firstch == "y")
+      w = firstch.toUpperCase() + w.substr(1);
+
+    // Step 1a
+    re = /^(.+?)(ss|i)es$/;
+    re2 = /^(.+?)([^s])s$/;
+
+    if (re.test(w))
+      w = w.replace(re,"$1$2");
+    else if (re2.test(w))
+      w = w.replace(re2,"$1$2");
+
+    // Step 1b
+    re = /^(.+?)eed$/;
+    re2 = /^(.+?)(ed|ing)$/;
+    if (re.test(w)) {
+      var fp = re.exec(w);
+      re = new RegExp(mgr0);
+      if (re.test(fp[1])) {
+        re = /.$/;
+        w = w.replace(re,"");
+      }
+    }
+    else if (re2.test(w)) {
+      var fp = re2.exec(w);
+      stem = fp[1];
+      re2 = new RegExp(s_v);
+      if (re2.test(stem)) {
+        w = stem;
+        re2 = /(at|bl|iz)$/;
+        re3 = new RegExp("([^aeiouylsz])\\1$");
+        re4 = new RegExp("^" + C + v + "[^aeiouwxy]$");
+        if (re2.test(w))
+          w = w + "e";
+        else if (re3.test(w)) {
+          re = /.$/;
+          w = w.replace(re,"");
+        }
+        else if (re4.test(w))
+          w = w + "e";
+      }
+    }
+
+    // Step 1c
+    re = /^(.+?)y$/;
+    if (re.test(w)) {
+      var fp = re.exec(w);
+      stem = fp[1];
+      re = new RegExp(s_v);
+      if (re.test(stem))
+        w = stem + "i";
+    }
+
+    // Step 2
+    re = /^(.+?)(ational|tional|enci|anci|izer|bli|alli|entli|eli|ousli|ization|ation|ator|alism|iveness|fulness|ousness|aliti|iviti|biliti|logi)$/;
+    if (re.test(w)) {
+      var fp = re.exec(w);
+      stem = fp[1];
+      suffix = fp[2];
+      re = new RegExp(mgr0);
+      if (re.test(stem))
+        w = stem + step2list[suffix];
+    }
+
+    // Step 3
+    re = /^(.+?)(icate|ative|alize|iciti|ical|ful|ness)$/;
+    if (re.test(w)) {
+      var fp = re.exec(w);
+      stem = fp[1];
+      suffix = fp[2];
+      re = new RegExp(mgr0);
+      if (re.test(stem))
+        w = stem + step3list[suffix];
+    }
+
+    // Step 4
+    re = /^(.+?)(al|ance|ence|er|ic|able|ible|ant|ement|ment|ent|ou|ism|ate|iti|ous|ive|ize)$/;
+    re2 = /^(.+?)(s|t)(ion)$/;
+    if (re.test(w)) {
+      var fp = re.exec(w);
+      stem = fp[1];
+      re = new RegExp(mgr1);
+      if (re.test(stem))
+        w = stem;
+    }
+    else if (re2.test(w)) {
+      var fp = re2.exec(w);
+      stem = fp[1] + fp[2];
+      re2 = new RegExp(mgr1);
+      if (re2.test(stem))
+        w = stem;
+    }
+
+    // Step 5
+    re = /^(.+?)e$/;
+    if (re.test(w)) {
+      var fp = re.exec(w);
+      stem = fp[1];
+      re = new RegExp(mgr1);
+      re2 = new RegExp(meq1);
+      re3 = new RegExp("^" + C + v + "[^aeiouwxy]$");
+      if (re.test(stem) || (re2.test(stem) && !(re3.test(stem))))
+        w = stem;
+    }
+    re = /ll$/;
+    re2 = new RegExp(mgr1);
+    if (re.test(w) && re2.test(w)) {
+      re = /.$/;
+      w = w.replace(re,"");
+    }
+
+    // and turn initial Y back to y
+    if (firstch == "y")
+      w = firstch.toLowerCase() + w.substr(1);
+    return w;
+  }
+}
+
diff --git a/_static/logo.png b/_static/logo.png
new file mode 100644
index 0000000000000000000000000000000000000000..d2b683d6e63b15ebaa604f5e54935df865e12a07
GIT binary patch
literal 139180
zcmZU52|U!>`~S4B;<ib)v?&Tj_MuW_iIk-<=#mK`J2Qq>k`S^pvhOp=mStv?B8jn&
zb;jCYFt#zwSpH{pyI=SJJFk2D^tvziob!2}_w#<<&vVX4;4NJZuH8p>Lm&_?%^TNl
zLm=FE2xN=a&h6m8WWphJ;IZrRjeD*T$T=;}zneS}ZZ{#2V-U@2R}8!oCi?8(CNfdN
z3u|);JrWSD8`6^>^1pA})pu^ccDcu+Om$7UIGw-4@+NlL``^51AdeW}wR>H1+tmBq
zor`C0UwjZ7<(_<`<HWm&SV-D6vop12AJXsXo(?<^s`Iuj(*N?4cgN#IcAfh`Who1q
zUp{}h0*RhWp!TXJ@$prm*Cv_D@CH|p1*ETjT*fBZkBhwAtMzaF#}6nm&1lCrK_Gi6
zAw2|VN^bmWzZ{nRww@cj%|D~6Xul5}R)JQS?s3kaFSr%>0}CbZ)BM=7G}~M%iRwJG
z1p+xzsqKXHqIZ}uLhQ<v*T!uaMzoQ_h7M6ZpJi4B%=gb+;F%HJuSD%)tZea9%86Rl
zGzeNPWMFhm^z<?thNDvHcM(mIV8S!x9!I5*NX+e}@eVt8Dk3P!RwgOC)I?Z$$y82x
zY8L#+pCNUAE5*A3L+B*%d`ow_<~nmoWZdOu3r|br{NjOVlx0Bf4hZB{LOp#kRim`A
z9TS*(W_NrgOs_=-hLIDpFHw?H^`vpt_HF(%A$Xon>{q%^D%i1VYbL&8a>hNlG6_3p
z2T@sK0gL%m8E%C@zB>&pIQ5n}U>3q$&PpVe-iR)pnAB}Cca&KWTx+1I4F6}3bz3R$
zPK?!#g__8g#N_=A4EIAO$%%O<3EuO9NEK-s=^zC1o;SL8bvoEICldKfEOL^jTfZ2E
zBirG8O1fCXfuR<DrTiPVSaov0Ptui&kV%SQq!v5XWZB}-Jydx({;g~TQwT+EX}(9O
z0Y4t^-Q%Ef*luL@vx94KZiRW45V>UCmU5=T%HA}t&<f!Fe}~Tg&rH6LG|2_YYw2Eo
z2I4C|ueKX_9=(c?RbYJ`K^}&$Ab{obs2`P9N0Y&X%v$|-D~b2Go%$kVZblm)4ZYNp
z7@x%c3d{*i^zS(Rw4eQ^(|x%L<(xjxt#<Zxz-?#I7sEQU-%{zFa|8DX3*bkw<@0yA
z>9~iEQy3S4Xe9oa0=c}jM-y4r-08VIL_&S%`gcAE#Iko6Uat(MD*r)FEZQ+hthpr2
zp!){DCUl9dK`EQ>-wX4AK%U6qCKparP;(=9lo;gl5|+LxJAHF)1Td0Jawt*p9xTNW
zdSDyg-_i`snhmR6pojdJSratWjt*9PIo37r)ve<w&%&gqT0Is3(>_J^Ollm?8EL@>
z{rZK0d1B2eAxyp~)z*h9cbtk<yk<N2g*=%X<~<_!jDE+4RIsuq`5-+gH{D)M#-`Za
zoU2MS6gsm2r3I*fSIhVsqJ(#*beK&30)TS5+sspePl`yhEjM#g?s(+s=sR^lPJrpN
z0r{3@JMns9zzC1wX5uKBullxmxpw{@xApKY98@W9&jfoo^D{jO;YIf;cA=T2e}uNi
zRqrHVrSW~5-I=stq6*uTggRHdYlCUuKL(q%BysQSJz_YBy*9F#A7OBl6Q7J>-cY>#
z*ys@f5KzZTOZ$AS?_+LzTifTC7iL~kZ(^*7rrdHWHdpC+xXKDN#Zr>23^&Ys=1o6)
zYmfV*wW|gp!ht#I<Y$M5PI8Gr<AwYNXSpY07;LZ$os4?b6^&5G#9Y_`4SkvJV@UK&
zv;7)VCw1nhRw08-RdKGK5#YSnpYa3_e3tbMF6<?f6NTB;%zyT!CTW+5LWIxEsM3Ou
zR|ow(1R}*Jfg2jY@isK{8A^N`ncHV}_YCT;tGS<P>~8~mOvw2yMY|;I{4*_hvW_pR
z-1k`NuSx{G7sjJ8k}WooVDnH^JTX25?>hE*MxF-(sald)Y76T#wQE1&8u??vU9#bm
z2a58zaVaUYW6Eu#jrVPs`fX4@`$Z??`4&0#$VGIP09H6J<{|*v%c8JK=4eH9NFxge
zj9gRQqvGzBY$PGvD{EYVaJaj?xq9bE-tyX)j5gE!XkuGD!FTomiB2pyuwnA~YtyKo
zlZqvNA~|0}uSK>r#)_+OiQLl_QerJYNg<zY0fhNVh8mrEJ)3<dqvT&3o%$-bi=gL4
zVzx)og5l#09mpj1FjuYX1Shxs8L>O@p#}fiS-;fi*&U2vI0)jARC@2+Kr9}-U{w<g
zL9IrXcI0>(Xs0jjVG2FR(7k`n18gLm)NCC=3>8=juZ>uYJIno<7CgU$4ey;)^zCo_
zWWu`fDs?u`BB&kEw_EoOM+FdeJ8G#0U~ZZ6;gAy*_K~y3V16aaIc%-&7(emsV;?kB
zHnGr~m+g_+sseNJS?DswvhRlcC$QC?(XFii!89_(r4rSF{pJsHMB&&>QEczy`95up
z+WB{2kr#*_!2sTzky(6x@%Sa@gNaF-h1I7~!>+R2_#_bB<|?fK-F13q1XYcf8{XmZ
zrq9d^A80$+yeTfrHw?~B3b3m~f^>J*#SYUnC<W9lg}!Hj-Tyx4!x~U3ZMPFX8$CGd
zzh|7~iyl!QibbRXZ2PkPhZJ--ObT(@?;~9B?Ets(QNuy|FWtJ&tXua^TAx~K;Fz4x
zcrUQsP>WJ%ue?P*S=#ul6I~J2QGv%wi)yyGl{I@!bk?q;dgGMlCl*zXSA+<ElEy@-
z>>qqM1SX9_-?X^FUxy6EBge9Gfzz#LN|g3Vw*44+-t)`OnCbSp=F+lC3WiQCQc7GH
zLktxX$%wvMmxK)@|AbF;QN@}w@ddG3=CzSeNYN{Z2>-We%@CCWS7M4d!1JSDA1O;U
zYG6)!XzH1xTAA}_=`6ent{t(YgIl7Bs(P=i{wKIC)py~!>oCF-7O#HHoX}4%ay`9U
zC>>bUg_Ak}>bh_sVyh4mEA!Pz$++D0=#~=}dj4VY2K8Vk162$l7mMH!2gkTj^oIC;
z!gq2}WowYA5Fn1Pq!-3YyDrMuLLJh?>nN(Msq`H%cL*fLA-dPwn(0$KTLlPorkwY0
zvugSY?s_J#uU@ze%uz;2X<^1<HC-TcgL<&^--+jL8Z6u~^+qf`^o6`w=mgITHw7rV
zB%D8F-NJyr3P-|~xFrpKjG%ZM+A)1a)AxWS1`<}E{bS?E8hFymS9jrh=ZhP>XY-_=
z4LK)7At=4m*5JT+{B^v6CdV9|&7Koo$`fE4I>CZTOEq9xl4D?MmiEyVuGWLT!l@W~
zZhL%wbTqAjAI+Ym2@{>KZkRYDrXTtAu-^wik#~d7^^4sYZ3izo#Bmggg#xP)DvV}m
zgH5;E4M=m-6R}2)`H}TFza#fVHqG&kv(WG)nHR0%!A>2f`4--*CHx^w`3+L#$=Oqj
z=5N0p*dg><ES(vUm$;QL+wBxIu`R;FXBGrrAK?9%eZbIO_c12k*4p{ybH&dx&YnZV
z8)H0Bs&W9#^9DHI(Zzf^yInwh!>*+sY^AV^yq2rTStjM;U-i9H_QYP-+zE(u@qm)*
z2dfUrZNMJo^{VV(jnWP~(pP)cH;;e^3&{*rdx8<53tY?MXrP|Cr)S8<R28%T8C~H#
z$IJ#=8Y#CcQOEc6iz<qotgj=5;$4Y5%npGapHuRsf>7>Q59MGmv2|ig?Ry#$t)MpX
z8`R$3>E6|@rLZ-xkoFC`jsXc4YNCw)ZR>P5#x=o&DPZdaO-Du66qLK%1SBl7#M1mg
zpQ(NOiDIpEc0kz8JDKxg0V6vo-&W_fhT?&~HbxUB0bdwxh(XJ%yYNiQnXgX+Qnbca
zL|&Jl<c+cBq?I&uC9^KcL~&ewGo;F^Y`Aru;y99!tg^mw(IOh#-*vmNkDp<9+E_MX
zrD?{L!pv1L0tWmuEI04M!#^xeJjLm!yBPzT?i7yt`y|IIRvd;iF$Nm8raGiT2v1L0
zDl;8?J5>EWBb#H8108_>|7-5Z4Jkb4*fV<kU0@K;tal+mc(OZ%!av)+JPC66*l5MV
zvK~eT0(n#Og)Y!lk#2-bHIX0^yq`EvzEjzLf95msto-++`#!iXgm)Raa-ggS<5yYR
zzHxe<XLK58`gLprdv^+kgx#RoLKWhdpNsn0ja~5sujoG%P81+RE%KG6Eb`4Or~FS$
zBFMCa3!?zujWOGn^sqAcdrrl^BTGGw@@OmGKUM+HnwWm}(IC-%$_EX#=-m2kry{F<
zYF@oT%uu~?v~F#B7#S6U0D)O$X*L_HwkOiXn<Mh#yb75%t!-yfgLb=4<Z>|HrUk67
zS4lT^<%IE{Q5Cb_$FCZ&$%1-h3rr;Bhn4fOn)-342;CbIBg6I@lwx=tY%|C#lb>kc
z6HV=lL*HpTm)S?pOO0cwxW?Ef44uD`&*3eheXK|?$-!s-tCIzQd*7e=!Sjf;w6Kqb
zkH^~7EeXLut(k=j^81Eta899e_$b*{4}|PWaH?uCRe}i=ILmd6E`WE|&PV9wo0-p%
zANhY7^j?T<=FkiOKlD<u!UEc11Jzc=#B}wu2WL3?R~V3DZf(E(o79_E=~R^MDQM(6
z5kR%|%0nTLrg&`?*N6Wjf<Ry*7>W68;kzvLf52bi0DpsvEx+1aT*Q0LR;lTi&9CxY
zhwb0-Gwsb&=G;k<G<d73-=LuI^c7ILzil2Nu&32bsUQK7@NyOXj8Y7JX@RdpXy^ck
z(V>P|Va3%z=8D~a_~1x#>~6dydcD|~+U;y0^n6GvWcCHwIF$B62B36cw1T;Of(bCu
zQTmt^h64%CGg5noMQ?D^l6)9?zv1j6`QR<>+*aEp#h%5{x*kl0Qmx$v54@?;*-GI}
z>LVzLMTfiqgtIrK`jZJeze&3vH5naedjdceCrT-tzr+N@YVHYqK!zHxL)EO85kB9|
zK=mnX3ndoT`n#T%`L4PNMlm+EEipI8$T@kfE&#SKIKE*yHJSF$%x`d}<O_K+ub~#u
zgw99c-fDDMO$NM$aKogIo#aDoiI`Fad1^eIqEK|}d^FP`KKFs2lPnEJnVXQFWoZcb
zQmmK7;WWRNN2N}_s|)-gRC&PU{|+^&Vz9`89aBMK>84i?K6tq2yN>C~d8t>Hn)|Q1
zoL=$+#RQQO!_ek}R85pkcnag4dN^`)*vm=w4M#QpM39(V)HiSv(>46$O$!CnwQ9H@
z8zuasU}H+$BI~}s%!2yJVhB&|94OIh?y17au5a|>lOr)FEmT-{kgIUw9#CQReSwEg
ztfw!-A3c<F#cz)lV(2eJfY~x!Ym}B(bJTNjAI+RrYkqrZ{NHy04U+F||9VJhXZqqG
zCq^8bU}U*CAaXidjl<Pl>LZNLq59^wLY~;Vj==&vNn#^d$81AS^IP3{o?1x^@KOtW
zmgnI|8-x=bilWej^*GrO>z6;DR$0U^POQG<sbvES?_OWzJB&<>FFgGy36n-to@wN+
z>~q}?mMEBG;%7#qOc-(UuVr;rxl;ktqJF2%H|{8l$;d*6s%=TPoa$OOa82h1*F1Tq
z<2H?&8Gg!6aT$yHa4&0fBj5dKCYSrwxP+Vg1Xi-4ek9aph~@3c#l91sj`{5>w?WZ_
zIrZBiI!1SW<`g1$he5ocMs!xVlW>zL;c;it4c6LyG3@_1rSFrQH|3F+w8JU!x?nx?
zg85q<60$rgu`&S?T@EMFc_a1Mq4p%Eva1dByXq+zg=j0&BVNz=<&;plQeN>NPPr_6
zT5R1`eIVp7pZUfc`q?>V>&gpq0FRf2k`$ERH%2CcurwU0tvAvs2>5+4oG4_H6<NQ7
z(*!c7K5&qIxit@-#j%>Y`P!>$I%>i>qj|ZxOWl5KNsp7XH%fx<_m4lL2Z=7MhfJh4
zCuCx_9%vd_dG<8-mnh&EC6KyRBZfc@l)4qYMy}>|NX*5lhNL8=#c3S=(7}LGRI1~o
zV?kgNy+<Cc=*8tPG!QAAYF2y$ku^Op;08>i{Mf=kqHn)ZfS9Fp>i8_$JD6*G-inKo
zTX^9ZHPpO->et{b25S4ZFLa?lIk5p-G4;ux0>^POZfhF@N=&#k6!9l;oRE|Nmx&EE
z=w)>)ejf{EXG<r85c*E%O5d!_9Y-HTGk@CWNm~{w*zUBIPPl0lPFNepqzAYpfKr#`
zG6v_SqjQafVz5RMQ{^GLW82>F4okXm3b2^ghsDNGv<8a2H{0EldiSsYq2Y-{jl`mF
zd*DiiS~ka1eQJjuHx>afj26heus*e7-mT>~QU}x4gDancK!!cQD(5-E7}V7EgJ;IW
z%~$Ja9cV5Nz6%=wCn&oDRlz>(+9+e9s@U^T0=kyXAvj=<e+U1O_<pwGKi>M0ZV&Q}
z&g-+2v-km3C6Sv(EgEPaBAO68mIqQo^#cNPnzLAZy|dYrcL{i_hS-b`i${9nuQI)G
zXye9?wX|kC%5t?5Mip)4{_i4y?T>&u-cXdgm-h#txDJVDJT~v^*$6cvD)~^HCUiwL
zsnUf=cUJ-;%DQ;EtXi(oIG5QbHZn;g9BZ(QE5}n~Q)%q&8M!oe8DgE|DrN(S<OMwU
z$Xx*0KjVp^`cWUk`0!I)@<Ga9gI?sDB89#cmDE`29}@KV^Ibb7PX${u?^Hygr#t}7
zcYaSNARi_hWh;M7H|mQWaeruBj^~xh>_^m`Jg2gV+MN(l5XE7-%9h38h@|qVssEHO
zG2ZtLJ~MN5Cn$ni={v5T#dpHqi&pQ$h2Du~-sC{tZ*3U4uw3kzxdrm1Sc1SF2q<Z6
zCktyd*JKX|?!K!Blq%ouS-r=gozU18|Jy}=wT#gJaFAsN*QIq(>=6lJzH}MuzDJmz
ze<)C1=?Sxn*dd)4SzOQdR}{I~n3jZeA)!O(rS^cXMe&ep024HaK%E!)1BR`-_A=ZZ
zMJy$LScu8M(C=3x%oXmeMX2YZ=D6vVt6u+UZvM<m-bTSHdiUdRN%RS%WFfOQ8}YkG
zGcoqocPq8IPyu9PqU}p{_^Gm)>~HC63PymOe!y5|Y89mhxf*tnvkOTE-KWe1PoKBU
zzk{G7_uA`zX4>opO}bV)&+dBhU=)$O3Ov`j0s5A|_9}z7tlNTAJBz>cb!K62^?>MU
z90wyGw~~=&zI`Cj#<W<8X_2dmnHUp<Vl9BNPOP@7=fmXNVv{|x78MH(q+q9`5&g*f
zuG?=B^E^>RX*ht6lOp^%f_#L~@(47Do0o^{^k6{NZESkj)KopA8z=S>l)lawt7R9p
zwg#db?4p@MQJ_B+_a5aUyF%47FQYnfqV>sH=dM1DDb9kY0BN0c3;64>aEAblvZvcB
z##P;n__iF6MHg1Kc-f?PvHIsc%>hC7ZD6hp^8pTOrYv^qS4zGE-HP#B+pP7k_2Wq1
ztt}&{LO;)|()Tn~0i8gF8wwO5kOOwLbV0I`ael8!?z>W~=@sk%6BNNfZG(3NJAAZ#
z39uKC-9q+7J#uT)jkHjz{DuXq+QejlZC!Uv?MZKG5PAIreR2L#XEM_k@TGHb(O>2|
zadqeQ&Pxb3768q3V6cTK*bU2lM<@qDQMR<_Yq3R4KZZ_gj7|RFg+|$|(`I}K<vrT{
zSCL}k!mO4RrLdGhSA(Me#36WE+8x^eZzvue^jeX4_8~T9GadI+&}|e2;A_zz%%dLX
zUxfB$kAf&M;CSSStCH&4bn0(D`|<W~;oFO-xG#g%ae&gUAosfJ(iwR&rjkkda;o05
zA+)JRXT<-Oe3IQ_JiU?=!S)Q0U$r|EDN8@Tlb<hIg3hlZ+b==qD;<46*pq(%@8zx=
zo>8a8(nI$aQ-$ZffUKx#3aQ|xl(}<`C_qPDv5y0uLSo*IC6FZ88Ll5aQ5Fu~>Fb@<
z&r}dF|Lv^C>@NF_AD%%ki&sPT9D@7ME${A4nJ##<C)0lQry?>22q1onE<ae9pz7U@
zCQ{`A{7;lTW)AR!zB`f=He42S?$Vfb=YkH0?d~96AE(nFXlw<|@Qh5(nSse<yx-y&
z$GxDg{xhns(G)hi7ePC|GdG=o8ONKtuQEX8Djn*0=WaB!p(IOGG+dokie`<aJ24&u
zX{bKjLQ&#KR&Kx)=BR<73IVvHQwUIc{}aeS#Y#`Lj=Oc&!};0hl{U7@yCH)=OM~YD
zaaUk;>b?msXCM1lLz9kjsIFt@$PvzCv-n?hXym(6|KsJAPV-)I{%Ppd9#L%X+#Wu?
z9gwP>W~)ml{^Wkffs)^ecpcYCySRgxQ21W^dqA0klLFt_G9^q!1u1m{sC7~>>rq4z
z99+yzHs~JP*BT8Lf+s4yk)YeJntP3+QZHQh7nfxrL>)YhT8L2hcJ!K6#DUEaDQl%E
zl}b-gnQ5e(#nH<Pb;s=AXukCW4M8V)4H7drL8ZqlgZv0Jy`*OWf*1TR{|=ouT+i33
z;vZLkQHO06$z5MqH5f*&Q0cH5JvHR6V>&jL7PgGb#F#6F-v@odEgv5<llws9dy8xP
zk8iP_F+k(e8k-rorWx;rPPfLb{uDVu!WF1OWBXXz*sRu2fqDhtW>wV64B7bK)j<X7
zN83)pBBrWP$p+;L*YvxfX_D97y@?kI)LPLjkO@MmJC8G5M&)WwGls{CBB#8^Go$hC
z)sKKGaU1IkI4qXlf_IoJws+T(v<uzEMIdr?0$b<)@3I#wvyiyR@v+s0cE*ze5A*-}
zm><Ot;))g$`xN}yQ2!Ug?@st>kj95Zd6{OUYQW6HZcEhAUO9_gvYEs;)uqcH!+?a$
z`=9XUFqfOIGe_iYa55HDds-Xb)@rO6$Oa-fxdToXk(hLgAMF8QMX)pr*^+*zX#`5U
z_wkKGIz#uw%IF<q`qEWy@y(!g{X!SUJL65tGL6n!EIzdcIlZWMulw)1;+M{?w`-t2
z=JYy0Q-OU!eZ1kM=!kgppz97HuJpw?o|L_s+eiQIEw4<H!_=zfXFvG6qLSAgU984m
zw?|VNw~(D=&N5oK2fDSxEfw77fn3ojY0HWCJJ$8Dx-1kL{$0)7gZqerwti%dmFlMz
zr?uOmxf`(jCQ?4Fu2`gD)^zPd;QW{OUXJ$^-^00TGI2V0rI`t5G&KA0spW#U)wQy*
zMdUc#JF=C=V%E_$YI^=mtFg*V{~hShtq@B=CBpP{s#Bk$U!MIa)Vlsr=ryz}mwn?l
z7OM63+b`qv+r+O>xNJ7unizYb!1ug9t{w03I%h*$%*H^W$V8;5N5nu(<lPVSj(JR0
z6*B6QfBVQvuj-DlfVx9o#DZT>Hq)trDLh0C`n}}^l42r)rKAVLQLUhP+aXeO9zLtn
z<ox1p7^v-^qofayc;$P4PAEb?lP9WF-@37~B~BV+vl4g6r2aHT2{kp>GOM*1Dw(*i
z!6UQwItY}&k6=T;?*jabgc`jfuZh*9<=5_JyOVEoiF82FdyUpyR^O}9pYN;0N~dF+
z*j3L*?eQ9O<e=j<7#5;im!=8rnrrhboDUW<+75|P*hBDp$n?oaq#2!cqRba}*!W+d
zKAYhqi-*{sr!|h<CSGgbUz3Wy%Ir)-2QH!SIXp+WB;=AGzNA_x9|C$k>6|P5rf<v!
ztmqWYRqQ^U$a7R&<eeDS^9fVD^-<b)33^{gY>@;V6j<y_Sxv~AC4GfBaR+gz&&UC8
zmB^A-<1Nz^i4Tg9CxM_lA#9TMPMclg)Z;RFt?gnShJJ#^>Ptu8qa)C@c@~A6NG+Nz
z)5}5F%8oIV9a{Qdi$EYwuShtaNu!;NNS<y%Qs7+rwg7VL&5o{#g1=86q>P{@m(4@p
zc|Sr7Sj@c;|L9n34vT+9O%c~J9dd##QMm=iW&n-fq0OqwyHHF^8-;to!I{)NnyOao
ze+h2Zz`A70G=rIE>GfU`m+ykkve>y)ReYrY9yM5t4IpnoZ^kX}>TvU$5IKI4qCp<J
zVxGSzDR8p9hqs^W@<BIm;rT^n1jfGox&F=>AD@-MQlUp4<74n7kcB8mz65T9)HH?=
zg}Xt&vSKLZyz5*%+Tm}-yP^6d=2gr+e+1%E3v0(MqSWX?!ax~N$pc^aNCsqZi*7(|
zh8YJPd;E$KIExaANVP1ZtFN~9!yCt`YWgtpoN~#4$%(01^oTtU6Xva3^qIu&t>{st
zso?9<WL^}-uFBhHlqxf%wJD<mFF-=GA7FA%VvK%R^xe*#@WJ`-rKhtH4)PNUu<_Uo
zlifY=yGWf;T-W2}(+r)qxru~FjLM^Ks!Lz$@0XE(7WpCoN-Y0_l3*n~w%s=1L(9Ho
zlRe1>ao@MQ3`;e@$jDwy52EbwByyR*PatUT{rO6x=v_|A0*0w;DF%T;_*tT`&&5^o
z0V?{F8&M0Zv<af~y3}v%d#Y2Htmn;MnrJ}7jUBX^c<IxymVkB!Y7i|Q-zNPsaZpL~
zsJXKH>E+S%*6J=$kNz2!I$xo}sa-p5QKw#b3<~KMy1hzf7S%#qF6_i%OMQu#hC5TE
zvCMK$w6r@&-{flN^W_)MY1A7pcWqfS#)1nFZz}5Xj~5%oS7(GvK`ny`PCcWjnyaL)
zIi$=--}B2l3!b7Pm9GtbX;!$YfzSUQRyCB%zH6~;p(QqQI`M$&zBT>P4fR^gmFB?$
zw&Zd!?Gaak+AA%Q97DBN#ePX**LRTROzq#n$n1Pcy8F=^{m>z%<zGyZ1}Tvt6<eje
zlflD&$Z>cSzPB2*YXE{&<=uwNie2>bZ<Fy$_U~0JN`n!N+lHjuq%p;LN}1Zw1BTj1
zLJ_<BtRfJnr^qkn#ue^3`Ex=W9OA|gRX7$ethD3gmIlB{;h!PZ%-VxbVr!S}Pq9;Y
z@tuBGuc)1gYD#9B-4DF53&%tzBKF0?mk^DYwK0WmW3%WX0nHQ8sG@KSUl!bNQV9wU
z$t{NutN1-mHUS6I7maa<*24LtzDXG$yzaC2+K2M(2I=>$-`;2Rd=F~z)Rg3TteKSg
z^hN_+au#8fn2Nx%1DeLYHs36F+p9*$3E4GF&#iVjp{|?9JF*|SO?E)lLZlEF3HrAH
zvc(}%O8pO0Z3=!io_pC3ny?#Undad;^IgrUYdtfpZ00k6H4=yonM=4I!(h$De6ly1
z=KuH?X>BN4FMtkZMpOS2-!^y4KD4yTq&2AT%$L7bY_~K0?~9DRb$7RuU3HGCM>CIi
z(A=>)>~yNASWP7;*sKOkDv>kUSa(InaAT&g&6r%F&;f{LjEB!a(?5y9zKf!FUBO)S
zSFH4YTc*tPhjGc_xCdF8#wYUqAdJ1${tTzv#1}qEiHbctRefd*9+sfIj{R{9I3KY4
z4Vy!B3Br8hsXhE!BRq!UB41nE_Bri7B<RI`-NT!Hf9m~aAWn!C=7dc%I&936FO53k
z<l#+JED6BEW8zD`h$q^(w6>~N%o9Hff+pUF!>j#GQ)PBbPs{8^1{=(4AC0c(hY{~d
zg^vnbvfy!cb<&w1pyh4S`ozKy-cTzuZCvv{BEd^6AzMy5KXI&(XtXv*j$ejsxD62V
z@SD4jMQ-c9F)gJRT9>Xx&R&y^iAM~Pfq3cF9Gx}4mIb{d`sGp|cd7tWp;gq>8g^nW
zCTl<01oPF;ZA4Dk=eTVqVPT#qff8MN3W3a;E<S$_jz(-L3l4Lnh2o}b?oi<aCn{{L
zT27(v!YHCs3ewwQ6l6>*NE}1aQPmb`MQJD{{(shkY~kTvPLYilfmZClnme=_9AUll
zv%eYOBftW?+_3TV#(2ZCV)NwlZe;B<P(9D@%EFF3jHSSx_cCoN<($u(pCd-1@q<Jm
z<(Rcw5Xh3*>bGjp^iE>0c9eC$OlB%$FA?LV)pvE`?RRyJ$7XR~qb+FOREQGjNVtE5
z%BmfWTN%kN98eN9_kzxEXtY&jY$wg2ie<QLQ~$b}doZQUZfC&){lp!&Xr^lm*D9H|
z`MR4oasD!0L)?c9oQu`HK^kL`>uv5)=TVfxi7-5CO?t4!Y<U%?=QF9(qNBoW17`EN
z42Y}Jrj~c#WMZz7Bko~+??RYG*gg6|)kcP1+a><XBxYEL#be{uXEfE{q#!unVDuO0
z+Wb3WcKt@&rAZ0GM52vA=|GIjscXe9JHsH$w@z*8d^EYz&_03-D5JeOt2|G3im3p?
z3B5uQ5O$bm2+97qqY*};RbrDY^tyHY+H6meR8~k)_;@P~$P=#f%b!2LG)bTR?vwvf
zdd_g1ed$&W^Lm*D|30LyKf>R+<jcb7ZQ^%~D_H^1`_1CvuFNN{#odXd&GgL1jyZZ5
z!*!s;_gF7rFA>nY|1+eT2g3yLeY?kEBY8{`XC^}A99l$PkX_xwEf_t^3Rmg;y7fKn
z_*agLoM;hGYt|^o_mJbE<_16i?wfj-TatZ^H=4?!5u(tt%@w81kQb#C*Wu>UNAap1
zcSdnMTM&2NQH#4i{MuWcM@p*w3M+g@@MLABN2CqZN=gPeC3)Q+YEt^M9q^173!R-!
z3tp#@cDtP;&DPVI<WHX7+)i7!m2b;0bi8*hn(6z<p#D~QC6;-)t>>HAM^lvCidiP9
z`joy|5Y|MQ#n8=cUTV7og4T2!dLJZ4L<#TpkPdJM14mU@g6cTql3ppPFhVxd$(Ffw
zM&6=ukM%mQ5VWBuyc?}Q43q%@e1o)<J`zc~N^dY$?cH#L^GR>YH!G&aqIe(7w@#@+
z9+P=t#O0M;B+sC&U9{ZQ>u!=3#BK>X>ipaLiSBV>aF{IJdACI=sD)N03a702QK}Dc
zANYjn?vJLL*g1%fs3u*a$I-eS=+&}x)y9weAyr=h3ei}@5KeEC*X&J-Mf`)KNdveh
zsYb)jgiY)2jlI(;1f)h5^<HNsdJ{UA{3w8vr-1h9qB5j1XidfRz$siE_l7F`NuX>?
z2%TzMs@Cloa4Y{smdV-5WU@}|3yv7&1FgH`Dl(SuVEc(|x7QDGJ7iaG>xDXfbek%c
zH9ZDfTT{3;@=&|Xj0HDd?INKyyX)AT4bvAY_#sk>hw+bp=P$g&>Uh{;d3_kM_Um^{
z7X&nAFEqj^8Z2dgM-o#v++P^)HD0Vs-{A3sqCyq+U5{SQ|J2&q(0D|^=1ba^-8gJi
zZp8BT`?NR7%$ub@dvH#-l9|4Wlgzl(t}J^*M^NO5;v2(J=%XVD3!H{outyXJg@Y8;
z#p2B#TOk=gM=TV5D5fPqvR*XqQ=UhQ4)rn<E@)-)-u46W;-0vtIUOCQ3_snw)x1!c
z03<7Na9c9x<ociSWYdwP5PEL9?Tw?sJSNVssCFXniu~wyZ=W*V8j_%TKNm)n8@?l_
z$Z|l;SN}N6c)TL(W0?S&3nr<^{uO2YS6Z$!lr9F<!rm>eE;DJ>Uc1z$a~mY)?`YIm
z6M4yw5fW7LvjL8AIVqSY?RJ=S7VlZ;+rVj@3EA7N>kf0TQ!+g&X&aztdFway?8jif
z`cMG9qzk!(gOT+jak(#ec2!xtPue$huVbG#iGJY`PFH%?w`AN~hX5l7nMl$VkbT%!
z?lq_8(f1S6VppYXPpwUlC0Vejd>ngy50Mn=hULE_C|cxB{A3(;IO%#3Pd__;slSRk
zzlOlPRd_ksJQQ$ogbcsy3W5WqxiY=n_FbYDUL(c*NK*RlRX$BpvOPtEf{YN3`(s7$
zJm8=Vq-|J9j|j{$Aj^bHzu4^Ifzm+PsAQ&yn>XFPU#g;#TWK0iLL?punwa%%R2LKu
z+6@}lLk!hHt$B-@krvg8Syko5hO|WT;yh6jg<bo$>4|g6^tazB?z$-z+(MR9xnlo!
zxi4>T87m>IUG+Vv2^yJxDT?1h>k`}#fqU5}Z5L;>>j%Ewf!EBDH|$R**(+R2i@)LR
zIqZuj-I{)x6YnrAi{=*J4!Yz#xi#uu#W0uHJf7U0x)(>S^GJwqt$!&ws314o#Ewtl
z`PROeS?`+eWHFP;?OV3S1$pA^z7)+(@orx3Wpa+jP}1`%Q!H<&+^9mels+R=b9ea%
zRz;h%Gcoq&c-`Il`g-w!ivN@3nyv-aP7$U{zf>%fk#Wvus#YB^1cO&Qorh?scBbJj
zo2fZIi|=~tewBXR;nG{@_<X1YsAAqormW^!NjI2+3nEN&ctW&Wi?pWnk_9JH{Sv^L
zqUF!Y#g^ptUSbaDCEnCi93@@$rkY8qR3pK?%m;lO36sX~%!dDshTnIw?|R&Ln(^pk
zZ7f21dHP#xGWn5vVk-lFSYg3JX`N(p={VaO-^&}DU-5NZc!Efdk(z;cs1t8_%J9B)
zcoV7nUeY<*T1DphAOxNv?K>$HA%GB`HwOi9;m>>Y83E=A+YI8#{Nd=N)tBjXh6+FA
z$c{Z0%AY{vai3#su6g28@aJn=^L7~2e*j0>Z~Z{W_vMsy9`s`X#dK14-L-(Ne<F9{
zH)Q04lIVH+IG|#@8PToriJ#rC3KzSzuWdhN9L*GYNiOs_Eo_sOlH@oR-#~NkyAo=l
zS9<0f0{KW$)WomK##J^)MC0cP%r&L@xpcZ2@UTh?`w_&l&BJ$Onj@4wCm2m^$Vgzs
z0u=s*(r<R_Kh(myu$@>cVo}H;Jr2RF^#<|(M>8LZn>8>Sm*LXH{dHA(-;CL%7qeff
zOMWS@-n*1Cg3vOPFmg5jmB19HpiMNh9g&}I`j{XJ%`n(LgPUgkm2&(1Dt{1V8Bx(U
z+6$zgB+3N>s=saJ6ZHWLpjAsQcEB&Nka2PrqG?^`R>R)VXt*xrVw-fHiTdreCqG%2
z_^GTt6*lpH^aPHAX6a>KEo?vg_r~_YZl3c(6>DEC&JyG7CB)uUHpaMTJ}?u{d8a00
zyh}vr9*wY?k8$U@5@<$)AGvEi3awAme?bm1w|{qOzTBKZ8f<XpBgoD93eWqX=ho0g
zt<m8|vz{Ce`!!c%AXPjQi!Gn~ivLm5!D~I}p<;ET6S30o5$Z<3m^BP?o)R9^Wpm0(
z@CZc4`zS=Apzq>f_TvolPrUJG)xz(J>o6<@Fns2>a{~;e&I?jVH%uo<tGw+wB8cKm
z!)Ytyp5?!YSMmu9^jOgcUY}l7A%$zlkUmk+*Q4h0ITCan#@OKTYB3ABSCMOoZ5jMY
zU?uXC8H;MXJobD|WHJrAy`P}^{tg7fciv~L|7yBfH?sriYmcN0HZTf%$}+J4Z62Jr
zx~W}yui`hoyP1ng$|Ob2qrS^6E?zTju|m8XB){d|L0f!_Z7N@QtGI}alGWq5y)Byp
z$RtdXZRp-DJNOX7wfn}BqiM!X%gSu`J-OUU^Q(5?_@@bZ<nn2|)NWE&ra`b@hO#kh
z>d@NM=PlckHbX`~k1z%B^z)|F5O7@XD1=DLFt7Et<NGbCC1a#d?_&J!@S&2@IZRmX
zsaJVyqq>&*mGN<i?cE}CQaqUFsPn7RwyS4?$DGFW{*Go!RA*kCMtk-h|8=-n<FBVo
z*N+~5|2*JDEY?d3#S%OXa=vLBuLfYxi{=kY>#cohap2r3a``lYRzp(7MocBUZ}*Vd
z6K6+RzT|`p<KZfxk2TcI@PWa`o6q{5WF%$FRl7<i>~2_i;d4aje*|6iCg^ZzpWgsB
z9PTg;Gq~ZEc!8!XZb`_xy{p;hflqnc$nG?5#EQUErt=;gHCFCb)4cke#$P`qXHBey
zB4+p<nvxfe+xZRAG?lz1Y|{zz^A;8?U+&z^5a`{-S&pb)=7{RTHPX#(c%d;kIDVIi
zk$!LDe+hK{P`>gloelr?{SuHx-HFrHym|q5;QZevE__Y8mnYQfKHAM3V3K8=R~9#P
zyo`e2lXhwOMmKa5xssdtP@<LXFtX@{dHj`rsOMSQO`*2pG51h#@Qh7N5)Jv#mEd^k
z4d}?yis?o@Z?Dp^uj^<0U}$j^^R*wdKWX8^MbNYES(^#dpnP<n>a<U0N~zpzl)N=h
zE7Ng3ow2gSlE_k3gI1R}z8qaR`$#P({eyzlv=6Ny3h)oFxxxXw>#Stq%Z)v`BjO|X
zaDsJOC(;_^-~GxFG4M}V69#pZ8FF}ZmQ2~*C(7&WkfNg2479)SyA=D*iMT^uBha8L
zlq5&mqj|=eYw32&g+YD`43CP~-nn_niKWL^L^nN&bzd^x*H@NqwDP^9%x;aY--v`p
z*k}_ykbQUC67UbQ;CG94eqO4#?4S{?eqy>fDX>;D(L29qV<8Z6)<oH3?IqzVc_jKr
z>zg4KU*S~C^+{&UCs(y0Ern~Ykvv*z54mD6ORjH`kLKx^0|@@Og&r<k=+Ll<DoyA*
zsh!>prI1t>>X8-*E*ur7dcA{8bRFi>RDw-`{>w@S0D>_zVJ~E-$OJ$_-(2yJFuU|h
zR4c8SM9X}G)qm|3Ec;GjOAg$c0ndV)(HqK#K<2(cEOgzC$?P}bG#Xz<H&!lCy{nOn
zU6;`pwyeZnk8>W4+cbug;~EG!`acvyo=X=BQjE1tCV9X=q#Thrad>0cymcpxEIny>
zIhvVb&*!#-ew_KeV)et-(d%`Uq?niFfcEpN?cBsO*LwzUk6D<{(D4h2X^Cdc{56lU
z*g>NmPsWOd>UZ#pjkJ6&b&Mw<rniOMBQfz6zVJNICdu`?44VkJlDSAeWZ);NSsiMw
zyke5MLE}O))FxC`ueCh%Z_JM5Ihy!4uX*1HYQGwtmcVYWLW(3a+n1uQ1dw-luT)GX
zdak83_GnAwP`PbiA$lxRX>S=U>#m$vO~?+EQNCf$HF{2)7Nug4qP(al{d4=1*y1^I
zH4|0@&>B@{x6p({CouoifuJ;qIoY-+ms4<gHC9f~#!ll7NbbaK$lW@z4`(`8cvlZ2
zqbc@!(jheX&vI?vVhaC|tf0wjHCyfTFK!DHzNrG<j-Qgt03!FUD-VpU(GIKpPDOOg
zlquF@=T^d~D#K-}-i@ZimuR=O<I;k4#mgqg2>AJIA<u13vfaCkLBWqToB|E@IoO(M
zMV%jfj##(Yjn5*T-jWoNj1J5`wKoGD)<H=}JLBZ6%<5w!ZM>!=^<Em;&vEB!Bk!qF
zk=UoJKaWG{p+D`@0S(*~-gs@`=dsUB39CT>SkoLFvqq1!tg!R>#4NsX`sr4(TxTw}
z_@fW*RdJ!RN*3h1)Kz*sI2@c#AH9q~L)DA)T68TYioyx(d<!!G)rsAZI}d81J3L7_
zBRhWOL<S*O>EkBgK6(=4*qYQc`iaC7-;<n@_JF#7x_3|J7ogf&l|OHA_>(#V4~lGb
zQMX8C%f1<(OQ1QyocJQ+=4}E0aaoN%5WM$#^Oe9WFUNuql_bW;5WZnPi&m?s<Ml$`
zQ)^`kxS#cGWhK$tUXU2uKN+xijhEwhtjUa}tqqm>I*-MwfUhkMa5l*MgYz|4NhA2`
zsQIPjah5li6c-Dl+OW|5$PC-Hjr!vj@fs9<k(^|{H+R*Yy7xU=0^~_^FI$C5ukndn
z%nj#?mNJM*E0x0VNSo2hkNdBRcP;LJ1LU~vqMSxF)2-TX!6|iHP;c5RHSUvA<V^jS
z<kWV*x~7o}Fe)87+4}qnB_PEvFeS_2{EAt+5vvio%6Oh6IsF16RU$!PHYN7_vZqiK
zjd31*O>C*eQe@3q7FA*9Mvg$!o|OCTs+g7R+Ok!?G8;)r><fCL*OiTpX-ofgl`eI*
z`IJ(ajR1)l;~p*X@q|a*{n71Pz+J%NebdPT`KpdQT@h%dc;2QF9C4&P-=eeox^MsG
zu&2xv#ZKP7U8=-5TmOUFe^GWoeb)B*B>B?39hO<7^J6D7BaCK4(W_3;`0k+`h}41v
z`v;KjIU`7pbpHt#9H#lZQ>}GI_pO+DQ{U)AUrvM#6~i&8?Nac*89o(dyC$lbo9<^u
z9-m(Zd9NyB#w|;(r;K?6si4uAB)xYYe%Q7KRBY80Wa}s2a~rhN<<rmTsJuhQzw1gY
zt8ex`P$np4U~$JncDDVkr_6TR-it9wj2U0*!M_lav&fUfdn>7Z;<SV-7MWkrSbb8V
zhe=&vTH-R3(v6$f!RBtM`qVWO`Fu~i)>7D#$pfnO);v+=WtE%vXV(31n{$eo@MYnM
z4me_-dLg`;**Y*EtKR6E?|pY;tJ`w$tK&Ui%|b!N%yAJ1x7T|tuu%^h_H=*b6+qBS
zG53M^yqjvugWwT<_`xI8mw8Tk-h3LdGNtgE{lVndPR%0UQB>kwwAjq4$g&nDc@8VQ
zJmj-u`;(aubNFiRb*=)L^?uy}y0VjQ4e#NhqbDXSRyQ^B8H48ObKk$>_3wk?D6<x|
z4%ZPKuU|{N)nkQxS2s2^bn-u%_GHR%)#VZrD6iCP@kg(gy(jLEnFqH_8YC~yUBG}g
zYOGS;-k({U>H1e_?nq^t3V*&~Cu)>DvhwAuz5yn3wPP^J7uV<L;Y%B{Fe^{ZYK2t2
zhk4MGai-K9PltGH;u=0BdQFxa*(miUZPm@^HtS_uQZo8JjC$2A8?H`(n{$<~fszgD
zjs%|y{TWYwb80s-+Tz7Jr|L`{^WYvJVbokFcnyDups_D_sCyIVE8!t+SEE43>WE=9
zQ<=t#=#V>|&PPX;PW@6SrQ7Gb*Q)c?%olj-3#+{KOuVW{(|9=<RusXjls`<u7bC~k
z#%8P`QZh<-m&cy@Ccc&!lQOSrXpj$7x=CSMrH8nkt_TvG1A&w()BbUa)6;4JMR8%B
z3d2TedGuTBfDW#!wKC7DC66{=K)LK^6NYAL{3>oif~#A4gg<Wsp5HZUna|SmJw^ei
zVGOU%+d=TWt=%KISMG&*cMO<c(4-rr<zj8Q+H*ACW?aGO6NIEcpBM4(_ptMFqqFEc
z*$3C!Z=dxYhOV(b`^9^-5I}Bq9E-hz7J?uOhQztgzpEcXIrYi<Igo6@cP`M-djy-I
zUIy;1z3dar@`-!`6WlaJ<yW(Th=FcUt&k`8cq;0jhsl^E&;8K45uZ3k%f$IVDb(qF
zw7Na^rSCI=zF*9E`e=zS0HKO8=8{qIpm<INWsOK)`S#tpFVI)Dosn6gFmI0E+H~jB
z;3&cP4SE08B5zLzEY95C_({NpohRRJA{_Dl8^T{kRzCOam=TXs;FHQDGAjWYYjrKl
zVeTYav9av+8S-`V^8UOVnnH709@b&f0?z)@eXzA>eEaIqQ#lw{-9$xL-^$7nnQyUk
zbkCgfa)cCt6>BLohU+cM=u;m_;A2$1{IP#ja4F3*@NU9<V^YO()r#aw2K(26wSjcu
zhnT|<Na`;l&Sy%^HNU+q$|7Oj*BEBcEqph6xZ#=Zxe}JZt!6H2=U`HQ=24xW3U|Vu
zy60264?E#q6Vo;%r{7+^EJ0lye>tG@>{;(KgO9LS+3xo559W^FR$IdFM~=uG{gSxn
z#bar-v{Zv3`DTMv>cy)she7cD^LqhOy6=DZfjq2{61*(a;^6)(T#>ZrqFv67Wl7hG
zo5q#GItaNVK@m*c3bxkyBVRzB@fzbq=BnJcdz&C<COhSf8)df8>02mDE^8|~BCRDk
z$OEaUlS8qu?q0(UUbq-=P1zkP8#~F>YfQ=^j(U}NA_kdqj)79I^L_Wpu@u@Di04|i
z_-TelqM<NS?Xe9wUH|KogZnA<A?ee5n86vMN6%Zl^1s@W*J2ZD-t+yh)2GMMhrPV5
z&IF%|f7$==4OBVUr9v!Zgv`_!zGW~oy~R|Mu4NVJ6Z83UQtu09f{K33a^^{0l{Kue
z0>eJIYZK(wY=2ke0X*Hev~QkTNQt+L0`yups0ePc-%~Es+cxv&((108zQ(T}$ThFv
zif<N_^+B7GzIranp4`|oS>I{QCB9SN^&`{x14o|=hI#yryJjEx%XUh&FnbkIX}7sh
z_v*|1G*1!!wW^fYth9^w1i!T&ED=nL7=KHmR++ebgz5+c<vDAhh{FkrnsmOka|vpq
zW2h%oxm8!f7@qaj&u7e{4Y>t!j6UGBHa|E}_A5O(%`@2>dtYq~d8|&3njK8AK|P9L
zEV^6QxJ(yEoXq}&cQ&nDY7Bm^xshvsoA}^#=P~wRB8Ep$!z*C3h>x7#YSEw=T`?p3
z2Qo5o-^n23TU^CiiB^`_oEr}%O^jp3=wfpRgEQ}}7`p5mdBJ>oSD@C}hq!`8FZ)HM
zoY7gO`#}q)Ro|H&gFu*qO3Yc;fVkD@eZ`vtiMw_)6%?7LOIV#!QHzJ@YBZZOzJXq+
z#}m%Jf)AQYdk%JWcK_vP)^pJLb}jGm-`o4&pYNv<hH`oR{Z8QeJ=-)6Juwm#$KJG&
zyE8XgD*njM5?}QGW;OF#>d3AE*<mHkywvb#vxcI$crApiV!+ol+eSaiiLl3MCRZrd
z6`u?Cu(`fkjSp<k5cYbngtlv`FqmDgdicPce+z0z9(U;ajI@)_a2xiJ#x_VldIzD*
zCFPJ8b}QcV8p3!AC#recINnzko2M7dLG}Q$xnLnS<L<RfGjGmDP3oS(P%6(mdLmw~
zj87}{e)Wk23ODkR;-c%5!c5UcTi*Y@((k`r{>CiO{_>C>+B5&MwFO*H5XF%$b$yv%
z1n$4xuqQ{sDDpezHco#j?QEFd1^MWoy{E0hj%Hf#Z?TN@j6EAvQx{8gPgHe=(T(+%
z1HP@v1x66Ohf*_rbCm=kmJ7>ca?zpN(G|0zuKAf_-1x>rdD#aoh>HPjS>qL9(A@W8
zgptLFlZWW}POpO-LI(f;?SU^&>_3Pycs$kqLVY;VVt$T*emdTJ`&+poe^1t|c}1FM
z@O48Y7|eB9tXAk`d~QVX>BzDNzs_o@V-?4sRv|x|-Tpp1I>Zj<-f@$eFA7eSx%a06
z_zJNW^rBT|NG1-Y&oYZ#oahXFFSVpsCSSv6-gOtgW&`GHSOCoTYVXRU1FntV%ho|W
z#Z^#ny=7yuyT0G5y<&Khrr5R0z5}|YBE2P&FXRTxwiezUffer`xSLxEztr~qa<+@b
zT({=r+-2qKsE>H~MR=WcBvZej6ElnX?krHt-K(d%B&n^W|89AO;A=hLM-t|ONacql
znkux~u`CkpcT8>(xy|rNan0e5VL)?Tm(@^WSuBj2NL0fl#>>e;`$gCnwVwZDzqkDY
z7C5P8+T=FEG+L`fy!Eu}!-!oy+18Fub+XxmajZ)|dNTXGE4uQ^9NvgligVjJ`$XbR
z49VJ<)W@&hyg1x$?c9E8tfW2r4W>}^^nP|QjCa-oe%opuWufSv{qgQp-tpJrH0L90
za!v`#?)oLX5K9Dfu_5w6Llv@s&x`w=UsBz8>f1{Dp=8Z;>f1N9>gUv1H@-?=D0pQz
z$Y;O2ZKkcibK-RQ+5ju#Q{UJa@zSQG>IXdrEa@PFk7NDB9gS$r=R4qglK+m&!iOmK
z=XRY4^Wm8^^z)1}U&*QD4$t0y*S&=;pIgaWBYCgES-_CIBijf?${#h)(8`(_PeIMh
zW3}%S6S}{;1W;vHrAr5v#{=tw*22Hm`LS#2?&P+Et9Dt1sw=-{^(n4IVYIz((SUra
zlX?~lI{i*E5Gf}HeshM7_Oe1az$}Qg4krha3+lgH1K+ogHTU1`u&6xn8nCcPinUX|
zt)T+JP^|D_7ss@R4nMFS+Ohq}&-9jTE!)%<{EttO3GY#E5kUr7r*ZZ=$<hJy*O0aS
zja4B-@_70x$+sW3w3r^<m@9DrBGvVaurga#J^9Ntz~pn?mo#Qej?Yk0XRUw1S4nT5
z(>BDDg9XJz^@CrfGe|#cj5GlEa)obfAN}5~k$kPO<lZ%Z)Mt8cX87cfqB?<Y>-JfL
zkMG^rf$>`2QB9^B&x5Iv-s6_*66(6Iy4DOAp+5Jegpcn`V7JTgud~hOwv@yKEI-=m
z1I%zH6E!nBeCH730^?{MO?sa3+;HhPo%yZR)V&mb_;WWThKJkNe#AAuiml#{)DS~C
zHBO|ybub-Sc~hVgKGS(OK1|`{AYZ)-ZT>f<J;c~o9C*#_kh`AAJENeSc{5$L@W!e5
zeKI$QXbdfN<44fzZ*_ThApJq@(N@V`xy^d==k;Ex2Yx*sp-CUQ;YEr)AqfA1Bs?7+
z&xx?tp;NO?Cg%Ko!Fe@gJT>fl5_>#CS1bB%^n^f}m9=pE8698V+8K+=87IAbyPx|Y
zS=p-4QXiRVoN>=hKl3TnaShl3N^MJ~dLUh$WpI!vb1JB3JfaJK4E@?_SL)r$(vnlM
zr4@E7nm&@3jLiA>Q#Ldh*LV8!Xz<5@tmkRGfnJ3tS{(CM>lx*N9T%-aSthYFKZ;sm
z1`jmpDAd$KtES%Du~e752gKAAkBmF#XMf2k=*6)^bwbAHud|J0qr$YO*ftbpG-a{%
z{TOcLYx*~1twylpzySPQ>j-y4u~9%c+5KI)oc^Z^ajy>Etn<n#Pm*q?LL!8>vfi8r
zNpKv2ZXZq13^(ZQ{B+*PkKijcw&9l2Kj02hmPI~Gzxq`PIxv6qi^I~@MLyz%tTWj3
z@B(pd8Q~kt=f_Rmzf^oO^~MkGP56;p$?aPxDHMx}itU+Kko|<|ona3;ySUble6M+5
z2tUQFT_-)F@{(|LwbUEJEPx0$n*T3|)V^ALvCFSp;)~s8O5X{k(Ws=eJ+<3S3j{__
zc++dp?b5!2Rz1>tk`8S5`Oy9s$%#zPUD$BX<o9-~<Q3N8ZRh)0lJ*DQZ9eSD-SJ0b
zBn{0X?BZAbWPOZfJP!23;PYOm!}WE9g0yNQp=B~P5`q@$SAznfxyQF>du*9`r)h^B
zmp^kVnE5S2EKN9li(hV2m-LwTzM2RI#W}&3=Vddca)z2IVVfrevFv(IF;QsVU1_|T
zqICR{t0B{-40SNv(;+YaHL_IeS3$eRadcqd^)^u<E7yZxtt`>HlKX+|*x!>#9!*{n
za9o2Q&bDs-oWz{~bzo+1_+0V#i&-8yM9I9dzXs{on<1GHm5Y9LuG;O;jMy&;y9C3)
zX01qOp5&26?X2;|8;014piG_H6~V3=b|LH;`HM5$;9J$mCjFrs#6n@>t6@Kq`)V+a
zg@roX&bSDOuJuK#sxp-7z%QOUhnP04*MJhCYank&Xcd}ZEgc9hj2rG4mTtP2*f-PO
z!jqcckhJ*0=M5nn4`u%Wi$U7w8r`H%{DA*LZXjn?Iv*K%OG+D?ON3>oG4mK*S+nRK
zr+$9Lvil)5c;m?DMnxfF`{Q2!Kf>NJD$2DD8)k1CQMQCMD5-=<_n@L6E!{InN_P&!
zsG!nFOK+vSq#00BT5=>uI)?!U7+_%LyTRvqANSVx{rur_EnRrWc^~I_#C2U}ok?1P
z{y9x6xgZ@*+eCWk=C?3fByNhbHybz6!QZ&UBv)~Uge;NR*S&PLzfzaH@@338uwc`I
zC*CIim9eCMXL4Augf;tYVy;{EO9mB%pk1Sh*&;8eAo8lmzCY3FAD1~C>E^fFyy-Md
za})vN9xGN!?s>1;Ta8l7e}00250eG4)C*ctj&e1?Y)b!HP}H77B<BpA$0Pcv%8{ae
zYY*1Z?rjFU-_$iO!PaQ)+^6B7tQ+qoNKjs591L%pPqSoV=-iqUKfVz%OI!$t*Kfja
z6$7ckuVqdpDK_V)Q5PD$=0D|%ZdfbfG-}9;w$ncAvvhjfx|J9VpHzu>e`{<TYgl{5
z)TK6-Vid(w?IY)p@A<J3lZ1X=&CcuqSr&1qZxS=+-{4|;K58HAS(wDth#J{68F+?&
zAkG0N2iZLsYP+ZkjHPO$VySl<J8q{hYd_*0FZtt4=aJ=j+_LNuX{UrJXEYW&{%%|-
zpo5b%W<)$fme`)2UY7~1ReGL%p{df9b#n?+l|=B{xLB!PT6}j4#tXBJ-FM6+4?gn6
zYJbgHo>nzCsSn<O*EHU)l9E|zWN+-Qyt(xYqV(96=2YCgWPWXGHRvPFT8Ev`ib-C8
z)gV-dG+wwt{){5#0bh=at)G@6gX9|bYD_g~1mPu5aD(@R<*2@sE-7ypDzg?;Z+n|)
zjX81;T&g8M{{p@h92yF+qURjs>BN<`7PnqywwP&$+q=`5o9zZ|umBgJ%Y<aPAI;S~
z?R(08f_+K!_M7`$irkL&lh>UP74fM*+iE}R<lYq@#@I0m%ue94c(V-E#SSg+jTgbI
zv)pqDXm+Wz3;|p&sO!>t$QWNXPKM_?+a=7(XI{oQn5f1pu!d(d{=4j+QKGc9e0YYC
z`oOy`bEZi;0R3b)LEG~Sad*iwh}dvh$Ue)JzVjsC#E7YJOXI3U<u`^eCJ}Wjc)A$a
zYQFnXo^iaZ39hhji3}PKXW?E6D2mvh8_y1kvGMu*lLGkTf~{WtDC`!RBY(f23!B@k
z+9FxoMK#~SB1Ce@#=H30N57ks*Mh0CDH;^$MB@8`O<kKT@w%y-#e#1f%Rg$y`6kMr
z5r~J$=ukDWrM@k`rVfS<l`V%&vGjjE;C1oOe$eW&eV?ig;y1^C-Je&AhvWvT48sdA
zrrSqd^uv8gwR3AD|5CR5JPrz@nk{P-P2G#$8WxFv4|2s27Zyv$ayYkWlQV9lX$PHf
z7pKWatr<is0{1oHpHNt|b^qtRolb9aH6R2d#CvHDA?r9qqc%kS;r-ihb{{A*R=6>u
zZ&o7VgZT}~@8nkZaXLA?Ms0`?ZcIM)H<5e?tB5>`&>j%EKjg47kutLRV5Ke`K35mc
zBIDw+g=$mW?uly>-<4YFBY6&GKH58(1oc3|NMGCU9}02PY~8-+)<WjWKI@L)%Cuz?
zf~-di8rMqyg44Fm9`;T^<9;2r-dS9UV4iin^`WTvas5^*;z<#qNY3*JS}R5xN*`BZ
z&RXUEJyKYqWU!t)?)D&s#ItgkJr|;TKK(WqVy#z;d&R`c9=(+kM!FZWWDU_m;RvBt
z--*DEJN(7V0G?Ql$Spy@AZetrL-i5BvUM88642%?g&ig4qd5C#CKO^yl$F+FTmR|q
zYQGW~Q|>2QOkV7HL0-PiD<S}mz1$Jw&G}Mo_d(HKNa}V{@?*&j7P;I03K{Bcc)Nv}
z^K9j-I?Vp2&x#T+VU*a@Dc*`j=7pp`G7n41aXPj^A(b{N=ET*$L*e-OtOfMl5MquP
z!EA;&2ymKgpgsyyY(x+xeU$;8gf<V{xN!^jN>zR%vU1fcwV*L!x+0{nx~vV-f{ZMD
zxi3~-KO1Rh1ReR!bG_cTHFCM7Ne%MW^zDDR&3_p=D67FC{J^z7))?C2g`@cXDC1#t
ziDNx<Zd7>2#O5gMSjdNYZD%0j(fj_d$nyoE4DaO1inIsohoudI=R(Yv+qHaIT!Fc%
zon>9<t60b^ZdM$$+uosOGD=HLl06!LG5`(iAZU_b;GQ3O=nA-0cj1EsVdc|Mm41PE
zV3t*R$om1jMVr|2eTi-YCOG~^gO%V6)3W#ST|g!bSuJ<_U9BemkK&|#<&ydF?f&og
zPfXf=3fb&oI6_U)J@FUG>HM0HD1ax_z|%SQ@xy8yC1e5dry#X#H+us>g_2EXm$?uH
zUsZi+0?NXa95Drf`C0}dIS2q4z0>fAluckizK9s>62GHik+!HmO!KApQx`JQkm5gj
zyvs=CYW-Cs9aNECNczisjHJ%GL#fX0)S(1ho|B1UbO-}Ob=!DRs&Da6s}uiov1jP*
zS>=Phd|4;ggFYsZRp+y>$mpzPelt$&`6lNh9AWI*dm~j}JD=9J?Za2=`Ta8}UZKd=
zO<bL}i({r&|Ip`>+hH11@KNva18WYRx=7Y2g+=^y=Z#I2@46m;s*to4iAva;zLJz=
zy9-WCw&!dWF&Azbx2iVvN7P4N$(l=$1`TP4Cc3p7rYBc>woYJ*gLkKXl_*KF-<{y1
zbG-chY5D)cBBx<a^s7|mDh+OMu>i!8CZJ={vEua-X*K%tz}qj|wAOLBa=U?*7##y6
z%ZNH1A3Y{W{_&^jcR@G#XH2}9vksJ6*<{`{<6ev7I<#a!k|Vnui$QZ1Y+<wf;Dc|?
zmj!wLe6HxVUf-Q=cb*{jCrajnZ!cpxQbAy-stbRa4JM39ULCd`Mtsg}6CBQY2_=;6
z<ta*&qN}Ced&3=+-F*i>Z&3SR?LQJsZ$T4(h&k976*)O8TR+>>r*FdQuM_%{^u85W
zKlzC=Jf15s1}nAO)yYCn$LXPTG;LQB1^L^MqoD~Bht^qqqBf0ue4RgL=MJLyC8*Xs
zwlKRGyLn~|HtJjY08-wtdPKVQM&rnnv5lpEdrJ#B5B=lQy!WP5x560|ta4;$$&JPV
ze^aUwEfml>E5BGK(jm;zq^`#MlqwX5bPVl9J9_F-pcMW3=|Wq7k&9!BKVJ*Xub~Nc
zF5npi`<3RXv3iXFEgvCnyP$}vyM(FB50d3MpXWpMXa7V_N{N(@@{?d$JNq$?gfVOL
zr2wInonon;zdvAgZ~CL$mXcNBy8s@7d22OEa#P>%H~4LB8oPN|4Ax&w)s{Dc+2^_)
zt&au1t+`K+P2)J}_V%HuV}g~{uV6?r)HiogPB}W{Sq@hUF#9<`#KCiFf4*^VM*GWU
zfvCB{Ci1tZ)kXX@w$c&{7jLog$A37x=Dx4Y*14XzD!C!@)qC^BAmbTVbGoR_JFvc;
zwjor`OJwRPAabSqIdG7*EO$Hk`OWxXD@dNpV3+jzJ|+{8fJ#}5w?jio!f|Q&adqh$
zK?>H)YwUAnF66lrOV9s{iT#b_Rg$yqy=a8(uX<_^_o)cK^WtUT4-r05E?K2ZmBY<Y
zxk`~^JT{TPI`p$S+fd<5?FoewwxCe72zX+zWktqtqRR74SQ}r$Xrj;Wsf&EB!_;0H
z+-?_jwLjGTAe6R!kRVX4W(Lo0)Te0kxZUeks?Ny8@RUjhwlJ4-$1aS7#2owGD?$BZ
z5`pK%67xVUVFjn*O;flE&ah9n33;R+V$Hjy+-|;`a(hUDl{pkYiGDO2xn^WmPpK4s
z^Z20rVe-$TElUx=@RwqKnHuj=P@|qc{K0>?=<6?SCc64NH0Aovebn(`(SFp{Eve@t
zCh&$WP;0m4ufWe;<=HsQbe+7mPd{3V9U)uX)>U3SAwFC8c`cG?yH&}qmWJ34)<O;|
z{6PXV`;Bdle!CQ6OiNJ=j<d+_9HvfJsTu%5-y>>WYL;#Zt&G!7Wh8D5=sy5!8!z~E
zS|r(cR!V;N_J;f^8q$!58NQW>j)O+itB7~fcgV+i#bsJhYA?n*6vP;33g*u4_f6f_
z-m`bU8ndLkc-9qFr?9)96Ja=5KlrU4&TFl(*T%9x-Dx!hfvt}u^o^OECHVqfhVlRe
zNJQ_109}wY=)F?BZ&*aBoCgx=?%`)@vXCCqP98gy&=MkDt1BzxoA$R3{|y|TKb3&2
zEve>hGXJXpMLTj241e-FB|i(E0^atSp%2NO>95Q%ZNGBnsJJ(nLcG)Ru(@9CWm(^0
zx2}zt$pT_^FMyol?Xr+NCm{Z~9$DhW7y6#AtUW~Cc^9$p3CEqHP_CZBE~P<BoMK>s
zc!SM_N$xrlj=B<chw%AJwGLx;iBxD1J^4ae`JE{c=~mbZLa_+ZE>k)xU3DU^e49-O
z>ew8wl6t>SEOrXHOYisi-I3p5I=}Kqy2D>!(dGXFi>zA+=fc+*(&_~O*lAlV%K971
zVi%o-50`r~zH$}#)~-l&?M10QPEht|NkqJNODXtP@%ik-ImH!;dQ-EsNr6=t!Vfr1
zZvsjtK}7!C^Fe0mCO_8A$x}j<xRy8o8j9t;!`miw>fdk*8#L5`_RPjq-BGIEC{;gx
zDVAW2tHWcze^XH)-w_)^78@T}c2P`<72Wnadf+k@$5-X7KgBOFjNXKO8V>lw-ArRe
zE7m#fs{iN7;W0PbySCSSu!)mP4Vhk)7JK=tZrOiFVrsFy+3=MuJsN)7FjA5%>7#%I
zerC{c1AFR@WB0p~UP>EaBpjV%VRfO@79;>1K0MrEYp;H-?KkGkJ#YIz?1QW1Zqi3>
zjyF0V9b4ZaTTB+`h=~tpmuC1fDv;`J68oh@5;6wsV>MYYaWolJcrSZ>?41hQz)C_T
zU<@n>z<7sdzB)u&Cq;-9uxeg@k~)~T<9SLe4}s6kU53!1Y)N`v9$e<>uS`DmY5KeH
zg`O-e7#Ysk5XjB6y2&%)>~xVMZt`XQe^947nTJAHnR{jMrUh)%5EO-rU^~OD{|!w4
z`4eZ@5B4)lbL8m%-!>1>`r^{O7YCZbqLz!@XtT|4kC!^QKe|ZkdGo1X6k{96EN7qo
z0HRQo=8Ue_IZTxH>e{(YSv@R$0XMA?&ZS|`FTBUwdw@}icrcOTqg~KY(72d|)M1S$
z#kabYUwgepQ{rbfb1q!xk9)=DC?kTg&}P%M{f%oQc0+FkF(O?ss%bG>cbR@i*>h!N
zyHR1Cs}dO#IUn@h4~G|o?xjz4ZD2|qg4OsnS>!&1?<{pwI0&YrEzCdyGU+XD^>iT7
zeCUZeS;6)G-}n2R{=}SoGz-l;m;~HORVQ&&t1m_5i?eAi90<LsM*AK>6vain>>JOQ
zv_3zP%el=XYx*gVGrkam6ge<?52KZD3Z9!!<VhFzQ1vh%`HK0B?FJk`6PxOzoGx&3
z`%uoy3)!(YB|>k$)`;=QL>5Wd#_~F5Mvhr*ZQ*XSr~}03q#Y??#>=Ai9qmECik+r{
zBI1s;IW-Lb>Gsalqy8o+jbz>e8$;GI1u=t0-Z6{Il)(heiDYJb&LL}O0(T!BdTt1*
zAsSD}e3YuceHbPgF^JU%X;9DNV3@oUI6N9@Ti{MOYg#9Bbexbt4#1&X%=D`nJ2wvQ
zM%{@J$iZ##_}B=DVA^=PwU=V6`Ed$edL>Zq^(a)2`mbsbiKcYbCG&(#*2zn+hQb4p
zR|Dq^y1<H8wB<zrj3ZO1Sglr*8rhB1Pu?C7t@ZL*qZfJ811i`V$Q$IbGhRwu00C+}
zkP*esx5bsx$U_>pgI_I)WSqHczQ_*{9<k#RCff_UwOqwB9#z^b3|2D!OcjmRa%)PV
zm<~a+vGI)%U~0nX{G9su&#PWI)iQHfso1&xi_`y<GAHkxj|?B)R_;R*`H*g}Cbk2e
znJPMtz|e5^xz7^z91W57=`nu`34X5Y@uNqR9+H(z(;*_b``&y8$uBL3-o-J}yMR{P
z&32^M^ql>|dNdl<soF4)H2hY(AZc}l0)IZIb6$sbbG`foD_en*-;(vqEJof-#Ko1n
zg=ML~+6v?F@ANKA6S6!pcCIarnVj2Nug2`LZV?XSzik9wXus9qp$Ar1iy@VLW9^^V
zKeGn}LzG0Bjt)MJ4aKLiUMN09;{KZj_;-De{5>mW%XJl?`QUDv@J~#sa7~Yw>R^1C
zZV6-0V7&5I4f?9bogwK5JZcpzDyHToeC_-I8d)<lKy|RTt!Yc3?>-wqx5v)fxkO7t
zclp&#v>4BQiuK)uDg$LMLww)T8-rt2;h0vwP%NY^Q~kB${OnSSMr>wYjSIG_#>Xzl
zKlG=3<B#hzB^4GA$}PyH44Kx|%X})>@L(i=G*0((@z*EJX{{d)@xynY-V&?4i=ka%
zSIYZ3#g^)$`6|;($E#zM&ah4$Wu<zL4<I;dqGyOj9sUP{we-9fKE0FKIgWF4Lw_&w
z?3M-AamRs^FW`p-S`1qd`_w0{PS)eo0V+d&DLHB%J2$543CE|3{KnnMeft9galpo8
z|35X8aQ&m5cd@fknzM1olRxo6J+vc%zB-1m5Xs%PjF@@;^&OY(adOzy?S210QjsYG
zGH$Np&r+J$mgu&QyO?+0(j>3=^y~%CTCf2s=Rp_O^)#gZAI~p|!k-?f2yGIkaZ_v=
zC3GPM6{5w~S&L7Ca*B~yn-I&~NDCF+e!3|Tb(>j%`e)d(Cl}QT1Um9_ugkjH%BA;b
z%0l9w3GG;d-b@e^A90VlZ!Vp&R#*Qtm-p`yxdF6Q%MRA9sV{zyf-zpTFSsVNc>>wC
z@JFxgx@a?2=RYV#Pq!WGb1To^K`MY|j=_<($gBqDxH5Tm#9CiQ#k5jn)55%0fLq|x
zBTp)T?;Lh&A%*47-M!CzG(9-*2?ur%LHN|#$zP#FDevZqGp}u_Yq2)HJF?h%58~YD
zT^sJW_8;0)gf4c6P@opZqxSEhUjIsbDzZOc>tGLtMlCydJ{k|j#^gC@D)Ky0i%{Z-
znZ{Hx?Hj6UG7ra03pM6ZDudIz`WTkJpSh^g7elQ)u+NE8@o{Y-YoEn$_!TUafC&Ae
zH0bkxe9Bhv-`_rwJ9Y9hy<Bif@4X!gb|j6;`_n>7Mn>mKTEZcU7iJT}I)H`ouX|$N
z3cqHNI|4ZU2^>y%Awy$+6<3~XFi!+4Qf&k_ugEvs;hVQy(@@{Pu<Maqv{UU3yvr%)
z#sth+w`M4D5MFwk!|#6nepLPnTN%-0cFsN3N4vc`F9aEu2lA1Vh@d<>-O0sY9Ebq#
zEbzc<HyaY8{s@~m!X@;7G$J6!9Z(EQ9#-p|o8$J{sdLu-woJqeMpKAP+|+Lv&L_|*
z!x;{EX%ToDGXfY+l2en(Q>`=DtPnd*hnR`B?`2Vj{+$P!ot_2`#wcpXZpsf)F*`D@
z4PhX|;MEXFKa&F23<A=&_1GT*h-<bcXSS9b<6|QO?BzyCF(H?>11Q2lI*df9QmfWV
z-SFW<jq+=c28Q$q+ryUObR@%{$}V?v8LDEaQC^X`4TLxAyBdB$D&fV}V}Qn)2jl%P
z?%hrS4B<fLNDj?0oA;D=3d}wYspkVG*h5;IMfhqF5Og|i@~<0vmYnfFOihB{_w|)%
z^#}jPn&(>|W~Dg6jHjMYHBNyK(t9t?JxT*HXb8AHPTKF*Ov}6KwcGk%c)RDX>FMfx
z<X<rX%tyhiBSAbXb3Wm0X*3OleOdlvPC)-D_II60n(ZP9>kOcYk2nOP836v}^q2@y
zp_FSywRZgsXORLk(0KM0a5Mgb5fe7x-`7JzVtJ*CY>0J3{zLT$wIV-=R^#bLl6|7Q
zqNL7rI6GqLZrlYY!4(9l4|c6pwmdQ>b?1&qWIJcvl-}GHZq3Lz3;XTb*$u6t_@={e
z{51U|c9taN(mO42Zf};73efSgr#ZB>>G3+Njorj(vNh-y571UHs?l5eGS*!?c>@Q!
ze|3+yS9#-8J${FM*G@}hJ@BR9tDj0FRl95@+Cj5TB(I9jV1MGPpJq3|1cZA`ttx!T
zIne9YUlBY8aZ-GV3$a&^iUu_bB>gl3UU+w<1U!Nk2}Qc5HyWIuNcMwKSMiIdXD-oR
zi|o;lLszIlgZ0CD%qt?<X2r)YK?hJK2lLHXdY-dS-XE~=rrVq&=*g4}#3wz%OJ;aG
zJt?qr(`9+(Z3~x4&(`0F{y$?N+3wgz;4Ff{h?Q)vf!P+@36ksnx+JvHpTIFr2K!gX
z^nlI(v=saY7(Twag;>{TZ)8a9cb~;iiC#fbgC{7P>9+7h$xBZ&M&H*x6^(AQJd9QK
zO-NNJKxmnjh>Ty|)p)!i>S7<(^RyyzVIJ4DR6~3oa5ro^Vs0L*2dZ^8FxMqH&x9vz
zUqX2`O^`PDEND~~Gl;18@sH%pdKY%cp6kR=#_|?V1B||VjWpw7gO|A*|CWg_F9k|D
zGHxW||CUl?R$v_By<|Rc*?gwTZ|3~QXGX@<)<pmD;|r3K^yZ1{8Q0zn*pnve4{A4s
zAiAV$nXaVpU$AJ+5bUVwPrNCyrJc3CkDGG|+=MwzV+WNZiLNV}j9mTT<Kboz=!-x9
zf?mQXpZJ%uVl;k0f+J#VZR=JDvTEEU`GSsLQ5AF31L<+7ZrPc&=pUH)P&(dBaZtN(
zYV!J8wWrFte-ktk67E4$bAijVYPWoiOrioiQ=e6KW6@#^Bf?@G)(<<Bw!c~^%SXH|
zD{jt83gci>xmfY6WUxeju-@@VG8p2KS#ie*CM*##clL!>`PdH1cZVjlf5d(1Mc&R=
z@`QA$;j7cO)kiT<MNIRahy2!TPo<A<&taOX;!$-~-E1|cFf^I{=S3DjUi#f@r>Qg{
zUs6}U*zG4G6VH}7x)-^!C{(ZXBq)8nEp)19%GZ5&3l;@JfIYolxb!R4Hyf(>Q{;8C
zRqyYG@0AI^4J;l3ak&_o-^prVkK4kIUxgtazH%P{*>tXM5291?4h11?w&{(q0{edZ
zy-xq*hZ+sn9-Z(V*Au>T=Ivc#z(T*s84N=shPl(h0WF#FF3||TR{KDja%)=qskC`8
zFoV}IGbG}x%iSLDjI9LD9#>9$m<Zjj*WOGtw{VlgD#9Ci2eQi%6hl}E>FVSe^Nc#Z
zm9e3ror@Iz?)5@Jt+;Y?fM{5Nl23ZQD#ZJ~d36f*lb5UOxsMcZDyc+iy^Tt~#8yN-
zn-05@B3DD`fT9*>%(wG6C9kWAAcIR#%9z%8>p#sjR)y+j;UzTx!A0Adq;_r+3-P_<
ztc*rXF(71Vy~1lDEnHH4fF$Om?tfIQ^Arayx31fXRksxt^?jWL+|BBnTPDR<f`5_Y
zOu+#%ZG|EAQOi>s!8HlbET+42Zve`;%J(m}1#d}bCLM}nKmz%Z?t}Fi5(gK3S{67E
zmqWLZaqdU37m3pqWsen#c`i2O81lh)%OJdWkPZu<*nRgEc)Cx3Qd1FQW>_qQw{H`S
zgJy4LOJ3X3^NzV2@7CU@svezb=3s7QjBn{Z{*CZ+H&{A?-pU~9;x%+2>xN{r?EK{p
zk*KNR1YWxu$q%a(8$>yoVoj;{+PyGS&5@9ta(s%lc5}T=t>sOoV}1MGsBEuBn*wRV
zvA64?c*LA=GC@e4aJW#4yxios5OcX-kLvc_%0+mMU-r~?ljri2?<3ovZ$L}nZFWZJ
z;@$v&o*BVqD~_hMGTo0(L7iLItiP(Ne)jP2x3Z>b`40v}OU2IgW18(H7ZE`D*EeUM
zUsV2|u332$jlE4{ELN`9_f?F(k>@C$=OJ6pU`?N*P<~{mTU{yop0tF6LbgGiLCN6S
zJpmt8uSo5fX^-gc!52+JolMC}qdJP|jo!A^fKS<i43U0H+eGtdE3Y8ku%@tv)pU^&
z|9Lyu>U$&Y+37RBC8`rJ-bQQv6|EL;rQY{{^*?M7Q*775W7qFsx7?voHH}pqRZ>s;
zXw~)-bQU%3h01H+ADj-gT4Uhgl>^4KiQdgJjj4j2^%Z)Vb3g5Q^`qv8HMW?RV*<;T
zUzcqfra7NczxhT<fZ>Y;#p^|ia&EhbtS<a_()@gif*8Lnq5khFgX$5t%J7v1RFI0U
zdIK_D7-k8-ai$T|NZ9930Ef=akDc6yt^wJnK*_y;tc<j!G~(Bp#m@H^ZB1^QY>n&I
z+hPBv2NyW#z)I}0+yN`7gQBSx%crPW=6TA4I=*=S`)efm8${1|*;Icx_@Pq;TQt~@
z2(4S#z2N4T|Javb!c+~0Uw>^Y$1^W3kw~-&3_c-1OT&qMx-1N<jPO^`>%e5F+YKn2
zG@(oR$>-illM}rcCV^D3ZT$i{7h)n4*89ZIZE;*$#tPPNOF?uj$2*}6*nW*CJ-_{E
zR_ZOA>Z^-h4+bXILxj}?=@sL#YIs6%d0n=Zs|H{SXPHN<)$=o3eSNWO*`xJ{LT@zf
zaj($p!HPJ}vfVZz>flKHRxTCPlQCy(r)hIWFTn|Ph)w9zt`au=vxoTXpofwl;fj4C
zwcos9XTB4vz{X5bD1sj@UH8cY*jZu_Rz)x^Sg{K!i}I9wuMhJwF@^vF@%}H_B$Diy
zQyXgW-u*lGf6=E-=92-w2gQT(wg((b${Z7MHbT;c8N%rdHO+P>IZfM37;d{e^evBB
zaxg^Yl=OVNZhOev`#M%v$G9<^IPYQoLF`UZrRY+aBmPp*quR&}83kH-Ck2h|vI2Oe
zv#R-7Q&uMY48Ur<AP(mV4XbwYi6fv8x=)pLqSm__K{<OERjt3epRjbk11NnRY}gD4
zv>8L1_<#>|8@rTpS<ZxKzenKzQOq{JR6lLBby_md=^nOu-P4>5eT381YenRWpLxId
z>Q?F`eE#mW{;!#qu7ZMTwr>p#AZE;z)asVpsD2BP*=QZqTAg)L*?yxK%J!fjl4!PK
zcCO%0+oh=VLSt!ZnnKa!G(abRmq|J*BDe+>t<*vB2*<n0Z|tw@dO#+=O&4tY^Z!~a
zBq}nIa%8!|o*xkG-X$x+uVpvagEVTTqhHMytj~SRePZ<Z>6O~Duv~WklE`2mc3M=%
z4q@ERxi1zIGu}&4c!GPv{Xm2x1>o<XgMx$>F#q9UoH}feXElRlckgRtLb{y@g1^#;
zaEw>6u5K-vt^B*EvQ`PYrrA(6a^TyTGiz75yU*8zqq=<}*6a2s3d(o5*%h#jI{sb(
zs%VpF^nEHm^8Uub%{Y6#Y}LiD*^3iSdf)yj)w>xfe^hiMK6PxIxf8Lc?}g@UgudWn
z>=%Hhbbstm%(%M(D~;?#P7p|k@jLxOZ`tFdBAUx+ljO=AEU{#a1p=u;i-%trt6hpz
zb6{g`2aUHv#G3E=AFe|9j7Mp&<i?rH8cLpI+x$>6(Qty1H%VwEXoxB_cd6{Q$eYuC
z&rYX~?ysnSJqNUTGxVZE+0>=(-ek9{WFAG|g*v`0zFv$W_!)dqwS;nHFS&U;inVme
zdz1`T<k;;Dlm8oCo+z|={T;}z@{oW1l9=-=Wd8DkcDZkkb{tN4sVQ6_3Vt(j=MHzF
zx8(kDSAJ%Ep*FC_3R7A9op(Zfq~s9vwF5iJmbP?)fARj53Y|iFvIu8Ymxre75y#Fe
zxfn5BAlBihv~ZN5xkfq~f?PEyIV^4gavj0mPSIi3W?*aF9FU4^XsW8z`%M}3RJ9j6
zk(fRJirT578s~L0HAS#j`eRqj^yN2RuNiK)MPDsZ_YBW=!>Mqzj!6=RwAWnD@9~~A
zbO-qEKstXK1B#o(gS@HKZ)4NAblmnZ15g0+%spKa?OvNndmu!S2a?@ss*z#(qheKU
ztMzuj83CWm0CA_+nKs<bsgK>OeObdhlD{ows<{k8bt!5G0t=%P4WVl>dE1UpMx#Tc
z(LBm#3@sAcHV?oMX|1T%YL>{)GH#3=HkC*GM`Hy%6Ol)A%GOGJ<hkGEWWgZp%z|DO
zd1V0qdRqNpMU-h#pY4$d?Cy*FAB=1%%tD%SGwEcY^THH|R)$YNsecK&l8|_wHg9Ws
zs(I%BNl|7sZ$Iwo3qrF>IY&fmO~{xT@)i<O_)m(vce6BN>|ie%g&}&i*adxmo1RXU
z89kQ~T^*-naWkA=oeQ^okEN40SA)w5Or~kauv|bNsdnyY5GtzFIS*$5*3DdB0FffE
zwMhWQ%O5hUrAM@#DW6@wXSoRf=+1g^k|y92^xlFWGb<K6QnH5(qi@%;y$Y~0;<WII
zI)58lVwsC*XVi!d!WBDha5S>eW=Hoc18uyrJ=*U7)yDI@TmPL@e?GS0vi`!BU4vD!
zx~|#Ix7n_)MM5d!A!k5uAP#!QhFxiDE#s|qlt`r7_TM)3LtG)|@h2+~zTH3+?+~_9
zv-dq7w*q?4nKWe0&No>;lMtYhivfjPls#a2o)JGVcw$+mCtn8TYtP8&84_g7qw(K9
zLA4-@1@pD-X*$=t9d&S($rEOCqY)dc|Ng{CK9N0he*41q(bQviF(~UL>WaLMs2zW&
zd#|c+lDBlVnRv1M$+7$*HMZ$5HY)RD^aX{~xZ;)Y`h7dV0H_UTPRtUI7_#@}GU6i^
zDZm@|jYTR*GGum|fu%R&1S51jTu^8Q;fYLic>-XcfB(ts7YQd#OLDydXvVhNlJ%7%
z)CHKrnBN$wq(2HrjcQ)=e}(fvuz&O){^Vtegl9tU&X~0MwU{tgV9}yH&ydax%}&r}
z7kqh$U_n}Ul#;g|qZXn+NZ+Ep$im?aUNhNaWp5xz!H(I;?N*x5I5u`UV;DSRsA<c5
z^TVy_i%@{w`1=CvW{iXbD+@jZ%$&v<{KmCBf5Q@?Hr|Okuu!19^bnPDo{xR6fNF>Q
zmkyzJeW2^tH2NF4*K4@oltBCvV~|mL<>KD|7v>$mqy6^_x%FTcI#%}BPfQS1$li!z
z>b_z;Q!Ray{*1X^)K0NLm2rB?0rt?s>Z|zf<Le3FiomYuea;&O#(C<&-nMj2h<V)9
z-cwox)9cQT>7%t*cJxAjvN7SefOhW&aS%oLjJRnfu0j*{q^lNhL_uIzlVRLpfHRL2
zI*<vWaDl;;?HzELfR<{~2r!!8dzHiAYyQ}4L+GCr%tpIcG_~eIE15MBK4*Dr#Uyv>
zVi%6?op3>9bE95%`TmuiyaEH1kitb2ZgB#@y|Q%vJ4n+<?^q)nYRdh*@8~;JQ31Bi
z#+v;k%?#r>lB~_UMs#B3B)`TV_&8d-(8)g34qoFU`x^wl_oBn<kGZJqyo+9}*LKZj
z#9#xPAwr2t91T+Riqmp16^;hv4LfBk{;>7LJW$>-tU#FX;tk8UZ%az(4)d)Pm%ty~
zrSZ`98~Qb*@o?qUl3nSwcTdO>rGD8X+l`}+_}wP4!jVcJvch)R*SIw9$xy(S3>LQr
zav6VhqtR(Kjh;q<8igm@z5naiC)jB^?e>*(idXY2>IEGH--wE!{H5}whX5JsCmXUf
z{#nc-p=I)<e$i#F$#Jv=Ga`Yg?N>vz1h6AuwY$dIrbE$JbbZ9i?yP`dZmYW9$mt61
zL$X$_F=>uN6)j%CUM%JrceB7<YS~BJ3Q;#X-f6ZBAGn*b2+@}WY{*c+m%8DWrZ@ns
zbucz5u)~_Nw4K0R=3naQlBFT4Zj)#M9p^rG=l~%4sd4d&L=4M(oY(hRa#$<^<TlbI
zh9vbvtYxrh96Ih@TK`P#6d4P{2tuhpE$wAa+`n#|<ejCq-vwLNV2XjMRnqDeQrXQ{
z&Knj2vH9DMFXt?*jURKVPpNwI2W50|=T-6SX@=|11k(*^?;_5g*}^yCpJVDvWcRRH
z$XRipz1}8pjc(-kCkh?@4erBO`X#{UQ;#rCt3)O~;BYiCPPUdJK&kpF4Nw)=)|8^P
zPX<*-2T%JF(tlqSiAq`6_#jF9?Tz*i^<&g7{I}J7k)7AHSLYb_*O(VFaUoyGWb_|=
z+?L@PP*CdZsTpr`k{Bx4-q%vb&fAG1V$fAhhs&~e(u9OqvwK_k@1#osBWRToVL8*w
zgc6=Aavz{jc$tEILkAiES0{@qnDi4Yq+jp(0Kj`f2PVf^e^bNn0OImQ&L%0unTskF
z2#C&F?EJ~2{AHCv9E~@simqlGrm+^>LVq&k%H-WI<O=QcJ(6@a0~Sx5^lZQVgl;jG
zgE3`mTjq1TKFeI))@k|&E4^~)@0aVBf_4f=!PeJ{M|4Ysu1BLBzK<6=D&g#NC__*P
zs9VV(-m&eMTo!P;;i;xnah-IVHr7>9WWvHz2q@bhRGS)^InH?+Axd5{&Q?vrle=NB
zc)5J+=yK2<7&k#t`*%~Sv2nF1a^Z4I)!PY=)5+)*vF)QA16*!ofaX0bt+Ny52r-Ma
z8?)|N3fls@G|gqZ%g})djaWazDuf>xGnptIfqkY<*TatOELs9`-)Ns@xXQF4pxxnb
z0sQ)($`W&%&xc#rkjSdmQBQRz-e6Hp+}AGb|G+nUuj{+Mj|2Av&p)kn8P&*Z|IW?d
z4anWUs`6Ys?2&Z^&j)a9Dy{Y-Upa34Z~&C4hc{cLZ77O*<wB#tR-6ZZHgVUsp@O;T
ztnB8ZkppqiTozKNk8f+vRiM$k)7N`0;VNJ)mF`Iv*if%X>cOgg9W;Tca7nL61f;wl
z=*YX_4rPaZzMj7cpgn)e%s{FV;mk$xl<T_x1Z5f~-l$rL70!oQtA!07@#bq_mkasq
zY~KsKiO4s<J(N?NX}%)$CflrkZ!T{@V(v#nPWRhgD4hsz^xqvyKMtAYeRqSrspiPI
zVg;5rvR#4M&0ZOAtu&Y_#Ec<4!fW82=nSDqKGH3!Z<DJVIzA{>|6euG(jPP=`TdQi
zA2h_#u_*XpZ-QD><SvAJAiKw}ysfDFV^p63-vDRt@tfL%*Y)D#<_n?)n<4`Hd`P>C
zO$))VC!R1shouvbQ?EI&Qx>|rX1nhx+NNo2J?5_Om&fn4nX_F_1u3Da!3kL>%VmU^
zFP8z5<G@<GdecGTt8V&=_t}f^a`z8(-X;5u&`jr7mHw$V?3mF_56`AZGwDjW6Y!*2
zi<f{9ZVxoZUs@{Fxugrf0VvqAt`e^|U|AdQ<CfgYn+iqZjTv~Lmd=F@zv7xGcIOw3
zotBcj@eC#RRmy8`WM<A4Z%#BiYiluP#<X9Eo>+UrA%a=cX5;kVQ%PE?Rgmrn4tGcg
zESHdI>*=GO=1T4dFPN~h2GXlO8(o0Fp*2W3dMJ{y$(#|oMuxP~(PAVzTH+YiPE^H+
zDPZIDfB!lZ{zhh6{s`#0jvkM`1<$*l)tXXTfraB1>9hns4+fi_IkO!2v>-AznnAOE
zf9>6h_`TZzQML8xW28Det#t}s{nr!FJr{?V4UEBQ$(MDF_#mK_sccDzbrB-&Hh998
z1%9PuBx_a=@MtSz$qSVQ(~d#))YEX>2Cy3V-xow;x_<JtybxB!QJ}nHz6Z!PI!~oo
z8$Mgg%YrQ^hzK=y=}O0;c|;;#DK&FGF-Wj&%@;NYTJA?ff)ISKI}Q!d=K2CSV)JM}
z;C|E_u(2dCM?xI_ypTNn-z-4&@m^{+J+jKM$;#HZXfy&Mz+Bu*xy4I$TV$$YVnW8n
z99?xWL*M0>_1YxQZBN5(ISN$lD*qcE@#sufb07-&KO8%e46{zcZ*<I~0#c}{6j4g^
z&h?QX(PHFZ_a|u7S}eloI(;QC*>J-Er7&)hHYjIRpBpno)8E>`+SR7jy-y7e%~~7)
zQsN-8Ib8Ey(C(u3^*eSw5h1-zHQub72!L$BOuad+d6!f?V8RqAyJJbMu$(<0r{T>x
zi^#LuicMpaO1q`h%2MJKlB#w~Wb}(AZ5=MI?u~9*e<z`QRNvL1a6RsG+4i`0m7bX|
z=*^O+ke|Iaj#~!m`)At5PkZkCctr;%ujReCQfr@;p0}#{zN-42@PbmM+=Z@s%lw8_
zZCLt#S2=8}$5&+9u-)lkH^cVEb(9y&W<z^c2f+pKQ8@Y=4|bX>5AXPU_7E+$a+gH<
z`k6>UtKw>Xmn>I|EA}LnccW2jUvx*j3XoQ>yseHAiV7?H<jm<npzN@{>f64lb=du1
zOL*R{RP57ENL|$O<Z8&Ru!3L>E}F7j^`xkI9b`%e_pI2+4<x&kZPUN%*GIN9rdsF-
zG~S{r;%J7fk#vp#E19pi%Ykt{Q^Sbu0T&ja%goZ@g%Gf{=i%_kDU#i8@Z_}b7p<;4
zrKI6TzoX2DfS-DIVNjSFfqwE3%=<E)i+&?$$z)j@CT3tTV)05BkUs5n3_Ks?H!9=G
zj5($NN#Ig}{NyunaKQq6rlm=rVE-+&$$MAmGXJotfzaIhgm@6m*$uI)mB(ya*D8wg
zF&?ngk=oVzy|f>0(*iI#?zA-xtpqcyoj4zyQk=hhPIoDRT%grM3XTsIBb3FM4USJz
zSEe%WoVoNg;F67CrwwI#>~zkaVwi~V8dNu6A27FOeKsOlhzEm2`d7fP@8=8ny30@Q
z5@%;MP+pMK<jadV-w_dm$?>(G2ZF$1;@0kfqckE;07`9lcVc$WQ~e)@D*w2k=Ua3d
zhjgEWLt4D~{{XGadxi8}9qi=MMrS1Lfb(9Q`xXSLKEW8n`hEOtp4}hdERQUDnEPU$
z+F#ySv>ag-cu~g19U<BMY|7o;Zf>7#D^_|@mmOdB)3I?9#-$%`8Jght?cT+Kxq%_c
z7AU2Y5q~|1xbx<BtCT;rz6ah_0PwD!a#3giON_dKZpiKU(rRJqvvm2mwjxu|o33(g
zwru*XWzL9GXi+C%yPsB5!VXrgL3qDJ%h`xG*Aw~xH|Hd2@bx&cy|S*agf5kPPxM(w
ziMOA()9kqMS8KNdo>3>q(7Q@qg}(~6ruL^Zk=8BJXl@`?U^#Aa)xoy$*i*B%lw%_E
zm(+pfI2rlcyHt$h)#TtZ3t;~6S@69wQ|&k4jHlQn#aptkg|EioWJX%~Q_qU#fzE7c
zi$)fKxH|?H>)i3NK{@eTJjQUP-i$GYG8h#n0}z|oNml1K)=jzeAm||Bx;n7e$w~C!
z9Z1yckk<Q2@XDT~+ePKFZ>tqZut3^4r2j;X#$_k9#SAr0gtmt7I+P7v3WtNa?H!+i
z@}SQa4pP?n?ZP>b?Ok9xMqne0U&Xlnz+Z^zTR@B>VLfSnCNm;ah_l`uQF*-F26?0I
zStG!JQ=(%z*65vEXw<Tgi|mhGPh6^3kVe6d%hLG}*A+S|oCv)@6HE}}TZ9-&uhxcR
z0S}8SOD^dEX`p(tfC@-i1ChwT$m3>5t`zbpYPy&oSJhOG6wR3Z3M9iK--JlK9!>q|
zh2b1HiyJ)lcaf@`HY!|jm!wtN@sOrsC?}%0F(%k8`!#HXG2LoNb__&S2s$KJULsSF
z@nAME1c`|O^SfcA-(dY61-5|dQGnZfull8??%QE$5B0O0iw&kL$Q?aeE4`eee<M$1
z1G!3)tB;4YcRU}I!3ub0#Rn!*KE96A%!=+Nr8o%hHCOHocv9c0(9NZ@u$2?PSjK_S
zYMpjT(d{NSzaiQzEV2>(w^W$)B-P0I@1*+7#w9paksm<5O%)`n-%Z=y$Ej$r691Fn
zL0;)q<C;|VMiuQe8vVMMlEgIWN0?iMYW~U3{qBG3o&QfrFyUmWl$E!B$GRU>aqe|{
z^cauwQrw!z!wU4}906{-9*J*?T;v71utT%dY<=$3j!{NRXZzT39^*Gkz2*h|&eA|d
z0x79<ICcdX$9U4{$@%W2ZA#<nd$Yxyh&7YwjHqLyHDuz+ES7q^RY<igHVOD4bF!i*
zidu;inxipt-F6?V`Sis4os#3}U$TH3*zwki>i!tk^FbxqGyf2Wk8^+7x#ZRZpdQ)U
zG+3*|>RH$24-^#xAUX79sUXevCs#%R$j%eD&MsZWKfkfLID!5?XihA3cj*abNLJ8`
z=Vz9rGVW6Gr4ecBb9N}s8BZ4yDSYTyuv$OkW>uY!?+s^iL8m6m{buSR8ET&k`t$9;
z+C$d$R@P_G=F5{(GFM~MS>!zb&WJMN8nM2?oa$r$7OV#Wrk}CEQfgvrx{wwA#3wZ(
zbaKHDhDs-{H2Loe0aN+mELVSvvU0x&kbXAgYufkk?vgNHF?2vnEAQri(^Hl5Z=2J?
z+f_V}Hcl2oR@RPCi;X&AIGTQq<n;XcWBa9$my;p?2)%6Boo?r#Ju!cp=EvIp!y>5d
z<@Qr0gV}|h6ab((1g?o5HYP?*X2;}MW<+Y^UwE(5hq?onC@1LpQ`xkxr6)DaaY+gl
zk=u-YUqBkPPCL!}p-!OVF&b@!w6o-Ve6Ty4Q%=BG{?Mj&G@?nq+#3z-t%KJOYST?e
z?1gL%6MVrg(xY?0h}kM#A2;Py(7=+rnI<&-HQ~zt4WNK<?TE17)qb?q{<qu>9p63o
zE9n=1c>>$O)N{{(k%M)56|Lh>o|NWH@1-24_Bibo(2gtfc_(Q4wY5hp{Tg+ciT=_;
znzyI1Q*3D%!ae{{#QIwy={{>t*NYSN7i)U??=bJmcv(8$@wRokO-YpGdEW;N1PJSs
z07<$KHKd@ZDP4y_or=Q@W)3a+4T1Z_YwJAx$8v~o*|)LP;8F&3<lVQ5*)a@O1?ncf
zKo6DxNWu$aU7Ly(8IR>BrGToy%V$@ZtIKR)a5|VL^I>iyWGcQ4eDN*bwQGNHT@0Cc
zxR=M%rtitq(1iVcr&P(9m3ppuGS#TDH-mYT_#icnQ0JQ9VtMK8d`ukLpi`-`G8E-H
z;>C2b*mBz=o^(e1)tc+6l3-0(PX|wL?0#sfKisiSudaMR$qWWd0?9qS35O;Ji%-E#
zeX+u8TbMkX5ihh%@=QL40y{aq5zJs$jm!TvnL^Sdd0LO|VK2{y|DzuJPU`VwY5kAY
zC0O;})nh?6P>-Syu}^NK$U`*WseLdmI#jXYjq5YFi?}2%3XsEM<x_GflU7R8M6>rn
z%ZY$)wv0=(hd#5}3dBJ9&%eOmn;EWz0nCsMdFYW%tB!_PQ+EJZw7hfbRY_ohO=p_m
z=F02t&6>E^auCrYpS;o&hyE9O=#r{8wE1%`MFh_MZ<(JYOdfzE`4c#D@(4uj0N;{l
zfrM7LcT)aobAxgBB@R1|;JR=s$9~u|xIbi_A@Fq+KjL~D&>N4@&6Qp8IZypveoWiF
zBI=_yCPI+gu+cf>Q;+`k&1o$Q6&)0?Q_>p4#B6{OliZ7621pW>$nzZ(_o$m4**zUS
zaD*y(t{k=e{(Dx&CTq8pN(%VXy7qeRu`t-u!Fwxg`OpRxXd-ca${~y$2V>cX^sDsu
zZA51?@+hr~X`6QR-K2@)kA==)nSdia_}TwD^9LmO{_|I0#v6^bV!mt(y$!WAvxvIS
zpPuIly_LD}_F<c+tCyl_*CprA3lz%1=nB>`cFT|tUtmc&{u}vn<j`6~#gvfF$kfQl
zNILmFXRmsK&MwhQ$3w>pwuWooU8|QpibEYNI#KD*Y94JTqkwax#600O)yN(i-;#BZ
z)?k(HxWDcIEXSGXwwJTru?U>dJ!3s|3olDr--Mrr{d@=na7Hd5LB?~c4lP&W1(vUv
z<9vmxs@*_#aw29ULEX8Bj>&n(Be*wJRrFF&WHRx)6QM$k=vX3qG)qqCLmi_n3ukov
zSBy~m(E9p`3rHtn+1<{}$}y|KAmJsvxvZQ0^t?c~!7hx-+bMN-b{gTi^O^%hcJZfo
zVtPdcgxhnHp}SqnY7uRf;+STW-RG;(&7khJka<}job)uwCtj?g^G{TAPwkLIwxTt`
z(xn?(u9&geV5L^o3Y!g17<H~6cXaHiL|6$|xd!rv22RJ1E;jonlsPJ!v_FJ%i}XF3
znE#e@nLndP^c-Gmj3BlMRx1(PukeTOqmDPTAn@u>DX#c7C;z!IE4MWWxV{P-8oHsJ
zH9mi(QLmO?LldrqhW8P)2^BUb{eM_hCQhY;t!ceB(jL`M5%*cDmh=f>Hfj4y=OFwh
z5(h8i6IK`{aohAm#M-Y7<*Hg;`h3pdg7M;2XC_d^*h^G$gLxctqvpDe+)odJI;!GG
z@SaNx;qG|O)UwC<rIGJKw<)G>6{ymQB8ZRd58sW{5bNuun!M-QHzwe7UQGw1<q5ky
z#Cap9sfIB>eCJoFUy5MCh^W?$U85C3!V}I(mu#@Zb^E)IXEzjz_P6i~kq<B$8TrJY
z{#O#^a&yxcg5kiKv}&^F5%<?OM9i(dbPtBOl~~mABBX)ex@=<<B_a9cuE*&WvX);)
zh%RvM%HAGxs_XnuZXZ8@(1**<jf1hTns(uY82_9JV!HpP7hz*9)VRBeuqXzes2@E_
z$L)FMyF`54>qwN)sU>EG8}ykQE#M)a&3BED;lt^|!Q!CR)_@F8vNxSO`D3;MKew%(
z;r+>prd^AkxsdM)NiKCh*@=>CkNBLVr{ZGnnEKefwplco?%}-b=OnnKy7Z1S^&p{*
z57WsR-i%F>7^A=ibAb)#(6+8ARbExI<GWV&4o<x?sO@o;%u()2@S}0#$KT^lDn&kZ
zI{AxN3U@=8-~`orKgSRb-*N#l{T;qPqufL&aqHU^Xxb@qIvU^D*uw}XPo#HE=ihrP
zo8OR5{diutjwpfGIykoPJkHaCbOFiKo$zPXhb|aLp1P^eE$}3Nrn_h#S+^93>J=nS
z3bJtJPO9IfrPtU#yKPwQ#}x`->uI`zbA5I?6FN-AE!Lt&o6s1q!oe0jY4A@EL*EXN
z)_6o1*#{(7<lVfaJWIqBMpAN0B5YfgZ7W`GS6Bj}q*c)vWkUc%vmy;XqtVg3^yfyB
zd(rS+o+f-RhU`Ko4G@AaX-c*pJ-Hp}dPK$UZn8SMakG=#s_Nx27f;+Q^5|Q`pILiN
zfXTre9&}$UxZcai#dEZKw3O52XSq}VHIEYSnb-^zgZx?z1ESB=rJIL>w|jW;L6;Ny
zj6LPC!T3AAzg6kD$A}?0I37VOD72TxadHxdPk=8aH${2GHyxp!2<_p?z12X8fNzz`
zCyv@7#$!Jg#fe(k5$8h{`tU&6#mR~;EvuiyRR}B2^}Q+D55+~4znRXsN^bEp>CtEY
z?De<cpl~=ihoxw)&xd25>!QA1wJM!p36k{Xk!5bLXhp-Yb;iIm<EH7O3veuP-*c@5
zj%O(8O>*2BM;%$C98H=AjxJ<~M)tE~G>3e3obFIuuTGB}%FY+H?~!j??FTjp%J7Pq
z-OqJne%>z@U|B@>)5rAbaAJ%3Pn{Tcn~&=CWe5Z{RLm=*OviP+D2n4<JSeTXdpP@x
zYgcq%0yl2~F>#`2EWIIq`hOpk6R$&$r)O0$W_{fsEsIAu9Y-D$Hb+<*kJgC3JD8@O
zMKS-P+KS!waEAIOeMfjs?j%UaI$4lvRhd3&QTn<zP=t{VaVV>Jze2DDWH87P04b1U
zsH4-3Pd~Hu(z7oy_bkKn8`GIQ%mBzwLRQT++^zV&WQA)AQe~;Qa#bG#o;YY*+}L9|
z>caS8$>rfqE*S?9S&bRk&9%zO8}I|tz$hL>Z|vrcFjd(;sxo<V(PfUU@AtVPBBijs
z%kQJpdw<?;rS#5O)QhgQO8fwZY<4`{YJJ^N@0o|5_Aboka|b=ZiXYvAMF7*+I{Y)5
z&l)0s9|!E{K@%@)#lc~4t`J)=^=>406&l68XGx3WV9IuJ9}dUkCZaf3W_7iG-anbd
zl>4)nDPN`yeQE{R-lMAT%;v;|(Sa`1=4Ru$Kvee)-egpE<XARvpN8QEz4?t|^0yaB
zIm5e2Lnr#TD;3%L)O=6A*~0*fEpy-ym7njy^fPb|mC<A|&(hs9YU`Qmp}o)SNB*|3
z&1(L@b4$)>0l5Clq4CDMu$5!}oK>iggFN+n1`*%4_g|FOJg<BMDfAh^IYR#;-|(C*
z9Kr9i#8(>T|6^xT>HC{Z`pU=Cch4WQCm}t}y!hwd7j^+@EowjG`})TEq6*0p%^1tl
z-udbk^QsjcROqdORcPLXN(6!TJ%8ujaH!X~i|@D$5j%M4*B@+TlO$g<6?xz5uX?eh
zvZ2YwI-cfo2HN-{qY<2hwrlcjfnx(D;$>Y}$-2L;<_K;OI3=jG<^o+|Au(g=LW=Zq
z>oGo+_G-+5^*f)>3w%Etho&aLu{3u4ii683duC8;p5m!z?W236X7wLWE;846vuu8_
zPN(<~+*xWL{M3b9@UPNhpigDoAl7oQ6mTJV!k?w@&P`8wx<dmN_<r8O=6MhB(8p=7
zMn7Zn%J6aE@+`qT%n)FcuKxHJq;qEVNqEg$RSf)EGRP_!yo>l!Rx~)Fa;alg6AyW?
zk09fS82dZ96(Ln5F0)ot)z2gD@YB~yk}C+{h3Dba{T*lvb`!r*NzJ@C3gR`czCnBl
zx4m{$_;duvI$>mqtBBrXA#eKPG=zsdj0-rDU~dtQ*r=M@&t*_Fd?BXrj{^KI0DN0~
zSvK@jy0V8;PuGDbvDY^`H^p19iXXoqyF?axxqG!qpuj45a`pPWI@=&AXM`AH70>0m
z!zBk9WdGnX7>f5Y8o+RkTuMeReExY>BDu`Ee|6YUPzCr$;i-$B#`ObXMMaXY2KByX
z=jSudL-L~w8u<}Kj(FKSKw)HPWX!1mwU@udJYU)+YV)vKCJ3mIft8&CiQ?ve7!~P%
z`CSlg{##6%hygmJFs>?N;1+E$AMiu6UPgKp9D>biB)gW0j59dFT&nbRf?hT7Hywq6
z`IxLdc%%95-wCY)Q=W|J?lg6}m-V(4evYHxivx<0A!}kop-C!eN`r(jUS-?bT;5*j
zdR*d7`mqrBY^JP~NJA4wi4;|09f_Lw^s7VSji}(c3`erpihYSczq3-MmzYoOiqXQ7
z)`(?Xl^I4}sH&=J`^x<|;LiE>#jV9Zx*kIUDC$odp@n``YiZ3J$VGjmZ4bBME&>>d
zcesFB+nFPr)*9<&qbvW6Vu5b%U$rlMDjQspvEj3TvqL5iJmZz*t>Ls0eQNN>T7xL8
zC*OD%5i7vsY+tY=sx@HlF+)iB%b#B^Vt3W+wF@Rg_oLDAu5=ISKg5R=O5pi|o20_-
z^9N7aNtHoi!Ef3sOXM3~T7V4z(tNG-PVFjwu%qz%p6(byPke0k{&{BqD7`#$Q7B6B
zyv-yrmpUg0@i<0=+4$!J4@jQ7y~L0#)z%5uq5XI21HM6R#egxVm^q&H=~g8<sE_VC
zO}vR~a!3xBgHW=`E+S~Koj3gdqw71Pn%uUoLz6Bxq!&>^ih$B<!~&uSs30I+nn(xf
zBtelP2q=n(G!-e*1?eRqy+%NK3B5>3s0oB5-;U>;dp-9ZA3tQo;fVX$d#<_WYWulv
z2(#R;V&C;cXr98Aq{GLNfhQZSSyP4e6X^d{BI?q{@>yMu``oL%-s0mWzU#W)GI)>f
z>B}Vklaimn!BG$HFm5Io?$q{X_X61|;R+OgY8DwFO)jEZpzPBSkAyXy(e;6UmCqY|
zgB`<&XL|S_Kk#+?$DpI{flSm5I)S0jfabA*TpdhNL&6fK(h`zZ=dH{K1QA@1jCX3$
zTumV=DT*p9@g<l3M8ype{9*U!77j;<>(YRGG`B`~w;4n23{RgdvVbS$hyfhil_xS8
z!K7sE&#E%m^y5mBSK?Xv6f4ou>47QN`>J>qGr`?Xe|+a9EZ>wEV9dolj@q7k6f^em
z%B_EwpAC-bX{?v<wtVfcz@vYoYzD9#@GirX@@iJ*SJztdJ{h7*z5#E-zdh3cf*yi)
z&%eq~MKQE!s*5+mw7<nvFUF&N-DGW?#NA%PsDOPr2N;)i5EbmH2jh85s5MBp3imCp
z{>GSP(faa~Dt6M9?OwHz-QAr(mpkkZaG<nP-N?Be_s|0Ue4=*kq3jKvFNR_gRZpM&
zFUJ1z9Y4vkLg!B%SRA5on+A;Q@oUAZ{^x@xQeoOv{-@mkg)^&0V8kb0W;#ALB`*0M
z`s!Q4rBXe%le3>Xl5FkKfI3^Acp7}u*o$Cqi;NxQrI*N@j+}k<R{7V$j>+R|EgK=a
zEa6V%pa1IrVLBE>|LzI}Udzq8UcbG@?(X1b_+4XpWHHFSe7bXB4EWZL;^hFDxnzFM
z!SD!l6r$cm-}~bAGnU?WtapV~yhxM%!>>PY8!iH#n(Cz054{KOKcgpVQ!0F5BHYoP
zcbu5B_bt90nN1e0?Vy9+)a7#x7{HA)g*DUVMVoK|KU<q-2>vUi%!_okv0mksp@oru
z_1bm;s<(O)kfO0&O>r3KXe~vP_=C<r=x^1jSEj7k$OkHpor<{OCPP74<8I>Tw3gl!
zpJaZV(5<VJX<$P84}|@-N(m5qC-0;W@i*;c`Pfuf%Gptg=j_*go3e?{ts4@uQqux<
zv5Tbe0x|FN=ao$8n-3AA(IdZd>S8aiJKzF69+!hG$G<Er6R^2)ob8ECmGzdMb)VG<
z=a#|2=Avn8ARdmq?RM*Lb*8wo{FXd>Lo|=$)%y*{#zQF{6LyEYSB($Y*xHg0y;s-6
zknH_Gb))k+V7q5|w!=I|d0I+&x`!vmVl-`@IF{WAy=L_CXRas6+k_V1EZDExCJlj<
z^3t$=5<S-mhN!@o8f*OU3rj4sOS&RJVm%%A|JQr$AUrp)d>_tGTxLpxKcvpzC9viP
zodK-S0P8_#?O{E_>$1RrOlXjQ4b+<npn9W#uD7N;YFwW^JhP5xX-#k*UMVj5GZrXv
z2KI3FMx5+uR_&5+`R}+-q5PiA0gycRcVLI($L()OM43@E+XU(#=Tq(tmCrirfUENG
zMDh?Pi{!(R@%sNZvrFJ>#if@-E-ZPcSC;byDgghWLp!wsy0RjEV@lv&<*&fAG=pcu
zn$_AaWq>XxnW+5zNjKKnhVcBQZ}iM2e_{Y4A3LQ|&$Oe`$ocs7aBbc3Ay?T>^H*Z@
z&`>+NJO!VmFD5{Pu*GBpS5uyY*$PL$hI6y2KfJb*C*YXVLii4{xd2jW6*l0lZGXKZ
z;2(rJ3xdO9=b@vk&Zl^W3ewNgVOQ4yL5Z5DP5;XWz~>}<Bm3d6!ILJ*jLXu|C4^eV
zy0I8*A?%;$Uw;7XN=%3=lp$6b2*{Y)EMFU^+)D$C7vHc0KVv95DSx+HzH;DvTbkx)
zc8O+lpdx#J*3QDm$whobk~Hp7Tu9EJx}!eqXb^q&I}HS97H$+cMUet?H_h}{r+3Ph
zt*3=3ibl*3%`6UV(q=*ceqki@lJR&5i`Wnu?!?&&b<xTl;AoQ!nfh}_i>xVY%o*ap
zS}60PHmQ5K_U){@`Ijk&iJZEG<jZCO<VAp_j~!z#E{xTOTk4dFindvXv4ET*{FsPu
zm50yIXj|h1g@3*3B8RfdBH3H%us#*G2F%1b5MP6y_k8*ri>U|q42#L^Qn6;he|5OZ
zP_)wb-Le^?1Yc%v6*}s5U<xNoT(XSaDFfJ_k+XaRICpNKYZX;~T?K*sZt7@se!ln}
z_jsnLp2X2yv-$wmI}yL#OMO*S5x&Jp4JQ~L5Ir*6TDI;1je$PWDp!%Pnf~XrDZd5r
zhl_OgYpq$<eF5H6C~%5yiKH3+Ui`h-e;MHo?hL$WPsh)Dd)X!P6WjN5<7C-GYv2_N
zzpfGhn(ilvsN`Mvml_IY1$wi{UQ?q<H&>u`T@d*HRHG+pId`mk+0WnrzgtA*pwAis
zG4YOld{S(T%p_u|*7iZ3s2A>-3?Gi}r=Pff?BCSX0X%h4W}G#5{;cb&Y`G1_UIyqD
zvB_6TjWgeiZ@S*yDE?lS_-hxLqe1H12QYFn+fj>th{iOxr|1Gd$|$bsn)+wj4C2fH
za4x5?#5TK4lJ>#*JOH<z$Z{X2>bk|o%yAp^UQn&<)9!%)FLrlrbk%|G`^;`hX!&y5
z;Te3R?&HyrWbr@pD-cK&SfKv#+8#k5-p#5<UN{Yj9LB^m!3(s3Ap-XdkY3mw!t&c`
zs2zNIk%JNEE%}_#)UKz}tGuQacTU8@|11J@=KS|3FLFxdQZywd?YGz-C03BOONSj@
z*;kT_oZ*zfWs)pJ*~<}^%<lm0(I$+`b4rA`!~G8@)|%TVbXn*ohoZ3m!Z;Pc_@i@N
znI5|zj(0naef9NHC@af7@is-K)(B8qDMJS8_D*da(AR%0y04Jm_N8TLtHdp#01=|%
zUiR_Yb?HUwKMC^#FJLygLCoQFv`gz`xBcq++ST;iYap-Wa@X+hRX&olr+az;0xh*H
z(ui~q0M5!c%P%`ZBL9OL3J|H39f(V*tl%_aX|Rds&jT4ki_SEHp)%%jFgFECo{pIU
z9_RS-z%uMI5rjd<fN#qFA^&=}UvaPKcz(LY8Ex~_{r_$v1ac__ND$|U-)0yEci<dy
zmAA>NES-r#ah){U32OiCBCPw%6NP|Czp;XNB1jfGd(^?V3fN&Z9BMKVP~`XTcw3GF
zM(aaFYZiVMlsGb0tX_fJ==0JuyUs8Bl2+!rFOe*is4@b0_Jp#u`SHn<KNCppf-#6u
zpfv>rS^rpF{V3pwVf}O~TOjf$cvQanrdNK_cqjA1tP)(3JnsiWyg*mF#+jT{o{~r$
zbos2u!K}|?(-0s%Rau{i{6*vceBaMxKn{m)_JbgYxOL=UC#kx(rc4bc{mgEwA2Ft5
zL7|0JfWglJAO#}<%}53s3-;srV2H(;=ititBmZhie58Zr3%(xan9S!3wu?_8buv1C
zP4r7uMyQ?9%NYu{o~Iu40>0Lk%<SqG&uR=jf%E<q#z<p(^6QB8+fZHjXp9C_Gz<FA
z@+iNHYyn(?9{dn!*_Z%v=-1=FJz*DDHv4aVp_73>cr-wMZwPbye02GNfR3~xUsF4M
z%2F9UbBdG`r6m2|pARhn&VH9c%wQ47Ih5wwvOd|Q)!g^}Q%)VK*7c?#kQEs?WYbzc
zh&a;24Sov0u@sc3qhdo|uhLG2{fr*GE!Fo~GUZQ&9jpbu@}1bN{)c?Ho-5d)v`GqO
z$I(9mY9fBd>)T&qRfl4JAT6Q(HTm&7%|%_Qo^@{<X&q0=<okF2<@U2hff$wqX;+T`
zB*ox7h+ig#M<w-50!}v;1Ikwqzi}wf5GGZY9QOu9L*GaoW0O@lg2(P{5x~bThyP@L
zh;$TiScyPFa&9YothMy)A)+&$g)Oq2HUgOQG?^3w=`dRg>4V%02)5I^r<^$*plerg
z&PN>I=d5?nf|AXDHxKd%NT=W~O{%UQ6KKGi21V?xTg1d3{hx64<$C_Myq7nk56<l%
z@GTuS9(zW6r9`U4@B5z)`4TIzv6rS7b<(}Zqd<vcDD4p?eL+p{LFZUUk>o;PiGOF7
z^NWdv?%abREcS&)ahi*+QulqUE^waNT>oQn5a%;s=a?vx(bLV(Vvo4r-~CXwfMNVC
z`$;2FiwwUp9l5c&dv{(Uzg^%l^gz@=iL)BX@E<X7-|R?*Kfj3Q7I@6c9p#X%9}9q<
z%&I+_wK_Eq<ry9Mk4PXaojOs<I1lAIY1!l1F@cJ6J>~#))LMLe1V|+nCI9PxX7lu~
z002ugovRp8U%OU55z7PhI`8w|)T^h=)ewDCS5&f=>;*zP{dt#`cP>+nQvBNC+K=I*
zQ6+@mK5aqp;7?`r^949l=6IHV#ai_9l!>*9jcr$F7d`EHMrLMXqXII!-pjWD7{y~c
ztT}Zej@$g(v#TH?+SmTf`rPi%$9#7Nf4Ke*OF*MyT|dxQ?*Tw`Wtl=%|AI&Dp?UvK
z;40dKy?>|ChO<d*x`%p6jta{8a1*CLEdb)=&}pi9FJB;6hiPs|X|1(aD`lyaKDPY4
z=XFs~Q_2}f*xWnr*xizM=@+b-s28Z6Tssz-?%VT8e7onO!G`sbKbJ-K1+Qxl4e%0o
zJ-EybC8I@pgGMW;MnIa&p|y6&C(MIrFXQ9WiZX9EXz-^q?4vAlfN8Mu4J8|DbI(06
z*O_coM7A|tys@FZ?-7b4)B^1j$MNSLLRPe!eLWp((GgGrx@@+GT*BU1@Reqhivk{`
z7ynbeF#ovh<NY9L!+?lrW*Ii76HO@<bYHNqt*%wvZ!iIDaP(g{mId<~1c_8XkNo0G
zayh*Yb5kL8AmW>35H@R_>@;UrO6fs9+6J?r3#_I<i*8W@_~E);VqrkivE0vp<uwkr
z7%w_~KI+(m%X;RatD>Dux<)v_zs>B4{&Rl>rP$H!SsUk=-)#%QH>JvF*3grYgR!~y
z7fxp9+?M_Z8;wnm9qq*&mdTq>*68antd&&aGzF4zG>G@JShHIIS>^`4<2U%v7y?xB
zxkWBixgx(g;O=Sp%F&xW%1P>f4!t`wu%kuoKQTSR>7IsfuW71oi8}`EBd>qj`*wk@
z_$FX_^(9K!H-3-^NUG=E)BHyRTS7TzqO~lg;H5&JC~;~TnLT?y^>c4EI(CpABz1v8
zJydbOZ><$Osr|XpdUbMK=D{D&_cL`$rCF^jDVoMw3+>j_J~p<Ika}~rA7Kh%X><6&
zLkV&zP5{t_X8MKmKi~Hl@mG0TFFz;!1QG1Io(uH?U3qy&*yiFmqIfho{1{6NeJ>md
zREU+FIOON3{}gdfqR$o0^M^~V*E|{|Qo^i+(-OA09&nU$fH9=UD(YKK+N#8%Uw2}-
z&fWa&GW_`A0-;k<hFMIL*RliZMu<|M?%CWEY=b&j_URsTz?7%^Do4nA$6N`0vu(!w
z*2!Q*P06*nXuv;6Y0~tchZJ)qW2g2a@ITKwDFv+Ui3e;9vKJz7o#l1-`ZDm3c7*Y|
zK5S9PvScpW9H+vQw3}kdBlpJfp|W5>j;pZyiY#reC(gN~Y=0W#Z;A$ik%O<(r!}TN
z7)356Q1Y^{E=mduWXeKVG)2yKvLUlVWUB;x(!B~f+NHmAdmRamyxzyjpQ3s3a7~&(
zCf~z<)@J;K*cgJ+Y;tJ53HNxwDoj>o`U)+7R!ZnFJiaQxZ+4J{(K*;l*+6iRC8R;Z
z!s#kD<)FpCr9uJ%g@4?EpPt-njmD3lGB%b`B41(Mel#Pj^xDben5!U9<`0281<o+?
zY&~l>RI~K_U0@jAcFBxj<u4(nSbT%7wtP^?s~?4wUSz6CD-}|5NhAB;u?c#A8uCwW
zr2Z3Fg=$_Ee1m*iDsTA`QhF8@%SFF{<_n+JI)%sRlE{>OoM6ubB33#SK*{`Ma9UTU
zW)qS8BO)C-5u-<5J45PQ{Hzchw4Y+li&OsUYa>!HG5%b0_HyWADVuSjV4J0(;o;2$
z>wX{_zk)naW532XG)5Dt7Gbb?!nkNIkFW7{iPjSwS8GO8mg+xS=N=EXZn4#HEG><p
zqR^=b_9~T+tgzWP7bHbHD-bq=<|Yy;YwNja1nL=nX0u*y<iXeFLzb`Bv}JS^P_lO&
zYP08?&kl5O)YQl>D$l3b+rOr(Q(zMzkjPcA`I_Q9y{+YlG_u?4iVf!FZwU!y=4vCK
zFRX+<=<K}z%A?28a}7b4P`<h*<^^|;nOECD-Sl$<rQh2&rS<;t>i3TKpZCnuG(;Cd
z2VP99sKkc9ikiD4J|x`g@nsYC+>>w@RAj=M#Yf0YxD5RyYGR_d&Z;oQbcnCvEU%Z3
zo0n#r6zajlvike-mO~3v4;q`-?UeK$nvwiX@#*;K%4tLE@u<bo{!HYY%*<E2IX^cU
z*rb{F!}mcSH*BkQR<^Nfo%>t~)~XJYrwiHPVW?zV(Z#UnWwO~ri;>Lsi}$pvkGE@!
z_ccflc_tyT2?Dj>78b+2D-NfPI5%K!O@6(>X1S654?ls|)P>W(p56_&#w|M2+}ud<
z5G1K`lIKk=gL1=Qw2<H%`Wq=aaqH~U>EWS~4s=|?qq}-SfW`9teideH7?C>XR2kS_
z?)k~;%&um0OzcX#9(1`nW1Q8Ld**II3Sb#{_HSQuK#h{ZVm?XjalCa?4=m$k0|a(H
zZCNXXHQM^;GC6<#HWWL(Ve*;S&^m!~cQH&t{l}OVd```gX7FSb=SW|Tw22(z)8zx-
zR&woG769j{k28dtt@|Dsl{)cai^=w~xr2mgd=#QM9!5%t#)NXH6Zd4{yI$JP=B5W#
zmXlYTaVGr|fz@|5)CjWP*}wef=e#7{4EA6rg56FqZrGc{zql-9Kd@{kfwX|Ll3ldP
zlkY|GYlsb;y&qB_pH2C~(qXAOgNFkJ&RH`A8=15+$=LI!TM6%%w=|n^X8rM=btP}S
z8Xp-13D8ZV#A|T-G6k;+Dx~?4nKOZPj&EWA=_kqfPipaJD`~nR>NpjSBXoSI1;>AA
zy44!*D-XK_5(L0Yi7w;_-kF{Kg}EzI`D3?-JQk6Tacf1v)So+$G(UoPr~!ybAWkcM
z(zyVgFX&O_1X!e9$n_+Hz?6uJ<FEViDdj3bD}Sjx{FSZlB(<$#9ZR?X{p8q$7I~Oc
z;INcB`^>=)xzk4_TmHG-qUfmb({(A`dnX;%R24`V=8X-){(7k3fp2XF1mdpL;Je^u
z0)Zdge+vmBpP30kqa$ZASsPwydb2H33~4}SXmhW<wg8!Dzh8+?UrVZM$9qj5VSR;a
zOrwImm0xXlj(m$Fq67*g^WKoMjz)j7OHattC->_ZuJ&ti*QP!G6j&&N)vbNw7x5tM
zO!Rd7^6V2FGk3J360S3q_Yg#%FWR)+(yX^5NI@-V{>d$>?CO^Mv9#z2)L?cJ=5u&q
zwmmIRwrgw2Ha5OotccsW2$0Hc?^eB$Lmz@H>vpcCQSm$i?&I$4uc}<^gcN%J6XrWa
zKKKKn#n0yeeRJa+1Nm(vwun<))fdr9>J&EJ@ZP$zdqsW}AMn8O9&zH~-?TGe>5&fm
zZ-<MNJ;zVe?ajXAl{dcc6K|cwPreX#0`drnPsz_6Guz#_vw=^CI>66HqYfS&GaX`*
zaA=X$lXpCkI@Hzvwn9&JqmSFoC{ZTNO9Jy8VeviCYk4k;8#I~ZZhw9MuHX&$BNNoI
z71AnCnvN;;A+BAv_aB>kuikp$QNjG1flMpJub&A*E;p7D4O4-JV`e9V^!hl**R%*B
zx@SV>W=)_$DkZ70_Qo0FV=0SUS$Pft(`$bC@F=(2Z?~Sc%%_!yLL&uIQRd!$wSJQd
zF8eptPxG5<#<&RfFLoPNW9OKoMaJyzXN6--22m9(vr1lEQSXM%_Qw;s)G;;uV~mnc
zRZCmBveDBfDsd$~>ogr*%$x0ANonNHSdwEmigo2*?Nl?p91sbe*}*lvTjLLR4)90^
zG3oDU9-}73!pK+rb)SR|6_s(9yFwAuiUhN0^sQF6?61#fM;dKLOXSwpdgs162dN&v
zZZo~h#%95fxhZvT={}46L~qjuxq8+;4i(Hj0#(JYHtlseh@VhB=$YFKPuimzL^nC~
zbk34h&>}79w3r?_^!C)*it#3o?P)i0Lq%VN@1{y41rX><*(s6hfv@X)r}szT*YjTp
z$ZpI;!c~k%eAAy|Ko?N6!nl;m2VRa4jLE3|^$<G~O#Vbf8qth9jT97z;JTaXU;KQE
zkmvZe)||1<AuaRpbf&n7<&2LvB^0f$JRa>x)i1Pi`%%(!?}G6v_Qbj{4btfL*^U)G
zM<T*AO@N>E6G1~{KfhtN<YW-?=QmVDyGOM%RctAH5HhHi7VVXPx6HAFxlUwFmqo8f
zef@gF%I9Sp>_hld%wDF+t~b8Ro*C}8850izoh@#odESYKJHNO?rL+`|)^HtlW4KMd
zV#RwH<{ea%?<6hD-0EiRMX@S=fliA_Y|^%tLVxWA<;;xmLWWPb+wftf;e`I1nnwp)
z%3Bw|xiK{qW^LRe6y}fYjEF4qU&caKY$$r{xfe+pdJxyF7{K||w}4R~3GArHt(bJJ
z;{<FHyRpDI+Zs&%nA#aO*rw^{BdRMoGf2KSn-ken6c*oL#dRJ&*LoF8tVNG>a6E0p
z-haE}rp~uvL{2&b;+VmvjcvzNXCb`0(|bENqo<gWtS?t8mOI)LYRu)zxIJc<CqUuI
z;Z&z+d2|k@`R(0%WN}`2;ay9nQvaW6-%_sPMAnx0*o11`KU}|pC5rW{pB8RiE2dK&
zyoV)j?%cP<sujFLbHl<=<u&v4XrN8zzOgS$FNRb}jwv{pF({C8rUFwPFUI!sb|!7y
zJO9|<zm*cpe*!2)D%Rb$nxHwEE+C0iKFn9i$dLVl!b#ldBb%dP;S&xX2Vrs-)KhEa
z)iae+)f^wcMy)JQi;I=KYF9kCIGR;w)nv5>g9V)~MFq<hA^fn|edmdga5R^jubY9D
zIi$i0Z14Q@j`f0%iLK=#ISqkWRvx>cb<!g`FywH7sR603?pw}Qa`I6z=2n331L(Ht
znfP+>Ju`j+W(nxSZ+HK_$)Qt(pPASCEniK;-uZc)0Pbo^O9S8hBL4qCR?3mmtJ&QB
zTOkLo+jYyq3hY2x4tU2>(c!d?U8JrC>GpG6#3UhQX>vs??o0_vfetMtw<hzzb6nE2
z-o^!dvXpATnXSBWc|T8w?ze!(l??8Y1$2#;4))?gnwt~eBX;^@-{<XBTK~~Y2Qor!
zPP~5PN73-Sprz?<nnGrM>Rj2QM_?|Exu+cF@vQ7}>Zt(1ucK#21~gB7D`cvd>BMK2
zDqBQR&7vSw_($C63i1GV7#o_;&6gSr5tl!z??195b91xycJ_6K)T-Q#O2Sa<;`*+o
z*q0W<<dw98lcEqk_JOJLkWhqz7#{H2y`V;7F-^yX(|Gk#|K1bEW*cxbzi9{ZJ&K_p
zGp*ixi<>*3Lp$6mSzWt<rN{~f1#>0;_OXFdP{<@v(h8_Hx_JiUxBd1p*ueSyV*_l+
zMGMVtv5YLbO;7gK&Z$O=ZS`Nx9~2Jo!$00lk`B^n|G(SUOpoqw5RCh+oJIQ=)nj}7
zqR-${-QL?-hNGOzrf|X-V@gKI5b!E0#x3$oVB5@u$5h)b<fUZ_OqJ_AzOxqWMY}5d
zcW?0IeDqk^HBzfjFT%WtbwHc<?>IC7I#pMxbG#Mmn~l6}*faEuE5|yhAIrm|?}7`Q
zUF!+qfZX^7fOT+n*xb&yLsoe)$cy(W$7K3R`MGg(?As15Okn!jW+kh?Z%oS&3Lf?O
zSPPQx#co5wqs<tJ)#}>~IRC}c8J_h7KK+>XDFVjT&NN*7JCDg*U&FE6OGFvu`z9G=
zoLcYB+?%JE5<q1>XYr}9_sY+b3X+59#tdz8kAmDK$mrQ`c)u^dwXi(&;aA|gyf<h_
z;hdAoAg`G3sJV#r`j_*C>`M1}&@|wJR{Z841G*U0#;Og;8nV@8HBPs(PpV8fuQ{0a
zpQzF0j*guI^6a-|cM|zI-u=Cc&59M<@vJ;~x2Ki{wz2FANi;XhHP|>rLuM87^uZ1<
zwzihss_b=1Bv}XGJB(Jw)QX3ps(voK0k|WK!@OFvOj?D3C?g6r-kse_tQwb19w1+w
z+<o>C^46AD$KU_pgQB%>&eX~*<REF1b+8vdPTXM)G@e>IwO$zD&Yt`2BVqD0<^}Ew
z@>=mlIntWdDReYb9ZIE;9qpbbS|qb5yWBh#_|{wWJwp_W!k%7@``Th3zN;|PW0s>}
zG9oqkG8Q%{+c{Z4i!RZ-8}8m8|GLn<n%fYD+>jXp-5e@pW%G;Z=5kS`gSzeiSoM!I
zLtI9hCQeGZ7rL(HIc+5f5XM)Stcz6}Q_fe$^Cxtb2&YXO>6HXe1;(Y)PrklaJ*<3Q
z8iL$1qD61%y!KCCFt%97mDS7s#gLsZLf*N#y1Q%dSXnI__**goZ}Wr|X53c{b32e`
z8$c^`;TL)p)K1XLk1h%SN(g<~DOYe0w{XZZtyX1nJvk5}@dJqD#*X!*U48b!*6LZW
z4@04NV-66!pkm9%Zoyu*K@T?br2-xbXYg1mAg52M=O9bCdA7$I%>K`g<xzG_>^(~q
z3#PXz3U%Pg4aM!9d9UYSRtTDN{s(b2)MJLA-1j@GQ7D&8DQK(V7|NWn-WYdb+RZ8^
zT4HW2$C{=KBH}d3pmL2Ej+&PCSCt^TZyI3LP|6lcDN})E-p7Jaa<}H6VjKhbeV}YL
zdIjU`Ly{$=+8>oWp+jKHZvEb}$)Epq*(I{B6L9~7kPJclrlI~uBa8g{0Ek?Fg*PFZ
zwIEFYP^;THjoCpLs)>1J?r^lB<=Bx~o^rJ3`9m+DG!`c=OM;-{NjP406E{)ACY|Qt
zjl0d>Vw5sgVy}5BT_KrS-)>f{+;U6U-->%4c#NH$-<n5AwD}OpM<4%RBV`Uf=cz06
z-f%@E+7IJv4`7sM^y+0>sCPxq+MsI_ij*%4@&IvLP)oIY|99^$j}1^#zlbWXXifGI
zB&a+}^$8Dc%du<@WlG()Kr9$*Zdx6g5b%BQgHN3lH$_O7H_5F24z!Nrtg6fZXq`yl
zIdnzh`b)Oxow}0q+8RDyhrnmhbpPz0MS)BD)Zc(}Y#9hg^iQcS9JD;4817++7f<+A
zE$9uXgKUDU!TG&w0D&>UL8D>?-p=`z1tFknLCv=o<rE5N_o}Q|+@BoTj9U}n_d7T8
zdc1~BzaF5kbGE4Z*Lq&HZ!3DQ<|C7;;mkOZ5fjsh;eXe1X9Op?Q|qBP<n~IU8($uB
zt0vaz1RPv!@n5mHcv$bmJ7E}S5W;wa7<ma~J=%JG2CK{luHuf5{?iNKW3cm^0PmtQ
zON6<O2c^k%7nh4S;8y#8o*I8VKgT{+z-<~`K{D7<rA-rYPV^>im4d?IuxH0FM9Oyz
z`}$4s=j(xMOKGg9Z)FAzGS@*Ly|$|o*CD)kC*RY55^HSQfa@G`;%dNE1dYq&550$l
zLDg^7O$ClFYxXHeaSH3{BoIcoBZnfp^ss7Z*~C`InE83Q0J9G0Z3q6iA{#i2bG932
zX=LKDXQUNA;^N&8!p?GrPa0-O4pDGQ`{U805H~ZL;-hfOry%UoixyCvSs*Sy1N~E+
zh}0V80xT)-LV8%8`r&%ye79XZt($|G*-?(-B8;8TK~EokZOW_wqe+KeOc{!Ft3aI?
zPV{=e{Qx9AE=roBJ~@7EI|6y<H{F8D+DNp#H9LyRkQrsDL%p{j8H=9^+;#kQ#I?$E
zGt$*)tUJGAuZvQ%mg!M!Jzy?30j&&`qYRrI@9VzFvoj-c{Y!`+(Ijw%xwi5S-j8I|
zfYWj**6|H>Ogq}&;y+5w?<9_g?*%Z{E#A*^?zd&G%Ob)*=kM;3rC|Lhxq^|-&JN#3
z_Fnu)6L!|BffN4if2F^8tdt2z6bh>{o)F%yyAy*hx^x*Me)hoPimMsT#ba<wN=Vwf
z*Tr1=>*ow3lBSEI`9?g3iH1Y1&JmawG}s<((UhsR_-AMnCW3|JcZ@_b`Y`|Nz5kk~
z44uzHvTzD}5UM!2bFlp3BZ-SOh~rYERYYn_7Buau4Q}S@|24^(v4QnTpK&~h%S%&E
zYrQ%BIT6_n+C(!7jC$qSe=Cy|(5aLu^)k(LJAk#Hb8<}!lQqG)`?@trhj4H#2=P<6
z^RdypU9Jl*$yw*8zLq`3Boro01r~UB$ak|h;53f4FE6(~!F}0!2h(qZ<p;EDoLccO
z|F3_o<@;xsu)v49K;PPms4~t8Z(A^eU4qQcj;Of)viyQf<FA_$-g@so%l+~-mhco@
zl*KXa8_YOcPcVYT6g5opt<m({6pix2b!krH7Lh?KQJB2;{W~7JwU@!3aU%J2;+Ly?
zMYZkQK6+f!&+=%u^eU^xkDil~_lkSYu};2ynzSn5(n=(sqk+r~t*}tqeIBOmp6-6%
zmHTTtvn}9!?XCMkN_N?E@8TbVdJ9N^6G<kIC!L@t1L)5=zq{(`$=yhgJI$xVSc~w~
zdGGo4A}Q-j!i5C$F|$pKOrc{z)jev+JMTzfF`aOUcDobzVX){yj_fdnXOl+jl^X=e
z3E7P!Y#(t0``zFj;7*i;w*Sz{OS(uUbsy<)fRCYxnzT+1N_=aYzC&jm3I2)$$Ht~P
z)Z|9swQf{XQI~>Ia5F%$Gnf6~BGuU~{nl}3P4q~eW18(z%~oH6#VL4{?Wq<txiWS|
zyU=0J)c~8Kquvhh@-+OBeqpPya#S}Ews7XS)S23dSP~}Rk09WpwE|;2CWsu=mf{ck
z8RHsQl-%1H^zc0cdPo#>Oa5-~%OEd5DofNoRtuU&2@d(3W#Qt+RN2puv1!LAWWFl9
z|7sj5;8;$0=u%j7xKb4YEuF~NN+OMBQ29ZEUr?fYOnCX_x*rwZU}X4`FbF)g6vAt4
zzllv}Y!IkUY*mH(8MN@sdif9j#0$?Us9mQ%H+5bkYzFLwCR^-YUsVB_3FdMFKR!H3
z!{Z`^B@vEpe9w%f*Wolr!NI8P(O?Izmc=wnnTkq=iHqeRePi^()><lBfEN<_0`+C`
zV!&OyfHqLqO-yl+D>iUU6M;ZmgGy+p2w6ul#PxeMWQ5B#@)@l~DowUt&P<Q}4?L~D
zDi<np_P6g`X+JB%^-p_%ShE&OKW7d4c~4qu^HjEgqfjYxm*}n2{+xbp<yrKDGUkms
zi60*B;5w<XC4lB_x57SLv$i}gPhHRleEBtnaDyJB77X*tM*K&pSdBrejOzpQK(`mU
zb5wIXUHAsBIf~XcNlWv(t#)kq95v=;B1Dy0VWa-iW}u^)sDLLACa@rB5y#FJYjc?Y
zU0|g{p|UPx-D}eY$bns9H&eZopfR*$uZcMeBRjJ;X6k`M%k|_o(9HE>SV3<eL6sJ{
zX;l#H4MuoAf@;0e*co3!FY$H%66{B`tzUD<S;$3BLmvTkKdi@;S_-~KlGhknvro88
zjZH8_&_gPkDrWK{Q?~p|0N?TW1%GL2r3a7H&l7`wX$}PP?dv+BaliZvfDNo9sWYrR
z4BFbbnFOfc%;Jdcxh5LH#n}dxdd(w#;ote0>v=R6=Eh;_mOM~A^X9P%^JE28mav)@
zG_LYc<!A<*|IwhVBE4dPk~x-E=xrh_#){Zg)E2rI^Av-C2a3m#Uq57gX$eY91c5Tp
z`IR<V2A-&t!T7J%8Y%pxXB~(367a#{gVF&APwoldu~*x4lTY?CYAsj<#$)GJWrb2a
z%o0V4)Lou-@_#p+tI~w<L)xs(zL?R?>_w-ZS=IEidRv$?ci#lXv~gL|RW`psDjKzB
za(p(44|a$u3Tvnlwn{Nq?^*2S7p@ZU%0onm-}faiZ(!Q2Pki5v$;{p;$Gku_2qtXo
zWzrw&L0!phv#Y@<(ckAqUW@C*2da<h_*kawxHC5R7AL)`YBM0J*<>_|2%o=m)?;QT
z@wHBkbPk@>;LE+ifvjfFPm6d9*?S>YR}Q9N$XWSYH?U4uGVs63zT{ksHGxJqjIAvl
zESs>5HMZtnpouEcsJ=If(ee!9ZC`f0dhw_q%Kg>_83;l<$%G33<;l*cA-QOAe)&TX
z?1lJK9E;+!V%d;1P63WO+pA)0H)GK5r9d&c`sAuF?yk&xv&BJ9&T)Zk?+dr5a`&Qa
zB%a*og6V?F(u)`uH?1a)n(89y^m?>g3=LL_`(Cm8dY!(MAD-4^1ow8~CJ^rYW;gG_
zRoWaDC>m)_=JcBl?H!y_&NoZ0VAQbBsFO^R-ZvMJUam#EB!^FFS^EX6f;O~PUuBP=
zaxc=NPHSOj%&o*XrOCy=i`K+KF#U1u#fWtDT!`y7X-Re;7x0RyhGy$VTr3SnMC8?B
z`cRsM3v2T_Pj;3$u7~EdGgs0;lo+MXhYCD-5@pb5%r+@f@qR1$SD<xJj9zoJco{<<
zBx*bE$Zg@mN)Nw}sS{9D{U<Ehp&J{kd~8L}GPvoc5D0Lx=cl0;{ak9gi#U76+B@pX
zRpB<%H)b(7i;H(EVb%xLoi?<f{(El+mrIRRYTQW7MOyCc=XFHgJ^A<`jeBFUynJRI
z`C6H3bHz7&aR7~q2>_#{s4x2|SDITXeL2GWag!J=a;#oDZdUiq>w$Mvk=Ns{qk4q=
zv4Rhz+u&=`E2X(PNK0kp@&K9{#Rd^nf##i`ya`x!B9A87X}pT~iT*`iG)D`DL;4!9
zN~4L@GWq1$LV5)rn$D?V&*vTkRYPLcPghn-H-@S6GJrct&@#+EJtNK^L!H3|b0CF-
z^D0z{wc=h}zqtXd%B;Dr{IQhZ!&6*%cwvP|i!L^0Mv~E~ODqUKOv@E=?rts;b=w4P
z4}m|ZOOM3(l3}MO>uTAvFxOKdd)__!9^^(!Gu4Ci@~?*IMg4aDfae2UxnNF1?gfeM
zelKCmhb532Q=i-?+vlSamrnm`w<<QJRqBit+|Ua4R#V4vs_9^}fCf@;*D|d>oiUHK
zk;cB9Z3yt_=jd9{OxmHWaUgS4rRb_ucCx?dz+}QK*U+7MW$u)O*c@H0rt*V()Z$wi
z5q=J2Y&~wtY&*vbmx1MzGV++H2~yvB=Xg+?8vcs|cXhxQ6gw17zwxl!!L=Q^S}6vx
zw5{{demQ@6jO5rvAlDp4Yy_J2LMv_Ikh{PT(<U)3z2R$1xUjwO?CW5n>Gu%TORI(=
zKfZmajt$;bBmfPKn)ZP96_*=M*~pFV^y#X^&C)ND5tm=p2ZXJ>GGm+BpDSGk7Rfd(
zb|-}&GGzlFwn&u%0U>EM=j367M!X_6uG}VH5#D%$|J}E@<Eln31<k>Bn_SXKUrqRJ
znDe4<pX~>S*sHWRcKNJ6V|wuX5TK>Hl~cMp?(j<9^c)bgwA-ovbl$V5;_M1#R3|Us
zhA~o4F}6n1O{q$dG>^JiqBm8;eqy)CXc^+z#;%)Nvlcni!-ADxFjfQld=?CKP&`D%
z-1c<lp}fVfV7v{~c-!eGmdL%1Eb6~fp|sRi{OKL*X|$95vUkto9z5CUTujLA7NNq6
zSSE4xaCp)~E^>$OkN+q&gP%CWOF#G;$C_%3G=@s~z0}I6Y~VD`ylt1R$2Hvw;;?Ma
zO0#rGca-<Kpg!BI&44e!=H#91<VB$wHlT<seATN*S{j_$rp@@#I~U_YANu!hRqG(q
z7Tsx?@&pmoSvqukp;ZL20}|A;XMgysS+OpgWzv0rj~inI#Lo!OAX7erhiZQY;4}^K
z!`@aI6(O!_ObPv-^eLc>m5K3mq)nu4KRgn0dx}`=dWQJ9z5}Aed-qTXV3+)o2-}Ve
zc7xvaO-QLT5QL%J-pwmp{kcfydK!q=v2Vue3apSAnw56Kq2oaORQKA$loksG8*Y9$
zqfV1*tMe9BdP3i1Rt)2&*@R(!tT7LB+zw_k{G#w42d}i)bZ4|L!o;^4%S5j|xR9F2
zd$YkeQry@Pc5k{-!bK{dlv=X(ZOGx!l*nj|SRDdOAMJ-rGVR@eewFDZ>Kt}bc{zoC
zDlo;`TT^dfE1*=gv3ciL)`bHN@k3DO<|3IF-yRMNSlLQPY;T`=u#Mhpd|;o^)Oyj3
zMu+ts%Cq^v{?MS5d3Yyi3MJ42&!F&05?unQC9N<}<jHai0h#o0sDBB+)<W&q<UhN=
zIcfMB#rc6HM(%ja$Cw1MlFK`G)ih#`ys5BO2dcdIUf&fow_4HVOTMR-3eD7Qx@60u
zHdU(;ep435B#YT3W2RJ;>=>Tu$<6MF1EQ3}NKoO0wxY!^${m+_jJZ~W3yk(7rzw0w
zc(xxvGAV76<10-Ypy{?CC(@kK(oMH|<M%v60w=8$7~#wA1`eTCmxVN|%@Vu+Dw7(c
zuf0<2B&_{8zyCT)S12rG3lZ>`32t&jb!?~cJAbm=&_Sl9;4K(K!Y>S@{seeZD9=$8
zSJIuD3Mv?F=2qi&!10rzS{5=VomMJ0Dxm{2+*Id6nBcKqtLIx~08h4XXTNwZhCy>1
z5orcdOHbG_*`zw97EGzl?&e6A-wT-A+0}5fISuout#;g9U~>syh4u{+3=pUj!N}E8
z;EL&JR~+Qi9tQ5ao}MHQ-Ww<4qPHGlA^xDstGHjp=b%*!K5mICjZ|z0?@&hkoc%KJ
z+ZNC;L=blfTh%KGTLPW@AUJvTsO`CPpU?<drf5e*mp+jg*@0zdtw-gM;L8X0&{H~t
ze2}(!*6jHWo3qZ6t)n};aYg5nq0-BH9--$46Hw_)T;Ws`mzdr(JJg@9OB0P<JDg2x
zIQj8jDv)34ifIzz9t%QaSNDR98bnmDyHBh8E2dAW?Ahf-q%M~Jt<)S2f7Nje+$c@7
zCO<DruP2<tt25QdJ_V2CtxNF0WUYaqE314Q8|-}wOcOUZPID3Z6nR>(dOZ*Di)+^d
z+KY#nvP)#6zm3IguPS~*C&78&E2gF0C#qp~T0{5j9?5vmw2tYx1k4{~>b(5u1gUS?
z3eonr7=9B&tye@Ygc_nXN^#2W>z3zHZlFIo!?Vl3$@oocC4JGIoSG~7Ur2Ra$<+xq
zCm3W0bv(}-)h(+22xZ)DXg(g~aM&pH0#n$qj3ENW>1<-m-rB~$E0C;tFlQ+XWXV|M
z`ozrzdg<ou8AA3Zfyj1$F@7qG-|eDFNXOXkNq`5^=IHQa+kMx`d3;22A$$9Xbe}=I
z?}v(S{V}!%)3swtZ1o7CMkT^e$brN)niZqG#s`bO{y=q3QL{VWlr4C}<a~qxb780}
z=>@!jxY7LTJ~q+dysHIG$K1QEHZMbySFBUxG{%V@*N~W;&BLX;Mf-%6?@vqTKYpM0
z=x(dr*!bza^=O#QzpbpuZxMZ)u}&tf^E)TqmDMx@L|Xa{LOc~dc|~@QQpzWUL<rPY
zvO|d>E&g&qZG};$W*^T0mlZdi+%Q9#Yik0t*xz=zUBaQuE!bE+8^1!b-z9dzLbz01
zfBN;nlo2+kcKnZ=mN}@rr1o<TC6v_uRNkk7IVfYjUYY=;o>qb7O=rj<tF4M&B!8;Q
zc-8ssPT@ck@0$DWjSosI<RO&DI!CMRuNi;ZCGgTR#cOdNMMA+~+ESw?{?65!j0fJe
ze$bnXH7Vg{3WYIW7f-d$HLD2R1$wzUM`&M_3UAWAkk_X!13?CNYyu5g<9-mLSwkE%
z-I$8aFbqBjNU<I#d|=OiP4bt+dtJ+^E4*7Dkoi}?7t^6uqaWmLP+);NHe+jDKhr$%
zwL09W@j1OdkRZXW9k({FYEyr(0XYy_G;${v{I;WlDq|{$yvgrW#v9Zv?jnC)r<E%M
zzVahZlf(dbR((<atB^O^Sd5L>leef!dj0pkjq;R0)^HN};HUMhs={F;;c%`x0AZ)w
zu<DthTIH71Ub*h9pxE*gi^x`ny4QMP^?q+7m;5aVX9ync7q@6vWWtRzzt!UJzbzsI
z^l6b7X2JQ1P)l5b=g(TG;*0nu20zsF=GxZR%zRvVZy4$J?e+CKbmVsU+I5vi&y!3)
z`zek=i7eGO_dNWr+HQjQ<>k9%;`91_yDwWoTS+_jC->Ny>Mi}hQbU~2DWtyr=fP&r
zKDjO5>sQV3BcHAHLNIH_gBc|oef35DOQMv>Cnf?@1rLqh)wvqv0;F-b$HNyp3*)tB
zElJ@P^2fxtl2KB;u?Y^}5R7$qb;1PqGpO+5qq;ChD#K9+U1bNmJqc@agJV-5`P)$p
zIZAU|PbyKM{zRiSw+I;W+NN}8=bf5U0)uv+?gYh;z2Tfgl~^>&_LZx=I%qff+!}KI
zuHod9r*UWcdPYDFYC1H$T!%GJuSWh>i`VWKD*V%}BranqZ#bogT6{rEY*^KGvyj6O
zJ#tmduHh*~XiU=NJx$6op&M$d?*pG}_>ZEw!!4*RxQ!6Fr11bR;++Bq@Sw@nZWSBy
zANsZ*?M-qnN_B44fG#;ZmQRZXBzXJEzin%@YvH&H3jQ;~m$U)MyzwvV+Mk&A#}l_T
z6#CBG6s(;4;Yk@Qeodv5%RWag5?DqkMs25Cw5%~(MamKtv2dRg7nax=@=zOAIbYbv
z!0X1Zo=8Cp2_%=cJ=%oSXKv2QLFP{5dxVQ8hN6-(A&}tvoQA7opco}CR5!;*C+X+P
z3&+)-v|XH}dRAbh*9_y0O1O?q**l*tw4YZdlhkdJ0b8(K%#%QvWZG|r1|o(d%E=q|
zc*&RI=-LXs<Rb#X_}0qjv|5*V@eI(_M?gGpS;nejQ)|XzzNH!Xs6C0~p_>d`yKbs=
zlqjm_*TK1NVGJQX*Osi72IAPrFiT1inK*zp2ZLG(TOo{)xh-d4G72e07XY&%a~@c%
zWxu7X7v6$!2EXSt@`Q^BSzuGCWnR?v?{@<&JH&d#R%rW$i`7mIJ%0Vsb6Rw}YtJP-
zpfBnzmo6~O3}Q#IP83VMUYD=I1~H4A0}?s?WKchHj5*znQ6>8Z-;|82PlZ!mdjD8t
zi|)ns!MZkyL@iC;>lWiPpJ|;6Uv^#HJyq}^MOZ>oh|Ny%p=|e?_PhZX`Toy&fnD|i
zvO=XX+S<pmyYf2nI?hlXA)K_~cc-OhK=30{TNAMJ1ksz%BSa1uV_RwX^=J=p2U)gT
zMrX_*!;E5{Zh7l%xhVNV@lyjg^ea^gb1<Q*1G)vjs>}O1_8#8dCl{Y0<`eJK7(h<^
zTrX>DiS$!34T03$rQ3?wtIX*xL(U>vVA5anh0}t_Z$Ju-^7Ha!4(&%><0UW?WWdN=
z54lA5j6Ing6FLgv-2c$Oa%cdZ$V?sZ?7YTOz^1)Qtt6equjtU)i&wFHEiZB;BQ=sV
z1kK7_=DJ0nZN~e0uNkp?Cs{WxI0&U3h`1ej3TnaI^pK;wI70De^WYRRg!pdG^Hpx@
zT~T!?0k#3JJim!^>O)kD6V;#lLSlA(4n6T58Bt`Z(K2+7eV_<cNKG=uWf836tZo;|
z64Lkonuc=pM{RLbEYZeu)b*feum7IoiWJHlKXBkZ0(txN$jEEm!ka}L`m-2*xsze}
z4lC-1=^ChM<(faykDSII7Qbi6aW^}nVAZtLgz4i!o4{R$L2(6v5uY8j>RxlB=-wvg
zdIhA2!vKcZ%VjmOoa8bHfjZ7D-bjAnS+JkF_fF{$if^}*;AvEYv`h%mE1S1k^`?e|
z?c%P=TShq-Ms2gjptNtZkC{UivZ=A!`#gj!jUC)OQXch681c(4&+m;_OlOk+H6^U|
z`yv^AT8)ea-9mzN_IMW+)zmqL$A`dWU#=VxKCe8!$bBDdZupuF;wmn%dZ*Xf5w86w
zd4kJXT1^ejAF~o-odb(bZ*D#pks<}%7=&wd&;{3amzw+Ujbu8zt!5|$gtEL7VL&c^
zw6a;qUyC~E3~MX7r8KPO?k5Ogpuw0rg_C!1>3WwXqqI$E2aoS?7!lRtNpTo~=D%Kh
zdwoLnN+dTJ?xL*wU4`H_|1-mDb80b%f}h!6orC%WnAhQ`g&qe?+<bA<<?FfY9!G)-
zrD9-ZCmHJ+%?`8o##8x)!)s~B9;M~Zm)%|y9$3c>UWM-<12@|d@8c%qZXqB?PG#f4
z8}iB|5Oqq~w1{&cR1!OjjpUuvO9a@)>@8R?5qB&!W2rxEKFwJl;|YI)BfbhX412vE
zF_3OXvywA*TD(jIBXto{_+U7#SDkZJ;BS22ky+VeHVX>wV6sG)@LEG4Iq7TV{2q!_
z5WC}tuqVG<W)`$V3@&?PWU@nB78BY9@fMZ`-MNbPCeJ>VW#dynmZ8BgBz=|6&F_&*
z9m&sais0^EOdo^h&CLnyAw`JcqP@hnnKZq2H<FZ_J}W=Od3BAyt;4O%cBk{a(l7?R
zU*KMcxP(QbffaP%!j}5P$-Qv7SLoFnNWD3};1IZFO$zI)Zo^lwX4}V#zH-d<TI3It
zngjN#3gzp1<#*y|KUowoJG}a9^hfZP0o~>Jjr~KElh8598mc{ZNWeky804b+X$GWk
zzyvdkr1aIL;;`>8<+k!4Sf6?t@Sq`NRf10Kcu+Xcwaob@HZG&9M!Wf=5~2F=xmZ!!
zmH4Ds=jsDgY7Kr%)Py7wCk0tub6)SvMJd?Xc>C(J!auw;M0pv^EvC75)ffd{+_7Dg
z#@A_0fwfzGeKB!<qwL<bU7v3wyC0AHMn2@(%jz{#E;!#v3M1bbwtVyt$%OHwBI~cc
z%1*|5!$M9oLoQ{gF1S-eIz9u2VJ;>bT{ebXJQ+m48-Wen_(nX*jur_ia#8abj|#WB
zqsOtL4&l0g;wk2O9)!KuWi*TmPfPfGaKXCB!pLAVn*yC?@>+YVfNuMO!{-NpDycCm
zxhk8#hIoo=D}5_m&R$+tqTiEQ<TPNL)biQCxFf|gLmbO&+tpa7BT*SNO~$nO9PvYQ
ztB54hcd?7xWpEkCE=l?DG?HS-O2v2LH6B_F0Z=XN|1~k{_(4oQCt^qr0vXtPX2S$I
z7bU$-a^RJQC{=<~O;PF~3?dx-sB!0NTa}-puw4JLw-VXjsp!Mu2H$sI^B%dHA63h=
zbb(B|r|HUO=x3Pw51ldOWo;ylNnKMXI_bf85PrqgnagYtp;gR|-A>eb6Vgs5K-DnY
zZ40Wc8_zbqM(vSOccV-%?mWI_>y0V8BLf*8MIRk8V+Y*n&#!&VIJWA_C(?=X6|uZr
z;~j;~hBNj&hNg4oX3)Q9tBkf@@QQ&^Wbg=Az%@4dOeq#P{*S(p<L)QmB9HG$1jV$Y
zS)wleJa%ejCUIKri_gL@a0WbGq6~G7F;}i_Up3^_D{ZCw?y`T6{S$BXYis1-DV`83
zo$<02BzA5VnH-m(SEC{42PyhmT`N_nKK62`u#A3SD*uG!iv+I$Tvr-$Z{r;)NUKNH
z<T$2;k1)#sc@m=bYOS=fzM8k<JcPYE+61D$ProCdQ)ZwZp8U%VFC{e(35r8$Q?L)I
zAIw{na@RN_*0{wX8OpIw4Q{meZ`z=RoG=GgAtfZ>{sE%HFpZlzg;(b~IQ;W^)(|VH
zMeZ)n<J>tC$yh2mu;dz+kI^_x<__btDF17Z?a(>hcNpfiNK2d@&`14tk7Br=)(0L5
zA~##=w*S(iz-sJ1I$jDo_OiQl!dXjdE7<=jHjw`goOHJkXSXjf*kn2$N8$%#&xa51
zUh^3|d<CO{hrPKr?`>YDwy*n(bcZ?5;Rm|CdP)6V#kN-4?^`^Ad2jRa7+ys6(n3CX
zrj`*0g~Hh(`nG`W4$(8#KkEh;%$zuhea~V+-nm>+RiP-q^L?YIr=!<5ofiG7>lyWQ
z50&4oF=+^<SnuMbJ^O>_%mf~TYV_?J_M-pv0+=4(+d*v2^&?Z`))L;6AcB|mMT1=Y
zO)cPM7qV&HKHr$G>AJEfId(OCq&9^nku=1bxWDNADjF6wiiJFK**@rp9&z)%Qj;<?
zL-)itVXwsXYwj@V1W(pqMw0YZMbpje09C;z49fGnf89;;c|B}V%FN36w$Jnmo}2y{
zB)H?01U2NG>}3puDGVE^@wbf+!E29{oiD%vaLLIS;i%evxT}M^uoi?3|1^;!9Y&J_
zrDC7og%cH^$M^73S#w+D1=gGfn&4<+EsY7OJ2MYnd6r;ea1<pjk!ti(&5Jl|H3zrM
zgpR~5BQ0NlNvQ6yQiZ%o8H?$0Bi8u^Ce5tnMXq9ApwC(d2d=gblX$7IV)veSS1+DG
zquc0@w~~Y~m+^m{ki_jBK128OHANgiA$6+7h(JDMGt|<YJNF&ayiu`EBTe;8YGvMY
zg8oCBa;+ru@t>{E6ZC>t8n0;p_cvUmm>y{tq2|HfdUws^iF;C=Vd%Nmb=%=Wk2Yb~
zG@FZZ>0@=Iw%5&M?;CH)iA^k!z_Fyj%}t|~;;&UQ*@p@Z5`!z#lT*pt94f%)>6Os~
zMlDiCvruD9XI$u?;MQi#fVSIJov(rztjgO{`HjK!OEWf=<_r*x2!BRuYz~T5XYu#I
zciVT>^hhQvDlkNvv*<DAcHYyB3Ucl065pkUqj0H1IdzM%f{$+<$&sXsxy_2Ley1c>
z*O;<+mZ5%yTHXd~ye<Cz26#D%n8Ho&4kO|pRuYM^JYds1Nq%rc9(wu$P5tD4M@$e9
zP9yixNvlgHDc>ogX?ofF(d?z3gV`z_*f-J1T@4>L+7I_N7he%ntV416vs<(03@25!
z1;;C)&5;JT*^<7KGDiL~mWTUyL$`iE!KIV-YKSLM$Mo-~BBfjDz$BhQAMCI=<U<7I
zKp4McUTa}t)PCJVhY&+z>M^&~x&~m7i<(%zcDs8=)vN_o<sq0Kg<|10C3+pz!n|5`
z5cp?wt=ua;a*0S9Y3I#qf#W_+FoQ-jZb+)!c5f)DxVqNOH}#IFalNYV1=QK>U+To(
z%w%G?cVd9r^fh|Oo6axw)#c?Lo*FOT)$a`7Vx$>QS8y!nDqHpup3#arQda<-s2kh%
z{o6c#KSXes{&^ldc~mbAF!{HmaLQEh2Y1Q?aDfNd$?nmEKV`5UY%Hx}yTn*z<)J3s
z+SYRe92ftNE|%|kcCqM{x2JqoAVE4v3ZP&yumTTQ677`v{@-$>vaQcjr@d2T@&mUj
zWSHSHpdsQ;=b8p5eAVpkx%vEk_(BPv5`kRMzRgkHhtV={I0PB+Klko9|7!cs_5a7#
zdxkZ=HQmEMdT-LDqk;%hL@5VpA|i?u>C%g+AYBMObQDxXKtQS>(u*KH^eR<B1ZklN
zNKK>#2np{Focnp6`+vQ;uJhrX!-q`vo|!dk*37Q=!K%*UR06H;@x{sJvPfd)tC<E!
z>USRfFBrG|$v1a8$9=_<5Pl^Q7e&O~fH{7fZd%|<sevn{$rP0NMzXm2exI<iDdMMS
zDM<A2p?VV#f5WuP`jD=Lv7Rp8887$1l={-u4=-&lns;95a540f5@*af50(8lK*4Sm
z2*<#sE)?-G<q%oBIt3ax9gRjlEfz9YJ@;=M*4`M~SqqfMz?`l$)?TR40RZJgj=bgZ
ztR%XrMtm5_EOBHt3}OD<6F#1s$#hSi5rj}%X<Z8C$*E>LvPZ_$xYN`50o*-z;ee)9
zUB8lB9=vNF;fR8Q+g5h!x5H<M5=<C_;2e?aab?0UVVjgTM#w~POfTGuPt4&3gaw*s
z?-|j4G4b5H$y-&TsCf>R7j!L;SodPb_nxI*Sy!%b*8PhcJD!f$Irbws%srA8ZQZ|t
zfe$q-@g8c`vKacBf$KdPIf#tAZcQ^vZz_+!h(9`hVuyAWL_xMZ>8@>r8z-XQP^ND0
z`t+`j0#h}O`6SuTkxhwt`X}T+E!gv}{6Xz!o)a~!3yM6+mC%#aM>q9uh>e3?3cF}L
zJ#;ZayNnAVixbfxQ~5d&nTVunUS89S^Zq0REaJ0S-I_HqG?B|oF2Ks!KaWP;)_fP@
zmh?^2+~0h#<Fl0^eYZnq0_>jLYayHCfgyldBr*c<NX1|OsrBgXcV*2gj@It|LiUMN
zII~h}>Fz2n%zmgrCQ0+T_8l~v@frA|n0+a%=O6Pf?{-ci&yvnGTC3-rV`6S$9z`;f
zx9LpTGkL_W=`=88`?j19@e0FLzEADmVdo4s7C0TiR<R_+0vcW2vmh1-3!>aRASge9
zVKOMjni9&AZ2N}IbiWB1bf~wX>T>4&@eXTy;ZDz+A0~FI&v;&X6>+Nq<bRQ9s_;Au
zrs+#inUnsa7I)b0=b>w*uJKSnXwW@kbXlhF)?cf8U;FYtJ_YH2dLk*Q**vLgaRVS5
zQ8L<G1?j0W>E@Y4S-iJ1V?dG;$o=F9^oSXU4~9Gh<J=c~a2vAhyjJc~`>N3SX=&MX
zml(u{3&Y_#hXyVma^d?%xMORFKwbeK{>uhqH&e_P9uq;jBBK2oz>kJkyu$dq73ryi
zqw9L=E*p!jUn5GlOtOBa9wD$A<HZ;Cb`JBoUatC=sUx<e^ZN|QH)(eJejL=WaU-sB
zPHbzo)r+qwa^!2TNYM2%NzsY!8}8~C+9R_}9y1159ma^59w#TzLH?;_u*DcF=boR3
zmLaz!z!S0WokM#!m+|fvgF#<J-Wo>)K#Q=eLki<euCDjaQzH6$<U9P7HeM{Btu-1j
zwqQrx`H^X>UYK7Rzivg2OGm`0UJE+0=}j(x{VI!O37yw9sKE?i-OoL<KJNRF3RCz8
z41cj)I}wqts_Z|Fn}PIC_u7W!?i#`~$gTn$kT1#Bahb2}`)FsKjDn6f26&bCneZ|i
zPRgWYP=aM$&>Oevn(S)BQ+0hmzusIUzV!>f`#b5__fEBy<G&$%G|6mhtCH^5^o%T5
zcE8$6JY9;8Ghfdy_+-NNajg<%Gf6NL3eB7<+;F7|^WYP`OlN#p2hrw2DlPF?Brf+%
zaO|N}b#sqBygp?2Io`HLY8`l$WdI=62_O3xF_F4H8O@C57q9ij5kpy5Hmh-LiHt;0
z1f$WJMja3WJIUr?Z14KJO?>E58afz_*#uU%=8_HBfee!|K;sT~`&i-A-meNBJE`X(
zzL7E+$hB5S`2E}PEw7<bkI8)R`10eWWx%71CItLd{ufNw_ydx>QAE`0>DM5=!qi8+
z<$TDc_3lXz{7QY=fi0}<uRoM~g%W}K`U4LlIhSf3W^P7dsNkIRe1<7BgLVgsKysWZ
z+0QD4h2hMgW5MO6OGda?0g@-d5AvP2)ktW)_ZqBmO#GX*QM2g=>SVGrVqG3_B*a<d
z_%9=CTZ(3TCcj1Dq{yNeMy~_&^__ask3}m0|M08Xc~XR;ByG>Ut$uR%0#sJ-Pu=Rb
zwp{omHn7xotjw!N$U!WDP;PEXWQgHh^}~)CplT5D0LL_D>`gv8TeU1^W;D4o&DY(!
zlhm-age?79r&HZIDZEAB^39swn~{8GO#RS*!q;KAW%uvBb3{Jv&O5kYsIdR`uK$5Q
zq_Zbx^?X+!vYO47WO=tNL7IqhaSI2bBD~wVS^XOr0^|Utj}j<-u6dRL>jK>fMZdD;
z2rElu9)UQz&>2d`+CC-VX%G;V6ZhuJ6_!%lGr=bwrdh6|v7v1PA6Rf;+N+r`+OD0@
zx;M^f5O6$y{1+5K@81p#Y0g6B;FW-|S!~~s%tNO@oknLp9SJt<wZ<U+<>MfWnf<YQ
zDHHNpy$X<YtFObV3IhL0iwB`Zi^%-c`z(*KR;lSLvyQC%Gpb_YU(NXqXLs`+i15$b
z=#MvF#VYu+6Bual`Te{Qb+N*U8$i}eb0VlUC|Hn?vBbWRXGJ=M7G?t*i7Ax1fXSVh
zqXAu;`nQxiN$lOFesbLoj|w(w-sxZfxVW|Vv(0~n82viL?FG$R$5SpdrYkn(Ga?Wl
z^PW!qznH1($wy?mv0(ENi28dy3PBc7@CO14$9dd+2X7SdSD;v-O*Q0~R2{`6J0lvK
z()_D0xGGu1OFLod8KA6pATl=j;4Y_3z%N_Em-o}`I4JpM6mx^?!io03lKxz9m1RgS
zGNRT!N^T>>Njl!;{mk>H&nEYLttu~+(Y-0w4>H+-;$v*dT|IVpVn~uI-BQ-%<)RK}
z0CepdWwR$*diP7)P)R=JLnDT{m96=4Lk3o4rL@ro*jL;BW7F^1#Q6;spZVxNn#5=+
z+$GdJqb<0j@11id1WC~1xM<&TIL=Pz9eyDJ5wLn%Xrz|}0zxek6dH>8<jUTqpRdhn
z&DL3M_L#vNgqD>ineHG+B<Fd8&Zcfv+aJzW`Bd;UCaUwBYb~X)c{-nhkMVmVJF+rk
z#(Ij@+4vpoqwBJ=X($G~Ri^A|4)Bmv+r7sU2W~0*u=(#<2-0;_cEXmVa^HYd^Z>s$
zTe%cSPZ|8aVH{8j><+?LIwG~0zLfC3e&Ra-SU9$vd^RwiBC!8D;T36{G*wHHN#r=Q
zu{-w@*#bZr*0@76<yLFu@P3oW?RcAoqGrs=B^?{eze&|H^YgGY;BOz@FQ3NLg+IB*
zySV<<Cd2{E$-~9{E+X*83yAh!#4*`+34U)DVSkmvXu34{kr16ZAM)y%V4d9Nkz+U1
zkerq)Q!va|ThF;5fL^}7u{p2ylrp)^X;`Iq6Cq9nT`Xf|svF%iLu8Ub)Hx$-mXvBJ
zX>L~N4Hrml>V*}m>bN7Oijp^OAF_Yd$Etn(?50}f_Ig13cUs7-LO%y-eihMvvh%sr
z>)q(sY^$8Y7%A+tADb+DW7l1tLNfyEna?s`PbAnJ$2ntust4iV^g%vZD|C<`87p$k
z(z!1w)1_zgXV6id$WofUkm|l-vI<BFT*n9jR~)EKTzTEBiIS-X<wt}E741BQH@L?I
ztT(TB8<5R<#jC|bn3V=0x^RX6b;1sue3S8M@DyfmU!72i11TIF+@094JG#K#$TT69
z@1&@E?DCf&$(=_wEi5b90FC(Y7}I=M*;?Q*qRISF!7TqsPCrgK>N;%^-6h_&{0bZ8
zc^9F{OB10_?^#uq%qgW0%I@)0zjBNTD9wf|)dx|e%DWT=mJ%SQzm5NIPr6B^!^~7!
zILbAMH$cE!<EFT%A4%|StUtxp>h?W(OxKsAy=J)Dr@WY?o$b9tkG&Bd--Y;%)`K@^
zQP4{tySuN+*z2ZDu?t-}DaYvMspo0z^@AbjmfmWZqIM3qKbECYIvvK;2KvPJnHzq$
zS~o}PU(Z^5L@|1#n611}C+{kG5{0zx>%NMoR=!9&WL2w4rUXHEg!8w8@<1MO{{unl
zfCvjiDb6gk@T=bythgxHcP=DRcy1_Uu^Q<?-R$i*{k?pA>UQEl&ZNZOMymTel9bIg
zxSv<T-oKuOHIJ+r%@rh2edu=-xM~q3_SBpyZz=co*jq<Dz;S!DYU^>nblqsL?DE$y
zqcn;^ma~Buou;0Dt0TI}DjPV{!5>U&Y$(-F6mDS16%dK0>aGyulSfWgO-IixW}4iz
zXAVm$=kJ-4&X%-=;uaDWX{+V22d~qxoz+^Xov-JJC1t&`Z87&be3VT`TfHnFbJQxi
zV_1l%&O{<9WIM@bUtwunqf?6}2H|!!?vyPJV!Zs_nQ<NVF1)f5r|X+c>z58_*bf`m
z%x9kqS~CGIw|DW8&7btT^PmoFbL@im#)ahM*<>kF)+ygP^H>%*ORU!BS@JmXaMk^s
z<Juse>d>)%wa`xm{#vS?0TDEE;G&(C<4Z)+HpeGQ8OJZnQ}rSlLQ!r>Uj+U7t0$lU
z>JT^ibm^}j1nWIAE!yxP(DBN~IAZF#2Q`|Jh+}{F_I8ww;Y;Q3Q>>6+i3F;rYqb>I
z7zGJegkqgY-_BuE+VSJNgQXks%B0p$bwA_m6EO{f9<Cd)T&pqW^TcQ>;xRkf3$$*h
z(csB(T4A%smtThQ#)-YL`QsV($L57!Pa;7uVj!HAV^ex7`TUSo@<O&QxFj|g(iCPu
zoKvUg1f@KHgAtP7sFF>n+c#7x7xlikKR8EpFz0kVV8Du6EllG$a~1vEvyMO<n@-z&
zyuTfelJkUVvJ2jn52s=K+hj;*yq{hSHr*2fAuXrTO?E>N7XmgOvQp|05_SYNwr_u1
ziXN-+)W7~7=N?1g4t9Wl+-n(wUs?`R-hAHUAKgH#@)56UWz;<Nwq*B-um4h7nxZm~
zvaX0>AO^S({@16pV-KC*)*oG=Zz<Hr@pHb5?bZTWRmv_G#XVjym@~M(F2a{Oc@_w{
zj%Ko3;ik)#8(Qj3;SNNeox$g+iZfmGBUTpvrIUSAh4)_*XH5Vywi7%}Ug~!{0tTs2
zP2~~GTz%K(JCGy!$MUf5-aa>P#4%mCK%3~fW2TEY6O=)MXDAp<eLo9Bhecmcs^CvE
zr8Dp^do5~{LBH6AviIgK@Dtjn_oNcdrf`lk?I^yKZDw?BC|l6)w?zdlwtGEZX5F>c
z*`R@)d63{1s^8Wo2yRh^UgU(>70j>|l~p3Hh(K?~;7ldvc$FIJ=2(U1FPE3$Rp(}u
zR@9(Kgd{pBz>gZw|I8EIgkA<alE@IJ_y#_tfRbCfI$XH0Id9m?C;lO&a-niji{B3t
zp;et`xRIylNwj#A6qC{HR)B06?0l7cJ`?~MmcHTJKGEfX54j3%!S#qD#<})5-&fgL
z8dc${OTrp+m6DWjY1%6mztuyT7{bOVJn;r-9)CRLBcl_#Bm%8qN|!`ns=*7EZ{?XJ
zo|QazIYSxyOQsS+TpD>>E?6HvG$yjn_njAOPKOcMb;LxnzEM8JIiShDPN9_WzUT`R
zJVw1F)=q$8sFvOPU4oRSZOL1xUB#v>2Kug9$=dtnDT{5<L!WOIA!3S2x6{VpKg!M0
zha}ws=a*{b%R8q`b^(G&r)-%VG-Xx$Rx@>LR7PfO%*Bo^I5zz**JkGE6!(ewzil#R
zrGz@S*-Jks%YkMNqg$!^K1JL+4e?rY+GGD4f5;%+TQ=nwdJ2XQ3}9qYY~GVQoS}(?
zHpqCX?akc18)&f4v_S`bwa(h+m3CRf4VUxA1_5P}eO`RQ)G1W#`ZfY@-mdP{v-|M%
z^}pTp1X6Gg(E#-w^zja8ekxUSm4>)+I*3qgB$j!*{`s}I)QAi`+Q!Z$J>jdxkY>F;
zqzE)Mn~58!Qaspz!?XvRbCU;6NXhf6F^dDou6G9;XO^##wM>0&@t+0JYkGr0jV)x2
zS+;1Yg?<}3*Pb_=$qs%2BA=v|`UP9luZQ{xGq3{S6eGK*VNhu9J>ZD{U`ogEg4SB3
zG2)O)P{wP6zm~rR4L|~w!caZ=F*8!$CT|0KRlXW1sf}r4JPP3ObFgVU^aD_R<|(!W
z>&5q+<+?yeYSAzLk6-1R5rBbZZ7W)SUs`yLI@S^O>pR|(F(B$g&a=ozd~aod*+11o
zgBpP+2-Ai8a~d^(h%iy!+*w_nVc|l~ymBB3!DAVup{E-to}Mk&>fcjbBU=i`c)AoN
zyG;5Wg6qfg_-@DZbnyktOVA?tGQD{zw%>z0aC`YKReQ_si&P7C_50Vuw*~LU;e4Y;
zk@|lm5+Ow4dqj2fb0T-#dY0peVf^V25HHDW`MfRSRgYo$><u@Y%Q^H=O7`#tP@5pM
z%lu<|N2{vN1m$^2&^~LwA?Kiq4s04(<24LfZg}rU&34151Epsh+Y|kJ^S<2XMd47?
zXuH&9+0&@U;-m331}_;Gcjuv+y!Nx3<O*-e+v3;b<-$$Q3w?NNSmhsI-dt#bF~3~n
z^>^JvT29mAV7gnmnT>qbl#?Mi5n$be`4KPQHP~RU>p&W_t0Q3K<jXR!0md_a)tB`t
z0z)I9{&ZS|L_`QCs`T8Sh`<8T>TZhCg-zZX3tk-0<xEztU|4~6X1Q|F&^7YW{qDm`
z+?3?7D)or-sqgn+Z-r?=qccSOTzAjS@wIJ2P}XqUoez9&2~NVv+cP#Hc{^WlKdoDf
z%c+kB(Ac(=!t2T%@2QkE&X{NkQCc5srevy6F$GA-4$6<V&+{$bLMtjvU<wP~D$D;K
zP`KmFry-&JV{DXla`)<nP3o37`Q)WkbBt!!r1sRbd*-dllFXt{f&1o{Gh$1TpG)7q
z%n5mvl*-U98`dy)Ck(O%%}?W??Ul%<h?!9rZ|{dzL0H$Z#TvY`6C+qOcTRt<!X_#+
z%cxDhFZcWhjNXLrlpe8#g1n-a=Tms}Br8sSK<XUq01emPe%#48zWHy3Dunkrj|acQ
zJGa8qjb~F9J4a;V4k|NCobJ$VLq<FS8hbi7dqqKD4_|&gxCkK;M~=&%pPSPyBQPo3
zpCt14c>*#R<M<)C=uujTU+vV2*Flpwr}H&q`_N{m1TKLUYHIdRH{=K0r>>}V!t7hP
zF5g|D3*+cw{cgB%k`6E2YY+U#;>TycgD~}Gn})r*uBWM{RsPeSqt2<{D*FazW8Nt=
zWA~Ra?W`28_S?6i!b14BBtVHj+};JS;P{#4V55}c^|*VkJUs`U&c53*jjff08olR-
zNg{@#_4?v2-!--o-j=!R+6qxIWpXZKnWkph@(%WJ%W*4Yi87GHrtGjwiSX5Z-S>Q}
zd8&s<PVl|tc20sosqA0v3no#yWYo603FFWP!9W_e|J0OAbym@2>P2CbJQX`DWvOQ~
za>dJa3v0v$1IeP@nkc-UAv|Ul^J!#F{BOfs(HrkCjIMZ<jRw+QTEAoN<;21kqi1g?
z;?gP(kdZ9yqw|z4-K1UT)aIxTcc0ntv<Ftuf9|MC4aeD48<NfDc*u$4UQb`cLILOx
z=p~fWS<+b)+33{@-}8`RRgEV<=xHWwyQXuJ&cfFgKukzdxZ<TkAf2t7qyq(?9bK*M
zIq+(TJ4m9RP*0^L8Oy_Z(ia!C?G)@TYg)-O{oKjxJ8tOD^op_Q++&yGFU{mJ*oOUE
z&1;=u=M>NptAL;tN@5jRDi9Nb>#0OFgIYq$t`nok7!AgRAuBHFxljs&H7`s!GR~Z7
zSWIG2dbIQN<qv>fzajYRyODoVxwnAzE!JxE>xsh!Sxv3m?b45pr`2mpJ`E^PGs{2a
z8dcgervnaPwP?ZJCAC-Q2ucjD2B$8fv`5D8V>2wCSe}V9o4Ydw+`->Yr)l5bVg~<P
z5<K<#`86@>QKV<OHND50SA)Aq{I>>|nN^;s*_@h~@o|8P<z=-il|j+Fj`&gAyl~xb
z7+<V3_88Wc-CTPBa1QwNOuk%}V7zr}={kIlnC3F1a>)gNebyTQ24uOMV8G-TtJKsm
zaPJ$!XJ`Z|TWWAdjM4Q2yPJvL0R0V)1u~wQO-qz+d!MT=S6?q4we6Kavph`T)_QG8
ze&9YsNA6}*)rT^R7H`RCE%r3Cr7vq@R`RFy$!2-Fzso_c<7I~9%49>OS5Z-}Y_Q6}
z^WdgA>D$cW_Vw>eVv_x*UF?rFT9YA!w?f!SXkRLLk55}_bSK(3I8?wfVDK;7ci2FN
zTP8g*NhoqwVax4uF)tb9#!RX;<BTowf5`{jxWKn?#pq|2o5rN8g&^9}eC`obDXgqo
zA^b&pXXNCY`n$AG>cT(#AyB%}b~a&vC44Bn=SkNJ8<(cm<)fG_xiL&W=j042IZXC^
z0<8tAJz$2&dYSLkI215k4vz#!C9d2sj5sHVdfv0}ln(@#AJOdOGkH-u|0Fvhw(F;B
z!9OO0UKb6Bw#Qdd(XW_yAN_py_%$czCrpQaDe1$pObctp4ZMLU+0E<4iyWuDkSU8j
z5S<?1{F%=|25pA+CV|M*8L<OS=-oe!E{rZ%=GQ?NaAIj|_XnCuCxV>nHfm~T8tX&V
zMGq{VwPf=!tD+kND94TXx@!aEk{qI<Q9W<;#&eW|M#dQmop(1PXu=!9P39*msj<T2
zGBWF#pTWJ00S6K$i!BnYNLWxKfr2l+2c7gN=J(bv?z`ZwEqXP4PVTF$qE$$xm>t00
zl_=jF&%BgYmQFdjn{*K>`=itF>M^h6*GH#GR#G$kZCr|*4cVnSC^(gWC+GrYvLs-q
zv${6j(W1i1%T1GeX50N<eV|HZn+y`Iox$PCZIufbJb;X#>+iFzFX32ohwy8CQM%7|
z-Cy=aMSQRP`{TDXLCc}Z@^*9WgfCL%pAu+^zXMr%!Mb(vf2s#<;7Qi^&5=wY>DLt-
zh68ri?@0TQ)b+80O|&f!(3*eN(ad_1NwwI+yjOK&8@KSHF%QVwI_(D?0paIf`LIW^
zRC)3nqLAQ1ez?x3cHc9<FQ2uTpUG7|xoMJ`vYdhT&m&l&7CJy(Lc0Ed0NdB{t=nAC
z6$o2Po(NjIKk3?Oq<ZsnRpOU>|NKV_fcx1lDY+{URd|1sAQtlCBc(_i*-{6mWMYtW
zD87Z)`$z0SI$e%D5de!LMoBF3(5?@Juf6PI7HH-`klj2x{PdAPxGY+1SXP#NUvOCe
zHM*3JrU{{lpQHg_<gBI1T%ccVdn>o?=B^GGObE}yO}ykVio8sMX2WJVD*Qzk?8pMf
zXgt&9aRo}}8d@0sG7!Xg@UcGQa2+0kR;XxmqJh>=7P?lklRpxI56}#IT<u;=HoA>Y
z$=#?a)x$O3caX#U^R@Y+UgH3e+Yod_gYNx~U^M^2_Z2TwmGcTfA3?XQvai>`O)=YD
zcvgtk%83L`k4<yn4=64QUogopZ|?Z=knrudDvJhYZQsDufB*fx2RxbO9lMc_vUFMv
zsb?h2>iCQMHZY{Kj=$GNJkKJlSI|vyBGZ%q`s!xE-XmR5+BF*xqxF=#_VP@9`3q_O
z4dE<QbM_y_&2?=e<H#UURg>Rm3FidWbwBFu&Fb~<d}RP%et>b@4^~OLHs~t923Yi`
zePMQnd<J_G3^U7>K$Y`Lsu0i1#k-<`oWABG*zNo?#BK2wRr^0o8$WH1le^Pzk2hz{
zr9LuF)yc4MCTDIk6Ra=@wg5LwU6LnqhGAK<m3!{zTBb}X0{_Hr-s<B&#wNx)xhM%G
zH)jvn2L1`JU<qE^0Irj(+zugH7fb?6zi(;<oHCKtg5v)8Hg2`p1ot>*Xq@H-D&{3f
z)A^N>{oY~hsw-qM|5)Vq`YVQyBEyi=lU{r=(Y2s&Q{LLf%iM~T?hTvu4|FT;3-f$@
z3*5yx%^`3c&-Eb7v8CN^V({Md>g1OYx2RyfOGG}S(>!ViIg|fH;VeQm-LY3~hDv2n
zQKCgtmp3hes;B=Rz?ds@BxIgpcZ-yrrzD1FyUBkV)-;IO+Y`7KPwEhKZ4_(;brZDO
z3YBY?S^llJf=-u#stELZa=dxxCJ=$)(1pr9#&+7|Kst>+chMq9brB`MRI@v>XQpa>
zXhGeWII{=<SLJ4y^@@AJUiDGFkvVVphc&zEk+n1V;uJv=!1cd?g<+C;RF>z<o(FLU
z*U*OxQ?)4UD9|3C9~WM9somqM_6MYL;lNw&Nh-gfs=A{iN9q0hC?2MR@??K12)e@5
zuQM2a%rj^hTGygLu*ekm#I7zsZ1ZYT3{9~Ufi%i1hYRy6CE~CDO$b}&FoC6$tYs8^
z|N9Md8q=slXgK49EOj93WsyWe=Z1WACTu7^i8J~>4coK&GX_u@R^^AG3GVQW>g;d>
zpxq>X1Tu<&Cae`H@-9EE>d?<XZpTO!FZps6RhgH|#H2_Y;Ua>s5HQho7khT9!+Qoy
zilu_~!w^Dz9mP1S8Sqs2&%x-@{C3fRlc2|_gCT&Yy7wQoT|z}=J`2U*i;9$=n41Ah
zKC|dtEIn)h^0s{C?Ut(T?A!H{fsVKpp*B-;0=fTwt%yAQ-&YOPr9q=TJTZ>V@-Ac@
zwJiiLSfix%S6h|EDnY>-lW$ZW96?s}Y$}X#6tuUWklBnL)H8%eXypDHaRf3=`tYo^
zC_77fvD<NUp{Sl@@$)4#W6z)XeB=v+jq%5D*E~E!sZ*hILh(D2pd+!*vI|;>Fmgl9
zARr%!4p`3zDI5(`0ujk=_=B5+bjow2sNEW4_;Br)!}wdDn=g8$=$nzZ4IJu|mX6RC
z3$F~J-4STGxZOwUVlJ{-y2HvtbN%0mV3j5K1T)sFodr<lEc*ZyoR}HG$^_13)g@yn
zK9|*egvG$O*$NZVqUqes8i4ePu@NabD7TfK|68YIC5@1|JNE#{haxGP*v<3C7C9b9
z%_?}jy$X8L+lO{UKHJW_&J!v=Zy`jV(pMj|LL2zdT%F^DU6>{r2AJV5ct&dmZZ!A^
zd<*U5Gb<FVAnnVNqyBn=BbgE%P^lX%uvjV)4_T}A{Sdu~r{kO1lJ6v+)n7sFC{JOz
zQrDbn1WtGQs;HhMd^WZF=KbW*l2N`RSdJ@Myr=ze&cpAl`ACPYwk{R>O`l9+*W9b^
zkBp?9o2z=9uEUNl=j>CC{~QvFxz93_3j`71CE~u?d;x;qk=7;i@tw0%?-n40JOB04
z>GME><ND9&CTl~SqzUSTw@)EKgEKJ?!Y4YQJCO~7TeVOFh{I2Pf#ON^56(^L3?dJ_
zyB*(kw5oHqFPD3$*7wt*jCP2Z#q5JoE5LE+G0YtvsImqd(`Tr{^<eiZ;Db1()5j39
zf@^!Y2d~2gC@)P}lM=OzH&!DrEhFLJ#~KEuwMn*lMr7hl%E*l(pbCy8x*giO=o!YC
zBgF=NW&f~B4B~Q9^5iI6%sF;0)XFa(#}QG0%G~L~Qw_pi()sV?pfdOq%1#6xvQv3I
zd@&q*v6Iy;8SvbL-v`F;joU{5zUw^@0~b5RddB_MN4U%W6Coc4IDA+5l_)qLg9RGB
zfq+4IYhJHPJ$UqQM0WF4;<@&QyopQ~!QvI2q+^}W${RA-b*H*It>JsO_^8l_M34f8
zLjrgp%qHut`j6gaq7Sff&<+!}(M@!k>{||+rf7NNo@@Lu6z%Z&v!rUE`+Cf)Xw<Lr
zUOzyGG?LD({bv1+J%kW<o6k*NBAKHPNxf;4{(MiY4G$o4$G`>ph3m(6w&%9Xu7*+J
zER1@9xTob4aQ2Zf3)-1VR4Kkzss<cL`!KIeOz=+m*9%Z)zOu#P<oc%YruAp!lfC7L
zV^8X25F5F69v#)4=#?RsuLQ_%8Cm3eiynHZ$bTLjvZPy-&Bb(UasT)Ov8NGpn7h@6
zQc~_oV9-Fd%Li<<QunDGmwWVsy}M~#4(0%!{269XS-j2IBtzY@HX%ri?!vM#E9uQ#
zcmmBl%(O=gDB#pWts9fhLsqIo%@YXbCV9pDC263p{74`xz2z9FW{SY5Ypq_o@}IhX
z=>_67dfW<TNC0xJW2pasz<nD)y5N&_8#E08S49esO?&toYT*E^1*Ec}x6znD{sB|Z
zKs*&eBV-QVQ~O8k{+XdmJCPmYSHdH1?bkIq06J=Jm6%%Df8a+T_mvwUr+=mgRj14u
z<C{rfSDQ&<$IaHmWGJA&I_{pxdi<8n8S=FedTi3K?J^nmOn4P%!1rf~62J*B21^U#
zJll1|Tm833NBLx3zzKXdscT=T|91a9k!OzM7N|qU4nD35yxo!Yju+1GxNxxn|B`s^
z$-g-xdbVO&qTl_z^$Q?pvUU+Eeg*=cE(`;t)fa$XtAPK=@{U1ZEK^fy5_Mxdt6FV!
za+@dF+=Sl(E<1UD()A>b-vepf#^wC8I7j}ekM;IX385C#(drkaQtOcu4la;4p}V;7
ziDyrua@opR$qzs3Alce{K~QwmA4ri{roP-hM%;$`m8Z7)|1AM#gkC}!@%cm#KKc70
zKlK&i+p?z!#WwjDhIn=34dwziZjP5W-z4%;C-^Hk2{fftdoo}R8RMK@azGsZJ-S4o
zGNw_}OCogYZvZ<4Pwg-+SW~$h%=k!($&b7&#CE>czTHp!E<u|mutEw-sal~@l&Ts0
zM2_R!ECsL#r1QU6<P+zfL<2BjhC%ocR&+oMx|YcW;QBb!nt_FxQ>(FDqvmxJ7s@yP
z$@5;<;kz@I+kDQ&n{saL(35Ic16A>eEm(F_!htIW4^$?EFRtqoo@}#lmem|;p>8GM
zK;ndl;M=BfP&MoKmvHAF@}e}q!^&;*_bs*3gwd888d=FEV&}O)n~D@UC;kMo98Hj`
z4gCMZ8=~DrVtG&inc}gd=fQz-YAbsbp$7C*^oJmiGyM{S(K|zQ%V%5tHv!mi0bs-3
zuC6|6&dr0B%f#6A?DWU?eeyeb>6P>Z;9<5NR=CjtxLf3HNXZ6X|G|CwwBXRaT?v#t
z`#hg&fhxAYgdpFY1wHr@-~<Mt)REhulCY0L`>&Kge!&YCC4y%yg_e96+3tv(XQTN_
z-U_qt;D7GI^LFqOqta9gDd^yKd(mT^LNzrpB3s?(ZQxRr*f9yHdH=|1!1}(s;mvo{
z<#$11Zck}%@rV1NpN3+_0m5)JH^u4MeOA)&0eqy(Pv^6~{tt*enMNG93P1WkgAazF
z^9hktbr;r-zfT2b=D{a8`&|D^6M|ke;wfYu5kruKP~J~cXY^}6LmnT}4@gGlHdL<a
zD2$G+zK>@#h9+{G<QDqD#SBo62fqYtLZms#oNcwUeZ06ibLED+neySLW}#?(BrcKb
z0`1z(FUw~LF+V#4UUKl~VC;SC?_(J&K=IF(Dqk2@mKau+PSz(y71S69{YW;$>E(1;
zIr_D?)aye^t4rsJeDd%Hny`ijCydTE<C9%jHh@kS6)ONFKZUOsb1-eX(phRIM9{h$
z6ZmiWRY1EJFow3A?*2e(MA>*w$7VkrmC_qhi1O2kL3pxCPoCmE>d12G*T9Z9U>2$h
z^RiaV57ZRiwC!VYuu4KwzCzXFgx!cGn;obbt$6}+08tJoWC|Pd7>i-lSmPftW!F#<
zMOKRW$Vv)`pGfET?37I+#*Ll{ku+Fxl<!{V-Cmd~*FgHyV)=*o_Rn#sMjXfE?s<o3
zo0*}^7A)zdB3!3>fXGkT5$f1N)bn;HQfOW5iY4&nB}y$ez7H_1o0G273^cVL#`6zL
zyF+0Ig8imqW1xq_QQxo2LQRwYsQt-jE<oAYaWr|;w-miUL(v9amg)ki-#Z>cs5Q5v
zbwUI0Z$>&bpV>Bd`qMe~zE7g*C$0E>U*ns+V}al&d(LbCQ<hUcI)4TMYiiia=^9An
zyS9p>;2-EPHN{y>Ep-i!{`Ly-+K{0q+Gp95_#S${5JGHPo9yIqoz1}$ZV`~IF*!!Q
zB`7mK4@lks7l|-CkzZ0ifu*w~ijd_qLXvxo?U`XN&KOY}EE_KzrGRlQGA1KrGmyNt
zNU>>(unZ>ta`PJky*_)|u(MCdoNn5-Np|6cP9S%w4ZJ-t`bes;98L$uaksyMUq9??
z*JUQtLM)imb9(?kltlBABHJ;&W}*Z?IM54|kkBX)iq8;&P*=*l3xp(AC+&*9GU(4N
z;p8d$b2c*^;Zn>tAOxNIfv!lUO)cm1wl+%0I<{E=#La(lCj9?S(-papFoOz6)<X2L
zpe!mPCdHaN6T98HRxU>=zi<`bWWo)$H3CsSXP+RWOYMK$;42i+F>Qh6`6UBd2~HeB
zPavgFRuyvHmH{Qc$^khIqJpC8zW_oJa(M6&uKf}2Eooyyg*+z#ZcZy-P=}25CLPI|
zw&y^8ss%B+)brWsN2BkV+uU-R@W|K&fE_ru|KMrCs$Ie-i<VN>4nPM|c9y2@9w<1y
z8u`GX2b>8{grN#mGs|QYprz(XFGh=+S6cd{NeF`QL<mAlV)UUXxmRO1-Q7z2K@j@Q
zAAznaeE#qgk&kmT;0FVS1Frz)iK?YxJ^p14C&g_1OLfAp?|INUmI`Z=MKG)>@j}7x
zx9CYw6aa{#^}D0OTCfaUFr%RH*&DffMxNyhFFwgwyG&V)C#b>?Rch5h5$bUq$>4gB
zW=jH?4%i{zsR{NoG-`t_7ujx}{Ofo!dmQJj;_`R<#j%1^>~waYU)nY>i93e6$8yF`
znU{Xu79+VaD7PVqVvdk919K5KCS4UhR^H7jynGe$(GWN>nENo)AD*q|rkH(2GAD(#
zX9)&v&I*x4M@;lI|7^y4b}Lu$rhPx@eZjfH=hsVmMu;K)pRJ~GVcUYZl~%<tOiPid
z@>6rO0~@=z@A~B#eY`V&%yAa7Ls|vO&enmouhgNLT)@K$m@6xTA;#WT=|2*yrTvW{
zd;&NkDjv}F1OW;*@R8Z)HhntQmL;XjVUzqvM&skRC4efZwN>S6L5o^!O2Nja_Q1A;
zg6NXib4eSZ^09kEx2*0$09Edc?UG_|325T1{kqaFK+{)EGCxi<7K)sVr)j>v+=bZp
z+x{L?CMywPb+H(DUT`7xr3O^903oi5lA?S;Tpgz@L~Jj96iI>*N&HY*?eu@c>7`RR
zJt}$+UVo&#?Q-&^dn02;q<$aJ80XWUSR@I01+H8272gOfbUXlc(Lfcqsi)ZRjj~B&
zA?<zHT`#{1GOv<N66nH>vH0$J{$FlEW#An->3Z%v)N@YDDz$e{L4_4xFl+^8Z<eQf
z-+Wxz=Atr4aNTTnmH~~v5&`FqICDs#{I^XolR3qGi8MY^l$R$shp7dXwOn=+%oOoj
zspqQJ;Ye=tz-G`qJsURN0dBdkJf=>O4G<LJlh8|;e40D+x}5RkU8@%I%yTho?UE+6
zEWfTre_{QpRa;A?Ex4Zn0#Lk0E$1bKj8Lqny$og3ISP0A1eUgwyA!VoSvKe_5cw<w
zMsw2qha&HqM2slc@|vZg=hp@n!4)DGt!`{D^aCs`Wnlv!!rf#;F~HUT(PGrAgG4@=
zck(Tf2*8i8dUtln|BoBo$*^6hZK+=b^Kzc=F<oS{`V*%hf_gSR2vy%L*!;ORm<cS?
zpb_*4i{sc1^5LyBEXNUy34Q?@)~es;uV{RDW3=By@$efE3?qzaz#jHIkf;S?uKkYI
zUH;AF64~1c(pu-=uk+j~gyKIAnA0g4;LH;oyCUn$e+1Zl?mhO(&)>*rdG<`r3Wh5{
zlg}L6nSAONx1^K5G)V~FeaQ%AJv|%|#JWD`PUBhkL0j(Odl}9e8wGtc`EC7=M+v@^
zrUhsi#uFS`rt;q-fKTjJhKg%tK7a$!sxv#=$%$BdycVym9g_OyM;J?%BQwDB)%-wN
zz&U1mQ2=E_P})uYidvV*5Zf#?V8aAxlGuKBy2UZ4?L7b6Lx=pTt;^w<QL+KnTF{)+
zVncoBe2l#uib{qo=sV8`I5QsIR$$!Z4(6*S_{YZwc2k~Sl&Qkh0C|{2;A|E<W#2Pc
z1>|sQS)(c*009llr<?SHG2Cff{Zyft(gsoG$NeQoS%2S<B!||nqKt^1NG#&DNC4E%
zyA2-DoqkAPOg)k}j(o9-5<eY*(3Ql^?9MX$(2*B^U|>WX65*bMJZE*WSDWA{Eg!vr
z4pDlTj{c@VAP4!2kSld+Hq~wO3hnNLA#x@SJ#X%@d~oQPyLLrjZuxV{1EqtfirT{p
zt+4ZXeIGvR2Ab=4T=Aal5dx{Z8i_Z7W&+FtL3j4Gz}^Y?U=G$9jd8t-3NLEEp8pNR
zEQ1#?4sV7+1cpIbofKu71Wu;5-CbKnTBnk}cui<fJj82EzN!y#P39_JiV%#U*lay>
zb<W~YfhpM7=JiRF^<wpR8gT$FG|-J*fT20V&UC9@*5h2FG~sAwIgsaAMyR{VAO0X*
zHE)`=TO6hBQkR;R^8x8soX*|F>+bVPiymItdn>yWY*1-Wdl>8|cD}@x!xlz>BcQXc
zL5wyEL4PDnIY-T9g3;6|x={H7-@mVx=7B9?5}c(%7Ng(|oU6M-;UO65Fzql*J_x>#
zr_2vj?czeTRr}4gxPpy=pi)j0Td^te)T2bG?Ej$xL+w7ubBJ9bBYfwJKxOR*@VHw^
zkHT{;y}Bn;BSENtkRk-Ut8!IIgKcC$QiV1_8{~{^elwfT7h5$A0*1!S(=(p7GCJN$
z`vp3Ls6X~-?1}57JPdx2r1xUdyV!!A5=OOpo+_LHCZq}}ExD(+<KJ7~9|piA&RV<B
zG9c&R54dfusY@kpRkuNIx4UD1JkWjzHBUDa?*cBF`@=;FI-~La8|ILudxIA=GXlnv
zmPW-#e_wO{Jo?*^>RFP!o4MJKY5fhInpM{Arvz(M1n7(#P?%B*c3e=8&G0;N*P-qA
zFn5ghv_~DK!imB0Y(|#piz+OGAkpuZZ-xaAr!GI(xncV`o@tKw+}(&@A69b;|B@0W
z(EBmg&!Ij0@>vxS@a%}W+s0BT+UROtJspm4vaMEhS<^r&YQE{bwq7=M(Z)nLJ6%lM
z8yE&K%QnJnbeR&m_^C;_0*){5_20k6hhQJLe4M4qT6X<U8>%#9w`Xv8edua=h66mB
zfd>rRc>WopIqvQ_ORz)t47>J*5R!eRaTyGpU(}NZooq25NicExT*ZX&2T=BBH_N}p
zj9Tif;UTt`eVx)=k=E`nZM6>Os$@CCQ*5HgdhOAeDOv)2foZ*Yxc|?niq9?9?{*|0
zY+(Px1eQ0q%Y8Z%Y<&DYK`bbz4zWM5#KYvPag$fN6bdfpNmHanVDcy!SP&RQ<q@tc
zeGw)5j`N7tguFC%qL^f@{)MM`%hZy))(P{Wwk<+2dwI`wAAO09AX+-DR?UU+e~=)6
z-R{gXF~FDd5z{R2?)Z;IA*KBB#O%g|NZn$yJ46!z)4h!WeZSA(NE`wEgf$Pdye)Ry
z?S8g3cQJZ;-Ja1QQbdVMt84KerInW#GT>(dqT-%i$`*C;*Z>)F83<!U-2V;Mu}-r8
zOQhSo{!Ja|>~xh0`?_h!WNO<(1U;b%-Sl_mB}aC0A;!!t@Iec`TNth*CSeBg?aN7Z
z9~|b_0eoQ()+tclNb{<Fr3NWQ;O>6ON&vGx+^4makc*5hMgDwoH`&W;XN>lj=ab#@
zhM3u-%fTDk;QS!f@`sTmdKd2k`9&VEsCQ1GcfO8IJSbER5K+?YPOgHKO?=`XadJno
z>Q$qmNIkav9=*J=+zEe~GtD(%!afnOYOJ!kjFiSo2q#_PL=~S<!0jw7o!yh25<am>
zBL?}D+_kVKWt|ARFXaV#WyeRexa&Tyb&cSxrhrAU9J6V4{bm|VkzJD)C+UT;1>~~}
zh0SXd*)B4xI&sg8cb2)qc(;EcZpBE3@Vy~x4-nOM$n-FB+z&uhXd%0|+p6BN@D3<U
z!K!WW2mBNFx^ms*g!0p>k<_6Y4006_SJVuVb^tZ+79I#1q~=I=^xx1Bj0MT*4w^5<
zqGKKveK4Pv3^HpnZDq>~Ab<>UYX_q1`vu-eK+n9D^nGUoAG3>@5(}#2G`}g+odnte
zk1~2%>)J1&KxRiPUZ6yHe(-o*%q~4VwF$oPy4V1oTUohw@VjCB19kYI=j?g#+|O%v
ze5^cZ#{i6Pt$p;j8QH*l^xf}8*vC$?k=@qL!UgT9nTVL@gg>x0F-2cd)4Lc;WtH8w
zBvNlA7m(yLJR-QE)yY&iz`hXp=fymdquWUcY;U;4MT(R%9RyPhUq76uvRWfX(|<U!
z(Vut3{5R@n_NWBjvo$Z9+8S{mMQ&9swFN=?xJtk~hgF8rg<{^olA4jVb>L{Oge6zT
z9ZYdVsrWp#p5AwV4b$S1toh{{-%_szKMbz>KfvFhftK)vRYGnWIyy+HH*zs0Zs|!8
zbS?S+^tto2ADjTrY$hME4A9Sj3IX%{?jYln+YAl6YO4Sn?^lMhh5)H$g{y}OO>$p(
zM2NskJOIvx5EfCq=6=XVIX0DWqkKjlDs#MZwiob-2={Sihs>8?lwtCg%mw-Co&jKy
z4qNjmt&j{#UylsRVm9z*AVy)A$x#cf{$1EUq?5TR-ZCt4TWH{+I6HfAO*(~>j9V$b
zH1*M^ADJ@cN>g26Jn}x68euHW<V$us2@BDptFgy=tm*D&43R~#cGowbG{Y2EiDkqm
zPH-s5I|E@?+c;I?s&GGfO0P+96MycOqgal5^QcLDY~`!tOMFQ%64sbP$UaYkOlZ#s
z7O-G0?9(d%`YkW`dA3E8p46uQ$IGKCwVKy@#sgPDsD4teq*>)}sQ)(bh6^w#fFp>q
z4wRNJuS2u9k<7o8B`PqHx8aUxUM7UG`2%1Q&D)p`FY!D_u9YDlu7{Iz_DK2i(*Ldq
zz7Vg0>S<|$a7e*#DZ@(mJP=><IgD%wN5rxXh!p{1!w$m#<-fsUU)M9v(wh#>$uVrq
z?{hh2uT*IHfermzuwtf9usjNMI-EEHT(Kn@QWy%FoZl}zZ4dAJ>VU$)UurI<1GG2q
zQH4&vAej@Ndtci`;T#wU$|_`>^9YsvtE+l(6=-0Ux#f}3wl+ubGO;w&l2z}oyNH|Z
zw-!^PB|*X7HIJD%Jfps(X>H-|?-$}VbopP_@i<LzGi$udm%eyqXochC1a!p;W1gAV
z>$&~}Ftnwhz7`UBzlu=F1=Cy*<fZZLvk7+4qxe5s0N$68pMx&}*)`T6P$D4))Ma-n
zqss1uqU5vRKXch{`1T<m={gg$<FoNQxKrc`Adi?G7=w*vCn9ck5zlFc?s6L5j16p>
z2t3Qs!iCcBH@S!cJ*(IoW=BSz?tje&W{Fda1h*1$$e`U3<#QF7Y7iY-xd6m=^E&W<
z<pk~^TI0Roi<I~$g!L`bC=|MD98?W>5zN4c|Ag=vTRsD4VcwbQ(*h#;GG%S4+siMt
z`Hl7Yf0Q>>n_(7J&|3d_*ontc2b2AXwu>0r%@-(aVy8h-b@Q@6)Nl+YW8YHE-_tdS
zfUz}vYcT(vUc@*U@rxfglYC@ZJVo0z**XB?s;lS;bGm$+BVbNOZe7B`?gL$m!g^)(
z)&*)iU=Il0j^st|)AP2{P<|9Xp=Xmz(BJj>PGf7`@y{*HS=B^Lz66~vi+jlK)#Z}X
zQKZ+2Oo)Ej%sJ<YbdgZmoq7$-@;^&)Fdg)-Q)-ebx|sM3xp~21DU2{j&EiX8BT@WY
zx$oPnIzF!)P~|Z(fs+e<KqFcVN>jPb8r(a_Ihm}t;MpWOv1GrOYv$W(otOYdOC=W8
z&<p##VUEA+W?hX=>xUlGP;`~)YzM-^_~Z9lR9y8JlDV%%(-=@hoabQ1nDe}LeriGy
z@p->w35TXTcm!r}@(I`nW@(SP9EabT6(L_fs#~0V)e}Rb$5Nk|(d7!GZV`2X(HCoY
zzjJ_7DJjcg^6u16PFpm<-$y;)U?5zxVB7D8j!O;sKIR=JirgdacX2abqw0m`Hwh$Q
z)+V3|#0c9S@-KrE>IOAjAIu?&vTZ3!#kmWR4OMrPRe^G28twmHu}dy6c4RdUP#){0
zuct&+_y2JR`WpRE-6qfH7e6WbR$>27`Oxjn@l*A~<)8+Dhc=cF&k^xUTvU+t7X2VE
z`Apa`U&aE`v}roos%hH#+I;qiA$RPRE15AM%DDypaNsc+yO(CMge~=SO!9Sl+HxHX
zQd9i&;`C=rJn^f9@IqTpj%T^|-^>KW!RWo)sPX0q8SNU%t_aNy7J}6*q@V_0Pz6K!
zKLZ(cb$~;0&v4N*g%2>qT(*x@+4#WP84!cS_#h%Hy6YqZ3vey~EyUqV+0&i%q$pby
zoVk+Jz%=0t5_I7wQW~-}`j%M5F)@O)D4T_?dD`6k$XhxYY8(Our>+adh+u;&uaO_S
ztD80)e&7v&&;~jDjR<pJ%dTFkOnGU&;x?Y%)nIS27XDwvCaR<T6yH6qkAXlzBum(g
zlz*S6m8?M8i!YqAe!C_2#Re<EAT1M<@340Zx*P_GTho6Pbh%{#rQ=b{ujaO~c8AD2
zm!7Z+EGEGrrkRW<nzprK|4m%qQHFysx0zCEZY*eDvyn>-18qdn^U_)oO@8yML)To)
zEOdo6$KkcoR3KBurh1R_EtZ4BYVUNF4U^DAZ4og+ce<E=`K#HZ?h)=EcMxWD17M<!
zE;agxn)0Ls{JQKxj{D`pTt`BnJuBe<r?R*whxF<{hZ+n$0dv8&BNq0mXhmLjDI*Hj
zqi)4bOsQw<2~NM6lkMuRyTMb%ZWOqDcFVSwL;N*yoW@I47n6@y&)H<{*<Q#@Q=CTa
zEM#v!3n>r2;Vnl&Xgms8pJ&W?P~IiyW;>JN%Rl8tsA9|E>n4OI@GOTt)1HQ}SYU&%
zaB=@ua638FSp&DUUv`cINWy|{aN4uXMb^Z(>f?`XM}jQE_EtT*!O2>&qVg*CF8ZLg
zPI<eN(w1cHy>9b=f}Mu1fC`6CV6W6j{oDbK!Q+eaV63Vm!YOW^t$9}PGU1rKx@>(q
zTjyW3eDHs%W`l10_2CDx=X3Qx(h5gPd?S~ycA)&PTi*RO$DQ1icmf}&IH~OIL1_%4
z__#h@xNyO9!m@>p8fDP2LK-1I(P5c`%ZhVdWs{PcrfHjW8+!{iAzdxplep=cn|nia
zGq5nv-BIHjE?KK-7ctMCJ6Z~3R~gJ>t4#+MZ4XaNa(5!n)NU!NVuhVNhJifLvzj`M
z-ghEnE-biz@IP>;?IvSw<7IQsBiN8dnSW$@F%}H&N4K>IH1%D6?F1SxjLXYM{iBO7
zv>}^CyL!fAah6y$#Jh@idEh<bKn&00qD?m2ZO=(rL;2z*8CZUi7(X;B@?VgMpf!l?
zeQ@Z#{e2krzP00zk&g{EWZia{(KC|@Tqt@ck9>K_tZa$P*m6L2-=pl(Jeyesr8!^K
zNk;O*{TigZ#{6h`6;(gPD6%U{`>#0X3!Lz^A(q=y>TIM|3j^$d6ihL**xD#So78qb
zD99q<yR1^mt85gbGa<T@u4Kr?>(<w)=NjGO66$k^p9Srg$|&?>Z%B=19t6h1;xLi@
zZuDAKg$H1I^R|%q$9R6mjAFt1`A;mwbG=b8I<SdVvJ0EQ5Aki+YIn2c*6gG&;oNrL
zH*nvBk{NaEj}bkClY$hWH%~eE)f3orG?!aw;CU~K5D#|EwMS*H$X=bf^~+oCm(Dea
z&lqPh#&#k%_}Gw5U=~lzxtJrl-`n|?z$>UEM!wxTis0W-A{4LITexcrga$q^Nrkl?
z=nCC&tZo1U#^mfFIiq5vu3v5`{n5GzWFKrm75b(<rDR6hbIYVgZ_-tdbWm#p$#ta-
z9GgKvHz@~szCO)Q?rRO?CM+Aw4B_2|g<2ONK6lty5F!6w_^|V?X~f*}P<#9Jb0B8r
z^SDXGuZ(@9?mctnmx8QM&n)-RV9`NWy7Bg_+B@<W6>{o6`OZ%GZWhvvG?4#1r*OAY
zxqFDA)N^N!+Q=b>DW5>0%3x1epcea-Vn8lod<<wMBwkO~<^n2mPw?81Rn{e-$^zYx
zbLIDG!)HSD+Z++3uVO;f@6+p%{UdC1n`^~(5Z3=2mV(LjLCAi5>1pKvn1xVUk{_m|
zv-fz7f0fqzhX&r^4mQ4aq0}462i~7R#ox8={pgx70q$MQUE~vT=G>yzSa4cBgRKcY
zE{r~KwU5fz<d!W1uOVyq9GgD(lp7~_T2Ehh3#?Z*O#QoI%NQ&hl(xV#Lf-GZH+jB%
z^IPWIym2{kf=c1!v{<yBb}j{>4%P6bjBDhovs-_UBIFBIZeF3HPZoQBQeY$k-B($1
zKxU{pV*o`qc-@JEvY!{!5qxc#D+`$Y<gchJV?Esq*AkTm16RzRU`q_(nP7uNhI;VC
zCCYPDZKCwJ*FY(~_vG267LbM9IPl;OmZitW{4#mdU67ZHy2^JsFpm>ahoql*vkLko
zZwTN4`|01S=()ZD4yn#~o6gyb+aO$()om9sp#+s#^xyXK*Me71<_irD0t%G`=r2+X
z<7>dAq(R=vP0UhhL%VVpHg)IRQ9IDxP(0guVvg>);`_AVb@VRi(r<yWDE79ed1DX%
z-VR0i+P;3P)U`vfry2Fvp+Gm|pCFx)q(jt|R#<J{B}VbX>YGGNfJ?QR%&#_R#|w>K
zu)`Fb3KuHmysCgRHM)7FQh4#vK)2gx>e^ed_+P4zTI;+vM**9qQ*v&UTdQXOjoR!)
zvGT6BUX239PuBF@F{{uah8Kr1cj}9<o*79eiB-~6h~0!S{FSWL4o?S1C~AQh<3DyU
zl*Ih@3Hm8D*)QeTFQwK6RLMiOE0<qRz@xxM43<~D3^bQNUZfQsGFqa#0C-3tRg)m(
zzz=+*z{51`@Tn=unj(xDcgmOY<$)UDe_+sb{zRj<7*L7r|KWRzy%3{sd>82~E)iJ6
zY5rr|#ZBPubT<iP&aD;mwE1!cEJ*jSqFhoFb~UVPz?kpqJI-(KJyYNyvua@Vt3+z_
zr(AeUjFn!?uItpGWUw1M2LhjzaU3ZUw`My}hqhFtI?0jP#<f<m%R2Hw57-IYm_+Zi
zff+XU?+?z0p8N<|-$4JFH1n7<^98t7mv1x%^OE^Thqx^WT%kV5ihfqVk5hCwRM6^#
zNh7L<JL@D)A_E{z*bQY%=bSWi2$7Ezn;-{*IiW|5S&@U3_ssPvTXKTlHBAOo>`XgI
zhWIJ($0b2fheGm}J?36qdd<y4{V>mEY?;zybKhOnFosR9-;A}d)14j=aKo1nuK^|d
zN#xwC&mmrYIQ`93H8|+Odvito+h$+1_EUvI#(e;CW4N^Q%ItGKd6(dctF1nNb4Loz
z!^(KBxW8|AHXdUMHeGaFt2(fD&v^D%Hr(!3b32g1apTj8`~p7RW?ydFVG8K}{l|k#
zjg81L5444LyLu=_Fitz>_a=AcdGmq*tJ)rThcOuTL~ZeMV9kLfLU&O-UK@J;6LbBS
zBI+xM<=98$%e}K{zAJ<qcqY=b<+_JHA)1>NZX1m`Zq-8pmvB5$+a=ZCzX0bkzA&AG
zwF^2WyF+Lq8+z0`|CajVLsxukZ?xGx3;WBrU5@<!*m}#TD7-Ihbm%UXk_H8(1eI=3
z6qQiv4yA`~hLQ#W1qo?X6zPzLp+Q7K5QZK?ItCcYf#E%azyEvhy6etbev%J!=InU(
zv!A_BBizx)%yb18KGvfk+PE{BM--xnz;%&Hf`<jE%4)SmQ+gPLw);^2O@JeQoV<6%
z18fzr(?EjtQxm5HEL>pws4ACT2c+7wT=g&1I^;6is|eDh)S|0%%ZeRwW0OOM;cT^$
zPpYP%Bp)RXFvL{TW#T!Gokx&hE3ochBGBjs3Pu~uk;w!4s9C@SpWGt8cRwplSBki9
zkzivv#Gwj%W$NbtPmS?!7uze5h8vueoq^w#M$+fGkDk>T9O7w}lTo`o#dTtWrMV$u
zzM376#vlt%zkgx!L2-gc3A|QaiRp9J&9lpD3H_*@j~^7?znXK7UYpBIg+H0C0wJ$g
z|Dj=*NK=yi%ig;{0g;gyab}`?Pjm}cEnTRU?I)ia;rvP%=T}GWVn>y4^zehFakhsM
z(gewAO_ORMTF)$CsP1?@RCQgzFt2HtJZ)noeK}-f<m~I@YiGtDNR!jT=53^R)^nMk
z=hY$m^Ol6OM_UY$7agMD#CgkJAM|~slFMsu!F_)ORQO>=KXedxTe*<-!q+f{blJjY
zf{1qeEaP)f{LsK7DN>5BnCq<v)7nbgpZm)cZq+(sy;_wSt$pJ1Lt{{DeJ^&&aav`8
zD{Hj89H@IqP|<PmN-11tRb35#z7!`SkE2596VCNy3*h-k>Yq3Pq{IOfe{DkY9{2?r
z@Ri<emTbrKl<#ft+oX2Jia@{EZ$FYs_fv-&sB&)*qvbt=zzIK!L(OL_&WG1fJE||u
z;xQ}6>Zbb!H@#T;7B51#Vui8zOa{nS0a#tyjSrx%&AYD@-V>4AF8N=`=&?DoG=Hkv
zN{u{zlD&I~qJO|7&Jkue(k97XWuh34^t{v5J=`YwIUxph17cF3GL6smbjyyUJxS-D
z-t@a`w59%~3^?wKdWjEk*mNn_TweuaqD`CEG{Ms2GTux+NPr$H+v=USmpH_axS9M4
zyWSeFAS%%Ik&vb>*_m90DJZvhE3RE}Fqog-RN$OY!mA+5Zv9_J%+lDLH6do9tE(c~
z`gEoioPg`1GM8Sc8u}wgC7@3r(&Jr^b0uGz-hx+x-Q`-E(nncx*7UEaR_q0H*_^>t
z+8!o-+IS*I5swb`AjB4bhwq%OO++!v{=bsq5$E(sSQI5_Roo`S^u*r>G9&HWuu!v8
zCHG^-pdm_OQa>EY9|0+}J%7!2uHf&@$kXGCIzGF(Zu0r=J4ZgL@0O82@ck=D@bZm`
zCG&ie?^|Yk66)I)mg0pL{Z16hFciF$sTzF!)@@1e$6Soe9>UaQFGd*KxPN!?jyD_J
zmKuk|62;dkv1YPv?oygv-|f0nIeW7r^s0E*_E*2zoI=0CbqpS5?3st$nfB=Kzaw({
z{@9!Vp0dB$dNAa%$>H>t8I9&NcHbrTGBr=MrSK-x+!)E25I#cnOsytK7SO=Awgg`U
z#piGsudJGtnZK;~H)jN)D=*dxOn3k6d9C*|hMU6*^>eWIEu16r5~Fnz-v81+WK@<s
z1(LFicW=tRnKa>$wonmqon~aOmW&1+&aZ3ONHAY-+-Qul_Ho{+ojhQR`st-Z5l07$
z<j@iC;Cf;@rVF~ac2AU(RCbw?+@l_jY@2Nb>XNpyKq(#?pJ+6=uF2TR|B`;k7`h^f
zxB9Bl|I%*P<>;eR7Uk&x+ZFGryr~v;xDN)Q;&c?6Vd6f2Js!_}-+1J<OY=}pasQF4
z1rD=ikq|rwn5`RnN#A=_Gk{vH@ggT0b{(k}OuX0;4q_QH<T7W==b0|o{UI)?NPof2
zHQ%Or!!}ne5&vT}n1un1FU_mID)(H#7%3Q+^^wB$B6GxO(D(UibzdTcFErI*v0dt{
zH9}sN%wfNpN`p9!ceR4eo82L%V}DqFFwy{7tVNLG{IK^hIu)L@&8VFDL(h9&AJ4W(
zCgJ)5x>*@^ygGoYd|n$QHywmQbf7b}nC?)fr4CThPu1<)a0I7{%f#RM`9bfH-2;%q
zCzNyrEoBH6EvJqbU<d>P86w9Bk*JM&H(#!xDUX!oyFW$UhR08gbM{ph{1^RG#f2xF
zLj;0Xe>KgvB>W7DtjcwgBMm*7KN4cH;YsyhBZFs1O-CrH79XNY=nDj8aLP1R6}Cbz
z=#;WQeTYi{psi;3H=TDHlsd45E^)hQ)=QdvNdynTak4X!SG-fKNORTUev))K^2t#$
z2^6A_y#gb*TF#btuJO?gc(_+oeyJNMd+Dw|`a*XsX9MTqi_lNdLte~4h-dGjMh=Iz
zh2O{jlK-ztD!6;qr`eEe8Ezx@uO&J*<gCN;2l{zKh5A6p7k7i;SX%vyjW=3<$-2Pa
zE+Pt&t9zx9^p_(_Vz-d~<(r&6iw>nIHsg(m0>f<(ujZfbvc0XK%(|jHd^>{cz`QLT
z(+|h<nNVTRgQ3b>9@nWN4mGBpum_W{8lN~dc&;_BiE?#vadi^Ka|Nf(%p^q(w!A&p
zjGD;KLK;LmwMfZ&Vfyp7r~(zGQoG)=5=@`ef42W10OqH>GO+DLW0z~O#w)))|3hi4
z#jlF1dd)*yh?w%)ePksWmo#J?JJcI+cXtw@_y|4l_Y4Bb<}aEfvx0q5WvT(2McC++
zg<Tq~dnTeBW>^*6w|&g00FT4{k~vqG{;i<wuUsTQ{tj=M!>}$X21B^9QR-DTq010V
z0t4UQ4?FBV0>k$*CavggBcx5f7j=t$u&8Nebk;lY0UPdCx~80|k1ZJk1ql!u?kZ(i
zkIOR);y;4xEh$G24$U7V<GJ_kf~uSPU%VZ<Ei6Ie(or${_FBd>HW})klR(HL#{pgf
zC&deQl2fMdpX5~}@>y)0UZ!(;ffv!BRWZ$mSi7Bb%ouBvF>f=#mh-Fqhk`p)dsGUH
zvb~K^4>JOfc0!z=j!yHFeq5VUpe^<JE~bAg<xtFL9J#_8wFHr!+W^P7gYNCTNzEI~
z1Ut6nmiLvW^Jbr7eu8aWJH{PBnQ>l{xIH<<GZp&|mpowKeTy9Sy?Y^<MZHtRV{9iQ
zf$SGKFZXO<DTr`ImcQiB{9#lNf<jGPPF#UefBD&#iOb?Oh>><5;1;1Z;s!z4gYV*G
z5a=_KrhdkRxyEU0x*@Qz5W0+J9A899^}$LFO1<x#Hg~m6z6emr`Sq&kZ#w9jxEI>T
zwu}Cruh`nwC=L}<p7~rdyLuHYCaWnHvK=?mGa0sh5{9H3$VYuPN`+@+Us0Cj3-gIo
zp~xlk=FsT0@xQ!#oTU6SF$;Nra3;{*XYU}RLr8=46`C|GRlX}<!6a=KY^3|0Gjo(l
zu#q^>zQ3}2-1gu2;ethgRhQ%K@r7<H9zH{n?$Sl;YB%E+$a-Z04?v-Yujlm91+<i5
zZPmYv2?Az)U7hhFPMygad6*t?Y@#rD<<;T_$R5uI$ILj`tOO~X#qyUH-)THW*|0A+
z7_W#04;hsz`jr1j<ig*njFeSA9^m3uZ+u$q-`7e#w+JpN?lDpwoe@J##w9M523Z$*
z9Cmctdk@y27gfeSdpzU5dQcYe0C>&x93ki3jn;-u*NKIP>}P8PI|pZ9yPZaIJk3o`
zq@z#Jah|wzboW2AQ?il#YDkjk_(Oe<O0Ci^5ZyO#y4^G>Q`Osg(ztgFwo4l|(cnXZ
z#jS)#WZN4QnFL>ri?qkdkamoP9>>a+4!(rH=P31b7o2rW*btNDhx2H}>E4rA)w}w;
zf@$QLd$3mEnhLuWn3V5JhYjFqg*-++9yB{KO-ItAQ)n$StSEE&&1AEqG_y52vmST{
zBAPrgdL3uAL1KRrI1Az(x7TUkc%`$P0KN7qL+w-IF?y?lnihW=dj+I^yr#5x&p)kd
zRPy6oPjTi<n>HgJ?EvS23ix(i#75iz?&0M*YX=a&t$)5u_jJvSyWxf()^?WEzbF?j
z;%Gt|KDn4Cgl^FNF9`S_92x0=Aq+Y^^@oPctL2nhVC+fg)xHMU2D%=Xf9vx$;vz<C
z=-hpYS-*Th!tb``zqWS#Vi6iI8vF+CcVs$u7Te<u#565dd#xBd>f42>t1G21J=|rd
z!gDDx0?2MKcp;-Hu{!9*@)9EdFo>r8Sa)jmqBRGJO4v}x*CrqSXsF^2Mx&sEyrFk0
zsK^z+xZ)<oY^y>un^oz46XCzV#ofmJxu5$;W2ad-OnGbfY_yUVu^#@0V7;8cA!$D#
zK!G&nh>dZ~dNdY?2c!_dbA*1swVkYnQe^3qID|En#MmFgtr}WNzQyu)V8h(BK7yO6
z6%M%7FOb>P`iUX9Y=WIjekLPI^N`%tdvE~tj?wAd-e&=rnGvh;ZyO74J@IBYN@KlW
zIew->?|;i2@=a*$)w!EDf2(?P<_iln&%UV6Y)KFq{Q5r6oUiwOCM<VIc`<UTT0LsI
z^^G~Fi@)xeCO8(CS0oO>*z=eL^<|=0C`Ena=l$zvm>-fXyLa?y;Ec5*G;AEU{lOb_
z3x+TADBMrCx~OB5!gY?TAK~_3;qN@74%x}qs_UwG2~`q!96{RJm%Yt$;7vwn2XvJ)
z1oU-(oBz&$<DsJ`pXy#BG<QfLy#P8#YxiLLI|HaUAA!BWcEQnL`l_KZqsS8cjmn<O
zz0YM&k_N>*iwt`)&*NsSr^iA(M>`E9khQ*SNrN=v2^W!>mgY;wwsn#c9RKwGz(=6M
zJ#qcnSD{Z7%<FP-X*~R7;t9dmXc-XNRitCHDs4?yi`B=Dg}bNU)=+*^<=XC8F%5J#
zE{o95EWQLRuOc0P{&9w+qtQXbuw!gnO@2R~@-cYA^>wcl7=Jf0NFlTRk&nt>S(X^h
zx}@yr&J9_e_7$Ydk<9`r2zi|K_I%q13yvb6Y!C5{X{`pDB=cE2F`9#QO^IJZP9G_Y
zWnODeI*uZnNwSJsz{8p~|Ahi|Lc4ux-idcC%MOVzMw*oSH2o{8yu@viLI$Y>r?m)M
z>0{r5ic@^yj#ia$eAO@A92$rg=+sm3ylx5=ag<BzNn30%u3@&N?yh!<&rxzeEfH(_
zu@VW!IeIv0g`r-g(M&{h>dG_1I=3E6=-U<+PBwR83npI?dwN&|<yxo0Q=LrYGqW>;
zhNSy;Nwk)?Ricmne7y%d_(bb{xOrOYgB`mS$>)6L_MJm|NKn0Nx9b79=FN{8ixXjM
z>XR4A2~;*FRkVs}zB|pmfkTj{K}!o90)TXOt$ZeD_;lC(16AlIDSpZljPdKZ-e@bt
zV>Se5e%C>WtKkYB7Q}B*G+%W5_Lk(TRB7Bom8ivd!%mfTycDvM6R|4E;#!ii*!bLC
zaGj2w9!W=+u;WROceHd~zUgf-^$+okZa;QY>YlPcsV?>V>T3cW=O~z1>R9Vu>aYt`
z8NPr!3c+mAw4jQh<4_Rl)T2S2I!StLq;^=<27>4f;o^^J^&DzFDeZb&eDb_0-p{lo
z{dePE8wvT9RzEu%E)4qg^x{m4eK@$r@J&pasZpMF!wnv*4QS7~P$F}n#F(O$w_~ul
znR```|Civb(b1VHV|yO)vc=-)Dv4p)8^jF*PItjp^9Ur=0vccSbg7FEumbTaF8&ol
z2QjO81Unj5mtq3;Sue^44G4QhJxo$A=-4G)b1uHBopZ?9n_aW>fhOO=z(<)Dig@J7
z(oUCA3+o<{;KFB8V?8?kTLCfP4T&+{9EU>Q{dMf?9*Z5{PJ6B!z=0U`Yndh7?Qw@t
zY&@R5p$K4g^0kah-%Irq)zLXi1X;6c5~!cN1jz6gQl$_Wt5HUARWRV=Z7(Zp^)Jkd
z<X#Pn&lts+FE)|?^&(VRYX}WcfJFEOsi%-iku1vfuqwm?!a7o?PYeVI)lB8#N?v5l
zYNa~c%UbY)1t&!9o|3j~_b19TeMckPLjFyZ=;tnI7Qv3<V`E@{?=`hFc9j@cQ_I*j
zIv)r`A8nFV$N{iTA{zv5>CPz=r*wRX!%Op-It9C0`on*<!?ttU{jzO2oSuB2S-Fz=
z;)dQD?lO-#-j?j~K2ilWO+rljl71$>;)w1M8;FgQme5L#olgc+i;jO9eOQ9sa970-
zN2)N@5TosvwX6p8uI!!sF_h>T{J_k>M*+5`FIdut(WU1`3J=ZGUQ)Ze#cMUJ_w~1<
zfWsaYrq4}@dMc$loDxHTyR4Z8aADw)yqqOMyczK+p3W}-{_H+io}Q_4?_UWyOu|{)
z!fzf!aF#Vo)l&34IW9Q?4U9A{Ir*^IO*QwzL4CC+Ir?uEDKzoLb0mMUpy5|<k>1bK
zj=FD;2Y23TY-i?dO}v=%3>XktBM2_LtQ`P2ZJ{=sLu<O9MEtkQYT}k@#~UM&=zVkf
zj}OLuCYB{x!khGR4q4rc1a<Se7hWV6ES;r+O_!VGYAhrc9$UQhzl4<e9@88X)`BD}
z@;QH*yl|$`sS)D|ctk?Cp-ZVWUs%9DU$J*8%25=V61hg!z@Zf0MkQRJ2(QvzmD*V*
z_Qw4_E`cL%`#?PO3p=r!8<3`D&*GJ~80>FQB;C`S&9UDUKMSWQ=35_>NT&^4<Na@W
z?$$hu<Mkt_2>aSmsoP~YUccljY`&~bM2oKjeh5V)T4_55uA8}r(?K(wul!|><jiOS
z$F7X=ZVXm2-B%iNNb1rpcI_|RWfsE0A>~q~v|%cS^e2F369?~mC~_K4#@Rq-|9GlT
zl3N7jm<JGzZikE;%O_^Hz^?W4dG(-^^CguOF#$L*Zm8C|q8Zh1z#WhDvpQNRk%%Z>
zj~d&d0pGd42+;F(POGCiP9Kb>RhjkmUgu3>x8({<c7C|Q!v5Wvi9`nFo<h9YpL-5$
zu`n34yF$oYT9E_5qA<5t>-PcEMnsx*H`c#9*q-j5BSTK~F>`S&%Cg6#2Ub=PgFDL4
z-8)~qF*f=cAS&wgPyjHIK~69~Y#q5%{S-1QLM!Y`wxNYZsOyBE$#s9LD;cRLGW@gm
z_V3SAm5G~No#HFhwmFS7WWe;Ik9^9;;f*#chJn~I78QP#JQI(Q{o94n=}vk4i`Y*y
zCx>4$^<)U=R04Tjfef9zl>|Hp13tm4;dObhPzQf?6CEEZ%9LNrSe9sF-TbW)ULj@n
zuANar4WIO7_%SFhZ1EV*?HNKI<{BDX^{)rGIA_L*-+<}G&Yl<B(cTs=wA|%lxO?0C
zlMn%v6yc~}-A@9dUKal+p1T%Q03s1j{Q#3WQ`_5}()v8Vp_BG+N`b-|+BaF;fmMtJ
zR|N7?A}GdP(A3~qkp|qIHk!N(G$2Imz0GovplQl!+V3t;N&Xo8NbqMFImOj_;<=X=
zeuJ$I-a!o_%@+#-8h52g#vaR1OmJL;h%B%85r?oje;>Yv8kLZ+f&*$u9-DDSG1?!{
z4xO72n=hi(I_!jgujl-$NVjJ?1}*l}1`Aw;7<61s+)6Tx3_xfktzhlp6?}z?dToJg
zWPNl<sNil;8%K(EZnFia%Kf4H;Z044C**$ou)+2CVS1UH$3s%rw`h{i4Rp`t-Ff&u
z81`tO+HZA?rGp3fjg!ldA+pA0JzYF>C5g$(ns6+SFX(HxP91)^+Nd4`Kv2IW;TSq#
zdHUjV4t7ehq8XIp`k%(WxA}&<g>_isFDVPDwhamd4{5mZ@PTPGI`?<ot~AVA6|Umk
zVdKRC0kg+Mr>tY<AOO}icEaGbmu6bibw(Fdj5oaMtRDk>bm-bo><uqv1P@WUGk~`r
z-gqR3xctw39Li0Z!p?-^@4qQOSC+x^@t9fSSv&4~Fnrqa-4)mzA8S-kDh;w|2fLPJ
z>7Np{SZm6$tTXG7ahIUljjY%KA7OEl<Yfd6nv8aGWqE0CA{t?jX*|sRc1!19qr3B#
zc6o9k%DTXV<zb?Ff=kEauineCPs0><*c5G#QCoM=84RH2wQN8AWH9}h_;)D2Rev7&
zq1-9Fj0Cxn!eeZ#t)>_NjPLtj{dt=zR)AI?ba~(VDIx+ULd8kL1PHjngBBklu-9Of
zcFio9gmZTg<IR03%%MK`Q=vF(VH&T$0rQ472L?cb91bLG87E<!o8@IY?38Sn2JSxw
zvI!_T4R2q<m2Ast<es=Lg%b0D9vr6u{$q0qG)!I+JiLPE!;o?z=lMGhvQUS*plX2@
zJ9SSbYpfO|Y6j2A14L>%ZpUH4!dLkFbA}E>lP{VNE&jNQFJOZG!23?>3lHCYwAW2W
zQ8cZjEml30dWFKFg<rZRx4;V{+k*J>-AYVr#59wc1Urc}_eM&gmx*yz5e@ARq}Y=(
zEv%(ej4XA-=H4%ljlExgKhZAKHx&}`u@oS`0Ke%-z6aYK9}qj5E<pfT*DPNQ{XqHq
z2Y`xR3jLxkC6psm19G18upTTBkc@r;i-3B)5__@$G^aAgTtIJ5c@l#7sMeC%M~UiW
zj{c88{kk|iuZy}TR<jbT6?SP4R272K<>Cf5Z8+WibZy&Ph<#i2m|*+&L*v&Lv}_Ne
zYFLrpj5_s9L=+}zIA^-NC%ljrrn6yr1<ZZ(Djr?<Q&}N*LkB1>bO`6Ny01|ljH0I#
zdz26+^65^-^6AS0?raMxx3v;v5_l~!?*U#7tCM`(nU_&UlQ8WU@I5tum87Hoy-sV>
zrczF8gEcG<e|f#N!Rgtpe_R$&s29?ZH4v3u1ccSpkxXYELgd-g_xY?`&mhpf=eYek
zoZN(+SMLs}Y7-KzE0I<h5F%0Y-hun1vD=m4c%^!lnQ?_MvpK`s4rFtT4|vE5fCV0b
z;v=!xooXFgRdmBF{jZl9KCji7Pb$W1hF)AcsFh?;YR4*9l_^FVOQmPGxZUknXk{A|
zFmZ7T*?~{;baH@X`am;!tI@x67n}t=Uv={4mf9~Ub%+Dt7kFJtHZ|`M8)W>w9*V<2
zdVmV*=7ZL(qh>>gD=rA3A$;`u#4IR{uq0^pf%>sFt)8YH4Zy=aB&2x5JK<rr+J>hX
zEqfa11g#uEv4TL4{ZK%s#Sz0K7K;i)aZwsI)59HZg1sl&`>d~7L)ZdUW2w8vb5n};
z)2|&-TXln?{7wAMs@{!u0R}7nfN_hdnXSUiw{AuT&viz(39^iGH%eK7N44CQ?CK?M
zPrI%R+l%uDkH^AZtV~V@q4&9xnTI&QQqYMJR(wPw<_>$PHLQCc7K{CX8x5Zt7!jkO
zc-VoeS>gF_U5IVT8;R-Nvs~s^Q}Kz8>B!Tn$scWc8w;@AXsrTr!d$SDEn^!%0bKK2
zF^N>?u1!&y4E1klW*jM8B6hI(kT#-(1c?&n(Kgmz<GuoNYM14N?vZR|Fe*kssmr^w
zkN;YUy*pX^>ZHyn3sp@r(Ol|zN2OWbxL%KFz|5x67W^9HizcBVcfI(}M&KV`IPD-A
z*X$Z*7XpJ}^|@=Z)jDkwP{xl3XIQB;XJG-~Ik<;dRNY0?=NkUz#D{F;g6SdOmSxZX
z`BdZX&7xkT{nS+j<c0kWK?#xltacroJf6iU62(^zG>RE)-<_KP21jE-<8DZSkTb+j
z$7G(-E$nDJ+~VtzDcHN*M@r8)>MG>cY@`nol41wJDK&+IJDll0ZhR;j00nN?)|Ftr
zZ(_Q1Gyl=50?<^snU!gLIoBo3T6Z$)%P~hk&F*z-^Ob=PW!RAixBRSt@W3V4)n!|s
zKg+h?ALXuIFOo{z+`YGVx2M&UK4gFR3sb~N-{c=$bTFOC*_FI|b`y+Amwmj-ba9lY
z<X$|0v-{p6(9~)-W~g3yb6)<jn)nP~*pXcku}RKQo3|4`zE&+gSLb<yxPd}5j-vYo
zVxlLP`LN`+48i(+*N@Azu%O@;F6{)A_lmyl)0jDb*pcxbq@|<1@XQccp?l&&bY*)?
z7vIgv*z>Q*F-}@AzF|_M)>_`uyOI94WLhsWkYZjSQWP)14IK1X0I7*q`(>mY1=Wz}
zlmNK=0l?++k1EfWcos@BKF*^h9;oO}-w&D~9-~_MkPnircsL1|)W!iM;*GLz>N22+
z=Ye{$=IErEPIia=>rNPHXN#*Wx%p4$>TYLEyr4b#{8IS~Bz=#i5VwutFWA74YdVdd
z$Q*@l`;fJlk6Pu0laO9cWwQAwJxk*=SkW=KP5?%&f!hANn)202#kx`x`e8cuGxGb@
z_hUtqsk<Z~WxdR6HA2vTm4^HZ!|Gdb6wN_}x#oUfk{J1b+jK2KymUDVHSTenEmBvr
zmqqZpb{hT7o9omqlz0%3CPzp1MT!Z)7D4HP5#eZP&zB;s)418*(38D;7x(qy(g|^a
zyP&wkAD2Y+`s1Gi(c)A!s)FKpxemH3qsYPMAwa14J6P*}73fgxgProBz|#c9eQDYb
z*V#5ts14^D^FPFTHa%3IKM|Jtj_E`kVeIs)&0+xqp?Zl_f?BO;(irs9y^>;98MsoE
zgTGfHFN$`0?M8`721WKjWW(Os-4=^=DO^{AloK$Lof>mzNqk-52He?=;81l^SNFR3
z%5A-eCcNA76S3c>9!{l<OyMxmhW=nE{r`(^b%)5G&tZp5dM{rbF6z-x2FY^rmqFK<
zIIaB|El}xlwuk6a)|gIJ_2$NG#AV>&b#k_-Vax^ycA9JeA5lmvY&RsyLI)#Rzdu)J
zAXwu@V=<27j{MKj;xu$<@9+=OND6zh32)M%jQYq8aZo&dynhXkN~iSG(Y1iai?IJg
zvCi7)B?MjwW~kCq(u*j8+L7d4DNJFe4jeT{gdI>wZr_F;cps8hG$;N!pK8Lqk@P@U
z&&L8g#D0eq91)JaOdtV8lG7RUPni;iS9|}LM-u3!{ibknxRcnqMrLK$a>0RVJXp4J
z!b#J5ehj#KD`;h4I!i66n#>8287j2o1svfCKr!%vZz?nT$5YYwv=vXXxbzu}D@HIm
z25=13_JSl>(RmKue7!@B2VPIBQ!)q53#*~ByN;HkpNG|DyZ^8qq<+(5I6y!t%zKhK
zKCP6Ua`r3iG<QWh{OuB?TS(%&?24Iqff>NLQyJAE5(fSLBSw)K0WxpK`iTH<UmT;;
zZPH02A;OpAX1*E!025fxdU23Wl=&*%fPhw2UGc0vKz^O_-ygbU{?MU^@g3oDsun3n
zAxXKgZtSlVlJs7!Sc8ecx#;?P?xb`0Vk|a1_)+}}aw~X&3`ISIEL@h%uTZ-SHYrGx
zJTW&a`RUHs5Bb|9BG!rXso#Rm-t^WGpQU(N)j<u6ZaWLi7*|0JrVuKw<6ho;!vSLB
zHQ(&uTR%OYwC_&ikGn3BtgW@|Yo4nGjtuo4bNmX?IN7YY2iyBZYbJMXb|7fh87}^Z
zh9WrLX<O2NC#T&T&RpDr>D@Cpo*sP(EA(x75Y+ob{?tEdB2lf`Jt%Q^B9Q&RS<%)H
z?yxeaKTPy{qadSkd`wR38GjQCY_gC=m-=4IA>>4urtRDV+tPIDO`0C%O!|33hW%n?
z)%b}dOU0@*1ft)WNoCtDj;t1o=;Gy&H0%1%q_6Drj>z-z=;;|Hl4wE-ATT82Uify@
zCTbV12OhOBZ0CYOuO!xxni&y+f*;+gX2)rK;}w5mAO#+GkLZkZ6?LmaJlF|u%6LG+
z>A6H=C+{k|TZ>&&kdiCQY$Jhsbf?l#?EaFt<hnxV%GJ`gx9reCH6U!>%Smf`^u=n<
z&pW6alrLci%9p04%bi3Rsp&)*f3*7D4$>0;m3(N!s_$*`CRXZVnk@GvXn#73Gl#*W
z7w)!s^F$^&HUx<?@5~fWSLl<Bl@|gLilUK_@T6;66;pE1HFiO*a;cX2+bz4t(~>Nr
zB>0fKapw2kZ~Vl+yJ)f<s!FJT`pK`E8eRsfm_bI3&PobT_5@dPKW}UR?a$_MG*Q)8
zKMq|URIPuy&r^?kA4n}9P9}ysM!zdWnoKcdTUS8yZu-)ar_xL}1=yLf^?a*?+DWkY
zJZ&gzfa+k#TfYOVXVD9MvwZoh3R<BXzr@o!fi}p89~XZon$S`I+!wsu#*(P!{{EgQ
zV~YBB$`#M>F9!RmvJag})-)%#+%l|lHZfJ1s{;xO??C5!u;HI0FI+w)F7SPW;a-^}
zZF7o-J%u=RCKp~il2=#;yJ=o}!bRE?7?e)|V1rkHj49Q^#RduHrH!(uJ}ksrWod`e
zXc0B|1>`KcG$%W76B;>;!wKU5!p@K5MRHxm?f;yCeGWgr<1@7j!Jw>Q^mYv}aFfMT
zdVREw03<Bv{|1ij_;`|S#~WDuv4FxuG}$OFS5n`cTpr=O1U&ms#~?eS^UTi(49?^!
zuC=CE%^5O_ARM2Y1X1+5O)Qr#R*{Vu{r$hS(C7+Iw*tDr8+TV`N!z@o;??WD0Vvf;
z-mUve3eoy9>U;l4mM|y~fY7)#kSs|+vSgmV71JG8^^4*DY9RtWI1X#kDvh4==b`^%
zR9vgjh-<M3PSR;xQTzr|4fXOslcQsFcU<@Cf%i1<-{i0}f@~v*S83C`>}=#IT0D{(
zy3@tOP16*ca1c{#6RqF;o~nNB%FaApop3<94d--A%Tlk$o7M*5Ow>5YkQa>aML%T^
zeIb^D?2?=A25hsgOn(Pq31ix(6fx&t{|HznWT-q;IVwmEJkq6=THWiXgNy`+l><3`
z18p3MqW>Z0I#V$hlQjzd&y4ZLGTJ~tu6YnlrB;h;-;<zjXKHf#9<v$_dApcaNZBeA
zr)E0|u{H6%In!|R%uktg_Z95|{sT(UlXWhxnjDSJG5DW$J(_4nrxzWUB`Pg+lFfrn
zZ_uPok(53@Ut_A9oyjRyN1A=eZuE=4rQK@Grthu)TJgZQm-DyT*yl{J2m*Wn_pS5V
z*7WTO+dwTH{%qq<MuOZ*m0t&rw+q~!lJNueL3#}s4GSQHXyreG(}cVStN|g>U8~T=
z*|bo(8D|hzM>m#i){ni(PN8|7k}OYmo7qkpTnO3@%SquMUq0dgP&Da*=hFx6GerPC
z#McCcV}D`oe+q{Zx@y9~41>ZvOYUtwvq#A0lV|CgA_?T6t2!LjHggY9h)+6Ehu2f(
zgPyQ9zojy3dP{Y@LrSPkS=>le%6R^DX27%JdJ#6JcW)R=d7w4|?K+#tp;=qGju^Zc
zBLVNe{<fy(n|NoAxXnOtLvLTK3#?k>WbnQ5#~YYUQYOwCIe}6(iijdtyR3}sSz&Xk
z!>8|yk}OLonJM)quh{*4o3s|qaPY{BP(Fo+B-rF!J2PWVA13N!vSNQ&7~lzPtdZa5
z)3ol`pXXcb!L>!V@>fdX@0Kr@K7TUw>(d~|<%NSgy^$P2Dw%#kca6a@ii)xLe>K(2
z1MK2gY3r!jMN;MZ3tPg~9Du)|Ev8)H_k)(EBnjuBD{KEFL3JmnuAqPXhM74<<Q^IW
zxKijJ`H|@fO#cZ!WJu!N*7Pi^un7`8<l}5djikp@9E_`?13+wF>?rWsE78t^UWfRN
z8ESgud`a)20TAU`(dNcCb2qrssv@?Z)0w0u?UF7zfaj)HdWt6HAkN@2GzauprTc!}
z%i5L|+#^R>I$W2+jH&>ysXCF<%1@M9K%CEC4Zh`{-q%!qEz?;xqB44|-Ts*p{zl{S
znd;O#d0$_a{U*abuI_q;*xi36s57BwKMuiLJ>6aa_SK+FKo42~qCbjSoB@)m$76ud
zUP1}<+ZCU1LP}eBOF(CC8#>XIMm(Y<Ab0j2s1oXVgU->jRyNgN5Zi5}-9z+o@D|`c
zY)NmvefuTfX?Z2MmbAi%Nc|CLY3+TDD*Yj(Lw|tShq^ean=&f;fOFO2?nc78nKP`e
zl}j~wdz)^d<I5%2jfp?)j0rJ}PEFFXUOX&Lu0tVL-x2=oBiLzV8bB4}FQ;PxHSm&#
zcxzD9+(|?s<#45(DP$9!xMA~eHR8f^&+t31P}aIhj*qS<`fGHGM3I6b%oQyJ;nWrI
z4KD|sEs8%lkBje~h{8O|=<;^%Bkj;F(^GIq6QOmvnI64PZ<;MQaw^jcy7j^$ZSHD7
z5YSohc-MH#J-)mOHi_s!8z4<fHH~JqF^WE0quT-P*kO|W0B~;*e$V~LwMM<5v}J)v
zHhLV2q?_ykw9ax>Z@y+W?E~i~Y1nRekqVp?KRuVK6?o6%oMk%F)PpY*pLcTYx7TlT
zj1i5QoID>W+8T@;vAK|iK_=%FlRswa6)a<1?=;i6U7xsnp7Fz;P$il!7J#i2;TGSs
zL9&YtbBcT}E$;g#x@Vd8sHK{o3EPif898it08e~!Q1xU-4iKWYqUynhSHb-_p#sM}
zr)HqKN=%`Mk`I&38#RqG71U#-p#j%n1b3I$X-3>}C%Iht{5X<ifDjh2Tku_=>3p8X
zzw7Iv8B6J`kEc(3DEaSB%ThhcG{8~6)C8&##%%B)r^&;@sW3EE5924CfaCzi^xtD2
z*EHPCU$bgd+992{Ilo_&Wm*sJm%`6a@gS36<uimcEP%DEn>G+Xy^A|PiU>tA>JQ(L
zsiSB+bQ3jp0Ll@SjB;Bd7+A+f;S*8jP!=UTQM2h;f-7K5IH&g{8Sw9ApGoVaT!YZu
z+W;p>LT5w_xTVZ36h(%O`$^N)9$dFU$qHf*;`Ac<q~0j1G~q8FggAHxoos2J+11{Z
zb)9eP<bjki*Y3;e|H_evI3qzv1ZOUgZ)%C^MevD?aG8EKyFNvgjC0lluXC~W!UK)!
z0MG*YGw<ybLj*ZmHKL`2Y&#UM-_rMZ`RgJO`vn-SzC)4y=yEA>ZJsg!l-l7MnmqP4
z2_UcHguTERlb3M^;zK%!$i~rYX_=uS>4;I{cA*|)-*KeHlgIspZo%EgXBG{>yzJm7
zM-;dY4_VrYtuwcie8y?lmxS1V9<w-V!8U3FS!EiCUcx&{+@i!JRZkoMWS0upxKyVw
z((2cWb!6b74gVRW_K(&5_9$OM^CxfUsr3)<qqib%ABe1qAF=zHtu2@OPGqVR3j!UJ
z;;;Jb8A9A_ZSKs`>9fv&zoXJaOIb#D%~%s!z|3&of<L8|ZHBzipH_^OBx43U1Zb^9
zsd0Qw2=)9}nJecsr-&1X)vE}p&0Bvkg)_*)4s4=VJ4&&t6e|*N2rdey?02`tE_wv-
z{UH10h3?u!fuN%Z$&*0w(<GKMYJ+`B+eE8>Y>mUq)@gIf(Z706Qdn+Hzo4D2H6;B}
z-m(V;k9WVAxrlHgN-l_>I80!e{An;RDA}+y3n<8*0$d<|OtobLDr9YSunW}W#sNO0
zE&-D#F!qsXvEV@XLD!+z`pBS%Gs9Ye(jww2IL4jQZ*}9Pwn@bKZ0!iu5SLI<DS*z)
z8~|4+%Mkt9WQrG1pZk*dw!cSHN8HgmF&boCHBu~fs3JM#>B&@>hrgNIdS7H`5NwfA
z#fg&05(jTiM|fOJPVOR(t;?|tKx~ks_5#}A0=EiU_pN^M%z29A!r->FeCKfbop~7-
zs~*GK1Lld2C$bVX53@?wJaKZw$yLQTR5D1Ml&>@utHdR-QKTrcQJ$k&pfaB**u5Xw
zJ+W$nRkhu6>FeJ3^dQ##H`rEZXsPG&GWx-h^e$CvpZG3SSR<hYrKz3){fyWqsvK&N
z;R(sUMon8%bnA<eo|%T`fP!)M(v$tE898M~h2*HimgLN{(_dMwm|m6^rpjR<TC%d^
zjnbYzf`4uVOs*SBFLKL&h_AeqIp@NOV5UNAzjOa#X~WN@tpc(B>Bn~9c695%d0Paj
zFxn*HN$`NxvIis8y(e_>I51!yhDxVRLTASP*ninJ5=!@)OI1`zxJ{Wc1zcW${!qyM
z_c&sPL6l6bFAd1GYjj>bcp=5{G0+?8Kp*z2$%zLv5z-1}Ya!|G`%4~+>wR{C?g*UH
zK;!;7Y%h!!6?Eq~HX-{;q|>RT&t~_n2v{%aJ${c?SiM4RtLNGXr3$)Ie6h8`Ugxv<
znOdWX)5vBLNCZOZ*mo4*ZC>yhGP999>^@;2<U2Ag9U!j>Fml9$3<-Tf@NP7v;z1l~
z<H2*V7J$uEqF?>tDK9yd<WgRg*1jz0VNph-Fr^KnvCWLT+A;vhz-)*7`+3eIxYSZ9
zn9!SMtSrqil_IXkd~irMc5Ab2zXtNbV}I%IDX=lv2S1#?(qJ7v$U(m_cBgkAsgT7Q
zq`-phjWeS>b|L9}M6Z9#N)#b*V8c<TJAjevUO^z0D~)UFH;x<?7MXAzk)`7c3(J+W
zCnE!b2{dd4x_R$$_bW`&*8j8sYeK!BeOCkWml~fYyNkgYBp5-6cC6M!gWeScQqp<o
zyi|a1QX+pkO$SSjy_ujW-fqefSW_R^QP0%D2@aJdsQvsT*dk*{`>qFR1q)GoxNnPY
zGN&ibcf3hk{o~Y_HU%tLbCXNqKwl6vC)*X{HFNiI*tnsZeEqSb+j`Z5E+l4-QH7r!
z4?1@{NQ~5&5xWN78V+BR>j6jFdm;VZ=8}mdvO)m>9I>x}qXYpedA(^j8l~7@%scVz
zmV`2Uv^k}yv`hwN%}F5GO;YXx^ab!70-&3Ka6YRRRGd-{vnZ+t?E=)vAbXJ28$kZ0
zF{GHkrSK=25;rG-r7~(<K`;TawJ4>Je)uObJo^<zpEX5%!YHq5P^JUVxq85OaAp{O
zMiozl=aVZ_+MMZR;x8p9)v~)I_@zchemVm<P1flkhXNP&C^UygGjJ!COQIYPx*+}K
z4yg+`(xp1E(WNrWBB)Y4AgHg`4Kxz0VPiS2SI&A@MpI({RcgT)Qh3G&Y;+jc=>5hT
zZ+=6xunPB9oJl!%m}KvJLI|f{ST+O_>tj}6YQs2@6?sB)PvjkwNtE3aV1x9J@2`u4
zyyM->ePT~&wF)LeY?%>?bt1@`Vs)0_+D*M=wCGjPtSTXxZzylkykS-o-IMk%k7EXy
zp6AwsweD(;{GKq{NJ|VSzD$k+cc(w{3AT*pFh&Jo#`c-x6js;9MQHdWF$jC2fKT`{
z6;nlNj9UuIXjHjgmL_;>YtTIB?u_nq&ichY0byG~$z%@@kIu<BSp^h4Q@)<!mH`Ns
z?KyuN=RiWp-yd(;wJSkT7+ucegY?e_>U2iLfD&iVyTyct2otqCX_-gipLHf<NTYn_
z$pb>cZ8EJ!ZNn=k;$R1~<&*GKcTf=+4VAa-6nK?Q>h(aqYm3}eA(JbU3Bzba;Jf9d
zrp4np&CK?LBR5TuMypm%;%t%@5vDBIFW@-eQ^gcn>H*(e^J~BMlm^U)Y}%DBQVb*8
zKUF)rBaX}aY*;%$OT^;UKSf`4Ly-lbFn@xsTSB!(E_(w!s5HIcK;t1TWQa)M=PjIZ
zUIzuv517)G?Z`49i3^MoeWBsiZkig@efU5mVgQ)ljZ(0z&&7()l7@kr4#@XQryfb{
zQZD(2pV4D@X7oa#$@SM0wBWZlX>zMe&u+-wIp_3yL|W1R{xt$rX<N;`Ul1)>nJ1f1
zcMs){j^AWRht%Am45Ju1E_&Q*JX%q#0KCNu1~Ls45@-R$AA{G@5pPMxwlL3^4&S?8
z)S>2|qP3ijJZlnF64i5@&3}d0NX~$cAtX5R^1t;-w)BOHdPRNn&t9S2L{0UKHx8&l
z(a_l7;+o31R4^5;PYm{32MXf<9-F7VkJQn=cSuf4mcaRBjhI567tbeDl%5@^i~b#8
zrs3e@-#@V7Lu#Ab@*E29fv423<T#_$fl9QOCQyEQ9V3b--R1U_$G~V;Z5{gu&*u=U
zZB@HAs8F5_CKPt+71{J@COE2~xStoIjccpPtr<`!zU@#G-&$9<)3y@wb`@v7`_wl!
z1_cQ&)i=o~c3%|4{qZ*vo!lbN@>m%$8iJ}ir}}<C#VnvGA1ylm6(-`y^6(5Hfwxfb
zrQg2}V%d{cMIXDg4)du>aUfWzM1I=gNU^#^j4|{cH~W5AJzTC3JO<9<u?UQ-X}vG+
zG~&j<3pCv48l6P;3>s0Zo2a5!HIF=>Tj^Mpm&GUTv~<#C2YZ*T5IOer*FA&s@!Ug2
zwMWcE4V+Hjck-2Q59e^V(BQtNV#c+JZ`&EX1t(6WMLI|_c&bC;q>Kb9KBOVc7Z=7Z
z4k9r2i)f+Yq&A7x68^bvb_*y{sN1~1JuZLqZSyO2+g6vj9o@ayIj?yxoN~jS24lod
z0o8wm;Jpil7;a1^i~x_wT)y18SR66>x-ps-d|}O5jIJ9Dy+4mUo3l~T<sxk8zJuct
zAK*|}Wpsr_km~&D<cSeL`hmQnl*RssU&u!;ObGII;>8;hq=b~Rsx;#?t^utw`RJlr
zQ3lj-S&;-G?#~ogURTi_8tOR*ksL-3V%<c|o*SbMqz|j$xQ-K&EEJ@7FGKS1KK%#}
z?X7O8B|9k%({v4a>wo=<QMlpjuYb96+$nOCO3x&LF9iF=+%e8kkbNJuft*C!aLBzs
zjpR6W^!a=XL-jWfvG7coLedT&k$<l3xjCAHJo+|uCA)6Nd;xS`UQeCeiT|MR-=bN4
zt~rNg#YFW+-F+ZhSsEDN4J~0JA%-aSv8Evv!SgmiR1+DS`Q?#kMdj_$4@;J%g&I3E
zri?8s3nx?{sRO=cCee4ODNzx~Z}a=<aXx^dNZ@?E?AhFb6}Gt9p9b+5spj}%1b+4Z
zDe^~*vj#b4$ePY$r!@HnP5eENKp1lTu5PE;!%PC&4XHoZUO;QkdI@&6|7q7<&4G4Z
zZ4P@k#53Wa;M+Ejw=f`cnWRhKrV!3I;la`7{Z*A=bZf31td1w;Z^6M9#iCBu-&?OV
zbYMA=!MF@@Bd+S{&v~~}PYDxDc5KVzaCffHZQykPX0`G58@Y>DgAXakz>_(|R{V|&
zw7j?cO^P^QE46@^jku6{sQ-TanYx3;yDcG`v_<pb`%5ay)6=P^{r-czk)<y>&QC*g
zs3aP|wuHam+(lyR4Sdbz%!7XDy!(BYJb<C<)>04Z`sBCv+_?LTpW4)a2g#Xq_DfeM
zW)YmWFMh1F3o`?C`2$BFyH$EF@$o4gBV_3Q^uLSP0_e)#lmBSICNRH1xx5j_Prhg1
zGr_!HK|PzlSDi)(lJ|4M_E21Ty1yyW5VKnNCLX7X*3oYCo?%d|HNm+0VSx+4+Wi5r
z>R<En2Gp^Uf70M=kk=qf5L)%bTmER`%`Kx~`L{o%QqNVc<m8eWr^<_I1@+<Rp_f)@
z)*Q1S3cMq&95s)5*fIiVAQ0`mx8w++Z|N^aO2d#}4@|c`UWWoe*i}cyj{EP)Q2wD&
z>G6uQHN*XlH<K<xpfKE+9F#@4ND?A{xq(Q@92*=d7;9Mx&6DbmSuMp?>B$ji;Q=kG
z7B@{;f(IJJ9}fQ1gUWOn)Ew?EM*E;9<j#){HLnMG?2pc$57q1vkXlH&1mqxKBc6DC
zAXgDA7phe~4*LH>bQ>Tt03Y+0pw#FSFmeC=`kTB+HV`|%>X;{uxWvN{0<e-G|APzb
zc?_IZX*>o{LXCf=-VYTUvyh)W^%ycmFqDmRv@tcjnv;=MzSKEHY#*kSA!ut^0XjMK
z^uL<hv5hs=<n6Ym61yc{x2D3#-$A_U#UIv|QVl05N38%n?gIOZi<ca-NvAm+vKg}8
z$JO%+M{`Fm0tcc51%}V02X9js_p#qv(;Ty&G7~K?O}oo-0K`|n>cH}VVH(zWkXv>Q
z0=g}>|0U!yUgt~PI}V>Il7|!Si$pMGWvLUZV`oBQ>L9_w3uhO9G`WPoQuNH^B=e_D
zpcFsc`!~?~JXSa?qI;cp+0sZ7yyN!sh_F4N9U+{|5Do`8NrLIQ#pT$ur`1GMnAjLF
zBH#iMtP_7I0giF1S`UPmv8)*%K>Tb8;;)D8Pxa=t`0|_$%1g>mM?qgh(ywW%I;@y8
zpc|CUXUtg|vDz}Z+hrcJsN3=rBgH31k{p%&a@{uj%#pql7k2`7S7sQ136!bOMM7ks
zSOAsQh3N}q>frv{mL3q=>-TKE*s)7@f^Z2eMt(ztX5Hz5$h#UlH#)RAsFkwD6CU2H
zHKzn#fTmLG(+?Xj>EKEaZ07$k74eoK$9Aaf`M6ny0#Na!o)bW}>*XH~nsIzOj%=+x
zqTOwc)a1=EJ3Gi$I4hDEX*t90&L8V{6XjY&glu%BES=v#YESAOXd|RN)q*DDP4G8_
z7e_!7n_!0`X^k^S4Ymx%%fLs`PMrCV#C%i7)7!(7g(#plxD_Tz+ook%C6z&W#Ddd>
zfKM=d(SO+9^Ch0Pa+MpjAp)c^&n26juC$J5r#8S1m#@&mXMH8LIGg@{WykaR9}W+*
zq3ssibS^J(ALmfKWy104OE~RuoQ?D0G2dL(gUh?KRlVmP4%d%qO0kwlHguF^imxg0
z5fljxgAyhy9uZ~F*+ulOADNt>S#T^p(oHN7Xkx8lx;d^F-zv&%M1mOy?B=g!)a*yR
z8H+d&@M;+G|A%5B5Nq&OLrH0vwan9cI5+s#r0}VYAA|D$+?XzYe&Z3rd<<gr0Q!t>
zrqWDjpvU$}V+|Gi!J!mMN(N-LY$McT<AhU=E6hiO)HP3&WMKe=Pnh)aR<-l>!850K
z*gt2B9VMg@M4Vp4AsgWjLW#ywv+e29Q>*ELM2byhL$eW74nPtSDk>;fKvB$c$hW17
z{6`?TI()ly2t;nFdhlhN&WA<>e|or?R9;)^|FsArw7NJRC}XWadB7DmzHF0w+vJ!%
zEe7SL4MWx7S|g05R<BU=fNPVvNid(%M!gsuf1h)@ZurActr~d7E$_6wPs*vW5jSV%
zqut22M%_BRZ_(fykhaW0NwYt1-BHrqi&F5h9`wexnB@nZVk_oV303fic>k~Y{Qguz
zt4}k8+*1JLP34#R__9sAF?VD*+5+>w(;W5$bjiVo^c9dD8ZvTfR<2<`GwMbCo&ai&
z4(_IZYc0VabOslqMh&1C91=bBv{2x$lp9bjsSjMX0;4hhCZutwSaZrO+C?$V9X4yw
z3BeG~5&UHlbmV5=EZe979dwmrjkc>MjrL|#HP_z~w07@X_+d{(`kj6Jiefp0<pX8w
z-S>_BHO>wYYgHK&J3047u0h@k2!vt9LR&rOzh;MQ@Sh0Kq!|&;0d3yY?ppgXkVoy3
zwmO(be*L$d(-gR}lq5}JE!pU7-|n3YYzH<KJbg)D@sZpk9C#}9oK3OXyI$N>2H>Cu
z7b@1*=9DHbT-^bjU^gPj2Ra#oSJ-dRnx1GQv}y|^H&K_1Q;95O<E<<&fw*)Tl2#}G
zL~~FeuxTco-~c3HSoy7Evr<5F9EzK=WWP3l2r=x02I|X;N@P1<g~z+LPUkB4LS~y4
zM#rHiXrP`3Vunob&;N$5>fVFws~qJm2~&;#X8t-BH+Y2`yj=e$o%{g;B{cRJ&qO6z
zr!yY)qzH^}S}>sA0P=~Z1WtS;NUQBJI!n~%E%}^CgtY^aM=;;rMh|Fa?oCt!m{fcr
zbpx@u%6dQod5nl$8v{Q#=&BhP3lOeN&+v4javov*7LiEwv{}nrIL>Tzr+J8J1rA&N
zjysK~a=Ks@vgPcR3Zv8o7=f%qme+ad*cde+ee!vI@w*0#gP2LFgI^E~{Vb-mhpU+9
zi&aJ5jc0CSfI;P2X0V!rHN5x6!Ai_aFW;1@6dcW(5F#;oTcg9rg0uiRd;C-`{rFDB
zr$chbiV3`f(JC<ciT)Sct9?~Wp@UwyN0?a)XbJsZIuV%xzdD*GK63JJ&q6}>L#)ew
zhCduV1)?S9f99@QC}qhE9)-5-+Y5T<Y!4=9iyQy)VZJz8I*#W&m29cOrYPJn2;jB&
zHyoy@Rn<Gk%5|LwOdw9CWT*iBi16q8?2~~X!H&6?br<xD;*+D-mgbppHAgH;`IGFh
zmBvjsFf#GG;Xco@)hr+fN&uzdjK-%0Ms?*fj`T;q`Btb7HhfNldOziU#kSthLovl0
zE#=WRH;^~=KRW1b6*)bDn6LQ()noSO<2X;}uG755kq1e)oQtoWpEWw?YdXWFPcH)K
zRPS;WF!}c91cdF)n33=`5b;&dKr!8@J|9g%(=agd-sYkbpatup`oCS*x@dkaZ-Q!n
z9ixQ10BtftNC9m*9^`l2T%(lyppK215y|Xcx%22Gtly=1%E8?yGX@pX*)R@lX#7`L
zZa$t*kCu0Sh)XQ=aYj+G7fSN;4920kq_~%FlLnGf`UT7zUi-V17C!@s%IQG<<CWBw
zEEuLN@I6-g6&K#>r7qTALM~f$K>BSu4CeG}{vXr70~Uxg@Q@Dwi<vGi>7mxwPUieU
zb1%bL7mTVp>Q94LGS=z%@&0{fPSX`BiDWzXE)V|uebvY5^M18J$7~$6PYO`8Tqwlz
zWrz!6ry5Xt7Ug9?MO8b&F}hJ$DIt6&jS_(9H)MY{LX7+kqg@DR;(Z2iCeq~0dXMjX
zFk=lFbpA!9@H3<!{+fAdCeNc47Xa_Uj&?lE^X=e<mSNwPHm1TgXD-OCscG@df92un
zhl0PNWnls0ukt#bJlO>k42%kg5q9q4%q~2P5Xk7_8Ez%Y4Q`hRar2OLasb?>3|-%Y
zAP?+4zaN25kL-v#2uPfhnYP2{{-t{_Rc`c-KFXMG=KAYzNB2Sl^CwthMTKf^4hlgC
zJ>=Ap&A>QboY)4Ub!U1Gsvab1mIdDF#cnwdpRalwoG@k@LD~#{a#f=(Oy(LswJe>B
ztd+qsEX`*?GapCLS62AO50-(({<qQpHMjUaXRv>bg7O}iAHOB-rp*`ub4x(dX6_FP
zsYjqO&Zq34SoJ~w^okZU=wy|T+B@vL3zxZeVJI>NcDwitIJ`yyims(Xay6O1jp%Zf
zG~+SeK4ODI0oitJK7lA@s}U>3hIw3DJyEc<vF3<7V&ra)R?d&}0^|M1v%U|f1J2+7
z89d)tc#?DKU6y~&!^f<c50u=89M1+K!T!$5FAYB>Oc-5mZ<fiL3rlelms!#4gY`&b
zpHa@fqWJ%njoJ>Tbd3jHIz~4p&5xU|%1M-8xnErXc$BZDc#u_Yr*vGo9_-cl{kZ7K
z21DNaNZ7HzN<D=r^^a$a2YF7=rg#I00P8E%DkO65z%t|RmF@8|aV;wmUb~@A%c;s5
zaa_StGjtEat{4XlzW4Zti%pTwXxRa2#eUqMTjw(zTgS_0&Dq%3pe)m+cE5)Nx#>DS
zVS8%!)f?vU=^T5KpC=<uHsd*<bgX3xDa8HG;0Gp8uGrUI!9z23na07%Y;2O+bYED3
z83V}b2lszS^KG>FCx)BxdY%W>E<d&XQ#aZdf&|TN%i<8=F*I^3+)aU!_jLy&G2Z$Q
z{P`OC>yIZ(1cbobNK`eR<PvL>MWFr#itTmxvt*BrX1(E4;#x-K((n7f7JL0Aax}29
z-a$!NKGd0B$GFoKf+5BfwBZ(PT*Sp>GH0IY1vI?XZwEUKWUP4Z9dmF{Ypr~Abe}o>
z=do!qg}1#rEveET7{$SrPM;7!w1D^)0%2<xgPg?d*t*vZPEeypN9`6-c)(-nV5l$R
zrk8=RGs@2bz)BHoHd4JXxk)X$8J{5ab}Sey4UYtXz|<n>l>~%)XI5aX5!IZyoV58-
zIZ7Qt7l184S{QrY8-#JjDxEwyN3pD(57ZNB2@HZ?dkuue{8X1Lg2ta_LYNV+O|bui
z*-T9m=|lN)<^Qnt=kZX#-~TxNu<u(!WpAOhAZ6bol@uzXEMqCkUa}8^l*-ml*0dmd
z$<B<WNY;`i`%<>SjK(w=Grni^dcEK8&+qo@FFkwSZe3lEbDeWP_j5nzIFH}n(*1Ak
zNJFw!lWwbH&x1n_x<v-K9e{;D02ZzY;sH-ntI~0T6)gXrSpCtLm-b6wKC+68=5;rX
zX~p=TiY+;Z-&pLr6MC&KLNxvbUOhIaNDRR7#MZ|bKU$s>(&5_5sTi7j5*g8$M`=Dp
zm|oc}9D~REV+O9>m+=qTugYkVO~L*d)^YL=@xHl;qV*MN|Ha^5&;wFckh5JKC_a|H
z7y&AKA&(i~5#7n~%6j}$pddItmFkr*U24xWkv8N0Pk-!r<-(6<cRXr)g1WTj-e1i0
z{ys}z1pQm-gbsKgm;vC$unnv1I35QEo>8vq&kU^;tlxI|s^^E1GPbulD^lztqu-cv
z9Zt=%-1RgN4iDf(S?%DoswpfZ<TlWCowt1Ub+0*ibuZKLU+^z@LgLCVbLxb=1V^Kd
z{Hlz>nI_!iXZlpP7f}ouI$CW1*P)9p!7?RE>U6^VFFek8!P`{4EJNDl3H7Uu0vhXU
z4F0+RDP`wzj5L$mbu!!t(%_sCIm-q(b@)y9Cw|PX?E^EU4y~P>Hjh-<Tx_;27F<Zg
zvW!zDLOiQg2dv-fo`6hUmXWE$7K_$0xd@*OV2R(wXRcE{!FpQgRsX=TG~%`lR@56(
z^k$VWNbzatv8CD9PWal;7R;(Q>_l4k#yq>}yDNNV@ANH(;WHaQkH7=NxH!3=wU=f2
zf7$2#A<1M{VYz<@%_2NEE%+O2`*hE=6-22sn=BPp8j>kDpU1a(5vAkBSsE4|Y{`bi
z+GCR*$BDyIFt58iYm}oC*k)^Z2IalB;s>z8ZT|BlE})$!&u{<sX86s{Dl=%v!(C(w
zgsG-5slZXV(;RyQF<WHUH>sgKSstc@J)X)l{Vg)AM$q1kt*KR8Ey0ilP(f4W-fuR<
zr-k&7*iIDz!hf4x&0c8fxQ2kh85LAFv=b@fnI4SqB&$4GMl);!3D?hgT2KxQiFE_v
zy=6wiu{OQNp(C8g%!8Ook54t5g^#M+G9G>dTMgb$RNJ~1K7ux!;PSpKcUUaE?VBPe
z42%hbwQL43v6R6JGV493UA}tmNRhZ*KgLSyjy%)Ho)A4_Xfb$lYhz<+wC2wj6kU^c
z!E|@{-pc=&xfW<J<no##{2Kd|`IE$#<pt`h|21`8%y^OHy@_MY?+3k|FOje-u#O~r
zv&cyJH05#l-s;_gP@!0W27;zOTbY0n1Z7EvQG|=2rs1IZQmv*ZciV?R_ewNV9gv4p
z;|A_0M`E9eTAp`hicxxnk^!cq&6@Q}expP(v2=M~ysUrGgU1cKDx8Q_mQl3Yrrleq
zZXXW9Ew!xz$HY(@Cz`dD(kjE^bKd+P<oo>}V-_xu8!&fNO5F_DR+a&`?#X*?vsi|I
zbR+UVt4EOMs_G7LJL}kwz{UA;`f5V4VVxFS?Z@IBeer;SiU9eKTJ%l3b*U(V|9t6H
z`}PJaNAo2~NoS##s{FoBbu0U`c%y9Cjbd5!?$!rqBQ>uMatVG_)cyrCr9fndAdWMl
zN58lAof+7YO&67$+Cw|I5&e7RiW%RJ|KxUzxc}dP(^lj3SMH?6b(nr|HhMJS==z$n
z(>Dl?xuxwAtcZiV)El_<rCpphaV!V7%g6lw0(ADZb9?*>e=9+pxqYCq4|ZT1^}l}b
zVCHLGLVeO3s#~RJ=1(=iMW58>6?y7}n*in)yKeV4jXFlAe5KlO)&Z`$iZ0|)hlVgI
zB^Wa)PL}`K;V=Z>7)}oq#a?-h$CuA?jEWa&O1~(9sX2KB4*0eYBtgPF;tiiyEK*Tl
zWnfD@i>Q(GANM>+kHtq4R%`e4Rk}MfeB7XWYfbVh4!59sU~IDytfB9%?Df{oNRANx
z8_8vcoCd#7DdY`n#T##j^alyoiP2x_6X4)Ziut^nX&G)~Pa7`K3lg<sO!WCry+dtB
z`MfVHQ{d)7sS7b^fsXe5!3Krz$ATCr_2NYhH4{dYq<-m$l<R9>31=ky^;v#fXQvH*
zZp2_$+ibkERT}N!wq1@tZ>D&1)@fFCeLbf(?O@VWxut&FK%q#C9fR=gd-ze|np36c
zr4T@{3CbEAR<@0F_Cd`%DYq7?iWy*GzOv1cg}v>W@9|k~Ga%ghI)x7T`gww*HK$l<
z5-8g;ZeSRPL_QJp+jgZ|b;$L<DTBEQSq+5d9l^uan<?a1P+n+?fi{7)SqvG=%yZ0C
zy=K_OKl7?AYBJo&0$pS(Zt#13M4|6DPKOvEVgdp|vF#|uj}rrs2=5xuk1OUePn^2S
z3;Ms_yANhyGSp_sW(IL7g7o_c*otMjGck`lQ6#=q;O-#kpZ(}2d2ySU^pBF_f2}yX
zL*f&z|I5is)!Q9=-4s?vtV>r<EtPO1`hJHFQy@R-0eQuYxTwty(C4g1R-_b2px0l8
z>B$|)J5H30aYB8Hti)`yK%b-p-Z>K`_hnpg$lfFG)y}GO{i8(dFCUKe(zp|fF`H%L
z-He4+`Uh8o9d}Xc$ltqJ*cyhZ`!kY4+W!kEiwjQ^As_tNv3bR?fFuqkge_{%uvQKj
zIvA_Ad*@#Ioq9k+_?tD%pL&cwv@)T+4t678aI{@^o%x3c1k$@-JM2ynhXn{fd+yO_
zNuxG5(`w~F-i9BxpN~1NL=k|w)m6qeOy~RA%u|_(4^*%39eR)QeA(o1hBA2dr_2*5
z3Mp_=sj-X^xTafrj<+4|Z()hZ`mYwibE|dve5Y<*+6r4Qh`KZxV6xGP^+R2iQ4TIY
zZuTEr`Rvu>vH_@n;`wpXd8=fn0Dsl(f=w)ljJ5&5v0E%L(@f;SX62JHr90>Q3^;a4
z9~K#HWei&A6YH)3OfazfuhCBHPYt8~HiaPo%qF+w2>q>pXrTV-Eow=|by6#<q45v{
zTE6C{R!60r8GuLb@VkC}K+&gmrPzC{iulkHbN>=le>1)ga7U2GXq6cPlbz~^>O&be
zi3onJTo+iEIQzEhHlft&MS=ulfiEY4IWquSF)T(sq0=+%Yl40)XWG$w+)X0@TXz4U
zIEQ&Z!#5Iy!(0b0#jRg<86m(v%-hfh33@6eSo}2Lv42<c#}p{>!2ZDH{ST;aH8?qF
zVK>Km2t#cO{@SBIwcAvD>;m_ij|~)T8l*`@M}%kNDnISImg;<wU{<-LwC_J$sGw5x
zCNzFmzJ-mTPD@tjg?FAU5NgJ<pkmn&%z)XTZ8iXs<p8sier?W5e6Q1Z_-UinFl~XQ
zKLOD<L#;U@-AC-(c#I_OIq-V+CA&kEncr)$!(LW6lk={%mrd_xptNv&qtHMHTC(gu
zkl2OOj1=+ahgH9DhK1U(^wl%te+&B-t!Dk|+FLF(1sth$qNHF6lwSK7J_~;3-b~(=
z_Yt0URRck>cUjZ3zEciC#yxms<DnOpD}EpHlKDUmG|^TtlP9!lrflc&@f=Mz@=xUB
z#FON|_~~x#a1t|pRpYX===i7{Jf*(2sf`pBQ1e)`Ue0GGvLCxOdqKi*9=b8_PrQ?A
ze*d5t=g)ZH70Q==T6BEuMl!BwHPU1MqXj)uTsYEHs2Ba2^9m^}@t@2!{H4p2`s$&X
zEK4gQ&H8Lnx2P-34+A#wS6a;<Xi`X<3};4il{!v@<~-l7l5567o=A4*=P8IfCRn!P
zJ>ARUef-Tqo#aM)##L^?A2lTE#P-@!1WqckfBZ@#hJ0AY#JmW6FKGSZ*~f!=#^14d
z1;ihpibk)<{Tu@QL9}F<og1(!<4|g9K@T=uII|}5nam&Mg#xo3Gu6T!dT~Xo!||=J
zW-I;q)V(HW>!hT_Wi{akL528f&OZ9rCb_K~--`9Xfd4bMAjic1617uf;QTX=(I?Pn
zS$y;5T;KIa5Uj$*FEJs+CyHLTBIoR3Qzd+ci|YMxGF|LHVhF_Kl5c&I0<moFxkBz&
z{S>f?`0ju`t4A?PkvBdX<u^ajz~RvKF}-bja-hTLj@#`$B4Z}k_McWLqg5Szq}}r>
zIe4>%bM=n;+2-8^3fCDCV&Z+lE#X{q@NFyaQ(uSb;4VHtlejD&!j%~G%FW{<Z4Xc0
zQR6022eduy-PCMueC^sMkV5#>)o{lof5Xs%%YG}hg}Hw`kCg^M7qTa}Y<xn=HxhJc
zAV+r|?~};YV?N4gJ(lqVsf1EsXCYXQG^aM4ow5h5c4gmp3Kh^-=E7w%{UpdIPqPqx
zA3VHk#psxkeGVf(fS``AUpw7Me?BOh?B|F{9D4@*7I@bZ-lUkft63pq!L#ECj7XBT
zAzWDfG^AearSA1QywVZ#2dN3c<i6iLZ@63ge)o(?rTvzYXY?-K*znxU(uCkyhY&br
z`JvcIzoF_Rv>nkOE>oX~yR<MH#&Xsn^S+A^S^(%da3Ri+=tx;A`9W9zoomo5Zhn7h
zz;Q_uQ&1%ex`MN^>xHm3g`>ePQjT$IlZ*-2Cs(-l(sz5C{!s^;s}&57yl5*b%Pm-O
zh_I;=8URt+AIXpZBj)Vyxu{9__bUA~TfSY)J+K1RyY;(2SuDw;2H8}G%`1J>5vZ0H
zGth?A@?;K?AGszdTg5Kq!Qy5(lMj8<jg?)`B=4W`!0lQ|v-GMq<I#sa6RY3zne*;<
zIb~p^JL#AGhV>-sw{<bM-26yOr^?0_iMzD&-T8Yb`pIhJ9aFLNH*PllWyHfETr1t9
z(lW(9_<YfwUfft_vXC1}o(afy?^C`%5zII!@t5Xw+jQU6C#KDGmk%1HS1~3e$H{-V
zu|3MEqewhszKeCvTkeQ7GfW4IjcfT&yiNjAnh_EGwY~@`mV)=$#b^`!x?*Yd#9Lh9
z2Qy;+Q9~f8Mr&h(WHHNv90$cv`z!9F<^l$;Fc(a!BZw|)e!qVt;nKEjTWOq`MkjOI
zMJF^41tKv|up-H)Pp3>5mAUE?S+zbg1;*$Hj%P5$6J=LM<oN1r{^s$mZ;QSDzMw1a
zB%*74gqqubF^HihdHHNNKaWt$R3FI`qH9@<WzToi>Q@W<7!xQ2q|-#0@dkfn>pn0L
z6c}Kh%Fuk)Q;r{buthBIWoH>foyKW`=+~A_tI^UjIC>XHk(I@ooxt{nbBSKL4ol!(
zO4+*e?O^D*$B9_Gxc;z{#nvY)Ny_}WPOG{bS=}MqdSPPI8rB*s4g!~wdv|c{W$nAb
zzEb^+mS2>uQE6EZ!PK+YSj#PeNDULV+YJ5X&s=Wj!i1Z$XJz1%^zp|WhMfxECZjNt
zN&y14iaPl``PCLr%IAvEB8NZ<Y2>%|;!S%Q&Iz%<DQKtA{_lmH2K}H4Of@{)gS;Di
z_7nQVG8Du&;V8L63DtHdcN|c})`olSJ7HA*ZNgHr@B1Pxdm$`O&WR(0W&n5V-U4|-
zyn9ZB@?jf0NJ5foq9aRvRg_vCNUk2GgqFIaTWXc!KCX?OZ_H1U+xJtcQhMdwvm+Lz
z1h@U#J*BAkg4`5|3*(LqWs3>Lvb^9``})x3kBooOeq*VqvAF5W1FNMIT&A=a-n-np
zJEDQUkUG<Ld)aEFY}qWavZ@TyDQhev1_A@HrQW#`<}Tgk$8iXBm-jQG@6>)Irz1;n
z;-Ks&Ur+}@H)fjLGCzVCWO92?=FD(g#Jc6h#k`vHvT7SGFQ<k{C<tCqe^wa*O*rI5
zf*|SZ?*lSU@>!J7NCF-5mk{jQnGYtp4{yo76k-fFH`#4QW=Jf?b(sPa!;EgA;wXK)
z=gn3iA$lofN>UnL*EAoytbrkO)rnKe1;Z7^Q#x0!K$lfUne)`o%~OB+b;Q+h-z{^R
z^%#91HbV!=IblCcgV9_I8PUUBZy}G#70kNc0acq&6e>U}hbyzh<-LPTphP#(DYv-8
zN%HrhPwF|zKJA0|XLmTs@?Z+6tD8G*K|ZwaEz^qK!sB-I?ZNNa!R&ab98+P=Nbcz>
zFPm%rd{nT4x>yh0`6|IjWqL3P^JgmucT_dSN>`b}osVbO|3TV?XKSwUn-zEcNeId(
zI=$DwafzZC$!f`b#Mcu8x|^X{rno(5Os8luEdDllShqT6t@fRQEEX~D`Il;%qx!1m
z$~~cLGU0as8?NLlS!4YyaoF@spI{7O<@JP?W}D2sqbO&24&A>0;@{n&DVE3xEStAq
zz`n8jTkN`bBdq#>_>~W!aE3xL+*<I1%A^a3J<Na{8vuPoG=1KX_B+o9v^IRrR?yf)
zu>#ik(233aMdl2UOJ0z!W;PHq^P0<lLIJY4hjJT1n9`E10_DM>B6G*X3olvtv*jAy
z`^fO9yI5`FteJB;pk*NzUg8h9e4Xd`7e*gx9sqgqbarEy(-+Csa1h3+ZZbzExCWVd
z!WOBUP@;jMw~C-EKMJ)C=v#YTS9Rr(T!^g(u543O3Cv<g^xOT~pap8(7}v!bIOynO
zS~p}6#{Z4zp8*-16nb>(&8a)w`O<j~9-g0_co4xRBiT<plODGoPeM%Dp}&_gFC)TD
zXs4%WOkPy8a>0qom@7~}eId-?7`+=uNn?af_aBEc5xPSd%zs7*fS&QD-GYn4;fbD8
z+lMZ)pVf{e4?LLpKZZ*j0!Z+3XH{L*f`sW{1_rU$!j$qZR+GVS6UibDXj$t1s9DSD
zP2YhWiIDk_8bAc7OMTZxEN%aJMgOf3LJaUvjZ25N6)#(e7H4Gv?&h9j5s^JUP_dwT
z(JgLxJ<01YP2ED~F2#9@^`Gf3m2cBhwl#qfAuv^2;6>CM5zo&^O}=Mki(19O&|KGI
zd)Ta6`mW&=a4RB3d7i?_p|Dake0wV--=t!FCg1l*MsER2P+QX~aY0$SETiPPSz8<n
z9`peWTL4n%pSQc4HpP0XVE4}-YU@(DZfG%L2|Ge(AQ$EMU7*M~pRxbi38y;`9_nA^
ze54w~Hj>*I5)i0dBSzB7lYnu*Q=XrF4!&QB$6cfkL6hY6C&bd7KOChv72nYp&og~@
zI0GB$8X<H1U+8O@#z4JE??J{1EaUtQPY1i_`T+}_-sr^$@}K*aWqOc=KU>aDnfZ4>
zg?2f{Oz?zXs*eeAJJ6H0BXQ=VA~^)^AFgcQFV!!*KkwNQea!pFp|sf>Gi6QAb=zAW
z2OZh3ULca%%N_5=@fc%9WUtCRZz;s}3npEBA|1cgmb2x{X}Oh0SC$etQp|Eu%L;sj
zJn~^AmqRQ}75b>$+%>&Ha+B{qL`h7cIX%&V_l)7a<(eqVY-g`#iU+%MvDA#*vM2v}
zos9Zz*p&KhhteJp4(Pl1P~B~mc8C7(HbAgSPxZEiE6$Xw&1VQpebuHM3z_K>Dg^fH
z&f(smda}=AvdYIV9rKFr3z}Q46BY?04K=TE(Hrs@7@<y1r5kK~>;QKe?koI?rxluv
zdkNT#mc-Giv4u?RYR$4S^5sX=2+E@`g(4@2BndH9k+yW)xz710J+bfkY|<dt{2c9P
zE|xGCD*QM!xNua@H%1`TyDw#mpe*gG^fcyu9%b5|#d<Wc(vv_DJ4sC2Yb-l^5Ai#P
z=-#^Q=tI5Xvqw8CQ(4a=rpRBqkKiyS?{ubqFgfrL)N3btKQ(#=;K$*R*-`#>r{}$W
z3c?=Ks{*nVMr_^M=xYtj@O|rkO3bCGdFI4Bn_a{QRE|@urbzMaBFLZGp+Yt6OMav+
zj@%N&HQn%K0Lc@a4GO*dCobc=f9)l-VBk(y9;^)yYiP|6BA!)yu{%{OFsx(xiM!8x
zf*ITB_FggU7e16#TIG{;-{?G?6~$1F&6!*3^0xd00cvxp0K41HrsYzdB)-!UFqGl`
z#a$e|zH4|J6lG7PFtwGxZn`kqFoqC+C+L=g&ds5`^s*oH1GNxb9YpP`d5=}rQ|8^4
ztfu~&kFG?D=%?%DL1b)uIz|oBao=CdJSntOM`e7z<sDrc>#OUX!#~_k`l2?qv%OOK
zg6inLT`L{W@SG^^mFBGx#Mi!BxwvUox4TW+TaUC9uwM%0p}KO68D?*d-v`D~S%x6T
z#d-6i!_vjW&1MMWYno5>I~R&P(6{m9*w2(48edJqwoC@d8q2r%#c#r240aGk-1AIJ
z)e`waN>~V<ByUHO_SaKL!r$-Q2s)FNX#RIO%Aq-Xg1TOjfpI?cJ#|63G>;<H`TNjc
z`k@umWRSgl6Zv+Mzm8JMH$3viw|XGBk*+bG+2$+#F`8cl*3qipcW}`0)e*H7m@d>X
zjOl2jf9Opzj3Cd@7Z0{+t%a;$wg6vpD~b-s3h&nXp_T#9CGj0$`ATV6;*Ky((5v@$
z%$aAyZ%LQq+xL^$k+vTH+UvR}S0<R#o`v(qwmQ$%=pI~c%D8vdy(yq$j7n0WFKP(?
z_44dX=~NXl{&UNUfYa(G==Kq6<rJ_bGRhIe#|YZ&TvI)Qp85EQdHm!?tVa1XrOB!~
z0Uw}56jsh^?o*w$MUd~yti6+9o%r|O5#&c}HX<-&&RP6AI*2yRQx_natb)z-2{UEz
zd3BXsd!yg7k3Px!gnCyVy*5SqF3%+MO{{1aQg=u2rIw<s-L{s(5w-E2>?u-yP1c!C
zi3Gxdagtr_#<&EI!aP*w=#70Zq@V1oQ>LBOYP;-=BBr60nN}X&D?aHhv%VLRb_+R@
zDg=uOnO4%v@*OUi;ciDH-%}3t6mYqXAlr_0HmF(9bSh4XTML+Vp?n^t^`k5tbR37y
zF_8+TpR@MvYA&C<F1{Hp)l&>ZN!Zn!;rqMxX@*x^#!>6Vt`hpsFv$#eda_v};g!NK
zwTfI_^G|d2HJ#WpEuX6K9!$e8gxNLtpy=a;si0p;8pghb<nZQOFJi(%h0*e+1`g}H
z{~rUj?8lwvqY9LZSU#P<X2uU7NbsGOPiXo44sI~eut++#poqF<MuYZtI4i#{*e5!>
z4M8tb?+ukgZ4B|@kNi5Ep5IGwY9uI{=~TR^()po=wP1TVHR)d)<k^9Q^Nj0^t{Qp&
z;8@;**3-+J?_Md65)X2k=FAW7m~-(AZcBZm(dXRia`7tb=%ss6f+6QmY03#o*$IXI
zLeMB1<9?j$A@)Z49;TJlu)N7Z1waN$Wj<qTc?@eUNVBEYW-oY!4ySB0!z<!xjphzI
zRk)D8*X|yBCVAox`l&z=N5&)o#W>;4ESgMX1cz8pQNE@!A!5f*Ev>VTI!NetY8AN%
zS9F@BhYN3slpKZBHrFt~cO&dgs25E=?^yd_rIbyl;?+o(xmVKENWvQv-l&8f;R)-t
z0u2PqiqV{hEfx{Pp<>LZN_hdQKpGFC9~jqDIx~04<|;=oOK9Bi=4T!qTRZw1w_J*$
zG-zuUOI4mbX0d6JKPBf<%q%Twt<PqTtg!W$DXe#sc)k5SULg;s8vSdR*zAxoCvf;v
zDp87s{g=4#2AOuX-UeP~ZIhRNp89>Ojl6=@`>i<GL(Mw8TylZl!E<A<@T|rZsVUIR
zpG^OeF#Cj#-T!t%i)X9wODXO<C)24F2XzJ0&1tnhXP(@pz0MmB_!u2#yW=nFvm_A5
zQBiedJ9kruTGu627mNT4K9e=r3ll-wNsZyp51{${pf^nNl!DttN@R<N7L3I`2$MY1
zn81)6Z;R{9r${9*sqxT5WH?CYx?#1xZpL(4M&ILZ<-SQRGQPNv=G8FlC`Gw$pSnVU
z4uI>bwG}hCliXP0?A^O1*IYM4DOyHb7RlH)LvR;AIFAZ#{rNP0cBUD?`xiqd<laNU
z3aq;%!#GE}N4GI8R!c{V{(RxypK87VCK1e=H-fA;ZamS$TAulLzYvco?P6@U5vGxZ
z8y@M%!xO;MjGe#Kn3~bK^JQ#v*J{u~jYnlW%R1@Jesf)O?+hyw8%U>;Dvh00Vp?x5
ziED9Uh2N@eJukwlv88jtO_94U+H^nfZ#``7xH4bdDvSm$r9MX1?*TvlaG$bIXu$F1
z<iLsrFCFb*%yy=guA(oJsB0dT@KgVZZK!`=y^`}#08f}uSpoaTueQmQ+k=LolQ9;^
z^_oRyB=IX1M-m>mbEZ0YloiT4IMP6&TCPyW^Hr(7Zs85xdWzR#ACqd$DgQZ{ss{wd
z!-?Mq?lF_d@avN-<BdI|c~sdak;-fW2UDBDeWGkk$K*>;&|&?St+$$`kCIZtg16I4
ze~EK=tn~x6xM`Vn!3@hpq*p@@lrW;=b5>%N&UKbWJNpig{OLczFVw_k1`~60M^bc=
zF$1pcX`>~oH9T*GWpr=?CK=%>!~vZTcC~uHeech8HDmkrogcR6PUz?@5lBgge3vMm
z^#PBsMxS9`Xo(<3yfm|wsd>E?Vl1P5^Uz~v;v>h6m6tzW|LsaJX<|k6tFV3^24ic5
zetV0hXdz96{}UotBgl()Rvj?C!wmQHo}Ya70@DPl&>{Qv7}broSe=_oKawyG5@})N
zd(CK}T1NYJ#OMBiU)xDB!n{Rjw6>Jiw0vw(@Dv&?|1&&mcxI~QzW3lBpA0FN_${e{
zDm>!WBK?JGk%R+QH%@7)3EK0Of98lj@&6}qG9h~(N+=Nqx!%bar&o6H6h`~@rDhuW
zyI?N4_hr@v-@;HtrzB}Q>v~rw8Hm=!Fuc;HPs}OQ_^*Js06C5*r@sTG&4wIGBe$)-
zyUcgrpFDEnIce<ZuVsrSsnK_ldG>_rAgS#x@!wNFa8tX5>4omTBI3ndSYdg&-I9+5
zf1Tp4JIVSjhMAbz4fJ4wm-k|2!r(;Sa5LEk?a73Ed2CD5kwc`1QJBHuVD)TV*S<`<
z#4XhQm-7;OM{Dul{Bj;b+Z|#X$A~kzxi(hIU!L@=4NtzLwr-Dg=l6JyJY1~RU_lP0
z%CxW2-tjFlFL_+i5HRSMFSGO3MVVOpoG=wRrFV53<*{`PKipczdaYuMv6OptFhSXF
zN}*Q}yk!etl$DUDA-rBN&JdRIlT0h)VKK1U#&P4`Xo`r(+-!i-OO+$Qxqt6F*MhHF
ztRqj-Bb`t2<xelG33dn^(%uAbl;uR-ljC!QpiQ}k3VGnp>@CXY{+>0v3+=yGYwiY!
zLO5>{YeT-HJ#aq!h)+w6SSp9C{i#W1->?zkj08T8CcoTvR|~Y>->cOl6?)C@T@nIv
zn+GWJYdU?WiD`Fsa7Ja6?hH1wK^rZ{&k-~tk~5W?qi8%+%DVG<Lvk-!5T(qBqZhpq
z-Px#nb-^>!^8Y{3Ewi2T#O6rbBw3M~YTC!kIISZamOx$lN{nsq-_0bGrZ~lnh?)KF
z{~1`h%oT18?>ow3<4Qw^DJKbymQpTswfY-!=>GJhz$t~1ZvO(Mqfl1j#FhwZFeSOk
z<UA5|lGcIHsNuUA8u!j;1TIdN)faD0hY7!7l)o_1cWXETU)}qQewT&V!nBx4_Vaeq
zF2Lne1pXm8wz2A_=UTe((Z}ZD>5zVt%L`eBJRTztJH*<V!rf(0`G$1{knMX;-FrPi
z)ec-xm(DmD>tRKZ=PO>Et92s`4<$|pb)5BmU|v-k60s2MD?+(mA24wxs^~S#_#@eb
zZ!AQe?U~?m#y)=k=YCe)C^u*)>>&61!^_tDy3aE&;*Pf)hB9D~{foWL(7BbbF1kte
z`^uj+ynT`$XSaL@LoK1i!`#jrH0bphfk)E8h7{hBY1if?8KfpoT^FtKPi-lUVsmww
zL-M|M!I5?WhJ+Y8Z2!q&dVFijSx)RUHSiq^<_s|{kgI;7Sx>5Z=i0fw+JA-$%U1LQ
zuaGH1O<T1;`bd>Tp4Fl8v)_VxWTLRYseK>c0kmPy=xv%&6zSxmV#OEJb)9n&<WKbP
zMUE5~)^nScYX^URa(fY$w-HI%h5UMdofGjn7!^NtEJ=)g9ulv^AztYni;U2nzEY!m
zV71B2Va@^kw4}$x!#4D1t<TOK!MjI$JofHiF|BGlec#AZRBND9n3Ir0PZ!SFzJp}p
zxS_C<kv{I&0eKAVW-r7&Qim`%GyeXhxRwK!```!PzT)k&>-lwST`sJln=c}9qIADe
zooBTMGjhKS9|t{xb(Yr}W2`Rt5h8U4dqPAUGc)*hPSsD}pPp8^f8*UsSg@-1sp(3!
zkv9pI0B^m%(#7<NX}>wAbh(*{Z|fJvGVj!{y<$M`2J^6U{TP|uNBuz{5zbtR#{<ez
z_MyGReNyISZ#zlwt^r7x#Zuq+WUV?t$H2Lyi;=>bynLl5s|rJ?L7M#L1z)E9%(U`h
z0eh7WGh(Ri6SmR8@cC|);X)0P_T!G!&mkZ3C>Ol0IjY1Q2H)FeeyXbIU}=NX3EXRr
zv0WJNGh~W_+{eR!Rv%Op>_UMxO<ZMbGdyb?K?X4B?(6TMbENDMmFLh0UuL|9J2_J+
z)D6K-GGD$4uD9;0`VY0Uzst%GiAj1qfCK4I)?P|t%e3e0@1cq{ye=VvN8m)IXTZ+e
zbhCa+jPKA(p0S|a-~4M=Ei8K$jQG}Sve5HfSh0#Mzo;>-r8RCtm%BwJhH?>qJG9My
zF}yCzxMa_$uY&%G*z4bmF{W2L-Cpt`;b9v1;-P=B?5KS4NV)6}a|{=5dOYO5IkQeV
z?!yxK_a7=1AJ%=fg+VHc9jklMJW`gj6MioC@mv{e{XXr;o;>%ofzZ|N>?{>!>FQ8D
zvOeV(Vk>+xTJ+(9152F8)WMOb3AF^ddFnuDP1c}1WP6rMF-y^HL_d4WPtB67{0F(b
z<k_OPFca@b(6MD5gDKtK<b{Ff@mxswaYKAZEcw#qx*G3CdR`8-b2b95`5L%Q%zvHs
z+e@=cJUl_Fu`2}_0K_4lVfYHii1*r|10AU_d*~t+6*B$^@F=s3-iyRbfw*O^Bq7Vu
z=A)KGyU!G7=_%EhJWjPz-e1S6oBgnP!3$f^@@8ic-8YKGkAu9dY|bOPa)S0-!UfVU
zZu^Qshcr6x*g<M6t9PR?t0PD^XTj&2YDuRH=u*PrW)h7K4x!8R!w~LJkF>B|{Wn>N
zaEiLOfeWX<Re7oz!f(QhqP9<uufjczpZ_W7q>Ii{moEHVk;VNXcO|SXi*I$=?(XCN
zY5}wtCv`H4y+uYzj4|QZ11DTof*<IwfS<SlVuF7-ig5RWdKNqdlKQrA4#Z(cB@^j~
zpPlJgaL!uD#0q;}i|aDQ^O5BC@Xk>0$KO!FIXLe~#yMV`8&4_nVEkC;9C?Ja1bHSR
z6?E0WWgHTt)sM@@NkZe}H{ztO4EIpwmg&FR%H8AhHB?GF>&wM7R#d<)->H5KRK<(y
zdc$)tzapcpg-B~2EctNNiW<bsur-qg`Xk>*Tf;N{48X)DZ-=z!wvM4c=xbuw+Y#ia
zHxh<M8_9lqBhm*^KUdanu4dBdMH^HMGt9F?5Z{|Q-nv$n<X)qCImQ4o?Oyily2(kG
zR&H1kEFVYd8My0Rv?TwqZEK*uHLGjm#3r~u_&T3k8z78Ubf!<l(YJZ1eU$0;udU3b
z7L93L-^u*rlI$!NIa6Obs>vyFc?V|#KYyG=I3W0{+DFs=rR2LbwlOB;&okCuLp|vk
z#X$RMuI&-zAu`4#2ZsqAPSl6=EubTWh4^mMoB7HS+XnrG8U0j&ZPV<eE{o{U%4Os5
z=WYa&?27I#1-%5X2TFL0o-D<m&5-jFgaHLlo?WUa)3C(m4PH4*Ow)b~Cn_^{*h2!R
z4`Jz><3>Egn|&|dcs-aB0+DmX4HPm@w_lF&J1WN7HmR1sW8k;KYt2xz|DvS;@4Z`C
zh?&?EJ5Wml(r}ifXMmLQ<~(EMKf&+x@5!{IIjJ84JuQ2tO){ciR16oYDa3XrxF1fJ
zuMXDpe`$Gj#<@57cbj9)KGlrkOe1VwNc$FacAL43w*2~eBi4(DI~LYz)Ba&2AY}ew
zqXxR=C(naPCn4B`3+-mF2<c1P9zk3_O08=M$Vo=KIp}2LB7ErOM*dfju)V1nL3~;Q
zVmE<c4tHM~cxIbF=ucwcP4(}0cUUH*MQ2wB7tLvzxU$@X<-?)rtnfPh&1pX?y}Ni<
zsmC)Hf=MmGIs@HYsvlRv8kX#}>H|b`mC%yfyUR7NUWc9-1pU2I$Fmk#10vN2(bb%n
z)(|s>BfCSQ8xe~H^vYe>L-R{FhxRAhIx9`b78Si^rjqt~G;3tOc5@+}ZPk>?1C;G&
zqYtx;ywNa9e0p;Ftr^k^6xG?{dHH~dpTq$rGlNf)I}UbL<xz4C0(>(hql04A{@#TR
z;;WG06bm)<U0@wEF<h-TGyyK_*qcWz#8A)7V$1}1UkeNeP%9k+DjLO7eW)Rzuh!G+
zf*05HNUG$*q7dO3aAh4q-kbWq3U#1;N)#0i^`^iP;MeV%fYgNSNZ$iVane(ht^Tz_
zA5!H-;Fero_TVY+2jPN(ylz*Sr$3pDqghL6ox*MFt9UMmmj8PDKojRUeoS`zIDQ6D
zMc0*O1bwR?w+oS0FSvCE4hpjg!*}NiG54VcPYgRKyu)Q0g`MzQb8~=fkTk+@Wp+61
z*bX@gd_1J>#n;hRQrR$rMMWaZ@MnczRUb7hkzH?{$oZ?QrFt&yo$B%R8?z*(SIR6O
z8s0ZYl{Pzf#{@d-iDsP9=-==EtjIEwuzUMa!F1uUcfrj2-!rXX%QDRae^uIfE;(*|
znIr6q-4}SRMVlEB^d_HN(a}JpS5}jlNXzs!)0JKc2}7$vb9KxwdSuM}r%_>%aI;Ms
zs^S1#m>TMyjf4mlPIhn4FLmio%TU({j};|Fd*P|Gh`+Y;MYI@NPi{r`vH$B_MXMXm
zS!x5C1IaaCEU_24uqqz8`hnzwg1T;gUmq*PNG`l)Yj_fA$OxGG^`Enk`L3}cG9$w$
zjDJVqi#%)bz&%DzZ(5X@#ArVg$@1Exv?VF$p6=(=Fvd+)?9Ih0HUlHR^!70P*6F9V
zr+rlTCElJ9*J;N0LuaVfQ<k(PUuXXf@_33ZjK!UcZACW)eb<$1R+^gHW&&(&1NqLI
zlbk3sp>RdM-y8x5d3(j!fT0z+`zHp*rfVR`?=Y0Siv`Av4|wAuK+he%cSTp(dS<E2
z_}3IGsc;<i+~NxS?cpYW?TzkzE^c9%_O}6r$K@^_surQH#vaQ`iUtId{EF@?yISej
z=d0Za&0fHbW>@mfHE>~@#qrAfR?-Oygd)h2@U5Y-f>?g3$3ocjop4cz?jK5%1!iNw
zLKDKz9vBm;Ujx<e9Hy06UxV+=c$S#;%sB$!!F;)lZ~x(!LY|nX#AU%L#+-mt>cm!5
zY*Qe^sUCiHWhXgZ&?-!KBb5BAbe#fxNW@n(ah7porG51zGc@AVdfCrQ4ZawN>nV&M
z(+Xm)OyJnq%Nr#h?HkX&1U^%L<+XWW8$oiux@e7h4hLV(Fe14@S%l%#@2M5bwZb-!
zuyk8Tod8L3?mEf=bO|#)RNR1%oJb6j!X|P`JRC+f{JX8+F)@7v&+5(VdQ*IVphg?0
zt3;CwR>tv2H$w`JmF{15oD~Kuk1TM$<56_~;omO<)6<Gpey_7$Pf&yG7TlkXX<c<P
z_%16nj?V{n1IZd2YA%=85Y*lJ7Xr^7)-hogTBxUSpC}R@cn&}Dr;glLHsa^%f~hBI
znyH`t2XkuVSIV%9>M|v`{Id+J3Xg%Pn5@`@zSUEh{E`tSTLzTkI(esz_9o?lr&6>C
z$w;FB)4}6(J8~&j=wYaCqPWVg#n;`%nfj^krJi(VNzFW~8feCiY2Dbu@<V{GVX&eZ
z7<>X3`gz9%M!unBjoSKupbj%4Gw4vK4KxUR?tt6xe#{YoNoHW`-jf7H@FHs>l>|oE
z2{*}{qmnnN)fEVd`iD~9wYPTL)ry5pU#|4-Oz4bLj5c2i#O|NHZbfL$JrH^p`%zp$
zSr@Z$-(szQs3)XngaL5~a40DEd}AKJu>f-ffG?OxS}dpBr6n*E8)>3<XnFSr6&^yG
z8xQQoGof&S7QUqDrsv|}gjq%PyPuZ$E<Wj;Aq?)+`V;)LMvY<ahb~>HoQ~Na_zbN9
z(=o%hL=s3OH!DJnL6DMKb@0Y(?v@hLjp`Bf<+a_mf|(ObPj$t(u+qXE>G73ulvhM_
zJp~;}AnF@j{q*MHt^Z)00-rMp-rloG*fO@;h9{@}Eij5))M<M_{NJyzw>@DY8al51
zK-P0;WJNRw3x40ER3$DNkFR6;xrrS5Cd^0CUX27>X(lL(A|D_#oKm@sc~O5U$mz9K
z;X~^3dWD?Wn}RG)-@!*)FZzpu+e+0DN}%n2n{YN^;miVUS%C|CK~7mPY%6zyc6r{x
z{m7_T_&#-t4M&8bNhAGRedcE9VYEt3wp`jj0wspBtIK`Wi*57&FRlp)9rn|A!HD+5
zIeh%W{Edq`2K_HELi1k73#F^8@UcwtE;by1q>)tl>w=Z@MQs0G^3Hc_y6KO{s{5v)
zWx4w>?#Aicc-S|%>T~~g{GJulrxUQhtMO!xu9UXU>rt#z+&)GL4CrWnY5<fyvf~1Q
z`8lQX_3{!$KhS_oNupOJ^Ds{>iC^SfxcTf{8uo#;Ktj)Gt_&i*BjqSV?rM7nK+Ws`
z4qdDt$<SzR9V}D>6qvfFC#e#k&2Ov1U-JwmP4!2Oc0<wRKCU9Yl)*TS62(5-g&;ZY
zzmtpyOyKUYTGHvX)T6VF-tQIXe)s*h(l^<R*S8Y}KQt)n&`#Vrvt{a)3L?!w{oef>
zdl9`MLXsmewBL7gv>6V}5ujkw)p?RPDZA(=hbY#t=U4T>-G=2|6fxT?@_*GTE;63z
zq5ChF$rX$e!#uo$I%~9xlHS?ClMYEHecI*PH$+^Ued$-X9Fj7-sO}F)HbifVu(eyI
zdFgZm;|f80PkqHa{wVViJb@m4$1g*3PYuh}ZX7b*@>fr@9NwXqQOl4Z$Y*NtdD!`Y
znN&I*k|q`!nY}lpNy!u=)HAALNV`CxcmiU*W+u*fn?64wClVJKD!iNP4Zn?Tp|%CB
zRTd=iiMd)4F!{N2mHXAS$_L3O?)^Dj={4EH*PP24+IE*$*nR&Tb<jTbar?gX^qViO
zK#naH!*5I}y~9l0#fr~{3jW<cx1&Yb@ss;q<0Tl$hc=LCUC9S_BGR?bTe<RK?IiF5
zSV`n2GrkhP5vf5~UsCYVZwX<%%(3(DrhFxBPr{EUN`R~b8md0FqeD1lix2RObgc|j
z`}66iJf-kQYT{xCLZwbT7I&~(q9yTdH*M5zZVC1v%+)sgXZbtd%>%1TP1g^+VYJ*e
z4oABkeI=@2o{vgdAE_@>I699}_Yz2zyT^j~I4!KX{Rj^_O{V0~zjw~^<)6%eVz(k^
zJ`g-NG4)3NC*XS{b)WxyZfd1KofU1rwj2I~l4i#Cw(g99uRnc^M?Q+HDJh~nPWdcZ
z^hq^8@xL8!p@7x<VCT*H!HjV*zT#<Z>EiIXZr0KJePY(LJTEL}k0qUB8P^z3eSJ(@
zv*5_Q-q!HVhkIx~1Bx71Kr8vCy1dW?DBS$M56r}f4A+A9C-FsusJ6N_z&mf~n-NNI
zO^0zfOP~rAD(rMorr<stZVi&BYEOSCA<L<}<dE<4`azUe(K8_o=8oGQ0*JBuPXA!v
zc(w}eZ1$A*+3Z~_QrSO$Oi*J<9&V4O=TJ{rM~&44x%csT0l?rw$y**za2MYLe%;k&
z?}ob#+GnpUf;tS$XM=mry|}nnEW~rPJ?w?PEch2Rwobfurc80VI<~$PV}Kd$?UUr_
zw}$CA$a8(4wVT6IbK)o;2vQCTx++Q8SvUVerp8R_V>i@EXzRm6)+R#yNen#S_&)M|
z^N_UJmFkfrOp}T+74LuFl*-d^3}PM4Jn1D<^WYi}+t^9YUcf;aN9Iw2`J2@Z;ZID+
zHk2`folZ3$bbEG=rgId@sK)ZM^-}$jZF7n@cp=C9#@ICgUVo3q4>fP}u_>q3n+kBe
z3xBY_*mS}iEe{S<LdhD=iCA6jVwbOUfJe|3sQvC(pO`OJ*>!>*czH_8w8Pnwr~`YM
zpTc09%!ip_c@aa+ob_t}s-4r1%x7q1Q5R~<$(!!N0n|?6+Cr9}o4J#$Jjs-EEX3{8
zQE%9KIYd{1v=eStvp%c)8s{D3u}RZeb3RKv^aKx7RNa*_eET$1E|SN8t_6W(NWJ=e
zAZ|PT*Ie84PLZPKX1K0XjH@Cf^xqr*)fSMNTKb3N%e(yjprfHA(3@B97KIqfvP|E-
zDP1|O>m?xdc)OD{Z(Z{;buW^^NF^P7W@g0vEHIr66<L?!Q9<n7rRFKtHo{sf=l>$3
z<f!yJ1SnEn9oj{q_=I{sVsJh@FCzzk2SN2rS22_^BsVD)AV=Q2H#8Thhuf?jS5jQK
z%ZsZxP~uk&k6^SY`YUbnROrFdr8Yz9sUt~u1knjkwmOv;c9h+lhH0hWw?-AYUJ|NR
zRNorpmGkb%Q)vdzIeFWV5wGMQZRaHL)lpKrM^O&-iQiC!DwPu4SWue$71mzgz74j%
zVp6jxLy6}@2Dc`UI#Z>NO$DQ?`!(=26b!A%RaM@>bfB(3MC@v>wyUjJJ>`aW%)(VT
zs<&FkxHzRz*u8HCCF=KDrBwXSrw{Lp(v7q^II$`-#0@`*N87PdZQi@t@zU!hGEXsj
z>RUa`=GAhoof~X-vyJVVdIjZkbzseyji8L*`v6xYtdfhPkj1#CUD^NN{5&L$p+Xs|
z%`6^ZP>uWdoC?+Of|`n(6r4TvYo|+aZ`i;;Sn2Y5HFoJ;tFjFW&5!m&n`%Ff@|lnM
zIhxe%tEVkMl9K<eOr>dX1lPlhxUrI9sqXej9NfwsUb2m3xluG)vm<oqxk%XTEaRYF
z{d!A7g?~qiH$XA-0R^KMVz*ltRVAx1QoM_nClK9b(olHNX0<P!zSfj}80&C!6-#Gt
zScP*v8-B}_)Ru`2dae2Cv-vZi-}VUZjd%ftle0Nfz2u})jNHYNsBa7KcA$)P#l>}j
z3}g|gek$ZVdt94@!~@B}A!iV(c3Vj5)8_oK5UKir3Kfe3$|`6{5zYkOxLDC0xjVhw
zp=j>F08xy*y?(6gTP5IiHW&i)XBZZJ6kIjbGb<>TdarwyyG~#^v8Khs7-KAdoWacK
z9m6~yLJ(K-r)3r;<%qe~mkNyaCfQkY)O8vp=A#Egx2PuoTl_vZ7j<T1e${qKGt_#G
zHFmeE#Qqg}*GX}=)VMgsi8tCVo^^?i`%++F`xCkH%g$z02br4F_;QsaPt~)io{-6x
zG+P)A^*%1fESHLa8`;)gTWaCzP;0WNS(bx~I{r5&f?PqL=$y6rMfqo%DWFVL638QK
zp)RvHoW`!GV;Ds=#Yap?F{?sBeJqs7)pD_&Bf`)0G@{{%_0tc;#uxsPZ58JMxZDff
zSVAwV{rZ%z(*f=<iK^bYMK$_QXzHz)pJ*c>j4C6Dx$=UCw?f`asULsAXnw8l&gP|_
z?D#yxv@+ukk6i%K#{&5s;Hh>DW85?PdW|QFq{1?_L@>MSU->o16h&UP<b$LU66f$I
zCAYPaC*U00#iTc`-gRwHyvKxCKgcdT(iWdfAoa0^p#AGwJ|6p^w$Q}WmNHGm?^)4}
zuOADk{>CA|h05r2GHAZzBwgL<62dd(u3U61lJHCX>w_zIt2TX+e<>JeJio42D>rU=
zSzMAqwn05GeXyrj{I4E&9pLx}mJTh?#2?zT@wdzpywZ|{!U+qCL(d0L{s=Z3hkc(N
zs|6mjL=QTpnxwm#l9ClqcJ?vS>OGKFi#l5oP!$78(z#evY=%GCzV2B|X(k1A=hLA`
zH$q>#Yt!VtcfS^CEy@%e>F~x4cIj&1yed}N(m-1o0h+R#rA-lkTlT-2UbZocqWR`t
z;2$3ORH<&<zcFJw9)N^j5yInUj{5&<Ix+dT?*;(n+6Z#G@-kUWyNDDOSr<I9@&Q^c
z#!_K6PSOu}Gz63R5_0e?7-ZeV$^qV~=8*1IV!2%{28cn2n?-j<6z{ex6<Sxnt51B}
z*|=?hPiHL<^QVhp0@~S32Oo}wq$V{Nj)kOu9^wFyDz+VK${%H=6Y;PgS+7gd+|V#6
z4ts#0ukY0hyyjS}TMc9D5f+{TI7z2Q_MCU=9O!X>pbXS8FAr{mvM-avNBTrlVlbGc
ztD{&DVtTBH5_HoJ7N+c|-;$J}*5D^LnbANv1HLh3Ix^!7(FoYL?V=ZM1UMNI^lJ2i
zPR*>fK_>1lC_fdVij8@)*!s8^y7=G3a2cBkaw-<ylhQ;-IfN1)#tKs3IBukmYV}_f
z8M+8Xe6dR)CKxImo~z%8{XUvI$5|&vP7E{kcF<Xa0(6**gU-=2dOrZ2W-s3=4Rg)D
zj%3L>=;$%t%h!Dfa-cn-P`%B>rsu-%vDR-<qj#3KJ?o-=UP`Uyg=exYxE<}kPwm(7
zA;Q#n<9B(Ph9^K@tfjXHlx&;5dg({jez3#Vg58(+qSTz2Ap-lC#z}IT@lQ+eO?<ol
z)^Q^fK{F~Mk@d0&!p+N3NZ{yw()%GZgJOy4Tx6uRD?yhXLgyJw^`!_hD6{4~nUS%l
z=y9=dW8f`Z0_d0*zc0Pt4lvea`M87dlxy`!O=0d_@7v04xZp!q4?t#ldWQBY=Py=5
z0h_RE8+sTr+?NSl*v{K{LDv~&4;Wl~j-pc^a5v68R9oYWN|%(Skj~+#!!DaFy6|5`
zuDWRYr9e*xd%=B`wUCe@T-(%P*DOi~JM2jI-&>si6Gg1)5I(`5pIFn!gpAqRC&L%k
z?+S+C4^HP0I2{XJp4&m^kFDI)77BCag;H*f+=kO(#Rt=Kzlp0$>T8kj?N3TqmA-HC
z2wZR89?TmOn|)zPQu8-^LteCFa=tI5YJy)0>(c;sMiBG~TwbK_>jpYoeK3ReHP{}3
zum7*%eTBl{j_^((5KGs<{p;S~%`B^?fc&bhr3A|LltNDAg60p%*}PNY`b;vwVT+!H
z=}_0;+LkKojfBFoPl*R_WM8V;n_H{+1)jsaU(nW2%x!@CvW|F0_$f3>&GaKaspMUb
z1cA5nYu1Lmd-{Q<4p;YXeY`E+jK5A`ePq5EgF;*$FbG#;oEQ>5zQ@94Yz^Oea&YJx
zc?@%6*4y$bB%n80rZV6EW*-Y#<B6BOfv)ZO1=7N3D;LRdVRLe*6hjC7eBM~RjAiQN
zp7GY5>!#1)1s_}vyMYaiw|`^ACMf^sW429Eh5(mMXKVCZGySOPOhZscG>Oh)4|v`N
zH1_JO5T^k98TO6^j~(21cC-7P;Nr7nbKKZO|3DtF5G$vle@5<5v93f#2cZ7tKDDPc
zj7;u-n1eJW;>d1iKE-u@L`huH3Lh|9DS@0hGq$!qA_dGDkWp}lj0syg1#9$grImZn
zuF!BEdkylT(j_&GQFauYE_EZsz~Y?9LOv(Rr>Y~&*SCNH#^8lA<;NDQ)n-L0$-7oe
zKg=NLE8&_98`N!mFr)0ak@`PHlEVqmU2M#sPvIz(rUd*@^)bykSmDySSTq^lMi2g<
z9AKl?*cG$KJrF@UX7A?`kTVPBqeoEj>xuVu_oskNUf)v3DuU0Yb^1r*EchGV>)>5p
z<ivDagkWlyjji|A_uW8`ev5G<2plZ@PGTdjBnSr|0AsWn3N<CBsY{N>z@pMqj6({J
zkV32=$Nc0FdaECkL$mkU+n?S~7*616-l}b}7c$Ub_|sjq2s~ie%@CMpxydjXjIXEi
zWu4E%H3Re3>`=gmDA^RjqUbSJ)}qd7hM@KXAlvnER}w;kWHgylAZ?o*^CN0hr~Kxd
zcOEr)6dZRgDnA3rQ78*EY$+{jd5Y(*8;HWoQc7Pett0@k3=7#atz-}k1#^NkDq%Zk
zB~RUS?7O+jg6#mvH$$M-w!aD6x*dQy1xe^(oy!U2qVl^eeYK8jLkQZmPWw#(JsnC?
z+XH&3_v<=KK&U_?$LU;012Ah9My;?pxB<@{DqN@f%{B8dL>V;6*3T*FU_yH<xx6e_
zZQY9ja#`X!+rJo9L7_+Yf^As>uz^CiEP?yRoca=+?s4Mifo1E9K0DLKw2sg67$*o~
z7y0T0QU`w?F<?X$<7Z9#8tH7#r+I9d5T)D@_ri+OX`eAoEo8JCHEO=rB)F(3i|BQs
z{xC`^d(yg8xhh!@+bSFgEa!Fk>VBXv{n17*Ng#;STID+f5b4=NVv`M6ar}dAKE!Cc
zT4xa`z++TDEqg3Ixl!qHykR2pp70DemKw5d2<+UiD&CMC)}#TVe0$<wov~Md%n!(Z
zl9CKh;6=%EG=vOru#8(?#-(Lueu6b!P>0UY$x@go<NghWvKU+Z=K>unOJ4$2%S~Ke
z{)Z@UohwDmy3KVtQPesCuSid&yRFwFHKnRhlofoy;^1cQZb|?=v0C#{98w*Ybk*^I
zC#)quv{i8xN5t~DG%*7`A@$fhfYhx(0qJUKY4%)KOl%&s6GFkqiM2zwk@><A454x<
zLXJ2WrfO!-GNBFQgGt#3V$l3hfiZPj34W>{Xer1^SgpMcmK*lEk#4vU3Z=eL@ow*1
z{dzYKF&5s)Jr)cntgU|I(tw)d(VN#KZMsnB87xJUIO;Os^I@i&05u#=q!;uO7;&iO
zBh!4j%gP%MRdA-mbdV<KC5tB|LM7;D(^jz}E;7wZJlsV#xwsFs!Wo9NiDZ;^&&WGg
zRP)kaPWKB*(L>EHU#H~1N$F+jG<6^RgO!7#d=x?kZTK6vI2PK-od%w!tr90TX#oF~
zJ!<pP%31Y{TG{Wry)VC|2=Q?gcCUp5f=AfzAhj0cg<#8ZJ~nQcpP(v+w1OY0J3#X+
z4V&h{A!m#Qskb(x(q7na(VtTGXQ$FGpe_6?Ev4<A#86rbM{riOKWn8T^pwt*VVBoz
zZ6k2d5?}^iE(~+<;{6Z_%I#u<_X4YfpKBOk{&?Whj9|jjEd{LM(@%mSH<7THqra5e
z^{Kx^zWJ^d>};TG^z4iF@@^_!po2F*cbH{J(xha*j!dH)c%IY4EaRIN{_DeZoM-Tx
zq$rJWHdqbZvP1j`81wo${&3@u$L4KHLbeo%`G_+iFOOe%X1!N<%D9w>CuYla!IdZ6
zK4yEr?862IxU06KwSncjX1|2$`)*P;rWQ+yvMKx9QsTJo>d;N&Iz_&7VGSR{VQ(S2
zhtX*goP=%Nx4Tbp{tlq~Mmf;7GOc_!51Y?K5Z*7mwiNcUYSX=3`L6fIvOgh+3}Srz
zm_@vI8~tJ)$i9qdd!k4j1CfAC+Bw(Ocz9Wm@Sm$R9voyIm?Rbl1+0G1j$AwoQdcaO
zLmAPouYtrv&V$r@wUXN`FL$jIQjefn;Tvb+HBuSjA{%!g^ibiKJVqID_DLN}WXtcx
zjoRTR66K@2&QvRT``;^81W&I=1gKoC-g<XiQFf){Wwm+WDW`DFs?&kT8JdLu%e}Qt
zZr!9!w|eseEi#Jf00$Cys%0hk?&I$N#e%22!L0E@J5r1okMSDz-rqdfNh2)uC4FR5
zYMKi}1g2LDM?x}~N!=EY=@V79aX3gky%U%XY+sgr4H3Q<y0~T+Cjk90C-k-DR_qro
zON>udpZoH}1gE@f&F5(OuhT!1WCl_~5^lH;9f=8z6GF>hz9FM+u}2$H<g!gnbIx%i
zs&d<&BNif2v;;;WNvH@qVXnXGYZ7v=tpN2>;pulYzStt~c0ksUJQGPD>w=Fb=^jN!
zFHGoIp9Q&5WsuxDf8Evr{jyxo?kzAVA^eCQ7p}49<FjZ~<pHa=tJRGdI5*y|mQo(g
zL(2JDjZBlTscs8`tl;I7G51RO1vyz&?w~KB_{04#9r{cDKf2yL9_syzA0OGbkZj48
zP*jw%FQr9`O39L43T4gC41<WWRT5d#X5X@}GbmC*WM774%Q9mbgE8ZG#=ZCZ`}}^t
z$M5~e^|<b>_v`&y&Uv2ad7ks4jrSdlR}}0=>miBR&hgg`3*>3(@cT@I?_}TVwx73m
z$a;gzLu{Sy$DDi#5#L+gxi1Z6#((^esv&T^m4U-nzf|1&id7vuk#y_r<x&5)vu>!~
zhZt)-h0Y>__^(9$6vCGG(Cc<uMD^u30&$UK0EKjdymog0nekLtT}I!}(OCRT0WT79
z1uAp2bC(0lEdM>4F=eDk`HzoT*U4Y*xuqt)c#41a`xrv(h2R@4HY|ixnjGDCQ-#%>
zL-hlvwSmB6zTPoIT5OT(13Kg+q$_`&y6ujp51wn?HTN8pmgPON|C?u`kwFQIm&`Us
z1UeCdSB2Z^PrZxt4G61%1i-b)+rrJIJO|rat~Au-u9oKHY&*zCD93+J&=T_Kak^Fn
zN&9u8tL$sb53yipz+>;K;}_Pus|Nhuo6QrI*YuzsFyyh$eo4kgvHg~NSPGdw9jFS9
zhgiQSz%0(i^}3m9qFrU33IxfgtM#qx#IYkw=oR|D0GrK_rD#5d!hf;>&r(i#KS<Q-
z$8{bv2CQqoa_{4|Nj|R5G9Q=i%ZN2$D9y}B0x9Fm`;CcLxmV)-LRTE++BZ4I{ZH|I
z-bNbZar1xcFxqXO!2miaNjXrJe$^jyPqnNr;NV8xa}sNtxmR?Cj(4-GzQybykbK0@
zWBzBmo{Y~OrymDj^7Xi;!&6qWOxu@385ud^^XQWix|#9NS0u<&)o~(%Xlia9Anukr
zY&E>T`S?Yjl@4ez^wReQiIrjM0X<YA(D%J9VXSp-#R$RRptbcKXxZoe4#;_qq-b&U
zjVFRr8Tl8teoj_C&;}a^=^!+JkaqgisltKKH!$l?%2H$glz^Jitf?Q|4aUt=T3)yF
zFs~cm_K885ILjUy3pqLwEuA7Q8oso=1*9KHLLt9mga{lP<hOuKlRv*{(_C&=kl#O>
zSD-TR7w*OoRyCu1EJ?El6GMN3Jja=5zb2p|znK86A6?Qn(d{Zz46?)&v+}_T@u`e%
zp~BAXmJ)nX(8E(17#zcK8~$6;^?^S=N}_p#wCj+;q-_^viLvScC*{4t5=y$(PKyQ!
zUoJ4}+59BQHTW!lhM)ivk05PH3B;!+gKgfZvu7uBmEqWTf7u#M48#W>@c^=-;01OQ
zt2#*`F68{;HR-UByxxg>5U)?s{D{$4(YA;5=!0cQR(+a^yuMurBRq=WI!$+GVQHAg
zgzLbCx=XBg8aYbipKSj%J<>QaZGnE2_!X8Oq*SC`_()qxp0{AB?N$cOO1?yFaZY@y
zwA6zuxxe=EHk%L)kLPtY_1Q8v#?ed5>?Z4`B0UKY&Abz(m1&zaE#uS#g=jLZUCr<E
zkV1r{MuaF!w?LG<s<mjB5gOtnQk41}<=ogb?Oa7JCnzCsw}Q~)UG&f1ay~$xy#LNO
zfUyGN*{S(J?v*le>6B3@jT$8fDi9-=h<+qY@`FEDFCu4oPtby)s$9~KY4`2hhCS&8
z6Gl9<_=O7xxW7}vicwC=24n63*}~4_3aKQHW~|iuzH$}MsItO*&P!q^Is%5NW9i4I
zcdyo{F;qvWROHr?%6CUs;mi4>W*w1u=ji?2H@Z6)Kx|>*Fw$3SslhN9(=I$y=&|7U
z3s-l3^$_JtbfJfc@L%~m&NXUzgJ*X)O>?eRJQ#Z0*IxzMRk!G*%i`LAk&-O7lmCk2
z4O1D#Z?1eGqLcFO<V=H0ysvRz`Oy|24zyF>Zw%Y>hSplAv%c5$Hr{Sgl_S>tDmISH
z))CB;yqjJnETFH;-=B|~0<gC+;Q3<4Sop5O9-K&}!VlA23#6vmrny4W(1*ftM%*QX
zeJ_r&xG;MhX3T&%^xB<6+YPa{y9fqhl(P#EQU!aQ?OW^gK|(_mITkk=#<nY)bEYax
z0`V;r5yoDFz6FR{=BY+pO?XB~<&7zl+3S6PaTwH~Y<|IkfV{^Dcb^(T$)?~dpbF!=
z17es|Pig!lpEr%1BSK!g{VgkE>n4Tg&{kLA?lm(qZhFCAa+ppM@}xA(9e;I|f9^JN
z!3uI7WF^FQ;B4F;fG;>W(ze3hXswPL?gy*}$#m~ORq@&`jn15)K#%_j{_Ug#>(|mk
zp3WFrMMxIo-pM9p7DPY<#!OH9Fl4hQ=AC{<dy1d=2y|`}w>T4|u2q!QYvoBK1yf^o
zw~kcGmc|^Nk<Z>l-a<>dD3sl#h#;&p(zK~3422`6H%H2oMQ&MP<|)MQxW?DmfFB->
zP#zIrRu!p1yH*stp?N>oX4LRAr~?x2scMy{qYW4At$%E9|CnO#21>vczD$?TL13Y^
zPL@!y(EQXDLXF=|CrOQqqLRa;oB@^3QO|O2e>dqqv1ji0_f<!;bwqQ~dRg_DdXo|5
zc0vk)m~)alu#f`XhCM*hcBYJ+ScGl_VhAZ`uscOA7@Aui^FONo>s7T<%#3_N($)?b
z5W#&cD7Pk(eo+?<(S<kZ3929xqE+U<8oN)n@^W2ygfT=r1rmxWk{gBLn-@-rzIh;5
z&t5-WMav2+Kr+D1l9Vex>dY|6^hK*QJ*-aPn@G=hC6%SsP)wYk{~*&R8l1Np<VLOF
z3WtbxJjozzr;rQ{>$tsT#+*xRB3d)Bx_s$}S05fOY&HIhOBd|^jlUj(|6C<tB$OXw
zb87qPjIPCAhB+jJ8UZ9wO;ipk`j?S20znE>^dP|~-1SOs2!cc*-noHAWhHIhKnFa*
zr#)DPtu<1blZEMNA|Uy+r(=iwfvM<O2SoG1zn4l%04gV;{N~c!*?@P0EJuLx+{ho<
z+mtZfRaak12QzS_ZDkFEDT~jxs))}HL$jS+=`#x0E{v?>(Hk$=gCgpRZ_PFHUS**^
z2vKr2BniMuDFunY_Z#uM=Y423xnuG0-IYxCo`Nb0C8qcYwF=bUzQ_4~staGU;;DPe
z$r(176JFty_jKkFkc}QSf%r?gBU~wRz<EC#cR5OIIY4j-NDzJoNkAEVp{wx+sDYST
zZ=QGUIo2NsX_~b@o}h|)Sb$~C4by1*vcNtR>ymwZq)l;>JaxN!^7=<p8g)DA(3p)m
zxebtQXsV=W!QWQ#tKL;c3ZQ03^~}ymDty68xRTon8T}(*W{;gSu%|&}t_8F_!^kDV
zyJ#c9hC(ZFG<wuE<Uk-F%ATR3_aG3G5v-T&jm~Y?-48vz{&QS0H^Or7T49W9$KIYr
zb*?9`ArIcAIVf?TucRC@=dr(Vu*GuloJ|#8_smaid@X<ZV424`w<*V2^e`g5I>QJz
zC=70z>xTgMb_pjIbBj)te>y-o5dFJL7YV64{%t8JQV3!vA`Sn-iNDMu-T~!~BacJ2
zJ)cZW|EXshNBhWgY=cE1S&ypbG*uW5CtH(A3uvot%p0L9D(Y~ux&m@h_vIcG#tBm{
zJaG2O7tE_$A^5$lVTL)M;OG<{@V{fY5@&QSU3Oku;p6B(N2jC_s&8N34#=mDZsD3L
z?it&NZ_N}9W4&%#2Ur__$at`vJt3S!O&g(s=?|MoE3{Gq<cNQtw9?qZ!~W79WFyqC
zva@$;^aAOHH;lRTmUi(LcG7W_dAkyD9g);mVE-z?`I!!%llw4N({}zQe?;M+KFUj8
z5Vvm5J?!@ozpdP`h9BGx9TFM>ZRAAB*$Y3N-nilYcy69DOmp@Q9B?wn3i6Ag3L!(U
z#LYCDS70*9>?z7utEWaP2mtxYf2ZfU?O%?QQjoBahCSwX$7e1t3W54OsQRGfeaeI7
zhLz1dIC}hHwq^nI5__W>*Ik707UsKaeV|901K#V6Rf)$|Y%Z}z3>aDM(8d$B{HR!O
zSRB*<NSepgG_rI4Ccw5TW_`%c{NaAiWQ7OGQ9VK7RblY@wB1YgWyoO}9}|=$k5xU>
zqsBZOA1qtb6(;ICEOjtOZ=ZT}-0iTDW^iOr1wwaPBH&DPTA(p0=WQkocWq3!iekj&
z>e^eC(Y?p@x)u4SGt}9Q&FZKi5F2lM(YfK<c_;d!NP{uE)i0p~Tm_j%4faMAdN{qN
z<QwHa0L8at83yZI)=WDsrwo&*ju6UUF)czYNDGKr%pJ->z@nL_OY5(h3ICQpV-RiB
z$@^xr%jwW?=-#W0k|k$ORfSp<KAO9j{RI;PyXNi_km@n3_zxQ8M(pG=xa3)(=*nh0
z3ZZwSAM-CsHpembc@;Ffn1-^63i`LbpMVT`L5<mV-A)dPpf+>l8Ec+x{g=hfD+aT4
zf9n(AEL#^F9){`(-)q2$R$jaMWqG^52Pv_5Zix8NuW&0ia^Fd8rKJn-2~#gwS?XSV
z513Y~jComEcDO@~0oTea^yV7S$E(QRr!(}FHKsAimen`e2VdfR%y^-eGNFMEk@(fx
zcQcm$#jZIC0M@@00*Wxm$eA0Pk-}EJ<hap)h+7&fj<812z(=52J}$uCWjv?ORgRZu
zZKqxzRvha^hE$8Q|103VPYN@kDznbeM@@i<pQ@|6dq9@>Vj}W~N&#GiB31jilmREC
z1mZkZ>`wuf!sfyVQIX{*9N6i!p^XV1os9-717GKRZjipX$K1vXWfQ{EoM{OJQE^#>
zm>!C&HuLN43NFU65;FUdDaFk>T4pkcqa#T>j~&xQd_X@^-;e9q0_NyQ)AdID?Lem-
zHp{U}I`=wY4GVvW8r*OCu9PPgU$h$Ukn4pu-_&?^l~6|$*|d$VQOd_8UQ-H;80pqy
z%HMVBU&DYF{0B=h7t>fK8-Ql;H{fOrxGSmtr+=B2A1u=HL(|vDhK(BhDR?gPFC6HF
z3_jK&6gR%P@xrd37JlA?@H1$VJMj81CXc|qo%H-~Sr62^7pu}WoWHqY*6j6Avvsn*
zdk@N9-yI5X@(Zi)*kNwatp^|X64mnR*Oo+h2n5Lsc29_%{|6c4D*<dBD>gEuZW5Mu
zU%KHqkWpUHbm4E|HYA`{PbX^K;Rv93aY|)lYR2su#*j1*F8D>U!KBhMeDh<r=t_gs
z;_3=V3+yRuX%eAE-!=sJ5i{XH=1RK?@Wl?wTS3{Z$6kC3n5wQ!`}67H<tv0#hZ}|m
zbA$DDdw7k6)O_CHz?7&;q{|lIfNS`ZD0&j|DO{qa?zGCsJt7Ktv&=FX8AixzUuVl}
zAqfbWh+RZ!|1ET#|2c7O<c(a;PV`g402J1R<?aWRYlyuC2`!}JkApR`Xjr`OtAJzK
z-DPOmd6nr70Y&SelU7NbTA`z=5*fxfQTLS({Py?VE#8t}_6^EK57vCY8@0E7H&dgc
z&U|jpn*;tW)-QC<K8Hk|hdNqA=MZKkU1~>=nmt#Xb?UwjLd)idXxZ(qEXw&?XgqtK
z{r{n$XiyswHQ@@jO<-zIK?fiK!yb~P63ijOq?{(Kq;$zhe1Q-`=^43YGD=W<FjnV7
zN+MbKaOnF&Jz+~~<*J|f>%JHySm5r29zh9|$ov6CRGsO0;O94^*)}--i3ffNu2>@i
zbBfR%iNH=MluF!i5aH(wN>?aKt6>WGWwot()>Y7d_hBKc4t@7cGeZ%$i1!qbl&4>l
zU~Kp{3G)xe$L3OFX@C-OWHe5CtA6#;MbMNbbqI`F_GZ!fQ~aQ?D&GC3^9->!kGIw1
zmzTT6C8_m55d19d3MTv>yFyejf3a5~dvoY1{dfS~`(eZul)vS?84S?tx1yySOf$;7
zZv>HJ=`XP7`G_d#U^NZ{3-SxM<jh8?hB`=BD6t095VW+>C~uw(O2*_ICm$!8diV4M
zjo*q*q~*athu3tTbKjdGtBSd3%oDd`yP`h3TU9|t_<Q&CD(Hja3Ci1NDc@pKF$e1x
zj&WNGe#2|jioZ-3`6%Rougm`j=!FbHeSAti=M%PFeYfe>N2zMN<>uwp5SYKLSW;ak
zOSJr*fRs|Z*uOx34wK|y41QD~f{-?gW3rUIr+UO#T9vm1{2oPIfgs6V=>^5au&0Gx
zSP>iG3fT`w#YUlxl%zvIx&=(CRTaGd#M-P$OSf`Pa=_8QlCu?v-$j6?C)BKyCSu<f
z?=sCXH&HdGy(y#bX+2k$+?5W>vY4Ya$FeXH$3q9-Dqq<Lbc8YoPcX4<DS7|`Y7+<(
zz>wC@GaAo;0uS=4lOcr*ElOM9QV~{~OSes`7yP?X%><owkxKomGXbsXaVNN7y4I~<
z6&Tc{mA{hibw;LVNUgUQeMWdew}ul?l4^1MZ9v&m7Oc#wFyFF}V0Rx#WY8Dhd^|L7
z#YdCw{yvyTY-Y_vtH1G;%+$9Icu(Ezm_;_Z;h*UzYWbkDK|isJ5Kkk(-IO|@Z%q9v
zqj6TgH<iR_IV68|2Ez$AIw7zD+5j6#b_ajd5kN7%Pr1lf2j3PVJ9^GtyUIlgEB=c7
z4I0aRzQW-VwFi~$-zdvE?X46alPw?gQdkM5OHXL+dTsd)prN_t54Z)CGRU#HZ9l|1
z1QQFI_P7`ro0(fvMi2N~57U}=P+#EEIo4zVnjliba=yBKpw-K+2i6qLk^e$Q>n5#k
zN<xeMyZF{_^F%HfNYsN44jvT7r9i%TD|&?oq*zi+S+ODVFo-fn`f(w;)E+?)aE?O|
zI}ddFPCRZv=wtZ<hh5VC?As8mmk~ni4hlMfgpv?oeGO3vtq1#gCfExGm?1S(IDv3<
zhP6|>4%kgQ$*R({kbm_<wYC`a+fft*Glp2{xfQV8PP|lM;Tp&r$b9Vd%Iz!+NU=kr
zR-uDH$j`nXM5%w~=uaz)l?-E}N28QsAjXGo9a;sTUk<qv32SyhDuRX*B+twUaZ4J$
zOlO3u4Cbglt46y^@7DTAk8QqT{qOc`zjdZbB50}{k@)Llmmzgu$Tv`+xkWXafh`tv
z5$5LL)^6bKgk~aXnUk<Pb__zZD-HAJ74WawIXOcgI_R66LLNeu?g^mXW@KN8OnZCn
z<;cCeRzovY%+0_2d_l|fl!a{1@Zc!~k<j)8AwFnAx5B)AgvA_(lQ{g~Xfm&f{49Qb
zdku3qV#c)%>))gJvfe|-IrlQH+(+dJ4Km!S4v{MhB9L(a6+3CRq1f_uHmVVi)nlzw
zw3n^#*6s8XGPi$hPL`g6USd!7@UdKH3IBH!g;A~<@dSXC{8N>BRVrgE!vn;TCofyq
z&2@%AVL@D6??N5L%kz1+)ge8xc|2^Omo#C>D)r#+YIY3lTN&|_Y{v7J>y(A~k#Dr+
zO1D>lX+s=q0S*6l1M!;-OjA^+dP#4+i%tZ&<pfYXpFPenL&F5U8$6UV1gUW#)UCxe
zBu|&-09es7?bm)e(#lbc8|nRrP=C5>F1XAaZ+i5>)g<nq>n0Ny7~6P9>Gu=+yc%V-
z5a@rB%Ys22wDNRoQ~5S|^9t#$QKJjnWWM3fsYv2>V3zA>n>DQkq<NZ{Mac=;+_2Pn
z9M-D5$-r_>{?S|;cLQlRK+RcBtUW{4yItduW*dKj#YfNjJbgg=V7oauiZlb)_lVAy
z4#YhVIib<SL8i8rSy!*&^W)G2Xzo6RGrhYaV$tvV^O_5yHDIoSX17|08&8z-b!}5Z
zFo~n6GXyeN$SXQ9j6Zu&V!;G;H#^s`(JJv>(ViPKs!l@xq&r;_w76E6)qb8Kq}pg^
zL>AMJLp@qBPB^b1aR8sl_cP9QE*o0L(7df@2Jw!~nEx;8ApP6EPmd^G`;Gdo%&{m3
zp435HlgH!DxqX-2Y3`mEsO9X<-RwUfH)C8e`~43m$^YCY>f>r1a{6itVbAD~y>1oO
z6%O34&#&filckyKE5SF8MMRs={thfYQlNYWtTyEx^rwEfsnf-K>l~UJ)jC`8mo?#T
z)a)69OJ2=gkW#=5KydA7<_YQ0i@c`70!#(5)Ysj(cMS&_dy(WhmAoPhIJI-=0NYqm
zpT9{X{t$<%9BuG1c_ys^Yh+p8gPW#Ku2wk5lll||p|>Ey*xDE$z_0W|$#)Y>w#swd
zS-f;I`@0wN77P{%Ux?D5$yqd6=xXS(x4WQX4iHnuTE?9iXs42L+)%TkjMg|i&egFc
zQ8dM9&o?;qSK9|wY&eC=)}sEpo^nSAWBorn%g)Kag>VvD?w~Hw&`C-&*Obt!%xo%*
zDXC>UJvXO1b%5a{aKRO<Gr7yKyRlzU{7MJLea6mFkFB;#f&HQNi~9Y0DMEbpQ{`xA
z|9&YTAUx?MW@wqBAT?QTQ5Dh>e&OGI`I&x<r=Hc23_F=bYp9d-xoqI?%r-|l?*nzV
zJbNN(AqWy1Rl;XAcOa)Shi3T8s6zWgFdz1W)Jx%$poeh<1uMy!aUGrd@IaFMHf6$c
zj1pHPMO*9j(f2XAth6ks*kJTLqInG%y_J&NlOe2i?zk53332|nBM4g;VSPoe5IkK7
zo`al@<>8@QceC}!W~!{;)Z|Dw$gKC+#j!PUbJs;E4)*Uy&p;XOGjT1qfYpvhEU}V8
z`PZ>rU`dpL%CZ-m>h)JMtakiK8`eXB&Or{l4bFeN+bCH3JPWm)a&v1N#DX}mqsF`N
zFiJhWFQg}>J&9qs;9?Z0I$YZh30y0BP97F})ygF3xrE{-VC+?8?I!ihf+Npb-?C~F
z6j`vM?+TBCMjN7Ahdf|m)oL#8?|$^raOkD)q6HG?d{w1lj#RK)#NzX+u!9wG1k_}E
zM$Ae-6bOX(7P2}nG?YzSqjT5N!(A<+6zZA7`PO_q{q50KP59uspQN+{C+j0fLJg;C
zcFH>4?$EIO$>oofLQ}|dIGm-^E;U?A+75VeJV*KuHJh?W7dSSZMfessizEUWa^($a
z=<3qb!BFuGIWU&chnz`$UocfKN*UfHWLGU}gS^xi;F_-%TtFpbd1hTr7Lm?fAH((!
z`Z>|O`sV(rk|H#?DQTcc1MxeG@IZ^~*{@jQ&sacZB>JhIPE+W<rtM<#%T&pxJBK+u
z^*zS0L-;X`u3A_i$HQn4)19|MxB_?}>?T=U|4`dO>l|VL+d45BZ}I&8sh-z9MCx3#
zziRR{-S;)%381A}^i{<wel3}4tuuZ1+(-Q8-wCnuLb&9~*pyLTXxMlqWJth>r!L8?
zAG7c4;$EW~I7mPKnqAPa`N>~wd4wfVjG|bBaaBKf-Z(DSAu>?xStpHydLSNMe>q!P
zY@pW=lAQ?aJ6F$7D-E-8=jZRngY*L`zm|XXBtf>o4+8~gIXV8W?ZFJ&IRM&#HFxT&
z<o`hdJ-ch2=uN|({HvO;U=q`qeLsIZq)F{_(rwQ3v`Zp~wCB1Zd0DLP{X<<mnHDVz
z-yNw@QIhScmy6YAZ(b!1TMJb6F@q#CcP#0l!_VFlrFM%)*Sc_G%m?#_&bs$DtK#V%
z7wBdczNa!Z3q0I_soHs_oFLY=i4nV{$Z&vS3!MoLt+vO13O*7~ZunhA9hO-iJqso}
zAUN{#*xI9yi3pSqMz<j$>bU0a+&^hMsu;`)fhUA$OKa%R6$V1u;JJ5Vd<+HZTu1+c
z6ehi>Ihrj|rj|K^4O{q|o}lYYKYehg_Z02*ZRWhz5K@CSE8mfk0R?<d2SoevtiPYK
zs@!DNl24oK2(T3Jhwe@Cdzo0Ngd51LY<WcbPuE+A?6U&0s_(e;W0EQd1Fiuj+d4jg
z{Z9+vn2bDzKnMXL0mEsh9_U~#JD;!lZqqD!*h-uhaP#!6*X-tjitBG#P6)jpD;zKw
zkMZwg2xB{lHJn273*-o2oed4;!CeXSHpTP0y<fKUSu#g2I`s*Y!bNwocaIp*f<KkK
z#5I&FM+Aa^<`Yx>*xr=6o6J4~fKVPPnIT)Z^=_VMeh6Xp3%A4AoOQMckKT#PB4>gX
z>WHVHZ;M!CFvuo_I}{NL^7X5qZw9zc(gi&UIN31$lgVs5pR;<`7O2U^KxvA(?nY$T
z57R_b8r(hbDz}c>+i+_%ZZ|ZB>v)IQdJY{UZi&|(M5K4SLX6!C8j7hGMOO|5?L{u0
z#?;ZmtmeMrb~fg=2@nCb3rcZYeksso?t70;Nr7ILf}}*s7sg(^MeL1DVL>sN6N8(?
zK#g6F2y=F;3yg?LEiHPrcS>WMXN7$kU(MlRt!SnDs;{MZJk6Do#q_JhIzP>JfU|l$
z`S?BWqam&&>8uDX!DXQbv8;dqBa95eE7&Srl1TsRn2drY@_W(u36`jmx+t>ur`4}P
z#s7c!vl|g4|C^4Nd?y4w&DDn3&<J;rzh$#aMYsclFwSGGRtoL|8P#94rr#&gvFjw7
zJGj!Zi?#Mx^4#^wz|-GxznEYj$$+>-B#7%2k#S5B;^<JWsBDn#__^43GuUi*L6biS
z6?aHLcWAFAbB?J1zdbX)Eu<y$%X_a$$KwA33h~~E#>{cgIU=mFK+x!NSzk`GrCP^5
zU#h&`u<W~AWB`84Sh-PYvB?0Ty_P;}Jw0#9XyOwXT6Inix@79{mvw#M1psxdC|sFz
zmH9qZ75j}6^s4U>cr0kUlyHllE+#lg*ZcfxJyyTD%3{GFlyarz*G9)@a1ftpRPyTg
z+a3Hn5Tu4F^(|&C=tl1XExAnvxOO&j<`?`gi|fAy6!?NewAblZe1|iwZ9+?R?H9D=
zE&Hy0N72WV?6IHol<jEcMI9U3w+xAeA37M#t1v(a3c$#wo*a2iQvFW+V6jr<1`8h7
z0eozy0;Gk{F=dtP8V_E}dO9Pu)7vj?DCzh|=Bt29c0!~ET0FaPfPR=O@Rnd@py!2U
z!Ocx%|3e*gEK}TG6Ff+aiW;de=m|Wa;uAr<ndHjA-qOHqHE2Y&m^Yr-_|)sG5!)Uq
zpON~CtPLLWkR!XbY_G6lw+5c(_CcxR!OLN`iNeNjfEUI<nP}g!BUfbBYaO(#{%DcH
z?~yB4StBIYuhoC$Mp~`CBMs{t=ph2F^-{wHLYJ9K&ci@yQPZt`dH7avar)x+0mg!r
zF>~_H|6lA>un&``j^J((!vtmf25!7FU9v|zh=zabC^8WxPaIEJ`F*7|^W7$&TlzG;
z^`2hg72r@tOM7tIARQXJyhXFDV}1gngnboyTKGw=!pQGa8?<Ye9=bR68D1B>J2M;{
zSVBME^U%ywymn%^AO=!P^bq@A<y!;RuFtz+KNQu3{cBsrXnV3%4ENJiJs0E3tjBKx
zy-Sv}FjS8NT?+yDbOeGr7AuC`TfhIWFxq^(*~3}YYA|GSdB2*dhpK!*%WR_Km{tcX
zb}#jw(X=yE_X4&9*R+}Y7w=Z=0uxI+J>R54B}b=^#Lcw_;D0o~_0x(F2U$}?POQ~r
z=$XX|d*BIJI#@aGr67N$@vaqa4jFy0{mU-S*Zxi)ZvUw^Y~3ecns$`c+$Ue)N^_r=
z!wocgzk(^mTt%Tnzk0kqH#>=ff)!p$2!e^Q9#Kl(+KF03eWraf!6SWVsJoN9+S<U(
znFGB7)c{X?Md=n-Mt3EK5Szn*(JM4Whexi)P8CB<l?F~n8Sn9m);VCwPu?dfI&JXk
z9TREC_beg!b6236y)0~5iE(`Hu<ToHmioI<hh^u}c{&fX>3VTUtAed$LYyxAjX?Fq
zy6@@}p`2%i7M_&t3Lmt#yG3~F*!7=V5-&crmWk*}M$W%F`StE<Tw1H0-JbLB`yvMd
zOwO)8vN>5z*Y&t_%uFqaVqKr;pCLAX$+Y!z2C~f3J=5`&rR#yN<^I;YDN>viy1RL}
z15#^EcPJO{2sq}QU$eNi!r${F2Z^uJ&Z!ARX(1$*)5N^yjB+<qa>$yQM%b;ixNH#<
zJnveEIw`B=uyBy;7n<sQaXixP&HZ{43z^$Vs;<wU4{jBNu*M(xVJ@)SC6x4Y+|#>D
z9xJ`H!ZjQ}%HrZgck+j?ZD>D5<ifWRi_vbzKS5euh*Ip_-Q(2vV%h?6vmp;}?r-q9
zoGZd&Wo!wPV)#-Zs1S+?`B>zAhudTgo2WL}ni1n!(2biy81me<bPI04#d_Hw>OZe5
zKA0EkvC18Xei(D~c=<cZfry~!(4#@pBdZ_$tqLxA8F!7zDO5!n&v|f$$1OFCSzJC_
zfSB8;V>vndbg9|Fw1P|yyM*CacyFnqioP>=4A1>zS}*QP?W@2ob3JMN_Hzc@VI!&4
zE`#Hl8ZwavUeitAf(g%7%;icK`ff}Xp+PT$s2S<gltCa4+>6TYy;uD`y6I{2B4yvL
z8P^G`oOG*SwQq~&-?81W9gqzss!@p4Q|O!4DV&IN7hl;OGo=%C=-jw_zxnC!x?u<X
zk>6QqDxHG(3}W7U+rY<q`=0IjBy<EXc3Rtog(>NeM<mayMQY&5P=XXL4%=C|S7&{=
zb%Qsv_MraL<E?*8iVv|+Qc|=ZR+T6uL@M5Jx~vZeed#aL*vdCk)gC4xmcl)PIi2O1
zwU_&~|0M7&a+v-$J;@K3*@$23T`vj7dObw*z$MNrL09qkYhgDh){Yo9H{WYoh1?TT
z6Y@<P2lhKWCB<nSIdSu~{9N#}V_EL)@mXz9>>-_~siJ(nJ>4xJSah~gvYG!hFUsY(
z^tJYMmGng__4j)#_I{mI(yzia7~Vn7MLMbXD=AkB2F?-u<o&i#daqVwrmy$E6o)HV
zC+~$iEK)z%tyjB?*Iq-^pTlT2B(P2I<y$Ls6LAcCx0#}6OH(oae#w?snET92!*SkH
z$}Lq{$z^6Mi8V0+s-FcX_?dgZ|Dc2IaulG4c;CS4<2^xY$E2<(x?KEj7oN1t-;I?E
zB6U7+VKd^YIB9MBFr~pQPUcw)UiS@aqz==>yA@LyX)(8suZbJCl7;=xC-);0&NyqG
z6+vs^rHnUge${EmHIW11%D`KSS*Lx!vOaq0FjR3!9$}S%%hJDg!_Tr`d*Y$(W}R?M
z&ckc3!x>x=2&P<5>$>Lkg3DvoS(vX%x9{Ds?Z~2<a*`5wOvHTA`9-uoP(lcW)-L@F
zeT063yphMjD~k4vLzk;&eRw<JjP<`R+s7Ue#blc`C8Ud}g6XY1^o!-!-P+?H)zqiF
zy;F8epjz$y6x>pm`kPYE%N#kli}-q4kAYpDr%B`9iIllE3u(%Mm*AR;U(o%{vJi{>
zw$dn`aRKuP%9%JOr<ul>`PO1JGV6OH+d-|S;spvo=NwW(g^oeHZ}HvayiZwv=WPab
z`QA2vI)9InbLy9sW%2^@#3tnTRwfhT(}Q!=YU)5}!p2(ug}0YsaZ<B%|Ctn?=ZlRV
z_qqztHVrdyd^T~5584-{+)UfL)39d}cT3cc)$lr)+uwN7;lux$yFa9PgTo88QJZiG
zc16mkpzi*fSxUVI2<>jM6emEX+N#-M?z9`Cd@O~$L4{qe{p;peG;yF2`BMIx?tNl8
z6h}PaxacbQ_!M{LsM{e|Zqv7z3%lkh)Z5Fd&PQsxE1_-j@7l;+bN4pq?z`mC&SF!T
z%Q|C@+Tu69D?R}gLl=tql<^4CWuE}_&#))u)^yLG-#ONs62CfiEyx@OWv*0?kUcS)
zC_d|`x!}We@-1)I<^veNr~QLBHnZDi0_J_irPkw8mkE9=zGkJq`hFq!mj)7_Es-*x
zy(T{Zr916e;eb$os~b*{Y?3z|T08nc{3WiUn&!e3FUPFVEFEPs_Ff<7SK-;h;f?dr
zn(vxhrFIddn&+Xbpc*hKf4ILg$oflgO73ZiX8IC(D)@-E>=Tc}B14J?x{_Zo-cPbQ
zBmE}Iz<Y^NR$(+zA@$h+On7R?s_1Tsl;qVpIzG>4GkH(`=V166r}-3oTYh*nw@cyE
zUp{oFe)c0#qmHr{8y;N0cFK@`JOee9wj6TgT6~}fJ>d}>*}k-n^iZ)T%U6V%2kgh2
zxBKX3(~jT%f%WRh$UEbZi$VncYy1vCTUa3lSG6L`x0uFy1VZaKxQ{5g)~+adqboEM
zn5n$+6IXTj{#30a{c7B?rxJHx;6FN^{B&eq9~bG($?)x>&-WE1@1Z7N+Ym2rLk(5*
zZ{K}E0tZ9xfW^<8zr~R(NkRWO&^FODU3Y78G=1q0<;{TxpQk;L#~vUSXKBcr?|HB)
z{9ryp6wrd7$dz!)+0P!%Gi66}V6NMXa)M8<KR$bPEun(VARZ{-<1^x$SL9BfaOq&5
z?wjdQ5k^GBFAmN}x`{s)J}P@zfZff*sgMuzDChuuLu&EL(f79g7xerMmTwfA-81g`
zLXWt7$>)q>IpU6>YZ2U-BgQI$8ns7tCScW_iamV@9+u(7MV&%Oz0LD)YFDS5eKsa0
z8DXZO`QFPjFi4T04!OAs;X6(kp~)NSZ~IuW#<R+QH2T=4OEa=uXl)b+Wn)`EH4lQ_
zs9`gBvUCTm$~m;K`mvulO&J5(B52!|{@@f@ah?fr2kb{Z3-g`**L~zOliqoIjlA(t
z{PWXO**?g;wW}?}?TpoQSFwiV-#=SV+WY_JT4K6-`S0%eh{#a`Zz#c8y6f)o(Kos>
zo(B+I!Ys8-ZpS->0;appbG+FTNr{eUPhIMHZxyNrKJr6+f@3y!L3SPQ@9O2+BiCGp
z^4>zw-k?g%x0HODPBxbuvg1D`Nb@zly7bvuSpO$WCUKNW=3cW6-mor8yfFF#wD{a~
zB;w_lRf%$Q*+x)YN}J=hhsqD=R4rkvwAY;O`9Oz%H~S;V<RaZ6e3r1Xa^=gIaiNFa
ze{%jfR68I}i|8J?#`H&FRJ^+yS~)i_>m3UA5!9L8j;s&gXORH*68ufshe(RhyO?r0
zZ+zi(Lh3RTR7`Tbx?8$Hs{BAtV5AuB4h@f(HeFGhVj5!~TsxXW26|4vy&6k=JZ?%K
zOyF!PeY~qLf0(+=l*JQywnqbFw`)$!Fk)YYF)NlGsFz1FdJL=G16O+={N}v)XjFID
zb8r0v3C1@^*dv_Wr0V#io~(W}QkHDqcp&3&l1+w<Lpg(6Jba1i2F;$Q#Z1=F`C<$N
zeNxXqqMjQQd2BrG6Nk^Z@S+2Ge=r1zzF;)%rG6*H&}uH5i+Z-J`wIR!^$R2M@2}Y*
zFN4^GUb(zM?1Yq5N48UHn>1&=QiZltSW%LJRAKttmaoVKxzf6dysnNL*ZdYeqBSzx
z&Kefl-qg>(g9+2F3HVm9O)9wq<$+QfUd|iuZWuNr@|Jzo9k;3uQzkjTI=0a()nVRO
zn9@%R^x!mG3b%>v>l^o?)ilF@G%0T@Muuz$eA>MH%C4;iQJH&cdzs3uGySnX(}GW&
zv(D^ltxf($g1T~{r?=0DveES;*JM@$z4u<*D|iil&-G!V=MsvGb0j&fD-0p1n3>S-
z8K^6|Qf;^Sp!%%VY1?b^sjpXy8H=7TF~z4^s+DMS3VCT=Pj#0Pl1%M5z+b1`IHS-Z
zbXcc=<D8n37rw(`DX>$#fm9uE^)f;8iEqJGW(V2w1>^c`x^){U%r5_N<f)S!-0!X5
z{&)=eRaDSuciwbZ@|E&d+3#ZC`SogZqN1;?8cSKH`JDJcG(KRRCMtl6EtG^m;lqNV
zKb%3LOX?i^*s0zbv8uT`&k$oAHYyWDKuSxTw6^Zj|LP}kKGC;X_~g`T*XxP<ETg0&
zQJ*|gJFIAHJYyL8Q*ujdPpG+2*2%5WjoOJKInC8tv6Fg)Iv1<By{K3R<sy-vbF0E^
zA@rwo=C3l1L2=UgNi&|c>qBhy^oW<;yeNFmo}fYnrPN9Us%MF;DUkD;x_RsSeGdv6
z#+gqWr*PSn6@O(=tJIrX)~>OsgQ0RRkGW%{0R%l`*rymAdFu8vyO0!1H4oWPLjEPp
z?(F1QPIN-}Z>ON*dom{z9h!wX^!BNl6D-&z?@C!X74iCF7Q-I>^!A@^^W4;Y>DNwB
z^>L?$b|44#@|%hHC#c~*u-7SH+4%-R_y+Yh^|$s$xxVf6?BqU`OyXp+&>Dt{12zKj
za_JZO`=!@d;z$!GLjk+!9R%h4dq3tqCa%@6zXqil#u#xEb8rFa>B8Qv@2SF{sBIsD
zpGdr^ZS5Mfo3?v2-Mo6(@)G*v@Ih9NbB@%>FWik{u^T;<xV!i59<sDQy6Opo{OK8`
z&{_kk0oQfK7=+S~*$;|w3gwU|t!?;t5&FL@?$Tb2v_6q^?7a<X#Zp@ka)EEoXY^$(
z%YcVpuT8k>1%JuXYMu29ZCB9?@-$@qpi90{oTa{R!SUuJ{PO^(ONc=IU_!`p1_zc4
zV!Ip1b~cnLq0aFZJ;9IeuPojA$7wqHxx#aZ`pYWYX=s-EmOdU7a}c#l<VlAqt7x<3
zbaVLxc#`<f6{o8^zh)n=85sKlUtz=2PA}*I2czYbI7*5#;xygL8o&6hXJbH1BiI<r
ziN?MvUT4wcnBXk`Tx~HHyJ^wHUs9v4$Gwu0QMBfV2Vbv+zjx*Rgg%MN5gxxv!9Hf7
zKd^3DQFw`^9(VZ>{*=zCDy}lVV1hokwBRlH4x8<_%Wucd^&qx1@6oP~pLAlW_c_m0
zpuCkEj3tV1?i`D~a4dvb5)<%+8weDJP&I}`L0v84YhMz2<CeeTEBPjyTF?4h7GSRD
z|9tvlx*{KYrE=l&rs8Czzjb%%l+|eZG4T_-)k{a`=WAaa`N1+?WY_3DQT0nmw~|h<
z|AF`U&|<sVhR%yEo2lZzJuLZ@E=?<juiWn+nXXW<Y4a#=8=S7=x42^h`*vb9J)d}I
znw%H6BI9&X;dBVzi<}fOhOBUDZ^!cl>~_*!eY|;;qG-H2*k^J`oYeVd7lE_3eCOL%
zIqSpSgXcK62VY`0d*OUd(It;sJs}{Ud?H$yZ$BSHI?=xC(5uHMu5X;!-`x0Pb?Vjm
znp&;|*pIr9j{6mn>Tk~zW&jRqKte^pS_clB)=O%jaL@U3?H}+TyiIK%tv3G&J6<pG
zP^xaALuHqsEZDUDm$cc1iHzS66P%At)JC<_sKxX$k2*JM$G2d`__2&_9!h;xo*aKy
zEFV7&Qz-u&=E<XkXHGs%J5{`VD7-8S>@)zsS_H(F!ea`=p~j=0{o9YwJGQL3TkMDo
zkzenxVydG*-N&o5@&{(sFVU@QO-rwfTEXpph~N3_Sg|s;lTj&fd)!fi)*o0Rv8zqa
zO@R~7vFZw*UFXlm#D2YM2KNdxj2=Gum~LHeHpaM_2djVOkp#Q<pRnf#3GEkE>pbn$
zD2wM=(yEWWPghQ<cz38X7W9L|lv-nKyiG!k<_KBQWV)*ZD<8WOw_j;zX@bVcfAa(T
zat()D6ljqKi~OPVmm`aeI6B0)Sq5AUXA%e2m}Z4Dpo{>0#O&F6{t2amJ%NQIR!=~t
zwGkk`E^0yhhHs8PFtys0T}}<0CJL)+%8*=#?KX4bPS;3eY$m><=@UDO^f9%)<JT7n
z6|tcDF2>~QnkmljNQb_7&*MoY{O>ZJc3-ito_KVF-DXro7f!a#e!8fatB(mbB4!A&
z<D=F`^4J^1ir3kATngB$$)bpeSJe93f}g+-|BUx!c4m)JaPML?9wB$9)Zn&#wfpz&
zd|~Jqd4ap}fl)~k-TK?Uj0fVi?W$oRZyK_~D3RYHb@-&(Z^Q84q!1fO1#QP|p8%vl
zpW<c_JBJPfLm?2pZwXQR?}^Hi;p7v!gyB&CbLC;bd#b#Fe5P`>#ojDouKO(Y$dWlb
zJoe|t>*_?RM%BYTf~vE~nde?;B_!!ZJTg#uXd(PUUd`cR@``4ZNAZ=Bbo98QwK*J_
zk{(-jYsm@DW>d9Ad?2%}?zQUb^Z|c7{k~$Cv~Ef}4xj8Ktc5SDQJ*+q!7=gm8G2O1
z$;BI^d%fk}C|7yn7;>auT53HFwTh?yFe_{9YueA!kyp;Q<5LUe1W?R~Lzz|t&69f+
z#LN-(ZZX{^Xr5VI-o}ly(|vkqsYbl&J+Uq5OI+*)y*b|8o^>IFcWJ{-Zn2eZot46~
zSNTLz<@F1Pf|2g;CE6)-qgI*S+Fp6vum5h59;Y9+-v5DJ`D!7OYazxexibJaU!Q4N
z&cc~2#rb)EwY_=KZI9x)+4N`Bc`I!Q6Wq@T2NgdX71YWqW+d=TaIPOptTbf5`^l?1
zLssI7wL#+aaL<Zgp+(B;y>nWw-uMoYgl<xZ@Z8k<=DTMv?6@(06eiQcRe+Ih{S0HH
z;x}f|AhBEhn9r(2&0nWGGcWgB?JQ?zp*-o*dG!0ykV^p{{gZdR*wx)w<5c6lYsT^j
z`G1cdq&46Xl$!H?gs1uqva1*!%OBGGr33j6ed`+bbN5HgX5%=~7CuE@_=eb3rGArP
zXH&S`)vpWbkbfzxoFZD0X{rZdK~+at7GCR~@b2?k>9wEgE$)^%hneNahPmF=H~`rv
zm->;-xH{%Iyc3SMJV8{EKYDJHBgAaPD@T^tm+9X`A&OOGGMlMp&l)xNSN>7Go9g38
zQK@l9=dY&E?IIXLJ@Y@0b>4gR_O$EU!@{B3cXijFca8j{xL3a;e>v{&0$s)r&(qw=
z(b4<IN-jMryxI|iSEV@R85NO{KUMeET`=@HSLktV?pyABMB0#Zba|Dwz$vRoZL;?*
zGl|87NYZxlfFxt=F2v~=UHLNkMLz{U@1dBQsu^li$)1{*q*qdP;f}<<X*al$Fm{Ap
zn$2)(m(*tDK1YGIgjJ)X4z?|%b+t!n;r2?7>t%<L`Saj<|9K|n0CTyp)RS?C<w^RL
zpN`E04#d!Sb}O0nK-thDZk>Lg3TX4%Eb4kB1fP3<WDnh6X?ns{rpB|c({Yz=w%whP
zPxq;1Ld*c9d~R8Jm}3TCJ%KuHZ_H*2mmH0c<t}7fq_b(`R;-sDMg{XNv9%RClS|&c
z>mCN1kr$clf28hH!`N+J@z1?ix-k|`+wRf|>!Y_1qmBb<3WXE@j#PoE%VNBbc|7wh
z%XOqP^WU5M+%Rd6hZm;cQ|LP{zc{E75KyJqIy)>L;ItsYfD=a67PHm2lz!(76<`>P
zESNN67?WdYoJ;8*d27Rg_4#f(exaZLTJ846fEa%Y=oR?%D5aG65w3&w-T^1(JA~i$
zu;*RBh^U|S!(un(c4OHAj=U}!BUiOgtAZyZMx_0lpA9;mu*jjEN}b&bz???iPYohR
zFCJUG|8=BJ2v+`5ZwTJvx|NmIC;p)8jpMc9G-#~&s)3Za_L^?pD7!o$GI@9BEvKPE
zl)*oIL3~AHzL8u?&DzGmG&@XUXOJ%S3i-nbmvl=xo^?oAOwo>mLyoz3Mdz2)(A{`o
zKG>ywfqjG(PclA7Ob`j$d4)s+>59Z{_F6O5{r^tkAsW)lILUzHY!XW%=x=ShAxW4%
z#$h^y^_(*R7)?sU@{%X>dUl}wAvlNjd5`eN1O71XxF5B`kDwg=c6obiw{1V#!q3sH
za<A!lf0<;793vDPJ9IV}4;iRy@wAU<$L~M34|6X@=n4Z34?&aDZb~#@y54oo+59Ay
z`h;x7sRmMfnaC(c+e7Isu)#Y|(4pCDOEwnb^};b+vx`dsB~oW+F7LaGpK&?lrb+Rb
zfUx^}ARJliN+RG9^>I@BxqA|n64!v&;002P;nskjSPp%Xk=%byzYWAnVU(4@@<z~2
z5GEaU?X2W{;R*^Shrcol_KRe_<#%K)UeT|@m>x5Q><XO;b_M6l=#A|0z<WO!dXpFB
z2lN*nZNo&E+3P*RE1zYFOx12B*}JSd(?o!@N5L9UN*>o|ACXqyUwIS9t80pIbE6ZD
z^YEYSPCSBTM;B?N0t=mo)FWutx2rJV;(C5SUkz(Sy?Z}q)N2@Ay4iSNzVrB>vZj(q
z##kl<Ryv6~wAPr$9M=uA>1`j4*J+|#|G}Oouxji-Y{?zu>MyGQ0h85IwUR($#J!t~
zvLR5+tG|<p5xxJ6YUiJlxp0PvtIRw2%ccR2|HK{?{{D$)2vqK^p}>1pnTDshEvHj-
zaGg4{umzcT*RGr9yNwS%ACTvR*4S8n|36}bh4Y%vG%B2R9ZqT4wF$0Hw>!&fT#KXd
zF^vf?(1^AlatAh%+uDoVsMtKk3$Wkrk5TTNqbYL#eS)3~h@kNPn~x?ZT2m2QaLwNz
zMjSFpX>a`^t7AHU4Yb%uqxA*Aa#@L^PSdOjnIG_{bzN_{n=0%049tq$Ml1^gtDv&}
z0Qq2=>K2F({-ZzIUl@o*xzVjl(@%v>8=XK{A=|s%(0I%d+G=rx;8RrxvU~6yDbZP=
z0Q>JzP`b>FQ%~Ep_CvGSNF2102rZJY_%ueIf!om|^v}`Ttc*VRe|fJ;yl}pe%0nVy
zo{QZdjtR@v|EG;0|Fd-r*cB!<J|i~2r=aexDIcNprA@fxeiTMX$AsgOi8+PZ79BDA
z-#FlB^W_g?$UXzO%q?8K1|#-84|?2)Hjq-39&v_NTrOKM8@3LT{;Y-jX0v^j1G{4n
zP^7;lCAcFFj0h}CBDH3%f+TvezCPf3|ES`cnfAfF6EQyo8EB9&x{L-1X<Mc61U`FD
zp#0Bt+6m#|tWP%^nxwRW<+A+Wq1?>>XDDKD19~`;jS<%(x1QEr-g?^CwPy#L)!fD0
zI1~T&6Ogg!(YY{6TyHrcGPk2$^sovT=L?<h1cn4iC?%Rp>dYy1xcN(^6ps7fI|iIY
zW5FQTQcMHmPKX+1LZIB-=Zg{ix$pmezjfXz*qHu*DU%cT5rKaf&p^LVj{50RF<(e$
z45ht&+@3wHAMNoiuwWASrTWa;9=1Lmx&4?;-IBt|3I>GrOOdTh<kuGI9QGuQM2#e6
zmplOle14}dqz@XQ_0nYig&quM&?AkFQn!r(htMnZe2jc~pRcy?Pw2S#-(!8?<USOa
zPa~dumW6#bJs<>c%Z-xHQO`o4pnaz}^bh5>{J8LnVQlU(g?3_%RQ!msb^`4F&}$<~
zZ#%E9h}`gJwLQSbXSA|-d0j;gfjJyZ{74=GnPID67;OP0xax5n5}s4{B6IEST+WsL
z_iFf=C>YI*t4Z@m`}sSQNC)@J<sDhK+*ndNx-M#f?xr8lzX|6;S?~>OafkA<>8(xl
zo9k1DN*lQfs6Xdd7>|Nb)|P8Kfc$}+8CU2$#A6y3DSeNhU*x(!m-1c~pb54fA{*`&
zf2+;BoJjpAt`_<Yj_g_9!g8SI-{JomiF6y5=B_NvXzq$@+|5p^f%Gl5BD;aq?`=p=
zklI{e#rARZPAj>pt={}pySmLj^2un&6ZF{I8MC!Bz!l_fgqDSDk#v^n#q?{t1)>it
zW<I%LJF$muJty+&sb9E5|Lg`1uv<5$J)ZYqlBxv#Kq2>k9^U(7fcY~Ej*4A}B3;d}
z5PZCt>y;4v@r6j8y=Mi33Tw&83x1Mx>uu9X9z}dEq1^Sp#Mc9TY@$N&c27djnj%tE
z&vJyAU8IeCb6eSZ)GUL>%;nVqGxy(ih@y_Vf=BNadi(Eg@KgFf0|4owb2nU_er>&u
zH_wDc0_Au%cnmq&G88q2wCTxb!A>(ZvhO0CFB||&AxS>EzWv+CA<Wf0z|On)jWU$%
zbZa}x&!=9lFM;EkrrDn-Y?2O!w|gZ>tEG^&Ubj!^#Tc1gBJ>OG+j;z?|M$hN1~k86
zVV1Tyj1blFby<Lhdhq6#W5~I87~{a&oHNhfF2miBV8!dd{L9mguGn_62^3Jr=M!k4
z>vXJc;Ar4Ie8ZnAXLTJpR}ItqsFdR~h5Jo9*c}s`;tXGDUq769(dael+?ou{&qqGO
z3#hSE^Qk>;#IK6~GaB(SSpJ=rxXq508J|UdhwDH7Iz1rNA+s7F6##@(BQlqNo@&|Y
zGrlc)HQe&@T5@S)8F%B(XfB_Lg|_&!W3fX9HCiHp_0tW1e%3BIP=fp*M(wHkqQE@n
z*!BVb$~mVS1j^-$y}R%Nl41aw{(FLuPWK6QbJ@AyTe?NYNM6u<uBB(-=@^pa>#}kb
z`T@|qy+B<IXV@}ww;>l6$0Cv9i|4XOn{)y$KbO3DeW7qZP+YZS&cFse>RUQ`$E1!g
z$=p2G1*=hyX=~NBLwikW^u1!l+0Hq+6DU)IyYaYWRpE_@|3&~m?+U^TD|rL)408NB
zEI76h{7<j)BkT1AwXF!$g!)_3-WVwFu^q_oWSbr(KLX4q3Qk8BxsD%_U)XSRcOs`1
zX}Z=9-UX*`ttpQk(MNAgtc4&@iYtf)(jG6R<Lgu%dmc;={>N|x>g4~t*_{I_)@Wbf
zohsZC$DKBJ*t3b!%Ydth{u48!x&t05-<bBSb+`^4-Z;-P-j75p)Vv@;bA-;&M*@Qe
z>T6*hNUdaz8t~wg(H@G^^#7bgHrsV+vK$9ixsK3RJrROm@=}sm50t1X0Tl67_dnGk
z3`D%KlM(DvZ8hpWZxEM+MgZxJ@M6WWb<u&3qcO3n;lgaaB!)3ZhB@Wq^V_-HjR`pT
z{ysK;Mnq7v2&#<I7La}^u#{ZUM1R|MK>s9Z00bPbgZw<vesRTnC|}y4J@UjHhnr}i
zRA`YK8L22BiTn>+!Q0=2W^e*1_4C?*Bo=P^F<Al_!BS^m-BtYq953WrbBwrYA*S{_
z)QAVgi)H)|?ek4KL9NX({HtS-9gM&={ICFEsC#D@z7+Y(ovf-f8sEw*rk;=ydrjf{
z;a5Om2E(RCpBRqQqsQAzBB2@ef6tqc46YrVNl8s^JG-ae(Bw9u&fV&@*Cy<iO$QZ(
z7$3M+*N#OzU;F*D4?XCT+Ml!IsmTNKliC1No$CMR5O$sMjH6qZt-<L28>Dp@!0lgB
zo4gPTbQ3hT!k&eG8Nwx3@eD}*_j>rr7Y)t1=2}UwGP_4`=`@UBSykAkCN}i;wC@GK
zxEc(}C|@Y3P;3peUq@wg5%b!jmDYy2yepU;Vy<mN6%5JRJ<A~_Sye1zp9duSjRl*O
z2dqF0Vuz+J`o5%aD-%Ej^7aL~YfvKi-*ft^B#>wR%XO2I_Yu_DG#zjO0rz@0OztLM
z9j0imvWC1|B&98<AyCn&lX9UH^%~~Dr4h6y*|g!^E(EqyWqkrHuvT(yHSO57n^f4*
zV!nxB;=YxXbt;=T8gEb@*e(f{Cv!?;3tTZqaNxf%nl=Vbp;ka7arhnkPhJ$=iz*y8
zwo8pAsv4EyAat{Tj63@fQ9nbJcR|X95=MNsiSx=UNa#&6eu`4D;ELBwL<PX13m7q}
z@L@4`Y#WvN*Bj^Z2VvGQu#0-LRk$#(nMOM<l>g6%zp;@N*#QBmsf+?})J@gJX;Yr3
zmLGfXme?&sy$<HaI4Q#8uT{Qbjx+8zHTf3x_XzqF@*oD{$EX<H?bx+1np{#HF5rXx
zZ2w#iCPtk>PhvMuduMj2yyOd#|A{kfAkC=pO^rh~xu=Gbps0Mn?0*iN0zy*kpQ)cM
z8^Og9%x}>HlpG_kg;@M#u95W5ULLH$gx5}+0&(zp;^UV-9hMdx#ki!9ZxvQvavDr+
z$Nc6km73|2)_3P^|FmtK{^eT_XsmI|x_=jew(6JLn&d<8Iwz3kZu}pKFHE{%7KBhf
znj4GD&&v=fmh00-dr|TNK<NY;S-$JC<*Va2JS*$fe)x|j4osK#ZF>?{V042r@Fhf-
z2@&*Mnb^J7xQM$}U$$dvRjQWnm<Yw?`%}(q0xrJ7My~O{M$jz3_Svg?kB&(t1#f@s
zg2w;AKt%H0c^dGjuR&>WV)1ikX9#sDLf!S6u8fEq(1Q>z(4{A&Vkl#&xby(6Oz_oF
zAHugi0P=`kt1oxfwMS^$K>M+pR(s|htog@0GZ}&rr_b;H+tu*1sI@$Lj&XYu-{$3V
zH+^xwNRuH`(b9A_JsSN(#VxjW*@z|I=-LV!rQDx-lbUiOeQq!Z!RUOp@>lC73lv`}
zWNWQglyu;H-w_GhEB{wj*B;39`o<TXa>=a{-R$_eRTPzS3q=tn`c*nn<RGW~GEp3D
zOxmbl7bjiR85W(=ZLQqavI|xumsO`oW}~o}*~;a+%<T8fDZlmp;~#t9@B2Q_`#hiL
z^Stl-Y!JPEZ}r^NllJ=SdSy+~kz@=p`13FBPwww3XERd#1tr7#`Cqs15|*AgSeX3(
zuK*>jaAQ%C?#hOI6i(PX%1Jh8|I8daFNtn0?q6|6PxA6{S3DqI{xnt$f-E%G4#x?R
zm~h)q@D_q~Z8e%kZX0UrehZ|ETt4fr)A{l(gGH=Ic75)9@3CvLOoQ;~fW2baT*3m;
zn?3PeGBMsW(S6WTP0QXbn`5NXvJTY-PxF^DH%e1|H!2#}|4w&EVVCfFL*W5OO3M@D
zV2;y2GB%q$G_NRQ#w(A0$rod5M$euY$>)^M<MdNrc7F&VCNgWsDifvsdsOzc?^uYR
zCb)9$ca{lepT2zUcEA!gf_-6He$JOwp%);p+D`0Nl96_mFHx{%xH9Ru-(zVr-W|D_
zVFnlD-&iPbDqisKe1kFL9g`lE`C@R`O5Q*!-s@huPwM`lxyh2SgQono_SdNYFEYyF
zK)WStKYKi%-mEA6+nivM)p^usi>=~f^uCn9w^u7%qA3dwOO~^-w|iYE3cs52;Mr#L
zj>Fmf1ilgz!#?`)Tnk4jQMNs)XmN-f?|4Bnn0sF6q(2sV5Pwbl<vg-9wuJpDMQ4v|
z;`oo!edP#6^Qg#EH<0dln_*2?ftMQ#8I*<H%4T#3wv>rO_r})M<#D3lTX*?rbv<oo
zJ0YArfWPYrC#e<e5+PZ|Ke+<`L=rYXDyki_N=mnPp)JYVy|d5WJ+2Pnx#q&A$0(ku
zu$r1sAvT2#BRZXYCHt>wvOAUV__AcWAlc^)WkEW%4HGv6Q5c43i9^H!lKcRl0~WU@
zKzUyw2KJMR5iG0_o@){<&wLwNSpTeCM|i?`ta8j#mxbIZ&6&Zs_xVnCO|G%`R;j`)
zkWtDo^l~<aI<>O2;^vBc<K&IprRZK1t~oKlYj&DZPW%9j6<Uwo48Ht{rV#ic6(B+T
z7j#rul$F#D79QFU(v)mhjJWWE@Dgnnm6}+(8p9qjS$*b0w(7Y5w^jugRU{u_Y6z7G
zR<VwF*~U`FD)h_$W6ZthrkC-i)+dEQY23mnF_-E!n>9WvX=1`-4uFIH&@_3^9heYz
z`&Od045H4pOZ3~fWWIK~av9v=)nrGAjph67Go#>XT#GN;)t1xLC_~~>@H;lel@Py{
z5bs>tLikDG<Vc}!!N!%}H6P7P#vAy_j^z+8Vw4RoZrb4Z=3N84Nd)^}oII#o+1kAS
zXQ)1=bJ?WzMaVDre8A;){adR`M)4%m&A$HjeQNOctZ6l(706)XEcL1aN8S`bX5_)C
z#joyp_`K~9SeVfcpgjD7IR^ZR+(fQP9zc|bez%kl6tC+hO<xQ^JCW$8maNq%CEvh$
zt&BA(FAufg7G(i@oM6Kb3j=(tV?I#Xe(2vYyAj{DDDns{nA2le^NOZ?#`CYX0IeVm
zRN0E}<dbI!C1ihn=-m5l)$!Wt#b3^g^~BpePgEnM{I3X1W6#+ExPkh;r5oAE!d}n8
z*h>OA5<H_NAtyuflenlL_Vo45cd9nM5*hxzR@*7Z68Bk*Ux|dI!6Cn@V7K@}MI)bU
zO&uJAif$2m1(@gIt1ht~-B-D+JgW+}wl0f(HIt9oII6A=riM7`r_;1t;DTekXX^_`
z9Dq;F7=_%wJop~G;jFXAvMV@;AYm!!bWCp<xW>$v!`$}*KGxyu*eEjs%g?+1dQf7H
z&DcR@-_Cj$6ycR2$@Q9jUNAOvO?1Is<NKt5$n%*lh&}#+(qj=jO=fs&Y{anzc$&0T
zYklnyS;{oopVeb!&LAFp{sfPAAp=OBbaUjnXPNzA>Jnf?nqR&k&2gNco~?a9cWHx2
z!c-mvR@K&=@R1glM>lN)9i3{Co25+iKt}nuT}5NnJ%aSJ$#(V<$CXjjWKnDO>iWrQ
zLdm4irJq1kO^s%T9SrP1Lc4@mGk>!tk^$i^)GgEAHprm%uj?-;S*gbo3^d-rFr7<(
z1Xz)9c4C8SAWuEE@(8VUk|$qk#9MOAV-we{<ZupQVZ<7Sv_;rE9k{&xTLTn-0l#rV
zO4&!=HcEBpEE~FX+oiTsLyCAjHJu>{-|8&^7^4Q^9D1h<W_F&YU6~Y+R#qK^3{Ynh
zAg|_pzXZ4OPKr0EGE_lFIavRw%1!f^i6_iTY@A2NL`uWRxj6P*FXuGTGmr{pYN3EI
zG0c!?4nGZT@V#g_&wKW*BD3htCne}g>W7bWe_DDsbAuNhg)3RGf@rx2D|*TE-#`Uz
zyXqqK#~!L<#SebWw#+KEG3x7@p(!PrkL0*<*z(uUX9{+E&7S9hJJO5Wd01^Q>+_v*
z>)*$ldUlu-h=w&Rp6|kzPWfP^+t4mi>^%)BjY9>s{t9yf7zV0ymJYU#3vuW8Z-{2S
zrozTh)H~10yKiBGKa^H%B>FhZ(pD25%DO%@-Z@pEHqh1m84^<WXQx~wId0Uzd{G1L
zX@owDDo^gIYt#+i`5!E1&-+$N)EeAJO0jUBE`(Vq>&9wwnTf|lyvi$cMj3P;Mb+`r
zX)+btC$iH(V2pD=(0y0qVT=J7!vJnpaNM)k=~N@f>PWMye8m%CI|~Sw^u;R}Ux#_Q
zI>2|Lta%)~AgyO&(g%CnVjf&@i(X+Tq6C$*la+NBbXej3nvz@#ARVS!)q-kd`4<_f
z!S`>-{CBK21*VhsI9SV47++sef9UbNV2%kNcRbFj7%+3yJ_AJnwx7i%kR`BR9Ah6G
zNp9XAJ-8zpLC%&Kfk*i31Tti#F8cj|h?kE-|HZD==^zb=5hY9Yqq~!KtwEt5$+0C1
z;s|f!KQt8!(|!qB2%d*}(lZVpjhyU)Xmr~s-qF1NYUbQb-p~t>hQ!6ATV>CP9Ugo0
zT#UEF|4+3sxEigI&89Q(|8R2aNEDzC-*9!Se9M~F@u+s%w4C;?N{q)Eq$xw?9*a~7
zB(c0WXVRU*Kz&Y&PaE8?b3EG?kap^%MCesg(I7u}$`|DJy`^u>ka&a0uNY9^_NsGU
zN^t0WT=wDCc+^BelhMWQB-y5z3+YbeKVi=Ut8o2N*2FFM@<hBd4JqtyJM89d8(T2@
zUz(z}=pGHJWQE>7Al0-Yx;UOB`l8S3P9|uxG*!!eqV~%xN0jH=#3T-I_6Nx(Y1zNE
zbNeTy^4W=;ho#7%CK2kco77*$sp`i7$!xa#PtH!ADjS7u6tO*P$)tbb`mR&=FB|9J
z(7$H0AK~uKS)v901$%`92ebN$PoQu)gm4=grylwawQ~k*m+k|6EjB^|Zy@E%IfF$m
zTYHCHmzk2$>FA73c~i5-Ve~{B><d18#`kvJu^KT)WD6VQ5C~^e^813P3lN_-_WibS
z&v3E6Uoa*M&G`D1H}uQ~zRJP*Z+v@NAMEk)_KMYIUwT(wPQn|c<_sdkv}e^wV&4j|
zR>#37d~fOg(*?dyKVt0AAu5(jI<}hs25@O5yp>Tej(6Fk=PSS^#94i?gHM_<SjavY
zC?1P)#w(8|+!m}mz=U*YKAtI8EZ1g*>(qyIDiSsV?KRbc&UDI8rHqgC-00Jn?ufXU
zA!$br?4%lXKj(F)uHkKtBiggRcJpG5Vu1R1*r(5UopykGLfWrV)?}gco>j%X?L&96
zmoZD}|76XSM|*J_1E=~2kfT+N7>&q8up>4QED5Qp<Z4ugB$c#rKw7Z7Vg>g|>`0QV
zmXyt1?vn@aELL4{gIr1%tdF3L<@6seN!dN4#X?g`0%LnkWtPazdH+>8n5nh^DyGz!
z;Zsxa2@sJ)x^s)_JLL-I$U<NP{Nl30858e#7fzMJPfT62#~G<dAV)K)n;88|Zi*F4
zutGU{;ZaCa#T+RYjQ4>>YMJUAKHs?-St)%dYRIYq$lL{mSl|r=kHh`bJ1@3(g_JB{
zNozkQ+8V`xWpyYs8xxJbNhNE03a~$exzIm~#1m2DZZR|9^uZ;-slH`uzB~Pd5e{vU
z&1B&m0a@wU$qHUoKw$^^;y6nhM;EisgXZk98%`Abtb*PqNuGwRV*-qoyj8!u>0HT+
zu|??h(o`!w1{o<)TMdU<P};XKl4_q^W2|eg4K!{lo%>)L#Hwgw7h=f^VH!6xKfmJ*
zMUJcLOcDCnhuLMwdY}RM4^zd5qQnygJ1QoAKM7_hR`-@w#AheGYa+*1%_dYmG{m8I
zWGfO>*mD^n-fn1q?F@luSdi1N2+|$=_|pB%NY%%$im-=YITTK=NwcDhSBDPJ7ce#8
z&72IE-QJ+;`6(YT6CI7mJTI!P_pF7m-pD)qTveRFap#snZt_+D8HEGKyo|n>HC`K;
zNDUZ^g*k7%l@)f-vlLwa4R@tw%6_p%cjLX1Tc0H3yC|LVn~u`7SJ^+Sb<OdYrh2U#
zZTObUSq<+cW*^MGPh1mcMXPZNQwkH$Tuc)~$FBOU_hB-bMSXIGM6bP7{}Q0_S(C=T
zG)%2ORAzpse0s`2Zo{{<%;l335eQ~WFovj0RSxoUtrf??_Y4n7@p0?~h)sDF`Gw`d
zc`Kbl(MRp)s}gXZ?7S*uB)n51=Zsp<RZ9$U<n^jG{+4mUjlxhiIyJ@z+xjW-F84EU
zP?baU%YN)y_*Kf}5<dte4k3XyS%bk)ht=k0#y5k0%wDZu)D4KGGu%hB8+=<dq~?p7
z89IJikjPEYoCmhVg1&egW1=$AGjBK+ELBh~Fm=>{6o)*zgqEX-ZCcg44}+&V$kOht
zX3LwBV^*kq@KiJ1lp#5)A*}<yx|91B@Q!FA#p$;T0Cf{=amWaMEfTOBID+d>?vTsa
zsY*_LD^0byuJo9@1#@uW5R%92OPlHqGIwojsb=85TD3PX(lH0r`S@D$hLGfmi}(3D
zxj6<LPh$v(QMWNY_^XQECR&JcSjl?-q->1z$Nd0S2)%>9RT!T|P&d@3Gm~&JC+jAT
zq_}n{6z|@D#wOvynZ^&SXq?Zj_~gMgyMYn7o&r8!eodZco7VxGxUBtaJDwg<H))t#
zfB3*!Vy9c`_qFtMW<)}&0+EW|8o<<)ZM-5|Fyaos&xBm(JFJ5XGZFtTe}=%G7_}Vq
zvPX?b-p_;jSfWR^4RxLsZtI2`1lO?q%U~zo;UH1N9>yy;6|EC&Nn_rzsVn!eOr~4R
b2}<;l_^U;1W78+9b@SV_ZQ}#4{TcrQvWe4f

literal 0
HcmV?d00001

diff --git a/_static/logo.svg b/_static/logo.svg
new file mode 100644
index 000000000..c75ab9f79
--- /dev/null
+++ b/_static/logo.svg
@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" viewBox="0 0 1007.29 210.23"><defs><style>.cls-1{fill:url(#未命名的渐变_7);}.cls-2{fill:url(#未命名的渐变_7-2);}.cls-3{fill:url(#未命名的渐变_1143);}.cls-4{fill:url(#未命名的渐变_91);}.cls-5{fill:url(#未命名的渐变_284);}.cls-6{fill:url(#未命名的渐变_1177);}.cls-7{fill:#0065a6;}.cls-8{fill:#859997;}</style><linearGradient id="未命名的渐变_7" y1="105.12" x2="286" y2="105.12" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#2edfa3"/><stop offset="1" stop-color="#0065a6"/></linearGradient><linearGradient id="未命名的渐变_7-2" x1="14.8" y1="199" x2="71.53" y2="100.75" xlink:href="#未命名的渐变_7"/><linearGradient id="未命名的渐变_1143" x1="36.38" y1="81.47" x2="250.38" y2="-42.08" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#2edfa9"/><stop offset="1" stop-color="#2e86bf"/></linearGradient><linearGradient id="未命名的渐变_91" x1="143" y1="101.36" x2="225" y2="101.36" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#4285a3"/><stop offset="1" stop-color="#0065a6" stop-opacity="0"/></linearGradient><linearGradient id="未命名的渐变_284" x1="41.38" y1="14.66" x2="90.99" y2="88.73" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#2ee8ad"/><stop offset="0.61" stop-color="#2bd8a9" stop-opacity="0"/></linearGradient><linearGradient id="未命名的渐变_1177" x1="214.66" y1="188.82" x2="270.75" y2="91.67" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#2edfa3"/><stop offset="1" stop-color="#005891"/></linearGradient></defs><g id="图层_2" data-name="图层 2"><g id="图层_1-2" data-name="图层 1"><polygon class="cls-1" points="286 19.62 251.74 0 143 63.12 34.26 0 0 19.62 0 190.62 34.26 210.23 68.97 190.09 68.97 99.38 109 122.62 143 142.12 177 122.62 217.03 99.38 217.03 190.09 251.74 210.23 286 190.62 286 19.62"/><polygon class="cls-2" points="34.76 79.53 68.97 99.38 68.97 190.09 34.26 210.23 34.76 79.53"/><polygon class="cls-3" points="1.01 20.2 143 102.62 286 19.62 251.74 0 143 63.12 34.26 0 1.01 20.2"/><polygon class="cls-4" points="143 102.62 143 142.12 225 94.12 215.38 60.61 143 102.62"/><polygon class="cls-5" points="102.99 39.9 34.26 0 0 19.62 90.23 71.99 102.99 39.9"/><polygon class="cls-6" points="217.03 99.38 217.03 190.09 251.74 210.23 251.46 80.53 217 99.12 217 99.4 217.03 99.38"/><path class="cls-7" d="M390,83.31h5.54a8.84,8.84,0,0,1,8.84,8.85v70.91a8.85,8.85,0,0,0,8.85,8.85h29.4a8.85,8.85,0,0,1,8.85,8.85v.14a8.85,8.85,0,0,1-8.85,8.85H390a8.85,8.85,0,0,1-8.85-8.85V92.16A8.85,8.85,0,0,1,390,83.31Z"/><path class="cls-7" d="M476.68,83.31h10.89a8.83,8.83,0,0,1,8.22,5.58l17.57,44.19c2.51,6.62,4.71,13.67,7.06,20.57h.79c2.51-6.9,4.55-14,7.06-20.57L545.39,89a8.83,8.83,0,0,1,8.25-5.65h10.84a8.84,8.84,0,0,1,8.84,8.85v88.75a8.84,8.84,0,0,1-8.84,8.85H561a8.85,8.85,0,0,1-8.85-8.85v-35.6c0-10.07,1.73-24.89,2.83-35h-.63l-9.42,25-15.67,39.18a8.84,8.84,0,0,1-8.21,5.56h-1.39a8.84,8.84,0,0,1-8.2-5.54l-15.82-39.2-9.26-25h-.63c1.1,10.07,3,24.89,3,35v35.6a8.85,8.85,0,0,1-8.85,8.85h-3.18a8.85,8.85,0,0,1-8.85-8.85V92.16A8.85,8.85,0,0,1,476.68,83.31Z"/><path class="cls-8" d="M606.81,83.31H660.7a8.85,8.85,0,0,1,8.85,8.85h0A8.84,8.84,0,0,1,660.7,101H630.05a8.85,8.85,0,0,0-8.85,8.85v10.07a8.85,8.85,0,0,0,8.85,8.85h23.59a8.84,8.84,0,0,1,8.85,8.84v.15a8.84,8.84,0,0,1-8.85,8.84H630.05a8.85,8.85,0,0,0-8.85,8.85v25.46a8.85,8.85,0,0,1-8.85,8.85h-5.54a8.84,8.84,0,0,1-8.84-8.85V92.16A8.84,8.84,0,0,1,606.81,83.31Z"/><path class="cls-8" d="M694.72,83.31h5.54a8.85,8.85,0,0,1,8.85,8.85v70.91a8.84,8.84,0,0,0,8.84,8.85h29.4a8.85,8.85,0,0,1,8.85,8.85v.14a8.85,8.85,0,0,1-8.85,8.85H694.72a8.85,8.85,0,0,1-8.85-8.85V92.16A8.85,8.85,0,0,1,694.72,83.31Z"/><path class="cls-8" d="M760.75,136.1c0-34.66,21.2-54.8,52-54.8s52,20.28,52,54.8-21.19,55.67-52,55.67S760.75,170.63,760.75,136.1Zm80.22,0c0-22.87-11-36.53-28.26-36.53s-28.25,13.66-28.25,36.53,11,37.26,28.25,37.26S841,158.83,841,136.1Z"/><path class="cls-8" d="M882.48,83.31h5.39a8.85,8.85,0,0,1,8.71,7.3l8,44.63c1.73,11.22,3.62,22.58,5.34,33.95h.63c2.36-11.37,4.87-22.87,7.38-33.95l11.78-45.31a8.84,8.84,0,0,1,8.56-6.62h6.1a8.84,8.84,0,0,1,8.56,6.62l11.78,45.31c2.51,10.93,4.87,22.44,7.38,33.95h.78c1.73-11.51,3.46-22.87,5.18-33.95l7.82-44.61a8.86,8.86,0,0,1,8.72-7.32h3.89a8.84,8.84,0,0,1,8.67,10.6l-17.93,88.75a8.84,8.84,0,0,1-8.67,7.1H965.79A8.85,8.85,0,0,1,957.2,183l-11.37-45.93c-1.88-8.34-3.61-16.54-5-24.6h-.63c-1.57,8.06-3.14,16.26-5,24.6L924.08,183a8.85,8.85,0,0,1-8.6,6.77H901.2a8.84,8.84,0,0,1-8.65-7L873.83,94A8.84,8.84,0,0,1,882.48,83.31Z"/></g></g></svg>
\ No newline at end of file
diff --git a/_static/logo2.svg b/_static/logo2.svg
new file mode 100644
index 000000000..d823b6a3d
--- /dev/null
+++ b/_static/logo2.svg
@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" viewBox="0 0 1008.29 210.23"><defs><style>.cls-1{fill:url(#未命名的渐变_532);}.cls-2{fill:url(#未命名的渐变_713);}.cls-3{fill:url(#未命名的渐变_532-2);}.cls-4,.cls-7{opacity:0.45;}.cls-4{fill:url(#未命名的渐变_745);}.cls-5{fill:url(#未命名的渐变_565);}.cls-6{fill:url(#未命名的渐变_713-2);}.cls-7{fill:url(#未命名的渐变_1087);}.cls-8{fill:#45d4ff;}.cls-9{fill:#828b99;}</style><linearGradient id="未命名的渐变_532" y1="105.12" x2="286" y2="105.12" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#45d4ff"/><stop offset="1" stop-color="#c170cd"/></linearGradient><linearGradient id="未命名的渐变_713" x1="14.8" y1="199" x2="71.53" y2="100.75" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#45d4ff"/><stop offset="1" stop-color="#3170cd"/></linearGradient><linearGradient id="未命名的渐变_532-2" x1="36.38" y1="81.47" x2="250.38" y2="-42.08" xlink:href="#未命名的渐变_532"/><linearGradient id="未命名的渐变_745" x1="143" y1="101.36" x2="225" y2="101.36" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#a45ba3"/><stop offset="1" stop-color="#b765a6" stop-opacity="0"/></linearGradient><linearGradient id="未命名的渐变_565" x1="41.38" y1="14.66" x2="90.99" y2="88.73" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#2ee8ff"/><stop offset="0.61" stop-color="#2bd8a9" stop-opacity="0"/></linearGradient><linearGradient id="未命名的渐变_713-2" x1="214.66" y1="188.82" x2="270.75" y2="91.67" xlink:href="#未命名的渐变_713"/><linearGradient id="未命名的渐变_1087" x1="287.81" y1="25.07" x2="221.63" y2="127.93" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#542653"/><stop offset="0.23" stop-color="#6e3769" stop-opacity="0.74"/><stop offset="0.74" stop-color="#a35895" stop-opacity="0.21"/><stop offset="1" stop-color="#b765a6" stop-opacity="0"/></linearGradient></defs><g id="图层_2" data-name="图层 2"><g id="图层_1-2" data-name="图层 1"><polygon class="cls-1" points="286 19.62 251.74 0 143 63.12 34.26 0 0 19.62 0 190.62 34.26 210.23 68.97 190.09 68.97 99.38 109 122.62 143 142.12 177 122.62 217.03 99.38 217.03 190.09 251.74 210.23 286 190.62 286 19.62"/><polygon class="cls-2" points="34.76 79.53 68.97 99.38 68.97 190.09 34.26 210.23 34.76 79.53"/><polygon class="cls-3" points="1.01 20.2 143 102.62 286 19.62 251.74 0 143 63.12 34.26 0 1.01 20.2"/><polygon class="cls-4" points="143 102.62 143 142.12 225 94.12 215.38 60.61 143 102.62"/><polygon class="cls-5" points="102.99 39.9 34.26 0 0 19.62 90.23 71.99 102.99 39.9"/><polygon class="cls-6" points="217.03 99.38 217.03 190.09 251.74 210.23 251.46 80.53 217 99.12 217 99.4 217.03 99.38"/><polygon class="cls-7" points="177.5 83.53 286 19.62 286 140.86 251.74 132.14 251.46 80.53 209.25 103.74 177.5 83.53"/><path class="cls-8" d="M391,83.31h5.54a8.84,8.84,0,0,1,8.84,8.85v70.91a8.85,8.85,0,0,0,8.85,8.85h29.4a8.85,8.85,0,0,1,8.85,8.85v.14a8.85,8.85,0,0,1-8.85,8.85H391a8.85,8.85,0,0,1-8.85-8.85V92.16A8.85,8.85,0,0,1,391,83.31Z"/><path class="cls-8" d="M477.68,83.31h10.89a8.83,8.83,0,0,1,8.22,5.58l17.57,44.19c2.51,6.62,4.71,13.67,7.06,20.57h.79c2.51-6.9,4.55-14,7.06-20.57L546.39,89a8.83,8.83,0,0,1,8.25-5.65h10.84a8.84,8.84,0,0,1,8.84,8.85v88.75a8.84,8.84,0,0,1-8.84,8.85H562a8.85,8.85,0,0,1-8.85-8.85v-35.6c0-10.07,1.73-24.89,2.83-35h-.63l-9.42,25-15.67,39.18a8.84,8.84,0,0,1-8.21,5.56h-1.39a8.84,8.84,0,0,1-8.2-5.54l-15.82-39.2-9.26-25h-.63c1.1,10.07,3,24.89,3,35v35.6a8.85,8.85,0,0,1-8.85,8.85h-3.18a8.85,8.85,0,0,1-8.85-8.85V92.16A8.85,8.85,0,0,1,477.68,83.31Z"/><path class="cls-9" d="M607.81,83.31H661.7a8.85,8.85,0,0,1,8.85,8.85h0A8.84,8.84,0,0,1,661.7,101H631.05a8.85,8.85,0,0,0-8.85,8.85v10.07a8.85,8.85,0,0,0,8.85,8.85h23.59a8.84,8.84,0,0,1,8.85,8.84v.15a8.84,8.84,0,0,1-8.85,8.84H631.05a8.85,8.85,0,0,0-8.85,8.85v25.46a8.85,8.85,0,0,1-8.85,8.85h-5.54a8.84,8.84,0,0,1-8.84-8.85V92.16A8.84,8.84,0,0,1,607.81,83.31Z"/><path class="cls-9" d="M695.72,83.31h5.54a8.85,8.85,0,0,1,8.85,8.85v70.91a8.84,8.84,0,0,0,8.84,8.85h29.4a8.85,8.85,0,0,1,8.85,8.85v.14a8.85,8.85,0,0,1-8.85,8.85H695.72a8.85,8.85,0,0,1-8.85-8.85V92.16A8.85,8.85,0,0,1,695.72,83.31Z"/><path class="cls-9" d="M761.75,136.1c0-34.66,21.2-54.8,52-54.8s52,20.28,52,54.8-21.19,55.67-52,55.67S761.75,170.63,761.75,136.1Zm80.22,0c0-22.87-11-36.53-28.26-36.53s-28.25,13.66-28.25,36.53,11,37.26,28.25,37.26S842,158.83,842,136.1Z"/><path class="cls-9" d="M883.48,83.31h5.39a8.85,8.85,0,0,1,8.71,7.3l8,44.63c1.73,11.22,3.62,22.58,5.34,33.95h.63c2.36-11.37,4.87-22.87,7.38-33.95l11.78-45.31a8.84,8.84,0,0,1,8.56-6.62h6.1a8.84,8.84,0,0,1,8.56,6.62l11.78,45.31c2.51,10.93,4.87,22.44,7.38,33.95h.78c1.73-11.51,3.46-22.87,5.18-33.95l7.82-44.61a8.86,8.86,0,0,1,8.72-7.32h3.89a8.84,8.84,0,0,1,8.67,10.6l-17.93,88.75a8.84,8.84,0,0,1-8.67,7.1H966.79A8.85,8.85,0,0,1,958.2,183l-11.37-45.93c-1.88-8.34-3.61-16.54-5-24.6h-.63c-1.57,8.06-3.14,16.26-5,24.6L925.08,183a8.85,8.85,0,0,1-8.6,6.77H902.2a8.84,8.84,0,0,1-8.65-7L874.83,94A8.84,8.84,0,0,1,883.48,83.31Z"/></g></g></svg>
\ No newline at end of file
diff --git a/_static/logo3.svg b/_static/logo3.svg
new file mode 100644
index 000000000..c092db5d0
--- /dev/null
+++ b/_static/logo3.svg
@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" viewBox="0 0 1006.29 210.23"><defs><style>.cls-1{fill:url(#未命名的渐变_1031);}.cls-2{fill:url(#未命名的渐变_862);}.cls-3{fill:url(#未命名的渐变_1230);}.cls-4{opacity:0.45;fill:url(#未命名的渐变_877);}.cls-5{fill:url(#未命名的渐变_855);}.cls-6{fill:url(#未命名的渐变_1080);}.cls-7{opacity:0.42;fill:url(#未命名的渐变_1218);}.cls-8{fill:#6466e1;}.cls-9{fill:#8b8499;}</style><linearGradient id="未命名的渐变_1031" y1="105.12" x2="286" y2="105.12" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#d6caff"/><stop offset="1" stop-color="#30a4e1"/></linearGradient><linearGradient id="未命名的渐变_862" x1="14.8" y1="199" x2="71.53" y2="100.75" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#d6caff"/><stop offset="1" stop-color="#6466e1"/></linearGradient><linearGradient id="未命名的渐变_1230" x1="36.38" y1="81.47" x2="250.38" y2="-42.08" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#d6caff"/><stop offset="1" stop-color="#46abe1"/></linearGradient><linearGradient id="未命名的渐变_877" x1="143" y1="101.36" x2="225" y2="101.36" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#4941a3"/><stop offset="1" stop-color="#b765a6" stop-opacity="0"/></linearGradient><linearGradient id="未命名的渐变_855" x1="41.38" y1="14.66" x2="90.99" y2="88.73" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#e2d6ff"/><stop offset="0.61" stop-color="#d6caff" stop-opacity="0"/></linearGradient><linearGradient id="未命名的渐变_1080" x1="214.66" y1="188.82" x2="270.75" y2="91.67" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#86b8f1"/><stop offset="1" stop-color="#6466e1"/></linearGradient><linearGradient id="未命名的渐变_1218" x1="289.98" y1="21.01" x2="244.21" y2="94.95" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#3a71ba"/><stop offset="0.15" stop-color="#3a72bb" stop-opacity="0.98"/><stop offset="0.29" stop-color="#3a76be" stop-opacity="0.91"/><stop offset="0.44" stop-color="#3b7bc2" stop-opacity="0.81"/><stop offset="0.59" stop-color="#3b84c8" stop-opacity="0.65"/><stop offset="0.74" stop-color="#3c8ed0" stop-opacity="0.46"/><stop offset="0.88" stop-color="#3d9bda" stop-opacity="0.22"/><stop offset="1" stop-color="#3ea7e3" stop-opacity="0"/></linearGradient></defs><g id="图层_2" data-name="图层 2"><g id="图层_1-2" data-name="图层 1"><polygon class="cls-1" points="286 19.62 251.74 0 143 63.12 34.26 0 0 19.62 0 190.62 34.26 210.23 68.97 190.09 68.97 99.38 109 122.62 143 142.12 177 122.62 217.03 99.38 217.03 190.09 251.74 210.23 286 190.62 286 19.62"/><polygon class="cls-2" points="34.76 79.53 68.97 99.38 68.97 190.09 34.26 210.23 34.76 79.53"/><polygon class="cls-3" points="1.01 20.2 143 102.62 286 19.62 251.74 0 143 63.12 34.26 0 1.01 20.2"/><polygon class="cls-4" points="143 102.62 143 142.12 225 94.12 215.38 60.61 143 102.62"/><polygon class="cls-5" points="102.99 39.9 34.26 0 0 19.62 90.23 71.99 102.99 39.9"/><polygon class="cls-6" points="217.03 99.38 217.03 190.09 251.74 210.23 251.46 80.53 217 99.12 217 99.4 217.03 99.38"/><polygon class="cls-7" points="201 69.12 286 19.62 286 123.07 251.74 124.71 251.46 80.53 216 99.98 201 69.12"/><path class="cls-8" d="M389,85.31h5.54a8.84,8.84,0,0,1,8.84,8.85v70.91a8.85,8.85,0,0,0,8.85,8.85h29.4a8.85,8.85,0,0,1,8.85,8.85v.14a8.85,8.85,0,0,1-8.85,8.85H389a8.85,8.85,0,0,1-8.85-8.85V94.16A8.85,8.85,0,0,1,389,85.31Z"/><path class="cls-8" d="M475.68,85.31h10.89a8.83,8.83,0,0,1,8.22,5.58l17.57,44.19c2.51,6.62,4.71,13.67,7.06,20.57h.79c2.51-6.9,4.55-14,7.06-20.57L544.39,91a8.83,8.83,0,0,1,8.25-5.65h10.84a8.84,8.84,0,0,1,8.84,8.85v88.75a8.84,8.84,0,0,1-8.84,8.85H560a8.85,8.85,0,0,1-8.85-8.85v-35.6c0-10.07,1.73-24.89,2.83-35h-.63l-9.42,25-15.67,39.18a8.84,8.84,0,0,1-8.21,5.56h-1.39a8.84,8.84,0,0,1-8.2-5.54l-15.82-39.2-9.26-25h-.63c1.1,10.07,3,24.89,3,35v35.6a8.85,8.85,0,0,1-8.85,8.85h-3.18a8.85,8.85,0,0,1-8.85-8.85V94.16A8.85,8.85,0,0,1,475.68,85.31Z"/><path class="cls-9" d="M605.81,85.31H659.7a8.85,8.85,0,0,1,8.85,8.85h0A8.84,8.84,0,0,1,659.7,103H629.05a8.85,8.85,0,0,0-8.85,8.85v10.07a8.85,8.85,0,0,0,8.85,8.85h23.59a8.84,8.84,0,0,1,8.85,8.84v.15a8.84,8.84,0,0,1-8.85,8.84H629.05a8.85,8.85,0,0,0-8.85,8.85v25.46a8.85,8.85,0,0,1-8.85,8.85h-5.54a8.84,8.84,0,0,1-8.84-8.85V94.16A8.84,8.84,0,0,1,605.81,85.31Z"/><path class="cls-9" d="M693.72,85.31h5.54a8.85,8.85,0,0,1,8.85,8.85v70.91a8.84,8.84,0,0,0,8.84,8.85h29.4a8.85,8.85,0,0,1,8.85,8.85v.14a8.85,8.85,0,0,1-8.85,8.85H693.72a8.85,8.85,0,0,1-8.85-8.85V94.16A8.85,8.85,0,0,1,693.72,85.31Z"/><path class="cls-9" d="M759.75,138.1c0-34.66,21.2-54.8,52-54.8s52,20.28,52,54.8-21.19,55.67-52,55.67S759.75,172.63,759.75,138.1Zm80.22,0c0-22.87-11-36.53-28.26-36.53s-28.25,13.66-28.25,36.53,11,37.26,28.25,37.26S840,160.83,840,138.1Z"/><path class="cls-9" d="M881.48,85.31h5.39a8.85,8.85,0,0,1,8.71,7.3l8,44.63c1.73,11.22,3.62,22.58,5.34,33.95h.63c2.36-11.37,4.87-22.87,7.38-33.95l11.78-45.31a8.84,8.84,0,0,1,8.56-6.62h6.1a8.84,8.84,0,0,1,8.56,6.62l11.78,45.31c2.51,10.93,4.87,22.44,7.38,33.95h.78c1.73-11.51,3.46-22.87,5.18-33.95l7.82-44.61a8.86,8.86,0,0,1,8.72-7.32h3.89a8.84,8.84,0,0,1,8.67,10.6l-17.93,88.75a8.84,8.84,0,0,1-8.67,7.1H964.79A8.85,8.85,0,0,1,956.2,185l-11.37-45.93c-1.88-8.34-3.61-16.54-5-24.6h-.63c-1.57,8.06-3.14,16.26-5,24.6L923.08,185a8.85,8.85,0,0,1-8.6,6.77H900.2a8.84,8.84,0,0,1-8.65-7L872.83,96A8.84,8.84,0,0,1,881.48,85.31Z"/></g></g></svg>
\ No newline at end of file
diff --git a/_static/logo4.svg b/_static/logo4.svg
new file mode 100644
index 000000000..afbb2c14f
--- /dev/null
+++ b/_static/logo4.svg
@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" viewBox="0 0 286 210.23"><defs><style>.cls-1{fill:url(#未命名的渐变_532);}.cls-2{fill:url(#未命名的渐变_713);}.cls-3{fill:url(#未命名的渐变_532-2);}.cls-4,.cls-7{opacity:0.45;}.cls-4{fill:url(#未命名的渐变_745);}.cls-5{fill:url(#未命名的渐变_565);}.cls-6{fill:url(#未命名的渐变_713-2);}.cls-7{fill:url(#未命名的渐变_1087);}</style><linearGradient id="未命名的渐变_532" y1="105.12" x2="286" y2="105.12" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#45d4ff"/><stop offset="1" stop-color="#c170cd"/></linearGradient><linearGradient id="未命名的渐变_713" x1="14.8" y1="199" x2="71.53" y2="100.75" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#45d4ff"/><stop offset="1" stop-color="#3170cd"/></linearGradient><linearGradient id="未命名的渐变_532-2" x1="36.38" y1="81.47" x2="250.38" y2="-42.08" xlink:href="#未命名的渐变_532"/><linearGradient id="未命名的渐变_745" x1="143" y1="101.36" x2="225" y2="101.36" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#a45ba3"/><stop offset="1" stop-color="#b765a6" stop-opacity="0"/></linearGradient><linearGradient id="未命名的渐变_565" x1="41.38" y1="14.66" x2="90.99" y2="88.73" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#2ee8ff"/><stop offset="0.61" stop-color="#2bd8a9" stop-opacity="0"/></linearGradient><linearGradient id="未命名的渐变_713-2" x1="214.66" y1="188.82" x2="270.75" y2="91.67" xlink:href="#未命名的渐变_713"/><linearGradient id="未命名的渐变_1087" x1="287.81" y1="25.07" x2="221.63" y2="127.93" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#542653"/><stop offset="0.23" stop-color="#6e3769" stop-opacity="0.74"/><stop offset="0.74" stop-color="#a35895" stop-opacity="0.21"/><stop offset="1" stop-color="#b765a6" stop-opacity="0"/></linearGradient></defs><g id="图层_2" data-name="图层 2"><g id="图层_1-2" data-name="图层 1"><polygon class="cls-1" points="286 19.62 251.74 0 143 63.12 34.26 0 0 19.62 0 190.62 34.26 210.23 68.97 190.09 68.97 99.38 109 122.62 143 142.12 177 122.62 217.03 99.38 217.03 190.09 251.74 210.23 286 190.62 286 19.62"/><polygon class="cls-2" points="34.76 79.53 68.97 99.38 68.97 190.09 34.26 210.23 34.76 79.53"/><polygon class="cls-3" points="1.01 20.2 143 102.62 286 19.62 251.74 0 143 63.12 34.26 0 1.01 20.2"/><polygon class="cls-4" points="143 102.62 143 142.12 225 94.12 215.38 60.61 143 102.62"/><polygon class="cls-5" points="102.99 39.9 34.26 0 0 19.62 90.23 71.99 102.99 39.9"/><polygon class="cls-6" points="217.03 99.38 217.03 190.09 251.74 210.23 251.46 80.53 217 99.12 217 99.4 217.03 99.38"/><polygon class="cls-7" points="177.5 83.53 286 19.62 286 140.86 251.74 132.14 251.46 80.53 209.25 103.74 177.5 83.53"/></g></g></svg>
\ No newline at end of file
diff --git a/_static/logo5.svg b/_static/logo5.svg
new file mode 100644
index 000000000..877055168
--- /dev/null
+++ b/_static/logo5.svg
@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" viewBox="0 0 286 210.23"><defs><style>.cls-1{fill:url(#未命名的渐变_7);}.cls-2{fill:url(#未命名的渐变_7-2);}.cls-3{fill:url(#未命名的渐变_1143);}.cls-4{fill:url(#未命名的渐变_91);}.cls-5{fill:url(#未命名的渐变_284);}.cls-6{fill:url(#未命名的渐变_1177);}</style><linearGradient id="未命名的渐变_7" y1="105.12" x2="286" y2="105.12" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#2edfa3"/><stop offset="1" stop-color="#0065a6"/></linearGradient><linearGradient id="未命名的渐变_7-2" x1="14.8" y1="199" x2="71.53" y2="100.75" xlink:href="#未命名的渐变_7"/><linearGradient id="未命名的渐变_1143" x1="36.38" y1="81.47" x2="250.38" y2="-42.08" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#2edfa9"/><stop offset="1" stop-color="#2e86bf"/></linearGradient><linearGradient id="未命名的渐变_91" x1="143" y1="101.36" x2="225" y2="101.36" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#4285a3"/><stop offset="1" stop-color="#0065a6" stop-opacity="0"/></linearGradient><linearGradient id="未命名的渐变_284" x1="41.38" y1="14.66" x2="90.99" y2="88.73" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#2ee8ad"/><stop offset="0.61" stop-color="#2bd8a9" stop-opacity="0"/></linearGradient><linearGradient id="未命名的渐变_1177" x1="214.66" y1="188.82" x2="270.75" y2="91.67" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#2edfa3"/><stop offset="1" stop-color="#005891"/></linearGradient></defs><g id="图层_2" data-name="图层 2"><g id="图层_1-2" data-name="图层 1"><polygon class="cls-1" points="286 19.62 251.74 0 143 63.12 34.26 0 0 19.62 0 190.62 34.26 210.23 68.97 190.09 68.97 99.38 109 122.62 143 142.12 177 122.62 217.03 99.38 217.03 190.09 251.74 210.23 286 190.62 286 19.62"/><polygon class="cls-2" points="34.76 79.53 68.97 99.38 68.97 190.09 34.26 210.23 34.76 79.53"/><polygon class="cls-3" points="1.01 20.2 143 102.62 286 19.62 251.74 0 143 63.12 34.26 0 1.01 20.2"/><polygon class="cls-4" points="143 102.62 143 142.12 225 94.12 215.38 60.61 143 102.62"/><polygon class="cls-5" points="102.99 39.9 34.26 0 0 19.62 90.23 71.99 102.99 39.9"/><polygon class="cls-6" points="217.03 99.38 217.03 190.09 251.74 210.23 251.46 80.53 217 99.12 217 99.4 217.03 99.38"/></g></g></svg>
\ No newline at end of file
diff --git a/_static/logo6.svg b/_static/logo6.svg
new file mode 100644
index 000000000..6eeade79e
--- /dev/null
+++ b/_static/logo6.svg
@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" viewBox="0 0 286 210.23"><defs><style>.cls-1{fill:url(#未命名的渐变_1031);}.cls-2{fill:url(#未命名的渐变_862);}.cls-3{fill:url(#未命名的渐变_1230);}.cls-4{opacity:0.45;fill:url(#未命名的渐变_877);}.cls-5{fill:url(#未命名的渐变_855);}.cls-6{fill:url(#未命名的渐变_1080);}.cls-7{opacity:0.42;fill:url(#未命名的渐变_1218);}</style><linearGradient id="未命名的渐变_1031" y1="105.12" x2="286" y2="105.12" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#d6caff"/><stop offset="1" stop-color="#30a4e1"/></linearGradient><linearGradient id="未命名的渐变_862" x1="14.8" y1="199" x2="71.53" y2="100.75" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#d6caff"/><stop offset="1" stop-color="#6466e1"/></linearGradient><linearGradient id="未命名的渐变_1230" x1="36.38" y1="81.47" x2="250.38" y2="-42.08" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#d6caff"/><stop offset="1" stop-color="#46abe1"/></linearGradient><linearGradient id="未命名的渐变_877" x1="143" y1="101.36" x2="225" y2="101.36" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#4941a3"/><stop offset="1" stop-color="#b765a6" stop-opacity="0"/></linearGradient><linearGradient id="未命名的渐变_855" x1="41.38" y1="14.66" x2="90.99" y2="88.73" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#e2d6ff"/><stop offset="0.61" stop-color="#d6caff" stop-opacity="0"/></linearGradient><linearGradient id="未命名的渐变_1080" x1="214.66" y1="188.82" x2="270.75" y2="91.67" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#86b8f1"/><stop offset="1" stop-color="#6466e1"/></linearGradient><linearGradient id="未命名的渐变_1218" x1="289.98" y1="21.01" x2="244.21" y2="94.95" gradientUnits="userSpaceOnUse"><stop offset="0" stop-color="#3a71ba"/><stop offset="0.15" stop-color="#3a72bb" stop-opacity="0.98"/><stop offset="0.29" stop-color="#3a76be" stop-opacity="0.91"/><stop offset="0.44" stop-color="#3b7bc2" stop-opacity="0.81"/><stop offset="0.59" stop-color="#3b84c8" stop-opacity="0.65"/><stop offset="0.74" stop-color="#3c8ed0" stop-opacity="0.46"/><stop offset="0.88" stop-color="#3d9bda" stop-opacity="0.22"/><stop offset="1" stop-color="#3ea7e3" stop-opacity="0"/></linearGradient></defs><g id="图层_2" data-name="图层 2"><g id="图层_1-2" data-name="图层 1"><polygon class="cls-1" points="286 19.62 251.74 0 143 63.12 34.26 0 0 19.62 0 190.62 34.26 210.23 68.97 190.09 68.97 99.38 109 122.62 143 142.12 177 122.62 217.03 99.38 217.03 190.09 251.74 210.23 286 190.62 286 19.62"/><polygon class="cls-2" points="34.76 79.53 68.97 99.38 68.97 190.09 34.26 210.23 34.76 79.53"/><polygon class="cls-3" points="1.01 20.2 143 102.62 286 19.62 251.74 0 143 63.12 34.26 0 1.01 20.2"/><polygon class="cls-4" points="143 102.62 143 142.12 225 94.12 215.38 60.61 143 102.62"/><polygon class="cls-5" points="102.99 39.9 34.26 0 0 19.62 90.23 71.99 102.99 39.9"/><polygon class="cls-6" points="217.03 99.38 217.03 190.09 251.74 210.23 251.46 80.53 217 99.12 217 99.4 217.03 99.38"/><polygon class="cls-7" points="201 69.12 286 19.62 286 123.07 251.74 124.71 251.46 80.53 216 99.98 201 69.12"/></g></g></svg>
\ No newline at end of file
diff --git a/_static/minus.png b/_static/minus.png
new file mode 100644
index 0000000000000000000000000000000000000000..d96755fdaf8bb2214971e0db9c1fd3077d7c419d
GIT binary patch
literal 90
zcmeAS@N?(olHy`uVBq!ia0vp^+#t*WBp7;*Yy1LIik>cxAr*|t7R?Mi>2?kWtu=nj
kDsEF_5m^0CR;1wuP-*O&G^0G}KYk!hp00i_>zopr08q^qX#fBK

literal 0
HcmV?d00001

diff --git a/_static/nll.png b/_static/nll.png
new file mode 100644
index 0000000000000000000000000000000000000000..f5f7d0b7d234666c443eb19fce5b4ac5a9bbc2b3
GIT binary patch
literal 55999
zcmcG!Ra9JEvjvJf!J%;i0h-|M?ruqN4;I`TcXx;28o?z<aA;hEySq2;+|K#$c{tzy
z-N!q|?lE?6=(ShbtXWknQcXn`6O9B71_lOGUQS8_1_nVF1_l-e1qu2j7!S7u1_ndK
zT2fL?UQ&`u&Dp`i+SVKfMlLc{3t3xpkT6#-DM8GtKoPYKm6|&s1zUHYyh}z7E(F(%
zDq?=PFvy08pt_)pI*J@y(9}}>*jHRiT89Xo=fi4oK;`4m?e635nExyPWgh3=*s?b=
z%r8x{L?F>eWGYo{syKAi{CH*glH^=ivVA!GeRzTJELPJ#uoR4T*kVWeZ6ge|U6r8R
zua|{aqiCe~ym3UB$UwfF^zqT)J8T#bfz~?~m;m6;=r_WMV!)?-+Avs&R=Ft7ohF5-
zsU1+>2;JK_L??6@h4*A}e0VT3WeMMioH+}^zEFZyByz=20!YYB*IHW#g(9Y?8UlH!
zBT61(t|(nOUQ07Y$_F<-zcIVl|18a*mcTl+qx2{`1?f;Uto%F>PEh_{XLEw@Alqd1
zbX)07x+y_~cj$roA8Z?mC-u(c1<4qV+IIz!&fe1rS>r+c@B-L^5Wc_-Y*Gm_rjghv
zwethwAY&e|V`#o~5S%+4*bAF=a}T+f(UvEgzMsLT;F{Q!P_xTuJ;L0tq@6<E88;K{
znODcBL4s0~c}zhrECa4ax`2uS&Gcgg`K8Nd4*`!^RGqfEB#)_lHkTiI#MWJR0!*{W
zr;N96;xy;ecXzNxiI%cyjb$0m1B1##)I9yF30Y(bMYGY}Q&~ULy*W6)?3ck94eiIm
zNE#>i9wVabV{{dXLWA4fHgk#s^2(URXF*&fi(84omG_~Ls2Y$4tbT&CuaM*i>}0BV
zJl3(@klul<uJmA(PPo8;H}9%Tg9oWdQ5L;Oy*6@M=>d)bfX{Fuj%sX@G)d;d5};V*
z-gVW8J1T#Wen5FT7TViaw~SOHa<Q=TMl~L4WT8+#tdL-R5Nl#A?i+7KI8`c|yKs*;
z<@x{zmh#CL|8cCEHDbEY)c7%<2WSlJ29f|$81KSxgCiT7U{9mHw{Xr6vC25dI)%e`
zAuGA(DHAFnr6&1(s^y$_8T2m<Lvj<a*TDqMa<A+}O$5`fJ@Af{-`}J<kuSsWtyfN9
z;oS!3u22nYfSE_pp6O3IM%Tl?6;#s*CaTx6Eacf<UdnwG{McwhU|s1y5-&7+wc%42
z=IwVg{_yruXg=)KHj=sfjp<&^)V%aDU>0V7UcMa1rZ-Lza(?JVt!E&)y7?KhEk@oY
z`h+R0V@-FN^CibCyM5ph?M`~qgUeW%P#F)MqYiCf=}vlmYC)oMOYV1FaveI%<XQC7
z^)=6@n-wuv-42(_SD4Pd?e+EbKWF;$@-R2vohSa=f|fl2d2h^NSjaXMz@g4>(hX70
znG29SzKXCR?RFxK^%FP2JFH^`i2)*zP^l1=LjG8Gxv?N=c44p)r37$Pla<2~7~{>O
zxd#ca!&t%ecUn8X(MQtiQg)=Q%(Xwqy$ZZFh7rfcm5{cI5HStXroM^b>=yG$q<RMz
zDXEdb^$rh*8uv322PK2#Cu)KL!eXR_crtMn8lpt^ZI;Q%D=LSCabVcbeBJLjUqa&~
zh=F*RvUn4yR&)Z0%rZpzihta!xPOyY$f_2l|H-mKe!xBHhbfIYFfPr2@9Wgo=1Pk)
zH#W5;{1laHHls~Gi;#EeRt|sEE3zS#9X92J+!Q1Uk;;x%?XuiZJV*1PH;ivwpM$8h
zV!g`b<6@xRAh5z^v85%*XD9&Qrr<T;y}=vnZktvHslLN?i<p(dY9i8%SO6b@X~CG_
zHn1kx-IS;%Pm1~=`B+}0*a|oNgV_7G74#=`IqwW;+>%A!GcaZX`56%zOWvcV?lPn@
zo@fXZxGEoHPQvKq*vj8f%O-nnTW_0h+ilZb5IFOC1{8|dl-{Y~0vA%6Q`}MnhIEGD
zhxE+N%{k3gceKp|FUWS>h8|N`fgbPuHU504(KPuHqm}rb>El9yv%*g0kly>$FVffF
zN;S~GEB@e~=A4$Dj{nK^QTAiQdsVlFFC|~v4!LKvzI^6Z<St#5`epjd`B(54t!b~B
z!NX67Om1-xOxFt6G>45dkK9%yMcjql7hGvp$CFLpA^OS1)+NNngH!tX%^xpgEQm0R
zGN-h;SnNdXNKcq5^rF9;SM<&f&Ay*nESf2u`Bw4c<Y!K?L9xLPU;P$I<f)oTlcViJ
zrYZh>7sb80%0w2;^r{x^MwQ#Ir^szMUN2rgZ5nNDUcA>LM^o3qS1wmt`@H+2T%DN5
z7^aw|B(_|9HWPvpHX*ktrxBAl$TG{EQkW+gTNvL+?YV8ak9j$G(AaT_D|iLWrftg2
zjvcb?S{Ls^oZ?YO!E;i^wTs)ugXq6`a^%NLs)`yX9+fbaC6#GQ^^}&BkVnc#aYuJE
zPq@9URM;&uEHkQj9C*&^*lHr`GfdQNoW_%E2dhVF^=&*?FM?BleXO04C_^gSoAsKt
zJu*GgJwoRBKq{KWz%#?cYa7{su|&3n*`RfzCdk{u)S}SBc=NrjrLEI1*!NMcTW&nD
zJ8_1_LGU{t8#x;J7jgk|H!>QsJ$@d(b`o>;_iocs>U*-jF3v7PZi~CF{`I^2)qgH!
zcw_yxZ1+aoDqY(*B-~KlGfyV>FBWH3Q_p_?xsD?pWuCMv(5qY0^1$+_@y_v%4-b|7
zDx2DG7e+`ERukgXmAp>8&cAe-u}9D9Q3Y28Z|LtS2-*XkUHFOgd13gEG_V_pScs72
zF^iD5yRLDqjkc5fmHLr&eCg<a*?kduJ$V_=Xc)2Gwcjm+`3B?Exe<WH#5$`q`$LS0
z?OiojHM1Niko`OLzIwHeqlKfLqfvz3z*T}~LCD1U1VKULf{ZOoOV|q0E)EZ8Ijuge
zkVq@_lT(O62wx!sZY)N$KEXZJ5QhR_1{<?*;ZIp1WuJDxQY1f#VoYpgMT{>AuYkSG
z@;ylRsrb$Sa)t+on?}2=S}&jSj)(pSZLI1~`G*9uj8=|jv*YhIoc)F>>~c88T%}1G
z$r_E{ZsO5N8ZGp1L5&dFEeGZ5Ts5wZO{w6>X1sV+jVy4cD_d+rT6P2D2|E2(&Cvr@
z%WqmGw@y0Czn0U5{8zsiVy%UDB)u~A(3_d}T`J{`e9RQq@zdGSX@g74mpB&?<nq?M
z3D4foCg%@s9&Gl$Ou3l6h(qBH<?l`#>`RWOYm7dYrY($PFw~q?I=~B$Zrr1}$Gad!
zuV0V%DFe$~L>JdhG*;K1Z%r5fK#t~$3X!guIxc0-y(#Bwx@$bRb3MJ<B`)RM9ox?y
z%x*R(<Y4subZL9JfNtS2j+$BL!eVYew!F5-KdQ%nw(_)nuZ$i;IHEIgtvJB?i}ok&
zX3`RR8<D+k2rDkr7ro<J(DSB4N$aGdxr+IYxlx*9y8eUPRn?Inx&GIhdb`ZW+SuX9
z;m~1+igUdhmy*7r%~XRjYOR^7(kd2%XZz7&ZlYm>Vcv?8^3?JZZK9e=r?%dQouiHq
zpFb{_V|-uMGuQsrbl8Km8fHWK%sXqxvG}3k;L@UNS5Qx)VOmS?W4Z&aJ@fK|tKMY8
zWe2-o*@As_o9*Kqc{e$^;Ff#CG-!JIm}B`@IblWW^vy9v$Kzpcr*N#>8#le1!(-+p
z<Qw^$w)wn<*V({x_)%nH<fAt_6h8%&1od^}r&Y_&YzDsUTPa!FisHH#HxITPoU8X2
z3C3kolnGV(`YfOhMIzE8sS@TLi_!|rcne&*YF!rZK3ewWHvhq0nsn!RIu3Q|+8fKZ
z%oX+Q+Uu@~HYAGIH`V{rbnMsMK2vdherDf6Y474t*0JeS0{OW&Q+mUGLSK1rAm}T6
z{?JT%Oq%W>vtrwk3OX7)fiw!9o}UVJWO?6Sw{?5N-|h)f2?2e-zf?aZZWJAg+P=;&
zZLB)Aa#q(}J3VbZ8Z1|)t6t>!92lObJk&pB4%U6$8yk<mM!41YWj~d^Rs1s?xSAx&
z{Zj5_et9@y^y2^PCidjWkJ(T7&SFsSxlwSnx!vY8{=olA<m|aKd@}xl(!<ZxN8pwJ
z{<1FchmoAotnaVq#Vh8grQojCE^!>$k;G>-<LW|UVwlziSQu_Z7!sj8^}fMz|JtB|
z@p2A?7reAQ9P?iC0D-$#iY4?{dwck52T~zFao=AM7_b_w+&RMc=wU~4fwIA|)Jchu
zy@1ggqp8cut`3F^&OI!r*JZdLKN;_upNZ;@eN4vHVKJZu1)sULyoItd%zNlP3Jh$x
zH4Fmu4i@?$fqr0M;PT=A_X&b-KK%dQ|NRFTnE5R)DHs?r7<no24<Oj%9OTvyz4sl+
z;Gpl?H3g=R3k2ddj5mbdkp^Z}><t+JMXoQVk=$tj^qW;~cf=Z^<0x%d0!Fm=gdvxC
zmQNj7&3n0Z9v)|>Z7svM&pxLs?qk`RmWPjHf}94ptSZ!S)Kvd=P1wU3ERhGnVZ-A5
z>xF{w*-BmfU;l(1_#P*rphSU+AO<BO;9stRi9(KleJK^(M`|ib#Z<BPfPa5$LV>io
z7=qNlT|cG7!qVOceqj4|d8j0#hyHCZ&<%cw48UOi2=~MIU)O~$OPKoqp@i?au*AAQ
z5j48~Rm=$JvhU&leLHHja2)j?P>X|b{&nd<dx`&1&<x%ADDPSmyIEh;J9#wman5m}
zO3U%wO2u@R(6c*Va0HJ1WU+kc*>AU_U9a2IlhQoDhi(rqntxg}RsyOVnL&nLO$3q<
zn7-sWBmb2gk5$jEd5QWK_mjgneCsfcy@YyArz<T)8XTS%pFJSGL@h*yptnA}GaodY
z&C?BL$^Gs&Dpm7DLXWOw{%He}P^hr6Nt#jLP`I!IJ@<0vM!DzoPum|)UDkYW4I=o%
zl2bqF4l*~b2Qh2Zu6i9c+f=p9`rhxP+Xo^NIzFCtbYz*u@$2drVf|BK6F`J|S;eC5
zluZUWOo@OJae(Hmwizbxl5HJ<=`$ElwFyh~sk+yhT1cB`#Z{B{_32VD4a2T!Bg{UW
z+}%;w>k7F1jrYG+m4iC4b@8%>zI47gO0|#G(;XzPP4`?S({nd-1-hTJRnQhDuSs_Z
z;Sb~ciSOz~K?UYTu{>WOSt9?mwph_$fB)mG1R)4Sq6%Q&*@_o^Ba`}3%V37N{ciK^
ztg7&J%Hw7<t8Vf7l&UD^+Qv$4HyUI8oigP=T`vC*%D}{=6@Tenlo&j=aR))`hjR;J
zKF54L&_<|ac{m2~rc6A=is}7CwMkF60~7C8@6%>dDz~V@f5zE1UP1wFUaTL~!M-PS
z?&bNPirlTk_m?&)GXMH;XV#IXDsm5QyBK0Tg$m0W@^oA`9(G$+(eSmWO!~h-Ay(<H
z|BMOu8Xz2|z(6tD`%IX3T<9`HRE4y5>nk!A#c^L8xn2anA<Mw~vT}_(6^H+lOdS;}
z*?0bLplGoi+zdV^)A>BaGlDcYr;$tNVdlNvEQfaNEm=xmuBS1&rmenM7l*d%Pj`O3
z1X|@&_zb%LG@zOQH1-rMOBkia5EQ67<t53smVepI8Cz|D3v?`VW;SNJbD53&pMOCW
zw6_d(k@YYvN(!El9P>0Cw8y5Da}<OB64Ol)D(3G|5@{s_l4O|SAZ?xyUUbss+B1c8
z7K7!(vMR&#FP{4a;meI{en*C02eiD_lf~7xR$r}4tw#_TXaBiJY^Yz*1{E|ypP$fo
zMoQAi#5lHj-@XYWM6YpJYquZc-@~A!^xVej4{K&^J(pmsFK^!c>Yix`mFfTJFTdcg
z@+KC(4MdCi;>aa|V~IKY+8}+TCJuy4>7m+VY&{tEIc~G4eQA{bPk)3jM>ykRogh{^
zZZjW`7~e+4{?j`?$U!AV>pI}U!48~7;A`-l{;u8ynk>_*s`df4e_8JgM660HtNcCW
zd-X^5&6ZM!-}C11mEr&0l^6m^8Z_)Nv)rPZP^KwidAiLPu(MWKTF<*#+RmbKM8NIL
zP%kTfJouKkjK)~L?sc`V?jNZh^M7Q6ikc7i7Y$*Q8=;Z)8PAwTKMiw~XUWziKJDtG
zrmFC-CF5Ys?z0=xtZjw=`l-&}fMMeQ;Tj=Ek9T*!m$xE#){4i|lw<$<ASiukdoWqA
z_shwzIrqA85$_G}IN!AY3g0?xX!z<lYaCz)s=NzG+*z>9b4xdH&onDq`Wqy|5HQ{;
zG^Fhb%4pN|IkIJ@EA)8he%WX>K3xnq{+2QNuLX;6_h-N3XthGkco^R4vF1ypf=$9}
zufKs%u+bCVqGjYe(@T$qV|HV?;M7mP0*=J%DQ&(S7xjOBC&m8Hj1cIKyT7UWB)3Q$
zW>565*#Z{l9HHJ-uJ<2K^S@vJf6k0nj&RwlmqUzIP0=jB2QW!_MwD}fn!7ZM_oM%j
zaK*N8Cm>_7SsqdM_$mSLJ$*66ThQR7@aQM^Xi7KoTgY*pl(%0nPj@qD8442!CBmXL
z`oA8RpZwR_^6S^yI96DJXC1Gr|6`noqbd){Fbt<6nOL-3CQzO0!DNZB_>tj%k%&O|
zzosTXcP&G4<qDKfrfY^8XAYBuYhcCutnIWYUc_1S`AXBI89SGWqu$lEa^LfxF?}uy
zr)zXBO6Wxo=ca0WJYRGAv#ipYcg?5nXsx58WiyI#IoJF2pfuO(5Zi-T_fQ0#K>Yvi
zBan`sDn5;F+?cki8D)fTE5?3Gnb&QYsV+v(Wt7{X%CK#@pVI&MH<Yu@<7@pFNKsET
zLsQp*{)^XdloV;>-v^~GjFnA=M{`wqTf#v^(L5c@=tV4q|L~N+uYYrtg%v?H%5)ga
zaER>>HM`t=wSNMQ7{Ytnd|}=9wBH{88EOA<Cn{+uaNuM6;{Rj)|9lNpfZ~(}1OE#B
zKN-k>)W`t0SVW5`MG1Ya0!_lZ$Sf{v4QnDT$B%#U^@FDJ$F7{66$ybmNzhKS=Hza7
zGQFz%UxV;9{}gim)wa0VTROfSDTQ#D2(vT2h;6527|%bQ?ewsdZbW7t|GfC#n)0CK
z{%E#h34dmd+tOMM>iY<XsZfx*{>mAZjxx<3=+N;zEBts|#~r;>UN?65aC>Hye6AkT
z4vppW{+fBw*QdHiNIxZc@LaC2_wSN4-63VpQ7&#z_cjCtua5g&(%V0>l+92BEuXaA
zuG*JoS{z>O>$LJU`0P&<*vAXsI6_I9%eQ_-D6%raQTm(*l5&6fU?sSlVfLnRJ4unp
zcPX(~e9et1&@$JPE~N#!_0>_4$K$w{+U&^4zr?KV`DV#}Bb3Pfa!jx)%dTOee$cPq
zq9B(Jj+zY-`}>aq#FCw8)^<m;{y4i(=ZAxD9xh}2-?q3_p$RUR-%RABe(JYjo0s`}
zL@4#jA?37CQp>LG?)6gb@3#4I0Y%7`s`iKXBNWAFCUhoY#1@W!k>`QlmoTuD?tCMh
z@;4^We7=RXzGaTvaxj*N_UwAi^uT(T6Ld?zp|VXWXRsAWudJFzGcQB&5O*;{qsONB
zD5JMHmP`m#-Euvvx8nc$xHe(P(snh;Jv!b1z^NJ+c`TQ1W&6!kJ3xKP+i|a~*LdWC
zvyTs_W#E=72&HT*P{~g|={re#?B#g;hSn@Cnwy2;D*Om6Eyo-lJ(#>r#Fp8ApvthY
zZ(MPAsowzTD?l6zvU8y<w%|+;$`zI?TTV%aShp@TPV=1Q7)$<ooEm@NWCbRHklAvR
z)2Zm|eSFKqVVNF~1L4=aX>@bP(+;csubEH77{uuVP^GhuH1rXze{BKdDmL|Ci7sQ3
z@v*%ywE7>Eq%RNB=N&^?Pg}P$OiRRA@SdT^ikr#oV!Ol9bswpNtLvAM^>2ucH_VO8
zlR3^q3<r5vItX9C!qhdQGgSwW@;iTycHJBr!O&wcL;)rHyguKGLmY7AVnx^g)E@pU
z`@Q0^w(=K(-V`Mc{$)eo=SG3m6g;-tZ0}Z}DdJ5JmuD!XnR801)zEV%RibrYFvG;a
zeZf4MWGjn3KBcseR;+~@CoqnxdOZj|tmYi5kR6teCufh#QQnrd*Oy!SZ#nMHP_F@H
z4F~RKTIM>wmQ}SM+BfeIdfcvfv<TlWb3R?4EH%>Mk&$;vkN+*~0==l!G$ql|h@t>>
zf=~+k0sRc$;zMSMzDXzn8O3{R6L*)?kDpNMDyrIUbnl28<pk3Go*#C+67rWfem`Ce
zvm!f?klq=8)`9ML<@Zr#)&Xj=pWhR)Lux3YE)X1=>4>-4c2Ju28(Jw$wNFAJnGg05
z%|GIVXaa{*W5maV>A8pU1-n&(X=A0u!-Lem$pmmta7k8VZ`ul0b9p>zY@yEuQ8%sW
z@%{3ljXSF9XxFo;_!U%OIgu|O+Rx57pD7bAA@%cVH=NQBMTQpRk3%a&hRqlQz(!SX
z(($-*w(N5`s<lI(Mo4DJc)RBB&-Tg8wre=iFR&riT=MEI?*HI+;42Q_&Agc;r90FQ
z1k!NY?sv0J>!(y)SljOkACHV)NzzsbkNj^JZCGU=<7H>f;)Qp$6++VYpka0P0*Uu;
z;-1x(RE&HH*E{=*srejmh6_J_fO=Uy*8`E0poA~_M=qhipK1Vj{N0|&8e@FqZ5YGy
zRysOuJG|VJA6oP6IsC0%%Sl7znpuJ48B^UzlF;q4i}V@eO_7mE3Mg3pt<afHGRPqt
z%o_XNJ<AU-@GI_lSO=JKERs?!wDusq``xNwe(87&{O##QL1nMo5~);jEYBhkqTu3V
z{pYqG9$JjK7x>}i8VRJ_?SxC8*__4K)2<GvfhyOE-g!a;A9u0B0(a6A3MjOG!v1n)
zyj-+?`8jvuZllCRTXL-wUi$@LtUGkb1<QG)7TKchUUcYEd50%fG5D+udkA<vd{#46
zE^n9NO78yzLES=MWOATV61m?_4vX||lbIfBx>ZvVdA^$RinWl26u$AU>Q)Q)yf(}s
z2sPtnM`}8%Y{3;}CmzbrTgUJ{NLCfGPRI8x!Mq7e1FQPbEi3ZhP_Rex`tcfwYU79a
z{mo5*Bv7Q3jdrN+(N*Qq{(Q6Kk>T?ZyH&q6=%Z^NZlgP%K6&HfmolD;V<uKp9r#(F
zwaAbq9Sbm%bxAi;W{6Et<D0rpr>A4ywkR=X)OFZSq*A6T5HwVJWJ-_2LqH|mwzjwa
zuebhPoG_~8y7q?(D<Yc_G-Rj!P?C@0)Y;NU%H8wewDg(vNo||b5S>2TQ6eLjRMgKW
zk+JL%&s(nR4}t?>6&LpmiCb%SD`Wu5-gU-W;URnvQ+Y4=X&7!Hx5<!Lqe!CM@V#u8
zII}_ep#HZI(@4wZ)xKNh!Lx<Nn4P9+Z=nS3_nj6QxHwjXa98G6i1ni}EjJ66lzW)B
zqe-n3{TuAv^C>TvP=>^6Gv3LC^5Ah}1uYs{Kg_S+Kxz4#6X7lUv`|94zS@}h=DaqP
zh4ozbK2mIhv(>hqzVqJGIfwsb?t6q6Oa6h5(v^<tmt0J99Q_NYC0Vr*M-Om`uwtvp
zU@V2Oz>}0^(Byy-!^O+v8Bq+Aeg`xaG%9*-hT!7E<E60|;LAzn7oj?mpbs$&yT=Kh
z^4;zi#S<2;xAk{ZrLF;hI&F;lYCF8Zp5r1XXlOLdD+Xg$FTig@bCUaUk4BU+iaGpp
z9R4B^(iPqDc=zt?=)3w*;2xks-E@7#nKec)$NY03t76~$<ClSiw*@iAY@Ln)rc^YF
zN4RpUHB14|_!OSoiJh#c$+~L=4pR9L6l0)qvepovg8%H~MW5o#tEtj+iP1Y>rHl_Y
z)FCAhXf^7Z@npu)=l!GUFURVwU488>ycT?bKy%5sNnw}~yC1aUa**kb*U+xF_MbZ7
zGtT#BS#$-$TqYnVwE2LN<xs;4pThtzI~#0YJ4lyl$PCcUXqHs-Q^Ss`;QxhSI=_(_
z;*ivMQ2D$%%cvk;&USM!^=;~g%Zjn`p{*I!`1@x{>JGB;iuQ*?X}l$z;18QUVH3wS
z-BgnZlJD!>z;2o5X*E?YMv+H1OAeG6T+Kuu?~&syVuermW;5M?>L8p^BV8Q_O0*34
zX(ChlI9e%Y4jwbTH3Vow$#-t`Th>)#z>yl^&k~U9r^PSV`a=w1gcdnCDrVg-n$&Qk
zQUN_^N*NhNuU3QwZ>A`gpzhS7+h<Sr3eCk?Yz-Dwgmi*@>Tzw@)>?{*vAnDE<nHq(
z8`Y?NEz9wN6j*YR|5JO4-6MRlx7zrCt^!kNW|A<@?%YIcY9&qqg#+W6y++sY9$RvF
zMFdIoy3rDpA<bjj9;)^>QsG+9&$d=3MQj2*kx<4Cn>JC9G8?Vt5Q)dKdO#~B8U6;7
zbmc<u+W<b^q2}w$b59Hz#phpy7|np-r(CoRbpVB$VU{(G{$!s_q^(HYdT2jVXO5lH
zCcttt)PqnW`-wt$<NM@!=*urEvDd+is;(C~d%T!uyBUU-PgT?tC<uU-%!1IDfF7)`
zy%}=;E>NP5$ZEGrLw2X8yzzZ+<yqcK3B-14VJ-6Q(CHY&u=SCSgWnTHRkE*DDasJa
z<_p$<=1Fy5t<|;vhS3DZUnQ8l!gKzdFmWz^A1p)SYr%bhVm0AJx>p~mO{bKh##Vn$
z`d0@V?A|uroJ&jcYy+{XWYkJfHFj1ZuxYSmm?iW_B3q<<{)AaYD5R?`YgkTUT5Y0A
zq?czjgp+!*p}GW~d~3s~UEBTjMgyrBBjs79bL0ExG*MGNDzY1VV*^3&w)H53AL8)#
z*_}Oqa)6Zep~vL#lt_E$pG$Fwyl=Mnw-Y2e(wEj>VjqQrU?bh5HP;t$FXA<Uj-uaU
z!%&azbQXJ!R}tnsgffjg^T~pFAU2AD##Mk09W#_LP(}Tu0%E!wdP{}YFnq4?m+9}-
zDPcZh5j6gQyZJLjnd9&Tp51SGfq}U%X!aq4lY$Qn7>?N!T{YiANW&$CGJg1-cOk=b
z(h&_%3URZCWppBAsnB7uv9WuK<@GT+ud45F!<z@bgzp#uRq2$p9gA&RN20iJ4cS$B
z(g<3iq`BaD^G9SO!=rRHQ@RgTY|0j2tWYw?0EhGtUiRP-l-*J}*4RX{KC{YjwW>#w
zDo*o>M-8U%Ri?$2Yh{?)nqyH;qOvw6iF7ddskM8Ca_0K>hIq;6;->8|B=e=9P7Wiy
zxt^X>V=5ne7<KR>8^XK}dn!CBB%3s3uMFxfjAmIZY&GvqeWH5>(P9x#smTtIYG0)H
z!@g`Y-u5I43pYT?XTob*6;EkVcJ=4PbM`bgcIkN?JI33LbWTN6YX|I2abKxF@uBVs
z5b6_K_5u}+HJ~V45rz8rc}!_lp}j?KI9qe>j&KUG&_I(rw9AQdBKD98CTBU0mJm^2
z{qeYs3<uxcDHjDB?Dbi~uP-^sN0l^wqgmU?`u3ZViHvV2d?;Q>1lGN@p|;QDZLz54
zZR3)Ci(0%jQB$6P!J(9;fROLs3Nnx(Q*0EInd=H;9)~B++)?tIB)z<)D*bRdZ*Wh2
zL1-7w?G$k#85<Q)I^(0In(^eSxuh$PpT`}ccl(68B9!jD$Oecr!8;&Xyx@m)V}WAX
z6eBl2N7AMNee$lucwo#h`?T@TTgCln+jjiIFZzs^5qjiuyKUyqhdRINbdx%?vI7x;
zO!d6=c;L*y_h{I5Bz3{S;?fB4#^*N6iP3y=cr=9|29yv?QhenUYNQv(UO;f|z$%?}
z7q%3jG<z#3{!c*$t_lEH3h38f8*dAdd}0?#5J7@~`}0h<E8A`s^J9z?#0scRz-I$A
z@fgFMlTMqQ3e*sZ1=)Q(AW?|H&-{sG_c)$jM{!Ycp^6py*rzUZY5bnKzbW9H|A=_D
zFPUP|ODGTpj}W*6ic5&Ee>{%Hzm-n8?Rb4YGbu9BG*ov7>K!MCbW2g9sA6cP2m-K0
zhfrlwI-XA)Djq~HUY;(;vC5l}dk~*XP@mYHdnJs**_jyQO~Op~R4>rS-ziO^aCXzi
zUJU~ak7vY)>mHEflT`#5gs`^W-KuFt;N9kvZE{QJtbcdM7kuzMm1Va;?e2KFt|Bz)
zgT$>BVJGw4fC^(e12JnEjEd3W3Bqs;7bL02C%UgwIEOgQ)44ixn#=}Fgd}n`{Mp{j
zZS~#MBRd8#ZqW|mo8y}4j08DL>GfEA*O6F6#w5WPieO*XH7Lj_iAzWaNIhA{7OW;&
z)|}fOlHA`F@P??_Tg9_|H%{o>hj%!{3d32%ynwp+K(+j1K{&<Y7bC$7*qdqd!#7oQ
zDe`^JPZa1o$q?h*0z`Onr}a#5Rs*U<b#d&5$RZgAz2Hf`W?4ffM|c-<4}<w)%rc;X
z-L|jKB{vij!Vw0Ew9Kf%?LmTcKx_Blu51C=mZmpTbYMk3;-krS)9K0&--$phrdENh
zTUIebB_z5a7R7}1vFG!@C91jA4=zd9JSeE*E9dJmYmURD#tk4%V|)fAu<x6PC8%JV
zoXdyHAX~u;w(o7@hQK1uUNuRgiGF@6GP5M?Mq1XNG%P|P2OdxfqXZ{u@a|ypx=x59
z2?MQvVIvU%<1p0TQ71?T{8nu1QwuU3?8GftW!o>ID8w&8hB#@5bmh0vAcW!|f)UIq
zONuh`_p^O(7EIP7ySqDcT-o6VVTekyyyUXF&VW+3Y7w=6(k3j}^PxqWi)7ZI<)ZLS
zMb{G2PDEhy=zC7%ewd&j{Kl*svPFYd8R+;6To<pE8;z=fh=iJgJc*5d!ajr)1-5Hk
z*=2!_U)ZK1gUlto)HCAvcLCv(s@{2oDQsymi@r=mHb<Xh_%JY#z`;|IK+9sR9UEvn
z1BLUkeg+r<U{jm)q%a>^JmX_pE%GTEFm(MR-Q_4ovK~hrmWU7<ErJ*dUlMA00tMi=
zXAE=xPg|71ur#0Y*-R5ur(wIp86LOKS3|DM#KdsS;)|~QFv0NBqZV!In@aOUs3jv8
zv+ex@soaY3g@evPPz+~HVZO-p%gD5O&>CWDWs@`R?Ab?A9pbV!{y~jL4Y3u9;tIpS
zFsN?3+l&t8P$iGlalu-H_+SJHv8sXy(WrOO9T*OY6o<~i5(h2NDTm|})05B^ic?RM
zjiFd|gU{B&HljX58Azx)Jaosya9QBNdo!#y`Yk+5ai+7xb5VO;M<Vg!9W&!vWL=I}
zaBVCEJ+A%_z_2-9SQ3P^#r^{Hs0k-C&*h1K3TC{esaLH>t@o_=&;D|A=KtElR*6n@
z?_XcD+@FD3McC?~^c~t<>Xdkn9Okqxe(GIwqBKivrFBV>y`eAg%^y+sV=FZ_rX#Pf
zs*kQ8EhZn&nAiO3r6>RtgdBK)^V|a-R4Awb%gu2&Um<0vcCo*1&r4%Ds!002(M_9Z
z_Z@5*#>jbkTHg|Smy8y&()dGeXzWw)RE|f=SP#zGhSZF#@Z{)Edi)+w(id{Soy}PO
zr?8C(aM-t@^K(qU>mq7%-V{s#k0a2vyC^;xb3!;-H@y8Wha{Tfj?5tNd5*DXACjT-
zfcGy9w{4WbckRKzkQ2c=ut!q!T>3?DUV);|o)*efOK6(2UbxYgWj*d%;_UXg$8uzz
z`SCf|L13(GCu4CWFAM%5x?8rWC`j??hg>A3?W}ZRYXMb(UCrJ7!aNwu%Vld9lRR<_
z$OZ|Ub&S^n;QH6QZ4xwqvA*Mp4rhT%L>ByCpYa^3QRlEltZ&ey?VFy&mi8=7@Ejab
zP6~M6(;K3ywBiUr(6lyFzlQ92ODb++ehlS%6o;-dHB#LMp@OWEZRmD%*=`D>;Z`|X
z7!Ic!{*iyGDJpoTid|)-g{)#l-X;Kr#u?WJp&h5h35to?JDO9xssLI;c9qu2BoJbO
zBlDd3Aas11+4<(QIQb~5mytY7iwt*Dyh9w|cf$jT`nV1+%jen43<=x77MTN?$r)$7
z8D`W(-RUF^k<d1*QG|AG(!q@1B6%m@ty<v$K++WnELDynA|s?V5}9-q&{l@u=B*LR
zNV$+FyvAl~1DJFOw~CC;?_O$uuhp4wHQBvomeD5ApG-6Zgvbn`;EB{S#yj&>Z_2&$
zHv4!NyLK|p7*){?LHY{zrolhG%|(g4)t76%NLAC4r!ns()zS@7ZQjCaPvLYZR%}%V
zrie#3@@9wqK23iN{f$6>){8fU;OnFRsV(Shr=aS7%d^aA=)(@dC75wElJ_a(xAgNx
zUv-G(h0mFA>Oy1kt7D@`)^9oeR&lCrhI<8p*1*~X!(8-l3L#gz0>s{+9tqm6`lfK7
z0?|hI5TT+NP_Ux0ILQx}m@?yy^k1PD6V%wRU8G$S!smSzj^yS8-VN_o2lP?;8K)wR
z776^mF?l;gfl1XQ)K=kXATkZuSfYbTpkNbV&nF`g&vm*_VZ~dKRhX}A78i!<Z{y3$
z@04{U+hVrlvLs(L{C>DVZPb=X3z2!m_GetWK?^Rw33@7_n4HB4;}r-KIKXs&Td_Kk
zbPR3hZt(iuhDBhjMq}OkUkuWRyC!`4t00qFY04TXK?)uAZ0B8kw_3*>BI|vSjuCoW
zk2TE@Qd7z69&+Wm<x*gK>Ej^9IIX8hdO~=j=N)<>nYqjVRPk$8H<DK(m5gcYN>hcu
z#8LD)iI*$Yr-S7`UG%$DF=b)h42W?&EP|o{C&`G0@h{6ZK~98`_e<qy-hAtEQFr@@
z9<r;P@O{t@Vz(_BAqX$m%w*AHr=6n=Fur}SQnpMbu<d1`nlV}t%GQHB-a5TWj8?~w
z!nPD~#BK;f#%mX330rpyYIALn1+2o>3mclC{^_|4+M)6^bV>1bIL6GS7;)tw9Ta#f
ziW7K0GBH{CmMN#}IalYih4qhQHsf*J#Wtpb3bWmW9r_Xd%S>Trl&oitO}-6W6>kg3
zO_X|ur&9RvF<V`=5T(gI#;R1W%GFy;Ch>CxDQ$sn3_3f$Rr(PtY7lEe6%%=9(RO3U
zv)$Wgp=esk-Ctcw{CQ>t74)aGDt}nng7ugTA<wOQ3nNT;gyyj~00(jlXPmqB%z9`S
zDD0)5F*=hXainys#A!2<9?5~=fjF(bCMKMgmZvPH5&k#>N46DLQ~L`%<qbm@I=s6(
z*4&12FcD1a`H(UH8_s?G;KqbeXBXzWv2AZAHWRvBR}S=X7n~sq3SJOK;%je+H>7-{
z$wTNJN^%M6YChpj5}U<&(VL06x7%2CXO6{BiGFFJ*!~vJ8!CP-nC8L1N1s6s>h6Q^
zhj6QJK7Ep>fehk4*a~rdH=AZ1VjMQUfwp$B<7QYVm6|||>kF{a+Q1g@-n6DFrgdjg
z;KKke%ckXha5#oJ);JOZkEc^~&)N4|5b}!e?Mi4(jFwl=B$bem(z$NuWhdjoTjo`4
z3(-(5p&)JDo!DQ9XY3*l?f1LPeNv#z^^I<4Ie;1qB&0|wAwGp&Xv`{pK<^^Cgqsc&
z?-&HgWn*&d*$Qrm6OiJor$q5o@{J6wwv$JvkfGb2`f5{o@Nj)(X++>#!c0FNiEPL1
z8GVa)Qt7D%88T#N3Y6G~w^CG+6Fi}#;LL87T#{+N>^n)Y@S|6?;ssc0%tB`0Qd)6-
zX9P{3D_t&b3%+2d5%p3qZ^K{zAe3m!&*Lso`9qOP18U7oXq@)`e|rL+f6L*TvE;aM
zBsf>YCfZ5QDRF9yT0j%${;mg7Ay4&ZNA1HT0)1M$eAOmbFFQ<`=E{tGgx;_;jrbu(
zLJMZUC1`WJ=LhpIlq^_^V82L|dg*`%3q{s>sd!k?Zn4xhJ#Ff)PLc8RiBESkcR3BA
zfizAMj&-H6LwJ*&bZr>gI7oBH`~l86gh)+16T7_lL`Zv{L6Z@Rf^!A7W*=^HsZFr$
zqpDYI(7ySGxg9^)uW_=;L{?wbEY~ovhQw+_IfZkn?WzAXHAm>-A0moE@r>xUbL!3r
zFts^h9rDnPATrTyRr{)pBX5*-ON@wGLjIO4qlCW;(fsR99H467!8DlQHYuH{q{c6d
zV|bE~r>>sMi9AZLoazq|4&gHq66L&UvE*RmMVD_1>*ioC1MmSDOQ?ERIFWgw$>h+Q
zuA&_$eusj_<iu2B6wEuE#`cb?eq$BAjO_rY{pwYYMxhkdT9TbM4p_R)a9aqYnO31>
zJhH`GX^BdJ^l1oYvY+wA={Jbd<(DNy!GGMDduvY^Z<)gH0N2*6YdyGD)zD<0E7u(p
zi8?JH^KE=2VVZCklONB)FOGq>V@h)w0D4)Sq0X0XPz0A+!@=ASo0`PXrY<CNa!Eua
zV4q6OM5*z1J_Z%Embm66?9)n-O>sFSyzWZ#*6d)%#zi`APG{HC|D+v;s2a1+B+a=+
zbgO|hG9lr~Q6&51;~kGP*G<a`Xi0s4oD-xT526C75`XI*4=$-$C%H56;e5L~?T^7!
zOt6|l9J5sKy{vc-UPr@Q<V~|PT`ypAW_Aj*{B57a5_2|f9^xgPr^#4g*tr%0($rEU
zn8dxcs~4W#O(O}b#QUGwrDJ7j7$K6e_8rcVE2cn>QfnqLxe%uVkj)iTPGCs|4oJ%+
zIBGq4Y1OjWw#6Rc#;BgkF<foaS>xOn8;l|-(XiC1?JnE#fKFcAqPxs~<)wc5@RQt+
zxx0w0Mq~krZ^jP^-au1q2=#G<a~y*LV`q6PagIZ>zq3`tq<U~Y`C;6)1NAyPkl`Aa
z_Pem#RCoC{0WIHZaEMS$ER9n}y<~CMB0;5O;P_>&8Mg9+@;n>Yr|A=K!+VN&WD6tZ
zLF4MUKhzac&cbLeWQ2Vo^_yTfK-g$<z&L=jwsNacTa87{RGFb<6%22i+xAC>H1wIt
zYU)IwyHkyTu?&!4Oti;7q#(2|dmpC7A4EMaf)%2gyWSO$jzsaMCSPB3JObW2eoH|B
z=~0;R{I@t;${np_mm0ZIxO1N$>vjRE_fSH|Hr8{7K|DvdQX}epq<WaLLNmQZ&xl`k
zJg}HBZpEQ$+l-85u9irVAt))DI%b_#f<#?x%Q8`h&lsP(Yu;i+O=hD<7);EH%~)1V
zi*?}5Hkp|i>If8c>WI`J;jFgYI0*1yq%cNNLS?n%Tjw3hFj`dykb>P1UO)~BRVaPH
zo5{D#2Uia}Hq1A;k-~-Dqd8act;RH3DW0n|{&zi;K}4zo3czmA#{?qB2pwYp)!Gjl
zlr4G;El)c#U%~5yn@k~E)dA7=Iv=*MX~as-4dGP>K*%OLoF}Ds_4#rA#z^-%fn-Q3
z;}EK^6obS{e%x1focOQv>y{=Hr|azI5;`0XB<(c5V63gGlGHQnc-H3aBx9>U%-l~P
z9QC{mvMa?xwBxNSm#Ro3LZZd@!<s@in<CwnWwVTsh|8}}Tw7*8#v^Vg3s&MD12uXk
z-2OHV*-G%fLz|E)gwkP16a|0cF&Qxiklv61^cxRCuvJKmw`7qJ0S4Ft82GgsA+QW4
zy>)~@Oo&)VisPnDhIs?mxp0nubCfulA@kav+r=**i?;C5atiwkp3WccKI#-1qoW#B
zQNqz(*5!_bEe)K0(MM_}<Cgh->siFQSu^jLkZeJ?>wXkiyYGEfXm8mW9}A2$Op|UT
zYR){((^8bVl_NOA(pudS5qm8@HHWv3G5V*#rYK#)nONiyPYZ=2Xt`<1ppf1^Fm+Af
zrdMfTGKUZ_5(7A_;FW<#Kz2as=FwdakEja$*;Gt~me6_?cp``$#f*1|U_9S;LpczR
zg7w?Rf@dG?MaT}o;<(c>@DEpI|1|9KvwE#kB+c^`!|ZGy2Fg_0YG<&erpn~T1t|r5
z4K4|3<DD?!1&*pR9)=u1WDz$-LPf5Yi$gJ<{hbO;AL`Qld>Yb3DNf$JmFdB|_)#j;
zaakPcKXFxiyoc$Y-_JAI%*<fc-U_ATMV9Mt74*(`=H4HD?0DI0X_Y4$dLPEE$m)ss
z%Gb7iK;{4NW|ipE8}$bEKJZ-c`0>Ck$jum7djo@o6XSN^C6#TQXkqWvYZX}2dq^`}
zOYCnR+9SMO23hAc@8>JM47|za8AFgHQ%ovXnZMokJ)V)dk8XQ+EMOQ(KkQeIkw65Q
zOubvY6TZD~OY8<`m!#4L*CYlWF_?iiYDQ2CRtkhP(kGKtSY`M)qs}qtuz<mS9cjj{
z3?<76L{7e{4n8??lHQ_t^pjTI^Ye>rkChd3O<09?(e!kSzqLC4b4ce&A>RTDe_XBu
zD4cm1xs$<67F5LRUkeaDbcE!ZRt2O494z1!arD7<vzbSbxPl&KkL;2V^EtL*B9DX-
zDn*ibG4CNUW{qe#uu&cQb?AaM($@y1Ag-cTt)2``R8gc$VBqX2o0Uap&bO4h@c@Px
z3}bceC}Mx>U^Qc9tMLnbvJ@+e4P~}^{H#Igjkq2aZzpX&Yf1FrQz68eQ~3^4rtYk9
ze6~aERi<P4q>p)W=xAT%h4gIK68CCgQHta#JX+|Vqq^NYVUZDNgk>C4C}WtjNqVdq
zm|D64X#}#}n_e+`pq955P2#BU>_N5@ug0HE7Jv%C`+gU86yt~iG}4`^2j=cx063D~
z+C1YNW$xj<5bKCp$m3+UA#gVRLSG0&fnBYZ7kgZM!23ZxH+WL|-jXBO4D3l;B~OF#
zv|Gw7<hFU|n3IDvaO>OZftgIBL&0;ElQsv=dInAo@$IDSXkD$KK+GEQ1T^D}@u|o-
z;Y_r*JwfE1@#hfRcd8I78m#Zb)!+mRL((0gI>I?;E_d7c&%}z&c!JwFeK0m_5%jB}
z)?2~7c><*IPItMqygJuOzk6_VKo^m1C6mLiD0soYcaRkXcd~z{(Wb}_(QeQHvZl6r
z%O8!UR~ntec=5MHY_<+O&JDX2BOc`WL+HckW9mu(^82L5mM2u#pYwSUHm`Mtp24K>
zwj<><X?bR3L4qe3J2^tT+y=;XO&8<`;G=NYp5(i!Tkvqr#p%rLf98W7xKNOy&0Q&~
zkqMMTC#SAcfh_7;xTt*|DGD7cOv%LT?psM4TMQW+v7kO~aW*1O$xP~P1M=tuA3OI{
z^j^Cb))Z=vS|(Ts?0YGD8|p3q>(ZP`G13L*hoa3m{l|@Cb#!6~%8PhSj8}D*)mqOn
z*b98cBggRKM7YJ(jd^P4jR{J+QmjrTJx3H6hu#h)Fwj&>Yz%)jrzNYj>`};4-hB|F
z;W}|3L(Uz)-a&ys(mpVQH^Z=mC|QzG)b;V8kKF>^O-i(G@_gk>C1st*x{7+^9dBG-
zudFkud85)f-?du>5V*J)OsmHg#m5#(Q|l|Q_%<r`NAic5j%6MJ54?Fd6(g&1O<jE=
z9sY26d$f+<rS|$gh<2wJuLpl;a5GdK9{$rOG89I_RuZLkl?S!$lLaRdcH4<h6ws5^
zncna^d|-MWx<6W$$|Cv4b*<egnynPTxr0_ZLnqL*B#$Mt7SKaz9Kq0bBnqb&Uw;za
z62l1abG|C1gx8aThVYYYn}*R3hcr<plY*!1)Yw*#y5Z&xCJp2BhsAZv+ajTU#i;?B
zA>UaSNR%;u(u_-=76`K>RDdAddWXk1$G2}^K9jv`{!>)K?0i7{Taqdsz@qlRWKZJo
zw7maC!cq@w(GbVG9pPk-xSBYxF<<%|R}GeNR{$VhE^o$UF;!s*B(U5dl6*R^9jcz(
zPE+-)5^j0j+fMJlt{as&clxL7?<7#^0O5;;Re^XTv7|s#`GPW1H2iZwCb2=5uXtiU
zUZG22fo1iWSR=kkf+;7dejax)VYsBM(q2~`fUOw0%2f&A0Fu6It1#F+h#+oqWQI?@
z>g4eA`?Xd_mlY_>9MVNX@4P)4HM5yHgOc3E<)1W0(+MNW1JX!4m+ORCA`Ys>p2JJS
zB(O0vHG{7!@bs?J)@@qFS8DH(7FOvnkQKxUkL88stc9%4SMCs0A;l1vis9Rkj>;L|
zE-BUcvg{Bb&hT*VdBtq%ke=yk16WPX8HPFO$yr+(m$fhqy+P)YexW6S(A4HpoSZDK
zO%i5_)E8XeJ;1&Dw%g@Cr&dlc7Pp)6e#9A92zI}S4F=z4Sl*+2-z=arAc^~NxQ_iz
z<~-8Z8-71&{^((pjT?yJyArw97}V3^M8Cn9Y+%gQui`}EK(;V^MVjUPZ+4f)hun$-
zwGZyCL7VM>UVg=z-idChXw>Tjwqb%psn*u02aDX#S31SRLg`GZ(VoJok56CpvFe&u
z#F|yTqf#!&nzCxbju`rTQ*2@^m4KPX$D?I?poomFfFm853<B9)%J9fU=Hc@%<DUmV
zEW4*1Ld3lR>6{nY0<zB~<;iFDWw!YvA}TR!7mHfaiVvyoU*(zbhIDP0YtAkWf3Sk5
zx~c3ubA>%kK0FAUS>a6@*e`B9Ze-5n>fCt|{15fEQqHFfBNU((Aet?hQZus}Bqig>
zFz!pg&coPd`@PQH?JY2<VaWqvVnd0r#8;1)3_b?HI)U3rworl%EPPE(cK=xP+~o!W
z;a7y-l5KYG#$YgI+_RK+?u2R{+-5HZWZ5z!NC%5!ur%W008?SElen*%PH31+`1bRI
zgKR*Sn2V0qjU{McRCo#YB2^(y#iywg1d|Nol-P8oiLec~q|{Fqp~}`fekld5v=|tZ
z>=Js9eb)e-3+Sp3UnB%Akj$W2Rm)<2bSX@nCkj{Q2h&?-5Od-h%7I0ux~^Ko77xx=
zgy(*7mE94i-J_TTLf2HEkcRf5l)l=<JZ}&Wsl$-ANxnY^4Z>P-9@#f^|5lVcnml>Q
zA3;f0!>&!?HypzgQXNQ|n$v1NKuS+EYy~@8{P(*M{QlHEa1;yYy)6RCx5QQ;ee9Kk
z6UksdNHzQJ9uU4M{8sx*`i{D|XIE!H0udO?eFyEjhv$-(4FAhb0JRVXZ0fdkvy;`?
zbv+^k6zrVUTVAne435TRPH~7?@{}Nti8TPwT`ckqnUe9yFfNAGPpFs(i381IvTD=P
zA%fS5J5~TL?8PAsArau#v<GKBHr9=GcY?`v#$5SbD01p~ymN1o`wQ?DrAmp~;+KPt
zxCW7_#dPQmnBj<6XHC7aJm)PI5dI3i1nuOzr;PeHE7qvgB8{Gz|1e)#obQnZp<{zc
z;TYNn#&s6LFtJHJ7=+|R<nWyO7$^rm#6E;e=x@h>7Hnp;2)&VeyKYt(G3qaB4*)jj
zCkrpfKwiaTKUL&hUMe~k495sT&L1H4^C#^e5nFoEx2}a4Og3+~jcV%WHZ#%cYEaqK
zisIDiGf*w7`*AVi-ZNXwe>eYWtpyPJDvKWcV0wZV?AekADp)4cM6kXc>CHY;zpA_W
zw#j=3=9~x9I#5Mh4!X@5%oWC(CHD!LsKo>>At1&z{Q4lC*-2%tz;=R$6x$joU`-cS
z|BNzc;rTK_35G}vGS0!IsKDQQjIj431P@t%%OS+KzsM98Wb-j2i^4nwxbL%KFV>5!
zN+L*ldm}V7c%UkvT8D`1$}yUQ9|&T!dWGcGCPvC493kCeIQEKiMV^t9$fj0<x|>s<
z_#kH~sCtvbx4@bMk>s-<oW*}sS_6H(Tnt)RZsPKvwtOW7h6#pn6nMKYjg5Y4Ni~oE
zoVhOU!blZ@m$Y8JE72wcAouFCk9F}V4rpQwHl&OVDST?z_y&x&;9A$cNOcz`>@JSw
z#-bTQ-4Uv?>ruNXxCQTsgzRW|&h?EKRCHw^^M(k>W<?;Ya)(@Xhqmc|Z>%?%c3HH;
z&`i117<!Kv=j3OMQ2x$nSRRkb44!9;x391SJeP|{`pkC6{`Tvb&EJz4&@zY$ZUNzo
zsa02mNgXl9L><yHAvuXXj}{mm?9Sbob9i4dU6j&H|4BDPS!!Vb4=E7jKS9i{J>aYr
zhYSvpq|cobpJd>yUf+I)5Ko975QlG}4N#ktaCEEqIR7i6s{2?U4nsn{ofUdy)zOx|
z#)LHldi0eKpe4xYj*-GnHNT$i^|dh^&8gV%(o*7i+~v^%M#Oc0f>0K=4##m77qL4w
z79%9bd_>CW0Xp|Ju%cvU>0Ojb2?AvrHB&G<cFUOrO{UM~Ccfbz{*`HdYNEMV-jD?A
zZEN+`utvgoeAug(Pkgl?q(m59p0)fya!lZ;T*4@zyI;xE)+#bp=OLr@sh)2l4~Ip_
zjyUSMkI_7*1Qr;LqQDFp=I}Cmil9ee^TWPndaIAm)h*NX#++fLs^Bh3#E)0{=Q3Ae
zy{S(Svu93;niHm>BFeaU;Mj%Z=7N%jwrBpbv64bM@??J=s<~(m&*P}!hB&bTYcHn*
zO63;6CzAO8!`@kVMZI+opOPFp99kL#1Vp+)KmiF+N$FI&JEU7`q(h{X?if0y1(a?O
z>5wkpGy2rM_xC5fYq@ZiG2b}*>^ggYHcp5rQ_JFGadLsnCSE;bGSN&Yvo+KfdrnKa
zY~_5uIWx6;quW@h+1E?sY{Z8&sLx^M)N3$%GHcX!zN&Ofmn-F))Fh2$Z{-$Hi6x;_
zagt77Ts2iieIh!o!#iZhMRhp(d5=2>x6Ki=L!|*?AB8n>Kf@R$+QZ)-E!qm+??}Gz
zl&lcd8tb&jTGvD6PKNCCBMlvEHJ#y#g|HlQ$Ma|p>e@z^m*o>~qEAv`&zZyLN*Dgr
zu3rV}ao*t`sYvD$juy8rV(Yn1(lQce6s9INYno}8D9JUBd6}5})gUX5sWXP}nhHOh
zTPBHy+X44pX713##rI*aCA!JcBcM#sDyI*2McJV4+#hNFWtv6%#yw2u?N(HLNCs;8
z=*$=UT=5wtk>TzUBRz9gYUphXM=QzU*7Pa+)bw0$>7@Y$RNjsl(ZfgC)Y5GRwtny_
zt@db)O;%QZ9V=aER^?GRQ?wb?vOoeh`nJ0?B?<clq+zAaXWsinjKj(uci{j&X=glf
zpdb?VWqMjc)J^c>f#PT|*>2caqBH(NwrE8WNe6ZC0isjpYc8vF)#Yj-8`kRc*?c=<
zcg2=YN3}<)aMhy<5-!2T#4h_*B`E0i5)V2wB9xq}47=4FpgvzD+}A{Yrw+zM^xazW
zm;BVjFAv+8Z4RnGaJ_^TnAgraD5`Q#$wY_~2Sq(_;0)$i*5HUr5V_a_MRD{A#~@WL
zz&g^&MHZ=Bvx@fus#<X&I7&bn-^;~|PCyHLz(XkQVIZ5F)gqhByil=u`#i>#)SLep
zJLXH$t)+cPpWx{jt$C1)M;&@la&8TWI<v}315tiG!$Q1yd3Km+4&=ivvIJ88jXd1P
zd+BoIp<^;P-RM|2F&=FWpG0=+7`$XPw+Rz#8PSI$HQmi;Vb9IoXsq909h5(&*zGBk
z_BAj~wAV>f7VMPvQ9BJ|_iUfkj5?ILe5|9Sr+&L&>_dMf4;kLVfi4{7K(2i4)4`tI
zn)v2?<6iTY;FB=iTpsQ+Ml4&JK3r+7T&ha1_)FHG2`@#fii>n6p$aQC?*w(OLT_)d
z=y?sZsek3JY2_^m7*8x}cxA8DLd2!x(rG*X#V8|f<MLSRcQebP1`4OQ_$SFA^pxl>
z5@O+w8VU>aZirGnKiP@-=ai+W(j~EU=x|OBU)%MA{t;=6PGLu7MgxL=UbUhhN5+Pd
zkA<y}T3dsZ6WShWtCCIn9ZbJ2#g*VeGHs0$piFGm&y>wd#^fH6!)opKZ=>l$MKw7&
z4a3Drk2PB+W5=m`Ml6B;15~eT*b&#}NOJm=uVbF}R93s-S`UNjJx#kdd1P=p`LbQf
z0;U#xX)&{06N@Ymtev2Y<^ts&0?k$X$!!(5BbNH%xV`So)-bk%D(6lf#U*V2{F^q0
z`Ivx9RkP7Ihu&?sZU-jYO#4K>7(nR>*Ym<Zcn{H>QKC>&ElQ6njavBW1;3?Mr3P`g
zfwkVtpq9^8WDlrXs=+{8dS6xHlz5i9FQ&(_%Em%D=do2x$7R$J@-pH*>7KpRMQ<jD
zlzVV7*13vMD^h?_&u6)s2~cu7V%+Yj87sbwjS5?{r}gce7?oTI@JFTdaXTr4wEL`k
zw#%P)s>);-LUU;&SUBp2I$?4L$rbXM)@xr%mJ%I;rMdzi#9@p#N(G?bD&;?cOR{Dj
zwP3E{#9)om_QX5!^1QL%2_#m=^gS9p;0v^+tOG#>l@V9k8AWpluL5WOQ?=S<^(1LV
z;RD(X+XJGlwi`6L!;mXpsW$U5Q+_}3qY;^o&pQ0yDqatS8fNmScCGa`vt&;7gjV?P
zdrfZ?#aEnBuRPYh;enY}TPLV2TT5%T5y-vl@eGb{bK~paSg8etY(0mA-7i)zWpBS+
z4Tg{#wFOs%tS6D1u{uH_D-}ImRa0KWaW>H=%n;rrO@qkOf~^E!pY?~l(-fP`dP$M8
zE4dc?CYJOEfefAx)k2ozn>nodyl0u!&^eo`#Hh%JN9DUel-qagmaZncygg|RZ0-=8
za<y(U(&Wv~==M(ZF4g!`JFtxQarcXKNc&<CGIF~lN53GUe8lS{S;Ku&d`yLx>O+3v
zZNY3PPSH*fNcX^Go9zG@?^Yy=g}lOjoi@KOB@B9yUg+v1g==$%8xf_$4}*<XayJ69
zzGU2+jjKv!qdQb#n9mfDx<aMww=MnJ`yiQvQ0+-pM@a9A6w7g>smFMXHkSPd@%#H#
zT2Igm+;nnRG(QQF;JmA0eSuf70u8dDErB2QalcmfBw)W`rDdHScb!Js6y8BcRY4!9
z;=PJq#|r0Gpk>kB1_dj!7`3a^lzO8tzUUG0chjJEvg+>3kUx}~>c}h$@TW6N$z9c+
zK93SyjtY2(x8~<*SvnBVQ}U!Gjv4*xKr*#v(9Dj!sskf{Sm6`flYu0M6cS34?^p-Y
zgdYZsp1n^SWsY}g9wuHkQGnPplo}WX_$~)j1Y(f(;j}xbX2kPX5nx19)$aF7B$R<F
z9sWRhEy>4HLJXpy&PMQ;zu>=F0L$k{V^wzEMRl>R?(=dRq8*Yl$sZugYSPr;^k~nx
zEA_BvKzt$#ZX}ppn(MU<S`><51qbQY(zv`QrBj9>)Y23M`F6|atdV8L`q7~BjlDKQ
zJM-~{yHAF+GR2D1(_KNrn8UydC0bkehl1mz>=Oyk`OD`(k3@huk^+Zwf}qh^&hGtt
zS&;7bQyuH-=Z2Y7kvO1>$xKr@7};O+VuZbJRcPL@DhK-hWWo2>i+B$EI_mQbES5eF
zcN7MQPFbrYF3aW}(^%22ACIQ8te<$^Jvs<xhlB;nG?{aB#K{>eysUs{UETTqq#U&4
z)IxEzF39&r_C}hMTe)l$!C}YB70W!7lUtCR{FW!oF!jLSyvO+Aa`-yM#(VsnIR=Rd
zaQgDX6wesnI*lBIXO+$^^Gk_QLH3-Hr7+bZrV5O_yL}J(>_E<NG^z#{mB1P@@2=Eq
z*qg&)$dP+*^u<`SYDp;9#v@P0Lr8m0@j6W)Lr*=PF{a`{C~wQJ?*c8K`BXzF`_JSX
ztdnVL2}6z?`i<IR7iu6GTZ_`#On=-|Q=)y`S6@5vGvaW#;M@mW`pFa!U!OZ^R~st6
z3SJ?}XKgQcE8$zF%M)$P%>nmjw=l5lCw>Odeqd=<SHAk55|+usHoGqJtFv*N#Ye7w
ze;tD$%vC`7>dpr{YWpM9#7_K0>|r_@A2Wsn7IY>(I!@iK33?VEJ3Y){d?R9sG8~YD
zK0#0IbOqfX&q&(GDG3YEktMjFz@^_=)vv+AXpe#<S_bYM@ET#YbS-04hGhyVT}f8i
z?SR6i%NG+}r_WZJG{WP+0+gz{!k6FbdLMrcqcwq%z%`PHOVJp8%y<q^S322andUS$
zW%UjoT@AR=`RHM(zzO*d6cm>+`abXZoHll{dvZ?Sl~`1UXR}t_s#4LLQii{QolPyS
zeu!7bKAvN5S#q65*}F-13o2p!_Tj57fv$x)ygl|z%6>Tt4{Ik$qd5+=((oVj6Um5c
zH*R9Ab1o+(`jn;x9!-jgve;}@>@JCvA~EW1aCLs(%^z?_^v}SVear+86v?o6?rofg
zD=d><zzuK~JWY8L2yrH81E}0c@AbOLEJ*9w<6CpM%h9-@K3+aG!E#riRPpWLo40ri
zDvWBc2xx|@1zS_Nub;+sdU7+(LrHypVypJj+NP<%9TEsD84vm)+JUIIODE0swxUSG
z78aZA#D=p~=obpveS{RVXWfDXSKn#aVNVe2_hFd6y$QQqS|sgwvDQXurJcZ8a@zOe
z5yK9;Y2-d!I^~S~veUDNah~8Ex$Uy_4JFsv)8>j!_NFj_AAw%W9?&@S!<dVuCKsaR
zVAX{t=_I!RV>+wSab}<Q^atTy%cL`{!FF(I>fK|x>=%|yKC1!xLr*b7*4tpBrdCU&
zr~8ykykz?sqF6gpo?V{zNj8ypIIV4SmvNucl_c03?=<B|i`bmYQg&W~jyKT*^p6Ab
zUsTRaIfGn=DEW!@_+s6L^E*9580QP1?X@S{p$yTF;B9w!IhsvGW$J4L+eRrL@s{Qf
z=5dz2Zt5|x=;`%59OIg?{L=mOcA=wvie*Ictmn!jsAnwmp%B^hcSQM%D$G6aG!F-I
zjXn|NK$<_M($^&wcoOXrC2V2;J<;bIWq=HMHRTW5ZHi6e4Z4?+_=c%`yE)2Z(KFug
zr?A3C*t^JBa?cks7M}1W{)6t=eI>WI;<f`^LT=L*wgZq;bzj$6`<}SV{W#k9IEV&U
z1ME9~TgG1l+k{YC>twPSveok#3{?|ZoU(lILjt7;PRHs!K6%ZrEqRi>2Q`+JBXpg^
zoNHp#A^{<Fjz@ZYSkzUxn~GR0&B4kOrR&ywcUV&hu31r-2wGJ)%ohTllD`u|m)elJ
zkE*(dLY*LA(q$JKr=gflb8dR?7E_|sk1&=D&A3DBgLbz*zpulOf*UrPu($I+!bgYB
zrVg??$z{+#uo|)6fr^jrkeqT0o|aq~`VDwP`Yb;UKou!#cN?~X;f8h@{9bFa^TWC<
zK4uSAW4b~#5{M5W8LL9UqzNXC-NU0bQ{CKUC#gTsP5mz6`AXdM>*UBB(NK~{pTr`n
zvbSUP(C4b`S=X<kEBugC&J0Z+vrChiU13v}2z15n!$TBG?B96uEM`-A*yfA~>{}fo
zDYK%%56NY4(Cr`$J||)V;;THtqI=FHB)p2Bnnr(RWnB`>zoGN<>LZRcaG_Pe4K5eW
zj?S`3A?on%oEtV{nP>k<iaNR|#;dKQXNQ08*9O{X7`1{iJjI=9IwT@FqfZp-d$|4L
zaB-$1MYEMOGcnhzC-g)#>~qqM+L?-V4~syU(r92(Zf>}}uV+N31kodt@XVc!KTRM{
z_yWH#=PuvY*&FcQ=5&UAU}J_ya;zlb)m63I_of$PBPckKF;omZh`W`2t;X~-2+}+|
zeHD(o{d>4kQE<?jm=z~J*6^NlyxVZJ!E8<x8*^lJuVVzpUQ@opcg`v<pVE(H@!Fy@
zo>OS3oYq#N8{M9f@1YdrT+_U)=Fp>P@rxgwTsyDrRjwBC;cfUiwWnp+-Ik3DdXHJ;
zH20k2EC=rFL^AAnY>Mk9{ZOo3Nl2Y!eMEH~cgX$yXMgUp1?jwT)qa1>{eDJtiBY|9
z{{{jBN?WMFuj5i*&JP$ln?Un@>wd}U8=QDxprkUx_y-EWnko5wf?!c$JCSrjyp{|$
z>eeB*MSJs0pg9272JXX6rmJwB)tu;QcVVWxZI8@(jQQi}x--OE!|2pXyi$$g-cDlZ
z-iMB6$On0Pl&_4E?#JOKhI_YgC|x~I_zWKVhiS~W#|i2Y?j3(MimHF9QLG_x2e)KX
zuMN9j8BPz?ZPNPaH{($FMue~Ax~}jN*)t+nB{x`YTX>TxHOfOJVKHSdb)Xfu)<Sp5
zQhn6<-7v?2p*}`*{t4=3865T0=P>a4kzN0H3f*7!bX?4OYLNb?n&E9kWVTsY!(Cr#
zC%h!w@Ka-#Z-$7F4oBGAZ`FEFvu3=DHor*LE5uHXuIPa%3TX0oR@?*47b?90ZW4I9
z-<v9kVV;ZcV{{rc(%Oj_m8lI2N$r;#p9XFTb_KYkiV<pFw9CbJ4ztH4OS$DJ_41aI
z1YOfwl@i-hVR}Zjrg@$l`;sinkPi?;J!M`j_KuRm1l$-NX}-V^*d+HRGjx|x|I!>`
zDt%ku%ifLT`UN~+<>I_Ep6OA70L<S`>=*{lhqjz(HxgIN(28U9@!AhHVYFq8Zv?+A
zv}YW?W8vUD2!C1-!7=4mEr1sce^Z$Ig>pb!Oaw39)2j5&WO)LavX(T3W}b9C(-7v8
z`d%o$Wpt2uybl#K2WfU>?Cn4mVJ!-2Z6UTi%?yX8(`C$5I?%@2Z5S3Vpc{CB#(S6m
zr&n1@F{AGJJU@o+&ftO>0ngEO4~V+fJ=V}1Vcw^LeTy~cU}BhC<y|25olY@2#_i_z
zV^e%uATZ7~ZRx|nQKy|MdqYM%jdK!cSotE*?q?shp&G`deDF!IIe3OA<d15_Rs%9Q
zFrB}^h&@&m>!NEyYTi_&u#886nn{tWh51nR$?de0SkV>NcDf<0jX(^=SC*G-G8{`r
zDr9{MDLe<}k15s?oMO2G2OVnQH4*!80n0}i{fidA!YG~J7_V8b|J;nrP0r2z5LF>5
zT3GAu>|f5lBi7Crs9*G?!hDk5oBs0aopI+3KD))%eXU=A@;iNus?FANuHaC~i4587
zuD$F@KnFL&21a9>ER>4?=GFP;buDdDNOGa*?Vo2C3Jbg)1C>|eia}c3aO9+~N}eGk
zp6E+%L5<u$dmv2x*sth(-crFHmj(Ktu3=8B&<qn{mF|ZyMXr5_Pd0;bT==_dN*iaO
z4fGPz!fZS3-*8S`=80XMd#lD7(U93|?-R7E_pmoD>4{43JM~4Bt}#_Mh~k%G_&#&o
zr#Hx>7)BqB%D=T1wZ%?2IzV*K(bl|%QY8oO)-}AC{9)nHFWGsXxJ99}JwR+}@wuv{
zX0CXqSDQ$HIqZ<5j$3{3cvY%?n=@*a(vq?0X@<@#kBSn8=y80rtpggi+@iP2aGzfF
z-Q2X!d3>$m&ll0=_O@yHxe3KinyZA0>6#uLM4a%D(Wg&hV1ic0{yjm}ya#)^%`rdc
z`*aI;<O_#;8>0Cm59K_i^r{knT`4zZU9U7NEI&&A?sLPsp6kXr<rv1Y7UWWw1Vp}P
zQ>}0<zD})%(`Id40*3zdhBd@(V9lMF&{gz(QImOC+`{dAO$Y63xfbgk3B^~}-L1tX
zUvB?#k8=rG#iP{nzR0};l?UT6Kiab3sEed0CUO~L)RRZi$&zNA$+-^<(FfQ?47;Pq
zyR@-r@T?q!COu!?8Mw2%BtU3V6HQ+Jv<qKUCk>^2o1t>Kn!xOC^4F4wbpp2hwZ;2a
zcSPJmxX1RQAZ;RTW>1p8n$+Cqv}3Y;GI>;4Os?7pGe2v3)DB^x;dK))Cug9+T2<GE
z6A|{^u6&OfnIvV74!df-<T^o8TF~D#KTKRPJm}53svy5h$+|Mv26@FpcW*%Yrv`4J
zi(77InAcM2vuv8`p7{5?qoKF4Do@rhH#%|rTTU=HawDs2uUjk8OWFHa=5-RkT`GXr
zPxiHlD>pC51vloy(;!jyTD=b)c~N(3Uviz+>*_y#$!L=Zdwt=+@6avN%O&t?%@P$u
zVikW!L|)C4;rS1SeY1mp_-K-Y#EN~uDgBPb#`el3CV#-YKaHo)<nH)#s+cZq+(WcK
z6Sb1k<Y1%vkXXPP@1zyAKE^8NW;JdNYV;`#clYvT+vAD0(}w1^8jKs>`CNfKyU?!|
zXivHxNDbj<u6|H8$t!%3IA&@kbW4sj+K*pG@r5q*>_SWudw0LCYUQI|G;#!!@G3?)
z{dt!qdD@LXQ8JlC#IBy#L09|4s`)&Ya8SfuuK-GMMM6z<S-9ZZ8@KbRZETrwYkE~N
z6xI^fR49ekbSF`z0_oajmjd0BFf>iQAhi@z1CE&f`$r=gmrVp6f__^YEYYHez3=-=
zj9`(Znl$}tr<ij0VZP-@kN<E{RLPOutDbc~aSfBqCusFV%faS-vc2VJ6iFCF&IKn;
zYO1K7VI6R5%^HE_B)v=FCLuKQAfH_-G+b8sg^BtgA)y8h6$kYp1REMY$=-n`qbO8)
z<DxIt;_$E-3!SV*EyvsHxUb;~um|Y2*Wb{g9HjEY_!I}kwl7=NT2@JeA(77O8J;%W
zSq3B>KU)J21J@(4t|Gd=lzduEEE7_Od9}3DS}Va3jW=cQP$~=7pqqEyn|BHQjnFJw
z_R9__&(H&K`mEV58%S`}Tf!un(DwCAV<<)OrdF7()AD@{7I@&agpm_&H3lYJxdWrS
zV=M;a`6}OO#4CL@*-k@3HbDy$!0bq}1A<bO^n0#{S>jmSP$S2OEkojN6=n6ba@hRg
zCXV8wnwkoor#{8*lUDJ_&{Qm=s)Cxuwq+F#tl_FMHP|&4#t?0nB`E*x((G9LQo4?h
z3O_Sip0(=ZFPUYAwq3E28P?1;S87CZ1fb(O+#H3J)pIY<f==I|y#V!CS0crpq~gG1
z0IQ>FKVmVg17L>sb7P~9r}QL>hvVp5qVdSy-?rIWp36*1j%IEXAKm~U1kOB1Cx(E#
z-W1Yv)azGoUqBNHS}{LIiti-!88V?ctMFltpc^OUMJcNoK&XYs&&RIc_^0`hsE|Z*
zKp&5yCl}RmyBGxMx5&eOUP=ThBF;=kxW4ug^Rv@jC?j&MDY1SqJU-)+yHB1xMq)}b
zWr6r4l^c~pV9o}Qr1l@W?KPdYdy4Ar`z&=5mJ|oYEei1B$|df%{`!S4n-Q<3x$?4)
zwi~^~oQ#umu+Ay;_?X!ze#F28AbWV*9n$fyEalvo=A6c!7(V{O(?hj#^*yys@J&d_
z0)(f|KObY7Os`yH*U)9&Hf0-81zr~8zhJ<-J+UmH=>%H(m3uMzaL)~;sf7GPy^4bf
zxJb0jjPXVHTe)@mdAUGOTiNQMZ6_4g@^yI%TOF+e(_ln=EkV5<OQI^1quw}-R-x!2
zns7MMd^*4<>1I<cYd!M(A+Ut2mSww$#_N0J&K*6r;2=B(fG@P7w~SS~iUce7Zg;!x
zp7sb}FbH^3TmyE61E6T=xAF3C>3RP__NY=IuxcwGK0VAD!^%RMx#1{kk$Y^q-{pz^
zh-0daphNpLER1{KXiBvhF0Bk2^vawHUl?fan)GMAI*yFSyN!<rF<K?>#zF6gxLEIz
zM$!Oqh0Y{N(we556ZgL99tD3`ac8fQ<*HsrjQ=)F#|F!0jWrR<uX)quIxO)E{*p$I
z2X6OJpr^FEzd>7fd<efY;)yy>4FSgNDi6F|1IKE+;ERj6Ty2EpV>E7sN?II6-UFJ>
z*KhfsRbDbJjmbWzZxy%NDz4SrSIwn5?76SHHGFW^_&wz{6}sbe|4Y|7KIpbq^b{;M
zh}B3p-}ZcjRzI3B_FA9&qj9KhjZdwD*8ATi$Sf<Aulg#gE+kF?V1Kcp?ccNxzC)fQ
zQsf}CkzzBD(fwwd(^`qqYnEzFCd#1MJyxpU5{LdWS5(3~O5$Xqyfc?fy5W&}C6?T+
zZ-b7bL)WJZdsk{HyI(c*oo_6@aI1etuZJ3v=~(gf+|L2<2NpNR=kci63`@%nLquG@
zR0WMU7h52IEz>WBK|Q`{@Al;RN>%m?WO@C???ZAiG2#~%p_G@O%hE!}!WijC)X<8a
zn4;=#-r6a)7S15W*!&4_HPnz<1oK5TxUyu-m0lGO8FJUk?bl)eez3#^I{eoFW;^pK
zbdxO~y1qh;q3DjM7=5k-=q|Inje9f~Gf!a@i9|Iee_{{Nq;k9EM34{yjLQYHB;VQt
zB++L{mZtIJ;yyCEXet4SWQ`ST!+twXBS?IWqysKOG_j2xmRX)cWjS)<%^HD$TqRBa
zqo!_ljGSg0A{36X57egkcOx^bMh7x(6A}J+gtDD1y#O+^=;r)p-CKlhlmHy0&_s3c
zW=Rhc@h=+7G{IYWMeDk|&j<_dB9jc&&U>&!B9jcGVsuS%)J1hCxntI3+|bIl-%PRb
zcYo!0{=D)%8|o){2Ar%dKD{Sem<a_a$oz!n$;?KchNM0qJNDxa3I|U6458r$D+^w7
zm21GIcG^P=`?25EN{F6&lfD;^2@qh%>>%TG@}pBg$?38m98g%g>5n5u<09jKVLGKc
zoFT@ZaqtR>;}41PSvry;y;k^VD+xvZSI(X8Ul~e}Qp1rbzeNL1#)q(6sWjGq#dH=_
zKyyD()?>?4=SD`q+0OKzz~1^e_9ixnI8+rf*-G-x_~ffWyzakRCM?l3^FE`0vEqm}
zOwCQv)>ESpdHMV$3WIz!B=zew&x%zr`=9mtpYLS_-%GX8OT%du*u&Fgxin4&;T~vC
zDN8PbaLUR2K3yN-M^;Ad@Fb8bx9)yYKa-cmE}V)K&z^?aX(_`&xjENZ3-L)$XY*)A
zSet~$fB|;jf5U<BhXo4N*Z>yaN>2=aj2W@W&1gm6-A{qY<O`Lxw`lIUNK0L4|868l
zCbW2hB!eSo{Xss4CWtAyxO7~AUR4BTyYd8(l>#x!CMZ11AOCUdDIyRo(M0|59g;_j
z01Ju7V(``pT_0edbOgWgL>YZ%ddFeq6EV@B<~THE#P#mz_<{pbA3#46hZGD0fW45W
z;(d!HCjmIl{@ro9C{C>n^&L!*#uhNdqEs=6xpQKE4hSXG#9Rq#|9Y#}#3O|=$C>is
zOC-ujD}CU=>GqAJA!^A3n3h>`7lOT0qEV`6b925v0nwy*D~>3dm83-D&WnWd&~uC}
z1_~r-D?4}sWjgml$9j!S*RG9UdSsY>#0NZ~+(OJqtz+8HCjxV;pwKTwT4LcQpB3%T
zIf^zBz_P)dNNX^k`M1*kI_rs(nF6;;?fWlr@53xRA`0kF;xV2Dr6J#g@u)W^F<t`@
z=|>L9IVKcSA_G)2kvGXv`v)cnoX{-A>!KwDU*NvX*xDb5J6%drF+@0^A%pGrO9Jmn
z)fG+EM$Keu0h<e^$kp_~>Y>Mo6h=MZzZUH-f5PNrLFR7(LT)?x_8g#B@Dzi2DyZyQ
zkqx!%9P_jbi4a_X@AM;G!_$m|iHx~`1)D{N29%B1DF7!1MY&->po@YI6*(sDfmlm!
zTVRLwBiJR$^180ibdRVI|C_1x?GK<2@hRW>9)H!a$qpiOxV}8amm{P8ISW76vt>&K
zd{99nul>cg=$jaaI?%w2k(24(Fs_qEO8e)<!4Uk`&oY@nW*ji7oX6F60o4lYb=O|w
zo_NpAt!gr5jrP`#=_7EAD^jE)u@p!=3y_lPTk`B);11Z5kize!E;<~l0`gp*0g-}-
zInf!wzICi5PpS%S52TG$A~J_rUI18^>t)s;gO(?VVl{9&X&h5;*by8QA2r#L^{t(a
z7OGLwa&xgi3rv@&)!`@N+pnJJEdlXR5<4D`mH^NaW@K2|zXdxUA=t^b;nkQ<tjG|~
zey$cTP0D5j|4W-fJL$0xVD8R!-PYd*2YbuhWs}(QW=9V`O=N48hJ^hUrTE_+m_&p%
zAhe#gd$tne3Y9WTv+r)V#H==Gi-qN9czpC+i4vFvR(kb?mMN(#f|v!aDN0rU(VOU%
z)Jwzgzkcg8#K!wdUugF*D-5elJWZWopQErk+ZZtJ9MbmvXT<d$%r!VW0^h`ec<lM;
zpWXVOpD758@?b)$LImcs<knlabr(y(s<&Ip^xtbNY6MA4s_wVipO3~=_Pzl=YuQ6g
zhJL`@DqWMVep77sHxA)PWuRd`GCcRA9}z|Srv19OOdq@FZVau&d~wb6D^5{upPO@Y
z1QkvYc3CBlCvmmkzO&u-{s}9N;NRKAS#6jA8EHVC_aj_Vo=DSUBX^ED)oK|XpN0y;
zi4ZzPcMd!%F?Rq>`VHo6{`iBBzv>x@pAACL|M!=_FdcwXr<(-6r$aebFc?VU*+qcG
zz8DWDOwI(JQd(EV>VWu%J-Ru%v<RAR^EX6dfecXe<A!;x3iPU6Xp>hifZn`2W82~^
z=6T|TbFa=If;a@he7w0iZJAJXBE|pf)sYE*IH9~05kp5Tn&cdj0mC#x)!|wqwi^Ph
zT5JEj^FNF6uMb%?n7ui<G!9bs1pdNI`@pO>*spX4O&)Je4ClT(UpzPj;wE_V$}(mE
zfDd=-U!ML=B*2vr#7vtCX+lE(3kE-d{>gG<nNGoJA#02BLAwQ@s$67mKvjU%#C5};
z`i4K8$lpIV5me~oIO6aS%yBwopmqU<irTZBqv6MU8w4|#upb~egNQ@8d!OBX+4$;~
zaqcz($?(5YDS3_HR^A*w<RS8JxWf&|)B+Bxy=8>E06+NMEW`afK`GdJvxC6@Mo_VB
zrY-ZGF=*Tj+O~4zLVNO|rgOuGj1XZ=`(>4XkTj1KfRuR*l6NVh2`9G(ViCU)=m#v;
z-|VaMAq2U$@p89ujxL1i=^292S!}7)0{-9)UC3#gUtqax1~*?<N$cN4_bww4lAuMW
zRLln=$fA$)QeJ&D7=em;RQ*PSF17`9bk}<BRBJqmAGVYS0Z0a5g25^It&%8(oh;mF
zD1XiaBxzft8EETttQ@$n$bGh`#qh_Z|5qCQJ??<3h^@X^zJucY$N&LxMDVFRuMT?V
z05fF?fl}K)FUW~@h~TJrG-woAi=9tR?YUjP)R}`Xe%6Yh`YySH7(;QtfKD92ncSMG
zEfql&C|st<K0^M}6!PDNkR%7%mJJnlkJO(n49T<13Rpx0`qII;JJQ~9Sge+k+rUo;
z&;(yKScR-F7gn51zpf2S6)aaPefYcR<u5`397)kTh3|S<X@G?~mR|&LYM1S?H&QZ}
z@5l9TSmplMt^fQyMj%Ho`tZB{K^;d(d6qwxll5X9`{sZC{a-$)YJfX0{IsoV^7p;*
z`<5t@5h^*&GraMyV)sV`C0`<hucSgI_U{Yur*7E<;kr5OvhDumlD~@;!JjxI?z^E6
zPxk-5?^p!Vxwk&Q?(Z-A_qE4KK^(SuSA>N8?`q&Tno*7!tUtqg#`3*?8})x*6>%&7
z|Dpf6*c8c<wQ|vL#0yjXg(iM;W_${4QKO2)brr!B6HK$cxMA~SiRbBV&>r)XlvTj+
z6*>jTcCzMFmubs61d|gT0#*ILz~&hU=IwBOPF<KJF~J#vbq(U5MZmtd5wMx9d;1;$
z$GZ`%PEH8t)jo*+>Oqmh*Cw#(N_1*%+IlO1x5kHji%IIQ8^u6@E-P0?T|t>Woqpso
z8en$P)B@mlBcRd>!*$MnZ*@e8+P)Fb#&q?k-(efbsZf?Adu$fYA#6TO-^ac2;lGM!
z|0|)Yr-&VK$?+6THme*<Nnp+i;oEY;H2{5Vk;eN-?G&)C$%rO$6>kdwj(CVfa~eRX
z2l$(Hlbdts=KO=Cueyi$zEi5FR}ZtWFe3q&JPO!pgoQycyy;|t-=5mBNCX^@D+IfK
zhr8rxZ$Sdy-Llp6_dPc*f)IMK5E2PO3ltQ9cWy>d^Jf*<i&ueL%{P%FkADgRf(8Uc
z`*S?Y?Wbpev5BzSd{9lz`+@Hw{-JCr5Kc){f=>7+%m=Lg2v>CJYsH8(PW==9sm{Us
z{TznAFhECb!Q&z$h7kduP2C2eOcn3dfAB8;Q|k0TE`S9TvEh|ou)U`rNkS8R9Ga7U
z@B+a4Vsw13k8v)`bir>n%bj6#hHf=p0AgksMo7aYY+b<fQ>XkdKxTiobZ}ql9$2au
zvRU0|a#0j(00CUNIC#H@$}vy_{v7C53Sj6S30O^*5uhsHS-uA4Jd}|f(`dEb<T;Cf
zvpKI2;~i2Ao@aqomRke|`4AdT0OPhWWS9jRIZ>+90X{?xXB~-5KLSdMm=a+(P&xOn
zTCV?nKt{v^KDA*hp&v0oi(|C63cLjIS)DAI7UTAyG!dfjH1eL6izmph1RG~)Ne5=<
zf-PeM2!xhp33nU*D`kJw(LOs^&}T$oH@=cvL^w<R0nH$*QD$NR<QZ=`1;Bgf{E$)G
z53^$=ZYITm)v^IZ?In`;01K;3x%7=go}}!rLLym=*ln-dvGvjN3{c!E?4}7)XN+_W
zwcdUPsT4xE##y9BKfsxOZvbO0CVNbQhfv;sbocLC5?3}$N2cpVENDQnU`3(zK8o!y
zHU(HR%wRGYc5+d9H$_{)eVbW9*NK;|Li{bJ46|V4ia~X^g`582Au$kwj0D3PZ22w<
zuAE<#ose_rwU)>6=2!X?w~`{j{;>I2nCFlqAL9b*66Cq7jsU8=%lYDe754AXQHTj9
za($d#&(SDCTY+L)<*u@;`iX+Wp(<$4s@x8C(+UDBzW;T9siPx<v!An8*DGxP@8&~N
z24S?aD->f%{k!S>{7qMTV|U-_``9B+(^CQPSzFczaC_|$g9sKXi!6XQj+^?ura|?e
z9Qy!-?<aL%;R}hs9DS>$3<3fySd#39aKr%xy3Kypkptnqrdm}zJ_~1h=vHY5ul(m8
zMDVwxQXIo;F~`~b!M?>{=m+4*ExAZ2q0#3XcHod0aMyguX_|AH2`U9M5J2JHM&`e&
zW&g)<pa_?|EzvRS5MH1CzO)xC!#E`fqM(}l7Gb%FX$pn8Fl@^C?uC^N2LvzLX08Fa
zmSox%*=OcWe+TFL2O|c}$e6yPXaBos{5GJ&0H|M(;V5ta2M+LG&+CujXGLt2XD)0N
ze;4@Q3wex~mf<l(-Sy<JD*ey1WMv})64j|(#3TII#rpT>abAQD6mlgIAT01}WiN#<
zyUbV%xB}KmL%DI!3xYQ`|MW!)62Kcyj2rW&S2afiBy&aA&BFW!pneBl0#|bvMB1*z
zEXqZdns<f&e#0ZPXbpU>hY7-j7B(o4<LZ8okGS}e(-yD>631MOZop5s2RMG1Hq!fl
zsQwXTp!zqEdOZR$$HQAeEVDBtxB~f2@CI+c9`Hd2qbLQ2t^o@g;ha}{6t(~9`1;@X
z*-MwTu}j5$p^dhg4rXr9055fMuwo@zq=8N*xDI%;UpRpuOt%SwCr`Y)K7V_F!AFub
zoP$Z96vjxT48T*n0OMM%1In8bb&KnWcnUbuls%_^(kcI=>InRY5aHaq-GsVq)?Kvk
zAR*tA`48Cz$0?W=8>-%`0Y-HHm!OQGfj40JM(MYX`=`^f82}dMI2J!V=>J&hP~a|r
z{e^X3?eYJ<kN<owz(<*e)QTTUl@s^}Gb2KuQ?Qba`ab8m%o|CsJ{Ha*g6Rh5PuB6W
z!jA&3nI12jB^jdk{LDwAyUz4Fd<>}h8T?wzw7}U}dHV~{NHt_kZrNu5=6*Iu8;rOu
zFWp6ZqWaT!K?-_e$~$rPsj}`lv-T+I%9*eI2GmJD$w?k(y<%m@3l&AAM&^J<B55}m
zkn?3B^w%afnC+{31%l?%M=DSOn>0iegJ9LaDN!3d`}X2>p_vjX?02BLsJQ`b|7;YL
z=RLFi_E^4(V4nv_bd)|mD}^71H~zZ0oCCELYO0Y>EQNOKR%Ssxu1@^=H^nSsW+giW
zH{KJFm=HrfRsf)89Z&3NL^+1$zysvsWZ-~!zYBZmqUE|8EAQd~50~{Dy2uWpaR8%f
zl-fq5wUz-F{+lDERhfUPWnOalt?Y$|Sx-<9su)quq55ioTZudsG4Q8m#%2qQ0$4DE
zHh`%n>8-$4{kT`ZN^}bBW$lkf^JG)9T2K+$*zA4D??FV#e1n*31eF)N+V7wnI&uY4
zTApg(O-{*-8*Z^ecn0F!IZ(^ArRRI?3@mSd{nr;doROr-cC5181c>15_t1zR6o{Zr
z$29SIdsxZK1}8SA(DxI-x{<^n5Z0_gNZ1E&>11X>Nt${em}JLm(27(GGGUF{<l3PA
zh(W+@-{2JFeMCly`hk(A^0jI>1!Vt^m&3j{#1IMq1~OBfg>pOB0C>dDQAI9X^elmS
zw~p)m4t#`_b+)yj%K+&1ltutFpX-ff&`Ed)w)_HOauR3r7zQEhJ0v~!@gR8`_fA7U
z6aFC+Fs;dTtEe*CdmMOEoX8{}5OV~^1HdqxMp?1ekvd@61+1sHcnpucmAJVxB=^m<
zfY1TbV$Z;<_kv&0MS**@3%JjAjsrX)I!0kk?SbF$M79@yr%S~`c}7QL?wh5WrC7F$
z0o2?_0NmQ4Dg_8-i`$&!+S5TQ0>IQs`e}^=ba}7X*<yg98CxD=kdh$!FKf=0wz(r<
z;c&wFaLwhh11gA>1>0T%^879WeGQp}X}>C5RG5sl-zYHW!5QLc1<?5wkWHn4aJXr2
zgGaCdvyV0lEUG;4Dh~x_?E(;!0s@vI+9;{#z!a;Q!h=_&U(&b=aGdttq+SB{;co6b
zJ#3A37oc8VyXL3=)bth$A~g68hA${Pf(X9$a$Vw=)+4cQIAWH;z81huYhiKEfYCTO
zjbH}U+#G=M?an$5b227gd|~ZZ!ilx15P$}r$A<#k_|fLqe#CSML#rTjVd6Z*ASAag
z2I{>8qYDfjB1AH67p!q*;~Gqdir5q3Cwn4NhakPP>oV`Y?y`Xll;!<XkW_Z!Mvf$9
z{n!iA8X`sUGK6M%J`tQXm@k$Drz(epU)wFcv?}jfF4bvbncx2TMxYg>5{V)Aitl~=
zavRcn)`3gj)nyi%990;+Wf1j#uEYRQdc-Z+6*oMv2o@50h*kR;Y@hs>hPCTv!oui7
zPA&;pvn{@tGYi7-6beJE+m&u^dOW=CPdRw2xGSZvswEUX6+(7{kgT7A6O#@r9_#N}
zQUUD);w^?2v2U=j9#;})(HKnTWtyGe?{*h$dV7mZ^G@B3fCy@?3R<2MijvV%Q+HiN
z?i2@ddq+^%0thNb(?P(qbO|pP4JKoLo}87%Sr1~XIRx)A7?+Z)hB)hq5D=D*8?MAM
zSavSas|I<*5GzGesNEfqf6<#vDq?uA07mPY4U3&RA(ESu>m3=FU~Pm>B&F2unc9W}
z_qP*d@11PG+y;D_dt}P;@QQE5sEJsxeKV310wLhKQ}2F*YT*g361aj4KSoT15$u<C
zr5i}I1jI%$DS^KN6V_Zi3FYg=eOxMv-FO^>XYQLuJy^=^^my6#)EslS;NboPGre3S
zS2^Wpr^(-)dmLo=`)CC147hEcowDVI$O#!2(n@RXfuk5>U+gf<@wnFfOdkqnnaK}n
z9<2_1eR}3t@k$#Wemqa<QKFQPn11<V(AY&ji3KN~)rBx49>Sm@zx6{vC*FLa?|odB
z>{;*L-AQK5!Hji~Wj@>Mf(}*GL?7;?UuhF9k8+yO^V~6TYEujr8=1t}X~*G$3>=)~
z6|-xAJkYzOlH2~m^nUaq@+jE6rQ5O69exPG-Yx6Eb9_McK->zu7Q|RzIZo0X%10I0
zgQ6FKn&J}0=)mVKu`&tw)YIX;0B3Zx{oizKsbP8~D4Tu)tY_s8k@f{sfzdnQ7$n5l
zmMoHd^)r~YGPf~L4}G+6l5K3FMU1_zCQnWJtsfSLk?DKmu8AEa4><%Zxj|0QxS2bv
z9sgSM=A;dSmZZH~^Wi$Dr4>pYvAVAvrx_VUMGXE(SjOU$#okT=KFQqE$hdSu4YA64
z%(BNq$A0{X7zWEZyAs0b=4^P@`6E=7Hu=o5UvNQJ=>7Nu24qp*A<>@$cANC6jTMo(
zT7i!T5zTKL6NU~D8FZSjxi&_rUH+crr|E9XVTfTH*m|2k<dtSCiV8Sa*saksIDwF;
zfa3$xZK}97z);V?&FB}qu6q6A;c+^!lg0$CeNm}YWIC{!EKxA;O6m;kf65aYq5W3W
zfamaUh3!$Uz~OfSQ_|cc<PFgpKtf$QXJR=d4d2S5`ZIH?Q-J_VA9nw9XcZkaU-<Q6
zQb6vUz9T)}rLZWNU)AOqas=jwyu*%VT?S{(=JjnpGDdw|idM=(YY7!3n%`E9KD4d)
zNx+J#^E+&~w!y?fG*s3l?wR&Q{S5S--0AvLM<=#QTWC3(F=Sf!=R@30-2A}gG7MF7
za&Hbi4{INkanxz;$mns@i)#flk597Uxy^eZ1wtnn$uWK_4D*37awKZn_s3$!x}AY$
zm6*=3&riA)KExcb8t0y1)0{x9lB|rWLJgp1ky;-SCmL<+!B|FtVQ|EdmLO8p+~5x9
z`&9`tiWA|I>4XMsC@>v3GPhYI1z(?i2K<_5rgMYSjb|I!alRNaTq+Tty^8*Ooak3>
ztOn#}gCEH6hs&aWy3boly`hHk>DR=3Q@6J3b7#~AR<Dr)9%*=LEj^*^TclXOpeiWo
zTca7j(8ph4Pm8ibr3s|UIJ+ozt7h49i3o~Fkh@^2xlC`~)1y$V;NGQ>xEuUTmUu$M
z=jYDkp6Hg=<w=_Nt})~K)Uc3easBs;VV)lEAvU|j)Y>WQD?v6sx0$0F{O->Qy>j=o
z8yl=5YtGYijs?q-lWX(75sqz{F`CL%8PeQ&?QF{Q;<<|TPwI=B{W^sJ4pS8jx4F}e
znUB9{Y0x-&NK8Jj1t}!%*~fTNX<|Ib!RBV>>5S2_6I>`3$HzyZl8}O>VJO@n`Y>oU
zF*r*NelaFvV5E!oHjkUr*)?E}2z=L?9=O{iP@E??XTI0tP2}VKUeftT4a}qCbD@+M
zIC}p&GokA^IPFiwUw(sG^42!g+cM{+wY4;<#S~g`Lk)qT38vkw$#bE^Pd+Z552qSR
zm0Pt+tx?A^Oxn`0@9f97Mc!N<{Nk+4(tLY5`<f*&n^`8!gnV2|XmcvW+UF_W-82WG
z2r(tJ_?lR$$tM!W{S>Ru>R@&0z?@LB>0<B8&|ZvqO}#T%kP>N){IKdMCDZNu<%_H&
z!AOcNJjFzp*BItIGpvGVhJIU2gRJBPd3u;o(0%wm{&wz~#Q`Iw_or7vtQAV}FJn8&
z?%xi__Rk6T%YE>%bxPJ!zF>6nwTN5GeuEfZMwX)ECsh7ag%j)P*&4b!pvKl~K6Fli
zR>-;&_RWY~Y4?W@bdEu0?X}{}gbm}pA~9`#sZS-GXxRSM1aIevp1gSM2sY-x9^ru7
z3gyCNpNEENnqVKlMYTr6NdoE~lNSFthBlRAQ}jqa+Vgg9@mu$wMXJ*1->m7+e+&OO
zVE2GW{?h5LycN-jXUnN)F~uYPxk9BfV$*yyR-$1~Fb!m$q{8_p7{>ICP1U}4mVGgZ
zV^F(V@2m62Sa`gB4%AcwZ8YO&+IrPs*I{~stv-nOtTXRuK~pJakZ-hxJ-&@*VW0{L
z$ov$|*C?*+lxq=8;n#r`mbKZ&ojQS`(OMyVttd(nnEdmJ_iM^a48n?bYJD$jYx`)o
zat?0YYL0~7e7nV>HPm*d(r$iX^>dSFid$QDxB>~9F^1l=P@Z>Dgr@OpC&kvf!o9<8
z1Iw($vx9PeM0w&Vp<xV-aiK>Mt)hJTjQ0W&^3EhLMji{B0h)QzM-nIOw%2+;Q{1`n
zOMt)C5iJ?gLT>I!ll)8l!xI*|f?Wo#UTV@%s^_o!Z%L#h)hOs`8ksXZ-Rg;5PjU3_
zR^LC!OSMTESQet&<%w}0cy@c;*I08p0-bi2qW@w)x`}&${VHhKN>8xlsv~tiN;C@<
z?L70ODu3`E7|AK5(vED)qg%c6vdG_6DW8FFBDU#g^f%uV5@cgp{8$Z>9OJFarUAh~
zg`b}9bEw`mJbLkpcyE;LNM{JuzdsG*3`aj-cXMX=S1N|v{dNEKj}Nxi{j=UOLETdf
z=0CkuOsBp6A-|r<EHQ}MPvt;UK;KdNy=?{9wR2IZNL$7_Ne7sLHTXnXQ_pne9FeIj
zAzqR!1;dkZo+VE+w0Gz8;k)2rQOmWAd4AN4>bl*$AM`?Mka?Np(?bJ!j!N<kXBP~g
z?nV@fH=)yaLTFCwdaC2C^4x}#%5fu;FFCzq?^FoT&9JX!n@<Kl=c_eee7rE@ITT!M
zJX+!&lsU+D7|$7Nl8SWVw7`Na`}}tQU8x`3!}%SPidMw@pZfG2m7b2LZOjasx6g=u
zhu44P{Mpqwzb<|%v__+7HHmds!YXOUTmO1fLm>65=zW&R%woJBn0t{wI>AxzHc9Gu
z<q$Yc+#c%A?7l@?#@iHuYcn;lQr}Quco>mBn8!zUVYS_KIZeW)<pd$b45n$Po36|x
zc+^8s?U8Fu?}yQmO-bLO5Vw1V#bSfmVVxQ?^IYeGBD#a1z4my_94e|f6B*e-(Dy2I
z5FQ-bU3)1Vi&xIoOAsx431KL_7DJH@a6L(L{!E2qZiOk!7dR#sNJJ%_jAvL>P3ud!
z@{zV&^hbjHkJsB>_X9gbeOtK2LsUicbyIzNPjPOE1eSaDRZ7?6XH$i6GY3xwx?{4G
z9+&SCQ{~*Zz%=Jfojced&$bKUCi{Lp3gdsLX2Kk7=Yes?+<on6n7FW(-yAq5FvU^S
zVL?fMHh!^`p;~556_sm0c(Rhzaxx!}J9jQW8Qiw8`{f-75YA<|csmLxmmhxm+?>iq
zD41+Tqen<ZZsfQ@N8<fjf1$n5R>E6?!GQN+*e`gxoeyQH@?a0<D6Sy~|7^!w%Jz}W
zZ5?R24D{(w{+LVOeQPV``7-Y4&#2=0Zjgn*75-KarS2gwEE(I-T$5r>(d5JSS(lcY
z4mD_dtma(?hF-xyL*YGjG^AAck4(zfqam6K4}`cf&pKI7-tC^XoZpr@pi{o4*pGD`
zJkQD$BO`>T_4DnWO%40Fgi3yf>xXHfQHB8TYbW4!?;wzmder))h11OPn&y|aIqJv#
z{&bSjBQj@|w4TN_7BnOIRwF#*zc9lgOJ02Q<8q|}mtdkt)1LPpWqUoP0PnB4wD%El
zzCS1rYx5_2D?YE-FATO|njIl-#hsw7i|!4L4BGSQORf}*a{PHOnvtJ`N#sUT?Qs36
zmV!5w?dLXE)_C%&E}@<6Y!e$gXtOcnqL@7%p{UUBdqizt6J5sfTHjp(klYkCec4y1
z^l~N7(CFPBm~I$jr^#!=Bue_QQ73P-htu9ddD92ARG;j6LncK`01^t5eEw&~DWu%B
z=c)m&w=h7uLK}TekXY}SdcXAGag5KlG40Hk5`Fqb{)c`RJML!li>{8Z!Y3yQ4X9{t
z_CsogzHOz>`yBaRm?`bXEDjO)hpW%?p*)azmOEhkB?J?O7kg*UzW7G=PSG(>w7#uz
zZ5WlIz#feLTyUZ|R##M3yyz9}f-I8Fjb1s+_m4U*UR){9oCXM5_X*dmqgRY+FDDIW
zBuH4-KNS#Y1|fCz6YM44mX)!vXBNN`E%Nwg(;Ne7CrCtcl<F&3g>cY2M#j<Eu~{l)
z#6|aB?$^{d6|*^XKYKXl%BmgiME!n%A(iJLEc^H~ui!UuwsK6MwU{>|c-L4r=k5|o
zN9#{`rh95%Mm(+BdC<!PABbjKvJ;xnkF*l204V~kIRC@d0s9l<`U!&MGynVjThM~o
zwq&Z;Z5-GlrC)u6ckUXs%EJnxw4On5V{fI`wpoNmT3ik#eOEDIV#f8A^K)Iq+zh!C
zDo$`TF->_ItpGE#FqbF=XLCW_Eq>T5jHN-T<y2$l9duC&$>+``uEottY7~3l<9XT~
z%hp&Q=DiBFQs>=12(R&?B}FS{YfCQ3vOFbc8Q$o@l$(Dpa?$BWnm}yuv&ieVRI?zN
z*GJEuL)-7T@FilcA!=j+=zaS5mP6jVEy_z#3Ad?gs3<i@2V0JHTE4#Va${8JSRdw&
zK|{a&xwqgRlzvN?HWiFsG!Io;i?-jNoqXmIa`4(JC)_eL@+$iE&)dqy2KmiD*vVy1
zbnUxHWu9d~V$*4tj~cr5Ki~ddWR7)54!z>D=n#YZ$WN7Lf>o~tbI)Q#+l<{f1C*u%
zaWgoDsW2D$;1NByBrg0){0_%OTZhx8GxjSMO`kbEn7{wMJ7qqy&yOE}fcCyvrc|w+
z_Q3>+rHUP|yPA%--;~&HRMWN#+Jk`r+B1JzO(^D~iQFy4EyW>Jk2Nr@)k+zDy5e_&
zC*C+Ca`Ayys7451l<N}u70H^Wb)1U(_5iB#tAz}|HnkKCPf`mWGPR{ICftw>!Ouu1
z1eA;)!*t%W959JYaTyzFd>K^x{H2@T`WABWB-e+it@w<i4`%(AIQZCPOth6EXc4#i
zj{UuOo4couyH($EDPzCl4}Be3C?u_6{_DqL!7Qa~=|TO65cl9(2RxG*wVgVdg)&Q{
z%qI2ew!L}%R^fQfjPn9{v5!nTKRnt#Ck(pbF*lVQ45>wZ-A10U#K_D`>mH7~?~Xr}
z*@k)TlKvWvGWLVsvw$igmVIMA7Mc>i+AFbR3m@#WJs6r_WcB)a<n>*FS5Mz^eXLtw
z8Te){P@+g`p>?~E!HGJLAYF-h{)A<TZtVFp%%XtS5b>jI|2L12)31855~hS9A7Z?N
zr40=3Qs$Q?Km+JuIs@xa`AlLk_=k{$kmK>nV;51<vG=Fne_(zyNrZIWQEyM)*gN+1
z59^F~h*(34M`!C&30tck_Bp9!!HwM}W>RGH;OY)2N^Ko2z3M9+Sh%HE&QU-v7#U_R
z{xf-y$Ej2y?}4~zyxN0qjNA59|7HPnSH5Sg7V@6rn&O_ckuEPX4$am2=^DXwZ}3bz
zA6K(qId?R~nIqp~DsR3_BYSbr@~rcI46HoO107ZVmSQt;NTSpzaSj7acrZX@I&m%K
zqntx#vll-KZ`#cRDAbnAbE-d>)jVCmI`Sg&DM$P(^yqDxoOBs>!CoP?owts0v8EhR
zY#o`XcUTMhx_-s`e)jPe3}fUy^W7cTZX489)v6AEs>hub&jKx`YVp~wVrY#`R49ou
zMUVf=QxbK!Rd4u70&;5ayI-6oM{%BpvQppsh=eiHl*)0PsFdgXEzOHffoP+e!pt-i
z<;a_5(`W~(O7Vx5H_=F?erqan)j0L^S^0NyV`5cMG{W%nTbpD<@R3QGwPJ(z&<@Eq
z`kGI_`!CY;@bx6y*thgWXP^l_b6#^EcI)P)@I9Vb{iWX1@WOZh(v7CmAt;YQ@KH&J
z@9N}xpD~<y_4l2}t9l#()<bR*_>3YIPbSJ62RXgD@q2xI;|cRqU#^7YnBN;=m{(6s
z?ZX=K;cD@2DQMtX_{HtLE1MN0Ce>U<XCTjv#U1b4B*uue>?um%d6|w9!NU|UcvQkG
zj)MPD=Yr?>7mMf;mSeiS*+&;yl!fj4#x_@INXF+Qx-pgR`<3>tPtuyFOjvG+2@Ynj
zVz)OhC3edmK1IJbOMFRrFAa6Yb0bl_RXKm`(khp6c*rmQ<lv=#ai84VPC^bsuehXm
z0~H;%%fN*>77Hv-e|5*M+3fzjxI+|kWAC%r7%#@gnq!rU5bwlEMG}q<pO2_-Uf2!r
z4{SIWo>*XwM^gr=%|`R#=Gdv|aiY!n+aT4ka6OOOB}d`sucr0m8~SF;tW`_pA5tU!
zAVKfuTvELFXZ3=7ocDA3_|%r&Gf2?fw~yMc;q92O0WiCRD0n6Mbum6JYWB_n0rE(V
z;iu~IrK8=td%jq0bKhRe*I4<aJTk!n2<7VIs_!mOS*vU0!;dRj6C_1OOInyqj|Wsk
z&tL0seOid<b<?b*e9pZ?LhQK~h+^%~<vP!lR#y`|xkspJu{G3j{_4rJ`4SVC`{%3P
zLDiy4Oq8@YF~<$lkKgFCo?JN>tMs4*t8usRN5&^dC1|LzF1T132@Y64c`&;R*ax*u
zHrxM?rL$m*>igO-ATc1)Dc#*lNJ~k#h_o;u0)mpm07G{-f`D{)3k;z{HwZ&_cMLte
z^ZQ@#Cpg#MXPv#*S?hW3Wi**|-P4Om2K3b&frtyC^*VFK6}kv5VzSJYuhlVvTHDYt
zk%fPpl6<^@+D?m7Cymv<M?RkSw|_m69JGN=?9yds^JlXO#$Qz07A|_8lCr)|j};2O
zTzCm->-upa_>x9oja+tZyeTQ6;VHGb9f(b8O&QQ~Yn%8_C9~XXN?Tm$wuST}V&y0(
z8N5HI{kK!nN~8KuSI56I6-1*M4ZOPvwq*VSmS5bzJ~A5q!efBt!@}?NG{|k5Hr#<T
z`cD&bsS^VLdAV^I2d?CR^hqvHAT)%puf%A(U@Lcm?AM>iWM17Y`qbnkTs`ToHE(<e
zQzS5O>7P=A?6ui_n+1}jNrrE;WqZ@VKG4XgG!g2Gzp12yhCQ3Qd`7QG+yXxvzS~+Q
z?gz!wd|0>eQHrf#{w_ef1hc668>AaD<2C8CA7Lh!ssvQa#3A-~yqAf-72OGtCMC-F
zH>NH{7jR-MzUJhN^?g(-Vk^g~I+=@ZmmWM25;TW66(sEW_Vps3Zz`!n&RbF*(m<_c
zx6sFwNO<vdV3IiXG)L^FU=Y(J&A6!I9C9HkqJ7yrrkaCeqsN;4V>Fn-%wAjQzT~n=
zixqbssSWGRekt8O;m7M%(eZ{Ti=}o7mdckx)}7dtMwa0ubyoR!O*>44e%@pOfUEJZ
z>)!t?SiFjGKBZpvD9ac6rw#rUAEkO%@Q8%p{?r!5Tvn>GQaSH<Pi*Y?R4vxAr+MUu
zNG4S|%R{o;C;65=-K}2(zHH5oEJx)U+&`k4Ely1YNRHworl7;d9%S7PO}3~k3#Uj_
z<ch~G#yxu=-_bmh0TP$H_@Ip_H?|=i6PW)CS*>PBsQ6MjE~LxVl)&)(w_nvlb4*SR
z0bPY>y%yoYj~L>9)r`<3*o%r5`TJzJE(-xhSW;~brJ~NI@)rA8O$PAmcGH0G{Wm_V
zb{eL4;(z_noJ7PkzX=H@X1R`~7he)Y0+wXPdscsGc!#~dS7utG9}UtRm|Sz?<3L2a
z#KRQ3cCb?hJcA#ZrEnpmDKQ^(ZI*3yr9C2b6UMs7RFkEEQc8jE_Q-Ct1Ej9}K-GS(
zMVLwLB0#<m%$$iQ=r|C&3PEP!{=65$fm56<p^GAaSMJyZ6Ks%|gS=0jhx)55|3RjP
z|9-?vMyDAP-OkMD`2%yX>np0(|6{oyBzHS})(9Ji#R6h=L6d$D$7W8JH1MMBVLaY7
zx=dw$Jc90=2_TEIZ>Tn}gnf5MZ9B2cr_W4dO}eGp=Rah+4x*W`$q$plB(#=e5^SWX
zlT{P;2A`U~ZIR`?68>sm0P>V-sv-!kocb4EYk4YU+9yx3lBtlT1?&9w$9j7yy<cb4
zl0-(0deF-b6V<u9;v@Q!CBYQ<>&AAlb(&!L6YoyZq*qr_JLud*MjhI(P5_uR63&vU
z?e;Jj9l2seehTMEptcLnB9kHs-@E^moeuRP)6(@#X`b0(lK1Zv3RMj}^Cgmal0mZe
zVseQ{SoxgU7#L}0zkijDbNS9$%;w&!2^G(;Izon@jWt-6$_cd&rxJhYD~`%~uT>W#
zWsUpJA*?-WKV*BgS$6R@4kX84Ec1#`Jen)Vi7Ka?yscSVIDyS3yc2+TdBI5Mnw~+}
z@o<y_AwfD1=l`f0U^^M2?-B9BA-uvA!^PMTB*0Hv5mlfNly_tkeGW$rwMx)c<qCV$
z8Z}3$)ie>RQcT@M!W6Y}5ApSGKCy)-cb0i}ZIFaYh3$=lhLbt$%NS1+B^&6Aao6nm
z_8i)iAl`X@)p^01S8*K{B)3Bpr%3zjRpZ&hEZCFiT}pw_ij>Og_!WkgU(vf<_Mq3D
z#iU@LqGR~cO0-g<XI|RrRGSSwLR2Jcwh*tOB5InrZM49uK@mPrynnUqX_(45D8AF<
zu%I2CxSN5$l)P$w#f##6msPbL%5^pmiT2uvDXJ<`e-c4ik9yOU^Y*j`60oT3$~`XH
zHIIS*x@ys)1<ZjubUfczbRPm4?SplbxwZ1}r)QHC#jl+tD*ewh^Eg!;)r4cz9#}J*
z`Z<TAo0@FMDN!~eISYxmrB`a!Wde7LH2yAw`0C~-9n`$~0r{^KP1SL=C89UvqWt)T
zWzI7mIatzIGSwCyz4$~l9vdH@sB2*X=-Gedj2i0z4$7lmv=_?{m)X7{FTw%>o~bCu
zF#ebd#Z-(=Vf<KO^;OxZNeF2}Bt&ElQo4_<-6+8K`LXG1+4XB|HR8nP{e%}UwaYS;
z!^gww5Vk_Y+IjK<&pIhq19I~1NmhofJdtYsKs<dji??%Fig{!BQR!mk^`**~-8NnJ
z@sln8Ro@$t$TKW+;s(1LSYnH9y2u?+ZwmJ^T4;^op%g=zh@e_BCh|psauc|pEyc91
zW55nWDm&(~gAh6JfS9rE%;%6u^Zo6z?o(tg=sYK=6!si_1Aaie(K#yq&GA%YS=}h(
z*^qFLwPfTl$-{pYvG5A-#FG(CSD2Iz^<WiqbH7J%N6AOig8S2)*8NwtSvVMhT08sh
zzi!hm7>}@VDaVwzwXuyK2%2Sd9Oz4w$)fibu6!>3`Y3Yv(#N+3V%vLCJ?~E7qT_Q}
zGKzkMcJiPiS(cZ0>d&eut9<)&EL%2Ob)K@bSKU5;l^79rrm45}a>(uDmcdxLFIJzh
ziKe96S8V5pQJ5)kszvKG^Ji`z7D%=j61$Uud^7XTozVQEGf`thVisDCm~#s(znS$Z
z2x3jaB#F_)Ha7hoJ-5(d)}3nniehLUzWG}wJUM{E)x+%51|p!U;;Y#X5*4;VdS<S?
zL^Xc`Cl;Rnb?Qv7Ycq(?Y$`YKsAg%)L~vCzNSfbXYsz-8SciR5`&;^=uV2`o0t;8v
zceH3$#NWFRhIaVVzf(F^1GNo5Pdy$DJWvIifo^o|mpt_kk{ctsCHCUNbp|2PBE%r+
zVve6wPT!k~s{ST<$voq;$3&>HESMs={KruQWnXSa&gJFYe$jJ~UvYn5qyh9!E(+<j
zv)n(>MB!^2&s!gPH))qZv7|Eb=A|AkkMJ@r3>Y~(_gldJ3vCRO!cX*2s^IKjZgeYx
z{-;5$s{qT4W~KIRKjrM<)WJ-jJ~y3HG8Kg)16Uf9y}d5>p7>z&0M~cz^Vk#Muy~hR
zNS`&=P1^UEH8vif$^M6g4)q@E0|UhBv+gL?*xcN!yjDAnGvnCC`=Xm@S3A?hQ?}N%
z8_ifY31$-x(vuIsA-1CVf%~(B;aUwDW_7CTO<X_WG0bgVt~MEN;3V~>Oe(onW2JW4
zh<DKR#EhFi0vuQ^N9d<Q`^!Bw|I%st>6#w>yTWUI5@gr2asZ*kwy3I0sCBa={c3A%
zJKYewPt<JEFbz5dx)ZrStvo>@YaGyZ!ymYumJ=9OYq7wKP_u_HQ;s~lklsP3`%Dk8
zU3aQW;I-`K{ej~qck-d^js%U4gL6)4J3zFf9OR;dTk>Z!S7tzbK-bVC+?7zQ&XJVF
zCs#p4&?y@8C*S=?#SY*7AEn=}JlnENHDn5h40&uravT!9L42P-IZ_hS@ACLxdD4Nx
z8R$5&<>}B<HKuVx<Ld)y%f9dtYnsU4V+bF_?~^|8kAFfkjS<-x7vDu{O@|$_K>O19
z>x2I7&6YWMOQf{f&qzOBU)l>)xA(Rg1f;qrlpVxMVI;HLWi$PzY$(C|^waLIt@DAY
zM?rN$tJSHXo;aql_i-UC(n;1LIcc8fb?j#H_*j?R8<xBN+MOHzykVU}W{;)0fqpO-
zJv^X2m8t9OM8q6asSOV|B`AQBT^kgIoki@BI)~DPfgBqUPTG$-wR1d_vF(wmN}-k(
zLTe!FeBt|)!DY$aC-%o4u~sS79AMJj@ZV|hO{U#OH>)+XPTQ`TX<rw`J5^?F*dLuS
zphD-AEVjhXtG>Y%tk<KlY-&-kf3N$~?Z>iI*Zl!=Hv}ufgEmIg=U!nK8)gOrF+yXa
zryt)N-_YQEC~|o~-lT0L1W>@f^6L&X#oiaM;<O91@Rj#j(clmiyI9^71;kaK4SR8E
zK^Qh(?<gRbT<;{Tc&UbcLt+@?x5NW~bv)=erOvR!m#2&U?PM8-jeLMVp)w%IIlJh)
zuXUI?f>qQz*+NR&f&Ln!)GY(6BGY}CEUv->i=1yv0juRQF6^9I|0+ZXcVD0r1*b`T
z+3>BoYyKNoT77<EqpfX2dOJTP)UAA<feYA@Vywobo&7f6n5MoXH6%6eH~Q?vnf{Ne
zm}a)CpG0Xb=!=Qy#I?Ya_HibTMD<!xp2Oe-<*PwI*JHMiTHFZ|mmlX39GMXJ+&?8?
zJUjw=oi6EYWT{wK-jaRdC^f*C6{#vM``YU|2X6cMg_s*#<6Nw<ge}!ONnq0Uwagxt
zPwG(q!w4;L1<B?wkvV`P?%mt2Am6Ugdb^1_<Y%4;nY|-CL@JvLtaw%*^msExYpu?f
z0~V8ycR!|UZX7u2csO-@XvA%W5;hb_*J#n83Mn{+EFLkqef@j&V+%f5gJJPMNZ1#A
zMw=EAm1J1dB=uLn{`SS*G=CoDbPa+oYyd=>_@Gw!Oc0Ykz<hb<_YZXf*|nT4^7cXv
z-hb851zc@7$!D+0`LY_gNYX9;unGN1O#zy`j90{s8&bHCe8^TCS>cRQqW;lt_fYzj
zjG_}`D%#vpT*W0U8lefs!pPC?rmxl6hKe4AzjpLS#GxYZM*7@yY{b{)A?B>{GlXU(
zsWhuw=0l)!-{!qrg|8jH#J}Xak(;_f90s9wBkGts?~aLL%@iSFy<;(cm!xm6tCLrS
zA5z-l&O8LpkKGhv*+Bfy#QvhNWV9+iy;LTBwRaHhH}p?~^SEyc@-aEk|G9fL62?<I
z+`s0-Yj_*=i#I;i#<UN+>2{cCjO{tTT(@~$cz!DQ{c3HtMg2&r_Q~83z3>n$;nMiM
z9S|;V*Aw@Ji+Wp0>QEZBH}(S=9ppt-VwB*%kH7!Bf~DTZqQb~wuhZQzyK^5?HWO!v
z*6f0h&XMy&QS%r28Wq$bHq@@(Q8OTADOTKmVJfV*o%5jmdqQgbDGSGIhEU>-QlPT0
zV83tHP$?*PP<V&N6l++U8|WgolEY6vw~&0QD-d>rR=(+J@mCe9wdeiF$N_%j6%=hO
zNX{OW6YU7NrsV5>jl8k2#9XuUHUDFAly`#zS$7w;OXT+J{#N(eX&fH`?(9k#tSim6
zK&3cjANHKkp!E{}Di#0rHApu&$y^3xo?SFuff-n>hKiA<&5vjIg?!DUkN+P}FAC+1
zIk)DY=bcvcC8bEKDc10kd{x<#jzvupS@~ygm#S~Gf^ADSAWzgAd6xcGf3<7uL&-i3
zPGj|?rHha#xKP9o=R$C0#*++@Cx;ePLAdU+OF^AK-%#t4DACHP)0KO`V@5UT>(8?A
z0Wn~ePrDWpO}Qmi<=-!Akiq&MA9FTt#U#hu+I7lLi1v<~CgW{7A5GjL1KYte>$PAu
zWkT&<ePWgV?X)Cyve*V^%$t*<eU(D=-Vv>b)-gPlF4Nt5Uj4D{#mi7yT43N=Nb55Q
zus>Q7{cQk+5B%C;v8s03_MB}arF*MWNN6(Ah>AOKh9jmw)UOq@qBu}bzeO-TDVj9M
z4L;{WFxBP3pi-X0f`k#B&>3ymf#T)L_qpybSNUdM!Hi*bIZyMuxGRHq!v4hB%<7m*
zr|AsmAuLhQQ)>M8)W~fOi=~i0su^qHI3QSL{F4@MB5(tIaQ&!&rVTvp(LY;ii*?R<
zWV6F`OhX0E3vG5yEaBZKnG#J($${r9^O=8Pzq1_2L@B@HDm6(UUlE=<+P0l_G@a&2
zBwTa}uldny!0~(+Xjf#sSo(#x=6>Kb!_GVKchjA>hgI2Hu4c+?j`KFVg`;=x8k9^`
zJZ_B@Z@MOh{K%MhiV{CuGG}rwu&<bjk-v(c>$jvaj;=1ctlTNJvC3p+XrUP6%M>`a
z=CQGO7+ogudZlv4PtV?G^)7Tn-q&oVswr`@`LNt+xKcBa;%;u}nMa!9thxTYWOi>V
zyg|qRC`v35hHyms?$=}R>6g%+L9MWrlVU8c1F8JdFH5jKA~2%9xcXvpvV%3}6c|M+
zDb3t;B@;eBzHb7vg(%w>prybqHcJj5)%~0}841A1$hG8dTxp92*=JbWZObjM`Z!Ox
z0<E~euIMjEL%BWy^rn;fLi&IGO76ILV+?XCqg`uY(?t$NAEKOcYywW$SUA3ig8N3D
z`A8cm6>0#q)$;hR?CE5&!Fe=?@5h~=R8tue(}c$P!23hNUFtQ}8hXhEe3n>u>?0~6
z{Fl@Wv4MC6_znH9gjgkTh5-B+QmXh89AvoI!ap7aG1Ja0mosx{Z<qbaeJHvzz65n=
z1W>em{Ny$n9t1Z6Q)#ELALuZA7J47GUD9-#-}T=44_-q{HOIvtch#HdjmR`p=W6-@
z=~-O2-KV@sPp4HA^yOEF1F!qqOtv$mY>%r%IO?@+qORXjZ~4s-*3ghr;}>s=1pHAF
z_VRIl?1TWuHGNlklSmFOb8~Pm$Ln&NKyosOA&{YV%rh_s-;GoResNmA@I{vlA4cMK
z!0o~AnNq}aColf6mK+?d;QCizn=cr*DnT`z@SY}r(#J5&P_2C_e{|^|ZOOh0{J0d3
zYKK8GrGeV10}9rRx#<BE=n18AQMZi!@dQ$x7b4`-S{>CV%T^imGq%|Grch4Y_21cZ
zc)o?<a*=55tA5zVmriP9*GIhfQ2$>*w+xN%eBeMN<BHK9KzO+^LuYt%j#<*v*h2KN
zYPQ_5lgq53LK9&zsyWM7_Wj3~{Qr-jjCaA$SRexZHcQE>*Le&x+Y@iQK2981S4R$K
z4I+l*=n917SFf>uI!@(jJY!l&1L~O%F`to*SLVw_SyvUmTOb3z1XT%*aRv+N{t@uq
zQ{U-k8jM+Mt9TRmZS5XkP0*6FgDc<iK?<s`*HJrMYGTtMEXBFOI`~Q(Isn}&7R5Aa
zmmKt~)SCnG@g|Bx5Bk}#4P;}ZEsV33&|hlhcj-D;NF&eUWdH77EVm}q$_Q;w{_5Jm
zczF}jV>|vlw?CDi=Gmy?lc|@zD{&k<#gNPy*y2|$r~Yoo(O4pZ3wCU3qc1HqiCnw=
zw6L|oarv6+(z;LA6u9bY<I3+v5act=fDzrwiaYj_QcK{|p)lGM`qt<5E8y7cPOhZJ
z?%gdxn@nUASH|-j$)Mrv>XF*|3Nu&spX5;2@F)TX3}2-vkfp}#PaJ8;8hU$2I3}C8
zTWL9}PEfEzC!AGPLrc0P_3l;xJ#eM&e8?v^j&hIJX}qH&NZr}qTI{0B^mqqdTEP=>
zvJUU37Y%%?8s%o|2g#BhU~Ll^_;4dZH>s7=l_#0Dr=2-agb&-X>&xcRPkH5-*C`z2
zXYpq?^^Uz(40D1a!QaF==d+Q^cXH5}OuFUQNi2O|i<y7Z*NznANvU!Z#M_pmx)&N-
zo7}%&RN2T6f+K`dn4%o3OZ~ErSMFLzs+P&pzU2Z96Dv-3O5giDIXynTWA?XCX@su5
zaGpy-IzaT<my%))@>Q_W@dk9;@LAN7(X9u@%Oc}D{Nw{t%}WO`c=f*AbQ+Cg{1~cx
zHn`T|soZ^JpZCd@n9qUdB*FbZIdZ0JA~mIQCdmZw+7%{#Z+|K*72P`*F|>1;ldUjN
zRKF)V_)XV;n4Z~~n~&6MRAPtMw5~W(F_(=egz5s{9M;dBS8?CrPBmfOo5RE?SCGia
z^K5lGZli36N+FZM)?=gt<-|pD!Rn9&&nh(2U40Vl=6h9VG`SsT@kC?q7r(Y#Fode-
z{1u^)IDT}$g*=>%dUn<WJ&Lt(NTjS>YzonX6ml-zH3BhVf29E@$12<Qi*Z2cfyv7W
zL73RWVSX#mcp|x%7yU^b9h=m4EB?{)+@FPmcp+f)PD@MQxhc%lIoR7`Y|wMt=dyi-
z>u_k@u#-wfHBITfd&TvuS5@a``n_)??~}n1>#Fz`S_Ti-vqb@bw{rgj*QbKwwIi_B
zl@-f()_Z_}{DEJV=ahw)gDjo;hLhACXiR?KMEZI4+7+$7lv*|mB)|7g$jRbxJ?Bj7
z6uI3kPR#f&kfDE_K1XvqU_|OZw^N!`+~&5*oMi{yH(*%(g7GR;VjSmPzq1x+aieqb
zlFyE81Q<Eea$~H#U9SI)Ri|h(7BqgK`S@`&cLL{;%GFexd##*Sor>zjP#AX<<N2fS
z@l5J%yzt_Q_3hCc8+r-4D0-!;$B3Mp2)g#g&TvToOtv<>PCr0O3RPBUxAU{XU|-mm
zWqX@g{pkB8haZ>StH5`yjSFcO0!Nwp7MHKWAU9#JYvbp5oxuwW^Lr858#aZg*&~k^
z{+%5mDDy&Mim=M6nTaY!{s!hRnexy2N)55TKb$J>XmLLFS8Kk@6?Fn#^Zn@4cY2=-
z#7kj8iY38=jjGHI0JN_#g3GP#oZI$_Y<2X4Wv^XjXPH;Y;PNDEnOaePg29(@-+jQe
z14DC$GN^e}lvNlQ)DzB>ZcjQrv6aUmzc+j+siDz&FtopkYT0iC&E5!;jiV;xH1`sn
zzEkckquY?U@O6=2@V(NQz0yY>&|!U?)cJ;PT9nLa{g2?%KKHv+N_A{Zj!(FpdS?!l
zI?Cs%-FBkb?<#;|Y<Q<lMY0;abTCh%rUHK1cc=nAj0l>9nmkE4J;0nFl%Eq7O?+Pq
zp%e}J**%dYnw1#8x#V4(#?o~&HYH}@^*F8lI4w24&dT)R5?eIu>UE6!4FDgJzI7q>
zL|*Nsl<7cAn9;fnccB}`j1{)-o9mj=x8@?v9L)raJl}Cn)e{QM+OFkX-Olb{XB(D%
zoRVFl-=L$vDp489z8$vvdQ`g&c;-M&ok`iTt2Cl|H)HWQvwd19(pRhABOA3Dl8c!y
z_Uso7tH~NC<<~6Eh0n^#hQEcBK6CKl05>L$aORU=bHN-J*B(_O!=oPOZ2C!5_>}iw
za-&x=|18iiYH4>Y*>fDi<fYBOq#e=pE_V+<8lUdI5L+C-nNl9ZQAPYF6&DNLY_M#;
zZGS=1l|bf7D->9`+Ebr>Re|_@i-3uB*78X^Lc$@uc4tZ7S3zUo<@O&0gaFJWz6Iv*
zE;CYHy#hXK&ohVpDLDjp<l|=QK$a14p!f`sk0b8F2UvQVapgMQUUO$44R#{?c0yCX
z=7quAo}0C^|4cAeyB}@nFV82XGtfR$MkARHP<s0FrYzL);W>IeqAp!*;ZgEv&avq(
zvo>qgP*SC|dG~wysw#8`u50#<BwX5N_^kQ*+M0GP>5bGkLD{*qvd@NSiQOGCEZ3IV
zZ;~s7cJz=yNg24&rtZGCYF=qq^$FYFo!oP2>Ba{t!c9Hu6QJo?mgwa`9kVx&6N%`H
zWfMi43l!vH@fa$@{YSrQ{XhmimgAoF(+d6YT(bb=b|~|(Q00!a>T-mZC2BR~sh{&r
zh^f`HTEf#NA!1s#ix^6pVG4AafnR>S{cc{YSBQ^D&AeUIylnfv97%aclj$+wN#lZ?
zdhG0*#|3+?DpwUVeE=CoNY@VlQ33Nxu-(fq!k7rOUAm{&nG1J7vX^Yxip2Z%B&y0}
z#eC$i6e`9Tb>2jGe6#c?X{KgSc>@JcCS^}+g$(RO%Tj;FkCI{OiBFPH#^A4pJRW3z
zDkxb^i<K!%7#(O?dhn$kx18qtes`YzK$iT=06dwqEOGu5VSXA&IsC#84<RuI8r_6J
z;1pOZG!xNTep2Uc?lROOr5aX$!d*TEJ4V&+r@RSbeBo-%s*nuRk%_8ZE71m*+fXRW
zMI|H=XqgYW5O-NRGM=zH{Jl{n8Q)cpN~s6)?_^v!P%r6cpC{AZ7pJpyB(2jcYdXh+
zMa@CyNjZ!7WH1JWuV4JSflDv=sGH`sKBe!MqeZP|Xc~<i1hM+`qa#=x=Bh(h-b1}R
zg*scgULghPz3Afryf!lmGQTYr;;ILNh1M#YJFhC8Oi7)kU|g)O-h;KH&;ji4oJsw}
z0jV;KhZ))O|Gq6_y?ta&V8GzpV{Oq&&5qtxrHMW&mVMlFGVkiHi>-CV6{zd<3^tpU
zto8lf+$W<hE&E(KiEq1_We{>S>sL9Fgv@?KLM4)XiW2R3c|4{VT-Dg#H95DMLotbS
zu2z0)LG+rvn!4<kQM*sOqPaq`lzHL5Q2+^a7JY2;Oa}@6d*SQ;vn&Alr`ZR)IDYwx
zUV^Y+!4(jP_=9_R&6THem&=~raQ6KKzdPiicjZ3;f^8bguy#M-A+lP+7aVpKqP0J;
z^0OaEU$Xp)uqaFPLeBjyzw>11!FQ1UO3~94!!l0<e%Q&7IigOK(EMSY@?<WErCPEo
zXHyrO`1IK=!BBFa%pMq=T03)^Z1Xq$dUrirY**#NUc)9^?)lF^shq4VynfcF;&#gZ
zZU<{nC7o3KL$m;#U^^O^cNa?t_t-=`H@}Vb;|QzQ-o&`S{r4e}zkdSC<GZp@TvN9R
z_fenZ&ph9OAa@pA8-E=w^D{*qIkUDL&DjlPCU=+d9+iJKJq$~p;dFZ@om3|`uG<ym
zN2~wz6((P^ydq^ZbDv^l4&NM@rixkkc<s57B0NKtzM;R-Rw3=R?e&`<<O-L*UDDc1
zfmDvY%_S0_1`ufoXA1jq<w`8mx@6)Oksh-FWP74&RP%gnBE7q_R-7*M8GtkZPv53e
zy0CaFS<&(4+eP<<Pd7Rx)R>8-8ar7Sxdx*x^_ylyq53RSyfln8UBqF{{T&)nYN<w@
z=P8xLiRbb=KRuT<V6}vt8duuz@-3y-h}XUG)_(+p7!;xt-tAqfg|C~xkK8xzJfWk*
z#s#+Y?;hOaWcsNFop8FqUmFs5X$nTxve0C}@_1$GgBF;3ARR)(F63mt{=7d6W5mYc
zkFYNO60VEIgN@gF*VEsbPU*`YP5j)1fUnc$QoQi3rDi7VpE!VBP<faMo$#WcB-hUm
z)aiLFa+A6)czN%+D%Ue4xlTH!1K)a-P!l%9&shFU=~ZtR0vffnp6)fSQF4j-#(dgM
zZRvLRGhovU*3J0%NA@u^75|SC_4~2ozF5*=LTW1*pjr!fFJ%WUsU~8<S!>h$Bz~=2
zV#oQ5!t2uU172$Pk9_;=E9<YkHpW=~RPDmP7@RP;`%Ut&?qe#tvXttv1|OGojVm%y
zI5!P?&zj#~QBhSVQDsL1hOzhl7*?Akl#;7jqZ#Y6Q0tBnFo&7YntTs4B~>H`2vvL2
zPq5Pm6LJuP2rt?h$`OBmK6y1E7Q@850|f*{PIjYZ^uaq$_jmIDHe2?ST3d`1=|M2z
z^Nxd}{uN4JG55ze5D^()60|r|2DsaK7wrJ3>ug<o9r{Q8emkIydsI+{b<A<no`Gcc
zo1eb(wp9Nx=#8tdELMg%Otq@kEuSO6&71X^Fn8$gGjzHOG)SVA>K&8Zt+gz|x{_p4
zaxw`vPixDOjuWVMBZ(^Ve>-m7Irg$}`W`sq{GEPFZ&K;PpMG^1h^EQd_b50e=2|=Z
zqH@CeflxYDR?p(w=AQbwKeF&-Ps93c(NAJ24k9U!+yP2;>p5OnmMK&3nTTZaA~7Rb
z|KJx2>yfX82gLxws~KMF`!Z;8+J;x;$Y)0^gT?^uk+=q~=vr{y$JusElNzB@zU<sE
z%=>m%66%ng$M{D<tD!cIw2JzOxtDGfyr;hzA<nYvMxr5@DGnpP7s{4@vtBt5_>Aq{
z55UMj2sQ6nrL-ESr1-uPt<e<r2-GYa>L-ys)GQZb-D^^^Q5q!ou-sp9Ul60%e23jM
z7btDZvw$05YJj@2RPY0c8WaGHOk@v47#>Jc6i1BXbKgSnI(=TUXLz{hZ~ov7B7!X<
z>_4#wO^@oB+@U(03II`xH6nsTBI;A7;F<Y2JRAt1?|Zo36%z)b!%FjU$S<ET7eAAy
zOI-_vT70WH{9n`B-JE2=0~O9b4{e)yY%i1^Cpbt%QP3IyoIIX27uQq}=LkCcsN{mH
zVh1R8T4p#p(B@oY3VfzO<*q>GrNB9gz=sgBd@`I;pxRb2faCv4H(6K~z>i>JzT`QH
zSsfr6ge-HM6lz;h5&Y+wAfoO=EF-riPZ$bw6gbaQoDPgQ@0g-DLC`!*uSzhBi+J$2
zhI|Vvv<9KeQh|g7>K?(2KeF7Sp)UM8V>8YhRE9hXRCWqfm<mm#uqC3*F1|lJ_z7z&
zkV};v-+%!!Xy}@7FtHGH^d=;K(l5?`(Q>q_fa5|U@a+XSnoi~~$M!4wvnoHoHA+7#
z0Ly-|XkCZkaj47a&a4Xg2obiiU&n?Bei9RYq=Y~(2I0PQ7W3=iGqH3+9M#id&so`r
z7AL%ohQ`dXGu)tLQZM$@v==wuG+@Gk6ftRwb``^H=g*pmcO8Y{o@=~KXC~|dJ)g7V
z!gl(WiHf6N8<S=N35gkhSm#5%VyXmJdeWCC46lQ~$pjNiLvgaJ`s`p!k9X9t3!=hJ
z1x#lxP?PdNF~Dnd0oM))Q1I@%xU*o6YR#`6u=bJ+9879KT%YtkZ3O>q<z?hZ?HDV1
z>=vtiaEhgj7C_;RK_Y7ag4)Y(eb?4YiZZQZaY9j+GY?3o$3CBvoDJifLBb5OfGwu1
z3RO@bl=-ue1YwVr{Y6)c!<1AwYZp;>v}&u9mJ)`Y3}Jsy|9MVNR5rEo6%*~ggs22z
z^e+j?CM7-S6naFlgyd_&o-<lzDrmec+^AzQw(Eu{#rb&{OKRB2+1X(h&Z*acy18(C
zU)ayxvXazvYs!>31@Wcg8K(_O)d(hxsSAW2J2m0cn9wI2fhiT@7P`DQWoB_k0~F_J
zyitrU^=oy~8<9rvktoi<*9w^BJ7aM+qo?p1k)nR^AN!;W2kDi$b)KeT&sTj(%r&Ba
zgtKq2mWksw%$|y6nY}tUvfC-+YM9^!5ig|Id>;!&&zIsJ`s45u6yGR?sR-LyMr}UI
zu;7Z<40(5~J=OLtnhJOcTnyS_8senK;T93&FLdu*MqaBw=2_?_h)G=c9aY|gS^2k*
z>JN4&H2d3uC+zIo8|<v-X8-$UiO`kh9T#7FX5*rV>e!fSfoQOmP`c@w2(@5#!9Ic|
zF0rF^!feWkOFE3#gMwk<@2i)%e+^SH2R|zQ@i_v|b_muv@KGt^781q0^L14gC($R%
z2no0#PmxftDW3ODj$&|f(Lni!9|-4HFzpDy$zMcOWbZeU-frx=Msf)b_N;I@bcecg
z)^RA<gm0e=h=&gkPnyKAsBo_E)qNiU<a#~Ed`nyr@?(B31Bkr}?VZH=leWz9@9o$<
z+YQrdP2}oe3$ezpRd73%T|(oRhb907Vu!ymWstYbwxR;YV8^HIjsnVKi6dYjn-caH
z8cr_CmjW+1KJdG8UQM-Ig^Li}UtdB3l1gY_%8<#EX)WYkzs||a`-Jqf?-FREM=nc)
z7VD$G+2mr&b7ODHCwXc~QB`B8EU>o^pR!Y)8%IduXm^3Bp-GhZ^D;!U7a9KVY$kBz
zEke)h&cZ7m+6lLrRlY|u>(2$!`PQ&;=ZE&=2^{h)M<1Sbak4qzdtN~Z!gs#;cT6#_
zf4DmFevb*YYzQBA50I?m&9Prj$gx)O(xm}EdUhYWVGfIJwDJjWlQ7&K%iPOFePi0V
zJLwBZT}}{Dxn>^!#2c-elRnPz9dk@Eq+4$%`GJ^f>)qDx%ewc>p@)>|rlN%XbOBY4
zy}nUymb_<Q=6>*id=JBfTx=;vjJ;gvM~+HCL_@hh+Ka%cBhSWSNAh;Y9+4+c$j}Mq
zP9f(x=e^Tj?0+P5VL*D`eK68DGl0UWM~oXcbk3p@3->$zMUvyM9W2kZ0Zl1VLtM?-
zVuB-J4Pw^@{mX_wPO2_%m~(gcp$ad<PkcdJ8XK^)bsaI5-<%q%|6upP2S`5J;mFh{
z`sAZ6(^rg1DLsJ`c<z^TlS5Yns4vmV=IAsP(4i+^lnms(Z7(6Smrw6}oO^QUx{rbd
zwLhhtV!;L^BL+*39XA?wmwk@ea}IB&gKWGEtiC$_22d5n>?Qdv1s|`kqm`Brabm_Y
zVu+U3EeneY;BQT&HqIRik|AQn>g@9-r5yZ@aB9<RK#DfAJ40MOr(f0SlN+nm)c`4)
zAFGJ940wbZY#|F|*hV~S1_UqFT2hijCm?EEr#G#?@IM;YY|*McR|s!z%u(}Cxx8QU
z;?37duOx$?2ME5jMSMSV1pYMkdx9J<F0jyE{L4Q7FvoosjsN%X--qf2=iIhK(CPD?
z)|5bQ;QLQ+_*;)pj6|Ix{I}HJMU<b_7)5wC83lNbs{mW>c00t_oIXCM9fYRjf25yq
zK5wn)5Y+f6(SoE>->SIFQv{PE=L<1<anqr3qHi;rw#WvfNUwpur)E^H7m2=ObYsx=
zm5WOW{5H|h)M)BH!xo_I(a%p`KR?ScCMkj*oMwt%iVR2(Wk>bJ)D~K`5(bU|uUp>Z
z3ekI%C0?6&_#)V3rS*-)JN56XO?@UNi5>R&8s*nc8^OcBoJbdk#ayKWjbAvc51;<{
zW*#=Sn(b834tQ1*lk;sFR^}l}Ei0aA?}f#{12-?OKGyf3*V507yJ0|X{(a+vcYkIo
zh#Xpe!Ji8F@cz?q3b1;ghyKDkTP4@FkK|A3Sr{G2b}thMKnMxpoW)Gu0ni$NxNe%-
z%OT5yg%~cuFL;}j1h11Y%gz0wp<#Bi#oq0WLP(i~r6Ugf5r+%LKqe>p;NAvNXHaLS
zEe%Y;VubBGFYUN=CeC)?PP|3Am%6MgeX#Z1jqr~AelR^`R$Et=A2uyrz8z3lminba
zl92bJW)rY7Ip($^PrPKqV$((o9~w$bbWi0J*cZtbp%4_p84$7dQs&JRqBxlQ0&i;C
z_4Y|qk7Y)ibdxr&sC0)JO2R}@+k5OPjf_GwH|ZJP7Ivms+PKSNL11-@I96;f`5fns
z9WqqxzC4$5df8@6{nQ(lBM0lI94*a(ja|^*#Y>G_Y&e#_7x)dkzM=ED+oraGUVjqX
zSW`;BJX#eQ3TWuoYKR_q$YKt|m=mYPuBz1--tIjhF|WNGCXunZB?J4i+%m|KlcOpP
z#EOi;h9+txu2A4qO3gj=@B^TR&Y2ZVs#X9jPFXr;i)a}d-;ZL*(3zpiOgt=8BXHsj
ziAI|!(xlx#0P-itgj!863y~@FD4GAWx32ogOlJI>gEqI&_|G?!+T$iZ$-!FPJU(zs
zcywXiV!Qd4BIk@!S`K2R_X92ShgXo;=!$6di3tU{CbfSY(Y^1v<LqXcW(R^>Wkqy|
zU)j$n>R%7-4}4x95gdK_6g(m7ZvS|c&XlXClR7as6s%kWTB8X~`gb!Z%rirOBOJX6
zvtDZ-PV4+Uq(?aOQQBUTHULo}IBn8s{p9-yB@*Yv+<j+Q;5GYQk=y}C^>EaQkM2$X
zb%`CrS4;y?=|20=tz_9&QU8ObQ61qh@l5aGv3rCj3HZfm8bjT0KAL=e(O)C_aT5!9
z|NU)xVcI+y)`j;E*=)k}%g}B1z3Z6Xqaa|K7R`FOP{`8L&^y@oBDm!8S*%A?bTxkc
zml(}=Zl;YPgH&W}?UGt9@Pf)-?@Iiz{+8iL&dzUpadVmb&4($&&Yb>;#xF%m2@j?B
zz^{hTRcbN^&4p5(GQA=|9Ed*AFae%Qz9<T}90?5)=6J`Jq!GNIn3o^quW1C#v|%)q
z9?+{E)M<ThJGk8&yXxVF>USqvWevuz8GUE3*Qg8Ho@{x!w##TFy@ZKwi0$>qYeQrM
zimHEb*!(4lZYs?$qo8>^|1dc*Q1`c=xRxPOlUu(taiZ*2n1;LFerh%E?W!PS*JyA4
zc0kfk10WeWMjti0Fzw2{fz!r#02fXA(XOK0jJYGetK!zH%ujTA&m38uH*8#oec$8N
zOu>6H$Gx=|3;v>f;GCEUEc8&@3evBI-mPxIU9~)&ld=J6WO7LMaS{iAF&@yuswP0R
z*zm)TM6!cKlT$e~TRLW<6t#KkY#j|fdU|;O!&N`z<@3=({TD=Gm#EpqU)6~zI<i)i
z5*PpI>Ga^@W)k_KFmE_4!i`SwIJ(mmnCk~`thW;`pU;C?<VAqL0?ilL)rRd^?yyLQ
zK4sw*Sk!9y7S?wjAq~LAVg6_UYU_*$px0g(|CbRTBLZX(Wv2_`*c19Yh5oky;>ezD
z`KESfz)=Mx-T9Y8ZA)UP7A6C&8ncc?^R+a7X|5}(<;Tx*{<n@Swxsa$v1X+2tB-<k
zz1IP6&$TT7CNgsrbLnziBxg$cxf!ksE}*Iymes_oQg2d5_rx8$l+<~z+Zbm2m})El
zHPpQ2^*-kQzWbXBLphl5vd;T*RwbqG;bZ0t<_LU_R};9YZ`Bbs4z$LaagJ&zr#)G1
zTPKk?E8qioK(&@XadnMKQh7?Z9fO?qx<baqfM&-}jniqvT`KcpqeS6RO)sFk)*`K9
zmaH<>=`iBgP>DFd4KJqFYUsnLD%)x?J}4B*Qi!F(N&Eqz@nN=&YmHoG;>%QmJ8URh
znj7P=xVu}fld?wr+>_waJb^Km<+0+iuNosFxpY7E1BKYhthqFFen}eR#yTk)Di`5?
z5E~FTg1+#>@1P^*x5jjyTvYbDbpS1x!Z03qZS9%-F?Z*5=uL3gY2>~vm|#wzqIfYs
z0G)4~8LUc};MqyqPijN#HlWspMBYJ|Tyy~AvuWnud$2e==NsyUm$%oz@17fBGcQ8A
zD2Rz(5p$jdZNa6$AIY24j8ekO9I&>mUC)~Otj(KuNuQD)5h`wr&6tf6Bnu1;`_2f2
z!MdF;`1TsFftBRS7(mN}DZW|u0ubCMwJpRq_bI6pzoCS;R<!`19>;t&gec*0LDb&J
zBH3&RdBrMnWY_Grh`|@!e3+y`7MA#}+Qw`;>w+^L!^rS020^WZuRd)cbD19f{aoMZ
zI$&lhbViRqvVXfJ(CQo6+2KT-pPQ@lDd{t<E`?KGRmdp`P_!|NrFvnc_<Cpjj|^Rm
z-B~3Pzzwp??58UpELyU3Pd^4OmQ*E4_b}WTxeT*=4)wSDZL%;(NV2H;8yb{KS<=E=
zoT<5zAWH}4vzs*Ld~c9|c-c*cpJ9CuC3X21@3G`qY57s5cGdU>^`h<V&`tICfr+)L
zuCRo|5z8FMT;vD9Aq&P$N;Io8%5+Omm6)0J?m9z-R(XhjjtOHKX}jT?nfb%$kSGUS
zcyLWz=RGYX=g(KMZj3>+QcsR4^N>y&xLKxcQ2S18-A+a~s{!sfn0ykITSJEXbn(-%
zPT)4^x)Q$yg);wGv(>Qk)~)7$c^Vals9?;}(n3*5!lRQ)bx56KUuJ01l&IhLXh%S?
z2dy2ih&>Hj=v3#G*bDU6f3kU#RG30_9DLzN#nuFL37%P;WW7Y-0{jFgC#sMD*<g8D
z#dxc5-bbR?O`>~4{MZnDl{_<2f;|@$4UEu=*9i-chGn7U!*3WT*}!Rhm{b8B4ZVEo
z8VZB4;`7kMMNhl5Q?k&pnvV=YnX_Yg2{vk-yeO2dB>g6Yf=~{!0KoJMVnKTq%?%bG
z*9E8MpYK(icBxW4V@{Y&BtaSo``4yoboQ9L&Ei6NPF3{AgD{5w%7|5QA3a#xG`o9U
zoHNB$dMv~J*Aif5Zy-LzHD`9b^^U?Q2h2vCb!qQpTCC-b%TeR3=%ZGein~$0z#LP;
zIMw>Al-q>A1n(^^l;!N~GdBbrL>7gZukM?Ox=dmrV=?D+%)|?14ckt!^h03_33HBf
zfaK<pU-@sn_jIS=&F>hSB;;}mD(TSsJt}T+WKjcV4}!`*<Cde|f%q3dKck6gLA$Y{
z5jOQeCw!x4B88ik7JNIP?N_Jdf3dsn1TA+AAHlw|u720Vp4M<UBZByC2CgW1qgTg|
z=RA-Mw1BB!n$JYxr(An1ma=_{WsH#<8dM8#U-`@4kZQi@wGktt5>hUc_&S_%>9BRo
z_p8a^CR{b|p&B&eFpr{P+<=wJXw2{DXPaK=S+P8~qfXLHrJFHwYayzU&tjt%^>~k!
z0w3?48QLGwRhARgRVun=;23WlAvzjib^x^%2q$akf2sO-a`@rveHV^QO_`Av<#?O(
z7_lIsBdzxG$M-y>&hVo;P%cK7%zG5HM6*24aG+-tz)=UM5RGgyT_DjgL`QJdF%X0G
z3@_V;*e+wtx@=wIzGPGoEVx?kqxZc;e;6@;FI~N_*qFlhD_oY0gZMSg$Hww6AH93d
z^Dbjj0~8Qb&r-MR>sij`5HfYQd{A36q{U@4UIXvjBBVKh_7f(6B9;ToK|+Z{$zHQn
zE^3Qi6v-kt89sc_bX)q6BWKdcwDvW-^fL2Q@~a}q6a^4Fn&u=Zlk5AB0`Q(bL-~34
znC05={KL>nvnqx(rMZ`t$(zlB;yB2|#u4VTF>*)hBwumwY#Io(a)l2ke4}6i3Ah7B
zB%x_ctSNMnFa<2WLQv0xslz3G00wc0m*_BlY5y_a6Wupj9D*&{X_(De8I^xIru@!1
zAm1taHnYFFx?3>S>;T?%)jUI)eGBA5LGg8-WzVQ}$g&~^Q(zYue5jPz@`twQ9a(<^
zdWnB@K^)si!j;9sL_z(LU4iU&d22EUuGUpK@;#s;%D+$+gLBUgfL27e@2nL}%Gyj=
zh{@b(A|}@p209I5SWGPeD;u~sqdljqe=)Dcv#_h$sEaB)I<Qzi{xYjoa^`AeoiK91
zlPJUx!r<49T*2T2HxlzJY1iCx&C{ae1y5}5Fh-my8^fQ6Sh452QkAK2PNUq{#YZ!F
z7D{2H=qR(RHee-QwH#q5k>#x0@n#s1{*)!_!@m1=1%7zy*1zWzO9nXsuPlh(`u*3Z
zLOlP-@yz?pSUDIkz=<jM?O$3(S66x$3QA&{k!;5@Yl$30y<9DA{s>d%hI(qk(mbix
z8NO51LB*Fb)6K&>E^f@G82|ms1r}<qxqkj(Zf_qBt}r99wb@Qc;xwW6c#syHu?@QI
z_VmP747Oe}LM;|2$REg-x<5x0qi4{asHlsWVS`>sLZ`H@a$nIUnV=lUZ~gOnRKj8i
z9H0a=S~wF71cDhhdqf!plgDm!t3kUE>5I>dmO)>AZ1uG>R}pQ_Nn$m>)^TJ`1Ddh|
z&yEW!=E}>cPcj?yXDKjn0_0=~ZLUOypcMk!QTqZ9{ODne92uWnect5!<cuix)vB<(
z-Fs6tbf>pMj*-cnTc_68TzJG?49m=Y-gy{mv{W9`k=tSKF5vyqi2@@;PQjnV?L6>N
zQjZ!j4k!_n!;Se*ZXl8&33c^D0d`RN^qeS$R#^=!$Fa`wGhHua!M6kvD~nNggK&8J
zZzqQ!N7X=_ZcIiXl>mOdpQz4ne~(88!8~{o_Y5PDN%(#@x`e)=!OM)q#^7vNrhyio
z?QHy__r2SBq}=v@m<a1;f#o);8F!S`7c9RHB9BRBldN?W3rKnN>BVa}<e0yzcx(9_
z5WfFW)TSC=&4V_i5mS*mtE4(6--e6+ZanN?k&Q`L=MPR4yGNf`$Bbh<(B?;fZr)qV
z1%Qj7({uzYq3D(fIOT7ItrK;~a7q3~&fRqJ=<J>IZ|ICUQw=5H!?`ttoE2_M{rAlu
z3Ilh}uFurE>|5vN!i%2oh-~4V0w8T)3XJ;zxhzbv`^#dKHlR>=$eddxgnRtEfpXcM
zF#r=|a7X;L=t(_%D)|@J>8D#AD0QLSkZuI>pByFXUvA>9|5{E8wia!jPe}>U96N>^
z1BY&6z_ai%1NSQR<Fv{+YX>_B!VV%Ti2{l(Y=>!CaxJy+N4hV`v&8#2-fCyJ*BwiR
z=l|LnEnAHqalTjl$yg*2C|=ihRL~bh|Jxh6$HF*%9--}%_gW6Qquf%wvW{Fb4oH3M
zZOb+Q9O9wly>II7<l#0%u{dLGy|^!Eljs^!l8Q(g2lT0vtdsYz|9j8a7=h57`^B9f
zR$n95+d`j2v(22c@G*IByX{g-*MtI6HQj#PFJBgkHL^IJctY<-{LUyLwLd?lz9EVH
zZ&Phi+7*cS3h$*2qlq&pnP_(>;0uNN?!TWn7<^jfiU^B6?YX+K)TGe<Zz+@Wjk*wB
zCKoEAnr2ih{OB?C#M5<t(GRNq^G(A_`0NE<z?5QF;xDoGwHNix?OS|K67h%2SF;u?
zqrS2y?H3*mY7z6R(Ux^h6W$B%DQB^v`)Q~(aAKx1=F%>GyU^;22kHSiI2{?d4BhMU
z{2&J=EAv&gn-fB9QQ;z8C2v!`KSN81;&p1=0Hx_Yc6C|*4u@Pmayv+jS2i!tBhtMt
z^BJUvzL5LaSVW@G(@lJ>zVlPr|6(r=#U4`!AR`XuQ(h-1p@li{B5I3(4FmC^)`(ab
z$Dk-32dUhI_x<V0;>kS|2l*95drz29{vSKY1S2+o&`R?=NrGaw1EytzI)#ZA>u%zG
z%jOj-n?0obVr96EhW27wRZ-pm;}FM3Te+U&81)EIi1O-Bi0j;SPKDf%-#0bsVxl=V
z42#<o3yI%1>ro5aKzMR=@3ivgK<@u~(etW!!sG{4Ya<>9^{5TO3veg0OkYq;5E+7}
zeDg+SP14{a2aAPRQ=cBv?D6O)Bk0|XlZd3&IV#uB#oOSz#;svhCEBT>(K}c*m^~yw
z51^1I-vrpgn(<Qp<HCsaO(Dk#A9Y*>t6R5!P{b4a3}6QBkNww#9iiSbMJ}y8E*%<H
zdmB2aOU@(b`egoP>EPnMsYfL)(STJ$d8bV(+xEM=S*RZ(*0bH(1Eqr*_B+LaxFC!m
zS4J^Vvp$3f;i&}tO9Y36@5=qMr^?^RNCDaYI=5NqI(c>al1uLkJ~VsTG#NoSe+2|D
zz*i0}?l~)7*(nIq1o<%b<bznJgtlNk&$+6w)4Y?vA8%}=*9}N*v@DCl3kXtKt@7+}
z;7NNg+(Gc|^Ox#i_xLgg)IqoEMojc!ZCcphVB8EfP==56?!SdN*<%+|;r0tYU;(|)
zueDKk%!ukX<05K4;Kg8w>`jntx-{xn?<-gfw@GS~iO?QCBYXqK)P_9)_eHEML2(=D
zh9K0~fz~5TJgIJpOEGlnAqQd>I)?v>xPJFbD=6CPKH}-2&$sRgA2fQfbnMamP$u&{
zKp_Nm&ZAhVP3)3q4;v^L6opPL8G&q4{i77oA9g#eKUjMGJK5sZe(_V>hMj=+cgLlS
zoEZ!GhEFTnn5PZ<ADV*8VE@ULG_^L2P<kYl)||(W*T(-tVzgELFD(A|xp**kp?MBw
zsQDB6od%7F+dVuCk3DEewlcqf#PZ_!Pb?=TUL92Zt^(Rfv<=SG=gA>bgNADnwABAN
zWX>W+`u7(~o){pQR5OI&x!#xBaH6!{+j`mi`g#-qs|0I>(Z4Uzvkq_YgyFVd#TL;d
zN2@8-9NxfFs*4h&DPv+bqF7pB-jv?zvNS{7ZN1!55aW1aaZ}0XZ&F{iw`W=C&%&EF
z`zKk35)C5FT)MF?(Ci@p6^!NtFb#Ggy<$EJ<?+7%5!xK1aimt!V@}|7VP@tlZ1_1=
z(5|wb|0FwSa&puwU=-_T`(f^bj`5x!agcoI`IZ0#WTMPgZ*Od<%cKLK#mC5k|0j`h
z${=J}rq_+mS)?4C-xnE^s;{hVk_`nas%j9SQ(SnHm%b7|RGumGEIH9;K=&!)#;b-~
zV^rgIvE}>+rmauwI0ROfh3M~ve_UA;Ku>iCK(5+<k4n31GS%b&Hl=EuLqLpklVd^u
z4N$@=G%OJR$r;x(0%E+;Ouajdz<-a$qLXhnW5(tT4&Hyl2|<&TzbcQU@Zli}7wePq
z(RnmMEaEK)C=_?{puTKXOq;5l^W@?NJ()h>u}W@8uKUBPy2afjTyo`hlwu|oZrmrI
z=@W5hjas847CJ08x+F&^qZvVHc_|W`{q(SN(h|&KT5NLYU*Ko#k0w*^!<j~Ol?I2r
zJ+){Ye2x+w!8?xKt3ha?`o=>6-VOUtD%1hMi>}W1DC)5FbLIbQ@7kZ5y0);5vCU8`
zVHBc84PdX-@lvKz1*H<IqHU?52ooNSK`Ek03<W}gn4H81R&2u*YpQ}kLBYpG-Vxy@
z!QP9Jh7c5K3<!}T1w#_Rkau!U_o1{6oqPX-=9e>PCTC~w_4v+Od#`+JKie|eNO$~1
zzam!O%=L=hzO%NXI>c~;CSNRg(i*NLH^-?67%4u{H+D<bGD{7{Pr6euH}#TdiG;jM
z-wKyB;I2M62pj20_CU{`{Cs`>@+H{xS*=8XBz4$T{;ln1M*ZhUw4RJ0$&0s51-xg`
z{o}n+KH&wLRngJ7lDDv33%O~&QT+)W??1`rJk02(0{ErBa;a$4_iUS*5=r2?G?l!Q
z0b=9y6K?5gRg8N+UHm0l_jM+?7nJ)cU#FUX5Ja9#lLoR!0tJJ5mT%)eE?>A!-0z3Y
zF62~xfSRwsC8(j|mkHgwH6GbMp`fWJ*EOw}3!4j(oCIT?zQ9_84}x(QUo%R5Qnbnq
zF$Y@SSXN@ZP<;Mar7*P;AJmO|yxLEan>8`$!ri(fG^TTZ@b2txG3ct3u=eS@^&mSK
z)+{8NLxb;683*We>6d7^r{EmCT9mqxf>xh}0wtmreKw3EOD-b_%4AjA;VAUTQ0jKQ
zfIn(}d<T3d-50ZIzpX1Jq<7QDm4|5$)@C+*=lQVz066^uz4eZGw#IEwyIYhT9`^fq
zzYWVgx1f0P=B?YQyGRKYGi_p9TjOh6$3I}&dQye>pjUCx<)h0N9~5r$t|Q`N5nYQI
zke1#}-l#D<B#>Q_nS6AeTntZ$yU_@vSI2uqmb4g(lh_Ug<mie-mFctq*rz<tQn3E^
zMWIDTKNemj(hi7dQ1G6}^;q89827`QD;AOm-@zI^xv1YZqNF(i#R*%K_uzlR{27ZJ
zVjGLamSR715i~HfkJ07tc;vsT^GZh`O98gl&{y6pwgGxb4kxMaAq(UOg1eTz<C%tf
ze9&Pmh}(MB?2-Fpc{oh&4|{Y65sJz{2?%EviOe1T-KZrBXz|5hY(3afWF}m0dck6_
z<}YE$Yl;9(6KD|s49Pys)N@js@j>~sQd*KKjPkA{<Lo0Oc%;kvX%?Vn)#iM)N(Xvc
z*_Dn-2w4TGo)cm=G=Fk<>3`Mzb90@&7(KX!?3qfxo|ipmLbB+@=o$h>f&<pOvhv=)
z$m_7iH@T6JwKsrSUq0y=njD;NsWlhWUSz3EEUHa89p;E~k~6S{&)izHf`Rz*(N|n^
z+P`6)QRjHqsc^?T*`&HsG<VrW;F2uoNk=?~^G^1E4v+Ja4F7f(AOoMSeonIght!mG
zMXZ|;9W(kPFa7mb2ez0L8&9`?1X%XtBh(KROajK!C$Y1K=y*t!!8S3{MmsT)RbC3~
zNl2y<Wir{l!UR^0@MvgHlrJ=nhE4~AHeaSjeEJjano_a#mN+jPG$ZK-MWkQI0BiGD
zw`ep$l+1L!!FS;%eHpqao;O_n(|UIdlgn9rpM>a1a@YhRcNT45+3|E~*m2M)qHD+`
zG8v$uafGVhNDZ*@Ce1vTrnFP>luw-`)IQ|p)t&(=kB=I`^QK0!P#x2_>;0eI0;+d-
zpr<eyz*rhqmRZp)aovtk<@HpL_nKGqQzr8iN(Ec1=+>del`Hv$xal#Yb?;y7{N9>G
zSJACnQJ(RV{P>}&%NC|+jCn}YjGj`&^1w3qBl}b!32VQCdkUz-^^Q1QW|^&NLsT_>
zuYi}!p9UouuO9-vsC7q@*2XrfAAgTnYmv*Q5m2`W#gigzTm}m4PM#HoMG1|SCyDlF
zN;H<Zrs(?)`1us#c0E*zNGf1d>EU>DHDs;M<%|IHwrQ3EZa@ua$y2&jy|?f(h0zzY
z+wx=9SEgfekDlMBZl3;-1Z!1r&ER^oo`_5nFT^RAceZ}-Y$XP_>t0f8Y~O{`3xK?$
zj-yjL>OCb>*=71q?k_E!I;9jNHt`)BN+H>-BMqL@I9}d}WZkTJ12#)YAt+h>n4CQp
ztlHe@DXOf}o`QhS$aH`Sw?tQ|7CC#!oNs4;-A*}3cyT~c6TJ`xJ&l|B=~zf$8=!hk
zAp&~7f&7+2mL@5MygzqT6Abe46uw>K&H~J<r4;m?1^%sdP3xxb@MqU=C6$d`^&wHM
z*Qi^VK%2oJl3Bf}O-j)fQ57lk<og>`+6sW`jPAO;SnqTYG@ciNz7~fdeIOKItD;-0
zlw?*PStim*MQQSrZ}Kh({LdwCzx7~eD|5w6oWX9Lhfs4%b7@-U`llM_fc}q5-9NCR
z&%(9Q0rMf-`K}}fg|(mK$`?4n-+Z|NZ{zZ5-f`Oh_e1Sxca>xOd`lO(aOOYx=iL(z
w0MAN~qP~d#A(z?t{5=t8heTBfBm!vaM^0`2(D2Rj1>o~hSj3Lo+Y(Rw3dzJUQ2+n{

literal 0
HcmV?d00001

diff --git a/_static/plot_directive.css b/_static/plot_directive.css
new file mode 100644
index 000000000..d45593c93
--- /dev/null
+++ b/_static/plot_directive.css
@@ -0,0 +1,16 @@
+/*
+ * plot_directive.css
+ * ~~~~~~~~~~~~
+ *
+ * Stylesheet controlling images created using the `plot` directive within
+ * Sphinx.
+ *
+ * :copyright: Copyright 2020-* by the Matplotlib development team.
+ * :license: Matplotlib, see LICENSE for details.
+ *
+ */
+
+img.plot-directive {
+    border: 0;
+    max-width: 100%;
+}
diff --git a/_static/plus.png b/_static/plus.png
new file mode 100644
index 0000000000000000000000000000000000000000..7107cec93a979b9a5f64843235a16651d563ce2d
GIT binary patch
literal 90
zcmeAS@N?(olHy`uVBq!ia0vp^+#t*WBp7;*Yy1LIik>cxAr*|t7R?Mi>2?kWtu>-2
m3q%Vub%g%s<8sJhVPMczOq}xhg9DJoz~JfX=d#Wzp$Pyb1r*Kz

literal 0
HcmV?d00001

diff --git a/_static/ppl.png b/_static/ppl.png
new file mode 100644
index 0000000000000000000000000000000000000000..a40a8289b62a8d3179964306a598c405424f1ba3
GIT binary patch
literal 25483
zcmaI7Wmp_-vn`AT3&Gvp3GN!)g1bAxgAWeD-Q7ZPcXx*&xVyW%!x^5l_qCsQpX>X6
z4ATtLcUM=fTC1x1CR9OA;tM=3JQx_*7b!_mB``2ZTF|jCEHvonPC@1r7#M<pg@}lP
zl!yqif`grzg|#Udm}F?8I*f+$FlLr^T#V3<92vNFI8t`s1Z1rPf^Kn1hyYY$;*f=r
zKfkQ7FsgD&Ny7+`d5z2!PrSZ~ifLkf=1^J7^R0LW-tRr{jrqKDUuU!KkF9vXfYmGG
z$5Lad!Vt@A5J!E6`yKsVDnC989RC0U{Q#0DIGxdGpidM`BWS5J`Mwd1)TWYGvi@!H
zT|XQ;I(z&RSg0RYX7c!`{{u1@07ISpE0{0!?r0%qNFKJ@Z;Bvr;Wo)I*4-xQu<2bu
z_Fqbps89Bv!K7*Nqqxw(W=mrVu^d=)f-H&p<b<<?V1023jMm%QhWSFKi5vVlNJH|U
zB5sJBSl$a#1<HmuO%RPAbxg!)6k-^GyRy%M)0dqxdKD%^!5C$<np+cGhZ%-zXFIZp
z$wn9<9)X97lgL)WFN$69iy{$9HRL&=4jwZx>Epk+A$gE_FS-0Sk@1A_>HkKCDO?=l
z{4(I^`x*FK>=%R!M4vk{<JLY*Kb<v4IMpDHXU;8-5vFpt{zizYcYX(<lmlui{41xX
zXM-@2GQ*g(WKar3wO9@@4ZM+R2*I`ERxbvJaagT}q6mkPR0f;(=aB7(o*1xZ0ksqo
z#3<#3<h@;_QLN<*3Ihq63%_4w0Sa#3RhVBTFa<L{yCgE2P$C{)Tn~yPj0O%OA;gUn
zc#L6D4$wIXguz2>?HJpKQFDqLMyFpoisZH7^i@0tUWQczl)!aj9K85MRFLC|qtO`0
z_5%8cw!4%4VY?vwd=bg33jBwO@nDzS@!U5v+o-UA`eK_v2>euF5+RE-<rfA-!t`&*
zhddDb0CarIl9Av^-knnt^$CQ6${H0oNMZN_xsU?<bpVX9k*J6sG7$2_WDmivh-Et1
zc3;b;-h9T9s@HKSO-RurOoqsGZHD5oMG?q@Q2j$2n!wM(J+@IUj*v=O$GZ4~_bykn
zE)pi>0J2R|2gECx<SA5dG(gD-@LPY3X32MEtR{>Zr(VdPL`8^+_5>?nTpJY=NND%|
zS*ye&N_`BY@UK*-oueDUg*jDZys?TkUl+5judgLlc~u(?F)S*4{zmi7-E4Z+26=ej
zjX#k*^DP9uTZc0AAR6sgPcMj_V9$XaEJ&51SoKHATwWZxlWOaVtZkWG?g$Zd3%-2e
z*R-I#&a}*Q&*&I>hJO&7a%D64j`<z!GfOS}f$W3W#`L0a#kOQiZG7!#u&ML#m)lzo
zwYyazC#_D$>vyoO{hf`CjmdMJ1u3vQkFHan9bWTZ-)uxvu&?-=(tUvrh{<}e7YsSj
zu4w{H(0g6bV}m$NkainLzl5+upy7x=$p%cCcRPQDR_;b%!b<REslqP<$1p%!fOq-D
zzXA3GY_QAXC!!9tdiVFAL={=K7pOOW_Xc3!kWq!jeuM}Z{n8-43t{aM@{A=WhX@r>
zieV#1Lm@>qfo3715m6(>7{bhhUW~^7CP#)9>$3B8D)fffE@qrMsN}a+5sGDClrRo8
z+7}753Ai7WJf9fEv3|=;I{#pA!7G=L&rP08{{i!adO8SJ5OHWwkODc-rJ=!=6lQ8*
zWR0m7mS{YyK`;lEeeGNZdE76sDVh;9Z4cA*OXN~CBV4}Qd{gEE-jhl%x^ZLvQmzf@
zUHmsH0^A)GBSbn=QjAoJG&M;AS_2v)+E`Ef%y)o1IjVEWoG4NgmU75q-(ep`-<Q7j
zKIJ|aBdqFdQPRWs6DfhbAE?19LbN31RHu}g<ho?e@dC6obg9(bbf4(*Y2gz0XcFm8
zm3VTTz8|JefoW!1OWl!5#JlZS?3nJ@>`-1|IB>f8{`pp2@SuQ7y_nFP;GDn%)C58T
zwM|V;Sxx14HB9}k@OPbo&k2mwuCzW%lPcB9hQ$%;u|@Q%i#ZO`yA?of+C)pS+rk2+
z&qXrD>@%!05;M^yY^oBf4Ycyk4VL+q?MLjh>Xs(#GVBFQqV-1g4)y+)>ND=M!$)dI
z^v+RF^taNtWJis&&+I>NbJ_o}U$G_qIGJiHy3~o!v&hHE8=ltr-K=^YVTSc3H+5Qr
z?W>J|4c;kzxpsJwX?gz~aE^9*DR;JDwy?bTv?Mc6H&3_NOQ%%?X1aRH@ObBlewzEY
zqs)G7MeJAQ<jPi!M!Ea6GnjT1_cwRXcBS@qcg|aZ<LO)f8^;^<1I`0MwyrNH2u5ED
zaIM+6tR{Gctpe_0&qAh9V8mBgMZcUPY$Fun*|J-+pK!8pz%!%blymYJ&sdcipV(#C
zv@JaZ*hj;S_RWhL)GY1f4S#Oo$dnq-ugq<nc$WS0UF17Ofwt_jEX?1sQPk1B)Khkk
zA9Bp*Dds7a9CjS%wM^9^bt#65R`%m@*27hQYjmvK)~@^$>s4!Jg-fAJ_vhT_tdEV3
zwT@vpRPY4TX*gy%IITk)5SH<mzci?yD)4f)(zi;t(%lucx3+hA`+Gf0_DGJ$_QcMT
z+3^-}F%iHMSQ79MIOCJy?{jl-b>NyZ(>l+DDeg;nIXXB3otO5UeCqZNswS_+IU~Kd
zt@r;rS2%TS3OmEOq@GS4TrJJ6C7!oT-bUe#GECXzXxA>QyCS((dt`b<2M0=|NhA*1
z1Ywc|RR`F2$8X?la4%n{>{BtiRzg%l>iM|w0`>vtSKa~x?g&1A8<=$k%mnbW83YJA
zoYvXaM>`0-3%v0=EjtI__TKp3Pv6E<8va`E+3uBs6@uA!ZTce7GtSA*6${Zbkyo)*
zF-WpfGZ&E_C{}6yH2Z1uQ$Iv|=q5%vCt%`Y0wbq!QQZ1#YtSmz9tsC*8HEl7pFkVw
zi+zA@0M{QH)JTMI9gIg}Ad58iEb^B>i<6~)hz2wUWkb1fWg;R&%OkvSIeBcwR~`Xc
zFL@8Tm*;2@s7Vwn@^w-P<Q!DR6p`{JQcp4XDQzsx#wSJ9tb=-T%#tX1Yz1*j@k)(_
zchR458_jg?0gaaw+jifpvJ}`hw?zFzo6(}>mD2lCotPqHk}?|TPCrwnDUTk?n-{9*
z-`i`h)UPD-`K(#$A*}~@#=X<`QW={LT+3$vRZZpB^w!+fY==ntEquYl%jThc7o2gB
zLBQ?bJlyPYop3dE6$Q&4$la4PJP;pF*%*EzM)4<#Mo)Q8_7E*Nym6oG5$y`^bKOR?
zXKA1KRd`<QL}OLW#r8~IF-$mHSb$jh^hp6j)?FD_(?jFogVWi~9!>%4-q=CLa7ME!
zCJUXH+O_rd;%770ak$i4$FHWgV=L?X+@spu=c_L}kKaE>VE)ydxRn`VtfwfU*os^J
z+>T|d6~Kr}Z>fD!19;uC%Ws>KF_kmjHPui0nXL2Vd{cSsO`wxjU1yW}ToXAGIubZy
zSAL;g?U+9R+)C6fB~_oTEU5gd`)WIy$Bs3kJHlC>UzS*Qs)1EqVc*{Ww0qpCVxqcI
zhETMkZK_e<bkqyI7G#C@$~kAlvZT^*cx~3b$Ez*eFr%)mnrug5%dqm~q&?Me-O21-
zx@cR~Zv8w@&_h7LyY1321DKgPVOgm!!z@pnxjP~3d_KzR;*WGjbk@E*I$>Cbxs$qU
zU&wBFpYywb9EHJwIY!hZEa8#m)zONckuN>B8nQh2A!}hRi0YEpJluMCp*Wb!8<k2}
z%2(m#xd;ag{X_*VkC}ZUNWnAf!E^1Tex0}XY(9|HJc+tI<-+lD66n~yKbB#hCFs_@
z-%}l~hZU`3q+{81;@#XaTYh_SZre#@>u6Wnxn-Y!S+YM{aL0T~Rq?3H>&1WZ)Qoq6
zmux4#YTcO#I37E_Y~($=IOFR~_qe}p@9}`V-{&LdqxLF#t9psu%smpceqUJLT(fUu
zt*X7Xf7yQ4U8zWxzsmMJ)VoM{s(VQtu1(t?8;`z)y4UeyJ`=l_nH=$3ixXsjD|0u!
zKAO;f^Qpg!JU#Yi@aBIo8`gep<Xvm-usVxA^tll@f9(pMihd$;^)~Y4dFOt-uJtR{
zm(-v0s()R&VR%{g?{4e<h9dDd_7&cs>JJVMSlc2v82cwMT)u3@f#GqVnqNcXWh_u{
zXi3>9ru_uIJP+@L%b(wEZ6T}d@c6vHdDUNn^(lZ$UO*LvkNhOyDIFe5oDvS$_Z_X)
zpT3^z?xeY5-AA&2Ux6qtp?he4#i~8=G#pn1M*uYxT&5aQX5YVqeY6;`VBo<PU{IhV
zaL@}E^a2Be_zm&jpFq<7hWzhiFyD_qQuVHyfPo2tNr`?_0f3*b!lgKi*$ck;<I(=w
ziqihMu7HBNg3-?)E+!WF<4<TPZiMy<*(1$a5HU*bpC3^76k#$GA>SsOkX^TbDdsl$
z(VbBb?7kdNk9XR+pLv{R@?MW+HnS?<@?EbzuQ|$PpV_&ps;jE2pHIO!ON9!N#=wH=
zn18*|BSY+K{%$c*MfDRUMv;X2_Zu~AUQQe&$^ZIP5(*R1k4*)$LZ(Le-z)v!b%Ug%
z{@;Z(xS*yKG@HH-!&Cgn#XnBLrR9eH_eK9P!C(_vQp9w#g`6(of7we8VNtY${hxMz
z{1zxe{3S8)IIc|L|GC8sMv{+jjD`!X->+hIKJ=eCW<TrPx30fdu<h{ebIZF<r?1s5
zTUEBV=(;SL&T!pl0k%R{?R=h>C+)D>55!s_5xIUZJC5^9bnYhq+e?_>ACAbuszD6x
zqD3+A^gPyHKa%6x3^*Ine{-M-w0C&e40a91caU`5{Hza9xw7-VPCxS}dNH^kbM2^o
z*;~py_uRK#_qr!ojo1IT7c|hJSZ=Ho-ARvQVap)E#fJr+Po{Ox-RF-x-e1pM^`EAa
zf$$@(Yqu3=C$-~xN?UH@9&2Y4B1os+ugChX`maaot^9Wjx!A1DX*Ey<FY^E66?#~Z
z3+W{#OxRKVW{JBLnomp_mrvX8*H$Z9&v<CJ5|rf*J6=!iT-9`F;ow@|ZY$SyFMeU2
zxGdW=p2@kkC~JOi?)vmq!ESi`KYRiG;fn&g8q(0NBjhU1U66lm?|X^fs#*uwD<*ZX
z;*w7~?%U6uvtM?oTJ_(ql4aoQ)2?nSecA&8R!_`0wH+AV<Np=yE>e=1ARy_j_#6d_
z#wQ5Dx69~}qv1!8<Ie6HJH1?Y>GYcuj8u221RsavP&=;@E1f{GWdxRl{jARW56dt<
zEUP!K(F^U0fX#58)6DFo)7gYyEw}SJMup#-jF%T^e*Opu`OEP((K}%9cXeeyVdDSa
zB5r%BtDK#Wqcw3xmB2_l&u!GV+w1K2Hr~^^=@t-DTvg)Um*Yt(?9*ESP9ok?|H><a
zZy(tH#zR^P3N5A|=v+LIJ-Ua|{X)%;qjq{@3W{q|RGBlGj$_=pf7YMZ?p^i;-yd6F
z@7H~9wEuG{6qF%H70hb&8o7^KTc978j16nTrA=;`$>;Hrd`OPGO>XL=kbW}4JM`-8
z{$k6tjd#`XZ{Ki0`{^hyVy|46hjyhA6E7!MRfg;AkeXujlSE;)OauZ8+!s4dpgcg?
zqB|z{*heLxxJHwCx%<_%ncd9?Gw$E<YWuh^y&<#a&@LWggW5KX#8SQ6l2wdKK>sA3
zon!iP0uz&98wgqF-nXUt&P1;V61?WcD_fY2*sIC^iGGz2zL?1xcBg<~B{m>ySKhPv
zW2Y_^;G(<7l1vjcTwIu?HKzG^5iGdEo%trt*)oYhvCr^tHnOG%q1>AO?eR|3U>5de
zkF%4el;$o|<2%Zfs+3t}>*}fPYL>^Drnmb(@46<7ucqQ~<PApti|qeIN=zI?DPm`t
zW!j`npZq?|&BVYE51mc0Oqh{C<T>~~X*WpZ<4L7^dDeQqrVC2B%Ele@|4B4*h|l35
zdi%W0u30T=zynd^2KT%KVj1ZHMnhc&-RGtGui~MQgo3Y!^2z>0nbkGN6|D`8yZ8T&
z4<_mdc5Ae2ltR0lpq@GyOKA$Tz1^k6JP&2`-GgDUsa@NTbz4_K$>}tTL-!vHZw(Fd
z;vzzOg0n0XE)+J+*!RaF2>~TFl^1<AMh~@t!-mFoo5n#xj|lIv$Qv^1E1CbXou2E1
zBmgj`x2S&hP~I=wM6Ms))ymm%8M*73Q0aAQMe*q}x$|ZBQoFTOXP0)h<N0h^Q@1+V
zG~4TLaWm`XKV(Bplvut%zfW(NhXO(wx$2iuqNp09&ks*BFE5>>N(sI#lvVHNQ*|XY
z-b4YKf9uVtZ`aqHo165UKbnL8_RBw9xgY>*axi~)cx-L)KbHR^Lsb?ZSe$lJRb%`2
zYM>QB=j7$ocs=oGr~M1NAA)F&`{6D&-9(J<|AT>luGxU}fu-z@u_Vg>C12?uq{s1H
z-BfHI!jE;Fuhs*ooHfONlLO-6n_QxDu~<!U(9&h7ejrY0%m8;f$O(3#%Jy`)UtWK6
zNu<~{{x8gGqxkVjn9|)@Hdw$wQH$ESjcB**3*0xaVs%^uo*k)oy5D!cKX>x3BnyIW
zu<HdS2IKu^TK+sr;OXEjFJA5l-^jg?ceQxL`tf>DVErGJ?=lkx0eEyqTXSBO<YHLf
zyT)U#HfuK10Z6kt?tNE#=GytJyHDKrA$qF2yCNV{s@$Q*KuU8qTj}k5{|l?_HwdI1
zbFUxS6PS75w$I?%{1kUR{9lnM4+6nkyIi7{pA70BY9AzU)p0$k9&|C!bXIen4A-mU
zU4I-?S5v~K{Xk}wH>^fCABZ8P4|bo|-`{NSS6$D)dTxGp{m20siYfalZ?9N8=@0iC
z)Dp)mEtwq_m+gn40M0$um)nn2E06A$?K;myh@KSgFQa|hJk=F&Y3YJki6o(1;Qc-X
zdlil<Lm>lNm}#7AJL1ef+C#_H@3YSDg!Ix3-cNP{86Pq4N)jbjS7|V5G|MIa`MkW|
zvFy6;)hYwh)YX1Q@5?h7!G~|IZ`av^-5<35MQGzHxzp3N4@JTWM8o5)$9nfI`OvMW
z4ce|bpLMD4!)=6ocIePPs-6Xe4{+hQ(;PR^LSa+A#eLk!HVU%Kqethx2cCXx*|+}-
z)UJYW#3G&l0Oy|*|5eYN`h%RhTv%C~|8F|`FdKBvm&EVWs+yDk<^TVFGk7KjDdqo1
z9t@4%q!*Od{8`2U58is+!s<-HgR03H?ntQXfaMB(1*NblcdG7I@Mpb?U&q>hp36PN
z7a(vpTnEn|=hTg*E+PgX4RM+uZTRcDt~!&XmFpdmE3bYC03SJk#GQf;{2yvuZQjq)
zu1Vvecg;7^*>2{)UvX4@xjWW>Z<+b#?q)tSdW~PkrlPjg>gGdQH3pkPpUHh$liY4?
zYu5qNr3nJ}gB(W;4#(0LO8OMiCs0{b_`E|UxahTomVNM#T->+Jx3}P}T?D09v1{&T
ze%oNMbCbCje7_dV4DEcwAyRMK!f5S9<h3>r(|a(lpMK;4fH5*nU4XmQ#r7Gz0M$E`
zj`H*n27(1{a>Qs>hTL~s(fq=?K(~75z_9`X{4I;!OE=L=cNo4d(L)yukNN0_*x>MY
zqWW*AufLex7jU*0J61Kg9fR~narlTUnv8hQvgiUGRVe%IrjV||KfCN3Av&S1xt`8G
z&-~#Xi0ivOmWedui9o#D@{w_;<Gh}(G1&4HhPj>uR;Vd~_^3s1YwU7BXO6CJ7$QDz
zn)htjG8_m54C&KY3BqPG(<rQ)L7$hwRWDG@%BC{lm)iTDIq{Zt-&XjX?DLq+GWl}v
z^X`grG2i)iJFSokxfZUMSA9|W{#2P6N>oXv2W(y0{i?mBz+9LGN_(-z9I6ff4^`eh
zDUD=P0i>-z%{3+oC^E{CN^g9WmmsmpOl?qny@=utDC=T`Ap{t>f-1{e*66=AEtiwg
z-gUfh58L7K5o2EA+ADwf#y3(-1NXGopVS6598wzUuCvH$&wd8;IwiO5P?(qWmaVe8
zitKuwd+6W)!D=r8gi)%7^&nXf1emUu>YiGzf`T|Q^<$(w{vg9~>@eZ|Bh@T5WVPQY
zzS@V9B7A6$t83dfK`Wj9BQQRUv6EFLihD1`L1W_~^60q>Mu6!{qP@Y@yHyxij$=NZ
zZZRyXUQCh;sG=s#OXAJ`1wrn)`ImK^e$2li;-v$`KNUV7nXU8-kS&XP4<fDbUk%>u
z!<-E85Fmnm<-MMa&$`L`gI=24ybJQ|Jyw30l^#;?`P;PMtMaD5yw{Pk>02kL>{fs1
z+83Yv5Wq7~>8*EP;o4sM4B~06jKap|Q5)&@f<sa~)pgJFJ27A5*bJ`AKD1UO?+gF_
z!@%UX9r?FB-klhcB>t5pzXieP#c>|G7PT+$$DMDb2hgN+2lY-@7EXz!I{9rQj~X{S
zAFA75JPzjGM@Bf=EO)Q>IjuBjYi5&o^-zxDEv?!p@RJ2LiRY-5UTxpj8z^rMHO#Z$
zU-m~(n}Xgz_>lmA|HKIR7V3j%S$3uQ(>&YD1^f_HpPwBxfI>I7ic#@1@XBYm)$Fz(
zl~)18%=fO@&nx>aov+KC1go6xO^*(x+N&k>5`w1|^8a)o(G>@x<18+5954`ki%B=J
z$l;^H2pDt`(e5+3Ez|OX?Gg7{uFyko&*$r}yZUdYC|bAom-6?Q23v_A;S`fXwoRz>
zhK=#rfn~(3ET7egA0>L*ByxE^ZPX9N(Cb{HA4re*<{5gl7NObZ7OwbQU)&E4gRj7M
z2TIh^WNg}JP|dZ}X3#C&4I}UKG~N1<{$h%#s|I3NM6e_%teq%*p&|os&z<iclQI<^
ztH&-M4PVRh+Fe8D`Lb!}gt$TWxehmklht-yl;x+3Q9EzPd%O2LyVZ|WV(I+p%Va`1
z`Q6TadaI~0&PXgd=i7P1uZNF*=Ax&LS-#T@Jysd82Wnru?(-SFil6s=-u6+!9X@1Y
zzGbd{Za@R>li3;ioF0*{*7R2&kMnj{ACNTOZB`%j{ygXvIy8PPdjH7nw+|X53e#+6
z6gta({A>PIyHkVbTx1~{7|m28z*C+;U|*2`^8PaZPQN=ib#>68p?jEpl~8z-)Aij;
zqHN{HP$e>JI_e{wUVx%Pfn+SK&)~+yNs=;x;;;rdL?~qK8i^T7=}ZaF5`s^YG6RMH
z`X^)+FW3|s`L}h7(=UZWL_wo*4#J8Dy&=z?c-PBmq+rYDAXL5*dYmZ!jMmr?fO7Yl
z(=@+Lw?L?%pxzv%+g`+`RF393DiT0ws`u<Kpe@bp6?S9vY$?WN5;48(X3e^cKp{%;
z@BMjhoDB@R>W=n#*%m-={o^qqNXpkH__!1QtO1^xqq=;aq~UfX!hG7haI${a&}btF
zhFR3VxpCNe5y*H`on1aO`Sa#xzT>X`v;HVcBOoj5#ntED)eD<4AK__o{rz=){BL}<
z@X{bqjwndVO#uJ#^`oU05T2WdqG7Ifq=GjBNJ~S?O2QC0DV#Ujvq-)63ciRCd8h*0
zp+J49eWdA^Pz{1ZJBRr62w*gJ`&+pF%NF3$6!oLP=4p|CJnGR%6OY9)gb|@g8Y$q9
zo5utLk2mV$zekZ+90ckRmFmBWeRYVPdQah2+kW~OFL(jao;2=sd~xL~A^cEf>bwF%
zBL<kfNp*~hoz428zdl4WUA1+xsT3G@)~b{7)kk9_{9-40dw*MhUrvgOx|q2w%yPGU
zPn&NmKst9O+62#K#wO`}MazEZ6X5gk0vtSDw4KwRr0==OF%Hn_nKXTL&&S5IuG2F<
zb75&d++_9fWLL!?t_r#am)PAVyl<LZ@fBRZ2;#JsdNDsO+&v4&sQ!cC$n(CK{krJT
zxPR*`M69Otw7h!e64*#;S$xokLMw2cPIgojT~(Q78z4xw(iVLXo|;WnsSoPoZ%|$p
zv!bi^wnF~cozRzu=b&C@dC~7D0U^7(ppTF*%j)i*2uGwu8l$|!zlf2^$UX9r@g<6&
zk7)Hj*k=nuCxklUx)o$9@MDp%O%o1Vef9+#s3t8umyM$Qy>u?{N~GNURu#-|4Q|=y
ziGe*@={{R3F8$pm+T~(Gf_n5Su#bR`*yj?<hq;IBOJb^l_Uw0}=Y_$ig<uvfnvOyK
zvq^1&<tTK+b}#EE`gr30wu-EId)c)?@6xPEqg}h!7-l1*VE+&>@fEdP*!Hh<7*83@
zn2YV0zmW7Ef$Er4!UW+cHB<4mQoXRE)Lx3ej-?MagNXz@jo(@-U^V!;K?SUM3#|lL
zufO9DwLeyagw>?MH+I%B=VYbxWzqL)lpogIRl7DM*jI<^mX*nl>2w9)?YBr!1lWcO
zHVBgAZgd9rnaPB5h}Jj+&bRKEyYQD)qq0&d&yzoa>TNZpyxC)p?Zr+OQ6OS}UEfr*
zEvq6n`-eM~y(4q)wp}8rh`?*4Is!<hULUjH-4+MybQ1G-ml4QW^{Cw`T$Va2rE7kL
z+OLC%o`?B-1VsS$TqcxC_iU}L!x1*XP}4+XXrqQm?v6uKv4KnqAGtI|X6KL`|EqJc
z%=ebMU2i0j{FCqrCOAx!)Fn$3`LG8!Pvn6g_Zj<q=Dm+y-l?!6j5KkxBLb-s0#~ta
z1v3fV=027D{W;jlt)ywDz*RSkc4SzQ)|1Pc_mMSP4<(z5#>q9kGfhcSI;pP**iW3K
z@h$OkJjDl!wS4l?PwAbH=`}R^TSV@iJLnu|=Ln;#3Wl%?U2?F$FRYWl@WmrK8Koc-
zCi)5<>fLiy?cuFI44{e(L1#$?!C%;zl$&f*2D;!b?yS#AQI)7q8Id6q`qjX#ir<ur
z){HL%FdPQ-A;bDep;1G^QW(UM{oXS8tBh+C!g@$P3Ikp<EG`lcUGo@|Yto9^Odl#e
zr36Ka?#Lz~&0(zks74A6{9sl^RJ5=e2OQ4%+}8wKKCMJs^ZWoT0KZ;4%{qx=-gO(-
z0ImHBB?B~slWF>PK70~iSp2HEn^ZQJP!gHDO+cngHN~I2iNemoFk@!Vr0IC8#Ty25
z5~<Vg^Db(i36=)lXQx)aU|=K?wZ|b4p%+VuTExBPqnbKuX$c{0cNNk8fE64-hBOH&
zix|37l~VKHZA%9;M^=qvAfZ87poY2#kkzko=MpJ0J7;Mi6R#_K5UU9;{uL^;#UD!}
znr6pWymCtl^gA&IHB!&gxD&9I^zgrG;b$^`XOd&X$!p4idYbvfI~j{Nb)Q@HRY<}M
zUjmlvp)<^pO$sp{?!Mn16aIuVVnzBEw2ytp=*)-x&*Wa*(IrmZHUsesWg2{yeCsu^
zSAY>s#%mzC=Jux}3$bP%%d4`qTU^8Reg3@)B%g=gT<!>U2Q8>?%PiZ#?tn4Yjiwkq
z^#Q^GCW!#HccsLNq(;H()bWIye5GV*L(8&Kyhk!+^?;*9r{yuH2T*Nv{Iy34CKG^9
zU1O#u(4}9?^?X6at@wymDdTnYIEOQQZ54>)BMlxUSkKFGBex^(9NMmwOCJW%r`?)G
zUcpKKc~vNXL$~_tZOi&IP0=RDh%^bmBy=IH3}ciYJe6A?YlW^;jQykJ+@vaXQi6zk
z0>&T7u9`k-@=l{l1x3+KajM!Dxgf+t;iYb&NpWn}v!$}sSn@m*hw>)Hkj!Iy(5$ZU
zjqRbB5ctnf$w~A|GZP6xf=O)N4S!oa_xNH~9Ff>DyXAVT5fOQNF91!BdVrZi6FpkC
zYG921VIJNwqZi$WTRwVvm9O=F%>#>&)lR94=6-}5hHBL9X(q#U8*d#DEKq){F--Lr
zn-T#7k>xaavFvb+Tl%l^nCq0c@!}Eewkc-M{6Ho`^3yjWZsDEgaj!G_Xry(tzdY(S
zkLsQ-ZS60h;g>5NUItbB%)nmD<|1o1m-4ol$mF5Fg~V>~A69!uPhUL_l=+qigyY)i
zgv$65Wez8Cm}ua=EdF2i%eOYrL<0+g#W}#C!^9dln<VC$X~4WhTvnK(HhBD&*ki&-
ziN=#aT0T@2shh)6gjyx=oy^4UCu=D{^Eq@V<S9embogf8R;WRLkkSS3bK-G_f|K+d
zi48_H9vwb~n}l}Kvq%$`&QQe1XxH>zh@rQ=V;P@OS@n4gQ)CfBdy&L(!+mpQj;xcB
z17g5%HQ7QZFpfz%a8#rkVqzqNeN4tVV(HvS4nd+z)^slY<g!j)JgG&?C6hDpl#}_Q
zqO?|C^E3#SNb;w0b~4TnOo13ppttg{v)0?q8Pjj^7Yq5&SN`o69B5%8!z3^2B$3&p
z-r-tuyh9Pfs~8M(0dgXf&YtD+jA9$r^RVROxVMn)Y*7w28EO>ibJ+VO7gNjJ9hw2j
z<*(1;q<sY^{Z8^;o#!>hK|LCJIHg+c5e`gcyH5f7*w9{FelA66{IPMksu5O;x6`~!
zg2tL0vr;DKKd@2In3XZh3~*_3CHjk&?FR|wkJayIo}O(PEk`FX#kCec=SR3l%fBb*
z{V`&u1@t5{DLhGj*B#IgPj;aj_jn~7Fei%rf+l?^Bi^dH$LG;VHA;c@nW)J?b}HQJ
zJqD@PWZWj41@ZX$VcdHz5%~>~O0Qdj`>t%}O;d{dR*x{M8KuiQuky>CR23q#zw|;E
z<!s5AYvuSZ2gVXC74J7Mi3}SS`|IfYr7rdJS@01!DhS1r#wP@DmhhNNydpICBr&Qs
zSucS0+{?NA`;Gkb#lv@i<6B!SY{0zDu!;6yai~!&szA)ALh-4vp=Lcm=<iB2Ex@ca
zsc%wrukmbfp$%LeC3bvM=u(~t9INIGM<bdIU`=}IuY0+i2Bl7z7+_F(j<Yk}x0km$
zj~<P*8)%>u<k4pLl9@rSREiX9?YL==HybUSm;f6=Mm(bFJccumK4xL^mJ>1~o9v8`
zPB+>Z1sT&XtvOI3-$v3p>yT%S!<sK$(Pnz$h&a?Y;5&$CGaz*STg7q^xxd7Sh@^l_
zASMfdTPy<befCEZXrc}fc{v0?zG$NTlHD4=P{QI1TBCBrpr`=C9)_s1G_s=C1+vy}
z(TmZ$bv@)Uy-yOoPQpsO5}O_)$40ZT6<5~W6h{u-;OycP(PRwwX{iqzT51_M=xJj|
zIZ?8d94Vab<L}MH4M*e_tg(I_8KQOX<C0PnHEbMiDNwJxHDK%O7;+eO_Q5yZuF+1d
zicHMe(ot47?IQ01ld;sksxd2fi1fY-pyJC>3AV=~e59}`+g@%@Ffvx=P;;IVXYp!_
z{JGthCccNbSZbLh?n$g$Z^9md1(1YNV*Xu=Xroc-s5c5Il^=D(*RLIOn?B+`P0JPJ
z+O#6LR4O}=^)MM`cjk0CzJAhxo=rq<vlKjr!BJC}ab#D)hOTBdpINZb!xFv;Wb`yN
z9D_#`&umyS6q35uJm*Aax``3G!;7}^U@Rc@EL03gMn@9$lEVy4gsrtOt=wfNOj6^T
zeX8iED)9VxHkULzS~CX=Zt)$5BEVsJ{z=xdd+xYFyi&)WIY2KUC_)Vj)|*fbjfb#e
z1_+M`PLr?A@@)tjL*{-LpOVlN?T2bnU8mWzafs*-$<&KPu^;h*XPl|r5(A><{UUF!
zn#m<`zwyHe?J7K|St&WWH)-roO@PYe{h(6Ts6H$vLR~7fiwj~vD|3QoB_2V4e3K&8
zD{r>)#Ws+%2g^0(9(&-HFes3fF^LS`<(NKiA=Kf=J|D~e<G9aTk@b1V&rI*BhXEqz
z{-Jh&Bd%f*;mZaL*4lPC$NhyyV=P*<+h1dNpB|aT1kx7^3`&l(sG}gk;X#tEnNr|a
zH|{)ZM1FOjX(XExz$yMcZoJ(rw{$j6Nw|PBq8dXuUoJKeQAOhV=_LG;tbJuE8iKBa
z9Q4-#F#k{iC_8K(H2<ydkFKN5cr2CDllp_On*AH8;$xR3IGHo?TWe8~7UNkcrS3=}
zuwY($nFxF=YnBQWSM_re>r6sr5%I5U$2iUkm}CVFlaD_l#P@eLDWK_%=RJu>j2M&!
z1rMd<ODdN)m3?qzA~?p2ATuk+=m-^v27ff!u0<j@5&$O)pBROaN4%WJwnafMIjwZ=
z&E4y&CpK=LjpSxG<F^`pzNHBkEyp8s%a?H2*yd5&7&>g!N;@XRs6LmQsvwtB(DBR%
z%k6s(Eq12etIfJCn=Ala7Rq@?&qlaEBW4@v_4}y9i_={u8MY^8LnJ75Xh>FX0Y#Xb
zWHO8pHXKUKAMJYt?`i|O*;{Z41eC2$5B!{;P$B>HB}4m{J>n(mGV5OV)OKYGgJp-z
zV+P~?%Iiy@ok$UUI>>F+_;w%+uscY>*rvUfSbeAzFv{!y8w5sFudm}sQS;5(N`~Vw
z7AC7!kesr7EDQ75sRSQcnMk<054fz3i6(~yErPX{gW5|JuMuh*6<XN$cQ8-aC?ivv
zAb#w?S?&NM{@krbgoB-2d@V@(vh_%kgPIPXgf|1COtw>9No{J23^`2nP<@**p(N_X
zYH&cSp$ogFF=yZ<zIIb6{Q*9Gsy9b{AUl1AAz>O-P~dAWnbz;VP>)~@`}D<At8~!;
z7Cn?@TSB<qM&7*?W9hb5?%KHfi1Qqjdm_3&w@DP)ciMGr_|oHX$J_7S#&K{9@22Yk
z?eKuIQ%L6PJA?&&6uQqc57T-$9%i&YY65-Fi5C>@BKl;flk;>W&lo4s<tAv3)IGPI
zyYZgkPUWVZaIblZ_O}B}^wI?si>b(-!hQ@MT0M^%cG__Uezv(^m_ldE;#7y<usoP<
z@DS?NQHkACWEGw!yc4+=Ds`p@oru%RQRq*^QbmC*<0@X%5euY5%hk+>n$96VEC1}{
z9PgDgod5wU>(b@kMUiV#S<kCn6LU-oz$PT)auaj1{nx-tg@N+d;G8_y%@o@HB^lFb
z^r=0q_p9Co!5WiAdB5=(e!UvD1~|7HIKpzCd4;4=p8ZAEz`t%e56SJyry1twN)+4m
zyYWxYmI9IC8k)HLCHlP~gs(nHj11W&>Y2voZgc!=`h%4Xs<P%@9qo1%LGVARQ4)L7
z7=!#Hi=A1jzR5g|U64jnI{TBBE%a>MF6PJ!leU%Q%4EeOet(3{>&7Geb5>vJWrKol
z%O;5}TaB&;%%;qkQGC_j)15)ULRnu3Yy{M4+Cu?xJ^dzQe0BZ*nA1W`H+nHmiyQ;c
zjJp=rsQ<Jnuh1M0cu)ZAs)<NQ$Z~Wm%jeAjP$Omwv{i8TgP=#{rjSN-RkMZPTMVyo
z$5D-X|9r2{DwR(CUdw@`Mk0M!dpQX$iB>FCZhKUBSff-{<OC4s25#RFGtw%~$=*?x
zTjn(dVNpv|^lP&1x){fkF3<{7YQigJ0%Y+|Z=bTHvX6`$s0R1Gq0`=`iVgo*N80Du
zmiMrU?O=tE3{OODv*DXPe+a#@DZ}<C|I;?*K(%VOqNrje7+(u83AsScDhMF+!XiAX
zkpld%mb$H^m-{ST-aO|~ay8N6PilDG(1$kLG-sRcKoiE$+_;M|-<vMjB!xp%7X+mD
z6bkj)fF#mLSde<5u?HhWIKM}HDIV8)>yu#m!5sFp7{$&KPqSJ{@3-i2yflYQZ$;+k
z1gR+Itf4ylSHGzW!9&3z{y4WLyS8kt!b(>KhvBbbH8VGB`H?13cf~}c>YGC$BuI3K
zXY*&$PtdmULtKHz6nIaChw`@q#=b9&kL|dGDBwy6lXH^xf(UfQNRJhf55KaC%1R_G
zjR}GW8lG_)mNX*KP3y&jo>;JhCD-lmW=&*vVEhZc)*O#|4UBGLbGPBJBjDg5mbe+!
zzK?L-TFb=YsNkkuAJr#%Z%GB|P+X(6nNb+~lU2nr&*JXh-dki3RPsy=4<l?3?sGv;
z1B>R9q<=ph<9gyM&^Ij|hVA6A#h6qoJo1FLigPJ~!K#O0&;gpE6`_{=MvD{#d||14
zj{Fd8yXHfKv9$OLP#B{^o85I0$RV9jLMIkVOnN-^O@g&lwicHqMWPgP6jIvUY_xLa
z3cg6s`Nc|^dTwbW@qMr0Yo3jX2@L)gxlX5;1)y?6LFpJQ-3|(~Do5RTf0*2XhisGU
zCP%v{u_x=DM^2r2QSCY25u^B()x*@MPRWaA$gUzQtJ+*T9&cl!lSr^L=5SRea2!9F
z(ydChb+@GD89k9!!@?f^XR<@yQi*U;jA}8X#IGf;CG9WH{V9&vQ=IEh!kYDr*)e|<
z7xwBFlryD-B};0p{+Q6h+l(l|mAlGnNka>a6n8;?DQF=%l7K?4&x06J^9j?m3<<hp
zaXFW|kH2w^5G31L@V{J|)6HM(gC@Ecq{{;%4v`RxSLWf>&$?D_No&k1@0T^p@5%;-
znwj}1i0T=?cWmeihj`2f5Ksn{PUvtmcg<$I^sUNzzTV+GRnv5!gM%Irv|k>rc0`wz
z+f<&D50()&zDnt_kkS?J*Qx4DJ0B;S6W6oUC9%FqA+ijVY}q;oyv=sV=*}!3iBUi&
zs?z3-&URh-Y_p@5KJbvHl9rQ3;gQz+tP~-n8U_!A3G(OS8Y}%6$>0>QD;Zg<K5_<)
znh`LR@pt#yfQ8e0eCA2SULo1Rdu`px^w(q^b_DDX-k!4b%6+JVkb2A#6$v-Y&4GU_
zk0$4==C1b4Ko72}ej_M}0t~FPu>7YUzj(7wXg;@1^_1krkO7j>%;;Gms|W&=Sd0vI
z=UvtO_R~!HAUjQC9JT{|n9K<3Yz<c%Aq;zrAFpxAguocgRVps&)Kzqj+chWstpK&5
zsRs3I<HuSUCcnM`jf#=9w++7N^H)=^v7#l+!sQ`-??J}{^7r?e_>W`L@GU%K#R@mc
zD|a2q(7;m)$lL{3IO|Am(#$w|#ubl8`dkgw$5%0hg*dKtr=M^yG)SX8vVX;*ZYy`l
zW1v$E2=J=|UIDyZ@7^Q&?a$BXO?$K?$No&y8=Dt>*(<$-TNGW6ip<$pd)6jsymfQj
zKu(>$klLlVPZ`0MkKdEtc2e67h#R=p`x3U`tBc-}!yN8Pmt73=Y2<}pbnX02y7uqs
z?b=;DS?(#Mu`8;NWv|0au^z8gQI}rpW<ex=b=xC8{;Y4t4Cm6`>(n(otJn_NIbU%3
z6}1bx!|R!XwDBj!n}D{doFK`N@2YXS^MY;fq*6;y6FHWtXU6#7SqDZL%GCu>bLm_E
zY!Up7HEIdsWEsF|sIc#Y1&z{;h@`o5#dL>kf+onlH}Y&F=^cYq>D={}#Z6Tqhh4Fe
za3BZ5dD8r29uqQh<tqY?GR#eh+_HNS4?CXf#@rB8#HD%n;pFEetiC<WdCCt`!UVM9
zT<;8V3B<1r7Uc_;y0WX>EXK&DsEE@AU*_94D+GlhTB$1|kdMq5k(M&CH5(&hzyPJe
zCN?9`!?2BLuCWcA433cfMIWIvI}_uOYvvTXxp)LDyiihhBDL1F=--xQ@%dCKIplsx
zKS0Tsak^DX^@s*wJ5gdYPpXozg$_SPXTynAZ&4bqO=<4>xoBHfhQY|^zS=(cV3H2e
z;FXh-GYT>ubH&n`XghZ`o$;3!PT7HY2F`8{uIIY#SA2rtylc`;zx0vvx_SQcx*CNS
zw}5&|s~Ld|U3b4m4ZPN1d9!KU<&VGho{^j@GR2vq<uyly!1V{N+tx>HqJ{w&F^@P%
zaor<t*708Rxa2g48QhaI-NPolE%^bk!hL{dqM)`O;nujlNFXAMUQ)w%vS-8#2u(Vj
z7*a$D5Ap*ZwVRVdEA{%NE9^dm@sgKgyi#~{Gxy+Qgiz&edW?7VC|9N5ryi-~MCBcI
zh&u39{DM1HGC#}3<7$TR!`_e&MYbz}{8xY5N4S4QV5Efb6hTp)x{PAyo3}Hnv0!WL
z(Eu5A={N?@t9X<~e%kR4!X58Q+N$fWl=5AFp6bNc-hM9C*)IB3?*?lh<p3;GS@-A7
z(mTx9r^m=~i+`$9u?&32X1vjF`(0TdS%%|U@|{*qcSv}E1s`V&VT1&Sr8r`4LOJNo
zh4@+8!d9#C+XAATo(c?M`D|z_Uv;@bJ^J8h9;t~+bnM?(#B~tzk0Z=`((AQq_h>(D
z4uD#TrT8{`?V2+Um8RO9j+#*6Qc^}?Y9ZU9iSR4+U49Cd?Rsx^SDQJ-WAJF<7;<4z
zp?<mDhKk#0V|7okrIu8NV=;5_tf!juL#Sx7sznjbXuC1L4CC^Jen8fw31dgkCkt_M
zt7y@@q4!@d9B9pHc`hNOy|j_Vp<5U$4E?eHrrFrSH0;rC3|IT!GrY&|OrP3IdCT2j
z7^DDz%tsxURmMG;E-`sgJP97C<F?g->kns;pLO^f4rr-3!4?5#_SrrBzG`7nW8I^c
zUeo;20?Zb<1;rkC=<w2<=C0~?oQX(W2QQ7Y)7~JA;4`Kij1|>Mgc*^q7R<b64XdsP
z60gsXKg2Q&iMIljcvZ4k_^mQx>S4u6rL+R8bwYds&Q;A>c4uh^$+u_Pog%atuk4o=
zO3w`p>c!*N$@o1Ntt2JbvLZ%Hh~jpEdp}t#f+PvBX*2YOQW~yUoH}{tRZ%tIcX2Iq
zV+JeP$M{>;kj+GHEqty!v=}>&Vc=-z+{diexes}J%XVFzFE~qc{;VVoUN948<gZ(0
zF1g9;tBX|WW#2m52Epg2Qs@p}ZIf5Zw3}08LkrP0X%4SdtBC@PgJ|*Lf^q=;cU}QC
z?#=T7^DYh^>wFy&@Pk{sEZB2tfIN+7C{55jT}X0HRh->4??o(t31#Z|JLMKfQsxEc
z;uccR6jgLI2C|1IwFI3iHZ~SMJY~Cdn=u^;;n30S4ZeuQN~zum**rj=PLV#i#}4*d
zO|tQ}bX@M)L{fI-jqP-?t-&2CLW<A|NYdY*VbpKSE<NH<kNh?n^%;Jbys`r>t4A-?
z+^d->Hcbr3IzYyFiL6*r0<++p!<_J()1u&H{Zb*vEKQV^mG=^H?e{oSGAIq}(_qs`
zB>x%&n^o?Q?sK&f)fl*9k;vs>l7D;UecDhW{9d)F9V8^4=%!)xt0cEi(}Y(g`L9Ky
z7Rv-}3A4ugMyX)LxA~;_=%>l&(rK~v3GJY#LfmxPJ${4$DbZM}ak^CXAYj@2aba_j
z4H<!g$I$$eIEnN?Ek{d(P6*3t#HF$`YuizCj;g$u!m#sma8MDurw*rC5JFmx`SN<i
zwA;Ab>2rY>C$#pXFp+L3p)b)vU1CJL=2P3+d99aVuQz)Nt@+m=_>qx~vZOkArIlo~
zGFmn|n$FRm0VM6l<srVsBN@Ov(bT$uPkLTN;8B;RB=(i2>uy1|Gxu}y@p0JWa~`Ud
zrzcc7Mi%T}5;MgdI>xVW2DycE6E+^chk%hp!OX|QauW(mBa@PEqd%O1_JVn!xIkrv
z8aIc!YGuM1;VG6A*AL+6{j{0mV71l~8v>qB<Q0BJAK-d%Qen`$$PtA%i4AQ+rJtd(
zxI`(EekJesF4r<3^D2c!ps+1EXE}?=kyc|W7<EnH(rlwue50iK11Q8LJfq@Vs@5Lq
znoeK}g{NiSC^)rZGH(DX-rmgbn?KeG=I_b__sC5#<=Fj`YTOQJy0+QIvVr=MIe&B<
z#6yx@V?UWx@x$}I@91Y3Dn)(!*&gSr_%duQDxl?7JsxvHLiAcc7WH%jn)rrwN+X2j
zAgj#>CaALP6^gY^)Gu99p&2b#)@NMG^Qg>1>edR1X)r?8G{kpvNmb%9j=x8e1c_0k
zYzODCSS}=cL|VME$vuT$8lMoK{#ukK7=4=IDluNUewv2&IUUc;ydF3fID9|qkt`}s
zsh0w5(|f6n<W2W{!U@iRjNj{j`ukv>`TlBKkzIGi^S1KvzHd~d;k~n4*5}b3lR~a@
z<W=j9+5_!lwHZE}xnr7Izj7xKvFNV(<-#J<+E;6eypQG<dfz=RUm1Fe4*}z~QO&C{
zE5*9dRS}SPwmVsR`PaTnX|YYfW@YSiF6~YUskmeuc$>bJ7(OJ+se*|tJz(O2Gc>hK
zyWe1n7YE5GuT+TL!hs3F&`h{CIllru{ejv-gn8UVk$19q+!7;4%6WX9lM|xHme%LT
zFk5fP)MSPGj8|qna#e6&lQ=q5n~((oDZG|8lHNGUr^N@Xt(i0VVo!dRigj$3Q((XJ
zBbd)tbIFV9o3&eMN9&dO7q{MCfLn-QFhXLcGkMIny8;dNDz{Z2T^3ce!5ol`2bH1&
zH-w3bh+DuVC?Kw0n(r|$dO%9sv_Kev0h;2~5H+@1b$tqEY2ofTZ$F}Z$@R<-O1oU6
zxMw+Oys32Y@TFpHJGJAbA`-NtvZk^>WO83*%-B&?*Uy1*k>8}Kl_PS_cr5+y?Jmu&
z3Rlm&u-D=}hIw9K-n&pjrAH=WB$4lY(S?1K3lDz0^gQK`8%sniSk-WI7kbi?V?wrs
z%Wst>cOYDwK<O)GV)e>-0AGLM)Pz}`d!(;P$|6+oj5z~F%~O^|c2d&5b5xjAX@PKt
zT5wwfGc58@&EQbJe!B%no_05OTHai4Qoxk7x{T}(httyMyopjR*Xq};E*!cxj||9~
z^jTE}!cjHmB2b~*vc98-!H!n!iUeo!OFzUEV}sJlkh(2)TNE=DTXru2HJ*<<HUU=1
zm((z-czX<=`@0l{zxFPi&zDF=(|DLEC-tpSj4?AlL=@8&_CL$#r>}s~Z{-Y|7@ka_
zv_wUnm-?A-@n=NuUKxUJCc|u}L{u7Kv_Bq|v8gC*GB#GXh83BMS5Ir7A~*Yn(9@(U
zEsllv2ml*?{RZ|fK^z_htKomzNFJM3p1I}n>Gz=s;-a5ktxzg7*2V6ayn}YoKGhs4
zJpIrb!l1|+Qxz$vE?Zk~&|Zx+N8HA=yZ>u6p%QTz^%P?a_>nJ%$6ahCnQ0$r7j$7T
z(FIxT|AyNmj6+wP`xcPd&@di(f9UX9tT#a5Z=b|H{#qMnudXh_mn$C>(QL6>|HP;_
zc}|yaSrnrhaY^!GPUHABC=7h{YiErD^It6#u;nU7NN09pxsvGopZ3+xi+{bKA7iiv
z*%1tlCTNzzy4CQ)3)^NV$n6#_fk1N(EQo}=j~zt0s<Ou2Zo=pxE-aekYAA02Ki@tH
zbVMLeemlllz1+j-BR*fWaCO#QUE!>Zm>&~2@(VgRTJ>~uV1q(W0BXuYUx>O$Ak}D)
zEV)T{`x!GyV_umWG1{KUE!jh0f@ZZa)Yrhib#kK{1&F#uqUIXKpzS=Ky>mQ&-)f;t
zwhENMM%poSgwpGX4NIsEveLao>ga(PWP7clU%?kXU&IJo=Izu*E)^-RHB_aK?H=N{
zbTN=Rx6z!n<iJbhMEUZ@5JK;~ZV5!P6Q($JexUmF*k=d^;;d4lk|D$<61HhdN#aoe
zxRvf3+fQx8)_8KImIwCTrExTsO)K9U&#$6riTJJPspkQ{<1sHeKY2YB7HCAIDF51v
zB%b)IvwP1*8gi=W>Dnk`ZCCJ+-tyv<ea#S56PT$?(md8%-g_U?i}`qDx4Tl@izMhd
zqqm)8YxY9CYk52o7UuTxFDiPP^|e>_J#bVfRp%PI*l^Ai($7PSB&%h!D4V(>%NJ2z
zvQ<VyC7(rL5SPZOmF2f*9T}MO{?_k-)7)gvs$Q0RX@SPbI<{leg)GKVAQk0Rzh33C
zR5zHlpm;v{ThFkL(^>wKnlGp3cU%%x4RjL}Ff*qa<lj%d9nd`QJxPK)?z+MM*0e;C
zeR)s2E@Z>`SF-`Xrf4ub^Y$IO>6^|cte`;z5ke!#*yLh41nl5o*LngLD}w6|Q80)E
zYY#SnDXhUJ4B|DKI172({Ac#<b*Ml)WbkuYjI64;X_FqscfoMwEds8=1EPIaV5~nK
z#V6#W+Y@ua)_`WkgBP<J=4v<nOd%XvK>Tm?imRDIf%=_tyG*2`d;yyuzgU9Ng*0P9
z8xW~5HctsSir}wP%m6W;R3C3D(!2)l^G+kd0>epFifL^Vz5~N=me4ybhnSw0F?x6;
zX7`T^-HvJG-GD<61qOV3%+WYFe77BU%8TVSHoKyaN5Sf15`3VI=?w4~yY*2P<3E9s
zbdENQ9no}gHY{P69)s_KW7!utTk^Ngwx~$#Z*|#92gl@Z<aozng{~}6)P*A`yOC#`
zUWlgUrKsEE&3e1|ZXD<iqH6O3vjz8JVR{Y)0)?_ypnZ<=^bTXv{uqv@cvFl+)eQn`
zQk=MdTCUF0uBdAgoIH549{Q*--AV;Q2vwX(-pyy+lB*BL1-^}u-ph<GuUT3DpK9*<
zA*!cs7`Pw`0+OO24K9t8(k#N#%@R@)5=)oTETF^!(hEz2q)LZNcPyopG}0;E9q+F9
z_r34?dH#dvr~PTqnKLtIX3orYeUL){Lhf*{X&5S*ud8;|p7eql;9-5kKZmT#Q$2$}
zvObjS1XD-8J|qb|K}Ui6+T?hVBYMB;)vlof;W^~IptG#AZ;SejY*ywfcZ#+9vUr%r
z(`@pakTlKJ?=mPZvP1@;VxT<%1V{0R#=cRZIh>hubU{R8Q!n7QpG%kP<rBL?_eHI>
zvY=_A7bNVOc2YROzvtp(lzufRA7|)Gj;2a?+G&OX3Vv;ZQ=}l-ty|*A4J`_WyMEm5
zzinv`-dW<U^L)%b=GZIKobPJ%I-wwqN+_r{LKMidEX|&^JGt5SzH0vk;&pg2*<l;#
zyaI3=r59{pWQRxfWbiKzbS(Sggy<@>(3mB#2$19K(%saf&+xS(?vWEpb^F)`^3w^%
zhLf+g@D4E>w-;E99<V<uPNY;9aOK*h#%0ka=UY*qoCQ6oSk#huk1X0(_qtY%($Re|
zJt(9Y7T}^UQ9d}|X_c0w%;p+Deb*e>?+HTH6q%j{@Y`}t!L8B^-1u18+l47k6r?^2
zpEn=%g2Elxg>SMHz)XAevsq;Vz5CCBi`x`a5l+QSBM*B~CtMG7yNynDsXD2*=g0L!
zjT{qf=l0DX^0j3yi|0f!-!YE(vT{aF2c80#{CQ-7zm<4+(H2-Fy|ebl!%+zuRbmdW
z(^`g}(xo=`d~{jd;$<Cu)EaMK?AarJBvi}DpO@c1`sI=3E;SRrQ*r9%$*Rghc6Fg?
zn#h#dRZOXFaxlJbCwNW=WKbJ#8WI*rNaeYq9@REe;92k8q43gC#FZ*}75kJDw1Mlf
zG~>HHkR}&ps85yBexxJzx{x{0DRinzpslDzE2GEb7-iyi8z)1)%8A0Ou(fJ<%Dkaq
zGk&U@P5Ja<<cc4dIU-V8qL+NQM?U-Eyy)GmhL7$+YTaw#F%2tSj-xLB*z5tKI>J%t
z{&z1xkb0q%Qifu2qb)vGwbi5aU$7@R#Ee|G^XoiMCt+n<AhY$mEjo=nNlnCN`n}Oa
z^a<M%e4j77lZJ513ZI+gNi!VD)40gcE3ZKD8ivF^Dqf;<^y;s6QY3Gq`*#D`<xgcH
zN9C_`iRFxSs1LJ{pcXyfZV=rt%p-@mRPCIg6iPoENo!0N?nf_O^_x7zirz=;f}~08
zceR3K5W^pHvQ2aFra3>CIWJ2hFvySIxFosYDRvF9*%%D=H9D+-SW5iYyUN)h2uS0U
zPcho_S$rjsB?icn_|aoU;5OP`1gg?6_J|%XS88IAoLx-nI0$2)nc2OykZL5JyO8vH
zG&$fN^Kijvy<MvzXk23RP1dU;QqeJ3`4nutGM*}93CL0c#Bru4pIdK)Hw9r2!lcK1
z)M^)IR3~{K6gixT(M7_)Uxo@&M!xnnBYFTp?&a)Hf;}rP(czlyL;O$G&A3ioq@fFZ
zhh#VWWAUU5u~0Uc11WhkjA;@U_p*SXi4X+Urt=fHmDBATlE|t#a#5J=;NQHrjkofy
z19y3!>s{`eG`k5xvR#4Atqo%K=;x2jETdD>VF_FHEpk0w_mGO|Uyy=`Aq21R1?If}
z$MqCd@8Q?}*p)Kky9Lta%W&Uk{_PAj314pGMV`|iLu4F}veuuo=iGGpk}Rw#Z#*s>
zl%SH7>1CuxwqHp_56z**7-2Nv3uLdZCAISE@RVbw^VcQrey=~~NQ*G>lHG@(>L1^Z
z^>>*Qk$k$y3*H#{mAGF=yt_2ndO8N#CMYE@^?st*+T2mu#x5V^s3%doz4jdM;ALXt
zmL-((I;XBCy5yT=0paUY=ldtI8kR@)mpjOm!zupP7ZMST*>>ZizqHz&cSap_`K|2c
zo<4E8a}b^ZgHZZ219{!s6h$^SJ!+y^t$wa2nMYAQ7!qZJ>dMui6Hez~@)kO~1e{fP
zJsi2y2T<a^Dq?kee-ss@sKA$4c3qKFRLAS7f0yems|pjB<>>AJ?l=)vgtVfd%cx&Y
zK*$q&Tt>>-bAHCUXz<gF^x4BtDkv{cA$e(_cq){Ox%BJo!iP=f_XT)|d(V%{>&jPE
z^J91!@I0_fPh_tNdDQ7=r-9-ffxQG-m*@-@&u4q$@>zJL`$P_IPM({H-*(yIcdIY<
zY_|ui_I4<6m5K>ue|qwL(4w4v5Kl&F+yrcxY&`v;#3BiIfy#UOL!d8qrx)>@U(6dG
z=IpmxXFZd%?*06+r)EJ#c&BFR0*~kUfM->DDVsq#$0Q85MhMccbQaIdcj6q0DiD=D
zdrN=Gc%kRUg%XiBqU}q229NL9Kb6b=l)X0!VYW59LeSo*6Lq2Xhb}4Q;Ae1}tcc<u
zNp?rVEL*N&Zl?G4Fz@e@ykl>V^P@Xnb}oRYLEmx$W2pVi8#)#u^BQcTn2Ct{0d<<2
z7T;Dj2Jo~vLfW6@fUb{Ba;4}Z0r>7kHQT)7qgA{(OdWPX#iz2?>gxNjg&HuylWlkP
z;#75%+?FDhZ+?@qK21<s%MvPzzuS7L*H=O{8Q@`Zl>H=fsyo0mO_ksFijFf}d83c&
zbJOa;z94^uIkGgOd@9<((EQ#vbhs~2*De3)cenme6t_A*y(p?j{R{|sOYCOMW$m)d
z#sAQzB_4Z4-H}^Y3MHM0qA-U*te+6e2ag4Dg~>wHx%&9^d6$Uz*2(FupMSL?3?Mrg
zx@uUq!%HD6Aq}+e*{!_kI{176Gfh5@1)O#UX`U=iz~SHb@g}r6f~1VN!WA30TtX8P
zXe-o1Tix$RJ)A(a9<OA_@p=0$Z}_*7XJ{;jbN7YUQV3F@T7HaH8SNU8_bsRa(fM#W
zec;m{_kMUX`!P`mK_3{F(-4~QEt-iVM3>%N3#gqLt9-9HNzy}~h}Y~S<=$Tjwe-%y
znx!B&)So(y;sv5UJ!i~7!3OUx<CrB6w<&PDv@g!|a4ul+B_q6}c#d~9)+dPDgo21i
zobP1`WLZbeB;3wK;ZY7qCkOQhow~<+Js?(3Y2ORnXFt-U3@ej8xN{rI_{7%yhHo|z
zZq{b9Rr(^|0snCdleqQU+s!+79@k(jfUey(qj=U1{8DxkJY}{N+10PiBDMn)By&-x
zU+ShzjRnAy(az`uf~dKw<y+}2`yY=F!ZOf#M;;Z1arA9z9R4BXSzEH2UlI<!3~Gs+
zY+6dDdZ;i}L>LX>I>W*w4^S^XW7YSDiGhhZ9Q47PUttHw>4N}9v8lW?n`;}R80bdO
zZ#(0p5LgUPMW)Kw>C1@Tder521hK~w=zOQJ!c!!@>+nb@%b|DA=4u)w1H61;Wlmwr
zWP~I%Z?1~<%?ROB`91_B^|Z}#*?zO+pq*cS541EV!23~^Z4*GFE*7ltTpp?3wI*))
z=T~nI$c>W25V}Pnkg2fsjGZu<Po{00D56a8Sjc_vL{pyjrDlFaA{OMbXyNgFPsI&J
zUvGoCh(yMSQDg8Kqxt=i!u+q=V5Twv>AdEm%~IGB1s)4KU7u+R?NlEKmDB0d+}%O&
zi2RIfK%0g`N8Et(TeanNs0Y25d2qB|e}+)-9iCaXOt`q&lAE&$)DZ!(YRrU8aiqE|
z41{y|ZXPz3=4LD%pZHtq^M!|%H?+HoJ86NL?tBm(i>OQhL)~94jo99+H3PKJ@A^N4
zXgjV1R;i&kPD=M8mJ;<9lXdS)yF9lHa-D=NzC3m*jOF&18;;a$oJXd(M>ogNMBrfM
z4AOB@W*J4z3GUxI(2rcGG<by&==5(F{Z3xcSwT#VS4K5Y%^D2o&;6PmM=V@6d~;YT
z+AVY|f{Jdw137%t+g8aOwk*mvSm`P>TYuPCP*fBg%WL3vpt2?Qw83LITK8L90Ee&b
zJ^Z=ipdjs#cTUeZi6o8&lh-a3clC{`T@*TUWlWKH9RWPX_SB>f7fZpf1dsOz>y6oQ
z6!uFF=?ERXi%z-Q)~L#3;cLx~yvT*o)pr@i2_(BSb|#lonz}>|nXAbLij%)~R>BJX
zE4|^rtK%S6wVCpd5aK#)kK!xX;jHz*-TKphgMln}pOX*7zNY(*b#&w=9_T<>kMHH4
z>&eAO)WCL=hIsK#?pjn|@8<o)Z3VX?DhOxZ#g<anw{EOdUjXTX6yinD??O6*6<wh9
z=12+70b$=AK~DBxwoK}RU>Kf9N;mFCCxPMlyItXr0jv#U1jd{Wry6g(!putt+3eVo
zKWAhm5zREUJ7Fq%%lQ|+-&a>h$Vz`vX%SRn^Q>^Y_&DuYVjJo})ijVgCD?z_WY9JB
zr4lGKf(Re&Kl&zfS9f|Zn4%+#g3Da@;GQP7Md9d+!lZ>Dnu4);Co*v8^ZkYkrL(d)
z9A1lBu0YQxt*Vb}(_7U=I1{5Xl$3BD(_kr=NZ;j6T{)Xr)<YPv%YoVOWec<`OH8-)
zD<($P#xAu&KkR)^`3bt1_L)7*kR2-Pkm#eAdi&>1_kGzsGEg1|>2Y=Om5qH%R-AT*
zw9%?mMR7OwGZifTxWrnBkxOP*y-<Zp@7MTWEJ|6lZ3LDd&1$-Dn6*ir@UbF(76S~)
z?(*9-;ZdG$fA}FyzA5q)<dpy2d9NEUI7q7Zxn9@dpa7^@v#gsIy>Ems5vf`i=-cwj
zrgHx$T+*D`w}(!zJ=bCl{W<;u9c^xn69G!L2sTi96ZM)(W2wW79U$~Jps$Jrr%;pu
zwM&27yr(iS?xmQN2FiuFnYs$ndM(ESt_1>neMCTCdJy}fDi^oQ0WFH^%e!GMEU#K*
zSasoX*2oY1&-|us!`~yyZIB&Kw)pMAQcIsZ^pjm185zV!rlm|0C|O)LmA^<3kUoo`
zH3o$kDQ_aPhnwCpH*0;gHLCLth0`5?i*;!8z=#_GX@XC&DpYS9_GZCXx@5QYx^5{D
z5tf39S8w{O27yYSlOXKUovNv?_S0Qpr);f+PMn55uv2jxKRoB*P`A<!>m-?m?+rQN
zawceAbpG~Ik~U(F`FGKg80?d`f({p5Q;Dc9&{PF^W%JT9<LXZuVAg^iM%9N|vF{rR
zR8Uz=rN`sh<6uX`b=w-=gv<^wK69vqn!Kko*V(vIaMVew1Py#|9^ziVcM_-wQlpc3
z`!(*y7XbwjvJ+`E{3s(WZv8_QMp&m1GSp;yJeLDT#~>m=uZ;3X(lMz_^r?KxDqjfs
zTaN&0Q%<d>c{BV`L2cR<AZ{j-4V&B0Yr)n(@6O?JTIi%x=|xA-!c(*J09<iXUbka|
zMbciM$x?7-K=8qASA};NBSJ}lum|Zo6M~?SQ|$TdF<}Nf9)|Z`SZzIR2ak5)LwEos
zFBZ!=^h$b!gL&YqpI!;OD5WfMC~v95T@<T6Yj`W^*7Gtc6*yliEE0y2Rnjd`176<t
zTj6%##^UAX`su)&i4u;Q_&?|XV_G#*<_zTZ%b6HEjXvx<>W{a@Jl{RGx$5n;RB!3u
z(n_-HtlUFW4l@2nW<{qK-|XM<n=cM2nVBA$CB<_ExDh*2UI2ccafv^@@5Qrnj>*Ob
zN_X@Ei+NoyoYx;u8qEp(fg3Yj*G8BwI8$4;!Xs5zTNN*@SrUY{J;u<3cv&6^Z`K6K
z>Ti5}>mr8u6WP^In!Vz4s;Kz(y?Mm%4HuFT=|vJwR^{RKu^u08$#g1dX9wJDw#k@o
z;bKZrd3k&t^Uh;0(uExr>_5td(jCq&KvV-oxfjv~TpI^WvR4i9sO@`$KgI$MccJ<I
zny%g7#nY(vF(bAtdz!ZKO_LxV<(%MJoX^X9)j`Z)ALZs=g&6BL5gruF-A=uPCNc4R
zJMaUy(7l4V3=v%=rCVh;yY|M&o7fM|5B%lUU(@Tit@dY@Cl>gR<k0C5Zg9}&TI*)+
ziIyr(SH{SgKE|PBIeOw9;T67Rp{<QusC`OP?&mSHPuag6cjEgvZZY$&A{pa=0B;&O
z%7%|1cJt`#hQ_=n9A#o_E_Z7H4)K#E1~_oz1j&Sr$D=fa!z@)Ss7%(wt{<p*RwZkR
zs`cebH;pkNT<H`-OgG-j=H=c`mp>P*dr(XgQc2s5a`Z{~UK_{z=#P5mpD7jL(rXUH
z3ir?_`X~V`r49RqQ0dk*TP|&v!WgI--q=Yl!$RWQVkCVlZ8MYEea@Lv4Z0uxx<Rkl
z1T6!LUKC1~ZBdZguDozIQ(u1QMTt^W#LgOUq+tCz{ob*(+aN}iZ9q6CUpPLW46!NW
zoMnlU>C3#~2~%Ku6)c&Tl8$ThjwW7j?*U9&gqzks94X}gW3%zlS{YGWOG<*xX`y$q
z?z|O$Zi46Y5F`$aVk1$r&!m@N)h(zOUiP-4@*Y-XO@MwlA)ZR5mQ!0OM(pM@D&&hb
zW)J~<;Wy9XCo`(`PdimOn;Q`!N}uHX=E(4pest*cG4t&DM>?QX1Vh6TJgV@`QO3b#
zSA9T)KGpk1skS5aFi&+Kl~pO6IRkcv!>@A&i9HNC3=fZm%*|#M@69`KP34doSH{su
z8I^7+8%A!qkVw)ln5<l3QEzl|k8|;QmRY%He_ou-J!dwJ=xY52;M7QaZeo$`$4g|6
z%-Op?*`r`4PMwMz7%Ol;W6*UC|6t+YjS*AQl=)-$ObY|QLFc^pA)e@`G^*J8qRF!?
zo`+rwG?JLU5MbD2tjGfD(#n^z%J1W~&$rwq=DOsPe*!skMcYPo!YmPXYJjNhEY%Bm
zcpOc;h`q~q(0P06hB&bj$dc}>RWfz>olZ7sPyD5QC2f9@LqdLVQGqbYFbjVOzMRQs
zSWS}$M3Q|>FqNhFA$HI=mw6}VFf0z}Gr%CF(a3(a0Hvn4!=Ma{4xE|T7~tEhCv|N4
z)0<mz?b9952eM@z>BH5_3idqZ@6Ih)xbcsA5{UPt-?*h7tQ+yfl0OnQN3=GQ0US#x
z`r4`=V#m!6YLc|Kgew8$KIXnEOcz=gN1fPau64v92M!ClbE+(B2M)PBK*s7gg4Ea%
zBy>jPWtQlV{d|EtF?*%I)WnK?*Af(za0Icjq%($voy?j^9mtimwHElb(=N5b4p_Go
z+*!%@W*4;eMv8op{pTW*IFA#EM>r$n&yRf&;#62>u$QVr=My2%sNrG)kgct>%8aX&
zkltm!xYvS=*pG6asxdkKGEpwKspHBDABZ9ZrI+CvH^@q^mpg#`(4`{4K8qD-$?ev#
zIh~*2w2+mpiM^}8ztOMKka`bZ^e<|rkFh0|IcSECqF+btbz$^$EE+M1YHSLgwl{yW
z0(-k`A0hZdrZRY1_}lCx(IbFz;SM={CPvk=gWbUmD7%;Xyf_|j+xGN!g1pr{)TpxH
zBddft$OHTrs+gw0g2440qFyvx5`WVu+}`L`1X<R*wmx>KF5{sVpkfltc*s5ZNz74b
zd+LQvFaTq>j1ghcNnL7LT*v+Fv+6?><&K=$$%T(5CN>@A0WMRZR{uh3>|f)o^B)gt
zDF%eZHQPV2cNMsH|MQ1?R&CZW1DBPc3+oNw-9x#=?3$NKEevD7dWwh1w4xl*9)Od}
z`wFkHmJ4Q69ULAeYWUKtE<geaO5-%QBl*r7nj1YrYU=uG<$^6hz_R~eC+6&yG^X+U
zOeWKN_j+*2SNrjdxwqR0h@@KG&LT<qVjj|-<FxUR-P7GL8%Hz5mQu+v*J-E+y*Qa-
z2-sadmj?$;(~rfyPpZh6WUwtw(*B7_CGS0dx}86Cl6;l%*A#Jo^-50^!U>gqAF5-+
z$<;P3VpUDPP4;ls82w0%^^sE>5u~Z>mY23Yt8XsVQRZU7X&Mo^>Iq?AGmyi;)HyT*
zVtS0H4eC_-qNDnNbBnfcSu~oFxwAK)(%wghreJ*b32TB!sqYG>^7=&Mj$7%4r{7<V
zrp%Jm6U`~jNGy^@vqg>NL<8$&J_!4KYF}$bdjs<#AVZ1>!t1F5iTa;YE4k*3Z*Hs@
z^<NN)@3rVV{`A2BB@}qPf@#rwVMd8Pa{L5Aur6ABKgpGNtvZb1FhicPhjXj?FO#K@
z*))?|LUy$@CrY%STRswaHXjH9Lj4K@j{J9ui-)uA<^S!Qys{2#W@MISE~*eeh>hl&
zybx5)smT6onrz`<kJt9Q^=5TgEWK{N7_Int@B;#@tZ%2?Dcc@?1X&+fBh2-EL7Mag
zn;xAjp?vc`HW`i2NQ~?7KKSHwy=h`Xv>~=t)_xF??B(W)5F^J%U!+tiH2>$R?i-A`
zf|gh|@D${j1hU_0Q*!!jBVYGe)zJ#X=Vw)n%KKC~X)!U7Pkuq8GZi*!C($XiHW`uS
z>H!~)&<U<zFsOva2_cQoRyih3uHLIjuul!=KB;Gy*sLZhTQv_%-Msp;l&XQp|F0Eg
zTF7L|lR3Jo$k}g*oN=h<KPr@EZf3<M2nYz=dv;HFE$v<J{kmgba7Qcfq&#|i$M3X0
zQ37i%!^e>{Vmp8JUI0sBStLo#67BB940`u?%RYauAA)C}^L!v5-M}rB*sJr~_PGJf
z<3nBdWp9Kq3-DZBzm~gngNn7jXqLgYWYmT1EQ~cnPkemQG~I*oVfG=Us@^$`b5H-Q
z0#CR|M^JJhTBv^VC!$f(=^trk!W$#i9ACUJ{N6)aGc{T=hFPU=4&(2aVh|t22bV<<
ziNZ9yg+nxn<Dy8Kit4R<4{EZwgLPj?0%m!8eM(=$?Ekk{{H<*C#YTOEr>kv9%ED&w
z<&T-B#kruH;yoe)7$Qqwu@=pcyq?`{M6Y7y&@b))uvppnATGU~5&pNi6mY8OF~(9e
zaNscy8#WL0k8(Z6)&NjsMbKkBjXF&G=S+#F4Wky*v(OR^vVNlTxgbZ|O`@IA($k#V
zwL}?HO_H<UYrxkL?mjxMmQI@905qP5fpadH&^*q?V!D$u4hxerIt8{qZ3K!wgnWPU
z?f3kBA<7v}!quY6{tYx*0EzH3{Q1NZxMfCc_%r8Rg{iCr{Zg*52-PZfIVaAWRnNE^
zj*Bk${Znee#~A<V@vW~96R-3<$Lm@p4|MlzPD^gn$oh;tf@fakiv7ujPi5*}rMD%b
zKM3yQG@*?*cknyWEd4tBrMYiE{y*6@jf=iimAVN;SqKXolI9|}79fT@Eh3f;U^k#Y
zt?5Ea7N1?dEcN<HAqgjbhE4sU1;dv~H_c)MmvOXWQ#}`IyDat(CnQLMDHzF>uN%`K
zlPPkkGV|L0flIm9#OjIC+aoHDXf#i?seYb}XK#bE<GUu7cx@ovw5272MoFsP`&7F8
zV-Ja8YAz~inJ26zgki677g`E`K9PT(I(w9FsbdM^h~rVkL3;>={%9lSP)W5PjjQmy
zimbr@=kJ@5Xv=6b(zA(&G%273qu5o1&8|?-tjnd}-8G1-JHtp$JwYQkO>w)J=>Y>R
zngXJi8_~C8jJe+%B9+Ob_E&>q#CTqeb&Q;`c<%vQeJtMJ=lm`c?mppsUg4LH!T8V_
z`b}CP;qaYb8sEOI{+GhW#gCE9Tz4&Iswl8AcA7`O$l7~diIVquoZNiccg%uU&|Au9
zYG#Xj4gW*@6TFM*ks-#tqK;lTTgvAsp(ahi;oaujoAf}+JS)ovG`t7|Tn?@9_@D1|
zGe`es=du8&$ZaeTpKHVZ=asNAje&ja3t{372@qfkg7OhsoRV$0{ow0*@8sD|EsFcf
zc++5f<FAD!)UfcdE5m5|_GTVea@~GI9>cYq!z1l|n}19-y;N5IFP+R(9%Gh~!mrRo
zq6bb|&Gs(Fvpl^yZEmg~Z`>J$qS?uH3{}^!)Y}j$b@bO&n&<+KLDBBjj+(^r$d1IW
zm%3N_&r5k)H|@WY3H*bi@Kcb@_#2P(*KQTR(Szx~bt*aZpuzu<sTtk<U@d4-rgaBx
zOUR!l5BTO^q9e!*BLZ<H)}?05V6<b~Xb<jPVbW+Erl-wdg6v%fpDit8xGAs9(R2)?
zGxNIXAfY0K+46wE<g9H6j8+SOZ@}-d{+F?)K{^2tFMwXFW9GUR6eGh?U(XQ+3+pqB
z#>j?sXyekngre<6a8Do#!F*{IlxImf;i|B5{GVd$5nwPuf=87RGNzFvrow0B*ogmD
zMO)l4&foU$%gK}f{ORu_n@W)?RBwNDSJCDCPaXVgdg;dupXol5(|>3G|MpbbFx}0%
z%L4h&5b^Jy?FbC)ReM^w{l8(akC?aY|J-d_ZgInicc+;g_1!S8t)i@&Oo_DV$NvXA
C;<HEq

literal 0
HcmV?d00001

diff --git a/_static/pygments.css b/_static/pygments.css
new file mode 100644
index 000000000..d7dd57783
--- /dev/null
+++ b/_static/pygments.css
@@ -0,0 +1,152 @@
+html[data-theme="light"] .highlight pre { line-height: 125%; }
+html[data-theme="light"] .highlight td.linenos .normal { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }
+html[data-theme="light"] .highlight span.linenos { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }
+html[data-theme="light"] .highlight td.linenos .special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }
+html[data-theme="light"] .highlight span.linenos.special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }
+html[data-theme="light"] .highlight .hll { background-color: #fae4c2 }
+html[data-theme="light"] .highlight { background: #fefefe; color: #080808 }
+html[data-theme="light"] .highlight .c { color: #515151 } /* Comment */
+html[data-theme="light"] .highlight .err { color: #A12236 } /* Error */
+html[data-theme="light"] .highlight .k { color: #6730C5 } /* Keyword */
+html[data-theme="light"] .highlight .l { color: #7F4707 } /* Literal */
+html[data-theme="light"] .highlight .n { color: #080808 } /* Name */
+html[data-theme="light"] .highlight .o { color: #00622F } /* Operator */
+html[data-theme="light"] .highlight .p { color: #080808 } /* Punctuation */
+html[data-theme="light"] .highlight .ch { color: #515151 } /* Comment.Hashbang */
+html[data-theme="light"] .highlight .cm { color: #515151 } /* Comment.Multiline */
+html[data-theme="light"] .highlight .cp { color: #515151 } /* Comment.Preproc */
+html[data-theme="light"] .highlight .cpf { color: #515151 } /* Comment.PreprocFile */
+html[data-theme="light"] .highlight .c1 { color: #515151 } /* Comment.Single */
+html[data-theme="light"] .highlight .cs { color: #515151 } /* Comment.Special */
+html[data-theme="light"] .highlight .gd { color: #005B82 } /* Generic.Deleted */
+html[data-theme="light"] .highlight .ge { font-style: italic } /* Generic.Emph */
+html[data-theme="light"] .highlight .gh { color: #005B82 } /* Generic.Heading */
+html[data-theme="light"] .highlight .gs { font-weight: bold } /* Generic.Strong */
+html[data-theme="light"] .highlight .gu { color: #005B82 } /* Generic.Subheading */
+html[data-theme="light"] .highlight .kc { color: #6730C5 } /* Keyword.Constant */
+html[data-theme="light"] .highlight .kd { color: #6730C5 } /* Keyword.Declaration */
+html[data-theme="light"] .highlight .kn { color: #6730C5 } /* Keyword.Namespace */
+html[data-theme="light"] .highlight .kp { color: #6730C5 } /* Keyword.Pseudo */
+html[data-theme="light"] .highlight .kr { color: #6730C5 } /* Keyword.Reserved */
+html[data-theme="light"] .highlight .kt { color: #7F4707 } /* Keyword.Type */
+html[data-theme="light"] .highlight .ld { color: #7F4707 } /* Literal.Date */
+html[data-theme="light"] .highlight .m { color: #7F4707 } /* Literal.Number */
+html[data-theme="light"] .highlight .s { color: #00622F } /* Literal.String */
+html[data-theme="light"] .highlight .na { color: #912583 } /* Name.Attribute */
+html[data-theme="light"] .highlight .nb { color: #7F4707 } /* Name.Builtin */
+html[data-theme="light"] .highlight .nc { color: #005B82 } /* Name.Class */
+html[data-theme="light"] .highlight .no { color: #005B82 } /* Name.Constant */
+html[data-theme="light"] .highlight .nd { color: #7F4707 } /* Name.Decorator */
+html[data-theme="light"] .highlight .ni { color: #00622F } /* Name.Entity */
+html[data-theme="light"] .highlight .ne { color: #6730C5 } /* Name.Exception */
+html[data-theme="light"] .highlight .nf { color: #005B82 } /* Name.Function */
+html[data-theme="light"] .highlight .nl { color: #7F4707 } /* Name.Label */
+html[data-theme="light"] .highlight .nn { color: #080808 } /* Name.Namespace */
+html[data-theme="light"] .highlight .nx { color: #080808 } /* Name.Other */
+html[data-theme="light"] .highlight .py { color: #005B82 } /* Name.Property */
+html[data-theme="light"] .highlight .nt { color: #005B82 } /* Name.Tag */
+html[data-theme="light"] .highlight .nv { color: #A12236 } /* Name.Variable */
+html[data-theme="light"] .highlight .ow { color: #6730C5 } /* Operator.Word */
+html[data-theme="light"] .highlight .pm { color: #080808 } /* Punctuation.Marker */
+html[data-theme="light"] .highlight .w { color: #080808 } /* Text.Whitespace */
+html[data-theme="light"] .highlight .mb { color: #7F4707 } /* Literal.Number.Bin */
+html[data-theme="light"] .highlight .mf { color: #7F4707 } /* Literal.Number.Float */
+html[data-theme="light"] .highlight .mh { color: #7F4707 } /* Literal.Number.Hex */
+html[data-theme="light"] .highlight .mi { color: #7F4707 } /* Literal.Number.Integer */
+html[data-theme="light"] .highlight .mo { color: #7F4707 } /* Literal.Number.Oct */
+html[data-theme="light"] .highlight .sa { color: #00622F } /* Literal.String.Affix */
+html[data-theme="light"] .highlight .sb { color: #00622F } /* Literal.String.Backtick */
+html[data-theme="light"] .highlight .sc { color: #00622F } /* Literal.String.Char */
+html[data-theme="light"] .highlight .dl { color: #00622F } /* Literal.String.Delimiter */
+html[data-theme="light"] .highlight .sd { color: #00622F } /* Literal.String.Doc */
+html[data-theme="light"] .highlight .s2 { color: #00622F } /* Literal.String.Double */
+html[data-theme="light"] .highlight .se { color: #00622F } /* Literal.String.Escape */
+html[data-theme="light"] .highlight .sh { color: #00622F } /* Literal.String.Heredoc */
+html[data-theme="light"] .highlight .si { color: #00622F } /* Literal.String.Interpol */
+html[data-theme="light"] .highlight .sx { color: #00622F } /* Literal.String.Other */
+html[data-theme="light"] .highlight .sr { color: #A12236 } /* Literal.String.Regex */
+html[data-theme="light"] .highlight .s1 { color: #00622F } /* Literal.String.Single */
+html[data-theme="light"] .highlight .ss { color: #005B82 } /* Literal.String.Symbol */
+html[data-theme="light"] .highlight .bp { color: #7F4707 } /* Name.Builtin.Pseudo */
+html[data-theme="light"] .highlight .fm { color: #005B82 } /* Name.Function.Magic */
+html[data-theme="light"] .highlight .vc { color: #A12236 } /* Name.Variable.Class */
+html[data-theme="light"] .highlight .vg { color: #A12236 } /* Name.Variable.Global */
+html[data-theme="light"] .highlight .vi { color: #A12236 } /* Name.Variable.Instance */
+html[data-theme="light"] .highlight .vm { color: #7F4707 } /* Name.Variable.Magic */
+html[data-theme="light"] .highlight .il { color: #7F4707 } /* Literal.Number.Integer.Long */
+html[data-theme="dark"] .highlight pre { line-height: 125%; }
+html[data-theme="dark"] .highlight td.linenos .normal { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }
+html[data-theme="dark"] .highlight span.linenos { color: inherit; background-color: transparent; padding-left: 5px; padding-right: 5px; }
+html[data-theme="dark"] .highlight td.linenos .special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }
+html[data-theme="dark"] .highlight span.linenos.special { color: #000000; background-color: #ffffc0; padding-left: 5px; padding-right: 5px; }
+html[data-theme="dark"] .highlight .hll { background-color: #ffd9002e }
+html[data-theme="dark"] .highlight { background: #2b2b2b; color: #F8F8F2 }
+html[data-theme="dark"] .highlight .c { color: #FFD900 } /* Comment */
+html[data-theme="dark"] .highlight .err { color: #FFA07A } /* Error */
+html[data-theme="dark"] .highlight .k { color: #DCC6E0 } /* Keyword */
+html[data-theme="dark"] .highlight .l { color: #FFD900 } /* Literal */
+html[data-theme="dark"] .highlight .n { color: #F8F8F2 } /* Name */
+html[data-theme="dark"] .highlight .o { color: #ABE338 } /* Operator */
+html[data-theme="dark"] .highlight .p { color: #F8F8F2 } /* Punctuation */
+html[data-theme="dark"] .highlight .ch { color: #FFD900 } /* Comment.Hashbang */
+html[data-theme="dark"] .highlight .cm { color: #FFD900 } /* Comment.Multiline */
+html[data-theme="dark"] .highlight .cp { color: #FFD900 } /* Comment.Preproc */
+html[data-theme="dark"] .highlight .cpf { color: #FFD900 } /* Comment.PreprocFile */
+html[data-theme="dark"] .highlight .c1 { color: #FFD900 } /* Comment.Single */
+html[data-theme="dark"] .highlight .cs { color: #FFD900 } /* Comment.Special */
+html[data-theme="dark"] .highlight .gd { color: #00E0E0 } /* Generic.Deleted */
+html[data-theme="dark"] .highlight .ge { font-style: italic } /* Generic.Emph */
+html[data-theme="dark"] .highlight .gh { color: #00E0E0 } /* Generic.Heading */
+html[data-theme="dark"] .highlight .gs { font-weight: bold } /* Generic.Strong */
+html[data-theme="dark"] .highlight .gu { color: #00E0E0 } /* Generic.Subheading */
+html[data-theme="dark"] .highlight .kc { color: #DCC6E0 } /* Keyword.Constant */
+html[data-theme="dark"] .highlight .kd { color: #DCC6E0 } /* Keyword.Declaration */
+html[data-theme="dark"] .highlight .kn { color: #DCC6E0 } /* Keyword.Namespace */
+html[data-theme="dark"] .highlight .kp { color: #DCC6E0 } /* Keyword.Pseudo */
+html[data-theme="dark"] .highlight .kr { color: #DCC6E0 } /* Keyword.Reserved */
+html[data-theme="dark"] .highlight .kt { color: #FFD900 } /* Keyword.Type */
+html[data-theme="dark"] .highlight .ld { color: #FFD900 } /* Literal.Date */
+html[data-theme="dark"] .highlight .m { color: #FFD900 } /* Literal.Number */
+html[data-theme="dark"] .highlight .s { color: #ABE338 } /* Literal.String */
+html[data-theme="dark"] .highlight .na { color: #FFD900 } /* Name.Attribute */
+html[data-theme="dark"] .highlight .nb { color: #FFD900 } /* Name.Builtin */
+html[data-theme="dark"] .highlight .nc { color: #00E0E0 } /* Name.Class */
+html[data-theme="dark"] .highlight .no { color: #00E0E0 } /* Name.Constant */
+html[data-theme="dark"] .highlight .nd { color: #FFD900 } /* Name.Decorator */
+html[data-theme="dark"] .highlight .ni { color: #ABE338 } /* Name.Entity */
+html[data-theme="dark"] .highlight .ne { color: #DCC6E0 } /* Name.Exception */
+html[data-theme="dark"] .highlight .nf { color: #00E0E0 } /* Name.Function */
+html[data-theme="dark"] .highlight .nl { color: #FFD900 } /* Name.Label */
+html[data-theme="dark"] .highlight .nn { color: #F8F8F2 } /* Name.Namespace */
+html[data-theme="dark"] .highlight .nx { color: #F8F8F2 } /* Name.Other */
+html[data-theme="dark"] .highlight .py { color: #00E0E0 } /* Name.Property */
+html[data-theme="dark"] .highlight .nt { color: #00E0E0 } /* Name.Tag */
+html[data-theme="dark"] .highlight .nv { color: #FFA07A } /* Name.Variable */
+html[data-theme="dark"] .highlight .ow { color: #DCC6E0 } /* Operator.Word */
+html[data-theme="dark"] .highlight .pm { color: #F8F8F2 } /* Punctuation.Marker */
+html[data-theme="dark"] .highlight .w { color: #F8F8F2 } /* Text.Whitespace */
+html[data-theme="dark"] .highlight .mb { color: #FFD900 } /* Literal.Number.Bin */
+html[data-theme="dark"] .highlight .mf { color: #FFD900 } /* Literal.Number.Float */
+html[data-theme="dark"] .highlight .mh { color: #FFD900 } /* Literal.Number.Hex */
+html[data-theme="dark"] .highlight .mi { color: #FFD900 } /* Literal.Number.Integer */
+html[data-theme="dark"] .highlight .mo { color: #FFD900 } /* Literal.Number.Oct */
+html[data-theme="dark"] .highlight .sa { color: #ABE338 } /* Literal.String.Affix */
+html[data-theme="dark"] .highlight .sb { color: #ABE338 } /* Literal.String.Backtick */
+html[data-theme="dark"] .highlight .sc { color: #ABE338 } /* Literal.String.Char */
+html[data-theme="dark"] .highlight .dl { color: #ABE338 } /* Literal.String.Delimiter */
+html[data-theme="dark"] .highlight .sd { color: #ABE338 } /* Literal.String.Doc */
+html[data-theme="dark"] .highlight .s2 { color: #ABE338 } /* Literal.String.Double */
+html[data-theme="dark"] .highlight .se { color: #ABE338 } /* Literal.String.Escape */
+html[data-theme="dark"] .highlight .sh { color: #ABE338 } /* Literal.String.Heredoc */
+html[data-theme="dark"] .highlight .si { color: #ABE338 } /* Literal.String.Interpol */
+html[data-theme="dark"] .highlight .sx { color: #ABE338 } /* Literal.String.Other */
+html[data-theme="dark"] .highlight .sr { color: #FFA07A } /* Literal.String.Regex */
+html[data-theme="dark"] .highlight .s1 { color: #ABE338 } /* Literal.String.Single */
+html[data-theme="dark"] .highlight .ss { color: #00E0E0 } /* Literal.String.Symbol */
+html[data-theme="dark"] .highlight .bp { color: #FFD900 } /* Name.Builtin.Pseudo */
+html[data-theme="dark"] .highlight .fm { color: #00E0E0 } /* Name.Function.Magic */
+html[data-theme="dark"] .highlight .vc { color: #FFA07A } /* Name.Variable.Class */
+html[data-theme="dark"] .highlight .vg { color: #FFA07A } /* Name.Variable.Global */
+html[data-theme="dark"] .highlight .vi { color: #FFA07A } /* Name.Variable.Instance */
+html[data-theme="dark"] .highlight .vm { color: #FFD900 } /* Name.Variable.Magic */
+html[data-theme="dark"] .highlight .il { color: #FFD900 } /* Literal.Number.Integer.Long */
\ No newline at end of file
diff --git a/_static/raft-demo-examples.png b/_static/raft-demo-examples.png
new file mode 100644
index 0000000000000000000000000000000000000000..d9dcaf61d933e84f762ae1a02fed8e1249fac63b
GIT binary patch
literal 324598
zcmeFYWmH|wvM5S|dkDc<NYLOe3y0wD?yd`mg$4<(!JPmBg1fuBySux~0xsX)=iPh8
zcyFJd@7JAU%vn9VyQ;csR(Dr-*9=vV6GuhDM}mTaLY0*Gt^@@Iv+{m_Mu2<Iac0<`
zfr3IIw-gakkQ5OiQE&vAS=yLFK}m!rB*CjF590sWPb4QNuMvdF3tWT2$dfLQrBSWG
z_$4SDN=jO|5Gx8h?;nPsW+M81vKFdpp(07N;yxG_%T$+cK>-~1vB_i2bN^+3q*ed*
zai1}+(a8)-A=)tf`(hy+6n7V|gN+50laV7*+|3sfL1>F^F#{*^uW!l82vDX6FY;hF
zfR;$35##l()0;1rEQ5U42dGfL^u7dpM-pNvD7W2^9x&7=kKb!HVwngm7-onyy-Y#_
zdn{uL70nUTUkFL-L!(gx(~*84C5&R5LuFGAho_7W{_vSe6mtFI2rE2L8sp97#es8W
zK9K9^m@lGMA7qnpmys3W`yfMbH)v*#Qk{d{SO|kio?~-Y^5z75t=}<`4mL{e6x$8&
zHZo-&-lMnnXo&b5GmbQ!lS67P)JC&0bZQC-WJHx#Sn}PF__&$=0S<~fVHgc5*am7*
zB^iMAd*(S0Sy&@xb0zoEpI5ELh;9f=K3erj-295X(?Mhed)Ro;Mk-Ge@l`4-?SXeL
zW<2qukJqoS@-<}oF}Sdrs8_m@YB~*5d*BZOOjZM8bdd?WC~r<EdEp{PSRBdVD3d~l
zkvK*J(Qn-T-HH3T$Y3iX0e+cH_~Cs?N->n(!ChEb+M_Nca2Jf}t40bnBZ(S!ome~|
zx29hUy!e}?_1liCq|eI4LK|<x!k61E!FU|Z@&VNOF_Y+M#hzYUAs7T0p&y8O6@UB=
zNiY$XJkvZj8Aat044|0@pm+_+XHX&~e%&K%vZ8(Y(N1#t!^B<ZM@tg?)$iBg3bo4`
z_`|C;ug4I&3m8W<1o*{^dKG>tE<$<?-XQ!=`AKf3d7A|jZ-K+)BN3nx6LScz7z*jq
z##00O*-!C~0%9Av2A(tK_<2sxE=*ynP0~&L-{JPwso4q;kv*ADh)o7jq(etjZRKe7
zfqJ?ZP%8va!EFw3W#}iDl}x_`&XA3p1yAoArGjZ+Ig+IGO~hHGh=lIQwBO?WY+T#Q
z_5Z3@t2!y}C2;U+`*o`ccpyUTUK=g|uelU;Fw4;yol(a76>24jOqIDBtgaY5blaP?
zml|yxK{>rcPQ(5_Fao{Cw8lebvRfMG3)>Hxi||+Otk)*Q(i_f`bd@9lS^YP`&Fs;G
zpB)Lsm@^17(m%wUzJ!(TqKdQ?kVRatJ2&=2T~PZCMSy@DD*<SHAAY=^rH!4i>)g?P
z;9BX74ZFqy>fDh)At?mwhY>OPdnU1$3?MdqWZZ!F?{xo!x+TPtjI^JQKK~bY1j?yH
zSOG<<Q>p@6#edX*VBv#nfOj@xjsNxrTLToZgWVp!EWz&L<Bgx20hG|E&%z=HAw)(I
z8YHeEm|IZ~B%Km4pTo(?{z|}o{hl{KXo8^>7EhWmpq7XB2;dGR6dxna7@*(zFcr`u
z#{U~vO0-r{Wacv<vMKk~f(t)dPKb3vrv{2%I4Kv)g1rgZ_Iq?L&V;oE&eO-!eyIE~
zW+u!LBB%b3l@@fc()~RPyd^NS{Y8~<C9tp4>J<znIL{%N4#2UV(@hW5ZcKZ$%<%E<
zq|Jl}v=$73Q2kEg4etl_kC6s~6=d$8H3CBeQMyosnSZBOWZ1{pBl+;G2D*L!8Y@)u
zwW$-tn361+E~_K9BC#UED|SN8|Mf0Tf}HOcU@+~AtWn&RL~q`0o^D=l-f*7X6g-wH
zmN6Dp!l#5USihxyOEnJ3C6Li1sb&_T#IO%wo5(w|rATLr>Is#R*2gf$HV;C7kqyJ@
zP5UfAm`9hVqJU3bK<)Moq@t=StIDTBRfwd5Q>a#Wrpj0RBzQ}v8g3<xnkSbp^>_Gh
z?Tqn^_6#Q*{^+kgap~W?lZn$u7E~7Zmc%m$R%ljq>}5+{wKTP2wZm47vx7(2N3Tc9
zv&b{Uv-idNO6RjA)*4nYRu$}fmMWHnQw_f-%O&RtbMZ=3avNow3ih~1&9%tsrArOy
z%hY1$I7%_Kq7QIaBZwsSCvYaM4+;)em%iK5+9tZ@yZ*hiJ9Rzf2he9!0z-kUE=A60
z$Ii@5K@<VK17;-#C&oJFUigMwUA{1&o+%7eGKOgPavr65`J@8voM!R+)H5o+e7y?2
z#5I~$%2v;(hPKjIm>0aKv>W-u`=}YBL-t>+RVF)avGu45(AohS0gG_UC=B>F1o(JC
z_+cpX1er`H>_)hcc&7NJ%s`_VV;RmnW^CMYW_~M3`rqywb2C=jWU3TzN;pbGN~O3`
z%2P_{e6@VU{Ga)rGUq9ZBP*j*qjSl%DYq#EY2vBl%;lWzwnBCG^&8qH)m+AxYg0ok
zgN~J^al5Rvy0nR^eyXjry4I<+<t}9|wYRDSIg!n>iKAj^XxcM#y@?=gQ_n%PTBlmf
z1>2fqLQ&jZW+|>Bj~P_YqSkirHt*IKX!szJK#|`faoxj{d%4PD<@nxaH|pXz{Wksh
z{XE7seG$aUT=}LW)>PV5QOXI^s^+OqA=h5}X8Y>Kr+rs_cKDvtKhHiNRxSJ)-3>2J
zD_*GWtnD9Np`W!4SVncTy~n*bJu~M+Ao@Ub&8Nl3@6qJ32R^%oEDx{I4qF$@S39U}
zyKh_d@m+D;nt6h5&#nxg@}6ui53krSO`p}C`EOXSx1Vh9J&!=!S%>YzZ>B$te{gLW
zbUX`=cQpF-`&VR7WdF*(Gbr5%+!)$;Ce0_sBfTZ{{>|{aS|s`>Mpsw2+Rs%3yeh5|
zpxU~7{iRcX7D_=-|4;QGfryI8PvSM-b7Hs2)Wz2#t3p1Dv5N*+sNEbLcrvRoH!}Om
zkjRCkD6nrzo+K}DSDJ4fI81N3_O*tihv;MYQgs4_QYR7%Gr71c87{q*6qGcT=6fbF
z_vKn*O0wuA?Bq!QkmuXTA7*{wwLHx`8)S$Jr_#?dVmIB3cT&DBFq7xwmG#E=TzXXz
zwo*D8n;Ac`aAj{6h)k#>Ese$;(Ci0y-fov|-d(hSO<M{M*A7h&$5~of)y!`$ydQTV
zdP3RRN*6S&bWTot&m{*XOC_Brc_)0_G~9o=g9Ej?=r#ur#iq^W&c7^%hKcGnW;R+o
zJ@;&Goqqn#B07~`fQr6|^Ji{KKM(!N`>Oa#U_>Il(l>ukW^*LG+Qytd<F<w*Z=-(k
z!Dj5Fj+@d6dkoAp?|U`}9%!in+g^|E;f$W^eQGl37ar;=`RR)d#7@L&(~Ykw{4-Et
zAvBq*FYNo5QH>TGXI<l@*8TdkSEzyHv(o;b<0K#Y&BSy@7dm>zmzId?6G*wSPw9h;
zJ^1PB4y%loqm;btOIb_XklSn&%V?E<m3T3kI%k<-**DE=*VCHX{h|Iuke;|EmF~PM
zhT>$)n%B$FZ5Hig!hF%kqB|R=g$n2H=g5`FB+f@$s|A~eM(67SdVhM(`XD>olDZOx
zb+MJi_ESk&qtTT072o3Jvi4Tz=Oisd+xe&k0M+rz(UB|uiH>6Qob<7$tL)lrC69%h
zQ-F8(9A5<AAJ0y`!?pePf>$4J{fD-?m$PRAG!4Q9`xu{!JPgb9*@gKY(2<p#-5D?2
ztAcj8`f#TJ;cKBSyJab`#L2xEV<<L<EI=GN>W&x_aLkwK08w17spFxd%xv&*>>9jN
z-Ygi)MC10}ig-)OHE$iS%e>}wJ7{@+%ttQ6!DnpJ;e99=&VpaoT_#{`)U`UhIIuV*
zoCohzo$qd|`<JD+4K;aeYR%M+9V5<z){?z)k16l>s<jvE^;XB)`RvU*b-Px7dE(y>
zofE8-zb1{g-#ItB0(?H*Y;1Km1iKMA5Vg9~fgz2}2Z*PVkILHxF5~H$wReFx&b(PS
zEzds>dIdq3`l-IQ?v^iC9qT^6`X^}%^Dj~mtBczzP7jdPdX)F_9Pza8o9<Y%-#rEK
z4gbl*!RzHnL7u!kyqOoJciDr{^`Bkqhxz6J?tp-=b;<MsNCp+Z4-cVQ%3yyaqKu;f
zpcE*-otZfbeYuz9_Iv9Idrs5Q`A~QZO=9=~3gaT83XubOK;<D3(7F8rb#%hJ68}}w
z%~k0kr;7l2vwLHMKm<yH9?A;?9SzNfkPxB-9oGumQK{3Lp+H&1bASzx=i=NBBQVtd
zL5|mfisJ5L#rv`W(OyE+2?`3E;-CA2q!RfB6x4?-OBD@g4OtnWG02v|&;(>;%HVEm
z|6Uskiq{?Zp0qV}HY9PkwXt&oy7PVgR}J8M`k!pZuO$Df;%v?LRYO*RL<Hn$O2W>-
z#=!KIABlv7gxAr;45;*7?7z_8-}t^-I6K<|85!N&+!)+g89<KajLh8J+>A^tj4Ukl
z?=|S1z;@1t?(}v}WdA1e|Iqnv>SXL_Y42<avLpG2uAvdg#hLHx*MAuO@9W>~G<CQ9
zKb-8G{!6TP0U7@(VPs}tV*KB<-%)x0$ptD{x|`Z)e7CeUwR3vs!OzUX#>V?Eg#TC3
z|6%#xP&NM>Dia&ae@FdqMgN7$%lMB7|4pQS>+4^+?+W8b;${5b%I8PIE9?+{{~m;v
z-{n=_;m3bI&O1PSe^CD$zNbG1v6|VqZ9zc^LP>rXQgQ!q(u$g_Hk-iU{z-)?HQN?A
zB{_Z|<IzXX=|m4>`+PUz6t_PngEaG9<hOUT({m_Nz7xpfi?0WaY*{wBa1X-dcZ59}
zg}hvBL=v$-VLn<#Hj+g&MnywtAM3PK*9?k^dS}$9m(n$l)2r|FJhQxg+Ruw=OjP~%
zz0BWAU&?Cd^*m48zwtxBS5ycHAN_xjko+gWE@TkKwPNL;Yx4C!Rrwb>3_?h^ER4{n
zPcSeagarS;K_ThKE&F29<X;K@ncqLe&^TlNBaMI2&klO$RvKg=rS?BH-Tw#I%6GW`
z+?;n(e)wNW(CYF1ziA@=k6PHW$He=eB>Mzo74pu_3L)Qw;(t`ryOK=)gLeLl{{L0d
zKbG`=mGu9cB{5kN+8KjTJ-W|-qNFgt$bj}n4pVu^0@gNaoLV#s<R{m6WZ(O<5kf>{
z1b7;}9NHU${-3&?0tkk3VUuIe#r7rB?Kx|Gi*NU>SaUv(;iBxh{B7~sYWk<%^RICa
z&KK9x`*B9EL1Q~E*4O8Q`QM2kcm3x{g0+&Akxnj_lhM~Bt@Gly*Ibvs-x3#Z0+L#S
zqB9aw-bSX!FY~jwr=NG)Eidr*#_sqpD_L2mB?5t8nK-s;1!hlHUwYf`d^m33u9~9w
z4|SO`2IWe_d_W_|o`1&)(%|sg3bePJANb%5Gjm;n4AXxwpQj5nk=SQOw_H}ux0xs1
zcj&Jjv9~_d-S5h0^^rxL^Gx_Y{Mm1v%g}D^rHyje1=Y{MchgwSHos`E74Z9}aqjLg
zx|as0wvKz*h+#Ei$pbI0v)s{3+LpB+7a09}P^$@AAN60y*JSlm&PTteaACX&+QI3A
z7mwY`Mwi`Rz7H_p{9g7sSgqZH18Lv4!w@E~y8m=LRabY{Z@p2D%Os1Ff{#GfGG2=t
zF27@V>633gyjBN$3pDof)Cbv`=eT@F5k5$JdTMID%V-uXZ)5p(YZb8<h#rTw3<KPC
zLjB}UJZOtL5km*#*M*So?(O*@#J=Uji_!A9R*UHH)=SO}J8T7SX3i7l?fo6&3k6li
z7PMMtJM1v$V{1A3Lw7(hWOelZ3CCSo1-I|9LpAy-AvDJ&eHd_LeVxRXX*Wa6+~Cz+
zek6vN_!ywd**`@7y5XcZ&F*8|Jy^*#&CoIl;$hO_Tv|)QBj!7)uXB0oK2I#yHLVz5
zx4A?~XZF$G)p(SdR*HI^0;RLC_xUoAf!EaE>Yt6kye)5TmK5{sHv6d6C>}>)hKH#t
zklRRe8sPC?WF2x>erFejyL5^)$elR{4KQXs2~?D&=Qzp2-KGlZ2XXaxe5l0L=O()3
zWXr&supiECepx*RF&}^%KwC>H*wH~t?dB7}*9B8WjRJe=;5jutN;PrZkeA`%An#yE
zM!LsrHgLC}yo=+f6k^e5KR5Je>4HqFgZyo6yW$>^U&tR}F}Mt$?{v6SAt#qXH|&ks
zyl7-#OcAHv^%-2WFB6}vmoCD&@Wh`XvysF~fb|GzdBR|AC^*~_b&aved?0Wo<U?+&
zfKoCU=wLmSul*0T&$xX1-!^3~{(8Bs51C(mlv*MJ$Ku}mPqQ>v|I>Z_BW+pQ@s5yg
zq)5!5wOklsnc(xOHRp3FK-Xu!tt=T64gjUAi#vCA?scOllIReFH}m^4pa`uN;Eru_
zH7V5Y^c*lWsIQKK{3|~<V9e%;@8^7%gR^5~8}3nHoLuwdc{pLui@dMMq~j$p_;oI@
z9kdRIcJmZ$@0e#8)MdoG+pr%yf%w#UOstH1-gcTBmI&WWRo;z?n5hC3B-Pl*Db_Ey
z&Ej?dehpe`#<Cy?hOFI{#_5#zQOiFtg;vOBEAY?5`57iZh&TQj>>ad3pUE%#6;7gn
z4kI_v#zz$+0+05XG0vMl#VDm8hdH#(uoTF(BBl@djsGRj)$APYsTjX2L60h@Tb7&i
zYyZ4E*NWGfiXmG4UeZYC^4XsxYNB7$wnunwPXYb9n7F?FJm+u1KXi*usA6WX2e5>e
zl<~_i${$zYjd-c!%!L&%$75~MZ~T_GiVx@-7oV=$wGItMCn9m+Z+<CnZ6uacKOLqe
zSP=T66=C7I^j3u719+zPL7H#SzS6^MMQUQ$6n<tqC#roJ%V{s5fazLLgB&eQ_-CNd
zB|1OL2S*fdveF)=^zsYu{U><Hxqo+a>TGpw;MD@??m=CzcCb;{+wt&+)*O(#oHUuK
z({*Lafr(pbFeY>g!*XZpd#A~4v1QG5#b5}1!TERJb<c5Cc3ne=PwX<LE<@g{B|^kQ
z@xHFM)t0~rJNHZILVW?LZIER_YO#KNTi^HLS5<3W(7t~8I_8%BacCVCY!tr6w`o^$
zaEDki?MbIzD^v>GqZS%UkJVRzU7d(YjWUny;AYVH(%6DLzTHej%<HL#OSbf+8MX@E
z9eOMO7qNg-$vc_>C9oMbz8Dur1(kiwYSVV4JCd}XEw~e=-v>|@pHq<;YOm7sExIkw
z?MLX~+WCiwa;lX)ak(|QKtf_;+SvX-`U9b&yjl%>UnN~BKQ?nkUj6Yw2+=DguKUGG
zDtJe|h<(!vQQoQJb^UeaE(8^g^w(@8CQ<L{j_up0mtR40S{kP=-ddKmsA;&t1Oyhj
zaeGx*Jn7ZzhN|L_T2%7~NjOj2&E)Tmm3|D|HGQTAyVN+t^{6R%r3J1nl7R<-aYk9T
zOuI9T#q=J3LYyv&QpZ@sgN)l>B{%GyfCKbe>5fpvAp6<9REy%}ouimP(9T!H4^ncg
zn08a$h!T+P_wjnZeS3<rr#RzSdjCZo7D)ClL39G8^YBSYJZi-Rr8p!JP!2=I3sz&}
z8E%^gY-xBuyRVQj-3T{W<eVgv(zf25T7LcdM1h*oQ_*zr)H7v8#6(&jf5-o%zE<#h
z^}NwP>H9u#jy^wUt!(OUHGo0cr#x$K({~P_Y2Bp0b0M#N+4AGZ{nmZo#beYcl@`3|
z@tged^ur7{p8&T(3e-;9kfJXQczNaL6;FTl<X&U$L$BYU?DPmPwTP|n?jT+r*Bc)@
z@L0!h`8^kLTBXP4>+R7XUeWcZ=C*4a3(n^wPGh(HkNrvRkxp+^30Q3?Z7chcE!JNC
z#M7<U{N9;vN7{^@quN<FPius5TITMo<E<TOzKgu)IzD9ID828ATdfOxmpmVm#}MGu
z)}qUuR(Lt@C|8skho77E39o2k3gB_Xe2&&RJ=7st)!XiiCX4NWU865dCK1}?wBEkQ
z8Fz!lr!zt=;~5b7%O~i*{SST9ehp=S<(7ii-~G^b;H;*b@fLyRo}O`N{#*Y0_<D23
z3+IxL6CVTw);3<6$04C^uiSg15q2efpgAgjjFt3-X7l3ts@!RUv=8-G!u&X|!QeHl
zBrl~&UGR32-f*36N3`O`Jze(YXOP~u&GwwA8(!}Ncg6nmnz+ETdXbKY=JedQ_u0v}
zwHx%T;|%lJ$SrbyK#Tv_^zlyuj&b>xKVLwqgf}fI`|j0a9F;P3!LSjn>|0>-@rdCB
z(Qtak_}+!IfHm9=Ud)Bg+aF=*EyP~86`WT@`A{j3Tmv$5^gi~v3mXIhLaB$-W**gJ
z83`Fd2ukD3pzE$^u)fDheJ6c(YwmK(_2u1ctPtp86eDKoaIGXdV}F!am%_dsYLyHG
zfYXJEv|n-#@y+KvPER>&n4GVy&qj$rH%$kl83dRbykH&ydVJBBzNfapaDAl|<Th~J
z>lYr${d&z86EqaQmO$Jzu?$x)F`nGMMzGL0kcM|b`IEon<RZ;@O~&AAfk&^c$}c#0
z{lag(&{|PU2rlMrUkd^c@P1GY!%;Kd8U1vAm|bq1%(Qs1;kxUwQg{--9m96r4kw=<
z0Hm3RW1b%JnP`9Gm}$3%qq#xG31SKGLv+O4a;CZ1eGAb3`8y*W+S^oEXN2-$co-*a
zJX1INHPUTUzYM8OCaZmpVCAt^!Brzg?nLU~M<vF@heGY_n!}F$Yb$#vj=GmF$B3GC
zmXi!H{{cz1=?-d)tOYc~s-Iy(_4u{fA#xU;$0Y66TTz)hL#EIa9l-C@8t$mN@FYfV
z0t;M0o3~BZn!3c<TP&|1b|g_lDkew-w~lyQgzP)S<Bq+dja`?lZwU5+-*w|UFrV;2
zecFrX@5uIU@qvmA{+)&Tsh9ufQ&Id?EPQuIz~OXv8~p@lnFp)~+|npF_mhU++C})9
zthxS9DL3}o1Z1Oa-NTptzGUZ>)eWGullUYc2j9xNitT0eeo(s{$P&He@<IE9a^ke{
z?YVit9ij)pn^~r}9ls~rRPcfrZ%i?sd4R5n6y3n!C9%bFaOYlz%PaWzYM3p+Rt8sV
z1R^rgu6am0tW*P(7<~Dgx7H82o&c`glAV_zztNGH57mg39}mQ>!4}9P4py6TL98I0
z@=ub>_3#j?<pv<n2o%LB{5(%=#?!-Aw7V>U4`vkS=cite8+Kz*+YzeVbBLQB+iN!q
z+I9ol>8U|PrdljhgpD2H_Q=p89CYKzGI6s6b<r_tv=T>o@rjmqaU(LLc#!h~>kMh~
z%icJ5`25pq`F`_d)pe}4i^oE$hv_u$3(t_}WRp3&XoXe)n((e^PH_9NU=px9G<frw
ze%T&*HKhG9=mLd?2a<zTf}&AeZ+SO@3rOB;+8v$|3UwtK6_`)@3<IN08^qLRO(EhN
z%pbonhN-ULf<g(JpY5pN6mLu}=W`mUXS;N)C&HB~dF%7e=<7U>x7)4u2;Z6rU!DoW
zM|s<dl#X0?m)^>5TEwg{O9CnQJa{z}!#~_&P!XFkEMllE#1Kn}9Li@_uLz6`7ba#^
z+)lT-Bn{>4&!pTJlw6rWHp|iU{*0@sLPqCb+KTpv`0(8_ZU4H3V{7Z_fUoc>O6`YF
z70h)K6@!+4nLMl-`tBF?7V!sC@Sm@#-R#NmZ3iYnJ9qj$9AoQ#8D5ST&?^pWzn@Hy
z<L@@{Uq1!z+u$+;Sx#|Ae||wlo!+ALFr;-b*RF1_v6K^3-V%lVF?d$HXPXdz{%4ru
z5t_T);|xsAOB*xQl-ajv2-REJY3F<U_1=`|8mIiK8GhTtJG*>&y~RsWfl8;B7bW>(
zxi|p%L-twg^Af~Gx*T~c!*MeA!$k%?df(0C3PtqKQ)(d&2JM#;JM+X5Bfz*u$vD2#
ztug~(WOw1)y*(N8>|`Itu4Qt#;!$&=A`kKF^4i?`dhKX?erF3fdP~3lh`B&!XG<EX
z6mkVe6P{Tx;w^Fc!&qGmo{(oNk$Wb9{yDew{+n}?|Ih&Q8g8r8%c>tQ9Pr_w!jN&o
zYH#TM9J=fEhp(uJyNKsiFE-ih+4E}C>7J^+CaR9g^z!`i{aJ80PoeIgLxh9zAT;Oa
zXmPf$j0$oJ`%_42Tk=&hi4zVThecyuG&y8sX0CN54o91?v_#<<*2zgAAN)F*kg(zH
zkZSa_rHM~Aw?i&}e~CnsGLCCuFa43^_}qTDWLK`kAdc?*K%3(eY!aC1X5*rt%v$sI
zH(AFJzb3%F@q=^jo~g_j)(wiW#~(6=5bWnc7l%tA`2c;%l>X!=0Un%Tp9`>AH`R>*
zZh3?4M|OruG^FE7kK5{=iUaQW3F3^pPc|?~8G(mT_x+eQt;+kW2lXe%oYI}r+8ZS!
zsKX#M8@Q@)o-}p1_QsNxszLhhWW#S-vQkK|G{5h2KVMA^F}7@RZSZidvy3|pr}<3X
zp2D6~?4Lm}MdS+K#0nqJK6WuWz!M$#^`G0YTy?Qkr<JL!OK>D)&-^a`OUV7cOtZO3
zV$;Ko_vF@Vf-CM-qhs>M;<}Tx2@o_GXLU#;ICeT}S%yTxw9Cf6Y8V={ewn;Yu8A16
z#?nm6!mRoeKj!$=(1ov>5X7flQ$5O6LL4~IX0zeAjolj7{ya8LYgj*=xyr@EjIy0W
zYEsa-ZhcodhOIGs-81jPYY(S{uVQh}-_)zW<R0`SX4oe<X!w0Gp!IbQga<AO1$w;#
z2%H#lBYTDw8K(LpaUz~puDF7@Vsbi-skYH)84o4lG<=V}#k{eV#Rn2&jfg6y#1aB(
z2)(qj!p)Lmdl<3M-M1ZsbX%VGlWU)f6SH1)?b9|TH%Y>iUId5j+HX^+UZ|EmR5Tm)
z30uOxfim&9UK_eo*6jvz6FllIDtCJmsLo^LEoa8Az=_){7VGgC*87~G1=Rf|R9wzC
zl0LKMf<Q@mTLaoxL83A@jZ))*EcAe(e(ax7BOQYP7v9`*7i!%=99au-<_Rb`a)3v>
zBM&#cq2Tkxtx$>=vfi8g8>LuqDYCy7v&@)A%i4Lj9~W@rOtWyR|3JbY+Ro0_JAW3m
z+NL--QiBY-$+8FlJpfGVGPv2EmfH)4G%7rT1?<1@dKz$@?8bflL8aR!yd0<da0OD9
zYa3}lew(XbR9Z+t?h+cGSyX?R=NMHtHGXg@=6H}MkV9@8)vxK%H$ufeH$d|#d0&#&
zmPO&oVKLn#>6Kl%ZI117QQk8p=eb7nk_MamHn;t;+LIk8cOboMt<N0=!4jS5lxGw<
zC6=JK{e)QEJ};ycdgjt+gbcSX^SnrPVB14g4d!+X0P*EUcJ`$`+CNXEkn;ev<{eJL
zj-Z^eVXvToUfZl0uN@ntzMS2#-3GnVDBU&aqPq*e_Z3?pLFD+98_03{M17fZdU6X)
zk7&Jyh#nULDg#BOoo&I*17x5n2|@?xM6cMhJOpGwZlcB;&ggz``!`wqGr2b=D+!7R
z4H_!3r7CoBOuA&R3k1F}G>d8+o_+Lcb~}xP;|PhCMvspbcXZyYq;VcZJc6UkMbOJ_
zT}Hvju{Dr`vFATA;b7cbhQrj^)9+ocq)yNN-uz?F^kKtsGH4$n#f0(q`CHyK+Of>M
zarp~f9<hz??hPDY(`FMo{JQPQ1BF%c`c`RqavsIYN2)JJEvTG&`A0t2PyHgg4S*Yi
zotAIP@F&x8fL<~&jg1pU%o;(u<6vu}v@dq>OdE6y#PGK+GJ5k>S)v@S{GXRNEu?E)
z05|XkpADdo!xcDq=n4qX>MWHEo*))mXeLW~HWeDHD`-tchc*<L)XIGV!K34V5pp|I
zi$68HnZ39C3Ce+UzuV9(IO(Mrx00Q=T?vD0p5*!Cyh+v9ZsLkSjE2FHcYJ^Pf~b@F
zxbSz%<03B`Tjh8!f6G)|I$(%1ZK51{ED68Ju?*gNsR_t|aSpAkAoQq-@&WQ8O5jfa
z+(MA3XQizGSA)<#z-x)HTELVpih6?h*nb7Y?ZQXNV0GPuTrxh7=+I!Os#CA?3+jpr
zU(S2dKu`oK<M%ALy?H__%Ln)rkKWJjg^TomRj@gQpD+oz*qHjzI$Woci&t(wT*eFV
zqn=+bKdm5!bpzDOmW5G8VDVX9gp-Lf6imnswTN7~IosGm&Nzo1glla1vOyW&89BYx
z<qvu?J#@?SrLA^44Hg_Z;QC%BuHo{%#6BSYss2&OYb<DBDC9uo&cU{#1L@)#8h{Px
z%dF)EkI{7h-71nN^$`ZFEf@jiPJm5#zpiFGY0Fx+ztzUUAInFLxwm}Qr_)-2mSojm
zBPHWY>Oz8Xl-PjsuV-FsYjhY;7dJX{UFP6{s`k20;;4Xr428SwKFID!(2m?QNU8tE
zK(KImCEz23(jRz$LTJD%WO6_dMSPGK9F2l2_gMr7<`Dj6SGfTG)bXs!9=2fU&bHb%
zMWaaL45v@M*)IMEoJL4mDLk{>IHuOwWCJqFQK8%&jk1kfLzBJ8w5E{mKul*ELt{1W
z@)*P31uNl}Abe1LC0>2N2v-<)st}Iy8r+#8EVu7=gUHK5N?v+@aykzX8${7h6Li2&
z*E(vn@<3cLsAnlr`Rx0I`ln#KSaWl3gJ4W?lWM$GV8#vhQP`9_i`yjFSX~1CJ5FCV
z?;@=MIBodUAP6b`m$3w=m&t;_gMTd$=cb=Q!&|I)=A$n9C7j+l2cdIa3rQ5brsrbc
zlM36$qsdmVVJ&5uJ%Qjeby5_r;wjC9!-U)fsbExQCjHW7gts=a(^q?N^(|CCj~;QU
zdiW5S6oA-(0bmB0=rVjqyI=7N7z~hk(H0|0Sv9&>w&=$0eRuSNyaR=MnL4~8JIR8&
zi8uRAqbPqV=ry1Y*ve>|EAdx`!c4$hBrUnZAG!V#_v$mH&SN}m*Xi#AiQmtfXv6|x
zo*EDDKQ_tmA7n5L`LH2Od$g+6Nlv;ds6`vHA1#h6@l`}~QWB>)VS-HtP@Op74pB-R
zYGV&AI4CtrLOn)2b5j?HE{WSOjVRpeWEH~H!s~Jbqz}ER^XN*4exuZVp1%bjj!6lm
zJh_&UT`e1uy*BmP!zTZQ{u-F<ZG0?PCSdaholoSk=_z#=(Q<vWZTf^yjmg;1!6Ep@
zAJ~SK8t;B7zUMxi8nfuVPay=<fpmC(I|$L9wQm}VTh5t5-#LvYQ&A>Pk7m%-AKH_p
zudMUK+_T>%QvUoS&zwoZxcW{f`5P)R)H3QS2h&3Y&L7{4{dVPnCPKEPQZv>dpA{iv
zt`-t1bv)Yp<m%J*hJ9oyg-|rplZrkWL$3<A3CA^DR3bCbV0Dyb0G9BGyBd~DE7wpj
zuNj@rnJ+=2xf2z_<WKFrlhWd{r#P~W1yJTZnh<Vsjo@}&(gpsl5O1lrq9SF}AjWa_
zA{$pNtG|UB^auj967`o%aWkY#%pr$CQ+-jMZ@MQEFMEaczgqzB!KrtT@*POS^bJcC
z<uY7Xq25&%gT`!5ys2}1IK#iL1NX%<R97^AY9_br0R&5+wgZT8b5M8*bx*Q$QSjwB
z`AH}O;HMWId1hV#vwdKI$|}7$dm`t|EB$(Huls>zYfi%PM;VbPk{E3(9NbS3O_$|z
z7eKFs7faicOK!_-Hp)>|e`-cP*D=BEoaWL|prIE@bj|hRDUt_|q*#iavRpIM^+kJx
z9Qr{;29wthT&J#znlhlHuGx0fPg=d46^(^olYUi(=xq0E*N=i6HoD!#2{)Odm~%;L
zbBl5CH1Zq>;!h3y!=pF}Dccm#>N1^`v^mhgbM$99V|(n_9|2-(^1Dt<$5=Yl0;0RP
zmwRJO#vAZx1ZCEZ#yyER-5AHX2#T$-`t>&9<uFpy7#wSEJyXYsc-oZ12MsX*t$Tw^
z(I*}X0<bf=%tGw+imM(efY8Zt3;qUcLmR)YC+X|ez;Z>a)JU-8=XjAMW4a#pSN%1!
zN?#UuF-o|~w7y&{-+VxEhB!4Sza%IrWz$XB+{&l<WK4hMQQ(2N&K2E2_Q~*h9lJu>
zbx|97dp5#b(eZOGMsDI?k@zT{l?vjI`{=SQlRG?pJHEFSvMMiEHPD41_7~=|g+G-d
zw$GaMjLeJBP{CTYw%t@YjhI%Z!Q4JySHQ&9XF(HHA-0J9c#!i2!G_PQ4BfjHyI>sO
zdB>q+`NtTa(Bqe?8Y`QyqurSNlN%|!-@0-(>t$4!pV}d=8e#y=b;Bb;!g3$|#g~<`
zuRf0A3CM3+_v{)B0&qzETULH@-)E{hUn_G`j-Mv=-z2=<N?~@9)4$Z2J{zOPTt^On
zDPRskffV-bxnOH-M`Iq?cc*qt-H4wljP&%y+{oL)hCnG~NW}-e`&@NfP}kM4sRTfk
z%+B|4Wa<^dhT|ZK%^(bTwx*%%Qx!!luX#Q9+VnM{0HPD2$2lZ{T2vC;?9`jZ+k&eM
zdGSV-{tZOHd`lngOIl*th3`ZQ>q6b;P`|2gZmy{O>xDhnCqI_QFkU)yJLgw~-Hg<%
z0JBJ}J%cU3$7RSCDqgqkh|Vs~7L`J*D%>pl{j?$|?Q8|CN-&wzYax#acH-3uFd`az
z90WK_r<M6t6O^mGmFXJT&DK4p3nF>VvHS`oJ+Dv@Qvq~-TDRZ+*#4o8bIF|f%66&+
zZN)Dl$d;AZ;>YCk4qEGJl*QiebxXx{Us`%B<=IG%b0=n^2}lzi8ow@M96|o6^TCP$
zOXyd`tYJJoXCbP&Y5Cd6G&rG?Sgc^!U>6Nzr<^&))Kb@F$9BDkuLO{!Oo6G`8;#0~
zzHc^izfZ4o<fa`b=2cvt5oIs!VK&)=%8xS|&*=93LenSH7pPB7)zA8`J~CVmEmQCW
z5uGo(^QAa|Zf0aQuK_r370~Q1B$l)FD~-Qz_481Txl!(;1rwkiRF*bBq?V9%c@%7q
z4%KBCQ#k8r7)fOs2KKyRPsa8!<7dm4?|OITFw(RHv}J4$nXpNoWBLs)B*u$VI!Vm0
z+7qP`!T8y(OckjSe?~o&Uqv_YxnrjCx?6WH*(8_eV*XzIR!F$AwuGP^h28O(Rr8qz
zA%H%|>lS{#z>Bl%?{!Y8ieixah_JSLV>MPN<@2B^a4pl`)ASN(?+Nu>TsH4Hi*ywU
zDd7lD_E?}4T@sr0pc_WpI{jHs*5wym?K77N(}=x9zG$lt^AdSA^!C$RsJ9@rNh3>k
zw{v<-yS&Gwi7G&C?R@!hH#Y%8<%m6Fj=WVrQ%x<i^w?=KD<iXh81ZyjQA}#l5YNS=
zYHSyNlk6?9lBuVIo$oyLo?U^X{!qqOOhV(At}t?d5$3E4F>hTy9=1cx!Rf%CXX7E(
z%|X+lx!c*L7DUmxn`;=Jm7Jyyu#hCYkt2nTP51}O`Z}mML$#7z#*>ccd$$Jx4Hs5i
z1S*1|Wsbat4aHZuYYmYc`{{<+$e&%Oa7mhQDjA$KL*fZ>QX(y0Qdu|7-*^kix!)%f
zRIhZz(de=#swsognvN@*pBqDr#A*`igp6tWEu_YeKTYAJBzLI-%@h3Mv7ekShVq*g
zd}hDbSqzdf5~8bn5tX?|wwR3VHLI=euGjKzu1}fzIGu%))<lpSTcDNbu)%-oGB9zw
zZtbHXi2tSwACo}pgoPHg-IxG@(5C@fP%m2Si}ASePF#)m6(2aai&xE<$`v)jqH4C=
zRr*m-7(&I^gbxgC^r+NdQ`b;lnSdD(X=HOl27InaxHAj~9XU+XuVLEU@5f4QR4wHG
z_TA=cz(|pYMA^7C{*9cJtL*^pw$D-f=UstQ1^le`y)~K(m-edYbNbm%x66{J#BGe)
z9!Zt8T%l`LyA3hjk+KMVk7c`ryTa&1B0~<hw1bhUvStZeUdry3_EF|qi(?!;SAO1i
zDL{sZvg5cEs&V4A@B@jqq1~o0;09!dvL=G@uenvp5oVJ~A-0y_z*8>e?rO@=0_t(E
zV&2%4`^E7(GADRoR{Za(8w?WCX#dc#2&{XNiqE5Mxcz;R>24}${5Ykv<}<sc(rfOA
zEP!zed4LWF+1z%(JSf_{unj88r9w8WCZOmRw>rVL4Q_VCiV<<s>*vW~cp}xZO<}dV
z%^i9!Axm?gByx8Kq0;~va5(jPA|-Qj=($0uhDK|s&4Mq~WuW+ec)DoV$tX@U^5jPI
z^jC0kTTu<eDEUH2%sAR`p|L+Cn_jT4%$mSiLc(LM_S`y~yK>O>C#T5IYmsHR)G?Lk
z%-w+UcVnd`elAGsW5U+A+%F>DO)$Aua_7r>KQrr2`~4J^JnR`D<-cva|FnYz@>(wZ
z2uld^Oq(w}g()mK@vRfy^cQ!i{UJ0~LaNv>C`cdpWekgBDp6BRm{3&?AGL%m=j!s}
z0#yD|_NLq44q|-cKmj`1={m=Uo8mqgz_Pwx&16+=h<%(D-~1%H1(2qT3NT~!=4f{p
z>4P8-qf8tYu@;aOqjZt|_feW6y2YMX=Mvn|6`Np;MqO0;BQe$l#f8u<?T0Gj2{xTh
z<wQ3cma0pL1i)i6m|6Gd<5r`TmEb^8?{4^hZ+moeccjbKUjWNRWW=WuvQmk?nq^?T
zDnO6@PVSYdk(zL9s>x!5&2c4^pP_((8*OV7Rddo-RYL!|ji_Pd>x$A*BE`_^3!SiX
zs?3F#|Kfs0$7p*qlY_FU_h<)}vAJP?H=?JUwtGN>s$FOn?>LL&h%H<JggMncL+9%1
z1l2eD#^3uD@f<$Vwpj7Cg(wh*76^SF!GUd}BGYeLr!AMh&6gE5;|#C;4e&_rW5kl$
zTTgEDc%O3>QlIMB8qx&ywz=3FY%C5<J|_%cq3X1dX&KH>Y@&eYd;&-G+kc%w>+&8*
zd`#c4>(n_yLR6+@9ZcY8>G=nl^SLEETV?@JS$UeB;B+fcy@Z*dP=cnFkHF}$WL~ZK
zssleW$I;1}b+w43gDrhn3B0XN7e87cgw^Rn2Bno@S<$2n(*8P@S?#{s_tyEQg+;%s
zKXmZ!qgajToN_i}FCKE+mia^lE9>{}))z92=HnTCsp@7qN+uD|K_{>8ylLS^3q+lZ
z_GJu1mmj_+U*Qt?(yjrBI~L}iP@XTf_m$Bk=gnXfa{z1LrE&fqNcAD|4c6}Is8$BL
z*?e)%#c*+aV8cVJkJMULocz>*jQubrA!APag~2_({WDWw5+dHuw2wuwVO@<~eRM3t
z5+S)mlDVB2W=YX(@lQ@WkuGc(gRu^oPs$vS6O1n9e{2pqQiBkF)a=Q|=dv&C5hoEZ
z`#r*5PFxKyQ2eYQca3AU@w$j3M^u^!yy@Psb#4EX%!y-p|3XzAmY&QK_bTB1&j=B3
z+ar_$ZfTCc!^96o;;Qgs2Nvr@2~j)vSp~YaYe0>Kc*aZ|5W$_AjhafMdY^hy)Fw+-
z!w#Wz)EENN(ZE>F=ArJRbq?!8_kr^jBYpbM#(&!%2Sb4lr3zQG$zG!P?NMrpbNt#Z
zo$&Ms#T*4bUHALm%6D!zt~d8r1{n?P!E(F?qb@oWcP_*6-Bn&~_LTUNPkQ=MB6Pa`
zgidLl-+bxF^r(_r7R(3ALK)r=$66!pZo29HoXNO`mkt6D7>P3VTs;VA3$~}OcWNkF
z&5tFh$+R?kGGrtl@zzr!%q}B`5&G+_z<~r7^Yl#;wIWCGF_~z_1WiKVE%Y|o^eF01
z=rQK@x;EF@_4i+~8X-ro_Ut9CMU>P~^~j}Kgi~-H&5h7@1>vtVzYk-|t&wZ59qA92
zb@sUxGfguOSx~tXivqsYuLotY_TYFO&H6=bH7!ya;w_$b)L=**R(f!+Ecl(?shJ05
zKMpOKece+G%itVQf=up{rt(G%QV(a{J9xRs*=flRNx#tKzK@QnezIn$Rnz9R(nNAk
zHCDQ*!kqKsk0%uoK*|C1de{ooC3DLl^89_>=r+o4^8OLmB8wC<1u6SvN;LDBM{0Ef
zuBQEM+?Ekt|5#%ZmU8|4rCWn2ip(jB|4|pc)d@D9N`ua~tLo285sa%l9mdJF5)Pf+
z_K*S?lwmW}&@M^R!1Rgt37p#&m#R7<kn5-=MFoFr*BJ1~jZd0E@p??n86+QNb3LPP
z3T?Vn-BK64l#sQgw4L2Z-CmT4R~Qj4)+>)NP9w4zd*7=F(w@6EQ+44D$Ij)WFd4hF
z@7>=_%{ocP$Ay}h>k0e{dG$W!R!2JENQn3y6=3UMC*efq#{fs_(I9gG{QldS6biR7
zgWfv6Qv$fV<pRmUR&I0GbB}@t;GXfeM}%}zWoEd?`wqC9d)2xS=Ci34<X;Btc3<v3
zG+fQkSe}4AurgzO%j^^9M4<~}x8=Z2+_k&O1U+siOz4gB(&At$zG}-UZK-J#icIu!
z&#8NIkM#BT`I=1=_rm@sA5RQtsU4{-R(o;jFt6{BkWKBsMn$hG7I20Q__xK~#&LWf
z*eV<dycyMgrB<zq%Emo3rK+!-0y0=b$p=#KIUPwf{eW>P@J36?xTRkCvMdT;$ci_*
zMF7H-!e+j%x<YU%uJ|+ZqPrZ=8u0|x(wHe)S-b@qHN^u|L$~|T9B3bS$)CkopIM-f
zQ9AM13lUj_lONpKmLH@Yv8tnEtQ#Cx9yj1>ZJv|}c1EYL=oX=s(8ClLFPOrR7()n^
zWbImQFsc16Mk=tl&I5)6ei7D%I*uQmI_6R4hbu@R>@P=1qOse}5e&#Y^J^du!aXk#
zMUefBAWt=mC~zryJ+oPkP+PlC$eGP9fJEnbwDXqYEtsg0@_2hRvpxI#ksm&qLh%^>
z$K#ZVA@0UDD2|Z*JbhgT)`$K{wTFkXx0eJ){uT)yX*$sF7Vlasa`r8$c5G|3M9ICP
z|6}P&+V5BIY3|4V%JquINtvga!JrTD4~tJky4D1<$lSDfSCbkyl-h7_#Ou*o3MNlV
zY)X;Bp!{Q>wtbC56+aW~u=98?{%++>ePSe@c0_4NM``DeS+{VC4+j%GT#R6@>E6nZ
zb=%BnWVR~xC;rTCd(P1gCuGCf#Q_Z@yAh=*wuCWB>{9g%(vW=^+8K}XSaznODsu=C
zm1Ze-Y5cdd`Di)RMo$HFJ4LB_?AbnW%K)(iJ$+g?`0H!}bNwMg7khmu_xRJ=5LWXf
z8p`OX%^#6exlQomQy34}<)y2B)y-!_U^hox`VcxTlSJHg53Y`2+vtlPS9m{<NEuc7
zV=*twL-)@3V0(={wpm#Y8uC7^{b?vvMCa~Iys`IQ)-c8pIu7UbUn#0K`xO+;5G7n_
z#trEy4x%%O&1D?$>!Ft<`Q(APZiCT~$w`+NxS@#ZB3yog-0rdqvg!X(=XRz5uEemd
z@{Q$@j;y~06}TIXHE<l?<Hoy#cl<zVUsmh(zIR;5yD3oFI(zQT_tF6#<NT?9Ru#9b
zfd|0BPD`gE8-5JKDczZk1xMNv_H5f$C4rt)H7Kp0HNB*J>hgG_+8(@#sWL|s!rWsQ
zO0BM7Eot~;@dXIbOjDhqY9Heml8c5Wi~ZSXA2tNme8eN5Pals+;}f*AyhY4mK<=6!
zS4o0v8ce1o6gI@QJZZFplSMFY;p4DMy79s}HD3x7T~NvsY?bl=53@(ig^EW*qCtMu
zt>l5%b(mZq;p3tj5QouizsMqQrL<VVZZQ`Wwf)v97|%(#ci6R{C#m?R7=H9V;hP!6
z2^-_aXYiNm8WHj890vSqV#`>m`e!F0rG(&V((y~Pv$agShPk<88;woHFV&5>WOEpz
z{op!jHaNSCg4q7>&jWKbtHj4A=^#7&?!xz_(%469h=VY|&Pz45SWibVW|7Tlusp2q
zoEPPKb4?WmIi8-iL3fOR-XP|KwM2Y7=}n1pj(u50EL_^%k3K*?hA?M}0#RR)06eN!
zML+0e`|#9HbR1J3{bJD3?(1l0R<6#-x+}TDO<qCqt0j10`A3}H<d%zSH3Tg+{3@K&
zgs%H6TlcmS%aTUmL!k!VkxrP@hDD`%2f1N!8_$vLbm^8DWB<m)-JH4BroQ=J5jv0U
z63A|0F99bic=9#J-<@l}z?*+Ws&-dOnagA2uYqXDB*u|Jf7+SGFQy$%;lsMZX~Wi2
z*FN((XZ#r01tUK>9*XX;ucuyhE=cp*gy=Fu^(rI)Yr-Va6Yqm3L7co=5mC5SCTM;{
zX@1;#P+Y#vgAAS9(4}iayl--F-{K(g95zxgEA-$lE)VXZLeDIF$d;p(!SHn?YI^N8
z_O;8PfZ@J}<91i#Po_2w7G~enlB(+*f<mS-Wt&g&S&XhxC!E`oQ+J@K3}k4cvg5z-
z&k~&Y;AV^7&$O=_^6hv7JRmnZt~;&Be|r!mY-{t4M-fe5j!p0V=)!yE*#Lf_T$z_u
zaZAasQO>)w4y3U_+f0#1aBtXw1b?L;Th29C9Sh3Rc32Sl0~|m`tok7_!j>DH$FaAR
zDZxQErLwZQ_YA)%+Yk~A&QJbJ@at&QMMehuZ8yn7wYKsz(gW%kdXiIMdC#0J_m2Bp
zn{MZ0x3-omPMvgv%<?{dVlniT$)GKp4@o9z{zJKCKqFEsQJrM0>yLQ#KmG2lsBK1L
z7fG{vzq%IxNQNvAIwkVdEvZ5>-1I(gEJ3Eu{kNX7&|J9kh4|is3g$nliRy01B2J~C
zdGvKhG~w2XOKQa$#BURKa4!j1FfjE0x$&{o!sC!^fFqQ6l6Uo3W7zhNT*R^TR**}g
zRK)XQv<*H4=0I${Q_3%XN_+dfU{bhwea3U?-#hjZh!K}bs>SDsu^!%3J8pw>*l3d8
zE{)xGMXVJ-*~Pgu@Ww8Qx-WIdWDANdRc#WO+u>SokXdY6yC3x_EgU_jtLP)9n%N9O
z>C*hV-Q=Juq-@dHIabcu&wD|P<g)y#=S!8riQ(PP#76CUdaSgJpfg63%k1|#mtHGg
z08vRoPgFNlZnLoBkr>A=EC4l}f$;#;D$g!~RB+AV@0Gp1RsDe5uMzfgmfh}r^tY1t
zQ?RZs{EJ_koo{z!66~f#8h7~6H}>^L#qZ9W)2<9Df4ABniarem6XstoT=W)2uQoc8
zeZ1Zt9@eerN`V+jzmU@B*FVmJ-*8aqPW{|wA%QfG`^bil&+kFV8TY~FdE})E&Rej=
zjnE-iA}a^kJMKVb(iGfYW1L6Fl6l@4x7|#OTSF;IFYKO}7iv`XGPu%_jvNpDCX<YH
zB=|{UAQaHxg+45)tTQ5we6_KOnNTsqVxo}-o<A>lQ8lLcJx2V|yPL^VyK`7a?L9Vn
zY*?-PQP-YXX5NpK{laDQJL||@q=nA;-_sdIQ=KdB{k~ZrYxslKI&9xVe{|RyTwo7s
z;AKMn_Cj{Jk(Ssx{c_GM*%vgxW>L^|z}2H{GOo>zujSAXkE?6qDo213ornO3#O~(M
zBmiTeD@Gn!O`P-SL%A!fW)~A7H4G&3*H*xAWv9H^0W3rxV~=1Z21_^l$NJ@ZY7{4A
za{d|8P0hpoidL1z5Ql=id|-LV%19#&o%O{1+;OwwC7C;0<l8`jI;PiZ{@ZuB&fb&0
zzbpi815KyvVy`ysg`whu$9;Wq+jw%1R4!I`QOx#&pvqye`)`Q7&Aeb>`hKUV#4+cy
z{1yzN&CBsef_$7~w~ZKqjV5QLu=ibX2}HMwr6^W5jlp~-7Xxkp&YaHi><qn~p~-f8
z-diLGR&U6$1fQbNmHkl*!OLEqTu1Tzdb4-#FPH8_Lt6oR-ai-gsLL!@62P=_Itr*l
z-E6*ne&`S*O5Vjex}E^^Y$n*<Xv4BAX0*~kGG28%_*S$Blv|evTI@g^Y6A@tT9W7&
zlC*An6QK@*`OA+#*cWSNUMmU9teFH(%&(S;lQWl9cMb@=ZKHj6)0ZJ<W%eW*C1}aT
z<!T;_v{e!es+nG0<!E(?2}$JusLk1fO5k1f`+3Yd-rc96QRt0iVeh%+%3=K}+$Dtk
zNF$(u83UMZPh#EM-1$*KX0K-T+)*bk0Qu2knZne9kK=+h*T-uFA3>$hcj)Mar3Mbo
z)T~~~^CVchZSP~>qOV!jhSa#rNZ{RRBEV7efARH~Uv0fl*f(5?gyIx;T3VpE71vUr
z6o=yO8l<>WC{moF#T|-ka1HM6k`Nq%yM&v5f7i3t^Wwg*zaV>^donY7<~Tm5z0k;t
zwBJ(q00=-ee?9^pS|r{@ifv00ZM$0mwaK^O+hyNZa+#Sw%eo$a7>h`Kd;ag=3S4ww
zfBVCh78g?;Rn2%nC%dE2f_FD(c@?NIwe6s_M7U`i9r5+5pDZxB@W81$F3KA}jW2IE
z6-B8Lj7>%=I1J&K8(8Q<8bkxtV*s>P?PxcaRKEtsUJ08YNp~Cv6F;SfH0IfiZPYtD
zGhw+I#n#`B4JR6@`b}HN%-4gF$};A}$zeu2laZw%xxA8ngaw*PRw<eWd}2u*)xy4w
z<Lj+FijvCGx-ljXoE%J67BXA7btl(GM8@IxvVWkX)$nni{xLkL=W08`sUzY8(N|ls
z(<N7O$oa*>UJlX-5vG#&A>TN29_e%(w_CA3GsZ)^M|`F_rQifrOCdEZ`_n=_4IwR~
z3`xgCShKs#F6pknaj6R1WaJn}ZatHqbRs4Ep}mWO36N#=W%ITa3-<<j@<m{zW+rYs
zU<l7q6h-$}lo&|o6^yLbOS<Plj7f&s$O^fGVw_{2_@p@xFos!rLlpY3s4IWc9dFMY
z*5kNxC+S)Xz8zuct<I`YHyHfLjg3xOCi0`k%ZxOAo}<(1!AM+g!bzvL3{*Q;2u3_Y
zXT>ye_^`YoWUGa^AW_mevM3qQ2D<CpRg6$;Wd_g>SoG7S-K1{ZA=fEig=j=&F-$<~
zI2)6u;Q|jt#vtb1@WH@X+r1OJoV1b)9bU`na;4C60)nVgFcolN@wrQfb`rFONwN43
z$Ut9uny!2dBO6Sl^O;0*v@dbwh(*`@8WZ#m0}wdn;c>p?(0>8$F6X(KNj9g|FJk@a
zXO=>#y96n^GZ&e6d5VU}`n)ZC7n4E*s@K1FeHtXV5<cY%XP0=YW{(Wrk%~j4w2V;p
zY;I$+58P5$`b&cA0y6b0WBNdU7J<TYfZ`t-J!%xXJ;4oc7=F$5Pd?*B_~#dS#q*J|
zcRPuVxQO|gufz%*`#n4a!aXSp{)Qe3o~|!sHZPJX{1tq;E8YZMz6%`I|B@_m(`A*E
zI@E}Ir)?u?4G-ElXV~XI^oF~I*22u_v?u!ge&->bkLJ;xFGwa!-0tHtNS&Qvzs@(I
zFxUD<-J~%2^X&ED&;*iioZ|kwf|9WNV6>BpU#~^A8)mBoc?R$)`8H#BS{1yHLnQYG
z$jC2L%-A=2Q73zH1vnyvt~j0^2+a9oAEKw6Qrv7LB~K`*_!7BX`2{h6U3ITKbvuXU
zczH6+6UeYq@<88gm-DTJ#Mct1OpB2Z{c4gPS>#J#7}GqrPD_js4yc#y5m#dZMm5^K
z*-i!~_~B(J;Iie#iy2h7m%$usoWkn?-T}Ju8{7@?yX|5JEO5lMjZiVCp=qyotkabL
z(1EXBnKGKD=xt!#7($PqZLDRQp4GSh%2!1{^{35^h<&JIlVe*vRT8g7T8oRT-3qw>
z+rt1JlPMzC7CU38;WDpX80))kOgoz3vjl-WJ?xNr29C<Wc;kT?Q$sp|5Hoe9oDCL3
z){R(v8|Do~@MrvzuqvsDnM$tqbSBh~XM`O)B5(c5;1{fFZ@IdX51oWltjIZAm*#i5
zq;tMPQw_yNjdF6Nb-CFUKNM2xAl0zAu)F1!ls!Me)w?wHzo{>J4d$#6WXQXgsrI-A
z9bRbiWsu!RXMBtYrWDKty)zJPhh_h&T<f&m0$KtJw5v8a&3nvc<#hfe(1^l;_NPAe
zUA@z%*S(YbAs#w6FZtXihlDfH)>YoryEri*+YP5u34S_kyrHm$AAc0mr^ZeEyngcL
zW9c|}Hv`n+1WV~YJ5!tI7Onl^sgE%f6$zx8f1G=|;4?XsyjU2|_UltMkLQ~uJ%8i6
z9WKz~a|gurx*Z|u&0)wCn_EYe_XwT=9VADr#rIA<p&RPbmK)Mjr$mAU83)I_f}a!F
zNzL5(jz2E_9YTuayVo#xSqR24R3m}1EE+>($0apUfVV@aBz^ngIMciz(YZVecYk<o
z2pvP>GJ}A)LVGeIL#~mzBGBG?U{*4)j6d>BT)*z954HNvNU-IhOqQVj3=lc^tWFU+
z|C|S%P&Uj}#NDL9N}-An-H>+x;c9hR2|Ov!+xL_=aeza8iO*bMS8bE~xi0M!0$yRS
z_{{Jy^`d^2ZrWk#$z3X3rEpiI(_GkBnsr@BLN>LRco;$?l*M>|1UUgy84bOd+Ko0T
z{e9#}wttu(1tz-HPa#zdOIM{IXozTxG>z=5evXb>cF)o4{*`I-Q9Yc}RF4REFF>6u
zLG<9)SOk)e`X(dIbY3*6RR=rgcsZo8K~q+_sU(VU#wm2|h?_DZfkxyU>+|`+m%i2<
zL-NwfV&OVIT}N>Rf4ld%y-=mmm<u=igY~y(Uyv`D(<&lQ>Fv@S9es%W^D>|%&esYN
z%IsQDu1m@5{0e#suP(i&XwubId&}6qH5<aULW*tguP`D65q43}{zuDK{3+OJC`{<W
zwXSA9m*(Vn#Vra{F{N6?9n7ks-dSFvv(%q=^7S3q`0{CNt<w^0053E-MR*k-S$5<4
z%V$@S{EQ4&m#6-E5W{%$mz%C|c05j}%qb#&3Cqjr)(AM9Q1#>aa^UC*jlyrZ`}m}+
zQFqwfu}nO1cT&XFTv>~+qXwsQ>r8Su=4$q^HL95cxeQ)UDfP9tkqebM%XzR6Bp0L^
ze<FF}9nB<L;Bl{9-xD^h?{^|2XS=dmoBvBsZ*PFge12c$P@-fABX3dFcidQTGf~l^
z?YO^0IboDXnPmD#y`$<FkQN_;!~^v|yXt3=y|ZnuD5{gPGx*9)J^n!pJ%G>8|2WE{
zT-V)`VYIFLe&m*0_*i6|!t>m`^13tNdV3&@kC%7bbK25tC(}FgdQi!~J57K!xGqkj
z^7tfejdYXqaU{lLC*vNl<`Ry(b<yU+k0CUF-_t)XKUjPvk=WjGsd&FRE8SNGoNc-D
z7a64RD;W2_WSp5!JIXkh+3~Jb%rfa=K7TedJi9Eugc+i8@gEBhT-$YCdpuH5oZ@r{
z?na^TBdOK~gwjwLQh(pI864+~T?RWud#tr&@)7KRtnl%27@_bXTI$`~@!H97rW^*C
z;tctry1YP)rkEx!{rUf~00f59JiTrVM*aK6Nw7PDSYIIuM^DCty$k19q4Nv={cGeG
zAvxOP{QooN3R1dELQ27zfOX3ZZte#hpU$h}oc6%$xP}n0pnMk68D8(C0HGh9&e@bf
zV?Z4Wqp32@3_Kvjoqk4pZ>!EFp4RS=3UZHrmN6amKz44S(M9rm^4Uk=!!>!LOwx+_
z+mUx51Ya4zk!>F&u}bR@YyCdLm2qg`wd*YfE$(!Y<eyp_c1%VGhfj#f^*{AYrh%J{
zzK@!%Vr_^wxEj`8r4<N11JnzOt01c#h~ZX-@$sxCqHEdbalWmn&}Yf^{#bv_a;#!4
z1l?KbiG43oPmoQd^ia98qb#~d)a@pt7=5PY(q6Zf*K-U1hHRmt7#HG;VpU<3=zPW3
z`sMmQWq#gwL8uZ+urS?&cxW)VLY%feO$vz(<0xN31vy7>F3C51w);~IFzs%NKNmWi
zaDhL3VuAQ)*j8L2E<pZF@2BY;jGlzhz@g3?yy}yJj!PvT3~e#m2_S_MD*)1-*7e{V
z4R;gc42iK2l91e@x0}BXYUS&A;(S-ElOx}o;kDrTr-7iv)U_SaBqhH2&&MSBM8I(g
zQcH<aB%$)3a6RDN=ZU-*BOM{WLiPRKZlt@3U2zkITh=mb<Vefdhs$+d+4jioSdtwg
zylA`=O*z;<>*N8q<wLv`oTApV*&IN+oq6l}qZl<;5&ZD+T=<Woa`63)(8|8@ufddp
zVF~0~!?k2F-zBqmnpQ*N;x~VgVo@;vM=$sm+?fM^n!V2w^DhE~u6RE^-eam_d-~rT
zx5kg%MKZWasmID~3<ipC(X^#0KCQQgrusqJF7wz6h7e|Nw0Hlxyj$;M)@sO}=2|M9
zg81s_vM3|gVsaR1P<~uf{i8P!$m!#dFGxf6ecSn0_E9|bYl-~khzthx{vjWJ5KthI
z7VCTN7qtQr9bP1rh^A{?zH!-JWPmk|;Z3UhZboREv$eUu4aO@%6#5S@!R%gq8NPAu
z>ocAP^A0+#m=@c*L_D&#bY{<y5ZQXjDs=zyI89CTpdR<x%l+GOt9tj5KwT$)je+#P
zhXFq#KRd-PI`n{>IF=Tk!%twDD2FrH&GWH%`kcrX6aJ%7KpH>4(74gN{V@SW+iCTn
zaCbOGz@8A#a*cS-1&o_b+~afNmFBe|ZmrgiwmY?cbJt)i3EIy_Bg;NemSc89M9uuI
z(zO@KrjshVOJHZdBy9rhX~sEt9<eWd{M-4vV3gFclAt1nM4-3`9$u&}l`@t0f%@?;
zfWswXgCd5k3H&3&;DG)3QZ#61a4@p&$EX?N%|rUg3jWf6a0(AG+LQ6M=0*SAN7rkW
zJg)Q9ic`^3nOfoFfi#0-cIwN_{SdGi^!Q4l<3!`3H&|Nl_6<5^GI}dALt}=hJ}8bl
zd4wRB=%wS_`~pub>=R6A@*YFX6CY)*{~sn}dn8yp+LPw8{bEMhWNdoOXFy1>PAf9v
z^G|xUfD>mfCrK;(|BlHwqH`f_Es1E}+rRp|w+T_DZu?@(UL_*Wy}b<OYtNUL|2-|g
zJsw`0|4P+YbRv|YfAox+fr0|CuIx!+VUIRW3X*Kw+&fn1R}}6jKeJ#sO@xg9co?0p
zB3`Ui%(_Dy9b1e%oQ2j-yfUwUDWz*VUiQ3-$o}@f=z(D(P2TsDbo|tj$bRwsewkk+
zeJ1+9@I>%XYW|%!@u$~p|5xY+p}07uJ!<Fg;{Pvf!GH7<apLEn9RBpB8}<JNANld}
z-<iJ9daeHd&N%r$6pw*VM%~X?1-c-E<MF4i<x7}@7xVf<DqkH;F4M_rM~K~f7!iZn
zmbjW{Fx|Y<O7oN~_HqAp;gir8|16<4eD6U2$gU{U)p2hn#7&2nq5oD#&w*gqrb6ll
zYLP5Bob=6clqa(9>RM)2y@#RBRrvg+TZDVp#fNIE;~3nhLy_u3F;7el`)ftr#yO$J
zE!#DLUwk+;fREIkvj?KWSPc}4*L{$_)f0!2G|mH`;|FJpiBP?9sODX(W=uEsNE|2=
zr!n$}oyWB3#z&t!-0xeQ+H~euvg&xEXWPRDJDI%4KW6#7rrVdc9&ehJ)Z^0LsiU`^
z7tZ?{;pO9-rHKa0E6NDqn#GHD^F2g~sZ(^)oS35x9Np5ksP8wfZ98~m-JNf(CB%0?
z>f7fwe7Dn2Cu6{s!pA^Q$EfT2d+L4+Pc;hiR`zk-4c2@Arxl^=i_@Gpj2^Eai!y<p
zuhna9ROu6`qOR8ERt*l6s4oy+m)-{K3#2>A)tuQK%QaA>V|M#Tx3F<p5|8uB`&{>F
zRzA<HTaRi)g-rma1;<(6u)%$%{u;8sXX*UK#`3?6s?$ZePv8sL@owGD65ea`F+;()
ziFIEqvh+SD-|@tz5#w&_yq0dgqGDYG6$uK<?`0tH9j;c!y)^Z%v>oQ0a?+v+bgMLg
z3>j$*pBeXh7<kK0k4hh0eftlLoGJbGJS#6x%hxt$Wrhw~2?Du~gTCIu9^4P-#ZHlU
zVe5bjubE0Mm&o=TNg@E2R!7TsmN2Ko`mEdkxC`4`UmeW(IK|dQ4%suj_D>}adT0rg
znD`am-Tg5?ZhK;q45=2ml7e$`pPrH$UnC=l)Cx-k#h&pA{idvhqb-`O8m`c0_x;eL
z(@^y?O7^F-8vjvQ2BGt2m&NE5obfnVe%S8@B__#3<@;+BkB`6`^*?3rqL7j785|ec
z&isZ}LLWdSu6oa41COMQ%Pm+BEa&PN#XM8kvT1#3&1ARykDMDnp0~e$iKVWe*c-mT
zhr8~8N7pEB_2n**?pz=LRhoz&-NmX};m5mCU^!yHkXyZ-!56~of3i9_D5WXJ`Qb1+
z&lfEOWB`1ICVUnTHCQ30C_TNf&Ge<?gy+9XIRqJDFOTJS`2?a%taicu5@zpiiJ1^E
zzVtj7wU2@hfso%Oa<q`dFKYU~hqBkbA+a6zIR}YVqm46A$AiZ>46(gBibwE_O1thU
zfc9}s#$Fv-aWRXpoi`(XkkfA9e@*8u@26w?ZA*0#U9rH3qT%x~&b<f_ml^iAj8oYD
zLPHmiKgpxN1%5fPXb1!mWFR2Z*B?Yaha3m3-e&xI#wO4u`JEP;1$pNlq_sSx`>Gy2
zM21ZmUb@~>5bNhx5rn28DXWV~|H<HRy&^&&j|x3rRj_rGmaZB;YqGm3w;q#{+D@pt
z1$7>=he`fcbrx6X^#8$W=w0qJ=D59LN8Sl^pF;Ipj#HM2YA=Qt%G#2s_nt=fxB`dC
z%xRRm;zy&YsZWHm{HnfEnD!X1lL{_q9Nhv3qEtj=c^2VGh({n|Dr7prAk`yuIlIGW
z{~{oh4AbGfp-y<u!N9v^Bd+Gnm6R6uEsfV3(aF!Jx98B8SAb!^2ZI=ZZWTSy@_{yE
z<UMxB+Rm|_e}qlY<)LI~XUyd7$Sa<T)R9OUhNZrBs5Fsj_z3F^5ni9g6X2lXh6*74
zA<?g9miXvGaNnctQtVN|KhD;|`}LHdK}zEge!5=q%+WKs0rT;!Zz;ROWUYRT$Sy6_
zk9{|cz6Gj-`R=}D>4EjYAJX7LStjLHEnEUqVQKN9c`M>Fh3a)_fFNtd_XmQ>hVUfv
z$Taa?XfpUM&2p&+L#2f^EV2ldeqRrFu^Or@`~tJG+bo_&OW`y$MH3}ao_{FvTRhH_
zZDCwCMTXjtuA!AoIwf`;iCfw&&h2TI)zsLk3iov}Ao~8s+zWy3)(z^rj{R1R`H0xJ
zZ{$__K|Ve&e~Vw5)hLj$n%hO1cFc`EFQ_j*Pwm=LA>1{6=N%U3vtD$2_$z{{aYS?P
zf1UB_0rw^Ewurq|vC~tS$wjm8YKP|+<L#GJ+-1(45svwo%;3<+Q8{+ay@`+%uf7a~
z5pbYfm+caSYWv9}5m-de)2hS;^~yhL4}JyHt-<lis&Z1*FnwMa)`|R0Q0A6kO>Cb#
zB*m~y%g~jjrzo$l@Q!;xDSB%~6fMUcVIEaKcHa?Om4U3`m@baf(A`P$MVU1QgV0>s
z+$(IH%fWlgaMUB(u%g`C98D^^Y}ohfG~L$s3*Qi>^7{m8-q_cny4q;H)05dJf4drC
z$)=g_N?RQ6iz^#asgwkdJaRvdH8lJGHPx*Ug2S{s&ANY0{+=<I&W6I<${ypB{L8z4
z16kIycw7U@!{rnOdpoQyoyTm>6MMP0EBT`1X?3AV<wto0f{Rzao7_|+n=D^Mw<WWO
zK=GBa1awJnmZMkQKS~(~-Xs+U+31K3j{K|XXq4vuk;UAM=|Lanifu$<ndMG<@(I%S
z!|x}~p~Q440xYJEQvz@8O*PRxK0@^qg-gQG3PT&g@G|pp3yb@O+;v=w!b)AQ_$Aei
zr1+6C8@xA?F$t_cplY@vqX5-0(Oi~M29M?-r3dY*uidW1Z~BF0H9`$;G-&6c*~cDJ
zPW=Ss1ho0CcEbe9X?H{de;iHf-qp|Z(0X{Yn?B`l>6ccyP}p(_tl-?Q>w^S5FUerD
zQp||8fSo3xrwJtynxq0LA+wsZq2d{{r!EfGX*YWB%f9&b>lQGp93YBejzLE|ejOTO
zsc?ba(_qDcN<cvgeQhx>25@JtdZHTFJ_z;?HK6(4GjPYV;n2=T8ELYUlcXs#TpTp>
zs!jt95cu8u`uYwTd`fnwH5k@=5ZcZ247P#hRg-Rq4psT@z9=Qgr-`1zrF6366Tmpd
z$IS01vfnm1nSa-}$C$}H;P{sE1*`;w*3~3V`7{(4whp*^gc{$c#s)erO&a&&mXk^e
z_{5IQLcUuS8Fz#VHk1?~x-=vbWNaiQgq;J5EZ*a{1i~#-HmOfW=`0y*k$nzwH<UW!
z#ooc%eDTKiwYG`QznZD5G=1pGgGyK()3Prdsfek+kI4=_u@e-IF}I$k>~-)%OTeTP
zmM3N}5qLIUbzDo)GkCI%i&LSTalQ~1L$TB_+h=+A;$r5Foi{$`8?j%w0vG6&C<$VY
zv{T0bH1ToO3SrfY%<Lj6PHwReBG|3q*Y|mF{{`M@N_%iVT@|qPX!rW?cf?S32K7B*
zH-nNVi|=GR!8psC8(EUVv!KMpD?#Di?+t%~=6fX2cXwaxWCtwm@|P{=Bp|ra0W3I3
zYd>YOZKs(A!7tm&a8vu6QFU9eeMbp?9uHI+B$Q^eZpX69qrqh^qZpZs59t1EzejN;
zH{%Wd%{=NX<Tv3EP@cm9!7uS;pGd-NXtJ?p{?pF(AP4@nIp`<2%f3_u(B@cvQ-usU
zW6@I&7UZgK6C5f8)m~?gmx{un6}x(z`67Fj+rDb563}Zqis2dUA$%z~n~nEg!Ot8i
zSLaM&Ps*bNWsn6P5>C*w-N+w1Bd4F>uY0*=+`R7n#g!q^`*F;T*HM<rmF7#t>PxV{
zAliuts1*#WRWyBEF~a5D6|x7LC$+9tm#WnAhb0Znc{zORHk8Yc{ol+0{QYbAq7q_p
zmUad`bEf%!o%o0FNNxKKrb&eQd@xz4mi`8B#uOE}7y2V-knELai(&Vvb7;tOYJZ^c
z*@oIuQ}vve4SSQVgcts}3B9?&otAN{;7lr$LyhOosKaT<L0-KJP~<1n!an)SCa>(7
z2cbe@a=s%;C+ZYjH+5VavJu^l+)zw=H18O7wV+HK&Z;_GY7b5r0v9bA3|T!wE@OUh
zQS;zT7V@~iW9dRGoj<lS$lL#PdfH^QT5DrO9vgyo*3Rp{I)y6npvLCKa}&7-U)E2f
z1)jKHRCPf6hD~nQt5@6J3Cn+jbj0H`Jfk)0>z_Aw`s6c0v2U`SYlD#mv%U{YqR7&=
zl-lE*6#Cre?Dce@mR^oVZ_Z?M_^j2y6B%VVlozl)2tz?HH?%)rj9z{V(S13r7YbDq
zT{@YSR`sA(?Oku-^}F9%F^_q_=Cj!0kOn2p=h|j5IHm#Fi%?M?p1bpjL7u(i<Ph8-
zl@Op?ud8!Z^IKowq+3mpV>R-cvh?i+fx|JgSDww*fN`E3%;d-K?`jD$kL-3I+3b~F
zU+R0@hpg`FC7>iVi^+JQyH0c|`eO2Im(ycVesH=K=<5oC#IzNT4^GxVzo3LBJG9ph
zLQrR+-NAB?Akl3|^l#089ijPks=72<2S>b<!N_<EpV23qZq~mNR5!PO99x#Q3y#i6
z%0MpkapM;}$7+oH<I{tD5VBdSEVj_+F3{rrfiT-f)$mcw*P(0$|8Y0>th3&NgNM23
ziitQJoH)sqxR-wj_l1m&s@(6TpEr0Xprh?aNk4`bj#mddm2$E-;mO~$)jPlCz_QnS
zxh)6de4P&U=onEbkC=&58aPcJs|^)x{lkTOGWB0~ZT;+EZlU`cW;tB$xFDEf(N%-(
zmt|VxkX<RM&+-Fke8aaIa@0qm?4&$?;B!k~MN)-pPNIsZk9+uvJ&shQs^4(R%E00W
z&Qp>oyzIbuhf39J|G;sXuVVrY93@weUPcLQQclbBp;GeGpq-=`pjb$gjE=D2z)6-#
zEAp^qqcNS0@Rv}3?KosBL$Wtb<ibdYvpND-i;nra<Mrdb=TJ1uLtKeMse{^=U?spS
zC5qin^?XsQ5DaAAeMGqV77NZ=E(JH-`abs=L>gcCYF&B_m`e=Xt)zHD7!#1DYbQF0
z>c=snLIWzeuK3j%W{D%;LZ9)%if;}6jD1ODj@;4c>oZM3V@E9B$m2>yij`UkNiEx)
zdi4%sr>i<z6Y4*2@IDC35y9Wpyh1OXCa$;y>;7a$99`b!N?^EWOp_;Q_00wS?DEJx
z4GIBDn@>&y0mP&+LbRNsoR1NYTHc1q)9lqo!PLy;dr%pJc8mS&hWqs9i}N%iN-vhR
z<JZzlYZLI)?vZLsvK5Sc2?mPvHU$!sU|;ln|BC}o>PpX~zB%61%SVx0N_)Gch?MO^
zpFi14Q7<cni|U@;<Jb3Pi`zLQ-Cgv&i074i$|?Muro%GlYml|`Xme+z>^*lmVtd)*
zZ=IF_v!2Y_-AKG20Go-dJjGoS%tMs5n{U~=g75#&>iQs;3_NQwa=V4J6H9gh6kEj=
z#UWJDe>}6&@bkqAPBEUA3NfQ<91QOD6X!`B^!5$U1$;@FroP2@O_+fPHvq;;<A1TS
z(WLJpvm*O;WA%r)_dJOyugVeP#)oF1L*&TLyZI+4hV9`BA8MiP;@geq*V~sdq2C+Z
zW(BfU&AW#u5IC{(7UnOES{538`Fw0IvwSQZHdi<kKTx&CILWpbM`Joc%pWIdDWZBo
zj{sv364e?s$|MCxh=QwCuTL~d^48yJvOZknA;i<Zk-?=RyQ{`-&y|TS1P0+aULT`y
z(Jee99Ge|bK%$0QH18sDv^%QnZhgt>V4FhzLQ=tt85Q69F~|e|(c??&aZEAA07pmX
zFB<`op!7VR;-Hpw5euxF3@@_RInHrd$~?I*A2DlI&qspL3^hb4aJnDU;TA)6w+R2`
zCL!JRnzX%bmc>on&w2CIZgQp5U0pY}*B_9G$)5H+Ke-=XAKj=AG#Hu43=5b5>ud+q
z1V}qiC*%R%LHM$8WOGA(#V)INx0~aw%@OIv1J_2MkX$Zc4Ug4bG(*?mQvV7YgZyKc
z`u9WeV(Ry*1vA?P<e?N($(sF5lx(~)Dt)qId3GrN6ZFO9JI}^Uy6kD)u3vvCv1$2m
zKi@4?31!D3cztzg6|1mm%5R7qagN#zm6Ec3Ci8o%4G$K5$LQ`KXZnO5hsy`CBPA8%
zJr(9jam-j?Ef9I!GbAquoo4yz7_IZPU3|(|e?N|vH+1|)$TBFCguC>*?$vG^-5}*O
z%#_3{8m(Dput0dG(eF@!&o?D6^F7}ynth_Wq3M%2tP*kK23WP-6pgKZ?fNhr%{En3
z+8k(K=*HoTV4CG~0i}#R<+n?1fer4A1DlBGm6CaWy0Mv*9nw@8>j>2$=h^!bT*ZK}
zxXAy@2iR1T?_lS|DQ+!tg%P_J(mGx9_)n(;(o!u8Z6JO+xw-Y0{kY8LPLHreNq!5F
zCXu844lEIrpQ;oRo~u$Yk;4`WF~SqJHQcJi!G-Hj<aNg(cS~5yYkLD_xo|>YY4AZ}
zXI-XDg$9@WLe@jmVi(^bL52Q>6y1Yr${}6LPIuvZ#FUi;N<o{ej2XA8=X+q=<QlMK
zWw`SXQQ6y|!LE#vt3EE+nYK70tgnt?`q}oBzccc_N~?dU6y5NAm)85xoTV9w!s$FW
zg3w+<tpBbgfG5%!+$)h@i2cBo$kj+Bt9D9bE2ukZVH)PfxkyiVHFF@~Mx^7RdGCBC
zu#zpENC*A_NPqF&t^HpIz|VWqN`458v{1*6UBhALLysJxqlb*M&dW_DN|?NJdxA8i
zHl}0qOsJy?Gg+qW;Ubj97n5e1<~+aha5Rqyw$2#f_i943RdOOv2U50}zY5Yc&Em<o
zKo2eC&P}O8C<65i87LI%k~3}s(LT$#)Gk-ATM2Zdk-Pt-|JJ0+X+|==VPvAI+4zW0
z#lT#qO(crRJL8qL5rye(y4>UE{6UoF2IwnrPwUfy=Ax+C7&waak;-1H5?}>ESCKu1
zxqE-yv)Z{15>$A`S#eEBKd57S)d#^*psu|>4%^HvM)l}Gu`Z4W_iJAGE)3WyKwDcX
z(_wr$(Rr{cy#|U?k6JN1Eq*G3js|IE0kUF1y?8*#JKgo-3_dB!UlPP|WBIydp4wV#
z8|A0kS%@q@y3QmVrERJjHFj<~zA*<57Q+dg7c8~!b=p2+2|m$|c;${U6sctA8EcD>
zN$@a^Ptn<MVBc2qfIz!x8X1ZgKSiawvd*}MG_aUt-u8xSTOSE?EPZ4q$SC%C!F|!~
zp1@XxJ!)eXdp^-;sh?-?BiNVLen+{Vittc~8QhrBKxZdschXp0-hce?m)fXwR_>=}
zROYPUin8eWF8>OG^TpG^M2`zvOpd5aY~!e@iw_Lmu@W+<?n}>S?;8~)0T-@?16YD+
z3>1Nwn(rWUgf;k9-#KD~kvZ<EvERvFU(_Nn4j$j6L%vBjHa${++4Yf_vj-PZ;DKjg
z4&CwZ-Xl1YE5*l-`)@*&uylzn&gsfW3r<n*Daw!$&XcwGV{+easCXtjx$@QRSNte(
z5plq97w7bv!`<(`%|{)21kxS{F61FtY&*>Q`#(MT;aN_*Y3V<!WA=EHwj6`JAZ+M1
z?3>D%94~#pDxT5a2R8qSOSqq{|Bc2_7uMcN7I$bQGS~AAn$*LH2`#DacU}k-!41YK
zh**5S;3TL4e4YUJ3rP2{T6SR+A)6($(@8D<Y<Gb}g2le>hJM^AQs~N<)DJ(-y$rX3
z_yuG#=u-`_B?R-ncw8iS9j1U874zoxv1wy;_eFtSTX8upD_j5ek<cHS&SsvRst1+#
z@tI}ppS-G$Mgy<~UxXU}=>0bYQtWYDb5DRW|BfpNa$}$MO=Rs@IgWJ!bL<;0(|-uk
z{lmj=ztfJ}Q`Q7C3KJc8!)nld+*+w%qEz0Y^fgphS(ZMG4Y`k(kOSMR+&ZZW>|{&U
zmm|>%U+Cr#jU5)8M~6+*&~z@mB=0eHl>jG=8=3Hv=`x~(<l*9I;ig9~H0g{<$oFGo
zNjg=Xz3d&c&<#4v<^EG8oFA$|Y)p!9qO^OKA2438%Z3t+L@NVA6=l48@mvMN;umkt
zXV$yL`h++htddC1`-s9X`pvuE%lY70<9?dzW_;Q#aw`^CrWh$_^hEZwE96#Xlr*q3
z1!=nP8}g;o7=Jnns3GF7#N|UP{<6l|Xy20?jminC&K3ThSyE9^r#^sPsTN5el9K}<
zjcWyiy|&y?OsQ26n~r(wDY%FTp&Xv$UE(imA^P9Dv1vtLb(TP!+dt@#`hFe<uP){2
zbtMW+if2qi?C9d~BX4)Nj<f=jjPpf6Z-2lX$aK;XH!W_P8N%0oU~6e7fpzRunff>&
zZ1D0~#LZp)745RDmQ_r$9Qyuc*s6(bfz4uje8c-ZY7WxCr+IjcslDj|+qYrB)zhu3
zUJAv$xwOcel)Kl5F>ploQ<49cQii9$-PMwU=|$pTNsGPLvL@|85;op8msQ#KEi}gD
zHltZ0-8MDj(4i9qlW2BO*K6n8MvGlFPE)@7K9_>oBp^k|0VoAq4K!N7^>YxXj!KlH
z+f2{0$;EpDB))xBEOB#lGT!$g|Ko7<+~cP>(F`R}Sq`^_K;a**QM%=0dW4fFCt&At
zKh3UWWIa{dJ?%N~Mev;xy-kT%jJf+WUtMO;yC-+kd-KVc+n?Om;}h{+ZG~jDORQv6
zpCkyeRT@)mutF`<?D^EoiF<<&XT+c!X#U-`94$q2D%KrM3=!-!A2-R(DRO-p&LG}e
zR{R6P5N1+lFI|exO5F)#H#EYNws^$b6<?-@3Nq~uYZ-oiWZF)fpQDq(x$3pqd6AcT
zZ7sd7qJhv_tJW+9Htmm*5L?~XU>R1sj7Gep>e-s1jyLvd7Y|$BO%aMxij4R+1ks?!
ze%*Nhc&OK`p6n?q%>WD@JY)?XIKn!+AdI8gf~KeR9Sz*VBNTur*&PbV?c3oZMNMiL
zv$4QIqJpOM(+Dfx4-+Hgb{<hjq<U=lv%dY?m|{rqIojB@9!#UjmZ)6(-1S^tBv!*N
z%;vEB%$b*oI8!QyTNpbsq3Zc7t2EQ6b9>WPRiCUqJxf?{2RKnJo$-ZOI-S1uoG<F5
zP-HUPZ(Oy9g)*rdCV;g~gJ}s*9Djh(5+}!&@&b<Q!imfmj$U}`-x|dZZH;AU$)Lu0
z+~s=Y7n(n?GVxj8e$N|HU$0stx9l0&T|iICxBSdk)D$mJ#39zc@k;~fgq$#Jsn=lD
zuO=){(0k57&r4ddvM}=j%$K<~%cioO#4~oi!>(tqXdi$mBGOa5)4G*~&lRHjTTTW!
z@vj3@uPZuc9a2_z*MS+(#q|nJ<R$Ay{Gr=l=1@Z&Pa%f~(^>Zgc!-crq>tQnMq%sH
zR`AT+P6s72Lx*Xvjn3*>fA^p0JiDpQfCgZ`RJmh+PnE*OFUve_ts$0!L9<zHKHdn2
zlgXiE(;iLl)8_nw@$h~CJ6lHI^fK3&b;QB3i^0hi)MmI~DLpvj(YaW^@=H|WD`o@b
zjwvxyT0DXArp(L!U;dOut!^$&qWZWq$_A$KROl8-^2-4f9rePreNLWaS?z<gAFd>)
zckOENvW8pbG|wE>`=^=(q>au(jlQGc{tB&irP2SpnX;*^uNs#wiT4TkP76IdeVFW5
zu8}==G8Jy`QURTVeT-=p!8s$9fCFCWH1P(pb(^h?m4BKOf;(pU;diEdWDN#Q8MaYo
zj)+D(Ji)fK@$MEgykGm0Yu6ny@%`97O|aFfLw8}$^oD-$<#D`6`qbG@&%3*UU8i}f
z^r|w$95=P3$`HvRWAM*!)|QzOh8>OPN(pu&t$Sp7Zgs&P2S{z;5-fZ3sh>adhgDxY
zM17q(?RPqM%!3ColHMGbP&vE@_geCHS)ahS+;83AKj@-)4-!zhYrIPT1)3!NY#tTP
zXOyV$iT`@du))XmZMl1wC!^1STp{$CoI@<a)Veur);HKovNkDroBC6dxIHDBY-Z)f
zXm{9jK2}9obY0au)l2un;`H!FJ^m~ee&5M6Gcto`Aerr|qNyuhWelc2wY}+&%OQ+;
z*O-oi!chpjlI`>;*e=!6yJ6Z11J{0XSy8|Ay@wu5BzKuq94!CZFtFTm0wVuYs)J_t
zn<7K<_(JxN^5*bhr%wSmqP9h6mW_o4J3RT&&r%yXuq%n9u?Ov2Hed1EMI|mSy}nny
zTRut!0{jPTem_33D4>;FZ#kJRt(6Sj<Cqe8?6s>MMDh5b7wL!!peg$GV@%gI!-_(S
z2H%f9{LTm*hxjZ81}RE$!8#)5CP^;0=&fr;sLE1s1wMJ`Pfc0&!l0aFAw&7W&WI0G
zIWNBKpeDn3?t%`_KmzJ+-Iy?0j<`-Y?TI;`ji;?c9|{Z@vshn*|4%i1q2_N{V|O8+
z{Wu=<xH1B~F0_Y()W)mM)Xve!OZL*vrcypjHjgkY@D#W@3)Wlb<N-3d=;ic*LP?pY
z$-`bO2HgOAnnWtx_d^DOBGLKzr#~yvi$Fk1-`xG*Z?c#&*Umz3QA<H7hxAk9Ra-Uc
zCFL(!Zx4FWm3&4^T5xvi4i0O&wMVdY*~>d}^gJk|-K}r)BK5~n+R0SNo@5&mw`@g2
z(~i#*cPvj$E6P^22iR-K93=>jR;EnDs))@Und9kMSoJl2d)Pws`)T^yz)=a5BFIQz
zI7|c<eBR5{-<<hmP~qBsYeg#ko&nWyycwze8h`bi`ItC}GqW=5UY-XlSfuOmpG40M
z&ga`$LL>HQae$Uv!b8u2WcIJh|Nb(&Uf=5ERf5fs>S!VDkUZH`*ruXmx6Jn7*%*bo
zoko<|aHzFL_OP|J<Z{j-WJtZOgZ=5^I~ixXCE=SrvPobOR96_^R`>~x3gyP$J+t*N
zC7nah#28IMXRZn>SaR$l9gV^;k5o&bPCyVsW{K+#x7B1$&$y^Q)4%H+NPC3%sJR+L
zZ-!v+CL<f@f1f)~E>R&2++IokFN#iU?+<|i{UKMyKM2L9V@wUoj^!cs!Ac^2-?7pk
zqKQhZUHg@AcSw70P|YWsycy3Q)~R&i<))n78#JU(`gPxR4zKLZl45a{vgU|uHclN$
z=&hmY{?E>*G0gAp-i%pakk~k*IUdn;G|E(BO<sO<*PvDKy7y~5`}%})zSTGr?@!6V
z7SE`5HOcVJbVDvjG-FWP#8-<lSzhh-s3-|gCL?d~eJ3MU>FC0Kzq?ZKIR6eGG$tti
z%-LdN?sah{FJCnFA;QSmOG*D)ITfDal$stEY;{;dhWmd1C8TngTIQu*3U~KI+qI$j
zuJa}p2-_7&6;FFQt4DVzKAB_}kKS8q8D~qqxLo?@W+{nE#?{DkY`Hf?lsPgA6_qE~
zZ;;z|EypL3C`cOL$_0g-&gr|$mkrY|ei+xx(ERSH{4Oo5-(&L|Oy)B!>rG6wQnvcT
z_PYu;cCqL5g*PBrUg~GRrn+<|woZc6@DGp4-ezB+>UBjMXS7)xBI~{T!9I?`GMBeT
z3#UwF4=tT&428kCQee2t$9${lXEcTO?H*?Y<GC#tY_t1E0yz^AgJHb5(M`)4_U|sc
z7gYIUL^}UU>~gU@rrcX~f6Fs83|>%h+6Uhv`S(B1|Agn(rnZXyFWTmqnf2yH5J{=L
zLC>C61wPzb>uuxYaz*N&&SbI)l^%&RNmR>Luu$i`vhm>MSa%uecdmsARO~|Z04x%o
z6O+)FYV`6NRl>o@O2-LZ9LS16QS3yMwfZzeF*w*u_DVa|Jy!ic%O2h~PDu=CCBGxv
zX^oo1>U+C7;<_)qfF{IG{w=6dG`wIi`(u5p75(eIUVnvyOuB`KHe2vA^X!{Dl`pr?
znT#lCzNF`_&GEnZEB`P4#_5tJ@3T~qg}Q07D$YkuHcxRQpV)K+SY@x2V>1$Q$>xN8
zb#w>9`@yDMSSbG|k#T(4k+RTbo8JDmILNM-Jax;LH<^^JD~@}L`IoxySJZrW1nUUJ
zs=VI*pjfn;fl#x2mg41S?S_m83<bGoS!!7Lp2y@FWae4ZDR`xT)4jl;kXJft(&O^S
zJv4cKG_sDX(r!y(?yL9jd)<Xzjoxsx1$9p9uQTSauK1@?dgEzvYWW%s<}ER8?>JXH
zLyS?-*>PPhJ_3v6b)yE*saK-E&qwqtdhs0?XeD!dPtMx&M`?o$o(R0$zx>7xxR%v1
zaG{z2xn?{jeY>AI)?<Ia)Y@#fiDdtg{W<cIG36AQGbO$VKi$=O{LpQjB?N*i)_6Y(
zM3O>jj>5Wx2Sd<BzprieMNOs%DQJ<qD>m=%Y_k@WKb`wR9q&8U#jcq1d_&&Y>b=E-
z4m7x$dFev0Cayq=AC=`yQ7v<ai#MgC>^oUx=5{^!J6PY_ZevtFz@zOKHvdIs==m9;
z)T@i)li%#|WDF!YZvDwEOP(h|{5*LfHT<OJnZfu)gGY^1-8I&hU&sF53<sYV>}l^r
zQ3X-x>0Vzlt9ttC!NWtBwT$Ssp(V8aayCm?h_gE0RpF7sKkm8KcfZD`%*R=3hooGV
zaro9`RR9r6%X@(WPPdVJ#1;XjcbgOgbu7lMKo=1{@H6+U6s@SMk0j5Da&*x}zk=Ie
zqg{@M<8`Fw*z`NP=zdPC5V&|<$!L4y`}$0>I{GA;1Wg<YzR$V%vM$-vO6xpi^o;B7
zHY0KKpb-LrOOOBm%%`+Pl2kC(R(9i^2HT-MB?{)8I{Yv8ZfS*QZ%5C%)HOn>NaFMB
zQ4K6Lg{kcae7|MYIgWKSuH>J;o$dRuylqF#?|-n2qO#U-bqNTq00i;y7hNtyh<yu6
z8HS*JO<mxyPygxDAMwA=?(e_+w-J!o@3cg#*qIljL!Itrk$zXmgW)N5U{jmItG~Sp
z-TGHf7jwwZ>t1~lWZ?CC_1W!}urzC*BpVe`0#eORO+P0~1jE<L=UYskw<EIk4hdj!
z0LqRwlE$PI%1xBL@6KhOyMckDHiZL;EDkz#Xnlq%FcCcm!BlTRC;<Pv)w9Ukq!uz+
z;XJbkcT=00gFTLZaHjblee6le#$K?cr1=h!)XB%4tL^V%mAF&-M{_A3Z8b1l7<?{u
zZ-_NMhznc^?`S_A5pP$?61(lL$bKYUR?7RD&{Zy<TGe=Q_x^OQr^T;hjRc3!hA@xa
zXI$BB+Xz$K>?=_)qo;US!ykk!2c*uS!lgQpd(!;d_q-6Rk7veF&v94#Llpp-EYGv~
zz%V?np)AqFlAc^-HH4vwAJHcNzgj&0T#NU)A2!CKIPlk-do$=M>E=P*iXfj=>Yt8Y
z@PDn{wTo0DkfnM0{4U!hdKIq#J)jg^l(~H_$^RUfH|LI;zO~)zlhGu8ZzNe9`qp#5
z$4e9Fs|GX=Hp+MA#?a(8PY7DAhua_`cijf_Rd|zdpY+ZwPq-~8_Y_H|TLjl*X?+<>
zos?O?vws_MmmJAWzISOjOwwuwL6TI6#)jNa)q86V8Yg8+b;=k+CdaI|m%p=F@TD5x
z&3IW%3~faw7wx`GKY97erJ~-g?!J6E>m<u+5$^sYsPh)EH5rhvwf>j0-=!L~$x_Bk
zBY<XomM{0emd{zHG0M~bfNukigvND{I;P4i_uYGvu!OJApP}559)BmfO*Qch`iQrW
zdIzgX{$I(aAYfr;O+RP%3utw-6F3+@sorB>-9}H3WXKo{Ia7SPcp)Ub6VeWwE{Hje
zoIizR4fBrH<Oz&_zurp%4-5Pl@z~z9W;$vHzAbbT;TH8(WBT<cI@oRp$AO?E`oGri
zPfnzv>|Me0ic5mRs^*S2lTR?YqtPsM`nC%;Ya+a9XRC}pAGAW<e>QC=Ash{E|LcCy
zNnHy+@}Mi0rWQh?n~JJs?AOKZw7F(lQv3_T;3530soXU4WY#b`lL3&07yc~oXW;s6
zXBSO&EcA8Qkes&+`+1AkMiEYI=}7K;013bujo3VU8%WkZV`!{h=Nx>UM@W5Kq*A0`
ze%ICN7s+M)sbu<~-wD>LV5-6jrVj>>{zdt*=#p2g#*vbJcJUx?MfeiN%c%8Q)<7Ez
z6D5(#POj5U^t&%DE%Wuy@ZaqWA_4^ZLEr3W9Cg(FUb$8Sx{}NNG~TPQOLkmT3hbG!
zlkQeOgc}u|neCZ=(k8pe^0@#Y+~*yxmhVEo{T6=CqeWUGIiFS<ov?{tAB>!)b*x!`
zISD!rQv5g+X=>*FP?ygq+vMb>AI=`PM%$_Sji;sHwKnw7=7{6#r^b(to$z-~3|C>z
zHmO3J9{(qZJsY_c)qST7Z<;Ks>(V=_(o;dqs?n`_Y4+5_MT?Y?i`F0fn@3(OyX6$X
z@mCe8P@0{d6-p(hpI3K>(u9bxEVVyV9{)tXGx`3q4{i(PWlAvh^WdM4S8O9n>{|XJ
zzs?1FUdpJnQ@b=-?QFGhzQsvVQpBCxiyr9Qu5Lu?vff^2H@tqWv~f-*ok<_|Eg0)=
zva10}^mt!)HnQ7PZ0IUOhn<$U8bp|&cG|L}^Z0Doe{8oB0+!vprt#2~i)YBFFPx<k
z6G_i<FlkuL46Q&Phff6uJ9Nqj=C95I7AI#N`uzFIIjoxDc_ktYTcZsN!Xx(PJ6stF
z#MXoVwF7nqF~(N|nuL!wT%0do00!#b)>`SZt#0SmaMK*(-Dg*ZEH_2@jW&}BT5q@u
z-z4XstHO6xeVgrt!eosqmw3?zcm&8YgcVTNM%D`~`|=#VzMpiXT(}JFd7d}f*K9(Y
zqs{-NW9=Ls(N`Yk7q!MiL1K^66MnaGsBW1a%WpxZ$@Wz+RZoeumt;)b)L)nGQq5xM
zum?f31C(Vtu=t+O=hcR`6rmTXPPJyLhUmVv`?xa;T?z0{z#=<cR%P}Q=d<a2!(rBf
z3FwjAc=zo)2QppUDTv1(vw=oZrnjNXjaGmu!{<*QEWL5cNB$xig>MNP%&(1%_a-FL
z<&WM;k3?n!X^16n$}|5Pm%drmcw}kofyeQ*gqSK1U%zr`+#?gV?DgD<h#DcrqAxCd
zP4&v+=&yt^@^)y)+3Xv_ovX_j@Cf2fM?>kTVH=|`CeAtK*e{+A&Y!{)h)GfZy+h9&
zaLjD4CpCbFnu{9V3@M?>CCDloYZ-Oe+r7Z0+P!q=TQ9|F^sB<W){^nZq9}T43atQ|
zTK$ZJn&JxIw%g}39J@lkESiCTvG+C{{JJ<x(>WtWNuj?ny?HAQg1Y?ksQLhl$Tcfc
zu6}@mB$B|D?xoI|Cqf9_I;cQL_Y$gRbUvw!cf;-ZHj6ud(_5WcWI`g{GsUrOYW~K@
zDM%7IRW$Ntbl#P?zL1^ci-GG?vHprU>1NJHX)T56XAS21D?qOrz{;V2xciBgA{JHd
zSY5xHBwA*r$y^HLHx)@SWm{A(ZO^a3Mm^b;hZJL+ZJ|h-g1pizrO_bqxGT?5q!{3M
z#6i2nqUGa#8M1^^@;3F3&wgST9B=AU<wn4dpJG>f2z{>*bLdr|S<l+zh$OFSMDZ7E
zACnM$S)apOcxE1s8_V(}C!N*vi(}2Wllwj>B-7wrJ|^5aYx$$)saNcypFd=vxcZUk
zU?qfm$;0+-i#XC=Mhlk4i*F&16%~%CIv=?PGFbR3mmp%*RO$LaxZiy{TZgTG<vOM>
zw9}~dnt9os9QbOiyynS6h>t${FbviIe74hRr<A$&Oc2Kzb~kXuZk0{xUhyrV?e)r+
zowvi6tV+7Z0Gd3kUdk?xo9wRDXE+HN0U)25gNnvwA>ZtIc+kk&i+}+ILKcbPj*|Kk
zz&!tT-n=NfC<>vn!N~;Bh^u29ujO>+RT}P;sjp8R9nzlLl3~Ir|4EF{irq4SX6M_e
z8{E?!H<v%pgIZdn%;}xVv->aC-S})m$9ly>;8_L=iBOzYQKJQZl&FKjOL-sVm=Di4
z@-*#Et=IW=-Sq=m92+?;?esKnL^~3K{!MS^>a-oIAjYoKlpV%ydH&CW5=7N13HT<&
z1uz@5)5#ttP;g5AyiI3*WG?rArpn`*N0g^RJ$u%T&3zJKZ+hCoM51{do49}y5X<Rf
z)GL<qI~AT<;6b|5N*w%q`ee9xetfdJ!t#eTjfi=<Ss`d|=Y%ox$w{Na-{mM{60T5+
zjl}smm~f}dZlgM^7-Bar@D(@5&wkk-?*ea0^8FS&QapsaRMIFfRy3by0!J2i*>q;Z
zr&s=Cl9szoi&we~YN(RErHPNXaQA&v@3H|WL8;-tfdv|79kGqJSHv=%wOiENwmx<O
zm}Afn@R7IH4o2k40#Y>UO8|gU$w%##2-jVbg5$}@U>d5cYtiqaIH6-ro?>O7@a~wl
zD{d+@Yn9lvYpTd8%}6rLk!3_^)JT0Z-7D{QO0x6Rr3unaGX8lcM}uKWXt)3q18Jm+
ze&<A0H7x^XlKV)rU)mXDyf@9SKsLzd*>Y_)^MONJ-7^e6A9a$r{H6^9U`N&#rY}ch
zT&fwSd@GCWtPty$_V4z&F{>t_x%b5k@i`dj4AOu<rZ?7@^W9C-P~+`p=~oVR9H04D
z+x?2xh4?pVR_1fjFl9^I?D7K7#K|kI_<JGx6G&194>?y3WX81=mVgd+)6iT7WlA3q
z(M{|4DEgFG<8_0d|2<fGZ9(<}9hO<q&fb0GNug^QL6EjkGi~6$iI`KTCDv&<T*ERT
z{*!=oz|xRI^J4+_5*ZcgairM4+YV%?lY((Dr5wo7m6H)94Icpfg&DPGYFfM?{0W6;
z;Q}?aQy%7Ri>(;HcJ>M)x_qcBlh>BAlzKNCQ2|7wwq!BD3_ZuS&Itdta99Go_$Iz{
z1#(cYk$V8M{-S~i2fn?a;qK#HGiP?YLB=y63*lC@Z{&}SH_9H?B$3Rja`$a~{CU<+
zMmYpdVc#n({`pPh<y}JMOkV3$ChpM_)oYSzuyj_w?T(KmgP=mHvbLI~A=I!5@+nVW
znN)PrSnzqz`5Nr0=auC~pDv_YfVE-6Jj4zsJ3*%k*snBNf8|1r(&qGoPbx7KxW~bj
zc$$FArP5Ed=S^gN@$xpe%uUoZDdIA{)=VVb8G~!wJh6be99A{*l-0(EG@aXg_JkX;
zgclj*^jg_6u<uWHpQTk2BioaqGPfxmxr#Uyj63h8BJnvv4}Z-r`Vdys8@gf@<=^N$
z^ynz~vH?eK(#=kJ6e`WK%8orp^lYo;ZylcAsy;a%_&Ut~Ez!N4Yq=O|$O4qCm9(Va
zH(h6XAFM=OPNtj8?=E>S5SD(0|4sc`#M}5mAX6SYx48H+WIVi{$I7ZYQd8)u4S4fz
zuA#7bL;DlhsT5>znU|_9X7l)WiTD3u@2!K{YM=h$0xeXKwv<w!NQ)LN4#A6i@#3Yp
z1P?BySn=W-9D>sl+@(0hf(3U8?vN1p<-Q+%pXYnO^ZxzLZ{|$qWQLrTYxi^Qo{QZT
z-2F7A(d5IlhIcuVahv>7;oR`|OS}OlT38ukg1cDbo0b5xLza&3BXlpoum5+O^FYXK
z*T)0O?ZdsfDn|i?m)-Fv)duyEjDR^FkYFhQVjk_8NV~(bFOrepyN*{L2N&z3Gb~2&
z{b@5mj=`&UGBV4F3^&&6hDS;n4T<{{(d_Rr#Pj`JJS+U+K#D};N#mk5ECR#3*ABpW
zKEdOld8zYXH@+|ZvkHrcy_Xf1Y)SP|8ANCJAeZkx$Y1`;I+F{GSqsB~QUsYD*|L0i
zVpZw2jF(>Ho+$?iemPq*rRm7FP)!>6MXy#vIO0h-{6npYg{nf`yEP{Sici3)i_%fk
zK$B{EW@7@SjQVZ@(wYJ>t~DYMkgFZ8jhAAkA-D9knPJ#Btzl$?qTdi40AEldUaLcl
zE@GL$uLA$G?)mu)Q(n|375F-z%FdfDe$0?Hoj~c5xgT2B8o432o&<{hk9dVgk0hX4
zZP58Ny@VTF4Ul2&n42TnmaY2tkCfHQO0!PupuumR?R(bg8jXJXeDhVuYYuptx=u*8
zT%(fa8oXmO|KIX1a|_i?9m;M}W&zrlHn7E(Lv0V{EZ00Pkg0hw<Za(&?Hz;rx^%0?
z<@kr@g`d{HoVk@j2KxgUWUejvTD`@9Uj(7ec!JyRh!Q~POa01ordS`y!3OSttmyLa
zjevW#T9Xz;A(4eP_Lg|iv4+A|EAJh_&LSgu*-lF|_sJw!*B{W5m#g>VsC3Cc)KQ(5
zL>e8rqlwz4^SH0)a%zkYd1$I&jx!Q|6Ft{^zZ26+Jf!eYlq7S&!N1|BD#!Zp<Xl_p
zBH?b@PQ=yM0K*EdUk{v;Eo>SRS6D8LZ9J<kN-IiMmYkRRzTP0fNV&4L#?`FzB6O*8
z3Rq0rZR8bSf))t*jFUpHA(^|}^%Js0#%w(#2ieeCa_PMD-LIn3xq2*Cc}Z%5Xv1IP
zdQmtb-NQ5A!bypZFllU2(&2=XzcQGpsga=2f~L~F;(Qoi4V5S-Vjaqi?=)W7OtdG~
z&6@SD0rttBVOV!#U)a%ijN8UM1!)x1+5aN#?EAqFrnTfu-r}xT{H9?wF$6GYnJ6y^
zuJd?}JZiNT&%j^F%gdmj`QWSnhy|xzr>m{<HM=SWEuKJvQi&0L6xSjaq51FPmFQ0g
zje||Dy4+lb`$R@d)i?D(h!N86u@2$HL~;YpgASq>sk=Le=xf{XxoTZ!c7RIs333hQ
z?f#AyG7e(EDJ!27JHjXf!dp^SQXj~z&>&*H4j<E~csM0(pHB6Z@v7t%ZI|Vxmvz1#
z6}?!TS)*#&saPzqXq+UJyg4v;yv^ua!MgPcnR-GKr%$*ul&aV=)q>q1bKd*Mu64rA
z$;a<Q2!rNkwUb8wexi2^4$fJOhprSGS)9x65WhO<+k&-feibo5AfI8OEef$5@J3;)
zoZT4wy87qo8Chvk)SC5O{r=HqqM7aqwUlY-bPrG?f9Q>GS&?YetaH8}*t;=TDQk&_
zOov}~z^ZK>{`HH$JQ<`ht9Dy(<X#blQKxA6=%FaW-7IT}V5T7gEt2nR2BTJ;CN2Be
z^XNeTxa6t%%_UJ#=^3*~0g6s9VLcHKI4w;6DqF)ZMENt0!SC9~y3^oDgiY<vNufIW
zF_2=vcKwIt9F8iPe9Mv#s=vc6LJ+=0HP6vsUAZOqsxyz{ltVvRc*W?vhI%DYfZ#$d
zX$^KO%6)!{y|h-5ye+g?xRV;p9kpMK_Wd1lFxvLRHKnyQ<NXKMAE%4%d5n+x$dW7K
z&5Tb5>qK`&A2$rvFFFexql5gnW@=*|&V~3lg+Dz$J*f|U9Q*`1cI{rDNKVAV2wQ9-
zIu|Sc@z~2UCilEc#>URN;F}-XdiiVT9?s+A;!X#9<<`9}rx!0@2G!fG^|DIlUOxcb
ztw-!Qylhd}^%KU7v}qE7jNHl9y~vPMjPX&iYt9Z-U6ON4OHCj6b&G@B!)(cxvGgc$
z=|M2hQNE~r4;$M=XIoxX>xPJyO8x6ho*%d}f;<~(CB%n3&ryZ95W>jF-gS|EXds(`
z>~nd`2fhM^*OiRc=G*rPS2xvIPHnWCH3gaWzBShi*Z9mRg>}hq`_?O2=+Sjmv4E+p
z1>4U<*Uo!W&yveRgVCKIMs6FxmUzSyUO#NrR=%1oh<JP@Cvp)XRBWywy#>UU#Avny
z2a2zYSqQXZ#@><Md{s$lj6_&Oy#dT9u~Aol++Dd(|NVSu#R70NWj}j#T3lVddH2l!
z>6r4?**SK<`<&-8<*fVBCjfA?|Cei?P73=`geQE~_?+Vte_eGxB4}bC2*!`Di`~sd
zCx_mu=tt}Zak$E5eVC#y2FxK%p6vd9jkE(AZ0Qm06}mrel@EO4Eg<h?PgwWSaN!N$
zG$S4?YwZBAy|s>J#_~gxBrzpIN(dd(^z@%tEoYl$6~&_1olvrwqN#2F(F?#tqSudo
zAn|0)f$>Xde6eJ7oigeZ^Nq0YboJ{Q*{G-zpd3Th=VclIQ2b6MCr?X=YS)`d742lo
z=W<=}ZQVTurehkau1bb3pz2Wx#HUMcB(AI2Oksa#?>%q1FumqF>-8OFQ+}iDn})R4
zIp6A<s$M$M-7qA7Res|;J>)BkikaJ<aSN)yE?FCq!>ds@RHg;%x~5(1H!##!UTtA5
z<x8OP_cMCUPh|hvWlX;}c2QF<;MNiLE`c5tv9&}Dx8?-RAe5&GE_@?F3uYU#&SUMT
z5BYh^+s62+Hy9m*u(4_GVZ)!v?di(;-W@Qn!bQ*A1`Wtx@V5W<vO9CKgppMGe-JyO
zn6HYUY8AhC|F0_>{V*HT>apw@@x!m;OjxhxKa4F)zrq6AQM9e|BKLm}S=0pw!^<-x
zaE<;rqJCJ`gGLR0+Xs_`VPQS(3l?W#%W`C1s7KJQq6N0sdV8dCr)~x+w|~bHsZGjG
z@{IE+GLG{beY^MYE9Mf%BNq2>4&=T!7I^O=6ZZT1oY!Vj?IkFZbKzZv)s~o|b=3P>
zvXw9lDZ1jVR63koi?~-f9g<J)VQ4X8@zqC(2HVfinBjSCX>g369h2fFUDej|sBb{q
zvK8+<36K2@&N_x7PTt+`#K!vTNB06Q-h93FKsm~=Vl)$b|FdJn@4#dR>~cKMxo><k
zsJ<6EIzAS>t>TiN{!NrVqXZ5{^5#(#Rul5aQ+4j$XCEvukEoEyu3E=#JgQ)h`6=1E
zq?C8-M&mGaZCT>GGF|;^(*=dDzfQOAuU32~Ce-Vh`<v)*%2-dWT(Gxe>IA<?-NUAN
z8h8)io|C5I(Zk<w9k1?{*I!a0Z|CW;wum7kIT5|{O4BhzpX1lXnLIFmd=I<fsa3%M
zy~|E#zsm<4GH%ebKdH`72n2aS9~FxfoBtB*FWFS&?xBxWQr<m%bU)e%qfN>d&zOX*
z^m|#t+O4scG5IQNaUG?S{=?+F%op<P;~WPPnDwTqynmHY;tK**$EZSWwFbi6Q(yN%
zo@ZUuSpp#9o4lg;asT1~bL~rG9)IyR$^&C2U!H#LV8f*oz%2T|Q~w_~0rh_=SMl+|
zznj&+8P|XRq`}1)*B9|`#Q*RFtf$`<@8Q<f=PwifZJ7W4xBqef|Ah@ea7VeP`vdge
zR^k~t)M9k`8<n-WgP$!$xzV*S&?hg+>jhzMhrh^=VNnO3DD;7@&y-$>e>s~bwEWvZ
zt*|P~rl`D7UoJ~>v-~bYCM8SufV5b_-3(IaY&jlg<=A#s)%*BG^goH&$QnS%ThDvA
zC=)Y(N}N`M;6*qMX=$SS$;qEvJK7cyKr8b&FDsw{BL?*uxUk(bbk<Ov|0R_he@o*A
z7P<PXuA@n>uY9b!;->0;-e>HEq5j@@Q7zj(<ez3ji`<dIzzVPLD$UDob1QkgSz|;Y
zXUd0<*#D6m-T*?*`eHJhn45L3w3=gLyb7-R$*K-EpAdbP@{}_W@|O5eBWBG$7lES+
z2&wt4-v&9obE_5+Dz{vESG}8;B^@ML?S9I!O>uW5$#{^e+9JnY(e~r<msjxDs#&cn
zI$7@{r}T_xVv;cBxb^@%T>elZ`+v>NG86si8{E1o!MgE5OQq~zOIJY$0uI7Jvpg+8
zpuWqsFyJ`O9}3+Wt~pfe*MfLdDYf36l$ZVmDxQnG)RPgF5R*Dl;GHIPmaq9&S~%Wm
zWNqZF!CKY@ZeTh?0L#O_5axefuNUqEelIh`a%4F^$G}t`4{jX}59HF?b%eA$Khi9b
z+Acfx);uqCDMu(8#lR+TLMn6>6(ze}g|%2vP*dPO2Io(hIn)v`NAZs`Wl_A-pFuB<
z`{y~#zaKdz$K@v?y!i)z$V(E(qnOvTz7EoWQOu_q=Z+aSyx&)<C(uwS<h+}-vfNCJ
zFv<goQEi&pec5s-I$Sc|y2;c(ezt%H6N8tW9LmuC*_r+>WSD|xGT^HVT*DNu<&Nc-
z$u#o;|4Yih!S#}1lpfzc4BYm`6t3*1a@OF($t%L&??G-2+#&M7oioLCr-t3+K*m*V
z31<&mUH3K4R$%<(UEhZf)|HyPy(4oGzj2YJOSeB%l)Lc)^?3xFAN&t5|F^FJdBNh-
z@9QUz_o|_BP}oGkqZ+*A@y}gYcs0#pbMRs`JYP+jd{46{sr8u5Q0$7M`mkKaM*b<u
zYkzEl*N$8y7ZJl3`}C&PSB7o0B0?1x@A!_;F5~Tg(G>cBrx~`9jc^P`6!P-qXSw_H
zC7S%vA{q-Lowwft15<20$3@Gl?Ru_VcjkdAnN<41Nw-x_<)`{fk?GLzq@mZc!N`Dm
zh7!-Dg`er$zsb^f@=zQ#OYO@H_f}s11Lm~<tyFE=!*sV3bCskWGK=BB3FQw%LA!AG
z$NNYREP~qG|A_Jq1a$BihIAbTNnSqXwM&z5$-18Lgd~;m9mQdyD7FrhGoesFt)n%w
zlwvsOeZAm}BhJ4=?Y{x;e|`RV``1!RQNwa`{FTLuuP0ujQdXcl3qYvVR|nl<p09$6
z3SyV!^Q|J5+~~_8MG;h;7RRj)x!dlVN={1iiL%#CBNV(;3c=5Jx&zTPyg1kQ1#fYQ
zvc3cTT*m#^qDI?b8~$G2G-NMvE;|eguHNmZhSAhipDklC>w9q+BK=8)Qj3Xn1P%g8
zw@)e!(}Y(NENiv+9I|6WHi|}+ZpovJPg?ierjIYjytC#dajtZ_3h(kYZjbFxvb;)O
zcP?A3b^8C8h<|S}|M~Tscrs`GyHs1n!&MDA#h|N|y0zqHnKrOM<a)p(o4dMM)u|a&
z<FzkxK?~Z{c6Zihy{*GhC}xNfm=xRBw-~_M+t5&Jiv<+8ZX9rIH#%IL;YBR{=hOek
zjVxIJySbx=C#<>Yn}UvrDzC#{K~G@06dHH5`|e29_1qmf-nDc?_%|9nnLZHvpaUL&
zE<LY=O&XKS*p_oJ4J^yF7C=~bMe`P)s6Z`SzyjW6<!wi1HX^JFn68p~=XosF+J7uB
zx~wCza#p#Kr*=ZQ@f_sRflmTM0S}kz|FvIXJ>4$6-<Z+C#!5{-bwdGm{v(X4*;>XC
zJz7R^fcIa`PYl((d4Ox0;t0XkZ9Js~SPU<XMU{#^j^H}G&a(?bbv$*itos=1f%Ok&
zHM;-Ml_+oyA{?<Fw9d+xDEY+b`g>ud0B5RhDA8!fH`sP2hX>#5yTrfzmV;1K4zPOX
z$uMt~`KXSQm$;&_IAO8*x)RD&AKbGd6+-b38+pKiLN>)K4D@y9SHdo{<fuFxS0n$N
zjiHmhPD-oRnK6+R=mPp^a4~<$XGtVr!igy<p`-wk!@csFNPtHoyPfS{;P~GN7cRcV
zSx@OsJV|As5a?~0#BXk=)Wl=1BxK?7sHJINg?#x**KAv~*X<uIxAuWQsLDn8;8v@A
ziukeNt@+z94FzS*Bm94`etT?6J4aN?CXq;@pI#Ru!BLhA<|s<o+BY<_et(W<CydB$
z`ayE?$A!a#d<jw0bV5XGzcG|Z{9_yADQjJ?FLZvu0=jarxXxPtZ~I4x@1pDHVitUR
zQJSkq|F9t9l%|eQRg_%;!tyRK#E^TgKjb)q54IYXte(S_7W`^>`#oyzO0jWcfO?Fu
z%C3z&%PZDm-s264z#B6AtB*k_j#rx*%wAhn(^eOtuA*#r7A;W_hvC%-eJ!LReJQtz
z10C|_>oV=D`EN=`_y0k1H*Cr;j`7|lsa)BAJFoJkM^Xsi^i4FCZtXV<^i67ha;Bu!
zrsO4z=+2ynU@zH^FKvIv@d@zi+K^`x5`i;zLB$u$$Cf3XhkHrX{KT7vz+QmqP2}@T
zOqBn>cjT|_W764@`1Wub`y!7tzw22EfXmU2>t<@;PPGK%>0Y(nhc9Ay`bG95qBb>_
zv(HHIUrSkLk}r)WnU@{?M{N2(JetPw#nWvr+`2p(jvrG0WR~{gnDEm6N0=P;zv2qR
zs`%EMdx82}{ZHinPNe=P4Gdz95<F~Qo+99SCjYN@eT)YSW8&alVNWdie;ozgyWfa~
ziGv#@7~Vem7X<!BMQ|u4SpYL~&^-DlM)_$ljPmzM(P91TH5Jc3#RTl-r|S>s{wZSL
z#{G*?+|=)M|6+;$UN=`NOilsHZ2n63PmC^n#W1=+LmYempL9MfiP8CRJn4^rFh22Y
zioY18Fnb&K|J)G%i{z;1Upn_Z!H)Y!MsfaPG*C(E`9J9#K!njbK-Eb6`9CrGe~sh6
zTlD`mj{hF`2K@igIC3bSTLE)=0(-~W<B_gizu(CJm&^U9D~`52!uVdxDgV~QzM%_8
zl>aILGvNI9_n$KYn*P}6tDMl{t4JQ)T+EwJ?p91os^bYP%~#L9xG2*4Zp>*|tyC)(
z({vizSL#`xN7`ENWX4;4Hi`NCk?oJ`$t2YJ0%VD|M3(hBn~`BjhGXnePhHd;#dZ(L
z$S5cfMZPo`I-*P^uEToG4YU*fkWBEtkAZTH>R$m~w#G9Ih-jG0ZV?w=M?GdnFJLD2
zdQ;((9Mq3#Y{woFL11{Ku%g`CoP#~mQ2ipiNpdeg_+7yR3Zjrkuy$2Y_^?2fjIq-r
zk_$pm(rLtOc#Y=ms!ZelB7M(KuM6;pCQ&2tK-0d_n*mAJ)rLBXt{qHL&vO^+<a&MO
zpWDH(mU(2(+a@ll+MMF{Lwal2KhY?S&!D**sD0m~=ks6(HnZ?M$<zH6Z5wNAWf|S}
zTa@{>7`@$KC|iO+Mz&}R>Z63o8<O8lsspyW9WQOPFm;pP{EpEDx!n)`k~f+iJ5-}T
zs@OD!T|n+Xqa$%|VvN)+VeVsBWg@#pTT6x{Km2}UBCP>*GHzzC=T8%a9Qo;Ka<hsz
zVa^HChJ11u+E`ZSrz(lb@PX?mkEM8QPrZ3i;~F<HzHR#d9<F&TU1k(*B_9+blNxwv
zHu1#wS9f>z^s{Qa)Iu6DVfg4W_K@?FuTkjbdw+I($rnzgS<1^$tW)@sHX`Dsqzj5`
zevLE4iibJ;V;W-V^3(oXt|Km|G1?_YD8)3-KfvlSWapaMmwx6nF>{0OoPs-tw}N%B
z*|au~8-n!r;LV%4n4<b^vys#28?}aQ?Z$l_l=bp-ocWn}%&FCRqb?7Vh3BChe2(Ej
zn_1tPNI4zTIE!<wu6pm*2`Hdg2UP$15%%mVMO&#Q_T}E~H{7cXJv{QACi-1#g;~C?
zt!H9T#_Y56IN9ynEaBzqm6q`EVzUdMGyuZ^{`p?s;d9Y+V~CJ_eXtvB8he%R*7F|P
zZh@EwF>x?GB93~|XD=*3{^<24iGUbbDV}Op6{^hvf$K<cq=iekoWn4F^c=zY`SbF5
zR?HotNZso2%#w5FJs)J<aXSO~ZCA3@@oy6<OT%+n-Ts<07%3qxQc`Va%(uXQrzw-Y
zVPAKPBKLTtcWfS1x>I#9MWuU%6CiN^wrZ}iVT<#n4%JATtgUFo+T6Cqbh!U{P(kN`
z%k>u)6W4_vB0*mbZ>sbJ$kx*sgWgDwJ{uKcbgaAo)-kQ%FEy?xhhyZa@6?%5uV-hY
z%f5ammOVH9Pup%)%1lM~LXOH%n`u4CqR=(j&4Si8=P}!{FOH(a9HMWyB0GzB15J4L
z9S3tND|YH9LBAulC^b~EpEH(qOSZk1W2E52VV~QVcxys03ap#Kenx41)e<Tz!wUFx
zA^zp4UFqvw#41&b@HOl1R-)n}52_2+Oe`?|@mU(FNOV(w2}<}-&8`0)QOoV%=t;?d
zeMapsD4p=}yWOi#*Wnk%P|WcednhwhyNt`0=DYy`TfZ>ZY&$T5+)8qgzk*Qhd-;@g
z$=B{6G8x4`@rKc^WYR9CoH>O09?JBkPO=;jMMZ5BT?-B@l~kT>N|NHHsD7G4BR?jP
z7<d4X4})IKJMZN0BSE9f7X`Lih*y#y<dD^0<dTQkE~Bcy6M*U>+k_dr$26~vvF=);
z+eU1JkqQPz>hot+m$bBoqIz~man$2oj2zV^mlH0Q5cq73b#d?0%WW3wRP!@3Wv#8k
zYMbQAA(`yLeHQQy0ncI<zaG6bI^(h>RQZ>}V4Xg}pFi3zYeu$;J0rr!#tn2)V<W7F
zi&=d8WmSB3%+O5F8x`4$>Wou%Y36oG`z~Xo3?3O;6u#2WxERW5WFr)~cT5+|!tn7<
zp`m64PM}Se?2B4Tlg;p<CJNLSN{kJvvAfZE-NNmI_)_X+*}!d;<7#J^L@l?U@YOAn
z!34Ne9WZED((LvU41R-!ULogq7!x&r_-go6<@5B~G#pSel495D@Tr$1%L+Bp{c!B8
z3I*&awJ+<ji}Lq`!1}Ros>hhu+zijH_uX2WhxE8M9;f>ofK8Rgw*Hj)Yo!#Zh`nh#
z`HbIUN*#W<Wg0S4JCKsvruuu1TbJVORN1zySlZXFu%yy;JrI-4`Zan|KgO+#W-Byg
zbW_X0Oy)Cmx&FyVHV;t3Dj1)3a1N{4zHpce7cV>=tgCoD!gAZLEtA$+Tw1pOg%8s=
z;%;pab)1bZwJmaDG$_ftB;dExpH9#_)riXH3bIMs8Q4u~Qgxb-Ha`3EbtTrFC8Osg
zWraQ@HRh4+WPT;U<h7V*>NG(zx5)E}JojIg)k^+yOuDKmjYD=Gn^lFDX`2&Qc<M>0
zKw5vVS$TKBa6_*J;mYVPB}ZUfS*uOdxnqCVR9cA_G&=+WmUauY!t^BiYDK%Ydn16=
zrObq7QDIqF%BMGiC0GsV^!l>WHFa9~yV5_$$#snsEMaLG?mk@^-}7Yh)|_iDpF5uf
zAPSp5AUs&`^VMGB|3brF=N3$GV4GP@#j5*7YhQ4Jvb|QHd{L#mXFw-Q-=Ee@i)xl_
zl1z$jps1IL6)6?3<E3M>6TqdK`tm?zg!^d$w-DylIQNzcZq?)|Ykbv-LTMaZ-y}$x
z4AYiLQ}aHs`Qx;KXxUuna*~z(&stG^8xaBT5MEOx{_ZWDNRNo!r&F&b%<{1>IVX%R
zQFPL}vJzzZM$p<kPx%_p;toI=E{#S|3&`N@^H;2XR2=I};<5moa*WM<-d}(%`+WU9
zv|+!pH@X}1;6!{&+RT&0ym|Q~*hn&!28zzH1s_cxTqZq&=Nx$Ne=&VEm(|!yZYVsP
z)-WU_shmt}PJ_>4s}N}?CLr~EcKzZbu$Q7~U3PL{!MIu;FvmPv!RoEgxud@F82wCV
z%f0vvBxCa6^wD55C32X_-ZH+JdwWkQRe-vUjGgoV-#R#Ygr}}4)EoF&4Sf}P=g}QR
zWp@6Gn+sD{P(`1O+Pvnh->~}nhwm8>Q8|{jP}+XWeqslCGi&LP)d#NeJjmrY=>M?L
zq|$goNvOQcam<eHr7#h_QMH5tOVfn=w}njJw6v_&?(w;w1tlPluc>#EzPVWK#EytK
zz`siHgJILBtOqf(5I=FtyHiuDExTh#;W@v>ul}pvYOf1Aa`!`KA>m9BDbwDIY~Iq$
z%OEl9TMIpb{qfIN#b;=Xn|Rlwo#Se&ZJixP-Yl=1lUB@tfZkRD_SCz`ZxTF}hpxz@
zc<-Vx*cpLw+NZHm2gbssY*+98Pvm~nhebt2{F4*gc~b_@X6hw%x@V-SR^a{+jaK)p
z{a6YQ<k60uN8_<ugV17zYfrh$c1*A_BqG8gTMg2zVhQ$!f+PD*2F#ZcYr};#KJ8dB
zJ)>*WzFjG<wj&N#_hNVV+}$gDV-zaJ@ocKcx#QbZ){^OnhL;P1Ty=3n{ri0MKc12)
zlYObsL4G8#zbQ^5ns_xXPdFhrfet}tCa&iaLh4J7RL^cU_a}-}JlUA?VQpGcc`_L!
z0y!XKWs-duV~6vv!@Fr|O&$TzEN9kp{ZktfUYDmA)NNwd8L>Pb;#m_oxtj|@MHXVa
zVxkhYo~JxhSaW;su$N-8rX$;uv$o^@SYXewYo7N_T^3*gG=DVyZQ$F-xTDv!f`9yo
zikG6>2!AXf9r?8;cxD)v0juv?rH3x_>5OGXRU02$Z%a#d2T+%5)|G!NsnvE7Wyr`z
z5(&i1_*uRH`>RCai0Y9SBPWxs=A!l3Ir`rtUl8W!u1K4;oMWTyFOzqob;T6CJE6p$
zd+EY`6>HCdW(Hsf?>~~W6pJ%5BpjM*3jr2{RYc&6^3(0O)d=g4X~e&3w&4c)n1k?>
zaOM(ETMl|x`ZX`VsUNXCHZP`s`2nF=ZOpbPLcKECdTri2enaO!uKGSkDhj7PzlNu(
zezawftM`YpY1&t0C%f%cft0d3B@8S$3L|7+5q6sxYF;AvZ0xj_NQ%V@DpgE<K{D#b
z#XcS@1vdIyBrgc%5W-q_`inUwI+s;etOjiouA<alP~SC~*<&M|8t1!}Ec@o3DS7X@
z@`4(DI3=~RFH(BCDfIMc=YW?(JAS|rHBte|tIsTaMJ?REGTkUI_F8pB9!v@6LwH=5
z!0HddGUGLdlao25(Od}k72e&sMje(SwGBZ_Pq=jQ93D-v=pwQ0L3W41a0aUmXVA7m
z>QtgP;TVV!&5;N!D_o%Fn`=g8IZ+w<c0c$~dENa-MMnTojpN!(>>hNhecrE8s8xj`
zSOz#IJ)zN3b=#X6v<DkN0s=!v>6cn&U{p<+KJK2QF0D_wk;Uhetwh8I(MVYT68Q((
z2o*>nKXZ}m26>5RWf+o_uu#z+`OHvkeAL2;vaDi8j)a3A^nm0nCm=9nbVDbS5TUDP
z{A7FNEHA?0Cy~nv@PpNY59oUw<hS#s)z5Nb*KhN+g5YRV>ar&)!Y*_Kf|L8i^j-2h
zEEW8jVs6Iozpg#8P1e;(gjT$T480h?&I?hOV1e9qKQ4j)0Ysc@u<Q=lR<xzz7Qjv$
z_nNzWc`Dj*1P@|K3G%QH{L73Qz8QI$a@4%I*b=bMr8e@HurltC55f1Fxo&NU+gP;R
zExipK&X16da&vXd$UdtpLAY4Deg}W)2bwK-E4Yw;ypxdTX}ofsJ?V0oU3vufKkSKZ
z1TQ$ED}+WBAebW6x!xEAalt&1*=2BJ*`vz~{+aqpNAO_okkh#S;+({Y11m&N>)>j{
zKCe|nsk^Xt!EKE24rFfo#;i_waJs%;fZJiBMQ^nz`P2H4!lz7g;aKjFaj6nn0y7V|
z%CAI4tc^DhWXWp<6|Y={b%u!8NvhR%VBG!dxOuZ#)_H0zTWOi>whHmnRn*w@ByGOp
zT{!?tDaGFMilxj`u{W6;rta(F*T)?C3G6XRC|V~=qcAQ}TeSCz#|+r$B|`9O`A5@T
zI(uMPkFPdJtDprUvBA>hAy#`tUd(qlbzPoRdceIWLU;KUdU>%cg>vj8yAPlYU>R9t
z@zvV%?H#&-0yb{n7Q(}Oqdn#)ZQpWWJ6)gU_#Ki3JQ>-*UrgcOoH`Ch^bsa5lrmPA
zA-pO?pVED3?Y9M?9c69JU!553@2HUingmRMsr()9F2}R(M3Zs6hEh{Y_gX)uCTE5{
zRorrUH!Qf{oM3u;b(?cj$i`U>68#0eerI`tJe9x8tl_&$r;?>q$zMyHE+Tv5aJRCR
zGe6;nY4+Sgx)JA37K(uKG#i^Lb0BQ-RVNu?g`-^fye_NJVYJw{TlOkb-0@Vb&8G43
zAO}%bt@cXG&97_Y-n1aMqRi(;N@xmIfJb@|ujitWM(q+;MUl6Lo|#_P4YQ%SjXPUK
zsn;XTgr%>U5u-UKE9>RHtMwax)m)<Gig_sl&`aH3s}_c?AD5<YiJS%kGl|XLpUzz!
z&sndEDriC7H$I>TStNCt#fYw4!A8L}WYP<B+>ne`1CP1aFGG%-`bX==ZztU)W)_ne
z40CU;QPyssh7i}^H;UB&+JkQ^T}q{v-k(NZTz?<FT!|;`UF=oa6%oq9M>`w;zTq|j
z8tzd|F~FD(qpn}#m+gx7SF6&^v#7um1S2jxjYQD{;@(9K)ck-FvDw<9U-{L*>ssq=
z*+uuu`0-$Qqf;BZW3Bs?F~z3D{9}4yJ3l9QSsu_$#gx9+Y#fxf)OGk&vc}2$M-y4y
zp-iz{P3?=ug!o<5v)$-;#_4=fCS~jJ(_^i>(Dg?jW*zH5i<j7sCHoDDkt+ywf8gSS
zXyfyhpCi_NhWxTH@Sdhjl8#z*b;nR&m=M?OGQ(i$%|l9V-&kHtgPeGx#+#zu6tck&
zfb*pDi^jX7tJGS@1}$AfpjL9_ug&j7TuqUvBq+H#us1Xg8gF#)LwUVm!8Agh@tzBE
zQfO;k#s2jcDN$W`cz8z7B>PsQMpx*k`T5d5W_D*)IX6TRxHjk<pTDAI=I*0W;UH9!
z(hmXOUd?fvVfGOsOtQ3|F(DQFIm_k!jV&--l8w}De8if7!-bLsRY7v;vCc}aXSdZc
zf9_>%)TuX&qM_PVjA8b+b!F6lm8zA7QmUWx@;L6Ay4fKXF)1a|g)#iOC5j(pRIWMT
zSOMNsIH<F4-BZkb-zjL3>LfTo^<AqP^Zvwd*m|omFWG5kBWmy>GEN%n!dg|6kd6*B
zdm<eWf2rv0u@V>{48zSkd(-3E0q{~$FfKwK?EG*+#sg?{r}v%pEEh}V>gpOZ3De(;
zyz`Is{!o9yA=Vtz-1eujfvH$E!D!46J#8A{P-~%tDel(6x$iybcaX?ifLBR%I7z}>
zW*X58puY6V`dH(>{b?#tp>6xhY%Y%wkGyjyS$H);q8JLBt~B$P?X5gJ3uInWv;dsU
z9)5~G@u^kL*n&lb!mTg28dlXZ#`h6MXpK;U$t!$@64(A}1W(aE#lgq@c+AXnfuE^0
z729%@SrbbtdJT328_uNz=Q3ciA#I@10msONtaZuKHa|~Hx=7aUK-binylF;N(dyMj
z#-ARA3}_~I2>IR0l;@WBdb9b!nYe%c^ysgt$z*zdCmn0uhTetfPO+;BL>==bOxj=R
zx&da*ucfPRO9DZZW|iAmf^;XRS6W;A+{;@T;aPGGpp#L^THLL-e!LQ%l@lr(RN&0G
zadh^}hiP|G{C;oVjli)p@(j&8<bb$Z89p>(9OAYI=Q7e-`FGU=lW+I7UXQR5FZ?{a
zj8V-m^RrJK#)Hc@KwKKk4e3E<6H}aA>S#kqeJ~lb;ibWr1NCy4_smuC6lE(i*DM||
zf4&rTz_mvl?Cajk=8|;S4(RsCeX}%~?I-I)+O<3tJXqWICSgE&334daeV>Lono!@+
z<&iN9Cz+A<<L^QQRbBa=Fj^00(5**gRMgUP&+WRAp`Hx9ej6=dCfh}we3IO{Hur^%
z6(A}|!DjwZ1J$r2umL(5cm>6bCBrTAu&L1^tvWOqGB|uWY$hwz*z6_7^g8cbOy`v6
z0hN7K^MHq0X6nH{Jnl^oj5dll;Doj7C()PfAQ?@6WbdCPJKiUP8v$>fU#9nz?nXCg
ztVJUgHfJka2jEY~Jj;En|L{@>RvpJf&x&!3CuBb*aJKm^7}knDu5ynO9jHGPR2Cew
zVoSZ!iKa?VOe<d-*c&ESGjHs8f31O7zS**5uWBlGJdmPDHQL(ByqMl>fTq0=vmVvt
zZrUhI@S$#mX<i(vw0hy?T($ogC>U<y6EjIq-K`a~;Y*K~MHE`6_lj3fxc}^&>U99E
z6W!6coT?%=8B5QGMqb+XU#<6li%;I+o0~Klh@o=zupjVL@nIbP<PcGUdfCw$PT}yA
zy-KHp)|c#zJZE)+qam){8_6}Y{d8U>!-TJOCT#FnOjYR3Y_e+ya*k0Itkx5jn;)<=
z!KYgWuXYl05!q6hdIq>1z3AOLAu&pdKM`?9iO+%kL%`%S)8+=lC7P|zMMY@w8sl>s
zH)?2In=_u3o|N~HeZNeaX4yRTYc#-3g1!NEBMs%puKn-wLms5hyAQ;sT^AAvFy8=7
zJzdE>@ScFJavh|ap6mAp2NgwSZHgw1#DCY>Xj3rO$`ktfYeUps{i|vFq{Pn^NotF{
zNcVydy#p5#<=gT_@Xx9u{WzczPh<q3$gV;;vCVP7PTii<H1<^YLU9#e&eJuT3ZL+q
z61f%0ma~XJ{{iuQ-8967xlo3Z>bZTg59`3kjEg=8$HAf@&TP^{m#E$5@+5cj;F6`f
ztlfkNA9DA#O#*y-Cwb5Z_ePCE3Q3rjJHq*;vF2+s)eIHAPxE?V{YO>m;_401=Nc=5
zuoXdPf8m}UhD1>~4U$5i4Z^p%^DLppp&`%Ywmu=tHC)F1zQ`c4WYYT-p+J+MV}oa-
zK3YckES2ZxcRNgGO?e;BYjHTonn(2fjGarWGUe$9ONoix-)k}|k#Q;vy6c8%GO~6I
zMsJd4EeBNsvau*|C=}~hn_*6QgBi^wV>9FUDs8UMM|=2Hkf~6f4}(^95e2F)EX{*>
zt_z$~s%q49E@ZWh$z1t^wrxv3x5F~gb~o#+&{42jbM?xg04k8Eq<BL&g&;(iJ}HH4
z_3j5W*_wng>Un02aV*pIFH-ufqc}Kls){1_8PE&>&T^TBTjh(L?ExR+uQIuL$?Ix@
zIX$Azi*hQaN*1j}A82wYs)GdM<LOMLSt-qb3-726d+7fBQyS<x>KxM0FyBBurpHf#
z(#s@roQ|2F`jRY)rAmpNEw-t?^BD1gAKx_7qL0=T+_kT<KWtO#*0%GR-QC(ZYmgLg
zj*o!9x73c2>zzJVxy@#_rQ=s~1qypA*WiQ>g9_&$)1!(gl`+t-Zlm&?fvRE^p2JFL
zNvUlV?la)1hF_rd>d9;XpmoQ5`lxDoa>U0~uPiRt&&`;ntl7R5Z3f|ZS^Gy0jadJz
z`ASHCI_6G+i^|B4YWws$|MAVxPZhkI2)wFDp+=eJ%lU8*E#n#gA3FKRpYUyFdX!Y#
zlWI0v*Y}{?Tsx;PP>}S>TipTsr=7}?y0*s+ym~NidQ15^Ckh7PqH<H$IpD-f#Sg<f
zm4uu#-?u6ij#<{J@$s4;>gEJj$-i|Wm4S~aaqDXH=cY_tRwnYA8exo;1AvS=>-eg-
zj$I{dkNAEb6PG8Z5Goo%wZ;E{rb5O^rozW3g~JPEOrB3IBAj<u;4tCAG0oQ<Dy%}P
z1*+upnhIw{SqaCO|Dd;}SEr7<4Vg_S#Mmaq!8(r{$z|;Hf(EtuO(9{dbBq{wfRa#y
z{>`9g1N2QY&p`8!4XvI$i=U}UHAS(`3;h}^9-`yFnVP)#eNH}<dJ*~06vfnzXnFcA
zas&7lt0UVZvkfx`&b-IzO}EIj^el^s12(cMgnE#V0@I}ud%Z7ljD|DImvn6>r>srG
zcYDOr`)R*3=lSSeARtt~HkBH>8oh2NGrdg51IO=`4&40SaVcZ`6q;O&CKR)(4OI0R
z{mi?ky6C++xD~@{MkK{CE2p7{K5&t5DzjCCPP?MaXZ<HQRBPKNMH3U#r;eP#C{Oqh
zZ^q#Pm%NMNFdsu;8-A*{s2`l^g?34F^X~9_{t;%U-5};VCky$$`U1ZRuMya#Tmehp
zNR&>k2w^I|XP<L|wQ+@sYPnXeMX>MN8#aUKjX2>vyXYSc3sy!Fd}AvE)x+hZ+?cfT
zSJxj29XT?tjeUi*6)+RA3Z|Wf1{GZ~j5}#2%07AHGl-^8m7;vM#5wcP1r0}frq!+o
zQ266{dqSbXnE}l$KJ|9N;lR}j|EuEc^_?6E$&?YMJZ(BU57DHXUSIB1NQVSHiDDX{
zG(!p<b`t@ska|GU6j>@R)m(_BnQHIfI=HW;gsKu`B<OwbCSA~brIN70)KX@-U+TO&
z@x=72k*QBYjyn09h15T{;x4R2x66Q22eptdiH0kpFfPXA_j5qLX=PQz4KLeYi!g7e
zk_*~+taP8xAsuYgFNEIX3;JrkBfRFQ4`RfVke2@8bVUOEgs(g=cp4L8yTQ>Fv7n|1
zdPROSc;Z0@OX`MZJiAX;+sf}Yp}uqK>Ltior8agusWsyGIK6;&L54YPB948-&!GRg
zZTBi;8RMQmOBwy?5QSKe#fu~|@Qo*$dfs<v&zEjyxj;r}y)9GHi~bV99Z7Q?&v^RY
z-gChSFs+en<~O`dq=?ohk1`Y9jMh=-+EF1FH0$M=PuXPGQs+9dw!K|<n{i03a2#F%
z6O<E0o^xA3cr)>VF-(7N^DznAG0zt|#s*b1uNBBIDZm?V3JM9v;2&>>_T*K-qo6K-
zUWCNdnsZv)QVdeW>B9N4>pk_OB$M^GP58145q#q*+HI)U<F^;#>t*4tA=L~brpf1)
zpW!janfU(GZB<E^JT8~Q+#BEqcFR|438zY_Q16zBemgl>Y=a{HACzaKmr{-5;zl9>
z|8DY}_B?OOJJU@hv4Td7A3w&MQY%EbYD{oWkyPrUQd98#K%x22af#;xxD*~M%m&%6
z@^1(zkrZIn@8UaX&cHj!3O5D~{t$T*iW+r(nbDt}UE7etak0-e2|6E2DSUlqQ1LE#
z?%;k~=saBRHo!&5s7ssIWM^vRkJr)$SgWP)r;046qJB?-<+Dn89hoHa%pchV^@ST7
zUU2~>UDteGeTl|}UKR007GG^moo4o=_xafQ`+tiCd<ad1Voebgoi2rOnN?7<(mlLP
zO%hy)nFMhz@WL*-7_GI%OdJOv>u;T$xp@}n&1=5Vc<}B}aac<F#{>9GJM#Faae)sy
zT<5PyaUNIR{{<jVd!9`10JSZYY5B027;<J69^LzC3}a0c^Q<flZD|)Qm4#8HnDkAZ
z#}96HrFOY8Fi+IBi(WjxnyIEfOp7vFq)1eNKkQ+y1ibgw)~(U1EgrhnN!3$#%I^mj
znJ*5EmB;6*B#?mm&kx<gValK+$_E;G`P~kOw!#>DQ8jdrJSWDj;~P`~Ca5*)fN&@M
z;s(BF1M+G`RX~|qt@kfH*J7Rn+ebm_zQ5tzlbL!Q=Y<oxQ`<@BpVgm7*@T?LH$S`N
zRlz_Qr1A+qRQ=l<^CLRdDj`nt-I`OrGlul)-QfaGRVgLs34X+yq(>O2$yWO4e(l(r
zL+ycaU!cs0y;IU+`mS>s)2iS=Lc`Vf=rnZptgi^ME<wqC%eHziR`3{1`TMSHt)$+e
zCqkClP{PmtRZpvWi16x;9T<Bt$bVG_2YAXGf7Kq=Jy=dKuYzDBj;z}G?A&bg7kXhq
z6m8AwJ;zCQ%)YBgTsWNj>0gM`?79@pP3C-1s4Ec-;!{e+S9Pw9o9ae<jv<4R9*4(m
zl~qk1Ea$UM`&mQBTyEcUq1jLhC5Fu9Dt~6Rm!7QBN#?OXZTtWcT9_0bO)DHORMZRs
z0hsIixDGB;7#6py_7JMwc2f$Av?I5QpXRzvly@#j&LUoH)v*sUFBZX<oH<klkAUH{
z?EImRF0;O&$Phm;cm+tQa8%X^!Iqa>TU;6R+@0nH#Eeg>p+#XAv^2Iai9S0HV(VF+
z=R#C#!+dhzZewHSZGLbF*VhRWnnu75<0{Tv$c)OO^%P!CFK=iaFbg+`C5cg)N-GAF
z@zfvp6n*JQOqhg8sL;<-Cp$YhC%b0Oe%E=og%;N6Hv;rI{MG_T7RrX3&HJsHGb0b2
zR?MGFY9(r)v-kikx7#phG0ETuVnN>?<|ep~f{={PQ0EbRZZ+9;Dv^Mv!9sO*9uzBI
zsju6&jRD&!s81Qed$)~+yV1Y=HoKO4pvIHWizd@+dztwgK1OzTl1k)f@g*3>JY@0J
zGt@tL)tAt!HupS1D`}S6qR^M~Os>Xx&i#YD^P_4GLoax={*)nJ)0D8y0@v-!((0b3
zHkDjmHqVmkq+}%*(W8Cm99PrKTjxX(nh>VB!fNEgR-7A<PHUc2Dx<NK{~6!(yEds~
zztWgb@Rw-2KW?RP$ahfcfMv7H4wo5w@|i4wT@XP@Jy$()wzAiN|5j)^W{=;xW$PVx
zvFn5D29sAP1+tUx6M~}08@GiD6}*+r`+KS-U<8&_p53fkw9X%vzL3x-9t=}9i+cG=
zcH<R!ipf(&j;94IZ_TYFDJxG491+a6q**B}yOjtQ2YIs5K&aqC*V}|FC;HMd-`jy<
z)K>(Ko>{}xyvvviAabvAG4Q7*d3BB&_rnD***ymvhi9GouX*@S#6To><P+TXm1$z<
z?<8cc>W0%?r9~WkO&x;sTCtS;V?k*IpPDoLyVEdzt3q!&q!XRm=3KdJX$QIR3h{z@
zEJeO=-E21hKpEq$j+_MWRHa?Sz&Zvxm^2zD4<>lo`K|Y!3>QtjY23(RJI{|Um#mC6
zpQh0zzf?0n0RBo|4KlBUgVceFZ+C56OeIhFr*&HDY9+$?yzd9<r|0ZjU`2$}kXQhu
z>1ChY(C+<}O{_A}wB-D<Z(i5)p-f5`q>d@hE2JQ$EM&Zt>MHH3gRD~UJ-X~InD+*Z
zzx~_-FT4DCe}x8ADwgItk0=>`ngV=&7S5<dqacsMqP+f^ta7wS=RWjl)&4Wv>;4ws
z39Fy(L^O48qSDSC<?PmXqsx9RO;i!hRW1ta|1NhZz~8Z`FGeTYc#5&Fg<JsIg$?N~
z*S>T1!p=_o1;M`lEu)~`PhxPX+WPvoeSQU5IkKC)^hWM{gner==hZU>!F=|Uruepk
zEVo1v-nh#8p?dF(crm@DnZZWgt^1%n{l!U!^~J_)jZOQW{SgOmYw409S+K=B%TMc-
zftjUy0N+*S*}l(u!9u91@<VvOAqSR*=en5SXqY_fZdiDi6ZNUgrtFMgZX=L;dL!jo
zw^P!yQ?swL_sV91*&ZZPN4VfXc1s-RPlyX*n&lnQa^URqOvq6FILZ1}$g{GT*A!a~
zqD)=HnL1Cmk|zY!8F~Zi$4Vb3WEQ<Y^G(q3tyl12c>67-PMz>hwAjn-<`*p$5q?R&
zV(D-w(<XdJkl#c-<YMBuahOWq2g<wflSk7b_W6vv!qCZMN~srfx%US@+ItPtCvSj9
zi5f-Xb!r|L4!5yIeYcssiqE)kEOz)PCjfO5K^0*RVSRS5!JA!DXRT@-WQGS4v|H@s
z&~~#vAC426(aPuNBEvQ@M_dOH8t4~$N{v!U=BBL>!tfu>=Pih}r0L*SuLyF@&g6V5
z;h{rhB28YJWY}GndGsQ^n+${zA4BY5S$l20lkqn9s!Z+{+uM`g-Kd<iDw93#OY8B~
zOoavdltiXlAOKFjdcEFFBh^#}RoxyBBiAYj>_0o&;uI}#$*Dc79@o@qd-f;(;<xBl
z5oj{(nBsf&^Y|4qq1d;hb<+(f@0}jmXi>yZGn{M)Y}ef_bU;kwZDQ-sD|0-9U}6h8
z?gIJ2cX1tc6`hyv&CLh(ExzbvmApz=c15Ag=*6J7g`Fx}U|Gbk-c+&HgA0A;)x@R+
zVsqu42}-wa_fCbwNToYFNsZQi5t1lMP3;TkN15|F^1~V2NG)uaEyb*F`DO_l^I24P
zLqi3Vo5a^Qw^e913dwV|i)Ks_JT1YO+Cs$o)j(l|P;JGP!RxhQ?0(ZlElL8E&<SKo
z5Y%mac-EUu=@%ry&AO+QX{o;HT2IO79Z^$MKe|#lFnRRuw0vi6o)|O<U3MnZ32lHz
zfo$S*Jm2b!3hW_uR`5hG)FpOex&<gr9ro1P=If#H3r}mUXalX!o7~q%*!LGx8KG0H
zI}zF4&>Ie${G~0<eN{#=ff0-HON89MZe4#*t+voWuhx5aJdRwPB(GykMF<WK3zx4#
zj8^a~49vPFh@L3sqWj$BMnuz^m3}G}un0J=B~ie7)2Je=2PYK#Q;!t8@VL?aA}R6%
zFI?YQl>I#LGs+k`sNXR6f_jt8c+n9vMBk<GJ2M1M_bnqhiPK%JC$FD&3+otSnsgYw
z!q$X;yemu;e)UP@*|zFK(~<9FyHSF4yBa_78#Z`_L@Xn@ZYAINzso9`uZHxLuyR0|
zOH0aIsPp-+`gA`kBqySfJPJ>XIXWXPhF5sYRD}b5J3o;vX1S|spHVztUk1_~;wNNa
zI!hOy`r}+LJXx@@84*6Wj%hRA-$a@5&TR+!UX&zq88xVv5@Oq~i7ng~aX|*k{SR^u
zY)dZHxWP3k#|@sz3wX2UQ?Xv~mT;ly%ml{$+5Oncr84uqglQggCK?y=%Q5rerg7l>
z&7S}touvgY{rlbH4)G8B;<xbL*Kd$n)e#JjewO9UcF?JCmC*whSfpURA;2GqLXz(7
z<$vcpJl$)^o=4z3$#<SyIiqr`)6&(k6>(Ye7zxd%#M7xBFWJwOe0|1a#@jMqZC#d8
z4C#Ex&C@D?@(#FCdThwQ{!3Kub~RdW_YAHRIQC&IF8y%J#X?etdG1`ObYCS=FRszN
zo0U<umCKTZ+MzP3<RoqqQHLWR7c$B6$%KE7QJq)6iI44kj}6qFON!$cPc$Dxm?+x?
zVXsAK_C{X@o=%53j9dnHfF)aaB(L|mO=`7<y`yIoyVEJk_JzYs_&v1_Cr_UG%XX3s
z>>tgT2QOR#OpN+i>85qq-N%<L2KzTopBc5n+aXH!hAr*=IP@_)AHN&2+)g@iTl8>Y
zUaSDXRl{BuE8t-la1!r@`r7PrwIgR^@qEdKJwcTBH4O=W<S0FzAe~VXzkg~vrhLiU
z+Lhj&Z*YL6yj_9=ut*qZ2>P&IM%@jxpY)~@`69+WbQaqvvvV-++?{?JMbr;hM{=ur
z?u*!k*GTLb-p;@+DLB8GyW)?qmd{_;Qp&JRuUNhuc0CcT29DlvHE%KqQ3UK>twq6d
zlUp3NS~So1a-SLa#d|Sa8*QMCRAmJ(s|S&VCzeBdn&9Hns?8Vv@D+=g69}q)AP9UV
zr+KWxs%x)x8dl4^WNZAcZc~+f<#+j_s{V~ZT<oIQyZCy1`^%~WZEE}QCbNp!MjLi+
z8=-Gpw?%F$pYelM*bYkw*x7U-t5umNlu2s~N<t=Cd!MX#Go5X{I1fR6I~u-GT&UFH
zUBq%<LyL0f>C{<nYi*YNP=4*Nu;ijiK{R=y>V{3$sA@yYn7RDUsAXMvwZ~x8IGYVz
z31{m%cdl)J#s2gokdlHt&L5F)Yfl<nRt22!Qb@Mcs!f+}7=ymlN=rOOSr+K41H3qh
zjuO2HM?X??t5(gIfd5?L)n+UdpQx|_fiuRYY^lz{@qQyFGV>utJn7>lXFq23l?rqZ
zc6bMqT4r?%lS?Zc!#~UJc-;O-#PRL#h`_Zy{-~pxg2ywTy1~4JQfB`(SX8p>-B7`n
zwI#wTd+z?iwq`-LYR_Y^^l_dDsM+8{1yX~e-nV3v?5@_nb{BW#b_;3hu(oOzcWE)`
zOAB@`Sf55#z_U3zEpDbFHthS$Nf8a~N!}{ptUZe5MD@N4UAEnY!J9glEX_K8g^Kpp
zVc{dA!Sj*g!we(C9xb#cu!i|EQL9;&a~fnB1}-mcZk)6pN%*DIayiM^Wc+6&e|VCZ
zQj<~(<)sa^*I-;@fKTgrHEBouM=t>HOa2C@f!~{JMz8E&q_4@i7^j_QBUITheTfGo
z9E5B(H{s(9LHg>dT*gc8L9fj;ol0*?w18`aw!&!da}~6H*L;aF=b1Us<l8x;O`*By
z=mve{cdm5=?lX-b6as%B<2HKQ&YMX2TwVj1zegH8qW#i&<N76ToiNV6eFwdVS?(U3
zQ3U+XFlC>`y&2Qu!`|I-kPo=L)=7QDgFRb5hougla<&aS{ED-$@_%T1%c!=xc47Bv
zOQ95Np%jM}E$;561&RfV7ccJaUW#jR5AFmH4yCxeyF+kEkdScF{l5Df`;5K!_v@Tr
z>(5$QbIvg{*OdFZZoR)}YikJ-uIc>e+49cx(HA@(OO^WiCf~l6t$xz;zVTnNs*eTx
zPjOGB*t`QfnNJDh=i_)kdG?LEVO{xCdPr$swm9FcK3-XFVn~AH!>>}(w)XolH@z?q
z0AAq?a9dTgdn#A#Bk;y&p!$ar>-1MYp+6B-y=T?dR;n8uW|z(D4omf`A{ol}Dl`l#
z{!T+p*9>QW<KiqHgf8uZi&^BvFFR1opho^JG1p+-9W3F*DgTB=Wp|0Y#p*J#n_F;-
z*%Z3YELC*D-cxdR!|v~GE<h`ll%FzH*}*@mz%xW}ldCnCyB*e&-Fgr`5x1-dOQiKm
zOJ?Oix@K~9(yJ7Pe-_1aIY=@<24t#U7b+v(^t~4(c(;&-?Oc`39=%Y`^2TB&QrBc3
zxBX!uJYJ_0KGo1kIJLuY0`)&X?JwA#(p2A{W=(nhv9}|s?wgCQWff}McUIi#r?8$+
zk+zc3Q_rG6u{ZBo$RBg$78INl{5$o%u5!z{Kr~#s&7Z#F%?;kD+_+H%m|vSwcjAtV
z4=>yMOq}L{+Fw#d5EDB&(FSE4v=}!!k5znfAI$zDZTb*LW#q>{7FDS063lK&k-yzO
zGSgo~T9M!-ZX+QD{bJ%jJO1I=y%yJF8^^gExUbjmyj;;8(^hMj&i9xWC#L>qiOU{!
zqCBGjnh;+%EHg;nfYGkSO+hOk>h6-x6prBe!U^ya=|d)U>9?!yrXR300T-L{?b}c+
zP+H4C*@ZvcBLO(eu5LHoFd2LKJMR$ya=O@8U9k6Gmjw;HxSDpBWyg~lU1yK=m(S&J
z_w%Q?^F|a@c~0$_$9;hlcaG4h`dEzj)vqfvjgXH2V*#J|(N*ay;J&C;t0IORD)bXp
zjlYxXb$Jmm2$<ov-+yN$%AkbW%2Mke7OR})U&>u8e;wPhCxyyj=?6lTl`nhH)&bZt
z&h0};`hVwjv-&#bGvFe0^$?qVwjAn0yVnlQJJfnoYo=Ici5m}`bkY9|ZSedQR<lBz
z=MyXS-m_)aD>gnw*D`3B(vr@)+!4ZQ7ygsMKh94ap-v32TOx#v`*qRS=wzo`AMeo1
z*i795f<dtS3^Alraa(DVGh!neU)SB?c(G8J`5Cv~43e`jWVFIGHMA*#r=bZ*QwO4&
zmP>9DaYZ_M{fWY~0ucI}A30HGRdKOk{<<)CWf$vn5mYrDy^S4SU4eEfWk)sYnF2mn
zB+FatpE{YjhbMo#e<hSL#a1O=7ZDkAu}8+8$`@)bUYhnwggE!+N1bbyZ$OOjv<3Ni
z{N;FZ_0fme9>BJ{(~fTK(4Cm|qD}?fwkmh?pGzUAE?L{zHOs}Y!P2bsMPv~iwzl15
ze#uhiB7LWO%X?DFfg6HDAv(*nn^p;{DBhZkp~TXON%@8Sun8HNxp18K(!ZkI2TGEM
zh4GV{i*kYwP$n=6KBE@<SImbhc8rE8rms6$mdn<vE)GfhB?iyxeNq5v#-q)mf~giq
zICD7dA!XB4#dK+yjY!Gtd#AJGw|bhCx(^&j<3?^?E+Qxll6F{f8rMAJvyiB%Kep2Y
z@(JVTE!%8gcV{NtO5m*ep(gJ%ph2&<lz?OQ>b2^q{Of0nzb1p;T4Rm-z{q~kkLlZ;
zT%8OSnlH)vRxFO}XnT#LHuU#KHQS39p9+@qTkE}6(2_auutd|!!NK_+H+-0n(c4gA
zmQB(lv@+hRsZs+C5DjY0bnW8B<?j^|e3vaQYL3r*Npkb-grKn*ducb(v?U$zwF7k=
zFeaWg-a^hyVXC|N=QXPFwh+F)Udi2JcGrn<?aMw>7jD#%JuC`@#atoc$}PNYwuF*A
zv$ie}aPs`As`_(Hs}J^&)}AqTrA8KbAUu)Wt0-kgowD)RP5|$Xy<Un?lE%++=sKZu
zd{C*AWox`^y&>pU9+5L$Tf_=;+5Unx&YDU1q0uT{D)m>@;qFXMt1K`y$Y;zFi<vvr
zWA237z*7Zr!z)5#2d`2s_9?G^^BC4q(d*@{WVPPw=sZV;0r|{%_@=Q{V|{x@5KlNs
zgYBxN!qVtUWzR(mmgEYZo055I+xlJ9cG5XJZJE;F_8~v?B{29m)e9F-l;~tFw*@ik
zCiD^=H;)@|a#n-_H)q>xV!z=L5MJ-Qn}@lB%I%!zoeZgXDMezGgttwY4){u~)_>N6
z>y@3Ye;gPxt8^lH93+hPo|4<-_PxcGLi~+*V=#Ri!rI+c@F>P2Wn@&;yeJnL;+N$2
z^@{NVpT_Gf!P`ROYdY+%vtbq`<Zcak$@Mzx$nwZ&4+JT{To9IbbbTO~mIND_khqSm
zE$O&rG;f_GWbVEYyKbHo7aUJ?PBL)oINUeVBxLaZaF7zg16_8@woJo}q&dA(FpVX%
zOKtTj)~kyg7E=Hz0Lwj8UDcBk*fhu0yp~%*rb7ZFzTPR?jG=7y4=fB6tlcA9`iHTZ
z%i#*!qeM3_Hg`F@)7SS$B_D|=;6@7dBaTq#4ybObZpZ#;m54M$HvAd<uz-$z>W_aj
zN1Pqp)EY}2>GlML)jA&o+cX&~Gd#fTJQu2o@F+Mh6Y+_-j_65ZJ2qU68q?Mbekjj#
z{}anHANu^F>&bOLa8=?N9Qs0f@NfY>bav2B<hSlN%pG59iHNp^wNJhG`dHXfx39@z
z-ZD>6StX!Ov8cpY!qMbfl@~no7kCU%%(<(isK1}|wwM~9U+!e6@{OeIgplR@7k9vF
z@_$k~^lM`wVtzKdOD=ZoB=X&7_Y@a4cBqhEF!w_NVL_y@?C}E*6aha9yBR9fqhinL
zvP2gmupvGf4pGr}{Mf(>nA&(6<Oqu74!nJrgW@hjoc@CaTa5}(zafKo&bca;-Rrle
zLV($y(nhPC+qVCUOrQlVaH)7<12zswfWUove5oFLZ2hU3<Ux5@XWwMnaIknqQ`Xwo
z#r+sCErQ6{6AJ!n9C~<K6@Od(+i6m}>i_4|$brDcVL9tB^qJ*EVEG^XD!t)}fsE+m
z9#ckdlVG{U;D{p!IM?L=@SrZ<PjNDkSxR^P@MD*ljP|I#hYeCvpRYi}7SAu1i(9P+
zQ;)3+e(t;R^B!Lgt8;`!!ax7E@(J|H78!CI_BK`Rl@c%d)yL}dt<>#`{V>G;_Vvkz
zEzIyBDwrsyh2t-;>#!&3@|PWbLHDZGK)%HLKP5aUPB4%Suwm@={m&tfrFud_ywP^?
z6m*?JuDE8i61wqZ3Hn`dRIhf)vnrWagUTa5<jr^O{AYBL8CLpZsPs}-5OFc=5adbA
zq{|!$!xG$-uT5JLrJ%7STdL0}oTZvn{QN-a|87F@FNmke{dVx*Vx#}VCkE!n{_a-;
zA3SNfX1JQB_Z^urt-lRCoBr)$4VVr=!-$YL*)tHRo6id)=fAfr^YD2aEi%gi=NO_{
z3zQ(DL2<-H!yI8){|!)&3;Bngta1??U1m^T)Y9i(vYTV-O?*BdbXf78#&Tz1r=pQ3
zcf+*^%_FI|gC9Hzp0_Fig935MZEB=G9P@{bk*j=?lF0;o?%OR&lSaFFH*xp33U^m8
z@c(Ut`fqT!%lp3^Z|=yClK%lk{1+Jw{7XiQ$G&_2zmd^I)PHL<#1i)Mzt!kpROA2G
z#gOR0Bv*RAAiVyBM&$nm7t2ch<2%U{a88QWg7LQp$%s)CYQHMccOp{sImRKqZJ<1c
z|6}X*_&D+SfHdd1q=l0Cuf|9mJ^9wY^Sb4c0dE4g_IL-%#8ec&cx4GmtrPz2#7wj*
z*NCw4FH>+TJ$9t5)aQ>x@8Vc^f0wZE_;j?*l~nY4zcW|8Q(wurbGA`x=b%YeyogZM
zzTkOlXU-FrpRPH-#E37REj<j-G9CAr*5+=d@La@vzTOWsw(PYxo7(WxKDgeSKa3%>
z%U!+ZCG+17ROERGWuM#%?_Q!4o)Bxm+KZyG1bIcVbsRVf-BAI+M^b_*HLDYoB+kbX
zlx4szv2CfA@eisF0r&bMu^Ni4YGT)`-maxSrPmO5`MK#tdBCsOD>r|h8)lQGBwTSQ
zq~Urw&A8;9y-OWtNmAK7GX50kEL>T6C(jjt(he-O3Gx^(5i^a@%icP^_(0)1Xge;1
zIOr3nl5w7|EE-lC&7<1*DVLby7yJURQHJCRmj62kHGEmE$WT9RZ9S34<T)QVwW)*C
zaIq<6Bf^`5-9hWDn}4BaZ6_~i#T2!|wKg+`cZP%#2f@Qx2XXxCezmG*fAu&T)JvW>
z-rg26IcwJfdral@js;gm&?{+cud3}#pN#6`(FzmM>wESrj(X66^WByk(=#)@iJ}wL
zAJCa%?Occl0LZ+!T<q3yH=!QrRO7<n;?hs_1DyQMh?QB4GwvqFjyR#n_jdR+fNj=q
z7CJ<?cvyN}%6Al>pKtJ5!8^3kU`33Ti4B`oMn<NifW6@f5lKA&6}oE_U=K$8BNHP6
zWzIZzaGrUclh4*G%kiS%1^JSyvF(DI8EYq8L`E=wdxatr!#%OzzjttZQ<X?(R(QCw
zR5Z2$Sc<EqM@I7MTE=@tAl*%4(eyH{|IB6QbJ}qe*WFSZR;JH~He=O(%&9=mPJ~>u
zx<*G=y8cY#=#ji%{Nmr>y0yVZAjqG+)ll3X40eVqgG*;fKZ+io-xt73#@OidtK4jX
zxF_JgB6M20^#iLun)z4pS29VQ3~hL^KFa1vqQkr2M`@Ir17cuY5Kt%X$-dZj9hnxx
zhxQy~WlhmrXm}+VL_5XnA(D%yV-(vi*%j4c(joB(pVi`sWg_cxt_N7eQ5LxLrkYVh
z+qI#=XG=57e1Z=Bk^~H(r^xa~45@bhi4zeK5<-1Q++23JzmiC93;?sGjfuL7hdZ}^
zncC{YQF^IN5~fg?nihb?XTb}Zbk8Th1u`%&PpMRpQCFc~W|8N@5}33Qt-Cck+Uy<I
zf{UH>)U|NQt80MP^)ITFwC$LmxoZ3%5uxyE9P-NBkvtmY)ycZIb9f7%Ec-Humj~l%
zvfNyfA|m%{$7&4RGG@7nWl|LFOIl9K@TN4L$Mm7?-<Va*I_PSva0&V#b)w?7ES2Ii
zXESGcrWK^~XUg`=c!%mr5+QI-o4E;Cv$c5QX;~#?O=-fcId`I(eWcj|CtSP?=mdR1
zc@{uv_Hjb7Xt=X(o3YU%P6uGgaB25HwT(->X)(X3yG_eKB%GuxhOwXr>UEw9p*Io_
z4*m`omf|aqg<yZ@612tB%Mq8JodPGdc5;MIdIEhr#yNc&`my;jep+Z6Iyje*oAL`n
z&{a^P;f*=UmV&ys|75{pc$>6ix)g0y0y{m>YVCka&3?V|IJy*-Xbu)nZ#XW9l>W|N
z)at6jW$oiu?O*AMY`J_LX9n9A_a$rO{rD(^(D}?`)uXMJvwHXrk20sswqb?|cvUJ=
z983H;=%weg;ciwn_!i$1%#5`(=OqSqL7nuQ>47eU>x(sJdmn>l6!x%G6Pdo~i^5uX
zE`^DfxjZS-{ds59Ca|hzm;A6^-cM`iM_3N*XOx_2&o5FoZ7c9+C9DvO#=(9(ak&_%
z%M2j10#BYMf4bx^E;amZDD;T3OWqV^%;BvYQ7)Iz({(ychjxo1+OA4(^K#XQ8xpML
zX_9PLWZB!BQ`9PawUE;xQF)WS$AC-qKuWv97R^Q#<8smf0HP)e0btI{<Bqra3-_?M
zsoCxuNxmP>xs01O$Iso2CN<MnbZxy^baX!EaX5d)bl+!C@{lG6fj_Q(QmwLT4Z=vQ
z_%{-yq(CKfqO(6q*Gt4O9H#_qnzJp>@x@?ei_J{mdn}R%rP+1v?iTRaukeR89n(I(
zv@F>=-7hvUcdYbo_(7_^Lhg#fX{JRD>c7UW^<MX+3m=azXO*ZgWZvwPr7}SLwi5^c
z6RX+x<vl@Gr+w4WDSaf_pPbdyjnlUS8O#!rm2MT9_~t$xYm}5axMY?NtsSNwlxAl=
zJ1K%_mY430`s%L7Lkk`QxJwV(0_$&X$PWGC4^%BQqv%~_9d;{UKf{l`=0flB3=yg#
zON`xITWI}_m7uO!b(s9G&WcDTW+sdNQ}i+9Q&JDeJJS2rNXzptzAJ{Qnxm1k1#)To
zu}5Qr`%wj5&wlE}25rPDnB3f&kwC1hO45(aYTK@K=4#Aw81t&8AE%a}_S4TB74vKM
ze;&vJ47DGsNz{~~84(0WQ3>dicfBbSx%+2@b@m_LBMo!^18e_rjuTd=<Sn<N04;!S
z+Le`D?2BT!H3f{dq2zlOU#;kQSkII_pC6^&s9}D{K?X%=b!IE+_}wa!cJBCo4<h@k
z;Vhv!*>I|`Ppb#eUvPWQwPh^B5G&vt=A6CJX^WNiWt+2qQ)t^|vQQrM{<&wZr9RGd
zvjt0G|6+5}j;!dN(pOsUvXnZR7sLOJ2`Rg)BEl(zVNN<}R>P{c%XD)5pJs3Vi(a;6
zePd`h;r`uG?CZ^Yn)&;!NY*)kaA#A+d|NL9bM?cbml<s|X|F4NMv(OP-eaMm0)rY<
z5|KbWoo2)~a(m#{bpxLJg!J5O8J^dD@ci_`ua%pv`&O@6bg$yV;Ni1(Et_EfJV|tv
z)mIF@W~l%*Z_-o|hzrpd7+Q&S{R~@en7YI)o<g~(<f_Pg#7=;D*fm|ze!?{;_sUFS
zfX=*A?v{U1eb<QFw)#pgMB24!aHsC2VzD0BAO<%PAwEY`=|WDZ=OX#wCN*%-X}dPM
z&hL${HUVh(;9mr1Z~n#i&EOR~L!s+c{{^?+of?(#n;V%=txJLHfV&ENKYB+$Vb274
zI4xv;C&_+td%NQgV)Lb4ev^H7L_ty~RaG<6iu6t1(MPCf6im5@lR%UKvDuOFTHqUN
zw5&V%(=HsD=G$fY4^i{kQWaC0;+JSPONLr^>NJzkWlbUPoZD{iYeIVgrisSK^Q(!b
z+3Vx7zH>T*zqpdO?d^7a=jtyj=OuoKfCsl@dsQrKTsrA|w8v9CXIB;UBFP|lGwu6n
zu7x%VYh{?Jig1s%U#DnyPfb=dn8l6`xv|bfq+UDwin>o)XQ$5A%C|Shx*4}n3U6J2
z(_W4b2k*t6QKiEy<@K)z%KgYaZ`^ZF6>ImMHF{oi>P)>!IZV^pTPZl0)JpJ19HWi}
zD}7&kkI#0taXBUIVa$Oi=|{tTe|Z+|#+I?mkaiE3T#B2;wDJgi8PzpNTAs>+&cO)x
z?v01vB}Y(tA!;US3EbUO_i9`3C=0iF_+8tlGy$uxoLbAPUd|5+AxlV}HW#cWi^z+T
z%#AHnd_Rd!tOm0oEiKlL>SdAwbs;)?C#rULHWvf_<n5qoic!wLz|;H0)$_r~%08{o
zz8=MdR`ZNRcF0<){R;SiF4xGff8_j#<b8gsXs&CMzP`oF@fqnoU4x$jX)8l?-{`ex
z>$GiW#j0~iZkVwVs9f6f3e7YVYfvF%hs5GEw<`h(wO93L%Cw!&+ANwTYeGj6x0X7*
z-Kb#3B?Hi15nN3Po+OQd+FHh;aD{1R`<6FyV>;{!?fDm7Cq7_V&o4c|8DpV}MGP?C
z^tKCU)z0Y9m=_e4E4Q7C#|}+x;+^|(D)k)5+>5)9gIeQv58LK9)B$vbtIXn8+3Box
z3|c6QL`7cST3}zGmWpDgSU0Io;)>6jhJMA&4x=zC(7ZaNx-~Dimt0zfBT6mb`L5t_
zZswvp>jsH8bT7Vg)TVlYox!o{g<@UiMn9igx|HRs&64x99ENBN7w>c!dsRCZtsTYG
zJ++i;g?IFR<MR8-TklaqlQJEL;gn0~!1PP*EU3+Y;i0u(5UKtUORDt6SBV^wv+|)6
zSzo?yebZ_weu?BW<DM}a<K7T){Lpy6Uen93!(G(G<stUcCtANM@B1EK=&s*Hi~s#6
z25iFQiy->n@4GR4m?n<(pWBIK+H=dj)xPp|gbzpbQhjws<-xP0`X;ipJm5qONn?tH
z09Ttq-27k?xeS*vM*useS(Hw%W@dzDe-+<oWw@%DdiLUO@==C>Ep*u#+5f=|A!P5X
za#u05d>u}sn@zr#b#&F<Gl%%6W|(-`XZQ`<5gncm-dOhL-Mg3N>@sVJc7|~^%CyUN
z2Z$G7sgW%+b4DEu;yfr<Yi@K+HvDk3^2<E|<zKk#-h4wucViKu(^r?IN7qwT98%UI
z=-mP3IQz`!J4FoWXz0Dn*JEx&FV>gpueCd$bMAe=ez#}#9p+Vw#CtTF$4jzD3IVg#
z|9I`eo=|pl_W>GC&pBh6o;A<#`*-x)0(I>cR~*xe4J_wV3dCHFBCXEUGjd%MTgv!+
zN{wDA^tPGa%i^_K7nO4fPpgJvu^CwKn$wtt0rkqyUc2gDGyB|&riHA{70|5((h^qq
zD_dojS<A`XUm}*B`7#P$Q8~95R9i#+JzPR6pEZw_Ag8m2+qib}m`9%W`nMci?5yQ|
z>qyJALx~I)Q$#!k-JO#$Vm^h(D|Y^XKfU_=SiK*&6TaR5HWue{kIMmiJE<h})+QmS
z(ME*H4dXnKP4y&?sVZA_L@>%}68I7jv%F?Hi-B+Yp*{O$3iWA507%7_x|ml_T#U-L
zJf1V7!-mVYc|Sv`b=i-RLSS%GwMBuEQf~c@0k6mO*fpg(TmDcRibD%0C_6noqt@xa
z>=N;#aS!zs->vNMwCs6IDu#WTVXr)zfs&pk44sYsrHQ}*K`3+Wr1Qm-E>j}fzw7mp
z`j}UX;?Cx%BYBNF;)XRgu{kfl!_(g-d8~DkchHUY$}(6nHYM=bAqullIAg_x&-ThC
zB>A4NdHw#jQfZ%QqUMHTSHy49jCpe)s~Y^c1piekJZ85eVl0P}QnaU#ZD;TitRpAX
zV|-u=B3Wx{wCroXWmjuyk||#EYB1oK@T|&*Dgxbw=tKjG7lue@;o#ay&ssV6qNEEB
z&lQA8wr9zs3&v+x2|3>pTmDS;rQ3X1zN|&8XG;g)_^#W-j%hYmq1n%YMa&|SUopga
zo5TAyd<+j^WTH&g?w6TPM4RuSLq=6XJrCmver(gu5%E;gMJ;YSh?m6#r4MWi^}__Y
z>EZb@_jzXOULw7SE_4|^dipC2vucGVX~M>B=4bXba$a<RZnauw(d{`+P^8F}VhnNB
zit(Dg+?=-FUhIAv9}Z(+x<0aSRK2iMdc+|ip7w0DB>@@4#J}jOU1Ms6O&XY?j||So
zx<oh@i6MCzTzyUIIC`8sb^PGXV{q_LqwclIyY)|5fPOjBVH*l&=~jZWwnC-%YPt%h
z?~3n}Lt%_@Fc!*z`|noUe}1S@{cKrTu+7MR0upgpN>+Ga-uE>$q-B{<07m+KroY$Q
z%}8GLf9m*Wz9Vti#q#b`Tm0gj=e*D+Q<ib|YDxxZg4)E}mI;o^+4{)X0Zo$6a))8p
z*wA2J?AP0BUo(?;>Jq7El*}~CxI3nfcP7~oT<={Y)7W)R#d^O08#{aJbUzzh_K>^}
z(Ezk?SyG8mCWwC`kt;67?%?*Ly(_H+DvIkvUh4U8)mJvdOr1yhv)lmiULh+L$6QrP
zdJ+u<6Kuh$XDKOLCI-UWVs#&w)4g^4#VyU4I&6+~Xk${@P16!u_$uA0>P*OXVU^&O
z8&QTc{yxrXXO9;D3{Z#AaqWb&<@9q5y?E5Nlpfdel7WScXZ02i*+mkSf;00@$#*&Z
zpi6F&5*DD*_oKEeioyemN*6(!tbHb>(zxA`Cgwei6sodUJ1^R=%cW8*`P=8IhO$F+
zz0GLyC<WDrI>u7?mPy$2%Xag-%6<EIpj}+;npWsk%qP|8o6XA@OFn<hZnalfq3o`|
z>RvUJ_??bn2UMd6i9(bnYSzc<x}u{!mp-_RgoQ@s10X?P>Ue!$2p{6_vgy$r-VRPj
zvykz=9u`MT4syAVoa2{I?!RMN8Epg}&!lJ{HK@#zNgyDt<orgiLq{}w>_!e(;|CM(
z4`M~t^$}15QgP{@KIa2;-I^meW2BRgvVLWCS^~4xJR@pEuK$e9Rto<zjauDjYRz{2
zPPQWEU|uei=fF5||NHUvVawpD5=6q;)4@C=9Tib-+2}NxP(<4gx?)q+@h5GfvUI>%
zDB_Bf_Z^mjo5=YUjhpy#HaPB`FIcF}uq!Uptegk@>g>qq9zC}_Zo>^(ZtEN?1A(B}
zlvbLd;MU>cXoXL0^6Sddl-Z-yX8EzHHESFjaoD2!_j$o1n@h@?*6LSI-qS#Rxk88+
z{Yf$-%VVAHJz-0x*S!!m)lNXJSoMdAy>jH-e?Ua;=Uj}%tv<EmyOB~?<BF;eur4Ug
z>CfXQg28n@t5$wxiX&hu_U^D%@LoQ9hH-~w^9fA80b#}a<J5o~@A0_Shw~;RLCW*B
z>1W5rP6h>EIfG96lrx6VG+j~JoiO$Fwts5x3wM<d`5;@>%E1*@TN2GxzT*pV41eS`
z4NwuQMuS~fAjQAy`z-$x`gS!BY2)blPb13QV_^ItImR&=0|ZORLtfw(uJS_d8^NK&
z7^hahSy^Q*bAwa!?_c-0PKs`(ZZ~7|OD{XgyZb7zQ_Kg_2w)b(`~~5k{#*e_2Hh%U
zvneh3r9<dS{epCk)6lnQw1O*>>DD?O!W+C))6{h|?Rm>LcgG1dx{>JRchkh$X@LZT
zB37_;W1A?U=W2<qF3~3F)O}8*>U?=m;|_s3uX)mgLyu57d+Qo~Or=MO6E^qfJ*NL-
z>xl6R(rwXq{CO^dSLb#+M%?D)qp|jAG>%3$#;I(+*?XLQaQ5cqF6-6t<+PBwoAH9o
zF9l%yd#|?{F{>`Fu3>`<)V@1Rt(jd#kc@U0p9`n*3ZGVlqs92OJdX~JH>pS+vH)9q
zHS+#7FNBgefq%b#KZhaO@~ve=*;Q>ft2kwwk@VkrY6h7d+ew1v+_B%_-vKQum&;gm
ziE6fJrph@Qp6iQYjK)dTH*z#EA!b`|RX$X?+ot#9o3t<Vwmj<zRjb~r<w1FvhrD~o
z2`4z1gu5QQd5(xeNeuK?FR)w0+g5Sur!VarZ7x_e*p%&@DR|B7A_zTT{DoKRbI8Y4
zhU7mkZ5W!XsBRQ)3ch-P4BrdbY6|%RcQ)oWJx+bEe;%ti-TRm#RIgM`ztvu%r563|
zWXs^aD)x*-o(!L9CuS*iXlt8H1qZlcaLN)-4M7unNj(%X)Cqp{bApH8mD#6%EYi$!
z&utN4k?0yJNZ`iR{Fwso*vr}f6Cbp%m+^Pd5`J7#y!VJ;R5YeeOtR$GG5)>~tN(G2
zbV`A3)zs~Dgjka@{|N`7LQ8$2ca^YP(O7AAL;R?04fn@C(dAYaTMyj+EY0LIA7b4}
zg2M~p6<=7yd3vj&917b^IGbE1aA-1G8#QUgQ3;4u*2A)}$F+-IaXwSFeFaJZ7TMC8
zZg@7uGLflt4Re`8h(kk|%~I1Yhp$=I6=Hlg^k9dvs;*O19#)Y?rGzQdGfr5Y!0Pqv
zn4Eu@?wao<1j2le=CTg(${q&TQo=H(eC~DxH9Q9@5*NN)3h&(K;xPEcr#x0(38N19
zPR4oHQG-lOz$Lv|R?04|jb|-+^^3&7Oh|?f*vS%TQd`%aLOx<E%dzuU?qr#uAS4Hm
zqgpoNwrg0f0C8R|y#asAW4l4;o)v_S41DT2x^a)S4C$niIED6NGKFT`YtDJWenX2T
z5qwZfr?9Dublc94DXIX`H+S!@z8o>D!olq>f&DpFa;*%wGER@tI-%RmH+Mv}89Ud_
zV|T@S#Z%1Mp3h>G60m9PmeZ>`Y}oGTcVYXD{(A{2GGhLp>!PE+BqqV`{G4#?vTIp)
z;!Ho3b$l*j?vqb*=NhZm53VRW>yPRt-0Ur_T7)#$AN~kp`@1Q~mS%gx?jq8gSj72y
zi0N2LFpld&l7KhDPWx2h8T2Ba=bz@wPp;Tgg{-UIs}re0#bco+GuRapc?ozXZAjdY
z3-->&J5I9ZzqumgHzptVD;YeDH_N4P4VAt3({9Q*LsL&nyA@e2!xxO@$KGKh4hVMH
zlxWhFzzS<JNzo+^CTkg&JUqH827iXa9zFfco0?0@*0oK?iJMIycE>`SqjX5t+PrC+
zchy=VTW(y?R!c;=Zxtw5H+e=YnN_uP2mjG^SGgQ=&)uXWt*jhRUJ4Dm<=37=b|?22
ze~G$lD4|lRbC4R&<Z6n3jNpuIXhW8ZpD6i`e*};c&`rHj-_P<YO3p}y9o9Z`pWaW=
zVU?filk+F-_G^WX;_#=y22)YsoDey+cu#xaQ0zg8_CWGNkn5$Kd#tj(h2TR(+B}kY
zXT&W_(~7{)TiWi>Sp(I&N@;wDhJtbV2;uqE-9g9kmHg9ZyJNsqY=1bOtcqhSB`~wA
zV{4Ih?h+2F-{pi_az!!y@!Uha)Kd}_`FeLoH8XQRr1MBHcNuPgQ{5p=@u{~iod5h7
zao5YRa3<>bv*=G{_r(hQ=}_(Q{m1aXgTS3F%e0&9q-DHOkF8E2u*8XN-E((XbUn|6
z!*=;KnUyqd{-QKG-H;sINBlF{%GPY-z?)t_#;Y+Mwkfptsh&`YDTBVvBFu3X#XYYv
za@=e;*T1K2P;wiFJ7Y%r+^8s=Kab;Ut-meR^*r-*Bv_ScXq7PV&uop0ixsDujV~q)
z-A_?`TEa(#A5ZCPu8x4{)dq*mw*y)d<bew~AHs9o(0B=g^9TyU0;V=quA|g)E!0Vp
zm|9*Bl&G8L>3z-{)g(|nX`Q<p6EZ*Xg$b?;?!zI6_kNvaS4*)CkJcv*^ZM}Kj@CqA
z>j%9qF-}KsPouVuQwLxchpolR0}1z+jK^dqozgRth2|*L?)N=>+ioedKFc>%;+WV{
zJ#|+nlv)uO{`QyRy6G6>qM|LYS}VA1cCvgp>u`WxOHeB<eQx^;<Gru?T6>}1>pZ6*
zZoW4FYoB0}p7rP4^S?QC37mx}`~2?5wvc{o$3ZjBgL<f+$e>;PjkDM|XFCNa$N0LF
zEa-+UDx|5S!bi>ZY#9H7VnAVhIVn5C<Len(|1mEIDmz=^AC}&1eqmQE0Xt3hwxKv{
zTn=&8TF~=k)@O)pAMUiXpSvZLCru)}W-w#M2aoK*8>3am*<7X+*eTHVD%r)?ysm+r
zVg^12(`dh%^!_1=Rbz~QWT?cZe9i1kN@x2fyS||QxVDb2FoPR1Xc4%fI3zTlC4Dj0
zheq<DyhFXn*C|<id${NRDXyWeTcfLc9cMz?xNhXK{OiFDTf?HUuS;CCm!!DTKX&7r
z3bJq&80pDQzl6D3!pd0he|D~j4!re{=g!Y^sq^8C)@cmnt`T0T<F$M%*9@i0;q1q`
zj27lQEwyK}!|R&LTZ{}pVZs6)g0WkF?95E`ic#5oD?PzEkz6o$2710wN8}p++P$m4
z$@nE>^Mt@f%7z>kIQXhKA-Lp#bu;wAeTZkT&%$XjvUIAtgl#^#5^Y^zwP`A!a*3oP
zI>>r3S3T;bNb-C3DE33s!c3&9WhmMD$A0y7rhigX(DwB{*%>b}uz$xK!Jl@0Z9pwb
zBS+#jCM!cJjs0}XfsC`_c>RY%?;HN<3izT!Czp9p+-L2YtqDt23)&vm&;k;C;*hic
z-+-1V9j<!kP4lPEc=AR9arNEiZ}z++i{ls3son-+YZhPu?!w^|Ysi3(X18XLy%E_R
zR~yI4?P#P?-Co1`By+Ou#EB?qbnc4x(o#5|=R0a08q>68>Nn2c*9eHi!WSmYw#wL1
z-bF7^WeY-M_9qZvbfDsl<PW9F+#OscnWp3ss+IlBm`1bRN9U}wmq|2K=M3olE~v8i
zb{;ebxF84|4*tjEBZfAuuzds^cZ;!{A$2-RNLb5Y&!(Cj5Cv8}X}ykEBD~+E*u$)N
zQ}D*`4d|NJca<^44b=VJTqn>uD)XO4yGR41c8jjar=~OgV{9$(VT1hM5UaPgS3FVh
zY1*km#gz?AJ=1Rw9@HYVxjWM+R}WCzaRRk<EY!`GP@Kc8w=!I~&GX&A)g}Z~BRw)S
z`p2kUq?K;s1*2%dnA0+N4tfJyJ!5Z8z0aGLeAhrIAO7iul20ZsOJhgDvp$N5@ndJI
zxThzv{)RuxL9MEakI6fju0HVz%ch!fnWWH(sw2M28R^UW@#SrVhPMLUJMXc9ATeoB
zu1_(RJzbmzKzs614Bs!zw*Yh-o?N@_Njm1WNgb0!aF;myx8po*>00c=C=XlD7U?7B
z+GYCM;chyB)g3qN!YDL!DE`_k#KyeDzR*G*o^>EWS!f63xWG@na8Vs}3cvR79GY|%
zp=hQOd=b`dyK|0;@%TeXK)MFdu6uPagcl-j?I0NJN*X0YuBxGJCgu8YefAXN&!K9{
zXI$|OblR!89&E<7Uu4|7FxizD-*ucRY131up^MsOx#qJJjAs}@t#w{g0&YHIEeBsp
z+oaqNg{B98^K15*W-xe!_U{XjB<maZ2oow$TMQD#XcZwYoV@D_$vwA`(k8?VtdVjV
z=EP%}fv-d*AtrI`$-<T)&+SpP`WIDMmz`5$VVs$R^DcY~p_nxevxhfLD|TW(7&g+x
zA=S+$G2<U3x4~Uy<*E2p+~AQkK~taauH<)LZ{m&X9A)E8=CYsvfQBm-_ug8=1Ty6&
zUegy7jCr|^PBI4@B!4Ze6yBXuN;PcO<Hiz6PwQsws7o(&>8+*(q&6P8^tK<ru}B$7
z(s5hCM-g^Yaq0kRxse@?=1NQ`(H_|(_l89Qommf&aFr8;EM>awBta37Z7%k#_cTm3
zbY@m%uXwp-CuG=V3L7fIgW;IEJudcXcc$*~RsEUt@+i5TY~xY?32qAR-bsWlKRI&K
zR<Z-+_07-MQ6=+~VjdyyE^y6EE9C>t)UuJF)x#ak9jak2=^%ECGMKk*|0EPW|7fWE
z%x0!oKO9`{!Szr(sZMzoC4wK32=;ZL|Md+ZXl(9hX$YK-VW7+rbD4`IE*q*bW$xn7
zTTUhz56SZXQq%E$y-*zeIWc|nE?sejgJ^Q&qsdG#HZ+>BwPLyLgkXMB7w;7(EkQO5
zmFvN?m$LKNA{-D>5Fg0|lHfJVeD2Qxft3(^{tymUT~7~+?-r@2p>*_F)e6VI=Xe7h
z!y|@M)1DSX8hCgi$x^H>&ADt2N{T_I5v;7NYw%>L^Y&e>kB3<4u~Rv(wdNg~Qr-2M
z#$WE8EG4Ez$vnh39^px@fZPzbWzW%9ZxK?%jUGG=?!gQnoC<yRqL$kghc6O)pFwZP
z#-6Z0o?x`yX4$OQbSODvViaAxnFtu(!hJ@+()CL87gr{J%;7dmpf?2ORs43Vb}ZU`
z40HIO^^fK>$7KK9@;yCe4JCE4RlL}dJ)SwaS|hBnx3QWO<wWP*7c{!zNIVnB=D_r9
z<e9-Z+`4JTBm*&U@WT?a9!E@LC7a)$!RXuD_MS;Lw=WWEki-NjoH!3q2zh+d|5@AJ
z=!*mA*>`4nZ|bwb4UMRE$F8=XC0nSK>oB@!)Gfn$SZuYlRPLa2`2EiukYGcWEnsB}
z(YLfkDI7AwIX=B6T#RwSInt(1=8Bcg5(%9x<94hgcOuLnp-p8+vTAp#Hd34f2+fx+
zOT3+5rR*$tbG0)p!+}hEUhX1=mFe(uSZ<k=KQzYo)@=|n5vb=uP<`cj`UlVP%4Jf8
zFLH*BHd90Peh7<7)tyw*jnrtc95E~}T-2RKpk)UD@Z^F%rlDBJQ@m>izeeFZKGw8s
zZ5e`TN`ChwJIU;>uXD3kdGI<`*MnxKp_uyQe1>+f?JLT=Jv-|N*A8fda-(X8F2?rd
zAHKv{^P^||Y~MfnLqEubE=3hupvGUP<zfY&7_5;@4J{>IvxP-h-7`IZYh^o;)!1=q
zXnZXMp2nT*c|rrmlU_!%JoOsog$u>x*qk-{20V20m3iLKU_uLX=I0e&47q`;jH^Ku
zp)jNEp}Q{r9%D!zG+W>9ddXIo)P6`yMTY+)TrIIu)kZjsI=iG>GPK66Y~?;4=l$FC
z(G{ts65Dr(HZLSk{&8af66TUCwO*s9y1sNG1dg;kj=<d*Rx+UMSJ2j~&oyioT~g*j
zsp6KT^;XxCiDsm|+Ja!+!dc6}=FnYSTgH0xn4q>9l2F-=4c4h(m}SPg*>yW6LkpK^
zf{BjjiEQiQoJ)|{=Zx|Na4MI{QjcM)q*%Y+v-#3rG3YxB<~7^QQ(#CN*hXNR3vb;f
z<K^8q3<iH=3VLA#vG5lX3MZ)Ixo8alznAY+@n%w^nwN|QERbBdko=p!o<TSCJ<j*X
zQhX>YZo<e}9T<>JL8RJQ&9?1s(cheP0D&hNYo51+v?-PEa~adT4#D|5hv2Iq)%|F@
zyUFWgQLp8%S-EEXN^W@X*a61P_v7^SKjMN1uRvvD!tqpqeWw0nPIkVY-8u#ja=&q5
zy*;i@he1L-po{%D>7Qw<_&?i?3nbq;NXN~!dkR%=Ryv<ATpf-$ZXUJ^UNOo3(VCMr
z+6HZd^z2Ws-qNei>GcCfNy6Na3t@{i@hg&6p^ZSjy5S*e0enH-kB0lv<FfuuGjD_f
z6QpGFC@8butxQ_@^P7TlZa|qjU-#aMn2nW0#Q~wuU4Bp7QQIf4j-ZlHPKVp45CR=B
z4Hp6x`(pa}kHBeX!#94-v}TDKnPWpr-oMb7H>K~IKz%tO+CNtS0!JZ2AdpDA`^L=v
zI-A~2{4ayGT2<=1YZI?EniDppJAVjwuE*Uiq9js=%+!D5vzT{X)V+oq{L?{_J1d<(
z>?`F^<=HHPZg(%RXr9rdQ}?cdb&R{E(fs87YgvB#D@L2rqj;Uw(BFn_+-#m|M1IcH
z8W?N~MqA2uzJplwK<L-9wl4P)csSoTx$jd;@u~KPn!l$uMA}|jgw~gIER8u*(x3WW
zdt`pWvAx^52*C}*0-QxJs!i}E&s0e;^!6TU&m#rs$_{@dl`$kXTJ{&uSFx>HDJOR)
z7v47Cx}C1xBOx9c5N<MkbkSGYV1?b{Bd`+&CfAbG$;kWIUCMf->hmMAi>ejtWEkUM
zOt(<;nE0EJ+mWhV?BW#fD5^f@WS*34W+K{f@q*W{LYL#r#wXs#Fvrrxih<^9NQAO$
zOPeL<1^N8<DbgSiD|@xqGH)-&y<s@mnVy6sqOotN_#{Gni_O%5M#OZgdvlOtxtNZR
zEB6*Fo&uzP^^6M-K#LYDl*6p25Zy-upYed12sB4#$df6JRbJljQD3=eOgg8pvL^PV
z4q_Kg9Ft(qGUeL3>aF7VBBtv&`?GTqsi};R8QZtw<;q{m$07;aG??x?&+I6V?tg8+
zBr8!|<|oPTR6~*kh>r1@B$VrZNrK|y5ou`(FSGDZM58Q?4EaA)8Z8ppX*w!?jYOl9
zv=Ogu5)N}@jK(oV4)hE#K-U^nI>vRtDnLW-_nhJF?tZ0OO49t35&1X5gW=Mq0oL&&
zu0b|c`W8kQyJF7hBd<z!)@;@M__lYmTl%4pJ8s1e$u$rBcVkmL{G7#T3k~}}*P0;7
zvc8fs^9bR*VhYOho3Gd|k+He`&aU4iWLEIGMw*RVJAJY?0ky`ux#UeEG#cin1r%3B
zbjk->tKZBi<0PObq!au&pkEFOIr!oW0k@7i8c(n?FI=$EalRflqXL;^$8#BO?zzQk
z?8tn+kE_L+v4LG$fp?6lo-Jc#%*OQHHW;ag6AJQZoT3};!;`UtJ5~0kU;ui+9hHND
z%rc_`-3Zbu!jk4kri7gvNRPC%ypG6zqS6NJ@v4K$uEofp|1+=Z53d*>p=T%YML!p>
zs%xNqrhago&C={>5NLA!!t5>o!R}QRrVGo^oRot{I8n@CGK2H0P63NPbnSQ8i*Hb?
z6AD5J)IR|ef5>gx*F`*$wG0>ulXZ6N(;SToD2DlEEIMIH?bkB@Mt;-x)d&+d6!p4f
zY5#r_k=KULzJcJyvUX+7ii$ql%+fQ6?W$-ZW6|z%_;A$p)bK3d{g8BB`uENB1Hv+c
zs*STC@G`2^>jKQNd{D5)(GL_QChGSGv_zcY5L2f<#$;|=*z<QRyf~>7%$O`^#})3t
zw`DFCie$C!8SZLqAASgw2@%{pV!c7%$4AA2oekDRUXI~C-*74D3}wZ7*ewHr8LLMn
zHTp*N$ZE_nDj(H0?~^I3cH%6Tiwz|LFEU-OXFkUsJ_Fn`=P$TVO}CwC#_0W$AEzTS
z=)iSGQ<A#RmHY*K7Z)axi;%HeVd&fn4)i6P(L@?1er<0ryAk!^Es|4Lr<1OFkQpN5
z0+A-L-sy*x2b?m|3K{E0&BOIM#oh{8uYF>ijAO<bpo-i;4CMq`d}*T|8!-42?hq>5
zd@esy(IeACiG^G5;hFM6?2S0VItGp~T4CoA^U%Jx0)v`_J|;hHC*(L#gD=$@Y4CQ*
zd;O4ih^Xgt8b*idoTlesQ49LTaF=k7xVezz19HxpoFRVR34+<pZta`Th1V3bK^&Jo
z9th8w2B!F{?K$0!agI0w7K~<`AbLd=%qG(&Lj*O_nHb9;L#T`Lkq-Si6S)V+ZZ?GJ
zEn_6;bw?5dTF(_rYD$W+sWBeWeP~2>Im><gHzl;5&18b68y%ighoI5WStyY5#xF=>
zi222fr#}MTz5DMM38ue~yrjpZ#pVC?L{j3_D>StK=`!{H9d4tGK>L~0VZQj+eB`#K
z2>B1X!ZU9*w158hr}9v1DAtJZHlr8yA~1JIVDb7_$j+l0p1mckLm}HasKw{O67r|n
zy{z;8K}Or6necyp`M*jOnlT{p^l7_QcOqW-bJk9`kU-<O*(zLC<!);%XLFUE!vCp$
z|Ee^hlmxjQVUUvv{fwtW#uC&?GP<GN%0X+VV0^A*m_40?3yZA~|9^e?I|kGQU>Gy=
z6T$y_%io66Mb&mRIHK+UewPz9swFdjf8qJxn+;Xp-@&K_gKo)y@c;F$|9=V=^4lAU
zAciT~J^!D20%}L4-9Agr>cK-jY2xh2_h=vTNmEks9swx1P^T$6`1(rn9ns(RdY<v=
zYnf()4LtB$%>#WyC6xcOgtT6Top!S>YlmqODf_D@PdmS6y!F+q6ui{oo%|!!Ia!Ew
znZ<25zoKgB7%=Xnp19(6y>eYwmYE4{w`o~hs4POxa$6zV7!m2V&Ef8+kAdm#S3=Op
zFLg6TRIDmQS-#(zzu;;;Y%0J50(nNbtvs)Ftk&AuOTSQocQ0{xPa~47hFO8*8h>uq
zl<`)<5~H0rybUf#(hbMMz@fMPzdzyuD5}<dq){0I{VUf}wQSO=MemPmDIl=4-3FQ9
zd89UL^Tk`&Gj^nX!)_?qY=t;EzhNdZ8;R>FGCSQ9ZYQJng-No85^*~|eoE9qx6885
z6#V;qhA@c<0q4x>bLB6yN5;<oXhHT_0#$T?T4ijq4qtIFGQCJO%JBS7J4<W0snl16
zJ3uQP_4*A)L!q-jk`25+r>k)1n@koMsNqv%a%8@qxFkTUr9QVy(bweG4_Xm^<ZFEN
zo_ix>cDqhK$y*%l_WkFAuVeQxP2GxAlUlJKEa-)JpMJAGQ3%H<DMaVwfcsg?z8GKB
zP7(RiI_troWvO&-DQzqCIPed@ioKvfpQV$h#Jg!nn*9XsXop#Dd(ald>UHmh3B8mD
z&29x3zk=&SDbQE^cgn5nT*Km>i6I7_x81#LoZZ~@&{l@xK%8$#3)jBX6X~D6>CAKa
zr+!<@alzKKuQiPH0-=R*n&Z*5x)?1i&+Fx<OukJVG5EjxIv?u$_+rOnvV^<MiLUYF
z&SjNoUKEI~i%IhWTNrJoI$!Grr3+e&?Y-4RSX5fA@|m^eB-uVf=^W)f@7<JEfC$Xw
z09TL&YmMFp7gb0<-x{IOZmSr9l@~>UtsTqx8KU|#&wXV1NYP{JZ!?u34YT`b&?7KK
zRX+xWlfmT_rI2GtQjp`orZ@*$mRIHMjC1P|G!IgfJ+fD7usZI?9(SW+z6xsn>c)a6
zew(!7Y2|<wzs(Ujf+J6X7jxdXy1<sy&!E$iZEz>^WW8^_Z5c@p*QH$lo<Q+~>W|Wa
zrLmO@#p3fQyaONa)HVL&X|lKaghMXk7q%~+@=|{(O9bTvc26rk*4nXON6S`(nB*!^
zOx8U}Ubm#k-Jn8&Y^DGo_52>Xbe^>fRKR6W@%E_%xFA$Z<${ELAzbZvF4_}#9}!>*
zBxyDeJzb8H%>lkAWqd9Br=)1qSTYoHj!2e}v`;f|0Y#HGa%6g*yS1*>BgNV+y`A?K
zvCo*%j9d;vREoW$s{~fiQ?6+H2c^3=revFeUJYNw>mhas2SlKeCMqY;7`9F+1MTzf
zBROu*%h{?Qb|8PP^>(WuJ0UM7(=MrJ;+ok>@y$j>+r!%lz3CrQ{K$_3;B(?2KwB1g
zTaA951wRC4tfi6pWWc`%1CFRs^xs^>f<;Nb1j^5O_Bn}ryx5r`4E5I{*OCa;DMuA1
z${*E*W1Gp%eIL`=*8&JDK1r{hC{A>?GC0JjYsOwUd?@sWfijPRb?W*>Wmgb3O<QQD
z&7};s87e;v?^s@7OQJq<d}wK0Ghe7>f^CU$U0XuYd<AhT-b;$q&vgE$3jiKQO?`Zu
z$VDoxRud|3v?lsmpr~$ykCd&K?9K&O3BgouTb4M2!%QJA;8VNX$<nbHj&rf!-snGb
z^fR0DM3T<~e>}7sZ*1s|Ul@_W!1mv2?ED;US;;nMy;*#{5mPrR<Ch^6sA}82xVfiK
z8p-WAT$|O3hmI?vV7<WJpva1Ye~yR<AW#2?SI=+o#WlHL2`(jTNm(7CP0Sru!<Msw
zX6v;*Aq#^#pduUR%qs{@?DBAU_VfNCnK=e<vq-anYNd)4x4s0F@LblR?S94_q|BQF
z(AcIR3xBIGo`(x|XEOC6tD^Ux!+RVfD|3AGgZ}xYB-<>fDD?Q0_YP3?4{2W+L8T(3
zN2k4vXrR^M-UYhZsIT0b-RI*y8|?p}&TD-ylPGlO+~7K#0NG~%e0;!~ab15%Z^<c>
z^&e+uFPS4)HCj_Am?fMRCwWs;2h9=sVfvdYbLJcG1-0v9@%zz&_14F{4i%^+RCc}f
zrP*l-)i`rIiN_h4j9NF6e)fLIQT<)XlSY8Ph@PI`ovGE7@|r{^(QO^Z?bRR0+1VrG
z!oIREM0yosp+4yXd_OeMZtB>TH5FQwp&hPhQ&uW6Ymg_|urf@l$EcS3GI6J&=wBbs
zi7%}E+cEStOEHPm%1L;M%G&**7{7xIk{51gWlMLYnz>|1EIu3|HFQolkG$0(1i<In
zt?s4j`fWk|NkYQd6L8!i{~G=8t4EMtBpLkfwwB*?HTiR^rddw=zJJ-cp6bygr|YCw
zgA?p57xKbid0p%_{1CErm48WqKI9O}bxFSQ_JE|w_+8qcBlZAFF4r#8X)%(_3D+&b
z3Es0?w5PTNVtjh){^YI`ey-<te6!DhWqkl;VACx??kEh~WLYva26UiKwx3<q(XqS!
zb&lPFC|kIsWv#}fE>Yk2tibAdmVU)zr?S6quPfyF!nc`OzV{mH+mbSxp+Z85voD8q
zaVe<IFHBOF8m&{j_Dm<B=T0`WXwaHSb`0NLJE|a$X8Vs~z8ih45Tct=wG+UZCTpCy
z+||!3Hsx7h`$u1DP&VfS4t{_=)!nsAvER*~8V?IqliiG8^A6qtwLp(TdG##+*Ve64
zViF8<D`A;`01v6@Z^)rUAvd=$#A4Z1GXj!Eq&%N9>I`bJnRB%gI>?Fd3``WPo=?XG
zm##^SKXrY_Or+GRtH(ik%=W>0dD_svW2NY~g6cMd(pO-nMt+e2PnTX4mxVYc*mT&L
zwG6sT5THHDm&S2HP<GFqFmvI*REWG;c&xB9!wB%S6~+%1pSgI{Y$&5|Svyj%oIuaX
zzVR#ES;s5lI*4%FV2AWGDXYW9ybJGP_TGdnU8;Dq>&Fy%pR~Qd1XX&GER^LEN|5Ze
zeqEEc$;K!j7a95a1)JtpIUL5wye=T>dFmY_A-=(D0Y}B3EILP~`}Rw+r|Ha{snzB9
zku>!$m<PkCN*#d@5k&S^ld2HreT4nQm7tBbY%YOKgr0ZT+W9@D@$AoVL#pn;|A)P|
zifSwT)_zMV(&Er!L0X`t6bWuYN-3^|Vuc{ZU4u(06pA~+-QC^Y-2()d;O=nJ{q23T
z|7V=LbNA&YH*2hswbsmf=X{>uobQWi<is{wmsghErY`%q)egOb2%=W6`i8^X(frQp
z)Zwuj&48OQ1(y522{+9wN2cvt&}knXh9wfaTsmtsZ@u4bchi#}H{}DK1>a7h$G@Fe
zmNcJuBO$QnnHzeDcPacgiw*Jf!6r)ZfY;L?`>-2#sc9D`?`lK+YeuIA-OWMl^mNd6
zxFa!yfEb(SHVO3b@O#naDB}T5z8PN%K3F?-{uoS~;Qq=@i!as+kde64Nyz*|2q1Ac
z)U7%9QCK<@;DmO%Dvf&O$4e08B0{b1mB>M*P+0VhpJg96e&uGxqhjb!K(qT1RDTx<
z`k)#~by;XXk<9_ij;3v*r_->?iZ?2vr6;Y%1{@T-uhGjIO<I`M(N|;pF-%GaUb}A_
zl5%W1F5Mb!LglfU!34tk^wD;X&|mKm({&z!CZ=zCd3Kj!XCLSASXx@TvLmFo*b&ZO
z(r>;wx%pMSMZaLU#LlRi&=D2tjv<39fY0L&U)o?PJ5jcL-lfx1p@MhZON@Z~{OJ}u
zPW@I1j=qsJ;u$xX_cH3e?p^vkm?Sew^S~QapXp-Xl|!T9x-!-(mNzQrw(6@bX#B<E
zPTGb1GD%eUr1?GgqAjVN^j+&UO?JtSK1bHy7NuF~gEx)^%vzVm8~yu!b+^H=k)-1b
z@%aN~s+R7ylv}Q2|4R#w0bF*#)qch3HFj;CbAL_I3^FZlsBzV@XyJguY8((sM%bi?
z6o-aS2?D+A=cZSP6cc<GTFhYH7Y0(hfAvkXvA&Oi%v<v$)YSG6{ag(>4RueXT;$g!
z<gM-0ROtBI_q%(G)|{_1=|?A=N}CmCM-q*iyjIdo)hX2*-}WV5;-od*K(r5jX6m`G
z)UWG)m&2L(;By|Fpb^2I#<7jy<M>@>?oNKmlBhWZgO@%#!wQLuDs$Q46s$M0!v{{I
z*D&BPt<hM1kTV27k6Ws}wDRaA8F_WuZS90Mf)dDgt+p@+HA!u@;2OgfNEMz;dx+e0
z!~O}bQzq{#Ok~#9skrwdI|;%=d$^V(3USXF(#W!|Iqh*cBgcu5)f3&bGYp;BzutwL
zGBM5+_H%Irhe}SQ82R&f`w*<5Ip6rLg74bnaYoJzGYf;Q)qx{PDDpBA#tA0<uETP7
zGuD+2+LRM7H$|`|M_-rfpWoxefQAd!bKyA0ANK237w=a>K5NQJ63wFZT_xmu<7Qnu
zh=HB#4~Eh4nje7rl-(KxKtJjq4K$hGa7kIN*uFk?1eZM9%kl`6ueO}z$_zTd3-g&f
zs|heLMm)EnU~|U1@*gssI*}Y1fe8~ulbE*Jd0KuORrAj!TEDX_Ray83I1Si%QyR6s
zm_`CjvK({3)iq8fTf9PeT8|BM&Q{<>o}1OX>oZZjt9?fuX@i9Y@Y{9=U|A-$^Nc0i
z-qGOr4MBJweOgFiTyU$6sx!CfC$S;VN(Gi7g~mUCvD1)0)n#fM9=C^vK_MvWohkM^
z&Z0m_Ynt^l!L9>+x6$Tsr+t><l3(qE3keMJZqN3#aqc5RI~l2rW`5*9l)bg)f7ZKB
zpzGHV3&Yuw6>S;0{&JE=v9>wX4SJsZP5Ul7${HKtUaKo{`;eM}qqNwFJ?QU2)8)M$
zN-D5sgKLjLq!+aui)IWfd@`8B+riduGE3G5gGR^3*GUfg)P`iszi4MosyMFcxp#eu
zja<0yJChVc7U{Gxl@F>cim{JO3E$EGNcg)br7wLuG;yah;VgR>z=03Wbl+xDF9khl
zKeKELNS;^q>NHv&`{s`V050zcUP#*l(HA8AoT?(+Q>=GvuuyWw1v@`u2`4(4;xGxT
zUbQ??jhUs&_uHNsZN>B9+$zI?qV$CxKf&o|X>51H5O#uI%9}s;Sn`ZrxUP`#-9Cwg
z{4q*}MS>MScT-(JDT!`YI3WBc7=3`jwb*w^ObM+%Jz|E=UlTtp`NJjz`~J2qWVwNO
zvYxW5_*_nPfz3-*0o?(QC$<#Jm?g>zd`#lKm_Af8@4@?($hqPqN5<#KF&Ptes}X>V
zEzW|onNnHTX&<H`4i8vYnNbr^W=ox)h3qGvYwgnAqfq!u1+{7?fJlND>Bz*NO`o3H
zQ|d26irNSFcgrupe2Cy*E!poCBylHYMz2`qSz2Q7I`Admb-ksu(sLe-^$zu1N)Jt=
zSWd$@CTgvB=B{Jkj_OAQsG&!<a+J=>mEQPg-QWI68!mKXD<U!+#~q<HlOPeebYH=K
z^=p#;q9Ngg%1WI?i^Cl6sVr}O3GNMS@kj~`!}j59_`SnvlDDEumVMN~4R<T{4?4K#
z?D!29zwNN19yCk+)W+j8l7yGN@Kk3|FZZ`1r8{=|ZB(JSKL^aHbU70%c$@bugtfPd
zhgz<+R*Ry!|44KxO1h`G>srV;rpCLdXyY3`snS0~Ub0-D7e5L-@5ti`WSuMh@laJ6
zE+Ct<s>kzVorEt(vC<EdBZJqEHjIo9OP2xl!G!h)$V(a(zybT)R_Ax3Y&P?15H{k!
zQ&yUMVA+5cJJ#S?^%XUwRhk9MVKZhMJZgT>kexl;Op$+c=dD#|vssra1c^DUz+9<I
zG(4K9&J5xZ6=(s)9<2t;Bt--W4C`~MXOV_|b>d?;ru&P>w;^7MJ6LBvyGOh==O@TP
zqj3%k5~$h^P!4z=-3N8nlqOkmx+<D16ZNNE#KE0s$|V(*P&2Tm-?uxvJwRM9tlILq
zYS8n<u6S^^#_?!Oc;kNB{S@sFX5*N1T;}YbRPmD!=of<R;d98sxQ^53w`MX&!KOp_
z!CP-~pxUKd0Yt+t&aM4{BBaZcTzAFoQFLS!4)JGv-6mb^T_d-f+~E2uj^5$o6&w49
zJ~rQr_H3uf3#C5j_tQ~UNpn1!;Ux6QOWT+{_9Olp&%&za7u*uVrkTy}k4tKpo$kJ)
zw5UA)qjcb<MJ1rkt*L4*<Sbx$LS;<sQTj+%mR<VeAlBTtOAE*`P#4p?ys#_w<F5U&
zydqLUL|5{kfcUOEboH~Wqv4?Or?H_*NvZgEZ+HfV0nD6!KPP_%C{xhN8EqWz4jFXK
z`Qc@MD~LLvIx*;4VMac&-dAKUOb*FoA;_ZKqfoVNqR%|ouq^*HG>i8aaQSe;QVz$J
zmdPL7WJ85apjVCqJ7d-lnFFf&-h0j)ZfAxsrkjb6vh*6=U(;taQp5*kr6^ndCGLuQ
zw22LF=46W0d9SmPE~tRy9<*q5LURYDEc)n<KC&;!bNRciEjBQ_guN$bo;+CD3q0L}
zdK>0@W8)>WpE0%HO=0(i<BmN4(A$~|{A~2<&+n7m)umOr)uDcgA5Vf@4B4oDab&jx
zfs&P)ivAX7#=e>&_r0vm-*GX3swupOi~dKjB2wSg<VqNkfz7eu$Tndn!6w3cTY3<3
zQ<YfAJQk|eWPxl`X(d7D($g)=onyarfFMSA9^xHi72oe=hXeCBQuV^Ssbq87Y3ja&
z8K{xbXSU4e!*oB#sKzVa{Vl|etjX_N7K$0Im2UgIDe3~LezJ8rcGj@akUeS}0HW0-
z&g!_#T2QCekKG<z>UQ_dG0t99dO*MR@O_kBj#>avi5;SONBl;L2;#5kF^VWOVk|`_
zm4=mGIUXf<M}!@(Jg-!28V>~Qvg!MBEMl+wGk7eUL=xx&gamY`Qcg3RYu5c+0BW(d
z$zcScf({f5Io4&Dr|-63f|WOiCn)kXlMY9&xM*Du3bE8eK~BMh2ex&9lb^Y>#xeQ>
z)#ov%K#47y`i*dav=0!tmPk5z<=MLRUk~eb*4LIU&1fCykX?LmvWp^3m|Y;7gM--3
zMUeh(f+=Bl{*VA;HxwH=aYt*xb^KyKvhgFaDLkt)m>*_qN4*ux(#$HbycwXM!+M`u
z@e|+AX+h;BJMT67wtdhzsqxvod-02dOLo)7&~&0hP}m_(ZE4wG$r+h<I>DE4X<BCk
z(yVF>HY;`<5kyS0w-0U8W*sNTOOxEq`l=Jb;otTHbc)LIci&mypHRf}FNIm~sDMGt
zd%HbxK4lj5+1tGP>n4MbLYcvp*v=2twS9W+8>;xE5&zzEUn{DGv0_x_zc?>UJ{Q-%
z_M7I8;f!2oX0{o^T@6uP5On+)s&{`aj?16#oEsNR3Oi2cT;m742F6S>W_oX0{>Dpo
zGm|~xN9C^4E_d`<uJR~dcHSmww$4ATK{?tsp3VR6Z1RUEL#tN#2<ED$$f8Ei4m=bQ
z;KbhjJ};?JIbciYJx3Dl#m7z}C6Kvze4RTt0k&=hJ=(rKJR3`)mw!K?M5D82Ne*s<
z1d03z5C|?Kqw<R_`Pk}dI8jI?Z@8tw&|)CzKS(rtB6Y<Aju(+FO2e-7jVDMN*|^tm
zwfX7N5MSl8rru3@8K%MZkw8cUoevnb5$q+i<syGf7Gj$GA||VOz59c8&jZhv7TMgp
zrlB2Y`+2;6*Zdbc7I~!THA|)#o1^`l$Kf$eM-znwqnBl_8-3On`CDtMY|bXn*8D>0
z7HDUxM>Vq-a+}->Ul3|NR!jvbovhuE4G=+xv)noli@v7Qn)-4Y4hL<Z?uNmH-bE%i
z;lB{a{Ac*T3#O?1X`oiaX^(>w_rdGz@?&!uTOK^Fm*GSyM3r-RgUqTXDv(?dcFYh1
z?5>=&ACEGWVqi&K!PyhM7+%mh!Kaf0gy#%^{&L~-fO`VsD%C8<=`Ge)r-lIHZxos#
zWIz|&?MxFHeFf`bmR{3uv<*0XmpJB$3Ur{FA+gf|)3{9R)s#<Kq3Bz5jH<r{s_1(d
zc8t`2gbe;NVZa(JZi02Wi+cQaZZL?w4sZ{(uZwWkB{#*um40Kkvx|IBk@T|fVe_bA
zv}9(BSQo$wG)<Kq0aIU>jl=7^9EN1?9=R6XFia2K8zrtB+Tn@!c+lE*dxbETCd(e+
z69Z-HUy_fNC1m99D0e5YI@iIZ^d#`1F27ynzux0kU9mY|uV0W+xW}Ji#eCM+Dj&>!
z$Xc#93urNFx3iZaBwac1-i%<DQ`nuxDt)ye6D@Hk0cp1y3|f^w?v`i5K#%)Mh~xZm
zYNkQXl^XQ5bA%giU-x5k=63r+yJm{uqQgDzP2EWjp4;5KquR}Z@{RkTnh^)FbtCP8
zSGvqmAv~-0&?*$0ldi{yDf0RTSu>!z6q)ZHFQAtS>RNuKXY%g3Y}X__h%3^K%~;G0
zW~VYB=>bH?5Tj7qAq8dJe-D^q(ciWteL~ilw1d;mt5U@xdOe50o?dFvZd@rO7VCSU
z{ulzlSJVDAB<0{fjWhgFk`^m|;*s0YmBqD!ckvp#oC~8^F5;of^b!dtu;K1M*-+r>
znSc;NzU@->DNu_jSe*~Mjk1>=dc6(vAJ7S-l4DIjPYU|x?Majw&b$rU$aUxr_r&l=
z`v@j`vaMCAk{yFR7p_I1Ss!&m+Xg-R4_p~dNh_+n9Z%I?t}l~r9?e<e-^YIi6G*_$
z7+Wbw$4WS9niNhlbR5J2-Pe8o<q=hwgOU$5sOY15DJp6u)kexB=ePaT$2P4F;Gvue
zq%KRy{yIE}hlwecV?-)QK0n#KHB*fVmIiaX4W=7#!(<B@6<!q1=LB|~V?RJ8k<cLt
z+<D;-)zZz6`$tWIov{WyHc5ZYfMeMb3pXR=f+Zmo(8Ie`#(Q!a7@?2AG6xBfR-y>@
zd{|mv2nueg-+R6jpYSpzMXFmuH3c&@JUmXP49JfCv-ZdB%CS#n)BR}NrEb>=QIr;9
zHu@;$%>1GVaeV&M;^=scWFc(AaR9m`r1Hx;A1fX|+yYeUA#AX)G3DPLfL24beWGRQ
z87VuiXBD@dgu`Y@GUi_30oDH2_(izMCpkDpG4?)HdMBmMH@>J5U3D#^-%Ug}>n$=3
ziRkD@VZum$u}%J9AE6g)C6cKHx3Uz+3MEO{^$n3uvJWt_*NYow%@!OETO+Ny4TR6#
zzah-@8%}{lPN`o$oa&C<)K1z@^paD8s6hfeO0kDF)+O@vkIp?h`4P;#aX;sG3eqaW
z;^>%9s(1U(hxkh;iH~f=H{F8jucexT)_G1C1KmWdldeD_d{C=O`HdG8s<TM|cOHXI
zDbpJ0*V+gx@%W&PI1#IlEdA)h4+nEI))<?fS>2qZ4BFkLd?vH3cM+itwLB<77yH7#
zhR)^Z;;{Gu)T)UV&T(9Pklt%ZxW3nC_M~q{3aJgUsb$=EM<&>&y+qUpQBEA=1JBj8
z)XJ*PmW?dw_uYTr;)Ld1Dm4s;4Jn*)@(T{tQ*Rbh*DJ9@zb%_GqD7NA@XoXQOxe~4
z+Hc%u_c^}=DZ~jQ%-qO!yxmH15U}@ySJAS<c<{kBz_%{+A7xB<UD!(Bj?*2712um-
zfAA@3k7<wLqjnM3XVCn6R=~8Nk@Ari0<4GO6<C)vZ|)>+AG^&ucsg-BSXDhWJ-bgC
zk3#d&x4?%@Vdw#323C7brEKBD3))?H$udr*V8R`Yqh$S+wEfF~)o7?SL!cTvW~!bk
z-iuRjr_2NI@BJUXE`V#S#%7XTJhF;Ah0yuf*&Y)FLtCEhiLQFO*5Cp)%7Re{AqrYQ
zg8j=aZM3vpfSAG0{y7BC<L2OEJ3<QuS6rNk9wdjkspwNu%tvpcyVvY=T&$OB<<90g
z^IF&t2M}^J)o#ZchrTpZ$7iFy>gG08er(ieWnt#)u6e#JC*$J|Z99j!Z$GSFWiAnr
z@@rd+gJ$7~ZZZzTbWg;X-V&;eBQDPrnhOgG|3X6p1X*{u>RZ|hSLFxI?1}Tb<gYlL
zI_tY2=}ib~G?79%zoJ*`I6lpTOfB`X3E~_&;#gbz{6Q%Bvo*({+rnC}gN)2I&Y;V^
zoKai3JWY<E^|biX#PYW;oa-{z$Es&|_nw2d$zlS!3-x=wl>VetPlsj*V6F`BM9@uA
zvu=!4+hozR+VwVQ2;(-O$}J96D5NcuN50CXVr%W{CFwA;L$%u`ou}R{%3(|P_$=AT
zR>x6YRyEG%!Mtl{eGvS(^keV4QSf!PiyA{b2<=qTE{y#!Hc>nedKdiO?qe^hXDRSX
zy{59{!B+QI`iZ!qxPHCW6-r*?a9MOnKRDs{2j6~~#$-Cmn&0O%Pobz!2~NU<Q(*Ta
zJJ<PL16L#QbbkfZ9c+Ps;(Tpj0Tn>k-^I)1h8S<}vc%?@H@`mbLYUqd+5PuHwlR`}
z+Y=S6;OjSnmK7ZhG(<-Dbw_f%BJ;D8De^`KU2zPxou@g9qt&`~jxsu}uuK^4Hvp=h
z2vF&P<urZ1v!UEHbs6I}+M!a8K2=W8>hO~Z3IDmRy7Yict!`yz=Nf&q#MZ#XiFV8C
zvgAjOkDx2v#`6R?%Mu@bG~jy1VA!Obt?!b8e^LC-jV-xSmwD%P-20j37Z}U0jrW`@
zf+<Zk720+usf1A<{Vq|q8;z;Es$?Xs>)ks!Edc95fFf^VY|bf9)^&=_6qpK%)!Xf4
z?$T!0-;jQ9jvcb3_GP1Lk{!;|ee5#I4*1)CilNVRF1EF%>fCL}CX_oyAIBVca|8N?
z<;yua-gS7T1j;O=7OJi^Adj8$l3NQf^Tx!9T1T^rivirU;3#&}kbx-;FI-h}aJZkQ
z=11qOLf}~NF{OxW?^4(PDR<9R0B3{d)9g={Dt5eyHft2#tZ0gr>H4s%t1fP3SMJs6
zS)Xi1rZD?n#$_Hw&cuo}fG!_A=FmU4D7(9n*HX0!y~ZJ=k%Ja_HqeN15$?8c&gA9F
zF8SQNF{H3?8dFoxnWJ7$=Ux2yEfSBG6vehKM0ZOy%Os5sX~>`fuGGw8r<XyU>&0Nx
zl^stgThu!(^o>`eaIu`x&esk?pSn78n!Vn}2@<j7O95szYhKKaPt;bOjPZ4Nxf81|
zaF%r%g_Y?gcqBi)J}zl+4R(U3CPxa@__A>wyw%ZZ?E6vBW`GMCA2bh?>C`ek-TdiZ
zHUEYf9?~$k2xr>h+x{|e7uc`mTHnj=TR}~2U9xTP_i4mdMq?pP_r#^n8GnLGE;KB#
zaczNiX#DkQ^Qw#x_-5hAne4egAW8!yAXFN#q{&J>Oj;P(Km&NQQWY;CG{#(l+%<Mz
zBfpBG7nkIULQ_G{h&h$nvbT3WQmWFGV^9mtp~;_XY}j~d!C}&V$v%eaEd&3eg`nik
z_?zSuoY3YY?G_#_Dmf$~8s}FC&Nzj3^2la3p12`5__ke=(Vh^!(di|EvzyeES6<lV
zyH<20SawYYtq5bXOAJ!-flKZ%i@Ax;CQ`;zs-fnt2I?FP;B(kflZHvm#2YGt^vuN(
ztkmXtC#exUz-ZMdu_v^|b0kQLBZc*dS3*bQAS$M`aOk`QK5(f2y&%Nvt-4u)1(spC
zA={olY?co3ib>cieM0{{-92NlDQo&Hpz@Qqf7Uu9|D3wfJYB<ER>gniO<Fir%}v{B
z(+$))@bPC-wcdjZ{p(eb;^q3*0&b;AI_O8rL^YCF#OdAXuRQOBWdcHe?(6blZ4hjr
zUGb8SQkq2=29;UZxLjy-ySv}AroJ&(d!4L%J>umx-0>r=UpRlj%Ri*X_RE{BUJ1N>
z+g7_(vt|?{bC;^mT6P9+KvVP=eteHioCClu=}IiCB?`)zBcTQzAX1oa!#+zD&?gvD
zH1pFChTQrkkfBhOcu4#u8V*}F<VCrgT;1B7N9DkP-dsN$_J!>r$R8-JV}GJXA0#~E
z+35Fb?FTY;1|GMQ1%T8>aIO)}{x%;DzyFw<eJ<`X%bY~H5p)w4@7`SEPjJ3Py-b|t
zb3`kE7bWGSCK3<Z8?AAe<}2{=?U$_~!RhW~b#Nt(4T&A0P~QdO(OP(<6wN=mHn8Vl
zig6lbFU2_U?8{W-83neta+q0-LFA-3-fA%zx~Pp|%=lNDN*RqNl`9HYJ%U6Y-np0l
z9Nh0C?ZsWsIY<nSSBV>mVP&UmH1#4T)bCJLQtL8UbzefN;Cf)1`M}2KxRb<|FO9%I
zf7j*G>?%fDZ@*cisR^J@YF@&H2qk5&O<VpwOKw0FPZ{~$HfX`FMn<VnMd365DyZm^
zU^lhiB-v`aGL_*AS_vn9P1T|4l2+>n6A8&fTO)|^5P5d!ffqFNND<vS0HS1ECfj1_
zp(K1bik|cwmE+*jqUn{c$}SyJSGOM!+`Y^O$u<XHbOE|5_lQ^ahJ)Q|T6jMXv?%{T
zX15jSoI%-SlJV<4x{E25Fh-`q06L8mvWR%cncddhLtFLt_|=8SbdV%Ha{9Bl3kFhI
zt?$YL<xz;2+SVG??_`Nwo*jK@%&2I1zUcuRDp+@9-QuyH7x&1ZHhlVZWFR59xUxq#
zqE^A>PD}<}JW&2~ZG^Af)46(p_rmen2{r68Cv(-pNjAlS(RwgX$f(*-(9TVMC#29S
z;F!Y2_9@f^o*qm?{td^Np^vMy(OJPTA+@ZR;*e4*if3Q&F?eW01fMh=Gs$oB`JZ>f
zBD%T+&ZTM>mtSJ_3%;vNw;6#|t&SL`v8ovBXjWD92c(%bH(qUmdCxt#4kudAPyHsd
zrYc6gI02L{qNts21hwreH#RY~P3mD)&OY6(!R)Zh9i0M&hnuT64i(>F`IZQpF}&uy
z9rPbM-zC+s&Dp+S<GZbhsHBJ+l^NiQ<TjmnroqViDy}Fkk~|jCD?&PD2R8vZ?^^dU
zu`7lQ>EpWlzJUBZ!se4T)+(5qHFJywDLKX<-MRh?aVVMXlZCo<XQwp8pA|kO#;wNw
zXjO@{=B7Rqit=D3<xkju)pftWIv#)!oX>7p&R3!ad419DdY|%^)#4v4U1pe!&w^?2
zb)`9~z)-FE$7wd28T7IEck_p-a-HIDBsQ50`jLcMNXYKX3v7Q_>E5)MO70)(P5J-e
z6_?eJ@0SCQZYzfUI}&g0YTF)(=~*WT27Vhu4B{4nhdzp||1h6#zH6hjVqURnRHv}o
zMEO+yzfd=Cz2JWhJ8WmZ4kB-{p8t>UYF!VT*^-zlQW&f{WV)787fE~lym#+W>0udC
zmUTX~rvd|oIm<}PbVnTh2X4vA{_jHjHVUJq@oi{+xp{0XjP}JRG`&VLg=jL>JWIBT
z3p&@o5RcmsGtE6~0mYK|R>=_m#=VwiB=_^Ejw1b^*J}J<*V<#lQ1XA+$Ioa4AMa&V
z!U|P^&i=>Zlg$oGGKl!ro6a2!q*wUA8BkjCe+#b3rS$55BcHd(8#Y~MnaF<{s(L9<
z^W`WW5w@jO%t8+Ir#}bc(Lw$fjHtWwPt;BCB}$k4kKgv+r$m0~lF}lN(i;+Z2mb#$
z>i?eWzv0{ed%ylC%>Mt1FU5kZEQIT}EZ1&RIW1PmmFskWt?Tr4z`BHLeHQ0`{B!>c
zn|}5WA;!k8523L^4nyGDPKSQKxS<aA#9sbopSP}c!E?2oYbK=KA##4$n%Aw<pz$zx
zL!=Tpn7wTPy&ZdxHNBuIR9f)c{gcOJVGT=*s-WGu0GH{#>3EmIWnP$8vn=!7KyC&+
z$r=f!GGFJoiYM>gFHA4yx3uiqUc?0T@U@;nK_s%@)hw_7EKE=5wy=(i0PvDdgeVml
z#_gxCpIAlY5_qLN_A|^>zz;7k%Gr~CC0{(rd7LN2AZi3l3u98BJ{m^aFt=y=#@$c$
z+dqC?sy!W@$_K+$;O_f|JguXCq({f<Cs%@$KOV^X?uwL7sZxF23I5-ZXVUh6N!$yL
zT0*fs7t_{)yMASab&H3Sg`M$;dfwqvWj>-K$ZTqJn&@Yz3-vN6|I#vFO+oPTDW9ux
ztr<{Tw2)RH{K&?&UND+}q8}N>v~3Q1v@BTN1tSujENnh-m@R&RT&tN+i7}Y$@Hv#;
zu4MD@P4XR_0}GMUE>+b=ZW0>$HW&5F9Y%!<`>C70G^bTHI9%8x;x%(BN;lz`6!WEy
zf*y*NO((E_>j+0XGC@+T)#j8Iyjk)W<abWvFh({^E;W7!r_9fvGgiBv-XLmboThds
z4^>(P&T9kY)dnhGhp{eh@bTmgj#6jKQ|x}fXiUk-YkzZMeFJIcU+>jXPQ5+z!d`ac
z$z)N5U&G+acwW|pH&iM3{??D}rBi~T)?rQ8M~GVmsAGyaVGPMG!LRrfdU(Fl`sm<}
zV{d#ka?`k*?;`h=96eVI0yi^KD$b>8J)vZU1wCPED!77UXVQ(X>&^Dly+V%!+gIm*
zV#5C;Hd!O_k*kX2nvrm<7`4ZjklQew&<$3kCkHX<-ljd8Y;v|6ci=i}9E}rczZoAJ
z&tNG%E@I(`+KKYqOr6?*rxMt*==Ei@0?ak~=hd6dChy@YUYBD)+JG?rJ;!O=Qx>lA
zagDGSek7KAtp!35{c#50&<t*y*Wm9e^BPv#Eq}be@uWUH<#Bnmvh|U=ZMsf3+qVc;
zmZ?<>xcVoYZWji#ULW_RdnwN<$RymDh%+=@({ddOd%Sm!pVm_`@xM;RURVB7Yvm?Z
ze0zE{p6PV(k0|e#XvsG6+Us5BSe?6CZ+-3$yYg{LeB2f^o2kMQl<4oN-_9?!zTu%w
zu4e@tN_e4jP_%nb);wV#8lFyPy@L(1;gZgzf>jpRu?bfX(5WoEOzf6)$EQ=_!lw&`
ztGwOq3m;kXx~BBlu$#Dl^w;#(rI!L1$M=ht^qP`J(cu=$=7@>7{@Fr_BaMufTY3`m
zqb0nzXE4vSSgo_+zoM?o1<Ym_9rLSiKd%~_^=K#uu6}kJzAmM<rSD&Md{~o(j7Di<
zX!WuYezJohcRV!PI%xE$mrUFMm@U3xoW32@{+4>Tb@`U%r`;@(<}}I3k#)SZOs=n6
z>$@@g9qMU_SGjaDdS(5|?9HV7MeXSR;8agDoOREa+je<)ZwmNfSSb<dT!S5Irr=kJ
z*x3E2EmwXGN9JeKc+0!WM1s%Lm+CFD7b@s<bh0Z9?_(CKcJ|DMBjB!8yF3k3iIa~?
z2lZZ_CTicgJ7x2Cc~Za4%HA?wcs*hv+Igl*1gTXOYOI-?T^W>F2zeQwPY4-`Sz!9j
zBC+#{x5tUJw@6Od9A9t!$a}_1fe1#nb)3`6I?++pKA(v-GWKv!I8bXcMJDDd4z?IW
zt1Nd=2-v^Veh47&Uw<LrPgp*`?WC}?Fm6n4>Ma_8mbYOkBFw!lDy}Wb!y8W%zh9h8
z^%A4<={&%RRzzQCzsWQ*q6fojkq|I*^G+PSoopRD6c{FW;-sCjVl#R0N66#aUo0MU
zFti6@6z~`3A-3m^Q$L3s&XN>PHgUGE5{pxpJE1juo?;+FAXq?2yjcNjdR49#?Ix%F
z_B*WR7f;jU$sMr7O^DAe&#TaU1m|?qcGvShgHn&2ja>>^XK(Hdv!1m#+0R1jPEW7&
zjfDy{4<FVsdsvz|;@(F$!F4bxJZ)6JerPq>eYzr{c`tou1=TB*d|>v5x>NDKKq7SI
zXk%NV0Glzd&zeX(9S7`lnQt0~OBI9Unmv5DEgHq}aH)hp8k?qK2CTZNbV41{-9Hoj
z-Br)o#q1qNR^rQKpuGhNjN#QSyN`sH*#q@g@-~v*nTG)Tl3-3%4eas5o+GifFrjVv
zr8gKqHI9~&jbQ!08cDbFcr*_0xU+3u_U#Fgvi-s@_2c0n{akNwob|7D0EWvGS4|qO
z+9VZ*)^CZZ(B#9Z?Nyf0yohSN+6}4o`Vj8@$l`WYnbQ<AR0`w#naiSHs2S>rY#0!{
zJwBAco_9l%aP=Cxe2wFuOU?O<+?Hn=8eQxa9Xol2AYVsML$*x^sV489<@Uje%zxLZ
z8yA#l6*jW-7A$q)Uc$;+lDKokXVErGW`hcP(KkK*2aDfqmhjhH9d>NC2I;DJd9?wc
z%VhgX(vL^9FP4Q?`!`0YS#UPg4A=vn=r0x!_Rou~V0yLrQns6kZ^u$GUpr_j`^*w6
zH|kfx9^xT^RbiCR7sykls;AeRZbdIm1M?oJH<qZD_N!rfnk=qmPkO+3pDKm-@5m7-
zND+97kMsM^;O4&MhZG8ujj4{pSu&Z>8DX7=(`2Mjs_iuS-kU(G>itO1Eu_3s*mInC
zX~<|(aq&DvWiBAQ&LIRs82!f&BuvtryYR^N`{hKe<7INtENSaTFs*h=G`q9KUM;87
zB9%{k^v>%xg04<H&-ZoPQsy>i4MunMWjlNK^msv$@^3kO^q!`!kj<s%GV|?YAVr;A
zwYyL<hrY;r=z#Un0X0s?hz4><(q4;NXM6xsrrV4w|1qYy^&}tO!-qx1(fMmXLO;Zy
zfAEEoOHoNYg3kfausE{5g;eY(y2+ZdjRwX4Qbz(iZhu2Up7N;XkIJcNgk6@VisTy+
z<O*s?!yiTD8O0Ij^#}I&;Ol&~#B<TY!x_9;YSJX{34_y(!nbMpOJqv&%&9=pcy79g
z1-o${(L5TdJ5JxYh>+K|b%veheSU||<uAo$$sx2ubS9=pB?8Gs`d^cEn_>}|^aN{m
z9npI>jOPvrRM_76(jf6(cXiN3prO-V<E+p>cW)-yx)Ibx72<g_Rz5RMJYw{6Ab}MS
zG~li7VDT`RW-n(u$*sE6m%bVJ8d@Sw&-HRj7TzJxvFW2m(rk-cn{6syU7SrX0JKi<
zT9_BgHz?cTUA6NaiM7_!Oj7aXh%o{#ukBD|^?u2WGCbsQ=rqp}c+<dtagv!FTE)R;
z*oVHwt|;Fe#+YW#fZ8wp(^Hnk=2AULGzipEqtOJqs2q#uJ^WFtTy!&Ty+_8yaUR(#
z%_L)pw(ogK{oLc1Oc`u)fL}&GZ#gpREZ7u}Jsxk5pwvel{Rg3*uW^(8ZgTn$$8_@-
zw9nY3Mj#RW1VhFQ3wO^^Dn|V$uz8-ROmfr3HWKPBb6E_E;gPptA0k84C&-JeelY#z
zSSDab*xG(fA>Z0+mphQn>hK4|Vv<MMddm8%qhQk=5CKq0RSi3DM$R+}wK<F;hfzHY
z>Y}<yzveeBaO<6Yz`5JCOxP=(nlU_^-Urn<d+QMgCUEU>s<d_sl-oTb1kDbZQ^RtO
zphpt+RH{e!>lD>ir$>3MmL9M^QX~C*`bVR{6M2-!ph6Z!#@^|ITF(c|X&l2_K`&*`
zSC}C&w<iK~&aGXgF4Ih07n;J~rl<A4oA0?rjIGcq>Xhu{tTmghFP_n2pw;a#R|(|x
z$lU~C%9(Ok=1yuN`PD*rO^x`(uk4Rzn?qsZ%M#jI5h8P4k7jdSo>z<1d9{yhK@{1&
zG+#)Xb`~Gu(-N%?xmPcfGRe-kmG|9u({8toe@q3WI_3X<GTpzHGCQT#DPc*y?B9`!
zvpKcin`AEVk3<=-x4GR!{H}ZoGxIu3GxJE7yEwlrSYEcDHmodBo>Wd^fz3DQJe-}8
zD&!XwyhKDnXY+#KOOzBpzR5S!v(wI|q~>??u&6z;KRQjn;hN1K6u#xS=$ElR{{4On
z!_M|1K2-^}(zS$^sPhfavs>V^&*`{?*9(l+*LX{H5ZIp07<#`}sjGYYQ2`$y5F=$D
zfl$`T>q=)ndeB;O-=couADNZU_;s(W?)PHy|6%JR5?))xQq5%kWyw7z63+P$a-|Sr
z2w7NP+Q-X}eYjI$OyXV)Vh24whYYKgqHC};lil^t>L@B4m?>Hc18e?hC*MSotDyR4
ze7{w5e)!8jx11ksi$CJbbyFD?1kh<c^E30cYx=eMYsP5eh9s}w%k@vgPE)y+lDuQS
zN26oDv9{3R^i8?qwvgk=rkQw1Dpk<*T@S(Z`J!$1b${cLN<LPuoEpi>%efk%wz{M9
zg7Q{juw3ikWExzIV9rAnavffz29GM2`2p-ueoHaI&nCSWF!KSj#ys^tPd}#+ul3@2
zlJinV!^+nctf%Wjvl$0o&W(-H=^18)g49%jmxisM!B@kj`B?X&Uas0MBGoC+9`UV=
zb<#?nJ+9n6U?68euJGzYR;g7RA(WloR^QYP7O)(WyV!$Oh(3Ht;Nn<`wpUesWdZ(X
zM1FbZo?AHhbj)LRa{aV1t*NGI@;>6O{487bVhD@5;GyrO&S=<0S?OalJkPHeFkSa_
z<yyTn_|g&exZd!IrTQtpGbKe;1^?-H&~%L2@W%mfF|ry)iREX(m~Qun>tVGYC-1I!
z3NCr%e@{^3nW_|$dA&N_R#U;wg#Ttm4v_7{2yazzzhvhM_|ttX)w<~J(=c(m$booU
zWe%nq_1+s3QdKBcy^Zlw73Modft*dWvbd1D6I$MDjuM2=&&Cb8Qfq+yw=aF}(s5Vk
zUfm@N5*XY?tp-tCFbD6$adhW>8eYYCJl_dbNV>urD-9{Y+R)$c1*+*Ja%re|V{;$o
zsf{*Gtnc~`wjj5p-n~e}AXijCJ2m>L`m4dIEP3?@*EjH1g*7jxjq~)&@O7WjT69;c
zz56UI7DcEBF+yP0+EV~w-1ZB~e_zOj=u9(r6r3$`{cL=Zv3z^aG^QD2P0KqkmfWOM
z-4uM_twc0}>S`{OD=!G~zny~ED9Y@wnlz_aa8W*U8-qe*ojSS{@^-!K>Kpba4@DC`
zO$z<7bV<ViTVfHuIFNa>lfvs-Q8gTsP8sm&2s)4JzaJ`GtMhPAse|{6QfO6wGM$pV
zP|<%Qq8`OOHLWB=4uz}3oQS3H@vj-W+s2<+5$$xZyXsP%_1(*LjM{uljJFg!zF5L{
z(;K3F?jBb|SlySVOmlXVPaCgFo1JWP2oH!9c!GvboF80u{3_)$a8X`dvwy^o*5yW9
zUMK7gr2O1wM04y;Xq*z=<}dE#?0Q-Bmr&|Os%Z^cPB$LI&Yl2iu&T0<T5%`~UuL!(
zXH?O800^6jt;C$0_@Ev=SuBiLEk3RM)lP8p$;kz-3GXx&lj*yJ6W9LJr#ApWxWCPd
zplzb;+B!UPJfEvw+`5ljtIVA~4^=foe8As(JN9ZlBhu^xZ)klI!|rGBRL4L^_mNOf
zM?UGGKO*14kMy~yu?Gy9_#Ds_oVO#XLqR#<r$`rd(Z!*%1jNoe`nVY|Xbo?*?Y84;
zWzFU6&JN-6<s?MG!-CBI-!eyNto4OAq5da@dmA^~hox^j{OOtr<a#s5LQBp&WrzFy
zB~pCA(jUp5$QHhiGB?b<Zw35)&i<#HNvl|uL8-_tK&qE<whiT%K_U<9c7az&O&|vr
zWyI5)OL5li5stZ~a`h9#g7Mgro>`32WF)W~+h!|Ejzdsl&&_=qdW-S9YZuj8>dkPO
zZ!?DJJl&?Dj`0GAHzgnfW?Q1+Y;PcHV~wW-L{cdS4o$`@ZcFxTcg)YNko^+vRFWmv
z98mf>;g1|VVDcbPHi;#c{#TmLJ!Oe24A;M>w&0NXG}T+mQLE6Ep~>qJ?!m3UtI0fj
z+r#qB5y(mR*k=y=sF~U=SA#uHRNMzWh?^VLF*mi5km1xxrE*cz&%*P7MP;FAjMN)B
z_zw9{)`&73I|a{SUS@W>+@5$BP%w<z(&XXtFu%^CQO`UFnoWL0eTnf60u6WRTCDDV
zY3&~Xyo4;QQO4{slDSVdWSUksg7qQ8lfx0HlpQrwInWe^RD>C9aqef5#es61@NT<)
zVhg$EQ<9_gR!zN9m8NXpwYs!X#sRwlex|;{Q-=Y)Q2;wn+ri{usA!R7&vD?X%UP3Y
zgx%tsDPy7)ec=ZyI;L!YM!|&r1GS%Jg9}0~-wk5<qAnlZBj0B~1Xu~@E?o`U=+F=v
zc|tz5HF0WC(hks<(`wNGp-M8-Qk9d204Ob3-Kr&(@DL_;?h1b(Vhl03KaJ=Np|BC4
zT22I@imoeiv@44jd-{*4d@w5tzAy3*O(mvu->3tXRqXJt-M9_pc2G(zWO{ECjon#^
zIMvi5AZM9w3-tikLd9dryX0{y=Z&<Dn^i%w((%`GiKJvr2GhjWXAmn=V`%1B-AWaj
zIxD+Y?P%$J=I5S@lx6Z9wP^YQ>Kl)%g}ZZ)!q_-*VHRs68Jd%gDUHlR2=en8HQQT_
zGZ@)hLFwspEhji0>Xo(rY}KkH#4(L+7#Z;D%Nrnhg`b{9k5QdN@$8(44V(4Tc4*W{
z5jYn*N=5w>&G0>wF^i{c>R;0Ck)*ckSCr%|5+1Te1jqbE5)|umzq_Y)&536a$iXV;
zbG`hV#Xc<IYG-C@ZV*4vk$5u1Er%#ZfA>O|LcQ^GXD}EnSdWfaiu)xW`KDpS#}Y?!
z>o~c1+p)q&^7!r__2glG6zX8$6AIdZH$8F#tp_?AzktE(fYv%+Pt|I#wHjb+9>Ayq
zh4&6WX@2k-#~4@jnK8Lxy}?Zi{wr1kHI9_vhC1hoWiqZxhpu|Ob1gw*@$i`hAF<x~
zWkdU*x_VY0UMg+kgA(PuJ8iC7oi=W-lLCk<<e=%TLRYs<k+M{p6*_(7-E~^i((#|l
z;5t&_;>X-98Eui}dV%1Q5){ftiu3y=*FR!zS`XzIsH1uH?dJ4m#kk|e?z(8swN@jX
zr6F<HgKI8JD|Q>JnEa_NZrpU0eEgoTay?N~6;t*dCUjMB;kZv@RqG{5t&*j`9@Y=F
zt7$qKyDl^6qMUzy=_2E+H>xn!Kg{4J-F%4~=sln3ckdkP32#p>Y+++Z&)oO3Cvpz8
z`+aItts?I(tUod>G7mh#r~;Ku9t^(fCv_U&7IvoH`JM2V$}eCY08jAVQO9m-#d~!2
zu;a@IbMq~z2E`}oCpd9+gRaOJ*_TeEX4t~Uql~?ojSwG81xkco&N2GQfvy}q;J0@)
z{J563clbF2RSJ5V@(_T+Z>JOLi}vqsa9EP1y`jly*9S%$A<!bs4E^#<j7idq#(Ec6
zJ%sb+ZmRsKQ9i*1AisIM;P#ZpI_udt1<gKI-V(PYpQ06~&iZl?HNs#Y`|KQ+$-8)q
zv+$@eXVoRucEeSIGAIj{?afO%2jl1^MZ`-2KYm=^`FZ2g+cVN@uQg9;-0JS#iNn6(
zCPa|av@Vg@cP1Z@OwyCt^Qi^9d$Og@Rwne*0JH>)Ta$M7Bp~OSkUKA02yD3?vHZ_;
z_H@2leQiqO!+A%$)}D@+rfx2wnVIp6z)DHS4)<FyHZs3R0EaHBM{gzL{b^l&%8{r=
zTr$y@{*XM4PbjO3@L=9S)lE|(6RwMc`1&ZSlC{khgQf`FFZg9zQ1Ihl8l4!9Gzd1V
zI1wLz4*2J8Z-j70D<7f$mR`(qtV8kGx&B~*&^HpLRPV<Hwnec&+`9-xi!V;wr<CKG
zXc3rj7|7E9hG=CGoCrm?e}{|}Om>ft+Hk|Py?Q>akviCr$t%EJG}@fz#M_hl(b<QM
z*~WFzgR~<wg@~dywvy2HAh|$Yhc%Yi5OgEf^+OLY>akEuC7<f6Po?&py6rCye|74y
z3OjsgkT1`fKWg&GxaL)ts0wX$;c31sL5tfQFm63*XT)yX!`iiU$WnBj>-R*&tQF~J
zcso_G;Q%V<3$J!kKEpr5263rR2YUY**QMlOfhD}NuDbM+Pb2Q@xYffFrs$koMHeZF
zLWE<1?DoH>jg$C#Kr;CkUTnLOPpJD$OB7#yC!6|XS8FgmYB4ls)h*d_sW+LeMkS%i
z>|ULd-trcg7!0)jz4-4O{@lu($MzIZwES^JGtFoaGgyg%na5bA%5!<aqvEp!zeM;Y
z^T?fch;1|n-}c}+kRMlYJi4B4%n#?_)J{3=Q^CGZNkqip$@VrH`0hQYho{Ng^iDF%
z$`hQm96E#j;FcP#Z!UGSYf`Cm4E98VHAhCr0@891;hn0)>Xt-^ryVLQu<QKJCxsMz
zH_={_hI2ocr*~ueGWtg3?;a?0CdN${=vkbStSNDx@Hc`k6%A@4@qWBt*t)NRcJdVu
zR(58x*S_@`!qL8a!5>>ZmF``Vj3rETeCNYT^xA9cD&#!7xZxIa&r4Te*<WfR+c8<J
zUT#uI<;zz1$m4B5XK`yd<u3=R9jSnZlJBMiAN;uz1gMQs<JmdXvZenmPZ*`#JY^9=
zLbPooz4LW9o9p5wmWxXD{qeQ6jt#kgYi^RU=mtly){lDzK2WCiWAaF>i=sxIcAHn7
z>1AV9PQ0@2eTq+Ca?ur93@T<2_@7w-ZL2q>rd&{tgk22<p&agEmS9%A$##S6KVi1)
zn9|ML)v9v6t@=OaH~T`w7?%0+n2z4?9xfm$Kg-g3Tp^U+i48nwb`mzTX02iN#cDWt
zx#9Kv=31uG$otuFp?=VsraV%G#Gel0yn(Q8qd8vhC4{2Qzjj5^15U(HiU|uk7HSb8
zd^`|r=eoq4m%?EUtJUI9l;kK*Fs>c@3$m|P9k>DJFIWlPzcynUhU-&ke*F;DSBpNV
zF^prSxpkfPwa#LEARlM}*x$5O?OP1<zW?0(Sp#CVsmmMumhlj|oo{1OfFwR-g_gGC
z7zdQFK@nJW_Sio!1<Bp6c3hg6Fyd`sRk)AvX@8SCU_RD$UAIzhqxhF%w5BDUlhp$%
zif@|MPLACPAxm9-!;@KkG@bh5A_lnKGv5|R_1H~z@2=aC1K9nvO@bYn5n91d8_fc_
z9I2zDc%5e5#6i5>M=!w?8(~DO<leAgd0brLEud+{z~0K<0_{KsyGycHE63m4*_3e~
zI$QdRVn*J6mqV!(8cJE(@~HN<Fs!Sl3roepk!wIK!_<ji^M2~j5xC278K)L|2gEzb
zdNBeKt=yM2hSspJ7&J*RXed}<-cgZdR*LOOFKpw+&=uPIlG1EM=q36e(UpZCX<DIg
zUe@_W>^s2P2klp{Gvp<wg<M|x&f&vxt#N9Fk5%jEh)C{(8K=3obCYSq*ISrCi~ATx
z5%ukssb4(Lgxd6;99*o+;;#>8O*O6Xim)+0FR|mv^VGNws=y;On?^E*`1hKpg;b=s
z>?CCnAJ=STaHjFGF~2Be27Qdaqz(;CD)VJ`Vo=ZQKMN>iH9ak7e3O&dIKF2(tALz0
zortJG%&;WA`!?3ANq+7zXcfTZR376iQK%yXQ%}Ri1D+w}@e(XgePfW2mSINECtM=4
zxPypmAxAr!QDONAy<5)AU;epn*65TBvSD$G?MkFgEOVC)hgP@}<<~sl8;ZrrZL@zj
zGZq<Ei1d+aWMpHC@ZNp+Km`>+FwMJHMFk2qmVfMSG~thNn{=tdi5;*ym0x^7XPU3O
z#nSiTKP;iwBNM>zm3!4W%A`g*YVy7K+I{zZ@8T=fatSMo?0##Yd7ohXA9+iI{KQgv
zK2HH^g%)rRb1<+JIWe(Kcx^>Dduvz4QT)@OLEo}pMHgG>duYGG6e|t}ZFp)8_VMPf
z{dhmBB{=<=56-@V&#LU;RupKk{B`Vo+s(nQI=B8O7Ix~P6Ni{6M^*t3{aHpYgEHE#
zg_-0mH`}tzESzIUsFk_d7(h>hmc^-o+yRI0RI5(e1&4C*x^EA?+CQg-?Pj18;Ospa
zz7G)U3<K<IYf%{d6!6z9hT~p|Y+7>mR*o!Sar7k)$JOrQc5F23g(`LtLw!or8&TxH
z!E<2?dMlWwy~PyTdsBkeu=qA?HXRHXZtQ{hZ&qz68-K5Tc6%XR0YFl}b^UGi)A*C-
z*Q%SxIJ+yj>RE17Ox+%s`^_G8y<3UhmHAj@)ZlG%+wJdACOP+Ksw1DZ%aw6Q!W~DK
zee>}BHe&Fu#R-Si;jE8oF-yQ&dTS9Knki7q@&%Tp9m)6|HV7K5D6o-Vyxu%t)G`FJ
zOu(!gC-E)<p*xhNem)TOJv33;g_p4dTqmTxV_$*p=meN{eAj{8yTxJ4rO-~H_`J+|
z%k_?!414#<8inq<cigpBp4zaHnk>g`B};qHh5YkPol4m1kY4n+E%aKCS2vV9yuZQ_
zVefjifG-}}S%qpCY=6X!-f!Ic*aw~tP5g^`Udosj5`WEX0?L*1>%4cb=$~W~jFRGI
zXMP<EWZ-0hHmz)^s@nV}?t2;EWXCs*pM@mLH`=2r7hjd=m^qMsv~1u9@U*)7DWUwV
z@5sG&&W>ky-=G}a`lPI+pBIlJ<Tj1oZ2BO2`|ij(YPoEpnZQ{<JiTMRYQm#x@$mR-
z>EQfd)0mW1A4y*f=HSFeX=~T$fs63v^UfnuyMbL9hpp_$_O<;C=iqSeSs)pfTJWRm
ztXA!ZZ=M7|r<)G;IlKMwfn!=Kl0p2QbxO@f_a>IC;wwjTjg+&ESuH21at(fEciAjn
zKRJ+ol)l6#(BaHRDwMOu*kExen~4{@v@D&e-f-#@PLve%q?@hQFmzqkU9Lg=B2`Rg
zL_%JFF_;mwtD8gAXGl1Fp9uDpjWXhhuly@YH5bY&k71=fdS$Tga^eQk`Bd7j(<4EZ
zvLLDqkAnpGZ=ctLIu*HV_Ww+hIgLWzjYm~2Mvi9H>~il8Uil`r+WY6o@$j!Abn`zl
zyJ6Ysg?<LfFWOT)_@eeO4$2$2hYc1=dbzltO*~5s0}?pJxGYzMq@yxDD<R~0_aK%C
zEYe?=RVWj^VXE6X{T$~Ft!%+DoA-E3Xy9N1ZPzOU-=4fT@LhW+Oxy6QUtp9GOG0X9
z$tOTh1KT6^J*t}eM-D!ns2X-m6MGhhKfZb&&esRGpcb@|WvkaYmZ^baugc5!65M2>
z@c&9z58s&LnfALu;kbBr-WrCX8+tYix#938DndE&2b*0qX9>n(ts1R9f0pK^hB_Uy
zBmt!qjfGt4el1Dk&7-bWRTo(dfo^>qh_{^tN_S6rNxy?tjNEH;&id2MHyqq}ttaI%
zoTcBRbH3&46Mg-oy7-w#{kN|hjXKUGvQ^lN__$@Vf_R{llj>gf{?vB8={mcvmArdX
zZFV0;Ek2T5G?6>SbD~)gP$Frb#|pMBU(695q!dac38rj;;wel*$X^?)C2Woe2qJfF
z3pucFC+fP3X8${ArIon7&>uhgzKj+1YLdueOyJ0gre~U~F5>n!<W?#UO+*H;bV0}D
za|Y@Ju9T}qvTAFg{%x;2i7Hrz1}Gig+i;Q^QSGBP9hZ*laQ8x2PEOAZgSOu%zn}=B
z27c7t1-+1^$He1^JQCHg8%P766UKe$|0OdUNo5O8c7^BA_j2U2&5vOdZQE*9H7wWv
zKYV?4SRBvNF2UU$0wh>)cMA|KxI=J<;O-XO6Wrb1-5nN(;O?+2vbf*<o%5aho%_#y
z_K$hCr+22Ny0*Hfy6SyTNpt83d@O=^eG2&}KAzT_+;Q54c6zMvr&4!AN&$)atAku#
zVZ+^mdW6YA+@T(WKoS~pD?>=`g*d-nJ#M$N(Z1Kn36+;|3WdobwJCdHO=bu51ijC9
z5m<!XR_;$DPTTRCCMQ(R!mZ;cUIaV=UwI8Pd5L7y+ZVKwMtBrZn8|(W&y9Gu_tLdS
z4$@z1uD{^5mAq03$r**f@XKFq;|^j1h4^OmIN$+g_L&bKSOhAT@^IMp4=8OGnf_Yo
zZwao-jv@|;lm26KK0h|bull{?U0dET0N8nJ3SR&m49@Z50)M!_2=K9%+7&o2Wvo0N
z)709YLf{IU{SFG9d>V`hjm00YfqF=;(c?n{v>P3sCe1i2mPd%DT`WuJ(=5)xYuk;A
zpL#rnsM1l1oKN61G<%M7QqLuYGs|LNw*CZChkx_TpD47=@yMjoF0glGKPqlxw;$Tk
zAR-O>M&DSOZxC*ThKR(d?>Q6=NTL4)D$KhV>{5%#R+K%yL}p+u8nQ52dE{kC&oN)v
z36-VXF=hTz@oMpfqUX|0pKt`Pns9w<;6usB9(aJE<9PS&5+p=}W0u_m>SQd|oQZ>Q
zNeBqpz-N9le^*y*+gMHIQmYlTRjBdh$HhqD^bSiN^BR%VF{4keV_3cjE=g;(p(sNA
z++%JoeAvZ_FC6s@>c9UaPE;l0!niCI%}$L_)sY=+=HO@X;7I1pA>%`<jQzImAgg*2
zaoD-^{pkK>oCU$vgF{3saHs#iI(%lbKkcFJ_r&pb!?9L4mY<FoJZLVZms18l_Av}T
z_bnAA1N;8w#zPk1W$3AWxdbPBeM!`P6?1FLKJKZI&&Isb?T>a(IKW+UZJS#cH494m
zkUW9~`g0JH$i59)+K~0JV4L)cbAJf+M)>5`u#<R%v-v~vJ5F^i3q4x%vTN43^)X>i
zoZH%WP@bN_7qp2j1(DoeWAOGV<ULl#N^!s+hD0=m`#bwa8iTbskLZFIeNJ$)JH5Y4
zZ~%3aT<AqePF%_Gt1b;nk9J1DsK2!~?qr!BHjHh7qR{Jm6<GYU9=p^{5t6hCAax*<
z?+94Z6u!3CUCV75z;!O=Q}>8RAbcmZr%YElzC&{8-}?69HHL3%#&u#W=mJG)u{l|?
z%keMM2Rr>X!Ywx4Hj$s+ju@2#4&~%mh&7h>=eHP7;pjh-G`D`Ai!yXH=!fxVFX@%f
zhw6Xb{l~DKfxl*&r`NJ+0A5<jh#S)0o8%2J>&;$lM(MoU<zc4cl+3N)3lA(^3=CGq
zXv)s1_cGLaF-Z#h)}GPK)8G-*sE=p<<xLgoHNh=l(B+KL{%(c$VKWOx(5i6A%i(Q0
zWUe-WBqi8&#5Wq)t@rXAAs?I!$Av*K;HaIAB9nS>!gLPUKBm#tOX4K4yk13~F!BmC
z4c@7#7UefyrWel^bbTE~yzt2P;Kx=Bn>Rb!s~q%u{32h#I5ATs*l7kE+m(IR$=aD$
zs@31f*5s)Ljk3K3PSE`I;@F(tmB~%qG?)QdgSLBcMf*qx^1KM@o_&cH8TjlH-afW%
zeeT)r@<oD5L!T&`YTs*T_;aj#HSV}C0igkFWZTv!om634inD(foS9oher|1Y7Kxta
z<VjIE8iPOr-<lh)sJg%RG>maUv1gM}JX#6|pVfV~!jIpuKiO?!2c*WXmRSt=qU|Lo
zI;DfW65+_tzPWX~b+(qMCWi-+aSzVX1#$Z#u}6sVWj9Y;X*X7DQ;TZU<}CRFl8-xh
z&RoebF@ph~3Exa?;w#se$<jFDJS%@~w%rMM4YB_Q5@s1PGK7>um^+S|reA8Ivh&Te
z`m_)@GN>)~6b`r^fGHz(I%PtQw`fIyULtm!ozE*5aBlE%im@2c0u}?;lCuv3Wp{dB
z5+LN)%h*db+s(tlzeY}4`12ST0oztbT_xH|sctV5DGs3_wj`MLY{EHurt{7o&oEw&
zpr5Y`Q4o&LQ0ngGm->0S$BF&a>ruQK((WQMTw@NNM<WeL*=l!z_RrS+7PqN|uiSqK
z9;DWyFU7?>1;;DxgMPSKbfLQnE}^NFpmha6CI0@SIw^u!yMKb^KUx+{xaNz99K>&{
z2Qps9xVj3DRZPCEWyER^8oXW)W-u|fMRs#gXe0}D86xk*C7b)Bb|21?J>|hRkY{1z
zFR+&z0x6N4Y!&`m^iOEuXh)k(;nB^5?v)t4nEbD|sCEyb!Y7l_R8^K#@9>e2zX(vh
zM1AjQOnJ2=JSBl4l(p~WD(nO~km44|R4*KQ5r?Z)AOkOH3=>|8s{m`=RJXf~{<&OD
zP^e@e>XXk|ghMkC(J6BnrY2z{ZUFd-zHX!J4nUA1qKTavYm0~zQO%B(g<I@_mb3;F
zRyj{K8PKN!m!DG|3f;I{=#6TX%Xvk7QaW@=Hqg$^Ujh%%vy&uZ1myZMd%QUcZikX!
zkXr}WRes-UuaiEVA+qWpHE8GwuG`?Znng^x#2h7zCKVo1=tj#A2$UQdq?D{Cw}m=;
znR>8v=k_$=qG89uHuwBN`E*V_$s7sp+2gbo-gSdkIfd}R;1PnHc`^X;Anp9>r(uB=
z1zK2M+XrxL>q=G<E9PUR%5+_hYBpj<U^mTpt~@GaK^ZjnJFuV`HMoIXiF`u8Yot!_
zlZ!q@vKxtMLc-BXZe4ID_{Qp=x08dQ*4Mz-J&?@VS6I}av)P9C_FK3Ua#4SyCO)vD
znL~H@`DnE0-IsQT#$G*Ix-*ZB8DA$a;p_uceTL~U7na7%-S*^=H4tirRXwpfxV$Hx
z&XW*GLpjAUq}GDk^yM}$)z+Wf1P$q4cIX^kz~sfYz%f?SZh?aJIt0uRcI)Zn-H+^z
zCd7>Ezv-P&KfR_5GlD*K^V;SyN|<=83N|OuMcG$gUtEP`1p}^;#<N+UL&K(Brs3oM
zz`t~u{oTMK30=$9e==L-5vDZn1(2!M$c#7(^p+7|LP-amb(T&NpWjBA?VVuRN8e$d
zMq8VoLco{yt_J3>lF68;fok~TIm(A-x^xDN1g9%s5gx6;;@LJ`W1eI@4l(VKH-;)3
zl-(W==9W3X<@(LfO^RdDPKgpHg@S6>ZJqfVkIeL_&m4ArGySM_DcMJEP7E6rzd?}Q
z${0~vQ3sFV?08xg>JFUtR+`y@!V@h>{p>b1BoEiEG!LtI*Xe`yMiqyQY^b>2Mu>Sx
zLMyBjL#NC88ZTK^FCS*^GG|DvC&gipd!bVX(6ZoTK(<d;0!OePuP88BFM;XlG2*0_
zETr}g-z)m=zSNGBFQs{Ut)zT@Y%5-h-^u)*_d+B1mNCAqR~0Mm_K6b@w+NJ<5+|m@
z+YVM$#gaO$6$Hm`h&IzP9^WR2VD_QRyMi!81@nDQh(%O;r&vz-)<+ZOhOAe+wjbz>
zWdQ66Uvi1!^xt~dAUq(a+`Luj7W`ZTDg`9v)2Oxm!>2nS1Q$S_<c8b+Q9DQpWu51E
zHQl+B^~?M7r5=DyO9k#3uFsd_>{hx{lvu|wrvaarD9*2wEKhfF@5H{=U}~=dANCQO
zoBT&N9N2`mQTel(z5pb079+2fjIWj7z&W}Ez&Hs3dzsaPk)o-#SJbyXPOcO#fU{wc
zqENyj>1TCdcC+!E$??RXgDOGQ63Bj+{o+@%tM;7aO{e(B6AxvLh^5j!9xEDIV-=Xk
z*#x%j7+LthoUCReINIB$UT@}vdBe0U+JsgY_?`~a-grJ8+US2uvaBudk_p|Kc}@<v
z^86f|L_I&zcOn7a(`!46GnqYIYVOguTDILs*mGH@9gsq`k2<;zLP$K<;GS~FhFPV*
zxXdWMMjI$Z{`q>bD2I5uxPqc_0>6BOnPNm1_SE87)F#8T7$W&ak@7f*vWm&;=t;--
zV}p_i+=Z7R-<eUdPIB(%DHyQbvZ7D?8-WRUa61U(6ni}1itzz1AT(DaSq^|(;e{UR
z$x%6mSYVi0Y+=@d_#Yl7t7|${1GX;8SPs{8?GK~2poOeHKHz@Mxe84onU@lZ@_OMl
z+(MH5u0wnZ^D%Y{)CV(Rcab%Z=5tuJdVF$4HaNx48QL<r0jiSNK=w+sp|g>y?e+ji
z3WoA~e%6@kO$!6)`?ihnPCcRUdl?J}Z$z${YW2MMM4b+$v1YVg4Vh%GHwUf@ce}|1
z$X0%|noT&t;!pFLffE=+>*W~;z1m!fy?(i^8X+Mh%dDrY!Ladfwd62-T}Qu#O}2Dz
z49Zd#ni}weKDDX+ge;Whb%7BDx=p1+xlJ%MaE}j*;KPap!}G(KHdU2vY6-`H$t*M-
z)?f!9tS&e<lrwCGIhR44k~tmNQ#dSdGs(SoRP;dGO-9}e0qYtFmHNV(oT6Dgh5#!G
z#ZY4uKuo}ec46S}wJ`1|z287jQMM3*ZBdU?$Xeo;Y5k5+FKG1KH68ajqb1D4<znMk
zrV3v$NMbV?!ItAhG*zw-2mz+h4u1*&#&+5|Yd5Y5j$UQ1p@%hhW3%&o0VP<(3E?dO
zAGLaF(MFyQT6Qc|bF&N^@3ZxJd(q43EQKlY)>_)(1ts`bxtTVM&KtBLSdujp2bc-E
z$`BZKl;(TI!u|}}8(~WqPm3I4#0z9RyRp@n?7$}b`veHSk$;|o5C*|{4^-BBGJIDb
z|8_vwN62nfSD9Z;%3;E7#trz9gE9suw1gU_nz~4^tM!=;hipM_#sgI0zsx$LEMxO)
z7=y&+tw2&__c#56cQ{p|)%dzuHr6li3Qiq`fE7dD^Po^L??)EJwEl>S<#152s^;E!
z0NRymZJd0v!G}Zx6r1JF^*Y91V5DQ}0wnfYG+*SXP+I!GI4s<RF_(#LO<q8|UbeZD
zZ$y?zkX7+E;@ox4#RnJ+0ZWg5rg4gkZ!1RjABYAguL?Y=DTAI5J+d-aj)skW4*LI+
z`SZ_y*n0^Soluw(FJQlrOV^4vVHG1)5ZJ4zYuvlXtE4pEJDmNFZqACi^fy{j4rOn?
zlzy|sRhr+Z!QhERd5yC@+cwOdM<;m~Iqt#%UTrRpMIc6YZ(U+>Q&!{GLZ0Nucun{-
z>U|8;AMnOloL}RLTxx~p4-dQtUT<tFl}oBRuA}OlECK6f1s&L~FvW3J&Gp(U2YuKg
zN`@&z#a!_$6MlI?N@uMvZRc(K+A|PPw#{DP+}~-5`dyYMeKZTHOP}u>IT%{W$O$)D
zrfH`#Dq1h0Q`fCqbs%Mgv@EjoDs`71sb!?C_p9*-xzxW>i}PqF3)zK=n0OBP&R1^_
zh3!QZ;JVxo*5hwj0({Ye$d5Hr^vw4{rrq?==DteVPC%tGWbv*{kF=D|k0sG|c^3|r
z2iD3#C_IU00^<&`CXYutm8ToM71%j8I&Z6ZEB;QTj1#(7$*nY8tx5ojfcnT^9nh*f
zc4s<D&oW)ezDVqJF)ZixO<ZGxAyKsUL2gPnZbzkj7arb<2GcH>c27R&bwdUc{>eZ7
zEUx}1?ZpeXoRJ(;zhApWF{UwY4F!1jTe|=~{@cSKc-ED*!yBig{EgAh=rBRIhtdYS
z`S;sK<+^3FKW%|_Iq_#~Q(LXxK?E~#TI#-TjOeM7rrRCgd=DNW^mr1X7AbA)u+%1M
zkmOO=vse5~mcP4VIk&>8VAKLVnb?HJlk`xuGiQ@YzX2D)3;yiIamnQ`uN)SPK|+oS
z)ZB8R*dK=HC714^%Q-)~j<=kj=+#AzDx_T8i(TnyGL`^FqZ0ly3UZN>7(y+?qFZP?
z{!k$CZF{qkGr>u7$`)<2v%65($13@7F|ySuP`w;`e?sCeCvZN<BbREwOte8c@yE5F
zLD2OfG@TBMuMOMWal5jqD{iyM`elChcQxJh1)^l$>?6@jACG<UTEp!S|48@j$xld`
zD!XET5c4+<8wH_?CsfLJ&)T}Ds{;AHaF?c>cXhy?z#x)@i~q0Cbdpr*y&>UAvaW8U
zQ$9t~23G#`i8LlEl0Y(tqo181;ICPWupshz4-fZNf<&wcA*C>W>nu~5fUw7qULs-d
z$JkVb7A%t)a!Bbc1HW-vbRKDZ=1M_Y-<N7UGEZ*2%yfk8OT{tMEgUoSWEb+%4w26-
zshdndH~uW<g8+QO!LK<|DT*bu4GvzP#`M?YC*O0rT0Tsdx84mQh$4k^X(^&q9r-VK
zO^H@bQ-Anb2kIJXeqV$7q5M@RZu+qzpagWn2hQ__wW`UzL6&AH#Z&T<Kx~#S{WS;v
zbE7MxWQqMEEfy7kmyp{m^Fzqf8QJ{J8g!r4wUjO7-2?g1+V^*C9?9G-`BXkmrgK>@
zR;JBfOR|jfg(CePbJhqfDLQ8>7zwr5l@AR1JYUbC$hh1~u6S1r?^5V(D3j*X3tp%Q
z9;+rf>na-z037hJ07l^$Swfkzpy6MiH^2y+Tddx~XU<%7G2%3g%}5d{KWS5kyMQ@K
z72CDsAw>F)IhEkoPc-o}*B`k7OMn1E!FDE(_ol;FodZ(&sr(Y%k8+n^YKVrM_4!Uj
zITlcwz9E|}OX|MKb|$Mo()NGDQ!xC~#%!``OA(hcnSI)iHcu#sC5R#9-NGk(&qy|T
z$<VB<E}8V+ACS=RZgIm%odW6(Ei74m*@ye!E57gLEJ$1YODg9)B%;wIYdXkY{%NoX
zt7~1qSzX}Sqq=m=!}9lEjOJ|X@4`R5VdUl$#vkcAy6{jInYN=`V`MGWGwM8`AiMnB
zQk(vYp&05~9v8xN+|>8PG_ij)bF}v4<{0#E<{4shH8U5@ncDHiyHxpPT#*W8Ud<i~
zJBRtMHiiHXS)+8ST#|kWI7idfodBkS0J57(A@Dd&$CeuW8(v5F5yp*m$C3!_(`MSa
z8~W9eXc};alE}2R9}S4KW}o>s)>Ns#6RIgql7f_vU^I{}VB=-E>2Aw4x*qQ9Ou9<B
zB>y$I)azkcAjj!O6OAWPUfHvn=b5%Vj_y&&>dv_BVXCnq>L+k{i$yE(#6B;YzS5nY
zWEiV9u2eYoD=lh6kTm;if?gJu=Tz*{01phSuch>h3G9;)v>xpyw*1Iil@o4Er;@H$
zpF@dU3IDfw;K;Wnx=)dz11++nJhZr5Y|wm^Gmg8dmah)At;g|-5&4;_pRfWoJKEbT
z2_z(b?^1xP2<)ml>$>(oUGNS8PmeJ{dUT(JVb4O*2q>j+r4-Kx&}UaaSZ&u>h7)@F
zJ#g|XG)6Au8i#XGeav~kXNI0JSDM<Z>A9-rg=;Ly5X5GQ%y0BeB+xqi;>eh|ozriF
zxt|t(se+VIQ&yZ4m6x(%>KpuOJShL}A!$RL*xCOrDpcZ2Fuk{0=cEMS4gXp^Q(dU+
z<<GgUDboq<H<rH!cVeL!SnZqa2aP~(gM(piWT8$y|FnEDu26e{@~MGEO#?dbho83?
zL``SKiHVuP+@?)Ls;S*~v6mH-1Im_Xab}Bvl2fwHBP!s{jn`h<q7E~dk2wapVe>R+
zcdlY&n&*|?KbK)ka0H$wi|-)~%Mh0EgJi(J54lp9x$r#8K&5rBp&7EVFdVc0I!UpG
zSlh?6#nt=B0UZrR*^9{f&q0X-(pvSEK0Rve-wpq?*RCM4(3x@n_oV}{j^SYUcXFaQ
z-b8g&R#6_E@c;KBy@)s5&^8W0+d)2&|H$%>NXV#x?Ir$HWZoxnVlFRX9!}0x>Q8f4
zEfGjVFoQ$$7dTswrzI$;Sj^1-5vLb1^Bf`&$fd@Jnf!knS)n~8av_V?YZ<nRa~4VZ
zZ8e9hqt}LbK8N8{m>$#rddl2;EW;q33zI26_WyoLCB(+FHF~$sKiJ|o;@@_HNB$Vl
z<KReMo$SC3Vd`rAN9bHAOvtZ4sIe3L`F|AqKl=TTn(zPVbzwBg$M#?M_#cJ--vcfb
zL?sLX-w!|kw<Q1f^uM$ZQ5@MniW92jqy3Kz|L?v2^(_$kYXF)ZoYI)?xBp+SO>s!C
zt}w2n!T%k~{~n}&4>;660}e%>7Bvd#g+%E<pRk!5Jb$qNzh?CBduB@pxUOVNrp{Zy
zGQuuz4@dtNfn7DgbGGl+BgM{Xw6qkeQ;mL|0vJz5vyIIb9*E-D`ML*t9JkR92BW~w
zEcswSbp8^N>NBmt22N0}<@*$>|C;+SB!Yp%@pj}s6oEy_$aoN;aaAO-HYk)JvQbkf
z-{q_%_&AR+c%Lk2ol@YkOX~P8ejin@EPm!5;xn&*AKmIXtionED2;)7k$r7e=ewt8
z3`;^Mw0wzz7N|-)h~qwu%|R~AjBmF>fv@M@hTrQM4RCe^_OO@>zO;)j79Fe{dK$R0
z1I69{R-9vQJ&TOkwkx7ObS{I}#T-Ai)R0wC4+}|o<jISGtI~EY$lanE&7nZRa)hBn
z?`QMp@w&uU`X91?rw60kDh4|!OTuGBdNiYn@g=r^ni6KlNYEu6%xn7Btq<e(*)7k*
zi|z<f)YG<D21j*dJJ8)AS6Hfd!a%`KGmod!RIHsq`_V!!_9P0SVnNGnp#Jl;bX$=y
z)Tx4G<QxV}$^QJj@Oe2wr1s3~?=hg+T1A}qZnp2<(>GvKi>Ph&TV079a@d=GGgY09
zme3GE$p;I4TA=ENjQNPwtpuz>aUnIW$;{%*I)<$<G3$fI3ND57-WBO+d;RrUr&@-F
zW}=16ozx12=(@@WGuPdc{q(sw&*RMn+$s_0xxEet#2e9d)#c58<t>~%G2pi>gab>q
z-k}DKwL{@N+Hz{|bXuK*m%t#<Q|j9d$;zHXxL@4YPsjE4f-e#oHR=Z0ns2ccI0xFp
z_uj=9lyHU^JeDyp032wD-`7*mI>x++Y<s_$+@)CFGz_3zZQl4CiA`B1y8tT3rEQP=
zrc5(?XVQbMdZA^000qzE2*eW?KUi!%%-97BWS@K>7B}093T%hK9TJ4=aKFo-5~|T~
zck{CB1wB|?_21j$Gn}g8Gbx_0Q!)<<T$b{-gJ_Kp<64U9R-U*j^g9OqS9TI=G^2ty
zAKk(3yH9;7QB`0{h8|CA_b7Y4yXG&i{kDtFJ<K|Dh<+YoYGN7ZzlP1Zc=|~U9dKg&
z%P<QfJ@AFBmY)^8C<JKyeyrCQ$Orsl@Ox8y%eE5bY`O5)%5*=-()i6W)h)&8&Uken
z9X;mEyJn!i1^pUy>PpJDlwm7yyZVkL{=QcA*+#(1-#2I3{WG-+-<^f%!!?gT>=W)e
z-Eqd0!oB=K&3nnip~b)n`&&`e9d)|G0(JTw2&antyQ7QL4}-7zH9^8gocBnx0g6;N
zIYZ@hJ$UaL?}16H8*7_b7<bp|h!MR`{In_*?UgTkpOl+wBp{@|gX_Tezr;EUI?tgG
zL(%Iue>@nmRB6W>qT@8(VnJfLafENF(a=(c*At3b9N4VACysHy&bLHNru~BN9h^J6
zy@G~c>Vs4@ZZ`8p4jzkpMCz{$80B2;S@60=Oe7x<s^zB`FaL}?J&ul})ps`~hz;Ho
zLr3z=3{*vQ5plXGG>lA%JB<efCNo08@ks8d)Hm<SwE3L;(01nadX5A*KmOK1Q`|7!
z2?bF_(tb%&+<9&wR3A`gf_;KM&#F1h2T-X8#%5zS`9nZK6X-?rO6DKkVtkBGoE}Yw
zHp!McP(A{?mHg-%ki}A7fuB6rh%9Jw;|5E+Qu&qr%zUDqoYgR(`>#U{SfRUOzn(|x
zFc-RU9~KwV7v$R6jE97FlUdvg<uvn`H4InLM5g)+m6_=o4cy;?Ri8@Z>rXDFZg953
zzUJ<fHL@k6*?sCgy%1F(YrtFgVcrQJaei~ErmYy^7BkfrtC;_7?O$9nFg=TTjMCci
zR4m@L#dA=OGkZrg)Jj?v%F79v<CI`^!TFdjE8T_;QDr!7r;Hfv*<g&V5%GYO#P*lh
zhfj!s_*(toj1=HtUCR#%FXcQ;@kpu2ITi&>oMAnh$bOr}^(%Zvz6vb)i$-w`3FC(l
zYGdP-A>(`ZH)T1U=WOpb8b-b<|1t$bX&^$$iBGxdFG5#8a~Wbr>T<rBFYRZL;?y}7
zJ!;AFB+=6@H+imCUDU2K8ldfVyh^-37Nhs;`QH<>nxNqc{As2P<b^dp9V<M%IBd-%
zpYMRG9AUAQxULoYNM=H6$yJgQepz`V$+fN<%@A_OKZM$q^u0*$bpqT`?&LZ2KhV`8
zQZTq9@2Z(+|GHMU{v4k)exY8$AUzH+GhjI{#f`CV(hm6HbAQ0D%zeJQ{~@$~0V>XI
zCx6sPBHwpOknJ|?xe^i>TVXDS>T;XJAo)}czlrhWn#D^+@2(#8BtBB<#_`fGJM<Ta
zCNf?&)Vm<-ml+R&;fg%LssM+`_VW%w((;bFfDH%=IZ7%#N<`-~O?22Df^EfNuW~M0
zDT!O&P*jtETX@h#V=wkJ%2114Ti;I-se4;IoqI9f(x9lG8~LkA77NzFW~3{}2!|(4
z#_v69^pPCXtv7ZCINLSd1TmPJA(r<sz#b;%_oESUmL#07{`1}R$_)gj@?r6u0e#`J
zLk}#9n3{UCFt*W+uIgT$_*n+fk43<-AH`6z7$C3-lpYPWsKCsengPq{*G2BEQR$w1
zW5<<a>?`+O3}aw3<u1xr25qtFu^4;&i)*K|^uotdoLhU+Ah+~k@;a8s7iq^EO@Tr3
z&=(9g_f1pf%&~<&xK-I_|L~f1)~zRYw<+SQ?S*MWSALORV5jUz)GWcVt?<Bo?fS%R
z(4_?B*J{bi>tjJFXV?8?si|2*q;;M^!9i-b)^ykD6q*Qfp9M3fJ^y^;>ZU-x5cB{{
zO(TINaM9}ugmHPd`1+mV?yM9i>}}bu>1rp%M_<=}lHkm(Qo@}q@Zz-GExbI1BzJSi
zbrzEi+&^nP$AvQ#GPG&7{Ur>&FJ42_8>-FORg$J8Ag;W;JTOdvqRLnr<>{#=P;eII
zcV4chS*Pl0=7nx6&ud9*^~3${kX{q?0ekXNJOKXDqp}-*|3|U8a`QXW=$&#}H`KoW
zYhXI(n>5jQ3d3-eGG*r7oZoS7sR+6gug;d~5Jm}32z=n^7%%^ypE(w&Sxm%lonk)V
z$spsAz#p^dtq|bdA%m>X#nBkEE$)rbH5ZsHJ&p^fI$^kI1OzeJQ2a&n_94tVg8@u6
zOF_&q?ELdJLEMazvBXT@g|9W0A8nclpSJAroT1UZu{<s9V=EVtz3D&ad@I~YB3aPh
z2uv9go`KQfcojHxrZ$lc_POTSO$igGIb?pRCN%raSM4GuybO{!VbTz6u?u<%l|AZb
zZMR!;6;rz?$40;)3#t9iBD3_U+3nU;5%cs1zsdeqmiETGw1mn<TQ&?XtMyIZ(xP0o
z-X;FZxF43#dBp@)Oq*0MnF<GppFFdv?-HW5<f}5VdK0SUyD=g`9Zi{eb@avmsVTbP
zMo~ckAWvKG$xTN@E}^j7>O~%1VdWj@cf9L}Gv@_$skSFun>9od1Km+)%u4~PcQiBZ
zJRRU%fA~XzIjuiZhXVJ3ixR2Q#8dc0a~UYB(Mi^5<ox|y&$)$-Sw4zc{+QFI+d$C0
z1nrRo!{A8fVqCpuZR^#pe@d=`S~#&g*BcN-yAZnMmw0=7uPh{zc^SS&R4^|)>4KVK
z<2~QAU0Zh%`ncqEs$In9vHN-DMzo&fipT!3+uo1f!0UFaRFK+Tgm;6ds;*&iR3paV
z1^uztZ5FKGm1)4fkh+;v;m06PGNi))N91UV@F=3kbV&gJt@zHx{V)SoL~5P}6c@tl
z@@*{Efy~Y>JMX6AnPY^;;AmW2^&qc0D3k$1F&ALr{2I4$<rC_^H}AK<^cLp)CH>Ni
zX~BKBXMwHFX4L`1f5zvE%&coKX*Vv4Oi@Uc$Udd7Hm@By?omQ7k~HSXhMt{AL5SGu
zZ_v@hy1ZbOx^{z}ysV^^{~-bS+C*Kz!TMy4O}trJQExFP*>X7(#vLSOc<=V2WfZNU
zh7CNMwnvr>;@9{5eZ`N|CK%SIm&x`cnfxIkq4NXJa>2|wf*gMwz?xK$dm-LdRVZy!
zkHp%lMVx<=x(KK2&MIFM0`+1@a#ig<AE?*_N%7uii_Xy_zbz)XlRw%iS90Ah=V7|?
zY2r~Ouen4A(EhP%0gzQ<(D7eBY5C4*^((MGqcswUv=xI9e~#$HS-8k*;g=@UsPLb>
z?!#rgK5~_Gb9rCYhe(U16v+}>d7iwAb@@Z}?;h8$jN?5CTsainoZh<&eauyqIL|RY
z{e}d1D0P}sl-!OCIb4;`4zkBhjswxY>K-H0R|a$zHrAdBjLh{Lc1#e3=c#o098iM)
z6y3KAa40b|s3RPV(^2PiIS0$~5p&qaW(qjh13z->^7o$jk>F+oQr8z24x&r-v~+<v
z_=v%Uv<4w7JmGG&#UG+ScQrUxFaq{5IEuN^q%CBu(j*Axir<E(yjBHLl_Ik6mr7xj
zX*;pUH9m@Ou%rt8?ZY{(E#5HXgweF|io_qzV)IWMa&Me+>~)gk&>afwxW4b9x)q+y
z!vT{c-o|w1p02Y$JffPqPm^CndItsG<;yVW3C_Q63VZ6v@cU*ReF?Ql+BodTOnNTx
z_B;EDLjQ6*zi!rhZ;>Qlw|1UplJbL-v!D*)^j9-adGJzoH_W&JB(GBGS=lugUNaBX
zj+LSa8T|X?epG5v@N>#!&dxXi8e)5$%0UVMFX{WZs@60$b<n;k;Z2#rz;H)ZASqgy
z7PA`sUTULiTkuJ;@kbUhKLh3?>w5O+OjVb#j-R%858HPu*G3pkQu|;D6kXT+C6j?<
zaxIY9*ta3~QQ0Ch98hz-9Y@gqL=*f0e&XtonF**i@t%n?R*~eIL5C{)nrhcj^qEAl
zww1*Ylc&cfTHSmw0*q&@F8PG-13&#!+x~X~|CZJR%$@TQivtn7aLdCxF<mijzp-v2
z({j)BQuMVuo$!cSZnuU!iACW?wZAk98ac}0bQ8dw$od3UCfl?VD?FU#)gdRaH6`;P
zUh}RHb-S5y;HbP??`6YHX%n7bN?XQj>(k&|7YqHd3hRduFx^HoJS5f37YB@bDh0$Q
zMf}&l;A9tXhJmhTyM>zTP;~vyTi8B@WI#ThkNi$m58b!K`IM%+1&sy^xX9DWO*s7K
zaQ-qhqn`|TZW10>blY#};(C9QawD$+WU~5xS7@~8U5IH)FrWH!FTU06bMgvquE_zF
zPqTW3*neyjE)*PoSDD#TLI)|C;QbzO{vN93_3=PTwgqsU)P7}2e+T$N+2r@=9Od5T
zq$tAs8j<7|-2L~rR-vZMaXnhaymu76&G}Frc{5kl)WTrZXVzHejODRyVfYr3fcrx^
z65uZ1Wq@j-z!K?mi3h*36T2`0--^boYQ@&&u``7t9>zAF8S_xB6ir}OIxo}t^0wO#
zHIr!qG0;f}nynNAS8Tu$G3ubYxS)Mt%RpWhCw)qiDbAUib=TjOvIleOp_AI^tq5#f
zNvqsQr-#^$JV+GbUhvMB76AdqM92ftwO{8JB~tS7&0HX{=*2Afq2Jbuh2fcUz0+k&
zo!bB{t9uSK2C3I|*SGS|X`<_ujipH+5W0o@?^%|u=(0o2IO;>2iE@2G(LHrN&+cl(
zengeomC9PKqfPTu4RdL54T>k^gP*dse&)9LHgkOCV=?)r;w|0^?~{tCtRe@ZqbK|6
z5b|hxul6%33LanEOl<uJM^Qa99giJ~?S%mGm#Hp}I!sQTr42AclrV<fS6k9mJOwSg
zc~Mf^8eT!O@)Ta`6xL&<Y>o!Q=B0{Dw^4PoA-6VP;Gau{b)oeIC??XaR@s``32J69
z=fk5J*!m@-U?kYpoV*D>1R6gZ6VB1}A?lMp#@j*_OV=JvGm-(;^T;f;(*#!udwerO
zv2Gi1p6pJQDVe@SweQ$r!RkaVnpFXrov*}j2l7X-=!oxWy<UCRwkis|S4bF=cSu^A
z)JHzDg(b<Pp-)lx?5Xn#`KRB?&ydM)Se4lkXn){M&*Gs)mNpB$uMOkXRoMJg&8f5o
zsiL3PCWQcX))6|Uxb%h<T5NNeq~ufgJGxd}$}<q#ods7v>WdQ1<kRcSepp%!vu^})
zRc1@%BW$`Y3fmP`DLS0~Y2M<SLzbc?W<yQZyDFbbD8o;snzWgSGoml64*QdEs>4>5
z+M~NKv6lUAmhw@d-c@-<w`=N+`!e%+{VAQHQ1u;IsuLbqUk&TzP4R?yXc0p>TBx3J
zSZ79J#L-4QCxS)nG#lYm#z7^5#V2E_dCLuTms-V_%{O|=f?(^Z`IfGOexzYk;?Bt3
znzebGp?jjx3>aQIHQjSmEpSTCKuc>J*2tVoYgyI*&Rd~&EzGr8B<|9}i=3?8<`Ud=
zC4hx&1fSnL2Mf*XGm;$=y8rupfKFnl7b&jaOD2h1=cAJScxm8OWo#y@sZBt8+!&ul
zp%7BEY|>f4m0kX|Y?$s3_6e15jM=mwJH?TZ_N>F5_QJHjz@#Ujuea4Y?&>z2Q445^
zAL44CG1{8Q!23@Ju;l8q_LX{+_9fAhXYhlHyCk?5|Bm6~*bPR>X+}1wIeDoeyZ<It
z9`697x|rbxxp5`f5Ury+sr=%v5ERB?5g@{SV5(7y<N-HMidXj!h~sb01|bMZ9g@Ur
zWw!eEZ67kOP=|oqia=PNb7qC1ytpVr8+z`3<7BL9G(xPh?Tl7K!t$XOmMWfKiZZhU
zAR+!SI>JG7)Fofh;WbXl*1XmXA)T4<Z5!{rdw*n(23?o^R^_9y85t|Y8o;+NriL>P
zoaymOW+UzQ#st5a34bS{e#7|T8&C;Waa!+rSN~L?XRAmn5=25(LQ^vRwC*KOkj$O~
zn$(&@1oM-DluFVfzAgnreTpP*ia$I!Z}CrC9QnF5XO=527S<I%LUptn$1~R6E}5dN
zH>A%00o)LFSNNonq#bQt4bCqrQ(o*kluK5{UcmmAX!5qqZ~NmGYsFPNT{iYUxaU<v
zVkScF6fRrq{hKA?9K-APRMjkMYFZ?ami=O3hE-FbHR`S=w%;^7&tc#wVk4<tqBlMF
z57`*I9Jj8{H=wvpk)9@cnd|0N|7?Rxa2=V*M|ObwXWV|o#!OhkQiaD$ubDsN7tU&^
znU}*WH344M6>kxfs*9DFh&O8V2Tuuxuo|w|>nkJTOzb^h+Y(!T_uLbsn&v!r>q!=X
zMLp03I&vE`vH4TyIsZb9Z~?RXwm7y?N(m-t>B;Er6@}4f4?7a*Ja3wa+ss}IoUYE=
zz+S((T*}l_o9z^q->dTR5a51dccbm=ZCFczEbxjwVjI>M5{%IsRfjNuYd>h3%!Afd
zwXNkLWshYtO@QP9rn&ZeQ{p3uFkNk@*vU9vCB0h9q;;)2l5zB)=sQkgP!&`C!Ry3X
zF<l2^r%B}Qy}}M@nmt7yujBKtzR9IrE2yaE5o&ci3|Q3jxDfN3k)rJMXbIthlV>6?
zb+XZxO&K1T*~@a4F%OQAK6^T5wZ(R0EDWfPweLpehxSsj8Ls&*%!%;RU$-Qr6$OI}
zW)mCe1&Al<g%VKSPE=SJDp)v9;d-yrkZATtTxCEAp>Tmug*s&0VBoW6b&`(T`JWj1
z$vNx<if5hJmeD(moA{%au9F3=>AVA1pVhJ}BA^*Or0^!+Cu+T+=fhKC<CZn@oxcsA
zg0EqWnv}oHxK)d~%jbZIuHr>Ge|V$BP5f5e#~_UrEmb<6-uiL41`92(2=Hj*!(>B^
zt(dR%jf%JNU=l&uqDj4%1)PnuqlgWX3ryn1O7$S)2riOBhGcaANj<$?G((xY{{{QV
zOe8#YlaRdeD^A`gf-J5SepxvM?A5wWGJ)SSn~`kgjUrNJK|21`UHZjc%|0QZ)3})F
zn-YdFH7y&kQ0OYwu-A(VR)Y_%*r({P1)L$8Dh)F;0M|b`XZgn4!R|y}xK|%;i#vVg
z1kJq&)>kS}ojW8C2%6MzP8yR0#1|V?IlWVOAG8}DpH$$)(-*Y9dg`D#=z<u6dJsAS
zy!sdmcJsh@`zD3t`u%6?4N>OA2Pjp~thO60C_VS%L86{Dvq(V<DIaU|=`N58pa~M(
z+DRjm)i)H=o3u2ZCODsyAo#oI_eP%n!zOA6QZ<&r-iEJs7H#=-JO8#8X7;V$!MlS+
zwxfUbrJsZxGZ-cth0^z9?Ar&ge=R!CbQxr*dK#|R%62`+M^6_C)15;MWZucNbxQUh
z*K!jlHIE?$6Z2=^mzP+cZ`G`XY{~8GyGxP}`mXY8hX2Mgkz~6SC7|-Q6GA5Ba8S#s
zBfq>HanuHxqh>N{!Td<8eZ3MbAJ~j|OlILnpd?SCeK7=+u4-ky)A<F>Jw^LXVQznM
z$a}a9-0=V!eS750Sg&+G`&!$e<^%2M0gzo`p=ZidGadk2(yt(Q4?T^c7)P|Rt!BZ`
z`d$&Yil3}58O)g3h1npJ&%K`2z7%iIJw-1(4+hmse-xCBApagt_J>QfR%Ip+!%nCM
zZ@~CeQ#bH?5fh^W&HR$#Hlm4;UwSg*R)o?e3FZv@3TL2`=_#p|){!s@nnLgV12a9m
zc0a;G#wpaT!o~$(YeSqSwen+#o13V2pv>}x#AGrFj}0}M-J*)pjl=kpGiym_)ul^Q
z9O@g{8T2V!LtJFo_*z0-8Ds7;P?B6NylUn0mgFEi>ww$5x;}Mod?$<j*N;Z%%^qLY
zE-F+ECFyw#{Uo9xy(}kliv(Oyt>(t&5uhsQbEy>hfp5F+@{~wkhWRD^;L7t_i9C4J
zr|<)CKr`85C+iE9<Sq%*4#X=o^u;uL!bPy)^P%Psc(kTP`$gh=sk<n5qqk#Fi&(-7
z`H(B1^MkdPY}Fno55RF!QFyWN#FS&tkXgfNqUo%$Q?yg8Cv7!2qS*36W&7YJ-v=5U
zy+dC>N%kP-Ib_9~&Un|;&VCpilC*?0m;KW=o+VKP2?2e5TOhEff4j`OP^2HQaql~D
zI?YIw%*RBCLEQJ$KgLnik|yXPPwTqdr<zRQ@#xlO0l@C4LEUZ9^0f}Ge}XFl(S|D!
zlVe1w2v$|&loSaG@Cf(l*tK5#0Cs4{mL-yEW9g@;JDTVs`9I(4rSd09{nW1XDjClN
zo$C0Bi8J$CGrLOenHK2y@_=FTHjX*53VtH9(UYtr*DX>ew|T~SC#*)kN(suQ$C~Os
zxE86n8ZVc=T6BLjsvJ>FNI&>zsR2ujCn@>KWN;*}kIEw*NPi&d5Lcs_AwxWWe8KV&
zb>g-#g+=xuz)PThNZXF@Be$BJeHbLS=`1(vtq;r%B<iN_8Jwb%Rr6DpB-wwoP9{<?
zouHD}Tb%Sc-nQ7~3lsA?8qBkzKj`}_%>Lkw8v^kFnfd?n0ODg)>+=bEVG0KHKy>f>
z=ygF3+ZXqwq2C$`HPR~p1MAxRUDfY0o|D;)qBE?4m??M7nk_#7kZ?lb)dWKCv;3{o
z^HZ3O3FRElVl5VW7<$`B_&A;b1idaGt_+fGrNm<2w$sX)wGD)1efEL!!4pVJZEeJd
z%&f{~n_g8y5Mm!6y+(j$=g~To?8M5OAa0?S*3cwnK)M6VZMHrvsw6bNs3AYx-+}Lg
zz~1}{6E_|ExsS_`%`<HI!<f85{8ez}Ov{Qx3-)J_Xv$9at+9m5zK4<li8Ku1xg#)+
zNx-^mQzK872zMq=nT2E+c`m|<$(tvWqS}=0a0R-47SPaoCN$YVH(fuK3l4sOFnKsR
zgl5+DLO>^j%?evbG`ZvucOp_EB~u7ovez=|T!KZq!0Ch10e7L1BZk;VZrasMSMq^2
zR+SaWB*{|^jtwp$N;uKM!;CFIqg4bg8f*e^Hq<}v;I;f-&SGPXxh8!%Bn=HVDq-GW
zj5BN2s%?PvS?EJjyZG*}SZd$I0eGcR(kG0U&`aLEQIfFrTH!>h(173oy7$%pqDncl
z{qp8!S>R1dyC?!T_gjw^xsrpNC?_=&wNPlKoIZGAd0PXqiL@d5&dPCnMagV+lypv$
z7S%IRe0L})EB$2Y)$M6P1otfaK?7EbA{|#ihFb+gjhH<R0htvikQi+<AvRN-vRbzY
zEg9`3@7Ju;OKHg-J1oegCew>`DM7(5`SL@*uInzoq6hje#iP#*6}h%cjc&JFiz9~)
zHv7@q=!;mQD$s@BWD1>oQJ7ouLXlREx08*M7G3>i0czym=2rwUKT>{ce{%*SPK+1x
zr?O6=M-OO$#tQ1JY>K%R`vG-)`(Dy5Ue1YBktGh88uP)fhDL=>U8@pVLLZ^g-ubU|
zckZ_Gm!|rk#pP+`8CcczJ{weYDJ<{+zYBg6d~W<LD7YC|{cR3|zUItlFNTzs*;agK
zEh{Co)ZS7=r{=Z+jeVo$>OWWj6(w4>=ar5XLc<9*U^>L8$@wYw)p~Bs;u{{p_f%b0
z|1_W3BdYpbK}-IdZYfj2^BP!hbTT+fBgOYG=VbcRrki(>4FqM(HNV)L739!&MZX7~
z%U96|%u=d*+EPVDP+mjcx>m-k@62x#vu@<S5WE1A;@32VDJ~2eHMp&8>CoYkRTpG`
z(fs}!iRqg&?bFO-(uS1H?|-_pH;djGe}&3(ev~m5?G<n9drg@YtTjz<s|yy`s^!G(
zTCUR<@|I5!o?L3kJOdrgMo9M%RM^L;Ke?FFvk1Bn&uON9VlED%VvXcy{oI=4W*Sl=
zPXd-d!`rb>ILc3&Hpon(No*c`qU{&S_WN}inxpUy$k-T7dk{-aE5pCCFK+DWMoc>h
z-!k9olt7;@gb7;{OCX;`ai-H~Y53=_cAn_Ot22v~Uh6H|y+wFMazVb*{qV?kU%%Hy
zkd>xmMZ<bw(Z$rHSN<=T`Mk@l_SKHDg5LNk?hK`Tk^&R{VpSb9s{1s}?^xVR2`Ll3
zm)}yoQ`Gn~Zj8GW^qK)2GvB(EBCFJC7dS>_zfalgzU+2j<Oz75f;jE^xG_LKdjri-
z_>H(QMck9`5+cPtbtccJywVA+AWkk?>Qdj9?CjtOy{)`*nXK^n0$`ayslJ&khR<sa
zYIGI}LKm7{J`mbAC3sfYTKAkuU2$3wQU$S_WGN3ZW*Pac(kT+s;<Hj3J6w}r9zcDt
ziqh%^FHL6b(G&aLUsDX5>-B+r4obK~jS^okefF5_Gh0*Be~KF&WTyN&wm*(;*LieX
z%}lx2Zk&c}eC~Z;>&)*RhQ0^Pt&-o?58Y{dE_+kJe{^}2|FYuwmDr0c68(Zpn@G_N
z27Ecuv_jj{$r<7=jv%)D-dSq2v{65uTu?mlK~1L~{u}K^NlS~mfGu@6LK$t4aH?uT
zZXZnb<pEd7ggWi`FpDTC?$q1zSiqc@OVKFGXWLG;SS8s1sNDGmnQLGE7JEa7z6Ug^
zr11H^;kM+;()~%Q5+_cXK|Gzw8kKAx{<NU$bHckc%-f_+`?^&@ntOJ41fiT#3dd`@
z|0|_GksRx=p!2zoJdu>o(J6!2alM-_)CNim*2bUx3YQI=Cq1PMnlWnc3ZtHHEHKd`
zn(_kE)dKf2b4a&}Y_9pbW<H;t;<r=eg{gELODulcI0y&<xn@<$h)^pMe#Ktqfj1m}
zl4aAl;3gv`b#Tvr8@e7lx*u<U{aer0Kq{vhJWJhyl?=n(tzzVDU5bubgZB)i@T1hG
zIX<tv2#M)%9Q!J(8l&rabYx=`;pj-Ag+PnJC#5wY6LzU<TNH7!7zlh26F@HxxmYl8
zmCvdvQRC1)Ow~VMM77dvPsW^BH9b$9!`o#>n9Ddu`Vq0nFN2ya%PHo&2&zfE2F%eC
zp%Xh}Fxo?}L)GB>!V@X_*Ryg3*9bso?3#GF);G_s!~v>#cu!<skLTnDe=40+)srgN
zsJZsEo|(7_J$d7mEqvAud}|*Us&9t}oE@gbv25C05%{gCYJ++>V_npwpSL5TDksl%
z3|GcNpNm&q_F6uUoEW0i);<0W$O&P%Y=}_o>wrmln`9ySStQ~->1JtRTe}Yr&dk}g
zfI-tv@@N!d9I{e+5YO<#U1b!ilJ!(J@@B5FsZGY?UQ}!K_-NH~36*}<8m<S=Qz0*^
zt+sj6_!HC|%AQaB82D0J2T6p{@ZB3rbx=7z2Z!mBSy&=^=fTgxFvgw;O3y`md7BnA
zWD)_wrG|KLOq8ECm=)_CtxAHvO_*E`dhbVZIs76??m<!v^;_sF<P7dV`!shEGF+4T
z-Zd%`mQnRWk_W13ryYz}kv3p@9O6`3B!_#lcCSV$-iRma#UOf<B(mGh_^CD5fpEIk
z6f8(H?b?GuTce!KHO41F5NZu(cD@VZ8uxCp;@J%iB3N(FJMiezl}!fEB`^y(B9Y9P
zole&ZZR^V{!{f!Ko{Lr3!>7q>gq?TWcgm7YYPEkDN2F!$SJ~}4nA-eCnB&FQ*vD>>
zDDLMP;<<O9+niPyT|%q+xyo`Md_5Yc!s2ICb~VIbgn96Vamz+@y}8oCh=0=XrQk*(
zlseKXt6WC71I(<c%P1Q%<gZ3Mt|S5POo*Q1)6__o)Um{CHIo1-u8jvim+~uLTl>90
z`(hIYRb3Rmu9$@h1FWvA&Wqlf6_9<iHhcP8(qW(Be;i4ZG^;MSD-RoO>1%2m0i_>&
z+eOJIEW8^`;s3a*A&?*lTAobJuv~n29cdK$Ufnn^<2(rWPQ#+bP+<T=gs{v7Z6MeI
z*3Y;QN(g9U{}O}XoB2b`eio`z8OgJT&h<w6`X|iD8txuiM$^Jv-tY5*ulJz{8wi3i
zSq;PSfWHn;^x#IhTKl3QBv*3H6zVBhEw{fsTCVUQmz4=ex2&Ce&s>heTj7O0ljHh4
zWP;C_efplPQ@TyEWN<_=PnC1df82b3VX7*Y`;W`estf_NhmXF{s>Y(*Cu8vH!wX5z
zZ{W50+)6?6%-)tS34R4^&gKV<-|^Hw7_BvgO)WnS37Fh`)l853v)6QTmTutlM22wB
z`z)OTHFF_)PeVQS>z7AL>TW^rRP*nHqUUNb6J=nB;s&?K5%<?)v!yPwdLus(<;t5z
zwJtg+evFIDiLYL}TF)5D1xb-vYfVObJbJE$M^ZUK+{Pfkj&U>7+^dCG6RwNa@QKi@
z$S_Uc!?&2Qd6JiTR^{^(CoWg5ugeeg)jyu_xyeGEJ;wr_>RR{DW*#bhT*zDkiLXS-
z)v9_6W<(uT_5D*zW^2h3+%~q0$IaW*;V-eGPSC59rV~rNEIcP*QH@V3TO-9GrjL$|
z9dg0Ow8}HM$CnCA=PJlem*saa#7@}wb3ovexRt}+xsA!!fl}{G%zPZ|)2{aRHQDdy
zehyxX+&n0EbB+F{bkX1EFHWZb3Vq!&--mfb?gw0m3@L1hdIJ;%H|gD~uk7ou+WPm1
zFQ>TQP|0}cW4F{(9#ieO+wxe%+2`VcfG_;Y_}OXgE=jV(r-ZHJ=#O(rSYBGUx_;Y=
zZOIdaiWekm!o|QDlGmF)pQaYWw7eA2+Y>tqr`(P26k5$Eo;7PZHeE%4?cxu#R*svZ
zOb1fLFQx*gPTz}<8{vg?1Tze`POGEN6d~IV3*IcHmDTnXK?a*W%+@*iyREBkcg7(Y
zqMbPE`o-3>g+=Eu3exB@H;OAy$O#JHq(}9Jk0@vdPF0_@V&(w%=8()m^Wl&!VoY9W
zGPaham@C9t8P>g^O@)pnmOo}`s{Tr)F&TSqJ?FtIPwEXDvJErVWzdWVO^Rj;{U1bQ
z9336<E=(45PwwJt`vZ+^AJe7wGg0+^#!>48G=FTgV3Rai{otS6htZ8-CyDMITJ@O>
zAu<$U{C;&<0>)*EA=kpu{5`xZs>O_Jh7ar7eQBkYtIEc5nXZz+&Y2{}K~WfZBhMsg
zZ00%KrSeCka?Cv`S<k+)l~uN-<wcpVXrB?|Vp6ARV~O$)4c5WtIVqBFr2Y5bJy$}B
zaPKZbX|&|a3og&&A!KH@^b_ZcEnX@1u02PL;`P_#Ge$xdU<qRSMryfcW7sX+>;!X`
z<fw~Go7taNB7@{wktNoX4M4S!A+wGt7V7(IM?q0_PlW$N+gnD((Jg(vK>{SWy99T4
zcL+APyA#~qLI@rlf(`B#TnBd#?(Xg`cXG~i-uImI-u0|?Kip5OW@@@?c31E2uKlZj
z0n!;VHLn6}rB&JWYPbm-P7+66?Ql~u?fU2@z7n|u<!7_o+_xYgC@dDabsof0udssd
z5XWelQfbsy?`@6Zo>Dpm9ltEl3s{Zm=LEy(x9<$ISZWNHO=woXZ+)2lQ2)`pb(PsW
zWryZqT1C{|l=43N(!yw$ca)9Yo<9*0lveDXU~p<H&u5cTHrWAk9<p90ilP-9LfD}&
z-Nx(%1VGSwP1CV6j?YxIk&Rl=MrYA&3n6i?P$C(Zz<to;+qOG*n1pp_!?KO(X9ubj
zo+`Df>a<}WTD7p|h<KLWIjGe<)44U7aiXocHNtRybH5wcbf4b^fiZkr=&1r<I@@F%
zA=M>0(LOOe{M{Ciyw~J<0j26MK&fB(;?{}LMaxDE{19~pb4;cR=Lo~zyBbf~IbK_o
z2h`OR7t%zm)2hN@`03#<wSuC-fNMN;s;@gbx27L$O*A>L^I*=@%*1n_F;s8KX419^
zy8U$H_sZhr&#1UQ6rL-~UQRIATs%Zo%n(Xby~?Kxu3df{rzE@lz3OJczoK_P*&N_a
zAx7zqgsey{o`{LVWacdhc#o{TLN`)kvL5JpK6;zvUM+k4L*cnnF+QW`oUqXb328(a
zB$hl1B|EeYH7k=M(}bg-uf~eCQ(cFw*X3)C!5lYfwuGc1ve3#aa?rz5P0;F9o~U4k
zZfm1h$+xI%f<I9%7<isi8Xa)g%!r>K8EB-m+aZG^+jd-;IL?@{6IXZt`iVP_R!1lC
z!!<z3l@z4wXzGK(^La^H5oFYEvD~jgm7Zw}4YN)N$M?ztxq-?}A+Dj3jM-LHBpoWU
zk*oa);40$Nh_6b{?6sXqXdwKoix!%m%k{PE-4KuquT?eU)jq(n`JMOkPLKCNk>`Vb
z9c~t={dZK<Oc(#$EgtL%f4yrN6JTlVT%S568R|rjSQOYzkW%0Qb43UFEVh$Iw<>)`
zLmH3y@{?#_A*aZRXsek0vXk$pztC%O^O>t>_9JHxBbYnbMl{WoT{u!#Ks9-NNrr@J
z&8couki%igx@#2vO@ztSL$DYgx+GNe>}*@<7^CbHkDMVrh@015IQ#?0<b~F=h`FJ9
zOhl4yIp(&=C&=;gAwtxfvTUVYA}q}mH&z3>xLs~hF=~K!!|xA>kG-4XvWE9#(p1bS
z!)#kMx}nCD#$B1Lx@%Wr&-#{6?GA-C`IV+#A@6crVl1}$ZVB8JM9h~H8h`oGW`^15
zKe=ix)O`=x)?W`SES|HhbkxW0RIktX(`CQ8|A({x!+i^cy~AJjLeo=7`wzH0YY&C{
zfzR<}eYjWsAvO_~z8%B$OfB?xFKB;d);WJG0r!6ae=rjd)8MD;DaZToaQyqIj^c)%
z<<2*1b(c4_$doo!FcA!hG&a!@se#4@{_AM_bbrwA%4z7S&i??!Q3xTd{j=eom?g#o
zP5$E2|M3R}7lN}<;(H|7zyI^U33g`4SMY8}iSb{%{$sEIZ3jlz5$$3>ZGpS_{=4u0
zwG|))=HajUzqkMWA;EZlkdEm9H{{>!>K)m7G<d+Dt2UVa>mdK>fUqg}CWVE|`s2Sn
z{!bU~!9&mGZ*vd-w}<@yIAmzpf3S73wNx6vTcZXXqkLCU1kj;K)x&BfSa!cQdhIWk
z9Ggstr-Rn|s3oX3Zm&MH(jdl9@B~7-H9*gJe33ZRvnDA*7VBA*-}g;bDZzv2^+Umr
zH_2u1e*TN|mbd8n=V=(0of@$hsJ$!Mr&8n7(S``I&3GbO?KQ$rL|&&KST<a?MPk&3
z+g+oI9?#K%POzX3w}*)&De>+V_h0GokR{3Ob*e9NBR^YYTKSy9=kKq)dy9;M$++N~
zjzJ^l>3ZVO+HSAna%~oc<#XR6ki52Lki!j@xuuuSqj)ykB_nHd>z<n>o|lHbd&;-Q
zYE}nl*Fy0rM^XI)n5{BY;5>8Gd~AC;S)PV7>$fiC@gR~gJJU(l{Iy2zGuJSwB=ed+
zcpk_`)Q5RfVk0rtsR}Ssjyv~}`h|Yt1j{9rYqFU{8pM3{GH7^xGVbfH#5dK`2HFCX
zm$(~!-`F;H{k3ReOt-Tbq2NDmzp85v3Q3ZXLKY({D+3Wwi?6Z>7+|C=(QnrRIva)x
zRNX@d3XNh^WeNo0G%C1=UUQ;Sy6hjP?7u~fWb?zhG&~Ol?-dlhU4J_m#BMo#PRMv$
z;HpTT`1-hAc{P){;6mG$JLg(?!)OP3uWYf!7PwP@vwsI5@<`dywO3R9>7!6T^Z*6B
zbfcx{R1-5y^qq@45*A{5Pub|22QX5^q<nc`MclBI&nOiH43X1kine>GXE_A}Yu4zK
z4~JX3%)?0YM)#CK!r4RN@kJpaQyp^82^ud9kSCun{;+fwHYx?Tyua=@1WT7FgQXf>
zgl7{o_9F0yY8(jmN6U?FFaXVS6+2bg*u`yklGg%pKXck5g}V>a%KC^D4D(D)<nxzn
zoTGL5p77{a1K29#Z-Olyj`+z75pXPx6komEawYDb<ekQ^ytjejQYW3y>{umDTv6?f
zQSE%%-gEm;OR(*R?s2`&?z{Rwh@w;C>i7&|m`Kh<T6)_pG}g?f^x<F{93z07NiWMh
z=WVP(@K(v-+6VOECZfGzYH$ygM2fw6e($?__Z;!lJtJ^{y|f-)3b&;iih#=Ys6+{a
zmI6e~s`;ndj~BE)?Ij>oEgi)l3JhrHb7?<P;)IsCCk<Ce5(PuOTC1PLF%ZYJJ)5?a
z`Pv0%U@4s2mCZAxt)7iQHTwz-izZ<H_(VP}9@*>66d<^MZrNA`08IOeFQSuv62OAU
zn8-p_&e#v$Gl+2wSwb7Cd!w52Y$dw3+$PMIrRotQCaCHAzLcqLIm|<Ze~avA6r}T3
zR46E#pUW(9;_}1h8C#b7`iJqH)H<oDOWUExggC4A^G!_aNhBM?v1SpPuydnrMR80t
zy|1|XH2WS3pLv36sgK(z7;NJhA2M+XKw(>04PAYCBJgbB3YDcsUKHdad>l|nO!A#F
z+5?fF#MSSKCOdOFc(xt*v`7k8xitAy8{3bu4SF()jxAt2EjYuJ3zKp3Cjx~}GQt$4
z{Kc6*J*oihAH696#Tvi)aP^eI#GsvkLZmIN2x;2W?RF9=xBuMNcudnxz7oN55B9@(
zs;b2(5pQF5VbL~TrMQ&z0xOCkF{<c``CpkAAHerQt>B%s<xw^vBV2!%UTS<jL;w#=
zgXpz`v#YYHa3&%b@2}_*5>xp!C}zk{C|Sb{#f~)M(|+iAPas<F`k18O4adwyDnP90
zShM<I#=S@RV76@VWKEJNGQ4=0i=C8s+O{{9lrCVs{KXi~9l^1gMT{XcXX9XdpiF)>
zUv<bDSztX$6*JDAj?G0KIy9CvAtWs$ODK9##OsmLVlkd#M1x9(Gi{2w?wco^i(=iE
z@c!>FL3;NS(EEuG*a02INJ?+Y8I~a=5}SG#E2Z{Wutu`=A|~b$dr2F8S9%o#61}-S
zY807Hpze^qn>DQDx2vtq$m|-4?o5qPK%1xHF}EV!B44vO_Osse42|*7eT2ZpF}5C+
z_)P_^<N3PM@5RKSRwB|rNO0R_U1@I-s@QSXt$e>u&Si(vCzxWc-p!H^`8h?UV$&d1
zibLeO&<e(Z_IpvsZmvpfY|7Rux2No&oL7%Dl<sixrwUKwL*zTHZzDQ2dU8#sq4-TI
zf1&$_d=Ht=8I7R*k5v^<c|Yq8<g2_6`z7?$>}uUxofTUfbCucK-m`bwOJ7JUJ^sP#
z8Q;0nnL@s?-rQYw!u&>}XM<)Ye%mSSohfJqxP}!#J!JuvxW=CqKCnFJpeZ>GIvhz1
zA3gQh>sO$cckoH{lJ)dvGN7XO8SL`Hb@r`Qyvb=xkoBl(YbHh=EP7dH7I_t;?knuM
z9_(M_T%U!6_Yq>|_((-@&;Di{dwTk?S#Q6_*k=&4AvS5JYbDAf=-TVS)-KxNW{G{%
zpXG-!+Ra$YeE7KaY5k4YaX<P5YaFS4`)c#ix{V}Uja!5AJVw_0_tXG?VX7bFcYU|`
z8|Xf2aG8+rbq6f*V|Uvrx1(IGAGfBiR~T9Q*P)6o1DB0Laiszx<6#~N`g4HZ43g4~
zQ!^-(G{zPC2UdJV3N<7B=>pU{-c=s*L!{}jN3I_b<Ar7?=cviyapTb#QvP*f&qf1p
zD9(HGSiRxu-F5kGGRN~#woqHhsNGWUh?C`OmzbjBzCPb_)N%9M`_g0O8_>#M-Gz`m
ziR==Jdm=LUY`@12S5A1?<6|`teJU!Mn+o{$yiH3X+!8I?LPtBTrQ!=Jf@+#}f%KNU
z$Ha^nmP`(9NB2k3X&^*3xL@UG*$J;jkcAC>6N;_g3v;6f#qIR3e!+Ab|KU4@O%?iK
zOSs@^u<?Fkfkr&;f((XyQfMKC3<h;Ef{;lHcg-d1<90fRM3(jO!zB4dP;@gi{RIX<
zb)*gn+tJDe^+_g;vmJlE#>BopJO^GfEVjOF23)q^F1%d*0DwfW%duTzP!Ou<if)vS
zq#vV}Dqaga8gJ9JNhJR8^B&XR=li!1AN)BV6<xo6Hc0+;@tqb|<E{q*GA14UN^B<L
zZ}oorn&0iG9xVqgUTJAF%dy1Xp305KT$7(ZK~&SEM)tco?L*h1EE2IkLYRMbsx2wO
zGX;cDxJ&A5l2ZC1no|F!L661!sU^P168>^z8T`6qR&{1bwr_uc72Rw(IFI&e?K$dx
zCp$W<3BH|9l45$C#|(yK@lDKU=lcT|jmcgaDzd^0!oCcrnIhw-u@;wBobqe#r}*w%
z;uez{kGqTFocD{rhZSL_;ai%=Fg4_xv?5_at7pj(>9XJ5gu<P?iSy8Rl5%*G;#4Xs
zRf!;wOR8uP*?h<&(Iq{f<}1)c#CIyb;N0ae9i787VQK)FkI0Dx2rfZ*eF%qh%E-1X
z-7N!b5@d$U-od$CB@U;h-C{&Mdr149T~#_}YULcOA{m>T<3F&`1)nskvhLf%FjxFB
z!3AgSKT%1n`DrdAT1h5^j=##8wV^1=A_ef+BFAKTTMrf)6{vdpLH81ZS?-h@atauY
zFLh;;Os`-DTfp&!^JyB{>DT19oexYz3J#gbtWWDwHLt6U7Zh4Vfg8>_gtyuA%x~Ij
zkNEtKu{MXDW<R9-UQE_nJKxb+*S<{eBXIP>S+wh{mL^}xe#e{2O{0=s|IU$5%=W#T
zy%-yPgZ+--e5Kl|+jlCfcH^bPnK<Dj1w$_Fq~!dOzAe5Od9)*qmJtNxaf{_QUqoi<
zzulbJXrtT49W648rr2Zs#noOtcni$EDL*BiKfF0~nA7|LC^pHyC%iMbyiz{3yRpxd
ze%tPCPQTsogwuJ|t|?o&hax-EVMj^Xp*ms#<mz$=V`R8d#ggC_9**V&+nk^F{BXx^
zIDW_z(DCl{s(`E7vl@fjCE3#PMM?^r70Ra>OWAWuxLDyKKteC69BG1NHdY;vPFQ9p
z15I`vrR?Q#fFVXxYx?GqSUzNp1t3MO?a|#d9RNU69mz!?{QW^oHgL^87N6Cd{D?iB
z=hR4tN(66n=%XAN>j}=_WB&FjY$5(mlfs?J&B7gzdNGVaATcCo<=j=TMUCI?Ip%jB
z6BjQbbAi`5TYJ30LHRKXiI&*BbOXtmFsH$yn%{fpj1$3$ryE?Hy)3;fb2#ij&-%u+
z$#m+@3#x)ZXEoUjFbCoa4}?UlS3l7w+q$ckGBnI>*^W2N$z}`fJw97yRsf&T!FevV
zBt-gH3JMA*W3A%SjpT$Rt+R`?M|{2`D-GC68N)*dMJ-Kc|BB(z+Pqx=ZMg3YBC`rM
zLwt0ZkN8+7&w|AG9@bg)*19|!o|mvNnXOlU=s4F1kZDg2-`YPH23<pxM<8#z#&8=C
z&NDel9vTmZ=fU)s>dy^V_K7CD_J(7Z&om5TmW!Q;8Nm4vtC63`_Y9!9;3r&*wOm(a
z9?V7b;)gk*+TkeA=oCTXL%XKd@*W-u6u#0_Hw&hu)4SWwJLS?<7ho<<x|&;>F0*$@
zMq9=DsXqHk7x6^~oU8DbIk9-{P(DTXF}L)VZj=TK2wYjd(1%*`?B1?t25LCOS^Nx!
zz=;JT(We)9+4-K0rx5vB-CrS-tKXV>-{yimD^7aj6?Y}<sWtsP_nv&GtNUY}HD`DC
zmQUV$`tS@j<8ImLi~BVWOI3~knL%ss5j{@aVxotQ#=CbSu;%%Htd)Jn49aeJIx%Q6
z{aBy!URGn!?8>6ik_lKdDzUqoSTlJteF#ACLH0CmV`E<}465EWcpFH%p|(Hh5yvrO
z*I29~Nsx*eQ%;bbm+B;P@*>9kj)l$riD)xHpuIMzs%|7@Lby#-zrXD>!<>4nrqyC8
z9W-aBh*SJU6az=tHK58vf$cGPCX5{Ca5p)s>QNn(VQ8QR<FB5X*wiHhqXl@}HRN}G
zIUE3HwZ#_rnGWCCrfl*$5)2h4X+`O*&%#<gr=Qh@O$SmGVbjC3=`o(oZKoS(ImKNt
zpX+njrdO?&amK~QCXG;deH5IF4-4b2hGZEIyG5994m6z{;kopSi5*T}WwL*ITyXO>
z_N+d#*s%U(G>k0$CfWx%(Vy??JaqM^P++XML4gq|TDdhuE^TX4#IaSxb;0F%c}JJE
zmM>p=^T#dj=h^rOPRssl63iaVe#m`-tVpZIT7rIg^!8JN#;>kyTF%S!0{*QL4R*|t
z{^6A6E>Sx2UX3cVNKIIa#x%IdPY2=^tP&ZT!U$vB!}5Ilrd~Rotasn^HwrQ`<NlP-
zf6mfmw&Kn}TIu1GySNG5*8{xG-b#yYeZwP3;cUu(qB~eqVK7V|B%lrWTBv-LRBbuc
z+B2!I_JeEb^r_7VzK&2JXTDo9*VKFgb@@|F#OEj3J3X^x+x_cAI{>HcEp$@ZcM@vR
zn0nuez^Ue!)9#PA6>Z%G@@MDix&*-7@UeJ~FZNi+&y)T3<6gY;S5z{D?C8OM^yvy#
z;{g!n))voE@^>G9hDp5kcH`)m!SiMh(u8D^o+d1eGHISVcf`nBY3jzQEB;pe(mqa(
z(gIQxU2oICpl29wyTqwGcHBvi3)0-9bK`6@6ds4XAvDvBw^pCUWPgOhEgB=OUie%j
zRzXWwf4^F$>o$-NML0fHJnS}zk&P_iS#-J28_+e_LqVnHEd*<xrK5ZxY+IRU(N^{B
zLh?CIdMz%{(ansV=%i$fAr;(NO5*qQ?GV7$Gd2SYldg|Lbm2<AWpN7rXz`ZhS2!@P
z)}Ol)Hj-*^z(cBn&ac<{_LYVRT!pfw_M&5(T04icoGhja+s*7tW<K%qWUy!(e!3-j
z=dRNzdjQ81vyI++ifou=P;tPfjxphx&^ew*W&)zY{oxZ=8mVz3>1nsSm~4~L?B@E&
zj4+L#FbArVMm3e-U(Dc@Oikb~WS<fA$<-^2bumqquJ_IX>@|m^1vja`C4E9aL~mzc
zPY=%sO&?CVwaF5HvTJ~#wvL`r`A*)8J_qCZJxHO#1wI2<|E2IuKKdy=%(wX9vVgxj
z3O50Dp$z??k=K?aGJVN)ukiRbz=X?d9&hG-?4FlI<wQjH_+J^Rk36F6fg-UvAHSZ4
zVVQpxdAgv~(wL)OLX<JGIN7Q<oP1v)0d-L1P>^@&3-0|Y%$Y4S9DC$DmZ6HU*rMZS
zwq?vv$!ti~5)BrVY<nh|t1j{ICU%MRrMty#y(3v)g(Wm8tU4Vii&19OV_T$UQ(2g9
z7~zm1e7@~GqCa-{-uQh4#Y9>4qIH%Vzr@T&K+>8oVV13*@rFK1>Er{S*N!p3Ici4V
zQq%!p#WEL*wZZNcsqLo5G*GIvL!kCcPKK{bCJNfmmX_*1tP_z+Os*YVUNCHjE}t5L
zf8%q0MStM8giCHvGm;M5i_bz2%w-aJKO|b~dWf4Yj~gZHEoSaujk{diMiFIFdSXW4
z=2`mTZQ@KiGq(KLv+%{V#m1*xfs=?!UY&b5Dnw4(1Rq^=D~7_5m31}tNa?Y#kv~L2
zR<w`Kb7wBvG>|!a9O*k4<X10PZ>oe@nEsVNWHNabM>2a$G>5>*ZfmB?bL4y^z@L^7
z*)aTp746}av3mpc^YEZYtr($2l3Oz+@&||M&}9=ZN$3S>=^&jDDbzCxwN$}0h86N{
z-Fp6I=(~CJlU(mIAnP=pFO?$W^G+5=RdjQ+w+(DOUWyuR5|RD&G=-6H^6X8jjhWgB
zvsI9eyKJI>+m*9qmgK6<i%&EQ6cxdTKY7xxr4vwm846kN*gNu|<F!!3V@_eqtE&o>
z?cQO(f`R0Lqe8v2R2Fu(cX_#&NMF7Cvtr~O@JVFb!$rJxZU8pfwxhe9!W@Y*O|RtC
zkcUNwripgSh0L0lX=c4<9AC1GfsQVdoel+B)CR}3$Qwr;gXmFuM2nv@zMbu-vv=1L
z%9I}Dam)a#-L9KzlhYliNP#fXwM3EC5_GvG$@@+UD$f%GrFCl8IJwv4)~yFIbh2Yr
zOQf_#@L1(>_;hZ=&2xr1xdJD~v#X1wI;(eHj!NB4W+*}jNo<N3G(wiupx74#>RKOZ
zu`KMg@1P?VQT!-bz8ga^Tn;Nr`ZR+Yk<;R`I)u+jg$5;h^#pc~o-a-N@|T}P7oN~g
z7(&4OYIxKv+U`=eK|7Eh^CKBsuE_qtmIv$VXDXRDjO0CCv#3v7iLFD6(T%KfkDZ;&
zG6MF%;<K&YyeXBZhQX0_{R8^O?YGxN>z$+wZ&MSMn%W-nfwrVAhUyn6!oqFS))KC~
zZa~su6Mm-HaQJ=goM1YzM|FNQAioiiU?1y!%O#u{H(RX??%*)&7~N+^E>M`9JzQD{
z3~nIJORGtbhzU{fjT^6xM#@F+9n|3wn>f)XKbI!sHlSwuED>#?L(n1XnFbSupPG54
z1w{H*S{rL2|0gKYpItnt)6%Wy@0scu)lVBjXi~@#g^<Ej@CmpW!TIcn2t$!xWF}@8
zO`A^kd-ZX8$P9r?bn!-VE8GQpdDq3q8FGx`hM>Wsc=@73v%wvnJ$`8_{6x1MP?pK|
z9YMtuoR(LX&ZuzOzMHaD;up`n1-8Bo9(Lo#WI~?eZ%q-(u$*rDsKpE_1;8NC#)kRj
ze$7(6kvdXU=!`x)bMAV9#K~Q>(y-l}tSXOG0#*BMaXUbJ@X&7!bgVLtjA)2lwsxY`
zWixmdim#$li7`ZuVPC_al?`<owv5pCVOtgwJd~~t<s?Uq8&`Q+5oQzwkiY5LC(HmU
z@%D?>!y1j0qjL8T2k#xv<s|#|CWR=voj(tF=;%LOlR2Iv63_iPl`Xf9e9y6YUO+2j
z4t{~oPfG)ne(04ss1*eDPy=OO-O*$f`%1mKpG_Dt5b5aRvIr^NYi!>Mf${GJQ_ZBa
z0GumE8&81+ZvHRFEuwk&I%B~D-C}OYty#Xcj;)ENKnLMKGj<ia_;*ld;wp4b3E1iS
zOe^jFeW&sAytSn96`H>_uuSPnzl3b`DIg)6#XHmLU8mly@a9FI2X)0(F8o<gaEs*r
z79|Ysw<|TcObua-CD(Z{>C2Pw)meQZp%zy1pYXBHFFd{g`mZ!;iZjD;!_H%#Jz082
z70knK9CPxF`N7=cBJintag!-L!CSK+8Cu63_fe-EVf6&wtkuAvzGbo_HrZa~MViRy
z6r^V6Nvx4Au%>5iNi)Zc*70o)YJ3zH!@boF_=YZWgT`GYyeM1mR?<dww6UwIb7jqu
z1bcBZMY-UlwiBko-r5v3fn-1BAadvTk*b$(Uh!<gx^<Y8D}#IRH=|0q)7jO|>?hRg
z7>w|4j$wVL>eJ$u2{S6_?O?aP+Rx#?5pj__u{*3SwHBJNkRLDLX+*Y{Oif$6g*Kk3
zsE`=kCqVkId8R(UTEh{nexMTYUpv82NGBJq)!h)z%g*2$dM7=<r%0G;z$mNrP!fcI
z_0a`z0t#J2Wjl4BN-3$g)IdglB7NV=B;e`j(DDRk7Re2${n3yTnmXqu+Hz<gUhe4W
zyOQRhbg_sK<+wiaVE+sWSts%qd9vK{p!=Pk5o9|{_$ci*TcB~W-pLq#H`5a)IQA>q
zUkv}l^wgmKl#7rwy<~lLiJRqQH;CDE;}e=l{!t4i%J&npXW@a}C3^6(6a&2Qa|-D?
z9D*+=gc!%1+ECOwslNYm%eM3flo~^)$s1bTa;M)Nq~7aXt_S9x^fj%61qioSim+5T
zf1TvIt1w6OpRfpS4m6`l8Fl(WDhP`Q2<I;I$;BYTB#yu%KfaeB3gcNepWL+ce(Eoj
zB~wZzDby;dE4E{lUFHCc5H2}>3o!AaXGpjoJ3WHuGAk3Np3&+x3(%R4cM#&+IlpY=
zBNsHwooN@Kn9yf;s{WnxUZ|WXH)*9^A0_D>5s5e2MhA#!$+C1-eM1kvEvSC+HL%~G
zl=76&i4TFu{vCjRj#WXkZxKnB`_Ry#$>x^BO{b$?+JAhNr|VT+A6B+BMxEAzeo<tW
z<b=B2_tq#PoGZW6HB-WF@|UBPy|-&=I($$+ET`4er0+W%_L<?6BVs2)B6RBjxes?O
z(TcY|ldG+J{!9EpULw7(WI(}vX>%)ioRkCDkh3Ro)7Gg1uw1}}2O|$IF?pqfla=g)
zq~_(^6fgzpd+cG#fZVlZsTwRHjXlm<c=Sa!=)?9p0*^_c8x90LS+BqiIA7nQpy48q
zhvKz0bf4uFq1ybbniOIYpOz@T0nnnC)GeZNo%~fWyQ)IlPLp?>0@+8v+j6_;E^tSF
za4`dOmh<D6rkr>ma1=>%vG8EWz#d|~{^PBJu}Y8^gN3Fz2~5yj=rpArKSX-n%bho_
zIqXp}wEY$kDBeq516D~xW_bMWk6swUD<?GH`1Cb|l~h|N-jdGd*sue6Xlb0^ty=6-
zo>&C@4&j?e`Us2rqvPCed*8;vCx1gc#jtZb??|a}U?jED{oR6w81_%Bia<g&w=GFD
zSF!gEf=fB(#$^H#wW9+yqs9|1H`OZ1haJ9xhW7bVSSxhNl`FRV38ZtEr9aU4B-`_|
z-1;Gnl5hZ+@as{b-`*SLQ%Q8wLg9nPdQ@<3qyxu)4~bmO9bwj7*aAM+VwV#vSpNR%
zP3*`F`F$m~FX}g_^Jy}z$n35hyHv50|Ci_D1NcETgOfV@C;J~3`kmBL`+cfat|uOA
zmjDj&5DY@?B{DCnaw1@jNdi~$SCkw)w{%{ORY<G<>&xY!?J2q@X1>ye`UU?SJ!|Io
zXj#5QBfD{-IQUvw4jPxa*>8X#$_32^)FB{X@dSOf!=!qsF}p6|Fo8E(KP#Ch#A3)=
zwZ8PpfmN_k{wWqSVh=6hq890}j@J?_SQQ6r5v4;|Re42MxAwHm+%x^B=rqUiyl25l
z!F(8X3+UsgCFN7;!m*9&8*oOx<H3Fq@AP@lgCb-UeIAKI>x^_D{GZQxr?4lLQUY8J
zZ6Aie#lObZ=X#5Oo`e<LJ@BtNw|w`bHmWzynP&<}AF67X>M~{OET(rySsJyqTBIT~
zR(+ZnDH)W&do#xtD-qdKR9TDn!TE5;ba_Z!eMTtQIz&N#;gRhN^fc&Hm1Vz^&&ENe
z==y{Xyz=4pa?Nv?@O6s?2{kvl_=4{J6q+2&8BCpH`zen|GUX4QU!$S00_A4tjfCEQ
zd2BAD+~$zV-P4rI@OiL)*Ersw+(OYM=S@@k#f+WQc@cRIyfA{Di4v6)@iA`22`?V!
zW76?eU)83rR%`ZPgPoIMd|Kgw(o&$jkLaXQZ9ag=CRys}&=CDvpc;?!{>JXR94XAQ
zB#jiO_Q~xx@8I9J;tAF1OA+edZE7t$>-zt}QLD#QXwAUO9<qPj<d<Y<(PL_)LMu;3
zKo#K^PNYwWPYeM8yTX~O1kxxuofzq2*?!iqNrJv)6*9T*``Krf-y1(;wjfAS?2{$@
z{s7XKhdC<pcuGo6=qJCjYHh5gll!&Yrmrx;sJ8F|dzesybN7$Pur(n@NY!UYtN`!=
zb%{VEF#NP7Ut}H4d=%5vJrCwI`IECE3-!<K$b!fkY7mibs&_8a;lKYxCUh}=>TVyu
zS+BO`eb>-Bef>eX^!$fH{+U=xLge`D__vB6xN1_OlOI(ON0zfw_BfwaZ6{5inI%IX
z<8X|wKktue_<cE|p%haliAE!Eh5Bsl#0YbuWc2Z4rY13A!iR?f;^QEb;2pZNx}%Q|
zA8shd=RZvT$uSTj-5BZZWxZrtQ6Sl#@dC0CG~yGC8qndkKPj6ZS_ax?=M0v^Z_*Ud
z+LsEXD9{ufT=-`GNiq25;5{OvehHGX*%m8gN^}iL>%>tyaLUcB8^pEYXji8?>d>q*
zdaX-Z7c<gADr}fJ35h>56#nA7W<ax-%tN)UcHGY$m9j&z&ZEJ(jFsAL&XAbsA&TfV
zBD{!Cly2oOmfAO_1tl5VTx*m^m@GOoxH#mb!yOfr|4tgPVswraI0ZT+IX;boKVcVa
z^QgIDyol2IY!T)nsHy~S#!c&9*)%+ckNIZuCFVyhg)pCMX`c^l;_8=;=wXm{>X+@!
z>ga_BDj~Mx#vqf`d731R%`jC4C;O(FGa^4{d-ab`xx$1-0<);A9iZu3&9id>A6TE5
zkY~RE9UPeM!y6x_#+<^bA{a64I-B1JQ5X%y5y1LO+iz2N466`}cxgWhlKURa>}Cw=
zG8zncN`8mJ%AQYGq9^CS10_&lmiKp%ggs@Ftdt07%?q^8c_r(^pyq-!H4EssjPbnv
zmoW+!4uvyVgIzW=E4E9V?&?bY1%xy3INPI{_9o$u+s2hpP%Cj|xdwRoU&bcP)sLT4
zoPNbZ1eKI}&)FnLk*q1ra*Fclx<<dnDU*L=%;;qAGd1M^3$>T6nlkx!*t;mEdl7rI
zQLd=V`Pg6J`AMamT-6-X#lw8(8hU=2pS{Cr;2%hIBM+fH(X8)8KZ|9W4M>Mt)=oX&
z9+f(Ig>9!2C`K;?XP!Ydw~j4mog%$rJ1rZk8vOa^)b!D`$qk7J_3gQ<=Xn%+C4p2g
z4;>)_k(JKL7htg7JDs9VX%XCfX;_PD@=>-K8Si$yv8>^rY;^S2YKg!l)cBsHWqe;7
z>f;Gk<_B&WcqD#QzQ^Pu-Mo}e0m1+iU?}e8j*`vMg!`PDdZWw2fjGAM=N;OuB}32V
z{BfKcD=8{bfzC5*DRsltrb&ec6-8GL&rWTv7rb-)Tc)g10oh1uBSaj(9Rx3chW<Jh
zho*ZS-&~3rJHhpY<w5iqPVcPP(T|Q<xPk7|R~V{XdI8i&+h+=zJ<{Jq(4q<Usf<T+
zQe^g&f_9L+#evA4u>CX)J@biv4Fe#99NNr+IRs$i0B<YY(jgd&Z`$trv9;<ZcGc%c
zh%DGOgXf(M&;YuIXJXR|@UtcD`@Q`vpUb;3gF@D7eQ%5HnoQ~5E=h>RjBGsu=VfTy
z`w@HPesq^7$gU)7-TL*<qx<VomcNp)`C(#Kwg!RXQivBiwk@9I3P7)UhyzaLY({Og
z(XIXD_$t3GxM|fp*>lQ1GJ^22<Y`h;SVdc)DpHhqdH-QDzrfy7s%Yw$lGPc1D;bJK
za0o9BiPCpyreJWMnV%4vA^P-2JlS@_p}@PBj3~9{90;6EEY)|X`29eOivZcazcF*w
zF-dc&#0)%Svk<u{+$^C(qMWL69;>#wiPHbwA&;1MjlW74U8uVRZ;boiJF(qLF}owr
z``YLpXsJP4*;A3~sGB&|_SLb2BP}&}u4|}?{e@<}(6pJUod(am6BrrxD$Jtx6j;Rn
zTrq1?J)U(saHVgi2-r*4`t=qoaE`p|i8q(&-c=ZQ@Q{z5aqEI;8h=!aATn(0TsrPy
zHqLai3c=qybmnJRwZO+p>(?}NPt7{n;6H?tMmOFYdtsNp43y&C6|r5(P`EDK-D@MJ
z-c$9*O0V8}4;DT)INYNlPFHBGZ0?lr2a9NcV)wf$ww+@xv>tI%D~0SXOeLD(cqZ&5
z5O&#QgjOrzeQ^@u=M|Cbu?e?x_aeq!-pkb2`w6k29a`~t={Y$uemu$UfJ3DY2M15X
z_klj4(1JAGrsDuib<c=Try}u?)O5vMKlK6k&zezdp9l$^%)+)1jKfRYWgCRn@02af
zLWugVZCF{)eD5X6q9^acQa8uU346~FJ4_Fv;0gIbv`uIa-rKL7lr3xizA9Ttv2D)Q
z<DJRyW?aow*E^VE?g-l#v##Vh?!uTT^|kZxHKW~$I$|Un^%kSwa^q!P9cTk{m+B=6
z5k^~Ibnb<D0nf(}_Nyt^{bQcmr@o9^W)hd`_*({0Y3H{+$ae;l!EfZ}(8a%#q8DoL
zzi|fZDb#k$t*l;3fxJoH&0eY7QbDZ)sX1)+Kkm%l8Q`$_w)w<*xEH7YR?|UzEq^4z
zDSE8pCGarF-YA%iF=SV2-P>e{!`kVmSV~7qZVopQ959)Kvey$jc^p|2M<&J^cg*~?
z`1~9_(ptEOe6Z!Dv$N3n)MoFoi8zuz;KJk)V58B8Bi-+XtkHW#w@>83Q;Tgqp1^6b
zV8FRNhJ)zsC-Mpeu^r1uud&-a#0Fz~X{eJoo&8J#IDsXfZ6w<~nI{g1huA*zB-4m5
zZQ#wPdw)sACILjOZG?McQZ&1#K}L;03#q=A>2f+(XA}<M1HMZ#k)M5^q>UY^d-+)h
zQ7;G;>D!aL)G8ZhlRvp>#_fwxaUb$5KXtA{8r5_<=Aq~fo?(+@kcU}4MFMQ1rTKQt
zliagKd}N!R0-%Fe5NndPYPN6$4-C?)_tM;YHsL(7*3>!sy6xwga-^<KZ)8oiBw<K_
zTf09Sl0v^)6X1f^3&q1m1NX}In}$468(h~uxmUIh&&;Ci;YM%{n56DCP<@HS&WO^P
z@{Hq@St?#SS`o_CX5~3($y*Zg%3~VZdJo%=pSiYw*`;dhOS;doKUfLqI>42PzZgR^
zkYz+0(d9**Ug`=EAl(}HDC5t*$Wl<VXhiE_dqk&#!gox(+t~m-kISju^2I-<;FXH!
z0qZ+a4qJCvM?E*My{c&&P<iOQpifwPHiL=tS+zm;{7OuD?ZeAv8JV%*{0t1T*>b#y
zZwES#OR2#7lhIno?bxBaDWIk8%ov(sI|H+g_PSy72_3eV>9Xp1b(y^@(_HOV5&pzk
zQGex^z4$DE0X$t9J>mO9gSl>NjkVa*DL8x0RueB6KfJ=IPpY@0PYqrkm>S5_gm&Mc
z$SP_t6<LM%coJa@o?sxMVh{{Tz?AUszbIjppl{T)wBbZTb2u>;*xL*+m?CXWlyQ3=
zB?yWEq#czS{<QW3(<Ke7BL+Je(M}=IO>PGmEVobX7!y3<RCjjR4N-5kEan>hI>y7*
z{@8D|+nz^!#}_1aAmbty8+`-Xktb*GLAn6X5#2#e^6bK+GmIfMQk4tbo|c>8Dv<UI
z)X<lEOvEFDF5Zt|u-L6>$fCcu4fPQ5X`t=#h^-b~|Lnl8xN^uN%|53-`I~9$P<!D8
zpq3#Z{cvCUD7&|}H;mF&4xvg?t8B5hi|7aeQ~72UXW=~V+&6mIL+rLnUm{xoefAVK
zrzc9nRM>Q;|Hfra<}=C8SeSvT_CuB)_*c=<l6e_~YeF)(=^%(`jIknJL5{rFbzH}#
z7)LQRL(bWj6TBQ#APj*E1jDAH0X-sMA>}MZlzlY(0s|7eg=Z=VZ`AM^E?-L7&Nq7X
z#MkoshDT&*n=wYPUTm_(hw;q6@ZQUicf<G3M&msHBhMWi0^Z}t%CV*%FcP&M5x|BM
z^pe!WWFC2enDTmVQB`&g>FE9vvbk7iAG&A$C)r{6{0~9!Kz`B=R>A0X=h=UOufm<5
zz~J!LQYS`cFun@?>6b;P+GeTjt(FK8^%`9p+YWrr{4czUqv((x^s6^5ANvzOy5nXq
zO=aF@WJVf9T6%wzKO);%9Vmvl28Nhpz5k;uei;gO@)AzaViZjO`l19P>toJk@=1cZ
zkcFNVX74s?x+Yx&VoLGD40WX@lX%UB+}T&8G>!;&V3Wjk7ZR%*(5L&K+;9vOB>k52
zyQtj%<!*%;|KxD6r@4{9UZ=0Z`FKe4v-V%;!_yw?9T}iS*gfPhb9cU%*s}lYhk|E)
zGoQAYJ)eIu(}7?A+y~yPt^7meJM^pY9Pt_&$J1@tV-RA_&HabpMiPP=y4QEfU_DiR
zioRrBb6q<;3S^|=zkDKw*6iDZbstOo7eq~s0j7o3sX0b>{#|zbzeT(EF~L@|bj;3_
z|2hNz$sGT4EBNc56P#brl>T>Cpg&e#@6E@){mX~{6lVX&7C7R-PeIw!?8)(eXQlDS
zs<LAw<$uTc|2l^85uCH8OGw>y`dj7lKUPPWKY7sqPLo{pPue6{c}iE|@2vFySb-qf
zEdCB5SEL39L%HIY)Zu@Vs{cO&H}Mb1UBhr!?!N@h{|BP~bFMzV2d~&*%E&wO{3}rY
z<Kw@+>h;I!=Hp2EhreOIg|orRAq}F%R*Y1CXI1&fN<T1#&FJ6M>i;`z{{IQC7x=&m
zq~DJlhAfxQpI#zPfqA#8$@}}3*u^ShJ?BQEweBApu_AJEfFlW0=Kt^B<@~8T*#C*C
z8kvhJgfRv5&`Z#D&=+p%<ev7axStClMm0+|WkJb!&d@xMYT9(ijzsB&ps9DS4UhJ>
ztJmZ?E6v$ErqMGV{)*q*+`ShO+Y8k?9pJd9GRFkIikp|8M(=mOc^J>$^$c(7Zq3au
zVmS`?S@g>1e#dK{N?x*vLYXV`)i}01!7?B=%8G*d;gXp@qOINAEg6bfN|OFBEr6Vj
z-c4|+b!Z%&OmBWNg1hLqvSF8R&Jq`E0oi39d5tdDmP*RXw2U9E<oLY|!3_ycOpFH?
zPZU<SD{7B99iK*)dJ?LcLSqH`>B*(q{%4N+&q(_E=Y4}&@cmoe%MGp4>o9h2`>SCS
z6IKw_7*_P_l&0fF5i3jfHyl$r76sQ(fi$|F2*6v5X4RIM3f$33YreI_AO&_<*qo09
z2J$5$){~gYjJndpB+><lxPqg=wQph}XJeeMrwE0;_Ae#zJ{OpD&kW7meN`5##fBd0
zc*p3TxRZ9&{5?Geq>nTex^BX){Py$@cyCz4&ztp`VJ&{Gqg+*=+3c}b+Vz;+rZZiF
zJDqIG{8W9MmkYnasU%xOV$H3+;ySw62CEUJ&i_1!=s}NOrIXCG=cF!JeU55$e-y9E
z5j2xZYhAah+J9@&)@8Q|uq^VD^tsS+KDPI8K6^~=`Kr@4A0(-*;r2W+1s70g(H|m_
zbN?Kgq&kd`rb}`G%NWz!P1*<O#~}1SeiuEIfWvATN=*x6KZN4+F4xV~CA31tfer4G
zag5;%;e*frxiA7>!vnU7$xt8zPuJd*A)~!t`xxdiN$YtAZWl7|@~Fme80d6bMp3=a
zZSOV+ooaE`^I?Zt9p?BxIVlVxvI!bR`u8o|4a~-WiW`8o()5kbK*DJ>RVlGJxAZ#=
zaTBbyBVKx=N}K;~@HUTQ{y7mp{^ZW%$f+ZXRqIXU`)&M*(_PS%fRz!Bitn7=b0F1m
zSGoPXOL3Zl7zh#M7-gdYagT=|xX9r?95>0zQ)NT?oHzt#hlf+|yuVYe^wbYYKt<3C
z&QD)wo&x*;#p(~olUhWpxT;>I^p(7u*s|W4adnu{gVhjF0tX|S&06(O^=6?OZC&`t
zri)NV25~eD*nZbiB>wz+0CayNXgN4`jf!|P3W{nMu;6XpBcDnK(8Il2ts=uWN*pve
zEp#iM6r5zvwlA#k7UFC8vt60Tb%Al=pHid6gs=~$gPaYgs+Jyo4EFj<I=Y+PJt3T+
zFN>yHU#@V^CpVY-wT!B_nZYya#1&j9aQv$?Hv`K3KjqWCV6viTLI~jSg)PC0oWSev
za@z5kfiwyBW4IRi2M{}ncv)ivcMj@M3U^E9LF5JKgVdgvKhuRAe$l<P)VEm)Grg9M
z^;zgX6tPx%kLne+`=0XxxRY5BrL8=MSkOvbTKl??nrcmdQk8=>*WEUj&J&WEDzt9n
z>;sod{0~3Zc#Q2npT{yKzTAk(RM5Qqgk5uGS@hK;$KzbJ-7CGyqlp^pD{qr@?v0<T
zcr2$Wxk6f8tFCrF@loUVVgTb<A1Y4;E#o3&$B-g5*sby-#?@5Y4&k<=_grM|AK0Zx
z661=IjT2B|`fJ;PueI*V16()pKi&D+_%qn?|1?k7=EGf=eeVw^PC}6hyuA=^Qh0*#
z5;DE?TDU#j0ybK#H3D7Zg-4(S`>gCC25*Z28L$Fgku&w)mYX0zA_z8A1UJpNX&!3e
z_Jq8=tuA_o51c~g9KR+!^g*yQg6Z-LX2w;&EQq{M0MIBd3^hHqUCh#7f!F#xG0LcA
zJivSa{{<}2In(9_ie>DnoghI4L7(4Ci_d@qfA^G<tuCWZ=xXAGqcH{1IrL(7%PN{C
zp33Yt@S7IRDUFyJ!N8Nr@>uuHWpIY~Hq2&k(UtV%&KK|AQGnK!HQ)XvJwBgxei`2~
z0@R}0!Y}6x*}#0FbjD^56Rt}=Px&C55C~756Uwyq*vcclA^U%5&Rr31g0qz#Gz-t#
z`ROXGJ+DbK>wBa{4#bz8&cP9Gfk~}bF)MG0?WaJ~kI|tmCsTqf+=%-%Lt{1Dq8Km2
zfSMCbDTC0%&y|I}<CM?@E09#U!`Z~kCNFA@+mG>wzB|C${!jbtmJYIkfkj?>@F@Ks
z>_J>+y=S5%G1lJju`Zb(;^Lw>60tJ1R!ci%^)qtEQI-ln9QbgQOQrUa`ldQiGGak~
zNz^omuLG~ZtChdUSm-B7bR{8p7dyC~)BqDlv-V3%cBh0-$O1VgvGc^l-R73*15Sm*
z16ToX2Cb3+(|vLASIz1VCvX<0v@*6Mv*g%7bhh}g7K7nXX4vd9iOtnX2Bn03nqyRI
z_cG)H#pbon-Lot|@A?)k^SJ1431y|-9ih^35ybTF*RhAaM3e4!**xLF9y4XT*YhYE
z_q8i4zT;XoefxpahEki{C9v2p9{;Ga)0LdX=X;(Spj${_eNN<5k~<Ykqta^k&IoX4
zi+R5qw3a2uaRRC$m?JKIz0U9VtPl$9tfdTpmmwWaHqN_N5euN#W6Ug@$6s|0!+Ek4
z!UF{~KEM3cV4;!C%jHMxdk872&OyU}RD(Q`r6l4%ZdZNG<RVN)TZ=w({$l@9+ZvuM
z*X$z1pj;VOuD+^i0+Ca5LyMZO#YDC~aNI^|y|6wDt}0ep0H?Ld?o`!>zx*^Rwf-V2
zm=@EQLXCU(t&?G9mGn-~fH9d#n*@8r;E!C6E~CN2x8>>@^?GPr1J)g}Zf)n>t<fBq
z=h;;!!6)@%r!4S7<Id7;he0o5siHf$(u(tAQR86=oxPI7Eh2t3nBTUWLlZMbgW-yI
z$4@d?V;WNhp574zg?AxTQZhBDT1W7EHBvmU4<r~twB@=ISOgrx>J=J2M3oG8Ji`&g
zGAxyWGes@7w9zsW@Ps~3ClHz~uduWF{C9%#$@#n)e0ZDBkyg`C*A?ZhA<<m|OLc{w
zj5@E+FV_+Qc+TMJn<97kkE^v`B7Sfa*@0wTjCKnsogI$gEg*i8-Q9@jbsy5G3!!uR
z808G*BLxNE>HO-XM$lemuJ4RXUh?kbiCQZfboON>Uz6BXkuegVtMo5a2$gDcZMt9G
z<<wyLu3$g|t{Vd4pkAs9XqZlQ53YqRMO2_1HLnz^cL{5P+d3(#M3$Z5&H3x-zK%z?
zbOm0Lf_taLb<@QAW9eXxVyO5xF32Fmv>Oj4K;#%-MO0i{FfLK3k3E%ey~sU)YX(1G
zY_LuBluQCb{Y&3EFtL+@!uDQWBabrPJ!50&Q4<mH8a!Vx38C037K~e(K?bFP`-2Rx
zTp$4IWD7&9*52}|X>iF8k1j&R?vNv)H=E%{sJxoRzWaRCDy%-(+l(BxyCiQme#`7R
zcHGlCQ$aB=!h%VF{(YbPk)>oHyQzxc@Y&ow!$eQ5{p~CF-NTv3y@)S<=8mbxr&s!<
zM17Le3inK((kqqWt2bGnNBlc2_8?{~e;-P0{VU<v{fMNwyRfxZU8*0T8QSTGvVZG?
zu`57>t`PC3%SA06ZGbByk&uWk_i?-URG@Fpt6{F?*&&*q-d=cV!7_cH0jqfMHK=}C
zBW&#%Yir|EW_Pa{`cTh)Jl_fLkEv5%Hd}u+`bLwgd%cUbu+m{WEOfz_lE&kqrLy(F
z2MA&Vp?jw)H9vB5TQ8Newga|4`fa22Er01N793E|<dI;;hdByh#*Mr2;^JM|9%1>)
z<`EeIkb}~^{rG>A0{V`dOr>vwuba(0H%X1Y=R%s7=(hprC&=<{zGHV{ra-8#(L_aP
z_P(JeviA_<i0zjA>GT5PwZy#CZUjY`Cxji&=k6C8g`qTCObQ|J_iJhVA?&lnA%asg
zLRb-#Oo@wy##|?*Xytk-mo&{G;I0&6__hB3TbKlA-Js6mG;Mj2SgFS#88xsvD=t-9
z#CKjj4|`HuN0tFoe-~FpA=>qMw4MK&g~ffty#{D!JC(3Qc{BE@s<ODZ{g_xZr6fx(
zL|YvDI)RL-Q8B?-YwV5&exlu?AL-%}Uy8w#1A4nxB9{t|+pxD!Id=M$a@{X-UGP}6
z<H|Vjtn_w9^g(GIBLdup!#K&6_LGxD+d@@~B*D|$f;aYg8ThjIaaMxTe&25oeZ(bJ
zo*}~bDugEd{>VSP{*5I-p-KJdexwjm;biJz+xQ75iw$xyS~f){hL-xr$4(@)5)##c
ze73kA{4g#wbCk`{&*zim>LFYzwqyR~FF@C^#iwKERp-^aVDiSb5C^-b>RJ0r(Cngn
z*a``Y3r?OEKdLE<hjFvUslyk?{?+&EOw)r2SY>N81p&vRr>|#MkdA{hTfGh^F-Cj2
zK{?GL{m41N-wL5$JYo#{+rzC369)h$yXDx!g+XJRDPq37cX5u!)4{yEddh_jN}$}H
zPd=T$VvV289K1Y64CbnWa=zF{S#r<Q58LjC%Jy)ei#ov5G%$ALnA?N6@0DQZlNnjA
zH|ct(k19VVc+16kSup>q9JOx<`wA}=zTv>@^|*HgZsO1}wPbd0WPsWYLN9qaXD*^!
z^|x%ceE;%bj8%_&BeDFL1fQj^*X#7M$n7}jc^0A9VlBQ@rNFYk`C-#rC0rchT)QS$
zxAmSgsU2m@zF9?%`5s$&w!~VZdnt=ddMq1g!22crCo{z1&k*Yv<Mvg{HP-0{pCu}W
zyX|NPCoi^pH`hwe&_4Gf5T`T(F^gzqJKo{SKvr80r}A5;b8ld@M#NP%5oL9`)*{t4
zED{lvPR~hhX=$uu?WdU;j+_ss5H~mHsKLgY4I*pS&8+zCUC=gjQfHxyyw_&P(wP|D
z2(&nqjLlB1)RJkVm$qNMf*(<l3b7M40BAxFV#ytDJ!Csqz0xskXhJE-c`SYe_tOP(
zEAAO*AQ&fuoy3&u-gYhe@J_q;eujM3M|cw0{fI7C5+<F*{hSEX$$i+uFsolMEDygU
z;4?B3>d^P$4iR;8a<GSbS;DguI)l8mEI#&ThQDDfxf5V%yf_5@bkb6^gLLWj*jSKM
zXc_3D%4rRFeLP&rd6Z}v`KUN674~@!AZH@7alZKQqsVIqF23QD@iX)UlU__8{ET64
z2X-gtNDQj+qP$ABt(e!mJoIG_r(&Ro!Si}*TCeplX&59A^zpKEq@B_P-cb7ap|1P-
zxcJm5U%=h{!n3njfR~d9T<Ze9OXghE)&f163~^;F8-VqfWG)xKfNw3c7}}}@UT{zC
zJzCG{EU!Hc&W_1)z}il<dECWlC=JkxiDl6l5$`b$3ms5p-q`do42AcD%QJT^flHss
z;L)C_-Lol7tH8$`<#(?A=8#2>hY^CeUO5vHUd$Bx?rSJZZ@=|(@%6^V%S$~;WQ+X|
zIedW9iP_YM{AKPs1bede6Jqcv0v#g)k<(YJm3Fdc6r?gHpK%r-SNnMWYJcL2om-KV
zsh`{r>>A(qP!Gb~+E^t7KJU^nro!^__>^r&6?|<GNAN0Kz8DJC&m#HT1*KLnx$j8=
z=Z{0WK}o6Jzs)Na2GULD=W%zoYLCJpZ}8Wm^%{dlk(5yR4_INHgm*^G@XiiUNa6)^
zf2|%%Yhp>)B<!d8WwbIMo+l+-Y?})p`cO8jZdZG<5=#eqCV+lJ>X1SMLi+3{3@-rb
z<_nu;0CW(B)QsAOC;n8x&YW50J(YQl&5o_RPLPT+H#hs+pe)C+)(h7ztk2=u1HIk<
z!`)lAwG}@5-UV7*O0nYZ?rz13B}j`DifeJVV5PXVSO^l_ic4{KcXxLuSg?oR+57Bk
zpSSS*nRR8YH8X3jdnNbGXTF0kr6@l`QPssX?_9-ZZq)7yizkTCi#(FClhm8>MsjxS
zhsk_(L7y8c>0i=fr+V3(QhXra^Qwl{X6E*Kozcy<LlsKkvFOl%=(8MKUIs`>>$|Ye
z=QO#)y5z)srW9%Ie<{xCwSU$>TxeK;ZiK*iew1#V_FC8Mxll9HjIdb!vJ*}_w12|C
zf~@)F8J41)WbABeZ<ybT3J;xjX<6dg-FKKF9E{_Dhh#+$S!X0Zz7MgJ@4Td4&g7jG
zWea(Tf{9P@PuKk_l@dr?eqVu>uu2@FS=41lV98zAF43MND{1NtPm&JZFf$uuJWQz*
z54_sJYml3dlU_YrDzmS~4@t%4tPpLgIl;3!aKzVoOtsa%3Ba)jCy*R^8hR1bM#a2U
z78$sA24#LqAtA~dWI+yN2qqizn)qP(kVSRY@9{OY4H(zzkX(t^w-YO}oc7*tM)gkz
zb09XIb=dOGUk2!=YyE_T+O$)h!@VI2KF64X3{_L#dx|=;OYj_7p*~S261RbNfxs^n
ztS@;%A22O@kX#V0seuRPo_}Qy2e?47OJSMf8cbyg@7SM<hsrOlq-0i7MI^}f<JQUJ
z!bbk`+ob<vB`g{x`IyGCJMCVlC8~*xFA^LaOyNL=<QhdZnttjZbeEqO8_4s^@k)8O
z*sBSk#c?fBR^jYTb7gv+Au=506+o*<Zm&Ofj@+>wr(t;jqn6O(Xbaoh4hE8t#I-KV
zQ`#eLERP!K#$M%tqoWwSAnC)x>rrKF6tw1}+xXP67}D-{bq_XQ(*wU;@+9&2Xqa&j
zIBrHpMCPu{t6rm<p^HL|lagpKCI{Tf`F)F?VDzl%|I`eXEr42^`mL_zl~S&ZPLs!8
zqdfB%wlfjtqjqyejH~j4ZQR%A;D^~^C**)KCN`FkOVYye!RubO`MAJCjMh2Pr`LwJ
zFXs=r3-s+Tf2C-bk22rJu_O(v?AZ+x4I1p~?n4PrXIt1+JCLv%*S^*nN~N|r$`w^t
znI(0`PF7;WudS`5sL$kj@Ca5fURxv&)!?M>V74G(rfT-FmoKNB;Tv4!99!zO&+WAt
zT+5KM5YQy-&dao6c(ol|*0dA;Al-P4eYJN<tX=*kp&bFpQR|tmh3JeS5OS5pIWqRe
z{)%6B;9=dUUg&n5$neZX?*kx;5^730Ct%w6ibFVh^Y|C(NKAq8eyit_9xZ%PW!!uS
z=gs#v=9~j0%srjS?N4Pw$6c47d`9%x&AzT(nq~0hBnZnsD7T$gP-7DPgc@S_>$m+?
z{ylY&vWQn{SL|c-W<byO;q83l=ttJ;%(68m_E;Q?Epwl#rCLViwoif8Q_SpIzT7q3
z`*`Keq4(8P{!lx_rtcbJi=RaghfMO<E{H9pgt79)-Toof{9e;V7B9kq*3Lb4N;mzf
zEmi500mYDCay#^XJR?67n)1!g-dsux-q#Cg-fxvlXi?7`zWYs?fm%JmSe%9pGle7$
zlN?<{>MV~SKkHQl;W`EJYzR#})ikgeAdTvl9an;yotiX{YXj?5yYIdbCJQ`8PuyCK
zh5`K$=dtS^*XL=A_OT_=7~6J7)Nu~d2Izg`xx@FGtsJY71ussSjZpvYnYyZ6V+)2^
zrY!WZiku*Hk$&+u_zbkjOP25oCuLh6i}u`@6V{vlgMbxo7E5g|Vznz?fwGF&gTP2q
zfoBIcj+I*ZtmesgsqAHm!z&hKS>~9M92hfYOgNgkkkwPAC33tReJ?0kIbw~z5YH(V
zi}w;dHvKCvnyo9L`fNi~48k>*Kgsn5N$bWr5**^H|NRk(x1)W4g?K$yYm8}SC`KbR
zx`FcL4_GkCj^lcjmZ=L1T(tlAWv@)1^eNl+hW+-|amzboPsW+lJu$r>I-IY$dUm=H
z<fM`_{Slp<TZxw{Ue*ZbyGwjSre8Xe&+-IF%1@`~_r1>-Y#*K7m^1&+N#UyvnIF)w
zR+>VF9O1|GWarsDGhWIaD(VtaJtzT_2M=xpIQv%g?K->5&$2_3kqPIse6;JzTwUS&
zSyMlq9J2!le0_Fau00jJ$|h|Q=ZNx=LSdkS^5rAf3S3LSJGFDz2*9%EjlTtHuWDtz
z(3PEmf7?_H4waLq5SH);hI<0ZD8#Ns8KJtR;dDcuyv_-RJ}0A-|8Kch%6FOT$lr5k
zkTmcYX&<IJrTLsF9rEVX23H494n82U*wDX$9c^zEhto~|!xCor?zgqhgTeFQf!Z`h
zI+TbnhQ^gkjuOrHeH!k_s^88I{8d_7E-A$!8Ym@MP6pKz4}{`IJ^N;SJw*<UK#C=a
zO1A1&yK+rrHX>V3>5lv+@#PRBsK|*&-@nrLf`Sj*E+ta|m!hc`l`k8l=ER36sym0;
z$CkX|d1WGhDbTMv#fiMDa^Qo~i+T2D0-ZR<tv(y87Y2(J;&?U%Ql8^3si%#d4}f?&
zL}qUC*DQs{{_92|t5tB%ImYr|dsUWMToy>Q91ikW@BQ=g<~uIu<p$nGVCswyHdDjK
zuppe{I^2MP#G-1M7@6P}j;G$-p-ztiI(I7-3}kqpZ)xzF_0SG+&IXN8oUayBGkhXx
zUw6OBWi~rIeNn7b`{o(_n*3z#R&PBjxXC5{mi;IDzdw7v!%CI@obJV^(fnj5Kzn@A
zc7@M<k*LV55oAaiE$Aj??rJ8fEL;Tae8J!TG?=gth=l`#bpguh3WM&Q#$<kb(Q(*f
z?{V<8+cx_<yNM6}q>PMd2}LXTBE<uSwBy9eUMsl6J1%7x(c#cX@HqASoqP)dCEYub
zq=?a^%RJG4)}Ov&o|eyASXNxln#Wn_(A@OW&;3Ri5Dgi~TP?62d6Wbfsoejoz<xu<
z%vvmdZ2tSu%3WY3)qq5b*Sj8iaT(~jk7?BBp*w?6j++6`M6hs4-<q*;W%GJL9p$29
zM67Ih2q$_R4Z=@r{^s5m6?u);)(5tZ4aW0{j~_0IMwGbw&9Deoj1l%bvW6V_-D(?n
z{o-UlgIT1P&01X3;E*e%rorLzZOs>X543wT`i5_aJbH5`t)>5L43erqwln0EU5^GN
z`spXFmqR{zM0hDMPeZu+z#=S_5(m6LyjrCpiq-9;`ugAtTv!Eh>6hC6scTj&-X1Cp
zfDGVHet%}Xv!4O**<^doY?gIgUDs3V+b))Q8~_?Tfl|yjd|tCB5X}Hk_Dr9oxgO|q
zY_YKD!-?e?*wGD%jSCf<7UWgsUgc>7J?&j(-wmBI?>>_6j4w{_Pj|(+O~od~XFc}D
z5zEJ)_9ag;LS&bN8hiKT>VWw*ajjcH3?8AFNd=4TA<g<{SkC8S&UxZK58Ij@)>4?!
zpnU<voV`tLDNB622czTUEZETzEe;ypqUjJ}?hvDBKfnmjVZ|?1(@-fgcYrcpvs?Aj
z_}fBe9+3AF_v>sjjexa{QXE8$*#(0Ed-*MbQ@oRwBSdS%-)rret;%19YF+!pLN44Q
z9{!jjx5_2<N>ZMq)h#gA0xTohd#lOq8#3pnoTP<{Ldk-)U%Jp!>*!sYX)2mKgC%%{
zaxsNhbV*xc81@>fMn-q5>;OT>xIkfr!{W4mT$;*jLxcOtjgJSTi2Ni%(Uxl$CGkIZ
ztgYU|VsP+)kTk@@{&9uPCPRM~wfxTtQi|M#IXAF)-u?Q{F;mE8FNU#sH)P`PDqJaU
zbBo`PWaGkA*olR^WwYp7Ek{$gRTo<CBg(dr?7cz-q!_;L?rRtNHaw`8s3-|d>*z86
zMcn~5*wr}Gjr&ulpTD`SoyFOO98@~z8^<0)IfimC$iW?uPtUK#jpa*~llLLNo3|zG
z7C!9`r#Kmw{p8#!$p99f-LP={eP>16m)3U14Kcm$U<KOJ1>mj8=KOshV#`L^jfUUN
zXa_L0ORPpqQesaN9fF`=Cwg?IY=b?RvTTRrkVu!(l$LqVyRad0=r_aG;wh~-+mKMA
zh+`Q>%#0qwBr(dcxSJ0om__)QmlcvItK{>?I1;Y|`M|-KAo1gC%mvIh63Q_SbyQoM
zRCq(A_T+nb{*+%<<TVA-kTvhs>Z7z76#g1She1@+esA~h^;@w**itRZ)i2q{?yzWD
zY=tcHr@Bn1K0&S*iXC{rwh;FIrj&r#0X&-HIR<K5gVhr+T38D-YMWyaGKeHZ<IQN%
zah6H{SpbFEV)b^1GsLlA+p=T!772zl0>&%Ukc5XJYb;Ou(Cq#ElKAgWqc*0HzRzq9
zHI~L1VPdh!73#)~37sL|<KYA~f&z;@J=Dt4U6-3@;WzVD@BVnLYP3e*ly+sI;U?Es
zMt5xE-|O5){6G>;e}s-1b$L1$hg_!`Np3tjQyGlzh6FCp%9O5Va>Zs$iRUa#JyrVB
z7_QQ<3eg%=4jC07C{N<`>t%NR-l04u)7}bMvNe?4X#?#2Dm!r;!fJ)ajwtx32izNS
zM6k3I7o&>BI%a8WT7`}x`TmMbpaycfzjI2IV%6KFaZ&jvOmIZR8chzBSS3v?AIezW
z=uJcdhFT2n(<U-NY0@19;&^>~kS(v^3TX=4YZjJW&Q~Gn8z1i@LZ*~%5n5f?Gz~f#
z(h%=2P(yqUJ?XqJSV*^^IhJADEj(8gBJk~=BVWOXAuWYL&P|gf1WKguG4$Y)=C2Ug
z)<sBZ3?}DR{j~v*cECm8{=~yF-ibK-A$WazN_;PV56|(<#}2VHYimIOxL}lcFtn^5
z_<@}plf2U0EB$!s*nITNk=G;bzAA69<OK>29u@2pd*MOhtFIoO^j<Dd934bl>Qn$F
z<{fZk)q|Yqmm-4)_1Z|RQAm5i`TI`iUbAtketQ|%01BvZt8sM?%y!e(JnJg^YV2yJ
z?f(9LXJaVF(88O-cuf8~Dx^E7qksX%@ufVZ*YS>`@-IhB7~W+#C`RQbsXXkQL^FiX
zEA+2~a-gY&5-kUNi0#8XYrL&|<$9>#py@Bnot#!Vb;A3RJmp083?hyuI#R#y*P8ZT
zUOq++3N5XFz3!ZlJaBd1b#`I^8H~Bu3%h&`R9sUhBcEdI+`611$R%-V_RKQ$KcPvo
zmFL<G+CHJD>;&@|9}h>My><rc)xLM}2ruEtFdoaY<`u?DXC~@IS&QpPILj4QPXGIx
zIEPxX7ld_9t(Ort&=hVl)g;wolzcG?qJg?%<|ht2Wa4|@{IIOQW_$J;cyjy-nH-mS
zr>12CZB2IzDWN%!z~g|iF@{BAsr#-mYlMik)XT@-uwUe-Z8pN`t6Pxy^Y8P7ZMGTq
zLKi-ZwhFR*2DfGMCafApPkp|Hhd}x(!wUj5jK}wD)E6OTaNBy<xYjG9h*%J7FDsbG
zTx^|P`&U98t|dE#zgU?ahw_2!C|lNB%%KSGX98BktoOd_=%nCLxT##opHS<1V&N!9
z@w$RZy+x}sM!$Q~w(3hQ^T}AEnf+aJvf|afFwfchjwmoKmvYgE^UgLBj{(Y4@F(7P
zLNA~cxR*L3Lg;1z&!TssG}NBusysN4##wU}U^x(gs9W1xr5-O74r~}qE591PUIDWw
zI=uJ-2gy#*XhgrQOf6U)aEgGY=lP?LuZS~pivgTaL{c(vr^-<}V5b(QVe&}$)zDFK
z!o@tcGo*h?>^>K^-`Kd=oaI^ev-2o<po2eeZ0o%*q)g;R4C^U1qkxmP_4w|LXGd}E
zsornUlPDIMKjwlt-#~5C0#8A)-JdF>OaJB3k!~m~(vJm7iGDiYstIrZ#4x-$nY}&7
zNC^_gjOR=&#oVCX4m1Vuq#uwN6ny#1BhlbmRJC)}X<7CmU4#Nfp!lJU=A^!*zlxu$
z9djvi+#Tl!1#W_zk$D6Gd%Ogv$Vz_AhZa-v_fBrbP(O~Ey&(x)EFJrNfP4Bb%*EdC
zott+LrK6yxbWL9p`{$5Nv%jV=aXveSf)4jQh`|oK3dO>$Ekq2J$u>`Z{)SUY1dX!l
zCOo+rPXJODh+rQ=d^-{TZjw9MzCK1Y&@BaZ;PQv)a@E^1LHQhW0>)ngTDYF{=tSNb
z=o&`wkWCo0fM{sHq2`X<;Q5P|Qsh~6?w;wF-}WgCJ^g~I1+a@B8mQg|#v7CSHF=~P
z_4pWqlid*gMZ;SZHziA!opKL@{HY+#`SpH#F<q<djdDkr_#vtxw)r909v;K`DSINU
zn&OXSsB6jrKr;5zEeHm^?6WBu6;a9!EeBEZbeBEubOkbRO-tL9F%n~9u{wNpr-q9D
zCHMym1Usd9XRxN-kDkr!tX+uk)}03teB}6pjfqjP35IC4s!+#>Spv8vshz>XYtsC4
zw_K28avp_iv|A3*Om+-;QjHA(!l3k@Wv9sGLEaYdh*E$H%Gv_XAPS6l<Okmj7-{Dy
zZOL-^G{Jg^U6Z`N<X`VXjY%M5`G+MPUf=iuIRUql2gymFHR?4hPCdy3N0G&KoTLu8
ze+52aMj&{aQ98wBXXuc$ZV0%>Kirj9(ki#hd}VT9SNm_US+qjslVfBq0*PpSn*S2?
zY0xiP8nBL)!&_?*7KIc#-Cor+vP018o9;@|zsEHvDan`7Kv-K&9<tl01YY_)Ol7bC
z4fGd&_&wPGV~bJd4%Mq)!51TkWy=T5@wFCYP!IZT>$vvoOeP$@X}R-2{#`ZhEPoeQ
zTKBD1zRm>C!ByvhKwkV~`Wi=gN)LIW{Y}lDeMVTB6$w*<_sC57y2$5?nQ2=d3AqXA
z&xL~}BhqXo_ll1Up2;WN0(jU(Ql$AyVC))Bv$aph%3AISmk$2EP5Sn}YlL=Z>J`<c
zs#v3=jDEW(4xw%=evDS~^=1JjzBIFbT``l$-Hhu*Z*cCgxlRIFdXd^N-BMjKFGt*B
zQ(_DY8rGzYzZn%XH(cYt>NQdVI%nML)8GCEhO_C&Nf{TkIc9k@Smfk2noya4w%*{6
znaB~>FXXupS72fXfT6Jy6-|%oDM}D-xlaeF0zZ3&ybV4|45JmO{+sFgkdU^6B+1=3
zL}^)kFk~I1p0oBV>a*i<Hr{z#iu_cR(1UuE$MQUGxs%yX`EDj`1Ky1vbeSlIlO?JN
zr9Y<Bcce(|I&FHp1^If<O0bW3irnSzLjxRz+(%n@{95*KtD9pj^{2u;{85!3jLu5#
zF$4?K2klo^&Yl-@$SqlmnoREN<)RBOiIbHIh=<F3Np2#V$qVkUV#GE{VJPWuSZ~#%
z)E%74Q+owTRu-HR)Sh3#rT<#_`n13Ut5hzzb?5u=4_%XLiy}tY!|s}XDjp&3lT{~(
zKSc-}!>0yDZZ5V%ub67K&wuB}cr9>=|1>WfPPswcP#s)o*p;wc_)r+1BC&NJa>6+K
zJUP}PGIUMAkE+%;HlN9_r1RXD>MlB@lq4K`ss>cbp-Myc?IjvPj=$zN11c*q7KaeT
zXS{hieXi)IBp=`2F|U8EAnL#9yxr~V?*h*=zRZ@FJCCvYMKHhJOnzrB63=n&a&hX&
znVmB)Oez*VhB5V8*-CS5v0xCC@zg4)FBz(tud_W$h2(#9SZV&caOt(dBUROLLG=_-
zMOJ&ub>H#2D%Nm|A?grk%gy0r5CQ6_AteSE;)&Ncv0$qGpObOHTw7-c?Hi)HQ=>3t
z_nU#7QJbmPp~_(?#fH-96-<Ic=L9#R=f@&mg?e5seyDmP$OSTMskZ?>um1#0DYcNl
zdamX}PAJ$bfcuN|k!fE^SrB&^y66%#mB4l8KMPWuvNA!|T)>R}PX}f`c_OLvj6X5!
zMuo<z7eF_~0F0!iPxBV$M55_mLk0qf4?ZSfl=)1<o<zvS0#_wHPa`HB*Q8ZB{eiVz
z?EX5?Mdr$gfKGsami*g^&-^}p-O38mHdb-mSvBWBj?Yc{Dg1T|3|6-bQ*NF&6aVD*
zYe-Bw?&o~q@Hf79p$|Lw<+hvbNh3l`iW*aQkbinyywN`3`@|h<xrtEjl!hhF{ydyW
z1gtwA@l5r3RYL$Va*z9c+b!Dv!^gcP14`aj-9>m{IPn0J2)q**zb@Jj7X|hj^=@&v
zUzRFhxE_yJ$tGw{#R@p{`&qi_9B}>oRUgl-mVt)0b}fT2tem2XZ~hW9RkM@XfcIO2
zaSv_Ce>bV_;9v7$wXEuO^jWaAt6E3HCL{G#Jh9gk=H}{wZL(9Dw*N?@6g3gx4CY35
zSzT!$YNv$-ozj#rixH8E@+K(6h*vad6=!g<l?rq1`$<rYSRK(I{wJUsd1O9VG|Iev
zh?}Xc^|=p(MDA@LvHh4LA{hl?01XULeKz$p;oA?c$<bW0-A~vQkUkUItVI$1=Wb0t
zvr=knQRn^<W!CZ=Nz8ruYVoXxBEPG<M$p^a3%;nlpwTPK!-MNMWPx0A?(=-AF4&Lb
zJkydZWryjDn5TecX5s=$o*7As;0Gn9IuVZMVU>pri#VcK!evh7+e#s7&+?fTw-C*M
zYo?Y1VU_aWiI2q4f#3G2+%5V?dLLDZ*P2`pkkq>|Zjn9<)MLS5MAK6}qL|nCs8kTl
zYDxqWVlAh`RK|ENA)`3O&EP5We%>J>5E7_+d^)h|3JR0TFLIHzS#+rdN_0W2&Sl^p
z4G#qA0PBnQV-6|m@Shp=K7A<aT_V7{H?2Q{9$`V`rO(5%HJnr0r&#-nwrvwPoLNDI
zXgdiXw?#ghYs^jXkl9?(t!T$yqP6_+Cgu*?!IS<8vY?r@oO;7IhZg8O^q<Vl4L_nt
zClWd4QY)CJvMM{V6dw~BdrezxB>|Oo;ZK2mZ51J%8!7CAHQHjI4AQ@D3V}k9@(M`Z
zao{l)mYU(-VQ<KFq>xi5+gSzHf2=cI7M!JTFx$an5-UpF+nmFY*y~Wm8+A258IwnI
zvfUp^`7!h*;%l-oZrwiPE_4Sq|G>h<DNYt_&Iu?W75G4$SYm1Py`dfUey;-ZP2E~}
zys{hkE5UjL-`p-qRs`#xOwi>Fsbd1ABr(^Id%9prh&IVVqB(;3Lv<CpP$DXjSqw4#
zm-%3RSJ+9;*lEz#woO`Z(3w!QV+rVK=qo_+0-XDp^{Bns1n;WIWub373vHn|D0JIy
zF`y_qJ-hBd!W}dhh9;ptD2un40HpF?tPZqZi)0=trqF;NUOB5pfxcshfj1E&ku4x|
zzt5cKtX`9sTbCZ<LZ8VmU30Pb+M&@(TaVsBTS<VyvtilCzG1yHzKs%$cEFrDxcp1H
zq(Q9N9#R=3@KJBI(0P{t&Nzguu*Ao(owJA=RD&sG=_TcnVq$T{VFMAJ#(vBK&G>lh
z9)1;4EPmdR7v5xT<Aul?4Cbi_cq`}m=<l|7Y>^io#FT1T`c5oBk0GK>M!7Yfg+#k0
z!mMd5^Q`Z|JQN7}0_@ticyeBU5ms$70dD;<VMfBhkOc_816h(a-zO4ax$brz<qx82
zCfIbSs#;$rg|(aQ1RlX9d24y!nOg6Lq^g1s{O`(v(Bj(%9_}WLwIBQ|BcjV@d&IxI
z-X}TKb{3U9)Rx}d$wRnDOtzGhxQ4BW{PlOIP+L(Xn`O((DS#zY7*p*#_N8_T>*IM%
zEsv^<EtDwFsFK|gM>CZe87fC#$2AkbCIV!-s|X|>-ht^T5f&D3;1RBTllWF&){WJF
z3k=4wF`7QfkfT3n|D;G{T=c^1!m=4Rf?KNS(+IBF&kSL<G6$jeyM#6K$t-M2{Q8V@
zJ3*-Y@YY9%L@$a9Lw-q5up$Geps8+ISuet^&&a0bNrbqvs=1;Ba|3VQLP3<C)J<bi
zGE95zqg<tf`sl6rgVi^RN<RsVYb0AeWxR6@A)VPjSxjfD?0~;HGew?acI7BB&?7OF
z=YMP-v+Z(TssBvkF=M@}aJh7aeM+?F&L4hWi+*pod5W_=(vvvWw5^~)bS4cI<3OY!
zjs4UhoxTyeo0;z5GBuEXvwW;6%Jx0MPbbW;qcLW!@-)@2;f^4|z{BPr4XLF{bKgSn
z4PFv0#T&yUoW^8;O0VnwIBmE1$<$9@@x^u9fKrch@@GiWgND!VPE5AO8D^w4WN=cg
zz>J7;r0<KA^-^clcWY4u%Ifb@-;HjJ!yQ&PLhhAYZ6=69Yv0sZvkp;4)#4_o)xP$f
zXS;?F+{Rp0$ia(7Ju>a)u=%eRQye4g#rxxNd0{f=O&0#yi&X5ky65W<)^@OUpl|wD
ze^877R|dPj;^qQ&9i5dgLRY?otvgG{{;HN|h(`El+^h}58su2EUaY|k7lDosA5OV6
zgba(Mw<zEG2f4Q+1a}=}LT>w?l(|Osj3^WHsu@YI^63Tg>GQ-|Djou5DBTy=Vbg<`
z3Ir(R1XmW0gcD4cl*FS_C?rg}4SRsMz#L1CKLAt`4A;L6mC4vx?5;7fHP%BcMwG-~
zctkg*XSgUTsB1B9DitqfOe2(1r!sSK%)Kc^B>rG=4C9JA|Hw4n7gS4d&1AzJQZP1M
zrd0$-ty6#W_?~snj>WJ=aBHL^+~oi>S@l$tj~9ra>_aALIOXBMj3j?6%sF(8H#U@R
zU*amJI`#x;Q|wU0sgB}bzIOP4l+uBae5EcY&L-~CFj26z_x5L#wfGL6+l*-Q(rv0`
z;3c19F3(|~f{K@zpZ{nxV^F%`c;Px0hyr?ASzXFL6@>$?Z}Ip}H54y!rqF<6SWDES
z-t4f@um8g9zy0{N?K#U?&{xiyj$zfx1vo^u&y}(fU>-cmdxJBsfe#{@U6^?)^G8KV
zbV;*G@!(%mxe#bgE7y&Aj*C$7vM5>N8f>k)OH(7xgbf*>O^%8#Go>9wr0*HGb&dTc
zcger2Cu3geum;FEEbm*twe0Hc!<mZc{2dEBy5FppIKmcQPk+c4f;YK2ytOROSxy1*
zQFwf3m&~P2v9ujOM`i+|oFP?zT+YsvRzUZcIY{08bSUS*d!Hgw4jE*vwUftmE_4{6
zj^c1(sV+Psu=l!zw5o3;I;;of+ZphPG{bw7%vle9ify{sJ|vI{$wUDh1QNmt@3HT$
zMe46)eP(YCdK#nPfp_CO>AV@yhTKMBBBTxE0KbCf&a#fDufhZq?pkmy<MfbFcLg}t
zXB;>(mTexUp*602c~qz!&SjHugJZ@3!-cDAkoq@gU9rE(^cUfiki4{r9bVp~C9-+S
zSU(5x^YroaFsFkS!C!v2hhp?0+5$+wx8!a+?94s4@jM(vmjPt81+s%q1gL1v^E0V%
ztd>0pNrAET`Rxx{)SHM2i#B7OFZq`>T6SNE+BW*_HT#_07?7Ixy!HtoNtyeao?Cp=
z%qvwy?yO$pKFx;*7S6Lw9GBK9EDnV2ci2&*{#4>tY*VXWWR>Un2f0ut%{dp1`HEam
zVQRxnZM+Pl!3(?)QHp!~E2u_el=v1qR0O1VvHxZ)-iQ!o^jop#re8sQ#rT7S_l;yO
zIi2HT1=8{Z5+sgy#efAkCwvnv;)2dlIxh0lsLOtS*jH)ezyKb0!Il>2R5+Vj`E`HS
z|7r3Zx$;j#A(#rNj4=j#Rryh9;8XU#ps@M_i0M%LQS{Za_P4F4Hx-M+t)8I!CW*)$
zceKK7^6k&g!oIcAYYQTcR@auZaH8QM#83MjPy0$1DwmG!&0T5Lzv(`-0a#p2Pra8p
zSsApeK<yB=AYy@OVi|L$oex7eck19-SmrQ-#tHvxvD!eD9LrX|+OkQd(7m1Q-AAj9
z^zpEcfKHP=31QO`bmXY~SK^n<<TnMAk}zMFwa74h+_D=t24h{X)$-4&&w7)!Um&4w
z?8;^QwJPkY0#%QtxGR8HYLED)Emf=>-y%F@IKT=tIAl422Mmh2wIriDtM^yY#q>RO
zZ6Ed!(*M*)iSmJBIqRgU0Pj>{%a6w!1^;mIn;nEzuH4ZVORmg|JiuqNJf2AYsLy@g
z(jJ<BM1S^=i+AB%#!%++@6O%<$UF8Llk1%1!uuOna-GY_<{NvVySYr9M`?WrJ#3|n
zi+WTGle1i8!j@3hfycv^=BpUI$%9g<k67vM8%XA2fU#9dhh_Wjj6t6Z>_lxgPG&t!
z!O5TjJC~GO+e1BpLS}R9%U`ZrlRR<S%rEDDuqdz5Dj$7Hto!bY^FE-Kz2p}=G9k_s
zp)uwlNU1ZjBpFQZdS3E97C0@RbSeCh^p;-eMv7=Q((Hzr6!y~zdoIam9#{UuR9@y*
z19X`2M0dWx`J+`qptd#MN5Ab`+^5L8R4^Let>RNe<rU@<FK808nC<k&Ea(z21gHnf
z^9cb8G!fzvKi4unm(ZuC?T&YN2?`9Wt!UpEwNv);5hk2sV(vN`{#?>xvy361ecJtv
zp!2OW5{Y<ea2n}D*p0P|_8??HBKE#z3q*GluCK9})lxAz?%q|K(&~{q$2?{h4cq?s
zaeMVP%4bvb0pJ-@D{VyjaizSy(cf^PRrc9a;?GHFxHR=*hFs<t?YP@9`GjkY%<Ieo
z(69RK&MXOFhDSucV5;2Q-Y0y_U<aaV2R|mCIX&F{Lwvh4qx;&3QDD=ngb<P1H>wb<
zpsMEM58RK<(v{3-VzX?E65_7|w>L4Uf(;&#bXBj=(g@$=Apto`c^pbO#eaKnq`AAf
zN5qsRyr*K^EXgvttDqi;f56`^#hYw5?4bHxo@_1EJPOpL3Fbks$D$O_$2a?G|2?~A
zKelI05JY2iU+#J4A&J{HIh{8D(pJcGE1W=Nh161_Hj;VQ2bVUg;WVMj#bb<#bnDQy
zHwy7R<}<-{k3*ymDNMRRd9$yRFPZxS<bTq@Y=z0+zwzk{Cw*y_WN9+HyLh_+g%Pyg
zaLsheN@zS2Tg9r-F2(-(uD>uKKq({ESz3PJg~V!DqD{x!?*{C*6|l;QL27(=IMDjZ
zMF1q<ljV%mCN<koSaJU-4=+V=#Z2;$0O?MCT^Pjf8B?0W^;2We?E$JKeP5S!CY5(J
zCCz*@c$Qo9P7UC49sb)=>qdNYPVJeosg-ze$-SOfs<NC7kq{v=K(3u;91qpC`6*?-
z_LR}&M)KtM!q~!dt>>Mn;R%|CEtD`2@8_E{$DkmD?ou$81oi#AEc^dG(+IZ$6~>2H
z+f`bi+`jAZ`bp>Qe=aN49te?qT7gKS_OG9n5wQbD2)Xy>bTQcxeGw6sHf9+k20mE-
z&$lC`DBf3)+k3s4U@-o_2LttS9b+VD;V%OoW;|11dje8I)w*=MhMzySZbjav$&6P%
zS}w*-8Ytw^_tX5xjQ_fEq|_<uYb}16yN3C{4+O|jNuvHbPXCk9ZO1|w2$jiv9>|Dr
z;{?&k?zKfI)SZHV;g8g-s(hRT>hoP0IPKN7F1vIur2RiWroDM<8@hz{W`^N^S^b~T
zNuj!iW95tyxg7c{a3-*(MX*cwc>&HO`sK6cCH}{P_E!u0qXJ$_R8fDh{>R1t3jQmU
zT<W&eb6Tfi^4l+hl}umg@z8|ZI9FU;48_y#%l~;m|K*=I^?w3QL>NIo@V`C&j~V|f
z00DBDXEa8l%W(Sz-NFCIiT}r({{-y+2NwjWzm@Q~C-IE9v|Olqb&2jZL?h^(;WF!2
zrY0x17Lk1tCxl&Q9DUvE7ehU>a%&<^UQillk7)_IvZsw|le_yTIt9D44C+&Vi`4ql
zPMC{>#nk<|9QRSVA5{#NGPu#DL{-1VEVSP866`|L^bgL>ZR1zu>3@e`c@K{4?yvTi
zBoyWbMab6f%vQhCZQJNGnU!&9Y}`(szv?iIc$I`7>#o0ggioozB-DG&qE9%?pTDU*
z^FK&6+1qXE;aZ|XHFBXb$c=fHXKOSK^jv?E#MPr*qsxEnS;zYmty(QeGk(#S?0<E#
znJHwbG3Ra@`_W}D(&VXRqUFzGyppEjP57rkM9TX=lq0y+ltL_o4oh@NpuM9k>9#`3
zA4+HMB_H>4pq!wRwM3I8-pHOX3A@-v*{7Sn?y5YG@Dr>yAd%gRx=F`np2^0~0h4~B
ze--?#M;UL{b`UPV-ziur8z9*YysKp&P#@?MFL$u==SXY({y3C9kWEDyzLzXdW+>3d
z2C#MQduBsK!~N_V=bScXA}jKo!m9*xk^0%v=4ZKl$d}h>D*J4L;&l}0;F*l;$NP$4
z>`C3T?esgxN_Sk$QKFCg?RsQ5O8uyM@txgGz!d6Iu2ZP&hek>e;SGrq6qJQJSq!mb
zLgHZ(J0+YVOT~h@$^rt9;JBB-(1N#}YV&>tzJ1meB6=OHFzR4$9+*uDj?Y<RJ8NXx
zwM7Y;UFPb~_Sf574MGpsG8K<ps?%G78^{j!ha}WL9<h%-25GNO3Xb;7hL}%cto7Zp
z4&T@)$Ny;arjFd83j}`oCf3dB@psSlWU4vXCA|lnaAkJUkR^nbl4hDpgAoCEV(Ng;
z`aHH#+38ri8rBMe|Ll)+>n?gzBBD^Vs<X6v4sz`9&1wlJ`mm&9xs-!1yn#C>zVprz
zjgq#`uUZwZEJKtq5q3-4{UU6^wez0CeZ<ggi!m3q)1A@gPHD95VA{+z4-p&mwr%1y
z$`el@j#0){)EoZiicyO-Mf)N4v7R`ghOJPNv^^;lhsHgn@@HF+j7<e-5*fZLOpJZ)
zW}f)Bx|;K)d;4YuwGT%`6OQ%!Be<n#BQ>B{8~ob0KK0s|_My8*`auja!%gU+^7(}M
z>ex@ruz1V``X^B01TWZe?mzq3jjJ!iK54a_QlWYpZP`!J->kt<zdL)J;(#agjyl2C
z*Oo7a*x10FJ(0xg>(P8&|KpaUq2?XZvn_=V4}>;ZIGqOPZnf$khabRb*kbDBGtErh
z^5e^+$%i5Eh(nT>>WF*&4#B9VIQPCM<U(PRi_Q%Z$KhF&xLok-s)YKtHp%_v*%!X>
z@3C(6l_6#tBvpw-nhRw9b+cY(JuG_@m%i@ZYj4${Yvi6{D-BQ^#s>B}F(Q?u0k@Yv
z&G0nu;_7$=-^XjSyO4b74;QFG+pKQWvvt8eo+zmpeu3s=*X+W9yr+iY5q@Zo6TBXn
zn*<zOFnlho{?UV6a5L_7J;e-9C^h^u?RO2h)4-lNdJftXQJwC9E5=t7yFTw)-s-Ih
z$aL(yx=3Bj3A6m`&{8=9s2L!(hp#(va`3;t{30<51mom;5y6dg_vD_-XO7iF@gJz>
zNYjSKQd$6QI=stbH&thu*Krks;voTZ8$XBI2e^;9f7tdm&^R`8>xRCcTW}B2vw$_=
z)$goR@+eeLvms`o>MRkXt7g2e1`Y5|Q2%WC=hyUQ&0MI>@@AD}1alM<0SQ;Xta+P;
zl1|Ldi7X#yp{beN5CyjHrDtyhdru%pqjx`SqRcI`83eH}7_roNTI{%b&gIE+eB{xH
z;%t`k3dM;Af($`V4wN6b{?cNDn!lyJ+}N1hvOdTuSc3O|wnet`%c&51s8bmnwghFD
z6@5iD51OXV-an+_e!J%Dtup;ZJ8}-Y7Lop)$z3q5>_lBTF6|g8{$~tBS8`FNzc;gC
zAweTF@5W>l^l8c<T<Z7+Vtnsgk*sX3>~F6)J~EUMdlV6s{mTN79e00OM`6Swz(iW*
z#JGfFQ^_uMh^Rk(%GuvXR0_q!14ofCMzjj(@SSoHnK#+Uxmbn@-P+)F+pV>Ik7^pv
zd#Uy(sw2{85ODWr*tPZ26BB!il@&i<3V8e^@?Azg%vw;Dfc6p;5O6git-caa-=zg1
ze_<(0*Gl?xZ1{g?0njXb<aKD#k$5V8L+6yZ`?Rdw8?G)}eOsh#o8pyKZ+_&^v*2M%
za3;z3y{1)z-2lWMEv-E#Z1R|p+EX7nV!kuAPM_R*-z=r?eC^3Cw#xx6br6t(US|)m
zyHF#CZ_p})ca}dJd=dXni~Q177vbf#TvC6n`a9B;ib&sLjD)(3I5o*3;o$I)WWu!s
z@0Uf`M~Q4Tl}CYo7gWh)gKLK$)MB%p(wnUQPI>jUcO={<VQmL3Z$oBgVAb!5@Bsmh
z>mfmn8V^c2hOWqDshOZf2{&1UaSg9)x{G*8*Kj|i$evf$Wrv!ed0u+G$FmSeWMLs`
z3Ag`&yg${75}xR!#)%Rh6LjG!CqnZ;d0tq)%Kq0-?2Q}tDv?w}>_>Wr?~>eM<dI3`
zOS2(ADw`0G_z%ORP4>Bd2bu(xaVIRvZ81o_ihoYPl+vxjbPa`8l5ZUhdc2Wsi7*Mk
z04Bm5i6vTOL+>0+<RPus-=1;~y+9FCGLDJ6CTH#{sdzzXmOEvt2Rrl)TM7w@>{Fzg
zVP3P$n?hP}3=6(L%d@8Knv;(Hb=NzSL33%}XhpYAD}Vj-4EUO7Fq;e9uEveQVV$g@
zk+f~A?9VRah7tE(%Lydr)f-InVb$oeuSAv!Tyk15=D(3Fmg^q1H8Z1@#~tN69NWQn
z;AlMBP+}NuPaxO#ik_RhVorimw<ft2z0&OJBTVjRhVkOeSkcuoyqzVR9_|W)AbEY}
z)8?H;Dc`)~_YWmU%9f<-&)J1Yx^^yho0-&IavRB?HkzMEFN%XKBkv>qspJnuhK8C$
z?Hk+u=P{nUY!qZuSvk!p%j6_G1AN;#9o>A0=7&>St~b{e$R`DGTJDs%ui?d$eoFX0
z>1Sboe~h1<<+YS#S>Z%*ejxC8Zb+`&KtkYMmVJoP>l|WdIT2dRJZAAYBjjv#lZDF$
zNg0C7r{89FTvvXYlxl*J78Z=0)eLxk#2jkgC&lajRcS<2|Ae6^(|U?XSzQL0ZbP^J
zBl*tlnU#nrqPSIDVt?q`DFf9_WKm}a+~IF)|3>r@_s%C%je|rGkoJN*FMB?;H16@-
ztGX^4w1(#VDthCxnLY{_N2Bc?k^B#lnpG{j=A(MN_?@0)4dLmXq^|oK8(~jbM)o*(
zOGe@`;>#4L_`|=QaVMuOuNO4uR5@QHgQ{Af?)o@_pZ`i$0+SGi3UN(V(s!T8`IE<v
z`H;qSp51cdtHTefZZp|GD3iqJ1q&Gs+b&7e#X9u0U92T{SX>aq(Vk5>P73(Q{8DMk
z3=-@l#>=rBg8XQM*nA_LOgi6MDQq{K{Brg73(Q8+Z6fG260UW#_FiXeM|HJa-&-sq
zGE{!33d&p&JWZYRv_lsMNi0%*S*w4s+*XEDS%!5Mm-EyOIA5>^RD$lRL?py-Kvmw#
zb`rjL2r184h==S75sl!MjgSQ{K0O+i)7M&D&B64p>Id%VD#34QPpSSUxo9G&X(8<z
zX$hK@pnPtzd_jv*Z@U*vm(`}FWqOHxBx)|hsol*AI+vZOtSqmi<kyNu!<6)9r@RYy
znjdO#)RKQfHEa10t`R=7Nwr400Z4$;=RzTeOXWrV;3*n*cYEfUTK~1|Kf7*f!}ej7
z87mhuGpkS~Z|LJatP;n@x#0Jo;|?mK<kvll5FEX9)l-8IG5=;syJI9jh-M7&wnaw>
z5mRn+3U=|jj}+v*IUA<R8Er3sq*TCpyil_i#!MMO`OTiXKHdHMA9duPjh&`4t2NCv
zJ<RU6Q`yb;*R3av_LAMFwa5jzjun6GbB!-2OCy>ggIHYzMET%2m&OZ;l2)p%l~T9W
z)^KmJXImCQ0_i^LbauB!#qtAOu9siUS9I8tFTkv1r$q-Q-$VuFwSpAW108Eplgkgo
zvOY+SuWB4yFu9e~ZOM{HZt`jax3oeb9{&ooD`ks!=gpu4gXnOW>CH9$Vwp4BtTl?;
zM>Jswr*zw-J60ONPtq&vn>Kl2L%z)Cx9cNY6HnHrNkSu8mr2eyr7wTDDP9S+WXx@(
zzfZtqU#4NABPFkXgI~F%Lg|kZEF1-*EdEO6`nPuQDj>ee9;(BZ%4V3*uZfz5E2JCy
zcJS#8=o^vOys3xg09KZ>M{hShK(pO{H!tZ}!qD$s#>^+s(cC^fDb-i0m*05hc=*WQ
zn`u_$NoTIh@6R*aeuje0wQsbHE5Y<h)JXvj$*>((m@=QjP@3>r$s!P)>NTrl47QiV
z-mk#joLYUQ$AQMJj0^2dx~qP#{iHO5oSo(a|E3w}(>}rECIwI@_V4|PAIVpfs5Gi>
zSuc-<K8TC^{!>=+X?*ML_1s5VC)$IAp5EE{2}NE^+&ogLxS~*}wa)?env?MzF-YqZ
z+nL6420A>3>c_Wp&sx%EYbnB`Z&VC-`PBw8SVwi~F%o-mm<5zdvdz*yczhdBvOI~9
zQ(Ao&Kg(33%?5#tHjR(XITX5!kXl5!MSNs{nq;f&{2S3#_e=r{iez8$L;u8P6W4hu
zeG@KKa^VQyvvyHV^WyFGJYY$8ReD|!C!4>)3-QSU;|Q=tL@;{qq;uQ`jvix$$aC=l
ztlHjXla>cr47(7HrkCaFE@hDr=Qi%#XNV+I?gn~Yj`l|R3|RLXOHpsOPV4@&NUT<|
z3G3xvu2k>FE0p3;;d8-M!>&Kxs-9HDJYXYM%QUL_50#{mlkOqm+J-?MW&b^|S;*mp
zQ+ziHQB20^7qXS&g%#~dKajh4?=Fh;Szx>P6^%^nF*RFFlh~w<Da+*9WNb5joQiY^
zayC3lcx7(3(9X`V5BBq1s~X(f9&Fwj>W+pulkzx(f<YcO70Y1`&Ap|Z+f-goStaUo
zuj5Y)W4+AY+JCl8z5Yg3ACGmG5CQPPO<3rpFU-*vrmtpX-61S_NP!}>I-oq&|LFN<
zITII*{^I0C-I!g@yk~7ypY5x3B)bM!7p(#70QYf@R(ru;A3_dHTY7lnHrhuoVzdnW
zh1oSM(4RucW+1PS6U$|2RRE9Bbt7zpzYys5JyqFv0%e<F1iHr-#H07m1md#f_S%d_
zsIV&(%Ls=GI26En$ak==e=#_CGv*C&X7asXyHDYCzMu5xtw5}Dnd5qwF=0fsXX9ns
zXv|~unQx*v{rEk>u;t|Ohjsr%7riWh5~0>P(txU|%6UquZjoR(!4tFG+#ABL-?skX
z)xQLuu{`EL4MN-qCoj%_JvKzgUp!O+-@4lu+3Z!cx@3?z@nO$r+nKdeVCh2*#N;Lj
zKJd1K4x_>;{ECnGe36!d&ZkznN7;fnqfmtSNRdaq?t5Ct;tjKwP&Wk)d}YK_mj(KE
zx%rF+7Y-ItZMz&-VqjMS_=o2eqm<LPafV?*a|m*pQ4%8}rOnm$k}b}3mJr?p#Wmz^
zJIRA~c&$W?L9Wa*YLi@Hfm)w+?ZI=4w`CsGA#U%BkrDY)xK*e}yu>e&K$O7?cr4%G
zsJ+(NH|Q;pyZ0=|64{L}zj)qtoao*vNX_@kmjUKs;q(0NfV#$Oe<`6Q`!GfS5q&Uq
zkljwZ`qn%?VKp4n=CwFN*I1yx*jm*kzN~?Em(xr7HM1!LG!q$%h5du;oLJcVGXGB%
zXUA}j7Wn<KT05-rb)2g~%_}O#s}+cgpavkdc)~gJX>UNH@oa7FI4{bP4=qoAbYNfd
zJI)igSdD)U%Sk%>@$pH4Z1!hYLvia0P|Ag43ignCq_zAoX8?!t+?wv~#Z3G@p|G!7
zR%dQo-khKloz84NU)mHZh^Ye9809P}Ypr^s>kUm|%oyysjWZ;gb4Gs@Y+R--uQs+I
znMr>-Eow8I6!~lT{dDZK%i$5mlh)I1X;Ff~xB2A#!&S*-^K*@Ig9Shevv!7x`zq&i
zLU{slyA1&G`@~aVxiZ)Dp*RYYu2&DP;3U5e__9gxdqK5HfMm5De6R%&u@O?;c4u{)
zBm;k4WE&J$`4^%~I5UBsGC$;*$4CWs{q)lnO(#v!)y}1Mj3#Zob@~RPJqDEzPX!Fy
z)Fn<h??uGaZacj4AQt*sOdOM1-^64OA)d{5-0{oiRd^Rumpkn`{<@#ceAEn1SL2}n
z2FSmcZ(zFpHj4%O+F$n+;JM*17|TPuM8A5rim3osXwPy@UE65$4(3^W2@&3p43SQ$
z!}zBA@`4Yr6{Zg2o1@eiis#wN@DI^$T^&>e<1x*G4F3=K1-=*L&OrLG4M(gELN!x&
z8Uwey>TeMcqg$MagF<glmV<qeoG#v#kIg4fTJAZ*36G+Uf1BICz>|RZzk&`O8CJIM
z`N~SNa01gm({+VymDNoN1mB=dSx&b)o_TUUu0w35A+}W2{|BGK)=PdHKGH=xc^zkH
zLCH%$au58r=KK!`v5I^lcg=6|rfXd*g2Kvh?CYj&VS{Yr_noLy*I!Ax_HXd|$gcGS
zc>1pR7Xs=pGc0=Qgq>g}K0A`$^N*ixOdXfEZPh})O4!G~VsX2r!0qv=Rj)ZmVnBH@
zc8&Von_l@@_fEcgX!KyvVQK-Rlz+E%klF&PH=66)dyQm@d{g+&wszVZ!&zBVVDl@!
zA|B|SF~SR}+OoY=Dv$lR6~;XBD?%eX_swS~Oh^IeZ2%b(G8xIwQ_N~d)9%iIt772M
z-hB#Ih(FaY3a|*kY;}>(w=BV)YxF&mU%d+&_Sc!vLzN1Z-YBs2>}^ztl&K((dYgXb
z@M*c0f|QU$<)o@qsGyJZQO)|cM$QIoDBg|YijQZ-r9lC+P`#vk%A63B-3^@>obt&m
zft;teLu6}<;t~0e>1!6}bnwSCv#Q97bq~jCWP)(Za~n<0z7u(3Yw0j@oV~>venUl5
zRevcDoBthRaof~(x!>fMC&4<+S(keq9Q(_lZSwf-g>EZ@cz(6YPNtB)H`T4M_Nq1A
zH}`sL5sviWwXBbhD!QSh*xSi|+nzZTvu-uaAb~TUh44FTxU*478yhtXZ5d;|D$`Zc
zTe$UQ)mUk(REqtU`hl$pEDQQcw^tI%i7}Z|_unX1fy`Wo6ajy7r`6t$?_x3mAhj1Q
zQB}<!Ud8r=7}UHKMOcLZF_Rh+xT19!5Aek#EzAkYoN$1bbXkLPR;|VUx=-iv-h$(?
zQfgL7^)Jza&TK=Llk#6KB`cd|^N|alcY2LL7Ql-5B8Pj~DCygzA;koH8ZXVkAY<-5
zMgP>dBFO;{Usp=~_Vo9ir|kXF=;oNE|5p2V-K%bDzkku8jGqzIVZVry)MOG}N)vDU
zp<;;BS2}mXFL~=~m29_X><PJQ7WNA>PP=*4VA9&<@>il&iFNi0i+kbY9JsKt`{8<M
zj6pLMTPulnXr;MtcGf=^322&{3U1_VbZA%;F7u~VdZj-Z^?xmBcM@InwdvTs?++MM
zh5K7kI^`d}@xO7Iw^hkyCu-Z2y{r1H>2UYjps}ik?i&)aJs6{N#nRFt{n{>th4m(*
zNB;h8+4}GCW0a|5e)P4=C`mk3k2i`Jbb=ox4ECu!AF9;zhjr0;<C<OQ=I_^8d$<;v
ziqp!R_W3N<uN=svl~U~E&f9!Oj(&f#-F8moCrz_^-C*p^CoDh;J(N%uE>+M=j{!z?
z@9pF<-bw{ET;L9e4Cy)Eq)@KfynhG4y>=25apQKPTg>agS_$nhV1#j4_~B<<^7H3s
zy9LiW#gH!-;Xj)CuM0VXkkSQZWeK%UA3u{LUgm8_^V}-N3yRV*;rKPs7jycFHIRwt
zmfZO<7{3l|OC6*J-0uxG@oc?2jL=`L^^4ck{bwj}xTDhXC3MQW@GFErIbS;+A)wZH
z;}u+Y@p>n;I1Zs<ZQ63J`6p6JiEKIM1a)cHb0$DpJ8yox<eN1l{4*FfCdMF=P>MF6
znFQ23y*6*F9ouEoseSppAW(3*gB0MlxS3~kHaMGuaMV}7|E?~hMmYN4*HH!?`x?hn
zHl3sbj*)BbD{K0+iPmL9a{V}2^PD-#Rq8=e#m9sTj#kXV3m(afKndX?WY_G!1ybrA
zO-QLzNa2%pVe|gw|4pB9thJImUL|qnj0A}nshu!O3S{C7AOX<ICAoYiZ$UKakUQRR
z#nl9eIwr<xUEHx#Tb#$>XJ2EsTWp7O(j5WCP@UhrSESjh<wU#V&w87Z*1lc%uS=)|
zLBlG2`5XvY$0{|{Et~bSqkreF$4*k6jDF-TPw0XsiAHQQ?y)|Q0cYTX7WF@KD@LHs
zI&w9bNvA-~Bm>4oIVIl9ADw6@EQd$w3@JT0#)ZCrIaVw|otbj~A8&6N6-U>tYlmP7
z8iEBUIKd^jCAho0LvVKs?oQ+G?(Xi^xH~kByL0+^&b#*>`;77Z{eJeSuCcnSX3e!~
zu6f;e^6egrRtsu3h{O}^J$q_Iu=HNZn&XqkJN=2|IM-?}VC!H<^}+KH2vFt(w@^Y#
zlnzVxJ?XEbdXD)d9JvxUl|q-Z<gmVBq?QR1lsV7*nO7pg&4OP)4JGA*SPUVK21Bd6
z6O7zZCf37~&{paQM0OD@eG~a|EHj1IDf28W_-lS+=}EekVLgQ*;y~$dof&jQLMb!v
z)##N-N%(NMFhaV7m1KCyE_&utVo$;BMch8~a1M;sO(ztLl08G~YO?=i%EOZgNRH&V
zxf+P-7dVuMv9T<S*ICe?8re&sV%7Gm41kY>0S8QJUf}KFgyj)U)xU%nOPA8L+cO^h
zlwk3ifImNGW>a*NPOY0OLBJ_Fdy~_yEE^hv8}yP@qLQprCbajF2(xjG@^-@oGlYv7
z_wOek$>EKAr)DCLiF*0L!RQ7R?nX`3C0SnKeT%r;uniw;e|%G;c)H98#vx6$9XzMl
z{5vYLm?hEPU28wu%}^!9EWos*)3LlM;PS2O$pifzf`fU~UEEyUSmtIN+DNdI5=o)n
z@n%sCYbU24eHXXJkix;jmz0PH!%ksGs-Ly49p4x|OVyK0XFZ56ODJp_9Xd@ma%x)N
zx{8CHMsxOiM6!Izb?t22vyQ8m!|}VCPbtIVE5^UOCBP-&0pv+|Q+UV*P)*o7+j*-Q
z$U&5?6p6yQyiW-At@Z$&^<7}pn;0F^r)zcFuqwpc)J9)M5C^}m5BP62Gr4E%Vzbp3
zsVo5AwveiqN{uY4$PG6dwP#^jRgq`J6=>|o&Fz-|sV(@|#k0<Wx3L_CEs{v_Y#OF!
zyJO485S8D2zt&p9Q|bB_$ENJ=dvlarJH?UVh?Ma;o(MTgm^;WY?#;I;ae`(C@qg%2
zqZyYfu+AI=k8jgMBX3iiQZ43#D3BRGOZJQhYEP0@qg^XGcf+!OXnu#<B&AEw+S3&~
zW8#mGH<O6PH!P>gAx-@fJC>B{>}^ZAGU2o!<Tl(Eu#c5t17oB9?dfI0_8PQPi|Ir(
zm*b7wW8-T@73JN|ScE7^5l;w|WN5{AHRFG$@_u9|^+`4=o%omO!57-pOWg^kE9*I;
zvzZ%JmuQ+wHad|K5M8V%!XBKs1z>Dmym&%h^Xy3#aqLzF9E<H<S$Vl*ho0j8SlDOM
zcwnI5Cu6OAu3KS5q{a9DM}FXeY`Hh`X#Zrm_8(~KnHT4Z(a;EO^4sD~#y?@B!;^km
zh1~l%^1NS=lt)Ce%FW=Do%S&4sQl<J-DCP{M|{T0;|OI^zz1J9tcsOI+LdN@$d~e`
z$-o|?6driV>GNZq2^ft|3yo|Sm(8ee|9cZ8;Ypp`>L{k`hb$wXs>LwfCS@$W>C{5b
zW#93YKpaT<7R#sD5>y0UB;TO}D)(@x1|>f5(Oq8oURW0)Ps$*qfdbU0Y5CG)jaJf<
zcCp)D_cFDrrc(tDV(2q%YVG@$o)~bvA()lNqCtyn0Q7NE&mX`)40_s`bk;(e5|&eG
zZ>D*)709%;D0s(DLJU3$3e?*@&SH|5r(!bF@p+vtzS@73)y?-9_8>)Q^U59w4mTHX
z3Wmzlj>f(_dEWMZJOBc2&A{7ukhIhXJKY4SOZc1D7Zh}kE+dVJ<3_;&PUQ;NY1sI+
zg=;LfY+>~j)--G%hXJxM_fl7e7n6v*r&5n1S8x(Zg-G$WJ;0J9!m5}4-&rL0^!|o`
z&_)6G<@S(#C^n;D^Oa^#nr3t7Ap1zEMFF$83j*m$2{l|+iyY@a^t9J~ULJq$5~%SM
zhnxhEl!QCO5+Y-gH^zU)M9%v`qrHgLL)ll$?8Dzvt<?0nYA&0oahYu3^%Z?Ri%Y<R
zi5_LQFTqUj7Rmd|!CeiNRPR*ZL!`5;OVbJV?UI!`3;peu3-5ibr_M{O8N_3!o{`E<
zt3TLomqJPllJ{9o!s~a)u-{csVvo;nC27T18g<z3c)dNfPZ{#ktmj?ggF@10D9Vt1
zky0nA<cfmSRgdJd;8?a2Q2ul`K_{^+2nIiowk&#ZVOuSRJ_zq15`;FNSVK@Kks`GM
zN`L(7ZY7a+eQC#HwiQ`mosC;D1N`VBXVX;kIyEo}zffQh4{VVxO@rmyEXj)*<2E=h
z-k~h(m)U7h20WqXy&L}EY3AUK%5rXahb;cbDCZP4%WeyxnjqUh9O||DO7zATM@3p8
zH}u#2D?m3d^Vn8lB?*x(@z@V>pMeO4_PUU_Kcc>FqPuP-nctuH`=h9*^gb3|aj|bK
ztI(rRk5CB;{An|8(t`q{6KHx@0TRggCh)Ek0Uh);mQK$BV`l{&z4rXSoD$7|<|5X4
zF9<vQ<yv{O6Z3o}<gsIlSMJg*i)JbboGv~%StLZXwiGXFa_31qROS>hoKhjP!-Z=+
zzuVIu9xh)6na;Bk&t+}McMD)WzPHh2OVtklHTIHp|8hUZE%zv!NhjoPy9q>S5@cLz
znL9Y;C2Rl~xU_XGQdWV}w~F0+`at*XYqo(ZocJewO>c~JhXC`vJyn6n%n!OMhZRMc
zR>7LG-#00*ZxgjIlrVhq)@UReM9?pAwL!0rm;;si{g?ulqG8J3ZW_r+CQma{oAoy=
zik*&M)5OiMn~RNZyOv$!a$+g&jq?ow%gMsz1kZSQHJ>6H1%gSeHo9ai9EF{u^bT-b
z5vKvfo@*k=k#nNX?r@(%-$^viCP34y4hfjAH#`h*;9+Fv+J~!hw$}Bf^6}mX)}Wg^
z6hHM}$OOmw1n`8JB*y;~e?`X_C=#;QW*>uye#eOyHbQ`$GDAG*tB{f+ANWk|Lt0vO
z(209rJ%~L}!kPRKVa!To={r@--euwXWNn=BWFGe%x)RGJRv;<C<X<;<S$^s38i%t)
zKDM9^B}eP{Ve86vV7W{_zvBYx(f8vIEgsz5zoJJR<}&5_qh0KaCT*jpv=+fEN#aEM
zdDPdPV&1QjwZUTam;GR-nVi8};9et3)6+-uLAxm3`zEKQv&`IY2tScR@cpDK@lXej
zhEb)-2H{PBzfGlTy#zQ-61cV_m@S@UjL|O;)q>I4TBZ0q*`qb03a3ABGI>SR2aXy3
zhu(@qmO%%;uhA>j-KQg5g+ik9rO#wnPotO;ds6#|=M%E={?Lv?OA%wrXkhQYqw!Vz
zI3>aX?hi^0xgE)#xq;sRCW-i6!3;#IDnFc`JPb>wj;C?LlcmU*gMC7-)0Sr~FJzZA
zA@BXS$RXqNe8zo6oKaQ5iT14UNj8t)WuL^;Fd|P74bV^8#e+u!hzbU(Fhp~H`i<*r
zB#5ZB#8yrlCj+-PhlZl3^ia4ReNhHO0Zp>PYuvd|O;yw>OoP{5Ul3ntXz{743Ojhr
zshNTlW&hqe9C|A*b^ggmjJrBY#B*%Cgp&EekT1FE9ZEGeP{o1PNqe^FrSO9Ak$4Y+
zss8q@sharYiq`-}hCfKRCzUOU>4{I=V6+!0N)Vag+d}C}@(1lsQNI?`@MqYZ^b+`?
zd{huC7LdyOoVnw7-fFo1>gVHFXFf%z&<q0JbET73>zf!W7brrfNer!&XK?}w)zuIv
za-MsHa?p42!1Z$k4Z94ly(}bLY{~XUEzBD?#_b}dYOd&5JSYq-oU5U9O&5KSl7_wM
z82pybE-8B(>7`uHB(At}MYZ{q91bii)Hwz4ROj-|*JySPEDb<zTrIo)@o_2wYsh9o
z{BVGJg4AV7X|YAAo|4SFe1_1Qv_G`_L-SNI;*rrWDUVy71!|#C-^_d&(z&1RrXCnJ
z?S0Y=?3M7gpw?QVPkZaRk(EJCqi89v6G|vNmX+-e1ePacgX3z!Y0NAkAlUwO{*-vd
zBNv&#Q@gw%nYVsZ6i}qCy@EVNA2-Jf59xk4>Omq}l`0uxD_sF=%jtMU=YzY*@mFUF
z5>m9JiDm-pSV^kyaF#C}Mgo4OvN3PP<Vrlz{rBa4FkT;lRgglpd%f*A743FJds?T*
z?T~FTGV+byny)uJzL_|M_u^RW)zto|)d!SJHoOzks(7t;k4PS>DeXcOC6y@s`s+S6
zI3E`;TS@o}PP6#4?rfB@bFH(_s0CW?IkG~Iv-9!j)5wJdz8F#e<k^~OFIfr2fuSte
zK>MHT982J5A$!{osDz`}44PekA7phlFO=@rtQ<P^h<VEs425<^rRdVl=kgy}$BP6k
zd$|D2+omLbJ5==)aV<wEO;V?n=+btV*C!~)O9XJbNMDN@_0`(u5`1!9ENx-O;NG#y
z)iBj=js9@kJv*!<lS5@iT~nI!2O)XK(2I?dCU#V6+UmIiPsV_aO#OurVHCN}5M>=t
zb(3gD>4bgs1689hD688^P{hyb!oJasNSQ6YCBd;rq}2(II@`okocm)i8e2yX?ng!|
zN;{Q_!BMTu9BwGl9_PMM?2oVeKZ^1nBz0b{08A+O@Ti-LM7>vCE4ZI@QGqtt9{N~4
za7zy|Z`=2RXe`Rf{y~^$5uuWM2dfp;^|rtcmUh<5)w8@`@cuNuEEF50yE}8wUi+v&
zkS!y+9+ti03RM>6w<R@-7~H-lZOwB-B7)H1uTP}3=<L{-Qt~!yADbG2^7%>Ck86{d
z$iV$rX2Pqm&7V?l5Dw~bk{e1?mL=GfZ{_cpP1J|?=X=W=Gr^4x5uVI`(y>7XRi*S{
z53g5!H!dpg!*+dAwJSKvklAw%`d7y>Lbv^GURs^XSc2pyW)lc}QlC@fhA|2U8nmyf
z`k`uMeSlT>i$Wt#YgPZV7IFf?6?(#R{rJxP!58RIH_(kN+_HH9y0T{4pNxV5Qx9eu
zQ!6;G>iHnX1xzijzjl)6_Q>RVf16#K@!0YXSx&$3YjMrF*Y_-UCmp7+u}w6t+0SDt
ziPXbDKlXL0)39;gx-!$zGOVc{oR0-;_NFUJ(0~#dntE?Ur_{HV{OJ3PxgvjRVu*sT
z5*hO2rxd-uAEI4L1F+LEK6_X|$gn6%h6m*LkF6igicOv9B%HxHlk<AK3*PFNXN5g@
z_hLlH%i0ZM{G&XXe;=~d7xg@)y9{51!7iif3g5cWP9#CRl-}YceQ*>pa9X{yE6)8T
zYCcWD{3z|L;9KS6r}{|ARaXBEm{(PQ=%3X}>TmhubFNNy<LS2wgNg(dG6TDn88D{J
z<ydCE`}%Y?0;z%i<OGNx@XAqd#yUHT-QnCa+im?7ui=h9_Y+vY11+%J+TjUYa*4g>
zod^qEd`@lV1>8LT;yexr@~%wUgQ;h|V2(EGJsa<lL69yIUZl8RINbuvH3uSYHdmr@
zMZ4b4YEk*mOr3|=&Acb;6>kc}FHvj_XiOd@Yv7on`x{jhP|j3Bm$CO-9C-@-@aWX&
zkG;_D=Ka4r*gZ`eCS_(SA-Jfg$7mOyM-z}oYrcm#V>^kE-p|yF$Z_Eh|2P^+qR>@g
zLbU-YKpF_etk>PcYstXwA<=S*njodW(E|ICjdn$4qHqx0>Ld=jkzf+)fQY{L%EAJm
zX%cfH-s0qCPUiQV7joL2t69J=^9FTKXPZ@sKEo(|tSaX7@nW=~^ZR+r5@kfzY1OM7
z^h2>8i*d?xE5W7gqQyf{>-v_zeCA8PUW5x#Lc?6mmf|!5x@XnY#bm*20(*)f{35wE
zzt<<qaQm)+2FwmN*m-iK#r_pP%dejd1igtLVQ5UFZ*OS40nTVZ5<JPRLwnfGfJV+9
zo>`h=jf`92PE;EW<;2NZen#lsTv=6>K*)0Cb!mWmPO;X(-2dnn*{RO5&Yv&W`E=WK
z7QRiqRZpM<{{?Be(NN9!T=jt<r{9G$)17b;V9dI8I&w=>Ki(j^x?KomK*U1aZGs~P
z`&T01l^V~Zwt$ZZX<b%VzLZq}#H+nH|E{7eMN5PrxY$iqJScMCHXo*+>-y4KsiLgm
z{WW7U^>>$G%+Tkyq7;@PT_3R?O4=9STxxp|%!DhXHS#=QurElmq;0u~x<g9s!E>N%
zH;=lI=P?3>uVWp<8OI_3v?<?GWa|4MnQ+TkXqsUfANO;h&xNBmlYNNUf7}6AOr%#L
zXWz|f3Z4-AeC9(iUDqzDlsgZwE}FZDVk!H;j>*SM{8N-S^`wwc#z1{L1WDBt(axbI
zlIFx5GYwiG+R;ZjF?1mD4$jifV6xGY2!cj=7?bt#3&=*()ZigA+<>?sZ_7#sd`{y~
zt@78lk9}jHKF3f#w+$cVuhzm&*YL_3uYNE|A^&0DWwIlv`zMT31mvE9T3Tdx8b+9v
zNu9wuw7dW?dlhh-aWd@pP7s6VZ%9j8c}Sx!qMkT>KSIA~6>v0Z^|sjq<v1b1e|IYi
z%*0xJJpK80V#|xqdr$OJV8lW@qF}3r>7#DcRDK@Mt;I4;kBn>wQ^61uFY344i6g6f
z)92LQU2Z<SpQV{M&RHR&Wbu`Yb}{?NjZAo)`{s5noZe<GB3_}cvrxK19&eR1?t=3(
zGjBq1PeC`P7Fo3iyg623soP?B7uWXt_dR`Gxi!0XBd8Y-<7MN<tKJ_usYg$yZ24|}
z5%5Vy1c$c=Lj`fvP4kwM*eIzF$yi$M&eZ<N7c5d2NAYaMJj?l{&#J3?jIyTCkJQO3
z8@!qaafwN3vG3Egtu5^mhLJ@)Z=yk0slBD~jl=g;|La#txmuO&DXqG?ZSs4dGhF1u
zohy}bomO>j4m1R?;ByMaP)aQbG%X|$kuz9|XLw(`FlG!BMcJf!3Q~PJbEY_fa5p-h
zZcDy+D6;O`GGa@$UjosVn%xgm2PAmLEw~~^+3U@L^@$-DAKEWcc)y{(Q%q=sEPZEe
z#Sy1PkOSvEHsgUyk4HcV=VSUKeTU6yQ0d;rz!z&x3mvo8l!}9%p3U%p7j)t=m{HQ#
z3DiegWLPg#%mN%bFDV-9Tpl^PHNWg`z!*0iTVk*Ih+IvW<=F%B6c`eV2>9+YWX`CQ
z>w7hI6t*YBl#KibdRVD9_5y!QvC^PLE+3e()8w5gIis5<4Wp4B?~NmhMRlzyj8f$R
zMuV?($M16xM)29u=;=}_^`bm?kus$y%NmtRSiUgnmKT*RRNaooUhFoP<vVg=TAZN_
zgZ?F8V)iN26Ez7P10~qgcifAy*v1yCp^+w&Ke51EHM^u`R(YP8CE*P^5C}i;H}BGs
zOLQJfa>*}M<FVCdDHJ;+jQ6US_`**JQK*w?&%DhVW@aWjC5~8a;V|~6BtF$DN0-2~
zE8=V1`63rv86QJ^R?;o>tE^z2_^8LS=B;v9UuS=3)X@2%!}W=5a?5<e#+dRdk>(6_
z_?B;$|FenQJZ=)ZVSke8dO^5+@-Dj=JRk{TV4nWH;7IL3h1D*z?pVjgBngjj%=#_P
zm>Bi!7w_4#-Y#w0(@u1^Y_NL%Q#oVYJ?sVDSvmMscCAX!g#<zh=k#oIr6o1r;W3h_
z3u=00r3>rs)54}nI}=>K47F)J{-#~$pUvYKGoy66>tlZ;ti&LC{rJX>!V)C;lOV7z
zYE&N%^ft7Pe#eNK9qCCz$WL@1te-T_)A#l}{xFeC`mE%*&=aNaq;!?l=L1MceQ5av
zQ!R<i<-+@fr*Bp}`LdzP@@D||L?0oL_{CiA<Mla2TW34CHPcf#j$jva49~Jb%PwbC
zq#qv63{attH4OzkNb1&+Kl?l(Z%=CH!>C93<r%hsyQ7CVm-~J3!8>qqCg7#nXL*5r
zL=@|e+T-nY--Uu-WktU!9d1f1ez~Y@6LSd>$#{FlbRziDXx)3XY20|m`}g=gzo)1M
zLv<}B&L^#3&O@!IlsY=6!KJ@~;k`cA1B`7$V$%~fPnrg<I)-S7)YPTpxt0_ZotD{)
z($9K|ptFA8gG2<MjRAISI@VS(VdQ6eKl{>;)7D0zxo*rfF+64YtdnJvJBTtf7$l1?
z!Q%Hct_LTbL>vBb4gBE(_Aq?dxn*UNA%4FxFfhoXH~pE&)&;uq@s6Tj1vB~KGq(|9
zg?|aHz0wqhF!8@7e&{16TyIibV>=rgv~1<tL|avPBf4b2=cnoFbN99b8~aD<dkjc!
z5ru0b!Mu4m)Ii5-i;*Ux@fo)CFxa&e<=w8mQ@JcJ{H$uXFDp`COvx0a%V5zDmgOa`
z3&1ncwOtq~V@$A-9p#$&5`14N+lJxU-fX6uy*51=@Fq9NZ~7Esq+>GfC}8%Hatj)i
zhUB~hvzM+JCw?4`w-65|vlA?q>UfFy<~eY`uWEZS5NDHeY9e-XqO0h3>l^ijtco|5
z{ksV3mc!Q+W8=*DN|5-=XMfA80BzYM-Nt(Kv)Uz=qh-tMkPuOKKFZ}#G5%xp{YZTt
z0xZm&CQoqLRS;x_7Y+0t<(}hBWq+U~ISLk-ms|+SZA}&8KO`8YMK^LLDIt=%WyL0#
zG4zJ@A9HSIVmzr3Cia1lIMOT(7)_S16~4Tu{eA4-KLuxr!e?{esoy%c(V<8QcK|-u
zw@BNkrvDAqq?3H|+@L;o0fY<t@QR)|zY*(oaoauY&Kn3RiGrznTaEkvhpqABA)O9?
zL4dqw9M(3BsKd;DdLXc_K8$$)sOPbKxUNxCcg%T^?<@M4G!tTK0@?IK@?D<#4`O92
zLHyy)eg`l>*S5-Q&9?BavZmo$d2ea+-w@V62v5IYw#y?XL<~ZwWLXZ&um6)0JwBv8
z(<dXs1(Nse4in1yzx0LGRz{<QiAubwyx8`QF`o>=K}-}&u^FhPBLX7-8wB#Rg~?3w
z@JoKVW`pub47GY@52FSkBcob&{VxnCL?D~}-w|`OPbk9$WIlObwI;hyn$-{_4Nar6
zHDC?|>S7B{eI>coNh@CI5Vl<F$l?oU>3R-uJulJh|8gGLQS)EuRd-4BhqHN-dvnfP
z^#**6e?hQ+ue9x7v?Dt2z*Af99$2gR_rIhUpkFo4nfxAf2$5}BuD@to_VKD43SZU#
z=am2cAO9iL0B8_E%&X;w_S6f`uQGZAt-;ADce?>P_w<NqhMJmR)N|hg5Rz*XfxU~s
z@ROeXlm$uS_y0Hjbny>wMu>4Y_&;~|{|@y3j2puZ;fd&6kgO2?b07bo<Nxy!6@s-L
z{L=s5_w#>U>i>(VwU9#CEc>D9!v8t-zoYSgef|HZpZL!caCXIXfw6rU-Wx>M#2_KA
z$@iZNw}t2nVb-2MV%VP>)<`N@KZL3=nkF{8Zg+)H0;I8FDPxD{W;R2FU1-0cw!c#v
zfOpgS6S}^6esyS4SBL3T`&6&OPhm2_fO(FZMZ)%OAN$QsrHcXU1Mj%iYY<VmI%)@X
zP*!v^5xNK8gXE=6*IUXrmrT0g;arfb_ITH$vz=TzgMS0TD(T?bW3C#%g`-h@=KB))
zaT7m95?zO+rOks^YWiDV|KEMdfnM17ff!G;;_be8D-WOKhh>)Nkk8l}IJ?b48!iOs
zt86?}Nw2S7J3qIFv!M~f2g1l{2HHn-qa*(CYrwd3f2^2C4IM$ubUKyeXqc>Fv)j(V
za}UgP;`5Xre<~viG3*Nx{~Oy_1~~b0FZE{60qypt@D8(d4nORD-5(v=vq+1riR~HQ
zbwYYi6h3injaG*G_KUbS1RQMG#c;>Rp^3x^3m5!fIF`b~j7_ucvGXNEI7QZ+vo#S~
z00Tx!W8W#F@>++#deN!d%==$3k3l2rE1=nKZv=c(n*3~hd79NU`ZD!$hQ5<mWL*<B
zO%thjad1sPGfytmqfS|;$?n79?6pFUfZ%)Ac@*O5Ek#TF?fcn#pO2(~3EyMV!~W*d
zpy69SsiK6oLEq8!K>E`rH55C9qH1=U8v(XR9C+=0za8Wc?|BES1g`vwUkJa;sPm}m
z@O^5$3zF<db#eLO2o&VAs`yMa3SWh{`2nYGeDvE8*~y0ab7b<(zD?-*B+<uTPZ@S)
zW-pF;-<Gj&eSGnrA%gPVem@oabzib{2Mm0=W8nLX2z-c2O=Xj06Wn+8tD$^~rakRf
zk+cnf{S4un&P_C|(|y?^vLSZ_Ue&L0?%Gm34)wX`PZ@Uv^6!w||I+L#)L9s>O2uII
zI3AA)ILM=yg#3q1#`>S^Ma9!LjW@8qzCc%r=X{lL@`kzj#n4-e)a@IyPxO)6?djS>
zO^pZB$IiZ4Y;F(S#Z9tK?;*w9DK3>ALwSuG{(muG<2_OC4fHYDog-TYtnqjejpNfr
zAt#5nqhcMZw}T>vQBNyVF{B6nbBx>h+g!zVhJ$w>c1e%<Rq%Ff?RjX`M!PUGyX}qE
z<x7yJ&}u8Vf0O5`=QEW{vUpRI1U~qOFO(l^Cc#7(7Vuon?c<e1k<XtuRJK0{fjdbR
zU5O`=DNQ=%@0-}+(s`BWWo<qT)in#BkY{4OI={RBNy^;+LHKb}6kgbA9c>17^y572
zBIe<^heCpEr@$vQiFMe#ygmjTN(2VNKT;3*PeP2o@`!BNk}plJXG84a^T;LfXW<1K
z(9i4Bsf3P+A)af7_1_T5#sH!^C0OU4sb3SXO+%jXqOug+D|fdna?i}##9HpAk1uEg
z_b|duApM0kA5Upc>KClASMM9b5IdP5&<}jjoza^uXfS?N&u*TP2~=2O5TL!|%sj*2
z;#TmcEya5cHsBt|hcF0qmkf1JvnM~gzAvoeJ;?{Aa%oVGW|m^U(swF&+Keja<SodX
zFWIz^Zs9)_9KD;ej|Nx<Yclg>WjgR3MWf2N8`9t;YVUv`yPcpDe%DHli_7Di@?u%k
zuv)sr-$;0rg*v*9zrk6FHAdtVf<AZW2HrjIq}+>>orYzu?>RSEcqL6fq@&J6U$C_}
zg)yReNB9)5ESIRT{K!7pYS-*f6ZJ=8-$$c$9D3$TvrNTj>~!>CgQe^Hsp^M`MMpbP
z{e@676bP*Q8AqB=ZApbXmmu$HcLVf3V;#a<b0S;UF8!ybFDXk)uPI02ENjI%(PQ8E
z0t=?@Owoq1{2&m+z?LLr6UhF>y)qV%u#mZctuh#w^erG5rD+wB-oa{awR-QZOr6Xi
zPYvb?>*MlqSmumf#|YdFg@!$#WstIyq4TY&$Rrr=pW(-}Wn=;|AZT)lTo`3Md@_Mb
zU6V%p9m)yy2gSZ%R6OcIZC(wGZpTY2O}23CO&$`iU?j|4%C#S7?IT$ee-$n9^unz4
zh)VzJ8@x123W~04ZkGGED^$N@nc9a?wYE}qw3r-n=N>1LO0;TcA=u3{e>NbqLO_x9
z$JHsuCgAzp9t;9wYq7~h;IoOnlVox0QL8}H4qTb>Sp+njoCxP4x>hV}GN5?ReJ|=l
z4NT!ytm^u)Ym#LJ4;H#bjLd~D9LGk&p`ge4DTCo<Rc*GWjY7KIklRuxnEu@Gd8hGw
zlYUU!tZ+okEFB;|cB+^xp}R=c`q0VbyC`f(8G<IL_sF2P^6)#O*rQC|vG>hPFbEkN
z5JiWzE0PL;(tG&VCuab<oBxXz<3Dae5>l<+23#Fq2&-Rwt_?cOyo;ps*BJeSkg$a}
zlZh!JU|Y6tS~mA*VUzzd!1eY9M)_ba%Fx8>+LlaSfu@stUO@0X8Hi)pBKvqY5?yuD
zb6r=t#U2&%^-hJH0#ArT7zpqtj4#Jq%LZ>dL?N*u)Nje<SltSZL|XAWqd{YRnd+e|
zL>!kNc4J}<JoWoe;G5QTAz0XR_IJ|$W*Z&SeU6NrC;lC;i*M5-c24_*f`uFY#@`{f
zMWAmdXv&N*qt*ZLU30=N;y6K<mqV0y&;Oyv(gTfkBbzFa=M7w77m)j{W4>`A9qV28
zPw!LckoX!Z3i=nrFPn26EO=($&&mhx$PV~~opPgumw7h}NVU)k>2w6dZ8(M&w*HZC
zLp55Uvysw2s+Rn4{~6*R6;x|rOz+U=4)<iyul`P#s#AiFOw3)((%L$Swp#>oJJU{A
zXD7Dqa>LxsNhSP2pCLyp;?;N<;Mknv#0{M3#xR&&NZk0Oa~1NQ@e^t9Lg9R-O`N#y
zabYGG{Y2cU!pt4jxKIE|_>sO-Olj@$@Y1B2e9oa|p$Vm$+9`PU4()cJ*1{(Cc_;a1
z02Uy;Bvc<5c*1@Bt}5`ZnA2b0HrL&4Mrn^_d`**$u4g0p2)z>Q4GYLd{yIT???iUL
z6|BX4oOZwkhyuN5<*(_<%9R7ccqDNVcsqog%a9{TSDwP(Al#guH_Afrfh<9_Zi;1s
zMzrVOW$nLDe`{)!TKp=BVB4L0X7F~aI+f}dPT`2MF+F`*`XAB+*!>7jUE5?5Y#?g7
zgW3kuAv6OSO<EQg=e5)EYV4T0sAkC@SlHUSn_P1=Sn(|r-_)>|n$7K0=>^o;{3psh
z8h)|n%F_58rA2&r>z)jFWU(IZG|kfL_2?mQtkk<>V&IckX@AbpVnuAgOcLr$4zxN8
zAkS@cR&~3gcwM%^>Z4r<H&8eLe&t+^+NMAca;JF>`x~9Zt3-F4vylr&;J)!m-u3q9
zP8<gx<3n9yn+qfhaFM)T{@T*~2ls;DHXgKL(Bjdl=MOs;1DD$G#GT&PVeON)7S<o0
zFXY|N55fl{(GPuYQkRge1m=dYy><(4N`FjKt~T;fZh%f;@I^w>d>(k2daEt79NpsB
zKJ^GNLsX3--s0v;Oa(Z#jPIi3mncWqbUcVYMnP8i1|+r*k3PLviYE}f!3SKJZohz?
zL4b>j@4x~(r0Wp&jLbNwxfm05y#T$*<P9&lm%pYI-Srm3gml&0S0WTJz2=j^RPnE;
zk>dKFuQ!e_*>(1Va)vzUnGk1^Ge@~K9<4g+CPpp&q-0>NPdJ1^XSu37GydU_b&&*U
z3`ioq^4Yv=dg(qn>)!Q&I6g3f*{heReG~y1|3cldmsEV(ea60M@egeKO0-!x;V5y&
z<2&8E?1N!~_5R;oQu$cWv6auN_r^=CfP0w7X(d`r(f@K$4pr>4&MQms7;eb%1o@9a
z$Op9xoJZCcwh{>Nc-MW76b$Rf^ESNJ7@GlU`2{!1ynmvmh&@=|s1i)Z(Qk!-lt0e&
zpGul9SPeBGOcr;-B*ah46ta6|$TR6tJc%sLxz~4&8nj>)2i}d84V<leEe;b_T81@d
zi{`@3H{X0B3Y-;R5(b6!yVEZ|-Kp2HRK0_6%n#}nsZ)(m?pO+@qDoM7<(~@gc~61>
zP8&v2Q?xMbJBnQ>_+F?-49b(PagY@I0FK6C5?wZXCDyXh-abw{bMY~br~*n)nK|@%
zj~pZBfBmbCPq2CS#E=Vu+cfhYDHli{gHTMvQ!=utMqinjmuaEuPq-Yl##Fs5r?vok
zD=s$d;FV7EETuP(C+f<|;Ud`t-l@{@J0fFifY7qZVI7+(gnjaR`gQvf5&CPhTdGP{
zGJ%&F<T^U$NWgL~UPRgt5aPPyzl@h#ndoem^^&x5P4hF0?X*<G8<qWWtsN4T=Xt?p
zb@t-B$-E_b&t--CzTEK3^Q-tE??$(7>etyq;Z`X&wi2%^hbET}WT=_dPqjBUGB7I^
z-j!6VGj)fY*CrA28#(a~&$RFB>2#fS3N=QBEZNo5)B#<g1L%pa+fi;sb>?@s%qXf!
zfx?REKbq>YVY#53g3GFHv+G>L&-0zTH0cs)4z$&jPA-Bh6MC-935%y`%ws3_e+#?m
zblz+t1uh<xQ}K+2M+cdEkGy=C5N5HyM*5I0fwdB9Fid=<4AGC!xkfa_xCn9mZjJBb
z&2R1;N&?SrmhtT$d4jL^HxR=fWs3za^18tv5Jt$6M8L3h={#S^Kk#y)Qf+TrXkV`2
z=L)Ljv$)2B<BDGvNenMFJQd=VO*V^%U?;cEUjozk>`q$Eno;%tf^^y)9{D#QEk1+-
zxA|h9MU8tvL@e;>Wd1PqJXJe<?k1si&J#Zl8+#!;obp%}+-iat_`S0@hUD}i@j6w-
zYfKxzDf%APy18OCGT%%5TR(v2AYPa^NOPrCt(&h6aZTek|KL#;<H#vZf{~wO_qM)3
z>NGZpfO41?_5FnL%#+Kg&E4s*!^Fq19PXaJdeabPfd=p*+|i&6``MDyKb}}?%#^K<
zPSX%grQQ~x*9w$o!J1a8<3<gV+@ftNl{USDJgxCt5`tqt7A7{Qnn_!UPOmSt^T%GQ
z<|*hc2E1?MJrg1(*Liu0lRy__MKe6)<f+9y*c!w-Y;{meI#_db`7<(yx$4ihpn7{R
zKPNXPyd{CO;xYWgir0dn{I^3cVi~V(=45^xVEz|dO?4iTV{93skvqty4HRB@MuT+F
zpcdQy!s{=BqZyKvE_$F4w9d@JrMt(|QZUdTMvf?{^#=Te2BvM2SBoA^8z&H5p2N1h
zF)iB#qbhSMBD|vL^m^|^l!SGPY{Z_GEVdB5w|0&y$mUjwItT7b-2)FwJH_U@jh`j|
z9O|zOmd82&We@3KhWJ-L<5E>c{g~Pz%@NG~ejEjn=-L@N%bsCjB@)R!lbKs)K^8~K
zI_}4pCD}5Hv>+A=h8sjCMv2)9kTLLuiOx44Cu4<UEYC6L!81$9Z~OdtoeHWp?#%h?
zsWmK!kHY!1<NUo!wdr`?<Nf~f{QQ}}=-EqUzN5<ntTOM_R2qVSvT_%C!T3dHcO}f=
zHnjG#JVQ%j^Ws~RRnmoBzGuRjhyIG&{t|<_$y`P(fIik#EbXXJ9r)X-@zpxV8#aow
z*z0%d2HdMSX0gEaYTX{X$aBc6cN>2w=oB(#4qaLu1k=_}Nn*nhTz4vlVWl77ufj94
zDNoLXkC}Wk<j==G9U13z=V^UoME)km4<~O#L$KKXOE=H6lfgk4Z`N<OqV&-GQK1l$
zc9uQyT|ZU#S#6CNZOIFD<>$$B%<ce}?9IY$LYQU?5J_Ph`QV@@cw==X-Nohv)1w|U
z+bVBH-G*kpyqDiTPrE!1Yq1U=LsvaE^~vu1RQthOp!`$f%@>CZa~0{ov;ZQWM`zQa
zVeus2z4`3MqHLZYwy??GIrCR}&iNmw76khVQT`HYJ!g9s=4fMPoT|~b^r+$lRxOJC
z5)72-X{~$o>n4hRiW(|4OufroM14R;EYy5IvGXT-vCrc=-S8bgHk+e$%Zf%lF_N79
z8)fy9`aWKQer)yILn+_ay@NxvLU+l}-04Q|nCRiJtWWbB*i#Rg-Bff~+~!%oZ=^PG
zPSrEZK$Y3cAE+OE-qLwCvH6lNEBZR9#Y%W}WdEd3xDo9+Ps3w&o?syg8g5($-;L1c
zVt-Q?$aB}LF;)(9QDV*G``HK!U+S2xoQy>Z{)TO3KT!Ntwo=U-PkP39;eP*F%UNPf
z(+Lh)GdxP_gsgY-YF)*>nzCMyzv%DI0y)^n$l1NmDe5c1`^)cFQbKL{r+^k}Zo&}9
z@#4A^OV&p!UxHnYuQ<TKClL}jZRqrJXPKOGN7a#U(IdxrN3wja?42FcT!%Z-6s2v2
zeQirzg|@k(cr1C5Zgqkx8%`Cvre-ii2lDtW<}bx-gk_4ib&A%R@!`!nU$9C<7^Buj
zjLR7i1o65!xI{^#&PIHe@}UJc%|CcjYg!%u7UBN<IQ#Z#WrfQCY;pG5&fu^Ks`@sV
z7r1UPTe!WOc?a$7u4$T&KYogC&u;5IisQ#+dMt_&+1feKWGDYP;8i$wd)&f{680x8
zC*<&~V{);i#(_Oq{%qP=ky{uVb8XICBb|*qMyxBk-or|Z1yt0fe^+wSDg6BAb8FrN
zG?1^ifdX1Z&dzY2Zo}^HT;_Ojed-H;AiY!PE{LF2(3N{th!XNAjM-+X+IxzYoqVT>
z=4;c9b3)2BV|0afQ%#7xw(DWrxTTSK-WyFhzs3G|Vhd*2gI8?H_qX3|)nk>L=uqU{
z^%E(Wxx`!^C}kEh3)p3C{Gl)w{#HBM9P_S$)8)DTVZgkassdYD#{wnak8|GKDe=8G
zHO*nE8?tQt3LB6t=kKn_Af!GMWlYh+$<<tG1|^P9I=tE0wgG1U;Y=Debty^^cd$tA
z`U$$Q&=9#6RDOo<8rtsT&b3SqWPg+)e&z8IF=OcTCzsQBgF907-U>(VI5#4%$~#dO
zCDVG32bGMzZQ-uBAnK+1fVOaXsXo)_f*Nuakt>xrH)i5mE~T`|VxS0`rR_Kuzsquw
zk1c?A-6*pUZ5?Dtfl<_L4h6##jwfb`(#XiQi6SVaNa9ApZI*O4{hqf_*A|pq{Br`p
z1!LS(`&}*<_{u^a)1uuKEn(;MytvRYxZ|4)@43<nfVvY~N94kjuROgIN=A<u)UAH^
zypB<$gPU*dB6|ftJy!$%k-8r%8keYp{Gk$xzO!(LW(4VgWdksDy#$_}vAdM~iY{u6
z)q7ON+NKe}`9UV{#OQYfT2uM0u)gf2;M4AsYhn;Rxd6e2@*6WMZkjK>QB6DyLj+O|
zvv4OBouB<S=j+n`9`*+4k3|x)Xm^w*(HNZ9bHE4W7+o&S9H$?hku-39t!w#K4zgv=
zTrGuk?sK<C`%q>b_KKC}p}~IMM5`P>xrc|-A}%9_<(|Hy9;X8Rka0>-&2{i3p7`pg
z^qU@At{Scho>3<40)LskAQtGtN%uQXqY@W2?k82%JS(dY5%VDr<26EkS(oCEm@Rfl
ztJ|rUb*}VW&w`}W3jP@)GGkp=`Kq40MPpz;jbUQH-VZ?xj|+HzWz1t6JZs}QZN+ic
z#2YDg0Xtk;i87kMGEM^iFJSv=dF+snB!m2VN7OCUQ{fcszbQw-+lK_uVILwd$T|2E
zA${H%k=kG0#on`_a+g{&zO+PW%Ce2T&aduy(Ikyl^<rT8v?5HSQw#p~u+hWU<Mzqh
z(9pEid6alGn@4GRAGTY55|Ke~)zhwDsDr8?R{!%Fx)+=OA?DSoe;N%ZiTiaH8%sIo
z2G^JnG{-vXq6LX4Cm46=MRvN39;oRSlg&#bpfKrDjJ3^iU!g=xSL24mn){5sHbL#J
zQ5hQNVK&z)UH0=$4OfdHSkRul9Ge;4F-gYQ1)Q6dvB=a!{b@kH$iXW{ZduM_3pb?n
zUV-4VRc9X*!NHcNrOFKDMO5wf!5q4@k48p;Gg)Y4p})fl6gv~~O>Dz2hjWanN#iOT
zdB;?IP}xWG5Uq8tGKXH4AHlicWc37>0><l=<S#`FflowFMC>=IS_5%ugQ%yG9r=I#
z_qHkmzTG!PswM-S1xMDkPy`LpOA*aa$a-j+`Lw<h%-OrFEM+@Ja0?cCF*hG)Ar`Qx
z+5EmG6-ZqudGN{=prE!svd?fAbewIB|DpEb@#Yw5i8_x6$;-g;r1(~MCP^5c-b`ah
zKYKGe3?V=2DC2RM0T*Mdz-cxLDB^%ZBb)Si+W>;^Ew)$gx3qmp_$@^iI?Iq}NkZ9`
z_0c>*y3Sa%)}PPtiYX))ivI)q3*5pz88B>}C`2iAMF!LGCEtGE@SUU^rRi<J$2LJG
zUAq$GF}khXoHKp$Logw~Mp;)LX$Y4oyOC4|V3GxbqmuC#HZKs-*e{J}DC?0!W8VO%
z8ETMuhLY=cx|8=-sM(W#3--=Y)GH?S9Fh08NU%%iS2a}CZP<vW|2#&j5ES67`WCj9
zOH=`pqF*_zD4v`AL<u!=GdK3}K^f?J<V%&@k%>TnC0=-gYwjZJhjeh<j;;yUZPSa)
z5HYFLY0EXx6Zm`sOJjV>#7nm!?y!<}A^cE+d5w77ritCquI)L4Ql6qyR9ECE$5u=b
z&flHGDU@Ncyian?D!qYiV%rR);@FjRMsrNMO6nKE1C@hDQZ*+eyB9D<CiT^q!J04`
zsW{+rfz%}@b^zT#-2<0NED8XjZn!fzJUDkmdN<grz2)XtjLsnew$~)~H9JHCN9his
zi;PXDAcswCWIe%ukRlsV9TEX!^MlkHF+~C`LT3ryfuc8P4>%{Mimd)#Ppk?uISFPN
z2unNDtOKsuW6}$3{*j}i>upaYl^&IL0V95e<b-gS8%0D}N*$k9i3f$7T8n09_dn$p
zj{NXYNVY)_{H-d{g%lBTdI1WEz(A%?i!p}GN^(onc?2mI7lJ1k(&$cSx$DF63{M15
z&ds-NpWvXOR#u}hy&s)_R$wf^trWUcQrCN<?kC`q{CPOyzH8Bt^u;y7ocW5mtbb_e
z`5i@w7G{%PsEm0bfj$xOwN=<fvaiAKh(4lnomZI<6fGQwXP%#)fsE}(jf=$6i;QKM
z1<+LWV`Hwftx*Y4dji!x>u;Zo@eT#eWUthJS$WR;)muW%U*V2~b&`d7%%MfpdZyPN
z71zVu!^^eu8w|%~UY^3DlO$1v<^w?QqhdcMX+&b}QaX@7%&86j2}C(7y<(%yjx}bj
zsxAS@mkaShpCr~PCi+iOFtpl%*;?M0fB0DOHqrli%kT``DFy~|%Eq4kcBYMM`i45c
zBCf*=XUZ?r*AM%OhKT>IY_g3YF$74P;J{ZgiM1mQs<p~{b>#Rt@kE@fzVHCI_uEXB
znZc3uG@V{}CIkeWZ~6qmHWwf>P!5-=aeUX!B@E$CBls+RXI(itM@um8w-yhEv^6e2
zODhfG>xL%%Q*Ipk3UC2Cv7?tEbh3nXE{M7%?N-px@aW@o)`$#Nx8%%U9-Ms=APhI4
zPG)<D?~gaayxr|uzyIC}(J4B$H*vbiD0PAWoxhx=Akf53E_>}$hSIGtNGK<7T?Dp-
zUO<u<%8lRP5fi|3)bZP~?9e-dG+D^2TKNIvfV*354_UY^hte(e<nH&3fE8{OdR^2n
z3}gg6cSkzeJr<#V1B7s)>f(AY-dog{!1|T`Z_h7*EgEDbxw%%h(ND#UdmQ-kf_r#J
zzdwujCuwYo2`|E8H${#{unKgX*v1E#1frj96i;5737rVXX6&fT?}^~b)cp!04zXPF
zg^m?T7FnKw^Km8fsc@}=ew$38&q{hd_+2qCDg#^ip|oUN#Q#q`hB!AzL{>%8YN^FX
zi-x6TvJO6-S4JtX*V1x@?b}DBR^P?KAi-MuQv+T2_NX8Su7m|eKwe!zj*7Lvd1>UI
zYlJGK4jSt@h!6*3eQB3IvvNkvI6;mWPV(^VI!PI>29I_Rb#vil{efd4bdf<aXjOOZ
zbDQL3+~zGm!d_OEzj^*dr2W*wvsV%JLWg4%a-Eb@fhF1jxAU3CvMqD-UA9L0Vq{BF
zlVTA>Tg35k81c{%z~fh9S!qB&C_g`B7oM3gbl(^C*Vq54uK~&8sx1*V<L|Ol;};wK
z*79^;G1G@^;@|W*r}%+CBSfBLZrorpay}#Gf1c-b2@2fXl1Y4{AD|uI7BRodpcRf!
z8=NxEGqg+kV)(gixO5-)sc(>GyaZYCTiQ2>AR7@wT?VZ$AQu+QO2<x#crB<%`xpa&
zb9fe7`u3%0#vXxG9-L5O(Q?1;m0R5&x3;)XVUbqABg`g~_WXdTvpH}EbgciC1^3K3
z&7Yqp)Iv=Q2_gCRd*zNn_q-=z-mbUM%^=c^is@*CyIp;A4B=-^1r9sh788*Yb`KAU
zlJ8aRr`6B(VJ?4wVN7s2Tv2G#^VQ$u5x)LP*H+x&)e2kSqfU6_`0_<BcvT>#*SF{w
zo+azoJ0dQnKZ1>wsD_caxq3xZw23<DK!<8Sd4=$hDt-exM#4;3o|%_#gB)bi-!&3z
z8Y4MOM1d>K+qa%;flT>pCrv-IDr*hLDg@x!BPJ;i+s76Y>5_Is3pqQ%`YjlZVm?79
zw#V$fU%or~()LOKOMX(47>nV)3b20Q5F8VUxH9wP_{*nZ((cgB8wDLu9(m-fV(|!@
zCwg0mwy07Z0>pP3E%9eYP1_M7sL{6&7Xmr);9uF0PZRCe|5`Z|vLv(5ajVi&?v&tb
zTwQxiy28Y*f#J$I3DN;lqlrcV8p_kbKi#Qs3Zd!^i&%he98O#xMH@H~cTCRLO0PES
zF3?-~2*c}8@+5H1ix)!Lnope7`AlndlWKVj3i3KUDjQ~Nk7*m~D(}dY03!x7M;MN;
zNAAJ&Ek-pC_3d)FAj#N6*qfmgl_EQ2y-WYt+F0w1pDc7_@Ykm&st7D|HH>&NV7U3n
zonh(S!`VnRe33bD^^H0>d{lM&@VKqm<s?r+@b=fpdcS~Sp55ZT!s^IJS85G8O2?K|
z#IOu|+M}XJ7BkWxqkvImyv#U)y}U^Q%PcNiIyJ3$*6#l9PI)mI>*IzuG7x&pwS>JY
zz~b1IFZxr)$zy{KH20TvX>s2Mb0V!8sg0(V!D-GXiL>?(bbc^TUgd{psOTB?f}qv*
zyX9)wfU1jnH9f#u-FBP(eAq<o@_AJ}kAB3F6|Npogw2?7v|&@yuHt4aMm3YDjTRtH
zGx$wt#&?qKkHy>)td-JIko|ypmcoZ2ad+f@<}=)SL25&*YV0lSRQAc?pjJV9{+FMe
zxg0yTGI%+sl=hk|L*hk+W;-Ma+?hIA7R0>~ZbCtS{1otwFrF}Tm7RLE6eZ?}S(_{j
zf_ad`L(Z$#8Blw69tmzN<i3+A@G``$iRFPsKlNvejv(M+T5kQ>21mUrK5TK%+3j)5
zWalVoy5O1T<mfP!_Q(6=BaHw^+UoM0n~=fDc>Cn7)6Iq(^zhIohu0`PL7c&Jt$`dr
zXO5^|nLLTtem53(nO%i-Ps)0?P+<=@O`a2{NIkhD*E%>SjW709eLa$$0tdeibN5^G
zu8Zg-erTr&#u0Iy_6s`Z(AOw}hPoz3X}Tz;P*z*ZOL`oZ1Prqn#T43Sw`m8>=ZUom
z*=kk48Yfn`<g+0ZRpMFX_u$c~^s6T;P7bqzVC{6izdba$o_f7G$~#T{{4`lwm*Au7
zu>-cxu3#*RgmFfO=FTguq{*n&A2B#I)8UK6Io8a7o9=yasAwAOCVNk(|7rtH&w}No
zCWF8FBQIm_C93H%xYE<UKaI-6E3~|Q=3$jTXD~v?c@Xf4C$e~PA??k0;VG9{BG2Im
za)Z9anX_hj^^B7OSo06J_(d)&RVD@<l>;Cn4~=3&RCGw`idWK}i1NhQCA&e^(V;LK
zTH-r1V{}mZ*9fTDo6s;n-&teLp^o-Y8MxE@s8XSf^wXE&%GK|(hqm!GA-ZIAevYk^
z`j3i9XKYVW4^KVtom;Eo+{a;!P{oFgkpnk#C_q%AQI&v1&ANkY@0Q9-^n;AMqJP3p
zN5)w_Oy-X)B)7@4F73Mkl@EP*hCIaHyVOn<HC{NhFVXaZH7}V6;y6oh-KOnotyCi|
z&CITxvn@Wrf6dI|TP4InShp|#MOf<ZqF?6pQnuW<#$G?x_}km|{0rCRAJkt<_Ckh&
zxNNLD57?NMC!cq<(hiB0jM?!^68DW;syI0~*0YL0U$}bpaMPZ&IaZOPnAj^!yTx-F
zqpE1TMh=Z1EByzZuN(nc%QS@dqkp&-V3NOEBiEKZvIfK|0{nh-a171a`)A<Kl!PVE
zFCI4(71Da~h_(ge@Yrx*ikm@-5nh&ZD%T4tqd_JR-<%()Div&*Gtvz7bNsnkzx*A2
zl#x%EL)+XlgUqv)ukL1<-W*P{c@DPEoId~=R_Ye{sAUBoykF}yUxtzM$L5)$jcBiO
zmV0_$Y}8uS3Avml8h<ZK%VB4(R7(MgY(E~2YYsaGtg#MY*{&>%3Ksw#K>moc&f=pi
z3(KV%YXm<-T3($h2nw~`Z!sJdK|gmiVcQ^{fBsYQ?|t4n9V1|w<~Kd<O&jUpiG4A_
zN3=F^#D!6fY9@~TTo6PW-|za;X$r?^I*O=}fW<1*vk=7<eoZsY4p0*8E`8>B;ZH!@
zKTd<~+*Df9x@k)mm02(MR*%96ODjvNtK((AL|(329Z$qpbo-?!%AF17IPPhtM1$)3
z&l(!`!WYj!ryq4yzo5T(9F9`*x;=4+luFg&_psqvu(%P7sI-{xtU5h@&yYOvEu2}-
zT{=?X?ZB=1g*Ov$djpv7G*sOV8u}xvA}!(pSeP`fusNl~>tI7)rCzBt=Ry^)(nee-
z>#tW{!rHybUO3=Vw^Q{Nh*`wZJ$z}`qtvNJZ>jBDb^60@RxgL8tZ4rFgyJ5^W7y#V
zcPtewc&;tK0NR7QojF{UH|15|d*arA$>fWB)-zufMKv@dYDZ1gBkBX2wm$uBE?D4P
zbNY6)(S9L<3LZgZbF*?&6)h`&l<(w+w=e$Kwt%3xs}TjoR$$oS@_*QStAMzg<zF-e
z0t9yn8XyoHf(|l3u;3xMLkRA!1Hs+h3GVJ1+}+(>2A9FkWbg0W`<%P?{omL7aGuu8
znx0<0s=B+XtA165Z?(A98nTmL7gJvAdK0}U)Gpvn=24k}FhA91keHTThjnd8I6eml
zQk_m2Y%aS#i95*=g&;^j{J8d;*d-#{iJbSks)lSR%9^75yy$usZCHhUE`*5CdY##?
z)|5Ucz86}O?t>|?%BI<UKycA$u9B{ga@;mThvQzK|FR%&x(8Cf;Osl9=30_cGbSpZ
zRRk@0qmJKp0_D>*x%QBB*7!+~Lffx!9CJvrYi*xw3Yb~Midvk)Ge|QC$*czgY@~}J
z6%8$YO)lW+0~VdT%RV`nchLjreJO67D_c83pi-ljR{>})si6k(8;MzrW5OONTrLyT
z1hBh{Hw^k1%{rU6zl7>%?rx^1)^6_m9E;yQ_{=)fX+6ZE(0eq4p?ZzRnjU|k)L}G{
zn9d;)!r1;UY7Pi!Chuo%UQLnIUxV`in!B#WWfuK@z1%uSSc8)+3%P*hA_6xLD&Dq~
zORcKd@*BwUXN#M28&gVTY%~VXHl0z}5c}!+gCml;g@KA%WOo-hZ$&x+4t5;fw?b7D
zO<Fnw5288DcJYRhDQQkJR_5mkn#;t)*eKIk<lar}NMa|Nnqo)asXJL~?kwEaiO2*{
zPpCTbwDcCVqTz$aC*+SF8zwheiRh(`aMOTz@nS0~5E2HSrqzP0L+A6@FxfPLT6+A}
zqE$N*8>CGB%xi>oOb3}ATn+<{ew9#O44K9%_VhGG_H~;{fujiYt2Q1T(bVe0be?zD
z=s&dOIm>jl6wGe*NcHae)@RRy3vaotf6Ki!HHDAIf1H%6k|?R3Xu_KlqkMBhg(azH
zMkX%XFC(CGOU9YElUHH)c2U#_<O5T`X#RNu$FbW>Q8kwCDVj7seN}}=e6BrvEPHg%
z($f!-@4;jl(&n*d_TKM9f(BNbNlgHSpeVn{MoxNT7Mc?9lMTo#me8KbQb%mUk}Phu
zKi2UgSuF#t*1}kY?sSF>cv9u#ST@y2-V}wFir_ZZB*1_gdh?bdUu=gs`AdC7c{w@k
z5tW1h-$&+7sYM{>0L7`vuO`xF3;);5smmJU+S#OBZ9)!ci_<}VOZG+e9h{tVOiL&D
zGTB%8v#RKcnQ)0yU#xgbo61~)4K?Uuenkdvr`^2%1htlr6GVQ+<9kCS{=m(?*05J(
zA;$sV_z1lJTIiU$XyEbva;QSTIo;4)V9jJcb^AmC1EY=Y>pzs}sP0#JKK?k_Hi?`@
zd8rc5gC#Pr&eBYq@`w&}Wdyt4T_c#NfYFYK_XM*PswMNv<1E4tG;l`awbq6RFAHoh
z;nz~)qsQb#1LR$2@F^a7s>A`&<fTqz;ihR6$LmX|BL$P60N8l;bMrFD>?jTE&Xt=h
zRsCUTf~_*7h>rGC2#=k+$w2Y*zRR1Kvkw*<=Go99PlRl!ZNRA7rDOa2^^oZkgP7L`
zdBW4!V$`$dT8xFyGHNwTqn*u`0UYgkYWM5v3Q7(Wn69J4NUPHm2-WYi%56s;Up>tO
zv{A;g+O{D*zZXey?+aefaO*?B$|MY#^cA#lB@tZNSKqTACG3XYBZ-KM-S70vuIc$e
zq>t$00FxE&Mt8p(sb6X%^YR@K$CUV+^bUYpEudMO-X+b>Q@;0L^b<~KVd5N9(^oti
zj)ICw#o1hD8$4=9;s*t>qCC(9Qty@%e_c_)+HJd>O9r0$GM7XB$UbS~MEML$sS$Nl
z4#Qo@);DdsFI7Bq@-$<(9zqWYp8`;M&6CEu`eh4z=stam^V}Lld{hwpe9{YPY9XOf
zkU5A&b-Ed_=PVA(W7p(fLcAWXktYcHimH=#<at$^$|ny`j&HBqnCC@PZKrlbBrQBB
zCuK9kA=m5T+Bdp_u^?MFos8K|8u4}Pbm=P7iVK!!t}NfNc`)B$mFO6T(Vm1T7_?}&
zV=bIgVKE0yiWisG?bx}j*sJvot5L7Hj5rP0d*4)e7^ebU?h>tn6(DGVL}lLFBhec;
z;m%f<BCZN*kH5Z(3+wYpFp|R^`HD_0t3d^B=2~mp05t@(EVq8gozH9u8(Ql-F!s=4
z^h<l(bvnJdRB?6@|7drS)wk1;e~Eb+4CJPB@m*+CaF}WHpik@In^5K=3XdSauzgh{
z*B}vTbmX=B;MBAvg0S;sD<$I-UWKa;My`=6d=ME>?YrhhO;c%l!;{cr;a?>LMpx#?
zG0u>`#Zx(c>n<o_dc@^1QhFPV=j!X82iJ=M5hj>yfz5h60pgDGnS`;Palc_h!2`@1
zwZgn+q&Px89(<VgcN?~qH+tW{4o%LsciP24BJd#w$At(eaz_TlB>Mehx|^_qbc;?Q
z-r94v(^l2Wv8fmTbiu+cnf1#^uzMs-RM@126(P1YQjvlDSz9!MY;}DCtk7t5jU3!t
zAc-zwsGzIh`};n`22Ac-WY&`lxuDhy=GPRG-Q}n4qM9xXf$Pmy4k1(G9u%jK(+RCk
z8$%_=YoOwfZ`;&Td98M;3AIv>T`3bFwQSwnRxcug2(o;R))5V$!-v!!d#5%!$C8J<
zyNg4uRXf$KFUvQTm7`AjYs9~8k4{fVD_NUFV!M?vBI_kpyn`kONv@{pff5UqLNv^~
z2aJsjW8}Ydau_vsU|QS=A~%uJUr((>lEN@cglKhqb_KtYyZIvU=q_!S;PtYy*Bs{b
zSQAL3o1yOX`U7sm#*sRb3vNHpsD6)A9t10&{DEYfh@rjINEQ5j&S|*1-F`Mf<Mjj1
zZ^5PF>FbrzYUVP~>VvpZOx1kNHJZ(c3MnSb7c3ck$Kdm?eQ~S#RStu_6z@$=?7vY5
zUwe4YB$QmC!=(*<@hP|a*^_CKQ8j2mo}+TjA(K(iexP!V>p(4)QAbX;l4X0Uv^h)T
z$*jxIX7k)v?@C*d;oGigPz;G4A|Ic1BW!-Ab<t}qvO0FPB45p}xh-m;NTgv$X}Y2H
zF5S2YB(dxgous`#=DG1Dc&OLGsn~Dleh#?S4&bCKE~AVHtejhYPhO(N?f|6jP2f(o
zlP5$;GHQzQ;i&8GW%j6eW#oy0i~Ul4rxUgZC*l=I)8;PYb(>9Ay_|Kvkstbs@;4&O
zYab$*u8o_%L%roSf@RwuhlDU<ne`as`cc~GzWa=L+&W^qh~!WU`g`*3)gM9t%QIwK
zVYk}E^<Ya%IQwaS^40T!<`(8r)Nz>f-)LMgW}-o!8C*e`+dNM`s$A!E=cT=un|UM5
zA^Cf_eH*>Dc6uYX1TbPhF?HeKd0$U&aJD=U`DjH;3qS{Rh4#n8NTLZ)5hEvrew6Z`
zt8mk(d(sSb2Vaw}>F1SVAdogFNI#9^q7gTVPPLH*YdrVO^~kIC!s6NCpSbPbuEu^t
zcmsPf`<y!$G*${cj(T%s5%dDsius<Zk7RHQmJJjiYNu$cybru(I?kiEOo_CRq#d%z
zG=ZJ!f)WEtldsZnvMMX@fzHU%v^{~4OAREkb^L1c(Gci}3A{Rf3hpdBy~&G3gKtU_
z6Vv_p9!i>DR3qA(xq4Fx-8tn?^J64<nK&<xJtmu!WUc}Aqq0Xs2;Frv5ia4J*OT%7
zYavUL8~4`FX>RhvE$-3tZ;#dr$rvVS2cec>6-{c9IWvnaN3dh1fmN&PacSk*yOK(l
z(e{{U-<^0;UfF?v4wB^+rgO1^`NL>^w^vzx$W4YyU|N{xO<*dAXVjeK>IHLg_qB@F
zr&n){qW?kceImGRaCk5{Uf?5-p7P`~c(2c>BK=CG9Uu`(z$(O`WtZRJicz+=#I<ij
z8!GtkvmocmgBMvVUe9`>iBz{&*iet+>DjI>kLPE;hA-mfqIynavwPja8Ujkw1%mM!
zIGc~5f5-$Qru+5%_;uDt5)D)L_3SGhjq~Wco;0%D`n%MBkWtT1aIP|};`v6v@8Xfg
zeAH2Vc4XnR-5vsz0lJjXc|?TSh9>^XF%NJ#sb_QNCpIy?Oy}X&Hs$8DY@iZ?E=P-t
zov%qFnKmWURZYw!*W`IyRKV-5%1+A9v^e^#ydBxB+49ovt}6XF?eN|7Ks<FyCd3<k
zsPrR?E%jd$|0ztcp1Z~_pWB8szO3TQ!67Ln5MY!?AYj>GIr{^qGXdI;g%5MzLrGDf
zk<1aLzLe@WD`e&SV(M714e<sipMy{1DL&l%#w>G%-LJno{jV2cT&PV4RcYXp&VT*+
zubWTaub?%S-Z(PF{*#1%mdIZ#<=+?Gf<NA@Uvz4_|Co>eTuA@^`LF+E>ip5>+kcPx
zefl${|19(WazRmo4xUr-!;m}cYhgmK`itRnCMjXH!C|6*{zFiQB6CrhA!cc=O(cEj
z`#%Rmko5=o5{Tp|SaCgT)#LNRt@C*19^o7zK5o@Be>4k?;{h=dxkQ4q*VM^(1+0}B
z*5qlGm)&LR9=Wf{g)%kRj~{=j)?P4GR-canlEq~)Lgq?vr100sdr@C8@f?z3W<2#W
z+3YlnJ5Sd<^$}D;4oEi1$_BO2UtuM|iDm91vYS8ECEos)YFj;e{d?CF-u!tplvBs3
z->4Gf&S$=EU%B$RTc@Q?<+)C#g>?t568!7_J29{P_xIe(+NPjqn$2Gh5y3NZ&(lUA
zh&SqH+XEP9i<w^gp{K4hAVKagTKd0*?mtA--b~afQ0XEnAtA9*o}tAYJXx8$SrF-Q
z4h<lpPUa!Q-jbShT32^^vN~@YTRYl0*m3d*AsF*PB-?uL;yt>Q0%T~bte+ZKz1bBA
zVh$%ea*w|^bgJL?C>Dv;ROZg#A_6i3AtOG$A$DY*YJBGgjz~nAS3RPy8?WC0gzG~s
z<pC$4&DquySv=-Oz2ZNz_a~l@C&=$rP3sg>8r98U``w4|hJC|{vp!SS_rgm@M!ht`
zNM6EbR(KhZn<?4!;3bdBQfu#FF6h4gXLk6r|9P>=O-7)dOiCM7+5KXKXp4|yy)n>T
zwI(fL^J>)-`?9@V11EcG&1M_sJlqU?kx1lnfHg7MsNNDlV%dX#^J?(ienub|eYI*O
z+xRNSZ__Bn3fBVt#pmuKyWH)zj`$&_;Y7c6-TDX;m*{f9@tp=mvsfDvb2llTJrvo_
z`yP$h5_hu{d-F8O$#rxWkfMY40*ru0;}yxi-I>j=LG9>l(&nW@W3Sr<b~`^l2|Zpd
z)dyaOX|XN^zeki5sJ4gP!*0%8YJGoMvFkqUMb=(gf<vokAZvysePC+C$B3lKz@U6N
z!f#C#4`srR09H#r$}CZz7nYk}ovy&0%uR`!x6dnId1C;i{y`K1FPJds;)AbohJR;s
z=J1!)&Tz(XXgWl*PUQQRBP$@wSb~M1eSkSP(jW?zKp$XHKj!5ZAlirI<2}|UR^L0Y
z+8%$lc|<?5!wlP#3oJT$@N*thWWDSSYQ++oayWj=dShkh6D1<s%H{c?P5pwkGIviV
zGlj%bR4ITUDK}mtwB<g|8BuW*q*7Ezp?&(OtO*yqbkoZiw?@Q(m|6xg+D|Dx@1Kd<
zT>1WTLua%3*xXaHly?h3kg3D<5!FbP{p(fOp$n7|)wGLyL;-pA1OQ}u>Qw%teDsQC
znH-U9CYNs$Tk@5D;52{yPU=<Tnt&?^x%T8O3ApNMCnRK+a%|}7Doj|FWUSZDGP*NJ
zdEOd6>nd^r<XJF_H(WE#`mj2!y$=}zT@76jo72N*ud<B%wYU63YPgs9@1<Bd1o!T}
z6G5=np?DT{e`2e5p4zrnUz<8I>}s`yO^CMH>T$>gbDVDY9bOe9<~p_8DjtK{wC(LX
zA*yG$USlH|Y<V8!kxWid$PX|hCE2v)RkKN43&P-oY6agGzr(}=^}F;JKW_ILi5dq-
z7bj*yr`(TensnRr0{8-eb4Q!oi}H~3t;urc{9s$sW>0#w6JBx;o2!QXSMyASPPK^;
zrwq)Dq|No_OC|Y-N0TbqRc@P=YmA<m720>{yyWlXYJx2_vym<muXB?&4+%ovId#^Q
z8{NzX=c(A4F}??5zm!SGHV!&{irg^7TK$z6jvdgm5u>%*WK!T%40J25CykN<uRc^B
z_1rn(${CW-ga`Ll|C})Rkfp!&o%Mk?fWH@NkfJmcWY7I|UQxFcAOK2hnAFY;=3d4#
z_Y^fxeUhE9*3}(4k=9S+zT_RJ)jfk%$PadUK|B07{lr<4kpaP)mMbvIOyJd(KT+Z3
zg$XF`VO+Ztw{hrB{#wdRf`{OtO_S^yg13!z{v6@)*}d+V4f-MuCD*xP1@x#{_uYh1
z<)xd0=hz|#2rmq#89Y?I&-F{)6jgeX^4sy>ddB$B=Q|eTSJ96ralsjMl0qt@?4%e0
zI>r}=ZUYVU-2pU@T6}JCIc?nLJt7sjTW8NOzn^PUuiG{c&o?1QLMn_J6EJlR=P3Kl
z$?-ZDq~}S6Wam-F7JLk{+b6k!7Iq3KS(kid{I2RX4trGLJU*XZj@LZJLOmlWqwh=I
zm&Qxy-ip2}clM-R_w?(7JzO92jGEetwgF?3-5D1)kls}nmyB5_$FSG%ZpRPwx>;Pq
zOq_kHLH*r@3vN=^RJl>J*+>Ehnpvu8<m_KySYy?dwr1Ilb8>cK;h{3K2eL`(K#MEg
zdKM_UUx3SfI6Hp{_@<q%wbC7ugY?@#z2XUiT6e!`%CHSG{2@5MSkxA0a~#7XQKazg
zgLiwyCSF0(sw>ZMoS*xBzn#x*z67N;%LU1K=vOA5S88gVO**Dy)Q$<Qy<X(f46eU9
z;tJXc^Tb-UpFC`CMa5Dia_Wiu8qs@H(@McLK3l^^gTip<{kOa)C_w!3%=#J=k@#f-
z--ZhN0e#juX2c4fjqf?zlP<9@Hn-V8g{3eSKYZ}*cV9Msdn^Mxl?Zy=#<>q>v%?N~
z^)7A==W<k9=dyB7#}B`De6YXG)+X__JpLXTy-ni|O(o9&xV%rD=UVz<a4$3Z82Q3F
zrQE-3XL!Gn^uaSa*6Dtiu+ezC=hxNZC!cG$7;N15%`uwoP|WP_31KzA-*I&;lV>c6
zYj5wdt?%H;_u;Ho<E=6dF_??Bo@g)Z7Jx50-!@u~;sRgj&>b{e?i<Ms?|JcMNdrIl
zgyfq~-(;!UBInX=3ZWe1tT_|&n*^D=Q?|5yC}q~G^NyDo1G_xF>6A`bW<NiH0h}n^
z^mh;~F82~}-kt-%&`iCMgPJ2>q-lJ0osX@DkN&xG@E46|o9M7LTMw#+M@L~*E%qgs
z_qDh<u##VbFJo&01{lLijSH7Z7scovTk0A!m!mH-@+=tJMrW*^@bgn~!OkUUW|V8k
zk#WdYq0$i+HN`rVY&E5Y$HbU&v1ys^Yd;Pkr}yCS?YKOS2fLmGjqo(4q<&=#l)HZ^
zY6K{VfgOTWQ}dpPn#PN?Lt=4*lat>7rRG@^^7Lt{yDNQ0hgKKfeFgzOnh9N~ymnGU
zU`}|X;v)|-Y1FjEID?p{PajDV#ea2VoUo??J)<;hZ-6dqi%>6xXT$XR+t~x8_>6Pe
zp|LjoSXL9uMtAEj&_5g)AmZ*>hkYSuWS;%sQT`2+XcasewH2C7$!rUroQD@2vda@L
zN<hg3pM$-)@)-0oEyH!R6uEvjralRQGbTONRs>?-?62j5eS?y<TSy1_t^qz%JG>-l
z<<s^~1Nj#QattZ(cM-D$dq+Pf9r$El5n2E?$JK^5IM;XJ9n484@}sbAW&5BAYPd7;
z)>ME|a>Nj%3-2XzE?EA<vz@!=!6H&)nR|L0f^7|gcX~fV-vZPo=i+oP=|VQ0prq_N
z6{SGcm^(q#&199KIT|gGZS)E~FYOZk3~f)md>*O7X<9jxiD|y<NO~l=v9U?aUTdjC
zw$Wu22M{&?>>Qr*!^e2ZhSTlMd;u!@;Ke-`0ng=m+ylzn>lP1!<9(4a*-1yC87_ww
zs9U<xe#2aVxl7G#JMoc%@ua8fyssWKw1<2z3espAFEMi#NMDXV&#1Ft%p08%ID6yc
z8GBt+i~jkcZMXqk!&RlA92>^AngVPmJ|6M0UBN144%AuK*Nm^@s7)Ih*!w&cftpN{
z<)NSE<q@W3q>CyqT##SXzIV%b%;RIBPWZU9*f4Pce?)1_x|0_qX-IF+1JwUpzlTTo
zqm&Wbz9UyW496_lzzgIIq?mMU(2Mm8cHO@pfmyp}J3TeZt~$>s4ENX&^3A)qSVu8e
z@`C^*BzTWqGbOb9Z5KV}1CbVOb9(wu)_cqEN%o=-h`eRlhHPR&{HG?&Ic_~x<=Ssc
z9Onx^r#Ij7yxh+>5)FUhw8l|{BAs7|^MFM^dH{wyEBUW~sNSGOjOMuR8AV;>1fZB>
z9rTjN5s~!kfjSN+FH>H&ju3gq(p`3#Nt6GehQVzl8@HYBNYa8_#t|FRm8~Nq``ZH;
zjdFE>u?hw=xaD8RX>q*>a*+#n;=jo$ep(tW^}K0trTQQ?841CwaDaOAbM3T2G_sF3
zu53tyb@+ZPAy#%#1{O{-eFgqL=}>0=o?|6SV3IO2Rk|V>`OQ0LKqpQ_N>=Q~_(IKt
zZs+);V=E7vB2vklD9NOLJHES@$N}3KWK34M5f5TsR;JcnbW_?&tFss>vO8&i^eX2v
zPN^xeYpDUhsJz@?fF#k`Gp?zkm%E-yEH%zc_u@5o@^7<&mjl%`rD`XsRcv4E^iw3*
zqQz}0p*4jso_yrX^*!u|=(yUg!6HZU#2R{eJM{n5zMv=Hb?EuE-itXKMnn%19(72*
z`Kmws+ZP^gVL?R?KLkucyHz!P^k>`(``@wsg;}2hxk)v%NwmGbWV$!`nAJ=%d<Bw(
zcKpnmO%BGJIX^L(36CBx^yGJi?20OoBJw;^IMZK!WCMG^99)l%QS1Z>u)J<X?Wd+2
z6QbklDaW*taQ+n@tgvJ|)g1;y!X2g<tC9%R11TsdoWR7BG2-l6!w#Fpw~P_N={%lW
z=eKXdp2$!%iVm>%vKb^)EYF0)wB=xOD1$~dbplhFl3O~h?dc{D1RuBPU&@<KNRajN
zoaxQ&Sev&U>bA_Qp&sIBl5n9HKZv+{pLheGLsFo?b4OVSe8j3B<;%L%!0C^yw0_hy
zF$r^bnR`Vte5Z%3Y?-YA#<yz=OS9%Y2nxhQS2Yg#C0=Q>yJ8F0IRh`h_w4TavfD^L
z_r<vl8V8u08b_jzoOpcV8(6F}thMXo+Kq>t5nC!}kc%6qkS$nxVz3f;A*+`okKlbZ
zsUDEo$qH{;T|Dt}^@;fV%==qc3<?4BQF8^!y8U6vbR#^D3A_<qJGT!a1eXXw?@~8-
zF<Xgu`<b%O;E1U4ZH1Oq&ERZx*+7MwAhom18qVQxmOPT(Nk8%@0%RUKJ(uP{#F87~
zmtFMU7I(eOg_#+HN4@DVb%8gPp-US~p^})+x5)2ZdeqPd<r1E#kk<N_5ek=a+QqI2
zHvG^`l}K{XxZ+5Ul1N&`&KK?G;?ga{3G7k;;~#u<hr3cp>-4IZ&yPJ8rk5#rS+-9{
zm<ozZtG0+oCiwF0eR++C*?StkB|1COc30Z{(k|z99FxV360vVVtfSBsci-g!(Y!&M
zJz&51oJiOL$m@cLI@9=O<T%`DrkEEM$HlRP5#)=Frm{^rZDz^(4k;uAfgAHNk`?iu
z=ul5`JU&#NQ>Dr4Fy@6K=6TyzW8pUpLkT|#*j)C6#;?5jhziU*$)<-eCVpP~Wuj19
z3Ppc8q-K|GvCkuLDGff=URNEMHy?jqnwoj9Ay>CM%5mUT{o|S^fiGtHHO{{45L*2!
zZg-!5>N@|jEVk*t2le%Xr#D{ljvvUSmuHY!PGa<reeO;k^dnJxuv!h4TsTfl*4v3a
z#)20yJ%2faVKIDbHXXM+8=dkUowz@9Fiu>tPX+X*-UYQee5)lCRge|R1F14HkI7IP
z`8zQ~{x3UucO39QU;M-4uMUws%d!>^bxTo(UORHv-hrfUruq-eBZU!lZCTi5NO`6_
zh5-+`AAu~4xeK1Z5@d0xqejiV^CPra4@2m65Hj@G@j65E=#A|ZL>jJZINlv0F50Th
zZCp${z7{y1S9nG873V~mdNF7jYr*+Tex8arBSNIK<=giI^L2(300Bt`0}_=J{U2JW
zUc$LbJZk&<rW)0YjS#XVv)cuBX`PC1v!rYZ71ot8VggnDj9+}p>cyI|-J`(&GUWJR
z{jn)YR$k*z9Po-wq&02di|y_8<`<W$J=Ywvswz6iW3meI79F9(&AHg@;cYP;Yv;p@
zOibXV*;*VJq-SrQ=u|syQu?rZTFVB9AKuZ0e)N>zD-(jDrCTt8(oRbHKWXQpee6{@
z$y}6RN`BZ%RC&o5hVe_@?%fnqpslMI5D7VBj0*Wi;i-oM{QzRkV`0Bd11g{2L&7Ga
z&n8Z=#K~@pbkEL<@|!655kb(Ls?M$j-B0ng+3~%^-&uEs=hD+{*A?;GzJ$*_i(}zi
z0NiWS5JhC)z5aMAY6vBzvDP<5|3SZ;kl=AiYAz+yD*eSej@WwlbNL#DfRrVFk<0ZZ
zI^ID}o}^pF3DFbzX&2|E<dM;VGENt3mALx%VXcjkS$n+>8@!(>*@6?S-}(If5v9Sx
zrDYZ7NOMQAruQ##{Nax{j{SbboaNBh&*7Jw`wI_VocBj|?HS8AqLsp_``#04(LJM0
zyB3Jv{R}`SPAP2H6{tnh033R+*63I%ka)?H_Fkc5SaVCW_CCC(;<|(-A>FanL`3(u
zD7=!FQ6y*4mYb<K(;_18s-0d;FxonP{<!0KJ7;2Yz}(qzia%CBsVW%9%c(te8?r$&
z;*90TjOKWwV&!uXe%)bQ{Exj}9{sXq9`=(GLZLP!$}H3Fs1y~m;N!ISmScKVwpy8w
ziSK2+dNWBE`!`LRqw#7bTls?z9v}x%I6Xix`t;LJA8JX<*nS)+f41q(T{&}QrgVHE
z-vsw;kN2Cl0l8o7S#s&2F==@%xN{qMbA-Y!Biau+d|M+V)USV%yq_^7Vp=p`)jYu-
zC=JXP&20U-{CVedYK^=CotvGc$?PB2wD?jugqwfD!rj29Ht#y{{M2wRsLIol(P%MZ
zytk;cObhl$4_I4y_LkTM7;ou~H1TVIZ|Cy6cyM<jL%ZjPSK0@uxCHPWvMiz}P7C_n
zfBcI>fIl4CetoQQ;n4T7A)A^(m+$;!Rq$>RsTdnkM9k^XzRiYLbK+iJksSP(_ja`x
z<_Z&#FOS4~*~}8%S130l8$N%4nzEVGR<p(H<u^Np*86iaUFWDEN=jF&V4T)@FL}p@
zK7QbVbB-IIL2k(1aP`SM<|rV8ed{fW>bzNWAtLg!F8R79VUDfB<GX8?v5*sDg+JSh
ztH&kw8lfSW^B`oiQ(oX=sw@Ucp#*tDo>!~NR@*XQ+9t&DVPO)x2VpnnizR!^ANrST
z4t11QWj8l`+mqjge70w*X?u|eC!6;Z1ZF47H{q<9av5o$PupE4W{zU@$s4nysa+0=
zYLp$r86<b-wbQ|$h*;Jz(Pnk-Oww}2wv7~9{lWxQD45(Aur@kvGPGwRg*gnb*7;w?
zQ<8t%L$v&rLMq6KpM2Y?{!?m4c;*^jr3TcE8rSL3=fuwd_IJ=JCc5f<%Wh$;)rBIw
zI9uWvena@#>Ez7{6#dew{3BAzpn7xWNE0G$iyS$Qe$6QeyY??zwJZ7`03&Ig@FyeB
zO1q^2L3oieA)m#G@uO(~MgMf?6G0Njp2|mow1Cs$4f4d(s)98y*~a8a-{$Vx>v8d<
zQ4Nn9{xujcUWCP{_a2RcDV1zFba5zj_Uxgq#ym518IK_z>uCptMg0%Q4iCh#TqiAO
z#NwD!uS$oJwGDUEVj~p=zD!Ez^S95UQMswJkf4YGmf>TTrv)bq0$fOH5Uz{Rk{^>W
z1%{m3%s18!c!$2ySZNyZ>>UWA4I!@|?&LXeR_yAv)lAz~7s$94!k70G&EB$6d81sK
z0<K`!NJE?4QA<_}Jd=))D{)ny4_8cu{oM)I-3t>h(b^wp5Io9}1K5IFPo@uDw0IBA
z`B^@_$FDqMuu+#$J9NzaW?q+30fpVOJ0|f+)Kz~Um)$m;%_TM|lFgE~ASyp32MQk_
zN6>NnKsau&4MG2MQ^!#}Iw3>U>!GJ$9Sjf=qYK#TQQ%a+vmXquvhe^VDqyC%N^r4%
z5}-0P)1X~5IVP@U`!$i;y@B_9{yOXWqFIJ!0$tKGV=?${aKP$c{s36`Kk|b>9tG{Q
zdOv$lrQO`1e+=UVyYz?Lsh0h#@OWsb-LGFl%JFO44(1bu3U7E#IEYOyvI<-F#7zxc
z)J~=N(TIy8bd#z1dob|~EqQFekaj+-vZ;8WUVL5gXlySN%D*J%KzNh*f-1!|B-|%E
zpqwbX0C&Y-tCzHPAUk;wN>j-sL8&snp{f+poLx8QqvWbZkz;mmS>-`|>^S$TWqMBe
zUO_nTH#={4)drQjN#EX2(UH0x;0ajsKF2A-@_GESDZMP%AT-@MB>-xlwXi6Qp&zo9
z54t<f8ZZ-f`WDt#{G=?o`ab0*CahvBIl56;;47Kwtq(CWGm?cu-2OXwbJH#4?tPEI
zXk!arMl&dmYRmHYhC6V_-@yi|g-pcY4I1g<uc0Mo*{ANWR~5Zmn4UeA0V_JVUS!E!
zU=)_yv-l^jRBYPJ1JE5|>aHxdbERIx{@`VtP3UtZiOv62R@WIj>Az^&r98Z135bie
zjv1869df=x=v;3QU!6$jo3fU4{0)hcE7e**LIEye1^$%M(c!BK+RCf3uThzO3<h1W
z&nH8Zs0b{+8FCce`Pw%M^~ANjnd5zR>obG?X0}BW?UT^MRI=0fZHo>3G5A+1=;pYl
ztU8;9(Tn8QWgSt(opyCNUZ++OSDe;{zTJX%Z57_t+mlO^@qs~(%O|5R&G}Z5WACTs
z5>1!kI9ca7eG^;|%}oNfgyqC+gfCEPoN5f~&(}q?pjhBO@!A3IQtOo#C&(=(*JJfg
zeAvseg&AJjrE&dT-*b<C^KWQk%h^d3$XA}(zTdU5=5j|cFMv%SSMUPA>UWrvRXEm8
z&$dPNH4dqz<eAZXM`~u3h*O?@mCQ7AOYdAX(u)Vw{)CN!6K+@w`kJ(_G{$S8hvH?-
zxNWR5c#Q_txC`v@3T%Hnv!1;Ec+6n0J2a{9O@m3;AJ7qONrD9ud6YT7E3T8o8Qv@+
zd736QiASLCA>`OY-hfCQpU+sElr^lsV#K3?t%6jYBrD_Y5Pe?P(tcK5;8NF_)A3l}
z{km+80WE<v>UA(U_8jfGvAy~wl@)!s`aEn?^yZ!*cU{YSOVEOIV>4nEIVACFh`BR}
zzv0);79g#P+ZbvXi9=(uAl=fOa-F-V1_vB6@A}!=iAFENP88U(X7$u_paE#2QYc-D
zH2ek?A4}H6Q;%wL*sn!%Y0$Wo!jDuxfh>McBn^F*lef>cTNx64`Wcp1K<TFmUbsC*
zC*p5s6+oE>2WY)4Ve@=vIVN$?sii?G`%W@7*>;=`O>qnj|I42U2%h`R3Vx-0TVCp1
zh<J>jZ6Q-^=lgh*?a!um{v~2{`WhpE{uEDg%^@v7bTZH4kM`N%2=9_qog{`y6RXB{
zB853@px$Uuj~j~wH2Rm^h5aLUQNF!1X76mG!bP@X=PA51xNlZM7@;L%an@9GH96%!
zJ}VX=w${%f63Hp0Z(eo-3h@u2qwN1A&UbJgeFX_W;>!T9R!`gHWQ!S=m!gPB=?37K
zsG*D*|4==D501XBz#9&o3_)F{9$^eXOw^5y7w{Y0#okYA*|%{yJ8-urcF|T>KimOr
z4T6L&-b<e8&Io&5M7{Ljy1CLqgrlNDokHIgroNnKYHzz--*ww%f{vHZ(&&^9dbl@v
zfOVJ7JR+OZY9SWr>WvEyxD_NEHvXEmdP?C|xkmqh(Wn7wzzqFEUb8BGPAVGNB_A^^
z_IVe78l>F#a=TyQnAmfhFV)JdYA0)WA{mrE=Lw~Ce3W;Cq|zmjZViUz4cL~!cN!~1
zdc3<2xyl}G;EZD(l2+HbC=BX4fe-XAUCzvAsWi_y@KF81kHt0aGZJrcyRY3KYw7=m
z%^*MtF3pyB*6LCHv8+G+)vGrF5+Xv1vOr0>tZ>1hSkYO_;r`bQR630mI`C1q1>cM{
z>ShJ6G)pz#BKH(m$tM)OpHu?uftn+kH5P>Btj@HG%aRj~S|o)YvIQQ?h}hQoE5ic9
z1k+!;5XsuD6H?mBc?r`>TP`#(eOpTYwuiFVdSlUA<8-Llq`=a2zDyV>SbGKwkWQ`p
zqRj8M{E^N&D$7C%^SjAOYBGE<W*BeLFe(-w2w~AkD6P~%<R{t7L_J??HmFQ~-su(S
z{0JlG*4%Ykydl<>KDZPS(aY*z=NDLfFU&9R*vB9B;i4(e;xSfMRz8Q~W#!~hxj@<^
zqw=e_tN6l#Eg3xR-;S@U_-+ru3^#efGc>?e99?G6tx>`xf~DRmE<E+PDOWticqln~
z_A??|7Pa-%V)xU`O64Z>tZVE{47twz3(%vA`wu|RZyzJdDQK9eefvHA`Al>~5F@`8
zk}9$^-}#Q2uQPH59Cnz%+Vc)m_Dl9KdFdA^C;L;s47#?5E+Bp$6C~_8M{9;7lO?Jp
z8-V1Q%sj>%5}VOr(Nohi-WYm^pW8<9JYV^hbJ&l3GO1(KAp^|$<9I5}5$__5VM$bO
z(XA(1WAOq&^0!N#<u?>{zw`<pBc|^!U<Wq_3$dkV3Eq~J*cHh}_Xa%CT%czL0vZ?8
zBk;F9<1dfCVEZ5NiHA=*vf)41CxIm5*8mMz?q%w$+^$k&ie;uD(T2({i;pI_m!1Cy
z27pEN&vSL-h^I#WdTu5Lx0XY|-8v&nuKln@*1;p5m83}q<`;_PN7`JFGB<wr|Ixu8
zNC78wFQ~@cF#Fr`Mb7jZ>iDx*yN~A{zjmB>zC`-t<JTTkyTlK&v34A4FTtAAagbPE
z#%iCl+3wp+7dOi+k+pf*CDC@BKl{9Vy|zXx=Dv>neJ3pg8`LUo@SU0)=5I><g<)WU
z`!jG19J=bi#xpLCXy5u#vM)Y3dsVEc;}dkhlX%=5jesY^IT$@h&A=mj+7&XgtY9lN
z=DwY3OItk5@%^_J^cO|D{^+cz0?-utf87uj&eR}YY*a|)X7(fLo1NhobGIFpX7$hn
zTcnR)59(E3{wADH9yctI3T-egx|_Tvt0X()_<xKQ3TT3XlIRFFvJvrb8vO;(_}?#H
z8$M7m<{wd#r0_qr0t!NHSfE#*@H_s45&xbSEbf1hwP#Trc7y*zt8s)s;r8f)yA=O{
z9Qoe{=8yhouLo8c%Kil%_^*v2ul}g}F#Vcv|MB+!TI->|&xD|Z?-HLKru`pU36lN6
znE_uVh5YL<@n0TE0Q>_$x$92`82l}j{!6QU*gwD=1adpfe-r0FVK7|?fA~(O#`ZTD
z?!Pt${rvrc@aome%U-kt>fh;Ej$5GxPG%ga)yZKPQSx@y2^NN3Bkx9hoLju9d_7l(
z)z7J1#uz=!1+i|rJ=EZ-HXl(DE6A|seSQQ+p^#FM3w^qH^=;^F0j|{N`OhnZNZ+a%
zrWh%kn<adwNi&;+Dx2L<PK)u0XO^0aZUAclhoZyT)_PB>G{c}GjXf(c__%ru`0$v9
zHL0`y!uOgKCgP{~Z_0JdGel>&|NQFZPi`IT)W88Jo)PpvH2c#D`RCuf%Pr*pt(X7(
z8SesrLi>XjbBz1{=<iQ&DF3&&zj^b2>Gq%V^#8{*M0b#TIXpZZo>H7%K5W=HnUFnW
z1_y2`28oB~9ZD-mjOA#Vu&wvcg2R8^j^>>$9?|<=Mn)*6hD1g4W#s>TJ=%!sMw3{I
z`<#}pp`EteFnb$8Q!2B%o-h{s2t0J|bgcW!!DYN0-3x)2h`d8&;l(?7UvL;+rb|28
zQ$5qY2KH&&iTJM0bl~>Fkm1FB7>CM{If;3=5@+0l<t-t70HI4aBRIAcqc`~4_7Z26
z5hHJzu13e%rLoU<_K7;;YjsfZ+)dKt7k8!|<6jOk-jF@4pKs-m@>S3{J31MSI~52Z
zuDS_X_ofJ-KJS#=zQNbwqq%#Ee<9YieU|nSf9RW-_?3AZ@Pc8kcH6VwVPfW!;kkDS
z(|97+&`!D3M*4V@e~KuZw2EbZe;&Vn7($&F9?^DgJI<x}-pEqHI+nQQt`BWByB+wB
zz4bP0ee;sWpOnF~Wv!XZldZw%XjS)AX~^3kBSoXEp?(b!jrbjf8r3q@hmD@9fy5pw
zEI~OmeS{|WWLc|rOC+Y}vz{o*Rv<v&T}}Dp);fCIg;NLNr3#$DqOFm&yh2nP{^!RS
z%z~DS*7ZeCPthOM%eB`u9l>&*#>-wwp`ZHF3T+sFnh%|nc68}HP3d-dy)e_6$PUeP
zkUa?u`3O7>FVn6uPtW!C;m*@0-2b=^U?LY=+edHGaenWdcOBA^Jw;kH2Cuhd`2^eH
z-L6?0!gC$HzWCy5ab_g-%y8E&islxGM{{*cg2>|pMgiXDTEP*aJ=5KF28>M~JZS)5
zPlLlBuf&D=($hNT;+LVW;*NSb_s5pu?IB<EYeaS=FYBu3)kEZj5g;2CFw0Yg!SmBG
zkR2l&1=Zhei4;vkHYu-@dz3aG#KZR$z^OJZYu*SB^=+P8F*p;zB)GD#@hi4!{G?Pi
zjVVyB<u2?jVJ%5F8Y)WP32T*3bv-3$La={I)ks!G+oxZC`nJ4o@$D>byTC;!v+1B4
z1vcuVu3d1%J>z9p{6apVt>bQ%(uUXjfY8<{GA90Od&<-1GB=76X)0`NW#tc=bDtE>
zU1t{YY4Wz6{5!8h%D?U`+V-~RkED~VR$s>j{W5wj1>n5n_W>K}*!An8@#`~#mcD>H
z%l3%?DqXPceBatrBcA&AHOK&G#l(Z4NMx^cae?eZyWX86YLPK%u_UI|hzP6fjTmH4
z%A)KZv$}w0GA@h4@?#g!VU^9dYlE-?e~CkM#z}Om`>M&eSwlyaLcM5~N{%;ArEXoT
z!Ydn#L}MI-`+{x{Iy$n~sL}|4Wd_M7Xy!`zwg)BHHEdD=W;V5~^>#P7-Zx&lnzoF?
zHCZSW%BG`2l;~%c?&aTpLUlS`HDb%R9>zb|WU>8(BD$OrWm~DHVMinzZYNR{S6D1C
zLF^xotpOrPI~&`fEH{%*6wAPGAyeB40+N8j9MZD^*z3_KTQu*W;xNgEty4Kd+pDv>
zvWg3|uPl>fK+qZm<tXe}ZW<3|`L!PrSi4|zq+zq$p<zHb`slvsS^%bkdiVP5w!P5W
z@SSY4z0$nr>mx&x{K>LGwC}yHTaKd#Y5{96-BTNFPb+#(PEHM59Y;9v%gSU+wOq%$
z6;C7^c@DP(qm`xG`^6fg1N~w&AT@1jvCce4+5j$STD}F5gQS6vcBXTmIUV>@o~eq+
zN>cE_4>d*AqA4`mMurM0XobyT`PF=0?W+1g68q?c%Y$D906nf<VMAr#Z2d1(@}Rj{
zFV~9s7SY0lCMT>dq^gnFsHIj5eZ?W+zO4_OL?Hh;@GG4Odi{B5y}M^c{t36wTYCn1
zUntxg#4~>(N(Xn15d1tFe$r4wgW*>I$dBgvXK5H2@<;UfXQNzEBl+mRlpq8a9kaDz
z-29^S-MzM1XFjJk?QNGerV#*e597`1Qk`nwuo^gPH9MO$*l*+>)=&HKm8%N~6=zER
zDq7;4azPqUp&6Z5cR@W~(mkv1W;OoYSUNLb8=Mv`5Yf7TZ|S<Jv5_~svJp5zEo0j}
zyUY$`Ti&h_%C?_~QE+=@mi6r!|3-zO*dWYWVy)$>aN6`^?82A!rE3XnxljGkwyZLY
zwm|pDG?z7`B_7v7pA7}y_vM0CD>xozD^q%FA~J^1=^Udp*TwA1rFr|Ks8msW^lO9p
zBB8vF-c51##wiUc8ku@MC2`e~ssmw!FoSI-L(@#7UG^#`5UpEGEnFkx8$!p-O@ITA
zyu&2qNM>m!q12s8i|GVK-EGMGZ1IEHCw%fuEhdl{59P}TWSE*#IXvjT@<YPs1xSiG
zJs0ugHf2l5jPf!SCm^NOHpJ+(`uPX%m<S-HYCka)9J6B`1CzyVP7x5O@25Am<!Tld
zEJeQ&ntK`pv#iJ*13yu0n2}5TW%qYx;+?*aTUJd}43aVqT6<3Y`vM3#LWiQg_hMD+
z{dha>C(EO-5zwyV+t?*_VoX0csKx$=(rg1kXna>`$s?&ai&)NPFs!9YsQ~9rSzyKn
z=!M_3yvzg5fU`J)=($5j{2-T!b!32PPjvp_Yzx(Ee&cJgW_Fi;(b0#xjH<;)Mf0t#
z#e8C`@5w?|E7xM9hJdQvx*&#=imf(DNGOW!lM%w}s?Rsu9b-eG!YGgVHlH(BJ*djp
zh-spU;!n1@mrLR_F5?Ao!?!E0rZ?C48b__ADk{>j)Oyv>>^a{2njjA+Qd!W~I;{@k
zytDJNwI)NIV|I_Qtz&I7KrP+_KRe00&fRFq8-BBTEZJj-A{H}6&$7$0)>vw`axH3V
z=^Z|?zR5({nMEru`-x43SDGG+YrOT{I^zdgJ;t5GK?g-4pk+s{Q(?_PoO?Sk@KZ_>
zS1mrN?*JZ!wB5I%Mp(I*gFCTIdO%UuED(9ASJAN{=p=S#H!6a5qPMSE4$#)4EU#nL
zss~xzp5v8MG?>2g-YT=H%rMO{dyIBTA`~zOjl#cnxS<q%%9EePd1R%nkG%n&_pA$h
zK5}g%k9$t-tzsQRMzuO<fQfA``L<_PK4UAFV;xT^r$|f{R{J?b%1s$-*ez$-tGdQD
zSWsmMlbwSDp?d#jIN~M0vE$9iF2A2J^X;2#@V@Pi#q5iR&2RaLmI{vf7JVH-r}GIV
zn~{zHAdf%;1l?17#T57XaVOK_XmlM~aLR|!U1|@383{OX@zBBWad)3c02Q(tY8!c+
zZNwJGY6dZ`uyWHlLuV#^ODA$PynE>lXm^r13qsGoud|`pV%UE&NV!x7<eBYNlXM_6
zqhrC_$EaItpLu%NdW?Sp*Q2~Ex|ctASXv)myAQ>jQF6?E@PsoOee%YP@HFQ>3t7*e
zdj4?DpJ=0P%{m7z7x%fzSNGu;AfyT%L-u%?^4wm3F@}a}r)TP0IVqM4u~0T)UT<aG
zaj$qAf5%-(zcVa2{DNee>>y<CX1WuH?=6$!1kX&i!_DS>`w|cVucCwMAa=FSV}i!i
zvfinCN;Znl1VFdc$RWyxK*X|$;68YIa86hx<9}-bMyI4TyZK-?eYF;|vpeHW@Srfh
z_l-~aI8wfbND5&lLwepU)ElN~_<pa@Q8YFo<3yEE%|s2TF<1^9so!qJXkoc%fAiw9
z`cR7TeBbU}Yxk^mmVmJOP47nqLf29$5pU4mb4EDEBR*X9X!5)UQu7>$Gtq6z2?rE$
z{33eR;{uj%ztnD;bEi&x(0Nf7{WT-jVe}&Gj*LIZcrnZsfExC#2cvBS>b2i^r2BrU
zy-JGeze<24b`W2ZcQDd1d};V*#AxS3rUt`-^6al{s`8KdR;@V8i|%S?zbie%u;|V2
zl_y>|4l<*DKET@{N(K^}bK}u}dq)80>7-A>#W5+9h6%P(AdwNENpyIApwvS7kf$B-
zoa+2?|8Zqo!pSpz24|R}<t}e}Sb%9O?w9T^0(0orIXL6fs<_XtT_NiBN|ok){(7<#
z=;OR3mq|70F*S+1#6Swa_!<Ff@|c-yRKTaB$;xQm*dLVK2|v7@yV<Y%@Ar?Mgm#vZ
z!`h?<w`dXopB*Nrr*VdJst$yjQl57<hToESF2jrGW#n5`=Xtp8JM?B8hF_J=$RHGX
z6kAk_s7#MWqzk!Bl*UdOMN8+LJ4Px$<bMh9{{fr$6cwdMW%L9sb|B|~Sm%vdpG+od
zKy53-7_`_UT~xpO=oqC!y{BZB66G#YeyzaIg-N2g)96}Se?8O@@NBFU8?9`B;KPfD
zFv4K<s6u5Iq*+z!ybJcAmaMXV3du>>7u)dWQ=Nr)%DVof+udK9M$q}1DaGy7wg7g}
zx)m;u)EcOghU6B!kOJ<~*7@5X76QfEOqw1>jJ&0C&}Hw;6^V>ivmHB{;_MCB4hGZ=
z-dIyWaKwDJOEb1DW<_(d4`^<!o^rTgn{b<xuH-ORMxP@udj#T5PK~6lID~@Q0L813
zzAd*MS~}>!GtV_lQHcSo=t!%&aA#06pe{cQB?6uZKaI@&xll($XFxC3Wr%v{IP`MM
z23<V-f~qCr8=3nC!W!8OjN!;r&sPd@d`W>|%p%YpPi9-aItlLeoAz19@GQ13`z4>b
zVh@O{mvhfTjKi;;*}@)rTdfN5r0BIRRF#<7G&M_@3$?0%CQPSBl*#kYd3S7MDH4v1
zwWlIFb{;o6lqYDs;K}yJRh<L$@X5roAcJ-!e#NHH!bozH7;|*4s5T3Vc9Aj}Y4(Kp
zTAqg}Q-}=FpsKRe&50v**dsrotl*xqh<cE)yPsV`2@)@)sJXWCD5!it?sHAtd2-L(
z;hKbpN=f&O-S&VLs|vnN(xcV;6~pVR707Re`%$@tXf9n6JSH3C%pa)&Gne}_v^?CW
zx%DMz3AL#wD9zEMW))US-AQ<|V>=B@;Vjmf&Y$rGQka-MGXbMj653@|TG##OXpdE6
z+zm=QLVNZOQx;M;VbR2G*SM{@)+Jyy{WsG)^$ZI1j@+k7A+rPlHMe}Hy${1jvwkNU
zryOTrYgp3k9m&_OOx6PFZJ=o?>yyH+n^Z2Nvbi-u0S|^ML=PE!cVuZbUtAvDGf!u{
zcFl43lA2<Dm5+5v_J>%gKNBU^2RClhMFi_xzXEj=rFsU~4i)c;i9tH!7{Ri79^kse
zG2$*pNmaQCgExH;)W+_!=5*`@i6i1=B;FfSqlSxK7ut7#@}8p%IbnIixA&YI1_NQg
zoHzW<ImCF*o}$E7RN?8{xL+0xH-l<Jml6SE^zmg5=@A^`jer&pM|Uyoi(YVqkXzW#
zY+)}1G)Mm<m1VqAj4i3Rcf^C~bqLdSBm?>z>U}N%B#>6XBz{3AkIq-aVhiNNypmKE
zG59XWrZ<`_u}XIR@JBjo?`iMOVpJC53Xh_G0eOx>2O-<d6!&exV~}jNz}YE6kwoV9
zA&Wu%fE(_Ie$<C*D~;2|n?Q|#7#Zs^?B*~J+)Ckhd1;{f%GXgUGx`nMHa(waFw3+Y
z^>Kip-E6UEB=P7I5;IG#i^#`Q_p0Nh<e9hyVeWoMcG?hT-p$)06;t+t5n{JX!3NGP
z<*mjL6@QXmy)(Urhj4W6%I}ME83|sMuTvng02BUlmP|fP0u(0!BBnI{F16DzdQShR
z@M=GV2Q*(I6^8K1R1ZR}VYXR(rW8=#pbvB8z)~se*;D&B8w-f40Ld4o3?TsHQtunQ
zd_D|3=<%`Y(C#?+Ay$Wb?80C*H15bg)IYV9#(ii9w<`c~io>awC)r{-D(cLn!Ci23
z$x)}<NY?Wsx5(H#Nb)3z4cGTt+0fv7JSu%lb2?b<^srj`Va)h;x{mT*E1P*sPvb`k
zbE)0A^hy^=DecqX27>MAGDk%KnQW5%e!*bM)5dCDUuQ0(UG@zHsL%AM<ohWbuXb0+
zhfEe%#wo5ZlEhB6q<qAuTDQ51DF)9LB-95(z<{L@se*>G+AwaMvc%a~I<koT0<vP_
zBp5?~^rq__!b>nB?}jl^obMQsrn_43%|dm4Kv|u9`-<L$mqnhsR*v!Bvs+yAOFD25
zS??&li;k|H&-1nq6{rT>9=_BUSc*AJaOo2kSjg(}T)YwSpwNh#L^r?Bk}vBHI!>bU
zoR%Oivmm5v{geY?lb|1~PAHMO#;zSyw-Jc;l$NS}%3D{HT4OTly9Cf`NS?6S<anP0
z^e2Kf`-rr=MmkG<6W#&d38v<Q?)LSD_Jt@N7LVf<w*;H+(s2EsMwH1*_{O7r3>Z`^
zftu)CbduT$rJM-!1$TgOGr_tCAZUn^$h>N#0@<>i{?ls<m};VeSjJvm;<eJ{QSE4A
zM`@&^172Wqr-XBDhV9bk@3y-Zinlr+xtu2*Kigw1Aejx@rlz?LuxoKT><6}Qhm9>%
z8y(&Gas=d_Q7@<icuDzex#L+AwjHLnZgVblj}7P-SO|}JNwuPb?G6~tI&Z>n&XO#~
zd_f!ptcFiXTxzUo=&C+5l?l>&dPzbxYRahX#}>a5%7;iY=RXj;DXp#dwAhH#g_|ZU
z+ol2-Rq_vYh}a$99<1?e{H~O{|6h!~RZ!evv+kP&0>Oj32iM?Ea0mo~y9M{b26uON
zm*DR1I=H*L3^uqP@~ySbK2>Meu6;2#b2INu^}naz?&tZnnF(J1Ut|l0cM{>FEOEXL
z+~15`Mm)cjA*?dWAh%f|72ze7{<<BY!FuY@#zmIA)2^Q7#+)oN68J+$1TxTM$$Vq}
zESql3LRlAY3xNL`V?Mzb&BH7z&WiNq;?|(`C1+ffb55gC{lvB0Uygxqp)1r@(P};i
zJ>0e4JK8Vair+?z?giawexH+$RmQI-n(4qlXis#zZooSn9-}w-DYoIva6627-5DZ5
zMvvq&LKfX&GDUr@({T@z=e2u9l&h(p2PjF?Ok&`q)Fw0W2Y=XVkz3Z0IsY*mNGk%1
zD8Vb|zA5z3tDT0_yvaesBU-Bj*U03lixEX-eQ$@B1X*L&q4;i<7Tsa~{nzeI2ZyAr
zo&Dr75xq3bqq<efpXki*UZ0WtcRo_{w)-QuW;(*%&nI)RK0~c_UicquYWiCTT&J=>
z+;kD8#DD2sA4i^+kD^K03@KwMYn05Yce*8s8NaJuKo+%iDx2}kmrk1YPutZ$5}d8R
zt8)w>7uDS>CpDN(F#35)Q#kh1ZkgqYO(c<wd#t(IC7JTWvQs^c1o5Dk<;q#`t;u2F
zpJr&P(wE&vTOVV#RcC`!=8kyeh4VA~FHzBPI8<)7KPZdSCrTbM&Ni@cx$1ze1iKor
zQ^yR`cP~E$#F#MnW6pYBf8&(oRMM<)O3SO+7(Ir<|B_wwXj}}C$Ik{<9fdsgmYQT$
z0K`{*fC>YRxqeiywNegYu74ZhjFwgDEVj{;06jioiEk8)HS=g?%CJbL&+PN7fq<nd
z63q@oNi-a7R;4R(6sTD2w@B&pL{4y>6@@NgWfcL_HuH<0P8jk#Oqwct4^b%ENY(a+
z!Y`(3X@EReYHkOwnSkhw-9rqjhwqeUa9*|;HJ%Y~MJSS2)lXyMON8`q>_=q!@tK33
zx4x->@W&a_8b4>U61URhEH$i(Z)t>Y1IYZPhVfS125mA&Zb1IrWmQ@N)|w5oQU5s1
zA}H?TA3@y49ZuLcNmFXVCiw${N^&3}Bm|~Z#zHar5RU-tS$+OF3Jc{M@>HFt@@3Ds
znFR0(xMl!ym(TUia9MLY(ZtdYmFdW6Hcc>~qyXRa+xH?<E%Zne75gB<;boFiTpI2n
zaB`_~@y5ZBkF=o)cu4boh^eb!(JF-o%*AKG9t5yrH*7bKbd6`sjQim-t$e@4WVQ~2
z28|~{#H0A>vML4*KGQ!-*s2Y8m0JP%C0E_%`Qv+>hUZm_RHtAEN}ju%`#*99aWbtT
zmn8=V#yc*7z#*t<KvCOt#fi26W0UQVp<uNwz|rEi@2U~KRw?0Mj)?TZq=pk8F=zUA
zT96gjM+*uau<HRP_iZ@uMAH|6=Q=_=icmV@IlK96;YwOoI)1$RhE>@tHwot-#)88i
zqF0DgOjepRbZ?7nhx?(JlKtTN_(qxuxh;!oSD?U9gd_tX@l90T1yGvb7Rif9`<9)z
zUhB$pjShF}gh9yf?5Jvospzb<4z_QO_6MD*2TyCQRNZ*6CjR7jtWC$?_l8JBRML^j
zASiJ{XuN7eT&<dNYP-X^hugE}Q|{sjBJ##oFPRBAGt6>xpocHqp5jBe0Rw5p8_rdz
zy8gXtI9_nCQ$L}|N!|u^lE41Zz^;N4Z_{L^j%^81$)_`zY8$?1oXt}z)4(*sf09Al
zi@=^q_6XoE0^xCi%IfhXC(Tp4b5Pk$^-|^E`XmnpdI}bR%Sw(9;`K(9bazRX1)dnM
zqMT+DY|zH-^Hau#n|+Jf{A6NCYNgzh>)o|zidR&)7<{(BqvtA--QP+}-fB;RYb7$W
zDQ}1NaY@(cfKgm$>Vi(h$I8fN>3SpN%GRIRPH827Ti_qi=OYzfND%(Ki3>o9&e1`5
z!l1Lku2`j1zpEOfC>2~p<^6#Suq~)sts=6dpt4~5*iXldX3}~1WvYAbCP)KCX>H-_
zg3*)lNB9$la{O;9x9GOOxu|sgKDDMN?Q$=!$np)V_95FvQZ%{SOaCw+$Cvdx%K*ZU
zyKQ~85z`S%^c19`s#1g$vP3yMmK`bVDKfv2p>wl(tzPDfI(#4<R^Z5sFG<oZ&Q5j%
z`IEa2x0rvnBE6G*{~kxg29IynfG=FDK=#utf9)9AmUDZ$XE?J89cY&2rjfNCtp>+x
z6<!Js%i`B7`_GD$vBhML5}Fu2o4tAi*cxpHUscZ6ZI|uzIC*<ND-GH=y_~?CEZ{)h
zB-Esk5+5ZoX+A_9(?u!^);tCs^LA3J8^C7NyH*pai@d0eEQ%i54P(w|D{hR#Q61V$
z6Iibfajp|gkc3@<+!}8BWn*x#YZbQ~rs3?H*np!<BUN!+s(7U%(+?}5n4cxCn0bgz
zwa~2bUl@6&P)Ypwq-;18_%*4w#9^%xvJ?!tKkes+jWpTzF9Y@%E>-Yo6f;k8z_ltQ
zi`Vw66m8|fCx?H}sXzBb1Q9ykZjN%OX*rmU&Nv#Q<X_z81Y1j|WLGn~rvw_bgwN_9
z`nQ71*fH{znY1ET0HEURbQ1ND-iNDG7n6a|X8qL5_3OR@fJtF98xPe<Aastszu7L(
zh!@%(q2@vaSLlHYJdNo(XAw(^S8CJ<^XI@nP%Qn%w5h_VAl76B2R=@kEZhjpSNdgT
zWsFT@ylE`YK4b*{<Pl#vQ5smamhwRPWA>@)xJyh#2i4c;TUO%urh{d))RQ9m=PDMK
z0JzDW64j%87dLIt@Z|`0{KtM4Kf0XgAfLX?n1m)Al?h%@S@A}#eLq|M5O;+q$DYU?
z4QT3v;Wgi5R*${)@@L&p-Mv7S3OTJc8k%b2Ur50;(-$b)_&VJDAY7hG1=;y;_O=`$
z^tgU3(YVNR!(!9MGgcWag_xFq9P(bb{tU+A8!Z)2P7sH8{%kUC5K!phu!|_VjY?-(
zlBbZ}a~ZKk3uO&y)MTtqPXONDo-Z#W{t^X!7(Lef?q#_F)A5^%qas=0oNxsO6!sHJ
zZHXfz)BtI5vO*Z&=LUOyNGB)3eY@$m3%9P!%NY+P^aJ(WfRHR=8fQ-4Nw*x`n)!-^
ztU_GW+%=7<ZgW+_>m6urKGD3$$vVurO;U?50y7rVjew}0VK08-9L3_;;`k_SYbZ>^
z5XTlc%mfOcYa4Ruaqe)=O7w9fM87Nnqt7?mz?f)~AtbM#3dVk^fURpacf@F@cagr4
zg&8I~`XO)(#;-yiwbj>R39RMgFz>;G{MspTgh?_BS_oAhmDgEcZfN(QqbHE0{b)5c
zlqpRak7Dy^S7uT~JZVtJg?e7K#ZhKeKCnVqe@r%o9d0`x8RWwNXM-{5@N7zu(Liv>
zZQ|@(4AH-l3S<z0>pRnH(RP(cZ%#uwf^&WtFtcpMy0pkI+Z$<$v%gmhuUr(Wc%=dj
z^`JfhsZCdPsHV2CAhUDF0tR6LAKsRrK3pH#wYl-$l%Tu)p4bYU^xl-lrMzhQL$S(?
z7~+6I5TaOt2!e)*iG}6BYDLV8;hD~zk%xJRt&Gsk6xA<eJi6qLXwV(kb^hgi5<0h4
z(6Cl#G~ltH`AhIXpBg@sy#T_KyH^>>BScT;Y6koT!y_D3(Rk%(Oe<Z2%$)emYo>^X
zv6sV{i@1VwkInYNEHOrv^+BKG1wVa-=L&*z{`-l~ZyVx5A4>oHTRhsUwcU=f2wQ7Q
zMQUFTXQ*li2KK$V4a?F1^@Hb9qU;gpwcB`^jD@!iSnODUY@=vp7I~dDwT~VeEthnv
z*UIV;J1h_GEaL%8MQ1z|+3_SJOG?Tgp)^LbI*%}>x&%$QtL&X}Gvs){V2uOw`1+(A
z)9P2;Fp%4Dz!;!C)fhrI$2`pPYq>IMRqa-mJ~l>fEDiO$rXI*HR0OjQ_W})_C@z!+
z;le!kCK(L`f4ao6J~9%WX&f*)%ylosbM)tHM$Tn{NjfgV%2p5*N9!0!rnt64$k@`Q
zH{t0QLM2LNj;F%wS{1WEl;k58&^g;3KYvBg4iuhlzIk!L+A>Qk3^K{KFO-i0&L3^C
zH{jYS{CrD47)2eF_J=y2)*_lwYU#*+-zR{Um|^$)Bk(YO0FHHZqry#6=K}sd*6Q~z
zz`lN~j3bhc)cJ!u)KHek!1o{scf;KC-VtJgfUQE9`-+tWC6ppEdy!LNWWlG&ke-J;
zKZ29KGRX2%s`&L-{SfT7<rP|9`dI0#Zz99M@Z6gLZgAc%NofREyM84VDs#K`V64QW
z+3~iy1IE?yF@7|%1J3LUr(Or+voOy$7}by+vJke7Ij><R&Q2YRe%b;?UYix2PM>k9
zq}Nlt;|+ij?MPTUT-e}zRi@b*3J|K?VQso$;qa+y98ceAaYt8EcCAHngqjCMk=^u&
zyFFKC=df<?_k_{mi<0`q$3dJ12ERCOEtHAvlDfWO!}E8`6;)!r7SXtF-6)UW#A>2N
zqmQd0gYJNj`hkT&$IUHxoDBcjcAC%{RgkS~lt(W)K;m_nv{W9ChrF!aO0n)|VJ#Ez
z{^H{jzUo>EkE-7czA%(8lS^7gT7?>1Bd^?lJtW&bLa=yFKwNQhAlIrt@&YSMg`S&n
zCV{?RY+HB@1WhD4%yZA#*LkYcFDpLhs7pF)OSxm*GE>VvDFLCd7s;fDkT2P2u}kS;
z<!ZHc;&(Ai<z=|>Wmd^fZcX`hQ!d=egmtBSL-ahZ&hG|?=~YTZsL5FEL^_6pOW^D%
zE;hA&Qzfew!a=1icSE?H!|n_P@8TI^2DH)K4Llc11S1{`q_D2$(`*+rtB}f%fN_u9
z+Gve&g%_xq*b3EhY%R6i_x27|Kf->*fa29yi3RpuGUXW4U++iTfP-P!3TlJY`wd^^
zDx_Jl-{YM5a}-oYJ&u|Sj-czkIXJv49uJv|YSqp{!$AjuzQD%zSn=wV5T>cmr7B64
zj~e&&nDc@U;wH2F4ej@qHQOvZIEQ@kuB};(p~Z-xHdFTut55fXfU*&-c55Uxr5B+N
z#^&-ssQhjUn?sqQ8NmT>)NAhjtPumo>d*HI8uWu9VlXPy=c@FYN70=&-!>A1bo0Xg
zlq5v)W%!hBXHS~iz_rUt&|mmo`9B2a05g}Yh^kqR)32;#bbC!q6VB5zZj4zH9s{`W
z+rE8gby4OvTdJ{*rB(|xcSDn$VA(o6^rcU)G;&lzt&}VrC`tArMP7ji(e9u01kIR$
znhkY&SvyzVJIs5329J*8-;iV--66FKYLY(gT5Q?zq-lZeY9e9(sE0=e{fs=1iQQW=
zZS{kUmF3ZELj&C#h_=(=m<|sd@h7=>W>o(kaCzNrR0~z9-YdZ1Oyr$1&#PYIrVMAD
zvJ7cdQ(Z>Bk3|K*O^3tQP&Y$((`E;B4u*n|%koB>3C2cS(ztB^zFRrdsXaTNT^rbQ
zdp&+3{d4deHxhw>Tp7boV$C<gLUvkgXx*!M__XdhOu8-LL5MH8Ehs^#pX>0)V6rw0
z#~%z=8ZSnd5@HQV^bRpizcRi)oK44v=Bt4kh~mj4-=kyM{g|>bhy=L#_vEj8#u8t%
zDK#B7<I>h+ObW4TMw-#$kP>8N6||F@P)3;q7h=E-Gg8)O_y{=S+X-dmSM$<f^rnj@
zHsKO=Y~MB~aL6v+gJjzn($(Zk4YOc$y&HIRaBQTE@(OnJ8hN}xiSn|6?k$F}D;YU)
z*`qBwA0;ZFu#wasc0l2LGBwkXJD(TLxgVh>pjO@GVwL&)6Xlc7*olio8C$|%E=Nn9
zP7@-~v`)}ZSQ_1_Kg{Gm{0;ssSaCu2&?RE>*JevO1I0$phUrsHdZlEkv@FQ~<W+Zp
zW+sVhr*{}jF0yb2q1aepiaQtMTVQ$hAFbF6Rs`MqaU=F?lq=vI4evJqUGgk`CC;+T
zWcw%nG=YdPMM@tm#K9Ja8ET5ugW`AIv^!)*THuIkdNY<Y-b1wsgsL*LFp&3Nekph}
z{rw@3uEtgE>+bRcW`&VZH2%@Qsrb=uPY<ci=yL+%qxC_pD_C&zb2Uev+4?C*l$w49
zM{Z?tp$%|8ZCL3Tdpo>MvX5SeVzog~+cCgs7XK_)e&$bS!i>CF8yRtGq$PKC@%ySZ
zW?gK7^ofg(PL(+J`q|oF2yf#|oLQC2RcN#|5b;z1Z*Kmm+j0`RR8}80-0&WAaz_kM
zBHl(_+wh2#-z-vT;JoS6GKHHgzwuraa;*bg1(WgnDoVUPI!&6-21Kg;n!p{lDUd#3
z61O>SWAA7eP2yIt4hY0ONM5$eey`wNY9Q>UcCVAqtrz`mju~7mpu-iy@A$7~Xav@5
zZZfJ(E1F!0<EFz$;)6&@1%LAJ!9(0sp9UrlFHCuAK}7NP$@0x4_p$3`B-+(z)`?cM
z68{R3?L{HjOmhEEeMZ^QIg`>4&Lp`$Cd0%PZPlmdwIg{O<Gt%FU3`CGc9GkzN0lCU
zYF+FuwN)4`{isUt*Ez_g-T#wfhcMm}Md;Pl&-naJ2nWm_=<^;N5F!LvbGRlh!}RCT
z7ld>Mjb}(#SjG0`#j9RVZ}Ik^7I}NYJqn#SW71lltLr=9&!^re;I(2(1U1Ho^X9Y-
z;+on|OwzxXrfJvKP$(Pc719TJu3cPNoEwbdXUMV`m%4!_hu@R|8BE^fJDa(rrO@f$
z$5s_m6OxhblPvaKZsdc8cqZd2R2gg->E&U0djqgqMz_#P@>ptUr68GVNzHpAl>ZX4
z*l2(6X=}~y4qy<5Vq|*$ei-@8T(zHh_`hP=LdP+K<yR5Zhk`>)$ik5$eY$$m+@7G+
zc(VeJrbN_6%|!sN-SCL42IRbqB?8fm?#t?Q$PMt^9$6wN2FvR`6SKh8>2dwZsAp$p
z6g!w;qs8@ie-AVa<92c+Mi-H_WMG$Y#Sa7sHJAuenIK+=`c;s6&6IQ8&Rc*`wuzFC
zX%WkGclX0~!7&0{pwp@f15c*Sa|C6awx*g-5<TBnRuSVOtupj7NXug?%=K7{&WChg
z)jnL87(?~8pUy<GXs{!IIinN+c7#IYlX`n3?KbAotV43k-^7lO5XY@vcT%fn*FGd)
zh=pe}r7nlIWYmB#wx3I3Z!Kg%+55IZCd2J^Wif~USlswDO6x$ujxnN!*FRUG<;(Tg
znbbx=cSPa9RPi8*mr2vt3`ie2lxQcSJ7)*<VTG08cgDQciyf~#eAOL<s~UGGcl#1c
z1f4dW72uFAxe)`?20=qyw!8cH^HmPBIlpUg|K0+zfx<i6E4j?rNP3<0aWT5dR+LUQ
z7MEX1V0s8C=VKSes*><~F|FmVNiqmZ;MxOarxQ5nasMk59P=F+7qES$`LyqGjZ{-q
z8DD@zeNpsp*nD7VN-gIj<u7~hgZJTnC-vqkB%=L(>OpxMV_XjAht8nZ4nppX!io0&
zCQ_BcP}?b1A|At<b;AbDV|A58j2VOZ@AwxPfD@{D;s){rCcSe+Rn%JDi)bf-Rs+2a
zPV(oSm5+3fr|dvaV50=5>u`4OD5f*1MC4NQ39|?=7V-ZrmZg4xJ3mNNh+re7q|B7%
zr9<L*PZAB8Sz?=<y57we90oTq?=idJPA^~BRHfoR2^zJb*WCaVvd<e?KZJgrS(JlF
zquO5<jc;LLN}%iVs!$;7mzbDU>bf#|II~2Xw2^r+a?_w!xXd|%*!F&{Z_y?xdG*9o
za#F$Dl&`IAg_`|}oC9>JP<HHt7<LFSB^fK-=h|Du)CG+BSx02hy38pB(!>a6dawn<
z4yVK{{wsQ$%Jw5b-!c-lGsm-bN?6-TbhQ^!OU&0pci|c{c5g#|AS}QCrI)kiMtI+o
zdIju#@Q{SH&SSBb7#xjrS{I&;G@;|(9x7z<2rc`%*AhYbG0t;x>9y&@CItdY->85H
zW;Kh0Ot}Zr8K5}NX22DdAS^Suwp%)uIZid9dZ|I{MKDp`rYTHZEc8V|K?3tK)ZLUv
z=GEfpKS_C11=Wh<WbS#KX<>Jypfs3hRMUV$$><=W^h27>8U@^{oSnk?iMIh_c1LT2
z*`VyKV6^qnXyr#>&{wse*gg@%Y#>I=VT7#ULY?|jTZiN7l52v*ZRa39GJY<HZ#?Mg
zluZ|CVIwNmg_@3)VW%8(gTjf6*vm<IBwi8m^=<Axb>R!{?lD;!kGJ8%_>V5{jTQPy
zvBuU-Tsi{Xw!~W}<X(KGf@Ao`SxW>Q=uxP*-fnJqsvz759DJSFc@ELv3NX5*uZ)sP
z>%vL$9LDL8OldD#ZTI<v_(8c9y=s(;T`oMuH5>{QUJq1$#*VRZ8tuTLd}FGyAX}{H
zy!BBQ2U#&2XR#ZZWsQXnD+EGuyf?SNb<T<h7($Sj>o?Vu0c&~E9W+^-U%ZM)o3?Jo
zI<~;pMw^G;$J!GC5O$SQ-~lG1ek4a&BzcMz8t!i4vMjyJB)SndgNVks78v=#VHg?I
zd^800>Dc8nL5kpe=q_oRIX?X=JaoN2258v3SGBF8f%g7R|HwRY9UH7^>DEbY5clYA
z!N@D7wvmb3Pf|G8jyVVLKHr;&Jc*AS0@jvV+V=@LY-S?sT6XhgxUNE3GVQuPp*E4t
zb3tKk>*-8e+nYdGFDcahYPGK%7*DJjOU)0Aj8ndo%QcK5G=gZ4{TqEM#>{V1M`&vr
z*}lEs+bp`!<b)5k=I{yV^oIB@lIKHc9nYk5_`1P|;ZpM)G+MS3bwkAaQCHMkbNnAS
z!ppPX<5WdbV$Kg4Z??vO+kVpGyRTJ#sz86oY<P786Ow}12d0Ww#>hg|q!mA!KG5X%
zXr`pBV@`~Z4(sUG>=DOl)jHWltEzoVn-o$B!=)hQ5iE!$N+LJj{{-ueg35n_m4xuV
z{XN;vv|i{p0Jd*@_|=nc)jnW6kN)?4jgp4f#sDo~ceukALw(s}5H>e}P2@UeaPz#w
z3?D`!PP54#2|Ji=`K-&gbU};~nG0+1`JNE6O3%yCB|qUxUvCG0E#Ht?EB*5gm=RFq
zkjCya#Pj3)Q*T?HK!oYX1L{FgT)(R!5~RF<SSZwXy1xoPR<fZi5xg>1GjOnw&$9}6
z$`9jTv85DdztHwzU_!un6WOVNtpF!nj4_>@0BnDdZ535jzqc?rKFGiC>BvdXv6X6C
z^R{&<>RSSAs=~v!BDSImz7Hc4W`w7++INR%6SLOtJZM8z%e@W!V%@brReTw9ww=mq
zv?1jE&X|PxxdqNP>_bqRb25nKr&m8Z3>J-RnIDz5JN9tET25Xy<>J=IM3<UD<iCkC
zmGdidtbhoCnevo{ELV+^*S|7PAz;552x%9%b-DL;91Ye<^O^uZ7uRka%A{T?{paG0
zkUqwlM!ydyh`C&vU}2)PEH`Ni@X~Wf?Md!ZH~y{b(H^&T(6c~QbqTygPxrPEnPx)6
zlMHMJ#ZIfk`kHW~SKI&SC1+IUwbCYJy`tekr4i?Ls;~<l2f2rjyFRNdP9Zf)=u^6!
zCtBN2MJKHeoa<=7+q~UaKK}Ht<`Mu{_<=9DIoc$+v8HcIZhWPUtN2^0uv!992)|@4
z-O4OxXibuDoOo=*(<f3`wHY0R-{^*==eZBzN-JofAn|QDu5Ap<z4&XoLR63rp)c{T
zOpd3y2wZH3Q3J^<8#4PfLF`Ds?--%f$UfO<!45!&^mS7Wbu!pc7p5<0YtGboSY8B}
zvftI&pRqo2(v#g}DjFMBj+!B3bB_<=`LUi}8J~__ec&vO3g7<708H!7fK^{Hua-qh
z@piXWPLa}lWs2ua;=4rhJ2l;I+44%!v?6*#$Aq<F$zQ8@atARh?ToNldN6n>QYZS^
zF*O_B7&Sz&**qu)B4PY36k5_alQz3agcL#1KxY=QrhCBm?Wqb32s7^&g1@ominN<7
zr7Y?$Hz58oT5w_TKuT1W+c+)Z^s^a6>02_zms;i1*`Y45#5IWhhC-i5y~=Lfb<K!J
zXgNidTbln~z_5o!x<0{-F7yjRY8aa_m*;IUTZY|ccz7h-)&?%TPtxr;xu#Sd5>Nc@
zDY}0*+1<7(rPUmjV3u;-5u{X~s#_0Md|zFM-APcIMv1!d<lSvpwc)nAFPo}o7(*p4
zLdUam`eOb}pBEPX-6hcz+QX>fl7JuuIn^)@oTl!Id^GObh)pEAqqkVHadT)ApI&vQ
zY|^eD$jpre#_+u+06$<>Pk7I)aQxw~3A8Fr^@Nsccwss6tnKJnqqlJP{8iE$K?iwP
zwHR5?2<;#tF)z@fVQprt@z?oZQHm@!A-+0!t(;Xo>9I6D|0QykZd$V)!u^S6O{cQL
z-FF$9?n)qmyXgSP0y9_h-u3ETztcPdM`NXv`5riu&pLdn8e7S=nn_fwotxZasJB@D
z498_P#4>>jzmoI%8O|#wNWr5_kamO72nHRdO{oI%O^kxQXk6(7%z2%;iD$f7F-ziE
zZHjT~k3{srR#BwXJW>-iPckNM4i^+4?rG~bMIvX0=H0lvz7`SK&xN+-T`VB`7jI?w
z^ugS*eOw`^&Fk@FEo85|%J2qP)wB-vLF(AuKRXef#oH_e@1!kw<OXME!#Dk=CHdv=
z5U;#iX64R!HVB8}&^!^2n5$1XI`Fit<C0TGXBdEA5QYI0CyNR@)c^?z>k@iv+gr&~
zaEbPMS^YbuB9j>+s!qc%A}yY5iLmFc#rC1JSXGrbt<?~--lw9f)!wAZ^|CEkAf4|=
zCJux{;7casD0f@u*koQ=A?%Uc`Rlr2c5#Hk!H2kTswGe+MPWnc*g9>}pZbg5uDcxw
zXTE?I6bUbS{Eo&h!A(*7F%8S3`ng8w-^uT0$D(OG5Ke32#wWGNkb$(3n4p?JkkCLQ
zC8tCvorV?G2C<d->M=ZIhq#P;Y_7vo0bYORV#P%8?F;`_j7a;`-``hj9ERFr#WvA4
zB{D>#)87dh4+OS@JiI{*4+;`#q&$E5#hL`ocnNO$Lip8<t|Pwr)ujG8Po+@NJ^#k@
z7z|ZF-|qjIW+pQSb)hF0OO|<-0JT(i^tFUZ4bElZw&|Q8+i#KS7p07a&Cb46=IF*X
zx9%5BC&z&Hph$o1Rt@Esc(cJGFKv&|{a>BRYXy2AMy>^FHZ;DAzI1ZoYu(E0cm?0|
zyFfZ~I|?#8nv2%ic+NQM<h5`7YSJh+xPaZ4zVX2lH%*WD4$n>L*IbOYQ`CFO6IVR&
zC*6BE5<FHs$4B^+G=NHj8@%Q6g0tWW2qyQ)iml-v^f$cwj)tIw2oa%P55ldQ#9E{@
z@GX@tBHiX_2`n8WznrlGU5O;;#zqV8G*$c44GuaLNxrTVw_84a_=agKxwz6>r@6K$
zvXE_o9e*woZF7!gees-?Ep?*(z%%BS59g7LRgHujj|Ng&icwxY2>h&2;VJwk!?x`D
z0j!*)jj?ACAySzgg4;2adu8l(U*{M;1m&hvc<y1oCCXo;A}#3aj(w9>MBMy3dMm}f
zcMWbyzjRA;2*UU45uOt7p@R3P^Vq2=jrXmFU2GWCpy2}1oqIfvO7xQAO}314>ND9|
za5`Ssp{&(!q(j=RUd|+`gHRl4C9o13_4?!V&n<={^b%2hF8Ar%#9dnU-1^87?U2t`
z$`HvzdMBilDiEFy|9pACH;tNz?%}$9;+CE%UH`@}@J}k$s(CIqMfvbO0!1RU;eMV~
zg}Ej}ei{;IqrWQUhqA!HE`u<!n%m8YytI|Xf$!+HcYJ_x6cZcWd%%F(R9GH%9B<8W
zx_-ph22toKh?@9_e}uqE$yra*e%iK56}fDv=eKnwU^PA`>#44u;#E)+e|^=r<iM(Z
zRr)@oVT1bQC$E&OW$nr_re5<fmHNP<aw#vPR0e!VM%^&<A5UZ5-$1j@!L;|YiS#FT
z5D;=(=@q|!m+~E!4t~#T9HMc8NW_O%m5CFk;!sTe;?pQ(-azT4_?)bK*_tV3i|cD1
zn9egdiV4l^o!G+UDsYvjD@M3+oM_LMmEXwoxcHD?JYY}2HDOk(nS0OYTsG-t`WxU}
zKR)>LFIwbeM9Im=gJ9!r$d;9oiNkv8A>dcU6Zj(0`Lib0#3^6bl6+cWnZq&fl!G@x
zJd_~wO`v{JENR_BYPL(0pNjjZM49$A9do<in#@Y;cyMlnJOUkmfg}VjbUWt0uDj{?
zcxP_&vz4ov;@4tyEQ^~QrPht#(pzFNgJR;K9YXpOA0XwP2FkRRrh_%Y=5hGgx&=ll
z%cru5Mh`oHDyXJY(>d=F;+BjDthIgaHpth;&n$o=X1$nz{u5S$2L)vQxgROJlzW!`
z&iY2*R&6)P5;&7CyDEr8&PtEs=KApWGj4R$3NpWJzm`NrxW7U@Ye5BZ#ol=JG^}&5
z&bQ8W9o$l7qC8NGb1W}x+^<uITXo1*8T_H09?wl&x}-fbPY-BoF*F)!RbE<!t-FNR
zp<qf~Aet3aIw;Qw%H!GXE|>AoZ0}1)tO$?Rb{$Ee^{RM!I@(`1cg$qWXkezqwhs;;
zH%md^!~aD#KQmDLl?b4pMQgM<(|m;lSGR8UVsT5Ovb#&g)7s~99DY=lV53<Z19J_v
zMN;0Un`mhwH;gz(H=-o(rPlD?QXhL4KRl7c7TZwb+wmW>PURp!*l+(h>aQL$5>4vb
zqNPbhNO#_Sme&=BdO`|xXH@&eDg0Aa*de^VMPisU$(AAm;@`m4NZNK;b!0b9H7N*a
znuRSiX=fRFz`jybkf=WUq(i>OXh26>(Ssgb)k|K3;3FF<Z@=Lk$p;J<9jBm81Is&z
z*S${KA!?$>gv;TsEbF8U9_<m2yw-ePKHE88m5?YMJMJ!#G*<gPywz@gs~(D#1DAFB
zWiY{J8?eJwM!pyIgr&Y0z|)=s48nk!C~-tBtR7I~6F-gOVQ$(DzhBmqL;S6fej((3
z|5DaJg18daschi}MK^Ww?t3CyfL)<0_Ec(QU%3A^KH@;XT-W01WeAQcR-ap7v^bK8
zzFYNV4N;Tch%^d6czXADm;$3PheRySXxI>~wQiKkCK)gge@g+#KkDxdX>zFvb>Yer
zU^(jSs#PZ?=jl~nM15;?-E?yI0-AEUn{PFzfDW!*qGC%51%ptly`7h~0El9l0IuhZ
z#=m>GGQ{c@$RvTSjRkl$lcNLH^<GmEp}t1rikHa=wjW<)@M%|+I2`b|9*qz(aO|mO
zmd&jF?5)0Iz2?atm3PPUNpcB35D=!T_c42#^zyGqRbr;}ZAEr?tn|`oG_RUs%7$`t
z1phK11%(;jRma;iW+X?L9aZ@U=9>I0Rt)kT2<4ZKW1PF`Cm>j}Ak2@X)67l4c7eg&
z_#)1Ud~n$x$pr+pzE;^cR&zP+2?iQCWct5odzxgN?9#$^zE|Cr@r7;xq6r6{#(g2m
z-bB`5vp*h#jl;`?ciL&XY}O=bk9%eneasQRB733emoga|mwztDyMKr{o=ZiItGlfY
z_u-z@=h397T$w*(hWRKgvaqAeavZa=_j3BXP|(1Qzn&tKuRFzkI4vtimL&XPpYNIR
zytuY;b5($m6-@mVzSJEp(d9AYD40M~C9POu99y7y%@op8%4c+gYXjaK_OlS+CJ#WW
zL=2&%D$zrqSdC-ZZ9W`Um4r>z9RC)c)i!O#RXDbMGgE8fXJ6yT6uY#9gxsdpvSUvH
zSr#Ue)6zX8Ardg<;Rxt#4pTW?akM`QfhwhA1?F;2iDpVySnmosVH(-Si`U}{L<^Rp
zL>*QeY7W&JrLSnxRlcjzrSC?38~GGj*Wx!WgfzHoE~ewtsqoQm|Az%&pTg(k?Ijl1
zR*@i_EN~;s5!wFpUFo6Hy{~(CU{c%>p3YfH*k-$8eDLR?^I4!jMK_ldLG_bH!_^T3
zMA5Ovd%P%ufcN*;GjE~v`?XV);dhPyKeZgbEFLZKQOIMOuN1u2n!njYMn;u0)?Y%R
zCFc1^C|=}~JHEDiYCk^SM>@kaqV?QH^_7QKl^!AGDT^B(+Dash8^T+T+T}->RK4X}
zHRl(q8!Xvfr;pm1yJxU7R-%u}=wYY22C$6<4)=eh=S4k6wgxLhc8>>)ERLw7oqe;3
z%9bCaaK@E)V>XbMhsB+=v-^;T-Au!21hlv4Q}gqy@Uh0(mp>hh&{@wJ>cnQt1B2qT
zTvbZLzL7>r&9Chd#A(Ui*plq+#`@PSexR~P)2>R!3T76cf@kR2lTH17SO6FBN1}dw
z!I?Z`7(r31^%K`R;K?ftFu^lO4+;Y=<=AJO1X}U!M>q-P@Kw#&9uDsBvInYeFwR&(
z4^qNJRS@I86cE<SY*-f(PHYqc_Mun(UvI282%bkiXCVq)Zl%aK=T&wRTodCVK1xXV
zZ5~c-nkp&+)*rAx+Y3DN3fU_no8^fJ>AvB2PwB!6&5XWVopml=v&~%5-WDYMI*9N3
z^-VgNn(lM&4h-3lh0dQ90U|R&en^As@+mVl2b(s6cR8EXQ)_P;!gp>=Ul;`h2ZY4R
zE{(2lUiC~3<4MYXBeVJ;Fm-U53X~3v76{km<q5sX1^G@jMvBh@2z8&k;VLAcg7Uft
z#13GyypFMt>5wLUtHD`R&4fsg$Vf45gh;8z(e$xxdm)S_%%V5!Ji=`mKv@9k<zGY9
zJ%LCjxQ1T`+8Cqe0)HEPgBNmCAN{f)eD&sCSQT6dUbu%JCVCaxA{0_{NX_wW6JWa?
zrr*DzBpX&v3azKR;B)G6s||hAeXq4kS+wj^QtfNhrZ1T#ZZD@enTAOw=ScL&0h?eY
zGOx<Ngi##=7~x-|e~le{t{tyG4!V4Kcc@(LOZ&oNm!WvRW=vp%PMkl2-0fI51@Zmv
z>7L)f@;(07E7!Q+?8m--g&wuZf1oFbkgR{rzGb<F{AAwIDhZ>Z46U!(m65m_vSNHR
z$i%fc*7d73mTl)S{SJsf4T@<D(MkNfzh5`1c?r&1HGZS2UkVFdF3Vs(xI4uE78g$Z
zGG>}71$IvN4{}kNOWf=-l5J><<%@WA7uEm0*nvx(*t#H5gqP;4j-AXC<A?uB#S_2^
zJWL2oqQa$Y$$DzhaBMS7xKXn5C1*>pK?im;dxT=T09L#Z1TKnOq3i~r=_>BMG2sQf
zRSH7xv+ux2-C{P18+Skg6LEX17vlJA9qAkPokdPMU)h2rsag3F!gw%#tfy{CF|Ef>
zqwwoDkBiKWW}fKpS^1U6QZN0~5vX)BbhoXuoe>K7nMzMkmn7QUkX7SeW94TG51p<Z
z+f%5TYr1}1xz;>$?R*?bohrsd#krO*&f+P!FAvnkTZp5lH3*(jZNje+5GG{fW*K<e
z&hx?E?*7`f9rRHR%Xkc%S*F`g`H;h35yZ@Xc<UqA<i)2UoJ6e3b2hRn8<}K0ugo;a
zjffhuWfOX$Ukuo3bfr$S6hHXX^+t@7CM%irV|?h31EiBNziZ^WVJlk5vsvwZna15F
zY(W83EB{S3D8TykM(1q$`ng;8?6p1nF3GA}>=$>Os~+8qc~y4Wwr(u8m-+YO;>$1j
zUrlu{FE>$CIbhewC!mwk6Ud*6<6lK~P^ojm@N}CdrpU9uULbRR7W32X3XV>%)OvBR
z0yn%-3%)>_L`cAZJg73iEb!Y%mzIV7>i2y(f#I)*YqvL>B3bW88&UElUVVV2!oADX
z{%;71DbA-lv~SC1v1JX>m=q2lr7qS4v(j`C7eS{0I)}-<gMAVHPjNvX!0yib>kr+S
zx*nuWP7^2Hgq=yXHTG9UeAUtw64)7wL(XTExF62}_?7S<$c00*Yw3Nr2*ZpHg4}N>
zcE0#TgpIQ|#EY}m7_gD^u1)+Ql$di?h{Gd#T-luUdP~u}Tt4t~^g5&ZAZxCFG3xe@
z%i`nu6~~39X=~MG>XODmT<hY^^NQ2g?d(P6>ymh{AwFjL#YatbE1HBn{o|aG!kaHL
z%%3DOn>g;lg4PI!eEXZO=ay-D*Tqe)j#GV#r_y{O?>}7QWs&9b!TQT-qhIp^{jWV)
z&IV|&$6=J_yY|9APbi8oAD1~jce`r4`H9=dxssxe^3t+>*}Xgi+?ZbC^}EowQ769;
z5bo8kX$1)$a=rd9Y=z4gCFJ*nFuL&~ObY^Q>cXjbJT@wBfKj&6-o9$Rb_jmSM0#V$
z7FqN&c{6dk7>2w+7evIpc@U8N*zwPMM>8Da*u8PWeVHu{qx3P<-IelxhsH#R7!kW{
zcTUm|?w_U)7ja#bT6@WgaR@6iimH6xL%kjH(fzF3wmaHJEin531!#1I_;hJ9eTGb=
z;PZ~6MV9cCY%bV9r*%9gY}(&d$ZY|@3?sf4*stmrOHr7_j~PEt1^;1ZsHi^GALquQ
zGfdn-lW!7Q??B(nJC5bHdQ%F1-i9%z_4>=R7v@*<ntBfu?8GEPAbMN-U!UrOj`Xob
zxf&AW(M`Puzu@pn(e0Iar`b<1?V?IJ0h|UGAMO3H6nvW>)bI@(zoA<T^|}83j|LL(
zgP}!|;+BcX=ThXIfHQ!inN|CL1m$!|{R8S`_$QpH-~MkAKK}#X!4ZUR>$Q5Wy5as$
zK>g3#|3B_ey8jpFpcN^#qfDd!GaBgs_JG5|Ke%wc&Xsz|-{-**rPc^?+~AolA5ish
zc5aSQB;rj#t0pmWrGSw}6W~P-FST)ElpN4T&U|~ES!5L*iFCfqHw?~WK_fakBYCkm
z8M?OQx-W(EaoT*^2nQj&;}a}jF8kuSNyC2uNELr3qOH@B`{_GLX#R7JZH3ciexcj(
z9{zDR4p{1uHozroHd@NyKRCV^ujlRQwxzB@<#J1*5-Jkjz}RAAarkCwvx32>2^eo0
zR>h4W<gKI9YPes%82EeA>sh@K=!H>HV6(uePDq5q{c6?X&PTg-_Jk#71$bQL?dfPN
z_}Y1X9rO-H2r*4-ye_t#5_?#b0Z>O!!Tf7M>p02%BWSb;+I%32m<-qn0BuDgZ6P7N
zi{&=C|C#x%kdhelZ=0`*0fuexnc40JYj^L(*D34IsB&2cOUfd6$_AWcCbv5$e~ToB
z;Abe0=}sjnF?n4LBMyO7gpc#}9<V(T7H1A88mf5r967*VO%eUO*R!hDR?buYMST7T
z{zr*UHxvb&02Eoh4Z;st=ocDrSsf3223_y`tQsWqPhZMgf(<HP+q+RveA{}<&F|g&
z;-LS-(O4cXK5ro>a5X{0q3U&8s$<0<TveeO-%^2Fs2_b_*1pkVf%bvxduTdZP5cj+
z)46l8z@Q;C!3{_4JVNUPz~EX|sNq5Q16>a?3!W>FMugqJZaaIM-O1p?x1@J6#V-z7
zsEwDBX<Rach@{G^+?jWyV`zm0a$g)QHIr(_VO*qvAiSojo$~Rn^olMM$~M<u9JiyD
zP{xAurG<iM)~as4^QN6Vn99*gexLa-e7_s0#yoBR6gm#l%xK2;BP3i5C*DzOS@NYX
z$#h3CuBlpEI&S}A{aji~`VHLVk{GLLJ+k)#)iTcA-dKyTz#vZ<)abrcgr8KwKx;(?
z4m1eLM5@f>=M`g}VR&=yX#r2Fp)MgIWWBG;!-;GIemi})`2A^e_cO;iM;bA$de^Wa
z(V>7G;gQIDO?FX}YrYVyUsIKgWpSJfv7__mKtn*w9pw7wD9f|vtr|QL8fKpFmqmEL
zs4|c0c}v+x2Swq;d@NcKeH&eyrl)^s5DwRXl}O%o;yo)~jRDurJP@D9dLELEVno+N
ze&Pe#BPBYO4xv*K=W0w@!5@z~(+4zpM#!m))`>mt<ehh9r^gAMQ4R0w_Up;W)=gn%
z!SV+AT4yo~AHkDJIW`&9DD?QI9NQPaSq@jT`p4-dRweCVK6r+RuDo|k`JvfqT)M?O
zH)UJ<9BX<*P1n+6Zv#(r7kW<x_qed5o>!~$o<+HFZ$Jev2puU6uEaNlDG&&Mx_0GV
za<1_VmEv*KXK()};U2l1@6?j@kzaEG>pLGe&`&3VF_J^)fghL7i)wWXOLfS!>HWrE
z633OP+7#lgAEZ^G7)|gzLq3TE-7@b%!#=W-oPIJC5YKKC@;Kk}OMfFRPH=}&{8*G{
zwI`+nFoM1T-#OR9OOO(d>FKDn;ZOCl7uBO_<2nr%P5o2Tx44;I5rd4Hz+@$IhV!#A
z&1ZaUi~XHX!<$6_3u~Rg73Ju`+s1Dp7MJu;bcAu7_9;D{Kn67ke;jCY7N**i19jiG
z$RdbWp=)QcA4B!8k6Pfk0k7$FBh%zcPg@n#(C8Irdcwt(j?3P-W?Yb(`cVvAYRy+N
z8-N}qOW|&QX)|C@{Es_A0Ti3qxs3GHAizR@-?tu8|Nex#TVBQR%8KZ-<WW8_97;zi
zDw%)d?L;Z1m?6_nOFA%ILNXQnb(lQy$sm@%dVjPxWD_7VQSQ{FuCVVVi}Q))kcJn+
ztPd?4;#>EcBD2R!k)0eXs?-y~`MMvkHmu0NH+`m-iFX7zalM#vWmdkTk25!`q|^CB
zkaeN?<>XVbrw6Ofl`@^CFbZ;TB+@Z>E{zJ1CSJ~m@plK)txfn;v`S8~E%P_H;q?}i
zXwW&>9_cB==X_9eN{%0bwdXHA++}-}lUY3%5;<n85U@8?wy>pAQHRb0+<B^cbO67f
zUx9JuAdY(<g4_th-ss8`028xW%e{n}tt?bC*3Y|o<)70bOU`6MtRna+R~pO^e{;a+
z$*mU@36=HubJi|2GUD;d((rikI|wE`%MjG7ykJm1uzo+JN;LLazn4kWBpLTWodW7_
zF-_^`XSHqvrv4QcIcOdF`NM;ddFvA^-9b*GUcvhl)2~Zy4y`4}2E(>c5hr0v^XRW+
zw?7zOM&D2FuR0)#aY$&5B;?}$SjFQA{`T)@pELJk&s8W0?ETfC+AemGFzLe8?l3;R
znWju&DDl)Dg-CIkyTVEh&hZi7%!O>aAFVc%2L%`8#8i>h9pg-CsuX_B_#Sh?ma;N*
zf6C_=m!a3sb#P%=o%~dLwXdgTBcenV1jA~}!V+@GmAgZ!21$`D(TRXYf6AlGJpN`I
zy(?o#&)CUHP_{qN*!UQZ@R4XKE1f2ewDKleJ&As5KAb$Lrm^{pw)=@e8A7mE>{J!J
zfVkwX8j$QIhHk+Th$n1##Rg(1Z7gfMH}PjmZ_T@^RB5#3CF2gQ%4Q>1Uez0Rh2K#Y
z7sW=Qb`e#;{N=EMQqaqFA9>@lTi-p(#-G-KhV;HbH{JH3<v#s$g`4V2R31GkxWcJ}
z`<G$F!{CU0C=7QpkQt#Z_#oz;*<;_<6M@oa%uk}$RV(~bpzQ@Sc5|#wHJ-OM)95Zb
zH>YeznNymXjB{Fwv^`2p|Fjw;;bENs7HIj5;Rv7MM)_jBg`^~vr8wJ}rj+=iEWBa3
z=?`P@v*QOA;sYS2*pmL(pqTKd?oMdxQv{>Gn(Oh=);lf*J{L91L5(+SO5H&qbeec>
z*;VVO@RQyjgYdS`Vh3%BbA^1rw^kU{VXJ$`Or0X_n)ppX>g`q&@;WJUel(h8hOG>W
zYDl)F(*eK^_X_>liCV<L27k*^dOVrAh#*DmPd-!ch{=ukDKRMyqm{sG{(<HYhvLDx
z_m_mq+t*;ahtzAj7=Bqhw%9pUp$4NByl)Y@<SPMIP0sX#1tT8_=5>I73K=b%&f?P(
zW)o0}_bjWvc;h(|6dK{vHAs~-rf_G-n@BxKG@Uy><JP2D7*K_j-~N`ebRZ&+Y)jBj
zHs)US7Bss<M!P8t?@*Em-ayPOsonr1ZgOVvIzvN#gWtYfBA~{p8ezu~Y9~yPSF3K{
z3cZY{ow?Gk1bycY#fp76{jM6w(Chl$jILQAnFk!m<~!#h^i*6FRm3>B{O4CoZ$uIw
zF^Qkize&sGxc7x%M-158V;voCRlkMB`G!GN+_u8?ifU_dlW1q4Xy)Kex47*(G;5JM
zo@JNjAetnczm-TnFsU!DaGXij2f2|VEu=uiE!=M@&%ps%?gXWjU&~1{!=B10H^&O$
z>tMC2ztZ|Cic_P245wo{(N=Y{4HsYAFJeiWhpUE=c|4RCLV<hBraB6ND7mY+1K+)q
znFB}p7bJpHM9zv9h8~XP8X>5_UgWY$HJ$CBZVJ<udXh``eX@^>6FQP}M}fPDc*Re4
zeC$P;nN5PRTh0gwgjdO{KdId<$C}MfafUnl?D?_f@vCj#Vw|ZJ^@}D{;sMsW&#GDU
zwi!<TC+`;d!*l`GC-1w%<-^Yh`42e__FkGvN~nBv`l=^T;!oJiE>(3i&Gj^SGu~RI
zF8nf_Vl9T&kLgVQR%?0=_QtcMm)tCL*t)2f##*UAt^}_%bejy;UP{VqWLU7Aao8Y^
z%YW&51ONlL4hp!f3#*v}k<$!1Pw6g1cuuQRH|uxQr)sd`_r0H)wJ1mOH@PO;+1orD
z9B1g}eraa8`5JD*rp;W^$c^8<f~zgzP+HILz2k0!VLHZm?DoS8>ES4rCaOF5WjzhO
z$Lmt9ejfC{z^z*j**<K%_w52+nzZ8`Qs^@KCR{;QG4dgfCrbqB8@TYQG8UjRx_+<B
zQ46))MxqwU;bGR{`HT}n6Qb#^>_&U4rH=??JhGkPHh~oFN6~Ms4XJjNVo66E6u+z{
zK6xm9mR;PSa}P;2cuVh~(BAEB%7`7WS5I4%K_lXa+X%`k-SdUoHj}l#mp?h}II#Bg
zay|ojjrteIkH>4(m=-p$I!)C83O!`i2_GFcpS%)i7Q&Ni(9M_^*8kIt+`#O&ky&Bg
z^B)N#@XW<0;25U?SSR_0J;oL;P`G9s-+AX(=B^PvSU*zo<UlpfpS3e}uye<f=As)Z
z91gjulp{pQi1)6cFiOo$m}P~7HFNqK#+q%)F{2hKqVo_^I$#dT#*I(s`2S5w6`z0Q
zrCp!dtf2yDviXSKvDsqNy2-%P-)tY<Y)tBRkP1j%@!h~bns}-ep%q;i)y!5<k!rMy
zObZ1l9};RFLK@m8)1J!U(tb8Mt!axl3-{JbF9c3F0Q+w?Q!VKY{7AQq90>}_uY>Se
z9iB2uIRc*!RUneq$V5{dB*T^q6)NmNgC*SiS5Bl%n+?VnVJZAH%P4Y(nBiq3+l|o{
z{oN!Fhw4oka<75-L1&sI)VY+?*^mZN^zzoNX_DeF=_G*^0fgqz4yEkUVy4)>3j*BZ
zuz{s%Y*7QsiZ~+##pm@jJ=fD;)&>xTE%k<<l1cXb?X=}jIAH}NssTl5rDwfpM4*V~
zQ2}$-Uk*eKyMbn*&pZw1l(dx`tsgfljxfPg1rulF=AgPj(YP}}+(6t-IAWxscL?m6
zl>1pYpsIJn#Yi*&?;f;N{k?Ak(R+!+tzMD+Sc&T+_vGtM(RZ0FH&|}yxbAN>V#9G6
zIQK!woH-F$5bSYIJWLcF6hoxs@4dS(7f$sbO?hXXzjku&0Qu4{MO?4eM~EC&Z<;z%
z&MU^4rb{sKuBz<ZFL?Y+_;CtFWVQ;#bBtt1(zyg2D{4*TIH#AncuA(ShO*-^N#)n$
zvre)vzizKRt)B4kw<jFX$m%HaQ>=S-3uP|QCtK)sZ6H0cQx_0_O9B&u2jCl5=J^Rd
zEH-Xr7NsIp*2QZUN|M~}lG~&k{>bQx4`9B{kxLKiyy%1`#S>6{-Z`cOR$Tg<@0q|N
zb}+KS*kkTV8uCw}UP}4>J|1<x`Mfg#E@!P5o;+o<tNvxmN}RVfkhU|&uVt<2)CO<a
zWKo+FM&wM&0Sv2X92D^pgDDuZmTy|J>Ioq(2Vi8{ypN$4rAt*^@2ofod`+09lD-Zt
zQ7UA<v;i-XV3_{#xU5~-fcxJI8W~&X0MSh_o<WU=bLJ>R-P`%K(7cR_%zxb7czJu)
z(mL|*_gNaFwHnXOQL>imISB-=%HYvmGt_?T>=_N0=`S#`N!n<>xm%vLiv>m-Ar`=9
zVb<fFo>Ex!oC#=ji*agJYJIKLG0-7<Hq?``st>yNECA~(!-hZfTD6kX<3!zBY4+4R
z4I*213y;`(xHeRcDG;8K<8iEim!2E^E4@JR7-)KO3KwRmvm<R4GXDRt_tsx=HQ)Lm
zgai^a3Bf%CcXvr}x8UwHPUG$&cyM=V+}+(88rMeBxD(uAct7{fo$t)b`xneztJYcl
zOP^D<>#1GO-skM64#Js9iM6juqSSO%SHB=l_FNY|xF6)Q(sq;LtlloN@^x?kmkt+^
zx!$Qq#peI6d2B6gOl-9FR$-90U|u?-DCJI`myzm-Z;$;&t1ZKweo0&T9t!oS>`Clo
zsy<99Y0CFkwcMiN>%Z0CuS|!;Z}x6hCGn;=J`7h0?SCrr{&lX^aenS%Uow}c5CUMW
z%0SGewW}c`UQMiPwK<x6oojVgqGS_&OU!ps7OQ`(AHVEKYRFe*7nsXb6D?-<OQ8Bi
zgpW4Doq0-q(zf}he^e*u*?T^}-amG3YrfIz87_Mz_CuOGxZ11W{hlvGi`J#u>!IAl
zqt0hxY6Ah3euq*_syJv}d5Sok!Z<7;3*w83+X*0qM1dTW*!@|j3MIsf{%@-DQUNT}
z@aPY3O0!hZBN=y-9A<R{Y~+&gV?`(&ek*%=ma!cpd>+i&%BMN9-}okNnAeC?T1_%c
zyhN{oaVo0NBPB&3ds8VkD79oS%LiJ3?{7_BFvHe<57=_eBz4%U<Uy%qS>hjz95-b!
zH{OwhTdW(htCJz`U|sqI?VB2u(TeZ4yjRxKAidO#e5|N)X*1}gXNR2wYw}FHWiB!{
z?{lb#mG2&ZXeSm+5lAKRKF9xg-|n;h8CXHFfk$j7iLaxoCr3hA7T*{rr!78$@w4)l
zHHvBxb|M)Uk0T(nh$cRg>br26fA=%hv5Wy0@6aTMV6A1Og-14S&zhVt@;#7NFLSsQ
zR`$nHa`J+BUky@a)^e8cY_GWYstJYgy&3mAESt#RXVqUFUo;pRs4VGUqzs5`WmSCF
zHH?w`+zxrhDYQ+u{T%6U@&+gogKaD%XaIXnWvI+OQbiILg_etlY^r&iH77X)AM=hd
zKCL1^_|e%Q#x$fT?YjfYS6_&X2=Ib;P4eZxeR`WjFe!!rzqnfC8hN#;z;}QAaUi|9
zKK$HSFEsOz+&tLN+d-FqRzrcj?#m4`8m%?R;ovE=3WX3`Ua*Y!+lKi$xT5rTNJP|v
zs$I><m_6z|jm#+dnb8oSClL2U$Gl+!r}2|0n_ehc=m7J<+lStVMC%4<+9Dv?hQ!gK
zGhBo<)0@pMN^~cG$+rJyGVuU(a#Lg(-tc5lYCp)po9>`dbw*Q<Mu?=j+?*0+G8U~1
zYP<VjibAAwSf$SfO;^G(7FzeZk@(TZEEhgJnZJaJX0h1-KxaFTYD_=9j%SNn%9_Y$
z{#Zn#Fqz<z0gOM5rJxZ!2qkI&^Z3&#+uSC07!s#fV<s6_d}_76JU2Tva4Te%HQHXq
zRSn53M>KD{rQu*vP8gjQwptwgo+f#~-C~H6Zp&H0&;5xfY?I!>M`B`v^xZT>`-#v%
zyTdo6kE3n>YS;<?u24E%@#%etaUHGrVPFU!ItUO4kGF?PpnrIx0m`SMoLZc{E2x9y
z2iR{lEI4UA{lVxcH=*je6!y}#Bm@zAmr$joA4O4ON&Jw7jufd-W=Ay)di&Q|^%@o?
zR|Xwut<QdTaSOiFa``6JTGh)FzU-0~?;i-{Cy<R1W>N-4KfHGs#NE4Ia>{WDNXEo;
zc%r8=LVzBXesKGqhhjA^HX3-0l-vIbj$nyp6xvH9U})#+dWCH!sST!^YXykxuV7dN
z_M6Vk&)pcfZbv>ZHjpSpRa(0@A}c`U>g`i95S?=u3ne0ly@>h#lvE!F0)t{~YfY_R
zqUH2sp*)s3z9sA}2ik_v=n3?Y)xZ_@4}9Xr*7y?B3llCQD^Gprun)j>N#5-|Y=Sf#
zU8n9zQk}{+?ef6t&(I92=_+}ihl(TSTP7EbODSo_p?7q?x>ZAczxrwKl_#}Tp(o1r
z5S6n}9+la{WZt8Gncw68)B@B4dDlbF#O@~>5ReRJt8L5;n;Y`v2R=gtBSu75j2c_Z
zSa7c7dQiZTgh{5?!)X}J135gxQ7Z436l5D08wQINB^p$EU@j-bo{MZW_PFV+!c7I}
zL?~_jmTz%OyLb?$n$6+U@zB|f8I1$g0qqKqGO3Y)wrz&B3Zw$58<?EF+DP7a5ZOow
zv#^MEwa$CS9|Xqsd^XPY&xe}u&R0o9AAB(dbIWQooLU06Yvd?$`KZu6yDEIwr4T<T
zNqlB&k`#foQsYCQs=hVKXEH^b2V2G2#GBakg*2+szY=>-dQ@H8EEV0kfysrW@!j;n
z0X89g={{E2-(Tu^nr10v6hSmk@Scuuv`0^>4UgD+AjXFj0>62^Gn?hGkg)cy%trLF
z)fnp>I6|#j5nYTLdf%E?PqR-pSTgfh>Hv83-QFk4zN(5K9W&NtL63;k%^F3ROFR0#
z?|#81&4dR`Vt!ut&QgO_B%pH_8!msBLXe;R0oaONn(Mm_Xakv$G12FQPQS?AUT}=W
znjHC99@buD5GrR_t~teo;#M(kAe_q7OLX;c*f&Iy#D^whs6b>67WvbQ;n<CI&PLfc
zhue1X2Sy9*xZ7pz9SEPyxt%|d#8#tQp&;xd8I#Jee#R860O~&SO)KALX=3boj0}c?
z#mY3dZxzg%npa%W_qx?SB|j)ob{x8pyb*b1Z7868RsAdATFaRd{VDROT5<K<)mLYa
zN)HeWjkBBuEmSNYF;c=oZCm8Wc$wy}oU~kTFN5>wP3yG}g)})Qo}!u*F02y;gJ`EJ
zIrhI<5B?Y{1kSlJm%BT|dTR0cf%k-kd1|_%Qq^A0*oEtId>hN9m#Te;ZSSKDjB%m)
zG_KX;xFUH%psG};7I2PmK1~LPO~qo~Y?yAa^jQI{)JM+vF?%#5PFF$pAYpD(qnuMi
zmlyq_k@o~!xNDd`Z%uYx(wkLiAm|fM1?}*y<-*m-!a|V{DkymZ$(uRJuIxw6l5(0i
zJ?alW6PEdHx62b%U!^%8oGD#}zTqy`(1?uy3$dz%P(z_Jbb+@L1S@kG3&*nU<|8%L
z(y8Yai}G`~!1k4b1|q8fhX<Rf{0-R)f_;rb&zjv-r;s!mcZq~k@C`OSV#srjIZk)&
zoA-1dvFcHNznp-w0V`;!(TY4LIriZ^tcr)*4z5EMSjd)k#?RJK4wQlLb1hpFUNxj9
zHz;lJXIfg&DKqxeayK+RjdtftB^~KXWa=QsZf0GI-bz*?hMAWfqSxDI0uTJD$I#~8
z7Mx5^ckiNDcVn-q(I&btjrcTMh$D#T8>)i?-^;6OC1&erM-9emH*l}fw;)}epmGUR
z(;N=C3_>yY(t{T3${(ublgY@%k*@Y^SO6(=ZaWEkJsK}bbQ#_gBW<guP>9aU@kQ=A
zI)wmMG;kgIaN+GO)*XKa<r2O<+EeQ((A0praxzTG0M&Fr6@;5v80sdS?w#H^rZSJI
zpVI<(=YCJ|bXsOx*h?=V|DBk}!7`B|!Ts(H5u$p94-1ihhrvp&S6)v4Ajz_ybM1{*
zp6zGdV0rT2(+qZttRFA|Ka%EZ6$Q_}z>$u-tf*rmp-E|DNO}0=!kcqP`3d3(N@ht3
z!CoPNE$>|OK@Y8(7>q<AbuCqW_NxKYv&44A5fG=*^k1x0?G<G1Y)-vY=D26q#Ku?#
z?u@m{Gx=<Wq?wH)L`W`Wf%aJX9mgZ3m_*yLbTG_wh52Gg;2!P&1N3GShHZtU?h9xQ
znFr_i6L$KhW+~%x$BlPPx*1=(tk(H{Pq4YFn_mTsvs4dJ5Ke+2Xz*;>ZR!**e1qXy
z2LrJBdS!s$-E(8Mj^bGh{~Q+@1Sm)$oL^tZfPPz&AF3eHkvjL&xiGB`a5o2+CKMHB
zrOF$JNW|jo;~T#U>9MNm-k)_Kbz}@7;5CjIo(Ll8<FGlQim|M#Rw|rDum2+Bo{bP^
zD$V2I&y;SE<iw5bF#I{u1}r1z9mB|djs}(m3&{~uHX=F_xU7GT`(T6#g|&&;YJ#_5
z54T_oqcMIu_Y?2=V!`DZG*WO-`H?;5V+Kei-I!6sdkhdhfJ!Ns@+Z<0Zcc%Cd_!(1
z-o9zEhg4R```leRxe7+=cB4BrjENOix_;hMU<lkA>h!#^elxu6VWh{HT2in$($J+O
zQvH=@k%))3ke@70(>`iLtKVoksI4VFyl-&!UCcLCHgPRH?!8x;I>@SL<1fMz%-KDv
zDualfIJ4&G<*gB$&gZY$K?nX}BQLn08L%5gKKmW(YKV)}GEf4?TH>}9;yI@^y<1lM
z<y=;C1g5WsEXo?o<MwIaG1G+5$;MOGSx1)YS*CGt-%Tnhf12ljJEQAHIUk|wsRX1s
z-Pg9-WaRu(`mYbZY2<HPn=;fWlrdxGmy6r800xu8g$x5)tiO7pN6*M14-)byxo`CW
zlIe?khCU$EVbop_%`u#PxnS-zXVClLLtbm!fqe!)I6#VY<8<(*_IUbI{k>kiP!PGo
zE3~t8)$2tE*(IGHY4L$Snpq~`?{sbq7C9eH?_F~W8bHHVEtz4^4`-ikLbCMT$;|mk
z%2>CW^tBTTaCYx&whXO6*3l)AGOn~^t-46ZH>c{G^TdA1dlq24%vltu^tp3VVR$vm
zfoMqcru4|*!`KC($W9C-E5OlI9ps#(ug`6iHfx)LkRG%->?dH4O{*W(?In;_T@yPE
z_2t;>(MivL9T|dYRfK|*sS1O4;@-mT2_LDN-UuT!E>P$pz7nDtEEi^$ABTw`^<2)u
zP8gr681hYNOMPlO(ndFTYzE3tlB5(y;NLo1?|)&_t>?rY$w<i>6Z;W`@ObR>J|J$;
zw2`y|@Mt>y>!D(5Kz!BOU7e@bL2{tyZPy}&RgtzXOrwD&yd{$rpxl?>In2lgLF&yg
zMHq+Je+h|NS*oV=E)LNQNma)Y&kg61<>9T47<J94)Sgd8L6S{nWh1(V>k|q;cAyGC
zPM$*NI<)u?I#&?OBy%=HK-apCV%2_UWP4X6+grKCTB%1~lgy5seak<zmbshw`5U`5
ze~5iLvt%mg%zb4O^NY^7$&XQTM;Tij4^L>I<w=}Jx|;2y`81-KO#0Uqe)utYRL5Qp
z;`%eN4k<=+7<z+?efton5la<Lg=_ixB>Of^*$Wsq^~ZnDLHNpPtvrP@v>gdnEEg_y
zy$yU@1k?s)r0e`bybx#-Snm}71IEI|?7fraH#=Vath%g$FhFV;(HotME}xdgi<ui?
zPeGl|)>64y)sk{Jp<-qjIh#DWi4^jrLSGlRm8it92CQ(4-l~<!B*y2B;EIjN&WUZ@
z8<^OwNK+|e*k%geNR;u^_R`sBzu$+<8`-F$+2g?&i5Z2RI?s%CGgs{H&h>2h9F-Xv
z9-;1-Twv8X3fW}yJ3{ahH%$ieeo+KObV*nKtz1YbL&f_Zm$ZQyS@XG2bjRU@_{iI|
zJ)R$s-bGIMBqzL=5{1#x)<pY0&d6~4!Eq&LHNx}kDmk@nT!@|Qwii++Ym}fpC3(8Z
z%c}kSOJ%PuLqntqrLyFY-<|Zq$ItNpg}FZV>F84S6h2J>tSu`d_#W8Csd^3@@)ovO
zM_W+_Br-0oSogtDTN3<T?Xk-hxNBxZFh1T|y}ZFGYs|rc6)2I0i%ws3@2acbj`jnn
zo)$F?jnuR~tlh8|E4KQE;vjdxL@*EJ<UBJ2w~nMx-iQ@OxS+VTn4m#h=?KA`CLhii
zrg3OI9Q9bivQq|4m>Ah(c#zxQas@Mb#0R^`uFK`)T^t9Bjpgj(mkYzZ`gstzG$8QJ
zyTMbe)A3wS!f<s>1)C7)#vQYqHTPR$A9WoHCNV7iPY^WElNKhLK{WOFhuU$G-ZmX}
zRhil`1eZ<7k7TG~H~~^ph)jWQ9h9A{Qa<$y{KSAfllOo;ylyh2nozA~+&2{pFzQ%8
zs=aF9g@9}`{miH+n_>9Tb->Y`k0LQhDz`>uH$)gv6OUU3NG}V<X7q+JRT8L*)s!aA
zcb!O$yag2@r|WW)t~?16)hK?7_6rz@fj%|OT7C-Ipim`-P)}v&ROX**{}&+fEP%t#
zVMqrEiF>hls0!2ZJBm30tNR%9Oda%7DTlbfv`rkHwF=YWU?~p!aTL_uZVWws<i;!e
zR~Q$+{PQofV!W>~eovR<gl#sFwo&`Oo#zJtW6YDR?MDx}DQz_Oj`$<6nuW7aNL~)X
z^G$iub<_o?4VRW?kd-siGp9FFZj71BB~`LO7%`DeUj;ksaL~<D=!#@zXDvemmkY=`
zD_UlYWvwS2($;LpoXj$ItJ-Q<eAxt4K<D;XEX_5g(=FgiCc<jx;naaSeht3d;5&Xz
zhtx(i!t<YNp=NtK;N5@x-lzYU-*bNVvWdf_qpxuJu<ti}IJkYfoi(FLCz{MDmfIOS
zBGaqAG;DD|uQo@SaC{fAr&tqIg*4aXY|qX)H|FlaE<Wqr3XG277r4iLa75|@h$797
zh9&MS4&R7E<T>;x<~;1V<p;F=;!r|r(q(>HY5DF=SyflZa3!_L9M|Zk-d4JNMsqGj
z;i8BB5-WNs?~a*By&k1!lL~^#^9%bn^<cPKE5^-)@4Z58&L?d}uOH24$z`7GpBVs0
zHam6L>RLq=oNv-%jEjemJ!v{|waBXb3`X{%VO)s)i9j@UNM?v|=v%hOl_km3R-M!4
z$^#a_$QF{5!V4_{4I0d5kn}LN#vE)ebVdv&dHbWwMnLRVQG~xcXBrl!1=}ZDTw!@w
z<CLilPlL3~<ELKc+_;EjTi&dIHQh@UE(Kvnfy;QcD4GUVp=?xa8eV9X#Y!jBlKj5%
zm4m<Y$j9HtN&Zm9mC^%0(-wFs+uD9kQ+<Xg!@awTC>&}IcMkK1Bh^?JuMRS_=Q98+
zU&%gz6Ch1w8~J+T9@pkWx8t8M6U5$W_AT!3*&T~bN&}y!5eysX4U%h#u<|Qa6S<*f
zLVaCr4`QIM7O~7<#GZ&+Ao<myuZtb}YXiX=O_$$d2yuc$1h~&b>^G%$#WQtlx=(91
zlIu)~l5KVjC*UudcL!@CozelrPZ@DDPv&R)Q)7~)C^D!!2{`0J(_D|htZk>F?q=rZ
zN2i_z7{t5Sk;-aEQ#tjnSSuqGdYCNaIUBth70477vS8FIXPl}|#XCx&uFSa|y)^*i
z#Yn55=2v%d{_h~Fz~RI3CZOV5bNi84i_QJxq&$WQNaiBlT(0H4L#~we@?1&ylpA+7
zqUZEK921U+Mz<$O*i8u!auF|vqhrwLHLB08FZX9{dahuRsF4d(!tuTY-SITUhKxal
zw_H<b+tuj{S@^BH@#sL?d5UIDN$g$*XRDa5S6`zIv0wSDLug@C=^Ua}P}IiRQ5xA!
zolg1YHge0JUOF2m??2e=udcR<udZg%O>8?ys{^`!0LM>Jy|H=c+#!QqZLNYm=26nX
zRO8_I2w`N4%)`X^y@)=T;nOfJ(2DL11sVpmG>s;sRC7UV#3|A!iOh4?s(fm8GvB5b
z*7c%|YksBp@jC@=BnXvGuVKs(by07(#E(jQJ2*|B&F%?b@TN?;Egca_Q6upmaEjmM
zx?22%q*!yiu?YyzMnd}Y;)A)<GRpdxvxZQ1()P$g%EPx~_g=O5nJ3jS7Z2aQLC)BV
zcse~S*;tNgaetM1Pz>_85yPY3m&N7a2@x|C!CeXG%EzC;p3(fsrIH9o(kjy3tDEv%
z#-MCHb9U5}we%qUeaTis-UduMXa3Kg9p3G8(JPzOSRig|APVvll&j}m{z;i!Ueov$
zmmhVI%ZyR_&jz!I8G7M(V9y`+s6})vCl>-JQ_}>Y<`CLO-9u7bhni*>o69pv#;GP-
zo<<|}TJoH`b`mWsCn8Vz9fIJBM|b>S-K5Tc)5dQkXNR$9QAs#RrgSLt)EL~vUEDoz
zyN<9xQ|Lv~(OU4%Qs$dWML(4w`1C<?R@UVwIgTnLqRgdZRJJ?9sH?ECL3^MB+CCkh
zyiKRvYUx~xVxz35bO^Q41!<W2q-&1>?iL+G7NxEkbkb1CH>gzed$U-J%1-x1+N}y@
zm%i2uR?|RDZf{+W;f62<U1A2_;L7Sbd`onP`BtReOwo)naqNAQ`H^J(6q*9KkU2l-
zyK{4B67W$8I_ibGgdDYIzH6+rc?acd?d>iL%p;PVIe$lGcs}qE;LZX4rd3$8KllJK
zZFxF_R?~@juPwj^Bn<vUg<e6f{@|5_FhG}jk2_2*HE2E8U(#TxPH<ow2Hq&C7-$Jj
zr91EP{3P-47+dSr5AQknd(Oq5R?<QgF;?}}N`njDmAME5AwSyS85uQ2K+*qf>!y?O
zjV1S=#q6(STv8nHQC-*J`O{z?J>=_Hl*l|ghrBawK;ErUS&gEVm#DwR*@jq73KsWA
z_@VX%vNe=zu*7-}l-Co)qU)o6i`lY66FG&DEJdYtC@tPY(W?lycB4082r5!Puw1dy
zor42Ta^Fpx(y7tB$uzJRHILC8AYr7Dr+Z)Pupl4W^hp40z}xTlV}{OVKHa5|^`%;d
ziGAjHC9qL4)u|v}V5VIrr*LTamU-gIS&VAFHfJHy$@y$1+oL2zKfc@2B*HGd?F+XU
z3=_?R;dbhMinH2F$S;4y`zxG2pFy1Kq!@+sBQotH?4``lK&(J&yxr3$N2-8}zn3X8
zIE9nKS6qG@BOXxbx(ZDF%APktTO$x-{ItSl!}cxA32shbE5u|vVlSC;88PQAsW|~3
z-BAK#@?0>&4-jC)r6~KjUq<G$M-3_i`fDBsJX|G@pVPRSVZgD6EN%rcL0z_hrr;Dn
ztk7X8xA*xJ$l}KQdD7j8Ymn;{kIvmsp;2!rF|;Wt;7{%Y+44ZjxrDn*Fx27D<tGCF
z0HsxUYASA3V=lZmqgTxyxsHY3_d?$twn)x$PiilVmrC?5t8F<$mO1xlrXnO183HN-
z3HBk^wQ@UZGgP^D>qIcD;vENG)uM|Uv4m-l<!lPD4}LnMUKQmY9g03YuS>~Uc+A!-
zE6cHy?8_UK5|Sdc6+_o*wlsql(92X5s-NiT$!v-$Np$7x(mK(WV&^e2=p2*C#^y3t
zspXtw&9?!}o1JQ~5q>k`GWfG0>3Qt)*Nr{kp?WaVOp?m{S2rHrPjR%`(l&epMkYGG
zMvu7K*Llix`}{)xAqy@616c=1&>k3rzS@=fC$orNBbP~~gQ>{K*J5}1fSu)}nuD{E
zuTLq_mNr!$&gl`)>2&GmZws0Gunt72*bc@L$M^|aQ#pel1024JebtuU<FXnJFom84
zm;wb7s@FIe_qgcU7@k!Zq`PW!^bK?X`Z~See@k=J+HoB~j<c%PObAN|934W<UbO4F
z8w`@bJn@eNf(ufZ?~k^RqhnJH9+^8MvrE6cQ@W6HZK{6XY|FcJZ{W3J!`^#c*gS>W
zUJH;&{Q<A8X)jprwFsZ}^((YQjVT>Kedi8t?45#Y6zPQ2t#_wbye0Z^zI`YT`2^!=
z+d_rr*v}r;ek%Xxh`P+vl8ML&Cq;gg=sO9G2E1BkvyCR;mLq=pCK^Fm>1qDQFC<lt
z+h<%(Y3LZ7)b4)chKFz2Gd@!4rr998i;sq8kLHeCuJD7$j@g;Ty!mI96wuajG6?C%
zC%ZC<7#x!dD}u2*^xICRrG3u$Y-%CW;E&vzxh@p1zbLSmkh@cAc9^e3Ri&XHtgLu+
z;v4KHl!#)C(*8$h`nMeZTgTRgtDi8*h|OqQ_5EP5{27{4IQPS?d}pi`z?D8Pda}&P
z^}54;BlYG1bAIGCyW*iXu=y>!pAPS90Ot~|0Ui68ZX4@zb0-x)u!yTifw1(d0WBUc
zvfK>UxqSN9?p;GH$xMH;om0cxgAi9*nDDi{^EUA`e*H-d%iSu$fj+p)g6|U_SBu4!
zJ_88Jhk#rrRsTAT=3e&tNTDZkmfU+LOQ*#C=Upl!)BcO4Umh%V(Y%ydp3fVC)1#K^
ze=Oki<#68utuo77{8Q#qh2`Fx`4oR@Od0`8Jc(j^y62g$b~y8yq~)i(yhU{&F=K}^
z9(zA)<v1YHq#Iv0`qEZ)A;(MXQuG2mM=~#rKOQGam4F*Ldwl{35t)Nuv%RGOzWwD5
zS9+t<0`G1ZsPmlih;ICou4(f4{fhlYSR||?upJiDacsZg9d7JpV_wLZ{nYr9U@l)x
z0;?ro)#bIm=#hgZKHoKuEo!%0&Vkfgo2@+Ca#M-ol9=3CkJaB{>5OG)=~`?lH5|c4
zVvFoHW@+<<FkX)TVP(2>pXc<PRj1H_PL(Z1+UToK+a5_L`3ocsLw=I+eN*kUnhus^
zN*a$7M>4pSALpD#&y?T<p1iLRw}c~p)AH+RWy$G7Mh^dtfU0`ZS`F*+%!_h}JYB-o
z<UVJEug85+IWaz}#A*rhW-2)>B|73<*DH@_?;>V?w@~>&lBnT&Q13uRd&0`rYGF+!
zC8oSeUtfB#HSD!|VJNhxqXXW#Omb}0>`VMG(Olm3#O4Sa|HR1xj>`$Cg*0Q>`|9^E
zp48BqSGNYdxB9A_8UnoNE`LT-I@!H;LqW!auZhs!52GEp6>@QkDih`%`59{xRl5)n
z5D}K8RV##@fgD0-N*B3QicFL8kBA+sQICB#+8pnvwC;0%?jAlznl5z5zPdm8)JYh+
znY|T%^X-c<EMitrqwy3QHZt`4d3-eu&P+Aq_8B7_U(u>cUS{SNjWHuPs`Dc@T|)+q
zB?v7X>vfpwGZ9u*Lp^!8*4H}3e1xxB@I(}0<lnkJe|_Ee>fNI{eB+m<U(A7is*;K<
zq2Juu!kRmGnom5B1zf$a%huF;S0HyhJN!H8JJw#x#qIOCn_R^t<ZTR6PGDT~v=?o1
znT~O+$jD4P{!4s<$pbNOx~ih1%_ke5wZzDegWs`CZ5Oc%kt7*|hEr_4g9$_;T!d*O
zv{uh0agNP4COJ~+M`}{u=vUKH(~NEx=a?C<;kk2Y<-E@J!ScKt=-r1R*9MxK&!rnW
zM&j=cffeN6rMM*QH5Q1*j*jQ8S#G#ENZ`MQUfRoqM40bT3w_aJ`e`&uHL4Vy{=A<@
zI)dg*#Az76RpDW7HFV@wWQRZr;TWO2#z%9$k1#(WF<#6dvii>JjqbbDdgYqv?eGjc
z(=`**=rnKb7r8ldjT9!_Y@dmLC44AH?L#xigY|sFo<4i^F79mQaG7#wgsIQ+dvB%6
zHfNS)dS6ETl+-6tidrH<)3snN^}%T(`A&1)T@<cG&u<l<e|x9CEyQdQM!_)yY*IvR
zB=Q&&Ie6Kx@Ik~e)q^aBklxj<t%<4ucZIGqseD{6Tvvc3MncWgHDjxO-#ajNG%bhe
zq1;zO<F`K)u1ot0{G$E{{%VO`i-)%i)Q*a$5ivrxbTI$j*FWG>p7UM5d7$LiR_h@s
zC*<~b!*c`yC0D2MiJhyH#n_~v(#QUh)F92I`a+!-K^SoalCbyCWc3~smmpEK8gEy<
znZ=t`#}J{CKh+Xb3C#fyyYE1S%0e>kQse3!*UQ;+e9s>_i$!cRndyJE;%mgD<d?7%
ztZ#Nmd+?4>vn_AMR-*F^tzp!>5>CoPplUr6<ppwnfM}6VMqq$)f+Wrw=ux7RcJk86
z2ku&#>T$y&I5Fm#!No*EN`Lsw5YMi|Ky?U_3nx?!C-tVtbPKAw<vnUf&YRYn*o{6(
z!Yv!afD@&sTQhv06u-82TYY<kFks>QEM=a6%l+3R7;S~ZAG00HN*NB%fsCCC^Tsyh
zF!_=bm6(+E`Dy_IBKPHG=2W%-W=(9BSII;z$wqp?OuFt`r^)RYvG)7=6KkV+g;UYf
z`5lGJ`^+P*5`yc(`Otw$EFBNt;m6Q7Ts+A`e5E$Mw>I-x=`xWLpH2;G^xT$j7gcSs
zL<Ut4mbDQ`&=Eof^1q-g0`Tk|9C3NKDMe<{&c?h0DtWZIG5Obrd+JO+?3^wD9dr|0
zH%c-K%GZmiKK5oN(~uOzoN?{)W|`*~XwOAC2yDANi>`)P&+NiyUXSRs5V`x3>M&UZ
zHjOZKmJSgf_yr^z7uR+7S(dQM)3~qPt}?YnweRH1uQ&KByM?-+w-(R1#!3}OGB`{%
z2s8KgvTk?BG7Jc+sqF_Mvewe_D&m~a>n3trt2Fx-?Xv>3TehGUavi~Q1uRU;<>Bcj
zbyRaGyGI5}$Z5O?=%u&Dx2K1XO6~qehK*tiuNzex_}M38=Ev?9>P~Dh?h7iPj!(k$
zxZ2Umd;3IdycyA>TlEO2AK#Z(;#caiBP8|-*@~sCi3SPpa5IUyjh_VCC1~Hh?U!0A
zqF|-=38E8Ec=-a=^jL(BW|{kyj=!BIFJBcoYXXW=21<}%SZ-HwDcl8AM_+tsB_ku#
zSTl4$I=fv#*UIe7iV`cO_Uoaz6norG1m0FP-BYAl@K0oIjow}PRgBWx$KMe5iI4bp
zXF$-G!d;cqgaZUCH<LICgfBio;9{W6Ji#o9y~y(2A`uD;li6}Y3*oIz2C>DW%!4@c
z!suxrVNVY#Im!Z(Nzf=Po{ir_XVd3Rw2ph-@q=E+Z4@&1GP;jq@LX*jLzgu0MbTM`
zHN#6>+lkVLEzD`H?HU3pZKBBdlP#wrcfWj-7&hM;)OFOtcJy&f=I+8sz!rEyQ%jGT
zC|8@%HPS}CDOb?KC;5ZEKm0|$do0m9>4UPln5m2MH6K^6s>=)q-s)IAUj*O{Bt>RD
zin^|p(DOBKGfn4=>X~zag84V^Bmt|diUxD|;%lJud1>XM#aQChSO)urZJv(G@jC+V
zcxf)W>*%T)<{^b!Mw!#pHT{>`^ar_)(v(>A;GbTYDT-$`DPeci$qVip+CWBDC>dx$
zZ}=dK`>OOk-(6N(@C?)N7KOt0SqHbWF?~cvOUeC;BmK!`4JR+SC%$*8>U$vj2v&nO
zdp8{w#d0p8!$aK%y%myXoPxp`fdjj7$3A_o#?5ktZ_`|v>1g0+TS%);Ai4A&eq!f?
z$t|NzVHPotBQr!fo<@2#g>@t<{(L{$l7B?$Qow^+asASBM#7Ua0Z2urnKS00baio%
zydE~;^ilr)$w}XU7t9J?8n@(UvrQoJ&>E&El}HIo$b!Y;XjAdgwJf$!GP3EC{e}md
z23M!JRREc4FE9L&s`1ozzR)YUn|0TDc|uVQX_4(>=-AE`&}HL7>3XKw^%ke0GOcQS
zNbD3L2DQ5Xw-x|up83eAbK+Vqu0b=mGvl~g#MsTL6HvG80-oa;c%UP-F8c;Lp0A+o
z!rKQ@ry@h)Xl>=>F|R1~V)V=`#`B&^OoBJv%Qg>?(9jD@tVA3~HN&dtGf`p>z{1GD
zOa{e!?w9PVqsa}}nqq^r<t6drcn1Z3QjVNdwS0!v4CDx@)y_5^v6^-B8b7A**4)Ay
zgHA{PAcM}hjaj03`i>0yd5#mnS*`S6RuLWF@Jr!=tVul2G2Vr!kICwC!4CXg4u|na
zLcTy3szWJ{Yj!~V8MGi9TDwhrW&M%_)US`d6`*A+di*1u`a)1v>sSm>H$Nolq<1qA
zb%Y27K$TV|w05HCwQraxn8Fjs-9QxaBxj~zn<TKU8%Q;UUBBO(%_|!%^6_Z1>Kz4v
zJJx#lY$M_}-8ia*0IK0Np{uTM+~`O^)jM@wnQPSHs^9td<>B02nT`9Edluvz?BYg-
zyI<Qqc9i9>Q4#mJm5hyv#8MZ=(sf4o@+4*)t{P*2G}oU<I%R7ouRU%&Hn(rnu3t_F
zt>kWVW|4pSpJoI;h68<a%JsY`cpy`btL(1tp^xD6bdz9~%8IA+DB0j}j}=cdmmFb3
zZ#DZN1mGtAW16{nf?GBTOHW*VwK1@%TWaI(%E=4GTarUw3Z!B|9s9K4RuqLF#i<0L
zX{V`lrLRP4no<-Dk?&mR>Wd`zuwE8}FWhuS*)G_SOHac8*|f|%VkNwr+GNc|c*#O{
z+;`?tS=-$9dO_i!K5<R}=vi$&6via=rux)gcCNad7-XLe)Mk+l!vF{IC%(%cz8pYE
zw<R3VGp~|$Mp7DC^+$v_^Tf{&COS*`zwedWE(lt-vs}N+_y4Rj@bh*%^pN8b5Vjh^
zJvB4ryfT@bJ{rpMj%AHlQ0lgnZ<M~tR$%8)5AbGjSfaFa8aO)$q`l;WhI&1G*1hum
zF~t?ORwxvO5tk(B({M!x$%~{{C;zN|JKtNmEGA@ntxtjZ1$B5xR{ciIR+5d<Q>a2n
z6w$$bd~YS>Onxs`pfW&bZoRBgAPt@CBZ^BY8m@`Jp*QQ++|AAS%(GR}^lGTgUhQB7
zh9f7;4pNiZMF0JYTT^o1xtK4tX>9t7%{lv-2rK``D3i~K1grTp>ZpZlo#!I7(8i#%
zR>zJFkK8>;%%s`3m%n*yVH0E}6&^9Rt(D1cm8N&Q_4<zE6f#1^V05XmQLlQRAN*Vn
z(A=n6JqVu8Ndba3;rXqUerf1|MD63q7W86cm#oB!ckH<Fb^$z7I#S>|>6bND@IfpM
zus4pR9CWTq4idoxA%I>_P9pL;8F1N9H^(Km-Egh94_ZuRPYbMqmJz2%TDkEVHv|i|
z;(R3MpJA%0`cI52scUi5k@s~|lQ!#|h>gwo`L?U=af#3O?H5Lkz6CZUcG=y{!{BNY
z=t4}YdC^YrhAd;5Ms6OR9W&U1XkI{Bbmex0eF6VDUkj)Hn?`{?m2;yWMsoMh`h@a^
zB690@bB4;>al4K@Z9`WGgD-*<8$if0l-JoFabuqHlPz8ZMmq4lJ*|}68db~b`j=1g
z=#5KzBAasf&S!91LiRo`1T;q<71(_re61LBMt-2pFx83px6Lrd;rb)OzaPT&ujwh>
z8@Yd$-Xc+(`o$J|(quwUYXA1N)-U19b*sHxZMBal=TeP#)VNyA)I{LYjg!fX^bYoW
z@H55gMS8^}=+D({N38Z90j9^qs=|&6iOru9<F9~>g&DzTZQg^yYfQTnVUlWPp6{;U
zJ(qK-XGwisT@lAh!AjLvZH7w*_)o3sKa<?1jpsu!o^U-gb1I&>que+#biG}!4j6P{
z1GV5HpA|G1zjiPw{~P~A3b<fx@R_aQ8gEW)kOZ8Y*C76_YAw_)PPFCzel?~DTmIMF
zrToIsTa_{jrPe*6-7=C@xi(RMz4o0EZo`8hcN+8$UN0lxrGK4jAp+yZ_gT|gM|-K}
zz0u5|SpR8m{)KdUu*FftKY>@AXxx_X9O!Xh_nh4u>1t+Yrb@M`(;knNBBU+V#-C6U
zU7pjCBV9+SC9p@>LQ|70=<Os*m7ma0?R&b$zMmg%5qU~+>H_az&Lm-(xq+uct}W}o
z71i9$Hq<g;_KRWhkz(a$tiqN&YG;3HZK9vcOCCv63ChPF%~MwJ@G{PkU4v?JOi5Qc
z+QUZESCID93jzKmV37)1TE}fk`V!mY6>K~8?B+hMVBeUp24r)`=ou}`l#Z`!H-4{Z
zZSxk|D%M}^xJ7|wrSlc;wXng-*{)CJonBz}t5-zPa3_oGPgZJBgOR$uIbxC8##Z^<
z6So+<Q`|j}dls*QzfvBFbpE&AG+*$*8k@vU>YfoXVonDv74Ov@;MX#x&w_W6`X~+$
z5fFT~OJ=^Wm_PcOlt+E0W2P*P8AH|`#&B<<zEbdmclDXog<H=BJ$1sn_T*Avx$VAZ
zaoZm0RYoYBS(CGg&5M#o_>Zd`DCb!r%0sL1SZg`wV>1P_Iumg}n0U^uP1@sxGaBXS
zS!}cXIX82yWfJp);AL=TC>xQ0W%0}CFsR$$UgopywTWHE1`puq!>l`uZ2Ra3oLx6d
zeB<whAvW)etixWX<ZDA@T{Dn&m-n~V_<El4^3PZ0=(S!g{EyIUrZ0;=*Atm57kwpY
z+pZ~Z8t-N)*ja_zy;+GeVPxXJRkRkKPxj_jB?v`tzUZ|4CU9?a-wFBtxC62<p2gXA
z737W%^JPl~Y8x~plMC<rUq2GusjBlFEk#!}-vamVT?LBvHdel?=}7a9)fW6DrY_G0
zkl)imH9I-2cf1}mEIPIW<1KK;%TTh~Q!Yun5#C!kzfx?f_xNZ+1(%@k9CduVh^o*F
z=t>@D2P*ouxKBAVu-0?)RiC?ZC=W{)z{?j*MR1%b!e(Lj1UId7-K)ef<U4lrH{N#_
z&QA7of0`n9A<0R+8vEqDq;RkPY41b2K;QW<2veI_S^yom+nBEa^Dv|$JNc_?pD@fO
z^xNjE`h<&8nM$rCwmY#Znd?`n*MpiD;U6B$%atESVY;cWqeybMt{+n?vf61Ua$X46
zRGqe6VfRigXI+A?T!_n58}!FSgn1q&?Lz=t{jTQS*0_XH*2H7VqjljGq{uUvxakBw
zRjF-t4~`B2^stSOZ+|K0bd=4OvG(WntBfGHPqswtIi-!Z`PWfdn|ZRZn7cfx$u05<
zi}uMBde-Pa@4oR_UO5n^?WTu^C_dfdqSe`-0~DaH-`Uc{1y2h|bOm>H4pbK8%*SL5
zW&fbP7JgX-aIbj@{Wl`NpL<{(?Wi220`DfaMDARN8jQ4M8pUl}kJnUWoPl>QY;H3i
z(s?3+lka}!s2y&ehJle4rlC^1#$T@_Pn{?iea_7QY5pC+$<wPT|9)khjqCtjEcaeL
zD^{lcjC(EJ*9=T&LEt)#qJ8~B99EmF0$87sl2kIURno>kz9oHY67Vv-l<c=$|JA9#
zTuH^@khx6Wyng}rU)vBVJoPf`TTFR>xVL_PZ=Y<35dM3*|7+6!&+D(RF9@S}6~{HX
ztgj-rrLC9!ZhEa?WL&yfpOuPpm@-|J&!V^FIp6a-BInMTTTkG_Av3~%p!0v;<v+iV
zVqOMgSbWv*I~||?EOVYF)F|uB1G6kK3)cQ7k$=uI&E^Qn@$8Jq<oyzr{t#E-<VNwy
z7F%)o@BJ1=B7gCI8Ue(1xsNAvN-B?~>|3q#)8ZR4($cU2TP`Ax=ms0DZ|YChg?lQ|
zFd3F!h6tXHznejrkLu-K0q>fu5AmFOOB7<&no=O=vdw9ZjD-|q`tLv4zIzqI@^=ec
zfL>?Y%TU8oEfyDQbL?e)Vxj)aLv|=%9`bE3tFVWmhizHD8?`Y=vG@Jor=`yEm5m5&
zz_VbMs3dP7PcD91Aju;mL!z1IbeNO4BCn+vackQpr*K2mz5RT_$IEj#O;R2<*9Tkj
zZHj3h&RJgdrap=9oQjX@01$V)Z_2-*&NA^B%0_{lV4)EUod|W^4^|VndqbptN}Qbr
zp#6PpaQ?EG7P`UEQ*l~Ppr9@fi$}H}ar4U}p6<CHRLN3l*{vMA|537Iu|Kw9AvOE#
z(Gkg{BR+j{LvuMCx>U;8oWR&}tQ68rVfPqm+tx(&A?q`*00?93N!bx(B}6NvAtmB*
z@I+Z;Z?!CUPPr=VTISIUTPvERQrWBZWn1I<WSeW^=}A}a1AI3|tw*x5W;6D7iP`gW
zWQScsZzNR%Np>1y)|Oic(!X`TK_dI|lP#LvR6H9ai@A3OuLFGhz({?Exl>M!#8YI&
z!?n4rvaY9mZhIC4B)C*oZ-Zi&bMoX-ef#J?g`BE0krn-_Ja1M^2y|dBWY>`tCrT$o
zTj0ZgKK4MgB3Jw5PC8Z;M%zvwoaVKe;(&J2A<3oBoj&)6CjZRQi)*b<y-92#65PQO
z?-Khpuuw?O%GZ%-%lSEeBr|gix&QMt*6pBj699=ugXES{sFP1sl?Y^edwXH3n(!Ek
z<qs7q&hMs`zrtDmJEGv;){(s`&$s6j6<Zn4<Ll%X52UQj>*VK7$B`fKiQxk54kK9P
zMW2<rOxx|fP#in_^!8i074HJaf%zYRK)}t39!sYy-=z<)qA$~B6wLuGPDCC&i&&<5
zj<^@~w~J58)rXJ760iY7;(kNyVBO<Q)$iB8E%*cv^)8h&O|NTr>G@<}W<O#RyzR>K
zr#qc(4nvE;Qe4W55F~)4zMj^IXXn@*1WrbY+o7NDVxcLVzP1VeUp$qC=W8u7SVQFn
zm(wbTW6{f2QewRz{748EPupD#eHJzj6JiBujxc3@;H21jPGwt{HC-|<lXX0I8pCgO
zT!Bwz<1dMyPS8tCOZ3*xp{(rb03SB*zc?y8>*12*N*(6d1o!a|!@wB~cOKgD8OI0q
z;B;-6xK^^M_*#Vmu~XAZ)gu}2BVS`0B$CD_<cXUYnF~pq|6iOB0r><na-#yk@VOJD
zuTBJ8G&m^oH*a+JX?tDgBuC-Q(@KJc+B}iYfL;8+w9oTfe0o<EWi7kD6aRNLB~!y#
zduoU0yw*8%SV&VkM19p@X*{*(=*IKP|7|a}%H!R;s6`tyVfF&8<Q_SUN<j28i{%UH
z#ff4g$+1`YZvW=5HuP_h$=<@UZIcDg4CADLnFa+P(FGj6jo2DWlJvk}R`t{`y`L0X
zjO4Q-6B&ZFgRqPQjFGE_-xviTb49a88JD(_V4pOk^}lER`*PaVd99@bE2R4stJ1W1
zpB=o&cJ(Ks$~OY?)(Q};ZP1i$f1p!<T>aP1?asW`(t}-LGod}IIRPw=<K<?UPmtbG
z_z8*}(KVLkDv%Hz*(s>lrhpR8e5@YbIyqK+7Km-uEf!pLL#fQ1->es7W)(&gw+e1i
z6ePnaMNbj_Zi6ei?0ziUnLo&AkNd7;Dw12Gko`QFBD=#cN<#6Gk#g~AX}!p7o>G9{
zIq%l}wtX;GRmu~uzk|2t8FK9T3@1b`CapR-A07L|R9<<+TN<t{e9+Aap(sEvar^5A
z3+K-%^^et41EhY+Zh9{hqbv;CK$lM@uAeu|o$`G?<Jq0+0oKA8v9LU?Rv%+41S?CI
zd3`?EAZ7?+GFl;Xxi$$}6qb&=WGI*PFFG;@T#e;LY$Sh$ZA<F?8~D<wCeDz##A^(v
zQrI6boUzky)7A>4)HBTmz;atttrlJ5+f(ETSrg^Apz|~l_S$<Mr;{YP!sZWRb`$3+
z?NkYBjJOgAPmg%hqhSL2Q%&c2(mcAb?M16d1^f)>VWG*0l1{Os*@wu1%Ok+5#4{k*
z)Ofs*K+XHx9joejS|M|s`I>3LfBBbDP4EBlpO$J_W@~v7#@9~4X;jN*aDOoVzPb4c
zYk!`gMdo@mKzL)^)z|#}#!g+4*jtid`#Ii>A1ZmQB6Nu<*}saMWHfqngN1)}xeJ_~
z2H6!VnCTY3oS;t8IORSsU?~_rWi8VGk+vxtXSSm?fjINQIJN!sr>V<8s;kA)pKxjk
zn*JfRrqCR$R?V1eIZqeet#vCv7sDjx_pSTvy;6&5I|-GF1v_5R=;ymZ_M97zTazc^
z)w*zTJuk|w%kBOO2@O0~t?zg{jbBu=s;Q8hUjlMP$ZzDo^Bd9$#CLUJJ#yUNlb(Js
zwwTG2#$qBu3a(w3zqT_JsLxn9?|EBSGT%+qlO(j&yd1wF6lZCrm#^<_#a+(Qkv~l-
zBzGd<*6zr}VUw5wojyjOob+CUinBc#(Dje@eu#RBBg?EypTKcWlUSWqWs!_avi+Sc
zv2Iq)3nwZbUWd)>hN|rzU;7LRoX_3AYqc%;fL2n#67vx``?2?<DPG%D41!-y1@DL%
zS^#h>*pBKww@Ew1+1>4)zixFMF&h7k1KI=%m(nbMxvSUC%0E??WlqxC=Ptyy%h}tg
zw5J_qM?kZ)D{yB!$zLp8oml<oyr(A|u9qy)dD9&nb-kYhD^GB`ZS;}m8&<5e_}%xW
z?idE$)fnb*S{;`Lcp0)hI^Ars->f^*4Bpt`J5PxfejKmKUg0)h@RlYBJI;r>GS10X
ztwtvOvh{I)&4B+0!J`7&IaMa%Bb6tvBk%3`A{_TPUdUG(#XAb_RkwA|6r{fsVyERi
zAG0wS*hcv7wxxMI@AeYQ|8@0RAiUO6g%z%^IW<M$=W5p~Rj#@(Da)IF{_Rm0vyV?c
z8NK2unq+~9TdzlAmpd?k7*^#gAirmt>1XL<Wt~5eu46NoYD(TruqBvPvD9~*;poCN
z?X%)9KEkIBk?F_y?`Y!xOWd=`F)c8W8>3Kh9scgdUP4u!7g~9svlqD#@ORq&)+FVH
zPirEg?)($B|E=-jyl{VBbr14}f16s(e;8>Ze6Rc8tpn!|5rac+ge0u~Wcat0ocK-{
zmT1~#PyF{j2xmKf@arjeEZ{Oi`Ilw$%k(A0FvVj|NBWmZ`d?H04|f086#r|A|FOXT
z{P_PPD*WT$vU4b}OvX{4L=e+xhEV*?h566@lOs5&w!;xTp=*yG;D6%W%JTM*bkpQ<
z`on%VD-vGx`+*xdm+2|3@YxOd@p`eHhO4jt#{lo$K1G~G_Y55kRiBPWR;tD2Mb<>D
zfVrZcEl#CMB%0H9(KlK`*O=?HlQfZwn?C3+7Vp@#m97}WO4WlV{(K|o#PxCOF0>sW
zdKyw!U6JB51EsQVF<cJ(n=`MY`6rE6ylLU-4{-COOW?nL^|-(4?(feB*!|*hbLa15
zyY>n{c8zW#{TAj%o={F(fRp*;n%a44k-b>#7*6y2;ApY?la9L4GrPJ@=)0Y+$QF}L
zKJ=LG%DL2PT&nJ1`gVrPpP9Yw>V3x35!8ES57;yN{o;=WKIb-QW?rEgw8)D8B>Sk)
znf?-}t}%LK=%w%~-iK-b)u;{fFPlO`<~N%!UIUr+GnXWlS3zbI7h@MoWpn~3nC{hv
zXD=E1d-Zhj)~md;zy$XEU~^}@m)Q8RlV#>RkOx7O-S`}yw(X4m+G^erlOgL5hSTE)
zN%JbEv%sw_?*ez42UT_*CU`cyvLM`66VO5)*7Bb~?nUX?y=z^J+kaq%{RF~Sv>6%b
zbo^_$&H{4zZl~Kmgze_?hD_^X`rXCmq}jd=ujfw>J<brdmAF*dVS}M{zS+!PNHw(K
z4|dB@r(kwlrGzOshcRhtzn-l~Apl4#nC)Vj3zJI)u8h^Cb=Zn$A5T!?--N4MoZi46
zv43O0IUHGUwRp(08$6pKTe_I+oY2?v`h}Pk(|re4IS$xRzn<g*3@0KL>!*LMFu+;5
z!7)3%$-o5kh>%P~%z}&l?EhN4sqiWvEIaO9z4*N}2pC%I-kVWuj;$aNN@}@)OqZu#
zlVsj`Yjr+EmCZ1rK}ibeY&}<FRh&sOMQi6P3Q2;4W-+_0+wED7Bx`#9p~la>`cpZp
zlc|f`-zC=QbR=9cqUW*a3|en2^r>U`L1}~D;EzQschqQEmPxIoYZUtFkKya6bJ1B-
zXTw|*4=efuOF<3e>+=}f6?sw)_r?twSHPs0v&8bDmJtW-1V>Y`#t9<2BTk$5Q7sr~
zEk~24P8-=(37%gYrrqp*BfScJf8*8Y@KOnk3bBmplqh*h{Ar`l)|&Xd3`=ZUF<vkt
zU$^1~Gx7MuwPxa?Da`vs*=I2ClI4*_U5}jz?NsGC8WorjLYH=!t_Y@EGB)iGqBu@0
z5$#c4g?@0`djb>Q?rvt$N2Wz5fMdu%ke>y6jI3WS$mfwbI|S9fq#E!!Vz%i$QprG2
zg9S{5#-G(q&+3HBVrJz<Qo5PuKP^l791q8+1a!2@^(>w7hhKLIDKJRoMgBhEOKJ^Z
zT;T`g&o|#aKFHpm5EOmV^XX<y+_L8F4756_<xU2&iJIMSS!GPeHuiv@Dh9>1eD<z3
zGm#J@OX!CWU5W212G^<;I^)nAA{H-=my}HUtGP%n?UOh~S$XE7qcdICx0)j;z9SM!
zx7m*QncH-Qx)Qf6#XWf}?Onfoz$HzqA0!@qZ)AA(ShC)rbS1L!t7lL)zVTC=+)$W4
zkF6&V()pOPqw<)k9!>G-S{A<tl}jG};Uj66XXQ$|BMSrIsmCyxPWI)E-i>E04}Cvz
zNz$q`&f}rdAXUZ0-k%YInkKunUcP%WQlAmwHn+t8yuBvLNvq@u&(x^@s>xQl%$;vu
zfq~~fC(C8MV_LlCz$kO%vj6DxyVg@6Pau_JIlwPm7?Xj(1AC>adew)iRzhfVG0(v)
zO?35+Y5%sIpxD$*ol|KnipWd&=hu7ACx6NZpNu^_it+u!`E7IOhVm+^w3qbTNj6c)
z=TD2zFNwQ?p;4;S&Gd{E-akZko{G)&vG{u;sb`h*DYcoNsQS{E2U<CLmgdRiC&&5a
z_Xov>2Tm<du9J%gi$f|Ld<|3o7hPu=(B#|p|DOc{3JL<!D&5izB1*$(q=v*8wb6`5
zML>{H($PqaZWuiQrAA11Gdg6WM*in>KhJajUp)6~UR}E`j`KW@^Yi(hR~-*~d%iLO
z74q)GEL3pC&p1js#S?W|8k`)z=cmnkPJOvDnM-Ov6x|GV#;0kX|JaRgjedOrKh!d4
zXh<U$!#qb8Xx#5fxXcMJNJGD-G7~SwX-)=iJHyjU;isKTb2VNn{6hzGqM*_8KCEQ^
zDbv}?>Cb5cs2yG4UU$_8FacyZ%)hSfHO9I>@0U#;`l&uwX}di>8t5n^o`ZpA8qRx}
zxJ|Dj3O!HZ<{Yp4#C7NIg!qzWNzRb4KdhaeKn`m|Q(s|Z%w(sD&_R?3YTfqAy=E`v
z*<A|KyBoshWE{FOrtMpYwm2jHe9YT+Fn5`<$(XrgP7FM4c{+;71Gj?p{E3m$&)>hJ
zgU)`_mKtu#z`~crWg8yWB*?+cqJHfsM0iApA}yEcrtpNQ$XW57^}+9~dkyi%bkykm
z!%0kMHbneTz*<&rt(f_o?`f0@@SJUwL;l0N)E{jMQ`IbN6g`&Ds<pi3#0ErD`%TF9
zDtdAAlTpS8NQ*w#k!M3R?HP`ZV%LM}8f&?%gJO(KYu;lY%i+%TIq-&u7vxHulF#_l
z5e6;XBCgZF_9vk>w$4ew<=tDTZ?B^whdHBsJw8Y8`p4ku?gl19rE3djN2aqjg~(pV
zVY_1Zcn$Eq9~m_z2J_NhIK0uUg=%%6#leOEB(;C5^VFZ(gQ`r2$8bIuIJudBoQKgR
zTe00syI^pN?ic?@hd1_TXy1ppKqPv5k#sY?=&0V5vp!-Vl}`~4>b*Ggt(|{Z$Eo=u
zAVR3zwNWT=+SA5pN!DimLs#@zDo@lj>r?co^#Rqd>9#FuBjV9WUO@t2Ke^DV$EXfI
zTHG>|^QM9@ichp|m;ezM0B@e)+_LdWB!VXWXJ7J1Z;bV2hGWh$09B2(eMXu%UX*Jf
z<>>I*l7q5_|D)WPr;O44(9~e&3gQ>SLqmrrIX113DcLddkqBgrj+tl88n)YWL@n9h
zgoSNrIxC{C(6r36wnO&Y^R&KB{yEuX*9sB)=VOgxA&T~12vacv+&netIc!_k3=;Q*
zNIB1ctFymcJ(|Z6X{ejz43!115w+6}L5gegY#6gMY-oSaV~`*y)yzBg<k|Ow8*#&{
zh?3&b^8_PW?X01bQejI^Lg&t^;+;KgD9Ckd&Fkfmo$2!<_5HZ9Hy0UnxZ6w?Qeia5
z*YuZ-Tc0}i{x!^WhotpiAl47FrQ{GpR={JELA33l2Y)=K)9NLfg=Z}$_dQHb1&^OA
zEQ(xrYzlJvJL;mE<f=c$FASR~v@X=lIcj)XSasQKKYYG*^~z%K;w#-h^Esb8kB$Ux
zlrM2#RF&u_(G$kYkLS)oRBCj7&7P!3a(QHc_VJ)9O{YJh^TN4wY8WGaAqfD{)vDr|
zdVl8Tpo@Y2E&q)8na=_nCxMMS*1*mO`(W+h1ywW4DbRFY8gs$~g5^Ggkf@V>w+PK~
zzETYYXXFP0i|)v|pZ{Lmx71Fike-kjhpQr0`$a}^?QZ$!yvK|=Zs3c=UU8%0y^i~9
z?q7lHmg+gm0i4+&-&#X6hn1w=8kBC*$x#-t<@`71j#6N9>L@mDe6HE~<DnA0wiT|j
zUYY!lURCK#!J1Bv?LXQb0rY3LfXt80$k&O6ecp1mR8*!FEqopim2<}xT%V8?Jb9Pi
z5wzmpYz~zfW3*qKkQ&vv;E_zOyI4*Y%|B&TZvncW#w`(>3^ym%q0H%4n7mqz_sJv&
z?;Qy8nu`|!i`XLa@GFjMvu({a%L&(cDHAZAF>P*_s?}xDM(0&8(bE8v4-c>6MUlM@
z+JcR)syB|#t>Y6B!w#$;&r4dd*uF~e$o*n@hapeL)e956AG={obte3Eay9Asg$B-k
z;WNW#x!hG~z9y4)L*?5PjhZ4K2V2;{kNejO7H<8bIP{O6&kh3{INQ9>mn)u9>B$ZR
zuBbnAByB{^wsxg!&BO2H2;dQ}xE=yL)>q?XKQXmEzfpQ$+U-}<;p9d+c6h(kptfl|
zL0^lHx4gmo(K==^gyq%R1)f?!aDPl!^w%5ph_uta2lT?9wnZ&yq;#hMrEI(@<h$Y^
zgR!P_uJ2e<^e?jW`}t3m1ag|c_{_WqU<L#%WuIuV2+(Vb2}1IEVLlB{WhS<x<P)x2
z(3>{_z@rnw20S^!vpBFRa(hg@XsSga>3CLPV<r|mc%IgqIj1aljaN=7VUmuEXdKIY
zaUPJW-U#B$H=P31C?u-;IxO-(Eb`H~U98m(nL{^}76;WD&e{phh+lBrfETwvn7hiN
zO~(T+?Om@L46cujcuaKF4T-l3jiHu}%bk|PwIPNx<A(FhOj=bxj}sql**TW1)XggU
zg`$hPo0!cNGDlqQA|aHMcG-!mY1ZixPM^mC5w|CEK6HHce*Hpgf3>F0uCl=2Wcgbk
z5hDBHg!1Erb9XJ<PR%b(>gOd*$y=Hq_+go#UzQ5FSq<LdwVq0Nk?$MqE?=w<^hJ6O
z`#&Yhp|m2lk(X_U$qeuEQ-j>p?M21cP`)oZD(RgL&G7sa8g^H|7@&g;%<s7*=d9b*
z<4YiRI|uZF(@)-iI@rl|NyWcExhc98@N%-G-*p&gHW?_5ndiK_#*3<2Ay+!N0b?+{
zA)@u`Ye;$7-8t8j_hJzO`{bl~*XNoa1NSq72<$Jc*@VQwx!hx*qTir#lXU_bJcc8=
zMHYSeH1)g0gdyDs)}JUW_*B!_V%1~u<0)tWdQm_AM%>|JxmhdmWPmQ1FV#D-+}914
zn;ojGwYanP=>TiJFa8hD{8ox%%W9^X@WD8<gOQb|C42wu!ST#DlO5_+e~MReC)E>o
zSSDG$wpTBh-_IKQR6XTQyA7e?kJJ8osXX)z=(e(3;-7w0WliOM6s_^1;UVlN$`%aY
z1SuqX>baM{uDjt{t2h_F`;E_+y`AiahLho&kZ)wDVPr?0s)L+~+m4n$jl0e%*tQTV
zQcs(mRDTE0zfY_kG;Oc?cFL&AS76RlrVd<;pjf1hr6QKIJ4z;`u}fnN&V$s&=Y;9a
zN_RMiGu2C*PhA>Oy0uLKw&nVr+d?sq&DNr#%~amtzhK6XiN;zzl8|MW7Zi6%i&NRz
zs5SQQA1auS*Jh-Rkk~B)dtzt`Gb!&rPfFnUr?8aX;mY1=k@6LHcU$|D(8mrp2Pr9c
zRCuBZr|CTR8vR?NH9=1|igv@lFI$v-WSf~fdgBS;!Nj^Oef|aDo{f5UI^VGOj)s)x
z_nnsS=|A6u6T*Tc3?4Y5`19)heFtLY+J|?dh`bqh-?j*?%OqKB7VHDE9@@@yTBtnJ
zIT+<P0_ympsT2ehMo5Vs<MDXq>krxF@Mrs^r+~7p^7D6U%86r=f9jL4WT>i<(rZ<l
z-Wu+h=V}?p+iw~ZS2<~Bf1-P6=|(y!svdv3SI_z?;)~<r2ziqes)c~P(tW1s6M^<3
z0em#RP>Ej2Tsn1$mZzn*Y;Lz)Nd<e1&oz6_pM(bE&Ij`TP!4Q*(9*vi>6EEo*9L-D
zK<rcPLLmXTH0rQ7OyU+A!PVL$psc~dffWQb-RHIWl|B*9g<V!+6S_cu7nR4vQE+EX
z`19EV?e=)R7TYdy9IPx}+CQpg=1q;y*YF0exh+xrf;l>U5IR&<;$d_mCwVo(8C|!s
zzj1eR=+7#h2ukz7&>JzLToHJ$sal3>3cxud7*E%AzNR+fLs9#BO`PUz({Inavux~y
ze(qJwvEXxDi?skqYqcz{e8>rA^YVaUUe%)kZu0!MFj|DVq(*p#_^G;ExeedTnzkf^
zVYMDPnEB;}`nko6)E)XY1>e3GpOug!Yo}?QOps#kHA-M1!z=9#2M!0_tM^-{TTh`D
zCQHzcc~-0Bvn3TyvxB7i{`nFl;OGy+W4{--6MH%q+^3~&(zdUiep-`PLy8q@kWIF0
zbjujpu7Kk2n3wmj)?|YF+{XaqQ(v<i#Xc$e2V|Nzf+WKTwK7*B(b0EGPg0yG2iIhV
z=}r0^tt^BzIcu^lvot%zNF|#b=ar~32?<}n>XLDoi8=c!_ZGDOdEIiJvNSdHf?$Vu
zor*bsK+171EgR~WeRQ;tS|`<<j^#I1V1HoBHB0$!)F}FtX^b~i^}0IrbxP&$+ra!;
z<4{G|pzbC+y=@uek-Fime{L^xq38L^%=749gL?Iq3)B>sKil+X#~3^usfIM{pmi^|
zjx~0EmOr77zJo_7Q|DF=<nFBJd+44}8M8fCTK`P6f*Gc?!4HoqYjrz6;r(5|>2q5%
zG)!(pi0)T$Z-*VH`gRGJAr5PdIl`xUPXXOCfZqO|`56I&=~X>fov^(&;&6*A0g;8#
zsJr@B$xA+YRNgs3;W;D7)!LAl`j(4trK0_Gbw3aybYf-gUz{5^?>kuWrgLbhu`*p;
z*dt;9iZu&nPn|{AysN==ghL_cSE6;oIEcG(Q?`Qy4?l4q1GRVJv3t7CMckNg@u&MK
zs+07h#e#D#63)I1ch;i>$eS|2v!q}PMc%&NTDZ_d5~R34Z=RnT8F<ZQJE|QjWykas
zKPBWLrU!l$do5FZ^=)Z}1SYm+GNkhz(-bn@8Y{vWqRB|LCS96ZdY8Fc2jDUP^O}1-
zsx7Kk@#fQMw;n%f2S{`T0A+;%$k<1Bwtp~>Z!sYcdGyxQpJz^P_VE7PrF5|OWHZcf
z3c#Lj&~QJxcw%OB-oEdZylU$L5X|zZ5)<VA#XSeX>hSoC=dROuBpqeOP<GGc2x4Ye
ziu(6>67>-XRoh$b@jH3D4x&-wqF6m?D67{rjnSx2t<sy}Y{c6jqE5QWk`X;?4W7gO
zSTy@W$rHD^;xlr;nT0dvCbw0O(jbr2?e0Muje%Q*-H|nu_8^6z;NIuM{DNoqR8aL{
zJ+o^knbo~H-32^O*l_)UD6Zuj)DAfn*Z{nAX(_rk(nh(EcRy^r|6W7ujsoBH9LR9^
zMS132_>H={G!5R^(>LA;Gg*z3UBEh^;!+ZS75`$H-swjAqZ7IPsRZ~J`<sB14!~pf
zt?a*rms{+cZ&Jv7A=Qazf6ENZLZ=m9kYC>R-p^uJ9?*P}Lx@&e-I~SuPiCa0H7D>+
zHji-Ey$8tOAlOg-Yi*bBKds8DA4nP<6j*FxqjIu9;z6<o%t`KgV;>R|u*TSiYt@!H
zxET1EeeY^NGSlIgL_u4y>ybqg94{m6zvUurHKC#u(w;Zyb3d+u=Uz{KYMes*<;V9P
z>AegCPptN~?u!O;qcX}5i3+B=M-@8v5?LsT@J$^qZ`;O(hKSkwV)D0}Ha)8?BqWn_
zrqx+J*0Pe44gT1Qf72VzHxT3wukym&`cTBn8*Pf3%AdYhUmO7F@fP0#yLcrMm$t$t
zEFHz`a^v@p?=e~$N<_l5E>fiUyv?gJv<aL9xluj-(f|y=8fM+(`&gctYPD}LbpV<8
zXtL_^U!TzMfsFR==1eg=C|)pZ#VhVUb4_7T@o>S_k9{2>e<_n#tIrmZqjgVt9i57z
zdy#1&LZrho6c&))ep?W#iY2HPoxJK0c=QPbU=ST9&MV!Y)*AK|z<91)T&~3rNiXEp
z{Qh9WZ4GE<J7sf}%S76ys#1WR7C6(P@Ou?T$8$%zUnDjYl6~C9a3N-$6R5LHZyB5y
z>^ha03kMKF`D3CYBP`}~uQc<a-uJ@XGENwL9$1bqWY8wknrG}9QbCjkxJlvlVR`x-
zvw-pYT(fcP<4Jwh5A<?G^q$WJf43N%bbqO}xOtU3tg3(4(CYV`Mnl`Ij*T;y36xLJ
zq@R^kc#BTx*f5FVGbzIF7;ymk*13$}on=c?f4*sT(aS%`@byi2C%@?B@}u$(I_Xp=
zJF#o;0%bl(V<LIWo;;!6uO?%D&TQWJt3b=0RV6Vz5UoZtBz<wDpHZ>U#bbTnJ_!aZ
z(LL9giheB8C#)e#xpZBt;0<tjh39>$7i+#{g=8>(rg1oB=4Wp^l^>_5C}PcMrTS90
z$OGOo*2Ko$XJPTso_sGLTSx-$!51^>uh)pli_F0)EkCwZh8r5!2Io%-!CmQPhSM&S
zwd?5l$7OQ-)&rJD6syB(Yv`C~>GYQzYI#K#x)2HFz_SV=kKso)qu+F6qz6$~l`Cs9
zyvoJ<>BkWNwT%25c+CYR&UEpY)|$J|^Jaro`&h-P)1%9JzjHmwrxg%QN`_j2RBMNR
zKcKF=AE55D+jYFd8;g4%Ajh%NssrMr_AO1mG#KV)LaiwGL0;J8m-Fa!#zkoGXU<y~
zHO`t71&Acm20A3mP}GW<JS?1Xah@^gMO%RnuqRIS%@nm;VFuWD%d`Q6kE$uKWM;_t
z<3`BuM!E(7NBXRHyqd(w9$1E{x(k#VVL}#3m_zhNZKZ^SnV6ddjR)yCi88vK+bJOJ
z<&iBAeb{EM+4RGq-qvdQ$bEMn3_H-g0~=yG;K+jrB*KoiZnuGY1YCN?`Ueu3^Uqr$
z&zN6XTzP2N@kNvpIXSrt1od`GJWSF)fPm+36TnjN^ml<Rk$VQLIu@77ak-CWwHCQg
zFe&z|r5Oy8dqYx&k3COjqtk!<?kz};+S@+If3gB<SA!1jIoW;Kt7|-gB>Y@nr}@yh
zXn2h2<%~$ViIzKCq(%HDed+WdD#Of~&+?-FpdOE$<;qw=J@jdhy1BG;>+WqT9p6c0
z+HEt9^Ur%Oi(FM4;U(^Q{nA5d`^grN#IBACld9AXBa4`fgz8tCqc-3)ooKu!MTq~X
z7+*?b6fX54(~f||z1mx*?LNgWP<&B)-o*oth;g$btK!)h*ch&wb-;gzu9ru4sCe!V
zBwpA!VJnoXB+NtS;u6=$B(WfRzzgPhG?U>k6*-{2o@&pah|HXAVaiWZ<vgL3mVj;A
z%l){;G+yP5>b{LV5@Ed3PGph`4xgz&<)><Io<hQ(waG-oGfOnt1gP0@N;6?`Nqe%n
zP20?mYLzlBWqtnCip9vu&Ed5(^m}f3Gk(~k{G2*(MZt@c9(?ZMXRoIw0YU+1wA*9h
zTubUc=aCg|Jv9C6Jd2Q>l)s%Msp-^7`Czkro<4B-ml&VIn`)lG8R28Ts>S!Oo0{2;
zC7eX0r(96Ydnek7igE*`-0RKU@UzqkPxZO6QOL`sB<Ut`8xI%X2%G#=gGTn5OGE~O
z(VkE(LXD`$bzoCveLp?+qq9jv_A;6(saDHOl`PAYAK3CY7vtQve%@K@@$(mhw4P#S
zxFUWH@-C*aPW!z9`6B#@vMtlvdG5j{Jx@CJLhe{U5NalV|MPBv?2ef>rN=XN{YKQ3
zQ9Z~bjPN2*j@j3gwOpG1G?`q|7t+MiC<AI}5LR~5qRHwpNAaPHBGv-`)b27ND2PrS
znREWovwR&~?AC)82uN0iGfy9oks>YYO;?$C;fddkei=WKTk%!;)0HeB(JP+#MnvX<
zU>&-RZ22PsXMkUpjh9^xFo%a|eK^QAypVll^T{BOd1q?*f~?&)m-WuHrMJ5aE0Jzb
zER!2<GaE|%#=&lmYqD<U>q`f5(kr9Q8SR$bA|9hVgZtyPC%WyHKPS^M_;l*)wmGc9
zV9Zsskk=-aP7NNGuefna_}@3-rL`A${?4lNoRI3m)f=B(7Nd<?Wzi6b1`R?IB(LkC
zo2(Czkg;{A6)Hla-pUSy+eQtGw-{IHmsKe^eu-;-jcX8DgsgElc%oe^RuTe{E}Kb&
z#o|7i;>_rk9&v&}=h^bG%9yr31$5s5)9+#zc$Z9W)^DD?$@r)mgrvr#$uNxsl4~G5
z=d^#xKY|wOIOTYcx~G)f2Jo)>Eg6Q2^#Kw$ciG?Sj%>C3eakD`P(Hr8cJy8p5Atv%
zZuaU_T-_S#M_>iHOY7d(xLcMD#HhNaUG%-y>Jo3Ixp8tsP~j`cvkbKsyQ5{{g_`Tt
zu+;V;xyk)N2lJ~ut=kS;?7Qg=uWR!o1)iD?UEbF-HHp3{5IoLD453glSu;9v3mMYs
z3U(j6a>Q8Nx@J1iH^vijFC}w_r^7V${rx0Q9((xBnY-+B(y##zpVw1Etv2&Tj^ft|
zI}1dBdWW9RgkxVn7zhNrCm+c?TL=iU3Oao(VYxeYyvl$+cZgttqCgIraN^U{pl?d2
zTs2rB?G9r)r_kc*uoqh&Gi=uB6aAad^$PQNukh^Jvo&Y)4w<{Z?8MFlU=fv{k`cn6
z`LxZ2w@h*>Gh}urG}9amHmkR#EF}WWoebO&t@gurH7Z$mNfs=hSPJzM1OCJ*n&hW+
zd;e3Iz(k$zIYL+r3?Earh6$4yW0^f6Y{0W=V7Tm;=LKy57(-1hDAaTDsBR%7k9sKY
z3%z7*X0>H{DV@fFcqKLu`{ARQ^KUk|Y_fJE{41u^E|Mc!q7wjPuCw_gJRC;h=;Hv5
zSaD@6)>?X&x4hORQ3Can8YyFmS3^drh-yzwT%uq3M$k@FTs*=0yD18)z!OIbkAxXp
z=k(LCH0|H<!8^AYa`y`5+IKAgW^d|bKGlV?P7801CwchM+il6qrVC><miGwMj%4uw
zM2~m1aq-V~medP})WSwytACYzrf{=L;#!w4dNEAqPA*|(v%?*j<F4{yQNoM}Dt$2~
zeb^gF*H9smTBFr;Nr@XE{|qsLIPc5ywOy6%9mb#XU4xYTTiVL+t&rYypwy{WVUIO#
z5Vu`8npCY`a$U~q3ILnrt<(a&N88w-C-dK^<`_+T-vc|1qoq0=e;3Gc&w68gCvdTk
zM+m*PU^B*0=M)Y&v@o^Gd^_0envnrMFB8l>RM2@zF7X&FtSUY?>|kD`?PxG?XYjX-
zQT5O_0z!B2aYa;sKM3&p<k3N5f1OZQ`j&0vLK|9Sl63zmvfvN14e)>_N0IxY#!rn1
zd=Yfk>Kietu^hHTtKaYMi)Xf7Tvt=y-yv#XbYDX4Tjq>yoRkMUO9ZUskDWlz{pjwc
zIJ5)}=Xv`Epj+HL)xVNgKH?O3Djt88%cT5Jd?Ro5y&TsDR?UoPAHJD|f5=AcHIZQ1
zlfVR008SaNeJI;JWm>gcJpUrg?cKio9o<=4Car}>96!M&g!A=~VqBVwTUo4sU$qqR
z!_}ETt)sXeTpYJ~PFEX$_(`VpMp|mlbKF4<!61`VK&jXtbkFhgHlDLo{43X)&jb6+
z{Whk(No47|?J)n!(^&muuWMml!fImJL%Psh7HW7jaadf^Ttu?lG==GB{mk|~<$SNv
z<eI~ZjJ`fexFzGnC&TcK^C*|^H+HEBVaPK%;>i<2V?8xbG_SZs^NEOs@pjG2GBOGz
zWxo|?)xtM;uzL;FM$QwxzwNd0Ql3;JY<IsTtl6<(E9bJ8zeS{sg;sDu>uI4huy^;(
z#fOUtndW6>*I&*q%7yLU+Y6f`fspL8Zl^M;<uoFuXd+C$xm>%^)XL>9k59xay6dq5
zcH|aA89_&SC-2ccwNwotIfC5%=_WA`wqQ<sUh&(Uylh6QzF+=QP5Wi>i2N4Ah6Y|>
zVP&qu+~rfCj3+*S&GiFv^P@$kBJY9sp!P@yQ|?Cc+x!%0<Fex~=zfr$dZ8UOZa*$n
z_?B_{Dz~u1m7Uj@P@<TV!sfzllKj~ulYk%`8NY($uA_O7y+8vrT_jb77IN%11B|of
zc~%-$b_r~kvx(kbybZj{PWsjZE2Le%JoJ1dqruaG@smCIfSU1A`SMal!Ai`2JZ$UL
zd8bzKK4+tL)EG{KFc9^}Znsq}n0eK>-qq{9(<O)(%yv_oZ+Oc<Pa9#@GjrAZik;#n
zU~r7+V77PJW||x3C&R&|Q<#`u;N)j|mU#I@VeH)V!U)P1FgAIPaZoWgKNmLLl>ln}
z`QTmDjFCt_`P;|09wmkMp-IsTx1pYR1)KEP0FZ<M!z2=t`i^Bu+1Z!XH|Y1F7qiqk
zR!O0<w|U_`P2+itC$`YVvwAZxRiDIZAm~JH^_TWS76@B7UP=f%_>mZTr-26Xkj&CO
z2)Y93$u;ZDp+xL#UxB_if8?Ic?brWd#-?WS$BBw=`mH)iCXYD^(K~4bv~!*TI_!|i
zsH%K^WgnPd={nKB{4wRrv!!^@@OdD%xhV5z)LbVL{l3}767uYRhWE_(qxdzK-~5lD
zinf2|)n1GQX;>KqD{?mP&egS3s#}k{umM@PjrZ`R&q>reg1XLNLgnOqI??wrVdi<P
z716P}!N*rZMsLPon?8xo<*tm&S>tnP14~%=#O<ZX;4j|~^4eMaFK}i&`Hyash<%ZV
zWz^&@D^@)kQ^_`ZDDw>17s`L`#_>Gk#S`@`e4xK+<0`-;NdH_hQMZIYk#y#{oMDU+
z>pNWSs>N5IyN6Z(lLc@`&|nYqaEX3ps)33L)9XSrlXOlj(-)Q5)K`!Y-|bU4fhL5)
zEPSvdt<j!j4Kuw0cc#LV6ZV%KuIF6`->l2MHVBoP@|5Awa0C{M*70RpRW^i6_?NH3
z&V5Hk+GjDq1(<CoPB3-<Fb{fG#X@rZj&S&kRhO^)n#^zDA(xez8>semuBV`UDU3_v
zB`^LD116k16>TD3jlyy+u=6P@up@t2;<x9omF<kJ1mFdwQ_w!X(@1Iw7*j59-p{>_
zhT<s4j6j`B$HKfP27A4=qxbY{!Q7^z&I8(o-)QqHu6L@lCfTll$p?DceUcuNm`5mn
zb_(f)tT|#$VR6+!u=uHzR-aD-!w9K%wD3w}mFB#gNqn9Ww&lY~s9`#3p%PnkE?;3L
za9tU6d)+wGorkei+9MI3B(}D>)EjzOM!XQY7Huzo5b>F=3&IH1E<bE`pB|T%;c=O*
z_G!^=S%1=%i5qajbhbB9z0vC}+X4Zx^YV3;l@1ZVEM`x*crg_P6LoORc$EK(sj0ej
z;#tX-BhaXDu@y^dvM>u~&GL9mg4=&QN+<hAkTDdT-y8I9VQ8OGxd#A0WhxZGm1S@e
zA5bca0DAQt@dH-Fm~S$Jg>-N2Qrs@FZ+mERQULiSmv31_1l<^wp9pV|=np_vJfjCb
z4wRIrQbu}LBz?3TbK*f!YdyB@k;Ka<;&R|bj{5~{EjW54DAkmnv)Z%hBMW-+yU;OK
z>g%0>MA}VEgXf5raoSx5lFZ!Fgj&<GYaX-)zZBO(SJ@w9dk!<CA03t?Zj6^{bJeKk
ze9B=AX4g=SZUVL%$S7jr-4U9--VxPf;-7lx7php!>!Lx!@+~Y0@3G%`X#w<pNZo-*
z5jgzReX}E`(GVLo#!*Ft+kw3Fi{cj*SM?7Ax1?g#w{+n*1|h72$MpHmv{O6!ne4g}
z4AbK5l=7RL${355hA2PyJm;FrnR;nouap_|+tOl%ZM7B~zSDW6NSY@%C{DH*Y4r=L
zagpWfMAExE8Qa};9$mGb=_;yfu%sJ`&$xq*5*}YnhD`rRy{oWj|D31Vn>d}}FcG6%
zx&7`INV`m~a~`s1`3wqO%$WgeKi(@SNHFm_tY3)J199w3w@lb5r~ArYP{n9x6FTk&
zy)tjDbEEWou{ZgjappsDjMG`?axZQR!u)JApE?+9SHDyS2#PawozF5+%DWlz)EotD
z740Pju^6gPGClP;LwUPrc^Ik441BulM@IR_C?Q6cGkWU*yK2!5L$tUj>SAx*g<Mbi
zbrioxCs7hLX8hIsKJN&srei>aZxjgSN!xR}cz9qt9DPE!>B5^c8`RVKwy~=Ba!yXk
zbsvXYaZk=mH6jL@iLZOlT<#pB)4jWnojto<kb%B5QYLv4y*4qUF4bVIv#EHP#rbQM
z>ai1*n@OtB)=SO!_ge`m>43~35bRXnZ;ejR@wlGU_}Rn_6iJ!J&#RI<C%5x+Op*&u
zU4zEJPPg%;6CQMCVFSJ@Fo|xLBl)i;)+C4<ytAHns<0jVj`_i#(ni77r&A}Xbmf9y
zan&n9i1jhs-AJ>>n=c`pi^r@xQ#sEyAJ;(sFhTwq+D%T_4z0X(Yj$DPWYA;p2Au+K
zkB~sMOV<ETp!VLOZ1Nb)@CqL}fy<1vY4lR!Bpi86%j+>&Gk~uYYrvl=Bb#edZanFp
z9=x-&R^yreSGI25|ELLj6%)N^8auI~4nArwq;nEMT4A);lMd=Dq6|aipurz909QRM
zOtZs+g2$AbPZZ|%OT9@a0DCU2gXZ!G8aXGX(TC+FY~~VVBpD#Y;O~>8Ox)X^?Rk?^
zc!CH2G{^FdybAHsRK<J>Z4cWY?})i(Z~j>HN_fE&%?ImD0y~kspuBIS02`HYAf)-k
z`p7kUpM3^}QQul~D3#9{=Dd%|GrQmQlB}TVr2&K9fUDQ(QA<$5g5pG6vN#c`4YKLr
zb_G_6{#cxBeOQ?8HF}(nEa7X6X4cST)gn2d%R696*ajsQmppY^tUT<R(eidcxL2V^
z*C_0uuaHN|oNb^yh$6NTK)+kT3&P}P#|yW@Qc2bA+6PkeaLgX8IDqHJz#o$ivLgWF
zB3{eOvIs<K?Pvx=kT5Avu)2vS#o#|C<q64Ka+Ww;Q>qot(65(owLxFbS6SMcqTGXj
zYBXu$Q|6`hmBwSN1;f3;H!BL;KOXusGX*uQT2aWTofMKy%tXqE=WTQdWwbQ>PPDYm
zH|LJ!F_xv4VbW>uf<~0wmmaIZukhHUcw2*vsG*deF}S2W3i><vAxvYD{DUxkFm^{f
z;|>QWY+H7pJElKt*eBIt&@J9%2~?h>Oszf*OXPG9>#vn2)^$&_)M&wbHTR7#3rk9&
z!h1CF3f60D1+|{nAW&wW^Ju$6r&4fCJdpd2l9Pt)8*GX0s>ov>>o!#nWyNQ#8%NE^
zrPU8NVz5EhZCbO?gG?q4$bJ2(g07K$;PL<lx<Wj?^V`qg0STO!ot>1^Z)i6T<Osej
z1}JxIu4kSzoeQGV1C7bpNPs;?0LiJ?_wkwHKp;bEOG)TcuEAXlJXNB>N2B3u<7+!#
zsEKqcI=-b1pVhO?nBAXH;wqW4sC1c8+8Q!=8Llm^E^hb=yhVMlcGcy#JMMOi$76l3
zUr}+%^?ifL3mst<yE**Yi6}0O^i|Pjz8#NRuoWJM4&)@ihdDE}OX$iZN%fg4YTs?$
zGrg3FPd7o?`mmTFEjQ(h-Dx>^4d$NANZIa9*guMm=${_#twCJiBrw-ZWe^HBK8Ix5
zGVq}x6Hhrss)6mJLHOGqE0xqwyL{cT1B?>=9*<QH9NKd_qpGeFd@&hu@y}+oUgyXR
z@MXr8r{)ws<?ZK)rQ0a5{JB67c*CF)3EWZ>!;oj}h<0qrvV5@9?;n9{?o~5+0@x3?
z-m%=R@xQ4RKA)39LEq=Ix_XpNHn0ck+Vo)^Yi`L*iJ`(AF1xHdEja5lt(6`84l0uI
zqfGa>f+vhvS%-eAA$p&m&+&{~LMko35$Sw4VPQuK!?rCEK2;;AxHoWYR{tcI+S^W`
ziYL#fD^3-^P_)URE1CDJO=Svx8rNOj1(yOJ9b(m~_;;?w&b-DRF6Vvw3ndo9_~fem
zD%+WP5hh1{+dfWfVlMVo7s_u~CRQV?1?TghA%14K=TNUE_$+J&na+KQBis&q7*|n!
zt;dN-TM+)TsHO7yP3Njsm6u1+6(SC*m#p5PXJz2!L{B(EzzU4jIPxA7&U@4*U3_+R
zr7vfYDMFb#7nJ;`$o*-&_WIDM>dcj7X6t!1e9bHa^4h7(GB>q*rzOsK*-_Xlp(@#^
zWmFyVy^pf*lM}&dxy{P>zqac}$!+@iIkjSzhUCADWW?L$6xe2E@!)oe<O7%g#;i9)
zZx_*8YS7MA#rcr2pB)Yjxvf|T!PxFMZ%ZwQm~Ie|PU+dzyEPGMCIsuBlS202TJVeZ
z+fjt%nC;LNK3p@)1158+_wJ<<|CVhx3T_uASGDhgdDXVA0rgk063F10+2YM6SSkDb
zS@_mOm&8IweYV!B7C+Z)v*sV7f9pJy&)x=9BsE@T7Hd`KfXluw`ztF=B|VUOeDhzv
z&ii9P*8d2`%C??MBcix*r-FStx#ou}?0KNi``>x511s^nMavRCYMcKMIGzh4$TOE^
zg5n7JAn|WC>5Jrzsr`>#HKi>dD3K)b|NDpClpl!0(@779_98!Cnp>Tk`A)&VGY71G
znT-r7)okavr&u&kpA8WE>0q)s3oN0u+ZN#simBaS6LjjOTo-Tr`$_(PONx^H4|gJe
zz`L>wFI{u84o(HF&n)#B*yU2W=xxe-HN=03{BVPpkN;1(YFgk1P~5cEhS;!GS53@o
zUw1{f<DSz#YF~r)m_ajySOwftJ9qbqE&Q)vl^@YA_kVHpgkSHywiW@U*G|9qB>SKI
zs{FrxmBV+}|EQAbQxW_seNKgE6#vPua`@|4rEtyuN7ZcfuWB~h6WjfN+6e>$+BW5H
z4h0A43E>y40xbIUecqcoA(emkPVEIo%J)U9Qa-^&(YYNXp}Vp~6sbe8<)%)K!DixD
zPQ1cXxyoT>!V$dx^MEfn|2i3*8h^+i&aH%R9LR7Ee+-w;qAO0;T^ggkqf-381pc0o
z{)v@~pX=_5=<1F@hRamY(4T4b#=UiqLXCadk>Y8EvEgz{?i>c<^bANj6)aOq@NDz8
zAJ$&vh#)iX<+u#T@l)ubW}T~?zxRp(hD+eck(8proTppUvH1gVwGjk4qR-3WZ&Z^@
z7QP8}_;ZFp(JOpwis7*EHDY{rI;(qkwDpQn&iqiaA6VNKIb0DzW<MW-`reAL6+xNJ
z*Y;}dj&nL(B!nJX&8wPT^~o_wm(krx)4WUw>zbjNe~G-DAWk?rvS*eZ#ps!~coQ9|
zsnE^7v;weBL(q;ZP`~C4V9Im<L{{^;^95B6+RMn4YsrMN>yB{I<4xEGd|PRnn|vmH
zH#?#?azC$-YJCVla;*v&q}LMi0)cN0ajZB6T#!azF9~ky?A!C|OuzV-1K`pH$y$l7
zSvr9YMNyaQ&r6VQWQ5xdEzUDl3IMA4XOzzYkgBXFI!_+NzI#t7^&&TkC+5Mft>V{>
zX`VT{xYw??V_NXJAfu_Fa*g>jvkDAf&=g^K2dFRO3F7=LpTBrmnXasu49#%ex;Q+?
z8=BbSU#C}AJdt?L$LlD(akAc|w36;D{W|7Vl&UJT)>F{iywJPNvg=*%ePd*X_#Z_Y
z{Ru<Q)Id2+-1dLF^*^7t-Qq+}M4%)83*av~)ODA%kou-QIp?;yF?dAHD|=xmU0-f9
zZ913H_VcT(Q?4#bjDX=TeuFX(yd9OU(T~WVA;9-uxv{Ie@xfK2iv}*Wmfx)?NZ(%8
zUAz7)M3WdZ7^<X%L?IVW90=FBJ6;!Fa@>764=2$9$H*?v%ae?v&WMYvcKon&$`<xE
zg4PDk({G)9TKl+W9$oq$SIY36m*%;O#Ox3#BF*j06syoYe(~O7WU@%peIf6>P694w
z<dga7fNy1=LEP~1{p`&zK;TT%g(-T0Ao9|I5)(VC4lJwZLDEbVt@&N4IIYbNWrZ#4
zvt&)_LZZ*28;;MIDio7}Rdo(C{nypQy_Ob@w2uL<%)$6Vvl3Vj4{*lQR_?u%D9vi7
z<FBIuFJf7aKkb<>+>hdSm8+`0UzP6i!O&5f<KV)uxtQ0S4d;DQkHv!acyb!c=I!kZ
zFvp#ggYUG2y_Ef*)o)k2<wtD@$=<(nuiC(?n9D<aJUgS$n^~1-&gTz^0hx^!pV~#$
z49oO)vr`}cgGyj%v<^5P2i(*|?oVvif@bWm3j;YxUM{E+iqp6nRcoM1D)0M$gWc>*
z)Ge&(4zIO_-{Y-Of9;SyWha081O)<cZubWCu^JQf&}QsofqN-IPCAb2WEJbg{`wq+
z88o&Hu1QoO)mKAU0A9<mt=;8$1m#Z#2``S!V_+tvoc`mle@^OQKUUfwoJmGr-MgKY
zm@avI<-KyyfQ#Z(w;Tw<J7VhX#?ME$5C)*vcH7mX?HMLPkOi?sqhYH|)R+ogpfgga
zVLKw}!k)s*A3Mq#=17rS{7$OAW%P|G?5l3$<y&mqyJ$9&+zcpOeV^@C2^~OgCxtc4
z-zYC(B}VRD{xA7obtUdktB<1Hsf{0Ooc^tyrw&n%8}Bj}*MZAw8r!3JQC}-TsSuQI
z2D~$dSBlclfE1Zzc<7+0QTeqBh9B(LHg3)vOaatGSZ{yvhOj-AR(r13?*b{3-)3{D
z#ay-Jn^WkEjwMf_$ESnlGECcR{aVsYD`Z+qM~6#~ri1-H<k5J~g>D3{u3Rd6VF(w%
za_>bWzgn9aHGDi%{`!OZb{y+FB`P2LTd1ETqRf+8iRH$6V>0K3+1%EqJ7ecJdr_jX
z$1U|=uclWl?IWjuv>wP!U!BQ2E?;jX$a|Vx!3`?iSa)T0G_<4wHxLr$O1{l9d#y}P
z_Wfw*vciag%lTT*-IGDDI|UJrzy(Dwl|1NZP=MoG?fq8Isb646!IcGNLSRJ(A+X|p
zRr}xQ7jS@v5LGm>_b?d)y6~*U52qD7E4y`ld6VbyDtYyUTBTWQ?T0OTxQezw`Z*xP
zp<5wPd&C3vnvfF22~8Mb=|W<n6I1_2>}-Y^W<G$4{>^(cnVGG(Iw7B~<a{ZGc+PY7
z#m>U*7DTj_4XCZ|^x0M)WcT<y-S^<ZmcR0881mncHrAP9YpjXF0{A=uJGh^hMuG}D
zSn%{1_*IjO|6IRmfIBl1DPMD)7dW4P-gtuMSZ0qn^x1smvEnPUx&zX7X>|S*O1Cmu
z;m>Ip_``Rk(#=oc=o%Q@8{JYo)wbm1o|vcaH-ZTfsCg{PG~Ms(C1cpq6#I<lM;cgg
zr_0y78<emPd-G3bH#kXu4kD1#GSGvRoM>$cxw=r#Ga+(!_xGUdMm(jwj_M)hCGA=G
zL$rYr+_f=DREqP)pqeA?3@-a6cKtAJX=l^E;NZm7(#Kp8xCN&LxW?G_j0v#w!_S4m
zW}oLryK@~6l@nbU1w)w4%_Ah<m-!XO(SPV`T3{K>_8`8WFF*a1^RMwzTQr^sn|zkq
zvgn_-rnY07;qJTQGiqR{%f9|KMX7S~8(tvu>iw0v)A(-Q{;YVSGSImiros<j?@LG?
z-R4YO+w3N7kvFSce;h5dQf^RJ8Rv~FgzF)WrLXnq#g=7^@Gz{J(2h9}-2g>mGzBcg
za}_WIdlRE!R_G`(aj*l^hq5%TP3w+!0f3UXhRZT~Nd<Fw_h%I6v|BRC1H8-EU719L
z|2uHPZ@Vbe<@dDZ{kxx75jW7Q*yPs&?w3FDm)|u65jWQgZR@Hg{|KN@F8LPNEl@(=
z1SD6Kps~(5jY`p@D;G}>N4};JnTOSt^5z-OCmwP(?o3<${(?)dQf>eP8MY-+&bh|(
zRwjdjeyfMyJ<jfDXWs!{dr7e#VFd+IG2U43#-O9B`t*48t_bv`A@~ZT&3AhuJU<FU
zWy!9+PR^`psn8rRUey<WJle3(lB7&-7c6O6w$=fecXD2(dCAhaSC^@mnYLJ`D81V-
zjV#_sOy@&*?@KhOH9aeLQRr3J_uvWkW;nW7XK-+zd=G?jO$Cf^g^!N(L$3odgfg;}
zlT(;9BVE2zveu|IjQJPIFUeM(H(pk;T{qgneEZ~q*5)yV-MRmABy6M7&24&t-&|oT
zJEoV=E3$rvOm5I_`rp60`m%cVLahb_x!#p&X;rJcwr!%k*51E0nZH|CG6kG^2q`yN
z$b%#fLW)J2u>*bnsYtiENBre2XXY=GDGv5;y>U0H%`p+M7}zy9DlVN?Z0c{ljj~eE
z-wfwLy6(QZ9XB%A=80f`Fy-<ndJ_~)<o58OCAlT0$Xa84-gl#^)h@xp#7+Wq@gx>L
zm1?0O>VpdkjHLismMvRW`P<})zU(-E^Oxi1EaBnfyw5M{_(9x)x(OL9^SjJFXE*i(
z^7ma%!RoQ(6)t&2V>sx+p;KA0a6abbo~=T_TG7uWG`l1Zfk4l=M_PU!*lB0t{y2oZ
z!H3N|D>UO|Zs`q9(Q`k|#${UM#hl$JE=p|v)~!)ygWjk{^|R5SM``2>z|*j=LM9Ir
z9#Z<CpzRa4a1oK88&mwTg>qzptMRSK+=cuRPIAA#ulssE5{(RJloHgPKE_LB*Mz6r
z`f<wHILncatvdmrC=M-S$#?}1-y(uCv$58vxhPVXH<t<ODd#<r6~d+MZtXu_+MpKV
zjvsX}Cm088Z<bv~(dtc}Jw5u^-`Y=D3c&(o3q!9{nLs8dE!LaPD{Cb2(X{hL6P9Kt
zgb`w`U43y)RaxI|#IB!N_jT4iv}|}@*SDTdE&Q<+=bD5!LY8BOKh{R#F9^=(5eL&)
z@Mcy{C`ohcRzu_d_?%)6DunZwgZp#2GtOeEzh7xwkJt}}=EKJY#xd2@M@Ef3291U6
zdF{<k3fH%4?aKWpB$5j2c!jCF7oN~HB`whPU+XE$lP}0r{MzQ^?e-?{yj~}UtNw4V
zJKQ<F<KLQ%L$u_!kD^^n42eb>9L;v?SBj>q6urK{sPDPl5(YbGr7=jlRca`oqB)&h
z)H0>@_fPVvAd|b3w<~gMY!EJC*6Wm~=28S*lYT$DuKSy@<lIoV5W}0wuuH<R**Q<I
zax~r6!yrzYsG3<Rt3~r1MG7g27(1GKqF5ZRUxC@TcIR`jdK<!wqzTa2{&uN6bCQ$#
z$Em~X=Jjs5m{T=Qf7I5;;(zLo-5)khJa0-3@~omJHq3spl%MrF>ZoLPhhuXN^@FQB
zcFzqV*&~xkn;Wu@xOkYHvJgD<RbG!<JG1FGSBm(2n>-9FjdA-UHl$37IH~z?!|wEq
z_MX?Npq0QUj1_f|)IZ;go;GLGGIE_Hf2wpIzN<2O=3(;H$Hjn2peHuqPbe_#V{zg0
z7>vb9So;az8(Tip=Y#R!xyXaHSZ`NwISLU~<Vo00kwJy?0rZeieQAz!VRqhG)i|Gd
zvoo2HH+5>$^?1pVvL>N`GwG)IerzEZM`l8M+tcJN7`=p7d<_y;d7dC)!9jw^PJ<aR
zM}&bB*jcDmGC*&CX&e8e%E)#>er}^|(oD!7<4t4A2??%R*&k?%=+5`-O_jO2WRHCm
z8I=y!OnpB4@c31`^?Q<*9!*~%z0yo!Hs0BtzC(=cbI`PBxbS)$JosQ(RcMF!wG=%s
zFOR~eap4MK{$!tLMnKLbw>@1{rzaTZH!LA{o*vN9oi7d_DIXCZkz}gQnKz9svqC+$
z6^6fcRe7g4J-~itI=0)3iDAVV2gMFgzv5vEa|B(*eNUDRm{<rb?6|g~%72Y%*04el
zS^s=*Gv*X^GYO;%w?%gR1hfE+st6Y!AOc4MHI^prFQL7rVE^ApDnVf*G(dgIrj9|Z
z!Lc*FVFS{Kr&pkQDZMKNCEnPa=SRImPibl+eLMM#o1Qs0RB{ScT|5)nT-zdcO(Ap9
z#cmxnbcgQBLsAT8sNp0{v=u?U6~JUaE%jCy`O7-Z%SU7+ygrBTN11K**W_$zeWxqG
zdy*-<5Lg3#hnOflxX=bKnHk;MVLzGPyT#?pVeDXBSXVNtNO0wtg!!gfT<e_bG(QVm
z8}FH~jKI(Y0Q@P!8E@O<(p5Xe#VpeOrLNYU@L$3~3;4FD`F9JdS(xr<-e8kOPe+v*
z)Uju8x+g|YM&_=yv_dUwu11>%CE=*WpPs3C+w6PpHTMD_otqmara_V}qXo$m3J56_
z!j?BGc@Nb`8a!A@md2_!`(PAY-XK$%>Y^}W1w)CG?JXAbGLpjkyx1-Y+7Zk<q!`2y
z@6GJa%k#lz%)W2eZqQi-{M5cz2Og0#_?@sDK)*$-YD7jq%DXFsD_je%2X6+-5rAa)
z`fT)FWHBM|U8KJOEI5{YIo?aSKbM;VYGxfM0;Z^Es?FGK{!Y*CUJG_s&0*)E<&ZD6
zLr-n;*@>|8mONqiUHKwJlAq4nA}12LbJ~E8m6!PqopiXdroB9Sq0e{h?(?avz%U*D
zWNPZ-%19;WvZm*X$*NyPWRSOk#ZsKG>_3ta0ea1c5UzB%1n}-1bS!^*Jsa+c{+vl1
z)4b9{93*~ENl4YmXyXW;kg9LeW98NSDm#gA7BK|2I17JrCSF+K<gR;>bU0I%w~%xm
z?3|an#TAO5%V^zX@{VY_;$*^_kR{O(UI2JD4e<|v%X?sY6jQ>5V<f$&wQ0lr%yz9@
zx+`ljYLYLfr830Q@1EI?J#zCLtro;v?l$nF*oTl`6uv@T5>){TSU#$X|KK+a_CM=I
zeN&Vx*N~6><F9*Ywqr7dYi}BAG~YeKlr0Q944Pb3E3lX28UN^@IvjZs=vS3J7h+?c
z)w}R!i+g=HQr}2@Tk<IK%AcVw#2pRzm{Ord6iX7Fq0whO&$s@I^Jk{L-adSxycM;4
z5Ddm#)9sBZHkx>q_v{#K=@s8|H4RTR#8@;=M1}`hcMgg?RAOB7{_5B@cr4YNACx)=
z{WPvMy(m5rEiHqB;|^B!^t?cgPu`;D{tl?-IIYNDz0j|pYa^9FWfp1Xxowy)I~WhM
z*65B~oY`TElqieu_K}DCt*_Ie^r_!Ivzyaq)3@3{(A^OL*w=+#k%TZ;DXSko+<Aft
zpI24O!+Lb_Hw}85Hc}Ho`WOBR`}9vk@?DPa199hrCU(k?@as-!Yr@wjm&qYRHG-At
zVW1Y|W@rl%r~gd$UE`;m$Ul>)=@iEtEPp2XonU=^e8(IT7rAwh`%Vh0H2Lc!2xoj%
z`;sLkMPY6_w2PbgmyJwN2DYPm(RMAHu}l|;5F`DS()V%8PLK*sCq7|R&<(p%Tenn(
z5&wDjF3&k4NEn<~0EflVu=<dgzeKTN0~#?}-1qv)qxXlI&9WEwUT{82exL;B5`?aI
z;NH)89-1Jj%%LsbM`xPS>@nnY41jTqxn>E9ri%&er#@#lL3HGVm0QF39&@TfnQ$Z`
zx^n9w`}Se6Wolv23=5x3W68*k4Le52t>^~+Cdkxx-e;pv@~^b*Y)!ErNBC4977^LX
zk}B|ismo;cwS%keC=Yg6rTc&e9&$v7%lX(m-g-cpZb7Jj5fNg~m-WOqD`(wLOzfsJ
ze0uc4<Mbs)8cz!IQy~EC$xi8UpN5FMsq{wQQN3N2#aH(X4}Oxg%VzpNBT8G%F8yac
z_tcd5S^nrajGpx|CSb-LiOgSm+1%V*jcUQEB`WsZvI31c@E#%7V(Uz)sAeJEHq?L=
zP(O;lH{@NVqI!6<>+RPf9lIb{Agy4T+E^la@WF4ozq1B9eYs6Z85Nt?K3an2_I!eK
z?de#@oNUETy8AiVW#a{xlX5a`Ww~Bfg|`!Po<1(VOR4*v`#j>Cmx@HjgR%bY==n-v
zs&MMtUsZn|fCGl!7PRntwv{JD>Io&*AHS@;_?-Y(jDCc6E?$^)fV<mp=1H>c@1s3-
z;M8`l=k0L2w2xtKxt;ZePM+{gf1-Ds-?+u~hI<eh&)Lpik`)`SNye&&^ryJB*^Tb7
z2ljlK`K&q@WHXbm2b`8be-mD!@6t>EQw2uASA$VQN=VDbm%XGC40`wU6RCm0l>Rvv
z_hS(wrLHpK@gLC?f)@I$_3gTLUXn>OGA$>v4JZ9>gSiK9ep}k(xiXy9k<FVTlWZAl
z*t%cns46U}KOC6bRxqcZ`X^3Kjwk!XRJJrA+1gk4>f`z?P_M7nubBF*32;!(MNeaD
zf^YO3yGTo<xtufqkdy}{q>Z!8k|0AwRSjd-^aL&1`0-Qe0%v`ncp&3VtUpt%`E-s=
zZqd{^3(ar~g+B_mnBQP}Gh(qIejhzInN{#HSk0M2N9DAg=s_CMihQAKN>B@CRze$D
z(uY5I&}4)MY4EUYSMjDFr_qrob*VQzz3Mr>TTSzbp;#i3ZcL2$el<6r?^1|q&eA7l
zY@MgeI<=hTPl{~FAk4dogUoX+XwUAu#aHvmBb`@Kk#XVCCLYOjhVMmmyl}EvdHoeh
zo-RXYY80gFrGW(Z-$*cU@>`ETP0TWA<;;lspgvWb{i}B4#1=5I%?D%%2V6U@%fOg&
zv_Buu(S)rOPS;!Q4iYLz6_J7`{^?Q-tuebzoCE9sWA81a;##)A&k!sSEJ$z*A-IJ=
zBaH_O3GPk^?gV#&JHfSa*WlK8aCdK98fmO?8FKG?_q{jkKl6D$%%@tM)tr5*_Ac3b
z|Eel~6p^(??N1(HHqK(rc2KH(<pj^5<S_*I$Tv?5Sj}`UTeGsN`eQ)BtcUT(NB_jC
zj9?8wv5TlOphBYVydiLTe@($4PJ@(7wM*SO9X45dr(JGzJzjoAwwGX}2M5L_wUNnM
z5r0>AE3%11Y0>jg{7e^*dhW5d<6#V$KZ%MIUEDK{JhwxY%b4Dvn2xO^^RS9=KC^O8
zz9fECvte+{z^PY?X<~&a=v<fnc>tX<3fC?E(?Y=FuZ=`0LGH`$Fb03x46Ky2RRqdk
zbJJ}k<o7DBK+}pYVIs60*GM&#fz;OUF98~DSgAad^w>#`#6&^<f^%zfwJi-`v{6*+
z$?+ZKgWxY<!SBvQo9<Od?~HT@X_~Sn@8UBkxWS(#+niRt8x6I8EeJ*n8<#Iw`2=Sr
z@b{EBD+f(*Gz<1DX%x&=iNhS29-3Ny8Jjh3HtNnST|StZ+!i%wd@3%s4Q+ey64z1L
zCVMb*m9H;3{WU`KWFetyU<)Q30Fy#zyye2Qgh{<Lv*H)PBFLt7dP2mcyrG{f_7Gi?
zVoaca!gGIGj*MEhRY;4cV|>Schu;<JiEND#=Hne|rcfphS5wuQnQ&7MwO5KPo|ap%
zpSZ4QG}fxJF?Xu)pRCYTMwmdQT9V1ycmgAd3Yb+{pKw*#Cx!F6t}<Zx`T?7&!U4fx
z*CBSjH8yg$of!J&n8;Zo#ess4{%U+bSf56eSM=U@jDR+0x4C#7Nka{1?V~3xFO3*o
z&dUxaSRK6GudrHSi@00)9U&LGmwW01VyjA(P?()wk>1Ajo5Mxt^%q@dB{!VjJ58N8
zNniSHOi+?#m`!s%nDOX+0DhUI@;jbQ7dX`;iL~+7G|fGte<+&gS}!Fil+(+Zy&pX3
zL#!aoDc$P9xccpVG81;J1!Dh;U2Wn17tn~zZ_1{hqu)LM^yW(}|3t*sn-AFzu_9D7
zPW2)A&3KpvZ3@i~7Kh(wRxD=A@A;zgNoUq9E=nk#wyT>@7ps&ML(gxTzZbwI68lRK
zAN4tl&vnxj2SN+Ua#{%#oyYB3PJf?}vP$lN@Oi3J&uAjE=WYX0XXKaH;4A455&q!l
zSI!}>D@=r=oKV4<yGpuYP#ITbB2Gr*j#&UpM;N{ILBZFBFqRWj!OGuKHbgnf>o?Uk
zHM!`NT{Lzz3dAT2wP)>EV22aM`?)&5pSgGmeQhh5*b6c%!V#xWq)a?e(lge{b?E+d
z*1~UzG~#&YFk<C|o4|Q=>zA57n;fY1QsA1rpdX~h%7VT#7`4`s_!aMB>4gYveLx0s
zPjq(r%~59K`;7t*&!Ce&a!;?-$DyrzU_h+1o+<cuci2zlycs4B;Fy{|DEt9UwIgW+
zi)SPSEyxjg(~Sb&za~*rTWz60m;Nje&@<?r$sG)&m{#4GkcOud&zffNIrA3Tyzi2U
zchJr~GYAtI!9=@djVb4H&+&G()7Fp%`46H22peA^)etTDqv2AGmDT%wwWDz%x=Il8
znN^)#Jmf<R?L~Ludo$wYDpV1>-I<IOU*V@88auq-BU63cftLS@?xU-qB)9uf52yZP
zQ#q^p%()X`T<U#?qUH{9t5GstDt~II`(o1VdxdxM#`f+P1|pE#>FJWeRu!j-o!fJA
z8XZ5eFDMCa!k)?SnMgQ_<2G}MG2A$yMVL0=YH5brWNkHYm@AWgo?EXD_van$p&ln$
zOEwls521E~dOdYLLtSGV`N+k%K4Qh?EcaGIe`TrJ_L%BBwj?!^nR#0_s^+uxI7QV;
z48>)MX&rUZKPDqyn}HYP+~A&|PoB%gPVR~#OB&#EUO!QlY)f?Cddc9@I(IR1J@DG3
z|5Si90*ns{aC#{NuM@IsE5SEx1a1UGt^c4zw_-COmN@ejP*qN&*LXFXA)aTE^H$}G
zh$19sIN?#)MbguKX|DvFzVaZZo&R%(#FcDu@ZgYbGnVi#dx>QMQg&wa!tR}IV2W|W
zhhsfWd`W^EyW9-|FD@c7C;eF?&QELluj)R7SLCZT0F;1wVzh*W-R`>KJ%E@Zk1j<@
z&H2XKUa)orqeRAc(>-%!9p`*dW4rnSHbsO^ua?1W0Er2kYz&yYLazF==cBeM*!9HY
zN(tf3Hxa!(0yCG$K3RK#<a<0Jqq76zLC5XNM<2Z044^S@MNk4$Hb``%5_<||V}UVD
zF=jCzwcOc(WK-a7e2p$2OD5TFw6;(nrQdx-71U6b4`m8rZv@r72<|wnDZ9Z(Hc7SD
z#i>FN$d~9ul@sINCSh-pO<_{RpP3a=Cy?e($A>6a9c10xZ1~e-f9XNS@8Pzg0x%z-
zUaYPM+_cs@HZkcrvGxro<AwGDNG}aU5SV8#x>zdp`V;_^GSTnBPnV<@+r~4@iY5q;
zJtNt-YO;GyMtAm+$%Rq5*PSt1F+*ll=w;6&Cu|w)rY{kb!5FfJCHQ_?fog>8QHl{p
z%Za_ndAJ-fs8w7L^*w@3O6XzS4(btW_Bm+XfXGJ7+8fKE*wA5-$zQF7Cohh`3mPj#
z@WwEE<J+<yW9LGY>uFYm^RGzj!X^>6n&aBW1EX{!<4gTvup4gorEEB6$J>Cfx|Z%V
zh$yEzyPHTu=aZl?XKRLbL$=(ctCTB2D|a-R_bJ#1ztH0e?Cv!*H5CJ*)_(FVJeSH`
zd;U%|k^&ox5z#<y{Usqy+<r}WzbG~&H^UuN`zr)^_SW=Mpu6c6O1!E-KPx|)cSehW
zW8iyW+C%9|E0p4RQ>#SFDz&ycFQh{(OL!*YI(TwcyK#X%TM1!4ez*H#)JW*%`jR4#
zmg~?%Ff@2y<@fG)sb>{<r!^dAA3crkhoX3m>W^5AKDOTvY3j3m+Yh{wn418)wZxRK
zw73Y~h0yn?k$u1Wd|DglCeF7I($`r|SK)oiUJshs!|&yPHrTzaxe)AD6$sE(Bn`JH
zk*LY0luqC)Db&aKNLO#)2p#ly{mo0~a4R_B7gX_mH)@UOm6|G_$aI`QT$463LZlDT
z5)X;Jjie8nDSnj!>a*w{hOAk+h{)U&1LU)j$z#u~E@CZL<3{QFaX*B@UhvH}n9xBy
z2IE;}V~JTW4~E|x&KGo`d3WULfPMs|`|cz6*&(vkFQ`6#u4uGm9-{Bv%TBh-MYek%
z!2`79Jt|_o*A3t5>(4M=@UVMiYPExZT8!bFdzOrH?9yaesY{Gm$~rluc}5Hm@+6gW
z-k(<KBF|eYZmBAL*<@sW+Q)PS@V@pe2hf5!Q32W<x-}8CUnG!*8+vNNHI9KA@?<Nl
zMP-DHOhkJ{)(nuhr@^@Ih=b=ow84a|;wTktEQG8a`~~{ZOEdT*wlH-m0Q`_Iv^JxN
z0A5&nURSh4Qkqj@b;oDBqCZ2DM1w)2)yjL>`Q#23k4hQ(yq4E<x}ElRuBey1pXY=K
z7@5{@I1>}1^8m)gOUKftQ9r{37qHy#+mcO==3_(pP-n}$Iw(phTej}p9+v3D7Je2G
zCebzA0KJ%-e2qQvzOJR-y%>+xp6|DPclX6rOh$#n>Vq~}H4&yO@x5oF8p7RMrHr*m
za@OeX7s3v21hg;QdAE+}(<*?}E;y98xBG9-d8vG{+sW?-e~<zit}*hA4OWvg6r+sv
z648#tjY3q<OGNi0T!w;#6uQ3o77ZpSR?x>ey{V37vU%ybDCCMl-A3c5hE66KUB18L
z$QBNZoL0&{Ztd$?F}3_`g2dF76NFqxaMzW00s1b<E;FIRad3sa_}o5amMIe9Y2IuB
zr#8KZ0&DQOA|i=$6<5+&|BOfuMSEWd`jI}MH3eAI?I(0uktA-_9+b?Mz?w#0`CKg@
zhC>c*6H-;xK!+Zw0d{<R1}M!c6%`|ghym$Ro+z?>0Y;dp`)BU2e4gKo%?PZ7gt~lP
zYDHPDWXs{aLM>oX^!Q*q-7f8P)tY6OxH6q|!m|HyYlq(7c|)|J2{TMvldw|+TQ=po
zhLlRsweZ!qB`)GYfWr`rET~~K8(gcvRx~iYxeoaLdO*3*Hbj@7xy!0pXKz9{DLV^y
zFad3mXh{)UjrkhzrkZ)S4l&q*!ntq<T6{*<6ch=5nIJlGi;~$)JC;l!Cck6Ab5aI^
zE)5d?Dt0^#uz@!_h!LqqyDpYLZgjaE9BlFG`ng${&Fyju>ip^yYS{NTvea$lYzlF~
zJEg=PlumQAcg+`}^jx&j&Z?SRLy`iJNSR-f`{E6yz%}cTZfWXmvRCt(pI2O6-1~5=
zU|YqvV3C2ouMj&_`{D*BqtUq<t)6q4?=TrhTpI9EE;y8bsY5udtcX~gY2iCN`xKcm
z#GNlg%M>b7xleRVc8q_@AIlNk^(I;p6pjYLNzFI_AYbJo%a&LK<w?j@G))GP<|l6$
zngaBrIpjpd#H$cmD99|Aowf@GIiOpp9yLpyMQT!{5zko0oDfemNVze=bhgsUG&Nfu
zUQQqzTlOjzJf?iG-EYD{+nL<v^Y0O;<gFGqR=$&zl=ksjcdb8(2||nP0oGZE$awU{
z-u4W(*zY?|X@3&PdWz`G-&7l}{)`Lodc7ZW^XYTqRYN*+S-`|m%`8r1ung$9S!7W%
z(b9gUk&f0~-1*2&3v^ohQ!2`y##&D(<qir(vi(d8`g*jBJZX=|7~E`S^emBpXd9H$
zl$I9I8iHbPtX+C0phY?8>3sfnjE-uwPBpsTRlox2LXN7sTIaCo=p?P2)8T;Z-jc>Y
z&uy06&2_7hVkLU!KHB@G%j@-tKDe{orKd%WTZ0JYAP{apxctQ36koEGq=7kj>^5eZ
z4uIiRFC2X??ik96Om0)s$bO#!>tUICHb@wRUy1onjr#VJHnz*hF$*&`F(FSAfyc8*
zN58{!j$P77QAnu0ppxD`RO-qzFz<k(0tNGholG3K;rDW0F+fs6U+?SQB&VeCgGd{;
zRr)1iIqOT0-Mi2UraPlA;>|+l9dgF*u)&9r%R|$-juR}XVZ%gT`L@^WT;QF=ArO%d
z;{xG|Nj8zi>rR<{Hz7DZP?nPafurZM$rsH7vKD`<X}XPP&E&n9Iz*>*h(ce(93z9c
zvrsQ6#w7dtLLkSZhgw~->2-0Ey6L)q`uKS<tn<@L(Xb0*nl)S}M-*RIkw;ChVNdnB
zhRd3%$mWyY-yJaQYQwF`p5Is8v}l(qp<V6CCpAR6eP?e&oix{j49M7c7l#<ql$8ZG
znI#)0QLc>0<2C_{QQ4{9oF+c#hTOsNF_3JaRHP*)2{-SsOO$)cHHL7s5an^wn$_se
z()mX8I!e4bMMiyMG!nYU;V2<y2hU8lk%g4lTxpuw+wJqfl`fyd-=6y|A|$z#B(GbO
zDe&@Q2#|jukKeh7k4HjgY9Bg4@xPTAS<8YtN_>H+>z$`CELdim)VQvTM_b3R*7Ke&
z|K(UqZV~!Iu*jQjpLfi99yVIW785@o=Jg)U7GxF#125&g`0hU>5nUOHJ9uWD_{kw9
zn~lG36E%x6ub@{`-N9DnIQzZBzOr_kE*@}Z=wuI?INm4Tcpc<sFlIo`opwy_mf=bY
zlbm%`C%L<iFVTZPLyQ_HP}z|jFbmf@v-h7RIOE&89s|SWU(|`C1%x&6eJGd`x7<5%
zQY#cRAz8A#9&4_)2^GXaWcs>8PI!CHtCsIxs|;$!hrgzyNhsGO8W7Wd+7(oB>?-?)
zgu#@Jz{PCAz5kX@ouo1MG)+D{1Ts`Yvk^9Bv;kci!ppGQC{Ub6ge)MReR{vW>)GY~
zEIwj8h!lcB{Y$H&c!-yHZhd}50YVa!*&-P63de8JO#3<{3`;aMVP#qGtVWl^;*;=I
zh6dMkGBw;--r(ZQ<a)T-fL#_PyZ?K_F0J$CdQ{*gePv*Haxoz@C?l0a-@xwr-GdwC
zs^}Io(b)e%M02j=)|inGY~&~F_38vQief#NyonkxbbeUmDkWD2<b1l&tEB!mf*Ym5
zh3saj_*bjTu3fyZMlsx3d#|NpDC3O=hbymA)jitU&vU>XoE?x}MY7c1zT<TMs=<If
zJDaHKGN$bX8c3gEq?9S`{Q`8T18vXOV-(bZ`j%SSZ)f6W%k|+0&y@6=3SiA}9i>*i
zF24Y*pFqX>O2UJ@?v71Tp}rguo9?@W<@P%KMxEHC>k79gK3AUIPkze&$4FN*99M>)
zlGCib+_hC2(6K#MulmJ)&>vJRxk##$Q==v6dKX+zez2x<Xz+LnRqPMN!IC<^S>0&#
zd&@J*3x9duv_-93Pc*|2i<h}NT=sQ^#ub#>=21q+|BCR_%L~jb@GLv!(Bv>%l%DXm
z$kK@|mBffa-r}q-J8OckDPSY;#8PL4#y?Sr`6w<J40MEEY+G^o1ZEGjTi9^AKzK|7
zIJ-XcB;IJ6x-%4eqvOA6EVYf`3YRRj9-lP1ts>^t=ymaZ=cIZxgvpMcmtL<%8TCvr
z5)n2aFuJtA{RKh}3WUvegfttp`K3P2|76(ZP<4L>+`8F|x`2GNsS4botcD`ZwGys_
z$~lZo2Z4lhCg1y#idK22<7^<h&s+I|&R=OQG5R>XEI91E{WZA*&2#>827T7mnQ{0m
zo@P}6Fk3ac6W3VcXyKpsz39iig&Jp+m1MaH#7%A(^z2@i%vd-$h6@YwZ1~a|QJ2f{
zl}i*6i_X)j)E7C&fRQ2;Al>bN^|T!N*iPRDa>M{>o%NFnw%-Z=8afA<NpQJP2e^wg
zH*|b~Ee=zZMN4E5FGUCY!vo)U{_Oocf-@6&c_UNX=#}9~$NLRh6|hS%L+giTwxdq?
zU2ZT|PV*6|MKcH$RRy$mn%reG_mJ4mh;;m*-R7byYtwhO{>CI@@{QxXVq`{W(X-d}
zdq0;FNbClsHSm+<4t?(F2^()1Yp2a#&Oo=Z+E}4sLD*F8SE2Z!{KV%!YX^<DCj8yl
zZ)b%0PS<NB^>N-j!+0wsLPg?>?WR16N>H5rh7u#g{q*v<>Z_^4xzZ66e(R2(b`c{^
zT<%tZ%WwSzPw_8&CZ!8mLdunr!3*{0FAK<8wixBH-?L0FHg*qf4X`A!Ze@g;{OEt{
z?RTh*yhzu%TDMg=^{j+Ks;i@v2oy1}p?u#{xnF&AWTKASMx#N%aagt^u&VsJC*oyQ
z?N?KDVl51m-*t&!Pr26>ujSNMI97M{qT1=m+o1AYtoOMjAlR*YQ?tsf`#f2?SWwX`
zLM*ugLTXr+EOHA-iaU|y)d1xr%g33Vmagae?g9+R^v`iJ`F{xp8T^Vog-6CG@RY}Q
zQ`EW*^-c+Kx0UEK<hz|+LR4tV)&|Z`zfF<ulfGQ%`jL%o1|z)3OGsVV6Sg!_wn5f&
zpPH<GEJ<m-0H7zt7AL;ejpfVxaXYPeVrd>q+gbGE$=8vMPl>ZnLwRvt_;i`%v|s&l
zp(^(d{w+&|cVYSC)aqpBf#OFIVH_MJB!s^TSxlrRT}TDD(47v~{_SGg^Wl}#Cb!KS
zxpb0%Sj895^CU9&#Z_E8rp<+T;yUdW9S3(tSRJuqKveSFNZqTd+@`b!BAGQWYJew3
zj@Ppp_ILr{H&an2k&YQ#68{nNBrVgMCR58h7sdO+I8kcu>`Scxmkg#eq%VTXmXBsE
z=lx<`p%~QAe4S9Jg7rC_GYMkH>vzRvx4K+0EuZB~l5!6aQ4ly9Fq4XLMdh547Azj?
zZ@MD}@eeYxj^<l*RrI{C_pn0Z^!`hxFG7Hi;6pgconE=KDX8vStY))q3eh5pGA8=3
zt&f@a*WQX5R~gRy8bd3m`8OT$LDNZL6^<>7`i}<ioQi>~g0w4{g*-`N4C}mJWkr0X
zfgqo1$xoDrCd#dfQUgDHsHk4Nc>Y(B*7{=AOFKb8RZdShck_oZmT%6g{xO3CJG+;y
z`VYRV=6LfsmwMv9K11(6ue%2a`V0QI7XGD-r(zXfQrP)kJED<lXMFoLf$Amja!^7A
zzGra6^b}V^L4W?$mXw~0)&c!*!T*x~Pf@`Be15}y+EA!a<rZ?J*!6hD$aO|qU!*{Y
z#TOvMEY2VQXF@B|ZdT;}Kg<8iG<*gBe5EH=<2U+$*YF<+|EabuBMlWwy;7r7p5=cY
z{?jD?{DbEo7yO3IzF+>oRP(p-{a3Y>C7vV}*dKnH|Ifq!GSh!Oydirs0XXkwOxAze
z{vSdA{)qOX@IM`4YazahVHzHqav8rkMGZfBMnc|3QFP?Jt-y5>b8ML3{9dl8Vo19j
zC8C8M=W%xDK7OHfO_s~vbTzhxp7eQ}66v<rmonga773c`Zav@B;znDRli&v#Og007
z-N=lR<k+2}dWo;B3T^dMnL?nP8KF}ZE+C_NrwwL_jT+=;b_1Fnwc~PHCZbj}Jj$$?
zm{4hnCjC<%k<9ntB%>stepC(n?xwBqOH_rL(&OC$g6<n?Jsh6v7s26z*N#;9|I`Z+
z?T_8(F5qe^(L@UzS3nOQon%Z~98n)Kt*^DOglmk3F`=i6_39hgd>d}@*Pdl0I<KrN
zt0m$R18AcE87&<q=_LiaPa;xzS1>5bQXZoj5Gc<JYYm0UjZM5@GG2!JlvBewovu@a
z>J&gEx4#GDb#2)6o`V`-2M-$dIB|r3ot6rchXLVSDt3GbR7QtNNNy|p3dPSC7KPKu
zq?nmLN?bB1#q)A*&ZDntaN)St4b4fN{!1}=Vt;y4m!cvx@o9J(8gYZGVq+kk<_wn`
z3aLz5>#fO?n%FAAz<E$-6J>0yx9lkN#*+2-VJYx?D|Ac&7JXY%T9kUhvg;f+m};9F
zTZz+fYWb0spaRJ4DpQN=$Xa>18vr9$JhQ$+t<^K*x;o%BU%X%_UPsJx)6Z1pu#anU
zx-+8f$lRX>o5Qp;;=gp|(Dq;B%e=jnaqQwvEMd`4DszQnLkA18oYf{z+mtt2Cs&mZ
zj$KDf%5Pg;LRG5p3)-lJ@$eC=<UTX}sViYQQ)FZUmtJ81RKZjwgIwW=)q7$EvJ_mc
zzipig{YiD=Su_w?X{7Tt%AyW$w`!$2b-z?(-yAN8e8$ZH_JHB;uT8p&gBRy!0xgd+
zj#|HWrkYkRrw<}3N+hix;?fIo!7K@qG9P<<zPxi^m0X?IbRngzMBeegY{}ymYSaKD
zhgHydU!yqgHLiyYuKm27pdLF#B4TTOh<n2LK2*R&jRN|1Cr;!b1yeaQqYWQDttM@L
zug+Yq%0&L9nKY-r;;bc4-fA5I%0>o@2m8utH$dN(8l^sVSlUw)duWy0^Uwm?%gM%O
zaT2yQUWL0JnLP3Dvj70kaCiLHA}S$}VQ+4iXYwp>(VrvYAvm9Ilxisty6;aFBGi1W
zkk_Do@1}pssBnLN5LL2g;hkgKc<)}hr0ezXek<IFgX5>w7%iYOai1w!+uEY$cy3zp
zu`<rOn~q6kt3Z+mwqQWKG^#gL7-_Ap3-HKoIv8b|?ncYNTxKf-yVfr#%oavEnovKm
zbA8>+gDw<HTf6B>9ALF@kvJWx-d<87YqC>31Qvx!M5tywuGgAR^%*p(p*^svD#)7e
z?(rcFB4V3F2;>HqaLk6=abL~C$P}9|DO8}^b=A1~?si*9T+q(VtzQI<`|$hXYxRP9
zK!2kT2ebQuFOby9cq+@eDQzp7M$j|9X|1VE%3)Pl1TN9tfjc~OL~ec=hN?nn3D057
zJYpp;d#4AG*5L#8oT039w^eEx=~}1}Id%sm7MGL=R>%S`1st!{R)xb_)fe1xF|A3l
zGvon!SCr<nk@#>`dG|d9mdfx8wRS%b%KXqZA7>-$eoLPUTa7Y486AP~5wT^PBel`P
zLN%7Q=equo$~D3w4sVsb;!ib5v(g_P^=7G8uQ#HxYYNKK&!%f&hmNg7deu5#?$6nS
zoR^L(>laPR8#Q&hQ$V*Y-jioD)GBS&B3zVw`SR=G4d<NDg%A$=fzP?^J6O@=&3dN3
zbuqC;8N>o25!bQRDua=lD%FTaCW%d}a_@?@+U!zOY^8Noc@E#^Gz|e2@;$J=eM$;P
z(k((wUM+2;6GQ7>d;c&u7)^IHE-ti7h(CZIeO9NZP}P~B-`D(1XnTvS6#}mB-;2#u
z8h}f0WXWZpD!5e|32&^tp6-H(&9jjTanL0`n@k`Sj%=HGJ~$Xpf)%vm;p7pY!p=B*
zsqp@Y3v#5plo9*AEA#O4Rf)x+%t*6$9#jwI8n%t;+4XUk^jM&#YX<P^9g`1Cq9_E<
zSIrQJ-wXlfzGpL0KC|fcv9b5u4E~v+5U-|_8~NZ6bQ5Lps56`@l=n8MS>IL5z|6bX
zeZe4AXZEuy<i1i8lSDI*qAa~Y0--EiE0c(3dioU~LVl22_fdveBc#(FTPfr2tYm7Z
zq&xyf$!<TO$M6k#mIgg$Jbb&=oQ+uq!cvzA9@CYYS?e{(O#IRSHFxByBSl}9I~S4U
z79%zDx-T5S%@N^$WbW%9tLxk}47jPi*{;>Is|zOLbBc5`{{4ZThBl7L^7paA7)<9g
zZN0V#V5M1W8YZ$#xInt;eXTNE%xKwVoqCaNl;wG36(GRn2?2WI)U%n*Ve%bOPkm>c
z5>M1F?#Z;am_%;o&n#;zyg0Jy-@=^NhiLMaEAGvDmwVao(wMJg+Y~F<HN{JUmeBrQ
zkHq02uskj1pZnrf+~KfT`3lWliq=2oYpAJ#$gcv|Dzxvzgp0xCYik@MakQxp+&C|F
z(|B@i$UOqND6_2wendAUJmCkxt$bvePb}2gdgLB=u(T`v-SM({UH8??sDOz(+wZKS
zvlI<0d42=FYDspr1<6-dsujd%KlK;<BvBzEUMAsMrBMO#h^NdAE2rmm;2H%rOp>p|
zq=W9G3A3UyGcyeeQL28T8(;7G++KX)-3od<cfkQJDN4vvrk&C6V0uw%+91~@R@Pzp
znJyn3OTN$(As=#?Xy3fY0=M4CWWk+ga(_3$m!s=qDOk(0yz(WEiFzZB+VC}_G<51M
zCFl<~j4Q^;VT7XJe(=aUGv!CU7)~`U*pEI-+$;B}qSL(=F$KoWs?MX^VLv62NMJ)J
zQJd>&bCe4xYr@!iS)uU+t=(TKJFkBppXi7@Ot;)==@lEiC%SGdW6-_;wv1+Ay>r_%
zfz+=De`s%wYWmeR3NvuMH4SVy_t=T;5S(WCe4+jZbh7XZ`^OIoHx0^9=f8EtZWo9^
z8EqE<KTI(ql;3D5#D$OG&DI~%21~&f$N7k7_Kq7t)&w(>=XruI>jZM%FE<*+U#G9H
zZUCCC7?>%%VY>v>_7B}_%WL<TeaXEfkPq!xi7H&E3l5WkFio@X3#Anq;YX=F_qiq{
z{(U)rug}`0g@x1ofwY4N{=zVu3#b-T@9x0BCxIgKa21Aj*@vNGE%*=K$tBp4K8xPW
z3cw3WT9ftWm;SXmINL3s6m^nV&{^vlxl(^>olAh1?X+K$+YG+m5>vtfMt3cEHPn8W
zs{J*`BkRu&%)sKsj#YLf+l@~+jCTByJ1DwV?7}6bWYF8lo;-0a-Kai+`GTh3oM|n?
z=j58g)M;wq_5ya@Uz?CpG(shtB>hEB$XPu0^`>BHbMl5K47T)|`K$vIauGoM-k9+?
zDMPOkSP;>{VgM#s(&f9{Z<N)#JlmWjF40EBJ0;jL;mJ~o3arCG#dEns?AeA=n_PDe
zmj-e-dt}L>U<mH8dNO|#-k?o5XD!JXsjufyiS;-{U!)pfjWak0knHN3aHQy4J5J+y
z)IdzLz+X=pXUF2>I<IX1iqcxI(9`UwjGh?;&P64Al~}Y>ILxp#aYY%xNSo6yWr=&0
z_liH+#Jm|;ok2sf>+bqsDW~-_u|OZF!{M1GLpb4+q#6kxH2J`ys@lDQhw+MT$2dE!
z1UM<l55RTezSAo>i1#5}-`=pIrxTC^sg&)!$7ucbnZEXZh;655q%cqrD!9uJcJhJG
z&utMe%Rd<2q77d5Xngl{eZx`>APKbEH36kPC?v{$hSl*--?VOFLvLTR*<K@UBq+=N
z_*vRyUMj|42N>a32Qkjf922($J8&{JYsI!}g#{^J?bh!hzgw}s`(gcwhN*!<#N)@+
zr3yq-e%v*81T6T_{(h)bH)gUdKOHs`VtC#!*9S-SJYJ>o>Y5iFqf4v1nB8gkV5o7l
zya#YgH=WNe3EaGSrj(d>5!Ld^d&5}BEdB(1*z9?osrz-(jACUUFaXow;->?UYiz&_
zbsp@X?G>qZUCCx2(lFkC-8;R_1hY-olWehltlv7IzWm>{>*)une)%I2USaW$)ZzUZ
zivqpXmej*Dm%W+6AQzA=PxN4wFZaml2>AhKC^yB!SdbCS28fI*yLjd~Sm!4a?b<d?
zs*BXxevMjAPnaF|mb1Z?7a(98>(@`bk$_6T!!9d74_iJ|!vW)?w{Z%*8bPF;fRFfg
zjKmD2#f0`?UxHJ0PT)PLUzsw5cJsplyVnwsaYTO%)GJI>m^PZHDII5D*5wqrS`$Qu
zFS>iksx9_lEg^p;)S4}vj*!^zwSWF4<ctD^y`j{5k&_C?l5NSeRtsR__iJRQ-*}3g
zOju<6liQX@Mt(#NLf{qaYCKX+Mf$bm(d7gQLZ*P_LuBejL*s+fz?lvjMU|<xJ5D&`
zs1YBH?b)VdTt@`%uIC1w^#eb%{dwPZnh7lW>}ZA@zaRXXj%n!|ddnI-;wkI-nQ-4Q
zCFx{^b)U*`rY8h-EOXxF?qg&z1m;Y~V2;JUgxU4G$<wM;@xXP{iBPR1Yo>43jB|)*
z+RSBP+&>}~ji|M+`a8EvJ2IH484fL8lYQ8Fx(;6O=h_is=jm~n5h$@<u191}H^-*l
zl|VD*^L^-<EQv-3{KY0A<-O67mJ;E6oUErvFK~2ACJt~YC|7Jfi$k)Y``OTja?dZl
z`Jq4Wc5(5JPf4qb?=n?K_7x7uvYP<8!%1vysD!d!!{T1I0O&p0KBBY^l&snY9;aE4
z;!RvnV5i&bQ%)Vsgh_Iqk_2q+#gvDb<*ylm0+(L?YjUJ`@&@85CY5-^E4t`tecfL+
zAOrBNkG4K|b;tsU+SdHQ`>I#*L>RM<Eh;AOSz1N#e+!LovgCcr{$?xAEA286RNmSl
z-mUl<JqiWqJ#(^4y6FIeDi6^B%VLgtUgK$1cE>0g{aIDgHKCRbKdfO+#fy+Qlp{Wp
zlvuTj*x<;h8=?fnq@^AWw#RzAd^eZbP>~pyUk1o+oVz&fTyL|3j&Kv14X_2&5!<;f
z3WybPu5x_y^)sn3`6g(1hbTf*gICz_$OECX+0g$m;GPtKouHmnNV|m`E!TreXX6|l
zFB3c%%|`6ka$E{@wJIsm2Ub}6(snLzF3nxFscXE-rClb~g&2TFy6fkhmzx)1p6i=E
z3p{~#?+`3cmsj@$27!pxI;v)16OD2K4b)C%J?(cCX;))SXSAa@2WJw-DD{=J1>Z7-
zThW)FQ<E?Gjw#$rQL%%ciP6Ht4D6p>gX(Ccdb;9VPMLPnH>;jIHR%tUPY&$HuIQe+
zp`LC?T3;rF-^jbyi<w*)Iao-ptRCD3V}m?89F9k)Y|`i@+!@5@7qzq;P!hf_c-1KL
zYZjcE%5|xwV?M4!Kg^a@vN>)WSXSYhZMXO}oESAfor}Fh)>2}Rw;;GtPKppdWogfc
z$yWV#Q}u8EFB9X*wJO8N1T{&!eL`nMlNJD8>31pgt&Bn~6CM%1!s?np=3Vb9&@$|h
zXe;x6=~e`7@inv@2krqHKJib)ee7Qykc~Mu_B`!)N>+vkrScM$HuKke4A;fY4An=Q
z=&Gio>L_c3mH=#ykBGX*d{#@o7kbOn^LieEG%cT*b6(+K)wHZD0~+$m0G+*!Px*aQ
z4fC<5YZ>^yz@869VRMt^U{3vNE9=64xYap&->@t4v*K+L1&<!B^+vn%CJR38;sWRd
zI9`f4q#UQHFDIccV4#uug0x%V)5^3A=U!$^f<-b4PHVWijF)svViY7(xh8xEjjm+}
zPOeEeB!(G*;Y-;?bQ=?-XFQmT=(~s9MRdcmAQ#m-5$}G05}wQqX$O3Ih5@oDws*hH
zq8`%<J-lq_xM<QpF$ZVWVPW91i%iFVPu_!?&K}wX5z8)pMsE#)N_OIIIlkTk*WE{k
zk<p;)36dumQ^R=4LyVBWb#LUwkN<|31;kBI_6}2m<5*ToAI1#iC}0`Pp6rWup8-I<
zmn*F@@D>fg4?&)Ou3+(fWQs#+Bbw}XN37Aap?1_3QZtzV_WOV{th)KJn(*A~T!gQY
zGCq`L^QF#*{4Wmq={&1!#I%Gc-i3rCZ7KXq*K&W{*Nk?L;&1B=3=L`_R`Uv#xj1w}
zOhM?PF+!;^KL%xoJ^d}sXcdwgt76}_Xyl8c5Pl1`MZw(U>PR>$!<d&F7~((VWGk_v
zBGFBFvmfEAI|uE*7~y)$-X%E|(`CCbnH@(3B*kxbco(c{8r`|il4v-77;y3qQ2I(o
zL`<7i@BQ2yboPSPO<tMa?V-bVg8PUy9+}t3g6SjXCHKp-rQAxha#p7V5%x<m=8Pax
z)~$<#c>At}4$Q`_+<a<QwbnOA;v~_cow#~$q5@{?XT3F0=Sw4k+)Prg#mE+Lsg~3O
zf3|GhBCy#c*u+_`okqITZ1h}y^s9~N@Oc2yK}$j3jqT7qw`*uq{eOtP<KD!WwF|5n
z(RrjR0A!9Ba9q6p#Q9sMhaZ2T`P7Bm^;Kklzcrig!;I4^0mt@;g`8&@cHQCP`Xc4{
zz~xVhK8nMBjy~v@YQgQ}>@SjA`r{wXlgQRu@S;(H)Mpg$O0Jwp>g-3gTLe3q#y`u0
zkpB)V-aP*kB}5)5_)Gzc!7X^}E`G5-7l*jWuY^#Q{>jmpz%nz6^O|cBUOs%;6xedy
zks$d>%_+wJ^6U0dCxIE~TX$m=j0M6$9oYeHK@*No_M~bv$5r?mHpoOvmK4_h@tEPZ
z3szZS9*MT{+TU2M&pY=q!@ndbEuovS)ew?WYL8*?d+*?40+i|MoUGhuGcc?-r-K|M
zT$fDiOw?@g+#@BL+}{rvK8URe|Jc5JaM9Rs*pXWB&~USwiz{l?Z$%yDoaTIR(N%WJ
ziAjtqi?o8pOm5Jyo>P~DH{ZB9Xr7r^0mIu6k=S_&w+M<l#3B-yP!o=SdFT)m_4Qe3
z8R%wNp^}h~!WQ%8oNmP;1xWf|UK&^7bn7ZK%T0p%+YjP;)kTM6E?kEfFtmV{)R+A(
zbV>HwEw-P5f}3zWN1Lr#Drq!`@WZwASht7umw6^hJ*eNd!KdKGw>(nDd&RYGT{C>(
zhohyvx$&0WaS)jnHb6Vz)s~n9h5ty^dHaO?qkptcz9xdRo-w~ByqYVjqr(7jqv~Qy
zp{Wrl;H&d|^IypUDuSnowhZ#@9>tuJ;sm|99E4+n{MppE1qZ^0)|^{@-MtF(gcb+K
z4o+6z69)PfTyEue`7Ds!sIA2QhFXLkW<DP35Cj%k1*b{OMl`itu{WD~vA>=Z0PlW^
z&`e{P2$tn0HdIN(=M}mcqDUzBr2Z=BZ@IaElJAo}tPI0mha(&M&Ak-8xa-pQaFtZ7
zKo3_Ni^2>Q5c?>?KpU#;0*)^SH=xiTGtVDdt{<_)qn#b_b#gysAfz@JE~i+LC)V@A
zu_0G$3?V+!MD)0AmO(2j%_25lt^M^5>F(w87e8}L8x5F4TC!Yz&El=(-+i#{#5>!i
zsMt#Yxu5b~XW}lXJm3p!E)0P8J6lr1VhDb<ccTWE+JufQRMtLzg~Zm|!M<rUU+S@*
zt;pI?jtVb?oWh|K9M;i^ABaxty7t-tBf9H)2`HYXgv>019|Q~3h8jn<sF|I4y%g?W
zkc0zfEq#n^9t7&ON4Zw_>9K8~bUB#x0f~$;&Z6Itw_g)uw@Mm39groHCPGh~LLsNQ
zs3i5~cU(G5A@KD0Vd0%svEOD9*)ZF9O}!D+;JtITbv~lfsS))kS8o^pl+*$7Y}ckW
z8O&d}0DDHT)F#y{wcOY5`xV<uPcxZ(kqSLQh}FLAwP(|ZsDD<|o*{oy0=S0=8HokC
z{tUQvZz%fNBt?<U+`0oHbehh)dDAj*!^JtXgrO_kI_a`Nu&b%z)?36?(Rgx3t++Q|
z3@p`18u&mcL0U^+9%qRMVp*)So(O0Xm4s|4=8KgN_o8aoVT>CcW!U@3(@f{zkNUkq
zl3#X*co*qtHGdiR8FgHJZ)pdI<6?$t=!&nr*0#yBWE(OK9sJ}D=}VU-sY}BT`|L+<
zxj*jCQ3pjOWShDyE32mPSA6Ty+EKapN+R$hGpb%nfre{(>~-l!m-je{3h@($!Lz*U
zCJ)1%lj&t20>I~{UuY_>t0u@-jV_Hh?I*NXeh-9J;=c8K+K71eVc(;F@ffIo-WiT=
zxluEOvClL_B`xt8<9mhcDcc~RX86vlVzKuciW^=tbtx_<yP1GQqvw9xMk@{wOyj@y
zwF*tI4CVL}Kv>)2>=jzy&>Jz?-rKN_9C|_S3C$&?fn)BZcUL{PUpujMljOafnF{#u
z=J<l-C@<A9yzynRTJuWX!#DI;HXy9#hU#(!i$BvJYdm!3C2+xy4%Gyn!a-^0axpOd
zfY@?HE$2~!9QhK?HaFAtPHDp>s9<LqaoBaVyRfdEW8w5BNY|ke@9&HvkJ+DKa;}IF
zJExcZjU>7F+u_fyFq@XiK6x~a#xuD7T8F8GuK6y*Di(7xg2CMns}7gr{6mw+11wnp
zMW`Bkg94Sb3#XDI(c~zV+b4>3AMXdu%Hgz7??|Q+t=EH9Gfx_p+5rg%tK6@x<}YjW
zZVs@F$17>pyWT|y<wmC85Sqtr9RD)%N!so@_^3?5d=A14eW~#rH^sUy=*NCp_hEqS
z2+-z&<o+CybEytiH~Q>;<v?&9IBfoQ01D^)?!Lw&Q6VqqE4&_}cmB{^Pnc}xW9GK{
zu7IpBR?v?$cSNpN+|@847G+$$jy@{C+rP?Z>b3m4Ov&Bx&l0d=NFJ%a*g?6oMmv6e
zK%YhHo!v}I0UWy57<*zYbggMR@?->K{&-%ii0Yd~I~W?1i><Q)I9(iZ+ll)(F>I9)
zXAh`y#9UyjBHf=Rr9I{eY=Mo=spCSOo6o0x&(bq0&AZE7r(xHkTQsehi%OxL1a4cP
zGVI%T(Jhl!o`IrtG;M&RY?gQ4j?MPZ2g$b_HR+d$o}%vxmEk?Wu<?*k2Ik*fw_QDe
zy)gH!f{>g2Da>^WuSvBVDjS)wA*UF+ux~DDUfUEs2C^-cXxfwPu(L`(w8RBfu>u1(
z=2Ocg1@=g1p#OEco;OBpFRKJC2xmR65Bzt$N0{#^C!By)G0E=mT6>z4eRr_^)d{?*
z2?{i+40Xpiv|m|szgkHOIafBziHH(G6A}ORL=iR#A{G>Ov{&UB<N%0%*r;7aY6Z>r
z-0fKs!InIL+CSW$Hg8twaSuvkx2-`M8ZF2)(?yCmWHTI5%6V_T*UVU{Z}w=sNR_6!
zKPRL|wNy+sv$Sn9c@_qrz7`M_Rwya5Ic+_(7{g4a`3dYy9H#<5(U0|g#{phIyO!yW
z4%HdMRbL->AN}MaW~ssKGWRs+!=V9r7w{H+0WFcT5NG1kExxMSs!QdUQ!ehXHRlX_
zbibkBVvI{}eC|E=>AXw%i&|^P;mN!?Y<YV*0^KfRsjW(Bh2R*nfwi~GuS-XPHXo*E
z8$fS22!TaEvB^LpE2f9cH7U<eLM!UOE{0{~t|X+Xk*;m)Te<CC<+SThXbA?b1w}Ug
zHlu5(sreh}bj4l;Xx5p&Rk))$Q|<Dw-BWPyRnV#5mQ5@u?Xm9ivHKI0e0X+DUQ_VI
z@i}jNP03qnPv79SI*E<+u=0(4N3M(LWnm_eqGeARh?OR_AFD!r+Jq3{QduJeHK|K{
zo0?6?#FV`oa@&Jxoij}paW04?6S$}_VyU}V0?1IqbOGZUXIYVL3cph<cTX*%(OVU6
z%cqdT#2Db&SG4T;9v?48u4z)XQWPb6m(Zi>{04mz%ccES!xj1C>+<<>EqPT@%K7KO
z%A4|2HioSgbM=P>Psz;T5Bs&2yd{U<)OBs7Zt+)dU!RL|Z~AZ8X_jx<F&K%Tyyxp3
zrmF^1Auq`v=(p)xmD?Pd&pJlA^Tj}fJ!9Y^#u#EL(iv2`dK?xl6=kDAb*tx-uU&h=
z(^73b=rYnfs#xDAE6EqW4fMF1L+!0F&(%-4#TPzDO&%;}#%?)Er|ND)@q)U}(b_s9
zprvciYPc--GqHuh=2<fC*&%&A9gL8FEh|6$;VV8p-zZ9b;T+YAB?x7+d>uE&TQx0o
z{439K<$w+`uz3Y=hv=dOT;$KC!@Q5{36B5;A+Q8x*+`d@VoSQr6C=J|RK)KrIWDPx
z+bO}un7FiS5V!_1mJ-W?C*%l~>?dpibA~*w&MfC(gLhek&(!o?T`nn5W{w1ocb?EW
zu2?%zV?BA`qoNc~sK<Id<<-QNYtp4}UPzo2)WQr}mUMypwsFzq$^qyg8lD0?3hCna
zkZ{1`^yooAAjfFmNPvP`+$>!;_`hEhZG~O^K3VR19fcFBx68#aib>3=B7}tU(U{P=
z%Hz;?U#6#e=_*K_7QW3StBN0*=IyGd?zwTh-@Xy^7&CA=$Bl)D<UlW~{_uo>`<fgs
zlX{A(--Y%M`q4#*E*1F&zAbj{&6!j`+IyvNpfg6)KKLr$d2zIrQd+2oWoyDpOFDKi
zbfV8Rd+l2-;XvEo$s*0KUcOCK!o2Z<fjsP#K+zGCtjkQgq?MrtHRiJ7%HLspL|I2i
zZ6V#5iA7FV7~SI@U1xwqT_qi}=&Q|NLtR136-<-;mlOi_bbn(D(fCh}0IfLm17?r}
zw&HF&X3Kc{&&M;XT}+sHgb`4+uPKt3w3$!$Mdr}YDj&p@pt@w4PwMD8s|Hmrs;eLK
zspC~{D0*eeblH6IQv1Z7nK@ik?4BnSM;g~C5nnkTn-Vn$6M5lk?#KuYjT7O`N}~?=
zah`0)yirBpsbgqkXq@iq!O&i<F2Q-<6toG@*L%TPwb=gx1(JSgOM9TY08BzJs|Vhv
z2ay__9hq5sDuFY&%&-=lhZWW5YHs#aoNxBYO~s?~XxOZVz)?BN(sf!w>6h{(FHw4D
zh;kHc+*fv=n|-H1TM|pxLhVe*e(S1I<;GM4RuA3;5#@Ju;x4%(uVlHKQ%oo^jbcpQ
z-c2te20F5R-8-`N7M^tBCT{sgy77(Rc=<vl?9Kz#Q2?_=IE6pl=p!xF><88y?Pi3{
zdCW<hzQ-@<WA8nV-eruJSy_0Qc|HgI&Wo~hceR5}DbuYGFfgI}#=3uE@pzul@qUys
zj>|-Kx!5B7%PmL7NZ{lCsj>27AjPts%bHIE+STyp<uA@+`1jY?r1fb}iDl##lUctO
zn(JV-15y~|>ZP8};`++pm}MxNb@_o3{C+pUP}6kz<0r?-vi;8*nCDLppDtzC&w7{2
zPN;c~_cK8W8-cd9Y~>o1Nohu=Pw#R<{A0&-=`C77FZNRDMl}#t!0EukYWzB(UBC1o
zXTMg$O*lvFHk9Pp(#ow>0;0LqKQsn@39-WLL$|cKoI3bb4+xad+UC^}@Jy=!uYKz?
z)bfj^CbE+AVnUpvZYUMG58LlD&??CE$hv=1(D4c3yCZ6U==E_~ErusPQ`6GF;GEo#
z`Z$58B2Nji%+Xi7dwU8{KkYh#9o4s^l<RQ)mMnteH{vy$#BrS8qTJjryj3P=A0#^L
z72y+|76Zu|8WXGyE72r{06M{yf+C^UYy=hhf(5q~_=mI*+3B(89dAjf+>p5!L?;>8
z-SwN&hJ0F>6t}<J>B=^xCdjF5w4UK1`*FbMK!?y7w2aDxnu1$3^Rv*$yjC}u>JGn4
zU$NHH23symY26>ePEbZ#<2Zmp(CC(B{rn7NQMHhJbu7!lV3GAhbqBqvtgeuVE8SZC
zrFW5a$yJ)>>}~6hd+O<^RyW>0=}hupL3+&AL>GuNgjEf5B2PyZX_x=LmxF_vhNII^
zy}qx%mXt8~H94hY!0X@hW{F+c5*(DTxQg0IaVCiNzA3YcX1E;|NuhyJp)~(;jo;kH
z>^%8-N$@#cA|TlMa|PQeTby;sUtPN0><>&N`yC~E{gUIk4qNfS%W~D*j>d`uzb!%5
zly8526hJlh59UB{K%7g_c&)?{awl6xan=i^OohNPK5%fz!bg`@xmPzCIU73O9X^o_
z9jotz{)dS2Rr(V}NXMO~57^E4!>iP>K%;MxQm%fFmV~wcD%c8PD;aEii2Ff{^%td9
zB|Q4`K|?p|(^E8Pf@YO?NngyHy&*h8dfnOlcaqFheJVuB_A1Ct0^@_ZFzHg&>Y9`F
zNY@t(YmQR3jDO8Ae^E|E`cErpD<74lXntE9Q}2h2DY}Q9Wvq_s`j#Ru+bsjG?R5)s
z7`OA@SUW?cJmXW1jYRclf0RrU9xd<ZddUq_z%P{^MKSA{6Jf49(QY+2`%o>5OJ0Ln
zuCADgaU*R-&HU2@J`K++N%P|j5}8jj>C%l8h7{HlY;iBYNvr)G&H9@Ii}m}{+*n63
z`({4v;|q^vf{q(i!qRJ%NRG$OOw;ry)-UyG*pPfDaebZj6aZ;Ls#fEaO~2dhWNV9y
zTeKs1)QS#L$&0;uC2FlDV{h;O&_gzYa4xPT&+j+O-#~XJ@Z*;-`sw?}DZ^Czx$vjt
zRtk?tbdzPzG8ERP4|!Q3L^WLtV{qKbag_I7hCbA$lE>&zIy&}xKNgZW%zrWO7xEC%
zuq?m#b;uI=uP;Xsp`!j1Sa=bKgoiZoCbQYAyhl6w%=G(4Xa4T!eU*ZWr(4qPC+ojj
zT>c3f!XAOQ%x2>>o`n!jtykP{nOR5gDOmJ+X%COwt)NdQZ%`_<EdrV$Uma|xxuBzu
zvlL8+wu;#UEw`m4IvjmBI=cT$<Y8but%SMRq0)t9nJ>OY!BnE|GgS}0n`JN`H%g|4
zY?ksRbm;vG5j4|9>!{?No-nHGc=M%ZH7=M?Azx%0@`KzJzn<r*oM2AR-PW|Z=$1*@
z;Fm#Jb)lh5PuT0)joMNY<E9@S9RY+oN_p=hNyf2@Mt;V5dR8n{-4c<ZnZ+cj_^~Eu
zwgOM8w^3bR{~P!GzgZa5{|nwX*7k%7YCV=yX373f!~QRor|(1k1OXzb8i+pKulWD?
zHsQxVc;9`qmp|YCbh7`&0DTbxpUe$KJpiro-%QYdF_!=EaQCEt!h5irSJ1y|^AF1U
zzki_+{Xr%53Epk~uVVgr&0iwUZJyLL)3z^J_%HwU|M<55bjkm}!TvwrU`>+*z98%A
zlNAkU_RFAj%@-bHIZco~g)T`fq1pl3yjTCz7yi-rA9FF(@3X@3&8dYbO;UQQsOZ{@
z9OvAkgo|12lz6+WiacSsM2O?JZOShnj6pYBv@|Ml{dpzLVxH%ryynw)Y=!863`NL!
z;qXgq#eFQjLRJz%`Xm?U#HOpj-{vR7iKuNFkd{1(^m`U~HXr1sHm69()2eF{x~*WM
z)pnM+XcDD7TXtW&Fg{-L;FcS}O~JF}S;d$4aa67A(bT`mbW<D}o5QGLv}H0y7smd)
zZuY?j9zvQ}WCQb1V!7?4qmK=?_)IR|y_wN8ddmb~t*R)-<8Q5Kg>PD1+z1kWZRs@(
zfH>Eh!Bti-i|p<wK>i;8$Hw3Z_yY+uXgB83YtEK5s|?G5lpc?lZ$)GIEhfm|V`{xc
zPfT)ANIZ>b7cZsYu4*&)bny=8PS1QR(suu8ys~(r^@NJv>0G*Ud6+uH{PwV#2a8X)
zRf>7;yIYj?V+guO{>4NqKPPFr5XfY0IK{Hy^BZZ<EpstVNdey5Ow*^@Mg5-85mGZ`
zZBRW_)6{8PH-fL<*&SG%)awK}m5{R?sO;N3mOt$RsCnmNJY5DIQSh6thIkKU`JwVz
zOz4>RB8E+MPxa(OiU<r@6(H7La_Jlo2=G+TM<O$aJm3cOMOVb1>BgmB-|;}cxp?`6
zOx&>|*n9i37CJ=vSk`p6FV_f5PwmK=wtBo`fX}sAUbF?#b+2i8U+_D&Z7nt-ddt^)
zu&%g}NGa?7XH*e>6khcM5~H2BJcv}vQ~y0AvPVDVh8Bg3%VMQU{x7=TGAOQ~>lO~~
z7A(OD9takkK@vO!cXtc!I%u%q4#C~s-F0wxcN=U5hmSn3+*@_)`_)ya`gHZ_y<5)i
zwbmAVdPAfaE+El==+rO#M_=%}mC^<9yZZmVdM14S)_te2k*I(xeLA-{L4B8}_gYMJ
zO&13{K!e_l+*VC6=wSpY!l->IeK^kTM4=YEOKVlux^4eu{N1F&WGrcJukj%KyE*+Y
z7?G)>Khqw~yrQN)#rx@xuf+-NuSdD=i@Tg9j*n$+W(H9sfJ*^R=m`jDaAC04u~DKm
zAmJ|fkZX+fB$@uaVdJ7#z^QJcMfqaZeiuKPuBEF6s(i&0;9JOW_0G@Gxh3U1;}>$t
zKp9*oe--$!>b=>qa`AeurZlKm@Hxxo4QML=4`>QJ8hvNmL=MCUkQgvnfzHWGDxack
zTL7_%-g-S(^1ARXSjCSWz`~MyvD6l;ZSBEUHj76_76aYmhcm*t(bsz!$ox>uxb(AX
zaEs|B*ITkV>WoZXzW!BblMF?@lxA%VSS;1+Tgwa7wT`XvJ_k^^-uOY#Tw|&2^NGFl
z)S#oo<^KY(Y2&N+^2%AJ#{rk5oY#J8YgeA`9r@!OE-B0K8C!*`;DE;203Uwo=ZkjT
z1WPvg`O@q%G`a2?Xzd6;=L+#5mzH-?B6G8dwZ-^}aF)-wUW4eDWFZ<;_9}@p%|{(h
zA1Ls62F<%HL!0l5A<Y1IHmF%oi4hB!e4th7ps?>_f6v89r`mb5oyP3Cx7Ex6QRSK5
z%N8INXxTaVwZH3L;f+J~_=WQ@>YxY`<;59v(5pm0m-Lc_n^3}JF~ge<${;9(A=U-S
zVN3dV`H)RFVhW1D<vzZU*}RCvF3z(eaX+QAeECTnF2HArp|hVP-_aZ}F*^0ttuBiB
zVqjW_ZRxh-5?7BT688g*kzwrwn0#hVWc(^00g;9vE51?VdG{vj;p%%!_YmGtUWLbZ
z3KF;1BO@@9e^AGw_n=q@hoHk_!DQ7u7A2smnllY43UECko1t$v^9IfXUR;t)VDOy~
ztf5b=Pek_L)O#~|thk;`{fdsG{jl10=&(?U?Y4DZ3T$&3myzphF!(GCT`L<e*tGnU
z1oikA$}sP12zG+zKk?rU&A!8+Zmk--TajFTpUCL-A9&xpH3f$(?f9l6Q$MqH=gbi=
zxSEn<f8qP$^&pueEway!2JO3Yr>-U}3HW<FBM_K-PjdGH7OhtD{9T6%@&;}ng=wMo
zjcheF3@mFA`<lGQ=faMrn5}QN#Y7x^gc)~)m3r(>;WvT2_(t);2a0oA-BpRf(7N3;
zx4qzOz~KEvb1<76@1uvrIL&17Lx5~N!GU5Ri^NhK=uR==`*i4cVKHF5t-o)Wa7cCT
zhg`UiLQNOo6Bsk?`;dkvhxs)OM25pUn^ZyG$>Ak*b(@X>8MEF__-!m+u+5smZ^khr
zb}HO`{CW4mt?lWhW^PvX?`_8$vTW*~Ly%fY`<FSI{1$SgHP2~z3HlmCK{mV8QG4QD
zhS8EkmVF0(EGGGQh(L@9p;A`?z0!X{uVr8+vGn@cWNU+9X(rH{a1zD`(t5MZyTDs_
zgMx?vyXu}bm3@CCaqhIYCuJE$hj~<d592&m3N-pHWZ$MCga+B4O;K{CiwQ<8>%Ez(
zY00l%XZU(B<$#GdGn32_M3*~#h$N*Hg_fi`4k?8as5QZFL|NU{FmeY)rsukvuzS2|
zKQx04O@_hG1A2>=-0TYtK*$SJ^D(*<D<#e&1|?mb9Mt2;a8fSc1p4l&X}At{xCt|B
z+<$Z#aOBZ=VY-1V`tUptbx>I-$&tKlQgQR2eU7o2A$Q8@qhYk~dY>dv9?oKKLCuTy
zwsqI~eyU_b#xp*x%InpQ+42w9bd#Zu*C3*49ZLeQE`cD&{1tZiEID7MiP8A%)4E9n
zUE^$)FV2KlcTp484g)v<ZQw>4#*(CH0U_dcI?W8!c(kJ*$3U}E%kbCyxz3BB&{pZ8
z00}Oi!2<>-cmM+uM!AFOMb&kor;p2Ub(Y4odj4vr9{@rLYc2Gs>#Mgwb1@)r=O)X0
zY6)h(;t;U7ikib{Ahb$-UfLAjg}$`XCCRFK0xm~}3t=t~<p81(rHT{bZ3RZeQAx`a
z+j9FKcQx|73*^gMm#oH+0uA0BA|f));p_c1)pmoifVT_S0#V%o>xlY??l-vI@<784
zHJi@9@$8$5x}0vf;<dH{B=LQa;!Fm+r;pu$V=Z^0<y^RsG{%?WN$=uGH(CU^%Vs<i
z7Ucmf(KbZXvH#GD@I}msNAJKTwv!mFmN6D*;V3#?X0&DYtl~cX=nhum$%aU2OBF8d
zW|k2hwwJK&Sss46Eq8*zjS>mS{z0nKV=GQ=^!-(~L3z1Ju9o0uvv_WT1h{JRm8S^d
z)jU*uA_w#4oK#z~lPk9u_jJmM331!xZtE$o8Ku!4EmlV(=D7T)G@5UD7n(T_x-{F}
zY3yV7MYj=a4-KiRN0>u}wHCP<4f=6n*|L#=L04tT_Nf&@=LdyW60L$BO>40Ym3UW^
z639+uHd8V}pf3#K-qN2+SN#CYD$*lKVo8HL0^^NhbXxkb)DwvQ-4URCYKRsEb<EUV
z)~}sxEUMl_u+i&w_Eo~yB=&<Tn4>L3Es042Qo7#W^8YdfgIgH7O_5r9Jo$`?)x*40
zc5$XDU$lFL-jsK&TSZlaiwDH)r9ka(4dLgV=ei*!2ApUSTzzV;x2;0FC#lrd`uGNa
z=k#OgYz)T6=e3{(@T=mQ>LqV|#Vt@zkadXRE^i=_u4BtaUoN!ASIODtQL_)R4sW{c
zMb=suAT9H_PJ007##UUcgjBpN^McE0R=!xQ+8s2FZ+hc(@6GYd23c2{nLB19S9)E1
zs$ju$Gb%QFF(?D4!#)uzYnLRZ>y)qGKf9{X#UCl%%iJ?i&?~=1J`8rAll5k<ihr%T
zdgpv*3@+I><>LZ{)$O@ew!FkDl4VT7At7kL=b`Tsx8=GH;qyG*aPQj%>x17rUl$$5
z@g3dFQzj%F_wMft6K$+R3p5jrbML4D$w09+yK!~MQ#g5vYsQaWiQ#Hwymz{Fw$F5s
z(9Y(Sd-xw-BW>0UBdv-QS^VAwh<Z;8vWf<8DQ~Xc4(}PN_!M|9I$t$_x4mS6S36V7
z?!RW61C+7o@(lA@JbKImt^$IkfZ$8A4^MNIKCCkJa560CwhB4|0wsk9Ue1u&$|4jg
zS%mX*`Bs-xd6o_rWfQ+h1AMAkIte-DxRL>C!LyWI_7AI{u&j8Y4!L|#T{2Q^H7Y?~
z7}-R!f=EuVs;7@%9is}H*YAhX!%jstpY!e6BI?jCEXD4JcJI5+(gPQ+;z%1hN9UzX
zwUW@C?2hnWyhu*Cvk!<ql8@wSB_j&=xD8vr@uu{|)8MDg$NtRJEb$1EM4xQsh3q%_
z2+vLPfojnk1tWW7>BK9w)3o(3>IeB*?I%=7ie-f=Ap##Mck}yIH$Z$7)|tCB=jm!W
z6f9K2=Mk-S!TnQ*<5%6o6JaGJf%_KY1^kjw;&Wf5$ehtTmuv}GAL(e;JF`a6KICs8
zM0>IE5bdp`z~tx(6OUBHt@v2_d7U?p%XT$O*9%f~j7ZkvvHj&O-10<d(@I5_uhb{>
z^59*+vqw9J;JggylLRy<`OCM(&Yh~bQP<g`_WaZz&Ic{UEKE0e=nc}#dpqxoGInmc
zVwfzitzs^|qjGu{g4`3AUe*(FyGCeoVPHOQ{e%O!cAL=_O(nSj4&a%$I~9j`^*s5#
zzOQ+0`S`XC7m0wK{L+pU{+ByLZMpnS_Oz(i<xukT|B!hoL=0tHSl$Ixh<MJe_iCuv
zGgqZN)N4N6SMt);^}fYTM^OP6U2w146Ee!V%;?7&HmP*$w0x*XxXH{$GmZ2X60+?9
zKwrfqndB_N7tD)$w2u-UI?yCNL^7miRBMOp0<*cwmey>%q>@Un6Wh7`Js#t*kRKk;
zZ{#w#avkxzKWL2ELPK-a3$*GK@r?<(q<A35{IW*o=&zo)T*dn$gKaKYUi@1nrH&4f
z_;9vi`D>CDO@qGyYTC5esdM}d^|rc?_y<{RT3WqrpM#<l)ztnz_$_4Om*PU&oF(VL
zR(2;=efgOR3$k^fk7XPCMlUZXnv=JO3+ruaqz=3IZ7_o6zAvADG}>RSW3Zo-Mhj$x
z={z0e&ix(%DIfg%Q5(>ldonE~g4uh0`m&*{H_&kUgv#Q{VC<C<GD|qJLKp)QA0Pyb
zhJ+=bBakE4?k{``rm5Cu(+0ZSz;oJvI#kGt8+>m0Kin7LD9{lXmAJRKimr1E-@^>)
zu~5*nD<=z7Kdng{^-lN4NqXMgE%M|kj54sR7Bz_{(MnpEi@)dk^FXPuX|cOUpoiP+
z8U&^B=_%TO9ol~W!WTc`%&3d3s@&nVD>V=OX(#n}j5rLcO+{#}mGXW}fB)JQ)0}7O
zc-$Lz%&pyUxD(-s`8a|*uICzMI&}ppBDnj}7)g8g$}Z8qV<n$krgO5K?uYHrJt)@N
zoV+I>LrqH6w)4?8?Plg$L6OPIS46uZv>7;D@+G5<kp))c`6i8#)EjvPshZoknB%hG
zOf!76HfdqPGeBa6d$Ge+>xMnLfUbksTuW5w%GkQ+xnD9!%%e$ztE#ZgY9ZSQJUwOs
zwXB1~cLNC^Tix@4^<O*wCN0H!p`F>q*6qNAeO%;X5p;Z_($zzC2!UFmjm;cJfjZe*
z21P=R$(byuu}Z7he^$G}c0#)#Fl3Jnt>gv3+KXTj%p~5cS}@WyYA(XY#>(Lw<)yju
z-6dLsyEDONT901ep0%yzi{Pg}HV=tKWVi&m$_~s@p8Q6uhCrrmim>%aQym?57xOha
z4aM*6_?IxY-*1JPa&_L%KmBP!@^M)&a4UX<F95+q%Dv7Bc*X9gAOpPDut@AYK%Go?
zGo+XtZUdw%bkc7%9BndQyU_CZaOTK^A4<0Ewi4G&^TF0_DkNvj@cGSBJ&rU(p^5D8
zqXu+H1*p0zb$x;^(PlzxT5OiW<$1r_K#3YBCh%9r8P8wB6QLg}P9zRUyWpMT2beY4
zl+>4`X?D;I{BXZeA~^{+55Wn9v)pZYdU-~^6L()~HxGNI;3sciGV<uZ!O3=(wFp{B
zWvP!98j78c6<KygfxZmzv&DP})sr4k{;7TqP-G-RCrjPdjE)_KihNtHT7_8QYAx5r
z9X4Q6U++-`q+&9#xEWQy_w}tH+B$WfP1CFR#i&&Dy<ag$pm4~p(M)69PLzXxuSW?k
zH?hCwCU)%sfq!pr8m+LXzLD!>uGZgkIa_h<uP?zeI2_tV1xK?dwjf%#7BoAI58^)3
zsiI^OLSYgh@7KqHJ7h_G4rJVQM|zJdBbbEaS1@Sndh&Z&g@QktbWX*{*~BEDOo?;2
zF^_mdrJG#$2#b5s#sxoD_!mM))%AMSk>h`OFzlN(E^-XlX7$e%rEPW;aC8$T^H>!v
zavPfNiC_V@;Rh<s49OskJ?g+glR1?g5^XEcX7HQN$Ukz(*3oGa9vgOH-i&rR98s4A
ztrZ<>J-g4bldgiJ-21x8NE4cVdDruWM$GP(TiHFAM}%M=)?HN}ser+X-mjZ6r-tF+
zuGGmr!rjy<;fa=-)9&GzD>wSk(_D-MsTWzTHrTtpZ#n{&f3ygmT>4|v0k~^H;_UQT
ziDBCgMusy~jd08==iKN<GdQE(`ln^h9_Kgq9d5}J)?eFQ+c0pW-eb7+^J(i}kZ@1m
zqbgpAR7@0lu}wU_YeY(7xq%3c6X1z!@i*PKVooSW7GG@z8z?}d!-sW(k2NfnPHl<Q
zGTsk2S3$Y3R*3mSzgjs|-D12pd@L&cqTZodS(I^9`ncuT*q=b(@0uI}YE<jXXtXSo
zD8VW$??Kr$fNObU{y0Dm>Mo@OJ*)v_QWP^L+CWgaufv~u6n4He4NgvZ#NBM-0#A<e
z)sv5fW5$oiACV^7Y=xIXY8+5M^c=ZjLjmGsn_E?~1l7a&hVB)I9szKjO*Wx7V{hz~
zxTKkJ*ZVH6_0NeeJ;l@u)BrXAaM?PmDV0edL(iU$Xyg2)dCB#N)aN{tp0$$+w<iuh
z1szTjV<cxWA8u6<awLbp*P8viDjjVCx|a`5mDa8&^)rJ3WzZ440;)3&m-<weeV>)I
zV_!jkzG(~}fl-XcOIB+M1^p#y>K!ygn4wh67l#W(_utMw^VG>+!^BFY?!I7+6gH+2
zI0tn0!DF;)TQLD><%GR`56^8&oL8DA9QP2KeJ2)lsqagH8!=`ncZMeKdH<#=rt%}b
zjFN@Rhjf$=UeMoSF`{e+W7iC{z|`q9<0@==6b_e)Q%c~fuhD)$#2Bfzg5mwurJu?S
z9QQpDZvm*i+8j(KNY8fN!;_7QzgON0DH*wTNWDLZP&Y2xR^NF(jlh`)a<}Jk2l@Kn
zFbLJ==G}@6uI7+?zUWV*J1MPG^$6U+#7nMt+Z=o=l>W}mafwYvy?n<hK)zHRQrQZq
zxBE1ku}svy6*|^ZkDdX@P96%TfbB9WeIw&sIS|3o*XtFEV2-vR)_xu%4!9LoC|WwO
z?pqsxX1__sFE*dNHkpXEoJ0r!Qps*H9VmNU@zw;<-+vgQ2h?{aXbTFrs{2TcgRncX
zGH<<<=Zia2;Qm@Lx*RjB+CE_upZAM@#5r?8KtHAdHxtpw-Hkf1JyT98FV5{`(`zzV
zr<LVCu%A$)>29q)M#$+ZcnMOP#KpuU+eJ~uEfCK)J7qVz61<$bZ?ipb{yppP$oI%)
zU%lUDS$%=KJ<a=;puv^&^2w%&yZz*OGY_fGzdOdfMc;2j{Rk8bF`+ZU;_hK7Py4z;
z3|(pBF1dzlZp~xoX055`#PAruqRcQOZkS{04}_n*j_}NJhnw_zplNkG`+MSP5f`<4
z+0>{0yt0fHHO@vSfP#_KXJ&5Yc@A0CPgiZTqvDY(1%n0P>Pz7?LnwfR_fxZD?)TNB
zM{dl?8Z926qy68Ire--J&u^|LZ^);#ma0pyVhgN9+Wu5LK+>nV<SgCJ_=<A(OOn$u
z?@zLHzwSD19)+JPV|ev9$wn>BUSb~CT<o++qM{oUWo-XL6IxDotxaw0^l7p|_*wMs
z%Sm@podGeq626*CN%VIUKtIdw4<l3J0viu`)7a~Rj@jmuOzqRB@h_T<Y-?RDhJi~k
zreA?wV}uWxZyJi}H`b8-K*o*q533`=)=NsX0a(t64W<00dQ6;eTZPvsVjL~;dt^X6
zbwHpnnYP~qaYPeca7t8t^&V)DKadscJNy0=`bc71O@LF1yTMZIXXO_S9g)L*kjcp`
z(9Ks(6S{UGq^JHNXVQYF4f7&n0`bC@5jXATa6e{nHg72iXMc;lt*O|J+W>$(5L&S_
z=5dyGqHU0#5p}BGH<vDvwhAGHm*sfw+Jj4T_CCJ|?aRZlUB-Lu3(dnkJZSfHC0<7f
z2g5xfkAzuvA?7;;Z>fmnS#|E@q~*Z4{dON#P>=j(`6D0|xId-jL^6^2o+$6$H%QJ)
zUgP42xKBJAZNroa<e271h8@wLBP`yrBXHc*$jlkvddb%mf=tS=j%zOCFG_fF-#^79
zvqs^VO@2$+uXJ`j0N~~<Q_K<1TU4#j7#T!kKO>9pCc=GOb69oZNoZtZZ~z5<c*;Sc
ze_`Ar46Ob_s9X6)udUz@7gjjBvgW&MAP1m8Oo{0JM+-nuz5WsrOY073aTj1@U`?d}
zUr^*Tg7nC)=2ClCn|VWPV3B(xEQBK_ozkhWfh&ePOuSpn96<J2r~y(3Fr(fBT-|IR
zpNYmbDr3@>mc{jeT@FV(!fnowk3Rw`6(?;N<+feJ&2o-E*?WY-5-iWRxixt6d-Nco
zKPDUu9P%@ve{h!(2wq5ykZ4N#y5rKZ`|(k=;|U3lhGpa8B|EvsbIG1ahN^+Eq|Vhk
zwR2~aET84*jd?bZL_cx-rA6P<GHjgWf;Npeur5Eae!~!vJI%NDc_=H`Jr!b#rg!XW
zXvfHz156=9mY=Cb3&efSzdQ}0Nky^zL0-vafS0N&_znd=L)Ta&;zv`dijY?AH`}5m
z^slIkmnM*@pY7gumruTr0$lULMxdl9;#F3DMFO6PrcP0U`RiPT5xl<*?ETI*#!X+%
zoIJZ}G7qG06+j5)L2*4-jHn+wHz8;)sQjLwRJ*^03l@`}ZQFt_FC0CL{o1rALFilg
zj8>i}%`uy_)MN+Wt`ONzV(5%I8;x+(T;HOYXHsJC!sXvI5G6KVMbm8>e+_NTDm>PY
zgj|@5d0)Cs@Hr0Nt?x<*GHhM|+`^4(_3g}jh}KD0FI+2_Gzg72<ih+|0N3hUSr>kw
zbgi7*oSn<<^JYU5zGiq-GDuuOaAp+L*M6!rsmMEYrprWu<ShR=BV5%ogH_+kqAWNC
z`lyd`{Zsv~OTY<b5c`H^0dhtkAcuA9v2HQZD|U*a1VgcX38P6&-vELB$%H@ccU!4T
zK&|l}S62iJ&Y$g%>q#lre1OJjAmvjIYs22a31ofE%QHgzu*m72>!Ykz|L`~3!cEl^
z*Y;;inY9sdIRo&IUUCvNOXAQ^%mvREMe$K`g$9?+t5i>Nh;sIS?**nr*WK?h<gh15
zc#esx=)--cd6Q5R$W0V9Gj^ON1>{&N*0nPNhnY$XS!jqH_S_Tvq0v^c?WTYk*Cz#Y
zt?TXlxls8AMIZ7#i#h1eah~5^{VUl1!zW_{RDG3P_fnuwz34wU#INhk^e_k%ToPsR
zBAC>rY=^&2Hg4RnoV6g(LL~|5tfp`R-V~lfhFnNyFeVmy*f;Y%*2{>2()-p;xk#K9
zc)bsOLVZ5n+jAZw)}%v3$?B&g7>F%=%g&E0-*%cJf)qP(1GI{9&US9@$|%{0bO<#T
zcnKvbaOdL%b*+1b5=ANkAX?>86c;z25|lGo$Jx!;gFLz6hg=R@IK#Z)5HkeVwTLY{
zyb7?XxAY6XJwkuG|Gc<}N!MCmj@l~xqF2gY=B_Ms{A4|8nFJihCj+$Gg#*5g_09mR
zR-qT7m1Bz`7WVz2YN4;|*S~n*@UZH?r1m_NZQY*8ESsY{B#m@Gk&U`P{WjVyoc8=P
z(1S+jTF4Gd9p`VIpOz1Gn}`>Ztq2DfTfgU&YQWEe$}RAuq!YlA3D!M}$NP;a4kdsN
zE*#wN{yO;0kR|^%AnP&($Xc=!xrIBd1}CvGaclPs-iCa7VXaivHah<L9yhrx2PVdP
ziAF|P41#3S-VHCG4Iv1Uj=h!PWTbLN?_>RACW{b|$}1_hy6S=Q^B1g7O+esJvRVNx
z6NA{1P>i^a^2ONjU3)F~mw$ef6(1FqsqXcbN!6z|bMpuS5r8P{wu=WilwC8bwVg(b
zv=msob0(1Y7VhwRO1XQGwCbBjnE@A(;DS1~{`t&gc&wNH|N4x^p0KFK-3SZuK0JPA
z8As#sCd70H=a(6mK+GNKJtq(1tt$0zIeI)@m(Rqyx#qiDKRo1s{6s}lo*2iW4;}}?
z!b9P0n`Mwc8@JXdrvg?`EUwjR61OIlFKV1Sn_r2x&iz8&T#4bsXM=^_KoXRTdn3n^
zzzEPRrjlQc(##>Z4SqT|u3cJ;bq{^CTE|q%2<N}`YC1AvezKLhxQW2~k|~q&plzu%
zZKlS2AnmG}Yu@YL|9h7$XCcSMuiD9#5xlwPl)B5C+;bV*CX)S(tcHFJO#L^dKaXy)
zB=BDEO!&3pRytV}p{o%8$RiCU)_L2G;TgXQYgcpwA)ESQ&QPloCd{u+)sG4d(;jVR
z2|nTAwT6_!K*k_@>#&Z&;nTVg7Z-+hUz4KaBP_yy8@UZ3kF}46u-JWdF%xexZC$Uw
z*KK5}zmkTRGgM7u3Ak)G;Y_~GiKu&`Doy2bqdWUcta<&(F5^TiKufhU^PY{6xaV9P
zkY_yR@qBvSy2%=|Dt<iQ@G~%^JIX)dgeXGKtzrs7#5r_*7sndbD!xr9Yv|8H_N9Lp
zCs>S{lHoY>>T>FDl<l#Ta;2#dj+AtYP^>}kky1C@10l$Pj&vu_xgzt-A)>lY2aPUt
zB<ksXJdFZiQ0ek~XZrd(Z02r6U$?y<;pby!W|opl!6|OV3vK0;X&_@=`I{g=JsP*0
z*coxr(mdk+`eOeq@g@HiBaz!L_WNp{NQJEnf>_V3b=Sb8^aYvBtHf0M;MO$PwQMX%
zXn~LYNEni_e)H73A7s7_S38uSe`D#;$#kl)=pc;KYn}Ydb8y#RpVR;s5jpr`Jwt~y
z-RndO7IC!|NV==pc;ls>V&Cc@?q=vElBD?N2KlOKA(`*XVQg?rPkLr#$Q85V<{SXN
zWORcnyNo<m(Lkev%j(w_%Y3Su1w%)Do~epP`}hRRYVnb)L8Uyzb|l)`_PmLg3y-Po
zf<~XG^odv;kGhZCtD-(;xN!|%sqsRYdx^VnPva}3ST>@3-vT*R&pFWqSBiE}C-#we
zMx$WIp3;<pIUqo-NmTj=8~r??(0x#jgx*edNF<A%90GjN^Rlrh0l=TDw+5nfDjYB!
zzqBJz`eanI5YOBGdTu=#R#W=C#=JHM<70a*<dhwJP{F*o!IrU2k8V}%e#aU>tfp_R
zi{9WA*S2#M1QkOUQtEJAM&n!+oqk8Q(6noJeJyi-RZPA2gPb-f*b8O<ku-XX-lwpe
zVPfzK^kr4QisiFRgSx#*@?Suxg{UFyhSZVEp3se-(<Ds$j%RL*)1K`Q1zXd<QW+~Q
z_gM7BYp#8bmqG>K;>_RtEk$%2DM&*7M}V|}`A1AK#yL*ZPI_KKt;(o_#EO;OgBjOw
zCnO;rWU-HXLo~Fd%LdI@BMfHuSjQM9_xTo#Tkw4bIApjTqVt_Xw0k={g<f8`W+iG`
zsW*ko!E$9~oJg20a1Or?I3rpebQvf?T*Xfx)@j3vMrjEQXu_qV26Jihn*`)6VV*uD
zscA0rS{H*g1a3%u+^ok=J!4+D7NLVnh{%IN9hrm^gSU9JIZUgY`1M%SLJTXrrQ642
z{Pe%wNaPOlJ=?@>Z<XDaP{`WEFID5i4=nOJCw^(NdzziGIIY!mUM6@FxFwZ$?(B^7
z&pcwCL>M{wSjTbAy?5BHYJ}Ldlgehd+e6>4r3raWW|J>l6Eazj{Ufba0Xt&IA0Oxt
z8#~0kvi1kzNo?^7vQY7&#y%yVw^L)gd8lsvqW+OifnwJ|kU72Wlya~BZ{Lb9V37u8
zqnlH@a#xx7jkQH*RE<Y){i#3G>W-adSv?anB%0}Czx$0E6@9mkd-|JLcMOK@<AmFX
zHrH??E*k32pYvZsmQl7z$>dhJKu3A<^jOQa-0REL$ff*LHGqsy$k7K(4d8udyHK;v
zfT*9&c>Xu*aZETT<%6E8_k(wWT~Vj0RJgE@(F#Et?WE9ptroX(8_p3i%EKS*qo3&*
zuv-<~ta=TsliG+_(UZoQQv7Y24Xsky`bGKmM#gFk*J-^)TieSrwynyO0R3r*j^gR2
zE2Q42<*ELEY()QTQIh?&X7<uq>`1XHg|h^*b$n&(FT2YCcHfbezvX2SIDb{#_=VZ!
z29tw?+vLMYiioFT`+^rIV1Hl5pGP_+WgX*Q%ZGen?d_x9M^z?3lgkCbW04@@9ds8^
zubhH>4;s?ur>-2Ji1=UPT}Rxb{}!r`64A?>mUe7MGPHL`QBcr(WRQE0VEC(6>aNAF
z)kYjBCm(9u*S%MFw~4Y0Le{c4ww^{XF4dorD6hE019kRuYi$o3;ifHsrN=k-tt{I>
zzZd`Gxt66xc)_=YkfoNay-wOK$^YO-QmNYQw<r^^ZwjMY>Qz_R!H(xArV=Krwjd+y
zYyBkc_qq?`f>}Y0O4{u);f?dexe*T>-u($`<*O8;CJvQk*2SRn0>7F&QwhB@EuDFg
zE4v%;O+<IbOjM-I!f>n($*%IT3OHghYSo_(Np%U1C*hgh-ySGLABkQ2(_L;XpRtV+
zh>TM^btgYW<`@+^h-gR@5^~whT0_*Po+FWQZ`KGCm-Fi@@?~W6xlL!nnnhVbHoaa)
z)zjF=dseEH@}Ht~jEJiq-{(m3?t8m8_wd1a)R8J>iYBypZcy{hBSh1{@5f|2@Nyy7
z@D=T=;mDd`tfPyk-ecbh)a#2Fz)($CN4*gnu)G-gcOI-KDp$8dj?TrxFM1tHwq0i-
z(M`#hc$!_7SN(;RxXBWDd<_D2Yzi=vciHn&y!T&U;lB#S0;X)IcFnsiY82eA(FO@|
z-J9n>KkxetH1+#Etq`@rsRyNXGKs=8P^WF;tA#66bI9eLS(}md9DXyXlj-R{%3QxC
zR_Fh?Ht#V!GuVzUlQM{Z%?j>>@BfKM9UJV1I=cAPyPNB!zx_aWjo83KauW$4<yi~=
zNLg7<AqU?mbYXmE#UoR8a%RN(xSHkKtp_q#8P4=uHFD=*Bk0bf+Gz54cQNgF>g2%v
zo{3l!kk@lGk>e&_`z^aUX#&ONEfs^&D&U8mQc)Te7F?n37q4%KsHvCm?GTHqz)Q9B
zI+6_7Al<97J^>%a4EMMXDT)^oBm^;itSL6Kc7HGF$;pNGc2aq8Ulc5$4PfY8Z&@B+
zq?Gy}GjBzX`RB=M$D2Q`1@|`{Ro)3BddL$?lHt-5gu%S%`Y%}Ah*+pwkLIaBOKn|c
zJyU@PbNV?E)C@+I#AM7j&Fy3-!#KWfLDAMYTEPyw9(4^XJ{)pfLvphboa5VSSqodX
zA4HYfm8>SLBWwUxtR2mh2uLNT-LmPymg$;&r!+T*>CM#dREWe#VRDTUFKWUiUgu<K
z^JmnP=MEJ#p!M2RBf3BW@hX9J?G1)7_FSn7w@H)!+HBjejrD+I!C8a6Nq@SKw^b3I
zLJC{B>(DJa`!b4LYTTZ0x@rpl$A?I+pL+&P3vl^n`1&7;Pm*n)`GX4K(28&riQGt0
zL3G5MZd--mjC0GwI}8M28KXGxp@D;li}oPKx6Mhbc%jju>z=n<_E^`oC{9=BHK>@F
z@7G_=&PA1V1rOHQeK~T)g^-fAgQh?N1V8!itl+mL)2CY0zQA#ke_KaSN+~!Kr7NNO
zT6XmJ6!hs*B-oN<|A-D^x{YeGTeH}ABiR+_Ck!TDe=lQ|+GJez>{hD_(w-WNV@4~$
z#qwTCno0q_u>(K5w!9xH*STYMgilN|O9#tTNzm0Z?3SaG5X4#Rc{+JKn)0-vIc&ZB
z<y&)TQ*tG>4Fs|zqyM%QoY2mKi6H$Lj5*4^Gom3OJ0FmiXZuLx1UGH+B<a_^Bt1%Q
za_5L$QJSkF)b7<F_cD+k{fnCPaR}eYj1`Lm01x6`F+l0g7-IywuY?~w-OMwfWx$W>
z-+$4E-jwhI?W?uxF)_dR(fnR@^WU;#5Pjp^@w5#}Q+c3BS+T*vcyvq7b-n%Zi74BC
zs7N5L0Za(w-M*nsoCPvIA%gBB$$TI5_BYU+S64Y(V5Wa9y2`swiQ+5SIC{ZuTyWG3
zXbLnsBxAKkF)!_2$d@{LxK+?|pafZ?-<l2!zqldC_O96Z;JI;2yH`&)!T?>!0+og+
z+OE?Lw&scRETp4o(hn0&7uj@nZ?*YFGI&vOnK?VgGH{Xx8_xY|Rxm>AZV+dBOxh)2
zUJDKunypDUkRlR7hr%)9RmanPx%>d~jRrAVjc5k$*dswkn@3534UxBAQ8{9=OI!Mk
zT+4gX4%J>p*KNTdX)R$hYAR;$FoAkNL%i-glAH4qq<aJo2luf1-DOU%-fwK^(Y@Y7
zT0G_h4%?X+XjXB_a;;n4(@UrUHUvqan@b8;l8%YOFbk3~IduJum1_t12;~U_;@{SA
zm^Fe1lp_z3ZDkW+oV~gw?2H3;EKrme$$t6AAhG3JiMKQTcRuIV-3=$=uS^H)XE!uj
z>=}$X>(LZ~_r6ZD8ddX0KCLaw&TBimp28#+-wT!N9^XklZXROwU4JAUarAqkv=*P*
zvWWBQ!^DyYk=V>g7%(t^=}c`$iPeAP<5s_*TnkuE8|q+L6F<6S3c_z#Y3<mF9BIUz
zJ8W+sC3^lMzxr6pX00~<mMvvwFlbD@EK`gWwN1(qNIZnuG?;%gKF>$eqL?6%X{z~g
zLW*HO!zlP}bFnl1BZDKD011m(A5k6stK+>#VS2XiG%WDpkH*eg2U6#RUabIKm_-yI
zcbskXJMs+dlek4$#|a-dMk!@u^4qj${5S2>$G_ENErORHlI3DV#a&*r;xw9m3DAE$
zT)SqnQg(}yGsv@Br@%*yp|DyYLL!lW3s0unC)N6(6x=hN`qUcF<FN=Oa-UrqYAKG)
zptX|oZR4!L490;=fTM+ndx$WeX)uhHp6sUx=;j&#wt7lCbGQjtCRAP4tF0I6BIWCp
zb&E+qaL*DJ7lr**;Iof>^$KQT#Z}!M(8i9qg4*HXNHA4zqQm&1Y0^o@Bc(>$^%maK
zZPV}XL%_o#%%LcfCv%^XnR4&Cn6FWnC5$^o?(I9h7)rmS&~xwx>LXb<1*%+=7}@R!
zPhj?m*Jd{&Ngcvfj^awCdGyfi!tt?1b3_G$fOk5;EKo^kOXBUi3X$Ir14*-hyHLQ@
zedQJVqRo~EYf@kY(Gf*8+XVSqJHIWFsMi06tP(ToG886RC4X$$o#(n^e{bVqDBq;U
zw5iB!26>ZcHQB16Ur%F1srEX5vo@yzaJ1&k_!PiP2ygT$ut!2ih6V8_BW3^(i)m7Z
zD0f$`QP7Hsy(DQeu134<UAuP92&nV2W`%atP5pX3wQcBFMPx!s^fkiKAcX)DW;Gm=
z!X+rD)xlR2wOL2!&r2<DF7&})V4BO=ZBmOd(pAQG7~28WD_wj%P%hkH$qu_c@0cnZ
zV!=$vUo&Tm+&D*b-hD#kipG>pjDcvV&;mC>zSrtpS@B_g$71te8U^QCiHRqAfx+;w
zf5{?@>}r?5Gq~1U+v@^e6o-3T9>#v_;vPju{W~~%fB6O3{5dMUJ|40k_j9)JXe5Rl
z{3+&Ey?MK(#M-khfwSFi6s@a$LTeVd9KkZQpk0Q05T*PW)DN@vyG@lr$Xg6cPjgZG
zM>_IjQ<IC?wfZPN)l(HfmcP?=kK}^Hhuj0}%=YC^p-!(711N)0gy<u=syjj(_cCX?
z=uchGkYF>ci$YtXMv>|Oj#XRV+**&Wp@rx1-hHri@!+F0_A@nNqq&#tV`U)vkX873
zOH-jaUW3zNy3acpPx`7Ey;asM0B>Ra_l$zB!H-e%6}TQ_$#!B>#1y-WzKNLbq?Rww
z#6l3F&U|}Z0k?8@0||5WuWLLoKaYYwhf=Pl7RUlKysFR(Jccw>ACNFoiYh;HEIxdx
z$`A~EUxx$unsXe&nsxFf^BWw$4CT{g%i%bB$gi(ldT+mOe`{SK{98O4>-mG9jSdo-
zEXRlJhs<WTcESB|LjpN(mLUuYO8CIHj3l{?LnwnYXZg#k!ZGW;3DDN!<hE6nSst20
z4Zwn-WF~A7+KS@Ac+wSUPJ>KP^UgC1?Y5m2i;%(y_w(7CeE-FmHkC_UcUV4e6$5)=
z%Y~<632D?m9EP99<kwQ0IZn1<?X-9OUN&>hFiV_1_nML{LQ4TW!W#1P^Tdzs*bt;f
zK@$0k>eGne+*_*LJBtdLs9hxAHt(Kuzw=8+ZiK|SL+0eO;82#hiC|^bBTo`=zDM$D
zi^7hDzbvuujYTY@grzFD`X^#Kr$wZ<Qv~;=I9<F>BemJhu5=dLn$(Q#BzINkXQKcM
zbfqxQmTreJngn+dQWnPfu$}tzbZoxmCo%<F3%kN9H}<D;6(cx603?E}#93~&)SY)=
zqMl{#?DzA?KG6{mVCBONb$*+)s|@qO5LFBaF;?u*pdPq{o+OcW>nn5_F*g~>NRT*z
zbwM`2BeK%-xYRR(Ne3El>B*iLBy=aUzrzLlMU7`vk9<~}3QKUCNJJb&m}zz;*7H~_
zUyz;fG_&idTg5GKwSh<f=^gC!c0rCx2mJ!MCBAqW;GJvvE4vFyqES}$qZ@np)k$dC
zjA5hc2^zIu3BXL2?9Xi7C2PwWlNyP=_y?iPjd>BWZ^Gfyxn&9HZy7*YdVpIE_>kB!
z#l=+?S!{#$i$zrQ4<j>a!!cnvZLn=Ny1|3!pVORxj>p&9TjmQD6}6YU?QHlwqOuu(
z=NF;67qbfp3%4)o+&!VCkI_4Re_zex!T{LFndO*ZK3{Wp$wA*bKppCL2{Jzd6cy}}
z=)Mg}ScEtVNq?~rD%xy82cz!QaDAxlnIl-xEYlI%eA{N@U|UlWbiEu^-)VVPzo7f5
z(wit<3eRKHRw1O)G&su~XoI_z7kCERU%Wxj+|9z;<xz-)$muS1Sr9zw2G(}fyWm!H
z9Px(NGQHO;&k7s_8(qZ9Gy`$B#5{TmP}@&6n^#ZV=srBPZPeaiVla*yv7IQb=k!`M
z(@m6Ksuw%N%B+W2lwqr}n13C<lovK=JoK!VdVVrP)#I{igx}y<f1-8~WYo!UJ}dO{
z-f`N(iNSurfz^l-1{Gsv?HemIO2~G;*v**$!`ag<{r~pXPdZ>y6k>U|Dn6M0&f)*w
z#SJS{_$zzsCsxzbh&1J(>t{T+IBatZ3z$tlDkR=zfjD~UZ=hy^%{)t?JOR3hI<-n5
zcE$bhwJ#~4$YvcCj+Moa7&d9XErDLYyE!BHJh?IHPl&xHc4OBQ!F{8|Qi5-j8-qFY
zPm1KS_0!CwexDe3w=5KYlZn<1oPsi0=}H*$$6ib@ZSfW9Zb?RHzGwSIcCzVgeRQT&
z+qCNRiBM7^<l&gm2aZ?IHKzmdb3kPpjZQb)=~;fy!)SDl1+p?TF-E}48J4dsA$@V_
z?=2ucgw?pLDb0eNwh-jZ`I=uH@y85%y3=6pK6+@l#2DUX$TF2AubHC5nWT`{fY2`B
zQ_2_owO4?)S8sr(6m6vkTH^Dy%<;t+8F}a<jopmgzNLXZE4glplZ+<Lg^qWp@}&#T
z%2%cKeRV^RVb<B&Yu{F!;1%!KHM(ULG71Fn4ac;b2xcRPzqAxC?C*D!!pPXLq#t48
zde17wSSje9NgE@>%_Nw$PA9=x3$Evgt23C9ULB&2<Ft&n==`khRd4}cm$O?eIEPi)
zs}6}NHxc~`ur}+*Cc29l!s;fQ+Ry}jyvtIP@q(}^;A1J|+pox)_Wfiw8)Q}R!h^b_
zo?+~S*zTN^5Haa#7#M+9k?(j{+l<87kYT?2zP<5Qyu*nmwMjU%BbLPEjzvR)|J@rg
z5maU+s-ZAu6Spl6O6uv~0IH`7NY&85h&2B8C5IpHKr-^shSRbSVtBQHTo{!FH*-jC
zBgy^|K^{U@t`rMik>mPS*{a%OmP%Wg7NetThqQXL$mqK=DZB%n-P&tA6nw;{2-bpL
zhSSsj>%zaDu)k>;2V(U!(rv~n3Zfm2{*A9qWcN4!l%s>Kk74_LK|0vdeMhiFzU9V@
zRk^uc1Wg(I@a5;-hEQNYlQX8X<Z_@+yoH3w<|-0v``|`z>^zIX<Q=T~YY+Md;qivZ
z0h7;i{FL&q11y_buRLlX;AxvP>uWc^K^Kj{+3aTQ^mam2clm3B;v=+I#darl`Ww{l
zra**J7_$+xU-mQle+>6#zi3l<JGoH|IYHF?YP9UjvAwGLW9cuw{B&{w+HI*9eXt+A
zQ+g(Ue$|Ng+NY9#;}0Ni)ev9yMr>|vfXQIg^xWAtCee;Jz64^RsaZyH;h4(onX$6@
zZyZUyIYu}3BvgF5fde%Wv;W_~puuMLhV4}P4yO`BZUY8n0zAE=k&<$E%J)VYc&=1Y
zn6&G~UdNF>Gs=}0XcX{$BE1bFv3ZlqQfB`j<5GCR8-j)N)cSVuV!k#T+#4+3)>`hV
zImsa(BEJn_grKL4PjS-`n;~s{2P;xg?6s;NX8WiAvdIt=ykPM5o?LgiP!u>OIlMt(
zX1;r*&>Q{vuq@e_)j8RA*Y1S1EYBbNe;nJ|PZ%xUdJR=#V$(rL+NC;?>|)K{$h<ff
zfzYFl!j~>&j%*#xbECrl_^$_#bAc3E^a`;Y4*7e(Hp*U#9BrFlQo&>>|LvH5Kfk|A
zicmdULrbt#`!3UqH2J&X{VU{o1;!52mCfp8?lqqZjeR-7Lyqn(SE@=um3kY%|DIs~
zPcf<buU2_<f!F`>`~P<T|Eu6bfPMQSJgb85@Be>W5=O4Wf2(E5n8!f=KZE>#HSs^q
z`^oyfO*B4))Z*y>r!)VSfd77I9r#vj8P#GW_`l}j|J{-wx^ULs46XU_aNz3tM`84s
z|GX!RB<A~O<M?W2rtcI=I3ed7@VGDe?^<ro3S-n?Jp`V-^w@rqIM};UXYwO3dUJgA
z_AJ1ED3Gk*+UX4DY|ZrZYt~Wx{qvbu^9A7bnyr?-!M$U_`IL{u<Yj`k>Y{N`r*yE6
zBT^@PSwg2j5Z`aNce>Hrf?z~hNJS7gB8qH`1DkhH6QUvdnKDas)6vcT37^&7F&7Gn
zi2#4^UR$6SX*dAkR=yyGr$6eUq#onl5OERC3kZ(bAc=1HE#aE>e$yvvUWIz>*MDJc
zvo&jI7YxRjIS-_f=&V`lK2?T$%F-wP*$&Y<8`d4s3KcH6j5xKUNd<IJ2d7GO(Yqap
zXYKt&+uhYq>hwGqpp-l+Rg3hqBwPITV>`HXJy4A{(EPN2-4BNl-9P=5QAto1OS+`P
zw;0Zo7wGc*GE|egwd}KEgev5{)_T!O9<0u#bWHl_$JvId!_qFpvODmXW%C?{`F?g9
z_RE0-MX>m}^Ale$#*ulZHdJ1ZFD>I*-X-%;k>fi>F1G%lH<HflxlMY3jb17mIj0~D
z@asr}wI7741L5{OoF4(9+Hz7i)x3J)Gpdb$cT_?s#WE*T>v+65mp~d2OZ=v3U<wm+
z?R=cpiali>eMgIeA3!m_n$auEy=xq=d!<J2&~)D$Vd(rFzC>+3ATmd0(s5SPeAT5A
z_`4j}aJzKKOwc!W8W%gP10~m+T~GQAC9vvEUeJr7RKBEzTAw+vjj4D?qU&h|%T)Zw
z;=bHF_rt?A|N7L(5n6B1!8aRZ)O4T+-jCaPxHNv$nZc)@#MUwrOLgC!f_!8<;bPFJ
z9*Q92L8?t-V<FC`0v{Z^IjitVD*}T%eLX9jkMd5?lU+91V<Vx%+EI{vkt5zHliN^M
zF)Da=R6q5xx9w~$sJC1?DpY>0jtFtrcs{&pQ9{B%VyaUAw6lEnveo;=r#q9ZQ^N;Q
zdy196P>AX2*r(2Q?V7eny0eKq=ykJovNBFCv{FbO6-f?-|HBrqNJB9tPl{)fg4eZW
zFw$>smqVfK1y|bdN!RDT$DITwX*O5d0CAzWd1>b-99~X*>ny02p4}VKuickL!GO)9
z1v)?R$OvNK%U6WQe9^c7`?t;`$}y>+>XBK`)Z~AnOYK}pAgsnmU{qIdPd_o_=&Y5B
zW7O4KAbIF<2B0R7AxB3-K>{0)e7Hm24U=0muKu^LedjkXL~KoMOV_LuU+rRT`6YE8
z;?*t#_UQhp>N)?ex_McbV25!n1^e(F-->d^_|@dN*R@eFlF3+1()s)`fNxb@rHs?u
z;f#!YIUV+sFZR%TcrLk@Al3VoK7%-#WcW|G#L-haEf&cI8(Z@O>6b1xBygAcaH5I)
zxWrN3b1*+d#q0v!9hq~}FZJG4u#gH*+%vb5ihGD48)LyLu8rOK8KafY`_=u)cRd_$
z3TKh_ln(iA@^E$zWwt^Z>BPM{y9@FO<zMyqY0McLAfVO!L;8mT|I7p&?3cnUD?(X?
zmv^_yUCwjQ5178$+*tl3+CLgib`Td`@8CK1pxrbcxLbPQd*<CVUW7f*(rhu^3Rld{
zKa{=MLK}H6+PKP{^@Pw4Eex3^Wqv~L=}r$xL3Gkd2^qYgN~SfG{KrUPyj~Gmxx)n?
zitsj|C_uoO$(+|2FNrNtWbUykK4x%72=FFM>nCDxcVRI3rsPE=u&)A9r4e<Pze9Vv
zNkmzmZa5Dp=wS@I+dun6`GxbR%{%lkZ6cqcET0oBa3J@s%J(xQ3~W7ll*zA+<hyGe
zBtclsn$27<@UkMWn{O)6cOu@9W-oftic^%STqJ`7Kh%&<wVD^;KUxyKhuLa<y)^fY
zN4WtZJ_sCkv{dJ6+F?z4k2Is>&zMXW&+n%@Pv76de)Rdc+InBxpz%K3lXW%Ik+jT0
z08NRj1B)g6nQz3JL7B9CP^BzH$Hu4WI)iOekH(dNg!lEw22=1~CGGc~J$no5$Qu)#
zVhrGVq>=1IuzQ!!`;A|vKMcStbDHVH-4E__UEuDj6YZg2N^p4KRg}l|evbet=;#q#
z5*7(2)m5*Cf9aIPeK1IUPHL8=W_;7&Zh9k|zB2J*TOn$JA1SBRv}=*>_E)B9*(zqP
zf{RQ&{;f;vv1!?@Y-p%xfVP&r&*{I-)@aT0$1EWJ8B`hG<?6$Vj;K97!wdPHvF6HJ
zao~~DhW@<%#SVn+sZ(11J;Ub@dmuua;i3H9exbUAg6(CCOTWym{<EkcH}IVnS0^Gl
z$5yCQQ?f^oh>-t_R3hf(LyM6ain60Xl#q;`7Yq3n{=F8Of=?4Fl(g*k-`si>U;v*-
zmvD&JqcpvHGrGrfuV}K1grp}?P(`j*UH&1BCy)2g*WCq9?seFE$}hZ)hBO(#dbv+E
z6PEN48c~4*x&7^@&l288&ghqiGq<11Sb9mFwdqZg>~2Lp7rA0YB}TMn;b@R#C0jVv
zR%9uWoK^O;UY6<POcqdX2D#S>$olu}3mURreRNo^_JZ^CO%_$TESwwlUXjD^U>Z~?
z5AW~TfAwLm5IzT6pXSGG6@hjdtaz7`n?RMJ9fHrFuoaeS$(#TS8>$_9YgTpLY_kiV
zmU&!pjR}|y>9VcLuJ^6s>z`@9o!GXnquxfmd{DKmcxR+>#E`u4?BM2);fv5ohQ{*h
zh1te^Oaoj=n!GQ7I|L7e2@p1rRwyuC3M8zRTZ8<c-_5*ev_BdGTQ#LWd*D6V4B9o9
zSY)nhN1M!$|G=z<T~yX2ZY@Q(zG%{!E__w716|-OR#x9|2Fmz3w<5?3t$T6Kg4aTU
zvmU3pP+x=FpT5~i26~6j?ecZ?B(m9CbG5G#7nQf(dW|w7jlH}iWOuZaXohNs)_H{d
z+WK_J!swMB{s5oXU*R~Bpx@`dts`wU9E-79acqaUG(*$^fD?$!jz!{Owqz)BUURTI
zEeE|}VlxKsph-Vny#5IKah|#OjH}8~POU+aNqnBw3KXdE*7GLNKte`fP4Ns-l2yr!
zITb5oxa=_Yhy$pPDWm>cCRfc<UpReCq!;j4sfR{Ro3SV+%f^I4AO4KSp55^Ieci&(
z>mG7<2Ep%}qTHX^Rh<@Vb}R0)Vp$H>5({_dqL)_El0Kzb>z|N!eETR@!0=3Eea77z
zk)*_Mpo3IHI{4Sxp`z1VaD*V@WG7@TT*db2TRkyS$1o4*b`5FFSS6E3Zl*&0@wd_X
zk7hQdooj{x)qS1A<jS2Zd6H+`TADp}+SWmg0N`{opC<b$yo?xTn_PieEA;Qk2r+d=
z^X7fJQsK%Ctxfb8kup_KTgUiAHr0!O_^1(QVg3^h{okmBmhJ9)5na1^LPZZ7o8H+e
zsiVP`mX;?fZfbYTokrkQS_}zg*~6!Z7RP32{(%=LEqg{Ry+(I}xGZ|1YGnNK=9_0S
zSInp)_pPkE8NvpMQ+mVdVVp;U_wm_}cIWlze5PZ>G&}>Ys8-RG7YxwGydhWTb=wvH
zd4y>#n!u6UPo4fo30D+gp3T=x>G}l!{BzSARR$DI_eV8t)uOG@qy#Psu5$91Z=U<s
znlw;`Cd_Eid)f`f5Xq|v$|c8+Q@a%QQ1I}njr)G3RHdQT2fg*eM1C@tr-9o&5|QRx
zTc!TR^TI*K6WT)Oi}(aut)XJFlMnj$g%f{`rNIyPt=oG}lS&?5qe>leCc!70&AX|{
z<f9dK`(FyzhA~}xn1$Z|e|)`VR9r#VwwVNiI|O%kcMSvxPH=aJ#)CAP1b24}?(R--
zcN+J`y>ZQSo_C&izFBM5{OZ*|`qVk6?%q|k>$-Ps3d*qZbsM!zFw<2g7jA<t%dF^h
zVbRjQ7t$1_^+%x;71o|E($a8PL%4bE*No?cK1+T7OT9}nd1?}6(QN{A<x9<$BSnen
z>M45_nJGM$xPgu>9r>f+Jc&-xd1|UVhQ+Yy*F-vZ|IHhw{Kx&Pm<|jhcvi-fU3|xZ
zx`ETKRaW)^gAyY4EB%b}=DHUL%qWM<88alN(Y`V2H#lV-yl@k`QR0_yIj&8oC8xeb
zv!=W)XIFUQM>QAKV@WeQQ<XR4N`1yJnZ;b*pzmr2;X+#WV%+RtUq1df52Y-NzN~I#
z(B<mv?QQR2=Ez`bb-|isJmFUgkGvO`&Lj7kIomXG((#jIvteS0wfi5Jrs3J?jAL1(
zF1InFsNi2|j%z<ERZgR7eD|ZIH-^a!#To^%k4(I}HIku+ySE_JdvADxSX8Q|!7=Q6
z2Wu^ZOJj%hJ68~PCF)s&n4!(9C#GYDZkMbE>Fs$Bal=vaPx2SD*ZaZ;ZMN^v#hdX@
zvY3%+H)vEVz1J%a;G&v$C$qBlW+Z+0gi~&u7JpvRtQ;J~Sn^WAj|p35{yd29Ocspr
zW}G~>+`lN$iQx97IC+2aCO~*Ze^J+0uSZR0VJ~`|F#z}}W!YEZZ*p`9dQ!!9^fXXu
zWSi3AkY_+Gg?X^gQ7z77mXx`rsmVqrcU`KDgWvLs&Kc3H&^MWjTRXCJXAMzz&$*>x
z&Y&mo$m&|S5?7mt(T5*xlUcjITQ+!}NAvk&(Co8@5#?UBYmr5Lc%W59X7L;-rX_%a
z#lKzP<?l0wt@})QFtCa_V*?g6Z}_OSffkkCxy;?Or87ju!PqqdBmId@_(#KDhUf8`
zCas)RX*)q@h~6*F2_S(;ZnGMZGH?BEnE2ia9<lKR+2FlP!fI;GLs2~9;>E1-RBKaT
zrKFsXvaN?%i_T+!LKDs+36TGg_HJs`fPH}4_2v1qXy^rdsx{5PnJY$y?R`ZqE#Zaw
z61M<V%8|Qg2Px6>M<fw0(>45z$?Z0aaeI;7BeP|mc4kSTwZIs?E*i>@jZ7~~+l206
zqV*MY1HMzvb4#bCW~hD$<C5aFQ}`*ZEhKC;2Uoz1q!|Af#y6)%f^w{;8X2m9YOvFo
z@xD2VjkWf1G<F4f&~wzp*vTm9FexO4dhoJa#7--JyFE*)l8c@?h=P&a@$*>i%)Fn!
zNQuV^5<1WEfa7b+e8FLIhqL70LC1%2J{1lg_+K=egF8lt+@fuYkw}9Hu}6!RzNxCo
zWGEqu(nPeJBc9#f`^`6XyNEHfT3~EKn_uX7Q%aGOG87M$ZozAbZjL8|W+x$VMlQwO
zCSfB>RUvE(a~C{*1m#JQhG|xr!3C(@EEZyHC~wgJM^2*tK<%)Xk7YbZhW-7T|7`7Y
zsgrK!YeOyF$l`PylLx-u$^9`rr_-r=y?CV$*Fsvo%3Q@4rVGM|zh8=@h}>v+qz%9p
z)t*Nxug^i~0QZkyV?RSmv__RADXGMqe@)Zx@V*`K3G%KzG)#ua*JhbHU)B$H(!%!=
z%1%LU_MRAtLSHgEY`9bKx*pw5Lj1XB>ULir4Bbt0xLW%L=REDs*+qn_tBcYA){>)K
z8jr@|FLT$;4d`C@%o$F@h06YfZfkm^iU1w3o!2s5b@-GvX2Y0$;K(w=2iT6_Inj?)
zy&k@ZB?rHTiTfT|Ouzj)(Gp#i`a3|lR3_`6e?70cc=JPBxj@CO`6=ity0rJNcgTMF
zIqkJ!)R5P8UC8Z<=tm}RexLd#XX5q|{u4dGi%}fD_1+VEB!R<+!~O6F1>P!TxldUl
zxkk4HY?PGUaMM@1oLHOcya{wV$+J64x5TZgLrSy)ROi?YF5ZFFlt7^w%_tV_>go@k
zRmaG*mKLRej?X^G2og(onRr5L(rx`7Nhr&`^|ZcQq$yV`q7ekaYa0x*b*8<zGkX!_
zyr0O{w(2JS?g%O!!{jPbHDiL_e*@mVP!a8xQ-H(Bo_eTu7@meecYS5~9ZTE1p283o
zKznb3gSa9kTIt+n^HPvo^<UM4F=F`~eb(k4<$3Q%J>HFkJ{F=@5Qtk|y7w$%JPT84
z^t|;76V{L+KAW170VVgHM!j9?fC;UTLnfoudh*Z>5+I}XCpTzOy?*pdaQ?U8Zwiab
zO5M4i)uUQ74W-xfT??PS(UgnYxTp8qa2KHp5$H*5JNqr8%2+9Y38*v=e)N3RY&$3P
z!<?I1NYfN6EsUfZYwN|3UGX&W#YQtP-eX@wWU=%p0jun1xb0hJ-qAJ*i}4v+nLkY<
z%vcF*dna+l4sE_UiLp2c5@Tor+^o<>e=q#54gNIcw-eg(R=rP9NMy$>--wO4xTsz`
z`nu3GIgwqA8j>Oxa&_pP%}d<KiJR9!>nFp>|1sq<B4;U7v&yydHrRF2xz(ne<I*%~
zY~RDLd@Np=oC*Tg+I|uH`}934_@Fv^tiIXN$Xur=i~a|(^q?ZEQS9^9nOf}H1e%^s
zm1+^=NNQPMlET!-B!Q?VYmxG*3iDf}A@~H#9-XW)msZL3ZTrmCx)2E=I*CEH8HVp>
z*dsU3Z7J7DxBis1fUS(RLFbJyrCLPv2o{^C{pUYU_>f8!1u>gxZ5ny;?;Xi8{d`HJ
z5b81h3`vw<PA{3?b4q>AZe!`24?X#`s8@&zGUiy5&yZR*mL=-~HkdX)d=fJEo{S)=
zDi8`nDcwQPGw55cqm<xpT1YKdy@s%cDlYURQFf{=m8#;Ok0*`N-b4US5FU3pwMrZZ
z%$7(%f^4%}+ak-`jG;c4EDHa$RLj4Ho2;v;*1{ao!v@^c|EOu%MHQlF38Bd^iHFvC
zOhyYw%LL)eA&qGwl*T)Myjgqi7BWvfm51@jT#7%LpYV6OBtA);b5{`bZESRdzSlw(
zdFf`vBRyH0rg-yP*^!O*c%jme+H0+?rhdaumTbO^^_aQ*+tTS_<YG?5pyFc%3aT3w
z6uTFyusMry)PzFy_&>#*q%65GqJndT>?UsCWkF|f@j8Z;Q4P#vWy^MyU)GNbm#CyK
z74b_bqkWR1Kwp~%t(ZsZWgB%Qw3`*&P)|raLdRYV0;tB0UPe5%Yh6Ycm5GZR^t)3~
z?1~BUCv9XQVO*%0mgd9tAh1#ZeYK5c3Uxr;weoPV{(*T&d~#L%g-U1FDhm0jH4;2~
z_4NwpibeTSGGDlkn<7aB#D>nwtm4gFBqG~gnqmnLu}|eZ%gdBnk12s0c~LR~9Lj3~
zK*D+X<Nu7q>_{P&>EUpanw(E<W_YV_oR$y~S?9L%O<=X|PKTH5o|!n{ETf}9H`z^q
zu=`%^ZrVEK($x`LM}BKpXOvIxk@g0EXu_sv8lJN%FSMhwoU$Lq(%zqAK6gK-Ajjv>
zlZ7SBkumL#b_`AgI<XYtn9F*v9OiPu+ZG4|1r}C#qhwV4%0(FEwqhJ9o4U(?%a(7-
zHq|<rO#vJGl5pggO3mptGJ^^sj1+y300WT%g`tx%Xk<!O<4>%tUo_`DHsqkSD><cZ
zHAFB(f_D@q$tTr;E*KpR&z#akUpLS<vEuq_m?W25iS@=x=Snv5Q?=a+aQiKE6dq`<
zSq>j9?@$HSI5Wi3#7UxYkOvBnWx{zNO-iXrUZwE@lX*3RahaSQdF_vCJd|$F&2`Vr
z&&oP;vqY{;s4MlkpEC(vEgDcfbyQj;b2uzm>OI{V6RIv-)P{;KGdVD?9=IKb0;s8h
z->#VT4>N60`En#}RhPkVdf=prE5R-bhj*z&ZkTf|NW0lYCZIIy4}S~XRP$7}s;AUl
zbQW7)3LJb;(_hzU`bHT<#GEr{U7igGC%I^ck!zX7Y1rUUQ2~C`UH4!N8-f=4>QYkq
zm!u=iIttIpxu)}bikn|^8ih<mg@vzWo@<C&p{wshX^f=PdfTqmhRhV0-^=gGU~{z^
zG#ikHf9mjs#HLY}#F9<j>cn#-t<lQ!p``7G;C2W*y2!PEPIj}qN|>=mWy=1Nn?*;v
zor2bib+b~X+O<5&mpjxlPOHZWVDd&_(<O>J<Bm*YRam4cOCrx)gvJo*CTI*%@J_J!
zV&sC7fT|wyPN#+%YcF|~WaeV4flZdbPV!Ol>~~^4TbJzB>sqkw6tJN$>82=_!h+c6
z5!CT;>eZSBcy)d0L7E%)Du(sxKsQwj+Z)gqHUFBPgaVA!F%+k{iO`=4^hrCj%_kjk
z$DR6GV1nkD+ruKZ7Mu`|K39cdap0*t#$b`w!owc7Cxw1c$IR(>_g0*(3tEfqW3kg-
z$POf|ANRD#s8?ZTimDAq9h41Ll81L2EtD&&2=w}VG#Ju1VardvY<wy(aCG2LHMlLk
z&ktkPSKhw-)Ow}%2p-hWBX_o=$0#0LBQ5US>|c74-K=qwJ_i$RRqIa~_AS#zK-gxx
zgpwFqdtC=5`{h@{V!`4L^RofzT$MH>Y*NedJcfB(#XPf0Tn@3;z~rEX-xv?x_J<a2
zk_NJFit7(>U$l!x><#^PVDqu8Mnf%<N{lLAWYU@n4&tK*IoFh>m=m_14-gNM{waaZ
ze3}skS~*^jZliFg@!`CJ$*iKy2i_YKJrv(ez*@S0?<}x`&O{EN4}YmP#G&*y;7pQg
z4nm|8XFgCA4BblCq%!6|ONh(P5rW2tum9qMGKNUX4y-)|+m3C5Km}Lt&)kqdzKyio
zCM0ZC>pkI{^z&8*qje4gL(0ov%Gt2|s_ZGDbiJ_rV_A)Cu<TY{NIzd1+SC&W-7=ZK
zSEU+VLLO#L&g;XmC(H9YpZeHp5y%|TW3I%@Wyrzoccgkoqp1t1Ox%9(7f6U!C=6GY
zk~Em_r#U7+7?Slh9h^qP$DA1=5x6ksI91uypJF|WN7NO{dwi@d1-Fc7h7Uh8D#@NM
z402#RNr7T)A2~0Zo3bcT%TX<Rm;UKOo|?4DfMWD~QktDz1tgVX5qDKT)`>V!;R)UG
z)T}9e4Z{i7z;Dt&3imDQZ?6nfY$Yjm(-~2KCUl7&B6G@=^O#qDKxgQ$3)dbhFG4JB
zzptQ2HwOKVG0I^m3uSYR=ftF<a4}=`RazrNl3Zw+*GR5G4bsz2;(ynh{W+t%l!@2T
z&#2)qe}qo5K<HkNTwy2<ot7sYE$E@G<MP!+A%sT;iur`5CAj3%1|eL5(m)to9Px4#
zCnJPxzWaou<@l{USiI-`gA0tUNss#t)p-g5se8YPG*bCeuAk<#ivhQl*$dZgA-62a
zO?yA31uO(B;i51{eNwp597-lrG#!EY5+2%VX3svod+4S*g}2{+&NJ{HMsowFFzkeZ
zfreFFc^ySwsj=eaJ_yUPyJRVMJ;X8a`A!?Q!pYwO4ewI|kh=745F$V|QfhRyvh^%P
ziMy)Zha-gggGR$eQdq^nTmT}lKACN)US+8B(3-<lk*Lt)yAx5g687L91k3G4<v$3R
zkiJzJJra3MqYsCzzd7lqoQ`<9erS-^U#Kw(F4k4(#X;2Wem9(O7xXBj<J~ES`y`lP
zHJqAUDIRi^`ymrC7T=^}!-K6K^F;y75vzJD<oTSc>?GAR1V0?*;_LP-^43Z{S-`w$
z(RY*0#^gcvQ_Cdi((Ne~YhDbh!?C23?q;XYM2p3WPq{qDuOk3NPk+pwqIjz+2CGck
zk~sLPy(Z;Dyw)=eh(!|{%FEt^Ffa}-MUTzB1&bctM?M6qp{fQ<04z}|y~V7Bj`>T4
z3z=PssUDKE@q|lDYE&YMmpEE8imCK|ys0|rGLYuHm8zm(kpJgQEio%o8qeFlbIPGo
zz4@~g3Gh}x7|joJl_JZD@j~pN0khT5E6!q0fr|tV?k>$|a2tk?8L$5Dc<mltS9aIa
zw??mgD*@vkZ(m2skhum=$WDuWp=$rd2##T$$QoE$fs`jcf;#7ciqF<1_3|9`O(2P<
zql3K=r!KqVYF62|ce45xRwYZ?hTFhm3%Tn`a8&t1^%a~@|NV=2o={p`53t*JSq<=~
zL!&~v(WI5}OU^~S$zEas@sMB}gS2RCz`zKCiyYsPH)6Ya{qJ-;_nn!;`sMn=mlXeA
zfgYfxsLB#Tp->%1Bmf)^NjDb~k7-PxH2+fNYC8S%AnUtW!5j}@pkruxN26V__#3hU
z(C1*!U@%gz_8?lNj{Q7R@LcNa82mZ^*%zu5Wy4XT#N*4AV&g=F$C;ZTiyR6ScMIP*
zbHijh@u<wIDl;syfRW5UC)aRz{HH+c%&Ou%MX7u}1I$SGI4(oAExI6-dtRxo<sRs8
zRq@24|2w{b_Uxpk0E~bg-A|^gB#^9zkhk0TUG+*)P@zY6(_HYzTa^443<F(ai0x7C
z%3oTQbM+2f7du<{%K7K53?4+bDU%h~V+c_jyG`rI83uD1ST%zaLx0ti2<I%l*MqE;
z0%Rq%68FT+TFG*&47v1*+_O^p;_^+Rr4g?C2di8oaMKqSBWXgAAl>ZriUw3rN;%)D
zOep54gQgwlOteDluIseNO06oOGnW5I3oHsqD$H>P2hF52XN!S@oZu&KIRpKn)XF8t
z*$)Q7q!6;NB(DX@8p$cOrZ451k&Hqg$Tc|GxQNXUP_@o!hE69%bp8x0u|2+K+FwO&
zSKc6gk+mDI;NmOjc`7ObMYlDWcrr@SD?M+^$*bunCC}*<y!_fY&T`GPX!^qQjDZ>h
zTzBCXd<cy?rnBr3FrkX}YZ^%LoFJ;BjQEnw29HtylOIh<N#K*B-=@IqG2b8as{%;5
zxxsmjz3^QU4NU+WafW~%3tjGZodmG7#`Z?Z;c$vbIA%fjZnHcgTK)o>Gpi8K`^=oR
z3v?X{;$cN+EWQ9wC&Z(Cx9dGF`NZTl3sDd`!BX5dR|?71uDwC46m>rJT5p`~`i>R#
z*%_%~;ajX5%t4x)q%z0?B%%f@{h#o4V)!FZDPwOo_0NC}S-W*gZzu<`W%{*LIw~!Z
z*@}XBM8`m@c|ZD`4Gr{6LlS#JRLw)i;HTHFs)HjRH^P><BPVaRYi{djFt4%(|K=@C
zY&J5bYp`OkkXmE7dqy*dDQ=ezXaeDZ$s1#(@a2ECCXwT0U|x7+|KC~w#*R6^5=Df&
z5JmRQGXjxQ1$ov7B{KOnU)YxAV#oyz^WQ!0cQE#({w&zfXhhk+Y-ZESnFd5HuudTY
z8=EXNbpzMZSH$heeeOp2q_9X0KQ<qzj<nIO@=C2M>$2PEG{kyzHA)Hkivt*Ens*z}
z(mE$B=Z4)Y+$?_f=sO5yg5|fmXH0X_XYZ{;R@rZRn!hi7$~#SWs7;)cX6nL7r9a72
zPd4`z1M)Vu=9nC^7JgBd`UcklU2DOyQxM54iSTuq(DFGvbXg@OEmXACVO76W>k?7a
zvue$ceu}W2Y#n_~s>thKQnNr{6`9@Hc^-YW@bnjEKrvm8`36Zq0@cg;J+0H-0+hEU
z)|a+76-g*Z2nM^uJa75Sv^u*3S<;33aXbb`&t%Zq)Uu*TfwYo7i1H*tKjmfj3ROo^
z;t2+e=;1xw`LkrKsd{#iIb4;;kcHx>aLz3kJmrjEN3bhCq<D^m$Xue`9BbWMcWf!m
zw|#NmsLDxN+C=NdOIF{yx3#y7xFfmBNczh|(oKolZ0oyR2GDmQd~n(D0&7XlV2MOW
z_bu)Z?rq1Cc*G2V0$d5lMq7QZ!~#I)0hT-5=2zKcq_Y)6NS>~3@AuZlL9MRo{zDxe
zj;Z-(m$#1@jUGIx!Bjf##!iva6MA?D28se1IsUn%TE2woV2m$a*AHvo)_=M&uYBt>
zWKGQ*I@mm|myRlABlQ2iY%`w>UzH;?85*-W%0i|59&_G&YAb`Fd}uN6T>NFUrLVgH
zC8cAhbf7GUyKQ9b_lRL1Vw3kCmEf6AzxslT7R|v^7Pueew0Qg}<(a;`T$~{CX7;){
z2hAF=RXreOpcl$T=b2S|%nx)mVm(;J-a8hxW?ZfPi@;hLQ%4ow<(s+q8Vvgv%gyVu
zA*mvA%&%96g6+@NqsLT6$pTQxyHo1CqCUI=GP;UbrL6QjyK;!8Fo{Nr)_}t=-|MoE
zXA<rG>#_CYyn*3eF$P;@6#c3-A^!Beoy+j%^$Er*-4t*Pvjdb8d!vt$mLO!R-+gr#
zb{&~B*x7zbu)bV0!U>949IbtoSl72(&@vp$?wQ~_qga;YeAeHz64)fPE8DRbT*-1@
zrQRsIZt5E}TYLvZwoZCqZrD(~uRL^eBtxR%P=9szbG2t-Pl{G3ajavsYI^QZ2IyGP
zORv9=E^C?jE8yhWnq~F+*tSztnZZYX`TV*PSyE_tOiep_iJtf>)91nmS<~9?`4E(i
z+=sS!qR_kVb{;ZtR)(F!zod?c(632LZ1&b%BkY?1*T5XGq^|0_-EfD`5tC$07O+Z9
zD}>v91Rs+XWx81tS()zDuG#fRP1DwQLc1aR%#XpOIe%R*^h{R7p;N_UxaOqcjp|N*
zK-r_^v8E5^iUqd4cg^3$Cd6(@;HKG=jLd3YGoN^rb6D%py6YVx-_IOjp~TfhAH9O7
z1Pwt)=oZ)6<)?-~XPOW_e0ui>Zc~8H{lkc-#E785MP^f`hHNwp$Jrc1hiCOf>bJ>o
zjIBZZ%`J^Uk2VN_Cn#s(Y8PkmhV)Z`eL8pa`7jrW!glSBYXD8sHDIwX#nxhrXaVw|
zAGQRk#U<~lUr@0r)VW=|{q=QO%>C#%^mggj-V=Q}W0x>it<{l1^igO_oQ#Kvf~|lP
zA{=WgPCb#N=txy*u}u5ey@5?|0*}}RgMK#u@4lxfu~jSKSGggnPuXM9b8ntrUt*5R
zcEUa6-ABH5$XiJ2;KAF5)|A<8|BQ`d0cQwbt0Wb_oCoiP(Bw@tm$KKel5q{8UCvu)
zW75nqos*zri|?uuhm-f}Li4r=q<d@Wxc+|Uv9;iq$EC>vw~UYj7kvwk(Ml2|4bdHv
zoW~S7<tJ&eAJMo0a=(;Lcnm^B0zsmWa4e)V^1g^_Q;$gk!O@(~rqxMWBiVZdmmfIU
zEbyEk7SPYOp-QAVKd(=7TokMS<F@*Hr&cJiBbZZYTD8(?TJ^qcmxX#nPn<fz^=iUa
zD-;ZV{S4GDcB1ro);|lF&(!p8cLbbJj!sn9vDg-E!f%cvdis@lujh-T^y0c$gyi>)
zM2HM<v~N{y(y?R+&oQjO;$_9Y<JC9`*<awU^9l8S?1|m?$X2%Hir`n$YDG^dGr+W2
zI=#erz#XoVq456Y;QZ-U1t3PTp;wKZEOH)N8JRKZk%>fMIjS?aGnM|L13w)(OyKbq
ztvS+6b{~_Pjbm3r6uMEtEB!R4%pU_rqDaQzm6Knj6ehmEp?hnvQivJat&d(C=d}^m
zjLvY4Y1CLDrE;<iY4FJSG%#pZ2zUdlG<gE4ifSd+gbgyTlEfrDX8BpvOj<%_ApOH3
z^lVx<%Ad}idF0D(6Bv2z)?~7xo2+Z1cNJ2eA@D~)q_(e04AYigXNK*uBf`Wlm-5nI
zq16UrhQ@TY9Wot8m7B3%=DV0OKfhQX+a$lhn`|WUW1GD8@mYUI)EOqwF*XXevM9>1
z;<mk=OjXFg%}$XOug_psH@T5kiksT9Ia$iq!m8&eyvRFkp4zmKk9S0A^GTN4;5gsq
z$L3m=jmgVBcv-nThFd-4JLt@S5sltO$+6jmTQ#@B5d_&WE8!7c`5DS&StkHwJx3r-
zP8k7x>~VaXi!-)Pkh&D6<icl)MGmy+n==OeU$>*$;ln3in-PwYG7As5A*9Oa%Kp-A
zU%yWBZHrpJFlQE8M`qiaRJu%F4--Yvv9pDy=xqgGTx1H4O^dvACY;(eh}n(hgyhuv
zw8y)f1iD%gr|<#qm}7?iN|V3(NK0GS%VzbU|IR;Q^dOhE1Ri)+`SE~Zc$phQK%zRc
zw@-&nU4fMcspx5Qqrq2D^lVUbswFy=gUssB?&;L+H7!@JSu646a#^;)QPU}}g2*gs
zlcIS(-AF{mzMcw=UUp9T5tegZku|dU8||($mbH(+ZStMssu$<uDLEB3y5L4=VoOA=
z2vZaRM55b;=wTp9`RFXGN76zolsX3RSZH|-Sss^?J+;E@5~<p0kXk405v;Ji_n@<M
zh;VUTrk9(YeBRY#hZSCS1vMGij;*h5D*_f;`WZ<vH@6vS-wRV_&6l#W2=ZM&Ti>#a
zc-NM>Mm1xK+A|72SUuMXc>PtHPphrhN}~<E%)C8wfK{S*?8s^HJU6c#PoMnb@gnHE
z@u4AMCxNct8)!V)O+t#c>9hJt^z0IZ{@1!mhz6xGNy?uw9Q`o--t&{o!P>o@X?u!@
z>F#;~@I<F_4hat))AnM~fd4(c&kYsqJVNFMJQZsjqI+HSmYo{tkN1?q55La#rM_&t
zpV-fT--)#!4e4s@YoSK1o(B<o4?AsS!axc9u;K*~c$YQH{#}a7A6ldg#qG*2l5hB8
zLvH(&)wtT(_oxrR1)H8U*2@{AoF82~YJ%0jEXTj(FFP*3&W66&B8InbcGQbxZ)M2?
z%MG8&^{*2S-e*D4ypB(B#=@0d4+CT;ktGg4j}^VIaz!kZ!Dmv+h?$aW<ZW^Vdk2h=
zp%w-=IP|eG`kU~S7C&qGPeW!|95i&`b@V9(M4-<WcgTz=a2$lIe_160Rsu&!7A6X$
zIb~V-t(Y@V#C7M?MYI^&q|{BV=yhA5M>w9ru7&mbhd3&G5;_Sm`1ovL4&Ta!^st0>
zG&W{KbO%vT$Wk4}M{x@5k>VZkou|6s=Um^!v}d1awp>(G1#WiORd17nOSnYAhD(34
zS8yHYk8+`u%59>^hIzQbb{xR&vI(5UbHScHvW*KDPQo9xl)EWjeX)kxvWZFj=o1EV
zR@PiwdOOHnL`DBF`TfUn(E>a9*X>OC<j?Sbaee=i1Aym<sW0XBy0XOpUoyj%-Bpr~
zdG(Ga>7M!Nzm1{CCVE^yAXNVYN*`_^_@G5VSOdP6PrhTN3))m}D+)+y&c}&)*iT8A
zffPDjd60qNT$yH?h-G|*ztTfNhQJ~<KWIXz*u4E&0y%^zhe2&QbX0?<T~RJNu@*LX
zFL~hwt<BK4`3?7;^H-Z<LT60qF~c+}#c%1)4u>Q3s%Dao|0Sj0HzQ;&wPWA;P9p*P
z9Dly|_5M{81ANIst#r~eA~bf7u+R&p>UxxX9*3_n(-Zi-`4b8Ql{*ipB3F?82U_DA
z&U%BB#QWQ0)tf~{xV)VeFy$aGi^kV1<2LJ5v8{OMb>vie<JpUx$ofiv(peYtAqux-
z<5n;AvfCdC-We!FO8zctm9E%Pw)EF81^)r1t`7!(bA-4#hbA{s%1O!OCz}+kbdzU2
z@-D^o_kYRDJu1v`*{Mz{Ow5{ME&rE~{=9-|ob#LIC8~wRXB(!v_kbz5=OuG&JFHrF
z3(9Hs7H7=11j(()FFJFPcRe21Usa54fs8;lhybY&$Zbe=u;c5l!WqzP0?+CGhv6Ic
z1uVx1+6An_-8V^>7T~+`dbmn`AUcYdae|@Gnk3=d>p7;=ZBmOn+2ZVCw;mD^ZWV~w
zo^m>UbjsMYfCO}h{DJd!^qxtaub^v6rD#_6y9L>YgLLD*gcu2%QweJyR{=oflPXhR
zRzk5&^g5z4^N;P6+wm?J%z^)XA|cA2=<Odzn*;C)mF56p5Q>M9tbO^X6a`OO=o9pg
zsda_P7m78}IBmmZY+oM?t!&I8Oa3@@!h(T^9p2u5htkPQ^{w-S@9C818t+;y29qln
zdTs)HqTrpS@nIN-4Bv-Z5As<iQx~5xot>HWnhUGN(dxtO|A!@4rTp>i=hlrw`dYS(
z<1*QFZ?ZwZ60!&@CoO`p`BL`<fJr@c0N>fU`2c%^7eX~cc0>woKK{dho7W`&QD1Kh
zWhgrxwZz>NMKT5F3O<gWt&>#|{x7s&CKF+Z`62N0FhaxUWl*Kl4GQZ~QvtEv(fdsj
zJ~%k=EGZYR^<mt9Z%QWW^B8=M$)hd32<Y7p_lqoSf27C7(k5-$ESw<gFR-DGU4N=J
z+Yz%@^&?+J;I#Dp(O*u){K;HKmXl$X>-rLY%$178L6QIZ93<uchot}SJO3Yt4fGoZ
zFBH>9CQ^}$`ESeral8DVg#CMo-)dA)gNM?DAEWSp4Icjoo&Vo!eqMsQ2Uq}-3?l!(
z#DCO3;-BZa^zy&I``>%{Kdx!=ufZD3!khm(1^#cl!~g3;kmsKfVg9tZ_kTS@{~y;3
z{MTTwbVEXcmeG@5Dz~q|OW~&V(q;RQ$>$4Jk}9yp7}<}Iw`=N~XUh&iaoZY#o?#ly
zYr98FUK#TDbDWBY8S{s;G|j_uwA^m{hldN3iE}GsV~3(qab%oouMd_AIw)F`Zz2_X
z?6@x9kNeuglEu(B{SP$^ED7IcVQ{<ZK035lpgP;oby>LhhW#H-1djH%Td#-5F4I(g
z@XpqF^R_RN7pgZQub7|RWdlP{0YV^!YmyyVfUG?Fi5qAb<Ixit;tY2+(-U|;8}NCx
zLHQ`R6T2XNLQvG8!K{r<RPZA$n+T$7Q$+~SLw^nt_Gh@vzn=f042%DY@V*)WO7zvY
z-5S%otZY8GU3e+xuJ;Zp4QZcNWSp7-EH;9?2&8oQKt9{FF1o!f#Ucf#8R&H{tm^=L
z1m=oG8jZ6M#WJjB(BB#L=<ioo>xF!2kZt@%;Hx!WmxZBkoW;&Ca=yLK<91o>4epKz
zMqBFa#jF`+d068p^|bHdzlY&yV*P81ePP}Sf)e_;$XVw%G{dV1`(?iJWT=UC9<U!8
zl{p%0l)R@BOxgP<EPO{SNpfgcP7Hyniy6A!U%f)gaQ>seXoP7LzJ{u!gHJ!FOn@7k
zy;J2%c9>-AcbLK=$9jg?HrJQ6U;ZL?c5z|0WqQD?|7}Eo15@X?;fQqL-DQW1SKP(0
z-~E&#?6gDkeUDvzl62J&B4s?Wp<43F4JRPrBf4~OxnhQGEUTd)?{`ZyFrBm6*zj3i
z$q1iU7~O@-?QZkPgWSag7R^@F^v8b&kWKK*BGS<n+1-%}{kk5y(>c8Y*)<nSw=>ji
zv8imkI<tAVrANDfN3=TW^NyyFjvUt)Tl&?!>rOzptKsv+M(4Cs@Sm`(?%<K`m7=3m
zxp@v4wL^TOtQ)AbO@{SP^8>!=YUE5i3wwW=%I!lXY+lbuUarYTI;M2G9rfNyhvyaN
zojJ|Z&Y}w4ykoc{jcnZ_TpI2A)!S#(BeK}kX&+yS0?{<8(-|rbxjTG=4z;N0cBlW?
z`w}}{1YfR&w=qo$nl|WI`Fodq>o7bYJ@7LYY`Pe;6&o(IWqHJ%`2@-jDz^2?s9~s4
zKm(ZSJNtB?_#{IrxEB9AzDwE920bIsZMUn+#QBebU7wQR+0nxPS_BMMu1$?dE$-b!
zl>3lKgpQ`zq|od~niWdeqMb4;6`72p9Ut%vhM;}=l{N5<0+5D%OXF?XvaQe==rBo|
z{sk|fLTOt0dHzM|d6jk<vF-0uJXG`gD`pjW);xOpHQ`BQc9TjR<YoUE>aD3@e@8)b
z0SG_czkcT9eQ;qzpWDnRU)?OZmG^G>{miG@S4f4dIz}I&w+dmwK90x_BsY?#OO`F}
z!}J2v^lk4ClwQcN`8oxCZh;;Wb335%-T-z*_>o9Q2!AKf6G!srk5lUn&l@Hw1h2io
zADjQVnG73`9jZ>ce-j;YB8=ZOg(OyWrn>TZEJ=M9e8yRveys3SZ9x1R*H#$c*KNin
zj~Xo7Xpn@DiK+4-pE0>)*DFn8hdx~An+c@?vh(#qI!xh;Dq9z2-NSy|Wxp27bfLFs
zhc>r?&|$8WbQD_zu~(!0<F|@^kLi=+_QqJE7@ofMlovjni;BSf9BF4B3*?(th%~<I
zvqx5nT2tn)I)g%|%PmA}>#!6;a>MhBzgxKY)6K;SL4IU#XI`ecYYjb5S1ErwGsNUi
z>SzipvS3=cIdS0hv4$AEZphWM01d-ttl)J3vz<nWI63h{Mc|=6_Pn}=_?~Fuvf7Mi
zY^HvTmoe-c-U}aK-JQ7Wc^U=Z*MQ9RrvBp3tH9!)bWlCJ#qRTE?^5*_3!8tG|Dt&j
zO|#e<a0>r$lQtx(0d*7h-Om%`OB(jVZ(YS$wi%MX@BfkM_)neN?5Az=Qkt^WP-{d@
zLBj{a^qOR<sj3!tdX(-v`P8gul7Ca)z~+K^@_ete04L}HaUoTmMZIxT**}r(HUjxJ
ztAb0UCCFNAx9-<wVqCOz+OXyBNoIB=(2gVCU`)VgRKL430%X$(zkNLm^u5^O)TOSr
zQX*^d*>b{<bDT#Pcf0YekZ)8qI9?}nAaMEEumw4+Ga$Xml$pL*D9#2ir<qsnBezi9
zPt};Ej6Wlwko{^~U-4tGLH%a*3!*3+(<eV(eq1THrON>S_Cc)E;#X5(=kxE<%TY={
z`p8nZK1TL;!BX+_S+H-<XoD72bjKyRo9$8wAK{TCyw`7LR7!-uA!=NJ@;Zi2gsU&&
zh@St1iJb@jJ%9Y$cSDd*&L<mvt5I0v>H(Cnx{>t#>5?h-fl^ENdp%=R6IhSZzxynK
zwAuqr(qrOBt@8PTYG$pAP-ItKaoP##8P_N7(v5{(5K*CHGTq8hw$o!HU+?|3qsnn9
zwsnl}5BZBmvZf<nR<iJyBg75`A$0Z6IBsspZwAgM*7zbf;p#fx&!nop97hjRDLs5D
z{!e+GB<7Qn?FV;=t9vk4mDmbBRAjDCc@ECYth=PtS9h;x*q$jczN{zkaTH&+Asabg
zuJ$FC!o7sTd0vkxntNW+-{`*+F+_^Wj1`l87I2_aNt#X~j>)#xGIZW$A3oqMIYiHr
zHwMMW0!hCmJcm}Lc)kL$&*UA7!C+IuMN_@P!;s68mqkmjD#Ye(ekX~#TGvK<GZ)=y
zUSN4AV3{B8hK(XN3Bf!-E7ls|54&OD6~gNkE#c7Kw(QgKs9h>rFPRkV>G2keC!v?O
z!Z@LTeN^05mK-@+PnWrq!{<p_uEC;a&zp=ZI6%VVU0*qFGDBJgZWtIxAMZ24z9vfU
zCSi-_mLCZ}N~|QK39s1rW)P@#wH&^1wdik=;D%|`_4aL=hK_u~Y1gDO<Jj)19nD+_
zx&V~cyPTl+`Tg{ZN1Qn-fdlGoF(?<v0)PG>JZ}L_yA*&81g%;|Zgel&(bGrDM4gMe
z&y|T1NHrnzAh3KQ6G(Jh+npF-EETe6afNmZYcVIs=kGr63itfF{y3jN$;HPBuU*s&
zlM_ZSTNRArrISsN=qQVz>W8(8Qu0vXHPo!?l)jv$u$X(Dn#9LHRvypa%x2aIm}wD}
z(@7ZX*#p@OTHAg3l=10kFLkiw@$OE8wz~2WcWO0vgXWZJ=#D{lm2Q`2v%Pl8Tv&h8
zEK6~)ouGV!ldf>7BWDh{VGoqsK>fEVf|rJUbBCkpi&@`q9$9%`YVAmtW;Tv`7JZfg
z^Vf43`1-v*=+~;>kDlxJ$G>|~BeD|h6GAp*U_fGJj#s;?;l~MmBP5a4mr2JY6JWdQ
z-`e_(S$l(0Vf)I1%qO2SyD_;R(<~IR76{Yvv&?4N-f?>OPizDiHIkH=$thyJ;LkG4
z&W-qC>y(dxr;10wRowvKPlap|+=x6p9SyEE#^fkjK|xvy71`ehjhF55cbcCXJ(v@-
z1o};%Dwb*Kc5ab&dseohvzkLIwf_<D!lUZvCkGdtcy!b&6Jc<^kVVYf^piML%MfvD
zHlu{*pw$g!>Q8f#+1Dy)L2*=(U6dm~8~;O>|JlS$u8Af|mqgRdKG@qFvUJ_p?h4Y}
zVA$K-al-z-A>f9a!Rt_?<k1|FZK+cI4v4nN>w|5sQ#&_$Uf0FWeY}_8?EOE0_a9uH
z8(5DQ?vmsnv(7ZAvyQ|ln{VjlABEq!dy(XW9drrQaI<dPi;_0rSTEq>$=Bchoucfw
z%bgwL(wuq{A^Sz)Ml5gfBUkRK)rK_^p9_R}o-bnA7jswLCi@(e5S-0*EbF1BPa1vV
z!{;@r5Ohzs>BSkBlrXNiBPeK^95PRb<6X@r{<u%VF_``z;{WOn*@`vYbI7g8da)mB
zV{nr*P`ht)ULikt{C(w~woP*EBDH!L;y@Msvp%-*xweXR(3cl)D#LyjCcXZfijl6^
zOMa=Lv!E3obMsNen$2n?ps@XJD`R-0!~=7`%Qw6>#p?B&-C_#~ZWz8>a<iL08f^EY
z8#!o`=Xsq!?V|>3HGc_5wO%xZYibOV$DA?FpiZKgcE0k0&4F;X{Y(P#SYN(K#Z%tw
zm7%qxa+=!Ad7H1q+l2CE%vf@TtLI}D@s32Et{5xN!ooQad-)$IPOx)^`WBVuQ~Y+`
z3(p&I_@^rOMm&89@}i%`tHk7J1bptWE@~y7(inhm)D_!^uy^cG8uhe!VpTEpn_2}D
zAeYZoi`?jwkU0%{O#tl+Q^cBMScnu*=y@rov;VoTMP{b<I||{LHHI`UP;w!xYt>6N
zUC!0#Ejt-kyW(dykBDYpxHdGh+(Pu-p0H8Di#5A>9(S#nSYaOfR|T}YzLYEu=Ana(
z9#2F?#HP1P_;kFW0~5LUamq{ic80zlbZK}7^`VmRs_sD&ww3W<>iV=}L!d(6BB}_a
z;*UBS-%mb|NOhhiK3IpBImho^jQ_M4F@?o3yDV~_bMqiXa}a_QzICw@>nH<=ed<k}
ziQ7%(84il!snJMylOD34WaY@1L1;MHmcPB|9E3)Y_|r?Bk1k)8NBF21j65V9B(sbl
zgP2q#qJpdw9redBb|L*ZcxrJ+i^T4@Bp~e<GOy*<Xr@vdZpdJJcnax4cex;^%j?#S
z_l#>%q>nJ{LV{)*QM)x>gxrHVfPJ)TcFr=sb0Ouo57)uj{mAcSHKK6)>eDSlGn<_I
zj>{jcTu^fW1BtsIgXJMH-ZO)e&V5tfI-g*RZic;U3s3DYhrw~)$?Md9S5YcS_bx|a
z%Z*)Tt_RECbc@~UX9=NubkU(i;HSH}!EpG5>>f!y7WWBPq>vm9<<S?%*J-D5Z8vj_
z^a*EQ5S~K6+%r}DX;EHhr^TdPduf#HjA+_bgjkqP!10q!bgeyv&&{Bh16xeTklP#m
z1ZXfKO-!#_P5ch;szd+wj^Z;}frRc}{<13_s(17p1T;Q&136P)fdr2uKL0S%N<kJX
z3acdas7)SBbb~P8xr1B@EXKUH1%qf{Ig#JHv{>MZ)rAnsA{HOF9M|Jz5(+Wjo~Lcj
z5~g2QI}(NHmCf*{VS1TGMePCukjItYq0mR-JA5yhjmoYtf$k|6q5YQWZruTP#wl5v
z7mt1^NZBYiVQ+DE`k41oIRPiD(7Z*Rxl_JRnWud^y?cDf+y>(L)i5;2dJL867SavJ
zTdwB=igg)@Ru_AqHwm*qG+A{|gvul|E3!W4sW<Nyx@jH1P;NOd!?jh>0<D}mX-{|X
zgI5jjgoE$9+7V0nOA8S9HnF(-+ct*zxD%!Jo`gH@<T7oflWUcm-VT~&_FH`C>?!*9
z7i1v#NKn8TI(<}TYPTI!HSS82y?w!vR32?uwM3?)j}uZrmZv!dBzl#z{d}+-TfIPb
z<c`q46Bbc$^jCmTf7o)Hnj75u%-G2wCGX25um(nJA+D``z-#N!O{o~+3JM<Ip!qs?
zp(pkdVA!>aN08-1O3w9AlwdoX;ozLSw_#6A^)<)iu;3F6-T|sk+4o+Tp?D{78Q`Oi
z^2$lMn9iLh;a?zZE>s3eMeBCY1-)7YDJ$Ec{$#1Hq6Mj>@!<d|JFNOX^~XnFycYC$
z{CE3jRkk(q`N$;6YYwFiDBlUVi8zPmw*5z!L9@T#$E==vvE~DGU=-V^VCzu-^)%5C
zXOJPZr%zktqCPpAKinLB9Mi02am1-VP8Y=8Q(KuF)P5cD!WGHt`q0LQHqJdPeb%e$
z{dLK$t(ZvL4Tz{0Bi~`EoYg-VNt91iMVw*|SAQZrUJ&wPj*r4fCAc;o*--@GmVQ1)
z)3D>jo*F~AE(r*bD3aYb<E#J@1L24cqMEoJ_q;UqRsWPJ^A1|6_UBc+X?NOa24&}{
zi=C1n)vGanV)07vazx3To#Jy{{;+^Ik4T4>RL^3UV{u)qw0JD<2Nv#xv0o7S%G<Au
zu=m2qeWphHp3V(Yc?DerI4a)GoAO6UW-vwi8jvnCCd=;9yB9QT@>X8=TI~;*K6KxN
z9F1@?3lZ<7VXvUjq1c5-9I^Y*nVFw04cox)WYCvvmZOU>%vGKCJL%StU~!eHbG@Lr
zbk3(3`*1-}ybgja<5F;}z;cDVfg#|rG|#LZ|6N5cHc3IhYH7O6;z}0ytmaI6`jrVr
zm8q{heLrfgzlfCo?HA4nJfIi)z>}Td#XcqsCuopAMR^fs!EHHZg|vxsYEIz3O8XFI
z20pw{npETQRxBtJy8l$%i!5S8@;!7Y35<2$#_36`ine<!hW@vqxqBROrIB|~?YUrZ
zEiv03t_<g^t%pWHKE$To!NN#R<kr4NRK{$%+Kr?nT_-2~9cgc@K-Go9SY-nB^nY}*
zO#OL=PJ+A-6PhUx=At(nNl2{8h_=hO5J+L)8b}h_Msj}C@QN(w53h=XK|ZYeS$aHP
zD3`)!uS0#dN;3-Rq~+pP-7}#OP>c!wGmRTf#4)D8U(pZw7_H~SD;*$;JUR1+l7ixz
zg8)Ps-T^x`%Je4eGU}u0B2z1wPDj3gu0glJ#PW;Qf3ayVL!VddHfS4KJorEngsT7P
z?j=j(B2;jP|AYN@1W%V3V4@ZlRjJ;faXKDw$Sp&HRSePHJaOhw@r5OGt_^neY6AW2
z@!|Ti{m2lSooLgXJVxGcF9Ky2VH}wwzV|La8SZC@>rD%xb4OHsLmOZ=O^U|)^T6p?
zYyWFbfs?ADxgDfDJ|<eJ?M%0(X6LcmWIHWbjTUmEJ%=$>RM<q&#|Oja!3hXF@bKG7
zRl9uq?(@Yb*&Z*5N>NEMr?DW_x%@;nJ9t86_e=6#q`~_0R3G0N0lBN-8Cp39_aSsV
zfGKkEau*$q!?=+2J-cgcZPK8GFteU~j?I*deJ1zCFo~eGUsS`g8qVZJHfTMaPY?Z>
zF>O{_m3tx}OR9b*>3vAfBzL2LU=9OKUXH=BT2MyCW4Uks4BbYlCOIAWHMERBMmv><
zxh?<a=5<Dbwb^^OC)2qMjz*_#;sQ|T_G4;_EQ;uISH^GKU9-a?_iep3-`cP#eXWLL
zj`TdiguNlCPZ9rAm~G?FY%=xltI3pl>&IP)gs7Pm*RO)kw#?DY@R(807DwpzEOMh_
z%igkvgyG}uomBGNlBWd2aR{4kB5WIQxyeUfqsja>NQgZU`?K-5(0amUY!cq0gYkeP
z&ar$_h1b`Da)KMQ{QJ^}9k>Z|6IfJPRM#ZH!(rqI*P8kGM~BiXO<$^Flh?s9<`fnH
z368pQQJ%;oSMzO;J--4E`R2~|psQ;5bSZbNvk@yjd@k3A@t*G};S1XenxfawZ_ZS=
zci)nYs$j;wGYR*S`e@qu8g*2c69(eCM{X5A<L@tIp=7w;U=L;fx~EIHrjTuCnnJ2W
z>!4nwfpb(8zE-l0$A7!NRP=@$&Qxf(ea1#**LVj1D9AJ{<K3<V4$j`F)bpLVcXjYF
zR3DTDj1kw}U#%FCNv1sFDD8t`2{=<mb?ghw`s3VG>b%vm5WnoJ4u64m9t_#P7!3a}
z2BTMVM$F>KElC0zf?0ZL5)&--lw|v8$bB}mIZq{(BW~tV?HB078ANi>M<N68CnZ)@
zLrJ{Gie`k%)ltJ;+Xmq-JAM-D)ub)ZCXXCS^>mK*|3NHMszNx(gm24>pJv?&apfWO
zSIZ|=+G$en=Y;7R0b00)om+2fLv$1dEP~R<soiR1N{e>#aT0mD>XQ_Q=IXLntT*mM
zvIZ>*cr*adD^?0O+Cn&`#6pJFaYqo3&Hw4vx(;v9Hv{w=LZcqWm^8jO=;8yK;L^p<
z=wqq9$Zk>?vAMy&P1e@5j=DhCb7gJ?D3L&mC>kuvdgHp7bsQ(qvljM=`6ay_Kep4q
zW%0I)=>7$qY7BgBIZ;tMyLz1M@O()%J?c_Ok;5CYkY8x!%#S~2XvtQPYM1^wZ<%KK
zerfpno6!fzL~`Q=9^6D(-GWB)*1IWOO^cPoc5$P@iWM!{*x40pfGL?nmW}DE^W~XR
zCWAhWki)bN>1#o@#?IDSUTO2m(~lJSx6T(QqgOo>8VALnb(JH{rCnRRpE|lO-L>3%
ztxz^2v0CFzs9M~i*fkpNCp`O}N+CuGxA^d%>>FroD#W4WGCp<=KK`FYAB-39M%=<E
z#*XT_B8Ar$p)8xvT=siTy1DaGUo3@2{k@wC;U-v($=+n;<voZwpw>8w<Wj(xtRE~a
zYH74_9^D4!JV=yoKM^QhS3mk&AUIt5?Is)JF!0~B9-C7nk+S$(uL~)}h3?=L-4yUz
z;Q3V$YQPcjvc5N!&9lWhgXYh9cGta#`KE+50(Pvw%jC<C>U<M+wsAz%pi6<@7VB;p
zG7Gt_HeoNA!brAO+!}i1S#>!T<eThw-eU~Vk(26H9QZ0iz7dM4m>mPO48nmPHae-A
zk(-+5=b0-tI&)wT&3I}o6Q3ZvBQHihw`rn3njouK*kmk?zKH7LmtWBl4M4j5T{t1L
zpX_`DGB>?dcv<<T1ORW$IhS;cj}yDBGb9Kj$qU3#R+`rvMA@_(3yw?#vmBiY_Ip0p
z6)*h)mTO1teM9=|a~+{cX^7yZTC{G{q`8|D>FXJPy%GC`&x;jf^MizyB%gH?89tU7
z(>3e7`AP8J{(IQ%j(td(sLoUc@&ZvMk-BGF0@~-+rR`~$?CcUXpRGbomvoYpI{<O>
zs6K02^nQ94blLt9aLaqtZ;TR4;KN0qWT6d0mPUBaA|7hXNv~<=_8w{i%w;F9@<GIL
zqKiXmEpuF3Raq!-gf#?q!JSoW^J2~4Gx{FO*@UCW8=9M;<M3;PvgON;GNlL)*#w8A
zc6Rgeh0dd-qw9Q+avtlY&+CnN{}x-!l=7b2(^V5taB`5XEOEtBq=v%P5ZtzYn`y3B
zvZAO1-#V-3llSp_4EW(*qKJ%QI_+N_H>(}@REwm<8d3Q^3SS)xyL!IF!JKU;WbYh&
zagg&-pqQ)G62HK=1&q<Cv%!3$Fz`*eHHQ6Me_=E@cG;FL&&M{8(LOHBXhrI!rmC01
zG|L`C^J)J(N*4l*{pGwA0JIdL0GF{^C|deJf51W%5{)J@vTez=qdJd{%`$Y#Ni&Bv
z=&J{rQ0Ha4m2zp(5nB8J<A+tsPcc>GK|TJ++?+>vEyEqdq%^q;0fQ5<fsd`DF0LC}
zzzR^_$?>RP{+I(a55aq8K_!F&gfEihh-YQ%Y!FliPb9U@4W#q2#VF+H!0+ro`12-r
zmw%7T&%TC9B&LHEv9UFq7Vqx+WnlMe=xdqpmG7QLUL@qV4mBJ=*KDVMWj715*~a4c
zlnCc*-BZDLA!Frk$>Ge7YtCv8&a+R&xMribpHA}UL^CBUE4kA2#CVm)z_-8=R_AwN
zlAn?dP{J>Dd>|du3fmdKc{N_Kx54GY^$W+=xOAR$9zA2Ft(|;RGFkNt{PXnL5(%Q7
zqmQN{B*`H&ZiqRrv;rz0Kibb2K9;hH+c-L=VY`K057<3Zdr}#xU8$fk!9}(9F?yv>
z8ME|#DH*-|k%)a9Pz)Y^f1!8=lOxW0@>S%_+!v#sMbM$umk9IGiNJ1^x$cr`@vQ2A
zTExD-wZ@pXGif%NXV<g+!`Gjm3=n@t2mjjuCHy+4o^9E6gH1F%`s#Vg^QMuA@6CYv
zr!X(+h5z)m<2_3gvmVD3#xAM|-#e}Dwveo?;uw=QDDVl;^m`e0dh>me4%#M`Xa&dV
zD1DiE+fDWVq({S|7pGi3HyUZN-f0rOA?ruX=d{09JrMo1;=Rq#^KHFfiGNPO-p<7_
z&lsr{F16+ErOR!$YL6UaxsC|gx<KGnHm@U!nCFU|X0Y?v9hr$r1m<zl9mdxG#ok+m
z#nE;NyGbAroCJ3Q3GVK$!GpV72(FDg!QI{6-KCKLjk~*RH}1RNZ|3`F{%g-Qr~71|
z)<Jd2s#Vp`y4Ss?<4Qz%Mn`t|s#o>%u1cSBQd^$YIwV_r9ELMf?I6)dqww6`&$fmS
zk`+crL_F)u?Yg(|#(-C|<vGt?ma)h#cYrj;U?PfgmL#+~zZ3-_$G<qXDzh&guctT|
zlQ6W{L<mp(lU(P0xflD)yG=1VyKrDTzdpy7LL?^4#FpG?eUIJI#e`wo3e%3R5VUtw
z_^QFaqI(~4{(Fb-2+GlVRCb5}(-$OP<B2};fmp?HNs^0IG3l9yk>3V+FVof?c<%e)
zVVqipLr5k9k_{#I><OnT4H<D{nWu8n2AVvoY73HvvQ=ubR3Y&hwa6ZbkuSs#&G)@u
zSz-sj2)}0XCS_5tR%VusCZAVB8R5Sq-3SNLxSsK)m);~g3L```V_(DfJg}3e?bk1s
zjP^U7IaE$48;hF8G$(SqcB~B|y-WmfH#*3R!jvdcs0Naa7iuwD5>(c!48tm|*-tdq
z8L*RbvXuX>5aLh@mmfNKQabOl0mM_iJojAGyej0&ujUrl>9UVH3KEL7)t^a%WsRSO
zLphQd87mo9EVwkmD>q@C94qm*W<=ID7=f8vvjWg6;wUC}a9WoYSFcO8r`S-CZ8zxz
z^6U;=OY8~3gzIdFbBVd4!a_p*!vlCJ;hCR=EB0=8&vc(heNlEbbkx~%&zmn1+nV^S
z-NgKoEvzN8_Y5gq4%63@zHH&~3-;L=wWHs1alJ<I<lq4ZO|#)FbW*PoGpz(Pq8S0y
z5M`@oD{wiO&8QMZB6agdpltD&lqwtGS>Ew@6ma>Tmmzvvp&p^NGD>DX+Q3H&85VLe
zNWIZwPMAa=l^Rck-CSGA(XDgtGUaGlW2#*qDa;=g|FCP(+7GodI`Kilk9n|LTJRS<
znV(|D@u$CZlPs-I;|9*07k-2kniyh^rc77n`Em9Q23HP`f_T~L3@zmpBJ3KyEkJ*3
zgQahOXL^w_)+-LM&Je??rNA*D#YTQo?Sk;C&kXa&QAmh4dbY&BBQl<s&~|khwv-u7
zG{Mh2x7G00h=Q&faS!ZV?4drlW1Hs?M)T??O|WOwLpZ%{D>*6rJNO!qHu_2#k$Gdb
z#l&a@hO@-Ic9_mI-IHxFc?V+eq7WR16Q%WHi{&2}?d`r?xpzGsfuq1;?#lDrbh*GK
zqj1O=tB^S1gO=N7eV|xSKPMxk^b>)K!U${au7lQvt$xF39B&f>pp_b&G*dffGXE3#
zFDH^}7jYq8GfU@oma9#|XaiVf(_iYM0Jp%D;seiWzjv6wM6p%f?b&^~Sx6ZTinUtg
zwBd}rC(@$8FeH~(iSRnoxwMp!^RL(1N%s#L)(rEaGu)pEfZx540*FC#sJ~*gJ@kLG
zqNOcmDv%7`GJDrFwT#Ei>uxodQ5M<V^=MCePrQGEfEgvT0=2k1{io9#x*ADfwiC^v
zQ`-owQ-?v)kFFK1i9%pf;vLN1eURBz8_q|ljIx2?A5A3ZVLk26TBxQD%+yh@xqQiM
zn>d-4G22KTNqC9_h!;A1Y^~=!{JMr+i+pfmT4ElU{oQn+HS|pH_Rm->o8`K@npNsv
zgNMzw%J<Tayol%>Q0!Tral}R=6gDIRD2L#4{?-+=t)ri?iS@iq;4rQV92-J;6xWG5
zo4->XMP8e+p>M1<KmBtwj)nk3a8bL}fVa@a(jd7#b2PvrWK6<mtkt&TpfGG7FKgCV
zNBV+9U(eb96X9HA&7ZX#7NsuMnUcYlz7`ZUZLXmS?&E_~&>p(&h)XH_=7r-iwI;A}
zj9jy`bu7mVrf+!kCqHU`$i<Ak0=a)+PKP3*p5Ucijciz(3`)nQH1mP^vyG9WJv5eq
zgL-`$`=Zsy5ZlT1@Oy4@B-wf<9m7*}sY@Ts@yE(n<yg6eNHKg6!9qF4FDN8TFc{4~
zH8Hwm+^=%AZhd@s$~AtT3auZ|zm`~zYK~EYmwd^7S$JX~Ja)oO#;+O<oj{*QZI_aW
ze_B6OZm+H`jDt`!Tf%UN|AZtS7sw5T+WrmrI+LMrj6A~GJ$VF{U#Quej<9B6UCn(~
z*~-FXjWMqHa7Thq#Jv%?Li#(|b<+>l@<Ez)HH+F455O^!z*(G`eK=-rIcOe1nM<ZM
zI*$XYX{_xyov^8A05lZO?Ee#hnle3iQFTp9ItvNfNov|VAHDYBIq0)Zr<66+?`6dS
zX6Z@M%Iyg;Qz997-{FvCtvTLooYy~o*XeiF)}I#gNFOCk;+A^qPnHc^0!BT?o(#hZ
zx@NI0;{m8EOBU{;34nutx}QHcwY~kZF0C_2yR$0sikWHT(q3B+*^KHF85M9^^~MT7
zOzZr*{J8`1DPX4;Oy{7izM?pRpjivXKSD7YC0>Ha$VIgs$eD;k1RB<telUOXs}RXS
zD#oVed->#dO*RQNVXirn7JHBpym~TtXUE6GkggXeBPt)$OTmz$Sf_PUaSnY5Kf7CC
zXn6Y0VZpciIT?#UFT<xPS(ue|yE((sFJm1p9(G?x%`wL3>}x%obLQ?j;W2_qeQ+y@
zUif3Mz4O}4;N+3RP6!;s6aG6w7!n@Q&7sj#$~TU|9~1JIV+yI_WQGM_ry=erBb?O_
zn!xbq)+&gWEvyZ+X;?>KLfV1IzvK7@g*o=78H;SMz9-TLnapFPe`V8D8zRpT)iV~|
zF?`gp;N>yR2-!Eag-@M1ky$=|I;i2hu-!UPEq<TO3Cz07^3P4aX>i21|KnYQgB{(C
zg{{U8!=st|_>ZZ%IGn#Co3q2l&Qn}pbQVi?)Ru(wWsw_4Wg>lsajA5ixq$JE&hg84
zCz(%9`2XZ^OAWb?-~WRT{qNve4hs68PIzO67PzLm>UYg=zU|egX?!y?7d?N%;d<qX
zyP)NE525Pats~lh5kZ>1(2uG*UuiM2{{c|b4Dj8x0|xjTZ1-TIHXI7zf^X8<LI}>0
zkTI%dJ_}g+l8XETzOeqooQ;FLXn6mJeHfD4onwL~S_!l6%fz)Q+P2>KKIVt$p$wW4
zNd!*;3z>E(A8LxAX3m<#NO7hbS1Fkvk-rb4{(<WJI3%GiT5D8p(|_Su|2OU?G$cW)
z^^k1`&Iy(H+Kw7{^d|5gQq}o^$d-QcQV9jqmd>|QRsYx!>ewbwlgBR*oI&~5(Je*%
zbHb?}V*k^@;)MyF9;RR3M@#g?m1+D$1e+KxUg|UU@ryFb;YSJlOWPYG_YbX?-Xh-c
zKUjRB#Z$2j!A~A&Dk5|3bBa&W${COA8-e9jZTP|KT$qrmK26*0Z&P%%1hF#zLd-Cr
zS=>=1c>R9^@z6gAK>r4%qfJsF{tp-UzsSAKK<KGu#N^)n*QWnQ2L5}aFWNusCk$Jd
z$p1n&3Vq^&s;N0CO%D10_n(A;uID_wi_H5!%He<i%>Q+_<e`^r3<1yme*-FoKHUpJ
zZ?}L@KGpxYkAH9U{r@l7f8CP*PcNB}2Q+)bqL{l2b<4OnQDUVg-nC*|{0b8})loQ{
zAs}7;;5HVNkpUu$FZufa&Q$-9vVXTESYj~!FvJ!5T5!-rck9CA14R!M*TgUD3u#H;
zMlezHpd8exPa5il?mUh|d;zZYFW+i?L9C>)7Od$51xr8r`U`*dAqGG1R^8Lt9LIxc
z<|~8swS~oQ>WvJ4oMJxMr%kfGh2Svv7FXctKj8Mr1!VM0`LXY;{PhDbf!0=27vl16
zPu96|!cv@?%|%zt3?)ug7srR#6Ub&6?Gx@-9S95Y7TMny0;hz%+kkR7)Ih0kXbi4L
zvV8Uilofh6giuTR+F$Lq(a~<S^K4I&i19auYnkjCDeh_d|GH8C{rQL?gg!Lihl*!>
zB8eG-`UFpXv-xWmKZg^O7xmQ(5+kc4{|(vtJp463%)SaCj1#9f%J~gW#wANBXmWjw
zO5<ZJEn~>%W+<}U&I^b1oZIF^o6ymSLG133$Wr+q_*QbyI~+E#JI0`JTQ`kV`G5Sy
z;uBFpu{++dFe>Bx)Ywp^GXQ7f1H4}^N;qmy7hXVNb+0WT)%EuiJs+gpIw~|W!&9)K
zs}tKrEt5THYj08UmTr}G_vbs#j#D1=!PPfwuG%sw+UShM9PriVfH_HsUvddiSnb_L
z;LuRC?{`&Z{^gRT6N7cZjStCp=N}}0aQ4axrWN_@a7I^qA(?G>DC1`~6rCRWL8E|&
z=EK34kB+%dyfm;X4F6`Jp$6=nChq)2;r<p;XkIwCJHF0ctX0oB*;q3K8vZ@it?}ut
zY`U}2>)^1aig=t8a8z`fOqf;eyHqrHPOVquuf2Qw1bDmnl&<_IB(wg6JjL~-S!Jrg
zmA=Lh^5##{eP4Uc;JtF=pQ^affd1Ha+M-(B0bd0$h{sOlZ7F-xf`xIIm8aG=Bi5Vk
zenh&T7aVIGaI<+czub(uj~G1W^hQ~D-civ`2`VJ%1zq5!bZ{U69%NV<+npvtMh>l*
z&@+wG9BqbS>`sikISU7XnAx6VgBK%xl!lXmfT9}5rY@wbbEXN7h_4&Ln)mZ3Rf&qt
zmJZW+2}Oat$D%Vz8qbI1iuW!=I?SH<6>g71Ri-s#q};BEYR?ud&$qI)*K;iVA3KBj
zQPN57ij_U5h`iv76k154`Z%qp2wtagvH*MHxZaxgHxo&CotJ?^4GtKxDJt)DA8+di
zo<<kC;8!^LwsqIF<$IG<C;I%e1q+)$LfIZgCfm=l1kg70CfOB8P~|$8`T8Yz1DLip
z(bIzo!#_?viQ^}vWg`DAt9PMp53NV^4e;gBi?s?~!MXKwQwIzDV|#<;thklmwyHu;
z1a<h<$WnKRGt!_-iyzDFut3Unv|y03*G?q?@7z1ii4{0jNxY_bd;<4$V%R##eVPRU
zxcDUE`I8Jk_+_c)oXEg(pA4UDV6f#~UtGTa_fs13Z^^wRn7;!`sh`*+GGtrAy`0e1
z-u9{8EVf2@xS?%*z5eFdLa)A<wB^^hPNJV0$Uj7Q2l^7gcJYls_1wgwO6)KE>Ni2q
zwOb}hj^}V{`Ye};SXxb4Jf1PeTF;rbERy2l6Cn&;Xn}epkGjtG<r1xki$(r0LH4uu
zqi0jn*zr6tVV+!jBl$z1>xxe2a6)trVC*Za!iM)Fsy3HXLT%YhB@eTtUfT-seh06H
z9Ox5J1B3R@byE<p&tD`8&*%=5<nR@o&cXxJPD{F{Ncyo=%k&k=sEYo^ohxm}kW{<Y
z&8kL<jQCh=ULxb#`eBj4>&gdGf={}#PA*5xQ)HZ9_x56BuXwxJ_ZW};TC`3AL|Lxe
zB0qCM`Ya^7wuC;Gvm9m3_BYNa|AV0?SpH&J!oK;+o(e|Su+6Z*a-h{uhP5bOi<`yk
z3!-;i^3#l{1hr49F>=hOoWWL`T2{X0Q;+2w0mi!N`#8H<B5}z6?E<*!D2?clhq$!p
z{$6E5BouAmbizewRKQ;%{|`nz{q$4jx}yprUf-*1-R%}snQu)3cMoc;I#rL-Fmd*R
zgu;uD%EOZK4pWq^5Wa^G(T*7}8L6GN8JWrT%=42`9ba<{VG)vCd)bj##t#Octa92w
zrK(K>7to4f#aANaw-aFwt*PFIq{auLM&t7-Ymy@scJ;+f?2Z{LmlozC2atV4b5;cJ
zKD4mns#A1+nrF{)e!{Ujy;m}elsUzT7=W@@zl*W-zKp@wPN?+0IO_ARb%bhia?Lhp
z*iu<(vlZX<4jqEXd*Hdk06}%_f*$?+;bud}b02vdpXP~ZO$U{W#itBtHLsd%BSt^N
zeBiqoUfprl?n9dM-%U$e&Ar?&Ug$;E$081hmHL2A0Lo}};!7kM2mQ2i^3vP~43um{
z?6Gg=xs9Yt*MwytQ-xjvv_z{LI+e~pw;_Cs{!epeXrwmO{3|^pHP+Ng$L2rGLsuGB
zcyz9m*7WH32tiBB^93glrm-OQ?25Q+UfUgHq6>-O<J-|Qm_bH&wQAd)y5X&0FV@y<
z7UMY^HO(do+mls;+^cLGDA+9q&7^0aywyz4He(Q{@l>JHA>;rQBMTls-fO>J-n6mc
zpf)3sES`cGe)m6Ee92Z-t=&@6*13yJZ=Iop?5peSA7ylJFfssW2$@*;pY<4$e>G*4
z#k=o*e3IqLQqzf=-E(CXtvSD;v~IDABCy2H(od;l&q9o6XE1B~RBWNm+Z9D=)={=w
zR0)`TGzSh7iVZcguVwNJZm<NlDT3_w+w?#}mI7)cdOA)1XN&sD;K?vycmWIX?k3Tm
zW>6zhXzm_nlU@ZaqY*8lT1_4VRIM*8jCoVP_IuU$V+_1reXAVEEEVaSXj~`b%K7^N
zva>1E%&Wf=#*jX6oJf<=lDpd6WQGeq6^sZeGV}uk+P;SuBS91;gq@sF%R;zUU15ow
zO(v~58kJ0nV#`BlQ)|J$a`o%0HlHm=I=>Z%Hj^#0cE_oB(_SQ^S~ln*j*8^s_nfM2
z1Gnw{DSPpfa2_FHcpE+1@^+i`>b;XIpkIBEo1XPe2QoXnMWG;r2HYx2TN`BVdD;Fs
z;&?Jt_yej9T{S|vIqk1??^MoY0n?#Qx>cm9cD%~yQmoCYcft%OyEJWqaFwE$m#2!2
zs_7<dn3M~JFP4elYX<X;*~L3;-8&y?_GRm_Mf8UKKASAY$-XW;*WRC&{JH#-qVo`E
z`LsH_ZpV0Ms`~kZFxZnET`{<TW86BspR@F<1l#Z$b6h)mZehVP?t31#_a4`pFEbnp
zU2w$9s}?j8)!@<1`{0lcTYT#u;_h5e#Q))*XV2CO%~VQp(`BU2d0IJcMC)T01gvAe
z=S6}JRMI0!Kk!D?-j@rmRj`?#-bn{X6R&*?<G5fd6a-8$XauK6?x(D8i{^QD6Ex$G
z|Hg+Oz3O@|Sj}UNcmv$C1=0!<n7G6Lc{|N`y>zP9ygY*ToJ!EmG}D{*4Wf7pQr`rx
zV13LtyfpUll`7y>Q4KYAi#WAg`o8*kfB%dDF!BpJIHS62h4h<!Gc1~VN*b5W*I+|;
z&AyzSewR&A=dTb?|I~vFb<^w)IGoPMY7PMUMsU&U<&&dEf-4@MmZ`wQJdFZG`14_Z
zpaSPZEIR=v-i!Z#&hW_pd^haZ4!?L}HRkuzf(u9rPb&c)Bmbw3(AW6QXvzuKH$<=D
z?|0BYLooELt3@4fN$x~Q{>1xrKfDSr0&+g&$<a^x`=~=p84_rnWX>aQ?lz`J@>FZC
zuz4Am8CjFnZ$h|$r%#93KclvXM<E%qLvzGbA<L`?PBkO9&@lp8RjU~P7Wcn$w|Z@s
z`E3LkE^G~VeSQ5^&UB&hnE><}mMglIo=v7_t11FBbX??tP7A6&f?2G4V872PMF1?4
zQ1D%oA2OmE2xpAABmv*-io(Yx0C=y^58wCTnIfy(B%=pKhJIIhAKTqGbl$fUH1*!D
zfJ8;d2JGzvMDlYOIFa2IeUaf&)5#MOiaxf{TuKVAH2fU51Np_fa|C+ti3|a|hn_7a
zGQo)(={b>LmA0V-906QYAVl&F&GT(sQwv>39#J9sg#%Y48wtSKUfi3eu)^2;?{*59
z@^K_~pM+GJgxZqtQyDU3R2%lGhS|hH(`D|WMQ$1i<uc?zLtiH{=a;R{%l)n&B`-N&
zzcRk#cYG3FmjfDdB{OR7(>;U^^{`NWLA1Lea|Hj^Vq2o!2r#u}k>5Q3&kl)S{DuQ-
zVF#(oirzWP=u3{=c5`FE9Fnl6$K;%?GYnix_(z?2G&0fP?W2wC9coBWjWf=%E}zTc
ze-QV1<D2j#TMF~x|869iU;4rIXo-2F_<#CHh)%4o4~0v3LYvAGn}?S%XAQr@zcPAU
ze9%*MJ_}d)$WE#X?MIrw>bBs&+dlD;dWWTbLEpr$53@+F(75j1Nws-&6(4fO{O1Mf
zFE?KSykdDy6XcV1tYVM|ZHybOG^VUS+83=rx7lwEP2aFb$4fa*1P50=K<x86j-9vb
z2nea^j1>!>5#V&=DDM4~>U_F!A$wp7nbIVx`)eVhl52_P{+fVYNs7~xn0R@rts4@2
zzRGzSI{Z2ukI)?o9-nMPmwh1ByVkWj%Y3(o$oxDMr&|=`qEOrQ*CoD2N1K;GVO{Xk
z5M{WTg`>WJuQlxP<<$TR#yNU^?%AYIRhE&y5c~$LGptW#NSXM}Rst1nPnYOYs*~FE
zW8R3Bke|gGJI?9JuhijH_LTwHoH=70u@OQsxx5`gIJ`A|daY&q4yLPkYq|!7<YH%(
zBlO=_h;XRb$-25!gBDGp+h0RLO5V$jSij?rkMU!;f9(jFqY%?b==%iJ%Jm<L88f&u
z$OY4iY96_nT}a+YL#Q3`1E->MPReM?3iQ4xISd^OEB2iYp@u?q6U#twD4im!Y-Sa4
zUlls|hBGL~eiKVc5^*6y6s*OTvmX1BZN32Eo^D{5-CGZ(mu%DnvOa?5qMKPebToAF
z+4>uCRQWoFFE1f}4Mgi`*H?!V8?f%92EYTUcFGNqC;?$zt+9D)i6lRopry8Fmj-4!
zWTOA*S_X&r{<jN2!r&@Gp>&CslrHP3!-jA|Lefp374(`DYnVmM_Dyq)Z`o_380*3l
z!AR1v(fmbl-*nYi5J)oi&B}eeMr?{{C*)Qqy_K4uwy$5gVyZy)72+Elk1_~DQ>A;K
z{`^A$#8RmVDDO)e`LVH>742Eh)D=0plEFR@;AvUN>?(OdC7`#%$cP+uVUvYwxG2%E
zsraL#*4-N^hxVro_r!Rcyb*B&j)7#)7G4M8^^?xA7y3d`TmAv>uk|Lym&xmqqRBCT
zlI3Ft=}&nu8;$_f-8s9>0C`5AbKY%%VunFtcSEOATV_ZWn^hDV5;aV!X1HY2JsPF3
z`xB}=S){!bar&YZD?|IPM}=M113z(;hzvUxNosnUGur+2hk*fiEo{dL(zslz&d)A?
zQT9fwrEj-p0ViqZ>Tjki!GpQ1hm-HVvnuxK3^^cAv_An7KPF<ljbG%C@R6_-{}~1c
zyM@NJI&s)=+j~E}Xp-Y_p)CLQoY{YZIq1lt@^LJDGsr(t>(1@EN^e>GEQiC=9&jT5
zSfwBPDzW4DeIJNJ<sFm74i2`DD1UG8y)O#bD+7vp=5f1n&I{1TQQxzoMYC<q=3e#W
z8r`;DUuyJpK`Ta5*C{wlGmtGd`~8#dsm1hsYRZ6QaOxz3q3hUhlyIFAs*<OrRqSSe
z%N?htKDGVLTBZF$Gs=TYQot*<ukiuWwDYq0{KrSxC)^1pXsTMteOxBwKZ7|08qD7v
z^_F)8*{vVCYIU4X54rDa5)<k@hG%;kB*t}i8Z2nJAmol<0DEv?=NCD6%rD(;M1m7*
zxBk<ARTXTq9CAO?mMIi*e6nOdt0gKjHxybcoxVK*CA#zp=^1W-P`>LkL;SC^Sm!so
z+_zn~19X2bhi<yKObd@5mKg2N<@O}77e+6%pd=w6DH|uKiuQ3hil@a~IH={`cmX4V
z?Slm!TDadoLjpe7)jK<p8oE0f6^9nr3`W}JzBK){Z+0CPe8PmAfKGvPLfMRUEkZj7
zfF$9Ko;4a5(An{}iN&9l%IiJ$Wc9WFDy!1kB%R`8=yXMxnws$FJ*y{ss!jQZbt_rf
zMM~doCbLhhR^ESKg4Dv?6c@M!Dr;yQVq!(#IGJ4PGCyx6zCK2!?9C3dTooqFASti)
zLbBCahD4PWDWzQFvUNQv=_q~rsZ#}h82Z*ZSI?I+e;(W8RvZ!rc2&3)^bgEFoG7#P
zrshvYu_RlR=7(~5h5!`sO7HVfSC<AbaylAj3UVDe>_;@PT*QmcvjSm$ggs%p*9<t>
z=(W&hmV<yEpJYj5V4Mic3I}sU-HIy|44E%G(Czfk>1a}`AY!WlPEWT_Rd;~|ocXY~
zW}4@#nuq+Asi<lQ{;(VI$l}K+F%Lnhbu*^Ky!H1{ySgr6`2M_L!)vECPW88sTN!9t
zb|jPxCsb%m<ODqLR#YJHh9t`U5g|ZGcJ;At+uAq&Xw#sZ>)66O52AiW)V)-olghSR
zJbda^LGq61?^K{?!;I;ZT;~M9WQ=?qx{Q=F6XpuXyZ*kIV^%dmx-T;M3asvZiVWYu
zwNKG5+Kl&vE$6sDmR2!c&_Fq!Qm_twv1-*TsL{~h@3|$?L4!p>J=bHJlOC1LWyL;D
z`2n*_E;YfC8;=ax?6`7^H%sKIPh~iDeEqUxxyr`%!(I}Uau!OtAu8Z1=X&$YAU##X
zZ1qw4VW)8;VM*`#>steV!tgs2{xyy}=URUz@kz^XE2eip$Ey4<fc7!|gwdJ{IJgx7
zQM5_wS|n|3o4hH23a4*LC?gv)?>$h=^LKb(6;Z5(c+f!MJA)UsnFlPV_V^znmi}fG
zyG5&e+<Zw`akAJU7G5!eJcKd$_|Qu7AAJ2lqJy{Xyn#b^`0?+1ZU@;mVe!!rP-%bM
zMvSgMOX0~U-SdL0{=?M}2S&eu)0~P!+n>WMOM`YqnoKr{lIpuM`*DtEgOi!w@i_0+
zsH|8AY4u`?t%l`C#l}<>pmj0dijzPby;}Jh(sx<g0`ErxeH5Bs^bSaFOsQA@lzMxh
zo**tbwoX?~K9x*SAMG>}6cdcRfE+KBAC(8r{0UC%PCmr&(js*l#ra-Y#e9xWp(v`j
zN)*5=<Lj)z|1!1}e8fM<5wdV446~7WlORk6%yww$wVDAj*np;_(A7QN^`vAhb|~s?
z9o4YBMO=RVwJ#>om&YA5Ne`xA5akyUN;RT1c<^r}-pPbWQZAzciTxx*3EhSUjdFW)
zhOQqKbVnp0y2k#HJ}Z+UO)w^_0N3knQX&t*g0($9gndP7YBJKMN3+g7%%j5d%K_PN
zO@9{>QCtFJ;#=p5{znYf7a%Gn-VaYR_^(y2-_$$oVg!b1$364!r;o6Dagq<XcvOnl
z;syV%2764`Bgb-Lybv2<@Y(SQGCOwmm+xfiw@Q4*mE~Rkap56SpX*hR&Jxl^YP}n=
z^KJ4E>#@K~gGr*`uJSJ)jdwh>v!4&z5~l^b4mPkVsciSsY)S!x{DHx)&(vMsg${qJ
zIa#QE8NTtfzy0ROwDNSh(O0uBp&ssWwUJ~wWg__ntf9NUO`4Fp%aTuZxE0t!{i4If
zxbtk<uuJ1eSK6^k5#Ne!E}Z>=`l=*^?}mDAyCU<|wM+oCVY=MTmY0!An0^{%&C2le
zDBrt~qzX1?oQoSZZG^Ls^SgNfEuzOHRtclx0o4#`12+=VgxnPz5&lgU3$xC{>9G`l
z`4a6PeXJEGue*SA(mnY#B_twd9?XxKt~%=MwHRpjLBfcyXS#QoTq4Z}U?=_NRg<Tp
znVwtC%n5lP*RPK1@F^LozXll!$}iX}Mr22o3yE^`sde(|@}%s>qN5frNB?dptExyY
zXEfq@nD0McYMu_?ORenJfk-D}>lds%gqvG$mPoH!54qtC_vZ&lX^AF|qoq0TuSWIu
zpDj(_C^P|uSKN9J<b?;*?*5s+$nIe$nu|14zFauAG(H_><c{z0aP{-)Wgd>2jKYpA
zk%5#+CH)%Q#wlX%U%qf_UbTQoR{Ri)o~XU*$N}BSUEFDokCtN;yVmXJrwNw3Qcp5y
z>p3+6){Po(Uw(!-9Jsi>!YO0k#d6MXVIwHrp$MUplw}lIT{^7<dRk)nLBiPLMC!hZ
zCQkQRsi7cn7LlUJhD-!cUNO52vOI(3bJs&Dd4^LZ$*HtiiL^{qCy;i;*lkSEQ8LcP
zzibjwjo?%trlPT%2!3z`i(O_HbA1>|fXGBs$R)6<Wznsj&vL+Kf0N)-)On$iEHCPj
z`o^`%A<(^2>Z2~W2wZK!8yx|{^6%njPjmBE%~}|78vyZ|KoSQjCX2I5^T7S9RET)m
zOJu8d7X?x6NmjV3sMUE5mt9V1yGrk_#&(AtAS!={>nuEyXph6(8QjAb87%*Kf_IpI
z>Ije7<R;Bk;E)M9UVuz@@J2rkb)nTLtQgMApT{xglQYkFm)Z)$J7iLJd<3vp<}Nb&
z$^{;o{naJzyDy2No}IDMe+<pGSxMV%u6a7VchePV;+ZO>;4|4BO3ktT3O;5`-c&E(
zFTa0AWKnm@qPz0mneU0eOO6Kt;zvf6jZy;wGLNmSGR;KIgC~5ixCg&Ce%>Rb8LGMx
zveC`8pCBR^F4m0i5+;h=JGRFaaVJDt8!j`YBmua`x+j5y<yqHiF_)RmiF(tMA9xqr
zNn|I2k2*Le=sEjs?1Nl`-svlSRw+(nk>~c}MT4!QaC~epwFaqsNvo)-ECOc0de7*h
z#v2qyq$tBb`i6$nNNC$Xz<WCBep}+3+MYcxcg9b-N?vqzm~^B;ao#a9pfSgyI}D!7
zCdHe{H=}iQ=V|4mq^@=nupXc8d;ggO1_=g6WM37YVV%mh^{>v)SM3dJyI)o(k*Geu
z)MJVB$4&M!G!%Y_tO<YJ*d#|DpYAE3zUY+E2~m%U98<S_nEvbMg1;%Mb3ac_e9F7(
zhoekzh_HfyHorBGkmI5mN1E~Nxd8nhi+_dfjL$sHP5duab`K&6D=W)5*O(^zw-kM=
zfUD8z$nn8x`c&gQ<?*Ryhl9^4rG<>OEdK=R5t00{h8?~=W15PY<)8JhOkm8p7A5m!
z#yznl9)kzeiRAMDOuWI|DR{)u>P5pF++F~k@;cSeEQyR*shx?97DFEJ%%`Y}t)-MZ
z%lj|LYv$_wyL=rs%`?9nz61a$@S>u<?6jo26uT(M{djX;ctlvU?B4P*xiTL}aJO|N
z#xgaNwTS1<F<W|PiKZBsk(i*qcpDym_3ufW!H3s}^lH_B;y??wBk~`TLs@7%xI{+3
zT?psZb=K=G=Cru5+@qDzP#@J2_;Fz^yqlLS`L|3Xis_HwSNTlg_6~a50|?eS_x%aq
zC$NB1G+gk^C_JfN&OMxOvb()^vw!uanyDJ0x{*1Yok*${C-k|&$2~W&iFWXt66_6c
zyUL{=%04>?&Yj}{hK2W^TMu4??`7#66&a)Yjv(6e{D}2E2R(Hcp~Tt)bRp0nr5O7C
z-Pb;8_Cm*lT<U<D|Go2?le^nbIyZM=K-m|gjt6*?jz?Vuq-2qJU8ChVUpVqF$+=_<
z$Ak2oJ9=dioQ|!ck8}|@9?b!#YmjfE%2~%Ud37TunSw+Y+aY_ne6-YMgnfHr(Ulo6
zhnuD=vxi&Te3bK!U1GY(5$;nsdLF&Il!}85RGOn1S@Nh8p?j$J1>W{48EpIqs67t6
z!{6ryc-2!KXJp;r$5$WA9L!;U*D~AAJ1t2}{y-^^UT|b_RvVYT`KBc4uCt)ZMrE9J
zzqK9ovaR+*wjF|nD-fcG+Hge=W6~z!Q8%Y4xEEhXFQmIM=Oub#Hxrm!%zylZ9joN0
zeiau;$8Cb;cVJj?8B8}l*y0+DsroBR?=%=zGRiJNV?}4t&lInH0i-@U<hJZ&%T~tn
z4B)Bifmgr9B&lw3A*z0qDh!2Ui;;Mu@yS)r994j6z|U7}r+Efu+h^Q}*z!xRN~en*
z-XWH~f!}?@$t2MREtEs#Q{n0p0(QY%=d`RB#YULG!cz0~SY%52Zk<-uhwZIzU0&R)
zkegw_6Dfx!gm$#b_?}Pyy#D2<i0A**KVm2P=+K4AqfyY2$n#EFjITIrIglSpnW-6Q
z@bio(oun~IQtD9Cf#+Vje(ispbyVh?;T?}gAm3)LJ!Jub?o%92<S#^yB`IU1O(p9W
z?oabhlFpL3rwy=@Ddm629feK0;E!&QFVHzOu?)riU9mPw1@^{#WF<3Q|5xzP454#6
zTEwi!iZHd%<Lxd|cm<dm9~V#z_!r}-HUkbezA5Kg_P(Qb?voo$7`_$ioZ`-8YF>3B
zP)0PeXl>1Voc6!2)69~9KOG3&gC-s>lAG_C`pIGwL`Mkm3~27=NiaBqgqRQp(K{2o
z2lQ;T<}TQy+`k}FAqMyH^saDWwd?)_X4|Z&_)5^xk;xNbcgHmMWuiU>k+_qcc@01{
zFx<6@MG`RB>C!L@xIL6pQP<$6oR7!*z#G-B-I7oTQ*I#=VwuErY_8R0JF>91w&w0j
zVn@||Ak^Na?tyN0El?YI$P#mt>D*$9qI(zWR?wk>he0JvB~_@Zcc3;tTNftuAza;Y
zr(_S#erj<Fg@|R`&E1|LE?{DnA`-c=;$6fdO4{eA(A4r>OHDR+=?7d|tzbc$zXh#!
z$%cxaZD|~OY+w_PjG5At2Bl>sbuSJb9`<=5p`}C1`-owk5fxe^Xy)pY$Y^wX2Jh0I
zcOfOf={-OUh}Ru$HPj;PORo+PtVl#Ylj6y1dFO@>-s5If)2;{*#RdT@S<PEMx-4pV
zA<ZB#=lm2dSqCr;@8&PK_Hk0#I#ji0+PiXFUa!YeL%juCd<X;^3doC__;;t9Q1P3K
z-a`I){xZQof#M>vR3dU5mmc#th47$kF20RSK3vTpogEhnqhxt%_>FsKHJ2V&hvnyZ
z@X_nc9OafFo)rgIci>X&XxL18tjF%M9I0cn&c3Jg?v^-NDvYjp=9f5Ur-V(4Q-!LP
z!&5-6iuI&TRWURfeI1$w40eDEc*~+aqh+`;)*Du_YPOx{Rd6s2Ka=w-K)*V$<<og{
zd^;vK#72K8%a2Li1)A)1i4c?3dHt<irBZjO8czM2il;t1b-hxB8D7;omu%-rjdqbZ
z^c4`@mbM!<VHYj<>7+0yi%V|5gX&v?rbLLiv5fsayS{tWbVSToL~N(X5<%f9hxpbU
zm7VTreZ3>L)_Bg}(Mnw#DN9}U;BnRcZAT=Yq_M%mHvbUAG3^?BH9Kq9+uxcMk1i4q
zkyHmQZgGxsH!DjyK~u5Ar<XOUK~X!$MjI?=gcik3Axz<I6iuJwF^z;YPVA806;XB-
z?Qvy_@RonliHXG0Blo58?}`Xpzq;;2IhF0h)6xk^*7l>O@0t~#7}A(xu}s5PTORVA
zE-D`;oxkd<DYCPaXOVet*Y?<Mhe+(Cb1lfn9wK2zt=~cO{je8bJ~7R?o=*u;;cn^;
z5$q%^;7Aq1a$|yorW1;hDUosW=b5|5(#!@KAg19xngmpQb&K4qxT2Ul$(9P1+e`M3
zZKkhr@`!H2`z0WRG@4AXDnQh`3x|^5cBh>=k9`lMeX<&Kt-OTyAe4A%;iH_g|DzcL
zj4Z|z_<$>(`II?*Z4g(w;QsKr%kS~eSd-Ty3YW#!1Y?moDc-HjPt2a5uckG#hL0UL
z>)&B(bWrcUmcqjKor<`Ik{)Q|K@=5~E;zD%>%fWH^)nHspZ1cblL~0?7g4|^SL6OE
zwy6xr{X>NH=^f6z@Wh&jbjP@Qh1Db~9{Vxp^p(r;E_8TFM(k$~u5*Sx4m2yS{nCaj
zL*8uYrn}$aQ;juu&p1LsQ?fMu2xur}{1q3b$2LfYNx(l2W_w0q6TJB<YHZrsS<bwX
zFyfDJz_e?lFLX?6CD(52p26v{&F-#CC9zE)0rP%Jp2~X&o#({e&dqM-m8i|$zS+h3
z$F#wowT6ujSE_!HvKIn9t%j!mXkT-O;!f&t#y{^y^SUWr&%6tfhsnH|xSy+8l9`~&
zAOv@i=ETDn7iOkoaecQl@j#nNk@_Jzlc<658xCoJp&lD<;LQ>j2CXQ%zQ4sZ&gwo0
z8jZ%A3}}*%l#FYqY$IOW8i)oL)9*PY+Q>(^8M4^ki!A?~6d<XVZC*L|rOTIIZ=&KQ
z{Vfig&3`bn!#ybvGx%pkg$w<YwS*;aoq&)P(r;^#+X6Gk!&Rrjdgm5rxUyp7v<k8u
zp_%O3KF8N3v&J^{j0n~rNY(cEUw*+#BS0GO6R?YACWaRAMlU5T9k2=KbBnJ7iY`t7
zhHlwEhrh|E>r0&30~Pd?9*l?d^yN>!@(jWd#O64GM{2`k!;p>e6PdROV@cK1#FPnB
zbCmSL0v4!r$fb@Nvn_8+EQ^4>loWs4d~Qo9`GTDVdh*(glhW;s9j_9QbTj+qspgM(
z?w^gv`M4-NhfZFj7|;k7sB*c)ct<jQBoufs$3FBiEzL5IgzJaOD-aDIvJgq;F-wsc
zw(X*~xGXUoY*zb_GSA(WUhlr8wU;_wV#A$Fbr;LmaeTByaZt(IYA3oS*J&`QoIz*0
z3br64c3wa7*i4-I60BhG3<^<^At!`Wv|UFRq;Zhn0G>$7OB0Z{Kgcnr?l|gf`ted7
z6s5k`=}wUs2A&=U+C}S_A8yL$_p_Mqes4|&n~o1eByB%Bx+uUwt3>I+Q)<Jf^x=!?
z+ZdJiYu9%rE^YPlBu4}hR?~(F-m8x|iCL6ye|Ye-lETwNZ4+i7aJbi}4cY14(na|h
zp&1oL5qc7MWp8Ux8%5u>dDk)0ka{+^7(gohtAQeC#GnNci603()ppjEVVogxCgB?u
zYcEXqq(Cp|9Pqu|ux5fEg;kw@=tyeg;E%A_3;5umOPa>f2DEw+myisKyjj@QGnV+G
zL-+7o$rif5+HrT7gM-kUVH>^MIBq!srpHJ9XG5|PZjP+eo_MT1_(bdb#6~vW#ccMm
zFyJF}z)SkwoM!+m_vARGZs1bIcqrWB%8R%8!}8joDD|_>-bJKy%O2Y34DWZf&n>-q
zyi<In1^9E!;JVs5LNSvJ@O6a5H<|0dj&DWC;%QAkGR3k`ZKWRU!MGCm0d20lKz+qm
z9ccsntSIvQiGGZ!VgKIBa**@R00C+-ZhNNIc>ZwC{7x{x<G|?5p4BaI`^uK0>-+fX
z{r2Y%7i;5y-)T9qd7nwbjN>>{zn{LS!^kAv_UOZ~MtvAQ3L<HlnK|KekmTV7zWcf}
zuOXOAXi=|Rk^dMSt!DHL0lto7xFYJ~xM6zon6A6o^1#-2l6mXAPT;o+F_<1Gr7bew
zc#Z3C^R7jQ$T=uDxidinacxQX9-?XHu182-CSz~j@pma`@bP@!(rW@0pRc#Wj$1wd
ztH97aZ8XnKlqetEaEzeDK8n?bKd+i!?H;@r#q~_%vUK3ql;bPDnxKRHYK+j>$TiE)
zxcDw|Gw#vz^onJ3PuppXGWsAq3EoA74H_h-Yn*H9m?O7c9t_`RZRir;2&DoiGzzV+
z)|MXh$avD$IvqhbvvR;d`f3FVZVJkGRqrhLr%q?pRCzo`XlFun4>@9!*&f6m;YQ~R
zaImmj(9DUauVQ+ROArXqtV5hL&$c-*%mq6(Quc}Nt2sd|*SqP(A(M~8ZmfyvA+Yn*
z#cd%n_5R)uWsIIpX*M%)0XoaNX{ZM*so_y*1B`zFK#%AQvBtCCeeaB3H0C_IFdx=K
zI=n_yCmvY*akve8Oh76=vJa)M2&VD_qtN$Ehv&_5#`_D5sI=xWPA-sjSk00K_m(vU
z{WCof?}+22CDU-acNJhubMVcL1`lqDQWt{*DP<<urN&$wiwSrpvr~}x2@j=st$o^>
z1sVTop{}K=gphF3^9x1C)x$W?DkKz&btKJSD(<DXzRHm=NYZeYc1kf-Sc)Z2+~-_h
z%jf;*Xq;M7+~ybaMWx#uNt`}H8l|KqIL<UD$;iX}=miW@(^7?zC~BV|k`ql*9%2s=
zp3VW2>{{%46exOA1j(cNYY+jui7ySLk49os2{R;wqdIDo6;%NioD>5QT<~KClX)H_
zxj^wDcZ}*fW7CYa>HUgK2fyM7<_#ja5G1oc-q2p<{LbC$rs62q>2xOYTdM4q-SUSu
z9&>s+^&-e8*30<qxodZNLO;Fo<K~L&Jw?I0!f5g2jk`qN?*rM256NdhuA8hvi*^#i
z$}aNEnPeNgU*H1JU3z9`;dvL%{TS_zTb_NTa;T-*tBY+$^>iDFynErDTV+MrO=JW$
z3~I)wFPw*V=Oz3fWWEt(m?w{`SZ+1vpbXddT0f<EKpU`67i%v_o$|CZ18R!RnNL$g
z0c-kgE2XyF49Tu>uY}wlk;<ip2m5F2T6s>$7a-{_VZAJO?`lWPqI1rjsoOo%EQj>A
zBYI9+K=GV&&xK*Z!y59Kjt{VzrV6G95QPe(${~T5$5IKK73~HQXhX|DNxf%BW}QW9
zq>7)$dmr_H-F-D5ALD7qd;JIIHP@K^uz<nmdDpf^FNdz^!Sw{b$;%^w4n{y@emFil
z!rmm`V}t+_PN~mf`DBN$AYBevtwby9P|}O)Gmt8EG~ePr<~{z9P^?8o>kzTa;22x+
z5&o~zzS~oPwHEbZt62rD-OrQr=@5R_kG1k|U(PV8b|k*v>2Zs$1&*4f5tL=j-RYFu
z#X~|Ike&TkxSd7_y?rd!TpozbO-Ag{n_ig?Wq{LX@hyCWCL;wnu}zqV+zF!7%qg#)
znl^SdSQrA;ptPCtgqx~qesm=HV^L%`BBGBni{Hl%>3|pb*4a8Onq522$#8nLax3?I
zhfXCpZkK9lckTyNhlF01jy0tH&;B7E!)+EN*(K2iNBrIg9nvoWPQoi#*E8z;De`)S
zEr4>y-}~rE7eG_0cFC=nIwyj_i%D;rYi_OjIj}FvW{gcT%DuGC*cmf8YN5g;>U&S-
zr@uS*0IL}najW|JC%gsu$(0{kz5e!>+HJbw)dxI1)Dh0YjkX{-$gx`t7V^SoW}{C>
zZd4<h9-i`zq@l(UVdSCAw^(jCeTepK?Vbnbh0eiW_ed6D9{(e31qqOSz<oBWi=Q&;
z3x9(&hh6%IE`WmmL^KFsA$`4ho*wRmmeM*)4bLBB-LGVE!AS=uZX&nlrUL$Gco8DL
z%BAFPon=g7LB39T1oQF9f7_~GRa{8wJ8dxwKso0wAyeSdD{i(`DvBIyh(!vYP_{N<
zW7HdmfK7;O9)!Sb^5h?X_)J_W03r;^JRyiE!!P9=Jyg`=_>bQS^`B#oPr-IJ@I+Wz
zt@UcWo4z4uoHqjI#!EJ-`;3PAe}Y&C%YXC%V~f`s3KBw)+Aeqx5sE(|uR(S1fLm!9
z0uv9@;%FRt?psgP2F$D!ts~wBzXM*Q;}4d6J8h6OoUB{i`S{%T8!&KtN}a~=PeumD
z?L_jrdzN4iwq@5@I0CZ_@3x^}H-)8qVtltW3qrmIxCrpySba8Z+DvoxR=8wJBE`u+
zd&&l@#9o5(#fembw8Z}Q6VAk4%#|#IwC9vfzeh+FaQUe6zuL`y;Gs>Gvx}8K1?_dv
zCC7R`uDbI&7qL_-6ZH_bn#tr`dtL${({ovEr6F77Ba4d{dMtC~Xzp&FgLtUVIJ<5f
z1BI?_J5t=Mq+W=|ZB>hVBwr62`n%v)uU|rBAA&kPKn;ZNKjmoB;4ZL*2hWta?)O1X
ze5`tuz)WQ_>AN8VjB6K}%o3yN!0p40!x}t+pJgFa+4)SQd@gH_wR9eh8DFkdFIhCh
z68G2*AFl8=H1`j~i;`c$UTQi%^6+?0Or;9ITk+tnaaS4JxSzAy6*>_fKcw|dw+p3S
z*<)OYGq6xPzAGN+>vI%!hMgL(_LN{inFn7O-qmG}B9nEsLZ&qUIFXqsF!UzsLX<Us
z&q@vfTY63%<oQ184U&iMJ{_FjOl*I#QUkaeeAEOFiT><utW~1@953qv+hwy6B*7gw
zY4CwBgTj7kV^_8b*6_Kt9w9is1i2`sWDT2ph5kDDpl)T7yOUqDvv=AEgR7-ELMS7=
zKF{aqYQX)OsA%tO>4dwcHw;c<UmCE3b>g3<is+sKx8{~o^W<pz(mLLBz#R$#kWqMY
zwSaFIJwu4R=fLvf-{2l220H%gh<M*@O1W}0t-ahuA@$RDWzCj?c_xi2%UEPnCil+w
z=hpY?9;U#S%L>HnN43J&9bi20=4iKb74OLs&ngen(#*HD-^jA@zU8HrX~H&$1P7VL
z_U~Ce+c}=()M%K*b}cKHE#SR@pQD4JVHsRtRZ)l(Uo?AyVl~9uji}<p!&u08hKyWZ
zOrZXd-&=^kEY*s{nZ@17^VU@$T$^-lZ%%&M^4Q7J<`o*)%@sv{SKl$3aNPiY$f&K?
zfd9%R*QN7i;lXtuIl+L&y)T)=oRjP4O|Z|SKZ?9@jwmZqmr$?@JSJi1;c24JB9*@=
zCZ|57ImqrYSp6B*K|hPTH1fdn*$(UX?nN?Q!snfJ>mjs~@55PqZr-U+tRX-UbIXEx
z%TGyI%k}$bi3#ENQYT!vr4_&cCCJT;eE{DhrkovirB(t5OQ?&Rpru-<GRt@4h^;5|
z({F%D>957Zo!UU6K$F0H0g0}M({CX7w?TRO*!&FPfGJpn@vtm@(*gy-s1Vc|%ZEF#
zI*ZI%=t2$cq~GZM>=*dMcV#nVjc$ibqTHlIOS?IpDt|6PjXN(qI0wjshD3(+?sbx#
z&+No}H&i)U-$o}fmvxQC6>?NZM<+{Arm)=F9y15XJ@+D=Kbjw&=`}A>2iK8u<FoU2
z_ETiDm~YdZ>dLc7A^?%2)=r4w*2jN#NE7EGA20-JwsOV4hx>+vcX8G1ZXV2$D0p@v
zn|H3+wg1`N1x`6`!{1(J&$6H`A622N-D*V4ob07nSSFBN_hKzguUDuZn4nYp=vQyi
zJR=0(8A3}mm;;BGQqTNj!ss=;xzIT}%@EcS^2`kSyL6J-IbmwXSbq_?HiqOXJ^N56
z5y6?{Hm=i}@%h3OusDNaEqZis&>&6h3&$;UlvvkLWx8?WKwb;hTT5jUUP)KH_guhP
zg;#amH^{6+)KC?2+GG!nI)D!zhM3?gJ&e@(-<b*n%&8qS%-RI;4Adi_qt(n50OrX`
zgdJw(2fwoZay{T~5I-DT2xm!*t<%AOn+EKRL@OOVf`EAE((C7ICVG{-kT3fCt6aku
zvvr~{l|N8@_31p;4EIM-DCmFdvu~IWZGVH0SLSDAIUpws$FZA4XnNoFvQ}J!YeumX
z7VePphei5GBO#p9=Q*ae<9%_9`?T26Futoy{(Z<~@6kjKB?V^InWeIXFTF(?X&f5S
zMK00uVrOCWBSbPawcWklHYkkJhEgEI&l2YIt@-Ce@TUe64{f%N6`BVsTox4_`(e`L
z;^-AIAVFl_i}6=PxZdW`B~H$S&XcN74b7oR&9$EC;wvF%P%R)46+L<Yyr`z0V-T)`
zVfSJ}Gm#Z0JxLZhL+7)RRPSpZdW?o~C-u9AM~R?wi4DK~hKzS|UP!*L<N1}|Yn=V6
z`aU|o7>cvmt2Dvy<pT-KbS4Ar98uq7Gafc<F%0@c!vyETCeZlE3Aswb&;Bri|4jJu
zzt+X2M%Q{5@7%GS=FAQp=P86<%x9w4u*r7Z9l-Z*BGVS1eB9GXVl`({eyzR!SYBMa
z_kDw9$ns*uBhI;mR{(iu@-y!%JZZ8I<KCB~m(b2`FR6&G{z)?t&4w;r`Xq&yJPjRJ
zmdc4&9g1Kqu|r?3hO9CsN<D1jiMNc!qi4CEb?5D7w2iO9$V`_o8pLaNNZ_-H7h61R
zv!<|}m9BYy>%eGl+I`IDZX@v_{f~5%Sa*?Kq=akZC5$N)^H;x@Ipeeoc+&5;R$a~)
zeqUq#Ot9`4-=pw8mNsSZ9?|_0cUrq>Et{-}MUGKsp7|RHr{vIZtv_IXu0dviP;?J`
z^%&^M@fE%ZeN|SSp!UhRP)Y7|Zuy81>Sg=6=9302-3M$x+)r+VHvAm{TCLaE;MNO4
z$~NZ^R*6(6=a~6s)Iapm{G?{jcIhDZ{rTp8jAl>hnqJL=ChYU40&iN1ERV!RM0rXL
zHF}%&XIUTI*ZqF0%>w+dL<&sl>Z=?Tf^W<D=gQS@ve$1zE3iEHXx=p<sKJ5EG0S%?
zhdyhZu%KnE8*2q;8FOQ+xgYku<`=Q3Q(6|b!FMJh78-_&wrg<jX&u*!sM=y;(E980
zWcu%p1u$zeRGvK_RdrMDj*=6dTQa>(e{t9S=h-lb`SPy1|AeRdD4$pPbTSZAF332j
z<5-JcVT_u-9QJWXa4m@HcC{h{E(Oc&I|vc<Y;eIebv=VOEEz!k-lOr;F2;Mc>a9s^
zW^1=Vm`W_OATJhvTIAa|jG3y?PuU7uR~gvct9G~cXZCjTsb<Z!eC?UWTfpXlC~H*_
zWMT#ze8=yg7AqWrB0aLMrUwP1%?uo|69^;o@mZxfDcUEE)c=RNw~C6p+17r81lQp1
z5P}5>u0aFA-QC?CLU4!R?(XjHH168COXDu5^RBhm*=O%D&fRzMUH->tnO#*~GVA$0
z695&$#}t}3&ocn9bq&d7<_OjIe}<pQ7#^8OM7iAS2}<eViyl`Lx2IuLZTO-2w~rvg
ze~E6^i~mBwzc^=DUSD>(*Xy6%)7W(4;>H!b(&w&dX{&PzJKUjpmI=c6bOwwmDkimR
zG=bKe?E2`9u<OI!d~a+%FVwgZ$1!@j2$<K2z-6~0;kO+-f1I1v%DF@D?eHvvMD>^C
zg4K2Vd6pL7*)Dd*vy4Ru>IBV(y;W43maKr_?kHyNKb;|XSP>TqK&tG^yt6ED^<(h&
zU{JkJfaQAbc0>($2f)@!&BN~*`<ZuErp+1Lx<%v5<a+;%HIFZl)l9pCBwn&g`4bpJ
zx>^j43$Mx!Earn&Ty;lLu-Ku@h2DQ!5xcBpu`?(+wH0{n)3i}ac0N5tJb&tW!DvX*
zHq(2aCZB46t6z8%(fVqTnnrs%&3T{;LK9Whr8bprXkR3~v(hNN!G%e$xl^t&d4l7R
zLG38hc5K77zn8_j^$w#B@$IlKZ@zL^dPcf?F!!nZZl(1S&q?~6cBCh}`1a!`UBnWB
zbitB4$K*siT%zM^Pg0M&t1UJOFaPpAu(KL&H|$qJix&E+`(;MoWDl^T2^W6D>4DNp
zSa0ci7I&o+>3cW#+V7Ye5W~N^t(zPzK=AmR{8HD{?2jFxD$Um*x5ri0Yt{90AJFf~
zFaqBXfP^L6xpwPqj_^U($?;yijJpMJlSu>#9#4gDzg0zH(H$$~pl|gn)?9MPhOLVC
zv0TsYs&<`-Z%6q)sbqIjvrpoAW)-^`bUw<#Bg_S=YHW|z^+Txxx!-=_O2W=xMBf3)
zV<^B~<^WwSE*AW{@}xuGdEXRAe0*T*iP(r;<6lKASL;=c(nRl9H~sJNmd!yi#rUPk
z1}m7~DS-bj$wENf%VkuAnf^FnQW1sZqWZ|ZrXg3Q9&8(gtlJ{B#`QL;D3d_d4G#T~
z7y9B?^zmy$?)kXK^n;HkcOC>zC|zZ<LPV!e{pVIlla7q23Bu&Sl?SngVxI<?IS#gP
zLy|&w63!1YJS57W!^GA}5!$xk@5bi^u%fMjP&eG$wU)Lx8!rO7Xi&04S4X_DO|&#3
z-J@FHyJdnm&uqgZ%d$RU1#1R;$lx?zX%e~o<}U*~(faZ4eLh22&nNx=n(qGomHx)~
z@7eEPk9Mu3dPFuwQa?2K|NZO#*GvBOj@FNm{}+BB6Xcn%6?z3pUt1^CKn&uuyFd`1
z_Y(EW3YlP!AhVsIn&9&|`P2`%?6sLHruC(=|K31;z`$S5&9{|T=NdKw{M5vc$4Hku
zs6738Dc8)???F@s*RQ&#8r)Vk>SS_l=g$UDqa6J);S*Ea!dyV;Jx-1`2dnjaMf;&H
zdMr~UV)9w*H|M&CbAV@my*!TD<I+$&>4)2k5zsg$t4b%`zU?w8_L_sT%E&_HM|ZdM
z-K}J|E!20Zmwi>#@M)amt61eL_CG;_rFZbtzWrUpO{VG2;l;0xDY!alk<Q-BsQt=u
z*u`_1m9KHhD!5ur3dmh(u4^@dFRF$7r6FV-6E)b(J_GER<CDJJl<}`~1n6eKcf`xX
zTAG3F@h+v)=(H9|df$gtgQZ_)kIJz!Lute|7|NBMnqCJqpf72gOL&Xn43pzH!j2Bb
zjp2NjBxsJNj5tY?&n;?QX0@L=i2%SWBlKY!$cm~^F4r|Gl+bXaQidTNG>WdrS~XU0
z5q?MXajfX@OUEo#Xa3Lj_3(_3`z!4SB=K3`1^Lh(juX7ik_Ztz78$kY4?{oNg)nXA
z?DLb%d7I!gO`VN74nc2m{7rgR!ddSsbI132KEAH=0aDqf9qF<YT-pXjLP({{U-suk
z!E-``tw};+Bxa=GtfPRO8{jD^;iV}3o9ID_lLT>Q<+BtU1iOd|+^P1}Byqc{(mO3n
zMqJhF90n`^KHWKo`+k$P>2X``F-Y4wBa%K!2S+eTol!t>F{_iilV(-qGH`AU#>!&2
z0{PJFEkXQ!H&w^W$4Vkxa8h)0*HYkaj$p3gb+BS&g%S1p_IN_m!_}5Y#>PEfGYjYu
zcG`GD6WRkG(j|z%4DmTUTJSY>bwJtmUFlb!DB;FkLrCnf%e0&}g9g7-|8&MX+27rV
z1dAdCzw+hxLY?=}!}sV}O61$Gnnr37%jA~|{CvU%KpxUB$El)15Rn|dq+oS8Pk&ge
z(%2Ph_g-I<sQZ9?tcACH#QW2CJhRaij)BP{Q;zE|q&Uyq{(>i+3ERI<&SPawGd;7+
zEHvq1OqnUjZn1}jckHJ%m}Or;zMkrZ_ctfL=(CkawlxzxkR&1VqP|R>ny%jqysm3i
z2d*9a#H(B<pXwiGeA&Y5#2B%vc<oz~IUK_NZUF-l6`z^qZxuSsxNhi}AKL&%KCiag
z6yS3&9@#^!;LFt(8MEnj(*OjnaE-GHR%8bRbb6lvhUwe=DkZk*4Fk69xf<o0H~{~$
zrAmNo!_QxZ*A@TxA=3(j3TthF1B_(U#7lhpUD?c2LHFHRDji9tof=`1E}r|NKgj9`
zO_z8LVETQqXFTX=9nNcw+JeqPH0_}y`C8U>_z9y*HxY^bAUaq?^j3Zv<@LiUe0^=J
z@7w8ZuGP!HjEtGBEsk&{E&~D+2{`dcutGkiiFRmC{9OuEBJ#n6Px>*R@Jxcc?`U|?
zH}*2uMAM1dl`Rg+yqG7j?vy)jXhMc(^@?NWRzA=1AjR}Y%dd$x<`C}`RA>bm>1D-<
zs{+f6(monTnd%)^u#WoBYA-!gy8Bmm7Aog}>5V)Ez!U|9VR*ALKS_621PDB~On2Dc
zYdyNyKoaRG>mqibHae_zUO<w!_4YK{xld?st&3NfIG-s>7hkY))u>HXUtOPeI?<A?
z53LnVItk-!rTUUpSjut`TM*4@tSl`rbmj_P3<QP$QX4<k)0AM%NBTW>odu?3LK6NU
z8=)U+!-W6`D)-n`p(SdvJWpCheh_&Lg7=$WOp_d*wrUq<XC7pgDAl1SIvw@L#$14r
z)lT<GG$$2(6?K3^BWaVC;gr6NngP+7<S3me(88n>4Q?hYpm)b9Y$w4oTyf@r@JBRQ
zlE17lRW}2wIJ1;(46RM5c55YHowMczhepZ|Y16iWg~FkEzmpV!Y>vJOw?tLYVf@9g
z(W|$9uQka+l&+uYR|#OFm6p%B1r(yb-D2F_!ay8&@CIgVFgP%KF%8dDyUq!=OREGJ
zCr$;V2w(I9o8c0R&en7(G`2o>xIDC-hGX;xOedMwEK$&VmaZw2p%8L@(O6MTL4ECy
z(eE3;>2!+2Wnc38t=sp8(2|d(Dtg{xpYFD4d-k2tl{Srn!|qN__p934*Qe1C<JyQq
zdYjevhsNUH#*NOfC}9gf_g_gj0F_-rm>X)8Ff+%}!Ylu#b#l+;@&uv(i`cP8aHZSS
zgay1qGY|#v+CTyi7qwIwP#TQw0*nnm`sb`wUM3L)%IdBT1eGL(R9U|%Zc`OlSMxTs
z=YmO>=}<d|ZDZAFY{Sa8vSf}6l7X<Ou6<{Bjh(kI(+~PETAS8W1)Sp@S!U&BX!BJI
zQfr7XD4$f3-rHjj<~)N%oQ3PcGmLz)Klfe<@d`kzmw!;_okN<@B@NPIQNqkYf=MS&
z$zOvWc7LqY=!S##8<n||D;{0$Bbm8lzTnpTOsA_h+9A*pulYyjeSpz4j8ueG?~Ytw
z<*S4iSD_lx`{Bpw_t8}&QoG#%M%n>~jhZ^Gxh~{dxGo>4;E>!&obYXVGH30ev8^sE
zCs6`=N1*TAfEE%aM+EJsk7w0ZI|T^Ej&cjtM!DXba?D3}p$29&wr?_kIXT_)6_?0|
zUYzszU)^J5_N&#|RPA7-E7sKh<gp7~G^_Oo<2TxMS2-l87U?WEgddMaVj)U>liLZF
zn&D=g3yk3Nw}QMNORBKFd+vA7Svgf23*E{<-s*WjlEZMiV4Sdb>NiDjX(JJ@J08b@
z0LvMGio=Yf+`h8hzFcYTn_io8onsN_1#!G{kKE;amGmd!Y<%Jh;FrepKs&j&q;<Ep
z=GSZUWo7riXQB768d~;SAfW|1Q9up*H0X+Xy9`6n=U3LKwej{|lLlEQ8pqn3G&78o
z{3dGkIuu*`3pIU9u;6R<gf_QV#D^kFjr)rbxuDw2ro4C7>scpJR(c38^lFg~&97)H
zDP@QQ$zBci)>FhC(9R|w=)Xaxk9OT+F?XFa4c<9Vp-t)b^(NtyRbVlx7>{;jHQyYc
z4B&*>eiDYoPsI66ud2Kl?!xHX{cM~!nBNC=e#_KzHN<hUXWjv|B^W~|juk$Q&W09|
zvv_WRfgUR;Yds(%f9U|O4gaCd?9h(+e80xz^s^3%Y9*Z<%{t44tHBT>u+u+15>hiP
z2^COvi?&1qVtEee^vn3P0n^7I^+ja9Of49(KU*p&q*p#~pa6;=&67PTtU^$h?(mXM
ze4X<AX;AgLvMz0-7=wxA17AYTP<ErFXZ;1MFV^0qO4!@GV`?&yy`cOjPqdREGPQ}a
zQ^<a#)z_bzVJ=dlqEvHP>?N$hNDP>UU6@vi`*Z9xv5&x2jsNljXt7yn{8LQTX`-)M
z%>MgWsIYxSKgEP<$vyZj<P5T&BuSTv8u+5#6}1>DLQf|s-kEeBi`>h?*!lDHZU6lP
z!4qoenFHS==Vbf2__DA4OgK{9$MZ<%DcA>!`qQa>9d?iU=ELZzjQ}6|M*k%5bC*NQ
z@4B$nYml(d?duCLPlm<KJ3QGGr|iP91Nu<Azdqb|E_;Umiu%CN2aO(Aho4*NFjB04
zhdXYE7%pG|WWNqdqd_ex2o|~=np`~%U+pp3h8~#R7)pW9HN-r5%u_M#D^C+<4w!K6
z9^+9<tTVFETZ?5~Gy^z~r4jT2KbtE^2dm#OMAeM3hsV}OYosIJYvD-nY-Kg=nZp1-
zqs}&fLY~uG?;pvCO-{es9GytZj1R&_wEx0bEaB;y6)KKPOxsCDiSKToB)%2V=oEfX
z1T=r|CC#u$VK_3YbTPrXW4#)+M=EDhJEzixkMNyX`Y7X%TBY`lmRByZmE6ow?{x}i
zglCL<H8!}Qr6<DeME!)MbbgDKm7@cT5ScO{F$UOW(^dz*<p>oxPKcBpyGi<Z(PK&g
z8>N+HX1Bnowo9*ZTRv+sbw&nF_2jg>C~z}tX6D*i`!xTvb-xu~b+V)7$}#6rUz6SR
z>9cO~KpW8iW!(L`b=r2+>R_Vx#!B#(_$!mI>LP}@i^W0;*wtgeL-Q`k^|8@>)(R=S
zjxhCI?jRoX?tVIIqe}6C>I<>%CYXC*DXelcHpVf+85YTz_*0&`J)!;lhY6JZ6b8%6
zHgBbHD=G`B@t%f_eqfo^aY9V{EUY+7m&PkK>Fh(YvcjeL%}zV#pd?dhLFzv30;P(_
z&l-U)I4+oOZpI~nB#z^CJrL2i&)~p?fWvm)B&}A;0u`qI$iXxIJ`KtmP$yR5q;dP=
zXq6T>swJn*_szNtMQrS#{50$XSFlG8fQCQon}gphZu7kK=yKb>=b#&J#jH(`CVMmN
z99Gn9zPh8AN8Kb1@!XR<vvD6x9bCUiHzRE%ypk=lORZ0t;#$!ZEj_)GRGNf67B&{(
z{vq-W%-(@)Hz9cA`j8_7-KiXT2QRn3bomv!(o-|3?eo?iqZb(sy2(O0o%L!14Iqud
zsfzD2DoVTyai}wzV~2yL$qff{4XHfBz<@aZ2ClMjSPwHIj=X+#6khr^+uDOy!yrWx
z*QAKMfD|Rj7p9JlKG~jTW#TngrsU(B*q2*ndu?H`gDQ`4ykfk1Y0w@mkBDC+-{pzR
zq2agFU+<@^lPZsf{C_DdSD#0%Ngl5(j$i`ZI6#>AC<_Ok6FAGk>ByE67bP?J3{+;F
zUHb9pa!JurbaL~@oj6aIsm*HijMFJQ$X}>%mfu}ntTDEWmV)nL!~td~$3aghIS>y?
zuG0Jm`3wWH(;!^}eb*DWq9Cfl;5{Y<l2(cVZ8y-X{xqgnnq!!raC4Jwf7vgtqSEZU
zDe6V28~XT5m;+DR7VzMOD)VWudGZxIQDYVuw8*%Edej2o1aqB_Q#YvmTLSh@_5)dj
zuatuEDdct@4lnL4*A9GOnxr?N3zdp;OFWZfTSv*Oh!yqrWoD0`Z_-(m(WBCQ64m<_
z={l3sxO_XwX!CR%Uh-{A5JCMeQz+&q4FZuNMv6|sgJIE6C3Bs&F_7Sj&&KUB&tT8p
zR;)6>8>OvcOnng|YKaFiJb&zpTgH9n)~eqjadjeq8xqWr#n$>NI*Uzitajcx55giB
z`c-O76?N|s*`+r*HPn?GwLYU<*kzQPLq@B<NgV&E6$(*(4dcA>C@YHlrHxh3vm_#c
zB;bwzIOrgjmMh};X`_u{>_^p-q*LCaYA8>9h7|^4hjk}DoB>z!E%qvnt#|kgGm&rc
z#oTJA=q&G*G|N}}K}ss&gY}M-z05v_RT!x2BSkeN>z(i2-Ys<!am#uHw1}Q)$0HWi
z%hftZjr5n{s&fKsg@dM(f_h+kR(6)fij@WWaDL6U-ULT~BKTY5C=+2FODCkFbW;#s
zxe*?~9U5tWZ2C#yfnROE)@Thaa`Cme)gmPGP+6tflAYz;H+*TBz5Q#4DH}gP;4>5u
zIK&0dj}FCNE&7WLoFYL&@+1iSbz;}Ux27Q9v+TOQ<pO$ARep(b)q=7BXFAo&Zy4~^
z@=0$&w8^2u8Hz}XGIh;fCCxGdM_Qwf;u|p;9?E@D^I``w^KHP$6WvzOr-7{0jM&gf
zBE)U&!MzuuzFL>3jS^7d1a#J(sRf~VeJ3lSP1|{k+Vq0&@t305akE}|g>$9t46~0~
zQ5>yf6>eMEY2;B~>rz}>wd4h(ND}&J^u+WX#U7%&{s1iz0YL<fTuG}uU2;oDEMTg8
z`4137K57)bhHP<Z4~b<U!&p%T<e*Sm*}a80=_al&&2l^A&+&PsW-WmN?a2Rz>wJr~
z@jEWULwsGO-UK5($SK3|<m4RoUH!5#j!OV?c;xuoMm#59+Ngcw2CK+^xXftv_(7Tv
zvJ4sv^%l3?9lI!0f>ezB1hRQ1OczH&(<o^$L1|enaYM74;-4X1``hdvsuvy>xy)*B
zInsbr$6&8Nj(X4Zix4gloZ%PZd)xLJbts%OnVEdY6~fo7D5%^@*&EH@O3(V7qCt<Y
z$|lJ!?tS)N6+Z0~A9+yL8dH)M6_q&&P<=Uv9sI^jsmCBsBC=o&3s`#{J)B1v!tQ0e
z^@x(J(~SYEciqe88X2N*_FcmB6TJ|WPE+DXCpuPrCE?fe107}7<Uk*f6?W3JrYZ$S
z!$fhyDCUStIy=FTNuP@4!$u5wZp7K}(k|Q}A1@QCmDrn|D{>;~uhnx(o$!Q<1Xt`a
zr4K<-9!LCr5U4|C3FRM0gc<KhFD0M;EZ*XM#fZ1MN*(J)T%hIei(L5-kyLNJK$H<`
zEK^oglxt1(@<+$9fv(w$;*MAwmVLUn_lC5-q|%P2$O`Ok^Q0t|VM_H~ZZ$K0qxzy0
z<pW7!BRjp#jBU&2#F0pQjg3?LHQhXL^TSD0(@e1NG=I8_?n*wDQ6G&EI63<#d;X-a
zb<kL0U;-I><0qS3U_IuS&+7}W1nDa)9?|mR1L&?WFw>hyU**X&oQ;c+VemUQUrMSf
zoGlunuwDKnVxz<$L*vIN4=%gXBRMVH?EP+^X8smMh{0D$^5@VYmy>9`GEOPWTU@`{
zv3oAG&~8~Ll!MJ~*nUeoJd#a#NAp}=BNb)jvd;--;*Lq%#VZ3(g<3t4fXP#fdhfNx
zX=os_T5uVa?x0ZjT@a6wgTQ2)l*ns8lTp!&T6q)mHA?K;^>p*@0KjUyDBbBJ2ou-n
zH^BrqC=?z!#LftCKWqHqF|ICwJztz67MPN|$m`hjMc`D978YiVovNtMtA%ogL^H`b
z-u1Pr&aE)Ty_ljysy$R@G1QnPjO&aV-+ymo-dqcaI#aUZ){Qa_H$EIxgu?7@FFv%+
z=3bOW=<-6uot@xM0ATM9l|otEbm<<C8;wX04#;kQa!NR`ltCpl>{R0VDsLwfGy7Qr
zRWCT)WC3Ow#?67zb!@~XaWujdNZa1l3_CMqu6G}Rlc=hY7J_Orz9A-~oXEMx9!u<Z
zqrH3+G;$ZZ$RTicq|{;vZ@L`x>gG}SbTVlD5AJ$Q6*p#ga620`4{5`_yk3IGA*ULN
z!Xe>u@12!+li^*VFUL*eIaQDXKpCCvNOK~9=(V`Yv|LTE8@u#CcsROKQLV3F{5=A4
zr~!|%9i$^{2G1|Dz8n44B>CqYOT}iEam3Q4{%kA;d=5IjCl=+q<WR=pJkv+!tSLxW
zel)dJ+bgrNBC8*;A)vr1VUrI7oe1AaQ3nfhat2=QatUq1axoZWyp>z(APO%g)4mZE
zLdhycBI6V5BY&`+Q=*ZiTI1pvqI(y1*6`<U_2UxW5zH`}J^dE8c%xP5kMeSoL2u#o
zyZT4aj~9nr6uN!Q(lb60C!R6*DLmwiV2<$zO8hzV7_)ktTc(b!2mQWr(;9jFPD|oZ
z<XK9Z0AS7L#Lv!tFpKZG)zdqNm!Q9Xh0KF~y=&Q16z|}k6!pEEFG14%jnM9^qW3*P
za>SvQZQV517%8hWWeab>QL2^%%|RI9*#xD?1b(@!kcSA@F)32eiEkNCTo_@AzMCWP
zoN4`Rt_#KgL)Bv6TkLtnz{FucdAJu0!3<-PQswgIDQ%Vg@djUJrg4geb1EuF9bOXS
zXu6AH3pAjM?pn)L!7bX6o`QfnIL)ibDI$&9H4!7SU9sOZa=1O>_M?-~AVkPgGN5}L
zqc5NVhtcA(gJO*LA;3t;jkLqBZjqFK9=0S7bC6zN#nf2ixx;h4CMLt~WP1YeO-3vL
zhkM%jK)6}DAI8b<!$;F7OKxmaDn~DAnPW~in7}RbYmXiHNfT_C$+4e9n@1_fRgy`|
zNQHZ@TCzcfd7pl|^%0tUwrq}Yoi1&dNW{70yhWClp_9pb{Jh83wR&2gc~LaJIz-BM
z9=+$AX9R0DWjgjGT(oh~9@jMP9>b*~E33SUA0VB3ufBw()rq6}K(wor*?PV!BXLf8
zHIvqXa#e@{3ca48b*2xA2x!L0C}bx*VO-P<@HM>={tXr(dXoITWq2Otg^j9HJ<XpQ
zBYHe^;?X;<?{vlbBK^u#4k5=TGIlirL8`0dkg6w)Z#kD$&>blb<g6yWN_1VhnYOWT
zLZiW8ANd-}NRt~mgPcAe43Pq{fQL*pjTYt&jZD}jf9-<qnX~8>Phc{28P!%E&$Cc`
zUBR>68?vni&65TmFDIn-Kb=y@-4{=cr|N9<f-ceSHDy|Vk3hL5$^rG;6_e8|tp+g+
z6qn^tqpCa(HZzCfE5nb=?aF6aHTACdw{IX2Z~}(E9$3%j+$bRK2JXsnr$^RVIc}?o
z?n}{dp*jiJ?0+vJL}6rp<veTsY4D);=xUrfly^Z)BPqtt%VOy<1QXM&LxJy}YB3m*
z!$8q_OII{I(7Q)rft^o4)8!hF0-F^>Zx(8uO~HVnI<5gZ6mJhhT2|**XCZtzd=?kZ
z!?eh0{-ynaUJ{F$QhVQN#+c$T1&}jJ9!&7Zcx;kBI@t{lRRefA1ISNNhqPId@nh;@
z<tc24sB?>dE)Qdk=U-s{%cfcB3ITZBvFG>=h<;e0dd#>}cN9SH6D`_{yPw>^@Od8F
z^BYCY&nnI|I((>Wo*by$yt1y{OJMxavU!(61JrMlzWYPcmFd1?eBW=5Q2D%EU>%mg
za4u3L&cuqhTaWSfxDZF}bJ_+B|9jN}teyOQeTv0H%O*=v`58_&X|kAX=L1J~dXrv5
zU5!&)ohiV;_{EN@^v~p4z-)H##3U;omwX0-qD?z1Hak?eUjbMBsYzjD{V#l?{_y~5
zmY=<V&}+e?AeUciJB$Ae0gE-UsqRZwywG&h)=BY}f>{Ug1Z}T1s&Dp3J@uT}YuOa3
zv#L#Aed*zZ7}|xONlJ_4RqOaK6cTo6O|@eMiJgFPQ+Sw(lR8MfQS4`~vvx!l4vjwn
z_6?^veG9!lq49|$OGQq-qgbZa&7P0)5i-q%8tEOANu$!&;kU?9tTMEnH*ufg3Wy+M
zwVw9I6Wnz&F@Y7XF%sZ@H%|sKi$j5Kz|#SgAgI`DTZ=2BU%=JIXm+9b$jDbb+5H0k
zhfG7{p_ksX=dt|57-8N^E%!;vcs9)(?PTT4g+pYVGN%sNk;&G};TWEciMt5&cekkz
z+qa~g9Zva4*v-#o+ebI`R+%8p#btk}VNLIOl!G6f?W0Z-2n&X5lKy5CE(v3Kar11N
zagbtKKU}_&7-oF!Kd|3E#FOQl2p+5PIG!%U(})}X<<}wOolvHYK0uHV3G28iG0>-8
z?e%Vg+H`r6LZgw08dms=Uo>$9KX30hc<Go1ly~f-*!FHvCW-Z)N`(tz!>I*pVrSn1
zKUW%X;5AA&>}`Hn-JUZKh9uR=pSppeiY!R06z3X^1}ctG|3l8WKOZGu`Et64-r2#k
zF7%ppkBc|O+c@1Nv-cXSQ+3=rVK12!+0SgTF52kq+n|&W?YSSmDCYUYB)%cf!uN?x
z(u6`t2N|zGpxIr~vH8>+HL~J!PY~E@ap@A0#6gAoa%>IF=01aGg#uB|KXHMezE6!w
zI0Hp=L*#Q?*essEW)S)Um@Lx~2sD}+ds(`{fR)PIA|$~20<Uv#F*FWV+dhovqXj$a
z++wy&tf}lLddvm9+=2it7}|_tyjcA`Nn@re8vOJID#r|WM2-jU_)I%^;;61hEE(;z
zUEca1vyL=^KKRG!XD7re{s_a-z>$btKDD>?#^6O;1kkk9n3}5d@J$pKSU@U5VY|ca
zb1-2rz}|rOU~N5p40s+t;tseSsQ$@-Tn7j)v6hj;REMP+8?$Z~P-i@$yZZ>b6eZ|@
zEhLaPedRj1mgYRqFu?HZ5H9k~b8J9yE|E3K25$1iq%2?*wy3n|3)T`EsSeMnw}^}w
zmEW*XUPCa<SO*3I6-lL6g?9(-sa-7n>yR2*Qurr5aUoD7HgyyenBqOyYZk2$Y|5-9
z%}^bl;ugL|q$#eZqox>MZGwl*;Gf|SL%G)=$0oX5H^eWpj3+&;c-ln$tBDi6RAEUk
zduasVi2)YF(DUrh|EEKWfRB}A2N49=E#&M|Oo*>)%`-=DusXKr!HCSZb6JB7#xD9?
zc7C1f?CodqmtpR>!W3i{{#8|fSGyVWw)Zl{IXfmsYx6lUo+PmX&Ran{C0*U`01z+p
z)3Y~v4oKpL07>(;igU@aefN%6g<A%u4mR?67)H2@<?YH>L`x}`M*I7FbTqP@OL3uR
z+C`JCdD&i%;-+j)l#avCh7bgwuNAL;)?x&bTyD4Q=d69^)0hb6Jq<v1+yI>JVeawf
zJUCI&=Z>cA)X?g|LkaZA4@Cex4}mP3_3lPJuM<}dZIQ4!XzQVsk5gDh&5GR_h!zX7
zCe+YXikWy-5&MWZ=c+OC>xD8UJgZ;!taLr9SaHBg!<=O^?%`COO8VqJ9tfe^u;6G%
z*(h-3H#<s_(Igm^8x&_HJ}UZ`E6_391nWFbv<?g2Nuet|p$(fQ%A?rIZx|T6YQyK!
z;B-;M1#V=G)r7&tJqk&VA=+rGUNjS*i18Bpb!;ceddC?){ki%8xx%sC4Mm-OtLvAA
zfewHb`KBJ2>`D*@RxJ7~OjRbU){ZU59?Yt{dsEPKWj<f3uDx33jRqDCo#bRv^T(yf
z>daO}U(F6n!|z^1opgSM9S%Cm`m@(38#-v^{+zr-Y~+LiV}FvG>B?^HIKhG=FAJVr
z0wB}V#q9xsef!ohxLPgi`#qH*+YHsb%f|_%?prjrE`jIK+f*FY>ZgSEG~9g2xT}_`
zMIuMEVi_F`aFk-;lg=mMg*J*}<cMkW0S;YL8K$_ITt&TnKfPp0*|g#viQ<Ey`14}-
zRo<*azbu!Ld*WP>(;kHvFe=ycI8U5Qu|K`iX2lFwM*Z~<(>I{r@r0xgIpx(P&u}u@
zx4GoT5qRN96rV-p<tTw>Dft~nVFY@b3_2679E$4lt$Z42#*94UY6;C4#R&}BEq~aw
zSj2QO#;O<<#l2{yqJqgsDK8pveU~*slsJZrdM&`X^rW^5x8Zo*#4#<7M5;f>l(XSg
zTk(^)K3A8w_}am;EOZ~+*<(&G5?~w@?|3u-ic50O;6CzFKyj8ZAQ2IQ4${8Pp-Wyn
zuEY3h85Ny%|8;xY%dv)a0#2WMVRppW%W++*K9j3Gn!YY$w|FESg|^1j`FAhC!Pv9;
zDE`{6Rc5+x(0ku`ewc?E+UrlBH0X3c_dufP@=5yEc;CY%1_AHHi!1qvD5iAANXD{P
z+IHuT(LQN7XI5-cahQ8(Xw_wF@h19eZng-N;Y~HR-AILEwg%Q;#q+0hDGQj_NFTw+
zk94a8(VZLX9?u=p;BdDmT<c3lk?RT`9O&08hp{kpHsqpDr>;O|l`3_MdxF_pPNFhU
z^O2v|{q4;*F^SMBXi_s)N+mh93O=E@ait#M&8GL>bOhU`YY*QP5v*8mylArSnxHBW
zmsW0N>cV)Y_PSTuJpYmYd+rgPQ;bIX+Htf;b`9;B;jr$<T{VyU6_rW-In|n^zb`Gv
zT6u3|75hgkos^mhKM_19JsrR9(NyKst9<(|#HG_Gu5B4G6b5fr1@9X-p2Xo}^}U1m
z@5FRV9)tA3XU^-rq*?$if{`e&AioarnYKW{F>dvTd2iggg=GfViY#_xnv*Ut?!e-H
z9+1Lt_1x)Dl+Wos$HV@``1z}h_n@1>Ge%Dn1Zxua-bDpoVgXTLJqLYC&ecFOg=v-j
z7c!r{PXEpnRxc5v$hlf?@lCc8Zqu{!G-Z7Kv;i97QRE5`z-eOXIh{y?g#MmWP3M@)
zc;59lI;%oVm*XH~3_&nQ-&G(os*JqT4##ZJ%-muA4v=Q25z@sIKzzF!zHuPXg)>@!
zo%pRF>6c6U#-Y-0=5WxAuMJlYt4Xv6>ISnGR9Xh?siXMk!go7$Wco-87mHC9x5GhO
zPEUxve0DuWC!FyI*U~ry6oD})p5Htc(cM>RFbB&8tbA-2>X0S(`sNZrpml<SGQ0b#
z)iVRX4OGv^Te(ETmdpCX80cN0x2^pxzh`&BpPV1-eRK=TVQbeSZc&dV=7~*Zl`E>^
zdNTr#JRH!D(@0qJ_u>#aq<-Rq$?MY2^cve~HuVF0b8|=Ggt}@@J`C)gSq1e8+q-+N
z!!VW8=6UXKawH?@ya<WNKT`S~f-q{)q!>}DfprD14_h9lU}B_%JG5D~x5V`r)>s1y
zkPu|belpl6;iLZX_Q!&96Gh+?7S9IFMv2@6?H_{jI%iN;00vW6QYC6k@ab1c_sgsO
zSZAHdZ%EvITR}IHMCySKumD<p>^|ZtRVBs;ZC4kQf`cgi6p)=H!m|{wN%`{mT;H2(
zF-dVpZ_2<Du)bG8c5eO;OtEkD@5=JB3jf#rOao9huU?yfA-G~s3L9(=k`Py+V5f^f
z+ya%ixY4LgWM)056e8?rK7W+3;LCV5&J3)i689TkuQKu>TBNQy;q}40%EX48R+~!b
z7MsX*TGdZD-m@!kQBi!^;-tQXhSgrh3_nhAuVJo|AzbzJ+Ak3opf=R@b@<3k-5iNG
z`!$yLKY#|g>6jsike@p1e6?*c3b<BYH_cQ{KC~T4aY(3J2C}=@mb6tdY^U<`Qu#_;
zIK*+opY$Uh-FbtsZJ+s#D|`o_R=gE#jJQ&qQ5$;>B`7=tiVvh>Ni>?YM~R8)DV)Cz
zlTEm>WS~i2I<`1;qN~o$%K9;KhIt?yTUfiN3?*iaTIr8+XnJ07(pM~=q*}`6p-336
z6yHGY9qKq*r`7%8vf{a3YdqfjWn9zz#ZA36T#(&$c@=92|D?cY-vz3nLQRI&%(8l|
zxI>5^Ok0?)z{cZP++Xt8lqsTKsk}?3R7DPfjB91&-)RYa8aQiT)_i`2;{4&Wfwd?x
zx=iLf4?ecYII}!YctT&*T;ABW=G>%|H5j@hUS0b6u5?GDyuph2#~L$S5Dt7~#fFS}
zH4^+a&LX&1`M#y({A~srAmm2#1C_`9$ifwKE(SYt*H!Q!mV?uSTWAt<W3N~HO+U<k
zFs^fME;4PEDAUS#G_LD9I;ikk<gL%HpN+WeisV~n)R|4~<TXN@Y;+KmiYOF?`TkL|
z3H#|dz53qPjc4aO0)EWC{LbzJU2vLEBPTMQi~VVK>?D2UzW&d9c|f!y9Y>Pxx&-^U
zk(016&0PZt&6lPjop;1&6clhBGiQ6J<!kN|L;I_XiK*5|x7_wQ(J`-!PuQ>V6_w5J
zKzqMrvWzm4pQyM-6tL4uu)~B-1|^JB<tTi<sV%uYO}BQ1rubTLlB<~FzC%Rw(_S#a
zA)2bpzKXCw`#&cwWz_4qYWv(-g9koucayk`g6~>QPsliEW~z@r1GAkZ@7pp)DR0*{
zuv^tlU2%Yk$2S7&*+;mE=Cf5DbtbpXON=Lg2y;X@zVf&9$-98q(+<9m?DDC<{&9(v
zz{`a9G~a(7GMzk1zCK9P52^#<=;K^HxPeM&VwTeAm?C%acCf&Z&;%^I*%k%mBVt^h
zcu`X3^9FcaF_iV+=|PSaOHb7*MYlAJj~JhWrCQL5s*Tok8-T>qAukf7+z<X~R%c;V
zC_FqduIbg?up~ch4o3l(ZUt@l{+%bPsUtEVZx<rd&pqC1kMcTB^Jm@A%eej~;m=eM
z){N75&f_>PVuhywoqtZYE4h<?_DjCz-_o#xMiG%$gC*}hl6b!ldFyYr9KpbHM{_zb
zy|Zjd=?5V@vjoj@v~Rnv4+o7~aL^}jIkw-@26*j_<vEy=S+z<>GVRyhGR!isSWlcQ
z-#^Wybn5xf;yNg3pqRDD%d+*3eengF;chcu<p^+g*gls&^0(4VYM#v1EQ=1ZWfMNo
z^IKRAj%)0f2jcx^`!2l{3st%w+V*NZ1PI*_PKiwISk`>QGf2I&b~eF@zQ84>&?_`x
z7&~6Jn{>YN*4NKz$BxiUZHinf!HBMHxqOVwR8J1+;mq!F>JE1J1nK*y5-H33m)N-x
zf`-l}t-sc2Bvrq7qN<V|EE!^a%K+7=Qeo-DWo_>BPd>Ksh<?62i&0`nYD^F0!K1IO
z9yfLgy?;i}_5*{Wu*T9?G@pu@ko$ekkIyn#PhgmbzpL^-?gWq3qSQyFcL0~4wM+)#
z^kEm}fUg6eo=jX#0<Zw#GXv^;ktyn6S-;c<`q03tmU^*zTxN^dEy1kEmz@x>v)c+q
zn?sA8*oOWJ<sysqpkek)_Gxw>n-RO{=e`wjjTyg;7={8M)TfZFAK6u-eJ0)1bB?Gl
z&t0V~L#>gIV4h`4-EGp%ooSWQZLtVWwm-A%7&EZrGOJSck}C4o>RJIHp)&7*qHxRw
zd!1vM0DI=!{0>DzTKcl0SYh9H+->id(|H;H{q#Wwh{Gwoj<x0pJ-kG7sW-NVIuG85
zpBg~7RfZbB7filxe=ESZm0P6CsoZ%gyoeB6i36Wx=i;oJ5PNr6rM-_^?yo|pauX#e
zNACN#k4j4-cra1^{5tGXeJE}Mru1W{!{DND0eHyNWSt3Bs@wpR2!-0PX0T^yks|)?
z>2E%n5ystG!?|xU*o*dF>W17|@DtjK)Z7?HRc__MC`gV~wk0g~u`iOW=JNMB()M7k
zw(|wdQu*Pn_tbGKSB9Pnxzc`FV2(;Fs)XvgZfd3bsigBH4Ze>1l%nV_l=4(Eif1-n
zU|pf!w8V<Go`S*wuZ!(6@SbEYg*o{PqJf07&w__*?1)dxnI}E$oU(jSy<*-`i%$-w
z&drjgRc#-t`RPeEUIJV7ivXizI!QZCrX6Bz`}?15pF!n`GgXsQm26~ZK@zh*oC=M6
zlRONWzZb()iY)#WDb<htj3BioXiqjsQ_@?~n7``MsP3Wf;F`aP)2boAfB)XStc!$j
zZxjaR+CuE6S(D?fkP-dhxtK~22qn1sfP_gombOh@NB}r7{9HwUJ&s56rHS5cf*SPX
zyn&|b{b~<^TwE;BqKpKU4&P>FHyH#&9DF6CWfKpFNwo0oQ5@8|naa8@L6N|>W-{)}
z-mcKebsT0@%THy`Z-v2NO`CF?-6P!I`tjHYP@^)<h|ph<gv3p?8)&@pMkdGJt$ax7
zTI=n170z&9q%j%DjCW5Ej}>C`v?%(-!f2j>j{-o!gXN>S6)7@VfDg@*e4@2AI#D`o
zlUZ!|e?^+ze>`gLn@~filWwkY$Tz2+TpHB)Z^I4j*GHV5-w#-r%QumHm#l!KSs_fz
z*8Lp$3D&EHPU|JbCoNOHVaALNAp7a}R@-5P0B^p5-%LK-X`iHtgTL}%-A1egDaN`w
z<X;i&C~arz%2-)+p#9wt2L{@xH^SY(K=YzBdEu{hK=1<c{-ddGMT9Kb2n#*Xp=#vz
z#i$O>Yd*VT?W5(nOvUwlgR8;l<9HDTlQX_<ZqE$|q!;#S#p~+S=kEt8wO-Q@p1b~7
z2A{cvJW*&86~GF??4-ZK^^W{(hBRzHV9*EG=K0@!)^43*&7K4?iMs#Mx~EG*C-))W
zMPbeciE~bA=wcfRJaQey`1NV`*ISbJW=R6AfbeK$I0S<}8eY~1_E-l;D*~bBfjGsc
zo`(Nu<y#5<=A_pp`)B?>?|$^lE7(BGYJp0O&t#RJ^L?n-Qb<h-uqOY9Rr!~C-iZj3
zgaiW-kr4~^r9zt#WBtLG<P(ONE<EDyYu5tDm;SXXobaNPH>scsJ`^Y$sH^(?xvUKD
z*#Wp0hlK%y7vK8f+I5BN`eZxl`&IIsaI%_ZbubOHPWZQt_dgnq)Pb3wPoy8DklEih
zJ#)#OGdbV3EovIht5~k`@#+Qyo7;B$L=7U$H<F;$m$LtRWDJlM4Fc#m5y|@|9{YC8
z|3*k)qBxfFt~y|9OlKA5&keirEOoIg3#09JmfB_^A=d6I#m}$ZG_q||{=x%)$iXfD
z1HEnw40`yThyRQ5kRyzNeR&*dx49+Q-m}#`d)M3e2eSGXW-wKOE<xlWzxF;~V<4*i
zWS2~8u;t|NZ>mF=5VOCt>0cBRM7ZET9jgslH{SmjsKAF_5#3v;*q#cGnwn&iYRIDC
z=P}O3WuIbXI7+RpZu!rbkiNs!cRC9yfvT{aZ5P4^?^kE=keVg5><1DIZf^5Lr?I}C
zjN9a9`EFW={kuV8aKY^<g_IKcKcI;J^$7Pjr|`R`M)&`pk)VeRMF*oK{V~gM|E<aY
z6XAa&EBy4J%ljB`37!Z3Cy?<!zWe`Ri0deR(X5!u6q{|@(`vuLzswH_Qbkr>{{O>=
z{0-;k3pm&cRV>%elvhEk^w)1S+fy6;dJ-#K3*<nOTjZLPX*>fx8lEl%5D}W}>mk`?
z#=zv4!mz;!>~LZo%CO*DMbJ-<5?5W!`wc$T@|V|l3U+}yp`XpIU9@Tk?avR%ZdabH
z0}%<Jut!bykdIAT^IVU+Wb-Rv3*%YUPO=~^r=Y|YFmq7|W-iv4EI?<5wqn^1x9i(H
z0`96KfEw0H9L>x$rl9?N#Fc@}2Go!i)}(O2!K+reE}J^$4Q0$~bSQ%*308RLa||ab
z;Scw$UV|E^9u{U#FOhZGTsjS149D|kDZ)=Uk5lH4uX>jkj<R9QNr$&fIQ7r%61|@H
zIjx#<Gg_C1%-$0gm(OGv{exNSOhL$pLVG|4w0e;MvMG&cFzRtrupQAO?A7D*9y%F*
zMgYR{$2fW+uH5nd?6iU7nYhky{#vps1}ndE<tgrUaEWZNQjGZfH>Ia<YlG}aGb)_-
z@slhBw7X7|3Lj(VMaDq_-{RG?WB>L+V+hkdB~Ndk{38F-o8sd3I)#Jd=_Gj!10>Bq
zlJOp${3o8Ba?SKsCGkeu!LUW8N8tpLNdnMhhjT_;n1t~nA5PhWeXlrW(xKyDF2X&_
zeuAxNEAL@ikw!u0i_pUZZQA$3^xFZP&l=9@&ymrC%kib&i*;xrOpdgs($ntz?#pu*
zN$I%fVC;crTx%8(nj8?zlS=3H6niY6-kK$%n(lQrM4bMaq?P~G!)7lg<~H4*W{dA&
zOr{8$-@;z<DLv0f|M-bRmgbzEY5jvAnAVWH5`7A+Twd}mQW!FP-gioob}mJnToY4V
zK3W7j9ZnMT6lp%k<LJqVzSDm64#BuA_I2Dl@VafMX`;EbUUqv32(EoCvh(sdQ03`E
z*1&rP5c}ml)Ag^C6!vF=vv}LBaga!i^8c)K*>Pg?0#6mmZWGD|W!D@(Zq^#w22`uO
z#4)|^WFC>CNbS8=7r&eK%+bH5<37Gh;`#g=Uo=J)>`ktCRtZC_Y!e^kzFvD4QC!_t
z%j+pw^d*0K->IJ%O1ku~gQNS0!WgH00)q|Dt|3^?wCJ#=iSZq0esGitE&st4-Xwo^
zdZz5+67m_WeB&A;<-1J%%v&Wlj<UKuL)OXL9OopZ11Q$EO7;kSSlJgCQ4*ifp;&my
z0t^Ig>0;Ayzu}h=)68O><+`Z0a4zV4xcxRRVaX35t6CDFNsPEZyz!6?LG3@z?ZVlp
z;2~3g12*3G|GCUM=R4H|17?1x(V<si8q+9YVx{=W<1S9Cz9D!5YuHj%k8r%A+?_t-
zXWON1zAI%`DGS1|BDY%+(nF)j0HeXyv+c&7X6}6LdSM=iy-36UDZfVxOWCTs5t=^r
z&cm5EMT94rgTiN-0$cDRPFzzv^teMNFTYOTKNQ@b9=egpz7*eJ8<r&-ot%<Ig`I)N
zP{NRgs$um`OY%9>#Qa7^Aa9R*1CDU{fysdlTAY)B+dfEh2;@^LrhUG?8wDhxWpXA~
zZ|Etp<iZ=*C>?Nj0lU3ZSY|ro%Q~G~@d_~078Rb-nm*1n$jooy)|-6dd>FGv*qF!^
zIxJfDuJakCp(1h#N(&!3@<^%}yW~!dPEnRtHlCo6BC#JWP4z&T^X1ze<G;32SDk3{
ztV?zFys`_t-YpeRCwkt#_>)md8gr$M5^uAp<+#&fX?B{~lI~SB>h5sla8Od)A$i_P
zQ;=9X`_WJ9tgCYI@+;K93QN?NrhqbrE3ziOHt$hbNE`f8LhhP(r>k#Gp!i`%D9!x|
z*2zDONSte0xt>$njg3~F>h;_-8zFb)HzJediwdjS=$sEf7kSp$iQY6_11UgGB`}Ye
zxMGZ~3lh#XoUfqTgqFBlb$pj1jQlZ)jMldb#oiQ#U`5Eb@j3FT6STuG&rDg}Hkp^S
zM<U(Uq8=w8rNlI_`Ao@4n8PN@gmmzyn~Y?JSw$k7<&|0+_`RKpyR0ie16i3qMq0jc
z6M7eJ9+IZe)}${YXKCYKFSSC1a8R%~Fnxv6*Dj&#=ZUIgZ6c@&#+|VY4$i+!FD3rQ
zd<qrv4-QKU8w%?Mz&?%yzkn;1kTR*Yf6W%GKToBje$UnjN^3J|OSW?t_z~>ItXJJ^
zK9?}acC1s1Rq=$w`R!ucQY>8G0l(u-KIAWg-8#7QeF~-<>eqyPQm9L^g?3As3gMF>
z-OmnN9Dl|xouBQGbr7(k_2EIU^SOJ%^=C_{SB$v4KiBNW2Tlv_8zB-;40yH38xIEU
zxBa>r!daiN!)XBf!;#hKzPEI^^ftKO#)k?k>}uq#e~X*=61e@=?eY^AnLY>?Wxl`!
z9Wcg)YbdOEz$M}k%Y^=h;NA1fJscztRrF|ml2<SM0-Y#M7C#iY6|7Ty!EBrQk6OJL
z3pp77qa@+z8W<rF-7<aqGMZI^`TBmq#+-&y6>~LiBs#+eV>U`(@oSGw`T0aU<+)i<
z%YFyhK*9u%q!5TCMTgomBT2>az(T>}mga$4c&w@v7G!8m3JpdWa{>)clRn@i!UW~%
zp6`j#it1jIIhF)vg}$i*saaQ@9`c*ddfzcUT?Q+d<Uo4MQtNHyK=fs1qx9bhpNE24
zx#gc(;-71Nj>DE$5$b$?OM8&5ab88gTn3F{VN^075++F0rUe}%Oo!YyFku><ykqGm
zE0JS#@8wHT`@*Om8~NDqWpf-^-nI6o$vlN732)Bm{Be!Bc43xVv)f6?ZSg27hIs`p
z_(U~t-SPAvN<H~8mSY=hmg$XHv>kHXcwCNGd6OwECLhB4CXoPFvu)<j|8CgW!xrwG
ztV!4&5aiun(d=(jc?Nv%o^=f-)-Jov4_-6`LCWn)sP4ky-|VWJ4Ia$oZ%|)FUnf}P
zTQOGKRqT5NBG6rZb{Ip;WJI59z~#AA3yJ}Ur9#40XdIF7W(2IU1&o0O@5J8z8%V7`
z_QdyOn6h7(2#}?QVP2A~s*X|e-OX$Tbl1K@7=2h-z95Mw8TOqo6Bmj?_}`F1n??3(
z;1kC&wf!PTC&J^dWeL7~!bF#sZH1Xo8|3km^5O<%*iil|0RmN(`@wpg$Aj?d>XZ+u
zST<6&baT#yf&9O?ktM~C!UL*Z>^~Z*#LJq(Dm=*uho_A>fQ&Zp?W<z0MU4r`6}tDx
z_VbG*(Z=uk>FtLPjo8H}OyL$W%If<uj4Nz`6o`S>3*p;1!4nvc`@T|0dB(BjYl!`z
zfhL<)VaB)<9)m<T8$9B{Jj{BdGCk5IQ#t;kYbNuSJ&bU*GeyB5GehNewFL9k=TNDt
zP93kD2kM<qYI7xO2i5P{KG_6JP_L9pV=OIl8v+3?=Okj`(|t&H5CK&cOSE5f_~AT=
z4KRQ4aXB^Px?ZapzI~Byk9FT$@V&~%xy(LRSK9b7v)^QFd{rGesr_*=q=<+A6Ikv~
zm6`nVW3vl#<@1z!cGe0k3Rav=7Oqs$J)0)>qqw>!`swgslO_w;ifH4=(Xb!)&=y<n
zUG6v5o$kw998-dp(L_?RxBV#E;Rx`%l;<M(lv2Y89^*tz-fz<CLCd8-Gf|1$MfufT
z%*2HpJ~?eSVVssLSFha{)c8NsO0&Gqz`pdqlf+=GzIyK48ETkDYHhNmCeHK@77Hc&
zUDMk=TEnJM8EwzrFxsVZTIF}-?9z#~GG~CVQS9&y7%C#UzF*7drT(s*eN@x-xW401
z1sdejWaW)Fd)fYWDGruF<<H_fhHtnTrKGE#Bnizk%zbX1w=D+A2)Y|~6nl0pFMmBT
zq1Z5u)HtedEy)L6sn~e*?jLxK5R$a8Z$*M2PY~k+jxIh*f?c=qq{MxQyowPD{1>ar
zeG&#n&l<tCBX9u;2ScF9G*sAOltIK<ETQ;o2e{sGN>I8C#FLH8h!ctJBI6oXytNC+
zagT5QkUouNSiC4W2`MV0mhyc!N9_@V4{UarJAeOACROlB{BOM-XsdIZnIG|!s*s5v
zJ|=kD>w6rb4HTVhTldtl?HsF1p7aKGjmU^W8U*g6UOg+#duAp-?ykq#MZnIA?FJy~
z)MO;HddJJ6V1q_4rleO?GUkm~ogC&bx10lae%!&#NC8G4?yEh`0Wfo18Sj}->E&xz
z0ORz#ktD-_%#5}>DhA2wa_uDZcmEr|l9x1{WlT4;Pop!!n*LyG;uYT++;E?O6m$Di
zT^-nnvyW+8LHd!ecvE#ey^X_1IgRx5s8Q%$;%()pk<G9Q1E&J!RB7R3=0BE)6eE6V
zD|mz<W)~Ux+U?x)is@zL!cVlTzeOcm<s(IbE%|)*<mZZu_ZhU$=x4bdp-gSB1s$$j
zyX?SaaJT~l11FZYYjp5;+sGf0yQGF<nL#(^ez=9z#Y<ZmVS2%7JE!1cF-QLP2+tl@
z2?(k#A_wmH^_J2^@MdP%*wIRCQ4EX<=42{<83JGFCBwvr-9)=NY<FX#4SuNN<9G$h
zYZO(^SCbIo9=^<R=X-mMuq<Jm_{gT({Q~E7&9w%tzbZ_)*j<phWDdeVtUubM4pQ5N
z9X69kUpK+Hp7Q=izs<%Y@J0ylv9F#|@3i<UmFHrFA_D!2W2CSZ;d%FYNg(0r%Q<y-
zcPm^LcCX508CA21C)O!=Ng9(s+H@px_1JG%%k3WZnv5H>7uL8WZ0h_#YtwPP_((d0
zZeM(${TD>Ag&2Ky_zQkS^LmU%ej2_*Sd!;kJfY*RxX73jkfX_uUC>+VV%F3~7i9f^
z3VZKpIJ-4$T*@nwC=o&QB!pM*y@wFddneKR=w$|@1Q9I=qIaTqGkSvPB}N!s^lmU1
zV;J8!=lp)_TjzDw=MT$bKl{G#>$>*+>@wTk`+`7NLdomZdtj9)2YbR!jW%;Bh)u2l
zw48z|<h`3!s&dX8l-vgwJ>1`Zsl@)v8(#i+iUmn(iw^P2BG%HtJW(>Y3!<>KU&4E7
z{BfLC4|2aFkJX3G=vdnUcX34TkoSTqG?wcr<gEv+Fw)`tq(k&?%fHQ=cmL{#a`>o@
zl4knu30n)*BN9#1ZV$AqkdV`hZW#*<(~!VAkC|}=e_)-GM9334DUo@*6pjeV$I1f5
z6P6QTEyMLFpQLvo=|z&#$A4bet7j}XI3HZ%Y9xJW%@B0AFWPO9optz=B39(Kqt7K~
z=93YV&O2*(_dEa%dp(kAE(K>KF+pUO<IWe9;KXv}V_Zy^DsbE_ESjD4;7fW88!qc6
zj*09gvPj!8QXI3o4pMq1l#t|-6ik$(vqmXi2Q<A{=}s16aJ*9}5yV9&Ju7}IC_5vw
zGTVpRYq%}^N2Ccxou1Z0x*)@r-`jTv>x$vEOSd%tU`1DD2bXFaok(P(!IjQwU0GzZ
zv$VyPcpB)E{`f^Hw|A!YPdTUey+aMi``M6yVhgFyB+w+=bHNAvD>0762U20YcI(z$
ze3Z$Rm);JKaQhe9p`f?{@=WUjtHbk6uJtFu3a(v}@DhYgXxXsgOmF72kl!>H%g+pP
z5oC<e)86B*Q=64KS14~1;V-LCaT}4x0k;Wx1!wWZKZV&dl4R;_2N1|w@ub!8DIPyn
zAQbnoL+xD`Mj0r&`mMW=>SQ$i!_!ST?~sei<*WSgfUz~!B_jAGArWs9X88ev<1PhT
znX&bdtfhg&Ob*t!!997oy!i`v3V~Uw9QQZvC?&0?!*l*S3h!Av(pX>-?{icRsa(*U
z!VDfYW{>u?)X4fWI_bCHN^L)^0&99@*-6z&i{ud9`$h|1Q!JJGkYvme)Onj1%`;;2
zFlw;fI-mV-G42SvIhgy$wbv12u9oVetHIo*b(;2%>88yrk;Efv)cx5_SY2Jnod0ab
z#Lqj{O)O(gE*A&lQV&6^PJr2J+%*nXKjT|<pc;>edqKP3Br6|Fnr#g*4LIvQq!GDA
zH$47JJduqwu*-_usK->2#p_EEcOgHwcH9O4xwCm)ZRcYC;na9lev{F#dvfS~o=1~9
z?W_USnhUfI&LgFr>fbJG%e|0_VrgvIdW)a&J3>W-7b4-S`hAeI+plSc2NtDvfG7JQ
zep|tcGd}m%^cXV4b)Cy1&0#ImlT~M83!MB;{OKjnt>rJT_A-g8>2w3WtH1dW^o>>Q
zLEgC5o=T8+IMXu=ucAh=%nxzY?cHo$&y>E6OufQOx@(m9x&BdPGTxp0?}+*zax%U7
zsPT+@DvjjFdfTI6ke!MvAwrT}i2LynsN(r7auGDTKiy{QgIH)owPW@%ZK+nM{Vd51
zaNrdZEo~kiRj{$j$9CLndSy)Cc<8LzMlPFJ9GKvW-8psDl`kNwT%2Bu*}OE1SnB&S
z9w<Kl96CKdDhLvns;=Iho+o9nqDY`BPvC2XJ8+XDq+f@Wq7_g4I4C|;st@Jf?`+gj
z47(GaPH~~D;gWjGYc^tRE_vP)S@M2v7ib9KYn%f+AM6*PN2fK_pd^DC4umeRVJ0DX
z67=tHKb=!BxRw9h7jztuDyT&}Kp!j7L~Y{TyO;4Hq~9#TYYAr8B}&+fJko=mLY4yB
zLq>_U2^H_;Dq=I|?IMLgS*o$m>4Wkd7kHR*@-K69JuhY{pfSOUa``D9hWXqzXl@2!
z^gyM6N^j?xw{I`x;=$lVDvygp8+~BODkR!UjL2oiQahC%lRHD`Qbd|CSt1BTiq|#Q
zSVU9B`x?gZ`<a2;UcV|kkBFD73f#^#ce+Xt*C@2x<D%gS2%U9IUV8EIM22T&yvhHC
ztNuZcLqSntT)Bo5osO~IQ)v%3b_Sqj$}R6ZZHxv&q{+7on|a)S11rS#lYa~golh;%
zZ(bOcl#hj{w9-0*O;AOQn>1d#hNUw#rdZP_`~*oCuu#plX*ZYr&(PXAyDsITizb-a
zH+9zhWC@1!rO(^?b_ccNs;p*NJ~_zj5aNEhwTin#EkF|zr%RhZS7FL4%KdMjPr0xm
z-^W8KBW0gfa-v*3U@f%KRyNxy=$08>?->W1`Rz}B6MhpXHGz5ohK18fF-dOuQHMs*
zujfKd(GR>dR5##f^Dgi_AbV#%Fmp%a7_Wrau4#ZeUhde%hQ7%Ycq}7ABb;U43irWZ
zY=#Yaz}tl1XjcZ`yTADKm}zhNvLNr0<*<bw+~hlZ%g8IJzF4>yXIcFDh)X@hx^0mN
zDB057nsvE^XZVrXq|*Na%}#qKAa@XE+Rj(}CM_yr2r>CA=tqU%@vmkrgD~frH}7Vg
zK5#-J7@q~cMzuT#*!5wH^TFSQpT1UNZ*`~?c<}*A?(hAzGS$G3hSP68YO}%L(nsjR
zP-v!_Y^^dP!Yj-owsb_Oe04?d6@4HXgXmm+PJJ|6d$dSisS((K=x=*Gy0R3C1#kFU
zz<gb%nK~d__;1D)F}hO(0)<b~bo>6Ax8Q;VySa|{>Z_~h1^MuLPrAU|*<?M%S}9;%
zKoxeOZ-0BY5Uqu{u)t6Pe*IjlwynO8MB+XMvu{>2voqJ)1rR{<J4WEQVI?o)4IACS
z53VZt0<6Nc%ZR%dCFWE_PMsu>0iI~fuiUks@#K(l6z+`sn36hBe!~Kq*5TR&vAJZf
za@3C#-n8`|rJ>zKotr;YT(9e_+7IaK5{(jOr`tr?c$>}aHng*}KlVIh6N_?_t?P`2
zlB}WbS-(IW#|d<B$Q3(^|HUt^><Rl=Zvhc~F?hHqvPl(UqZ~GHsvHcReZSgH($(N)
z4+F`M@PU2k2`=p&=C?mq#m{pVuY^u9YESAU>Unpcf-?&tpL}yHGvVlEm&^`jWOu}I
zV96zBPjsaQ&kHqG56U<J&H1Hcxk@1~430MEym!*tY*wdy=UD?qsVeG*W;He#K2dFA
zve5UgTHY~3(A^z9_!3qcFJa)Z0IK=MMr%TRWcd+)LKP!60`qV(4Q#O>3Z6MyV~eX0
z=^muI|M&3X*g|rH>?vaFPz+vm2q}Ro0Gddhs-8Pz9-4%np2FQ3iaMN7f5A*73<4p9
zw+LN0^FL9L%TqQM$$f5I<!M@;42y2?e?_fhcm-oCs_$$AhWI^bw+pin{;cZT?<UNZ
z2XEb9+evF@9X;*GRqwRbKSe`R4btDj0e4g5L5`wZe7(nDq;vevEyuKu$ov3uMw19L
z>OTfad7;A}%O(g)CjWSsO`nM|YS|<wb2`ohG%oPnR`E4eP9v+@l`x9KNkv~(&aD@c
zk5TQJ&Xl%|hxqB6zbfB3h7KIZlN(itJpv<L`In~br~SFN*fW}F)}tJ7+UTgqX~O}A
z8iVJ}KJND7Qh@#}6Kb+W7v0zV+48JoU=wULt>oEQ>KLBx{Kb8ZoUBzcoz*@(%q`u4
z=^<@ux<>X0P*{*|2QcW>GH75s>Qr?_2zfrram!=CIj^|N=3TLSvQIqJ_||GWQC9=<
zvl1eH#2^){MgO#<S&-hE5tNYOhB{m<Dv$Sf<d1AYf_G$JfoKwa%~BS>CTlm~;W2P2
z$;;~0QB<A+>5LzIz=nx^JRqivj%>kYl-sOvBdAJNA`vK&eeRPV<wm9yEy==ZhIV6n
z=AHGAguBsF#&;f8)~^F8e&>r_QS3;Exv>LNB7B%Q%80A)twyH{fJ2*WJ_TmGmN#7R
z)H0by6(H&821BNkxJ(D@wg*1*WLh_KC#)wtJMizm|IU@hjl3R;r}sYHhUX2}ug+J@
zCDos#)!CzA4^Z1m;e3T*))#{h_(k`7A~pr9K<MT^VYharjym=h|G5wY;~v4p@Q}5F
zT7Txo0=~0KE@3&d(Z=G|d&dgJl*ae;rU<QY_w(COe9I1NQ8{jFpW&OToyt2Jv)MC?
zQ|`?3wmZY8ZRa3HMYF4&O!L|cj~+X_@3XbJUD98EN-_PkvE$xwKi6D+pR~ou{?W+m
z8Fqcirn6?e7A<9@5RQIPXPH6X6pdqBVyg-O7Xma$JzYy19PuIE{kX*G@?`K4BeH=P
z_+}=JXU`^p+RdI7sj;i$0w0Qgx7C83Jdq+W;V$7DEL<;FmAXIblg+*tb{@UmMC%@I
zzG(h*#H`~THeXiM0b$QHlzH#`ZhYCA!l0?pY5&kteUpk!cz{2k5J*(Inj&T8vhh4|
zVNb#?x;WDBo&3%u&$_~?yr^99h`9FT8~WN*5xNuxVxnW<;-0StL%4^MMTD>6@<Mek
zamJPu=N}=Uz7F8E#_CuT@7$a7w;98wVKsvm`nb<XLJ4504b+_{G5Vp0U4mW1^#P_s
zX}J*+t&Ds&zi-)N(|o7*D(f*ugK-~QV;K};{jnr|WwvcCd30_{eiPlex}d3r<BLlj
zRh5MOn(I%LJYvs~-vkNXxjGU_)>l5!>rtP53Nu?E-Idmgov?&w(mpYwGX;ufXPt6u
zU#&S{Pj=R%LPO6X99K%b-wVk1C*+1S5A4<JI){wEZg_DYzQrhrdGCl#ZoVWRGZY8G
z!v;Td!D);hE{gaV+N|x!bRbh^3ttWO4xbKi+=>6BgdI{SRRWz)FHU+&Pwj9{6mO2C
zF-S(sB*zo)BTgEnW|xLDwWQoMg^$=3_xYC7#t3-%inoTLk>)#95MP3cOSMEv*5Sy}
zlM!x1>~xLj0c>S~sQFH#ud@?FpPbkp${R$ZKTq|Pt6*uv>D?^-B+%UYyzfX#^`7D4
zzoa9MS;uS==JA-{rAy|Jb?^sR4Y)8d91A&wp1%aHTNLmY?ZnHr)t4*N%%zxt-)r1c
z^4~u9@wMfN`r*mE`diq88(v_I^uEXAjL>RfOzoG{pLuw)wJTtCW=%t=>c3K~zNk}`
zSZw*0pC8_x1L17+G20X{Nz<n4uy7;k>H|@RJMO>=e8fp?sOLO?sa6@&jS1G;Zs_**
zp2xE~G7(Lym{j^yb?DRlt~wO6z}zO@5>ioz>j6&agz^1GQ+lVf2)!P0F1~$!>g9Yo
zxoKTGkJ)LvJ5Qb;57NVFh-ow*`EET;jlrJ%g-Q#bD4j!F3ekxTRcmS_P3DQs&mlgC
zLBbZ7EePR<fRt(y#!Jde#ODT^JmO0h<%Cy|lGYQvLXH#@)NbqmIfJ2>%qXL5VARGu
zZu+{nLx_5obX<@9*+w2v(6$yJ%)UdBQrgMKG9C~wu<;<xmQxll)L8jjZc-Qls#hsv
z#Cb1qxpY${mno)XFoQ-hWf=@b*&`bLAbuoLRFb#)iV}_Nr--!?9#8ydvdejpLPkf;
zDc|nmtM)nt&!a|<*;_3~o20c2X9knBP$c6R+!g%;3d9b(uwUp%Pf2iUTsrz~r43RC
z93Hvw@CPlz?5a?qP-8Fl-BF)~NQmzAbK;e4htHtqL}!}}d9gR%1t$TV&6OPs7O!;C
zb;$xOW^nNiE3~a*Z}bOc!$grpCE<|aV9FRi{I9X@B}0Q><E<2*d7CSr^O~(L>q!ax
zP#avw*sbskQ6Ba*+=rOKOIKXhejdx_b{q~Wp`;`ml$}K)D^0I#2219Y7L17;JGwSJ
zDfv*s6|tPg6=zKOp-;RR&gmvRrgg*36%;Fd2kSr<#pu^vT&IH^B#Ex=yCSl>{*ayE
zfghO`ZdIn|JSO5KRLS1-0&5@3OszWs-t!Kf4OI%K0;0~SLy$Phw94}Sg<&M?;MW*T
z7k8LKP%rqj$j`}Cj;@aowMp3T%75&U(u?oH!4P6RO{RnR%lPW@dFUf*Z*81V?v|Nt
zG+&-YA?{l56wqXO-E$=JhbBYQ<5Zg756miZgHjMaGeHq<jMl0~>0YRCP#YSV3{>Ca
zOW&nLNK-XMnK>^UPX8{-IJRX>cQ#iEqVfvb14tlZ*Bi^{u>n#`R?Z$TYU}m|E<zj)
z)^%3ePX#FN8n`ju`HaQ>BnSBZBE+q9(q#R<M7*aN8X4=I=oKePFP9@VQE1BVc%c3?
z;eq3{hjbZJUa?oYT6M;&CnR0|?<oH`qM+1TJ=W(FPG60Oog&_R<<<|kKZ7psC%ka2
zQdBK9RkM$KGEYM_ExE~=E!7X;(V<vg+H802_&t62Fh(uVI$cCN=SiZ(T%H>VJ5Bs+
zZ2#kIeJvt?O@^b3TAEwfZ&t0OOCX9KNr3GThjxN*0q0@qYxrr&hZC4Fy?2kZv%Q*d
zESTEcdvfymwRl#O%C=+(go|@^XbtfpR+i-=90#_7TH`Q9J%Wb9?1U;mXsYXOYSuc&
zm&$$sFU6&9IdP*cz8}$DoJ2<9I0o{~a}W?n6k<aieQF-W40`S>v#byUG0n}#&*?H3
z^x7g@Ffz$@1gEUtmFE+EoNmfux!lY7Kf5MvyJJqeHu)=U7bDkF*y+;MA~T+f`o6It
zbD3~kEF~$M(&wf5{-Bl%!Y8ir_3Y|6OLYfiSr=a}6b7wPCl<K<B(mi&Ke_Bad_;~J
zSHgPAfguT94B<>OhBPlbT8$2sRmPq0Yfpa6a$-&z$#=^{R)cuDR{I6+cvb&sC$yN^
zyW<%7zWdCOH5v>oqVgr1M$FTTufb1LA4;<0=Cqn6W*d(OWO&}NX8-~uDp8^FYhe2x
zeoXf`NT~e&Ih}po{2uOJR*x_uGD*;=^VzUH+DWeBJ%@YC0ID3TP1<t`Q3O*{bFSPw
ze%ck^&6z;ieW#4rb%KRdQt!Bbd+}b)iq2jDr}Ol?13fsWdQ<c%FfTGy<WIRcU1zF$
z2Q}(7EQ4|L9dms_*MKZOoKm7p7&pT`VV;S_8vYe!HA%+R(X}%(zQH|aB({_JK3~5B
z$=*d|+*(Q5(i%AEYpFg;E<C<Ix8%GJW9i%4nwB(%D<rSV3vo@*ta>YLycS;lIjZXD
zLevS`$Wv_Nns18tf%PHOYhiTM2jaJn0=5=y01L+uG`25GJacWs_$6)I>cj<Ia$!Ub
z4`7T25XDfnNlG#;C%>;h+Vc13Ej=<H#wc2LTzSex25BeGDmQNYO5y-up2T8v#rV$L
zWYUyUjpA8>)|ai}OH4~!(xg@i#2oXDR1!L)ocsavp+q0^egFo;E;ujbZ2?HE;<9Uf
zjvMrE@3-%AH#eaX^3rWKI@|<O?v~sS9~lmP!G6payR0U1d7R7x(%7-d&d~XGGSlD}
z@EL#ga?P<qEtXrZ{yjHPcqrK%l^c}m&rav76k)m+b4IdR7-ihRwU(+3e;JSh)6nva
z+T`pXZ5>4xH8rnNa$03eh0kvkBg3j|k*)n;*(}Jw)5vV+j7Rgd!Yek266f<{u`y6p
z2A`KY55MLDCGSkH-~Kqdwa!FsZ)%S8!%8s;U2y_$9={(y;+6H!&w$PNqtY+>^K(Pu
zOFN@BoDuc|@pT*JztGT88tPq?P(X{x<Tk!-ZzG*qhZ)y6L^<V_w%Wn?RSdh(xyuTz
z!?`KbzFvfkC)<_FSnbIhy26u%{K1MfVPO}gl1zPW)3a*x?BmUG%}YMnlv|Q*n776+
znSvwx!E@1J*j#!owF)y=!-8@*=ezqGE$+=^moxSV+O~p)6ggN?<zmQ7cPW4pdn{j@
zV-+~WLw%?!Cu4l8>tfpad|6ja@{Ru<)R9id(o+Dt*0#^Zy~PVOpUWMz$8FkaPlbQ(
z8|oGNDOpwC4(i}M()k==qbx=zl9Tey-ZW*Fln>bhpxLrMXi*i-L;jI$#=aFJ_&yn0
zZD%#l#*(gVXfu#d53E$4kxIz205_x@jsXpqfj=&i7A8%bC7FQmwOvb}`NylN0&<0^
zTTFpph4!S{zWiQrc(Iqm6R9O}`4`CN=khF@G^6!v1q@S-q^Y)*LwR<2^|EvGLr=q5
zM*fGEdj{ZjqZzyy`jI%qaMGo!@7X6-N1OSYnW`(R$zRE_FJ;pbQKTz@bm!_6;WL1W
z>EAC_GIzW(rK(3~>z9jUPWzrg&GlaXW&wE`n=kav*al{+NY@Rrwgi50MIOsK#5o-g
zwmKW4&ktfTx4A5!9olftK;Uf0=xi>e0DbXgB?JXaNyLVI@saFQxSGhnvT9WA%Ut)r
zUw&obOj!JW!hOWq@_H^y2yaB{+LTNa8EE6lS2k<Zg0ovL6iJCe!W>bpkohzaxL~f7
z!>0<ZvM#(nThKESXW83FgQ|2CDHJo&`{T|+Brt10?qjI{9n#r5!xSu{_RpO^7@p<V
z=GtSHD0A+QXOSTM#4LW*ckrXI9MwlQi-kO1zS_$C8`0_Ijz%k!D1N=o-s0>CH_kvx
zKt^GIU}y1sy2~u1iv41`PvDH`ugpw!&wROdUVA3i#)d7<euz=lhLXogKliDYX_w$0
zHMiw5KBOKno($P2w2nI}3S&jE+2$p`V%O4Jz0_O9Eo7Eb$`!0dO>O!C9VCN;t=)P1
z$N>gc7Oy4?wU9b%kNM)HM9s+3D_zd6Lv33iYEvziDyl|K<UDGFe=r6FX>$N;wX2W%
zxZ&H(26HSx3>BiF42>h;i_v0TU1{}9UVW0sVc}wm@(zLZ_OiaNawd9cI_W3HW^Prl
z|42BK>YGGk**aY;<{cWVUHemxIrysIbKX*S2S~>yA8vUiTxQNmfAPNEEI2NT&Ab>9
zi<a|z1Yk6I;)vASw3laCWoJ*A8E=q(&OX|7rcWHQ{R3*-9&fbp2Au^>`tZ_sP9|U<
z#$W_UbctwHg_vjI_I@p67q~^S`f*t8xlyA)+lIK^!XkU?r23N-ZUPqS?h{JxVg-F;
zAtoAO-=8a6=aPX5<-bbxuS#$;jh<Kg$0}Wig~3|!)0JKM`)3nxaMKNHCCL26B#on#
zH(Yg$EVuipN}LpAZeR?o1gEPerbupVJ|k>j1~`C{sVPgK1$fYn{xbnL$F#<$NQ<kB
zn3A<rE}7Nr7P&n3Eut#}cQTCBe)id6avdf9AZ=)P>&g7tk<Aojb1XKd=x73Qxp7V%
zhZ!@!0DUifAu(vKP0(z^^Pzt;2f{jHJ}BY1<bX(;Qpr9@Kgy&}T7bF53409G+~V%q
z>sh181VQ$oS2NpCC5d|4jiG?w)2a2=({5Mf?X!Jv()M!HP(64rh(ZNxoxW4Y@EvBk
zp(P||0Y0Te0qG2-n*_>6m*<~a+M33T(Qdr<7RzWK-^;W*!)5l-(OOU1#W$Ce4nkyA
zvH-7F2B$^m0{U?m%UhaLueBGpJsPG(k+EtX%30CLyf*0Yr`>Qbd6G9gz>d}PLM&lk
zvpXyk($xC}nOx`*KU;7oQH{&=Jizu>*ubYdO9>YoF1;Co3*5@q_lL)@wYj~0<*A?C
zR(BrSV8jlZc5L0--Y+>fU;NF0|7`yR3FYmgs$rw2_|#(U1g%@)1#Im;d|Mkew+8N2
z%ifC9fcs2_)BlVml8?#en{P2-m~lI_rcAjS{y3Rpl%=G!R#0g!yx`>+hEYBt%-Rng
zc&UQFQ*<=+O`t8k-si`~#>MCdh;)$6u}h_q6nGN#a7S4|1!-a)!=N?NUozrm8Oqbn
z{1Fm+bxd^&oF9?SmK>65@TBF`Ys=6wC6n$xpjF6K3YrW1ZWsrJvpFyFzc=nakhnU3
zy^y!_)_xM6I~m+31zw7go&5n#a(<qGP1A+Y`z{&39MPL`O&&cm>wRrQ@|qaNrqe^>
z7dOI{N$XpZ^}hc6)Lpc&i`QOddxPdg9F(m3m|q=X-ZE$O@>yUXDoVVupz-0WD`H?=
zl5MWqxR)A0Pn_ezReoQEE>lnMX>31B>*O<F%t<0UlPIo?Osn#p0L;;LJ#$?eyL!40
zIw#w(ZbW`@;^|hj`n^-9@rZ-&PWaY3lnQPPf>6OH8|W`t23w2(*4a&GtC7f)Px;O@
z*0tV-<o%0cmA0b+Q?p0%etOxu#9xK>!=h-{e*f(-IOBxcHtfmfhMcG_d@kcjoX||y
z<2)=%_l^$?9GPRe9Nn1)mx>M>lE&Zpd4bKIgl7_!MpXQPdOdo|GU~a!L~?tAOeF93
z!9g`5a5<8!?22n0hLXd)1i!vuMe{Ubg?E|NeqXSxMz>7PS1*|jqC`U<W|EZku3xk_
zJFQiHF+0jLgRWUFl_|_ANwBEskP&6SIF!n}Okk?IJW3UQwZI`A)t9~ILymY@*u-{G
zm6Et59L~5F89xipR3m9ZD!1M3%I+ckeBmd*;iXNNdcpR5J3^L9!}h-S9ItlfWfJ1=
z`RAG>F4NznzzZcKkHSf?CKbl#KHDD3LT8ART{B#P^^8BB>?Epb6(#=a#|>HBf%(bm
z0UhVF`*Lqdvc7Rj8<vlN77vqdZ31l$HfAT_+*corU;zeORDYLloSZlub-N|G>>PG)
znhpdcQfw~)1~Fr+&0b7?UuLsq(`znfDNFR)XrvWt0~v4?J9&5W#1au+t&Wv6DrcQs
z>dkfS4`Byy4utpm@h9UV$N1hm-ec74ozUr+0VWRoJZ68)I2}ZpOxOf7xg-B$@Zv5H
zbj`Mh@==fs=Useo4sMA?+MrFMVCyB~c$fuvhIflVAntp$Q2-kAJ1Gtu2Lnl#V5-h!
z6n^E#zXeEDU{5@3ZJ>Z9xSX^mWygui2^3WX8~kg6oI!%wbV`o9gbMMsD2oEs)}gDf
zq9!C%1nM?sxwWmwLP7ru@Aji0yt_uhc=!;#U$^;Wa`p%a%PLtI^tsD?%HqIB+`n0Y
zGT&?`Y=5WfOXeLvCY$hajTG}^QusI%%TF4DcUK`?#?EydbE_1bF32etz4)zAd$G8E
z?lK&PVJu;>X^8l-m+K;Fj3_6(lO{wk?7N{D5K0ruN!8DU6!darenBhq9~VC2T%gdx
z&y&e9B{-JG2zqUsa~<UKZyf4R1Z?bUYA<3ou)B?#(k=O>U*sj8;RV0CE4Bfa53uO>
zUX1%mXZ)Me7F&|GYA_&jv6C`!2U)3QZiLQ5G_o=3KJTZ#IE*(heNCpW4Zd2h?oja@
z26O1~kBZBQ=2hZ@k6W7$OSiUj_peUoz3B|9M#O4lmEY3P?&%$5H8+sg2N7O_gfqE~
z+kGv|BIkOJ7tD5l*Z4~JBv{#{%9Rsyxr-g<6HJ|UCU^41(G|^i8(6aX6IfEfQ6T_M
zF_BhR&2}$3ZzD^h271Xj2LdaAn_G124ZxjZ{StkKCN0$mS<xfuKPthREUGPHPs|XO
zK3)3KWkuLS28XWMYHHsGeXq(kImv}faB%uGkonK2q12#Y4;R|I-;C)reDB^OxQ5a0
z`Y_VSK=dPi@Yvh$G#6QEKrMR(<45=ApM$npW`7@(hzY@oY2EJNn!kHO>bqj?g~AnS
zF-lt%t~hTu@?BHR<o^L{cbxE%;|8nGAHn&5*Y4-)ke(<7IShlPa-=*KK-%+%eCc>z
zjzWwk{~vRa*W!WU*UIF?5D@MyR?@cu+Gfx*IcdjOk}^YcsDf%BrE=uVo1gN3-G1~p
z*;DMv=tSe*tS)L~-r?O9_k`>fTKu;?*t0q9d@Sx8vM6K8#ASvU?;f~O+xhB!$~M+5
z4!G8!H8c^4&@9dl3>QOmZIqN0?-&Ax94gaGON@8IAIDvf5}ZRsEVxs)rH}XjEvo<g
z^55)T2hBnH{f|&`y{gIce>eaC|MKtl)M~g5u`W`wYqI|h>A!gY&yU{?UIhJc=CDtP
z{%6?yS0jN891!7*%y|DI<Ntp0U*f5e#|xb`fC;&U-PpaI2yVGcq^N@TUq1cEh~+iP
zH3fUO-S$j?70p}mm)l=Z(aI;*e>x;j3Nd<RWuv036Lj1+1t!z-`1_y9%Cri{*6@h|
zdxWM(vsu*u+#?06arvjGuFO4(@g|plx!@90L%IUZlk6WUKBar0^NH?AvZS<}-|Z|N
zB{tv{CF%O<GNDp-H7T=2lCq_x$nEI30F%of1;?qbM32b;H~I4YBN>i@Wwxi8w|95W
zLe9V2BD7axTZ~Rx4?m}s$GBd|-bmfwVLxJYG{}Opbj>wv%?gQJ0O&8<egSQ167f{d
zg2Agd1uzqfgRD{8-s8oG9~Gh*_i^g`2PtEUI8rvp4VB+S|EPHRJLu6rn%p9&e|CM;
z_U+5qv&Z<5v7DDyAO1<;J0{$+hUWgZ^CM<_h>*q$&HuK9(@@TH9G-%A9WkI!KPoQ0
zR0wat=UWWU=NV`Wx{3$)jS252Ej`AWQpW3A@x+aG82W(XhOlasabZnO1mu3TbY=e#
z7k7gtJ4@W-GX_mD(bV`51=vr(4H8h`$3>oR2x9zO93MjZLH*ZF=m-S6DaO&C7#Y3#
z;HY#PBi8;+>Z}R#8w&A%yNBHa3i`cL{uHTk!;>;OTi1Z#{%sqf>*hp?E(&{R`|N%%
zdtr>D<POC*_7o?L6=oIvvTVZw=4_FXxT#;Hg=*=1uQU1FJbTDWZip#JMiUp*r^mEp
zI6WrWW|vp)0ozKa%<7jFx5ev5$_F@O|7*<tD~HLn;L^gR3$?R`zZm6+PWHtN*><_*
zh?aO_r-%{IV`Ket`&?RFu%CSNMw`YsMNdv{zjVJTM!eg@%(#>^jVn$6PoAztQGt7>
zI69s@7QF#~+$+cCx^t)H5BL6=k8mpdHF-Tm8-S6tH)brEy6et`+&V-5so3lHh2ymZ
zjN0UhZ+30^nx1r#rJILjy&z5;&J|`a{)cN$xX#m5GY)P-9PjQ-+V$Yn<BtCyt_k5f
zA7)nby%|$y)pv90fcC8hFT(#3I^Ny1H@IX@_MYs?&Cp=3fyIh3hu>^Ll?lgMWcB?A
zH$!uI4J??BOzvjZn6DGA{M6mQZe|U8&00rn<nx<dBfIYWg;NM_dUXSMcHnE)%zL{Z
z-RxTAHEX0ovj2lMX&kV7RhAY6H@hZ(dnJ}ubXkyB;(wSAlKFq?{osnBEFYDv6rhEN
P`%-$XCSUpLUC93dj^%XP

literal 0
HcmV?d00001

diff --git a/_static/raft.png b/_static/raft.png
new file mode 100644
index 0000000000000000000000000000000000000000..ad9fa1c92c06dd14584c03ea4b18b4944575c087
GIT binary patch
literal 181749
zcmeFYXINA1wl)fgRO!;BcaaX#B!CnJq>4zD-bH##2;~K&caaW35v7;VJ4mmFBAtMA
zDWMZe2%PZlwby#zwZFB$Kj+^$xvqIVGf$gyj%Pe&jB(#lI$A0uL=T9tu&_wfRF!qH
zuyB!BSU5Tac(*we+lg3MM5T^ON;+yvN~}5_u6B;^Y_YIZqf*lFbzT|K{;?0<ic2NO
z=KU<WCGf{ba$~H!Tda@R2+3bW27h9guyMaT9TG_}Gx{3)uEE#-Joda98f-9}H72r3
zkmg<C6&YPf^Uc+b{7r_mXPfV0<0b^WxhT1b{q*HZbdO^tCDv9Rzgl4~Eyok?)L;f&
zho?A?pW@yZ6nZsGL`aAesWIO^gp$F^tdtbs_CTXhH;)GW6}ZW;9^x+J2678AEx(Fv
z`xV7yg!Ku@7KpYrS|Qwfn$E32q<p@J%cFUCfXkyH`aq-ieMA+-H|#%nJR3i7V)fta
zhV^f7v4Q#_1ST=(iWHqz@8J^@V&?=bDl@oI3jS+RSHti3k2ui+_8|bCv@FieqFN22
zMU!-1t<~~#`TEr$b-zq6KT;0mfdR@Jo$g{*u#b>9ht7|@_ZQzM_}8?1wLPkS%3EcE
zGInPiH#}SKtXofP;A?v3jizzHl6weq?G(Og&pcKaRb2PvdNEV#xbebnBjwxke)IrN
zJRnHR1Nif?Ju%0owx``&!(STrO!m@LmDW#*l>^PhjxPtNA&?8WUl`-XOzE0i(8XM%
z7^bK+BRga4Xzm=xH8&%j0MsHiq%M~xC7{eKq<Wl;607c7v#w6OBP}MO+!n#R@JFH}
z-n&HXaq<)_JOV{X3wA#xKP=ObKt`!WkzM1(539;V25t*CWx$a1JrgGr87%I-ouBSN
z_1u;Sa`bkrt1JbU<k<N+a3@yiKn$zE3!8P&&cC-6>mlpr^^pWb?}O}mR*O2_8pEeZ
zf5M-^5~KLs9XK;MD!+&T!DIwj-&q;E|LFMx|AWF4-3!q~;;CR99o)JQrcaE!ckw$Y
z-0o##X?~*4#!u@IA0=@Kjd+TO*P$+qvluEu5aLFDH&1apk<^ADL#d9O$|J)66YInK
z(ZMU9IaP7hKEH90)x)J$S|1H8d~kX92}w>w`BT}^fI_kdA&XCQE$?fDJ66)X!!`M?
zk$xAybNPT>iqMQ`v*Y5xuHjyJNJ3}a0qFdJKSSzwq2-tNpC;*eeH0wT=~0Hh{9b6N
z$NnP3IPQ^hN?x&!cOi`)&o=jq@CVNW6s^bf9^Y{e_IZZ-!Z8Xs(LH&4!tEt}Nzg<A
z4{nRvQ%d;MZuyW-Ggi9yZr68Z>kq<1!gtekhuJjlY_M-g*kK(r$MlRtdkuL?<t6a4
zLQJFJ@8NM}eRY=LO7zT2M$&?KvA)LT5cfB4Xv(qbiGA<;l1|4kbOdxKA3On?v<S)3
z2z72Qs~?J=SD%|*u?EI|QWoZHAP*t*?<DG^=~Pf8UdFThL-@HK_X8?;<Z}1@l_ys8
z(0E|ddE|MjGW%9iwi>aj;<&Lc{g%|0Xi>sRRH>o_NA-Za4KGOSLbx%`KldtC@iW~^
z059|-|DfFq@pYVa!u2mWufOXCf9WtV{G#)Dv~aZK;6>6Whr)i{H>K%c42yxfhv6%g
zlnMi?QDxn)jmz{tdl_$)j~9x*o_#KDC~XQ#_J}*8Uf+BqMa@CYLf!VFZ4f(!_z_y~
z{_EVqZ%2hYMfgRuMRy+akEHj0{yG}%(C1)UL-CsWHF#2Wa=n<k*j~@p>pnd4$LE5v
z`cduC{ZV<?cxF!~S9-0xKD>0%h~=kU38<LxC(-2gq)SPFzI`F8u%OuWm8))0Y4DWr
zgyAQshIK;kv$*Bin%VMdP&KiAZaVrY)zssOdy{X9Ku?dGHVRYgQgc%cozkZ6mRUMC
z)fM<e`$Qp)5Nu4v@y+o%1<IS*KZwI0h4!dge-wE_$I+D8bls#2M>m-^S^M?+QTT!U
zPW=*2qb@bj%4kZo51Qf*Qz!_p5w8}H>>gpbMwd-jPIyi@OKfahUaSP2IK8Z-zSsuc
zCf)P{dC|}5z;u4`BGE?MXmQ)jZ<%`%2#L@U#^DbmkA_{+yEfuC6p!H>vYx`8*_)9Y
zdLxZa3UwAnHYPufPmIO|LivyRv-r)8=nZ>}ovXKNF6|H&Id%b~m!IRJHUH3rXvT~%
zn24D`oWl?TGSN8|@CVIy&C|#YBo0azdB#{sddw&xRU_l??d9=NYD$_^nqI0$=JgY}
z?RYxdu;%d44?26HP4ORPKXTtNnpVBhYPec|yPUagyn5+he)#75LZ~MSezMTuf`Z%Z
z-8%!G1?-8EnUa;Or1ZVg4$yAfWP>TB^74gwe>@!=np0ny{B;_P=rrqbIerd8foT1F
zFAC3<&p)C7hm*&eM`fo>hn@!}Thki}hh!&98$oGbdY0Xbr8%&*uuE{B;0y#?2CoNe
zhYp8uDhM3^0%0H5!N8A~+}a!hUkMXz;JT6pGFmZKJzkr>J9C$f_z?|&xR!8)1aa?6
zxBy`ZA(&;4x{d3~^1a&o$9b~4BBFv-PmYv6`s%rV<ez)w_xhxkE{@Jp!;@c7LKLa9
z`&%(LOooQETe_RN8*XIN;&7427b8wnARt$*>`v%AFaXr>*YnrBqID6jkCw^oP%G4+
zTZV#UKtdOgE6VG{4ow|?It?#%el35<T564~<HK2w#>XJTewtKb^Mdn&UCoQc2ibm7
z-T;9|WpYfp%(;Qa`Rp&Ky=@%q`C-2)vSNt4tG)`V`|>j-;l3cXRm_MH&8l`#we~W7
z%Z^r<OG7|RlhE!)YgNfZDQPjO=H`}lR;$t98JDti4e|_*^y@aaH_C?)COyWrLE<$}
zEWJ?=(Bc<PcvF{UvQ_a7g5Gdf&=jZ#v_GX%<~lGvXr8rn6S_oJdza|$hmdB5H<G((
z3iOwQzVbL924CkZ{eGbqV;fUF;}csd*NMs8cY%*kj{cr28|4D=CdhX^F4WGg)g#vP
zNmlDC*(=wG<2uPy;TJ^eiwMK2)fYtpTZ=i}IpaB1RiCOhO!1oQ_AYfHZvvY;OY+)u
zt39mOBj8u?Rxse~Ct`BWq~`2cZn+JNLL(*h>0hTXcQE(M9qd}HegBc0A#cDkE~V%7
zd&hFjayDs{gK1K0a#B1~+~n=Du_I!1tl<m1uO&fz+UO^uqF|bq3BT`apZOK#%}@?i
zFwdN}sP<HjDB$(%Q|E{8-ajyS9RHY0av;+>6Rd@w?c(b0?07zTG~HF!R7P#D2YV;2
z4#*1fxq9=XaZJK}{@^127>!V$V=;?tb1Sh{P?_%Oh)EN=Fl!ne*arB{B0P4e)A$N1
zL@OjrJzJ)L@MXU)SJHNIFKHj~MYpE01;he;T&#ZcU4_d2Mb@IE5PW5!$p?ImfXGnz
z?!n`l`I`IM&#nUJ(9>j9EU~h0Pz1+?2hj1=ah^==;y_7stJ%gwnTbufFWPZtA_CoV
zofc$seQ3*iL%(jc_34MjsptG=nNph)T!^NP4#*oQE%D}{;3zkxLc=WQa_G2XhhvQ6
zP6by*-_kq3l56CF49XYrh3$*>VnI6#O6U9~dNoOVSUV^8iyT?J04fBfi<~@h8mjMb
zP3WEI>+5+#=vzhol#Pgo8|Pp{80#Y)Rv}#VrqZoA7joZWo5kWoN6<<U1gy)5U8ppy
zY|`y==a*F8bEl+*NUa*>We8SXb@GnazRYZaHUZ6r_3Wms!dblv)pvQD;)t8_y&z-Z
zq)FzhiECmaKD^+esR+y=7AxA;P|Z$L6YCM?o&XCM`vKNn%pEr7lEY^B=lyf+hgf(1
z%*VmPig3ij{YM=wOnUpnoPhsq^Iz$m58+t&m>f#X1^k5b-?ed(pYHtkJ&q2h3`^ma
zlA0POeP!)oYwP0a;OaHrYD|jBAaqkT^2EZT;kv!B)pR)zF#S(E>KS?&YHB>Oc6AoC
zv~jhv6$CoF-S&e806xLoI@@|#vI3poxp+PS%Ci4i;|b>eHd~0D_0K9^PO|KVnmVjX
zt{%3m5`w~l!t8QHtgNg64;#BDy2{W0(H!$nmfgY2%k7Dfkgu<=ps$#qtB1Xih_tk{
zkg%wbsHgy@hJYu?#mf>X;Nr>gU!DASKgzbA)*g;-UXHFVthfDITDf|A$+ENGKIosn
z|9VebpyR)u<l^~{&%%75&~1s3h@i00Kl{cs1>EL7(QyRYzB5vGbjHXG^9(t0X;HwR
z_5Vl7zn=J)mWKaoDIzK<{MV*`Df;)OuRU!&lw6%L&-9Y}*LeM-@m~x7(GVbX`{{oX
z#ed29&s>b4<%j@6{~R<qqCK37b<8+2Ix1`HVN#5i-Tp9IhlPc?Zd1&K9mq10T@Ay+
zdWxl{te^+P-i6<7@K*80`^99yh!bW-T?!<`b#mkFadgMw%2y4c=HVm{Q4eNU2&Ljr
zv?Bb<8K#@B<8a6>uiv7YWHqqYa$HU7dDaeV-E1lYK|IeI?O_nv<B{Ra+BPGxY7=wC
zG&TxVT*_c&*8h}9P~$XnLF<dQ`NXmaIB}w$vM2tZ6FuAyGehD3q3|CK-zj1$&A(4c
zHPZTDAByQo;{WgI|Lby^3Y$Uw!k3@PaqA?E3j9xc_or%A+LAZ?@GhT#@7hW3kUshS
zNt}10i5y&z?eW6*d^y5*>E`-uIzRE^!+%zE%p|a=rI5P({iXR!{FAp%RN#HT>E<RC
z`OC4AINNWz|7WfLhhmq~<M_7jm^RBSUHmY~nlTZ3eEjS`PZdm~4w`2crl~6~FZrtt
z|M!ZSY>C~AnF={VFaFmlqfZ*liPR?@(%}71Gx^V1ZYYMCfhJ;w|F_p0WUFM0YBj$;
zN`u?W9cSz{FQLtR%bZNBWq$o}$4=0F>8!aM?pzP|xo{Rk9|YZ;1;Op*QL}qP=GPv+
zhcV_YsATzzWS6CYGgCbZpH9(kyLy*orD|e4`Sb5&O+o=%>TnPYycj(xNO2UQY>~z0
zHf3b*y>zu9+3cB%T7X}p_S?%|+~n;07tFi1qnqxK0V?MM&fwFnyY1&aLQh1u{WoRI
z>!;1YWsZS?$5~51ecY~(bKB?ba|5PM;3lWG!syHC_C@sd{>^pf^rn8#<Drzl4Phk^
zSAr9Iwxu1^B(vAB`Y7r2yidNZc|g;p%3|P$)N1VBGFcGtEGfa?Mr)zO6!A9au?w(^
z;m1Wl_0@55?o~bt)qaDD#_v4N-UQhD{(Pz8odE3Su#*^<KWl_}YXdu}nhe@6Cq1J_
z=s-WeFi*la<<2sWQFbVN$42YEwB}^3Cq>|OK64|2<^8EA?MvuANza#b_BZDVH!}g4
zZ^^y9XN9HE@$lx0{N<NDkF@4DY~-+deh<AIX7DbQDf9-L4A>E3!}($3=PPZ*^{M$y
zE$Xx)NLIY!7~XL5RmgPy2Q5ZcdwufPhyE`&TO-`s*9BZ>O_zxb(j^ybvRJMUB4p4s
zV<G~q@--Ksq)!T)uhxZ@Ojjt=(i`F0>Yg%JE~39yS#E&N6Fj;89%Y*bt=q5RwG{yf
zae-@C(Dm^ivc5gwR_`lhb}PFcH8y0@pTFLs%Dr6UX`axQZw1fKFHWK7+tDT|xk0A}
zL9bDJZ`;7eb4#ejv!<Ynw4klU!+2jWSHFp@CWfW+UicCkzT{2Y^6PGatZoobpBCV-
z8-(g>A^ezC$y4xN{^nSI|N0{Lrm-$~x@AKgepk>f_!8xJ1%mIGO$Thh%H1ZnRwr4*
z$!IjQIjsmhE!j&{BHg-PmB0QnRTNk^Ls<VcRW?`s-{#Q<s}5j6LQm>=xcXAE?JzFj
zP;e`zbCH&^ME-bKjCR<X?cwEg8?qLiQtrj&;+U+>hP?D`N5P5Z+Fx?q&#P*m9dVK1
zO{k*ItG@ZIBL5_F9U~#90JejTkBx~c=sr0l^qSPSp)1>q2s$l)a{*(~v=_TFyjOCq
z#B!+<gqa1^LN@dH?>3)IH(p|Nq%r1E=UMv=x{b~b?d%GkpF_?`JVt=oE-r6Z7wUP%
z27&DWii<le=TC#sJLc^|%9hS&hp3v8wu82_)`}a9yyveIFlxkO8xPoP-j_qq{_v?A
zC?L0P?jD1zp%iJGMg`B<bBxN}rZ!U_5G^mzjyK7l{|ZGx{o~h{J89e8Zax+OzTQmL
zx7GffcDdLLyqw5IO158&pk&V$JayaK&%o^l^CQNz*B@VL6(LTd4O*<s2@+oiq0O(?
z&5>bmGCrCx_-(11&!rK5cpY>$L);ISKM$2pPtH{oMc$y7Ze(c!ez^9`R|XBJn~g91
zelEEkGrgSs_+ttAPI6#5EG;U+bme1Id|P6PcMI$O7j5c1s1Vg1&&B%Go5F$>aj60A
zH7Aj2#aLw6W7DB`FwitUW?d=%q6`_R{u+PBlk!K<zn^KIv5yLfUKgv_9{^51ljS_T
znr_|mMy-HEC%`cgN;m0!0J|gpFfboExM#vjE8`mlICgq9LB36BQS<Z(E=g2GERm8;
zH6<9c9VA@d$|8k)9=zYmvmWIk5H>?xO81@2|JNNJ$Ihd>LO_KCWv>e0$tlc#Lkv~L
z=tsH9PTQ!;MHXnZ!q5E+Cx*vXsQsuM?$lL{hwF0KmbJEuS!UJ>KM(yOv<PZN^#c(2
z!A7ZTldw?clTVP(<}Ea+6<GtdF;)&iA85>JPxH6yy*A*jl7N*daMss_f-XvDa~@Ox
zf9oHyZ&v_)PYGZvBNHV~OE(*?*Wc=PB13l{E*id?SlE*!zeiy4{=x86=u9#`u1BJ-
z5O{XTFMstdzB3(^zI)7%5<K|A2XI=XjTk~=)e72iA#r%nzc9M_<9(ql4D@4SN6eaY
z92SVIzRtbg%Dr6qI=v6=W0oC8e`7ISe`HY9?ns51%aZID-i_Bi=2y?OeUio2N{;+@
z=9l&~N0|c8m#IAT&DxJrDo!vH`eq`n?BV4~*#ZNZCNA9&4Tq3H(<k|>WinKi_;d5R
z%fe);-iUkuwncXMu~sfewpKl^xS5bgPY@be>O)>WR=0R8w@MF72Mm3J?{+gMypNFG
zwM-U8v`#OgLs_nVL1@hw=z=1d!+WKFxW~V=s^IU*TK!sU6xAx(vP_bay;9_x0ABdA
z*+cz4VaFp<Hl2O1lVF-=h_Z=h`;uwaE0kfu;>gVAKD({Ra{Kj8`=jqPAfchA%PL0T
z2InH?gkcaW4+)x^fYhlsJ5X8wO0%7w6yB&Ua0S`@9(jL{!1ZbPnaR@SltjiQDJQ4U
z=iAlUb5CHMr58}eG3Z_ahL=z#IUB3SZUQ!?(a(hZ#?)G>PTGXPeYImp;7}}3Kg0d<
z9fjRPeZHXrp3|=*Y`xR$9|iEZA@w_&ySx=`(VPafF<eN1Yn4LK@s+KhXT}SkivZ(a
z0~&4Z2Vp|liA9(4IA#135adiIy4*DNYfVbFq)%;dT43V$H6hn+3D(h&{I$sjhyn`=
zc5?;uJ`H>WJKAif`kQ?qfrrBoJ@bWAe#Z!wNf%=IdufHHK7HVirRZ*mT_@i%W)BK5
zM-uHWotFC|6uVhvuUynPhU=a{BYxhqO6KWHQel=Z7~xNwsg9+kz!`GvhWLRn3(Vp%
zgd)K=v93A&kEhkRl5>qtvnEwHv)VxtuS9JVnxxlO6uTkH6z=`9`T2o8cU<Xc*pb~q
z`{1Q{^VJ8_h7?w0V{Qd9aL14?I_PE&LB~9N&nFppyHAuz=7Q;co_*S#pGj+d#%G2R
zAydG~7*z;h?%ZVYC{=9<F5bs?raoKmakLlrbeC;Tqp8D^x0~73!~=d@O%Cp_;xk(&
z^5(5@cmHxXneI|ryVKGza)!!0hy6*rY0H)9KTd)FOy-$_NuKRyT+}HG$96OryYw?S
zf3AET)p{}L46F5RJI+DBuJD^AbTP@9aXP%+c>Hh`YpADd*^!B5hz|>WvU~1g;4W6q
zo6CxuCQ5M7c|?$5+SdV;CT6Q_a<wq&bzDkQ+&!GTSi&r(+hbA6H9V`r!-~7JtE2;V
z^zhMDyYxlH8`z6owN1>Huy2zA;TpAG{7BsD+Pq+{ZKaL+CE(1<1=AT)taWw1E-^=M
z7aM%{;Xz^zXE{UX@09xWTp{nSF~Xdjz3A-NkU*LFh-oC2wfct$$BmnRpNsm>ExNL!
zO+Fr%GiHu4xpz~TvN1=KNJ&b;_q#n5EsXNS&YmBevk5d+ofPun6Kezt^sfIdYFJ6V
z^RVfxWmDEvgtDn>p+E*T3tN;HIe}g8!kX00jy{;q`nP>s{n@HW<E`~9P6t5JbbHL)
zAN(*}&hPq&=gXeH7NSMb_q2pXSDTPpUXhJ)X8^QPtI(u%Utox7hcNZ#fb3>ey69xv
zpxw_vHQ7(MctV>PG%csEZSz{tP|6WX2R*%r)I$mu7husQl$Iy@n@a2CL%5Cm$NVx~
zX8S6i3#iv_7d6;&9GAbtT})AW{^!Bf^l)$MO24Ivn@(!>B`D@4b48H>50%-)ph5Y0
zNA>Rh&wyoO^WTF<Gnx3dyhFxL^x0CxuC}^;w$l0m+H4LR`HeF{@zNXV#(js3yE*S$
z(qFl+8clD??t@P6FDq(!^;lOxRCkpJEP*|IW@9O`Bu+Vv21wsh9x)jKpU*n|Gv(yL
zwJw5~W!q;w#QgATXK?(uoZOIVx8UnKF6IDoIY#};J-W}#0NK9t^G3Lk>U2+UR<Y)u
zE9}Bt9z325(c5`B&##`mvsLfOE`D<#+epN%sq9>f#7x7D-OWfJo)RjxZQYYxuwqC6
zlO9N<@NJ%lKjHBs(mM4Yi!u9|g8sSD?R4W_>@}NJQs>0;K*ngAuz4(vn0Gcx3}9)v
zG%i0>)jmg-0wvMfNr5;WG9&t_90?Jv!0>xBP-~Tu7%Hcss^Rj)PSe9D(GQ#75O2*<
z<1|>kpxFxQ0oU_E8gtqhY$Ndp@<pX}wL4PU0>84zAdGSal~3}tdjwEf-nK%}N1iR(
z3|xFUc#lRbnc0QbM<eF&iPc^~1#sOq*BsN^&<NLkQrVNyBG~f><vkoFs9k&cdFqhL
zLAn{~u$n}Dac!-FDf{qcXhOg{MBkBdOLNl-m%&cV!b|h6Oa5-_loYn(uRQT*!1fYk
zVZ{$gyk~Mh;XHDOQJIpb8ZL{NrLD~QshOQLwdA;mw;yPhqjt*qaAv1Jp^CpXPV~Zz
zl)q@^_s*OZ!-@EjfxHyPB6ziJQIGUV8S06tk5{m1hW0f)3%sAuWqd8tGOt`i%L8?i
zvDJn>kxPu~?Ki@q|AsDfIy$SF(AtT%KM_oLyRY>vuF_zOW0SvX34CIIHt*-M-zQeS
zlKOSL)?aw?TZ#eNadmDH8=q0?^u+VRYtHS^LJ)ni+uY3u=`}rcE9cpKd$aal;$3O`
z&#!{7MJBH|hzA|7u7>U9UiH0lek-dj4iTyV3H{A_v|u8C7`-1(V7KVJ;<}&Rys$3l
z38*CEpG0+i`*j><1V93T(jz~xamG)1InB%?yV@SkCNHf(O3)M8T6_nIe7(*=B_{CV
z-gPE>Yftq>E3$k=yUACMJ=D?ev&XFNR*qTaw%Hgn=FX)GV1*(%ZgI7(b(K+L)bD+d
ztXW*o!EfT1qqOW^TA36)<*7kEat_wu-PuCGASJ9}Fe{+g%F`o4XaT+FyR>eKwD7cu
zx~hZq`DSe%cY9rKxd{Q>iK2`p!9pcbX^i3c-%7*!AWZ`KbKMhKGQV0S)TQ}30{fMi
zb<<E!PAPi@>V)}hIzxc`z1G@`j884qekoegB(zG9KqSKs3Qt`#mU-ka29U^*r5H=V
zgzg)N8G>+(bSgui8UNu-RyEyrbB(&Q8JLDBN!F<Rf%6o#{@zyB+$(Sexy&i2otaWd
zE44L!3$VU%#Q9PiW2~G)(8LH5!^pbKaHou?>HfYrz~-$$Hjc~sR)q5@uG9WHwNYec
zW}no0YKtS;eRg~z$Hz(XM?8WV_jy=oL##GrS=YZi)@cb&mhS`mWTw7eU%mw{&T@<d
zI$H<Rrgmwb_*SL9b3Gh$1tx0}x2O`2!IzFw>q`y4Ysg|Oby)dgkh7`b4_>ro`Ucb<
zCT;c}@mU5Sf7&>eTuJ`$wIB_<ofcfiE&QwjYR4}X3?JA%kpJ}unulldZOP01lE(XK
zXQgP7ffU2>eaTY`SAE1Mas%{ML9Q9usSe&^*R4@^shByP?6d3?yP8ybbN|RSBv|5W
zE4@m@_bNl?TBYH918$;kRuUTaGzqdW!?vX<4w#ND#fkj<LuGIZKI+>C;t;ulaz<R6
zQctXKzgpj7B=#2v`#N>EJkQuoNQ5KX@mtX8D_Al(IHuJ-hpxPwgKRGCRW#_q^SN?T
z$|!#|Fx}w$rF?iq{Iaw)IHRR_H69A@Q&6M-yKPX%g2Q0PTQ1xDaD{*r-{F2jy!4>Z
zm;P|{u|RW1*o@L>Z9a48gc|sxPcI3F(8s+MTODmga)Bt=+$6%bLW}W!1ojBQq1hT!
zs5nFb0z6%ta?rhpjl!7~2COvDBt)AM$hV8du1_*MK(PY>5i?4USe4)8ln!FWpziVE
ztIt%LK1l5$RI(CU5qn2Rh=^+3n;ja6Dx{#>?V++ef&dp&#`FNnJ|f0+_5JNDraEak
z>@Vjx{Z;a$DBO!I%c5kR!^ZG(9}=(f_b-JFE<ZMKCjV^QQu+2|1tc*W8Wa3vX#ChH
zN|r~HqukYhl|h@Ua}FSsI-}ITU%%8wO`JZn?=E6VV+A|1#3x|Q#PM`a!j`=^C!<RQ
zJ?Z9=gHq~560K>{^%7LGE+wuxvsC9AozNfkF0+kowlZw3<xfZmUsZUQIZa9^1cFF|
zV;ZJ6PlH!LdW6`HGPTqfU&@)%+2!&+0o%0(H%-TD^kh_J+qcvs>9Loexm?06*;Pdl
z69J<A`*IlLYRxUN=8)eg_6>0e`?wNKT-YmZ*)pvUiE-|i#VvO~e+H0FrPCxX5)1%Y
zKzOv)e{@MDI=O3`X-O@=Z)$1LbbSEjAq^o9ClK!5PRxI$y~ehiR4hZ3Wb4|Y-%Gkw
zbRAvbI+Eq_V!Fy#bRt-*C&pWswnS~0ZFZTen;B<_ZhGae7=mN2vQ@#^cdSl?2ln8`
zLb;rwAWxPqBB|Zk%}M}A0}k;TSECkNO<12v60{nV%ODQJ2zRS6msKN9L1DK?=A;K1
zNbb;k$abj0%~#M=Rr@T4Ue_dmNO)2%5fmKdq`oViW0vc<%6j$H5r^U%LUvv%Hp!zG
z`#i~n8(bqsMHuXe3S8nyczNBz{Wo{}J2G4eTxg$zFD37I#8@onm*1o*`inRdEy{Y+
zirviCxP(*RhB^{R8j^fJ?l-_p%DL*Ye*<SMBQ*EKMCsiqwB#<<-p+I%BPpy7R4h8`
z!OKw*8mkmkxK9Bwks@6%CjvFv_GUaN?G`(;F-&d^zsriQH&Pa#RoEtU#^zT~Gf-)!
z{YNXoJrX?iZg*R(<3REX{@Sve`ZEfKmYYQ3?N9tEkmDA=bzbPtO8su;FN7DJ>I5R>
zwrjR;D<q#!^9O%6PYs=_>p!6FRC+k>#oC|FqnmY|IJ3$-7<y-HgM-9OAXaCqoCbQf
zA=wV{xGcJUnYvs%`{n@k*oq8t`AwqFsqPfF%78)9g^sSTWjB$}wsDoQQ{~24TXf^u
zgM88i_BGn<_=ahU+NEc61mtUNcXSsvvH4)mqk^PugWppY$aKj;U$p?CdMDQeqz;66
z{cQZU7gk*8L_7dVj>EPzOTGvW87b*KHa?&GRZ-dRW2E{kS;tnGTK4kfF{kN_KN9gh
zL=O_>C?80_kmGP_aWzK$l<yaZL@4O@2dUGsPlvDRtj$xu>^>F{emefa^C(<6%;-eH
zEPFYa-Z7Q}JV%uvYf~I*{j;wU#{NA;J6D9c=8R$ylEUL1(I15(7)y-l|Mq)2i6!5Q
zTw{m}*<uck4cvR{fh+A?HGpPCu4yx<oIeM9wsRrZUdr~URqv@hI7r2_qH^ExrBZN)
z@38bf;i==u$@}rui88b_=JAl|FX^)$`tTlduvbL?G##{{4EOVHAUXI@3$jr*o$LOn
zUqrMqXLp=A*8SkXY*p+?Y;cMaJRZ@NxF7~9Oi&ATp?J?L>ONx!^iWv@z48;pp_FbR
z(tMH_b{Xp>_D_T7Pn+PK{H=njSk~=EyFZnt9xgogApAO;;x4%~@P(KH{Bem*?YSjs
znBT4(M@rT3(@z@v&HGv~&I@Qu^k+(J=}EuqgKUd@6^kpj@j21MVe2bl4|-G!Gr4W7
zm8FWvFM4N$81%m6xalKkU;0zgXTRL11ieqR!eO`Gm`L4L5um$rV4{yRJ^hesY{UK_
zI6CRR0#kU>j@<s;{2tw|x0K+cz6Ra2GVI^Jz4AD0bG!T(2Duo=9Mhxa(BE+#7Rq)J
zDi5p$=AL%bn?-a(03&Bc0#*s%qui$BBf^h`Vb(WfQL+K|=tn6*F9i!6joGC}j`6<%
zJ_1)+>BGN|pBhpooWyd$RHe{Az(6Lp!udb$lexYMuu^@*9CSss9Wb=f*}zR^Mfj*w
zEOqaNft<j_u^e-VGt2Q_gUZHi2<!N}<9wFJ3NrT12zT<WRD@q&X}4KEY)25FOQ8se
zh}cqDKt?Y-?LSm~7o~xiw|UmDZ3vT83%&1X`!y;wb>?X??sIO#PL$|To;luf*Q*Ea
z+m&?fZ1Exfap&&qMh=c#6*V*Atb1&II8^Dz`i_kyEk|w~EWm()dEo=A*9H(1LM!bj
z#{nTTPREw9Bp!aMyHWNbyxS4e@WhtIt>rq;sJFlbkYH-sgP~r411KJk@U&1x){k^{
z@50jX9<;zTOmYpQVk7MAmg3KMrPToaI?QKXqRF%hZES-k<GO9NtFnR3?|fBv8!Kd$
zXSHRVH`mgZI2@>!8zoO99+yYhLnKN|DSk1YeT)nI@b}211&Y0L53|aV6m=cg{CfU0
z^&X)B(QcV*z>ev?a-bcX6<P4_iz!yC{D|+rbIveU@8*L26$c;C!|=sVxgnr)Ycg>2
z5bNElaB3%qTkOp7^##B{KY3=AT2I1;?EI}!IEvx2U0jdeUjW)?xI#~@5SUWbgB317
z(doq9;@rJD>9!KZ^PDoKt(+U<$gs|4`)F*HFs=7<$4a!c>^N^cH5?U7lJ@C1RasSo
zSjmd)`=GEb)}<6hGIr!g#kx`&sH8wp^|c%)vPK9^Fu;T$Sng<uTMwNW-?(5{bRn>2
zc0s6s;~s-1?uuyRv7$rI6SdTtx0?Z(0KNd_SGcDHCIPJ}_~Y+e{gnOcK%47yi(Ye-
z;}3BZg%_Wh#Zg$O{&6tH5z)JnUipiJ4~FpqCFT)`!PqxUbVZM(eb%=ka5pC%g1$=}
zxIE9qy9T5J-)j<YYrer1N@$ljC??vsAM%-(#qV*7k_9s!9mJ|dI<8gAc_N(5oB7ig
zT0dNDe{z<I_=p_$b6^WO5#n?P?yyDot~FjoZ0>qQ$*$!=A6+)`g}j`vp3VcrQ-HbO
zgmgb?HHv0vczdJpjpOqi@)4;}Ut(~oWnY8i5#KnI6<#~=x#`;yB^<s*mg!nfq{g>r
zi*eRUA;UK6{kY6oN*d%%P#fRa88JfIGfn2g5{V9;NO`jY0gxH#ADMumX5e|L2(RjQ
z5yjd7?Lh6n`SfSz2ufw35s#~a#bK#ap##GWw&u$U<h`HP-vE4iJ29K(_<(MY>JYU%
zwYn=X-EigSqWx|^@z<O_HE_Zh+|Nm*Jju|RF}mG8OvDQ2F>ujojPZ)Xgx?*v!=q%)
z9;ME*bI)I@n6W;K2><2pJ@9K2+X`eLJy0(=@f@!Yf1sQE-WABiXhmA!UX*Npr(qyn
z%so9wh*F3RKcZ(3Dw*1A705sdj>LIhz8-{-b5~7cUFjXSKW9%F+Dc3LI9|WoC5Z?<
zfc})lS>a8sG4b0TO(R$yp!N<`CK;zzuwCLuasE)z<K=+~cr!vW#+7Vp)<EjJY*Jk%
z!bdC7{X|y8j+29*BkOg$nL9lieu^)SnI#eT>*7986r1yYV^XElNbyGC+S#}brn1YM
z#Z^6S6gd;~M9k-&ENag8Ly$t<2SjAEqjTRrpqPJv6Tkn?evz4fJ3OXL6Z|!*)nnyd
z*43JfkP>9CL}AylUcqB4qW9Sv^RSN)YJ|xcY4uoIlEn_m{a6b|D?tghPx$fh_$@iO
z<vMgmthc6{8O_sjw1<iEX+-|)h<`_5<to=mb$Fcl_voBAg`uw%Sw^4muYqM{S(J58
z#LArDtc{crO;;ilhcL$ZH(dAXhxMI+t=aZRvHE5N2<A5}Z+6!?g<nt*TXHrHn(sJv
zS$BSDmLRp}NyX)c)>m^lb0@T7HqkfeS!LO6+LspQJyUe^MxBT4EiH5FVM#yy4Z!U^
zx`&p<n4>5}=>yx@N@cvIG2(^+V+H(_UFX%#*mc6=Q9h;{<GP9D8Y1agwUuCrnn;yw
zJ8p^j`a4q6n~y9tiNgrA9R^^RB-igJ31B7^kMmN#&Nc+-3NU#~kzIt}i@qbBs<><P
zjfpdV(?t6Ix{;F=SrN6F<8kd45+XorAw{E~Idg1=tJb{R7*L_~k^RP+Il_4X*(|#k
zQaEWx-q3|PT|v&?qM$66SqbZYyKxj?q<3rpKB9TXud#05Z1{{5>67<WyHC{OiTV!V
z3&Fd4HZOtNf_R&;e}`~XdXpO(SS3(&9Dfx*vAc+8B&;eeiw=vFbr&ddCUdoLF!8ZB
z`Z{}YKwgs_<!cUHXGL2XnXDyYPr4Y&Ki>uQiUKe}Vj?AYM_@I7k8_kPF7tSasg~5v
zcb0oX%ukW~E7Z}r^e978yQtSSBE2<SYu1(`-HS8F6M^#&yCGL>J^TX+Tp?}>v2lo`
zUXuY21$XkwY9J4(Nae@JbKlNbgI6p_Y{*1&Y2((x5}li-eV)VLP}U$_6;%b;LHM&5
z5YUE^_fpFYhYN2vqy?8StkN;9NhM?I!;F|?h~6_dqX+U+b?ETIT6^;v*P+CLHE1Pf
zO88p`$?VsbLb=pd-~dl+HGvATh~#0)nkYI+pfhP2tGcy7RI4M%jKm7Y_CP_Ho{ZMm
zPDa2Lv9Gp^CMDs+>inWD6QjnQtPJg*J{R6u8{gOc;re&4s0m@PHy08}c9;k?(Bn5?
z!eDgb)?ZrQV^>M!>=Gu1Zf3B2L>KYDek-Y%P-99TjI4K*)ZRo}^I9{wHr)5+k6*V_
zxW>S4s(U3$`XCD?NADXPTIHBJ#+mO|R)6%odApNpYvL#V{DsOpyiBS%0sEg*Z<6i;
z9*0}Xkb?)`HGT{A&f8~unOz)x;&jJ@H&s2xI%=2B<uKNEuf?j<5GjMh-SUXS#}yOM
z_KUz_E#n`lRk32sJFJY5E&ULW#C=CL5_!@?=0Q|6UPpm%#UC+~T+xqzF#vX>!wB~k
zvqxrZwkt?Lm&%9&l_`DRkv~8uLNODEfEK)~7VN2&<j9~}gPi;{Fxjs}8A_i{x5duU
zi~6*AM6yIM&!&m5VJo`DF3c5oGs7-APCM?96{>gkZBv?sxI^rB{n4UPtvQAl#7ovg
z3c?C2j`pOxmAKPwMWij6Wcy2BKQ8^7{Vs~L^L*DR(3XX}`)-#*W*!UpAnYZJK`+ux
zZLzeQ`ObV>VC;+DT9O)yNTgV`wi$4%mKsL4{GDpwmx&o-xl}U0up!~OGV@A6VnNHw
zl_xJNW9AhO^Yc{mI_K{mEvW=9^IFT>eyAnshVN$1*5N@6ON99Z9jZ%m3D%rlK5R&f
zP$?pId_m-EC6{MQL|3~dw2J}Tg<KQt@g>*8cY1YENa4F410G8=^V@?4Z630~&P5AV
zQ%>1g4l=EIWgWKa@Tz{9xqypAbp++eIwjcVHF@ZfRZz$Bhbc@*fSh>lq0nYp`_+1~
zth%Kz4L`9f+4E&y*$t-p58cfBZUK)8Joi=BYFFa}ok0O7GR)!7o9#+Q8)?_Ygw)3x
zI6VU;qzQC8B_=5|p)rB<-O20M*32<0^g|#S2B!%06q<lfZFh_6rKEMiil7&;*KZGs
zPX<}M4QH0{>>fkTTAK2g7Qu7y%69xfkmlca+CYL*2B`hk%yI;a842(6t(vFMXIQ<u
z1;R~q$kfp~j9(!km|FcU;$=^X{glO4&e?P3-Q2V5v-Y5)^R4y#?*~YVK4xQt0*lkL
zUFk4#!hv6@Jd2vCgk8{)n*+T!oc(q(fKZjLM7L+c_vU-^KiHkG%u2~pgF2h->3}b!
z1SID|xFS{apqj&$u4-jpleMVL9;$wk&6)MI4zN+hGTxZ^phDbgKNrwlLUNumqU+7m
z@PoES%vr^}lF=K=%n8`<ocJMw_$=*1#OEQ9nj0NqIMz0QcIfgh?Zy0XP=hiJ^lgPW
zZqC=5u7k|!mT%1A6y8G|@oY@rtL9mWyMw0_v8=rM0)tinw~<FN%TID$RI`hDGSO!i
z?~O=dI^j~(2A7ew%h(SHhFBvhjMreL!3K`m(aodF^nhtbkC6LWwqz<Eb8HO3LekP3
z?ouq!f*kxY^=~7-j|6q8XKgM7h8fE->y(4ZaH2oA(@G^SkJbF0sjYEmw<caIo%yN?
zjt>&42QSjHnd<Pjeyi^tpO6aeNx(1K=H=LtzH3FZb!it6r+v0hXY;O0nfPcs)pAEA
zh3JHwXqmb`$O+5kXhiCrVAAxTJQY))8LRT)%(bLnjU232Bf(4}^!v4OBIES}v>D_P
zs)<90K+l%n);CHKG_{;<k1=~N|40JTA|`~*POFPlH(An3{k}SjoqN!(-ogbbjCEQ=
z&oJ647pDJ1O-AY8@LS{rrNq#_-_P8Hv+0Z{QkS1|)907@zMy*_nSwdT(#CT&xi6(a
z>1~KA;xKM^wPaK*V{iD^1c_?<PhvRE8n7KJhf$H<VTj7*J^t!V3UVBLCNG^<*9j*)
zu+uNL$j<bH$G{VkwD-^a$^iuQQxr=n3Ud#t_A#KN4f5y}VY*!spIh*RV8E)t=h)aR
z)flLd7VdNDFJgPmm+i9ef89p$%jzS$$lsgeUO1~gI3#DhJ*kndndE=>9B(sFCqx1j
zai1BxPm+fBuJdn+LHKK*#lIl_emXbZ=5z9LO@r|dRzbz6mrD<VFvlv@M_+6=@4<{Z
zqt+|$ssp|NA`6jgdX7?-;{7CtA3<ziK&ieM_p&ym##+67PC^a|@V7FOQSyn+rw0oL
zuJTTPu$UMh)h1`OEaVeo|Art&4xX6&+IZ1>;G@tX`~zWG_$H9ySnxw`YE8gYHAH+<
z2kR)(KDEZQk91@-9(WIr_mBJ|Mi48@HCe?|+lY*0t)y)G`>p^;$ap<&qil|(uxBj`
zhlkwY+x8|Y<vaQ<ag12^&vLqBS<|%_l`oGTEIJ)99;@D`jmo=W^Xad$WQaQnHvkbH
zy}3@@64=Tkt6gKEY<~KLkU+e<+}?R)l(K@I%{)$MFZUpXihHPTd126ED5U)|r9x{w
z05!iP7m?2aLmd7VCbFTM{UoH`)$kQ!`II|iSPJ7~E*U16t<Y>@aaPDVlz8S-%lVU&
zio+b3wAX$USPQTz7zpXXJK{Mi%<j5V*pEnmDuze8&+rO4=UI^%A@)PcdPSt}OX4^b
zW>zEK@PU}X3XOkJ3}Y*!J$jrBlX|Sk5WM8CfX1z}iGZ4*mgya0<7Kzk;?g{08V3?t
zWdc}`B>A=$c31Cgn*okrlG?b|m8@W#Rtr2l;b(Cy3NIemn;gYw6VM-Gu!#+OCPf;H
zO59Q>PVd+q<w}pt{-)*~Q_2=2QL^2kXhaCPGzR)(p4Jmsn)srNWQ4*!l3abR(}e9F
zzCk|ac+8mN1-~v=iv~M0M@$kM=^b1J$}^g#xwV|bg-5WD`>76|>`bD>?zZx>j-oI8
z$;Ok|x<1)tlPnDY?SxsW!e6w1%~7qKV(31;*R>ii<6?Di5ZHo=BSY2aR^}oirdb)C
zUF7`byFv?6&NwcN6Xe`MO03_=@ltO@n!+6q{6FB`iS@N;9@bhQrq!l)1mkxVbGfq%
z46$2EJ<jfJV(KIv7<5pEh}LlAG|cf0$$TvuA%W(k?l~6Z=$}nTd%c-FGhAFt5t(T-
z&w}pK3=9izAZT!LkA`%6L_~{5$X0b9ltE+PKl`a{L|3s5Kt4*=_8v)R-jkjFCcEU@
z(D}Hf1?P^LL}be6F$PDY6}n!iC_AG)Nk5~l^L#_wk5_nyBBhe;J=pX~@@uuBWWW$Y
z)hKAxACddfu@D&5@0qHK6h90)t)hw8U4>bhxfg)*Ww#}`CIZ=d$YeR4Y9z}}wHymz
zr|o%W^;Ya<o$CB+*`*z`v&`Ln?lvDQGQs@HeT947;8S9=q2s$-M<FCDnoj^(Cept_
z2y8YMa4DMxMaaaQTFa{^{ptGOx*p6@+t982)=g{ri7Vq#$#t!fS2M%jd56QMUG^x1
zQPGf1A>^@0@B~so^i6Y`YgLo%x`dlk#i9&nVXV;Hn?$LVm;#~NWM1bdfHfNSvJ*7~
z=waIW7;x%?Uhw`*%GFyni&a*S$2gz|%wb3U)(ttFPa*Z3&}&Qo_>L@n8OI^XE4k-1
zy~pJk!~1P|d|iX5|GRYfh5{A8YS%LA%9U|y<LsfCtNYUVs_EP^iHlpP8jHes5}`dB
zDSj}9=erb8f5|UBlS-e#$Q(cJ^T<sm<NX?`=u|*5rGU{}4bt(^0FTNtZM5Y~(3g9N
z$w6#1*PQPi&A@-#{kaKBrJ)hO9Y~sVbU@EV8rWxSaugWOHk(DO=ppy!91r(iw!EN=
z11Np#p-1IRM+Af#f6WjH-+fE+8JG&)V=BY$7_D*D>85XHXGx-Q{>4NKA?UKF<6Bk>
zyZSU!uYTNZ@6egW!@Tk?{rMH_;Mx5;Wu*nr*OEw1M!(})cOcLA{8Le<{bM^B3~X0r
zU?ei)oN+$4xTpIp2lBREMp^=r>$5)am?ersb5Zg*MSG6)DV^V->a<7E)Ke<j*;&CM
zm*a2cAg#E&s06m(bLR-(-gSS(n>U;tJ9dUg^KC35J_q0NwF4@=YI)R8!h<DWt<ld)
z)y4Xp$KFsJ3V<e3#oFXIrD3ySI#!r4pTtUcN}m-u7lDm3a`e|GukRplfrjcQsQ(4Y
zMbD@TsmwcjI{u3=+N!)15@&!im2Mta;Ih}z0FQwcHZRkK7+WysJ)y>juihlRtHYuD
zV8^WlRAx(JFXs!H1pj*faLi}YoiA;!rPtTpIH68LbKPMegAJR}9W?YzR=LUj87`?>
zG4Mx<N)6W+C67MYBf{_9d`=hyZpJIKGtDQgRwh+4<Ig+#X)7FYZTS1*Q03bk{WjJ~
zWe>u;i4@0`U#NE|&uXl(8b_>V1sG859ch!kEdZ3$WOn~r7MJL)0X-+xnxB8wmZ;w1
z7qOm*2a`vfpReZXv@*|o1=KGrG^`}1I}|uGDw{XwOWE`vnA)WJWx-y-aOY-Z=FZsK
z`jG9fwch_a0Iep8m}fmrELp-tcWRLV>R$>d%DI^l6;f02fKb)PAZmz@Gc!DmHlh$0
zy0ADMp>aen7A_T4;EH>(Su9_ZZ<953(2Nw-r1}2v@2856sb?0F&-jQ|>IlQ6wg+{)
z${+5Fr8UxhFcpC`uN(?*X5;~V!)kVjW+$k_n)vYC4esrhCjvZI$|mJ|`u8!$4z+en
zIX~M@G{r3S9Z;EB?*8mI0&+InimGy8iC0+UZz3qvGV!6And#wF@x_=VvC$ZHsNeUi
zEm+S`5g)|#PeiKeihBJn++aWL>!;?01QrVpGq$Hqitm4Fw=A_P;(3R07JGalr#hU;
zs|Zux9n@VcC-)T>04)Zb$-cPE-HARk=Sc{QH1ukkzra4VZh5$9Wt#D8EBegCpuh62
zJO5%+t>7=y)R3ubofg(+MIWEAg`Gj0k<{?l;OTPW4E;V<grC5iS%yuKwL6>pj$zAw
z8!F|@$MmqU;^N^@tR(ebSS2F5&oYIq2UX9PZ^twrQ($R8Fm!onVw5>gjZ?>GmTh`?
z$%hU+(|rA7_}$<4vm=TbOk{_>od0uiw^;^sKe}GWAa!M!ZKYd5_r6eld@%P%TT@db
zyjf~8$T|mv_pbp0;k($%Zih9cxV!<~`Q83&4zo*DTFf6?sgh^1LL9HN|Kk7Yz+j3@
zD8RPb+2y+=KU<G;_IC$Go67m+5rapNu}kqgW~sg2wrXYusSZB2MLMs`RX8>i{%V1j
zV&T|ccpB85{2m-fm-9mrgH?+~MMwMs;Pt7fKFq$@pt$?heowFR==oni(Byuw^Rr#H
z6Mh5eo7(Jflh%X5K-({`O(dZEgAg~<Sy%}lLzJ&}(B*_DxBva3E0KSpCG?4ct0e8t
zg7ZQ@dIFkmvE6*Au=s73RNu5@aC4QfjRDemc%Uz(_zw-9f3>0v)(r2cV@K|ytks-n
z>6-PaDq1}%6%8b&%fVwQN)NXKU#y=@m%sB_Ou0ka<T0vXpf$<}w|7dO^tO}yS2q$;
zIQldnlfTP6iP9ncx;?lo?5(Yh3Yz2YX8wdJ2n95GSg817;6z1xyba%DQfL2;fi}F0
zTSpDWm(Q9>dxdz^s3O8lypB8Abhc=Up+IcU)@N%X`o>{@GrQ@OI?UQ(z+SN?>v(B9
z#XXW$X;$J3-})}n9VEoQXz!r$O8#H(hcVbKcGxNrk*$S_V#`_|t`^VQ474*nMZYF<
zB>8E5Fs(>-<_P+$ZHb8KsE@ISa5dJET@1!oc;V;VeDYhVm`D}!%oy@QA?lZqcX`wy
zf6<wz?7|8b^ORsJ+#~hBj_F_3;!ky~B&1Naa;4HYwv+zb8~zUpl$Fr@e`GuTzu%J~
zt<^pT1poYOEFsW~sx#&PLXqmFkUHqN(~RL;9KjXvnI1D+h6R~oy1*c^x8eau{d>1j
zF(O)?<i8K~zmvVD@PjdImtvW#gRXiBXuaI%hRNU2o`3hXPzINUb~FKZote(7IRA%a
z|4P~FxMaofRxW=*T>pM%`1cn`4SPxn9tsg+`QI5~nD^vFPT`;VJN@rt%YRF6q4B?P
z`?FdWL-3yzmRrc>ErrG93_s`qAKr+e3b|moEAtCsg9{Z8aBeZ(O+x<5#PDEz<{8tg
z7218|to_osnLffh<F=<#HXIC-qoS$8f1MA`CFmZNvFLZpSHUnV5}M6#E~lM*wwE+8
zamW7_Nr!|IYgDVV{<6`+pNu2c1I(z_fN#<6trsJ%du=!9eP0ZHq<u*i0}VsCU~n0}
zsJ&*_kdFkh?0>@en_&Ta6Sp*wogaYX(!JFB7TdF1@0<(<sNdX*oO3$K_;dM@kiZF8
z@?E+{);Hf`;NkI1GEKLx8~80r#Nt~JvO3rMmXFYUO9R31B+i=lFn)}U5SY#Te{rD<
zLAPQ9X^8b;@@9S!WP>&`Ce;O6?jbSc8<6L42?D%?GV}f6390sbG{OL)TJS$DF^9_{
zp$UjK4AsP2h0$I4PYO(0%Nkd5`Yf(@Jw;W&`sOWVquHSCC<Pv4ezme!Adhyug?v^t
zqu~3Ci1wRn2DX@th>(sy*(8|vSKvaV-RZx<kam19pj3lqV$<J>@J5U%);Vh}b4g(9
z#OCLn^W)a^MG=)ZGktPX81l*f_Vt?fwarG8sEF6FNT1g&8mei!{rVKn7QrZbT&*5;
zp0#xKZB@GEY!U2x%Th}F(EZc?H{P5~o@bG>smW#T%|-4WhEKE<olt~<nBGjI4<h7u
ze_$9cFy$64f#2Ls*s$EIH6HKIq=NGo?^Tocd^@o5OO-!l-0mKwI5;Qj=+{e280wM!
zGRZ{p<0*d(1FDpTFDusj)+^zA>u6~1e<mo3{(jo>3a{eckn*R8if92B_n5Yyt0B3!
z{>K23Z)q332-Nv41EdW|eM%N|-c=A7kKra=-})kM_I%r$g$CaSO&QJ8R`6fE!Hwoo
zyy;hwt7^yiDx|hb44TGME^h;$0WW`VJ{P>9{=C?ZF^IN_w_dMecq6`hwcke=<uJf6
zq(SrC<Xe!%Atre0Bu2|~-1dVYvFF)xF=Fdg?&VJhdE-rfmy+o;uG0WE1HY<b{-S+i
zqgTBhJ!Bye&@BKoIpOoX>OQw*qndeSm_TTe$1VS=)LQXSuiUREm3JZbF7KMFfG=C#
z<t_SJOzyPVLHOz@t$o3K$9Q@h!&MkM3&XvJk5FIhV@NPB6a4g?lhYc9ZQsJne)2_Q
z&tg2CLKTiTV=U;GHyH1u{g@hNBy#sFFifLU+mJ8q6FGZ&-jUCE#sN?9MV_}G#<}k9
zdsl!@d#c;od&SGg(xQ5~r>0QFg3;3sx5SQb?VZ%+qvof@24y!wwfS=ohZiyOK;F(0
zl)CA*E_xEf6}mRRIW<qqw2@20>?tL&9k-f$c1!yy(Fp9Mf{doArM(VZU09PxZ)u-N
zHjTw`KOr~$9WL2od2`DXbLl2I$u%mv9ogYer<ks8CUY^gMXFHg`^ycONjHtSS!TKZ
ze1mZgP0<Q{wrz8_mizGuIFYsZ^G^yx@&7~9S-3U*y?y*Ek}?`(q%<fEN{;T5lJ3rd
zAky72x<quefaGZD93!N2bP34l5)gRi@4BA9VAsy)oIB2WzuvbU@yqxIK?a<rbF&e3
z)5mNy9(5lhc~M~fo@ceis4?8O*Cna@f0PSxa$|Pg-!gR6>FQL24x7vA#SRME5w+@p
z*V=3?DT1ZnJ)nQe%EJD;aFCJy6R5gpDsIO8)$Q6{$6CeNpF{VM01JD4${blmi<g*?
z4{hiI@e9lYEq7=BM4E^Yzgn;f7zPsuPK9ie<+bP-nO}PiLp}A}@u)1?8*b63nDPKx
z(>NSj*NLXd%D4aCABQ}$c4e6-#EPPC*OvZ8B8xZSdDUj^crG;TKEvLmYGdaoGGfxF
zV%1KRn%xh<*KKQj!|jjE8!2IvSv)>JMlK&uUp(?33!bk;zGQl)0(5G1`)ED>iS&|q
z$D9Y|bGu)9o!x=5+g<<EzZzJ@d(4V)klc+%@U}aB5wpQ2O+ovjI#qW0_>5#4|9z@u
zO#YYupx_qmzAVNoK<JCdKZAEMpl!zc>AZ)Fk3A+&eHTZ}=kg9+9dZSIGdf~!>*TG>
z&KmT2OKAT!bLFaE7iVS(GfGTWe(RHcd6|}XGwN`Q7O(ruccLi@)W0O4d6d2L=1qvp
z&#M|C!V!(U?9bk9+I^}!$)}!Zv2=B5<v++Nh+_D<Wl$o%x7`BFY;P-hHJO~$t94a)
zFw!aWhJK(H#V<tdUjNrH>+Ka~+rof9Jzi`mAg<e!c)epFA+q-zja3y2Pi9)(6X)a8
znIJ@pzla$<gE^4)*DHf;+Paw&+_r?Ffw}w)=FvUCKK$HyCnMzI+E&e@KhRF?F?sa%
z9~&da_UVh++jWiWi5STS(t9+>cjlS)`IhUrvFj$ctNgH@_C#A{w>*{7uZR{y=kh7|
zv5>OEku`8IEnh$*mxITI{9VmJ-I`+vdaHUVpciqV_($N4>ql#%U)u{(Xtqw@UBV9i
zl_;IE?0!jTYC5sOgy&kDuibAi8#+5WQjJ!oH0v){ubP@K4So#kIfpd%I{vVk(NQ~f
zf4p+P|5*Lu>*wTuK{?aK(Al`+e)${w6>bR2_`e5*MDKX^0qYLg#n?_>cC5MKWEtRy
zn)d2&SSj|7R?d9*LcJHfF0%U%x|+P^Q4<(${nJYOa+ASsMl<>=u46jFF!GNg02DI(
z?eG0R?#6EXL2!>tX4VbB5Z**A*iv1ZO|=4B|3SQm#(uWuz?M`aOFF@7*6uIMb_A!1
zIReZyVmru21ccH#spcrd1c%QuOGpFQ{KG}8{xlV!_@|k9giN&9{Dfu5<n2tbuScMc
z9VOdKFAa|zUD7kBaue$}-Z)!-Y*Y>wfTtPAFRSAC$CUJYMLG2hr~Z>W7X`_f2rPzi
z(Yv@zPnqta6VG)>-X7;t*g^!(%0ZG(yA5D*``iC}0nDBogfFCjlMF^fl}+Dr`sw8b
zFfp>6(_{1-Z4JdSAylrD{O+K$i+@$cLHzg2i(=lpwCikidcMOqUB|e0)-bLj-^~5I
zP+vLwfNy{3Ys>gvUscBFRrjl9fo6tu(`GT4Llv5?yvfrXZhziSKL=%(T?Z+m%KNeo
zDX;&1t>&2rX*F=EeaG?p81ajcQ_7IZ+_;6<9#R@}yMIV;c8BIq!y9#6TK~O*MuN=5
zGNle@KS9G`Ga=Wi)$V^#7(5O6uIrzjz;~BE2cfu*L<0r!q0ACLRL-B;3oL#|)3s*>
zUzU;oLx)B-l$M-x4rM<&fCL#ww!B5v6ql|Uv>sc1C6I$z4~?@JTX{-cHN2~TT_|RV
zJQL{K%wi=*s*k;!+hzr7$YC+>J4Vcobf(p<-K&g`;UPq%tIqUfq`9-qL0d)bE`*DP
zh=5kOlH^&c8MAAO1&S3sU?BS1rmw~eInz%c5xxPAKK#i3wSf_I)kFhRg8e`jZss^c
z42x(@gA$c<HcIE{a1Hh~!ieZ&KVq@0AG%YWVE}e+jFot*s;0d3zLNm-hpR~Gc5yzn
zeeoIkp}a~ffu<Cof&zbOjs1dtQr)ssIC7(~P7XGF>soE6YRzk`Y|$}9%V$3|ysE%S
z;XD2H-G#?p?__h6jx@2jsr)0lZMk$ouiEUO$XYGglYsRlrL1%<(g2s~R||zoFM_-p
z6@&#~gHP{j9)8z+U+y{%I$k<PBNyQRj5kfZ&rPX{g@QQLWZLRZ-6i5%eBD3()1l9o
zsN0ckR&!w;#aBFR-=pahP8w~W7+O!|`zn;)GzD+-7|9rz0~&0WBXKyaQf;;v+I;sf
zJ^A+sES45=CVuB+&zjwX-l4_9lN7?2SEtprLuQMW&71-2UUo0ys<|2~mE^P2hg);e
zbiSGBV*Qd$vczveyRE_M6i+j!+7DA4F;BhhBmxZVS=%;;8rA4|y$bMWLC@=oI8puv
zS!~{RAJ#;jG6<0r0#%%=>1H|id+|zk9TtFS<7(Zvq9ybu78yg;U%Yj@##Jn%L$+&1
zKYK^LFNe+E@DI{XXa3CC1y#}aa}FEccYGl?+){uJd!lc7ddkN|728}%%}30$?7i(&
zF26KfWT2U795W*foxHGt7wDuk&)l*rPy4q>a%9xkz`pjN6t~{rt7)xWuHzVOQkY&k
zJgN<rQt+J-<1O{k<6<ZI%$u~z1FbOgAgXrCmcVqOYS*Po7+m_Y)sv8_%YWC_b=P+L
zg}H6SbG{nj_{LltzPnxGH_Ky@AY(DYNQC_6nGt07yA-9@pcYv8&DU7}FG%;Qc0Z$v
z_PqeRo6+yt%s#GGPh7QoGxG!=D{vd~k1Cx{HT{tevM*Ot@<-=xKba<~foSKI+lCn<
zxl71Y@UZLYXv^r!QS85^zOwp$#Xl*o&#J$qv_1KdV}#_w&DeQQ=cfYAENNLe95K`*
zv^Q!*g`B15Rn~<Da+dndc~pO~M`k|#-deECuTRZB(ge_xq+YsOdkp>8KG`RO{tjsd
zTZeym8$A92dhTV1r<{q^7Z7#wd)bFViJmD*QV&t&&r8Zh|K9ABbRY5qzLa3P&7?;O
z{PkwDEvtn6BcNrjvDyrNV&P?%R=`;-t3=5W(9@`W;B&f9ew0bZB5pJIrp7NZt*et0
zdT7Sj^lRHG8*`Xwn~`ffwfeupgUr(nbmh0ga{G_ytR@yVTg0$C%N58wFx~zBNV65=
zPjKUA>h!c7^2F=QzZ7tpbv7y5{)o=3n><_YckRbhp&eJ2ru+{GpM>PVtSZ{?58G5X
z!kvzJ7c7@R8Goce$39%R)1%|XHL^mC*WAg>Xv$zjb+=Gh#od@g?VU@|l*yjWREmSx
zr*>e-NCDk6y+!8QhAE-UJ_N_^xbQnpWq-nY+1m!Wpm%gJN3Ep$O4P{D!5S}cgLsa5
zveAr|S@n(9F_QAWT7B<hR7OSXWP{k9o|CWzuYcy19^18Zovp&+c~u)44>LL#6W_=4
zG7SoKBfWhCST~6+|J)2~C0LY^oSe^ypq<TvhkKhe<(2#U<XoWkCU^4FU}Z&JLuBff
zirAA%nQ;=b<`aX%I`yK5pLmS3SPj1ZY}>9mw6Nj5y%`T)4LVh`s2op?+R0DPK{oxJ
zw3s!zaOldEnEqe?riW96|LTq}#q7e;H$XA(9ld}*aPN5VMM&Um5Slf+hB$zVKCDzf
z`Gx%N?~HJ6n7Nz!S6)YP=*uKh3Wp{w#(`h6<Fkp9b|99=#Z;*H^^e5TEqB4-v|Q+9
z=|84<_OHe?e*@N`bl%K(=FmdL>QLB7rchwD;#?NjM|6dG_njMU8qnE{)UmTO@Ii~(
zXCGWq&M>}jwKSO~Mi9o^s5%JNe#jP$MU5Q?bkgiJy2c+1?YYLPv<~<F{2cMjH<bd;
z;a%-7hOY45@VeQ4^$bY<BMCP8*5QuwxuyHN*irlD;bq80qpzZQNA{P=<iy9F;)_@A
z#(ov%!(l~@k4{Ryq|11OkzBOMLND60k-H@i@zsOxRb@2b0iT85KuNQTpkSnrKc4sv
zL~v{mk<XkYVomwg9$9~I7O?MV!u0;8xp~FjFy16`aId_&xn+R+w~AK@f1SXuz?Kj=
z3{_HpksN$cbfE4+Gd_h=dMpO-xL@%p^*a;0-EkN04Su{^lM>x1Np$hE-%km0>b{KK
zo$C2jtfw66ubORiFI)Wc$NaagHnI`u<KNx+D?PCQiH=II;UPf<lA7Eq-jXN1mRTh&
zS$Ip`B4_p%8HMtf?J(3NdoqYH;Z#S{Gr<?M*vs8w+tIrEJ&%du)lv24bJkKGCkg{9
zEXn{3<!OPj!kM}0v;-#o5KhGr7OAQAXw*!PHL>0}MvJ_pDc$V^PddK~3irJrPS%8>
z>XzAW=C?%gyFbIbr8&Z@;Tw&_*GmT)Zl38X%eUPbQVI<7{`G6pR4lhl3+$!Vs_v)^
zF~J(biuK{;PC=wJmGAGR?aO~K=&ydKEw!c2&zuyr=<_3W=%iT@;6ffz%==VH;zrwK
z(b?)HeSn?xq2FugimV#YNf=QjSjS{R+r9?DOrK@nIHRrI-$d<7^9@uR<hM9dP0XZM
zo#JHMbYJDmEVcfGs0{ios`%&-U$(l^-Vi^zgF|q!b{-lq=P~7!TE!Lcck6}J;#x8!
z5M}eiL^C+)fzF_`lBx@HwOb+V3*tWygbNP;slPb-GVa&%hqbsb@E4xY<wBnmw+}Ty
zx^$58TJ&|4>-X!#5p<&^NH1a6;!_5qVRrkFyEOCk6-4cKrX=xxHeRdfnI?V7LFelt
z$aXMDaAbpO7^B8F7-gJs21TWKFJBX$N!M<)1gS0Y2PaE-*W6wHa1UhB9Klq*B+pjl
zANxIJ_>t17Rm``$yxm4dY=F2*q;dSuTMG|gWB#M>N}qiQw)!8DNw-^8_Q~rDBQD7D
z7Td2v!nAo-`9Bwyj4Wa;zr4aUpi|;HP7^-nQ2b?^T!TnBg*$yuz?*75{!D=f8Ub^e
z18QF3#Pb{MNHdZIfTMAqsUt0keh9#fC8jc6SR*8U@5Nt`nTK&w%seq0dJc*rF1w=>
z<<Q7qGsGS(N?I+kI0*1EyR$E!?>1RUPro(AJez*T5FkM{a={*;@oYOPJtC%#X{rc+
z(k%FDI&U`p7q<eBt<MUXdyD(K28-L$i}($3C*9kK`?bB~zS@kEmark3)syIlw?3`9
znb}d1dw%RUj@jDL64wj$Jc85Bv;qz&4*BlmBg!8xCIzxQfO`GM`RaYFVJ*O89BTGx
zLo~w&%mmeY0j207=DSpdD$>KVHor^l!++gsKX;m#AG{jnQhcn-+H_vBO^3z0N$T@_
zRNtL_z9dn(V<0{ynRi*~@RN$u7&N!p#y>ixyOkh>cplG>IX7*K5frY(bKKM|UJb_>
z-!JWxX4}_v{Z?B98g$!fEIDp`5`Fcd(bYHzE^fL=`|pB|{dm%6mJ_XeP+R-X!nqfH
z^V)nWPnI6)jdTj~Xw69vZdY4<cJr>jg!ijCO3`~t@(yLT^KBNinYMPd#v?!Pz6I60
z-oKznJN9gaN(Q}MN_A_G_`-KJ1se<x*34$+TqkaFV5|;kg=8v$LumqLT8KTi%;0!_
zz=}Mf#6NC<T!hbOy><1)N?S{d<Ug<4(2|<YW<R*_wx;|zJM8~>ZyH)6(0prBylc!h
z_x-^Rlxg9O1=*LkS$aX9_G7q0g!`+RYD`KNzw!!T$>1jf5`y|swgTK;3cBsc9*Y*n
zYjgY@_~lfOlk_JFo`s8+*it|1o0jfl_&k@B^=!CO%>$3Cw*x$SYPK~A+<I(5x`b~s
z5!)&Y$_g9}QcM<2?*Xc$<&RzlL~1?T5}jwb>Sz2zFms5eV!YJ9zN3$h_#UY94M>-v
zw1K1GD6uC)^f^7tE5SiuuiB+p;V&BGOIK2RFB_XhHK>YBDi6-V4IUml5eCPf0R)Gm
z#0oAJYjZF()(=q74<#T!cDVvy$QWbWO!;eK_Y<9Knf33?ssswj+VWJTk=m-vnuA#q
zX6*)d3Mm7})x$fYs;S3vXIzmizAss^<z*4|LS({!fVjZavra9`N~SJa4y&O`fx%8O
z^uvYwfWjRlpPM7SC%rHt(07YwO~gq!k`F6l>Z!r-!pd>*waU`Beit?5f$Fuvp?F44
z_q)5=<8@Wi*suY`7U$*+j6;%Y{3wB3eVliy_i+PSSCc}QF#_r42dUvMCq?gjQhQ-j
z9~-qKoBM(kvpXv^(}iuxq7t6Sn^R~)kUq=!KVl$88lK7O@u%N?Q9oJap6eO_7~`!k
zG!4$umw^v(20i`gI0tk6x8-Zvu?}Yhaj;$bfEkC2yOfW(L3O&u@M!@-$su~bel2O|
zADRxh5e!VJAJK$|C&wIy5#?6RG~JMjU(uiE<IvCM{p<nEZW(bKz+%h4_J46pzJw==
zd`RY3Fux(^t;7r4TjfH_fOP^lwr3E&FU)$~PqpWVcQ0%_jIA}PF+dj~*E{1q_TlaP
zO^~QmN!l6Kz{G?@iA$RFoHt=-N=P=v+4MU&$)%&KEkZ4u-v*@c`Z;gJ!-X|u|HV=2
z<v>JbRt%wH6O|ydRJBb%6o%*VtsU7KQ98qZ5C60MLqe4*-l4aD!g@P3C$*`L39FI9
zuUKtBVxQ@IhD}y}Tsg1P1kPSS0HquplTS`voIg4`sW;U6Lh%G7w{$lI@b6K@mj8>k
zvoK2JZ=F!Z6kC&{KTE;0KO6G}SZgj7A5$%17znIS)FmHv-W}=Qk<y7AP39nQ!k(yQ
z!6oLi7If!_`w_PJZOIe!L>B~NYDNmWcIvs<9iGZ4#z5d5GUAGH4Im%%X_O5oC7_Cs
zHCc`EszK(=Vkth*2_Nnxp!sOe9{F%UzoIK8@lQ527X5CbTB2q=)q-(r8(-+fpKhG*
zqnGXX)w7wwajzMDBq}uMO)eIiOsBMYf(-KP#rNt@CX(bdFaM@2M3|g5Rcwb!N=wa@
z?nrElc1$0G4GCcme-*UKgzr_B1hi|+8f4l<LX~rv962Y0_f{`<FLAx<Q%r0T<)u*L
z19E+jTEsu4i5c6i#%M|e(s6+TF1O=AR>lq<S|AUroNQ`*<GJ8_aKHA-sZog0kK(Gz
zArZZ_hd4QfYhMh##u<t(Eh0Ovc76JbuLomHr`ImG3Le(!7ai3FRx2oN+Z`IGm9m7k
zg%4j|7}~k2$k|{775Rqfljg0IxSq+m^I<pnG4Von&*ZL~wuR4L<@G%kU$+;aGqM|9
z|HN$O`B>XWm<t+-P6GA*6#V_)6Q|y1Zrg=bPun%=7!H>tQ9rCEo+DL2<KKzWfvLxT
zG^E^@noj4pBZFOaUT!A~l>)-~W8|LkD2i48))u>9^h*vn?c}#kHvKkUZUY{WC#4%&
zczYe!>VSq>?rPi5@7<@~cEJ4P*e5BAlcn-0J+WmwZzWkwcGUvj9D&8HlVufeCbeEa
zY;+orVEO4=C;vBXSlH7_y}%)N<dlKFBfbB%OvJ=`R<<!1O0(~1N)9NX>)z~?5jXts
zbnk?Y8>Unn8BRZr1EciIXIVRm7XFg1`SgH~IP|SZqSxkkxv}q~O-?-WC@NQud+c=I
z?0a*LdV7mhBbnhJ?kP;f4LDCf_bd%Llu{OTuY(x`r%9poXR<Q<B9W8%!M<XOCp?i#
z?Rf?EwM;0?2(yZ1GUDGe-3)w|vM-$_a94I=yFsZ!PaM|^)O(COqF~&F19?m=z{=hP
z?N;SVqX?hu5{}Cb`$gPYT>HwMAxj!~X_LA(Us=wgBM#X!KMo`mKkTIhy`_EKc2MKh
z!a2R5D7NRZl^2yE2K#k3jW6Y=99ERUW`y#*do~&LETi;YR79^OF}kHt%Ww-Cw4@31
zaRhF()2|eLDswQT|2U&PGUm|EStCxuz@EAVz&_)}pmIZnijxjV9QFfB*Dj{DzHR1Y
zwx`7v6#t%<`BXI{&asBe?^()vj)<ib_kL$oaLgn)q`&pu7m{$Il}&jjk^ySrYZ(%3
z5m3JWRm9!w8Gl9<&BVL!4M?dzqxW5Z@=>R<!vQ5@Sf2ZwjAc#I4!+j*@3F?&R9?&$
z%RL)`ub>~KGfKq_!Sxdo!V+~Hbv)<fj^?RT7W+q+kRedMT@?S>gaJq$APeb~b<6h>
z)YiOHjJizs48y$@UtrszoYx!w9pg?Z6c4auo$`mnfF#8G0(BT4juUbMFPBN{wIAW#
zzTV0xs*!&-YBP7x((?wHC4gB{_J3xwxfGx!(vl;=EKAf1lKSadTcQu!PkX#!#TYBw
zn=v(pGu~u=ZF|&(crtCP9j0&3vP6)>LNL>+P7PdZ3(}OlXB77z@8MJX4gM+8fK$S2
z60tCx!tw1;X+oxm<>}77>hLocO5{gkdTy8jbEXtTTRwZ2YPjkeSH1C#EkU0dt&pM`
z%8BUW^))2#A{p0~YHC!%mPDe803E|N{XWTemei~`PV?z`duv~iTxV)XGVi@8sGyE8
z))aAip&yIwi>x3~p_aL2^NxUojl?Mo#)lT;dg72wCh4BMHli@s=0>Nbd!rR5l|Eh7
zR^>O~x1uwQX7Yb=tIQ%rTrbCB1-<o}k%V{JH;L@S5|v8huj(5Tdj_8i2+HOP-v%dm
z^cy8`G6}KKEd<KQ?>)s;TcDaAO2vii`(xdNvxM1Tjx&D(?aHy^(MY%1;lPwc6dlP(
zE=Fqm7Wm+bjK7t?x|EX46oCdr#)9!SpHKF>C4d-)5^HR$+G~#qxtmOvg`42@aIk+`
zz_7%!O)|vCH~jAQ^^!oi2RrKZwxzgAAs8$5VDaRWE}<>$Xm7>Xb<=&W*F5{)4BLds
z@>{MT4p(*o^`K`mX+&V{EfS+5DTGobX@UU6tYdt^+WJ4(p6H8vjhxhoi>F&}-u6Tm
z!qTfCkq}E)CQ6Vv%P)0T)p3a}YnNn4_f_{Z7DWHX`KpY=RMU29yW>;P{zyR=)zhV0
zr>(blJOt@4&J45G^;htnhS<_xToy}W6eZT;tTi!T)y{96%BOrWkIl^lpm!}UNpvTi
z^aX*qvS>K&-HU5es%MB)Tyi*&Z>1E7_D0c+U()-u?+T`lEF*kL46kA!uVKCF_xmN7
zXC&@i40~s-X&k8we3+7p-6vK<J02ukW*T5bg?{6T>Av9M>E|}lOf3j)Q(ClaCx{^Y
z$w4|8$~ULa5pKWTe=6e(#KauN<^^@paQVJ(2z!sKkQ6?=R6?;j40NI)9*vi`TxW19
z(fDCC(-fv9UmV?&TnWIk7wxB4-|VQ4t-wX1tQ02w(68YV{7}F)Ku1LDza>j8f*o1W
z(N8fLO~`&|Sdz{xPl6D~Ih*NF_n-L1S9YdHF(iskrp}=?Pllu&YMF?3@ySKwpEB3(
zJ*J0>%-Lt+B%K7R8Ih9Nz3KZ-U~edj3B4$0pWTsyL<sdenf(rb@-1oT4@2e4Aeq+<
z<5k-}Te7|ujrD_^H%Vj`-kb9LtjR-#R(!z%AdX9$8@{)gjEed=BueFkb%|EfX;Oe$
zt=OSytJVw^c_ul*?d(9RHDUvR)H7#HdJGw@v;?dFerO5}w(NX`hr*=CRC9a9^_wA#
z*Nt(r3|A=yQ&)u8t~A70xJ1{3u@at)Y|Fkn{3*=7S%4&m2wxA2uME|*FCyH4Gh{)y
zIl}=g?{ktD5?(iye9A|oiC%ByaW=!Xr2uEHOhRzu!(<%DIz1K5j@4+hr*<2I&5DxZ
zl$AZ+neAUzvW#i@yw_K2AziAW0X0oO|GN}7T6rrr907VoX6959(VCy&G;)E=1pI!*
z_9c@ObhSx$@oH6SSVOq(T^+$K&~qW^u%wD;Ez#j<KLpY%G^&JB02K#RKBFQH_}w2F
zaY3SgagMtkK4f!!;R63alO>0my!10x$t6rW0m~gIOB=iqU5d^M-8!UfExS^S$>v#M
z0WmB&q_*2xe5Y0d$>Xeis$>yQwNQSB*~`Z;v3CSTL#_hhbfx2;FB~JpGKj|iFuX9P
zA}dY}NMO7oZeYvqJ=Y6i8l2fW9h9<4!wMSz4ium_Li*RlZHwNv`y?-tr$_3aCR|4g
zlZ#Fz3A!>u?YC3;1;PVqq$2VY*0z5Cr`9StmB{?W1TI1%FFNiTR74>DEQd~LO2gfA
zTFlLn5{Vy<=-BXek&Q`@yw!aKw>cf^ncJ{oAf=y23Igt9tC+atQBE~qwA!Z^ddHIU
zB?g2DzW`J8NDvRT*&s?XfJa_s94zJj2~p2o?>;v6tpiQNE<>*AC0PqBXvkD-V?sv8
zLj2DwoOywX=@=oK+~B(vdHn?^5aCiu85ks#dpanxFc$~wZXpmS;{oU9p+9}Yegn#2
z98I_j*%FsdD~p?$?zuqcy-RnbR)yP{X!_z5*9=O1Nd+X?ye(Dc4$j%1`z`$T>bscR
zd;}|@jGzVQTF!h!kteFrJH!Eohu%R=V<gy7iYhN%v4&F?>WeX{7O-7?4hYrvq6J`c
z{qzH0WTlPHQXl6+ozFWat{GuCc^f#BDifr?Fzsk?F=cxN$;V-=aR`7Q+a~eyl^Hd@
zJ-}LD+acMA0fReMX4^`9gqaRIhU&k4-KOZ(B5YQXAMQf=RRJF<1#7XR(b7_Ic4f|e
zd{6Ea#W-)w!nDO`EgbE^$Nj7{m=$Qr(e)ra5gA^^18{}J{?YPBUj3wD{q6K7PQvf?
zg1_)i6vh)9^ZyR>+wt?&{ywqK_>%dDOUZoSg`6!S_f}2}N#<y5$)^*47&7}Rc$22g
z@1=GvhTvrtn6a&|j9$&8O$qc#<tHBv204~01#>j|3OZR8ydaNNq_0H@NId~KZ`Htk
z%xyLpmFeNzaMjL)GjhR8Kup2Zk(n~-s>-dP9G|U*DGAll4bYuAkGz4w9`O>~Kp+G3
zdn<{&(r`;T?(^UHAr~wdi^QW*6_r&phJ4k>lSg911}j>FG`C{{6FvDpBO92LPF=58
zgn3}4k{NkSv;tCRt7*~jSlkmmrCORa(an^^roRj($kz^xS3sE!*bHgFEs3?MIatu@
z<(^?mGhhFOLU9Cx=I@d7-;l5wY?erm#h^oP*G4%SsZKRAD(kwpG=!U&R<qROwvu(Z
z!fG_c0svod;$Xk542_Mcf3n^)xgw80?H3xhOh&0dPO1hXY|&A)sYt7~uj^z$3KTr?
zxIJ_Z`RnaEy#@?F;?7*y>L7)q+mUpl8#!qP_MqhT0AH-hYvBF=nh{zNv^v4%2jW$~
zF3mC#2DUx_%?l6Y?hZHN6*ydpwta!!0Z^N?I~(RTAP1o4-NW1Xs-~&eX=N!txb3Zm
z+E2eQ4~my0JCmq}s~(>*wnfBdJm|zq-2W-A`#cDAshp@Go=@#Ola~fDFRx6mnZ#gt
zs1Ki?C@43OZ5{28+;W<PUGwDHxB`fML}gwu`833RI?0CmFe)sjwI^u!Doh*?nTxG0
zkXqNN2b1vg3r!O&;>rjq7$54jIk+5hc>9I@xWX%@{fnqCn{IIqDX3R7XiFsShpwS1
z{iEsRUbgg3f4&EF(e4M8RG?&=N2(}Xnfn&!ynj|VC^&3OI@Z>3x)EID^;E3s4pCUk
zZ-wB|8!OP@HG8t?_GLbKeU4{R!57=y<BfUuKwGhzW-cp2ZK22OXH($eIl|j-=ddhM
zi9x(KOu@nxJZlu&F=tLk8t_F1A!s+X8!h$;1bp1O-*n<V^yW+*cQPPU*~X*Eq7>ul
zX8%$PDka!(o5A{O)zOD84D@N<_X=S1B_d+p?=sFs^Cx$pa}IzQ(`tJlv-VpVNtw$&
zY(6XG$65@UzZ_Ce3QOaA^FEOZbh?_gz|4hPW5w5JrTOGj&YL*a(LUe9G~-_J<srNr
zpgekS47)GCV7>xmEc<t?sT*?WuGSo(Ni&Y~0!tkozjoZJd{dN;>G%jv)EgxEz_2>@
zlT~cE62RBqbQCCvC=KGUu+pkQp93H)qc>kkE6>ngpQCo8(IFIU*0WUS`s;Yu_6EZS
z-d)Z8y$>c+%gxT3>>*>7FCw4cFaUXpHEP+U71|L-lvAduP3^}rI<G{j`ayiffD&-J
zfqEmy(9ZI5zpD4g+@F*Pa$$JZ5DWH;A4pPzks)0*)Tt|@R7@CYRpNPVo#xnUqb;9p
zEpb@G^uhaG=tua0)<hstUCQW|4C(R1Z-Y+bW23Wao;$i<BaaYmMQ2}t>PWwCf~FdZ
zDjN<FBeRAX<O}o{tKN<MR7a)q)@~30`{_k{YR}qK;!a{+MnH#mXpd4GzMP$otMP(`
zWURy39(L&MY*r-t!=4bE0ANZ`0p|uH#b;Z7H4@j~6*mnZqPtRQ+UqP;lngm0nypLl
zJy-sKEJAVWp3>iLqT@Dc0HW|@-H^z-W<d7N5|dfIFCEw3_I6_WJvuI4szv)<17U;a
z{HMPq!=>|xiAf5!oX)FYOIf<@3t7<*QRL?|H|!eFS{wg<APh$ks)?-jfIo}8*#@n|
zjL}$?qsjk);xI{e3mALuP~abVX<8+VN#4swAz&YOf;?fv^#V&Q*&n8FB{jzr=O!`9
zLRFc!Z=r?+(dzLkrCAGt+N5fFivYGk2R!&IOs?aW6ED*}QIW`OwgBhO){o<c#5R{`
zi%_3j^4Qbm{@f-heAxW2rMrfgWD=Cz<~_*TWR<kJEZc-JB^0BwjZ1CTLeT;6t`fws
z>wt%AOn-FNvucbhd<<21bc^fQ7;e5fvkS*toJ&I?9=%a$g-a;}SJ+Mnsvr*hmCy+^
z!Pz&tKzF*C?p05p`PC!uJ?7dq60-YOU)yH=xl?4m;z+FO&40>}0K7DbTbMRlG769S
z?vBP`rAs2aA!ju8FZ&v(TsL01GR$E|Ub0<r%y0av2Ec5sexRL`p2rb~>mVhUGgIGK
zl~Fr}|5u0rsKhZQ9X_H_U2RMoO<4@;4a<qsHoGbhrstYpH&DU3r9W?1xTvuhf8+JE
ze#ok^I(V+#lAK9(z#w;-8igx+n9|#0S(|}H!o{nC;>NgmsXYYkPOpr)^v8N8pum~I
z0Hv~TiKw)2>qHxVLO!-faxCD64q&oS*=R3qXR0q!0`(E9L5|mJJnlQ?vcX!W*NYCv
z&U-f@z`x(jfgNXdIt1E4dK`3j2V|2WWC|EuZJUjifWOux{aBO)gw+453gBWklVjIM
zCK4d{DrX4W^MUgX6}Gb6D9{z!5TVhJ?=*aU;uzlk4%u=_$Hw-<GS1bU1ITN=CT4yF
zs``bGt4;Q?R@wU4rcHVv|64=~j(z|GwI9v-LgeB<lTu^;xt+z2pC1}*29t(<lkL3W
z@f8BG`}HR7$Jq4K3vjjX>@g;!V!g6{*}qzg(Fi6a>$7XwWK>54K)b4hjf%tG)e`6)
zd>wJ>){*tsNlCGd84|c36zMTH*$+)A?kg7ukNIA0vrGK&A8tf@*2c~X!{-c3U|PXk
zH%)Jz5Oq7HGxfc;#Fqj)Gl~iZdZQiHB7^tT5xVAq=+<c)lq(aI@*~!S-dRSznjE&<
z&3LvfGR5OUm(_fq7vr#S+a6}{iYt0y+FnDtc5pEUWxd)1bM<A7lVCA-S$p+CEu(pd
zowqkDJKJcfhT^|epTkL<jp2Hve36G6b2QrezeQh&8s;#^oApV)Z#kYmph5#hvM}BL
z1JKX!636n)=K#IQULP9L_m@b39<tO>DwKYMas`*@UY<jY91A<OgUVXh$d>x4!yMWp
z0h$B1!N+qVmb{pds}QzR3no5+g@lWcmsbI22KVhOD0nU*qR;-@{Fp`6#z+s6B1i)n
z7K<u!0rH$4vvK<6NdvfFsof=Q!!YLnuQkI84XV`m%4M$@yDtRc8pMA&GYgDjB?1sk
z6QiWfBm53SPnp@XZgo^&euNx<0_4-eSAE;1FJ7)vfPmqcu4`NO?V=4IJ>NUrXY3U2
zTq`@E#fmmcR6{10QgboQ)UZ<SuV~8DIVZJHImM{YGp^(0-3Ff6&o&J@puue1v<DSm
zuQOQyKqKg9K%)`8LOmunF%gcTfGXuPq}Y8}b2DqCjHG(N7;vP*d0gVEGIG7aR}F3R
zAAstiXLK*@#_3TFK53V)EF1Z$j9R4RV{44LbPkM_7~nMqeNh)+?mgTsZTI4<^ZsX-
z=HIyclwJO(buZc#Cx~%1IaDVJ_x3LE?jX)VAn8NYsk$`br+2bKjWOjdrBzE}aZtM3
zwzX^;1n12}_CBF<+kyo$U;!U(LL~PxOOSfxEwBF?*OPq;6H~(#{)Fn-GJ;;px6_Ev
zu<LPIRF6bELVdk_9KDvDX%H+>KNmjx+RRsvm6m)7^d3_pq<!E0@J0h#SHJgh*iJxC
zQvHNuUNsW!R|mbyx~?ODkm0Hv&L7KkSFylK36xrgN3D9c&kPydBXCYf-*;W3Z+{h&
z3&&OS=!WoDeoLpD#0J>xscM*Nj`ZmSYyQ*8{X8E<hdNv*cue%Z90yiz7d?FdCV5E>
ztHH0Z55KwTM6zJ7=F#3Sn8GqFK~H@TWS8v$AI781+6PcDr4vOeeVdonJ+CnH!Z@;U
zYcOGyJHpclI=X_=zmk}YI>|9MsGrhQKqwo~sYO7Gg9!G;hZQaHs++~{`y&2kJlg8S
zWLtYR=8l$di}xx7+!vZpufHibw4`m{l})`)W9!mpbZk8}ZBO8O_~3sE8u%Gnk*A5T
zg3^$)KkYJEjbw4Yo=<pJ$g}x4+wQD}YlE%qTVHi?hYlteK5C?wc0umQDS)5cZ{~Aj
zn~1Q!{L^;WG7C_4aEh6p)iy3bSv{AJv;L3=Q6uzcjg*Fkh;`nVBFGvYz^NF_ZY;U>
z-TP|MG4K)y1tV<g_ogwvBnD}-k1$^Cy@co*GCP9NHa6Si{dvBKI`nc-?ii2iEF+#4
zW%D0&#D;B{`;)l;8-nz<yoy;*Urc#zRo<42sxi$<kSrkfze}*r=$Sm!g+z!MrWB3D
ze|J}J2Ej{)NslwB4J1fh5vy%j$xE@;M_WM_rssjbr<g+a^{N0bzR6srt?@rcyKf81
z(3t=k2bS(z8#Gq+&xD$~=#9W@AH2f@ace#JB2-=meIpo-VwlwArZIa-#ArB8wVlod
z&X!3g2{=O(X$AHZFnJm?`0VH!?N7&=jvM)_3x9g9zM9g^_GL>NbL)pkpfSUepEP!X
zAU-SxmyI*=Kp%*Uwt-viV`E%uT)VnBLzj#whgj1n8qW`s?iz;ZSFYZQ&+Hq4BPHA&
zip9x(LxeDB6#9{*R#@9AeCQJu_#;|dck_<jl|;mdfPDL+765N$_&`}d2<`kWY8NkN
zKp(J0vjq8iDCF|~ljd`R^G4%Gc35*S!tJ&7_^9{jC&#}KK)8mXi?zEFBMX3#+^~Al
z{saJ2!*h{AY0t<xF-~wh0e13#)>E|^dd6EqdR#grt^jkG?f$z|q<47@{j2Yz%em(l
zd;)wK&yW$)xB^f&{t?A>l33;91p@_oU@>+EOghK%-IBV#ArDMRv{ov{h(Z<7U_=qE
z-7+~+^kj&kOGoBMN*``m<{zz4ZY!g=d^8AgZR4zBUbb%jSeapP1+nTNAS>(xGB{(C
z1%?x~gA)`7ns@s-L3ll1UxR=}y>rVH9<J{!KjK))Adhc(P@`uC`!Ory%Lu5Q1p1ou
zy`wb#ETQ(<#^YZe=Y=T|`w4)X)>`B|nkv6#?zbxwf!^<K&|EFcGo=>r-)Z;(HpkB)
z=Zr^ZV+WzM?}Y7dOpXW8l^{sg`epUC(yH+&+VF*0326yB3AS6aNnSaS5kCxi)<w{<
z5<>z)=4-SHFWZd&+3|VNk9fUUa!+;KtISq!-LzU)%YReNtMFxGg9@JJTN5`4(|DMz
zk?<Rs4)(Ret2O*t(xQwN{li<8XPkN_jMW_^mCDIUHzTtL^(xE+(8Mj+To6XUctO7E
zUJd>INrg`bB`{?I+y&8>Q!5NV8f0+runJZ;d)JfX`1b6)wC+W5X(RuWb^RBEJ$%O^
zecs&2B_(4)!{O5M0f=?<Kf0CrnhMal5&|0dhNV6Q1vJbGb&rU+e38Y#-Tu^ZsZI^4
z0`iApzBo4h-ZAs*iJzk65VOSlmg^NDOi6NzUzU)XtZ&R|n(df)K2|jnhl=bj;{#t#
zQ{`)r*lZ@AA4|Ru<2#W;$e{(kR4YV9fdy7e3}X*LiCodvEC89&EevfLS#gfUTuK~?
ze#C-}MbtH2Z}ySka|m~A*FVR9y{y{ZLi5y!$zMwj`{JN`31Kh*yoC<)<$~aaw#Ct^
z#5$mLO%Tu71hV3s4r;~8dI;unq?E~+av!vvS@;0!qBl#SCl1pnmtd~iTBkJ+-5!1K
zKI3c8U9*(7P`}s^52m=y#P~}5D~JMP;$s!zxb!*+gx^x@Q{TPBnaz-6C2y%RgA>}l
z`7PCOycceeEZ-*a6!`*fd#!cfSgehG63nWLU?khux+k|_@mth7a|7Pe0Jh+?hjr!k
z=SG?|z<G&EJ_#xGtG93T6FzLGQ((LePdfr28{y%SAA43Q408x=^x&4uUQebX7Zm0H
z&oP&uV=TsC*`{Ad{V<#V^=A^ehZcT`<OzPZ=bb_~`x%K7lcHZY16D6$I%z9I<4+Qm
z&FSA0XSP@N%v2PI3?RtCBZJf|%+75xsljA>FsTsYUSg>wz`z&<w;w9193E#zup=b~
z(hHw(2DRy#4v__5X@4oCXRf+x1zIQf?n-227<FhPxWVxS`GNIKbjJdFCaTq3_1N(-
zK2+;5LDt692yC2Y-y}w$K!9-|91~0b4zlPZ7<-%1deXsNDgtM`OVh4o@_kXaLK+lJ
zLWKW#FilQ+_*DSB__GqIb!#(Defe7P^}t<bdaowq{I}<V*#Gi&!&gMSo!e=!0U2Q|
zt2vn*>6A!`1w-cgZ<}(PZL-tB{Dc6wfu|b~@=@kLoh##<QVXmbWh%GC=XLSyVHc%z
z(2Z_FMX&9!<&x=!9&ci8>_awqvCvq<G234zFqyuMj7fsX@Zy!vM=s4H=A`I=mL6+K
zYt0vDu4<^E2ya>wDdlV?Dbr4NLYt*v=p^Cgo0wK*-Q47d=`0szRAbOk2co9+1y6Q*
zl6SGYf6N0J{qy%0u?y43pl0%9iRey`sbcLm2LqEP*fA1cVj{qAeH1xAI7NN;(>6RP
zmbl#J4ze%!)}WY3eW}ft%dCs)1J_NO&)N8&3G4+L^b9A$=(|qFgkRzyxwTU+QP~>T
zh*m&&JiC&*e3Qo>zEE#AcAdzDlLz9IZ$aY4Qd0~2+S~mLTB`mm#|Vd1!z1QFUz%}h
z1V~J;SOP?0kmpD=?mJsy{)XM@m>TUC)XtJ!a$w`><2pj^9bVFke0nF{mG7XCcUpPh
z0_g!u0~z*L90D*``6Moy)wY+51_|xLgfujx^xBa2Kl-7%ZbOH#`%8!u5&|OFfqLeg
zObXGyBMGaPm7EMq<n5D|oS0v*hbIY53`&d_$S-|i?WQ|rOSVm7g&h6~4NN$VgA#{j
z+omx8hRrJUr$)olzy-(Y=kWSIBB}Pyk-M2FA&bJd7)CuGlEB};Mxuz<Vmiwun#O@P
z3&IGYGgjP6v;o@<KfRj4Jx~qnOj#^=*DL&wbHph`>H8h7eh@!~XE1enV-X^}@0TbA
zr8%<NYpG9gM$mfUSQ7XBLIZGb8l}rp2HNYJyaL$-&Ws@)_V}B$n3QN2$XXPP;MM4X
z7RZmN1m&it!#^ei`pFUW8cdB#3VUl!0;v|;i$EEogWOfyQiGETk|-`BR0;pL<4HVO
z`+5oZ_O#)Tq2#iN0=n|AlYW;b_nP_QeO=rUlA*+2$;*tq^GvQ(*5eG0DW$$nsOj)0
z@l#;^EPk7<(adL<G;*)94jdgNvD-G(0gSYFc?zJf!WsY3@!9Kb2IICMYR1T)x%7(`
zdwmD?13!!TEEoMJfRc?pn5m7ut0rfZf3`7>xC{wcvy5hWW>qjf&`on%t!JN<`Wit2
zq(R{pyx&RktqyjPP%U)DHzU}D?M8UD4~`C1J{f}rX30)^!-WXG3J}8i%BU`ktsE_J
z3kLFWqs*y?DI%zM7S#=DfB?qW#+6SJ@gkTyDcK-F<lQRS@f!r8g98e!@9#(|gR^c#
z2FB(eMzMdT65pp@xL^c3JEGtps%22+OB<;8-A3X1gn|s6?|mFPn!|AnZ<y_#L-;Xh
zM&^vuUSgIy$d)-P)dMjyLK*Ua->~~{V#x$ku|FJ#D-o-ZYe&30a9**78Fjhz@u5$e
zuBME&77tY$QYYmcK?Q!tpvy6NOFrP!Sqrq*q*)7j{E9=(SLVQPf3d^x=`uI!jLfgZ
z&<I&X$7gMys<s(fn!L-w%^gSPL{y_0e5T9R|0X%NKt>RvSx(fe)HQ9<c*x+3`0HEK
zDlM{KbO?|NkI;6~k&~LI+*y3c=H~AUt!uYZ(pGiA<A(WI8=VnH4lmRr0e(9jIM2t~
zKEA1_i)cu&qqNr<RDRdOVS|#{l{jG)MtG7DQgSP%d2#s4EHwJfCiN{PnL4jA*VsEp
zycT>&9D2U}eugq!T6CZ&^t>|u?<sXC<qgEO;P1ePZF}*#20vRadHRgWCKCQ?-4nSX
zZ%qpZliiAJA@Ek&ZXmIf%@enR6i$2_x($pFGMv-Eln>lF-@d{(#NWFu1v|A$$aSAX
zSxcF^p~17?G?KSg!02K<dPCX1hdKO~`2%P$rNmZkq(yp=1mXhc_VC?l46-nV{+G&*
zZQizxbMzn_J#h~iAenJmFR%GNjjjxQS;05aTa_}MkQ(TvTlA4NMl#J-v1gw(>(l#5
zLrYMU4-JI*w^C=&`XZ?*co}wVtCym#JX0`8@^y8A6`W23!%``ZQubm9yR(gISJZkT
z)9{^WOOvKymO){!myr4d^0Z{{;CC*RG#&00tKV!}8^c&r#XxmIa0^p80=n;i2Dl4|
z<I;GF$AVe3a*u_QZPxY^5YO${x)2*epI%QVFLybnCWC37*RW!*uq!bG*~66;zD)0%
zSepjrpwfjF2Z5B9v;M1SR=)FN+MDe#53C+i8v*qcUij<l@a4?v&ZyU+1-Em*h!xKI
z=v{Zp;tnT5NN8oAEBZZeb+*7Y^z3MSld1HArG9(pu83=jWyw#)Ns3?p?yGBoT#8v~
zUT|yPFHEQJf`@D<IT8d%<R_n-T3utJ3DLOMkG?$0T;ox~X-)RwT1)Y-FPRTD?__%A
zyF5dV1qztQkuN<dWf11Y6_I0hqk_-jEFHTIFGNxT2<Q2d7*rr_{PfdB#!Fd{2{__r
z`fY&EfLVa_?*<kV;(%2f*^0(N#0VL_OldGAZgnj><t5ZEw=d4%rHR67;qg#KXEi|@
zD||C(_=z!e8Y<Gc$Lpn6C)G}rICN1hloBxC4u5<6(c5dm>Ag)wOf?I*HM}VzV_s^X
z?LGf$G+}rk!QwBfODk)0nlm?Gnu)8!p+{Et1Kuv(*4`%x+rJY@?dd`J+}+Ykj6SmS
z0CD{n(aCo*&^Q|ayC%sj@;{S|&Wg&ZIk?HVs10ub_m^_6MDR}4*<|7@zF!4*6IuCr
zX5N_Bo;y2HLq*>Gg%8=bEfLzWs7O+&b9s05y4<`9oHv;1cziatszZd_aE{?d_a?*0
zw<@XEONQ%Ti_e}^J+-XKrU6)s;mLAkF{?^R+BZ-cQCG+ilet=gRfIiW%48inqNp8+
zHB$+nQl6_DI!=f=bN*$5_NX%#Ui7;|uw?yO<Ow%`)l($G)lfWU$>0yOGQPLk%DBRc
z`T<g%x}U9wzJ*oNsWT<LtzjOgR)tlNeD7;0W`3g769EQ_Oi$UaYwgJawc;g~sNWyf
zADRb{Tbc8S@fN(5?v0zDf3NwpmnTDHM}eWSn5u^aV>(FCGSUh&qQ#rckbf>y(=krs
zn!uA>@F_`zo|&L|<HvY4+~D!v+%pP#*5BzT{3nh@{4Md~02!2D=@V{joExDOoiXAs
z8{afzsmS{r<Lep>z@G#Z`QxQGT$Y3eqCKSxjAJ3fC)#CObXRh~v0v~2)1S@*=Deqn
ziYtX7_srU>A3tw{o})u&Y)?jDei@?@NB(VNQ<N851w!hX$UsbZGjSLnC#>{Pi?-;k
zui;OFWbGHWE?fk59F4>8(Qe=!6=3)66E2p*2gNx>`k{EE?L`u-wiB)*j%b+RX!heQ
z){iYZ`?5>!Gqt?NQIouBUZ7U3aCzaO0I6{Ve6&zf&0FgoF_4^9UKLYBE0Bbt55}|!
zsDb}ym(qpK_R+2>_M5u-CN<+$$kVPmet7tWj^tT$rVGh@qzrMWHg6wiiMG*TIKLQl
z{xRlskNQkzAKyCQQ_?LU;`M4(rw^xHQ%0%Y!jOdxNuO$4f?X3-FdCBR@H++KG9)38
zHb(51d=1Z}6$&1s%xJtk%YLAW0#ReP=$0`bUcGvhA?-f~xO*xYLDEm#_{&&H+Vkis
zuC0LZB+#A-Kh<*&BxleL+UbXny$vdQweWRSb@RyRtaPo6Xe-GFE?|7!7=&#}Qyez>
z{zWQPWa4e9Zd#n}&-aTT_o!2qTCtyshM$7i6zN`D0p{}T@c;z=DGirhMiQTy!DkvF
zud9*a=#W~N{-j)E%3!J6GHKdk=j&hV<)5St;@h7(ca1K8wmchMc5vhmUtvd|!Gh^B
zBC^V(n+X~@rORO3oI~8rPJgh9Y0&E{Q}P%t{6dDS%tG|ol{>HZWIBv;Gup(ir%5eS
zsa$sVAy-FFm*<MU)BGtm&<ejG#RB~NZ%W=+9+yt?d-Q5?LMSFifxMis4o2jwIC3o6
z3?|ud<<JQxIu>#jw=xVy`Erct>)f@-#;@nAL05~Vs{u!5=b-DOpeym?jK_w?hkARB
zzhK&&-O<@?#VvNI(TB>$b4;5tF$(|WV0VB*&ZZxt5Kb$rYlbzR%3hERd@+~m%R_je
zd6*xV;Zs-H__-49EBzVqx7O`6DW8{NtI%B1%0Z6e>i+n2B|3|L`-K^A@QaXF?DKdX
zJdU)?_tqUu><mt~AJ4+mxF4jJ5wb}7mo`arU5ImA@=GP13EottDIp6NA2`iil_Y}?
z;_Yt`T&bPms==Zwr}_avUAX$|xyvKyNtpFYoX^j`%8QH@+Tu08=Uvm{RXtvK+pg`x
z5AykjCU^OX`;yaMD#y9PB>GdymY@$Sa@lmdBq1?wWXFF{aS|@CJu^Z7ac&F9YUVOC
z58l%2aQ!0i114{p5H%(=CAQybe!SuhmOD!|Y0M2NurSH4>M77n$-TU1kl1I49Z7N8
zoA)YD@ngFGv)ttd`E25UgE!|9_<A!@DskGcYOc~<x$<d<!rPqjN>LlsUWgS1<&Bg3
zcFlt=>h35fIQoDu+Oi<Y#*7@QY=7SPH=L{a8^bPS3-^7b+zH(UzAg0fXK52(DF{8V
z4yyJo*{hE?&nvTp2;pl=L^sqj?*<pTUks!w5SG;OX&>?wvSs?Z6YS<}O(gwS0ZWFA
zZnzZhnYe_c+*=ZY5!fjq!m}sg570Fcx!bkigWrr7-7UzSgf&<Q%rc(dTsWM74faFh
zR%C^a)Y%Brbv3CK7760flO%%$Qk8<)q;?v#h%r$R8U|cu*}5b;7K(y`Yh5;igt|C^
zNSqr+&Im4IoCo#PY%)OY!1KQxZD&yo!f)G}XA=ILmX8yUwR(v9eO;HudjpdX%){g=
zR@B5GW_g}#nu>!hHGoqcWZZc$;$9ZcPU?j`=d(LPGhdN$x;M-cK{Aic`5O30rbwG(
z23PgBus!vFLH9_}`;mTiQC8YdUfkWCOFgznt2khpPXZ)xOdic}k_5$Yf6d>125SO%
z(V?!|wm)l2UahEM*?d`df?#3a+lmKUyil0anS)S>XGLmXe!5&0`TFzh8q^PM-6O)J
z5n;;074yc~NKl>{95vu8;xLiF!}8$={X!46(Bxx#hcKb<Nb54Z{UwS~gC173R~7X<
zfc;Fz;+(dpG_UP>fF^BQ`t3{BqUID66kV5z;3yWHuty`9aY7>?nYdUo$^FJ*G#CHn
z+l<$$kcNy1E%_(F?6oya)*$K}^bCpWTcLXMUg|K%q#SlMY&%_Qt)BZ7bw;Y_5Vfkq
z`>XlSjQRBHyrzEwJXW(zZ*X8Ph9jnF8=h}qzC{&9u_}Z$2vnlY!;M7ST-_A%oHbQP
zH?sCj{#_VP)Hct99=zE4p}e%DzCR>!n#zL9*`><^Hs+q?t@4|@GhaLy`9H4iIxMO#
zY#6;D$RJXJgfvJ<gVGFLf^>I@NSAbXcMF1wbhk7TLr6#sNOui6FvP&X**x!guk(H9
zKj6AxHhZtNSKasTHd~?)f#`a6Ax_V!US;zqi_1pG6@giPfo;f+qphh5=hPiHF0aCH
zde}q6-jm}?nzS1Srb7TiP{dhUBhsq!Htt0jUW$4)L-p07`z+4za&`XxAA@h1=6$mX
z+i|C*&&b{{?D8EoP_#t5l6XUkuRj}rf83qg#5btySt=ZFJyE2HC68$wt!{kkEcK|n
z|B5=1?+iD5mMlh?fgghMHKIoOvoAnykmz+u>?B?NJC1F}&PufxbTk9Ur=@ol5lQHs
zz*o^}7Nb;X6V|w~J4k$91K%$>z*WoqO<st@ahE}2WwRylSa+mU;WpGfqc=?!B4qS8
zJ)Yt-`RHJze{-H(<CEx_ya1)4;34*OkF!3tSt`~_jV=Z=6J?Dv=y4H{N<Nhp-ozLd
zX)!n8K2j}_L3+8Ri$*L;lIh(vP^Cbw5^0sgvU!NEucz^tDFb`Z(R=2;@^?k47VpBI
z?{Qt&9H`%pGHibKm+;jWsNjOy31+p8+^p+s7PoXrzN8Br{E8xk3@*G>eo}`$3#)ey
zO}u`!uaakCOZjd#A_TWA>o{HWI_?jLHUy7Kzcmhr-jp{+tW0^`*u~7O*VcZKz@3{k
zMWnz<-W!EGz@xqiKD3Xg>at|tRh{CO_O(#WG3aIBa?KE)AIxuZ%rVt#?*F2^!&jWd
zJhN~+Vfe<@ZD1DRE8L)r?me+qp?aQf^E0=lce`GQahN0@UPSRk-}?4t{$_5}hGt-G
z0K$^4CZce(?o{M6KO)<+x2bs=c6RYq^c$YLa86h)9hUw@@Bdl=F{i6<qda&cF|SKk
zc75rZ=^&O<lJig-cgt@)c|<=b*Do`?p=9210{ro~K2=Syt;KiWt^Ih8$$KX^`G4R`
z4!>OZ;Buy^r+0aJ9-GI`Kb{zk61y|-UzzS(R+QwXqbJ(dzS8+J?$ad@KgW;@>wc2D
z^Thi7^Vsx()wkYsuxCs-v~d`zK})T)rzA?D=pWz$zc$H%oh4H-60ZA<wq8*cJHuOL
zA(!c%S0Cxfa4IRZi}L2`y=?5}SV^sZ3Q(9-yay>b$bB;!`)SQ1`H0ts2_q|OOG%H6
z=(GPai9A#;j3Op4Ku8G`+*tRz;O;G(W4>xYetdU(U!K;v6KW_os2v(XNI(gxJ*^?8
zqddMc!mA}cCM44*3=8~w<zFAzLwYUcyj%NKhs))4A_Skek_MmM`tVlsbH<lcmh&*x
zn8qep=mw1fvVfJYb-I=Cm?LuTETQ)OFS>7<(Dp8q-;~y8AfAuPFrJ+BV)x?^+{k;_
zvA>_*_#5lt)JL{p>KzBiH%?=zD*2aJZ7w!=O-B(S7*pN~CoTqmP&o%?p);c2lnaj4
za|dxZgt-Jz`lkp>?_2{8O~j49_v_*-WU$bO(pr8t)A5Gv_CqKaBLzn~R58E2`mXYh
zja-_W=PsE(3JczAfU!aG&%*Cgp8uv{t@;3o8;ciq^YJM%<urU#Z)~2_eL7H&cRnas
znr5V}0xxra3MH~I<>~h5T5D;<WlAB<W-GkZeHqzQIGjF7H}EBvbc&gt>NrJnq;Wl!
zN`OAgX)6|3l(XCt8i+vMsa&l6O)B#BZj!O0bo%B6=gxUDij4En3^z!yCGmF!Bp`cn
z_ZL;TG<F0dNFjA_h5q(=r%JNi3S?Nk+7FM*CWUz@G1mO~^_7;2rDWG%@gj#N*<0ER
z0$Ud;#Tym||CL+;wvuv}cVkbfekD5PA>IpFR3eO8p^|nX2(MB<<l7zK$Uwij;`Hpo
z>DO;E#}C<^?Pr@UDS|u^=5I|>SMh0zmyn%IG;XwD)qKNB(GTE^@djS@VioRLr`?JT
zxZ(g=5Z6B?%D}`>zAq;B0TQeX3H`>$HJZ<TVYvHpg<WK|&Fa_IVheb|Jhh(ip<?^C
zo)2R=PlJKr1ekhlI)Sf;{$Y@Tg6Gd~Vz-X*iRG2nRM<WqIr^pNIcyk`BEQ++gO)*l
z0~`bB^S|u#1}2IM6V@SZ?8669BCg_z&NRnc)(;{S6*{pilxq-d>E9SkFCG11;`h9H
zMq$bIlh;G*Yhkvv>2!p1<b)Ciwe}X~eAHMFBJcRZ6w9-z!N;(-<@Ojt8*yKS;s6An
zJhtNIA~a|!Hp*8gfueQ~5Y;ZGF8**1WdXgCCw#nSg8}$a!U_bY-3y#epZabptQhF0
zSCBUSQO>Lu)v=>k^JEy_{uH@UlMp(e9%m*n=Yl&xd9~!0mPX{|Hd;|*WRa*H-Z-p&
zqu+L14<E4x0J9n&3<Avjl8omSpB=bq7j%g9lxRVtTnUC8BrF06FyW5_#*nw=-};wF
zSBQeCu+c~qxdQ+Fq4<4B_K}9*m~(KoY}KjEsn#XxbbC--!_Io;6+o&iO2dyG;vbPd
zTLg`6&;(Qtf9YE`pliXMG;dnEp;@H~yLO47q6y9e3OCSt19hd&XuF-TC6m)%yn)Aq
zh!sTsBy5MR5d9?ik&t3-?Ni(BnI9Vg$?Rq)39P^EwEB|)#XB<+tjlm;$BH^843PJc
z4Whuz;3vkmdQ(k%nW53z^2!Wtu&g^_v*HT$o<!KMeh>el44~NpebAzOkI>No)}~^p
z(4so48()6cR6?EX{JK7tU}FBqaN<+=8{laX)GWilI%>wZJqk9Y;H0@amI+W5fvb9~
zo~6qEZP*AAdF<g&l{yET7(8BhiB={2E89%HElni2k=+geI-<9q6v=w8t8!8LM`*rl
zek63CZ$2OwWJXDvL8(Ut8}pT(Pl#89I#Od$hQ{HdjT!F%5xT0a_3I~z18B{oSYvJ*
z`Ih+sZs%?W5@v1T@yA(ILk9TSdrS_bR3pMw{J5zzcaUhFe1o)Tv6}XcZwrLNn-s;c
zGrehbR?70mC#Sf4O6tY8Tm#1%rk%GBY@%JF1XaqYQ>x_^r?Oj`2V*ck<Je8|?UY|Q
z!lq8+<;~?m{?4V0nYHvP(M&$+B^hKDwZG-WS+eT2!ieK{u8cG2CiaKdXf<~Hb=co<
z<IQG+K!<2M`kI#*S9}eeWAg_l%E;J`wpc88+7ELwAg!J3!lTa_KGIv$8ncEaiXF=3
z)6tdF3?!Nxn~6C@!NQVh`@{{?8)=1Kt}^zs>a~9^a~rX~Q!5J{)2OUfY#jWT`(i|+
zS_B=b*72@!dLdh>W2SU^Ay<*{bJHI&O{X%p8A8X>0c|D5npvux_~CEDCQiR}na>K5
z(ARB8tuQ$&V^&FnGPk`Z{<t$nJgt23Lh5u5L{XcD6Tw&gs{QKf>4lSFUD%h;bKoO&
z6Zkr5W-JgrG2#g~2RgL9DM-sMpMQLLPsP*rOdVlku)ihjMq0<pU1$)J=>xOy?jcHG
zk*7QjYgrkM0{W0M8v0C3hgMfxp<_T}WApOS!lpOOJz@`J#tm?OY+FB5Rag=50S&I~
zPlT{1R#VaY{jUm_Ym9dLfmZny4X_UR&Dk_Z%G2db-=F*_te)k{v`wEN{w2$>6{;K=
z#2@&_aCvgD!zNrD2+)ZQoT0a!-?-%<`~;#6Gd=%7n(mWhz~F-?cD3p{AhKNhkP~lz
z2FW^kH}^bP;?b$D(>@#G)92E$W2L1RPB{uQxH8v(ScZto$babn(F9l5TmkB0Uh_5%
zxFyWu!$5g=5#7BG&v`5{wX$&a7630aO%OxwY5+MnJ^~Dk4`m902fzP1NF$M*G>32s
z-L=6$ia!Kcg8&ybH$k8PoVeRfRHy>C@)f!py;`D6+LWqz$DwPU+nCT6^~`=5_0TYG
zu+ZrOd4WHWaL<L<4O|d4qZ*l?a=g-j`in5==<mpwy`fV}oXGT%Y0~n<EnJ=!(w%U4
zYbag82u;kz+799Pp9^{T2)hITm9)dNVafu&0WR+Ugem`z?<&O6-yKwnL$d#GMi}^K
zH0jm2p0Of$e*!Co*NDmH7VeT}4*?~KJMAfOpHY(4&;Qp>{7MEqDvx!@cp3Bm?{i?`
z5*;pW?Y~Sa&H(eRDc27^6mEIr9b)+#5%JgcE6k|)cE%Jn<Kg{#N)$0D_`es~E%(4?
zzAP9?-g)4lNIMlDbc{-ueTtmig$RI+bc+>sfVFm?lg_-WHw?I(vrLS~v><Oe6_6$a
zpH<Wkd%_M=ck+B4i7fDiHj`YVm5eONUt3N8f7YV{20tDe@5qCD1RwM4tX~?tZmjmN
z95TuMyK=iNz!i8q5V+_#>Ji#x{ouRK_mc~>3@I0pz}wHp?0zo)F}bMoAi6S-q~hmW
zZlD?RE1tmt9|2_*NJxF!0Sa<7R;UPV2KA(#=5`g!>T=1`3GRCNNi%SUYvvE8@|l63
z<Nj0Ywf~G!J+z10|ES@VjbyTZT&%4fC{^wtaOMYKe_Z%Je?wyo(FF_z?{Cj?YPOrC
zo2#FfQ9l%`cmNU?Dp;KYZx;TT-u9cOJ&1x^gA9r6ku(8U9<asx2g=V-AJ8!IRZd!G
zJC3v;G6>VkjkZ`*SN?AtcHuF4#s1y4(_N7$fY-4-;B|ijZVnB(AVLjBhqk)qE0=5!
z{M)?p9OtliFU9g-&JpiE25R*ONRj{fNttfPzRt`Qpx{hyDx5wzls^kQw0|-NTmjza
zG#uyOaTG&>-g^8N7$6*aY3`i80fblG2_rzQWs3uUi4~0k;rN)Ds4e&mFhf`#?teMv
znu+_yh2oZH;H4+U=`Q_u2k$q(3dS7YFjboZ9IqWoN{;r?&)5?Dlz&#7&i9oo$Vxa+
zRi^It!s*_F;rd*9JP){Fl0LVH|DL&r55|6#VT3mA$%AOvdzOEHYrCMsbkFIU9QzDV
zoB~^7w3?U718<rGMU1w2Gqa~<JMvD#nC<J)XZroU_BHc`Uym;xc{Hy96}uMs*~T@<
z*_;|VDenNX&*chG?oc{M4~3Brn+ummcmB5-KXfvCV)PGlfZ1r=``{}wRRgb6v|b#O
zlaMtLz@rSIPjcwI>gU3Xr1sfWi|$+ln4MEK)H3daxfrSF#}Ng?KNUxms^ZQRFRNi!
ze>X66cwxOv!#W|T<?3A0H>M_#v*YhD<Tdlf*$XJRE@*TI&!kBZ-Y;3R7QesNIo9fN
zE6E6X+>$5Set&m^E710DG-5z>o2T+UBFp*4AMlGP{O%u6Jnfehh~9^PLFdU<`+*hL
z!hVywceRz+49&(#kwAyd9N>DsZ2E4B`b~5DTTS4KS^uHHJJ~+u+Bx3N$nvT<5LrBQ
ze9ibDCj9k9a9zEQ9|z1hv5Y@>=b_MzVOSGpt$Y;RiKFMT>mDVy-EYy#0mM^#11zYx
z(Vn92OF(lSM|*ei&^|cITB>Z?4;`HB&~?DVqSIej-A9$y?f%`tOy;OWA1fvZ(Mi+S
zXkgUgR}!~;z3K4rG23}!CQ{PT=v6;*|Dp83UsZOj9H5ANkQjTjt@Bz~Vlw!~%iXnn
zt?yBnPMgpM(oR5M#e48YBLN`IRT@0X%Q3vz_4|C~?r`O0i$~7F;ztQN;H<<gA9Y}j
zPug|WAGzP<RL-BuyuJ%Cq&FD(4>Y1O^!TibB)s<>^VhdpvHc{+57iJ(W&=^Pb{Y_|
z{HrQl5||}o^Y2Mgr0CD|4TITMs53x04%p?uHAT9P+^cg4{46b03!Ci!O(ItT=4o_|
z3b0#?Dy&I_|GSrsQh=e$Csju+L{@g~Yp5mw*mXom021{pVY;%@$}!g<-CS;Bx&ZM!
z&~gzz#1=KF&E!(xF<z~%uwF!UAZl$hRaJkMo&gP11W-k{-)KJ0d4T+Y0fr_IkUMY@
zUXx91GTZH?gkCE*%L3B{H;>uU0fDg{q!nl_jc{xFNa1(3opXAJk0z5nI<TLAbRV#&
z(p4){&Z+Wcq;*`(7k|cc*$5sgTxZ5WLEmq3_GQ*yTiVsYU5WRg_J;Z~@5e8)D>_^p
zr=iy;=J>vI&U;FxtqvEQg#O!>?#x`KyHcbt%gVD$EuREO%V?>dQUe+LoUNsCgH{{^
zuJ3Pg0PMCm|86heQI=$keL5Gv9DnCKlFw+`7067xJEg^2ca;aR)9Q!!xC2432{Wh(
z=h?R9#;+;YyYk|#3|b<v{|3J9n1=yaw#EUaftcc<bz>K!cy~aUb?}LhCGYOM*vYu&
z#oanf1eFouAGB}zX?e}4Nqs164s>wjlIZ*PyCoQBW3GDXUqPankNR&n@Y#3+mqhnO
z;r^;Gz=k=_CeSi+0%|2;P4I7cce^?YhvO0v*Of}&UAG1BsB6yQY)0nF3Ta=7A==u8
z6tpdRd%C9f<%)J4rgC@f$eC(KUd4_$`;tsw+|3N{ka!(=`yl}=AjA6P7;u0MJ3PjO
zcAh2rZ8Ajqn9p^0{4v0QmJIfQFBj`8Ylmx7f<9N=%}4lOE;Un$i@_@Dn-n8N(t!#n
zRWs_~(V?g_{b0P+pAlDgRG<G!a^&kDvtN=MCHu@AoA+9@96Ld-*!ie#BK9o$sKKC%
zMbHue!!GZWU4N|b%=jN2JpExjR#eM@k2L`iTrM@cgiL@}KgqqY${cd^_2_dkQ<Ui5
zJJpw0F8bAGnOD%}PWL}-(H#gW@4e6Q_XA=VwbJgnV8JTs)1f2~I~H1BH_+uW&5t{{
zYVzJ$I0@|@F1p+wjNjSMS!Eih%q_?*J*w^27@U|s{5pqY*Gl_MubX3*Yw=-Jen#4{
zD@)<c*95Wb(BaA?yV`gA;K{K?^NP|t(~cx;Trs4s-d<LCzMC2=Z0+Hbw>|OY_SLzL
z$mDa70&HDLD1svPJLo=f_~(5q!uTWrMzp&!KPGc+nEfb(*X6DZ&-`!5&`FS3jxp3d
zY6NNPvCine^IqJb)-CHdDNS%+sQmRXxo}ZBdXR=SzJi)_a>^tcK4oZ^u*|@VmmV2<
z?(A0@`ZgO5Ek4nvdZD^Z!J2+nn}{+j{h)Ha6L`0yfjPXs&2t^A>&dXtHH3qRZ79@7
zMWnSd=$E-I`4r5)fN_|RzHU319AGaLCgxQ2TPIRtWLIpQvun2Z?lMSgxGXPN)v4N7
z{UF&wQdD<kdXl1J9A~%O*5IloS?&*%yI4|s5kt~$)aG>!2)QO@NqS`k8xgf2_ft;y
z%a+^oVsT@Fz<HX<&uxZFr^Y9GFK1}pUr%OsPM;)qQY`PE#$ujsqk1_y=Vg5mQTC+)
zAGiJ8^-q6R$1fwH%}tjLsT6St%=@y9wFv81WqFQm<#;tp;3D-ZwVUa5m~scVKD%Gr
zMSp#>{p%T3mD0BFTRVs+yByYEXXcgfOWP(Z6CpS7zg3R$p>FPj#{0>erT#L3&s9w)
zyAbZj&_C)1rtVWl+QLPh6xVzC?#vzL933myUsr5UmWzSND*&)9jtwYRajdVBv4^Q*
z(fB!+D(cH{IJS8Z<%yN8kT<*Ubk4}<$A5b^Q>;=da(hV9+*)q5AI??D&dp7Y@SKxL
zQf-WK2!s|W7p4gF>#w`8$&2RS>UH@YmYcHe1l%ubaJa>|XhbhTJz>emP%#uG;fA*F
zAL;z6rxO5ku;ksT+y6r=AIw2z`9f^6Dn_G?u9iA8Tj-;i+Kc*&@mmI5hXaz|)Z?A(
z$WpN%#_<CWPq4D1Jlqxb#$Tk|5*(scHvToDIp=I~_Y);MNkjErfidi~lz%0g2E~c*
zk1<Br@Le?RT^H0k&8`;iJMNGQDeEM0&(OV3xDfEvg~6aYp6we{N{u6*P7#m2&mAcC
zdhQPlph`21m>KoM7PynMh!-dg!`6Qjhtb=j8_sW@xCNd|pXDLR%fX$<D)%2Wmg=)B
zSMjkkrzjNvPM?!l<%a(~JV#7NU`lc6z0HO~oA&cAgpl3d11-qtGF?1`fn0I|W{Da5
z&Vj!vS@Xs8auyZ#A<C9W06z|-INK)Y4d;Wze@7MxwWK_+tln+b9y(0@0&C;zQ7(p*
zQ1C7B3z@tjRZ^hQ6gc}Oq_{-lt8190D-JBn(iDT<ijESVyuFJvzK5VX#|>X5?=*KJ
zDw*WTuwu9Lg^F9B+*wm3BGAK_1adv*vFwcR%eBfIH4-yaQZo6kNBzulyQgx^Mq#1+
z!Mcsawp;v7OYeOij97%E8|s(nq{o8Y2p_QB=m<4pRP|q?{6y_3QhBYlQlqlGD<`lY
z+qmm#hx9-4KZXO=cJUClV|gXU?2pr8@L#dObNN86k!vpBe}711)a9B7!~j-_b;jD%
z8vwuCSFdgFxJk;&W8n`|)vc2ODw71Sb-3do9~R5_W9pM2XtS#5X&-;&r4>B`Fo)pc
zbT4Q}Rdg(o0QvP|^V7%Mkq=fa7h=%ax5{{F&S%_r+3mSFGM(@DmGEj%EVT2acPLka
zWHR!`eL&O=sHO8(e^R-27z4}d`FQ)vxc0V)e=nuN6GL`|+13xr41xON73%dN7mP&L
z4IWw+%|q!>605(7)`gOeYFMt1E8<Bvh8Oubvc~Uj$M4(yPC{wex7&Zc@T8nwMDukp
z99rRjf*61H-wgIZ3p6du)ccaQ+oD)J1_r5kc=L|BI3KEw{us749*hL0{o009z9mxL
z3yhy(<qS32pX@qi+Uo%+si{C8<)Bua=VQCzelT^^RZktGhB>}1)65!MNvS=UOIg1)
zmsP0yB8I`|<{)t$D%-dui16*Y^iA~{{KFH+`iR3<zV7LdO0oUd@vMFC9Vy-7YFDD&
z+X3d>&(dw*2s7y8#5xM39ZXjCsO6gx7!P&H+R(6IyJ4I^=~|v)N$$VMn#3p0%hoa!
z-_=&8*YxMj)sNB;Cy^iC8?8zs_bQB-I)}c6!0Eytsui|10WCT&4eibqy~puR<n?EV
zCGRPFC-B*Vg|&Wr|0P98H*X1Tt6}#I#oHyTYDYw{I`h%+<%!_wgtuX5Xg{JVqXm=E
zgq)Xm|Kl!=QK(lDUXNm!pC~k3hG*a`&nzbd>oQtsBwo>j_jH1tQ*_gYaQ)Z`dQv3v
zw4)sHD*LNLoJArwOO&-C%*SqCqn8O+PNnW^H0nnFX)hwq*LGOmo3U})WwYw3PDZ<4
zwg1L-<Q4^LF60BHf~+~UTquWcAf!rNUfWZYYc1d217#E*5Ag(!wF^b-sBRH;J#Cj}
zWPhBu=0BsF)5HV4`CR^Y=~i4jIN-S_G?SGbpXPc{-ay3pDJREWfuNCq=ebSJLc_cB
zLWRrQv%r7I)PT)xiL5s5i8jdBhH0LVI&Jt5D(l&MK}JuabrdiW{z$k17OY7M&gU^m
zFa8+3mMuoL@$a&%n{zjCTip`AEzsp1yK`Zy(=;E3P&8$>l(>Hr3AZdeNdJ`8w{|n2
z7YDr8PD&HC>Kx0P((k77E*ie4JmF|<c;J=aZ*PhPt~A#3kpIrTdYH~#ySyB5r|?a*
z((a6SShv~SeP`X{%X&OBV_<fN=xm8;ln!^*muzW_y(_;Ls)87VYri{RvSR!-GPwV7
z<fwV2r^WDVnX7<SPcZ_2sP&o`DU+DZ7th6NDZ3J94ZI+vS7vCamVAMJA)1}v@H@5a
z#kiN>9{#q$TwExh4$aY7^le3fKaIs#B1@CN@-TA|lcH)JtB&;4z&Dvh=O~`Z*m0XJ
zvP`zSD#Z(DV69kOxHN7JpEAt_c>!g_7jAP9Btyt}%a}+ik2#F4!Ij$&oUvx@@br(P
z*ngNv$|^MD#r@wdhift>7PV$3CAh!h<d5?7`%-Y76Hnk4))HMFwq)1p$j4VTM7{j4
z5jtV8^{e=5Gv?Pbhsz}oCwZEFO&X|49?tv-CZ>POO`TpZrFZ-{@Hv~S3eET>7}jc|
z!s~e{nY}t{J3=|)S>5^lSpFrX#-&;7Z*_xtyS2kqW+uf5sbkRBZPW2zZ;rDID%0)A
zD>#^QouzWx8n#?RU2{}D7aV4x38<{t2G)rgq_Not6WGp5l)$0pjj$}yu|e0B(u)dB
zM;d;C-w-2U$A=kz`yjxPUOYLzA~=K}N2x=h`F`4o3dOIQj5e6X&Nyn09BeRar_y#p
zWAY@!rz8C`KB#bue*yb-{DIZgigCd(P=F>SpUVI655uvIe=<8=yr8*-eNue~tnGY6
za)WzZebH+`v{JQKz!-$ddYuhycZm?8dlP%c8?-TBuxpX*bX;~yxa^Yx)$SajC%HSK
zp1)-Y!dh~c&x;(A8E%OZ*r!1xIE9xWjPK8cm(2(Rrwge0^tG<7>OdNAITk})XO<w_
zz9F=d-<Zto|J7NQh%}~p`X7_d0LOW18lTd0=!s)%dFjN<`d1fOo<p!lNL<lzF_WCb
z1%I>)Lau<)2)_z<T75rACepZ;0kZp!vpOewgY}G00cyjyhMD2I1BN|ft3Mm2sc5{^
zLPa@o5}ljE+<(ahY}K~nh<Kv<aPNJVnOUd+Qaqkm5GYfV^H=iI2HQuzV0<zRI`nL{
z>UNOI-u~l#az!f2M)3Nx>!LizmrPl{i57_49r3P4Ou7_q6AnqHUz+f9NvX2W=k>d(
z%gjl1%>>{<3K^TFA%{I56AscZb>*GIAx3cDTZXKC!{P6oVjJ12ic%~ZlYfh`vl5I3
zb1Vu+u^S{uXhR8JroDYNsCPQa1TJN!(bG^vFoY$oB&mcM94boas&p|#gtV`gv>p))
zSoFR@6zJJvy?XvW-q3&H%&UtuHyKBH@AidyqzpXQIgvf;(oci^dh)5pxYIlvu~e8<
zW4Enh_1!nWVHlMQ_Gy~Y*&hwGPRoNh??ai_-{LN;&T<-(ISG?kCpZ;+#z!Q#t7zXA
zh!U5$_%~vPG%!<a+KeUP@uI7jsOs2dK#vim8(HXBv9#dEg6raHB1VZE^(xA<6qC5u
z*R;3*5ANG6gj3rS?bG-qclG!<z(oDHIuA9YlW6NdL9Ff|Q7Rd9*f@*%q1>AuWHHu-
zXe0B4;KOxje#B|SU?>*zlyHnX<<UVA0FpF}ql;2r45+#o@hijI;Q2TvOFKewt}D2&
z6nGUGILB|=KvyR>4^j`4qzXtiGuBwi%4EXVIgzE23L$%p%oM^kj|M%h8x^(t8n@cU
zCzl@-E=K;U<UFLkdHc!ovIDAw!L7sYE=eGTF^a$IW=FtE@~NDfSQkuSen!-?{R`KI
z|5+^mU)>*g$wmCS`B4;IO42u%AL!^!x=J&~lTj)8+b_>*K94qUD_dH<bcs-S)=aam
z9~r0Y#@Rnl#(!B>exAm$UY_p*4}al)ve?=Qt7>~YNxN+_zGhZXfse%h5&H~{)cZG)
z&GX}wjzurlDj_;}44l@^=U?~n_7B_a*$>5YkE9-bVIE=u8J<PJza;*wyhHxI*%vjg
ztTPbXPPVOGnwMCf7^pD3U3(~Gip#~D(SQKwFxI;KFhMqa8-J)`V5}Bk#Z7~759vZU
zl?G`sVtMHbjUKaM&nL%~s%(7YSpDdiECA*0UW`b9=hQ!5@P?l_HsU!Pe>oc>m<qWY
zv36Ot<GlM<y}e>{I?$zWCnd%!PT$_F<t>7g|7q7bF@DaxlKwhR7B%t*nb$&`@!|D&
zg$+bD&dth0U;bX~76@Isvpvv@Z0Ne46gKai(k+PNiHubFc5EmR^62d)dG9vY7C+ub
zHgDFON*luI--3*8rEK2nE7zxqoF|A~W7ND{IPZldRpp;a?W5)H?}<Ai2TKR+oahr?
zWB;=kF&6f)H9tL1Qd%=tUqbr`ghxU$%v>1By_+2frjz>>OW|HF<WnCKYeB~j@X7Fx
z0^I0p!|R;jSS!%@=^}G5VX$-(c<*N_QeK(oSvducwJJZNk9zIPBLy>}B)7qa5}a5z
zsVPyQ1j-uC7YT?@Jfkv32&N=u%kQT$L9Y*z*qoDBvgR~N&~Yuwbbo^Uz<clhK3kWb
z55sHzq<#+Rdxe=J#VPKflPKY3RV9(F5!nr?GEb^aV-AV69$rk<c4HJ{O!0}7;vOBK
zmW&LN{=0vZ@Y5$&<0tQT0rV|icF}a*z(%&XpU+wLJUR>VkiKAgo_|}~)=~vKVKO2&
zbTxF0i0nMk;8@ZA)6%cK@WdaMSb|70jg;qmREIQZ6CW!MPT5@JPNJ@668mDmuCts0
zw0f@G(>llSM4lKme5L0x(ks@hkFkzds#A_Tg7!#^{O9ZY{SN3nLI<T95Pf;V-5C0>
z<7qEIecS@#=fm7MTu<@XS#A~Hp$p3fy@=tBd*Mq$GpN5u6!7}-8O9t!v8Fca*L1@0
zz`lg9!mN0h(ocumWuUZ;Z5P9Vj|Me1!BaJH>XFFp=sd5&N55D6<alNT)~Y)t@7k7X
zNH>>M;@+zl5tXIOCpNqG`R2hWZdozEyX}`2t3;`d<5%QR1jO#78D!w()snp#lcFJ%
zUn#Djbvf?>rIpd@b=WyZ@62n81z3M2&v|oN^;ut@Otl7<KB_u>O6Jyb8Sj3Y`~<16
zx|)rUK)sGZCe$`jmQz{Yqr?6fQ@ksPXV6O>VI&sEP!ZK6=_BOj-={dP*d%h1;*qr|
zx5V&&9w8a@Z+a5&tfeZ`olo#5D(v~4?fbHBe2q51Fia_?Wog=Mit|8sxhEGC-t*#c
zZPPOxsS#1avu_Q_&fsY4j?2X?f&99nXwkx837wB_sbchjh{$FyWX4JcMUvEVWX`of
zh!7vheM?vM49!CKWYLe|{toOOq?%?-)vV>p5K=pL`EHJ4A>`sOo;vmBXPPAAP4W+&
z7jfT~A+npLi`z0*bol4+Z1@-Xir?G3ePN@@4)~@vyQa(0-?$>#BSyIK=xHTKkC&&q
zAeq)IkNrlvhDR?QY0l}~^v`+h7=G{~(SQ8P7Z7bW*;9=Y-%P!>)Ts$zbz8L@{dnCB
zmCVU%8&U(e)azU;X#N{z-r#f8q27-8HA=cSW0-PrtdG?I&OJA@J9G_ctj<Kl8v@$^
zIhmmnnVo!~wL*t;E~}XKA#teC@l(h+s%&u$Y^Taq4ZGQnOS0?FmODs}G_qY6^0yE@
zu4ig|y&=@{XIpc#I<f24ipyzSbT-ieP7g0|`aS%H7oCbe#4n+YBqXo29(Nf8?W4_k
zIH1CGO&5H~h-TRox=s=p!Fo892TG7-KkQh>8|0iU<0M2?n<T&brVa=FB%zAsw_3Dt
z`Ykhj^;JV^FHs_>D9QID=|qAP2F?+?MPj-HCG+?5=k`c7RNGq^tBuIZNt`lp?amTy
z@w$JXbP@)rL2Y&Y+9H3M;V%*NwXq~LN>%OqGmURIW-diRtUFnAH{-334?khAmEZ@&
zF8Ru5B!$v_8=O)Xl%kOO&FMjNpaK~y5!F8!DQ8zV>V2LN#6yxx@HWgfWaZb0KlwIF
zEEdS!@qcexpwM?@_ft7p!mF$xkUQq$uV=4DP)OgTCQSk@wNjy%ZpZnC7_AOybzN|N
zf^C(!xn|eQrlk5Jm|%fMcXEKm21B5WK|pT^VFmBJd-0TjcXzsq%cX*H*i&x@;f+j~
zi={GsYNzT!AK$aDlAHJQOXM8GTm-L3{G7B--x=-24?m>h2$ctH#?HvopZHYTW+r3f
z)RB@7<Hsn${KRtdxJACgNeyU<w8wyFU6a?c{oM)^x<zQX4vWy>Zxh0T4}E%)Vdsj-
zIajWPm$Ct;He*o%AJ{K5+}K@1zs-t=#+`7=idXQm>{Yz*5j7_tIG@JSv6LEKRBSoT
z?IAhCDa6FKv)>!oAn41yq_m88X(6sH?xIkQnzG43XJ^$5RG2+IE~$30G(K$$<X8Do
zu~El?Xl{8<{-<^LSG*|lawK?L@Kv}Au5OOt@S{ZPox2Y2F3nuRRd!LxX7SteP_ms0
z?17C($)R!K;qFoCO${s!8v#q1*g*W(_d8x{5eQ`ntFxSFIa|n>f#n#eu3D9pj$-e^
z4ws&VU*Ko7T29200fDci2F%nyDZleopu@7^qtY&hzJywX%mK7_Bxl(|rOZQdg@WH7
z)z6IpES_|G>~53?9hd&u&w=N*l!J7qd3^NjD{0y}vJ12;Me%f7Pc<gybh<#!8F_D$
zs8JEC0HWa<&M5BPCrF0nV*f<R6tzlpsC@Sn_Rim(8rJTDgr3tjZJIkCSWUa!rr*Wq
zMl8KxyLO;9g$4Q5C4V^Jl#k%m#VEDH<vSXXyq~O^P;IEk?c2iG5Q}2nqXlMl<5E@F
zuq81E)Sw^XHAQA^OUd}66Cq%T)?{Q!w~I4fjDYzSHvR!MP>8CP#VUOllVt9ogGs^X
zvNF!1nge8+&(%Bz_Q1_WUj*orzS<l46sdGjOXEc;dr=hgiy&(#EfU+6n=%s-MKoRh
zU#`LboykzWc&NizV&x0O4dF2cP_oF<y)yJ-`SX25Lr0H5eb@I#^pBV@f_&~?Bz=!Y
zAGOj;5*uPH^rZ%~--T)hMzB^eHn>(`R2TC*3%f}Xm2A$dx5$u1Eha>*TWP1{Q;jBU
zF-zV!Ga`)XtT_tCmAen%DZ^21c?b@>_q340YmwL}tUmKw2CuO65Lf!$^})r4TjUBs
zWtxF|!-g0h2Cky`_b*o6w!Q2Y_}C;mZeh7DU_>!3(=<^nF-FzhZICIUCuq?|O!Iup
zD}U?R_h0w{41P*Ed^W1+xoR9G7CT?M1(YSNf0N4-5mpkv*(8RA?TSx)JT0w`?5FO>
zTO(epP4qtaX`DQ~FUUbVN-z>Ox0|!BuJ?mR__>9(c#jW}Z^A-9q4uV+UZ3op7$KT7
z$esWk9IG0G@m};YkwA>e^5z_QToD@5z+D~p4i`OuGBdfUNjl{zB$!{W2NClqg=<i;
z8B1f1em8Pr3?yyoW5A77NB`G^dk>pU=ZATJkrP%h6P2RRs*R{^my20HX`kX%^1;UW
z-5Ei)q)|?FFKe8|baU{rc`_Qaj0=mu5}_}sJS#KJKi4skc-SkN$Lc7?<<l76A6O^n
zS%CFS15p00L5NeK?X)c*(IwW1Z!Bp)s{8KK!BvO%J-^3ete90YEJZ5<D}|)2DSCf@
zQnA5QGMvG0HAgXC8>(}3we3@vg;Ca3QP*meZuqj~XfMKGUhi3bj~Vb=$j(@VXpuCr
zb)8P~d1AUs2es$elmz7d-$o4U!I3sjHbsuLV?O+jSQ(1RRl=?l1d{n69NV)PWr!Gy
zX3`x*Dam>NrH&*w)ed<M2p&!aH#Qq}3OH@M9MO)&l&QJKkd$JIHap@?RFJ-h&Gr>W
zfgQ*xzJoGt(DK07QIF1>%?{%<pCk8OL6w6_#yz}56ZS{?uy0*iO6DK@O>Ns#g}O|4
zd~#$z6mw{T#$CvVMvx@`99{|uB+vxrT&|bAzGh>W*=SL9dwfYce4O7od$}NXWU+#<
zLh(-4B=l2$aEwbOfP8Ln%Fk%zgfhW05z0ZMK)(-=VoJLNLi8>pv`bq?o{Th@QN{D_
z*Mp;<y|Kv=+2iu(7SBFI7qFAX*hriyb#a$UNBKx)kI>`{?klV8_5!RIqFDZQsB(-=
zaTc=S%=f?e6_J+@(Gu#h&$W6j5-olQE;e*?X@I;u+VvviAFmA_iL#|G_dqNTP#}`Y
ze)bF7=cDjM3+W)egh#JXz?~s^JJ}fYTo4=g3}aK8()zS2(7A?9q2?<I;AyH?C4n$>
zZD=%<ZwUGPYZ%Bx$|_7Ld%qwOA5Euf_#2=+y%(0!x#+`a-{5%=VaJ9|=KZ7^$hkb$
z@1yP#H-BNC>7_#jW@refFA(RFJs`Xa>CTcnpiAiSNsyRLEeLA^J|EqL(qA@h9}CH7
zo;fK!Xk4a};EjPjL*xs`$MhkAE;IU_O_9py6_^qVGT6VW9dLa-B#5VwR@1%~YV+DY
zt|jjiMe*!?2XQU)k@{CzdRM#o+()V#_!C3}32X0$jN;~`B`%@K{w<IdLiBAW8FSUT
z-9Rw;8~oZG1&^fcsVMNP$QF}peNgWtWrUH$2Q@K=m=Rkn3QnHp5&DatN_)wi8a<Q(
zY{n$tc`|%0#&fA;*2&rm@8teQ@TK|4A~!BasI~=umE?EE?7qp{A$Qw2hXgX%HN=iS
zB~ht&KC!p2Wd?6Pdagu{s!vd<FM^$!#LN-R&Xa_FZ_)p<UFH+%;bv@hm#=LC6|Tj*
z=*zoy>WBivBY1nzBxemf%2IKkjq4r!ejU<E_p3ehXO1=(oIq&^bOsDklloi*<F-n?
z@i-hw^-5dhnIh+Xc^jkbLOa+swUXxkmT1%{*$yhD1n$_mH#>B`?Xg-w$-WTf(HaT<
z=DuWEHyPD~S^^VCwZyr<T6Z=SP5xgt$>pU-QAoZY5q`b<@pdG3aib9dCNuJtZ2PwD
z>`-e4nnRm<O+Oj-OKK&|oWI+iecBDj8pmELMR>gA5KFcRg~zqfwhr3RuxErWAN_sA
zjoyE9x)Wp=kv`BD*bsbe{Cl3!UZ1fxQUa%3)nABnC#tUB+6f_-0KFspNa(xGk>NEz
znROW5;UzAHz=3tNA~=D`&LyW$pD94jQ@ZjLJo})T<Me+|27OH|#*aLc+!ZndTo^^w
z%wBgWgh{i$?iUU|Ar6^r@1y?C{(f?c6n(vY#+b`O^^REd4m|z(uGZBOjRf7#bPR~`
zO(-X1DO6&q8%zvWiV38wuP<n({9eg5dnH28bz4vP#WItB{a*(Y0UO1i`1FHTd4p4-
z*tt+lQ$2WSuj_sbn7~(|r6y@Aa;@Kq`a9vjdZSyQ_Yi7aix)`j%BPXFO+a3W`a3%H
znH;~F5S4yyT+WSOJAmY8-#mAFs)u2jq$==>?tH}ME@3oy=DDu|lRVvP`wE^oyJ11N
zysraLvW__VIuc#9NeoZ4$bSrImlRKvcnF|LQ<LQPO+evwt2cv_g$k{^SnRoa`t=6}
zD%k_>yZ0Zr$op)PexT00yTl(-ob&*|i;%cJVgyLU2_Vt&P+-Qe(9<wT%#^lwB(BBc
z(m8|-akd#rxIT&4Cg6>-=Ei5U5-iPG@htDEcQjPd!>%o6uZZ4^o@+c6asnJ1X0AGr
z#UsgY<G+I$>uxip4{Xi^d!jHJzc=b}%MnPzB}CoK=_)X5ivtG=DHM7Ii`w<{^|Wgh
zKZj*;HP#@t`OhAeSN(d#F{!4jgIFiyQW~P%W`fmSAxS-Pl`0L{k2f98%!G=m&u6=#
z!fU(*HfYx$dv&vKE#0Z0<5TOGUF4)jxk>8nu=51vs)sJI38@jDQjt_~0GimqjEdfG
zI|_>6sgTCEgFNmv`~x3*ZKJ^(d5e9nlT69D);Z*xXx^0>bM*j-m7F1Fi~IlEi8uqE
zM45A)s$evOJB9#BAUg#MISF%{DvADZNcgaYC|B)HhoFbKwOdOk6Et=1u@G~YNATr5
z4E>KV;%uBJ=kzqpI~pv1Lx~^N0FietC>7dHWC#i?nqi4dE>e}k|D84|Nz1&6KvREl
zr+tvCB62r7kiT;SELUV{(VtPLz4y0Rt^t~sN0w#eR|yr`XBLT*Q#AEoi%ozlOYj%s
zXTgW6gJhtu!){o{F97bn-u!jH>U2BQ>&#35zcx|<4@Vir!Ty%eFM&+ML+=A^4NDO5
zj$lUDON+#CP<W6#hU6W3Gdd|HoeX^Mo<jDZwuxKy4^ehwu4Q$76xg3P3M^l{qy4t=
z(16gFJ>JZI%4Yt_`<29yxdMq}Ad00N4x258`4aeoE_W?A1eH+npZJTz+`nE}phyQw
zi#w`~t*QDE34Edz{gpG|+DI*)>M;ubl({#}J>BB&&n82E*jGs_AV>tE{I6tVO|i+(
zwp`IEEjSxZR77G#n&za(k|{Q$MH@TEA8hB>6n3H+!8h8baZuN@^iIOC0l;$;9IH5h
z;Qw<psJWKse2!mA)Hef6-zPuru{nfIxen^RqJe>r>B~q(@@$R5`>BPR#9f6dJDvHC
z>2mdwIpp{e$MnDSjNqxwb1su&0e%9ck#^5v(-xUOvh6}YX_)d(H>C0<JKpbQ=<0+m
zpCxYP;}$)%vkEc?js6?SqbkN)Yh%&Q`fu^@>*?`PW4NhOEcmV_u}lYr#D72Jgp5Ue
zF=Fllx^xW#ky32!qx;SO`_+}hEjX@j^h)yJi;#xQ+>aoUb*XN~3IYs^Vz<J+Jj|(t
zCd@JMwyLm%{t)=3d@?j=P7@Cu<9Eene?i4<&^h`V(37tcAmJMQFI{UlSObm|OD-~W
zE&Wy%$P5*e5!B{U+-+GwF;`IqDY@HR17EC-s;+!VY9l|z6;wo}>kp7)&!v?MMih$M
z!wP<RWL?)yF<+Uv00m0My|y={60Aix$PM*npKyvD`q_!}<N<FjYtiR-uWvyZ{#@*!
zCQpk*-WY}8OdO}HXNK&bOywiC85?)#E=CN&wV&SGK8E6h==|05f_H$2!X6dUPEB;3
zwinNIvHDUcN&(GZa9^a7!~S4bmz^GX`A_mSe=^JfffDrhd=kh~j)32(@DHhwX1^FX
zE}CImXRUT=B}?WxTq+E(d_kq-0@ZXow?Lb)T?WObMp;)Dnbk4H@k#4bwNMYldv4QR
zR7k4xA4Zemz^G&<G;-MZ1AAPZ8KS}N4S0~Na7>~8l_;@HUj#h(3exO#SKBiqPl|&q
zYYl(V_UjMh#m~_n$^lVw82>6nfgN(WJfO!xB}q{5zmCvZ5F6t~n)K3RwhToSqf~OI
z>Y0nZ${4)1Im>>knD4HtBZPbMl5q6w$*%!Stl_S?__|FUiu!$94g(x=Ya_WMBYP0t
zPTZ{b<9|#bk&H!tuVT!@5qIT=%$7L(Ztsw5CI!}GL!9&94=t@G0f+l4bA~1yXhWNA
zC()(2iDj=A8ouRYL@&@_l6V%9_-?<2J$gS<&|&wj@zsSH={5H!v+FFn{|P$8{%=v%
zh@A!esW5DG?B6lrK@w<YwH6Z$2N%s#V@e0=FZG7bgYy}?lrlAzR@R{d1!fFWl;tIG
zPaK%+0Z&`=$we4Crre#&NbLGtsv7WjJnq+{>c0xN<ywX(I&s<5%7J#e0oq6oEr*2i
zyJYu8vzK!0j)2Iw3fqRl4KjacX)_QcmN3x&*1PMnFYm$)L!@V~IDeV_XM;Fe{CHcf
zD~Zcmq-d<MJ<8jV$k*jM7A`Uw`Hj{Ya#&70p63<YdNqBKW{+We<(LyWXGur6SzZtc
z{w~5*JMg@OrdORa<ZZBlf6-D*c{{Rdhus5vj*x|WP0g2w#y3HJ-*DObI(F_yi3-l7
zCuNbYa8ZBk3eYx?aF{%?!N9D&4om9YKSRAr%-R7BmTf`HM&0vfOkgNHq}fam%B$ny
zsCnix^5=_0HjvCL<Ck=BdPFzl%E~^XM|<J+sqYk9SN=r5i|iP1;$AByiB;rg(wBtH
zwJp0>H?Q)Rbmp?5jy^2_1-X`xX*nO0Pp3kJNJ|fxbg$(8?&xo5xW@Bz{s~!>-sXyc
z0UfzEx(~WiP|YAnpdO%^aoS3}_2U=C{kDV5oFcG*`LI~|sv9$VbFR?)ua9K1X<V*)
z;)!h0m_mV#q;XRDe#GkYxG5k}X`T(<K=_grC+<v^yEaPxE}0QXt38L+&@)k-XVd~)
zYLK7FpVMS%xJ`wgh;i(`sU?rhib5FW8&xKPv%00oDSaDtM&>DZ|3vyl<KmooH(1gz
zciP$2Q3_@D?;BvjT6ne;&fbbcVz;n*t-&z;$~(vqdtHLNsnM<(C-W3XQ-)l?38N)s
zaR>fHzWL_im2ro6DFTh84Iah6EDs28q#?KZ!uH(#Su7u_xT}Aja!fYv61B1YVLk$f
zUNjE9VGVnR#w9HMx$iAG1VVfdl{``M1ma@d8@9{xZ=pSj?$coL|Kh~wo`GJ~Hp^AM
z#jllD*ti+`Vd;Vs$fr<H)3D}yG<{Tuh4W$nwM<xS`T_eniQ_m6lO(ZRI(tz^HNjmE
zWWWgfFu__a3B*X4L4T1Z6!o@`l+iseXl|ohXrKD{e1HZ-*>NnnsjTOy-1cHt<<&YG
z-rc`+s!S}MD6D{6kb=ul_|>2w)MNde?xa8=A_Xc=6Nta!pRkZn`A7&OA!m|h%UqTT
zw=ElksW0rpL`A|Z;zvOMIQrYjnx;G$TX^2VNHvV0{M}Z}WhT%M_aPz2wp`=RX^Sqe
zq}aQ_=pisUf!i*MogxI**55#cs#u6ra$e+|bOP43T$Bf;U`PN7iohf)3&j4j=jrsD
z;cLN~VMcp9oHitiN7;*Eo~1mm?ejO%R^WYnPPLhMiIUc!%^Db{DAY8SI0|1E<b+)v
z1M9C4AHC(5E9M-Zj$99vyjj>#zkYhL6<8<e{1yl~@2Sg{HBmj6r#z%ota-~~3nt$`
zrY#tWN0Al%#8S7teJ?^H-YIJNK8wr=S=_z>e`Y1UN{@MaWco`x%JoKO-uccclnw|X
z;JMXPf8|@zK4q%IU_4NHgNzB4kYJ~;wBS@fR>*W+PMulKW+s!aFxrfLZya}<k|FlO
z?YXRLtr!#TM(Rvcr^$7GONcV4&uG<7g~qUe;cdF6&?=c__Y=Ij<dwImQ6mry>eZkC
z#(4v7(!*;1WsW94y@gl<6oamO1T^bnsv)O%Ytg(u)+#D3mfXqkW20dQP5b_m6aHzL
zHlH?U`cF8CWgXfPCx^t_oE7?tw7Drx$zOw0Pf8Kx`18aqAMTiQK)@IPd4JA!rXzkV
zs84vLoaBb;$Ay?vx`l6Ky`Ai;te%uBBahth&^V%9{<-2b{gN%X5n;nf;_F<~dATPj
zTcf4s<fgnDq88f;J~W1>doN|Zyo|It%{<GwBPI#nL?>}Qdt%kJ_LExSqZk({&O9V}
zJ!@sAA97e>HtHs$kBoA3wddOL_N2BL7nWe6y4gt$FbepATSO%dRh42UBl6vs8GHLN
zZ6fP4-J~F`%0>z<T*c`tur!u$O5(Sh6>P0GO%*{M(FjG)SYhz$P>Rnw+3#KivyU>y
zc%~O;4k5idbL?_;2xdh#aCHQ@Q={YRlW+!XHdp?OeX{?Ekuqmk0EI<^ahR<b(xM0H
zLBvw%QcQUa98%GZS@ls8ztp=~K$353S}hDz2(g2oe1HC-8%PL&65EEn(DdAk;3kh{
z4LA7*`s~q$P2)jw2-gn^XBAQQIA|{x-H`ozvvvk>iR0er0EP`wtRl5POslU-$v;Pi
z-vja5Hkry^{7j=Pwiv5wB?JYQlBO*OKJ>vX4sgaGj8a--2dPi4Lm+eFBlg~;7aNAH
zB<0M(aW8tVES~j5G$rHP^#wHN{3Q-)+DTX&v6C&NBnC<4dA>diGeLzC#5|MIgQ&~8
zu<RtTmv!Aqo?hex&<0?{D6pszzv$~1toelVhrDM~uLUYl$YDy7Jz-twCbfi_;Cfar
zNYYJUDVZkpge0t>NxC5-J)_}u$2!4$a(S9%Oq-YFm%%BQbN|Z@i4xgd<@(9ZSl;20
z9`uAL><qMdZukqV0Axp}T?L$`SQO5G$MT#rN@9+_OhssGskbpe_Cx<EwGGi6p+BVg
z5=B82IC%K$1XGfx?P`ApE3BxKa<6;t%BoVyOpZ58HAx4Y3@zq4^i@6hH+t;3*pDCb
z@7AilZ|>t)y8v}+ccxWM;zj`rwzOw%*y3`;G>W__uh^|wST=C6y}yW?6GC11dHSZR
zDiAE=ZHG;Ee{x==xD1B$x#^fh0V7enGKJeDP7j8Mr4`h#u(B|?4-Y*SGsK&JzPnf~
z8dPI$Zd4P-UG>xMkRXj?TdVKeHAVuG_I+dKCGs;Bg#hIBcO)meYU_44OSKDWUmm|&
zXF7(Z7m#>hN0MbZ$uk*R(8%t*arXBrD85B|0cYPG$b_=CSo(-Zc3G<6c5B0tBCiMe
zEp2LnX@}?b3Ova_j?p<Od2-Zoxjd?Cpt?+6?eam#2V!`>HrT~{v8*wG@KSy|b}qZ|
z2Wi}9jRURAVlFtW-~%ac@p<lt&R2?CI2^@v9x7Z5cCloR)6)JNr(c@;)==+^CsdFq
zen*|7Qt|65Sz2sg6OtN0`pV@#?=v`7_*%w5-!#sR!jgSz8Yg+OdT;&}{8d-nKt9jQ
zjDOovHMjHyQ42dQ*8Pi%jz_v5bBf__122!W-xS1cPufV*#ChwrG^DnmOInUbU7KbP
zf{+CCGTxzTr_?`suBPylf}!44(WkLla#xp<)Z1w_MX}d1lwr6QTG#1LFH5J28|OT)
z^#An8249<zs9QXvmHfeRvEa1Ah<g2{jj(LEl*=1G-Q<5QfJ&<xI^c*c%AQ|!<>+kI
zbSQtDnS!=ns=<2%hdcD9L3HMyoa(s}ukKap!P#y!crG>aM#6((TyX%RnafWwK=yV(
zh9)WB<)>*Is9zL@;-;=SN{L09aYVg6H|JG$I%NmfeFsqRR25+xcx$JnNtCdGuP=KX
zrd{%m?_uK(x_BJ4MUSm=XA}MEB@y(+wf8u3VAwQU%0s#@dmPNXGF~^?@bBf>Q}Tgd
zab|at{E7oUb2nmfg8xM;MR$Lcf5`H%l|(0yjGr`H`=(~7MMxZm9h2Wb6cQo>t8_Ww
z97ss_L@`u6C%~R3slUJRA7|=gIKcQRN!djd$^HK@cGh80{oDRGKtUQ25CjB7ngNs&
z1_VS96r?*OrON^7?rxRt?iji|mF}UD8NwlkW`29V=bU@bz2|$M=l74n^Dwjb{(Sal
zt@U1Oy;j(ljD7}u4%5mZ3iIQ1EQ7&FDgHa+AQ>8a&XxFK92xQ)`J8M$ELuFYlF!ri
zZy8HpByNdEaQob{M`CqQXhLqzDLzZtt=T>!qv%Wsfbp>VaML|#JZz64eeY9O%&D8G
z#NoE$GR)IgbBZI5Rj^$39R4ufj_D1mQ_jsL1|82@GF$`hNWmu9Oth2X&aXf5`AO;@
z?Vb>`CT5#P_Wfyl^O!`Bm<vfit_&8lzuctdRpE7XzDJt>xa2ED`A7Mz-3#&#)^RHR
zdv{<#B4_k*NbhGcrZwyV!ogtk*^t~bYCLWtISyF{i990iU7j%B<8-pHj~bJB<e#jW
z)2bLXE?*5vwuLk^f+d)xR`{W<BmQ5J!v|z)&q#9@zTru9RkN!ZQa*ky`25&`7R=?G
zW3BgE-d`&H7`JnrpFiUCX@8ALHSy)IY0(#~(Ixj-3>0sW`8U2A)H5x8U$0u2Q8|5V
ziNqiJpXVSP2TgWSj?`6;?s+^8$IijeYAG*=vHD%$;>8GdB6N+uIS3OTMn*)562W3N
zBFTFX#%J~~9%3`~bB&EiiN6}~J7{WMlrfZb#S%<T%5Gtm!KrRnK}Tjg$Dq$1-kxHo
z4ZdakJXcjG`4QQh2js=r9>a?>Gv+CIZ80iasvW}d2+uT0)G;LD73WDvGry_x5+tcE
zO^OO`_1=LmW0)T=I_5CPSdm|dN8rYtB7DRiJUmm~U)<_0ldb7U^~A~-o3K(8ara{J
zC2342-7G?xW7f?<!7F&^MUHJ`{?}w}jl6q>X7Y690n=Acck*h#!-h9`#Nc=rP6~Pq
z57%9y0uFGND+J>cYbWz7Rg%pHi2WxFO^$>a?y&xRm<4|<(8Gl)b3G*=zQ9{#Xyx&L
z7FNkP0t_O894?ys>)|qkuv|#Y!^KGKT80fd_d_VgVhGj^_8$}%+yB^tdp;yw-eR{d
zhr!#Bqc#x9j(d16*xy$gye{4j<Mo4$Yg~F3KZSp`rH;H%!8)M#UI+8$)#fG5G3<+^
zUnD{zpX4=9b5~i0KQjM}XN0xiTi7tWm45NV&lFoD*>M=#Mp}Q!fZ!<+)g^xcK%1AG
zCOxS`R?ps(%E5c6W+856@xrm?1GtJ$PMx=OM8}+s;(M-GCyn_z5Dy`1I~1onsfElU
zp`&L@I*d`GzmqFY7saZwM0*b-If`{jko3?UIr)1%Db(hyey?OCJ^F{YZ1=3Mr>Cdi
zdPu;+UVH8kU|W?P8{xm}lYKSVXUJvDZG5GH3i;ZLL&JE(Pgl?u!RE1;_eo@e(AdvY
zFY<7wT3q(kQVy2~cUxfDt)|-jWDmW6W+1HtFa<=x%~HXn3b!8kg2<oA5^jEb{pgOB
z@!o+T>M;@*C*JJMeT2_bTb!K8njXSe*b|19nTKWD<TIZ6z|>7#`ft(~qI)<D%C4K^
zhBEIC*3aeX8AQ76Cy<Vv0;}%xzSJkj=8A9OB%k9S0}17xdXfI0#HM}cBvcyoJ;)^F
zk-3-hZ`VwfmZ>Cz$v%L_c|9NbBGY<N%|kcOgE>dp6Q|1`Z&O`t_H&T^(=S4=h(*)l
z{Zv=af|t%(&XQw$Y~SPTxE-{pD<8Vz2|f_SW{7l?){SZ8s3#s%RM<pnYrKA|lkU&v
zj}ss?U*g3DR>K1%$${$^Fx;36;!G$|McL!uB-&b~#MY?<f>Wngyy1#^M`Mzs;?G?y
z#&C`*=rm0CbeEPv3Afw5K0GE@pO5Cvus(rR6lD>2JPKi@?E2)w?qR}1cf1~!!LG2*
zI@taoVGU0}@M$<s?HU8hbxm{``EG!r9~D~X@bMN)>w2Y;y!g*fu6gY%7YQJl@M^F0
zPC@{0=1u&*=u?Kt;wW2ML(caDhUUZ@X?6aN3I!)80gnw^nN3(?$XGqt*BU=OY#?&>
zzaOypBIj`y>}oLz`ZY?zy0w<XkT4jwKJmonZr7Q=#}#dQTJ>O9<<Q|wGmWETHxOe)
z^lEgBq)d;sZOIzn%H_OM6<<48Pgt_d5=5e|>+10G@YLKrSwFiZ2BnmK@odNdGA<v_
zXI(r@#^m_i<Y&F#IuMkDzu%4FQ{=a2Nd?bIiSG{t<PI7x8`3Z3XWGoWSbZuS@$-Mg
ze1(vrmFY`+gIA)F(sQN+dy1M*(Zwwe-kzxcJ+LGrELxg6{Ugo$D6JBEhBA)(3DkPS
zD3UKfJVeq_L!&#m_yh4>0Mo<%TVOoWo%~Y8$={tWKHv$Do_=%=KMvGtjyI%ii9}=l
zIJdaGNLpj+*j_|{Svs1~aPi4P-vq_6FtviOXD}IJz1!JbOTZ1*HSeAKvvXt;JpbIs
zRU^vKde$)tteu?Phvlf)7E*MuFp21EIRANN8Bu$CAt_+^wbO50;d4zo20iq<UY%X1
zd9{qkX!92*I|A_$=0~A7lxhmo_*gH59?tcfbbfVoNFf*=(VT~Ag!70+LSx%V>}6Oz
zOmfx2dXpgJH`8OecS>p=JBTg{9Ph0()o}-LnEs-00kTdfHt(G@`@$bPF4P#L{T5I;
z(cfeA)c6uPlt&OV=J7MlIO%EQ8qK-(3pa*kx2>RxA7S;PJJi$qIZpi<*N#Wge0p>h
zw5Y1^o5+;hA&qiLF`uvzdh(HpjrT`_-|gQIQ?pk@i)BSWGNBtDvHP>AV{_C0kE+4Q
zz;-S+73M~K%kbBawq$iHWSQ>D>T1X>-DZa%)4$6x8OZ9o%j#muJra?9G|7xC{-aU!
zQ5oK;4L8&*BIR{??#WjPTKlR-F+5OxmX!dszS7kk7xSufIqnR_%js{8=W@l8m`IAQ
zW)1B%Z0l{d-5{Ek0)A(SKyHH|pY%zyPgsl}Ey_Mxq-y0ZzoXyqGk{CSY0ku?p)lR4
z#MiMfN22h7f)LQO!u3N>RY~?yjVyRsr>H`e&0ZTmY8@e_e>)zX3_CLw+IHKHmyLbN
zQT(w*b+x!%x2RLOs9kx;@pWOReWCB18j^o4NVXu+Vn?wkLyJvFn-=*;I$r<k03`cp
zRF(->_7UkeJUOA6RaL{rqu(uIm`zn&6Y?lL*J5t!k58OaxIu|I$cP{kJdI4VZ`APO
zzX#uUDtw@0dDjCbSmJL+Ju~&HRt1}{N*kGv3bbYFe7;^>-seDCHmFjVqoIgm`<P@w
zGI<f(s-A%02aUkX2S0K?k!9+p;!3FhW^7xcA{RaYFnm%6<d%|(!l9MLy2Ht;Jl}?+
z(p05tY87i`wSGqGk|Hjqs4{5Scr@8q7c1bG!-9@RuhaD`BKae8E%D`MQdCQKdIw5}
zVDOyqtwz}|HBPB&%tuW^`K0kuZTkyq=$gD8^(mNYp}%S&)?(FVtGx=Ro>p?#s(|OZ
zA;MR(HjP97W;{uDb8!AlUH|Mrb2}N<a507Wf%5WjCI5Y4&ZH~1METiLl1UT5eqU-b
z9a9sIgMVkRgBq9L%l$c9j{+rpWS=6fSILDAa3&P6ZK<O@fwqU~^FPB%342NdGyLk0
zhdc@@CQ(<eg~_2G?(u+rA0Zwei&;WjW%gsXWw+~FQ>KlN+FXjYoTe%vnFB_S2DPs(
z_VU|KJm%d$r@cs@itzn@A|UYnXj5{lLhso81FudWXi!eRv|G08XQS!0N}{VygG!dI
z6tAzSo<N=))34B0)u@)TM)E&J<mQmbGd~BQZ?<xBB&I61QgU=UioHP#{#@=-)O!Di
znt`=KS+%F7kw-*H7VKPy?KgEoGbhBa=7fD#qk6Q=F^8{B>kWwYF16PMegiC}CSH%H
z_<gU!(l@J}chcZ#3C|yi`OW<o@FLEE9v-o;_Y}rfWFmF<H43Q1J~$<+EzNR3j!cUU
zHP|=ANAu%MWUw%#oFK^zd0$Lg!+@Hxy|JGM-&0@X^ZxHu|9uv&;-Q|%vfbZ;R*4rU
z<Ze!b#S-<Jp=q0okt9E(G-!}6{B;$x%b`o!yOm$(6nz<;RV%Eg4dtc=?1d8}u6xAg
zTH?l>MX@yQsE-$E-V+wDBF`ih8zD{2ZGOm7jJriejxIsf-^><##Qp8`DHgfKQjuN#
zSS{*SnCuO!=D2V|yT!|Y5Ilq#@0eFYyA*3RGe-h%4F$V?mnrPyl7rmQE6Qpr(w!ao
zh_39R#p`i#Y8H@Oa_Z)d<V0y03|nqiFs{Www7_>tvNXyoW{nz@q8vdBTDB*--^2nc
zdzNPnt#9J%(!Z6|YP_nBENtzT1<yBsUM=$0f#)d~N*BmMAU`>{DI!vc0NjGc+bN^e
zy=S;#uhl||aZ8L4Jd=rUTAmbWZkIm?t2s}BK3A_4Ne?7rFIkKyZ!ZBCBve3qkh(n;
zUX62ygB#P!o+yjrXK#W3FqKlv3wxF>_7_~FqJen!@!W8UBSQt4z8$o*2M{m@nD7^r
z=cbD}+@}H*p5LgPX%19eGyK%vwSS%_ytZJgZJiiv?>nYW3e!!_6a&`g%KD}%ZD48F
z2Ug5hS?A?Wi@IUIY(&QNNI#|1kybTB11p?MypkNNpzIL{S+uy*=I6dWJ<?Wc^H9&c
z@kH?RkyI({D0Q2+^n@@W#i}Zer_Zpw&E2r&XqjYUduH6^BNsY^de*ty)im|5QqFG_
zJ~guoIwDqyPWRueEKKREwqQ4$TQ1x#7O0h-B7q>IK(wu?XLIW?rY=;_E&b0WU*y3j
zlWKyV^#F9DsWb#syjNbQnH3@7ZTr&m^lU(@Oe{Bo&1j>mF*<-2Vc`?S*E+-BG{vSL
zVkEcY+MrOF<o;4W+CA_v?yStYrTF~OO9gcar->Qic|HJeep~dCjbyRJ0m1ZO#Mib#
zh|29`yg^AINw#G&N7Jmdw_GRPOEvyS`k3pE#89?giYgpi!y}&WdKAX)l22=5dvk1{
z1aj4w5OGTEK0_2!7{t5&(l{zrw=K~VKzj@0<!O##>Q}EYiuL&dw<UnVpu!ULa0<^`
zj-QV=!gl`krsY?u&}^B&5B|P^<U9#sJ8@PpS@c5%k;>*(0^84oC#o}2I44`JTN*9P
z4$$_>D?)r}HEZTK?UZ(!;Wr2M6tUvm3ihp0w~7Vi<hd;8!igKAGnUCUMUOLyK5&YT
zPbprxQh#y$D#oXD^9|>{cm`p*Ys~B%_6{jlB_6{TL2o)^{l?wvpOqd-YNZf6@$l@1
z$lNAk%Tm8E>u`{z`3Yz-)>$*KAzSN5Lb2|i1wxjk4B}B2bZjL!|IND;&x`k;NS*lD
zZatJQ=5D#MSOzf7B+1<Tqmv_kSYJDB0<ELGlM!%azSGcV&G^%^eJ##YE=}8$k!YOV
zV4QlsV`P+<dlRmc)-|#FR!MDrkE_-N^$GPIY@}W9W3w8hUX&ug$y`KVzns5x;2Q%b
z91ZEkJJ%IK0CR5p(9G<GZCd;NjTmWKO}PE_A<*aDdeY;vys*3`_}%;Gkr3e)p#aN_
z+}qx$!-KzC+A+tLCT2}I>Ukah0|zwhcWaIE-Dw@PrIe;G^>pnqxn_yUfkRSL{E+_5
z;gNBI{a;%da}K7G1F1Hr`Mj~v{|)^4ALtW|hPp6Znq;DYIXj>Jm8JWSzh=!zfGX#F
z>x+`F|8YtGjV)?7eG3RABa5z@sQ>tczfR_VeEMG(HeI;mbup2ih@t5$@ov9ziuBSo
z_ZpDXb4Ya<HU^R@2HK1(_uA0_wuLh#p#1r6-GWU|n=atV^bByQ1y1EOtaWaL{rjE#
z*L5I#F<}=zV$2g&?F7h7e}Q-0SMl(a+7>st)NcTUL5axKDiP=s;LZD*Ep*Xsr#kDo
zYzxCQ^-ddS5-@Gx5gPS3C*}H{2dAFp*OEBRsM>N)^@c&!!8l-U_FiVcW+3GCYq78E
z)G!UMFPV4=+OUghlV$4N=}eYnwgCVYmA?qUe^~QV90rSq2pp~LfHq`zDRU2?y^tOF
zj+Vju0Q0AG8SN(qn9tqHzC=r%bt-M}5Z&Ld0Y3$%pzf0!{fQp>VkU_(N~FsDv<b%P
z28<?afL-sa?|<zR=9jv_(1eD3?$c^S3a$me-7oz#T`UiW0nc#gmo%O3GQjLf!q6z5
z0Dy_o5@5e14Pjsn95VXn06t|~#WD+9>>%DN#cjV~^LRqRc4Ow?X8)j$A6D|Zf<p=m
zjNuN^0o)HN9{=_!3SN$gcFb)bPUef&Qp({Vg%?)|H=zKYXQ=#ozB~cLop#?fqGQKE
zkA?ueizb)Hj$9H3Y_kbG0UNp5#g5L??uQCi*E;}@MMY@U|KZ>V8;7{u7ipyHS4U$Y
z;KQMVmq4=-K>HJ|b_a+gKe}ZNZSP2VzWdL;YH-B(G0M^!{s%BNZeK?%^<G>7+#jSJ
z@Gr-L2PbxCt3gDnA+$vfO_`A52iJD9pMr$^W^PnTZ3Xx1*8?%w<AQ00I2awOB8ULf
z1I}M#Rn1$B)&8eI8eEi6uJwoz1Z<+9^s5c#y#&`1JHeAl<sh-EF0qvOHZ2-Lv5Q|(
zdCzf=B6>Y6#Q9gGPux+!1b3Yc28ze~o~KPj7(k7{)0x_lEYhEnwOrOl+np4xe{v#D
zRxe@b8%ZYrY5p!6!P>tt98AbS*>J)sxP9>tHOrYyQV)Xl`{3W=l1yiHp+}fTenEhx
z7L>fSYuc4BdQw=LR=yD=ytE_VwcoTSYQ7-5OMoFksl}?a#{C^ttRj`ZonkT%UnU{S
z^EkW9+SY9$q?5}3hr)y7V*P7)=0)>-#YM{|r|sjiFS+a=2K_#lklS^!>z@w-FpyJ=
zHaZL*X#xN|5U@{(JZ@R1IiM^9b~f}73}6DDMCY!1uN3y-Yd^{9t`Wq%z~JWJQXncl
zWT<+$K8zlH19Te;EdniM*59rM(ys<804U-I^)d8CIhtzuPoTg{-zu#X0yF@L;s@SW
zXLAU!<~Lhc0CnjXPJ=~&z@$-yvi{W(0Ev2g^*ud$yLYJO4?suyQqXr2WWg%Dhph6d
z-cc;I(@q`~4rwuw{0923qv5~Li;5+N3iA*!IWtZ$Jj}u%Mk%U|zqrM<+d8XV51}<f
z6>UxZ=r7t9=~p*}L+RbrFXQj-=Az4itNe>sBnjSsl>(X8q*l#oI27mI1$qy=87-SD
zrm5SmfF0{xzw^O%zJ#<d=kAc`iP+hSpSEt_!G5rp*KbKWDjWMS?HbhE>FM?O<klbh
zz&jn(1n_A}<=^d&z*2DB)ts2q-47>6{9T7#s&EyfHVH8ZvHadR##jxni!t=o*f}0I
zqbaG|i}nHO|GuZc0?@yBRt{MFOqHj8bXNv~Ti=WVF{|2zIR_NTHjc*7Ce?xx8I|t~
zMfdOZ=?M&49OQyLcKj#fyolWv9O*v&JX7~NQIB(g0bUI?yyvlUSa{0ojLy?IVC(w|
zo^*Pz-V><r_=5l<Pho;eb<pF-W$v5oM3JD&Rz`ndyytW%hJmwSkK$RSDSWJqAt%6k
zzuPNa^Y&(Ar*p(afD+5YG><vIeu@-}olU-*6CV4YyR`kel!3Bd8nR)`qYMB=fmoC3
zyO{<RrD>*YY)jv0;nmU%id|2L@%0^_d^RqecIVtT;d1;QT(y-B+^hPKGy?m9t=GbW
zEE5&;4u&(x32mGz9i$I>qU3ob;8nfEF%;wJZxN$hG^OsCs9$;}%hWRnDRWF!!u57K
z;t0AVX}wN8h*tmzOu4k^@bvlZ9`<`P4XD~=$D0Xx;nF1w?I`s+JYk*rU)#`?p8Y5c
zW??f#!QHU%rUd}e5q)jHOI7ydEUF8`GJ_t<Ca<F(I0H@mq5%90gsy4^xK=g-g=+dn
zxui5EGrG>!P&Dd<80`iV&IcZ*J}VqlZvv(LWaAQbgg}>hUeEX3{o`a*e9GmC3<G^e
zi+ZA4wCKqX8gPsOa2+Utj4P>S*JRrjs=(ZeFjH`wpO9seTEM<<Bi;(o7@kB4?3vYa
zan6@({YbQ^oUsub9eQCZu<Z<={qk-U48W66W1BD5SNL}@^;M;4@CH4R`T8Ez69Mi0
z6<VnsQh&H|U#xx+Kn|rfdBLOT%8$AV^V76M0V|q>0U9j=uPw(^QA2Hh7}yhl;z6%W
zb<^n4n0;U!?gjk7)@VGJ+ol)Q007QxRkyDGQ4RFJyaWSkhKk3y1FlWG?MYLHi6-{%
zp`EW50V6=$`X>NXV##-2&@?|)#$V|gcW?&N1rT!sB#Y?tbhNp&KYPf`nyRmPfxu@%
z04#Mf4~v8^A@okEiVVT5(1X+FZ7@_e$yTJ#ZT-x610fLv6pGBnotk+7IKbr>ivV6k
zE82-^f6Jl5YmoIIO@l#d`I;H84d?3E92%y(Yq-}LBBys4dxG-^yS43!d8t6v+ez&7
z$a^yjfk7D4&FOgn3h#MRJd8n*`3>P}5Tc|hH331ozp)yWbP#D5#YseAEOOBmz!AwQ
z{QP`_S*$^Kk|7lX%z1ostbne$-WPOW;*_?vNfVnev^}@(VVGrl2B5Q|C_I36)9I3n
zFwt)4s$u4GjmTeQpk1U|oA8}|(xhbq@ncAzAG>&4#sO%{sGQW*n&SZQmbzLm_a`{+
zBl5a8*3ui_{o=@i=7)LJ-C^|lH(%^uD`3DZX`oEolPfd!O7F4<<Z18l>#jtp*5vyi
z2@(<Rh0`38wMx`1zlsUQF<%b;xGb58!T<s*V{E2kZ3NfDU;L<w%qf}<e8_~xV9zeK
zJ&KqNj~-H*3&w6kacqX)#P>L+Pw|=+sJ+Akx-8uxE@p)*pR1D<yT<gI%dbFRJMc^m
zkFP`17li4QXm0x4rq;?yZpl>F8%tiV_ZmEVmKCetYOy(Ug>soWup!@<tmU<UhN~g^
z-o65Wm8>I4G+DM(-{Gib47iRn1mojJ$h`UbnR@fhg(zT}X>|-4#*zVSp&#lyGlo#|
z<cVtQzz3W0d48N;3zsR+BjMOz*ks9kfai_kgg!yLvJYIdD!9lV@lmZ4`JC1Ffy+on
zdhGz^cOtk2h=J@KjLN_E(YjeN(vgJsnhX<QPJkc_lKP27_m)^xzw_UCa<c%@lBp6(
z_B!~{G!tO`1E^hQ;=D_gqJ`qs6V+^8Cz2CoZOXO}F&+sy%dZLd-AeuqBldVJ=mkSE
zRw6gBSC5{Lb&JQIY@|>-^}W#pZwc2QGX%dlpiiYmpY=gsZWp#=kUVca1Mbjm9t~JX
zH;~v*_&UsF*OxAaozEKasYgjd1*K`vH*k5p$p|DSKlI`>iT2L~&<y%S<8f?+gyTC9
z_+0uBxf^^WGWM@JRPJS~pyf<$X5PcJ+YeaRjR#D)Y^TncHElRg`<w*BZ(X)sHnnN<
z-qhg0v(WUGF~h~z_lZ2x)9Vs2vz|j2*A6O=!+FNWmbFORN)o%q1$sGnMK6vpex2zK
z)tGx^2pD=c|2tUlKJ8Zn#)GW40AA1X-UdXZ4LBoPO$AVIY5$Rb#aOm~x8xPPSWbe5
zW7t%|<T<n^R6LVG7*8{TBRfaQv{<3&1{vBmCb=ubFc|xs%s3rG0bAf{?We2^2oDvU
zX}dWIg3oZ#LXC3mb3kh%04~U)dkXMU=VYqj<u(PPZHbU!m3xt;jkh))sMmKsL-+dy
zEIb5wlh}Pa*_f)s5;7X1#AQr8fbE}6?RUCh+vfB}y;<~Z;g{~1?zi##eKvm4uQ#fm
zg|0aGE-Y^G-^F8`mRbLU@ntoI*`n6xabPZY*M=2!a@I0~XkXRSI5YY1<uQ)=Ahq0!
zW}>`m&SHM%4u0=mR>I*#3mM{@mwFU?wG6~uZRXdUP|F}ZMY)~*dn~%0HrxTMNcwR#
z01EFWk;dmR%tkLACPH<9_=&`&Z)Lq<4BE11jn`|pk$|D*{?Dg~`5$#yJE8P$rRb+Q
zDaFzxYAX|yz_W!alOa2R3nBn>YvF@gWTXS7j)h$k<?7LNH-CaaAEOaHbR)6cYk#0Y
zZ$jG-OVu|(e{xStzBa=ZPVAnbfF1NxEzxJNo6XMsXj`P4wli3YF|7N><K`Yn@LsbS
z<H6!j>JA<qND@fj{VMq6_B(3N`-q-LcrrR0Txhpt$l`aRjvKnRKp$kseY!<J{V^ZA
z7-g%_<M_4gjMbnw$isdyZ1!m1^ivRr&R$#X?;*f<vMPH7esVdZUA$m}FyW2Ri1gfj
z_#G@dDvD6kSsz25h!P~<;m2k86CqUhKm-HmfLL`8U{sL*JQ1z^+6kq0g*-KFbFIns
zN>tzk>C}O6;68?$DmTOAC@%g0l8t2nRDtVb&AIh@vI0=-RB{w4c5@~g!gszeh!DG3
z7nA1-{{0B;YP5Zay~`!I#XeI5bO@P!c(3Hp)sSds)3$Ah?t43(0Txzr)z@k5QF>b%
z0#n64rO<Nmja5T*8v1HmfG#9vUn!mASO-N9bDh}n_Fhz!0HWK4Sj3M07!&8ZoGpB$
z`EF+u5l|m*kBETcZBZLp$Sqw3uUrZe`g*<4Cw&MR`ny}?7rrA*R@s;IC@cY~SQYU(
z2e`nt>xDL51M474s)uf9jGK}`x>s+rc-*Mit!L2dunjEg%AKBcCY=L3TcPf>V-2Ol
zxlOed%*hu~A-Nn80yZ(|_zKuvXNlz#f$hc=#>gts8m&?K3oOq|+LHq5TIWPZp|@@>
zk?6J0n-*7yk(d$wkl60!`Oem~o0D*CsA017T6(4Hw5AznK{totK?`t@w;S>;M+t>&
zI{Mps9wo;y9*~ULFISovbdUYreUbV^NE!suqco!a=x>^iCtQeV6YK)7t8D?as27`*
zYy*s%%|2g59HqY9Rh(WQ53YTVT9ikOhiLZQCBZgwh&s{HC>oL6t-SSR)j!$l{I(I8
zLSQQ}s>QsUR*Amk(4aT{>9}~BjCkZt3^m3rXXfsF@QF<n(RVZVZ2Uo6NrG3)Bh$A1
za^Rq^BfgJY0xGNZ0#g|208xe3>p>xgWx8zDWK26+{QAr2wF_~7^X_GueY>CzlBItj
zz5e}-!@Qjf<UTaMz{|I3sw^O1Hf2INWSs~smUHV-s$a4L>p9Zas&bpfru_4P=71O^
zuC)R{V*;*E<0>FZF4>L0V;(6v1Pn(ScHfOgKVTPOmIoZb)~2Esumo63tzF9L7n43-
z6*!n3M2yEpV^9BJF*+NgcFe;tEZQHoV3XjLCTi0j$PftT9=7zGdWR9TUhkOmVozD~
z>}ec-R^DlO&~cd457hn{^uT>vI3mTL>CmATP0ZV$rU{`*`gG+?>Spvt?ex_3d`7wG
z#i(HRhzn2B*%za|Z+6L2yGUmtQPX0&G(a<m3#F!B7qnSjmQlYS*SYZz&g1{KrPw%D
zu7(FquxlH}&@&51`d-1OT4yGBq!w51r8huHs^gdVgEl>qZ7~qg_#6OvTwE9I2*?Mn
zf~Z*|<{lkO!zh>-x$V`@kxrbDb}#jKOP;V6Jl)?CT^MrkI~R{n;$+6(jA{YKt1!8U
ze(0>x98l)URdMW?^ZjH!U+j7Z8|za#3(Q>{i-Aj5hnKTr-)-24c4C|euhSg4CO|HK
zb>5MAT7Fucj&>Kk_?@kNPV98#4+4C50_AU|J#At7S8H9EKN@fbdj^xbmks+l(5IDN
zV1WUypfg=I>2-)Z5<6^D!2_<0r<W?V+aS_)HVcq|fu>x*jvd@77Bdoy?CLw)PQt*h
zwWAeoV*ZD1{O1{8fyA)_rbDqF0XQYIdJL27<Bxo7&kSNN+pk^J&!aS>E8$uZCKJyp
z^*sPa0#X;U7Q!r_b#vAyW;1P+`2^sQS)PqO{3S(sT?%!-hYG#;HrurJsC>5M?D=<&
zqcEc(chV|fw_jm;H7kM3qJ<dgtP6eVg)V1BV*)lRAF?scBJlN&g6)`&#_C$AC;$dD
zzpL*INAQL%i>31m!V9vWABY}(AKTTZi8c;~B7kVP3nMFGkC|Stu0<^$+Ru|nwVoXl
zsCn7C3~8G(3AmsD)^XvI*06L&vNK~vyx*9d{>8Xcll4{WoG4JZoC?Tqd!0Wpd|_?P
zP);@HYwTa46{4H*%H%r1RkOmLNlC3oRY@&Yw~;RqG-DdM<`mXuj0R~2B@A;hqrLjJ
zB(KIU*D&l|TgN!hu^(rE7M6ZG)v!)KyA)327^+<ZNN&yw%g{oe=V+b!Ut_H0Uwhc+
zV^A`ZS7RRgj&U#R@7DvgSkZwlr1Q=7bZkzFaiaeJ-ok%boy36lJSNncKj!tE;Yf{s
z3n08YeZ6dSOqL)3!n=ZmYxTtgA#V`?GI*D&aO?A?7M8Re%{l(&Y(BLmiM_|`&x@!!
zXll1I3KME^hQ9QDdUUf1xIu4ax-Vt8A^S(Kt18np%~y;-97vXf;5eQTWOe$r6$V2+
z%N-|B+=UTI@v7c<3!>mcq!Dx?o<uRM^I1@pFOPB9MaiZ6Xq*@AfQiKrCFZ`!+D8bX
z#PB~&fGQ(UkpJa;M^dN#{o}>i0bYwIIoH9`B8IVL4wY~bKysMBv{7gsDYc!r7z3nl
zA>=1Pd9y14OYuFx$6fwnhxE4FZ`kHf_l)Fbh~ani-cAKhjlc!q;iUstcGfuz7gVEp
zQ<FVlgtnQ`d&`FXD=gj>kWfW9=hfQG^ZOA1t?R4hWhq01C%3{IkK@}L5HEapGLOUj
zb1d9Fnr;C@w+3;TRI{9so>gLUUk%<%;?-Vahw-nCfD+Yc@Z>34Lx52O+o*@>rHqS7
z01^|-hBV>H`gsFUG)VZ}-9Q=S|8W@8`h0k|t2nIr{pmJKR`tzwukBUu9Xwug3!5*;
zk&NXnQ&Y0Zk}fQRINtV}S8w0FWvQ}5D(DDwhBg7AFA+v?OJ*X$%dFo#G^A+KXae(d
zjhrE)JVpMrIfrR48WxL0KTD(kmPk#UZ2!yu@FSciSxO688t~H&f0{>OEWzyuWCJ5Q
za#X-S-42eH?Dg1sk^Y$Wa<-3)MS%1+N${DS`yY&)>gL&T;8v$?J#{~GR{L25@&w8V
z#b_Y+MDg9&-zWKE`Mlf%x6ubDO72$;Pk24~5Ko6!!5Ar+71Jkv%n2o^GjYErX(;9@
zk)t7SzM|&uSMj(jgg8%XyV|@xKz)1qI+x9E$Cfh^Xsz#S%SHpj;4&VEV4wZEtqUgb
z{&7s7R4^==bto_s^%(28B+QUOtgl4(kW{b$UD7U=oxR#ySGgkCxR9L<s}xLt@<e`n
z+kD~yQJb$$GjF44Uctr-7I68ousHw3ST~+gG{%j?Wc^I&j{4*E1}CZE>P1A<Dliuy
z$|~o&gm<(0Z+$qx5l{yD5)T?g=;MECrpsifS=r5r6Kh7McmTSzHs8EVEi#vKomVm5
zZHPDYB|B~FRIfejcw`_dR1B{ca|mhoL^!kl;R@HVc}*$0XRcuR=9ZXSN%BRq8>7eO
zn{lCe5zeRe8utUH(#~WUHoV&vW-!wRSlg)W4qTI(UJePH-~6Jx{<SfO!P37I6&KkD
zbi(O3ifQKQ7g(f4MB`3EJ>RgV^-8J|5zP2LT@ta_57$V~v*}uZ!4>Hmf+y#;2QW?R
z1{`|#s}9@{YIJl_T0_TozMWYmywx63QW6gg*PU%gxcAnlOMhmO%ufZ@TlBwk=s|rl
zOJA(q{}4e{xC75<-Z&s5et+E+x&3%=OI@w{qDaoPH?hRy6+gZ&rR0x}@Z2X)Vm5D<
z0V#*s;|hp+2mMtL6Pg3gkS~%o>#NuG9|^nb*Khg(MNQBk^AZp*T<qv0vl~f#bEXqL
zUbX74K2ZNmd5m0dxqAPf$~=&EbHdqWKor&Qsbt0q<zCpZt0K8nLzeO&<<p+Iq||`^
z%bClD&ZQ0FOyFR{%s>f@mqVo6_il$dsN}Y(+{stT%2-t%s8<R}QF>RR6qU~3Y?@F^
zo60#wTBx$wrZQfvVmW{fDVWyF-lkx0+I0|*ceSdRTyNq;%jU()s?^;-nKdmGy`-W|
zz~P+gELLIfRa=&=s#UPoygqwmkKoM7r6uCL+;2MYZ7xnwD<lujjTJAonxZP*tbID3
z;-$h~sAOx%b+YwS2{%sZU6c|hXxrVbFZ|vD+^k$Bwt5j5#sK@oR#?iuqr<kU%f_SY
zv|tV0tFwWuXQ-@IsidctMkUL!6>9XIK8B?1DX}4;Q}y4}9<QsoO^*aAg<SA6t5$jS
ztea@jo@@V2d=}C^{}eMB%JQymUez+rq)~hB#XAn*Mm(mCM#4-(yh{>;uvvX(CEuFP
z3Sz-4x0N`}<xc1{i&U3-O`1oAJb*kE%T!|p<`<V~n?@T<4*xa|3~4{a(2QKZn(Put
zc%P8SwYX}^Fsc%}m)xVkE7>j@t9bhx@T^fLz_%1BG?*@(d*2tyz@(F?laQ~rRX`|~
zDQCk%=^Z(+L-YUNx3nw1?$&F;hSESvDOBQJ<tz(5JIUkQ6$dcR5_Y8u=njj5>v)`O
zOZO-7lq(|-5)5H<o`bE}wp;0%I}t8t<MCoKW0D41mh>ealM%^X{;E^d)<G~RRQ)$u
zB21{aB?)9H&+P^^sKalidr8xId=UaH)^nkw$ALAD!{nKLP}+3ElPo_akjiIrw-fTk
zEd{GXB~MLttS510(B}f_!I<eCghr)(W5z88S(Dn3p}*3*J(P%0qvx23MSt^fOR#`!
zteUS@Yc)MiJ=O{^X9D5^=)7}WOWzWAfCyIy6CNc1%4J#Z^6yuivNv^+ID&avw*H(a
zy<0T!f+9TYV*mpVSBrSRtXM)kQ^{^T{x@F(L<_foFRHu!`}^h<PJQjErT(P_Ms1w4
z*)h)D&KtvIaW*Z~o3t7*Zk)}-G_js=42$z`KP#j?(g(<;3cnRL<@{U!<)1%v;>9G9
z2dG4flmE|;0!GysVF2}!zallm{`R>3=MPiyF?s8%dt>nD|HoBfB#d23Ope^BW?iTD
zYzG(t<cVrSYonGN$Etw&kRvDZZ|9N3XC^H%8_nrgH7VEE2T7pmAsJD9Te{X8pZkg0
zI_DWxXg#;u2MBi9MF-K~_igv`6yzSd<a+DMVWC*R^oe_+oOP1-q_EoH%$RWbZsQQw
z=Gx_fK5Eh9bgwE!v!GVmpy2u(c)nT<K(<g1HD{JyJ7(agBm#_q83HDRL^u?_>P3|}
z9-9)*Z<JJzvfT%IDZsYzH?w`BYLUYG^;H14G0`Xr?BozL4uW5uET5)n0qgIah)FQG
zBeY>`GtFCkLztWefT+G7FQ>CcYLfIyy~ONinpv8Q=^XI8gC+p-8r6_?Zk<>Ih+q1N
zm=Jjrh)~QM#uLgW=WW#HYB$we4-*`Iir-IIs>@s?T|e3uL%UdxoqV&r5p&zVZ6Z|(
zSyjS)U`o>E#%3y|I9!nNjYqnjW9v0<zCImgnqfEdw}cGeEEYC#MO;I*SZ{_5-`kE7
zSRIq-o6y2Ko3YgI!rXW9Q>v$j33PRJYl|CgXZAftSDVkJ+L|1<hLMgnFLlUI>KYuk
z8|~M78^*-YiylDgeP4P>!Hz%_kdDbA6T4@D-t-1Ily?H#l|sBp+~j^&pfjbgyNAy}
zZ4O3X*7;Bq3QIFEk8OTkFOM0|gO(*H^v#UKo(mfHph+6$I&{rzsTR@~5U*M2WDQ-0
z7dH_gChZAoDJ?4YU)kg`R26p9SWAlIG_xPBm#vo9zf;bCMXp|%>e+s~Y>G>`NjQ+e
z`R=kyU**?NF(g(-oIl54W^^Rph@f)vWc7Y{q+gmd<Feb>b>V45t+$wM6nL>0GOb}X
zuR%iaX;{x4DTNtt)bOb*up!;+W-jg#V8n_SfH4VImxMI0&cc5I>y^hG%(hk^W<+!B
zF<GEU8w5<Hp9ft@hf}Un@mbUe0ss{Lku>mvxO~f>PP{T(SZi86rJ!BL7)p0dK5XY6
z&bJsw>5Ep8o3r(_$shdIKbXbY8VBN@e{h)5af~q9rKcnmyRtj|(?!&jpsqJJ781tQ
z*dIq2&7-+snrywe(&5J*v3R-SN3^J_uDhA$wl72x*_K7i|N1qp*A+5B_KWFdW4-$p
z6l?q<FWKf$@F!RBR0RKgf6L(@`@TM!%~Pzi_6j)X?2~kzzB5M<f{#DWl<T(z%rngs
zu)6Q$CS6z_oWi^gqV$W^IGxWolbVXlTN^v)yUvd;5amshMBq1$IgL%{Tgfyt$ev&A
z*h=NV<1#;b`iEM;W_PUZSkBD|v;P0s26zOw8Byd^lL=lKnhZv1RF`Rp>t5=cLR$@$
zZs}#SKANS?%Y$t-y?*04vY61jFmyYC2ioH_B5mK)nlHu$5i*=y$ULQAxcdYQOhYMa
z$P-ozJMqgU*RFj~Y+rS)OP1b7`s*@}eBMJk(m`1Yb<d%^jy!cQna?puV0OOcddJLW
z#UFRSbRcdCKZd#EM~L7nx6`Gi#^RDEty)%kL0-p|2wvmhO9?z~D<MwqGQmR`=ik%X
zocz{0KLf9T8}_81D^>?7{La){JwTMVIXz^S|5DA+nbNZYtj-R6su-FPq|0$i_rf$>
z7(j$BUN$CnqqdT5)mRN}w}!vl03|6-fR*_)9}yw2JR2w9ti~t&JHu}=Iw`0ghCTyZ
zzp|j6w0LAb?2Lm&1_1vWHb|Y3JrtLYInD>mwCeq>AIp^U<@X7$xg^Lxd0ZUZOT(#r
zev~l~RGqiqB|4U^0p#kXhkWWf_5pr1K8un)^?9BXiO=z#{RD<eZMRO?x>-ew@6Fdb
zLYJJhuWG%<&O2O)QH7;Jp3#2AWsMe<RS8J!wkYKj;@xjXk;qeJ-6ki9PLOuZ=T#!;
zyhZW3BP2i7sVKT@0!D39-@&8a5ny@u4Q$O~<%Ll%KH58=Qr9FwRcE2jdI>V43$5p$
zn>YE_@^k#cQE@Rm0IaMXsO31;%OR?%7tRojN@OVg9vli3?RbXJHd}4qj7Lo0cPLHa
zShLznY1lX6^v?U;`%(ka;LK;++++!I&XF=vXcHJ+3YwR(iR+l2nYcJYS-?^AUzE{P
z6pfnA=-Qd3kDM!-#H*&Y<v0U%Zu_fC99^D|t}mu*6qZy*@>6qU7v4NukF@~9TxL#O
z_xQY&<>{c4YlTaCZVh`g$<SM9`LHC*TyHmK2Kmvctm~Xqr!Pb19OMZR376Au6K!82
z;MObrNka-1pOo{0XQ|<NL%EdN;ufLogu=r#6`SOlG6`NLmHunn)TuD_;2O((?8$k{
z^1aJ|Cb~DYvLGB$96a)1ZL7+SV=>sRNLujxLwA;(ilVo>zf5Uc@u^<+rvk+~%V2v&
zVz+4BAx`xV57@M73<4!=fn{^lf0P=QAigW^oslJx8NwSiHly#=(ubrejBtUqdbmNa
z4u|qn#UIbrm@5mo9<zM9fvQW2vrLt=-`5E@xxP4ABGZlfNJI-%8Kq@payr92pz1gh
zvSl^nvgS>RfdW^7l1{vOEJDdNKZk2S&vojTf84Wf*i7t^akcXyHojRIjVP@*v~F<s
zg#KLI`>t-53Q2U^`fet29z0JB`5Z+HUKH9+SwG-|#Zq+UNu6@G#^C)SN&qT<#14Q0
zCY*8+sT15!ErQgJX^xSOX~cJcOm{k2P!QMXhq}7Q|2iq;IPR$an57nWDnZOvX1f7q
zIE>)uw6^Wu(x<PDT62|OLTTK2&U};CE@59MAaMx2irPo&zd2%D3t;345I2PNjAZw+
z4(pL`CPVYbZY7$f`umr9SlBk1pn(qrM)*YW79}1NV0Nj7(>k4mCivI3BHhv{^0g7p
zLZ_}UP=k+cTWXOy;lzh?U@Ym@bbJ;Tbz#pWjneY_Uko&o^!7f%ho)iCESK-!*!B7N
zElOJ~m@w?E30)PFYtMkuzHZi6U&|##WR5nD0Z6QFeW<OMDmqAZ6DW&1=CX9_?>PfJ
z1v?;^(3OX3*sW1bSDPu!WQgCTv*mD#8)R+vcQ;UMXi~|swd68VjhW3}Vj0t*T-R%?
z2m#~NPnT*7Ps#vZ2G@wDfwulC?Nj55!KWyRyY$sU6v0#EyPbHT<4}4Xrt037I~(}O
zp{L*5n;DGiR-b!ylLQrm9T(jm;8mPV6C8iEc&xCT#T#%-%VkrQ&HwcX%G&;BxCQ4L
zX4geP!zK`~CmBQtDQM<4oXWf!EMl%rp1g+DLym9xeTyTm!wdTjz&vY`ZCe}EwU)im
z*FbtR@@Ovr$Bg<xZRjgA_nwPBzK9?Jov{u}djYcW=vf^}yJA{V_j+yHmJ{+*Q|(Lq
z+mh|X+*tTNOr$<+a=Z3Mko5?GiykfkolBM*3W1&aBKz!)kxRf=TBE36wBxWwMWpe=
zuHlyZ0a;R@Ync3lv@q0CpLJufj?dN=GXFAtZAYZkV5@_&+DMAOE{+OhITlWRus;*Q
zQ1l+3BCk4kSNKbDD5Nrl<1|CK<|pMz$D0|M92xDHAh{yL12fK7<uMY!33~cEp1#+n
zbFUCqv}k@hyThcuB%Sm7vQS2KpxK<Uc4!(gtQ>q!L$M|%vyAviEc?B-QpOMObV)#!
zBG9yg_)^x<Gr(eMiH)7(Anvojwd(bw&RYUqNCvJb(TmRtIvEZFR9u#&TD6D!794Lv
znxHw<GaiOtDxa1Wb<+-Jti(2Phk(f>uAF$w^*ql_?$fK2^e+16Y^0RDH;i)p^csD;
z5CIzGT6>hZ0S}FZta$s_WJd@%u#~dR=RP!~In^e!DELfon%El@8Zb=nIe{Q6TvJqA
zvnJ`AGupQU8+pcFVOfO~)>r)5SbUnuI-^XMxTyYvwDhb$_NCvMA8?Tc=oZZs-Jme&
zaw`wkKNl&H1FHzY;xP~=5!y800Hd+SNJK;_@o?8_qv>s>Sv$&FP+AA~Nb5kH<>Eg4
zLJmDrNGH|myE?T}8}jF?0<cCd2KZa@hEX1J$aKN_C|QhCH$Oor!H96^>9i_bG!~2{
z{Km+mwZA1)rl4bOm4+$)bs>eFMPXe{44cKf5vt+lv-CwZv1#_V_+3Y*u(7@Duht!9
zBq977;(vt3DiRq^<SlFDa#@(Vc;sBKR}i*~r3^IBEi%K$*9};aqq7ib5AU@8I}eSY
z2k(g6A(0wZ9G*?i2m5r!D_Lo^Ks($F5N3eO9JfbBpNTtN#-iqeTh8eegGsk=ahdfV
z#^_H4x0D`x2RZs8?gTa=%=1$dXIc_Bo)w(hw@7jJww{lspG!};%q#$tR|eTLBWBrY
z-jSKJG(A0_d?<6CAz>;)e<UxsiUrJ8kzrh0<b1#0VYfZ%%+lTQ=<RO89T6TJMLk}5
zz(GhOx~(W{zhL|?yGeN`A^3H|GnCy2#xyboR|TQX_$Oa)ZiE%_<M;Og5&QIU)2sgL
zEgbA$dh@#>JVs04P_SSVOMxgZX(V7VWn#Q!eg;Gg9EF0uM9I9=`QU5KuAWrK0H<OR
zm79LPWBw<Vjqah3FiW7KFv$n)k`smS+6tHzE0d_o(!o$mN#sRS92pBci{J%O=ZaZ_
z<v?}cwf6oCOBS5d_=8sGQ+Do?<WR=bptMn=;rw+eI1RLm(B>RI6&m9vHvD6+m$mMp
zO=j}6?u0etOs!Zg&iU$$rcG}OOvkzb?^@fYxH)mgQ+a{8XzyWs52b2n9a?L~y3SM=
z7C8>TFh|+$doX=`%5M;1G4xIq8d=pbZ9l|*<|K-s$2KdUNzjohQ_#r1r`<FSKQfi#
zL9}v)mY>=vy0>hqm|gl2*>gh2AlLX1$w<$uYa-etx<`E$>wJj%y*Hu>3BNZc=)4@(
z+b+gE_|_?)K5K@guqb=nB5_g=5f=4BdI$5%4=4A8hnUYluc#%mfuhU9*^5MbJtOUm
zb*9<qBOf_pg^P_#Y}nR%$Tj4@NE)Q77IJ92v#u~XV?Eh+qNdvu1$O*0OfjJgiLv^S
zgQX>X#C}X)<07taPmPHtTZE}0B;^lkdFQ~@4-yAtT@QiLn0>Y86TQrY{j|x*i}%L-
z;#Z2=D^>lzKh9*B8kihDqzWSMmcN-dj{DGTNyYY}is2^9eqK#8c!o3*dBd?tud8uZ
z>CKFy<Z?a_ulHstu%mx#9k3RnVO&s!u+SrAPw2jTUh>h9DE5b^G6*>>I31|uF!S@|
zlV?P0_RFlOn(xNXiQjHUEU3jA5=~K4^7htaWr>V6viDlwcjV3nPvYV>`<0nIqY2)|
zb-Z=w8>L?T?T1>_iM0P>0Tjmwdc~XJ+b0dj>P2&kyiCsMepSyfPOc+oIjy4y=_~y>
zJSq|jCi_^QFCXV7cIm79<SqyrSS6PE{N6!yE?F+)#z(c%pkg4P5&a!vza==xPJ7nl
zSi75@pyTVg`u0JdadXe3n1(<%!3Pp{X2T!v_Zq#{k<kejI;CV+$)AK&OirASKD@v~
zO*}G-cf<{sOUA4Dc0tdh=yy*?bm}e*FU5=LXCD1)?e>BckB#i^R1b1rgtEUJcxzY9
zl=8qnX26MHbuo7+)uy}!h2mNG2Ewy=66Kh1go{nPjg9<P-78`yKf;K5<)&_129EB%
z&*CW3+dwiIICp*IVJd2;CxLRXUeURyUglPRWE#QmpCh9K^F@LkBP>5?zNOyLM~UQd
z>YO$Zda^4)ly%3>X;>|}Hic;+J;;;b+hJ^VFV{srMUv1Z2n~z8eDSZN7vCpTJVTb$
zEf<oSoCb_5?Z3~6Y5Hw(jr!^)7dJBX;oO74)^#UObT*KoK`V^d&s{^sshreMM8vA#
z5;5i?%RO&j5!@SwT4o1T@~#$_OYIUk9+>-3TTQs0kligWTu?2qB$)Ydxx{c=?9mRj
z6&@#!3=3&rsIXk$yrRaJM2o!-4H=s+WPZL#@ZPMa*<#Z7ncR8d8a~GVg!%gE4%VA-
z6_R_iYK`w~vg@q%BRQhIes9&o>XBvnWG_C&P|eq5;>6Th^zBnyQ?|<L(;qjQ-{yZE
zR`Yr}Fv2o)){-*`ad$(a$*Kmvm3Lv$Q|P97`G;2c9bS-fZZBRpx!^M2mv5Kv@nm*k
z3<IM{2+LNNqq^9L#Zv4YGEVY_0ixc{!oAy<?1al|J(HQ@wL)rGWMO><##LZV53n9y
zS%Ss19fjx;2tq+<vNm(k`9st_vj_K@s1Nq$Xf5k{qbT{YA8e{s8i)qKT*GTtZIcbn
zIQd7D%Uxemy>WU!Z!<o4AgX6{8P<Q>P@<}$C~uD@hjy-DlxlxENzZ9LK>hc9_(U<0
z@S6cJ{QMOgobD(K(7rumXyL|+^1k?n`<2{W-A_e>c>))h*f<qrKht05SJ20C6}1x=
zqyZhD?+mkm!r<|?ZTbh|>$uMqqL062^r=QZyUhVK2yK!pJgM}?57dIUp|8sg`>oeP
zo|a7z;}?+NroPskm&((gmvQhn;JCHJK67NXTO$z+zIYAmjSXl1;TX-v(&Vf-y?03L
zPVCz4)O0i|?D6du4wnq5|JL%+<%A4`yisG6)QyF!>C4VqW&WrE$BzzT3HIm4-;$oh
zS=adSzs?Xp7KihwNz*`7WGS(8X`hLU`W;8CyNMH$G!Yu&@&cs>rU)Gw5upyAccdh7
ziaeF~f4yG_UZpUzdCt#(d@3ZL=-xk*x9c3OV7waeto=9R(TnJ!=&XWmO6E~ILvP6M
zpB<%@y;MS2+I7o^o`KOoIDA)%9Q|+`q<!C5@R{H>9k+qE`eIfzmhvcG@CC7r?-53?
zc(+Yhxic!bj;{a@vUDm)|Huekb}ddP@_Fe`Nu39KPwVcm%&Tr6WxOc!^A5?waKy`n
zrwu{}KxdcA!QhiR<!zwWPm$H40S37dn1LKOY^NP<FSk&MmA%uE{D|bSc~%y@?ebaO
zMeqB1_tstJi>~uqb8@5w7Z)CedfuSE+xmv`iBkMLL4p(#hgq@nT`mJx%*wmmOQ%<t
zFvNM|!03BzL51)6lzoB5W}dG_9Y>L`8EHXwt2##7ns0Z-wr<Zn2!~1c8~3|&M-hxC
zAq;})DQ(&Jz=95hN5q9`o=5eOj(UWOfv$D-)?3a@qoD*14_pYii0gm=?x@?)ysy&w
zU(%X4{YT0>a!)Gz2qg36Q?DM{8G5Av6&FG>cG)vfTm6|*ZGkqH=ai1E%upd=)k>m*
zn0;51H-tP@O)7jzp?@@*u$YuhoJknq@|?c+c6vb?1mG4?MdZoovIq2!>L?;ON{mEP
zVEDt7As?kr%@~RJV^X1vMqf%d8A!FDVGPb{;Du!gc>#x&FW=7(Hrjh(MZ4HN6K*O3
zv-(nSah%Es6|th?yKkOnFeK+PUw5;)%LBJw`?0JLdt4sq^p`v6L*QPeYh_>w>x4}m
z{dmcxTuMHe&Co2zB@GIHvWZxJyC?jHYrt=+PO%ZU&kigqEjtuF>wEB#+G%>^=v$Ul
z$EuQO%5#QQO%k4{Vud!@+f|`t4Az?rj{Vcxxes{b0|oFse?|P7aK*n`u(Wr7-(`Jm
zcJph7S?U&cRYf5=#deXx2?ONrH?hKQKCmkJu#nd%X{OLS<5@gc(x}czjeyeAc+!ty
zfd-P#73`hOMe@UH3UiS{neQI5qD<>EwT>M`$loV@j%k|7OBP&XNBvwowImV!B;Hij
zYr@D%qeemk&D%nL9SMt)+UPZk9#f^@gME;tJPaf+k6%*Hp&XES{+Yx%y?@2_OSr{U
zal`a|+KouEn`pOgUH7v+=S9nTW;J{XayrgV=t04+We?#R&SNBL3<#rXsm@4LGL%Y3
z(R%uX!Wk)zAH1QS1N*&Vu^MV!oA15zaB6Z&U4M_FW@V%k-{D>YMVKhLdBK6k@9YP=
zur@DO!pxUGf(9`-toL|cw{Qs3FBX{nS?TcP8O$P6eUL*1f4B>XS14JFExc!Ux+UME
zUIzvQaN%`neD@8-r!ZD|Q8``=U}UkuUc2`d+YM8CURe8x<SY><e1Pb(_y_xO_`No?
zA2zI0NEcN~w=QkWyR=Fq=3yZlu)*z64esft-K3}cs4b3fOaOUijF<EJsbl}Q_f`VE
zfl~jEvbTzgtLfTyg9Uf@rg696?yd>JB{;#|-Q6u%AP^w9yQXmq?hrh<yX}?ld*%7}
z{!jKnAM_YKYK^sOb=9mn=Y8F-*g2XdScLY2%xiUn;fmkl6Z~QjPh#o1+3*%Hh>S*f
z;cGKsaTA;{91>z<Fb4uvb2-dvY47gH@E1f|`ypflA>E-yqB*c?Q28<JWC`6E{5xEJ
zU1~K!`l^R&A{6|pXt8tm_j`tyWF&fyeYj-E{lZ2c*S1nW<Zd}n^k|7;<%@0!T?XCv
zob#%HZj^=ZsS+zLAC!F|2qB|!h#wuKHQv7bRwn0LCm8sd@-D(T$QjGk$Ri_g39Es4
zz?gg{BfG0-A&im3<X5w!iOQL1w^0o3xj-f|2D3N4BhEIStMOB=c&l?oiR?FWy(UX*
zwn46?o@}jOuqZK5418?nJ|*}zs><TSj_Wc<>f`XWzr??v=e7)SVTnHjO0Y93#>+3(
z=Bg8paX6d9W3Gv<o}#_?k@$${imFD*rlFDy+5RNShcsKnC`=ca-WWYkyH2*|ej3v-
z?0&&_`T>9XEKdQ~>=a|<IgDmOg5N9K6o`*l@fWotESK|sl|!UEtG!&`R80)3{n4Mf
z*$pYG`S3j>-b4;AaOAZ#vNipoou@_cj9Dj?)r`>K{NFSD-=loV8(5e3)sPH4#z?zO
z?A3az)<^{xY>jez^$Mz1F^S3)*`f45Z`-vpf4uZ5B-2obuluUhUi$OjbxF_bWx06m
z7fCHusO9<QW<Q)!T4h--C*H1PwF4S2>PCfaWvAb-C!>pde;!X%!W46lZWE{I_PLx)
zYuAr+ra96@9`(St<B}}Na&5GEid}mi;07!+)$7NoY6X5YFX7crh447JbWLe0^}C{D
z%=0uo5f6vjW!L#fr}nkygR;mEUPB7x%JOlox1BDpt5vV|Q$TanwC{-5<^6;RE&i1+
zgVoGAk>S8Ox~sxkHa{1k;JHM+0zn(MieTno-@A&FvC`?Jtq<HbuM^#3L+PC6cgIz}
z7jjb<J>gTf^TMy<7h|lP)4qAHM#fH(?^xqBv=5MP2J-x=(}gb;TSW$%U$1$5{3Qq<
z{4Pf%l7#GzwnvUx6bWE%H1M`eCzN^aPfH&wtn$KEv0B~~Z%O9um}fn`^LCpuAPDI^
zayxdF>vG(h@+HU%5HhnP<e<8LnKzt1c6wGCKY7+_6)__t=HgF8;$Xf!cVDM1zVhMy
z<g@5EZW2#R{)a!%H=o@ScWmgA>@+!;%Z$b6VBaNX;q?a;O!6+eU<OH{Cq33XtNdFZ
z@|W&Sp0SE3ZstvF%ny-+`2NS654bOl3&zos$kUQkZjs@$LCl!v8@H1MLX*Ao={sLa
z`#N6@hPbgX-}AZB#aR`=>3T6SD`PXKU@rwnO3KwEs5C`?kl!{>1xyQVw7~OXUHXM$
z1#pHsU3lLj@v8PaXDDaGPvPh7mgu1wS6TXUMci}(*N%X`lt{duPL%nZg)nsTliT?q
zzC<MEwUmHx1nRClyywWl%Isrm--IcK$<5hQS}-kCmyvoeHCXhiajgT$>(2CkS2rZ&
z>1T^69&D3DewN%82VbybA|}R>n(<c3!Wb6IHoEvoPEn~J6%8bLn}9h+=0{EoW^DJl
z59W-Q?P1(oEWa=J68jPFIQ?1f*9ztP=~l`x%{^4ak?6&Fv^x@Qdf{5tY<w&!`57!e
zV{f4dEqx_*GP&=!<4LJbhuO0k-{T22rw&a$Tc&Zl4#)JfETu3~c2|;0#uM#*W$ow$
z8-C^<y8(%&scY{&_V6$6(mz<3RAefNmyyO~507}5A(0;nJ6T7I96Cea23nQKKt9!6
ziZgUm#Xt>UMv}QHNmg%XyW8MT=GYalj<R<aqm@!j+u~lo6~y5H359*ofAOg*dARNd
zVFso|WGHUCWpSQd#orHK#|d-97ycRJdv_^Un7WSE$?nGor*J3wnsGS83eC8u(=CXR
zpDL@dw})`N@$~!Y!xUu~x(TC(oAPmereo?b02rZv#-iVP@*D02>OT&`>bD<OH6tRf
z=J)UTi^Jg~$*}I8Eom=VJN>pvs3zbmOE;RtAVvl72m3-i^L_7?`KTu)q9oBBr#gxj
zE^k*!;(hO#+5eBDP(ZP-##(dabh5oa<p3&-zj_Lly7Bgk-n5z>RrBOm6l5&CT!k;L
z-j@}onu(r)HkNJ!uSvo>HmvgAPNNGc3y8;`8K(}mPP<rZv{%=>1gH~wg=ug9WYR%}
ztnY<Jf`3S&h1t^9R75DMxAUuPhS*}L39b3j3IVDAV%lPn3(MBi7chJt(W60Y4)JE2
zbo-Zg_ot{<nPm^?vN(@p>4a<>&)d~)C(2D?XYCi01N$#U9f1k$a+h2QA7gNRcQRns
z{YAo4I(+|Br^GzD?T&6gMIZ<iD$H+7VFvbuDd#4UIIlljWq~TBrI4SOx#qK0986lD
zUuW_r{2%y^PUI>wd~Cg@$09!j?7OX9kqU70{TXMmBE^rm`vOja>OCzeek_7rqUOH^
zY1Z)6S4SXL-}pY#BsXN3<=v><C>?(GdxZhR>|rt``0Yi7AaB9557g{Us&(qDuVJ^{
z%8ncX`SYQX<QHu^(KG#<(Ym2ZFiI)>X2MGV=l2XxL}wb3t$=ZNOB9hWnk_G{;ESzM
zgO2HAjP1uYw>DY>8Ow5!*&v3sJg>J>5DgtQxQ_hQ1bwF*1$>hYuYoeXQmj5@`A2;6
zG^sQkKR*$AKc|SG-LWiMb-laLoE6CCDweRsJGHK$LRxj`9h*I8(4JPd0%4M~4?<x-
z_;Dc*o<%)!`>0AWIWmRAPhi_L4JBC5xa*h7;15wOH8nX18RY8+53oquwJAjxW=)`&
z9_us8C731Mmo-(THE-qvUr1UCJ^c`(7|j012(w5bwit5xa|8Kd*dQnpEhYuH?PW$+
z4f%KSQt5OrHeRxLYs8i{<HA6?R(?nnDMji?%^qY!L0o$$2KzQC(~}lg%}Y;qSd<~q
zI!!V7DRpYF2^q|HVyz=bXrIEyNZr^Cy{T|gjJQQ9)k#aps{(zGyNkXIYGn8+PkGaO
zzHOd`<@OtPD4FY_w*!$N$nho{VW$Y9;F#vbvzHx)f@T8J{8O#)fRu>O7KjrSJ3`4@
zo5R1&L1J(4`}Rr&3F7j?Can%)ap=g0-X)$#U1Gid;{OM;F;sI{RwK(8L~RJZ)>}AY
zH^eZGzbx~<J?Zq5Wu?!WZrvAwRN;*A&2x0_dC9k30zrbuh4_#sLXTtfNk~nKu5B+D
zN5;(8NUT9mlEMU3!{Vy<*UKakZU%hw$n4|VmG4Q?;i=SpclAVCzMnhf91CV6E=<W(
zGOM3ZJ0-FLS)BgAiT(f0?;}YX5f1MUGz}9)zH@*BuO-?!IXwh2!kobA2!3Bb{H`H3
zBXZ4RsB`7fM7i4jeYxy`7>i(q5`SoKJBcvH!}J)pt2`fzg9&MxzsEK4amd8UJEnLU
z(+RoN<FrHGMfYj_OmoHLr}OQil;f`@!1;U|B?lKd&cl&_MPeXn@}r$4WP%WIFK=sj
zwUA=J<(|{RHerL5c{?3>S7xoaM_LDOd|#GchLNV+*<F-6ey$$8#+;u6hSRP5vV3=O
z%UB~}-$L=0XUrGJvi+J&X{Bo$?PFTUJFD6@&F6;#y0`^%Crv9x{E{3Vt@T7n7@isY
z9r=b#y#bx!W9M&R9y(*3o5|+4Jd43Q(yomwE)oR@iN%#|OCe=wVarb5eM`#<cHUI6
ze|Q}dyrn9GddAb159tLM16t2|3)7MnE)i~X^jg^UzARh$h_|=0`h9E%w-Cf7mJvuB
z$GXXvPa3y{RlbJYt_BmCtT;QMfBx3|er%^w&ZqYkh2YO>FQNB%L!%QRm=Jz3Q^wlI
zt4m-?Zf7@c&K;r*!M)LF`Shw{Cy<BVf>g=el%tR`*;1qD6(mHRUTQ~apk#v0z2bvB
zcKN-s^C4D<HofeT3*I5Jw=tbq-Ft%st15xw?qds0xJOX)mVxWc<Vp9#r5;@ouS?Ug
zK4X0#e_Cqv3|wpOZ@b5pJ4F7M2p2Ssk1;}V#Kg}O83K-d4q4Ht^Zc9dF@xis!5Mf-
zYeE>B^u{Dsn0q{t)#r|#1}TgSQBH(Z5khupJqS)@#y+H#RFcirVa-Y@i5!qN56mtV
z$<PTSIirV#F9+>lJVF-Z=B|xgLKMxGKp1!rd?{m@WBU20=2bA4kDFl<rcv|$##ER!
z%f}=}mJ|1x*k#h8x3xpa-LdOvD?i!vcpXEdP5SOpGyQdbd94)5o}bwLE|qF#EedUq
z-;r50H5sVjtLJ>LG(NC^t7-bp-LSucnS15_j+F8qN?O{ZKllc#7(-J4vwqE&wRw`m
zEhRs%d4S!*!T+ZN{agIa{m5>{G~9FtT!<;9Fsas5?$Id~OUIE)65NW?&zl9@^bCK;
z80I+UB<&N3@0GXE{B0`x0BWr&WhI$&LSte;pode8{Tu<~=In;Spx{8^)P=Ky-->_T
zwS~ymNo-mEAo6L{;5`N#qM73(O&r-}oQG3$q;p3MYLk3_7)P_prdJNLglB(;RP-+Y
znqa!^!ysRntE!-wUVY1M)b_<*^SbI4`usGI8=3U)T_xgqSm<Y%fj+sZ&%5DR$D6T4
zw(G~&cS&?drGbK*=NNad<omhJVt^S&EHKYRPx{=xy%ch2Gv^Svs{^;z@VNqXJEy!R
zl;JcZ#w5S#%hBZc+@I5=YJ&DVZNinn#BZ~9nCq=#izf9rc)p5F1~z{D)N0K))mnZI
zm&e=YY@4JI4b!C)9cZ2}z~)QG+qorZ(yMH1#;~XwtB1LDjJYdmp24=SVxs&2{k*Z#
zJLZ;g(l<B%tWdou|L#U|op!afLg$_hb5v|x=$~g4LD(OHHTYjQq@&8iAyO3mak&H!
zN-HT?hm`BF{tNDMUepp6nu*fbk~VP5n#fg*sU#pOS%h`ga!rJM7M%t3d2vnxL<Suz
zG56YtA|L6GA6ndYkA9*$ARkW~)GU)}azxu5(4KEsd+B5+eK0|1u4ixQYYmrbg~hVz
z0vZWy8(66bcpy#t(pU7WS_(y5BiQ6su@Nn)u9pfIiMmo8C?1y;E|0l7+x3%<^Z5Cq
zz0t(Db+$%o+2EyF`E%rnIWXr+;iLTJ6XlvN?eXIeJmnsps5{WOk<}LGN7#8%iLlm4
zDM?D6m%XSA0l8<SIc<;YX>gd70G=^Y(x>!idw<V}tin)UX~7>iIuT^;8osZY{ye{E
z`?`_X57FLdjep2H_*`!EYCrk%9GR%`K?G5o<yXNb`X`TKym~5@{8#2sXU0`X)%QO2
z3HLP(ytVo15ZvpOna$qkTj#b2OXjhN>v_$)@hNC%7=PnuTY^5X)|1el6DekSXg=ri
z{#P8c06qO^U58F_s#)H|lAiGo<@8tuu-S4EQxcMvAa{zkN}T<LW;*R57_j{#!bGN8
zcK>s|YOtcd3NlwI)T%XD4}?)ogG&6+97N1uFxksyoR0Y@$70Bb&ILLF{RY9ya|C77
zEo?3!oCiC`GO^acw?KPRnSG3ekPSgaGz@SN1+mhgjeu_~@oQc!X+yXQ->~;Yq=%(S
z9*HSp%0!2<LajuSYkt8DFVF91r81<euxUr#OuREoD2EV-V044vKsOQk8ficRLz3dW
z(PQijPBNKDX{?Vr-C#767+gQbThk!wMh#GX#5G)4D3<qc6tQmOJS{zKCzj>LJ7+*9
z=14*A*IZovWveFnpjHn;2FrFCrsa>J1-WO4O$aRsXg4lriy$%f@0<UO@VOtOwk&sW
zHbXD&@#E6t1s0%6GA=Tc1Nd6(1~HuML6NBG%n|~;w-F|Ul_Ck|5E;lY_$I<$uAQ5S
zyk;Qy%^({an1;xa9v{;E0!YX&kD(3wLk>~rq2p>g{S<gL8GAuF9^DkzSt244%Y>Eq
zj?dX;qs~dQ4r|WBfn1GWA!M!&9bj{cLczBi<F0ROF{XH|cc^bbG%YWeM!GD`QEe@m
zUBV&G*$)(JLtZ1kaX3Y}<zr+#KIU<lZJeXgXTF`92!Q*<Hhl;wI-c>?%C;D^Tr|Nl
z(4;#v@Dek!;Qfyjf)T8w6O-C^;=z!NcNE;oc&LcypVRQXA3AQa^ss`xKbf>7t?@mj
zN-aHOMor)h_<gg5L{ie|$w`;DER3c3Izah;bi8%QkI`3Oq)DR9q;JBB;q`KS64cX=
z49AOE_zY22U~hft8j^4_M3asYH@@#%-n#RYZ%xQPx2$q`>~V@Dy?YU<`6(`RZ$OVf
zAYxjc>vx*h>~w##Az=I%dgoeyXxVp5EWAVBG`YS*<Q%$*9L9a=)v(z!sabwIs5S)U
zG=#Q1aWpI7Y5$P!NBrv3-ZsdLH$N}4Mw;2$8FWv$M0+%GsNF%Ppu@;Q%F^c3ztdK=
zk{M|jF*e6&*K{rn@p^JIcf957l9=ogp)ND?5!VyE(R^#vSNTXQjAx>#*sjWB)}wZy
zY7en6)$4WBOGsravOH{0KjgA7);yf%`W|D;wyPBpC@VD6)4ojrksWQ_xWVdI*9%+4
zA7-2vU_MD|bv^9u`0lUsVJEhQmg84AI3bBit=g6R6(wC*&#~c`vSj;CeU>%tC)5&M
zSc3v7r*SuWyY^+!vLkchOSzgzrCG|*94oc9dYw6Hmr{sQoWQ4&a9+QjW9)PZ%*0Rc
zBP*N1it36Egp80e?uz;Vp611<>@&v$Ni1p-Ry*4B8{^xVsUI==j#7W9A^F1=U{5UL
zzCcP7vVqZ`4GA6*>J#^l?Z1PN+n2ddCR6a$Qn1l6@H?lBk0Kq;Pii557*$r7m&OD!
zw7253><u4kdXX+Sw<2+vo2;4a3Y#gXS;ZZG-A=_E_pB!PrrZ@Sut(O=IMHHKd>LhK
z^-KL|(P<mnyt&Ce)FLh5U^qJOhG|q<R*Nv&ZpJI@tYaEe75yN2R(zR#=lTi7L19dX
zL{|x7;O2aQB~J=r$u>;>!lqV>;4dPrD>dbPYqG=w=CBoDsRPsP-LmmLMxH4i_WA1n
zSj|4(fZ2OfbJ88R5CfC}9rg7wH1!VhMdZ~tw>|(4+XH+|TxQ@4Hj?o!q$JaWnbJ;@
z`4O9#j%KzE1P38>S1e>sV(<7CDMi&`hYoxw0?iuHQz?Q60?+1+&6^yq729joRq`xv
zN8`{h|7n~G6Wsv^cv(eq;^{jvpYz}{xkM;5ydz-mYh_^3195tUl2BUp7S!jSwkJ!-
zPgW{C2z{f%zf2M~U6#HGRA%kSmN=?b2@ebwgqrn#3JJM$%Flab<R)T&3c|EFEu_04
zsyQ=OxA?P7k9fY)`vNK2YBS2gjgIrM`6S@Dx2feCA9V(v(^**iK7rkr#1hh4Th$cV
z4%(1h!0)|mh!+{c8@XqK`{cCa{5<Wua4X(d7rlj@WCUZ(Ff+nvO5=D`ZJiPUrYV&T
z5W9#pg7i>sNG*`7GsS7rHqCASs&p?}mf$mxGjK}ow_ZXQ2|IW?l%-arpqJH4Yr&vx
z8eGi(6I^>p15rnL6thGd#wex~d1SI8q^|sG%dUnpiCjv05Fi^jfB_vRN%TpeDxqgO
zI^x#^?x3I@%(7f|&n>|vv+@zN81cX!<H0a=muL(|Je~4XLPd%c83gt`wO3qa_?x6Q
z#dLvsXb~((Y{c`f$6YI;rl}dr8$!MOmF^=6vLoI}kB0*`U_<Bmb_bdaayiUW@bl&J
z5-Cu8le)AOD~T6F1gRis;Ufi$z8Qu$661!PKz~NF6@@TyB=gOxyP*mJW?cfWegcxZ
zXwmtb5rrenpOs5amq61<U}`u)g^e+~_?PCG1VRcqQSB{R=v0gmHM`Y5St_h}jCeJc
z63?NX+5-!3DNOI-uV1j-R=Q*p^)@#sl&Cx|_k^}8t^U|Lf^IVM)FIR(Q6yAiDrrWP
z=6cNrn6GueYM3x%3Q!$8FuJLfhAi}Wb~ZMayMMxw5y6%r7{|93epTWXp^%X<^fY5h
zQlz8VBWx>uAT{>}B-DFgc9x=;YqOq{`e|i!B^5)88<0jRItM#BR5~8Z0P%w)ptYxF
zmlfX32*=h#ZjnxtR!!Tg{y6jBL*lTgQ6$yPA5QI2tp#Q!2wBe)t43`=M!D!L4z{S-
zXgJb3N{wVeM6(ixB2_5e&|r!}|B6Q*9(XfIJoc!;(J5g%1=d6{F)BQ3t#XMh*@9{x
z)FV;-Xy8vtOAmj*hGVrb0dA|MBBD|Q^{6%`dh*y2`_g|Xb&chi?SJGDrcki9G`zXC
zhAW6mR;(5oplXp|>&qHZD>-%^sX+X}g%YaQCL0UyW@;Z>A)swdS&Ic!irdkXl=#(y
z3JiitV0kz;ck9H!u*R-nQ>*N+RO`Y%^O}X)*<WFTm}Bf0!X@b(xwn}nGc}$PB`IUr
zUE3Lr*=XF=o`i(GT%aSPP*t5$G6c;9B6-B;OlJ})?h&^FTTj28zEEWQP~)DV!>x=9
zoF1i0I+ZciNaNKd=rJ|kQBL+X7ohxPf~dk{LP#&rW-+3djG_f<itkE=gQ=du)^;%x
zfAb-ls%4l@uVr-Jwtw6`;E+=y1OK3`?|J*mK;A*<Y$@`k6q`SC*9`QMgrW?D>6vcC
z8LL;i`9Yz2_68ZIi`pZL9}SVRj`1V!Vvtf3Hs_BbeawWAp&x^;y1%Ak1$}j6g7Aie
z6hOSSuvpqwLj+1`>Pe`hRT_XFOi;Y2cmZ$po_C{dJt5PbL?E7x8sN9|_EHt9$e!yH
z+EA`a=3sV#e|^qqeZmjiPot3zapGR7DYN+P&5aH|IKl1c9|1~1NxX@p%8!MFeGf5y
zw?DrqaL)IRXP_g5#-R}2FDc>awHWT{g^P#ANT^DQSD&*Lyi3a9zR1E%j?J&UOLa0e
z+r4utyBUaRau$+`9C7PU4<|SONA>g`fvE;U#(=sKfkWPJ(Ocm|hyRSJib9G_KMA6&
z#7&gMo4aQpWRm4x@eWq7@I1^cl){c`%oqhfT9#KwA7;}81u2#}dUDJ$kbTjNb*S?i
zNyz709y*cl*Rsjh6CPJ}8}oS8E%XS2{U#zZ@(fqNvZmaqRl6#F^p!4n25Gs64QL|B
zh?1k7F-<U?LEaB1wvr38X0Szdy^L~Vr^#NxcUN68+DE=g#F4kS^<!F%=kBZQA;IP{
zDIb&&yWJ}Hj6R<i8PeDiYs^=bT$KH;9T&fBM@b(fzMRrbbuXP!p!@-hRW(Ri=dj=x
zaxlm+v-W&dx9ykseXA>bZ~Wn|WRmX(x1JltsPWIxj83$qA1=Guo~G0O@Ld_-8n-y*
zTf)~pc?b*2=YD&HuIw8-X`Y`Yofcw1;Zv{4j>oRJSUGFb<NM`4#~)(<82v3idJ|m~
z@I<8^wEESgpD-l;bWVsG-@>wdD`xlzU3rVk;50%Dv6-_t_gDK4iLu>ieE!1w3&g6o
zUMhQ^tfU+5fm(8jaUa?C*P!o@r)n&Td_0YPISawRdQ)Tk00p%I_-4=}QkU`ZcYm}K
zHF?V$G`d0uV#Xmj?*0S7zDIdgbclqAHH9&2fzp{y63(tjEVfjVzBG%zvWI<AgRW-a
zlq$r?#&jK}5E=4m@H1ihhR}grEzOoO=~)HPU}@8z;1I(opP%t{yph=_>?&-=oE(95
zCM&_Lc{Qo!J>r{oVis6LyJ{c!Pb2n9p5+A*LO2FJ`sv&jbz0%vV%4ykdSpi8zu|R1
zK*%y;r%vh6HLP^4*0AcyYa@j>-&R1t`S-|INc02O@m>_pEz;YZyFEMUh9@KJ^NxpT
ze}iq`pR0UXYIib##yp3+KK9Uw?fdc!V3JwM6V<bSTZ|@acj9XD4F#5Tp0Y3J=C&tg
zr!_wW-*NIv=<%#Nz!`tX!s@ME!o&=s8lfS~V!SnU{+orkHjya9f1ZMUVAC?T30os&
zm3WARJ?`%J@=oh~5)~zQJoepIMd>rn423P7pL}(`Ms;F}ZJxZpsS}Cf5K7v)kTlB`
zYEZkcON@QIV9X2I5`xR^=T@r?o<oA|SPr|s)2gPY$!ybIJn8)CaABlN|I=7tcJ^$&
z1?pIeimwmGMIr@5?EN~vp|%{EvPOVyZAnr`Dg?KQQ4TVV^nv{74Yi`>x{FFz4lGu1
z>Til8LkY1OmC2gH?j=J+Mz;^YTv~Kn-Ep;ub;Tm_lxJ@Yab&%}HlBXBi8f@UyDo{u
zes&8gX<E{ZWKWUZ46f>Obf<Ih?hb)Iq^EnOU*dkKn^-MIGWUzavN<X@nxxLui&?Fc
zfA<wGfmM7ezHwNPxVFQOZwudjT7k9IF;CHTkA6tf8SV<7eqU0$6;q1P)|#P5#@Z~x
zwloBbfY9f)7($esN{CJg;~)T6X#c?;=Jj5UTH8Z%i0-Oz-G5CxGZbA@^tku~dEiL=
zd+4}Md8+e|Y6<qWtBo3s%i8iu=PA{K28i#vb0o3y<8&IV+mLZTJpZcLGb7Gs8LcF}
zYJZfK<7ZKqUI|WV77Hnuuu)KtHQM`}x8=UlGyCg)5W)R@P|`q%40bB5noJI^wLoHy
zmF_~-^QZ7uKL4lvTvhWK6I>kMS7z?p0i;FGh5(mzdhhLl;p)5g2bZ*3{H_X<uF4n7
znnUzeWkRVe`}tqwx4LkeUE-0f;s9*BMb8zoVP_AAHn970pKe|C`y$`3Q^3G^n~K_t
zf<1i3@5DS-X7KyS3qIqgK2g&6qfUGV>~kJYe^>X;Cnx?>ckDG=E*Xh1lLID5a~h_s
zp=2t$p#PfZ9$()yV<Y2bP6#v@%7@LD!x+oM5ZQ+Ox85*@OfO0c$-~!O({>0#yNalL
zt!LCbC!L74@AA7?<>lqgIuMP|7buAwafPs8T3;*_Ogz3w5_$PtF>#C*8~()nPDZe>
z51tHG6x&{KB9IG4rXr?5L?B2p1y^RL7{<lqYh`DqO{gP27~PomQI`m5tH=$ir^%h~
z9#Y?n$qjLq)J1amCEo=q{-A=Ta14u(E)MNbK@r1(g<c9*Z2#s;Dx~KO&m<cl`Y4VC
z)t@7Y3_G9hG-!dI_CRC%BZA)@;J)(Lz8ag39L<y_m%X=8+Tf`on1@ardTC09kbnuF
z!fKN&A{NdTKTG-Dd5#gt=W|z0alA~&B^;-4bELsO?GYtR@Uk|egV@6&c;-92-KpPc
z|9rlhj(-uvklWw{N5ftdDoEMfQNdH)y(#mk+%6M&iQA_6d8IqR6&`|keQ3XvOaw98
zEN_m{=-ehuv0zE64q8XVY^}7I$nAR=Y>`jXh#MdQ1>0G&?s;2jJ8o?g{#6XMZG;hT
znm?3xUM33$hhWBtRK?&9jmmRW+L}6j9F<t-Ni-6!fxWdzrd7Gn$}K<{_R>;Q$Mx)j
zNAC|j@0MJ;AFUcqUXToWbUuJtMSeHeA{CU!7N3D{B#UQvLrbVcAxAC#49@m-0@s=(
zYkYZ(Fw&bYiMD1LQjz@ect4?MMwVojc7aOtPPU5csBG+Hgthpwd{=M2`J-c8QVkFH
zgOoT%CFAHbAb@xr9ZuQspi<h@!HThln)mc$y)zi@tEN9?+!=bOxQSg!;iZ}~E>OLv
zGj<*J67l<TzE>6Vamp}>Iq`!8!&bKUrRul=>c7+nP!eCC5I)Em$7jLVrQy`|t!p{J
z;H>jHn)E;?WHH)}Hm>_U;t=8$m_AZU5F_g}j^t3U8fdKh0OG}y6U?gCPini`vc;Dv
z^zSh6cJ{Mo)K0`Ov5g`$rhCgmYm*u^`=`ZGIsUL<%KO&8L6*dO<ev4=qvME1X&~5H
z`*Ks-rlM_AESY)Ejg2@jeLP{Vs!@P^TYZrIEz919%Pcn}A^Gz}2kk{rdF6%&hPP`&
zb8dyyb{rVTeQL~<+|2VymM5Jc^I$u72ZwF-7;scOxzcxcy=5rd{#p>zl(P5%dF@y#
zDOR0W1he!fDx*IE^6i%&CSh@_>W~>Luv(c6RV^#dv$yf6*1Z1n_AZ&p8rct%Tso)Y
zWO6?j^!<hUicq^R6AQMQO-W1fL&vj1W!#fZ84K?VkZ7f{$lV`MGQH#QhGlG~LAK!&
zX=T29NHFD<gSjFuXf+5nv=<H%%r5qL6i=PiKP{)f=oAJo8lX1bgk){*wT$aT_^5{v
z?+ikF*K%w=UoG3*2JEiUf(cZx@)4jK3)m9Hr}S6-1BgpeqI<ZpUTihYl`uWjuUbCo
zys7$Tq7UM*kx9%R-Lig*CNIZxV?s88LD*D>XKq?pw3aoy7azS;n~isEU8a)4=Mhfo
zW4`jz%6;YcjLKuuFUV2Xf!94oY<IX#w42iXQ%J-qWgs|orIg<-`nXY3dfQV=fqvf(
zLyk}5xhHK_Sd-kjUI0T~HZPA<q(|thV*UEQPLSBq>>SuSZFega%!NMFdn14S*Z}lk
z#H5<)A1yp(-1Fn>HhkW}H^HFLYXZH8_S_==>d$mhRC|#L8oEP_tScVZ=uaxINI)=+
zgbWVLzLhOie*&XT1u%sZz(UO{aC{Uep}%VrxLtA1!2!h`HIn3odb1(WQg@&2XuV5{
zOl69!;lfozKnw<|Ih>1OhpySUn!|iDhFhJkCo7A|c6#9;DL2iVgEO`;X9oaA>TKWj
zoK}@pUT=fG9kyBiaFpewdA&?=4N|m3-$U9CgB~x=WYD%uGbd50Im({#_tT?@x_b^_
zOIRA%nmEOB%>!V$aa$4w32JHQtV71e@l35uUfOK%Tg^|oiO*O^kAiCh6<s~;jU2lP
zADMO!_xa%*8?q`TOe7sRccW<c>dM5m*wH~uAhhNaIjcQ?2`knuWp1;6-5(SxSYoXx
z^Mc{OAfRAE>(X*TZCUg8yQtR;acQ(tmWOY>szF7i>+l?zTcPY53LLWvO56Ur-FaJ6
zxwSFtc0d=01>Jg$iH!SDUwX4ngEJzivL+$%_gg-+#h&(YB3trr>q_)vf|*#m=8byV
zc|GBmu!2KaNRhb}3JfeFJQ?`e!KP?qGM~j!#HHV(nUILT)fRR}hxP1j-)avbLEeK~
z25M-u{lS;S(7fDz6^dP3Pv1#v*!_Ebu`lVfuHUf`uT9(cCwu<%;QW}Ke11<dd@<a9
z@U^C`CgQ_3$;JcJC!bzK+B;9!Mg64~?0Nhyi(KcN5D2v4F3FxW=%ia^i1w-dH^0nx
zv(mjOhQdErg*Xr@NK;CUU;7laS^I|9HQ*WI?L@vAdr3p^yak1t+18TlaNOS=Zv$DG
z2+Esrh<kzG;_vYZ3(;5+Z;GrG(9?v`R^dZ!8t<4`VILsb;uPd;j3Hh044yQU)$H6_
z3+I#MHtK_1+HO2Cj}#KjHMnc~3#<7LLXfTC2}ih4k`fN(dG|7+ui`yNTj+Nx22!jo
zeSGN-7aCXpdt|?$RHX)B%g9X1o>9X}l+j$diUn>)pL&fGq~55MYy0%|qlQC|JrbMQ
z33^UopZ~CQ>TwYndr^|Ft<oni<&-`gqUTZR0R?=Kr*~P26KHmI;e?k9IjF!myqQ}*
zp;$c=qxv)w-OMY-^Vvc!=VRJrn4FH{ZCcQoMj~RJYcdJ(=c)>q%jR6!8vA<CvuYHU
z-Ie(Ayn!{ddpn-g*n$22M>X<VZku756FX)X;ph~jluwuqWp+_JYncoey`88H)wX4D
z&!a8QdYYvV!e6+VZ7!+|)TqIx-}tS0TU=WWbP}FuHYvH6R#lykh)buTOCfG{^62W`
zz|CmOn8|aRy?1SjD)iQcy#Hgbm)nZAVrGG<6}!451)*K0^|@|n)8E1zY?6?AAZ0dI
zs;sa}=qHQe!3g{%9YTK~V)qohTDWy>5olO1yY0Q(PY8aaChk|#cW<kXrJUqN+o)d;
zXwoNyR`Ap##xTFv10LPDcv+6xYvgp|;rBI_b_>CQ)bd@KM0-AsX$PIM2S0^E7@5Zb
z;^R8r0rs|`M52#LX9kP0eKoE27C&)x39m|YSS(jqiHHDD7VejKL>e!_QNcEMN>3F(
z8*}+I`Eec|CxDBxPsRTF{1fI>6xViHxK{g9w|TC|HgZd1n;j<Kh<nU3y-p!%Q?M+1
z4rA2$-rCfPOiw^X;bm3uU?>f-{kBN@9{Dh)Y^12=S4op|NN*M#0|bc*`GD~QNs6tX
z;$_O803VZ2BVYl7oeXqTDe3Bk&?}P-t+F{Ju%6#70|wr=nZAOL+NjUas2nkHpFo&e
z5#Kya^22LHQ;4$->(2>YiB0uY$Hi3hL5M$oW-NRYL{`JwE^z@Nf=INe;GM#x+}IwN
zdE~1kw*~OpOO!zaY_hZ|wh5651@|$bTm!0#l|gvv@=+VhL121ou~6*BS-i=a^2I<h
zk-?56lpq*!b;p;dO)^}(E;$U;gf|#i|D%T+)%se9bSIGzNN7TW5(DBcftkg3HS$5P
zjKcnN5D9Ba0`dj1SBJIce(e;)gkKEPL2uYjp-)JCeX6|cm3jrS-aTu?ynBsZfqlNa
ze2f-@<)rzvl0nOKhr2g!Z&9uW3S+-COgoEj<##*!?f1-edmFNek)J?#6@7^Z4JAeG
z41WRm7*a{k@fL&3t;l`DuQJHDR-hRj2L5n{z<sVVih(g*<D9sfcHOYA`W{*xTs|ga
zBRc2k=MeWE-J}obfcf%>S1ofqczga+PS#)-8(o-PR%v`$^fUW9uM7nt`?%C_!ATz|
z160JP$$mme6Is3c^5dOJ+|}u8wg}nIozO0SszypEKQCG{*X*g2%@f*;ny%_EG2w}L
z$1w{u4qDJAMO>zJl+xZKaw)v$VDogRR;)~026BBge|Ky+)o+bKL7)w4VJjp1F$Qy9
zsX9DHLJ3G;^b>62ysyRd?{vXgjLZ+%$v7GyQ4Q05&9&9j1>|y)>+r(U=DR|B`18oS
zT=QHHMa_x17$G=JLza~DQR3(3ED|JJsFd#JDpby>n@vA7s|*a}hV5k(4}uVfol}Z>
zcyF7{&)9}X|GR48p#(C>9p%5<pZpfyB{#yd;4xfk)NyhLh*#w+nFOQwHtn0W15TKC
zFkmTE`pXLPoesh2$xL#oRjk(y4z^V<+)wkZ_uTl8XcvWLc%(i0j=Id*<MS`q^9E&F
zi%0l%u9r(vJXI>uB=jR%U&V<EmX@r3dR6YzWVJHS{WwgUe}9vNlKRECrKK1E@Kt6>
zp_r}ps2gPG7Y|Oz9-l?T#TUfGy_aa{XjlOe%c>RLS)RlJY>OWCW3g6;r+%==gw+(U
zYx2`kao#HF&#+dW{#cbEo?QMsh9v8}G45MmZ@C|;cHdj}MO05bEu!8bAwiOsNINw-
zP>Q$lPc&MlRb{F*q|~==yb5RB822dN4IK6ZrLL|1gIytv%A=8jq8bFHnXs&Kz6ny6
znV~~Qa^ItqDM3V*niZS8%Ftm7r*HaVyb~|a=<eSJ<0(v}+q$;Hkjyc1_C7;(T7Aef
zUiO6B6}=Xs5hCoOl(N}0*5#@5Fb|xSIf&2wf-F`fo27Mi{GD6E$WBM|#Mg)ZTe~q`
zD-==QSGJH8+a}HTO{scYY}BTX^5UD)2qJib^-ZM+4E@yxn57aeRsNa#I0_RT`BK)3
z(oRQjbyVx=zhEy)ak+LeT`MoZ_g*MsroA$NnJCEBz~UiO1~}BT2r7EMj<bCR3CIun
z>VdG-EBQq8)9$w0h;D{oNnTY6>STikzbRpGPD?6=--P}E;c1<1v3${JsSXH(MTRB5
zB<zia2ha3zVK-pDMGX233x>OP2GRa>1jW$BXDyEC_`Fu0M=;afYH1b7Z+MXsQ^7`y
ztW<+Y;L%8a4@t_R=(U3|gKJ8F|Kv8(Fb!T6T|!D`jn;C;a&uCRhv_8RP)`sZ3e1M|
zYKp`x?D+sRzW{vid0kaPUUPartT9Dg4O?+69d6I={T_UzZ1&**6c?%1TQG~R;ayA{
zlLx(8DDD2{A#?~5Y>y}SOqF&(ny()*eOy#h!cJ^d^iOrH2;ME7x7QzcY{eGyuPh-c
z$P5yVby^l(zmrB!sk;){p$5;@Rz>yLX<<Hv1ldFkzWxPK*cA+<*A$Q@&@J9qZ$>pp
zL6ojwW#zWs+?Qzi8z>Fh^Ek)ckLqoZj1rj)yRG=CVW)mztvnpqHwe7%)k>W|PQD+0
z`JGZlI;x96D9bz8dBmsCh6^*rn60X4wmh^TdiC|d7$sONM3wxcY@^+X*fcvy1IHQ}
z0&<Ri;B)6?tqqGpP734ZoG>vFk*yVd+rPYdJ1SNk@@o%h^k{m=5p&3M3B8$niD-<d
zPN@0ko?3hBhHbvttw0rJUbr!s5M9=xCwyjrK%_6}YMwg9LVJtE9IZf{F*+g$ve^R}
zqRly)ZOHTES8{gK_wU%`f#OIHXI}{xe>>d?5LT)Qtj6qdC54*-1ospY`=AaGr}6jS
z!+e@>g_g2o8?^I-_uzfM^fO#l*sq9OQ{Oe!UI4;xDs-8}hZsmS_)?92$C2l01bZHr
zpA;4W3;&PAn-<$&8MREhw#`LI0x_&zO!MTxPI&Rx#jGtR$rT?bZ1Kgj6n@=)VS(EY
zySMA<>gDFwpY4cOQ|5lCaV$1gpKsEfo0OWTSoUfdy7A2$t-7vwff)(B^}I!%ZUlI=
zd2(M3TyZhQ51VnV=VY_)+8X^?E|d{j3t4m*XJ-p6_>S?3&Hg;ftoze8P2#T3eK8b^
zy)Cm_TF-hZden)|d(c6V@WcFBB`EzEc`f}|TD5jy?NPKC#buqQJ%&;m?|}bow_BB~
zxFT+PNMr_~cyI(Xy3=`nSM0mo7^R6qXc@i_TPk+_BsmJ;YS16~gJh03-l=r3bdqx)
z56g3m->Yv?c&BDA_+C7t8%(|xL&?3zDU711Jij0Er8Pq{sT&_=EVAp%_nY}K49(1A
z)L(sd<7DGtKB`&P5JZbWIIT5HaKL<Zt;uYZV)JwI<Dg_p(c&T6fnpqSe_C_Ur3r>T
z9h+9bTR2!M`gvizf&>DlauD05l;4dOlM_XsCGGsMRu6<L5srLb4KLITQ*E(z@7B3|
z;U_7YyYv#&r_e7j;^64RClcf$6b6zM8Vq_S<GO8&{OhSkR^iuPrB|oc0n8^D0~?8F
z*hRSh#Xeh#fq-r(7{7S}XTM^t-|6O~=>2$8DV=b&vA(KKFXj9^jOyR}JZgW#<0CMi
zu~@yZt(Sy{3hW?Ry|+4MKIJ|G#gW?;mrPpH+J>Kr+eRI-`VQ=5gnK7YFqW$-I2T`0
ziV5bQg_{NnGDE1FuZS2e7`yO)qeI0M2d*}ypj(^;kD6Bgh917=5SAI%)csV{g2yJ)
zGVbJS_5@ONblqXq<3uW}FqRi9bgDed>OF;nH3OXk-P<z*+1XTPg0L*jY0`rnXo4|&
zP0`^fC6#JOwRxHU3vK!f?O`H-)<|2>t+rcU`&Y>%nkbN|27eQp<-l$6JRJ=-0z2#r
z2G1XiGJG=}yDud=CRi@}q7W}Dgb%{d7}o#~$%|<1lUhP{>binJ(7y++cEDsaw%^#a
zmHAg|H=THQgisbzG!oRz`)25ZkCI{`Nyzh##+O*>u@ir*HL8vXUl)RBe(K}L6Myh_
z<oF&XkTm-rkvx+x0zpmFRcy=k!rz@fun#&gV}VkDl0AReNkE^8$S$_z0s&(>w%M&&
zVCyZ~WcL##8c2?wVOgVY7{-#64R$#44Wg&4sHBguF?kr3{7!j)6!E`dBw?aturUiG
zFj2n$g4w9HAYkMnf<w2V33}Fs5ZXZOc??Z`A0Ly#-XMzdyWGWN<-NPm4!B<_*XE?W
zPOA@Wd||>v!1$JYH;3VMNxbD!^CyC<_unCj7Ruje!A^o>o%BDCM}Yyl*`(vVpIEq#
z$&@UToXt2pF~25}gDIIbxxCRlOw^20Gzexe%<K9fQ>VVRr}5#xnXJD!4F))Xvcv%Q
zvFu;fRS-F}X>|MZ&GPC1xqr5O9qrv09(%h?k;?)TL!$rO<G(;%0Tc<*0I93RV}bU+
zO*#MR<o!3W14x<4gNz^ssRT6Eh5w$@{{emdXPAF|`~N!{6~Ys6){RsAQdZIUrOcqi
zEAuNxVP5ys2uoXe8Nfcu+^P6~U*aFv)j$Mp{l}CgbDu4#kXQX*c8sq7^R<3B8YvP|
z`73~IY6U)PvE(9a`kp(fb^v?EFbcE2^1oC4e-7TA6!sTeHW*Q04)Cn9AAvH->PKDM
z9qu&`ld2CR^Bq8Ha7-zOf3^d_z}np$E#@+n8FqE*T30k&;IhuUV-2;ic4eQ0p^+XK
z_&>Ye0)|~J_MP|rvz7YGwtsVV5tMNb0Es93^$DQ(v~>DDJ{)D3_N=%}D_xQ?CU}`n
zhs-PksELcTkz48k3-YCIYk&=Qfy+4WZlt)ZY2djNloGu2qdf}G{N<C8I?$xv-XSV>
ze?4#5rql4bMdPD3wSDIOeNv6S-~G;WO}e&urjsVS^&FEwSvdgw>CAP~R7DgB(3smT
zwqN8ws{6l9Wj9t>0A7N*1_@N6%aFE34jlIHSMFt-y28KNWG`y*3IB;3Rql4X>Ylel
zdOb!+Z{7%1$w`WR$Z=oGjM%1YG^8+$a8c-wAxs8r)952*i4D63rCzn=+qS0^Se1x?
z$6DU`a4-ZwgOYs!&d<=aibzhxk;1s(<>>YZYuBiEhKk7ZJHF%kX{^N5p=1@|%qB%`
z!zfclNg5geWUR;GxCyC^N^i~Kwy7KA9*SbD%vb?HokJ_@{;Um8Q`I$qIug-Y9-?*e
zO9Gv`OC|#VB(X*!tZf5=YI{KO)<y;XopiF=#)`q@G@wkQlAFB(K#_)C9#2o>0XmpF
z?dYj^%rjsCu(b_LpOUiRIi%2TwRPyPfX)ZdriC6El$uGZ(>xy4snlF1U)CFT`9&h^
z$P0C9*!`E`@sE0Q1BQbJpz>VuK@jDZJ5G~$o9)<782RCYm7{EYrG6WYH2^ka)%+$M
zTWGu7n5!3j6~r45aR_9+TUGMyfk#!g-FjH5FsJXXZ-{;^1YC9~hLF?gM+T>b^2H!!
zJe96pOOePNO_oj4uQ=<f4sO;@?CvKkl9lyCz#X4%0~|Q{1%JPzs;-xtkm)dDyNth1
z!b78(oO{JY>;2KJoxFYbz<<!tsKB9<sFvUuR_^UN5rQ#-xhyA2!4be2Q)~+5K$bt!
zcOK)~7esSS3ttE}P2uM%){hQaNbcos0oX)Ia>+DBlt{S%Ksa}x!hDQ~&(;TkL|x?4
z?vLO6LmKek1;waT!YGpR1{Caddy@0+^9)@Juwtc0L4_^^Qz28qS7{sDA1%}j0aK8Z
zwrKe+R&Wt+G5mq)$a_)>pE?0Kln99zf#t>!z)=ax-2ybRv}OUSWxy9NxxC|M$uDn=
zXI*fAwoG&F;%6AD!p(~F#5Tb8(>LP6Me6xjp+gOT(|o78{9oG#7`GL0tc*14%$3$P
z`_36no2Xky0bcgTkmihseyfYz_FB%g;jV0*`PdZS-*RoL1P!^@pM0yF9g3R;@+5+w
zc)pPp{RPMxz|G4eaa;IA)|77Omr1ju)%`9zz5R<ZD?Iq|A8Xcs-wY6J8n3ba-Py+4
zj!NeHWPr5iRZ6gph)v^h!I<#o`e0V!r0ex1yYuh6n45rP28<0~^&?m`pm6h%Yfg)@
z<$6xn`G)cMXNlUkq@U;_E2nL1UWcAiAjoI!F!B8mY~`~@Qo!YHh}gbe%?7`(MuDaI
zcaS9ea?;;dMm$Bn*{<NldbPEIHY|#?K7iFB^a;RrJsTc5>2y~r*kDE&t@Jn+zW}Cs
zQP;`G#e-4;qa9J88(xHy)t=+Fb_m&{2#~B^!_9VM1`6q(&O_Ac9!)L?Cq59elQB9=
z7<1s}<JCC-E9Ux70j3)x0#gg9q*KewnvDJJ{(*^lq_p#GE(hF>+V~vD7nGmcMSqK7
zi5`jBVEp{cz%zJu`OAu$NMRl6CW0ezt!O>?VetH+q>Ksjwvh3$fGJzp*Q@5I*e`C|
zrI`WM5t%34<@^9v)wl-BNwC^Qtm&_>VsJ;Fq!^F_0qhS?sX<38`DGON=6}dW$tI)U
z?tvDj8MhTjnf33XQb(B4SRAC_Y&YT>O*PKZga1&G{_+0qOBzK|+(cxB)kgPm_$vUv
zoV${W#IN4q!tgdbPPZ4Rd30Dry-rMlE&!tIO+>R{C6GleBGes<*pS-?ZSDE-3ldtE
zXs07!dAdD*necz%TxWmW<UjWP1jIJ!9{BFE9l*jaBIF*kz-aTLMgM=>p97x|n10TK
zEq~O@!EU`pCHiC6d2dd8o$-F^Esx)m$NYCC;$jGjMyq`hmjM!AvFPIqD@)Ug(?|>O
zTVDF<Y(1(~fz$S_T(1Lq03NVjpOIwQ9E`?4cntInE3^1_maiC09~X7xvtXK>0E`xg
z;g5%fozBd<jZC)b#>-?rXJW<Wh027EIV(=ijU%(O^kPB%E(wRf@`D9{h03{+br8OV
zm!JB<Vffd<V|5>%Zb}=!G9f)_4o4?{;yb@_Ge$KOuHw-OKNbeH-~p{DuBM~f5&ph{
z5j@t*OJE<^9!T06UY(^!u6QleDEru2g;mI#_*EHV`9Nx4IO77A({K1Z`*a(H{AVOo
zygvMRlsM+0>uwX7wstNOa{X)~92s(PjQ-eHWdxorF;NbZb7KMaDbG0eVmtbq?3vv2
zlYQ?v8KF_Ahn_n+ze3$lp}12V8)JXrnmARHO=?j>?gypC{0}r{-+?#tl%yG;Wv(~*
zXPTuUwiQ)LgrPWsy;cVkDRXeVm8t69j}EwaKS`sb`Lqe&Z(|o9btz{ilI~$_X-7nU
zaLW;JcY6hFBC~OR?6X`&I|p1M__(jSY8qD6>O6WP;q`1s9e#OdwQ1mctwGz6pT%Jo
z#X9mbYSGV^H|b_Z;|2s0-{0?k7I`I58TXGP-u<Sxz4~YJ8|jWonWnwOY*(F$au<O4
zL|e&-_u~DnEb{D99J9ni-swnsv-a=U^PiF3*CdR9g|Y)_H@Occw7DJy?yP;G&Q_{N
z`RX{@s!S3k-{q$OzF-vwR%Uv_*Pv7R2g<e8;%Q-KOCk&frRe*5mc-(|+I2j2OgTK+
zsKkk}n}&G<g&SaDH3g_v%5Tv~v}9f3(h(8)BrR!JJMWBp9@P>;P{EYG{pyJVozJIT
zuYSs;rqFnvTU&%!uN44Cv~-!e^4C?Hp!5oVl3;|O94`$^HiXoa$&r$j%}DgiQAo^B
zaPL%2{^-TXyrN_b4Vdux(@>nE`EONtGeEO%nwJgBGl_5D(L>ABW1Hp?@|Rs&6%QdO
zA+58U2EW19v%X!jX057*H0Xr9uVG0mTA;Mck7(Tn+$z%TiGMUE%t)DPa=85YRiBrC
zv4D%;aDvLSl7L~zk-JBgikY7g5rG8EI?zd6pnU9NQtX1V)!L&WX{cjpWgj6UC$jlm
zcd9no39QPUUwqha1_hNk2r=vVt|LgUv`f1HW~#AL_wy6LI+_#I63!<<lyQ*S6Y_7~
z01Et%j&=^?k-x`E^LR6;z~C&`#xa1J*O12<+PZ;n{q^sX@c)h6t3?6Arlwx4g4@vo
zdtwrGB10dxR1u~(m6=gKdJT-)Ij|=%VFW~bSqu`?4)>q<GIcIDSnpvIiYm$$ZDC78
zqMhhFv5XJop10&k5%R<pKX*tnkau;5ze}twNmz|aO<ZEmx4lnd76c4ea$^KHBPg+I
z?>`obK}1bAq5l#{arCZ%Yh(v#$522_6^}N=6|*G<d6J@Hi1;`r&pbF3fBYzvjwM25
z84M+oj9sttLM69PXbz5PnkgCjOZi)_1-#d&)YXbK)We)9>Mu8=S(Q6q9u8f+J)EDf
z=O0b!fsjaPkRT?}h&qPZ6m0Sih-Fn?M4Q!Uox@JqAHGme?E=wAJFLQ>q--I86kWbe
z_C)?P=t$vJBzQf5`f6;94y}IBSHGVB+AjWi@?*r=0-%+!oClCH`xDH1KZ{ha&l#<(
zL8Ot3gSTdc13-Q`>hc#wJ^BUyRLrI7=N6o0U^jcjm`J*0ArYB@k-P0h4i>V@XBIpQ
zC_y%BR!*uR0>^t|chXKOcfZ|Cvo=}4>#}Rz|1P)M>KcvdZ<G)}ih5kF`F~xzhk;WS
zws(Sfd}NH}Oi7|)GYmFT^;CpV%QtFZ@r*ZYF5upuE>_veZ%g<bHYzrR&Dsl7BZCbJ
z-n<J=$;u<}PVk3(M;-`l|6#cqdn4XEI5tTu9;jC;8?2b3TXG7sjkkbfOqT?z2krii
z9=rV079QnczvwAGbI+JK&gn%E%QD=so+TLg5+w<I?T(psq9RA!Msr*S;j$a-bRHps
z`VpzE6@}Hqw1P*`akNZz`aoyf!w1W8y><StFE8FN`Cw#N-OJrEYy?UsBj``C8$Zv1
z`S1qpXRunr9-b)<jduu-Q9?+EyML*~kz;$4WO>BUAyAb5wFoi==+U>XwW#X<4_$8o
zmDSpH4J*<iB_$n#N{N!v5-Q!@AkrbyCEX<gN(xBlO-o9Hg-CY_NJ=B+zwYyX-~T=D
zk!K8!W9WgKy|2Bmb*(k$Tyv2!aMr+Gm$JuZ#X97(UD@tD2hZHhmM$eu-)9&Kizrl@
z`tzPkRPUaRCLNBsAe=-@@N=f_6s%fMqIzniJr-1sG=ZY8mSxxU3G?dL7qnVvJj^4C
zj|PqMk1Gm=yDr<j5nN7nQiz@>e)3f5dhq@Ap^L(y(HxypPMGkc`<-Dss7hK|NmA=`
zFUc`+U)^GRRa7s2eFHoQZfRe#B+zsGsS9%J@YWwwX;*GJ&IOTa$e>`W8S_{V|5_xL
z)RVft$h47*f0Ng=EjM;y_BxfClC}x{4QewM=Y`zIU2{fEm;98}sCrB|ax&!(NZMo(
zlV09rdJcZBXjh*o5ulr3G|r<k#<KY}N((;{M)WIuO&LwNmg@K<Z-Zdkao7k`R!!Sx
z0&13s_llyw7vn7jmqSJ+_9-Si>7$hs&>9Y>J)3mzb#IPWTNAO$nwsDfaWIRFWgy1w
z8aBKd#%NRLf2&<b@VWgmQpA0bn~D1M`bcc-ljO+yRLY=He*({&x36g9zJIMg%+-=B
zpR$zVC+Iops^){35R~+#%WusEYvNlq-BeDInn^ptzTq=T60`As{AT-jPhMXEJ4JR>
z6xRRnfc}5!u96?7nzl*IYrW+R-)7hTlsm=`updKjluTklIPpU^u6ae0!ciIga`I+v
ztqz*f*p4+5F2g(pYTB8ytT{2lYifQ;de0tTHBQ*lA`p(PG&@sQ`tvHaM>BDYe2cZT
zuI=zLiHaOULY3mFu$j_ASJc@#Ec<uNkufnBu~%PGOn0rs1oF%<(O3^caYmYS4<)gT
zh;BT7oZ`2NoiGY@(Mn9v1AMYwBXQecPi9;CT!Tr)p1Xq{jsC88zEZFHzY8<{9=Cl|
z-hOr0nULjbkOi2n`QNwm2^F54`-%UKEKvQ<cvoE(W;SiCax32JyIhEnkN23*Mx(Q0
zkNk!rOulCtL{6E@G9K`2IEV&|O~IHBr}JSgQ@k1GakSRp69@Vlx(=Gvk8kU)RI<H^
zN#QJABHye=q;O;F-@tmDf;rt`i|fQw_fv&Vc0F@A5cle^;zmz>4)ImG2;ZCbW&dxT
z|D)DO2thhYDzPB$1j&h666tDp80fh?c5f*UZF)V}=1hc=&mGX?f9Hh~m`tiRRuHKZ
zhFwz--l$rHdQcI1h7z-rCB#FHvD^I2u(`HhZE1n+d!uV-uF#^?`h(;$p_zW>`KkWs
z;*Uk|yxR`B(BbjkWxA)#JN)X7a+h(293#_B%>B=Otc%pWN=jeQzDN5kBvP{W=Tmm-
zE0Gd8adynLtH&ewe|@+jdn21J*5Q57wMZ<wNnUytQ(Yw!&eDj(y`bG>eus{3oJC@*
z?pZ6D_pe7Kq!MGapaKi=lDofpsTMXul718P^7Xqu(tfFC127AgBj%fXH_oPD#O1lM
zcz$iw2xR&nZsy;;A_9hG0?~q=qZ-53P+|{gMSvM6uYb*#zaxTQCk~Q0M9_s4*R5nl
zgunNtK=8pF<GXr}j*5Y!eoSK32Y_fK+)r^-WICth$Y2r``$2|@u6O0JNk4e1r*cW&
z3MPtwMxgQJ)@{dyK%s0QyCO(wpBLDgFow~GPM{kTcRR5<JtVmqF&gmU$_W7(#~#@b
zarlG5=RbH}8pc|Z{UG<PdzCehvsv^zRM;W#15P)K_!Ya<s>}OhJR^GX**iG4{*R>M
zTnCjOx|^)5WXC8xGADZJtj3JM=#_aHjx>Jno@At<_s03y0ricEt6r#ro>~B`6<4tM
zn{k*8NtIwY>gr=iMv8b?+^ti+bc?s0%1FfF)XUc_^s3b(#)Bo-=eD93oL;0NDC*Ip
zjBikGGZ0~j-6fd0ea=|KEJJeF=H*<x?W>Sqx+=EDw+=$5g1_TrU3!W7{hd;K8gcET
zgBt;yZ^Cx9i$@vJX*OTwP?UxcSP~fIzrt3YMbp#^oh$xcQe}!L0h_729iAOu64<i*
z)aIPioQJn(mQ@oGhQo9`0Ss|&cKOtfKRRjL7vs3Q3`F3Yji??eUbCpnoVWVq9<Mld
z1U^{qDhVKK<ixsZ!MezTO6Hx9YZBx0;1{pW=r`6qC7WQmCL|ts-P9iT@YfT9h-kJ?
z2w5lSps@SYLHDkOX+BKdXSSZMraV-5&=V3Tb1ddlF?#b31NxiV6c?=tK58Kq;-QT&
z5mrawW9=<v-Z&eh$JNMk@(*qTwcA4aKRFZbt<tyj%;X$_jM9-k4O@Bz9~}^j<V^1|
zZ)Y_IEicGUveDxf&@z;>&EN41eGuAIj_)Sk)ww;|V~`r9O!o1qA-xIRcete|ayq0B
z*5huX-(`&?b6ilGC&aiz5IFa`u5L#-&<%{^gty5CUna<}6AE5ciE+7mfCzsk9@dm8
z5t7T%+o2Xoocv01J<|r2Wjx^1J&tAi8vPHfK0$iHp=?a<9c>H;4wP5lg#X0w{ru*^
zqbH09^ZnI_7X%r$X-qW%6MP19Kx-Z!Zcgo<>w~XT`f~<evO>y_!ux}&y{~Sml5WLm
zy%=U^Gr?b{$ECvxRSr3pH7W9s)Ot!!ahXCYfPr=9D?Ka6JFV0`dc(0DpNC#uVz>)0
zbuCWX9l<yw^p;yY9^X+1Tc%JTK`(mf9}GLNuP;=zA*e~(!@dlo;+BuRmEM|%{RBV6
zP=T%p(`PonSct87Ezxbo(#VL6Qk|Ey1v(6$5P3#oaFJWiAnn-`Jh_+w))N01_ao`p
z@nN*-9&8LpugPu+s`i0v_!aNgFh<4F=bzEAP=%ckv>3vnWc6jXO>X8lW3HK;O|Xl2
z`n?&HIO`4rv9A9Rm;U2$jYK-MP6_j$br@*J&U47bdQD<EDt~@z)^%f8TRV(mXQP4N
zhflPFhUuzI!OnFvOs;V{>|`Em4J(vjCBD@=^PXt_bK$IIY$TqKu<mBoSoCK-SV-wl
zhdPSEPfHI>_g`}8*9S6_p33O8e)FAFrQh}6PT!()M|n(C(<zC=k@Mt>uwD|q)(2cZ
zz47Z?1J%|eW=;FUyAx%NJAa5O-YwmG6~~l`Il55zRp2Gh{FwN$a;f|7l8B@9%_{{Y
z74t}Q?`diQn}vPwN|b#pt|zt|XA~=wf4P9RLYTZzdWmpGqD#W4Yqd@x&2ER|vrWja
zsLDTy2|@u}FFN!sa}pYLUKM{uBuKA!YKI8$hUtYmlTRQgOAxoc3bfSYZR{iRj19Sh
zj{9pjNC3qc=gn01!<fOS<b_?br=9Gq?&!NLb#x(TVUE2L(ONkl-(#KES!Ms~h#bYq
zesa-S$99g%=G;N!x2<ed0+^`?tM&SH-4VX#IVn*HUfT_}S$E;J;z|{wz&hu;w62M;
z$A=Ium&(dt+0&yvS|lJcMuM`Pm@#gyuf)v4@$CGVr0M3QETvyY>$^@35v$axK;5%R
zz!F>B9K<vZe!Zj51vl#AyNPm!p#~pdAn^2hSYtRp);XM~NLxk~5V;26W8<wY&sRqD
z%=8N`035SaQ9F~oadaKGG@xwLQha@f{<lmAu_-lWIDb#gor4=UzR(FehGIR^Fzg~H
z!kdpeqkm9vrHzDyD*Tfeb8s_)T64o@hJN(X62bT-Kaw87`C~L!;~I%bA`JW|)NdRq
z^2%u+t6Siod`{+JWnty8YJ~6RaeNYdXMMWF^J*rVG=Wts8aa^wv!#G^C<{e&-<2yB
zTQ6UrqTM#VWEom`$8+5ccm9wd7zsR{NKVy>WfwDLelCGIEDIvvY0IG}AYqre?z;le
zA1nmnwX_g>1x6*jf_J%JNmOIKe?;q$!?MOH^LUAeqo<A-Bk6ZJ{u3zv5?rpB@J?W;
zy(QV;&7k=<eL*2>g!wA;gUZA8CMf+x=tLqkOu|SWRJ;nEW|+Pt;D0yId2QHWh=g>$
z@qyMiK1>GE4YMavwY#9%TEzI6UEEe&RhmiaPls)$*mH!<MS$x(R_C}NbPKfr#wSet
zHpA~2kucUumQUs95`aRs$pi~cRqDDKjU<{kt-v$0_l3cY;%L|Pbm>hj3|CyiSw9P>
zhYWY;H4}$Xw5Ow`U5`&lmTxSADGX~tx{2qV2ctPH;+WrOwB~e&M|haPP&>dafR1sP
ziKF&6E{S2{H$MW2C*4OcLzgm4*|)+RLya-{=l_U1AgJ43OOiVt7!ql96pAmcRap*v
zHy0c2)DCQSbQA}m?q<5F*E+zLOdZt$cdiynmD8!-S!7wbr2IJJv6lZhR|*pvQTXC5
zg401fUIK5&hSqSoPjjQc4U`Gd{qJok-Ljwd+Wo@4WLVP-V_F77C|<Ck?^+EN{-C1F
zirY;<eGu!^S}i<yB`idTyDDV&`V9l#FzK6ptl{FJ8oy@oJI#d39%w&$u(W-&z2F_Z
zIGamVLwxX}rqk+!icz`z1(BZ|g(UC;(Zlt))|Y5^-hF@pAey+$CUWAq<#a*Z`8|<I
zomVw_4a*%^a*ed~o*F8X$K7OLep2z|2)AD)fBxoktS9(^^dIY$SeUTw!fuwM2$Tak
z<8~-Ij!h^btrHsd9XEmEO=Q?YH+dIFe$Iww5AQ3bSb-m3LL>a~Zs@L9pJHQ4JR^8i
zogE`G-Y2ZrN-v_UFYd8TQ>%no@BnX!oTC^#Nn&Zp_>8V4$FkjQqC2+0F~u`Zj(OX8
zMlN-o>jY)&we6LYOw_<VT<`3bo4uod5ZyGg%Y@$TY2*($iZF3b9_)!_J@|%BHd!7u
zf}F?G%_u2O8LEicq!jGrG+J(aKI%w~0fPfXpIUOR@;s8gJJQu*luwIB`OZXcxPPlS
z#g?DBj35?KbXSrpjN8ON;hB6g_x5fEZqE;{asPSte8zSJzQ7`Z;;OjY_l#fO2P9Hv
z7@ak^%bG5N8E<G?Lr*ZU$R~>zcIp=5$C7vPOe1BB$Qa1@HrR$U6}p6I5me(Q7ezyO
zG%+9Usucu{sXWnHzkh2i^_8Lm>a5DpVfbco4+#rN3^B!gu6){(PuhMFao7*uF*o8`
z*=+sEL|wa|@3>hHjY)7mhp9bNQNK?}ab7){VCF*LxA?Bki6Js|P+#G8y<8aDNzWT@
z*G;N_09!)%^xQx3^}V;^{D|q75|O+kwPc|j-&w@Z&55mMXNIS#lg!>{Y79AViIrE@
zkR)vyW)jhwd8DL*z<Z^{ar9fiQhP*2%~d}B7%U)nI&aLxOXb_E%#vFyQSGv3seTiW
z&6($OTu|q&fSNgjo~}*Yzm6cKSnad^)uYMLV3?U<nKX!?5xlcFjsSd{q}~(ptjg*8
zyT=OHd>CH`wc|Fv&?t2@#S?uqrCQ2X3omuq7$3_z82dZ#2d*2NzFdvSQ{BQVaXj|{
zNAJ|g<Dez1=5V?>Ivo1FMUK_QrPX(4dRE)j(f^G|78`o_A+)JbG+5RQU#Ug-)Im{k
z_13{_&5g{L$%gTt-f6Dgq7PufP0)MyfaK}!1jj_*MHliv!!mwL!AJd;`g&0G9JTrE
zKv!}7mQR|mx>wn6ktH0?##t<3HIKT5`Ev{Ni$KDEszEVzY2@VydB}=SgoT92mJ%ts
zJ+lv?cDd<V#-f<Mu5TX}C#^5@L22{84aNnT?f;=YNclh)p*tzvF&-cA;$=pCH_h6H
z>B2^3pUCtL%`qLPkFO5=%m_YWuO9EOEx)241#JE!p!OwI`Zg&SQO*BDzN@1kN&JpG
z>{C%p3qJu*TnF0AXQbkZ><E;g{&z#qx4UV5GSY=T>yb`HX>d>rXd5Kisg+!mkN$JK
zC=G$cqO34xR^|J-=%3SpC~~eg@f{w^?;Ry|^Wjepynwp5>xrc=h4ZP@av-<V{7dqa
zO;fc0NALXqS=|3Y%=2#pDWj<_{c+s#zpgU!b|m53$o!1$|F=ZrU!Ut&=npaj^(XI&
z(=Lpw|Mwro0^!>T$q#Ct5gh%@%a{ZnZ!Q>*a!KU$m&9QwnPDBwCL6YG{=KyMHsI<r
z3@V2gMY^@6qDP>4*hJc7y3BaH(vp=u)2TWdmSJlRBW9^@4ZvvV&wkhtBYHl12F#Nk
zva&*;#oS-LSv#bm`}=b<tXE~jLbTu{I;G8QG|XA|Ckkh47OqYFf1VTnTQni&j$G`<
zQ0?FTgh%U@eR@y@=-1j!QEfpJiheD)kb9a~0b}B5+8yl-Jt;NFk!pI~(bUv4+t}}i
zG9@tnc(-B{&7&=UeuLz&7Vo=mu0zitb&M2nx5DVotAmaJgSv9IoK`5(pzuDVatvhI
zr}0>!j~Rk-eH(`b^LMm%|J#DUrNU&}?wBstuAJQZ_^y6@wJ7NdDQi6)=8NB;JlL=U
z$Mt`hG8o6>3S|WUpgC~pE;{=lnA;rEd{_s68z$BSRP6Zc{WUhlUzsYp*KIz(=0cK^
zb{}4xtK8*S)`59h<2ANxSl?hISRMf!!+e`aBjwT3@;2<vldYCRw+haW5C20J_@6us
zR0{P3`8yp3FPYMX9&GGEmb@G}4Y0;?Zwzp5qYby0Y^v}R{4c$p0YIZvHi<)J`fN_8
zMrasWX-#j`yt~qTG#`=P;|af4Kk7QqB+>)Av~fpft5o;9h-};8=MgYR^DuE*?1{4j
z3B}%7@Vg@IO7m%sc^^-JFFkHBp+yTUk;jaH&Lh6I(fhx?;<6kl>gl8Hox4a|Av2X`
z<NKgMn1T%Xhv};y6LvG|R48>3pnTcioc7#uhMMEqa=K@F11K=2u3Hx?uZo+0P5s$h
z0@TJ3@r$}-8rMn3oK4jsz6Yzq@Ijkhl=jb&v+!U#ODzju3b&<-`3N$UY_}y)eBT3d
zzy-7t=5J24|AR>W*Gg6(SCU4i3BvntrYmEa^aYpRJxt-Xtw>pSLSwvR+J(F2fg`}D
zS=#p$<|ew*xSFPQIi6;9Hsi*XxG@E4XC;dhy}5~pr`-GP!DlryCTn^@_LpiV7QqE?
zxP#2U^J*9NHg}m?(lhI-4I`$Ue9PWAwR}~j`O5f1pnLhy!!glQ;P9+DqouiQ?XY3I
zwl)MEx14A>J>?sxCAA2ehZSz-2$Xru&-E%c-?dEm%z60lF89^K>_9iV@p9vAhqT6p
znK*wjv|Omu!pkc2EE}w(+r%ziMkV;qKl}z{=?VgxN_1;Ia;8%SoV+%;c#Mrsf1!ln
z)7HRvMt7xZ_g(X?>B4U^>K0+q1*&<++ero<1ucdB(uL7EvPn7_BHnF0`v&{BYR^s%
z4|(=?z>@PDyLG+ONU6TO^=`rZ7E*Mw!te6nuirgL^k~&+3x=2E-MXgyHtz+@e*F5V
zBrl%HHM{VhFh&WP=3BALYtu~)iiA=2rw)y}?tt_Qe`^44@=UKl$1m5a%g1=7z!?+b
z1ZAnXTXu-oP@%dYjdGPY3^a2A2}v?aGh|iY28H=}tCki)#kOuOo12rPy*=M)n&Kw)
z*A-nOel%{Rs4ChWNwM$+{hGlb$gN-}K84=W@^BJXqMb7_-+b&m?lTu~I~PIIOtT;Q
zp`mJ3d6%1KA0%Ffo23Pf&C{T&5voJdbL*21$rbluLr?Ig`lo!Ln*0;cb=|^3zsbI|
zjI8B6Jw51pTnC1)wmb8B#3d9=x)=ZX$CxgiWq`1=($Ia=%BOu_>GbLxjIGDAa-6sN
zr#>ceW*Dc4+$2Z^Zr4)27KvKJ5oazHe(m?Lk5}C54{bp>!uiP}-u13k>ecCmJxEnl
zhr+KvO$S>u_Jxw4(gbUN^v1Kub~Atanp{zA*TVQPy6OtXRjq+s`Kdxs2x>P60Y=cK
zMyOqG)QZ~;=iC5HhIFhg)_Y%pN6Gsc2<Df@v?Blen&B42tT35u%lYi5h49&7xR>`p
zz1<8I;O$|W(Ma771#f|h%vy@&YF@5rVO9amyP!CjlbA<A4n%B^qU1|PxgF$zU0kv7
zE%|wJbCdVULHsj!^+L66XH<w+Y2SW-9~clGRaH+5E>?5Q`}v8*S}uk6OTe&H1gSP8
z2*toD<gS_heO0Sia8+cU*Fex$0jdKxC5Y8WxFlZK`)brV1@93Emkk}m82h$BEDab~
zNy%*yYpKsCJ&XqTy`bvAgQqE>YQck7`ydEgTXYQHp#$V(Kc}9?kfLVv^mPwkQTy?f
zK?3i=V)i$A>J%QlL!?$}@mt`bz_Z2B679+^m#ygyk|MQiX1VU@e?4bRrZlnd7@phn
z*!Q@<XnesME;F?Mz4N0QCWRhfcR=ubj;6x0X%F77@z}S**SxJ>_$p`r%e?Q_bKgJk
z%6Z-?{2-SNY{|KRVYAm*qV$&=ds}9>XY!rrr9=W5{!fG6TG=Vr+Ub`S53pYO*Y1bE
z@`H^t_5tc!?B<rU4g297S+bMEmOr-aI#uP!c~s$~>>dmvR8F{;)r0ZK?|fjEPc{Q|
zXWueIM}ng=^W#Sd2E8z#a$&9g$H#<EBR>YtS4oSU4+dAn`{YP$w<QuR?%TmD<pL8x
zC*X2vTLf-R<I{>v?kTP<6sRZ;e2C$Z*T9pia+w1K+vMXl{KDN`NG-hagi8>RwOQX*
zh@b!I6!-5nOZD=5g>xYJPHYQwUDXe1+&jQ#)XI0r!y_`D^)1iBwe_>#eZDn!pL{$T
z52C*Ryl3wr>_YQZ?4m9f-CDc-TLl?jJ7HU8R}nl`gXIvibb;q-GPOAkQ!_4*`3yM0
z@1ldpHJO$(fUW*Q=b#pJ;pVC5*AOe%zck~uE1zN2Y%_LO_x-aR43B#I68eII57?ML
zy8bh?{<GG81B!GvU>DP>k4%6*W}qG48NCUVDGDulC&Dv`!sTlk1q>B#+w%s95eTk<
zKI;0$B-mG}_`e-$@=cweoum{VMe|<6JXjl<kxAls0{fvl<8+bb;RL&`FQ3c0mf_4c
zWFOVY@tu^GdJe5<So)Q>5359G{LygM0i1Nca-K_^Wy<ou!~SJd6%FKd373G#fnLwB
zEC9n(G18;$D`$$ADqrP^VcIbFV6r2~QnqE03RU41!QIdIVmyCh*IrXA2n%atcMi3M
z$C}%1NiB-z3qqiX5S<_#vaio)KJ(@W?9BW*cUAo{+e?m3R@Sh2pH;I|??K+cHOHVg
zvWaY5JT5=UH0YSGe=Wd@%-y!qZ-;L`1eqcqt8$j)gO*PU|9;wZvB@a6DAo&16r-@=
z$u9A=AmD70vVNblWGdwBnqyg~!>WKs4?N{^t2%?<08)w6I=ntsSt1^QPBUz|(RhR~
z>BLUqvrjMM%G+O>c3Z$J(@?#cn3%YfsRjRQ9$#R|Gbj|hTg}q#Zg6g!)qUG9$vnjO
zr@(1&@dR{05eAO=Ru9nvx+zZ%Hg<U!Gu)RbX7-`VuY`e{-SBLu5jGG1I!*>j9`zd8
z!<*IIH4Tc?^#u8rlc}w{LgLO{!p|1VUAfiop>&1ebntyc5<`tU3Tw^O)^OR~S9s;)
zGzM_xWz~7ohkr)UBGr&hAZyH>QCh%7tARuovGe1T=Rg>M*S8A_hX*5FkA4|9e@(w`
zfz+Vp-tJxqipS<rpW^d9_iEV!q<*~2XTQj;aHT1ipVP2;@)|?of)ni6duh{D)E$)n
zeme*r$@sm)Js5PG1L0ULR9?T;bk!BofNVd%G)SW9mp*KRED$WQO*0<^Wrb9FpSTR@
zJhdwB&sS!|*Op1RO=7z&Ka6@!Lq3Ifh;=iOLw}XT0{bU)+4&3-7c^`DWaV{zbG|de
zI|G#3AsHu6b^H$CZFT>pS^wPfPnwyYA!l%=RV$6KhZ~DAnum_*FJz*~(Teb%Hypy0
zo|{?hjXzt1XjCU|-b~CQSJf(kJ^|0)drCt7Z0Ook=fNj`p8-}_Oi#-5EK^}NEf1Om
zGK~FvgS{h@$Y!aUE4N)xyLx@R!PUM<Df2-vORn?q-Wf<+n~ooEK{tK^9Cc>$voJP!
znnKfh$x>#s&*$KtuL#^<H3}N`QLq1<%WC;Kxy;xaeFDo=pWfWdQ>;|K7YT3m)8lMF
zqz}S*<co%zSt^<ha#=*tD7ibed|k2mojL1Wd8iAN=Gw}`_k78!T(=AcX5J6qnQU~g
zBw^F~R>hI|Kn%_Mjl8P{$t<Wd<@zKgd?$`e;21dF>Nt)0&xGNhU2}^u_El^8b<B=r
ztPatz%CGl7N!;KevluScR_i2d>4T!LQAW{N8Au)Z0dW$PyUi&t{2O5B*5}=?vkEV1
zYbaDN(&^8V^0u<@@TgDoJ^!6s(2t$UJ>;@E`To54t~Em@I(6YCHS+wQ`5<cjZAX?Y
zAFc6Y<;Zct_Oq&~oNxRpdNF~{pMTN}pUL-5c|7*hlX>g6FCit*B~T{2td-Py-1Ml@
zRW2{N-2@51Z}YP#R)-S(hJ${WWx>B+MHA$!xadt+{Am1IX-?k$NLA_U;im)R*4qPi
z_dGeU<!Wrk=KTh?hxtCOQ+fG5zbqeQP-~YK;!>npR#Vn|`tjLv?}lGG${;9|M~nKR
zefDZwhMrrac_ix;zTm8^{?Y7h_RK5$h2mUpp(2Tvbz#LhGiU^t<OR6Tv=XAUDFj{D
z%bn-0dtx8$EcPyZk$*0N{_kWB^If9YcXZE!ee$}L&#2~~U(fla=W4Ug4jg+Fo=()*
zrtk&Ry%w%dom|Y}`3+^M4V>_T$I)JYr+trE#mbYaVm9~cem%|b9==jY%C2)%bbsM0
z#oZ1{+r9O%dp}1ePDIk@p%879;bdq30?FEsDaV$WnYq-z0;hu&8s>pQiD-)tmEZP4
zx98iuucfQZC!k^;%aV#HhMfTEPRRn5_pz-(v)6k*@&YYHmgZ`~_c*`5Uaf*#aAnDT
zJWn=d*7ge?*h6ZRDC+C8wdn!u=gG0tEvI;vZRvxJiRH9lp8dq1wJd~*s`URWr^Td;
z#e)l~n3I@$c^5lBGcZ{WO38N0bhGL=l3kVCSbrxdkMJ)0Hx0T)Pon*`=5#MUSTc*1
zyCw@e*pdDHlVWr{sF(*T3j)u`{#qBfs=POsTBj`!KD^ie>y;C`<Tpq_H!2&o+@t?j
z7=d+3HAg}9%p1G1>_Yh0i4RAs#0w)%o^;d1Ee6czp?}C+ZLWTbW87r4zv!DQN-nh;
z)Od5x%mn4+<xPC>yJ8D38HLX~$TX1*kTtwK!CSxowTl1zHTDcuMPziM#kU2S;Os2-
zWt{CP_LKB3PPxuhg9vgAs$&v;v{nPj*1rS17~KTUJH7H+sO~&LAh!XPe)W5&Wsm$3
zNCA;O12w+nqKjxSb$3)clH6F>^B~eE(TFo^7Vq!J5W9h<!aRz6%bVb&<n`FIn5c8i
z=CSzp0h-b+kOSgLIP~}@L_<{_{rW^siuYIg+d*kKC#qQPdOIDZCG4+P4Kww%m<+0C
z1Q}-yw6=`z{TwKH_T!_=pR+?<cwN~bonXm*mP{oD&tAyFj`*)<ucAWtGXUJmN!|1F
z3Tgp9sG*k#)=InYtg>UAww}T;Dma#vivQP-5gg5-zPvxGTVWQqzc$jBtB__z!m6>i
zKW_(&C#?c!%Gt|g6J#9Yby|{BC0T&h8{352>J_j9-B+~pl`?TPQy-NdLPgsxax{Cz
zu+jnQ(PndF_;Z7-<tWHfSfXUe6$ogpH+P?8Ic-hU_9b!}+CfLa&e~|@(-=(^Nd9MH
zPL`Y{F?P`ZzC1BHW50KJ2LMSt0sTjnVgtL<+ZS+mU!hOP3ZjKe8()TupwId5<pVIZ
zJ8o(6JRDc_JyvDasmlM7Av$9=)adTy`KuK}*eGJ^aNEBLj;}Eg9Z$JK1x=Hl$gVrA
zQD$IE$!qg#pvq?KJ321;?8IJBwEO9Hw=K}}xi8A#R&$ptjapGuKez%)q`NiUw9$sc
z8L5w6cF`tdVIW}oc{I{;P7;`4A@0^EPELb1ckIzR*Z$D|jvRKI5$iz1F`$P5kk6q3
z%8T9BMc7Ja=oR;~xwm2t@L-(H<`cWx*oTm82EemGHSRP3kq^G)Of3}Em{&fm)U^F#
zzp3)>QI)VwI-y)De}8SrkZ<d+x%Pn5-@7LVf}3@Wn5=tV6CV>;15)hjdt2}Wn&@zk
zPfy?pV{A;+#(W)tNJv+_gM;jlp<9=s@}7ZwX+^KU^VfxsKo%u2r>U~{tMLy_0*xRk
z@%|X!(}lONlHywOOEZ9iYKsGO<}Z3Whu``&P4J#xHA*X_X&(>a&DHsp)N*>G{lM(R
z0pi{)^4Sn51VstKi9kgKo7nzGneUnP+YbtUGiYRbv&;xjA&v5+^mH0vBG~%{1f5r}
z1*&@@Bmw~16SV-|hydr{IW@MNEM#2N*|Xd()b(AE7@pKPj6c6)`TgCBUZeQGvr>4i
zdg{eg&^>9r_mtljat5g%C-K=ElznJ`_8WglsZcw^Np`n(e|%aaG1INF2{VJ_ff(5t
z)vT}Xxz)I<P{mF$14jpMJ+1aNP6Hvg#khxAy6zj*BmT0SP23ePQwm9WqDj_+))u5l
z;_h!rk?CI(g8q(Hzz$Ef)TM(}Jp&x9Y|Eh+YBojOD2(WaKLE)m&BylRRhF%QMpNmB
zv^`Yc;rd@l)&JQI3UYL}0-6ALLWcx>H+H3Q_bU!^Bz^u}IT8~PwwqrwFL=2DMPbx)
zZv8{3(y@?-@qnbdW#VUv%ag#pQJ|Sa;)|2+zVv1#@tU3iOmXW!KsNW~{j>@(Me{VS
zHO}}>iKA$Gc@-YUnpg`3?~uc800urfgBPSO{*cHE9aF0>%LDITAI&%I*4{oc`djU1
z-mNCwi%t?^;zu8sD3BSv>xZ;^Powp{_lHQ)JvDM<67JkPp22KqK3SmsV=-6nl&gR+
zCVyhQ|2k+u(zM~-=QUR2hqKd79@^wDZ#-9=<2GBHX=!oo71r2|bil5}gM<J=^LEK@
zq4;B@N|7Q^HJ$zG*Hlu4-}_5cR>g5WNHMT0_c>I;={=k40e^>Cx?gz6<jTA+-PZUX
zI<h_rQnha)w$XZ%G&fwuZbM~kobx(}#FmuTzuZB_D%e|PPNFXQtt<M~loyh<stqo-
z4NJ+^S$*5~-@kt^h6kzv>ARptd7kO{c8sr^R;|Na+mWi|7QFgaO>IihnrgqtQRjWU
zhk$}wa$;Yw<@5(<HvD*-^B=ZirmB}WWV?L7cwV%OsAw<m{u7xsJ>~I=hLBUkDS513
z1x3-{oqSf8n2tw%f3WT6TNmgZ8hvs3ik!Mh0dE^e4b3_TzuD`nLFA&vf5v~D3jOkE
zo^#w!a8k@hf!;Qo+gr*`mnMmajJWcNY?5{c#_=fRFxh9<jsf+%dHnf;v-6T)JPV}R
z(*P&}GK05!8v$W)3<xJ1&}#I(t#DkB#z}cJ3Xzo^RBV^yZy&1I2Fl+;f>Sjno#89P
z<F)o%B;k;DM>h+OBQFSH+Gb8^(3E!x9Jt;i%FcM;=UU1}%HIS-Yv%27pFZJ3x3@ZQ
z!mJ2RUzm|lA)ljKS}xD7+T&cF>Ak{_O-I{6^&B5Nn07^OTNdbql&R#%Y&X2?@GQAE
z3iCZ-0VEHz%T?D_h6+?!juq^yrj17RI@&nenmMKs{=)L@L-&K8L6sks!Ln{*jVzIP
z@I-+y`lGP6u@7)y-caE)#SvmkgjF*j(Ou0uBjwqp1=sH=rVIH{Q}FIwzUTm>M_y-}
z2tBX&@pJoVg!Kr=$%ehJHEgf14(-_K(ClHs+4#$U3aAfKm+*gTD-mO|kvm%VZ2(Iv
zzxS1&vE8xvz!*{#gS5QWj?tbB(Uz3_mz)~xd3W~0SklX({vD}z*;xI;8BMlcfTEb0
znryL>ir}*R&NSfRu{KgxtJOJ;$F4K__M5JswN1|nFbvhulhcPy%C=&-=HXldX~)49
z$@zuxr-B1nU)iau+Ez!W-rl~>YqO$i2lxgRmPNP~j8PRDRS`p;vNO3nvt7~DYjZD$
zCsjvA%(fCw;F$9yj@aAVTRHdL%cgZVjUzRl&BcBXD(30I=Zj#|yyp!y`47LhZ7riW
zHHu$r&R9k@Yb``hHMn|)0vNV7YKKfb3^nZ`A{%hsGUVY>4lO*i9x07FU{`q;3cB&n
zg2Zcj;Inw(e8MfS>6cN6L2=QstWAc*xoZo5=B-9I0Fc!HdKuyT+nS}RorU|?6fBzr
z)Hi>8O`C8Y#C>3#*j51L=FtfpUr+QoD5*0)$W1vqXKc%J&vPoPnnT{Vruv-Z_|_{`
zCqjXan64<wlfmgPDSRi*tD(9G6Ded9S4ehY?4XWka&FPb$7s*7ch%zZY>_F9Xvh)s
z_3h;m=M_Dx;!G>izY8l!^24y0oXhi!9@D4i&J{AqtBOXDvRmokY(f=N4gI$%4+MiA
zBM`>z=w;srUiAad<4Jtg`HTDk0;=h<3RTrOG#pa1x-?L!jswH=1h}G*0)M-o-*cu1
z=W~(_K9(D<w?pIS=GtyS#WtJ-2m8O~Q%o#cKn>VkmEqq6C}PLe>L$;F@j**f;YroQ
z`J*^Mj?`aalap}Vz3`OML$WHPw$@)7p;_y7ZdRdHvDfdNT-cOioHb<z&{k^+N9#6$
z(tx6>WxjY9Y#<vrl?;^gzRAl=gp+Zm=S-+|PD4&-enU2K?_)jW7w^l~udNo0E0eei
zI4udSy7Qd5g2Kut(uE%m<<tAm*)D*0#bz6D-bt1pPEqh*`8#okGNShe6m@!i5k-HI
zA;{hP0gAOFi3@5R0x>!)v~!$Gm;U?_^-=fyzhm6j(=c9N>l_ZoM`tQ{OMb@~mDs!^
zP~a`OGMB#7gZ`qdn<fuF`wKi&;$mHTVL*I)DOA9?jq1)@leg!)!)MlZee0h$*DG)Y
zKWX+=?%B9HJMZfbH4XByaN`)rNuf%L{g<z|S(x1M<(kS1IbXwHV1{1!CB{T&tVk^W
zUp_$$^>*xwz$R=eZfEWP`YlZBmRB$SaQKa4XhgB)ntlPf91&vleulnhLp~=PeAW;>
zA*(I?Rqj*9g0gu1x~Na%7m<^7UXTzDs}@;FU+~LTrN!C|rs{sO5n6DgBD?<}g<Y`8
zePdn$aPWkREc=kf@$lgNyT)y($li?-Ku~xcX2o=$_1;JsXJ7UCdzo&BBJb^bvCPIQ
zu=N&j9sqFZ`5DQKpFp3kZ7_v-Ib^C6<|(%3DK-^hU!ePW`yk!t*wPc3_<!LUsRB(_
zt%|*S;XY*|k~N187Ec%GXVi(I5zZa2&KtpjUTJHWZ{iqZ@PYO)dC5O;@TW7wO0(3z
z9@<+%g4psAO!UkGmpLSGjjU6l_O$g%hvdy>NL{n~NuoA#JD8U>9~Z)fI3a;y;A4~~
zyAPn#FWMjLL|umX2%N9iJBSslChQv6bz(N)hK3v6cfYf&JZd7WgT}dqRH@e@d*39o
zdo=B%is%UmF8}-MW<0FQUGV~)->A$AsTN573yQ?iQG{s;FiptX8<dwxUjR#fAIz+H
z6)LK1TVVaT>?llVNxgxIG`uE%|M^O6Q4zj67IGG_%kj^=320li4}lRquxNVftd@!a
z1h0kME7HIL(`CDp?4fM0zROrA(ER;$BP!VpbSVG+QVJLA#P&4w+71%l?jYv1o2*kg
zBt7ioZ;z}w#s2h33ZWyj@A?bi{_5R6(HUsNMyYK)`2-E!4+Vkv&YrL>C=!C(R_^hH
zuKBD^itYJO!{CNe8Ap_@=95w+h2sO4c{8%KdRV8E)yh7A&<(0;Tlg<#gAv3RJ0I~}
z*m>Cgx-L4DPE56T<qyqeCh-v{CvzdR;ZvG+v>M!6FOYq(@4b(|-9hak_1NiUnShvT
z1&ngVm8<-B!WAKkKCLF}$hlU8RE$esj;(tlxKX8Ay)4&<fgoi}p9TX5QzWn9b9Sr^
zhe_U3rh)b87?2S(08uh8Gy_49`|5xV({tm0Kf>1{Sa3~;@GgvlxY+t>x_j;y;5ml@
zGq`OR%4CBSk(db8n#KqZLqm4=Pp81V@A{?!!9zszlkXnKn=E~6-7Mta654$=bY=b*
z{R*LP%*JX_nZdC8fawjQCsfXZ%E6tZV54EKYlG<A3(ZuX+=UXy45`$swjAg}&XL!J
zCr>HWGKlr>izD#+86YNw?O8WVq-z6NX{23xhb^OsZp|b#>>O<Q{+UQL@c4NX#btHy
zmZw(6THFT^%$R~T$CKxIFju&&7=|ux?;fv}86K!_0*CIV0F3%Vq9o<*PAolW_w6^a
zhhj4<(FCMh4%^}56SJtVPbO?@(~#XNye1u(YmhgbR%Etv7UNtLwutljjbkDUpn%C1
zJ~&!%Nz@GiD53PD*!;LVTrPQ{hv!ChlK%64DPADvH45^m_a$b7$P7Zqw6_|zL{gy8
z&E}zS+%L~%s}+KGPx+fJT$QPAc$sD@)$X5IiK}6cDa#2KcG)UFt29;|D3G8q_0LjQ
zLX~-2VpEO2G~(~8VnQKdS?_l=i6nel_Dbszq%U1s_7DiH<k~*?>*g3^p2Gg>KTCXV
zUpM>mccPnn-QaNUr0<_I<?<;o@yIPY@*rlGD~Am@TCUYmP0Bh55#%1T$q|@x2x}Ox
z3GwYh1mxE#Fsb$!)_ywHa=u??#np29k(JkL^p0m|tyUQ{XE!4@ii24DzQ5)&#U(q0
z7VaTZ9gECSW^c-+WeDDt%2nfT3sI)e;;lIOscDb~IoT)tiSa@hpV(_BpFa|>{d@b|
zVMp#BP2zk|r0~_xfm{Y0OCE|drb)2z6WJ3C*h2MOgN*M~PM_nhTI<T9bdU=|0N6<P
z+ziXYN6bWvp1Y-3=Dx8S`cCV6Y7BM`4RP!5Aof?^uT^1ffoo89^O^O-uz|o-i}b9(
zc7PTuFEjD3W!u^*UL7%J1TsiQc>RGl>ZXBx-CXbu`wW+I=63ZaiIA&~(~yz~q-R8-
zLB#Fr^btB<$>!RA{Bh3%?y18%l3H*W0DxMpkX`j1tVXEgT;O+&gGbZ#8P;k?@jm;<
zEn$D3yJ8!d+ycg10Fy<{cL21R{ezg}nRXo38l&EG$1|jR$Od!=L^vX>LgcH7{pQ+3
z9tLOlp1&^viy39^ajT+|?{m`ComP0;tMF8cl6;IWhlJkybGkcH-vp4sL)NnHsT``>
zCq6wir;tTUM?9?Hzn6s1KYNMNVepb}{(^7GigYv;MQ1%22CzLd1TUKigA-)%Y`-eV
z7Cu~c263$G7^1C|+_Q==yc3IGB^H^@Z;UVx*xe7l<F?oT@S%C+t;g~e{zj#KKTq5F
zX1s`l`joS!DTY0NebSLsSay4U^6wB%hl2O`!_=>2HAlY5@&N_Gpc{!=40T4zKc<DU
zEgUqvXXV;K!$dfVQYb^caWm!xj$$j?U=LBcgRATJ9ffrCYL3QVZL`DH@6W$4-}=4n
z_M%na#M+yM7CaIhaGKm~mrUsI)$-#N6<5dzZTjmvSa(hG=sk*ZnqXgPgVQR0!tkiI
zL?xi%C;E7n9tgXP<D^Kj>^iScZt!VNO^pYzyuqTu=c+5Rb(j-$jl>6hR$B{pc{@kL
zIuY1@lr)+i|7byVyDD_gk*W+yFxLGj`1bc+4gH`-|80@&;3Xvosl0^nd+AK6_%-s)
zyN6*-P)!Ykt8#m_(l8H){ZJ-B{KhL)g|M(I{ArR`PZjj)x^zwp!nY~bvVA|NjDMxB
z$Ty*W=p{tdlIk*Usm>;i@R#Glc^r!X@!(RuXDb>R`H9EDI>o*DdaWeEL=lG=<Ow@h
zIswgb<k>N}{T-j$lW{eFl4xkrYjpeLe-fFr1DckS9V32%lf;2c1p3YAo!V-+%vihj
z6WpSK2dYD6VLxluY-J}O&C?L=Ism|12^fJIr7ppAo!pFLH+BDfT%a%*1L-#HKg7G}
z<-Ua?FEc7@SJ&}{(YOnMCG8FW#%tF$IAdPuib~ghBV?4;0^DQ$wP|#i<ETs;pzMYZ
zH9J`{yzZKRzL8=L-Oi@Yf!F8P>G=umulDokK6>5ezR(?03nw#c(k#7h9+8MBbUpRw
zkR%z^`bT_D)Uhhi8zBq>RMu!Y-Kn&s>FYR`Po{3-QZ$GedX=v+Em3pSbWXmo4AAld
z)FY~mH;l_cfMpJSnQ<qs={1U~-hd;Xw2kECC-bozc2sSKfcj?ZIvkL1-zNON{N^c*
zuIWBcvHF_Ed!OND486d|i++M!BS?3D6BGiy0K%!&20b_vl;4{BELeF%e$ln2?n>Q7
z<ao_E1@%T}g|nt>bM!?km4;b|`OQq`S$n55!H7gc@QvAFJ!9~8h~@~zVJjDl`YcSU
z?a<HUHd!}Is?g6f$YLur)0sI-;drGq(hy865~3GeRuirJrFh;aX%whk7Aytv?WDJ-
zNyO~1v3rvXPW{RQoFozNW7IzX6xYd;jRzpq^N+CgtN(GWRxb*=Nb;UeIS3@3*Ew4|
zur|Fou`}N0_e6+t=#bE0!=83>FlJwervFx{+S}c0))?Lgl6=PdLz?*MMoe|DDVF=7
zC^Rrk>ap(be4qNMCZeY`K4OuY^8klqh{FRtMwwhMkWEAC<nPQ|^6fO%Y~rX{_*3(!
ztgwc@Ab_?Jy_1@zO@xZC38n5!=0n~h=++ie(XdMD(aPSLmEZk4U7n9KQO{T+U4Zxb
zT8+*MU`lRqJYeH%e%?xtWT;Cp5yNzr?Y@2H645-XVG<{6>?M4g5xxF~!J@&<V|B~V
zu664A;87%O;u%^VI#Y11YXq4Le$SjdNVYD;+BDHun|#Ll0j2}7T!oawX)`tW^cQdz
zEX81jgLiwqH6L)Q*7)o|q@&a6ud_p*X`<bi?sDEHn*|qzfkC=&X3)yozAyHNi-z3J
z(e8J`vdT(?U4jSmB`1?sLY8QM>zk#aV{9Ec96z(vKA+1>^3Br|9pW8+R|#700>==O
z=Z2kV#C!weq<SjI+=9{6#{1QeMBT5G{+~$=+~z8h+taQ|+ub13*Tul6u}OcyxHY*t
zzoxtm`CYWOE#H*QJ<1gd>#%#8`tD=*F4Jz5HsK5JDLrq$aWzZ#v{Tg~3RwzlfG8`R
zqpq&}ss3Bgv~0Phx;Y{<IcFFQ<$8;32kB5%1sYq?F0H+eTklQre*4)Bn>8AMs3u$1
zNuq9kxaLoX{3JBLuLD-YJ^5Y#T4d=`we7?Ge-}AW^Sm8K%?<Y8=PC6|z$1OkA&07X
z=ME1?fi=G$yh|eH^7~<AkO5@OV)1;OpXXZ2^1QKZ3ukhxX_LAD(yywY2YC=jQRo&#
zG;l0L2y&D*Ro^m$0d~DN0=koml4$Q{)@y0Gkm)JmrfbpbxWtjveoGic%bK0nsagp@
zdZFp(*s`tk*hJ`7u*Bs*YjnU!?J18)pj|e{NsJNp*b6<|=6|L>=`XP9uTDqIIqYq=
zy#J1<RZ62T*h}u$=!@2POVpT8WIg<+K<}toJHKgi(HB+3i)b1@*Z^CqFpjp*eQ}Hr
zhYy|)k^NO$5%}FFwzx~`sbyjuwa5|8Wo-d6Zm%@`##jp1okd?0=>dwTsR*|l8|)Ec
zyng79>B<{FnLDx%_jH62n?69PXn4sTz?P-8z~wSF-<D{c*~#;0=f#t2kEw)$v)-q?
zvjxjIUI0DJ6)gud5HA>YH$YluFSuSVA!mX3eD~Ah@%r|<2?NgE3J<TQ*OH9Z;t!?R
z7C{ky?H<1AHb%_qN1#Zl$Lf!+v7gmInxW_H+7#6*_^8Se65kH%+<<fGuwg&sfh&iO
z_%6drL7FV|l%wwYaLis3*RnRFb_W32Hk-^7Zxbu35u>Kfb)y!OEQ#J=J*e|lhimGo
zx6R{K@itsCtL)D7h#dDX-|_GMzPmM-({bJ1Z7<?4h~mR5g8WLR(NG1ZToNplREetb
zU9{)D)5bhf4yOW3uNjB;*@|wgt%`j?)nmfj^e0b*Or|Ge5uLMXPo<X_A%j%ew}F2h
z`Pw~mE#LPm>N`|ql|s!1wc#;k-FVCqf_={^%SB{=K!a1b<J4@cBI3wnu=8UX81c1E
z`AQ%HNeh~)I$tIn10RNu)Y^|zlVkM7QXV=9li{>O#{61v)5LHTasyLeWd1mLzUOxo
z3hbQc>*VeIIu~2EgGZ<0J2;y$<gp#IHlg!&UmZp6<?2<I<5JvZVm^IKn;ypxM-)qt
z`d-6dcmD!&Fib4|QTq0<u02{z5g9R&8*YV_`0VtG<Cl4pjM<~QmxXc^eh{ke)my)e
zBHuO%BGCwCP~pu)DNW7rKG1gf)Ohn59V)}7dt{wZYk`)cLsJ@5odMkepNKQUhiqZM
zM5+ELR0-J)?7DYLrybhyT1$Ev1drk}58y$vd9dD@g1p>RU+)#+N+B!Lt?eT>V~Chn
zA?fsV198KZ1!rJ>x~yMzT=Ba)G6NBd0lU~tQ2Ol#xH#Qp_KJ@USpLG^XEGC^Pn&t*
zsd&Hj^~e&2NZo1T?P|h<p4=4Fr*-oW^wL%gSCe@*^7I2WYR8LSQj=c=`?M!^HtvUN
zEM6ne*{LtuNNj%0^y|j`)g45#l543Zi17-($m-AH7}VFu#?5GNaJ>Hbidd{J;0D*2
zIkZOfkg+15Fk{gfr!OzhaJ&z8B`7FYw|jbmBkI%(30l^{Hxe`b1N|V=c%oG{Vq<yQ
zcy=?BYnT>=g<-0+VN=J}tLZuIW*ZgII;G?>F*R1VwF+KFB{`837mt){-6&#rJO+jT
z>pOiyyMdrgK~*#(T3d%htX2PQ1aH)xAc9byN$6vZ7;X*6W0gg%9zWNEtUL2{0v%al
z6yT7qGg8U}Q`L-5U5Pmb(ari(ZX@E!1JYxeutT))$w*zDqT6Vx%AcP?0gPR=;`&!n
zeVH(y*pHdev`-)N3!32WJ4`NJbI7E;@$~DJX3z+oSnoI?vY$&)7w}x3?;%WzI&!n4
zHW!sY>eZ>TF}4dbPw!~e>$qD{oTxkL!PfQ46MzrXVtz#}<;?;T^({E@LMfc4{0Apr
z_UVOw`lOW7L@`=_bJ>xBrTcRmvY+F?p}+6k<xa&;{iianKfQ~xry{kB5&P~m{7e4H
z>=iu>TvzN0*1OF5o%yE8EA3u9on%jZljJ_zD{rDFQmSv~xILohRs~QxiEs`t*Fgnd
zowZYBS?$!p#6;un=XhamsSH&wVY0P#aD=EfF%^##Ug9ctzp7JXBOo<0*!y>_!4Qkd
zjb+R^6<H%6^8va%$kCcWIV|aC|Nci}B_K21lh|K`+HK<r%`-e!3tqc*m`=RuE~$h9
zsMw4se8%_B)Fk`ZWW^Vuns@G#HK}5JJmw#;SW=BPX>9FmEiq`^jIIqnjPmFiugw^k
zygML+<-gxwTaW39`4-tG<h(IHB;@4sxWAm^Y}e+4Bd!pQv0P*Y4B#Ekp-5xA07i0V
zpffI07=8YIkinsoL{n1S-m|vq`r?=o!so#eLkgFMEu^k#I{I(2){L5#9!ulmlz~LB
zQ?8;wJTMRRY^>OlX216rIhMFKzt3gqqAI!>u`icaaZOnDhwnEMD-*^=+BDGUeuYXR
z?=|(v6nJ6U*7CS^MvvvL{jPYww92sVjuQ9Uc2qyjHpRB4p!VFoSbSG204YBlL(;>Q
zD16`zHKOVLH)KfQ7t9P-pNUm;c8kFQz)Gph*3cS!Z&F0KsJIE5Q))oo^7<)#T2@Wo
zQtu>&{}l8OXY~w%J!Gov4n*JUCHoGDC!(S}??XLNaND2Rb%YWk!Q0)y>l*Pm9^R8>
zEH&kj$G+~a;Eo|CFPU8_V^QFW;dx3^R3^d^KZ0J`h*=*Rp036e{uZ+yGliy|ONaOC
zopxL)G(rA;QM~|!%{plA3?@`}BA!x2bM%IKhNkgEjlI&7UA_Ps7vn-9ItT}+F=>ex
zZr)le?k|JM8Nb}4NMR9004UqkzlTVTrh9;7nuRQb2)FRKdhSh*1NciBbh|^QgT6y{
zzC1xS)cBEi-~F|$xq_NrX38X|@oOWQeUtbPQdlO-OKQe%W0!zXWae6}ez6}E=_27y
zHesqEFT%cc_tCceF>W}uPO@kk1M&0iJVq~*{od48owDZ1KtuN*B)c~4+vgBz<aEIY
zpioa&@S+Sm6YJ5awMCniZ#dR$;VOkfHdMQc<zEY{{0SB*l)v2L??&Zh@}L0Z-S{$c
zD4O76tc1<k{&1~ENd!yOP3M;0@fb<+x<v*rg{Zrrnc!!hd}#YJQ!c7G$2t}>kckn~
zrw(&OQ}j&)6F>bhqecX!O}Y4UJjIbAjSJBGT++)}o~jX5rT3<Y#onGmni+*#i0QF0
zUR?<}if?%3yJ^7Exdb=CCy1PwKUQ4>P_5bV2(8e!;W&GMz-!#?ecZk$R&v4P`tS$y
z(dY>+jyI8C_>EtXmUYOrLgz<ToVXSe8Uucj_u1kjT{s6!(2ev0)Gt^qBOSuna=wzD
zwfYMp$^W3ycfbLCUH-CA+U*N=@BjSNKsZ3>mQn^IB!6dq|MM??x9I4M?XhQn4z6Ma
ze%L-z6z?=opQ|y_<Y_tQ-R7e_>Cyrqt<NNQ0EmYqIW4hg0RUG`0%K|oCZ*X(^c>lX
zpljrZCY)l&&Y>gk`~1-b%MG`9{R*sjl<~D>BU+>k3Yd*z=#9R$wFzd`G?n4cic@zW
z8y-WVxX?ouUAv$61rTRmqYszXCTi^sc*lC<1BCa!-4y)!E*{yB=PnQamD9(2KTRR{
zy(N-ejO*vTaIq5PfT+0<W5Bp6=syXmf;Cd<rVl-90)AyRZPTDVu?#@Rs{jc(w`(o%
z6eVd6ZCL$8oEwc`iun(D0oW_!<~Y0$B~Wp>OgWE?0f0!veu@P9kXnyh@s?S+B^M-2
zp<*|P^WBW`T|XgPACMLsH<^s-YX((h$Dr}rQkZYz^Y|C8hc{vrI}g1<X`~z|>#u7Y
z^bcreyahPmC&bNb2S|fFH$2G>HG<zrDB7a+5OTZjDL`-Rt4NW(1gbo?7nC;0>c#|1
zt*ZPG4r%Fr01na-wl!m2kV&(G7nlDOfTIV{{iFkd+Xi}LndzU7f9V6AJWIG85DYTw
zlD0v@l7egp98G)Q-Okx6xob_M1$gThm_IduqC#lc9c<rBsOMMbUJ8ap)FY(<+S}BC
z#rF%>0S!8EHMI0^fc2kucONbcW-f~SW(s%Vn|dCP^sH4DN(R`JArTm>J_X@@HE?y(
z1-j#ee0w^1h*?)cxK;h0xhhn>zd+a%Sy~tQ8(>HCB4^z%I@NXw?Yb+$DS^UzjsnuO
zkLnPxG!x+boFb@5d%@;HF3d|Twc_%S6WbP_Gm}Q{;tF1;+6grf<k2>)78dCZPKfI4
z=)27Y@j??xPW!lm6v{sR`q1DMfJKs@8j-3l4f!^jg))YV_nv$Lk{zm>*aU~n8qivH
z*I8;k&DjBf^?#18w+A4tgyvH!BpWy|$J}rRp1bQ1eYy2i9lg&W0%hlnLmK}SX%&rx
zz_zF1eI0Q-1^WA|DagKYkjAAp02p#jrap{<ISqM8TVKKv<OE3fHx`vFz5t6H><uV(
zyM-W0D)#!7EUvF}aN+iOsn28@Y4)ub!ewdUZMq?m|Ao}#CUlQ}l=i<GZJOW;5~%AV
zGHA6n9n)|jqQgkuXn_8|y<v|D^0ka(%qX<X2vzO#(p)Nl?-FIv^M<u_`3E~kcAbZA
zk0HD3n9z#{<!4Jhz|pxg1L>>|1(JITlmHvRSCm7sbOYPTEJ$h=$sTM#m_)V`bNVRQ
z)>sYQ4$rX>kY;$PHu^Ak?D7@#Q3$K4F5T=SblRQ3(<IgUZoXfR-o^vhO-f|Mnn#Rd
zdWP1Twa6#B6>LLOGvnsdR1Ei1;bK171bN;ureUPMOIE$#=YrG~GKzdTl*|X8AkQS9
zWMJjr{{~%8-yl7hCADhp%%AmFS^@r13nKXZ@;ZAx0r_MyXECDddaG37tcJl@JIU`c
zsY(Z+S*1vSIapWEYFjYIK<&p-J7fA8jqK8butK_!fS%WOXKTsPvu=TrpxST1h0SVs
zI(};2TX>+)51_&>x%>RfS&g{&_Xj}@;oJ_}mjaOZq**Mso#_5%mYN>x?0Q2QpQE|i
z;b3>$x^gm5bdZtD*51U0PzIi+nG&U6fovh2lou2N*Fa-HD_Yr43L641!v2udxD-)&
z_<FVL^~-&xMQE;wNdM@DIBD@u?2TFZwR_)XLy(9_9p%$$b_PjgWxq|O_>KVVhnE@`
z5)N=sAS~*Wk!l(FrItJQ(1!#9^&vR@5g8~a^#y2<#9SNXDrpU!5P}8ko)~TlNeaCF
zun)%xE==IESeb6}oZg02SPHybi;D}#gU$3d(hfP2M)<zL0wBquD7r-GW5XJwPOfva
z8cFUWkJ=3C<A997a6S7nv`y~WTDg+@|FQPgQBkk$->4!eFd!WwodQFNbR$T&z|b&&
zpoDZQi1ZK&(xoUJ&d?<>w3J1MG)f~W2)y@qpL5>de$MN2{yJ;zwf5R9mz!^0_jP^h
z>X`h?fec+-61y%2%lLMlzu@y)*(V-n5H5Y4(W&A`h=2Dq8XP$H$nZ-)K!=j!2la`c
zt|bkp3gR6lTV-HhiTA~`YmE3ngFhH;xz$3?H&ZlMQDu{#Z1j;~@efdwNplf?5jA>h
zV&e*?&v+w`tP8++<0gDh^@Yr(Lj=L!pv3fFogI^efw43x!4tBo<Il@t{&rrp?Q6fz
z3zxb2c%d`;v&<ddKLf(U((Q|V%LJMaUtj!o@bFp;0=nQUqt=LKLk|V>jsuaI)}B@v
zL=J4Fkn)dN25wHfhe=F(aff3W^KMo&6$gRBae2M<o9GloH#LLy14B3(oZ}+y8w98|
zVrYMdq>L+u5TeFCz1cUG=l{z&%Bt&tL&dk@e>f#iB&rkp0tHJ|Dgvy3IB0~!0NI5Z
za^sorl-g<*p7Y7tYAWD!?bQK=ytyVLOhzYw>^eBSn`@KMk&IxzDO_RLBdVTH)Zp*y
z;It7SiF``-*B0ej0Glqx#9hWGry;Z0{|iwjP#qwBz5xbc93WFR&{o@Zvj=uq4$m#%
z`5ENvT4*UaOfVaq0C}$ub8y*xopPM9qBCA!r?)3aawt2%6*G8}4MBMmo-i3tSTh3e
zo(hBa{iUw!m8*?@4mpBsV_~qsT|XcZTM&vbTyWn)2HKTtEkpdC$wScWCfB2!+_Vw~
zQ%mkQaj@~<@RhMdRxoZR@rl{5egM-rGFKLGa)=(=jdZRbQMS2KYX9vBDkduui+1A?
zQ>y(6%DQa#E4Kbe9>pB}d^b$`V%Kb`fRK#3{G`C+s9<s#_<^WZsoOsk<U#{s*}6`2
z7vqhnKAx<Kz7);Ii6we3Me0C%;&7h7!1~bJUk(U(UVisqq}8NJy=ws=6JTjZd|~#r
z!t;e)XbgQL@LrMPBz%d|u5ysj8R{%d_<IdlH(t6N1BLof6aS4EAos&cH<BOza{EE^
zUuvcp4i)zXpg14Thr;-y@)DbpBS0_RLij>T=5QMPh*tH-Tt$gtyjU^=uv%3lrl)Au
zYzV+Xr3f5gkG-XmPHvO12i(jHF9uaLkwmy7&a&hLfiLv`%e!&HrTY@g=voJ?pZLHC
z3w@!)V5>rjBEW=l2-sslEyBYB@HX3*9wkTq_|MtVcmGq~U7FitP}THL@pPH2CSVK{
z)C3A)Uk!FF^CDl~cW7!l(kVtq`qhKBYt<qUaJDdWRA^hIJqM1{Xx^vA_2uYbj7M}Z
z&?A0;5{#Czw}lR|V1o2;AYzi0_6uq?&jy{Va7=sy;mQB09Qo%bS}!2ULF|=0^4&dP
zhob>{N`&&eyVa)1Xb$X0-uGD)mEyw`ieU!Y5*(%(n?OBn^XGVXqvFSHkELGpQliF6
z7@yLe`VA1DzdQRFkYb;Ji7HKk)~NOT#HsmvrfZ7?9Nb6$RkGY|P+?<x$c7;pyLmAx
zAGBR|(w6WcM%Q-W2i1u6KL){QSK&NWd5pR4ROcXc;UjV@S_TW&pd^po1HOTd%*QS1
zQWubdJIep3WR50>A9-?RSBLrC|8{j$|MYjcPfxi0YDd%9A=~qONofTkh<b<#{Q^R|
z>1shx*DY6uC;Ok>gXxk$4U#Nt+G^CpH3dBAa|bi6NFJY4(C6WHmSfrJ)P5jQna<?j
z&_+RkjKYFXF`c4Ae+HzN@O8tXajF9-ODvTI*&nb_`>K&X0R^uEu;8g<+JIfW;tY!W
zR`c0UI%i|RQSE?=Y}B76`+^p)w11-^hggh=LGn%z=vF{3hGkTq{h@d+hFxg~0=DwJ
zAJou%QTQ(}yFAX|k17}om5KSM$|)d9GSHFel084R_59rT09ZUu!FRLZP$ydjR!azN
zX^262E|V`<?hYhLr=jFlQY>me@|3>oplUj(e-0UDdQSa;@Mv%AEeK5+1+w2M@OBN_
zE~K0FLn9IOjU~fB7gm5>eF)I4T7TsAo5Bw?lX)2HK284`mgaSpUbzu77%tibzCIUl
zb~Ky->HzIMW-_+bbRp}>D+iw8?=qm5S}_WBUw3UE1#`Ul8I}ax`9@+ouTZUweTR7#
z&wlwSuz6!3TsaVU1xo03q^5bkLPL({uGr?6zm3tLyUxIU7DIjb1)q8632@x37}<$-
z5DAa7qToozu|ZQ{@GDtz3I4OIvN!IvvX!}F<mwM7u~aZUc1ZJQgE!X<VVhyrq^#Fp
zzDl=0D$Sxoe<iFIe83ZcN9De&>*YIhiycd6ld!Az_cCyqEX@N)OL08YgXq!!hi(uB
zM&XZi0tf%)D!_%Xg84K9|F0t?^h4ceVBR*IdtKcBU;hY@E_(jIbaemv&;K8#i=d?c
z{hu(I!M~oP|NO8>9>7REZlAu{|NWH)ZSM<op~QbFS>URvRKX;Tciq(6A_s`&AZ^_t
zkp4_U>Q^9tuY?qV)K)?6PP5Ja-i2$3E9BsXV1jB?6#&q?z&CFOFuoaQpks_O{sX>`
z4RA_hfJ$S6_kA-`c<@dV1XBWuzCHAgLn5ckfrqUHckwSyIrfS{eogxKeZ9Yd3$HPd
zSP^kAkMU#|ROJco!Ym}{gj_70@B@ICsF=ZXKuAyUqg0U<-K#c0U2Q^Aa!5o7iNvY~
zZlfSm!ga|X%%|XToWXt;pP%`ulc1uP>_2Y=k(KDX*CpPycYT2B)KCJnIze3yYQ_DW
zi}O=TJT(7?cgf=xFn|~8Yy6WBmj0mi_W7d5H36~}>832aM+`T?9-r8hKp?<WoQ9?x
zXb02-7+<yS2TQX!Dy|(+vpa<(z##JvhnGxkVn-$gGzR5Be<$kriTQ5qjP-M3XB1y;
z)M98G$hQ@Q78%)s)#uZKU`5QJ2+HbR&Y^b3oxYYozciqt0F6@#XX;uP>sA+)z~GY+
z_CZO>>@V+uBRwvU<@nuLKajV)Fbo2yY{1bPHTn=}0h55Ay|3V!nBw$fs2RAm#PUg(
zbl2^u8MHkcfS`#z9W`|-zeUc%UVKOTvbkAL0@Gulu&see9v@JYZ-%Z{_WNK!OTcfs
z&9Y^qB>@C$E0SKoXs6WQUpE`j4-sa^4Pb_CSOl!@If#6E1Id||`Ppm)A{GDENWIo0
z1M??|&^quVRf9kZmC`%Yc#t?R5YSkg;Bm(1^?{er6?j?tD75IP&H~nmF#FK1j3zLb
zd-8dF2N1b>U(92mpNeJu41orvg(So}bC97P5+s7LGG`jdlAJN()IgE+gn!ghQZ+~k
zz62TG09v%U0Z^SPK!Wo#xPIc>YOg<^1<Sn})>X&*-+a3Ew;OyPm&a)QE?ff0#^Tbf
z!4r+kWr4&_U9_H{6@3M%dWG&yfLC;;==GSd1NuHRy}f;%J}EmFawTtyZ()_n*qn@y
z28fD%J8F`ah<F}Ar@oi}6A<2E`-N~iUCrYqAA9{cKq;NSKRK!a7B#`ZqK#qp?zY?8
z#f3Q^v!A;~mhOWoIcF+WR{GCQN5ttqm@_Eq@V^1%-br%$wHZMmag1>s>i?`V@*l{H
zRBaN%a62ypk4EmiWlW8!=+y}zXrd^816fePNz(xN(eF7|6_}D1pv{}X@Z(Vzg;v2v
zfN5d%@MlLbf%T>Y@O$+Vk;N(I#Q@{kH^_kseuceY-ji1Rz<PdL6EQ#>uN;o8T&ehk
z|Gs`yAM_(New!fTx=39Z-zjz(ZDTwIm@w+_pWxuGSN{+($pRq%-V%$y-1DMfs551G
z!TgFjZO&3XsG&@RDg|LjxB0y_n9>OyI|)d~_1xJzuDf;7t)LpUkXKbC`cw6PxKAn>
zc(6v*RRnIikihO`qqzFOh1%F<wr0p24mz0tFNZBKX+P>FZ3k;yMt&S-P5opdZFVDB
zj>ILckJBfiF<7_>rGibY#jWfC?aIy7ZjMxQXn%omta3dVI6F@t1K;WM{XH;hRs!El
zN4;qHM(4m_Sq)>B6_D|WzbCut9^9o*l0&Qq_Ch5%=J(csy>L{36q(2q;SxGiMeA4a
zZh_{;9=MOptNHC1-E?@%s}susAW;iiMRkO0RJ%duUrwzT_<kW5F}zz}Xd8XE8UHY1
z<E6G}&C^OeNcd0iXn2tfG_IjAF2l9^C-G1Yv7dCcUjq8p;o_)ed-+Yzx|8;6Ot=ZH
zG}DTFcR-F%S>;EbcHznh2sFys-aWzs@av}eAJi(;c(*H~=S<8?t6Kr{@=sS=&tW!;
zfh3Rhy=hGkw3y7TVXoe^(`R?eKp|)^hd<{!!v~?#hl@bNLBy{g0YL%u?ww@q7Wc2L
zeygodqdEm5<SD;6aQcY)txQge8kC;w3rpSO_Y1%Hvntmh3JItMAAKwO)a1Jj;WucY
zIOq&Kf;Tou9-Ot$#B~7s88*vvC|%5TfQcsTIn2ZV)WVL4zhKq-iYq^7&+iU@5T>G&
zr2@?rAd4t!)l-MxNVLyEfF$B?<La}AN^J$7y$gR-f?90lHa3~%CuoBW<Jn}n+CXJh
z_WU?9MDfiQL|u_0NX1%!w1sEY=;QU_keeXhufq+TCwRWw0~EHkK-@Ps5!wNOG$ztZ
z+@RDdQdTEk`jPJ00gfar5MJH@_oR7l)Cdy?cFYK50mxQjmq_ayg&b~O=f|e2I1`XM
zwJxD{##8{vsTu5TmKZC5W@tRvm+5*=oQR)v)2)vI&Yf;RH`#+x+=r5c7$73~mTW`5
ziOHq|@8?=G+N<Nt?AmcSfjY=U+a<L-3+6%KyzlsY<=WM-u{(_W#0tWLBN0ECLg&F-
zu(4p)KkYrkBUo?f<feh0OUZ0FE4Bg%1ZzW6-D-|N!Uav2ZJ2@161g;|O+_4pzYVoV
z)YH^({{=t&D)lUe@{rUU4ZrKx*0>Ik(TdygB(mZ>rgD1MP<L?Qgrk;>!hAUzH**xk
zw%#uVlWv2)IGw_v-a)CdjDrn_b3u#T2|fCYHmG3=dI_HQ1obm)H>!`_TFECwSlUFV
z4(Uv3iOt+G{R~btZr=e4FWfSoer~a8x4W;;^Ha^~@a_${j}4(G9%ObMLXg&+>xf=k
z=dZ+dq~1zJ-twK?(mkL*j$rT^XE9Bi0mY-?jQH%bPz~5bi2076gJ_UBo_lUnGh#fv
zs}BE4&C)1Jolw(o%{Tpej*O(Qi;~1}0X^yREr16<L4b+Jv*)M58+4C3Sib1(;hiRi
zkSkcIX#N~3Fhdf$0fdN&*PVF(a-&#vb#Qs~$M0r!@KMZP;=NtgM~HG@9W+T}IR}H4
z3MQuKG7P*M4$?IL7qYeUE#~wCz0EezZxEa85~mKA6%nMC+23$>XVKzsz}LdJRV%YW
z>5706^H%R?>QFMdj!2SW3M!4u2kcz!i@SIU9I3|NZCjK}B`YwXsNd5Ywrx<RK_0qa
zUmYzFRB@BKH~j~E{Z5{1lq2B@J>s#`q@42u=A8^Q{a=BB1TW!&7S#E)9%>tkXH~K=
z#m5<^59!H%+)H50ZX(Kp^+@n;2r&WppmmPmp2u0f))p_7lJSah<5lWsT{)lcTm7wH
zA$`_J1`Z^Wqwcryl01ahpl(I+?x-ZOvBEUJJGZjAe39x_BhaKsm~PvMgiF~OM13#@
zC{!t*(S0>X8&pxcw*!=m`4;t{_WGz_wH7z$#{8GHKCv~?y+7~;W>!vFU}X1qiKRQx
z-Cz>BB%`Lzf1NRpBE|P1d{4V%o<;{v9^MPldl4B2fqnpX{rrTsBKESaUKw-jJl&v7
zq0mm)&CABiCp@t>Z|qkAK*s>`*3PCmqMV|-tk<qgm7#lt4aTp!lO#2X5zpha00LsA
z#vNc}qYtM&7{30khai2%13mdgeo~S|L(*u56QH8KazL?k0pLO+1ObP8UNPse>RS_V
zoe0lyo5*FqHhUmV9I7(fk1c-#_E-9QL+DcC)FUCA$L~KJex&fy5B@mAN{D=tUWW74
zzf-;wk0$jK=-N3pEB2^`h&5yuF0&df(3x%xzlmg|L;C>35tGxCGoVY5q4)Zv@RED0
z1F>yrtm<3|Ci~^?OByZi7u$+CLu?Q6^6pS40==wGQ~xBGj~8vqYeqlT3`>Vxao^Q1
zRvYj5@~MhzwMKznX-Fe{oteIHHGz-5YD~?$=C7<}5vUGShrOaP`NpD_2y$}LgG6+Q
zycZx-zGds;WkHvmJOU)riCopB4uUtYk<9a$C~(F{=>e1|3YSiCoHrr`@%V~NG47K#
z1`!h?>Qb{U*+e~N>-^J_P3lr?S?R-uvE?lGZ!dLmXcjXFZ^RlP(~w*BQEw0fET2Y3
z<LHIZV2<{uCzZy8Y36eRZ${q$3f=+1`$3@a7e~tz$%ZXDGtiEXK9L-IE&JC=jgkYP
z(Uy%jak=4Pq)Ma=szxE!$+%RN9so|yEDlJT9%bLKAuysfBeo&A%$bbDTE@Il$u8#D
zc)2Xrb+Qi9QFDI0Ix(H%Gjc^Fv#8oH5{}kpPmXdu)2MX4dk2!sKXOKXE4l!_g?4O1
zAgJ6bsSAzx_Ut@rMipux!H-&O<NhNcDu{k=2OoNn%FnqWl)^n^-<VSSmp*|R$Kc-U
zh42s9DV0uzHb>xZNdwIX7M!0MO7tJFrvvAf(3@tE#JdGlx(4~zC$vE_BZYuSO_)q#
zj@S0AOm_xE+e)|0E^U-|e}^3{8{XRO2c?2(1%NXl_=Y?~7SRUl%@A`FKgNDVXsc`a
z75g+x^TGCaIYIsjuvfNFDHXYOMWb_)QmF$%eufjT8_zp;Pl7=PVs0IXGdtTAjm_Z%
zNw3q_h!~~C|IDSQ|BYIg$6ZU1Pog*^Z9spytINE&3iNSi!PRsRHADyk@u3oIlm;eQ
zkG$&@A3M1<!<VjcGn$jzqG>xq<8Rl5eE^6b9Iz`areiP7+05Ly?`^OqIlI7&?C_JJ
z`Jg#<GRJD_B+9k3_xUoke0BBE+`^1~fcXLYXgIL6s+664K?&TmAx-^7$wq?u2IH@1
zW*)UKUoLSvRepT!>cgOkL^PWd6(R}U6EqIuB=#nmOWj3hLFsAo*~~DGbUy>y*V6Tz
zzq2^s66KS~525cYP3cNI5-4|w0C{ArM~?_1d;7i^pSUYbMkyck5E8x@^}K9eoO%vm
zh9!hfqAl?SZPA-NMh`%oT%QmRD1)<^R9+r&$-B>5Sf<HNw?T&xZRFMYPzoGwwoHH&
zU(Bju9@O&b_<-VXytThIBV{dOGzG2eb^M?<Ist}e_80JXEU>LPz#KhorYYlaCsUCw
z=Z6U|_C6T~(t&vhQ)~qj>IN$TFC_b{l1lpN|0bN85P*SBvC66*FBMB$^pEgB0^Yn@
z#uaC>IY&9FmCKm3O7Aq9*kGRha3%I)?KftESSJA3d-S-Ua;nArl~|EKBX_)wWsD~5
z{Sq^d28Okifr6OVNz*qlMkS?!hOHff_T?;I92ynuwGd$8zx$Ik8cd1%pTI_}-Xf5!
zrUzG71hIf}S*x<qa}l~3geeJjue3I+W^5uE)?VEY6~(NnyS~9!VK<VG;;3M9v86iv
z(l1ch-_uEfn0i_CMiAv%p-;?2TKEt2sh$mVPySdyzl@#;QYYejO~X<W@ng6j*u3Et
zYz1yjN#+oiX$!y(UnZ+SL!llvV7N;V!e7sb%0t9Try&ni$cd5jHLf#~kF&`7sz$Tz
za)x4p5y7%xti&`8aG868#`tjj9`In}j0UhP{~bLmvcjPf*kB1EO^w~C2lJOFWj)|V
z&J$Dg>g~XA)u`tM^(T8}qiZxS9d!|1EW4jo2VUSdu{ZMpczGTiH?ah;rvSbjzmdpp
zSJgq@e_df4$T$W~Bt2wP9{htfU(=9CWPD+Fndb91S+m>M0aLX5*}@kjwbwM-lpUbg
z6qQ&l>GP&QItW>>qvm%hkg)>&uN_U)L?OUL()4JAL&~Mr7HEFYZ5dRg>lNSdnYHrg
zsu_ja+oTwt42ZuRNx?J#q0L*}=(gyc$dCv>V|84%6>Bgcu)Uw`9@98#-wS;asr+D#
z!mJxdPpC=YG(9c3lPC*pBy%Qo<=(q<;MQ{nvBp8_p?rC>LOt|Jsab3XcmLjSaC3c%
z<qWd1)a#<KOEV*$03kMz{|f`y_3zW_!|~F;R3VS<Kvz=Y(IRU~1FE>82+41q-gBo4
zEbPv=JL?W=3vUG7_!N9Xi$;2|OUx5ikLck$at0C*QLi7mo<e-v(shEubpdx0Y8?0{
zyKMOxGt5rXQ40RvRjvC#I{=ncE13Ztsj8{;Y(Tu;FcSimR$Q||;D9hVwU-B!W%r0a
ztn^V2;K_I@^qKEh^3{%Gr^819lJxS?6yNIWp0z5l7k6wrZp3{5PJ)92SYIei6W{{O
z#v(LPo`atRXg)YcbEHI_^)eK#Tsc}YdV?v4GM)f5_aH}_ZPpm`b$arPxtmBrumk@z
zRVgyo3}`6^!sR@l4Nem`I)3o*X}syRv_+siP=4F9&}(D-vw53H9b+{}yAr8fowcxw
zo3_BONj5U=Ge@IV{hDR=>VsE|LRMVAHTWFQvF0;?w3rp6oKX-50Gv*LwUxhq3c2+n
z$Csin4U|XZ{rCh)k$^?YrTc~h97TwI$weSoXL`G{>?fv;)Q0rgxAOXd@}-4@rXSjd
z^!SRdWR9*4G(CdR@~$r*xc+{?2opN&FUMQ2vD&&#7k;;M9textrY8rcJ>LHik=9eD
zR7rObp|bl0$~+(8gQ%Avr^=^vye}f{$7tvwm<PH8e~J6+=_=iaG&zGa-lZ;d?rw8!
z&kjM^Q2I=&7od)6%AW$D!K3einXslJ6{`op87j`@mqM!d)$hJ9pKn~eZI>TQ3D+|(
z;3F^Dy2C*6H#9d*5kfPx=4IBs$chv%p~Mq}X!sT@uQ{aIT+IaLT4#V+@tcVc<ycwe
zVk*6i&F`9f9gW^{1SruurZoqj!xZ<Ph}r*>1t2pzR`V_XZ-BDAJ%Ck>Z!EsE)LzAP
zGwTkYPBK$UHBWd*%Xf?}>F^Gy1q0F$!Oz3Lh>xsK*nqI?<fiA0J03Z61tXSGXy*`&
z#Z4eUbAY079Juaj7H19T`2Gu=ZUCU>SxTI5KUoDu@JZi32HTv4$gcJAsv_M1|J}n*
zWw3bC81dKNdJ+7e-|xmn2e9KP6TH}21b|euNnjtf<{vq#zhFba|C<W9fuFt`-m8D@
zeIVc)l91t%<TUdCZ43i|x!@Uy6%ad%|FC8LH-H*a9EAe?W9eMG!Px(Ox2yx$cfcG!
z4hT0_fPEWueLshQ^iC}(u(j(u%JpB=fs*wa1^D|9(${1D_pj*9yaNi3*~R(UCK#FO
z0IE8>0;D>U)nDqhCBXR)*=Q;Oi8PjiKoU5C!mzmZI^f?*-%mx5BJ*3v+aZWFmjBAU
z14eCA{B}^fYB_D}NABFquiOg#?FsJp6foD?15HXoIVj9pNPZS0C}$dsG9r*awj5Aa
z<VQSCk4ozP_ix84s0IG^%!1?0l`aSe1`>}V88Ape8}x3pX5!T;{jh_ECO%*uhZKYm
zKLugP)g*F%K7nWu2LQUbWXO9NYh_$=^@1>vKfeKt<N}6=N&p3YUjcSK&L`hRX6RW5
zz=VthN`qIB0_AY0gJcSfD32hgdc#UaX~9=-bKDYAD7e+RE#xxfFIRr{9vE@~thE*o
zJj*=i#GN?1au>f7AfZ)q0PkAezh0}h`#?$jnCHslkEy5N_T*7@0Upm4Y$ot%9szO0
zQ+X)<M;d%JufVdQ&mafi+xBu2-%Xe5mym|A7GNq5eAlb+?>xd8I090x?g*kANI+{g
zAI#cZ%Un7gl09{XQdhEpBBB;c$U6>Te|lvIB9dsF5LwGC=^27_ih3b8#9YlQ40AjQ
z0ww5AbYVD%DwZ@LMU;{C)zUkLy!<ZDiAAGu#J3X)NafJfcgsrt=Rq*zc(Z@{5hhb6
zEelZIrj<#j4d4sH01m(uoR1o_jHq?MkPab~;u)kw8{Y$m83c_yJ&i-Z`?7<JZoK<R
z{)c>9CIAv#KtqAKLLx0_x{{x!0C;h;>RAOMAc-(W_P7<&fh_KF03Q>7|4jG+3vyp=
zhe1J#ApG#WU}RG*i4&4!+*p;Wcy^7!>%}1GHFi1uX`sM`S|GE$SIW8&0#sVIAU?K9
z88uR%oOjan8$9T@B3nSVLXW3S5IC}zL8s!Mx+AN-y;STz$o=8!g#gqbyL>Cs+tqjr
z(hMdu)|%XaxhDp52!XMVGnQ(Lck&m@_WoU<*6zL5BM7XoCI;!&z@JFC=n3?X3#<D4
zE`WT<heUsRx>#0caCxYUiB@2$+yIgQ6lx6=@oDz05f_hFfumsJpbRJKj;{o>`XuNU
zY{4gEqK0}7rk}6S(cA=R&A0iC*jY$DJUk}%8yAumLQ84$oL04^Htn?f&fpi?ysKif
zdWeDr{iom&_Tc%g$!s&uDubE&g9Xrr?P^3qX&;~$dIbDFIwJ*{n_!Q|14zFYRH!kY
zdD!t32puLBVSILgod^f5V8EnrQ7q$&`wP&1YnRGC&fN>#&WFw=%V1cqixZm_Kc5Df
z0~{9-q9dncPyQ_x`SvU%m_l~2A7uSdf`slhG$5zL0In{6?9$0Za~LWRk2)NlnZg1x
z<R(bPATw7bWNhU&9$st;2PZhUA5q;LDsAprv1iZzo2LZz%lhIVIVH47xl3s8*nb{q
zUAfm;UAri!)djrPNBN*60Q+oV8KYla7CYl%?ctx=nvK(Ip=CEybJXqbElqAT(>#3d
z?>0Z4UgKo+b!vfh7Gc0jzB-bg^Yn(#0eY_Q5-SPKAr{<mVVRkyZySvhW~ikj%>GRM
zq~47yc|79Xt=uaJ@Y_gW9TbF6sS(5SV_R3fpTQKyFxO(5KN_KJ+dTjNOjvapE>}`&
z?=5Vlx0i2&gn0@qNEQg~`4o+6TMLt+{s_VSJCC>Sf|C2(Ok{cMdH5lK{%vmp)fWgm
zJZ!K%(QU;=+zO}+>7`RR1i=af%lx!tcx0utDkK$Ftqe}A<qb_9Zv8#I##qZm+WR_J
z$yzmcj{VzhkyM*NpXE%^Gf>X^HfrI94Qe!#(MNsUiT*7Fv|RJGA>P88>S1TW#gMK5
z%zADV!_oS~i9T%oJq_b1t6$IuH}D77QGF*JB`N<Sd41hS?s&hVo|c~6_^NL?5Z*Nb
z6_C)hTR8CGbs+U4{z=@J^>@%7P6i&qZsyGmW{O2U6^3M~0uqoS0g3MTeL6?qzStb4
zVXXyEC-uqY%4l-7&*dU6H6up60gUPQp6ESlX+u?v_QR`U`ReCFp%e=0@@*V<udeR!
zK%i+5tJ{f6_jSK<h=<G)9>KFba?=*I!6xhm0D(v`V0rl)&1)aarS|0W1gM>^t-s%N
z(jip5cXzUTBrW!}27kl5`H1vQ-m|v?^cP@`L@I_4WlNfsj+XzT>&Z_ddI8PU)7v<;
z2S2~|fM7yf7t=aSO1bGuS0Ag-=-6xmv?QE1Q<>U?-Lh4Gdh?fS3)uy>DcnE)RBmgj
z^OOe;L8Jr6%|P_j<4ZzdSMXrB712G-;g08tbL3Koj8~JX^c0%nTA1IJhB81#-^M|W
z{wz?P(h4W7Hfd#=Xvkiv4xbX=75)B8Vth3BN{->|awolAi%v8z|1jb9Gl<%p4&dcn
zc85LoEZy92x&K@97x~s@#%3{Ev79$+$W*esFOSIiOCguFS0wxXmEc_Q)OWYSUECNJ
zBFJy1NVM1daKI?yu$(9<QccdAlV6{28I-aG{dq5prQQ9cGfh{uq%AN4;|0?Tw$c*J
zxwv21GjRmo`kvO8fN^mST7@8y#@b)5X~1^I)&VD()GYH4zU@1`c6(bUrDEjM6a++c
zPcEBa&loJVK7e=@VWAcHUS=yt{)(Q#>g&^gx1zU_<ZYAS7~NmY6Q3vZUv=dOh=*P)
zP4Jf9Rn$XCUnIEiY<OU<j|xMH4Yo}qG201Tc=%8@ZS0dqyy^TPchkp5%cu!K=W(=S
zvBK#DQ^f&jgH|dIfbR@1p@UM|Mr8-jduNHEeNi9+JRh!l>onhro?pBVA1=e@<_OF}
zxfFY)t{L0vq>zf|5T?ATrvRBl=Qg0qk{}`jf_lcUC9Km9_i+3HvplNMI(aSq`rD=k
z0c~5khpDH(@SeI|X0P`nTYtcJP<fOs`!%&pV-aWpD529QBpgxF;e+X-MiIegl)uWn
zn*2gCz6rKK+W-VLJ|&s|9(8%Vav4`FK0gkX%q+t9woiRu1tk8pmA*>Tjx=@;_Poqv
zdVP^+EARf?Y?4NngHL>qU#s1!b)!zVJtPsO7K0e-Hdg;f)U!{_b0s3CI7YRF2gMDt
zXr!5b%KoG<`cmJ{^l3HZctFqrP|UUBhhg{J=5LARN;`SGVM>ee$hLaGA@1+S0w47i
zl5W%<0%*$TIQsralMzlnvnd_n4ZWVp*gMrgI6WqiBT)9_tUrfG>jO|4yu5NyHH@2!
zAq)l*pHMEVkJ&w^E1+Q_JP+4vPPj2sIk*_nEwu7juQnUfc7j<)CcYe&{$@vfYzJ52
z<>SUnEL?7-tLaGRA5y~-w%F|nSs-R-tC!EMQu0-hdw+c~hT~><z`)lP;ftf}yHc?d
zXL&nEw@udU-@lX<=G^C_$_b>A^Ncn404TYuG@_{ZV9C1iL(o;lFUt24mBl9=l)#MU
zzk|&LL|tv)9T7BvMo`n%cKWS#0CvuygV9*pwACcIHsAMX;d?yDE@F;n@qFcUO7m<t
zqrM*?TfcW_xyGK?@!I|r(!G+>99xX4Yw*2S`}_}aJ8Qy`eL$gI!e07m7QpxxvUdDB
zEZ?Iep8kiU^VtQNBV|SX#kzJBEE|YdscGQmCT?gx&dCN5rGE_j4Ae4kRc^SQiE~w&
zuT#;_?i#NC`WD$LKacgq>Cji<-?*59^#ukg%yuCkCy0HH+WLSW0f2VXAHa!fUVZxz
z{~@EJv?a822j&Vaq9OFQdP^xPHVQO!q&G?H6UV#VbzDiOL%H6+uO3Dw1hPJ5^ZWr)
z6Y)Wx?7zhbaOUr~V$o4|&EJ*pR{;rJBqTuCmV~3zAG?ZB;mMzb=-#7OrS^uq8c2(F
zUhU*0gt^pd#vH8^&@6~d6T)N2@Jj}MFFhR+cbfr}`mK6zD|5kkS|Qf%vUiX9Ob5s<
zzM`U;c9F{G4Zv{6W@X-vc-MGfODRgmRfR_e-)%qpAUf7tD_kSjk^TECFydS?#=TXy
zL7el0g-|c&vuMl}*9l5Ij^XO;Q=H55mlSZ^?bBexYR+@BGZ_N4s43=!QApKBDhe@~
z;-kb*a>H)PBdn*%1W9Q~yCJf2&gi*&D_3pw=}(jcCZ6@0C;Oo73K!N#B>@G<CjZb;
zAdp~K4L+3RJeL%TxNBWO0V6*On%*FrviUfz<}zk9{=Arrfxy*-x^>59i^3N8e)EoB
zO)lS1U6lXs+W50zJUCuhz(3b5i`JzX_Ql_oIEA7{JhGLl(?zR?Oqh(r<}=`p+i2AG
z!xv+Pgs92rXG#PJeWd$cFQ-c7?#k;b4h*#LO62;`Qp^%$eV?}N&0nun#_#V&-1S<5
z;l{T&>Z&^AeAbrC36tr(eg1$I!TButh<2OV@0Y9kHhFiXBz`phVN|3mM=I)#dM;U2
zL!k#DEA_V9`RFe^{DoU)E%<Z4cglXcUk_M09p45*93dTN`>kJD?mDslT4T#Fq8tn4
z2Uf~o(zhes<ql&TFS{omr$to;rH8r=R+($+!dK7r89EJHH0i{5(r!`TE@&36^t?T?
ze6oM7d!!4t6tuA6PVz8F|K@)x=H*i$Wh%GC)r$!SOgO?A(cN%BjJmpJ?Vp$^Y{&Qe
zc=!S|<mP@QPX&@LQA@%fSPziE*vl-{slwpoB-UyM8<EOoO)t8;Zo(c=hp{vDa_TC?
zP65=&W6Tbd0hdqoOF4PY93l9;tqq6lGpeD@z04l1sTRJ@s`$b!+{uCFMU)HtSRz4`
zEZ8wY`6dqSsNl~G{L9s-WX4dMJls6?)~cd+jH3l6{c>epiCx!Om3+t|+f;MK-y>pW
zI#rQ7WQC-#ch2odAroB|I#&G@gG3HaBqeNCL!jYKz1idDTt%43Rx;l0v*#MM+so!C
z$pHfDndws~{kZE430u(LWU(a6tkU&@8|I$n^ob7f;mX2g$m|=g!EZHe-zg;B{gR5W
zc55+cK{aQN6+iEaB3%%MNV!^`6nGbds%aJPTR#<ypLV!QgEZ&l)!pJX!QFLc9(PZ`
zzh}~Kx*|E6V={k}SNWv>H#?`|omF$(G;Don><y#JD0O_335gKF9#8p&D?^bcz^(81
zl+5DBbN2fj5zW^otAD@my?V5-$x^xFx*CA!eej(rOXFf@M1)br5qaH#@5oO!>C(NM
z?AGdg&sG^>m1XwuZRX<wtO(hx_z@5(^;<9vH9t=j?{P1a{r!dD*)^l-$42;rru9Or
z$CaUbF$96tFSk&O;mv4m4Hsc1nC%-@de$Gm#A@6se&uU5It*)PHWhje=hvd;9Tyn}
zQU#-7M{_qmv-u9EW!>z75?{CbFQ2?T?|!uxWGmDoUFe!|ia6mAN;#6-OE^*{xAqUF
zuDs<f*L5h&<#%o8sr)BKDIe!}sy!^z8*`gjv)VuJ-C}<XH{=s~(p@IyoM=_aUvp~x
z_bk4z@tQSU#=Pqb4xX{(xv&KSvBUtB-0oyyW@OZwl$~sFF9gE6CKdjUaCmH6ghD<_
zZA)<n%E$s7W;h>uN9j*Hk~5HT^;bvUFVIC3uye^1$y2t6zD1)yX~RUyzM=2GK_9or
ziPrQe@-|sJyo^f5?ZLFQn{rdZRnbM<Uc1ZUD1G=J{U(Gz%d{@3yMDy|lLT)|qw%QZ
zKTL3?M8B|088x@<U2m&AWKsz4vf>#b2sFP2yQf5vz+$2-sL<KH#&(by{zMC=MBTwv
ze~dj_@KArUmFi@zhU$M@D@cuxMi8+MjTC$ehV#Mr(to_~c^FHb@zCAUq6twgSn-hY
zzHzE;EZ1TzGv4~0L-V>c($40UCwCLm?0#ofv}bjzCpQ@OrnCs_pj+Upvig|@mCG}u
zB_qKz828PgA;5-{ZIcGVOm-_eLkX|e6R>(Yz!o$;4wf_$z1}xFB$Z0Jyjx`r<lhuh
zam;KmtUyPSX0Ke`stw=v$5lyqb7M-G-hrsz3SQ$c$BfUSL}NRgrM<wASI~3O@15?e
zOP=&YV2{HLjjMr_0tqhaAj8DIPm-Jrep+TA7VSB}9k%RzaU6?IQ{pPSoYc>El;TH8
zaVRcUV_{}P@{9nc6viGlq&OPo*5{GtUN#hW2lHBM*1!2X>DoiKK_U3Fq_pn9=+otX
zVkOK*R@bA9_1nRZGQ<Wnx?c&zRl3Y|XSayYTeJ(G0w8IrJ9jnri8r6LL)7rYoE<U#
z=%t5Nk8aLOoo(_`wi{$ucU4lO6wRYcDJ~g2`44kj0pLV{SgaRaKj#1rS^>!bI1Z^#
zQHNqm*OlemV9rQ)*TF~)>J)|vWy8-A)m7}5;Zuoigbvf?m9P?hE>^-hx~Te;?$1fN
zN}rQG(RrGRG%50kcKhJ*)S-^s*WQRF-&uj@kY-ony(BrZ9LoQ`pR1T_;#f%wQb@h4
z(H|{Ioa(DzteP(<%*SXb{86u{M#M-TUQ|@?TCXsbsgvQN+tgM)(wuC6r-faj_Ux04
zS_E%d6fT0~GZ%stZFfrCQT;mqx-(&UYW=-h=5Neiq)bFkGSriurwc*dVovFu3vvxj
zw#!#gldo*Hx--J?X>D!#Ym(L(QbI_icjV(xuH83KtZ?FfZg>${iDPCc?Lg2E&<s+w
zqZSdApSim6#lCi%lg5y~^;2Q>$`3`cA|)_~=tkbG!ht6CX=ZJjZaX{YMAC@<N8|(k
z?>UcjayqKzEK*%8suqxj(QOfd0*aa#>>njnjTEdjg_?ZNNaH69k(y`TPNN(&u!{Y<
zImj3^s8E(qsH|P~*nifK(@l%Nn*DC9dWt7F3EguXqk4O2sIyQbZkYPysF&P1%m;}u
zp6!?j^20Q-zC_ZOIRYOlx5-bE;_9NJtdeNE#|5A6gz#Yns1u{SVht96C9bKN?Q6>&
z?I4Nvs^}4?H>xRX@A!i6q7i&qO$e=8FJ6h+a1-|z5?vQ~AAe^iSLiwSH(hhuzH{_j
zDn$O{IE^WzZN6BQLzBIEl{+!2reBJ7rMZ-SgNp6llYfH>(a>pXf|5Bmu&Ae<TX7ri
z_!4m;UxnYnr<ebf!1~!sTyb$*6Wt`YR{<}kWh4@qsI6KK_moGMZoM10t+7|X^I$$!
zM7x-&>{+F2IC#u(W;Mz7gGwwTYIA88)dYuq(jT*x!(o6assmA}s{9?ELUdGDOhT8H
z=q))<QC{}@TpRA?-sj(yX?-i#P6mW<qmi!b6X&cvtO3>&>IzHkJcJrI;bxWLc!BCU
zI7b_K5@zqm9jFiS^YEDwoAxw|xTjndOp4^8bc!0qPpigzw33c4opBcm`k)sX+P%ph
zrBfizs3=THYtWbOzL<kOQ|Cbhvb?bG+5zDch;&Yx@4#nU8Z<fF%zemTfkLZD4V2q)
zXOxXTLMCLokfX?yzIdAYG44rsvD$55mejGDz*x&)s^-TFBngp;ev!!zwIs{<fzyFw
zhK={!qgv)$G``d6kF&j#C-))ic@&UJTJ@e26TOORCF{eH;IM5kss6TSF0NglBy2@S
zSGC=R4fvy-)l~*_Evf3`%<NPWZGN1XY%OWTWVUkMb@<OF)>N<ozPD`1!gdn+qbHRk
z+~{Tallh=9KR6aP#a&kV{om+e!DFa6%!ei;Z;=@kBIcaJuCypS*mUB=cfC|UBX3*d
zBG7Pq09QBZrL?@^L@>=KL82kid5sVzi!{P8y(ycncPl;vcLuI+fK5TFiRe#t)z(lQ
zA~wj;ObV3iH_O`VkL;oiZrYHDEmNKnU3OFF(;?g@5n^d5XR8FhI=+eH00cuNny=2f
zJkp*tG|=?eVR$)R7Qk&DLOA{@u%zrya>p$7WHbW2*B?xkYzdDhaUat2r}h2f&F<Ld
zRwFzhv~V~0&|TYQPlutqoQbh5BMtT+UDFu98<MA07DHO75k2Wmc8Dg@+7+C)0w;iK
z)`B{DYsI(IsFuP*|1r`?zG&I6rsKB!HVqkxesdo`?tOZJWi%@7V-?Oszin9&sTd05
z2{{0Gtm*z^AB+<V5i9+kZvm&x`|;ng+{2f~x+=w@>$=uPdt~9i^vNlwzoa`9@8oE>
z{$gh@Rx7jRA6^=L611ypFylr1pM_cHpL?ItnHP6aR&|Nz0Y|sk58T{>>*+-O-d{6i
z3l(Oz==5Ocf-8|i+(czoCHCg6QNO$VO>9UsGNcs_hom$_^t^&C!erp*iGCet-@5aw
z&219!7)LGZvlci9rZtgxDQ<ngFrrw-)%;F|daD#CJOfJem)TL42+G^{-1xl0GZT*k
zs#Ssp6fuHR#A#$^#KxK@A*9dOxpwE#OYN+var1plqfey^_0tHJJf4O2Y~>X?*O&F|
z9#5vvlQ?W`q$a-q6e~-+xersg?Uel-#<*RUVSb~^CFS%6%awIwB}zD7Lo2&c0d3#`
z$A&u-;$3@0?lmTQ!~5po+ljO{Q~wxx$$U#V8b2jWCMQlGL{N5y@^(pe5K1>_iC(oy
zD68R)zoo`S>H|KJjzQV3gGE^8NlxaD*6smmwn1!n#)^04_+ozpGI-TVlF3I6y`+)n
z@!rg!DL;NM9?MEncXTIL-y&@L)su|YB#|d%9nwm7dgx5BmU%`04mbdbB+wa2Xv5Hr
zOuUd^Y&1qxdLMT;A#{(hp8J8D_|`1p{YIwrY@cj?>%|)?af91TgvGVh2xLzxautfW
z6cUwLCc4hLF)ta}+bKeJ1rcUsOuE$FAjJi%@(7oiJ|aD+zh`4K-AEYN8O6p=@51U~
z<Bupo>-K07SuUr%+IfyA5vo=TGV|DSj9dsYzg4X-^<)qph1(-OmVG=cVCrNVAN8I`
zy>36ViR+=c8jfeYM2RBq|ETCLY{q147p8(N+|?vaN*Km}*eRzpzAU2z7ladIX3K`L
zL7gM?Fpi37l(ms=*<+i8)kFcUFNnDzlYx^;ek|je_AqQSiS|vS&`RTeByAM;o2y1)
z2*iW(g34yK=I6?ko5F|1mbRpaZUPS1S+>YvDPN%Pqf^aq@yyKPkJ#+?4$rDgu3zDC
zfHXx-`cYKue(BrDuf5%|nwz85%iGr~FyXrgU;IbTT-3Zj&S2run67Cz+n^5@!`HJr
z57#C+Uy@)M-EG#Zaz`u-dj1FP1?lKX|0EedB-_Zwy8&APG(|Px6KdQ}jkniRL{Adj
zsgw>Xuas6$z9kw(?2tN@7U-*wFG?@0aPYzwU{n<5MUhU_C`P1W{-r}u0vsccL-uo-
za0~TuXsPAx8arJm-n^l=A?YXXl%C`aTmW`%ts~ETa~l#ektXI&LMr!ApXh{eW<5ua
zME+1j^s<TzvJ*4jb3(xT3j2->T-2kcADn{HYWgt*3dO?{d?{J%VQuySQ0^(nKu*ZJ
zjkG7Xt32rcZ4R=PI(2caI+@@R>J1sJ^~}<Gk?g$eNQFOI+w@92-C4d=m<iG>x2paa
z*X8}X0!f5ozAt0eKe4C=Yi8qb-IdhIDFm;%Pw^jeXn-$%az!JPedV@qpR9Qq^PgX{
z?UJ`MrJu{r=UEOUQXNKBd{Y=0uFM$4GVXDj?2w!B+i#SsjW0I8s(xEgR$t%X@SP*9
zL^VCce}!zV!4h-ZZ)cA7-WyRN5jDq9j+=|URpzekiYdM~vF4)gIt{B7L(f)eeD^ob
zJY+{hZo1anB%dwUS|AdBr!K7Ty7DhU|Kd=V7?vHK0&~+{FBa|XM6A==f5vEqK+h7s
zMDQxko)ESSNvs~<k%&LQ7m7jrpd{q(#JT(Gb8yNYdgLb!qC?ES+wM*5jv<^t;SdM(
z36MF&^>g;5hHGgtaTyyr0E>(g*DU^~JZ^jW-l0ZWRy||7J7)iuIlhMuA_l3IMDFv7
zZNo+{Z8{!<RdDVdk_ii2lJ)zhr?PAk&Qifqi%7NS>n8n*xb@ZHsVZGah;jXECC9F8
zN~ez-zoSix`QYw)%GT{1B4$w&DPb=GT{G}rLL#u9RhBJ`g%s-wbg_oG%=P)=&ao!*
zL35lpu8?c&@GFKz7_eO6J{J#R$IN9LZuCdpe_t^LZWs&KhIL=xSw8lZKe{c7Bzz%I
z%fwxSW0?6KG9wp1r3=1}Mrz(LZqRO^HtR%!arGX4K{%2eOA^Y>s``k=_<SG!c+&Q(
z<L6BAcYGBu)NpM#s*naQVKNJ})K3Wn^ZSHpGhc1I`$R|%GsL~^XI(7W*R^_|v;pJ1
zNSk6CzOC8K)0N34<vYql^7)QM(jF$IGrU>%PERNmOtOXfmP%IMWBerUuOh*vN#5RT
zzTReG7nGR^0*2}`W<i*a`aKQ`9Og657G%b?sFxG#9`r(KzK?w5p3IxD;9iP7mNHz+
z2tVsimg`c6XBF1>a(zx{6)}HX@CBH7#Ya1&86RR~ne*7RN4u083Y^yK|1|^4+mktZ
zVJ$}qqc6r&$XOAe^4reqSstuoRMn{U!(Hp)on&E%IcpbIni=j~G@KN2Q~wC)&3nN_
ztp&W%A1Tn4#vdDBKCs=;iwq}JRD6WugP9^5F=-qRaj9uL!>|W}e9ol(@jkme**!@9
zCPC#iwav;M<uWynn+^$kWYtkkks4I2L^wB}$iKF>B40rxl#UnZaj%{jBM^=S4eBzJ
zU4$>s=lIj7A4=Ry(JHei<wg)8^HXj+Bg4bpI_V!^BE?_U94-b?A9ikc#HF;L-IP@o
zFY$?;+pNir$-~(rEJh!Zli?~{apHly12=cTaB*CBbJquP4!)nju5%0)7kJJrKT&K}
zE_>pq!|vN_j&~8F3^dJWd4BQ?VaaQRIM=9^Vr^Y4QcIo0816c&sfDM(aDS}IqLGUd
zYSbq5cZrx*T|dQg<Xt9kO2iLE8{R@94`SVV8ltq+i004wnjPfiqU`nDJW%3<N6{R%
zl0$(k(PTewsGkW;{~}_y5y{q}*9jBPxoQ3u^J3}uoP_4EWT<v3zlF!x0B83cop~xz
zZ?oJFl3<NWJg3^u7+=+Z)_0Qoas<2VmqNjiPm<`U8q`+S7J;9Z<FFN7|JdZ^wTb~L
zCFT`%8!RpSRg#GCHbV<S-a2rA?;F`<l9Q0j-uYn6V>@B(qZ**$>3QlYg9}gl>=RSf
z^Lw9x*Jd`@?%RX-uL867;hL34R>$cLDY(jIlaJ_*uzNpwkGfL*Cr<lD?~ZF>Z)_{A
z?FNTXa^mR8tV_n?$6NQN@;{t?cX<4Twc*>ZEqD;M{s$~ysB6GIzf`-aaQ8}Cj(d(m
zZ}_JeR}0`>6AjsGf?M@JZ)+n2Q=NWi4UPTVPW$>Hnd5ydBg|eYv-(&*w%0V<?(-g#
z)$|pU0dzgvQ{szs6(R-ne5NOLR(J~cpF+v5a4a55IRqplT1%5=lky>)cRG+2z(k>^
zYL8KB=B18`V(A2n6h?PrBg3gRV+|?=)ei{kaMOlkgb>DdN&eP@DHiwHob%uDR(pe>
z^OZ6eg!6Ua!LMT$Xr^wh`h;}NSEfWEaR(`%6Z;B<g_IrWu=YQVFZ3~SV|84(trAqn
zj_a0X3(p}-a=XynOST>G?o@_fb5~e0jS^LRrMBr%C(VXmJQ4n2&Dh(FdVYs>9X?Ee
zgNs!XMa!i`5wqwUj=evTOj%fOj|tSoUPe;=A(N;#roQMROb;=jmsv5=3Yw2BLgDK<
zE*qAeO>Vu_%E{jtXY(1jv#6=}YUN3q(wmE|Tp_|9{2+YupUse0TxSETCBloIFUg_G
zN;6Yq)^cGM{KARjKbbo7t7Yiv4TCjHC?Sx?#bIjxvGSV`Wixgy`Bw$}NZ8gH!kY!V
zdp<uC_WB%1GnY7G+GH{XUL`J5Ezso|)0uUXl8MVauI?D;a*q!{pgCb86~j^j1p2_J
zvq0{8R$rTNLzuI%{)?4)37Hq6vS3-d|24ZC?}pj0uxR*@zZg%TdPr!{cFOLQtipte
zveUl3`t?s=PVS_(S!bu+4U<O$gh#gFvAPkorc<5|QE9<~HfB6~B6)1v2D6{>YhvX*
zN%UTvKD)9<tPpo$`-v$``+QgrL7Q;rd|GU;)f{EF@&v3{#7T)8Ab1M2x3SGa+$y)0
z(wCn4U891%52??Fi~VE^J`Gy?k9cFvLR`9Vbfm6FQqR<aT-U^f?jDm7wKg0Pb@@I5
zp@ccMVYz2<ZZKRo_D3a0Tr_;^c~-)RX64GF+*uS2&w)yapLeW1h&kA!ei^HC7^%#d
zVjoqN50LGms3(!f$ucxf^wioWmx=pp+0pjvl{-E83E{e?u?`7r0XEO>+I&6&iX#=F
z+a{Lp8at7B@B7xSyi80}`|0P?ADXz%ars59$tTt*yD5&R)TpP_-_X`9U;@^F`=@<3
zPYAamCnP;XlYb<fHIEF`RJ^waLn%($E98a|oanpGXW5yknXz-s`77aSJUetB>k37G
zA+)+Qr%%jV#(o6i9Q>>X-gqS-qQF5wYLl4Hp(u|LcMaE^i{6RrGRsg2_b%A?RVpY8
zr}t!}>2l|%6f8ueHTSO6A-05gD+_J2B9+-j2m9hg8EiGmR6CFE*Yh8xz_W`w%())o
zq<A9|As*(klX%z{_OSfD(y3qZ4nq#%?voBoc^<{JZib!YWV<m4`{IjflF^4zCKIbT
zFLOYF$mctH)<sj1@+a;iAy4sjPXb00;|Z@=zHO58LpzWtCKJ}N@kHpz5BQy1DuaHg
z1TejGvNq{yqFFc183g-jq?mDj#ShG~9aJKDgU!gK3>a2+fya!#gX~@H#INfs8gXT(
zuC+QNPr4<^S<)MH*DS1Neg{eP1TD>0ZvWYkz^`u3rM<Z_F(5ohk7qC&hv;`d+rPN7
zB2`h_r6e;MCR5Stw(z0nLPmWM<IHp%TNHImLoN+g-TuAwj>E^e_bRG*=S&@`23-|{
zds*;9(izkU_eSP_P=t<+xEbzF=S)X2{QSLXDA)>4#P49YLA>jGU75dwV^OoI0@nv-
z*am5MS76`rysJGw5(x71$EKN4D9!RuX24ul7NURSOa|HLk17WVIc}}kxi(n4x+7bd
z`6X)YMG$v6a-!b1o_`G5wcEmZuoX|W8Y3|DNWAK#W5-RXUHV+VIhgjH=#!T>-mk)(
z*XwmvoJJZAD==P+p3~*t3ulFU9?5^6n;oU0uB}i@uAl-vg}Th|^cod^bvW+8(j2UW
zmdTuhtu;#n!i&PIyI%*&4+qNcuYEhLmh7{5KZqybsqk?+qu+d5*jL~V&2v`aS(cVt
z0uwTIlecM72U}GI6D%I^e@Q9Jl$i^)%xYR5I@@Sie(Iy4@2N3@m{E>xmS}5Az}Ri=
z!bctCjtv%AUp9#3HrQJS4S^NiE>=N8IRY;h4$PbRk3@g1xyue~1#tRD3d(lnJa48r
zqZ#tXaGuxq(rjn12AERjXwmEq@a+$@-WQ~C8&EDhe(I?=O#9x0R>;^wDaJd!RC-nV
zd3C=`Bdy7=bi4RH#$UJ$qqQzl4v2$^<?SL=!3wp<ZhGNh<L$){2PqdWmiEiZgW=yz
zZ9!TO()-WpkmtIhNuRs#`bY6=#fyrT@s}5eFG%W1{F0@%Wd<Vs$1m>*xPNR39v2do
zhQ)H(wu}5eUEXyZ>}sTYKkRMMCgS_Pj#zXY{exC0LVC41Q2orRa1`ww`tP1incR^a
zYpI-QT&Vo|CAqcD$Ivh31cNi&gEQOammF^fuBW)J@DHqvny+j<UNN#-;opqg*H-uK
zw?GZ@^_hE}V5dWerag+imP6TsxY58ftaQ)S!hY)!EnZ9LqZ30QZ@k61!E(LyzD`?}
zW2V_`x`TY3?N!F4qb`*LhtacJRz-f&N}l~Hp2mBLNLFLF{@}gwS9$}7pYZ(^Gb3%H
zw(06><MYI<8t)rV=ZUl!{PrLhO5D5u@5xf>wIm3SD0%zVmQ|2}+)d1C@7QytqDOB>
z@;r>fIiJcnmR*|P(EJ@TdD%62MUCoXrQjoHT9cqa{cy*#=Z<r(HTJ%ruZ(3j?X8xb
zp?`2cn29u`q7Fb<+vq(PDd;gprO15~s2;#h;wy~UYTRRavvIu`*#+s3>)v3$e6uhk
z?td_=gK!b^+C47VJzJ@mYY-5Pv7kWdmI?R^?8G){cvYtr@TY|KnGPpSq&2aPIsOJ%
zUXFy=Q~w1Pz7i@*)Pc%>B>xxO#W~N_|3?wrOU7)=u5f(jO{*<FA4c1Rdfq9`evk4L
zBx)a<+IJ6#%;L+EqTn3<6OUWZ)H!=@jTkQY=WC+vu4_k=RUS;Jcu@PE_1mw6-Ndv+
z5tzc~@rAZk-zj48lMB>Bj$QcoI~}6Rs_gU)xf9x;l(eMQ7n|vw%ETs)xyPmEr3HRz
zom!e0j%dEjwXat(G&7rGoKotirR!q5*P|U&7t)d2$!kL+LjMvkLz-pyVZmJ?W3a?#
zgv<}9a9x1Xd?gwu7U9wP#x55X6mbx9ZA?rD_g63g1DXxH4tr9WiG65l$D`vyW*E|n
zbe{M~AN;P%InH^;aElX%nmP~3%i?pHN1Y%lDnvgognaAbB))7h>&bA#(R8S1gOhA-
z;E{xb&iMmiw*6j)e|WCmUpT&1%yRbncI0MBj!fXyhjbg++<jbYTB1cWHYBBGA$ANr
z^<ZHqh4(E|XSd>)XU%=zj$4XUhOHm#_Yvsne7-Sdm4dJK4%+gfdi9yGzAPiP9PW#?
zyj;wU7%dqN_vhtJUGMV^zq2}T+64YD>drDO%C2Ag(jd||-HoJxQqs-P-Cfe%-7PtE
z3ld6$gp`1EC?TOBF@PZ5DeszlKYKs>*~fnN_xIDtaUe6ZuIpOsU+3>sp}dLpm&^oA
zX6Vy<ov#L!n#E=hYF)okcY3ECV|P!u$D<a`G|^s^n5vg(AC8qyW=j`z#`;6By_pr?
z;V|LOO2j@PIePtsLO?#_ii2A<BXi|Y5b<>={VKY^f4t0^Am=Z<>k2yL@IiK`>y;&Z
z^QW@1R4CS1L`z;fmmN3AvC^S*xdLD&5+G!#+M*OhX5vNO$1xMaj%e%o<?nk@DF|V3
zJ$ji$7j^~Rb>I2?&thyop*a8Ff1UG?qiZI-E7U4CY<1=VgHNkgs7BmvKnyd4CfQqX
zOQP}Oc9b{Lo_W${CCDdKNaIT6xjbVB;@22ff%^j$T*#sA?DshbUsq`ayFm|KvK<|N
z=k`M$YVrCtEn9~d5^Av#L|a5#I8b|R4|;5=StPG|Oy6Dvw2U40h+J`|(%OV}B<L<j
z<&Pg6N2Zhra=m!2$Q1waVo#2!vN!jE>heMe5rr<ai$(aQ1v55RwXe|m5iued%(khQ
z8b<8~6sJZ2ZQRVD>PtTYY*O;F#Ck(tFmi>)jwQ!%ekAlnA$Fn<j5iCzP`=6I5J|G?
znFMo~ARkG82Ls+Y!hY_P5GwD5HP)~X4o4|@yx|Yo5|32t1<pRb<0|vzyv@uR#=J%<
z%*)|*>rkgkh5=V>?CDtC9yp;+LV8uwIhT0*Kd;o^7ffvcU7$gCpkPDW1DPUhUFH!~
zTq`rx^-hcthP@CH5~uC%_AES-s*G~*ftD6fwz4#r@U)E?I-J|qP=VUsIKK6(39no+
zI&QXro5hh$lER}F7hU8R)kcIz(_O$$U<62#)~tdzi_d2N<t~B1?-J<pL6SFk+iLON
zVS!p;qlshx0XGOp=P-7|E4NE*7T?I>2(CpEgd5j;w#)CA_{YFfzt1ba+4%Xdi~o;n
z0FmmA>ze*@9Psx$WrRJFUID|WR$#PQVh%st8H}D`U-YCfv5ujVc(>2Ktg9SaTg^b4
zv(0A%9(8mAoKN5Q_U!-5>ipxbR!hKhwWuTV>E>SsSDZ)$Q2^np)t_3nfY>m|mu>~%
z0RT3y)uHXz?U2KNJqZ7J9>9w@H~s2>3HRme`3d-c4?sT3jvwXp?oavl61eriz*qq$
zUn|k=U;xtp`n6c2LJ0z(;Zp=z(V_M%_UI29Uo+Lt2WC}K0vUUDW-3PN&HKZb=&7WR
zMS-gZxs4zP3k_THI{IIPh?S=*1wt=;`{3i<>T?rp8~^LK>xAC|ok5y!9K69~-6+s0
zFv4L|IxlzU>$rR$ZJPM?fth9oFa-MX2dIhh08_>HWpI`aKtw#Gg2`?<k;gir+~cDY
zJ^%dD|7}b7j~k5_7p$SpnP$?vO)&aT2W0VCKpP}p=(`^G<dZI*g$sbQxdWo)C~zuf
z@WnL-L^haviyZ1FpBZKT5ny3w1vo5NR5A_9Tc7^<-2LOy|L05CQmFV>ynxEGq}$+p
z&yePiW&yjvEFuLAs=XcW&IYE>aF>P@d0;3N4`+v3zdc5uOYVhZBGCD7qSf#$4na-;
z1vo$S6)6lLrA^KU!*C}+u{8paG6cbfK-voHdAP632ik<Sj}PFX`Il|rpFf$o$rV?|
zb^)(<1>gq~ZbeiCR2fG{egIWJ_8kY;A6FDepVENRNR|c_j{R;2qUF=K$K;qeVSwU^
zW&v00lzchC5AfKUggQWH;nqYvP-mbUcIEp5$bZGTV_36pv=0|xhD%BK;eWa9{&59*
z*AvCnaRVx~fZRkA5LeiNWo{-gNaQp#A!<Ooj`80DT0@Q|+bN=M0D~Wo>!KrTfHGc?
z;NW@#j<CyHZ8%1h29uOOS@<OO4pffZ!O-I<fPHfT&4w0JVaR{&f&cvS|JWGK++>Sc
zVnyTy9UIrM@OM6fil!2L=9h=@LO!D70K!(oFm<`h_#nLTBx`}K@vDKp1<tKMoH|A0
zT4N`>w}v+#@z0axf8JXygkhyL7ZyjrCWoS|0bqE`dE&YRmciypu~$pGrfsP024>`a
zt^w$|Jmld|_%0j1Q;I1d?n8kur3{<`{Rke55FOwEdR?>{us3D_cpRu%-RodLb>@eG
zKa>G1;)n6%C};uyU!2CY37_K%NOPtAfLOC!wMg^?s98qA_#qqIT1P2|N2V6Y!_=tM
z$KJ%>g1O-_Ak$+^iGX{nyj=i-5G{bmDps2UQVkWj>T6s?tI2H%f%^%n?~Kz*XZv1d
zb;rwt`iS?y6l+(oRfS1Z`UDusSp(To&d;ka09d>TaK=&I=3{^^yn=ZJ*L^gCX;eiZ
zJ8W9NCI0Wn(Z5~_A|XWk^xDesH10Tj$vDK+N>=C60X2zp%@byINmh03A26UO;$WFr
z|4U({83xdaO0!?(OJVxcKsP8Y@|g*&MBr7ehI1UgU&#o>n-OlC(zqVahMNB=w+9e&
znSuDl`2^TJzO(~(ZCn7H9h1&&IhYZ;56IvO8E0ehs==%ZIgVfqcLXS)bK2#W{{UbL
zbeMh~r#d`O!PZO3R6Ca|J;OUS+8y*Yi3}sTn?aQ><`AwIDE&Z8z!<`7H`^{(YGo7W
z4rqE;(gkh|aKXqo5k~oRurs|(UlIGCH^@H^occ;!aj%w2NiTu`Ap2K^qbsG$YttTX
zdmh-b1=v|rG@$C=!^2W|r+)*<fx;}NQz?GH2pqLb`xf|@7s2V4Wq-gfLC|8CUHO;k
zmgo-cijJOnlz5tZi*UPS_A=0f=$D58R$G{O;s)@`WWs%wV!3<8z?OW18&Dmccjmtq
zCGE6ry$KHe00s~g`<80>jQ{|$RM0G^rR42<s$FS%b&Nh&I9v+C2Xx`CIDHx-9Lzw-
zHrBiWe?-+~!M_o|+FdUacJ@yE9}h+XS(t&`o`kf-6Ny=xr1IDNL|)Wd)q&_8;H}y9
zEt6jGlzjs{tCUKJ{jjbG3Nup!9GAnm>qw_rSpAY2XkV?F3d|Ah_-rTh%UphTD*$V^
z#)Be2Ad6e#d*WAlWA6_9Agq_ce!QFxIMR`~niVR_z+N)MdYKESL5>O6r=rl`UWZDa
z8k?MtJ>yePyT>*Y9F*AzMB_H=#=s?IWITsgDj#s`7*i5uCxLZ>eb^6_mxyV>w)t+d
zTiu@lirXv2<OdLj^IX!P&wW*t-+uVloBF^1q?ds{D`epU941Bpw+=s+tn32l-&$FY
z>pT~M+3k2Un&jiE4ltO`kzThHgK6I`t5c$qFiW=vWMT@lqQ=q9bdvynmQRZ^@-j1g
z;}4kS&aGVTO8EoCfC{t0raCupini1p-VXS!aXOrL5epIWk!sC+)ylmvpd7^Uf)y?O
zr35!K91mKTAbPD|tn&oH50G-Lm*sd4fv_3Fud<*=9QECDmpD!CtmWHB|JRO9Ctmyw
zFW_|hffT&}Pz6L5@_DOjWI}_We_^yo@4ibl0HdJZC&0qX>Lj}D2LM3vT+(i-%`le5
ziOBN<4XhII1dzCyI@$6E1zjd^U8N;%aCZ1VCI|HcfttaIs|>4JftuC27Ft7$y_e&V
zYJoKaXdX^gqwTipsMgNLb>O+<1;CwaPXPVRW^>K-zc2Q`?)NNA$o7;Cx=O3MGRE~1
ztzbIDDpiL&4EiULtkK^WOD0QZucS0L)05ZYP<76t3!Gw&qn`!;=8ez2-(|X>v6<QJ
zt)vnId^Y{_-VxNY(lrFy#d5_gTxQ~im;%Z3(L88Oh}*wCXN$ApaYeTO8$_?L{Jzr~
zJRPIsAMx|wjY3p`i0lL?&LY!TrS?izax=r>rmg8<WZtHwv?ESrHhxAUYT*-Hec7~&
zfU#Jnu<=Ai;49owS&D-BqOtgd>z_Nwzn-sZQt&s8_SEXOmn^FR27RP`k1Y;?j&YP&
z<oD$$Zh-HP&+lB7SZaZmPk&O7<vn;i3M(UkBG5ly0BAc!fi`)90N>v)4nL^uSz}>7
ztp9I;8T>t8AR{ne*vrdpzGIqvdUQZe;0O=pc>r}cP=TjMH+msYEF?GuJ#oy~8mi5M
z&HO$EtSJlH$51yBz6G<0GSCRniE>qbw>}6fA_&k|yZ*PwL?tae43rXLHGOf6o_C=%
z<`b+Mv)E+I49+wM#HVtVPa5R>8CFsPc{Q-ey}>-be0`6txjAltz^ERG)p*YJ9BNw9
zoR>TF2_=K3?fVzV;3yaPO7f^(qy4+XsZs)c7Gi-j)dn&nWgrICjp$(Y(LOeUi=J%d
zIpjY8b_7$FiN9U=y&$GQpLiqI^dTVCVvFf!E<^~!?S-30MS)rhf0R!AsDzaw;C*?p
zz{l8;9$dkxo(9VyroVS0(jf{Dr4ul7ZKd%9>_~PXiekdPIE91P$Ae-&f}M@oqe!T@
z+d(6#J+=!t3kh5Uk^zt3)xZ7RuE_CNVWspJz_dKw3WyROGmHaZOrcVta68R79HmwK
z7!JSGa2me{A=3y@VNx?p05n{i6$o=tHciJ<3qfF6CX+y<KQIR;CN{hL6ix9JNP9E}
zxAqk1>^(+VtEwncIihjE_c3ai;s=|tYY6b{7y(27nO<U<o@3UwI4!I22r}-YSUTVe
zFaaa#Nr)<G<>LpyxGPC)`~%1Mj=;5O&>`TWl_AqX8x~8D3Yc;)36i~)3V0Z;9>5_N
zbz%4tUJ8eSFFyPimY*-cY+6s_;~+MJuqj92&;Jo_TS$Q*R|K8W@eSO^Hq6Qn98o7A
z^vQ}NP1E%aeX+g;z{r#GmgNLMT6BwXjqj7L%%=c`17?tRTi{R?3}&)uG~HRNBsD1m
zyr->jR&zwo=Vg#mr2{J@M&NKPPX%c0_>pk<F&V=_A}$;>l}?{5%)>^FE?wUe+}yeG
z8^}R3IzhZpE}0CYE_68lKwj|~^BKMmJe92eY{dhl+b}TCI_`@e{VQ;<VJPh><iOn2
z1(=9o_&EX*9)@(Kg2S}Rs60VFaNSS<n&Lee@GX#c{j&Cde|Zqy#uAhbHHP`>{wE6{
z<-&fU-kc~d0;~0JdW*8%5ePS8#J5(HUN8k;J*n~pic=;eH0*ImwC6Xt7<B!OEFi2E
z*(}#pnfBt2z%5g1k%{2xaCKFR5wg5sKTQ((Phj;M3Ilp{h&SL%;QP7t0GKO+Pyb_g
zoVZpHrdoae339Xa*Zkf}1}*l4!KOf5h+np;me2(jJ{=JL4T%Kn7JiRui``<m^lmL?
zKB0+L2i?wXV@R2l$R)U8R8v+Gkk0i$d;l3}<@vB+kpP3^oYSvt|J}n29s<NL#>%KP
zk`~s=$f8OoA@vb1b@x{d_5iQ*zzJwkxGfTBc5D!X5m|C428gGQ!%J;S9oZx@M#zam
zm;bpHh_i_p|B{*Wl9_Wx`Ucrk07+4G3zb!^eQzo2JP+V_sPkyf4x09w0h(;>MPt{a
zfB#vCh!9)S77X=P{RRKszWw(uF78DO(s3`^-`7I_Qn~t{;p1E^$h+szYi<A6n!tY@
z2wKFT@}o`MSXlS(Qx))e)RN%aYtpL!xBo9boVYgKvw#c7Yw#>vgJ2H|h{D>|-&#29
zsDrO0a$h(w!3F#54FmWswf8G#|8h~nJ|lrMizqqxOy?g<I*W)56nK1a2Zu6{HZ^`i
zC7%r^(lsUE|L(|ZH&+b}wc8j>dRIp?wfH|W`$}1Ook-S|*y|f_@Pjea{mr}QMvaB<
zJGFSLOj{31Ytyn!VdxA7VuKTHb~DxkQ(_}u;?m3PB5oQVnDT~E*=qIF_L-;hpZ45y
z%jvwD)|=sDVM&rRc#f`=|6HMP_p6i%kXK#<r5rPWC@9Lh+kqh)c+eBw1A0dWnff+4
zt`A@f%gp33%uGPf5ew-Cxqt-RbH#49@@uJ`qoAU4nl$ioVuP0yo`afXe$_j!8+e5}
zwh{mv<oT9fz0ttD0cRFGp?tESI^A8!=ak!QHzVu4V2);|hE$}y`vAhq$csDRHNz8h
z;|b47tw3^<3LvMATI}x2M&M*cqymcBkoG~3+D`jh7kGW<R~FdFKk7(Gd}LP1XK{Gn
z5ItM7O7<J@4>C+W*&nJruszCR?$F2c_*Jv#dvkr#(Bz7R@XR{zt8ZBf)kFB+5!g+A
z2kTvp2rgGO;%B;1%dLFDlhOH~!*33IxImeJh_3>fNPK9oIh_JrzUfM`+1B9fKM9@>
zSC)P*3-uabE9<Q5n$M;iFCMdjwXR|LzB6cF016x_80AbCd2`XdMvNx4AyNwYQ)21Y
zr(3qrrinrFv34hev9}P{48o3}yEx%0%jBp_`)by!x~|JKo!_;FI*p}04?xp448R-A
zyf)**hVx(}l?J-fmsMayWkuk#d&D0<0B76;5Jo82Xhc(-fY(xdv(tCJz#hPmg8{iW
zj)#25WD+@l8mKX;VS~zx#t5#KEE*OWgG=Xe^cg6HZ2kIi2Kd>iiPL^WW0xR983pba
zNIBP_>5mR$LM9tfZ-@lB8;D_wi~tCM$9gnF>3ruqFgaJw1sCwT=51gbsM&17t3l(l
z{YaRAwgClV@DGyG<pcrI7T0y`9I{C_I-uZT><65L`gwx^1!N49t5faIxV_oqB{92c
z=|ym(pg^6mg|js=u_(OZJ}u+PW^kK<sOxOo87XIEnhSWg|EvP;m}@>x>-GB+UL^!s
z@8gr1t|w@D!g-xy4~GwbRDs>oY&HAfsXIJSE*^!~D@SRKJxPOaPpx)@%|Zi@=5$FV
zxoVY>phAL1BVA1f`>a^3zKs;y#36r_RcpjFb~3k7vC~`z6H~>scG<AcwS}wzSt#wT
zK@9b*=wHYf%w`L*O9sTRD)u`P9NU-d-S8heQ}C_=f8d;)fm%lez3lM&R~vy!zrVTf
zs=OTKv6eTWrCb6TmUVF&Es@HShpJsq&=?t4`z1K*M&bOQ6m*cAHE%_+;i~~gZx#h0
zRrlGvp-Yp}C)GJ6Rw)Jn{6M|geb6TlAT`Vz1!sOK`MLTDuy2;Tx(C_MlWDcFr^hQF
zeOz`{x;{EvF-I*)l!-s9$%|>e2W}wndLi@@3Tp{vn1b`^W(QV)*&_`&cGAGnso%2c
z5yWKA+JNTQYeUkv2C76^0O${ag9%{P4aqY~XwuXqADhuIK7yPnI+%>dD%SCR3=W{3
z##s<ue2wesbA=48FHiVd)YgY4^({&7xuN%#e1^M~(&$Oxu%$8?*MxdKIsFaK-2q;u
z@e{M+?=WUG4iNFvJUvFBI|FAfP7H^<sxiPVxdQ@7X6!boVGznZx@}1rf>*(qFqKRM
z8nVLr;iV06JbE?H0V;f(jc#h@W$JyZfucH)d1!pzUwqZrwmy(eKs1{GAZPr*vedHA
z<~?1)J>5ZFr}nK1zhePfC);6Vq>`OEsM$l)35&eHI~p4_KmK{E&wu9A!z;c5Q1VJt
z>{mEilE>@I-%;&8`E@Mry*x4WQ&Sjr+Zc-RWn<g%iN<F@bM8=JGS&_usjW6F+F7t~
zIDs_o=Q`+kT_nf^ztI|PdILEKs)A9K9He<|h9hBTSUx7<B7iP@<dfMesD({paI|WA
z9i6-zA*3&sKY+v8Bs#)xB(_TWI_k?&X&ke^vsxTIfI<uOFH*uX)+ylcb*;zfM7ySA
z8sy~%O8AHh09VpF3~(4ZGMO|SUck)))$8S!9#KzuOWuT-)h|%lz}?!ipuec3zAERY
z0Mr+n=jviivsM;d&&3~>fg!GBQK?2EvIVeG-PcL#Y~(a!mn;n|JtQ06jz?y`U=3gT
zX-8u~YB+D?q3rwGo}j_@^p{G<2MNU7e#)}KQ<Qp0`VQ{?_$N3nhaliOG$B~-o=TFs
z$`TtYP~RIE3AR~QJTs@>fF7wBY;%{L0Tb=g-+rJ=ChkS2rZqrCUJa6@5qOn5P<Ae1
zs-0+GsZ;)IaVAn#7B(MV5#Tg3@;UFcy|}-n?l9XsS(7`_``O)?={%{r;3l_dSI~k2
zs?boYTRlcH_Ii-<6PM~%##{MIIk>TUP*!O~6ll>gaC(q35<HY>n98|yY3qFB3cQ92
zf2lJ~p<5fYuio{0n-*~E2u!(5v;%XB@xxm$_nE18ZSH49yt6b3>2VnGNQ&_dFIiM9
zwOeBAt1m`mI?V05{`fP}Z(nV$;fQ5ah+e4$Yj^VNhXl2YkNl|dRN~}BXqgMS+n^4s
zFI5K4?`|n%p??|bg<-7&4FfCK$y-6#O+~14FVm8(%*eOEF1`gv88yOr7}?aKv)v*9
zmLL^4EVQep^D(tn2w>VHn?K9$fyYfj;FmZLx6?a2K(U7c<PNGH5t>CZUpL6pB}%K|
z9W{%v3mzw<zrMq72JA>Gd{x{)LTxs0i#m6aF^N4m?_*Y7Uv-wr0tfa=YzeJ$eJh=8
z6BnHF-Kp<tJ6IQZlJMlp^N+_wvKA6jM;=J?qW5sJKQ3?S2j}0ZY~uDZJ)2@e%l%N~
zsr_Tre76d8Y0PM8iZRn=sLq;s)nC+0?`tx4T=%mWU1?fyseeZm{9VC{gK?>SPpe*)
zBdvNw0G(DmLl#R923#G8Rt>A|&F7t=OW7oq*M4n^?l%OgOj69sqxMSyZ{(Qw&sHuT
z+L=y$);vVgk7q(g;&ru6SExjV1p!l$*INBxWSLr7OQTjAPt$f1QtP)l_rC4n>*0*e
z1S?)EkC@WjU2V|*xmAV3Lmox&1Cm0~8a-?JvPHqgpUV~1rIkR%x2l}G*MU4IV9%c0
zddh3dBzAw}a5D6xeG{TfK=<2%3miT{JRXVF2q61XF{;#tS2fe&M$77R3(`)XwAvh7
ze6rqyszA<%XNPGO)g@cwnob9dtc-Z;m%t9WxKW#tUp+29CA!@h$I^YPt3E_)WbBK#
zHuL!PYxR%%3d=%Xl3maR!-YAv7GWr^az?C@mSoqgueWA-c%N@ch*x?V3!ZtaJxrUD
zF!EA#|7%2#^c)%6kJo_Dz^mSJCXJE#cw0S>U>Cum&mgf|4J`S~H!Y;E@tx3$02q<`
zYebqAFlS<SE4o%^q(TaY>_r_L(WU)eaKofTSmE7c%lV$MpxEKL)A;50D$3tkfSL)i
zfSVC4EQx{K=G-;w*N)3{wf4JbIa~A>3)z(h>ZRT#_<R-Yh3KFvms4q=QaX)6+qLAP
zsMW{d5t7Q^oXdI?vHmRt)*4rxo5^CHzw+^}e1)_WwRC@+N|lk`eSjmi4N=VV`l#P3
z>v!!G>tS`4sd^(PKZ2X?gHIY|?M_Nu$Wd28x^DNM4n)RnmD1|+k)DkF@Yy%}#4J@K
zE>sY@XR1Z?gODH81wVtha$eY~j10G&WyvV-Rh8UG?{`IA*E9~SfG$JEdoXcE%SBWN
za<GunG)BPy9Y#U|?Rn+h`FBM>KxZRPXlsO&>)w(ZKcG2E*&7sE)0V{ds>)2>7HPi_
zQfFq`ab~^r@p_g9hn2@Ik+KPJARB?vU{;v{b-3F{hIsfho2nRyG@((9r!@1$bZo}q
zM}mOzjn5@5kn3MYn)9FwTXOR8wYWMw(Ngz{(H#U~=40vNN-a095-MSQi#i$NLqBb|
z9*u2WI1%kVFi!mR0n_=A<JHHAl1q4qG+0T1=4gi3$`0jL>}fNGE?|DRys7R&$8?(O
z->|NZ|L90MiOU{bqD%40Q)BKYF<r#<LWW(%Mcbr9m*J$N+Vb8?UC;Q}=1Bit^)qDc
z?;@H7c4hs6+C5Y?D^;#1YORy_(;6bXwbSo&lr(X4AQbWVYK^<oiX*%Bkz_Am%wo4j
zCK_H&MstP&dfII&Kdo)`Ax_HW``r^7mJ3qMDrvy1I0c%^H*+_ifvO1fVNEupTn4D-
zUgw;hfg6LF#uD>_)$ynIvvwz4)r_@GckYeh{%;1p{gl>TTT~fapouY{IDfFXOD-*|
zaV+CQ;Mc`z@~zdj6UY(0^|AX_ANIav-!cEB_pvO2%^hoha4rLzy&!qjzT{He{JKOn
z=W~508s|3`DE#`?)>XtJ{LigsVNdN%50esuL?{Z4e0%>i{v6C3ras+<zWSYJzfqIA
z-ek4hlk{!s>K<ppwlT4QPy2(~t?uvYu-~5IUUx4)NBELEh#bW$b*XPqReH_VLJQ{g
z<-TcYliz3ahN3u?7Hqs#&Wb+I)cPLBHa2)Ouk_p4Sb{hp{D#_F1cT&2^s@8RijbqV
z6k)CubQe{kaj2;p7XnWwtdm}RLw{|(E`0ngo|T>N7#)-;T36$UKia|jhxJ?rROkLD
z{=fzmS&kTakC4(2mikG``AXtpXHdkq>kiJ<7ap(>zX8b+mmZWrn-}W(MCG}Kwo$(4
zsvFUZ7f`Y%Aq#RG9A-oIkxRkIY+ChNbT2C{p%oLA^Tm*5-3g-5m-{7R>c1rmno=j@
zU==ZnCneP+Je<me#C6@3v*x`8ml)V`xzFu1B&`LVBS5`0xO}p+4&D>_qPtn;kxyMT
zwn3P6Wn-Fr$*{$ywUl1OMwmf@<xNl&a3E+`bNWIh1%i17!vFm;7tj<8o4h_VzFXns
zaJng6oHtb1=DT^zv+Ju}9-pw!HHf-z1gfQdzOvSZw!BWdz&%zt>{v{PQsp@D$g$aR
zu~DnVU-$(pH(CDF7jML!PvRIiDWb+i)kh0jWt_fyOeWo-{`6jwpNpO;@Am|id0?F)
zO<KOv6{=sc5Q%;)X4%26ZM5`Q1@ah;fGv?;OYIaTV9PhyRi)vpEP0eCtUq5@r#P6T
zUlr}9Jg1gY?Ts~+Jeuosf@iZ~+Y6E+U9FB*GpRhoXokrP0S=vJL64k;1N?DIExTSj
zc3%_aT%s1vs6SLtj&CIH=n-c5r2<k_M0=bM;(ncc?+Dr{lO+M6_wJaj(zANLKH1`y
zxf9#>=9U@9QBo|ki8GdC>jt+CZb=XxFF7wa;Jh>JKj@eB#pFZ<?c}Z4ryO6RZ2I1(
zBD!B}6KMxuSaMPNcDuUE=x0M|R0~lk_Fa>I_Rb+X6+1WUGqU#!Bv6`vWvaJe*E5@J
z60Vf%I5-Yc?1u`e^WUJP<IG~!y~~(z8AoccL*JEM{9#+TM!H&%Qzy3GI9+6iw%=cd
zK1JnZlNg3||5^Aiv|Xvdo6NS1i6W?S@E!JBx8i+a)3W6GpF}Zdoi2XZ?ARpueIm++
zV*VAOq*F)%LW7(iw0AB^n8)(ivcm{H$Ucye^Sxr}kH*2=NBwB@A<WMtC+`*4A<7{A
z<g{<}Y1O)h?YBr_(?qOM3#VS+&!{~6O^<YDV{DP1ICKa;ama#C6BQ5Re!-}(>;r~^
z_{J2`GgpE50l;!cA)`(4q)I*~w^ov}Ov$ssK+DM}8>(K)Ab8#|#Eb9t(IiAFJj2K<
zvQBFgmdG4nH7Y52P#kceY(kZ0ig+jrMgM-?^DqYSa470#tgeVObYJ0}vSBzJCg16%
zrw(LU+zgqII+M>`UIfQQ-ZAWL>&oP4`qeQt$**u$O_d@%`ZRiK-)C&ttOG+{ehjDI
zS6Y9jp53gEDdh%Ii)zOehg|y83f9kvj5veZqxYVx)39aF#{(6rz=sIrbDi7$0q5m(
z)xp;jR#Be$X$2!Dt;>Z^v;1db<m?P<uqs6UNb~sKUBPsr;RyZo%?ANL{0$p5Bg;WM
zEoG}ys@%%r++g(J=Ic;gnq`jr^%uii45gYxODcbXYuE8?SR=9N=$*}F+d+s=1&k<&
zh|8~AxACyR2Sq_S9XK{clcy8++_A!vJy}@iVapvG$~hlq4yWFc&A)>FDRtR!6Aun*
zQJprQ?DQ)sjPdT~Qf&OP>lK?BCUi-8e-qP_4CW+qj8X6*I?2@;3o{c6r2_`jr7{VL
zvlJdgA>e#8&Hm}hO%nf_@tj>HP&>daj?=d$*V@DjFLw`=c5~cQ-M`5=F;Rlaxd^9H
zzB~NoN~Vr$#gvO_lt3mGT58XR{POsB(rWI2&cxVJIhEnGFQc60UmoE+o({@I!f%2}
zTW>*oKv6)Si(w$m-;L@*e`HL$fII3v*a6S?@{0e&FQO%k5uOXJNOZJusdqZF;<xmW
zXxVH$57bdD6`TuEv`X8KIjLSpRT#TDj~IE}B9{lRXOyUVgZx-}|0$}%wXn?Qz5YHg
z6GV#%;>ZLaR+|iN8t7QV+R*2>M@S#7M&u>(+Q@*nC=_e-_^l+o?13S%{A{Yk321I)
z8F20D>c@#JXGHUO(UT88&!$tZWY8sHdxg8etK=8h4eHtWjL3aKjE+2RUiA6!LcRX6
zS3l)aNudU>M*KKR;>#tw&1bdp$;k9|FBlQ}u9T8R&Icebco1If!+X}2CgKI_$aY49
zO8J@Y8QC1%I=(SPrSb3bZl4`&2w%U?yMlqw!Bz4NN98SN9XX7qJkgF&QR*i#e)h|0
zk0j?0=^+1qnN6W)Sfanr5OM(WKX2qhrxu*tk%$|f59km+24Pj^0#A*$sk>rny|y8v
z#4?kPt_#X^Zse^YJwjca5~&cjh7_pI$Zr@Z)jhF$d_5$r+~xP}tK(Os1U1zy$T>qE
zv=yI(mg+j3_|mVornc<!xN_|!=5Y>xTpb#396k{(WzbtIv(hs6w;co4%g)zJ(8~mW
zdYrc?@|HcPO7M{Zb56!1XETmE?-r@N-ZGFp=G;NeAijLQ%ttEa#%E3Vk`f+6C!5t}
zkQ8CtyAu;QAF?lCa{Y)Mz)!_^PRD$W9ZK#kFWYx%mHQ5bU4(tymfP#fjN4RENFVOK
zoT^{9;xj(!KgF4p7pWM+u#6B4@7ST$=xf4})!{qE+^3t8L_FURJOoX@Qh-Zj4}C|4
zYtx)JY^d<lDzWwwJ&{`E4gx%=^PbL++6r?YV3=WgdndC^#ZqB2Kqz9`qmBuFy3AIZ
zmll@wcS=wVXuj;s5g0%G&}unueu2b+GDy5|v_dPk<?I?i|D@A!fqHk%xbJFJWHR7U
z=X-2OezZV_a`6KxZzAz-ZCOa{J7nWuQq!9npUYh)!SEGskx*(KRa<29_{1zdws6mE
z&fTjzD(Au0`=-AO@7m$w>Xqw*98V(`Hll`1yF>SB_qLveQ31cBoC&hyMJuy6@mbjr
z2QB*uXx*K~P1n#i5>k%Ez>ru5&CTu*#+=6ItZp*=VJfzS-(t{T<hiyRSO`9OX{v64
zqNbNO<RePP1Nv7=rkyE{L^$}r?X?d@B=<KLU5scaXKCf+QJ+RGOVEF5!)f;#eFIEh
zSa&Wmmk=4$Vk*$E9}oQH7W;t1IKipY__&mPlA1i)^z{RG8QXplWH5?n>4&FB+!?d?
zzD}vi6ibfsm3Ca=@1MTBi(X-@9&C(5m28^_ouQr0H$U#Yi>x}F*@N%zlaW9Aro5J|
z#zT{AxE(5DnD4$Zh-?1wAjbH6Hr-u^mu811xF(S;peMgF!#sSV^ED*HARrb=lzO++
zd9{r?ZR`aX`b)*3^ESB-rlQyPqOjzoZft|(Y@=oY5wwt0ag9N**1*ah$ffn}aCX~X
z6mb)}xcn+NMADp}=fGhMNqEi;!jwE3T*+#Iqmhiok#Gi|kbrkh``$)H^>7$1o=LHC
z&r38nJhDL=cpK5a@80szw?!<Ocw0UZx8vQ!VvV3Ide_iC#DFM00DOq?h(3EKf98f^
zkuRP^XD`r;@v)@vWcmbJ_opyP>UmneQDwwP11XKz*Yx9h^C;hX6ajK-5e%*-`)1Yh
zguYG}2FgPWX!n!N=~mh4&jVDL-p_2C<c)Lr?gM5N{rP;{NN$%G>hV5dTY%<o|5j44
zRr+kP#Qhg<J~COlK|9QABPt6VsNF8Yq+|D@YZIw?)4<16Wn#;CWgkDJyjOfY<*5Or
z8mc4sv6G`|{%5J@lpCK08G?po_W;dKd3M$BZSvqR*_P(*(#8R*vL`{Kv<wUs2<-Z%
zp)W{253KZn@tAtQGBKLzD)Kz%+10qILsnqN%qyJ?$+E)pO$>_V<ICHnU>6>eJ!&Wk
z1T8{I6IKy?`3T<`r9a>1aZ3V$@y|9c&YkSd6_vHP5!qHV^ajjUGzi8ZhgGimtzMfI
z*urS{2c4_dqzfJWX<;Nq<4MSTN0;yJP56Go-OR6Q-iN?Yt{<gJ8D{fZ@Ss}|rS&b1
zY*8UHVZ#v2<!<x%U8rJJx5&w3GWnP;hO@j_>^F{sw)nRkUOK*x6G4}uchC;!9Di}v
zP#!bOVm0|lZV-1Mt=xK&QN)AY<5|o$aNg6}5<R4l4@Rdx@c7c-`AATbYIYk`>66Sc
zyx=_T{vhbq6OU)7%pU9;MV_;!j2S#TZK@$WMIJ@mpVsig{3Eb>B6tVVPzKpyn3sE8
z;A0T;3=f;GKU2vj$ZqwUvPk$IiP|=GwI!Y?SaACzqWAeUB`@Lb^>TH}(c}S`VR;)k
z=y|ttEj7e!IU+AG%(7ydh@$l5ShTfc#(9gX@I`&O^EiSxb()&U1M0ccf#t`vYXinv
zNs8E?aK^1>>$IpVIzir#;wN+H!*u6Y%8aXh3o$hNfZRj!Gh^&y!ZEVWs)ErCR7wGH
z%3ew3ggqTCOtd_3@Wsw6OzCG+S`@xi0s_T#zWjJJoH-jgNF=5UZS<6fkD7vp4)?av
zclTI5#3l6az7F9YUlg!;M{!^|u(~=e)JxEa*kRs(_8?0pTq6Iz@CKWLZ+ad+M<-=m
zQ`5QKKs-g8gt8K_(AwhvjE_W->hFP^N_%8`h6Y_CbfwWwBfT<-mCsywD%|?=&ptJp
zsT7%IL4A}vk{SNG*i3{ioPPM8E+<wcx^aHF%fzHyd#78VGyaX(x;8XD(ZQ3}>b!}K
z0Y6PY@=`7tgCTX<p$?fE-%Wy5b%f)#fb9cPPvKGP8TQ4RMd@Q6hfZf6_VF1RG~dvK
zO{AQ1*2YiRKLu4mi>KgL9-+FUfsyN6$YJsa4AeL#Aw8FrK1?5fx;uPxk#FdF;podl
z$*wg$jxJn=OFjLW2Ajp?*5*}sWbPaud33c+zV;u24!sq*(^dg_M7qt^inMfHyS!J5
zj<-j2B3(%Z7GG*Zp6c7~v{zSbS{-y=kHCJ~aMDZ)Ay*Cl@gIHi$F2ClW+<f@*Ut`(
zLw8q_iHiC`C+)zO@looeRQL%pqsYby_04L3Vvtzj+V_ls1Ptk6VFx6yq+2}v)CPF(
z3%T!kt%J2_Q?aFU1B!u6fo;9K)z^`#piaT;Dm`k@vdfteOIXbo2)}AdlCyT)Gxtfl
zPWEmF9wLsDBV1iMt={hCdu^OitJzE!e2Z4;G=jZ{^2tWPPc%o<c{H^0SrswgNi`FB
zEdAC_wwjrQ^Eor+G<p<-ppn)hU9m&baj(JGmcTNPkk8JMMShB!{BZz9C2tv0UTW6)
zb@FfI9K4OOzM=^WKPf*jc(LfKNt}gIVnx0O=3H1Ye@gz=B5T*3E2Hv?-~XM{%#eoL
z`J0xhmt69V^Ky?ok32J<(Y2&?tjHEaebkX3uxsWiduC1hsh>pLzJE04YJ-392{J^O
zZ(+P~ahvJ`SCZKWf2m1QKe_yhGrxK-EKBSY!S0`6tY$puWc4W2^A0dHgEF01D&D0z
zfk=R%TqMF#e$K;tOOzqsDSjkUC_{FyL1F6=QW{FYFp;l`@1Ok?MEAciz;NLKn41+>
z3D=~^Mxv3pVtXFiR9zRtCtgT9L4byi%n<wP<+*}O+m2YF+ZICEXY{+lE1M@|-rNmx
z6!l?RFwp@rZ;=^Vri)lKg~@!?qBJUvacnahFIOVXSH;U__U92S))d*u6xO^%d!idS
zoqj`jztFH>x2g6><u10tMuqYMR;R(pOiFafss#gTBrStFMfRx$4<UZCI}34mtnCTt
zttO-+gR4#oOrmKhp7JwYu=6;@4>jQ_*brpe;@<xFW)o|2vx2U<Yh%Ukf$qC-sz{oW
z$&{1Dbg?IjOnOb;VCxmJ`uaH9QO9MNbPaKM?6;?rt(Rr3c;Dnr!XH@sw&(^1-Lu*}
zXnHq(%muHQ!HN-Sa>{#(T?tvYV!2K*Sc!asxwCu~G<A!JAvi{#NEj4NiaGv?ueG7D
z{yL)Li?cRA&JO5B;xkasc)^xqGpKVP4=~esjT=e(Ef{ZqL>Wv@oqV5FiL?*X-!h{Q
z-lo-m=y;5IbP!P)DGpPMvQ6k`I<XCIadWW_MM9>$oEe}x6d8sL8|1ah5Ah-un)Tj_
zVUM$U56g<Idzn&9iM@o+OQ?||6TP&+K_heQ)BAP0ZedP|#vr;N%%ks7y-;V`b9G^H
zO8R0<E@SW64p&(XO~>?ZmiFAL!i}Udfxdi_#3T&!4YMl7B)bI<+6af1Jca<0;B8d}
zuNBU~gl0!ClQ@w{#k8KA0lEY|p6xF>bg0rt)o!2P>tFe#V8_*IqA}8uiwY3rphof0
zJ#ypv0sGI*n+>WDX%sY0`Up=Ovq{PV|D;&GG08~S-SJ|z3zuG`T-wTxl%95;PVCX*
zNRgDTnO_RevP7Q}p!q~h;FDsF8+0bP+mX@!rn*LtDkEXr08@VD^T5&pB=MYzhhg?p
ze)1RM)ii7mFZC$oqRokOb>p)kORN#eStkx`ut}!+ss(E@l6M#Ji3lne8Gf8P&yk4W
zm03Gt$Xr4mtI#5w8~s%l^qns)rA%~gjiH=1R)jBry-XjyDK2)10ZVp0b^^=eue=Do
zf~x9R5e7V^Wz0D>K0ZeSns8uI3a!TwxBf;nH5(<irNXBhh013jFRkZK)aq!!QBSfB
z`#x%(KFpN-Q@tci+)~{IpEL6{8#y_>6b<#dvOERXeS7Bsm6r;EEBm-QR;h7=)MIq{
zGO}MkjRYF>!(lX(Wwi>b<-hfoE<vXoH;ZZW(-s?3Itx-78KFMk%MW4fHG4*vzqlAK
z7n*YnoO5TJCJbuNHm|X&C=1DB_3JmT*NkzVIyp&6A|<wGGScdiS|pz+PY__FospE(
zEh>?b_Q_83^*CB=T*(hC_=X3XN06WmQ>Fik__1~!i!`m}wabn-NQnP_=QdHD{*PKr
zB9aE{MWEzB+1T4Z4QG!BqEWVL0|%%sX|RX$wqF)-?ekAOj(t2zbM#W9d^XFCFj;XI
zF>bVt<GC76i(RH2{Zxd4ZW`Jt-T>=)<du{L2R1GQ-bYR0qbm}id0O*fL~>~qKOn`Q
zLV75yADKRa#G&#V5n1o9hcVYz&LgLG<FRo)2An8Hc?<oYQuUvwy#~SpS1F|_Ks#~x
zmm&lCb%Fo8W$n@!@rPsVHL3x!$;@f<>UCMeSfs5^`V^IN7vI5Clr21lA^D-~_;M}<
zsji!y+``*&1d8>cDMZV#%cqyu5!n*%FR)WYxdqWmtiu9uWy&UG_9R3l8a>36_qZpY
z5U&{G#_x2{mcR0f5Xen2aFJm69!8Y?d1GNW`w$Yc_%W1wtg`@Hp5$cj_(ypIBj@eY
zF{B+K`W9a?`fJ(c)L2W=*GSfj<|L<dtiwbk6%^yNUBgR5h%xj`bC~Lb6m|$BD#34r
z<po#xiQ3S1hDAmbap>5w+jx_|sC<*4U?@X^l~KlmQQ~jNo!h$l)=<}1;^geE=&qnc
z%><g~jYWOXKm17%)&A5rgscBC8LS515|W7tyfuAyn#_}!jr=LrT@sy@^j%MR)@6DG
zP|G^8^~6z2ru{y0dIuNXUXE2%e^2D#omH~5K)v+|lC$E9+;E=@lP1!aGC*&Rfj*JW
zRiSjVsxqc?v$cof+7b)j8od=}Xr@YzZJqLx@NcYMHVCqKVZ3{kj*s22FA&Q#+Rw|r
z=432geB!V}3lk*~i8e%u6783>DrcrH+pd2qCocI~jNzx~A7{rDqo&4`y$G!)*TYaB
zBu)?w9t@#W<PL-lbV_S`?{^!k>kbZ5@g1o7wEU#~+)<V5!0wfmc_p>9ug~~R4T<Rw
zA=kOepe&}*J96DUYW$xkL!zs{ZyY}(JRP(Ac;{4?(q{I$Mr?S&gf)yXJ<E<fpl&x@
z=);3gC4M;dFAt?v(zo{w&uLiO<V|$EDj!=UryFUGnWv8<iu|$s^a_R}EzaJhraeHl
zOV7lCB$!snjK!}4vIcUg@2+YOfyW#L#+DX1_$W`4J!APbGQw`KB9)Y`8t~${*PG^w
zo;$AUJtxK=RorXHfuIrH;X*Q!A|K_2AX-t{8{fT-RP30<&HAx{<6cTvaj{8zkrHaQ
zkqDFGG>oh)q;0|f9sy}>NjY@<ks==UL+S?SHYp=PT3XxLVnK_Sh_+y{{wrDrak^Cs
z4laZ{EWZv5N_#P!K2d{bgNwg|#Z;!%IVheRk;Kd3fA=yMdG&K!s2(KU%&{?PAtEv!
z17Y`*`P=2~c7>lYfldBz=kykPm$I#93b1Eq<SEv<`AHW)b%Ced19>;HEX-3$3{$>z
zk@vkyS|3whquopViOkpSX%meo!?k+KGr?T@5VEqO;dI+)nRH0+^iHtjUZt<nsSXlY
z$#6cEaDCP?1D^D8a=SR;!4`#-W6oHN=INy|ksH*~k((!U43q*^;rTS!(*sOU$77l<
zB}vM4E+$=)-zl$CNK*QVvN1`MVHg+#zT<>6K8eqpg)9(Dny(S#*vZM;V?}Q8<Lpl&
zSYm_~(FWNfvE6949S=VErnFLwyB0MNr>lWgsU$~!dCjxUbNLpj>`hftgiMF?IJxi}
z*e4?a17-Wa$kgIwP_mkFP1~_We1CrT-4c`+Ym)zwUMN<!p4eP&niwj3lX)#@LZ8BA
ze^+Li|JjQCP<$mXqkK}CXQE^uH*Wl${4aeoBYJMzY0tzdy`|TUzK7~Wev)V95l{t+
zrLns_>`pG_O@8IQuc~~qy%|1@6LIhTNO%eTS}T69{)c{&g$q$JMVz^{6vH1vAiTV5
z!BnUFBM>@AT=9!ZF=GL9WyJm+{|^bV#H@gu%PiVmKg_AOXjN$S9zG7*eG&Ge6l^EA
znn-9PQoD91t;0zBle^ZxOP77JS~889Y%}e@t5EIEs|IT)yAyqLS}~dg6K^7KIB75K
z$-A{x^&(|=?G7?NIF;XDaP<3Vqd_O{Z^rb_pUEu9cc~YA|B&QZxAGnxw!B&31It2L
z+52GsRdRABkl)Y?>T0U(T7DafkaJV}W=Fo~-Ii3Z;o%0-3Usl_ChG7O`H4?DH1`G8
z&#{||wD=QPz~Kn{If>00`gwC)tEupbL9f(al1n$EOJys>xIVJzQtL3v*6zoPOno8W
zzOIXw`Av>4)h`n{cOE@Kn=nf6Vc*1iB@w6Pxr(|(wPbT59o@s2?Zft%W_27hRU9&g
z<&_}Svkc)M-=f|a)z6zW&dZV;c&g7Wz89*5MUU&K+VajK>06%P`K<-tj?m`0WX4g4
z{5s#%otA0P!pJ=GAae*tK4hLXY;;EDQ809AYRPYLc(?jA`{bv_;p+~PCzn22T%p^|
zE6RLEtK53!r)3}4s}d?)L%u8XrJlBk8NTmKOZ?(qH*?lMXIK@LFl)G%YUuRT_n>C9
zI6*}H{7a&b)beb`cedK<5fs%}r7r&^b_i#Nfy{F3JK`3iU27+d_Ooe48GgP`2?d4m
zn3taEYW6p0%FCT7htLi|Y^p0Q8D?Jx<)Nm->I+uZD&~%MY_S~+QQyJ2B&f_tKsmdy
zZ`M(Ko=S0E{wjuXd9AcU;_AIF#_L34wPhGGSHO<vMFDq`oau7?qN-nAhaa}}ueaJ-
zr5qXMX^i^E9)6#b=qr(-DL5PYaha<HdT}|c%EWb_eJRQ&A#?*HjiQ>m(2Ox(o$(I+
z(JaZXbyixuPO;Zt{Yor{@?gO4oqbI~t6VZ%SDUf2FIz{WJtIxAryQnAlFWz0QpN1w
zrc*jT?J*Lf%kT2ZgMOcmB^AGuh^Jc%ktw84<DjxR`WKUKh7xB>-HEl^#EW^B50tzs
zJlZY#TE-lSXpA(cjF>%5CSQ|+S@O<bPds!yC1NaWTq?u2EVVDD#okETHi_$0`s2&I
zMM{Ums1d#myAn0w@Y0-=sQvX;*Yi!j!n>G~(lX_%+OjYfg}ioK>g+5(Sx+QtVNo99
zRhjg=49DRe!mQ`Q>CITG8TQE&jbnc0I5|Hv1fHzW*QiKu6_V}XY*eNAF&jVfDw$xl
zduQ_F_o(ykh`vGuuJT+!gsuUU=<(0I=XL<(_3_u5+|h!<ea^!XSuM;&=Nl_GqkXbU
zhSK8=pe$QePnS1C`jCh*7!Gld!$;TO$)N^|-y#ojj}}bft{pJxP4TNp5x@5LrR?;H
zs7j;g<0V7e9lEl8t=7kQW-V-c+^tEdtd7B`uUQ1mG704NZGL^a+C)@tX&L>iB$>sI
zV0Ziwq7xmwQILua=}Z7n<SRGTa@YQ*{`_n%7CZcJSH66<6u%G@7r68S2gOe7(q4R0
z6s9WN=@D2Z&AQ}t(zW2~%<KIYo|aLHziKY3l^0oxhDOR&z3SK4{cs-2TE^(Y=*@0-
zR6QERm>$k-2Yo!&`LoJU%=mD(M9$ZimuTcAcA4f#>$5+8A(hCUnOXjfL^9gn`xmq|
zw9blYgy-F;u0`L~bRKo9S1GA$u%2oEm`^0$qTemo$S#es{zUHHeW&O1Yfam?haje*
zX1WgniIK7#Oc;Io&2|4iEvWA{zgJJ|Twnc!Uw9re9k)b<H5Frz4?E|X@)+&xze3Yx
z43!E#bJIOH8M~sbj~N(irizXBc{W9SPk}Zlj%I9IS!VVNXSW(=Od*r;182&Hv##wu
zWj2>q&C7Jux{dLf*%&_mzx7pSsWipue(ESHhD!TLPm#1JA7kZ@yJ|$_Dmo^Yd^t;y
z>O6d_r2MCbZcAReWH*-8aWAx3XepMGZ`M}XB7sfAWT7mfyl=hH_7696yTsg|_2+@z
zkn-^qS*e;#;XTodQ;QdB6tV-=?-qRXgig7tNJph#yilow=?{t)`kA+Mw!2er6C%&b
zrg}HsF1+r~b1FgTtV%B5j^C{j+>1SxXvFf;N>I$R7?H~N8c2&ltIk^EkHJcALia|z
z)YvA?MAPj=nxQgQ%2ex@X}K`erf&+hGJMZRv29uO@m-mAcek_<rw-C!2?{p<dgy>D
zyZSujdt5yCN$P@#G|!jO_4rvPs&J0L670t*xy=G4SgZ;X1%E0oG}jwK`At23nD4yX
zH6X+fSirLf4F|QVPtUykB(>%DXzwke=hV&TuvBuiS7Brr7(ogsAALIT*m<$=EI^BD
zbxT-;xzPDn#1hf<hFR!kXu&Xjxdyc2r#5YaT#zy&1_o;xW{NZSt>P+~;{BJw>c>%#
zJ#)FQj#wph32v7H&oWc{5InTp)H!OD#4W-^d3RJ8lrd9t*_xHK@|4KZgiePH5WYJl
zHpNGdLiWBMR)*{6)tFSK(0(2OC$d@iz~gFLd5Ti~P^)-_J=*=WL1q3D(zs3|{D8^<
zC{sg>ceuh{gKWH%K7PP}%o6$z(~_K_(@JU3HN(X#r0JpAG2xNH&CCn^xHhwgg(Dc`
zJ~ci7&@U7fb1$jv6UHTAC<pSU>LF#{NOa$_%HO7})6&dEE{ROVRrwF0Jlj(4s}f?k
z=*1m+HcJWwaPKj(nrqnOL~d60f;ZC^EMGsnjR4b9euA7uJlXuNbV#zR2?lNWGbNL&
zxjJ00XL#juTlE?Cd<!jiz6)>`y_2-Ssp99AN>#MLv12Ri?WM4#A81uu#t)Eh=4Dp*
z<^+Fv#xat9U#~Dog}ouBS>Hg2;1(+XmnI*+eS@-+Lx|n{f(_*%<z66uN}1ea(j6lQ
z^4(GS5!!;d8;;%2khHDwLR;LH#c!-`@l}sjY8Eh`kH}mEG?2d5v%q<xe^`d3mp%IY
z)V%gw#tuIKCMS5|7)ly4&IZ{-oqO=i`2(jlP=rN=J*tS8A!LQ@9rFxcv0Y?4e%05)
zGDL?;?BgS%oNN>aykz#RvR<=ZBMx#SK+AK`CRIyFD)3YjwXfiofO5L@w!R6UH=ws~
zD-}{Ii1}rK@6zFoBl!zkn~~1~BOo}7>lL$`>vM#!wg~R0;oE<7W@HFSp{kf%l}p<Q
zK0c$cWR4CK(?Gk{bey5Uta38Ur!kamlKsz&-xuP|6$z~r5716ZP9vNU7zTXP(Xv5%
zcqU4a<Uh3H$2IGnbhmSeCD4`dCO-T5%e_BnX@BS+8qE>IvX~}ZN6jY{SkS$8=&7#^
z_p>|1;|B*`$<;he3Gz)Pmg_O*`d6OWy!{hMQfGOP_?KJE#*;y;H5tY$ix)-|iJhq$
zn^ElL1vb2CqEbO5=XSz~<iRkh0Ht!Fzto~0s=>`LL-1~qNMoX!=0fX)p(!p1sQ&RO
zyfX;(j=rO&2#H$|;y-4*c=pPf)f~vL0~S2VuVXp7nw{QBj!qI+bbulUvFx4%>R*!C
zTkkCGdWa!=4WDES9+T^JQqLej?fD>%^7Wr}a~~TCIo-T0!`N+Ic-{AgY4<f;e?>z)
z!=NqXRPvKnla`+bS74x@DQmIK^37-yY0u%~kyg-Qu`U*R94tG&$dugC5El7YQz$Dv
z>r#qV<@zy&y6w%7;~pVf!w_oj2M&Fok#X7F4$vz3sxvz``kj9j^=nv1lldr=8<iCK
zTDR~KH<|>5s#4MflxRPhKM6wH={@cB7z^66`Bs`Dlql%9xB0jwXCWdqA|)C(>Wg>*
zzVH>LCV#Z0s73N|@2MHz<GAGb_Yl?1aG|@c0h4pQ016E=^5o#qlp%FlKg21^J?V(+
z@Z;XFC)da07GtmP{N_!4md`O%>U~XJaq;uGi;V`+q*-HBSP_V|Otr~)wJ<a>S?$P9
zs?i&UuAaMR`N+LTbadiOSfHnhK}-zTB3QBx$tG3X(Hk4B`&&!I{v4SI(*iWl5`vj#
zf+3IQ2192F2hgmZq$#Ipx@FgvQEAxuLhc)TmHqO&o2Tw)nJ_NS5{=g4=~4F%HXo-*
z2)M*<XdeF7I<bGt`Iz={T$IMGKKU|Gy3<XJDI8mujx95KgSSWjkDT3d<r@_FzJ&+}
zni)?`6*F!(+nzt9ht06C%Q=^jLdTJ@&hE<dm-39i-v{OkRoc78dEy3Ty>;B6t$KyL
zs?}BG4RK5M-@P&BPa_)Ki5BvHCYiJP>NR)$PSzlg(?f0365FTt_&H)aT&@=GXxnmO
zY{P5$!OtH~>my?3e+YDPd+LQN3y&3bAC!2HC01%{Z-`GW1(l)2unOj43yg+-Ns3cb
zj7XBnuL`Wd9NB#W5;2<DXZSU&))FU?GBgZ2{u1Q#)OP49or_;^tHT#w0>VeXY@()@
z<J@2PH0tA`I2D`id#A9`3N;HHmvo!U*rZ4M)}m<i@^riUWZb=|corN^dF;9c)d>rh
zQVt&ov<qCqT7UH5=uCT=&-D4Sa!IJOG{o?03^Ov|voKTo+U6$lCWdXN$wMke&M7sL
zll1TX&m%-Q22uIGq?m>OkXo-~9TsV-n+LV%@hd5^eg@`-36fS?rdjv@N7q}1#SwMu
zw!tB|Yk<bxJ-CF%f?IG89^Bo%aSITjad!!U;4Z<P;O-VIxy85lK4;(iJohJmimK{b
zRcp;P-!bO0H-Iy2=<xc;NlBq_E+G#>kuom)eAhrJu>{s*a5qihv@s*CU(A)MF2eSq
z)9hUwq7^g)8o47c7Q?gG;Kb~1iK9lpzF9)V*3SLu`8$*PM`9fKNSsR?=i=G4Ch^H`
zwhlAuMPNC5(%;IS_AbLp^F@G+<ja@95|ivfU~JclMOCSRHTD?3>|<Kj>2C*i0v*+l
zks`U6>u@(P(j+t<*5~RN!4UX(%uom=4<jG+g3`K<aby);E`e3+F`Ww{+{Z?79Ia1N
zkcEj1C3l)JbYfOGk!a5h!8@8~AMC8d0<K}$v&7)(^y!|cc;a>ckbtl6_P<kl)GFIi
zA_PdD3zC&t>KU%a-9sPX;D%ys#gpC^pjk6&XgQ@}A<a^$CP+J@H_1_JSXN{q!BgUB
zZIr^nKZ$(Q8?u6qm}-KD2*?a}k}e9+XsS#ZdA6u8?|(R8p(OZSS;MNDLC59R+ACU?
z>o&=u+9#)J0{a!XagI#^pq|!*N6wD|EFxYBYNi;HSiu}x4YU{!x(2Ebq0fzt`sXSI
zm`M0L6crKSZ+HJ7&2v74wk+tpolN?T|EF^3m_0Oo`L1Cl2rYEBvAyxkGub=(9o{ms
zDr#EkE>*sj(*e>onc@fV@@XYboclj2+O&(n7;S@&0m(Fj;%OK6Jm$WA<9^Vcnu3TT
z0uk9C>Z;`lg)es|huz-LBvr6XXMthFA?9r3_FsWdnig?42n&rQ94`^Gbbbym!HtZe
zv=6`EWeNxg9F`<S-Td&9ng}HH3j_~jqWOt^EJ9)yni#Gx=p(1^DGd|fDT`}_w#mP%
z`y&+>qyBk?L3R!!#)b^WMA#b$8*geUp17SZ^Tv=x&g?zKT8GS+PfIR=5qFbzMp6mf
zbol1S8o>2<B*5cB`9AT{7}vh!il~C0k>O+iwH%7%fe8&M{trnFE}&wS#FE~I^jKb9
zjuouVwolxq;nWR(2tyMMj)1=Hv#SnV#<4NR`tjj;Cp!JOMH0x%Lu(HQP;J`dAo6rU
zD2#)2#Z)kALJ;qxKQPn}oSS;7JLG<7s>sz?qnwstxq>qkIpY^h?C}P(jq9c=u0HY%
zhsIwYh^N;^er86?ZKxUz(d0TpJMHW^wBnFK26ebTSjCM9lCgiA#lK%+`#@kzixT7i
zN?g%VEWH(clQ{W-BGqY6P$QmI1!EG8)b|QbR^w`flrkKGRWD4LlCc2cp{mVobo!u|
znGl)Bjt9WMP~xekA2>u+&5A^^^lijgnC0urx43FJt<)W%b*97DlEBb0%EeudJwyN-
zs0V~KhT%o!VZs$<09Aj-8MUR3eI*N#Q~dNgqqOT-nX+iq0W3qlznR*nZ5+?1(Io8C
zex-|pw#d0=P1b}12$T3oq+r)8wk_fhJ?I4qP9cIgV?GyW@pCzhA#HG)##@QXkGeH1
zKxY5w<6km{y`&-5xW(S@oadx_sHkHI;NgaNkkLF=lcvuY$7Kc3LBrw;Z^KhJC>{L{
ziB}`{9-%LWz#8J27-^+%NY+Mxf!c-2q`hUXStx}AG<=B75@Bl2ss4qKguru}BEb%6
z$kQJO<H#{mn^P{6&*3&*Aqsz2>)h?dw-$h|Ps2zS-D8RTih(XsHo$Tc+6Pjo7I4r7
zf7D8-$UE%8RBpp8t@{3ld{<TURMPo`?p7##ys1^RrL-M9*=+=0Q@+aWW)Twt-8MHU
z71uIYdfH{d*NEhb9A%^M$D>)iz|;E_7}>t_V5qw*brYm6^sSz>Vt6_rp|pjyN1I90
zOMYiAy`2jGp*+b#`=E?TEj%wkjIB+N_SjUob}&NL7}lbg7WP~_+rgvakqIYZ5}o`j
z+CrF|qPZlvyf9jP*`WQW7PZBU^6xj;5&R|^wjky@S|SCpt!}~<uSInDdz7v*WSoR>
zUTr8&z~e1@whkn=u8$%qL_~80EBHd1*7eXsvbM0KoD>@Ity6a8ClMQSY&M_B=m!wf
z;(0uYfmVITJ%~t=Hm;u(=N;|{rHsx=`Q>e`@jyIUsO$+k1f9KP#>rS}Fg&53x8#6N
zTY@07>OC({jtsQ#@Y>HZ-=W<8YjH!O_?A4ycn8drooDa1dCudYnyMErYisvFYS??7
z-G;hB*yWQo{lX)sW}SiU+_LB-^0IeFBuI8VCNZhw9-?Jg0-|}B<wKS!!XTxp4go{m
zKJsZ{Iz~&rJ8(jIo)9szSF)Izx!s3taUWIofm{u5GonEtIH_WqF#gsJ8rQ)LNCEXk
z68RXExTrr82C&7=Hh6Nj4GNT(?ySE8w!m*VWUfYYk@#Zh5Rgn9)V|m)3%MO>?CDgo
z)D0FVTQMl<<U3nJc~194`@)vGzXRdQ1`gTB;jQf-ym9JYb|;*&&|`ICQXgZqzsAZh
zNTsu_2xq%M&Pz&rj-2vc02qeTeFlb$NKO{xJtZ3D9p%*ufbNMW!T8)wRyH(^+=7Eb
z`240COJ0k984lZQJMXeE-ihlz@pIAX1MDRJvGP?2R#0p0&#%$al=qY<i5l`qd$Jlq
zQE<ck`BG%1j&^34PnBVakHsB)chk}WNV;@E_$!NM>ypz^-cqr~>XWcUlKL)~@l}w}
z|6u_n!ELc&#G-0&)Ey4`Km^I2Gb02dDc`B3km(n}xiId~J7siv8ubNrc%`-GO)@(y
z4;U`blS!v0#sa$r9cE+SnL{;Mc;%N$*QF5EMYnCng5d;@hF*Zpx(@ysw~CLXKC_80
zNliI?#@xJqwFr!(L3l5dB((1!)%<0`z4;rkQ33-E*zT)RBow&vv2z0AWGgP+9o;0u
zxvE>L;c}~-3xnsX801Q@DfgaiawfS7>zlJdVl<^Q$4FN@DzQ(oniwf8;wPNS*Zu$+
zfX^);awCHXp=I)Rt=}8uAJKpK-@SLKGT#m&MF@u7lX}8g^PumtE&5t?JwemsRFr>v
z1$c2;a5y7VvYj-1Olld#O~c~6mC(Uk#ASrI^bexcG7&qtFmtoi<U`FBh-C;)>bVz}
z9%gVbxalaFks5|%bmM<`fh9Rg|GwUB|L5`&f`6x*<<od`=+5Y=8Z@W1Ye>U*=?iC@
zY$zDGl%<d{`D9&zuZu<~IWD>WLP3hX<Qoq17`o(7>t5PdMzn&e)<rDHe1NC;s6JsJ
zsE<o><S46Cl0=n4AtA*8lXk=zfrhSob;~(ytX+vy@Wb^tawk1&xKP&VScB9zeS#nC
zY#}$XNG|$8RT4A;tN~XAtc`?nqF@uSz(t(^r0d-GJ(EUDj6CK*QYuz`DX+R{k~g@=
z01E%LX!osL%Rhsh;J7}J*hl0fkH+DxsbZ@XvL=hw3a0Vk$Lh6+#SqufZ~mVMCBO(1
z8egZ?{Q3q>WPND;R9z+(*b~ZFt91$>u4`g{X);Cj9o_cy{gTi53FeCp2Z9sOOUGFc
zd04b`r|MhHoub<bW6Z%QG{1FbmwS@NK{0?#tnPF8jR{-~o(4-9BIw)thNRGtB-)J&
z1>CIN`smxmHqKiG8tPdo^XX5te=<DcSn7l!9f3n3$mS-ah{+oxxTdBF=$P5n0)dIx
zo_1G@7rJj2GTtCC?>V2qQt1Y|V!Abqd$9Cdc<r<o9NB)jNqiHmVZr^jIc0K>nvIZp
zgAB31n}(<i_qGn5bKoFxtKNTi1o6!wB40H)DTvnXLr+JCr&<K?<sZQAV>oZi=#c)@
z9jU}O(4jC&N_o2|Eb)Q!gTg2^-91xzS`0klHvSd8Aufky+@d;r42*KFDzefES22I=
zp}Q7BU4+J`tP{&?OEn*?e<yk%i(2Ku4CC!lx~y^0ZCF^Cs|cUc4z4uf7_ES1M=0en
z*>pb)X@+~Am9U?Yh>YcxRdtlqiExMzw1^+HzWCcEVV@mib}|z0tU)3kt}j$Be3}nx
z47&8R^sH{ZLJPs4q=f-WAViPv(-61F`!LMXozfLX!UZgV4vaYBNSa(FfO>qwucp1+
z>&a|yC}EQ<g}LPdbe6@y>2M{i_~nWWybZ<LPQCjg(C4=o+9*%u<Q3nB`2mY83~{;V
zDZCLYpA6`!1z?t3NHSwlhn@RtrpHbg?uxKc8(y-B{C+S?GKOOzSp0lle0GbJ5?5Id
zbF;P})Hx~VgTV}*-KlU77_X~A?%+mSvJ5NF1*beKvf)Dk@xL4!U-CTId_r7*>vCM7
zKlxv-g7uNf8#V?Och)|`;NWb1eJ0Lpw7)s{6%-iK$N9Z>Ra<##Q)OcM$}oSUEuxPx
z)LQ6*MR(2f<5Eo@j9StyyqCD0U+yQXL8qBK(2f#G!o8u8<JNV@m15E>fniHHru`ZT
zdr(KlnSbFgiX<9078C#-`g{KCgOWkh1m8#)@Y8iwP6!irFjrypnN@Pci%vJO<K(zw
z${8VRJ|<dab508P`Q1K}pXqkGTdWu8bN}ns$rl?-V;A5kS@ub?B0{1R-k3!<O%NoV
z7IGln&VvJw6l2gu4f&2;J52iuQ;>}<KReuBy-s-qV1XA`X{3muFH59-eN)_*_C&?p
zuelk0K}r{G)g41y2$ZO}KGqGCX7yE3Fjt7fB^jYy&NK`7l4z2>us9B!`I01&6_O~J
z8WKLo12lAXTh57#l5RLXCx2wy5gD1kW9lM;LaVy;*_(g0>kL%q^e5DmR_ET2HR!6D
zu10+I)-`I!b=?FyhFU3BmZZYC=`dIWYD2|ihdI6vI+jMH;eS!YUa1K2d~^+()vay>
zxAvQtq{@<mt0Z)mfyU!~%5%!y&RmYV&_;ukTKH5uny0NB-*6gk%--1i#X{CoexKl8
z`WFpJ8^}#5+jb~I(~ka2XahDiQsS4;g9afEwsrNZ8A(cl)S-qaeQWBY?;lif2pLCp
zxH#e`t(F7LX`1P_18umaUC1b&{OhhE0{)$ovFd!>cD45EL%zLR<TV*z{~Ef@X?C-`
zA+?1*E+W?%H+n_Q7*WG4HD>e6e9!Al!ptrbp!_iyJiS$KZP_XPgEoyIYG%=j##WJQ
zVtOT22`8JCPYEmp+C`A)569Y`w(HqOsNB_^PZfNJ@}}WEBAXo)G@}#Pz=5a?#m&f$
zqCncS;#<&~)Yryv(y=2XS#0em3a0Q2!+UqXNF>>*s|zvW%<TIRkDj+?)qpNoAVmWo
zNqGIi)@fRMxu33xDj%4zJSUbLbHg;>P@GC71^yMcXVA)FRW|tY9Qdxj9xO77s|XT(
z?W^LGwcZ-u+Wj6b#LNgEl~#QX*rSPt5J-{inpJUG%oQ~d>SQ;l3{2s%uHi|oBKgKb
z7c+_{m6#?6;7H9y)s1{U{Fyk9RRFUMq48Z}DXoU+$Do*8iQO%lao@`ku4tH_nU1mK
zTUV`SaKuE3oQ0t5pt<0~3-yUvNUcMM-wY7(y=OH-M#UgynVO_?vIMSRnsUjeJ5uDr
zth2s7nEhDvsHX#ISS?XpG*<7bo4f8ddlzASOm^&)3}YYlx5G!O&$u+82=2?p@XD<h
zWBcMNGLl-hFmdyVX2DD!V>>?WOl3-D$~{(*2q81I^H5))`+u{xXIfH?=$AMNSgIp#
z*)Mle#}x7ai_=fj=OtU47R~`*nC;Ncn4Bhjz|Ikvv1MTTaP{5=D-`cEh^Vj%@}5&`
zHma8~d3tXRrLF&~OorG$b}LdkPRDI(^5d}?I!>8~u^CRm&!KU1ab6TFEd&SZG<I?N
zg=_-u1>>O^pPPg|9R#7<m{h|wkFCn@y=pExB(u2ASl;3I)j>tCQ{|{}B0R93K&U$~
z#4=)E@__Wv;*sU0vq3j4osyS7M1L|h?Vgkmw<@=-y3$~)eb9o-A$MNNby{Mm&N2$l
zZBuYlbmv>#y#Pj@SF_y-cKYQ{?qQ;B;^3)<cBce5?MT^mxU~4cW*E&OEv;zkIrg&$
zj)R0_-)|I7rus*IkF>+Q{|X?_HUhtyjrX+{tU*e!Haq_PqHDNR(gmjsdc6ZWM0F#Y
zoz|ZJE0usR^GZOa>o2*H`Kds8YJhwdA6A*OH?2YteL29C%QBOvYwVI`79>--<oO$6
zY^$6OQ3DSYXWCQHI2o|C{`7Q%i9cLHWz9K6WLJPsQrklCUV2EI&z!hC8JsNvYqF1a
zRn_?O!%iEpP8#MKP4}J9lyodzhE-I!0rlTK@=L+Vfj>c)Or6n9Io;`kp)0J8_rvw5
zK;vk`-*6^-$bp`5A!ZT^R~?pGLOqD9YC`|os~1oUoxl_%IkUVYH_B6&L_;D1>ymiI
zWU(Z<?-7OkE!bdrXTJMBar^BM^N^-`e=(ZyB2^RkCyHzq{14^4)Wvwp=BqJPl*6b5
zdr{O<jKRJoCj)YnyK90kWGIFKdW_9lvdbYf`3M>o$^l!$icg0T-(9(_n{Oj!4ND34
zsS{ZptM{3#Q9#x(E79~aI3Zxny?T$&m)8b~<+c5$-tmk#?@<puafgfJ(fu9+VI||x
z?Uva{+{~!Qty@{obPt?^Crmpk#Yq=c3VH?#9$pb-NhFFbAzp-%#Rc#wb4UenF35Ho
zrdPpNIz&>DdqNK_HvnQCwPfQG_G_cv1;pL7L390QI3aA`a_kK;4lNc<9Zk_%NdhHR
zqCWVZ&}n*t3({d;XriW<hM6r4p3cZ%bX-HF1-jMM;FE$cdn_aGnUiQHMW5`79#fT2
z;lF@(GMf^VCSQH+$kclySMYoqb~~WK=|zx5x8D#P*mUU4fIi4VBOE9zM$;hKVjm4y
zHQdVvG$h97dGL9~=4G77G93b}Hgr;PXu#=%@yvwh(`3Nk<cIMAEN#ATU}uT3MSZM|
zeQul|5UNto;`7UQ!-A5EClao;OYz$cg5j~8!AU$2pbgMT_zBHqfNbDHy13Xkaze+x
zC{)YJNyWD;((?f+Av6S&^e?(46Gk?bzfnc)zA8PAPN(e4i;3THcu?^sW5lxW*$n0A
zUo)8l6IePy`_n!7mF_9Ki6EFA2Dg2GAWYsPM0kZ9^-ZA4uUHJUsIxB_x^nJbWLS4(
z+x|CYIE&CtOTkP#HvpPbt4*YkzT|fiwrCt5%+a!OI^qX*Klvu9(GQOW<#i!fvjUd^
z&CHI&-kVr49`SPgNqI{P4UgAk*Uvh-?g&;xyn7f=teuiK;{L&HiNTkCwX&ZGs77Y_
zx1VkaAd)w@=Y4<3^MH?Jy-Fbs$9}#I*}a&~^+$=fguXzKfH404jXB*@`XBMB5#xzq
zqWuSseIED8kv)-!x9lW}mm;FuDdB!-YLRWQljPQ<{0<satcR-fmSn|ney9(YIJR1B
z_Td7oqtD0!;`*0_lR+s`IX}a^NKCnm0yDxdfEgpQ9L<A(vGLzc;$jD^o8RDEt%W}$
zt-pU{Vql#&ilt4_zq3uXy%A6FioDY0&hK$V@{Q#k61(`Fxs|zVAo*#}wGqXXD^W0_
zO-K|9-N@J$`43G?B^8-P!cvHMuKZha+kth>!3TpI3%!bEa(*pMGMOAp<b`o90~r8y
zMbr6XNUY)})tB?HY^I<`yd4m-ug|^HexXfAyH%4VEr5^XDK}DX{^aMTTW39^=;t=d
z0L}RpFv{n^Qo;VCkw*tILB)B}l_M;ZQaj`0W^p~ImCVbN%Qhx6CuVV=EH*rp=by%#
zuKR&c(=Kba#!;@~`Eju0;VO0B>}F@U{z`4UkzFfww`m=o79{mP;PJHO;>nwtg<WC3
zO#REdz43qQ10S=Q)UC$TG}AxHYdOqo9L-hfeetH603zxSP6O=Il@f5eZPwhn^=BKq
z++K>0x7x&FFNC_qs#3ZJmo^za?Ug-50cORDs;mD}_C{HmNBL~<9)n!vpNE0)<_P@;
z8va`CQ1NjM`}@l8)v=EcRu6VN;LNhtzWnMmzXy|uVTo)2<fL1qU@`UQG9Qyzwy`iH
znA~pY0-M_Wu2Q!kM(USR@Btm8Bfgs%qx{=w9Ml3-m0xj&h&8@ZMCUfMpHe>Kq<(&D
zaR95QG0yOOSS+9N1^Izd71Z91V}KUpU~^@mo4?^!$0kn`$>uF1C-Q6j`+;TMw+`13
zn%QK&DF*|^m?SuaqH%mV&2VWH8Px*j4pgRO?ds=;xck^7<^pq*(gp_{CQYlW+2jD#
zR1sBc5>?o+JbCn+26@}Jf68biz;NNbj8o?wEvc?Uf^=TtpQ$I~99rf5{I}dIX2VM1
zyABY!Zd841Ad|6Hv`am!&?K7zQq%kdukeFw{U8B(#VChz8U6{~Nw%(Qy92te-+Vhv
zg~DBzWC!8gG^(Z_9`DcD<)4vHO+*WFHL-Q9KhAh0$}`6}49W2SETHk4s7$TIY@Nyz
zmI9#g*tE~m{1+x@`-&Rz6kh2JFtqaUp${}U8kpSD!>p=h^*2?{0aj-@)n@5}xwn>@
zo4RMw8d9+9GGqCC30!%4-@0y$B{Bj$$vwMR*n;@V@U%QBIMayJh0kO4avf>!q*3}w
zHuRT!pR%kV9!S>~pmU5za<Y`!(T8}os7a1Nh*hWHn+r6nqo2Z~4Q9s(w4;*5v*Yt{
z_y}M34QNwjFx7jrLLVUeS}1*AH0wo~ZTBz-I+MH$q-!`LuKWvF<wf_tKBZ2k5Vyx`
znc%2yDVKJU$0H~<aNYkG2v!3&=#$)uCM>)15wLVXMS($Hwl(zAe8eEbBT)5m?E3NS
zlAdFMPV0({6?}vz<9&b_E7Kg-aJON!Y~RqtQ)#L;a&=!pUCV&Yz+TtFaQx}S0-%^O
z-`(M>b@9}dsMUtiTnqd+9dK9tO}$DHzDyBPk=yZnK^9u7=mE3rCHWk%rELTA>Y51t
zct<A$%ZnY9=>5rit)gbQ_E%w1uDpZ3`%pJiREus@aVY>EdzhKr@eq0Uf$2AZ!cB(b
zf6+6ok%s*TK*YLVd4z5``+d3a*2y}ilTkm)NsV%o|FbNj3=AcAVLo@%+{AAh%}Pvh
z5B;ya5z{M;kBLTv0s#J%s=*hRdiSNWchZD}qoY3LmfEP&J8Y<>Y26~BUTe@6ha?5O
zAY!I@m6O*sl|O&oXs)uW9~mL&uZjO@`l>Y0T*&!&66ITB^HQ3CJ&R$eqFN;sDnkF?
zEWxHx#-q>F8AdAEMctv%h5rHMifg`#6f~)i9$slmsi2D?SqcTrrZ65U`+DPlUwE5_
z>yN`pOUL^)cL<yQT@64$Tx#}(W13C+C+7yxosGSUs!tZXJ|E)6>t4wubGMPWK2%4@
zpv5dW%@dpQZES7KigwVR;D3xJSav0*V)*ZUi<VT}OaJTg(OyZ9h$70zz|TJy{l@)H
zRJFG+fwB8OmBd9mAxw97lf}%B_rO<i^b|VRemty(ZkwhaSUK>AQ1;s!;MOs}fx;Hf
zTAE_+zdtI|nB6V~JSrR^8QwFgAs9m7S#@Ydw1vtg=Klsn;_(Cj6W`;X=P7Us#xSr%
z<hl8M<mAD}Lpb$G-hcY_oZhAGT>nO%C4d)lKVjhC@O8Ps$-3K#N$(t_2mpW%&F*D^
z2yn2PGXO>Iy{Ef5Mid+l-Ybr5)n!sdP=t8EiLhnT^S2Byi?+BBNpOwDXj+@YljHwB
z2LV`_aNzfnnFkrZ`ybLHB?YXJ-)*Y!?W{UL#>11t)kG)ZfdFM*u|>rSz|PXiVl#IR
z!22Mtiv@Z^ffYeObz$DJ=8hWgVS~@HuJb<!`j09gs2}ka8JS&u6aN3DaQ^QTnWcMu
z{*->4+Pl4C?*5m4<$wRBV+X*Q9qSI|9V7qqF#qo({NJOC+e-qrkmcExb*=xu1HCqn
z{LhK8EJ2n4L?F~CAa!GZA{l=!<iI(O1ZDk9I3@0<<B^}+Gds6=AMjC3-yR$tQQ%GX
z<6qaA*Rjm3$K8**Eb%t~D(chh%kNf>mUolvE3zrO<Q&H<**QLY;OEu*rE2AS;fK+0
z=UwUN^vYUdbyVZppX4xBeO;C;*4&_F-m$kTKaj_A#Kdm)x#CDiuO9cwy%n}ZZ0A`X
zH23oX4a>YpQ(EXCfXOb$3-Y?&8wP#|c7PYzMdou-=e%rPVzK@Nka?2<v&1NKO$Sv7
z5+*_JIG_FU5TIuP0m=6y=nocc>;7nl20@p-A;45Y39vdHe$5_X8n`Ha6;Pd@;thJ=
z$(ZGE8U255A@A#NW}~Q$Z}<z;E^u=77wPJ-)`Ux&uoj-+<e1-)&8m4_$z9hfr-z$a
zc8q=C9|KUh)CrF#)ux)W8D(duKdB9ATr+;bx!<Co?8+tP(aD|n^*2(v|MRyKXkab*
zEULh3NAyphHg8ZVk1gl1i~8K}p=eCIhYsVr<M*ns6^m7!H`AAs2!7<Hq>&TnXpow&
z7p=bl380dP{viA)Hpe3g@hGQmievkJqiAI?Zzv|upQ(${B-m<@SIa>2=-6!)D#xYA
zZ~Mgb!Ro9%rm}rjEzI)xawO{K!LT~Cca(iJ^1Kt3apk<N<Q0HfAB%IqE9eU_mH#6I
z0YIDjo0ebw9}nvRRLm>OzsT?D?g}tti~<);n{ME)yV$BFV5WQkye_=3J^#ai<TD?F
zV3hzZh0Z|4_pSNHD+cbL$HVSbg*VCsK0v(`ZhJcW?y^>X*|jP1m~P~s_?iP#yMCn(
z<Q4)P%U+pvz9T+EOiW;qu>vB5!baPLkL~WK?L)`A`2QEp9+dU^ish~7lCL`TrHwOu
zuq*`taCwSaDAoG;ZMlefRrELZBb1BIMXbjy!`xOQKb^#eMveSvLsI)5rBE19m9lse
zlU+BxuP|0tlm%<f6Nstr?~d6EQ?Mf8=od*DIF}6NYI9OE>`H00G?Ab+qBD<T#n3J7
zPXQlqQNV(^P<S$FT^X8sI;Mv^pq#(kHYt|w*kthGj8{u>V%e5m5O7(%br@lfK8$nh
zU_38vM0ltKaE8uWE6w$j1FW!13sNX}L!ER=*&`26C)5!Rbza?GK7Ci9DY@Zg!#rOq
z?tmv&8hw}Bx!vfO`(*nY;)-DtKmagZqD%qcAblW55=k@|00C4jDyb_^4C+5E{RaTP
z5kmvLZw^4(Mt)7PzxKTW_VX)edKjS4i6Zq`JiMZg8PI~*4BCf)OnRbM7bJTB%R4_K
za)3<<YZ3rZI)?q9PU*8NH178wMcehGI`cHHvVZzGl*=cxtD}_NKvWZ7LIj=+i<W`l
z(W&CNjxxaXQHu^*G|Pbj`7D}5ggsW3N4VagYyZDr^U<%`8kU{`V1vG0Y+0n#(ZH!6
zzle>eL@|>J;00-^h&*TwXY8iweFU%_9!o&%(gGmFGU7$_Q>|CXAFW;At#B!$l-p~5
zvP7Wu5S>H6B{G;DFriq}G3vZupDtbB$iDvLZy^>?jN^09Ajt1JTdJr9PHlSSMJgAq
zUxk&)W384QuB+73yt|thGFog_Qt?pSs5m<JlZ{Pz0{Z{YZ}D<|e%M!}y~C)Wk(=MX
z4w^n)^RQ?ym%`?oMPF!JE`FC%%R+Nuw|c++bTx)V0bkYuP=kjLmlNpiTB>F>qsDBF
zd<+b=Pbi;{DoD-*+k<Rd#^=M#4N{yQKZ_}5KIl<8GZ(@d7{q<oi~2!GWm?nkeLPa+
zbIO`s)h_Rb-==ac<GoW_7@vgc+JU`%5cCE|1h@xnNpAwQ!<+;SiVQ$-00&k+h)8<N
zk)5_G0^=PMeXge|C9>-6_q%1=2EN6BeYMp*O14CAcXh=6HRYx@_tG>jQTFx&DzCG)
z{m}iZSNYE?#c-2*=U4Ujvn+&P9H~beI!cOI;27^S0zJD#{E|lY4qLD`Zf?wy4dV7M
zSqgNVSiqM92{czy$A>}7<OckVwA>n5NlnZ+VgHrW{_kPQfC(!ja}uX6Sqh9@35(Fl
z2nv7CKWMt?%(cj;Q~6R8GE<h>_H^0?ZmDE`H@|pIe`Qwq?LOGomUpign&UC$q@OgE
zR5rY6=$8K|QSZm=Y)<!`m2XvrO|8HtSyOA`Su2%dok&RWOueidrJR5yf*Rm%x+bh`
z=hcgf6_*z!d^WHu+c6@8h`!g93&Z{7K=@DJKJUB#O`UKt{8u?s`)bdel3mBPN1Ifp
z>*mgv9WPqBy|Pm&*rR%@mSs&~g;x2^S4^#}>-*E)IGDS6)ph0@`Ksw|*CV#+eYeMY
zB1(Dmk1B_GAnEam&!EUMN%6(CW4+Kb-|(6J-6ZS6rwYng0!yeWs)mznIg^85Q=(_H
zJea)l`s4QFErBI(c~d1Y4`P70V(FDe8US>7iGDt#eDDJ&y~oikqn*mqv-6)|5;7cy
zIGb--SKoyQJF1ZAz*1G!YMJC@L8c$x6kznkPbqG&F1EbBq%A0N@|?q1TUOONHLh_L
z%9apw=Oj^>kr^=zJ(jKM1$Q7)&K3q5T@tSGv-cwapd8O`QB%Vd6BmL0>z81FJ&Rl4
z^Ojjdj<k%_g(iZL^;2UhZ!<;(yzpe-YhUSq%|BpaWh73V<N;)gayK~|6Lf>MR#%e*
z*lRxpc=IV9DMLTSc?R+X!z}9h%TCVA>pcM@#|Ut3ni_~n6}GFJkWErr*Lf{T$K=Sd
z?jS@pLuKYeD-&v7BEP?9_7>lH7aH@*Zysxxs}75yu}k+2V*uIeC$0R+OMs6RY|UKb
z-^+eYwajGQLWuk($q3}oaRf5ts*qBk3T7kB<hDZIWa=$4pAeSeLe1o|L<13XSuFnB
z@bH+o&0Fx9@T$PwpFKaE@rn?sdeO5z_YmoN@yWGfWE{VW6L!Ud?BtKOXe`m0+Rudd
zGp<~<<YiBGjp!0T{+2S)kCKnF_i$X`yP2E0vDgnui3Bxn8+n54G2TXyzyE7=(79wO
zgFPA5EEMA|^5UAs9+gAJ(qwo)&WFy+b5`hAQ=7X)Wn!~X%9wPYaK3$Ybn!S+k{8}^
z_FYUo<CLZd0K#@LG2Q)4y0Dk5N?%wkeUFKtQtlma0!(AA>-~EAqd|AbZLjh3UZ5zS
zfrC;iU(vAn8J9{{U|VqiEAG6$`2E;XY~qTc8`dA0u6xM-($t?<9=|@YTY(T@g<MFl
zgcv@3{>lwHGN&wX=z6)+_~p}!&JV>@iCMmI)Q{pDgsggLo{Xz4BV2{X16F51?yw~N
zZ(}h}v=N&F_QOwPA{c5Fk{~9D6KtyP8F@M=fDHE*a+0J~C8!&Oy1nk2Ju8L!U-k)*
z$alTsQD|@eBla0#EP)DIY+cEm;%{v_<4?C5nmGUB)ArAiP?H!}07bSY)Eiorvonf2
z<;{9XY<hRSJj!B6-@QeBSITjyd4*@Q!5Fb1L-l!3<U!*jYhVbDz0WPRNh`BlTgx|e
z{P_p9evukq=|vnWy7m|zQS<BZz2m7+dLx?p>+v%6nU@P_*ZJK_SdK!^H>Tx>?^GR=
z*md`&Q8;k;wqm!NY=DEK(Yi_Z<7SHNhQF*J!C*v!OsVlp5fL?#Nm@rGAAKVhkE=tJ
z$X6>3vC;r!q?(Bj4&4CJHthX*2RGpNVPiZrF|850Kj_(TC-<zT?<ah4FT6AWrk~q`
ztIP{bf_D=_$Cb5lSB*HLUFO)B^o#LRs(6lAAu(_3248*y9A@PTw3<>=W4b$MsbR<1
z!2iD2K}><-sg!s10$7GVfS{siLIY?}X){0Td^%H6NL4Ls&1l*PQ0^&JrJm{F0!$bf
z^o%TalQvj%Ng@#=z6p=kt&z=OWbAqGZ&xR-dCCRlm3$v@g52s-JTw}2%l^ElJyMw8
z0MpFA!FD~KtueB>=49_6s!l5SBBS}0R+}u0`wKb8U0v6#_e=9~{Vd-SX6a_JFP}&U
z^_6W@KOyRIRfqg%t$?E7k)J2-9eMe+8^ctI>v@-&I!c58jv~%`$0tQ7ic$FOG|MQS
zz-z4`Jsxj~p*zPycX(yAw&hi%^1o)F>(K#J1rDD6YFS3-q`OqTKMQK2qf!Vl&=DEf
z0tf_sR(8^RiOOzG{W}S4Yu>fD*b>gG3-BK2PZxeVQxyF4bw=JT9pkuKUyQ}zr)uFi
zR+y3BzmPBG?7XY>n-NQ0p^P(Ps5QP<)>+w!9RbX9IfXk6uIs*oP9RAm0{<OZyLz=B
zQy=o9uhsDH=9I-ydu7f#S?E)qWfs+%Y$N>NpAxx43P0(=ALEda{_4aIdS2J&?$yXa
z0!_!sq$J&;_tR!vM$b+0HC;6;5gtq`3t_zcJ&R?RQowiDQE0<kI;)!OLS_IjxyR7O
z%{V2>F%t4)b{#&Kl!+bw!#|pxZ`FHH_T-KhOD{N&JIpKDPOLq&uwf<c%((TVRQxOQ
z?t5IceWOMO(<#fSCGv&`9_)a>WyM+r%XdYRTz#X3>kd3u#$$np4ML#Hv@dzmmYY$9
zl+O;TQa>Ge`5RF-0}mxkjoG*OIKtOA6ruiWgh(jTl?W96-zfDvqL>-{M`dKH6`nMv
zOtZOpqAWtGOFw{FeAHsD@i=7-D9_MeRWjvM`R^DUoNfdic$VMw5)BI?(^;@4tXz!r
zR-U_!PaBdu<bMLhF8ec@49>ODXA-MOqg?TqWoO#IWkK+g=nL8z{$8G`P3!?i!mt`#
zMUO0ggZhP2t&<z13*FS-U!-+d%#wZHx#W7JCZu604v2gC<#csDPkNNe^PAL#4x4I}
zkaL>$<;{9CEPyqg*W4D0j@8559(L0wHw?Nb8&jV%OI@s)6AAG4f2Peje$ID7f*JZl
zgL(ByzT&5wXw8CUvudm#In5MitlxmhwwPhW!!d!{$>p~UA9Q=EA7F~9fl4P<sS6je
zzMs+lP)e*e{<b@>xB*c?f+wiKZ^>WZ9=%-hT19KVqHc%Sb^$WIfSw3Z3nO;$ZPyQ`
z+64&HKcMUYKCfVi$+rF2-Db6txYo5n3PWuS|DS=3$`3o2w8&hekP)c*?hLTLWBw~q
zk(u>ToF$N<7WQbxc~*vO2eOt7e@<sz4)a@Se89cb(ES6Y$-&dkuAWgG)S+6X=3-iI
zcW!=?C%%|$`il(Xuk=yiZ5Jceo0g?-nt#p)^LnJQ-~3W(zQwC+ksTGg2ZgSqXTfh&
ze)ZG{9w)&)JsaU%P>ts_H|do48;j4z4;~k2fM&@274ltI96W7tdz#XKYR)5j6%h?6
zy?t-VmCK8l2yu5Jr8_ARyD)B5IKCoZksU3o*J~luobjQ`nb0!HAl$(kud*-a+>i~E
zr&Oi?=*PawV)ffz@ZZK;dP%4?5_A_^?FGA8;Y^h2Zs5+I5-C3qtb<%4c^5nXL*wPH
z*|5RHXvg)a;<#EW=2ijQDeMlr%SSJC?vUv@k+m5o{j$Ar*2sSMW+DadrFvl0<84Ir
zejqNlPMg3z3o%8Z42}_X6Z_mdPUNYexBndSuRYa2{&JfXB{bB^?wS)VU5;&N42>Ro
zwyOEaNbGwig_%x8cr2Ic^z86%8kQW(i!jF5_au(LcR`-wB6+CfKpfX{T?#*2TT|~}
z1?Rt^WAz?F62ef27cEJR0^!sA+O5G2FaBrcxfoDDfm>>!dtmf8tm*8g`<+I^0^2{j
z*jf!TG#2RlZc>}mD$-YC1bV{bbFDxPG-hHcV@+H4Jcjm8<wC^?H;f00fFZWCh%2N#
zJEzfGR0GHa3HdY~<YJL{1O{8yy={qVd=M8{m`<!TA@KOv6E=<HBHmC;-+2ZVCnBzd
zk?#!gl)43*(zv3da0TsD;xW9BDXW#j!dqll=x(_2_pb^+92+OW<Lxgq>^r~Px8qle
zK0k7@ePlKoNWs=%C^BWVp<l<{^96JgCGsvk&qVycC@p_BPYKqZ0vt{?vIgK9;UX|8
za+%P*-e^@~J1Z^Hm2pzC3u4v|(VMjHuK_kM#(ab_e5q;>?NBeWoXlU|lx)=OAas+8
zSffbF0kV4~rj*zD>kq^~Yp`Ny1!=HGq1E;FG=4<aC-mzTI_Y22cFI`N%+0bZbDRc<
zH{r3}OY=Cv0$_*O?T=ptIBBmlsGVxa`RsHkHClAGh3IjDJNzC7c$J-UfMsq;%;z5Z
zJVjJs`%<Ci-^t@Dswc8I=Q}{L$UCu1tXXqDSD`I`WYJQSf9$9nHkn*r6crsQQ<wlm
zX!_?`KeV=ppU8w{Kz}1ki{8^zGplu67=X-jWN(deaL5bp1Z9}Tv##Em1s|Woz3hoE
zdmPWR>VN4*C>l}*E6?Y+<aKB+JWL<9Nw7Jy2aD6mQf)`TGVDA+zhRq2JsEcFJH;iX
z?G>Q1ziqnxR;B|t3X~_gzxtfDy?XahNbMU{1%RsHw5`B*%ro@g@O9vJgdzD@ADHbZ
z&IwP;b$jmro@5!r|Mi9^5Hd}O+6R_NGpjzYIxnHi(Vh{S&}b(C3q92_iI3R04PpHm
zQ_3@f1sCm4{e&)WYr=t!Bm#+Mw;3_I_Ef)#&=*5t6Uo-|{`4M+o;y^*68;gdsODli
zXhf9VieWzk#}My)my+O{l^DBKFz!;FGulS_5&I`w!N`{u>whIwSdcWVU}YPsX#1Ms
z&yHn}16{Z7L}h;d%xILM?Y1i~jpqmEb!^bpU7@{GDq~6oWMaEQzPbqo?;Z>jTzjLY
zmV~Qg)f5_%KZXpNThdC-nRqRl_}v&(X%0Tt&YmCR`93P#|62jwFrZ-n1H)sOG){sw
z1D!eyv!1my&~{ulJtl2yA_jSiHwYOCW1^=f5UB)6hP?Wazh_a=<L*i}zQ~B)Ec!O^
zoUf=;Db@JwEnSrN1cDk0?3u^-ru-ci1cqzA#^O4M_F&$RRltw^3C8-dcNukf+3Jol
z*XpeRUGmGi_{TtR!7V5Ni5}mX#WQ0-cVtzivCLVWlOAx2uaJ{V*YceA>KG6-N3L#~
z{VUPqoQ&v#NARz6d*UJ^(&Q20xbp8YT~S}O^Un>OAY=^OB1p7O{e;j+-n^X7DjnJS
z)L-hhVXhN~GS$+^Sknue#3BAsCn-T<IgxqdM(x4ch#?Kh@EU9{&lq4D`+~jFEhIa-
z<ZHCC#%--uw%t(KMy4`BUGdvjWnx~2ibU1g*!J*|m<D0Hv8D?}djxyz4mjMba&m!r
zgZ!9`E?QlLy!XKrw|YtL6H^~~JQZ{*riX{*qeI#vwKjeqferVU9?mc<$=zx3Xw{_I
zYcW$#z>g6Ihbhb<GTEve{Yji=)~L*%$aU;nV*&ldVWKQ~W)exIEC`e7Y65*Yv>kaX
zP(KLj5jZAbIPrL-c7PR*)IH>gLPN1bGSHT99vw}KoRg!ONg|t37n~Knd!yIY>DMbl
zMWsq-%8>CV*SCSe@yj=-Kb%9<&cn^W{9=6+c)t4L>5jXe2TO0`n^x7_#o1s>Ie{&<
zeNEZUsQ;x-jlCie$=qgD8M~2!-ju;?;!VcbAG}ZbbP6T;a4WN6ij@xi^&5uN9AOl-
ztYkX1Vu=Dw`nD7+e)vylXgFE$w+PDMsnF*@&!9#ri;*9%RBh$D(AB7Wf+0($ylJag
z#~zDqt|2zH*s&OAOdLL~K9Q5!UT3cuYl;Qsj`TMoeJn?vZ77)$Uo1PM_zJJggt;Pu
zr*`x~)0DJr^syBGz*%`$vWY7tU%ln!A2|qer{UO;8IUVT6lTHc`SGmtKZzuhIET*h
z_7!M#k=cf-UuAo1_zUPv-|bWX!j0ni_!j@DOs4(yKwg*=Yw?IYo#lgf&jXgnYhG1o
zZEnS^B`czQ3YiUA-16QhAt{K&u0|O0E>Jj%`~t|dgAR4_Mb3};(P=5}qd!BgR(&Y%
zS_e6*TM@0{2AanFB9P!>Imt!8<@Ws~m#ha=REPV~f2Tg~jol!Hl4~Y|IJ}4^q%u!>
z;lZY9GY;e#F6?_*-^4N%xD7$JpalE>TJix?nV&`pKP6>Ou%l(_(CAi6DKZxKl;iUV
zGJAmlcdASLFsia}CK@~1-PjWaG}Vbw2rDPl(@^8|q*PAZ2%9Z5mJQkGDTaM!JAgH!
zu?D$mU+kGdpI+vp>zGrd54fy(0kLYjcjI!T!X%XJTM@6InEr9V#oQQ&S{^8(;FQf`
zOq{iXm5YVMBq{W%lMVLGidr<br0)VAOdlt^zpJn3_FCCty*-caX^mNR@s!NT_dI`$
z-d)f2q9|@$hXGkD?BE|efjE1fiSZQo3JeGNS*sKbw3?0DyBtP;E~qyxl)~%gk1H!&
z)`VCpO0G(OP<gd<ExTmUbvNx$3g7JMZ8{6KQKT|JD0d%+mlo^gwp&-6jX-;BTK=Mv
zDV!<bmm=pylN`%Ks}w&g>m-WaDkq-DNR`EaLiCy4XVCp7<NYdWE}lu?W~d!&!EbfJ
zqEPqdG~COMuOSILb|xog;t>YDqCI@elb_D1$_CGpu3;aBg?@OfGrBsg(e>Z(naFSO
zPIn@f`(>v=?46#j46ULnLsER2LuvB;WJeBHzuzLZsi^muCnU8WTgPR8{$A<{#*D+h
zlJTdaKL~a52-DuBF+15_{OBWhy2OP(6SsoaetMUZ64$qiF@k>^|9Gp{^dp2L^IPXQ
zPR@%)Xi4%ic1))*+toQNBZpZlU+8kwyL&V}i&;>@FbYYrn<yhl=S%CnW|!1YhpF04
zGJ2K4sMK^))P?}mQ7x)?jvk+gNQOu8W3wG^;kIj06Fo+(!%-OLkq!Uz7gyqdtjN!c
z;yW+?B<JW_3~W%g7AHYR;QPFT{?QhVew#c*T`}?$FurD$&!|*MoCLA{1BFv&eF?vg
z62XX(8sVhu;&5Kfe~c>cQebPN(M<Y1o3tNqp&Hif0nx&O4~5yznBFHsNNJsv8rh^Q
zbeGU2NH1a5yg;IVw~s#^#;2x3DK<#HLu(@qZuK=ySd-$K;sZy3Ph_aV&XLfFATEm_
zN=gt)NpHSv9DFV$pfo`xUp5tOAI*BmUzMr4$_ab&6i}>#qeLTn!bxnO=Xr6w-leCz
zEjQ-fO|jJm_lb*y7ogNz3$7V)KIUvp@ULeSY%KPwZ!xijd?OZp#)(5M;=$UKgsETw
z>CD$v^xi{s88Ru=C|l+YJVdbdTn0!_tvYi2?v@VE-~N%X8~7XiCMZ;blOP})l#gsu
zWJKvtnfIWyB_x4PG&sC!EcN*_O(122Dux<u`lZIuQ1W1F?1F2)+f~<gyQT2AiQW4K
zm3R^IDe>v+x5IrMEF}AwP8cYmR|(hVW2Znd!6}u?NRTfSS9)4aVGd{0O-i180R1Pi
zzUt3l=G0c*yyL*_5B=oFQtS3rZ7}zE%Oh?RUYr4alv7~TUN}=U*z_ly=q*XXyH-NU
zT7b^~6Lm|ZJQ<`xjC+G_AH1`BpL`7ir!n|D$RbKav|_0}!7n3e=+XqkEYw|6)L9?T
zC06M%WV=trznOi~^GnYIXE6w~jp7;-bWFTOXu*Y;IpP=U!(mQ$JdI@h*5Nm-97=qs
z%>Lx}QQMGo;{UVGN)7obDRHu?aXE`M7&D9)-X0`JoZMe2)k_0UA5}80_{lSC-7vZ^
ztebM4d|G{%UdB@?)$F3;YiMOv!Ly+VZ{nKU%idBA^XZG9`Z{uY6p?Md%1U9&8rv%E
z{(cwL9hIuKp7J2(vUkOlQog79uB#!<pj8sM=A|g|xcQK-%==u6vAckkvA(k65ph?S
z$GOdbGU*sbaE%D}B=-|N(`YW<5~u!CT-!^QjpT;x=n9|B(9vczecs!D$&a+*i}|09
zwI!di>qnZ^6_9y`a3mwPn;AwYu)XfTX>cljp6MZ>s>RS+8AQDvmFDwjh{!(2Oy!Dj
z>0z4_j7xOEP7ORQx;mida$-#s=_dkpg{BWcHlLq;F-s@e18(X#szL?y4dSH&@b^Bm
z%W^I6Sm%Xa7dc2Kh9}x$mlQ<J()xvA?tl0d4W_|v_Jh(Lzprj_Se0p&fuESeC7&!q
zT^@!Jr$x#4?m*_1EbT_<x%hHxU5Fo}-Gj62)2R`sTBYmQ?`<;0_C8ezseCf`%YG|a
znS3wRgKR7BwteWA$Z|ZqfI;wPbK8J8&lM_0`)xBBO7s@IX3sQ9Hr#A_98c~?yzWGF
zR8fh(fQT_(murRJ9H5d6YP-xYM$RWPq>L6D=u}NA@us)Ck$&_zmdbH2Y|h29M=-qx
zB(j&wi;B|2Z{nH;ZnYE_#DHhI7f$WFOc<@7KQxk}=R4>pQjB8s4mHLRv6?;$GA}S2
zQXtjp&I5-0AVvw`P;$pm7JiVHFvXPZ_#wHb8^2a^y>4j6?~p)^@ep)XlbTW|kiyvn
zD?I{84Re`HRM0oEWJ8F8G=g3GzW90tCTo7{a~t($4POkgD^&Vw1v15S#@K?}f1kIQ
zO|wh)XA4W}&F@mFpEA_a#A<k4Z#1c(aMI}ExCufyR2Xiie6G0Dfg3z`F%nsUGG3rI
z{g%>PmBl*c6qV9n#fn}A@)guGNXCiKE&h}jT&D4J1{1H!giAd%UCDrahGn&T806Tl
z6&2wtJ(ywq!THt3C{b|(Sux2woa(O$^rJ)}oIE(R)bO|q^oX6I%+4DKq~Od-SX|gF
zCj}$%NQ<Y&jOK!AeY3c^Tgub>l!Gvq;HeN}X_F5D65xVH^NtVH8tD*EjmUnYQ}%?#
zZ3%K4Aq@kFJLEAjHsNW%o!%}(kmQ=XM`xHfBd4O@t)a#kHNWpqff!{XRi~54yo5=#
z4USr-Z6t^KmvMkZs9ic$k4*~kfZVPQen#~RR!ex%kUq(8-d!f9rKK=K-(|)m<!F{M
zKwTOqlF%fpu}HhlmH~5Z?Bn`43EoTWwJ{%iAkv!9pzBVw22F+hDWm{>+~ys=lKepN
zC1JahhynxaYSfAHy!D^>vuP|fvI^c9)+|=UrGH@XUqkx^IVO!W;*%N6|0>P^)0^j8
zWDOtF1Qc{-aHgwtJzdWbjoA<cbV$SQFB>@o>{mfP&S$*Ik1zeqMch)wR-<Tq9<V*m
zp`UyUcAmN;$ud802qWVcG(_!TBlL0$!87m~6+B^wWoNqDQn#f0{ba4Su3+D)xJhCr
z9p7lXBp=;j-&46%9?NVRt1|yN^bAK-alQ8Qg2=ru5u+?EQx4@YBExD@4<XUc1U{8V
zae64bWGPwki<G5WW#C9-O)A4n#?<|?JL#^KZKi`yTyCEc^A^+IF*QBOTiL5(6aFD=
zirdBQteuJBxOw?_1SpaDs4i6zC26;)zp^S)Pn>eszn}|a%X%LO-Nla-+bWf_t{ZW$
zfW`Y9s^E-Bvz#54``NoXaQ_$V2s0TfnWWMjyu=@2E;Bedf%XGK*d#@_ozs0{B=QX~
zYEh@f90jOhOI1Q2Xeh2-PW+ymB!i)7(d9n_SS}N$0v0f=`<K5P`mySlb(&ZH8Z1S4
zUjggfyeK1VsavBR{pB5X$?khW*~PuLey4>lI(jdJr`XtU)4{C7sm=l)LOhlPOH!T^
zQs}qw5y_pFKfVm=*g`-mAgsycBsQFk6ViR}Kr)FQ`@uu~dZ+>4prPI1aL*>_9sOR%
zdvP=5dItnX;evOpnv^D6s#PQRp{W#%u(d%026?~CoJS<n-cix6`jgscFo~gfh@K4m
zVhkGg>_I%~%pN+{vc|@aT#ezOQ_(XH{hC*-(mI0Ht>jTDrVE6+VN*Z4PCH$wCU5>J
z(8&D88OTjE=YA5>DIWE~(-H)HC&m9_b%sfXPc7S}x<pbn)AOM7;_MJ%+6(ZRRAKQ!
z2q3t+1!21a!w^;?X*l@BRfc2&o;oB(35Few#M9DqmE}zpSvYPmjYb5-DwJaDhP(Jn
zD~J^oAaGyV9<vLfxx9yn9&dC7OkgKGO?FO4V4;@gP*VR)?eH)CeW;!yr0Ru~IQ*9~
zPbCaHJcG8B(p7-&k40)IS%MEdlxREJiHDs=uX)Ylhn4=-ceri|AK8RXPV}HqaCzhf
zL!<H?3}>%DDC+U>*_MEk1(cy$lHUQRJhU#rhN2w?$q%5D%U|{mneVZ$Tirx<k=`8P
z#oW}09pP%JTNnRuNts+X*@1*pJhd8*B+bA~?D$KEu)dne?B6c7PlMZ;?LkZS6RsIu
zeD6u8pzN&4r6KHxSiv4Opkh4A8&bfMqj2C4@0I0ftlTMt2O)yT+s?41!SQ*%Yi1Oe
zbm|JmiuKxBB;ZOruA(wc@TZM?)5l1ohlWy&Zrt58*4c{WBqk+86sMy9h_O|TRX2FI
z8UQj@eGbB3(Bc6boPF0%Sd~)2uQ&8U#Fo>enDc!rnZSn?JqZUFsa2r2wHUTYC!N0S
z!u~rK!b?u~#eb}<O{;!@x3`lr)eEcsTlK>U=ga!*7TD{0s320jiT=*CgtN?a+8qFe
z7C$&-G#U#?mkODds^|xp1XygL$vKtv&=*X{xUB1rK(^hVyW}wkY4g_mscsFE<e?>z
z%^1#i<Q7TH^=t|St~k1In_=y7!?&AM1I<Pyy5gs@tw$xa`7w>Dt8TK9DKYtBHMYb%
zM)CS;i^D;m)~!ZabwrvfcsM&ho<_psKg>Kbu%wL?5`OOS{~m4eo6;B(TzXsjR>A*g
z19X6te(-HC=!i;2#t~bpBjEa)y@x}|=KpE$EW@JQx`!`F3`h(xbobC9-3&uF4vlmR
z0)vQ12@Ks0(j_GzsfeUBiXuoW0t1M2H@vq`ob$Z@=YP(p_v6du2d>Lu_P*D>_S$Q&
z`0d7UUa1Fj5Am_v&nwH{=RL1bI(%7Gz8(nw-pl<Kk?qh%#ek1DW{%Ax^=lp=Rbk|F
zlaD`1j-ettv|+&@nzbKIebjSL_w7agZc=`sYRR`qw+>Sc%)T8a%AHAjc?#qY|NUKk
z`t151IomA$5ee}OMFtpOUWu28&e=Lkc6!9zFNw~dr<5+#fdo1iq0#-*`^}jeLn|ZU
zegvuxou;v9dMvbc_@V(iyv6#Ps0uMe$XIpMr0Jl!PP!nFU)7p4G~V`A{R9cxf(|+L
ztt{>F6@1di2;{Dv96xql-9b|$M4J96fH>q1MpP)k7R1s4+mH@*^P%A>0!en(h%Z~g
z<I{qsa8SQwL80g8`_Jn#JJ#B$7jD1_6pH(Aa}S&=bafFXyN{9m`A>nYQtbp?a$)b7
zr0Gc|m6`0u+DuUa?4V(;Am|51>wMbbS5{ZM%{x(O#hcekW0Hh+IWc}8IgbZVAbES6
z$4O5shC#|NwQPw+)%?G)dJ;7DGE@o8MzG$I#A!`qf8F4|EbA%{F?bn9_Wa{`)_uwb
zMr8u-#?GAS<=nHYC0{<x&7$5=P=M8m*@PH0bKg*N#yhkkj4m7~$Y+vcGG3QP#G@9A
z`<489PiDaOFdlt&MJNoR0Jo2KNK!#T!yG1=E}_mn66=#Sb27<pGgs+H`Iy*XH#84A
z|Ev&#ysf-B%2SWKr>sZ`BgNB=IOIi1qG!MR`;ulKd285@tCF0=036Vtz@3&~ckZQ-
z!heihm6=FipWZ-e7UK>X@D;Pz2M<oVuLxQ2$|(A&xWob)fkj&1n+4n~B@u4HXKxV(
zwvKbE)kHK`tPBg!e#o-@;2l`p%c{Y;u&=HZ<gI_In6?@v`^}qqgXznD{23`R@f+RU
z-Pr5xipvUe692Oc8y$SVAwNaO0hy;@|3lqlIWy{B25#l^R{$KVhunf{01Xb;;-I6l
zg)HS;1da{99|%L9M@LGOz8#{e#2Dln4BF4^!=*gDL(H?kgEw<^+WtpYTW|-&E7EI;
zdR88WzGsV>craj3ts1I6RP2bLZHMbVZj>Bi5=tP)dv%ziX~a#ZI}}3A|D~oD6vv7>
zd&@ILED$84eYVX&SXaqd?5+k)qpQgj(ck1P)A2|~LffE@UU^YDY89`pzq4Q@Lbxc?
z0Z#-&cz<jF=_;!xGVw&ZMx4_Eg50*?+JBhIN<(P-qBAPwq~p1Mj6KR(6229;*dV~8
zwGMNq#gj-AjEpXfb?wk+X|5&nBOzCCThNF}Z8E5)yX9slncS|7Df&sG@ysS?gC{QB
zwGkZOM`ge2$3kb<rdN4HDfFgf!fIOj>2OdB#5viSCDYcba<?LYZc8(qBd@BeQCG7Y
z;;vm;0Fq_@w6MIRV@TAgip1hgJ-Kd6vGNFe<fGPj7tY^rx62+QhZOC2;av{KMyU}m
zzH0>4&LZEMg6f=Z-IC%|hbtlk5!rUirOG}I=MJm7s$2*rrhEGaXE#d{iYs0p7~0Cw
zo)&qRLFoQiNrJPE*J-yeG*(m?-d}#6dW#&q=1Vu{ogh64OWDRtR!p8RV@$mFlZ1Hy
z{ZrZ!%TA2o^}<8?w6!{)rv%4}K?6lS!YN-XK3|sz4!krYa###Em<Z&+6`$otW)Zo!
z`Bm|T7t(RAxbXfMLvTmhk-wNiHXAPgVv*Vbky-hP5|$6_N!@piKYk#tT~|2{X9%z$
zeH?!<Z{=>E>v*8I`dFmEBZmFh^XZUp<Wfm{zaqlZOV*n>-?;BLF#<)3MRxKe22-Kw
zN0+gEg$ff6h?v(tNgxRu^~2E6mUtg@M8!=5(QNXwL^y|!7VLl=P|)ycg%vf1*;DJd
z4IEM8v!kcSo9Qo_qehmlJumgmh>~4-gQ?YTnDMEJ!92m?PjQP$NTY5JG&a5H3tQ}D
zO4LxNybxQhTD3Ia4@!cyu~ZFj3ZrFVh;9m7v8ol@0?1Q`0M&516Ot%~?JIIqzu%!1
z_d-6ZPUEJsB*he1R<ay6ub(scV20en3N=2%VoaSlPu9B1{0*P_&48ZcF4H0PmntgT
zL_vr17=JCG$Eh@}UC^+aTys-MMehR)t7IC!h-0I!CKoeZ1>1D+vU^9$5jUK?BmmuC
znl3Ep3iS-i(3gqu513@amQ+1Y!iA1?h2lcy9yYp9!@Ol7!D9c;kE&RAGrjsu;@<_A
z16OtFn<j8ZRbc5!@VctV)sH}e{9<#VZeAop^y$XCR4cEl+}bSReTL7Q+&Nj|ukI-y
z-w%iBw;pqmaGgyOiRQzo^wpCW+_OMkZOWL+tnYTgTsK6}A^qDfzWGtF=g=bXFtso?
zb7S4Edf(36BG=|+&18B1Xc@LcD)t3u5^tIKex_;YZoLg&H;$KYO@=CIReDr)kd#~>
zfy8k(i|aQ3?jy#5rVx&{o0RcedntxlUb&4BvnOpmTESCtf-xa<wRjK3-0t4j>o4$!
zD_AgN+aiQXM`w(;Tun(Gz3V2x==9hTt`B2YWE*@m;=^0Xw#1pBwye$sY}^<(Nx19#
zw=84)q-XxIuRc}Ml|k0)I%Z7530`@zDAgip^=#2|4a4`E^60Lv;JK;}4i=H`{1ej<
zSn&%ERu{wig<S)(b-H8MH-qt0TEj6RG)naz(>Sm#x=NyQjdqmcJUtKWjsu#2gaq<^
zkTek@N?qtZXq3&Kdx|A#Rxg2nQRJzj8=g?te2Mf%f_}+VI>rj&+^~iU<y0qmA;Vy!
zcEV2)VJHG|Gq3PRwAW}dUvN67dS7ErnjwUnncyUyV5Yqwx0K(&dh%X=r!(3g{+05A
zJf*Gq{FwXxNK%Z%XNhGf+rjJji;B1@+vxU&gyzC1S=a!RF({CAiUFTYZ89eAXowlR
z+jy|T2)|xc&D4<nxZ+UV1{&3VSxgf?>hiGjwmJ6`Qn6Si-O+bywng(HxvW;S%>F3c
z)zldDHE`-wwS_XeKl`C5IrK)E>(H3Xmpf-VY?Dpif<Sw8DHe%`&ECraT=FBkd(H##
zPtpgQR`i0ROP^@5O*L+fbVQ0IUMX4iY7Ec#;J4RBO_!KpFCI}*jqN}e_Tv|95gipX
zOpxf7sKv&!js0x!f_Ld5`svMpJOyRQTZbozg9sQ7L!9XD+BCW`u1}Lj%q@#_*n#Vq
zsf9U&T3m+rhoX>O2e7P`iyyIC(2~Y^v%F@ROYukzg%~DJmn7=VlBmQ!B~!Lx`*v2N
zbcN5AFbw8pH6=VgWG(7GWRL0IyiKvIa5mC1UzZ&__h+3DHW2$#-^3u-Wz2T|QOH*X
zY>MS1__9}4L-!2Svcg0bM~(by;%tt>I3A-Gzn7?g>x3&5YW3*F+-Li;9u)q}k+AH(
zKr(ejn;>%GK1;Jz#-@oXn)GZ#%;lBQTcV3%XB9E@Bjd{UgwIkVYChpBsTPo^pS5^r
z54p8GYkQO%x2Am#K9WbY8JbuYy|DUlE_eUhS=~YBxH)IEsBSb|50O#88oDhcSu(~{
za7;!$ZdA8fk~HSHzpM?NNr>Jr8ZLYM3e;>H-$N_zl0tv5w#7^ugWth3nN4PdyDc8;
z8CN=QWbo_Ah|wvUqrZ!4i!~)<MDxBcYBuD{-tQtGva(xGE%3vflhja)D*Jn#PnUv9
zj!6y{ijW(H-a9YtYU($UHh7YzXJY0=tHJP1ZAj%EymSu)o!rx@`;eYlnOv?uW;@P^
zjK_Klhu~|W*UWRPWUlUT&tu~S*nH73UA0*MeAG)~FTBO{Em{y97G`&?42tMxK^{BL
zu++S}U<}_+pd_mQ`C*lXms!8D6+f|+p8eYhzDvH+1Sy5KT({cK>lfZz(yK2-EM)S<
zFV@SwW`WYDMM;uW=ymZ@f{>8?!YG$wD;41<s<T_BCuAfNrX}Atex3j9P;T4(BszuP
z?0QM!ntQ)a_`5N%r`T3#W&+IjNGT954Xlj!JvDn12{v_tG|G-StShu46jBN3z*eXT
z>j}dzDuiYd$btIo9gwMS?VYY11hi}z-zZ6B=3OwtLzzqcS_*loKgg!ZlzcmW9w~8t
zh^xUA<6HPB0>Z|d+>%mPoe<>)F-N&Op6KgetUmU00Q;jySrY|*l>rs6tGWAl+Oy%I
zX>^{s;(oJwXUh9-irmL~?IbaH>k+~PJ<Y`(3CxX6v@lvxQz0Xg`p7S-uZ&+h#y?-o
zb}CR;dU$X`n1L=TX}3n`C&<>C;(t)M_Iemc3waU_D^~BrOvbh8lUPmncNdD0?f=Cb
zNtbdgr11#plY&zSc$XLWt+>%(n>IJY`6`{wF{h+Viz%=tbf;MH6sBQ7fWk$il$_!}
zJ0Ql<>u5`OCuvYTlFXL)D+^VMaLBIbdHkN^pBZ;<%8+m%O`3KK7X`LRNpK<~H0DPI
z?|j!wO?HZZHAvM!V}5w%+lWX$+ViZVlO|~3t?oTnN6l+iWt@j_I6}uKD?1G|JbSld
z1M_%~aC+%%DkEb-Gqt_l5%Y?KVn*nDA{<kBkSeT7m7MxH__amRmU=&fyFXVYj)p#^
zp;pr#tmXifs~#79h4DPfp!9ahk1lMppvF<kf!lNX(QhuQWs{64qvqoz(bE|+kA%b*
zOp+yytK;`hZ<1zrt@#!d(PK+yCz%UDhWYRI7H8syBLxjWO+Y<aHbboUip*dlmQBmM
zsA?zqQ$0zHfymeCPPpc#!#y3dJ26OSJi>>K=QLb1sr<aVk=``L;IC?GU$!mr8)Z#m
zs0faxz$Eg-v5?qFLmi(rv6|b9h1Ic-F3Mr}k%R00@LrXsno?oucWmPHrN7Sbj{+M@
z&G>t9Y~Fknp+Ggn7*E_Wj+$t@ZwF0@tah=L(_5T3UtbSw0<wT=HpPsQA6Gs%n>0Tf
zt$CwUWIX6J*x;5W7|zm?aqqZ`*5kz)6`pwn8=U|^g27k0QJimQHymqa!D)1bdrBFd
zZHlEWd1fHE9zZdq&;o2cShiZLI$xJdd>Ej&JMZbdEP3>4u<H6;O@FX~F*v&>==r_)
zLubQ@c;@#Ac{z3Fr`Ctl(mTKAX}>fH!rkMCVRFQ2N=i0Fh?g!2^2slSaGmXrK3N-X
zEBj3AjWXS|CaibHP6(h5V2>Uaw^T`g*$5eKANdtPl@q>s^yx);V0|s<QPiihiBZ4i
zDviw|BUX{P&#DARR7G7$#%DM-3Q>$zWIZD>?8!j7D*<Xx$^6!I#HdeM9XC7G5UFUb
z7_oS`X53d(@kGT`@>nli(J`eBpQr8@UrbxbZQK_?it>VbN<|qb!S|b}j^y;QTKdk+
zuRMuMFeUVzo~Ggx`>n(APf0QRTOt<Z$lK)w0se8m3@s!LxRSTVK-uFZs)=bod0KQE
z2Dpy{fprYAd$t+!>OE5wlgz;$TW^+7mn;tU#Cmx_{vD6|Nxck)Da;IhS+}Cow;CHf
zLp_;4G2n#E8M*G;@)k_9(goaHoqTeSAq6{7E~hQ{!y4~!HJcFIKDSd!L25gtTes3@
zyjy$&b=P0AaeFU{o1E)Wp-o*v$Jn?2{^=`Ru|BXrbYf6+=!jU7-W$sipdg9$K4xp5
z5x2|n;~0Q95Bn)s5BD;-uUTT%T8r8CMx~)*-6Icg6AL<M0T2n+L?$dg?)%%Rae>?S
z=luui2aj}Xbr4-iXYsV6MBj+EGrooYM;%`Jv)_B<UT@Ma5l5R=Ki-H#13BUT3eYm9
z$+<G~yB>R=#=9i20`HNxb}93gnNu1q*>`p<Z=G$C9{p=}7Vg!wFtx(xt^@+?o4ZWO
zvWSyhbJ~NR*hPGLCwdYt)Rfo=s`q{|cjMFp>Qq&?dt5RmoD{4-F77B-#b=~aO?3~=
z>85&BeN$n5N!9oL+jBu%bc(cEuN84J!}Lf&!1EO6IyLgp8g{y;v8Qt?i}h$4RF;1S
zTOGydbZ6<$&}o0=QMlM^hB*UuU&>1UK@Z%sX@_y`9&M=s-Vl}Tjt*BlM1TC`zCQlS
zx5DHYACM|6&V!Uh)L@MfPig<oV@T!Wlj=GYqq0~rPM99w$cv$F3C&W&4?K}yY76!^
zwz$iw{K*V7?Z1w&uf?Tody?m6@pGPk&*X@F=!#o9?U(H%;XBw}i+)WyioQk1<s`y!
z{BZV_%}*%%_pql+9ibeHk$5F|_*U<NmdGk?LzX8ePi`B=KlVLhSCQ5w!8l9kU(`Qm
zAh)c?=~qhc-1$hlp9-D}L0%T5-rajSSc8E~nG<~oNgzL?ydYlkdDX+!y^ToD*r_DO
z!#!7?lEn#`HNWyt%uV-LiDE3Wh~7|C`5vH|l{QqP0&+7Zk*#24=J=Gs!n``FwB=Wd
zWc_w?{=jB3QNFCGdm-jh+G=#TDdqBUT)iC4h#hA;SngiMSL?;|d<b!gaA3miC&jsI
zYUhf!a?D3CM4bNOoBRZy*xbQd^QP3NuN<;igm(_xcJLVSblq6vbKWak*Kw9t7i+vf
zv+4S_lC;I6%nbkWj*6R!bW^DUb_x!SB7I>}V&(arSq4N7IG#ZUH^61YOKdI1^5j)9
zbJUm`7hz1+3#eqTmu~rjX@MvXBm<i-Cqb=Am<Z~Xf{O~aCEfQ5c+AB!K{#L3GLc{R
zd$x1BlKV+AsOGT+)FDX*WppPHiy^%zc^4WtNb#vqz_W?V1hYfpBlJBJS;H^(C_Fm%
z!Mj0sMp(6%&UQ5_gcslB-ce?4m|L7qr1EM7ebK&mgrLv`kwo3=2bDgM`SiwPP;E5&
zapQNZqd9XRrh;NvX4sEjs07mMa7i5hRan}xWXetayzAi|e2%DmwQD}8ZQ>abH&hML
zzX9grXMS^_J~gdkNnHoFgcO8Slv%}!g2X_UJv{45tH#i`#X@p3oM&Q^zL4jrUdIbs
z0{O`>rYK}T@8ZWFM#*Eu&`zfKCm?Lo-lI&>T2Y2g<VOWIl5358=>$`f`hnIW{*nq@
zZ;pN)L}m!}GEGmy1!3LB`uDgo+<DH{#}ZBx-E|w{B1SrnM9-ws>vf~VPL%Lt;bCzr
z41IwK9ca-DaL5X|VV}cAD_1^A;z!B5L{KW2RnZ`;mLv^|{e)uG0l0ctdu_stnt2T>
zt8UeGJI#%X|ElY#<6DacVX+_F@9gM}<r6&&YOV9!&7h_FOV>a0M`?))>8!_|DueL@
zz43U40kfI1gd7AC<`ODe32Li(flHszIFVQbF?fzGo+<&MYM5@hycwC|7^4iYx!Z5n
zGp*i}BdP4N_}u4SI>ZW^5dCyDfa+&^qd4+?UGw@xnCH~VniW|jY(oRNCsTa+QgGdL
zJL<bx5>{Wl!ujecKqBN@(Bp1C7|kq`QuTVQHt0NC(4d#WCPT}aIQ_2TPdkx8t{$I)
zPZf)x@l6gVxs)WYZzFL2D_tz6)GmOP3f&E##a$a_BVVel&K(TC@?O?cN*Ex1{;1#p
zsEf%bHAuRuVehFfyYA#(Te3^me15bxVV#sqmAaqrI*?yuHmZM{y9revSwAW_T&=nn
z9k~#|GtZ7`<I>z#7nOg^-izD2PUU{kIEY6;j=lB1dyryPjgsQ@05!;e4z21m!?)VU
zRM~%gIaejcD<d}{X)U29u_&(y9TZxhsday1`^(?u`YlzE25rJ=Cq2^8AMc31ddQa-
zeXH`Un(ppsduJMWPSjl5kbx+5Y4Lut``94glws;sw){1?7LqJ_RFVkGOxc}!EBc`S
zD|bMk-sn!5)PRX3rLeV&WZA+<h_piK6b7vaq=OVs77jN?3@=KA^gWU^>#UGB8<Y%C
zBI;g41_{FIoNss~>963=Hf#69l(}2(!i}`}ByqPX2e@+Ufdb~+G#uP&iQo(gJ*r+y
zdM=pr0Lfe3&4!=@&l1qP4Ik}5K3p1e6_Y9Os=RX^2Fp<eR$vSyZMh5`#b)<BOt;j6
zHCa|~uKnSg`ZR7&*)%=_(WWs87Gta1l5wmn6W^H9?OTun17Te=uqvMZ<=J1H{f{R*
z#w0eXF9SW&lS+Bb&EZ>K(9+WR9P{PtAz4xf3YCgUpo2D-!bq(Xx<stO2Qf=BN7JxI
z2TF&hs4km5ta}i*LX{xU+BnfyNkXGS`=)$Zt5G(A5$6of&r7FEkb+lUuw2i-l|Vj*
zw>W1VHhG>0o$-3Bd={)%{A2K*^X&#AA28l29)0hb$|C&9>@$qp1}T_M3;|FISQamw
z(cY5xjZzka&BG?b<<|)?h9m~JyBch;mUG?<mAy#7Rk#(>;Hv&mJ4$9E6s%3P6(WII
zZNmF(SAx!Gvdlf%T0<PVMoj*IRWfoo*(-94h%UhLjA-eurrc)+*p6O;GV(4%#0K~3
zl%4c!j>o*YsX!I-AT#rbdWCUdWE7)eBXr}E;*)hSbR6i&<ofzux2UMLde6Wv^%v{X
zuu0iZ&@e~Z(<Fx^-7zLV{e(T*TZ_qc4A_jTBA0T7T45qF7YP^8!x&Nsj4PJ56oQ_7
zJO}g*W-PTJio(-S7E9-YDM+upeM=X4dm29meZrmE2Si_?rL=aJzwbt#BMi}X7VgwU
z=PpY{t5^|}U<^sveNf_cz(c>Wyyh8*IE?uP+vWnR#Em$EKR`)Z%LOChLa{)7jAwX7
zc)&;752x(mcji2|l+5~8=0~3BeeeZGBw57yK?0#ttvh-W*tdiCtqjl+dPc=6dO(`n
z5MS@J<95BpcpUY1*yf$Dbap%`5pM61QEg75pO$9*)G8IJ9*hn84b`_gd3_|q8KL5v
zWHM!zmn0(Ylk|=Ar#z!!K4YGv6914bbdE{d1axbz=hFN5PIi&9i(Xk+2FWsf2e*8<
zij5SWztigE!1&qzWaH$I;y?u(pbNh~GIgLd8qYfSr$U!Qw4_cL)=!yl4w#{-sJoFY
zA-lw|TUiQ<KazX1h8|63Z7R|&9g-0B`d3vOinC@;8S2WB*JhN^#^W8H_gLSR^!nr)
zDsGDAB>OBYcIH&o%W7pI0!r6}KX%~p-F_LpTF-(LJ2cbBmTgOJjQ=Sbnl#}nAw(PF
zEUSREc>X+qzedmyVpDOJ#k9?=N=43Ks8t>;u8B<|hc6z>>!*D+bluRBeEy;uIi^p=
z+9J)4LH>|7-F%LF$eWT&o?*=I`wKh)$w{Lrw${u9Mg4HM3YO9zge-)%46mQ?s`Y+c
zQ3xz@FPTkr(BeL!9D~Wpz^L&|+tsij?tJzV?2$1ObDA0*!GkVc5z%h%OyPQLey2o+
z_bHQe?!%=fhb01eM_%8^%5z!ptlcTQrJ}ZKW8hQ6_62W*^ZJXxQF#qr?{`;e=HfNH
zi#bE8+sYU%*=nAkQputgW_Gy8S5dO}Zl<_jn#w_z$e$!~=c&1`L{y7r$xhxsHMQGR
zV1sdgikD?Gc=RSW>rw`}xudt`;j@+KIonO$XE`sA#b5Pyrt^yX>tZh**8nnC2$7A8
z@3g=94Jy1yQ**bGZ9b=}Yal9a4qnQXVbz|sGmXpKi7vG#Fi1ap8ecNa=#>`+e`bJ%
zRqWoSZKt^sv&n;7nc5HEdM^4U>Fm7+UEV#D2HBQ~S>(HY!hT!kv+%pCFAdKHQ1sX;
zOn!DCIZ1lSYiX94;^f#}=OYI(U26-Y?>suxXDOabdadxVS4^AuYmK}t?6o62MDoLV
zY8{F+ibl#yxaq2o5=;S$B$HwVMxufggKTD!%m*=!wZDr^XoM3i;SE4om^2$8X5u%}
zL5WOv3n=goa&k@7P-^yjb76J~<D(v01XFg5KBc$9FGD1{B-dW!G+->hV`h0mg*}78
z0ESVSeAd}s&ZT8;I>ZRX_+CM;Pg)dUD_P``q@66G<(qOAW=Gt5+a}FW)4_=AG$R0&
z1OiioQ=x6n@&jQ{LYpp+=R8WyX!BR4S=mTMS?zA)%F6FB;3|HEf7N>R7PxTC`zpu%
zR#tHp`i=JkFpX&nU)xoM2FvmE8aNrE6q(Njb9^i%U=!3M@r=oK-hhyt6w4rT;k&bi
z5(+k;XQ|fdmc`E|(5v;XQ&=o(95U_^r{Ax0?ExK#_^I4Ce1ye3a*5lM1vXe!KeN|z
z^YE*qi#&)s5DwZlvI^8O5nk&$QfXu6l$0BA&<Z23x=UaS_F=_0*(UL_TA>%`x7%za
zJP_L+lMyUcxV;R?%^oLl)Z&LT5I7l@ec;LFM1I_*)y>xd+|_yE2&S^uQKoFBb=Jy{
z_3P3ZN*1=cWW*V*LHdF)jub1Ql&z+Cp{~dCcO-DOdQ5OIR2g5<6F0DC<8Td1CQT*#
z^RA7YM%i;<(VoC3dNy&9Ns)9(ZB!>z{lq8hw(%F7aCR)VqK_4HtyVBW;o8=l&ujCa
zByej*A&;AcxM>f9i3b-4C7c*#?0@7N3}#h(w-=tj?}`?h)jJW$`6HvJCxdAiI;}J+
zucMo%OGbk0GC=muS<3+Q_1BB?P_Lz)bFE1)%!Z8y!c<JUq}z}b*iYLIp}qHhY-tgk
zVPKR#MoQ;H!yB!yGb74IT4}JT@^aI#+!=TRoH&^kwYxIsv=BrY5&QG00-1`NA&Hl)
zsL=Pav)s=qmq!QksdI^f<}l|2zUmK}<fbgkM(K7!3a*Vl@QGwHaQHWEq{?t3jn=y(
zNL%J8^DOQ2EGsVL3U#Hk=erUiysL7Kyhsf~gj$}+z0?P<-kPt`MvJB0M=_Q(Em`yr
zh2W}Am|Z-Ozli-jRQ1X7@`)K$?LnY6ei85Yikqj^dFJBo1~FQ2KHq4LW__=Ly|=Es
zzG)?dWA~Ev5w^K%;%QV`&;5thVSP|}Zc&Jt6VNKLns1lKw_<WQi;VpYk9lqtd?Cvh
zx97>x$-stS<C($3rs0Ql(|&!@l<e9Ip{wN6DoHN)e52OiR+bwRE@NV(sS~K0T!5>l
zY}kqNjGKcToTt~Wq>}e*Uc&_0KRl*PZboc=OhK_4#viLBmnJhspW#=+W$yP53>h4@
zbNhgFIPLjt*79bXM#(=(7%p}Y1I0JJ{AZNREK{G)Y#B!{rE*LCKSPfAE`FR#Dqulv
zUe-qqUa(;68eGftM3{1>H2G)JMZVe|pGe5sNnGLF`KG@3<H1^?0<&K%gtPuLuCibP
z>7GiuU9L#ntAnPWF$i7>cL!=9mLt8LoieT8E{<4Uv^)}r^Ib@rjao(?rGMfb&ve)l
z2qMl#?c=Dv<_$|zSWoo{XxxA~E*_HL=NOXK6S3VMPzY2#b(7<YRKjHrZ;T|%4o(E%
z)#?#pqgmaRYj-tDH)TV>h}o{EePFhk<AD>2EN~*p9fUCWUQV>VDtV0J<GmauXG=Cm
zP&Ag({;;Amcvb~^x0hm6PKSS(HTU=>iCCO%lwHHZW?;n86O-`sa9q3}F-!Oj7)v6t
zcQ<w~qG-4m7DpNCUY_LMu}Roz7dJc=)s8mIeX2&=gU?Lw!idJg!o}fF(h5%#L%HV%
zU;7z2=Vg;(*BiJUpIS#Ep>vTlD?lZXRa42!13J+vB*cI3w4&_iy}>Nr)udWDzQTn4
z0BtGpuaB1pOTMJojHN*N**D68rCciC6_PJ;%#c^sw?aanCVCBARVFF!!?YBCroUc*
z%{g}3SqS(4^d{J39+Jp(1Buy%d#i5I)G<%4`Qq3T^;rsO$hHj83^-kS)y}K3v}#Ld
z-9}z1ymGt<*_t6D&G2I+PF{Y@Yj;0NvLy@>dt#SiV~wvqBm>#cSpi!j)*WbT9nQm6
zgqpFhdW55-wsn&n>BtQErZ}08lA@9ca=l9>dl`vIvZ`VIus06zk&+*M1eS_$PUN;r
zOqVV6Pz~<&{ee`lB)LVJ)l+bhwYoX=!4D%%Vz(Tz`_lLO+p{9c!jV!)-K79(r#S6?
zZ|3E<PLgbb#CT8i61k-qZ(wuh(tl(ye-L9;Vb!0gddCD0uw@N_Z46{c*mFZXS?NB;
zaA0X!KY2d*R;#wUQQta*&Iv^mbTyvnf(%Sqtt&3?Fzqud^1ier<0joY(C7{`sVM&Z
z_TZep=&eUkt^IeOjnZFje&2MjA15)^1)ZCoBGwW@-Sg5njMaGcD;9y4;YeSn0BWO`
z_c>n{NIGR`k2QFRWK2_@raKfwt3CCb<y?%kax7;~hTI)Yv7Xstg2<n`IzFeZ>Bb+S
zq+h+PQEMBAtMqBSBZM;?laXlu^q7Pe?&Tlf)SNT;m<Kku70bBIZ21wls?ah=a=R`k
zvfLYr#U;U6Oz#s6?(7S~xi~@JA{B05MUIq9?%>DYhqy5Qnk~cR+i)1tKi<peH!Ti6
z9GFL?4m=+HvR57?sT6J!q#J9Jl*xZ%*>hHcQ11Tf*+>4Rm@e5l{-W|{I!tg}`DNX^
z4_{nX7EItc#&qecKJ47ukm$A;i~i^w!IgNyc7-gK#e)h7R>kNP-<b0?)=C-knwGc6
zR(pYk#m7dGD-z$JYD^JNs8FOJOuo*W)fEdpbkcdq0#)uf3<W~7CP)+K4OaJi++n*-
zoZ`97eD^>-!p}vs$X&u0#=uOk$LX~#F?is=4JNNRcrx98J=kq%exiNql;PEZx`otA
z(`<J2`T*l2SXUB($)6Dq;ZY<hpfq6kA3V(OF#3$XPS@V-vs0YtJHzGflk%yMqRYG%
zt%I^Lr!v`O-&h|~FDO|T1=-6KQ=oRaTb^kM)g&$_v_fLqUc!We{U;q{XHm1&ygm-s
zkNW}~`9aaW*U>%D4I8EF4rekfIuz^948@VDa(_T-ePYqO_+GJsX4;Rc!61*HGW8iW
zlzTzlLO+Sq;-IGu8a<CzxOX{Qk{{$>Fr1L~6XS5v?1aznP-!|O`y`|1UtEc`UOqh}
zJ8{~^sPJv<Z|)U7536q~+)kgdJO6iOOh`Ckh^yu%pBw$+foT&-VHH2Oni>O0Y$D`6
zq~Rojj>641{O1vANF&qWlrrsEmG`mkaJP)AJ^vXF7Ynj@;)(gNX(AgZ>X+$>JovkT
z3AYZ#WD~RG;bRS@h3e7lH|#3F>H)DwU_BFwIy(aGDGx>TywvMxmv70x@x2o>opY1n
zN33GK2nUL$tJ=E5aM$)X_LzdAU^C3O^;>5D%z+LoclBDXEUmXwhqS8vpqDL^yD5r%
zf5!mh@?KSl9DX!M`2tjShJ%w19~Fe}GSfOJxZe8knRo>$h(KR$|NXEX{LQWoyW|~x
z8%M&v^=SMDv0wf1xFU8xS_!npFVKc`l3@UiYJuP<XiQdBcIGHFWyuePqR-#X+y&TJ
zs)}~X_QMrQcR>SeZX_%0+|>;*-?fwJ@83bZzd)Yf^q*k?7g*Oln5uOd#zL~fRtM;+
zVx@HCg8ywp(xab6w-3X|FN84<o^@#pbU;+N7nskY24x>r>?5=(>Q4&yF_sFqf%CrX
z+(U+6+&ek$i^|Vr5?^z_-1aA7hMb)0M&9=(Pg*wBMZ8&WH8={adUdxaJ;4yVtg}&*
z!!5fk6(ovjNQeOvr3={qtbaz)5PxYg6X2FU<A{7qvORE>Drm*MmQt7U2^^2cXBAiq
zl?rmmk3Dg~s-nfTa6(DkBhMOcMM<V6CRv1lH~y^ODBLFztT#T(dYY(WT?>9=rO{5p
z%^|E^#0?RuEU)rqjCylsbawl-p0Ee^yjbPyC5x?b9%Ff|!FJ!vEnk!`;xpDjl;^zB
zuve4vIL9!j$=!E{V*`selIjawy5po8YH|v_YLOHhnb`~V=gdDS1U`TYmzarZ)Xy6I
zhA39Ok6Wojl&LWgjp&cdPX!RxcKZGOYY+bwo-1tlJ-3;tr@<=L{3M0D@4r?BNgr31
z-)pAGxi{mZYuF{L_tlY$$ktp0d}`rZoKyDW-97SvPKq(yZc4%x5^}cJ#txcQM+Snd
zs|&Fkr(ge7{rPV~$L~^53O+2YO+K@$t;U)jkiu@!Y*qcKeA-?2Y`uHBt?se>nDQ<%
zdWPrz;b8vPh^;0w!Ei&jKTgj7`9&<49LFrAz+3uX|Mwr7{uP8l=Kr;#aUSt+#ra=8
z3kIH<GI_W94^Q}CB!d6=XoU(WYr0w`Z{-EeKdM>(a~gjxsVG1vqk3WIQsot^jOn5`
z5+D0QW86klul7d-Y-XEW{?XG65D!5|zrNl9U_qb<HO)pQ0egLGE(PBQ{})mH^Q*15
z!GiVTzE2FRn|A=Vjb{%YCrUk)<Eru-P-k?<iHM%Y$kcB|o{wiE7S6kPy0`qEt2~z!
z!1mP}RM))f!7uVZ*|zC=hG7HX<pi2PF#a*&|9TW|MnGKYi7x3g%Zc*WRC4D(O9UN1
zWC;SfB!mMcEpELyqWAl-x8f!MP2<mfvJ|~omT9Ts+X!H+QMLEmBI@l1B7inDv5A+a
zcj>%82(3$rx=utz?*FxDyD$|Bhe`Y^dgRV`OFTa|PoZ`?F8Jwu=e=xNS2B3cC!9Xz
z(?0}$^&C^C3?I#tjs((r6zy9*Z<o+RsqZVx^Pew~e)4)4-7e5^%90Co$J<Q5vhkv{
z+5OR7bieKKB05!fv=X!J(RvC)wPF}f>(OQZUkVynnH4}R_*a8|iAp#CK@>K8zSTTs
zx$fU16DtaXGv@*jznaglJr+dVnx@?;CyxMrTeBy?P7EI_QC-vdJqv&)#zH_X%qDaA
zS!rEvu-X^6dNmS27L?=54l2D2zPVnn&#-*De%N^P&$8(MO_6x1{^1^Ny<T|X+Mp@W
zE)oF1I~K1_7IZ&728IBDM&h|ZuiGNfQU@9l=zJI5o+5OQy4Ny@>LJ@P6Ud`;Il*(T
z=JgA<1w~wXh}r9h?ba&QCx1rZ|L4d_N`)Fgi9+bbYK}*bfhHfhW$@Jz!t3O8IWe-8
z`PlK}b8`O9%Y8xT&xMC*t<9Bl2?S2PDue9)M*Chr?Ro-hy>Bmx>ZKP`=byC!|5L|K
zN5B=Ny_UyrTpv6EG|z7mP@}W$&dEZm=O<a;a#z{?Jw?)v{b8Qq9NQg0pYEV)<p6M~
zi<+QKWsX~e`ZN$vR2s?P8y~*{1aYq?-lw5u+BP=aXX?Iqlgl`VAwTK)Iq<F{7=uok
zfp%mfn}xs=!s?d|8jn-w{gI>{yFGUnLvR_?PP9P)XK5#wV(<+>B3k8w_?{oGyIj9B
zFB@!A8kT{ik39xn=x0EaU;V18j))hVg}!W=b@kGx(YQj*UVRwXjU<*kWm3Ru7zM<|
z@Lco(h3nU<baTo+jxCM=sRsaxmyp)uo!8r&wS0emdNRh~ORs6y>Y6XhwXLhqXTJVX
z5dU9qK9&NL+ep7`6No}<0R+DjczcpseI{_jUP;e!`{n+~5P*#=xZJS1(SU<Ew&q^z
zmQHKxNGzi8U)Ovf4+3ar_Lrx7>w7LYzpCr<e^Zu*B4{6G<?8}iB{CAXXh_VEN9R>W
zg_(OXozJ@PWa5qTcXp>dAE08N<^j4RcBEy-F~@$XH#*q0WuJ_L8<N7?icTqO(Jz~Y
z+uJsD<Itv^TQWCAg9hcdIB#Oh3xcFbBu!lBrZY}8a?<7mzomJ%Ek@9$7}c1S&55{8
z3(nXj%94h!y(rk2sLTUMSF9w_nbMH0q<vlYI@4bgRns{YPc*)?`R>T`En9S6pKr{|
z{q@?6rNJz!%G;eM^*2aj=l+%C2AG6s1TuQUeTCFNNf7|z8GdPTpF;!DWWE>)UJBNg
zBF*RG9{DNvjy9%rM`aE^q;-WwrJ4y`F(7w3Z>~HZW*VT`0rIR17p&Sbo+cU*6b*5~
z9%=_rkL>ueqgzVcE%0!d#+Dr*SY5<Ci{!70k`xZ@=V@j*%*MR+oJV#nz&~;a4TOqt
z;_=b?IH&&LBg}gI+J)eoK;cI}KidPG0PRFv7fB>~5FRsZ|D{x2vL_=@!IJx>xGzys
zF`ym~8$Jk+q`C6gTgfdB94B?qP5s$s^?xtY5M}ZZSMAN;8_7_7i7F=mH8d3tINfa`
z=3zO8oIiI1u(@1*6UEB02vK9|yt+`v$_8FiD||_92x-ptd8)3Q`qxKZGjbQ3sZbfw
z`#3Hw3qcp@ER9BHE@kku%@lwG$8oP*N1I6Y@Uy{)J_hWWN*l1YQ*j7s(36Vn<(Mp7
zMK~(pX#C8NeC70gr2czURTVie+mB~k1EEb~tv^EVEG-?9+up4rlkETQzmo^;w0i4y
zyg9QjsZ8+15c``!RomRdwk_RZSd<(j=<-uakgWf))(h<%qhF`j|GnS-lbsf#jQ9Jg
z6~8DV75CxT=E}Vc%!Rm&0iGIqMYF7rT7(IJ`@JmLY99lTBQ(Hq(e{mhPH1Rj5%HX;
znBZQ061wX)@drLT`U$t>Q`V-VxVxj!=G<36`lCbg8v~34+iG6Cj$Z1KT3WT>{ZYm=
z_dID|{$?EqGJh22+UsQiJ*2CLP`T%oYCu)g5OBH}xx6G2C#6Y%<j<4x(U{X!6T=@=
ziKk(o^XLq$+(Jt4(*Bi`!3zT+%x#HTqTji3Lz7uQ|9l-2)egJ?SwCPyPXTDI0@=b2
zF$nh<vOHa&)~z4~Fi~|7FO>?YtF%_QKuVDl(MLXG{aAF<xQ=EfIm<EAx!l#^_#?^S
z2twM7If{lmwqO<#R)|={nF^BFXAWSOrt_~6W7&0ZziNnX$0OrFrzO?X1595XJpvWh
z@<RMD^k|t*o|FrJ_nX|ysI%Vhzrht1Z&WP$>i)y+`8UGuUoVIqgaGdGE{<b?Bq=~#
z2hEgwEgydyep)wNkH==j))eE%&dDIw*~?&0)hE<2usKy{wGU8+*K~I3n2!4s=yP5I
zxSq0E#l_f{-;g7Gc#EHAS{MK?!9|E<k_Wz-Joo9Z)E|R&*9ohNy<W8&uSbH5AesG@
zM0;o0wExA{>(uRUZJ8`Djm+Jfn=}N5-(B7~6$Cz}!HhHW42qVGxtEkGpqZgBB*p{3
z<cVsJ73h#K7GZ5+FPIK{^8m1nl`Rawbu4(jP)A#V@b2LMxo!Ph0D2J2uL;v7VC-LS
z-ui63P`>*Xa476=cHS%}aE%!%r90r6I#6#Yo$^6h(PHwQ^S?2mGJy9)&VU4Ffp7Pl
zTU$0L3`+|JraWJNgKxVt%u}ajWa^M^A?-T(GCMb5f2dV#_XW6K+`8+T1qLCRf7y?H
z+`m$&v{;2^*5kpNNEW`To*2LTfE`Z(9IqA&rd6j!N~?h6LY<KoYy%;Atv8>ARx_vD
zjwa1DS87Ut%TNx7!e1jlbcRlsWT_ske$=!{s+%4m?hC(^2KHO;UobNNaFwXP+{u4u
zuITGki}1NSNJGtj%1h4Ch&U%rR2lNrWQ|CHHBck`P#%QL?LO<9>nlz2#HayIRcl_5
z5TF%?4s*>9jAQgRW}6>WF3|=9HzS2F<vpIQL`d()I_d3}m^M1L`b?TVs6kg%*T%e(
zfQUz;7XEKL_@~EOX#0C2#$#Nvz(N4vydDL-fR8eS$C<0=>BGHjP;)e870mZ0z_O!s
z_~FK6jdv;eaI3cXH<6kS+7neB*&c5|88-mHCz1M4`x*PO-06=nl#GvQGgEpf^K7xc
zq;K0+3r{l<^YHH-+@h>!3h?nb0U1Aj$cQ+WJ(>6Rp8NP*l;`ZE^Cr(+#~-Bo8!PR-
zVew=!9JpnNH`TlxFsNnRH0Jf0VZ(ni3jch9f5IV|F%1)@{f}py$fCHSiU5Pka|RfB
z9+~J(e4vE~^)VlSN8|NQVBpdvY2}FVwVlay-fT90N&$S?6}4)$CVgJp)7`}(z$q$V
z$ixS>K!D0SYx7C=#G(N(xaqSA2MPG^0rn^-WG?Drx!G&EA5U^iFvdQPMwA;kHKcoQ
z@V0F>;C;2%C+z^N&QLclW3D;HN#j<_90wVvj@F6(hXN8-6*l0=GO`sZXqJ%7_eIF2
z2W5a|8%ac;0-Qt2+Ek08o(nT{0>2N5PFdXh^%)R-;1|Fxo0ocd$27}|scb^!Z;U)7
zgj;!2(Uzpvv@_@`ZQf;}BQSxAyCs;L<v1TWW1y~JG|6P!!8aFH^=-AD?L(=Yoc=rQ
zzFfB4q^5h#%CQvdlEvS{2xOMy?rM<dS+@f=xy+s+Mhc_=iSqKdWd3>#KtK^&D#-*0
zFLa|ha){k~<7@4GX+nQ@;H<FSwkh{SxBd>|(F!&C#~p!pNt>NF*TJ*I>Cql<*I$Ti
zQev~M!ACoP`!yT<?T256_68UW6khzseTGcL4FDU6bFDMZ6;byL;55q9+bze7>!?`D
z3iolFgG=9u`Dg#txsn3=AAUaY>Nrc3FBP69(+X%R6uq1S{H7G(-PKfFKVzEpTgw|C
zegknF6o(mK9B<_?-2O_lVf>;USkNwBiE27PsCA7&qt}FsGQ|~mnzB(*AQ4!~YUpsM
za%<@+Z;p_i%Iu}C{FMkh-js=5yuk(5__&TVpXATNurE86g_?bqEh`r>RoF^tp-wW6
zMaD2!E8tMmL=Qb>k{CKLF4xJEcaJpN1NL-2avW{L#pw>%#abJH70blmsQDmw(D9ch
zlb*yRP9;{twA5|&L$(dzWh=57o4AyvQw+8?29sDQy1r!!<*5+%(93d3ntd2TC$Npc
zvU;-DM;V6vjvc;<LtYJHA%73Rj0$Rt!PD5^Bc>_B7jQ`O7T*)g6W|jQRYB<BPa*(s
zS*CW$_^%8lRfr$lI+^Wwm4wIRuXh(~tSSN^m{6r<{D%3jw?CD_CqU$aJTZMr1oN*y
zWrzcup$%W~%!B`OP5={D1{f}8VEDn`g%h~LfOrD;IQhe<ztSK;_D}|3xIViT(fI#5
zQ2a4Rg>*$A)|bAHFW&o?XWuRmBoP>Hseb?YAAk4X-&i=zWNlkF{3x8(_K$+D|F+OE
zs=#o2CVtv~CjkfWsd>~tPVMi2YkUoHu>N*(2pmkn+TH=2`MS9KhHnx(!>8Mp6Qp~J
zA{l5LM{`YWZnua2jg%^gK)+)OMF7v~j(EK>y;P6;-@ZfYkWS@#<8%zKKbuDWhv5cX
zA_X4>;5mJMvcDet8?mC9t!)K}!NGqw-dBv@VXC6cd6mDBSW+bLeq<Cb|86h@E})@?
zk(Nl>zf%VWK>7A>?Wz9FV0!dG%*mMk>CoTG$q$H%joj{K{JX*azmWdl32FBx?1`ad
TYS@|kE#OC8SqEMVvkv_~H}CVe

literal 0
HcmV?d00001

diff --git a/_static/raft_idea.PNG b/_static/raft_idea.PNG
new file mode 100644
index 0000000000000000000000000000000000000000..b3fe93c864ef87e848557fbba7c0f42943a39b6d
GIT binary patch
literal 121926
zcmeFZcTm&a_wNgWfJ(C>9R$HbQF;%c0!meybd&%hO=+Pep{Up>QVm_Ibfwn>EOZh`
zfJhBW2@oJc2oMs;jgQaodw%Dff6u)$XXY})3_E<n-dVe?^?t3zD+@D2uHyp7nV6Wk
zZX4aa$Hc@g&cwtb%fZHYMUBk$=NHU@_YC!!s*yq~j0UTxo~a%a)91t!v<F8S?PE`j
z>;jpX&RzWT&3yZw<ZmV>dhzX>dRC#%YX>NAp`OD%`Yc`YjZf{#$+tftFX6y<`M1`d
zKj-_bcH#|7<FRj2yx=ox1~1Ob%G7_-I}v@3E&6Ed58ekIuhiJya5#N<ezNtvc&~%~
zd9MdaUDr;0jXGE3Fotn*!eFeOv}Ys;JtXEKb^a!g)(V0xwxVL6&KSvmeOXAvB>mUV
zOiYh&@t*omOSg(c|Hyw@AJ6I;{ingyaDn64e_HuAtm4f7X^p-Xll)I3%Kqdt*8j9D
zuKu6stW@Ng7UZcRe1Bh%vWk;mR6AVL-~WAYW0hnzwKla^H?X=;=z~fe_|X)*?hUvm
zNz&XzFso(AnK~srDVb@L7!du1`gcf<_N?M<^+0OhOC>n`f?J^GDkA(QZ|(YmQM!Lm
zm!WstcJzdfI*pXe3D&-Zqqpk2F3V`xzkAiKirx!xv_XX>y%3XZ%{JiOA|(tuTl(Ia
zFBv3mPwTtL6#Z)+<i}Hby~41D7$f(_HhG<H+*5A0>O@=M7a?lBPmgGZ0k1;99?e5j
zmVKia-R42_;y+69b2T#Xk01Dw#p0Fw4SK5bas@&X7i?|=_%6z~e`$c;n9$YsJ#2YB
z2#I*^<IY9`0pTM9xS#cFNayR=@oGL)VVzyZu7%3oJWCFJ6_**~nxAeZqu&1+(v_Gb
zCp0Vxl(&BUbZxRWj9#6?=7viAh3Y?!0>yjt0ILR3?JlX;3B<37o0V+h7+6vvJ#1*L
zzU~t2SQ%+mo#)i)XE!q}bB?J?oroCRf0HEj*Pd#HcVv4e_eWDbu#IwxoB10v&pWcA
zwpq$jAztafB{=wpu5^eNcjXW6XsJtZ>4)HJBqqS5dmjG9bpvpL+?1J4{&#UFBvx=M
z>zd@IO$5a*Isq`Q*poW*YY>X=r>Hh`gq@)uzL!~Fq_s`j7kh>tIRGuiCVg-}O;(Q=
z1AZu3L)fR5<rxBhp-9?O_q~OE`?gY^wp2{2#4P<gME}R3(nV(_kKcDr5?i=<Zy+kj
z=<sphVV#;wneSFjl)H;L+AedK9x7Qlf;$u=5r08<YmW$9XTTK|B@qkx+7`1g9{pOG
z#TCuddc6VQ<T_-kqQ}PHeKIF>fn!3RRowhKM_KmA1GY)(b5>D%!Nj7IV4|s=h7n70
zJfQJN`rJj1W4zC9^2H=INc_8vqDXJWfC%1aS~Frr!jYE-9b_kHZlBz6GcBT%^aHZQ
zgv?w0Q_omA^b1bt^-liRp6U$SS&U7R7IhDFMi4KB)!@$Zw<yb<5M+*xla^JVeT1N{
zP=V(h=gX%L)?t;xtTKHopAOUx<#1v<w=KDGS}A=Fw(gfSHvEe=*I$}wIlluRU_R&>
zk+s`ua+e}e#uIV^-wr-WSD-}uzLB^6r|@(uo852#)PdycbPghzGU7j<?-UTp!zb+J
zbClW9p)CK_qeG44Md){&%bCc%`U|HwL1yi*Q9lm~-ns(<HQMxglRAetr!o|3s#lC8
zbmD5*7pdNE3hD^Co)0yB5vtOFZm7d~pz4bDB}GAX!0+3O^UFD&w}tZ)@wtQWAMf0-
z)N6HJ+)jyuvx;~nB`Q5_%w>{JuMm6_@Z<1m=~6|+I+$HQW;iE^-!43~rtmE%SJS}n
zjndH4XR1aiOJqsYMAbD7gRyvkIVf#?Y4-J?eX#5NRMx7!-}BO(<b#AFk(up<zG1N<
zNU4+O{<mNnV-JoW!ZDfsOQkG!n00i+htTf&NwycjZM<Qv)syLoox+ygdnJ(92+z6P
z%AHf6*NfkN%VgiwJX#fad-Vy9tZXINK$pTGTUXy^w~xW%f+vQd_`ba7tpl&p*$9qV
zg$e6rqVLq%at!1}`tE1<2qG$j(vQu!{CI?I^HP80^0f>f=JgJO2qg64C|>AWc@^dK
zU^{aHeJ@k%@^WBmb0ay!`|sM$wsleB2>UUeJmJ043C62jDl`w7Hg<!|raX7(i}IOO
zxwOm2eI$K+&F1_wE$0k={L_l6r{*?I$;_Il{J9Zt*cKRtwRK-2(v)VPD#_dlct2al
zPAUu~ibU!p6}v>PRHz2Wmy9d2B)DAS@M1H1gF6P}wwYosm9T3)DeCfIcr}|%+&lp|
z7GXLd6x%O~or2odKdVwNAua(`71h9dSQdWe==f~k`%Z=JgAaLLiVF8_t(&Od!xp>s
z&FpUC7Vp&H6YC;2J;m*755Bo@xAjAix{!Wu$vrr=+~lyy=^VH9qy`ikYTpPjYztmh
zdy&3|oY~HfZ4=pCYIwQ7LHWGR{3^TZX~f+fQo?6X>j4kroje>#q;6IKTS6igKWq|}
zwT2sN95zO(6I<RIS&koCTEy|Zy%{!COhy%$-0v#ikPXxkUaVb_6^-{cCy_iq(4rjd
zYuXyK>xnI*?=Dv))`T49I7wv^6c$atw$~dWa4o?L43n)`jQYP)v7XJ-+&2}*?iBhQ
zzEk(~_MWym&^+?)?*v)E9eKB0FbX8%j_by1m!a1Rb45zJ5|0hgLydp1^$IEMdUT4m
zoe>-6g=jB&s&EH)wQ8QlE+Uz;oGzy50RL^I<IlTQr-578WAp1@%VJSCW8-lMZ~U4)
zJ;sv266^6>8D#Mr;|+VuE+q)iPHT|lN*$*P+wE)f&tX^~>nPum>)da9ztLLPphl?A
zWQevYBKoP_0P@@Mky_N#n<cr#-UpaNuHZ=eZT1YqZPy5zzi>Nqh6OW)@a?!h7a+}v
zc+osIeqs&=<FdW*Xv`u1Yrt#0-kBCP^OeZ9cj^2&{)s%>J`IMo_P@?^=i%NWP^uBv
zOo^g+jT3<+1kW#5RQ_A-yA(Cpz2epAuT5_hU<-P|a!xFhHnW-FN~GNw%K%ZK%afq}
zq3{IzW}Nz<%#Q-XY#9X`>d02zzwk!;`mLK-H{vcZcqd`dvo5a5q$mvh%a%M7A5oh$
z>btLr>oCFGEN1ux>rB{Q;`rMElwqxQ2Ah<jf|tl=%*LzP-PhAOGj|3;X$~tar)=Fr
zN~uTqj*5fuDH{64iPbkuv?+IYa9U}f_a<p-XSYPgEv)-Jj6?Dy$s74^4?XWT;#4sK
z1(=<dv%Ep9VW!m0C-xo)OjyL#$QMryiWQQJZn?>1UY`g9^DV;F6F1^m>c^)_$6LzV
zKYB(<;ZH7Z3fn;{C;c@=v)YxV6fLhnKkm3KS?_~d!Ojgawk?!83wyII7C&a+7dDJz
z;jlaPPONBMaJj5m$_EEbZ6)Mu(T6MSZ6}VpTi=eYfG^%ZfnNA}C%nTN^1g{GyKgU2
zInaKro4fyjPdz80+Iwn)UW5%XPd7y@3i3xd{}jLUcOg2Scah^LYl6xz0c8Ez+FDuo
zGLMxsd+c9otJ~CJHoopbiAnMjlLn}r_wwr-1Z{#oRz+BRy6lZ;WC!FZvYF}hKdv=L
zn$*nz_@)NnOX@}$=VUX_zAv2@($<3oolBRv?{if1@iBe5wc8|1#hF=R-{c#wmOL5%
zQqTNz@e%v!Cw<}ReS1wb#b1@WhF$3ZBR};9v7$RYp}ei3Efc!81}Yy{Gv#rAG3=Q?
z|9d>ac2LI&=f5?cucYPv6r=*dmgYX^ho%QNUqjF0GtOEo+$h@$bD%@|CQ@b<-Yt<Y
zb!`1Av6&cmfE<h@55+x&^=kT8K;n<Qt%g0>Vc}DcHt^>S@R!SFUZy~-&rMNNPaN)B
zU&>l~Z=LJ7WhH4SZ>lRT{{%Rh7_G-&cFMKN-oVPfQriV27csf73A;?Y0Xp#2etGY4
z<730w*7@e@apkR4<EHxT1#Ha<g1Rp`wtsTPUx}@HUo>R@w|)9-b3$Q$<(0A6Brm9h
zhOOsC&!OePh^WhfzJUbC16_+^?-T#bOVpdC6KvwpBnoIV&S|^TQ7b7;c+Sta&qQf?
zKQ`itgCdKE1SOUtf)p#Ftp5r$^&DH|Jf7KUqBpm)9;9XDxT76Beh+4T2g!EjX+l5&
zL>gxOUX{=*_~cmHlKgJSt!@!`5WBmquwGgF+>q#|YZWL2r_QC{-7{+Js^OvtCjR<L
zo^~DBaLk^CG3CK-!T36BeE<nHhC9?M=BNI=!>J9~x^F^irAeDCEQ>L9ULw5_7GL@t
zq$)dK1io_6B(}tYG&)CT@n*Z4998vK?eC~Mp&O~8Ghwo_hJf<$u7iV>mQc^~bhP(G
zeLqAmca)ni*{%~)n^R_JLfs^5+pW7xg;xgfUq;(!lr5_v-cg|;<Kz|1d&bBT9+`}g
zLm3w$<%plXZeCFU;LGETJKF8TLk>1+{dz_f+}pVoJ9dl5Y^kE5zGTj)ILhC(J{Ld<
z(CdvZoh@D9$>^S>y0g1B{mM5*&iR*aH+wYxN<5+-o1|$z&rEr0cU2&`o^7o*HU=3u
zpn$u+lp72i5O|3?!M!iJA{dbH8g^?b=jj+qY$0hn2iMo0f$lo~sWnE#Yxuij8_Em^
z7}xal^k|%3r1+Gcdscts#aA96Ip3=J%VpD=Y}&&kkrVY=4DYBRvG^nM8lF^>k)!yo
za%)i@w-Scu&X!hm>(IkTO66-mv0TEjy%2AAzc56A>wYfFtB&vp6P4-zNB|99Ul?RV
z)*i8*2ntzhIBktQg1}`DSoPhREbFQDN>IOM<d1IBZ1zQ7&hj)a`|$QmK*LP<k|UKg
z9cLW5*1kx^6khSY5k9TWE`D`aq7wIeD3j~%Fxz{=lp8O4TQ$Z*%AP0cV82zdSs>gO
z9L)3AhH^Rbe9vS!I$L&(*%!VvCC%5i#3U^>-C!eCiro&e7><O#8d83bdo0qo(VbqY
z1HD2r7h$-JC;3g&rN<ol^nig0aHhB+&ip0Lq;M9w4?+}UggN1RY$)ZM|1`0<zDQ9?
zYXD%cI6tlEc!f2r&P<l0?EZk!YV-op-+W-JTkk#-_WSX*kReHvBau#ZW*Bb)xcgQ{
zpTl?NaK(xWf?6eignPP`>5qZ5a^L0dW|{U$5`vC$keZHW%_;*#27e@i?xY!m4mJk8
zm#D@*N7wb5MFvXLmxf`a4R336*M=7g552%@N$c^!cDc6e-$t9Kt=GR9(aaup1UpY{
zHyNWcxrZEzw_I)Rsz1d~+$2M8?hAJDsOQu#?714M;M!B4OPjkdHC^ZLfG9)orL$X@
z3)<teF-cMYHe{--9QL5S@vkYHA93g}UE<V#!TCNbS`C%&`bw56Jd|?_*6eXq(8-}j
zuXi{<O8%wIfNK!|+AvDY+HH1EKeGlLV&$B5>;ap_CfNs?r;E<kv%$wyu3b1ya?oa%
z7<${8kY9Cnj$mJ{)KT&?>naC-lAn}CCJg4OQ}f&GOG;v`dW0cdwacl)>0Mi*?G~;O
zuk8mC>oVX)mp<lIs26I*d>@0PTHfHl%Z)P`aJu?qc~S!<E+e{h+_VS1;(Jv?XSd<D
z711!&s4zvsE<$VE4!`6_@p>oX{ap;`3JE%#Epv4E`w1PD@TwekQ~AM7U&MgDwIC0%
z>iv9I8-Y-76g)Ewej!URBH&{N|Jx&?9qfIB>bYcCk^$p}b@UaDB_6mhb=+TU{EQF7
ztRbKBDvW3(uk&F7aGA<edtPQCq$8fU2Dc){HGj7=!%wm7$z3GboVwJIk)cF<*mm+g
z7HpPTOJ9PS%r>JB%{sUMNew<bn<Ak5t&wG73!}Ik(#Pzus`LO{g`(2w+Iuq=9dCuo
z2`MtN2M6nhwZ9)aso+F<Kl@1+7{d-<T<x?8xlEe#cDbB*rZ}F^es7F=G8&8jM;ZuT
z>r9%4?wO?HeEN2}dV+4x#i(=&d5?)nD*mg4{63o9CRTJzXqma?%79ULGuu1Nq87`n
zXcM3Ot+K|o{QR)Qhvr01<BSu+Su=s0S9U}sjWMcg!eL8b%P_qGdo~G{Ei7IZ_6k$<
z^ZX-auR}76-N7)$uX$tLC_Tk4#mPAIF`pQRmwIHHoOhOJkZgdYvFao=&ru>JF(9K}
zJE@qkdDsCrukx9_(_a|2>!zXD=*>UzI#V-H;T-C%5AE|9Othm%_}q_=@*MoUPZXJ#
zHo!SyLnM*DL&9et`<`sB#w7)!(wbRDcudmS0V7g_i9wI*rIZ@S_1X$QQ|J_V!?AQY
zz^;)Na_NfncV7d>mbh=H41U3ID1*<A>#&QPPVw!zj_g=bBU)q9O9OoWN<>da7={6A
zbSg6<Jm%Q%+E2JIWHD2^zC=H+yIDNh%61uN)}v?@exD7UjgXFVMPNUz>F$-Dg}qE{
zyQUTx;2L8-W9)aOfz#l{qP+HxDKm^H(Bo>LZ{_y#@#E@6%y?1Yp%d-H)6X$Unf#&f
zj(e-bQ%fChMk>EZcH-p<rxNEU^g59GGC57jx?Yhcw{=wY`odvzv`!3iw<kPNQ|8lM
z7v5U6<#tUzlmhZtcN@vxUPphx0UR!yH6Ad81a*~g^ZAvmoPwr*l;*Je{PunL_Y_1`
ztBylc=FZh!an3}BO<HM_ulsN7W<7u)u#lU5wX3k_k5<hL?X9@<GeM-KMtA-^tos#6
zQ~?2;|KAOR5G&HYucakB1t0mfd$smA?w4DC0MCU|MR)-&M8?TWKBK&;T2D{&QV6wJ
zD4Z>fTv>jRdabg{5<BPF%9%HLhv!+-D5ZyGm<0r!I4!A+DBN)v67;Ebzou{vy5deu
zp8NXSa-zW?(m^nBu=W|8d7j%teWYEt|0uUUo7@EUJFC)Y*xL8HPm#==iq~{5?Sd6M
z7&}U{>BrgSm5ctcE=X_!6*Z+-B<dL$gI9in<lPYg^>rF>>GjqG@q;eXskiD}oIjy6
zu;|WyYuL<|6}o)78i|lgvpoox&uX3J?oTn4UxZkxE+FC>tHprD3is~UEQjlGU#qC9
z@M}n~5eakr1^)bDu{Lfi>&g}BRor~*ly*wpV?aSk?Y<ot^`t#fLN<YJ5|adHmx}Ch
zOFKqvu@Ad|d0}9-IQg##Q=lHFXY`%(NJZ6kDUv_;g4!{T-1;I4#4uenvCm@ex;usK
zor9)1hH3sci(PocD5NH3hho&22S~ab?Bm8G3S*|V>)p|`m}nk9%-xA-Jrm5`!%nFE
zO)xs6MW9hx81b3x7*QTUeFQ-Sw@X)8suo{=V%hy`aXVaqqlw*s-2k{sNf646wL}!6
zK`VfDw>NCPE^2JO-u!Q6-XV6DOfnPR!(a@BM<VyGRe+7TM%Gw3_Q8S>n>e7d_uIOc
z-K>TC_ttdu{v~RS$oCZ+t*=;FQ+%JxRGm?vXOBP#FVMn(UUzAVI1eiF${Tyy5hisS
zyA4_zWLmxGf6>^Ad-@%9giO_2ib&Ag=%vL!wF@oAeScgGyV4ftk2g6`O6Q6RO`PI0
zG8*H1CiL#J;1#en_<F^vMT)G|BoY1y>o!8<#AgWl3;QBY5D8ZUKp{>$fcwElbq8kX
zw)kv>|9an-s;ld*czN^RvbJfpcB#^seJA34cvMlN7%)wtfD|5(&4@pTc5NmRxa8+~
zW4p5!>o>P&D8Us6*9L~G)a^8lrZPhW8;U0<R*>u#`UCFri&(}vVroZiI3T)!ca6pq
zCh@b=4jE3Gh1RQv&(g-0prY?X0@nJ{h3s+A<`305k3D-vgI^|erJWl`Z!pBn29_@j
zDa>?iu`Fi(vWD##?mgCBZa<-7RkKn<_U_jP<zRzOAGd$+C{Da0M_K)BG0`c|k44q}
zfxPFnD%}yWt)DI{n%i5A?*QY)UbFm5B1JO$xsL=PV&1Um_BE4%1Yg&i2eW!)v5C#r
z!JDyD;~-B}Vad?NW2dYH9+{=~xr7hv921)Lk+=1hRlgrTo~67u)uOZ>#;IRIxegUf
z{j84FH27MiXhuHwY4msByUK%HR=eq`UZT3<VtBuUjjp|ahul%Y+oWP|OwY{OqAkYd
z<yr_AE81^@j{1l9hs{q~;s?vVv@$g5LAsSi+;XI9ULUB$<YQHS&a%yCpWMjQp|S<$
z6W9`#n5#(sHS$sMwBznYZU&Jd-qR^#g;g=F4Hu~>;w`=AYJ{qT6!dxeKyJT(L34JH
zmJG0Q1Gm~CdeX*5Gj?-kYFPe?sPEYZW>I%#qNc`{t!^f*?$T+EWq+t)i1*vRFA0U6
zw(v%;C(RqXzyIBPYZ^^@y+#Bhl{`T5EAJaCu7u8{x3dBn6h6&>weym4r<7jO9vG*K
z%>L01yYvp)01WqTN{wq!*Re}%u~gd*tRM%9OR?j4?CofcBO9e}uSsInKjK{Wd`%3J
z^=^?>LC^@PtdV>dQ@z@h5+S(VAGsHeql8<)5h3O!xite1M${e_M=#xMp}8Y4cknrM
z99<e|2nRjVup1zTuvQ`;@NV5ig$_Co_yZq>-jouN>z~x)iv6<FW@J1V<OO$l7p5s6
zCuCa_HXy^e%ETN|z;J%yrZSe%@x73nX<tjdw38OKH$HV5ROuNV_-p)D@yF`f0&k9e
zI04@`JXT&ve%lfs0;kThMg$2fNTrxN_lL|jP#-FaTZVhww?a6OYvGOJ?#1I`z{*mE
z0;OFW;P@TtvXDqkFJxOfI+zeAY=^AB>!s6Vb7YvYF=`f7PuZAdYQ9mRSfW)e!uIhk
zw5V|NgO{J&UVaP-qL0;+57t$JcC583t`~6)cP;nq4wC&NBP^vfMVO9jiCB*}xoD;J
zIfR7yWbA&?u^jV`C`h=zKeX2WFWhPFearwoo$Bvc#n~f?wrz7)jz%%%v$MZZC?^P~
zKm3ERn3$Ta7|%3I+SgJ(X?uU(?6JY0%Fj5)KtiQ5F!iB(m4CNY#tFtN8DFWl_}|3%
z1A1v~3>x!NOft%{Sd?ibnA`aC-!2_h7{JK!i0nk}zIf)Vz?BRZNseQ4R^N%SG_vx4
z9#Z)8kWXGBkkn`Dbu+ptt2T%i3bFey@tcV$`v?Ok$rUXK16M+<rRW;3>Eu)YC8#sd
z+|fVpH|nX`_+BSk)7y)})VB8MN1gl+f{L=#{l9>A{eOzq{{PPYj}XfL)r`odCd}3t
z7WZ*IK?eE~pjp}|N+4NG&FAc$j(jTW+&^<78G(Eu42wsT1Yux7=(EQRg>8q{{1&YF
zqZ+_~4e*^S&j^2bGvg~?D<(+|#~?v@9QxyLU2Uq&5`dHtp~(8ieN6g)e@ISh?|1xL
zIiYW&`}IK511*k7@LleC^BJA~EyVl*{or20y!qWzdPds}kd=9IupIM~l(nwmHkTmN
z;oY6L!-D1d+o5h4t9Z^Gwqp(ig|j?pwxcqcHSP+t0GZxbj;!J%=B(o02G=TdA{6LO
znzB8)qTln#C$L=LAawFPh)}CVH<GGO?0r;Q|D-46mcce+Mj!)5$*2t*M&)b%{rju>
zyy*ES4<tos0vQbjo7GJi^e47KDf*SY*ec1;o!CL*_q-2`54|iMlccPP>Z<&2lrM49
z(`9(`^>iXSz{2CI6@OKstR3Xf6@mw99##tK8GX|c?p-ArDm?Xdw5b?>^@o)~91Ykf
z!z$LsEqXrH-?;V{c+bg5aIDn*0Z_3{me8hOJfwBx!LY~~!Iyb+&}KO}!f1!S+NM4c
z54geeukF{g*z3@42<k6Rh6(z!dgM3i2{|>T$Mg&CoBH|@@@~N%EfZTVfJ6zfc+p}?
zn0b126A32l=nsC}|Fi9^U8X?B=VkvjUyBXI@~(;NUt?^)7Ar#aB@A|(FRQjke~1m8
z^Re(a3>_6YFZe!vJ!LPiO_It7mOp;-*-gO7XO;X68p!+XTZYArtf_mr>!-jy)@gZb
zJq_r{CE8(1$r=D5eZvE>7ZF0^dHpy3nj269>GeVnK`w_xBF{oUodPZBl0HD}Bgpsz
z1|7_uSkC(Vh%L6qeSN*)tpOcZH&{M_2z_z!I0IS!PhB<)9kk0?_I-6iz|#}HTUEi|
z0RXR+7;%+HyF22qpvmO<rRBJOH4Wvm5@m@jp?COIA?9K6E{VNpJ=%j8{NE0Cjn&it
z5NNPg;Ry<KnC7i@Xx7svWo{qPJ_6bOvD?>{EMU91;g15Y=kGpfM9y33`1H|Gw)6Z)
zZGk$1)csFb$S#Hq%v(^)?O)BMs<p;=nUb{=iG9mWPjx{Ik`B#HAA=q_@`*9^)bI5s
zj)DrrX(0iTls<~A`naO$4fdM=;oeDF^K=8e6fBeoTp5a1%BiynL^C*0yuW`*Ps&SN
za4=}^PGltGmd^`>1zCo8&bvpji}ZQi4d^gV=Yd}~CeEYum96)05-GK4Lnm)=PC8CK
zp{1JkVe97?e>a6w@^z|8k>(yQ>%q<`*;+~^{a;H+nRr%H1%%ty2mJk57VW&z&0EAV
z<Lce~NT_K}&zl@%<j5NljJlB(Jhy*Ame3a5IxUd#9FeGzk*eZ8Br5bfqc5}Ccx`Xp
z7}i8rxdU<nR1SQuXiu$l!HDH}6SO&gXx&dl&MbI0D8q%{y^4F2(0W)XL)WBIVX|eV
z?P$W>w1&uwWw_r^$5~hd`kKZ@&(&Jdq>JQ;OBo{ei+0#V4b<7u@6PJ($`Nn-f<v%^
z+Vv@uVLmRcaEa==+F<{p>c~>$t~nTP3tanYJV6(NzW))HS>p0@w}Elo*!vZPmpyj+
zSJ<A+$r3BdU3t%~($X~Pc~iw@07#pm)VXBV?s!viDfxvk?UnO4r*dB>suFPkA|t#2
zT3quNOr@@Z!eIa}o)8g{j#O%2r%(HYqT^0pd{Jz4%hXxeo+4e_WXH7>LbTigp=I07
zIFxSZ+@&}$yyOXp<(X$bi!w9g9m>_F10`<X>J(>evU^7NR$yzZ7)rZD-B$|rCKk1M
zSl%vH#K?E}yhrp<q1nzXw7J;a8RGEj0ADFYdKWb&xZZT|rF<n#Ngxn@8Hw5LnSRw;
zqx?$hG=d(!5|-d=VUYLH?G?TDhA!_tXp`!}3sH5F&ATfrj=d=vg|ZF+vN(rbkW=^4
z&Q{F;!{tBxRUs{mt%%!ikk{))vDYqWLpsVL)f;k;apc5{E&M!To-^4jU4fLDNS>iw
z_Xkfmcl;2}K0b2SW{fyEvz;moYPpOt%bnU;G$MOv<31$XHdF~?^BF;c`F>Qo2Aj1s
z-{hTDfex|-_unl)48luslw~-YvS&?a&DJ=Z+8taxJZxWt+cQq|gOFC*V~q#}|B^z_
zl7+rp8OJv)w;8(i%KE$dU(Czno6(ipx7n<rNwBKv4`DSyp91YY=vA>=^6tBqv_$@q
zsKw5gCWAEthjH6Y_C&0^6o0#r1wMnc>+y-S>jk@u*Vv*K_#hGy^F<fq1kK%L_<sZs
z{jOt&&euPZqJj@EwaHIL=XL8UoW9`CRC}?_yt<pNbE%L%(lesub%o*mQCrUH3gjX=
z+Z$m)RBW>eH=(oh^_n!k3hst+%<u>LJa9_)uW%D+zFa4`Nw&mKYu!I)-jvV<_5Qv4
zGaRC8;LQLiK_cI~`nmI*2Y``>cOjoj5*u`c`!*`nTL4(?^Nz-cd$)@5ooC6ynjYi!
zgi!NzsCmuAU<G+6s;!^J!=MCt!+-24jG!Xj=A`I~&hHydDOEV-1Pz7=_q1LlNr@?6
z+NBSM_2!FlOwfq2YE0KJxtUwpOG=*-o`0w@T*JcY)N{Y`jz&aTtB1}ZwcFg>UGZ|>
zoeBnOB%5@_T?KRL+qB5F1wIw|JOCSi+`k!}>%MDbQ-^SBftV(CpaUvpuf8m*ULPe1
zt1G2t>ChQS*IXb{WTb0<52x`AiIz=bvjbz5wRii<yeN0sH?3N<O;(MaGLSRZd!`!(
zJ$yhq>NOP#WhhMwHN8L3A06_ur%V>)gnwAvvvwLRW4o}iW|P_2Dgt5TG9+uwg(kX&
zYKECjWyF~m(y+}XgwaIdo;LL?wwHoUii?1svbeD|o!iy%{hrXHT6f&4!=MT5fA6m`
z#zqHjarYLO5^bA;l2oK!cS2gVbsr%V(V5y4WMmJ^uBOCLXRF(%ds8{_yDgtD#cg_i
z7!2d&yi+K4sg{Z_h_KTR=3bu&qp&>p_+;r@-L){DF47wUj>~n*WIZwM^}+L4IJ?;H
z(l0xk*Z^psqQl_v(1gKYM6Y7UA<jrbmG`~VIplRsd8HYRn4~Mp&NFWiAT21x`8z8-
zSZCW2Te*V{U8Gl6)2^pC0XO@iWoC?aDUWrXrm_c0Q7zhPdAo;WN{<+`+p{L7T_`6o
zilu@}x9oRnDMz%Wq6n13`LlKAWphvd%qPs9huHV{5mVig7?6m)1vYeamUe!g2O{q)
zYwujq>s5*>NB7kEY<(yPB}X}JP@f|#comML{T7NNjMLUbDa-Ax{i8j({J)9G;PCtR
zS|!95DO?d!50Gmeb0J&pjq;DZr$h2=Gl`I1dJg=k`1(Zp#v)Pk2R3r9jM_2t*W`N-
zk*2%~{8}d_9)prtx!H>YFRN(hPm2Wk2C*ifCsF_&Y{D33<O*_^4+ir|h(<i)4>}=K
zR96(6^G}HOHCLz?P2Qm_%EHpQL-vmdt-KwsCSEg6I#=}k;VkTjve;zu5a3tzRFwNK
zhW6RGJLJrF(Nq=H8G7oijl(<-RwlPIXn<85)Bf|giXFvQMc)3W2XctJ!1i2dZtkw*
zBn!z6CCozeIJc#$Ubk48fBSCdto@dX0~{Qk`xXB*_j0d`2h0&?GM>AKnHXR_H66V!
z7tAgm0he0=v9h~(m$_hY>*^d{X^UTopJFz1){O^KR0oU(>|J@$P1AK=SKBgVO;_hd
zofI`Cyu|v+X718g#TKh$r|bp$SU6|A_{Pn0=XZ*)bw(_pK(kv;(bjqFO@=m*0jIKm
zLtqynMe*1o<~n5l!~L+3?n3_i?S4G<rNqRw<3^H1m45!BwN!34l+Sn3qpRnAb`A|r
zdUxw1#ftgYf@JaMM(Q-T5QT0b&NNtxx9&F3*rNc&!gAK*T3v)q&TUoP=6#kqr#cs|
zU3G5;xqTQ@6~-K8epTV*K?qM>-0aYI#d~DxZioUgk6`V3_?t^0i?Dj4u{3sLhB-%p
zS%MHztC*P`ohxB?X!~O~i^jl)#h&t13d2+j{2kK^uZ8w6s2mG4T|K71197UnSQ{tW
zyQ<gOg}sMJR}Wz)akqV%BmVqi6hUuq#h*AT9+)9(cAxBL$cv85J-v*&M*eEm$A&C2
z-1Ps!T(cI;{a6|mNIi8%DW2ryoHJG<w&09DS0?*=vJ~5&A}6{740Nk0Y5X@daNjiR
z)s^A^vY#F5e29#yrKo!6ZvcB_B;E@**Rx~vn`&8pE~(yOy<tBjF){Nac?@sve%L1g
zYu33?h_g5Tn#Nf&;YdAtwCq$h^IgS9f*Nj0!ax+aYgQ16Yt}yY8=U!x%Va2Ki1kF9
zXGSJ}7@Yr(S@^N<p<zZH%6|O|wAX7Q&u<p*zX!yywPM5^L3U4EP=2|txr)On5a^x_
zZ#!xEV$PeQ0l9KtSECI8<Z>2!u8>tAZ!J3bGmf81X82kW_UWqJ+0ie;SFzhaA{YYw
z{Bo$;_pcdC$aXukOqB<EM)csoHP5lcE(hIDl{d<^pSTR<hoEWWN4;DW+QTk<^piue
z>WnRM*H5m+om>(ccqPsAdQo=buZ5y8PyAw%=rkgnKYz<N903o_l8H;_Zl|#GVC<5m
z<{ss46GhsM*$=)n<;Es4$F}`T0@WQIo_CpHA?|F(IV=Um?8_`{#fW9Gr>!@g@tYU(
zc6X3n-!{9mr<-E!9}BiB%T~}Apq@CA;jr%RcOHG4+j!ogjtqt1`8z!z<!VMMsg2pp
zuM;mP+OgFogaYMVAmyhi|7(Q&B%1;7A*AB!c{}ri)oKa-w6a(Uc)g5M&!RErg;<eI
zHJrzk+6t+0SvQvU&*do38mR3LpOg8Su8|M|U1f(&1PsNPpocoPRsSAE`-Ev$@rE=y
z<*G<gKW-bb8{ZJ=p!xB=T-9ng$a=@=$<l3lc2_q#7NgB^z@^=+TCi(5H5~sG76v~h
z>lw0Fc6kUD&I5fS7W4_zkVnx1-d4dzedLd9S5vgktl0<_HB^`%Cqcd|Wd0H4vxI-N
zly61gPZgIY|FZ&cWmh>StkZM4UMS#jS>-E^282fp)gSkcST*t7EimG4hx>c)fPFn9
zB9+;}u=e0#tHjIMKU}d;>2XCVB$z!}A2#1yuP9z2Twz;=GfKp?q=s97p<SN(ON*HN
z))q+uEyhN7uM!>D+(Cb-A!>`k?o;pXf&LyN`SM6>-g_8DVRCXspXLQJeU-j3#c3o#
z(lI=p8YEA0XX`B3O`@=m4#C64n$3xF#|75vJ)R^CdBpI(0PLOA_^~qS50dj-afn-T
znO_Iju@u-0t3C2+lGx4Pg7HiGF>+>>RnZ`T)Br2}W3_cvrltbjQfPFA<4NwlLk@vu
zGiVP3)3+bO+kbKTgw9IXy;)tEPGRuc18uupRZ_y;KvSAx8oE`6I5`&NB*g8aeJy(G
zGkztwtn`ncffWOW3q9cB5|WE=*hN*G$O)?xvP6&-IlP<(YKaQm9%u=%Ymd^6iJ##B
z)VlEwc+{y9WW2OK6(?2{)@k7TocO~kcywz$WJ+nPIbyudclUM#fOv!R@7h?d2;_T4
zOCHpn=wvwOg7PSVqvz*^YfXli*<sKuPJ4}*bdRLZleL*Ti)@<OEG^PrQAK#MlwXtb
z!XFWK*WUYdoG$m)9o=t=0$mi5`$n{;`I&{xqfU7;d|UQ7`Xiol1Lkvq!U|mlt?#2r
zdN5P&mHI8ic)i|kQHNPQkZ`Y`e8cnsd}uqeE%EDM5d5TCMb%pYbc?67gTm>99n}E3
z`BkT(0CYKb1Rr$#;!q*w&u01cS$pj8u*R>ZXmPB_Gi|>iocEqiRqsuj@adY@F|zC)
z$CgC#0Rmj+voiBmF-bOdvaK81iYp`hQOG(?KK99AFwwphYm3Xn?-V~8r%8|bzw*3^
zUmBT=O|r2+Cp;Aw411MciiDL38;hFsPpoPDb=jFpW5j^`<N3senoS>VVr5Lw9Qjaw
z_jpYDxbF0)$sW~m3gp(gLk9?N`?M=QKkyUD0&K3Smfz?cYw|h}F`hp7EjB-XN`7(p
z<ifgILN@aVBQA;!z^OW{FBeZILd@4a*_|GGN^_L8zhTkC^0|ZN?^0<jY+Vt_suB2q
z>}J$aw6^A8c-gnS{T%yNK&*rfjHr>YV1H(@h9N8<%Lr%bo=Jb~XTw|TmY(4xCrr-m
zc;<81P_fb)`d$p!bA9|SqTDSJh1i8z%cjxi8{Kq^6Y)Qp1HogEiZJ)i3{_*bL&anI
z`$u%GW0O{ozn}11ZS@b6i^`4Hq_;+X>Mf5gtPeihxBJEe(eCPgpLa{BN&a!5+=KGf
z@}V+GsK(Y;B&K<GQ%mgbi&hc0&8Nza<R6SI-7H~m`O~xKQ=aEUog;?c+Azfb472!m
z5rPeQYg2*8y7j!i#g5!Ej=M;V_Tl_i7aOT(ta!hyvHSfXLOp4Ip~FL3)(eg$`g-$x
zBP*Pi_IX0qJ@*SF3s$D-rsKe=Hd;|wjoFm3vaD4{zyAB}XWUVGQs0~MM9!NZk%`aC
z?VpSW3-v`U<eO+m4j(2=M~7<9<;lpZGvXS%)b(9clJ?eL;y5Dt=@&<6Zkbq-T?r{+
zIx(Z@sv{if7O_)ef~@JS&XK!gyx8!3k6H+WEaA*9Lr2d)6Keb5Dejckht~{vpIT~T
z$*sDy;3uJz{wI+$%NNhJ&6uFXf3hHMDuN>G2DhxtJ@e)PJ&T9)8MS5o)|m37pb7f}
z2_0$SXP;&(K3*K4uabl#YTb)VHL$jrav84jNEmCw7P|5UVe7#yDL)3Gc~LH^$?Yuq
zwNQ<teD%ee0xXSS>wSBJDAG51mrBiwAk;cYJ@n{6ZB#^`grV{NIdO6k#&rHy1VCD@
zybyDBW=zrwr)}^oYG-o5fq|%$)8J*RdPWiaLp6-L3y>$)ogN!JJG9fLUFhw?%QWbY
zcDuPKOV%up94z<tNw!Jrnc?#S)5o6d${5Kiag-(OMK;QpPp{<TdWrj?^WR==(bETU
zt>{_9oWwe(>sPfRB>*~+vOW7(|0IO$no+z<-h<qy6e|I$#gw1AjV`kNqbS>{E<<Y@
zsimd`HEAx=`?F^qiLpNGeK9&NPHc)q?Z0xXE|^0<9XM-y-uDnxc&n1xx#R57xAu&D
zW*L6TVBdepp>R9bS&imN^{mTvTyJ~VrdUqXR@8fo_IAC?>ftct_Ua_t;^7y)pvrZ6
z2<NxFKY$LXi)&b6UO0<?J5YD_*%|dn5&8Afrh-Mn@%OkOUq<vTtEdxkg;Llpwy-wN
zPz;hetMNYZ5gYtox1w_y=eZ)$BFJ9?k&!oHt(*fk$qn%>Pk=!)Qa%SU$sR9XqQy@d
z9Ny1ZN@tw&Pn;@i(-AEhk1OpFjUl;9jwvEBQ9XUN$rHp{6=9*U#bXI)7KdC#-N#tj
ztEwc-$9SKO>JW>GnZbUm`$KKBv8xYCdjN-#lg&G$>Vhg-jLL?A9d~1HdOMbTZIO{6
z{^jgm!pBC+ZL=J{Waq}E1et>nOrX%Z@1wmxDf?dRn#M-6>?@I;HC}F`HakE5o^QO2
zatFH*i|db)0(xYA@@d_l)<_=0Cjx&>T4*$@Uz03s>UMYL={|~N?Y*Mhy#K)XPr1Q6
zpuxSrwG2K6J$Y=<aBhXm;#6e@K-0mijdTAxxsdOgwx1gt%*g#WH;BXJ8_rLigoQT0
zhh%)UJE1>GOrt(kAs#LKX-{kdQu>;wWwWsRIe=A8{Z+yT&(-$K#{N-;M<^>Yy^~hj
zscTMhlFBRheWWG>4YM}#l-JI>hrN!Sw=)=F2-(%RlrlUaU|AC=9161y>rTCxmJaw+
zvr`x{*Bh}eg#CV4T|g*+KmnQJ%4mmh+e(+Le-PeQ)0H7eE=iuaT7W$GZMvg1iFXO*
zs-KKWdd*z|fO{qsZmyfc);OZhvGT=79hm(QLUo8l)2WZB2otHi_SAJ})Ws|jCyJHR
z)!=x(?KLBmvfP+iH`0w-Fg0#R5V+YxRl9azJv#%|k#oa!M>|`~n+~q0-an@W#+7xf
zxSj8)-s&1vw1d=Puit3S296!?(QH=s__x$6`*=)JXP8?|lB#<j$*g5SR_W1^IsK1{
zmm-*t0u};JzHhhiE&h9<Gg%q(o>KGV=KHmHpWG%ek4j=xkNx3<;Z7wChd^{;{j6f!
z_TE_YG}I4)KlArZrO{%NPjc%Ufl|&o{=cl{y>*?>v5H3<KNl;)x*yk<BE||iy^KxB
zvu>lzSns0$<%-LKj8UlvG@n5L))*qZY$#}E(cph{ws1%rTXrPM3BUSNr!Zv;3<@bd
zK-&N9F`H5D)EPF+$ZeJv=`~X7)wyH|Rm@&xjQ#Ie%R%}OoutJ$)kSLeSqfw?VGyn%
z#`L#*>-TfsXROHDCd_)YTOC*bk7*H>^5+Z?nsHpusK)s0zol)RFg=DL211sAm$-Ex
zbWF31Pvl!VGvwdT#rI!@bCJjXS1|OiA(@!|->8%Ozp1lM3i*M#it0{L*eSYdz7kIe
zkn~4?r0Xd5)-cKa4`z!jW59w9M%^Dl6bO+&<aoqziED!4=QA2SXcI79$k8V)j2ZiD
z)+ajP(957gSB%7jR-onV@9*t7=`WCcSiZ$;hs!2nh$GEF4*i{p%{Q9mckglyAYA&N
z()gd0!I0y^c89!BwmFjmE*1q7!JYU)cs!tDC}D7Cy$zKFFA<M=Lu4rFZBB8K<WT9u
zO2a%xn(wAIzt=Y$_I!r2hL&EVC^m=-8N*mNfj<wo9lZ|@K7xgzU|G+{Uj(N!*=svK
zU!>;!qHn~0eg5u}3r!8wSR%Mj^0wTuZuJpON?~>sVil*pWu7Lj{6<A6Y%}6iI^j6B
z%8#+40%KJ;d&M~Hgc++UZS>4DTIK5W&%KDj?t|Nx7+}ylB#+Yo`!wjM*deRuBqK@X
zqeke$rj01>)!tnj|0Y6t83HvP^tFH%{&C~0po>cy3<2oQW9p<ZGC?m3b6jbg0l!l@
z#W|O2sHu>r#WfIbbii2ecsAejNqW4s3`LFUN|^-51Va#-Wa)I_0wgy`jDNZMMobbJ
z!-!cn8wMu)NV*hQnigw-+g)xdZ`5=SRUBacv1)T+V9d^)MELo|l<aS74TY&3@1!t9
zh*x3@?&*Wigcyae)ebZNTVLWF6OpnhceRxm>4KaaJ28xoQs>XM)YSa!toPaVwQ<V(
zlY%?bDuZGqoDd`w_I+vjH~(kSD3?JPtvgL_nMo~%nA^{UO+rK$?AkK8hH9kQYlChO
zOIJ$@_Y|4bN*O0ty^eD6L_^hp$ux6w+ouc?@p0XT+xgD#6WG)P+cyei(n}MXoo`v(
z_(-}8fIp+{YhbM~bo{xPB>esUR(P1pRuk!&Fpuw##WPkJE{w9+iZ(GlqyA&Ri${5%
znXr|N9rRui&VKtz=Yn3ZZXP#~OkrptDsmNkE3><AbFeLpk@Xo)WQ%5uL^qqHO$)3F
z$gF4Q)s5FN2Om)9edxH6P#qjQY}3ySGKz(Pa*n9s8SY6sH}8a^h|`~8+RB+z-_s7s
zN_s}wdgMj&7JR`Ue@s(k^3%ZOk8+4g4t*PtUQANz={e&vaSpE-vs9el1Mtn5q-eQS
znE}NR=RPWVLoTOg<zW@{oco3Fv%_awsP|Dmv`a&%h}s}0Xi1YEEdI^=T*;?<ez(#~
zy4?rcbuP^i58nhn*^sqHM3yG*=2k*p0|;5$;3P$dO+!ycBF<#$n9Ft;Eq{M#aGL!I
z<m5A%s%c#XmS#0-|Du;DW7STr4iFZ>eK_Rl5Uu(L8K=9>^28*KF)}?Yj_LIt9w$hg
zdd7K^ck7JMH@!AdAUyB%teHLK>W5nnlsOCI;-5yTFD5{T#qPr}xU}g6wAw{Hb&U^-
zLQZhsQ+Ap07A1TQUZK8g9CyERC*!G|rllpv03JKuUAVJ$;9~Qh>llNC{clXP+n8Q+
z=zl<2xulvj$Rvk^2ZT+=F;-vW|2)q(pL~xM0|qnBS}T3tTEv(Am#pF}C!ehWFg2l)
zS;2;le@OnDQUwaeqWxK&x8T)XBpk)SF7k-UhQX3C^DE_rnGUzb6lam!$vKAWyMpEO
z+56NpF~wIB4X|tb1w!4TImbE{Y)@Ns{6iU;=Cajo^?LtEEWVC`DhI>^TR*WDZzfe|
z;=#u6w7q6Vrc2dRyB{3qQ^)1Y86}{3KdHcuy>9Ck>Xf1#sXeKc!1kd<x+0NSEZ-_}
zK6xas)Sa9spm`+E26#aOfJ3wYNx|zd{~AV0*1EW0L=?dky#&q81PP$IcQ?BgVeP?V
zTCY@;w4_o<4(Ol`paxzuGvca3R-i~0(hR)p>JG^@Wk)$WEOXu3KR^z_;TO?FwYlQ0
z{Tvfxo+?^)YMCng#x^!t@HhcfSesAZAxi9~gfXTKJVpV<`!h;+H;pH_@6<#(E!J05
zpHC&$%%n`%M*O7QVF>ct{I4>+`R|Med|px5-f>UeO+F|9la)j7Ju!1al;Xq{Wlv*&
zWhtfQx~YdXQ^TpKiZ+oUn%TDSh8D)snWzi}T3z-WVmOF<zJvW!Pn{#0e*_-b?1Z*r
z4!5U1DFkH(kF&2Uy;(&|P|K|=0H-Gcrlun<TlR^4@9fytw(Lk9GB2-6m{rXvV{u+<
zCUl=7dxOHR+GX&qn$08K?uHreYVT;&fE#6fA|n)=wJo?}t~8bVHT$JasU;$*=%JF;
zjIXLWUgOpCwm=ur^4s$<?HEgklbQ0`R<o1wxW%;bm8Sx!UVv0IIP0d9*L}&yC$`Yd
zdD!xefO`cZpSP!X86`LN+uZWr*R03B?(4WxoEk-teTkxt1tEP4RSO~)EE95c%5YfU
zmj~1lkXCYWj9bk*q{K?#et21mTWg-Of2qTEQ>#|hz|;69mxkd*P$9V8>+o?NVKmU(
z$<b4|KoXo!YajRKt?k)zDE93uc~f^lKjQ0QFroAC35>H$H2krm@ymU3ZbV#Els!od
z!t;qNJaWAp61IELnr7(aFspW>#<%%gjvLReTrnEnLt0fNK{}?cE;S{(#19<c**Br&
zk`|K5#wuZWc=0hKv2@h-fhfIfVCpP?#NNFfK#!=@pr-c?blt@)9t<`2(xZtR6u?;`
zpSz4kHwFfK-_BFJDOu*}P^+mbr3lUYk{Gd_?t=M@`HaVV_dQ@T%Kl55RVp{C=!zFG
zg3oI8dcQ8E4?5Rra_lf_V#}Wfg?Q`yLCgy!s(%pk3HBDKKTd2}8BdgsML0Qx@V~r*
z<}N2)93H|dz-T+5z*<4*bk|x6>6`g;rORmL-+p!eQi-GS3!=7FxV344rm<iYfi+}0
zX%#s<R9`sJ;PmblfTSUy9($u@i{vvQXxfylx-GndO<k9kKnD6dMqHViFl+gM_Js@z
za4ydZH1V}*k|#&*&^C_tEUm7IcBEhS<R7219xLPG=w4WEerJnNRQG}U<Cjb_66&>+
z7pGuuL)?`@H3M&VBezSo@>Y5A>4k&Ds>5HG692@c=Fl6)Ii8d)^j#6_mOOf2;2@(u
zGV`)rF)6%;B4KF0xJHVWX+fiX%7@<D^9c0TrD=<3t|P1j&q<06X-us>CVLhXg27o~
zrzXq>lHEzaYuFW5eJaLOXhtd|g$F&vRNuTi<BQMNlx*B_c379RK6)c&$6`^{(?LyZ
z-HXQx^pofX&0H}Z<1vOa-MX9ek&sa(8Lu5yif_E2<f1_%RbJzI0kONhk+D75G_rc<
z>jjrJ`!Of8JEIcsW~n+=P38xbLPE19sjxajBXMZ<_#_iqerBLM*gxJILkZ5h*K=S(
zQF=sZG1J-HjxUgn03$vvwVF`&0gq1*Vr@R&2ukwn2m24h<tH^7#DI&97n=_NGb)Pa
zjAh)@F+k18h1U_BRmOX$S7pI?=fiW0E2?lY;Gc5fVR_N>eWRM%tc3FVpQR!G5h}4s
zj9bxQtmEbaDSa>!)PP1;#z9wAD!_;_V++J<hTkKljbX~{u}OxKL%Ju=8f*S8PuR5b
zkSZa@D%Hs)T7k8$$$5jJe8ny`6!H_V${dKz?^8_3l)%%^Mvbc*1+D2NyM&r;whvbS
zlm{1T`0Ou*ljjQSgA5I7E<-<EP?Y*H_bAkC2PnQG#0^L)ZZcd)*xL87*X~Fjb3I^`
z2G6MKF0uUt1#{xS5)ugc{vB#4Aei$X-oEmid$j4<_dm&9#GoLB!PB}UDi@5^c@E!$
zYjJn4o4)?RLUiJ4Z-EJ=z%Do7I!$3?=KehZ3`5^q#9;)9)6!{`g6Ik9eYUMXrL)Oj
z7mR~uLwC`wmwO#>h`=YkRhLld0~no^p$gB5r0Q+KcQD(Kec|46<tNvFaXcg<rB(5(
z9yCTqyp-x>6+@Yl8o~g~e1o}Pm$fU3X0-gJ;V14?lpWfjM#JA@K2x{trRK9V4?y0x
z)otvFAYh+kpmy)@VLPX#nW#>hU2x=|e^Nnl+81P~XeyqV=pF-SJO*i-Xgyh8`!lkf
z{`HPV6!v^dSXx|SiqhNFl3<vso<z@^jVf<-3Sq1n@&+pJ&@_ENb?DJ2phb7C4@{b_
zS+kpOayZ5~48yY1?X*J+6G8L63;f_Akjs6RXJ5_To8}9ok8ClvPoagYI!2ygKFMmW
zQW7+)A}xS#pck=NN+g0~;W`g`0?$o;($5=fxgWgv403NG(;mG)n%i}a+OLD!rd5pQ
zd9_&M=uV!q8x{pVDs-xt9Dbi#893s;y8duKDu$a=ntY?60<!)Ad@p8oOW$kKl%^pU
z-_58BuLe(d!vgPu6UPu!T^a9iF8zKAYWb@j>n%Y^V`C!Adhx)#4-SFC<iIx5_uhcM
z>RtXPqazxkuz9c2tg_P%;T{$=VV2mXrT~&z@ju(}rlLBV`?jriTXWsP{i`26z56tS
ziyxdM67OT%YOv+ie=eQ7;TzQDz<9ds6*rhh!0uS~blw1WPzmMfYM-UGj`i$*;=Y5o
zlo8>;#?$2|s(q)3k-~wdsqTw{MbT@$c~hu$6)iyMJqo0xROzFQG!gk7UqvcgZ7L$)
zvrJ&rirZN4risY-^&P%3R5-UdqenR`HLT4RWU{1ogiDdktu&s8*#8Q5T^_HOsn~BV
z%xkc);KQO1fACa%U&&ceo$-55E9nmjs4_k(F6*(B)kYgfH?lU-y}}5fEIS$Pb1tIS
z9i-wQVdHjExN-&e{-Xbfz4Pp9YW>>1ZWL^QTeqShU_(TjQl*QCNSEG0rHPbCCxHYN
z6_qAkdhaFl9)cp$TOdRT5Rno{fB+$cBqW@L`~RHh8E2evKEZkGD_Lu<J9FOix_+1A
zFXS-QLUH*?^p*9n)DWfuBJ_Fj&bK`Si<C$<5P_vJ9J1c#&?xZX0sth@<3D};i}S^m
zdd_`gYC{hG-2LGW^F_C-DQ>fmnY78EBm$nz3Kmpr$IX4J_lsh(hh3{0PIaPHmvD~~
z_OTJ(v+Mz<J|gU)n^)vW?W}gm%g2pZ#2bf`oCztZm0CRLXtma}gH2u`N)r7Xh=SD}
z(~!v0m^1mt?a%wBNNawP<}PF0@DftWa+6?rF14WvzxU$JVChU>!r?=AxexaKU=F9z
zgVUqR0DP4{`!N>DL(a0TRpbQMvJY~W<3IkGnGDL}n>noJ8<ve!Q@Z)LdkOnmcTiS#
zw84ik&5Kd(((r5f$x$QwayR0Ujmg?Qg1<ZiFV(CV%j$p1SLb#(Z48Pwp!FKj6Q0b9
z4M&qU2JY2!OtnM{4Mv!;kM=LQWn~|3)}t@N-9ReNDao*4-_5bz%(~#@)%X-m_au3s
z*ZgYub!Aa<3Oq{N-s#n{(%+$7L#ESyE{;>*2GPgK+%tGDlsWg%-7S~vDp|#+6e8MW
zFC@qZJ#o-g`IT&Po)UJX_sNG@pD-8UQZheM@G42=JU^wlXaFny2r@)eS&TXTbWx1I
z$`mt9ZV#6c+@>dher~3}lN%W3oi=w{%2=GZqOyOBQY6yI7AE!kxx=*_=dn|9N`LK7
zZGM_`b?G$YSgz>a(LP?<f3m=h2i*x4Jyim`A8b{)ljK0P>epN3hx1zoty(P&zc6f_
zMdbM<s9-2=1mQ$1E~jz5&7{8K*kCf3l4#9>^-g7WX2mtqc>MdWH!ysVT_GJ(LAWI>
zM2I~B6_jS&bC9>VxbiIdsbNMwSyQT&sFZkqQbW4i?sm5O(UV(mgw05<5S=g`{~8_S
zZpaOLg+6EQ+#QF;K$2=7>%Ow+RZ_ZXOLBW7<BzGFe&ndXH~(5phDAE9LL|*Z?K?8B
z0e19?`wN>l*!{5=(qZ}Czlqme`km}~SK6DE1a}y$TfWHo<uHQKRi!O2gVR21&B1cE
z5I9)_mZdCXiE-y+6|8$|43~W<<(cs7kOtq}8Pl4KB&qQ=3O@K6sL~lq4J#$vNA^<*
zWI<Qy<NO@??I7D&QaGt+wd_P$*)qp{v;CaSmk27U_<N>j>V7tir<KJYH09V#t+Kd1
zW@B$>@d&-Y|6(8dPl%|oR{P!AI9e9G?{go|HTfUTPy6k%63t4)*JY7}l=RU+Gcn&o
z8}ERI&dIcd!;5)tLc}ni?VnN;x&*6A_snM#8@ZA3skoG7_M5`Lo=Ee@tVioW!hR9I
ztoMc}Tb`|3(D=Nq)as>ZV5&$Ux3u46l(O;pr?)Cf_u^bpmG5Olo>kfaml|bxj@rqi
zO=C<|L88(puVO$o5~bfz8@tqXD{R@BTxU4)%JvDpK-#$$U(weuxW<<&h^XyAakW_y
z?n}$-3C(IgnRR$PGN-%c!!6lm*zY!k>f}f9v;-n_Wp8_qkwtE5q*73Om3vQ_ndF%x
z+Ey3_*u`3y<k8dWtwT057%9+5mG|lQdSp)$3O#as*VCl&Xo9Bu?Qcv2yR0go(DPHb
zCCRL(-K!Ese||uou9cm!lp8$pS2ImMQ8}Az7Vzr&Mqt}o{29|^9-XnFnW0HG@%5Hr
zFYcOdca)gqZfq>qKQ4r=GtO`t!9GMx0bJyBFmnXYm${1KGg?qpJt@uR{)zD;kbG-g
z_a*m|><6iWCk0R6SyA|+P0S5w9-F&wo5^9C>cYNNIy<}_$Brf}$T0ir4F!x&-?EeK
zNLWb^NoU#67?*h*>}(NCkyr~oEWAM#C`C7$x(+{u3OQ*#h*&#Bu3WE_MW6&BSL+E>
zrrAeE#?0jPD!Z3rU~7WC2ptRBat%vso9GO*N%w4`u}s<<D;C>+haaz1e^xgA!1*yV
ztx{>KJGbH0PQJxjyGFgc*~i4JH&{{f26D>w@Hg(@nr5R7dm|ndtelq(WeMJI`;Cy;
zP@62N7*f*kQ*j?QEqv%65};nyNktX;ciL<`h_k{eVPD=bS@-q&p;XX)PF()U9r<Ik
z^_>;!3-+KTpD$d{fNYudk%zS|dRn4m&H9QalfLjP`k%fKlQ6M&qV}@~=y831?7oWC
zJAwcBUoMY+jtA%C@%z!a$~M4d%n{C>cZFF*ZFszRiZ-kL@|xakR?V3q-lhvQs3sSa
zn*X;eZ9obyw**EhWGPGlTjP}h7SS1>osGUjSp@s4s>{on&wNId+(F`lw&b62ehJFF
zsn4O;LN>6DIiE-yp!%)xeZ~a$2tNG_h83$mBlqFSmU@Z$;seWOwUeG-VlEDx>fy^k
z;mV9dQ+?9JBdpY&#~}BcvL)m&T~-{X&$AM;TiPC{2DBdzihtiu?YtW!uV33{)nr$`
zQ#b!*K<v${8dtl1kJ*x0^S4X=vq5lobeo9@?|DV(g9hP@WsK<>qfIySyI?|fb4G{y
zqfjPBgj2?E;N|UZ{tF+ZX=SYrPlw7dj1-jH<9Y+HK7!<C^(nQ_Xn|z7ZkM`Zyz5vv
zDi?OIn`Q}xXVQ;xwP$Of6(8UmcMpgCZhMyD-~fpJ>%~*exM>XeqEVOkCqx-fosHL$
z#lC)NHfOdY3166(LYGWMvdur+*B>D&dk!VHG*-;F<vI%L;U<%BgYSxm?VKqX3U?5V
zz<$njp`*X2D)6*;xTVUMQw0O61lGBWQ9k^no~=C-tDwK9!h5P*OwR8xhiu!cpuG)u
zE8Mr>(udML+km10eq+D+9Xp{IRLNZB(p*kbk?;+B{4)N0i-Sn1FE8hLXZ8=+uf6$a
z^OM;|kuwTaanpC5ve+xG>3<4>r>#Wt_%%Tp?_gTr6Y;2FCV%t46_StnA${3GmN|eI
zY8dd>ORh1+9+=DbjOF43FD~77&iiiuvw7&I2`hcY1N9>`-U}xBGUcKx_*hHQYc|cy
zua`mKfvwro;fs_S+BLmRP#SrgQ*2V%&0I`O{QlT1AfsvfZeA!L1#aGcq7Gr15e&%;
z&9Ect&eF_^jX^VxZA;`26mdd#R_AR&wYG+XfUf|E&i7P)-VN&HVCdU^h3&|aWy^4X
zy6(pcveJFB^I;~dE6}Ef_QGf>OF8j?wzfU>*c!fQhxr69s+1A)38*x)TfAN+P(36R
zM7(LT)4DT0(IgU27Oe?h5S4e{%FC;e@8(xz>tLqc$bhtGiKMZwRKEqKv^_z_7|h`R
zqsfBsb)a9*hQ{=aht5w79Wj#@g+1RvfnP%rTmalm^=Is$yri=_`ie9%s&|=EwuD${
zro7GbTK+vX93mj~XqI%@NySt5d%A_uA@TCIl)V{~w2QZZelV;2MT$&T^3PAi+yoly
zvuZAtbp=g-c^S(`G`j5qdttk7FBluUfXIH)uMh7a$w8iXB6|>?ljOcpda(6FkP6ZI
zo0O+R#bypbRcOsYEoLYgnq#;oj*9b439k$^kI=hezl8XXM5HL@EO`7@OmVy^(d{bf
z@YA4LeEp2MJX?pA|G%0r*z~Vmd4xk~V%e^~2JI-v%d*+arq9H?l4l?PFnnfZctQxD
zqUxEd+MB)>f{wkR*@h9{{xQ_(P?Dx$k$JM!N_?o6;_rfTpTGFIEETpyK4UZF@3D?Z
zDB;`G9h$H37T>o!%xu@5=u1#@Rpco2oUE~5qYeQbooKw;w}18nO3354X7<7hG?(uH
zU=OmvE>I=}#Ei}`QipkQkD=}?1XM^xqhj`R<aP@f7)GADd7ws;y8%cUg#Rp!VpLM&
z8ePDUA8bzesff&5r8&t2I;~*-cF#uDFWUa!s-a&yZ$3()dEZmf{T@i8vCvtT3T<)T
zax^+!bJ^kI3dn^xJwk7Ij@*!W6eASU-+OWpu{>kwcF53^?V8jb5_tH@D?eJ#y0gNV
z*_FRsvu|uGH3`J{3%IW*`KoBZ>)nyY;Kl7XpP%|C;Dg=_i>}J+v(`%nPc2G>H<UV~
zLhPTNY!&d#>E;nxHW6%3NoqB(8m(U}w0Ith(BZ30rz1QI7^kxq4*MfA0!GS{A!?I-
zgX4p)Xrb4l^Z6t@{bY_oDbH`o>2q#`0-wdZ6e03E8;7Z9MO7^q4-J(bO!MX}yrjuT
zyeA#UfZKKnDq8-Yf{#Cm&_MIOeSvzC3zoWR|1f7>#icj+c-+N%ODaAy8E{2Z{s*f3
zeb8zMBe6w6k8JCDPmVc_ixuswUEHdR0R*J4Y|-1Ii2+^KT1qv0yU`&al@e9CM$Bfu
zb!gwiA}Qokl!B6@#BI+3klk|1XRy5RA{p|Cb<AWBJ47V>)~<c_)a~!ool2&TU8}&C
zwo6U6Mz;<cV%L|=oVP@Cx)=VoRy)7yT4`el7X3Rw^#&&t{2ko30veY(Tl3t?Zt#*W
z6`WYoO*ZRc%R=M0x{+|#Abs&Z!jGiYc90I)M9H%aQ~zwXM3gfkJ;UA1LgBb_(g^Wv
zF}O<NBy#d~b(s{+B4E2t52n{|-mQ1YciQ#o4X2{tGR2Pt$aT?9j4nT|Lv@>D>8n$u
z))USMK3Nhpqt&-J`}s%pO+f6>&M%mjsZH5P5kpSKt%b#ua=Gc~a_B+T!$*_}2Oqo7
zz-y?kAV}~QVuv#LR_s>mQ`+DHDRXMz@k;eB{exM*(=v|Q7ZUnNa5TPg%<4Tvj(p83
zs9GWo#Jymd7~ztjast(F;dRCg`?9}e2KzqULUQ2l=PAhX)C`%$jS8LC6X5;zZ}z67
zMC^USr@snU*)v-cK@Z34{rJ^uoZK*ah%u)O8$-_*iy=9#dh0z;`Kvn~@Oyff>`e`2
z(UqY7!0(3{{?@UQm6T{mFM5GLmL5t8rG~OQ<=S?E6eCo)e$jED^MwS?`({fScAL%h
z$zOWf$8La=^`<}SPkh^VE+`8#(n|ej)4p+pPYDM|f&OB59+;v|LdV*cp@1aQP#cz6
z0hd1O=W8m<Oegh}FE2bY4>e3$A%w&Y(zt%`rfgYUxS+?zuw$?nRw!h3JV^#8>lz|0
z98XLA3S#2TufT)1w7&et84w$)ij)Q!w)ex5_`jiki`6|<bte0iX40jbW=`D_jSehL
zG^&S19Yrs`L&>b;UcE?lQD+mKV63dZb{;c4E)qP^3aP=HAu5jP)+)PX&%akHo<mfg
zBO3s>+27mdMe_|-i1npq=|?86F$oOdykHew-$+0NxWU9ueYfn!RKtnBJY?04b(Z&Y
zD4(sQaJt)Ljf~ZoyM)(~5Ac)oO68_mA}Qi%J8D6<-4eg)ic(UyGurGT;|-S?R?vO2
z<#$_L8&1zN2cytGMk+7!{P?!8zoafD2j#d)`MWHsSCQ~ka7==I`|mNifka6T=dDCh
z@e?S~%a#_+oWo<JVC*Y0;UhXAnLD-KiK-nk6tzK@qyVeaGGvAG%D@(&LhW+nz3n>$
zdFiy^O&y@i7~(ySlDyYWmZ?2%An}wWNN)Q_VQJ}`U72QH5&cV;y2=t)7okJxl?U?2
zt#68p$1xoqNxo{nXYT+zO@#kDdcc@KpWiH{{KrDq;7pjQHr__QSiAeKe3bj6_^Y$T
z^AJZoc6iTb^H=A(?ev}MB)F$8s3SDpAE{hWzuDs#4csv|E3X~Ad<7;d+k6L_lV}xo
zjJjWk`sQ~#__*r3JBx7l)3@j2Zy9c$4Ll2%$`~H0lO;*)sxRTXvYJE8<xg#r{QU#M
zVDKN8os8&N#ZrH(ai%)lxV7YO!8&$k13Rj_i)HxPVaw0@JDx=zQM3Ee5?NJu7w9i7
zQrf0-{;dwNG=Q$cw~$`s7WQ`H@KMH}8d07YK5SNd%GiQ$QlFy>i;=GFdjtOITJ>^6
z$u2+y-*&6<+*7v+@d{TcoAgRG-WF5(Zx7+}r-!H$SaHO4A7QaOH_V?#?OAT8L8@vq
zGO11!lf9?kAr@vG9k{t~gFEdCYymkG&zA@f-+c8&#mP?PU+B&!FH@;Lvsuskp1Yt^
z{<<-v97b9>1U67Hs^_X^_u$h!e2B<zTTod$I7S+O0n8~QehLM^oMI_h?US-fW}4A2
zr_*7efp4WnHEu3jAMvaK0uo=COyj~(igP#vSMUhWIq=aDQWlzoteeHV{*@1=V2k*z
zAX38aLyiV-6c6B;vjPHc82Dtvs*EMA>bdf-&Xxh&u{D#96Dr-JK5BG>M9?RJSwdMA
zM94U}{X^Jiwriy@N};(38v8qrhNcgpn68KJvH?q8^W05Sz0Pssp~{(y7tQ%Oa5D_X
zIjda|^^RG)da>A=rJ=Z31K&_vT7SN@GQ~KFa`yR<)f+Uhy`#3goBc`KsZ~7fz*!s`
zNKu&WUMXl;f8wkwxBr1d%TG&|RH7a1##6tDx2iY1<aBG_)T+qu{hOkBTMc?a*ku5c
z*dKp*`)oED-pLVYYbZthWqzsZ`7}%<Jd-QL*VP~ia`S%SUddo{dY`t5T+wg7k+!t>
zk-GZ6eB`P6aqlkmjk%&4>=?-kxmad15amIQ@}K%jE!v+WJE>tf{kz)MgHedbJs{$?
zK_y(rWH|ek6|c@(BLOiX@zl*57=MuqUz1*uNRJJf7pJ}6C+SO9DITzCK&Ez80Go#S
z-Gh@-b{-ghjaPV}T3H;XgUCg-<6b{?%~g6Qs#8K1zP!uEwZ;$Fqiir>GcVC^sPeRP
z**=S1dC5L%5@hJ{`xA-p-wpL0oJ(&p*PukkT~|SX*Fo3Ebl92HM`>Fx9T6R)nw=YW
zzY$Z)^by!Tu3z*EGP$G^R_b39?wk1WkQ=hNwXN`VvJW)zTT4@({x-S`F>$3ukcT4&
zN)sQGOKusGGek-o>tb_(^T@3}MSZ;b$oAi&$-OX~gO#ojgd`ZJ>R;!K8dx}MVQo#-
zdC`x+%z}ACk?tOO@8zZovZxL3@Mhmrj}{Vs?sR^dfaa+V4|r>Gm;Q5fx%87@2h+QB
zi39fOI$Riq`JTuPxSppx_GV{)d=Xpy2e<o0E{DLlthV~^kW)Q2&xjmzCpy^u>R*Uh
znoaUZvaPHa^sG)^DI4Ep86CO{J#s}?7j)VH&<Fp_FhEa!3H80)-J9L7t(!`swCt}9
zJxWwJ#>0T?IRD#hnMeG!hd|p9#__`GMYcF0-f`7YllArjlUs$_f0HQ1b2pTjjelQk
z&B^L(5ve6y0(5wfRYMbIc(|Qp6uO^kP1zX!B@ct#N+Zonj<~(7I-_z0(*zY}iCb7(
zO43;CvE}^r$h_u8iPm<#v~FNEuwIJ@;Gitzm3vwS-tf=Qs^9wn^I2VAZ2_pJN`VzF
z41`i39|0$g+1MTSZr<N5@E`<ETQa{Mco|uL2L}1RIkWT1Wq#^_{*%B$z^o0CHp`gU
z;_?JR)A~MKRN5$KPY)@g#?xV#Y9kd&)Cn01qdDJ{@vP64${!MDs&qejJVLUP=WJK(
zk-p&6H}9V)NBH7z?_h$Fgs8=$=vBh~`));uS<rU{>8iVMYO30&qe;@iF9CQ=VJ0yx
zXZxOYsh5ou9;jl6Z3xGi0A<%4PSFJhtbj6iPtFW_JI(H4*?YQl9h6ftZRg1ZKpxe?
zux++B$)M#{kpS&ybZl55?6Lni-1?E7{qi=u>DuA%f3E(+>ikj%>#=Nzm}@W5Q{Ytg
ziuYLmkks{3eekSTe7AtA6!<Est+XCEbdy{|ReR+voVVV!xb%Wb$vCKN1z|`XjeC@N
zm1OvDHy7nzFm8OXlsiXP#n#F@`@n=C$S>NWnvEhSdjkoRunFWO&0CM6pW9Y|&ClPf
zOZ91bnpkllMb1z#HiZ{p^hoo(r@8foF)&1d`a4Bh=O?}C%p~VT!DuDLe+VFZl}}{*
z&B$5hm0vC{8^~+-lBGq5^JqAZLeQNZCsD*usn{j2Vqa4nc8CL#l@7^`-YIex484I|
znGb5cX`)yGQjRML?P8ou<l_OpFC#7NnRJPwpDyQ560ezy$6F&w@a@j4y|Sw<p^u@A
zfO}ero#EV5-Cy*j5xFMD@{>@_f`awI^DuLYMhK~p+Pl_mt;@N;yHS6M2*2yr$HFNs
z@tV7JB7Y)0JHWI8uzgmbO#JF)&hJpl{?NdO7&?Q-T{5nZspd?Sgdm4+!5s@7YFhg7
zg)(U9z>cUHwWRI*Q9(3w?)zTSrNBBfuU^JwQE^ePqZO1T8_gfx$|3VmL*fcP!cUf?
zUV$MsLmxdK>}j=TzC#ZUn(-1uBR0M%m0dmbO8V_(CpWje3sRQW{f8=15P2TX_C`Mb
zZn>AhWkm~Lda3-eUa1m9fZSCVni!&)_ZjiM*1rAMGht>2(*<PQJKl32t9Hv!GmM9h
zY;F8eU&KmzVHo%7!H`_J*WKvG6J(F>SDBNLX7Bn5Y@@uAAvzW|<%+&&FgrItJ<gd%
zXr5B&u(iPyZVj~M%?IpS(76Qha(nkCbc)jKSPct+PHinMa(eBPUX#~{`(ttwbXTJr
zabM433YBUqi^Tr47Wubg7@(}eue8bo%XCntxy7d~NaCK+^`>-wVck_HYP}qPc8J;9
zkGl&&pQZ{QjKY}Z@cPef({}r2a5W7q#@Ly>@MvN{Ddb;bQU;;aLE~UY7Fx2;Rw_Oj
zx?U!a#g0_C)(|fqyIDG~b)bPZ@)27n6aMt0wv&_U{ArbZfljz_kV4qH?Bbux&t#;z
zHUDD%1C?CLPwE}y+|;&0fUJl)43H$F9pX6ReHwnKkXE7m=`-nD<Bg+*VIcrmNg-!t
zVG7<T^B7VQuW#k;Nl-i|o`=uncssvUNv6*gx%`uzR!4kQL(*A4<`W3gKo+|@vfuOV
z3)T1ka0#)l9J6}9evn|PK~GXWtlncil>~GjXt`R;M}pNaQuIvA!XxUO0@n+R)-N6i
z6pZdwGI29*6AC6$wIwYeO7XHiHgnS1b4Rv(<b7!KLDRaWV=mWpOFA?(A<6!kX@UBD
zQT1--T<L&U@tboh$2Mi?1I^Zt1?%9!jGV}kR}O}$N;TWq!QT*<|H!n1!-V>6@nSsI
z@6WxAZV+%aOFehJZEXxF&FhcTJiZ<tgw=HkgzGx4kp`@OAo2v3>+O!bQoVM}${G^(
zGWCB#pWL*9{tYl_TmQVhSg3ORvZbLD<vZj~Ek5J+%E>y8PWD2G-a%(;)U6+|gWLW!
z0MHls4cLP!dp;<JA5~G3`gJK<J`)hM?5^Ae=mluQ;Iin)+ut26Gd<UfxyB5&6-#mZ
z;L7i)_vckC$-9}6-w(fDwp@Lp&A*a7V(8|zEX=005PbI6fqE#acOgExCW9K-9b8jN
z9-zn|a>)o7K#D4p5mm$+5=dD;F*=@6Wm0rtC_K?-53X5+H#c6bodEyACUi&p>bxzi
z&|gHX@fe!1s>iA__D}Q!$3HO9NA}H!iyugd7Vc!Q4!BhA98WJ!q+!aIV*dKNxcXq`
zdz%~H(>K#-wi18$jOL%s@RJ*oHkct6tb4{S@!)`8EAWuRRrIuF45Q^_$_cN-Bi7sO
z-I>;plX>WWv6VtfZrZ6~!D6L?nSEQ>j>AXIkYXxr!g;tKY-bd?naUX~ghJArL)M0#
znwF^w&Ga*TYmV}ZbuqyE00ypw2>UY2#e*%b_Vx&MR63xg@8MHsT5k;GeSeu-85gKg
zdo=s`6J~{l)rZQI#0jJQ9~SVg`(wTJ4j{MBx9o(=wweS7?W?hyuOP~%2iVcV?mY7B
z_`FQ^$QUEn4zp{fJcc=2TvE?9>TjhLdL<3>7b8hkUA@OLu|Y`2uKhupRd#WVWxZa&
z@)*<XKF-2F0k=%IqH?vjRV?6=D*)mZp4I;BL1M)WeqJi6&VoP{UJ6j`egN_8L@ppa
zd%(0ypx%(UN;=4OK3qF*IG${@xL+j0?^DtcMBd~_H=OGITzPW;7KW=;CNQ(!RZ0Gb
zDYF(`O~0pt3f#k{4LRgt%+gdT0I74;zx<QQr;?KbX13oNgu->_NBv*?S-YyhQmlL$
zo&-Fxqx~k^|MQEVx%25oFsa$~n1EPe?G(kl&~Bwbd|d=tcvh~(VJ!dl*)tsq06GfR
zd#2=E(lgEIR4*=X@3*r(4i9r)-#&EcRP6~4XvM7QsZaCc1tiW2`o_&e={#!}Q`RP5
zYPXurpZOvT(VliSzc*W-fuL@uEZ(n3@$H8HSY(lX`hQ@Aq3}RYlC~eJ%*@tJvLr1M
zv2tf{*P%5XU>L64OWJ0WIM?mC=jk0olf9VCq!)_O%rk-Chsev<_huV#99C{3=6=_X
zeLVH%Gawf}e8{;V?B3W6X-NIho6oL>x&e4n50x!!iwU1x{E7NHvpkcuS#;=Sj^_NW
zO+Dm--5?Y4O7(=PEI)QK4Vok-t=Mv!gI^7F009p^k32mlAKX-FU8jng|FUUZ>{2f~
z{K+HsEShxjm{oNBeVb`Vn5XA5HGw^y?NIvDBUk3oFm*Y@v5Dx**QJi#Z#h7BCHOoO
z_nw$<jktGuYv7LV<jPufxpl@>m7b~00$($iz@gbEcMolAxZ?{3)IgGlTa_3ZrfN$P
z7U1J+!;QE}aG-WUw<}t*1*-Yb)sTcpZMH52ZDx3tt@jd&c57AdhDZyO_tCU$AT(te
z$Cr>2^+r0hDD2H)X-?6)F2o{c;S%gI^UhS@Av~bV4_X#FW$K8xX*h(q^52ha&bme^
zP+H)~vyQ^RBUgqmk!M>#PRoI!O|Wg_buCgUpNFIh$Dw3>cfjZksX~6O@LdGtRBCM^
zIQhyRYG<I+4*F{{=e-|s_hw4U8<LKkI_b(s2$bN`xwi1NUUBNT6ixO-H0*4E$Dx<+
zfk)DQ<jrS5(Y*o~8V-HF%ks5dSF=q(H*)ywm52Zsloekfs-vQ34*)L8X?U&$u4l^w
z|EYARk*$3diT?eNgr%YbyH`o%$}r&OC7e495ZL0amg4UXP6qm1?{B0SSIaw>48x1`
zT=-ab@!UZI?M>G>RpfY3krN~5Iq(=K=a|<@HGNH2Uikv=X+sF${IIJ(k_V*Tuim<y
zZNp|q8dT`U2dbL`Z%mI~B}7(pNX*%tI>i~&yRg@CNs7Sr3#BvL$s^ee_POkCc-}b;
zvbjnxr(ftKq^?kJ?3BS>T=;(ImcyoE3xwgVxUrv8v9!5^OTmvE;9K=wbATplu5gNU
zs|<3e=`^4Kclb;Fk<;n7KouW9blY01hVjD4<sfZj&TnGg!oeXnmgp!&TPW}tYb-qs
zsNg#kK3n<J)^9%I?bPfC{f<zwPfmG?c+q9<Aix{3^?Ha0S^l7wlRFEi%%_{mzJ%!n
z_%b6Hp|HGvAOC;M0`05yxD+0h0W1)o@zf<cgKv;+a`tB&Q?;$z1Q~3s_Q;>wr4>sB
z=%v?QHltxD`{4SM7ArN-4nO-tJRJXJ@i>InGm#U-C6M?m78r9%Qb@`j)}>@?;=QEM
z)>|6493rQKuJ`Y!Cw9ot&UGw4`2YO5|1(F*+s#9b^{}08?HLpJ+wf&IG+{9m<2z`x
zU*5AiY~lPUG<oVh3@YBeiJUv*F3__B+(Cza+E@Bs#5OtoGq%%Z57!~&yZ76m)t*Jw
zC5-Ckv^Q9)0zicMU@6oGB>awqyHjrJYLCZs_aD&O7-g<tI;!&PAIHagFHjM7asO1&
znC5f*^D?_)QTM<tqXvAp^D6dhKqlXXIPiP-7hiWRpa$2(hx;j49dhIfNxEqeD*(Y&
zpza8s9GBjhgeZw$^X`1!&>x_|vYL76U?TrMDveNBbuKihmml6gupuwC6{MRg41}B<
zklPw``scK2dW*RvuCU8VrU6dL7Jg%YFD)Crs=|k-%0s=6T<HXIfzH*TCB_r$b=D((
z%jnCY8U!LkMYKevMt2|b;-dRsypx*`Ida9OhUxTx7x+S|oB1gYO!fBro)X%pYbd}u
zBIWp==Yskpp0eBjO@y<(=-`mhxoaZt{`mb6vn0E^&Ev&|T@TORfqM+40|Vi_E^nxN
zdke-utIyCz>H<syO+onGFyL}pc~57|xlOVzg|ge8Iptw52MFNJ1Q{_wpM7?Laeec)
z9`l0b{_kV=-(NWfzV+u(68#QhuCh3%V+#Vs&YH;oKt2d?h1ZSU^9*zdiBuCO0Wdir
z7wf7j2Hu{Tw6;6n5(|Hvb`r??nhm;f79tg{i0nm{%dO{&FhJ*>3HZO=-({8gP#axq
zpkI~j{+m$8lbM~Cy97j_P`OX~<lf@#?=PfxxRT+UJ2-CTxWRtqB{!G*(Y@bhodx%p
zH_JxPwgRW44jNi@)A1Ha)(g2R{`q*i$-&<wd*q6K7WlD?+_8H*9Bf2~OX@whfo!Lv
zp7kgFouNBUtkwYQXce_jY8n|A_Ko9GK4SoFHt-^VKcwP6=~`!8aYs4t%Tew7OfQA#
zHSEypo%Xw~UDW+628;zr%!rQ0)+1L`xOg7^jsz5<n%piWe7Wxvy%lq(!3}YI#eY0b
zrw}y1%YP!6WcSaUlu1g~r@S7X5%%u%4qp5=`>HXz;g6S*Ka+?Z^Wpi(rAKUGbx#t`
zQP)?wZ-sK6Z{&FgcwkGJ*U{gz5NdND$%Jy-{i^vVPslZZ1*DlQv+jtCHIvfMx?RQU
z`upuE6WL=IgK(BzlU~3#P^QrBzbLW*umUX>U<pZKQvDlx4!2q7ew~UE%1kEu+mp}D
z+q;fvp1Lof9B+iUz6GM@YI7J_jY20vfimGykp^;m+J8#J{azq&vG112PPkV=C<VZ5
z<O(u?MM{4i$FUx7ucr&>u^#fWd5`>_YoYeWJMF{U)8c1#rMGWxxFi_-bT)zhB1<h?
zl<pbn7E+RkoQNKHXt)aauMIa{%|bpU>(LAfpan%4sKc?h`UifsDT-p$R_|W;3Donk
z`>8O2HPij{`PY;Pz*GOFRK2G(RPdKaqi-D=oj%{P_0JYeWE>3b-nDooEt3>vUx@(^
zT0{nCsPp*UaQd$NJg+fLY9~JWoL%B%Mf#z6yno6k2e!h-2<1Y+y@Y?@FEt-IvSD5Y
zSc&XN@;9YJm0!4~A2zF2NdBq2BT2VEqM*UgkuG%PipI#rldiK-RvVwSo3>S4&%#TU
z*Zz^qej5S2k5G2G*zn5>w3pEw6Ot#`7dlNj^V9ENet&sEGa#DSXVSwcc#Nm;&pR9*
z8nOHhZnnqEI^0?JXSZ7bjN`q+GY?HudIh!W){`kOQb@mJB(o284fs&%;y#OYuW~PI
zU%2Jo!On>|%y`Q1R<0eDcO3mE-P<a-^=6*-5ATWz*K>8b7n6>?e6PrLR~Ptfv!%}y
z9wm9zku@?X0YuX7*F+#<>q4mctkv=RbNzQkF<Gzm7l3bD_21`un0DKYO{D1b2jHJ#
zGyb0s_3p6}hxzd#&}N~98spNwW0K0@^}moO^pyT}wUm*sx&W>ge`&Vw|DgmvbQM;C
ze%)EV&l1c}8QrYxjZeA+s>7i-pi)AO-e&8p9lZ%INQQVk*IzUa5jOUS|2S^I#L`om
z_?(8<S1uxeP_$4%prOx@ai*H#asN(slFqGu-D5E0j%sJG)E3_UrOa~G_Jh&twF{IB
z;1Uz;@9|<F5+$p59!l0R-HMq<02j!5;FZJy(Y3&l5-5v;;@`O4f4WuaS)uL1;4U*8
zWd<(`o#_)Cy`w@Y^jz_^_=73Ieu=y;xwF>vUWF?pHM%%hnlpd0G5gy5+H?SJK<)tl
z+fumJR)3XI`LMrmYl)Snm$`d@GX?~iWTMZdb0gdG88%v}M`vFW_jY%j{Z^M!i#?xP
zGI!$N)WrH;gS2gy=S6Dbq>;`OhH7Rco_X0Bw%NaJA080<als}@v)x%ob(Bz*gLd!Z
zYa9t`^_sNRxo7VJ3e#8_1>xo@eEZRrd2%tADl{`_kCMYJ#-zN(OLX5m>FozBDYcNN
zR8pXcmFO(A?A>H$#<J2Bjq)?aL4%_P#a!KK#S<NT8D=|oTgKM&Jcfu=-|M9NO;@kA
zs8#=c`th1>Xi4MFBcyA-G`t(Xxs)v`m|}%)S$nz2Z6bfFh1{*a^~+|cUX|qOy%KP1
zLeUs%YIc5&91ZW<i{Li|o3;?R6Xjh=DP&&4Vu6`ipG^#A<h(scAVzDb{So>`F8H$h
zC?gp~H_l2PlI|y6a<DgnC2`yEgJl~lhnv-U+~qy&kYi@>71$S~-Eo1>`}xt5kRqd{
z)`+dZ4=t!k{}_s4hlcGaw6I8JP2{sNtNl)Hw6i=c^x2f)M`FAb^lfC4VtDLsAYUU|
z#cGy1KHYBjy+P5qP<3StYSUo9;=#uB3k{rQVJm|hE)}3^yW0ed_m~etql^3$OB-k}
z^O*80)r~|{2|lx((cCiJQ<6qDy_p4uM&`@~2)fE|qlJW5c2{N!@^(ADmi*0!_738T
z)<`Rso)V3o;#u5b*yawFRi+~B+G;9sF99PT#R$4FxJ5E@*N-^CT&UW^%roDb6mpKZ
z(SFB9c6E>J3JJe5t%=O|@x>D`=cpB$6<zmlY(--Go6A~u^j(=2H{eG{k)yg$0b&V1
z*Aqf{5MS_mF>ZVlq}I;6HU^md`HTG&6yng!8F9e=$}o^0V*anA=DEW?j5#Cnk@T^A
z-md9ib|w=r&;8&teloSf{G1DbaJdb<&jK5s_Y%rrv;k1lGMtFRdDQy`<I`KuHE-z6
z4NTf*yvf5z`HOpAYH2K!A_sQpM;cYnZ?FpR$|p{M*6?vPF{AD3o}d)=!sNEx_S%`A
zCN_}(=M~76N-zgO(1ss}-1oSSI6N>j?Pf!mrbKNdkQ%p1#qZI{$#w%#MgM+Tm1&bn
ztCG{+S(qw1zhT}u-F4JKb&!<27U`^b2^@`a{$V4rU0B)gtHpH2UA|O8WN5Ws8N#Ns
z$wB@0TtlRT5Pgwd%33vh@E!2XeyuISDBzyRmIC-nz$~j5@DG^~*I%PfA4bMw&!bag
z;<&7KNa#Peu-o!=dPR~T-^KdquZPWehPB-H0{5YHvg^Gsyy|%#S6*F#C5C;5K1cQm
zgl7bV4~nPw3&iw=8`|n7-GmY)6Fp+aC0B=H3fOiOZ02{K;We(`$ZD||tjfyV^Ogt|
zt~GSIL3#H)LbuW6mI^B~X>?&}w!UPmY2n0MCqfKh2?F$}yNM?jEBa(36Dq@$h<Tkn
z)M@aPRna+j1eAb$sNiPn01MdP=^??f)KfqXLpyo5@*%TMd55%2R8{cUx5F=8dQwGp
zXjqTtXHhz`V4X2e($t6LmYyr+^jt)Kj`?^@6ly&RZn%ch_%)y9`i{~&A}QJheY-2&
z+2=UK@(`SAjQDAuTxs|M335;F9ST6rvODY5<C^g<?3`ql0VsOUPEjSehj9gveHGJ!
zAllu4YX?k9W+ScuUanZTw5@GhRWw*-v+*J;IL*LiZySS&Q==!a-<p}X>{Shus(s4u
z?S-Wu?LMPI*p2$;Ztn-LAdi{HxzI4x4Ry=SS++O!6OmaPwb)nbTNxEdh=^k;=Zh>G
zwI3d7H9J{vT7sWkS8VhW8b-WtGcVSG&bljIx*l~_6Z?vgg}<ahp3ya7e+Ub`D7SLG
zf7@!Cr0fs>1%7z&1=#zbua~=yG0<ftQZ9Q;aMUwhXZmvE*z0`19Vh#OSh;LFX(2<?
zqly2Z<~VA$cOF@L7v=kh028s-kHk@KsoF#fk==Yw&tZTXpI{tfGPAo<;?ZsxlY`#7
z$)K-9hbtJWg&51vdv|5;d#(L)gxCYlmlbOM7Am882VW2b`p3AidJk&tGA?_8zL?@6
zjDLxW>umI0(I>xZ2>=^{C^XxFJF4kPlce#e{&?~3v0#;@!~+r#{tGhSiQbOd@m@eM
zBxZye{G*#?58a=^0P|h`{saWho9%S(M{L7ae5De9;ny*S6|$hxk<3gQ(Bfof%fu0_
zN*!Yf%TE?dXf6RhFRWpv85Mu<Gcxx8KIfS2u!rzfwD&dD5V#8C#^fEHWFX^CS-dO-
zXUHKGf*%2Q&MbE7SO@SrO2J`_y9zKSmV>w{JIW}Io`^{Q_SEb24<{EL{`eMfTK!NC
zkf_mL7y{0`zE!RCIx0<ya&>(J`WU=~-(3-79wByO?Bw4r&Bc~f?8ehg_t&@SK}2fO
z)Nn#40G^=@IGwRGFyr6Js?#9Z*S2@KQj`ii!S!~6mhYM=B`PLvJ7!Ck?n7GfNPDvk
zgG7A1@F{;4<fiG5I9yZp&9qzl^0|=f6(B$UmKSe+g*Edt{9fM62<I|go}am^({D@6
zTWzmmj|evluM)j;vM<rd56T7)MPh$<%-?pSN}|t~u^uvbySmTV;S=i+k7>g4k1!aP
zamvo4nO}p#&NoZ~%ap4C{}V{9bSnd<+0y3AD!sTeHa(ie*wAP(+)<LSwz-6wAaH2L
zwYV=&*lMx793m8&)28#e?rt^7K9*%zE}A4Zrc0AQnTmnTDxZS<Us}Vk&{$CdNs)0C
z>+8lVpU(D9KvDm9TQ;aJP`bO0p+o7!w~!x+(`d=w6zrO>4iVDo04Pu36IjoD=!}2s
zQ=H(b;a;yexR_frm=~%wpUUDA<tDUGd(f-yuN&(&vlWy8*&E+O1(O>W9^^XO>ak_*
z2OMcQF&;OxsWJg9N;9aI0oXE*XDGPg<l$dU_VS(te4W1Msq0`RwTu%-XXP8Hd<mlt
z6uoyt?#<_?wR%qVOdjsZGPT#~ORgAR!@hGX3obEy(GE(3Y?(gv6QIS=nbA`5b)gV7
zw8J2{=VhS8kOm#5MG@3zy^vkpy$3JfC?;lB_(0w-F|Yd182&R3z5>zd-VF@*((kXa
z8(wa2rQF_lG|Tn`Pf+bFO(WAC&TYlRj#V;2r6k5PNfik@`8HfUdQvGj75k(jN^6s<
zFvU^3r4`)I-z~*(YR|iwbu=l$LkTlx`RuhKtTTAfe}5!lX2&w38PmY%E6vdFdo6#A
z_U?yS@L*`6Es&96Z_c-2NlC8tkEy^&O<?X|gW4R>B9Nn;CV|AdU{{p~Nm!h?_xfyG
zp{vCE#rpf_2;|JEm?wn2_&QT;Ozi|IRPnKAXWg`+3y`GL>Uyf^nDiCKwSUb$Wj5AU
zi!z@|M?Xn5Fozzw?>Jc$Jmu|PFT&|x*%%0z0ywH7OjZ>y5Xsh6N_CQLGC~_3symq+
z+?*0qle#*8tSTsR?QXF?H~Z4y)~F$~v8W!eVuafFz0+NMYd$vmaK8Zpf392V+{f#o
zELL_#xNowJzM@a5Myz%&knlLu?^i<dl5|bkuQOO~CKhl^8Kx_~IX!it^(bE<f$_W@
z*pFj+vt@Xf3Fz6`I8R^InH4YGser*Ik<$|wbz`bqeLp8f+*E6+M5=~6btuB6Uu-4L
zwkW0jjqXUL_fj)wZ8yDHjFi=OL&EdKa8@4ZSKQ&Z0j`U|6i^1mNzrDsQBM!f1%&tE
z6Zp$dB-sl(Ff#+{OzDs$m`bGA7YbE@J+aNYhH;(AkB`igfc9ub3lD9F-PaB`1vQ6V
z6E2+>;P|*Da8IW{YD$?>r4nDxnA+`2tXQVMx7{~+9>GQiR0vF)dP8vcGrR6Kt5?i5
zNeW5;>_TnR-C42qp}l;X&*TFys)9S#97@Hby)OGsago<Qq_Y-|x<{PES_j{zSG)Fl
z5#QgH|KPHf!IiE!m2(;0Z$rXVVHEF*BaIxKu93R^nZ%Wde{PGOSl+?<NBNA$WufJg
zcs3e|5QIWy@RZlWuzf>lH+V3y4$91F9lA}EENV+rUKas%i@Ib*TU~xl(7_LJNElaC
zR0WGUAS6gU`q;;QIQ+s|yiI_)Q*V8O?Q_AemAJLT3JzE_erGMW4y}R~uE#afrqtZ!
z5HN02t(?cWzZ<R2!foGG+AMCL8e|-O(VDe^2CGAKwP=4_MO*CxqS}=<Lx{pe{4fjp
z-b^09jf4yvmhStsG7<c2M6)j+`DyUWkLUyOo8J@Kyv5>OCbz}<TwgC<#SSG3>WZ0n
zgO1qwItWrN?IK9ONsrUykCilx30kbF>=!JSW^-=nU|;>{Fe9<aAd;(L(UY)*dOUGR
zNQcRs5g3|D#-!R9omQGM^h6xJ5?nDH?4fqQM3u(eZOg3-DN@76(XbP!&EAP7tERdg
zO&>r?&CgTvq|DB^$Cp=Ax~n<Q=aH+Si_${j?|$=tmfXVA(~Q$f^rlJ00~Q*mQ8FV!
z>;eP9<0ll839Z<7qnq3DF$$F<(P5GwQdRGu{8>dWCDgku@4&w-Vto`Rv{?X+;qnm2
zOjHbw;b7jPH~@2-lFRgyFMNv8lD!`KofClBE{%QCdw?ryX#1h`3VxR74j2whcY}Fu
zj_dj+wd<{=;Zos_vyu~?CV@{gs<$t+0$Oa_e-^aXY6;^%qM#NkPBY^)#e^ICC3OSA
z@+l^M>E0Y22EO$|^5KdXo+}>Z$0*O>971;rA>ns_?UoA@P7Hr-xihr3OaJ1Nyzc}z
z5!{Lmo`ELXZ{n0(oo?QElu%o)2^svv9t`zQDzf#z+0uR(voACCyTd20(9T$ZoG<TS
zw=zLx8Pj%_yPLRrlXuD#UBprqRf-JHIdC9CYaW<cfOS~yPWSN24PBa^YX{!pP!jL*
zTGFg7KH*(QK=ET-e;S^5$C|@t5rgHqLw&HwaZ+z`nD6JB)f3pxIK?ev!A*YPkQ)oN
z;?^tzLHa75Y$=1v(%?$CFUt0hX9wZ}nhm=wjI<E3mF(jbej)myB|B}+#r`g=FgMS*
z)}zW~6{Sh(3t-+gtS0dDP-=K$>P%;vR(VU!&6h8>`~yyjFr$A#TB5hhuaPidwYd`V
zx9!i-hyB#9Xctum>-!Gi=MGN(+7$;8wF0I!hWJ<Cn7Rli@CLF7wRKrFv(hJv1+}=B
z^XH(i3Xt#=4PGHnZ1Qzs+L~4iNtKc(?(YPpC(?Tncx0U|MR6(D%JUT*X<I$Sdgj<L
z%klzTK$)0Q9}J%9rWRR|Z*OGd$I0z9XT67ut|+3aXKD<*Jry@A<8SV_f}gw?Q}~8x
zd0*7?QmYqNyk08xAC2(3s(}RSX32_j<~j8i2MfoioY$lF+OLLZw#fXg+}Ul~?zotx
zEQ$5*vH7d_S<=IS^nJJV2!)3@-4A6;=Y08rtcu=x5ch{0{PRQ3v#7MeZ+tNlnQa9}
zGFnxOn$$(T#P*Op-zGomerdB99xSJ3yfP2Zu6=DlOhg^NCR8q+B;eTK`=n2E<nk+7
z%efGpYNej3ucuy}_p_GH<#-t9SeK=TVmoG}pp#V^!4BGY=g*LBMF?i($%H2Ge=x|$
zbX!U2Yu#MEa>m#3lAoPPVW*9ynt!R9pnu>G6Z4s07mM|kDrTHJ6VhIhS9dd`p)-r6
zCZiPbc2*JUeE5QfkbdjBI!_w;3uD&dgh!0QIH&nGlV<3N9iZx4arpdR*JxlnzNYN4
zWq1RA!?`X55z3l+7HX3=WXjDOSv7^Kbd;wdLo%x=$hPLOI*j5^3CY>^54}{Cx^NOI
zyyVAHz_hfv?(u<P1!?_L@yHPM&wjVX;V@~WiQnM4v38gKFTLE8tf+e4up%$C9+M!B
zL>$v|C8;E#16<ql9L;S2mUE*ae>MTv9l~@agA~<%O8@>eI|)AVU;Rp(=f|!zt#2s!
zV5z*KxpB2UReJB9Irm?R`~LZ^{9xn|<G5=EPieT{sX1NY|6PxifUZ3!0rrbcB4iMo
z)cr$eBYWE4T-GKR^BeY=79!-JN;p~Hh1w3EzbA%Dn3o(sULovq;$`K_rj91q#k<g;
zh|+GW`mAYp7P%ffS1#Alc;6hY$cTuVq)#<<_`>hN&6dP)&=+dY3fD|*)|n;t4MFoM
zde=przT2!5d7ll2@lTHH(tqqj?GM<^g}JqSn$j0CO`5jY`znfZYuE(c-d$)<EPIR^
z`n*rtlTTTx;e7#YJ{U0$eJ(XzFs`wCi9dBLRGpQ$j`IJ7ffad6EXl9F?u#UVyJC6b
z-b-98)^@S-^W%zBUJXUHqLt%jqQtGBEK|TM^as$CS`MygOm{!+v+j0nk+>_o2Ud9S
zH1mI)nZuHmOw@2N_>*wC^pfhs_=m3zXVP%}0g4e`r_bat{Ga2GN?U@wvX%Pp5wp%&
z1#Ow0w|BPZvaj(Sp9-F^pSMHDo-M^#evx?X|HTFZwWYLbQ>O#_7mp2MFWykwWe@R9
z_fkDk&Zq45YmccsM6_vN#XI1B6M@{^@8{z%O6LT_&!+}d9{923aGMKvR?dwJB>t4b
z#|>B$j$Oa;wb!Xb0(>lY{)Qa*`M3SIdzx+$Oo~w|X$zMYx46vaw$uhW{;h~026Bey
z(AMJoUjp)Rn&LNDOz~9iVz{{z_2xp0*}!mn8_zg<@K)5BZKdiuuKopAS4LtXc-Opg
zt$`w*aSj73*zz2qY~5r~=`)(71}8E)S0|=Oic+f+<G$l#ovHc_S?H_^hNFQHL(yUj
zgFI;KT~B{*%Ew0W>LBu?6vdG-)+FhknJ10fSmX?z@3^Q}A*524#NTyEi=TQ{9lG<E
z&{%b+wr(^fdp#|Al!+Svf=VFFQg9$>-TNp%*{XT?C8jx%9FJ(Ln|kjRS;oMt;2C3K
zcTsZ34Q|X|>ecb%?~XsBvgIPoV+RzhQGLcqqi)bY_T<muP`%VejIUYD7+|>}-i;O@
zJEW3=N!KPlUy;71^f?cpN#(u+JDsmRK;`{gZ3L2jRKFb>$EQ7;h4WXo?gb2%@T0^6
zRX7S3WEDR=PK68;ferihHO)GO^@_p#4^<qmhH3!6&A}aa#mV{EZ5T9%w`gbZVJcXE
zu5ay9zF|{Z$15vHS?eKFc>bqdP^L*&M&Z(|Zc0mRn+F0&1S$CxTE++#mA`$(pk)3f
zFaC^Ar2oRX+VXh++AL)iP*T-xb>FgF-8aN&zL=$<T)pf4oV12?^2rueb^iSazc^oM
z&D{Ze2`C1UpHF>ZkR#lYBgo<8C1IRGQ|Ptj7OQk555!Y<@pVY=U_|IdSb-_NXWegE
zJUj2SWq5M|rrQk3z?NN75iptD`+RpM#Lt|aqc54cN3;oH>eG{c_F;s(x^vzYX1F7_
zr@}$)^Sx&i1S=p{f&d$y`!##s@pkP9#-|mZY;9P*m(IXh9IduU(%NQKG-7V?=A1$%
z?-Wl(^wtDXR?H^3un(iPNc_++AJx1QrnpI6_5@%G<%+WolhnOS<Ff^{8*o*T`1M6B
zTxPJdv90DehU#Z-%Sh~NRM|0%96^HCdr<Ie2UaOC{TRTHAlZ@n>M|ZO0%E6LRG314
zd+LhKDjOGai9O#>4X)YSD3pCv+5m81zmQ<v*5WHx^{Y2?Xx}!NE<h?$ieF^<rQWIe
z>LI3YqkW7*6>h#arT_>j_)*zXpS0(<WhCbc2vl57JAi;piurSsBscA^e$89kUkMBC
zQ`S2s86Z=*^}Hpm_UkiEj98VesVDya_$J>wjPiBE6q$ZFzEq$HFg#B9Us72>%HWbo
z*9yK(y;AH0OpYGh|CMaqf?bHi+iTd1nii;H`Wgh)<J@x_UnLHWQ(j0KPS5&S<ZBZU
zBS4K+{g<c9_vsd}m4J@bhnOgiq+@@IOj3Qmh%_s_t2R|%`?$ZPaPFs@@gIBE&&pkj
z&&^<ddCMG5Q=y`9@4HdH$7xa4Tc1)&(o8?~CP0-q(XN|H>~WJvdqH@8kw%Be$}FI8
z{|!byK%f_2`!#H%zT(bX^4aRd9C)%|<BQ&`QdF812DXB47ml+<amA4?O?N*>E$a=x
zt+S$CU04yTaKhh&IvDNoY4U&RP#h3Ps8-Tv@)>S}Z?^|?`@CCdH><oh@?Vpw<S(S&
zfcyRbDYh^IrtedX#z}o8;KV_-YpJi8HA@)l;iRCKb7w48LtaQVAD3n7Cl=MP-ToNy
z^E`u?Cxb)FpRM^OYgU@ncndSK7tpKNluW;wud?kXP~QIyv=g;&<{yJj$uSU8OWQj)
z;kAGUIafB9bk?j@OYftTmn4efhCTsaJ(}Dy1pIanq0)bUxGK#$0Uaei35BEPO@Qgt
zNbso*b3fp+51<s?ofQmG1p2@iWu8|kACdo_Z(CoI_%K06`TqG^y?$D^>o?<DLO=A`
z+x%N_OLqxPS;ZxTvVWnBxGmEbV=1Nf+*!R_Y2uTryTyG0Cf4-K!DIvV?p3%Xr{PpW
z%j{Bc3rDM+zgNW~`C!GI*_*GTedWngpn$Y}>NNKarW51tAY&<xtdGLp2`gSs1?~q*
z*K9v1!typ!!^JROC^2Ma86-U?R51SmOWN)6ID62$`or}J)vA$`g`#|fl+YNbxvmzE
zc9qYLGq?!WG)C(tcwKfu7G6qe7SCrMbFNIAdQ*v{uuSV0k6wvyxwDME<~j|;XT5+F
z=Gptmg0=El4on<0?Xa6=t&LyabN1+IU85ArWS&snUdCgb_4@x}@4NrmZvTI|wR9QX
zwW+GwwQ5tewW+E-TdM?7dnKXT>azEywPWu+Zf$L`ViRfvK`KE=B+gs!`}_HxbN+$z
zIFEDk8+kvHE7$vVy<XS*bv?(v1`L(;T2YkA*WevY%FN}i#Okc~9T%AN$JKZ_2B1=N
zcbn>4YdUWJq{#?iSKvJ=J$Uw7llf@?#fpS7+!#PrP7giw?n7}iHUSbQN1y0e$Ls`}
z)<oWQ+v74}brykxINYoC!4gn=%|Ia7gZ*zgyB#~p)!V-v$6HvrSY0o)UZ$T%{OD85
zp%x`^$7c`46l48t6yku^Qx)qd=F)(=d)v@Ui&VO1@Kb!UpHzJCOtC9e11QJEuRXXH
z{pZn%*3#H#ea4jj_O>6Vk1XZU%fqVRK1<Smd;7I14ztO;B)W$)qiqiIgTIIhBuey(
z<6XVa_7Ca<E=)8eyH%D8CA!m;`my-32O$r5jZ0gMn%}ohbhi1n8LmTl|FF&MnL6FS
z>G3Pm&8z3O$C<RJp!l~;a<Lod!R@08I}kd3gdYEC`E`8vDI^%Q5HtglrEgeR^LKBc
z%MK=^T}cj%L<i2P+J6$@rsWXv(fKWK_(J{lI+(F?PD81yeDHpU0@M8i1z~J$uG3R1
zbP|U;J>jZUZDWAtMsk~UfTgh&^K6pdX>ZC3eE<)B;0MS_XJ!u{e)B)@OR`^WXy8m|
zb8@musQI=lV?4^|qj+fHJ@{AuW2CRhBlh+sqBMSIH5%k@{JS5EF;X0Y&L`biRkmqn
z7b)Si)MwTvI!Q%Lr;-4r#^6;zX#?`M!b_jzOuhX`#Ts8^nI+#a6ZLfCC_EBj?)Jnz
zm_LAC+u!DwkUSuPDB9a16tGz23Ljvb*EzJ71YKzC(*Vcm74p1tW>G&OK?0r58o<m9
zPF8;}q+YNbF7&RT^^v<x(3HDgqHfje?8BzFWcentGQTBu&K8sHU3JIOgxA?`Gxmw7
zC?d@cCj^kZ(OTFfFX)R&KK2jJ*XZ|Jv{`Guf_9$y>*S9NTpYMNS`dfBzcRJknMDk=
zTukvvx;(**XZxn<WW(}Z39fL>?^5#{5fy14t#znKKHJSRP`N`nJEReoiH<ERpfSLD
z@%Hkggld1;^ih3_gkHuv!$EZ^28L_3bWIDNoeXpPL3c;xDoLnkj=g%dt2U?*42;F?
zwXLj}^W<`eqm^5>zG+bW+P6$v^OtK+9wS_Ih_+VS7O^IK9nPZ+Yb(1KXKs-Cz8dG#
zk4tZES4B&lv#Nz?KIaD;mA@$WsVL9DGBw#~syjDk>goaZ1J%7YFSegplO3Uatb0Ag
zTq5!{35mQBgv&Cs;=@8rj55qymlj0XOAH-;`do_1FQHt4sh6+T{0=OboRRjAFAH4S
z;7LB{hr*V!=lssb4c4yP{HePIEyPFgqqfq*?aA$+X5os&*h{t?HO;ND`OfSc%+DvU
z7viCR8={fiqU^ftD;v$pV(l%<7v$EeJ+qdI(nzLCx{Cpe&QqV%*kkLSB#gA?3i;OS
zYNBLWx;M{ttlVpI3|;-|B7gI1B+7%5mXaZ5daI|x#HSB~F;db?SR*%y?yvs)SbX*%
z3?yJz8#$fYva<m<_X+mn<o-Pe8UQKXh3%VX1+CO|sg#7%L&%#9#+<WFNm7j8S6w6c
zC{=!mfDUMvOUGSUgv49F-@Mx)b0@3T&0l?Jg-g?lU(`mG>l02r4lr%b9%3M^X7}G)
z8v}uYub#^_WD3>0o3qy2^yN5tXec*Ujcc1}SwFIR+9N2P2=s4WbekGlYqFeR5cd2m
zb3NMacNn93ai3)}@VFLaPZLYj9=~nvYP!w(z9xRxXx}1oG%3K$XT-1R8b_QN!=u&n
z>Q=U*uM<#*XCJJ!kxRnqo^T{tv@1XahNL~#<ey-k^rvx_QYXJa@()mqN9a{=dn<?6
zh%s(od$l2ZV0Yuqh7jRcGP_CXwUReu^2S~i0`f2~qo51iS3Co*^=2u+;_sDY$)!6g
zRYChrq?03?<T4MxJqo<K3}$4=BGFzh2^CTF$dI1BZJ4ymc#GEZWbD=FFkFsPhj>@3
z%?U|6XWJmq?%T{aVMOTqb)GdFbXIw9b(Fnx4rC}m+EtV3<=PMJ#&QB-NT9*=G$%e8
z2)J1jI<fm@MfkI6B}A+^zP|6YF=?8odz_(2cwBs6g$C}XRC?@+Gcj#mqE{_;dqg&N
z{|FFZ)}D_tp!XVsj}T{)6sd)D$v=1IUN(lJ#<ZM%G-l0b9n=;byxud{yNG2S18uX7
z8z1dMgIfU89O0`Uf@A#HNvJlf+`1Y5a-|WVzmKnj>F!%-%0I=a<Q*fGTr?0JVprQ)
zuzZ^?UhKHhk#Uu>;Q0}7uWVViosuAdxvi-Ywpu-j^uD4A^V^$#*<6O(_d~Ygad8Kf
zAzSqkDf!aM?^Hg=?@1_evRb#$$&lRr_HvUW8Ir%vW<Tu3vh`zCQm0gcn2+MuW3%pL
zw->O<odQ6id?I-=SckG%t3>Dlhtrsnzbw+xmC|WA)p<gNi@seT+M&;?^gN5@Yp2V*
z>{RkKMY8uf{^vu5;r0IS2lTpjI%_Dlp4ZQ8_030v%Tb+E(2Zhj?_)kp5pAp90j@b}
z+b1?5k+}2DrmPTw?I9m-9Jt!OK1KsV=zyxnK1=_EW=$j@86DM7)yJ$+E12}N_VmE;
z9y!Rzx|_((u>1L>-&o<Y#PS~N(e4sRb^=v8NqZ7h@GaWdT_^#XrM)gT5o<D_{YBmW
zxs4Zj{PD4V9T4Ky;1e*E4*T#%>;<K?`bdJk&t|>=pawY*Q(v>_1)h%RHA)F?;|ZT=
z<ZQEwF9H>}%=_OlJh^bJ7anb=m9@YeU&&{&QCR~>O~IX+jT}xW+qqqFlZWpftWn5Y
zp(y~ttSZ0H*^_=Iz)O4e<Hk<fL(uuR&48yvSUq5_>DIL~e~rz~0Q!9OCB?W(uYLKw
z;@*>*QF^$K;d&2NB%kLQQPJW@F|1x{jNGhV<vXs2T3ZN2dsg_Wy8^^G@99bbuXzR`
za)A}Bk*?>jLtNM>Aj^>^n(y$S`BG2O>NgfTx-xj6F1y!qK>@o%8K4-RsA~@8;VEOz
zn-CHsvjYS|uTMjMj~$;|O+4LKC#75N2Tbi_gDlwm*^qP(5|ju@Z)U$mhS!(%%uoAF
zpG2?q8&{K|AJ<n;hVZ&LySW?ki66By_w;6d9d|B&o(3}g%fDx7HxCX6ElG0g5KHOr
z2{TE0tx5HplNI$q8fyE1IM~unNw@j90Cy^jxH2DhXH^4Lkp`2``)4i1m=q+V8gnAm
z$=E-H!bvo?zT^OO|AaD<b|Ob(3fLT?z<J9~8QodEZ$-Q^J!7$ZoAfAXpvQgwE-Sa<
zrb){*q!0058xiY%=MKm+B{+7X*gYtR_)~M*$fvh#b2h4|jXV|MFWb1tl{-4(C_^-Z
zxT8e0EUem;oZ<L3H=P7R*+DrnzKb5-%KX$V(D%kl`1O<qMPUHqgqhk5ua8<XnDbp3
z;3`VZ47I!aZnG(?aAxfwI&$zt>3%ozH9&=CcQ2HCo?$|2l8eNo&pY{J)-v&9kprk(
zN$*UEt)0eA7|CFHr#7zvd~{fnR?>RQbC?Mtzt1Ltk2y(~IFy<F3|Oo`D!<<b7AqEI
zF0ZePdR|Xx0;eAEj4Srz>53ViPwp&zlNrR30)5z=u;i5sN20Ih;2IyH9_((m8FqDY
z<=x$5VmsM2&y{_aoREZ%q{_go9lHz4geF|7ejc%9(zhX`^$qgvrm{Q6L{c$;zqza;
z##!IUV#B}Zeq-H{PflW;>^s40>C^zZbu;}62A1{kd-3|7@wFP)R_)13U-!)?9xuu*
zUudvV0z`{faP0)Z%ff4MOEP6B8f4flbs}=oz;L29Y2@o(>}ih4qbYVD@XN~8cTE=d
zl6bCeSG13}RE;u#XPRsk{mRrT<zZ~bx6H?^&y3UeZCUs&jP|8BvlY=wHXilPzBf|e
zGo4=#pQG|W!ye}@^Hn9w9{_{lx|<~HgJ~>tA_7LBr-U|s)JP(M6d19azKoSY|FUx>
zSKqtZOJ5H%%qV}yu1*VVxl28}T;}cLA;rB4vXje(8SVmEKWX*?L~|IBBO?+PdMxba
z>Y=4WWbrhPr~XJgD)Ae!oH)7Jg19IbHnykR1C-eU1HKtHo9<I>T*Ix*jJeotz6~zb
z$s0%qi~Qq8_ou4mFxc#`iseEFTf+?}Skdz!;Nu}D@m_|~>i8W(NQ&RLbzRYZ7T7{(
zS^HOq;$H{}k(U;1SKf|06gAL~Q};RTuyzV;Ye!o5TOdJ&3FTp!&Cd-fEvqx$=Ju(X
zhP{Zpuy?hr0s9_Ac?(KVX?c0iqjDM?pL&(!LML|N3Dj?(=|HA5u&X_*DJ!jc!<})T
zYVaRJm7Usj+Bs=o1;wm+@8&ih(pp=?Bc^AVbC0YOdfS`uYVF-LBB~bi+K?+nz(T-i
z06fpRU6`L+M9`7MSzR+fw_-&0PCx!xu6f}5h?Fx$%h0bbX|vjLzVYolZtF?cW%sO3
zfLbI^V1Df51wP^zrYX~{S8JPgvz};v4k6D+zBV&KnF+Qy7RV>A8v@-gh`IqvvX~}n
zv?^h85-unlh~YEog>eIwsN?}xHm7APeTm;do!8@!1Nk#0NFG#!TBBDvPkDe*5kj!0
z%Y9<YWS^i$kD2<#RZ|4_;d&78)QoU8!_9%H)WbX(d{*MgQ}+#m=(|(*lqH6ZiWI(^
zdpD*!UD$iq{+L|fE9C<0!uKW`2HK}$maSR$OufL59bp_X*tlgE%Y={JqRR!GiWzsw
zk|j!I7?aX8jG(wDIa5c`CP~eypOchJp#Z)p>~(UO&3bhh^p3V~=+--H==u@hMb3KD
z8=~hW&z=zjfF(7uhIsdGMSKe2h0uM*&fnVbJpV3$(v>K5{NOH6Z%2|=Tywi(+~7SF
zNq!{{L0++2o1<GLfe7{!>3*l(v)^JHb}9AvFuBC6pAJid2F*)2hUB*o)<Q{;shCxr
z1=nEtuVr0^W_4K)t2RPQy9on6AV5=ljER;~rxgxx@f>i8M(yizx9=#wU-<BTC&`J=
z%-!MX-4l!FFC3D2!89C7{9WEG#qC#f8msTcDf6zX<dG0wN7}GLBJTI#SDhl?j(B9l
z@fqZ5jt;b9=2A1%bAOYV8MVpI_F?mp`h2}FQ(|&AD?)CRUmy&>Y5qumFK-ob{q6Xh
za+vb$%e-}&O_r%IdqwTf+Z0`jKvN=y&Y4IDmwzl;zBcu>$afLWe3J_LK~VDCepCD7
z1we`$rJV`5EeD&YZR*$0)gt{G1qHx8=p&x_cnm^f*ogd?GSJm$)})NAus7x#*u${s
zZB*|92tY5kP2wibh8~q4xj$ap-g`PbTtTFSf9`bCD>E#VRy`PVo_r16eVy!evB~=E
zUpuVdno7TcQ9I#6t~KTD&fofIxoac2&_*do@&)dj1pf|MFg&PCzoraivDBuENDh<|
zzIx#2QxhsXj>We39dv2OFAHoi#}d?xTK`mn)W}{biH(~!)h4?G_fY%FP6TEZAFqw$
zTqow%FPj@9cQ}iL*}?bLLD#F^h)j@1nx?8}6bR}nlS0{+<I_WpK!Z*Nm4h1_?F!o1
zADbkhH}+<LW*ruTOudmu>b;J)S!yte;UIQipY3SscEytl0VisQ!@c8TT5g`}c=H@<
z48C~VgTsiAS7oyKVz~qNBKPWuO&qi?EZ|<31Z~lfm>j{zI1tpY&wNS<TFsWrNW<%Y
zTb1vlGas$p?icA_w3AE70sUOoImZrg(=3B?XgsN?t}Q9i*u)3ORCUvx3DJH<=*^aF
zYvrjfXP9A>^f{}6yg@g0)7rHc+Eeo5PRF_#aFAiJp&<05_>{-S^$?G2^+o4wT`bNp
zuR2mEoiD<VIzMC$qoE2p0}SG!RRf|U$^NIpp3wX#qdQHZ_7z)K1<f7NJ^00+_G?)C
zt@|kfFr~zoNPzJ~#q_n7r+P&Erj!&85aL($``vEHeoVY;^6T1t87eA`Q%{recce?f
zVtjOt`+V5u2h@9JL?w;tD{0X1wT^}oUb~oH&r-BKzi|ZD9V)8OR9J+%_U&UCj6c8s
z`Q~ckJ7Spl(am`h{(UDk)kDBVgDP8O&S|lKxVpssr?xtUs~Q}j#d7Eg_;CjXA!|~E
z_p`b&;u67+?^>H-H|GLa5+?DInHm6JDdZ`TmxC?*Xx;k&4|g+Bh8qq#KDSuvO&S^%
z!f&xrQQbNHJ)3_+Xo*XI<QF^)j8Pmty9N0*v3bVI(yPDcKb=~24f(mS8jzV)qdRW;
zw*12>`+1$YQB35bg)tQ_{gfB<kU7iG-q+M;7%bHHLKLg1ktoXyP(u*APBJF{>o;|-
z2x9+0Ge=jfqN#`>r*q)&xSxBy;7om=&!~Q+{eF@l?)0cc{*!W7W7i@T6&K*yL-q9L
zj>PcIQl4o@?Y%!bavpM+T{uvGdiKRqf^pC4l)sbyddPV?{n<DFC4{Z~80H4#HP(Jc
zvf)@X{$Y<%-DLX@0}S|2?f=ed`q$JCcx(LA>Dc~#{lDw_Gq2PV-)sZpv4KF;6P3Sb
zS$UMgjO`E5aPeyy1ZnX-3`jQegq~?SW?BY(>b~=rNaJ?*w)ifx-DbkthNHb<tRDOG
z$0#B=`71qDTcNf;_>TqW&9f>R-p%&X#M1gP@hglXslv<y$AHuOB<(8vr)~<$`7vo(
z{3R7j&+T*5%BA|c&KciNs72pI-;Xr5|GFC?*~Xq^^p}wvU~eok+Is%z)U&FC7eI~t
znA?K#e7i`u8?Gcg9ej)Nt<@jmRDD#{G1mX2`aw@%%A4p?;S}aqHsLYK6RR>;om>Il
zRNIrlpq8t!PyON4ued&Q*^^IM?QfyKA7_wOYz7YvSN}t10=$DZ78(RjN#wIhVp<@8
zjrx6OE5X<L>FoSX&Sibk`ipUQZ>}%A4ZZ-3X|am?+{SZpjVbe_qQm#%>$HtWfGi5R
z>rHyW`jxZXX|Ra;^63tq>4CKMSAWP(91`53qO{qeLT{q?olg}4We$^Fx0~{LNunIR
zrdRw2X5VM^?eMy1o>To>!!g+^hzm4uAJlx$+4HiKYOMZg_*!21m#L)E_S&ra?wu8Y
z=N91~q6KD`xP^1o?IuyU@IkvykjdEP(j@T{fb=jB=(un7m1|`u&VQJ2(3f{>eZpMb
zqUzY*a)$xY>w7$%cd+r$I1BLY0kt_)-U?|;KjkSYl3ai3I||xgep}MpDZxx8e)>Z8
z)&Kb`@U9E&L^e9lX75W1z#V<5pSp`HDJ95PF+no7SclOK9Rg^znGgxC3YfXkaIKQ4
z9{=Fbr~VqPi>Hgt5@2pY4%Sw>{fC`5!6%Yo^NpNe$pGW%DQ2nZkaBza1P*xmNmEKr
zN~SZ~@*azhM6@xtv^>%-P3<I?7o1mh0msq(6?~LIz248?x+9b_5y53YcuI;&N3{`&
z`CrJnq^p2Xk|1BlVSrWsue|S~L6nkoMrWF#kPeEok;){oMJ7*DUhp(I0zuXI7~l@s
zt2-{Mw7n<Lo;*w}!;onm)}qw6sY64Zrz4$&y-(Rg3jk$$07xdZ*-HRrDjfJq-%Q_<
z!tDCX$Oo&+!}hi4z{l_WcDs|9RC+##aM=g^>-mN5YP_gQ!MBreFc^%>gn~G^I`5A;
zL7tQ<W5c(*cL$!JzC8A_QxE}xZ#qa4Pcj{h2Pr=!gfNep!~rj-TOsdRRWvw(U=hdO
z)2qWiXD7>2x`jyGMcOs#{v2<jZvvVXxPN{M)lj(C^|zKsS?#|%!;W=Ng67R(-4d+r
z95H!ehh1NWI!6J=t}=+iw~I%xXa-5S1*w41*yV_Wpq@7UL<^8*!s$M-HjHC|WKU}~
zXVIN)J`_Ru$7sgr{V#I(`)+$rsCHyVwW1-B{jE0Ox4vHg`XS+T-wLw4vA~w>XX)T`
zmM$%*ia14#1js(*0CVG$4ek#X0*c2pQMDXsO+#NB#cd7r&w3v5T|f-Tgu_D<(bYH|
zW_mZ`!e0ps)U7);pq<j5l7#YEbUi@B3H6=jOQ;)Z7K-!Bq3<J1?(}Yb&d3g)>Sby1
za$~2N;T_`5{V)+IBKzEW1{PPO_7yd}>T4BQx^;_XYnQgX+UMt~iBl^dPs9P3(F_ak
zOH4UEdL}v-D$=+vW|lZIe(fe!Xie)NkhKYHXSTIC$z{FHDGiC~gy=6t`00UQC%;xW
zt_Qw1AENZHTTQXwvI5aVtcCrkf~Vg%mDh*xM4pS4<#;2HzEJqwv|Q}=O@Y$#f(^j@
zDOe={BS)K2j^6fU&FVH*q;{#+32KP|n2IO7YO7h>-6r24ER&f{Gzn6#IKE?<^l-(L
zK^aa5Jg9iBsXae<YU6etwN1q!`P7C1b!dQWEaO0&s;v3ZE5zzTE;*129(&RlIh5Dt
z6+q3_<kpE4fYbOH4_;0hP@^ffo{m5%rq`t8NA|(7p!G4d9LC3CD8L@?H!AHhshe0#
zFz003HrHrfX}!CjU^maWuJB|&>J=I_ufMBzx>h|d0J`?i5Z*fO*})du61UUJ9mFX+
z$0_$uor%a;&Eb)OOz%rBuKGW^(+Josg8h=Q=U8SveII0BK2H^FL@N}T3J_s)K_46J
z@r)sL_I`lP%D^9+6)*7DZ7Nkv*5k&kOHAHwCdkSpt%%r-5ye}!@NJnW>wZAfF*-Y_
z8PHKNVP4k*qJ+1TWiZ#kN~Pxyr)2G0`FlkPXna>YM^B-I+dqCN7t#!Csn<`*7Dr95
z^^MspX-X+x8@c$Ux6C~QvvTD|uPXTH2d)G3R)Fs73+R6Yz`qR?9!tE$CJm}z-|P}_
zkWe(4OiV&4if>dCKQE^PTdTh_l@b%>D8AABV9=@1S{d~cg05<X2nAdS*T<Xql8gE<
zUiuu9uoHXriQur3n1$Jp;oyz2<DeOviM{+jOzblEWzsE%s7VJ$8GOB?;nDAR1;ccj
zq$?wdXPm`TvDT?|FINajFFJEG3zv`dRg_q6y{-ZIAjv+fT?#iV=?|;i%Ag%(<REE0
z=M1Uh%4ApAh5*-}TOTH`NXs3xljLI^Ue+{6DTQTTpD<8sFY%rrIE)Rj{sztsd>pz1
z3hNwp^5tQ0ki=0E+D~8}gnTL*P_D_oYPu?qXiZztN)SA~+lA|MXSe~w>;P$lX<e*I
z`JXe8lk##T%%KJ_>>_S^A0YvP#e$AHLZ|ntk74>He?UGOa6lRTM91-u1xn)*Ma8iW
z2aX_@gx->0jCXL4?6!B@)RP>R!EZiH-PPZ;Gb8h`KW9$qxAF7!Q);Ac6Np3jt=w-k
zaEHW`rS%<zIDQ|#c7bD_pw2Ksvl5vTtp7DpDF?`g?zu2d`iX;BwomMZJ(IG(la2ac
zZ!({GJzTguW@sjVm~2e|%Ftd=?1Ws%2&Lr$_&&dZ7nA2HQDR^%4541-jH{6&C^z9l
zCyudU{uNGu<w!|CR6Ihz06Y6QMTfR|Z!t-t@Yj-J_@e{MvYzk3=tfsW9?j#3mL}B=
zLSxN>k}lnqaBQtpD(}Z)=(i%=7;f3wtaeC3a%y={dLm?I%dR%7xIGcIinw#%VQr~q
z09r8i0&eG+Z&0aJQox{D9+bCgS}Ny*n}m-x<d><4kKGOulD^?!nCHXO$KyQXJ<$w2
z9+>D6>&AEQ252B>4Vzk0E-_8=ooLeCwZgJ0e1R%2@M+aMrXFw}*vk?lRh;j1Z}rF<
z&720E=dn{DgIfKX+$Qn>i|C!dW2zi&Q=e*gnAE3MyuDSJ<r5G{5ErhN`;BWO`?VTO
zifHQBl4U@(@GOSsnaPQyUe`pxc)7dUjmWnnT8J8zK<Oxb8k)VZ$q4WgOstIb=WR@s
zguGwOganSZ66}ue!FOry&1;{jx5vLY1~X22L8XDB|DAQt;7wWWvm%>B>u`SL0!7Bd
z&{2By0Iue(5GGE9XF{Bi2Ak%k*O7!g(%um|*nLy0IcxV3C;ob)d~-rk+Gx1sSGq60
zo}T_c+TPV$i@Uk1oaQKG|7+sI5aR63pcE*OyoVMp8$nJtL>)!4(keYG?9q?Y7Ruc>
z7K4T!?$2%<L}dO0(Vvv(WSql(nalW0$EKiPh#mS#0vu<$4y!#J=!EusyTwdxAtnnX
zal+a!X1z@25Ns@@iSCXLm5N&c&A2&88<k#9=;h4jH@0AU=GMwykV%fcG$TvPuCLVR
zQWa%4TyY7Avlq0Ro99)VXwbsbnOccl1LtSbco!7(5uD}85XT3B8X4Ity#tM|E349q
z%S#tCKVJ{=92Dul3Y4ieRzs)t(E4MQzPJ1mr?wdPB+QY-icGmXtvmZ?xRGngOZlAE
znzYyV6Ir<wfPEM1u9B<@L!2X##h*4{Ol2HCj}@s)H1a;nU)zP&gg&*fUMQNP1tj7v
z+0hlNnY%~R0ur(tDQian>u3U@zEMja*?p+TWORV#YtJLmVfUN>X+3!>TBK`4T4#R_
zVdq<kqKzNrNvaKg0cPbiZzIO}zTuu*huJGHyraB8;6%{#Nx98-z%IFDMpV08&2F)H
zS?)Ukdv_KU73Dv>4-r=B!9O?golEndxsEZ#Z@$%ne9jIm;w2$2Z1RGYO3fqpyQsF?
zvHr<tPxDS&Hh}Gsl=h;w*w$I}s1RgIXz3~%5u=J8${bjDmBzYH?SluQa_a%7+NMf!
zC&aEah1G;0FEHOn#|@FL*!Y6j()9)uG+!|e9WKz@bL#sVzhZgVzy}{9B<5CYvFZo1
z))b4Ln_hn1W)okzj)l;??3<OD%D8z>w3k&!D>+mSp_J<P5W%8x<iP7hC0r+clkUy)
zqrN+cCHx`T;A~jG<N1|2<{M|<(DKqCevDtAKgvU)M}vLJ$$m7;RmQ)plyf>cz-{MC
zBx<t74b>o~`s}gRgr+yqJA~sR)F(P=NOdzKpez8%^i?{3F)FlZTH3iiRM+2EfGWal
z^@s2|&-kD-Hr*b+Y8$v3%%H`hgAqA#&Ws9s(^%s0yx9?!P?0dM=JJ_iOE&r)Xcs15
z_cuWWIu$n<QQ`L?(a|{c@X5|e${<*3f}DVRk$rKLHa+11!Gm8!x>v{y67e<rsgrHh
z81n^H<DI0l*5On`M5Skn=K{($wczZ+fn4RaOEMFmAfIRcy*XpS1{<BXPVY?NWeM)E
zT<N$B=~L)*=yfkO=sfh(L=!cFVeSr1^z@>A(%zc4qafv;=`{7{9qM9+wW75F-Mx=Z
zz4{a-J`X2}&^JXxabj6PbOL)g>s_cKgLw+-H3gEou)t*>)g90vIYDW8^?TehuCh<N
zDlkm*yu?%Vv^t()J<oYkQNyj0wpxDdP$TgKNol!P_;P5&MzkNnBqB?-t#~tzpii<K
zZQ^BpF7Ogs&czsrkC2iveJ+|JzAWF)ve^~58NElsoL5~NWYIR=ujV_9RyvMpWVrhw
zL86&IoOy01xz~L#``oP-N$2!Id8r!eoTDlW^NBaKwxvWqdDaOwHFqj5htaCN%KQPp
zLSaI~{<CCOg#GQ5rK{lie(Srjl#ikaqa{IBBMc=K_oDFP2<<DQvDSP$1q0O8Ne4e0
zwv=pO=QC4Ref4e0f@pQ$&u6s^Y4!8M^96`ZyewRBFyLCf;6bJD9=*W6uuDl&%TbMs
z?1Y5uvRN!}R!DJs7=C&WsEA+uRB6Sfy?vjKDC}lRtb-_nO`SE4O)}s`ZP!)0A8-a!
z`-TGvKu2F(dSh*^)%L!Cw{p_2c6~Y;9IfI%AUsD-b%VCvejbWSsp+7K#z{yFfM&%1
zE?xA;$FqefrJ-Y?FiO=|_q)p1E?od7k{|y=#WM~qp~7duzW)MQHJz9bIwDS7d0s#}
zfrzYf9xYk!ycsoz<xh&znQ_4D(VxcdR$^vY&_sf;E@4Yb#Oi?-C+P3MhefYK$ftnO
zqdPNfDo?!VixHV%h)nc$gpXM!OaO1V>`5Sg_;LsP!O+5QTBBSANvwLNbjgQ%hHCo~
z`?pOR_3HC<Wj(-3x-m4YEU%&wm!;)e=6wljS^=D*{cqj}sjb(%tmRC$#|pXwL^k`f
zeI;Wu(;y{l1Z~$1Q~}DY(HfVjO_Xm<u=`;hYwV^5_}nZZttq7X(;~8X4WO2;D4#Tg
z-SdNiIwHDow_hYr`U@p(!#gpz(M(Dk?Tzv_ZWd!O^69wU^~cO1QBw_O%XPgeOM4eZ
zqNDk*x$fSzVYJ^Pe(*{o!sVlK0|q5JHqLo_?=q|#v1@25Gotz!W(01&q>|e^Y<kB4
z%(w2x&dz9YLv@I0t!rK0hIw0C?U_7%Tk_Vuw4BM2o<YJYE=IF?3p_DQ8?VIq&DG4K
zAZVk?fM|MkGdV8LU+v9%%@Do?8~~+fu3ngSf`)_hSIe%udb`a#@a(ZP9KiWG-#0P2
zh4|J*z9#fWPnY{Yv9%?e<QA6EH%#1BPMEUXSZ`<0@={NPvRA%|C=(XVfFEpVvpKC?
zoJ`;)G9NTf@5ddHuUHTXtE2rEq5%lrJ--VF@<;d5s{K}i#_HC$Kg~f}t{e|}oJMS6
ze<SdiE_i}Z^5H?YZosf<Wey)kQ!Y;XnekR(2X=+ko<o=<9wDheUjm?xv=?wkG&DuS
zWXn${3$QqweHl0hOmtM`7J{mPXL#g?e71=X7w;ZcP;QlaoszCiBzH=PzRaUz#N!@)
z6z7aZV+tAJiGy{LRtR0Ds{E8_?xUVus{=XYFzsF@-Kdw0v;=o4;oN&jcc_q66v4#8
zn#q%(*?h9JnRQL}IeMJo%cVX-J!kLVpM9o-gSfi0Qb)hUB26BQ!r<mM#(m*AB9B<J
zGji;Xfrqf|#cn_>X8%Ta2eDkkVE(E{soE?<eTR3A6I{x@Z)N;ziMwm>PLR=0rP{0g
z<HC)Q%3=s{Pev4{qd}T0Mj|)-jLVW*&{vXSsswpJ)HKaWJYZRi<iB^S-zm^G^QSx6
zCu5?bp^NYa3<>XIKa8KiftZxwiWnfV={_ALxA96`<qjlh1mZKL?@jGyieiJbiVNjz
zW0qzR`yz=~ROv#C8E#xZX+62yzWF+}!&?+9z7+#|D(FbNd5lfc#=9xgJ{t`^D8?G$
z<PxaY0T2U`>=4FqSJ}ITwZeip)RZ1<_|oHk?!dSKr?GIF+s3{EMx{q`@b5&FBvuh~
z&OBe@@H0LJdDoH74D^a7kW)y+>_&;-p#$A7wWyq#jykR0_Mfw`^12EfX!Sar@g+=u
zat$Kw@<#;&CVeli-12C_<U#R7lGJL~)0<!Ehll#aHvQ1GVT*PbF?Y)&6{cUeVIDM4
z1fM;hq_xCslB+5(74kRR*OuP91hGqu9g}N;S@kkDNyXoa$@|(R#pvt3V^IpuDFd;Y
zii&Wa-H(;dG8XQ3AIy+t{*1`k?M^IEIPLX3Y#p*<!fCsl_5*X+`~_yg%d2MJsO(zQ
z5}A8QfCT%EGs1)P8`%S{dziZ4@o%ChuzP`V7Wz~<roj3J;{-XKVx!vH{<!}V2b2r1
zOK8)<j{ys;675p~Zzs}$GuMxSyKR?cn&CGK#Hr>1UMIE`+4=oS7|40?S7wPEYTD<H
zw)6}-Aq}jGxx`yIF#eR&SR^fGFEDeTD>v9Y=HsC$pY?U2jsQ>FGIW%eZ&+^MUf1C*
zZ>9ZJ9t@$0&J12pZA1(2k3Xw;Z@w(7nX^GAA=F2HP>pty$2nSrz!qh)(ngWCIr-kb
zBBuXvwu!P&$BpR=4;9g1^)QYc8TiT`iO)AvIxHU8X%K(@ZxU#naLIBZS4Cr@Gs{J(
zUTs#j0lIc(R?b7qtjs|>jDPJ6Sq&<e1G_jD%<2BkIubHKqp@i{L0DuyUGL)`x$K!I
zbeC7kcO!yY#$DRHLl!Bbgc)f->#!4eiX2@azyTB)_|g&ilhAQ}67_4cci=H=+w#Ez
zn}q@?C8U(Ydj^+2&>K@S)V|ZCQqC9hwB!&=mOe^|!rx)nai4>3u#XSx-xuBYO=XWf
z`9c>OOfvv9c{MW=eOD4DW}TTowyA?5EE)kxXVmCc+pxX(xcsdl;e!X+Rd`7bQ~X(~
zq0Yt@ebiOyQ^BvZG@st@``JZy4tqO4X<fba{`~~AmT5|n_36F1?4!8s$@D;Nkoz9w
zFS__auD}Din(@Z(+oK^wcEn<1W2Q?xa$vXSL2RgFg1(ImuBf;^X*or2BvGQryq3NY
zR54?5qhyr{@93P-*U@A&NlORSvBHH7xcW>EIo9cyW5a|ljjy-dV8p>oDO~oDqx(IN
zf#&4jVjQjUBm?NqC7IOhXfLw~pjjF7wdui%xyeq_JInBzHgLmcP+VeQzq21^eJrb>
zX;f~a+?y5MmszYIw>8OLQZi9rSZa-18)NCKRHq1>JH0i-HUPvzGf{*k{)xi<VW@Jh
zMOBHR9akIoKK->R8iok5ls|gnS9JM=o!d|`tKD|uxs=vPGg-$YKA+|0-y%u}jF_q;
zebUL3PqeSEw2|f9|1SR*vy&MB54bZcn!oC<Mh+UH4^*Un`|<}2+umEgEVFHE{g0~m
zwRSIUgGutCt<?=E`$t-(zDtraDQ1xZoQks`M!AgO_}Yk}%Z+>Ik5!n1s(8aStzCgW
z03du*jN2mvVyDkgw9mFWcT||kh_9~gU@bxBd!&-c)j^4rhJ3vF0+$t)gx%4<*evIA
zR1TgHnIMYjZk+1;`xGSQ`&*1mxhLvYBxf&o)xU~%wGHSuh(6EMz@eOs^w8KMIdpN;
zL_s|pDEB4ARDF8f*)H)p>5tC&)%;VPbB8BS!$u2Sn4VRl#2(zgZ3y%#hAnjxXpB;Y
z0d!PS1D(NG-nCf=8|1t}&QtFuIPn!_;`%;0%6^k@SdXJM6Re|7poL!tn+^L+r8ORa
zF_1l5?;+0*1!5JM3@$L9tHoPx?f!tfrgdK#9Q2~BQeGn-uvJF$O8I2mN@_iIWW~2;
zgxA?GIT=OGH#l=xud6dv)dk=B^>+=m&EYMdlqgIEXMwmxw#BCOaz`e}Y$E%gxLG%c
zca@HdUTkI{;Wg!)W~_{qw~Lh~k~i%l=6}~TYq3L!_-7{hXtA%Ok5}76)5PJ#v(B)`
zd9$DCiovODa64b8ytc=>`>}-7=fd6c;xIeQ8`j&sFOJ<uKFGBJY8PvB@|<#@-W>_Z
zP}4yB0Rndhh_?_9rnf@;txJ<>Cy;>4f{Wkkzh*mNtcYz4=zRB<n+GMV{o+a`ruxPO
z7Txr&!y%2&YEWpPEzREuAV4<nLZRiLhsBvxs+GUm35>MkKI!2w#R|vka&>@^$67N-
z=a$MfP(Mc+$5tdQiv;P%Y!4@9bPhL|TqGBLMP4?x78TbiM|gk>BDfac2Ph1&crrVU
zIvQ<xkcG*9K_S!ud^ew(+gE$J>?R-FBiv-LS7ugFDCNq1mZww#cdhfyT#Fz4h2LU|
zSCTrzUAyd)vKv49YK*$iB$LX>U|!KfJ&`if`RTCTil*=SObDQ0R#BBrH=iX-10{2s
z)^k$biqCnUV7$B7kdl(ic~0X^o!qOX7HDtRvmJs__<A}yeVPTZ%y`v5?n6GWDgbo&
zRBz@19lmb=29Ihom<CXJ;d`1Og5EGql*)yf=T?-_H%_><68irt4<iq20Av29?|Zym
zW~`k>Q5aA71mQR-tijiKMS%2MtJwsB&Yr<(XAkL_4&9mYp5DGBIRjh96r5&)hhV_E
z!h!C(n)_G%@raDZ5&D)&(0{-x>}S{j2YQQT@spS8k#MCPBGk$e*2cmp1MH2IBl?S5
zy={P-b)_l``ZCzgARh|ASajtIQ7wQn77)61%h=5eT-W$1K_|8kOS_Mp^^LQ>^v^E|
zZbSAb8U(&6E64H%-EU$DjNaV^jN@taBf#J_MwhJM_`v{D^32gG>_q<6qE#It%eSb4
zoQgHJr__si4Fs~XaKwBJ|6Jw!bWidjb!*KW{xvh{0f*_c4JEO9J7EX^$f3t1pbekW
zGJfd$2oNB9o`@<2B_$=L3Y-)vh^6LsPMX-Spq<SYoat^2EH3c_C<eueaG%7%QzjWD
z{<uvqN%lnC-ArY}(2(hw;nC4=-R1>dd~%r|ws#!wBYR2KS5bZ)X_(TIP;*@_syHx6
zi5wK?+g4Z6{UCePwpN(`qA$f@wb|-Mlhg1FQK_PgcvUlcdFOAC*|~F5krblN1Bk=G
ziWI_gHZ(Ozu(&+hTG}dHN5MLVyiP3eu|mCxHn%#e=y~^wrfpFywMAh)U`h`m6g4(k
zD+Dq)snK}aUbAj)5xAN3&YCU{9+fw96A5_QpQ&m7I78fXYA748LSHV(XpGdHd3TTm
zp92|{CbdlFy2!fsF8%B;+{!1*HZvys38$Pt55z!r`=67dbn#YKm*rV`anfcs`Kh{O
zh2p&IJq|jB8vw$BQm~F_86a_i%acQ~%O=a7CjsyLQvagTdeytu)xf|ASv)=Vq!PIg
zh|*F~-F%2s>KR#q*ksk)uQkIH5*^P{#d}oo;dkYq20^7oPcOCP`g0T}5<)&YAnkk&
z#RBbfZ_=kPLpwE2=6_KUMIa~6wSM?NhjEWjp#W5iw-1B4`WeA}ga|<#xuSMlkShBG
zfUK=}v|Ixv5XS#pdRF!HQs7m<p3*KRf)|}|AV3ULvzDU(EbuXaRz-03tUqK6VQb*&
zFZupw(3k&=nTkqgyT`R=z@>AdflvG7J{|k%rvPWFe>UI0Uvaqh;*NPz8kBCM_^LUO
z6F^`+ImKUr8FB-zU@x6s{NWUa@L#W|g8zQ(pMSVK{&!fXe?2Vy_hjzRTZPyEKQiI}
z8qCG<$uub|Nv?{rM4%Xo_vG^ZvZou$fvk7ULoVvasytx!USWA8qM_PgxAXp>Qr6`D
zaHq=mP!0tynRADa+s|D#|CO_P;gR_xwlkOC`0H`9pJ9Jn9@@ci=7G$i0tgvp1b3N?
zGHP~V)m;$1*Hm5F3U?_rf)`H|Oql<P6)LL#vw;NNKd?X8^z-C3<!Cv+70lr;?@aMs
z-GrEtnSvQ)WFP^`0?H?rF=LBfmbRS6E7a)Vw{lGXjy&)U*NO?2^_(hHDf(TwE_0KV
z(A9FiH;k`2&Z%0fp5b0Do(LUR{`cbOVgQYyP>5{Y>yMI76W;^8<yz+A$#K5FTftLi
z7J8`CUfkGa<|bJ#dcjr}6uhQMUcm4(s0tNl+cC-U*0uitOP&y$i+<N0rksN=rsMv+
z8HcsEt>9;bX2c-tKURI-dCs5;jYq<M;_o&3@257=3bo6|PMH0!{apNvU4u;X??B7~
znci+$(z+|hA;XM+lfnHP6X^=CatpM(alxX8#u;s)Tq|Sihh>&z%{9h1z;_MtoO2eX
z@w4fnpcmomL$YVMxAR`(Ke}43FuXAei~jFlxM}$&nldV?HYVSSx$ZTQxa^l!C8sFB
zNQ<_YK0N#^OS(6Eb<niD`)>eCnhwg$Y)Sl>)+9YD6>uNMwb!bs>_f-xQ{J6pF*2%i
zeo-1Y9^ZY8wepKWvc264R5TBGu-7^-ao?1d&w=ojHVXc~%&jMr=YiYr%55+0NMIt5
zTMvi71v)FCFuFW$CENkHep+sXo;`U)3IyBp)O`<waAUA~x%>PYq1AOnScWk)+s!|b
z%YNW-pJ8sX{J9k4teU6^`(Qu`ynXn7?c>4%kQ5!a{N}aovCM%JHpV9mqSnJZ(Kj0z
z{>_dd+9lE|8rmSR-h0y5{rz;a>LA5)T$b4(NXi^N$Yxyp@X-W{4yFKTyN}4DFTdS8
zCab<L*=HGRJvRm80VcQzvJn5en!R?;0`oOEo#sD@cKE6yqpHhxecWhGo7d_{>bR2H
z%H#8<tv0bQNT%(F_uu~cz#DIMN+#2CKDS>+Uw^AJ#6E3wBt!Rao(GB{-w^LP{?XWd
z^$|aD*1%c5DiZFUj=~qj*#h)O)}XfA&Fm;mNlzzXe(KK?Bl$x5U;4t*CZa1LG~(WV
z&jGxHkD}iyWlMje4%~CB@;h8$XvUN1`*;n;9ywT_kA4~S8}p^44KRm(B}L0^p<X3c
zj&}O;K6P*??V2@ZsONA#|M*p4_W=<0;SRxpj}MeL68g~2LQKX?C3`sX#)nI0-5z<b
z)!uU`#jVZ->u;GfkS9V*lug=iu6-6z3!IwYE)M-NidN;~QP!`wF7$5ardjiQ0w~_A
zyq3lGcqP~<!@cbLMb*(a-nxSO-qtScH>O|3bY+&z1YAk1JkpibYax5I3`@2LtcJ@@
zGb%-Cx(59!ugRR-l+j6X@534VeqA_pJdPh6T3{T?0f`?t{YqB?_3ozmL#5l=zeU2-
z^?aaN7ECyeMDSyb)zPZYT0h@BvRf)6?wyaVg)iT)ed5+^;iZ*H<^4%kqWOJzUrdpj
z5n1XWSeK#o`}P{~Pb~$OU+Zf<6^)n!8^nO$UaV^mxHaG<%RqIK|1wybT5$e;n}@`O
z6$jynAVE*d2TTEubhpQ!sYf9QAn~P&_R|Fj-Te9M2<gUzJ49Y(u8JzPYvf*rIU<gX
zkI5B62?)aiE_*->f9EcGW6P37y88Xl_NBWuC)TmuX5HhHI`Kw+^BEUI?V*{$@!P1Y
zT{IT`Ap+mbR$Is-vhkk5Blu>HOki&~$vJXdMc%+<)Y`+KGRoV;ot|TP!THF<k{8y7
zZ^Bcx41?TfpOh$_pK6ag=Qxp;#a9(5YR&n5XahmV?j`o~#=Hax>vci3Sh6zcT|v1?
zo>JLHx=<lalo4}%|H5+8ZeW=!Qxzs~dF#;wXuncY`Q8SuQtiJs6CLkfyQ}p0W7DuJ
zou0;OJ5tU{!FTav5)0z{+7dUY8b+lBSOF4$DR1*<N+7;{;p0SOp!&}+Y*2suw%cP8
zcAp{j++@&Q#u5&#K<BmnIxVj&%0G9N1>bWavAnEsA4E}(1?4575SKWfV*{t-j*IxM
zyl@fk^z0{iHner`q1G{p@-^u7-KZ{6?Nxlu{c}fVd7ZY0ez3}VFhOMhM!A^sVOh+%
zyT=ds4VBkVOjh@I3U%%T76tY%rf((D+Y6pfu_=_HfBfx<1<}7dQz@>}!T>7++t3#+
z;CDdiNwnWIZkMDh<#^e8PIc3|!$P9c_7#thduQF2Qaz32r2XM!FV}Wvubn*j<xvn9
z-dc&MZEn7Wm~LGh_XKH1hPA)bJNeMJxIgXRYG4ChNIGg9iS+7CA9Ko})+(2N@ikqV
z7Tn`gtw)s2DwjDo9aw*R`9M}$hKE#CSu3#rS{H4v$DYilWl6UX7j7VMZrZ0cMhF#X
z&>qt>D<}ZoCi4O?*gvmDZMk-?mFY+wGt_4SpqA6fgR}Kj*E3`(wg7KgKSlK}D?bYy
z|IQL?_?N-^nUgfYamPw)adk`MhP(P~<EV6;l7e00l+s?mt5!I_lM|gdSo#-XK4~@V
zyHZx|iU&8xBF42a3T?8%WnUsaMxR0y+=bLKR{q?b{YW*~X&J0EtdelR-n~SZni#jS
znp|+C5|Ze8{8AbRxAPw}gX$SER}-p7#JufqCub*hI#s!;cMU1^{WO%fMOrZzdW(D?
z9YBS5UsSK`&?I4RR@xpVL}pw6GWEfB)_+t0vG?G<JwlAKyovUWwh|qwXzfIC3irO~
zV}n;_#u6=$<YW9lQdun%*yHXtBT8|*-V&GXKfS-!N>NL$CpakjX1h&8S%PiQ0blh>
zM3rIBNZAMEwOrS9Z;I)$!g|tDW9#Al{H8WjHqA3RK%Ci8Lu&8XL=KZ!<@H|{kDn$F
zGoQuqT2n41xL<G)p!L}_rOtjS;;-(`fM|^>d-nXB|0x1(x%i8!ciWOKscN}LReqm)
z*^70Fq!!d4s{7UTGto60;WDtJ8r09AXMYK~t~p>bLC~qQgBzb$t#cp|?mhHU%)1$l
z9!vd_f~p9w(HS_KI%6miZkM@6OT93wGCq`Z&8-{QUChO-PF0lj+P)zM%?xK$)@WdP
zjSKr4nY631xg(dK<A0z<baTyIKa76Kj5W7N5Qg7;!LHJ?_XDfaz&4;Zz40iq)Yix`
zwLPu+FWHfREAo@Ec-2&nIo*`3h|Twh9;3eCzLleb@#L(9ql`d?AFu+64;-V4n_DgA
z9B*}OjvLj`fVt{+(`mNQfz!mbYI$vzd8+idAB-?~!g;GroU0g4-8udV&3+&>?J>1q
zEKtqXE`?){cwCWO_L-(s`SvsrS5EP@B(ECz%alm-PjBQaAr#FuJ53AjbIh35)MG0P
z1*(^Cj0-@xso)cN=AeYj1AV`e3j$O^mRFb7!H`XG0CmhkzD_bqlVLFB++F*%k7^w)
zv2VP%6CE>JajF5G8RE`2Sc_|%RC=Jy#7WGSthrqmS5)5<W3kVx<wluyS>1@)iL7rF
zw497H@<oKR;QTgTRN^oxPX`L+Q@hBVpe00Kg`fAbrL)tXc+HW-wc<xnsCBTG(}=U+
zob=Z79S{AOqi-1Q6d8X()weq8qrn+Z8&?fWr}KY+oWKMe#tw;|V+QFAWr|kL%f*mO
z+07>Lf9ZYcqfz{B<2h7P(-CRc(KoCX!m-_ZkapcHITL6u1m5dh2@c?8UB5~lR+O=b
zAN@r?`Hp(_5s#08o8TGlt9bt=5l7aKs7S^a8y#Z$_%~F|h%N;3*)1-)7$ocwrh82B
zvW|ve7{}!PCm$b?mDK`ar&eGpv`N4k%2zT#qhG%i=?PT_ZG`qrRW^tk%;$w~MJPoL
zW!-x?^?P*s!ZkA>ARanwHD?r~o%)`B&6x_64K)?^wN5HoR@fDHAM#6Z>TTp=m3Ep7
zRQ5AEor<5M?xy*fvuc*;eN{rjpV_E9?h9gmgu(r%s498@1!MV#{F=OT>lxrUrsw<i
zcQ!9GZF4pE5^h<(a{&hizD`)b0jZ?6c-L(Gz2jyj3nqnu_>rD0e$iongE!F(p-XqV
zvGR1>5;RmfEg>rOooCb&hTgRt*Q}P9+#L+O@%GYs<L|A2o%dsbfBlU70jaFNE+6S}
zZdRfetfF|Cp`B&__{*D3gYwnAJ*AUlA(_d*>u)b*lzc}&8@P34%CP7(roZG_wM%Y6
zj=G=uGR}1z(Q<#r!J9D>|HL!Z-rq;SNyL`y*tn<s=HnIF<j-eWSWwaqTN_^n^hJA#
z6Opd%@-Y#pQfz_Lik#Hn7!G}3#5d4r3-FLICQ<d<)3C1oV8sJmq)G#MsnI%cH2dj0
zYpz7N?Iqqo)t_iMT=3j<e36NOs5kVQ5Q(qzz^*+*+{F5FcgH6$CUMbQzdjP)eklmj
zrH;C7g!u1Y&i;1%09}$JHnukYnZ5sm(49WJ3C~0R!#2mHeGh953D+WI^;0*&Enak%
zlkvRcexziZ2J}WcsFgugpvmvmcv^+B0pGjgiPLn~E4lEI&LGVNg7(TrtMn|*<-CZS
z{Of;8a$e_D@2*WY&1BuGs^G;$eMPZmDrz|YE_K%H@td%G1^Kqo6=s}r_1Q<l;jtb|
zdzmIYn@_jk`>xWn_RGh-My^?LtvY`K`Evq?^@9fl%T5I!XG`4k-iv>+iAsKJ-&lNR
zZGC6$?H4*dmka(Y#Y>hw{rVO|fuu2G+k2OK^0c<Jquxa0%vL|o%L_MVn92&C8?~>p
z;rfl>HuT-%b}khMRt#wH$J9|(hwK`9;mh^Ce$FZhu+mfMUGdBtZ{sBrGV8zjlpSXe
z^{;fAltqE$_zXv77HH~i-Wyr_tpV^@m7Yh8q`eS#F0n1y|MW;~o`J)norI^en(d?_
zVSEumppOsVk%O42-`eZ<)x4iGRr4gKzpDKC>*L-Ba}4sszPy#>a6NL@)a&a;-jT&m
zOJCVvq!?*YGhi$FH}`Gi=w-=brnfqhU(gKj<hv>06HERi28Mv*Ra)w;huuhMz_GZz
zs!xmSiJoVW)L8{jq=(4LS^=+<Vo;<S`*7+^Z;y{46<WCUf~vr-f?i9HEqz*X(m1yT
z`y;hdC|}gcl*G|3Wv7xO9<XpPa1&}eP_k=IzutvnilN$rwbgO{M_c)k+VQ*NT)AIq
z0!A7OHaox1)gdv<sYAWvlJ5N!@8i{!lBREH6^-WlQfqO{$oJZZ!vA?(3D#5uDwhRc
z)yd|&FY`6wGbBA0X(#)7JO9x`tfHx&<X$EuJ_5D<)3!EOYUL+O>{$58XQ>;QOxceR
zXLg|<@0T!Tp$p*P`f%I5#GmZUSvGNGg0OQqeUFoyGsLBLw^`RLgCk+d4AbC!f6U!~
zr&+Y0?&G~VGi`>OJKgUUzYxTcS2ipJ&TY`;ZlO3c7h=Y0Xg%0Hphu5xTrs-}&OXtj
z<q+;|Q@oU)Kf2@%dXgA7y%}9#Uc}XbWEJ1fmw4q#Tr(3w#8Su2{Ke7eB6ITsZmh?m
zyp_lBARyZKK*_c5`R%ENelB};MAP@D-Mict;GV<&usx2R?W@z78S(oy-UiMu`-r07
zRGyZ5QD%U;WGnBIfqElXrk~Ipxt(DL)v~CA)njI%0WHxQVKv)6M%_R9{f(JMXA24r
zlzxPdmRg@?PshAUhN*>Wf->15dV!AnF#~$}XyO<mVa@1gdGYw&f3bo>M2kF`Mtw#)
z3%bUSE-$Oy*!+p-#45%eZ7`=~-AY(`Pb|T<DTuiAL=2|oB>_1H8PDB1a2C@Y_@n4C
z9WCG9v8#mmYj6js!)>A2##v_KDah%``<L|V4BEdMM&G2+gmsD)<rR(nBwFC>C;3&f
zT~vC8-uu_?(f14kg)^Ve;n>{}dMUE@+y7wiJ)@dh*LGo-4VC7SrAUXR^rj$ELqr8c
zI!JFKO=^(P5}E>ni1gl-UP5mYNFtzgh;(T|qy!Q=ga9G%P2790{eEMdzvtI^=MM&h
zggNIk=l$IMy6!>wHHuse#G6yK)5ikXxQu57__t(nySex!+D7nhfd2_A7!ym4Hhu&s
zDs!n!+6twzU{bU}ddE3d!|a(|>~)lQEw;RF!|tly%Cd>HYyT+OgsLSWF!`AWNR!a@
ztVpsndp)|?g0ah2&s8mG%%@j%4hvdKQK<^NC6$A{3$73LSkK9$ZX0LS4#EsJGfIBh
zDac<cxjisvti9j!+k3WT<?l`EskDt5+qv}6K563)q+PD=z<jiGOzwZmFyq7fXpH!d
znQYXE>C^cplf(2}zZ}5DS5$ZH(i{M#H3SaGIkEWG@pV@7@HIPrZ1n#TchEfkVyerG
zbvO&yrsRj5w}Ydy)G>6LF7|yO7Hs8^(v!8FI1#XiY_zjPPO_8(f6Z2<y{dTD!qCL?
z_4pzN;PZugAR1XJ6rE_@P7l7UT$+9QjM_*e-i6R@ksm3>lVo5TN(h_{^8<Es!-T&m
zLt-(u)96rT2+#GblRYCxS9IJ#pU+6+vZ4LJu#e>d%S%kZ-+|I7lbkynEuFXisi3)C
z1$3}pz27~f@>3lbIV6)79<pVbfd}B%T4S~%c+;N+DPkF84*%nM0M&UEj^&obUCiH8
zs$6-WR3kHi@p(IO#XbNb4&BM&!-kNHbH4zM5uWcSC1_)<En2^v{lKQWsb$jRXzA^T
z6ZUe+J2HRQRtA5f6_)<)<HWJ%+g`4C2UH#WO{yI5``3HZjAI^E8jbc_v*cLD%_Mv{
z^R6h2+xF+~+WWo}Ira_N;+Q!gRugsWvGzzE4L!|tfH~qmwF-PJ*}=C~O4c0SzZE+d
zTLgsYn$)qa4Bj-~n;Xideksu)i}<L#1)YWl))3%{X42MQ;0ch!IvBv;E)J+u@Xj7L
zZ;t{7Jt>curD1F%l}r3q42ts&!2Q+M3zF@ZwfMDTz2~s1En%DnvMRwVY!C0X&SuMM
zqq2JwYB(86{05Ys9JB~Bd0!R3zw4BA__+`|#1oZ(8e${odic00Jt_CP%?7}MUl6`K
z4?QDoJlxOKe&MF`KP;!%>F+<=WKC<rDuJw>zTEa{<p4Ooz<2<c`HK-ZPPq1CxY)n`
zcxD%v$^}$5nz8Z{q(?VjKkLnt;<@-A&szNyNciP#13xXMr7N?s!K;o>U%I>rUb)vh
zVu&+u{r5S(`<=er&)@fK@6+Ac19oyb@Yqum0p-eah+_a^)}4dwtHJ+Hk8Kv5!dHJj
ztPPHB>IKq$82<9wpU(e2)8>WW9|k*P0DANT2MO;oLGYf*8`VmOg?AbvZas5F&+Y#G
z)z4@Cx6EMp&!z$%%Kt(;|8JBJ0wk5XX7ID!No7tjG_$`RW;lgRbv=SAYlUFZ_wjMj
z`xjfXc6$d$DK&>Y`*h4xc&3RNxmY08z5$t-P@Tx)N}AiCP(7xlxp!v$llFBDPr6Be
zeAf9qb{4<%3w&#>8)3xd&tD4(i8*t|>T>qC*Ux~V7Sw}x_t;tw25%-sA=y^Jh^}+%
z)lMYQuSRHPX7D9g4WCVE%}al<5b-*z4nQX~|3dM3qn!^@Fsf|a*4Is&O3REJAGvb~
z)3%4~m1U#>gCNPIWHb;Mc@%7<xdajUe@sLmaJ0cOV{QF>jOZ_Kft*{)^cL4G3HN%d
zz$3f0Hn7?0PE0g1GP{9&@qlK=Q2J&jv;9gd^iQR(NsFcZ9y@rTEM#5j2{E75ROX6e
z-*EC;QL9?uX;w<R)`VHJEOrfRxdYbk4N2h&(e4E%HOJ5%ybeI}*(k9zd3fo0hs222
zVXV3~vE;?p@<O_jZ$ZL%G$Ke?YQ$*d&Lmvw8hfNEdXnw<T-wQ;ykkD+tg6Y$8u|lh
zw|Dz{{~x5Q1!Gz3N~W2Ju6f0zfyPm(MOsqVZg90+C1t&LyV-D4(0B8K7<X{&=aD^I
zMSo}NwM1(`HOht3|0Fb7lWy-LR0^ppPWs7N>C5-+{w}7;oZJQYjD=iZb?KV@UbH#j
zSTS01a{2SI@q_#`vbk;7#325Q9w1C+<24Y^a7btxf{nc>6oN|LdQSa8uls{(dx>eH
zI2M3G5?|frs>*BZ<ZO{C-rV~kyB*dr9k;gV-~kdKyR{xxel?n4@-V)^Af~e-FCRw=
zlCt}$OAncC$<SZi<xdM@K9VmK?SvSZ;7k)Hrj$$!d-rpsXqNlTTu`ugKJzp-q%C^!
z!0N2eA5Ry_4lHt^O*$g{m~{BB7Z*!-sTM|RI?eQXfrXOkt~H#%Z%i{6fBV#5O2WUa
zLfgbLSrnIvMmpT(?)ZB1(9Wr>bItQSk26_nH?)NgD~>N8aYuC0YGSUMb0kIO?kzSZ
zSVC?F1%zK*$*TIaz@G-bqzK(wc{0YF)BU6WXwDRq#ciw3U?tl9wH-jC>YhwFF`xWF
ziI@G}$^q6K_Dc&W5L0h2wxN1hXVw>xjvtJX{NK(^NlT5WHe)G^J&e$roGD5bFB|6k
zREwNB^kbDMWf&Ne!Rbwin7wtHppi!hrWIV#C_er<r8p5cx}poaBjY0rugj4f0$8#_
zTk<D8khOpBU>MXBBKuQ(-egOa0Vrlj_qS5j80Mq6XX#KoR-`$5_jdWm^@GKFIBPE+
zG<<>dT0H}|{<x|lkbzFIUD*p396Q?~sR@~~n|lu7d#9G!{tgPFe;AlyxANj5k8EX`
zkTie(^*d?%!Y7)+)TW#k5^F$f)WC56bR)ctq(1vo=6;pGyi(WlAGOSn?5~C*QUoQP
zN>KP8%HAfdD&m^w+&po9k)=NU>f8pzyuD}*QUL#WBxgEy1a41}Bq<i78FAdl)_UH4
zN7u~6pis4tz#;-8S~tK~Eo@ZLA)oQ-txGVq!Hl_wO%Jf?qrl+L#)aU~JXfz{SYF9U
z5a|BeOQXo8bw*WBU2Fs#PD{!<x{rE!KD7=1!2C=K9yd4TTe#xdEW(~jX@m+m{AX!j
zjpAj1wK0b(=wIfXu38{ZUYYG8q?F8t@NgZDMC43F^1(yOJRw!IyU%^^YQ^KS{{m#(
zwm;*roc@)^zgMQtK-xq@VitB2;Hh<P!KGuWYvm3@?t(dE>O#umS&2+hi{L#1Ee!w}
zTH?Mm;S4OM`QG@QufWUG5&K?iFEuFnx6%`Ts+c8+bwD^Mpke8Cc4^*GiZ*2jy9W4E
zLqrlNq2wd|-;Hac=9p!lS4vkyTZ}Im`oCH_j^bXhdwg}9_)P3lNh%NfHe4i~$^F@V
zT$r@DY&*h~-$j?Pmw|b?HJ1B!k1p)`d6M&ntoNWra@b&RO^zDi#D(w}LYT2etn}j+
z(3u^<Oax|m4F`R2Rf0yT%kox$&ZHeGEC6uwEd`jDO={DZS>j%wMF%ohank2F9y$dU
zR?A&kAQ|5y$6;EMXEvKei!{ZN8!czB&)tEs#V>e_x4xxOfSJZAaV7jfjo9ER4NqP-
zqSmkDE3Nvy5A8Y~=9Yt@<AU=;MfVR)nSUU7b#>%9>C9+qj{P5pp&5fqfdbsFyWSWh
za>H@>Rea{kRxrQa>e6c#)>qHnjXUr@3-^57q!vRa!Jw0_ayE&p{^Y0gVG8?Rd|q66
zw6L*w>d$G#34r+nbECYnp<vjCWZ`cHSg-8bm>P<QO2>w2XJrs&y;$VPYiO`nh}b&o
zzg{zKrd~O%SYq>`KpS)93w>;FwSFCMlg%Wo^UMKT9dF0Kkl7d{zgm(KBx{Fx(f-oI
z%8W)>VhdVTj*i<;-5@9^;I{GS?{;2WnEU!-vdOq1@RiGmtaI<m)b+H0ABpxdR(Yl^
zo@s!CF30wcc;(?m#@_ZwWhRBm9i`;V@y6FpiniIeB6i0~>%9&V><eD=x;?P+ofo}a
zz0}T4<ue=XFdjD5ZfO1ysYBn1?N5A7R?GcTsGFbn*TnKSbq5fkF}w0SO_T3XV(6l@
z;j9nkC)=XcQas5oC)(D=k|V|dyHD8#Mzgc4l|7{*k}BkX?H;W6E87nEP%SJN-Fix(
zZ>@L&Ciw$FE>^S^mI;N(=mF*hSUFkJZWn2uk-~~j0FtM*47vfyGmp4gTl41s$e;2U
z?yn~An`Vr>%K1pv%?atQ3-bEuee9*n@0HlkC@*nxM`V87R#CLq^B^FhIEG_D7346u
ztxr@GkO{isK}Z>Z@~5e^LRRw?KBph`SPrdAnyR)=KQ)wn=|7uTU8G*K7QZ1xuO9>L
zomEZPfRULjEO*{}<`;=BvTS)0l~*Tt6u5e<7}1-xg+jHI%3oL|%WTC;4<N?FOv)MW
ztgE*x>@-&Mg?C%j5wueL+DNLc8&tH-`iv8%!ZSf-W~4USshrCho?J#g5EwCu{)Nvf
z1?%;|988+$Gsg=Ezr=`x*|Uv)xs=EuaU8xz#R!>1`t|Zz5sWNqe?9|oA4HrGbFS#l
z9?^ewIB_eRe0<*%?UG|_+0u+YDc?crEy;e&JMPlytu;0}82h&4P1EU+zVq3LF?1zO
z#Om<4WXg{jTHb4ZR8tRTqsv>})@y6t^he$5-3Y<z5ncp<Q!Ic1@l7$#B@P~t_0?r~
z1*WVq%SBsEc|(6?PqNd)Fd1!+ju^ji3}c#>Ohew9XU)q}WeXY~_m+0_+Wzv<Drhz6
zis4h;UP*&6x4@|-k1W9q<_d~s0jT$&sl9-D%$~&X@F`GFpovJZOjBhJy|erhu5EAo
zI&D*KZ9`D{Y!VmUY-LxfLBzKdpg^oT*lvI{R-*WG?|@a`ezsxp|C@~^L)L%r!<FZR
zp{AG_10BNL%iQN?5mycshIF+?qEweQN_h)1ZD&5GVQEv?B$~%z9-Jo!kHgv`-6tiA
z-B>-GlU#H+L6nYSgj<hRmsb1UtopXh&SwKtRDVU+uFuYP<twO)_v{FBs7qORe<6ud
zJ?rAy^VrfvFR~Gw6G8&n5mOS`zT^*UrXkbxxqKsgGQQ=(g9pbnb(~OyVv^HiY8d?W
zK+~tqIs>`zsa_6vGh+FZ9L0J@>!d-5P5Ov>DP>`+3<;Za`@x;qu-j|nwf4Tj@EEQi
zc)HD8t=P1+Y)O{6{=9^?r(Py=^$*8|<=q(397veX%ig(P??$U)(f8c6ye+^!Mn2~U
zfOZRA%j1+1GSV;Bi7J(rGYfb8(B;vg=~zNxA;}XhEZ60CTg+AQP<zN9L`CV^@Yg1t
zZge`Oi5c;oft=%$`9Dx&5l=%}U;1_P?#P^*8B_C~P*5~k`D~<S86G(#2y&gQ0fRHm
zI`6HY*6p^w?P*(4B!{0^&{*<mWe!E*328(tm{*kPVY)2YMcK;{NTh6Zs{mBw&KAXO
zQk72-VpkL;U*JC=wW!&jkImHQvwLFn$b43tCi+=sOiknu(zb+3w`{gw<g!k2=r%gf
zGA67;oSmb6p;uetL~LK9goGS&Wlcgj^cI%|BGybQ_V9T+ztDbi&Oty%4?|UuF@)v7
zG(reXJUga?$`;d_SGN_`r_Oy7H}d9pGJ4WnD7b{?y(DLf;@6+zK{OrfpWtfYLYoIq
zlA#*~!{jmo6(TyGB$j>d1xYm>nf(np(D)T*wcC=_LrN>`Boopf9c-_^;k%y|eDnV4
z{s}$`!rX4hVJLP^!kGJa)>dN15X@(<jeB}oaa0Dae#7MSPCtDsS4H*aJIU-7id06r
zfuW)fynn4lq8%nK6QdKIKX{0_9wk+*DQ{PhkK?_kEi5|EHDS+WuLn)(|B}<cFK)hv
za`H1=J<fYbp~8f@LZ6LV?t9x1%ybLy$`R<9NZA@IE5;-GL+C%BG=GwiPG<>XOr<+?
z)his?yU2-(ogpi!WY#b4ubC$T$yjhC>x};63$A+hq(_=I#!HP_`PY|xD0!=xI#!E|
z4z}O|5nsyWc;aan$I;dD?|~30d)sxJXDdq=8(v7OQY93suN8z30g(E3X>r?C<(A-e
zm|S+r_EmP9c|6O6)_OvqJl+wXVfa?s%~7$e?n;AHrrw_EN#M0-SA+rz4J<<|5wOmt
zzd+`t?+eV^!?4pD#ZbGw*~9Vz`K9u_H&ZN-ZsuY=?|9hMk*i>KqnV_H%l7^GDA8sm
z_ss|qOnuJ@zW7M`u+0VwhSa781e6Zcb9o-G=96tN5n|pnC)7A?L@F{f=hFPLAEU3(
zr(VPSvASt3`jp9nzeP5Fc=KWLkO2lJ32{5rW94nOCTP~nfd0ytja!@|P^vR#==>xH
z>D7LHLXzAxN5fk)KvD7%TMlXS`i{hcnJV~IIz__Of|IX1YE3vFReJ-}lvaSiGwC05
z^tL*R?WD;fpcyBCUnV-o)``F1!F7w6y7yB8W(_2CpQ6z?XMH(Bn!aq*ZZE`slUn6A
zxJagI=>$?nVH`(VDU57`{T8@bL4rM0ZZhXn&<ect`XU(!bS>&CYhh7lvO*&q6lm))
zZ|{p{-XhgQO_W0oP8o$yR&+vvbE#~f<@woExyl2+kMGICt}1@qEvd=$a2xdk>rcui
z65X{55xBF1pfbxLze}lJ@fh7t(}}WsJIFCUWzz5mpOcZ#CFDUNE5gxKb`<ADHkMl&
z&q@+r*zM#TTJlo*r;OXXe!e?VaY@y(q}KD75s+~TTx5KE_$r$O`ZIs0HOH?k@F40t
zDWW*VsB3i=Jc3G`1P6#a4=ADfeS{6&|1PlZv$^oagE;&g<P}K3<#o=r!ltS@RcP?G
z;JD~?j+>hs2f4_)@*(>JD>)7X60{bWOYo|HkT@$ibsyhXrJ-3&^2~Bxd_~!PU>Fq1
z`ZUsS(@qBK(63!-r%;?@W2d^c{gquCK9ePmv|`QCA*&->opD05hljGvNf9dBvQ`Bl
zCyht2HcgI<ONL|K7A5g2c7%y|LMaz&w&YM`d>Efs%pOIR5Q+*bHEZ8dH0dnCe<NXz
zzL+B_v=ZwP_$nc<YA6V9NnZQpp8c(*BnoLw;$%Cit<}h&+^DrEBiov0SXyd#XOHQ>
zcgwp=e=L0jZGAej$B!wRUoG=BQ{y0Gyv&V|Forve0`-WoT@gdU*{2E+`8(qw9~{!N
zab=C8)v0o&f=o!ayGhSuZ{Vmg7%lw*JDgya=+oL=Z<Dg<9zVfPKv}~Q%@>knn5}^=
zh8u#J%i#CDXD;=IolrZ2@>f-p<aAO>KK8-9{C_FT1Qo3NEg@er7nf1{rZJ0Pd3?>}
zhxWrStC4a=F^$1+Wf1hO9bdU-RZl(Yv&aP@i>aLzW{abZGU8Tm$Z{h}Zu!BO!$wiG
ziaub;CnaC&t#E$oC4O+WT-fjft(hz#bpcyWyZrYhN2_~!47dFsR6`I5X#+WUZR+Jp
z$*|2@&Kr7MM?H!+OO=`1>ZDC~rvQ+#;bFg7oPfm$upJ#SyeWq6<P&28`$hizq23@H
zK4&ft?JLj(!*$X6?4rEO_A<NZnuXcV1rcsbdb2MinMX1sCs)!<ik8<3v<5e?9$zz1
z6ov8#amALNs*Mcq$7@+FEu*N9?QE&5+m*A2+gL}`LQbPeAG$O$K+?tv#AVsLoq^ac
z=RyZl`{7qTORaLD6+2<K;`q{vLP+j|nwyH<BN<NJpI<Ux!3{uV!`BYDH93MxA#j?O
z@U3}9znZHT|L*Mza*DpS(Q)_biAqi>)`L~H|B<!U!E_G&n3~@(W*MIUY$nWaf`P0I
zb}yxNmkO1R+0j5-NkVuEP<hP06qYy`2?enm?LHt>lO1G>0_%eB%}l@avGOV2zUkt#
zYOvL$AEtl!gE0gP4SQe1SuDqzo`hU?cq$*`;kqtAJTPCS>t@^$v_q_~o4;c9H^KK&
zbB|Of=?8YpG{bh5e}N<t8EgjjLc>g|wIuBH!K-CNvoXaVQ5sE<R=zYjK79G0&zxMA
z($^nN3H3o^?nf(-gJv2Y=^N6L;f6!>?$(N)3O3&e?-3Up8raeAxU{zebGN<fU#ns3
z{8*p^%e?cMvhoo_5=6_&cf+XX_yErXhy3N~fpzO0-LZBr#3!wp$7FR&n_q^LQcq`N
z!p5OqV5drt8z`Af(1!W0e?rT}F{njJs>%S#1y(;2^Hdb;0SZv8>5VHrH+*(zERetp
z0$pj)uNLzzSH-TOLALOlIBmL7Q-i<<x7N_S1kl7pp+okX+;-tODRES$Y+$gDGg3`z
z=EM+FQ9R4Im)8e<EvV{fSD)-qU3zdMi>!>D63ZpNk=qEyzJU8ya$blSg%qJLEkzOt
zY!+HlOa&%mlJQ_dq9y+;RQ*>=?stnB7R>l0`m*s8Mj;fFiBSQ=h-=RZ#ZdW|EdS0u
zkC$LcGK6KZPYKatJ0n^548=LkwZbcU3C0N+^eIKj669PqhDxmE)e1vqmkee9Aw=~~
z8E1(xv;4f~SHFi?fT~XFE;P;tu~=N-$%}j#)Ooo9jJG;st){ga@soi4c_%GFDm14F
zu<K_83%KZZg3W-sIo@7M-Se*Km8T_FxY>vJ%5|3B?^WN;YKURre@nV*((tMQzw&Zh
zlJYQ<kK7^p7MAeHK+X}Q*EeN!;lKtinarWL=&8x+*IAM>@SqANP)v$?dsH)@+^9RT
zvQICW|M>7xegB6yR_-|aS5xCJ2F*NK7=yJ)>Yf8$)9#5+m`>{^t;&4lrM9=!GX}`$
ztpl$n^@B)YJ`s#aU*iABV)-c2$=5;v1)fvC+%#M5_GJg|n%;PccDpU#qQ@*rc|4D`
zVa~jvTw$P@1!0(y^x4n^^4tetIKRU6MBP$8hB?@#((B!GEAN><lRA-*-S@+tJsXMP
zefa_p22HZD_&~mIO^L*1xv@CaxlP-~k&xWd6V&d?%ls=CE-t6<%Ii{{TR*TnrWt>Y
zYCF=S8$R-*Or?(%bg0KdLKz1u5LWM}s2@y8c%l=gR$tm_RWth6uZAGxn{Ow9hyDu{
zN?m8ibXK|>XS0Y`h-sg0VF{lrZroW(GqhyKviPO8Rj^XAM3dpnsS9{#V`<r$yb1P=
zls;%6KEU;vn*?iBUyDJOqfW0^ffH-fSgl2IULY_kv(~)&=+wN}(WD~eMEF59jru4t
z?^0Fl4thog==Is3LO>Y;`p+}Sulu#b6$ZT>)@KqZ*9t6DS=PKjc6%k5kpNE%GG??=
zcm_n1>M^2gs6Oq02SpkTNOuy$^OoVUB#q%bKQHKjd|hK^$zcN#X&N~s0yP^u&QD=3
z<3Qypn8NgS<5Rj%9&_kaGu%5cmMczbUv=JsYFSg#gt&0WUTmQvx|X@GJ>XX*td8qv
zN<?J4Fb(*Kipvw?pLj?2HAZ#G21jdRm6D2$E$$QKHaUxKXRJr_$veW(ZjIM5lJ=n~
z%{B4RL3}r73?I{rnYr@tO$2vz$)<!@YT=7+%*S-Po|}iaZI_Bxe?7E`9{kiQS$gLE
zc$6^mledf@LmSQ4;EdS6K-#LrrILfGbarNQpNA9DZi0Pm7pN>deQz0Qz9qGE?*t(d
z;mz_rC%d1X=XqpF^Z1w?*ma6R3<_8nX5#mweq`vhI;62<r$p?n)C>K>bE)o=*yT*(
zON0sM9%jSS#=gLHG9%5-m{CWJxsgFBFK>9NY|lTo(XwkDmq|ySiJiz=zSI}%8)a?c
zi}%%Ywm!rch8y4)QEGKeM~2o^cB*A(laVK=)S7L;ZmMT~RW8@bqHs_0^OhJ|Nf>$m
zx+Eo*^znJW_Pl5NOIAH1WVyS74Z%aC$>}=Ojo7yQcqagH<PCpNH85if&&E<U-a`e)
z4c*0H8ztKY()PY94nzEdSDBK9!|PUJeKcRmd3$N#eakf^ND0G_^uOO^Ri_YO%<_y9
z!(cdCVAS1~kUEeelAvL#@l{e-&S%??nz5^<*-IdmV3CR)LsRg)iBA<>ePp?cvI`pU
z!h!@@hi8F{ya`P=JYk$N*{Xb5NPiUx@&{#nitKvJC&KXKgIKxSW?jfJ?@Su+?|<`l
zagLQcl4Ay&2B}^JfSntwohh~iBi;Rzg{jS#o9q;YRAJ;7_Er*M`n$GS8B!zPv`EV^
zTqE0HU=pNQ?D?ecH8N^x$M(74WQy@Q*8u^0R7FuPb9E#!;z_{&Po@nf%H~WEaI`Z1
z!TL@^-Vc=driq^R`e-{8SKa=|JLWmFVfy0qPWrqC?RvVi+hi}h8)hRi7}#88rZnly
zF^6`(ZsN@}9+&GIa?+elyk&dsR92t@;$u|h10-r0ZopRweda56Z?SG8x)0+e10(ns
zJ{>c=9@N}S)aViwe@{li{gP!H#HKbdiDDtKhCFo;5zVs&nfZ>)PR<Wk`{?fQLpvh$
z_FAjB=Pbt66DS`u7pi90#v)cBYh8!GfX*>6(0mLOWEJ0~JFMGmsafyUMn<l{@@-tU
z57B?Xj{KDyVz#w)@M|XRf7&vFuNQox=&iTiAknD@2`j*xE1GFFrR|EGaOqf!voEiU
zy-uGn>^$cXR@;#I-^Rict8lyJF?vaH_4-3zA@T0Wj}ur#ipaI6qdYJ~#>a}l-sVpt
z=wiRs18%l)+mE}k7bi5}3(4IAR4Che`>as4XQtnU@jL+dYTh8pb<OoUQKFVUg**F^
zjrJP>ssus57F>aHtiftJ%#3jC`3Ddb4!3am2saLcz&pYkTtwett17kk+ae|m{S8jY
zoLAnQi>CUN)lsvEaK;ndeUS}oZedI2Fr_ckDff_q(psH?=<&(hwXWe%1yqY6R(_|s
zi5@-ZoyocRkdN`HNCfmNX}a60(m!DG*!_018GvA^T&M7<EWSZ*IEO*4N_YdKQ`ET<
zHcUEAUkLhXXJlycVXpK51C>qKvA4!|ZaIT?U=an=^`brGqm9M#lwsj^f^OW6kDefD
z1r|q3*(i46#4lv&j@)YciK*wh=do7tP_4OkPRw*q+(b&t(SV$^b8(~vVNi}xHRR|f
zbtzdsh{G5(zV<Qqa(@1_bUykj6Yf?)-HL%C4Q=YIPgCdJTnQ7>abq&^DKqwurOyd=
zqb^00wZ<g>$f3{C%rfSLFP76~8EZ=@c5BJ;+)QfUUj-)XDD8nuXA8q2P8<4-$R8<|
z_e;m$?D-csIZnrI9Y{@pSJ0weiKB9pSfRxn(v;ZR(wLX9I2b<K!8PnU<-qCZrcqBf
z&6)Jph-L!HJ?_qd?VEa!nVIq8TS#X9MzKVJgm;i}Kan`tirgiZnasx*WE#&tA?j@m
z-NM@Hb;to47#ufnD`Yzc%;SJD^HZ0w=Tqx9K_WgrU2|xIUEdvReXlXiz|Xdq(xGWE
zxR%eAK-|l{iv>~wJEb>H<#eg-2C6*8OKB=6dK;s0&NQh;UkakW*fbBz_y#jOZH@8&
zf&6JNYCuRKFo(s;N;kLYO03dCjqpuEkB}~M=QaD<%S=Yv$NK+FGoJ3wEb2*8i<#c-
z)e168Y*<%U<uG*zmD&JGwQzZIxx_d{*7epi^q@jX)lteMSNT3XwZv&=P%d_~$miEA
znJ*8UR#f=l74;`DE~{pz&!y}aRl^alTOrk4J$;XGo?K!$!(8}i*FIpxH+Ay6LBX_4
z^2Scf7+=zaHyl=;?c-)wyV`X14ov5T+G;qlDFKu17XUx9*W)y9?LhEa(j@;Qa%Rl<
z!UUq!r``Hnty>BnAaKnXK#64hGzo(D$68gQuG872zPx!q$AOL(4|nsG7-|8OWXfBR
z$#?pYwBE>Qy!80dcrLVP2Oh|Y^%z}$^&h$GiD4zYNP*Wy%ONz-6!^{1XMH+Le4t#<
zC*?t!e&&C<-xF6aTCiAzRy{b_9qF;$f%d3aj3>oo|48dxGk0*D=rjP$_7uKCx5#~8
zZbiXMh_Qw%igHhDkNk38zFXgi&*4oTrKWTz?AGY2aujE|^#d4(*uvz~ED7#g<cb~f
z^>mHxws=wIqJV)Vf@o+OxQmRM)vu@Kb^%58Y?L@SJw*h7Z*eME5i|WTVMJg1L!+a$
z^$JOz#ufpb&$vPtE}Qu3nLO^$iz{^XW(POxVKG!A(8?n)Wkk9%{=LLXCMTUmVrhO~
zU}rc(761G=gjqwhqI+-b4SSH2TCaR{q+4;M_LbTk3%XGc+yzuX4SyMS?Y)){3B2qp
z++pXZJBbtYC4TI?fTA6{%D>VcYO?c921QybGxaWADnFQB)YR$RxgR+U#CBC_p=RXb
zfWwD|v#VX}gQ|`t_Z*3A<G&q%)$JF1I})T&I=B6qrJibh2vBY=;#%D7Hy`1iyctKD
zmhPtweH6M-m@{=?sy=zcJcjTij0juyJ_p1fl2ZbNyC1JxWV}z??S<>yp$yMw$9j!$
zeA<gxMQ45hrAh@FG7hjXo!|ve)gwkjyKCW&m4!@<$65rdKp)8Y&I{F5B>RB+HS}Fq
zmCYwCTX{}RujKZ2P-A?u<J#{9A5}N@m*KO_CO$Oi6QTk-6}sl(>kE~5!yJ`c;mg@K
zu(orHg)SISyneteJ}XjJ?kjf7W5s40R9#s>h!9(JYnQ^5E|ty07T|%l<wi?KglOV{
zQC0Ses~uS(f-o}oZI>-fZ;^K%sV5udB-2@&I3b;IBLXLb-xe`C;Os1CPYdYo-9Q`_
z^X9xKp6bDEJBi6kx5^M$?xznSl)r}8V_$8!L;)j_<W1Y}7uNCqTNSRGlq%b8Kimzi
zlU1a#Ei9YRs&}X+94x`9?luFc>7P1OGQq?P!l%#d&GhyCkVu)uecL6bI%2N-?N&4b
zAOeS7OQLTsM@P$jACsF*d~3D#lS0pOQfqD#D_F){!k@3++$3uxzca5ex$o?X5%3s`
zKTSJMuiv}3|M>nYqGC=iJFW5G_Is^G4Ce!X=*@FCC-3r8dm7C?%{cOsEnNA>r{dSL
z*-!9sN`)U~%||rO+;4CjG4u3oko>+>W~rnsU9`wXhcQ+%N+KLWtc^|v?hev6l3}Wa
z!8A*q;*nyZEH;JX-&oNSB$Ww`4#bU4+)J;-tWuaWF5Oe$WY2azNsL{V!Wu;arGvIS
zbg6UCFt%1Vhp;;?_SxK>85J5pMPMT1DJIapMZi=%?`uPzTY9_1>b*+a<U3So439k3
z)ntb@N$+3k&#P#--{>1s#6c@nOPAx>=M9!Vse44FRw0O16dk`EWZS=fghMde3Urr|
z+a#(EcI7lanx>S-O7Sa6fXZgEo+}DHUSAfVI+*>=IQB5PX~Qb8b<CSY>o}nlwjvww
ztg~x;NzRo@UEOwF{}}=-<v@p7f7|?$iI+xWdsNe!)gWA%eE7a3hhleRF!soOO!Cd4
z!b?xF0U9Dqyt=a9+=zCeFkxPHzQyuJ=OLO9P*<W}Hh@p*UJLBDLsfA~kbUiOr2}&d
z4}ol#P?7FvO>7>V>Z=IozKwtDKI9!?u+wjw3q?VswSxbQ8=-~~je4F;H)q)N-3;g!
zi?Mv(xuAB4f>shs5Kyoc;3%S6Di%OAw#j+v+22a>8AK9rvz)ey!}|oZQOp3W(OX9n
zY;A;<(M`V&gwy`fld|nA*cvOkm_SUGi><;V@~UD*J^v0k{Euo6*MU>zw<ie|?G<-0
z$;ERNDtte4W?D$?vC=bvaGN`*o&<c?LEV-SropGEUYftXRZ<x-Q<;K&_K&c6Pmq5B
zA&?{119c)0lrv;@Jvg@X%Ug)cZhu#W(09j2wO><vNI;tj8(*N3c4aQ@ku2yY@|{?r
zBNL6;E?a)9Z%=jP(^k!#PF5mj>WUCt=SZu=l2^R{0R2*_(d{PJ$)DhcqqP;B>9KNE
zoV=s+*WJnfn|j69^zDadUrO3w$_h5ac%MC*C`oS|)YQ<`)^~2(6BB<eMsXVJcBH%G
zWn6H`D~FeJ>*>FFQkt!v#>Kx>Tgntd>sum=7)ss#nA{S6!k}c6=iJnImar9$7SPOH
zqevS?tsHG%XR`DhmE!YbKhE531O}tA4t-S`Bo&9be$EbmOSx8Tn4-)`ipo`y(jVj4
z13lV0S7P9kKM?h&Ps`Pjf;4*Z6{iw;iBuZiPUpD(#Wyp3<7=B|BP(jDJJNW~Ikyc5
zq?g#D-1Buc(pLB*PA^5<u&6bpB?Fh~P(^=IZ`6Es4!-_=H!x-_qz&NcZY7Dpa!jN)
zB)ACIdNOI_ivRpL#r)EoEFOgV)v>dDe4Ae!L81Dl)dzFSacvKE9!5n6l+UFb-ImS*
zLUuf<CUd;Da5CD8pLwL&0@`BbVIuBtEm2I{h7{dsDOp77Nkx?=zhzG%s@6hK0m+g?
z&w!YD!{|QmZ9-e$QL!bn{Aiv%1@AuV7P?{-#ImbA9!TxGZV4S$wc1YWaP~q?({J3Z
zS<mvb&`yGRH;j0l5Y1IRO;T(DVNE#|R-xa*s_6+5xWq8(Uujv&v8z!Ka7V-G-iR{`
zAgzrnC6t<3Ds+@qCj+S{j>(-BFXbo6uH%D#DSu)me9_?Z)!Tj>DF+P8q?-9CJsD;*
z;CQ6_rE5gVDI)6@UzFcehZS`a2Vz3R)K8GM(@K9qd|+8yO<Cg;6iX|=WFl4dfRczS
z{VTLi96gUY&u_8*?XHkW3|G|tb!*G-dcw#r87iBZ38u0!bnnFq3AUdb;}3L#T2i;J
z+FP8j5nCVpu5Lmp)_VW(?}0!j5oDLeivEh}FO`MJ8@ri}W7`c5LmJ?}kPL4egbD*;
zndIbs6ZM&0ktC;|(PH+3c9ZBcC0B$St-A~XS<P&;pFR%WhHR*&Fh58MkRsZGjym{^
zD)caqnR5r~8aZFn7+9phlB)Sg16h?Lae|Ca-G2MpSKRt6{2bF7-^EL(0DVd_(}Ag~
zI7muMs}%l-x+;zI7@oDa*o5vpgkkt-Pqdm!1dyDgRrKRKRlZhduM=yQ3*NB_1xRj-
z-PMK7p&ewXw_5JoOaEM%)5wfg1??;B5pLC6{whfN8_=yRrMuqVYFw0Qh$1#EXh<$>
zY1gt7{S2M6GA5hkZ1oJREh2I-rP-TOHD~`q*$_2XG+apoh%K=?XZ}=d^GwqT3fiJ!
zLj8^W!4AQ79sBQLDO8=5=TSm&eqy&DGcK17Om&xhC0*j!sg&Pq^*Y%euMLyKmI`aM
zKQ+XelCL|dDGFp9P&^fFqprSST+2^66=<X>w{(^{mSy~30!^Ze^>K4P?B7`&N--U$
z@TQP*TXOb^+%Jp!A$|WRskR-}eM^>cTP${$H`NM!Rvwu}7~E-rFv_Byn`l0RMECf*
zqZjUfG&{JnSIqr#Vh1ji&Q*Cw+hyry>CB1T_@5O-UO=K3lJZP{OY{_Mqh}?pLw^H>
z41)YM)+)5&k!_KC|0|^iH%fX~U}D{j;B5h;3)TbL{pzN<l@s}!?D7h-+1gq*F8$cu
zR1_<sF+#IWVS(vC2_#(=vl`HJspSj&yAomezr-1Z;A}_v7nbf#Sy(zKxZ%BU=<(6)
zMfi#Cp<+R2=>?wt3|Z>XC5hZ(8tor1XO?L>og^Vw&M7ndI=>4hY8%WAfeYO>8BZo7
z?e<`{v4Srf*L^MG*}&MYjzQI+BfA3X+yldt(R<f5YUiWUZ<$;Y4r}GFIUG1#UwMPj
zp4MF{43zf;godl!Iw8;(=uQkIG%(z^o7KA7e2w7T^YvFYJB6vsce0#mB8c>`!}4q6
zAkkBSYV(W}Y!&AU@~$_yb6O5O_H&O@yUb_rhNYuTgIa6ZN#k~VzhjbZ$pLqg8}>}F
z0YyWd8Rnl?l|@@FYbYY7nDx9qbWaKyk!ti+aRj!WDu18EprmlNT6pe4V4CmY*ORE<
zT7L47obcQVNO{0jquV7;$sa_Me1uS8kZ@@d5P~gVyhG|}GglZHx|w_+t?J>l7Rf0e
zbnGUR4WIBxZ-rm&4cOP-Bv!3{s{4gN@*SM9E-?GCxdM8SM2O2(roS}7h`q1^5$=g3
z!x>D2BsNB*Gt4<<gx%Vk);$CF%Z*d&*Vx9`Bq}Pe?=|gutMTpnK)?x?A-;n3f35l5
zaK4#9BhGMiF%;;0s<xZ`d7T3ru_~q*_lcYFJ!jqUvr>6QME4Jrnu!6$pSAg2iDw{5
z#P@+<V1`7J{^3_H+l%W1-w4(a@G8tqlC!w<)TwFd@#0n(0I=jJM*_%2#^t?fr{5a^
zo8{!5adx~781vh!8F{trSGEL^n+d5|9fin`kr3edw2!8^MbHTEVm7T)kampsvj7Da
zi23^qxQgzOiPI`~R^{QbF?M4i>yaSkCo)Kxm}c!`fD>(@9olLybU9f&sr1RBC(0UJ
zY-sn_nH1u>!C88bVOy<o7>0^axxCejkki`WQyqk{;4+H+)+1(*W}nQAmyVB~ZCY7?
z4<yPo>SfK>7S6<uem;1dJe&O(`^YkF25@$dkd6M2s)*G1DY*4UNe7A&cxX?`ypr02
z@U<vOR59p%7)>C}A1L%11Sxt5bJ<$~2C3z!*qGV*EveWWs6+lSOx#QvkNEk}Y|m<c
zeEm}pN-SW5)n+N4$ha+6&}pUNkYR}naDos8dBwF8_rh;Y7{b6AH{Z8JQ)YXBHiw*Z
z9=*Dn*cYSQ{iMUWuB|i`+DZ<u=`z3flS-qq7?%oVS#hW99xf?TZTH(K+cL$J?WwUm
zg)QRh<B6Az&$8T-4ipU-^Xs>s6`%8Q*y!3nN?)BKZ@t%Qe_M4d#`@}(=AhzRa`E8?
zY|Wvxk3zHs)8260=uh2IvC1I9Y@$b<d~OCFiWD=&<DZb6o7GK93)Un7!jIY0GzXkh
zUAP=|=rFJ_>MkKL-s@0WO1=cB0&^_!2;TyqtTo>N_uz_lO}}yH2SAHba@p9wA%8CO
zVqO?rft(+qpYlw7GZx>qy3@to{E-8lb@y!{RQH0W0i7e{<%Bd1M<nve02=fjN7wzM
z(^%OtIdNBhCQ-05j=!m*{&oYfm+v{I))!xaA7>eCUC?V}ZnlOR$6pUE{c*ofA#p-|
zr*WY1{`*va&S%$CO5JzA1<3wLmNtP6!V+_xCQA4rqX#lYV)=94@;AJ06^UuVt*AjM
zs56I|ds5!k1oMlW<=tCaue5`I9Tw~twou9OAKps8Hi<;;i>`XbZE=!*RyE$5GCQtO
zbX|NS=TK-_O0)hGLTP<lqra&V+g^2sv&_K*A5w>wp*YU^mU8CLik3wB8-{qItF*c}
zRf?w{=~682xdklc4!O*6^JcMgX_UE~Zry+<V?ZI2AE=pEijgeSb?|1b^(Cn@0&b$+
zv-*rBQ%rO1)?>Ml<q_Fzk1Ww=1?zVX;_HiPLYfc<-0x)yV-88}-~QaNuC}poX8BC{
zd&gA2ld54fP;2gNZ+jMHJJUWXZoWX*{fv~Zzw`sU^1BK{nA?kKk?_{+*k*MVTCg~N
zWD(!n6dQj(hcUOwMS|H&4C|BKTR}1Uj0>+{p8H|CN;F^XUs)Ie;_?IzXn+R&nHwAq
zvjHZ@W?%uDlYt-OPyg9la*aco8?{SYIfE_MX}v$0CIkXRD#+^5L=#jEN$=i!m8|ik
zis51plP*J5s;?N<!=Tb2dNHpUxB1ti>RaTGd7T(1yvOW%L|EH8`xS|&<Ag;?I{WkL
zolw)F8do@~tjJ#sJCHSAPatEGDutJJ5WG~`k5%lWyCP!&=p;X!*wUd!-UzIc)5DQw
z6d?1PxyT(buvYVjpK0O9;Nuqvg$da>lB%_in4K;;6oyi4E<dN))Nkr(LXf!Z><E^`
zD$UUF)aULL9#9JFv~r=Cc~<?c$%^aA1wG??3VS5MfI+NF$(q=yCG?H?KI7HFBc9k6
zQ27beOFGV*4tTm@!0@mm_&rQI?USUzYt0LIoqY?p^SiN{k$DnF0Xk>cq-|Z?#Jy|1
zI9Wz?zbmeGwN?5obZ+{N27>lUZ(Z20PmjW8*^=#jAD*9`+KZkgWCPu*<tSd36BxI^
z$awyNyBI2?`f2mLw7SGr70X(Y<;mD@l#ul=*EFC$AIN6AmG)r<X+85mg#yit^K_hw
z41>9wOY}g9GwFR(2cMGPv=)8<SqKgEvX^HuO9aZdnCNu+0z5Yj2x(`e?x)p}w8rv~
z$LYMi$3RcN6t(cMOpT?PC7YJIJrAS`1FpBc!q;%UB}L9_HX*+Q8JD=N<_M_fBIO7G
zdzHs;c;jw}l+eeS-to)=I@(eEJBLd%&99y|A<8n&&M>vh$B;CSdh<pigFbz2d6b3p
z6>!SwYN&9~X!?!;b3ams%^f~3^=hKW*I3wFolPQ01myiCpZ(KZqVt!xt5ALj6IQiP
z5kMB+nrC5reJI+sSaeI`-c69F6uQz<+7DSS|4fbpy<J)2WhJS=JEW?(+D+IkR;cRz
zXvB=|B20aNbH@gP6ZDJ57h8JkVH7J-S@Gwo*|eqX3*$T2D`aQb{bDNX)w&PWAVFiu
zpxprsUbtqey+zo+Se+@GHUilOW|mlQ_N;tezCf)WzyhGJ0Y#N@D%5m?TKIH!Gtk(T
zg4Z?CzV~+lzBQd4-py>mC0R+eOdBQh4F-_Ee3a`rSk9hW@WY|K{VYUdGqfu38q9dZ
z@HRgsGO=+Oo-?weliNpdWdQl}^+KYN{#)&&z-}s9DE(ZC?_W`qK*=#*m&1<dkt=Cl
z7O+J~L)m1;B6vG57>S6a4~LHo6Di*P=hue^QoI8|JlExmQ%z2$o@Hmu5A;f!X%hq^
zfuLye?W#gIyb9rdVs?NWT7<sVmn|ilAj<rSzhKSpb|f|X$pjt)#;~=Ix|Roy>Er9h
z$<StmeZPr`!g%1{+BF2Pala}t!UNE@N@tT{J1WPTaHjcKd@#_y!frG;meb31Z<3>p
z@*p6Skv~uF+sI`W{FArNe#KQH2FM2dAWG1v#o<2m$SESb;0j^XQc>RkrMkQ=r=Q4N
zfekVEZA+fc6lI1G0G!^J0z7?^#eT<3%GG`Y9ho-@uT4MWWag(l20352bLM8tFUZHY
zaoge*^s~@=#4Ds*??a)+O_}0n&`NxlRwTv)8PB$#`ECQC#(rkt8K<DuT<;l|j(8X^
z_6`VM{>XNoCoW875AxmQ9%43R(WyOtPIw2V&)3p805lPn?XSmi-$pwYiUGa2S;M$L
zUbBmzZ?7i6{tCg?09YaS1Arg!HDpE}glL%m&TAFA9x=yA{dq6{>g%UMQa5tY$q_*L
zNvK-nyS<;=9)5U-8+u@LPyANY>TEW^f)xBIEHM5cH?DP4^Dfkg`g1QB04DnPMyF7>
zg93srW6$Ur?}#)<9Y4N$?aSgRLlF8E&fPRR>WP)cen$YHg2ko6K(A|nE15xGd=^ts
z$l)5+Sw|Y%y#`j)*X_*0wJJjM+8-BJ@-ZAi)BzY4w)*8v;yMpm?A`Ue!8AF4yyo@S
zZAEE@ao>L38xG*1{d)&S=n?QcXhol`W|_KKEqHs)eSnDUZhOsH&`!PdjXe2#H9?wt
zw#R$LeHehg4FHzErJp|qmdnJiedCw-t0k_|>e$5Yw&D574IkS*hak@@kS>`{%H81D
zeZH)$E&|wV;xkVzWaZpt?oU+RJuAhnqU@$gai*ZAF(%+8nJb>YDYh}((4+DujX;K<
zW&l&9<4fC<^v6*~P3ks}lTQlg*ttR;ysxKIjElO~>UfI%Jde}rs4Ov}*!ix1{OaES
z&v`ONlh<Y`9Q<IK&2sKb5%++bw=h+euaP^^=m4MtomJy`rl&L$%X5+@fzaEcDjcuz
zc?j?oIQYI>*N+`3Sx>gNquc->^ucq8*2NY|Mc}~}u)TjA-aKkOC3#F?1b*y?)iZe8
zl0hh766e1T(*KjgC<&Rtv97W+CUFN=;RUkI%>H}4^$VM$@fr#)JwLd(La_gNJOFNU
z4)`vY9VzEuAj3d_u+vYXjI1omXnE~_Fm4haUyc6{N+0;=Qvy!l)&A!+uT8x;LKOno
z`nG;t3XU|QEOutJODJLli`JeOG$ypp)}%}{wjw9Pufu)@1X>w_i-U3k64xgFtc`lJ
z;2+P$U>oq;RPvmgg8*+k#LB^z`tdsAFWCUttc!HO<%SH;vE*UVAck%(O0XAdttB=p
zZ(z;yd*>(e&eH39!LIalgu2PNVM&rtg4WpHY=5`cMr~l)<Q1lS?GtbMPJW)OZ~ju-
z<oVC-p4qfZWPF&N)peNXtn^*Pwfd~1L2xWTz#di!Mx#d~w>OoC&+iWX?ypzi9yhqU
z4UE&6?w7{_EC=qlO|091j(7?Cv-Mf(^GhFA(Q20;_C$!xGK?Jr@b8d8LRr*0z(dn+
zb@=CcUzqjeE;WID+8I+xGjS8;Ope5Cf78=$Rr#8u-86*HE}tiR^yrm`1Bw`Nql-CR
zgy!Asux2bt@t&PapgsfmF%NQ&qM61@*DwmOj+H3McFq}N8Eoh5uU;~s)EwQH!|koa
zxoFH@4h{l$k7zwj-LzHa3<vRtY&5owD&};p|6*d%3xe#vZ5{KydhbfQV!~RVfGvP9
zBx9^@BK}LPI-u0ey^!5^QK`P#h@m}})h;BKI<!2D;t9d(ubx!j4mW9buLJtJH~jhM
zg9^(RZiL?ZW4u*ug39inS_Ni9K>Yv!Ce~kcVxO1pxqT3ZN{;-QN{U~OIB?YL2*wm}
zg%9BpFGI3&Y>zEAe-qjNeny)Yk{BP}xbRIc2>`qCk2^31Gd91v-P?FH>!?xWisiX9
zfr}Fq|9P0uhwfjodb!sm*sI#=FiE*{)pRq!E@+P*VkH+I5(SsF1h94a1-<|^$qAWj
z$rikt^qYnD_eK1bctc`EO@i`6?3@6Y1W~mjBq*8jq1M{@9It|xehf-oS<kaO&}tL?
z^y|tPyN#U{u2!8=IhmXO$=!WfX><>THkKBU<i@%K&b*_WmVGPt|9kGtydmhrjwRWa
zNpUbPBXP9+mgC|@0x)_y92)+jz8;_u!PMU(MoLP6si*4BZkgg9{SC2p9l&87+?-*i
zG`7eJlkhB{n>O3Qu>e2WNyv#A$D^O=xMasap8g|7KYInZLPtz#Y7Td!>3sjFdcNY^
z^W#WM<xY*uRSDjkDc>i};;S<ZtL{1<3@LTB%+oJk;7I~tx5u+5O@9})I+*Um02Z-V
zYYG3o@m*6Nr7j)HV{a?19-A_}^*mm7;)MNZrdayYM4-H2|8u`{jjDD|uQVGP85Z{S
z)lG2?WsIFp!Lbova)g;eGu>|*HE>@7Kz6o>`Hi`j6sknM@}+-~1fSkEvuZte{vyB;
z$aHfxGpA~R+xDKq%I2z)3gZ`oQWs!&k2b7C)0re@jy<oYQaj(X)8yDWQe|0Pgo&Tc
zYPKRV7(UnbroTy<2^Ond|2&89!T@Bhe&#Y*$9oZHnoE8jrSzm$OFua#7#oj({nbqW
zP-SYRqR9HMuPq6voa|)4jO<C_KTnxq^FNI@p%yMkksX4z(Zej8zlUM{J}(-9JI|Bj
z#s?P&OeFc41l0nJS9`syQ$K4%=K;+1>nzKf;MjoU+3*`Re}@{ny8~w1thLh-SJ~Z$
zVQTFIu#^|3UQV0E&40kXuMX&!cVjYk?b&YLY*e+S>Tmk|bMHTc1(-vRs`zZke0E;M
zCt7BP0@YMq1}hM4n9W*B1V*m?0|gq01#!GI^JRX#3sAJ|%^sh?0X|cY+q5BYj#FDZ
z>cdsQaejZYh^9vIl`3n1kL`~Pab3|8rLLF0!@^*Ngeaz<ATyeb8i>!F0^nsRac?~Z
zAPuv{^9=ZA1!r|4H5;MGmHY>ZT*sMI6!iFyjt;D3%`4jtVw_7KVWV^xe)x4(W&M?^
z<Ni;}&1OuCp=$i60B%~J%I+j|p~jIKMD3sTWah*xb=`TATEE-7Y2Kj_#Wj}uI=lWW
zTvF}=kMbmhzyBzKL)}WBx}-a|S-FNvOPpPvcB9N>(iN@xxU4kaqNQ<Eok^xP$s1~S
zE)0psFh{q{Z=GZXO>YOPg_UnCF)%p?>WiM3tI@^$o@M*{!n;~c+zV;zTc2+;F=h<8
zgtk`w=AX6EPO+UAfb{7tQVaVOTjogYBT98#L7Yh>a_j89PVRhsd$h=r9Q&H;DL4@b
z^899@86h?Gjko8)jOUfwoQM2nQWRI6D32R{1B=})elx?&=%xiV@fWc?Di1TI0q9O@
zD)~_6A+dW`tHFKUfjIA#-$dWPZ|t>r{+KM!VAiQi$-R)jdNHRUGB~z9^8Af|Y^D2M
z0C9R|&mrVM;-vcT&Q^z#=(aiIjjo!Rkkg6F|2jTD0AQMqIns82r$V<6r<FFX(;Hk)
za2qtjT;*!5UmE}Oy)$2!9|7Z4<W32>-1gsm%RS)MZaa_&ia+rApXTSn<Eu5{0QFec
z{_cC17WufbBJa(TC(R^XhQ?&uU-sW?x12BvfoXR;=S|E^0Va6=?;D~MzDU7A@qcdn
z3^Sur7w2TaIt{}PU>F;G6R#OhCro@rM&;mbPoSIsHFiIPf%($v0LYQDNSo&SBk|Kl
zyJd{i3R~<avs)6?p9MS966Dvr{cl}T3=C2|#g6WZ>@)qdAon+?$p8Nw|4a0J=6@;7
z|L<^%M5w>iwyzB2HyV+k+}|&D11PoMG)DdA(O7M<2n~>7#>eLTM-G3Q{~7v~_Y~r-
z#{<>|jNMk&Jv7(<N85QuHMM>1-gBfKsaB+eD2NDX5Rej(CMZak-jps?LXj2}6hx`g
zrAzO<mw+g}_hJBr5FqprS_ru-o^$^1d+!+ce!B1ef(($Iz1LoAuQ}&4e@`r~R3hx(
zU#T;ulKXYx8r6~es<h03O{v%X=-=>(f2UAu{6m1b0AAIlDzs28XXz+f?|&(ZXI!2!
zymcG;4_Sb>vCZc26IcRF2DoGpT8xEGFM3x8NbsQB{eZsuFRHJ#_5){DAV7jNp8<rB
zXfiG#vVcf4K<h*{=(~?984z;QBJ@%({yU#uJG03u0~oUqpd|kOt^bOR>hc7EP16?f
zK;zUO0gbXJVu+pg>l45;dm`=N9I)38;#`^ptnAwVqTUD=r~v0{JmQpw`Nt?D;ow-|
zO;q2F!i;9WYmGvnJA8tc=?=y$dikH#)27XvPgUb*8oz6S%zgXrpZqu@G0Lp-Cljp;
zWub#xmc)T>`^36wAw6bK%Tq_m!?NTF%{!CH)yR&6CrmyK*Js_yz5%w<{;$gQMQ_Q~
zT=^Gg2JYIrgSeoUAA&FA>kmS-Ob_gRSYKvSdVHqLy=<8<5#iV7EJBJW%r4~y_p~nx
z_?I+gY&~sxg_s4~G}fq)@5#PJSKt2dV_&z?Q2V=kJaW#DyJGWlnX;$XTq0w)+b(?Z
z=rb#PpE#G#DFeveIMW0tu`vnV#DOQ768rtT&y?cQ!u7pY(nhI`IGSaW_UHzYf>q(7
z7iWi1Pq!97kh*<u8HwE(RZjD%<X%p`syr~CciKXH@rdU+`iX{rvqYqBMbl@Ph)=hK
zIxaab-|l!h4<`A@6nUR}(KFL~V#!8@fMl($CaBv(uK0>Jy}IC=kPV=O0rLburG~;s
z|2$p`v+f}NZph7?S585UO>$RQ%KN9|CbMYX$rco4-@VrBpDSk984#I9Z+VKod3Crv
zZ{9%NPAS4$DL>mSA`9SbF^0a9-Pxr~^klJPr#_E5iqg$p?S3RjA;m5SMglrl_Sy0_
z@6}!rCL^iG^n&016mukT0{?`i^a5p)Fn9X!OiW0!#rQ|hCOg?O=D^6`sPGylV*f2K
zC$ypwK-$^ljxJapzg%9nF=FcsTM%>VS`QBwG)NANL~bVKv^3s--mDR6d~5|8o<(nw
zbB(`L6NB|6q=??Y@JNQKt-p5aY4AtfEJBcq1m$d;>$j-#2l9`8Cf|FC{@$FuzTh%^
zRE+6JiQ&we+1%f1J8%-}ez|RTE80n>`=s$dv*vZB%Eh;1@+BS2iNaT!O_9dIQc?u&
zbFMT$Qli5$BSz?0Fta2DT(T6^SuH;aPKJ~I3*=n9R=Skb%<T|_`?cabr<rgYA}ac(
z#vOn->7E)2F5_J)yWPaka+@YdM`u38ub#K_XVBoxh<EMvh<*>WB_c0Z>vQyFj96;d
zt`TO66p?-LcL-~_)2eaF_?&cwpZ>?1jM1(BsN|qF8b4O0B#6HlO#`lXF`kd<8M%{n
zu6v7VG-ah^$e#Yr=(N#RP~T>DmixBkSx;W8acSL@2wjaLaZiSmwofeP{MG@H>Q4=h
zGt7c=0hqyxu@L1I!?<`c+SP>8l5yPm?P~F^n~%h~g`9(%y*-1k&Ok(ll9C>}uP1et
zOYlCCCe&PS+%j5jasm`Al>h3&GsLaRQwby)myv%s#jFPYaF_EIuLS)ZN7wPXU!uBK
zT5G;jMei=Gx<z73cm75|J(p1Wl#o24>HVbHtzuC)J@$(kN^#}@8Ek<}qd!{`BDJ&D
z&BR1EvU{xXIMrA8NaCtPF3_FNpleMA3nZ@h)#}%4b*}i&#h@}5Dlc_@^gMCAM9%;3
zGMX>=gaML#w!(H8Gv1${K^(PbqGujR8uY9PzEH(H!~zHymfG+lMhmsA+l@A}f$^s<
zk+{Q~EFY5+<=>J(H1`&W*RgyrD}P6+R5K{H*?FOW^bk`JtzDe$dS9Jatnf!ki{|tl
zc@qlNg@-~RTn`*Q9$XhV-1}L58`0g9R^uQ;yN~+~a2HtvQ3Ak$3nW<RbYpJ{?ak+U
zjMRJ?@&1)-&K5ht=JhExO><j_8;lMGSL0mmlvi_nRFI>%@3WNJ>_CurA*2$Km`apd
zfvc=C+s&t|F;Ow8%_0WUUH9z<NQUVyp?vP~Fos=o!MsM-fkM!j=4pB4w{2kVFDT^<
z3rD$W^T4V06$Q$@f2}`Q4n&x!Q=j^bmCx-?0-mrz$iFWqSL4r|mQxAq6=VI`8S{Ie
zk7SFZ**10*WK#xOA5|5DTuBd}%U-{&B*=eV%LgP~33xSjq9)|<2<*a}w@zc&e5uw4
zgWhDEENB5UtZ$-Al@f8Btz+lAsskbwv2E4071+x6g491>@}O;tK&O|x!bG&}1jH5l
z<J=z9P~|UGayk$^*P0@(=W?p{4lc@k{izQ7GxD=NLq)G2OuDqVRTub4-+2|!c1cA~
zkP6&@M7O4)z1m5!QKoyI=j0yZCUAY}C07U&LjmI^<)GEm0PP?L;(UbAi$@a|HCuAX
zh}E9Y67ybd{&UT(Vxg&#bRSrb0SuhubU!74N>_tdyeMN@H*H2&*SY7xrPf=fwbGjA
zxeoNIP~+lSgaIv4RCv%BgcduyWq47y2TS=xNj7L>(VPX_n}O^6GPxcBgT^Q}8xz?s
z{~{e$HHgq)CIUr~U3_m%b?_DZVY4qDwIkP&oWj+~Y3Yf##%Z|SFMZ<exuf;^uZ9C;
z)KTpwbPGIYD)EDFPd3B>X4k{A2M^!zvKoPXomp*8<XlLb$P{gp*bH~{GmO5w<-Cf8
z1|4bc{rL|>?+hmkOitpa(@I1D8%WdMGf)b{b*G0<$Ek2~i)rYY;M=3tog91CPC&y2
zuud>~ViyE6%vtw2e^E!7t><n=FK5<t$LN#X1zBtF_;Y6-;-eFJY#ha*GtnDj{1tbg
z+}a4b-l1X^S!YD(#@OYzRe|@(d(d7Urr%J-j08tt55WZbGP58w-3RQ@U8-qhivd1X
zg7Ud>ET^S|zS+Mrd?whdhdtz6!Pwua57!eXr5ZqHYPwYlf;EU6jFak&H7TcKM^2H#
z$Pew`X=t)EcPtbp8daQj5sj3#G)8A3&<Foe0MEF<z<bS~hzN!+!rHz}78)gTTEfe?
zW03m(AXnMx%!7J!`Bjf$1D%{T-MDQ&^INy+pihal*?$5%dbh0Z=@17gj+%fl1C<_+
z#~bpw5W8W&)W$eM6YdK_<T!vU2zR%Utj~2cZ>^;sKCj0|m7hXNP2zVg?)=C@7)MhM
zPA=WCI2{A?GuU-(CK_e-`FDcltE_VTy_%ZXIDhM_D1}lO_Glyf1N8lW-o_r_Q_sv0
zaW+jaCCXP22Xcv=OYy6J9jVrsh+-D>9p~vE0Q;4LR6u5YG7a~+o||SkvRiYW&3+R6
z;Z(2qDqhRm#NYc31|1AbWT(1W*IE+WH1znUKy0cb+hLdFfEIqJTZGC$V6YN<P0$&N
zadsXf21Ea3;jS>f^uF&&W&!?+-zFr~4#jD?QgkEO*|h6$!~S$@)oMG$AuCI<7(j2u
ztQW&W)o+ui&g%$0)Izjb3AR>`hicj+QFPXm>G_lcKR2;oDs_LTY?z_H{bLF8;lJy!
zo#0ajtx+f&F0$|^Yt%8~c)oe%52jmAD>9!@M(}pr&{xfE13F<LLz|%y2aN7Zq2$GP
zIoJj@!&0jsv!!q*|Jcjsp??a43?qJ2_Q3$x>!D-Vm|3l*OTNSBb+tpp==rL9e}v}T
zX0%!z{zQ{qWFMa4tTR#hMaxWr%0T?@Yi|BFzizJoaJ#>YbhSCs^xik5^##u<S6ePJ
z`XhTtJ}5%0siJ=HuM*%iu(qa{TJSC|OpOQ0zFXMbTh^uq&4e#rF7(Sw-etu_C-f21
zS;dL`t_l+}UWNazoegfQV80xoB2PR8?s=u?Tl}`H&cw?!2cGUaq8GIL6=xerco_Q8
z6;mnM^c)MdN~ik0qI+R`sKChMy8dz$#r~J0na7mzbMM|z1|R*9cJ5JNSXfQK4z4La
z9pirCzZq0(3fVI8wyH4=JH!A2za>hllzC-T>(ry-T15M^N0j4U@gg@Y)B__mOYb||
z4m4qB9%q6pngQz+vC&VM4a+&f_%`&<_g+F;3Em~MoLSXsh8oq8e;pn$?n^}Mi|`WX
zg}Td3a|eZpKaZxyTmBux=TveM@4e@j$X*nsKgAw}S5^FZMOUa4==zl8Jum4Tsxcqv
zLe6q@ehhExfw%@g7@w)YOHtM3XFD)x%}7En@2~YpipEmFBSr1ab^}y&Ij;-7vf}7n
z2wmqD{@OFh@ArbE;#VYS1NH7UxB9uA>=A*eT9lduafFwt$WX4;i2&1kRZPm{C})<$
zu?xJlkg$7b^^mJ97&)#|8W73X?9)5`#0%ARj?OxkQBnpH5Q#YmKuD6$Id`qAE&sKa
zjX6HSC>VlG)rL8Xft8$Hp(}E)=BLo~5l8d(ot;V9?O8gZJO2*kx$|$`p3!0s7QVmX
zFqgvLWHH)dm-pCqSlCQ<VD;NW6GmtL-uKxa`VYQ%0NzJNo=wlL0%hDcc<w)<xLLb_
z?7e`KZ9^WX{odFPu<7ePN1in#4-1l&t43}d413OkV%yK3wDqn=8+v6M0F<jp`Q)qL
z*8Q4M?&$=!xId7+tE(4f*c$#4oNl`iOM!I`U%ID?sjLApgB4Jr&Ck)|SorKL{}g+|
z)KWZB0X9Yj^E+r|9clj<LcF?V7g@K2PixI7;Tu^QX@5606TDc=2oXI@Io^<LgJ-d#
zfmVPY?S*?_=EN{W$CI5R45gDhep7v?H4pnzn@&COUPZ6gyl~=SKQU}qwhNkwfZn1m
zrCYk6;4_@r+dV6;9Tz<6eY0WgNjPf7MVdx*K5{31t@f*lru*b+;iIP<`HKiP8qefm
z><G<jK~43xAk`iCRLll~K*W6usF_@9D|$RiuUl`Q(zV(92vMT?sCR!Z^eGP*q}k}a
zR6<cp<4Rh5yLuXVXG$9JXIXcc&jA7F9<jXO%r-+=v4HZ_<~hC1{DtwKVQX<D<4T!&
zLuJ28wONI7d8t7033CIq>IhlNWS))uOYRh=NH;&AAX^fiZ5EU>>SH|9%m+KG!S}0~
z;UXwfPN;jRw&o;pW~pU@B;)k#5Om;5c+)~OaceU`Wdjj!d7I|YT0la%F+S?N$HKC6
zHI!}a30y`WfWAk3jCp*zz`2y{U7Mm|$-pg1rvT-}_NL`zzU*vNSlUJG<$Wotd|&Cv
zqqIuw&g0)woP(>hr$|Kx$Pf6$$jOHL!@+044c*nO-@TW47)|EayEF2QMgk*upY^W^
zFbS}&8ph<zjR>`4Df6MXB8a{hFslj1pMmQt=DO5oz)7vQE#?OBdDa%n&t7IP{pXf!
zS94LVt^w1U)Cgr^DhCmg2BQ?hTqyIm#G+1|f!vc>Z0T>3u#Afh03K6p(o0qVRpWFC
zQBnvbNo}sSJNd{_kn))zc$T4{#fEJqL4Ml=J$d*?^kzV$baU8I(_+EZw|mKcG01pf
z2wY6!3KCo3<*Bo5w%Kj>LW4Jw=11QAOe%>n+TI#Ml%#I(w4B>YCY2hqLK7>7eh1mp
zda|X}O^Z^woLLiosglczo8I3M&nO?W9K<c%p;1UxV$j0wYQ|@?EZ+!x(*{%=^D&F*
zT^^v&0kzO;!anPY5t5-dUJL!&kgqC!uIDJQtt{6?W1#lL@4cx@7nR(LEwQHt=SCmJ
z9;7K*K^?nRKu;(TZM}WEd4iSC#(<2#m5sszmiV-A1-l6`PI;h%*n@PM#1QvSqPQzg
z`3~=ZY|K9kVk?tu7j+Id)lt|BBJMWTxg|2m*VLH;v+9J`ded{AD^Y+0xVsklYr+2N
zTZfaD=o#$4r49{t@IQk%vj+E6kAIgj{$QN=U3#SC;<!C3zj@8JnRy4pK)5Dv>Wmsp
zg8amNTaV8dkOLL---ias+Az|{Dt{*<Z;8ghDBFTtzi0bw9=Cr#y^g&!Ewr@`ET8c1
zU9o1SDgz7Su1&8x&9*=YGDxla@80k+7nN6=FIH9=ia<Jxo!nb-6;xO-PHBqUY2#eL
zg&ml;Tm_ga&rY&;HuKASPA5&=+OqT$<$7Kj|5lhD7=YwZLMfE0UeZUYBUjII5gAe?
zt=RU)1Cz-L?&-IFIPnnBmlxw{>_dbc)t8+kdQQOh#ee?O1bSf+tYOlu-TR=_c+Pvn
zbhS~a@X~au<Zt1x$tPT`lIwB)jg7K{U+6eigi$mgj0Z?8NdjxM-sHt{0$j=3pKdK_
zR>f6S2q2+<A0YMFMZH`|FDFg^rzpTRgUmw@4f`3ijk>jVukCXH@!Fk_nZlZ{b^QQy
zpGP!%{Jj3j%|fDTktFDXqGJEYXYRoU04LrtBe)d(Ka$d+AKp`ExmR%)Vh1hPbpZt4
z2U?K5(|oHPz_WLgGGojcy>tAo;Y;Hz*CwAaF^76X*WwvJ*9JAGPBbRA-}jHfDakgE
z)^S;Y{bJ-X#Xna7eEF{b^G{&ZYxqBx+U9hvxJKXN+%r_1bB_g-@{LqnSx-n?=$jw^
z{RE-NkgcVj@vW@8&5XXJhbBhpU#qJ(5f$)3dbx1o^f_tLP3!rOy^R?N220jBm&%G#
z$<j!(#Dt{5_<hW`B^yu5cC|18BQw&UYxr#>O8QaiBI=p*%?W;Wv1~S9(yed{{P#n|
zfJD=;LCWb9M>Xp$WaRR~YZ_mFGwb;~qWs|ZdPt1$-$zs>G;sqP7zv8u-5r#~#l?4$
zYn5=&0f+J@{ro=o6u!}8f!Xjp9a52wTL)}RUTOFPMsh%bAz0x~(0|yyvH~>WWNyb{
z?*A~2y?y|u@#unc>*R9#2dM@2^s0~kDa0$A8-i1JTL4J;8vrRYli6**s}8sHv{#~&
zAtwfne3FmlX;ixLXRwkAB&*kBKCfu@K>k@G5&gj>iBzi4BjbAGd_lIW;r!d8$(>%j
z?DnvsveDjGo2MLfJUxa&EHSwK-@662#CeOj8@_*#aW%>@>|O&h7*<kVtpweVmK>3w
zmJaMhl(ukb`}@&IcY7Xp`XL;6mjfNZN^}z&UBUAEPF@0cC~cfmq{`)zZ$d6<Wx|#I
zF#NQ*FzP@OakL#biF^G4z^^Q_UCbnFzEWQ>UEiH${xl@NX1MQz@ZVhf*`emmamfL!
zN41_IYab^NX<ge$VD~t_7sr8PaP$+d;5Z>b#0iTbQl}`){kW3zZ`VmF;MbhjA#a5)
zA$13{e~9Jm25U8E45a+>NkRTal=9hzuvh&^!*>f|)dq>+p1h9Wy#D8nz_%Mu61UkK
zaKE+}klq4|ZO2ZsYI|4yzEu%+3q!Pz+H4-EC)jHvR2@^aOBF`{#u7hL?;1TjJvS2-
zHl1$s*7QO%x8TQpJakKcA`R0qwV)685||%b2mJAW@8dmECIz0c5`G(-khnowD+_mL
za(gh3whamZ!C5vQV|@)ee&BqBHVZ_P%x7aBFkI-^krRFu$qoR+6ZdgLj`Xqm_r(cn
zhmohzeH>(IKdG-#8PxW&0Kd!sq~1T4Z8dOptM_)?qh)2B$>laM^X0J+Q?epdh_-C@
zlTB9Ocf=zi0pmE&Y%u@EwB)af5Vl@dC;RnDG(T_>QTMQP_ts&j8F<QaWvE<`n`$*L
zcQVMCk81i;yB<kJ%4gwkym>5|Ly3}~L?0W%{o;cA<VdQ6$Au3!d&PcwiSjjilxhiU
z{nchlx(J{|i#kipP>@koKZ%`3aINQ$txyM*^RJwL<}FINbKpqp<?#$Wm9r>XIm703
zp>Hipc-V9xt66ht2IYHkie*vPa=cXCvOl@drFZiEtEOd)zX%_(d$D9(>SjSu7=J1v
z+?`2QSqa)K>Z<~{!up5;d`81)QSnGOrdsS8CM{qvG7}`D-Mk4|0~tlR2Nw+p!KS>X
zwVQ*;?Omq?O;UM81I!JT1>&jyX&h1iL_P6T=2U$P#Ea0EIqbjoekiZ~yypKt*5zg#
zJn<NsklZ4)lVeFEUu!JISacrE|A3)F5WJScQbk)*FT!ZZxIQXQ?(*xUb?;lTejKL?
z{)}f1&!z?zr>qzdAA}F^-BDuL{Zt^{-zU`@@AiIMD5DwI$3gyU*wETp8389e%JDeS
zf%;)pKd(*u+j<Ar&wyo3gBiSO))A)|9rn`*A|J80mB>THN9UR{Xc6^@M*by4Q?b!r
zuF}pgc1_H&VYModC4#?NDOF*4VP5c!0qVb>Xz)cPGdHXh{f+gkY?mB~8?C|*j5PUE
zC6BL3p4&1R`vsxjp9EGA$rdIL8AgW1$(O;aNvk?%Iz*;3;~1KHSH%|iM#fYAo#Y!a
zP}yfxvZ{Oe<a3}H5F7&lZ%h7kH0kqlgY7~L-JR1~A{<FAl{1`%K}@o}cOYKgb68J!
zFa4a^uPLZ8PpnO&W9as9lL`~;NoqC9tv3z3)S1q^9NR(bwT+>kzF%-y1_P#4W+ZP+
z7RbUzpy4rv<;rgm@2As_qT>Ph9ZHqT^S<Z;?=#x-iT>`9NBcHXCNU&eOpd6GEkdi)
zB3=?my|{Dt55{Y1&^X_RlV3J2NiDhG3vfSq_&2y0wm)8fUP<P4%ZFUEMLf_vn?)fD
zXemu{Dho$z+oXb*Yuy`Y3Z<rEM)f-yMQu@H6ywoCEx&Cs!g|6@DJTeT^Mgz;9uG!b
zCZ?vPQaguB)Y>t>sfaIOIh1DW4wBwaIy?NM{J6>IP&ICJUonI^x-t$D_LR@-_?gm`
z_1%RH_0#%^muiLc3I118sR94Cn61I@%UUv!Q(KLTKdXn=Gs+Mz4iwU@PK8Sh(wpV`
z)$H1vYozgqIC>SOyH8v=Eh*0Kc`HU4^OQ#iT8cHtnY{BhHIo^umu4wc3LvBn6y$!{
zjjE5Kh2;6VASNs)#dFwS8dUuGCQYY)XSrKt3`&kFjZB;s$9tRUVuYS<rnbZ_Lh$ap
zh9+XxmW^`dp!2>-S`S-SVPZNfG1waeGm6h;3pRfXV080M%Jm+l-5F}N4FQqf92e2<
zMVl_PqA|BB-BS=KOhq`x_e;1(eaTg5J)!_T!}x-Y9Oh(rgxOdey)nHx)hSp?`=$!d
z-Yh`4UFP?EGCiep3Xph9{`e`ZP*kLc8s89~REme3%En&wOPx9(pMwpkzL0ivml?Ba
zUfc;UU#k41TR@}7&ym>2`i^NZp=;<<nAwRNhiTJw*tImEq2u-otK+fo_q70M)H7#%
z$-xt?(U5DQdj^t4ESW5PN@Cg42d+9JAJPjv<IY-zChlz8E}J~Y<I!gg>R3c2vIT(4
zjUzAikNbn&pF<78M%YwlPD!@3Umw_ZK(=iKs@uk#9!L#Ldn}?ED@y5Cu&I4?Ii-ti
z2fcfgMf%)RAw8vi^BS`a;@pjCfkGXYVgW1-83)#5SI6pqMCKzJe1#k3Bp4y|`KyT8
z=~s)9S5j@8flyhq<wS}XpMNwGhg`M3l=aA9<e*#tk8Rfkr00>k+UvplqCOYE^8fRf
z@+N#G`z1`P<|40|UYJwjFE8&C^hLW4c(3W+kd{xryXr}@X;fzjBuSNnO@r^&q#iSO
z#fhhPn1`wR`{oqwr<8-lWLRn;!6bs9dKC0+e>U;NjgMuUI>o6+$B7MC89{0int~Ki
z+%UdM_3ToSY4>m+hFwtQagXCRnShOYZyR*hqkH-`ih6X98o1<LOvNV;+K$6~)>v1U
zj~DEE_ebwMc_DSBxWtFbq^9wHEmRTxPCC~Um^07%3!&nCwoVtHynLPGn^=rFDZcvF
z7x4*pLLRmRZA9Gi++azSRBqn)IyffwG)8CvwT@%^sYkqQ&Y)67vd>^#6Z$dhVsElf
zcsNE++&yNeo$=_v&CV)&kDw9jv*wf<TQy=;2eKujG67W$)z^Wf#Rb+RwVsQ#Vw!G+
zJFO;lvzL>$b&TLWK&?D@loxF5o-tYI8Q$em;az>Q+W<?Y682TOck%+j-8KZql2bp~
zP*yNVx6(;Rp=x<Y=-4A?|K^9ja*0^ZbjgAHNZYpBpGa}fZe?!O8gZaB4w`&F?+&4$
z3!sHisDW`=sZwV52$;Ga0J4TZ2T>T^kMa-gIZQte3lR~g@hQ&Va!y(tzIg21vj%lv
z2LdV&{`;Gx9L&U*pib%7RKDgpRqm-#CZRLKpf^o-{`-NukLWpN5iJhEN|rWgnrZK=
zp_>__gKDI0iNUFW>PLLV%#*X~@7x3{WEg$t)r^XfpGLWn+S`dm2M{U)E;F|+Rz&Cz
zkV>H(Irp(xkY$VB(r+LBVylsGR2HS;=loJVyxrWwC>AFs3}L;eW|B7b$j+sTIjDh_
z!V83LY$)X4y^;JLCygI<EwBrg4_`~or0p?s87p{++=VU<KEsw?o+g@)p9Kl1xu!G2
zLJ`E3AG((=CBzt1tM|`W_<UIlD%(C`uajgHnlzQ}>kcf-C>0_!iI+-T_b@&d1I5i!
zczcXkH3BcR(e!yIF!hlj^nshng_H1jxUSgJelBR}6yc3PH{A3DDzEgiIU;DEWRBw%
zViR3FO1fNiw{r;P5^iE*Z;VMM<6n_V9=TYK+v1fafc(a2lIzu+>c<g_wJO6O&r5V&
zj(cL`ad!9pZRgcgrDt35K+O$MN;YJZA6J6lxvkFy{!o6BZK3b+2`jtC?lrhY2`wn;
zc9}N)T>V2{wI}8=W~y2*e$|tIqE~=Peq~ySXQiP{T4+pU`xSp)@bU2z6RiT8p!Il2
z0A5VnU*FwN-m!Cjeh(1I5=HDh=Gx0ob9R-HvvgA486~iN>B8gNs_~)6wY3L&IAeU@
zQOTMvk_C6a&$yKJ7+rWkXcM1Smy}^09px|OiFB9mST2U7uh_rce0Lw;1JVLlJ9&9f
z4RpWqS-4^InZ5JIZcwKzXm4GYz~Zn-MP8_nRZi(DhtX}0B1YPZo)7Lt^#}6E2}z_h
z|5JhX5>)#eui>Z2d5PgnNMCru+?TU~+-s+u>a+Xv{E(NkGGcpO7cZ7Jc(vv)=%gea
z1hCiAx=Z(2_ciBD80n%9z7`6ajnF0z)9MfsYf?T2rIJz#@udCCip&+_XcB(-ovC`M
zS#F*t@9_QHd`hLQpIxcKBY+Zlmm@gQGaojfeDQV<&BHu*>p*v>c;9w_C9!=C1k9o7
zvIU?NIl%m7ZM~~c^3#>{u#qRwCTXOq*PXFup|2h1ntwzd8ZG@1YkfK-H=ix4^ZDVO
z>Lrm5_CmtW=t6GHh5{{13T<RVZ6P8p5#hg2qF=(*sU0roMGhQ|y(ABNXi;j6NQ}Qy
zbJCH%cJyUrFO^85LIzPvB~W`ySjO;fnAE8mw!1$8@=gxI?kn-!m#r_R>~rV>^__vk
zTw2kYvFQ+wu2BpjeJw*_HxXtiHiag4T7NK$4SmSne~uwbBs5iz?YT2=>8WGkvGlrP
z^k+so+K`<1fEH`ZY!y3AHq?gya{iny+RBBt`^Az-XF8nsFfXR0M5M$$TtFgj?D%_3
zTk<Zv!FmFeNb~qB5*Q=DqCQI{jEAiWh>OUEBunHyc*a>#b0+KgASF|(jraYp)6b5E
z?ceJWWjH9V;jL;^tMx~r3fAfR+456?cO9=zuiQuu?eH(IUL+s->IihtJOhZL<HapT
zYcp-kd*&X87;BSM1XpMJd%u)PAzaotu3@f;y=c^cxPN?FkZPaSLg#$j$(nAy$V{4M
z%R&ns==oS3^ZmwC)5?cbE0!FK>IKGGDb5WoG(Jv+4Ki5fsPDz0n2ZOo-nt56r`CR1
z8TsT3ygqCzxS`myp#eUw5yUoF`@Dda$gP~#gdfnj^Z3!>7kx~-nu~hQK&=0&Y5N+2
zn$p$V`+6aQCU5ZR5uDLiEKFj6Ai$PIe`=!W;y-obryVZi*b@^kZ8r#!AP^AWQ@=X`
zSwG(Kc#QJyvbUB9Nxb{|IoQW@Li=6}YZPTbFcQMFZYk|C@7l{np0|C-JFu+Ky#3;v
z9{tXKyM~WaPE$~J)tSWaeMT`*C97N5*RxL+<!ct-S-VK3w6`jRJrj>U0p$XYLo9PV
zD3t4#4Y=80Y{sbKDnUoJ-K+QL__$Fjr)539S2smEKV2v)xha%_RqTwNOONK^6?neO
z5?Fv%x#SB=%PbO@Mc`?nljF`-Hu8J3DX{wS^KV1<+pX>wgRH}8L$`OEHjJTSg`7fr
zP7>3&l04x2Uc!_noVWU}nBa@14JicJIEK_{Ukbg|r57BeP+45mQQ?#z#a{=v1q#lu
zjT4v1wVuwAbHqMxF<UlVG(;<?3az%_LsEZ8Yg14U$O!v-zlf9!wkz5{;9|OSN4`4F
z?^)L;ruRj{XYvzT>kOMP3PX)wj*NydJLUsqQEN`g?v24+h0aA3E0)gQFtew|o}h9>
zHoFD_)?X*HX|!i$Q&#nr62nG{da+_l8u50+?$ScG7snB7lVhp#X5+98e#_>V$eHMo
z8&H|zZv8U91H4GTM2C|7PWtD5@971V*q!G`q^G12!O@fN*-#H3*ZlpI+!)C%o*0*H
z`PS^J6MQIirK*?{R20`3DFbo4JsHsp<dxf~i}>s`OSb1kfei_#unTvuU}TfOl+?c=
zC67qrvveWk#SwNn5q6poY4!bA_vl}ii;IIhwiKv!pG@@y!`CZNb2ahss>QouEZ$VD
zBI28`9%Ix@1LKd3Of4F5t&_v{V?*?rrxJFuT%%OWm)8GcW<SkfcK~e%nhiW0QvIx4
ziU{X$tGY%m1DM-<nJq1^)S<#egfT=~QsCDvi{x4t<7mRTr_%Dg#67#@4KimHMzL$3
z37D%D1t5VtP2agtYR9UIQPgy1s5lh+^00L&X|LQh!9ZQn`IjVlL@`B{rC|zKRcf^L
zi3MhmcE{Wz%t+*+AkE6kQRn`L`94qcwVS!EIk8vlpoRT+V~Xwn{QI5wdKLO3Eefsc
z92v0dwA-@yIdX8!fC<l#!arU@$@Zq}^V!JqU6*P;YUhI{aZL+oLO(;)<kFjh^yG>$
z%_6nGq^T9rR;RsUJVKu)nJ?b-pt`AJ>l==9qLc28F;E)Qw#JSCxlH=|`tU1#8P-AX
zRmTv`2vSI(S7a_px0-KOv_0{%j_$kStQyB-QFCOxZz5##%gSk?D%`qPO1+^DKCTtS
zpPl%fIVAUHL^Gq%acZ~67d?A7N9rx5r=KZhn@-}-1#((;{>U@g{<BfDwXykfOHtg&
zMXhX;!`d<Y1}D{OzuxEZz{&SLdiw-Y;!}ej#HmB^DHlg<0=PUhhRNu(nCtrHizs_4
zS7Ze%%bUMx?f)UgysstTcXs#i=z{~t<?252y&sRZJS_3!pZe4XyMUt;4U%SgbvrJZ
z;#(6XtTRYvbSRebtsB)<E3qL(`qXa19fa9)zS&j_x2S=4s)5a>C-ptLwYxz#delDF
z*)~^I=&Q6$VZHLL7WVUQH%Tzqf5^IDC?zh$hN}EM;FIjVUUsHnwCfzKIS}2sKVCnn
z5~zKcyuK(|D)G%uGNY+OE;mTQURkI4-sbOK{xrAw1*`D>axt^bozE64UrPRpK}^Kl
zOq(bR>1Jn*S~>U;59zdT)`%Gv?%jK<qp~9&KS*}XpM8J1vRLR$ets2XGT0t6G}ONJ
zG+B@TN$Dn6OcAaRE%@eQr}y<RaiVdM%3u-q8eipd0)H}sJ-4BcAdyy`dernyOVI!E
zm_YdCDG6GDoQWAf`t6KLS^L>q8rqy?n#(p)sU!a3CA0%#cV17ODZ;*1&C=gqX@c>o
zgqlqB-PmyXc}nQ&-lxx$!*}l^>D@KAVwy?g&Nf;U34>02+6c*!dKKlX*(X7{ZeGVe
zpo1ER$NKSS5P*~r>35#Sz`0p9(X~0Or22*u)#IlXzNC458CW5ulQJg;_xlP?8%;}j
z$IgBsYA{1Nm&GC#o3B4<9j39HIErKvVXYTJN^w@x>017(N&O~$oYK(`n)td8iziZV
z6Gy7prz+zFSs3i8IF|?le2s4e8(-$cQ3PGPJTm<V^FTLkF6s|_?eS-bXqsrS_7O!f
z7+dw0n>HL)KBCDHw~Ya~0(f91_!dszQXf!JI&!mr-)FVoUM<TfBv2XXwrPyj>6M1(
z!WzbEfd4>Y2Hi`^k_V<1&q9K}jTVc0If&I2SS@U36-f@By}}18v8zWXi3bT{K0m0p
z5A#V3SyoGpBJ^c_fDiidy=>Br__Fu#$X0rGZ>}Geaw>jHW^EpC#v|!nqVt?~_^M#o
z)_C>L)lUwR<0j<Ze)JgtK)2T9;ncLHvv$(^dvxp39!%lxiI{SmVjF3p@|fN<?-IY6
zouo%z=Hx7uabuV&dGEO5PVSOUeI9{dYkZ3>6LX(Jy`Y7G^L>U|wU^oCHwhZwebyk8
zrG`Cai$}V7a?s^+Z>>1?FHL$ybNMOqn?~|b$uKg_Z{@yrHUi@C*~Qi(?^E#~GrjN~
zQt9gVK5>nd_d1>RN2D^n9|MUb49@$DMOSMcuL3&RwPmRZnkXN{m(AS!0tl3&$I%|<
zO7ye|SFQUu-{tsVRBv}Tz%TCmsTii6h+vtUlEd?Z&d4QpoUc?bf96`f-jTS;62$sc
z6|FVH3%s5nS59nCir9Zxgw0IbosTD$nX0$wkDVp{Ok#vMiP6vkRkV+O;Yb_eo;wAX
zkf(Z=C34wK&1m+x!AEXFxKtZst9=DtZN^ZSQF0d*^yj@!4_R-yndB0Ya8zaSe(`Ft
zXYNANE$z8R^Zuv=Z?zei&iEaAn2?VgaCK5U!Dl8H?B4emE3zo<p5m|y&aO)CgueX*
zX4Vg@Jmk5sgXI&<)fq`eYc@^w#D0}QEs#^3#Q08lm&|XOCc5>v>?<}e?%7`W4|i}>
zz2LTwqWFy3i<E`P)a)M3P|RHqvkov+9&DQt4tJ~TcxH_#U5iGC8|eiah_{q&iaUH>
z#AEaXtfqE+`E@ebv3Uflv$gs1N$@|cON?Xp*<c>CvZEvDC(OaSC_CU#y)CDc8NOii
z)rQU2_|soP`ex;-VBi+MUZe7Qvw9|Ta}|Ws<1m@E*38q=)7sM}n?ID*pGzx^b|oNt
z+xpm+Y3-tsbs4Au;RzLhB0V*_f6v^vgkJbt|AuActxpG>Y2$aUYRjK#J1H*vW9~D)
zzyb-ztmx<*MoT26(fwpC#Eyp9w2I|(W>uHt+cM^vQT7Q;cf4D1vv}!UQ-9ul>#Az?
z>}Az3z87RL^NnC_&duvWwux+6Fh-Q!&(o###!+#{=EPK7z{Ak-!lJ^Bai+I@)B%*L
zevvg!u|ePBWEj&9{GjU5^5uV_Hnq<Q%NWfZS%+t|G;~D6in|N_!v+hIUu9D7fvTh8
zoAO(*#>*f7^jsBUn!@UtcvZ!XXpfjUqu*8m=10$Z;$y^Y8~R4?mkr~;msD0CCi&$K
z{WqwJumk*Pn-|6<f?-3uB0Mj*Z#mnB4@^G$Gd%XqreIU#x*_}~oFq$frwy+eb!4b;
zfU>{LlYy3C+#%?S_7<!79ox4FKc|c8tqLi*d|cI2V1$Bf8oheYG6SnR2X=d&jUF1u
z<D&<0{>|L-^JdX++RhuinnaOv$3A7(TH9HfuE%<16PzA(_&fDPgn?e8!)ukdzbT^l
zWvEyvxcFQ0qA&dzTpOke8f&o$O9h3c*QFqftny&+=#b+hah044=kIfu(vsQvf<NKG
zkQjI4+M8i}%PMrAGK7{9D4b`v+wO4N+A)u4&+V;?@KbFcMr>zD%h)nQU{)Tpuk~#h
zD)P$rqI)7ZT}1Y89l0cq-oLo)uGH?uR*(z3%yadnbAL}}UdGC?!~XpM4|yLah6=3~
zOG#69I=Fv_H?{cV!GUK<#+O9`>dThPUQ)?{27G7lTsq#%8|rMbU0>XWY2U+J!n=-h
z7usLjYqZAb`ALz#xqy7X<Xl?WIpsCZr5R}1CJxUT2ulxET&vft<{Bo;Bn3#rK8zF^
zV>pz_wn5zBn&G8n=ln20*IFyKm&%`HF3x*tOz*r#>CIf;a4U%^WWi&ag7CbYVm*8p
z30len9Sctq{ALNbChjv#1p*BDBhZ`F-3tNYW<?e2Y@S6e{V=ryhR4g^)_K#XU(e`N
zaiCbXGJ{;6`Y(z8n1v;|Pn>s;nbx1Ck;Z(mFp1MEodGj`l@P1jUOkwlIM5s_Vrf|`
z+ogqjrK(9_ay2Q{7O?tZTnPKq_?VgHhks6d{>s&ntzhq0>M8Dof=n98ROjy>Mmk6K
z(MzF07yCJ8s&F2UmTkl?x<|K&nqkdo&S7igo@vSt?_|;XIj!}DSJd!Cu&fDJ8$C;T
z<u~xm*sT-a4STkq{UQ-e^HxPgh6~e0lmR<t0hCieQjLb&w-=STPvu`&4~=-caOhx`
z6a{|wAX2jn_9}0L0LBz|;Y&VSD%^hkEQLZRWlK{~2^e-sy8=Y+a;Jex4R3&y3hf8f
zy#f8OYl>kl4Z(A+a3f<moLkAYW~G?%_a&pZHa4F5)bUp@QE-iq__bI94_bxAdM|^8
zHQGdEw|LZE(gwCvyC+8^Ey-fA@1FP6U4(4889N&1aQCQi;v>1WoF-E~C~jq|=$dC;
z-WOGIE_0_nH-&b`dd)Ot{!z#(k}KKUMS39G{Z7T<O;=0eho|bqF5Y_>x1wkc&Q>uW
z(W88d%Pbz@TIr7x$^BwSQ(RlJW4Z6$pYMNJ_|kJ-znG4DD~!qh1#~lMMz_XonOW*N
z5Btgj&QJ0-_??$?<mLwt12=<RKjpLr+U#5EC`9TZk^MUX6%K0%)~J}24Dohb!(Pi=
zpORNt%XV9Y8O352?z>_l%2<6=9=E&FT8XY@8*O^ew3Q3?7V0@BVsLw>j|+d!=w0Qk
zSbJIxHd-ko!kVU7V2HJg8GMiN=3BeDxNL|_t0znyX!piQ1ewc>1cpMJatA<&tr=7}
zCwX;(@Z6MbM0fS-Tzl}r+(ffZ=TMI9)?pPa^Zbv%B=#}`p*cV4C}c6)G+b|4dP2fq
za(DFHXOAX@QI41h+e_TJ@vQyOsWQgalP{`pfV((z<`5_f6vDUP<eD+Jopl&aX8L&t
z8rH&@7tJ_U#u3WW4k0cptF=F@1ewh_Og&I?*SGgjaH|oo#&!<sS>jhL?)Zn5`gvc+
zASlClLn!9jkIFPrL7@n#_brpBH}R!BR{CtC+Z*Orzc~HglqDdLE!-d-&_7SYw3Stw
zHf-p_jBz$g2JppDr1JDX9Iw@knhf-?;bHcw$h8>9ANijL>i?mdA+@ad;&y<_ZoG0K
zLCZTNp+m#H36SS%Qc{)dA@15KXAd@$ntEyFJW_;wApSl&S4>xr0*VOQ^H8{?%|Pr=
zOL>s;DWYi3+0%GLZ!Lz}mfs_I^rkUlW#tLbOoxZ^TFCf!o^5bo@VKFsQnC@$4QD!z
zmp+ZFYNxn`<g7eY2nty2a%w6NqY^LLZRfZD=;O+oy0sC<>fC>1_hk~p5*wx<q}(rZ
zGkrQN<g6#!Qhpq|Jmr4mW8o;o{lEp}^QQ*{p6Wh@uRxr)obl~%icWiF5IFDtBC)mN
zqmU9AKY`qasnvQ_;*2mJLDJEnC!3D)RlRRIK1B$<Md=>AeJnsN{-K1Uy#Cacd66IV
z1=LS0O(2w1ObJnqgZUo@WK1Z`x)Q|DTHO{RN69O9t;V5YJ}RyWS~(_{C^+S1QyEL*
zxtK-oGF_?(N~0faK9Jz(!3{~o`%v9*Zr-0$y=2b1SBkWA-`*azQNt-;VLPG51cx8X
z#}dK}{I+sRs@2mD@t<bOjO7!n(focHBgXy*>zNLDKCiV{*B}C=rq!t8$zFV+VTqrF
zKtO|thfG%0xL@B_`B<Aazi?AtoWw>#YCW!Q1YqfB*G{cOMQ=Nt670XZH@l7Z9!A}M
z`$ouJu%x9D;m?V==s8yT>2c?svKYghqK4IAn~Ime(25vy?ft;FMOs^Ko&+ubQo|sV
z-r%!X*JpFAA<Gh_akBp3&Iw-&HiWB&?kyk(amaX_LJFvQEiMy_Gu?lFtYXSM($7?0
zxq)Rb2)t~oDkU>}kg#JK&RcN1U^n0F61Ca6b7zYcJ&N#YWpe~A0Js>&$|goxf(6y(
zF2h@elAHH4;s<`7mDHz|^X5BeSTnB(Ls|*0arymhMcLhY&b(T21($^F_L92LjQpkY
z+pRC-2nhlc42_)heAJ@M!zRvRgnBpXbEwt84c#C2VJ9~Cf}bx#fb6KJrQGmt3d~?N
zsb#OX%=^{W?I(`YDd7QvY`y8;=`zT>c5y6k*CoupNOc>2Ml`EA`x1Q_`b*>M^DNQ^
z6XKQWQ8BLUaZYuT_?xpA#aCAZ8p^!{c;9O{Gq`%76(%^|wdKWtey&0$`49CJ<i3vL
z#(lk>{(w0qw=Z1q9a|v3zQGEhZf|}LD5gb-LnX)W;<02mT5L8D(-B3AlUyb(Z%PC8
zES_0-PXGeqyXIt!#yDqIejy~T%KnKi!Tw;v#o!Za38)MEq)ewjmMDGbg&9wsDF_V1
za~PU(V;wZm_xyV$)2PX0-@j1`M`&^}*fAGsZ(8*Y#rO?cE)w3{p2x&L>~VQjKd{Iy
z$7fi&GJ^yOLc?2Zg=AdKF#k<8HIw<@jo#+(h+CmSs2(d(=7#$<m&pE-ygX@TXxjXO
z->}r?Ku^R3{~Wsa>ErR-@uO2hq`mhGLEI}}7ABkxC18E(2rYV)l~}CW7Z4;M2a*O>
zH};>uF(`6;n&;Wj2tO)_eRi;``E4&~APF(*ldn*WC8p>c>^h43Wtu(KIr!!oc*3TW
zhA{_qyy?`5w98FomrhM57*2B6QVoev<K$pmO{wC6)e%yO1k|B*PWv*Cc@+h5X<DJw
z#pcA7C7#!<2=ddzVfV3k$i4fVXXodmLS%fVOh#@{CF0~rt=xj-0g+ucl$TA|=?2Hz
z%j|{*e3FJobm_%E0*Tr3gT2UcAV>66vN}bjCVPz5<<CPV5MG#|%a`hS`u}N_HW2tI
z@g5>`i=LuAhy}-1%|UGhKdZdgO2V|9&i_Zfxmv<|$l=p+t+C#2YwAAsQTAe03Og`R
zh|CBcv^?t9RCD?A;GNq*HvE7DfyvvmH4HO68+hwUs$h4ACOmp~b%L4+!4jJo<h{b4
zric3OLoEeh`J7oPLi^5_Q7T5^wYrw4K=yVl{X%Da-e(&VTGaByex-7ycChxug1$Gv
z>Gko>a#^T4V=N`;y{|^B`N%}{ct1RAbskCWc(nSS#sJkJ2@jUagHq-9*23FQ$FrVx
zuDCHpO$Qu*Urkm!F5@jcPA<m?yARYY{RCSBw%MaHZt8*`iLj2@c}6;~?7S+Z8Efp%
zvc7_?@cxa&g|JQW(J1z(7|$dLw5QoX9(F?T0Pgo|4-c9ZktX&dn0Z^<N1Zn^ldI;6
zy>QB6?&dW$o^((;tO5FG`y&H-A2rW>T)c!(nst+L6l$ru01fK!V>z=GGKFT5+bk%_
zB$x4dv{JymQK830Zj+UjlLrhRjG*O<lWid<5cfz6sFs2m`ni>Wf#)T1uCuW{MRb<Z
z^jN_<C&b#9@}_jb(GLM1xpHyduKuwvH-u+h8u#s158~`yS8gbShq;pr3S4xJaCv?y
z?^DWFaW=e@xZS6Mi%gS~j?TyxIB#iD)qWh1v28OKti=Swc7ED7f^@x~?zyLY?N&WX
zsdqT_x4$^p+YJ0TOGgvlbrPN?I*0ywS_lzy*yS<0-)X(y^wE2b9RE(L6aBMwRkx}X
z4j=3qYu~Pggpblh<D>f|zjJ1q*0g`L!)0^a+|P}t8oUu4AJvGu@Ky)?Ii80OHVDf3
z3%`Ki%otJ<X5Ly`nA*`<8f#y6_Dn%niu$E9w>~ak>0jg0=NC4!y`#6E3pGfEr_4}S
zB42n3&BTCrc-ivLzH_bBwW&vcNI;vKk3BIPT6f$Wg7$M%U%A6NlIKk|?p&P^@#UJ{
zJgOJ7Aus6_<b!EQka4y1nO*&G8Jyh}v?v=xfWGb;85b68d0g1Pvdin8mVbikTsxyP
z<Sp`RQaHrxl202MKEOPk{ZA3+6{jtDuho5BWIJ(Rq<*B18|$-^d1toi^IghpiM5S+
z0(&<6wHl@M=yGbr^$70#=Gtr6kMO~<GIsT?y9Uuf7BFR>kf9H<T7rOa)OrI*tZvFN
zKuB~c=mEfy1!PE71`CkA{;=oUW2)XZN13;m9M+(Q4dQ|$u0^mMNc!NI0Ayw_<sN*V
zk1G73tU~j>oMN;jk{XhLt&9@!%$y0AsVd-Tl{9S;I$MwB!Q-=oo%&PoTEb8npX3@v
z*{(7)k9;{$R2mzC3s(U|FpamImh1qo<^6bY^l+9Ik(w`YiR;Xzp*C4$<<_x7gxM=w
zj_$QmG0SwI6sk?@UYi$}Z3+~(Gj0BPM=4W8%+^f%O5d>~u8(mnGcdX$Nm%_Q-d3@L
zvNWI2z}2o!y?hUjrZ4?G87OS&ohVUc^hlGz)w!d0zG*H6EMs+3py@jvXp_^aWV@sQ
z^+~7vf0GA?ZEs8{i31U0NYb;?RJhkkMX>zgAFn+;t`#1*HN%uhSeBNo!8K{OZOM9R
z%`xhc++g={_vP)ea8}@T0bnGzDTP4ytsfWO{rlWN_=)&E{AREB&t9&dzDaF&%GMfL
zMpY^b4^Y&7B+$m|-Z0MVF5Tc^_`vYC4-C(P%gfT&J?nWaJ}RlvZlBRGU^r}a_2IL*
zIf8w8{6c{Oo*GSPVh^dQjAL!_<!PU%6mr?4?HU4$fFJ&@eg=CdwZxQH=Bul^8-!0Z
z7{4P~w&2I=&g|W?M56s~h3rBh_)=#H?t&tf+%lLC>7Ue<DhRvpNH{HxDccf8qzl|$
zf#-@CDnS!_Cb<?EA2Xk*eIE48!o52i#n`{Y!3}Z21%(n<rz8oj95v%Ua}FC_>**HT
z6)Z|BvJ4~a;a9+oHoHgysDWd_)t;?;b0VA-e(sbu`$q7sCms<!O>85dHUKT7vqR+Q
z<#tQo*4QlnDE!%w|C{Vp0hxXBG?SDVr1=x@-;Lq_4bAmG15fxzr3+4$4pf+=6!722
z*oS|GzTo@+Nxd`abpO9jmM-c8@xe*fe~AN>GNt)+BpS|a>_w6iJI|4jWhpENRyg@>
zrUqEX$S!2eD?ZzF`S=ZUbIM-N1$V*U7k)Sbo{s8_^9I~hA#`M1u{G;zV%aZopJ@g+
z^8MXII4vKEtS{Q&Sj6zC$BW^8Tka_^@;Z3>!+-8wD>=|kOOKWjQfLHN$vKvRo=4lC
zz+W4;bpP<)EC4y!l%2}xE!{iZ!^m-0c<cm5DphP_*}Qyf`Q&TIAg}so=R5s<X->mu
z&cD4RZW3Ld4zF`{*YXIAP~oOB(8BovweMfMPXwTalmF*g-+Yq&3f5yX+8nho<KjsO
zd(qaFrKV6@g$Y48Lxn)6v!|txy&|c}823Uu4l(iE>pa6-oiC}8^roys7k7V*l||%`
zn%}`U7vj%=Rg!eL$TGZr`?6_>hPn8mRpw>i1k^^>y#NaB=1`)+_Pdla*>&DT87gV~
znsML5rTy;kBDIK~z_$8LA2T~3(f<vDV~xcnKkV-nX2rbe3kE9lN7GPh>4yZ+ux9&T
zdfa3fx@lx&)tC4%AQJtnf`ZG)^$>s+*hQ^E=av{L|J$cgSTa{Dp$tpE=#M(sp8`T<
zqtfBYEipGJvTtMhVu?TMcRM%!nDl`FP=ioa3L(o{p33tAZJay_nQ@@)Lh}b3sB^fP
z;~ox5dJgalqzrflMB?g;0Cfq$(R7{43otxDOb!l4vj3alwT8FS4{q_CTkN$Pz?<!P
z6CgOeFd~D$me-&TKD=e>!B}}7^tU$bo5@Z;?VXfV*alC=$RDy@b9J}dntLnK9u31U
zzHI_@px<2|uc-lLh<x9S=zfcy*h!2bS}piT6A9hgLR9(0LSnJZq<`t`CN|y1!Zbp0
zluG&*i*vkX7oM2>Ot=5|JM^}Y7-&+p`b%z8ozjZ%0<jbZ4`BwE2qd>+RC4=3r%fI5
z&wQ@qB+2?DS`X(>@$N#~2@gq<SWiyK8rwt+Ds<o(PabaYvUv6Kc<<ViiDTHk!~W2E
zp1~Q{2Gk%Q&SK~y_VGW+CDx~`*RkuFf75EkEL&$O|6wc{9Ad8fP)%3Hts_48m+ts!
zcchkfIgL{smB44ewD=$LM49US7g%skRuTWH{yzC%#jVN{zhVQGl^81AIWdDumMeJ9
zmE2L{&#WSj`%8}FRXuLk3W-$v*frtl5`NJ+39y%EZ+>Gtpz|~iw^P#6dw-7pmfxPI
zlY!M3KT){eZY*km!^2W`kaucZhT|%0)U+PuNQKYM+@<ej5`#6pF{|G^l6OZ%T=pMr
zq>1^jiV$m?U5mpP!5;of-?Z1UXoC9PAEn_ipr9C6-bN;;B1x0#1itJAztoa^US@9W
zO!Hq?EQQK9va(RC6CbC<A$Rg3d~)yfOT<FwC3}@H=1;7P8@gr8CYOe?Z<R(8r`M*G
zCL0Ku4!i?IebXh1ptUC|P{!?DZHzI(`^o3yhJ)psro7k1O@%Y#jVj_kwUt`dPuPh|
zUKqErbN&#+Cj{VZx2J0JQx+=e0GRK)8L!V+kga8KczR)jDVy^VP9tXXRQk#7<>Uw?
zYh%26v`L(Z8Jj!L???Fo2s9XDHdhlazouCQm7y0_%v0LTQ&O7TCRg3=HS=C<%NyEM
z7;25(KNzH}a1nEOplWoCDxXo)+jz!n53y#R)QSrw#70j-UJ8?b=F~N2W@ooJq=Cda
z<7a?0kt5p9C3OD=Ogw^f2_KtGXa-W!WU~3kns<v`4Ny{*tTAJ7Np5+bo5>YfSqDEq
zT3;$sHK`cwQbRX6iuwC&aLjjqkQ(@bU@7R!NwM02`27}Uw%7;}Aoq&*ZyZ{O$^#9z
zhtn}FU>;V50#X)E$;(}NUsN)I(6>3w)~vIWVw|R8*8LB=uxchSl2>3lWk66qEJ?Tm
zFMgOB{X({DF`ip<;a2^<hn7&Z+m-~8db~X1yStAjavk-Z<$%-jz-~H@aP)J1t04QO
z^CP;^9>oxQYtXPuf=UW$9`LRW?xJqonl(T>t~W%AFuyfGhu=ofOOF~RL85#d2gEs4
z*nLnHx;80)zgab2gL$2W_63-dh9S(jq>`n|1P3{-CP1+fACn~FlggL>7kTg9)#Mhn
z{n{vs6m=`oK}0~BQl$$jy(kDsS839bDlJh!ut4Y_NK;DaNbdwJ^dh|zkP;w3fB+$c
zB=17q`#jGWXPgi49{+F*k(B$sSD9<hYyK{Y0o6v{K|h(Loj5t!sMj~zf%Oxh=&Q*t
z^Z!(kkBL~3RhmE06|K}4J(~ydhx|mk92wC=I06yV2Yg|5sKvp`-q}Go7@gu9D;Q_i
z!0^2l5Q~w$zacG4N`I2&^nTtqU#T9J@y@V!`;O^9&pra^V5?Gw>AIbMpSL#O)3zN<
zW@+U$aJ<-2_WrhGMDRWI7CsZYn<t9xAnR-WsJNa5<2gHJmJSh`#Ab@J?wQ$7ZTZD6
z>|N3rAh59NW2-NYSJy>XD5kcP1iZREQkB$1ijIoA2>vjRLQDfPE-&NIHal7!rJCt5
zIXdLJmNX-<K6uht!hT3ZzeLi;kenvdrxn?4ZE?1@1JZU=)yD6spGvaK&wUZEfYQL}
zXkY9VC4?M_o9BiZLx>ZRpw1w^k~n7{e`)U33qsA+`6nia9`|4u2sMg~taeDYr0I97
z3|^nRR-!j(_X_7z+P$xIRTG`FaceF>VEfo|$6bKkdgmVcDwJD-+Wyh;<{<=?mI+<g
zP>4*m0Y1s^;xI+sh8bDD+uwk_{DknmlwMqJK)9|}rx%;J(c<Q<vlRg#zgtUKGIPu_
z*N?l7w=9GO3g%9(AA`CHiFuWii1!p9k1W#!YuekJ=$dTwg(<B78~JB}2%zhAwY9!A
z)`z|zyTwRZ*Jx`(Ft@*DSOEJFSln4@Zp7nhY5`-!?5D%is#k9n`39A%1m8INqGRME
zS33t3LIX#9&4<+MYdhbTW;rrcWpOF7(hq%9O9YWxz>zTLs5!%RLzU4_z7dl8Im4jy
zM*Ne_)(xJ=yX?y!u>EH>$HvRs#Y1RyZ1+aSkL<j6_^wxgwEaOuyPrvvK>Wtyiqw?E
zhK;y4pu6t6r6Mmx)YQ@r17oNnh^!N5@9K}3x%U9;^O>b?Fv!<=*K{=4tCg%JS!LK0
z_Oy*gyF7Bnb?Pj$2N(}349y)=-tT~-SbPNadfumksQA#rwn75!VRvf`(JTHAj&(C>
zUR!HFHPn!;yJ%|SMkujssf`&Sh+H4Ec)4Yx@*UT`+K3IlZr5?PgGM>|m`g7vy*SA$
zr)rJEh`=b`9@*Hb(eAD_{$_WdCPd`P;r!ch|2v{NtbN8cx~VJ1t_*r2xcSD?v*_yW
zC0vibDD_ICU|D<um9y8BA_Y3DAy<!|Yz~4Sn3w-dWg+04K5FvHv8N^Y-eBB$*t<Gj
z*g!L?zO#T{P0qVVFxO=$vt(xSiid0Xc3i$T&e$Y(LvS8}_qsi?6~e9PxswAr$`g>*
zYzWf&7l@WLhk4|5X<-K;JEbgVyG!%q!Rp`V1dd<-TR{f`$+Y!8JZ;7nKIZdGY)7c{
zg=6TC;eDkefS6jswFX~n)$0-Owds41)G3|;m*BH|!9f`~z|1G~gtLlU?c?Bm^R4P0
zGcRL#Dvy9#PHPJ*Z}AdC)#hw>g@5pTpxfTLARAAxeLC~~M(zgt<_-HP=Vo4mIF*Q%
z-i$bPRMXmg<Tgo`$)fw&xMbr;gyzbRXA~;$EKc3PCx{w7)nORe(b1qeFz)`+rz^!H
zGxdx-aH7&&`UB001%5Bp`%9kmZEjGRDi(+$52v{g!}fub>d@F5Kd*}^s>Uh3J_P}C
zCi+G}XP1%(jHL{_pp3(~{vsEVngo7A`IBQeg|rJc?2aI|gUfZ-703P0$Cp3SHIGH|
z6ep!UL!eJ9FD0TK7I~PHKFv4KTC&$vUJaY8%+9Q~-zWYC&_;6;0?c$bYUWEv^+!G*
z<`?z;iQn3l7`9Iw49Ih>ht|EgfAc_Ye7^sb?9)4OE7g0%U0G(BPY`9FwOiSpp~sSy
z7XlW34JL<EARS3n*V^#wUPs-J(DXCRQkj(B=I%zmer74lJRv~zj)QH@mV6Ys6w;6=
z>1p}B`(pB~oK^h&<}-D^uPR~NjEnQ}r`wcq!e`BPyEdmRYCoBYEmCD`$=IFkTE6L<
z!1z5fC^c0ngiq+*$+IIEL0-}C3kGFtDS+cW1pz(N(8BMlm&F2hW-k~ky5H)5zx;$>
zY*kwE%N)*4T<P`Q=}UWATCzrHT0k{*>UHoNYuyp|c_>PqBy-kv*$G(s?f&wz8b0)S
z;Kfn7Rwq9~BeO54?~-gdrHl3L)I4@Nhjn7tlyOwcGD+r282_m6=)vmZk5i}^+*PI7
z{(98S(Kp{|*q<D$*=*vuXE7to=5KCOMD=eFLHC~ujCpk}<MYa^=mGt&hdmCPqTkMY
zH$CT|Xq`$dqfKdZV1ejLAH{?7fb}03M_fYKe-JAD&f;Irl3zgd-l?D7G24F%(6+bV
z#Z+4C2Jk)htF!4Q*wKoZU8a`Ux$(ANE`Zb!<5&j$HPSrk0``>=SfNO7`{9llQ~hw<
zCpVZzV;tQxij-}^GZ3#dt8tq;eH{iNF={4a6CsALZI9LdD8}@UEvxf-$j`vK_xT;g
z`ERq<myc0*sS9$R+bh$?-20er2cRW*)F-T;aY`r55#Mf75_%}|@aV=NlH_{)nK<J<
z=3K-Lp0Sn|d3#W_CNapkt}73|b)v>+=US1s+n=|aWn1rZ_9IZI!|u~r9=@gW{Kols
zgJJgnjm6W#n&7A2;SaYeDJ|U1_;-YY{?Q=6eMOSp+>L?*j9+6KHOp!trQn5b4chLz
zjbU0D-KYFzSXz3ysq=n|t@bL%Fybz-&aXNsR|9&8wD`lsN1u^y&M7*o45Ga|72>qL
zr)^!tD{I8}OmwQ37E0OV9(FZPs~eML8$0R%A=OH19t{`fw_l0t)$#T22#zxOt;Z^5
zJ{TStjVs0PNUOiEs1EWKU1wKS3cN&{c$pnIij?`?r`X1Zb7oDtD+NdS`d;eP9(+fz
z->&~2<1QrP(X(@pps&j?P}FdATb|NVZ~)*S4W&}xISZ-Ge$KDb-1<vRBfUD-^LV!g
zIt_C@Sl^aeQAr`c#OP6V9Q-jpI+l=kE#ceXV6xkNJBwsSHZK{yFcK9Hxqp-Sym%$<
zLrh9AKWoB`qEBLq8ODgc4gkMJh$x!4lIi={0H`+3VOG;X|BhGKvX3UMI!jH3QhPC$
zD^lXFTWgeG&Qv~+bX#7*Q8U`1gVg5Cebj@7_nM~Jc4GU4=Jf|jCT6kgMJJ_6Lop+E
zjVh4)o&2dO+l3@fYgM-B@pg+P99xC1HEz*bkYQ@yS7)zrN3dlG+JB=8Qu_uMeg*{>
zYbzL-q_MC)Rf5H?P0}fk#E|pyH!>55Qn7I1J5s0|`ya%@WaMdKF!d@FQ9wMSn<@9;
zMt@3miyB>NTHoB^`?+Tfw$@>B(VXB$?ONPNEwg9+4_c5xdiAc?H4zLOwz#RxHm~5{
z5p)}@af8QIrUxI;*XE-jL$6ecmPA*gZ{LZiwC@41oCgY-xDQX4)UZLa(#KE#C@eT>
zg~E>zLG_h<iSDDXY66lfUEtdOL%-hA``#z*E0M6JiLNupP_<4<?W_1GUMDp}O?RCe
zeMe+;QgPQo{p`F;C9o?%eR)2y!N$1D%ATg}I5sM5$RJvYi|p}x{>f^(TV*fB_o{#l
zWcT|i-!R55lJ`S0-v0cWov0fd-5s_Dtx`o4y_Njl^+vK2BrWhgI2SKMvsM)x#-Wqx
zbiJ-&2S>fP2p5N=bqjrnUb2bTlKy?;8tIis^ewa$PNAG9Xej%uf{3S8y<b#}Yz4i-
zH#R*m?yD0j;G$+45!tG(9$ZYU^c?<o%k>0X?s0k4JXDYbXABu4um50;vNtJFIaDbG
zt?z*!6(aS*lQc2F&(f5kLOjA{#RI-OG!E19og@K-+7pT7&^F++ym|$GuJ)JfM`@X&
zxnE}HRh-tl0qs^2AK(&C!x}Aj^o&HNV{;4pLc<(7p+kekcWzOAvJl}B>XW5TlBG}n
z%+g2<U^!yShkZ&>@g595@P7Wy&$#{qsmZdo+E7MBo4aXa_;SK~ShLp$)jFStSDr-&
z0VfW>#%`$PZeTaTbVoTNzr-DR&c5S-n=CajSg{IGS%fc}2??Q-SKjWyaZ2sUo$qUn
z`ccmNztm9-olk!1MpB+@VvY3<Uuw_^Ed61(0UrR<ZwxKqIG#WBcly~=D_{9Id_$6?
z$#AO6-nyJci9?Z@pW4JE!aCR?bq98*3Flb0i8&MTQQa`up|CCC;*R|hBs=A?rK77(
zg6ywT(h=<j(8)6}iS)#e<tsapH>i^obW_E|K=3fDmB$)Vm_qMcm?q92(Q$38fW-y9
zO8qn_%CIiULtvb6&HZp25;fzXuHJ>EQ(on=u|wMalaU|;|0<_R^KwC=3_*i}ttYnm
zHh%0iDO0Ter>B3pSv(gtV#?I5pcuPIL1xuJ(=t=u%dseKw`zpwh-81}TsUZ)#rP7?
zm%q4Kh`b-fTS&O&?A#y+N~~GUAiG(E@w*0V0XjXgs{|&Bx}n;-SdRz8jVvh3da9(d
z;w1J;#g;9fL4Bl~P)TK6Vs7upM~}9vKYI7R7WuAK*q-m>l9Nue)X?kqe6d_3v*mGW
z#k`07-bANKmBzoD<rz>lJV)ttkbO=(0FS+6OXE)?eh@Ts#+TNZS77@VD|7ix!Qa%N
zl(6MOp(320PW}^Di^ei}!GDDvrhOgO>$eFzY9mnk(1gKnR@au~%TOzhQvp}HTy;NV
z3Ib(@J&44C9p-%viH8Ei5ZRe8x7>VmB!^0UJ+$p-a6S+Hv^)LET5htc=l~-K*8St%
zM}Bn%Wi4Yn?0B8uA_qTYpNaeas72vir8|ROUb3O@*BPNvyLofZOy?QUb=etTJw1NZ
z8hnj_3C1H%I2(jAey#oL*~f;h2Bv~6*;MJk=73LR@(;y4^^TAAE$R;zA)E`F&hq3r
zS|-b>*!j%{`|5=kjymk(L4hkknqe)XCR82WSOnu?-6*vTB~@On_m+W>_>SOaW7?^(
zw=)eKBi3u6l`Ow3fLXSG*vCSiX1g5mf^YoclR9ym_O7tF+S#}-J_A365F1@I-~U-m
zVsd$qJ@~c$u@#4;&E&=*#6n38j+ICBzF#OZ;v%OGl}6$bpDcH%(My1mURq~t=^<&m
z>!lU{4X)eZHGs_uC03xWHTY?>ce`}Xh2^eQo01NpGx%ic4_^K~4=H$<>xjQ)571h#
zFZ_hw2Rc7$8^Buh_1uZCSjpX)YplNvi{@$TQ?`PX&`8~&Lrmh_N@_0T;b{K6P^CzI
zF%I1qJ}>aR*ihJUdMIff^vwEeFGBUNo%dilVs#ul2AB|U%^qadSPZbtSrxuaw0IYk
zu<qj_%1U{qJfTUp%<Zwn8|gQpoIQ_*Tze6k<equOOy6kHMBwV!e~;T8hbXE9T&3?c
zojH&`(*fq2Ui1ZDo=ljW2?|L)>O7PrnP`5#Sr6mPG88Iu|6su{Wxd)1xG;e4^z@66
zx$xe*GrqhYg2wM>`tD8^u{~RVbda2XmMv-au?cgk&tKJoISoX_1^`<eMmB5jMXUiO
z>XW6^E{kLYkDqH2vd0A2H3~cS_B^$P=rt}sF%IstDtv4{lTvH)Bn_}pExNxXJPQ29
z{Dtq_m>SBxX_VpCU|kQC*K5tH-!-=)x!x-1PO2tC&f?Z~uP+&C@2jV80#SIg^a<FQ
z`+ho;A87Hg*0u^~tG8MfvskR<Ta?)lg01z_9BGWW_}HM!%OEl99R*Z2@ek=K7MI;S
z{LZH6qs-u8g2^g#&L20K05P!GrL#Tsv=5M(6uVO{Ed_`U*8WG%C9H3lNB%U<11p}`
z!7pKVnChgGNIj-fg&ETl_A|@?HLNWy{7x)T51Gy$PZezVGyD0SP<DmbMJZLLZ=Yl&
zulVdvCeZtrCGSH(hrHarmx>`UM3h%{s_WmwzYz~+cZ|RQe0zbgpZGG@#wAsoeL_Zy
z!8&ei%0lp#z7fNKzC=-3%Epn2nb!E_%d81ocs4|}5hJkvIh5l!4OLtwUY5jvj$Cgk
z5wV_>b^Sa$`<hyLRkX<eYh!Xn?=a}UTzb=eDHRTjSS#{z6m^A%nNU&!B9y1PBur0d
zLTuXK?kKT7jnEIhn0whje%I76<BDsq*$@7{8Q0FWZ(nBgRtQ*tMrko0eR(ghStax<
zwlSyKIKz`9n`gzrh9h&9l3VajdqlaQsN@Zyrp0td#bk7;f{}r|RO6L;`^8VcB+7rI
z_``OC4rad99zGd~t>Y^h?$i_)$Hl8Uu83UbvvqkC52?k+7)UjpQQ>IsBDlaBC(A~(
zQ`L3<R37dh2TA|il%vhmFR#q~Y@DVQ`0E<*+s)`06G)tUf!sx?6uC`fbt>rh{5jSP
zkPKIkah?Pv&Pzwk{maY}9xh6M{%=+Qlo*diy#iI(q_kLEH48+f{XCZRx8#)wh}z)e
zf}(Cm|6c4GudrZrvUc-`WLFHt;@8;0Ju;b}$vWokt(9i{dBj>*hXP&9*#YaQb}=>V
z?JIL@3r=GYTHI+$Z?_EV3`P-ZdFmHGgBiF!Ckn|jj9>Y6`!4mbM#3uRGN&9H^$zFH
z*;%Z4L7l_!>gqvhUBN`r7BT24refF}F=A3Vaxi4F2)#a#3yuEPL&1a?#87b(7RM(q
zkpFVSnsETia*Pw)A<nLiyJC@9BBM5*B(<*$Jbx~AlX!&vQ3GU@_MKg|s+p$fEnXX@
zx^v-v4>gf5LHEb?c`S>&8E*8iYlbk$4gY!N(cfo)QuV)o-Tw5qLh<`2eE<Bd#Qgr(
z>;DHk4$6o3|7ZE|pZ`@pRQ_MMB)<Q<KCFA&S3R`EB38hNqifraT#jj>kT0G--Pq@l
zG=^2gD+H9ZUGy@B{cDWyvJ1v(eF;>#jz~O3gn6g-jfch>MkSgy-nzv9j4@K|PH~kp
zgvc3zQ<0N7hn=&5O5;v`qWbu*;=e?yW2g&C?ScS33tZV5O>u2Zun<;=N3+_$%&lX+
z(m_$+RFx(G;@+w*_mtNJzpIYe66>AbH-)KThki4ARDU0{+Xg7WgM+9U0FRXdNYFs9
zHMSu&@&D^r&i^+MnkhWy0N0g-yn`pFA)?UaJ~G-X2u)svqWk<oRjU@f@l$T<ai0Gk
z`%9W7WeP4M@~A7M&FAnl+2YpdWApjD-bkI!<>^V$$9pSvaU3wU-SxB~9pbL2xXzxn
zufyF?s*;cMb$WF|cXY4W;H{=DyWC9@=X-eaNP5wqScoK_gT<v(I0#cx8>ZvtWk~C<
z&+eMXzVa_GBlM8J+}X`sUwS^p6cqzy&1SJ-3asTcl@7X8tW<c=AKG1+66;1ltKLN2
z1VEB7j!V{u>c%el*mG%<@4^kFvPL#X%L5gjH3{_TP^})&l0}6asX8WmE~dP<=o%C$
zlfXXg%wBS(>%I#q&kZarKqkFlPMdE!E|OYIWb%2MTn5ALuUo}hbULsR5(k=0^^sW8
z9*!q(g?>jmcLJZA#E7awyg<VTMgwh3TG>NUZJM$d3fyybBc+j}*0ykbJZw19dfFfU
z^n>CA7SCu*swY=x@!Ybo++iK22il@0L_`z3%(kS^GeK7t$uVD`At1D9QXZcCT#yA#
zRbauym}MV3(y-&Oo)|B=X9w9C4dgNHP%m4VFPz%mL%$M)Y%BhWK#m<-aDM{eBw1GF
zSeZcnfV)|D7|KX(S&S9$CppUF(mJ1NxV6$GhbMk}d3WkOZPSpHwcK)??jsGKocWV)
z-?j}iKU<b-*Gt@mJ~}>BlU-5>krKGhAR_h%x2|>x+BrrVhA=TBqx>{BGq+Gd;W=rO
zdpmCX*X9q*8-+75I){~}Hta^s1e3Is_U+6Yo6N+X&wCHj)9v2cle~ZJYRmSv_GZNq
zk$k#N)#m9Hm<dqNnFwuQ-x+lV#j5ei{dw#%X69{*qW=T@3ujqNhO$zzR11p_^IeP~
z6@7LdrGrjV_=<W?>&^Rj)(-rb0(eYIqcwTT&4yOIWF>+7(WR_kMS~(G01@;N(6Exp
z75Tkq|4xz&YQ}5Sd2lo5^#GXVj^FD}YUPoc*WA^V?7f`_{fzdfT4Gxnr4Mfie6!tw
z(VfRb%Tv?ZlTWAYYT5*-dn>+9U5cn9rAmn&Tst4Ig>#;r%M!$-sLAiYo7f<wNPKqM
zYbFQtEY>(62;5v8P0>V{Du8{y)BC})=*hY^WPfLsqo{d!ws1{0{N+0h+7XF8>ub~s
zyYV&wGZRe%`xU(J$<!yA`=h;IC2;o7>L(H0{z)?pWL@=1;Js!e<UZ37s6WY8<kCK8
z=U)#P^R9PC5C_mNS?ora&?dcFa)yM%E0Cc=y><nK^te&QTr4htOAEf*!0M`_$#cu~
zbY@=+teoc<{;-j?Ei0nJ!x~I^eL~KF<qFC317l+a$SYmDk#u8X{bMSGy*<2cqmq+2
z>04WZ%=!cij<Q5{eSpF``x32@%sP|%epd(r<=ak@*Jy9)KplGfyVR5LO7T4UlUv;0
z*d^e%n^#blC7t{b^IpTFrFVI!tenB(0-K!6q|P0-K4+*b2^TD0fWtn!lTYten^&y*
z1tDbr?M!mh)SDi3T5DCqUQ;^Te_m)c^*5B-XW62R-&>U2w^ve=&0r97@A?wNlAVsM
zj;J_&GO2oTmxFHb9;E^z@dv#+TFM%otyqT&Gakn&N{9v(OcSo4u`F-(1tm!@7CF$7
zdJ%6UY}IM*uJ;`rr(s*uPSU%XvW@j7C_H-AJ&;Kuy~ashPu3|)7Y|?MyBzPO-3oEE
z<X^85O>)c&zPUQ4PkC^$EYhJ?qKS>8c{*(eUb!}VOE*F5XTc7i@a5q+Z<rP+S(F|Z
zy5|h{O&cP$omw&-&>fI2bZ+0LCwUJI*fZ#%yfF1Ly4T_vpUZ$W@UEux9J9n?H(Z!t
z)!7!;53e*R*D*b|`FMn03m!mz$kwI_oiJ#;JD?J{m`AQYXRND+J9yOSCDZt)Gz`i{
z55d^Bmq}tvP|iMuhM_pbtMT8m)@;vaoR;0?MtJyf8Yt$TlYM%VYOuDriYbuM3=<T^
z#LObP;fza`^B+}e9!|XcMl>_2xMOoN94Rj!^TWT#OH*CxeMtYyY5mCtw5vzH{F{VJ
zkGrgk7bR*XVtHw$H|B@oWeG1_zV57frtlWnPE1xD%S!NuedSz;)FUoK$8v<xiUh>g
zvr4atpL7Gdm&ZU~Jky%pRPb?oaM?K)+*HQ+p(ju+{AICw1wC4?Cqzw3xl9oU>2?8@
zP5I&r>+enUI$zgh)aYDrRFoj6iD3Rt<zt^OpHAYr`N#*DX*jxwxjZyrDbPn`iyb~5
z;gB8)e*D-ZW47dsa%WoiK8j21(Pi$N43%uGj@xf|C%4RMYayC9_|EFx3y#Gb>)&ZT
z^fq}Z6nilwpm7~Ka<MHioO07-`CJVh;uG06lEM43S2B2_ZY8cV5;b<k{^DOMrPdX%
z&PG8BvC~_gg&e_$Ga~WhjYGJYs%3F%d^L~so+4UDMWJUZ{YyZ1Cu;6t9l)Bb!zt;V
zS5~~CtsMMJxF*#L%~~J+1MB{jnE8D<jNSv0<mj7c2QGNWO{Q*}M~0WBN@jUi@S1WY
zSi3lwSdhU^Szk@`DgCit?@p=I-iiXP$5T;@s}&F1RIT@2j5}KhH;47C83t0?>S<hB
z7%$vSiZds36OwJI6k;S+f%#pnkRx`r25!-jE3@&5g_JAYXib5Vu_?CU@w#qkDuTP|
zERU^5mFqPkrz}>QGzfHqGdaZKS`t&{Qp^aNk7Mby@;hSMoys`&_D5a@d%B{XQ98j<
z^Beuk#oP8vq!8bZz|!r{(rv<nl5i@nZ_TcR<@;hLV#5DX<l7g&p5h`OBEoule}&&G
zG2#p3%rfP?(v&$R@!~vs=k5sab%1rZm!1Uh|1fo?s-Vm2nz}(z#i*N-Zw6K*oDGns
zQT?)poFew>G`&A49+Y{y3H3eY7#tCUqR}{2_I%GT%a^3J3DaMLqJ&p?jTp-Es-y+_
z)-p;=DA{#7C(=*1w1aO<x58(nAYaazrHPL_7ZP_tjOdOoJ(b?p5czNH95wm+w6a41
z8C`9w>hbj`$_eAN=Hlj(O;y<1xtqX&R|;=B?z^`gAqX8yn9dZ@Bby|KrY+qRy8kni
zbD>lw&DJf?(nZ0K$HP_1+7YGZ;E=vLV}GX46kH8OM97+_$zbR~V;On+l|T2kAH<+a
zl78V4N3~R^i2VHl_08Of`AWHhKrK^|!^zMC^5eU+UQB_BY;UywTp9k$OHS(yNClS%
zLhcLu)fyoy#E+Es-5Hj#Xo=2LT`?T*A;bvPo??~Q-)O9)=9LzVdoeLee5Mq-V;5Lr
z{r>XRzjNuI4G_yo$PYZwtz~ZDK7#DkZ=Q>K_5E01wBU@hcL0448$Lbz)^x(?45oro
zhy~i<;0?1}FXV0@O{!u2fk#imKVIrI|Lg$f1ysap#&xf2OhfNwDZo(a^v!QQ%?8u$
zXW2JL2UfM+uqE7eF{<+jJVJ)r*PHO6$^wmC%TvD)<eE@{>|qNDKtYRx2aGZH;+W?>
z93kSX{<eM-V;PU4mD;cT9bH~oj_)V(<*?Q1#M`$&R9%RiWcAqJXqz>Uu6_vlcX{{+
z^Ob_D&rIVb@ACJ(8Scb$H9wWgmfADV8CP2>?0JD$ysxBV`5nQuL+i`jqKfiu^{@Sk
z&tC;uoZ`}ce1(VM8JjEkNLcUuN_0}%)`*kivCpxN4j(*>6MkfkE!z8KeCK}=_kGbM
zZB8MlXmn4?|1M^DQnWlZV0{201%No=@9HPgrOU>%&Q<<%DP%||$Z<$StpAQ6;7{FE
z@ES`8p-J702W#rom9tM1^g3@v`l2#T&PUdR#ZM9k7ll5}-o3&1ud<U5sJ{g5kF#T|
z9re=268!Xju1=<PpFKcpuiujy`qsMz;O;CiwqMuAqW<wLtx9UbtSCjmtSPUpMY~b}
z;52eC_zn}QU+Z4pi0S(c6r3V&pOEOM?Yv~R$4i<0bwy<O`~|+}XV`h3*CyE2MB1Z&
zkc61uOueO-20gMHoLRdw%>qEHjMmnj>`fam6lpxI+<NLAE{EtCd;YFrO7Vkp_#qpp
zmfIK0wCspbxuP!8TF@GSe3rFOaTCWStzD7e;h-+N@ff<1@9ca)|N5}v`bOk`k4i>~
zK~z)aWgFepS-F(>-As`l+)alF5%>KkhM}502wqC5HP}v588D{mc#i$yl%_(wyJs}h
zzWnh^a3hI81Fa|OB~2vE&-&=HxHsgD7$A&l89jfst@(F8A0Be0-(8va<tChN`S8fE
z^Ukh(W6uE5e*D0i?Hk-V$=}_yI5+T2P{0pLuJ}UizP;`x@&ifKItW9X)m9}8P}SZQ
zLR}YP4Ub{fmV(ylnxYmX=MEw!UKK_tM(L%?xInFqrkVhVB&G(O;@rRzq<Ltnd4peJ
zIc~2W9~`wRx27F-T*V7^$w%ianzVk`8}$+Yd3!q^+9XS_ywNKcRl1tL2Hi$Zt`6FJ
z2vdJwjSe0`%Gt{FJG!G6{qt4^Jxmh6DejrTzO67{jMh!vYooK>>-wp9$$C^_!?iBp
zq0KN`;M88dabU!V_{|(_xq0upV0#T|qtQr|Bn538s$dF-1PJyWg2c9|J9Qb_o#F2c
zndPER;t^{ejQWO$i_R9sHoRaPx4+}!z_LAo4v!=^RgbZ)31zlTB)&>FTznyG^IiYf
zb7Ks}Lru6+LPEc1fA-G95uGakZ5_35i*l>S7BwdPSIvs;(Z26tQI5t#d$!&%(k~z7
zv}rIu5)W8}loqJ;h-1uP@p3t*q_)0a&Deob6I!&C%5qT=^gb7I7RF`I6u^ZpRWIiv
z=+7<REx!x<w#&TO2X`T4cIEA&ZldocY-4d-qw5Qe;YqixyG8dksvPnKwbZMt+&r^Q
ze$6!6F9=ZmtA7TEZ)wUG3)nP4-H7(h)K_0T5T2%?(4>S#wzkKU_U14}!S?;wvZ)=_
z>*lq7=AAz2o%@HOtdQMNn&oI!fml(8s&Nr1j9h$DeDIR^XZu0n99E;5bGLVc4U$S>
zpFA@KGr<9Gnc@k~k0oTN5c`Nu<<kL5CB4mR6wes!DDz^P9I;d?UXmxFA-w5px&+A!
zncOKRvy5-|mRuZqGww4w`#^A1JU_MN=mz0gcr~{L)d+!+Hd}o+G~)KAKOY&!!!B7W
z#D8Le+^BcG9h#;UiXThC^aktuWj!&K{sbk@y?Kb-P!~$hak_4f>5Ek3858kZSFyvH
z<Xl+4%C^`Grz2d*g__0Zr)~@ZWn&f2U1=mzRCHm5aXcA1Y|)#9jZbqieT1oNkQh8r
z7~@m<*8}hNk==O41yvf+1&T7(pr%_VRm+xjxl{0ugHJ0fMBQt{VSZ~~!C0z0O3BIG
zxAdK0^@1a<K1jNsv-`$8z3(?k7Hru{QDiHCU~<U++6!j@E$fMw1%uqi4WnmYnB(z*
zYeSj0Os{=;0ZfVph1IM~EeQI3uwu8xzr!?R^yme#p=*5#n@G+5%<i=iOu=`{%&q5!
z*CCI$M!u)jDR;nt<=rdN4FN*<<G<;!muW}oGmZ7>H*)^9oQ|@pdIXb16^7gc!^HW>
z0=>W&jZEM5G^M#`U!MrDi0@Y|v3_o&OQ$%R&7QJ37&@fL);^(ikMa=x+y;2dBJj{i
zhk$5ewMf=l(WJ;c^uKR6D|jR`&fXj0;7_arxWoO>Atv=v|MW!bTYJnE!r8L1Wd>M;
z)nrn5;UowVPX@L9y(~ACqstn(6-|t%@Fi;A3#I(_AkwJVB5`~AUAVST$A%r$d}qe~
zjsu60&GUe}p_myuRr!3I$6o}$A+)--dl{jNm;K9sCrpUvk|wo`|8;1%f9t;-4fv@4
z3-W-%+44!Xe%i&kHnweK>e@9O@jg`iTY2(@1p_36+X_dOwU9jwsg>@ky%!lE8&mlf
zt|yGg*N9(1OORdv-t9VN&@&Q8t$Znh)4CLfRM-oS9Qj&%Pb5HpaqjWeBJ)7Ou5==T
zQQH-xgD~Tlk6N|(M*h1_?vp@|>L7}TX;0QNYzcNqKPV2)mPliNA0-%3H#RVO5kWwc
z1Jy4--ZldAesK`k9aL~$6banxDz816mwu3b^0hv$%#!fY*Jm@nFnvdX?-7_QQ!^CT
zMOG1rsLpS2QP1i%At}O86HdFUiR|;|J9E-0sqM?0JufK<vNLzRm~QHU$cpqYvitM7
z5W!y3T4axhZ{Ol^Do!d7*VV+)k9OkM`@=&D+oxJ6m`)qZMiuoq8>jnDb$|5pt>p@z
z|0Afn4IC$Lz)+nK)UP*%qeS}HHLWVG_*;9Xt&CN?L@#LE>MJr6Ng1%6y`Kk1(dIde
zsP87ei?i$f)&z~NRPFIFPQH@iUmPeOQ;9bc^|QM>;wRx{U$U-Ez?HR?wK5R3c9AN{
zizz0dNXVAr-}(3p^u{kgHgv(KzMqF?MY>X}T#r9Ev|kfdx%(knTd`X-aIzLC-JdZ^
zQQJoZDKHY)H|UjM9xl8uYO=yGx@!_1XF>b{4Y3AgcKLLwcYkiu?2$5e6G84v-%iJp
zdGyKiP(NUkx$$;TKB*t$Tj!^(k=ODDtThFr=)_S(lY_U~o_43pFaOD)-Es({9f;=p
z!Kir6x$qjO%wkSs7LAIVVuY5S{`uBNwFK0dBH1{WmrUZ=7CA}QCK28z{tphdxL7Bb
z^Y>pz!$XEk)jH5lDYSvC%XIO{NFV!?x8=&<ioksY+0(C_)V{SthKor)#Y5ZABnZ>}
zf3Bcr2V{5kft0_5?A=8p*LuW`W(4f$!?{SO29tyz+~jX>CMMf@yhu$~C8VQP>3eiK
zKGp7vgkYjHUW_h)1>z%(C@D)&iWvRDNTb?jKT)a-cxJnIo`qXl<ABCmzCq~Zj6!ZU
z*gx1|?Nc1QTl($!n6(md9NcUh;jvp6H|&!yK&7X9d28IyBJSW-p|%{@aeA6vR$5Oo
z6}V{jkeVh&484wWw_%gMqWmCj+BSU&_%cU8YJE|8-C1h|z!I;bxc43!FEGv5wfsG1
zZHZRK(>Kk0*!vL2nv5JlWq&Si2^}&KZg9LUX(l3k%dp$aKA*$6{a4qG^<}WA`Oe@F
z2LeByr8ucJjmo8aaKI+2?Sn)*G)IrGmZ+8$M-qE)?{7Rx#iTXF{HO_hphZ$QlR54A
z!)&l`Qq&Gk<q93ZnwX{;A|3>uUwDnyi+DdSsq-*Nj6c(-NR(G?)D>rfg?=}V=W?EW
zfIoD*w_fhmNJ;5s@z$x!ilQhsVb;glEB;yoPSwO-IUak!EC}s>(2FL7b)v<D7T+Mq
z`s#}hKPa3})H%7mDpP}WjOk+4ht&wGziKYIKCSg5PM&E+B-XduW1BeReYlb5BTNY~
ziJzC7{4<Q2U*BWcP8r~4i#-u_oaobN@o5!f(~YqX=Xm`vaW~ymmaAvEqXuSY<I_+J
zSwkY&+NZu{WQx12B-%O1nn$F|8OJ{;(64iTJDIMCY`@Pyt^FuQ=HWVwI3MSnxPgcc
z47`opv$nkfVZ+oQ)5M{-KJ$!ueJ(co<(lg}os=-`N@6s<>J!AWTHS^TQTnpCSk*wv
z?22}f?)AtSykJb=7t+2LY)psxdZ9QGNSf+|adtL}dv)U0KKZPa_D>V`1D)C-l}{-e
zClZfxV-?lov7FX-=ujVDh1DQw5(ZM$$KP2uSx|CCLjAcG6Od(z5T@jb_d-@C*qs=;
z_gUl1_&oB>q@>~!kbM4?zQr>Eo`jH+ngOK|2EO1?iEtEB9Q6I2=Y?3v(C$-!7*wV*
ze4xvVDK&ED&exsj8qYFaw%@lM$GPB~H14RYDpLecL$???`}W2AR+np-j@twW!iEpb
zEuQ|FSY}QjAm<|Xs>(0s(==L?V%`*XP3S_}=K3(hh$}~FFR$yNgC(Jvyz?qDIpJ|~
z@=>)JCw(_x3vQ&SeWRC=a8NB%*lnh7J>)B?o|j#}kcTWjNX2Fo2FrH&4<VVPdvv?)
zU5Pg{)f$?$5NC;ns!t#{3EW5>YvSl+PoM#mgHRN>09zR&PreDK%|~9gjQ{p{DEWYt
zAf4u@moyo*sJc2vU~livflk3`^Eiq+(VV$I;k|a<xl)7bME41r+i@EW;J8slsFS!r
znWa&hX%i#cB7c;IMdlUvr3VT#r|(&B`ML?Scew|oaGS%hDK^j%a#Ns7rHZ+Os$bTu
zfHB&JLA1@QKQr{B{X$iEqkNQKl#|!Qz9*-nSfrIIa@{DnIs0JdFpZU}r8QIo&5@J+
zi;Bl3!nN(x<hx{8<xwDB$#T#Zv<AtqW+iVlT;ya*(V~uGyvgSHqNo*~lfAay7qbSf
z>U^a;sV8ydqF3&-K6xhCD^@gfFYi#!_o~DCv{fvEioLfg#}Omb-RZ`VyZXLQ%FZCP
zf;Ev8p?5ETRZ7cGpHNR8X)+w_$*_w%Z4E#5#>@B@8RkPOe9|6K0#a6@0)4jiYCh>k
z-;PMf^s0I^1F=5rk|rgt%C{xuLWSX0>#G*?CQEl8DWNpF)xIBKL=LyJEwAgdxL;0j
z^iqKNT>47I?HMJj_OW+u7OQVOWfspQcD)89XOiP8)!b5}{}xZdGEGNB9YTFE5YJDe
zgJ@tneBTmOTKZwA)DBzT8e*@S;|{3Vb411<ne2n^y=|ec><vfsHd26`pGTS0y3dGW
zd}T<6QemTku~u^H=8oDWj6@_zezBy1!@FPYxeIId?IW80L%m=e+goa;m%l^km@)LS
z^XvP?O(g{s?!kTOAmQUWCcHK5>l0Ib-;s*MD~>HPh*pX!(S=)mD?`umuio}w9CfXu
zJoR-~FF49!Wuq?F;~*DLG!;`9c@**d20C(EQ%!fG?Qtqq*?tGPn~#s$elOE9O(^GX
zPSLs7i{mx(f|Arp%r9l+?J2i+{1kKD`P>h*(l*eEdM2~2bQRi8yt>tn6ww;cCrp6Y
zTXg80yRC3piP>L>H7H8v7P=@C!=#SNwtB(@n7vO8U2_AN<l-1OqCf&??e~Ki`)c)8
z$Wn`0OC19;*}dqVkg_XpeUI@!D(>3s2B)lI&Gq@9bhv&e=ilk7aWq}!dmKg{7QWIM
z@v=P;wys}HF<2UJ5_EBHxx+L%LwW42B5>S*+k-QLe`><><Dy+h`|chOo-MvqtIhqZ
zKVQA4oY7vSoy^&<2VY^tn%j{g)W3ja^o7sW_dZ0goQYuNCtn0?WYaEO=yq(hFwx-M
zhjYG+zDdlMPWbTHOUE81ZT`Y+Ba_ek<cT@I6{U8U63xp2@0T*IX~z#LnZj$cAp&^O
z5X3Y|xcW*^)EuNjSAKls#fFl{^t0Mo{#$@GaV%IJv%YBl0S>|_Rr0sXZCKLQ_fw*o
zJMCMxP4<13VBCjHcAQbMF*DNd%{j#t>^x3iykJvo+)BJdsN$N3yu2k3x&Rpl0uZtO
z5spb0L^HDp?5Zm6Qc2<;qRHJmE2T;s|6CiJm<n5t`)Zp%qQi0d&q=J=CxUZ9ddDa-
zqym25nKm>eT~$>s8MajKO=Z1LYOkW}b4JWdN>V?ek#)3mC2-X4*Imyf;X3<_7HcwR
zMy#Kp0{z1(_3i3U=eLG4OYA&@FUdSt;<~Wrer9jKr2!#yPbM^0wk48Lzy%Opvj?W*
z3i$_mJ!BkZ=2{rbycgW(e&4k`rFPa)8)z%09pleA%og^hbp;f2_V@{G&bH*%Os2e-
zf3U8uozyFzgQLa~=P1`gxw&td4m^1(?RAvQ2d(-qoFe$%rxlc6ztin4+J&_JIb<re
z!FuZy*Dl|b>5L4>_e^WD^GqDqSCOWWEb=%fRVD+7p#?`bY+fugjBugY_XB15Kz;3=
zGcLq4r(0tz>_*8yP{luEt@#Jok68~Kk*CM4|AJ7TsswyM7^n8b{cRn(zxVFZCFq<y
zKUb0Hu)-2$cR{nvYk`5YpV>>AR3l_6K%<wMobX$XA;v(mFkuiCqPC%67?4TVZsrrB
zUSYU#rq9C@3hq@8#^=i|udKF}UvhS3IfNm$r2})Du8*qxE%c)=Z~Hwih(bTr9?eSx
z*<aCWcg_*V9TVe~4qtoCbc$14J&+pMBOO^3$V`<F+HP-sM8`3IIb9vk`cDN92NLX{
zI+LffD45NIqMmBM%}*Qv!CG-A>8*Z+aDudCTYo@a+eaztJWoi;3TL2+K4s!?Ww_$Z
zCV-*pz()Nt_?|4(Go8*$Sd;rBAb8mY%KXy13rI4zN3y|DkfOVGN}u#PwQGfKbdFE`
zf}0$*N^^zJ{g24i$1B%3A1FSG+43JXg}^=S?k-hbfD|<GzE?~v;(RnPF!Rda@OX%F
zkU4+q>=I7)k<5l?@AI_OA$n{4;z_12(1H@d%;6f`Ztc<f<#&`EMYNA}Th-d1$n&ce
z{syutmjiu}wKU2(6q!!;sBKnCF8yGyR42`@+D0125T|_S-p+Cs(!&gr=q9jXxiaWi
zw|uq4J`Z06#@5JqW2<~m2u3F@f%skjA}Nh9(6VL1ojQ*kWNGS?gm{Ym83ac_3DnS0
zY!4+?o4$ReMy&nKN}b3#dsbDs0-;~be^AUX7HF!a>Z6)EOYtmatz$ccZLDAaBzpzy
zda3HCm42CdnzE`|Awv499(lq{tVoPdc!JuTqTVE@V;Jn!vg@;$N4?mo=+?JuO{q?&
zA#odhouKg4C&9^3Z1jksY`yN04%&TxE+TyVRRowsF1lW0&o@PMK;rqm{CQE6D(OXg
zM^hwDt^LC}H?~t;CwDVxlL$W`w`Z&B+B_wpaOo-!(endKo+Zz<(rGJQZCo3vnDF_r
zJ2Mp6+Ygk}DpqOkA;SKqUKoK3m4_D9<y~MPw3rXUWzxy5{t<Xwdbx9V2K7vaDOOeb
zgHM<*hDmc{Y!uj1j%dPp%l70xFmL{r7Z`PWy5l}8<wvb_E6^*ochkK<n?G|#U-L2H
zMW~it_U-eih8bDxvmK0PiMmqBfAxqyP`Fc&DK9~tgs%Hufa%frLGKU;nWjTnt?R2|
zw45K@W*C^Dhu@x3H~CpM#%aAo%37;8{URIyw;njjEL8Zw#@xB2ozvQcVal#ExW=oV
zTA6uQK5qucLAU%q{g;^fs=|*kwoyWELd~I}8^eHF?dUBET11}buO8u_B7HNq6!x)5
zN0X3=`V2W3v^}E~f1P!hH0aQ+rv7)=0lWl}W5nJzatc}Gs7-tFz0d8|yb?KQ@9TTo
zT^le(`2xnrJ4SEv5s5cTxVKgFHmpwxX{-7n8fMlqaJAHi4720)obA!<;<O)pY|_Aa
z(y)za<RyRDJiM$?VR|e3f?UmrGaKq&J2VY@qvefi!HC2e`1vM@Sh>iI>Hak4kZv#Y
z7&Z~ZLu7hFAX$tMP<4t|R_`_g5J&FCl3D1K-ptoh^>{%)RKj@fJ28l&8-}1PTP0|H
zW5OzQ*@H&eWW|6(xAo4&gX7^#dlRbRslE)N0+H$Y4RetuupSm)pzk}ag6q=aRmWJ}
z5Y?TwR5(cI!RKt;gFxN~wr3bAsR`{_WZD3r<xWn|G3UJOLJeEnD>bM~AJ*@Fu=uq>
z23w!RWqVSlOqy6z`X7bKtQjCHSm%@zO2dh3K4oy&)R|zaK8RoPxr&OQOF(O0k^2K~
zkVI&le0(>2ELzR*1}{H@N7>cB?tlE(*+%Tx0fpI^D8J{IGGgO6g|IDEDvi@|V|%d~
z?4GTv>7H`_VL$iICE0hTi;8`|-y`3Y$@K5mzGF$X*X1fm7~;$_sY3{NXfcS29(2hG
zy6skVJHf55%8LS1foskt7!MEDBqmi|w(+xPce~0f`X4QUMg)@+Vy~VCN6q-Ep9vod
z?zi1)(HI3trIM}-&oOMf#w)pU!)YqL3e-tyZIP(p-Sa{5{)3R)!(PlYZ-&m5+J6;~
zfsmi&VxzWneu$k>u7de#(<`$Vpa&bE*Wy&4Ydx0aCg+JhgL-pa+fOgFSvo-$n@8t|
zAGEnUHcV1+{TXeWjaR7O{?qc0D^X8_Mg8gNoDP9qyi#gKMkt%pb%U&fIW-5RFf>q!
zr0>*MUZ1J|CPS+-G!xNY{+4H4=HQB#BNcNcw=rQD69@C$WH$`X!@6{hO$KlJM~Hw3
zng1e?LHm=$U!s5l4D;&Y5VWh0(Ha+vD4+42Oj!LkD-YT%Mg<!$$OSMa)g|~?wEyZ=
zf1xlW>2T2KqL|nS+pd^@6N{6BA2eLgjJ#&!tDl{nKtsMqXJmnI8$v4FgRe9jwfRF$
zs@AZ+71v8JxpJG!rM2E*!!IuBN_&WjXrJnQ#Y(`m?&M<FG2O`xdT(Kz^77F((4nFD
zf>E0nws+_h!;r2+A2B~zh4T=+S=wUvHip|oas9}WK9;iK{tQQf*-V)0GRMI_dVu<b
z8FC2HjfmnR---#g1lyp%s^1HHXNWEg)Z25rww9c*Da*SDpb1>`jGhOb^7U(OSqZPg
z%FBv7**(@|h=E_K;3w^Fwq%RZE6*6%8<^9lF0BLKbcLuzs5rLTqD1cF4dt<L2;y3`
zMN?w3j`x9i4s<Pvh)nh6v>G<Rq?RlN@X;DH^--)CM$Wv~rr|H!;K&a@gEiIQlmYzK
zkN(;Yc>jvoc#b>*tAGG`MSNdH#lU!Et27zS$|U3zFQ2VS<9=Cv@LKB4dq;+ffFPO;
zbeVZyXG1I|;Pg%&43r)_F&8VG0?s165^HwA2{c$o#;T|nuuh4#yH*QQ+?mrRzYo)~
zcd+#+PHO2=p*BpKpVGvV!o5VaX!5>r=?-U>_o3-N-H-1!o39IfWN58=u_>K0h%HUT
zLtFZ%C+2C#Q_lnoJ9o_rjOROl^hjV1Sv#l;8NF)HuL$MfD8zd+G+Bx~f6!(TA?v+}
z&6hY3H<O{g(FIBb%(~K%zyRfH0lyL)%pI{wuo>txgugS^(6y0%qM-+ua~5B4Q&_K)
zN_-XOYOVO`kz9f3_nF?g-ip&_t3_lKbHdjiwlyeTJk?bAV**;vq5fX@PF+C6lleXh
zCfXDA>Y901JpB1_pBj|Z^bL6gpW~9VwyUEJC0hmiJ>C7TpqJ$1I^GMbmJNCoaJCl%
z=3P2^CYVnQlB5q|PfY)Y)9?{p7ymQ2nnX@<MfcKOuk-w#zR?lB^3yv&_tE^_swK;g
z<<wU<Ui2TC&8lV{c<Va6Wt-)9peoCOVmF)x_}u=h{P6C+UFFm`N^}Khygq<UMAr7$
z-nhEvcC^<mAy=siM6pMCHmpV6gM@xCCAH;or&DJiSPuK9l=X#~faKe9hk)^+Tip4S
z_9$SntB3HfDdU`94Tvk6{ob4*!z;pQP2_77YpBAj0EM7^j^~}oUwSJ)SmrLIm4n9^
zXugK*tP`e{o-vDA6}bqszLfv$;(k}GQo~=Suz#u>m)qkMTIL++k-f<Du~UP-6He<0
z9mZ`$wwokkP7P=NJgOFP2~59e!?@9&p;unrDN;ohX_?rBD~RaYT+0l2-Z}Un9cOKf
zVyc(Z_ibHdPBUN{s7eJ%H26Sjhu93M>3NOeyR;b($ldEo$p%Qa)M+tpBy)PzWS{EH
z_H@qn;l`wWyz9CrQcyPqqKNS<G@o|-)rABb?39Pu`2ls}e7+IPQ=jsh;|t$qW4F{@
zi2#r>Y-wNN(`S4J9lieNVF?2xf@X@?OJ>J1q$$Tm@`{5BZu7z}S*yWs=KM{))(nCA
z$dLyxC-CtTKgHGWsKjh_-SGG8|3If&K;@;3Aw&z5pBjgwiV|hWH!uF3Wj05MsK?+9
zfOIwo^kd9yBdPFM|Nq{5;eWsP--=-hAOtp3fVXQ0h%ws#w6S(TZSM9zKJ3pWG#meC
zy$XKt|6AeoQAI(xZsL{GdaBwtGOu5rFuIS*dfCM|l(M7?Ecy$(*X-Dj?%2yeit%QO
z;r)QCrdSc{TuibVo)9y1*!P@LK}c;eUj_rj*UY?sep~Xl5d9@h&fZ?BS?vU<!^UO1
zqf}kRfwOb@@cz63wgG7#WkvoI%rgKHdkS|C-94Fe$B(^~U8RRcOuvO-8WOMV9qI}P
zivagG`2&3+)LsqFS_PACV(Wbvxyd$LcVb#5*6-_45@CDU%9ykU^=3DZBYuDVMYg{|
z<N+1cOVDok$|FSQ?y+M(`N6xs29<=#wHM5Uq7A4w6GO;%%h%|4UAbKK`ZoMP{0bys
z#YjQ`(BASEh?>-0%Vyy&8!0%(rKI-H8=&>C0VDCvm*<JKZm~kFsc(+`^sn?0Ss3rr
zwOMu<H+5O{S=s$v|C#;#iL}X>Thu95q627%7s%zBU<i_~n;DrYuUd$$!U+OKyf&lc
z8hu6{>RLsQX~hvo(Zjv>WcwCs<<Galoofc8_bx-op+Nsz>SK+gKv-(}ak5Nui?abv
zJhTuVd~_|qGFLH09;?vKVcT=^>~1g*6T&j#Q_G}Y(%8OE*FQl~`vq?1&zW9xbA=&U
z&C;TyK)4{o&Gr2FQMps<6xX?Pk{h8Yx-3#<TEkF%(FyL;sc(>xtsPari|w;cqq>DB
zkntdOFG<!=!R)yCWz!8qS?m*P57Me%<b{gP$=)H(t*z^hV}hO9=ccTf{F<(}QPGH+
z4J>xzCTf$UsK=uw_F^7F^VEEEjFhy33>YI!8<WsyDm(=AMYI^U1<i)QdDZFY&K#=&
zVHLFi^?1RGNYoph@0hUK1nHcf<4ppxv5>S|iogMWYl*ngISg4bw!Gy=39ej%5E#W<
zrbxIQdaWuO!c56qa<(b@4s;!(Im5YN9G-Rckyn4s@W<NX+p^W7@tj$UpDDTeo|v9F
zy0=F**t1Y_y5gcDimaJKY9sHE3dqy4K+&e*c?F^f)|PXq;<*6MVUA>6y?~Kr$E9Gx
z^H&d`x2$$?qD%2?C-rKL5;_Yd?#)zR0MfnKd5ky2E+~qR9dx{yp5l^KVK`55)Ro}^
zDnnM}Lp$xXLjtrNarRIOj}fu(Q&Rig8eB{A-=bC~#V_AIJi<?3={5a^2g<HP0-KGn
zu*@pGT~`DBfN8NN{`a1=5D{`n0wp-`k`vQcsK;-D&tm{foZFqF%Ea;3qcgC@>`fSG
z*lygvZGBUz{YX+zJW}P8;vPEiAJuibD2x>~{Ke%4YJk9JZqi6^3ku&EbU;b%fr7at
z<8QtF9{ys;fHM<mQM<K?4^*I_+4$))9E79MEOGnWxx8Z5TV9e9I(vscDP}kXa*Fgb
zR^6C1R!6LAeKkQc945JV01CYHiSIY#Hx63zyT7NT7sYiJ<vHOaQV&{U!jT+NF{DCF
zUf~b(+~0K}6bzA8z420paXi(hf4u|cED}L6YV&wAulsI^I`Q_%wY1Qc`_2V1er7jQ
zS)$j)aXxw@biW34_KDxK?buwPuSyC;S2;$}&g61QFJp3Ttq!))`1x1+^`~8_a=v4)
zh3yP;Y}V$SnKmQafZd$o;(c;MhZs|;g{6U?V+We(j)iF?y5h1D;;O8+t(>u70-rr=
zXk9pwvLx~z0w4ILDbPb@qoP4_5b$}JT%vHg3TmDo-DRi<6hGFB^BDoIeWr$OY6CGt
zkM&27iY|g3E*j$toe++NqVbdD!G^k;utjyGmqCsnCbYD3D1@L%+AcB&niC#B4pHAt
zQ!GY(N1-CeH`OkA#hP4z&<l6jey2K}=!2}v(Zn>YU`Wr;!$M`PKz4Nm)Qe~EdmD}s
z+{XTpNrI2BgNP;beVf)TTkN#OlL@C8%~n)(-<rkwh4uZiPhFg!SLC$839vME+T+B{
z#nkm9#Ulj@S7pLGm+$3aiQuc#c5Iq}k?CXx6$X&Q_D**kdXq8+{XD2}bcvIXM2cO-
zMOLe(x(B~Bo|nS;!wz&*am_&!FGGtFxt<vaj+~Q-!lceI+=;Hj;(@0v&hHO8^`Vmh
ziCKZLJrxKIsLe{ETKJ0nTv-`h3AjFp{B*9BI&XKy_t~poN_THT$@(Yn+*^ZZqh+%7
z<rP#-uPecw6?Lf=ZA!4g@GeMQF^mY*Ka+tPsaX=oUV;Me|E$54er+{3QfhZqE!)?>
ze^VXwJU$EH{`B;*lrYO`=*-&CO-;Deo-36o?uIiIv?WaapYGnfFX{e``>#C9rnbz?
zSyrZ|mN_A{GBtPZ6<Y2+ae(VAE6crSh~^-ixs{xm6DO$TBzKC63rG;XK6ZY8zrU_q
zaNPtx9FO-nj_2$7)(Z+BLVDQVaZI{-mqn?h$4kFm!(_D+`|6#f`$y>X&*aZ@^qStu
zQ`|`<k)ZG?KWMcfhI=|NDuK3%iE09GHvjB#KPzo0_+@gd(NU%Rdib`kcD|yM@KkzI
z$=frNn8xKFE(Rg{P3fvSDf4Et6wMV+=e0?>{WX>Oy473j*P!F+-b3%AKmQPjaeDb>
z;V;?l{%RB^5Q*FyI0oDL8VQ6bbb1|}g%y&k*8~P`LhVIBC3z{FVP&)U8U(+tVfv;^
zUT<XzPS8mFX0q(c*;Nx!G^}{X@w!YlUVr_^eC9?PsbW7iYOAq~r`rM|^GSCa8!+yx
z2m3?emj|g2?~cQ{_KY>=YHfI@+r0D0Ro;UZ<2@M-3z_-90+}<DbMuVBUwNw?CvJ@F
zA~c#iJFE!n+Yz6Nd%lbejn_cU)r6m0piaYoKTSCpZCIQO{Zs!m8PQiU$=^=#)xCn6
z#+K96)F|h;bTzGj1t&dbfL&!uQ<7AQv8)SL1$y!8@~=m?N|PZnMf6{OK`l8X23;ji
zs02u8m^<(C<IO&O0ZMOPkt&n#mP7K`)V``aQ0k#+Srs<5f3fcYgy4^<W4$tbV#js_
z*{H(o2ztWX_(8I0qUvX4%{%!l{``#Lo8AF``igx#au)XWpy#)==v+E0Xs6-jkSi7`
zwiHn*TIiJY$+5K%IHt@1LzoNXeyiodkoJ9HNtIKYUq+Euf1Y{fa^y%DuruWmcfX(x
zUse^R^2WcKE1d%-rosM$%|~fK+QYJ$U8uN8MsclsR_Xg`OEaV5anaft>fytc#zdC%
zv4`=h)}4V71|uyeMF?RMwq8Pe2-US`qo_CU39(yna6!hJ(W+7VtON7ue5xmZ6QT>_
zbccfd>2V*yDHi=X1RW*Oy)Hz0G2GlW6Aq+R3aYO=i>QQVCJ_oN&|%&#tu?yb`by@>
zgjRXuXk4BKM5G9(@E{)?QU(lX`~pgL{1nB_&dmZiD<LFA%jWk&UhiIZjF&Tb#>v1)
z9(j6z004-?^<Qc~4NSp`tTP8^Tx)3f4^b4fs+;eCYndK6&Mx%wo9PZxa<sj<Az|vF
z%9j{YfP+l2BU^t@hy|tsM;xw7YGgtZf=8aSODVc(j!3;Q5~M(a6z>x%J}jG3@GJcG
zc2WUbU!fMNBhuShpgq5vSP6b9(N1{Y5;mzNfE{2<9oO+w&*oYPAUlZJmDoKc(_YXJ
zu`4sYGN=-J*Jw0IF2OvGdJPoY-}7}nk3^DPFt!Mf(LZ3mBgzf!zEoxld%K)pf1K+~
zO%6AB$OTxyRV-tRq~^$Kal96Isx-Q{)BhweBCVj*L@ig;F317XQ53^sZo+NB{pVE!
zm{~G-VM$%EYfse^J*z+&H@ClQJ1^D_O}XhS@~5dO^?tKY6*Lr+zeiaYTnzdh6Sla7
zFd$4zfdt86!@6{xhg4)#bLrMFgKf)@mG<@FvISBw^q`Ss$I<k6==r$^Gxki)L2tpL
zl}0c3^3D3ZY&^wT)4N#&-EaxSjPPE%-Pf#e57wa>iyW5z1mj^y`4C_#kg~$iNf5+c
zg}e-Z`c5PiydLBz+Lk)YlI`4B4a5YN=G5N(M2JBDtwVbh1<fjrlsXuEaqlD!kS2i1
z23*i3lul?UzdlEQVZ|Fy$v#~tJEdh3YOv5iN6DW4wM4tfKdteL$bLxDSBS7{Wf(j8
z5Kj-M&Vfj$>r;QW${&PWNn*EbI^DzSt9!O*!0}Q0$B~>2I`bjz)7Pkves{)?c9j!n
zJ#N&mlhUvCUD%oqF12!>4!wnaJnoz`Y5~i|WfjVDBG5j0)l9R4Jr%M?G-fzv$1&Ta
zVv*<V+4_;pTw08MU`vz#YX5tNl}1;XLSvI3Saal-`S=Yz`4uo^ie)VrJ9Hvxpy<`U
z!EU890ZH-TajP1=HQg~^d93Gfw;33}B2*DjR$r#<b3d9l8QB(29KJZ=ewy%cU)A+I
z&QsK2payxpuGDTsytb^U5NmpLX@_AJ6gXqb8!nuiRZA6p>-(If$YCNOVy%>1N=kx^
z%yAHE3{TK%tZEnRJhqjyLNGcp3eqLP#&$`X<}6390$Xs-VV{{i6)V02hcjp2Mt|GM
zZ=Ag(^UpBxb8h<GnZAdm-rVlZ!~WmCwY(Wv)t$X_`f#|cv$HJX=3#Utu^<nuAjwfw
zr&BQgZR0AKtK^bQ%gC00wQj);hx^Hk^QaDeGa~I;YPoKlsNcW03!d0rQx0b8$DDG#
z%38Va$USlr5=CC!G}ZmwZnV=!9oGL?)H_aiDgol{aMfXL$RONa?|uko3Q1*12VN`V
zH>5|%2h{doW_7k8VJ`4VZ8e9)i#4;9?EHRbvPsZ2T@N*mM$`&=8w7{A6huOYN~)an
zwdQhe*bHE&6+0Csfos&m)S;bEV0H&b;mU*rdpG5d(812Z7vpc@3NEB4SUMNkBM6!C
zKX3{Bm#E69+H9Ze?o#M(8{yquo3A<yM}tIT->=fuZEKOm10FYZG?TJwFQeX9W;NBO
z^CcjI4i3lS_gjU9s0;xscd1Zi=K8vpiaY%1ZOXxj!1*x=uS!(APxIjciq1C|T#8qo
zr9_An1c}luk&{D&IN(~4lq%NB7puV%n?WRhX}~6$yb-`3CYmJsT-5e_rtY?}+Ya$V
zIQ_L+caUpKpO*tlaeIs@S?}MUZi|;Ko@En>4UvndhyQw&p622by(mljA|Un?VGm>w
zKP%xMKPyvA7=$4gw*$@g;tU;zX?Ovvj!D+jjj4Ib?7m}?yZT}ZY)||`2TozHGMac(
zG=)3UXiZtI6Rbud;vt5V8UbQQ^1=#yax3;a^>yb%jE>7n1H3a>Th%0DN~jjt?WW7Q
zYS}@9pY$!Mf9UUFdj49dVF{lRe@p}G72Julw27kTo3~j+e-Tlo%2lp?5PRR$D}iG*
za#=1lzSb-kTH8cJ4cuhO9GCC=&ftHgjbp2x08MhmEET0NktJ)=$I<<Q=>hBhc-{ts
z&TGgw+}?jW6i2HBY<<9xHHR3Goj)ZY|2Hu7n2=C|jbZoVijbvqHoBTFD{@N@aMzBs
z9O4wqPR5l5*5{xVrQF}VQfr=yoQ`&btr7Dcs69952&4M{Fq=4VsxHC(>Rla(?M{Bu
zXZUDN%`yKzM^V%As0Nx(Y=%`VAuB`kq70j%9Vz$V6*sB7++74@oDVKyX)fo2Zd93^
zZOeXX9$q|7uwuS1Y-{2h*v(y`8@AokzkW6>D_&3Y$~UiMDeJpDt)61`NwLZM71y#J
zhvyc0c1FjXq1`HZa4S+NLTdbZ4Yv;HS@!-F#S0}VRX)LV6{;-d!?6Qm{k!GHlEk}?
z4<{jU$|H_j)UuNM=Yy0I76TFC;5(JyY35a|*ZZCgM(yzaW3JsI`VxYziplfr$>@xy
zmAr?o8f&n0yFXJy-u4ys=gC_XYtNr?-7-?013l9n(u<1GnzItK#3;DMSQ7B^fzf(J
z+p<1Z=5oL(tH8sjZx*C-k>hz2g3>tsBHU3xPHSKtVJp8Kq29dtxWzqy_OSVEEirn+
zGRriylEum%IV|6#my+|%xtu6E>^D8LThJqt923rD&0W2~7?Pacx41`fW`^-`ctTW7
z%(q)MBlxC9C0Jzny8cY?;NJ?Jqyblp=cgM|ZAS^kw8u1n6KO41<a~jj^KVUZ7Dk{o
z)*0{&K}gMecFVMyr@g_q_bi?;RR<EvE5#W7G$H?PC$WcRDy*tgD}6CEHqztV?<6-f
zo^9(Ux{fUEQTupp>BPVG%@0;2GOR{d)~ZLiyi?5}dF__9Vp@n>1zEqW)F5RiamEuN
zti;$;S{&SOA7+NUuOd08-gjFj?x||kU!5nu4gdFCk<x9doHyB_>2Ddf4>KftqR*DP
zHNRM}W8YRz`4IxjpR*ar96fRbh{^f<ZisrzlLd~W6X3Jf1qy>TVA7`t9APyPsnyIX
zJs-0-%#Ww+t54U`0B*fY&)R0puntrPz>87ndt(4zE`;)tX5*O-v-jP2IuPr0cGn=b
z9b^=(aKO+x_Z07xE%Gj&`tO+K_>O&nb=IikS+LIpc&KD2b)?n(Ku615TZ<pD0ix5}
z{B@2O_a|p>zY6Us+bg<t|47lbCoZigQ4b&Qe_(PQeB5?aizc>gP>1sJpAFuTOL4SZ
z^W~Gt&EV;|Ta+`N;h@tS*jj;$RA(RBvtH_~YBKfd6I_Vswi(+Uo6p%Cy)=?w40fHb
z?&(ruJLj-e(f6rVqvuZ8O=H$b5z$3IY974Vs^`+ER4V-Pzp@a5FPHH<J@bjmWM4Mi
z`YRZo*7-{je;8$aOA`kW737su3l{9_6m<jhz~aE?Qo)(tC}6VW&*OL7!?OIC#KKaM
z_iEOi*<D1=qIp`(R2tj;Z<WPu;hBT|BvVO*hVh2jVvY#Dd@*e?P^9E+U!cSg495;c
zR|s1tE}S|2q#(s<h6oK|Cqkx?o29kK)=3}ts_fT2W}v=nh~iQCh_$>LOW)@aj6j`_
zP9@?lO1li{KTk$uwywOrUJI|BYix54y=>`$wN55DQE&|uL=5?L21Lhpk~MK0C@~wn
zU6K(oqK5eY7g+Jjr&VjC4XIKSwYR@=ILRLOEuga27Q@wD)5l(&-zuha;Yy2OvEJj%
zyL4HXf4(Y1x;|Cb&2@c4!v5Rg!)ubx)Ye$_Z&ifvUe_FF{#g5YM(ck^e&_|_DJS>E
z^6mQ0dmLXJ+q1D6p-)ZgJ*_Ur*6IPNgNcgj&%{$wVspAT1kc^*d|9(LEazg^l&SaH
z3nF1`-L9zP4gXN`TrJ(LUQzv;yM%d;u1C=BCy~Wh$|cMUA&+JVJWN6>Y%ST!BwFdV
z-S3&q8q~peV$C&06#Q+j28tiQRl@&vRkVmKFEt85Z?7+eKg(?zdwlVqB+8C<e6aDG
z3r<5q6>T<9D~w3xi!HQz=MeNl0VnNN0^f^m8ytMJNnRq}yX{jX8)*C*GS=4LM>&N8
zUg`KRR=OV;y>k)ZzqU}_OT2f-VA2{8+2%?JD;KMV83KnmiGMn-a5t1*Lv$%f`5%m;
zKCU+}LF^6T3@FnjrY2EVmuH71Jw2W-8na!T)-JvrXR_+w_M0ZFm}#0&k}ED~pmt_!
zbzSs<3S#+rOuf~6!N=oZ@BSZ4G|3R%W+Dk=2nFCjxhYOF8qrtY(?h@JqJx&>h+F_Z
zNH-VMoV)V$a%dGCG*of6c@vzxsu%1gm{p9LfZ&5J=y2<U#}iA~H6>9ih_X<0*WlE|
z{vE}}A4xgyrZmU-kGU+3&LE1n$^Ie0W*k$VZg$Z40#hkV%1PJ=T#s~7AV0fuV41Q>
zWb)X8awZ(^d_^5~*G)ACp4tVR2cVFPLc`kTm6pitf#p86#EJ&k{Pf2xfd+{J#6SZr
z#?sOX*{!+EuD(!8hQis0Utmw^29&Pu*-^nVkeUYs>2do$3wtNtg7!V%GS}8HQL)#F
z0#`3p$-mi~!nvP*!?$U01XUY0)ieCV<-!ccviL@+%PjG`)bf^RrM!bzXS}~sAMl@u
zF^04iEWK7pUG}A3&Jz-$u(6`Om(ZWOC|~$$J^sJ#SVv)(cF^_7-Uu)=niT&zp_URO
zTJ(|Qizj0_8Vj`#1(}Y=)E=@ttP<S17UrU|Fp>3AVqfLCC5>7kd4)XL_d&dF0Ewf3
zih%um?xH(~$ZOs?VU>1bLbQ4JpVo(sP+1PUnd0T>@BNxZuV-ZMyz#lHsCQNII`5h(
z`v8)GtCYk#J6@SiU&v5P7_FrC!9Mc%2rbS2Y0Eon!P~5^xmqnJ@%Y)cVQ}4V^j$S8
z@vamD`Q(n0v%(_51R00Squefyufr27s17e$<C=p=cZiPlWC|Av5fWdL{%Z=+>J51>
zWV~>XUWz)l6=AGoy2au>1J4Hogw>kmepNZj+c{H&z*o;8M0K1Ys-Zbk1(L1p-#LoA
z3(BoAGA-eu4KvV*`@9fiGq2l%&vkWegFjmz|2H8eO|t51pn3;Q0l|8~qCT`4Sd>it
zlIU+F^?5Dq8^rej&<LU+5wwxbg!Ii^{x{~qN2rLdIevw;VpYlZ<>;CLZKzN>_fhzF
zhV}SvZy;GNj6_OHJ&h(qz1fFBKuEm4Vms9!us0z9vYle~nhj^I=UXOaf6an%UF{&`
zYQOjSq}QWNDW{IBR&<!Zr~!oJ$xEQPs=OflXs%y8xb4$F8zV>#$ru2eT2pf?;SKuG
z^OBTnOA-=+$^4i#6dt5Nopd^P<Rze_LGAAjWbBcer!>w!T^XPdDe(3P4^P|svfB^1
zdE>=R7R)M<;v>2R(Gp$bi#4mEID;V>mnZhz@izC6{-EI4Sq!m0W}q6kcB)K`A$6^9
zz<@i){x7IhBlI&lZznzzrvA+wS5Tl}@Jgs&o-<_04VxI-3JrTK|0*WGF<0K?n*X4l
zV)eC*lI72dI|(h)O1Dc0dLc1O@X|KDmvpCX6G@t%a*bb1UIFSiH6nZ?=*P4Y=+u*f
zh3!wPw8t?A`zp?A-1@;+|4H_a5%JQxq=F)2K=13VR(0woB&ilVav=4x&4tV{^7g%p
zI6ojm$?jp>%f<!-IDHwHfrZC4IuUpw=_@!CqH5NDhS<m<i>Tqq))FmK<PX@rPj<|c
zqXv56i7amAYF7*GpzUk!94G9DtrdRYxJRLB_sN+xYS{(RuBD`STXf}lIXQ*+g_jtp
zLJxI;+id@<*f%w+?<N9y$%m`Ui-I~o(uU3=l^I0iDO&xLy-yroHlk}QC=x1!g-A?{
z7y0)&mV+rR`K;?6=9SVqKyJfUQwYA?)Klwq^y}Hh{RDFlzeYuM5~7vj+Z*|A)c?i$
zYjZVP>&A;Qw?~!4rB^re`mT#zfzTrC-Uz*#sFg3VC28GjWh`p(YA(o^HE23c<fAxi
zB}XWKd=RMrZ}E_{N%`O60cj?8mGG6-rSx%4mU<msu#?zQjtK#k;B5vF|2sbY-@vg*
z_D9Y$=W2m4z@Hzb6J2+=OJCQ-^d1udYt$(f&JL-tra0{pHQk>H8M%WLIAsVL%_Mla
zR{q%adyc9{tI9~j0X5S?vr_Lq!N^ZiCwvOi!p~J#=@j)c`{Zh>dOXBOe<#%aTC!?~
zj?$eZOJC1ut1LD;^fxQEmj7r#N_z11Up(=dCxi%E+U(55>DQ9K_D!6u!M&<AFa;+I
z**n*Ta9OBhWNjyvc(xoi_G8`oW<70UMJlOth7fGh%eUa#K+^%7f8%Zl_Rn0GRTF~Q
zIPKf`JXj(PzG+U98aO_mZ2BMnre75NG%<3cG3-9gE_EmXY0{Sj&HcFhAoe{^=%4{a
ztcxxbeunH4G#=NEb^-{1S-wvr29eFdf+H&ltF+G<E)Nz8&BsB*Q>xiA)B>jeIm=Y%
z7}GcP;ne6ZbDS3{IR{3*HM)ZwqezS1n>ISqp40Y_rXZy6_Y=UJTe))oWndz+!;Hk%
zeCs{=%~fMB-}E=v^laGrK5)i~d|P+X*rWe{>j`I|5mCK?)!Cki@nFed`UPEe0zxGc
zZg#HIlA<d5=BvIFpM`mbh<8)>q&S1w?L0%682>?9+)fJ+H`88lQF3!wvU0J1_xH@x
zSZGi%I+rZovnc0IsQa*0@n<g%h1Br@Q@>^<P7xF0|ErBNkD9kPfms;#%#)RaD-5=-
zFJFhQOnm(G13s&;?c`rnS=*@GSM8PFYIya*Vtg!z@jPCHcLq<7@6i@ox-sj@%{qqo
zhfx{Q>D=r14=OvmFsXz&U&eXKF#?q3qIjH>u<SkhxU1%U;M?^dPJROIv8UCG_CR>@
z$`D!ptEy>QN9bfyfECTw%h`A|7x9b8X1kfnDhZv#oIR_UTpx>AJjGhj_S}E;M6TvI
zx9;P%$GQ?%cr;zj5=Gg;_Yk7Ybly^G*C)q%+c&KD1-U-WyZvLaEHB!3+e<`OYg*%!
zqM4o5s<dx6(ivgjv?89CD7M=2!7n>qvSVC{Jh`LoKX__2fSqJVmtptW!D?xGnLXO|
zTcp7W6!Y|74%lie(&Qq-9qSURglr^buLD*HcANF!nZ}#+=i%+_Pg2&ku<o3KapGbI
zx7)Q~P6MDgeHq}wa@43(zJCbq)`LtMQknTMoF*wnYrMJYde<)6vtrsV&5~8iI)68h
zywtSxBZ-&Fm^)!KbvNfEHrUGl{30n0^Qm5S#r)~O!16~|?T5f<_b6B0=A(`|)gId}
z#HfDSzxdpdm<b_w%Hkd^Cd<l-(D1X3uCu4`yc}mtaC?+^U^EWsTT<pH-bXQDr6o~7
z>mWZ;J94oWK3ubr{-~L+oul^{K85&3AlzN{_=3mQ{es#s-X6ntbW>SfDQD3KQpO)y
z(u@Ei6u(|S%Dn#frp8W6`bY_A&lg*0p<LtCv%HeY|9yMfNC`3Pw|czb?~LZ+g;%*H
zxdM3@J+jUpS%#v~t7lGH>C4H(a2Gd@GlUMiE24jraYMg$t3-U7Cpw*tL=r`|#3a7Q
z8oL|jFA#p6s@d_m;9C>T5_*s@o?V!5wV0Fe-1qcpk#k07?lQ)V1+#Gj=;EhBiZ#d!
zQL|)tpS{e(9F?4P!p>eOM_rGYLfFT9sh^AW%tK6k$niTjhCHoG_$S(~C?1W6<ULtE
z?lqB>F)ZQNEIu$@=)?bl!S~S2gFNx)6Vh)MRQE<Fb@P@P)lv?*BjuO%aP@n}d)<SO
zC!LKt&8_%hxRYqAmV=fmUGY$m@Pq*ym^x|f&;z|vDD@5D{dwm@$qJ)JvJL&PyVruw
zcRr8smeDy~MVv|5`lpHF-hj)^A67+%jECND{Nv;hEh4HtW;ME-w+k~Zkj1}rK%6P1
zedR0gz{4&m45%b|0E!QCsZN~Ku?HTO>bA$PSE|LUm@U99_^Q<Ad;=vuI(KdM*SNaz
zJrgT2Son<Q{bGYN&?vAgIB8JYo+nXMEEj3Us>LgW=BZKJYEN!EKB@U;c~Bo4p2B>t
zsmgmV%V_%c@$cS5pNl;)jslYF(@0v=ddsB0O^gIj35S?&2IYGNXz83!P<JjgabCwo
z_@Y{)yl0WD&<lc$Fo{B^V;0-_M*S~3T))RCNC8>`qY9$c`_>CXtxjXHmUy5`E4ArV
zPH;Pe99uG}kRxajynMq1TDIvWBB?uc;attiH*x89h2AyHv+|`bJ;d+JB%}Qiw+T6F
z5+42$9CM6X>EPS*1Ku!;OY!~{6did-qEG(r4W+jWy8cY)dZ(r7e?3`%WEcmXc_QMM
zTU~Z{ZadI>GRP?swRtFlxX|zjv~0gVUI%??@<@XN9Rkb~2FE<d<1pjMg^v-kw2&QB
z_M*p~9d2BKfxnsqRh@v0c(OWrhoZS)dGAI^qjw@qd_`!*|Ik%2+JhgHVe47BYM75E
zzwZ6wW8+88%sLGi?$UKKsCsc3F6Z@m-F}`lA_V{DKU2Zz4qb|(9{M8M)^17v7yz9#
zt4D$7hA9Ms_%qpP439^Dk7qYAwFxjm!DUmV-z{x$C%@mh4;+Ij+Jr00brNu;2BC%6
zJrpZtOCtW!=hSJVric@sBl5z#$pPCpttY=2HfHTx^nNihd5qz~O!RWQ=_<Ii?npcH
zmW9dYw<3htznw%48DM{%F`JbCsGXXeTDJG_l8LxoK3QTq5Uxu8eL_RaOz)9kUG-5Y
z<h_H;{DzMUESH!#mp_qqSj9tvkg$hvfEO(m=Hf=y*sXsAv5E}*v;2SxH^noj;OJ{}
zbfKNVsQsTVW2L-*#HO4qISmw-hg#{4!b-B_TZm~PExzZPE-Z5PBD`cLNVpKw0_w?*
zjc@XII5ZE91Ri`;cZp~k=s)h1v(`QJrl<BnBSEO7R%qnEJVoa|!|LHm124d$ZkvBR
z5$iplJpSzhOs4#%1-;Q}^fvcjnpT2kM_HhC7ixd}n&84Ktn&HejVwQ-+=n#3GnwCa
zuchdM`r4-hXH8#sq-<s1Li~s_B)taWWm;KiEw}JpeblnHFS{PBG~I&&x=(w3WBt}S
zQ4cjNtAvMGS6x%;RN$?-&}{;?n5E+Z9*t(EF@9Se!mb=z)C5LW;){-|`J$hXZ2K9Q
zuzQ17m!G+xI-+_kYPx!x1^Sq`G7@A_T1AB!sq$<0R1{)I%}+n^)iJi*U6Kxb|It*3
zve-alE9oQ<q$NbAfca(vYeq1ZaH6-kxvIehJd^Xj!OpnWsl3xIvLL$zJr0`wQ2HnQ
z_#1BlN7;OZjf4@l{<{Xrm1_>aw=mA=<{XyNR;F3%t3tjd={uLfYiRn<KR%5y=Cd)m
zADF#*sMS)wGP#=kiaM*g8$DR-ZRP0@5Y0wHphTL@w%<anIu(;vf8<Ehe{|3|rZ9jY
zGhc0TYSZI~S7g(vvYS(HA>66%*ZA9L);GlqkIML;H@>PMhBz{$V$@PVoxk^`<)%^H
zi++{7>t2h|VM-%j1*^HT)fQT}^nBQ#hyXB&3!o{~F^9Ixz(AFiY)|@KkzM1uk6A$W
z!U}y?7o1IAl(g5kiC;zgTUZ`!s$46)zH6S}jblb23yg9@3?MvHGhgM`f=7cuO(UD(
zX6XIWZHtW^kG;!JAJ0)nwPL1?i<D(|mxDxq=GLa}#5}Mfz<v&Gddvjc%%*IW0Rz)h
zTUt-*^RDHP77?D^n!B)GgRd%5Z>l|5XTVP{a*q64$XAtrfw1)>$~Dr$FvJu*)?XL;
zGdHm0mi;#~Lc7IIU?Z;I?CiE}kQQCaAg-`uo{nNMGu*8m_WFgY3^c7>o=6G@?!VxH
z(|3vetCq+r%ReyO=WUBsW*)Y|dP^PmxzwO~0$s{(&)g~PVXzAUsgt0S-}_nGqdF>8
zH3Kwz4Xrrixww8w^cy^0Vdy;&OX3sj+F!yDmX+`ciV@ddXXNBRs0`jL2%OSE&t3MF
zkcUawe56=apDo<hGhnO`awkbr7e)iM*4q1*cGu;X?X)yEmiAp4f%#jL1juha&wbb^
zC2FqBSE6gedp(!^)~cAgZ^*zmq2mU9cG_qzQC;%~%R)|icBn#vZ>rskl|C4mw+U!}
z5E_1yN}KQt3WL+NJZ_Il4#klt*u`#%<wx|c-P(HwoKDpa52sJgSRr##@-TTE&wJp%
zZdLNGJ}=0EDJjbmjIi|RV)zT9joF?7>%d-UC$HKC2HMx`Z&}vY=HsBy=reaD&h)X%
zGnPRhYJ*eXj&}Rrxg+A&-^6`5$t?|0V)_?e!~G`)?J4lW{gu0H!bS#loCfP;#leGe
z@g;wS1_G<Szqx_eng#WPzokCpO=tULq-Gm%@AsERaXSj4zHbqB3VhQbE1|l-T^oPz
zd9zO~Oq@+HS8G3^<1x-_MaU+xKRRLTYD5n5szuPQ;rreiLqVFVnJUoBfOlMW)|n8M
zJc-65Ln=J&a2nt~=)=SmcUKHfadUV+3>VERnetPZKet9~L|`q+9s9f2Z#yFZNu*}s
zDAeVUH1SKv$uB<yJ1fcwA!6ter-{PJRJK7+Df1q1zw6ZED-&dG=JM8iV%K0Z(^fyl
zf6vExOAB6acu*%YVkKf+@2`F0!~>kaUi8bgs_WH#j<8(3t4c>fimmKSAR<ZVuiYku
zh;2S}+wf+!F_jn4%-sv<f0oQqq%nJxb7mQ9g^XRQb{{4DFpM9yC7k74H}E_Bdd9h7
zYUwQs%{k7BqI^Nq`=`v4+nSuTuj05`zEf4j1KMh^&%d|U?gH&((sH;&O2eN0R*5+j
z&a-a^>iwpU=@ykv=A-yxxRV;;r^rjootY&eUvJK_uX!_*d((7~fBMrpI*)UGD~rj=
zqt10_QXgqV(fJc+?r>H^Aq24Q4nwW)3wLAQF!nQ)*Rez7z&Bkge@kV<;uONJlmGU@
z7-C{&lKt6&-IJDY2I?^Hc7Bc&<4)ln%g_Dln(3FtxWB^<34oXg{~m*^8nqj}!L9@i
z9q1OfSrtIBJ?TuQL8b_h5g@~6_h$kVBd2)bMBYVyoU<#|g&g#wIdM|G)=4`*x-HqU
z-s+cpcd_hJqkq*(H^sLvvL=QEhD@{MGDhvV2cHv(!C)~P_O(E#ZCOdu5)511@b{cU
zYVDo(ceG%)r=l6gJu-;#RxZ00E8IZbO*e&9;zrLs1ob;;tR=g!&8UrTYW?z=sI?Lc
zWLWrHQ72=#99)}!5$Y4|lbm{dDf~wTHs^2<H;_1b6?mmr41oSpqh0bD`rd1kbpdZs
zi#FiuH^TZpTJ3bmgL6V#XT>YXwLuI;GjeH!XeQ-Lc!Efx&Ud!t;`jyCTyYBhJ&g}b
zzHxHdB?tij8WcH!DB4__zd*Zx3d{h;zz3Y*)$KQEIVSwwKkAm!W&2TK#|ZluH>!?$
zIhh(vEx}9y-SK5W>g_o;ctm>>;fKpaE%JAL4!J=kW!=?jWJm3eJ2kOO5)6xp<zEf%
z&7rjp14gj*AaGp08jJo4cid3&vwQchHkPK;33K`j(V4>LZ9<)iNb%U2k>gE?%cf%V
zhB~cI_}+vlI;pg~rO|B6gX_vvT!|HP=s2Ah>84zenR_IY%moGn8uZB}gvycBa6YdC
zQ+Fx_+lS0umP&hXB}Rw5+((tX@!dmHbYpFUov-_jYqeyb`+$9-3tO8)mUVC35F9*m
zM2`LNbDldlc$w%R6l}9?<WtBQ8ZkV#%PL03xHO?&>kOO1Gn%sv;`CVS1_6ds7IT-{
zbEUEBpZ@Ru9F|<4V*cr_x@3k-Pw^$^B#bN$=ae6<w4H9_;0;-t+y?VLASOES^$uy?
zJnu3+Fb1U9+M_O(kh{Il@0?QAx_u(*-;<=7V-R+$%Fl6&jX&rJV&EyR2a?rICe7To
zSI)?NKKDWM^JX}as-4yku+*Ov0I1zgnMJCdHm1(s5H^(QN5&_le=l2(ELPY8SMBMZ
zckJ3<Wx;<yjO}K@NnX1^d(fj7YzwG}%86}iJiXsm=&KbV%ky2UjyJ!tjh?dvJS;mU
zoNqaQECY|A6w~I@;e*n`Q?N@XJe$xDKlexOH$1#Rn>MwOv#!X!@TG3D_uG}S?iL?X
zYr!QW+3jO<%&k>7jRK_&79t$pO0Ur<t3_wF@VY&EaY6{tVl=1vQ!F(B;QA84USr>%
zaEu^VB{q58#5PZ+yvF$o69aEyLvgS2M-GEgsX+eH=Vo=8uTxsgZwwD9l?(*-8Oztd
zDPNN#AAb7QJCw}kyWKNOm|-Onnj@24@ldJ&s1EN(r@I#?3Ng7Wz+M)mdxJr!8UHr+
z**fiSSG1gVAKA<3=9Tsw%ps09<F_zpO=x*p(iL~FzJ>#qs*jJq8{Ioivr;vLtVU`C
zTFq~?RT%<CMLRC5ZjEN&v5Wu#S)<_f#a}jP_fQ}s(ar@F(HRfjLYgX&KT)PCq2UJ4
zW98ife7}_Ja!OwAh5p(Q0Q60A07Crt7oP?JHjzW=i>mVqPC|K(dyE&?Qm?k_$2TJ5
zU<bVy@N4E!uH7{mE{Kxs)T&<7Wi=<7#jTLJ6x^$gxWoW|Q73u1JOXfLO&%K4yjRVQ
z34g=4^~?UoDeg&jP-}HRUXn`d$zfdC#59XYc(XQkmTl;n+@A)q5usYU7VI{`_T|?b
zHY4ZcZ|`Ok99d?+pCMJCHJ8zMw+6+I90APd&wf?p9L;#6L)Xmr0$jjnw&<x5XVV4#
zwFH=BcVI`E1n0fpj@?>gTtu)L#sX%6a^eTu9m{(OxX{!JS<IRY8F9?BY3Rfce?Va=
zh8+P^Lba}ZH*t|6X^w|M8fZ}58K8c}{_3n+-T^*pWLy9ITzjYFhGTc|O#$zd1+Bm%
z+jxv@qnLr;-(ceTL#00O;Bl)|Op2}R2Ig#%yXL||0#Ar8IpE>aN@k*deNI0!dSc7f
oK2z^MVb;z6XJZ`4FYX>3O)<w~Ic+o0hY<{Q6`iN0PhP(HKME!Ql>h($

literal 0
HcmV?d00001

diff --git a/_static/raft_reward.PNG b/_static/raft_reward.PNG
new file mode 100644
index 0000000000000000000000000000000000000000..f69f56e81df52f35a72644eb8f2ed194297747b8
GIT binary patch
literal 221269
zcmeEtXH-*Nw=M$G6af_w5F%Ycq=O2CE~r##(naYV=_N!23@AmUCP;_Si_%-Dg7gm3
zdmyv`0YV6Y8^8CQGw%I)&v*Xa`y*qo?469U*4%SG&#aL;T57a3tTZGfB(%@fRbG*h
zkikeuNGqr)h<9++ByPmZ70*{{N+e~2cTvPI<aUahiX<dIVy~TBT_t{}c2_s{Bq3qq
z{rkD{{1wkW3CU&Na}`B>KZ~7a$`9;ni39?H`R;A@wd+!ETFK*?k}4`HT1j7ix*G3k
z_WJe8Tj!TKanE|*e~tg{IruH7<nH&J_f<TmH9Ug()ad4z4VnCa7mn2#&*H6R#j(Gq
z=FR8L%|VS@-k0#vOUn2jYcZ$>-@h(Tbtz!~tp0UlOP|Fa_8*UkmqI<8Oo#vatp;2x
zF$Dg<|JmUG|M~w5?Ee`h@!`yKXh>!3I4oka*sY6salrfhr}$nI)Px^bw5i<U;+WPC
z!CUw*A_var0@euz5_;P_md3{Dlp>sn8*->w5S{O63!H0Q%Nabe5{;3A8i{K3t$1Vb
znl8y9w#?tez<0X`s~+K$mTQ}X+I};~ICgvMf0qBumCOg0%Xp@@oomMha72MNL5_}T
zCu<#pgJ?Np+|ON5i8s!88{RCKA&dc5{agrO>}yz*0@PR>8qkwB;)cunc4|G*hrlCX
zA71ArW_f`Z5N`D+o346+3Fis=u8Nu3A8km_JHff+J-7_aK0dE0bP*fgQR-#BY3yMN
z2O!SQ0-P6`SZHZ!?<c;vpWOGTq_D8ZrLU^|$B&U+j8&=VuPlr;P{ZTY?{=BfgJAa5
z7PztP0(b%P_j8zrtFRFVAb#>)_DmfX_T!sh7496>%RjTd2(DiTZDR6*?10YAyFHKv
zuPAH=uA*Tbl#_S<`Y0LaGK3@~VXYSWA%W6pX(1b&>>phKa&^)NsMVs4jGae-WTaeX
z)|N&?%QIgsj#-!(r4SA>LBze29)cqaZ8=0hiaJ*UHnSk^BdW~HMa5<9)l(MRJD&y1
zS|z8<Dn0SNg5`7dmdmBp#pQ?@=-@=nV9UkXw#U5pRK0gN{#_H}>Lv<FSOMK%^anW3
z2G#juf<*xcF(hagROoGaVq+rAp(m7{?z|xEFNib^htgr4{aa3Ls(-ox=PK6;NDwau
zZ;;fC6R9*$8ZtW61-9Vc^+SYGu~}GMyE7e+({nQ+SFlS<@U%^G{V}s;&`6sd;?qQG
zuA4L2z)6on<+Xy8<>snV4K%ip-g&l$mOmp%>|*f}5B}rULa4aBz@u#`@4Ax&9B-{B
z{o1d2eKD7Pj2feC4#ajjS>k7scZ->fw5zk-4kv5fEV18Dj;;@g37&POo*Zva(NS{v
zE%gPxj_N2O%A5~SBwOs<cZmONC{&j2=-=`iB9`bv03xxBr6y)_6NkZZm$xTdz+fys
zWzneWt*k8+C9T`Hf;aYmg+H`f`0R1yg#q@rm{yf8K0_<Jq+XslFKGJ%1T=u-!2o2b
z-$i<f-+B5VcDdYtz{!2*NB7&&+~<FOyz3rxa};XExjDPRYyO_tUYo^iHjs2+HAd#f
z#&^#fuFc1WQ2_ac={k>9_l-2)JqZ`V^B9@EKSE8sV?V^wws*@K_hwjzTx#b(rMe+l
zm@WJ`DQ?`|bkt%*4iz)mX;)^`tv5OXWzkI-5wA@hZ1TpeTg!Y<iZ~iVG=efvFZOWR
z*e)$Rj$5t48{P5N%ywd6LwddY<)iw}#Em}TmO0!jtEYTuU*n6WAfhDJDA#3h+aIqz
zYW>QoJZCxLR=>`~T=s5F8Tv2feQeu*C+dzUsq~zChZdmM4%so8dJbpO4yB-dFO;@s
zgcd}X2kzB(<Hmwc-^a_Zbzp|%7n?H-*bI-d4wss-3youti4{~2{CzR@_q1Kc?!u7t
zg5RnuQH(e(oDRbI3mxcgu&Kv-E$F-<eRMwuK=>Vk45b5Z!bXd1GB4pLnS~;Y-wl`l
z5$A8av$qsOkUnylquZn5psB`%@cfM6nGid~WvsMIprC85^+gaMzaK05st&d%K>LMB
z=)F9Bv0>@Y;P+Q&Q|Z=)<!YaJBA1A&*4^N}?X&&vYix18iKP5*NN~oPyUTn7W31fC
z9k^yH@*D+vEjv`wfIEku*Lk5{8KoPQnW1`X(W_h;fi28KeUM}Qc)*$<&V8ZDzf88=
zxU!3|E{pCscMFjp3Q93vfM^48JaDbF&(UnsFuGZX!`YQzU(Ef3FK`>af%1JBCXOYP
zmq0PVp6ekbr~r!zqrSC2`ck;LAM#X7Q>2;h3sDE-V~;re&kwri@&0{DLNb{2qy<PX
z#Kfl)a#6ol%h-|*Ei0Bcqm}>!pITv?WbFffOAPPSU6*J~IecR*!AQ=SuBY|nD%yW2
z@a_)oBR5qdPKgI|^*3|+By1{uaH@s`v)73B`I_(-<1B=5IsmNWAg8(*05QiSw4UmC
z5l43{D&jO*GH-Fu)<vA)x*4huHn(UA22C_=VJeJ*!P?R|RiN2VYid62^c~=<V)j)`
zJ`TN(1Qi^Xz#!-L8Gj)nP#X7q-570`xN(lo#%2i*EgWuOiT0)IKZ!SPO~A%YegXDp
zxYhJQE(j&^xBcaDAp8vEvtH|yfFpZK80)OcB6YCk#ToHOPL8R6PZ#v8+*aVWz+(=H
zs{+4b2A)2%5|BWMtY6b{>hJ5;m9?BD)s!qWZS>h4++N2r4+*;Z`eKTZ*Pu1W=X_e<
zH!rr}u_foC6N7&r#8=IYfdn;#e?rCW8d~1M=zNV|VlA%g;S4n4mm&jb;q;Bf#YOXh
zMa)9`5u2dMDBvPUKJz@7Xh7*J^2Q^U!7>0o3@``@wVQ*W(iO%srs9MC7{<rS+80S_
zwS1v*8$jnwKWgzC9>*Kw^zcOP>c+WLniHW|Q4ei*DKrBGI-|_OHxw~(Y8)P|#h@3V
z+OZn5V&JlWQP<l;mS$Mtn}Vh90b7MPB$ywW)m1L7#g0JUA?SM!yQ?OP((*-q?K(-&
zN|@F-Vjo`i1VzI`km)Yk*d0&`I1c(B2f}2coFvw#&P8$WzF5sqe@4vcrJVDMyz*4F
zBzrXgnHb#NGi<VJF+U2$F7HfN`c8gGp_{c67#fnpes}5;8a_gAWXm@T4`ei^o^xhU
zOH4$WqBn?a07gtd0&EdV_=?Y5kI@^0M^-O41Z3%8FVr|B2>vDlnqqo*>&)q23BvbP
z^%VCt?G_1pzKuXd$M`Z3qSarJ30ZR=kRiVad0@f#Dg^$tV_P*zIP;fhbI@HmRaJSv
zv6^%KTlQK|v0ulMbE$L~kE7d+ge4(#N4@7p1D~h>B@K3iacBU7-`y*5#BLGdu#r6S
zMHFiY{GMlK;qD3=QZiFNZ;T7X_TpP1P7g9r9s29Wn#;ww;S9m;2Zz{4nAru&U+$My
zTd8&afU{L=xe<s<A`pHR<|vd?CPaGL1i&D$wllmqWuU%)y6C%4WCP8(;i|?pK%!N1
zhGfjg4?`URcDM>Z*Wg)G4hf8=U2*fGHT*B-eJM;iz`)zH6m^2ej0yB-EyD%F_TYo=
zVYgb<i#f-uy70%u*Je0E=dYIn%)b855(AVNE-Cx3vmh14FWJ7;I&S#bt)u4!(PV(K
zyr2}y0Y~7`MsL99LoB)$yVk*FPXrnMb=44CMEFOn%?Nfuqmc)EU$`jz<Cz(riALz3
z9piL_ZtJIi5xV`f1M<tW`e9e`XXu!=Ji#Wjo?pCLn*wJ3`WI;?X#%Atfj(e30jGpq
z?>NVmCdRsHZ=fhU=s4bH8k=7P7tdF=pF;{Qtcl!qjtic5!;)8f4bfmS3)_@W$YD4P
zp<!`vu*j#}q}))704cn1U!%kp!5uw(><k($OO5)BRELlJMTNmJ^5y&menY{{FgiK=
zBH8WG;h;Z21HmExmUR$%pz7scY(y8!5ta)!J3xG@rEl-O^w!Mde(T<O@1rT^l4a}R
z^Arw_zHo8Ew{8=Ge&}fNaKW@mhL11Ps(U`atTc6<VjtIFdtK?c%F%NuFesqrg8rPU
zSFzO-`YO+y>E@{tM_<K<?yb|BqD`>SvDx8JmjWg+j_AS~{vatJVN^>S2(`WV1A0K%
ztRaV$>pyKMT4TGkYTzH1xU7DN?nqMj%X|hH&1yQmv2=r@tR#eNEjbKsUPPqI4{!(c
z;kvQV?V`kuY_w+M%ix1`V~~WLJ@WVkPG{uWjkLIXR~@OXV*H+ch`QUP`jFqFl9#79
z!MgD0t!-dfZj|zels3Mxe$Q83amL$Y0L)p|ik<cI659&a=Z0`tZ_0uA!_xh_uTx)i
zXt-s??H+Lir+hGSQ2JdFkpOcC-`oeW@4x<(2!&VNI)5F+XTpTH-=mh$WRL&X@$TKR
z_o3^NzEHKzoCf*>SnX>xCxTYxHHZqCPxK{@)~6LJy^D~_WCMe7!b23W-T3pBK1}fy
z34_^W<DX+nq`6BEc~qZCv|08_<ZI+V<ZS)*@hRcmU4a*}(OvJZmWH)Ic3`uIhxJJy
zv3x^}B8GRg!<URXt$SrfXe8ou!$cxPz=Ps`JIfCW5~M{-uaKtYcs1g)M6Ed69^3k<
z84x$jYL5^jdcf@hOypz(dz`qgmfzPUX8F;?J=GM#+=v<&V!N&XFQK#9*Jw=y*=J5a
zIsChE&^OB<{|iVcpO6z%F<)U&cb))q(XZPjAz$IJ7fH7aj3{RAIo}jB$LPD!<kY{@
z&#hW|BD(6Jcjh0{C>3d-V~ku+i1|J7HtCL~lo6cF3eI``W`7#tHb=T-Bzj|4!{WC)
zuW}|E=V!7c|8zELi;1Y(cLKm$LbMW=EG+guR*cs^()<Xqok|W`wY2^#J8V{n?X3QQ
z^hL)wQLF-IIxj8L|4Q1CXzV+8afi;tm(U}OkiA@WlJKWZBAWnqJ%=g~W2s4fgB-Eq
z6P@aK)}VYAj?|HHDA_>k#4{Qte-X{h<kA{mToD6fF%6SJJj*b@gXgZzU8ew1rSGGi
z5@B=sGlsMjM@uEBheOd}uRhwoutpwNRvFQU+mPn(5%Ec|kLaWpzosXWDuk%lT;jvd
z8BKJw60>ke!~p<K0V~xTftZ~A-Q80I4DkpXzmx5V-u!c-CgT9KuxuI0NJbUI6mm%!
zgT<FuIKJ_{TD-wSUj2c@Xy~0}#f>MA>|`_O^O0qlHrtL`eXF-YK8c3U2{P}FFTFqt
zjIoX)Y@5Y2A4Rs~3omjUEmG<4DU)u9doo*1hQ@N2P$~6kfIsXu)xhRy)ETv*4n}%Q
zF-4~uRT9HNHbhPdH+V=z0h?Doj0VTC*w<(;Kqz1?dXkH$bb4e1lZBhgLLh&sVdu0W
z+^|BW=UQK7h%KI{<@E5ph(we0?Fz-xvh<t&UH!$Nu>RaQz&o>8GExf`-3CV8>=AP)
zWy(WiioB-~{$ii_H~BLvl5P|1)x5_7v7+T@1#ud)_^A&bpRUKQlzb9%aaA&nR8A1k
zOu_f-`R%kBQ@se&u&C|aZV3Lqg=Q&~O4tx}DFW{nd>&GHNj6~g>|ZeZ@lEEm9M9q@
zGnBZkH4ssCr3^69mm-qH9e)OnkNy?(WFd_BHe}QgvPAvXKKvp)ba@J|bi7fl=tmye
zDLO&gplc@>rpj7tAbzKj!Ft=+c04L+(f;Fr?Jeg9nC4W1A$PlNC~h-m=378wx3m82
zeKO7Cdh59>wcJX4-)@%nI~AjCsM{tDUZ3onr^w(a^rarE)mNpF6iE~q>c4@pN!I=Y
z9sjN+s)fQyM0tiCc#I@Py!Bx8H?R7x7kqsGC+vkn0r#ncgfSDVory0Qg~wPlL#KJC
z47`?*Sr|BQ2C>zHS<g}nSg~9gi|gdgRpcJ+(o~lgQfxQ{mNfg`F<dy4PS`Mm?Em7A
za|TCtZ~lq+TBI0mukn3!Mwc-oGKm6K9!1pdiqBWu6e<JN5eLFzf1^=6Q6pT6kVGhV
zB&s$es37kz#Fgqv5^>NcCg`u)WUMh;re7Xp79ygi6Yh12OHfM)YDUzSsy6vZWv}02
zkd?mH_9>6NE<%4=#mQaMdV*O`5~e!bk8JE`Ew14`b__1Itz|TF>cybqd9L3*N~$YY
zSt>0FU|l{MQ7^l=x#D?ol{RNFl09ef^J&wquND^%a<K$jR4-+m6vHz~D*CaaCr{XU
ze06@=W#@$;1-Z!v9+kXR6dB&(_~v&0rJo3LheYsjPA$T9fm4X^t))uLsQgv(ZT$hF
zC8TlxMVMKB+^^&fu&n)CJIr{dm7IVuFl4FFiC(izRcvoc-4%CbWfmN{X+b(yQb}Ku
z67y#Ei)>*Wf3Z9thRlW8{#lunF8z$*sD6vqKyqQ>%ejKuA_Lk=Ak1o&PYJ1fe;&;G
z@r!(n^`9(tien`mPx@Zv8>)|U<vVUDzhA&Z7Z$n9Wv`%WCN#L(v~}jFgCG8%7#U)V
z{rZ24kiJjtw@z(mSs=;^!Hc_yS<%e1M>KmY@Y2>^7Il7F72K$Hj@Z^~mC!)WLR4&(
z56Nf)cU;U<3Qa8GYV=EfuIY=nlTnE_>eb;gcn6v-{m=qL(sf_S$VT(=)SH!!^(F>g
z%=MnsDdYnbH=PUoCS^mddleD}gP8@YL2;Akx4yOEIg557nI|F%PO&RzZB-Wq*X39b
z%)yF3&)bk*g+A4*HqHM&NHu9IjL}jT17pa6$55514~^DadOhZQBIcq#?&^<C2;?Ft
zH+3~Y_+f!`cHy4AJ_SH!c;}1X?~sVC08Vx<s`ioY%%DDL9SbQ&4}wW9XA7NRQKU8Q
zRe-ssb<gpbe!Q7W2U~e{U&7?PrlyTDUzzBOyJohY{-3B`Yjxmix$2kXBDP}`G=?=O
z01<E8<v;R{ZPd{5(F3}6m4_eaO0Re>e6dE};mJ~n-*?KK&a?UN<MG^dX5s2pKT%8#
z{8+~&{h3>I#|NwFDsI+0{Hf1!>!HPGia;-pk1yq8{Dz&(#K$XY<g*pYYeES0Q+L%<
zWJnw|<w?ygRqR^WN7yf;t&hz*$vZy0G0irZ%an;S8vA_Tb;6bJN-b{hI@?$Mt+<35
zE(fz%ziyMxSqp`q>rXmUleib^uJ?RWjj(`yflp|2nH=94HJ}@_<$Nu2W4UgTtCX#^
ziq>7E5H?wWe3Q(P3=e^DvZ#pg1yK}Dm&n{cIJM-SZn}G=ycn3lZg2p#cgV!`{{HVX
zYyFsF{AGlnYv)9S2VaqaH9}jEJhN`ak!n7Hx9EU}9?poEs&P|7)nw#IoO?%T%Y3Nd
z<2Jjo6YjpG7&ZhXmr>eSx=^QNkMi1IR@VYw%T{}&eA<4ipy~9!fFR&m#d{_$$rf|b
zYpBCQ_4Ks0hZa&-4!#A_3$w=hUARPSn8j`u3Npxfh_0x0a|Lgj%}htwML(0xxzU!u
z+IdU+3g0;+su`@}w~A#*!(aJm?M5C-rzml%qa@7fc`wA%9pH*p6mcgRo0caVXd|N^
zKT-->or)`g#k?A03L{dV$-hB=cnS$ztPTsqqxk!rI${#zRAsytwLQCf(fQlT#x2x6
zhsIRpXAiR*I;h;$xJq}?nLm{ZO0LUMjeHHQmm^uu<fi(r_GDOJ(+<@{m;D~cPek3|
z+>K^7l(C!Y)3a#xp|$sGy6T?}n;N(o=wt-MOQ#OZEcnaVKIF!<rqptR{C9!skV97P
z7?LfjYVcylGWjg0Y<ed2d+$3cgahprOqDWKVp#tSf#Jpiz+;!eOrFLRdh&jkGs5kZ
zLJ@0`@REE$>-U%c0Z0d3fl(_VcUvb+p0R2`bRmwDtWF7kT6aD5B0H&92821Gfzm62
z7ttb9Y>%U*>X@_J2s(vN684|JzHMJrxaH7H6<9RCqrJ(qMk{}Pq@W{i^dp0;&*PA<
z3dVfu6w1^A)r7CtWrw406vMi5Pu{fihrIv8K;AQT&?HTg6Fby;v{TE;j(L>3fdKUU
ziRG-{T#U0gf1b+>y57c{q4XFj7d9U{B63QfK>Z!nN=WGn)p<MfLiy~#q-n5qA%e0H
zwL`+uw{4E~&K=THnaf@5v!(P__@9u=sG2yhjw37%iLB%d5wn&Mn}UdqC@4;EmW^~V
ze+7En`u62AoCqOV`wF>bP-j5cR{VqC`WeJ3sKjP<1jX)<x1^EW?Sm)$T=hq*>gDI8
zAGZ-vOK0))bNiqK*T>X}sfgH3B!IISCDS^vc-57_IXVmebQEKGkNuJ@*Z1-Y(Kr;J
zE?e`^egAAds=%YS^;JuT{h+_{YyVS9ROkjbXpNMdF73-w5`e)v_F1SxO^y6mj~BhQ
zpfXi@HAKXmWgH$hqmN>L@n`M$o403Nf+X6m8}s(x#}$9e54KPp#>Xk+65Ly<J$V!)
z+4V5ZoB43w$cZMOj#oT%79ZE<d>J_C6dv!;D<QMg=g0Xj!%{vN7s^qsXA$CVpu#f^
zH4X5j%4cN0X;DWg5A_TX73zy8eTi`dy!uO=vssjV-xn%hXVshZhwaaDbt9+NYIDV-
zKNTN2sJsqFdq6dnxgIAuG`Ao2yqh>m-ssl0jCAjp`I;UqJ8GrIrHiqU%q!*en!NjY
zX}lw4Pnik|qRzr+Nu0Ud*9QFZ6Ri4h;C-I9ngQu|WBd1l!v%gxnC=I4?g97!_Ylp_
zxLMMay2Olg_)_h<ahRu@5^hhPyJ>0h>vj3>nLZ>|mBP!fhT(<E7Tb=(FT)z(S(FSC
zlG--HD&B+-m9dn_xe$F#w_2BHCPRVtGcg4kb<Q&pxn{8`>q2bPNjlSJU$&ywU8e}5
z1GxjvEu(kqaT0<>9tI7PpcnmB<<KwIhw2}Z)asNRDk-w$S))%Y=xXs?R<BJw>6Mf`
zl;@WVqAs~tY{uqD)E)Zqyv{!QjiXK5(!c&ISMlEq<`E<VDqo*JzV)VvPnf6ast1dQ
zgUsmhwGh?S4~5kcWtV&(MbW%w9FeC{s|@KEPsz#q>X1727B2?03xTnWpAww{?Lx^N
zVtPfpxtlQ%^~rvZP+DHjkffK#SE9`pb(#e$O*;F<Dy+A}#sXa9DK}T+w`G|bFZ80F
zf1gsNgxOMN(Wb_g_>rSZTSx|R^xv$yIay)RpHdxZl0pdPmqVN1eX~$q08TS^;VXWn
z#RJjYiN{}=w^IHG;wb!Ek$Z2Usa-tLFf*4oy!J}(Is(X@(v{Bljn&HuTv<Enxym=L
zZpA1*JyhwO0+S^c%g2tQ$R9@oWvod?VI6Vl&zA1`k*u_ptBU0c-)y&kzdH5xVKc1<
zS<3)y?9Q+Nc{@$pBObFN+fm7R?@6=NNLwcgZ95#;@lk3LgQ#reOAVQ^A+e(OtPn<(
z=&*tHv9DSss_&=4_t-rjw(FKF`~P9nWjU}_dAA7hIv0+61Pl54rYO#Kf?txVK1wE?
z*%uB)*8XqlhX0N$jlMQ~+p+6;*oP@~0b2ERF!@N8t~%AXgC}mvk5}%G)A8n}VHuVN
ziB7z`^ymZ~xBm&SQ(Q~UYnq>d4lMVr1*;aGS*Yl{!T{mAgY!}GJ*VzD$A3+b5*^sO
zyq|fU3Kp4KEXHT=I8Ro92D6Ap6<YOD#h4A9S=kmIlN2d_*5{MN<|jQsD^^3VOkFLH
zJWEKbz4zx#wpSSVX$>m4-@Y4KaeCI0a}ilgFDHEsg<Mx>lWG8FU1_V1Yh`sn4A=c1
zuJ*@679l}}MzKb^K0th}<EuJ(JGb8M#M-9M%b)QRB(O6ECi|@pFSktNIVl_9h+5*M
z$UK|<*>2El?d0B%L4Qn!OPD5a={KQQYKRHYL_3#MpTf;b$JvXHDieRH-9u>d;)#_B
zQc-VxIw<9Ni(?s6B;a_dTW<7gmp7^6f?t^OM;gq^=Z&La1VETUV!~Q%w!@Q(6;YqF
z&HcO1S1wU2DSg)!HFw<at!ZvrDpo1N&VT<KimeED*LSE%oP!@wr)%tgJrf8j)qe8s
z{l+!|i&PFVy3_T#!CDlq4I-JXDe)fAoEI%)d2qS<s@ujvxAY!pYitm-<pw?W)(`Lo
z8}C|CxQ8|DI_VlqF^(v2aMrJGs0*K0q|S=568OlOL8R**V3nhNRbyH!0&DISt6C4T
z>|`WoFw%r4o`So6SDMQ8>`AuFCEEcRKR3)D_|1lnYrZB1J94g#vTFEQCEyaU2bHgP
zf4LCT$G&iKIyA55V_`cpc%Elsl2!_4Ah3SYjF2ktY^Oy3TD|3_+VN>Uf_ef<w|<$@
z9bQm}GOHS(k|{{@FMjgqKayoN_dqEDBI7SLk>cr-#|KLtnC`uvDtSQnLVq_jR=WKB
z$G4c`^72J$*fr)yB{VR9r(zRN1d!@80R9VeUjJ)1J@Z`7sL?I_s^W)VUzpJ+Rt@=>
zqHu5Db#(m<sQBT1^mfNl7KCF4_s@f4az?aF%dY17)aA1M?o4$c`M`?5oa(Q@LASmH
zW(@f8%ukz_1^}j^_~YA*XlXPV6q`k<3A$G#%WNR4JvpC8=Pwzk?~*{F{rp>rwy6|p
z!Py@Ns3tTk_FN3J@SYB~J;fEJ#LRKcODWi%((&XDSMJb6?t4p|Na#cO^h8k45ql!&
zWnrGRRGQk!PGy`H*{3nTlD)pPjw<oi=-*o~a6|B?E#BXO_r|LXi++r$6YYX=uMNWT
zmEoJ88NJF4iG0n}52dM=3b!X43Dma+mFHf0IF66=tT(E4c@;h{!CP2cXr12mYPqrV
z)<XDCOhK^R3hhh!fqKSeVtBN_bBkfZVeiGv6SJ0!St@u&<WV?3Rnj@Tho-YJ)8;kY
z%g}uu&Me>AFX~3}(VNq#i1dvgsfPRYMGv<7`ugRS0^ny|P`SVw!68jeO@qaug{R8D
zPX99{_c5VuF4hb)YRVyUe5IH<b1`}7yJ{aA+l?0y0)3x|!OZ&OL;z*0*Ci|oy^EgJ
zc*B$s+rEER#v;RWji0e;izmd4^v}=O?~6$fgo$~Lk`-d}!M(p#k)V1rO=GTz-;QvB
zvt1_&VM@z^K4)@0+LN2BSCclL-|Bj|dbU9Y<0Xa3Ye{PT91kIGbypBjbfPsCb=Ufk
zO#T7G9{~+*3Qf)@x+KtvhQd#RiqP(47<yng=Kx-}=6kcltOdC1R2Jw|Y8K61E=)g$
zn+PhXOQWOCR@8~by-Z%YK*hepoO~)VOu;80e)B$Gbm9DXUjaI7;i}aL3+;N4)%&Xh
ztBaB8iRNo`;2q@`Rt-1?bID<Z)V0B6qswYsa@*v&Mj*xm?&N#H%MLuzsb7<752ShA
zfd|!}2JOs@o@PFH4wK??@r&%!oR;8isx^A){}p^)A~1$n9?}8ktX~nqGxTT(%UdcO
zF5^6Ho_3LKrJvwDF|#unu|t5>+UDmp%=Q-qG+25#h*1zcG_hxDMFA7~XJyl-;E5Jk
z@C}|#?JXx(2@RSy&TBOMM-(b;{!`aOY#FJgs3aYwOyDnJDb?f-Wl68zpdJ=K(X*`o
zs`Aru{e?!|hGB^*R@N$FX)T8N59<5Z3bhEUx@QXWMnHW~&I?6D48Km_GgI$p`?qd4
zI>_le<%_^IMe?c~3r)0*rLIB{qF%8BU+3<Ei+T4wff{LaFW*XT+zPJr2S|07!-j4W
zD&2dyG}524%+|PYHrXZ87Fl_Tl-p*=T67o1zHRjrJao#Qc||JO_H^7QlBCX+;TwOM
zlI#VT!4l`slM|~Zc78|2E-hjoj^>7WP^|pA^|i!O$$oAnD0{}qJk;{@AVh%rMQSl>
z;uY9SCqq5ja62fOz_2p$>h78#*+4om)5erAWG_H2a97O52`YStSba`uLUpv8dbG43
zdN`O18?YN`bf>Q*-(I+B@00X-N~o78R!0<q(KfT*x7uq_;2H%i(C7d`EG_B>&(gr;
zp8vxC3ts-NHHmAy5)$->X?W+sXz8fjji_Fp@>e|WW4@$TLP&Hv9C9y$TH@~Meruv&
zddK{#+Xfz|8b;1A<rafyx2MrjqHWK{jMus^o{9x(eG~_peoS%V+g5l!wdVUJR7~+^
z$mcg3A4c=EMZVAH>^&H*AzLx8ctCICRhs0gyE@<pNRr%zV6xuGV5|(^^<!zujrd+a
zvXBLg!RNm0&PqKUj+zKFx_EWVcp=nLw9;gFXF=)A05PE;t-Jb?N@^ePj8f?wQ{zPJ
zj76v+IF>7>xSwngBFfJM+7G}={(I~N&L)F_YuJ0MNj(4t!ab2>o2g*{+i9v2_~4%J
zzz3{;7IlNUkTB@P@d2~IkpJJhxRpQd{hYDGqptk_$aMZAe=#~u4hlkp2EI6Lb&4Pc
zE|Hf8YXK2*dE5ZL_}i|T(~{$(Ai@5BWN*N9T>@LS0@|o;CKdxq{w{2KzYsOQ*l*F_
z^7r4ehUKFxiT9o-(Qam=Z(g|>B3eZM`RzxVj$=HtRQ-rXvV1M;>K-iOczaUROz%~K
z?vQ8d-Tec$&^N1RTzkrvwwiCLH8r$(Z#Nyx)Nka&e~{`(ntE$G8#p~!@*E4^zc3Rb
z=8j3GLv5MLSKJUff>n)lEZu!UztLi6yo@^({ef*#4~=DG*Zo;*T#MCy6F9YR*c*FV
z2Ddz^+o&1+T@iLr7n0N~Lth?n=b*EN`!4=Uf>R(T#|(4NZ*GPRf6{M_+RuJkv#wN(
zOnf)|CbsJQ-o0=mYvUWev$+X*;*4Pcb)EW)s#bQ><z~O({!C+%%R@0t?AKmmN_dD^
z(!BPnOE~y@0Lg7~Gw%t5(_g~DA)LyemO>~w+^=@O?B1f(M-BxZ{9>MW|6`Zt&2(|T
zA-}pIfB7W%C<m|{3qI$5ggAXGt%QtWmoYmROH0-cjrG-ruX-UU2SAcrI}0nLa>Ob{
zNfHrnCV(24jWuG&(&R1T8!DbFkUgUtlam}cDEi02I<Kj91V1i<pFjC8%fPHpUoICE
z5bykBA4$SPXC1`Cqw4w)>cM;N?>F7Y%|<6kZ+1!aw0!8)Y)3n`>vJ>R3{y(@C}vm|
z#6&meDJ@kN82O=7gfwCsqMp%!{3$V|Ulk2vpXI`SKQ+7BhVU;ehsL9z7<o$9GE{y$
zxzR^S#nfUK16HW~ocNS<ym}n&W4A{R?nU6#BS5dV@xbaMKv9ht#<JgUz2ly1>uBCJ
zQP9@)oj-N#9=xIc^P!G6uLd~Wuh4gSX!Z0V=lmJh=vC~!`b!C(skE@8b1XN9_wA1j
z7ZK&B5~Gp)+-ZSwqs$pf4LLYE645K-d|~y7w9gs)cdj)+4V#BxAYKusH@2G5>w*)v
z?dbO|PVZ3}xoi|)mnbH1{wZ#@c>=Z=@iG@8c);?OkH)m6hv&M*hMg6A6Mn_&xoj&h
zCz-moJ?Bpgn`++kKHagC|8nh7T`MCK`1{}lB{T+mAd6dzzae%te-St6yFcI-d~5Nv
zOL{-CJm{+YX&ar6o?fV3x__5Vu;slV+2BUrG08(o7_od}LBul}aUwOFG`J-)6MG+s
z_JnhB`^T_wF3`hlB7+ktWr6wU-@3`1Imfr3h$x$n(f6L~C~>dZtLROTG(5R!FlNv~
z^a$jlAbU;ZsYZjU!&QtnXRjpPijKV$RA1{3z{y%bYcEy%tYz-^o_#uG=l3fa=XUw?
zL7zL{)iks!m^({T(jx3hdM-@)N|n>2ZAT8MEj-@U9r?^~9q1ck=$K4*b^Xv_36|UN
z<lfbq1xS-cxQ7wz<jVB4`2K!#ZKl-(4ZJp;L}T&U^gSg~efrap&i&{<2BD0w{U2ZG
z=MVLG>=3t56*+N5Cx@fj_Yk|?5r9Veo-eV73$=CIRDSca$>amCr+kJC!4ga2y!HYv
zGWd!H{1hf0yvYU&Od8yL^U1EdS<JPCM*ktPnlx`pl#x1%2Fq9~%}259B0RAp%(`jU
z#jPWfKRu!xOxS*Wd5)Ry69xw@+)n0Ols_KVGw&kytu%_dEyzB!M9fV@(U{<|X!HjI
zfOmso_2bXT4&Kt@2Or7~7O)84*i2|`O_9l6)iFo`Y-S#ONpMxNCmWbgY}t^50yUB!
znSXW|gBy=c3lRM)Nj>A?FB+xfN2EXCl@kU{Xrc*DCT(Q@^>LrH2<cfkYx<Yj65U|@
z0A7LEzvT<>#iU=j>kl+E`Xx7MgSa<S6>ao_8LTPp8=W@ttADZQY|-wO`C8&nVakEH
zi7Cf%*T{JYWlOU5c$pfKsmWh^C81R0I5_U*ZHd!4XkeATl9Os^LpeHX$dhWsWsJKw
zT3^cr^_BU?C{CTaJrAN<N~m!rHwwnnQ!Teed2DIF`S|Wx(Y3nms%M8*lT8xuhik?l
z{ho(e7)zDQ@aPCODNa8h3X2r?8@86dG^`s7p(+@KB$$pvQRdAeK7H)?2d|D7x!Pu-
zYiGhaR!3~QWxKIn>helRj|jJOsqF4Dg`eJJXib83WzZf4?+eLSrEX&VRa8EbyVt>i
z=Eh5eU({o3aIGtC;!b=ot5n1J<bSsqyTFrJ!djk+h`&tj^^_bgu}#Z*8)RZKHvPAz
zOMKVz88`zG!*YvVSaa{z*492s8$2_iW#P9R-e@zrL;22EVXn?22VZ>Bn<zL>IeW4@
zYyQkAex~nH<8?RCw&?-*5(`cdeb+VUv@uj}8{Cq6oTN~h!U480O^D6ElPpa(kdzlV
zpGu@M)1<v>UVy+~?)#TJXX@1FdEkF^dmW@RN95ho2ScitJprFTW4CLY{Q-Nl5`o!t
zdXgNPWR!rSa4$(CN1))a_3KE(>ALG9-&5n}4xBoZ!B;fj0VASwv{h_&<hkp&D%HCQ
zH&EI5pC+9=3U?(M_=*-{@k(WUgHyMqoK<YrxvAi3A)BAjob^i`>fOQQBgDk7-+FS~
z!V!BeT}G(&aePFdDFeNT_^|$No<h8X?f2<cX=5P{;ia-M)*guK8+M$|*9SVpB#E&H
zvV$to`x=9sKkqkF@Tbks!Z{xOY85VPjr8WH<ppNoo2>kFigkxpDs1VhKewoJD+d*W
z#(d+xUjlhLViX>qz9A+`V%HFt&z_=LRfgn0zZv@yLAua#_JL++Vyf~9?ezUx<KoA)
z^50s1HkHl{(u@_vxm<;@ivEfI-4_@~xAw(+J;Zk^NW+M9An3{7_4U=0*?&sMraFe1
z;!AqAW8IUQ((_?x5IV4ox&+9opIYcb2m2Ot9Nx*C@YXDN9^grAm8)YNUA(-&rabmo
z8FYun45hYvmC+QHJxEC?tSEx6FJ7ETPkH&YeY`4jtwIA3A=WZ)fjQWD@CdYY4L&jM
zIslV&ZJhDA00w054nP9ZgGDk2Ql*WKvxWe__S4@ITQsc%*n%srNXF^<Kj4@e1Eti;
z;S~KxlI>_Is`3x#MlqWY3<n^GYCmL+`o17r<dW5JD(lnI;fs3ydy&BFt5m$Cx|zsa
zM*%jQMKZ=Rh}A8D)THdAP2T&K+OK5@$LfxF?z`j6d7NbQ*HN*4n_22Ze%Dq~!JpTl
zsb5`#V??HwU`q`S5@zBPTT^=}s^o$e@3ybGp2dvH@pIShuSP)FBfXZK;%@APR-C-i
zccfF6U9sm$1~XAxJpzS%n5^iDzQGokuFnP=9?z}OD4@|7d`|E3vVquztn&s1C!-d;
zKL(+p6P(pG9&#&$zN0rI802z8lNvhWt^=uhGA-1J9?_j(p56^G{Q<Eu=l|y3Xye6)
z(-mzZ*FMZ<=)U(f`GY^9mR=hW;*R4{nKk6jj#ZurK7=tMjc7-2qo51|+={%QoH(#@
z_!BCd3&V7}N@O!DuBNkMhr_FUj#+f~j7+i8CTICj7ADhz3%(qI9~R0MmA$RB9eCer
z^l%8DeHVXv?iST0!6NCH5P}3I{dKJ<^lbI73|s1;8_~e2<X89d6+z=W2G!8N-Hzt*
zMj6iueWR)Ehb(h;7T^*ZD2XAIkf5j4#vLm=`u(E@Zq??qT6I<t^~CR=Vv2beGk9X5
ztqz#!I9d$MN|kbXAO0NfX&v0iofInw$e%oB4xE@P@~;TWb!m}z+tM}>YRZ585;#}m
zl3E-+c26&0a3bjbC6ulhufvy7D65eHWM`9db_WiilZox1)9L4#@)m{OVZD{G<G$Em
z=Ud6Y?1s5DuuUMgn%&IHEjFJ|Q|Pb4r-8o-hs$P4qq&i{e%Ky6Iw#Vl4$ULd5UXcC
zT|fFD3-6~;Nqmbo*sXI#z@+Dc#V40NfVqXPJQ6}AKSy^IYRqgq(`CVJp=y4oWuEZ5
z$=17X<p7l_ikL5bV&NMlou)VBE;r;ny8&UBfq~uT{LQu&=77^>j|t6oQv)@=@IFB0
z?32b>+nJeN+CK{w?0M{_!xi4*lP-nbNcQOaE&(@sR++PcaLhVI>>sL-m)xz*Nfi$w
zWbUrda5G#7-WJU&d*w-$y^VmlpRoUFdt5UmXm0Az5UbR1Vt-MW^iald-NsY-7K}+t
zL1GVE5cT-RNag~I^$pHeT9oEJAj!JkiJHXPo0;W^F8V*VQA))2_BUVl`~$gJYYx&K
zdg)TP*;s@zgD08;s7LVf2A8)clPKIUB>HIC2O(oIGy0+Qx?p))>{?dku*Avg(Nr4=
zD58yWI2k{35s8xhFstFKnbA*y&6CIHNSul!Y{jS&u4c4PHd1C=ji`*bRNb7FI#`ev
z1iL9Khn_b=rxrhI9-EH5Zi!y*<v+InGicX}{{!D>b?e+c*4VBYC|GQuj;1{XfQ@Qw
z4(<<V%GP*nQ`pg(N@oYp^3|4GY#ea-CPvPq2>`1XkcFXG`HQ;Z3SIw4a2LzmsqT$M
zZ&uGYo5NWYF#PMkY({^-;1&;%=#)HC4n~S+cz}eCm+wW4egKjV5GNfPG>TT`bOJEY
ziSERq)|v8_&2u8#wG(@oBe%;M2SK+t66F_WbDziWC%2scl8iIhU2j|@xHfH@qu={<
zGN>(`uEzR$JSY|TE!Or!1F!&m*s{Qc@xXZtF(yw=92rfqOS^t2;{^|_RL=zN?!8q7
z1OOaB$bM*<EuJ_e@qWmp`!?wH++HM-urY<X_)3>}Nv@8#dz)YboU;i$j6Q#A)lU9*
z)*`$(NKoUY3*Ig>&l1h@>~g<05IIgM;xgJ?2ky#Ak;O1*q(14}1zXRk^naIP!D^<V
zYWTJ50N6!UfL6-K=VNwnR!X2Lq7F20tt@mM7^z>TJ#a4P0_zRZ-e141ks%iH5}D@E
zZ*+uuc(%iNrX>}PNa@^<9J6~%6AN`mKh~(j`?T}b<^z9&s-PxX8&az|m$sveWA#Aq
zqTZ(9TX*)NyEWehjloeYyRHo<2`CLUj0dZ(7ljL-=D`CYJ^Oel@_7*O=YuiZ!zRdn
z>o{K2q03q$t>-jfqbsThdkEa?E}`^aNO&I9bv@&PiX-iugyVa^U9ro|b!<ug$`uRn
zHw@)a8mf=UMXV!5%~mV@GbL2px-umRTE$X@YFb3;`JLxRrbxbH;;^2v(?kHq{L^}l
zM!4zbKA-)_c=q@(I>L#Li>i52(I(s5L`j5(S~al^a*yK=+epFckM+-PqS33HtJEo?
znnTZmzpsoM6IhU%#9l0ODf6Ms=|c~VtXTd-L`LH>UBP$4jZS5%gs>`mkuhY6{E8xo
zDx;Zx;>hl?Rpp*~=SF(OuUfiwOs3^J84v;;;~!ZJayv|4zg2xVrO0VL19VL))Zc{(
z?LFEL!QBPCKNy7d24*~E@a#(avOh=1dFOyarJ4|;SnfhGKyU~f;H2CmAIRl<=}T;p
zDPcJde%P?a@3+&%km55%n*uy=1quQKyM$7;0Ci10UPb4Eiw=Pehv}8J;CcyxOY82Z
zmJP@KX0n4mIR9iDkF$4{F7UlOjsqL%$%5Bj2&0>E=2hhj{J>xX4#Ht_i^P9MWN(TV
zrYtS=n=fV;FD7Z%Djg>TD5DY2^fc`V1(p7-{mSsgNq?XQSXWmw_e`JWH5l8tSPv)f
z-64GiMi>DHahQ$OPi(I4-2GcYP)Fs<v#FL9%gbX+k9KPQ$LhLzIYFiQNkuODwBdBg
z4vmt?)i?*!fwlyumQuouVQC3)-ykO!^i<2R$u_t-KV?rEtqcuH!ba9isA$M{_THiq
zJQD#}1~>Aj>=|8*jp94;!)QI)7T_Q9og(&=?lX)H?(}e0uICV&Zi&TdB7%-a)aEHe
zdOd;Vju|!0meso%UuqgFpI8c+eBix?eA{zAol^S&HyHjyR@f{m@7l(jp7#3d$#l#s
zZ4ldK1RniMTzjCSDO}c86tgPZRsDvV3DcB^LGJ%iO=<kva_w41_CR^SysJ%9H}5sO
zcMKp=w}V&x7Ik5WcMI=MN5e6j?~uZ-eU_M%AzOMgK6!Wfz7w;J?W@-<D<akxf2o2G
z{h*V~?=yymt})vn%NzajMWgVQt59E@Y}&ybZm%9yYZF<yW5&^#f(sPwYt4Q#u<__u
zlPboc5<2<hwMd2oOg?twaK7&y)5DY@yWT+4s=Iv~^3gWejsk%I3`>UgRFi50#aQzj
z4o#HNi~-!<%y(+yhiq4BqvhU|+{3JLrv1~a7#85h^N0R#1(>v2ScVp)=<}?~iR>|4
z66x$>IW?yz={i!8^YuEa<9LCJN}akxiFD({%B`*Pop$@TYb}spRlLlX5hu4E*01Hi
z4(buEIQ?m9b+F#jwIUcC`pf8oj>+;kaOU#u!9<nQMbxjU3&5`1>!A6b*ItChAE#Yo
z6{m2;s<zi&82X0|Hust4pI*P_R@g!M$&D4Gt7I*o8+xLg2M2td38i);wS>F85r!An
zzI4+V)<1jUt@dUjZ?Wy%RPD_{o#^ZL8|n|)rD^)QGNNsKzlZa#4z_1E4z3`?EBwx(
zr9bmlpS?q!ii=?Pu8Fvv*T44LV2MI|Jw&+NH#9#X>pY5u?Nd~1Bs-czEfhU;lox52
z&vZx+zxIEdSt$692Rb*o-7%C8W;W~x|9pZ@@z;Zn%NV!jV=Ei#PO+aJUfyun#&>qd
z)vkzMfR2f2jz$_bvH=h=E!(Vau+#h1^0b|KPJ#zEsY0K3UEr%;okq^^);o=XKv|uD
z`$r!(Yi%4K71LQAAD@pVBf>H+j`GV2i(z>vvqidI0DAS{fJSBFyAgefMA6V^H=4|k
zIjmZFy7=*<spIKPhxv?hr_+i_lZ(4~s_^zgkzKvIJ(ua9Hhbl&TVr{khV&M=G$6wg
z{Qypkl68XP=rD+3%s8DhBGz^U@-Lfd^J*bZR}w2m8taNS3Zb;WGic&3&v8G4f^fU(
z*0{i-a{M`;d~r`nZ{@{rBz|o3@oqeB@pd?VV{~4K3RV0nl&xQxkkvkVBWyi@JW-4_
z^!9pYCrC2_6$^hKey^J97F?_S(|{=|=cdwb`99X{7TQ25SGFbcK*}-TvpZX9)9i6W
zj%rzpF^g_C@cn}=)E5eq2RcwgV84cc(T^Y0L$Y$r;&yUTr7|@lmy!)}>hI3vZOmYu
z7N8k{BmdTl`)h#-!L+0n#nWDLQ{AQYmKlRncOgIf7lYtcVwyurhHqtI6_y|pbIUrO
zR6Sii*Ul?rgYPpDZi;yviEyEU{j}N6#q=)rREO=kg1l&1-@t+^(J7Wn$hZP(-Yz{P
z$jPEBV#o{?nc^#;*<^`i&QryRI)2EYiFNu+CUFSUR3z)o+ofVXwfOUcW4VTz-qqq$
zm*lmPiY<qP<RKMVtYN3#CoaT^Ai0JjL)-h`B%S1gUc*;z?2PLh;J>vsiw$!}e3B&S
zJ}DI*g(1RZ)1?dKt@M|C?|;jJa2Cqk?AgzWiwa`@rYn1Sw8~Hn-eBRs2aK0O&a00D
z4(7m$K^G8u^L+*t=HTlis+TdF#K5BANSsO-f49BwGQMw!3vp^0S8%yrNXfPA-6u(z
z)4t8?t$}u+6rrW0pF0~pd-CU+(+$Pd;f@i%1vQ<9tyFcl^DE!BTNLqpH%l$?E<Do1
zuDC~|Y={ZRFK(wd+SyuS#e|juyyl-rxUL-*33KyF!1J4#7(cq!`;dV0X5O5=816et
zf-YpnIOGU-m@H5VHLQr+n_P%wpA8J(0}_W>wWXS6j<krwbV6t2xf3nE`jdo4GqjHo
z0a0UeVRjkZTfWWKc4og9$YQ^zN}4CLS4RdoZKWK}*9P1L9gX;{vMokPH!agX-YUo$
zZuE6y2EJ(}x$8UUQHL3n@of(UA`ac}V_1y77LJrH946I17SWdsMksp$<2MGFWdUHI
zYj|NRO~6$c>Dsy4o0kVZ<3X(F)klh3lyoC<>(~uVheNE6&_cTP-$7i~yjPj1jgfxz
zP%3ehINNefoG>vy)My6`s%JzKC*95ihL3UvkyxWIwfrf3ky8XG`Ww;v)9Uw!TOY!X
z5Tei7Q>YXq9BuM$zCo@0N&))ctZ@kn5gA{CE4b&lj%MW2<vu6-m;g{ad@wYoe`}&_
zHccb`!0B7y6@w(o#M-*?h9xbs9Q$Er+9Rd=VB0~Ze!o*{=XJ`gRJ<;51t+H2w~Uk<
zr^+2Hm)PibZt0g3HtzM41YIqib9QxYPwT?+W~e6Id@Y$RSSvXJ87;bENB--D-tEID
zlp5m8hEp$LQa0Q<kJ2Ldr;Y6cP%J9ZZ2k}B{aD;orw$z4QCXswbq;D=q{8LK48u27
z^fc90D5{sG$~Te%k|*6%eg-qF*u;+<jYr?tWQ(>wccY)zc-`<_K^j@A{tS@V_J%#0
zo+W*T(#Aw}nD5@(*Q>1;5!$mEc<MxNh+-%@*O9dM-6-qC47<v2*#!3hkrk8nMJ*W~
zQyQwYgU6B`iL}dzEHW|fKQua5D<0HV+Eh+Zo3Z%Lw~^Ogp`zf3SsK3**3Zo#BFkFh
zwcEXY!rLTT7D8+k)EH|8Ugr<mI=AW;NuT26@R7(~E>~7}*ZOo23;~4ub?U0C-F-gx
zlTj)PE1#gCDQ_>5?e~YFsY@{o(?Kt$d83{3gl}_Hx`Qr1c02(~Z`231SK!5|SGV(D
z%Adt-n4kDd3^cQi)KotbX2yWDr4yB#)nx6rQhl_${flc_VCzE;Jx77uf2#c@u+79W
z4_8Pk?3083Gz`XiokR8xaU8k5HH6GuZ^pN2mECVcW-)RIwC8zaf1JG4&N|~cX1n)~
z<KR>4-}FI5U2k;@#;{Gvz9NWy*}%~$xCIkOIKfSmSBo8=6$hu1YBGLb!ClB22d<4R
zt^T&Ol;;qmg9U%|1#7&VjL*drvv0&nwMmoivv(#x(|K@3FgiwJzEpm*$~L{M>6pi$
z@iMz7mes>Fvp$$`mWCJr#Dke;_NYhAd%cEW3JcL^4<1vo)tABUfBf7|&9$ysTycIt
z?Poo=USNrP+}=eS#;m0BGki%Gj*>Q!bT)J3ASI<{N0p7GsSa{|XjA-N-24vlUNWZF
zQbsWf_aY~rg0~-9teZ>iemqO%PcG9IAkdoM12LPDU}Ufozg;yNJD3>7FHWG?zjLt)
zOQli5+~}2%Q}HQ5!`n~oS<Y?21qQnb&&X+U-+xMor@xmQrNz+8gRW-dFQQK20Pcpp
zRI2hzigr^(?CC`dS1^*5hm8sXgT$d;-khkD5}oH<U9RQ|wihqwDT;KN3-R0bRG$<b
z+yFdl+H~d7Hjh4PBlbA3lFuE><kp^(pd>AMUk5WShXYi_kSh9vk2%_?+ISTs_Tyuo
zGND-9ya!~js;p;0xLLhOLhcv0>EAM#xKs3?05%CH?@GD1{$ckdgA?}j`rqCbs5iwo
zmYlf3JK<h4>bZIl?liI6sI|67a;VO+uRwR2U3Ak_iw#z=5vv`6#*wWwf@4q1E<HrP
zN8uU2$$%r*h&`K#bx1_nne57l+p;%Z7Q>+<U4Ee(GMMVqk>X>0dSCii6D2pe#&4{g
zwStvok>(L)r(EBP5uZ+xXs+yx4mLz+Skl6&5RY_W4aT&}t+w}On&&DJ$sbc69P@KF
zr*ACo(~?h}GZXVwQNi1*nc`(~w;z`>yU4@W6*`NkIW#OAy?V|>cgffL5E1*+4<dY;
zI)ieIg*5#OgQy=J3+-GPnvyv4jYz`z`Z%q}E~Y*R1_{p|@!>6q1?D(0grl;)IKN-O
zayavx&ZVGXiF;^t!SpsU;<g!|oP1iLXV3&IE<zQ+fQ|ul8OGJxk`Fr|biM=b6T*6-
z&i4AcHtvgS{mc@U`*YW{glCcNP~kAdJ;&Mn<Ek+&;`nS03Ww4>uP2T;Tbh>Mfvk;X
zdTn>;gBm_(8Yj@9BWt7qMZ%>3zh2@DBnCm8Y#yz_X7#3Cye)d4<(_7w`KvB4n{mBk
zU2VdCi%zX*<f!H4#a$uHP<zJz;_N-#nq0HCZ#qHg*r=g{q9P@Lw9tt_1O!Etru5JS
zrAQ}2KsrbX2uO)^REjj|MZi!30YQ2-Aian7-tIm7duHaH_nDdRAINbex$kwYYpwJA
zog)T15Q9P|EmQYyD(TJowC`W+S+b*g7_YyL?zT*x&HHf{Dv*4}K{wc|By&dJ2I#@O
zI#v8#o!l)7?P>1$`wDD{+G|1_@TDVfzb&w<q^}wEV+dSB91t22iiU60rSf$)dAlic
zOLydtrzI0VyUe&<A1wT$la`j*eMY9r1$OPy)u2uw+JEIXpy-P6N2fX9u*&&RkX3Bq
zirfc$eBWSNh(Sv*x$=S=daG7&(8zt%?u1M&<}KWg$bf7GX@%g6KS>8SQ}Y%R!tCY;
zP-UmQ8~rs+wTDWBRJjXSR8=F&qURa~d#3uQMb@^W6mZgAWtpYB$QYJKw&`~)`06K!
zs<}qT)VS@*D61i{(lYDfN9u5LaUvb8&!0Vhd-8Sn_W*U*QQJ7PBXi>fy%rNK5`#fQ
zQoCEncDe6Q5**@}(cScNk|B9P#+g_T6yh?2vg>46+;!F%i<RyC#24I7-$6rV)Gvz3
ztI8;=8bFTTTm;O;3R{{{?by=CI*(BYc{LTIs1clDc^%QMO;o<+<x9|}dpToSdbHP}
z<d`AZjsl9S@xtcCOgBi*ax%@Nkyd6qIfI(@Pab+n{ha0(UX39%K08ioD0-Jeu@=pn
zDqHjSQJ2S~U1#2+7AGUECY_q<)8cv?NeT6jtLsbdU+S~vIKQUWv^`fN&GXisZ?5)Y
zv&2Y_?|vo?S|@>6QE0QbUX}+oSFJWIad1By2=a^Tmu!l+xLv!KhLZZGIn?G`Sf9|d
zZDY(jC|x$$w;Uljy*<@<%9(8`yQR3fj!{UR9eG}&;Xc))<Sx3EYD(F3nQUtD2;fe%
z1AEbVDso(Yd3S}Z86>N4^B7;||H`4Xro-Cs)p>cm(DrUwfI`D=iA&8@I7Mo(n9Fib
z0oi-^EwG1Tv~TlZ)wePc(_j<2bMwlBfTHl&oAaZud_02@8?ZyjyBC<z7?$nprCW(m
z9IEma>>tW60A5e2b!cibLvr6xixN2*(0WZoY|Rikao~|5A=9kXGrFzLL)+=rmq>y?
z;XE~rLcqwN!<r+R(P)^*+l@Zk{(X+`-jBmLpOhnEItQO8ogeu0xK1ts+96g#@Z&gV
z|DZsvOu^%?;Z;15?ES(Ug)gwqHpWgoQ)v`X>fzq^JF4St&;7k{4A!C2xIQ^8Yrr5Z
z#WhbolN8DbXWACrw_Q3Xe#m}DuNYa7W*O(Acnf*tYX<N(ENONlr%Xkz-DaskWFMPu
zF5J-0L^mcSW-M9{)>_;7x?b0^ZDiFi!F-lswC9|0LMfGtvJFMvn?VYuA97g?-)?ee
z;4y7Pe%QS=D^#W_|B<ST?(EexE+*Ul8;#u+?+e|nW-+c`S2!x{VlxtzQfSA)ZwBer
zSX$8<jhr*U3hCLlXYAbr^IV1~{)bVjwaad~KOWpVTXn^{%`Jhg>cJXn{W{bvi&Uop
zAW%{ek#)ErG}ivPyRvu`#e0u7F_;Tc3b(3FvkuRfQF>2)XcDoQ@#fn}oPfk`0^ITt
zVvAKYl$icjaTn0p70{v9&K@TzMonJRE4rPyf9E0ChhpT2ROpOFIl<l*J;y`d@6r2`
zsrL()y|qXWoUcdw!OIfCpT-g%3I>bd&?6}0FRX`ucv7WSKUh{P|7Xl)Vk^@-IfXed
z>e*i1=e}Gfg<Xz@o*=;aFyNDOuSnnS44Ru0t>iIMHgH9f_vz_9F44uK3}+vspFb*O
zM4fFLG>_}-bulXs|J*48H9l25SFzNWw7*s->u}X2;N_A$x?xYxot#GeY^CwEhcrJ%
z72<ntWuz>#=R5f1r=}@((WsGl-_{%$8<}GBHD<R)gfG=1eoXE>DGAQOl%x05AjOAO
z>k+!02VT~X`Z`D02EF|Wb3?Vs!}pIIAZR@N)V?e1oKXt8_fZruvp>>n7)xN-y=?FW
za9{5`s~@`hksPqWA#ZmTlPYdD_Iy6|KCu2QJ@I}`?p#*a>hmcCbE;%L(h9FCJmo2_
zWeJVW+Jm*IfU|4@_$xjMek+q>+x+R7_3SW!+xJ+!jXk+8W*!AzL};-O4DnYiKd(zD
zsM1qp@!beXH$SY1D4L+l{vLE!2^M~hVq^FG=C8;!v84u)k8|yZAU5d@c@^F{Jr*N~
za>*_Bux}ZoZ96=z<zAKMg0Vwjs!z%<#Ct4~J98=EvsW#7GMG^&EJn(4voqz&-}Ery
z4!$bfiacmfiEI(*9A-k8<yueSVIhGIYrs0Pr<(8oa{K{HkL!0a)XhVRS%(JXL9HlX
zzjv}f?Iz{Ay3gWAkS0fX)|xAqX4AcGyu?(v$1lw|v+d{8dsEzgS(n7#N%<^mAkVL>
zS3H#^3%06Qyqe3N%>BLUl0M%|5pUDgfdFg6s$xKL{tD$i5j*{uC;tsi;X4F99-I)H
zI<zA|3+}D}2TvE4iJunH6g>9oTh6@JxbZpBAN!bUT~ykL>cudHG`pu2W_Mi5Z=G1#
z+sL^IpAvcqR`KF%!UMLebADdv>Cc&I;{1kSSd^OFqMD?F>$PCKZp~7*%+T3=_|S8|
zlN&*)VGH*HVEvV58661BQkmDdOR`QAAw#ZJ>N*7mPrB6b<IS8NJes<wV4Xy9N|lN3
zR|u-_p+RO_2sL*N;U#EkD#GW@QY_)?mpZI?LgwydKR6gaD?C*xaKt$=5wY~SBNs6<
zd7%vdhhqa!tXRD4-gtCb?X7$7Z5e=MWP(i)FE|%Nq8wOIHHBbX>BCc(A!GxN=fHK3
z`m;}_siCfX76h5InI!3eiCN9J(yxx#;6WIK-W$Py;=Lv4xs%)e0S;L3dfz_yK=Jz7
zlSdTra?`$VWtaW(olb(3An3#9z%78M<50q|9YT8bIylUD<y*4t?!eB%?e&*VJ)aLh
z93M^$n>aCSUuYJdp>}U`!k4P8EY_SlESy(FibLgdlo5AFGB*Y;cX8uAc9%4-LJcP@
zc>S1w-cQZl<sH}{_)XxQN|obj{c+&N8Ko1lWrcn4CQN)VP>YT6J+tc_uXE<U93RoD
zTb2QXyNxYq5&z&VWDqK|o|dq66($YUy@l}DG&JNR9KUDqy|GJ8vir(<eCtM8f#SPT
zb76R8N8zhWL$(%Ob2RCQM|CVphjPrx?4OkVs=fr;4T|qaCIAihv-LAx*5?<3esWVF
zqnU495&rONg{XlEsm`?@8VSt0P}t!NGnXE7xU~gB+JbfaeHm~)ijni(x_k|%4k(u+
zCg$DNbT77e>OQrtZ@Z{uxN^OX;#0btIg0O)-1z$1<6#qr6gp(v(Yy7hp_ItobAz14
zk3(H&xsjGoWcOET7MmCwA9@Y(4JA8|Q7cgM_Ri;Q-+{&)@$Kts>`CJE_%(5VG{TMC
zpP>L+<ip;bB4G+OIx&DZNYg#LK(k}7Lv`}`{NksJlvWtU4_X1mm{wiW*Be?YYpE_z
ze2Ap)Ibkd0{be*gu@3RRYcH9mPTdn9=&08Chp?yI(Myp$d!m9JVY&B#o5?r2&eYdM
zFaS|ud)O<W>vpw|@lsk9@1rBy%-wY+(+yGKkuQ_YK3Rge>zzTw<id{ULus)lr`G(s
zCTkj`&eJvTPD%27zdEF>il0p~+AY5{s!TiFhYb|A59k}z4{QEG2sz(v`|Kjg*}LUd
z>xMu(Q*c#7)`8{8rlLGJc`E6S|3uo}%k9pd<K2iQVH5JsW1w`q=kzY~JqCY2mFG-=
zc7`4KIeql?1)2!i`wt<V`3T9^aNE}-+dmFJWIR31_~M&#oY%NfKya^Q!)7-gzfdfG
zpSqu^<Il)(b=`p;_cY1|5^(?g3eAJnhSxv1B=2V!0lE|a?Rrsbvr`lB<iw4YKB2kw
z+t9C&Y`Z&+0X<wwK8x&eh#@%MnU3%(f6di@92|5|p-PiU5#3DuIs$rgv`xM%tTuPX
zF5n;g;9#Iw-vBa;(A~&WDhn4RF*G;ATMQ4fd9MQ|q1jiSd>s2PDQv$`-=S-U`>{aq
z9aCrESzszq3``_U!#sMlf~~{dwp7K&XAs>#mDn~>eH|mY1uuGcshz0a(!wB5i;2Th
zWN!o?(cIDFHQAC5AxU$bqAh+L?XjQ=TN)4DnRy+1w^Y@q)su<IUxn_0`xt2;d33A(
zv%y!4w<z5x1dWc`>tksJB;RhaNqgz=X4UQ(pa462cKCblR6ntI-tx20W3uW)xY>VY
zL5GZtnlvqRx`l-h-QAJ|E_{Wj?Y0?!v@-FuPR^ZA`&4MZ`Qni^mLpf9SH-7UGi7QL
zM29fAeV=@lBYZy6lkr(2a(+_`Hr)5Tf!Vi+0l|{_N@Mi-iT>Ml@j)=?g_PI}i;UE1
z!%l*C?HnlgaOo{+K?XHxbXADk%0VumzgK6-dNvrtm~r?bw>2h<vq#)R^2A5R^-^oc
z*`>|+BR!h56sY8u=-jg$NRc*+U;Daf6vcI!Z)|s0mUmPgBJK4#=+<Hs3M&^BHrR~M
zOp_c+%Dg!SDxL_x-Pd;nKfjLF_`s3rFZ?3ipX^%BWW>(anNsbvM%!bWJvTlr-@D4)
zzPUDIb%#wl>kLyK!{iy065^|(4IyI=H`;St|3p#%@}+OKIA0sAt=SC)ibZI{R_8hZ
zn3^ve!PT6~K3eS4Pq9#g+uqN)+J7PcfDMaS`nu)zx&64t?9er@H6Z1Za=D!%C+8h)
zBcKi5Y3OGt6}(tVg;w8QIymr&KYY1&y8Wbkee2yi`KQlp*o@`Y;qlMZSl{!njyKI-
zzJ6b4EJYWps6pUckPd3@Xh(tNTj?BUi7Rw~f9M)ji{`C-(QdZ5XE3N;=@;Zt4jbNs
zPRSK(#H6q<6WkF68nPOIv`@JEuijjwI*PL=;U?wUj!|5jE9tW(uN*&C;73O0FLE1F
zu|!M03xG?7H&U6=Xf{&9lLH0-bNJ2|bO*v&tGPBrr$e22^f?@YzwRU;-+=NsA<OM!
z;i3^mG^8r}-@bK1YjN+!Y~7g>Xz1rhbIoV?rJx>Y_)dHbFXxdm|55-+*X?yxpN_{S
z5+e~9%9(8>hKALb5sf(7WEJs4oObPqU!NEAnY`Fy=h~jksGyer(iy`pFEn;=9q=WX
zG?Lm6O0^+~-QtTu5A~MPPHW|P`1EQ73$MA56&Y5d3#0DC->)UFGp?0YUwwLIxPtPx
zCF+4f=ud*fhut-U&0p<PoqOS}O!-nL?Vb9gM2>9n7=uA4PbbC89ipz=m37A0vMuWu
zbS|YXpCnApvC-w2&g-1xnG17o`)Us#n&vQXgYdFpa*V(%vciXcmXcyEtXEma;X)6N
z&@^WmWX(b|bDj~x2;(D_R(@rX@?hCJlj1^fHhWjHTWx`z35y-3r1Fdcty30=gEZQp
z@g6dDOU03OXt-jLS39kmu1Rpz=QC1~n@6NoSb1=Ba9wP%&0)#Xe)K7<=_T%KXDOzE
z`b4m5IMK!LH6Lk4pzb>B*=awn%(LxLgyzhjimd(Yr#6E)TeC7>o4sopz?8wV;(CW^
zB2+M-6sjpIbTR#SN03+)7rmKfEdO3%YsrDIF!zS==uqFba`g?gbnJz0+}dXW@Uk&c
z-}Yoj(CqX|*|E36o!Y5-hHd}YuCRs}3<Bln4|pl=-zWy?R8#Fvcl>v<j5%n91JaKl
ze;X``%joVE^p?H9w9-1@cMLIC-CR~p1IM#YL%_rJ=5D{lLpQ+(aZ3nauU=>D(fFbK
zGwH$S*0NQHY!O9=vZIQpG4@LRB$ABh3}08hkZM_w&`<4)nR|azRhDoe*C^m>+Hzp*
z491$<Lj&RmLVuyT<7bM!6zY2C099jta*!>yt|EK7f83<VFp&P3Q4}3Gf2bLlyhvJ4
zkCx)2yKgB3w%RYtYXr)~gzk3s=#@2R{<^boP@=-S&O^5wP=Pm8YClGJCMwcLUea^#
z<lla>SZAYI?Fp;jHogtsGTC>0VU{`aeU}0apOLlk9ORk^DZ&m3ms4OR^_*^J9=VqX
z%J~xzpR(AR2I;wTIdxVN@Ki;{*G2%=jQ7C`qv@Gb%i?{|=Z3&;Dzc?zp}3ow(H;c;
zT%Zw_-KXc*HILaDcpcX(&ysTQ{P@0+#kfoC<Q5CRv@ZkUjk$);KxbUWnLnV##bevA
z)IrbfJny^C<wg-NF^RGQoWkjv{eUo@;(a5jwc<$bivZ7b!EMnx52jFu8V*Sk<<)oe
zQuljUX=Sg>I6*ac#%Wug^YN&aWJ$J%Pp0h{sLkc@@YM=cltXjJ-$*#EEPY#bKOHK!
z_FqcqIu^AU0V`Fnq2q0umX5zTL59Dj28TH-Ft8^nYRT}{yEURGYsZwx9z3utc)h}_
z6D}=q_p;|A7th?n%}mM9=7)u&CMDTAR}l?DE<S5IIcO<cY+6mJyx{i-+g5Paa(mJb
zZx77}sjm9f8hI9@A4`}e(d>uOc=mmhHt9wYZsy(&FS#`-+V@R9zruRNZNKQ`W{JE?
zqaV?dvS)rfCc4woGqSzv{&q@ziCmv&0Ap6)?#n#)d6mJncqnBO;G=livEI|DU0<nr
zxcwS)@OR$ObJ1l3*Mn^~zxx$b1!4#b@h)RzXiRd?(TB<``)+_rg;y$<$YZz*P@J~q
zm8roO;0Yr}pKP8>`dxSE=fpp#t~ozJeXX^oR9TsKH*QEcMmgcJ5xlLJNxdcmzhf2B
zntwI@2!;=<Ir^|z`&xTs*n70I5pgl+9Ml!kt~zJaa#>ji;>6}_O*jE``MBM7T9=dC
z*OGnNx`d<B?=*S91*y4jzp-e^t1=BJWm9>rabTD*Hg-B{ah?3gwGoZndk(Q6prm;f
zjL4gOFB41yC@@hZUYz-&v)b-)8$@)PXyFv{A$endqgfskn!YV+e)Kq~r~yIquKh-u
zL!-%{7LmJ1fKRQ1PTrFbw_3#b<$(X%_L|Dh&%FAwqM-JD(n8X~9a^@2HO4mUT>k)U
zpH?Mg^lLUTYmr*2RRX$AB<hX0yJH-Brf5gX`@KzUyE*7333C~4xuls!gD)t6jU`PK
z;pb<Z5IR_uyk$o<zlQd{YB8xL)j;P~dC};vsOq(cY9l8kGZNCges;7tn)kdlyGF25
z1)47FOjX}gO~$wWWV7!yH{})hGTtwiprOxx#=%z1VytAGhF0P-ZVqxJ`d+ea7^PEU
zi8{H}(!<&ZpVAOtONOl<@!`VH!lpl+DHVir%X>kkuS90JOX`Iv7IYO?K_<GMjC)=Q
z+P<|P_G+i}Ou62lo=j1KZY-fAs#o`PZjAKt1IfK`U+)K*D~JW}5%=<`PDH%<@saoP
zP=PasWc$evfS7ej(?7%E%ztm$5qxQG5ckinb1F3F|Hn5wVw030xmt%NQ5GsdVQgHV
zP)SPSbGyh?UMpQK8s%@wk9oT&U?R;<*b0X|(0NW?iA1V4KYj3{bvtctfP}Y@(4~ah
z6zg>);cC{KFHP&%MQN`vInJ*yMG!!?=3!Q2w=BlXXadbue7YKWxQ<AXOAY5|eoFJO
zJ-<F1r^(0RbA>0L)fW-N{}92i<vdbgK65I}aQn(G`jbjR9fNFORyjCEK#h(w@S*z^
z`ePwAWWQJ20L5EN&?6VpUN3gn;qa!Kv&fxnUG6VL5tz#fD~~NBPE%B^){X;raS(L#
zHH{s5Ss8wl(yz9Y_NPbLRvLrBs{X5<?r1#f(iD7yN5+(>uVy(JN-Upz)t%x=cG;Jt
zdQk)>$-5S6EguFB-HD@0p=lK??Hqkx{SD(aiVy39lo-N49k1*B*!rAdymx&^2Z`3<
zIiM(K!*0D$?18|N;A$S7WuT-ljEodLJLcYYP_oD7(zWj$LJJ>h;2no?%S4_+zUio;
z^!Rmtj)w3CIeMBKS5BxXat<6r>69cZL{G+F_b7riX_Z{}(wIzkl#xaOTr*f9o`!8B
zimwTAcMDXr;~v>gMZ1pDBKx2rc358im1N$5pG|2QIVCF9-U2NJs!UP-<u60oerl2{
z&;$<humvSr_;x$QrFSg(F8jmouD&|>#u6v*K|i}Cn3%Yp9`H%gm&~=fj%re@-*;Eb
zU#$OE$~<5n`R%UzAC48Rq@3ViIKMrJU9I2?H@?Gr@Si9!mQgtC0urRL!&EBxW&!ns
ztbL1~Y=Mm5-I)SItKoB`8#lhC%K!P4A{-HElOANzrIz@)l!wDUka@Rmwub%%#p8=D
za6T<McKzQH_d-pO(;O4jPar$2PeZ<vYCxgUJ*3i3HezQsfD-hSfZ_v5h26eF+b6>{
zKnYfPwqP4KzGJH9c}CYo{+_pXw^KMu=F$_+nK(VZ*ADdD+?+b!<#Ek-AE_Q(Q7f7|
z)$in~KQKV8Ln+^E1}HVXgq0m#3)*9$zr{0MEEjJf8`$JS5wG9LcYEF31a9in`k5x9
zehFAdq5Oe9_xwOYGH<#k3(j_(^Z^v<vCZ4I{NsWBtipL(Rka@=Eq7;)T)~{e$|62~
zrZ!0-i`>&t3adyTg?N!umd%P8Wbx=R=ts0TsrYPpgu!l54w&*y9F^tFc6K`xbV%!Q
z>~NX-r(H^w<$G-oOS}We(a1fIt`peORFKEsuJfmp`GME5-#A_cj(~*lj60{u`9(0F
z;+Z)U-MuqBfVNu2E|5}42iwwphIpTSQ03?s3fU8?-shcX;-yKu;!UgcJ@O_|_?l4w
z>CEnwlJB-B)6}MfG{R1N^TK4`EhNL+SJRv4nR*+jHiLgKiheo&@f7xPS**=r(TeY#
z^HQC5z=!_&h_#qY##5@&Q#K~b+|Zk~m%E~-`0NU+;u|c-OwLTOpC7t&55Bi)_(HNf
z$MJ%8_OXA2D}jO{Qgu6|@P!#x9WwjDSJ-uA=lZwVBc0#X==f>Ys92XM6<R^_?_om#
z@lxk6NMOA<04qgHdW9k&<qS?Y%ExZ|N3yt9m?vy%S_k)PZ!`9?h0JgD@pAfKm^Ik*
z1RtVpNGeY_)9*W<P_o%h1#czY&mSt&R6Dps^N`z1Kvdz?&qHc`Gq1}n05$3Dh7noo
z9n{TZ;g$gxb@^YyxvC5TAbKU)t>>^_pY!qVch#Pq9)aGTg&r)6G6@GDJ-YP*KP29)
zX5XxRgG`8`2=B71o=*T7{<x!ej*3N`ni)xjz#X2#mO&|yGIiNl?t3E)Gn8PBgQi=#
zKR)MtdchNHZ9wx>nkN-7;2_WV)Xq2v_mLAHe{Wg8p!$4;W+f&==XX9c`o{1zalFlY
zZU@apwjR{)^VTm*z}62Gab5CfAH`rx6A@O2*V{zDk()iD`1U2v;*>>=+<UU%s7A0L
z2yrsxX-Ixi4BN$sCGGOaJoC?Gh*9z0H4t_Qh>NWaW~@mhsiWu$ZHLjsZ;pARjQTXT
zj<1gt%cSahy<GHBi!npl5rro^3t+&lmw*L!J|E6y_nmROF8&C5LgA{Ctugn!(qA>2
zxB5#)gKT<#k$((QhbUsjHen%4tD8D(<qGA0zpLUAL3N+^+cs<7i2nfOHa#XrW01v&
zH^Yw{*S>ERa_VS8HsqpquM3kMSCC}nZ5rU)4=V9}Y!fi<3TY1RIk#=wWvr!b$E4M7
zmyMo>j#YDX%^C5|NPN?{qY&H|)L$Uj0N|tJxHL!<x4m+WeCan0jQGA+jN^yu9@(4>
z@DDc$n;5a!E(9crjzqo$cl`y#T>le{3DmjU|0}bU&_E71d)=pjeWYz7>NeOc2sRq_
zj0YbhaA-!-YChf}q(q;Z*J2$=HH=vW<o>*q`^r$}bnY16*oW4Gpa6c-Y4cclUn$)C
zo9e``-W08D(^Z+=HM8ZYK5y`oRV)S45@4!EzLrAWl8J3vTx{o)u|p0xGzLvKBrwT8
zV~STtUZRsM_r8te6y<Rm14%k+03VE9cz-CFfCaE$s!d;Zru5xD3^QEBQVRDUzZC5_
zW6KRO>v7msw*Dyw>O7+y5GQ*ipqAV0+M-eFqFz%$xq!i<*oJvVTUJW^&JiN+z?RIy
zJQhK@z$l6ECR5diOWBnlpc%58mcz-|XXF%4=wKS0VOIR_oG4a{S!jyD)TP90S?KR^
zowixtk6bX}fzq+uo@2^8Im866;8@0F#%UZ=<jOb$Z}pO4;`_`H*KQd9J}_^R^N-%B
zb>Xeug6h}I8wi_kD7WRrsPZ={&S6KlcBE_BeppdF@sr66JJ!k93F9KO)60G57sOM>
zqqHYr#}ys>;gW-5%R6M30nJ<`a;iD37X59jKZmKX4}<eqK3y{ws3eo)DzevgKDhM6
zM?4ejjylvaG4I>5X={Fv8-$U!x9XS;%iUsWzE`<O<O_AU2RES0A->)MspJv~%r|oc
zGgSKxPu(|CcSZ2|xirwkoWQ0UzZ>!1ET94CSE8rv6|TTfN#sWxlQC4KM{EEtK(l_8
zl0v?$<y*<4-D2pCx$`?=iNLld&H2^Lk=9=r&;aP4IM6?Ry#T8C|AqhAHwD8_UknV@
zSAJU+Pd;8;q?U%(dpqq`M&cdy;SM6!Z87E=)rTA^WU%alo04!?3HP>=imgl5T@rU!
zs`Td+XzjBb4wF8k@5hHOpp$qx?{4U?8cgUn0jjy0pO&DqlS|n@&cYZWpspQ>@wOH8
zQ$Ay<F1jQZCzH_JjD7P0PjZ&4(3+s<Pcix-n@9nQ!sjKD+>={m+0+1LD-$?&nOH(4
zJ)hjmX-F|{{03NNxw+0>s-0g~sd&@biPUKc!`!jwwWB~_mJwc%`|5$*c1qtX!Y#_r
zZPt31?lEewQ7ulX#5SBG6n}&X;YsJtkUJRgc1B|a46X-^cicFe9}gBs%B24&a|i0p
zScnPT2X@EKQDqcWfmF1;0Dn6$RQ9@94OUB$quTTatNP9aa4VkW!1}zQv*o&K9*Z={
zE@^Iloq|lOd*#seHygUuQGH%4vtsLN;z=I@2ttgeC*Eq(j5;nSJlSrIYQ7A9Cr#$D
zcgvLJSFTgUm_>9CjxgcMRK!nheKjcji8tKr39A^HaH?Gmok1`eZ)hE+(6>W>Vuf+y
zB-v3>qs5%Lz(x-eip9SCGLTu?`17to)YFKRPR(~1wTP4{@0;7EDhA~i&bLCAkBd$0
z`1K(yFGBPPw>N8&flboEFMj@*D0SbnDaOeYTebaT$^qoyl^0?-YX<Jo3R(d!k*^*n
zVU{Yzr(P{irXuXQT9rp+KYDLidIQX2ovqhFHl6&9SM2sU{)RCAgDGSQ9HoEZq085m
zC62s|2GwWNicFvNMSzd5LNz3tq7bpXQMejE^7y*5pNMeYD0!_E9el;J{?qX_b=bYZ
zAytr&Uv-RLHmNa!$jfn8Pj;>Z*6Xh+<=5U0Xe<lM_!>5;Ve5x1@u{}l>F(^=7a4fW
z8fgcj=}N1vJ;1&pdf+H-6rOaWw{7f)5*eNOO<I(he<{6E%4ai%8^slyf*2zPA2)^p
zw!!S6z^aAf{T#`pk%l#Pn9!L5`&U~z9tp(Ldh5%&uNZ&)R4Qlz8$``^$B3hkvlo*{
z2l@2|vJ}x3Q&AWZ*Q<~kWyHof3vdud&BygH!MTE_OQAB=+NDQiPmWsi(>=+8p9D)1
zQA!!4@2SdWYKejLJdOyf;UMImvY!dOfW67CjU7aA(exRvi;ecqbhyc8-u<%X7-~W%
z2HLJ(c83Alpso@q7Gbh31iM?HzSVE+T-sx|hD9ZfhuJN7v>frlhOz3OoGt>STYARw
z=FFcx0-t%lY_Y&Jm?L&|Qj@HvgOEEXDN=MKZIZ)#?&Rc@zLU+bz?wT>)sfQh7~TNT
zh5HUL_r7Z{&BUo+^rrODq&i!fk=ZbR{i#c~RLb<S-16s(jzLHHk#^m)^Ryxr9Un-Q
z+m7#z1yARVqRhQ65MJwYkAAptr$(%fEAw$q#=9(h)Ns$y5LlPv*oNGcFj^>@PrSM=
z@~<D7bA^9!um9n(=~GVVQ-^4l-pL6)9q6no8l91SrwmbW=FZgO3$EN|n$oc=#E&OZ
zGYQfthu|gevn+svwx&r-Yo5ApIgDqC<<(m11t7qZZ%@pS(!SK;i#5IxBvsD@wH^<1
zJC29uehLZH&-B~rWIL&vClO!K6zacFoyeqdF_E=Yuz_AF)ogdrU}^x_&nU2~W836D
zEtnk>5-9S`6aGW5am(%cTc4|3b>56seFFW?4}X@fVD3FJ8Rb?NyT((dJw4zXf3-}g
z-vb*qA;HAoC4^>H9jvlyT-}ONk|)1lmXNjFD<rADQ`B_dURUad#-|M_L%|cyJol;B
z5sUl9d?tmNKV!}EhnR;nX)O+#g&n#t5-*3)zX%ymq*%TD1x#?7QkErH<w}I{`OQA%
z-*(a5`9;ew+DkQjgIuaa<56RHf6q&i!E3uZ)TA==)ae-?$t4ea+1DS#HB?pDZlvkP
z+mHM#jWK_1_f75-X%L`@V0TN(GOh#PldfI&A0IjYIINMY`P&oNh+>CVmBihXAnqa_
zk4ofD8v8&>gS+e99*psm>a8~~d$-<2<I7{MXVzlQ>98<hg9I}vdjb*sA-R$9J><2a
zm?<EN#VP>wb?uE?Xr$`&m@lVvW%Q3ezer6;{#dV)gxpQ@Q(3_C{x90^tY84MyelyE
z_YL?jhyM9wXh6LR8;{mT^8wyUD&%?##~|}2Fu5>)<uy<ODm%Tm&3_QaC`+s(HJTj@
zVYFwwlv;KF2m%{?lHWMqjB@6+V!kz)oIzG4|2qE(dH0iceA2lKC;Yp?ba=g5Hp5`m
zZz~vz%kW0#nhMa}6b-UhEzfmQmy#+nmDrRAKjmYzAx0E5q;x!$<JF|vHc%An{y37;
z!HD}Bi<bAQ(uaX@Lj$a(h)J0w`PaLQb0<9%>drW?AUe2GZow^74)=>}nz2R2W(?X|
zY9xi=eOm@GrgeoczptS+s+;Znbq5C5OEvaTpMf%J?kidwhTBHFFAWE+4&zVZYe#(X
z?Q<sO6CSPdL^m;<Q++bA5+yl>S==5nNud``)vPZVzipX;3z)24T`N1xps*+ve8@K^
z#i(cQb7?cglF{^tk|HPCm9Ixb$6f6!Z5T(}hdI5PUxWENT*!@AnqEB=CPBIEAB57_
zG&dHN1{EYQBKWXe>8Nz}TKm1j#FS7IpZS$veO#lmgJ}xXp@pGoBl-!((p$x|!{9qt
zF!I^{9o^;hDIz~oN>oQArL6tkSq|!Cthz@em9K59B<(+(9IrI#`Yr_Yx+&xCIlrOt
zqKEBD<x779-x})CN!B&Y1LlqBzi8WOest)<g4dL!^y0e%C68|UqvSRW2aSUtw)*h*
z>1ysYnKPPg_#;DriB^7&c=KXxNE}XyMA-?d6f$e)$^nlpYlz!FW!)?cJ9wvSA#ml~
zg-;AG7g>TrxZ}WQ4G*;8*5%D%l9jJ|^$qTms<-SQpzngScw>hW!Q##7h)S#A_+lw|
z=Q0=B^~d#B28HfE^`6vG>S|K)leG+jeRUeq0aV>cIPulbRSTgsi9|)$MIM+^e30QZ
zj*;m_<z3fjBbTovoBA@9El?$gAKbM29_h7uFL%VBva3I299H^qF?Yx}qtcI&e(2{H
z<Ag_~bs9NuGV$*re#=^Ia>KHjHHzhHYyxIwU8G4xCpp$YYg~U7YQ5s0`9suvnN`lb
zd*A2M4(kbH&k)$N^=Vt}TD3V}!yfy0x8_F6o*J;a7dDwO^5{lzSSK&Qf6nfJ*X!m5
zxz&wXV=tl=LcZD7He^T)zJ>H-Y$@}6oXIrvUJ-VZ>SEp!7L2##4Z03gD14nq!4w|;
zmtqiiA)-Y%<}kIh?mfr1ecbwk>x(<p&8VGXuc^|-#F_4}HC7QtuJSSfv;W6M`p=iz
z{U50NK)di4w;ujjNh0912Yxi6T=$Bmd&mWg7j`rK5U+e5+QHXy&r)G_DN{v$?z&{u
zp{*L*cg!Nmn1da)N=~f?@(y(pIjjHJDXYb>oh&fMZyTWr(KD8q!b{Pp7>1-a=P7RN
zZe|9hTf|>>k|GTajBd35CL3%qSzqLQ4{I=dIU!9x5xDT^_oLUN0%s8R&#_v_i?5TW
zo3FBbeA07XSQE)e%`WE^iQx8b;KfUdQpf<(5s_CA<6K&Oc%Js<R{mzt6G}E#$hA#h
zuFL15X|9iB872GC9NaXIX;kISw<82SPbT4WILU|GK3wYN1MlB+s^v^k?eXOAKf^3m
zF2uRGo;(`7XaH(Qu%Qptx=Ligo9F7oiVyQM(tNudZe0{7MVlK{e|9v7Im}DwM`4aJ
z4H0w|8XaYax&%2h;fVp12VTC|cN%fMdd^X;XHB!#nsPQP+@e$DR;Jypyid3}Dstxf
zW!A5n5z_8y2*KU_dGK17NdaBW8`81FnlHWUV{yh*Xkx&n7<cG)2o>c$^Ip^sdfr_f
zvc=I4A8V>^v;U0fa`n~-;2_!N%BEV9&eOykqjJB&yh2OA#Qgq?YqrLZ{r`mC|8FC@
zF><sRwBA%gS2@g~fmQb5ptKYraB5W`l<8dc>NJitLdjqnXHtj%z(jCo;Ptm)jCVLa
z|9d4mkC7l~R{x4oGl+Ff&FEM)0HuaV>YtKG<JG;EPvuK5$IpcEu=$a-2$d@3*%z@#
z3Os|LjSWVNZ}Sgczrw^FLB#?3>LJR?ZAb9MIi6#2JZR<gvwCmk=GKnqefEAh=T-}@
zUHg_(x0T~48sA`D>3h4yEsm+dPEo6)<qTqO^`2m{>PSEGJl!pg{`uy!&l&L~8ZyZ<
z`c=BO-rlTc>4HO5OkU=)?`SI0g>r;4&&U_AW@6Q^ff;@QLfn{lGhIbYPA0R&$Hb#y
z@$Gj_Z@D{`YFIv*vK1a1R7>ycYEPEE);wBx%#hO!9Z!Nn*c@8{gWIvPyzH)-VO5Fx
zJs#gJvrkLO?eFF)*Qk!|qhrnn1LWWt)4URmxL*$%38(916?_sMxZ^VKWQq2YqpmIF
z)QbQMJ+xPGhH91Qb}DXR)*}FEAzL}s-gS2V*4xb#KYpOL-uVv(tMi&~Z2eM7%^Odx
z!S}rn-93YBf@AeA)($8gK5mO~39`NN#<UT*#6&8ofESK8UXZsi-V*79A~!Z)$3~w)
zaNAInRU5lK0D1j2i0m!n2Cmw$&_cmdb1JP6qP5?-uyaxnIt@`Z$QPPa7bTCuea4cM
zMRXCOR4vFx%Jua$_ffT{D=s}L8i<wzA~`a|ts?*&pugdlY@C9NH+`4E2zSF?8ceg1
z_0A2_1Pfzc$57QzadtU8nmjeYR8?@~oqg-QwkY;g)&ri7!Cg&qWtR2$Gm86U@WXz}
z7Z35iKcPDJGj^2^7<QjrPWsIwIuZb4KK_hU`R5tnKjm#4e}B+FZmbO(6fSv(V4Cv@
zEm}8;6rY1ZWD7OMt&Z7*1?<No?OMx)MN=cTOdHjKiMDBDv(>>zSg)7ptVm<{e7H#L
zO}3bMz~zvB;+*U36Kw_xV@V9%*DM}@QG-z)-$~T?%wjfXb$R`W*V-OGsKumyYEv#F
zs$6?pc>kWnqg1r}3whlOe?(A}QS}=ZS`AcP)=!1fcR4{RdrxeMzn-J|WSKUBE4S?2
zs&LGSOtz+|M(q<Zy~$zPRAPwIu$B>A-8=dCGL!s$OZXX~VdB0l=FeQ2Klq8}zjH&V
zfDD@7Q_+8{;QyGL@-@bZ5-!~ojsS_U{nQ|zx04)>Sjl|Ao0r&ANNPZrtyGSzus`3u
zy~bm4da>&}1MlsXC=ku-H%~NqYKaaot}mdju^IsWsqwIphG!%Jux6O;(d6>ctBP9%
zNYW9)x?>e#l<#vU)dGpPao(SK1@!Fi>d;lsYEU~lcc<a)eEP5Rtw?|0l-J*Rx2@^!
z?Oz#)AqFJT5HAuHMk_eHUhv|CZmJNc2_|O2Rx=YQGcUI0JqRd->NH=|r+MS{atsw-
zd2qKpQTPA<FHk}W7o}5}TTG*+d4?Vo<G*4Tcq!OA|MbO0ybbf3l&ErrmHcc$E7SOf
z0&=PT@|E$+GDn=S(i2g&1Vr4oI!Z!^Iz0@+TaZSdUC&9--mYF3R2$UHdN#7IUvu&v
zwRXUYdwH77=p%^BV~KNo42<3JUyDo4t+F58j$$kfviHbT(<a5BSH%OWPu)2U|485Y
zomZ0=Ho^d5gyY(l?rhw25^NLItaax4;LREzy6bxveJW5?-~qR`57q|WxDC4Xfn5?v
z|1Yh3W<Wkj<L?}kKZ8L2r#t&sfj@EJACGbfn6<1Lf}kXGVF(;A$Sf~H*R+Rk($?kn
zd3HVcabt|#yc-Izd|CZoN^W!Ue?DO&NJ#TdRH2t3jP&<cQko-)_m|n6WtmKi?B%o^
zl6lR<SiGDs0aWQ&$;~-YUSTF)ZLQZz!ZsYvJ}kgNY)~#!{UcQ=B8@X?Kj@pw(bh47
z6zf=F80=pkGJH%;g1NdKMm5}pG4H}G#!g98580K|3r?Jf1$K5F%clT>5dZ)8Gkz@}
z82+38b2<?4lvBZ#GZr)XIc(g?r9$2%u_u%1nv<Gr8v_v%bTb=#j0}XZ`N+?_Uv}Ta
zE>FCU?NrOVQ3}l?>A>F4S4M)~qM224A_}V{nW}5Og=s!2(||yE%3q3Zlx(DHG&7f#
zd05#q$#0sVU$^Fk6EIg_&O6sl{F(S<ej}^EruE{a*R*eQ#a8OMCe_1`A8fgI8t7Uz
zD4wwRB`bT`kYQ644cs5WyP9pMGvZq8u_FZ3p9w~}^*~6F|L>Hj|785f2d$!%Tv&>Y
zWoDu^NK`ga9tY$?W4B>kjU1Z_0csu#OZvCxZh$oFxw>lBRb_!N7_J(Ut_W7t$_Njy
z9-={8IXmr}nOos-lv^Q9X^1w4FKxrz;1vZDb1v#u_bqrjSk`rR(_~pj8Kd!*4#HbX
zjzx8@$aL_zK9%P$U6iwLjQE{mv}APNt}xZL?WIwa^@F;r<~9M0{o5D67lpe?L83DF
z_Frn3`z^0%jg2rg@b&xAt^5wISx^TudVZ&+{h8$TC;aqxhL18Opj24V{=ZXvYTtWH
zZZCbUm{N{h>@Ln<pL*J+y+#R$my>;&Hwi8(GRIdEdk-(G0p!&bSGJ8`hWv8F%pNKu
z^HN2A{C4L&KnjTAg7CwMKt-fZ5GE}o)&ge8ADg5fou<d@_~6PK74eCQjuMrsye?m_
z)PSMG(qjDlX#`@G0zSQ-D}j|C6<hF|`RwQI4+#`xQOjhkpiISbXY95n23Q-@eNZxi
zXIj(W^hECs$Xg+_SV#r?CX!D~{xR9<KdI0M41<7xW97_fGSan`w{0+#_si^H(}&g|
zv>6Pqv-iuByI~~;_COw#$QJml3m@B#Oc9?M2%>Kmqggo|zg9Z@k(ue~fm`;NG*t6F
zH!2?)k5KfEMkrH8JF~H5w1%YJAjjU5fE}XyX|*WP_5QkK@aCNDkfbdmq`2Rp(qM(C
zb1eT_(b7FkLnT6M1tDv?TDB*_rM!5z-_XEnH*Z2AphEMXW7z(6Tls$_7*$b8E?9DA
z%0uY}LUQyQW;9WKN4I9<T$Ch_ue2qf%He2)_}+TTWkbrDAkO^r`QZ){^4*3%Fl^ev
z?^mQ@uKIL&hzVS^n4H-Nu;aM*h+~WRu9q*e*5>1=zx2b{$AvK%^sG2N9q2+4gS|@s
zyg%vXvWr^{hD9U#3nzd6ce?|L)&$2=R*Z`zC48~ymvJ@bYsJw1kD=}7GI!@z#lEJ7
zJRp_pyNr5aI2j<`==2Y1bN~D`^^}1UO#J((rtt)ibtcqh)(WuzYlsOMmGhvHxH1aE
zXM8xmU41lp1VmRSr<+r{`QPHiUyuzLE-x<gKi7O*$2f7Msg~@^p*-;srtvn=?pGsE
z^<gIntO(vbAH#RU$u^_6Nz^0CoHD+2Htx)F>?tV^4=yi0O%g}H|3;h}hiUPMXQzL&
zGHJqr9dL!@wvRXKvd=JiT9IfgE_@L=Hz~9ZW!52#(-)>RR_rVs;z`jS2IW^@rHU!+
zh=AstYcEr0c9REfUp9K{ohdt4>#U0LjlXR^IqcK<x1_26;)Qi*z<qfS2jRros=26;
zqHxt&w1%KDSeyDK6WM8K@xvPCIU2OXNW)hKq!yL!j!RR_EAc+<Yj9ZdMA$*bRF21@
z93%a0<~yq-xqLCIK6oQ0to-FSkZpY$HX0eSM;)Bf;^sv)j=)vA{Wx!-h(@w5_|NM3
zAKYYXs0ouS5Y-!Xi#ZpG8sw`ofjWGN@mHeAu=&B=L~F0j@Vd?EsP~q0$6cV$^1873
z^>a5yKQRvkztpQ)T1SbV(c%VN)J%R{?jm&Z)GTkqFfnQsPpPs?M7eLS15rwmd~_Q*
z4(;DYd_VTo4wMToiUJIDAZzcB48VUqP5p-v{Db5*Q}WMhNcmKyK?t99HBl{T*<9uD
zm1dm$bAgsSsvk}|GA;2nXW<=>{evvJZy9FPQ+54v2(oXD9Y63xT23wzpI~eoD7)TJ
z^d6lBCmtnKx--35VJ#!VLtzl19CB<i!KT=zc$xfrRED}(fW7@6p>%)b{wbAP#+(0^
zYyAC?=g&6cuMtnJ$2yy-WaL3J*Du#W-)c7aO>JeC7F!|NhmnA%5jRstj$lxh{i!rY
zD$Jn3ylY#}K5!e*1a3Xdt)%8*_2!rbycP#lnSAN)64^KUn0O)iI>S>2aO(*}KIG$F
zN{>Fx5l`*B_qU%+D77CitN}*>{BQY`e_LGqxx@I)y8XQq%g+L`-2A$v|B@#?zkSxm
zGmE-)r5z-hAA@d{DplQTQH8A&1#l8mC?MIQ#o9az!+>i(VPv_M!$U21zEW}L7Si?$
zIsEf6Z61J_w%&C#WRuPjE6i$hu*yo^-BoLmzwWU21>J|Zs|_&T%YWpjG^UpU|7q@j
z^Pf)a#zm6lK1n?kQ7%^Y2WH6f1Ves6H>3|YPjmn!haB<F_{gJHOa#YD7xt#jV4<AR
z2RL2clmL-ncpq=EWbLEwb3DWF*Y6zqCEvgp&)cLq$fQwIJhRs;A9LQh?jbsPT=hr#
z<R7VU|MJrKSF!p_kF$*XQUtXrUM!xW_xB4$^&r~-f5}2ifU3-nFhvM4X8>=u$7i>v
z+P3PYmub`gOLry8<AA9NZ%!eFRb_@$%PcJM&z_gNrp$7piTrkswEQX=nbF6t`gCM3
zMH;&GqnyI|%+JRP|Mmg<=P4rah!x%eqVF=yGm&kp4tg2V^l}<s&PS;-Xj*@tL4|U{
zzw+x1bD*rgz5}g9s~F0W&;&_vd}nvVDmPJr$u3PsTXiC+qwbh!6pv{w8VA|1h0))H
ztY;G~1;?qBFX&S;#Td}pdHzv6^8U?l_Gc2`-~D2BOg5B06b&+VD)o-nTIx;#65)lG
z<}XZigx6L--<6I!-Wxnmm^LeAbt+!0fSlC$-)d!0@#%%~r3SFi-Via7FVKh=aJ4Z=
z$p_o$Xpc`r!`c0+yt%N8HkG4^*Cv*Xf11@FI{lG-3XEIj0BOs=7`6YID*4x-9iR7d
zBt2<n4-b0Kr(AJB+P0q&IbXleCt#6n1LY54P%;$BlK;AkjjZ8h$j0w1`UC_oH8*AX
z!+O2J!IX1Mj%MK~Hw0I2qSDp;NIO#a2i6+FPqbm&$&E&m)NjB2m>?Km*iHsg7J=N&
zKcW-=st|t^i{H%p|A}7wtpNK?;V$YSolLR@JRX7b!{)rKr&NHbLqm2$iOk9+Yfq`I
z^2-1K!&9T^&fnvfq`OTa@aW->n9hxAKx=Hbr=la{^vIO~UDc?Te?c;|f>yE|txb2%
zaHh)tHRvq`Hn8-vqfOS;TdRtF*}!4;XFTHX^v{1={SOMs;myhDH(O&q*{qIkd7~-U
zjOIR<hZktuzKGT?qqmWfpg1#t^O%4;k*?Gu;Cr1mZtgwp9h_@mHwOI^Dt-|gQV9+6
z&!-?l`&p>f(Brsau4s>;kElLB>TcV&fm#zxAO|Ajt)qA1DT`F(*>sAaxE9?XF?v2E
zPq{9e>{6;ZWGEGUq7z77qm}f;jHKc+CqnMcBP8A)T2}uO#3^`h14PpQ_-8iv%PL@N
zh|qq`ZP1kZOo4}6Fpm31h>mnVeGqQ<`Fs^o9Py*sc6&2U!(*X}XskhjMgF+S)luNi
zOPZ<S?=*(;+P+62V$1FGi<y7UpE~kH_}21sSQt2bzSt9xMmjVp1Sp9gN*X#J@^mlN
zQ#Z$+0*E*@;eHMeMzTlp2aF?*QpJR*t?7pR$P*)ELKt^x0O_0R;7_<=TC~D_;AsAL
zZ`VImq*fy!Nf@DsmXsvV$+}nR%N84U$#tR`Ln6U@z0llpXuMOOuPyW@I97lrF4&OF
zw1r`<dY-5hdQ(valf(wE`b77AlQ;f;47dL;6<}abNBC+PH@u_#$dey8NP|xHze4zO
z7UDrL%>6OUH}R1PFsw699>XbeAT5rfz~RJlok6vAx|MN=D=2pj)W#RxmZV>%*{Tqr
zeB2`8ti*#aKIaRWEe={N{rBJN@2bY{Zw27R1?3A1iY)F43TuPHf+guv4BowAeei^c
zwj;w9OG%K!vFfZF4-6d$i|_aYfn*HYd}Wq+xOU(xPZE~X=y^c?qPT7jO4j6Ndv+ze
zd@B5)!-da(!=$Z8kY}v=%EBtU{!Pf%em0Zx$s^>pi@M}mV_c-a9F7M)s}@4WiJBie
zp+S=~n(ru5u-w(Ca1@WnIAQI}`n8N}Pe~j6DuEO2zj}=j<XFd>cYw6qX_=r9HR<JR
zJolai_xCH?sC<vKqtI41;pem4%E*2QNw0@?2;>_W;AL*|A*MCww9A>W4Xgw6?@-P}
zp6qRCuoqy;%NN`Z-sfp}^JGwy_mP*-?5;Lb@(USQ@vT$wxl+)hIg-t!rMT;FBRz=5
zu(jhGY^4Y19%P>ACzhRF!_}QpcjYW``<y;$L*k0B-Wu$){$-J{QYvKAXZRoVHve$+
zY~ZQT!L5M=7;{7a(<FF{C7Z^$b2da=XF7Gaix&OF5Q`8Ax<0L7AFTx`5^%)v?lxa{
zQmcK=8gIKUq(-mlu_?q9>BVxbQZq+>xdhcwPffs^b4w2MdT;WyEF{^$oJybE2kT<)
zY@3{uO@LwD-e3|nnoOaW9k}XPCCb1-*z<ziqqXsbXvX|ZVN#Ym;nIQztA%LdmajC2
z{(tm+{;pTsz=O~iF?{p~DlC#)HHBw+R09RB3gJ(w(Z*09I`}}HPChB?j_?HwKo9kS
zbhASplv%^I>C~w7Y^F0d<IpA-h<6qL<;ri{NZz#yRc$+&(cWfSd#L(#UQ_=em1siV
zD@EbWF(y)Jo*ls*<2H{(P-@ip`xE5?(xRm6ZRB%nO=x^yHuJDtzYZGIaxqx2Pj!gm
zmRUQ~HSEut!}9;~iCbRDVP&~QlY*equ6~_xuDKMc2MH2uj*exm{?J56n`=`=8dGHV
zffFweL=$~(4D>j($_1o|^BGW{hGh|0FTWu@ApRk^DETbd#*UvrQQ+H`H_H<kT}<1P
zeWquP6raVya6_QVV3TI6S8gIVoL%2lw0r=;>tCccv5IpRWjuaGvp@MddAMPI7B`;a
z#>fluZou7{)=|mm^Qa>Q&Op!quF-E8Q(_6OsSf{enPrHLDOrlU#GFRbnY$SBPVK-!
z3k^y?jumXP56om<&-plx+^T}09zesb_MQW{?n_;VU=7({K&9rm+wUtk`x+YbGYQP?
zea!1=@^DPULcnlK5yI?r#a{B!WFta{^*wGH-N7u37m`$YC6{K$)-U8bQp$K2FTL~X
ze&5XL(OP>4ekFEUTK$$z)^+y)mN~j>^<X>-6Kos22yOXX#(eR=JN<!U>FU>UjfZ?G
zPG%uq_*U3uGwNb5lFiU|P;hKkfxaze*m4gm_c=B_K()O<4WvJyU3lfUO;H+|fw^mT
z!HZ&mZiu1fNLRVSgDk$I5+w42>3~fS5^g=&MSP4PvP?G43Irk;d8yIkmmHr($*Eql
z2o@$4x37{Ji(E3g(Q<HvdD6OkAJ!wsVMoPQjq0?)`lWe2`g$u^NHqI2<3Bo@{-K0(
z+O0i&ZFMM_Xq*0OBYKLvX0I{UY}Xq(;>+6X1a;?Y3DsUxuYN+BT_VCHdL_oI!>;Y~
z88aX<*O1?$Lxr8cIrZv*=)#vFeqrEPj<Lj__Ra}{%;M`2My#=|9tyn?A%T}FB;0&=
z4{n3^cz|`k|D(VXUyh>_zu@+EG^(gKI}5pFRN6GLf48Hy(!mt0+n&b;moAB5ml4h?
zIJ_}qup;|0?Dxg~f9r+L-S6>wT;<r@^6I>pJ&s?V_%nyTcxTRF^?QARD-DOoS|qE)
zEr6%HZ@D|4p!*bGc)zsi(4lbJzD8&vCXUcq3du6YJ}N>ib#EQbQ<GyAMAk#<5B+>F
zr|MOr$_DyBKbt~KOOsxcvZf2`7QE2*TdiJWRCA!+>NxXu$6Fbz-BhwSkbfK3q0n{2
z?PB(5U(u1lZgGDFv>(8^($$lyk{0tWb6&(H`1;{J07tJ@dj5b-ng@uM-kb<fAM`$p
zbp6Q#qh5=t*yA=t5VK4?W#ZkZqIRM)c~iw`2l$q_YY(Y`aS};9u&O9w58O%j<o&X!
z4M3szvTsZCjD1LHdarM{8*Dgr$1Jl^tgyuK?*Ad}E5oX4w{{hflvsiwA&XFu?uNB!
zP+E`<k?!u2M(Kt{gHjSo_mb`|r8^}Twa7Dh-*=zu`p%DY_TJ~uoO1C4=X~Zf#(3^A
z?)w%lt<N9xH%0qEEgphUPBuQKYqt<q>;Eng(d$b52rE<P2xY(d+_$O9RsuaKLgu`i
z2_QM1H;GK-Uk<8Ri>L}AqSB|E_WCOUBe~5Lj>(dO;w@B`@a%1$twbojUH<$_C^q*2
zJZnAmVAFDu<+icy#Icv{7;qYE&KeB0iT5T%YgHVvwzQnt<8v^$U<a2k^8W35kIj(B
z(0(P)rkZ~}o^>14jt`~Cx?ZDqXnjP&--GBb8o4;V_7HOYWB<O`Y4Iq^En6uI!W!>~
z>^JConXDT?qVI0F$#c4zh!K(N91*P}cXLsV&W+eUn$mtcjMzA5u>CqK>NMO3tb=dX
z^*U7koX0%GIOf#{%JtKUKPo=Q;^_GfwD|8A#6+1H5O-+UnR#kkeS*_mB<>K}9E)O{
z$<HQviI(-{Yd$f2Y^7e{8Vv6F1TZCW{od^X6=z`ez{Oi4Uj#o7?K^l^iJuW6wIXe~
zBb!2XAjk)Xt{|J0@%bbvi+~v`g$H9yKkfGjq;TT7_z1;EV{<>-^HmO?7{E}N^9u>h
zKE4b^@EX9p&EQ2l!ttgqwqUkUM#X;l=~9iiU4IBhs5I_rYMXVthdpD`2yPgs*$CZK
zhscaHC1=-1oqr)4*oa2CoyD5E)N_Edq@JqnojryD2<}i|{v9BKt5%s=oQq#of(~nu
zmqR)oo6hK6{v<Jn9R5VrY;-N1P0Bi6e1vSWn9(~9Z~3N{S)h~^1xWQbHd#Qdo7cz_
z4V$yT9&4cU`QpVMpKS?nL9~+Gzj)TEx@=q~+&3}q_ij5JrUTEsFWSLfIt$0d7@trS
z?leg6PzKVlW3k7_m7tLazUPe>fB<k5;2ltVZy3U1K0<X3Xkz@7fgFQ$J=PO7>mApJ
z5T8l)9hh)$$vS-+x{3-icSKd8L4xYtPsbCXN`Ecw(bH8DcCV2WD3ugb*R*N^+)QgB
zRQvL~Nq^EFI;<1XR2*C_EvWx925`5_5}I^&(78FEd(ReYWm+_&z&98Y<uaV=p<rQ}
z0{ejem~EV~>hXv>+h?KVvMEg+HR0*2@~w6c#Ks=l;MC<vU>x&qdk<%~y#mt;OCyxB
z-KSV&;_dh3$f;s6M{Y~&cvN>jvfZgoko*N#E@Cm-{g65gEA42>R~G$T`mJ0_$zG_r
zy<bP5CoP%e_Y#4yLyyNR2uRS@nE^F$AnE8Y0Ok`3EZRL6!eMyF`iUVMo9#qAiwe7m
zaKO8fFVVm!1)H1C`q$F6_DSuz$HKi22H+;w%Xj%45IBPJULJnPWYJbqIr2q7Vm!c?
zWHG`Nsra+`-oQ+;ZxXmU72kjrNo$b#tvM?k!oA0ed})x0;DIfeWh_f%)Y7Xq{bLOg
z!<p;%sYto!B+v2Ybb6J|4;daP)w*2Ua>^3%T*!6PKnwAxE3FL;FInHS=v(i-=G6M7
zY>9p|nUe4k+7SQhV~kKZa?nn<Zs;hrsjiG95FRyk<7=}^+cj;@s`>rr&HA8$LBF>y
z1j3^8?OMQBZ(}^q1=)N45a3kRHw|C=Z2D|X--HkF4UV=nl^g3%wrSHp2al!5jC~H2
z!llaFh~BX)KvFUZcJM?-`fDn1qUHyCC|jT*z;{L4#`UBX+`~q>8}Wkb1W9tU3neVa
zzQ>!Mck4cd(JD8c6Xy>tO7Tf`xPCHFCh;sp??(67q>}T8&dwBEB|s^>or>z&`-j|4
zEGBwxKqFPpw~M<8bderFw_B-B7!6B|vy56(@Yf$XRa{eWfGJpVj%0%3^ZSmpn7Lc1
z>_(JClS`yk#f~KhPs6u-&{1}gb79itBwwup@`}~gOBMMJ=TV~B<j0G}^j!GO5g=sU
zvseqicdNU7m9%LIVb07mQHq7zvBCESuSSy1?hGv<){P10$Jq?<K8*zf-+8+<e7EH5
z<@*q9ka@l)0q|rE)+83i#SS)o{?Bzx(SP6appd;#VCP;t2?JI;?@Hcs@|CTgtdGWg
z1LqUp>8FccCEox)WU`;b_hNRxKU~AYf%jG)7twbY`02iv+l?6-doLMP#U^ny<gye`
zpNjx?OGo=|aB?M)(*!12V1sQ_+{G0EYL-ujk1Ph#EZ;qvY%sv(&2~m25Z&))_p#{w
zs@5d-%FD~${9dK-+RXHej)OJkhC-`{oSQ7sF{6VL-}X53NXQd)+7{yu=A|fvirGgn
z22t>cJ$3#XB-fT6$eOu&xz3RpwrTl2MKmUnhkk~H4!p-;(7U#{H(T52_?1MKk{kGA
zLI%}|WK`X!dNv_elZU2(JD9Ab(QO|t1J&aI84AmvHxD1GvMU<!Z%zG3o2K#!`R=|C
zrzUlLq3lIqKVud9SePAEVLG7(UwM4?bUGrGQtVLp*>;#wEX4Qan8GkZ(3@CZgCTcO
zBFQITVOVrL`HSHvi%}fMWT>_k@~W(cB|hsBjB!|dhmYP*_tyRHDrUGG5DJ16*05wT
z>orXHr0lHfBRTxiq3Y__k$YI`C;s=&t9|q>!2vL)o#?{jcSo99N+*U^3Vt6|zcD15
zEjp{;9=DXZM*U`uk4`uzX4C(3YrmE@1jE;{JC1KjHXF{|yS4vvpmrdEEU)z5T4jVS
z7VPeL0jX}WPqgw8GFmH3;E`=027cGD?-`e_+q}zuIR|})^OKhI<^!P3XVSNVAC>f3
zX4Cs9+@Rwee&;g{r7<W}FVXY|+<OKlV{QQfxp@{q3a{?Oaish6N6G4x0HLR_)7|Mt
z#3R_{Zu#BMDg3+v5>u9^p-G&6<z8IRcUQ+;Ic&cw`HtLdvE^G$mHhT7fkFYt16L(>
zq{xAU)@bN$G?}Gfqs2I#TNJcriw&NEZ<W?*yK?h3*a4KZ!Y*1buF6mSX6e}&EG1-5
zFn=I=&5<zoO{xW*wYs}}a7^b(w=mlZlC)&M(F}H;bDZ6F3{gw*el;=?TGSVMZph=e
z-j3GqQ0UGmT#&98?%T^OF0dpQ*TNt~5IveLJZ!Y`IiJ;n1}y{+^1P03RD+FKQV>D&
zB5xt&qzX{k=bn7uuu&OGLpcHN%{CpOm`I!T(c3c>ywvOfcz@=5*7-S<FL_i=XX0YA
zNs@Ih=R^=5=GsJlY&JixeKcm1w#izAzdZ`6NtC(a8hJY_*!?v$2?}r288w@%IZZ1(
zW*!0Tl2#S7)^8HebIz_1UIE+s2*A<ECnf2&U2R?KKsZ>#>uwG46}o7h+#5X%xnkZ2
z{Iljoxw7sC?7VohNKQlF<D#cx3^w6^%dkpH_5N~R*-~USl#b*anP@mRsdk$E@VNkn
z^tpUaD&Y9#A2XNZ)OVx?K6sA;ELra}bYnESkGId&?4oP-_LVuK>W;g?dt>bJnxU(U
zZs<Nb%G>-Qa31)&alfH|()AWd)vn;K+Zm*Lm0Dey-)4qegE*)6g?+>`j<(-1)DnjW
zcee-bv$G2!p9tw2Mr}@$op}(4`8GE^io3-g6E#Vt)z`>Q7r2_!lEZ3O{nxWAr|QKt
z=tB8s`Dlj1Ipa9j$?kEi!TGH+r3zHTNB14V@b=>D!hZX0=P7d%tJcuEBWqS?#n%dx
z!D*c~{J58zO~wz2S#>@?>d)-V{W`#fvtcxOO@A~t7kW7MM@y4u6H8dgjxve$gLil(
zH_K}1z}w8#@n2-$lDb4vq4Wn5Kqb${T~T>k2n=|7lk5)<UX|1}J8shm*E&<*d`~%V
zd>o!sx^@E%ec7-69_yyk87q3zA>6|+{Mlme%g4zFNz84Pa+`;BotpAc7K4+=Q&0iQ
zQ~U^VB+QPQ5+GN?^f4c$N?cNt7Pr7CxvywSw|6c1+hM+8b(B$_&!^^H7?YqLY{Zbp
zXHM$E!|o(Toh!fHZF{Zt<xJ8bW-wy5b*?o;?^Pj>deixJ15_F^^}v*IxG(&+?)pej
z(St54`UW{a--Q1eas(N93n8Myno1;Tjy~?*)jW{!%Y;k2)iw<S`jNmd4>+q2N4!|F
zhJj!Hx>u-yU7!3iimpGvjdW!*z!FCI4qQ!n*EM=Ti|7>Ru}-=NynVlIw11!wnO`GB
z${&~|npLaC3}3fcsz(?pB|!^jDr1nVClJ-n2>*?iM8GHL$;5U!WuNzD&*dukE({PN
zPAmKP3c&(+y`2sNqtLiJwEQQcufk3?#8kSyw<N%hfu~y-`i8!6^xw+<gyw*pJ=8@(
z@s#AbpFCuoNq$fts16sGOYMA^7Opq=Rsm0y?zDvyx?1727>T)Km{plWw*~go$%x()
z5`H}y@S?WHI6l0bCm#;zFM>mXc(y!K4%WA^THLE$+=rA@?GI140Lz}=nCAU5-ryLL
zU{IZaUv1D2Xok15xI{pq2bH<&<&}JC+!g?C$)0Kf7^D!=8|@f>wIZWa$XD=uz(uzh
zy&Zqh?t?!U?M1I-cJV^k{jM~{vHz?4#Y}99Tc^Q~wXwzJ-u~ODB8}*-n_B%~tW~>K
z)bqtajQm6WV%^G0X5gcPg3D~yy^yNeWzZh;*{kncmAQlI&S1(f3aP>%!u+A1KC0lZ
zo2p*jEMAieM1R$%b3@J>&e-WJ$2)HMO#Z-4&a6crRJwUzmiz^84$9^hyy2)DvZ-iK
zA=e4Ypu&Bspmy|Z;S+WgQ}7oF%;d*P=;OM63Y1)H$_ZFm58L}`qdRlSm0BaPYL_xC
zD(wr#T6Yzi{V3Y^JJj^;`l_~5IDP1DL~!vBk?TGaWZzux7Opvte_R(H|2gl?x+OK&
zX*nIvErm1a%{M6(r!JN+zW4dX#Kt(VISOcmHX!Yjk$|_-#nEa;FZ4LEZ4K%tuXJW;
zHF)*bM7tVxn^DS+Wb;d}y+)d*+Nt?A<KJ9KK~8Qt@2ah(z?atrKvY=lthr?OQ#_ix
zu1UI7xJ3^8*=E;An`Hk6ESS~gilPt!aD(d=64t=t?f3_3eoq0@$Ck=Uz=P!BXHvD8
zEq|44hM0->B7h0dg<{Wy10Z`{g7146*yx{shBRuj(6JaXFjg~qWj^_~8t*dv0Pz1b
zN_SuRv|pZN`Q}^xbI#M#@sZ04Z%rMA8_~&hRXK3h)YGgslgr&oU*VJFw>F!@r|LG1
zfcwb3)DCi|+~(5!t&Qic*3+ByT7+Tg4;G!;_p3>R-0QK)-;H*qLu}U1c&+EFC}%BZ
zd2KpE2Ia}i;d{tOzs@cqe>rx#eK$Js8sWd@pUCn>shMmB>4ZFwwjuF;JWD_Ep?_st
z6K;dSYhxlDY~`B2?d!7u2v)uuQlpz-E31ag4R|edIPqD}e8$yreZQ){(7(a*tHncP
zhp)UnN}+0y{Ci}}tT<rp;>SChoMxaMvZpj3^1L=H;1Z`LECggHR==;T7IH?XdnPSx
z{NQaQvyi>ti-KnK?d+Sn`B{I{9v=`3=^Q?p2QR7MgCjuL`X(;)yGuKr^+J#HG2e&A
z#!wqgC7nhGoGH7xpgkR4_UGh3JgjE>oo)etV5HHSoI&PkHxKfi;X;<jB2$yQuN&V1
zm)|1N1<sOHv>^e0%Q$?jx}dQTdTas7IwShFQMpp%UFZhOPZ+rx-4yfFRXPU>p+z#w
zv!ZBkgUqL_P|wA6AG?}H;Ji~dcBQpw-6DMDlXQ1xdmP>Oy+y}t5Vk(mJS;dK`q@g!
zLu+|w=ADj?)juDP@JNxT!Jx|GdCPgdis8G-TcFgLFSSDZ?bd?GxR~37gu-w3?Dmal
z@ME+u>FQwg2*Rw3j-mX@na$|Q(C?5nv|O@SbvF47+NWI2L!hl<@W&<fd=#2j>YjH%
z<{OR_{I-EPQ?Tko24H3u20AQ$wh&gkH_7Pa$~LAB-~DruPQ`n>LrL%hV4on=2J7{W
zFIy)OtX<24S*kV${lwG}5-{{1mX4?k=8VyzS8Y$8@Lebci6X3RH9xQQej}Ja)C0Vq
zWz7ZAR0eO+3U4uYFqK?xVx~C$+_%g*AGcn1hP|1^ncme!obUC=E+!$9FfX4e$9H;z
zV=Om@$31_V2n_5Ct<`KqPSjK1fOyip<I$LPPRT)Ko)n2@Wnv~{3}JvEdK>~8;=G3S
zbE)0`p3Q%g3|c(*`r>v=TnCsTTH-k?lUgig!60GiXW~nu2<0z9%B|*tXSA~PxYA8W
zk2;>_|9C=On(&dGN$xAZlxj+yZ{BOQp|N%8pFYUuy}{#i{XsQf#CpB<UET`UtFF-u
zmV;1{z2xEJbuGwX7X9wlfulxE^L*`DZ>VoVpLf>ob-l>%&{7_x!o2TEa}K`u0n-5=
zLa(3h?iJfkv|aI2#G<8d7SjFz(5m0QMT2>29N>U!h@#tgx;_Ekfd}`2<78|M|HmzM
zDXib`cGIuxYVppumdLqy>ToF#-}S6vx_JoqXP)VU+d$qdmBb>urXy+ZJu8FuXx|fB
z4X$GQ=LNf45{jx>=C(u7JLK@bN%o;P#e%n3(b?N~tw-Z&g>1?#OXl#XNe{^WwWn3d
zOg%&qoPT9!0qEu<`GHk0OM;uhGtYA}<1^ux#9QrZejc-YkTrq*CQNhQ!N8&-;ju4k
z8>7y&X_hAw$XaSZdsNmE-zD(Zi)yqltr*$DDI?t1Q~=4&_QmA;LuyRcTOvTZMYLhP
zzU@?S_*%pZ;W2wK?Fy8RenBdlkf<i0Mk|9JY_(JtNpO!b_j8>yFcYLCy8(?B>}{A?
z3@ZTFt`uRGU2rqfM)z9M%Cunh3V_AEdr4yEt~6da*@-szvc_#vigEh0zqxp<2z_Wc
z9~<Q2zwr>$3ZKrLGN7sk8pm-U<v45*dwfylz$Et8MCZKYYW0**JApmJ`)=FA;)Sk4
zb?)fL@bgKs{zWeyrX(|)t5u;@zOUcX;EN*jm4G-HJ`C8utAn9laD%;ZyWD$UWR=m-
z3U6VgOQjwiblXc}-d~Mt^JeX4Ga9iBm}(5L9VcGued~;s<~#T8yBY~a+EnUo_`DoZ
zlZv`qI9*;agACZ!PHGEk0D9CvwiV6*O9h=6&J>9HB(r#)Q4v<A{38j!;Ckdn+Pw3N
z3h4DRJ{~0kI@dW+d;{pe+;H<WzxK_t6u;Z_yT!=5oG6%k0QF*`n$*X}9yjn5G2aPQ
z-W~?W^MM!C5u57aa=VugMz}FA!?zbiD+{#cAQMgr!)UO-f0mH`80*lUntyhfUF5-$
z#_ebPE9`Iq4ve~^z2IrC0I6U}AC0stle!a#72&fNlCal-#he}lfV!!lUk#&u)oL{O
z7SD1Zgk**fUnzt08NYd<4mQ1Hz6y}%Gr_sZhAl`gSL6?(4baC936gKSEJ}NJ3B^3A
z2yA!=z=~;+%wDU}1R|~iWPHpLMZPR;eZ-vJsL3ybzM0XcI}ig(YqTZcpej>8pyMS4
zOfG#X9u>X=Y)>P^FXmQd+vy-_Px>sv(DfTX5!V+PU%YkQZ4~9IJoDdJ7zoV24co91
zJx21}QC!z7nk6CI1M?RiwxGW_0LJYQdM@z>CmcIEuP(?40FfmHkI`D?p1WmPc=X0_
zbJ{g^*7nt#b%DFkye9l+z^fNH)+zy8|6o&NCiL_8Ujj@U`d{SBHwsz~LJu_ZTv`jz
zG2)6Y=lPh%i5`%hh>fSdLR<#q-4Ya6djm(8eb{_)zS?Ht_efH%uGZyypV>ydzFRio
zNzUdH?d5S7ZInUFDat(tN_d&`nwaXWey5XX@v(2um4uv##?6-Jh?-x3C8UJT5o1}h
zn;;2b`^Y{JUwuQ&s+y99!)ecxl!G%~U;G@u?s;xU53ElIRO*FtbZQ{+){Vcl5L^mf
zO7bBf<XjcInKLgU<KZdVR`|FR?%c5~o7q6zTT}hOAFD}!;6egcmsHOu$6Q{3!=%t4
z5%SS3dC!Tj&6!ipa0bkqX8F7l$$AqWCd0EBxRI_FZ!md0m?=cU1-Q6;{gL^-fJ}!V
z+anXTChU7YLKEU4(wWRina^X`Pz6=!6CL;b!Zs)^xGDWY6`i(z9YYq7t%)0tMh&5S
zoQ@=Zq2UKQ@}Z~oIo6i!F97sDP}XDKzgAXw+tT#pGGQAVRzs^>OVG>3N0q_RH)$_@
zev>7{mkNoSuERQl351dHa@lz5%hMh}&0EZn3Yjs`H6|OAmN<vzz+(@lEO-J5BD<(5
z1<)u|gfs!I01TG(c42_@9V((HWnf+ueI;z}3HDqc?hQ?IO$MsS60wikSE296S$o09
zId(bO=ZFPw^BFx)vnyh3vgTSIKg|8Y&kn%&6tWNCTRvh4WI$ZOFDz!+o|d(K<X$P+
z=$sIR`m$3sKjsnN!}q(8z(~3Tjd=u!;?B{szayNvN3QC&a>Lo3x0LxlMx&!NtbAsg
zdmxC+N4tedF$|=wmHz+?#;4=#fQ`z|Gh*GiM!rt!pAV6reGDUQG@<Hx;Q!@Pzac-l
zFFfEi4@M-D?(pT=Bg@R!J&sqjwoc6ha%!$6(oGk{FOqN5c7?)Vfv{3L_3-WdHGWK$
zz1FgP)f;@HQ!xPOJsi4|zFvFx^SJltj2Qu_g;8jVx_AQ7a4BZynLvk<@-c<Mp6aU~
z$MSIM*q|2yCU%o}$RqB_hW)wCpE0|f3w22ip(*D)UJJqlDW?Q|%p|X06-KG7aH#aT
zw3%+LZ2X8&QZUlf&gG07<CWAgkgU37c!98$68wq|c#(*CpM44oq5Nb%;Se_Lhu;o~
zKFNer`K(~SU=t3h8~gb2{6)%Sb=;S>&>Wb`X((X-*2;TOUxE6o2D=wfxXJWNuoocd
zo!TUA$*@pX)mvDl?7L}67}i1VGO}IfS5k5-c&Y!z;V8jZ_0RYH2K>ka-VyneHnbBs
z;{6MP1Tc;g^~~@VJ^*FzOg4+O{&Q!akc|9pazEye!+pA^%RHR1*ABQ$ll@CQ`{MH9
zQL4i1rN2swJWm2lMm0MX?4nM)iY3Y;8e_@mlF4_PqZUGJY-t&Hb2f`F^W25#FDTPP
zP@{CE+ic(6Zj!T6DS5GA6&MKs7`k^fRwU-fBVlr4DX0qCsY{xM=@#`Mi(j%aMcKJC
zLOdChA32Mu?LAm<oa8#IgONwQOrYqS!n6aBC7oP$2SFJ=`GW~u(OlE29i2h}Dy)n|
zn1nK={3^#XY4^kAJ97405<uk&b$O)`Gd;XVKg|Q5I?^L;`c0~NeoWV~6rd)XX#U2H
z>cWZc)j9`9zt&vG$-4?KBKFY7@t_r5KdDx(1FN$pWD2s!=+`I#4K~uD&*{0lkY=Kt
zd-c3%{5EC{?e8jr#DX%3?T{?0XFfD<7u-Cx?K)7esGoU04q$4DI9$ewDcNjJ;8{|i
z3RT@s3TCWCo8^9CE=_9Q;G1|-^g!+0Uv@Yb!ZnXNo(;wOQxdEm$u)lQrf;*<`}P=|
zj}9wU{c9BgN}1q0#9#BU|C0Ur9~=c8Wk}w3sN_b@utY|_QcMx?TPQ?OdeDl;x{|Z(
z{X}<;lbULgX|2=dkjo&=B((w9p3#~NOOX_y!NW64UV-<|k=(!>uuU)Dt<;qI$Qs^l
za%()QfLQYJO9cn%plZ(ci32vpM|@R51o=KY;f0V-DzswXDMxqc382obV{(wts(t<r
zo=}<+A?#&~<~iA-<)uIsQb)9anTME<V8xvj@^vKwm>lPw)jptacYtsnmB^-^Y?tq|
zBa<wknoA~F1ZFzV?*Q(J;Fab3PUH9E*x81tV&(oL{oF!_W6cEYsrvJ(9aZXaAFnFC
z#vKCsxL0rjw$*KN9&lojfl3U**2l}c=dM^=qg=vOEPsWG=504Pb{;?FuuYQD;L=`*
za9<Uc#Icp;JTiVUeM8a~e)~Le91x!S_2XwD-vbSA^lcwM>FbkfOpPJ&DhJY)uED~Q
zf`{PHZzc+)S-TRwCr=>HKNin%DWIrK3n%aTKA;o(F{W#n^0T4Cy~1?Bd!ZN?7`u1;
zn{ETxq!;`>h>QG(4fsE)Gk}CQ_fwRc6!X+Y)_o@J+v}t@7hg!!^=0PTYRqfFseia8
z6};2t4J}>+LIh#K-p{xnK3xre7Mfj9GSeanEiCpgC$P>f5wMUQnX7ry8_q>A*oZgU
z0jF)+fr1`C{}>dL`k3-#;XXS?>b<$Jxn;s{0NVoET-vkbz<_TVIKsJORJQz@0`Q0n
zUh89{Z9ZHDW*qq9V!LTb-(WuHeQ-o-C+U3zZ@A4m(u0d}o1m@8u!?eSdVY~RkZRbW
zez8kJ_4_u!s*lWy3?D@^IEhEjBsHh#z-a4zYIeO5E)9=ih7)RTuf+H58L-jb`l7Uj
zB`%|qZjk-&8(BUU88r*ZxrvX??M||penh=Z<go*VQuc_~m|aac%OkX)Xm?a#7)z4<
z7Cr9jHel_OTnyxnY7AhVQKG0l^bXD@t=q-<#bZyfJ=yx>(yt-b5-Z$aj;nRdzp@YR
zD^!&^S+bMD8J&pq&3j=B^EcJF-$-QUH41;TMDEoC?o|Q*=CK08S=M5m1NmUWo1cZE
z=6a95x<^2V(xOfcwa@d}S$w>a2p@Ndyt<l0))3b}<cgBg?CndzCyKg*so%EyVFr$a
ztr3>I<JtAS`Ic-R@~MWiOfMyy6dMk0f$y5Q6-iG;>G2|<A@GNfBs3^5N=on#=KCe9
zk24Tkg7h!@1TRqWMLD;sWv$L61l|2Qc!EX@WoSsRhGjPv{VzJ}0b`p$<DFDlJAfA3
z4-GsisRc<!)7U@J-^JTJk^uzpMO78-^beFJ&6-JjUxFnZuap2Q@=x}+8JPMhf#&^W
zUlfF5l6GPfc0m{lWo)K-$0A~7^3N0i?U%J718Pbi;qfkRJizHZ##fg!G7k^6?3Q_~
zn^n>SwZE&OQ#PxGbh{KTS&U=>ICy<AfusCuuec*UF~;Pek!pc=%vnL2*-Qi^b>Dt}
zqtY0D{h)~KR}ADGToswVtrA&u>DTvRQvZoGDONjHhp-Y)Dj8MJq;|vCzQ!e0_#;Zm
z%EgZwZy|Q;<=^g!W}Z*~qNLu(zQ0F%y*>V>xcpBx;iGjfn6Ioc<6};EnGa+DP_;)I
zT{~R*_l3TzwA?gre&rXjSYY(G76g!KxXh&o+@5X-oi;o3n3J94M_{XPQPp93R(McX
z(xtuj5Q<ACrcUv}VO;g~?*n3z`&1(VzDP(mQo1AA7kzon5VHW7rkkIh*G(<lQ=)AG
zF$SvMP{g3dI2r&`r;;Q-D5DZaF$5cE<|;Ry;wy|D<VjR|H$-D*0C~s>bqoV5>-?JZ
z$OS6uy-W`%T8f~w>H~whp&&^>)oIL0X{`mHw(g{|Q1?nNsP+zT8#OQn;_`H5U6r6L
zBk+v{2IK5Bm8LgITv|)uiM)74zBYB8bl??dhWsuA2Gw&<OnX3LU;pA^(6rB^(MbbQ
z&)X|dv?pQC`Li$Wi(zX4eiGgX2B5+EdwM+WOcjOUIPs+cM3r>Y$$lgya%%U|n3I{v
z{t3Z+)ho-ST~ThU{asjZ1EOLv@G-Ym7m|W=gt^|N&n&Uk1aa^NXV}A_ceC``<tDVq
z-*i{)Bq!;w@TT~GBVaYlrpiAOg}&f~@0Hf=MaQqVBaSZiEFC-$$D3>IO`ic&9Cfga
z4VgcmUF7Tk5P>tS!J;)4#N|9F)E}GA0t!k3@s-GY-yofqe@dt%<v6H;m0~B!zN5(s
z%-i&pQ&HQ$RMJ=APY-t=v<_0z4_<=Zi33J&+57A-0x@AcJMl%gDWNnbfWvJ82qeGi
zg24Kb=HY|;D<0spe6HBYCv&Os#T|>pnb%D_x^4tPU0;q_xP%>Ck+;qIMsCf-$LL<Z
z4%ldnO~3V&1T+=BDPKQ(=^f?6Hueo3bTO4h({ObjRzMg~@L1Y?R|&Jg4jqWJoc7QJ
z3~*5P8I`2G5vOc*8uxT^7267+I1q<x1Ovu`4(?toEx=Kub<s<_3AXX_*<RH{O7YmY
zZ2_ZIO9l^o6)tmARP3XLxwrP){UXVDWT%gb$`ES(Meb_5B^AT=BqY>HNA>x&Wpuzk
z^7iItF?fw^N*p%0pV$7shr7A(gx(X{I3MACyfEzZ*-rZ#0*mcw#OB88c+1Ld?%wN8
zBcjm1^psW&i~)A`!wAdi3AYi2=WUC&rZ{|<@9HPQvYVbCC7FaJFNE9g5WGnC!+z$$
zt?Wc)fB8d}^%@h<$ouR(^iGzK%|&?!{V~mhN3+v8hDXLhcTlhChvrfa)#DM(otbO3
zZc?2KAGl$;Kgf-yJDw5pd>VZg`KG#iq4BL$XF#DELvFywM~@`clpCBEyqX+qeeaMP
zEvWqsOU=nR9tZOX5xhzvA)&^rz4UqH{=iU^JI6{WUT-bMLUkx+>vLldl5+*qizz+K
zd_Ss98HcWlZjAG{VjKDNSWb>nl;RirUT=?x=(j#V)<_144TbPVFMPumTENTS{A4E9
zYaR8ks9MatI^1s2M)h<6DQ-DX%(bhtUdP^)D+;Ieum@!N(ag2-%Cu<NUOv)JR)CQe
zUjK!vMSo1cHsntu>$Q&W<Aa<)g0ma^TqS*}Ci`McXOI`o)cjVZBmS|TAp>zQlT5;T
zZGy_GTP_j8I`IosHKt;c>&}O++<-XUVdj~u&PLOhva5=<b@_dgU^7MHCee%7)#G=&
z@-;f)rU*%xM99uC9XE~yaVQ;kKkBI&A?&MvQz0~Mgc*0OGstxPXV6CN$KVLOpBPvQ
zIFze#{k<-ctL<S|FOusOR^zTLU~3IC$JyPvAbBhCN@U>NR(OeKPAVaFt!9U!VxM@=
zer0Z;<oVCY?+x}1n8nGUmSd8j=+DYcRze+o)(@Qt8jX~es&_%ryOHJ#^kKXCt@y-m
ze8W{p*oYh^m(V^a8{qog8l%OBQ2D~+!&-$v(SVClK{LotgxIip&dwKg2kl);Co5Eg
zE|^Wmc)hSTc8#G;?!~(ytU>fon^m-qQP)Ep8j*#*lr*mOrLd^EYc>M!n;&7T5KL(Z
zO8KJ8VWt4dLZD3cZPIgsJ|6q1;kv<@u-S5a=QX!f8sXWJ;P3Mdwu-~LX{XC>a|6{)
z(swx{bC|91aH(rC`VJ?_UsFRaPVYB#*n;^oE3fwB@(>#Z%nqed`(*rw_&)I+$5(HT
ztmvhGze_S)2sIJ2bg59#FZ_Xd_i819`K?%zu*c&C6?*PYVJ$b7j3>2{=6xLMwOQuB
z-%uu-j0Tx5%R2Qa-~0|THSAd#2{QG_zO@#w<c=ZvXPr_;N+?oBbRQlk_L>A6G(6a;
zS*s{REK965u<m+timYBNy|FbatPwV+<PNY<Fz5A)A@_-W%ET?u>!;YE@Ux!X1kC3?
zR1yvHzNQH9w;^y90WFIlM-5ifYNana%0Xu>D5E{}1U@`Z946&KQZ(1E7^Nf~`UsWK
zp2)mgcl%5fVcRk3DNqn>K%Fjvw;b>q(`qv;<!j0Z(xeKuq=UCJTVR?wV;O-1?|!-h
zR6`u<;b{e>wri;m^-~I<LJF*8gyP$aA66DGHi=k>Ir!b41TY-5-Z?+G<``}R-%{l!
z-=*J0sY<?JF^9)u4xJJh+@S_p*IUD>y|E;^YIVRII^451Xz<wskT*p1_(ppu8YQMQ
z?igjQzSrY2a$bvp6}1~nhxuhB;Yj54deGPO`e)^ls1NHs#4IALh-L=TkoIcY^bM)e
zR!;}Xlan{{YA$aiqf>>;)a4b-i9@(UA5Ui8|AQG%!8{4$ext~tdZ_sE=4g@igkuf%
z`C|<}{rCuKkUGHXzU@NOtmK$M4OWIhNDGc$kNYHt8f*Xc+I_Z88`YV#=!L4h<un&i
zve+uHp0V3M*kDV~I?!fJ5Xf@}J(T2FLC;!Fpkr0SCf7o|EGZQiwtJGxh7P#al3%mJ
zrS{JN$HLOk!>9}p)Eh0g%#8Ng5oUmvktEs=FYA`HSB+62{%b|_bh+;mVkVUN-@VC5
z{}@I`k8M&tI>#5fPtlQ*K-aildz5bCPuDLe-;pxg6J+h<;_(?XQNeoMh?1yZ@FFTb
zp+9t^A+xWZrtlHnS-{;2SY_IQX`*_OqTZF9#<npZbMPSir>;_&htYI1am8cTYMU#G
zpuH)Hkv-=NgU7Z+u<N9R?z{&{s|bpY<L;Q`XoKCP`O{-(4_jdpWS;1T$P&~oa_Cs+
z%bPf*(G))9U~RzLpT!T>U0ydGHcZG~?}R2Pz{HvI?;{crnzT81a<raR@N?&AQT`K?
zr0#cv1r-||HxmXOx5md3l!By%GTE-up3e$t!Omoaut8K?f@dI-Kdis0c27UhB*z#r
z(dab3dM0JANR;#2P8sUmY+ge?5Sk!>vnv^{7S3N34ECZ4J6^<<egc+?lPM!U`r<}#
zY88GMOAA`8VflcU@6;Yr*tmDM1ivPcG|G}t6CqWxja}<*N@k(;h>v&7$H?#P|2R9B
z#mxFT_i98puIZ(#foo_-h_udSa)f?EMt(SQR8J}fN#kwS6n=Qhv{d{1nQ!+X$T@rD
zntFxT>ZctXaKRw&y*!sRwneL&o$%=<HDH@<%=qlSgW2JN2y$DFMZQsSNv>NXLSPKs
zJ&1wu{pbMSu~e@SBUeq$<cQ}S$DY+j+~9v7KytSQnK}S*1Be$6e@Bh?;!f+L`ts&w
z6#x2aO(UvkQ%0;Q?GenMzhmVwg7~XHoF2A!wyEM7Qqial;Ordsb{supU&3Y9<R<Z5
zdfnh{MW*xfS6jEbf|6vBtlV!nUTYHrW;gt^{n9CB!wm!q+hKu9dG{0l%nPO6@gcp&
zZ=mK|x48(<57S(isP<5w`an&yaULB@ukD1_w0zzQ_h1U5VQlBMqG`It2t#I=UV*8t
zuV_KZ<TTdwyEaIc%}Y$%Uypqc;Iz8J7&(g%_yS52RI^I!t$j<BjgbXdgDWu`F**0Q
zPT+bBB&t6DCLkIL#E=le5>bH<;hBXDd(Lm+CC~7RT`O(_VoQQgg(m@)%f0dIm|}_@
zMmN^pmq+%GeNJDh4X1~FZ+X)}hY{K9H!I80aqytIfzcfSkzn7(1*TkOF8(|U+n;+_
zAC*RPaZnQmhG33X!X^0sxCW)yNKo##V7|j<EYMEd^&!aQXJm#augQL|AHHlmxV8eN
z=}?(h)`ShZfEMc*R$R%cTl70j8VBd*V92tnv{tW^SMCQ8Mxr&D4inNylx1l&`U09F
zwg2b|f3m`=V=>8nzin!>88H)$ro_8^&L|v`;omfp_b3pOhIla*cb!P`AK>`rZk?@j
zlTzsMCJ6^@8!h3s-~ic=y2*kGq_#_C3CG*9o1@U5jwgune%5-;HX8FS@jBSe7pHeh
z)HkdrnhJQrFac%!g4ey#n7eH00rmr2g>X1e6P^y${)Hb6x4?YwOj!9%%V9+(bNPX!
z>13%;E4d?ITPpZdPf2iER8VOiWheATDy-~3Tf?8Nwmb7gr0)(Te@%oheItn_9Lo6U
zF0awSrvLb0J6_po(NNR9d*Ki)Lg5fS4#Kw%w|6u^bApH-uD?e^wUz8)tw_H*9?~Ol
znL}x9E{*j%QaY4z3pih&JtS6?USlDSCHqO@pGt_3vOXqP2|9I@fEAc8*-N&h`z7ea
z-0@C+a}5Lsx9<lpZX3JTT8CfarM0`EAVR~+4G1&C4$y0i@fx1v8bk%a-OJYA(x1}~
z)IO0DFbOYb=BE659J8K9#~pqj(e7`5ms*4{CXC|`|Ljb4Jb80oqc6`}gG><*yh=6-
zaI}{4@p>I@yF65vBi$){-9QPo*4yYcSY-xal}49&wdbReX`jR9aKsLvv=q)PZ41d%
zO;!eyeAfPlaGiF)YfMRzAu;wXQ}?VhDK!y7|17(sRlXd1)u4;A(*%|@O89H%Qn$fM
z;OO!QOY+qwao?NL60aWQTV(nvb-;O+bo)1r?6Uq;&5NZ`%~t^;-k1Dq`nODuw3L#<
ztF-PkoTx+b3|Ghc_JV|b8E<|jZS<A?H2n`Yr99R25J>eV`%Rrzg!`yQ7<4~)r&~!<
z;fe)RtB1pZQD#v~-Bv?Rv-8t~iceG`4OeB+L3Pdc7n(o$YZufGJ{v%uKTjE2C(;I_
zK=;YI)_}J6KY|<&f-jpUUq1vHKJ7!*@55?L<2@9HZTp0w;>TicduzPM$EEo~(i@C^
zBp&3z(iGBp*y2uNaN<X1<OL0zBav919&lJRmX1k=Y5u`EKb?#VZPHCYj9xnO8tgJs
zqp7$a|F=Uc9f4Sf04YNh%CN+BXIhxxbJV*U8v<Q8<%-pGz@i~48jv-e=kY;$aNb;A
z@=rD^(h8+km6SWTjkLM@o79v?{JCtQw8kXA=V%zv(pSDvQtGfVq_6!R9Ojt!PEgqj
z(t0-08y<&$w&4ZlpEjdXr=KR2e7jAgGWalx1~jfRM`gkqnr%wg1s8foi_!A=GI)I1
z<Fsr&uF_?ru^;=KhEoKT1vv-#lN~N_Q|3L1{SQ7-u7@(bPh)>moii1thF;v*hS|=d
zs9c?b3~I|T`*CA->nN|!(hpRo%5`QHMnP4x6b@-T%hh@q3#b@0p1Wclm2^}?pX!Wv
zzpjyY(&!s01<^^->0Iy^Gs2-?9Gu_%_&8o@&z@oG4gNL%u0*@66<_kjrGR^GA?<Ra
zYNmbm%i;3&P@{$J@3kSh`7P=}wTDAuDPM#u*2mo1(o{RR!$#J^6$g&q){8UtLxJ)w
z6sY6wZ+};DK+%8)oNWG17Wd_|b(gGA0M_7!|1TGtr_>!~nNIXHl&+0ecwDx$xfO%x
zVKX)nAyHQCyHstJH*|{Do_oAfWr3K3HH#7xhR-FHC?YT7Az|Ol!tIX<Jb94bQp%qD
z<no{->mT8cWnLvp%cS<1vI6;X-AEb_D49}meU1u+I$}l`-*-r&xm!T?t9AWuKYqo@
zZCx~nafgoP1eL{$Ox0X_XRV~yjs(5^tYO4BB&SFl?@=`0RWU|<ty>H6O8zeb#^%7y
z*a5f-v**-1-j_swJ=y)4&6G=e1_Xo@e0w~3JmFOwjN5@9n3ry%-t(|9{w*a8P@2uO
z=!h#SkZs$o;5p3Y80C9iJ=Hg{n1ejZOej@n4j|a7{*xSmajf1LRp+o+t1K2TMPo-}
zogprUHq{8!t-1qHAS0;Y&xgLNQTMllxYo_M0%7=YuEgCqE1c)}M8A+zIyF1K*UO>L
zIQ}TC?J#in3XEU2N$8u$kx%PQS&1qeY)ZO2-Pl%B^c*V@!aAu3sU_}j;lOnrD6A7v
zJvjb(touqYjbkk8y?ABBl6eV@s5wRIWCm9rIm_MM7&o`#$#4z(P}^2!Dvj|g^9zxj
ze#No=)puhmRF6E4_)K2pYTQL$%<xD`e*X36J6-@D4X2dmB@K5V9tDB90^cj;U$&}z
zN_34w8<av%j?8XA(*uquaUMsv2+df`;$Io6w@rG}TA74O;%H<v9x09;TA%t2<hald
zLUq|lCxyn{Wzs*I*1k7jL<RSU|H&|Hmo)zzlW^k5N5N-nz~NSIPL$HJDQQ}{{^jw%
z{iLEYK;cWrpF_vlQBAb*h+Ha68=^=V!5Y~k2eKvd#&y)<j`~B^+yClJ>L(fzPeN*r
zrect(h<7>mLK_1nWGkw$yB@P#htMGne4o^>qtXO&NULTaIE0?2KZu;Qc~$d?Jd)S4
zkutZyFVKBTCTw@3-@%8^1gmkdOHB*i4X#Ogv6UCC+sTcj89$ccVh(4vmH&5l5+{wE
zIX1kBm@F9~u38qjjWX1`)t)@R^P*Ti$SkTfgy#Bq-uU+nX&47S{^;B=2(RUl@8)W3
zqY?c??II>MeB8PgZ)JCy*rC3K*$EbZ(0`n7TO6OA(NX*R!R;VaZ2v2?zdve`#PL=J
z&z`VO#ngit)ebn0y^ZON{P(KD>^#^AGk(uYxSjKdF_+qba)h=yQeA<=HnHl5FgNNa
zGr9cXvLA+}UkHyFXcAOY(TI}k86^>PG@j+;Z6vH>Cx`xr2fv6HZ0<1b_C;bEx9^s!
z&CWGJ@q1IDi>@LU6jX)t(qS`>4TM8^^-r7j&8RO9NzpiTWRRt~=>_CVA$m{+G=e5R
zpc;@NzQ6t(rGbBxC9u<}lDXY7hc5KMI-KM>TrY7Wal<IQyYN4(tqBUOqr?52S<($N
ztO<`cCuy_~pRT+PB!D}S<POnOHF;Q}@!6pt+%_+}i9B-1`WCzE$xF5`^mthT>KgdK
zgN9cGl%g{0LQuUZi?)qd=(EDVCN1o5KO`AMz|n69S+9#Fp~73MyG9>h+D`*sbL@Zi
zvXgwn88){YMk`!YZHh*Se70MnOF-Z7ooM(46wrh7F$YHCU9d=o#jZyYVU2f`M}D>9
z&WIVzyGD+kgLzb3pV{^-ZtNfQ9=@j!1AO^t+^#Y)?S4H|39X*zOB+44`q}8fH33C3
zNrIa`&i(0iG}tA`!J`bg?kRFSn6IM9!BPk~9}F@*>RIXfcT+es*xl7L6R9q*j|vX5
zULU~UkI%|~0X<Y={owZd+L<OtCwhqm_Q~8E+EK0-c-U(Mn)`v%cD75~;>Xs}%qq>l
z@yMO4tF1gEl)qR;`J5-994dP8DQt|-uDxxOwL78vE+WL2MjE<>eekUo#BzykXiF^)
zCw3^)?ywG|P#R(z2v<-%Fzi_r0>UC~IQhRFBfXb?Z+paPz^r<|@L(%p_g(X1iwp-_
z4@OvSQt`Q<HN{5L?t5_L3Z04DBWf`B7YJ1fRzY}wJ}!;>h81tn9*Tg9|2WPr4fss|
z{{CR}57x!;JmlH!8#6TpN?gAqT|Nxsjswo+*Mc5Y8^m(2cK>9Oug|$!4G#qVXK#2!
z<sMYorwCRO(b#*l+c5rCS(+kJ-B_<<zxc!P$|n&I*>_4Z!X6LW^hv0j%Ae+v^9C=P
z5~8{NiM$97C-CEVmKoONx!mD~dLfA!lqqORozp)SA8P+>b@jDD_VkEYO07Msuxq=C
zTyfZIrM7g!<`(AA3T70ZGDn2d@%3Zmm6>>@yRn1Ozne{y@I5$s4>%B;4G|;Gr7!^O
zAg-i5w3K3)qQ3lq8Neq<b>0(;c`%2yJeUR?Z*hW#v%<}t>8Ib2P7gh$uGg$$fG-cY
zxvaZo{t4*cEZLI&O`J!)X>=WQw8dA5Z_1`kC9W!I{HpszAb_QSBi6}&JZ+-y*2NLx
ztG{(Dq!8F>_MgNH;3$Dy3i?7A?^sqr+2=<T_?UrPPcEO0rrZ_Mq(w;3GW}GgTn+?r
z#=Jg*2Dojq9%e#_#TKmy6t*JWTVK>e1$qMmR3xbftCincZ%E*AckI`;VD_O8w-;nZ
zbjLirVfRx>#8hydOhkaFd$;9}S#omLy7z02W0I~YdLcx4A>IO9Ia<OMbsRZb6k`;e
z0Ek2JH<a_QL;7C_ph`y|t&`I3VJ6aXr%G|c&3hY!zrUKBGk~noMkFoDMD*M0(bi>s
z8J(2QNww=tu!_4AR!Mz4&s(nEv6|fDBk;t9mNN7l5Kgk2M<0#Up=t5);udzobIfa`
zD$9)g1%SXbD686QlJXUUD#L?3hM()4iLdsr`gNJ<0*<*nVsUrx(23bfKq?H~Zqses
zswm6caW10`AFQ9_?ns_s5BIn1Gm<ow|0nN0XDle6?Kk?v#u)U&7f_{4RmgO^r$CG@
zpA@O%G2A-&gnQI%$T>q#c|L)PD9-VX07oW_FsTAK@+~*~A?kCW`795|JG$;Y3|aV;
z-Tyjl|JOqxb1%==;5u4&Qn=!`G30@!m}9CO@rYo5E*V#9#g26txykX9*l#&9*u=-*
z$zwe307&r}VG}+Vd#?_7!2K1;f0Ck|a!SaL(^E#6DYfQ+E1m@&DUb;9v?%V=;lJ>q
zzX~rc0RTCPh;>C1w0w!BptuV<p6z(Gsf{fltJrh!%p?PK)t+0!x3S0hp+b$+pN5%d
z-6}>5+^99W=$V+Wf6M4oUE9VX(P;1lOt1>rmL2DmBQJK3!n#gG4`<-YcfZP)?%XgW
z`R<-Atje^-^WU|b0jNAy)8(UzEQ(N%-XHFE{4cQ)mo*HsJ&`Rv=y(G-gUooi{h^M3
z*SvX8RS99i_Z8zk5FY$b&3IXfH8>`!KLJ3tN$9}Q`hR~}&J&+^1ZRJJAnnGb;r?Bm
zN|N+P`_1U@rH1-G2Y2rp?awh^Fu&Hves~$hk_c(cMSpUzN$`Wb_>aUgDaPpsT+Aj<
zw8=^WrRV+^rHvpL2&gquC+Bv<W_f-d+pvI8`R)w$eza1p_JO~yPBTTGe}Kl2vazZD
zueWHG0;f7<89-BH001qO0TdZHh1gbo05Ul{P{0@7qxtNAm+;E}pvwO|guMB`Ab}jm
z;QZ+HF#c%qp=wRX%V>THb7@-<Fg-qx(%I<?A1XGyPkExo$y#OIispS1p;v0X{qELm
zXv0~77g2izwFY2^ZZLI=n?cacUF2auAHh2|q8vv2e$AlbHC1rmQKzvT2Poa}xp?_d
z(1vBefBS;pJb~ptdzr<4r;vxqfBxJkA?voCu(ij7USRqq{G0QldIiq64gtclUx~c*
z7UXqo7%N!2QqfcEGVWjB4F*APc_e+x#jZ~r0FdqZk%A6&gC7))%^>LFhMqv)kHmML
z0#x8=J&mHx)ZZe6vA9kfz?0xbMZNkvBdOZ}bpMUW18ctiH{>b(zwzV$S8MC9W71aP
z)5%<h;g^uKD*w1eQAg$U>nbqK@heI_9HR$XjvFh#k2U#7^e4F|H2fkj8yGztCO*lQ
zNUsfsYQPC@|0pd*dJr$*04d7J3}|TuY@KBg2>S?M^Ucs+G=Wzdv>Z#&K~=&@DVT#j
z<xT;dNlr^@w1%s-S?O^>=iSl2Gh91WWA6JhsX)uE?{7Ra^55@2+#|UoDoj@F*<IVx
zWQcwxDVqMWQ%-vCiFSM^^O)7?H{Ovc2mNPKl^{QX#|tvUKfzawZ49fRNsWHUyD+D5
z&@fr3?&9vq;ZOfCyBmg;nq8W>5x<QW6AF|cb=$A69n5Ly@~G&$i$dwU*JZ0)G&_>i
z=(!QQ5%?Fl7IRcD?B{;32O@X=f4=aQAkzUcS4lNTMnaHXET(`jI2BCLLkJO`_DLk3
z7?NT?WeJj)rS&{>3j;$@MptE8{r<dK%e;Q+$aaGz2LA>+X~uTRP!ioAng`8Qy<C~~
z{k&Dr1#0#Jw{6odO%5ep3%glU#*1^|{qjt;WBK}P;8UFmljZI8EB4=U46_1`rC*W>
zt(pR?|NoXjEq!Foyo|_;1B^%HN5dE#C>xZiXt+{2S?Ni9s0Cg+l_L<IAiIrB1MpzH
zTs)}wh%c)9-c*QNe>6y0ccB`COwbzbV9JI7;)%AR!6wGNuN8!#Zy<iH7`&1%2|DUg
zm!qdZ^$-oY-XBkvs<GW@9REBq;Ff4YqW?P@VIM+4wWYHovnFj|4D_i2|GS+GBw?c(
z#?K!E!{o_=^XVxhX--f;K7T59f;QH2Xr8f2%=s#S63KIO_UH=ju4sd7_(bPieLf{C
z2bp;TXqEw-4nVl6w!BfUW25U1*jFjw%hsYp>nt?9L}CmcP7wK;TuU5FiplWF`crSF
zeELmQ7K<19XUpa_f5fkBCY>Gg*f@a}w3xtO)k59q=OV1ZwtJlaJ}$QXjr5ED(*pXZ
z9|x2d|0JuR&xEj?HCnp9A<|&54tZXo7+Pku=O={ZvrD`Az_~z-yt12+!x_7P?3*qc
z&S~P<`jB6OVRZ|(v}i;r&-tY7$M2}!cOgOVxRy3XzkFZ&2!Y7NOpaO!JW$RXv9^AP
zE2+6*-0a`IUN76Du2@rZoGvavGQD1Nbe)XxU;oGJQOv3HP|Xd;`Ss}I9A3-q%h+e&
zzlfB}pFBN*Y;(W<QtPk?MA!FL`BxjU{l`ibU+VG>^Y{lmPHl~~;L|S;0l*%2Z(1hm
zL&dyj*lg*`GanEPN;F!ZNto@60ZJu;@mU8e2brpI$w_|2d<TlY9^!Hpk3Wy5iJv#~
zfd7L1{^gRaXR@v)%pt(!2HA(c2KNfAOKsL_ME%!sbp!h+N>|%(D5eC9vu?q*VD0wV
zn&XtOGpk;~%W*5RGA3@^c5Y|ze3$dnm)ruFF_>bb4%Q<?M;HJyG^onJ7GQpaa-k|i
z^~W_1vky#V4JU4@*$pcV2YT^gGj|VSJFce7U(%e&(D3rb)!`C+dKP)98|mKV6&6T#
zL1R9jp8ljlr{wBv{J9)IzLR4~!!=d{)_=V(oF0vcOJghr2ICW!Ii+^FmvQOWT6p44
zI_u2TDIit_5r*kBsl1UVy!#qjEHIBU{!qHXc;;Qn!=p6Wum=$&!!p7-_$!J;FyLgU
z7_<&ls9SYU9j8D4vaW;#xRTR$?^oGONzK_dH;pv;`nWfjhd`x*O!w2o<MaAA*4%-t
zE-3j2He#*)ulk`U57-K^|HhuEN8tz2;uC~;f45H!R-CqIP?Te%AGd^H&nvF-4hq$0
z4l&_IPTR=qycUml(yF*B-fbgu$838bK+o+&BkXhG@;2w{!>in~0&3&r-@DTRZtY2j
zg0|Fcjx3UUQ_{bD_1r610IbHFqdWssImD+t6lz5Ap?nBS&b8MCVNjuC`_{aRA|mW)
zeKqm_D1rWiM>t$=#lV&zd+r|`MZIdFjae<~n3gY>udtuRI}Zp5+d5`($9WL^^$hib
zIlW2JaP<yT#iW;ZHKN!g_32?9<@8j4JSYS3$lLDm_1VftP7-`5|BJNu3~Orb_I_<h
zlO_lXktV$ggc77k7ilU=FVY00hu&4HNRRX;B1OR<MIcn^M0$}90TOyoXrXy#*7NN5
zJ$vu>oV{Mxb*&Fx%MZ@XdEfWA#~8o=e>hE_N3ipa0mDjeU%5R7JMtX7pX-ZO-HDiu
zdVWsV#{PUJAeA&139ra5Sqdx245Q5;{D_@%fQz~P?;uU#3!G6>Kw|ThaCQ6t?NVQ$
zzbkkD=#fQHSS)2rjWyedjWoGnnw$eut4BS1L!3}yf~MMCCzbDZQDN*@MQz>%VfqH{
zv7ZTcDvqkxuPQ|QA#{S8_^D^=N!e-A-I!=noq|$XoT@gDNhY0H72WU9q-U^xy0Ce4
zr_FmS^r)ebdQ4y4j4tg7<r77iuI<AEpnS^n{$JHqCM<A;cKth-&mi&_fcpgBiTAV{
zF}!Ah>=bZ{Xm4jQk1mGSk{zOg<-@61Vj=EJ0=hh_cPlZXPCwuCu1r-6J2&j>o#JwS
zd-#f2&z_wedb20=6`o)%I*3!3GQvC7fqkM5gks#s8HQa3H|FlU)USEn9&3-fmkTb#
z<yXALf%yHV=Bz{8VCdt4{PkP}TZ`9quW`-MF@}q^%&t?hKZ<`}D5!@qz@gcWoSuiT
z^E#qbfX?F%b<B)6XClL>j~+jjcTWdH8jucrAfz^whSPTKBLGD6_h=e}9sK>Ewa<mF
ztNAGlfNxlbOYn(trfUmnpSJv&2e%Y474rs}z;Gcs5BliQLhH(6%D*hChPChYZU&$3
zcB0#8b=uUqvqFQ<*3IG9MZLGCIq|OY+^&qche~Kqro&>6+_$g3_R?SmUs3ytutW~*
zL2PXVM!!)vSBj9B*HG^Y%Bu#TU+mm-Br82DAORv)(QUP~nVW7ATGZt#IJh23PQ!)W
zwnm`q&;?mSP^mnwJ(;_D*;1nl6tint{P9bx)qlQ_{om|23(>%1kG<7l;<Mv@ccO%a
zO<cdcT7S0eZm0Y?4U<!40<9OF9&+>nBa56#;rI!l-8a?XA>Z2{GvDR|1w3j1<ab8|
zhc6{6S-QeZD4E>RT!;;s)Zv)S@PmRosql|R;6+563#WsMT@;76sZk*0Y0wAex5jsg
zc3&O*IQy!i36DD({YpJ^Y;{G_F&rSTXgsG$o-<6Yk1B2*%sfUdOM^9UzvU_NG_}Mn
zBO_S)lMY#3GUaTLo;6M{7cb&|{ev&RzP4s^XgFE#i+l9!*<#ytPny{7PyW*9owvO`
zE<wk?S@gM$7)arFESLp)8vGAC5X;%`a?Unv`^r*TaF*FR@@L>axf91q=b1?-wA|_&
zZH+q(F(=Ow2du}QY1=;$W*4UWe5*pE7!<Xn-}^@LfU_;`Hs$nzZX4BBgwH@sq+Ag5
z;CMC(s*%_Y<?#A^le+2bLJMgrd!{(O!sewbx!Xqkgx_pFTho`UIhnX_H_fHbrN?~N
z{7>@-my6`LZr_igqMJO$<4iS{`|fRT)V7w#7lajTDeZPXob+AJiXfP^Z8x64Y11@2
z-)QFo6Mg`ptcYbBNlMdJDN$Cks$T|k@srRO9EMk*y_w6Rk&x2QJE~jtSFhwysqXxg
zr$&`S7Vdc|p0?yr>-?s6n^9Dz_(ig$qYAcp-xi}#`J9nG)`$r^rHu$a<<iA^38p4W
z5}zQwQ<Bhc5l%6|&>yO>MbX#)C5ZgDJi3Tb7T_oT&k*BZz)*JtX)ocIvan|_=N1%n
zP@F35f&mSg$I3u6X>W#eEvasywr8x8?Y!JI&$KY_jq$RYQ*8I(muLKvkpuqQ4a>EQ
zqUXRu@IVEdx%Y}|gRnkG88C}1i+G2aU`%(w(KCJhx&R(<XB<ltCM>hknz^`TPem1o
zN9OSLs`Bgzp2qQ%`UXFYPu%#7c#)ck_~qSrV;WQg_X6cQn;%v(QdNdmF)@V5t?fAr
z#mAd7tn!IebrZGls$6c9(N>r{x{}wCZE=sQY=ic3&qKecM~rp-wOikhXI|X$<&+fA
zpykqZRdjEDLzD*MIutP;eXsO2wcNG^d-~JyvyGb)Gq+CmMkrTKAhO)<Kud!_6{acb
z21`BMwyhc|m%`jLI7_>si~a`I4K5hy!7hUjUvx79s9V9nCwZob_FKNW^1(;*5!!ZK
zsD31P5py?-MM<J(+H1-Q=sJpM+69ptbgf*V(*)72blr(-vxvAIes+jL2?B3&@cX0k
zSE8on)a#;X!w>jJ$AC0{^FSVOq;uP8CKg66Pdb!(qJl)T=(j34`y;3%m|+W&_NQTq
zm&qU74ZX#`qZl6gY#(qRYO2&*d=o8KD=(NO$$!y2U^mP^p*K~e_4}8XMg{7Vq9@1f
zdjabFOZTi8CpsyP^Olhwkd!_gM+`SrxQXng`8RkBz3}~3{}~z&xf{C_#_Tot-*Z%W
z;Zt82r_;IAR!z^?V%A-VT(1=)bMbEGO(94M=ZOeKHFr;zxkZ%>rDqnE+4-X%QM`5f
z3_TWB=?}{=;@l8xnQplz-xuxg8xy9Js~#@i4ct^81_$)nwR=of+B?f4gj+H0pWfR(
zbeC9k@db%;*z$q#R{6(b)wd$L8<mDSHm~8$9_J#N6*P0EQ@WAMFZ10DiW1aD#mRi-
zxevv<nS2Ki#nwGg2SM6f4#|?o#f>?<V8iWaECosLS~2IVr#G<>Xt!WBP*rWen~N8G
z&W0v`m)DGCI3rjdI8(F6(w`KYpOFQh<8o#WQ&!DCKeHQJx$l_dw}n`C1?(5+YS6K*
zw`?<C6BN@zgfPF^h{J^BB*k+i`f}o!C7nUpfd__TUmiN38f7QHJ1k$}uUp<K?-qhP
z2otke^7HaTz+Q(RM6}W3@JDDl;8)OcYnvnzE6D{QoG@Jpu_=RjYIQZ3#_o!@v~n~|
zac^@(1Lb~`(qc^7Z;JL7_N8voMxShmMgR7=B*$`cQ4VGmAQE(~AZ9CM&Gww+ci&7x
zodti7&s=rtUI#4o2&sk-Wa>u;+xs=3I@nkO1ZV?pI!t2D4GZjzYFGEUd>>n7Lc5x)
zq&(w~6`-g^L+PeuN5iqw0{bq`y1n5XRPgUJ&Qq>oyIV_Dd*^~Pdwvy~)i__R*KM|<
zb?*ET$bcabSM_%f-NtgwYM9ai2jku6uh80pX;cu7;G`~#Tvg{Iv{W}cYIx&W{#0Up
zG3TLpH&gd*R6mF#v0*rI5pewQy2Fi$-nKoMw9M=JGIzJZ1wcblwi}P_6y1Nd{EMk>
ziET1?tXcj}z}zSEOQ`A7i5tlau}zg&{p&MoF$=J!(H)C6Unwf-71n9s5WXpD9qIoA
zHwL2j(9TdCM83(PFF9V<eBy}ZbephAu^2P;HnrU7(3rA@X?+6(lz9$gaOYWVy-Iv{
z24nEELU&x4mWK0*1iG=7B}F`2x85~tx!uz_7`Wt2`(bZt<{iX%+Flm%8DL5@M~iSe
z&;7W#6<{wUcu2c`QXsQJOw~LmCv5TUlkaj;-_73`)ML-#J`(3?(D%cQcV)t;*!i>U
z&H>5f#kpX6^{pggMhHe?RxFvSvCW{brq4&KI<$1ePZXY?0Zqk(oethC1(l!ful?%A
zzbfcm;OB>57YLkL^<>0jsF(%b)yAAcned|?(fxd5f-~=F8<t8uljYAJAunP^`d3j#
zFT{lf$G|bgS2VCmCSy|JnD`v4fv2vbeNqCml`@GM$IwIv#uQ|poLFz8{hs+^d_V!@
z>3O2t3yqliZyTR=hRwTsEu@<;(n{w(e)C}r>Hll%<o(cXF4Q`w%X0i`)y<_A9k@}}
zX~Om^FtMZZbS^co4?6b&u<Po1tJ51nl=;!~onVz~l4X!B63r22Gg-q@SMm7A5YMCU
zOqYxwdZPlpud&EYe~&NszaVd5Tg(^JJ)M*5%->NHx20p1jr8>5%#B?7{*7FaSbNAx
zvoP7La1}2IDGkGM##Zg!BxgYGwl4`7MI){Y8Qo>VE%!Emtj<xKnIUGc2?}gG+#+h~
zZz{8!<ThA|HP7lVeQQ7;f!wTjsV@wm#&yPEycYQyvZH}_1ns)pO9Qdi0Gab`Jb_Oo
z@y~esx+OhdUDqxhUTHf0o_5|tf9F^do%9>Ltr1YID9~#-Dh)xG9-PjHT+UC>#lBy9
zbolux<?uRVXFRB!GxmmZ&`-s__uN9<!=fc%N^<c)d0fODuUs2L4qKzE-%Y;cAC`Iy
zmv{Lw1iwO`_^N;k4kf!!`29)Cw={xB58V+E>!8XK&+rZDzS&nm{161tf>$27y7THZ
z`+$1~b77Vj<5mrta?C|u1B7L3OLz|4t2_?RII<q%KKR%yZiX{C|1fg+x>(yfL|REC
zmL`NK`Iv$-9#~-o%Uvj~UE@29;W9S6)k!~Ojq-pOi!@=5tJAWTL}Ptl&Q5pW(!utL
zCrME^F2T8nzvPdFba-~SUzxaz5ykZy2g$w>UV6KDnt8LSw<JW;{wAZ}*<}hz=a`)(
z!B=L%ic)xrkze{6j1$5JevQ8!jF}@|PIu6Ic)2Z<`U=QnB(o$O%wrr@fp7t{$pn={
z`#(n<<;6l%+x7KKI1P?#hu6W!!oedS<#FvL@^5VN^YgbgG^)W=hU)_FHohe7*`i%P
z=R+RgTJEm)pXTaSCv;<QkG~E{35;~zSr>9NU4DN5xaX;^5HiQ@NDFw!$2W{cc?}Xl
zYvsGu0lJjDqZ5J2p5xD&bC;^?oj`T-qsA`cxritWMV#>uGCs9O`LC>UKPl~ScKY6(
z5RXIIXZtRt!Bt?2uw6Psx;V{F#h>PO&VWS<_k(`<-vML2$yD40g+<oYIfU>5@8++|
zz)BhY_=QZg6g*>pP+;T0qp~^czr7hsmX)@<R++AS_?Vv#c?QF;Al%YA7fKJLA9A8^
zq-Q6a+4fv_ck9VfGFw(+T>eHELuwwan;2?#<CM|Y5a>RK6H}^~c7COn@#Sa{(P=%T
z&z+5-=kHvGbTw*StgpJ^)+IG-6{PmV{KP=7T<Ylx;V=mqO9#pzjZfbz1yQ9i%b6lR
z@|G{~T_mh*+O_`E{cQO}x2!2x|NBq&Ci$$so889T`g^uj@M3v~dTbJ<ebSq~{k7T-
zkx%ltna}4XLCzq0Vr+h1OGV>Hn*_FMPCS^hi4?04yBBe%swP3$C{YRRf@jsoM9PD9
zr#!zFCTFzX*zA`YZFcg5J)@#cM_q~%j}_>CNKZ^RFMP70I$B|?a~Oju$Z%e}inU;n
zZ6x8Po1i!t(h&AsHF&@)m6olWx&cJrnss=cNwuDrvBmrgqgci6ZApZ-tp<Y!<L&r@
zDTF3*zS0-imjO(5o`<D$;#R4Bjk##7MeYM&z&2;hZpoFMNxiUvYY@)2%0zVO2{Lr_
zI%;713-u9kuTVi&%HCVof{y^*yX-|z$BQuml3MEMw$#M2-^>_$70}SpW#V9fvQMLc
zR<cvB((c{!l9$Lu1biyJ*PtD*aW)@0)3#@mxny!UeDZ{MV13k~&xkvMF1;ibSlY`H
zXn;U7_zdFnFh$?|GGP6VE9dQGdA3A>je$Lgo(x+gOLx?ofK`?oPVyITeG*RvEs~8L
zH~3(qvu@Raic)|Z%z8ZaY1AFDMdl_WlNMc7PQ-vV)Y|L6bQL*TCxirg#yjqm9`-e5
zIn$l8PkLQ|0qU%&_hTHD*|!Vi*)=XAfKlnwLkeyn8BYgBv`%%LjhCyV=j2TaFDzDP
z$olQMnhx*fCJD`%*am^MS|ie>Imad#zX+yJdg@hk#<(%|e|u!TnG=iNUke@JPxp>9
z;EG*;iS&umTw;0rX#G2RX>R>ZZYqUov`1*QlQ6rxcbly6D|si5`s1|@h%nCH#`eWe
zrStxTC5<uhST>dxBz^pihWhdaBk9C|LEz1-&<0qPS^i`oG3Qz1R@b#|e|BkTEXzbc
zPu}5Of!Ft8C;M6PSd0AorN*TYg!Ua4jOJOItV!Z|X1jSlv}nLn9NTQm?$0xjgICIg
z76tIyBBg|wl8L%MYU{tJOTRYU|DaV;!OVbu;v<sf4KEl5;_A+IOQnFe9f~n29VvYy
z(_%cU1p#S6zbJD^KMieqF*9)JUZ7W|oOVYA9wHJfAmXG=NSW<=+z!DuD~7qd1!-RN
zOGIk?GWRPReM3Egqq)a-OKr1#I|7xEGCW|&!i_@i`3U@}ms)VIy4<(K=b6qr(wETq
zOBVc;YRxmYIf2u=`1=^IULC<KwcLTd)l<9Lm~OgJy17%;ibuJ;zXe5{O)X7q&)LCf
zDeCJ1tUYX!E}~GES>TxctrURMdF5Ql$~;V=uqhK7vyAC(Y$WNu+h^{ed!avMnQ#?n
zsx!wi412Uc)k?TAHq`9!EIaXZFmcL2WKkFpk5nWyZeeN}bNEl^BhNDfMh6q?`Il?I
zKKU?KJeFxvn7{x9QpL@eDbv}Q(nQG9m0Cu`e4S=pA6E7=hzp;WAU~rjCP2P#O9@Wd
zAj`pMA^tiCa^<h!D!9+sq)ZPIL5#^nzJn}pChCMIZZw*JpgA{g;ednI571fw_2PoC
zEXp4d;M*LAyS6fdX}~(esWT0Wu*}Hgu_K#dOYOLi!vhwjP3L}NA3rOKrBNw$>>5m~
z9SYMSW$e_M_oV9St6qI$ax=0k$lY~jGQO>%DQf-m!MSdCAx-QZnm&cwmZewEySNk9
zJs8*S0l^6_OGDK_<K|4W@CZx7@iiy-Xjzek!qTwwxB08(+`|me%ZXgxV6dL1;jIVc
zy<o7MgRurt=`)Bm#Ed!OwEaEm33u2FdJoj1lgnp=K~D!6ybU=?@U>xj>U~eGRxGl-
zr;^c}_Yq||Jk1RgQoo~1rv3JVAgem1xYp<X#iiIArlr_cY&(1>b23zhH=@=hvGs<m
zF#l~#<hE_2@wk!#N<n?L%O=tIU#EZ~Y}*zNyC;}p4|Mh4)PI}i^ylG#Je}1`R@y2|
zzKSkc-}v_-5FM5omcz!vYNFjZS5rbouT?r_hYU#EWh}&h<{VxvhdY=7(Z+tY2&mN}
z0S}q10xf2kTNpcH8;e~O-&N^=L`5;AH<G6%UK*ytp{$mMAIO>Aev|H7lp#iRjtdU;
z##I$HxZMj4gx}v=d=3fH_DZ?U81~{O68|9EV%j70MS3d-u?e?1xUs9;2t52?%bXXe
zr8agQX{C_%I+m=|if8vHVI)m4;=T7$V22$1P03`7tn2AXOwgi`h?SxRLl9qZ(oYNe
z_e^@4o;%cfBuPp$`oquoBEsI(@1AJ}YX?m#HYP5L*oWt{6864f=c9r0l%?1g2K1OH
z&=>haV{mu-eAIz$CTZ_4K-R*4fjpHyApE7#&doK6umW{O++g6Zn(Q8bmDQUs&5%bJ
z2_g5ptdy#4t6It2;2W!F<jvTF`C8{;hBV5npM=XRSA#T0Q%HJGL%7KD6W0sG2Z`fI
z&ZjMp#5!{iz7o?+S5uubRktm4#)kYxYLL+PhYI^DYYKNpD+AI9ElQI2{nb0`CP_yU
zl)ep^XhoRmCh)Z+b3t6WAPz_+yn<i#xE8kll1}}&_dA)^V$Kw7&QC8uBOLqbue}Gv
zfb$YXcVCc@a3iM@;t8<%@E-<7;1zt}ByOYYg!GElMS88|;L!s`Jb0^I!K}*y8*vBy
zO^OGIJEv??5QChLz-fkiknUUouPFXE8+>1FQ8L|eP?N0JT3V})uQ|dQ%P@>l@=;l4
z`$Z~Zu6gAm@fl6-O``AYpl1T_e2QoX#r2dIXX@BCYS>PR4>}ie4}Tc%Bt`a6GdFea
z$xd+Jr*1D7YSB+dt0oMBe&8@l2FBnfrxC+e;HQ2WIIVhYOW|wn29PM{uJ50Thzq>?
zOD*`!u7LkpxJ^PkTf1~vL6se2M83iq1akAyb^9&CN-vUkcNGRmz7XI~g=%5??=xCG
zf&e-l&ipSB{nG945OS@bJ@?NKRqT!vh~@2E*+s|#X)*zZ>lpKz#Juj><A|6!RF&*$
z%I`PN4-AeFjOtUY>=mj*_#H)>G-p3pd!l|H1zDvuya6=l`=86e|E*$1fXDy8$rC)e
z7C81g6DxyX5qgi)f<tjw3G<9C^*3jW!fk5V8sP=6b+!#ZzLvwf+Sjjn_$~+K+!~Px
zO5qj09Q%anS?fFZF7?^oMK=ZKJ(U-LY?UX1K)TXP{!7*OcN!-I=zCxz>;-6%c#Bkq
zCjBCP_r5EN=Y7P$f;|8eDW0~KMc#7^JvwlrX$q8Fn4q=s4+YgBI}(#a&LaojKY*yR
zu6ypxWek+u{lkBJn(&V+vk1NFbgtMq_?fIZ=2^wK+d~tXbAEfl*jqKA6CP=PdKu8I
zvf35Vb~opZzn5{R%V_L+NK*{J^jCwbd9TCbzAa&ESS63y0UnO=Kl(6VQ-RLJHGshS
zLJ(VD`xk@tZ+7Y9zh%DvTgPU%rAX@>+rvnc-sq5ZZDK!3@V$AQ!s$}5iuKm3mC;RV
zPg#oHHn|;#MtTLtx8E?;sS4=l_cs(Z+#crgxs@|ZXr~OBFf;sd6Z9`auh-OWNiBUp
zfz~b6ea24Zu0)TDOfBs~XB^PIG!3P?9qrl`OtUDp634Ch1pC5_{lOiYFBhBEJx-Pi
zvn}{}>jovb^VR>^VmJ*eh*S~@azQ$Od#(TO`5Luy|L@1jOG}Y1;}_=F^;>eV*(6Ay
z+nr?~D|W+5atfGhXplUB5i3cP5&q+L_FtG#dfmV^&Y!TOABWTuBkY;-q-Ht%G}BGO
z3&vFn<BN2mR_2N|K8;KkqguA?^cu&hD0kJybqS{M1lp2ZO#4>lhAO#Hhe!tdKYWU&
zf%g${CQ6vMu9=dz20Qy({CA#TsXnC}Y>i;`S*8hfm-8M@302U>?q{_p%DpEJ<~ga|
z7O{aOS{<`XFmdQa={%5z*r5~BCH~=qd>qJxI28Ev2m@J-dU~b<HglQ<eM&7te1xx;
zUtn%h8gU~B?#ppIcyj9Atd-i1C-*yYWs2;uj!}!Zw%;bdH#I#k6J~$PY>8l1%)<2Q
zX+6%F9`^~?V!|l>6N1HK1h@d&_M~6!w`I^rk@lpT8d{k%k2oguR6i`QE6E7uz-HdR
z)&w{*Fl|z!8pYqlX0VN+1HjSS5k*x$Dd_1`P7f0Ck17+1&FWH(lN~&!QoM3vQtuDd
zhW1{lzxt<r76PXQR>_69k9r(k1jrgHTbcytS{iMI7KEs*PYk^>UDiIyI{lB}L$@QH
zb4CR~P9@{RTJO#Ls@gR<qeEFVo;)c++L&jCL|%}pUhN;9B~gmF^-|kzzaTkHRB;5i
zCG1WMG}KV1L?P66!~M(kw=P9q?k|>VGhc_t%D?tT$I<K~$WpQNmXycCH7v#|Ry65l
zq*|@e|NGu5L29L_OP4P8q9=S=lB_rKYf5IaGBE8Ls|b4|@&)Fe;(8nrX<kP**l@yD
z>*NoJtHO?O=l3^3DcPnL6HdurKbi@w_W}lm?9Kk4{s}2mzymG>9)O2Ff%oGO%>9}&
zxWnSrs}|jb%Vn<fPnEajNT-7@xDM+@GRu=S-#cVA0u#Bh_6buRJi5YHngG-RvwM?`
zW)@|K{;=0bcoRUZU!}MTW@?mq59#DIX1nYS;Jk7R51554vSeHt=i;_QcJS|BQ8HbW
zbh-HZd+-P6W_^+~X__=<OW_Ef_`pR&4uAI*R)#9&9p=@4!r$?d^%_<bbSq83+-co-
zIw^TXe~p_;=*RfXTVE+P4cme1`Q}b6Dubg}U-QLkY9y>ix4d7@VMtFnaY&=OFni}8
z?Frf338o6$fVW-6mkY}=6{#rw<Ak`Y5GwPw7nQDL3kh*-etQqDNdBgql|5-@!AsR6
zXb0wm`>3u1?F7tIF6g=Q+VvnDFV~X_&Nqh-{>jOq_1La2P43oXxCClbJUVNjM}_&N
ziuF=GV<=!hvf)AY(YRW{b=ld@2cOdwD9e1YuE*~g`$NbMmGb+jl24DVf1ZO2&5T)n
z;{MTfha+sMWR*mgDw~Ck?6PFM4(lb6sKB<0PF4_<c5TX;vJ|aHk_{nl9yz4>)uXJL
z%RAQk`H^y7t;@Z<)DW^owXB$sOp|Y<T1pasc+>DxgrV$$D1;;Dx)DU|boiJTH~k_V
znp9;@nM31lzdiPnfAjn3F$v_S<BVb(qJ67Qb1;=83n2arMEDte%hGc)Ds1$?fBF`R
zP3&IGRDSx((~GUjM-Q(Xj$Th_Z2ZLrMpR>lj<~6=dYOB6*i7)PJqmlVJzM(xxCGMW
zMyW}Yrp@{UJSyypvr?4}iknN$`8eNnkA34eOPemLFDGxK-$evJK99a<&0j#94ne3c
zwVwK%CX+%x{Lv|5u&D?O`y9WB@f!lmHJBb6W{YTPC}EFD&6dB=r~B@9LIilP$Rgun
zdFkI$V@^|FJ0?YJz2`O0)!$|i@jP(EOjk*_1qIz(3cOk$aI2-xD?(UN?6A#+CP?DK
z&N`O(v+JL3>GMKGSl{P$q>s>W0>TtM0+vI>tcu5WzpRZRN}n=f+&i|vQtG)+s^$Br
zE1e~x&m_eFBDbwWef8+&+SB^ULgB`Aai;y2V7lX-iE_kqvK=%1xyC?hR3&GNTI~I;
z5L=FOa->v>{DZKN|FYYZm;t;!_J8zDX?}U83X_Zudpl2HD@Js6<`dMIoAcu2YTIhG
z{ed2HR(`c@yG_1r(JqzjUc5g~CHT|r*VAw;ykcXB@1scK-FP$}d*gcYwF?)1@2M**
zJau&Cn(zZIH`c}>Ijt$?$af_=ESfp{;!rbIb}?n4ojDyJiouFywym053w4Ledy|Vh
z%Ubj?hMZ|9F>0S3|MX4kIlOyuy)~SowS&nedOL92{|GRh-&UOTZaOgySiXlce351W
zJpqOjTrMf=VANOuo5kSW<i(x0fez|B{=4?h`Vj-lD{nuRBvn7t7Etc7*Jx|uQ5B+m
z%x_sXTHJ|I&&loQ7q;5!Nh3MEK*27`^yNkKda&%9j4H)(XLsfPtanE+1{&W?ILGrP
zywy{wKizS^-E-6ELMuN#e`-AI1-B*b^)PbY9?I8v=P#0vSls^#%|$8%A3fqsVHcX2
zPSv(k#eY^)w!3OUsoR(ys&DQ~<X)hqmWrbu=~LPn?Ip=MB5E<N3iCdkB~g=mUGsvO
z9Vr~EWykl&k4Kw}yd(|`aWsJ<ISGU0+-7~3=cnEKO*?LSAFmdoi}GSyRGFMGMj~Uz
zF0vayTlxr?DiI6Qk-G)dAtm}`aGIH3llDM=AN}q3`H2<KD)$gfcli1}nbsKKlcMfj
z1q@>U{agKVk?5b@52I&nApM-iVtGGck$#dNbT)ojt?8S8JS;A?8jUt2`Le{CDRcAj
zvK4_*3iaM33%+lBMNE$FG2>f<ea_M!fjIjvl>Q$@#pahR*CJI5OTK)8CjFq2JG!vl
zaF|^Iw9l<ZiXPttX8lww=L9qBoB489O*!9ltmu`HI^AjA1;$*o28K*jyO<N^^?Z*$
z*hZo98&(T4Tmg>F5*X&wL2v9he}$)M(7G0wc89R4=HP&E>O~Ah^YTq->Bat{4%?Yt
zslYBH-!<iOx$JwUvJ02bmL9eP12pHgZpG}xNRa$K2?h*VNSbxk?xaJm204q!Ai1%!
zBN}N%=16x3-nl=#3zbOZA{CW~J*D|n9-EqG@ms4~ot79@xSniQ*U@Xzzag{$ZpsDJ
zFe%#sbW0H&uBxRKqtPAj=DyU^JYHrsh;X+h;8vu4?tPF>QlSC|dm>fD$U1vdv5H;z
zYlAFM=~a<XeGEo{=?D28TZnwiC*i^mz^8oP@hNV;!IP>50Fzp^j9dJ%Qde486dy^B
zZMKd+0wxTQ%oFRm{c$+|4Wrk36BgBnyxRaY^OMpm`lQK?T^hvCPnYHEBG}ulMH{yI
z<74}6wz`RbP)JU{YNfn)jTq#5A=_txGys_2<m|7Zm@a9C0s>0z#x@KWd8Z#c&q`mh
zUL2|U_V5xF;>$$p`EJI|QFPN$ElO~wHalu1ym#)*Thiz3-j9s4&yj<mOOk$H<L+ZU
zy4htY@F!pXDApb21mq!Y=)Yj9Fw*>^V`lz>2ebYzV|;j^jb(AXx4I)G{n=69&hJKg
zt#(=arJm6q!d%azv+dwGNmM{Z`?L5xKl>{XlYQWhpBQq!Go?wY!V~`Z)dvfjdLq~$
zFx1({%J2F7`sPmi<-!kZuUqG48ZWmnRVyEjmjQ0Bb64zzKGp1DGTqsqPD~0O_Q%LL
z{OMXd?FS~<TX9v3Xe-H`9tCnXRZIh2dJ(sUHk%=S57R>pI8MR!*E4;K5DE!bl=qVo
z(MC=---as?gMCw56;Sy1m^(dW{p^%^ADDMNXi{Cy7@oy3U*3+*0&HbS<~<!)*S^ue
zIamk8pT^38a3ul75zmIjfXp(K9l)pZJQ^wA()s-7&u$F%l(i2yx*fPTsLcW8F!b{O
zRY?XHqY|G<q3n;ikW$)s{8I9XVBurmg9+QF$MG5W_R@i1@A9kKE_{bqNCku)WW>Cr
z=|3e|@uv7lY}{Jssc}W}l;rW$KJ{-~tv07Ig%M|^(A!nY0jSv>VpQ}oB&U@eN}^KP
zkt^l3!WMr*51;=0{pKGBOvHcyef{SO+kUU%^b%n1*f^n26j=%6+fxQN_m|w$DB%U`
zbo!^+IDoii8a1@N&28Ynw=%e3rup-7%p4&Uy}ik7%*x5)l|@0{d;h3UDUum~Ye7O@
z8Nj(CrAzn~-7mL|eHC`QUNv*KY1YfhM~CHI?O@iNcvmYV(VhHL3-w4V?4*NN7f}EH
z*>13d0yn=HMO-MYwkysSh$*pf@kEpzl!=^MX*Pgjbx(lx+UgFSaV8faR)!;%(#HdE
zy@SASrq_QyLb_Mm)d}q9$ceBt)K9X*%pvPFJD<2yvSuygJC4=aBoArQ0kVEtTlvLF
z4CnM+)E8cM@Y=UqEg8X4A8@;oUz(seTWk=XE*<1Kx<*0yU+VyWlkH9i8nwoEgqu#G
zlGN1gOGxgVXtM@Cgkl0G_}0+it@00=sdt;dzu2l>hmQj^(at;b)0T!gW24ve`$aVM
z=fC*C#V=TyPh>v&A1bhZ9CEqyrW_roZ08*MO)zXp8c^1_(?1UN8o%!t7)rHCy!}pL
z$6*<9LN8Hecf8d9-yZsJ7D7X~xWS~e0bZW6=SS$WQ6mRLf|YR1+$E3Iq4pc$0IV?m
z-EXMb9=FunI`Ua}cRqqeT454ydT#?jbjMv=DQ09cC1-u^bi}cs7RWRwX$Zkj#SVHQ
zCpF}-CoOViCSneS5QA<?Qs*Mzk=<G}l{43ZrYLBt%v--h!|Ad&CFn$IV?!-Z_1t?Z
z><pxCAnv-(*k1N(4TFIHv>0KRs0h=Rahe2sY>d+wJ0tSf&wp-<drvt@ns`mv_+S#$
zrj-(jLON0;!|bQ3oMu=R9f{87se)Hp(6RR@Sps%0e);l+(cJ~|;9kobbvnB+%?^{6
zx#N=56_=Tq$tiE;K%iGpQMcjA^uwWo6S^t!8g+iMb1(9O)9*VpxlGNp3AZX4F8ZNf
zN6z{)?z?m6O6<@;pR>j)^Wzz@htK{fnI7N><01!<&+g2(x&hN=t=7L*ds(_~ejn@5
zGj>{RoSF_ioXJZ4`2FRtpPsNqFc-+eD@J_O)jpUo5e~nc<4tu*?xqprQ0)C#y|{sB
znlO=igzlr?Te>?gC#3ZU>q}uTfRW$MZ%mA7vTu*kB_J*;mKC7^KNm^GAFp3;ImFm)
zkdx4BscAD0e1A<W{~i~8+L8BE;!m%iT0qZ>UP^ikprd!U>b4gM=n-#W=0Y21AwRyE
zD(Dqq&~f)TFn3B8fMtl5Y~@eV9Pj}ERA!Q5wDjvbd9)eKjKSCVmm(}q*Bl#piKTfV
zY=?{TzQ;B*$-R<@Z@rV+cHvGE;ldUJL0VVX<?_5Xp73`&jnWJvM1UTgGI-Ned7s=g
zWuo@(D`9rASfkmvi}|nqOYZe|1<?bTP%d(&tYo%coD7hD5d$HE?;MEKb@n=hpf34{
z+E~K=-ex6iUC$qYbJ6v<zShfBGOq~C+YNR>o^*2!#Yq~Uz+xr4+EuN3zqZ#AK3j#6
zbvaRgnHUFYaf{|vF(G`xAywsluU{!Y?-(m(n`ex4K{4;XV_^G|{aUIA2{~aGUCg~x
zXoIs``s1BbE|^Z2N~eC!kQE=ZNXR6A>6;^5gjVND>#<Q4=VvE%iY*FnQp58`p{P&K
zpE;-l^|n%%@1$DhAkjS<!R*b(>T5Z2l(Q7Z&)qt|&Ove!LWFkK);H#l2bpT_cYJP3
zstlktm4UK8Jg97%#Vf@|PGpS}eYIFp>Q4h8;oNP%-t@H|VCe^dYpZ6q%)||n8?m0A
zp4ysNHi*NhKu^{^hF`ya!RMfL^ppk5m)UNOC4LzDG?qX8b<zcV=Zf2_yTBBQMa)H!
zL-H$mfbYRZn%&X17m2iEs6i-~vdJ4$TbJ=`%~^G9K3gCk7LTxMS3a2>-z$l^!fWDP
zancSU7a@QA_@4%>^$oRc%TYlm2nbj$j(IcG`glo9@OGw!U)mHxTVqT2TKR$c(s$8h
zRXxV<7;zgXs|Pzq$YJp@6LjbjFcUS|EwdybDlB7Gxy_<WPs>-Ri?4_(ga}3*FnxQT
z+?L%3%QG@lAx}R$$w+a?LfJIDOX7ESW3Yx{PWb3Lq=11OA>SWisoUnz*igB62N^u+
zanBeFVR3w7_adM3FXCn_Q{v5HCg^4Rdtp)ZPR0XsZ>XstFBsp-Ja)ry^%oz}s)IIs
z*nr;f^OIa8i@G8GjnzH%V!BVIlE{l%kf*^*alJn+q@5<Ob+fPb^MNaT@@_SfVihUz
zL8C0O(X9*j%~tv|6J1rdZNEQBNMg{cP8%?hC7k0_O@K;bnqsR0__(z~Wdfo032plV
zXr=opuvPnZRtFY1&LnXeCX?(u95`ICI`qI<gAwf`>!)Qnb?dr}5Q`rlcp_biSi9iY
zUag71!UHT>8wX~KF0GOWOMUW=dQ7W*fwxrz{}t|2=dY%D_CfxoZ9r6!9~XNm7ivat
zsCXbcHOC;Iv=`Gb;2B<<MI{3u3hw~*tjn9RT6RiAPkDb7uQ}`gW@BVet#fs9<Wfbf
z!fAJ+FtP(SrloWoh=--aT@hqj$70KYOSSAsLui^F?*sEvcrkB8Kw%kCsC-em$x9>j
zH`jPcGObwuTt%_oG`H%2cmS!9zfwE^(4fS=>SS7@-g-n4s{tb(fK|}15H5Vo(#Ax^
zZnydQbtcDeuN{7U8W_1$6eeOy4UWx?5-nI1<ySRXI<Wk)CzEpV^3|FaAN*npx<B|_
zOsIZGk)S4PGxeTX%nY?pTrR7}<yn<|Pd3#{MdhcMMFH|8RG9ayZ(;YlJDXS&SaJ`{
zLfS0GO<Y`DqqdLB7%Z-Z&9qZddTIl@N^%*>$7U=67!@Sq<|R`sZl9)?ik!h{d~l0}
zcm0P9uDn}*8(WfOIsrwx$NzlRf-O`?FhnSdIT?w3r|)&#F&YLHq<mGuP5lKJrR!<J
z1k-B;dl8Id{6v0de!WP5VE!QQKu&UopOPBcf2eTm=V_`1(X6)Ps^(7NPiR$CV8%n9
zY|l2B&TyWYJhg7Wg{_ZQ3QBLQW$UC;PnKNdZ*Vps!?quabsFc&<7i8tH51SeP|h6y
z*^*8FG=LFq1VFghI{e8-MVvpv?&6%1h$$<g&qsWw&($W3D;OCJ6h(Hco6aZ-bAqI`
z0XA+dE5)nhmhGJR@8gtWnVqRFjN9-sa#0<|h2YiaG4Kt{EMEjvT;6!I!2Mr7K;}8~
zT(wrdKg;4evL05jx+(?ad|zqP>G<lsnB{%tkY{_RWU2x;J{2j1as|J>ZiaWV;URFE
zP|fC5@#n{wevy@5DI&og`mfXyc$aB*i7z13WF-18Yi#GPD|QzQlHEUMUI}~v#|3F+
zN)1wY<eP3LyZ*9^Dyd1PG;>H4ZaB9FLC_WGUP%-<7XboLL<GEJnT)>x<P&pan4>eD
z3-DJ?y0m!zU$~qh@c;DFg>LKrYn>+0Lc5|?zge_S`wPA+Es<(1XN^0K;R=>o@Jp=Q
zD|f3A=pu#;v&0?|HZ<VWyKWM0qhPid@%-S?fQZ3+B?{Q~#*+^x1^;_zHyu?uO^v?Q
z)HWR|HJ9g>gK1|7T!}8sE557YtWl7A<MNi;{1{MrvGpK{l4Mkw=S(qq13ddQX`OK6
zk!4y0Maq5v@R@5p*Tb!1Ay?HvZdaTUi}d>65U5U;;Eg5nMd5n^cB^FwoykEDh?8cg
zdDzM~85Zq}O1Uk>ADMG<g^E`p^w+%HV%9i6z3L@blmFoJe*JxbT{gqRb`M6g!0#+e
zJ?U@93%f977-qLVO(#+<GA){7x{Rb(d$J!ALab#{e197pXxRnPcJp`IQ{vf~q&<pA
zdly*)XON!aa@{;&|NXT^KfrtFCqxUPZ)K8yA2L8VC%b5KS^?+=pg`r43K!jq{(9}X
z>o0$@0V_>v0h-Pqp44NJ`<y68{sT^ps!y!X_Kxq+F|Du0sCjg&je=YkTj2n@F3i%D
z=qFYREI^KJIbH*bHX^`e=`om2u*{R$F7!`m*BKzR_>Y5BGDDcQ@2xsp=hhdc@M~^i
z+Ae<O)ZbXY*qdNg^42RYf1LRUm6!|~%-fDVWn>XLRf)&>Vo#5LSWIFNE4!<YHadRl
zbrUpsK0q35njg}qV$l^ve1aDXmd;={?oaF3w+_I*?vM%(z+<O*z`R5FcT90#eeEU%
z_-{IT(hYXS3&u~UpbW#WKaqiA#;?+3!wNK&w&l)t_Df1>jva}d>UGL8$%*L_J<1P<
z+~YU5jC;v`Bsyr{!Vmgupx+5VzL-iku=B(7@A&qr15zs%x1&oUxoI(bp4?-<y6*57
z1kCP2il*T!zKVpvLts920wi$6Kbu2{>*%%6u7>}PMi|NTJ=?~yE>v~Hm>sZ6w~ktA
zg4=+@gwNJQPmD5!2+ta^XFIB4=yD=K;OH{HHEIHiEo)hWoQO(rgozMQy3XXkPhdbY
zG&ioYUe?L%j(W-LGxz$4Aj4vgc3iYr=OVviE{bH(Du1=6@qj<Y==nRM!$B+Ro_Fgq
z#CrAP3qO;4Y3``jM8q*i6m;gtV>|>;nQ?@&t9!p}M7FH|M?gHcGTTshVw3POo~jt*
zR+CwOHD+15H16YjnE@Y1#{PFEW$OaLf<t5yxB|{BL)@Sho#pGu2Pp0)nYa%3gaBE2
zV&V9dV9<xz?s;4(SDrU&OvI`~03uvv9Yx}4>bU3=(=n=bn8hs^?^HRmYxFU}ebIC~
zWR^*McG!9h<gBywcr_WL1NkJ+6O4HW;;LJ^8Iga>_ZQr2W4?J=GtqYPr;M&qZ2}ck
zHF$|i?RY&fQ7OD*9{X3sr~1-iVXq1l65@i~zn&MHyp%STU``RoDo{=s#0SYcKt?=}
zXFbUPBYI-kH27(*d%@`$w5uKuLhO$k1L&3|7?O)5W;SO9QN%RSe+TZmQ;roBUgH+T
z1ktdwvleB%B6kjeCX0FfPL&IkF3`$Yitdgc-ZZgQ5UTodR~8VxQDxH--BB~O&JOso
zLr~R8P6K6wtr~&)<HS6-;BZHxn)z_1X~{IYKDvTVX8#h}kdy5u`Jmq?DD?Tswzsc?
zB)#}j3ZR7vqWzh2#rLLsQ2mK#jNtR_rr&ZxwJYo!z<hc5Jt3AilP`Y?AsufI>lvPx
z_j!g-n?(wmJPq+BcPJ~WrTitH=#tbXm>Lm=TYhMU3tfIG5bt^#>gKb*uP<Vc)pyQW
zOK&eqc8439G_-yB;=2)M3Dy1u{FlguP8IC!g2Z&!Q(g1T@nP{Xw=ao?DodQ#v|x)@
z?KM;CZCCdn-W9AymY0C|3Qq@Z0A1J?M3DtdSt01VX(+RUP)KNNRvQ16vo4T_Ny|(;
z!~_ob!+P>KGoS?kI5_GFRZdV2UaETw;)HM%cFdP&=8qxgiyr4?%2x#akixcau6i?}
z`C?ivqxXrt!fd;_#ti9^DbRv?-jo+8a8<`HjBj+RMjKuU;)_6H0TpDPJg!r+FRaug
zL1{6rpdVRs*3=7*TXaIZf3oU~CFmLn7k2)M;8Qe9*>bjj?WJ!(NyL(n$`V>WPSrD`
zZ;M918lzubrgT+|xidVJxK4cABi?~@Q97n+_VI@smkadgAGqNGMP$u3f0}SyFp$wp
zCm??c5CN{5`qz!h<I7w8fqEn2vr*9TaLxh{wZHODXWjGe0Pd0-S1&JoxKEcy_pV3n
z(8dgD?jW9xIyP}@zp*7VGH#kYpKR>B8tOX+H#K$Ot#fd@K*U&g1kF-RqwMqa(iIvB
zujf6XV&a>pzQ&9V$B0w6&VoWMl!(#2%^_`9t#+6~2AWV8>ajnIp{ELo+6PS1*FqKz
zfD^Yb;}eUtM`SU0Cm{S9^P60ysCy|(FEN>8y)2qOR5d+5=jEMF^|sdxd2MX&Z(Bfr
zV*-cwtxYr}f!M7CuLyuC#n?(~BNbk^c`ibL2wd~<+n#6Pm^~)t#+MZHAYXOgXF*6w
z0C-MQ)Yvj6O<odk$f_?>NxjUMzz>J<V+%p>9(M%7II$E1NHaJj1LE9`C_56k8Sxs8
zQw9>Z)*1=Y)cJd1{VIXgc4?Z>j-!VH?@%u=Asx~;yYrH(ffgj9&uLYLZ|1$ys9P6X
zV8$tL1k%IAHURotQ8P}VJIX4%gk@(~5+LE}wY2T7Z%F)=?AD>qgn(($+jGS7M;99p
z8@3TPj47}V5Bn3KvDc)TE@WCOm8*9xH5-uinwuE#@NNkK@tx3Hf<H@qlWO^@sl`r%
zOOTXIOWUGl8&Zu5@+WL%_z6{$zoq_M-F72W)5*+srqq&#7%yKM!`Lk@Ac|_C#mh1T
zn$9rOX_fq?^+Rse-9mLW45G&EM}b9E0Z@oNqxrW$_7kn6n^v8{p=Icf2h1!ME!<YS
z$eOa7@?HX5(Z8>t<(MOAwxgA|^^c#A6diJ8fl_wSortaa%F{BOZ!9d(9&Nvh{{CE-
zd1U-$n@^B_my-Z_07)FP?Bv(NL;c~iimLA(fKY5791)PVW1}j6m$G7W3U1oup8sG$
zcmR{&7T`7D5&ZpQ4sY*t;#gz-W`J46N2N04mZg!Lp;?7g&(A{6scKWn>I5adNyBFD
zU~orqzXLNr*1aPCTocH&kiZpnQR_?}KXTTCRn_jg6+Hbxo*UKQVPR1K8p_!}D>1Qg
zOP#L1IV@8#dJ8*SrtJbKB|oZH**zI~+*S50wk7(K6G4T)4u%ic^tf=q01n(!X!yt>
zpo}Pdm3QmuQ5hl&#W#47X8~TfB|TRAv)TLTx^;IRy{7n^vDU5fjTagsNX0(6b28~$
z?F;D+@YO!pOaQFqNkUJ`a+}`MP5K{OWnizu{Bi4Q6R%%GaV5Ua9_-Ber4gM2&dt}z
zMTNO7WT9qR8~VcC(_^pzoUi-y4<>!B7<Kcv4=k$_PbE#|;@CPCPiuo<T><Vpr&n#$
zIvp(v0@h{cq_;WI*i1~}XA8)DX}*AbY=ttjI{N&j;gs>mU8C<HF-`ZbcpLMkixy){
zK-Q|)<%vV&qC9qKFTx!loP{Q~(#ea7*}%JRj8W>SaKtt=n{Q3T+<T`a#GOOh-CfE#
zS#4Sh?egY4OWT&ks`k}nov!-jLk)^{XIc<QApJEA#})*91{As#*w0S^iAnsMNO`{%
zf;R8PqApUk@hGTZz-pZx)sJ~MH3S+;(OA5jrUNxi!Gq?nQ4=J3gw%sPOppe2e>u%I
zTLD-((+|Fa?niXkbhzD%S=a1T@T!5^+ePO+W~DcNo|-KzDCj7V2}HBgahJ>_gT6vS
zQvyAx#$x*9!M;jiC%T%`X|l2zDAR5=4q8i^%&2~|fw6Gt+Cg8j%GFd}>J6atV@6up
zF`5>jfxl7IC31qY%m_DR_V|Y_sYWF2P3MH1)LeYsmYTxksTW#6(up(=daZBnPbBt2
zD}<#BqTCWOrGtC0#JNoEZ;7H?6CB(NT9q9*`QcRCQ{G_?7*c=#E|V2!B$`vQDL7az
zHL`A71j2UG>uIi5bj(YBjbC4K59`+<(>jlD-^a)Sx(q+_A$;Xi4YLz$Rk6IT-B3<U
ztBqFaJqM<ksjgfgWv+l0O%3Y-!l#$&!7sMUFzyN6FV%%$C;GJsae%@sqep-;rfD$)
z`T+P_xzVg7C1-wy2>!_&H@dnF3xY2ABa-rJ+g1VcVvybGiuA3{eQOZ0Vb{Y>8#1LO
z(`$)D@2PC1Rm*p9`sZSOQmqu&)|Q7fKP#t`J?N^W<bbL;p~am!jtQDB2up49sCF3Z
z0}=t7^V&T8K;l4Pd9*!Z8&L7FaD&rap4h>=$gaX97`=@<FgoMcEmqOprCeyBh}PU=
zC`KK6o~Yan(mfY^8?nRYuYV~K#GzHaD_36c_9Oh$z*CN4+S=@Bht&sW?#Gje5>R>J
zYdy_hDaqA3((vo?jaz%t2_>n}ols20y9ZmaN+1nRuNeK6Q^8k|&eaW+VAi$Vf5dN_
zpUv|aiag^uR8O}sERwI7Dm<=u6R-<W7o)6DkQ>RtpM5y4P=tWblc}H|fkPy1KS+o;
z0#JJtO!0sTin};2Uhrz0hl!va@=~Xbc#JMu4jFiBt=vfDxnGpBm)N{8KH$S6%_lu>
zSyZ>G02@>22UclvmTAyH#MifurkLGZ{pY6)CaxF%YG`gIe@yk&Rj%I~(%1^I_M&~E
z5<2$bftfkus*{X4*_sOwEg0)xR1IOJk+fqsig7aO`~-xSj3zzYMG+$K)Q%C`wGb`?
z7-v}E|1z2}40}<w{qZ3%@4-IT9+Z4_e_0?`{S!hG$bvqeXL5o`(XQ>R!rip#QBe2P
zzBs`852%z(cqS9YeogmyVX8Ch(NPf(jCAtqs2G3-OE)=>f!klL;+yZ*HDGgl<nSQM
zOjE9L>`_Y7DN%+~)vfb`>L!G}iq8gvnNPE)sSRit2ePAkB{}0kOMnFq&v?~B7rmr!
zEsx1;d}}ock}<=|?AMm}$u`7{0YZSkhIjqHeJcN^X^b6EiUkYdxWqo)S(i>hm{xxV
z48J&sxxX6h-0NvN-!GLqoOYi9b8K*R`wH8iO2CGbUBO0P`-^Tl0a*2v1A<u&;IL7R
znO=FCIhHt_dzVVvt!1Z2tlvZMBgi09_)$>*=aUu4I+U^2v|1z!u<ECRww7BfR(Sje
z)Q|alqq4w;xyVN@qAB(QsT;LFg+6CG!14}vt-Kgh@B<?@Kmp}!hzLLR!^-^)w9gZ1
zuS}>taSPo#UpN2V(r`HYj*tWa$2k7meEn0v6)7Tv8)oXfe~O4PwV8+lEO3SDHbln~
zk%M;Ih-GHUBbuz3)jdDhSpQv$8TeFcA28d47%}*tfYEq0z`pd(|L9{%DBCC0Hwy#^
z#7onOu*8?hPy)ho2!wwjz*o1(?TK3-vQ>&PBD+8&HssO`V}*q>1eKBzwmz|7DkUX+
zw<jVwdRdyhDqA0#!OpLpFx!C!)F$VDs7MIm5(e`MO+55L$4D-eRP?n72knv{%A%Wz
zS)wmfN?jlZvP;npDaPeuYeq&i{*fRd#bIF$C>kL3Fx47E#3XAL7M!fhc?iwuSBs4R
zo?b*a4T6km!0CU@ko%)$7;;g>y82mP9KKNgjPzuyPFP~n?oI%ZB6WHL*E-IFneCvl
zG@;;af0VJMrUqJ+7{yC3LnU|L&QF^$#YBJD$8>ZvfX80WWOOrKRcrVqW6ZFJJ;xzu
z>gYfCa@GKH)0`(|y}k%^jv|4CkqU5|;U(PmEwemF-CD4~W%D)D{yw)np0sKjs^M6U
z<T408+b=tqbh!5#8CW%LNv<Dh1^^))cjzdyiM0}*U9gw26^G+;Tj7nh+E#MWI&iTV
zIQOA#*Cu1(uKh}5ZOke#`T>~rHX96BDJuQjN|7xN7^<cAKhFF60Nl8+FX9-*$x_Vx
zKGUWF%48vJIlQ&Z@k$>25wrhB*%6@u*~Uq{W?=%<F=tq1K*%IZ4`4BL5UhfLmj38s
zdvi#XN^QJb8WkO5mf{#fbC~zp%~H4Yh=Rq-5656r!>fp%)5O)`0iT(Uz<)jKf1;)S
zuC<!h6T8uok-68z%s+ryu{$?74zhHUTNDz(8Ae-3We6}{nIEiGGQ&cK-#jvC01oyV
z;H?hUdODmQk2Q4w3EWZ}<stuKdmC;($Y}29K5$GciOU5fP7QVVIGfK_eG<{e4TJ6N
z>YdR+fQOp=f00at%K#D{(Eljm2?4^?#Iety3WLRzQ+2z<lq0&}a;}}F!Y5O}eg*Eo
z0nl4kK<f=%!=gl1{Faux_6#Uz+e-3Kn{Q>VeR@xJ9Z;Wzu_Zh^N_3M^#Y!Qa)bGm6
zx~Pbp(5^c5^#30g=ihZZ1FPq%KlW)M<$g{8UMv%I)4nCa`g&T9BIHk-NqfImO|CI1
z!tAA<2(nz@&hkj)zyJHFQ+@}iIHc@ZK4ho@;(WC6U=>IC0(p$?r|R9~|HHTZyI!^m
zFuL@=(2!Kuft<jDiB1Ukb_P+x6fbzp#9ixjMNJ;+ZY^pSGiCS+f1v@Bs4;y(F#-gQ
z)EFBgyH=_O;2|0R<V*wL0y3?j9Q9lTfe{?~q~1NfCx5;hr)>v)YN!{qvM2Bf3HCU3
zfFn8r(Z~c_|AUQ*RR3-TtRE@C!2G`j&%Z0R1Mbz<V-x1**c9hVLo%+uLZspY+a;~0
zqlI@_B#W=^_sbq;((13iY2Kqt*Hl9XWC~Is*ExN9{}0RY@A@8pP$%mJAi()(u|8MS
z^QxD>$Ot6C+S*#CScR(@aQ2Kl*KJPAqyq`;J-Oef?)o=tvQ*u~0arLlR_a29`RQ~(
zRBzYQu*kYe!UZ<I`d@a%-+zptJmlwxQm^`fxcLTX@7y4m1unpd5-9r+VsTOnBmn>t
z&<<o9Zvp0UHs)!!qGE2lJut|1^$mavPkxUg&^Lre5mVd%G~Vk97I5<aC*R={x;#Ho
z7v|WoKk9=;a?gZ3c9AANn5W|gD&XgdD97^nohj`oO|pIhsRvxR(j0Cqu2d*RP&&yd
zcE83d+|jhFy7q?_!i3c?;8F5C$8*%sd1RbixSpC&{O;@71hg`)U(=^MoXh0TR&}y?
zWY>!FOf+oQN57Wsb};=vq`h@ilx@HEZ-Rmnk^+K&64C<FpdccWic%8NAUQ*~NVkI0
z3?QY%&^>fB^ngS6Ff<H94f$T<z4!0kzjr@-J<oIR|5)I1DRW-u^^N2B943Hvq6zr8
zf`CoJ)Xl-Tpg*~OcmMXQon1(0j3X*XH8l*FGTy@LXD6!cqXCf|QcN&%QX!Hq5SUKf
z0$#2~^g0I;v$!b1O7NQxrEu;leRtG%S`oXH<o9LS^$EjKP3wXg<EMYu`w5I?hK)F*
zXjPM++-?mf?ZhM47(=*2oE(C<@24Cr0nD)Wj9|tvCg4JX((}~y*a1>lJKb-9h~_#l
zymQ8Z8sjvdy^^>Bgb4ljf5pFAF9M!&CtW&w4?uBBp+>399RlzpS`?oN0>@wE16eNr
z>P)?a2e3JLk(aHGI#<p=sd(^{V#OIl_n%p-e}n1$ES4hanmsqMe(Bt*xQab22P;Vm
zC<0VRc24?g9MQFem#Z+B)8Ra~*RJ&t76FCt-!3Xm1KIAy=ZGW)@>|gr<;+zs`RL`4
zq5PQ>p%0%Hy)Ftz+Wl{yr-DJ#=0%m<tGV{#sVQ+08vIhk{EOASPTWnkYD%)l$L^_r
zuCUwiJOI;~@gJDZf3+S4G)8sJfDSy>5|MM8(QD@!450OGNhrI7XZs_5S#^JIlFiGo
zTLeFb{08I`s;q9_|I7Pwyv$;8r<K%|wF|#H5k1>Oq`ECQvHwA2ex`ca9|;75=xA%d
zEF%ihReFAW+yx*^fHvjt(X1^?WwGi0BrSMRkqG-Tsqc|o0iYz*wn%j@ZYH4r(0F#x
zim$aM?dtlZxW)Qz5_1w?lM|OxS=9abY{(k1zPJzYg8zPf&o1Py!5Js+ycW=SakecA
zU{+F#rvSuU>f9+X35^EC%7au9Ktn{YaL<_%UmTtS0`%u~E}J_*FZZG8)8zzyqtBj~
z_+_N8S@LYQe{;EJkoLd%b>LNUz1sr8q)c1O9j4EK66D+un1EbuJVo(=nl2N50Dw^~
zc!9gF*!WrvP~w@y>$`v^FZOa}GQ3X?+yEr&ABbny7v=;pM{~Tf;hrB7NPqd?ye*XJ
zh?!E8G`{sFi;%Luh-lJKlM3X5R$u94aE#!(50uT&bzP<AUmbQ;c&owWdGhROhZLEv
zuAk!W^DkULIw{lNG-Lm#hD#(K?qqqJGf8_DBJEwfP6u4d+qGeChg;JiY@bM7TyNUN
zDSFIfy%6KiblMcEQt1R%&j8jK{ivZB{M!;}p#R(d|6gwTWd*HZLlL>Fj~7=r@IriG
z53{p}P+Q!E^SRF2ywQGR=>u3`n(<zks;O)xnkwL#pCHG_&;Bo-3Vn;!E++%tvP-4v
zrTpM1l`tirRx#(bx3?H2X`>}~LnspQ<AxFdepkm^X49<)lror1rn4?@IPMfXAi|h|
zdRp^?e5v~%+?oHhVpi;`*!7@5LacTbX7^#)vSm<cKd5VSH%%}nbfE*ceYYy+CBz?^
zj8_Ad0uo`Rc%4hQrwmda<rEPN`UgOt40f~yfQt}Sy>s9IyJQaSda8uM0Wl=<HseU8
z?R%%qiP65A@!|B-!5d%zm+;>{n}1qSKH#{4{V^Cn82wz7K)R)irg=rli37uY#epE?
zsaL}JDfx_#LV&p~71&6%@+2lQ;mBi!aOZ*d13frNWGBbHRnL||fW3sTL>P`Nrm8_J
zvyF}GyFH)rig=*sJGvSru_X#vE$C!$-D<@%9_wO$im%&$_X1mB8oVRLuJOzT`MuJ9
zw!!V<>`))hPW)C$4j)7K?^i`NlzR%g+;KEXrQDK%KuWME8SgH}j}~#Q<_1&r-Ei{l
zTMEzyYJ(LCW+L*RH)M69)h83w-V!h(4g>xgqp=HxCMD5H0+3Nr0BhvX#T@(rp(4^w
zKyULYQe=iU*}g#xKNCa7xZ`Dxz^tha&o92lnJZ@$uPa_pO+xL}ke+h2y!c8iaC-s3
z^?(1^#(zGi4HF~hL~e$wM?B{z&~wGSIcR(Nj-l-ar;yAkZR;!}Nf9%8*iiy8WGa|P
z-fE)t-UY6&n6-TqKy~L60?q9%eea)2R~rmb_OVw4bywrBJk@~V=J9)RSd;!Vnb}Yc
z>N`YS4PVL~ET3gil$X=Q8YCNV75%NQP^nq7(=5>Z1|Q9Rw8D<OcBxiD8Nn~s-+@27
zM;E6_Uzl_i8~x0;)&@d!(iLN;)%mUnIz4toxhv070Kv9eq(Co|BIWsn1Rz$57;gaj
z0&kJ(KgGR<s@Qd4+W!ooNDKI!x*aZL__GsVF_c~v{qc~Rww|q?mm-Ph>c4v*@?Pir
zZY)Usofh-H_vLeQ+Vti`LcRhD0Z!^AdVg?6^kI8i@NEh|tynpZo;NPfrYyhy?0@PO
z>8-=rD#7lpMhj3M@yb2N8JPPKFdIVKowFs|ED#)xn<v{kw2>>!0hflt$-ySxq*(Jx
z6g`}^vgDdTOwFymdu8}ajGmp<;QFxjV#VKi*E(|UBH_oCe8vbzeE6QYV?W)OaGKU|
z>$c}MJqItZt&9mU866eR@8q?oNs)igP_?J+axr`qYYYL7ViYh?;Ecojg!J6z1Gnf^
z!upubcY=>*4ja(RZCN`@H_9SpcDq>dT+n%7<>JhjzKY1Lqq##&NX*T2<JNm)BSn1V
zY24+1w?K*itnn3S&pG16E$*MMh9}%IwgA24<&e$h?Wb0o=lfaIK6W@uks)@gCaL`n
z`-0HHAQ>`~7wY_3Xcnn?3cDql^Ir~kV9Pci&{3csO{5Gj0SrNB)&X!&3?3T}6gGt?
zE{;G+-wND0A^53c<54@G1K#e3w}t=>&l9KZnd!bExsq0Q58a7y&cD}+>Jj-QcaxiY
zlZR#8=+h3!S$Scxzfk08jW*KUyro<OiQ**AozKS^{<sJ^ce8TfQ`SAAYRIF0e@ewB
zqTJl;@FcI5u8D%_Frf7@3_SdQFCTXe+!aN-T#1Yi^KT#4d+fJJ9Zvb<oosgEv9rLO
zUKAL>PjQsLx2^b|Ea!-)h-#$c#Ue)l{&H(r!>OX2Mgy<2{ySBx^8E<;4+K#8$02Pp
zSDC&twPNXaR;Sd4u(32dBh{UluFmMW`Dn)WRwn7FN^kq7(fY|bzi3Dk>z%`u&gg}&
znCQjC(1FFXqS@f>P|N+XNSFWegaZ*baDUZg=S=6sme}i`Sv~@!?3JNRLcr-l#{dim
zq1mz1){Vdty&77{pd8{)<P5yl@CC<mXMj2U(*HC5{-<x<o7Z*+qSa~8bIj=4A%fQt
zf%B4*A|(<&h((dHd@SQw#k%TP;mOumCPLb{up~L4&S`?U5MmctpJo$S&p1o0AnilB
z<%4#4+gX<4aC_X_<>A)`_CL5Kp>ArGima6pTfN*^!K`=h-WdUJJg6rR8ez({d@yOh
z4fJ$_SzUkt5kQLRf1mRPVrInu-ARXX`fE$xC{97FpzL#lTYLqb`Du+|6cCx)CsL00
zUDj*aT~=Wd2OH(NH~OSX-qVz9>CId(giKJKIN3!mrb*pZ^V$&wgCebmmP8<2;d@f}
zE#ZE8w@d2im%w-SPvj9(HIAvD7U0X@wG2T11k-?K2pYceD;KiXFp96qBLLww;-*T_
z#+#ECV!m(LO)E7Iu*1$$+gT5*a#g-Ym@szJSc2aJ8?LE;>xChBvqsaPNSlM&VUCW|
zQ2ZSsVX0AmQZnN8Z*OuL3bv)d<<IiO&z0RvRfdO5eMNG!ve1{r6s56o19Bs=2|oqt
z-X$e4=tMPTzN9;IT&4JZ)R_%KT}(Zdh-9m1HmiKBMK_?t5}yDaj&+JYy$FTr`M{Hw
zQ`J&u)7ra8S%hv_Tqe98w&amj(Ej@Bi#|HP3)nb;!hz_EYp9vAY=qqyc2%m~rIS(L
zrI~TsrQI&o)p?!JMdH#0gWL6EwYBI^tjOcbpB+%%QtqTq(|>B74RJUjv*9C#reHo4
zkZDmpWXhRBaEta6UbI1@@AQ=(<bwAN8^>f0mr}*D4%?Lvz1Wgap=KoFB^u0-3y#Cj
zL!ShB@grg;6RJ8K6XT}X@8)JC*=6$Z*#}uYK#(c_)o=Vc^8Y*vu7+~ULucR)R=yX-
zQ5ul@<;vp&)4CA&i*Wg;?m&p0Ok&70|Jf#3o;wSaZceebOJ%dUHsuILtP`*F`>>jO
zA$(P@p&%`knf>HdA~o>?h^%dBu_>dXb%bjRGI8W?I}5ql>j&<yUSg7DqC$j97*zTt
zoRD4CeaZBknx;Rpdc$8YFCHv9FY(Ciu9wCQBwM+A{lbRgVwiZ+L<Tha6`&6zJ_;1N
z-0|?nJyvx;a2-Q;dl%=2z^BtDh&`n%=yjbZZWfAfN9aq0?`%xbDM&kt)_UzdXwc(V
zJc@qJj%>pk>TuT+Dy4D$DI5Tu6;l9Z9D_g{)(o#g7TqjII1D}`Gy)pAW<W!O*5%~n
z5ZmQ$?9LjTvdT@Ja!tZdqNg}2yuSDV%z&efvH^8{ef>9n^KuEZ0c>2qyyh_s6=ZCY
zt5&tp69?a<4r>O)+7sap0bhVxoC~bL_WrtC@kwj{Uvk|4IMC<4UNkyd)y5$tRU+Dj
zL~$%sOIZrBKd2|8;1nWk@LS7!M6MCJU#)&X0cvZuPI>JFO1qW$(iHwn0ullvxJhq%
zaeuDlp-P-0<k1~Dy2aC_CkFmgB02LOiw7L+Xiau!O}@rdG`xmeis=0y*qE~C&GwQo
zzgjd-(EtpE(_%FpJHJ>8EGN=*QMSE_Mm>EBhSdgnc<Jm@9V>786C0jUkgU6q9juhb
zhV)6v+f>TtvmZXsht`?ILko+y1%ns%;?+9q5UAA6QRXkFVodDsziUw|ZaM#T_<@0w
z{QPFsY*g8RG(En5rGM=Lw2-bm@eEVJfXNYX0sV2A0|F@$U@OE401SeI89i?B)$he{
z@dL>NO+a486x_;;=oS;B7eCf-k@0y<Fq9qZOnO&bi$xzSmdIw?z1i)wiV|{?0aBm-
z!LR;*o*FC7*VdLo9=;G*72JF75xRn<Z-4vT#2*q;slN7(qBFWR{qk9fc7XX52fv)*
znpbHrEM@0Truz3t_t<^yqa1mmc-5$fKl<VqdQAfDe8Z+XJaP;85RTv54V0aJQN-9O
zFdFW5SMe`j7r;7oNIul(_z~^?u}YJunxB$fA@(tOXmJ!LI9{)4gB8cKYB%Yo2hr#s
zFnWC^+gkt$;#*d0svc)D9*J-jFzq3O9k*M%Z6P2lLQrBxMl9_XgEQX*X(Ras?XlYF
zG0xDW(XOz>vr4~MscO2U6X9vH&3@2sptjArUbq9FVPEKX-ENH!Z}^7&Vxqx8C|kQz
zQu;F9`mf>||BU$HHy+m6_|7)9YYOP?OMpRxG0<(fUOmTK2BE|}OmURoDYd;m0g3^>
zc&q-`tDhn@0quwicKEbRIr@7=^j7D-Q!Q2ERx3&<X7xsn@SpjpwhIuTNb(;6ivROt
z5V#`1KB%XZ62fH|5XmlE_N<Mj-FzvWEYtmsul0>N^crTBooO-h@|il%`NlQ6z{D&N
z%j?~n&wFvkr~QMRBJc8BF77z8Jm$ND<p}_Ljn77VlLrWi<xu<rf8ILXMR%2MM;J$R
zn6VSl2bLxvE%kC3v+<YCtW;d*vX!EgM%$x6;q~we1D-taXabU-sqCYLQOCmF1YIer
zSQ^z13u5WkPHczKNWqYs$=ap8(+wR-u><C*lNsCU`OSlB<H1IUt##ifOucH$SE=YW
zzX66%(-1q9O&athtZRz&#K}HV3St$gU%MToskR;XTWuq_9x}yL=&=z_9DGWZY2g(S
zxIYTk2N!}3fb7+<y8rw6ln<K1AEH|&*9*#~KFI*?+S25HQ+x!3`hCqN>yIDfU{2-g
zKPlMaar``$pR-Q*nOxFTDx*F^yetJroTZ>SLuM+u>xw2`zn^Rsb?ghpSl!I|YlQx%
z^Z)1gL_H#dygGAoAxa}kH;A^sw)e<=%|L<@J-u+fZxkiV^RWZc&n`rcP5h1=SFU~C
zP_w3tC<#%u$hgsSdU-cYGv0YRRN>b}=nXSn<S=LDp7cvGd!fqbwU@z53@M?seTtHG
zv$ek!FgKC<B33cM*GicJ%?;d2zYk9vny)W-aqs5yXWP+9r(BTFq+qBG5=ZFOBD@(~
z8+~aeNbJLv2=s$y52Yv?-BF?pn1$D(S?GDuBb;0tNjvjEsIvqmp_3SGVeMA1RP9Ef
zrrKs`ecBj>r*sVs$v4~11}{lf-!mtMlLgFJ<GRsUHk{FNf60zT2xq(OBf6YzxX`$L
zMsu684xB0mUJ1RF^JTdP`XFvMaUlG!aR9zw!uv0RnS5xaJP+EwFw3PhrHDD^aM60r
z@630VyLUdC3iI1)UEFc=dMY>~8q^I)T0Qg;LvH6u@i`f*fUqbQ1H8uX{D(y0Uw(~K
zKDH5CniB>kP-LLR{9;9j)JTRApVZs#1se1hwX>)BMELm=e3<!heJ4<Z<aDvS(xW@m
z&wrO+wL1G|w^HG2p<$;FX{Sx%HaE?}e~#F<@1iq1G++KA6_0g_Wc$^g<<m1{I4xz8
z8e=6z9=*@<jVAn4N_1ic0@HrVx6_PVkU`8gH+B2(4CmGly4%o5?##<od+%+o?AQ_N
zogIuGb$)?6V4A{rf~C=m8#9fazlu@<W?_3=ryID|q2omF5>KS&SOo$ymJe?b&FY*#
zLWOrOWNE3j;?DXzv;EIM`@$TrPefSyODIeSvCFv8rUEqmvb6oSe8YwyAB~grb$0#f
z8|5c1Qa3!MDhLm@U=l7HmB|c;_Fx0=0P3*S8wMPo#edSMqIBuE5k2?yp+F?@UyeFH
zvdXvM5glpDeRDwwseFa0*Kiz-7I!r%@xnJwjUAw46l<Pr^!3%k`2ue%44^*|#lW%3
zkZ=g#^C<G06MwIOyZ4CcpumlvPg6Dq{R_A1K|O3$-T2ou2Ok#n&x0nH->7Gy2+O{L
zslvu~vS->ZgS>tXxG}Gtd5j!~mxyZluN~8I7bN<V0#@BLzapcx)4SRu>Va!|RLcb6
z{OoxxKYnTku&w`!<N-soONUY%Oz(Zok<}Ou;W>Mv5s<?z@w!_y2pna%bD!M!_A?S^
zgL;WAq_d%kN1u~djpbvuUC-_slb!R8uWlB<lWB!AWjuUwcObZOE(k4_!;Z<FegX#E
z9kfn?H@amOvf|Y`2GS&m2%RLz3hOrfy=yljuhwq)``2v-vwARmHc68r=xp4fPZ?Qv
zOF3Ta7Coz&T}(j^v@)IO1vK7<F(4b<rR?tE45wt4FFT!-3mTlPJ-n47OqPj`Xvqqs
zC@FT~r4Fw6iEff(ne(1MSWik^Igyt+%|YKBtAhKC<xUz{?oZ9QZDa-u*v>y7Pog|o
z>XTD-bCNB)B`@%wA1yE*?I66*6y{MTY=E><mV-~cQ0kx|-Z})r<>r7-(P=gizu7$+
ziuMUm5t*i}+pJ8czj+AEP8~^YrC#j1)%Ic$i-K7orpmG16OH9R3VLg^V^i^IKH^oO
zDdb=CU3_%tKhJ-2<a8^9%EuJODQ2u*cHZ4ecq4o-Hcz@U0GDRGflOOK6tRC8{CRoh
z+`}~jaxTEui=(74S8M7Vr4SfOjw+IO=(znxrf=vFi^%2{-dwBwhW2uOFW^$|e5DGc
zvoJ=bNqA&rz4{}aK+xLUWEWPna~^TH(-Y741J16t2pMvnugyTt7oG97uNI|*8cHA1
zGfbh>H<0c!>@0`tm81uou;4K}maT)PnQ2|4XsG80v}&CZn%&h=UypPROy)C<rA;|l
zQ#`^p$)ryYT7wQlBnHit+1W`tdrxm@&?vY6XeU;We;<)z$bKcpxcJ3$)54He0`e9y
zks^-w4>;%CXT%KKK>8#`{v&~b@!gJ%EoN7Xn&t=2@0_?uTw`wpUkOrS;{K83!!t!}
zK{>^67a95aJ%N^COGUze4!bq%i+Kf2GtzfFfkI?@gdJdFIkLjp5Y~)d36CK;@b=F*
z<y7&}&~DGI8bZ8LlOmWB0>ss%(o-GMG#*{Y53)p%xr_<}@{u5$_9YkT5l#Xo&kX@~
zgTqOC8Iim#DlNSdQ^O#yUpd%UK%Uy_U+dL4s((e^bW?p0$4s}75<(xxqiRKy$*Uc4
zXc7Hll6li%mEN~Kkfq`Ls3Xn6s8MFIfM%ezTdKcF$D6lB-kY~0yT~vi{6zmV*K9Sx
zw9M8cmap5J;sP##X5m%_{KI-Q+AeO9*{U210_f4u^uiYvH>w9{e1Z;_VbN7NjgiQP
z6@SJw$?#55kM$LaAs;shJ>d;QX!x4^RlYXd*+A>Ip^Qer#4pjUYzhbEf!Vp2<YB>k
zqO#Xeeq$%0^(EUHCv$^P!nqaT??gNwJj)muuU?9jSxl;ecJyW1HG7*dwjdU3=W0Uv
z7Am^E1uQ4XV8U42h;QrXmi?j;X%}~;4fh`AjP|y`jst(9%31wJ)Uv0R6ehn966+D$
z`RGGcxMl9gd=n8cJja?!C^SW0`7pEKz9Nu8$m_9he{_bWnJYPDYNsJ)BjQ`q-~I8@
zx#Kj8A+C88flB)8wL1Ny+l(l?JC@yKXRdi=<QD}#IRts!`Pzc7X6cR>uy^Av{7gE8
z!=memjsbJSLK{cje4Yarb-|J&uj)iwD8;mY;(}FM2s!lAIk2s32ci_W3_gj^nwNgc
z`!xupn!7aD@AhzE00rhm=sQ)!ZjtX9JbUrnYX@LD>wTgB!w&i9so4}{M51R3_vD!*
z$$f1o$ujQ#{;*Rn;;>Mb`{BZl@Iw_)L&w|I86j2cOsaAZ+jvtCTU=5|ce|&Djzgu2
zf#{35pU}oOl~!?<5Q|J{GE5h_-uiMaZ>9#dci_OXcUW`-qyK9`g%4xp8MOQrOdj5t
zI}nQbQ8PXEDeU}owb5IJQ*@!V!)g=U!6%K_VT~(yK3>W<RIZL`L6`l#^Y}ea=gF%l
zhhsD{z6TuS35_kFdetVYqSPqIlhk116T^T;l4{Beuf?=#mo8t2#hs|?#k0wswlM|F
zX)YjX4>xdgoJc-$=%@86v3%SBP4&;yyS*l-7EpOZmb~3>%Ez$Rkxf=$*|M6lW5bNY
zS>v}_AZuvQ3OIn_IkFH=VobSq`a1>{Uf90dn9TRhpM2D_B0==RL{f4?U}_l67?CQ_
zuM*z^6MJ||VG4Da2bI@4Kw~tiSHXS%xiJ!Ic+Syp-kad3z~oKYi0iX22K;Kqc(8|~
z6#&SXbvD9#o}c|b)6Q6`iUoWeDeE&m?oY-ab#HSh{kp$elNIS+a!2WzE09N)4%pJU
zD3gS?TB&u2mhda+m9Y|5yXtuY1|VJYk8kk+rF7q2N)dt8fq;C&@Y-x0O^(uo?1xpD
zQ(~$lz-sn%Hwn;*mord40;oxMk;7kM%X<}$I(VJ7`CkDX2JGd=Nt8+ONp^e=jIc#v
zz6{>K9xs3FfneZQCRRbIf1P5t3Np*Ui?zFH(qhZa`<T_l%#@0P;hYrDTECjmgg4E<
zEIX}y+^?uR|8XSZwdi6tvBL^h!6@3!)5IdISGCSK8n~|yfqOfc(ep#Rs)R{&u1g1G
zV8cJtP?Aw33Pwd-Z<SymOt^^liW0H7)v+FH@T{sN6*SOQx0A+<T=>yIvp6!-*!r4}
z(cjyXAw<7+E7VqEGh|v~JE&fKJ4#bc&?fNr^g;VM(8`?EGKF|cY)7AoGkub)WePHc
zIS8qG?!P6^^ljhYJ|%SZc9RhIJ$_(7H6*TgZ6g=kVPeoKNe&kSg4UM_Zru@&+hw`H
z)W3|i&UxCKY_qr?le4Oq6(Zwsa9!8H;qlF?kl&O7k8qB@<cZKm>vL8uokP>cOA8gM
z<;x0Vg)N>Y3dVyD7_>Nf+z%Vh$LkOERlz9Sn_dDN{Wz2FpH1(oCfNxB85x6Ffy=<c
zco3lTh`6#fhU)=|ig>+i9Zs0@u~ot81EvVCki3M<!KDU1*v{wVJK5i^w0b4xE)RSm
zHLom{#x2E;0Yf4E^#=;=`jEkq>Ob_Ko;TNI_6Fs)0u?jdfz&Qu#E--&5v%8#FuFsY
z=ay{I8Sp@s`bJ<P_mznv2~LQ6PaZs#gD(QdyB5@&&4Au_ISjD%?}=wnA!Gw{X5wEx
z+ntE7f}ZRB94vTCLRAX|-3q5l`rcubp$@ogflb%J<85kiPTW%IxL>Zat?VMIJ2MUY
zxU@aeBTPKR<7Ig^F)R7=*CTTc8GGZKf?w8MXO&(gl<aUx$}|wG`l8-$jAv81Y}HD*
zY?jsm6X0UBMq#l6T1wt+%WoPoLtH3+Z$+0@?9IBoy<P~hVf{6W!(&!eFBGLn)e?x0
zDcZ2{h@JsAH6*%}*a~`D<v*Bob-~Fhvc4`X^rvJFn$Pqi7GyVD=CC@~xG*t)0Ydml
z?qXP>b*sP`>zqW)wERrW#0(*sj9WIMcg8=59{o7x^3svGSGI0${Cb*U<Tqac8oiW+
z@_9q{Q7LQiD^0Wzjt>!IMTe-lzBN#q!N$NdF3U&{hu?`Ed$H)VP=XVpO!5)#@Y=gP
z<K=m6?c<t!bm{p9oF~@vzSTR+WkPx5t#($OAmF)F1_BQN@o6^=(e)l+1&(eo!jClY
zahoZij=QQOo50*v)#)5?;yI8iNmJr>MhG7;K~AG~2jxir2xtRnxKgqyY5S7D`vc-M
z8dAxfPbUUgyebUXgb^Z*kAanL^;WIdhjp>R@-N_vyu3$;g=j!#NrOLZTl5-4#{Hzn
zle6c)oe6=&kG>vuD@kfZ*%cLZdU_Tn`Q5U7Mz>wNZnL=ugz!)i!oFx3!PO?@4fb75
z?Da<rSPeg57ZQ8KciruAbe5G*_t|?OcQ8jxIcHu9MGAy4alc;$QXo$)@&2yWiWnJv
zHE^3VxQlnYAYnJqn861@RAUFkfN5tAnv%w5$vDcOgmoZV-wc%m>t|5<f{Qri6hJ~l
zk>vG)J{S&oyZ8WK-M)nnOU%LHD|E5CQe#>-;Exot@AH$ZM=M3}#4xCL#DkfU^w(0L
z0VscGsv3i+jUIaf`IkWrBUNsZulv1~y1Zp)<)W@mzVveTrefn|2yWazlOS2~B6C?S
zC3aaY=hBs`Vb_(Z5ph`udbBNA{+QiZJyx29@vyad8k_lRYxI}!!D=}Z$(qY0pMwpf
zBNTCG$cp4;3)q3D-XPjbZ}_698azm4iCx~W#-t|i%zmv}{7TCB3|XR&X=$SwJcKvU
zKJliVxj4S+C2r4>qt(~cQXAOi4u&!gXsnTgFC@ig3B-^D<PLs%3J;KQHyN6)vY>`$
zK+dF;&^)_V`LrKt{j-B`VW4%!hsHTP(0YcQyrW@?HA-k*uuxB<;l$=#%CI0C?&y(t
zTd-=zBmZXKy9m7ay3ctq{h25*m`O>%uSGW6C~|4%<-Qhv2jEl?>v2<5aNrB3xT&<j
z;J{^I+1rla0|ziPhI3aoSu&fPZ}@l~PP>`}-*a)8HQP<$_{=HJx#H%PGfH^x(CB6H
zPN@W6U1#Y&7_c;}R==?<{gURk5Nby+pRv`c9=|Uc1E+-A2rq@lvv>eIrhbkf;8K$n
zX(Ee9_ouEm5StxeC#J6cEdY&WjYQW_ngB6CXz@TYxxVQ2#m;Eg!yrDPX^5F5lkV?I
zAjOj5v8tG<*|)|SQd<6|iVo~dHda0VElKxxaM!vTxNDR9WP?foNVYv(j0Gjw&bn+a
zeV6e7qN93QqI(1Zz#*bK<-3}a@SiAB>D2+RcvDapYDq97#zy{h=o@gzDhmWv;S4y0
zGx#OEr_<)^;#cSEJ}z13$}``5_YF}`YPK&X-0XC6&`IU(72tdCu>AIZk|#8M&))T2
z;iMscc;+?omZ!<MyU4|e>Lbm_2TY6ZcrJA~HSdeQ7Wtn#LZpr#8#)S-Y7#W12-ALh
z1?Cf!w-1Ojg-(lahNPyA-Sm_yAw2N)u-_PQrr8*D_S_hDCM;i@Av$rLCJlCxB9mEn
zxlgj@b3e*`izDS^*y_k}8|z?p0e5Ikw%wVlb6p%Kt^3f#USzVD<%${KQxkCPcjDF-
zWg<qyD|*D<X^sAP(I2W=VOf&${&rqlUcM#Fa4BFvM7KBIQ`GRF5E{3hA9whn!hz*v
zAk8qc2D@~(;i+5RS5HnoGfZu*ZS#-71VP^?LE$o_K3H2Z83VQRBx;Ac6q$hvU1(9|
ztF2HyM@{Kjb8dt#6tR8_C#Mulau2xfjPchUA>sSua`*;7ty$A`Ioe4dJH|`d06~hI
zB2tIwF%Q{Pe@hDu`Fm2%m#!05?0ajY`72jiVMqCdzL&F(P;!Z#p656;2-rl;r8xFC
z;~h7EFc$fa@$JU2Q{o5tNS+c&rYPNf){!>>81agAZp6BC77vd&AYxc=z-FbtoivG-
z(yImp4r0&nsw|Ohc3*ikJn=<ypR=YLEx4X?@f^m^kQIsp4?gTTTt^PHgHGUljT`-f
z?hJMsaNQzGASJDSz7*VLhWf$`=V|2zQX#vmU-28*$H?_7D3Rb;so+dvs44^lWyE>}
z@@&}4to_~RY<!wrW?HPmv0x2b^it|+jAKxQiFneM7Uy5~@8ZMTa28}YEBpaYgG88j
zb^W`sVnYl*Z*r=bD(Xc;7iZ0nfmpemMR9n}aqYNBq+Sdwk&3rZ3}4XH3%kUqb-MWe
zMPDOHQej9@QpZEYZ8~nW)RJ%s(yZRsgsy>pM)pD^D6CuJv(?Nu@m}=?``{0Q^xH3C
zdKaIbxp{s%TA6$1R$y<TcX1q9ZP{cdx)5kmY*D2-R){4TtC>_a-jJ;maoH{<1a`I(
zB$Oa`gLuK60SjJ*6N|~k1GIh1biG}tfBjZxET2@UDxXXww8-gMG}1HOA`Ojs(?~N^
z=gS9^o)g~lTy(z0FrwD~9+Czov2PH#CP&!4A96d!(R=YW=emE}QpzlCQk9R~apu6>
z<rN8a$Pg4p+E$zaOFI)#-<oWC2CkQzX{MK(iQ6mXmVPYSjJ_Bq2;HZDOk01_bAlU1
z1{^Z4@S!C>9flklf0?4MS#G|HswL>De@OF7vZ1eBFAE+-2Pav_;&+;VZbx<@&YZD3
zAmCWlY0CLrRWuUuKHhs+=*0kX$&7pgp0>d4s#4$BQX}uYB8LHt?=^hmwiRQ*p>&h9
z{b&)+>U(>s=)R5o#_vM<C_u80Z0U(lbJ0M7-qOBt*!2v-18)l7SGhRam+qIUBtJO<
zUb#W6(UQZVvP8cRHYe*^`b{jQJxGA!>QufIK#EnLAFrnh$V>oE)XmykfOf0DKhzDf
zQ7UheA3*n>Au$+gon&9O)NspLze*6&@!Fj;X3PV=zjWShl7r-Gb}O_xuj#k1uGSYb
zy&iV>1%WdKB^ao6R5&eG<cMW`l?&=&HsqjOvr!oaCn=}(^4-(A8HX?WB4^XEHP3w+
z0m0s)@G4f%{CMwda1);mybpL+yEBdjJu^6h*RV+q!(z%s)(mn73~zjwOLw!Ri*lyc
zYL~6zM8GXGs;5K#JZ(#1n4aVutB(w8H9s+@G3e^vO0^|jX`HqVxAa{;(Bi{rHMRwW
znsmIkG(~a<h9Yaq>&ay%r6f?P_r{SLrXxWFGc`G|a4vl~XE>lAC|H{sYUm_IK<Elg
z$R#$Tf9va-G+G>V7)-mg+OhDvZkCfCxaU!ktoq#Oci)$Ui5}m*b#BQY1i<=VFDfX_
z;12uU#GY(YOau(<F8|hXk@;@x71bj=%+^(RuE9B(7UhhVv4efv$1sJ0^r3-zA(8Rx
zbX*5t;f`0>dZg1-j$T59P59#)1TruQcfb+|@CCP>Rux9FBZ8i*<?#02Hj>b!EzSL2
zd5H<?7}&?P-AoI)DPrP&5hk;3>fRN$=xxN;jYk^<EI)bys0(c(0TG*Ts54R@Suu#Y
z{8L|p^O$4f$=HXhBK%~&R>+nnAkVDdUJ!8p00>w>=TpCOCDWZP&;Xpu69a@0=A5Wv
z0a=j4tRILTF;w0A86Kxjq-+a%J82fLUyNH$u5&bn^8~IS)MPdTel2HqE|x+lkXANU
z`ok)w7-Ls2pUECEF`Nn)myq}bZ{y8{8V4ruOV(UH^b=khW98&t{}+CJqIO7U60f#b
zRKEA;K(QzJ50TEE4{2*&%@TrNFbYK-#-JdL6WCxvRD9fW!5cB<-FDQ2U}d?q^p|K6
z{f$=^4Ce25C(7K?FLrele9Z@LL%BKuhDjvi|A#b*Q_-|r8QAEsW#F~e!ShU*XRRnV
zHjZCJedDu3?7u=4tF!X$SrGLti6A4i`oD>wN`H-4rN%2@wYNj(P!DVXTqjh)a81?N
zFM+8`n2>Q9(6Im$Xj%yi27}`~8oRTdCwhYyq@w8vzp-SE5Xm~4D4*pUDZjBoDTsBw
z6wL8m(Sno9V`Um#8wYzcmS-c<qWdTf5OSriINb&6z~RheI5Y2e<kYV)($qmp(vsET
zQIIfzc}b_zj~EzQN|7ab^)@=_B{vy1+r~3;8z25<W%b&BiP>GG8prA1vHkd=w|8L{
z>VwX%LGLIWv?SR^N9X+@ylB)`GP-txF2ddZ!Hm1(7g$>UYd*|=%X_Ky0wB?Yw@)<2
z8nBq@;JlI_!ok0v96(IBpZ~#il&*6dw%KrIU6NQy_um<}NUI^tm~nEsYX(GVrjELW
zJr;)M2C^-NU&am+7B44C5?p^EzYG97KTp~>PS<bj$>*L7(+=1RjlUfT3v&6sAO@If
ze8Gi@lm!u)U8dQh?)4wL9p6~Ym+Xn^EfXgo>jCs-b=d9b!pS|l9^0N^#$dFtSq~-B
zHcp0)gUK8&g8HJz8O(indOhv%+m!jHMde7>S0Li9IWBs>HJA_MUvSo|X|tqpmN>Ay
z<fs6EJ$=mGS!I?Yu>55w)MsLSM!+|e{^Ah*?8*4vyb!T}mq=wFjRJL*j@!U39#Ajy
z+<6_(v=fX0tCZZR)DF<n_7AKPqm@C)NeUm6pO$R9i>~2Vi)RH-OroJDACT;>odbrF
zw`gulF_A_$lG=NpQO_hx5!6eshf^4cQq_eHPW@W98>{K`sWxwOrj=6a$qrF6yq3my
zWhk1O^%iw_H5G>CisP?t_i9J-ev^4E7Hr2rVOy7&LcJ}+pye;&Eu@>i5YaM07y$Zn
zC3SDiYIuli%V1~gi~<`o488kp=E-K5Kj{_;o>`RkB4wSSd-S7Bw}-hXZQ!}u7ENI3
zNFArp!@@?Qp8L23(&MUW7XT6n6zV{Jsb&pc?;84?h>a@AT|pP%wh<D&B2p2S79H<V
zc-Et-=8|eAqOtWJfUxj@Pf^j0S*4}zgo3`;O=%uS*}&rz`!l<@?;TyTkgPj)0hDK(
zaLFUtrQptuJIW@+RE?9S*2MDjMLzpnt|}1-8jeFHt>-s_^wb(JRrecOj)k?7gAVa*
z^Xw2%_nuCUJIxg-9}i#sWsv1kSjvvUuK1p0Y!2=rlIybK)`;8Lt?x2)Tsfz_p=QDB
ziqIIK)TwaF!es(a6`PnT5>)w#Tg~#(e*_Xk^M`zKS7~KP=MGUlg<w^s8N$dSN)E!m
zn1OSGsrpq%a;&o|0N6;Kgt8X;Lb?kOeYPCx0`ZfsA0q&009e6y`7p5nUOXcU`KF{>
zYxCs#O3lLouDiKiop48`hzv_Ha!Vm{9-n*5BWj&HiLHa1v8vTZrk<z!k>Q|^MX@{B
ztze_s+{2dFe5Y5?q%g9o2-q}0Y_;h~KC&6Q#%-Cv*HTWD#(Q)n{;=^*u$1Esm+b~Y
z67O>wH?Lle1T8@tF(REXDUm!LdDiyS5NRUVi`Xt0sE1Tc%mT({XOS4^NFB+EhUOr<
zJz199IQ!zAHk#!?aB{PUkxPmhJtt@-T}U*$)m7&EQI25c{{7)=O~lwMUPnXMZrwd+
z%zHkEOR?8zGh+y8GPP!tuS#)Mg@$URe!IfMZiTQdu%Fyx%V$)nT<3m&lap+`{@a2|
z-6ya_qpfN9GM#lniKz!g5;akd*~K~_NJmzfKPL4tY{L^1bY7!eqissO0Qk}l(A|mE
z0l5fMdu%{EmoV7@{Za4*Q>Q|5+)D0<dt=HNfmVt^4`4MecT>NPa%*=N#!ctg<*xc9
z#eo9s5KbSJTWM|-rOL)4lGT6iRM)1S3{@B8xKk1JaalzmV52}buG@`93zRUw%7jqk
zlE=eOr<5(s+f0oL3qL<4vEgw@aX?0j9(tc0mpMc@uF(`={^YMT;0v$^g|r+_?tP4*
zN-T_~$ML^H_%@7Mdg@6&R<V!>H@rz_-~gQTVx|N44{iVqKDj>(zWc5WC6I0Ev+_X6
zhwe(d$hV@s;(WaMoV%-8u;O;L08RfJB{k029o*@FfY-5v+k6>%=seZs=q%sj!-WM&
z8&v9`?b*P+FbzY=>%Lo#%ytYQR>;N1b+0EE#BWDO?|*?Mkjfj-UT1kiGe0yGz0j+r
z+L<AB&}n#lcHN)-_Oq4t9eGND+ktHFnWk%=9R#)sC#8I+P5TOq>aYfV)*i;DnAHo&
z&zMS4`q7517|$s6Ia1eH>3F$9RKF20uffR5s6CH9On7;fq>~VXOI4nyoFco1FR1;d
zI^{>j_Ph@@oaluJVrfw7>%t*}1vE2DNlUs6(O_FfG25vkZA&r5<iCcC59<0$&7zHM
zfwf^PAmoe&0CnQcKSWY92f=nKsS2^AOtv+PE|qkp@)&?&r@OkUa#X^fOFaLU?GzHz
z^4*^j<N3<P71af$jPS6j1VfyL6`}HRSLZFgCfxd`qBHy7Sc2zRI#yi{KfT&i;#Y0Y
z&B*|z#sMbL>!~mYLlvN?v<m!e727#_77G$0QmMR_Jb23HR;UQwJB|gx4DsRFNq>E(
z0skF*hU%X&<R4b8N+6pj2XJo;4`aT6en7Yi07ca^x*3U1c{%#N=O}f)BN|5LuYX*6
zj{kEjXl|>9(uG^_&ys9sV(T?8tv@ob4e-DzCGDllI3{2}FAKs}S1$ISIx)Bls(PYq
z7{>N3#>x=qW1GKzkCkD4L^lF6L6|xCpu3)ka$!=?vADa|h>~}xqYaS=<OZxATF2u%
zbwf{U)YT$FwlhOyu@^ni%*CRY``hc`B&s?lNZ<Ud_;;uR`;FA5-lNz(cM8Ca>E!M}
z`fH#@MvTx7@wjl>f<6-K`zGQ#I3@k{&tdb?BU(`q@A<ep2czeSQbZ~r?pQ^^lGknr
zGqWUCJ>UNN$P-;PbXXP$?iU98um}Z;s6m_05|&2c6*mbZPpHBRz_JM2|F9bXG*n<R
zlXjvStKnRX`Z6oBc2xi%s7%+;7Lm$_DL3>+%bdH}RE%nlfjaL^TS6K3i-1w`D`^FD
zlX!NbTmY0})cuWup(gTk*cpt9GMc-%VCoG&c4|tl2y<%b(Xg@nO5g}MM0If*9iU`8
zHC;`;VIf0JW4zp%T?i*d%r!O@r<xCMXQ|<fch--$qO9`zicz?<E`Fl2quUrMr<tYw
zL=*;dPCWxo0N277xR;f><#402r8(;#_U19)`==`d0AQ2q(sUe*u(n_w0GrDG1vXuZ
zNPD~^)p6KT1^PLOr{;`03!}blD;^8Zrgk5OfwV=tOmr>T#56d3RldqWG>%j5?&=<G
zrM#D-u~(*#Q0R~h%MaG`7d(OZXL_CyoyfTIA8ge!MXlrBpL#dVVs=hQGX}DxY0vgg
z65a}5-#54iYm(#b4Si`t#vH#LvlcpXvx!M$%uJ#wr4~oKGf(ffF{-dLo^taiW^9)e
z-R9?U6q#|t`=+j!mnXbiCR9cHoR5!(Q+YgR+!8v;g(b6wEjxr_W<tB1;4%3rT!9mW
zhJCHJ-qI}s7n{LK9udvqCTM2%L;WY>&~s$MdyVL=ZRIwx=!Q_kSwY$=((_ws{SdSp
zR{kAP<eeU8!lkdsnQD`-+AW?h2qN%%gyM++Hv2efN++?iFY1eZ%kTi~WU)LLI1!yd
zbAfK?oBJ6|W9U_~t_g4cWZ<#-P^9`w-KQO-s*67AXSt2HX+9c{jX*wX+p3l{C>eBA
zfv_`~cfVQ>Yw||O#=>>bF;j7RIjiueSapQxXDc91S-3C<i8B7$uYkP!2$<^7!hTbM
zy6~HWW@~-&V7(_9xXr(i(@K_)Vh*3iJJ-MYvh=7PNOiZ$P*kN-?2m!taY%nrfr(XA
ztQn|Dc5`O+$_%oc!yj01wN8s*)lIxJhmH#G%SS7<eqxSJi+*1k{{EY1o^NStp^uPj
z{|4_Pv>_U4yu6Tjw6jhs;IRAD30F-Fbvq$SIql&TEr-vK6~X;P*V;`SJQ@W-xIsAR
z!t?OQEZUpjeFpv0_NTLKBu@;bu+2TnUz{;&0c`D2DS0;Ja9_bL$9+^nn)1;#U>n)~
zIq$kEWl0JDK^sGK41kiVVI=nfmqun76pbL4cC|lS;eIPsnuHZQ1MxHw+uSuWu$dR|
zDvvi|J$DsX6lcQEXy5HiQ;%|Ye4g--lvz_cJ<kv+v6yTSJ>C+*q!}y<EWocHSx(4C
zBsY*znDe%v4Yf>$Q|Oc-b=qz}+vBloJENyCl|;jEE8>p=(>nLexwA%1mVXfR=6!5J
z8kgle_AigP^)K`E8=Wr@0H8{v)RgeCmI;Elqrxw>`K%s9%R49Vy2E75n4xO|M8VU?
z1P8U4+bIYuQ34-vm0j%!v2gDZp(O<pms5>`*y20sX8g9bk&(}VyG2T~W^jbQJvdQw
z@95LqV<xxm>Lvdus!c%k+EwuqtbV_r7|%1Sd*+R|zaoF&>bJcIy!pDHIgi!H@uUK5
z&c<ZJSe9@4_Tzn;3ZU$Y2OL^LpHDH@doKrZ2#voBmt;d&O}x9g+Y3OsO(~NUXSC4n
zLRw$wj|sSPY!tOJe(|3--0NWms!P8x+1T@p$%s~)KWWn>FkolZy8Ctc3ow!Z5WJ+B
zy41IU4b~UgSQ&<)G#zFk5+#59PkMIILWk|2PIel&1;}N;S-Qv8LWnqCn?7or`XsgG
zM)|gY^p$SMukKnt%yFc#YQ)uP^$pOYukE)-kBpnqD^fKgP|p+bl)?4q8>8-g8$;eA
z<!cS>z~q%bYTNw|lzxJMPsQu5pmk17oYJQGTDQPw7T)jZyz^eBD5tka55CDH=>>$m
ztKqeXM6zKTwB<5UMFX+x_73{aIQ@k?AkNkJO#`k@S9Gsqv^p2g=N)QsaW$NO4a-S;
zKFY0!u;zAY@VW0&+aTjob1ux_GI{B=5G{I&ZjT<EN0DM?k(~?YBGT=da?-67tqn~S
z_P(LHL2B}7*W8S7?ZdAcJ5xV7S2zz^R(`py)?2yZ{O`J%^#VB6A#RxTf%P7}2;o`h
zKzq|oZ0hYqQF?oCY2s}!X&TnXZ;s@iuERF4i%}%4fIGoV0zVa>ujp+)A5mh`MlmXx
zMzi}aFuO~Ywt{4yE*t)Jgpr*Ru@;M|d^E<pz1Owh>Z75thf`;lN@BE^Awqv7GTJ<V
zfBEI2fAD(u7mSl`!*d~nD$vt3{N%RpX16<A6@dK_>EWtFlumL7vY~U9_U<#J#<jd5
zylEW1j8{YuktTP#Gb~TWr0cCZe--@Ky~Ef%vZRTYF3+;(w8A|~U%%T>^d{dRjOGb`
z=D&Xas}oMlSvesbrfO#b@f%@A!CF7*Yk800-X)Y<vUqsU5LZQ2Q~7RQXr=7#hKjD$
zqyg$>!VhUs!i6?S?!v5G(TiM+no(Op6|g+KdE|jn9=Rs`Gn(#sL5~hbPfBcSn1o=K
z2$Fm3Xe1kJbe_Z!x0o=!ESB@q=Tlel>m6R4G$K2@DQ5pmqyU43|GsF@lwrS4)Bw4y
zmDv2^@(k0$5@FE%7~zYBfRouaKatx_zqJ~}?m>O7OPr#GHzq2XNGOe}F8cf*dQm7$
z3EWbck)~9bexhxYK1MHk+~^P5yC=7kFA=<ebR}uF-(rX`oP7v)ZRCt2^>tLs>^?3z
zyruJdD^^tOpyaS2kYbD8FLsMQG<J(Al!D2Prm_CQR#8y<fIh;oyFsSTyZFX@eDS7m
zJ&M7wbLP(Zf@b&PjOz}uqUW0bRZLg0TJg((oLP>3Eo%5wY=<-6zL~FLx!WA8moube
z6=Cwz%1J_=qWuo>6@ZFG1K>>M{BWt;UY&vE`1lZBQzbUyULXcB6flstyc-_>{*8}=
zF*BQr)*zZLBPDUrX6F9|di>+0AA<_nSZ676WpMNqM;^?V^0qNlSk>;H0^{w|j;&vJ
z6yU^tWRP{(TIhYVL&aY-$K=$b;njMgH@8*qMa4_IQ@cjOuG5envHI)}^MU5d2y4hq
zeNT4nF}Z#;Kc?$82UjzpwBly8Y+5R^a@t(zR@hPkWwkMbVw~&~N^-afB@qC&(*uIn
z^J8ShAg8E6^O^gy8YKNhv9#;;cR~bu2QIoDV{5q-;;1t2UUM#ulnvP*zZ6c5-oX{f
z(+b&%tJb*QuRL+(JDnXpE14ekE{+W8y{N0Dscc-kZsYon&j!1VEZ(muQk=3L!_=OO
z&czvtZZx(!wE4<Ag^isij?Gt*v=|u&ivV#7aC*^|xMwWNg!7-0v6UAp_8%JQ35^@+
z2|mn5ouzCdGK$LTP2C4b&g4f(njo7qWb_Z<{w21_%wsR2YR-h~i?O48uQ<>LX$@n#
zzc03vkV|?eA98&gKX1B6A@^F<p!5$~_EaQs{b+4&hU=%Qv66)kVh*1p0U$JE!GXQd
zyS*W8TxKhqJqKdNh1XG{u3=XKo)4`h0{Ce6@o;D$BuxgC*_c<7z5xU+**zFzhq$$P
z!@E!2CBsyRpm7JruXbPT{3E2&|A5CTih-7amGstA61LIwcj9_4I66b08-b;jh`5lE
z-&iyv4j*6GvcAW93AF)nl630#cwG>oUiwn>nGHUq`fN^}@9gy=Dbl1E38hBqPMl__
z2={CvGd(V@<dTXKibOY1o#n#9?m8Q$OWTn7aUr>o&3<5(sF5{6S=F~fkV=-){JfD@
zSQ_<9UE#-$&tzkHz5wa5-nr|qfWwHLWh_0<_a{INgt7=yf4xiW8-bR#87~+ZH3G@;
zP>=jJn)u-XJ8PGfe6pmZw|8`7W446P1&CrQyHUCM+{W%C4cTr${S|kmvS!!pbgyUT
zsHANZ(GYF@>r?|7?chFusXj*Ctgb}5>9M+uhMQmpAWRZ<r?}{yLv+U+pHv$JM@Hj%
zLFT)4z|6aL%zQqAx(c9iJ&#OshmPcOUyEL>el^Z?*otKZEibZydRkdQ$E>ZOq%ja_
zR1^7Oblbp|JH5iRz$Jz05&Dzz^<Z1*y#U*yt>82R39|QLwxE43RktnPNs>ma)Rc&u
zc!z4y9p2}9C&G7@=@NPFb$k9;$G0q+$%9!Y;4ie|59}G3xVz(C1PL8BS;1=b%oz|R
zrG4ZEoDc$1U<LtXJVr|&|19fIm8c^(5}RdH@J4l&J$fwWb{(ObB-D)pQeh;AZu~>6
z#y<wVq2DnJjqgx`CvFpQl4~mkyzLnr+eBRW9?V>Y<*_OuAt#a-K={mI|82p606X?u
zZ0LwcP`?<OVRF-@YI5mT2f<ZUT6Sf17JqP52zSh-jFhKh7OC8rv-Ny$B9dIQs(>b`
zJD&5X)Vz;cg?SUX2Jd1doht9vyL{|tdlqqN#m2wSPR>sM$rWs+$pm!OjqOxts$@|$
zVp52l^{mrqA?N*$_D-2LY}@3T#pZJtGrCvw;n&;phCbRq4oZD>N0f4ucl&{QDUF6-
zj@*P&RGGt#!Es&*%P^@2Bpw_kNsiq{b*g@9xhlgwn~GVfT%ZcL;KANwNs=_-Q~LCA
z8~fJz1@qJLM%qSV(J;aW)t0u0Ale28k>ft4w!!O|>eX1$y)@SJ?wQ`z$!jYtpwR;D
z(-(8tJe>?}O0O#?Wn3c*GV&iY(%8QJ9?g6{LEG+p`$yytmED4y79W!xYAfZVehBl3
zZvGk#qSh42zaF@oulwcwbO86zElB)3=Em|XVpJ7fKfif??|aE-{Zvo$?#!&)C8Wx1
zXg;N7N63qXkE7Vyt7?T&3T|swOG?!|E}W!!dJbI<hrW)ISUJPwbliGZQ7tVQ7yEY^
zGFWFOT5Nj7)2Y*X(@+NOZRacdiCOXv_qeByhDPSoxPkGjd0Ocxq5?b=<M=**%FKg_
zC*u0$UvDN!8CufEnNSGv8I<2v8Qe&6GNqpwy4KepUUl1jWPcp^N>Sm7&@=3C2y`Ic
zY^7i+ly2?S2g(E%bt_cyYh2@-+!Tpx+_#SeA!Ucq?%yF(6AW!M9mZ$f&f3Mkmkpxa
zsj6%ni5Jh<7dk%CW8At-juxS$7+lG_G045J)@%j!8FiM8M_<x|noCnTY@gYw{h%KE
z>U!fR?R&psE{&n$YBTy+Ta`*43Bx!`8vQg0!X$8^2fr1C72?iMd9iQtU;3!5_3tRO
zR3b)MidV__sYR&|->E!d;3;U2l9NAr3?=iON{UiwkO(PZQZU(psYmEA1S)84cAg|7
zO72V2gtrI^sLsW<TAuXXXc;Xt-bqd@IogbPImORPHrr|T<K?9x-fLSiLXJ?hjxga!
z!bP`x$=iLkvy7)3KkhhWK-Ry1cf37VUUaz^JZ=8{^KTYA?neyJHS5>|_ZRZS=3A11
z>*minnoc(hIP|1gU!{w_U7-1_<vS6%D3AJCGBGM?aD~lQNpu_PwjA2<<QW$E)$q1Z
zg~x&Mo6+D<szL953nw}42kKyErYl)4okY-SjYwI`qeV#kHMIjbUIUnR-{{g>`NdK6
z#n1HEEzej>`W2@k5;n^NuMFEg(!`rdH0%>WY?dx6;-!&x2}$YWVwg(i=vwJy;H%Xx
z;3o!jL;JYWTd7n}haAbMEK9Xse|m8_b%};K;3G}N&6lFr@{%_Dep>!|Zs*PDLbctZ
zW+k(57T^jxsg!s-fO>gxv)Z<^-i_tR>XiBT@JRELV2vnwd!Kx9zBlhp`xhsJ+8OUA
zWQpv!!IX2%+PiE<G3oBVp>HoU1^-%`FHfbMkVm5X4=m|Eb-R%{q0Yb$PG7+_ziDO`
z7F>Y?;etnE`m<r>!yW;Qe9fIkcim1W{b@@$6!$erAT*)$GLaouXE$25ue`a&sAi<*
zB0Su~Gxl!qQ(dESTl)AsZ9PA=iHu6eU^|AEU*rm_`E_Gk*(YJv+{KDcsbN*w|A(=+
zjEb{svWB69yVJP42X|@Q-Q8Ui+}#Q8?(P<xMiLwX!5u<^OK<{TXYQHjm2cje`Najj
z7O+;G>(t)6c2%*|pV3_%U|6)*uuLB&_Por>hP_rcN{!z8EIqQ^Kgz&z7iwxkS8a5z
zJAG+rGIdr(gPV#O--$>Dz8A#;9Z8uz^y;GI5=C_Kl<iphkk~tVG7!D&2d5FtXr)_t
zjdE^1t-(y2$OL|z;}U_x4=s68?P}z<bSIN;5Kz#CEb3E7l{y<+7i~l=OHNL<$0T=l
z8`iQ%%-Z-79tZ2@U4K7bR2kO+oh~7mRkV5&(J?P-2?<x8EfFM&+`qo?d^WQWwLThI
zj}$t7a)dmUQ9(ZOv6Earnp`OT`a01H2qLCb=u*Ey71k7sln3rM>3LF?Y3GlueK6_G
zeqcRE?!(9KOaw;<QviEC>79S~;m*Pj#9GN@P+chsL9Sa>t6J&ST0~E1&O22sppgwp
z6OxerDmMMR^#e0mIDIj(+Q-xcYPlke45IJ^pOW4SWih&N&-vSKkD{+$dXlJvT!T7d
z@t&b0Mz>_{2yTLP62b5uj7<WxsEw~p$h@p^<3&<2U4dp;Pg99wSxFX`ztR0Jwai&l
zZcbDeO~pMFs2smrFLvm<R&^n&G^{2Wh$@W4i4u=gChI-?sQ`Cf)WXD@;KWToE0^|X
zpQLvw6)(*Cr|?DR=PBRWzMm05{V;OZQ6()Oz_Qk0ey%M?<Y%`;hB>l@2JM~xR)f;D
zgEd!Y?*v|U%g$=y!$Gf)9(Aatyc#=p-MEz!a@NBOt2Rj5{}6^;&lC5M`I7d~-j1m|
z>b!|FD>T1MYx^p2T^mN5P6l<0=~{RLA5Jg>8zl04%GnDUH}rRD(iLR<QSFkoNx-ev
zRG}ytq{~Un%IXmIQd3>ML?jC`*>Vgr*&-aW;$^?^?XeKo^lzJU*UgL|2V{Uz`?9sm
zv)KYqk$RR0HkV)j9M4nK6wGuSwi5`OtB00kRhm~3a$DzluYZy;D*XHO!4mQbTf)yL
ze}1vlFbi%0hG424ewxAq5+?xi_YS<YcD!9nDaPSea5MFxhkq8=JGSt?_}q8SL%;#V
zE}s)8Z=Pcx<`qHse!B0J)#Pw?+~aCy)i}W%+9gg7t)<g;sci^c&G-p@)&Ljj#_AFA
z!PFx|LmAQnjXQ7mnF{T@G=kdCp8X1`f8(!FRJ{`q4qt~EuDVv{$KpBzXr!aouM`@D
zq5935&dg|dI|Lw+z6e2=9>H16)nO<a5=K>iuQ2soVW`{Aj2el@vdma6{k#O7TlV#5
z_Q#sy>O&=M$+Aoe`gWL@>~wpV>CML_1b)5vLD1tD7Kh*j4>B4wq~TFdINJs7aDjK^
zhr>|jF_n3FQHtez#@7#QX_w5qZdWutFuNFyhW9Z`X&1Jcy{&S3eEOP6EDW&GLi*84
z^M1S7Il-SSB31&+5Wezr+jd~9cdykQQ0<o$y+}M8$N#Sz+KGeQ8ZQX=Tdh#zu{ka=
zaZK={xAboGfHeCvow|#~LZu2BY(Xxsi#D0uov(Xh#lV25vCkoH`}OJD1KHa<$l~lb
z7elYjk6n;~jkrzj%|(!XIo0Vlez~^KY7A}F|77JqAnVW{qcI)Dh&e{7iH@a26_9IO
zx!37SX~XQ_CxI5oCz;Y80{V%I?54&oE;1f6O%d6YE8pFI*T>=dv1|G$;yAq;oz76W
zyIJMM?)gnA3&R!LdD|PFbmN-b7|jp0^`13YB~mm|yP{}JosxOPNf<_m+R#EpN)0U=
zSJq-Be!yO(1zsyn?8N>EAPA${!MVwJ6rZ_e9Rc!AFe|`{6f?1j_#1G}A_g|?EB7zY
zEZ1AoHo1Ud=8#e0RooT-9WLt&MKEalRAzub87RbyWPvxjb9Z><{Td~lh3$>f4}a~}
zR)*JJkB69A^u^d>xCPoh_Lm)5bDYsmuAc*;2l1i)xdBa}jgh*|fyjOhp1t~~uM`c^
zfiZUJ!tv>Dh>8B_n-gKknhR?ogu*M4_4)2j^u;-zI{{zOeVpCei&rUpEISEreYOMJ
z<O7^>RXg|y$i_xhUz>w{z8HT$pMdq*`&Zh1q%b}FJtMYtlqc_OaLx91&31J6r3{OU
z+Yw9X?c(jv0FvZttDLuCw|hB&QKU3?;O%)}9I{yKZ|5Ne1A|<E<fHN*iesleI~k;J
zwxQOu_-oQ{pR$X5+)A7v-!7L?t<NPwLTCJ+1(Df<3l^;_A$m<-GL*)Vt1HK?X%C}c
z^y%FeN)S<}<7ZURpD2sqZB*bHCsZfgMzjvtWb7Eb)~$>O_T!8^Y%I@}Ln@yujdo@D
zc)IQSQU+N#dbI{KcF7ok$C+I{n0h&-!V?i3EE}Q%O#(Zhd5^~ghx1Xm{zL7FN2OS*
zSS22#`_QhVj8!SN8XXfm?@#LUKLP~2%+X^@fGnNq*ujq;!yL|b%*J=3u@~wSZoLzc
z0=aEwMrcb^RXcMe4b2GXW~BCAO<~K5J(EZ}MEKt*UfN^pS9+KNB~e8cJE1@I{@{$F
z_oL{R#V{+_wdG|*%~!pJhhhv#>A>gyaYATz*~jx$G-?fBBY1U-^7fHB<&H$WzHI*7
zHbclcqpOEP<7GwDxMzP(Y2Ndj*_CbRk4CBi-y~Y5Dw}d@_CUydR;2fcv4*`ii&bh6
zw=LZzWb$0C@l><L_}=#G*|w$%UDRklq^H21mA3f$?WORo)R*(`Saw`xRaFdi4UHwn
zHfAn=9Tyg60V4q?2!KNS9|EY?yl4^_q%x@)XO=u%at)9ie^Ih4g=^^#nT43FZ|C-)
ztzVo739*uLQUt{IOKJdU{0R=<$@neiE$noU!)!1u4Vmtn7F+7ZuF*%9njd>#{hU~l
zYaDT98Me{iGy=2w9MFst0<qEIL}S0^I*&fwV!Ix72wwXBW_mauWQ4IfAL(xi_+s}3
zx2h)jK^C%}G_$k%!To_I)jGHY6CMw_zmQR9q7RzOJ!*E8MLZ5{G5d!~hyerFi&=1j
zC4usn#mk|(B3-5XunV!@+X(uzz+rXxj`{?9y)sivi;B6II@j?@&dy$_1WIaOHC6@a
zrH9xvMC5L(ovL%5)m1NEYu@gm)C7t2nH3g0^8y?!h4_=n)n_smrgzs1h6!ti{K1kR
zt*}}~JRdbdBIq=LdrzKwp&`xAtMHrhwdPE=XNM2s_J>KKG9#fE!%LNy`9^v$yD!(F
zh}IjY&>t=C>xhYa0=F*q85r5G@FJ=Hu897vA{SddWYgNYP!(yQ=EA^%P*YMZ<13oY
z)SD~L;XMd{fuOEi{v1LsJ?|<(uX9cIuLV;OsYQOe)v(zd7BZ2p0q=hI2C|J00I3Id
zLCE6PWXL=Cnu6(|ho~S`j3>zCL~}^j;+Lr@O_2R<53v6U%C~obH^%F8mhnIN81fpc
z{EIRD&B<Fl2avzp@DE$E8QE+)Vqfwset-AEFqh`1%-`Y0R|19Z_qft?^yy@H_*?X+
zJ1sjag42b;BgSI;a#d3P1I>wAj{A8Uvo4rCn4*n7J9}KcNT!Q&fprpqaeb9bwT!kB
z99Hb!Zsu-O1pJTvh-iFvh8^knE3zbmj?F|maN*c<GnuK+gw0BFJ4v=xc9mUyNWGC5
zXm88~`CIX>6ly6BK>gCJtV3hi9CSw|G+!BR3I*JsW|u-O+6`!bsUe1?(pp2IKg2LZ
z^MVX!i0$#@&iqz<uB9Nz`Fe@?f&+}}-Ggh><vU2-^&qW5{)i_S%S4o{t{O)|H_`0A
zY}ZTV3cRn22VeRW1BFE4urtrbz}qAv7cc1hH$OEHgpI@FYqYu61w~ETDr0Q|wtBoS
zU<YRPKm4gQM!G^vlELN*B6-vjONEWD;Y>JsDN;pDKOH*2SpD0i4_Wcdf3)>Y#nhC%
z0J0_G<(ufA)X;^u*X=jKz?<Cdo!Z8J31#I(cQBbpO{HkPK<^fH!LL&GfPa|K2k_$m
zj-Dbtxx`y<_MAYZ9|L&ATRj$%|8ig5@Q@Kab22rwLN4EshjIVAFK#Q;M>qV3Il|H<
zp78QF$e8%fDES{{SIfxCu`<*v!+)hXP_%0!cE;QsNsh+&k{yW(tex&Hhtleh$m`GU
zsH!3L3;l*QMhL*fURD}I*yaFsxN(g+=0^8)Y$u>&y+I8qnv^Y7md<RE2d_u|=+Nf1
z#QGozP)ZC<^}9ANrT(C-Yy4q0t5RNF^NF~`ur%5^O;DobcskvyI&iY}Q%0)3c=`>>
zC~%=^*S;QYg<3?9CP5c3C31XVBjRUiI(>cW2ZVS75?0N&HP@0h(xIr>tF7<*S-q5q
z0hN&@vs1E$qP$2TJemE5*?o44Ni!HyW-&P}&N`m+8q;K6o|qQ&b}!_cVVG$qnZR~Z
zB|J(MEixSbnJf+vG?jQN(+n1&9@MSJS>dxqgMK^G+ey$LHrO0PBhQ&UetT{`XD12X
z7miBNvr4KnrM-OCwMm%i0=BjU#GK84;n*?AD-#2e8X)m-`1dH_z$0R~7i{v+9r#OA
zFanNz=F&Qn#Zoqp#X^FuKWMhSh#=dp<pVPIjE7Bi_3cdVE-`sX+z&jdr_jteR)~2t
zxpYi*uAJ+X|KJ5-MEg^os%>;zAkjJz#&Mj;?DspN?|2~w4f0m@gEZxEl)o2<`FJqs
zJ5Jlr6W$REP&>m}5<2lp&_?w!*5GKlb;-^NiT=)Xk_*x89w$}awJEoMR_v}^po6UQ
zBCLC&bFSVS=$odsTswtR3bF7b>ZsJ*xh+K7t)NlU#R|mg@nB3g+6rfZk7H;1iDPW+
zg@Nu*kc3${Qd&58Sd>lSy;Do=-zJQEa)(RKUx}n=bz9}=h-rYemOyG7lVE$nwp{R6
zgV|ymIG)qPqgFYLI{b6Lt#iY?3hMrNq75rg2y7R8It<|X@Oo?VriEJAcbH+YNGY*l
zq0uvMrY^i)n?CE8>RpjA{5$*KI3$O7tKYxC2!vL0|5XDZcNs(;fS^ed<mtiPa>L?p
zy5$;c0)XtXUPZ6m=)MSl)(@j+EZ0cNRPWyt;##G<PH;!9*NnO$ScNl%77j8*&D7L0
z90b<3N`>t0UQ;2=ppNp64}hb3M2GjyRQATEQ#qJTsn1sf>5P;75nJyR!RXgQHRMPH
zAN8s{7OP2B5ml?3*BHP@&nq(?o$Es6sS`rqovFDQ`|`0eh!9QUUD3E^@7J)%xn<w`
z0~+C!#uOH2?qi9Z#rNFTG<#|)zDCN=haOu`A<hifGkKbQ=~Wf#7a-N>BpH60OZB#!
zHrr@nCZOteKdlqnc=Vw(-f~slxdDDZn+B<S(jeLE&rvo_6j)XQoWA1Y@=~!Kwk^lm
z{Jk0OCmTg%G;9cyUakTltwEY>G&*Ssv9SX8iUFn%AqV3!4L%!vew~aMhFm`(%@>j%
zKeiuFFFLmgIs+8no=CN#mUBlq18^!Jnc;tM@^y){*eV9optXeX4|$>id90b(JR!Z>
znNX?VNZLx=)d(L1G;}FBIS$8br#8XU1?{FG)>+~+ZNwz7-r<^+cxxYKdp+)#Dv^-6
z4Y8?pQS_t;(U|$%af^pBGvI&^z<Jw^mlP!c2{{#?ikcH+RmslwgSLVYG2G>6t=k;N
zrh}R?o_CZqWYvW4eHP`DIb~0j{tSE@b}tqGwh5B^WYNg%XQ#DF(c6NbjKMnbSZ71f
z$OE4T7>0xDDDD`IrGjmbp01KQsFsHz6Hb8dUk?jJrgS@-!a&LC#}2SQ9dxmQ-F?zS
z>~i)+k)Dnjj+?Y-qI`kbYnf1;Ux>xKqvb@y;=L<bJIpopa+{%}&{e1Zby&G%bl}(S
z{oMwVb_ZULg>wYHc^@QQK(71PpqKF=O=Re1wfe$l4$x9y)${$uIAj)e(cR?1o)~kL
z|9SNqy4G{)@yS2AAmN`}pg4D0@GlMJ@6rE@#{c115Rg7TTy9avY=M;36;#q3MJb4$
z&nj3`X9byrCT;B?;I7~LX(=nlD7&YL=YWowQ7Am`v!Uhk5317l=r)9tsu#8?O~k`?
zXqw#V6n7s`v`L>JY<VDI-E$)Zaed76_gFFTLKF$zd4TMj=X7w8H58f-iL}`_f<^rk
zvyXa98N_27GNt2*VBc<|7=1+T>lxLKY3f35-H9M#kxAHiRv4w%?XL5*FMf+a6CeCP
zm>T?!85ct9V_F|c7F3fT_{)i7+*OphgVGJ$zXY^nq&zrbCd}3kf|*KOzg=%;b(y1A
zFvpDt`|NgEdm=RCV23(bOYa_MCj)YtyL`Wk8*&Sh&Nhk$zg$bFE$PqAQL;09`dU2A
z6BCxK<5(;IobG4feuC{Yu(qEve=hOfujb$5V(2-yI6SpLhR>r}B?Rf)TwVf*Zj?~-
z@Z4>9sB|)a5+dbug@P<|&?)+Qrqbs9yxZe_7SPxk(*Psjq*(b6-GyMN|D{wy!h!8V
zbo~!$`m1*}BTNqc`gE6`#=%XDi#G2GEeJ24GiwTNM%B%`X0R@s8qEvv;PBff_Q$x9
z>1}(YQB-<B@VK7=Zv-H-+sp#|Q(B0(3-!ZxQpg<zensK>iGUrhF7{prW!U+5tmAKp
zqTm4+qZe3P!`I-+Chh>_JKQswzyRo+Jf$YYD*l-kUez@xfP->OszN{h*qrfKkXk!s
z!|^Cp`1_>nfyfjZ#4P)UiqFN)`QK)cjflu=gAFv5X%q_(Eaio&!{uT9k)DXnWz=F?
zm4vtj){?bXqg}kL%wAqhe`jMekQ0RTbG~ysix)HGWI|b<O&t03OT5d>w`qgQk#^_R
z<$mXRa}}v!iQPzUV67tzY$-)@M*Mz~zwuNm-2+gu5(`!58%!{6%=#b6Lcq3V;YucR
z!q27mDiSdLR~_bq&+#8C^_(qk>a7${6)qBm&-rXR*VljK1A(>5sml1@?Ea0gMjkFh
zkP`lX=uH1Fq;%r?ow36mBdbdbQAfoJy#&^7Y5a2Sj+cGt4G-m#D5TgS%Is(4F0jQ6
z2v<)MZbk%W(RHDUa!(dw_6p|k{pP(v_yON|?HP%5?UkJNl?yTKMaE*`2ZOiw5tTRZ
zwNsr?7a$onnck6A^I5KjMN1M_7m3+zi^5N<8>x1a=IH!Pi!XjT3!d@(pc_mw>j4Uo
z`WWc2TVCm@4NTM**Pe*()efPJs;D8=22`l?enL*edO0;8n1EwKsxffwtwW6QxqT4f
z$G><Y8R<$Ybp`jz8uBvXEYEJlIM~3&NEdI{@x>Q>Mh&ZU9xviI>$;7j_Q&<YqKlhA
zUBR<GK<XoA^zSetuTYv`aT+OZ#gggiEwo1o!1LMwr3p9p848`WBIfj&1ai-5;`YCL
z*q_s5g(>sn@|pcsZ)5#NY-=wnpjZ=_)c<`gcSO)gvGJdGV3VC&{AN#beeU0gL`N|R
z1eNywUmBkv-Y(se$JlN|N2{kU%uaj&<=K~d{+Mm}frzg}F2kk+A;2R}1Pgx|vyn(O
z;8SVdGs>v#KrxVZ_URc_%w$0PC)*sq`weyc@7!>?{?46bCv=O~epm!X6mjuM$VGHK
zXYUZXG>_!aqZRzM`Ui1`Ak6^r!Vjz(4L|i7V};Vpbdcoi=Q0+`PorCXJTbMC@3@w>
z@<{u#^u512R`?1(NvZADkV>G@;XOy!{NY&j_(^@aiVHQ7y{1|wSD@7{kesCz!Dili
zwH68L@_A%MG02M!yxE(xaZa<PZpg-t*`7%p`pCKO*8VdH!PZ!q8*d<=Fj267C8DcM
zAniN%;I1V}C*&Vt<tn@5)Vi?gMbTo~Bh0OmeOc-Gx)gAzr-QWy8g2bs;CMdN$u9jG
ziI@%>Q`7i(+P-WIe!GAB6ZBB{b`w+s8wlAKLGo9vJ^opJ(f`a!8t_}KUE)mi1uR<C
zowMaElzuKei&))PKloF==$q6!<Qjsg@c$ixa-cdACG@ivz*%%fpK4YA!KfeGOyPi9
z5buoy@{ct84S6a0&0r#(fib%xRP4iEz{p3mA=7Pf;7?8%(D$SfnC;vv%}dU#&o+$x
zf>)^KHzlrh*9fsN?^&F^)lKPlTMa$vnU6RH_&eumn$<Mv1`jB8F-iF~Gr&@a8j&?l
zl4-Gm63qq)&wDZpWty+!$z}2hC36gFR#CxB7v>DRLfVxh5nyk`9z&(h{-wyN7T=Kq
z`jLA?f~<~x7-PJ_-un|_U?(-+ADamC-zD@yU5o@lTY)gG;DxJ=C!F8exRD~97OrPg
z2!`BLI6Bku;P%$xQ6nt01MZTB?7j>%A9N?s&JQx^EFQ*?OdqKt%Zr*~r07jgj;I;J
zk`vKe^=L0|p`3p!Um`)0xe(;&Z-6vTw*PEyGaTa4Zm^{uZOtN9Xr5{J6#FsG{xaz~
z5)KLq@{0FOvh`DE_(z!eFLJo|qj>6tc~*%(8=Ii&t?X!mSXnx+plj_bCZrVLi(^RO
zby(O{EQl_u-|kmGk*q$Z!n7MMY>ibAj2l(H!0H6>*Up8dNsx-JK>-`MxmX-LxdF0E
z*}XT}QSerX&qd+F{zRGwZ=22+1J4AZlU)1=IZJKKrjyAqOt2CF3W=Dsesa~Jj*`59
zR7yjvJYrae-4FLu90mvEMNC)P!d`H;ZtYqCj-;{g6l2^`9WC&h?~0o95&yU?>4pN=
zc_JsAIuQ@}|Ktt0DbKD?p}M&z@wD*eRbe7bHQJ9Pb8;gnv5YNq{Cb#kVLLa#jDz-|
zVz)J>9t$Fj^6Qr!_SmMgSo;wMD6agJrxAO=iK>g28L`{yW&Y*GHe?P#QI|NjIBuW)
zgE(KoD-ULr3NON7OH`&QlbxlYjwUp%R<^<dTg}23`fu8Rp}iF+RsM^^;TI0^n~7u-
zSgyWwSouPapLK9jsF7RCPaDtrJxj-kIZss=xkWz3>ywb+c&gYE{TG?sIB?Fg8?URF
z5%5Wa(PY2aUs87GQpF;YEJ`dEBw~Z1Sun}6<wBR2Ursm(0#Y^1XSz#!JqfQ3e5UJP
z)i!S|3Qphs#D))4Ltcw6NZ<AX&=xBX(fXknHFd13E^3Ac*xU{BItg!~1rYJp4-kM}
zq|Gm%%z(qjN)YbdWyk*HOc+fBh}^$RxCb9aqZ1N3z)m>YreD?*Z!ug5Ws*IS?Vl7s
z^*2K$N-F)%BP$QpHkKFUn&gDv{7jANS&g4(!C0U*YvGqF)Kt|(Y>r)7g{0G+d1RQK
z5p2EjDJsP!#S^tieiR~vHrOh);y&NhTv~k~EO-mVA7f3_6i$BQXM!8ZCrJnd<fLXu
z?vWG*T(0Ja^-f50b+!#NbTq{}-8k7?fkOZr9kz7mhL}nC{R(5x4qm&>T5??hb}(Bn
z42ZDl(@iTzfH|?$2RDKGF;$~A+0~fi*UG)0{f#OctiC<$eULIC|26MBB_3_>7l@Q~
z2^ogF2U(k?HgVHXbN9UAF`NIbvM^A;y1y_lZ{vhewt%O6uHe5##}5v7wv1iseb(+t
z!b<XpoG=7;#7L%CAG!t;M#9edPm?{c^5qi>q=$$Rr-j28H^DZ)ZLvfjlZfs-`YG>U
zGcFO8>x{4UD^{WO<phf(WDvhm*rz_U#XjpRAYY%%?1=pmfjNTNAIBColbcX~@wxxd
zi~T0p3w)#`MGXY<x_m~Y@{h78dLu`<A&JEl0dx3ox-OOS&$+OVITeJLGfXC{=K$jB
zrEa6K9)LlwCw255)m@Xxujz!pE~hl8oB*RvJR$OE-M9t<-Fps1r{+|wX<+`mwsNBv
zY#}2<M_ES~r_M>W;9%bjzd(TzTGo{j*0By&bB#<ARGw;&^!jM!#*vAiioF9@d<74$
z7|zfBqS7q!LvyReKtWL)^oFgtB+s(FN?|<&eyz|yHulK|o1xOVy@faO-CA)_7PV;u
zy~ng8mAF*=r7*i+WtwT|(dNhFFtUw9gdeqRn`QlBm`PG9@Z{d@*FvqNA%{P@g#JjU
z8bGpbfl~We@yp?OzElPi2j_!wSXh`~eZec>-Md*x_djH6guGATJw$XV-CE9-Tu+rZ
zdxBh$j}q$;f-gJ|ac}(9R-7O|%K3{W{wW|{wuI4|jFKOIxz8^Yq`xc~od20kt>~VM
z#Og~`R%NOgQt{$ZTH3`SmY~K3)M_xw<NSH<rm?$#8WrbWt{cLFrj&jZj=gVmz9kpZ
ztnctHHh1EaT54FB)i{P1ErH+-)sBZfa=*}LLXW#?ye5Bqm2lC7^^WZrHIaSxF)w0g
ztifK3ai=b}-^Lr*zP!#jMa0<bNK^qZJVAReQ<y*^cC;jT;n<N97|k~A!I?gi(eL55
zPD#pfE8?{+x$1V<vcMPnJR%9k35xnN!!((qsY5uocEVbJlT$#tp<?u}P7gJ*YS9fq
zIMg_cW}Z2lg*e6j1}l#v?sEpOUb3-%+(Df>3^C(yOO|0TQ?Sp;wGr?=8``@cBW9;`
zJy(Ov&sX0gahQ<=BHR70n+1DNXg;Farrfv5Z**<b_|YuIHe<;k*djUAxD;+OJpQni
za(+Uj9vdmCDJDozN7JW9H4Ki9Bjr$nF{udo#dRogRR^ir-331823J(J>=`?PrJ{;O
zY_TE(qx$c9E8G7~S=Rn<mF3>kqOUtJK<@*G*<0FJch+K~^J>)l!)~qMFDtBJXQ*mE
z4SO}3#FyD!{FCXrqotFh#+}xPvo(=!NEC>@m)(4s9(UdXuo=u}^|rQDmy4zIYHY!J
z37kqcu5~ZKzqMh3Lj%Qx{MgE{g5d)tUh}RuRg`Oz@M1tb6>rkmiW4@w*<=E6AQ0<(
z{Q=4t*$<WTZf3C3PNA9aBcd^?&pQnsl`+>s3Lrxf|F8=eMDRiXL9Ga~1H}M9!|9WD
zYE9Lr6l!N}2*e7Z@ul8na@@Lvay+L#fo{ZT`P@(%^<xtb&|})QQq|Mn0hN7Pz@#Ep
zHw*OlA<8dvg;IcaS)tq!C#isUSPKfxl#K6Y8~KAVkxO0wOt~8J3KIv|NSi;r05{qx
zF)tp7hd=j?^677cT1S4p$S%wFP|FT%f2tiZa`K~ed#??4!cM<`&$3qmDUilbX-bZb
zmVWNmyB$2f9`w&dK8%7=GUokXQCo_fX~FO*Xl))QGa+TWg9i%={XrwW?bgP#w|Bvs
z{jHW|{ols!MN8~ky+2FwxGsSfdDJJ0`&GeJT4Jf6Zf)zZqzR<$o*_)3(bu-#+OHoQ
z^B^hQ`9S{`$9^vDvMVIW!hln`6G2s*M9SZc_<P~?XiFMyRx%&5OXNU=jYPo^C<J|{
zA>V7q8u}|Dv<3Tj7wR$i_%Ww-=b=<HmEBFF3-h7mhzE*&8JI&@3zsY7GM#d&P7(aa
zX;|5Mk(his&}Es}F-BoA@E1f5mNZEUAi{h!vu$cJBy{WOFxbJ_=}9ZHqdyz&AV}xg
zCsvHgDj})Ajt5yeu<#Hw)sCZo{e>W;9nAT$zDa=OdJ|I=M(KR=muqqNXyQgJ9!z<L
zg`6ahv7s06KCllxHIt@h$^3a1Whh)~*V|VMfOjespYZI_ELaK9vr?k3?L^FTk7hY?
z5y&cb=Mtb>G?y%re_pf$<$`PWCcHV<yNqJM^Re-#jw940Vp2}nr|LWjTlPN~kP$-4
zr8Ycl+gjooNcZ&qx+4C#uZn9RBV(p#SIlaa<1O9y<P@(Z*<|V83jaaFvwvN=r@c}X
zjrhea-g<=WWD#n8|1a(Zf%y)93j~ng0`bIwO|ZERBd*CPG@yos>(Ygd<#E&aw9u)e
z=G1Rv8$Tb#U1=3D9OX5BEpCIEvG(Jf$esM8l^c=P5*96t9e36(*hCw^l0l{y)9rV(
zuM!h9c9Yz=^Ew#CVYOcmY&9budrTO$gr0hqI3M*;o87sOxiFvES_PCUTm>9^t>m=B
zs#Q|C6`6~+5VfjV)rO8*`QNO7$M`c@zT@RpGY+>sRA&L%5kUdcJ7Y?`7GH<T=^X{A
znk;qtVwJ16>tSqR726_gZKTTcd5Q=hZJ-3xojAK^cgq1@XRJN`3Ry{c-+W2AknQz%
zysBVkU)vxZk;1Xz8dkPVYo;Fjs+$|Ji^E8gh0H@w!mwFnIWd_5>Tc^t&+N9&TyV1q
zgGz;C`!N^y`{+bTraqR09i&s>;$HQ%#l_MkR-WYF7`{g4M3svQ?VS~ZDhYkO*+*D1
zKZN(`JT;Gq7nsj&@hichr&G+-BNtK$;U^zt1>>xO+3SW|<R6q$F4UzzTMjaT{)10z
z%~`%16?}IaK)NI|f8t}K^=za2BrMdzuJKsc8DsOrpa_xj{(2h!YY>{|l`Q71IXGg>
zEKvcK)!7jzFRTi-NSDud1tA?7tr_xg3F@oa)XKD%!&$LF#J?%<?gR*d31P^~u9o9u
z5TzIA8S8w@hj;8DY1FlF9(c|RSIVO_IKq<3VRgdN^>O{I%WFIU;%O!5Eqw6vcYnZ!
zv4l1CvTF&~3D!p}46gb-y|PXx-vhh>Rb%n`cz&mFYe&A0ZV-N=DnRhZ0xu_w+)&J#
z8T{bkHcabtgHi99L`2?IDN>C$wWW(t7LLNIeZe+}4UvR39Oc1S&-<u5|Dy>~zQ-3v
z&IO5S?Ad<&?8@GV7PRG#M1@bYTX5LZ^!_KGGA=K9X$jhVwoMcD3w?o6Cz(Yi%`Puz
z$()NZK$=x~FioZO>LIiF!S+V2PEa*;F@;vM)BTlpA5=uWvWENMrdwxpYJu%X(2>%U
z(r807uukn_RN@54%wwv3U72GQ8DTpzVHCrgo03zJQU4%B=wQE-BE+u!mfj|>M=rK*
zG>wHenWVSe_b+<>PH9$=QzZDPSp8k`QC~w1+BZSNn*QZ=$x4mNgkKQg^549clV~|e
z(J+x|^10$e%a5M0At4~!rHMv>8S%A!ojc<>LRATCj76=OY#^wAaEBGf@@a39YLdfg
zIm(-2Pfv*SH~aaj51sR>2m!BiVQ_P<j9J+>JIXdYaVn4UnD7w#&SQ8c2d@t+#D~G+
zdEhkudK~b@aY|B6gJU9>Np)F0>c@BM+x3}r-SuvIOuD2s8d5-de63N@7uF_Ad|CJC
z7T$DZRLm;0jB{{HK;d4968g?M$b_O^Nb7Omf58(!s%mWq=ThU=aCPscbEgrpb;K`y
z)#Fq3A|G!DNpeT_NMG@4NXoG`P8t_^Z7(d*UBQr1;h5RaQ)-mL-01a8n!WT}x~r8=
zcy=E>Vp<lZ!A}v*KY8hp{zw%*oI~Nk=8Kui^C;tK?J61}PZ$u}wkwb#K46wCbugvq
zXnxgGEA7;087WHx=O!E>iP?d!EO6T|^@r6A8f0Ji)7B`jMAP%*!XJ#P>RrOMpz=MW
zXVc%gO$G=q<*9eOOho+qC3{^vJ^y=(IdJd+JV|xY8hn)_qjI0G7Fqn~>L)?u5%*>_
zx(ypvzA|Hk-9H!Yzge07HzQ*k<ekmOhv>ake^s-5h}p@zTRio0Hx!fzLxGjLvFI<U
z1tkTv!)|ISUC@u{jT9oT<%{eZ=5w6gh%8jD8#<B^8Sf47xOxu1zm#~m@n*in+-Nz}
zM%lM-477luhm2!dKOeopWZOYUys&_*34NfnqWMl{5_hJzj^QJK#4yV6rDjG_P608;
zPzk_sYD6`1b-z?El<B7&Qwv_L=z!aOkwDu$+yIsDeez&kVpogdZSU<S=Fn5iBIAem
zL&jG7re3}9^$fJ$1#s+H(Viwt&#VeBe8YR;rqJL}7Q#9L;=?v#;iCzruY7fwk<h0>
zIv`Go-m$3UV7o-Hb`%b+UQgP|=Jw-&Q2B;J+$fP&i#e4U8UK{%kQ=Ny8+{ZwrKAzC
z3EALcePxpXD?EivtSi?5R(X2@;1R~*9Z=7DrKbk(BKc|g$GMxf!8g@ajcUi?ErYf3
z=$@AJmTWV&SG5pu4B{<}dIgyEYG!Im%G?p_*q#W+HbKjs|3k+#L%HlkjXsOc22&T$
zA>dzdHH;QH^Lv9e<EC%aE(u=xvOgIFtJ)AU{D*4)-zt{iNz@#BK}=qMcE=2l*FYYF
zT(x!mxP>3;I0(Ad-wTH4W5Ln%C{eT3s1K)e>b-2LkKR8mvRXdAmwrkbl-<SzQmOL$
z>`!-joNRX49EEw4|8^6?zIwkE_l%8qp@&q69T?3fD<-WaV4uT8qcm+_uCvrbKT_H$
zt`@9dibbT#|NADd1zvg2A>Dgjvil)EhiRgKD{}t_@!fmIEJ`lKv>&*K<_5#kI9m7!
z)N=d!8_6+-02yK)U+tpwB0J`lZ;uycz0#xeKj*1`#|n3EMg>=ZhTCL6+dC<zaTz3=
zEXT)B(Drs-kw49#jkLE)%-MWLGvpLR7|6y>sI``gjohtSrRefi1~%l9#EE-gxjm<N
z*TqkPlZT~s4eDL_%a}0$hnt23S$3MBtWlm9aIhIVv_j0d9yeXuF~(t&ZgX5$vD3;)
zwJ#Yidn`$MtjT9=ts>@oglX20uX(>6sio*jnk%%=-DKO#v>bnY5TLFfn==@$d*8r-
zAN}t&%xxGh;fpKytO<!9XXlMEo+<-dHHB)AVd95A5cJBWp5^=xdi`rC{@)yjAVG-Y
z>GOZs#WRVxa2FW{Iemxp*Y(xx8Ag+jOA#0lh9K46p~Zu0Z;ltKjm|TXjb1xQ!96eB
z4N`kN$WQ~e*a6|5QAgPeceAkAVO3y76GZ8P!VkxF%x%umh}lG9hsT68btlwZ8bPh4
zh%AYCy0B?ogYl%F?dXSMiX`T!_+qYQy*<!hHGO9i@)YtCd`e-2`0?k`Ob_!Wj(XYH
zEnPrz&K%}p!)p3kHAT@vBbZ(GCzywc<g{nBNkTv44WV*ig!)xRdq}inRHWjNWsn^(
zJo(VKWk|p>uMAy2tHy@Hf|$!QF<5gbU~W#zt3uAbc^he<bJk~U-WMPEh61@clu;I@
zz01)C!#w)hs7*8Ro>H})m118RmTTubd3RRpo>skE$(H-oCQLtef}~daT0+LC#Rcgz
zKq3<3FYOm#5yPFk>=Vgp83yl%$?G9@Y9%y3ruW}CKLHEF$9zL8Ui$M6kw#7laEacF
z0!fb?V}mqUac2KIyDu^a5N&;tknwLDyPs*UexFo7Wupyr#(3#xEB~jwBS_koYAMV5
zT{@~y_m7}9GNSS}Y@#{<+XfmH6BBy#Rgap{c3C9KLQk92k`T8@W`_uI$?03sZFN3A
zTt&3RFn8cZ+5~onZ~YWM<Ls><%Ax3$SBQg}m$zb$_E{7w^&ts8d~qz<y(n%&Vv_pR
z1zPxSGo3ukSV<b{M<T}=?Ka(9)h+#yy^Qxbyy>M?{V;OysTdi+$d}JT2L=jiwoEc|
z4g`eyM056?)dhW?bUi-~Acg*8EA@b&_^1)3b2r<vM7kpmtOrJ#7makfWN%th@od2>
zjblP)!6uw<DV4kV;a->izi+(xPR%y@|9C3mbP=C%5?Taww|{Z+7TROzEx2U0KFf_^
zOY(;+{C$fsu^Quzcw#zAQMOc&@_=8gw_xGY$`HNSVb++dA1{qE>~E%s$Z%@Z2gSJg
zD8`LqdyqgWr!|8DVidzFZShj(L*#;;1a1P6arNe{y^8r75$4S>>8l&U6Z$?Z{M5No
zIdsaK_0bknsz*5qM7<5YzYa0TuJo_d%v;HKs&aCJWN$=2_nPK?Tg3kD^r$2-=jHDA
zeQzreQo&0g`uW>?#MI0Yi_1JdLtak2fFtSB?Nfx`*#wETl~KhPQy9PVN#h+V;dYPp
z)VyCSA);Yw1U7C$RlA!pjUPrgdn)v=)326<@$S^Kha0-lGE54=S;O@;%#&}r%B-?g
zW#d%?DXcYa3e_(riB9(Et5Z{o)YQ}^#l^}~OysrlY3q0>VQy*>xvz>>I2i=kT#dw!
zOL*_zA1=gsNwJH!dLjJZ+~~$8oX5OW`_#<QVNjaYNydAJo)8)dt<!@*L#n049h0b&
zDE&48KukPJ^?IP!!}Ve$0ehNWOvn^WIOL^064jM?Ia8e+g#r>s(V7^5V=1L@><yc6
z;F(xb1$lWu_Hm|ow;bCFUqm^An`jK(6#P>b3*OOe&hbLUTzEtsiC~r4(q(dup&B6a
zMGK#v;o<NcP|?z%<+gMT(|t49nI`MEd)3A40Vqg_j(y%gX<}VRKk)$UWF96pO}XYV
ztH{)HJ*=MslnkRIprmwDN+%D4N9P%#G{Z_F%^~Id9bGHN6a`PJHgW)TCC}~+*o3~x
zUwH4MP}3i1#p}ZG-F`3!qbUxpbAGg-=T?Y}1nH2+?(F_;74KzbIC|5=&iKRYBBA3f
z>ey>aAAQxVL}{Nfxk|8d@IuzI%(L8Q;j^0Cya3zQ8{Zk{^C0P0@7k`^mhnF{)o1dk
zxrE4ga(Y;EX_Pgm@_u|Ud->|$XK?bwH7$V?FkB>U*iji+k~Aud!{Re|47s8CQ!dr#
z%*)5&A(&~0B*aMTjvmD&UlU)B@ygN{_f5rF_TDo#pSg1M$4(B#lh9pVuYtQxZj8pZ
zGFm0(gPf@&1M-3qC%yIyW>c>oj{CfSjE8wYRf2%eZd^|@2t)0hpm`T<goZ-Qc!!Fp
zof<t3U0_P&a4exNO~Om*)FZ{`Hr&bTMX52hQ(yiFE4?=nI%?`FOr)*tk(*nUQ|k#6
zgfsKMDkenzGQDa*_)p6>ukmxzZRzK7>P2#@u88dJqbgBcr`>&_sCYU|?^~29DOi!(
z6P5ruc+etoZL8&GR9vRP!CZL+8476{J)h@{@yL#@&m^q)+^K(J5SO`)`viw$!D2cu
z8K_r9_8f^Ce;XLWkMw=u+!80p`l8mwU>1iO@AX-$yBG_GPcj({JOh-qg)AyGc-Abi
zQbd#F0fa(Uce`kI%tCsJID|}r(r_qg`v@pkWRVmi?tdVYpG^dDu>GS<e^(@DoeIL5
zoEyMqeuHA`nM8|g)e4r|v-B8VScp#fEJ4h;7UopRWB#i^<x+Yjx69!q4EQSu8}1<q
z{3-YfNHHEF0(J|VHAVT<B;!n+(DUwMh={|*^y@3lN=h^AbL>6UKtF~OR=?7zJ^tZD
zIdRQbDRYz+su#I_frm|wBK>~Zpy~-){F=4cu=CM|wVE+k{i|JVv{_Ad2AN#q`eVa7
zHq$d#y~FHz5W~EhV^s%6vbz=bJ=(b5J1(kFH4P(1AIUFiUY&A!n(MwzFYjNcMDu6W
zk7%|>vW~xOfqi8J-`kTQC^U{OyRAO3?~s!krpM10j+pIixEhM476EAofiZ#Zr=r3@
zkPYnq_wp-m>uEJb<cW~&b7JBz3^KkxRA_jeJK0NDKRH=2bY_*X1}M7Bq{*LVQMA{y
zFu3Ask8y0}Q)V)CVs;=lzb~;?Y?V#Skjkxav613juloqhG%52I?hwQQNAdsj4JJwf
z<q0>g`!rD9)}~Mi06_b&K%X5fEdx4)gF{P8Go23rLL#Ci#*z)sVVYS@w$cYW7~r~<
zEN4y);dDN>HU7Rul?>^7l=2jtYvuD>@4#YgHRwo>nw_zIea+i1E#q-e!@JSM4ONMe
zE%@>02SL|`y&4PVC`JbXjI#s{W@P06W&5mRs04)#zHZ#8JnS@_G=7rkiNxVjgMwl2
zkO8w-Uel(xkH7;?gfOY7N0gx|)KU)h8JQ=vLHak$uAq}m-TZ6*n%vlAoy@R5RtziI
z_^~d^0ZBCQndZI=D80Ckc*ZDxXh~5*p?+Rl5e&J`BQVgDZm$*`cKeb-z!wbTBtPt+
zwLdlxeSx!1>0aDh947@<IVqhCr2P>umMB(rQ$K=Mru6uc^9LDWFz7*poUZ1Fj{CM$
z6tNLA{AvA<VPT|ETp{&UCI-0o*D}MqCP0M!qdV6{)+N8T>wawsmYe7fi^o(y_3sZ+
z$QD1@_scsilHb$$Xsd0<ehTaMzUXn4EK4tL8)Ajkn}~d{aG~E{p}w+GsniJWQ!^DF
z=z#URVQN17V&|Wq71mu$;O%mQDAaj}LHcCPc5av)ZICf)_CRToz5)%%+0Pi+>2LJf
zM;$2D#mES`@I15_-{q8@pdY0`WU4R2tE|mc{k9QWgGaP*lJ7lv1O=(6jVYNBkDTNx
z@1}ozcAH&rjep{UHW{-nttr+U>XlSksF;N4;44y@oZsO)jnQ@m|CWvCSpQbX%tNiw
z@h)ObA>ayQ4Prt)|5u+4L<JmoN8>uA$4d=b_=#_zrngu&?Q%(pkL}j=DYqc(;3qZL
zk_XlOP1R0HFrT4z|00q^mplEC@-5{@^(-SeOZ9{TQqU^BS8Z>?%{#=-yv9oHP3jdj
zL^k;Z2kzfG)o@BPJ|15@uEzKI=cr{{irIc;W@;p`6Lb_G6}&|dn*PEfZ;cTQr7dn#
zf{You5}#7eiCgE{ipA^?<KQxdf_p;2og=0#;s<~O#!aj8&^Os7gFX1pw*z@AX#YT;
zEcAmjW%@@pgpsd2PcN>-MyRq{{&>bCXN6+|D)(XOUDn-ta$FwG7d^b$UY(}o3u4RR
z@%F~|x6S+@45lCk;&(?(YfBs2YpcGV-Likh9Fon24n=Yu;W5-o-^MGAYBn0G*vZ)_
zHF2916^b^ATFJ0SVd2q~P<`{ZM&aeTmKCSHk=UrZpNl-w#AQ?q0ioU?anTU-%fk`D
zY9z+9ZU0K^)$#e|zT+wmix*X4zL(<4oHPw?=1j1E=qO)`fOvG=YY3CoHN!?uvBLBC
zPzym)x|oG=)gg3OFA&d&-;zlDGsG}-BxNx;#U7tWmYAD4hU9uabmMTh@3bBEaPz%X
z7t0Mv^psU>F2<dh&nT%^ANjaGTxU(vk*da#+^$LBtg5Md$FB&AOvcjGM}i{sy?yha
zF9dao@}uL$zoHe`WKu^fjwc??TGe;#q;M0fKD#WwyZ!kdVH;wSx)1mKTM);DAnzt`
z9T>=$s5zehs~tcPFDJStB(Z2rnv3_0w<Ci+E$$CrIdp0OBX&;-_PpwHm1EV}l+vcr
zo-~-k?~zzJE^WCX_{XEaMuIa*#>lL$RVvgXS~aXZO(X!6(&=7?m4ZU^dm`X)*>=vu
z0i}=dq}%IDcQaCDC&|w!D5r3yYR!A0eb4^oG*2uY&!P8~3M}&b7+)OS+Q|p_$wDH`
zyuD8>fG3=2rM$|6<lD`_HnZNrAKt(32@*n1#vwKGc|b16?dE$fhuJeV$_f`u7;pUG
z+(#;Q$3F=1mt+a^ulwv~`<5I1XU0NAkMBj+II&IgU+{(^DERaPU=HhPnmXGF`{tEn
zIByior=frrd`!@}6aJ-yzig?Z*H}_6+|SSdv?H9h3+PK3m|_CG@@PJ+jmm0nA~XM@
zZIkDd79kKW4%f`>&E?62zqVk^mP{AItaMKIa(J$W^#h9TC*i)7Mpy<1SB#jvPXR@p
z?InW!@|mN)dp=xDj%hexxozf|!bTL!#HlXth<P@{%F#{k=H`WrQbb*DtZimBF_Q)y
zN$#%csdYJfqCCwcf!<)RfGy;N7C&TmzSAUozphIvZU^F+Trf(S>tiA*)Tk~yMkMeg
zSFt;tMtqR%g-uFtVO~4XU)6K;)Go@KFgD-3abS+N6srb1M}T)R1c~khObQNQDIwOD
zRgeGS2_aF~pbP<7E{$C*uxtBaBC)r}Z6ZL!jDNqwD?7}SbTntS&$Tb#Wl69JLC5!l
z8_i$4|HMvb&%6Sb_B@)bBowF;lO0ZxetPt`EH!vnG?J=X4XyeTqX61;PMdf)%bz-R
zaWi-5JQYV(lUJZ45k?O!u3qmupZ6l8>w3~nss>G+DVFnX3&~GGSb6zy57Hi*m`N-y
zVTRK2!*O&1+hqSWAKg4}RyST>P`=2W8V*qi?KeTlYp%3u@HRclB}wG#>${~4O0YjH
z>mQlD4i?C${kfwnv%M%?pGyTtzF%=Wy{e}S+i}?3bjb#pNqb{hnw^@p>84J~J@WhA
z1-fgW$mxFZru{Piyt+PBTn-pbL8Cj$mfvz8B15~U;WDUmqObcvX{y8e;(16>`x%u5
zu<)DhO`~S3G_q?V8EojItmWf85^|B5iYBgcSgL*xv8!)!vf8h-kLWog*&^*A+G^{<
zP&(+lo+`dWv*F%JB0@JiO<{lWeDaLf#fp0#SPRQsF59_sLECGXz2B*r-Q#ex<BIgp
z3qLucDY<APkJrNp@zbH*x&8Re!F$QquQDO7YR}<$q`m9yChX3XOO0-zwqz;Q+N&Z?
zI~m4E#LVW*okBT}-IVcrl{%5weCB;=@D{`1g>8mAos6-<ljiK_IIGBIlpBe1;01(P
zRsNsoxUT9yHDUb4$ERkPhs6962wL3no5(M~xR)+d+$N#){Fs$+$wgl7VB{^nrAv*_
z$Q6x^A=XcW#`JbH9$M4LxkdxbaJKRm8>l%)`oE_bJ7ckZGPy>ouh9eqE2o_6k4Xy^
z)~KB|OYx^wmyqdX(v=a@f}K)-x33t?+1g3^XfPn1+<el=jcA+L=QH#~@Qd@J>Zd2g
zj7Wx{m1rfs++B|ed<W%*q2qlM^Iy9vO{Q;vsdY~LAms-Pm+}6D%F=?-G!_l^H5;7B
zM(j@;>`376=wMYyiTJY{0vtTb=!z2dmFiBZH}9QSR-YfO2>%1ecj7lv!Iln+T}6O$
zTA`RY2pVDlO@YTNJ#A4?f|7Z!a7Yq`<#A3au5`<_5ShYLKy+zWbzF^0v%)`u3cbd#
z+br``7ff5+Tep??Y0<v7^&DeMKM1&<v|J>yqi}0XSfSOS!DuxGrMTZm4CjCn!Hu`x
ztl*8;gPl<CxyWUp_I)n)8@JMz21hY{I~Mg>&3#ofn!6(%a{gN4W`aWm6(*UvW_L#s
zUBKJBQQe+Sj7jVkb~+(cNYC%cq7y;aN2&Ku(=j95t>DXC2#-+yu(QE#C@P4gGm9V-
z#2>UX$tQIG4YQZ%Q{tmw;%;2khlkUzpoc3;;xDxl%2K7lpK<XrTr%Lb&iN&??zmp7
ziu);;4O6tskycQ(GWWhS+)7Sc;B_i@e<j^(WsOVIYR+EC`Zz@KYKxvB_{V91XK}fJ
z(Kv15mlb9|7=mmN{|^H8|2Sfv%r#LeVnZwBCJj=EZZ*(n?h~EM6sY)ubtU!M@C(X^
z<g~u{_6{eeUMo0xt7YO&F0J6hCQuc?)%AhKZ+gY_-2V}0xdaPeDlI;G!;+`luY)f>
zWo<@FS87$GlJM1e2_%K6iPXDBeZKJXD1)5k&YDXvchNY#ou$3Xxn5FAfnQ!<-u3e%
zmwT>zvUX-9I6bPV(HotzC9Iid8vP`xerOrKe{gRz?ow^9@XGpB&|6>+c7y1J-Fd^4
zNZL6s8v2#Wb@N9cu-7fzFbQ0xgv~w_^nMIpaWE@^%Izd0*KNN-4!FyUMHL`p0i%<5
zOAy^G@{ZkipIwOX8rZ*agPJl+z5fAJ5bT^DtdN=&+?N(^HrzRw8~h1i6*Ae1);Yny
z8x1n3w=vMlS@Kwu3`yvr%Lr&a<w#?n0VnQhu8ekQ8wfh4EKdU>#^njwseV#B>lcI*
zWnK)o6>*i}>6FYYFnW5>*V`pq#F1!Cduk$uXYlH)uFZv}e3r&p;@XL~)&9t-j5I~U
z&@fuiD?ehHM$UK31yFatvT&#EkCy{4G-)y;iG&iYqwNY0fV_Rc6hg226dUb~u-KyO
zVe^I5W?xExhJs8G@tGicdkSf^{l>y=Y04YoDu)ihAK4YPiK?+#p52OZIFd5IERo%~
z*J9R|A_e^{HYaAXmFN>|7^QG!@ibF8el(U^3zcW5V{aW-JFT0bL@vzhDXbQ@g8h;{
zpC-?d{<1q~vX;ZN;jXoDbHj$%2&mtS<ms3U=r<U+$rMX%mp(bf_)iYsf6D$nnH?`{
zVBHaN9r@x~%r^GyrK;;uy6Ydm*iVRS2e7&yD^qcNMCXZe5r=?L%VkUQN)Bv(mQ}t@
z^<aI42yeVO9D{HDhC&r5MRr`*H0Hg#+B|oNifCrawPISeTx$k6reBdYsZ7K}eVsuL
zepL3S>GbxDlG>vTWd;1<2QSnxM-iqPPu6w(PgL6HCW((dKlW7If(CQd8uwP4=q_cy
z`~iIG+;1pi=n}rv-uS$)t+?uj?7YT9klL{ojhIVkmM3<Q72FIw6vYGokv5B-g(w9?
zA!Y{S7Af0KU0(YjZ`$Adz1TNs73Z&1KPnYRV?l9O;u}3(4mVD~mOzY#`yS|kcTW~0
zk*{5TM@|n9bK~?*dz2iI#)CQ!tEIbEJh@aT6me;m8cA+Ewmo98yF9m_6#zo3cf4(L
z!s4ln_oyq${un*R?0Ffp`O=lPaZtef<f$s#v6xs|M}AMj&c^7G!N=q-vp^Kupp5AB
z`ClvEMNkf2M-~Y`z2qJr7)8)0&D|lxaF8{;3WhIuNXD5&qOML=;4eqKo?S0$HoNvG
zjk<1cBsY46FnI$q{y)0@Dk`q9S=WXGA-KD{y9Egrg1ft0@J55Xdjr9OyK8WF4;CP}
zHUxL~Z{}TV?XlNC_CH1+^;!3v^;A7obzgUYOg<!)#`w4d0XhKzzmb}hN?2xKgoXpa
zt-uHGh3DpUG=txnZ%9O5zG|<sb73#|*AnphG+^=mg74Ls6?2@JbG-gR8tCWpWhOO1
z20n{4s=`ugs}&_2!*PdO)}~@+(N45N<DJ^O83uA`4ZZV-HNkEy8^5x}=rM_yfg;u4
zJ0Efxny{uo$@wyOW48@V-nabEsaISd-ih-YC3hA`q6xlf_nZ&B0<W9L|7s)z;>`a)
zg;;SC>D@u((ALDahw((eo4&GHNZNnu62AY4OR6351MaKo(E5+G3_&#in0u`C#aqJS
zLH~#!-o<8rUYF9SK*9p&{6WCv$F<3jd_=ofoi|^ADf>YZeN$PiI!)KX;j^b4|0NM(
zg68+Qew^Xj?j7>&C<uljepnX2KrM&RyCcQQ!DcBf$Q2dSh0It<TDQM&G2K*84(40P
z#$b22mYLBoQUcrdXnYBfDXCg4S2|lPuBy`ZMKB6{CC&J>f^IBpcIf5T1qs~k+lkO%
zghS~1v-_e|*tV(`n=iscM2x+Db2Bg2zZ*wo9mJFB>N$i|)UnzN`6z(~{-_CHWD^Cp
z%+4$tqz44!JUMo3!++s=&*88@9J9tr`tlyuEO!GvMq7Z;U;ZAe53%3Gy1ASan2v#6
zXy)}L)`H8(bvF&F=T}wuwOp)eHq_S{+)a5^CZii$Gp(BH7G}bJdm&v`Fptm!@v8!~
zgXX9hDiCL^i{NEg`u-GHo-3H*{)HhO4qs8+3_hVd?+bg6&mn^D`!#$!T_*vboemMr
zuaRiVr(5ZrGaKof8yl%%C){1la&Y_ZYn2f-WN23g8hnA@#fKY=Cu!Cr7&y3u(E|Dk
z158l^iT;Wkk37@Ir0ii0u3-l*2#6>Ry|1|4PB=r|JopHAG1eCB;oDEqapI1B<{&c|
zv-w>c-I+KCw!~CopMe>9e2A$_YT9y0Q=eCCo7TjLWpl|=68%vDZK$+W#lznPwuzKU
zO4OdKk~L$_<vn`ryjv{iX!fwnUPBGh5XuVXh_zB1@3R%J<kF5xHWLkZ)XN$uR>Wgk
z<SLH@)suI;b~#!-AOwj_#(ZGv!TGeT4Ryc8Kyl0ae|0J73Jak3e5bCK)%*{ZrViL5
zO9Wn2=KVa@S65?xa9J9@OV2)S(2D}L2=5kl_MSb~_oNoFxHYJHAg!7X{#sGb`sp`9
zf0JJ+sI30X&wsNLV<A~PI{B^QC3KQ&Un)9l%JJ9Ivn|i&V1GTCeEs0I<IfmuAf1(0
zs)W)*>r<*aeiOHDEac&?%-zeN-C^`?lK)J(@jPjQr2OL8N86tO*4t5ga%aR1C}ulx
z+M_VfYv081&v<dgo?%T!!l0lf#n>~ikbJ{UBiKx=*<l}0us#!k1$@qi`wZlieVY(@
zZoA3~5AJ@}P3Bd)K}jK{6DyQ9&lfD~#zy@?OyW-6lsCgbJVc6DOO38fXU?n?4yA%l
z@|L}&cyuLp+kr6?(|y;)e@i7SaF2s~Vi<+y<{zO2F&gH=nA+;iLfgDmXg0~;ppH2w
z2{-8~;P~Z`6f~5eH*EZ?38BUMNTJ1Z1$(va#cj3m%1{vf8l4o?4;c(A6ifF$r@x+}
zW_x#}`2)Q_*cySWmd}$zlUP5JOw}Jdek~$C6MX>4sKzr;=IVucrCsRrDYD!Gg#Wg1
z7+_ry)g9+G7j&YbL3AgFHc}R51bxnZj(Q{S0sS|!B>ZBWp*K^PNE1|RY^W8^T{RDd
zZ}|I@-^vZ}%uSyRY8NxD8rqas4#hR1-|xoqB+WH%x*A|z+7iaAX@Dy)y$Vx#$g5gV
zrJ~B~+H0YmmMj1gPS8YnRIQs5h9(W}_H3!(FO<<CI6b9Jg*WVuT0b<>H$PrlAIwnO
z;tyNt2PQUtkZ<*0IE>wLKbz85CmvOlNaapl17dJ0=$EgdX8S=_KqCznA{;wdXyMf2
z=?Soqegv$D;CVb#%4~8W-1%lvqqh=)ed_U)eA<=l2`-DLD4r=c>A%{^4fge5-7^c4
zRBZpefK&_KpR)DcQfa`;Rek9zlO-OhO1xI5Xs#rr(fC$s@3yIKPt?V)%(&V1VPP_*
z1rg>=^WD{<bzY0;^&}WF;LpNHNTOP7P%FD+c2;8^R|@6HoK{|<)m8PJLb#y@ozgN}
zzAF8auwu^UW-6=UR3V7P8}?~{Eqr3E)h1)arM!hpa9rRaCi)$}4jq1zmzUz?0tQ4F
z+*aknA9}XM%g)7xb#ZY4v;#w?S)UDL6ck>fMPO`Ov3ahM`_*o^{^1CKH9)MspdqyA
zz2Sj&bT30S9c#fH>$mJFj~&3r<|cJumgL2=D|}BuK_LrPVSU!=u<|*R%<#xk9!vB)
ziW(O8o9b&iiZvUf==Xp0uGb<fBCn{Xz!HEr@N99R*u2-MiIR?ifwA}bdsAW7!>XFc
zn#xjFVV@+OH{WWk5><9gbQ0YBk%`Y)427R_^a&o1$J*f%Jg()COLJ}O@o8ofEN_mS
z=eDB^e>r=MR?Es7W*6Vj$fhx_u+bM9wXs{sXs#uP(-jOdUM_K`3B4qOd7L&-sQghF
zFa{(M`y$AIoW=N8x=M@4CdjAh4eO(l^1MqjGnE2Nsirp<quyA&K}KX!6!rk#psD36
zYUPwRm@UpPlxExgUyB;+i4|y3BY=;6kljIDNNPL^y|Eb%1BERA7KJ_J7n$3viD`jp
z$H)ZMhvlIP4P7ZDWbh>kxg$UR%LmSIL~^BNK9zUZJc1Y(^oG&CII~!<F`?~yrWRwC
z+EhJlG`>jr&INUE*P@w>Q++?$RMkH&<W)DK-e{_LlcXQVLMKkt@-f$bs;os&{scpq
z2_3>(eSMhdB16{Ps?2oLqyy8C0mEH|=*CA~e^QBZ+eDjLDKlu6S7ZEZXcYf|3j>RH
zHMq8ePrDFV+p5GZ8h@j_udKFdTi%|_IV!`kwwB;x=^H%1VPL>Z$%TO_t8nzMfsRe?
zXQv~MjxLH1O1?n4^}(pQ4I*Yh=QDMRIW!<{qK6fe2xpGz&3)j&r(gCr<OZeT1$SNL
zPy08$+p&XE2y7Bo>vuevGX_?c;e9PbbI*RU2TnOYw|~~pAxPNg<$@0W8wN299A0=q
zGt-=QVg}3h-;MI03us4>NJXK0=EJntkHzaM2R{EKOq%1)*Y9a5UJ85E3@;@?0ogY!
z&X_yp>KL~zovJRIy1?tNn7BCmRmri?it9BmWhno)>RlGIxS!Hdfd)4Ina~BT{-WxC
z{JxOTVlBg+ABUIMRwuF4C1H1lj?Ud=ZqjQh@%!7!mt?p-jWxCKE3ebC1(99NBhThN
zHdfbyQ_qnPQ=eM8^!v_rln~PE>gxW*k`)mqxBJ<;eb84nJRpQb)y)R0+DwWhHfHa>
zL1fz(KFrOMBqz(YUjWp~+k%&a0^t&mwhQ&JUfI4VF3U;c7-D`^U-;I$Bujk3-z-Gn
zPplmqIx{9#Z`l`r&bEYtzB)RxkByBb37ib+UBAR--)NfXT2q|~=Gr`|{P5@C;P8h^
zU(Y{tzpiXXoLwd$a<-bRHfU#kHxVcP^wPT5so}@4_9$$xudEypA0O``W@&AvlyOCR
z)yu`knwt4u2JEBRf1%C>zvoOa1<$xj^c!E6VS`cS<JsiNg7%`sbKTw!^9w;m&ri4+
z?ZpsKnx<!XrppI;R1)=mxTWeOcD|b5gD-D|qLb|W;WfFeXcb`pShzkij5OjfQIWM=
zFLpSW#<_37*wZM$q-2jgu-L?d^BUoo03VqiA~3#QBNNGdMlj0K4=s`s9?XA_#Nj?g
zEH9THg|__`;QoQu2<(T#CfOBH(w{v<)eGJV%tG5fJ#jZ11Ddx^5bBr<5+-E<Racmy
zu?Gb}51O@dirEa_UOL}9m%XmIH&}hTw_4!-wITpFu7+zmpHOuFde2xnh1`>pNdiBC
zB5G#{gNX!5&~Jcy8dr5}xx`nTE2OAkMGB4mKyS+`QB@>KUP8e0rC?EhDQ-8IsjhsT
zO)Qp0kB?VXy?S0yM&&fs%0xJFjZZ$&&78Dhc?FR67vqldtIbCh{^DxcD=EUieVN7o
z`4<ZL?e_gMKRcq)@E+v*A~AEM{4jLj@Q)RxtrvT`i}hmljy-Seni_$aR6p6?`i(PS
zLUG*t-so&73#j`3Cd;Iq>rhJiVuC54Q(o}Iw;o$nS`<H5Ze&`0NTGt^r1Qay)=M3R
zbg^MwguFY57C9Lv>SDMz=rNJs@PWSyRx&m*hb=z&H9(R>2l)i`YhyC1_lEfP6Ci1v
zf*<<GWp;jk_+C3XJ~a2ch4f@iUtPXUVd@{sl5&iPk5%-gjk$&El(irUwF->@mE6qF
zr`1Z5t-OcJt(Q#Yr*-sgXGeK`LtV+gS>uoD?I%iYY=+zdfm<ULyG-t7+*JNW&mIOq
zVS$3Ah;Q}cs4S+cMM^^HvgQwvDumCKxZ!z|)>Jxd8(}R6Z-@)IYQ70y2@E`aG7`KR
zNkR&B%d5Jc)_D~+LM8g0qv^0E)CDNE$Jv175>y-DLeKb?o|BlE@0EJ5^)fOd{ju<3
zvp1NYg+=3v@&+x)axnKYQv`p0{<1$YyPB}`nF+r?A}&sIfuEoMu*Njo&lO*g5QU+j
z_Q-p=O0z4+AIEg?hk$!Ew1OY}1L#DGiHS4Wui<Sn%I%CG%-zT1#U=NkEPLzVh)Fa}
zb6MK{GYgOTtJiY;UdQnB5j=MzP*)l{;1z4U(a-R-#aYes?4ap5JO5lNc;MR30!`Ug
zHPivcW{L)5%@4RwY}YU2G%&)fvZllc>Cdp%(>qVQ0(58YGT883$2dD**kKgS$fQ)@
zwHYn$@`w?D3ZmIrn(VBuLwXH+KDVsNbW>m{w}hxf!MI*ywJD0Im5r^+x(wLC@aVvN
z!wq`?_rHUhfFfo*S|Jl#xa)=Q_QU@x&=W8b8+@TWLgn%CuR;g$9bq=IH|!mpNQ9pI
zdYF^mkePW8eatzI)ipun40J3OH0^e=!gr_floiL}+jd8z>6h&{6G4nym~13}Shf%m
zuok7!PzL7unVCNrj49H$y7Qpz=E$!}_JVq(^jKOY!H50vz?Bj;nFLr`Ebno`9m|NG
z-m%jpdah_Fr|31M36#tzj8)!`r`qFlt}Y03XVkLT4HC0)qMp>j>9+Z(w;b)6``jq1
z?>JqUmnfxwKS<l8PINKeu0QEE?>#I;##@YS*54)goa6zU1M}DP__xA-w;d};`@!<s
z;E@ERiyO5fyh3yLo!!QS5_Td^EQde^+6_nrwY<^+PKzDzG1w8r^?Za_B9Us*hw?gg
zNX<xg>sL3VtKD8Aq_11#q<Sv!Q;11jKeyo5w*{r3=F3*%%>1Sin4_ip4_lD!7PErw
z8A5IRzRK%z+*;s5tV}+{CCyQMYA2|9AeP9{*}8Amxv(j<^AG%WWa(I!W!2IB0P}O~
zaK@NUzFt6otei2r0V!Nx>=?w%UU)mIzWdm>5o)A*+w`4pa2vB2DqX)U`FvyjPY|sf
z7wTUP<3E3}nHN2=k($W{h`T+0Aqj^q1xnR%@4#O&W)$YJoMt61Tr)k4bQ){i7iR(y
z)xGxYt0*t8-aYl`u1|Il=2>zz^)ITntOVltmb?AC56Nem3hW!#+RVfrA80IEzJdYL
z+<&5r{O|7W)-aj>KtGyD%u#ha>f5!_TZsSbp1<Tqm+!@X0hp4V9bG~~qC8MyKKV-p
z>#t{Wr-bYbP#Z$`%ZbCqKce9Mgo9tON`d57G+UQ)zy5*YN7Gd@Mu&N)!%`E4U7?_h
z<E46ymh9Y7%j+Wok*TZ6sAeOQV)quy$xIb?`=Gx@h_i>*T)wBC6VKaZ<brRe<tK$2
za#DAI?6Ydw@Uer6C|l4|^(6q1{?&PutX%0b5xsjhBRX;c*rP^`?S5{pAt=V-++he}
zFtUx|Mf7`KqtR2HIV#Fm-NgAeMslx3tE|&z7#$6Wm+<P0(@D-0%7sl(D$}y=py?!c
zdJB(3Ul=9m<PZwS(GOOewQ`HZx1*hCQ~7yCQ2Z9d=O)kGtJsbLlGQ|k2}j{b%&0?-
zffHMPF983CJj{+XgotPSCVd9K(@QYyjT^b27r8g$X8Uxw7r(;<KkOdC{4uUKB$jvU
z*4`bG5KnimdR^hccJTg}AUoc;BJ*&0r9&ORJ5r(xR)Q+?rglc0sx1QH3ZZebhci4i
zMOuYBp<Y(RW#&NwRbXPdH(4ts6I6Vst~{2>?sHU5*hKPL8Dd8nNO_O3X4?=Nv$@j2
zR7nZPTh3bLElEL#E~}wnwORIkS1SpftZpqmGs^p~AS}m~DB;;)$Co+jXE(*d&wB8W
z@<_cJT&B}nldXw1-3J`|cP4_5^05FrI%v+6H~6cB*=irm>V*a+>8l@awx0jH?v?Z?
zo!*jKK}I55Or%<Se~x@LCP(rLR60am$F`q1jS5vZ+T{`f@fB4+@%}7=`fF43Fx{8M
z$08<7M8C-~{VrcV?=3dlr*nKD!)*5B>SK3&RJ_h;lE`}VyyDnFvTD*U$|ASc6ziTR
z<1QqY<EJsU5xNs5O(u!G%F%%B;n-k)ZD78Vzl`$SkN0TePcQV*{qS`~nZeqoDQEDY
z4^v!XsGy}>b8J+Hd(?YQI3Ja0`9$+#*p}+;9#w4;>0}F7i?5uzx?8%^E1gnkRkp$l
zqJ2Y`pEjq9eD%F;&~vww7U?cT2C|Tml>GGb2NNhNanRDw4;y|@EM1R>Oy8(ISoK@b
z*}9)OLG33z-zH$wUKeZMdGpv15fNPgrmSTYRUs2NOq#|whx6qD-T(Nx=e=<MIJon+
zJ?kou3)=)K!g~n|f_re_vt1!ZQMH;+<K4J*&_^CtyxC5(Cf_s#UyjZj4%r`|MyIBD
z_pDxHo<9;@j`%>x$QGkdU&M<4D@3LxyUoRxqa8hu7wcTaTm;8EHGy_WJLZ7ETK0l2
zLb})FHyN48@{PwI&1+@Llc{oZvdPdPFa%%X#K<GP&+EctykGbiv8X_E?8|O}%&jHl
za)SW2^sFdvs8)v`@1qyP!)g|3AfhMz*v}zg0apFgb97w*-h9N<!sUHjDv6Xfh`dYY
z4N4A`wSVru%tPqfbG2(17nraIi$blwIVJ7GBj4Z{h(HJXz|g;!+{z|F+nkT!g^?BM
z%Ur9kJMSU(ak66YiF=MvuqLXb@`n3z)ep#RQQPGboobKY#c2}vMvh@g*NlBuRA9Y4
zLR4PWrtDFR{bWm$PE0&zgY(&(tzc1lQ>-i#i`FCO_WJ<+zWG`0dHCTs*VYzxu)^p5
zyw4Jql&@A)e`_>>4;7geSLbD^f$?`)BAEk@cj|VQi|I~QQqci4dt}*yYN4BRT-sSl
z(fpz%=P6Svf{(Kl^4fK93llNDH}!6ac43?8O9}VjAEGVkx%^z13ku*N4KAwl6&;g(
zE&<}HS6BQT@vo*Xvci;c$&`=g45pmXbpl$SIlE!Ql@_)<WZwTCPmkB5P}Z4``mM`U
zPH0zZ>y}$5(S@fO{+zvB)KULSI(;(Px&U8VMa@lB%zrmb?DsLYfDX*$*da+;_X+$E
z8~orvGTgb>XV~>yk+^m(Nt2yzm|w_aU&>mnzx`IkOk>P}ab8v)BTC){pO*y~WJm7<
z{7Zi)quv_EV$3F}2u8MhOYCg9>?Ri+oz$H&A3THKU+Edq1wun}D=V?xWZ0TnO$<z1
zjlm6sl#^bChd(Y9)sXXq7Eia&xTKm6_|uu$vW3!h6&~dBs_q5)HY}_2KKDR`mM80_
z>IHbNuQw7dN&c~<fh9>Jj9Xd~eoB#JX!~Q6ZDpOlS->#%0`U9UKCGx2B73`+`dIdZ
zi~v{Gz{F!N^hYPuKeDBedZxEjz%B<pJv`u4TBFLd;|dOZ)xOA#{mOV|JrBALSLlw8
zhO(YEcZh>kc+r2F#fexDbOQp$JFG1Eu2b~;Y2*Igp}Uun5yBUNmqe7Z=S|O`uTng7
z`87gkH*R^cpMFxD9Z-!69~IqQ9*`_m6nJX|0h#3~z0Uy*tKj$ciI7Hg{?)ZhpSAU!
z<Bq{Gg=zo;3P=Eq$X$dn1>61^3o=WK0xIB5^rNY^3$RW0woADpo6_(qobjz*P^!*2
zQJ}YF5YzO<Rb@n_>axOvcImJCD;}U5v)6gwd|^CL(Bk_<lC~oPRp1k5I#w(Lz^buh
zmN^OfpCyfJ&)AVMr-_?PxQCt;`pg2AwvYNN>iE5xD5u0JTFrTHwT^p7%gyMkS;aNf
zXb7urtiCnC!#e`AWAW`_lVa#UrqEClVxBA<JV_7aIY>pSJF%qFWmwL^@$!7CJ-<AY
zSq-nco+r_BS6XsWo00N}F6hoa$&nZRru}8<<Y4D5%D&*p`oQYhANLWpv+;@qQ}i}W
z*A8`#M^{;SDWssp7CluZj74g4v%OHB8D1!k_Zhss?(m)Can|bNb;DE#eat-{hCAy@
zczE}6C;}j(EMJ<rpNKKm7vMr(z~U|e)FT9W_>J6(xvV(C;fRh6g@}zT6RN_yFVJ?l
ztc=tb>g19c^vU;cp+Du>H_A@AM|}(Qqfu7N!?4=t(`;okF5PwB{Tz&|O%h~A;SWnj
z_y8}c6CM9u<DQ3uB)Xrg4x-IZSP8uq!tV)I&MYN8WEm@1M1kIq=VTiCmild~h3Jr_
zWB8tvg)v9tQ=z=vvJS8QCF^URxR_o0Y!ugr)(ChsQ47yV@nG_?Y|Y1yzstsi#tBXq
zdIf6C{{_J%jKT}bj;|k^XZ6*1!-i2*lZI68!BX)_HH!bJ$k_$jG|(~d1J<eVLO#TV
z8sK^Og2^y>nieB=(dQQoy%hor_lzDjAVXE_TOOL^!j=wyeec%7Eyh#I3>DVR?w8rX
z$Qe?PyIe0ESL(u+XAD|dH<<;aPi>c8L!sr68Jv?55l8`K;LZ_%mME;h@prE;Xhpdt
zfgFgwT3Q;tOJDcv1m1E3gu^TlsgVD^HP>sMMDSm><hQuf|9p97%XlI)z4Pq)N)h9>
z<FbBMs_*sKN$mA1bP2M~R6~FpLRa)W1vyq!Vd(Qj(%~Ibl&b7zzr#r;FZ8+?V8KJ>
z#6{b-$#T=S*KAm3w))9aFu5G@tBrU7mu@Tky78Xf6P6Uo4;JgW#CRku0XjhpSU08*
zgN;#UYF}@&M?6JoBXSt*h-)s`=o1ql>z|J8cj|mYY3cQ6mwb%vgRGGD?|EyM`?6L+
zi~n)UNPpXp=f+4PaeFv%lqH032kky9ef%}QhEx`g$~2;8ZLE8rLFI<dn|bAt^7#-W
zES-yCbIY!|&#g(rxKd+X?rG!<#$L=@wktc?YXIo|RBcaDO0}<2xpE@&JFj4*e(z?w
z6#l{r`68WgesWo<@yw%}Rchr--rg6qbL>^Qp%!4h61Sb;fXVR)fd|l{>hCzAd6{FR
zx<63hvx}hj<ArS9COdi9PWV->Izx*L?Lc;OR6j6698W5(u5m_=1%!|n{94{GD6Ivi
z{vj2&u@*S2VsUKv7)hcx2z#vZZLBp3AqC5Vsz839zsQT;foN-4s5z<0OsVLwDye~w
znk7@f=GXWQz00;@AVoJ<i2Y2x9lZ;DV-K+wsishbV!<lGo+Q{*ARNEmU8{O(%=d$z
zYt8e8d*aqn!D1JdY^}MD<|~0nB-2&-1jPgMSc9W<wR3;o94Znb5)wKNN$BCBY$Vy%
z_h|lu>sd|YVUj-<)l)(WI;i(2W>Qe1-|bmn6bijwVXaT-GnsU(v6iJB9JZIbiVPWv
zUK9P-&FH2rb9@1)y9MsQzL++;5Roc2sScF0i%Z?t3Bv1xS-9C^1vUS*ZQqkWdJpp9
zyE^Bd=HUG`_l?+^?$<|RE4RXBp(z~<9frkVnca9D)b?Fd5;EUwj4%|U;-T&%ieOiz
z{lZIqEpuT=9aDM#tjWIP4HwU$H5n^ue)zI=_j1ndk*{x&6>z%m`3(8U$;qijL(R|4
zW!AIJuNXcqU#%zSX7dyR_jjq+2uE&cy!WPS(?F6c=syI;L%G0ssH^x812Hcz&xRVd
zTcANGXQp`FZGMV1Rrf;=bH3uoGgZR-+pm0jM>Ahj9Y`SLv342s&F-5+XFP#?0-}nm
z&f4bfpVit8y1JoV0GM~0p{FLWqkMhHO^rQZ)*pGJ-QmEo9_^1dB`F*Z*H79<66kI^
zmZHZ#<BmC0O%z@CYUcPh>3yzwO_Y|}0WU(V7d_p<UZ_qoFmtF*jYDgE5~$I_wr_~M
z%q{-eg*>c^B*^-LjIxy+Bjqn`3$v+6dy-KgX@<hP(bE8pcOh4A$G4hdyO|esEy_#%
z)_WZSlPJklQzy=o1;NHF2iAB`^x8Yb8b$KWJ75LEZ>K&skPeLsiGbgXcdg46iwdi)
znVL=;<We9u$_*+g`RI4i3X><>!!*mZo_Ss!Fx)4&bc9DNrOY+Mft5CP%l?GpVrmys
zIYJ&=Pv(LxHfY6_@<StFh+}Y%g%RkEgAEo-WG$AiTVb%Fs!W?CCorC>XfeYm_<H*j
zHun7HXVAyjjuT+8Ip|B@`c1l&!_`d^`=3UG$@UIn&hSP~zN`HE&u>bndTHIAwuzw@
zjMqsfZ!(=QzA3GVn`Gzw9>b*FP-`{DTPJ#eLZ#{nYWdDN$mx)J7m?MXfF$}|d6-mg
zewM+6uW-FVdR^ya2=Qk_YLE|urCXqC*D%FxfkNw<WkrYrk=}qvj^`gy9Lv^I>4QJC
z=+@sa9h!AElP+f`MO3BNn<JM)IjdNEC`}f#SZZ~2mD5tg_amg$v2pro9+Vb$#(HrX
zL~t3Tp2b4y6~+>GL^ievz)S|#XiZM{Qol+|>F56Z$$mPSk!n-Bk(*oTE*m<7UEvEy
zCHbPNn)q1NMa*N)rz=!*!~Z5->w2IH*k079<vt+g=fz2(f{jIY`h;@)U-`fOO3V9e
zC4BXZ%LbJ$eDOC6QqgS*5<R%BZ~xpit<nk^0{H8x_>)usJ}gt8F8jT8TZ2w+PgX82
zhRKX_ul`Q-Co!n88~41MAD7iOnWFU`_BfV)sAQ5AdA7~IJtBY&%$RDSWDEb`62Woi
z`8dBkbCaC=sT06}PuuzCjAggf*0SGdi5Pl6-2ubdg_WDV-<9^I;}lxozM2Kxd|s)g
z>i&22m=joAUr=98MB;J8u?}iy&B->|r#ka~>Yhta$N*nw`gm&LI@fc>e`oN_a3dyg
z&Pq@Vi0R<s`#}Tx>5`C~5&Ub_)qHn^S9&`sBOy5RTHn0b5+eR6b1b(Jo~Aaln9zjs
zO2=b7pTJh<xFPt=ae?N+3OH@Gz9`0^;0GrNQ*r#UL<Zev#lXFs_H#igZ8yfUZFN>T
znE0I|`(VA$@X*>z4g}I43)G<<cv|x%q{bzw!C^GHGi9&cLQ6g<DkDA18@glRN=}HG
z+Awo+DN1J+o?2Q&=eX8=E6!t_EQ$X?Gj<*M)uMz-K<ij>B0+Fu?6aiiC(io8zxh9O
z1fyaLQOzyuN!gVqnPzhocz=nkk55X0Kv0*zx#Imek)G79>RkMx^1(Y>j=Hp$oM1{W
zIwb2>PrhWT3#^IwyRQ)wvz)U|^eiMYWzgXaOXmucL*<Kh%qUZRIXpQ%S0bc6YfRhh
zUq$g@ClEy6_)YMe$T6`*;sy+IBsULsF+8p+=;fa{NOSCk-{qVOMMw)rBoz=Ox+hxt
zXgxAe6AOi+`MQ}!eBPaY=uK;(hT-+c3`~f|<S<rb&o;z~T?{sDR@0#$iGNmPLeFR$
zx0K{@=Gx7O{}HT@@I9R>{p1gGRYS5WaGPh<So%%h&oBE!3iK5DRrb-v%D--EG*f>x
z+@~(uN-?{1K!geJAfJwgWNDjK>bdliX`#Uv_GHRuESYbd!yvn4T)@iArub<syZk7*
z`D?_Jnx|@UQH;4|X5!t9Lw3Sf8IjpERwp^jRZb-eg~O}$By5=QP&r@4WN&^<_Zj`J
zjEnfTL?#nCVs+o!sPM8O-UJ~h;$5ffeU)!jSHV*c0y`(eb64eKRf{q4qicAJ49@&^
z`vlo{AHl?J<8gp0kgeuFVTMHT_Nbz-U-{P@bMx`;41846O$r%`?H@~K(xBNgGJy}q
z4$AAgWv=X)>drqGeoGhg*QZG7{6Xe(-rL$8GsCjxQO;|3`5w5goR!`0^W7XRC{=|$
z{X@B{Nr{Mve8{k~!ZPJY{yydcs_R?lEbrPNC^9+$?`67-Jqcc>Uu^UD(^*}<WS`|`
z<fxq#H&>LL`~WEt9acHCq?3sB*sXHCc^Ez2HD#38@OBzBq6nRQc~bGr;n@M>+Ye`d
zTUKp983ARN_qSs<6kOZRgRGSP*rrIFgx~JY)E{3i*X!H9rRri)8$pWv28P7PzY*t~
zK`@9hkh**5?GTl|IQBt>3><$^&m8zDOqfuY2M<P;RNhy#Ls>=LAc49E4HGi2)M1j@
z=Y*v;b;Nk{t_RC>LD*$KJGM-r$1L$dLz-iTckJ~~ZrlVrZd5utUM!xi_u%fST{*g<
zA7C@KeK82@v79jdRQ0(Nw+CUFQ|nZXu&6MysCgD?ZsidB;a0wG-lCwHuC_&bcpi=4
z3;NJ}fxO4apxzi4uX7Jt<QD!iR_eoKrma{IPydRA`Qxv49i9DHfytW(|Il#M3C#(D
zQ=~HsPb7d*9hHtyN^zB2!gI&i4ku(FWZ4>^R^IIz;F5hFq&-zJu}6;Ur_A9K>=Hyg
z^TmsGXp*Gjpe0tRi-eEE%?JxO^LC`?IYsVUt`ay;eslZvVokLE_*H;ZpETlvEIjY+
z8DC#W1mWfHpz#RwWPOa59$4NmJH#7s=R(?`GfCfI^=T9$$oPVZK}u^K5z}8SKE(fJ
ziGx*hEPjzgcbk-E*UC~J8rLpj7&2~pzTb@CaWzcuQ%j4<R>dFi)D<_7ArfLLr45Z)
zU|PpUOWF>h-oU~_&!*paH~OR<>8(G<Iv^E2#=$CCz2+93>OL4vl4K7lGj+GnW1OO4
zE3KQ6ju%vzBa}*ZXjtT=q`(M7Xve_qMw>wxjYw`8j(DPvbXR`{4IHlruWf@;SIyRk
zWIbbXyoZPC)}=dW*f4k35RjF)MaNVGHC8=Sz5jYk-9(+m$S%^n>cG-eBV1gbQhSCc
zk$F0FEOKa$N!fMOHR`4bZZJ44;bJmgtkYO820Pa2RnC=pcn*0C%6T}cW}BK}YW%qJ
zD=E7;<PF`Bc;!gjd+eH|TTWkA$^K!p7}-6QX*xK=`Mh@S<HlfD+JCU0F4csx?g2sh
zKWPw&019k8Ql+T9-&%yA6=wndyI5z26X2YH+?NZ$_)~KaJr&%KO-#<{p%VO6ObWF7
z%jn|0%29)-mV(FbYo+GY_~qkWB!PTe&WO6F?i5Exkuv7O_@cH{QGWJK^FS)RXsYa!
ze5WP72zd-uxSvh;a;&CO_wqz?F3??DH=E+|=G0l|(^qLP@>&KTp4`c*7VD&%C<!QA
zi=Fz98!DhOEbz=t3Ejvx=D9BmX|$WX>>6e0<nU8x->l5Cfr*0(&kmkj^p~~uB8!(f
zXKd!?c*$!zNJkHfwCgx<ZTrwDX}VDd=M^wh3Dq@mwnu(Y!7G7#(BS+BuG4;aP@GGw
zc0PdQk<%hLpv*YN{JABcBC5%PorEDyM}Jq&?e1OD>bz;Y6kbTTS`+xxy^xpYALz>&
z5^CjCD9_kxYQ*h>7t0Ng>CVcll3f>~NOrVooGNN?^&(CfZK5!;asmis>M!=2!s{r$
zy^q5olh82ukWNUbqPG!YbD^d+PjB)uH?NvhLBn)hmtoe~knOJYn=`01ffohT{Ouzq
z+zf2_{Z^+}q!tTg_s^r3=a5f%Q54FN&0zPAy;%KXEv`x#|2<?#2)~OOSLroACI>UF
zps5YUVs~D8qDov@7OD$XX)uR}8nU&Ek=DWo-48?i#1?^58zZr?M7&HG^a_xq_b#nE
zy8PJwh8OnV86#l+v~?;k#CWinr%&+GR#6E%BrF5p%qJQ~wl7vz1Fx@d@?x&+_K>Sg
zhYdAag{_K;F1Ta%hR;wlYNRt9=JbxdpcbA^PKm1Ir}!}o=m+X(@MF5Oy4*trD86A{
zE}L8Sk;bklQSnAtrN7v;#D(YEO0r2UG81pu$<ZM-Slsq79l6ROu{Lc1a|5<3xTyxW
zJ?f7N4T>sTn#f0m?YOP^KloGBzdp$rRp~uUej1JbbR&N$Io8+O!qDvFmv?BXOOR?)
zhqtjo{w?@`HNmmG5Sd$+z#cnePOfVqGK<hAgSt9-WP&g5zLtE)VmG%f>cAfRZTGT6
z-Oea7jP?(ijAg9KX~%{q4b#=<H(@Nz7SchsMkm3SbQ^kq-bJ3>Gp?0N{qf+t>nk}6
zF^9M81wm&UJok8Z(W}cSg!FFmws`A%NQ)1yGLNoO)~8T~44$nNJ&=k}v|UJD{O_vU
z3aA9YZDu&Xer5<^gqZW`Z-xV7cedEqMzX29t9=6294z?<#i|%GrX%2v$EN))1%%M9
zPJ&0;IVDiwTJy5+a#qV<678b!gY~pv)Wha&xytR#)30dXsGcJEVbr;gjPozU+j(w|
zXZuZ87v>yb_~*3}$TJs>QoC+VWaFL<>a3}O*|kYpJ?x>`fXBmozm>^R7Z~hveQb&}
zd4cK96OQr_>It-J#B{6)GBfw~w#oYKYNMBziEHIZ9lL(Woi+>b`TtIP?|$mrP!%%8
zZ=>F|Ql`FIR?xA+Ak{T{GUW(S2TksRLwY@fZBX;ZS<hT4>wgm1-ePYKI*47w8IgDK
zqldE)n4Bx<f!R>m`Wu91&)W+D7|%6Q3){J{{JJaF5Q2W(1zG3{FjXz-%jxs3J2RW<
zD!RX2B($4cjEZ=cCZ-n-_;o0+&YfH^IGJoTl}Z+#M6(<nJyANSokp9QB(tCsS+E#n
z5$i%pqu>^4E8{{C1P(X07(mRV=MVfOmxTzf@OCyKn|!l%i7-mB1uh^wb62ognRk2H
zbXJv=HUoi9TVJ>vBF6+g@%FwpIeDYvb+?*|d^0+Y^aknmPTUjJXhth+#_0Jp<hINN
zH9omGo-J_fH<J_cvdRq**`jTR??bdn-JoZi{;fz$0o<i0#}&&9d|@_Z-d~9Q@gJO8
zoOTTbyZ*wjuXtcb2ZOhUTd)@W*zqo)UsqKNYlj%EalAQvG4VcHs$SN_UeT;rixdym
z=C!4?xIhsuCq$FZ_$7{(GdJ<+I*A`%o$hp6*@HqR7(9QoqV)5C_$kZndOa&m=@_qe
zD)m-ScE>u0+V2%rJQqAIT4&vETp^QGUBfS|)q>;Kr9xUX&-F3Qi_gqg>SY}M8un~`
zz((|tui;S#_odnbLSsIAzYt#PWEn=>V$?+f-!v*u>+qm&ODFheYq@|cm37^UvU+^*
z;<<7XUTTjC9&e&zKM5X7V#n6DDsrml?7t45#oBE56!EC2@x`<6ACO>;LZ`AdS(C!U
z(=UcE+1*pwz*lUcobn{BVEv7dXh*=D(sK}IqMMT0+-}4RPmmFB#N+j0?_y3(sf?nb
zk~teBgv58|B8oZbxtTAbB>X^~?e}PXlecZ;CfI-Dm(z8NtHg`_g4)@MAQBQHtmM9Y
zvWs1uTA}VNJ!-VFK9JOny&Og+I!QrpbqOXSS(;AD0XX&`Iq~ToFyT?SpIgz2vMQSO
za=C;_OFRP6s9l6hfm?e+A%c)4$+n{WS3LR=k?+B^5#;JgS|eN!q-8@aqF^SUPsZP2
zIkhUysamsg8Bx@Eh8nY4bxZ2(Vcr60LE`hc7?&{>UCxbK;T;P!pYt@^lRO;1Dbs0%
z&&Ph7cXZ3s>1qDih*1<<G*UAqr7VfKzTEa6BJzgFTWuIX*Q_u&zEuMW?D6Jso5e>4
z+~qkkhByD(^Fpdg+#xePQs-ax>pgf6QJ4kU(L}QKgU>tY3+H$8hnCGhhd?7l2HPY_
zckFf!;bMmoI+^Ic{zN|DKC%S-Kz~u_9be2_T<nGTT_ffWmA8aV%()S7FVvPeO*}PV
z@b1yIlxip=e1uN%VZ&2UeaEzgVGeIBg=mZ6Rf)f%s%U0wL*UnaLl654rT4}KSYC5u
z4x5S`oIs-DFgN6@#wqRal1w8+^uD?jUwx4fBz>X9Z~#g7KF7;fZWx6O{6gH(k`7*B
z?GD8Bb1$MI`bjZ%KwCLwhyMDJDE)U`tvog^Zn;U$b3yB#1JdBiZLvOmLv;0&U3Z;u
z2G4Q&G;5pik;Yc%(N$4X9)Jt~IROGWI6wUN%QJ-5rm}f2x+N7Rg_0g3d8>RlwtbKd
zx&c#CxXhPp*eKF?bMgxtywONKnY5^%Y^Ze(J`wZ-3(5dYuE2i;01NQ_2a0Q!FjsMM
zji6!Ej(F=@W{7U~O$D^c8bepxhe`uAk!S}qY(IGLbGG&}<BNa(7@3ySt|sBLTbYJR
z#xA|cVhHE$Z3Lm7%!o@MB}m)BZq$UhwJIW3%J2OW(~g}+CsVnW^kuEVdWoiT@|}u{
zEhBnzRuI`OobhY*pa;b<j{G9~2P(hJpafgc&>{-2i4U-5f{Gud;@8hUb0#U+_+ssY
z&4si<tQ#KRVMSaaFClSdJ-1Xovv2n$`a6>yVW%i!)2@(a<adc}ISX*U?x<SPth4}R
zO@-i+)d+Y4R&l1eZ?(+X@Vzs|X~gDcMmMmaWW$(jAmugB!B8i(o;@vP^TsE~H${5y
zxero!=&WA-ASTE<_|Z^E?k4;BFxNRm(YD1>Or0P}=IR`R@UHs&i?e#jxdRs!hvTwa
zN82lVm-L8)hc3r&*&}5@GY4CNrGOgf&j5#8m#~)Ay_Otj2e|YVVWda^wW;Ah_htp)
zum3-SjBZMOn23sY=Nt7ttl@1JeO@gMgHEr;WvQ#11tTpa3&#lhJsHq?2f@*wDm6bH
zPrmN&6?!Z>KgCZ{W%vxb{W^6YqR@LE6ps5WEg*0)WGsh9Ovmd}_pJYgMEbuBQw4Bk
zIuB>mz2Pv$^@)e})rF!pRY;4bWh#>NL>AS-U!~2_Wyi*iu}aC~%SmNX!>M7MwX=gX
zN<tVm>gsu#3B~wgN%`JSr=f(sJj8N^kvXRs3fvLfHN!hLMSYrK&Q)Jk5AC;*!bri3
zYb^P*RWByT@zIIaFxm$X94Zto(6k;L`HBTdAj!ij$>Mr#>CC3l&>&$oIXyQ<9bv*u
zZy~R8dz;?}?D@2v@pUwQ(A%(zWp%3jwh&n`J5cQ>B!KH<KZ|6c_8nReg0TwZo^cT)
zrNKy~w2*SR_^(H;GmWie>L0$Cs+WL24LW{5+Je7_tabBMRv3dc`DE$)FT3w8nCU-Q
z*Z2OAvbe!1E^t#_s3T-Q`H8?zL}Hf4sOyDa|G=P)Cjk6lq#(;n2`z44`Ko{5hE-Vp
z#B(9TJrNAs{JzVlvL5qO5=pmE6qt6y-nZ{1UDs*N?{0s3kH^!-Ci&lfq}sT^QwC%=
zu7H2i)%O2ULIeHo9=@<=CI3e(;g@T*4=FI>r#Y=wGn9?{lLqc<MY!)Nxi8JI(9r|j
z*FB{sx(kiMMB@u{0Ap3Gx7V|`I6OlS(UUa@1oGa<$f)tt@nn|3JT7zyq9$M1Al1zt
zVtZFj&P)};|7L$O%eMzHeo%vE4Z#i+INi0PyB`($7@|=+c<M^t%}$14l1E=tiuw7w
zWDv9@G&Mz70K@IyX)Py*p5h+3V`%FT>4y}AOah@qXZ1aGz?SEVMb93k)Q&+EWQH53
zN!Gvq^96ni-22!tp`H1Y*nzGS9NtGZSU3UU7;?v!w8M|3fKGe=N4_`U?6ha|>{N2&
zj8tkv!y)f%?8pAq1+;$WrtP&}2S&3mX;kfZOPCHAS_1_~M<s<wyj~S3P6`=O5(xE;
z{5Fj%-IUdPF`Q`x?N2b*bLbk8%`<vWskIP;E{|hIE5<h%{z$w%J<KtUn!?f`Lr>6W
z06~~tCW1dAWVRt?*sgwNnl%{MmySy;pcY%&!1>(51OQ$IW7<>cjPF<Bud#AF+-L&<
zotDxHLp(uXg(Z^lB;ruxY>?B!o_8%fACVNIjmpenfI_1A^;fPlZ4eCiE^pYAP5I@v
zxzcrL>}hnue@FU1=xRgsKYzDn;H+l)uWgWCOC{P=5mF&9?$=$T*AM=0cm81imj(a9
zcnYz(Rry;t=XnkTa{uQGsgFswaRR5xOg0y!TGl>yYwq(YRBtEUFTBrdfIWoK{<j5#
zlCW9lkGhXr?Tv@Eoe~=hQ~Q7Zx1vh`H`%WQDiQJBhZbVzhRE1C%Fo1#<6HH_Db^y<
zK|x#*NN5s8-175%9}#)PX^5dw!+$IzWL(Dd>m!H<kskiw5dRav_g)cR8HUe^X^U4c
zBf%&qE}{?LCmV2Lm3fB0bMQxp_zsy&w*e#@2>ta6H{yF9)M5B;QT#?fYROkVf__br
zs63-r3|{6J>^F`xf#;QH^5?H0?8jfI!vRYfPW{IuEf$`{f2gy*fUL6y<hnvnzClMr
zXIScx#$3B35Z|hnx0Yb~M^^d_q`9>g$Eth5jC}ZrP+Fealx2eP?hVteV;m2R_aQyM
zp(e;Vh<0zLJk)R1nC&n(im2+jzjuk}3qJ3fJ7pDca;nN+EDx!rwJ;O-T1w<kU9(?j
zbiD^^)h{`%9QXdSSdQ}z^vLOiAxFiq5MHh?fL<B+Rf_CgkxEjc7Fd;!V5CIq6uicO
z*dYB#xDBRqi2O;lNBAJX>r9pWADI$h`2g&B*AV~r6qhN=Nz4!=(uc-%x3)#<{M>_1
z%^#EEIE;ppTv6qG&y2+K!in5R%9ONDIH<<IlVakGM$r{c#$wQ-gV8BsoSna>^8aJ<
z@xayB5G^!!Smy=>TYb)8@rxq<w{eANYa@(A3Um@TPHt&6dN?O)kH`QV-KDCEs#?VF
zN9ZJ;zj+-u$TlY$HlpN>pDI-Y(V^JWGe|=Eq@A(R6VG%cY;hFTOa94-{$yPa{!U81
zcU*Aoz<y&S1$#i<!3YI1=$#GnbzcBtCc^K{@&o;O;O&V;_d4S{%1WWz_kXe%gO)B<
zIGhJzul2A4j`r*@N>~3088<~NLy_oD#JdUfX51ywVW*;p){Kt(yotb@758@H=1G(V
zKR<k9fsuydt46Na9&h~t+kxQ1NTXfm-SC~>+d`v;4jiK1=K^Oj((ZGyXfcg#5v#@c
zn#_qROJnziG%rhNjM$I)(8@(RyTMcm^x+S!rP=K*mB&GeZ_CxVdCV&{!pokd*+>xn
z>gznzv)?!ebL3<Xol|RzwRPem6XFXD&yzfOK%>6zJcB>4#SL(-_(B_HqdNerDxCH;
zh}ql4;5WJL_YK@YhsYme?$w;~`_H>c-~PvL`v3LR#ql`(#2XH6A#iUUZm!@%9cj-K
zf)u|=7r&N{vm<Z8J7lTof5&sqe;v>Nm$5ZTbYV6Zcagr6hU&xLS~}IuytX2Ifh(~K
zu_0ZbAGaU7KE2X($rywxeHR%dYkEV-p{waZ3)kFgr!q2~k7TRk@ny+Zl**sfK?*-~
zvJF=VC>UK2D*wxf4WCq;QcQW^${FPRrH<B~MOgS5VO{AJD?8GM)E$IZikueG3)K$@
zW3X4-uS^7qZV}fpYco@xF;Yk|TLd43g6f0g^&Q{AO`7+;X~PP`3r6)KnaxOTUOp0R
z;gfrvAO{xy$kf7*@S6?huu_D9odRQQ#`Kj+aJ1YV8F;wQs`%jkhz<~<i4ZXigoHLi
zt4>%oEY7W};f4Q3RL3};D{q0;I-a05XLwB7Q^%!oqO};YS4R8qbe>GA(CC=hQlNJW
zU*Uh)b@7YUbHNU;rLwHQuFp9nlfDnogQRs);eh>{(Wn5`<K`-Sj5c<3d)IO--U4aQ
zb=Gj5%^iWbr6xsD3x>?YjmKq%0X~-qB9`z{C|s97$SCw*G2u9D_T*>TF;NMF3L^Hk
zco}cUQVt3&6#BnsDdYE@pO`FUOq}lusQ~jePwS8ea{vsAbdvv%&x`k;BVH@+-!NcX
zXIh)#-pRy5S)}yR<N|QqbM4<#&-NC%>KY!zF3o-wTsv}cn7O;*YWr{W@Bgm9kx1e3
zoj=o4gegO8#B~Qn1+gF70yZXEhTQS1FNk9Np!(U_jj3TDzyqZyR2-(A`cKFM-!j5Y
ze*IZF2d##W*o}$@SE0sEu+ZC^@)1OggNMT)!f9FDL{nkH(L~LwA$hN29Mk^fVX(-(
ziwXrC493tPgtQoe=JhyUO2fetZ(K9EN6lj<k(j<>E@;(n&@~ga{Kxd5pA_5`#B2}6
z#+%QM*BJyMe)Wsq1VLT1W={RgK^oLRP!Hq+p2XHmtgH1lytEVGVS1=CimNZc;cguG
z+c*a2owu<C=U>>IzLfq!Xq(2;(z@K%LDsYFuB2yFmx!9W=G?kVCW`1zujjDnm&N);
z(f_I1i}LDW1R{g+KMS8;f+h!(?)0cZC~kPN=_h<DGf}xo#J?63gCcVf$R`m7d-$Q(
zk+x^t4WWnpn~0{W1VU;sf7a5~K;tN|ByPa+HS<kU){jNq@f;7DRbZaq4i(&AIP!lD
zx7-{hyW5Pzxma{MM?TD`lsu#HRzyg~6p2jG(}~-0k9tep3v(lTC-7em^Z$P<Izc2t
zCMXfsO5*3z^YlB<>Y5uNU@>`$i@42FTm8&i(588P0Z`%`St1YrCrP7Rt^bd=+`p1M
zS5yF|z!*cok;@%5{8yE5C~txAz|MrbS=yO&%r#LIU>avyYvms}sX`QHa@@9sPxEQj
zff}^#8vSnD7M>BlCq>QnV(?Yz;;H5C!gK(@KJS>U61nn%)_Dlbc;f~EdazCC;#2cl
zI4_k^|F-YdIV92nJ)I~5GFWCKvB-7vq128AC%!yb<J~Nclk+G25qjIf<(EN^ofQN8
zIzGhyzr)*IMNjNby&!q*7PvbfCw%{PW~V#;AU#c5lkF1k3nyis+pO}T(@I94<|A&u
zZkG3VFRqwWcoN2&G4}ZUY%maExjqG>(?Ow))8Td~sSAy3261A6STzgWIqeq28lQfN
zEa-Gp(AqX>aXDW9K`F9lDjVQQeEP}iQwXvhBKK0euI;x?Z6~4UyFD{SW5(>)i+6zS
z46h`cOpkHhXaN6Ai^{vFB9!5CPajfBrR2zyLuGwE@UkY8F8@Z7wGCwGSwCbnN{!|l
zpQ?KJcbr)IG6zXR3o6xU!rGEtQW})1+A@2iojb5wymJA?_jZf1nhe!G+MJa^5~z;$
z{}}RiaQFQLu5tg*=r>rr=g9l{#p<T(4@jVG&ok%szV5Q4rvLZZZO5r}gArpW2=Gq;
zjA-`BIDtE$;(rXYwbR^CeH(JVAy~;~;W%&p!Iqc^y}!M0_F&wxj#+0%gzz8_o}&bF
zELyU|KB6XEZQ(P-^P8@*L(OdAvu%3h1VGNP)(0yv*tUB~?o6LkX7I7dm0KT<u{z$p
zd%Knq7ge_=K1J+!uOuNH>b||vTe(($F__1;>D>p1!M`QVAQVv9hODkTHlHtG6G${l
z#o9DGsbNdwccV$Cs<;3dzt<PM=@&-MBnga^+rC(C>(@@zkC~roz31V<^LvV}l?&k9
z7ANpqH3-m%^ju6epxO|F(<R|CPi`h>Er5EJvaO@!!QyHv+MLL_-JGnsH9b$H))!(U
zG{cjE1lG81easO{BXC9Bp#EyIR|kDO#Uy2@L1n|j!?(OhR7!Dm8AR(%QU$$~tiVGo
zU=Th%r|OfD6s{9!?V)MsboE;sIU6gEB6{2LXw1B$3ncROLkWZ^@Gu?=2YOtg)Rrcm
zqf6f+RVe?mo$^dm%!sBt3p!y{WRCSik79dsBVxdqP~M2<Y2NZFy2vV##$`Ao`{+!6
zqDFqAB*OUq7L76H2!Vq;Wg7+1K7*N%qhgK*7W6v)^<w~wFQ+f<X}O-e5&wGP-Rhfj
zXca#Cf@juSLc#^V4FHa^h<kVa_DRke=l{j+uayd>Ehtfc-P3G*7kemndGG#u!GF`i
z{`;^?UG!47&#Im=qINWxXHRDlMz}mngHN}c$Y1gK{SYv+Q8|J&DFrq>Z^;5F0IxiC
z*YhwF{1(HWSqFANE8zMVH>lmsfZyx{u%@$OR)Xmud7E(OHggqeo$fq-vQG#X5>6x4
z<D$W_Pg*~v;y0MyYx32Ar-xn`=Wi)2vuv0Eu)Y8)F0ZMmEAHttC#i1<oawJV@n=v!
zhn<Le4DM|%zwV|hccg^frn2z|wQ4sg?&!kdQ&M7P<!$WG4RUQXSygd7P{Uuv$yoY+
zpA*kN@$N<&wAjl~OR~iMf0VstK%7ywC5pQw!JUNQ?rtGS&|rnTyA}k85Fj`S?yg1Q
z?(Xicfx_MORc?3Bym#kKPtTuI|Ib%vpS9NBXRYpQ^mmS{rBLUDH$RZuB6@1T^l(){
z3#mHcEn)a-Gdh!$Fmfoi+bs%98h()8Hfxc5U7VH~;m}yswaj?>ticCejlzR!p$io%
z(vSw|vrgqD!`{Ivz-QALV<%1gj3+mWKBk|Y{)>}@T)f(&g008Z--I5g#hK8ALyJs2
z_^5oDE{Y9>imbKi%datScrJ84NEeeEi_$BQk{_+zUL7q05bqhZ7^3;|yZ4m2XQ<W$
zn7|kEL9DMpjR+2#qBbabpMrkqe=xcK=^^y;{YfR@a4!RY++ywEpC5ev|7FkFy#||N
zdoB0U?De4gU8zs>6RuI;vt~ztosWR1Utp`okI+jDN_qlO1SBj(V9_)FEYHWSDD(%e
z2J7?dO;)G*T|u^qb_ytW{{Hq*+Sq^G5(BS-lwm#)kqh}AyAP!?W6KXmAI%Gu;y5L$
zJjyq+sPrpoF<(Gh!H2l|xa>LYAagHf+cHm7gAIUKm~hbuRzacOmiQSrKnZW)R|7ZH
z?>`w3;0_mM_!RaHL-$Q^sz_v4ONZ;xSebZW(8)(V2C^LDw$=L*LnXD<SPScNrT1mk
zCH)Fn+FM^)*)--XB>-leHxTs?j;E+2t4_GszKP!_!*d+FkJzr#aYC~xPP^{%n;pj?
z&1XZm&F5NY^a$ki{evnbrEXskAoGF<);q1%GQ}<PV{69Q+!m>FYgRb42&1+id;4<A
zO5X8|4O{ri;D7mz2@HJ(<EOEdf8LuU9l;U_wU|5VW*#<S_lDZ?juWI41irnbI=^y%
zJhS@7vsIlkTd{=(J3<5M5nuBV$&~as0i2Qu6&<Tu7NXC5GcR%)-wy8IzxN_2X(>$2
zxm0fL_JKYn%7<DQ{&&89)VZbi7K-u7{uLf}G?+DbXJyC{NA)oxJCzChNP*LH^xK~a
zDXeH*nC&w~Hxnl%8CXW)mPvoPjOJ6+9Y-DPvFGX<EvJG_uAB6=$k9Tct@NUGqpm)Z
zNxx|rjBBzVn<T-ghEY9^nMQ<dz9@b7QJo<6^3B+1xedP+WMz-H3p!QU`?G~#XEnp{
zrK^b_N^bT)YNBZg1Qn3JbFZDJfeoKD@m^lcohzu5eDAN$t5r!4wNtYb3jPAk#^_kF
z6l!508vZa<!CQ_r%j{MkB)~eP9}y01cx3dkeZBcpm)m*Jv0tK1I>IQuQOT(p8y9q+
zO?w{vs=U(t03+o?8Yu4aiQQYD?&LuTbsQ}G1G_H0=b}j7cFDkBFBwm!+)~mGRB4$;
zY|pT7W#iJZV=1ZS!{~OR)IT*^IY38;_tjoqxc<kmEX#rFl>eSd{vD49{InZsH|M*T
zNXA>^y&OV}c=xpyE@IZw)oA|r`**(X<Xwt!8+XY9GdzNLJshIY%=UR@X!cMG7u2Zq
zzYHg<z5Yk>71nHF{0>c*mH1c6jW8M4EIz?OQIhf$<t=J7R6bD-_y|vk$T7ZzxKyeF
zKv+X4Swh%D$vwVUNm3fX6G!}%&-7MG=a{+3wpdJ24+{%G+EL(YFtW~#Gk)|9ju&y|
z`+Hv#Hp^`pjPsrEVsJ?$!QoiFgqgn@{Fa5!@NTu#wJNmi&`K-~#uz^e)n)F;=I1CP
zk1a)Rmb!(C#$O_EKTUzd8L}2FfghT4h~5`dLydubBXmi_jnIvqULlWX>S{bwVTXoJ
zvcx+E7D4>-X3y+mu}EB5sps6gKV);INwp;TX(u;pvIdRC&CTF*IEK#6KH>YepHq@Q
z^PtvylSeyc_L@v=(Lu+I5hQj)wzJin&nD}(?`VqODAP8Y$&xuU5`SHs_>PZSIBVV6
z_MLbHRU{Z$L(=3h8;+C}EbePZrI=nUxmRq=q%_b&a)@QTGWAWDR}|l8m)2yYaIoPr
z96G1j?~_!2Nmo~ZyJ4zDl|oIX%YReaqke2bFU_yKf6Tn4C+)%Y|6eHo%kBFAYqoat
zM6LiP-KD$5r5)sk{q0CwvY{-IC$wFXnRaL?eE6mo{c~l1%sMP)3f2NC^FYZ!*W492
zDMbhm8(}q&UWdkTs*fES7KH!}i|YUOI!Fi+Br<{eIGkOZuy;7{A)#0*E^-i2d~6B7
z3dX4r(3yQw6kH$vhGyyfjR|sH>w;CJY8dNZ=m|)sfvXnF3=608;_GFFgqo|>Z5;Lp
zMZ9uhJwt+xvcz^<4JgDHCK=*pG&fzH!he~=O2dK_(Oi;-9wb^}xO&}~s8|}6Ci0x`
zKoVQqxIZl<elC~P$Sl&^*!E8TMzR&wk6O$wL=bQ{x_c6X8sEKv{~Utmij_(bfJ)K7
zon4N-5v{{fdEY^l_J)te%4sv?nflvPAv|FVA8EOrY;WOtkS2L-NMU<K7dSI)T{8Rb
zxJBj(lBQY9J-C`lyA?}Dk0R`eWVOSiRwFQB>4Iw6Je_Szdt=EdJ{?yU`oG>;|BLbX
zjqL=N=M#omIW_E;EoRERkoI>+3(dVoNmy7)W~?;XN*PZ1UrSUo>T7CqdX&SS<Gm`G
zIbjTr9&ssppJB&?MI-40>W!0L>ExGOsmQTCFjQk-eyU5i_b3(mc3o^UI{BRi5*~5l
z!vD3s8MlNcSTgXKIedz*RauiT#Nor)`KpCQw8N%l5~TVm%}bCDYa8!dY${sBQo|jF
zk=n{k9baX8*4yw89OoSe!h52kXxpNp-4t+9sHO?1rgdF3>+D?|A?vbetN0IxT}=2c
z170YdQmiHHr!f<Q`U<eTc=K-DKNgCQ<Oyj9rrT;P1Z~IMxHNU?>?_{aHo%4w<u7=$
z5y<6ooHdX3pHD!;gS?F}9>%fyW9};(_AhaT`j}O{v5j`4aNr7%*-o}CF>vX~UKjOO
z9(G_?&hN=1QYj)k;fhT}uzF+HX^cG`iG3Gbh%0I397`(Tc4`$YD@nRjlgs5p6a$Iv
znT90eKK7m9-eR4X<wDI-J*=Rn-v5Os{=1PbK<yumn}1m8dWl#%^BsA7qasy~<WH7j
z{b1D1;&1BymTM72zwsUdt}~D0P1sZeBE1WX-Hfe&>!Px|88iTK)ag)2+T%pywaf4q
zlsN5;{l&Vi)NH<)I&=Eg05`Cui7BoIck|mNdrSSHUc?&TyTE5;!<@U+_FEIvSxCnL
z8}TFB3`$B`^Z=eb;?TFz_r(Z&SnE3zT8aKNkSJ#dnSwzU?WKu`yHd0f^uazb<2yE6
zafHAAIM|45A5`Ux>}qpHsg1&PSaf^Vq)?ltKe3tU;%21}yBEu;AzC2SRQeY4$HNU<
z2Yleh)wCV4IB5Lwe&_?0_+YKvk&QIKgYyHY_4#qe2l-`R3|8=JIMlPM2lLX{adiX*
z>ZOL!4f(MhOtiTNwOofUMBZ9TMuk2XB7Ac5{AiyP|I-()5ZfCsJhGX?v8du5o&tVo
zMvG3b^wxVP?hef@hB`Ijo_DByf#KMdeu@2SAJ>3!&@@oi$iFe6f7^2Z#vI#d97&;`
zV=0dj78}tD&V>h-HMIj-BV4^(rBNX$_G{PuFRYq{gI@;eK{2WUNr`+E(YAhY-8>PJ
zqM7d-*4!zP9<X~y+hWhG7v%1QfOu{i<kF$F;+O&IUunKB&|tzoF4B1W+@TA-`9=_I
z+lr5U+<@&%>m7L?g$2t7tBsgQ?BB2xqP~@HINcy!uZubR!D(j#`hXzK&NBW9MNu%Y
z9_|L+TQYlm=D1-emThFo4VS1F)3HR-p41o~Bvvr(FFz(nZ3H#w#zziG_2e?ft)Ztj
zZ5l^pAY$?2ycWsuEMkv0l-;0OnBL_<9sgjZvY*g2=*_=8_q}CE+nd4<O`_1Q=bH1o
z)9e+S?zJ*oStj{+l7f&hsg0%m%DZtQ<-1#`x3KlD{Z%Gjl<3_t_eEvo&P>6Y3B>EO
z6(>}Gt}5?|jC`K~x+5Nfw0>zhq>F|UE8@Rl;(zb;e`9#xoq?4y;)Z93nDT?TD{}I&
z&m&3krmhmAFr_`K9U!5OT*+<1m~d<mcDuC({kunRj-}Tds&w#081Ujp+xp<1_P-Jt
z<oQI$d}qsxqxh^S!DG#IU9psI-9~K`b@`qQPB^L>|7dVIQt}+#3DUQGwzxL#Si~nO
zrtBMoSeAAS(Cm#==!iHg5}WQN*9_3-{qa_+k+nbq9CrA9bS?P1EIQB8V0X9(J>j_(
zMp?$&pAg>#=9q-Y*uulf1#69N$>ZPOxVhBElKo1g%NeuirPL%8g)f$1<V`f0?22bF
z?K4168HeJgNN>3Kxsn++_Omz?TkzZ6uu#oe3ugmz5o<Ny{f^QBe1yJ@JHM;vZ-yQP
z%B%FZKSbbCe9R@cQuv`+KZ*@-laP7xuHEild&{_*67Tp6#;oDS(mrEF6@6;o2=r~x
z7`6QmHz4!*vBLl5t$o$&k790}k>8Up?3s38c@(|~jd?(XdS3G>`(T_~Y;DG#ga;wx
zZ_l3bI?g9}u>m%mr|M9@vHc-$A-{gNOFh=HL?!xxMii@%*yZbXDM3sulAzc2ALu)f
zq^EPpR}DlpyOo~xfZt8;eiQ_oEX@(g49QEGN#u9LR}RF$^DH3nFtd=P!NnX$@dU7?
z{^-mu>=RP1Jq~N>rarT-l*>pf?qAa5|8~=C8u7%3-;Li7Rrg{m>>bFb<cQVzNS$q@
zn=9Xy(sgfjiIa{k#e3(<m1uwc()D$-;Zk2uPP>+vlJ+FOBwaO^Gxja?ZJ5gtxzaA`
zIJY;kaNUuaDbg(OvQtAAIYo%YW(L$JSrkG8HRcOwFgMSF0rEYJph#vec0x@c-<6k%
z!n6k#{?ePxJF+kJXS`qSdq0ik0rrBWT~h88%$|mk`)=19=R_VUb3;Dqk$0K$?`ls+
znDuw_ae{wZ^BXB;<_;R=y1_Q#2Sm~YgyIK8;}_y7)S2VXS-ih-WfM_Mj_!wg9!8QB
z8eRpT6K--6=RWmq?vB=P_D&E~Zgp7vWd8X{QV!L&3?DUiV+(h{6=edYh;K)jhmjPn
zZ0-hNM$1j{>+IucjV!72*q`?wx!}~PXW?RhWLH-VaNd+u@y?pRJ*z>bM6=cUsEd`V
z9=<owJTu40LR*SYd%-|<$3cVRUiMS0T}zPNR{x{^-6l+zVurY~3Fx5H)}_gi;FSP_
z#OyS?C%M(<9phg{B9C+UW_G}9>9)mMxmzPWhcxG~7~$)&erM5w%#>#*sccyoV<c^U
z8Q_oljESVHz{40?mpWW>doq1L(fbBP%>31W5RKkV86zrU?NsU!^GZg7=gpkY7o%?6
zL|=#X0WSgKwnxq=9fF!s*A;7*5?=Q&|Hp#=-#7hqQs|I!pcaH@ma2Cx0}>7<Thyui
zw%-4ZahFVp^oh=X&5aQnG~165FucsW8eG4<)<sZkK)EH4pa>|;F51L6U;<XS$bd+~
zQ0$mgkAKiXr7cp_3<QD3evhoOIsmMoaw`1&J>mj+E&PtUB;02##8hjU5|WXnV>lVb
zn-=G>0wLO;p{>{C8JMGrq=R-nYOiW5EzZ0TOkw>MP`p;Tjj*Mpo?VcQ%8V7D{H0p`
z$*C6lFkf}n1adM9MA_71vYbjN(TL?ym%ee%&|roZ(8d(fzM<@9m4qHlY1ZN51EG&G
z&I2Wt7Idrioh;VB8kT!s7wc$n0rQ*T3mCAYDBQ@`#N^Pf-htM6gT3cRa@|IB<f=oc
zJG?rZVuQI-*+K0N1&pOmXc{?(oTEl}UOZV%hLWnoJM|mkssCM{v>N$;<SJbnyyd!=
zdbYpzwb~;-cuN2@*kW+Z%6N|aao(&s@kbAn60$NR3Wuf6)&X{=w&H?5D}(!!srrJN
z^V49-AGR_6bgT7JBvSf-%=-jOD}cxVtM<FCl^jtN^n$h3$9PgI>+VP1-M((naFX88
zBdoW2M`PcDdVBosO?iJ?FybS`Xf?F<Cqo1-rGf;wRjMN>tM9Hcm$F>tbjyccwPgNx
z$wt^4s%YvI>joRaPPLhnp8cm!2$G07r6mY<ipU!=Fx0x6<dvKxn4~9PUcD)vn>uCu
zi%aY9l>3L+ye_}ULlZdy3Hr8n%YU;o9wEG|8d4rs2Ou)+>w|L`-csA58eN)vICbid
zUv{pU=|%Cen!N|9ho2)2it=BKG;HZP^+UE2>i!@O$5xVsHT+IZYyL8zCfs|>la}BL
zRvhab;D1ke?exK3pE#sBQPOavFckW}2T94d8aYIXH@7VaUC3#yFiuOE>ev_+^;H+a
z3dfzaC?Xo5@AWZTW1!JfI<UF?|46e_U&FGBi5o8T6F9GPvkYYo#fbX^6s=+LZuW{s
zuUVhYpSdNM&y2-lg+KS{$4P0cV^fOvgO4_iPl+K@#f>g*aClyqq!nm}j>}j#1N%ma
z#?bhRO~P)nMfd`*+&e6I1b)^@tY#1%W*w3mm&S)?8sztqotnOXI9rHbXzVYWF-@NP
zO=c1*3EMBFR`9;DLPv|p<ZhWIzKT;i6T7U?@BLsqW;i9J9Tfdu7?x{9uqr3DvogdP
zg!mH!Te6W$UQc&*PEzzYl5)sh?HA9*!=mXy$uMnFJ35Do_g6BgYwI!JvpFT$4fQcj
zUKmi(KSjT9Z9*J%1NC&;_Y{Vp-bc5LGVu(xdfP0Qs^EX5pcbZ%s$OtwL16`lbsdUG
zpZz)|XVN5ys(-%OY^?QSvxBgk&I(Kz$3x*9<c(GWbJw930f~@b^TTC5zTQpI!_r@A
z`7h4@J2~(V3F>b@aWSyB`9nnxG(q7qsP4WTEyc5lyi+cqK#sB#lQDwID?dUQ9n<n&
zC$w%;^4d~XV4>SBp<+*wWV-Yw7#A2uOlgkcQiJnuKSU=&bBDbCIZMZ>J3IGXlCuql
zN8Dr0uY)fizy1{1Nw%zB{m!Y-3m0jm>1c_X5$1`7ebhiwZ`+2t%@ft_*gl!mvV$c=
za?X%IsvA|&B9o_KmQ_-T773255d6q0z$ovPDlsFINo%TN#mI0fnC|umf_@c@kW-b1
zz!IVwP^bNNmLAi$xhO9)>x(}CR44mTHJn;0Pzi(WTa6tEkv5)KF%+s5Ah)^M{ZPFY
zW=;#8#)AgXd3z?Y4MEhVvTh!6Q>#YU*6%}pJ^)D2N)D7Yrz7C|6)A;;NDF04V)UoO
z(ge^;nHY6D=c#v8?~PdZwYYU3AVZLN$_z7Q1E^I0e`Hy&)Z(gZ#kWV>qE^bo3W;20
zdT2Gcrc;rmjXUtsevFemEp!v6(eJzpmk@gp3Q%Skihm<&@8F|1rIAcX@@>3Pfir1d
zI{2PNa~k8$K#AvtIz|6xTS!oZwqsVuIz}r!bnl7p>6<b9?ukx>hVfdDZ~(U9Q^A*{
z4N=rv+ji`7$@XB$M*_pu!OS=D6rnR=YEb<anmNNpY#m3xJ*%a>Pis1rOZ?4h;rp!>
zZ3%29Tt7?53<PeLjhUGlC#$v-5g7qzXSGjPHlAF0jMG)-hl(@fnVFgZ)E<~15Wm;x
zsOnqGffr0F_6`-Og&gdbfDIQ<x+wU->2YqLO=zJGSg2;cAM&JxJO#pqDB=h#>*)r}
zhw*sfR86bCrN9Yz=o6TQw~Y<PuMdQU<j88-1g;W#GW+&MFx)Y!yFSnCeE(12%WnGz
zONKMbkIFmGU8u6CgjfNf%dADD5LL=Q4hwH)cv^~}?Or-W6*SD=^t;^q;*vm>C)pHK
zP3v(PSa@;7^Y<0hQ4#ma#*8=K?+Q?CP9XS+h`>4PG%LEU^RU>T3_m*$W=_xapB@`5
z`r!#r1o2ItXf~ih92><OwC7S+&2UMXhH`WGc^VYbf%z&heV$nFtZc>(O_T_|<OCTR
zp2<D-i)(PAw`Ajlfmmt$sy_yb5OCu)x(Ynk5i$^}l6VjS#@MO3WbvbJX-vhPTQRW>
zx0j!<GRY!PhjGrv64Y#k0TCsD%EUO#1;mYB5#7jyd()P3i_j{9%K$&yUoJ0igFQMn
z%MWyig2Vk#VuT#3XQ?=!5puQ{*ucs#o^$l;=9TcTY}xqaaIhIWUu1tJNqlcGmQ0k?
zWBK;)l(m02$-me@5rqTPz@^bglH}+eb@S(L8n7{mh7=w*Pi&!}^*Dk;Xd~AxX8a7{
ziKTCuS~r&BjV`ndlc0sb5M+nPbga=Krg%rw8O2=^9K&H9fj&ef)MkhqJ23&spU}#G
zv<B+%tOl~D_8EQdaD;jY2cMJlZQZAjZ^jCTLv6EvY&CgvY1>R3S&PXvBi2=~{Qhj*
ztFMYRtKRL}_yIJDGpBW40~_yj(?(nN8FONUm98&p!&FTTFyz$`QsB=QGFl})jhTv>
z;C3`t(j|C^!)AX^ewF_CT7664=&3&cBKq?1++KcqE9>Ck<>hq>-XJ$_5R_bk!#kKk
z=hj|Qp@jS5=q@|?4etHPiTI}<Z=gRM4+A6KFNdv}T1GA>v1~Zm7%p?BJ89KE<;)g7
z8MU_}xwBOYthE$;syK<ZQ)RZD7Yd7*WCZ>NJLr6jf95Ob67a-u%<0)gTYgYa^$~$g
zlCKjvTvA(Y!NH0d`PVvHyJ_I(><$=DU?c<F1d}E=`~=sz=9QFOrR_66VhVDwN>-!a
z58Z#Qi{GF-jG6QsJ|G|kZZueFV0RmZON*wcM7B8xg@DOQOEvHQ#z0fRCX5rGV>&DH
zwKAn@*fEnS1wR5W-x1bd%cqFoc2|S1CSSmN>;rrjR>(Htvd^XqIYQMHv-g)?7Uq>I
zvF-${`E_$45KpYbJgd~RiovOS?bwM&nc}`%)0o#W+_UK&u*qeJ?0~(wx{sM8f04w(
z3OH!3ZaN&e6XVJ+&a;wQTA2XnyV^9A(e^;MHX~Fq=>ILqvyC<S*37hHP;9a)pC<!3
zYVHA;s!*o6^7YEjo;@p>`!#ZT_-bforzv}xtG2}=PP>Mp*d`+_M_P(yDs{Ndvb&5!
zotSM0hhbANhKX?$&$q&|C@YKu5@8i)c#Q0-%9`YplHCfE!41umo6xOL>+X}EO0d&W
z2!1k*t#s>saPSHBY4RUstJ{;nRHZ=<TRZhbeDA0bup5w#Yk5NPKJi~G54miBIeft%
zF~h6^vFc5ns(U^-1u@gk?`9e=bcOMt6dZFcG{PG$V{ESc<g*~l*Eh~YC+HIxLCE3=
z=*+_suz=cY9vIZ!ky+=t1xLw_je(r1l21AeJT{9_`6-P?O;SumB^HIExuOwmUo@`T
zq%7P-4GMvRa*0=Mn(O3G%&TlSNVR^4ir3%DLkhsDCC59d@!W6VVz6P>l7yIDh6jcm
zJRCKZ=|TcXu&B`1OELVbzdkj+;l{S*5Myy<!gkM%R7L@?=#(kLBU^1BTn|}P(L{gW
zvD1*0LZ3gL0y(hED;Z?u?7Cx4D<4Z8dX3A^a8eOit_7#bGJwgE6gd~sK&tt@@E<k3
zV1xqIqn&5WSNu^t?l~6Ocr?*;%t&GQ=JZd{MTJbOJX(1j+N7chjl%Y&Z00#)nGL5O
z5`a10_#Qf*b9Gm_w>m{mzoV5-W5ws_Cxj^D;6|UR{`~&e+9yFjVcQW$ih?nNYsusN
ziHAZuYtW|SOdd9nC}iimraeb~N%4Zw5YXdmsd5V=(-E>kD+O~S#_c{^7Q-c{N65G<
z9mxv^w-!9I2Rsha0LQMuusRJ!g2)AbIEO^qA!EJe)Y71*`JC#*&H%+>j>b5uoN@H&
zjM`{ZdiHd48i>)$R~_89UeFf)XxUkY$WY9JJ*R9s<Sh4xm?^<1l-`R79e<<)Mjj+x
zL8EhDg>-k}hKW3Igu_<JRoQoJp0SRcSwkIiQ>s^~yN%pdV8@tPQd0Fv>k>4gK@`I@
zzfM~|X>|)7cd$b$9wdVYPD)VCIy^Yrt!b5|sh4=VFER`J$HR_ReCB#O&;4=F{rRg|
zRb?1r!8!!f5-GeoV638MDT7}o)Z!^&B(v~KuC(RjSmcmqDg<WbNzBBVD;we7@h2d|
z!X`Y%mn#1c;SRJaz(e<RMf4NPzgAlz=%ypd4+eCIDFm2z)0$p7A*B4VOp=HP#G!4=
zGD`H}8kedt1!1iJM6lG>S{%EJR)g0IeuU$KMZGV9^XL3Vl$2l9W07nPlMPt+Ru7VO
zS+2Ex{1B<nBjuNQU-iPV>88hfLD1wqp#k(Pp9ZSb#{yp%3HFc!{fd?i96%?RO?%hs
zv0Uz#x?pC#!r$^==q&C?!|<>>QV?CWiMk3+HnbP<+d8$-O))yujRt#u<BOHR@;!n?
z-y?&fIrkA+G&x~YV1}rN0oFewA+tOg!{8UeYIRSOeHI{Ll^B7Lh;qoPh>FFksJ7pp
zh&mOdeD=U@IeyS6xz-g%nO{!O=DBAIP3|+ebfJx&^Ht~P(ghb~&TG<&kw1-GO64aD
zyO;@sDw_U?Ms@8~Mq6@UhZpZia!#pKH+$7a*psvEPcgOmCKwZ}18Ym{d8;sH&i-`#
zQjG;-mmOcxxNf6k<o>miVCRcf!>T${vVB#}T0BEt^d=3U0pKS*s1tWA0<^$8%ZyjM
zU9;K^zRhlg51$9k2Q4)9hed?YU!y-HZnhmy{+S>y{c#kIAVUH(pd)1S-Btudws@@F
zViK^X$t(Gce?7^`Nijca@|t&g^;$M|;M+EK<Wa_Q<WkwjeWmlTy0CuuQjEP9f2%)v
zXhUmlH;5(|onJ^pyD+u>%#Uj#68oEzXqV}g0bL`88fLl?h6{vQ0sD<v3??MVCyheQ
z?M5o-!KAZ>@9im4%D{Nic)TrR@kHEW_?2O_xTndY=jf_tPUYp?!I-a$>!Jyct7~(%
zz~v-rQ{!nf@x=*oyt(GPVf56CDcT4yb4fQWAL0g)h@CO%kS(yPqn6r+57*Sv89T)E
zy`ZT%uxkl@fXN#NC(Fu&CUsH>K8313Vm2Xt*$!hTo?Gk9+KZO%65JtegwQ}%jFQPm
zip&bp!hf^nd19Bb;sd)>uqk=J43z})GF{<?cI6cK$836UN`u>)mBIRiC5k2x%adTV
zcp^h%R10ooq#ZY{keahlQMof8mIP$dAchs(u8)mx>Tbp}ZIUZvh9MZ1IZ28t9*FCy
z2s0ZFhO}(^xECn(wNCKMf~*Z%=_MQ<W<!kX0OgX}e4DGA1S?OYli`Q%y;R^mNaOGi
zIc>37eStu$4SM?gF^1mqE%B<r3F`q*ef8R^(6rb+0A4e}+z>}@p!))QSOzw|&y9j9
zpZ{{(kPtl_7%S`#Bf=D}*#^DOdHay;V^|+yLv8HwO?i8P_XcV@^WU#W#NVA4b8&2u
zcj%a8Rm#fE+gI{rWS5Ih$&D*$<v?heCQl90il&#Zl&5)~fs?u0Gu8v73i_=IW=L5j
zn3nhn<T~ukoG!mQ-9U23(b#P-O9z@oa@l&s#6b@pqLr^`^3A(;Pxx2}RbeTqwGM4l
zeV0O!M4YnJ&2?Ir$$oPmB{C`7U~YGiCWcFoIr(rKeoZ}IFrft4pmAYi<cqnyqBQ3H
zbFDsyo)r={YUEF7T%trjY(+S-w1RS?>vSiQD$AU{Tggz!sm9X}8J!Yk_0VjtxWT7r
zwN`p3p6s=MCRpc|X+PvselBFykYp+6kSiR)wNX#h^gi@53i#-l+mVn}nU9Vvmq5P_
z%LLniwh^7ePhf97{7F-elPy6c7WFGncfW+DeoF0%d?R*YHPWFW;B?GaI<`=Lrpz|<
z!n=L)kG2V#Pu~=LPtb?iFiMB}`9EKV{msNlkpRQBP7t)#TH{jvwEG(S<TF|y_knug
zm-TSq93y0J=A^J#GJ5x6cLM}`j01QA?Ri^k4=|Zmo76~Tc`ZYBbnLS}M9X(OYp~%T
zTa0I8r3q;-EGP0#yk*h_5SqeHl%ma5Y1i@4!HmKQ@?lVN3Q<+Sc~H<VDUG1I`+uw|
zL46cL_rQ;Fg#Bjh^ha&;Ehk`KNl#y+diim0sIdQlQAvJg@K_qO5B9kZwCLM!JV(D;
z3)Vv_VpO-X>!H<?2`>dBY>sD?4m1RfpaajmAV^XaOfQcY1<z=dZ4-?!7o@>8hMnH<
zDY**ltK<*%*`tQx;|4HKh>Oc60VMRj7grEmOL9WlfZdn3&R_nynr}l96OW-R76Otr
z^rMQrPmPAN{4J$9XNiG=vi?%--IXEdAMMn4!YKb`P5Ia4?r#ooed0mM?+5_}y+Tk%
zMhT8RMJYN3Iam;BWGaQ3TDuygYa=&Q3u@ue#8pm14|5MV79CzVsM0AOOO1tPN%C5O
zW2&Ugjxc;DhUWEHDvxB^<@?#cI3Ck5p&C-9n_QP*Hy<1px%9PBIhBFO%4bkOA%jPm
zst<Fr#BR8z*fh5;&TeWqjC)DHQV*~5!n~@c#RV`g3q>Uz>vbb;%&V0vFxxr@ulf|i
z+{f(C9$N|zeK8LENBR(u*}N)Hygd9WJ$rppA@(H;xj;1U<kDa=k2|*3x*sjxY^YQO
z(Gc(EK#RY+u{7YWOMe803)Uq;V*#6(DJ5v9Ts^JNTK>6e4!=2Z0ST!Fw?lhF`;DsR
zNf;30K4s$j&r#8tQ3Ok;w8}L8EYt**Eo~w?+FQ-7p(C^A{2JXucXdlW4}EOx^jt;?
zVsNfKNonpf_B8vkRLMOV)%B`H-`@G8i-O{Ew@6P~{vfYJ_j_KqbxkYxN5aSIXb8JW
zA3`rzv0DC>I0H@hfAlF4ESD(o#R<dl#XyGpj7NCGgfNWUXH}yQvx1;{%c~aPfv8um
zzLx{&Wx_opWrbZ=x`w1IS1<w^l;@!rSOA{+={|2c1C2<mGEA&XfT6@3nbH0dn9t?V
zXgmn}6Om17E$FE`fH%PcCbJgEbZ250SJg+fJ>(dTI055S7BjXqg_+WDgBhtSEJZaJ
zfg)r(>DVf?YxZn6FikAbUM|H#k*y4pviUP)<mt6@sNHJT7+?gm%5pq#JdqM+r(I<e
zxwdi^te?EDZr4{g$Lr;%QwS85?r(PvOKEjW7&~xDE<3pueg^8=wDR3bDlBq+S!Nto
zGZJI$7UzZG6-$K$chM@-74E#n>KoJ+E-5|ow}(GpY^h5$t2Ndsu1mCFil84?Pg`rA
zHdQBQs({)5?p<LI$86JST!HZ@&NZj6DFq(abbFhM0l$EtY25fjtIb+ot5nIXTAcwj
z0LtUAF)N-}vJ<?%xrij@cc?`g(*k#*^X`YRDga{Z0x>MH=3i@wiY1n#W{BLQ0U%$8
z;zN1(KZnDmCrCwLZ<x`PgQ>4&eX=4VJ`v7PNcUWngrS<{VF;y$kJ1Cw|5_tMRI(QR
z>G|KNWZe%^nqDfyj>5TimLw>f)u|*Ku68X2#e-+WG?U$SOJ+}M>6`w%1PB~wD+_E-
zI$1l-JCb?X_>oO6b(n*U1lGzd)`g`Y!fXJDD3>@y^r?x4b@81)-Y{J{zgcXcid%If
z-pzY$ehd`E;evQr7R@ynv$;ZR58k~V4#FEw;s$m$9@}07AHRum3<O^}gMF7`C}e44
zIrPaNJM|46+w;vBBX$g6Ie5g=n|kDE*3UV`d`QqwyDY5(*iFubaYIQ{Nz=%2L8|m>
zkqxpQ`0^pGaJM|8V7oazFJX0jW~a3r06GU>4Xu!;v>Fx69YK+%*)m2}JP~D$C((fE
zJxeB7ETK~+xz;mDFhOdGHBE-tp9X)%8oZi3IdIE!I9@(Bx8crfd5sm@KA2Tfu(EK%
zi~-zsOg41l;%~*~!u%pIQHmH1J2N+gjFn0AKC1FIdDYe`K@DJad?VXhZyvIy-J;(O
zDV^lpCOD{)mOdOo?k9|%yT84jKPL-5n3(H!?HPr4l}Vj^__XZwuyvd}yEO*IPA}OH
z>h-)e`FR{L3WSC>$+!YqRC$_XFKKY256bLB{|;mR6Y%81q$K<Vrg&?1g-SAI@?s71
zqIdj^9<QdT?(v@*-i90sj>}INA<<#C+hs-Ba?42966Z)bkl@}`=pgT(>4fYJ*JDl_
z)niVgf*(NYr$h%A4=UH0jj}idc2g}tYHf=RYmHr>LyBP|bU7*5FHGx*n%Zk|!%u9Z
zcsDd%cSJgibOv~|<>s)ZTA11bpA9(KzM0pn&X!I)31m+zTTkAbrY~IMlFdJV<lbNW
zq|j6|12v#5@__6%?ZsCEx9`>ucY=hk24+Hl!wadv)hvp=-x-j@k+Q;rBXY=IYGC<n
zChb~jzlBJ48~TIg!NH(*Eijvbp<a6pkA?!iFYdjG*+``;9(rDYJiMO~QAs>I>4u4O
zfuhkX-Fi%EamAfSYol@M2~p)&h$x%#F>-5RXm~vmOggrl1>+|j!$}O<nz=fU&Ud;o
z5(C{y(N9$4dhz^V-i~@_x#ZpQ)w2Cz9-FPMYhrfy^oRCpfN(+C6gp1g-k>u2)kX||
zlT8OxZ5_|UG~@L3a+$+E0onQlZ$@&S$uh?5Tf5->qB#lI5d(wS_h1amI^t}TS8K7$
zxy}b?=&7n1btIc?$}ICXjw2@$kzSUiXBGX(_5#-7-r79gp6;ic$F-b`4jW|qPjnzE
z_!OilCeCgbe}z&j7(zKwi`ceoqz;MXATOIip4wwh1L-<)5S7D8ITcI-5>K0L)@2L;
z9>X7NViSzKFX#UTK1IR)IKTRoxS(8wM_kl*Q(<+<D=LGqTCKGJJiTzA$~7uB@l>l*
z>piUqUIo0n@k>xMS2OYt6%wuR!#OmE$*A?m@ki#{L7ONonfSqNa-BLkf7p^y2=XX2
z!T3r%aWwNu0n?Ia-#1L#P529~n}DQ(V0vk+VEv4T@Yn?2bNLK+cNwRnXZF~hcQDJI
zZ&ce71oSZ3L2*!?`<g?r=Ar0Nnp}FF4`OEZ&u_)J-08~)(0K9qlZbiJfUMzDHc3(e
zZi?@A(uqXYmC-S*+J|(d@{>8>f8;YTju$V_#6OIMJM1mYQMn$k7w+v0%BiVj%-@7@
z*s-$KKFS_54O@9x0d(Bdt)Fkuz)QLX`6P0+jy|EgrK|ZwjP84*`O91$lYq#yX606k
zQoua@@YP7st!p{#ppF;phr)en*4BJbhOZt(ln+1)BrMz3#++fVznPw_$DCld3pu!R
zqk^7zWr6DqXo+2`6o9;(3p|&x3*0353S7-?3Y5;ukhv@yl0D4(3aqT8tcl1Z3X8~M
zW2J(Vhk1&Hyr+IpV3(ezkXGdI04~n=P2!*!`%t>qH0j?k{usDXB_t>~!>}~=gd^Os
zQL~ztb=eNQ^rs=ZGexDduTHR=)q94J&=HmA!lv+&z*upAhB9f{5R$L_UH-e1;Kzd>
zH{C$*kV%2ArzrLb58c$a0-I6n)>FQvinh=}`4htI>1FI$JnYB`R{f-+W~7lh^w>mX
z(DH9?8_ciq!Ou}`GYRlPRa$Yf$SH_yiIosE_>@LlU;_)iPMJQV=V1w~2GN8q^h{`S
zAAV~|Y4grWX<1O~7j<t*f#GV-C@4rz1!*Ok3sg2YwomCQ2xOKHd+KGUO05+^EP#{C
zBn8%TJJ#{0^^CFQbec5qKTTir{eB~S9eZbz&}|-%o}d*=(oyw}=e(+B-rrHZU`Rht
zeW{5NgouP&SjIf-l?s|Ec@^S5WK-*}AupH~#F(-ZUZ2{!&iC}Xc2m%+)jKU+FoN+5
z=^I_n6X%>O8t%#WH16>_m~g0jxf)`6zS^;B+AW?3Za$j;tsmc7d-Q*;Z7@knSrwV_
z-}uUHCHR@Prk1C{Cn>7cq#>%!ry+_%r0N)Y?E5-QY+Ra6YEsq(wIIU$5*n>%Wxh+-
zdaDmZ^0oV%8tn&MstU^(lgdb3?Y3f<?b^|kZ|k+1*Dg}V8bRd`@&+Jv9l7qo7H=e^
zVTR%Cdnd}Yv?ZCr{=#VGGWv;93cOgO|A2y_Wf4uW!vrBto~6Jl1U>tcYOAWM8*h}z
zleiyE(ZL6RAI9dWqZKBMz)1TsfNo#|>|m}oW(`DH;Hd7Gz!LyIwH|Bgv}t_CB{tz;
zhd%sjVxZA__LS_oX?eNn!2t(22cHZ4)NCYBb^VphW~1<4j|zCwz3qCh8he!vS)aI0
zUz<}v$~qxrQ;Lq#En;yfov7eH<o4Nd1FtA&r^h2PDPU;|a30auIs+`^0f4*T^2uw<
zZDQ@@mtN`5Lmt+1waF>&f}LU(oZQC?D*ZrSvBlc_fDwG_x#Cbo-VTphf5z*O!S=~*
z&J6W+NqBDHYN~dFPjQ{PJ*S67>5HZkreTUMObnBOnUa;!X&x-&O2jXNJ3i54^V?0=
zMO~@IAr-so^9ExLDOFyzy)|*pzcu~<qF1m#pI)%v71Do*0`F_kVpkN^25x6*a!>M|
zKUaYFn!e5*tG?P!#E@^}QlD=<YifJE9YkWFJ-F!a+lRYRTH}>sKIQPJqbSbB<RV`S
z+bNR(>Nam$dz=!cKMs7>J0Brwx(O<NC?34N6)rfen>Rnm5O=x0D++pe{9$%*GiQCx
zQ^DuBD}b3<{bd{9e)dqV-f9<$CWZZCPsr+I`n4LPD&0hKt_^)UFARx}H^ISLI#Kgk
zS|l}YI{6$%_f_;*>H~oks2G+Eh$$j}rL1aSHThtuLd*8QQ)uZYo=o3V-52?rZ+gWB
z;Yf7fVuQ3kXqR};vM{eVDLh+%=OTqI_97np35$qIrcb!pri~&}>L|)MPssD+C{m5H
zEPkh~u}k&WlV!I@8kFtm4Ar#Od`9Q7pN`8X;c}dl73OscI3ecH;k`~N&IBeGJGApD
zXb3$AkNKH?A8T-rOK~SO^0$hJvg4FxUW%UcxQ(8!&+t)gKfI4RAr&z2%WP09SI_|~
zpMvM(s8hLpuJM@{DkKhhG%OFtob6av^AVR5_IE?=fXb)C@DsLk&ziM+D>EJDa%Ah%
z{^n4bruxz9!Mq)G4Pp12nuf}3XYtRqrjKaHBXQcRi#mgCa`55&v7t^@JcV|6M0QM|
znWc0*?Mm>caNfkjR?PgAYTrV0@M2jqWBvZ9nGUzpIew&Gjb!dFc>c^?!)hXe`hXu;
zoZ;n?zZ9)gR((ohn+48{;+b%}GI&4&oo3Tivl7P${=tWLsmGXNug5}pq`<ZkmX%)N
zk)>T>muEg9EmJ)qGTX2(9pg0gJ<r_+-s5avUl-Eu>x?T*V(WX#iS7$2%-<$ui?;B{
z)nld=dP%X&sp4o9;TSsi5l5y~100xi3mc)L_<e!il4VEh2a_x|3&SP-y9yxEc8gOZ
zKu1roy;hdCZl9vGtZ<D}9t!(AmdqQj11kkyD|xxCZv^a?0QO=y8+w`-r2aIbYW`y+
z^!ImH>_gm$Dc8Bi9P<CBn<5jiOB)z7Dvij+2{T&3gv-Jk(>QyvR`L#>lC`Z>U!#>v
z^>x-Q39GrboyFHM?#0q@SGF_;8*^4kpSZXks%MGbCg%=krscR$ke8R?Td01C58xIf
z33=4?D`IAq*Vh>;;qp>QMjBr#q77!90k0MwHY+L3eL*rztw_79%NDqpgI1zZ8L~A`
zO&BAC2h~jIR9T}<UfactPE%z+sW7{9C=}tz#)8WgbUCG4@gpZF`fBH>TgxUw$)Z(b
z6+?<WASSa_+o9GsTj$nu`<gyB9qOL@t1Q}$F11zaTJex!usl))f7*Z;bz1qzayV<@
zC=wgN(Nb|ekKNpW*&(;(i8g$*&5XwdgD{VI<q4sEhern&v|6h9+@7ksE^fyHZzLS{
zek}JNp0!8K81Z7y|Ekv%wkOMxSQuV%s|XnA6nLr=7kFZQ#<0{D#*`s<mv#WrZa7C^
zHLv<Z!1&`^H8p1tAudG?k;{U%b*BQ~*CnA^lB%ytkvgR(4lJ{0i;UcM_nyqlReCzZ
zL+XPflh(1-m8Y1{wCKq?Cc$09d`DTk%^Gc+kCztw6+!tv@>tybsjO9kwcYwp61!#d
zjklQ7l1i=F6m4u;_3d0@^THz9i9$jE4Pn7A)dXdLg3_L;p_H2;BeArC;VEc)6W6lq
z|Ax%KrKmHxV+>1dwhUEijq2Cy-C`yemWvAN(XMfe_pi(uwC{0ArWY{HrOhsLS#6qc
zW2UrO2PW7v&ZgDsaGRYAVCq!p%hjJ?WPTY`ua5pDVbszBLnbDpuLW8jyGWZm9kCpw
z32!$Kfh_>1G&trWjc67p(y&ZDckdn)%$x6fv9=bYQxJj7kAtCDG0@d^Ia}a-tWw~)
zGrQ@onB_sLu=zUI*Kg`J<mIyg#RI0^#SorA^<lF0()~9w2XEbZbAjx`FV<_7L5hdY
zQEjsjsa0<C-x;YDGGw$Y^OrdIP~vdu`kLioy=+nTvS88VyL@T)Tym+VH-FCT(|_^0
zd+D%8g7mT6!#t<V)3~Lq=0f*sgD?HL10mh*6(QaC)-rv{=m)enT0m2eJvnI?vRA<q
zh{1<7^A-s|wT7f%`h++?oFa{|W6vX69>wcpbi?eELY+ZNw?D0%3Z@~2?<3oD<0&aH
zvq=SKUB|?EHLrNM&Z=)Hod)ubdF5-#PU>)txlJ<9G2MLBlx)0N6^$zV5RBxFbV|f?
z+!~C~Vj#)FQtpj3qDu{G?o@{njb`;d7uq?fYrj#;th61xrd9kFoyTe`2T06)tp_|5
z*4!>&oxDRN;da|c0<&&DCL?jcrp)Hg)p`5PUQnIFPl%E*+~2tpK|8VkPRh7{uU0zE
zPPx%~&)=xSu@d4$+J74R=jBAY;UAsDM>H-cq!qlRwr!(*3B)W7G8jX<_7X%2CY)96
z-@Jy^3fR%QdEw;b<uu3rgfYbBhk|Wo^99V4RSL_@it!wcB{2BtatdE)3w70{1dn@v
z897kF6!ew(%MIrmax-eq-06mzS>Ep2=jQ<Uo6*;oYbu}<AGOXlp=(;L>L$AmGoWce
zkb~eFyJM5@B(S0EPRBj{X<}O7WmC{~#AN=kp<OSZhBZB#CLQyun^jrCxIxaY-(ypX
zuvb3pXe`iszQB6w`ZG80gvZ6@{K0n!`CPq$r{4PgqnZ!I!C`M1O5*3qxi<#rLGID>
zsJDb)ZAAzCHw?+<JS(DXIo0~t*>#yQWHNkjQPREEQS|OQ==2`W5cF;qNFE+kRj;2d
z3ik@fl_psYQgRGPDBCL`YO6C|O|X4LP2g9-1GaBbP!4J^9;Kf{E5$lz@X!Lnfhu}N
zNL<*hwV7=T1?{k-;sO^u@Kbh32fe%i%lkSG97VI3c+~MfQrp4iPwOqq(G`et`KWCP
z;OFCaEgRYvE4{ZOHh1<Vwb`}k$34+lSCJJIvt9ApDvP#302aNx_kqEd-A$35&f41X
zs*Tj2my-{mI+@k-QWcl+f5LED-;hO&Wt!69@$D;qyS>CB4vCz0c@-A!-{&meO#B$V
z?fc^&4bSRRv1*uiaMSJ;hDNO%VgbGMndMx!T~@_RKvN>KIhv=@=Bkt`(?ol(W4&~b
zc@#2_$Db9fXS~7%R#Rc%CGEOl7&M>3scz?laqZuv#HMDjw%WcP7hHL;mh{k|m0d6f
z|1b#O7S7vd5mtf4pR5aa3dPo}P<JO8&5m)spjQwdwB99dE%-w2175xbN<pe;T5ROF
z{dVAL_dHS*kKLli_U^dJ9{;ccSA6V%TRz9GE5ncytHN#QW%J8TSgl|9$^xXJM7HEv
z6lE<auW&+`lW>d#%7-0HaF3Y$a@=2XzZy7ub&_puo^=uTD0R>h(ti*}OM9^Fmsy(G
zHu8IQ-+KC1Z&lRxR|aXdJgbs^ap!b`)yh%4r%k8!s9vKMBX6Rm9BM*ei?avC;|MCh
z-HKL0KsDg7)Jtm5k+J`6J+}I#TPrXvemkiyaHdLu0_f$XzCbkyQ4=|s)@gO<eICY}
zelLMxiLWfL`anr-HZa@o0wt(Y|0Jj)8LC;04nK6gg$l)7*8g^e{;MB!`^nXt=1rMA
z(+a`ikA{BV6`YKK9&Ri-ZAa3&1(%VN#E<UoqFw(O!Z;_;dJdp@j_lTi@$mohfSo|&
zUvMqW6ast2cUL?Y#@sw(vxN4%x5L}*jvLMU#%->3`)cD%GNsA8uuk1vUD=G1TeE`$
z2@T|y*){~7ib08JMD!=}FKGQd!u+BptFY#6o#-B{6c6KkFH#WsR?!_Z6OH}Wag?N^
zlxg;96n)0w+;Elu`n?|S0g~R@D?|F@frmiz^1eX6s*B#tHr%sYPo$^mgQ}<L0q9}7
zqfKBZJJI#?@|Zx&>N@Zb=epW-;<&OLe13o19Uht6E4)C9f*-e)uzaGZh@|Ktk3v1B
zr!;m{Qh9r%R!LMWplJTejF`tFoyHmN@V8w`opFdccrDxG(9OCtSWmmkDlx%fv3Pb_
zw#LmXbncjLwn-0f=_{akbcg7;Xr!1cRfA@{lMj*Y_&BK4U457N*skfUGgRO<AoSrS
ze0s)c!P}>Pn+d))*7bZfkjt@;0lcnmbs<qkam?q@UU~|VGmsB3U!|v)DEs^Aw0Qs7
zJh(WA3X?umf3~)v(@Bp~f7Rx{I&&jI&NgSj#AL&Ru<i(U{X+C0iel4ab^mIt=xs%a
z!tds0igTzM4Q+_ug_o{*aY3h?#C~2+b&#%mF)=0Aw*3(sgTWZNWmwm%)?<~|F3oiF
zR%M1>WU;42WQtxs&R52xY%%I8u!dl?kQM|=6wO-SK4-Z@t$_O)<bAw^{)_FO^qlJc
z3SDq?U{T~OD_;P1S5Tk{hUG<7r*Q1JrHy&`u6;`GFFC2A?b5_qqXWH%#fC%RU<V^m
z<(AVGa9e#nE0TDqLFHG96}!BbPc&}Tv-Qi<qFbo7+BZ^)NfX>GtvIidNn9x5aG2js
z-X<w6{EA8TIE5#0b4~*M90GA&3(TEqy0vW;@<<LI(aRkip0}DSIo7Ty_&m&$K;+0Y
zc{3H!dRHn>1PU~AWo(-rZv3kBhp@C3fGt)p{<!LlQpNVZWt=C0nMK!%!=SNIja6RL
z-q|4Aw|egISY2?|eNM_tzl@-tW#Wl@KsdK_@Q2wL1+~X|*kdAN0hIES{HJG5^vBlN
zP1}~`P@V!8MqHjCX}=NNhtiNsFTcL^?Vj^h)*GeYh0xk5_g}*z+pJZdI$VfJTrn!6
z0YXf%qevj)Ic^xLgdSN5MIB7Bw)*VP*irgf<)WPQq*X2#tJ|)t!_!w!d1gxh^YiXa
z4pJ-=|JkhtJELNzTNovHouE@`?RKpFR0`5AZQTT4=={fv=n4~yd{RU&b8*ScZWx(%
zy-O>;ZASUIP5<FwedY{n$crrGhEHkKK={k3SN+VskK#ODYkQ6X_H#aYl5@1a&Vqzo
z9rITjowo@{)$Lpsx>g=z>valt52KU6c(G@*-P?*-MPH=J9%jaYM|w*8Nj$<=+XS?S
z>9c}0xj8}u#U*`17EawBJTOMFN_V~muqFohqz!;wHh*^HRKWxttoF)|K9m77^J9NR
zUQMT8*qg8$Fo*J%>!pn$&%Y8B)kSQs2Z-O);(vzDL#8t`<~_ev(N&@yf)3*}sj39)
z`ap|bwf#T&^`C#9oZi3n@iy}B?8x7#kxL1>6I%=f8r(9+BZ3eoQ5veQe&2@s*20|^
zc7jVK6;kpKPy-uitChs>*DQgh_9+@A7EC;Mj20|mRfXm{8trd@m2`pjHi6ybXnAYy
zG(|>Hu02IrtRTiTg;z^%cJ-X)<(OWAXyH$MF=Lot{$;CT(}WX4?^a8eT2r+A%{tSm
z>+oi&5o1U)m@#jRjGuFrNPc#IZ>+OT?<*q*NkchPJI5Ojk5fkA3y5H^hTLK=uZ(v8
zDFd?KALy6oRk#A~3`e#e+vQ)<tWy?bs0T3_6v%>1HYeW4jYV)Y$}(%VE4W{!gH0TW
z3UclD%X2Yw$vvdU9fuCH!NEZeiep;!s?7|TN6Bdnk+3NM%$+KvrxLU|!&eNMYviQL
z;ct1E*2@9sR@Z1DX9%$GEeP*Bcql0S(8mk)Ezy~>GH621)exbbdt%51zq)n`Q))h$
zYPflKZXQPC!#71}E&FGeY~Y{!9v%&CY@d+)o%;Jb@3*V}a}d5Dm%Xxs{cM~3b%EI}
z`nf^Ci|^?RCl*2f<zdFabin)nc@y~r&~#vl5nwG;dMRRIrj@IB(0~-0hh{AvWLK{a
z<kO!w__(cjRWwA{zmOq~O(xy){O}2Z$!H0XNf>Mbvgk6mo{9T1`6Xl+G7>zpT=wy!
zZ_CMQ-4JyVcXj0?Ry*ooP>w-`u5JLvyqt@{gq}m~yhERExy^$?u8UV1C?Q>YuWBuE
zA426@hE?v4GHfBuL|mX{vp)HdeyEnhj4c+~{aq$HzdkNjBFfqs5PZ=xX9t=H?8YyV
ze4G@3lsPqRy}WO_$bK(ynO*O?m7WXqnYYqz<j%d29U83C@ajAtRr-G@d&{sW->q+W
z1{9=K6eI^l$)S-R21VhIlz?=1cgGBalt@cAqI3`4NW;+G-ObSbj{DyGJ??!R&xhyv
zJPh;UI<IrB^{bVy)Zdf%b5{j7zZ*B+19+u%=0k)?`c>Nf649e!0yg|pv13ELh!jF-
zMp?a-y@n#xS#hc5jviViYL4IR#xP3fRvJSozwc~(_<yV=2HZYdW`y_x`(A{EHwi{R
z_V;n@*C<@i&soiX1k|~Eq~c6Ko-C#O=W+Z5?Gg-v0j&i-3_Eq>p!Xmiy7D>1m?sr#
z^gQM5)g{VybF@x!v)ZNAdCZXX^swYIvD6~z6OY;>pe+t1+il6SadaWqp%T#{A^b#j
zCl1)45#XXP|7yHSrbG)1SwLO$b<DxDow#tmQNxk1A^Qod<j$14_k=CuorLO1olLfi
z@YKaK?ThJ;wH}N7Cs9A4D%TS;`5QN?yp|(%(>jf^HCADSSW2{1KKf9jdBjVU;uL?+
zkBP*P%Q*FWj$(xaiXCV1^Mu;Hm??_|>w;-#nxC3i4cSy4^@WTQ0~7sEv2$ZWqyX;f
z5#+ef?7I{?dIo@!;9#Qc#g?v?i(6mVi~J6Tnda)3-={g3qA*0_#$z0uk{={iYgI1H
zUN#*aBI&dHmS5ji(CvWtOt)ast1#Am-II0<y_E8Udv*KojpY1q@a83G03#gaf6xA(
zjtx_`|B8-1e;Sz@a_fJUs%5N6S0mep&@3+#o0fT*WB}M#O8WQr;hq@+#2+sHXBfMC
zmCrxOdv3QVmo?j*Dw_z1+s0>`)%a;TKbsy!8*c@0{EdMIvsh_Czfv)v?+@8b9DJaa
zabB;fvQ2tX$?V7^;PDYb8XdFRVhi#!_UDX$w7^<RQLp`~kO{(Kj+*TrvU8m31N|Nn
zO_Zn7+VcmBDR<ksLN$`o6`Nj2xgD>@qODul9N(79sTMjhHnwo|s@G(%D=u$XOjQ}+
zs4~!uVBbaHnCx+w>V^2aVMWXk2@xf)<tZDadq|s0%v4%WjGFVjg)bXf&(z8#NxND|
zy4U2Pq+1milH1obBa-{6L~j+1)61rfI$l(v4@?kXLoQAz-Ai0C{Uyq6n~X`Ke<FRf
z7fd_22nXBp%r8r|_)HNKMnolrJqb6VFeuaATlMLKU&>i44#G7Ce+z}e|7D?Y<oJMV
z_==Xxz>gMoqHw#$;XOpeKbwwO8>HiXyQ6!GiGSZJtePLrl1*x%JCf^$^ct=1YvG+^
zwkQOI!A4~Hw<enID(+6Qp}Ce5MTtD7{r8_*56(aAALxIJ`j@7>L)Yk3N63;I@QO?p
z{Qb=v56dHqIfz+ok(UFK*qlcrdjTrfv@a^&U{b^#QKMuWGX<E$1mD^_fIS~RqKIhu
zn($CYeSyV~>oHaHb7{viGIqoLwvEmxZnl}44~8mX6UPTmL)K9<+Z(^&MYcrs5*w~i
zO;Z=qkK|5JRVXc5+&O+&M|<6AY;uj1*Rd5`I9Z{QdHKnh5Nx2(j_&*;2#)MS-i&WQ
zJNri?6<~lSAd`x{3_wy2^3rwE4i?Or_7(Cf<t3CYRFZcI7!w&OR*5W}^U`U?v~CAK
zB>mj!)|&Ci?d`*HBh#Qgle1JT+SZ>C2HfUl8+I_F;-1-%je)PyyDq5GzzS!_c)r>G
z;*RNj-DHcLSc3)OieNxQadHpZMp`Icnsr^P%r<FpVW!MErvLj+*_Tbb_6tY-;a8jb
zBMI1s@WRmB39VoGMYW$)bzv2a_qPW-l`}R3QL;hICl8I0%`ze~5?cB~yi3#N=D%#=
z$4OV)s!e&U#r2AD+Aaz3lV6~%ZpO6H=?cqM1ikL?MTy-09c_RBMp=E~XLa*Y#Wo%9
z0%%5|r{2);<<~~x_};C_W(Kb0_C=+ya|dk2NvYU%%ywMJnq`0m`|yv)qsh`c(q^@y
zmLi<hNglJsu<}FPNA^T?g9sC2=bB1YtMTu9sFVIH3d$w}ztpr_{#)bWzFLpwDrbjt
zo>Alub8XJV!z-62+s%pAqKT9!p`1EkZ`DLYW~lKFfr=F1MRRo};lmwWpqWYEEi4`H
zMcLt{19mV%6Ok1Y<A9U?&6(2{^GC@<&CVUW<{q%`z&-~h8FjSf+WH|~<5+j2;`S48
zEQ5{_t{XARzK~jq;a0E0S&w3J_A(9W@6fc+WS)T~JXp@D!VZpkH;<7S#f=7?uz}sb
zOcA`Gqi%_6mi)*5GUfm6v0c6_x0$QueA!C)&7XkW;_l{hdn7v?^DGuN9n-G004_4K
zGNKRt(YHWxM&sOF%<?X06kC-a@hwXCPO-S*OrylcWqVW#vvezrz5eh$5hX8^%uSo=
zGRyc>lnoatN%OV7SlNq-nD6cWhR_}z)${}}8^UK32^vbL=M-Ybh)x7%6ZkQKm}3%9
z7F)`{`6gXVbD~2w%QXE>TK;Iil@5k~8RMvx8Cy!Oo_6u;hq)zB*Yhctja#X9Tz%5_
zeo0o-V*07%g$iZ!(qQ^IbWdNuq|#nbPI+7`+lL2R4AJ~!m%VRZL_THXZ1g~Kq2Sn4
zE00WDW7DswGw0)gp_9eLdM&yF<+QR%HRh}@-%w}R*weBE33;-^=t2}oJodtlmzc3i
z;|f^fK6e3nW=1Vm4lyu$kgTW*F#q#PjfsQqg{7XezmJ(?+FQ#?26;E_wdV+V^vOl+
z?=@8=rL08i)3OT&D4@`vw;yM{NUzMP?v0_a$rT@C|5a|;Sx9dL&Bs8tHXGv8u7}cj
z?IkzhzHM$+B^?US${~w=|5>##>11Ci`vvAR!spnDp5I9w4SyP?1Vb2o*8fE4mtGLX
z+(PiLc@7kC6PdEw=OdZD1%54ZXjXzZdF#`*l1jZ5&zQX<A_=%ZkUTI$i`%iZUU4}b
z9fjy<>*(AybsYP|>L-T{#9>%4GfXuEy(a`TGvJSp7iwZh%v4ya+KC?t-CeA-KAprP
zW$^P^Ntp`rTt3+zV^w1ZMq}2w5gb)n&yWGkFl(T`#*hMpJkNGX2p{dk03IHm1})<c
z%@{^DG*mxzxVk{&(OF`S1J7d36J=Vm*>|Vx<eJrh+3{9DujiK`>+HPgK?b$Xad~A7
zpf<B!SF%!WtS<n1eL+UO!C|&CH-bu=iIHCmBhDN$BU|S(L#2fhJHp=H((<9|Ip^ax
zBmMYteieMLX+3FxN*G~!hl;;EEFPrtp(n_hgf6yF@nhwe9afd?(3q7m<8=ZH{+|_a
z!R_p;tMp=w7}7M~cq6tpedDs$<*@2>W5QN3yk?Hq{PIKJG^cd_ClyQ8qESE$kZpUD
z8bQvn_4%^-CseHg-|BvZ-;Tfv`L)n$$ukbH8WDRv(@;}(!)wZMuU<AkwkpMx1;%CA
zd+$|bBgh8Wet%;ZFTYF@hedwJG=^LAc^_-0tTgVe9koz3y{qDhX%v^unbq6x4&XMy
z_%B^e&F!f_udrm(K$HaS&I==sFrW3tuE7;|Ei|2Ov_TVER4ie+=Q{6YNgRb;_w^J1
z$7{5%+kV2T#h?MdjT$u%+^-EXv!0^Wg2y4=+%mAqoz<=8)XI$m{U7${9bJJj8di8{
z<u9pZcikSmICA3@>TNb%j4rt8y-ED?E1lsJv22=Oso-71D;Kb2`JRUo?3?3&KK^+(
z`prUns1t<{xio!un{UFoUK;L7&0rd=5krzmVoVx0k2A`K-!+UlWR7sx90o7ichY1p
z-0r*nj2fO=>8A*6IPMod-Lv%;-aUHYRRf(Z)@=fC%ZnU+kWJ)eUfu9*8TmHiZ*TaS
zi_`0=kr-BhZNup}fOqP=d-0T4M0+1{Fj>gHyR>p~v>3olz`>2FT!|Z#P!o9)_VxQ9
zqisvEoZ+Gv95~Y5awEh(vm1fOqnrDq$DuC)s%t0_K4!856||f|{KIB-F>Vmk&=pd)
zt5CjS#aF!d=DCOEB?Ckd?$dbM=9V&!5rZ6XeuTo|0Xm7*GL9<NMAq#{#pOdmlHp~Z
zmw+*rx`Dv%>FIWWDMQ@QS&y|wqhUf;k-H{#Ea7OG<9*gRspeK7Hawl6PCI?1E;a91
zM_5&rYO-Gcgv0rO$~702EJs;k;3Qxjd$)1hdi}HP{O4q0(e7>g3gh+5fskO#iUUS9
zF}I(uug!bsaUpXZEzY_8E93-^#7NtDX$RVZXUOkLPDIUjPv?|U>`iStu&@S=!bH`$
zPs+Q(t)BsqeM$U*$3Q-t*%xr8Mf{cV$CD_cU-5D&LdyB7#q)au!T~}perva%CIv#V
z``V2X>Rs;zbhHKJI@4}HCF-bdH*jo4;lLt6^>JWzWL<J9yT03eTsahP*CA|f*>r!)
za8q>?c9XNu-8R=+Lr9fRecr*&IzLdbyVCIVT%1IS{IXsD8|tRkaic%o^Fjobyy;jl
zP{;{5|APy{9F})k&&=SDhPvI3Mx?f)2?O3*j{ChK-J7l$Dea`q_uQ|M%vUMIPni_X
zVkKy!Ck8}8Qf@r34xRkJme@AkR@yXPn%o{k@5?ZIjme@P;7BE>W^bJ{J*hXlh)G_n
zujd4~)_*H4;j<hE*GH5lK8-~lOABE~n8Ib>U%E71*-@vNMpG2jr_v78cYT&o_}tB(
z5WsV^XaG<9!p_00bXnxEH)(##!TdHMCLe3w#@Y}$Q|ee(X*HKMb<jID`Q%rH*PP<s
zWQEl8H+5gsHFA-|G}FqpQZv&0Ic>AjE_<WvG|xReml)>8Y{U9%Ez1)tC<~OP2}jP;
z;xp!zR4t|gtXVk9ZVK64_GYtEV#mXf_^k4opIp`JhH#8$z<n687E(6J#cCvR&+@Hd
zE8(fysJM6PjjFcjqD9K!Dev1oc$S8fpmes2kR-Wljn)K7?YnS!m<*{(=+Jkz;BwTm
zidBAz<xkia=?T#RsJ~Q`OAz*gG!Fic7y1akzBP27Et`}TM9LUI{ul-g134P*2KIwg
zTvvNUTHQ>-mZoM%5;_L1M^cGmwBepVUhek>sjWzaQa+ytiIOlHzZl5BZu$Mrqu#J9
z@XUZ+f;m;#)kbZt#Zla#C*<^z1ed@2#t`G&+=6fWGDwEOcJ<R;G-()W>HIXZE*sl(
zj*Ly?+3+ilqF{24`C`6Ijtr398&PwiPG@Ga^QbN<3VKaf-<m+YoxQ%8#(&go3FJ{L
zm|VhDHC+k#NVfa}TU|8>trZ8~)K89tus#H)_15(U-N~UZBQIlF7f<%<4&?rG!hmO+
zUN_6=VQG%7uLSIa>ZU1~N^*i9$+ev#_nYqT(tELpgPlJ&zZ8q|^((^kG`F`0-Xtkd
zoJ2yh*nztd=&jOZU68?YQwu<)CeScl&|$U94!vidopJ^g#mptA?Hq6gNf4~O*7bn)
z?BWRZCRPzF%ru9=1)1tRVF|n<qS68OhAEK!)j+C=yW8tPhabAs6E%~LhKt>#S!Uml
zUgsZQ;2Fy5{OYp~ITDB(6A0x?71(kPrQ%Ydzakmh45H^X88!?3v#=~n%dc)5nv}jp
zX4`YYMak)rkxG3S3%{xg652^U5<(UN`zjkwe|74P&=(U9HY*tN+SFH!v>j1G?L#_b
zD$kO5-;s8NBhhLzYJv<QX*pag;6g6!QFA9#2qzji$Ga<kwVw{!&S;F=8pV`F_3)^%
za+FkU`t`>voAZP1ELEFf@dnW64Iqc*Vg#J>t-<e(5X$^gLk6?nHmTU`HB%XHL{0WU
zb7>uvJZfw{UewqB+r{ii(^lZ`AYuIg5)kc~=y6D0r}SBkfla7S)Tx|kYge4HONeO)
z#Ci?=abBx|Mn8N)O@O(~V4%w%^aPifx>{4E>*-|;VdPF@KPr{M5a1s)GN=FtzGl+C
zd5nL!3K@T#fNWLccJH8E-tV@%0T6WV<0h^k58>G&`($xG*8)~B0Xg4`k_ll72A6!2
z&b76v;(gk{+uuB=?LysZ?})n+Jv&J5*ez`dOj~MZ4d$afd+w7W=~d&%C){Se;;C1-
z-cw~L4a9TNgHg^Y&CX$N{YG*fk1&!VU+KT~4r^8)zCmsPAHvKy-XI}{^hhN`@6&wU
zhaY2g|DhJ?Kty=rU;-wQPGlvCgeVT(n>?&RuQiN;o`Ve_jhkSvCXd--vY;Htu8lR>
zM%fe78gQZ2Lt&cyB=$s+9nQ5p@#=o<xdd=2PG)j$yS0tc8zY0xKG-Lm24ds*89%9R
zw@+CyJ-2-(vH{6I_DqDhR?ep!hiRKy?ANO&3VRosBv707rese!u5X3nMZNV{)N2ri
z1<WV#XpV%MKJh;D>hpH>A*&Zwz3phobb(0h<l*AxwC}Z!qWkS@(OVN^HqC^ma9Gwn
z!`@}=K)qn1plzi|zqjN;qa;Ovd~aeVMOl``$*96MUsfRP=6%H7&9v^xifZvlOtDT8
zPhCo3R9$Kjk7sgmJ}z9kiRM;wPI!&|7Irn_v6rICYjXCFY|?RT=37mILK%Vu?Hu>A
z+?o1_h2)xpDpu=V7;)Il?A%RFkqqTYVKvE7FoWtrAk#jgFM-DtP8Tj9SGv*BW<M9)
z4OB-XO+@ORKQQTSrKPQME8r({NvNS-_VhN%#QRQk@c0?sLE4FGnlRo5^0eC7I2;vo
zqW%UG>uwx(9;&H%8+RMfVZ*g9siDGh7#6bA2C*rb7`i1P^GsO2`Qg5ANg+ZzZoR)!
z>n?h&A>T5>s1>%~kHN9st91?awx#cvw;d$pMBxVRZfejfLt7Z#+Hy#8L?LQ+u3o+a
z>2$dt1OeNdllDK;6K#9cmaJxF(@x|qs)^CL{yjvq1n!L3i@=K)k8j@h5+vUySkF`v
z1p+H9##oq$8*yfVKRZ{*<7{C&nt$@Tvlaj89O(dk`~|x{1Dc)f&j}M!i)Ce#Ms@m>
zW?U^A(4iIIm+CE+IFUtkD}r0FVkthfy|mUHNAbXCC;aiGsrMm7=x=jsXTiynzq&bO
zSE)K(!_wxOxa^GCiyi7>jZw<x$%T2JE2pQkvp%hjy=6hE=Jy&2dH2^Pa$4nl8bA1K
z>@Jtdko`UHQ*hKs<&Wg`V%7<zsMxJ(a)zxgO<m;?;$NpJ8(G`kTm%(Pjvoael`OhH
zWs2UvxeRexEn_LY$cz!LfNd}o&R1*jjLRq;xHmXYta_f#-9bkm7VG3F=|I#On_PbM
zn;aP5+iqtmcn_D2qz(CH{;JTQVTm!SWL8oxs*e%0+(O3jSW;iboK_r8MQG`Rm|z3f
zQnP2(?ogK|H3?A1ysda@$Lt8nTqtJ<UbGFo!AQm(<c~+bff+<)s3#iYsb4fTZY1hn
zP%N+;-rP*hZ3}1fI=Ljcr>uB-OGxw{7o6xxkvjw%Zw_43i?<vTFTz^UTwv=Eym|Wm
zJDn>5<N$7e`;AvF7doT=LV+=kO4VBuoh-8<R)(6N3A?Emu`3k-b?BT`vrD-^2b$D`
zJPnqIK||u3Jk^m|8x$K+A=gqH)%{%^X*C_zSoGp29JAUw?9gR95H$Nnw61`9u#-D@
zZ7Vq>&4@(OYh4a%7FDSE`{0;)B7TTq7?Oya_Lrta6YoYnRAnGf)B9{ts_2jimNsaJ
zdOA?t)|%mddpfa$3%&56bzhCtqX`w}vK$p`-cU-r`T{7qTRse8V`zzDfbSoytt9Av
z12i(GI4OR%VzSi2<w$Zw*RX4Wi6)PKQiFAyfA}GNHK`j*9p~doO^=g#zwhQ*eSM2a
z$c4!ACBL%Nvp)J-Ln-X_&BbaimjyeUdYrT|(mZ`XB_`=KPPEK<D=NzACZ*N+%gnng
z<+zE{>CxvJRc{UTMu+?2GnDji+m=_mT4(cm^2+sgL{4=^r3)Ene;yFuM6$b|a;J>l
zg5MrYYcV=QKa!u<$TU$NQ9*={)n#?w%2BznO;jQ)SaN5JdRUaqqy~E8vy+qWDCNS~
zIMYO5Hho>V=J;x`i~dJ+!avym{B92xK~B)5hxF>`e8(-<1M2UJimMDaO>>Rd@JUn>
z?G6!%pn!fxdJ#Ol*XxQ9IB`FtDK&-SJng6LKcJ%I2~&`RfjgbQw=RWy-4C6<_O@Ly
zxO3JY{<gE^9Nan@7fF9Tz6EN(p`xPKJ>ck{w}FG|c;JyQ#r!dqsPsgMy}v8meOs(2
z)MXKu4{m2jyBf$WW;`3v|L-B(^QF*>6t2_{HwmmwXkFJi=c|3!ru;w<`~o|@Uzlm2
z{$k3v+d;FjAqcsC_q%nTF?Ny9eE2ofE82x)e|Ndqp#1Ds9u6X%7VGJ91yn3sYw>iE
zncIcIuK9$&$i0zQ(>XdL9_d6*u#>9uJyqvs_QfMWz*_M%B2&NYP@S~Zgn@shzMIR`
zI-%P<wPN90siD=Y`CP3=7Pa?PdwHYZrYxO3n-Xdl|L9}%@;U5oPW3i>5NYuR?ZUMr
z7e8BlEFHm2K23A{kKJSqWlP#bTD+gb<vnGp<Z6{CLN9{Y_gwwu_U#k<J=UzW=MGqI
zZF&z*7H3FuJ8x9q+FU+ho2?g{K~*w&*$&4aKoeOktvHOZyYC<a@%g<ew(p<D+Ddfn
zUF&OszF(*DnA@F-J}Mj5vt5{TWLY17!K8k0F_mzHQhW_g>@{ZBj1#1;@~z`1bD(g;
zl*kwE=^;qlZw-Za1b05sFPEf9{J)<Q%y4}!>Ef50Ikqs{U6b_rAb4WBVO0J5`p>eF
z=q%$5AH{m>WNYhh(mEcpb4Kg<3#QQ<Chgt8ZyjBukl3C1o7+@Ze8b+E2Mpj<`(<85
zc$#)VBQG^Yt<?^Fl|WeM%BTB|X6(Te5VlBI70D&W>iz3)E{Ff79)Q310PJcYc9_u`
zYAUv*+H^IhH{^BHPA<;4_U3@m<dPt=`&&bt8n2+(H{!i)J?<C#wL3+JG<~4^i%xwG
z^&4&<_p^Rl*SmPU0cl?}koU44Z|x9BmfhLFGjK5xBT^V|e_Wn|$rv!(-dK}IW|Kqt
zah8R{)tD?Xy7QI9daebxHX$=)k0~Jfc_OD}V=W;~{@6~Hg{66BtPR`PH$h7$!|*Mk
zo%u+ipR&5zCtYS0TlYotT1aJK>9PPe9+tj?(RH+}MkOrhZ=A<F0-VjIT#_ve{+*@d
znUX~|CpT^m85n+pT{Ora=%v&IIv0j;-##JI<<pX`WLHqn>g~~A<60yDRAS&J7f_>f
z0vC>(MxAB?upDINdG)0q%akqoqul4&d8s0jNYz<fcZN`s*i4V}d5k!YEZeAI#niNe
z!;#LDH{smi(j&f_O{+iDUkgWP<AOp$uD5l!bJdP_o*u*??pn<HB-O>?VQ@{Fm%&4n
zV1li<9!D8#7W=PicC8gxQ|?gTKl3_wyT*v~3c@yDh_!qm>t0Zj68C*ACuVIV-pPw1
z9hHG>uH#fAgeQ`1umZ5NzfawHbR7-t>I!J=^`sEYsOxY5lV!i~yf6~Es=$_AC%kgw
zroF^a;QxN!->hD&r_&PKHw%4j@z0E~)xA3k_a5DV-n-wPRW%U>a&X#RjHNXsfnz>)
zXNs@Li4boKu11JEt%UO9WQK-z&DA;+Fiz>XE&1zvXK4a%<2sB?JBjXBS37v9v1Oat
zw;oTlc5bz6slKvGF76DU(?c*Yf-y!rZzb7^F><=N3L%Vw<ou6;nhB!lm6xvLd;pDy
z$Lh@0UKxznecVD;6ZuhlK_?59M)NDnm1LZ2FCb?PMtZWl`o_a{g*XQ1;p=>>A}y*<
z;}(;7f3^=LaJ6R+W-#J%W*qjUoP;B(Ik~YZ$iUcnpjP&EHY968Rc5${237H<>0_pH
z=D4bLMqi?Xg)x$y{=1q@Z$(c}pR34HVlp#-O$e|rBxdePt#GWY<JH(|NtT8{3nP=w
z!Rv{@N1$_ld3u*$13Wl_r#uf6+nN9Ug5R{``$K@T0$cNHOtJ6K6cV)c*nmW@c|XvL
zC?wiWs;c%!r)zpQt^2bAqcldsJ&KK?$CGPV0^ix}yE%{QxXy&*sW;ji9H~uo1?4zQ
z7`HX7bY`gN>7Us<`D4zOsnHS)@67XD(k6rhfxpk*q!y4Xa^Bm;fQgUbi8U-0I9)Oj
z8yN{psJ~p*!#GmdF`)n)A0&-H42*qutBg>*g+Dj*=Z)mxs98gw)pFG{WrLqa=}!Bz
z>d@OQdFL@@)rWLaluiyh5qDBQY`s>fYOl8QTY>MiO23I-ZT`~NrKd>~=<a5P%U%5;
z#F@RUh=U{oWW|sq$wq2T#T@PSHRq;HfeKh^7J*vBnPvA$_$ay1{M}4l?M!qYzV(3h
zPzN$bqruXj%vw8t-*#Z^R2!q7yRO~gC=MMx@C+Ru;~-Rs^3)y5JwF`^*G<f!wY47g
zw3IG)=_G5YG7iH5Zxu6ULo+lx=6-{SuY=h4uivH&A6umi7iMDoYphp&UAkLK{Gc(*
zW1{+^gx~I(CYG98CdwwP(fQoz8ipKW7M?#f5?VwRVp{^9SF6a?;e|N!lvKC_;%*}J
zKY>sdp#=5>N7SYr*7f{FBv0VBw-KH?d;nQF84T-W|7lzp?qYJo&CiwU?*5CeT6Y!)
zelUP}waNbVYSVUwfSA;UP`TfZl3$>}je$Gjex{A4{U0%FDc+0DA3#E~@9i0THrFzV
zOIXi|c_GQDP);3a6D{GV*uH~H`?zDWT9-Y_vp{<9*&XrUoo@7@(EH1Qm$K7ui&T40
z3+e6SsV@JtKKCjPKW_Xz6OpTS+rV)N>4o3V*wnEVd~Q8l%*WBvDZvOcrMG<jWFjaW
zpRG)FN`ZMuTE|vi3rbZ#354$*)774*u4}GGVch9n7$rvFW13T2?`zEqzperAyJG-8
z@CW~#QxvV|Zs}nIH*SK{Kh3Xk?@Fj+bz`;$_ay~{j#FIcDb9?rzLW;I>`rnqwH-+Z
z9g5P45EacIrOXk%BIF>!O|0MiT(VL^JAbev-R*RfkWcyvKo%i^&C6WN729KxFS)qr
zw{opz2q2ZvSxTqIb`e}rHntqIZKrciS&-uBgTEW--~CO6D_SU>>M7O<9hKV<oseTo
zGyg>oYEYJhoT#Q8pJ-o6-`gbCI+QXLyJ=#RVMwgY9602PzqbB7Q>ey~CuMvOtUY@E
zNn7|P@T&EK>a?v~lGjY%yH6b|FN`Y90X!|VeZl#RwN#Ha>l8UB1IhU(6q61#OOA2w
zjCz#c9zm^c`AYop?#Wx+4uoD<%V>D+1kN50>uGVVE2gU_|2PdxOQC3DYV>jp!TkJ!
zoITY=1C*cayK#(HKx81ctY{VX;OZ1}thyJ;L;{%USRQCrG=j_wRyM&`8Ao>??LXR@
z${|E8P>edu&lv27>BhHKS2<9YHT(Qtd>?;abEwsYx*KC}+K4or$NRl@M4&&lLZjm4
zR}+2xby!EBZyI4zXh6w&&R(|4Lx&hRYOW8KVjRuplX)qR@*-KiYj3I1WYXO8q*_n8
z-qF3+7=WB>T|`c!w(K7r3`As;@@(1PVdE0sEt5ArN;5dT46fF?6{T>Mk0$he)!&^=
zJh@W-V}O<sE{9a&YJ!w1aq%<GejnpmH>!?bw!3cq9omZ_Rwx>I0h6ne^e2TEk|bR-
zjG2%q)bsQ}!btJ~uA14oxs|r$*R$VN#G>q1#YXVF<@qsu4$*&dd)!`pDb%bY7K_T%
zAgmTh4ur(F{q`~F;WohW6xo7DuU+IYjdqsj<argoHyW^eXO6^+mg*}S8)JHXX}2gZ
zaDvO4?w?kE#g;O4y<>s71i}s+LK0md@A`x*I7L6o)+HA)&<qS7Y4;Ux*jo&**`*Bq
z@^9on)PP)ctX$n8KwAa-8#(G#>tv0YiLO>ZmFkl7Y|rnr_Pxp)aM(cUF>9<}QaLhx
zN54vd!zp|ND6oAGn&_ZXACe9of=I9pys8%%3rT31SQw~QyYL@Yl9b>=lrzrcq$nId
z2G|T4ZDpUgV}D$Iml!kk$$GX#f;F~Rb*rDI=f$A@lrW+sFR!J;JE8C6!wA#T?Y7=0
zf&{wSO6JEBD#;Pr*2j{SLfPEwkd*@aj;n``dKU^kEsn!)k5jrx9$O_7tr^LIt2k>u
zv#UrwX0$z43{ZX|xGpKTURjRboy#DO{weV$^MEyLl-%XbnDHVUfdbtf2KuCVp}Xe}
z-rty!6wwyHsDn<_20U~+^yQ|DyQv4$67!?OR@^TkpcjZw5J9hC&lWe0BiIkf-0PK%
z%VwrgVqRtvRr!h5HzT2GD<v2S!)%}6S@7-L+{lsXWRM*@M>tjBvD31h)V+YrkEkVs
zDuaDMpJA9FgIC$CPu50~;vAm)P^=p7ka~kQ$%9DZTX%V}ks!N&4;cRn+WTw220;|a
zpMkf#i!Xd+yYUB@bn1Tt60-!>^gRQBiI-ZL5*!IRSsYqL3@LIdWpa$0t8q!2e*Jw;
zT5$|G;Be)Uew<grRbL1Ys{}ajb#019YSVJ1ue5V{bOa@{orJ|rPwT}dh;9@1ldjUQ
z*lE-8b{Nd(E6>Y+TKV>rWsR;`Wf()5#7J3v)~qeG@p+Z50MDa{9R8!v?K;%Ml10#@
z(e#)&PFLGB#DdjwXx<1bCvYGHTit%naVxv;;ire*>~!1!(eX`cL6Kz-^7C1&iBY+u
z>(QAr$CM(l8l4mY$zc9{MX5KaLpEy1<d|*-(AZ~U&vo4DmT+4fjRfBUExY+<<5VRi
zn@$J5CQwri9<ef5%2Jn2xW*Ue-q}tjYns}cOoc8aW;}3sB_bB3EOk`JNx^S&^ZARR
zo(8ynV@?0&lL&60h@*d$y@(g}M1tLKpXD8McWROybHS0$M|J#Db^|d^E5>Q5Aage@
z*G{&6R2sy3U}9yTnc5~*Zg;(x=dOfw|2JviX?6F-zNA-=5}5d0N6yvJbwGG)zxO!5
zxRJ7vxEZ~PnHz*(au}sx^>w`acf#CXcjL__o?b$Q5ZM`<#vGxRLp+AD#%VU+XVN+I
z;IT9P{%y`!wmXk2lOrvEaECQ(Cuz=hJYlY1l>j@Av0=7FEF0KY#JE{))0rbxMM{X^
zQ$odQ<gB4^_;iInvemt)-u3o@b1qBPx-aXV43^`bZnvnuh26?nGtU)qOY^H*N<CWq
zfsE%SPb~!RcJ?MgQ6r*3AMTgeFmZU8tcNMO&VLKCz7vx04X$2S#Kfpu1I*CwqzDE;
za|hJ8*&@C|Fnbj63hE{pal>g6mfn}JVI@@NIczYXCpW`FXw<)uJEN0Qv{26}S1hZW
zsXQYvTA<gks`=+F(9FRu-j`5%WZ9s!_p=SF(uEwSl`~T{_IEZB{0y$eXvng$@xj3s
zzA^q*1IZ)1m4rg7<T$&lt!!Kq6pCj+gfzB1>;=~=1o(>kx(cro{iX6x{n@mo$C5vJ
zi_{o~CnL6Qf9FkoH<2kFSJId1dA_d?zF;(3<}N9x=H$lA-Gv$z7FEK>pR?ss-fP%m
zI4J?O;PQW>*cgf)d(xDASFFVBwBFSz>Zg_vXmO0EbN&rgNR89+V2(b*l=mq%&_L2j
z$-bg3+Us&HPMktAoR%im!EvpbcY7O6pW5lHe%v4)i*t^l+^wEuKs_OL;;GS*2Y&P@
zQP|aPKS>R;DH1Y2>R*mGpd<Or>*N9Y4QTrx59;VHL2@02X~L9_8Pv$T39<Rnpc_4K
zxYu(4)QPU;T`CUa^0Bs_oT-U4U{8#CI>aSiWc10DB7EAXvX#4kcxp7G#(UNJzYZEk
zjeWD(JxYij&ZMlup%joOJGHVp_Vu)6(WW@#&=R7h5T$Hf(Qw1Pt`jq~+-G?b((tT!
z%ds%g)iNV1MNSzN#x_<i?MQR|Noj@D$qVBGImsk-8fMMjyh+s^8oO=Gree64D`wyN
z`Eh1JxzZzlO{s`QE`+Vh!H5}kjgOuFbn&vWeXPgY@qjUhLR9R8LkANE_}3Huzb}nN
z_7t|>`(1slCU~I=So8O=w*QFwvD!oH`BJGr>kp`K-8<gD6D%Ut^i<Up(hslfWA+;1
zPs;y-p>TWlp4VEg;=|UCn1zm0Bd<m*56rgFg?Z(inH|r`7y;d0?1&DmNu&AvO2NT+
z&<s)~l?9uKk6cBw7Q9E2T!z+AkKtb%NX>Upo~|e}s=A3_?uEd9u&%A)yDc_1-d}{@
z4?^#cp*surw+pAb_oup7Y4?|Db|s+f3~Vm};N9O3eUaUbb&mallDc>3xYq8y-Mt6j
z?B-ik8__X(-9p<rzeWqxd9>ZvVm$6HH=qDFChW!aNyBiW1ymLxyEvf76nZ^nH{Oc*
zHaQHVo0}Mt1kuFH{|1_T{w{@WNsTxn@xJgShTa^e19-fFqJjx`-&u<W6HHSn5KlHF
z4*RU0tsM_QbxA}H#Q^D|3z1}KEHC?uMg2e;vTZT7vmXX~?Y99`HjlXNFW*qOBHo^g
zRV|9Onx?_{qG=YLw=&}bZVwR#0B+~i*Ra@xznG)YPC;4AABir%H+9hjoZrEM4IsUP
z%agqs)lM4M<i<mPjvVbemzQ98S0D)A&h(yz&gXEGB01p?3I1rPNX-TzA%GPXzQySG
z@E@#;0nuwKZTRKdPok#VJ=>i&Qh7q|OzLD5@KM8hqUHM}rMoD3Q9&<RgbQT(8~uA>
zjOJtkQwt|Bv4sK{fy@7tK7qfbZ))erS1d~2dsKtRm)mH69WC0G&&3ZP;?E>nvmWUW
z>0qDD1)bcNg?V3op&2Pn&~OpbWUEciBCSo$$<j{Rx?u0bCD@RH^bZ^I(2j(h+M0J)
zUScTQchA?1^H@3^e3YNR2{pEk=^6V%SyK2YIk(&IR$dU5B!b<ADRFLcTd=>)3<m!H
zmzF+qNfL`!Rzaib4`O}zGH=oxA0jp)W)d0-<04;WDgOn|N6LtLN#wF8zb)!`nwUS{
z@uMKtdNx;(mEU|mku9;r>{sJ)%NLpn{Q_cZVMTILKApLFwVX>!evAI;LJef@2Pzlk
z91%mua35LGKW#2PJ?N)YQE&2kqp9@TJra06Cn^N_^gm_QG&inb;RYM8de+}v{yt|*
zJbAZ`dQP`IfZ@olm~T&)gE@$MyN}dv?2<%b9-r-ij(}+I%Q){L_l*>CIX_%KJ=h8z
zXkOp%4-%&Y2>63wBR<BCKO_=G5FV(b+qoYNu0}wczsGF^F{TdQUwhxn%O>#=r4#2f
zlR=4b{@%RU5lfDG(ZS^g>08Hf*nKGWI&k%06!fKb!B;?cF>KFr>De^~4Rt&Bj2n2O
zzqL~C)lB#4?YQvvVOiR(3`X-n;x0VslyEh6kc$f-Ob@iNZ^2`HK-c-hs4DCSIZ}Ka
zApj7H6*IyCm{9N-1HNGmuJ<L8&jL^YYxKLJjXMl79xZyDFyJTOaw#<d@R_k~(#}qC
z(TKn{{-venTB@SNfqqZhg{T*h!OMT`@nj2|xAfZ&ob+fAeX7TH?q^(ZyILvm@a6?j
zZB6gv^2&YP!Yz9>>?l~5P)rEpE|L>v$B2*r|I--6`}Z^gY1Q}J(G_-w;Xpyp5)aiR
z2W3Fw&38~ivbv#CL0)ePL~^Gp&SJOzt;LLYEc^Bu70tw!QOe+&8g^o~Un?Tf$)+eu
zsLZv@U+w^&3KvBWP?mXMR4>M{Va#f~z&Cw8@8P?ic}4w{i@g5*(dEgGC1xt>>Yr%=
ze;B+kNJojehMF*=mNyGS(%*C*jC|VT(1*cL2n3kBBHAgpdJ+h<=6M{aJl-J^+pE>6
z=s0mO>ibJ|Gg3kNWODPP^Kz$T--Oby_07`KVCRKz{uUDl+H=zz+H>=!0XF+bS#k}g
zAsCUZqOKL>z3eo}h`df>0Us<oKgPe#a8YCqV)c`@@`je8Ml3vBCx?-ZX4uPXRRm%(
z5!k^#Li%fdgp__`b7|;c>!$+h{lx~x*xa*c%;yc{R=<l^a{u^_>tpJPxan$8at%5p
z125u_25$V)u|D8E0@%Z*U(yWZ{R_L`Hvif908j!j#HzKZZI)>KAlv`_$MZQS)+WG9
z@A9=Q+62$^so64d3APB(4CH-tblClLn4TR$2%QBAMbG&#kafK<!*XM<S^WV}!wXlA
z5QwJQ0f`+8FI>vGG3bJkfKRi)VKD5Wm4YjPCRU=^gpgOC!_{j5t%+b5hk~1uYY$Xp
zwp;3G^#Wo@({a<=ND_eirBCU3#9p3uqCX{+=3!OCQeW=JHJ)VZ$^M4@69^&)L;}3T
zc%<PCTIEDgLI6uwut+4U@#Q*y69s|;1;>-70j_1IdRQ<KxR;7dpt^4)5+Vpxq#Heh
zz4Wheq4;`lZ~=e4iFoc@IRE<+{GVf|5Z9tu-XdLdXw{Qhfs5+|_)uy${$<)oLw~@f
zu%OFQ$y}m!L9;*;mvHRJRWxTMwLwC*mRG5aRx<BnHtA@&ArE#U<JM|_cFvl#>1KZM
zjhOZZ(MSf(POA^aZ*@QfjtLQO03<FpZ7asd|4HlQN!dADe4F?YZwHbxb`*vc#JBjL
zHnAG(?_Ky;E*`n`=V2mWYCd;fmx}q^Rj7u+_nUt0-LVcuvlWP{1zhkH<{Cv@XZV%a
z^6mS;IlgimFO(RK%dw?UM&IX^C86E%KSJeuOf(}F>2lSXal{pzf#sB0t(rubtQn%g
zpP0~>5=?s47b-`~&;>hK&6wHkfNtP+aFxal-!JE!BmF|j?Z?>JX;s(scS2&Fv$GSk
z;JtCfLa$cmX_)WH)pu(a5@|-1by)rvb@#1($Tzi`>e_-x^IibGm}+S&;u~(f;S__k
zg*wi5npn^X9d+6-V0wvDv@1(<z#Aku+xUPnJ&HTBNRu~8+B6M*wN>&VF0-~*vN}@K
z*dUEwMHw;Fo4|wFp=d|U3)=#VJ+hp`XjPsyGa#OPOyJHa;Snt*PIc&E{+axHpz(cd
zC0dGazSFOr#y?T$NY+m5xmc8IW!%IRUd?Ck)ejyB)n@Qk1nsFLJ6ODZnBeh)8{n`$
zfL#x?!;RFZPN&j4S`7rm9w|z$dSJ@2pK(<=O{_S8%Q>6qQdCe{Ei$T##JMx=%rGPp
z1V}b61@lO#Ae>s1B+tsQcrWdNbnm^kKWXRCbGP*nG@sheD|)Ysf+ue{fdGvBrjZo=
z9@8~iQDc-6|4M59fN=zbr@n^)+|-`$(h@KL%Ivl?n4euxJuc`Lo{_fnOsK7IP=UHt
zWG6z-J!$TjYb=ghC@hZ9QbLV0ctWKkTsXW$RxUP0hQet%E5fP24$khQOl-9ZwPV(@
zzN~*SjAG8_H_0oTrU1%bs{#@MD2);dv2~ktbB_MR5aipe*T2PhGQWNA#u3DrZ}+34
zU#c3p9r~E4WF}w@wq?{1q?<1yso${c(jrXmUB|a_G;wyAu!AM%O!P{Ky^Qkz2-WxL
zVF!z2flL;Uf~`9@rL8+&Fj4>Hpc)ie#spYQi>)tJ5f74Ugz`2;3N`itXfES)a9&;S
ze4Ak_@Q1eDG<xlD!j@w7g;pz<fM)Q1#F_$TlIB5N8Dl3(qOHtR;jX$3r3U#utw9xl
zzK}6<(-OHM+>}TTRU=B2C?2nZ1w>?T34`K9Al}=O-#iLLey5Vw0*ZifAGYZa8ImkD
z-Oe;U91Qdivugt~0owr4)Q#dAzq`@Be<8=(gT`yD%?!e7Y!<xEc?IHl{RO6jUAL_Q
zY+2%|pWs~*<9oe=bR`uu&g;?x$Npr&m;KfDEAMXs5&&pR6eIpw;#<qFV_~-$YWSs`
zwS>3ZZDrW91-Wg%l79z+5QG$1jgbW7|HMwTbP?n;JL`MD8CQ<Q!<NfsRCy10eQLPH
z%~-2A?|IO`KO04U^#Bn4rWhP)ESllPIj&F;U4HzKT<fveD|f8nsiffN32Rl+&+X*A
zrk@<{(fK(skpL3VC4tTjTkQL|jwG9zg_E|J!vteo{pg7?UEBXVgcpkxj?a4!{Gi<u
zJcY=-owAhl5(NY^H2<K#y3Eq@huHFS3c$uFAyD%S_}~t6$|ZVWZ1s}BX*rn9$@swX
z1Qp4@Ik0k<6O0k7tfwSy`}Lo-z%M_*Y`f%7TXs{YEjU^~$we2kRFw;YeRWUszhI2k
zYv%Z-<A~RuT)@=i<t~T$5*e}*g24wN%eI7%?p{B=4UWTjA^%32`BGyX$F84xB3QXa
z;>Dsav!~a;Dk?>fTdQTSNt5R}b-EbKEG^@5G$x^*UbR!XY_}gtpWOImde|6fyO;K<
z5}cd|;qWqmheuY(clw{8ocb0P#`UWx4<AHS(q<TjkXRGBQRg~>5`GKg^bL{zS|?j>
zX#6xFqhWp&r!gOl)x-(+IePD?SwzkV9w#7O-{}sjo;t36vwhZqRXC@f71C$uRTfo4
z*dFnwuqzzCiO(`%=S4EurqQ7zai&&#+Bf4{<MSpWklNuNz>D8Bow&(%@r2`|!@!f0
zn<&^<1Zch#As?&j&`e|x@Z7yU&Mb=#ub-3-T6>mHdtl5V+<Vt?e|K<Cyy%}<Ma&D#
zNASqyX6~x(O=5TJ$izP2$yU&LblCNdZ&#`<1Ap*CP__<0R+mp&y({X(jo%UQv^iX;
zW3g|oE#^=wk};if@Z0GaDjY-?8KpHvj`v7XZf~Ya6(dA<(N+K9fe};%LVk21xJeJ<
znr@5%IcmC*tmCq=o5=I0J8#5Z)Rf3vhVb-18vg{hN+)i^jlkkAMLtbCd+9C4pZ+44
z4S`*AMJroSwCXFO6>!2jjg^k4cUr?9i*2typ0q!N38L5O!tha{zL-|-Zy4#{sM1Rk
zx@`j_hgaGUSOi}XWf=*Q!SuXlTTgj1qHvUIM|JxTp18B@hOqN8fohI>XOxppmD0(c
zYEv$;?R>T>`RUMRW~=kuV%lc3P4><Q?4)d0Q6-Yb+jO#hwa3KgUF)pT-EmYAM&S(j
zS<DZU=gsJTG-P-QJ?y8TT3wBQ#cBYwBSV<@?CCOHF-C6c)?>D7s6+pI#&}m@#(lqv
zX359+>O*(Roi3lo?9TqI?-P7`v-nJf%g!vUkASS`mFcI{J}PGe><0;8Ts&iC^Wy<I
zD(Qnl4uu(vjANAz+xfQ-ie8h1$jfYINT~kw5$vG7>ZId4&GZ)y8%JH0Og{eDNs({L
zF|CL^2OO`mwI1W1FArZ-I>c(y3hyQ6mUB~<HAS=G_2vq$v+lTt7)Fb)N<p0-K6h_9
zf;4}-^`h3m@a3*A;cd8X!2QqMr8hrMgq{$W;5@<-DV-(#_-u#5&<6$s(20R?=k$i?
ziS@#jq<~zFl2P@O@_2(Mw+YqmGyr~Z{^^LS!jE8}YzJ*~oQQhHwiqb`u{|?~SU6Hi
z7eKl8=lvrwdim{ekXWJ!05uw(S)Q6&hWgG*CL0S#a~hTccGcV*VP7baoS7nbpAUXZ
z?!rbf14v9bUZd0+cH3~vk!>;9KT5_O9+Rv|$N`k(vK}rXd&$K<Q6j$F&*BHt>sP!M
zCDYek?tLjYO7yvV67yr<#-brvqmUxjTGFF#T<^20D^FjCVW?VIk_PX2TK(StZqe|#
z9wy!cGT0eBqhEo)6Z5t@sADH;xl%q(u5tFYf$bS<4<6yrjGc<49QGVp3>5~!c^5T^
zJd49sY<2ZNr7%BR(>%?tIv9KiV;@Q2NXyN?hRegqPxwC;Yk5VL3E3isqbn6r;qnhw
zWqk^A*=5GPR@hew;>(fo9B0aPfqRW)MmXu_eP}6*fcXFOdHSO*WO@9xkL$TH4)28R
zV8&X0#joFBgkWsn^1zZ|fm2Q{-<t&-ZxQm{P=f2eT;D!Rm2j;rt^9GD!4#Ou_O}^{
z&BMwS=dMC==PuprsardmX$wF2;qr`!ZL3^Nv<(lS9(5(NMEV@~So2=rANLxg>^rl#
z!FME2Gr+s<y`;uZ6Iv?aab#p47!7|D#`L2oxn(x+#4@c0*9;l>)MP7f0LZxq{0E3Q
z5+q%J6fMIz3~(y>LVAePOfSFuB&<^8eAW>==e>e+I>o_|3@y$xY*`N6we@BSOvA1W
zP$A?61~D}=t%gA4H0MYMjpSb%uQ^hWhMr8JzA2OBvj=W{hNGPZKMGNQFJnxn-~kW|
zBAcG%|C|6aK+deNc7f(<E->BKmq0>&5>PgQTPAh%lPUd}_V*#lr4O*N)Ub#iwOa&%
zjt1G(zDmPKleiz_s_WevCvX5p*%B`lBxCzAC@C@Cca#ql7H|U`QS)sej;6Uh{)L6h
zpXvE`MFaWmtN(oqJ^mY*+@;gmf#79j6ndH7Ta<PDfKL6w2*Jz`hQ9sC>^0u@tRZh&
zcTr&%;B{$L!515Ac9=17x)FK4vs|>GG{R1vCw0!_+?^k{ea1g+)`fnj(PUV`f|7mG
zgTAfN-fqYUzvH|Lzmk}+-L|zADR$-W<!dkC229!27BCxug9y_3>g#Xc*ckgi;c%!7
zj3?Ot!#cWQI8S&j^M3~H(y?SEs(|I6b~bj1%-D5mWrSU0UJUk5dNr@_a$bc}oQOCQ
zJ%L8IDtd}a8SM3wXujK;M@okir4)`}LS2t(6VpNUyxL8h=rD{F3E9tA1$FY2K6W)x
zHn==y>rJFyB=fHa@S<Z))FS_6OSgCh)w>1SSJdE_P)w#W=e6sb+tn}!0AyFCgxF1+
zT$fW_A0rU9qR8J*QIySqHL{P-IB}|jeN2)W7&8sbTR(N?=KRJrtZB{YJg1yObz-g^
zC$jukWFK{fcD8_nfW+j{PtZe=Lq^D%?bWdQ>P{f=l2*27H;v?{$kDKR{MBb%F-1aN
zg>os}2p5bpFK|Q+sDv<!J;I{ABxXb@(wGUkPw&T;n|0jv;%(BZ(FuA5`nm{IKl<F4
z>@!>9@QSqHPql(*{2OF6)o0l$^>u03VJk~F{1bo)7tj7ZhqTtqX2@b9UwTRe{plNd
z%>R`GK33zM^8U{LOp_OwxcX$mr_uTT_8iw+j4iG0rR=hjEBbPFL$r#@Op%NuKPo$J
zIJJB2zkpwtzxjy|(haxF|0sS|d-V)%&)*(stOZsu4mVt?i}A9WeU{ez@FCuwnW}LD
znUzNF#5`3lG;}lFQK#`eqm!t*X@f%$oO6@B))bAMSZ?#9a=r2y!*ob@JHmudx9E*9
zS<dl{_Gs<FUSjR4wKdDs?@z{90Fz=xeFKTqa!<BnEzP?QxszfHFp>7Xo8fOFA13{P
z)+!w@lH?d-?dOd5=EWd#;el9hO#W=qF3-S>NL5+6qPR)nmmN5d(?N;*_T`xdQF4V8
zNvK*5F)Vv$O^To*4=J(co|HW)ZF58*?M46%!|107Nf)}xzfPE(G#PKS@o<ZCJtVXP
z1<D2l0!%m>FAOG|I+p~&O{<*bw%7$QObg@2HR$(Qaq3=D>ue9$TOxzpc)UY}J6B@G
zGye;3@WC@6uOKxpzd)s#(4Z5v2UFBC%gPfpc?Ab?K6l;t#}K3Pu=}d(+oZ5$JD5FR
zP*hcy?M55tg^QrS?bmFi+5sMq%T(=SC@7939vi?HVr8uO48v;S-3OEOB;bbMcxK}J
z7|iN}OENMTzG#ju5_Eij>wT{)(1j~l2Wo!Sd@(fu=TmbK%p{`;6anG^#LnV|iMxWU
zyI#=sX3GMKcXxiXMUp%JCeaIJsGk6$YeT37xe<=d413?|0SSJ%qnH+GyN86Fm^Uc6
z_f~X0TJ#_Ge2f#G_&<B3zDi?q5nn=SV&7y>CE^=fun_5yH8YIqQ)me4MQ7@evD@nB
z3E;)L`Ega9^RbO6jd*OcLYjV(ainG_>56J&XydFM+&gucY1rWUF0u4P^DkyC>d_}+
z7BLOKkz;w>sdaXdIAIjI%9;~-A&>+$`BJlStiCywKs&Cr^jIa9ld}Il6MRPL>Lu^j
zznCA(jjuB+$55%}jc%7u{op9GKsPK>7Q{{ATj%y#ig)xF98CtRM;WG>>9_mKT5{i*
z!^exizQ23-aAb|F?wtr}lc*SJ<`X)s>Lmb`ocL|F_f<AP%nlcWBmqrqhMZnOQcJ%&
zOfv(%T_h*=aLYFn#wr3B#qiU+s{thr%PeH?t1v*);DbS|(GtQRqWc$RAEIF12&&lR
zVUy&?C>|9hun$u|LL#=`!HtAW99MwTzC;Bpq$G#?AX4-eGrSodyhM;S6I8c3JbHv%
zj)gxZj_2;7BRafDn@O|SulU92DK^-&#Xw=zN&Bo4tLMUr0QV`-qBQRcsS)lQ5=me!
z-XNeJ=cKoacuPUojMgy%3LJVdlY8KJ8USO|sMJ|@8Hwmw0rYfJ7+6ad!D)S9j1Yl1
zF`c#lz=2XmFqkV56Z<WH&_sa6qCDtWmlYX0Kx=1C_P&@DteS%V>QdPL=6DNqzpb|t
z6f@+1Q9ztsXHqD%OO=PDLYP!e@0Z%C9R27Hu+=zAgUbAWgq?L*l<n5`XBYu#6(p2U
zk&>ai#{fiH1cuHb1f?YghT#DOY3VKrrMpYILAq<`?vC%~dG~(by^n9d-}kS7;Bnmd
zJ=eO{TIcyY1%fRyS#jOeGY>^y_W;=AgKE8B)w3u$qe#U+6I*fNg$3ReK;eM?D20(0
z#OD4<_3q+xnR@<dytEb=6)fEs*E5u>K(ko1m~p*Nrn>(hn=8u-*kGrXodUoSgUZ^N
zyBkGrs}Etd<xKm~O=P{ZMCB-brR#;0TZ}%M>XEQG&}>be7|Mv}g7L)r>98{SB3GiF
zq)d*$v4<*TGQ<uB#0~GUPa!oS#6BInThhc>hEyg~`7(6l_vz*q_rTuG(jEwNwn*t2
z2j(*1!djphW+gO0yxn{HlpW^7TPw~93PrutpsrBj9{`2hDBc(lW2|ZR(+R!oghFuF
zIA_00d<uTNS7D1LCf9Dbk9<`a|DNgigYD1zBT_TNAyGX`onq!s96N(w5yOL;nhiQ0
z$Jm|Tb67rL8wy6`QNQ=f`9)yfI2D#xPUP^Mrc+_M!{5GO0BaTF(ge@b_$Vh{a{uV$
z%Wut25`1El5Ha*4fzAbU=UP0>J`>TRV(j7lbMU7(=K+XCNgXP?=7aM##KFbw?|kE5
zf9LOSf9ENH-abjd+Vsgi5F9Q}(4NE9V186!;SdyM0C?Vs?re9D5q#--b`&S^ww`v6
zBH?O1T{ur#C1L0YU~=8$AGXK2k4`=|v9N35m-Xoj9n$1#>iQ<Snf{^689CmgR{^7M
z3m)cI2qwIcAXs`Lx?$`{H|7CsOQ4tOA68BOLpx*e8W<C{P%;}kXLvx4HNT$Sll+b-
zh_~&z#C0;BP4i$DuQk>$e$h9efO=AFt){cwL%j!8&|g+FiXvUK^`Cx{_+7&dZ8DIA
z+SpsTFxC_mAq^^%t!rASK<yYzCm<JtMLsLohz38FAmu;>4IbDwp^^?A8v`%@;5pG5
z4MSkyw||Jnb~@fGMP?Lb-Uq#<&J-lcvzh5#OfV+dxt^jp2yPPG+oZl|+ukP6l%6&s
zU>!B4S`eMCkCwX{AKbeTgsRW5KohXz=T^c4q6TgltpKgq|D3Lde?N8Wev$@uQx3|(
z;j9Ag{RK2QM$OfZ8_~hY2MK_BCR96r*Nq)lrLeDr_f*HH8&#JL;pxO}Wb3&92(ltl
zk*ut3e%lwGAHZT5Ar!RMN6uT>5Sg#jQJ$J5S4`eBlxzbjMO;Kq<I949`cGi$r)&K~
z4s15zhR+N6Q}1*Ov;zpr{Ov$1;*bec-DgcW1iV15;i{4QW-(Nyv`2$ls`VIze$Fa(
zS{Rs%J*;|M*D(plXw{E5eu3mfo!a#WV{@AHc4=zDg-K}p?e{A_>nw`nptpC-w-Z$?
zuFR_8$_|cWnBUFx)F-q)oQa8HZFY{qD}C0~tp^J;U0aLu7$>elx&jm1IbtNKaz<hA
zUM0nGLQ!8hL)z_n{+(ky9l~$vj?L1zv*KCfJrq0cSkehuo3rAYaLcuBkRbJE8}X^8
zine*fdO_%s!dJWFl0T&d1upfr?zeT7#VtfT=wFAFX_2?+d4nWGnN$RV-#jKyP4E;*
zW}-*^F^$RofTwO+l4lv=!m*wk*E*dli_}RE6HWkt?!7~2I{AZE7j-|m2xk6#O_+qh
zbIW)wY>%-z$Gg6F3{T`x_8U`8q)KK)2hpm+g%R^CD*6PFAFB_;#gH-B$r^Lv_*Vh{
zVOC^01{++aQk{rb{cIN<$N-5oAgT6{RPbDisa@w&t)nY0C@I^OiTDfXi_<OZKB#(g
z!4BL&3yXb)kAXYPBweyh{n9#hTOe4K`s)`v$4tRQpA+jTyQ2%ic$LlN`AT6Vs%5)!
ze{<0$9uH%qPiA@yOO-p2mmzMQPt#B^TmsQ3*dSPbYAyM84P%ff&c%~$T+$g#^Vw@!
zMk=Od;}}4I1pHNncm?juL_u~tuKHTAylHE!XLNKep)R9fI?03mm%yg}f5yPge?MUm
z3{{%ntzY;?nOpL?-Xk)W>cVf?@8XIl4_jLW4yQIv9j<;pwl~Q4cQDKxbeQUj+OWvK
zmT?RK+(34bW`lYA1!Pew<b^mxAa41K1+D1Kl{-eyh@s&guV+*>x!nsH1E{lUCLO;P
zf!;t!XFN>aMp0-z4GgY?OK_*U?*XAtw^&xo5F=nX_g^|PBIjVnPl=|L5=ai^#zM1L
ztssW0`I4>1sfc3d(8?f1_LPa=t}2vsRM2RCceUcQkp(g|^O}z(R^`;(ZOH&otp*ED
zqV*@H6F%q72Cox(YT?w5b|+%qm>B!Hv}zxR`o&_q&EDEk7Glh%^E}DmHFU=~nBIqc
z@9aKt={zsIj;CW@_&T@U*wy_QsmRlT^uq*bLG=>-ZK&yjrVXavG6B#o*{>%^CdU)|
zidF+LvAjaP?nK;c^n*h-<IOeaN7mHh!1FG8srIkW+w1S=oq@%o7p9WYs{8Kpkrk<`
zLq;Q6PR)zrhzX^lmtg{OBMB-y0|>V~C4XR>BiT0S@GdidBc8T=J<o>z=zO8HQ+<M<
zBfKcUqr0Y)O~;6sh`T`Jdk3qL{&&o^5di}-*0!<+&<-n+N)KVlt1K9f|Gwt-tk@No
zoi)dP#s0q>-<J<sUj*cEbs0DL`Mp4)v(>QY2KGza3P_3@^C}KrYx89eKcY~GM8rU4
z36(ozv@9B5zOv4Z#$x#Uex6OzFLyp@B}RL~x%D2(Hnkg^gtxJYqU{~D-aoX-B+_^E
zZ$duWv*Ri^G<e%~B`Qx6R&yeF-R-*CCGO0w#=b67Ss6U&2psLE%xenT`%i0)olKdg
z#Ab~tr{1<T*~v$aj}OY6Z*?Tt*900$Y=`yf(!ubJ0=Rs%xC11bfRiHxj~loPW0HYm
z`!+U=5%49@QWmMldUQ9M%Hz+v7IX7Z#TJzqw><;owkor9rvwSpThI91&y4xpw?8DD
zHak>R4XUj3Zt3()R!CM8hL~itp@_^8ss`q<RVJN}E_k{C=@xbovr$dAoH?c;$x?oX
ztSn$gOH|RQEnuk3di@DUO`|L{GNtRgza-zy?$|0`<?es_w#m(K3gD;nxt+_5b)?+}
z2=Ar=!h1^UU0)RqxpBS4Hb<o4waPs!@?+(A{U!EegT0Qv;8@r7=7Q|X(Pg9X@C4Gn
z_MAEx58jK*MAj?8H98-=C5q)4kGIssj5(R}Z>n0*vY}#0#+1NkYeQ+up|!bl^p~4P
z&c%GIt1rW#he@}WcbJ1iy|J^6EEa66>Gu@U#YWqd$^F=gmOp4`{%YpR(A~681SqtD
z<4vuoj)H4KdiC>X4vNd)36lnJ@y4G3m;L`Kj<5c1JwS9!wT)1`Y~FP9*Dd6M+T!m{
z>fMr1oSW6cwHKAbdo6bMjm7G}*1~cR4`!x^Z6*lWrYmIAA9_cTboGqI-GTRv+l{%K
zhw0v~#69FxB%vHCSkZSq>t8C{<rg@`3yXX{gnydt1yC)G+%a2a4=9a^`oONw`mox0
z^TPxF<5bMRv&`&kWykWXxVT1^qGQ7LoL|3VjqJ2%$XaT8Wcllr1we=~Pji$(U&fWj
zT}bpu5-1*8iS1jYUaY1N(#7`0(}{8V9IMB0-r6pMqp5?Fmc8zzz~<7}$q}<|a0TFi
zn?y#3IgU>*jA?Cz!k&<sh%bg~;QDg9ksUkro-$0wPxoP*HG!9K!rH1}1FxS|nZ6uv
zYKgZGP6|6oTiFCm($~~H+{PNV&{;ukY7XGQ;C?ytuLs6$30yE1nn<dBTM_E6<ds7>
zPvO+5dRGO$uDa3pTO@;OD~oyfuU!9O_wjpIHkU~bGCThEhM5>(v~8L_K3kVrd_D70
zVpCj7%#>}glP3Ks@kYB+O=Aj`vZgEf6IZ5Ue#LiWbqqcM0p_n!TYYz0?7z1LE{J^e
z26hHw*33V|v;@T~v3&L)$2q=}DtXEQ&<t-QgZ^c0H>OVlDM}?*m^MLWP5YY!sm4nJ
z?M8<F?`EPWoe`CMa<<GNrFIWdzh@gvXBNC_W4%wp!v5>wy5Yu#dzd<&Ttp{`22>-R
zOb2Z1qcMWl)a$x&<#wBN)y^{=f-Q}0@c|eEu|x@`-#VM>n=#ou_<UEKII&Z0_|8RB
z*71y}sn^@&7j4wl7jB!&+cpW)95ISTGaX3xpdHLgv<;Knz@e+5L|#khok0qrpTS&_
zhjy(!U0a)E)y}7ei`O<zf-<!1y%g_Sza>qx$#ihaC@dE;y=%2&6z9fWZGrK0aBX$a
z_+8Bbkc^stN|%1e-M*N7QXp2&#qhM02Grg*`=W<LD&~~~zHC!>J^3a0p6O>7#b5<@
z&8RxtLIwA9KyNLk%p~q!6+3k!3$WG+&Q(Xt`P_$bwwPH|zfw3V6}@d@bc6Ult=+XG
z-P~Te*WCGGLw%#+VLH^5l-_S`RryC#7gVmf)EzKV=z0*W#Q?9Q-RnPHRDQz3W_vn6
z^}e`#VsfC>8_spC#T|#UYjzO7)RQ8yIlAb>J%xDk0sSu9!l3u{H_YR>*3H3^%8*@l
zruM(vZh9Wv*!D0}$CdF2)kLE++Z|M4y&b>}m1&(93Zk&1?oE1{*X)mRfs^c@hQJD~
zd)7AE+sYY6<zx9{JnB)8ZT>u5<yiETaPvkoZHu{fCq}k;R<v$j+fXMODJ@fvK0e-+
zId-j)+ooN)KfBY>*R^C2Iv={-vAZp$XAEW92oJz4EmM8jH@(!xX<?3hJnwvLZ?tdC
zPRBHb5wy`*u;#koI)1RY-Z!I5C*}3+z>$$?$CImHZuy}KwS-JX3jhN?+e`gNF~HHl
z=k-@HAbF1T1BH3NX0sKIYb1?yOo3`dgYL)|XLTb=ELYwGMpu<`xy$Ujk((Eg=*+d#
zsJ;O%McwQSl9E3W?nn)quZqLbl3FFjC8Bcsdv`h<too48E@LPm{nS_D{q~JFO#^0|
zi$Fldzu#t#_NMN6m_(@3o8EXMF1nMb^;?J6l}`V3&2SCFnZV-RHaO>km)#A^+lukg
z$dPmT$gR6zS8nwOb;&RhfPo<3Yij(}R<Ty;`I-+cxkRGDHt>Br*E5w1)}D?}hn~;|
z9S0tqShtK}ngc{^WSEG<0w_j<FVgfj$-8<6k68~=w0QSUta}s5H?KQ5w-fFC$1yGQ
zmx(1|CmjV`ZKxrSx~(PI^EPt|qZ9wxaT_Oh_T^TwGo;>gB1gr@j`6EK|H-k*=N+`T
z?qf+}x~@A0qVaZxhf`uytx00j(HKq?UE6%3^}Rk_V&<TowU)Jhl0O<JjO7)dZ60-4
z?*R;23Eh9H4nP3%*X9+t^Wf?oB=mBk_qtt9UO_`8biWmw;a%s1S9z|he@1>%N|2rK
zIt74AwnHYQkP0WI8R=GNON(9;@VTvmb<T#Kh;bT5W<Y1tXoTYKS}@6ERwU|<E*>P_
zb!Tk76=qigE5LtWxVec6yh)}~Q!G%g9dV&*pg;J<)u+l=Iy>4uNl74k7DHn9Tf!90
zYkFXr$7uDWxaxSaiF`B>99v`^`SO#(jn)?hQTk$d<qDO;i=VU-&Opg6UliM%U#PgI
zo^9(Zv7!z}jy1K+V{8ShlYoCH)GD&HsUSHx*teDps!>;J4>Q2*(q_Ub9S^KqE{_l6
z<tCfRWLZsuJ!@#<o6Vgrqr=Rb=WFR&+`5~dV_Mg2B){$%z$$i#IGtl>)Z?9xISWPl
zI)sJW%t>bl?dt_;6sx0=`$D}<Ld2v#{um^DYT>*>D^x78i<7pGi*$CQt#~ELxUcH~
zD9vsD|5%Jt>a+Ai6*4eCv1-4*W^GI%k3Wr7XBb_VE3i+6efl+(QO+;-KJHuj+Cm})
z5UI8-T)6U0I1j@he4bLjx=So}6p_ieQFs5K)I>I&8;83X{)RM6C-m0vWe6aI>cYAJ
z9|MrkQuGq&;c`U&3l{BpCzfA8O<nyCE5m__hkQzju<@qm>75VhXX<xAv3Q`t^}cvG
z!cR-}mp4c8{<yaJHsb7+DC{GYsXfW)?=Rnwk`CvMURtmAbS@<(>dwPv4~(^_h_&^z
zpiIR_CFwrFR}1!cAsE^V1`o)|o?UkZQxlU=3~DoSQIL=u^W<?9&F64|UjK`Vyz5PK
zp_Xl+1I6H0>s`hZwuv2iPji?wHlMYruT;31wN+?yVMuJ`i>gTl*hTtmTxwKS{k3fz
z8?6+N4o$7p0TV!Gn#ic2eEqA$Iy<OmG^QjZP@)>RUTya-t~i`<+6cOTu`M5yV3+tD
zPI|_rSQ+PY;KYTIrV<SlRM-a)sB|3v9M_@&H;`Zq%BYisy9kfgn^}($&@ynRGbM3h
zGwbCE#m)Zd+7ZuBBDnz~EA753tlI=BY&#VyxGqKsANPs?d1(q&c|e-%>G?;taqZM1
z%lRmzf)3#Y7k$u%#%$Gt#Bacq;6i~Nh-3ZtVoldabinz9Of+sKAlnl$NOrEoq@q^L
zpI1b^>ACRHFHOE=IMUQ1sCK<OTL`gL2W*nO5!iUL78V{Ep7KmlTAMg~azak^Eq*L=
z?p`(+znm)}=f2@V)>a!=CdJYGUK?s$;%zB7(DpAT%CR1=_~~)GUQ+v<n3AYK<1c<F
z(M#tI2?P<e?OoEIz!Mc{`<AqzpF{WTZcfEx+&4)u6rQ}n80WTa;IY?7@A-(<*=^E}
z4#${Ifw|$B)nMY01W~KtuHqF;pN;0!MKNI2c<iHi%tg8I&S|X4MRY3EMpO>?|1ghM
zHVi~{F7fCtoT-^eFxRva3UNp@?N$Ha0q3@2f{|k|&_OrPPjUQTFJ=6YlsJD5`8nP`
zx-VuREdRZChv{?>QEY*032odtI`8qNSi%rc${mCi*>q3y6kIO*gj24<gg5F?=Nc>J
zLq-6HcB0{z4}VO|{s(U?1il(qGL?$<4oSAiwOjyz7jd)&GB^KylymO%IT}`6P+t^#
zwFxk2ErouneYMDYugdKFMNh}5lbssIa$pFAg4h;NX%UR2HgHc`C*RYe1d@Qd@N6qv
zKHuwWhvLK@zNVDOt7zGtqIax6v53drY_83mr_S`R&m=;+UWx)##F3wXsUeor;4ig2
z?aE1spr(l3!lae?KCLoKyaTn-Je|@nTNBfMC4l>-MUWUnb`d6k?=X-`l^Ijx<4n?%
zD5h2Gax@SJSS!F1+ZdNqF|tf-rlbTGbuLHKV2W~<&r`@LK|SU#^p^3C`$7tBgMn8S
zd0OS{HDTsaw<2*Dz*tf0w71aDA(m@gbuBlc`z5jR+>IN$Ewf8DI*EBnE3V+5UES6P
zkUCI1<`=aR7qvbPLY4JCzmqHqU0hD1-B?_wNT!2V0FcL<dejx~=cx+Ldcfa9vGN_z
zvr+uhW7hx(F5*8nUZ>)#SX**qis^+QF%e6eWkwNS*&nCR5?@8IZ`&k-yCXji>}e+)
zmM5)+<8G@QOd{9PkYCrM#7xY_8gnP!w`NMcAo{8Hi(8A9>{snDQ=U~fk{+n|K+?p{
ze=PJ*JTlVbS~T%s-ILU@<<9xS`S^AmEmh{*!$NTs#gn(`h{#qwe#J&;B~e3o0rhp5
z0HSwsKIx5(&`b96F`|W-f;#(1({&uXop%{+OSA~mC9*Hpn!wV#yCTdaXS1n<p4!LG
zJ<qCU?`yBn3iY(hM(!h#FP7HyqKK4F3#N(?Mvi>Q^}%wF3P2ckSpSu|4v>2Bj}%7!
zBpH3FU$om)5GDI~o-&#bcWP{Lz*Z8w*en)a3Rf8dV7_zZnWCY#2PF%KyKU1{F>~bA
zZRX7<oDRwPR+D9HAo8hpfnNF!<h#i$W(Z1QTOoE);NuYU)f}$GU2eh4-6jXX%40EB
zU;sd5Rqk%AwVH=9UIiEB>DFrA!s8@pEoW+kRxLVgp9?7jvpPRtdTdu%^?Q5yC{?2M
z>Ylslz^;AGv?z*tr0YR-+nHSpU8@*X?}^#7$%^R(-uRW?glxM$G4)$_@Nl`+Ea0$r
zc0noD-d^+mJ1QIwT=PiJyh$JALgDmXginI>c%)7gYgI;%>or$^HAm)=Adu8tE0uX&
z>lr(JRit1xsot~mDEUj1so7=d(S&BCs&n1|8LJAV;O;XEj<%XT-F4DYf#ur!|F55J
z0eKI#n!Ji|?o@Zm@OVv@Z7Y>ZxImw3&wgQ+hUoj@jTvQ!Gje~UGqvBB%<2Z9lEwv1
z(Pe!|3g0lAT}7AvAFqCV)@rCL$zWTjW0962%;kW%YAI3+OMe4EtXcbGoF0_~O0;dG
z0-Km(T*buaL1O5hvx61fU5})wo_ti2To;x5GO(ymlL>p}{%w$joT(#`WY3Uc_TWph
zjQgIwn6~TAnoYggpHvc%d|%U`{4(i9+h^2{vyH=XK?~&#d<3qVU$y>1>~zj^er6iO
zdllovd6Jk|jiemwnj_1^>tE|n@eC4LrYyCZ)t#<(9QD0V-{@o~#;VYDU;zCB_~}RW
z+-vh{_TTAP|HiJ`ECNWb04`yp9Fk&qG#iGD$5qd_0vd>KLK^v}w8tfe0;Y;hNnvL$
zHNYf}noeS!xPNhGvi(Xdr}}_h0u#V<2zUsfaQ^Sk@n5<Dzmt<P5Kh3!fQtoz0`LJm
z?9432d>tIJ*~zjuwG_1(^$L}edOJ4`2?I+X5@t8vCrs}NCmcFGrg68D=Bu{*ndh`l
zG%aLX#>g?5%*Hsn+f~D&XuuL}<1oX>t5SEGF3zl^Jnd(r^Q|B)=~m1b1*BX4n^G`4
zc;zcT@4+UbyfPo(-#DUHr3@OiOrp}$?DK(Vw>)e6Kt=WFx-#qhN|z6_t#Y(E8a`W~
zROy${`2nFC)Aruj#^T}c#`%}<SNI1R|7d?{{hl7BlDG&?B{)lsa5k(?SmX~ojOt{r
zBqG_Fzc(LPYdZ%G+5$td6~R2xvnNI*Z9ItQ`IHqeZG>$S6;7Scn;B@-8^sbYkeakf
z*`p%IOvImQBU$CnKrMxG6~d%G7V*UU#+9o8yqUTI8Bel#h7?uoYI26xGY$slE3_<F
zdrJpV|7s!@fr9wA#o~PeshnGFnIjV~T28ov=_GXPFQ#|whm@tw8_?jFZ3Zm45L%Lz
zJ%i;Qz&0l{s%<9^dK-QO=vj+VHy5)Tfcn1obH7)5*seliWK>^L0!H*Mh8h@>2v+OK
z*}%6?!V`HuZ&S4S?_jK325cO0V{K+-GB5@jWp2wcrt3@+XWv{-wri17E%TKN3UNUC
zGzNf}S~i4J-z6-%#f(@wNNE*o@!EA^)pBO~13CsR;U6pGA7_rBfbadC&Yumbgp0+d
zj}v8aL7xW|qa?&+PJ(pP(nY#dHO)D!j+s;=vO5`bw0l7rTo?YQ5s3)IUNbx(<^SI~
zn6qe2;NWLj5+WS4OjbS)*G$T;ACV4EoiM1*UUPIhoaPicy6KLW-DIsJ>(gA@SXH@}
zgj92qGCM6b0H|nf+%O|9F;Qocdzk0=+wy<TmZVJThJsxFkR*uJv`My-GAR&}r{x2P
z^{T7mNj``54?XchDgZTz0H0wkdCusUp)*^u2mh-Sgfsm<J@vkJg>{NaPwd0!UFL#1
z<H3iz(Y%FkTwEPZV;19u>|W6>;+bJ$3c`K%xuyRo)+DCA)ObrOus^Q@n}HfFC7wWc
zoEyKBuOS~L3Y>T$2vxr9FTHa*$2(JD%z=4HCTb^KjYsuj@!p(|)_>d>$$cum_Fdu{
zh$e|?vy<X5sAP`RZ48y*POcdmZc>q6_ZPYLa$8Ao{L&SFh*xZbY1L=AY^12W7g=<D
zi*WoTdvR}cV|&xV+PNX%;qfm<#M^|obV3gj_X}qu=~(M-4P>X7;86fMo@zdr;Vrrs
zd|=C{7&-j6Pan2?OCJZ!KQxPCZs+!Y6X^kT<ip6n46!*PzZRE^*?cD`$`t!9wx4n{
zfa2tSrPkiP@djS5A4YqWIa5Io*=Z_CXj@u$!*ibq;g=#`BrRt}JfqtgY)<bZrDKH3
z=`Mb&Vd2>eq3;&_;hH8ND^89ld(-?eum*LBq1zZ%*CRsswTa5nHZvsvGa%Q0&VcZq
z0!NX#-4c1AhKO!?V5NrMhI35ss%@5ndu2NN+Sw<a;S=42Q)?!g!Mr5RFO=!oNzpm9
zD&!VDBQ!5e7q}~nrpmr5{uok-t&IJPAuP;7=s9G(%|VvCdAMmSaQFY}tC(^BCx+)W
zPAs}{osIPaPj2Q%VI2lB>z^ODPt}*U&r@w3Q7_MqlT;lhZ+^snxluo>zEMdJF8lDg
z5t5}Vpb{^orHfc9J_|(@Q<DMt+Qrb&e=K3IuB2)!GU*!k;yWK|6sPTFN?iRC2MX@S
ztr6=Q0KBS5Q`@{5Q}<aQ@r8s>0Y{baT3Xmpx-7-C`_0vPr+VE3^c~sfU<mVYu-I{Q
z-BGR;HhMTmqd0l<dR1BizsttC&)Jy;_kx83aPI69n*?WEIBbq7vuTwL(FcpBUTjx$
zG+WKoIFDhZok7p9^U{cv4*~c!#W4~~A-Wz!;njryl`r71Fhn_#R8drU_HtRk^C-ZY
zc`{L8e|5IL9(9X)tOc2Lu&foAP_j;cDxW4N=H7Q{w)J+#F|OHdBzQ@Tu1*v~ezh$K
zkE}8fGufdA{Z-M+yaU3E;y^GE<+%QBw<NVe4beDu`Ub$W-qFG{^Z*nT=ndFyCCR<+
zy6@&p;v5J!$o*SE{Ye&Jiw($*JY^^C7BuKvy2XklxE+rRACDO|CBi2>n@}LhG_ce8
zVA3^N=f2eqAV#adIk{Q)*Sk!O<uy;;L7h}rleqcC9GXKqWh?8COY^a*`hzMU)}ZXe
zKe8OPyl`V+IM5x>Wxv{Aa2Z+Gt}Pd1hU^UVCeioFYXD0g-e0%9wbcn-0T3fX+z)5E
z?>k}Dx4&CrpN$alZIA|<8u`>60LgpVzNeT7uOsR=8RtEM@$;B;--QL1J(rj`E*VrH
za0|SCsrBdPZCnXul>2ED-QjRPijVq53kf-#1<*mY_dBk64tXBeFZi)G!kd%C`V=Lu
z>{qMdylyO_Z_b$gyZnH-oBx@Qf=5C;GHM@$^OTu)gg;t0nh%th#tWFl4QiI&`^7b9
z%C*@$)<m`tYUM2NZSpxjgJRZAxt`kX2m98MYQp%DWy0W+YpLrY)5o*+>5pe`6lq3K
z$qKbr0jL@0Di?XZAL);k2Z|yi<+aslj~_h5e>HQQn#IP%O_JHWoua_y99`vQ*8=$Y
zE~Nfb6O2n>7Wz=)tEzElsri*=F1+RqLpq{!QGUM6*<Z-7&u=l?4nOHphq`9dNQBUv
zT<vjfZ(9PXchhxDV)o%#JC`J?r8fb4?pax$h2Wkn5%F}(3wM=_rp$=iL9}UR*9V>}
zXVsN=Y^a$1FS)%@FQRe_im238ne|l75k-439K}M8QQihicV&f(=gG9{Y^aaDW4o0I
zm37xVMSad3Ns4d+KxUwq%#G+yfW6EFe_bGj_1vmzEdxXgl4Gvjtc4iwZ#SN|ijY}Q
zbspWaX$Be|cyklCgOUdlb!l^cGG8;VfgqipA|pBOSEKGkFn>hYBDnto$nu~9J;N8p
zuZP@+BG7LeNDnB~c*u4~b&+hFJ0jZ^m{-m+KDH^odVsMPtY6@GPahWJ<XEs6R^lsN
z*-ru(>>t9qZVjfVx6Wgusin1|z@)Rig|n3;5;Vb_wAfF`h+)HoH-#=GnqZ&7YT>(t
zZH6nFq_ew7i*ncQXjpw?-iMC(Nx&`$vgr0llERz&siU!ixZ|vwBM7}Jh{l5LyI;1s
z-zku+-OF^NGP1eZ6&5^@dsM<TM=8ZV<ayn|BXI}hb@~#E#$>L+BP9s^i^FYAcK#Z0
zvk<`^XtbFlQHQV50-w4`@DFiOcR){U+%EUpUe~83*0u2hSm~>W*4_?@>DA#BR0GF{
z8s+%UME2?BPcrKlj)6)g5e^2)(?NbN9C!2d-642Z_3Tp%3CAT2BAC&$hCCz!LEU`q
zzXb=d2yCrjzOuUc7rwl~vaIr^iYZ>Cl!z2H>PVO9Z{@uBex|$x_`Rpa7*D}oym-<z
z9PVdbDWBk<<*2m9*?FQKqkt>z=|8P9o;gn;@<=}KLA*1mxHUQQzDBBI>ATi30==<C
z&ELMx8_iKODhG0h@?du^XjEudsQv0Di*aDsI64Q`^K+9YYg$+Jzet4>a*B0pKTsT=
z2KA!0aroADD}pL=qlhg;-uG<NWG30WXwhUsN6z1qB}20Bj`)NGvfkoO1^&}KLfi>P
znysRXhyc{&xx-m4{-xQ_ADynHJ?Y+++@=le)pqGw6vP6w8L^U08|~y@_Bas~gzBH2
zEQUp`eX}NSeD=4*e^9G+PfE~$O!Hp<&Mj&(4X->H%&Oc_$#R*K?L0aEo<&YrAnNne
zWWGPdy%l89o98SmV(Pp|5nOQbG)udGb_zEezZ^jtkvsMg#`{<;kh3i+W?%!x6UlQ~
zM~$!cvh`Y^ZZ8n@tnm4Oqw?<bKv(E%G&X4W))yTqn-dRLy^D)(XoM&=J!%$ffOv-&
z-1?jMG+y}r^)BuKF#$Eq_|!VAxbios2apuH<3AyiOc45Z=p>DWRoDp*0Eg-F<Ho9F
z$l-O0k1#iokgH75be}&^fqnpiVHB8$Ldj;3efX;YL||1ig|(zdrYPK8rVHZ~V<eS_
zCXVbi>}ax;JOhn@Y+>9DgWzWkzApVM;1BT3kcaby0q=1duwWa=*^9(mqjE@a=Ig5h
z>-)V?e~N0zq((q@U<C6ZUNon%bls_QOeNz5{%5E+4rSpX`8`j`@8@CSAoBiz8Z)MK
z4_$BUP&&%~TxO=F)O@E&q^`|O2|b#ui&vj&&RaE@l^a)X&j{P?X^^zAjh0}48V$G#
z&6X2A<S8yQ3z{3ehRYO+4sx*2Ie@(AS<!;qL*Krn<o5@ausJ`$orn|##jesp=}n;1
zt_M8^P_SHWKpLM}e{umd;fv&s5h0i#+yc@-@jeyhehzhL#Uir4^V8^I`gMTf`5oMA
z{T`F<f%4VYId?&>sN>=!qCLP#j+sGV1LQ@F<Uelq;?a*^^z*9QambbNk*70SsbhAZ
z?(PPb#TTi@G%`tc=T0=}^TCD$t>y*6{6^QR8A`b`rSILM2!~1`b(y!In+%=U_3*qN
z`=;H;_Bwr;_8;QF!db&lXpoai2?L)?P!pdrbYrUCbdBT<evGr0mH&LhoO%s>_!9s9
z;oq_=fO+-N97LXMFyM>YV#$|j95%h0w_379X3Li1EzjD6?ZzJf*ZwBhf+-NC?e0?>
z?ttOQV%#<Oa)a%EVZQYFhC{XFS|FnQ{00yZ{kgV#!I!Kty`A*I?&3y~kLFr~@5KFO
z#Flub-{j5fapK6?E1&4el&re8)4HaJ$_pjcDF@%IO(Dmd`<*k$pWYW7$ZN%}wmike
zoT+sxMv=$Ws!LhCF%d&xv&Qzg6S__o*jI$3IBo;;1oY3MK6&JDs^z|%X!IdoTZAL?
z=4a+Y=GL^p^t;&+T6{ESSm&TvTHNHbm_AoVXqRzi#>z__5aG?v%~d_fmD%orS!(kS
z(2re_eq*8M<j@ANIrt6u8I;wSiWy(n7S@6LHgy*g{?=!AQI=9~|L&>*9y8*Z`xpW{
zj9fq1r6B%$dhw#7uvg1FoYjU$zTRXFIB?tw5b7tvIjHEM>c?a2^%u!l)^Eh2d)R_S
zmk%tcmX5fYy1c2s2X{S^7&sCKwJojqNthcP0a%HTA^B}B!3-VHojaDw;#21C6I86!
z<hD@v_NiP?uf0PtMGl@%-w@yF7_YBpZ^YpJ8ofl(OSl9xJ6=b&jcKo|63Hc8`V|W_
zjpUbouo@WV^&1%V_d{M2<AJ8wCYmKt?GNVKf<b8Q{U74+WQ{7>SzJpXH;1K!LV^;M
zMI}%oG9m?IpZ~V9lortbdlIKL&=JVn?P9Q_Jrn9Tp3*;BNj&lEx;J+X@Ge&Mu9x(%
z?&7?mcU;(?`5k+6OARqt<8(_941Y>{!p?E|@6##Z%B}{c^5gCGX6;Hp?ROk|Oh(3{
ztCVq<(x{ZM4TDKCGAxr3*ZZaOaD(s5u-UW(K>BvvJ#o+sJ{T&(2z@-?9JN1eavu;1
zmdGF6rc36S=9lv+rOl$BARJY83>c9i3Ft3Rq|KeJ<yEd<6;&lA|D44(B(2!~^?zRn
z;F6f7>c?=aYot4D2z=}qz@DgJ{jr}@s_HOOq0LumtJ=4nnI$5TzfO$9FXCGDdFmQC
z!y(rK!zDc1yDX}cZgmOZ-s79Rb&(tS2=a`a>H4_Y`gdD5lB$lE#oV8+HH09sXA_80
zzAb2COW*9ChW|-YXRKCYNuXV%+Q@>s5AWN9*$DTNZ``aI;~NcV#1V1CWPE*b&aQ{!
zxju9jK67th@@4C}#}Tfq_9gpaC7AVbCo8CTR<)UdP`xq5*r??<w{QA`g1gqQze%07
zx-?VEJ3+sIxEc*UJs01GZ-PqTJZ>?}0fxPGvM^@9N>ADxQ4sS!3~X|Z1qYFXV?o5_
zs8C`*J`mm$FG)W)Z|0tGT!u%iPv`oUzS7Nk*<wMDjsMsSS+^K^Q`e$EGV3cx$vPlz
z_znmQ+IEK;j8hnIIQ-@d`W@69tqScx(?k2h@pikeMI&+OjT!{VCDwm?a3<0ru$n{p
zlkfGfEqQ_2;KS{~4AqN1i0{@z_Zu#o=&m^ri4LC!OFsfpUrL-*_F*S4?dV1Jg_3s_
zn!jErK^vLBusx|+W!*G=PCQ4GO(UCeb|1~(n&s;dA`Iy^S84oS&SPW15eAVz=HGWF
zJ7nGI^MZ62k&%admHW=-kfF^f!CbHz&adyxW^;P~FL8O~eo)$t{!#{U06}DKQB6!0
z2w$T}_9mT9nQ=cuaX<p>>y_E}fKF+n;gMKK5!U)tYRA6zxgfXwlhun*4Fi~t8wdZ{
zy(HHl0Ft@ebYDM8oUW^5=Npjk_AgWTN5cJ!1qqLk<~G7lK|BO;e}U(6@HWf|NV$?8
zzAVh@q(Hr7?b;eB+S)N6TbTLCx1^@?=LElW)g_dD+b$w++rg9Ft*(~X{3JBUQtYkt
zjIgF&okIn&x!4<@NByEDN6c=OM=%-(3-?k>o;P}KAB&`VCzZ5*1}BCPkpg1ElLT!5
z$@TA8BJo5L@_u6u=icS!PPqsmkR_R+i`?uW@0}`$H=CQkJW{h75w6FlQz<^0y<VF1
zvuZSWy?yl8@-SU?#^C6OgEu^ouSxGCJX(Pt#gD9N2~+(;JtAJzJ1R(k>edY(dhi7#
zJVx!txM7cHz=$a;qf0ew`<hL)cHd1^LsJNIX%TsBA)zFnPse<-eiG>ZGj<g9s55c{
zVSsb3J09pS|HW1FBU13KnK~A5&<_YjiZt9!{<h?Y_*FyjO<HimlL5s%#5@PO2O#M<
zcd#y%P+&365lTwQN9a4NO)tC=-v!a{VWALtMS|qBTdt9T#AE!@C_0IK6UBHAHtZ5G
zmMh+`K0UxO6N4t3LF(!)ga{;K!5V6O9>}{NETtrJIHxK>T0xJ{A(Un=v0RuSQ7{XL
zv_VxHw?2~?Bp)f93bEf9VT132Xxi~ZUXB|?F_6oHh#i`%T0}@1@b5wh4T;j@LHsb1
zt+$CX$!iD?a5|cHfJre&e-mO%{3C~2?hP`OU*c`^(VZuDscdse$A-O4Bq4Wvv7Q%3
z-P^?#{EfBYkW#;=qc{;z;`{pSrwBJ#8;mOOHC#%g1LK>E^wAx=GyrOJkb1ix9nNZ;
zuE{y~Qo}_1zGcodyI$T<0&+i<PJiz?F}ZCgemmA!;SD)~SfsMCe0@g#ca94lb7^s_
z*VPZ`bi&GB<qdfp8=AAJH?XV80uT@FHGeN@+IC+Esvvm*XVCpS3yj+O-DIn6a^j(R
z<fkwSL@J32adZKS!S&x{4Ms|Qi|NJuxnoK4kHGrN5nLe}9iece<BtylYvY{7ISfa&
z*?)urxkQ@WdVM|DaqK$|=>Q!jRP)#@gWy6pRf}o*7cL`VKOItj$iT721Xf%hRO#Rw
zRELPJbDh+uvE5KY(x#J61Ps{Zwyh5M=#8=tMp_J)4NSauJM^-Q{$Dpzt_5h5nPsVC
zBNNu*i6VmpHK<NnXsGQ{k+ZiVk80d|wJ>%5B;_*K?EQ%94+$*Ur0NEErKiK$nZHa{
z_5Q-<ncS9$*xJ>A#NA|N@!uHE?W<(fO-@Z-hwVdqY7Y(ubXwssB7HRuy8Nl$2!5-b
zlA`cmjMb{D<Da}u1+BO{v|?<5%+1KGBE#zar8qKT_ACRuln8IJpE$f4dEZ2Jz)O#c
zyJ$fkp@GBr=;rDWC+1Rd0e$RO9mXB3vY>A}=EY=W4R@_y3ALtrbNh+BD`eeA-$#RN
z-ujU8ervN;;2<EsjhzZZ-HS>R2lpR_y;Keb{}#3Ld*RIOem0;%+z)<!KZsX<{m$DK
zKgsLO@#n-yJa8<$|I%K4zm){$ZK&vi4o_1JEE2l!V+-L`ItD{&zw&k@u_8>?*c<N9
zfGnPau={c6K-vYHlpxTB7$QYH?X?J4a6GP)<W4wIGCpJ-<g`ds(Yzo#v57n^nv>QX
z!rgc-r!khlDjQh%0?W^g>9V$$Gg3Ea2t?Zuxq=IWv>PQX;odzb*itwvY;GLF1u33&
z;0paw6zuoihlP{kSk3@RUGp${kdU(y>cp`!lz7QRgUj4t*qHb!;TutC_K-K~jbIe3
zgAw%;M`PId+pY^@^dne1>TN651-FKpToPuE;rv17=`#*!VBKX5|3U)n2D~{a<z=e<
z1h*A`tMP$Bxk75(K5@zpcSZ;xKokvQv^jeTM(+U2$L&vk!{ioXdxg=1K6?k&UN0H7
zWp4N6bpI`0;4@A(SIObV;bGm%dyxKff>8s6AeQX8l#A)qETO*OE(Qz|8~dTEkBUly
zuZ7ph-5lnIg_k03W6*hye3a`4yNauRO-`Ag8zspIp8yfTX{vN?zXEl%NCpkciT)V4
zZ<)gE3MvB!?&socsMBO-&mIzK*5=AR%t*faad(`P>y{<ffdD8q|3)=j$DHYdd#v?i
zOJxy~&VqzyOQDik&~aJ?#~!Nw#l?{q5KGlzG@A-HQOWSocMbXX2Jsw|KVyF>W-re-
z5yDZr!&~C%vBx*+eC3m+2_Ev0gfo%x;EvF-KFa<}$F*yYER%LR)lUb!Aa1P|S`pS@
zaA%h2T;UYV<kylBYz&6bV7Jg0G_`XzkU%*E))M7RQG=rL7ly{GAo5UacyoanR)`aY
z3uV&dsnj*QJKTqy(BNhlFKq9n5RlliVRZVh&BCVl*!}kYj4-Y;GZ#%_UJb*y621p1
ziq?Yoe&UgA4c?F*r#0x=kCI|-3iO6Pn2~tC;4PTgH~4X<$kJ~bPN_Q1HLxovWAoRY
zhP}=G=bsvd=5gTQQ@|{z-A<7V6C|^5c-Hk=CisM$91g-o;*am(zLhFuA|8$Yw&SpO
z{>1d~E{LiDOW(JYIK+oe-zz;F&QOSUu5Jqd=GyNgs<{6|XQ0hRy1FQw7iaGDEb(2W
zFeLg47eB|Q{abo*oiu4Oo^PsRsCqn!%@~9*#1yXp0u(zn3lMVye#o-JiLkX|WyL^!
z2pE_zFv21^6KUQ2(i@>C>D{0nv5s*B%oWBmB8`5A<SXPI2sRTrCs0E|!%OI3Npw8E
zvNtI-aCX8|qj9#azNxR&%&hy^@u5ED*gN+k>ule1?vPD|&>Y<%Gc<3-?<64$|J^0y
zb~cX#dn+*~O)%l@#Qp>R`QKKF0WXJ@JTl}T&JT^hQ+dP9OupH~i)JaKM0_l_iq(=-
z^)53lTRRHmIc%I#IBy!8UB6WC9gGp{`@`p1_d|rSTqP>T?QkMMqsn6SmCp4LKmq&w
zQ(=QsTbk<U<$H@bm=)bmqtDy<0KG^_PahA6gc3RL8%{k+@6-nu*k1nsK4tl?=bP#r
z`OJgo4jUz|e;!0cZ3A#bt_G9NXz-u)M@WhfkE6NE8<}EWm5?A7)IxC@tkzbo{Vb2K
ze>BMy^U;izYk*=^X~fgJ6Osgy&&(mQN$-zzTxS+-Rps3~mbYyYgrc^#P$pmX3Adm@
z<4=Zt-}54T&g!s~Eayg&4sf`ehO4w^-YjSKKI?;5dN~J8E0t+H@JsC;kUQp0h3_l8
z8J{8TKRiYA9*-D2T(W+!(1aF5<Ky!;FyX{9FSu4)E5gF;S8=zA2bp&5l4YjbJsF}p
z#A2y^u>`TqS{q`{^}%oU9Ft{q+6f5mbxDI2;pLP~6+l-Rlb#G!4Foht_nEfAk)SW@
z@}fD-X%JAWA9lm<2`OFldt)L>@ZF+2VOMVL)CnN8l-u1l+xH7$0&3ES<T<Tguk+yZ
zA@{f1e@jA>mJ-@M1VtgA3-1nafOx~FH|9T4m9w6-$1K&OO>&~3tQn%8d+reaDVzCP
zf(H;MdPA8Ij2xZ!Ap!%($OV18J7$LWaVa6KKS|H;BX9~Jb07s9=L8$iZ%x63Y2E=e
zW-in@d6#z^UP`vd6?Bg`zaAO#=N`O1G>F2r1&P|7l3Yd7DcpUXU?xH=yWoRm3~qqv
zlTyH1Nw~u4{FyHTIv`YrlGreaCP1LP5dz7d=R@Sn&EAWj`v#kKuB&}G7jH<c!#;oz
zU*Ya5O2Bl(zZW^_Xw#;(jJ*ZaeLnwG_ZbsWnLX;|#cpRbtJ9-u(ET))W8-a;F+rC3
zAU7ZWtGui}z09g?@S$5>USxv8NQ{oc9}*mt9!dRCfXOHAobNliBQ#gbYXQfk_V#gw
z!bgdC=UaCCHf3>XO<KW2-yKQ%*Bk?eWIlae^&Ppg|1fD#W!I0?Ec-PS{sbj+)5bD5
z6!YzSloN364*XxP-OouT(xEE%6%urctYf}}t39&LRvuaYa9lq6!E{SAZ{05e=iMh%
zE{4v|z(M|HOs|z*uAk#Y)@Ef_c;(4k)#(FC)$$c|vb*XG%v1yZ{(?ny`buMoPFz=&
zl<xiKD@#?)GdPYvLqwl?#G~SsP%A_vebNhn6C#Jy8^6@@93D1~<dYEBYUx-2%@pnb
z)#xXB!uWa-Y-UhT)hX0v3ig11U+qHQX`q}VPX0z`=T-ig@Dzr(vip#@0L<KgmW-q{
z+yYwzw~EE!g>4w=(&QkPnPI7YGS$lKBw+rI9cHT}AmKosjE7FM1l>6{3l(_he)+lW
zFHi0u?@+N=_*w`FCL-(&I0K#oQbsdP-R;7wgkmFN=yx{Z88%eE9xt^%>tZQe1Sc<k
zP=o5F5iqw2w<fA}z7|D41QAg!^m&NFg)|0aMUUBvcn8^B(9$qbTP!n9M<X`Ka~0Gv
z)c|TJfes5~-43y!ZSdvCjt&)AI~)F;Jj;V~=88|Qoa5zM{rwB94*nYqN544-+5D8a
zzyj}sfX~gaM=nDi9_)!Ac@m&Xm3#-5ssMVK(+y(JgSt6rhydY}9`^lUUr75Kpb6+}
zI>2H6v(qbfGNq)D^VcDsaUfsJx%C3ntt)pP<htJb;M4h|JxYN^k}*dDA$`hSUj-R$
z%6n}C`o`@8md4U?SWe$rFrH#RhNv{a>-~6z_J&8xK15SgeQDA<zaX&irP0gN-y%LX
zUXsy*aCdV%3BvXbVil9cm$SE@{C^8tEs1E&Qmr{XI1NS@Be>XI1qOAshtoAn+#;)b
zcYe|lUwgB!T|G<~$Jhg2ALuk!daV@nary~=&il4einXYOnGtixpdq~S!{LZvIpt~;
z!O?PK>G_$zN{`!L>@LPxnnck*9wU4D(}ROYg&au+Rp`mhG5jbo4mr&m=4V#_yV#fx
znyevD%Ws9?hEh!?d;3Rzl$Buw$>(_TY2-M#+EL1;c<{!lvq>dXSTo15*rY|UsSo!)
z>{Q%IlMCMYbvg4|b$Z?(hxdjaE|C5CrljbEE?q7&J3Q&~XBMIHauEVM*?dhuw|g01
zp9jvKD>qNeYLPd!>{R`JCRudE%W2-rCURiH&Rh)582O+9w~OLT8%LJ5uotwt+IPA|
zWX8RNwXZazvEC+qqPzem-pqx3FxYyd?AS&At~ROD?=LwjDh^`aae_9aP_K`;hh#C}
zm%NG7;w{`HgTyX+{M9Xr%ZW22*#Dfc0gP@Bm4)h}nFkss2=1<MKFLAW6k`f|YP+bu
z_%qr*7snR{fOlwBn7|Qa^L}EHGW#-NPG$+@Nk^`6JK3CX-9qH+2`|XUf(b1Cw~xCh
z$pt3x`(ItqTM%D@oA>Flwq)Ts#pn-F{(2Y}Ym&DqG$2V3Mdx3=Z>B@SwA?%V1>5hV
z16(^iZ5cfnoqw|nMAPB+dlq=TXWCEM9wHitgn(l%mFD*nWClH0ul%b3i~=su)V79x
z20h*SYpR{V+343D7)1OR#e9cn8Mq(v9NgcW-hIj?1@C;9lv3lG8CJJ=_J_%t33Qi=
zViBA2&Rbh7wzEPYz9+W*p7RB*{NjegQ^oz|wnvMSi6yKC-$l#^%BkEz@)ONn3ws8O
z-g-czDE?3VyFSn;en|etSG*hZ0M%05QO#)jPD)`(6``qKCWTQ~=>MTYH|+4*Jxc|c
zYai&<!X+Lb^N8_b@Z`0Rsu($Re&(r=uU7>PI!x38$j9lsIv3r7Vtg)jGn!SBTAiJI
zP9vJNf(9{Fb%mNLk)H;lu^$g@1wM4KSwOnjb>KvN81rxM+`I2<9{m688F$ecr7A#6
zgu^Rf3D1|5C69TaZTN(w>&ZpS#}l6Wy`3sOBOtX;Lmqf>#ZJ6P!FVRqgFlgyopl1R
zn7xzikNvq>YX_`rrF61xR!XRDvCiK0^zN^fSqn-@yPi)w_0uELxi_hZ<{rSz*SrU<
zb6q!S*jzcrXTafqjLD8o9;uHbpI9`M43WK7rw-$m>!e{wSaO0lpUPps;{{=56XUb_
zvFqPUZolYcdv4>pteX=Jf<j+0r=!Kf-{&~PV@{_a2Ly;$g-Z=|;U}jdRFTLBQBEMK
zK3qoPIlm&#-nTv{jB^D2UQ0ej%2wz*BNFSdGhao`DbgU<jwG9PXH1*uTk0GiPNae}
z4jBl@GzULhrPt-p+vCx{%ld4Nkbvq0jt!<y>ln}gaf2ijJTGP9zUWu7C2{;7=zh4t
zpvDP<8sPan3=$2Y;AV0Y&wk~4`JH6x6)jpGO=$Z}e*GoVc?O7`=s>@jxO-2|^&Ho^
zFfF1AEG!Cj(7(hA{|Px0?hm*iqHfh_4v!H_i(h<~z|4e>|FSrJr{8aq+e8rpdvkqQ
zyFFd)J}3QMP!y4|2`p?vC_VvoDDc=uN!Doq@l9wyooy|tV!lsQvOYbqK$M=<c>Am|
z;Hf>?Icd>>i$Tn?+so_1hvC9cto98pzXM8{pl~0~#V(XwD>b>b@p7boXUBW(fj1_)
zR>3<<&;BWVL8che<WVv;wUSlu^Z;I;aC`rx1UEF-80Ci+bOuSgAg#9EGn{T#@d7a~
z<F~+tOC>9Zx=Vdb)@oNNSF8T1C@CYS!7D)yS<#o1Z;T88qaDgQKd(0nS}p7E2pvd9
zygqE>`sAP-oIO-@Kvk7|gI$Fh+8J{GqL|Az{S$km{7%8?QKr<o7qQXM!ZzOdh@I7%
zcZY0=6GlxVi(4jo!4RL+PnL-Eoq}rRizV*ite<%W_aeAU883rwUOrrI2pgsP;x`>l
zxq;7%aSHr)8)#x+3|CLq=+5O*0!(A3cZC=|ObV@ArXE2UHgq<}+w2H={k{E@HeNS3
zWX14Enq$r#=0##7M9P0=6v+)!7Z{-ol1#Q9d`QaPXaZkiCs$^JKrugx89J@Blh>ER
zREzE7KaTV&_2iXyu5?;NH9u-7b&jt+8KyoD;{c*h%$kGk*=VLtU)~p3q29s1gh~*T
z;9u|(`9?-PBqn*8*wSiO=ooJ|5WF%Lh#UaKbb~8lXhW-bIi&k#=sR9~e#EyXGNT2j
zO7NQ+>1f<P_bJ=pO9jkDIpi&LyIVjG+Ct4$qQ!f)4(r~Q#%9#wigWAs-c{ll0@Aj`
z-rB(_Dgyu3<VJ}EHY1g(Mh{sfHkVkJ)dTS8+B_;9HAWc%qff}lPmIMcnmSz09TfLl
z0tL|ytvvHi?BtAJSX*g(59S|f@rt>T?AYPVQxaT%0~~_1qCm=yu5Zw40>!d8?XLLU
zCyocEOe8Cir;LrQ+a0FNF`q9tyLtAQAJ_Nq>il@#e`t>pw08UU9_0MOLdb|XgVg$+
zq97i>sd}AFrRA_N5v8b<1^*|adfmzTd+a!|^M@~0Ucuk1hDlRFb{{IlE>%9$`?aRE
zvX){VycUa6Ukgvn%$AG_YpUzMFcD2N-!L_s&P#}`ul(@oLSc|0IB2AhhS02xDr?pu
zCNE_tZ#0AEVdtV83*KB|8Fh*%9@W0~UF8c$>yAluwu(0G1UGgpl4<>M7iT0S0Pq%7
zIs&msXM|7NszCBdDu(5)Fc}Rp<ni960c;Qf*L#MIX7n3yK`)24OqAM57nr%wsz!dE
z7t`;%0P)I5gvX`lg!6RIe@W`{VkqSs=Z2&_#35E@2NGBxixWRyu+@vBl{KDK%;yho
zsGy*%xBi$HQ9Ctbg-dF-%~E%<)gjU0Ad{Vkk^#Kbeq8&8Ge+Hp292-$j_|w4B86@g
zm@X9mJXEUkUkbDLznT&Q(n?Oq7>~2NfOp<B#ALaD*0jJdjuGsa`|H=@|Do%xqoRzW
zwr>>#QBqVShYsnGW)ML-6qJx|kQnJ07(ik`>28pejzKyU6d1Z|2&IQ+=z)Rn_KA0W
zZ#@6af;Eeq`<%1SKG#0`ceTm(oIx$ffF`1x1Q@6rdVd=|!BNrXlBw$#wP4xJ8@cfw
zDA`I-7wONlLdD7YpA)1HG*l9Hfo;k807(ciwE@hw2#<yy2n+aWyVwLiGFjYj`5DqP
zO!#_ic>D}#kGcd}GdJGsux;=0$eS(5LfQ3gR_=@Lpu1fxIwWhjI8TSjIBwBAA@0K`
z)X-lyk9eYHf*@c>sd`_+6b%2+wG~y`_ayO)y8eANnGfS~KHu0GI*fFBI$@cgOCz5^
zjPqbe*wT{%!{fo*)0>&u8=IYad}A6b&bldQ%??w^4d1={*`r_(#QQUUy%ZTqMP5nb
zN%UEic3C^(x#JK)#gd76?n$tdPMzir<0(eGiPmZGS8qQk7m4AY6OE>t{+(ml(A~`^
z`&5FF--L$e@f|zNOZm#ZoJ2jrhVCD~x3c_fhVblaA7uxN#&>|Fe2lyxJUf`r+YGX9
zuWaaAe|p=X-wRTAyWrZ}2U>4e{}Itzyz3kqjN~6eFx7GwnZ)QjM`fXPuCotLZAh&K
zYUdMzjm-iJL3v$^VyBC2DG=C0Y~N1{kk+OyY5Yeo)iAA#07QT6A@Ph<?8+=EEcjb0
zA+G1v*_re`xH$mi`cD(mykNn5=Z{Ao>pr>*qqo8nD|Y=VG9rPl3pu-u-$U7PNs%l=
z^J^{lHYu5z)*Xkxx1Z=&EsGBkD`-&#RQ~Qf*e1y+s<Fc=WyqBG!~jix;#qCAw>tr7
zQ-S4h&|p;lNzH5BX{)4)(h1CCwdZ~#FPuzjG@mCm9ytv-o;wnCluwg`1N%R*JrM(Q
z*CH(7>yBAvKuaH*Dmb7B(b6#CQ<eGEfzYIE%JRL#udy1y9W7sUcdp3XZF)O-5#cdL
zA3WLGrR={pdWbzL8yE4J)pg?9Nb&10elqf;k>Nx-oX(0Wi4C^?S$dNLDC7(8s}R6q
zS~o9&&r(!nUYInW0p9f1*4zAi{T6>$F1vr2+!PQ4CZ0m>xya&s7S!x3&4yQ3<~#PW
zm3SJaLW<y_6OA(vQnLoN$io<w*0$Sb>SH*5y~#;;r|UL%pc-kyUjZgkkOe=TrQ{y1
zQJ2*0eEda$Q7l>QSN<mmJ4q?F1s#}a*sZp<DBI$K%f(dpksLa^c=)!~`R|L7Ff_wp
zuom!a4@Q6L?<pv)qGaf{2hkrhwR?zc>4=6JLT-7v6d%j2X3a^7>Ln8M;AKv{@-;p0
zq<YD*AtIDqdkdV_x_oTGmt2G(Gm0Kvw-NymcjnZN>Yu^iWTG#pBWhmgT6Z$glX~~2
z(yCd`LRO6I`CHc1k{x~p0^JzqSwYGG8`oC7m5u{Iml9XoyiIB#LSZ)CfAX%Oo;&Qi
zf6o}vO*v^ZQu30O(`XhOC4&cO4R>66Rm^LPlR~>rRY{;;!Lak?j1Oa;)zNK*IYT~X
zd3&_&Z~N=t+UZd*jg{U_2Ljw9`TrQ4o!T=fB&C2>EmQm!6Oo9{buKZw3vjKMtrsY=
z$W3#_J#D%pB2gX!LRv_;f`+*oFP^X6VIi~n)e>?zaLJtf^=x6LqFw0tn!!-She~7>
zy#Gt)BzPN7L`GAvL+hcA(p(z1t4lQ+P?B>mM(K)74}wM29$W}Ux2Mo`aqXwYQD8KH
zendq7ff)cCy@it^NMG%FKE(aM;+3scm$aIOczOoq_CL$Fk8j06zBZIN;5sW63|$=D
zO)t(MGCz*INaf|ob18qmbbon-L9;7A-z`z1C-_K)S(EdqQ4lJrB4*1jZNj=Ux~lvk
zkR6YY8nnK+FyG<q=u!UH{RM5^9sOFr)8NehweKI$&!XtWjkZNAb&KY#EExhv9KmJk
zq=iejK0dv=HGF???y6F9Xs>*U<R*%5+anY(WTi6}J<wjb#%9q}TU9@SnC4;n?hhIx
zboez*iXbvlZ+Q0^DxY{ojfmx>6=)_TJM7IFKkH9l?k5T(aCXxLKXq`3x;+;|ITE?w
zrr{mdh>5gN4N$o=$l3(8T(%H$7fT^>ru#n3foG9x5Gh^JJ)q+1wf<Pw;{cDEkw8M~
zH%`nsp}S9|lcb4Jo+mxhHks>5f;mqIwr_0Qti3quJR}HcDn{kh-Mj(1%%*{B&^H+?
zzOKAeDa!)ALqQz4uAZtMx6dmDS6~)OtH=CVyzCss1<<~IAabe89=E3uZzQ@e(I$|}
zHkyxGF}UYOv>@PY`|jwSpT+;-$QsK@=&lg`He1lL`=+Dq?AAeP<J`>+1Oqi4hpLMx
zS}LV~k~w|YQlk*{CSk@<J+g-F)H!cd<hd^S@B5rjKDV)+R_a%Ce>1y{|D^us#jx$c
z&#sSNv#(xzc5DP8ys%jkOKD9<^IT$}jmQ^#2ELteX9;k61IP8(%CO+qNQ$vf`k32B
z;m;mt;3!)?{?2@8xBcx^bWG2wl9uh2lEr;S1sI+2^t;)MSNwLU8c0nGsbhmO#~5HR
z6V!XkI?ORu9jDC;ek&TH_w&+Xs<p@(S?QdXMg)Qrwd^xLK(u`O_dxy)S-9RDr|AsO
z`bB@e+V7P{)1LHLHL0Uf;hm0rP@Q1HA|gVEX7}(7Xl7~O?`M6-9}~l!UfS~Mk}k&v
zFd6IeMQPjNo|S{9z1Bt`ljd{v>5@U=(-7A!UXD3FP6eI3CAp5ffBGO%Fvj#&_j`G>
zwdbBpX33<%cJ;=py;-okdB|c)M#0VR<%qGSd53qZf5J3JzJ{|4ryt3~n`1cAVV?c)
zmF8Tz3+|)!1+_eM&_asN*z8wVEaq~lb%Lnb(kPIjsnqnXK1gf*N6!n;<^bj3Gy1>m
zL<9HHBb0+ah4P&*MJ~!C!;7j`RKGZo4{8z0i5SWm%x=1uR2vxx{UrL7IY;n!)m@DN
z6T(7F<{er*=4K6I2Uep2_Y#}-`+NPY7ui&xc58H52gfoYdo%nOyzq#nJ)GThdCLUM
zcY68>p_TSKv)eX;CjPbxzAFus|45w!^rx~RWzi2wFfY=(=nIWxYZtRD6@C?xi21vZ
zZMK2hE<dHk&IzEbd#fCGxqW9p`=`{u<J-4}F0{jqam&w7uM;Dtdc~%;Xq{F!i9CBr
zVrI4i+D4z?R&E(yX*Hm#E@iCnvb|;EkcsS8DF$QR$cq+-lZaoMS(2kg#SSKwmvPZg
z9>u0$yhb6)GB3YTP6%d6mI+j-@ZL6lhbC`v{|3M(gGc`Ee0vtawVAq4?D~OC&9=}G
z)P3`0#(~I$>|LJdZpx5_k|~(3Ie`@QU;%C$l^iJ;z6}0Ad!NH#WJG(hRVr99yKOVV
z)yJ%A5?fqbb`>dGz`{Sqg7*Pzn>WkQR7RTo7Y~HzpUZMHmanQbX1odUREAG<_-Kff
zG)zF)UBo@=%m4Rb`S#<jeux3PO$KB7KUWv_yV9$Z5|`Y6?Uh@N{{|zz@V9$|tKnkv
z=cDF_MNpJe!=UTWyHo>_+pnC7872hU{wQ^gdWoo*`Y^YS9gLZTn-&$=2n=1ZLH4r7
zhWkH^2bb0$O?)m$TCSVTPlvYm-wQ9vsVI<p_C>D1b$U?#5Lvro8NaaIk4KvIQ<Dxe
zTPYP5QlyKE*C*$*o=b*avztiBTtT-lt!&M#v#(<vuKuR09*b_6!*CEIJg$9Hvb+}%
zo|i}aGqjZ2N=deAIucBUgIr`95vv?-Zq0}8;G9paXH8uDS9g3$urkMnOipa+zuvt6
zj-Rx!{nB~N+%UqXV!K*3<{_3-OQaAbl&O`U%!vD5AU`iNdJN*My5Il3?X}TXlP>Z_
z?{TM&y~R`;m54H-^+g+$d~o(}fDW1n5h55E&H)RtelhrftSvmbMGNl0yXB_&0z}hr
zcQjK<8GiJ|4ucvYbFB}&iSm1IMC8>?ZHqseISee&``^sA9Gm(gKc=%mUN(DdC)+an
zNcEb%WPN>L{}D^ue?Bwa1o)m>*ZKg>pR4CPw->AuG%&D1ID;=|&(Ybt(4Hv8;Ddud
za$gCFqC%i_o-^hbN9@G4_R7BweU5yX=_{R*x>--ShUyg9E544V(-h6Yb-GbdoyG4p
zI^8oZD&3I*J=sqW6G<9j)iO62WvJ~-*D@$dQNwqTAz*`o_Pmo(YAeXn2|-(5Yod2t
zPf}?zOiBZbHOBm66SdgUi3pEIG>7>@k8nuOg)Yb(gkOvGXu{Aj&C(g6apTjG<-LM9
z?R{XmgCpB6|5_p#x>kTv5X#Vx?_bpFn@vnXi;8M~U+-?d@^4u8O!k0%KCWB>sO<9>
zygqMkjjk4k!K8k_YkUT+2=6Ozj*KpNfKNZB$&@ZV@)AC|+;37yh<(B^L#}b!HnMbH
z-OGX2k!<#s@oqU&vi&ejcdxj4vkUf+{ux8?GXi}ckFb%gSNYkXR2=RJD-$F>dH#80
z%5^J<Sq#*HcwoIQcCvQb{1vBaim{x(HG1r1;Bqg&ago<5ESKzfEu!f7ar1`^Wq7vT
z5I24N6Lf8DFjYb8x8iI!*-1Biw*fnt<bMw6TWoE@dej!+8)B8=++5=HeNxwFKmHl(
zW12YnH!Do^Hu&exj|qv;zYHpko86HdC=B}17ze`{b9SJ?1>?YWRY&$&uFTPZ@Wmfc
zW#*^;h<b7}8d5@Xie&hGlIWj=QL&_BEfht(Lgp`39p=3=u6~MVNM9R6y{<A&XEv2~
zD#i9|4V4x?cttk<cJ_DqW_<Z$(sPJpW5#2cxb|-q8N$4Zq-NWM+KyvBAl^yXj&R$@
z2eR$rqh+|C?%`DR^Ky2Cf{spPj@?fH2~K@&%inSGb>V8$u)a&eOMjozxvsK@wC0^=
z$Fyd%d5FyUJZDB~@APV7iV9^n>eu1^Y~zHv`uWF94H~g-cFSyi+x?h~)bKN?sQJ8p
zMj{4lF4ZR?`+GYpI(q4jPy`^sn;5-_Kc~AB44o%}o1JzT*jL?+BNfb-uQY)h**mn>
z{aqskem4p-SIbH|egDwZ6Qb)cpE6J>FJh1s#P>LP>DHSZX0gII516w`1{TY)VVd5D
zJ$;V}WGCK*{wAXjh=-+l*WqEI*$>Dy#zrC^50l^v4>Kv?Ro#7+SS18mG{)1(t|8})
z<BZq=!)#M(=dXswVs>bx7(?V^S`*pnJPBs*mx;8=tXj&6jWIdZl?R6qwSDIsOX`+2
z3Akszt_#0nC@3jJMCN?kZZ}ojv?c0d%8?V<;QwS-f9F}Ma^5qUBm`usz5ZxO3I#M#
z*aZ`i^kqa7sXYxjD9!y0rwxA4p3yXGld+~_Tk<Ex*wbLK3c0iXtgJj<Hmy@mzFAC9
z&YqcAU;JzGxU~U0w>i5TV31f}nEd`(fYgJ$P8o`1Cv;1eYw%3H)}G4YwS(6pz+Y?H
zVBdDr14-G9fOc%B_daBhVoRJ{R;KFLPBLb-Di=ql+WeeQR)4e{kAe*6lJ%S%42=*`
z70^NK)a7TBsiXM4`ja)o*}r(C2(Xx`MKGMK9tcEih0u(XmJ2?q+9l)q*|Y}K{T`Br
zwV3wNDe#5l<w|JqZf#IR-J)v?NfThbSC87wM(y`1ET2O6stxy>2c4U6^_dM)`^wG@
zOO_2q+Z9|pPAYrfc!rDl*Y<;Gh3PRpSxe!Cv%!R9Lmkie8sFhO3)`M3SS#w)Lg_7(
z^t!@G`il$u$izTP@U#D%3#TU}GTKyykjXb!SCBx1Lx)$KRliVzYVgg_;jY(I{^A!Z
z@UfiEx_5q=|C-Gb-Y;$*wOWi6%02nCkT4O=*e5mo`kp&_udDudmGBoB1O`nqb@mMk
z5;)g=N@U)IZDc|`z$Y@sSP(8?MZ*G=94&L|bLHoR`T1UnZ-kOYXf4=W2AYVB508ln
zk-j!n3sA{vJzInl{#ShWdlk%sNj~2ys~S=>1$ti2{66t;jC>%_37MeOXkz+@o6ho_
zv#-C#0t|M`7@1_1XOKpw5FoyPb80>8_FRtOSXb<pNs6W#^8>o=cLBq|Uc#G?eLR;!
z|5~&%yVD&i?fZEfjIx9N#4?lKwdhVL6LWK_k``MIW#wl(dT{#7Ee1srLt1Z2I)j?p
zYWV80-;(Fi=6H7-hQ2@b9YSSTx#X(SY%mSxo>Qf&h`)jwHuSg5O5#mbTke6!u(Ez7
z3r7=Krvq+WF<iShU`(R1=+$l1&DG!fAuh4mF{K5^vP_YI(;c^tEU!z1=YIZhk*jRV
z9xhYn+UObJ@>$SYAT4v&8l`+%oa5h}U1WsZouwEq6aqtrr7v4$aPw|x2Bb;i8sdpf
zZW)X$%>aW`Bh&=?n7&{2A1s=vG1Kd!L%7MFDnAsyDGYG<H|nR)WJabB$iwWksUAlc
z<_c|ZZQECbyRV-xBJk-K0ce1aS$=bPu6!EB5=Whq>paEqOhU;L6oV|@Z7R-D4gT{+
z)MX4mL&&;x(qEhnX_X=EqkXOZb|JBu)d<Dt;UVgw9pB6)`n>Qr54r&9Au8CUl`T12
zq`roVwEvT8rK+0pQ73+4vhcF#AVJ4Fq^?mjUgmd>wwOdyv50nik*4bOBZ&3_92sCi
zdSkxuE793CN8w5`<KS;(5U0-zWqD??2pico_yU}J#Yal76AC~)n241rKs`N2&Wly<
zEIdAm0q@?2&%b2|2mmSC_iA|VhY7R?a=N?|H}1dWZa+IZ!@w4Km`KvfuG$P8t_v@1
z+4|u3qWT&b*IeYN&=FKLu^O}gX6G=Z%acqubo8t=hv!l++xDM-4Ef+iBx8ETizH%;
zp18x<0n@8xDj(#|(#`c*8ixS(K9@FnA)L&fPW+=;Hj#A%5B|%L#Djw0sAQlkRwhAz
zQ4xXv;_$KV-zNU3PlPc6v<l#x1GwyJng-3&tlVUlSd*rK{I;?A)V?XR5^0;L6Oz-u
z8LC|^FM5Xz<Cek-Z+oq6Pqgb)n^uZ~NXwjSipO_rzTLR+b=9ddEq=qVreub=xk;xE
zsk6G`9^gu~+)BC6hu}%EiP;v)t({7$MMOU2sl7uEEQ8l;fPC9yYquV<1a&&FtU_HT
z%y4(ZC@XnlzBPn>zVcwT-#z=&P{?8>nqXZw(=RZ=Rp4I5X0HL4cK1%}x|JiA^`3p5
z&HcQlu7L2k7iDg<Qm2G?X5`TEszN!5bd*TaDzNLptHnQh=-1c`LwZpW(V50(0@~+W
zPSD*83Q^}yJAg7cpLuGWlFg!{W2|MK4jcv^wujEMt5~P&<`i(r1Q!+7rKeYbWeAH{
zy}&HLy^?5YiVI+VpEWjSQ?4!)U3Km+O-S<wSd=o!bfhPUutQ}oH^!6vj|T@;Z0V+w
z`!XL6=JbECt5kC-T0ig@)Jbr|RA~*jB@V4BsGmGcw$00K2+vM%9hOu<_nF#+zCoS2
zkZ0Fm=})fSN#jJX4R*#~ONZ+D;d}Mc_g{!VRke5dy|iTNc|}q9Rp|WkQeZLK;IKDk
z*}MoW)LYEZr={0<S+`bsF2r)KvI;ji`(4?3;kU^EmI+n**rUnd^YaF1wzAPKb_c65
z1tbR7JhrT?H3=OsD3I^2l+NhKVpT%izb#=~i9Ei2xC%MA6hu=hxLq@c+i<`BUC3e{
z!OiNnmZVn?!(rEkVm_A&h|nxWs8;vy0#U4Hw_yp)pFuC_Zs(c*aD^<b5eTEgqgN+Y
z0SnJY6aTF5N2I~4HM6YE<h$PZ<xr`l*<@I~Xa1{lUVtR@*q<9ofzRy8pQ*DViYsQS
z^qM{U6%eRGZi8voh0D`A27ldZz7!tI%cVnF+SB!{Q&^XHMmho4iTxk#9E{*OQp~ZV
zA5?1{!|<V>Ui62kgnu}_)PW$>eS+|Gub}zL6Z)Peadq4&>87;R;}+9CmbY(7z7tZV
z^9(2+LJY$W(Mir(#+(iglE0$l9wC$lh>wzm-NjoH*G@f&aG9idd8Qa#b_(W<bAQs{
z_jYa_%Vs|BZqo!vd$H`*UrMcrR)76zJ4nY#blb()15p=Di|(|(tq!fWWW+bQ>mqB5
z)Jm12hBxUtnPKP?l`&yH-0ArCm@r!$vRlE7R=L5YyNxM%yqMihhsK*DPn_Ofbrlg|
zOKz=trUCPF{ouRRFUy{eY!j4utt~0sKA5@p#qe2x{O9RZWnKjY6WZE>W?op}RZbzS
zK>kMk*jWhbRiG|EJ!W(trYi3l;gygk_q>c-Tt3L<7JulWo-deW01N>~UfIFH#Ge*Y
z9{=3wVvKyi*fy4rF9qZME&ue_ctc-hBbYd^^;^e*xxe>plBoiW2eO*60_?&N24N%)
z@n|i{_Or6~J0DiCb1$(R|0a3uo+%x&Dp3gL@X~vW<C5P;P7XWaVA{cYo&{L-0V$R6
zf234T3AtYgdCj}@Tz&krOGvmUfTzMCZ>wQ>NG7&eP(2V?Sm7t76BVkN0C7IuTaelx
zwyF<xZzLqk4JHGNQ)(o4#utr^e=olZpTBLhOp();F%W)1_%f-Es)SW|(e+4Cl7wI@
zgB`U=1ermSWji~VBH}zc>QX9x@YpQvP2rZaf$Ye6iiKi0JEh8wmhI+<L279cD?k#M
zPUGZC^P^SVx%LF=UAOCY(RI2q4*6af(~N|jo}qRjVzTxztykF^sp-9T&jMtEQ_%LY
z!n`DJ$NY~nPDtmj%V2PAZ*P}d4?Myo73n#<?!9x)qHiB~_tNms-_9`?21erOmvBTr
zPeX&+CQ5&CxK@jxXQr$(Nt>Iqe`oieM-4EgROccexxe3{ELL}_7lcIB=mD4CL7IL`
zMXcJ>vDhtpG%RX->qsoRc-%`%%h@ThXktl>)n463H^)yLO%H8wu8T@LuaBa?sf@yX
zEN_#z;DtJ2IBB0>@A~zfOHnS>j3Rg2kd{uaxL?$gz46%zez4M@@t;gJvB=eIOyb?n
z0b-l6Y4>>@*imuW*wLr7VWfw!%?s(#iFY<KU)I75Je~xA7${mZLK?%DeOfxr*!@Gf
z_N%IiKU8&G`=S7|W5)%L=|Md6PHi#2nbKpnnIyu{BN#?W7Od6y;M_rRe2MlAtWO+F
zAId}zqI%P)eSl}AhEw+Qz9Q=BtNLf%T)*InT6=8%;7+2RJ7;`I#-IMx1wa{T%a!ky
z<TrZ@TGdm9zVP0X<$r}c3+WeWl7+c!?OZ954@v~i#`4zBbUrb3tZ8k)fNsu}vPZ3`
zd4&`8OneIeArI07#`MYWv2M4c86~owh&_7_(Gx$nt@u9Ghe$8ZbX5i95`lr2m53op
zpBoPQ`&;f!y3cR%hvfgYQ$(=kU8uymCJ-=0P0b|sN%DG-#<ov{6SjJ{?RxfuBg{_g
zDtW$YOz%;7G-OjDY!PKA%^y6ZHtUyASyl5+gD2xoA8;a2=N<g*%L9k{-3zk!Rg{qL
ziyWj4yTfg$p*EZW6!ZOTERc19mqhy9d1FGPH%(4N$46+kfCpq<ikJ6&2~SK?aWD{n
zv|Q0)t*9;BZY(U=PWM0N(^F-%RJwF)jo#0m61IO)E2LQ&8fJ$*HR|lK=iekNmYdkh
zaCBr&=a2DY%ITCzDy%KZDcX0eR**!?jt{4NVsEHdv&44w|9(TcSdB8Fx4<)L@2vE{
zR;3lO2sFOze9^M=o4b7f3vLHvJMMNk8{~Q7hF7tqriY5+C#|vViBy(6-wb6EPK+H4
zE4iRF800bzw~=?1-_ns?N}Y3Z>mtsZ6><9T1lV5Z=VJ+qfH|cQc6B7U&4eC81$AyV
zylAa7eE3_(w-Lz~5DNI2Rj@r|&qJy?bU!55=p57Hirho*7IKnZXFmjJGyHc}apd!8
zG6PZN#?n?*{KBv9;rXt88rO|JUJ+M*t=fGbnbZADTl2{^a)na%f~aFj%fgEU8YEeb
z2kk!FZHZy*&222gwC)p(W_H}mlOX(9y7Te@uT{>?)ds@s8X%zve7V}v>*(VM!TiYm
zDewgeR?|%RcdB=TtE?$K{q~QUbYYGFs7Z4T04Yt~LeJP^GBX<Ra#;SRQ~BYR$-N$w
z3F+PR00m*0d40vC2piQHzvrT;@H5kDM2`4$^Xazp!sSU_Hcwg`Yg&RkdVEs1<pcfp
z&Fb@0EFt8*+lokrbjjin!*y!~wA-nEA}-kJg2lN$Yxtsy1a|d~RtDxuGgBahNa0w>
z=d<?FW=}6c^s*+%-Lt;n)S)M@z%ewwv1Lr(K}~8)0hwu;YuVj8UFW$6yFl|NkIF_+
z5!~Qi>$8UUFASU^DX*yvB>Qvu!`F8>`*(prG}^Fnp(tW7^nS_Aa*5iqs{KlpIUO`N
zVFj@e6`&YYSi=ZSTGgS^r}IS`@tx~3@ex-G$IGu#$8)=mg=M-m4F1sfRbO=LlPZlC
ze2tspE9N`kC+&_WtZh!expnLK@O0}m_)s%^=le_Ai`@p6G{3&W6_fXiV1Dm?okgaF
z@lwN;n12k8)0W2-Heb*M9vhiM?3wchCYYm0h>30K!sY1|s>XPy%i)RUGW>U|^!P){
zwkXL9quySyEOvbV_%M4gk|TBW(dhB@8cv{$r&h;#4b>ri1pB;yH=r<BfG^N)j6Y8n
zH1#Dg?R@O}9^CkBb!$J!*xBf@$st=|&N53R{F9w0_&>zImrsN3_{9HH9LjvwPt4M9
zM$sM=(HjkVoG35!ZQ!jdDb)MmC*e^<3NfD_T<z7mz%3gCeUttHPI^u?Tc&R!q>=)r
z$cW;;y#=+u3`ZGuwYIi2!Jljx9wM`^tM_fL5?wsE9U5knry)*dXZubve7heO_;>AP
zuoYHirr+yvew$T0qYa?%Z&nr&m!`0uXqw|jCzsQ#`ehgSijA7C-Pp41x}PyowGG`G
zwOkR5c4UCVWrlbEp#Ns%2%E1u9@%1@*W0>N=$~$Rv2dGhyNda`G{$r{#s3_8puP}U
zJHHqW9Jis_{0a$|B4TSgMoTnV1p55#nc=@!9lmJ=dttXG0E|piMmE{WSKhie+Sg3}
z1DuOjTjZ!6tW$YU(n1AviEnCqIlI2oTE?-dUCl5BtoD8y)>aP9ah^IEd+a#t;5+w3
z&&irf)=#ea?2$vM*4qX_73Re8;^{&W9oZaA{N+g^sQ2T-amkbdSSQ=3TOG1`5kJ61
z)+j76@rR*}*|%Pm>>NoVTt<UEMkh2B?bP%zT(}XGW6eF|5Kn6NwDq^d_G0C14jzMo
z5(QcOk4KeIfE!fC@Y~#gztt4kf2G}2Qm6(xE(RexfQ4l*VN;yxvKv|Upr2bt%0Rh^
z#Xw+7i1elJ&s%ADxPOSOjV?l;v_T=+u?;u7f?+j#T^&Ik7(nS;Z;LGdEV-E*ywiF#
z)W1UBO_vxdcLp$yk``3Hba@=Cuxj?z0g#Th$GQc@u5Q|3{vU+IPh)#tPFRka4}9X|
z6u!RP?+9p@C9HbW@s6Qe$dQei2hSIS?DxX-l=ADU<PwgpFT&|P^{=CpWBh8jzF?}|
zlb5XJ0{gm!Lzmv-8ys|SJxrw^GoQ^~`?noR%85|%;Ey`qeLJK5&0T2x^|v2JZ5cbI
zx!~w|e|^wmi@BaY1juE7WGQmzA<bU^GU$VWr`{s#{$RA{RN}P=+p<JO$)aNWSu6e7
z4`lsCAW4N1O-T7iE<(HtE7>-4sMN_QVz)++e~x!Jx^{&?)|xw8dcGMX^-EN8C0ao!
zr}{6;VvQF#Gu~EZR`yrMnc|}F5jLygV3UPJ(z{C|wKXw0N`?N>8e}7gHux#n{|{ji
zI66jeP4~gS>3DamofHqC?UXIQYnr^%=L&5J0Y?Xk;khs@&XoOGnRp2NNb03=xv$IH
ze;eAEv3AEyYWRAA$j1%xNIl2p5CvD7G5@C>g28F`XfIOdGw*T!s`BZ+?vc2ExuVnB
zUBu<Nk2;%y>|JEx&P$&={Z>mI-O1{;9#$7g*++x`^tUeGW3=-E_PE*^r&r$=w(l?<
z)5=#k9vg1c9MDIG77_EeBhg3z&Vpj85uvxH#img3arVM?f7SRBQ3(|{d5qeHO^%(V
zFbH{bj<^4Am)dw`C`pc$_49D7oC%{h`4jry9<eh1jw?)BL7WqBhGVS?yK8GEGYO1i
zGQi}SJULVO=dOP9dr$CnHpB%ik^YPoe&dPP8Fjs49AcEtyWgjW@QJ<Fj4z&q_nciP
z1HczQQ{|`lj)3Bbw3T@=cJ7C#M>;SI5R@>u20)JVtChizMKChX?ewgoN(mXf&A=1n
z%*3MwF|5~9h7c!!{pRJ^fjjNb@f(Tc;2-ZeF66iV^qG`X*cyvky0!s8HnvkNXI9-V
zfId}c87;w{Q?4%Y)}|Npp{yc#ILFNKC3P=e!P7ersKXlLA-X{|@9c}SJkzP8wc6u4
z?%FFREIkWfpO=@+1SupOOf1FQB{LfRS4lBQvcGl=9TfOlu;dpefBdU3dt^-T`SHOS
z+WV(*7FC|2e$n4o7fw-jx+>7P6}CLWQ7Z!nn)~`RpJwVr7<~dT`YqeLaHnDOj@oSh
z7F2CYxuY@Pj(tbdu9qijlW?arF~|Q>*1|ZZc|2L%x3R(pOYgZF$gvZ32%1|J=`|a+
z)lp4Iuq$SzSJ$+!RZm%~z4-Mj>@CdXyhu*zy8He#X}AQApz-dg<G=KMt%w(>_Xz)+
zlnQver{o$1y04RGQ50LuI3H+uQ9(uO(3ge@t*L3P8i4HM01R=nMs_TLKDxcBV_jjI
zm6DcYR*j@;E|qH^5FpOxSXVgOY$<y>Ap2T=)7gGED%+RZrYP*zuG8<M0LO(P(=K15
z=2DN{rRNm=g-$2ne$?lk;r(%xuW4O+*7do)=S|DmTu8Hv&kr%5hf64Rv<3{dSOV==
zoNTkC+-9GFM@+Qd{JlKY#2Jd#A38zH4tb&Zhgrvd8pDB!``ra!N>pt9bjB$)Zg_s}
zpu~l(GdV8++LcIx?7hbYU%IZar@Ibi<+ZBp{>NnG1U9Ht`(S(_HS~Cx-`|EwUBfju
zngjWBMrlN0Ecd;~{-T87nt8YUDAC$3GFj=Hy%4q~N#c9U=j$1eA$6s^OD1xed;9mf
z%gz45zW@$B+$=ec`h|_K3fas032n8tTJ0wU(*vHazBpCx3M+u7XL_{7PD*cuA}xs0
z%9R77sgMB;&I=ebS_>~=5v<=9=dKz8Upig_=~=p-3q|tAL(mUf&p7J}a-Y7Q&pgX7
zIho3?z=^lpTy5?)UjtaJY~(t?H%cXvL?C0GhMNDp`q9vS!Yg%szJj<O*#<f{ZMQ2L
zOcvjV@G%uWxWh@+@fb_4(#bmllwT`&v$70hH6M@uYI}6GERcH<14wT_^h9Vr(F6xD
zCXz{kG(UBuB+s*{cC}m_m*SJ}jkBa{_)H1D*-{YJG0C`KShs?DZFx&wXX`la_KVf5
zU6#jH%6<<H%e3FP+IWNdn`d6%$JL;JX1QNKIE0PKazlMNJ<b5l6c8*{9+~A0lqA@x
z;)9cN5=f8UT#=6`IA&lRZTfzR*hp{`4iq>L{>p^m%>(CZQq%QP*YPKx?1reVo7#*Y
zDj+-(t^XYZ#BU<l&+LbO?d-;c+ZfZV9{KkZoPIyziWdp@J@`)OnDdix#zg+KUIxB5
z%RcKr!%)A?+fi!_L5+Jd)R*Cou@&WKYh@@*%4z)?wlWiY@z^P?A3EcTgU)?Mw%!rW
zOz~=1yYy^`B5<;h?eL*Ox%FPxUq)TdW{d4RT#_!JNE@yBQ7g#{PAP_-2xIuJ6_f-u
zeo9>*gZnV!T)$CmiUM;DP&D91vjeu9CZdbqyFCNtyezp4x5L^NxQ7~eVzqjiRwE6h
zZ5|q&k{ky8kZUHDIUZ4u)|5W{m9{j0Io)xcq(a*-m3Bz^b#0s`meWSgp0Jt9o+4#e
zX!T1F_hBlLRd)>Kx&%*b^*O@*NggsNklcJ;B?p_^dRfPm4wUg~fU{zc&w6N(Fn-)^
zNmqcNxi%}S*#!@Dh0G(#ir3`54{b|$oOztAr;^vc)mhIgd&Rwaf^7aX&drhI=h{vf
z5?`YL@R)_;65C5UV&`itd)B43`F4&G0|i1X5qzU}VFLxO1b3TXOvjoe1qI1C54;!Z
zLdtOAUDIXXn>BJpvLOP<2Z!*`0^v4JjN$8aFWB<sM#v{dncFcJ_e&1$uk4z<KaxKf
z$31#-$hknI)>EuwyFoZQP*&{?9dcyhxWisL8*6mI>2=NbzOPI623qLk`8(S__t4U?
zYWTX|f9c3+<I7$hL(igPS+-vWZaELWv*gw8$f9^yv+@W&P8e}RQRAqNUhN9xgfb)2
z?AY(j#bQvYwdynW<1;-p!s)qv6^O2srqcO|tpCm(4`NC4W`+8M{%y^fO9~0m)It&-
zA1tv^iyFbldQN=>L>EUoI@+>><!17@N3_^wjj5H+ML;{&+Foo>%CHJmc$%`^@RB)3
zrhL0lrrb7X4?Q>7e7v1qkii?ElCqi;zGe+bfNJrp2|X>v-V3$vYA)iTt^L2bG;^GN
z_F=;ciZ1){W5!qvTLs~g*<OEfjjr;cp7PnIeTGR>s0?_Eq6`vfhcPN|Kqhdk+tNu^
z_=1@mM{Qq{Qop!f<&Z@RTW@5#X)b>0U6f58uU$=`7EaFm{6nrcRSuXkX<AtGWTX(~
zIRF|9Sl`LryE|&J&yFa>U7ihd^d^8=J-bN62H-eB6mnmq9mk--n^j16Zh5*K!S2Ec
z9T0If-J%S(y2p|u6kH(plq;vN_o1oH9qI=HEXZ62*M1>A8yjkU#PJg$rL?p*tEuGK
zSAFU}jQt_DO^%f|T&D9Wwv@Qr!sP?6^9*|>gp*_Kutsr^aCX+L!tdlT7%fh7lM2(m
zp6;+-*x!2V7RGjo-9>E4@93^G+EGn`5|Q#_=LP>}^=CWZ1(i9xPQp>6YH|fP(%kqG
z3P#_1jGwOjo-Tj<y%)4_iQL=&yWlVWE8m@ihs3rWw#fQw7=9#D48N&E+TVI@n;=o}
z)g^s$&0jRib8o?GRyQTvuHUq=q~f$8q2dzh?^Gj8QoR?@jDq?6Wbs7e_~!_9XtX7{
zWCvBY#)E~8FBs27lKWtF{Nw2l;~Xu}ddzwDtdnM3>V0@th=`y`4(`4?4`3{^0*^-@
z^)0>e+}GXC<rN4wFQE&5PH?v|=$Ze2W`!x9KjfQv;kPHBsb;;K9W_wf^~nyr7VwDF
zDHdfDI;@QH`)i#cBtbD~6Tbe)Vby(l-M7X6n74jC*TbppHtn3^wDY`ZlZ_=x92qw4
zY(;B=h;#Bg2~!cR{$8~jGFv%?i?^;&T5=z>s%@!oeFHH(8v~)Vv}b^86HB6vo-(LS
z$-O4scn`-mIoi2qt_qUM?6_;Jm}6?Ev$Jf|hX;lMe-!KYet0-7+@_t;6x;nU+$P(3
zPN%x!ZlX{W<j*`zE>N|iJ@TBDH1D`=-=Cjf3O5=9vjU)z?%5TqeLS-d)p*PSWP6{Q
zdYTLiUk8P-Yphms;uyU8I02%=5AZP4I5ClIJRMB92drPxq;2R8KB;2=-SaWI-*>Gz
zM>1@#9C6bc0kp(E(TnI~$|E9L9idnLn!RBpQ3toKp0@-DeuHQ4X5Z=-0AzU|uIlcm
zvD&9E;~_RCT(p-g2DSHp+H@xrHu3znuHw4n^Rd@3%2rE_cyQa;(j`z~|JE%t2^9so
zw^7f_p=W#b148{)^Z{$!rbt8zIb@LWsFtj@cTwA&3m5mq+KQj6*%B>;PbG`~#MZn@
zVlYCAcRx?!449G!<E-gy_o$#Evh+Y=g;)A4OW_;lX!*_B5(5z1(5V|@qpaAv<6`ha
znfZH5=42hm9YWAt*}g^{)nXYUZeODwuhM76H`6c5L*`W4@bp`mQ$#+b-dd}5;dJ-X
zbUR?F?DQ(NL^p-Z+sr_lIo>`{M26&3kw$S`fWlDLs^7*6U#vl^BpYW{I6I^ZIR1j_
z?z%tyw_~pH?FPc#$p@3)Y?b##53O^LsyA4u$(u-Co=VvVp44hwef;#?ukI|1(n#X_
zNiY{^1~9^{<>&^R$He%zI$?$qPR}<fPB&X(=2-T@7BOuJ^NO-IH@=&o`pM^pbA32n
zRRy>19oq(&v19+W;q0Lo9ptRt3oZj*mxKBaP@Ybt7VpF*P61zb)@nTqkvSZqJY7X;
zvHSF8t6yue-*7CYnZqp9?<Z6ax1D8bd^i_?jN0{a#EP-!Nm!!Grd_x;X&&~-$&zc_
zBkfm~Sj&<mCbF<08SCX?f{(rk4_A}6k!6Y=P!&3`nj#gtFJZaQh#qecu<VI{cM1|Z
z>T%=<*=yde?7tK@V17Wf+D`n9i7nNj@BKp!d_=<~wZ^GA_A#IZ4XD?WB^XfeX<Ts9
z>JdS?(zxEv@kFta4Vb(XlCZibgv>?))o{`MH-(w8h8UG*a;+_%l?Tb=FT1%w#eZvi
z;$3U%lBg$=^h_kS1B1Z6p^Wliru`js0Y5v4ZFDse7G89$24hOusd)wHFbug=|At4D
zTmgZX?|9FB0MJl6kBgoc-_~Fb*}+kt{3P_88sXSY-OjFOse(?CHeGWWo|Q{Eq4UFL
z;+05ie#bN}!rWH}>pKM-P&NTA3$XDUrj}LNs<{xPw6eF(f|ZAT;4enbVG9BbM4@L!
zH;?+?*&4WBW=2rz9?bAcWcJJ2wuPBzqU37nDe<UPMA6oA>zy-34J;JKLZDBKp|0*H
z``f_6PclfNvV~&T;Bv%?WC%sgj;iP`iBA@ZPM`u7ud3+%_eblek{mYnIrxWbr`1|!
z<r=Pr2x3})c58q#w340HrLr~bU#IO!gdP`78d_2=O*ksYXWvG4sCu)nIz8mcz$rg{
zs?C=~7+1;GSTju4`CCqC;E$Zej83?eT1wN)4~@RHxoe2bqQ1ziQ&%Ckd$(Uzh2+bA
zg#<;);WO0^%nLTM87ou#`b-VB5kxlRj9MZ25S4xW9bfn#cw(6IkYJtUP0Z!P=%plE
z8P@c=e=z8c4Vkq=>%>^oF;*wimli%e!YKofTrGGn;NtSJE*o!WNknHD`ERVz+{27E
zUL}S+L2SfoPHA7QB>1d@UJK#fKMZqU)=V4y(oQ)PZ%Cnm>H5L&aB$23bL(poxDf;>
zWM5;i?%Kz&UpPeA5v&uVfgKB`9ZTu~PsyXRd#rGT0Hl4=$Xccmx#57kG=|7Q`xgoK
zEaZ>g(h6efFVV7)hP|G7i!UJXd^>d2%<!sLf$zo}!HelZ?ZT0*6BNNtO<i#I8(PQT
zlwmX1IEK2j?n#3!<A<Y}HlQNu=hY#}9!)`4wDx8+3T}Ih^aP`hpnro7#0C07(xy_G
z09^&0m#I7@OB(g{MCKCKe!gwbYqo&mWe``C7)MO~wn@{XC~aMm$M({Aa);mD?FGp5
z%O1*&nIlRQ<q)Co5+;mATQ8RuW0e8v@WSey@RG%lAscweP%RI`L^H2L`0fcvw=*y<
zO*PqZ573j+s&lE9KYV;pW1&l~%`PMO;&)Vn4wGyw?Wwmjz?2mxn_!~Dd)v#c5JUay
z)NM>_g|sI4T&`QZ=KE*6%9g!m1i8i*yst9}<CBNv(9>V!&MT>w-k$B~72}dXCP}~c
z><dq;75H-+5-xeK$;zi<%!Z8JJhe^1bRdzhVcs>(F@fy3vn?&uI)Dp5Y1T~nd2>jt
zjwtDF(A&>;+JxxfS0eU!t5JT+j0dsrF$&bZgjkEPq~K5b$t}6`1i$uylJOoB@D|HS
z@JDxkaMIo(%h3w9TZg<a`q??Lt_ZcfvTDHu={a7yft2yx8hKBPGaN@*DjPs_o>>6w
zEGoi|RP9tjk_=?Ea4p_@x_22);(aBPA@ZDJFp(hmDFH39QEz}*_kL-RJ^5f$@W065
z|80!K7br0N=lZ-dpd2(PsJLNFk6>JfaI!?$xmJa}Eu!jfv$|Yx&8_eld|%)9u588=
z&#TtFYT-jdhHYJgy@m>*eI()N>JRbcpP%i9^hwpQ6<_+`b^!lGc?mAxN{g_JeVXj{
z@{xgRze-(Zc%M4muW<iNp39II7}_{Ky>?-u(U7+_Hx!uP`|SRCw#h-Rs@zZ+>T=gS
zZiLr>jq3<GoX{R!j)^i*pkywR=3X+gH#NeI&Boi6Mi%slNh9%q*-=j9lw>RG5oO)6
zQ@lFP{aHzpBg;LWw{GO+>2H3RV6nyp7p?Ul+N>QV>tQxnGshh+;OHwK@Q><Nu7uS;
zz9+GE{62ZeH&5lMjfBMQ6hEIlmGqF`F>QrBx5JL}4Bh)6O6jq<=vsGC0b6CboQMOS
zoQRY_U)YOc9GZNP`jXoK_S~iKEp%<c#dEhSe@6_LQ(YLtV5i35Qi0>77zD9?x?o8j
zW{{ZC72lJT`tZY2bQRgK^PQ|npT1|0BX2HstS{#|GB}knJ#a}GZyY!(09M`MHkYpp
zX^(-S+U4}<eYbmoU`)?>6k)N-OIk2ovP?3F%)DpiTI;pFh)`2HuJMQOSa&q;ZFmk@
z0;>qrYUUUH%wyk2OkrzIc(h<qpbM$iXFu?novb~k#6W|w4N3Y|bXiy+B=EbU$dUh8
zk}VS}A*OSHrPEc{W@<(U32?jgg>+OvYIKE2qT(IPt39udjGRJ_%#SaxNs>ful&Lrh
zg|>Ujnfsj#R3AuKmT)6Jl4~R%2nDjqzAGNW`f>YLu0{;|zvaz(=aZ%~;l=q*v5)YA
zm7nL*+B1o*>@)mY{8fr|Tddu`3l7n;9>?W~p<0OF3T+yl(7hflp912{5B>HH120Pl
zI&N_L!ai)IcOq=51r;K}scCD;;q3ko`fYrue>}849@_5$9yh2~Bz3Kser;m*N9O;3
zF@0jd&jA8dpq=P*eH}}GA+4H*hd!-WvQ;SU9wSRNkl>GG$z_7epK<a|NEV2ym0NX%
z>(=X*zlU!`Jv&zu6@Z&KlN|O#V<J&-qpf?|*LcD(Jrea`yCM35F2?j-`6TBNOY*_`
z{0Sz#?;97dL~fszFsEC?=DsNSJ3d3iw3?}3RKo2<hf*st=f!8U{cni2Zxle6XLuVk
z3Hvh6E5)p1$yZ}5Outp)_cG4(g;K4!c!{6KsgHe7AEs>2DsO$a_E;F3w|M9m%=uMo
zesvEe>ZyRP_Byya8Xo=?FB3A*Di?cxMW_%b@A>rOGwIHDh4vDnAnB`vuA?By=9IB+
zO(XFB?Isqda0d9=$8_$T*FoYnbcsRy{^Uy&s{#N!%6gd7o|Ctb7m=LxvUx^|sN@mT
z+ozS}lHj#YIq-Z=uy5}JL<PCMrknMMhW_3w>3RL@=6_N>^{YbU0J}$G7tqUs{yae$
zgZ&+O@SDmsQ$-3p+MsMuI`H-7`_E-<UwMM84FIve?@teFGr#$WSV5AXTsKk*PU3>)
zwMw0xx_f)f9EeQt|1D>kIm)uDO%WmyVocY0sFFD!YlWRiQ#iU*GPeD4?b68WGU0TZ
zFH7<rbnT*CyJW$g?VGc3V8@)k!cidCGtoMnu;2bXzPqkDh7|8J0m*rkhyY2Rgjejp
z<`A-X&2lkvBB*F>xY1;`l;o(Rroa2*`j3*o$RQJ45n9Jl^++LOLPQXBJf7Kie^fo@
z{)*09znfGoi?!JSbrBuh&ua>#BiX2xnR3$zhjVy%Z0AEwJU#pXg4l-y4F$o}vYkQ#
z@94&u{oPf>v7}lD<e~BhEE=70?1vv$N*fC*s_*At`7qI)>6!>oE^gfRDlC+KP1?!X
z@r;An_yS5Ejmt1nfeSQO?Wr&bZo@G0;Vb*a)kcM;JHO%>8nJ$y{HhruMERI!WZ{@5
z#updtz<;wrI>zL|7H+fg-<!YruvZ>3mytw>rnr#MhDJnvkYhv?1*P2H#VRSGK68*Y
zOlna(XIx*{pwN+W^*`<9oz`^rs-vzi0v9f>p3;id@OeBV@yQ``C`&g0XABJbL`}AW
z{5%v6Xh#KsGBEUCG(^FlE#&qvraLK>$X9+C2J%dy9}h<-9_r-eG{sRXFjAx|Fr-Be
zAd-+nB`f7NX~}fO&spzi%UUMKr>U_v>F2qkFMbFa^bRRuMswr$zyV6L6objRegXp4
zAFG1Uk*$M_-1j$jUpQY8G^QL{r8GTSzGw{ZWC7Ff>sjmR7YxLZ1%vbo->g4|QfoSc
zN%Oq#coIxV>wV*&VGB81d%W5+xe>%|?k`hzpIi1PeDdAd@k>cx>Og<FTNiAyz*}SZ
zH}&RT{8z$6F$~E|>&EvH7@d%oFfLyA7u}i86#xE{!QUy9-fd;vvrj{7h8l0bM%Gm5
z$F=n4Z7FoO?GArKbX}Y!`^@U2N<pt@BwVok$>JbFNNWaDwld?5qI=|j|DDsXs$!dh
z;Wnr@Td`$xZc41_j!CJDu>{u|==9KFfBi{qNEG>?m8POvOz`TE>+B=Z$v8#jqUZYl
ziwSAit4Gn*(y|z)7t-pfr|RodFBrm=+APh!{q%3I!kou(W?nU__KAmZcUh!#%!k7h
z_Z}9$nM*no%B}><8mY;yz1kC6sf2Puz+_;VeRd`u9Mehc;cd^`!XDjArY@~nB5R80
z@pgSipRW@6Ayv=dV|rJ2Z};i_Un_G67-iAIa~(<vM`tV}$qV+PpD*RxMMa;_Bq}EV
z6soqXonG`_aBg*}j-nTyqoFODODypd&%|YFAamf#_cI7sSH{hPhcgb0zoS*|0P$3^
z8`$*7{_jn%iG?pK)J`+w&P8esCZrI!Eiy>hxHicd;a>e#G_ISPpJco(en5POLx!6%
zt)!3fZB}AFPa}t-<jO9afrDJH7!WrF1&-Bh{b$KL88Jtm5o9s+0ilz`IamJgP4EXW
z(`|{_RH0c3EyFQhN3d&HDn%7plS-dECku|L*~(1AD3nQK*57a5oh*m_q<@1-cBe4o
zjJxaNxV1_334SaYhW`Id-Lj6a*wcCZKYYlJ@3rwadHA~lXqdX)E5e{qg7~I&iVfpB
zRi#N?{CLEBI)%yyGHMHrC=D)6x6fViWfOv88r9wFC0ti^ei%l)r_m#11LwwQpqf4;
z^Z(dvNL561at0Z&VG7TRH6hiKy==l6KC=#6H$Lp@4$3%j;Y<<`MZW^g7%ttXM7gm$
zQ8T`PKuxQFHDY8H!1&(Ii;(O!30-ZKH_?$~PM?sAdsr!?)|yiZjwYK<1};k83*q-L
zV#{Uo8ly7TVeospU<!NZxvyxPVEj_N@C)c9gc?Gyx~O&CoNN<v?En5g%g?=WZrs&m
z3uUo8zcsMF7_0t!-v6)lr2?!kB<-fX4@8qW(*Dsc6{}mZbhgqwV}ZF}zk+j^y}C5I
zoQ@6ikC1iehxJk*;<$+Qf=tcpkKRLi&FiCzxBTt?$3q8=Lc;AQORC2}J)TP|zMeUB
zi0}@ZZyffjNI{e`&R+APn8qqTHC)wI?X8G}|M6_x>L4%1NuaozOSx341YCHmeV<E{
z#>gfweeZ>&#8tV^z{To~l6jp{0yI(kF%{BUd0HZr`<2~bO#KhY_QH#I=VKass}UlJ
z4ape&q<|)()G~ryFrOZU28GciWp15X9Oy4jrmS<*gsrdo&fk7_IkmsCYOz!>el3xt
zv%6e54}=QjwM5EbHxhCl`H!~5iUp54H53rUUATh>c@j4i|9N;98N1LFLEWM`aHwia
z?f`Y?61E)Jujl9ASs3{rqwp6ySF$nSMaT)@-|^ZT-VdXGvj}!j*6NFCNZDJ>puR_;
zk)-t3Z?+ZR=B{tP?CL9gVcChO<Zv4^Pe3^+N?}N4``T46Cs`o##ZDFeY1$kb0?uX2
zE4dEhc9#1O5$ZqADN!8ZHa&#RUnz+Jnd<+svu<#Vu6i9R=m_i5yj8hZ-0D9So1F-P
z@Go$|0s8=FmC|KGFoEP(+0~3R@<DAqqr#paJF%7DzJUA=<eYU}6NVbYNDaf>@4nFs
z;1s)!m~R?`Fr~76Ub*XqmCJs|lQsE3>;m1l-`)R-G5rYiUupRYQQqufkn<+~|FCQ>
zWb%=LpZ8f_iu6R&XKc4!wQ@q}@2*l&j4I?jQ!=R3zJs`vRi_}rMqd-{_dYiIJ#%{e
zh<Sf(bZK}I)qZr-vrH|r0PG8F_?vviwT)zs^!r-kFfkaA4E~>w@IPi8z<W_RcVZ;x
zp@squa+tP8vAu_T3!@=^I;7bU$zMf0DSbuNi@g6Da0XTnjLglV7Ph(Fr<;|PVjsDG
zYkn%}*aCcfQPFCr5}^E!=3%IkGTc^Xn5#P%Fs`(@&N?`HZK@K0*^s@F2w(S8{I9)j
z6A*^{a%umkVCw^;)EHZHy8hRXW&51kZZtq(CVu1zIl;S}@0|R=%z*|yWg=D`1r19i
z6{_o)KW;d#@{kiFJ}BWZddI&C7jXW;Aem`ALWq6l!P(|RA9h_@IPrDP4eG%u+Zj)Q
z9{vA_d+&cZ*S32&N|X>ah^V7?f*p)*^d2=MN+f2GLNI!ZAcJJ<Jz7MGo)C#HYA_RJ
zB7`W>OVrW(_ndw2=lMMM`wzT7@i8LTb=Kop$6D)9eR`e$fvJ_RnhNIJDj4yf$~GaB
z%6q}6S@VEol62_Q>3>!cm_!wT9UZ#@4b!}S?m(hSk3cJ%HHDx|7KK4y{LovS(8c^F
zA=Rl3ts}HjE9iSP_KIz`BjLqqa?S>mG;ENT#AEl}aTpuASa%?u^3UE4@H2AggNx~&
z1m2xHI#0@n9ha+Y3Sd_DgGB}$rB@Bm>4gSWCP#Wp_CKLt{L+_FjvpNB0&VOQpdSE5
zOpfaaSl%{B&=1w*z!qTrp92|G5@`YG(eIwG7jP<JOP{*ad~tr_pVxgU_WVw}d$iwY
zAiI*wRmIZrlI;_EMuAEkY^+=gzoZYZ?<<$pP+>+LfL1K?VShb{5ShGbDp<aTxKS&%
zOu~VG?EgEAT-tGz+QtC7yA4wrfvg{a<}&?G-~X6;SC^xo4DoyXRJjygFuiy7;aG=K
z=FJBkuXElgB)Rq&or^r!5GsR4V4TVRy;_0nbsYxcwA@=m!)LY5(L_wv8H&#jPw}=7
zq9?zbsAfkCkmya!S0Q|Dj0NTD(LrKjl1|*AqjtUvtPoVhR8vdS1`NrB(zGj3h8d?T
zP;q=RDFE6a&7_+&8_SQ3p~-iJmvgrzB*#jRHKHfzasT%i%K~G_wJZ>|DEL1J1bDRl
zH^Y$8BO}>*G?@uQ{dS;7sGVX=yu$)(JC=KE`zzOHLU*7)@y>VQMbb30XRyp_PN2Jx
zcGAt+9&x;7TPRueH6E_zov{4_u<3KGX(DKT2`)!DORpb+Hku}!pJ!iS_0Sb7*Hrg{
z8R{vgvQN5kWB)swH!uG)o2bnHxu*U*JliSuxE|lx*t>>M*R(vBGAAR;6{~$95X(xt
z`i9marjH|?X*q$_=psnl#8%PF?x_5k${=K}?5<lqVsVG{AQg2T`~roBUf^OE{ebg|
zal>OHwB@;O3YV+ZnZSf#)M9Zs7?MlkmVVREoSfmWq835JNhEKk<y9J2TALWj8nPm}
ziKnoT2QicAibNZj0Tf8xJHyuiKMO-!J<{@nzeMX}0SOid==4enTux#Mt}8?WC6*Q|
z?FI@`d)^-k4{KkDaQNL0+?R}<w`OW>7Me3~A;0mZ_lbJsfUod;ZkX_T1`z}~5mc`q
zn0I5C+@4hCZSTvssX5~IxnXf(Dub+&vFSteUAGMem{+8vL<g@z=e4BP;hHtWT(~Ld
zJNcO6Iu-E@<sm1erRH~ZS)$qr0ddYNykh*obS}ABT+`g6@9;^ai3yJ{&dHQNzqj|E
z7muE9#T$N(rwek9+~6m;EQOW#%tMJZQU++7hoh;i@itm15>3gPjEd{*t*Nu4H%d{z
zBOBha(lyIJG(xXyv9|sNH@%%=P1!ID@1~Bj67u5^dG(pRR9%%~$SG3eX!adHWf1jl
zSpP?jH$fg2{9*7g<3WgItjoX-Uh94PcNQU|`m{5xe~+I_r6N$XTYP`9(;FG5&drDk
zHO=0$4u2?mE$&gqi!~ItQjc59(IOlvi+OKpMwORU*x!^rjc!va9^2;Lk;#11!J3!#
zWC-GJ=>N<wpB4JJ-Xy_~W2K9J9WKL-H<uq<=&r4SEmS0$Nd|ayQd5w87vV_txasTj
z4r|Gytm~J0Q>suoP1D@5uP&XWhPZhs$KRu%a0_>L`BA@`&T{-db|=HV@wkBnnk6$x
zKfq*93X_HS?bx6TiW{!h9S8mv%&$M3eK}wYHAt71EF>LAgC#|ch?HcbA$b^ujN?>Z
zD=E&&te3m0N~TDGjJ{!cH<BimMq~RXZ(X3A__rZyij9cvnbLyQoNS{e8ia&*&nD?m
z(_E~g^bj>izsQpxZ&DWrB|sJhX16rZN1#NchNX&kcR6ibZ!>DwzQLoS!QGi5<+FZ)
zY84Er<}yGB3=%}dMuGMp8sSo^EJMP>q32Lr+OdDdJE7y*9S^>IIKQO_3^@#6<q_?^
zbA@(D%o}adbu|OBOm0@pbS1rcV>q^K_LHS<NAcWut!1t|L_H-n24Wu~CV`w#<|`)T
zl{h6<8Qf)37Jpre=vMD=Z2bZbMYqWYZ_L*(GVdJYzum^kieQ_%+G6jyRS9v-dZ||a
z^gr}kSoS9)1-Qi%d(Yx6TYgJdt9oCnH|5I_=wwvXIo;`0@WJHi$@WOb!q+R3%8>v0
z+#db25VqU<VFw%Ha`Yi?G_lggbMrBdnS*<o#jbiSX`dQC+eN;iluTI_qBc{8TYJUB
z@vYYcI=`+Q9*9gDmNY=O(1^C+R<W~t!<k$S8afBlp{-ZBRFtZc)g5q<nF1J>!(x0~
zWCOY04JE&)=OK__-aYfZYs#MdVs4028<th3VD5GQ<-s>cjd4E>I2dRGm!JBm>kK!N
za|KDY2F*p1@0mf(IZpYYd@9f3+bIIiGdxCwi?b5b`d3YPREtvY(f%`Fm)^~$&7S$0
zcig=NRf<~lZQ0@Y1&U5<w4sNvPh0`ZZ;k>w>2;jif_^%p7yeAd;CQm7C&8eM0wt`X
z;?@@u=VPywWY^kxdr^(CW|ahBj{4na!b1nNPz_1zh@iR4(zoHQn|%QkOix-5yCl~@
zQ<@BZTZ4v&z0EaxcTvxD(-46I6A3z;3NlT*8{b#f{xeb&AhJb%az09dTAbR@R~3((
z4*m_qlA{GOE2c9}G$qU^oxmaq&Zl2MrW(y<z59w2miFU$>>%qp=+ox=DVzrS&I~<_
z8FZ}8<IU-9+-zxF2NgLVz%U+Qf6J3jxub{D4LTD5j3W<zS^{$kWN6_po@t2v^`~Up
z+m-esEx0K!S=95Ni@(&=Op<GG&JF%lQ8ybiQf6JnT<E(w1us1v2m0jdn2Jy14{Hi)
zaQdQSeWF#54a94cgEzc#G1e0*E6vcdybeo1Yrn-Y+w$}N%P|5!9wXTr<U`A0MRhFd
z9hKLy8N3;|pHdan)!@6CGpT`IXWZ3tXy;@e_eztn^E@ev?c+KD+S!~c#A$q-cU=$9
z%lUomC$fnqnv$|N9V3@H@Wh}2<k`a!#(<>U(Ko)d8Fue{9}a+e`0itH$KXEi#KSK)
zzf=A$IBQkd@$RMg%W;_j=VQ%@Qk2k%U;^HE%O?b5bZd-Hve!A)9P?25!?3jN7|A*e
z#E6$H-Sn7-k`<kH{+65Sc{B}Q?8c=h@aiCC0?&VbIp6(mvLw1evGIgy_QRU?N`eaJ
zLH%qugJNoOmfvhN0c;=YSG_n5YL(`Q_()f1Df&Ft2KQzcvv~l>k+(KE>Oh$k-)4+}
zcBau7wDwCZ(<Pi0rrGyF3sm0H?(X&Q#{C=!J>NNff1gr3V9%Aiwj36X8Z=XMsyYD8
zQ6D=T<I)J4O<$Gx%ae@u0mT~(si;A(iWOi3KHF_YT)PxTLXK=i^@J$=uxmF!Te^B9
zf%0)VCg3YI!+Hk7ZP?(43@U24f#w&FQXE{O=h6j%i9GP=NH28N!BS^|OZsh2KFmr$
z3^eZaHy~PvpBK_t>cBvFrg4JsCYFpox>d;~m=D{1b*{3q6Y<ec#UrIBR`tQ<h%v)`
zNtn=IEsHUeNydDGDj-O_w8g{@aNsLC;e$C$x`J29+g`n2xe@AqFtf-PW8Q5ZK!Ncc
zP(H9d!Kbxm=`s9SUs(E%U$Hm>2T~SdY$0~*<o&?j&j)d}h*Wqtr{C|ejMn1zPsy<d
zo7ntn>l9YSK<S^M9|lxTevUt?$_5q(aoUzYZHC`Siss|3BOOynNzBNK2)g?~*LlI)
zepLp!R?gQPLQ<N=<b1|ug7e2^(8d;f(bZ@g=TXQ><=FSaed?R;B_^5=_0a1~<C_Xg
zSq~ge#z*lvj;lQ07IVbyerJ(1&P=-+c1nxs4<6Sqq+*<fr5HytX)v-6?FJbi0Kdnn
zUn6$DlS&F2f%{O+#T1&1H|3s!b{k*))FZ0-QWoy8QU1Hr#_sZ@71A01-Ssso(DmyC
zr;vnoLC{>sv-9Uv5<Mj5G&PR#d0U)yM=1S^e(Ju`YOg(LXMgyg2u9=gh%~!f1K6Q4
zk*2)&o@v8ZEJEzgUh871^Y*7hFHX7V)c83?4g-ZMXZ4_pL2&w%Y1oERgzJD9-07Zv
zpq0|9ScBEiQWR9QfJDy#s;Nqy?=nDb$Q5!pGm*nY%W?cEZurVSA35de=GIMzoGd6?
zzSaJ_k3_4h1i|r%WmwCeU0oJ^l`Zv8{;uGe0dWlp$8DB4&{wjlB67?yM;xTh;VKC%
z=}q!Bp=WzLZNVR%chOv<{w1(nE8$hizUhjOf&UJ}yPx9T_iQoRv)>-pd5*s3=(-*5
zMw|Reb45t}s-c#w|APW0|3g{6T4D#}#oZoixI5KQd7@32WX-N1huIP<^~XrS_I1{=
z%%G=TTgEUN)jFRxwFKYL70(l^=_FBvngKZzsB*oPiAEYg%b5<P1Gm5Qdr>{-%kM3z
z35fjxb^_Jv6tL|`juyxj!%HeRsld=bK-BSST^-bNICmZkd%rurp3bq1UZPlM%~wET
z)B9L-cJ$H5PsVu>oK63UHZEB_7HJe|lH;}fTiucFl+`cL>uBU%#nyqTjkKsTQb_`k
za13P$y*-(ui_p08%kRpgM%k>-kklLg@#?aFNj)(_t9-M0CRv4gcU&P#Voq<QD)od?
z<y(L3r(`D8U9d8T3X}sA3!o+ETCP0V98ZKrLzeZx1NELix@uSm>-^(oyyC<Bv9SD5
zT^91^qjOYI%fQ;6K#fUFE$R`IjmK_<d$4G$e7@V2Wh&)TWpX5Wz|hOLOjdUwRi^h|
z@#eu+9hcuD^4OXwrk-|OUH>LGjU8ah^AY6;NxW75u_<|@uv!0*WA1ZSAYo^_l;F~Q
z-cWb^pe3zeotdy-&WYMI<gfeg`1JQ>!AsS>zmoVQ5dofm*0vgEV>tH$b^V@t#N({)
zK1q@5eB>I7DyU;{uexk%AB_^Mi6FX${<)RGwg2ehJ&{9M{gpKrVChkLT(IBNedEzQ
z?9Ombp;zaDWMV&!gH?f8YeWMDe2KfT;`0=g>+#I=)E}Ea(A}*KiUxJ`M`xAjy!}_X
zqmYLEZ4g73kYuc|D11R?dyn)EIJaK2&u_RRx`im7C^hGwkE3xN0PYq~RLWm0)vj`K
z8y+>Sti_)fyiM(kRo#2Z(gyw-W5-M03VE|2<*yjon^gF+#MD!R1Kh^s<w->;Uy6KI
z7Z#Ui4^f5h!la{SE6!M*vl2}9aE6Khm>5u*`dcC^ar7EXG1+sU{L{B@KH#*e6Qz01
z;d}MUyaQD{jlMd|GEo@a2lA58i*x^brmM6yp94#n-ZfbUB(McM;KDh8{=PwGxm9H`
zZ=phpW}<oJ$&;g!D>~2n<&W;H!;kT1FQL1Ly#?{fA*x`G^`mNx1SWjATKO|yHisob
zq6__20v}YQ?DY@@-;^@=TuYqtS`<<B@a2_oQx?|&EOkvQU1EeiQ>-lDp<gLrOW{gT
zX`&-jMcqOzn>L=}F+0>By5%E=>bk#nRGr#A-SLqOy9~^|NY;t3rgz)jdg{%|kgD|*
zCrA<mC7*Vs8}SuUsjyK?p?M#u?vQ2C>-so^OizQoi}&g@(I~OSxmf`%r3{zY>y!}v
z5`7ZFi?*C6AFIbb9H8D}N)-)>dHWIU^uEet0p+JDJ;s3{XM3YwfmVki%?6>m&N^_n
zwm)i)gK6V6$HoQJclDjg5#^<d_Sb0CozIVKS8q2g+K8Ig7jmUIm^FAU%)`fPkTOiM
z4beUJqAq&Y;`LsI%3dMQxu_s2v1vFXSARLZdu{4;-_#n_NUQ|&wC!UiE@P?R4HNy|
zBrj>Hd3RYC9TG1rM+r@^Zwd*zzG(%<de)mqc(SZLKn@jEKcbd_fm*F<xux?X&QcoJ
zYKg1fim0<l8>mi6$ndB}5*Gg!bW5n?UfS#3ua;2J!kTdDfHR8~^$_v;6M0YA!^<se
zrRMhf&D;??bJs~_3VP#w<87YxiA*LRCpx&M3K_HZ|M`54%b4NL7M-Tdpu{1W5qVzh
zh@8(vkj@`PlFAlFv+H`|Y6$BigPKuMAXZa}<fXQL_IlR_EG|^x8biIo`!u&guI)h?
zx7ufB<maoCLoYx7vH8(5<WR9{8eqziJlTk`kD2`~sUBymsMy3QRE4r$=&#*{ZNVH`
zj00BIwh&r6mH}JdhBHm}3sZ^1#6r#`#UcqU`pu43jaorAwZyxFoE!mis}<d{wka<r
zN@sz5`F|Hl{ThXnPE)7>2kG{5e0`r0A$i0lkm=g1;>&PxsEx^>OkME&DaHjXn-bZ6
z*^c|+L=HMk#mv9=-d2~X++QkH)>bLUPE`~_$HvqZFd@`6Ytp=hoQ03Ntkb0*JEQyH
zF(|$o(7a6mxw+{rD2L7FNdBNN6q-U~(AfD^YpaThgE*y;KbU(NcE-bV#-qY%UHb%E
z`&`5>dDc71t^wa*AwQ9l+KHwwogk*7dG10se))#<-`U1$JaW^N7T#Ap1!=OcSgW6d
zWx47t6s+frOFBpkWCabHd<vLNQzz57yc8~NAy2QFAjg<K)xeIocsO(=Z5Je(lV@!3
zbK_&`{>_|CDo$yyQQs3oUDH$)6)pNjlxGA&tGFVLd+*ggUIJSCO7PgX1B5XSh=bWP
z%~yR}>EPD@GvCWHd*>{F!aGH%Qf`#EfKd~At$=ogc6(-FaqgZ(hyV6WbKt!;5Vxu|
z{$`WCYDXJD%k?7m%SqXeso+cp*wQSLIC7rgdzAroM_ahUwZqXX1zeRPjT07<;Q5-S
z>H?H~ef*krxugvwr#8-R8ho9aJn^VtGPU}_g=JW&_?UJqvbT6fcxE$eT=?g-<8T+A
z>XWVXN0*bCCX=DN0qUivPK!1S2R6@9zqO#hW88*}EZI_8FRtoe*j!rJJ%*8<j+b`e
z>}KcgrB;$qsr)m-(_77c?>h1r2NNmYh*TC><r$knHHMFPaA>)-{P%#g4p!AtYjb3>
z!P9~zoESu6#^9izCdIB{C!nKZIK|hTz20D6poDOn2kREdRVmg#a!ohsFmOgomhJK<
zu<e-zBsP+Alx~WBf5+q(e25Q}+w+*R+lwos74;HVN2K|2^t{T`G5&ksyYgE7$?iiU
zM8*dy+|}nBer|zs5%LI*6uPgi7I?@1ec(ePFu0<zCyOM5p+Mlb{=S2z2Bcgd^gTa3
z^;ouc2>vTnH2Nv0epxE?YM%j|9{ZsC)dvXd#ZLkRrw9U&_21#Np9pp<J~%zW1$V=3
zaWfTB^+N^a--InM8r-wC$Lkww+PaW>cMUWhB}k{b$!OTqRq6ib+N-=ap{O_**B8Uu
z?D+N7)>t@5&PE6^#l(w}I3Of;_c|+3_Tp>iJhhM0uVUu5yNyh_+c20C;{N3skq=NF
zj=w>3K5PX!>KaHfKJE;ts)h<$=cbtksE8K;XlDLbKJeXe5Ak6KMyd6*O7rosHK6wT
zmMf#KW3LMSiPC)*JJgV2GpG0PZJ#Osr7@WSr6kqq+oDj?W%RA+mVW4;@^Y$i(KAA-
zoOYu2Lo{c!stK=OnFgNJr!tP5{i)i>T!wRwE5YT-RsG$=%cY_e4FMoM^|G7IP*<LC
zZ#}EFKg^0g){kDjZjCZ*EaKd3y*MAKa%i218zF9@@oyoR{J?3ya`@VYgq+UMhQ24<
z-iu3y7F_lDU$^JJF5Nn-3O&P|AAVI`JT_=zG?s^`wz_^_w)eO;4LgvWB*_*3&+?0>
zzoAHC+(Q3VEv(-W7#!CAeQR6@V-kXA`~@BtkN%q#wY^oUA0GSS5r`vINqD1<BpTQ5
z6F|+8F`!U#yZfZp(kJ{GzAb3iwk1qEsQBeDanKPxDj%J8dUN#0kFmFLDw*W@la0Qg
zbCd&-b~{^wv7t{Lt5O1lvTs{S*4-ZP`<?66uRw4ITMDRKDSADQeEkIZ%e^(2^OefM
zgxArDhy4Y^TJ5vyHG6bj+>O8bDjQZsegW^=^<vdCSGVs3A>haYjpACo-WLE`KVcLw
zOy?KJ#W#c5EKt_CL2$~j9RPYi|C5o=EeL`@%FE7N9z~+C1plWdGODBe=+>0q-02#G
zX3ttO`ytV8aGdIS-Ps)1#rxc|HRt-+vt=FCYLIJ;5M;A^{2X}d7o<P}It_dhBA>i(
zBdZT_tNC{8(`L)bIDwLDkY)31brjIbU94?ABF_jq_<;vGtQR|~7fHQ}&{SiWvVah7
zEN;6!;@kshvOY<MY`=u3#b4Njp?cC^q$Mk^fBcT!KV5eZ4MYYCPy;&<2}vAS7e0z-
zyWiJAA;&)!7Fgtlv)*6-H3}NRH?vzE=OmFbvy%={f23-Gl+~#Dy_b1<OptgfT2G~_
z?Eie=(<fX!kG=y$HeNdUa^FJkE%us=Ka(zNO81|`!p`TDOg@2H&lD#}ehkGR>EQbo
z?-d;XYlu-y*`YMXUKDBp=PxygI=#6=mKnwj)w~|BA*=K1dV;t1^~k*yZ=ZKrxQsrN
zyYq$gHSA?oEdc_0*izUCE4P$d@bVjUwfz}haP}nN4wdPcWe7c6y`9&$zx=sdI~fe-
zfdwa&I~Ds%jf~&{NnRLVE-jt`j#^`~EouQ)tV7Ut*C((mo%8*HbtE>Q3k92;<L`+J
zJ)Jv!>=@Y&V1isj3rJ?XXpZ8l<RhYwzkphkn|33EDF2&A`_ScMCo*%Wrlj1I-vdt2
ztO`xbmE#?BadxnjQ2^w4-$w2{xC&UGe~{fJmraewr5fsvSIc%{c#wZ(M4wD7smQSn
zX$IXzXpO#i?2uNsbE{u?3{(KM1UcScTCzjrP)!qpYn}a7SoQ<~KT`zC(8rRobgOfO
z60JjNRdt`p(UBon@Ac)QrQCs<j91a!u^hj6hIU)S`QHELjzZ?y<*7fCiiTf3O*D|`
zb<gp8Fo1(jE!JFnZX8L@6)wVgzWVO3o`aPiSA@zU@HG}_BBh4=89(bh=Bm|RTvxhz
zCvS&C!E^pGM+MgeFGvs%^of+#?v#Gxekpe#W;eesQsX5Ssqy)WLy8sQV(6psOaJpk
zg;*f~lp8&R=lSW~Qy}@vBdy<bg6u#-hQ&F+<G9o4^vO;?x|8K&?Qz{EU`3vG=kULz
z-U<%N)dt~)(b;{nbpR&!S?_jG>*{q<V)BwKxt3GIL=CnF#hqeUyhW>;s@FkvqJBd<
zY8@zPhGD2JJVtV9jBi}5Oy19D>+!y3lF!Tjjh2%QrFiE*WKZsvCEi)B0|2(ObQMcS
z?WMlnhcw;dgd%2$cAiO2+rq1&%|CR=F@HzwV|Ig9NbyVYPj{wNIz4ZP9pC+-)7Shq
z+!Lg4Jo5<+x?qJg<PH^_Ml8g{`m5}i#V-8XB@$~cta2(!g5O^)+V9R6>{<}<68@hM
z?Qa{7LM%n#=Q|SZu$cZXxsWJ98NJ5!Pp>CY-;HGE^qrP$9trJn+vS;a;;Bv3>;qO#
zk;&v-*?uLeT|a6G5fGJ{?;48=foW&Yik!G9Ge<O35PKY8Tl52X2&cBgxbP$j_j4yW
zd?ymfH$OU#OSos$PGkO1$sCve#Zh`uhU=O>-~X(%HGi4B=8M>0$h=OqR3a7(!0ir&
zuj+}fm?HKYl)#a=1IAl+XlC#YY&~%L#9qK`H!cRPMt7%(nlh3*?@vF?%HSU&#Q`Ve
z2zO=(Yk9jV%h{J~(VDKNtnh16ud$bM<+C|(5J}JazF4Umpk@~lU(uhDf~k$g^}IAp
z#mKzH>XQ*%kt57ABhs4r4`M0sMyv%e$+gC7gw$L0Sk(%Y{WD(OBV)|wO;>pP+{bX7
zVOGywI*sAZ)Bl9$FAPbF%^fK@FbJS?$xgd@^YsdwDoF_0NmHA%e{x(^88i;u^;Okr
zJa$rzoeYTtH%CcVjjGkoA2Y`&P}oUM31hHcAyTKM6X3+tp~#G}kGHIt@_X-^ulw(9
z>*~0A`&gu~%{;-|JgP_`H(aYguB-ubz;p+n_ZA5Y+t`RhBG%*XjX%5u|7J6{)i-O)
zmxHBDHPG1idz;hKVD<`?0y|%^JB2jwjbP3o8xht)g>8i5j21d-6#_^KnN9bSRMKxd
z(+!-<?R%-`xqPMA-bOdoj56kT86@=6{9rM;LEpFOEL#YpH(XO{{E=Fp{|GtDV(&N8
zG@ZoCKa27)4MFL^4d{nX|4|3K=%;l$dXqQg_j}YaXvXk=+|9^>6Af9gckS@VZI05b
zpWnqnYUV;BpQF_MkU{<>oiXyqTHVoE#n0Y*5>2XXcQV#j9{bs*v`pM2XIA>>sFy`v
zEtRUt3Tykm(YTp&;Is89qmZVB@tz)~f!_6Z=O@z_{cZt+!W<coh){{RH+_=9+yPrP
z#VO7D%PGw~Vgt9Y<4lWVC;!wib1Ul>7@){|o$d|5V&MF#TX|gj6OX~*WVjr+m30>X
z7`vT%m)33nzUNbtlb7#&l@9&Q2fpN{ajZoXY5_YK-qc$qoIoiDRTO$F*(#~%(=2p9
zl)-&%<pRKfc6YshSng2KMlpCNrEwR`{CZyiw2+qG6vEL(#|FrWtrK7w6HlHO;=|$C
z%X!j%eb?N?|CYDUM*<~gg|Xzp6|@Rusnm<Y$5?Mcjr+Kt&8^O{Xi8DBKyig!rt<|J
zCJVyLfdOsV?FZdyDV*i!zY&Uo@8L!yLyujx&5_Z^bQdQm&3JbQmA-JAdJWk!E1@l(
z@LtOP{d*-h(RYSIT<5Uwc1~xy1ThJ4Ptj_<%yav61g)6gKhVZ-&yO#V%r%Aksx0Ib
z6QInH;j&5;`Baw(!@ONg()P{wTJ4CsF1qEknpceWL5X1CUH`_fB(Z1O_M2D!%wejS
ze`6#$Oqcu`m0Sk%PM<wD)riuZP$=aD{yem;CI!hhgKu`NR-XlYy%eoZbW))({4z{M
z@FspXe8EV}$jK)?*nU{i!8Zjdo;RG`C*jB{NaDNO)g~&oG~A1;yG@%h>GVE^rleKB
z*h&p<87gIvrX(m&t*dFiN8L`^=A@m}R+j3+Kpq+;fZ62Eq&|HV=2E)0-1dhqzaB`?
zuo|*VG@(D_BGR<kuaGJ@>&Z1BE(Ag%sxIq>;%Eo1?p_|6m*S>3I)WtvCA*>L3u3{&
zS#5UvI<M6#!A$#$s10G?g13L}B47ouO8%j1cNj2cbCSU;1{p2{^?3hD1Y82MdV|Jd
zj-NyH5e0d)pmLSyB<5Gbm*=439p=*^_ju|#-U=?0DOMKlhhNmYn!{>DcGi?}U)y|Z
zW!UDcnm?wWA<Rh{LnMr|=S604n_hW)Xzz|sY6fe_!8;226+*mL=vXgUlervQ7){}x
z4GQHQL8cVpBFL;=HlYd|9Apcj4c<IH{YSyK)vMC&Shm0?cV<Z)(LuBJG>NG7wfO!5
zl1h8>thL@$qSSuRhq8iirTy;H_31}?kH7`F0y@MfwwX*^xj}kriLnps`}V~GoDtk~
zyr}?9Xgyt!bZ*yU6S#Uxm)ei>t(Hth>>T>~<SC+}IG>@GqXi6J6&6dg^CkHV$Yind
zAL||89#@&6{5}S;v!IleIK#o~cX*<|y9VwqCs+r*H%U^VS6^a(N9u9EM9$IsS*F^J
zQ)c^NT9l@)q-DNx=11q|d98PW@p|^`A7;aEiF0tj68ERso~7U@YsR$xMJXmvShaF4
zohY)^>kyqat<@i8);UDmfwJ4^ApGFcw`(U0q^G;T%W7l6B9LkSr5Z&~F7hFiDtp@}
zu0K7D^c=@%T_slt33NuU=IlN*rB%4|tT0BYj@vK@7ZG;Jdj}jWA@=zluPsxPKO}!T
zz3NoO^lk#j8_tjFVlXi#9%gfm2ouC)g&biV#(}wE201l;Ct!FJH#W1Uolz)Szx}}@
z6@sUv>|r1=>S>soH8+9B1nqp28yL<oq4fgcMOn3{Rva+4rUXh1d8801XyFWHAuCPE
z9VWy>!(Xo5H#G>91s50%By%@3$3_2I?2=^QykCe{I#*2_R^WqK?`Us7O`p)@J1j+j
zli^P5#YwAwXoGRKmQ>BE3qgB6z^pRUUGt(jvi!;s;nUOdw(K-7ESAwkt@Xymy^ZDL
z)bl2qn=(FOYM#$n0`Djf&1Yy_9Mo0?i*u=wdsqimmYe3$<|)I*;o`-0UWxy%=CE#)
zCEemE);26GTh2`Bo$MvcrQVVh?6dc4v7f_NZTV4Gq(`q<AW@KDwZxyFyUwpqiPW^d
zd~i|d-uXwLqjbWl5=3D3et9>&dOfvvzm_E1vTpH;P1*yA`_AVEZi6Ph6nlQ$g9<fo
z?PDj?fphg$ZW7UN$$4!16gy93Q*VAKcuiC?8!-7Jh9<b4zV<<VjbZA-SN6lXG}gHR
zPpCcLO(6t#nQYT=42^Kv=v8a+TG7_GdK^wg>0|U1g~n@Iy66mfqzs%=<Hl}oHitdB
zR4yHonFYNYqGLYefAeb}&8J$aq+Ua*vN)yQa!}2}<&r9IdH$`O3H7!LiTCZk^(uI`
ze8>y^LpNT~YUC+?m*7cv+RHk-qH=y9vN=hA+ma?CB!P)z`fS73`6N~VdXMKh3lE)9
z=pDZHm%KenPcjibnL}cm5R3ZDDga~$DGiD`z$=dR&F$RK^P?kBNw-LSu`LjDb_MI^
z8N*S-M{VmL@4B_Bc-fn?_L74e)N!9fitbs#Th;v@YF25!)@B{+rajnFfRtVkbgIJY
zH%8o%yYH^aJ|AS6$x(rk<dRBF$CP-i(I^fcLZVl(Y`Rx;jSPl;2SU$9|L(t8d6B~<
zZ_^c$!#?jU$A`nC-iD|(z#bAMn|l{s+ihcbLQ$XkRXVA+T;SwN#*mYP;;*anD6IfK
zw~A$2Q>#A13$pDuXFTk)T{!r$rj@@GBQ+5Z`v6@kFKqq<B$T;-4c|UCNEMX%)&G)j
zTpolr?>4r|x1dV1QM|ykcg+`Z%LrP~ukmF_G;qv+%G5gJ&D-pqrrYwk%E;q)0q;J&
zlql307cp=iL-)cd>)vpox@5orJOBO3g&WIr@ta(==SfzY*KS$oP;lRQbux(O#5v4-
z-|65WG8rG8fw}((=v?t(fJwSv#fOIYxvjKMLF>GX6V^#uuT4HN;N?cVnKM}5fJ_qR
z(<bL~qmffP2$e<7mQ9Ja-6~3!vu^60HK!Ega(ofn_RIv9od}(@xrEr`;CEm17rnxY
zE{92bJg5>iX!T1UT9G!T(@SgN!||?+T(NMgqQWlgHr&4_GV@JH<`33wbw>dAdNmKg
zfH-Sm>Kx9Xtu`gzPHCwrHJQ5mPrp2SsjdX=q;}MBu=K8G&Z{<a8|Se8@fQ}`tuA}C
z_Wd%8ZM?U<=X{UhvwctrotT1G>rU%I_1{|Uk{7_pvX$7l=cy(9$KE(o4qT-vnu(;3
zNK{v@aWBt%BT>D}n5dZ@aZg%NBZ#bgcmPCr>be30)GJR3CO@L`*PFbqh=|gqN6sP0
zpJakw4s+cGylZp<AtVtf)O#cGVV&LLmRx6YNzozRz)}V8*B#GX%7RR;`>y8e#vQ9g
zOkaB5rT3A}$Vyy_ndjANLqwpSerqfR>WvCDPp=3~PR(-snD@9$&C*9_`yf*m|Ec#^
z!orh(+H|pADc|}XFWWMMt%mQjey==2O0@osvR^f9p#C>J3HsV&nN8&sg=TfnP2awk
z*LPQlj2OirMv1VBCW`z$CkpCmsA8H#n%cXc-Sz<=LwQp);&Hwr8S!$gXhe2f@>EO!
zM0_O?U_AUBy0lTgU$J^6!#c=YtWmv_qpEg_ib4-kr0&cvkQt<3zXKeSe2$*T)aZz1
z6slD7C8!g7(@&<*2CuXLRKGOjs3AgyE{fkB57$Q1xPGO0@>*GE7^cfWBJM!`NZptJ
z&QZIWPwD_9ek+pM#%1yIZaVAzBS;Mw9XHwg@_5I~6a#FjBlNV22J8xFSq-+L4F%_F
z*CZR3^E03kPiq9TSvasZxEzQP)~R|RB{*XGuv-qf0V(+Vdka05y{OxeeXqI(_Jk#j
zTm9M)Nq`gPvP|Z364T9(2;?|qspZ>!E5${`HsH8)WK0iPTt5vC)w`4XqJUQe@#a;z
zPneQf)!+;;(n<d1HISq1{K?tAb0`N=IBciOKXCn6`&f6LrS(v&@^e8K6&<QLFd3S(
zK|;m(A26*a?1nt?nA4aT$t5TS1!FcT=3&O<{FCHaqAqitnX`wNlQRZd-#?BR;7oq5
z*p646!r$~`D3I~qsx44Wu`_MfJniTnwMHtXQ_j3Du-S(fnLiTp{=yWz{oDD1o`j#w
zihBNNmk9IabVjVy7ZEh3C{qw#U0!#S`#v&!j5bQ)<mGkwZQ|H84>8DtqHuL)$3hvs
ziZFA9D!^UIII|NTRz-DVZiR)Zgj??RQPCS4Eg%?9Zas2zFCyPV!*3mb5;=p=38r%8
zbM@<Qg#?ToF7*?9f4&!}j1s%}F4F^b^Sxu`3+mJmzl=7qo8)^b>&)s7N@ZgKUxZJF
z;YakKW}fbCqUZdquV*KPg{kRh1rE}HFbsDaMoaJXzGT=X8=Eg=FPoPai#k_-hs)Tw
zE9zbQ7D50(gxKmo^U9)bD1{5^jj>WqRqU2AnIeIwxh4OqW^-)Q6kqN0>Lz<~@%l2i
z)IItV%Vg-*erZ4n^bSAbv_CerZ*?-ET2(VBurkqx1W{S)ZeJin<W+LFYSiZWs*$DL
z4$m1@Z<N9@#SK(hv=$4v90p1?x9?I}Pu0kLNpp5XMcp-UB;~Sy=SlTX^h7~Q9Hz!i
zNBH6&1J1pFzXxB^i$QSp|M$lV_W=@l>&f^>vGmuOnbq*E5dNp`Zf^Y{DjWs;Jq{}=
z+f9uHnkl)nRO!Mm+n@4RsZhJc2-dV)6s}L!_XDQA-{;hJUHi~8Z#Ma-c^m-q^O;Z3
zvB^~lemqg<p%8}{>qC{Dl5%yQ1S(obR^ShOBvoS9p3ho#T_p_6S3<E6j1HC_DEALx
z_L}$Vj5DOWKvvD)xRevP*3IUQ#S>t24VoJ_Zah9c+};QLsGn4lP4}vr_icR2co?EK
z!tRX*pPey%Zn~1rf7m-+EZW$U-k!>5vk@bDI}?B4sP7B>CFA5=aeRR%>z`|42Uh^l
zDJPR-Vm*Th0Z%L4d)h`Dau2|)b?X2^x)rdy%w8nM=BA%ihK)KXW9X9nR&6jplDgZ}
zinV3>OwCA`v{*vQL6}8e*+KLr*;}{z*-r_eG6+#xf0mY2J|sApu6yO4WZ4C-*UIz~
zOFOocvfaENaki6t@ewd<ha_4KXAMh(u4}rEjE*h^F6YGan-~Tw)khy*i4j#bsk_D^
zvmM-^|GF&CuP((uNpZz1W%5LCez6OLn0Sm2F!vp!6wJCuz@gUxbX)Q2>sbmp{&)~Z
zI|f0sOS28_dg$sPQgvw?(D>WlCPx@UrG+<zewJ<wvuRMES(-r5{@*G1c;)J0eGK>0
zE}9{wxmPle3I5_2n-?b+)oQX@)O1LB_qrLfUoLO|E;mS%`1IXaaSSlxUVpzfG-P=Y
zpAnc6p=pL-kxD!7tD%+LJ`kJa7m>5pO^q{8DVktv6?rwGy91mSKfjHI_BHsdKm1c(
z6HBme9m{@te$G(1J#20=MxYjkl2BZ<k#jvJ1WZ^dOOK}IbIXmss#&TjqnRe-Af<^E
zUoKiXd+wgeV&D?0FG|>098{ow4(4J|ME3UqNhoCmS$=(=NZHSe(}4@8eE)BMjbf?e
zKobZypD{$*SX{{N+qKs5hh+#nyomp{5;i^oca<Kl?qn?=DXc}R?NbPE@K?p2+j}y4
z;2ne6>qT28f|MldCIkLd_}@y*nArs7Dj(28Vt^V``$|;CDHJd$LF<X)1^H!w1y2v0
ztEk%doql3wR-MFrh0VvRF(6WZo3sGITr9HGa^M=XTqqace?NC0Q7dWtmk@(i6f=+i
z4@>KB8qD08z})CFf10hdWQASu-VYfChHbhPA+3ncY4aJ#U-8!_bu_iN`F?`3C5E<<
zTeYxot58S`xY;yL&=vBksx}JVFOml)0x!=9sWa5ZZzfX4W7n15SR*KP8vu1}XyLBn
zABjMpi$46gAka7wWxepiPcGvwV;67J?DCEOyT)CZa!3!=Roh=Jf^w6>FC09Im^4y~
zCzD{8BXB$3Y(j}rxHnSM?TX;`Iijf)6))Y=PC3P;iXX6?vE<}p4vkD`+yHl6h;4I*
zXaPXPcruG~0v^UX6j^qsu|nc9`T3ggVxe49#gq28q53aRxbekGDy{f5JAd;m5xE7E
zko14%*8^aJgn_Hjm{)6m=0N%k<}6>=&G?Oogv=?_xM1Phc#?_A7?@?2ll{-F*=eKa
zprOJ)j){`>kp(>I;PU4IK<qDkwTX$8rW&_omTlm{r0*|KYH<Z{+nY0BvN5nd07z<0
zN@@HBfCl&Yya|mjkxllk^!E1l>0+{;Kgl6v4)1x4{euBXbPa$-afIplr-C=;oSTi&
zY0wWB-99SsA)!@@KgcBjg!^F0{5s!`3;jtjCanrW&KJFQ>kAa}elF&bLf#9MBkO>o
zShEg_bN=v~(BnSqmiw>44lv~K(CVQgCrfbX5*z^KC!H2mPmtR#nTG_|BbWtDa<Y2Q
z0W&VeIZyZx0rAap$El?i9D&vX{S!LE!6Fw8)zq1&K_+sQn=%;pu1(!BU=4oUG@U#h
z6BvJ<ryh=Xt!ET8#yY0xgg7sZO&c8ie@+Cl;~R&8GVKgMMe_C1X;n`LL3zmm%a@MV
zIX?)!7(Ji9I0^ONkLCZIt{?<FSZVrl54H{*qn81JI9_i6EZ+Oz?6aO?>=P~{<ZfOz
z)5jy9364ekHMwVEj6l(H&?<8v98+){1UntE4_wb;!Oig5stif{ZQWD+{fy)>*t{RL
znBko*z!Y=gbRF2t%IuKp%|+Zny3#7|n*O@~Qv7iwlU2`&NbB*Bs_~1n>5GLwo)~dw
z)w30o9V}nL-+`dHocH3ZY*HTWqe8|N-G=Sws;-{yU7?3F=p7=p<nZA4^Gpn4?Ju61
zOr|l0LQ3hLpeXClrVd)tdc}{rUrA)xMRMZ<x9-NQ={YS8|561L3Aa9_DWDwE%1sUi
zm5+W*QR{wf)fh|5R%Dh6IgBIpbbEjKZCrnK?lp)}M(|mZ@;}=h7(D!YO!c74)vcsn
zJ_c5@%9+i92mSLDJhD%Je_?+yR!U$$r{1|wc{0n=njW3huBO`pTyn>Ox87b*7Y+QZ
zgX=ur`4DBYLDUO~rixEM@fIss9kQf2-Mh_=q7wgTDj;PGxmibubU&T#|8h3FQetMK
zVx&7w#`ZYAm#q^GL#5xg_2^>qI;N9|Vw`^ob~0W5Qf{cqK;P}f&5(n}qjNfYO8+;n
z`q(18UqAWO>%?qUsHB@N&QE=S*GxKGesQWca+4SAq`NskEP|p0B3SZ!AKyGt@?I?$
ze43t-?KE=e+j?SmZr7vgubno*#c_YJFL+3<GZpTH^zb%uAfq>$jUNEv|1CzJ%6W1-
zd}E$TANiX_`}l!xw|-$?P_rg}w^GqFKic0j*(ZbtC~A<rgUjj*dmDEvp9CE-QWluJ
z^%NjwGom9jKZxyGw;o#xBt?Y%&O+hz{8{-i!tl!IjKU)9sq2c^!Tlp~j0q%wPPgY8
zn)2qK9{gPj9+}bt^4kF2ySmOcefR>H)!)x}Ov83dOj4X*nVzXK$R0s(&p(Q!ppS#&
zK*f2n>3qc`wC_mZiHbry-Yg}f5LBcBE1v-VVMatk8o^|VYw9h3C76z^b--EurxVd>
z)Z^+-iv{zXbPh!CN*Fn(yw{4DKe&4;szdbExp?LZ+88Y8jrz_o4(HMr5_p|B+oPNn
zKm>R|U?A+Q-6XwHt-~tBd$d?cghb|#nNv|BT7?i!v?chp@4=wbzaGUK+oQ=yTn{1w
z=fwJ{fIkb<&vtdKT{i)pg5c_j(4e}hHHv+lSc%&`NKp8{*N&8q)L<G#`q<lEGfUk}
z<SYPs=Px<>T+Jsd3DZ=+7QNJQSeVU_no(`L4+!_YHL-M`a!=<4Gi(V(VE-cX?K}N_
zA!_?~HEuw3=*)ZAri*t;S;Jf^Ye#GpRI$}Je|-XU_JzVmu_JaFBo-^5PFYpA_9Q|L
zcFg_5MD+~SPjzI;o*Ju>^HZ@;?q)``ldE{dA)sVu!L{2x5k=ekzuX68zQjAUYoNB%
z>`|eTb!nPOtOJQIJajhr?-h4HE0smXQeZZ<j_PP<q-6f|;^3*X`EP+ygwlYGM&g+x
zY=R`;AMA4Hd`kaFzNEC^{V~CTCh0&!cAVgoC~3{qF$>({!TnGP*MuTTQRLV^oAblJ
zcZ^pA*?Gv9xcXJ8Jdym8f!9$kEnqX%#4}a${2fD8azyaDJTR-?O$<(=v!Xe6EJ}JG
z!K;J`fCN1F8vAV?3A}o-GDtekjfxXpP1#<_&e#n3s9|tpCa3*M%)z%|%lG|w(&4MU
zEn1aOF&&NTc<qon+7tG%;vQx0?9A<yr(EqY*^GMeXo2%S1L3;Bd@xN&CrBpHMq!z~
zW+fXU=5O+XK=xpfzu#VU{_>fp5JRBq5It#P^E>ukl9K(EV^*W~MDnP4z#oy@@zp5c
znvlmvXISW})HF94a*Lm)NX%~jCzg%;H<o<_^r>3c=;)1*U$>n7g)nd}zPgyGp8gbf
z)cNU>>VQmAJWCX}Oh!(SP`NJkd@QRDQecdwPeq4<t<!3T9yz8dOZgcvOz*UYoJo6>
z^C;>&%<g<iImLWCftf+>V=bg#>MH!47M@-0c*OjHFSGEHyYu2MSgK{VP|>ihYMo@b
zV-JT@?wC<~GzEk4Q8)F_(k^`c1EM>vBY3ci_A956iR@9ZLT%c2mu>3?Y2M{|z@)*p
z(qg>EtYbC`b-K0>DE|p+HU7;j{;~m0bm-vkcry97MWY2Gwxb1e-rF|^EM(|W+`z~{
zR65YKxmrKby@qVosULYj`TmiB4mQv83mwMt?2P+1)H}8LEuqzl$`)B`;_s(SZNPs^
zm(M9CsCt4qzo+?VB06HsGBvVVHK{3Ha9>2DG^k+$tsD<d)*>3Ahm<1ZxN%M+$`gGP
zf=6mAeV`y5k6EkegULI%eJK|p-<KlMAE&zjM~PeN(*+zHDRT5p_(4WdWmU`Z!<-6j
zE;<Kes*!5jQ+j;imM$TWgaT_>GLSmms~oZdeQ0HEcrzoy@3>rWVIDx-Fk&@^u_R+o
zCJ^Ekbtza69TIf4bL7D$s;|+}EX;uA$a|l>#F16>9LtHns!Nef8-5QR5gpab#e*#u
zLQDc`2)<1NBhmO+paN^F!R#HX*`RJLKNA60VaZ29zKI1!`qienRPTO#xKiHN#sH@w
znmwOKvs&%f9($$`%PU1?IQ4Z;8APdej-z)p8JlH9|D5%r2dLbz@A0@x`Fc6>hfxiH
zA6qqS^4~9WQ!qaoFwgZhCAjt1=H^gl34iqn7>Gk;DSRk57*4q#_0iTuV405t)llg}
ze0CJ$pm>&c;C$SqtC^@7z+DnloCG(;{YF_CoBVLrpi_C&PNcSmj>wzQwy)z79QF)d
zCrPR7;pO`q-j?Wce46YNB~bGTD<9*M?`48ZU)Q3Y4{Od3n$j%b46T%=`g!z1pm+$M
zZQA$cv-(k1OnW{{t4<QIaib$ml{!2*tF)!ph1N{WM=Alh@Eaua*?aqfi+^q^0&|9N
zQ`x@d&!<$N<5o@s2H}SJE3_nd(oV*gp#KT~rZ6APJT(tMzkV$s-ms^WK*Htq-eZiA
zy#$mbEjp(BDbo<2DxGXgeQ%CwF1ci%sE0zAm4PKBaX!J_v-*@V@hi1RpeDqN;Gd6Q
zd$S9JA9j*2pq~kwM5%gPY1TBU&hA9ItUT6}-DY%a+SdLBC`X>0>D~wv!2_!sDo=<M
z=U?+zW?4}AFR2OcJBf(7y0r4wBn1T~4PgE9g=;p#*))^VXn8W(r!;eFCq`fg5VwD(
zi$Lf6&)Ng#y)!q(l6E!Cr$1dI3e)9&9wk6NL{|({O&Ma<6G@;GD@@ag?o{bRf)KmL
z^;R85PcnLD1#=sQTr$BT{5KN{CYiW-cMwDOYtrQRy~>*PkZezqkO{%j7Em{rVEZI7
zfd>G3?dy1!EbyG8CyupS;hPIykAA<xGn9sur)ktY<JOn{CB$LBE5Yw~$mvCWjLSpq
zBo<;iqr&2+=bSiInGq~;lA!U&U_g4_*)+JWzMZW7-><Q)7gf|l!h_?F-}r0PHYxgQ
zXBmi2GsOI#ix8WQ-}T%vWB3NnY*|K*qnDR>Ji{kq*%hP*$ncX!Q39RuXK`55)n{Ch
zI(KsVrc99Xg<K}G!76tZy+_a;H1ohKJ+4SFbAj`l;%1W!quRy*r`djr#`A4+C<BoU
znt~7}eggOQlJvpy5O413pN<pdOdw0&_tzSHD=B&GtPPyOX(&XZ1&|kPghMTF4p4ju
z&%by>oZg7r^UK<e8f%`+wIRNoC<^NYx~`e`?8Ztz3~w0RGJ&q}1p<;HGJ0uFzbSX^
z`NW-um9&1q@{(gT3X>}M6KRg@Wi;@GJc=&>sl51-Wli{(>0DOheH~AmvfYjuX!10p
zP-iX_<ET?0e)#nHTINtWlv)>mI2X6X-#C%G(1>l<SmS<uXs!}UODYxED|b`x^+FR}
zJ3X&9mZ6eA9|-FDgwKTfC;oE~qd-rD|J@hD8bo0bYPWc;Q)@UPr3q_6n$M>YPenTW
zfe^rSpnh8SIN!RHe?`Ge$>cuWL{mMVd>sV5a^%iE0mkhDdTHI9bC5YK?OsEdZ5U+%
zp5iY85AsoEJ+g&`Z__07_&ZiPvyaUhDGT|af0Ar&oWZXujgk1cpqLpSochHxkJkSu
zKr--4)`KKa9?1w^u0F;Et!b=4O<FZTH~b}+fxb@3K>mUQI0^p<9wMn}z$MHk2g*V+
zhbhlKF|?QEa1+3vWkP+FkSW(|-vu>sJBXNAcvtRwcIgi4H$4&Um{__^kOzDsgpY8e
zeP-#h-j=q~Ed(^an37Qd>&X=<94$cjqdBry^i4q7+C8#7K`=E1FU*X)YzxsYAaT84
zB0ilGA+N8>lbOTKo8=Vvg@}Cqqe5-!N3H5-cu3>#XylJq0yB`nQ?jgh48t8{T81*h
z2gj(O>2U`SIyDM96H^{V?m7V0#4Vq-pKzR8v`aoo#M22_KpXAZTQB@`W2GSrzjjCi
zEyMp!`2no;TZa^jO+H(Y4DFq#e@^Nu6usCP&A6jifsLYnaTWpyy56y(vk<e{O;{6v
zf)p8ffaRCTJYds3Py|}8YDGS4zQnqC;aMrhzBcGu<NUl;>fa)EmAdshNy-Il{QT6e
z9!#Lk2(4?4;k86PxV(6+Lu*LJPyfV-GlFNmferjzR3q{|Y9HuQGM{KO-bZt-NbmHl
z)UG9n9faK9_X(aN1*ecEHU>B=XR92|Cn>PYV8q}HzVd+LBICc0356_0Aha!~Zy<yv
zNfTK}p{!Q^LeT<F@bkWKRH?iT^p8ol)<x!!h5Q%G|9X8FrW=O_q}X^8FsqIkdw5NN
zxk%AnLw0gA1MKnX`73hLtHgbru%0U?k{Lf3K0J?7XD1|F>@3xk8E}9=>^Q>+6i-!3
z{1+11c@%dpc03z+4)_4$g+-ItEjSGg6OVEN6E;NqI~Q5OsV9?WGF=qi^n??VPMRRk
zbV}5;2T?X?bh^+Hb|=PwmBk^*;}595x99+CX8+kfLx|wG^lQozfl%jcZ!3E<jEe4V
z=Ni9E!@R@7Hn6I>IcK9KdsIN>|I6ps6B&q8?h2t8F<1SHxxKes1nRVyK7H*u*8LM8
z$|5-ujD^pK;b1s@C#@Tj*zf>>n-6Xjd-<XV&tbHhs9QeUC-$GX>R-GTfQ83@ZLz$E
z(cFPHu9lYdYshZz$oPnbBLw2dly$F>hIb&+?H;&_Ilba968qsaVtNp<>l!C`aJVSf
zs<6Y&nTqp-iW+$8k>__YSW*a|+ir&s#?!cpOyU(Y>0r7NA$>K^So8UaugrvtNn>>2
zhar6aLpaNcUP1q`cr$^}>vU+mnur#$H!Ti88h>^i_2L!j<x=ba@y(QlQ7bt59B$1h
z&=3!4ex3*=m@T6Z)MOXpbkpb*c0YhDbG|9hV52LTyJS9>h3Hh}&;uc0V}Mk4waBD9
zOLUR`E~d<obC!dZe;`P6ygA((dY+xm0Zyli9SjB1s*t6f(F&mk&SdVv{{^L<6SW9l
z%~7j9O&NH)m2Chn2m{Te_jH#lvjna{!cc@ZwEoe8GQRPQ-!eUxsdI_^LHl83y<|-k
zoegpv1X-knrft$U5x3~`s|Pix5Y^-#I<JC}pK}jAx!d^rIn^40_UX?pzEH-O30!Jk
z6>W&hd=bNSRGT18fd$A<Tjtp$T0A%LRbiy|PCJz(7UPE*crEp|M9CwCHSsFZKVg6Z
zx7#uqhaQ25nM8o56(lJ%lVAZo<eDBJB6(dXMuDUZAyCTa<1_Si#X8eJ3X>ys1<A?O
zA3`$XpGM`HkVB@iE@lFFV0^#VD4O-zYY#XxvKnPW%bwcywqpP5U{uhExZHke=HBnP
zZ2qkH975;70o`h`q=hsdDxT{U+lOSl^DnmuVK>ddW9IgpP6C{r@zl?=;IQsPoX77k
zVFa%I;=+fZqVTbLfZa(HquHF25%ppZA_rZ3z7=<2S?4L9cbJ*?aL*@6HGJL>Rz%!}
zVyUZ#fej>`O%#MQ{Z#EE_cRvjd2#bW_dKh4U(cD8mn06P53t#28Nj1z3Qasq1yaig
zBaFvC$|W%oDRQYed%zX0XOF*|I_)_ieG#>m_2Pu!ro!LKZmemGA83|0Ts~;M2o_k0
z*WF$Eo+zYDqjVoGzCA=@kmey6JG*iYoVS(<E+f;l&F?MJJRgmGUt&tw_%9q2<d~;m
z7_>dVZs@&hKa!%*)R%cMHHj8{h5H?o?5L&G)OgCR&Er0SDYkwgnB}l2+%}bXcg6n7
zd4u~(Wstv|yLC?U>-Jhc5jIx?NnOjJvFNvc6v!$;F8lMTZ!c#>LRR9EJt&K17aYc!
zMt=R*|6QR$5=`l1az%;$w}`5w%GJ%PL&OBGnOah2hTl`P!@tc4Au?CXr&75VQ!M-c
zcB;TGw)Nx3ny66VCM^BGQNo8RF1<Lbw213@r#^5(r)Rrx+{>*ZZPlmr%-1n}TUDZR
zZ=DG^bHIAjYn__w1e+&`R2VgcOk|%K%@y4CwM_rh<o%zXT+Y_i3p(~x>xY3{(RBIn
zqgVI-wXKcY6Wg@=&z?x7gm1=ECatnq+VMiMZraV~ya(f+UFWWw*R!o8eyMW3zs=KQ
zp&tqjep@Wn=mj^wq0J13)dGPZQk?QagML--KBFXYFZTPS{nL%soCz&sxx{o+$n#U;
z&PU5CQvY0%eGoTkxt8MHPxC?xql$o49+&cNmhyLjs&={g)Ap7;_nIf_)@1!MF>Ke;
z{nLxYHf!a)eD`_v%<#@Tsx9cFr9Y(JI5Q}ys=oa6#NpxIy;HQx?|Q#HEo-d8viY5|
z{iT&cKQ&{#oo<CzoBXY1_;+ghozoUKME&bxo~|=mX=7)Xwp91T+|_Tq+%u;x`mMS5
zliiVUi<zm@>qBpO1((g=Hr;5?9k2gd2j~A*p0Tt2OqVoreVrky#2A{gK!x+3-o$84
zuiYYoJU0y=FRQMwU$V5C)2`>cyT@sHKb4TX?s-M}j}wlczwci5^PyX$z|w^VF||K}
zp6Fk+RVmUxoE#=GQ<Tkk!t6&HHJV+{67g(1e-{Wn1<n?n-BcV}@$~3#cI1|5R4a$0
z<puq_sbYo#H5U47e=5$4sn{*vKkZM4`lTSzw$=N^17zf{d5d;5bIfo(W+TE^{%)ZG
z$GcOj4{6Lb(mguS;IPLGEe#*<1ODGd47B-$@~2AzuW4>H*jFPof7>bT2Yq+jpF-v$
ypkovuPXl9mP)weJQXJ4ZgK=ILhtt{(e*b4I^w(J~sVGy(00f?{elF{r5}E*VC68$U

literal 0
HcmV?d00001

diff --git a/_static/scripts/bootstrap.js b/_static/scripts/bootstrap.js
new file mode 100644
index 000000000..c8178debb
--- /dev/null
+++ b/_static/scripts/bootstrap.js
@@ -0,0 +1,3 @@
+/*! For license information please see bootstrap.js.LICENSE.txt */
+(()=>{"use strict";var t={d:(e,i)=>{for(var n in i)t.o(i,n)&&!t.o(e,n)&&Object.defineProperty(e,n,{enumerable:!0,get:i[n]})},o:(t,e)=>Object.prototype.hasOwnProperty.call(t,e),r:t=>{"undefined"!=typeof Symbol&&Symbol.toStringTag&&Object.defineProperty(t,Symbol.toStringTag,{value:"Module"}),Object.defineProperty(t,"__esModule",{value:!0})}},e={};t.r(e),t.d(e,{afterMain:()=>E,afterRead:()=>v,afterWrite:()=>C,applyStyles:()=>$,arrow:()=>J,auto:()=>a,basePlacements:()=>l,beforeMain:()=>y,beforeRead:()=>_,beforeWrite:()=>A,bottom:()=>s,clippingParents:()=>d,computeStyles:()=>it,createPopper:()=>Dt,createPopperBase:()=>St,createPopperLite:()=>$t,detectOverflow:()=>_t,end:()=>h,eventListeners:()=>st,flip:()=>bt,hide:()=>wt,left:()=>r,main:()=>w,modifierPhases:()=>O,offset:()=>Et,placements:()=>g,popper:()=>f,popperGenerator:()=>Lt,popperOffsets:()=>At,preventOverflow:()=>Tt,read:()=>b,reference:()=>p,right:()=>o,start:()=>c,top:()=>n,variationPlacements:()=>m,viewport:()=>u,write:()=>T});var i={};t.r(i),t.d(i,{Alert:()=>Oe,Button:()=>ke,Carousel:()=>li,Collapse:()=>Ei,Dropdown:()=>Ki,Modal:()=>Ln,Offcanvas:()=>Kn,Popover:()=>bs,ScrollSpy:()=>Ls,Tab:()=>Js,Toast:()=>po,Tooltip:()=>fs});var n="top",s="bottom",o="right",r="left",a="auto",l=[n,s,o,r],c="start",h="end",d="clippingParents",u="viewport",f="popper",p="reference",m=l.reduce((function(t,e){return t.concat([e+"-"+c,e+"-"+h])}),[]),g=[].concat(l,[a]).reduce((function(t,e){return t.concat([e,e+"-"+c,e+"-"+h])}),[]),_="beforeRead",b="read",v="afterRead",y="beforeMain",w="main",E="afterMain",A="beforeWrite",T="write",C="afterWrite",O=[_,b,v,y,w,E,A,T,C];function x(t){return t?(t.nodeName||"").toLowerCase():null}function k(t){if(null==t)return window;if("[object Window]"!==t.toString()){var e=t.ownerDocument;return e&&e.defaultView||window}return t}function L(t){return t instanceof k(t).Element||t instanceof Element}function S(t){return t instanceof k(t).HTMLElement||t instanceof HTMLElement}function D(t){return"undefined"!=typeof ShadowRoot&&(t instanceof k(t).ShadowRoot||t instanceof ShadowRoot)}const $={name:"applyStyles",enabled:!0,phase:"write",fn:function(t){var e=t.state;Object.keys(e.elements).forEach((function(t){var i=e.styles[t]||{},n=e.attributes[t]||{},s=e.elements[t];S(s)&&x(s)&&(Object.assign(s.style,i),Object.keys(n).forEach((function(t){var e=n[t];!1===e?s.removeAttribute(t):s.setAttribute(t,!0===e?"":e)})))}))},effect:function(t){var e=t.state,i={popper:{position:e.options.strategy,left:"0",top:"0",margin:"0"},arrow:{position:"absolute"},reference:{}};return Object.assign(e.elements.popper.style,i.popper),e.styles=i,e.elements.arrow&&Object.assign(e.elements.arrow.style,i.arrow),function(){Object.keys(e.elements).forEach((function(t){var n=e.elements[t],s=e.attributes[t]||{},o=Object.keys(e.styles.hasOwnProperty(t)?e.styles[t]:i[t]).reduce((function(t,e){return t[e]="",t}),{});S(n)&&x(n)&&(Object.assign(n.style,o),Object.keys(s).forEach((function(t){n.removeAttribute(t)})))}))}},requires:["computeStyles"]};function I(t){return t.split("-")[0]}var N=Math.max,P=Math.min,M=Math.round;function j(){var t=navigator.userAgentData;return null!=t&&t.brands&&Array.isArray(t.brands)?t.brands.map((function(t){return t.brand+"/"+t.version})).join(" "):navigator.userAgent}function F(){return!/^((?!chrome|android).)*safari/i.test(j())}function H(t,e,i){void 0===e&&(e=!1),void 0===i&&(i=!1);var n=t.getBoundingClientRect(),s=1,o=1;e&&S(t)&&(s=t.offsetWidth>0&&M(n.width)/t.offsetWidth||1,o=t.offsetHeight>0&&M(n.height)/t.offsetHeight||1);var r=(L(t)?k(t):window).visualViewport,a=!F()&&i,l=(n.left+(a&&r?r.offsetLeft:0))/s,c=(n.top+(a&&r?r.offsetTop:0))/o,h=n.width/s,d=n.height/o;return{width:h,height:d,top:c,right:l+h,bottom:c+d,left:l,x:l,y:c}}function B(t){var e=H(t),i=t.offsetWidth,n=t.offsetHeight;return Math.abs(e.width-i)<=1&&(i=e.width),Math.abs(e.height-n)<=1&&(n=e.height),{x:t.offsetLeft,y:t.offsetTop,width:i,height:n}}function W(t,e){var i=e.getRootNode&&e.getRootNode();if(t.contains(e))return!0;if(i&&D(i)){var n=e;do{if(n&&t.isSameNode(n))return!0;n=n.parentNode||n.host}while(n)}return!1}function z(t){return k(t).getComputedStyle(t)}function R(t){return["table","td","th"].indexOf(x(t))>=0}function q(t){return((L(t)?t.ownerDocument:t.document)||window.document).documentElement}function V(t){return"html"===x(t)?t:t.assignedSlot||t.parentNode||(D(t)?t.host:null)||q(t)}function Y(t){return S(t)&&"fixed"!==z(t).position?t.offsetParent:null}function K(t){for(var e=k(t),i=Y(t);i&&R(i)&&"static"===z(i).position;)i=Y(i);return i&&("html"===x(i)||"body"===x(i)&&"static"===z(i).position)?e:i||function(t){var e=/firefox/i.test(j());if(/Trident/i.test(j())&&S(t)&&"fixed"===z(t).position)return null;var i=V(t);for(D(i)&&(i=i.host);S(i)&&["html","body"].indexOf(x(i))<0;){var n=z(i);if("none"!==n.transform||"none"!==n.perspective||"paint"===n.contain||-1!==["transform","perspective"].indexOf(n.willChange)||e&&"filter"===n.willChange||e&&n.filter&&"none"!==n.filter)return i;i=i.parentNode}return null}(t)||e}function Q(t){return["top","bottom"].indexOf(t)>=0?"x":"y"}function X(t,e,i){return N(t,P(e,i))}function U(t){return Object.assign({},{top:0,right:0,bottom:0,left:0},t)}function G(t,e){return e.reduce((function(e,i){return e[i]=t,e}),{})}const J={name:"arrow",enabled:!0,phase:"main",fn:function(t){var e,i=t.state,a=t.name,c=t.options,h=i.elements.arrow,d=i.modifiersData.popperOffsets,u=I(i.placement),f=Q(u),p=[r,o].indexOf(u)>=0?"height":"width";if(h&&d){var m=function(t,e){return U("number"!=typeof(t="function"==typeof t?t(Object.assign({},e.rects,{placement:e.placement})):t)?t:G(t,l))}(c.padding,i),g=B(h),_="y"===f?n:r,b="y"===f?s:o,v=i.rects.reference[p]+i.rects.reference[f]-d[f]-i.rects.popper[p],y=d[f]-i.rects.reference[f],w=K(h),E=w?"y"===f?w.clientHeight||0:w.clientWidth||0:0,A=v/2-y/2,T=m[_],C=E-g[p]-m[b],O=E/2-g[p]/2+A,x=X(T,O,C),k=f;i.modifiersData[a]=((e={})[k]=x,e.centerOffset=x-O,e)}},effect:function(t){var e=t.state,i=t.options.element,n=void 0===i?"[data-popper-arrow]":i;null!=n&&("string"!=typeof n||(n=e.elements.popper.querySelector(n)))&&W(e.elements.popper,n)&&(e.elements.arrow=n)},requires:["popperOffsets"],requiresIfExists:["preventOverflow"]};function Z(t){return t.split("-")[1]}var tt={top:"auto",right:"auto",bottom:"auto",left:"auto"};function et(t){var e,i=t.popper,a=t.popperRect,l=t.placement,c=t.variation,d=t.offsets,u=t.position,f=t.gpuAcceleration,p=t.adaptive,m=t.roundOffsets,g=t.isFixed,_=d.x,b=void 0===_?0:_,v=d.y,y=void 0===v?0:v,w="function"==typeof m?m({x:b,y}):{x:b,y};b=w.x,y=w.y;var E=d.hasOwnProperty("x"),A=d.hasOwnProperty("y"),T=r,C=n,O=window;if(p){var x=K(i),L="clientHeight",S="clientWidth";x===k(i)&&"static"!==z(x=q(i)).position&&"absolute"===u&&(L="scrollHeight",S="scrollWidth"),(l===n||(l===r||l===o)&&c===h)&&(C=s,y-=(g&&x===O&&O.visualViewport?O.visualViewport.height:x[L])-a.height,y*=f?1:-1),l!==r&&(l!==n&&l!==s||c!==h)||(T=o,b-=(g&&x===O&&O.visualViewport?O.visualViewport.width:x[S])-a.width,b*=f?1:-1)}var D,$=Object.assign({position:u},p&&tt),I=!0===m?function(t,e){var i=t.x,n=t.y,s=e.devicePixelRatio||1;return{x:M(i*s)/s||0,y:M(n*s)/s||0}}({x:b,y},k(i)):{x:b,y};return b=I.x,y=I.y,f?Object.assign({},$,((D={})[C]=A?"0":"",D[T]=E?"0":"",D.transform=(O.devicePixelRatio||1)<=1?"translate("+b+"px, "+y+"px)":"translate3d("+b+"px, "+y+"px, 0)",D)):Object.assign({},$,((e={})[C]=A?y+"px":"",e[T]=E?b+"px":"",e.transform="",e))}const it={name:"computeStyles",enabled:!0,phase:"beforeWrite",fn:function(t){var e=t.state,i=t.options,n=i.gpuAcceleration,s=void 0===n||n,o=i.adaptive,r=void 0===o||o,a=i.roundOffsets,l=void 0===a||a,c={placement:I(e.placement),variation:Z(e.placement),popper:e.elements.popper,popperRect:e.rects.popper,gpuAcceleration:s,isFixed:"fixed"===e.options.strategy};null!=e.modifiersData.popperOffsets&&(e.styles.popper=Object.assign({},e.styles.popper,et(Object.assign({},c,{offsets:e.modifiersData.popperOffsets,position:e.options.strategy,adaptive:r,roundOffsets:l})))),null!=e.modifiersData.arrow&&(e.styles.arrow=Object.assign({},e.styles.arrow,et(Object.assign({},c,{offsets:e.modifiersData.arrow,position:"absolute",adaptive:!1,roundOffsets:l})))),e.attributes.popper=Object.assign({},e.attributes.popper,{"data-popper-placement":e.placement})},data:{}};var nt={passive:!0};const st={name:"eventListeners",enabled:!0,phase:"write",fn:function(){},effect:function(t){var e=t.state,i=t.instance,n=t.options,s=n.scroll,o=void 0===s||s,r=n.resize,a=void 0===r||r,l=k(e.elements.popper),c=[].concat(e.scrollParents.reference,e.scrollParents.popper);return o&&c.forEach((function(t){t.addEventListener("scroll",i.update,nt)})),a&&l.addEventListener("resize",i.update,nt),function(){o&&c.forEach((function(t){t.removeEventListener("scroll",i.update,nt)})),a&&l.removeEventListener("resize",i.update,nt)}},data:{}};var ot={left:"right",right:"left",bottom:"top",top:"bottom"};function rt(t){return t.replace(/left|right|bottom|top/g,(function(t){return ot[t]}))}var at={start:"end",end:"start"};function lt(t){return t.replace(/start|end/g,(function(t){return at[t]}))}function ct(t){var e=k(t);return{scrollLeft:e.pageXOffset,scrollTop:e.pageYOffset}}function ht(t){return H(q(t)).left+ct(t).scrollLeft}function dt(t){var e=z(t),i=e.overflow,n=e.overflowX,s=e.overflowY;return/auto|scroll|overlay|hidden/.test(i+s+n)}function ut(t){return["html","body","#document"].indexOf(x(t))>=0?t.ownerDocument.body:S(t)&&dt(t)?t:ut(V(t))}function ft(t,e){var i;void 0===e&&(e=[]);var n=ut(t),s=n===(null==(i=t.ownerDocument)?void 0:i.body),o=k(n),r=s?[o].concat(o.visualViewport||[],dt(n)?n:[]):n,a=e.concat(r);return s?a:a.concat(ft(V(r)))}function pt(t){return Object.assign({},t,{left:t.x,top:t.y,right:t.x+t.width,bottom:t.y+t.height})}function mt(t,e,i){return e===u?pt(function(t,e){var i=k(t),n=q(t),s=i.visualViewport,o=n.clientWidth,r=n.clientHeight,a=0,l=0;if(s){o=s.width,r=s.height;var c=F();(c||!c&&"fixed"===e)&&(a=s.offsetLeft,l=s.offsetTop)}return{width:o,height:r,x:a+ht(t),y:l}}(t,i)):L(e)?function(t,e){var i=H(t,!1,"fixed"===e);return i.top=i.top+t.clientTop,i.left=i.left+t.clientLeft,i.bottom=i.top+t.clientHeight,i.right=i.left+t.clientWidth,i.width=t.clientWidth,i.height=t.clientHeight,i.x=i.left,i.y=i.top,i}(e,i):pt(function(t){var e,i=q(t),n=ct(t),s=null==(e=t.ownerDocument)?void 0:e.body,o=N(i.scrollWidth,i.clientWidth,s?s.scrollWidth:0,s?s.clientWidth:0),r=N(i.scrollHeight,i.clientHeight,s?s.scrollHeight:0,s?s.clientHeight:0),a=-n.scrollLeft+ht(t),l=-n.scrollTop;return"rtl"===z(s||i).direction&&(a+=N(i.clientWidth,s?s.clientWidth:0)-o),{width:o,height:r,x:a,y:l}}(q(t)))}function gt(t){var e,i=t.reference,a=t.element,l=t.placement,d=l?I(l):null,u=l?Z(l):null,f=i.x+i.width/2-a.width/2,p=i.y+i.height/2-a.height/2;switch(d){case n:e={x:f,y:i.y-a.height};break;case s:e={x:f,y:i.y+i.height};break;case o:e={x:i.x+i.width,y:p};break;case r:e={x:i.x-a.width,y:p};break;default:e={x:i.x,y:i.y}}var m=d?Q(d):null;if(null!=m){var g="y"===m?"height":"width";switch(u){case c:e[m]=e[m]-(i[g]/2-a[g]/2);break;case h:e[m]=e[m]+(i[g]/2-a[g]/2)}}return e}function _t(t,e){void 0===e&&(e={});var i=e,r=i.placement,a=void 0===r?t.placement:r,c=i.strategy,h=void 0===c?t.strategy:c,m=i.boundary,g=void 0===m?d:m,_=i.rootBoundary,b=void 0===_?u:_,v=i.elementContext,y=void 0===v?f:v,w=i.altBoundary,E=void 0!==w&&w,A=i.padding,T=void 0===A?0:A,C=U("number"!=typeof T?T:G(T,l)),O=y===f?p:f,k=t.rects.popper,D=t.elements[E?O:y],$=function(t,e,i,n){var s="clippingParents"===e?function(t){var e=ft(V(t)),i=["absolute","fixed"].indexOf(z(t).position)>=0&&S(t)?K(t):t;return L(i)?e.filter((function(t){return L(t)&&W(t,i)&&"body"!==x(t)})):[]}(t):[].concat(e),o=[].concat(s,[i]),r=o[0],a=o.reduce((function(e,i){var s=mt(t,i,n);return e.top=N(s.top,e.top),e.right=P(s.right,e.right),e.bottom=P(s.bottom,e.bottom),e.left=N(s.left,e.left),e}),mt(t,r,n));return a.width=a.right-a.left,a.height=a.bottom-a.top,a.x=a.left,a.y=a.top,a}(L(D)?D:D.contextElement||q(t.elements.popper),g,b,h),I=H(t.elements.reference),M=gt({reference:I,element:k,strategy:"absolute",placement:a}),j=pt(Object.assign({},k,M)),F=y===f?j:I,B={top:$.top-F.top+C.top,bottom:F.bottom-$.bottom+C.bottom,left:$.left-F.left+C.left,right:F.right-$.right+C.right},R=t.modifiersData.offset;if(y===f&&R){var Y=R[a];Object.keys(B).forEach((function(t){var e=[o,s].indexOf(t)>=0?1:-1,i=[n,s].indexOf(t)>=0?"y":"x";B[t]+=Y[i]*e}))}return B}const bt={name:"flip",enabled:!0,phase:"main",fn:function(t){var e=t.state,i=t.options,h=t.name;if(!e.modifiersData[h]._skip){for(var d=i.mainAxis,u=void 0===d||d,f=i.altAxis,p=void 0===f||f,_=i.fallbackPlacements,b=i.padding,v=i.boundary,y=i.rootBoundary,w=i.altBoundary,E=i.flipVariations,A=void 0===E||E,T=i.allowedAutoPlacements,C=e.options.placement,O=I(C),x=_||(O!==C&&A?function(t){if(I(t)===a)return[];var e=rt(t);return[lt(t),e,lt(e)]}(C):[rt(C)]),k=[C].concat(x).reduce((function(t,i){return t.concat(I(i)===a?function(t,e){void 0===e&&(e={});var i=e,n=i.placement,s=i.boundary,o=i.rootBoundary,r=i.padding,a=i.flipVariations,c=i.allowedAutoPlacements,h=void 0===c?g:c,d=Z(n),u=d?a?m:m.filter((function(t){return Z(t)===d})):l,f=u.filter((function(t){return h.indexOf(t)>=0}));0===f.length&&(f=u);var p=f.reduce((function(e,i){return e[i]=_t(t,{placement:i,boundary:s,rootBoundary:o,padding:r})[I(i)],e}),{});return Object.keys(p).sort((function(t,e){return p[t]-p[e]}))}(e,{placement:i,boundary:v,rootBoundary:y,padding:b,flipVariations:A,allowedAutoPlacements:T}):i)}),[]),L=e.rects.reference,S=e.rects.popper,D=new Map,$=!0,N=k[0],P=0;P<k.length;P++){var M=k[P],j=I(M),F=Z(M)===c,H=[n,s].indexOf(j)>=0,B=H?"width":"height",W=_t(e,{placement:M,boundary:v,rootBoundary:y,altBoundary:w,padding:b}),z=H?F?o:r:F?s:n;L[B]>S[B]&&(z=rt(z));var R=rt(z),q=[];if(u&&q.push(W[j]<=0),p&&q.push(W[z]<=0,W[R]<=0),q.every((function(t){return t}))){N=M,$=!1;break}D.set(M,q)}if($)for(var V=function(t){var e=k.find((function(e){var i=D.get(e);if(i)return i.slice(0,t).every((function(t){return t}))}));if(e)return N=e,"break"},Y=A?3:1;Y>0&&"break"!==V(Y);Y--);e.placement!==N&&(e.modifiersData[h]._skip=!0,e.placement=N,e.reset=!0)}},requiresIfExists:["offset"],data:{_skip:!1}};function vt(t,e,i){return void 0===i&&(i={x:0,y:0}),{top:t.top-e.height-i.y,right:t.right-e.width+i.x,bottom:t.bottom-e.height+i.y,left:t.left-e.width-i.x}}function yt(t){return[n,o,s,r].some((function(e){return t[e]>=0}))}const wt={name:"hide",enabled:!0,phase:"main",requiresIfExists:["preventOverflow"],fn:function(t){var e=t.state,i=t.name,n=e.rects.reference,s=e.rects.popper,o=e.modifiersData.preventOverflow,r=_t(e,{elementContext:"reference"}),a=_t(e,{altBoundary:!0}),l=vt(r,n),c=vt(a,s,o),h=yt(l),d=yt(c);e.modifiersData[i]={referenceClippingOffsets:l,popperEscapeOffsets:c,isReferenceHidden:h,hasPopperEscaped:d},e.attributes.popper=Object.assign({},e.attributes.popper,{"data-popper-reference-hidden":h,"data-popper-escaped":d})}},Et={name:"offset",enabled:!0,phase:"main",requires:["popperOffsets"],fn:function(t){var e=t.state,i=t.options,s=t.name,a=i.offset,l=void 0===a?[0,0]:a,c=g.reduce((function(t,i){return t[i]=function(t,e,i){var s=I(t),a=[r,n].indexOf(s)>=0?-1:1,l="function"==typeof i?i(Object.assign({},e,{placement:t})):i,c=l[0],h=l[1];return c=c||0,h=(h||0)*a,[r,o].indexOf(s)>=0?{x:h,y:c}:{x:c,y:h}}(i,e.rects,l),t}),{}),h=c[e.placement],d=h.x,u=h.y;null!=e.modifiersData.popperOffsets&&(e.modifiersData.popperOffsets.x+=d,e.modifiersData.popperOffsets.y+=u),e.modifiersData[s]=c}},At={name:"popperOffsets",enabled:!0,phase:"read",fn:function(t){var e=t.state,i=t.name;e.modifiersData[i]=gt({reference:e.rects.reference,element:e.rects.popper,strategy:"absolute",placement:e.placement})},data:{}},Tt={name:"preventOverflow",enabled:!0,phase:"main",fn:function(t){var e=t.state,i=t.options,a=t.name,l=i.mainAxis,h=void 0===l||l,d=i.altAxis,u=void 0!==d&&d,f=i.boundary,p=i.rootBoundary,m=i.altBoundary,g=i.padding,_=i.tether,b=void 0===_||_,v=i.tetherOffset,y=void 0===v?0:v,w=_t(e,{boundary:f,rootBoundary:p,padding:g,altBoundary:m}),E=I(e.placement),A=Z(e.placement),T=!A,C=Q(E),O="x"===C?"y":"x",x=e.modifiersData.popperOffsets,k=e.rects.reference,L=e.rects.popper,S="function"==typeof y?y(Object.assign({},e.rects,{placement:e.placement})):y,D="number"==typeof S?{mainAxis:S,altAxis:S}:Object.assign({mainAxis:0,altAxis:0},S),$=e.modifiersData.offset?e.modifiersData.offset[e.placement]:null,M={x:0,y:0};if(x){if(h){var j,F="y"===C?n:r,H="y"===C?s:o,W="y"===C?"height":"width",z=x[C],R=z+w[F],q=z-w[H],V=b?-L[W]/2:0,Y=A===c?k[W]:L[W],U=A===c?-L[W]:-k[W],G=e.elements.arrow,J=b&&G?B(G):{width:0,height:0},tt=e.modifiersData["arrow#persistent"]?e.modifiersData["arrow#persistent"].padding:{top:0,right:0,bottom:0,left:0},et=tt[F],it=tt[H],nt=X(0,k[W],J[W]),st=T?k[W]/2-V-nt-et-D.mainAxis:Y-nt-et-D.mainAxis,ot=T?-k[W]/2+V+nt+it+D.mainAxis:U+nt+it+D.mainAxis,rt=e.elements.arrow&&K(e.elements.arrow),at=rt?"y"===C?rt.clientTop||0:rt.clientLeft||0:0,lt=null!=(j=null==$?void 0:$[C])?j:0,ct=z+ot-lt,ht=X(b?P(R,z+st-lt-at):R,z,b?N(q,ct):q);x[C]=ht,M[C]=ht-z}if(u){var dt,ut="x"===C?n:r,ft="x"===C?s:o,pt=x[O],mt="y"===O?"height":"width",gt=pt+w[ut],bt=pt-w[ft],vt=-1!==[n,r].indexOf(E),yt=null!=(dt=null==$?void 0:$[O])?dt:0,wt=vt?gt:pt-k[mt]-L[mt]-yt+D.altAxis,Et=vt?pt+k[mt]+L[mt]-yt-D.altAxis:bt,At=b&&vt?function(t,e,i){var n=X(t,e,i);return n>i?i:n}(wt,pt,Et):X(b?wt:gt,pt,b?Et:bt);x[O]=At,M[O]=At-pt}e.modifiersData[a]=M}},requiresIfExists:["offset"]};function Ct(t,e,i){void 0===i&&(i=!1);var n,s,o=S(e),r=S(e)&&function(t){var e=t.getBoundingClientRect(),i=M(e.width)/t.offsetWidth||1,n=M(e.height)/t.offsetHeight||1;return 1!==i||1!==n}(e),a=q(e),l=H(t,r,i),c={scrollLeft:0,scrollTop:0},h={x:0,y:0};return(o||!o&&!i)&&(("body"!==x(e)||dt(a))&&(c=(n=e)!==k(n)&&S(n)?{scrollLeft:(s=n).scrollLeft,scrollTop:s.scrollTop}:ct(n)),S(e)?((h=H(e,!0)).x+=e.clientLeft,h.y+=e.clientTop):a&&(h.x=ht(a))),{x:l.left+c.scrollLeft-h.x,y:l.top+c.scrollTop-h.y,width:l.width,height:l.height}}function Ot(t){var e=new Map,i=new Set,n=[];function s(t){i.add(t.name),[].concat(t.requires||[],t.requiresIfExists||[]).forEach((function(t){if(!i.has(t)){var n=e.get(t);n&&s(n)}})),n.push(t)}return t.forEach((function(t){e.set(t.name,t)})),t.forEach((function(t){i.has(t.name)||s(t)})),n}var xt={placement:"bottom",modifiers:[],strategy:"absolute"};function kt(){for(var t=arguments.length,e=new Array(t),i=0;i<t;i++)e[i]=arguments[i];return!e.some((function(t){return!(t&&"function"==typeof t.getBoundingClientRect)}))}function Lt(t){void 0===t&&(t={});var e=t,i=e.defaultModifiers,n=void 0===i?[]:i,s=e.defaultOptions,o=void 0===s?xt:s;return function(t,e,i){void 0===i&&(i=o);var s,r,a={placement:"bottom",orderedModifiers:[],options:Object.assign({},xt,o),modifiersData:{},elements:{reference:t,popper:e},attributes:{},styles:{}},l=[],c=!1,h={state:a,setOptions:function(i){var s="function"==typeof i?i(a.options):i;d(),a.options=Object.assign({},o,a.options,s),a.scrollParents={reference:L(t)?ft(t):t.contextElement?ft(t.contextElement):[],popper:ft(e)};var r,c,u=function(t){var e=Ot(t);return O.reduce((function(t,i){return t.concat(e.filter((function(t){return t.phase===i})))}),[])}((r=[].concat(n,a.options.modifiers),c=r.reduce((function(t,e){var i=t[e.name];return t[e.name]=i?Object.assign({},i,e,{options:Object.assign({},i.options,e.options),data:Object.assign({},i.data,e.data)}):e,t}),{}),Object.keys(c).map((function(t){return c[t]}))));return a.orderedModifiers=u.filter((function(t){return t.enabled})),a.orderedModifiers.forEach((function(t){var e=t.name,i=t.options,n=void 0===i?{}:i,s=t.effect;if("function"==typeof s){var o=s({state:a,name:e,instance:h,options:n});l.push(o||function(){})}})),h.update()},forceUpdate:function(){if(!c){var t=a.elements,e=t.reference,i=t.popper;if(kt(e,i)){a.rects={reference:Ct(e,K(i),"fixed"===a.options.strategy),popper:B(i)},a.reset=!1,a.placement=a.options.placement,a.orderedModifiers.forEach((function(t){return a.modifiersData[t.name]=Object.assign({},t.data)}));for(var n=0;n<a.orderedModifiers.length;n++)if(!0!==a.reset){var s=a.orderedModifiers[n],o=s.fn,r=s.options,l=void 0===r?{}:r,d=s.name;"function"==typeof o&&(a=o({state:a,options:l,name:d,instance:h})||a)}else a.reset=!1,n=-1}}},update:(s=function(){return new Promise((function(t){h.forceUpdate(),t(a)}))},function(){return r||(r=new Promise((function(t){Promise.resolve().then((function(){r=void 0,t(s())}))}))),r}),destroy:function(){d(),c=!0}};if(!kt(t,e))return h;function d(){l.forEach((function(t){return t()})),l=[]}return h.setOptions(i).then((function(t){!c&&i.onFirstUpdate&&i.onFirstUpdate(t)})),h}}var St=Lt(),Dt=Lt({defaultModifiers:[st,At,it,$,Et,bt,Tt,J,wt]}),$t=Lt({defaultModifiers:[st,At,it,$]});const It=new Map,Nt={set(t,e,i){It.has(t)||It.set(t,new Map);const n=It.get(t);n.has(e)||0===n.size?n.set(e,i):console.error(`Bootstrap doesn't allow more than one instance per element. Bound instance: ${Array.from(n.keys())[0]}.`)},get:(t,e)=>It.has(t)&&It.get(t).get(e)||null,remove(t,e){if(!It.has(t))return;const i=It.get(t);i.delete(e),0===i.size&&It.delete(t)}},Pt="transitionend",Mt=t=>(t&&window.CSS&&window.CSS.escape&&(t=t.replace(/#([^\s"#']+)/g,((t,e)=>`#${CSS.escape(e)}`))),t),jt=t=>{t.dispatchEvent(new Event(Pt))},Ft=t=>!(!t||"object"!=typeof t)&&(void 0!==t.jquery&&(t=t[0]),void 0!==t.nodeType),Ht=t=>Ft(t)?t.jquery?t[0]:t:"string"==typeof t&&t.length>0?document.querySelector(Mt(t)):null,Bt=t=>{if(!Ft(t)||0===t.getClientRects().length)return!1;const e="visible"===getComputedStyle(t).getPropertyValue("visibility"),i=t.closest("details:not([open])");if(!i)return e;if(i!==t){const e=t.closest("summary");if(e&&e.parentNode!==i)return!1;if(null===e)return!1}return e},Wt=t=>!t||t.nodeType!==Node.ELEMENT_NODE||!!t.classList.contains("disabled")||(void 0!==t.disabled?t.disabled:t.hasAttribute("disabled")&&"false"!==t.getAttribute("disabled")),zt=t=>{if(!document.documentElement.attachShadow)return null;if("function"==typeof t.getRootNode){const e=t.getRootNode();return e instanceof ShadowRoot?e:null}return t instanceof ShadowRoot?t:t.parentNode?zt(t.parentNode):null},Rt=()=>{},qt=t=>{t.offsetHeight},Vt=()=>window.jQuery&&!document.body.hasAttribute("data-bs-no-jquery")?window.jQuery:null,Yt=[],Kt=()=>"rtl"===document.documentElement.dir,Qt=t=>{var e;e=()=>{const e=Vt();if(e){const i=t.NAME,n=e.fn[i];e.fn[i]=t.jQueryInterface,e.fn[i].Constructor=t,e.fn[i].noConflict=()=>(e.fn[i]=n,t.jQueryInterface)}},"loading"===document.readyState?(Yt.length||document.addEventListener("DOMContentLoaded",(()=>{for(const t of Yt)t()})),Yt.push(e)):e()},Xt=(t,e=[],i=t)=>"function"==typeof t?t(...e):i,Ut=(t,e,i=!0)=>{if(!i)return void Xt(t);const n=(t=>{if(!t)return 0;let{transitionDuration:e,transitionDelay:i}=window.getComputedStyle(t);const n=Number.parseFloat(e),s=Number.parseFloat(i);return n||s?(e=e.split(",")[0],i=i.split(",")[0],1e3*(Number.parseFloat(e)+Number.parseFloat(i))):0})(e)+5;let s=!1;const o=({target:i})=>{i===e&&(s=!0,e.removeEventListener(Pt,o),Xt(t))};e.addEventListener(Pt,o),setTimeout((()=>{s||jt(e)}),n)},Gt=(t,e,i,n)=>{const s=t.length;let o=t.indexOf(e);return-1===o?!i&&n?t[s-1]:t[0]:(o+=i?1:-1,n&&(o=(o+s)%s),t[Math.max(0,Math.min(o,s-1))])},Jt=/[^.]*(?=\..*)\.|.*/,Zt=/\..*/,te=/::\d+$/,ee={};let ie=1;const ne={mouseenter:"mouseover",mouseleave:"mouseout"},se=new Set(["click","dblclick","mouseup","mousedown","contextmenu","mousewheel","DOMMouseScroll","mouseover","mouseout","mousemove","selectstart","selectend","keydown","keypress","keyup","orientationchange","touchstart","touchmove","touchend","touchcancel","pointerdown","pointermove","pointerup","pointerleave","pointercancel","gesturestart","gesturechange","gestureend","focus","blur","change","reset","select","submit","focusin","focusout","load","unload","beforeunload","resize","move","DOMContentLoaded","readystatechange","error","abort","scroll"]);function oe(t,e){return e&&`${e}::${ie++}`||t.uidEvent||ie++}function re(t){const e=oe(t);return t.uidEvent=e,ee[e]=ee[e]||{},ee[e]}function ae(t,e,i=null){return Object.values(t).find((t=>t.callable===e&&t.delegationSelector===i))}function le(t,e,i){const n="string"==typeof e,s=n?i:e||i;let o=ue(t);return se.has(o)||(o=t),[n,s,o]}function ce(t,e,i,n,s){if("string"!=typeof e||!t)return;let[o,r,a]=le(e,i,n);if(e in ne){const t=t=>function(e){if(!e.relatedTarget||e.relatedTarget!==e.delegateTarget&&!e.delegateTarget.contains(e.relatedTarget))return t.call(this,e)};r=t(r)}const l=re(t),c=l[a]||(l[a]={}),h=ae(c,r,o?i:null);if(h)return void(h.oneOff=h.oneOff&&s);const d=oe(r,e.replace(Jt,"")),u=o?function(t,e,i){return function n(s){const o=t.querySelectorAll(e);for(let{target:r}=s;r&&r!==this;r=r.parentNode)for(const a of o)if(a===r)return pe(s,{delegateTarget:r}),n.oneOff&&fe.off(t,s.type,e,i),i.apply(r,[s])}}(t,i,r):function(t,e){return function i(n){return pe(n,{delegateTarget:t}),i.oneOff&&fe.off(t,n.type,e),e.apply(t,[n])}}(t,r);u.delegationSelector=o?i:null,u.callable=r,u.oneOff=s,u.uidEvent=d,c[d]=u,t.addEventListener(a,u,o)}function he(t,e,i,n,s){const o=ae(e[i],n,s);o&&(t.removeEventListener(i,o,Boolean(s)),delete e[i][o.uidEvent])}function de(t,e,i,n){const s=e[i]||{};for(const[o,r]of Object.entries(s))o.includes(n)&&he(t,e,i,r.callable,r.delegationSelector)}function ue(t){return t=t.replace(Zt,""),ne[t]||t}const fe={on(t,e,i,n){ce(t,e,i,n,!1)},one(t,e,i,n){ce(t,e,i,n,!0)},off(t,e,i,n){if("string"!=typeof e||!t)return;const[s,o,r]=le(e,i,n),a=r!==e,l=re(t),c=l[r]||{},h=e.startsWith(".");if(void 0===o){if(h)for(const i of Object.keys(l))de(t,l,i,e.slice(1));for(const[i,n]of Object.entries(c)){const s=i.replace(te,"");a&&!e.includes(s)||he(t,l,r,n.callable,n.delegationSelector)}}else{if(!Object.keys(c).length)return;he(t,l,r,o,s?i:null)}},trigger(t,e,i){if("string"!=typeof e||!t)return null;const n=Vt();let s=null,o=!0,r=!0,a=!1;e!==ue(e)&&n&&(s=n.Event(e,i),n(t).trigger(s),o=!s.isPropagationStopped(),r=!s.isImmediatePropagationStopped(),a=s.isDefaultPrevented());const l=pe(new Event(e,{bubbles:o,cancelable:!0}),i);return a&&l.preventDefault(),r&&t.dispatchEvent(l),l.defaultPrevented&&s&&s.preventDefault(),l}};function pe(t,e={}){for(const[i,n]of Object.entries(e))try{t[i]=n}catch(e){Object.defineProperty(t,i,{configurable:!0,get:()=>n})}return t}function me(t){if("true"===t)return!0;if("false"===t)return!1;if(t===Number(t).toString())return Number(t);if(""===t||"null"===t)return null;if("string"!=typeof t)return t;try{return JSON.parse(decodeURIComponent(t))}catch(e){return t}}function ge(t){return t.replace(/[A-Z]/g,(t=>`-${t.toLowerCase()}`))}const _e={setDataAttribute(t,e,i){t.setAttribute(`data-bs-${ge(e)}`,i)},removeDataAttribute(t,e){t.removeAttribute(`data-bs-${ge(e)}`)},getDataAttributes(t){if(!t)return{};const e={},i=Object.keys(t.dataset).filter((t=>t.startsWith("bs")&&!t.startsWith("bsConfig")));for(const n of i){let i=n.replace(/^bs/,"");i=i.charAt(0).toLowerCase()+i.slice(1,i.length),e[i]=me(t.dataset[n])}return e},getDataAttribute:(t,e)=>me(t.getAttribute(`data-bs-${ge(e)}`))};class be{static get Default(){return{}}static get DefaultType(){return{}}static get NAME(){throw new Error('You have to implement the static method "NAME", for each component!')}_getConfig(t){return t=this._mergeConfigObj(t),t=this._configAfterMerge(t),this._typeCheckConfig(t),t}_configAfterMerge(t){return t}_mergeConfigObj(t,e){const i=Ft(e)?_e.getDataAttribute(e,"config"):{};return{...this.constructor.Default,..."object"==typeof i?i:{},...Ft(e)?_e.getDataAttributes(e):{},..."object"==typeof t?t:{}}}_typeCheckConfig(t,e=this.constructor.DefaultType){for(const[n,s]of Object.entries(e)){const e=t[n],o=Ft(e)?"element":null==(i=e)?`${i}`:Object.prototype.toString.call(i).match(/\s([a-z]+)/i)[1].toLowerCase();if(!new RegExp(s).test(o))throw new TypeError(`${this.constructor.NAME.toUpperCase()}: Option "${n}" provided type "${o}" but expected type "${s}".`)}var i}}class ve extends be{constructor(t,e){super(),(t=Ht(t))&&(this._element=t,this._config=this._getConfig(e),Nt.set(this._element,this.constructor.DATA_KEY,this))}dispose(){Nt.remove(this._element,this.constructor.DATA_KEY),fe.off(this._element,this.constructor.EVENT_KEY);for(const t of Object.getOwnPropertyNames(this))this[t]=null}_queueCallback(t,e,i=!0){Ut(t,e,i)}_getConfig(t){return t=this._mergeConfigObj(t,this._element),t=this._configAfterMerge(t),this._typeCheckConfig(t),t}static getInstance(t){return Nt.get(Ht(t),this.DATA_KEY)}static getOrCreateInstance(t,e={}){return this.getInstance(t)||new this(t,"object"==typeof e?e:null)}static get VERSION(){return"5.3.3"}static get DATA_KEY(){return`bs.${this.NAME}`}static get EVENT_KEY(){return`.${this.DATA_KEY}`}static eventName(t){return`${t}${this.EVENT_KEY}`}}const ye=t=>{let e=t.getAttribute("data-bs-target");if(!e||"#"===e){let i=t.getAttribute("href");if(!i||!i.includes("#")&&!i.startsWith("."))return null;i.includes("#")&&!i.startsWith("#")&&(i=`#${i.split("#")[1]}`),e=i&&"#"!==i?i.trim():null}return e?e.split(",").map((t=>Mt(t))).join(","):null},we={find:(t,e=document.documentElement)=>[].concat(...Element.prototype.querySelectorAll.call(e,t)),findOne:(t,e=document.documentElement)=>Element.prototype.querySelector.call(e,t),children:(t,e)=>[].concat(...t.children).filter((t=>t.matches(e))),parents(t,e){const i=[];let n=t.parentNode.closest(e);for(;n;)i.push(n),n=n.parentNode.closest(e);return i},prev(t,e){let i=t.previousElementSibling;for(;i;){if(i.matches(e))return[i];i=i.previousElementSibling}return[]},next(t,e){let i=t.nextElementSibling;for(;i;){if(i.matches(e))return[i];i=i.nextElementSibling}return[]},focusableChildren(t){const e=["a","button","input","textarea","select","details","[tabindex]",'[contenteditable="true"]'].map((t=>`${t}:not([tabindex^="-"])`)).join(",");return this.find(e,t).filter((t=>!Wt(t)&&Bt(t)))},getSelectorFromElement(t){const e=ye(t);return e&&we.findOne(e)?e:null},getElementFromSelector(t){const e=ye(t);return e?we.findOne(e):null},getMultipleElementsFromSelector(t){const e=ye(t);return e?we.find(e):[]}},Ee=(t,e="hide")=>{const i=`click.dismiss${t.EVENT_KEY}`,n=t.NAME;fe.on(document,i,`[data-bs-dismiss="${n}"]`,(function(i){if(["A","AREA"].includes(this.tagName)&&i.preventDefault(),Wt(this))return;const s=we.getElementFromSelector(this)||this.closest(`.${n}`);t.getOrCreateInstance(s)[e]()}))},Ae=".bs.alert",Te=`close${Ae}`,Ce=`closed${Ae}`;class Oe extends ve{static get NAME(){return"alert"}close(){if(fe.trigger(this._element,Te).defaultPrevented)return;this._element.classList.remove("show");const t=this._element.classList.contains("fade");this._queueCallback((()=>this._destroyElement()),this._element,t)}_destroyElement(){this._element.remove(),fe.trigger(this._element,Ce),this.dispose()}static jQueryInterface(t){return this.each((function(){const e=Oe.getOrCreateInstance(this);if("string"==typeof t){if(void 0===e[t]||t.startsWith("_")||"constructor"===t)throw new TypeError(`No method named "${t}"`);e[t](this)}}))}}Ee(Oe,"close"),Qt(Oe);const xe='[data-bs-toggle="button"]';class ke extends ve{static get NAME(){return"button"}toggle(){this._element.setAttribute("aria-pressed",this._element.classList.toggle("active"))}static jQueryInterface(t){return this.each((function(){const e=ke.getOrCreateInstance(this);"toggle"===t&&e[t]()}))}}fe.on(document,"click.bs.button.data-api",xe,(t=>{t.preventDefault();const e=t.target.closest(xe);ke.getOrCreateInstance(e).toggle()})),Qt(ke);const Le=".bs.swipe",Se=`touchstart${Le}`,De=`touchmove${Le}`,$e=`touchend${Le}`,Ie=`pointerdown${Le}`,Ne=`pointerup${Le}`,Pe={endCallback:null,leftCallback:null,rightCallback:null},Me={endCallback:"(function|null)",leftCallback:"(function|null)",rightCallback:"(function|null)"};class je extends be{constructor(t,e){super(),this._element=t,t&&je.isSupported()&&(this._config=this._getConfig(e),this._deltaX=0,this._supportPointerEvents=Boolean(window.PointerEvent),this._initEvents())}static get Default(){return Pe}static get DefaultType(){return Me}static get NAME(){return"swipe"}dispose(){fe.off(this._element,Le)}_start(t){this._supportPointerEvents?this._eventIsPointerPenTouch(t)&&(this._deltaX=t.clientX):this._deltaX=t.touches[0].clientX}_end(t){this._eventIsPointerPenTouch(t)&&(this._deltaX=t.clientX-this._deltaX),this._handleSwipe(),Xt(this._config.endCallback)}_move(t){this._deltaX=t.touches&&t.touches.length>1?0:t.touches[0].clientX-this._deltaX}_handleSwipe(){const t=Math.abs(this._deltaX);if(t<=40)return;const e=t/this._deltaX;this._deltaX=0,e&&Xt(e>0?this._config.rightCallback:this._config.leftCallback)}_initEvents(){this._supportPointerEvents?(fe.on(this._element,Ie,(t=>this._start(t))),fe.on(this._element,Ne,(t=>this._end(t))),this._element.classList.add("pointer-event")):(fe.on(this._element,Se,(t=>this._start(t))),fe.on(this._element,De,(t=>this._move(t))),fe.on(this._element,$e,(t=>this._end(t))))}_eventIsPointerPenTouch(t){return this._supportPointerEvents&&("pen"===t.pointerType||"touch"===t.pointerType)}static isSupported(){return"ontouchstart"in document.documentElement||navigator.maxTouchPoints>0}}const Fe=".bs.carousel",He=".data-api",Be="ArrowLeft",We="ArrowRight",ze="next",Re="prev",qe="left",Ve="right",Ye=`slide${Fe}`,Ke=`slid${Fe}`,Qe=`keydown${Fe}`,Xe=`mouseenter${Fe}`,Ue=`mouseleave${Fe}`,Ge=`dragstart${Fe}`,Je=`load${Fe}${He}`,Ze=`click${Fe}${He}`,ti="carousel",ei="active",ii=".active",ni=".carousel-item",si=ii+ni,oi={[Be]:Ve,[We]:qe},ri={interval:5e3,keyboard:!0,pause:"hover",ride:!1,touch:!0,wrap:!0},ai={interval:"(number|boolean)",keyboard:"boolean",pause:"(string|boolean)",ride:"(boolean|string)",touch:"boolean",wrap:"boolean"};class li extends ve{constructor(t,e){super(t,e),this._interval=null,this._activeElement=null,this._isSliding=!1,this.touchTimeout=null,this._swipeHelper=null,this._indicatorsElement=we.findOne(".carousel-indicators",this._element),this._addEventListeners(),this._config.ride===ti&&this.cycle()}static get Default(){return ri}static get DefaultType(){return ai}static get NAME(){return"carousel"}next(){this._slide(ze)}nextWhenVisible(){!document.hidden&&Bt(this._element)&&this.next()}prev(){this._slide(Re)}pause(){this._isSliding&&jt(this._element),this._clearInterval()}cycle(){this._clearInterval(),this._updateInterval(),this._interval=setInterval((()=>this.nextWhenVisible()),this._config.interval)}_maybeEnableCycle(){this._config.ride&&(this._isSliding?fe.one(this._element,Ke,(()=>this.cycle())):this.cycle())}to(t){const e=this._getItems();if(t>e.length-1||t<0)return;if(this._isSliding)return void fe.one(this._element,Ke,(()=>this.to(t)));const i=this._getItemIndex(this._getActive());if(i===t)return;const n=t>i?ze:Re;this._slide(n,e[t])}dispose(){this._swipeHelper&&this._swipeHelper.dispose(),super.dispose()}_configAfterMerge(t){return t.defaultInterval=t.interval,t}_addEventListeners(){this._config.keyboard&&fe.on(this._element,Qe,(t=>this._keydown(t))),"hover"===this._config.pause&&(fe.on(this._element,Xe,(()=>this.pause())),fe.on(this._element,Ue,(()=>this._maybeEnableCycle()))),this._config.touch&&je.isSupported()&&this._addTouchEventListeners()}_addTouchEventListeners(){for(const t of we.find(".carousel-item img",this._element))fe.on(t,Ge,(t=>t.preventDefault()));const t={leftCallback:()=>this._slide(this._directionToOrder(qe)),rightCallback:()=>this._slide(this._directionToOrder(Ve)),endCallback:()=>{"hover"===this._config.pause&&(this.pause(),this.touchTimeout&&clearTimeout(this.touchTimeout),this.touchTimeout=setTimeout((()=>this._maybeEnableCycle()),500+this._config.interval))}};this._swipeHelper=new je(this._element,t)}_keydown(t){if(/input|textarea/i.test(t.target.tagName))return;const e=oi[t.key];e&&(t.preventDefault(),this._slide(this._directionToOrder(e)))}_getItemIndex(t){return this._getItems().indexOf(t)}_setActiveIndicatorElement(t){if(!this._indicatorsElement)return;const e=we.findOne(ii,this._indicatorsElement);e.classList.remove(ei),e.removeAttribute("aria-current");const i=we.findOne(`[data-bs-slide-to="${t}"]`,this._indicatorsElement);i&&(i.classList.add(ei),i.setAttribute("aria-current","true"))}_updateInterval(){const t=this._activeElement||this._getActive();if(!t)return;const e=Number.parseInt(t.getAttribute("data-bs-interval"),10);this._config.interval=e||this._config.defaultInterval}_slide(t,e=null){if(this._isSliding)return;const i=this._getActive(),n=t===ze,s=e||Gt(this._getItems(),i,n,this._config.wrap);if(s===i)return;const o=this._getItemIndex(s),r=e=>fe.trigger(this._element,e,{relatedTarget:s,direction:this._orderToDirection(t),from:this._getItemIndex(i),to:o});if(r(Ye).defaultPrevented)return;if(!i||!s)return;const a=Boolean(this._interval);this.pause(),this._isSliding=!0,this._setActiveIndicatorElement(o),this._activeElement=s;const l=n?"carousel-item-start":"carousel-item-end",c=n?"carousel-item-next":"carousel-item-prev";s.classList.add(c),qt(s),i.classList.add(l),s.classList.add(l),this._queueCallback((()=>{s.classList.remove(l,c),s.classList.add(ei),i.classList.remove(ei,c,l),this._isSliding=!1,r(Ke)}),i,this._isAnimated()),a&&this.cycle()}_isAnimated(){return this._element.classList.contains("slide")}_getActive(){return we.findOne(si,this._element)}_getItems(){return we.find(ni,this._element)}_clearInterval(){this._interval&&(clearInterval(this._interval),this._interval=null)}_directionToOrder(t){return Kt()?t===qe?Re:ze:t===qe?ze:Re}_orderToDirection(t){return Kt()?t===Re?qe:Ve:t===Re?Ve:qe}static jQueryInterface(t){return this.each((function(){const e=li.getOrCreateInstance(this,t);if("number"!=typeof t){if("string"==typeof t){if(void 0===e[t]||t.startsWith("_")||"constructor"===t)throw new TypeError(`No method named "${t}"`);e[t]()}}else e.to(t)}))}}fe.on(document,Ze,"[data-bs-slide], [data-bs-slide-to]",(function(t){const e=we.getElementFromSelector(this);if(!e||!e.classList.contains(ti))return;t.preventDefault();const i=li.getOrCreateInstance(e),n=this.getAttribute("data-bs-slide-to");return n?(i.to(n),void i._maybeEnableCycle()):"next"===_e.getDataAttribute(this,"slide")?(i.next(),void i._maybeEnableCycle()):(i.prev(),void i._maybeEnableCycle())})),fe.on(window,Je,(()=>{const t=we.find('[data-bs-ride="carousel"]');for(const e of t)li.getOrCreateInstance(e)})),Qt(li);const ci=".bs.collapse",hi=`show${ci}`,di=`shown${ci}`,ui=`hide${ci}`,fi=`hidden${ci}`,pi=`click${ci}.data-api`,mi="show",gi="collapse",_i="collapsing",bi=`:scope .${gi} .${gi}`,vi='[data-bs-toggle="collapse"]',yi={parent:null,toggle:!0},wi={parent:"(null|element)",toggle:"boolean"};class Ei extends ve{constructor(t,e){super(t,e),this._isTransitioning=!1,this._triggerArray=[];const i=we.find(vi);for(const t of i){const e=we.getSelectorFromElement(t),i=we.find(e).filter((t=>t===this._element));null!==e&&i.length&&this._triggerArray.push(t)}this._initializeChildren(),this._config.parent||this._addAriaAndCollapsedClass(this._triggerArray,this._isShown()),this._config.toggle&&this.toggle()}static get Default(){return yi}static get DefaultType(){return wi}static get NAME(){return"collapse"}toggle(){this._isShown()?this.hide():this.show()}show(){if(this._isTransitioning||this._isShown())return;let t=[];if(this._config.parent&&(t=this._getFirstLevelChildren(".collapse.show, .collapse.collapsing").filter((t=>t!==this._element)).map((t=>Ei.getOrCreateInstance(t,{toggle:!1})))),t.length&&t[0]._isTransitioning)return;if(fe.trigger(this._element,hi).defaultPrevented)return;for(const e of t)e.hide();const e=this._getDimension();this._element.classList.remove(gi),this._element.classList.add(_i),this._element.style[e]=0,this._addAriaAndCollapsedClass(this._triggerArray,!0),this._isTransitioning=!0;const i=`scroll${e[0].toUpperCase()+e.slice(1)}`;this._queueCallback((()=>{this._isTransitioning=!1,this._element.classList.remove(_i),this._element.classList.add(gi,mi),this._element.style[e]="",fe.trigger(this._element,di)}),this._element,!0),this._element.style[e]=`${this._element[i]}px`}hide(){if(this._isTransitioning||!this._isShown())return;if(fe.trigger(this._element,ui).defaultPrevented)return;const t=this._getDimension();this._element.style[t]=`${this._element.getBoundingClientRect()[t]}px`,qt(this._element),this._element.classList.add(_i),this._element.classList.remove(gi,mi);for(const t of this._triggerArray){const e=we.getElementFromSelector(t);e&&!this._isShown(e)&&this._addAriaAndCollapsedClass([t],!1)}this._isTransitioning=!0,this._element.style[t]="",this._queueCallback((()=>{this._isTransitioning=!1,this._element.classList.remove(_i),this._element.classList.add(gi),fe.trigger(this._element,fi)}),this._element,!0)}_isShown(t=this._element){return t.classList.contains(mi)}_configAfterMerge(t){return t.toggle=Boolean(t.toggle),t.parent=Ht(t.parent),t}_getDimension(){return this._element.classList.contains("collapse-horizontal")?"width":"height"}_initializeChildren(){if(!this._config.parent)return;const t=this._getFirstLevelChildren(vi);for(const e of t){const t=we.getElementFromSelector(e);t&&this._addAriaAndCollapsedClass([e],this._isShown(t))}}_getFirstLevelChildren(t){const e=we.find(bi,this._config.parent);return we.find(t,this._config.parent).filter((t=>!e.includes(t)))}_addAriaAndCollapsedClass(t,e){if(t.length)for(const i of t)i.classList.toggle("collapsed",!e),i.setAttribute("aria-expanded",e)}static jQueryInterface(t){const e={};return"string"==typeof t&&/show|hide/.test(t)&&(e.toggle=!1),this.each((function(){const i=Ei.getOrCreateInstance(this,e);if("string"==typeof t){if(void 0===i[t])throw new TypeError(`No method named "${t}"`);i[t]()}}))}}fe.on(document,pi,vi,(function(t){("A"===t.target.tagName||t.delegateTarget&&"A"===t.delegateTarget.tagName)&&t.preventDefault();for(const t of we.getMultipleElementsFromSelector(this))Ei.getOrCreateInstance(t,{toggle:!1}).toggle()})),Qt(Ei);const Ai="dropdown",Ti=".bs.dropdown",Ci=".data-api",Oi="ArrowUp",xi="ArrowDown",ki=`hide${Ti}`,Li=`hidden${Ti}`,Si=`show${Ti}`,Di=`shown${Ti}`,$i=`click${Ti}${Ci}`,Ii=`keydown${Ti}${Ci}`,Ni=`keyup${Ti}${Ci}`,Pi="show",Mi='[data-bs-toggle="dropdown"]:not(.disabled):not(:disabled)',ji=`${Mi}.${Pi}`,Fi=".dropdown-menu",Hi=Kt()?"top-end":"top-start",Bi=Kt()?"top-start":"top-end",Wi=Kt()?"bottom-end":"bottom-start",zi=Kt()?"bottom-start":"bottom-end",Ri=Kt()?"left-start":"right-start",qi=Kt()?"right-start":"left-start",Vi={autoClose:!0,boundary:"clippingParents",display:"dynamic",offset:[0,2],popperConfig:null,reference:"toggle"},Yi={autoClose:"(boolean|string)",boundary:"(string|element)",display:"string",offset:"(array|string|function)",popperConfig:"(null|object|function)",reference:"(string|element|object)"};class Ki extends ve{constructor(t,e){super(t,e),this._popper=null,this._parent=this._element.parentNode,this._menu=we.next(this._element,Fi)[0]||we.prev(this._element,Fi)[0]||we.findOne(Fi,this._parent),this._inNavbar=this._detectNavbar()}static get Default(){return Vi}static get DefaultType(){return Yi}static get NAME(){return Ai}toggle(){return this._isShown()?this.hide():this.show()}show(){if(Wt(this._element)||this._isShown())return;const t={relatedTarget:this._element};if(!fe.trigger(this._element,Si,t).defaultPrevented){if(this._createPopper(),"ontouchstart"in document.documentElement&&!this._parent.closest(".navbar-nav"))for(const t of[].concat(...document.body.children))fe.on(t,"mouseover",Rt);this._element.focus(),this._element.setAttribute("aria-expanded",!0),this._menu.classList.add(Pi),this._element.classList.add(Pi),fe.trigger(this._element,Di,t)}}hide(){if(Wt(this._element)||!this._isShown())return;const t={relatedTarget:this._element};this._completeHide(t)}dispose(){this._popper&&this._popper.destroy(),super.dispose()}update(){this._inNavbar=this._detectNavbar(),this._popper&&this._popper.update()}_completeHide(t){if(!fe.trigger(this._element,ki,t).defaultPrevented){if("ontouchstart"in document.documentElement)for(const t of[].concat(...document.body.children))fe.off(t,"mouseover",Rt);this._popper&&this._popper.destroy(),this._menu.classList.remove(Pi),this._element.classList.remove(Pi),this._element.setAttribute("aria-expanded","false"),_e.removeDataAttribute(this._menu,"popper"),fe.trigger(this._element,Li,t)}}_getConfig(t){if("object"==typeof(t=super._getConfig(t)).reference&&!Ft(t.reference)&&"function"!=typeof t.reference.getBoundingClientRect)throw new TypeError(`${Ai.toUpperCase()}: Option "reference" provided type "object" without a required "getBoundingClientRect" method.`);return t}_createPopper(){if(void 0===e)throw new TypeError("Bootstrap's dropdowns require Popper (https://popper.js.org)");let t=this._element;"parent"===this._config.reference?t=this._parent:Ft(this._config.reference)?t=Ht(this._config.reference):"object"==typeof this._config.reference&&(t=this._config.reference);const i=this._getPopperConfig();this._popper=Dt(t,this._menu,i)}_isShown(){return this._menu.classList.contains(Pi)}_getPlacement(){const t=this._parent;if(t.classList.contains("dropend"))return Ri;if(t.classList.contains("dropstart"))return qi;if(t.classList.contains("dropup-center"))return"top";if(t.classList.contains("dropdown-center"))return"bottom";const e="end"===getComputedStyle(this._menu).getPropertyValue("--bs-position").trim();return t.classList.contains("dropup")?e?Bi:Hi:e?zi:Wi}_detectNavbar(){return null!==this._element.closest(".navbar")}_getOffset(){const{offset:t}=this._config;return"string"==typeof t?t.split(",").map((t=>Number.parseInt(t,10))):"function"==typeof t?e=>t(e,this._element):t}_getPopperConfig(){const t={placement:this._getPlacement(),modifiers:[{name:"preventOverflow",options:{boundary:this._config.boundary}},{name:"offset",options:{offset:this._getOffset()}}]};return(this._inNavbar||"static"===this._config.display)&&(_e.setDataAttribute(this._menu,"popper","static"),t.modifiers=[{name:"applyStyles",enabled:!1}]),{...t,...Xt(this._config.popperConfig,[t])}}_selectMenuItem({key:t,target:e}){const i=we.find(".dropdown-menu .dropdown-item:not(.disabled):not(:disabled)",this._menu).filter((t=>Bt(t)));i.length&&Gt(i,e,t===xi,!i.includes(e)).focus()}static jQueryInterface(t){return this.each((function(){const e=Ki.getOrCreateInstance(this,t);if("string"==typeof t){if(void 0===e[t])throw new TypeError(`No method named "${t}"`);e[t]()}}))}static clearMenus(t){if(2===t.button||"keyup"===t.type&&"Tab"!==t.key)return;const e=we.find(ji);for(const i of e){const e=Ki.getInstance(i);if(!e||!1===e._config.autoClose)continue;const n=t.composedPath(),s=n.includes(e._menu);if(n.includes(e._element)||"inside"===e._config.autoClose&&!s||"outside"===e._config.autoClose&&s)continue;if(e._menu.contains(t.target)&&("keyup"===t.type&&"Tab"===t.key||/input|select|option|textarea|form/i.test(t.target.tagName)))continue;const o={relatedTarget:e._element};"click"===t.type&&(o.clickEvent=t),e._completeHide(o)}}static dataApiKeydownHandler(t){const e=/input|textarea/i.test(t.target.tagName),i="Escape"===t.key,n=[Oi,xi].includes(t.key);if(!n&&!i)return;if(e&&!i)return;t.preventDefault();const s=this.matches(Mi)?this:we.prev(this,Mi)[0]||we.next(this,Mi)[0]||we.findOne(Mi,t.delegateTarget.parentNode),o=Ki.getOrCreateInstance(s);if(n)return t.stopPropagation(),o.show(),void o._selectMenuItem(t);o._isShown()&&(t.stopPropagation(),o.hide(),s.focus())}}fe.on(document,Ii,Mi,Ki.dataApiKeydownHandler),fe.on(document,Ii,Fi,Ki.dataApiKeydownHandler),fe.on(document,$i,Ki.clearMenus),fe.on(document,Ni,Ki.clearMenus),fe.on(document,$i,Mi,(function(t){t.preventDefault(),Ki.getOrCreateInstance(this).toggle()})),Qt(Ki);const Qi="backdrop",Xi="show",Ui=`mousedown.bs.${Qi}`,Gi={className:"modal-backdrop",clickCallback:null,isAnimated:!1,isVisible:!0,rootElement:"body"},Ji={className:"string",clickCallback:"(function|null)",isAnimated:"boolean",isVisible:"boolean",rootElement:"(element|string)"};class Zi extends be{constructor(t){super(),this._config=this._getConfig(t),this._isAppended=!1,this._element=null}static get Default(){return Gi}static get DefaultType(){return Ji}static get NAME(){return Qi}show(t){if(!this._config.isVisible)return void Xt(t);this._append();const e=this._getElement();this._config.isAnimated&&qt(e),e.classList.add(Xi),this._emulateAnimation((()=>{Xt(t)}))}hide(t){this._config.isVisible?(this._getElement().classList.remove(Xi),this._emulateAnimation((()=>{this.dispose(),Xt(t)}))):Xt(t)}dispose(){this._isAppended&&(fe.off(this._element,Ui),this._element.remove(),this._isAppended=!1)}_getElement(){if(!this._element){const t=document.createElement("div");t.className=this._config.className,this._config.isAnimated&&t.classList.add("fade"),this._element=t}return this._element}_configAfterMerge(t){return t.rootElement=Ht(t.rootElement),t}_append(){if(this._isAppended)return;const t=this._getElement();this._config.rootElement.append(t),fe.on(t,Ui,(()=>{Xt(this._config.clickCallback)})),this._isAppended=!0}_emulateAnimation(t){Ut(t,this._getElement(),this._config.isAnimated)}}const tn=".bs.focustrap",en=`focusin${tn}`,nn=`keydown.tab${tn}`,sn="backward",on={autofocus:!0,trapElement:null},rn={autofocus:"boolean",trapElement:"element"};class an extends be{constructor(t){super(),this._config=this._getConfig(t),this._isActive=!1,this._lastTabNavDirection=null}static get Default(){return on}static get DefaultType(){return rn}static get NAME(){return"focustrap"}activate(){this._isActive||(this._config.autofocus&&this._config.trapElement.focus(),fe.off(document,tn),fe.on(document,en,(t=>this._handleFocusin(t))),fe.on(document,nn,(t=>this._handleKeydown(t))),this._isActive=!0)}deactivate(){this._isActive&&(this._isActive=!1,fe.off(document,tn))}_handleFocusin(t){const{trapElement:e}=this._config;if(t.target===document||t.target===e||e.contains(t.target))return;const i=we.focusableChildren(e);0===i.length?e.focus():this._lastTabNavDirection===sn?i[i.length-1].focus():i[0].focus()}_handleKeydown(t){"Tab"===t.key&&(this._lastTabNavDirection=t.shiftKey?sn:"forward")}}const ln=".fixed-top, .fixed-bottom, .is-fixed, .sticky-top",cn=".sticky-top",hn="padding-right",dn="margin-right";class un{constructor(){this._element=document.body}getWidth(){const t=document.documentElement.clientWidth;return Math.abs(window.innerWidth-t)}hide(){const t=this.getWidth();this._disableOverFlow(),this._setElementAttributes(this._element,hn,(e=>e+t)),this._setElementAttributes(ln,hn,(e=>e+t)),this._setElementAttributes(cn,dn,(e=>e-t))}reset(){this._resetElementAttributes(this._element,"overflow"),this._resetElementAttributes(this._element,hn),this._resetElementAttributes(ln,hn),this._resetElementAttributes(cn,dn)}isOverflowing(){return this.getWidth()>0}_disableOverFlow(){this._saveInitialAttribute(this._element,"overflow"),this._element.style.overflow="hidden"}_setElementAttributes(t,e,i){const n=this.getWidth();this._applyManipulationCallback(t,(t=>{if(t!==this._element&&window.innerWidth>t.clientWidth+n)return;this._saveInitialAttribute(t,e);const s=window.getComputedStyle(t).getPropertyValue(e);t.style.setProperty(e,`${i(Number.parseFloat(s))}px`)}))}_saveInitialAttribute(t,e){const i=t.style.getPropertyValue(e);i&&_e.setDataAttribute(t,e,i)}_resetElementAttributes(t,e){this._applyManipulationCallback(t,(t=>{const i=_e.getDataAttribute(t,e);null!==i?(_e.removeDataAttribute(t,e),t.style.setProperty(e,i)):t.style.removeProperty(e)}))}_applyManipulationCallback(t,e){if(Ft(t))e(t);else for(const i of we.find(t,this._element))e(i)}}const fn=".bs.modal",pn=`hide${fn}`,mn=`hidePrevented${fn}`,gn=`hidden${fn}`,_n=`show${fn}`,bn=`shown${fn}`,vn=`resize${fn}`,yn=`click.dismiss${fn}`,wn=`mousedown.dismiss${fn}`,En=`keydown.dismiss${fn}`,An=`click${fn}.data-api`,Tn="modal-open",Cn="show",On="modal-static",xn={backdrop:!0,focus:!0,keyboard:!0},kn={backdrop:"(boolean|string)",focus:"boolean",keyboard:"boolean"};class Ln extends ve{constructor(t,e){super(t,e),this._dialog=we.findOne(".modal-dialog",this._element),this._backdrop=this._initializeBackDrop(),this._focustrap=this._initializeFocusTrap(),this._isShown=!1,this._isTransitioning=!1,this._scrollBar=new un,this._addEventListeners()}static get Default(){return xn}static get DefaultType(){return kn}static get NAME(){return"modal"}toggle(t){return this._isShown?this.hide():this.show(t)}show(t){this._isShown||this._isTransitioning||fe.trigger(this._element,_n,{relatedTarget:t}).defaultPrevented||(this._isShown=!0,this._isTransitioning=!0,this._scrollBar.hide(),document.body.classList.add(Tn),this._adjustDialog(),this._backdrop.show((()=>this._showElement(t))))}hide(){this._isShown&&!this._isTransitioning&&(fe.trigger(this._element,pn).defaultPrevented||(this._isShown=!1,this._isTransitioning=!0,this._focustrap.deactivate(),this._element.classList.remove(Cn),this._queueCallback((()=>this._hideModal()),this._element,this._isAnimated())))}dispose(){fe.off(window,fn),fe.off(this._dialog,fn),this._backdrop.dispose(),this._focustrap.deactivate(),super.dispose()}handleUpdate(){this._adjustDialog()}_initializeBackDrop(){return new Zi({isVisible:Boolean(this._config.backdrop),isAnimated:this._isAnimated()})}_initializeFocusTrap(){return new an({trapElement:this._element})}_showElement(t){document.body.contains(this._element)||document.body.append(this._element),this._element.style.display="block",this._element.removeAttribute("aria-hidden"),this._element.setAttribute("aria-modal",!0),this._element.setAttribute("role","dialog"),this._element.scrollTop=0;const e=we.findOne(".modal-body",this._dialog);e&&(e.scrollTop=0),qt(this._element),this._element.classList.add(Cn),this._queueCallback((()=>{this._config.focus&&this._focustrap.activate(),this._isTransitioning=!1,fe.trigger(this._element,bn,{relatedTarget:t})}),this._dialog,this._isAnimated())}_addEventListeners(){fe.on(this._element,En,(t=>{"Escape"===t.key&&(this._config.keyboard?this.hide():this._triggerBackdropTransition())})),fe.on(window,vn,(()=>{this._isShown&&!this._isTransitioning&&this._adjustDialog()})),fe.on(this._element,wn,(t=>{fe.one(this._element,yn,(e=>{this._element===t.target&&this._element===e.target&&("static"!==this._config.backdrop?this._config.backdrop&&this.hide():this._triggerBackdropTransition())}))}))}_hideModal(){this._element.style.display="none",this._element.setAttribute("aria-hidden",!0),this._element.removeAttribute("aria-modal"),this._element.removeAttribute("role"),this._isTransitioning=!1,this._backdrop.hide((()=>{document.body.classList.remove(Tn),this._resetAdjustments(),this._scrollBar.reset(),fe.trigger(this._element,gn)}))}_isAnimated(){return this._element.classList.contains("fade")}_triggerBackdropTransition(){if(fe.trigger(this._element,mn).defaultPrevented)return;const t=this._element.scrollHeight>document.documentElement.clientHeight,e=this._element.style.overflowY;"hidden"===e||this._element.classList.contains(On)||(t||(this._element.style.overflowY="hidden"),this._element.classList.add(On),this._queueCallback((()=>{this._element.classList.remove(On),this._queueCallback((()=>{this._element.style.overflowY=e}),this._dialog)}),this._dialog),this._element.focus())}_adjustDialog(){const t=this._element.scrollHeight>document.documentElement.clientHeight,e=this._scrollBar.getWidth(),i=e>0;if(i&&!t){const t=Kt()?"paddingLeft":"paddingRight";this._element.style[t]=`${e}px`}if(!i&&t){const t=Kt()?"paddingRight":"paddingLeft";this._element.style[t]=`${e}px`}}_resetAdjustments(){this._element.style.paddingLeft="",this._element.style.paddingRight=""}static jQueryInterface(t,e){return this.each((function(){const i=Ln.getOrCreateInstance(this,t);if("string"==typeof t){if(void 0===i[t])throw new TypeError(`No method named "${t}"`);i[t](e)}}))}}fe.on(document,An,'[data-bs-toggle="modal"]',(function(t){const e=we.getElementFromSelector(this);["A","AREA"].includes(this.tagName)&&t.preventDefault(),fe.one(e,_n,(t=>{t.defaultPrevented||fe.one(e,gn,(()=>{Bt(this)&&this.focus()}))}));const i=we.findOne(".modal.show");i&&Ln.getInstance(i).hide(),Ln.getOrCreateInstance(e).toggle(this)})),Ee(Ln),Qt(Ln);const Sn=".bs.offcanvas",Dn=".data-api",$n=`load${Sn}${Dn}`,In="show",Nn="showing",Pn="hiding",Mn=".offcanvas.show",jn=`show${Sn}`,Fn=`shown${Sn}`,Hn=`hide${Sn}`,Bn=`hidePrevented${Sn}`,Wn=`hidden${Sn}`,zn=`resize${Sn}`,Rn=`click${Sn}${Dn}`,qn=`keydown.dismiss${Sn}`,Vn={backdrop:!0,keyboard:!0,scroll:!1},Yn={backdrop:"(boolean|string)",keyboard:"boolean",scroll:"boolean"};class Kn extends ve{constructor(t,e){super(t,e),this._isShown=!1,this._backdrop=this._initializeBackDrop(),this._focustrap=this._initializeFocusTrap(),this._addEventListeners()}static get Default(){return Vn}static get DefaultType(){return Yn}static get NAME(){return"offcanvas"}toggle(t){return this._isShown?this.hide():this.show(t)}show(t){this._isShown||fe.trigger(this._element,jn,{relatedTarget:t}).defaultPrevented||(this._isShown=!0,this._backdrop.show(),this._config.scroll||(new un).hide(),this._element.setAttribute("aria-modal",!0),this._element.setAttribute("role","dialog"),this._element.classList.add(Nn),this._queueCallback((()=>{this._config.scroll&&!this._config.backdrop||this._focustrap.activate(),this._element.classList.add(In),this._element.classList.remove(Nn),fe.trigger(this._element,Fn,{relatedTarget:t})}),this._element,!0))}hide(){this._isShown&&(fe.trigger(this._element,Hn).defaultPrevented||(this._focustrap.deactivate(),this._element.blur(),this._isShown=!1,this._element.classList.add(Pn),this._backdrop.hide(),this._queueCallback((()=>{this._element.classList.remove(In,Pn),this._element.removeAttribute("aria-modal"),this._element.removeAttribute("role"),this._config.scroll||(new un).reset(),fe.trigger(this._element,Wn)}),this._element,!0)))}dispose(){this._backdrop.dispose(),this._focustrap.deactivate(),super.dispose()}_initializeBackDrop(){const t=Boolean(this._config.backdrop);return new Zi({className:"offcanvas-backdrop",isVisible:t,isAnimated:!0,rootElement:this._element.parentNode,clickCallback:t?()=>{"static"!==this._config.backdrop?this.hide():fe.trigger(this._element,Bn)}:null})}_initializeFocusTrap(){return new an({trapElement:this._element})}_addEventListeners(){fe.on(this._element,qn,(t=>{"Escape"===t.key&&(this._config.keyboard?this.hide():fe.trigger(this._element,Bn))}))}static jQueryInterface(t){return this.each((function(){const e=Kn.getOrCreateInstance(this,t);if("string"==typeof t){if(void 0===e[t]||t.startsWith("_")||"constructor"===t)throw new TypeError(`No method named "${t}"`);e[t](this)}}))}}fe.on(document,Rn,'[data-bs-toggle="offcanvas"]',(function(t){const e=we.getElementFromSelector(this);if(["A","AREA"].includes(this.tagName)&&t.preventDefault(),Wt(this))return;fe.one(e,Wn,(()=>{Bt(this)&&this.focus()}));const i=we.findOne(Mn);i&&i!==e&&Kn.getInstance(i).hide(),Kn.getOrCreateInstance(e).toggle(this)})),fe.on(window,$n,(()=>{for(const t of we.find(Mn))Kn.getOrCreateInstance(t).show()})),fe.on(window,zn,(()=>{for(const t of we.find("[aria-modal][class*=show][class*=offcanvas-]"))"fixed"!==getComputedStyle(t).position&&Kn.getOrCreateInstance(t).hide()})),Ee(Kn),Qt(Kn);const Qn={"*":["class","dir","id","lang","role",/^aria-[\w-]*$/i],a:["target","href","title","rel"],area:[],b:[],br:[],col:[],code:[],dd:[],div:[],dl:[],dt:[],em:[],hr:[],h1:[],h2:[],h3:[],h4:[],h5:[],h6:[],i:[],img:["src","srcset","alt","title","width","height"],li:[],ol:[],p:[],pre:[],s:[],small:[],span:[],sub:[],sup:[],strong:[],u:[],ul:[]},Xn=new Set(["background","cite","href","itemtype","longdesc","poster","src","xlink:href"]),Un=/^(?!javascript:)(?:[a-z0-9+.-]+:|[^&:/?#]*(?:[/?#]|$))/i,Gn=(t,e)=>{const i=t.nodeName.toLowerCase();return e.includes(i)?!Xn.has(i)||Boolean(Un.test(t.nodeValue)):e.filter((t=>t instanceof RegExp)).some((t=>t.test(i)))},Jn={allowList:Qn,content:{},extraClass:"",html:!1,sanitize:!0,sanitizeFn:null,template:"<div></div>"},Zn={allowList:"object",content:"object",extraClass:"(string|function)",html:"boolean",sanitize:"boolean",sanitizeFn:"(null|function)",template:"string"},ts={entry:"(string|element|function|null)",selector:"(string|element)"};class es extends be{constructor(t){super(),this._config=this._getConfig(t)}static get Default(){return Jn}static get DefaultType(){return Zn}static get NAME(){return"TemplateFactory"}getContent(){return Object.values(this._config.content).map((t=>this._resolvePossibleFunction(t))).filter(Boolean)}hasContent(){return this.getContent().length>0}changeContent(t){return this._checkContent(t),this._config.content={...this._config.content,...t},this}toHtml(){const t=document.createElement("div");t.innerHTML=this._maybeSanitize(this._config.template);for(const[e,i]of Object.entries(this._config.content))this._setContent(t,i,e);const e=t.children[0],i=this._resolvePossibleFunction(this._config.extraClass);return i&&e.classList.add(...i.split(" ")),e}_typeCheckConfig(t){super._typeCheckConfig(t),this._checkContent(t.content)}_checkContent(t){for(const[e,i]of Object.entries(t))super._typeCheckConfig({selector:e,entry:i},ts)}_setContent(t,e,i){const n=we.findOne(i,t);n&&((e=this._resolvePossibleFunction(e))?Ft(e)?this._putElementInTemplate(Ht(e),n):this._config.html?n.innerHTML=this._maybeSanitize(e):n.textContent=e:n.remove())}_maybeSanitize(t){return this._config.sanitize?function(t,e,i){if(!t.length)return t;if(i&&"function"==typeof i)return i(t);const n=(new window.DOMParser).parseFromString(t,"text/html"),s=[].concat(...n.body.querySelectorAll("*"));for(const t of s){const i=t.nodeName.toLowerCase();if(!Object.keys(e).includes(i)){t.remove();continue}const n=[].concat(...t.attributes),s=[].concat(e["*"]||[],e[i]||[]);for(const e of n)Gn(e,s)||t.removeAttribute(e.nodeName)}return n.body.innerHTML}(t,this._config.allowList,this._config.sanitizeFn):t}_resolvePossibleFunction(t){return Xt(t,[this])}_putElementInTemplate(t,e){if(this._config.html)return e.innerHTML="",void e.append(t);e.textContent=t.textContent}}const is=new Set(["sanitize","allowList","sanitizeFn"]),ns="fade",ss="show",os=".tooltip-inner",rs=".modal",as="hide.bs.modal",ls="hover",cs="focus",hs={AUTO:"auto",TOP:"top",RIGHT:Kt()?"left":"right",BOTTOM:"bottom",LEFT:Kt()?"right":"left"},ds={allowList:Qn,animation:!0,boundary:"clippingParents",container:!1,customClass:"",delay:0,fallbackPlacements:["top","right","bottom","left"],html:!1,offset:[0,6],placement:"top",popperConfig:null,sanitize:!0,sanitizeFn:null,selector:!1,template:'<div class="tooltip" role="tooltip"><div class="tooltip-arrow"></div><div class="tooltip-inner"></div></div>',title:"",trigger:"hover focus"},us={allowList:"object",animation:"boolean",boundary:"(string|element)",container:"(string|element|boolean)",customClass:"(string|function)",delay:"(number|object)",fallbackPlacements:"array",html:"boolean",offset:"(array|string|function)",placement:"(string|function)",popperConfig:"(null|object|function)",sanitize:"boolean",sanitizeFn:"(null|function)",selector:"(string|boolean)",template:"string",title:"(string|element|function)",trigger:"string"};class fs extends ve{constructor(t,i){if(void 0===e)throw new TypeError("Bootstrap's tooltips require Popper (https://popper.js.org)");super(t,i),this._isEnabled=!0,this._timeout=0,this._isHovered=null,this._activeTrigger={},this._popper=null,this._templateFactory=null,this._newContent=null,this.tip=null,this._setListeners(),this._config.selector||this._fixTitle()}static get Default(){return ds}static get DefaultType(){return us}static get NAME(){return"tooltip"}enable(){this._isEnabled=!0}disable(){this._isEnabled=!1}toggleEnabled(){this._isEnabled=!this._isEnabled}toggle(){this._isEnabled&&(this._activeTrigger.click=!this._activeTrigger.click,this._isShown()?this._leave():this._enter())}dispose(){clearTimeout(this._timeout),fe.off(this._element.closest(rs),as,this._hideModalHandler),this._element.getAttribute("data-bs-original-title")&&this._element.setAttribute("title",this._element.getAttribute("data-bs-original-title")),this._disposePopper(),super.dispose()}show(){if("none"===this._element.style.display)throw new Error("Please use show on visible elements");if(!this._isWithContent()||!this._isEnabled)return;const t=fe.trigger(this._element,this.constructor.eventName("show")),e=(zt(this._element)||this._element.ownerDocument.documentElement).contains(this._element);if(t.defaultPrevented||!e)return;this._disposePopper();const i=this._getTipElement();this._element.setAttribute("aria-describedby",i.getAttribute("id"));const{container:n}=this._config;if(this._element.ownerDocument.documentElement.contains(this.tip)||(n.append(i),fe.trigger(this._element,this.constructor.eventName("inserted"))),this._popper=this._createPopper(i),i.classList.add(ss),"ontouchstart"in document.documentElement)for(const t of[].concat(...document.body.children))fe.on(t,"mouseover",Rt);this._queueCallback((()=>{fe.trigger(this._element,this.constructor.eventName("shown")),!1===this._isHovered&&this._leave(),this._isHovered=!1}),this.tip,this._isAnimated())}hide(){if(this._isShown()&&!fe.trigger(this._element,this.constructor.eventName("hide")).defaultPrevented){if(this._getTipElement().classList.remove(ss),"ontouchstart"in document.documentElement)for(const t of[].concat(...document.body.children))fe.off(t,"mouseover",Rt);this._activeTrigger.click=!1,this._activeTrigger[cs]=!1,this._activeTrigger[ls]=!1,this._isHovered=null,this._queueCallback((()=>{this._isWithActiveTrigger()||(this._isHovered||this._disposePopper(),this._element.removeAttribute("aria-describedby"),fe.trigger(this._element,this.constructor.eventName("hidden")))}),this.tip,this._isAnimated())}}update(){this._popper&&this._popper.update()}_isWithContent(){return Boolean(this._getTitle())}_getTipElement(){return this.tip||(this.tip=this._createTipElement(this._newContent||this._getContentForTemplate())),this.tip}_createTipElement(t){const e=this._getTemplateFactory(t).toHtml();if(!e)return null;e.classList.remove(ns,ss),e.classList.add(`bs-${this.constructor.NAME}-auto`);const i=(t=>{do{t+=Math.floor(1e6*Math.random())}while(document.getElementById(t));return t})(this.constructor.NAME).toString();return e.setAttribute("id",i),this._isAnimated()&&e.classList.add(ns),e}setContent(t){this._newContent=t,this._isShown()&&(this._disposePopper(),this.show())}_getTemplateFactory(t){return this._templateFactory?this._templateFactory.changeContent(t):this._templateFactory=new es({...this._config,content:t,extraClass:this._resolvePossibleFunction(this._config.customClass)}),this._templateFactory}_getContentForTemplate(){return{[os]:this._getTitle()}}_getTitle(){return this._resolvePossibleFunction(this._config.title)||this._element.getAttribute("data-bs-original-title")}_initializeOnDelegatedTarget(t){return this.constructor.getOrCreateInstance(t.delegateTarget,this._getDelegateConfig())}_isAnimated(){return this._config.animation||this.tip&&this.tip.classList.contains(ns)}_isShown(){return this.tip&&this.tip.classList.contains(ss)}_createPopper(t){const e=Xt(this._config.placement,[this,t,this._element]),i=hs[e.toUpperCase()];return Dt(this._element,t,this._getPopperConfig(i))}_getOffset(){const{offset:t}=this._config;return"string"==typeof t?t.split(",").map((t=>Number.parseInt(t,10))):"function"==typeof t?e=>t(e,this._element):t}_resolvePossibleFunction(t){return Xt(t,[this._element])}_getPopperConfig(t){const e={placement:t,modifiers:[{name:"flip",options:{fallbackPlacements:this._config.fallbackPlacements}},{name:"offset",options:{offset:this._getOffset()}},{name:"preventOverflow",options:{boundary:this._config.boundary}},{name:"arrow",options:{element:`.${this.constructor.NAME}-arrow`}},{name:"preSetPlacement",enabled:!0,phase:"beforeMain",fn:t=>{this._getTipElement().setAttribute("data-popper-placement",t.state.placement)}}]};return{...e,...Xt(this._config.popperConfig,[e])}}_setListeners(){const t=this._config.trigger.split(" ");for(const e of t)if("click"===e)fe.on(this._element,this.constructor.eventName("click"),this._config.selector,(t=>{this._initializeOnDelegatedTarget(t).toggle()}));else if("manual"!==e){const t=e===ls?this.constructor.eventName("mouseenter"):this.constructor.eventName("focusin"),i=e===ls?this.constructor.eventName("mouseleave"):this.constructor.eventName("focusout");fe.on(this._element,t,this._config.selector,(t=>{const e=this._initializeOnDelegatedTarget(t);e._activeTrigger["focusin"===t.type?cs:ls]=!0,e._enter()})),fe.on(this._element,i,this._config.selector,(t=>{const e=this._initializeOnDelegatedTarget(t);e._activeTrigger["focusout"===t.type?cs:ls]=e._element.contains(t.relatedTarget),e._leave()}))}this._hideModalHandler=()=>{this._element&&this.hide()},fe.on(this._element.closest(rs),as,this._hideModalHandler)}_fixTitle(){const t=this._element.getAttribute("title");t&&(this._element.getAttribute("aria-label")||this._element.textContent.trim()||this._element.setAttribute("aria-label",t),this._element.setAttribute("data-bs-original-title",t),this._element.removeAttribute("title"))}_enter(){this._isShown()||this._isHovered?this._isHovered=!0:(this._isHovered=!0,this._setTimeout((()=>{this._isHovered&&this.show()}),this._config.delay.show))}_leave(){this._isWithActiveTrigger()||(this._isHovered=!1,this._setTimeout((()=>{this._isHovered||this.hide()}),this._config.delay.hide))}_setTimeout(t,e){clearTimeout(this._timeout),this._timeout=setTimeout(t,e)}_isWithActiveTrigger(){return Object.values(this._activeTrigger).includes(!0)}_getConfig(t){const e=_e.getDataAttributes(this._element);for(const t of Object.keys(e))is.has(t)&&delete e[t];return t={...e,..."object"==typeof t&&t?t:{}},t=this._mergeConfigObj(t),t=this._configAfterMerge(t),this._typeCheckConfig(t),t}_configAfterMerge(t){return t.container=!1===t.container?document.body:Ht(t.container),"number"==typeof t.delay&&(t.delay={show:t.delay,hide:t.delay}),"number"==typeof t.title&&(t.title=t.title.toString()),"number"==typeof t.content&&(t.content=t.content.toString()),t}_getDelegateConfig(){const t={};for(const[e,i]of Object.entries(this._config))this.constructor.Default[e]!==i&&(t[e]=i);return t.selector=!1,t.trigger="manual",t}_disposePopper(){this._popper&&(this._popper.destroy(),this._popper=null),this.tip&&(this.tip.remove(),this.tip=null)}static jQueryInterface(t){return this.each((function(){const e=fs.getOrCreateInstance(this,t);if("string"==typeof t){if(void 0===e[t])throw new TypeError(`No method named "${t}"`);e[t]()}}))}}Qt(fs);const ps=".popover-header",ms=".popover-body",gs={...fs.Default,content:"",offset:[0,8],placement:"right",template:'<div class="popover" role="tooltip"><div class="popover-arrow"></div><h3 class="popover-header"></h3><div class="popover-body"></div></div>',trigger:"click"},_s={...fs.DefaultType,content:"(null|string|element|function)"};class bs extends fs{static get Default(){return gs}static get DefaultType(){return _s}static get NAME(){return"popover"}_isWithContent(){return this._getTitle()||this._getContent()}_getContentForTemplate(){return{[ps]:this._getTitle(),[ms]:this._getContent()}}_getContent(){return this._resolvePossibleFunction(this._config.content)}static jQueryInterface(t){return this.each((function(){const e=bs.getOrCreateInstance(this,t);if("string"==typeof t){if(void 0===e[t])throw new TypeError(`No method named "${t}"`);e[t]()}}))}}Qt(bs);const vs=".bs.scrollspy",ys=`activate${vs}`,ws=`click${vs}`,Es=`load${vs}.data-api`,As="active",Ts="[href]",Cs=".nav-link",Os=`${Cs}, .nav-item > ${Cs}, .list-group-item`,xs={offset:null,rootMargin:"0px 0px -25%",smoothScroll:!1,target:null,threshold:[.1,.5,1]},ks={offset:"(number|null)",rootMargin:"string",smoothScroll:"boolean",target:"element",threshold:"array"};class Ls extends ve{constructor(t,e){super(t,e),this._targetLinks=new Map,this._observableSections=new Map,this._rootElement="visible"===getComputedStyle(this._element).overflowY?null:this._element,this._activeTarget=null,this._observer=null,this._previousScrollData={visibleEntryTop:0,parentScrollTop:0},this.refresh()}static get Default(){return xs}static get DefaultType(){return ks}static get NAME(){return"scrollspy"}refresh(){this._initializeTargetsAndObservables(),this._maybeEnableSmoothScroll(),this._observer?this._observer.disconnect():this._observer=this._getNewObserver();for(const t of this._observableSections.values())this._observer.observe(t)}dispose(){this._observer.disconnect(),super.dispose()}_configAfterMerge(t){return t.target=Ht(t.target)||document.body,t.rootMargin=t.offset?`${t.offset}px 0px -30%`:t.rootMargin,"string"==typeof t.threshold&&(t.threshold=t.threshold.split(",").map((t=>Number.parseFloat(t)))),t}_maybeEnableSmoothScroll(){this._config.smoothScroll&&(fe.off(this._config.target,ws),fe.on(this._config.target,ws,Ts,(t=>{const e=this._observableSections.get(t.target.hash);if(e){t.preventDefault();const i=this._rootElement||window,n=e.offsetTop-this._element.offsetTop;if(i.scrollTo)return void i.scrollTo({top:n,behavior:"smooth"});i.scrollTop=n}})))}_getNewObserver(){const t={root:this._rootElement,threshold:this._config.threshold,rootMargin:this._config.rootMargin};return new IntersectionObserver((t=>this._observerCallback(t)),t)}_observerCallback(t){const e=t=>this._targetLinks.get(`#${t.target.id}`),i=t=>{this._previousScrollData.visibleEntryTop=t.target.offsetTop,this._process(e(t))},n=(this._rootElement||document.documentElement).scrollTop,s=n>=this._previousScrollData.parentScrollTop;this._previousScrollData.parentScrollTop=n;for(const o of t){if(!o.isIntersecting){this._activeTarget=null,this._clearActiveClass(e(o));continue}const t=o.target.offsetTop>=this._previousScrollData.visibleEntryTop;if(s&&t){if(i(o),!n)return}else s||t||i(o)}}_initializeTargetsAndObservables(){this._targetLinks=new Map,this._observableSections=new Map;const t=we.find(Ts,this._config.target);for(const e of t){if(!e.hash||Wt(e))continue;const t=we.findOne(decodeURI(e.hash),this._element);Bt(t)&&(this._targetLinks.set(decodeURI(e.hash),e),this._observableSections.set(e.hash,t))}}_process(t){this._activeTarget!==t&&(this._clearActiveClass(this._config.target),this._activeTarget=t,t.classList.add(As),this._activateParents(t),fe.trigger(this._element,ys,{relatedTarget:t}))}_activateParents(t){if(t.classList.contains("dropdown-item"))we.findOne(".dropdown-toggle",t.closest(".dropdown")).classList.add(As);else for(const e of we.parents(t,".nav, .list-group"))for(const t of we.prev(e,Os))t.classList.add(As)}_clearActiveClass(t){t.classList.remove(As);const e=we.find(`${Ts}.${As}`,t);for(const t of e)t.classList.remove(As)}static jQueryInterface(t){return this.each((function(){const e=Ls.getOrCreateInstance(this,t);if("string"==typeof t){if(void 0===e[t]||t.startsWith("_")||"constructor"===t)throw new TypeError(`No method named "${t}"`);e[t]()}}))}}fe.on(window,Es,(()=>{for(const t of we.find('[data-bs-spy="scroll"]'))Ls.getOrCreateInstance(t)})),Qt(Ls);const Ss=".bs.tab",Ds=`hide${Ss}`,$s=`hidden${Ss}`,Is=`show${Ss}`,Ns=`shown${Ss}`,Ps=`click${Ss}`,Ms=`keydown${Ss}`,js=`load${Ss}`,Fs="ArrowLeft",Hs="ArrowRight",Bs="ArrowUp",Ws="ArrowDown",zs="Home",Rs="End",qs="active",Vs="fade",Ys="show",Ks=".dropdown-toggle",Qs=`:not(${Ks})`,Xs='[data-bs-toggle="tab"], [data-bs-toggle="pill"], [data-bs-toggle="list"]',Us=`.nav-link${Qs}, .list-group-item${Qs}, [role="tab"]${Qs}, ${Xs}`,Gs=`.${qs}[data-bs-toggle="tab"], .${qs}[data-bs-toggle="pill"], .${qs}[data-bs-toggle="list"]`;class Js extends ve{constructor(t){super(t),this._parent=this._element.closest('.list-group, .nav, [role="tablist"]'),this._parent&&(this._setInitialAttributes(this._parent,this._getChildren()),fe.on(this._element,Ms,(t=>this._keydown(t))))}static get NAME(){return"tab"}show(){const t=this._element;if(this._elemIsActive(t))return;const e=this._getActiveElem(),i=e?fe.trigger(e,Ds,{relatedTarget:t}):null;fe.trigger(t,Is,{relatedTarget:e}).defaultPrevented||i&&i.defaultPrevented||(this._deactivate(e,t),this._activate(t,e))}_activate(t,e){t&&(t.classList.add(qs),this._activate(we.getElementFromSelector(t)),this._queueCallback((()=>{"tab"===t.getAttribute("role")?(t.removeAttribute("tabindex"),t.setAttribute("aria-selected",!0),this._toggleDropDown(t,!0),fe.trigger(t,Ns,{relatedTarget:e})):t.classList.add(Ys)}),t,t.classList.contains(Vs)))}_deactivate(t,e){t&&(t.classList.remove(qs),t.blur(),this._deactivate(we.getElementFromSelector(t)),this._queueCallback((()=>{"tab"===t.getAttribute("role")?(t.setAttribute("aria-selected",!1),t.setAttribute("tabindex","-1"),this._toggleDropDown(t,!1),fe.trigger(t,$s,{relatedTarget:e})):t.classList.remove(Ys)}),t,t.classList.contains(Vs)))}_keydown(t){if(![Fs,Hs,Bs,Ws,zs,Rs].includes(t.key))return;t.stopPropagation(),t.preventDefault();const e=this._getChildren().filter((t=>!Wt(t)));let i;if([zs,Rs].includes(t.key))i=e[t.key===zs?0:e.length-1];else{const n=[Hs,Ws].includes(t.key);i=Gt(e,t.target,n,!0)}i&&(i.focus({preventScroll:!0}),Js.getOrCreateInstance(i).show())}_getChildren(){return we.find(Us,this._parent)}_getActiveElem(){return this._getChildren().find((t=>this._elemIsActive(t)))||null}_setInitialAttributes(t,e){this._setAttributeIfNotExists(t,"role","tablist");for(const t of e)this._setInitialAttributesOnChild(t)}_setInitialAttributesOnChild(t){t=this._getInnerElement(t);const e=this._elemIsActive(t),i=this._getOuterElement(t);t.setAttribute("aria-selected",e),i!==t&&this._setAttributeIfNotExists(i,"role","presentation"),e||t.setAttribute("tabindex","-1"),this._setAttributeIfNotExists(t,"role","tab"),this._setInitialAttributesOnTargetPanel(t)}_setInitialAttributesOnTargetPanel(t){const e=we.getElementFromSelector(t);e&&(this._setAttributeIfNotExists(e,"role","tabpanel"),t.id&&this._setAttributeIfNotExists(e,"aria-labelledby",`${t.id}`))}_toggleDropDown(t,e){const i=this._getOuterElement(t);if(!i.classList.contains("dropdown"))return;const n=(t,n)=>{const s=we.findOne(t,i);s&&s.classList.toggle(n,e)};n(Ks,qs),n(".dropdown-menu",Ys),i.setAttribute("aria-expanded",e)}_setAttributeIfNotExists(t,e,i){t.hasAttribute(e)||t.setAttribute(e,i)}_elemIsActive(t){return t.classList.contains(qs)}_getInnerElement(t){return t.matches(Us)?t:we.findOne(Us,t)}_getOuterElement(t){return t.closest(".nav-item, .list-group-item")||t}static jQueryInterface(t){return this.each((function(){const e=Js.getOrCreateInstance(this);if("string"==typeof t){if(void 0===e[t]||t.startsWith("_")||"constructor"===t)throw new TypeError(`No method named "${t}"`);e[t]()}}))}}fe.on(document,Ps,Xs,(function(t){["A","AREA"].includes(this.tagName)&&t.preventDefault(),Wt(this)||Js.getOrCreateInstance(this).show()})),fe.on(window,js,(()=>{for(const t of we.find(Gs))Js.getOrCreateInstance(t)})),Qt(Js);const Zs=".bs.toast",to=`mouseover${Zs}`,eo=`mouseout${Zs}`,io=`focusin${Zs}`,no=`focusout${Zs}`,so=`hide${Zs}`,oo=`hidden${Zs}`,ro=`show${Zs}`,ao=`shown${Zs}`,lo="hide",co="show",ho="showing",uo={animation:"boolean",autohide:"boolean",delay:"number"},fo={animation:!0,autohide:!0,delay:5e3};class po extends ve{constructor(t,e){super(t,e),this._timeout=null,this._hasMouseInteraction=!1,this._hasKeyboardInteraction=!1,this._setListeners()}static get Default(){return fo}static get DefaultType(){return uo}static get NAME(){return"toast"}show(){fe.trigger(this._element,ro).defaultPrevented||(this._clearTimeout(),this._config.animation&&this._element.classList.add("fade"),this._element.classList.remove(lo),qt(this._element),this._element.classList.add(co,ho),this._queueCallback((()=>{this._element.classList.remove(ho),fe.trigger(this._element,ao),this._maybeScheduleHide()}),this._element,this._config.animation))}hide(){this.isShown()&&(fe.trigger(this._element,so).defaultPrevented||(this._element.classList.add(ho),this._queueCallback((()=>{this._element.classList.add(lo),this._element.classList.remove(ho,co),fe.trigger(this._element,oo)}),this._element,this._config.animation)))}dispose(){this._clearTimeout(),this.isShown()&&this._element.classList.remove(co),super.dispose()}isShown(){return this._element.classList.contains(co)}_maybeScheduleHide(){this._config.autohide&&(this._hasMouseInteraction||this._hasKeyboardInteraction||(this._timeout=setTimeout((()=>{this.hide()}),this._config.delay)))}_onInteraction(t,e){switch(t.type){case"mouseover":case"mouseout":this._hasMouseInteraction=e;break;case"focusin":case"focusout":this._hasKeyboardInteraction=e}if(e)return void this._clearTimeout();const i=t.relatedTarget;this._element===i||this._element.contains(i)||this._maybeScheduleHide()}_setListeners(){fe.on(this._element,to,(t=>this._onInteraction(t,!0))),fe.on(this._element,eo,(t=>this._onInteraction(t,!1))),fe.on(this._element,io,(t=>this._onInteraction(t,!0))),fe.on(this._element,no,(t=>this._onInteraction(t,!1)))}_clearTimeout(){clearTimeout(this._timeout),this._timeout=null}static jQueryInterface(t){return this.each((function(){const e=po.getOrCreateInstance(this,t);if("string"==typeof t){if(void 0===e[t])throw new TypeError(`No method named "${t}"`);e[t](this)}}))}}function mo(t){"loading"!=document.readyState?t():document.addEventListener("DOMContentLoaded",t)}Ee(po),Qt(po),mo((function(){[].slice.call(document.querySelectorAll('[data-bs-toggle="tooltip"]')).map((function(t){return new fs(t,{delay:{show:500,hide:100}})}))})),mo((function(){document.getElementById("pst-back-to-top").addEventListener("click",(function(){document.body.scrollTop=0,document.documentElement.scrollTop=0}))})),mo((function(){var t=document.getElementById("pst-back-to-top"),e=document.getElementsByClassName("bd-header")[0].getBoundingClientRect();window.addEventListener("scroll",(function(){this.oldScroll>this.scrollY&&this.scrollY>e.bottom?t.style.display="block":t.style.display="none",this.oldScroll=this.scrollY}))})),window.bootstrap=i})();
+//# sourceMappingURL=bootstrap.js.map
\ No newline at end of file
diff --git a/_static/scripts/bootstrap.js.LICENSE.txt b/_static/scripts/bootstrap.js.LICENSE.txt
new file mode 100644
index 000000000..28755c2c5
--- /dev/null
+++ b/_static/scripts/bootstrap.js.LICENSE.txt
@@ -0,0 +1,5 @@
+/*!
+  * Bootstrap v5.3.3 (https://getbootstrap.com/)
+  * Copyright 2011-2024 The Bootstrap Authors (https://github.com/twbs/bootstrap/graphs/contributors)
+  * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)
+  */
diff --git a/_static/scripts/bootstrap.js.map b/_static/scripts/bootstrap.js.map
new file mode 100644
index 000000000..4a3502aeb
--- /dev/null
+++ b/_static/scripts/bootstrap.js.map
@@ -0,0 +1 @@
+{"version":3,"file":"scripts/bootstrap.js","mappings":";mBACA,IAAIA,EAAsB,CCA1BA,EAAwB,CAACC,EAASC,KACjC,IAAI,IAAIC,KAAOD,EACXF,EAAoBI,EAAEF,EAAYC,KAASH,EAAoBI,EAAEH,EAASE,IAC5EE,OAAOC,eAAeL,EAASE,EAAK,CAAEI,YAAY,EAAMC,IAAKN,EAAWC,IAE1E,ECNDH,EAAwB,CAACS,EAAKC,IAAUL,OAAOM,UAAUC,eAAeC,KAAKJ,EAAKC,GCClFV,EAAyBC,IACH,oBAAXa,QAA0BA,OAAOC,aAC1CV,OAAOC,eAAeL,EAASa,OAAOC,YAAa,CAAEC,MAAO,WAE7DX,OAAOC,eAAeL,EAAS,aAAc,CAAEe,OAAO,GAAO,01BCLvD,IAAI,EAAM,MACNC,EAAS,SACTC,EAAQ,QACRC,EAAO,OACPC,EAAO,OACPC,EAAiB,CAAC,EAAKJ,EAAQC,EAAOC,GACtCG,EAAQ,QACRC,EAAM,MACNC,EAAkB,kBAClBC,EAAW,WACXC,EAAS,SACTC,EAAY,YACZC,EAAmCP,EAAeQ,QAAO,SAAUC,EAAKC,GACjF,OAAOD,EAAIE,OAAO,CAACD,EAAY,IAAMT,EAAOS,EAAY,IAAMR,GAChE,GAAG,IACQ,EAA0B,GAAGS,OAAOX,EAAgB,CAACD,IAAOS,QAAO,SAAUC,EAAKC,GAC3F,OAAOD,EAAIE,OAAO,CAACD,EAAWA,EAAY,IAAMT,EAAOS,EAAY,IAAMR,GAC3E,GAAG,IAEQU,EAAa,aACbC,EAAO,OACPC,EAAY,YAEZC,EAAa,aACbC,EAAO,OACPC,EAAY,YAEZC,EAAc,cACdC,EAAQ,QACRC,EAAa,aACbC,EAAiB,CAACT,EAAYC,EAAMC,EAAWC,EAAYC,EAAMC,EAAWC,EAAaC,EAAOC,GC9B5F,SAASE,EAAYC,GAClC,OAAOA,GAAWA,EAAQC,UAAY,IAAIC,cAAgB,IAC5D,CCFe,SAASC,EAAUC,GAChC,GAAY,MAARA,EACF,OAAOC,OAGT,GAAwB,oBAApBD,EAAKE,WAAkC,CACzC,IAAIC,EAAgBH,EAAKG,cACzB,OAAOA,GAAgBA,EAAcC,aAAwBH,MAC/D,CAEA,OAAOD,CACT,CCTA,SAASK,EAAUL,GAEjB,OAAOA,aADUD,EAAUC,GAAMM,SACIN,aAAgBM,OACvD,CAEA,SAASC,EAAcP,GAErB,OAAOA,aADUD,EAAUC,GAAMQ,aACIR,aAAgBQ,WACvD,CAEA,SAASC,EAAaT,GAEpB,MAA0B,oBAAfU,aAKJV,aADUD,EAAUC,GAAMU,YACIV,aAAgBU,WACvD,CCwDA,SACEC,KAAM,cACNC,SAAS,EACTC,MAAO,QACPC,GA5EF,SAAqBC,GACnB,IAAIC,EAAQD,EAAKC,MACjB3D,OAAO4D,KAAKD,EAAME,UAAUC,SAAQ,SAAUR,GAC5C,IAAIS,EAAQJ,EAAMK,OAAOV,IAAS,CAAC,EAC/BW,EAAaN,EAAMM,WAAWX,IAAS,CAAC,EACxCf,EAAUoB,EAAME,SAASP,GAExBJ,EAAcX,IAAaD,EAAYC,KAO5CvC,OAAOkE,OAAO3B,EAAQwB,MAAOA,GAC7B/D,OAAO4D,KAAKK,GAAYH,SAAQ,SAAUR,GACxC,IAAI3C,EAAQsD,EAAWX,IAET,IAAV3C,EACF4B,EAAQ4B,gBAAgBb,GAExBf,EAAQ6B,aAAad,GAAgB,IAAV3C,EAAiB,GAAKA,EAErD,IACF,GACF,EAoDE0D,OAlDF,SAAgBC,GACd,IAAIX,EAAQW,EAAMX,MACdY,EAAgB,CAClBlD,OAAQ,CACNmD,SAAUb,EAAMc,QAAQC,SACxB5D,KAAM,IACN6D,IAAK,IACLC,OAAQ,KAEVC,MAAO,CACLL,SAAU,YAEZlD,UAAW,CAAC,GASd,OAPAtB,OAAOkE,OAAOP,EAAME,SAASxC,OAAO0C,MAAOQ,EAAclD,QACzDsC,EAAMK,OAASO,EAEXZ,EAAME,SAASgB,OACjB7E,OAAOkE,OAAOP,EAAME,SAASgB,MAAMd,MAAOQ,EAAcM,OAGnD,WACL7E,OAAO4D,KAAKD,EAAME,UAAUC,SAAQ,SAAUR,GAC5C,IAAIf,EAAUoB,EAAME,SAASP,GACzBW,EAAaN,EAAMM,WAAWX,IAAS,CAAC,EAGxCS,EAFkB/D,OAAO4D,KAAKD,EAAMK,OAAOzD,eAAe+C,GAAQK,EAAMK,OAAOV,GAAQiB,EAAcjB,IAE7E9B,QAAO,SAAUuC,EAAOe,GAElD,OADAf,EAAMe,GAAY,GACXf,CACT,GAAG,CAAC,GAECb,EAAcX,IAAaD,EAAYC,KAI5CvC,OAAOkE,OAAO3B,EAAQwB,MAAOA,GAC7B/D,OAAO4D,KAAKK,GAAYH,SAAQ,SAAUiB,GACxCxC,EAAQ4B,gBAAgBY,EAC1B,IACF,GACF,CACF,EASEC,SAAU,CAAC,kBCjFE,SAASC,EAAiBvD,GACvC,OAAOA,EAAUwD,MAAM,KAAK,EAC9B,CCHO,IAAI,EAAMC,KAAKC,IACX,EAAMD,KAAKE,IACXC,EAAQH,KAAKG,MCFT,SAASC,IACtB,IAAIC,EAASC,UAAUC,cAEvB,OAAc,MAAVF,GAAkBA,EAAOG,QAAUC,MAAMC,QAAQL,EAAOG,QACnDH,EAAOG,OAAOG,KAAI,SAAUC,GACjC,OAAOA,EAAKC,MAAQ,IAAMD,EAAKE,OACjC,IAAGC,KAAK,KAGHT,UAAUU,SACnB,CCTe,SAASC,IACtB,OAAQ,iCAAiCC,KAAKd,IAChD,CCCe,SAASe,EAAsB/D,EAASgE,EAAcC,QAC9C,IAAjBD,IACFA,GAAe,QAGO,IAApBC,IACFA,GAAkB,GAGpB,IAAIC,EAAalE,EAAQ+D,wBACrBI,EAAS,EACTC,EAAS,EAETJ,GAAgBrD,EAAcX,KAChCmE,EAASnE,EAAQqE,YAAc,GAAItB,EAAMmB,EAAWI,OAAStE,EAAQqE,aAAmB,EACxFD,EAASpE,EAAQuE,aAAe,GAAIxB,EAAMmB,EAAWM,QAAUxE,EAAQuE,cAAoB,GAG7F,IACIE,GADOhE,EAAUT,GAAWG,EAAUH,GAAWK,QAC3BoE,eAEtBC,GAAoBb,KAAsBI,EAC1CU,GAAKT,EAAW3F,MAAQmG,GAAoBD,EAAiBA,EAAeG,WAAa,IAAMT,EAC/FU,GAAKX,EAAW9B,KAAOsC,GAAoBD,EAAiBA,EAAeK,UAAY,IAAMV,EAC7FE,EAAQJ,EAAWI,MAAQH,EAC3BK,EAASN,EAAWM,OAASJ,EACjC,MAAO,CACLE,MAAOA,EACPE,OAAQA,EACRpC,IAAKyC,EACLvG,MAAOqG,EAAIL,EACXjG,OAAQwG,EAAIL,EACZjG,KAAMoG,EACNA,EAAGA,EACHE,EAAGA,EAEP,CCrCe,SAASE,EAAc/E,GACpC,IAAIkE,EAAaH,EAAsB/D,GAGnCsE,EAAQtE,EAAQqE,YAChBG,EAASxE,EAAQuE,aAUrB,OARI3B,KAAKoC,IAAId,EAAWI,MAAQA,IAAU,IACxCA,EAAQJ,EAAWI,OAGjB1B,KAAKoC,IAAId,EAAWM,OAASA,IAAW,IAC1CA,EAASN,EAAWM,QAGf,CACLG,EAAG3E,EAAQ4E,WACXC,EAAG7E,EAAQ8E,UACXR,MAAOA,EACPE,OAAQA,EAEZ,CCvBe,SAASS,EAASC,EAAQC,GACvC,IAAIC,EAAWD,EAAME,aAAeF,EAAME,cAE1C,GAAIH,EAAOD,SAASE,GAClB,OAAO,EAEJ,GAAIC,GAAYvE,EAAauE,GAAW,CACzC,IAAIE,EAAOH,EAEX,EAAG,CACD,GAAIG,GAAQJ,EAAOK,WAAWD,GAC5B,OAAO,EAITA,EAAOA,EAAKE,YAAcF,EAAKG,IACjC,OAASH,EACX,CAGF,OAAO,CACT,CCrBe,SAAS,EAAiBtF,GACvC,OAAOG,EAAUH,GAAS0F,iBAAiB1F,EAC7C,CCFe,SAAS2F,EAAe3F,GACrC,MAAO,CAAC,QAAS,KAAM,MAAM4F,QAAQ7F,EAAYC,KAAa,CAChE,CCFe,SAAS6F,EAAmB7F,GAEzC,QAASS,EAAUT,GAAWA,EAAQO,cACtCP,EAAQ8F,WAAazF,OAAOyF,UAAUC,eACxC,CCFe,SAASC,EAAchG,GACpC,MAA6B,SAAzBD,EAAYC,GACPA,EAMPA,EAAQiG,cACRjG,EAAQwF,aACR3E,EAAab,GAAWA,EAAQyF,KAAO,OAEvCI,EAAmB7F,EAGvB,CCVA,SAASkG,EAAoBlG,GAC3B,OAAKW,EAAcX,IACoB,UAAvC,EAAiBA,GAASiC,SAInBjC,EAAQmG,aAHN,IAIX,CAwCe,SAASC,EAAgBpG,GAItC,IAHA,IAAIK,EAASF,EAAUH,GACnBmG,EAAeD,EAAoBlG,GAEhCmG,GAAgBR,EAAeQ,IAA6D,WAA5C,EAAiBA,GAAclE,UACpFkE,EAAeD,EAAoBC,GAGrC,OAAIA,IAA+C,SAA9BpG,EAAYoG,IAA0D,SAA9BpG,EAAYoG,IAAwE,WAA5C,EAAiBA,GAAclE,UAC3H5B,EAGF8F,GAhDT,SAA4BnG,GAC1B,IAAIqG,EAAY,WAAWvC,KAAKd,KAGhC,GAFW,WAAWc,KAAKd,MAEfrC,EAAcX,IAII,UAFX,EAAiBA,GAEnBiC,SACb,OAAO,KAIX,IAAIqE,EAAcN,EAAchG,GAMhC,IAJIa,EAAayF,KACfA,EAAcA,EAAYb,MAGrB9E,EAAc2F,IAAgB,CAAC,OAAQ,QAAQV,QAAQ7F,EAAYuG,IAAgB,GAAG,CAC3F,IAAIC,EAAM,EAAiBD,GAI3B,GAAsB,SAAlBC,EAAIC,WAA4C,SAApBD,EAAIE,aAA0C,UAAhBF,EAAIG,UAAiF,IAA1D,CAAC,YAAa,eAAed,QAAQW,EAAII,aAAsBN,GAAgC,WAAnBE,EAAII,YAA2BN,GAAaE,EAAIK,QAAyB,SAAfL,EAAIK,OACjO,OAAON,EAEPA,EAAcA,EAAYd,UAE9B,CAEA,OAAO,IACT,CAgByBqB,CAAmB7G,IAAYK,CACxD,CCpEe,SAASyG,EAAyB3H,GAC/C,MAAO,CAAC,MAAO,UAAUyG,QAAQzG,IAAc,EAAI,IAAM,GAC3D,CCDO,SAAS4H,EAAOjE,EAAK1E,EAAOyE,GACjC,OAAO,EAAQC,EAAK,EAAQ1E,EAAOyE,GACrC,CCFe,SAASmE,EAAmBC,GACzC,OAAOxJ,OAAOkE,OAAO,CAAC,ECDf,CACLS,IAAK,EACL9D,MAAO,EACPD,OAAQ,EACRE,KAAM,GDHuC0I,EACjD,CEHe,SAASC,EAAgB9I,EAAOiD,GAC7C,OAAOA,EAAKpC,QAAO,SAAUkI,EAAS5J,GAEpC,OADA4J,EAAQ5J,GAAOa,EACR+I,CACT,GAAG,CAAC,EACN,CC4EA,SACEpG,KAAM,QACNC,SAAS,EACTC,MAAO,OACPC,GApEF,SAAeC,GACb,IAAIiG,EAEAhG,EAAQD,EAAKC,MACbL,EAAOI,EAAKJ,KACZmB,EAAUf,EAAKe,QACfmF,EAAejG,EAAME,SAASgB,MAC9BgF,EAAgBlG,EAAMmG,cAAcD,cACpCE,EAAgB9E,EAAiBtB,EAAMjC,WACvCsI,EAAOX,EAAyBU,GAEhCE,EADa,CAACnJ,EAAMD,GAAOsH,QAAQ4B,IAAkB,EAClC,SAAW,QAElC,GAAKH,GAAiBC,EAAtB,CAIA,IAAIL,EAxBgB,SAAyBU,EAASvG,GAItD,OAAO4F,EAAsC,iBAH7CW,EAA6B,mBAAZA,EAAyBA,EAAQlK,OAAOkE,OAAO,CAAC,EAAGP,EAAMwG,MAAO,CAC/EzI,UAAWiC,EAAMjC,aACbwI,GACkDA,EAAUT,EAAgBS,EAASlJ,GAC7F,CAmBsBoJ,CAAgB3F,EAAQyF,QAASvG,GACjD0G,EAAY/C,EAAcsC,GAC1BU,EAAmB,MAATN,EAAe,EAAMlJ,EAC/ByJ,EAAmB,MAATP,EAAepJ,EAASC,EAClC2J,EAAU7G,EAAMwG,MAAM7I,UAAU2I,GAAOtG,EAAMwG,MAAM7I,UAAU0I,GAAQH,EAAcG,GAAQrG,EAAMwG,MAAM9I,OAAO4I,GAC9GQ,EAAYZ,EAAcG,GAAQrG,EAAMwG,MAAM7I,UAAU0I,GACxDU,EAAoB/B,EAAgBiB,GACpCe,EAAaD,EAA6B,MAATV,EAAeU,EAAkBE,cAAgB,EAAIF,EAAkBG,aAAe,EAAI,EAC3HC,EAAoBN,EAAU,EAAIC,EAAY,EAG9CpF,EAAMmE,EAAcc,GACpBlF,EAAMuF,EAAaN,EAAUJ,GAAOT,EAAce,GAClDQ,EAASJ,EAAa,EAAIN,EAAUJ,GAAO,EAAIa,EAC/CE,EAAS1B,EAAOjE,EAAK0F,EAAQ3F,GAE7B6F,EAAWjB,EACfrG,EAAMmG,cAAcxG,KAASqG,EAAwB,CAAC,GAAyBsB,GAAYD,EAAQrB,EAAsBuB,aAAeF,EAASD,EAAQpB,EAnBzJ,CAoBF,EAkCEtF,OAhCF,SAAgBC,GACd,IAAIX,EAAQW,EAAMX,MAEdwH,EADU7G,EAAMG,QACWlC,QAC3BqH,OAAoC,IAArBuB,EAA8B,sBAAwBA,EAErD,MAAhBvB,IAKwB,iBAAjBA,IACTA,EAAejG,EAAME,SAASxC,OAAO+J,cAAcxB,MAOhDpC,EAAS7D,EAAME,SAASxC,OAAQuI,KAIrCjG,EAAME,SAASgB,MAAQ+E,EACzB,EASE5E,SAAU,CAAC,iBACXqG,iBAAkB,CAAC,oBCxFN,SAASC,EAAa5J,GACnC,OAAOA,EAAUwD,MAAM,KAAK,EAC9B,CCOA,IAAIqG,GAAa,CACf5G,IAAK,OACL9D,MAAO,OACPD,OAAQ,OACRE,KAAM,QAeD,SAAS0K,GAAYlH,GAC1B,IAAImH,EAEApK,EAASiD,EAAMjD,OACfqK,EAAapH,EAAMoH,WACnBhK,EAAY4C,EAAM5C,UAClBiK,EAAYrH,EAAMqH,UAClBC,EAAUtH,EAAMsH,QAChBpH,EAAWF,EAAME,SACjBqH,EAAkBvH,EAAMuH,gBACxBC,EAAWxH,EAAMwH,SACjBC,EAAezH,EAAMyH,aACrBC,EAAU1H,EAAM0H,QAChBC,EAAaL,EAAQ1E,EACrBA,OAAmB,IAAf+E,EAAwB,EAAIA,EAChCC,EAAaN,EAAQxE,EACrBA,OAAmB,IAAf8E,EAAwB,EAAIA,EAEhCC,EAAgC,mBAAjBJ,EAA8BA,EAAa,CAC5D7E,EAAGA,EACHE,IACG,CACHF,EAAGA,EACHE,GAGFF,EAAIiF,EAAMjF,EACVE,EAAI+E,EAAM/E,EACV,IAAIgF,EAAOR,EAAQrL,eAAe,KAC9B8L,EAAOT,EAAQrL,eAAe,KAC9B+L,EAAQxL,EACRyL,EAAQ,EACRC,EAAM5J,OAEV,GAAIkJ,EAAU,CACZ,IAAIpD,EAAeC,EAAgBtH,GAC/BoL,EAAa,eACbC,EAAY,cAEZhE,IAAiBhG,EAAUrB,IAGmB,WAA5C,EAFJqH,EAAeN,EAAmB/G,IAECmD,UAAsC,aAAbA,IAC1DiI,EAAa,eACbC,EAAY,gBAOZhL,IAAc,IAAQA,IAAcZ,GAAQY,IAAcb,IAAU8K,IAAczK,KACpFqL,EAAQ3L,EAGRwG,IAFc4E,GAAWtD,IAAiB8D,GAAOA,EAAIxF,eAAiBwF,EAAIxF,eAAeD,OACzF2B,EAAa+D,IACEf,EAAW3E,OAC1BK,GAAKyE,EAAkB,GAAK,GAG1BnK,IAAcZ,IAASY,IAAc,GAAOA,IAAcd,GAAW+K,IAAczK,KACrFoL,EAAQzL,EAGRqG,IAFc8E,GAAWtD,IAAiB8D,GAAOA,EAAIxF,eAAiBwF,EAAIxF,eAAeH,MACzF6B,EAAagE,IACEhB,EAAW7E,MAC1BK,GAAK2E,EAAkB,GAAK,EAEhC,CAEA,IAgBMc,EAhBFC,EAAe5M,OAAOkE,OAAO,CAC/BM,SAAUA,GACTsH,GAAYP,IAEXsB,GAAyB,IAAjBd,EAlFd,SAA2BrI,EAAM8I,GAC/B,IAAItF,EAAIxD,EAAKwD,EACTE,EAAI1D,EAAK0D,EACT0F,EAAMN,EAAIO,kBAAoB,EAClC,MAAO,CACL7F,EAAG5B,EAAM4B,EAAI4F,GAAOA,GAAO,EAC3B1F,EAAG9B,EAAM8B,EAAI0F,GAAOA,GAAO,EAE/B,CA0EsCE,CAAkB,CACpD9F,EAAGA,EACHE,GACC1E,EAAUrB,IAAW,CACtB6F,EAAGA,EACHE,GAMF,OAHAF,EAAI2F,EAAM3F,EACVE,EAAIyF,EAAMzF,EAENyE,EAGK7L,OAAOkE,OAAO,CAAC,EAAG0I,IAAeD,EAAiB,CAAC,GAAkBJ,GAASF,EAAO,IAAM,GAAIM,EAAeL,GAASF,EAAO,IAAM,GAAIO,EAAe5D,WAAayD,EAAIO,kBAAoB,IAAM,EAAI,aAAe7F,EAAI,OAASE,EAAI,MAAQ,eAAiBF,EAAI,OAASE,EAAI,SAAUuF,IAG5R3M,OAAOkE,OAAO,CAAC,EAAG0I,IAAenB,EAAkB,CAAC,GAAmBc,GAASF,EAAOjF,EAAI,KAAO,GAAIqE,EAAgBa,GAASF,EAAOlF,EAAI,KAAO,GAAIuE,EAAgB1C,UAAY,GAAI0C,GAC9L,CA4CA,UACEnI,KAAM,gBACNC,SAAS,EACTC,MAAO,cACPC,GA9CF,SAAuBwJ,GACrB,IAAItJ,EAAQsJ,EAAMtJ,MACdc,EAAUwI,EAAMxI,QAChByI,EAAwBzI,EAAQoH,gBAChCA,OAA4C,IAA1BqB,GAA0CA,EAC5DC,EAAoB1I,EAAQqH,SAC5BA,OAAiC,IAAtBqB,GAAsCA,EACjDC,EAAwB3I,EAAQsH,aAChCA,OAAyC,IAA1BqB,GAA0CA,EACzDR,EAAe,CACjBlL,UAAWuD,EAAiBtB,EAAMjC,WAClCiK,UAAWL,EAAa3H,EAAMjC,WAC9BL,OAAQsC,EAAME,SAASxC,OACvBqK,WAAY/H,EAAMwG,MAAM9I,OACxBwK,gBAAiBA,EACjBG,QAAoC,UAA3BrI,EAAMc,QAAQC,UAGgB,MAArCf,EAAMmG,cAAcD,gBACtBlG,EAAMK,OAAO3C,OAASrB,OAAOkE,OAAO,CAAC,EAAGP,EAAMK,OAAO3C,OAAQmK,GAAYxL,OAAOkE,OAAO,CAAC,EAAG0I,EAAc,CACvGhB,QAASjI,EAAMmG,cAAcD,cAC7BrF,SAAUb,EAAMc,QAAQC,SACxBoH,SAAUA,EACVC,aAAcA,OAIe,MAA7BpI,EAAMmG,cAAcjF,QACtBlB,EAAMK,OAAOa,MAAQ7E,OAAOkE,OAAO,CAAC,EAAGP,EAAMK,OAAOa,MAAO2G,GAAYxL,OAAOkE,OAAO,CAAC,EAAG0I,EAAc,CACrGhB,QAASjI,EAAMmG,cAAcjF,MAC7BL,SAAU,WACVsH,UAAU,EACVC,aAAcA,OAIlBpI,EAAMM,WAAW5C,OAASrB,OAAOkE,OAAO,CAAC,EAAGP,EAAMM,WAAW5C,OAAQ,CACnE,wBAAyBsC,EAAMjC,WAEnC,EAQE2L,KAAM,CAAC,GCrKT,IAAIC,GAAU,CACZA,SAAS,GAsCX,UACEhK,KAAM,iBACNC,SAAS,EACTC,MAAO,QACPC,GAAI,WAAe,EACnBY,OAxCF,SAAgBX,GACd,IAAIC,EAAQD,EAAKC,MACb4J,EAAW7J,EAAK6J,SAChB9I,EAAUf,EAAKe,QACf+I,EAAkB/I,EAAQgJ,OAC1BA,OAA6B,IAApBD,GAAoCA,EAC7CE,EAAkBjJ,EAAQkJ,OAC1BA,OAA6B,IAApBD,GAAoCA,EAC7C9K,EAASF,EAAUiB,EAAME,SAASxC,QAClCuM,EAAgB,GAAGjM,OAAOgC,EAAMiK,cAActM,UAAWqC,EAAMiK,cAAcvM,QAYjF,OAVIoM,GACFG,EAAc9J,SAAQ,SAAU+J,GAC9BA,EAAaC,iBAAiB,SAAUP,EAASQ,OAAQT,GAC3D,IAGEK,GACF/K,EAAOkL,iBAAiB,SAAUP,EAASQ,OAAQT,IAG9C,WACDG,GACFG,EAAc9J,SAAQ,SAAU+J,GAC9BA,EAAaG,oBAAoB,SAAUT,EAASQ,OAAQT,GAC9D,IAGEK,GACF/K,EAAOoL,oBAAoB,SAAUT,EAASQ,OAAQT,GAE1D,CACF,EASED,KAAM,CAAC,GC/CT,IAAIY,GAAO,CACTnN,KAAM,QACND,MAAO,OACPD,OAAQ,MACR+D,IAAK,UAEQ,SAASuJ,GAAqBxM,GAC3C,OAAOA,EAAUyM,QAAQ,0BAA0B,SAAUC,GAC3D,OAAOH,GAAKG,EACd,GACF,CCVA,IAAI,GAAO,CACTnN,MAAO,MACPC,IAAK,SAEQ,SAASmN,GAA8B3M,GACpD,OAAOA,EAAUyM,QAAQ,cAAc,SAAUC,GAC/C,OAAO,GAAKA,EACd,GACF,CCPe,SAASE,GAAgB3L,GACtC,IAAI6J,EAAM9J,EAAUC,GAGpB,MAAO,CACL4L,WAHe/B,EAAIgC,YAInBC,UAHcjC,EAAIkC,YAKtB,CCNe,SAASC,GAAoBpM,GAQ1C,OAAO+D,EAAsB8B,EAAmB7F,IAAUzB,KAAOwN,GAAgB/L,GAASgM,UAC5F,CCXe,SAASK,GAAerM,GAErC,IAAIsM,EAAoB,EAAiBtM,GACrCuM,EAAWD,EAAkBC,SAC7BC,EAAYF,EAAkBE,UAC9BC,EAAYH,EAAkBG,UAElC,MAAO,6BAA6B3I,KAAKyI,EAAWE,EAAYD,EAClE,CCLe,SAASE,GAAgBtM,GACtC,MAAI,CAAC,OAAQ,OAAQ,aAAawF,QAAQ7F,EAAYK,KAAU,EAEvDA,EAAKG,cAAcoM,KAGxBhM,EAAcP,IAASiM,GAAejM,GACjCA,EAGFsM,GAAgB1G,EAAc5F,GACvC,CCJe,SAASwM,GAAkB5M,EAAS6M,GACjD,IAAIC,OAES,IAATD,IACFA,EAAO,IAGT,IAAIvB,EAAeoB,GAAgB1M,GAC/B+M,EAASzB,KAAqE,OAAlDwB,EAAwB9M,EAAQO,oBAAyB,EAASuM,EAAsBH,MACpH1C,EAAM9J,EAAUmL,GAChB0B,EAASD,EAAS,CAAC9C,GAAK7K,OAAO6K,EAAIxF,gBAAkB,GAAI4H,GAAef,GAAgBA,EAAe,IAAMA,EAC7G2B,EAAcJ,EAAKzN,OAAO4N,GAC9B,OAAOD,EAASE,EAChBA,EAAY7N,OAAOwN,GAAkB5G,EAAcgH,IACrD,CCzBe,SAASE,GAAiBC,GACvC,OAAO1P,OAAOkE,OAAO,CAAC,EAAGwL,EAAM,CAC7B5O,KAAM4O,EAAKxI,EACXvC,IAAK+K,EAAKtI,EACVvG,MAAO6O,EAAKxI,EAAIwI,EAAK7I,MACrBjG,OAAQ8O,EAAKtI,EAAIsI,EAAK3I,QAE1B,CCqBA,SAAS4I,GAA2BpN,EAASqN,EAAgBlL,GAC3D,OAAOkL,IAAmBxO,EAAWqO,GCzBxB,SAAyBlN,EAASmC,GAC/C,IAAI8H,EAAM9J,EAAUH,GAChBsN,EAAOzH,EAAmB7F,GAC1ByE,EAAiBwF,EAAIxF,eACrBH,EAAQgJ,EAAKhF,YACb9D,EAAS8I,EAAKjF,aACd1D,EAAI,EACJE,EAAI,EAER,GAAIJ,EAAgB,CAClBH,EAAQG,EAAeH,MACvBE,EAASC,EAAeD,OACxB,IAAI+I,EAAiB1J,KAEjB0J,IAAmBA,GAA+B,UAAbpL,KACvCwC,EAAIF,EAAeG,WACnBC,EAAIJ,EAAeK,UAEvB,CAEA,MAAO,CACLR,MAAOA,EACPE,OAAQA,EACRG,EAAGA,EAAIyH,GAAoBpM,GAC3B6E,EAAGA,EAEP,CDDwD2I,CAAgBxN,EAASmC,IAAa1B,EAAU4M,GAdxG,SAAoCrN,EAASmC,GAC3C,IAAIgL,EAAOpJ,EAAsB/D,GAAS,EAAoB,UAAbmC,GASjD,OARAgL,EAAK/K,IAAM+K,EAAK/K,IAAMpC,EAAQyN,UAC9BN,EAAK5O,KAAO4O,EAAK5O,KAAOyB,EAAQ0N,WAChCP,EAAK9O,OAAS8O,EAAK/K,IAAMpC,EAAQqI,aACjC8E,EAAK7O,MAAQ6O,EAAK5O,KAAOyB,EAAQsI,YACjC6E,EAAK7I,MAAQtE,EAAQsI,YACrB6E,EAAK3I,OAASxE,EAAQqI,aACtB8E,EAAKxI,EAAIwI,EAAK5O,KACd4O,EAAKtI,EAAIsI,EAAK/K,IACP+K,CACT,CAG0HQ,CAA2BN,EAAgBlL,GAAY+K,GEtBlK,SAAyBlN,GACtC,IAAI8M,EAEAQ,EAAOzH,EAAmB7F,GAC1B4N,EAAY7B,GAAgB/L,GAC5B2M,EAA0D,OAAlDG,EAAwB9M,EAAQO,oBAAyB,EAASuM,EAAsBH,KAChGrI,EAAQ,EAAIgJ,EAAKO,YAAaP,EAAKhF,YAAaqE,EAAOA,EAAKkB,YAAc,EAAGlB,EAAOA,EAAKrE,YAAc,GACvG9D,EAAS,EAAI8I,EAAKQ,aAAcR,EAAKjF,aAAcsE,EAAOA,EAAKmB,aAAe,EAAGnB,EAAOA,EAAKtE,aAAe,GAC5G1D,GAAKiJ,EAAU5B,WAAaI,GAAoBpM,GAChD6E,GAAK+I,EAAU1B,UAMnB,MAJiD,QAA7C,EAAiBS,GAAQW,GAAMS,YACjCpJ,GAAK,EAAI2I,EAAKhF,YAAaqE,EAAOA,EAAKrE,YAAc,GAAKhE,GAGrD,CACLA,MAAOA,EACPE,OAAQA,EACRG,EAAGA,EACHE,EAAGA,EAEP,CFCkMmJ,CAAgBnI,EAAmB7F,IACrO,CG1Be,SAASiO,GAAe9M,GACrC,IAOIkI,EAPAtK,EAAYoC,EAAKpC,UACjBiB,EAAUmB,EAAKnB,QACfb,EAAYgC,EAAKhC,UACjBqI,EAAgBrI,EAAYuD,EAAiBvD,GAAa,KAC1DiK,EAAYjK,EAAY4J,EAAa5J,GAAa,KAClD+O,EAAUnP,EAAU4F,EAAI5F,EAAUuF,MAAQ,EAAItE,EAAQsE,MAAQ,EAC9D6J,EAAUpP,EAAU8F,EAAI9F,EAAUyF,OAAS,EAAIxE,EAAQwE,OAAS,EAGpE,OAAQgD,GACN,KAAK,EACH6B,EAAU,CACR1E,EAAGuJ,EACHrJ,EAAG9F,EAAU8F,EAAI7E,EAAQwE,QAE3B,MAEF,KAAKnG,EACHgL,EAAU,CACR1E,EAAGuJ,EACHrJ,EAAG9F,EAAU8F,EAAI9F,EAAUyF,QAE7B,MAEF,KAAKlG,EACH+K,EAAU,CACR1E,EAAG5F,EAAU4F,EAAI5F,EAAUuF,MAC3BO,EAAGsJ,GAEL,MAEF,KAAK5P,EACH8K,EAAU,CACR1E,EAAG5F,EAAU4F,EAAI3E,EAAQsE,MACzBO,EAAGsJ,GAEL,MAEF,QACE9E,EAAU,CACR1E,EAAG5F,EAAU4F,EACbE,EAAG9F,EAAU8F,GAInB,IAAIuJ,EAAW5G,EAAgBV,EAAyBU,GAAiB,KAEzE,GAAgB,MAAZ4G,EAAkB,CACpB,IAAI1G,EAAmB,MAAb0G,EAAmB,SAAW,QAExC,OAAQhF,GACN,KAAK1K,EACH2K,EAAQ+E,GAAY/E,EAAQ+E,IAAarP,EAAU2I,GAAO,EAAI1H,EAAQ0H,GAAO,GAC7E,MAEF,KAAK/I,EACH0K,EAAQ+E,GAAY/E,EAAQ+E,IAAarP,EAAU2I,GAAO,EAAI1H,EAAQ0H,GAAO,GAKnF,CAEA,OAAO2B,CACT,CC3De,SAASgF,GAAejN,EAAOc,QAC5B,IAAZA,IACFA,EAAU,CAAC,GAGb,IAAIoM,EAAWpM,EACXqM,EAAqBD,EAASnP,UAC9BA,OAAmC,IAAvBoP,EAAgCnN,EAAMjC,UAAYoP,EAC9DC,EAAoBF,EAASnM,SAC7BA,OAAiC,IAAtBqM,EAA+BpN,EAAMe,SAAWqM,EAC3DC,EAAoBH,EAASI,SAC7BA,OAAiC,IAAtBD,EAA+B7P,EAAkB6P,EAC5DE,EAAwBL,EAASM,aACjCA,OAAyC,IAA1BD,EAAmC9P,EAAW8P,EAC7DE,EAAwBP,EAASQ,eACjCA,OAA2C,IAA1BD,EAAmC/P,EAAS+P,EAC7DE,EAAuBT,EAASU,YAChCA,OAAuC,IAAzBD,GAA0CA,EACxDE,EAAmBX,EAAS3G,QAC5BA,OAA+B,IAArBsH,EAA8B,EAAIA,EAC5ChI,EAAgBD,EAAsC,iBAAZW,EAAuBA,EAAUT,EAAgBS,EAASlJ,IACpGyQ,EAAaJ,IAAmBhQ,EAASC,EAAYD,EACrDqK,EAAa/H,EAAMwG,MAAM9I,OACzBkB,EAAUoB,EAAME,SAAS0N,EAAcE,EAAaJ,GACpDK,EJkBS,SAAyBnP,EAAS0O,EAAUE,EAAczM,GACvE,IAAIiN,EAAmC,oBAAbV,EAlB5B,SAA4B1O,GAC1B,IAAIpB,EAAkBgO,GAAkB5G,EAAchG,IAElDqP,EADoB,CAAC,WAAY,SAASzJ,QAAQ,EAAiB5F,GAASiC,WAAa,GACnDtB,EAAcX,GAAWoG,EAAgBpG,GAAWA,EAE9F,OAAKS,EAAU4O,GAKRzQ,EAAgBgI,QAAO,SAAUyG,GACtC,OAAO5M,EAAU4M,IAAmBpI,EAASoI,EAAgBgC,IAAmD,SAAhCtP,EAAYsN,EAC9F,IANS,EAOX,CAK6DiC,CAAmBtP,GAAW,GAAGZ,OAAOsP,GAC/F9P,EAAkB,GAAGQ,OAAOgQ,EAAqB,CAACR,IAClDW,EAAsB3Q,EAAgB,GACtC4Q,EAAe5Q,EAAgBK,QAAO,SAAUwQ,EAASpC,GAC3D,IAAIF,EAAOC,GAA2BpN,EAASqN,EAAgBlL,GAK/D,OAJAsN,EAAQrN,IAAM,EAAI+K,EAAK/K,IAAKqN,EAAQrN,KACpCqN,EAAQnR,MAAQ,EAAI6O,EAAK7O,MAAOmR,EAAQnR,OACxCmR,EAAQpR,OAAS,EAAI8O,EAAK9O,OAAQoR,EAAQpR,QAC1CoR,EAAQlR,KAAO,EAAI4O,EAAK5O,KAAMkR,EAAQlR,MAC/BkR,CACT,GAAGrC,GAA2BpN,EAASuP,EAAqBpN,IAK5D,OAJAqN,EAAalL,MAAQkL,EAAalR,MAAQkR,EAAajR,KACvDiR,EAAahL,OAASgL,EAAanR,OAASmR,EAAapN,IACzDoN,EAAa7K,EAAI6K,EAAajR,KAC9BiR,EAAa3K,EAAI2K,EAAapN,IACvBoN,CACT,CInC2BE,CAAgBjP,EAAUT,GAAWA,EAAUA,EAAQ2P,gBAAkB9J,EAAmBzE,EAAME,SAASxC,QAAS4P,EAAUE,EAAczM,GACjKyN,EAAsB7L,EAAsB3C,EAAME,SAASvC,WAC3DuI,EAAgB2G,GAAe,CACjClP,UAAW6Q,EACX5P,QAASmJ,EACThH,SAAU,WACVhD,UAAWA,IAET0Q,EAAmB3C,GAAiBzP,OAAOkE,OAAO,CAAC,EAAGwH,EAAY7B,IAClEwI,EAAoBhB,IAAmBhQ,EAAS+Q,EAAmBD,EAGnEG,EAAkB,CACpB3N,IAAK+M,EAAmB/M,IAAM0N,EAAkB1N,IAAM6E,EAAc7E,IACpE/D,OAAQyR,EAAkBzR,OAAS8Q,EAAmB9Q,OAAS4I,EAAc5I,OAC7EE,KAAM4Q,EAAmB5Q,KAAOuR,EAAkBvR,KAAO0I,EAAc1I,KACvED,MAAOwR,EAAkBxR,MAAQ6Q,EAAmB7Q,MAAQ2I,EAAc3I,OAExE0R,EAAa5O,EAAMmG,cAAckB,OAErC,GAAIqG,IAAmBhQ,GAAUkR,EAAY,CAC3C,IAAIvH,EAASuH,EAAW7Q,GACxB1B,OAAO4D,KAAK0O,GAAiBxO,SAAQ,SAAUhE,GAC7C,IAAI0S,EAAW,CAAC3R,EAAOD,GAAQuH,QAAQrI,IAAQ,EAAI,GAAK,EACpDkK,EAAO,CAAC,EAAKpJ,GAAQuH,QAAQrI,IAAQ,EAAI,IAAM,IACnDwS,EAAgBxS,IAAQkL,EAAOhB,GAAQwI,CACzC,GACF,CAEA,OAAOF,CACT,CCyEA,UACEhP,KAAM,OACNC,SAAS,EACTC,MAAO,OACPC,GA5HF,SAAcC,GACZ,IAAIC,EAAQD,EAAKC,MACbc,EAAUf,EAAKe,QACfnB,EAAOI,EAAKJ,KAEhB,IAAIK,EAAMmG,cAAcxG,GAAMmP,MAA9B,CAoCA,IAhCA,IAAIC,EAAoBjO,EAAQkM,SAC5BgC,OAAsC,IAAtBD,GAAsCA,EACtDE,EAAmBnO,EAAQoO,QAC3BC,OAAoC,IAArBF,GAAqCA,EACpDG,EAA8BtO,EAAQuO,mBACtC9I,EAAUzF,EAAQyF,QAClB+G,EAAWxM,EAAQwM,SACnBE,EAAe1M,EAAQ0M,aACvBI,EAAc9M,EAAQ8M,YACtB0B,EAAwBxO,EAAQyO,eAChCA,OAA2C,IAA1BD,GAA0CA,EAC3DE,EAAwB1O,EAAQ0O,sBAChCC,EAAqBzP,EAAMc,QAAQ/C,UACnCqI,EAAgB9E,EAAiBmO,GAEjCJ,EAAqBD,IADHhJ,IAAkBqJ,GACqCF,EAjC/E,SAAuCxR,GACrC,GAAIuD,EAAiBvD,KAAeX,EAClC,MAAO,GAGT,IAAIsS,EAAoBnF,GAAqBxM,GAC7C,MAAO,CAAC2M,GAA8B3M,GAAY2R,EAAmBhF,GAA8BgF,GACrG,CA0B6IC,CAA8BF,GAA3E,CAAClF,GAAqBkF,KAChHG,EAAa,CAACH,GAAoBzR,OAAOqR,GAAoBxR,QAAO,SAAUC,EAAKC,GACrF,OAAOD,EAAIE,OAAOsD,EAAiBvD,KAAeX,ECvCvC,SAA8B4C,EAAOc,QAClC,IAAZA,IACFA,EAAU,CAAC,GAGb,IAAIoM,EAAWpM,EACX/C,EAAYmP,EAASnP,UACrBuP,EAAWJ,EAASI,SACpBE,EAAeN,EAASM,aACxBjH,EAAU2G,EAAS3G,QACnBgJ,EAAiBrC,EAASqC,eAC1BM,EAAwB3C,EAASsC,sBACjCA,OAAkD,IAA1BK,EAAmC,EAAgBA,EAC3E7H,EAAYL,EAAa5J,GACzB6R,EAAa5H,EAAYuH,EAAiB3R,EAAsBA,EAAoB4H,QAAO,SAAUzH,GACvG,OAAO4J,EAAa5J,KAAeiK,CACrC,IAAK3K,EACDyS,EAAoBF,EAAWpK,QAAO,SAAUzH,GAClD,OAAOyR,EAAsBhL,QAAQzG,IAAc,CACrD,IAEiC,IAA7B+R,EAAkBC,SACpBD,EAAoBF,GAItB,IAAII,EAAYF,EAAkBjS,QAAO,SAAUC,EAAKC,GAOtD,OANAD,EAAIC,GAAakP,GAAejN,EAAO,CACrCjC,UAAWA,EACXuP,SAAUA,EACVE,aAAcA,EACdjH,QAASA,IACRjF,EAAiBvD,IACbD,CACT,GAAG,CAAC,GACJ,OAAOzB,OAAO4D,KAAK+P,GAAWC,MAAK,SAAUC,EAAGC,GAC9C,OAAOH,EAAUE,GAAKF,EAAUG,EAClC,GACF,CDC6DC,CAAqBpQ,EAAO,CACnFjC,UAAWA,EACXuP,SAAUA,EACVE,aAAcA,EACdjH,QAASA,EACTgJ,eAAgBA,EAChBC,sBAAuBA,IACpBzR,EACP,GAAG,IACCsS,EAAgBrQ,EAAMwG,MAAM7I,UAC5BoK,EAAa/H,EAAMwG,MAAM9I,OACzB4S,EAAY,IAAIC,IAChBC,GAAqB,EACrBC,EAAwBb,EAAW,GAE9Bc,EAAI,EAAGA,EAAId,EAAWG,OAAQW,IAAK,CAC1C,IAAI3S,EAAY6R,EAAWc,GAEvBC,EAAiBrP,EAAiBvD,GAElC6S,EAAmBjJ,EAAa5J,KAAeT,EAC/CuT,EAAa,CAAC,EAAK5T,GAAQuH,QAAQmM,IAAmB,EACtDrK,EAAMuK,EAAa,QAAU,SAC7B1F,EAAW8B,GAAejN,EAAO,CACnCjC,UAAWA,EACXuP,SAAUA,EACVE,aAAcA,EACdI,YAAaA,EACbrH,QAASA,IAEPuK,EAAoBD,EAAaD,EAAmB1T,EAAQC,EAAOyT,EAAmB3T,EAAS,EAE/FoT,EAAc/J,GAAOyB,EAAWzB,KAClCwK,EAAoBvG,GAAqBuG,IAG3C,IAAIC,EAAmBxG,GAAqBuG,GACxCE,EAAS,GAUb,GARIhC,GACFgC,EAAOC,KAAK9F,EAASwF,IAAmB,GAGtCxB,GACF6B,EAAOC,KAAK9F,EAAS2F,IAAsB,EAAG3F,EAAS4F,IAAqB,GAG1EC,EAAOE,OAAM,SAAUC,GACzB,OAAOA,CACT,IAAI,CACFV,EAAwB1S,EACxByS,GAAqB,EACrB,KACF,CAEAF,EAAUc,IAAIrT,EAAWiT,EAC3B,CAEA,GAAIR,EAqBF,IAnBA,IAEIa,EAAQ,SAAeC,GACzB,IAAIC,EAAmB3B,EAAW4B,MAAK,SAAUzT,GAC/C,IAAIiT,EAASV,EAAU9T,IAAIuB,GAE3B,GAAIiT,EACF,OAAOA,EAAOS,MAAM,EAAGH,GAAIJ,OAAM,SAAUC,GACzC,OAAOA,CACT,GAEJ,IAEA,GAAII,EAEF,OADAd,EAAwBc,EACjB,OAEX,EAESD,EAnBY/B,EAAiB,EAAI,EAmBZ+B,EAAK,GAGpB,UAFFD,EAAMC,GADmBA,KAOpCtR,EAAMjC,YAAc0S,IACtBzQ,EAAMmG,cAAcxG,GAAMmP,OAAQ,EAClC9O,EAAMjC,UAAY0S,EAClBzQ,EAAM0R,OAAQ,EA5GhB,CA8GF,EAQEhK,iBAAkB,CAAC,UACnBgC,KAAM,CACJoF,OAAO,IE7IX,SAAS6C,GAAexG,EAAUY,EAAM6F,GAQtC,YAPyB,IAArBA,IACFA,EAAmB,CACjBrO,EAAG,EACHE,EAAG,IAIA,CACLzC,IAAKmK,EAASnK,IAAM+K,EAAK3I,OAASwO,EAAiBnO,EACnDvG,MAAOiO,EAASjO,MAAQ6O,EAAK7I,MAAQ0O,EAAiBrO,EACtDtG,OAAQkO,EAASlO,OAAS8O,EAAK3I,OAASwO,EAAiBnO,EACzDtG,KAAMgO,EAAShO,KAAO4O,EAAK7I,MAAQ0O,EAAiBrO,EAExD,CAEA,SAASsO,GAAsB1G,GAC7B,MAAO,CAAC,EAAKjO,EAAOD,EAAQE,GAAM2U,MAAK,SAAUC,GAC/C,OAAO5G,EAAS4G,IAAS,CAC3B,GACF,CA+BA,UACEpS,KAAM,OACNC,SAAS,EACTC,MAAO,OACP6H,iBAAkB,CAAC,mBACnB5H,GAlCF,SAAcC,GACZ,IAAIC,EAAQD,EAAKC,MACbL,EAAOI,EAAKJ,KACZ0Q,EAAgBrQ,EAAMwG,MAAM7I,UAC5BoK,EAAa/H,EAAMwG,MAAM9I,OACzBkU,EAAmB5R,EAAMmG,cAAc6L,gBACvCC,EAAoBhF,GAAejN,EAAO,CAC5C0N,eAAgB,cAEdwE,EAAoBjF,GAAejN,EAAO,CAC5C4N,aAAa,IAEXuE,EAA2BR,GAAeM,EAAmB5B,GAC7D+B,EAAsBT,GAAeO,EAAmBnK,EAAY6J,GACpES,EAAoBR,GAAsBM,GAC1CG,EAAmBT,GAAsBO,GAC7CpS,EAAMmG,cAAcxG,GAAQ,CAC1BwS,yBAA0BA,EAC1BC,oBAAqBA,EACrBC,kBAAmBA,EACnBC,iBAAkBA,GAEpBtS,EAAMM,WAAW5C,OAASrB,OAAOkE,OAAO,CAAC,EAAGP,EAAMM,WAAW5C,OAAQ,CACnE,+BAAgC2U,EAChC,sBAAuBC,GAE3B,GCJA,IACE3S,KAAM,SACNC,SAAS,EACTC,MAAO,OACPwB,SAAU,CAAC,iBACXvB,GA5BF,SAAgBa,GACd,IAAIX,EAAQW,EAAMX,MACdc,EAAUH,EAAMG,QAChBnB,EAAOgB,EAAMhB,KACb4S,EAAkBzR,EAAQuG,OAC1BA,OAA6B,IAApBkL,EAA6B,CAAC,EAAG,GAAKA,EAC/C7I,EAAO,EAAW7L,QAAO,SAAUC,EAAKC,GAE1C,OADAD,EAAIC,GA5BD,SAAiCA,EAAWyI,EAAOa,GACxD,IAAIjB,EAAgB9E,EAAiBvD,GACjCyU,EAAiB,CAACrV,EAAM,GAAKqH,QAAQ4B,IAAkB,GAAK,EAAI,EAEhErG,EAAyB,mBAAXsH,EAAwBA,EAAOhL,OAAOkE,OAAO,CAAC,EAAGiG,EAAO,CACxEzI,UAAWA,KACPsJ,EACFoL,EAAW1S,EAAK,GAChB2S,EAAW3S,EAAK,GAIpB,OAFA0S,EAAWA,GAAY,EACvBC,GAAYA,GAAY,GAAKF,EACtB,CAACrV,EAAMD,GAAOsH,QAAQ4B,IAAkB,EAAI,CACjD7C,EAAGmP,EACHjP,EAAGgP,GACD,CACFlP,EAAGkP,EACHhP,EAAGiP,EAEP,CASqBC,CAAwB5U,EAAWiC,EAAMwG,MAAOa,GAC1DvJ,CACT,GAAG,CAAC,GACA8U,EAAwBlJ,EAAK1J,EAAMjC,WACnCwF,EAAIqP,EAAsBrP,EAC1BE,EAAImP,EAAsBnP,EAEW,MAArCzD,EAAMmG,cAAcD,gBACtBlG,EAAMmG,cAAcD,cAAc3C,GAAKA,EACvCvD,EAAMmG,cAAcD,cAAczC,GAAKA,GAGzCzD,EAAMmG,cAAcxG,GAAQ+J,CAC9B,GC1BA,IACE/J,KAAM,gBACNC,SAAS,EACTC,MAAO,OACPC,GApBF,SAAuBC,GACrB,IAAIC,EAAQD,EAAKC,MACbL,EAAOI,EAAKJ,KAKhBK,EAAMmG,cAAcxG,GAAQkN,GAAe,CACzClP,UAAWqC,EAAMwG,MAAM7I,UACvBiB,QAASoB,EAAMwG,MAAM9I,OACrBqD,SAAU,WACVhD,UAAWiC,EAAMjC,WAErB,EAQE2L,KAAM,CAAC,GCgHT,IACE/J,KAAM,kBACNC,SAAS,EACTC,MAAO,OACPC,GA/HF,SAAyBC,GACvB,IAAIC,EAAQD,EAAKC,MACbc,EAAUf,EAAKe,QACfnB,EAAOI,EAAKJ,KACZoP,EAAoBjO,EAAQkM,SAC5BgC,OAAsC,IAAtBD,GAAsCA,EACtDE,EAAmBnO,EAAQoO,QAC3BC,OAAoC,IAArBF,GAAsCA,EACrD3B,EAAWxM,EAAQwM,SACnBE,EAAe1M,EAAQ0M,aACvBI,EAAc9M,EAAQ8M,YACtBrH,EAAUzF,EAAQyF,QAClBsM,EAAkB/R,EAAQgS,OAC1BA,OAA6B,IAApBD,GAAoCA,EAC7CE,EAAwBjS,EAAQkS,aAChCA,OAAyC,IAA1BD,EAAmC,EAAIA,EACtD5H,EAAW8B,GAAejN,EAAO,CACnCsN,SAAUA,EACVE,aAAcA,EACdjH,QAASA,EACTqH,YAAaA,IAEXxH,EAAgB9E,EAAiBtB,EAAMjC,WACvCiK,EAAYL,EAAa3H,EAAMjC,WAC/BkV,GAAmBjL,EACnBgF,EAAWtH,EAAyBU,GACpC8I,ECrCY,MDqCSlC,ECrCH,IAAM,IDsCxB9G,EAAgBlG,EAAMmG,cAAcD,cACpCmK,EAAgBrQ,EAAMwG,MAAM7I,UAC5BoK,EAAa/H,EAAMwG,MAAM9I,OACzBwV,EAA4C,mBAAjBF,EAA8BA,EAAa3W,OAAOkE,OAAO,CAAC,EAAGP,EAAMwG,MAAO,CACvGzI,UAAWiC,EAAMjC,aACbiV,EACFG,EAA2D,iBAAtBD,EAAiC,CACxElG,SAAUkG,EACVhE,QAASgE,GACP7W,OAAOkE,OAAO,CAChByM,SAAU,EACVkC,QAAS,GACRgE,GACCE,EAAsBpT,EAAMmG,cAAckB,OAASrH,EAAMmG,cAAckB,OAAOrH,EAAMjC,WAAa,KACjG2L,EAAO,CACTnG,EAAG,EACHE,EAAG,GAGL,GAAKyC,EAAL,CAIA,GAAI8I,EAAe,CACjB,IAAIqE,EAEAC,EAAwB,MAAbtG,EAAmB,EAAM7P,EACpCoW,EAAuB,MAAbvG,EAAmB/P,EAASC,EACtCoJ,EAAmB,MAAb0G,EAAmB,SAAW,QACpC3F,EAASnB,EAAc8G,GACvBtL,EAAM2F,EAAS8D,EAASmI,GACxB7R,EAAM4F,EAAS8D,EAASoI,GACxBC,EAAWV,GAAU/K,EAAWzB,GAAO,EAAI,EAC3CmN,EAASzL,IAAc1K,EAAQ+S,EAAc/J,GAAOyB,EAAWzB,GAC/DoN,EAAS1L,IAAc1K,GAASyK,EAAWzB,IAAQ+J,EAAc/J,GAGjEL,EAAejG,EAAME,SAASgB,MAC9BwF,EAAYoM,GAAU7M,EAAetC,EAAcsC,GAAgB,CACrE/C,MAAO,EACPE,OAAQ,GAENuQ,GAAqB3T,EAAMmG,cAAc,oBAAsBnG,EAAMmG,cAAc,oBAAoBI,QxBhFtG,CACLvF,IAAK,EACL9D,MAAO,EACPD,OAAQ,EACRE,KAAM,GwB6EFyW,GAAkBD,GAAmBL,GACrCO,GAAkBF,GAAmBJ,GAMrCO,GAAWnO,EAAO,EAAG0K,EAAc/J,GAAMI,EAAUJ,IACnDyN,GAAYd,EAAkB5C,EAAc/J,GAAO,EAAIkN,EAAWM,GAAWF,GAAkBT,EAA4BnG,SAAWyG,EAASK,GAAWF,GAAkBT,EAA4BnG,SACxMgH,GAAYf,GAAmB5C,EAAc/J,GAAO,EAAIkN,EAAWM,GAAWD,GAAkBV,EAA4BnG,SAAW0G,EAASI,GAAWD,GAAkBV,EAA4BnG,SACzMjG,GAAoB/G,EAAME,SAASgB,OAAS8D,EAAgBhF,EAAME,SAASgB,OAC3E+S,GAAelN,GAAiC,MAAbiG,EAAmBjG,GAAkBsF,WAAa,EAAItF,GAAkBuF,YAAc,EAAI,EAC7H4H,GAAwH,OAAjGb,EAA+C,MAAvBD,OAA8B,EAASA,EAAoBpG,IAAqBqG,EAAwB,EAEvJc,GAAY9M,EAAS2M,GAAYE,GACjCE,GAAkBzO,EAAOmN,EAAS,EAAQpR,EAF9B2F,EAAS0M,GAAYG,GAAsBD,IAEKvS,EAAK2F,EAAQyL,EAAS,EAAQrR,EAAK0S,IAAa1S,GAChHyE,EAAc8G,GAAYoH,GAC1B1K,EAAKsD,GAAYoH,GAAkB/M,CACrC,CAEA,GAAI8H,EAAc,CAChB,IAAIkF,GAEAC,GAAyB,MAAbtH,EAAmB,EAAM7P,EAErCoX,GAAwB,MAAbvH,EAAmB/P,EAASC,EAEvCsX,GAAUtO,EAAcgJ,GAExBuF,GAAmB,MAAZvF,EAAkB,SAAW,QAEpCwF,GAAOF,GAAUrJ,EAASmJ,IAE1BK,GAAOH,GAAUrJ,EAASoJ,IAE1BK,IAAuD,IAAxC,CAAC,EAAKzX,GAAMqH,QAAQ4B,GAEnCyO,GAAyH,OAAjGR,GAAgD,MAAvBjB,OAA8B,EAASA,EAAoBlE,IAAoBmF,GAAyB,EAEzJS,GAAaF,GAAeF,GAAOF,GAAUnE,EAAcoE,IAAQ1M,EAAW0M,IAAQI,GAAuB1B,EAA4BjE,QAEzI6F,GAAaH,GAAeJ,GAAUnE,EAAcoE,IAAQ1M,EAAW0M,IAAQI,GAAuB1B,EAA4BjE,QAAUyF,GAE5IK,GAAmBlC,GAAU8B,G1BzH9B,SAAwBlT,EAAK1E,EAAOyE,GACzC,IAAIwT,EAAItP,EAAOjE,EAAK1E,EAAOyE,GAC3B,OAAOwT,EAAIxT,EAAMA,EAAMwT,CACzB,C0BsHoDC,CAAeJ,GAAYN,GAASO,IAAcpP,EAAOmN,EAASgC,GAAaJ,GAAMF,GAAS1B,EAASiC,GAAaJ,IAEpKzO,EAAcgJ,GAAW8F,GACzBtL,EAAKwF,GAAW8F,GAAmBR,EACrC,CAEAxU,EAAMmG,cAAcxG,GAAQ+J,CAvE5B,CAwEF,EAQEhC,iBAAkB,CAAC,WE1HN,SAASyN,GAAiBC,EAAyBrQ,EAAcsD,QAC9D,IAAZA,IACFA,GAAU,GAGZ,ICnBoCrJ,ECJOJ,EFuBvCyW,EAA0B9V,EAAcwF,GACxCuQ,EAAuB/V,EAAcwF,IAf3C,SAAyBnG,GACvB,IAAImN,EAAOnN,EAAQ+D,wBACfI,EAASpB,EAAMoK,EAAK7I,OAAStE,EAAQqE,aAAe,EACpDD,EAASrB,EAAMoK,EAAK3I,QAAUxE,EAAQuE,cAAgB,EAC1D,OAAkB,IAAXJ,GAA2B,IAAXC,CACzB,CAU4DuS,CAAgBxQ,GACtEJ,EAAkBF,EAAmBM,GACrCgH,EAAOpJ,EAAsByS,EAAyBE,EAAsBjN,GAC5EyB,EAAS,CACXc,WAAY,EACZE,UAAW,GAET7C,EAAU,CACZ1E,EAAG,EACHE,EAAG,GAkBL,OAfI4R,IAA4BA,IAA4BhN,MACxB,SAA9B1J,EAAYoG,IAChBkG,GAAetG,MACbmF,GCnCgC9K,EDmCT+F,KClCdhG,EAAUC,IAAUO,EAAcP,GCJxC,CACL4L,YAFyChM,EDQbI,GCNR4L,WACpBE,UAAWlM,EAAQkM,WDGZH,GAAgB3L,IDoCnBO,EAAcwF,KAChBkD,EAAUtF,EAAsBoC,GAAc,IACtCxB,GAAKwB,EAAauH,WAC1BrE,EAAQxE,GAAKsB,EAAasH,WACjB1H,IACTsD,EAAQ1E,EAAIyH,GAAoBrG,KAI7B,CACLpB,EAAGwI,EAAK5O,KAAO2M,EAAOc,WAAa3C,EAAQ1E,EAC3CE,EAAGsI,EAAK/K,IAAM8I,EAAOgB,UAAY7C,EAAQxE,EACzCP,MAAO6I,EAAK7I,MACZE,OAAQ2I,EAAK3I,OAEjB,CGvDA,SAASoS,GAAMC,GACb,IAAItT,EAAM,IAAIoO,IACVmF,EAAU,IAAIC,IACdC,EAAS,GAKb,SAAS3F,EAAK4F,GACZH,EAAQI,IAAID,EAASlW,MACN,GAAG3B,OAAO6X,EAASxU,UAAY,GAAIwU,EAASnO,kBAAoB,IACtEvH,SAAQ,SAAU4V,GACzB,IAAKL,EAAQM,IAAID,GAAM,CACrB,IAAIE,EAAc9T,EAAI3F,IAAIuZ,GAEtBE,GACFhG,EAAKgG,EAET,CACF,IACAL,EAAO3E,KAAK4E,EACd,CAQA,OAzBAJ,EAAUtV,SAAQ,SAAU0V,GAC1B1T,EAAIiP,IAAIyE,EAASlW,KAAMkW,EACzB,IAiBAJ,EAAUtV,SAAQ,SAAU0V,GACrBH,EAAQM,IAAIH,EAASlW,OAExBsQ,EAAK4F,EAET,IACOD,CACT,CCvBA,IAAIM,GAAkB,CACpBnY,UAAW,SACX0X,UAAW,GACX1U,SAAU,YAGZ,SAASoV,KACP,IAAK,IAAI1B,EAAO2B,UAAUrG,OAAQsG,EAAO,IAAIpU,MAAMwS,GAAO6B,EAAO,EAAGA,EAAO7B,EAAM6B,IAC/ED,EAAKC,GAAQF,UAAUE,GAGzB,OAAQD,EAAKvE,MAAK,SAAUlT,GAC1B,QAASA,GAAoD,mBAAlCA,EAAQ+D,sBACrC,GACF,CAEO,SAAS4T,GAAgBC,QACL,IAArBA,IACFA,EAAmB,CAAC,GAGtB,IAAIC,EAAoBD,EACpBE,EAAwBD,EAAkBE,iBAC1CA,OAA6C,IAA1BD,EAAmC,GAAKA,EAC3DE,EAAyBH,EAAkBI,eAC3CA,OAA4C,IAA3BD,EAAoCV,GAAkBU,EAC3E,OAAO,SAAsBjZ,EAAWD,EAAQoD,QAC9B,IAAZA,IACFA,EAAU+V,GAGZ,ICxC6B/W,EAC3BgX,EDuCE9W,EAAQ,CACVjC,UAAW,SACXgZ,iBAAkB,GAClBjW,QAASzE,OAAOkE,OAAO,CAAC,EAAG2V,GAAiBW,GAC5C1Q,cAAe,CAAC,EAChBjG,SAAU,CACRvC,UAAWA,EACXD,OAAQA,GAEV4C,WAAY,CAAC,EACbD,OAAQ,CAAC,GAEP2W,EAAmB,GACnBC,GAAc,EACdrN,EAAW,CACb5J,MAAOA,EACPkX,WAAY,SAAoBC,GAC9B,IAAIrW,EAAsC,mBAArBqW,EAAkCA,EAAiBnX,EAAMc,SAAWqW,EACzFC,IACApX,EAAMc,QAAUzE,OAAOkE,OAAO,CAAC,EAAGsW,EAAgB7W,EAAMc,QAASA,GACjEd,EAAMiK,cAAgB,CACpBtM,UAAW0B,EAAU1B,GAAa6N,GAAkB7N,GAAaA,EAAU4Q,eAAiB/C,GAAkB7N,EAAU4Q,gBAAkB,GAC1I7Q,OAAQ8N,GAAkB9N,IAI5B,IElE4B+X,EAC9B4B,EFiEMN,EDhCG,SAAwBtB,GAErC,IAAIsB,EAAmBvB,GAAMC,GAE7B,OAAO/W,EAAeb,QAAO,SAAUC,EAAK+B,GAC1C,OAAO/B,EAAIE,OAAO+Y,EAAiBvR,QAAO,SAAUqQ,GAClD,OAAOA,EAAShW,QAAUA,CAC5B,IACF,GAAG,GACL,CCuB+ByX,EElEK7B,EFkEsB,GAAGzX,OAAO2Y,EAAkB3W,EAAMc,QAAQ2U,WEjE9F4B,EAAS5B,EAAU5X,QAAO,SAAUwZ,EAAQE,GAC9C,IAAIC,EAAWH,EAAOE,EAAQ5X,MAK9B,OAJA0X,EAAOE,EAAQ5X,MAAQ6X,EAAWnb,OAAOkE,OAAO,CAAC,EAAGiX,EAAUD,EAAS,CACrEzW,QAASzE,OAAOkE,OAAO,CAAC,EAAGiX,EAAS1W,QAASyW,EAAQzW,SACrD4I,KAAMrN,OAAOkE,OAAO,CAAC,EAAGiX,EAAS9N,KAAM6N,EAAQ7N,QAC5C6N,EACEF,CACT,GAAG,CAAC,GAEGhb,OAAO4D,KAAKoX,GAAQlV,KAAI,SAAUhG,GACvC,OAAOkb,EAAOlb,EAChB,MF4DM,OAJA6D,EAAM+W,iBAAmBA,EAAiBvR,QAAO,SAAUiS,GACzD,OAAOA,EAAE7X,OACX,IA+FFI,EAAM+W,iBAAiB5W,SAAQ,SAAUJ,GACvC,IAAIJ,EAAOI,EAAKJ,KACZ+X,EAAe3X,EAAKe,QACpBA,OAA2B,IAAjB4W,EAA0B,CAAC,EAAIA,EACzChX,EAASX,EAAKW,OAElB,GAAsB,mBAAXA,EAAuB,CAChC,IAAIiX,EAAYjX,EAAO,CACrBV,MAAOA,EACPL,KAAMA,EACNiK,SAAUA,EACV9I,QAASA,IAKXkW,EAAiB/F,KAAK0G,GAFT,WAAmB,EAGlC,CACF,IA/GS/N,EAASQ,QAClB,EAMAwN,YAAa,WACX,IAAIX,EAAJ,CAIA,IAAIY,EAAkB7X,EAAME,SACxBvC,EAAYka,EAAgBla,UAC5BD,EAASma,EAAgBna,OAG7B,GAAKyY,GAAiBxY,EAAWD,GAAjC,CAKAsC,EAAMwG,MAAQ,CACZ7I,UAAWwX,GAAiBxX,EAAWqH,EAAgBtH,GAAoC,UAA3BsC,EAAMc,QAAQC,UAC9ErD,OAAQiG,EAAcjG,IAOxBsC,EAAM0R,OAAQ,EACd1R,EAAMjC,UAAYiC,EAAMc,QAAQ/C,UAKhCiC,EAAM+W,iBAAiB5W,SAAQ,SAAU0V,GACvC,OAAO7V,EAAMmG,cAAc0P,EAASlW,MAAQtD,OAAOkE,OAAO,CAAC,EAAGsV,EAASnM,KACzE,IAEA,IAAK,IAAIoO,EAAQ,EAAGA,EAAQ9X,EAAM+W,iBAAiBhH,OAAQ+H,IACzD,IAAoB,IAAhB9X,EAAM0R,MAAV,CAMA,IAAIqG,EAAwB/X,EAAM+W,iBAAiBe,GAC/ChY,EAAKiY,EAAsBjY,GAC3BkY,EAAyBD,EAAsBjX,QAC/CoM,OAAsC,IAA3B8K,EAAoC,CAAC,EAAIA,EACpDrY,EAAOoY,EAAsBpY,KAEf,mBAAPG,IACTE,EAAQF,EAAG,CACTE,MAAOA,EACPc,QAASoM,EACTvN,KAAMA,EACNiK,SAAUA,KACN5J,EAdR,MAHEA,EAAM0R,OAAQ,EACdoG,GAAS,CAzBb,CATA,CAqDF,EAGA1N,QC1I2BtK,ED0IV,WACf,OAAO,IAAImY,SAAQ,SAAUC,GAC3BtO,EAASgO,cACTM,EAAQlY,EACV,GACF,EC7IG,WAUL,OATK8W,IACHA,EAAU,IAAImB,SAAQ,SAAUC,GAC9BD,QAAQC,UAAUC,MAAK,WACrBrB,OAAUsB,EACVF,EAAQpY,IACV,GACF,KAGKgX,CACT,GDmIIuB,QAAS,WACPjB,IACAH,GAAc,CAChB,GAGF,IAAKd,GAAiBxY,EAAWD,GAC/B,OAAOkM,EAmCT,SAASwN,IACPJ,EAAiB7W,SAAQ,SAAUL,GACjC,OAAOA,GACT,IACAkX,EAAmB,EACrB,CAEA,OAvCApN,EAASsN,WAAWpW,GAASqX,MAAK,SAAUnY,IACrCiX,GAAenW,EAAQwX,eAC1BxX,EAAQwX,cAActY,EAE1B,IAmCO4J,CACT,CACF,CACO,IAAI2O,GAA4BhC,KGzLnC,GAA4BA,GAAgB,CAC9CI,iBAFqB,CAAC6B,GAAgB,GAAe,GAAe,EAAa,GAAQ,GAAM,GAAiB,EAAO,MCJrH,GAA4BjC,GAAgB,CAC9CI,iBAFqB,CAAC6B,GAAgB,GAAe,GAAe,KCatE,MAAMC,GAAa,IAAIlI,IACjBmI,GAAO,CACX,GAAAtH,CAAIxS,EAASzC,EAAKyN,GACX6O,GAAWzC,IAAIpX,IAClB6Z,GAAWrH,IAAIxS,EAAS,IAAI2R,KAE9B,MAAMoI,EAAcF,GAAWjc,IAAIoC,GAI9B+Z,EAAY3C,IAAI7Z,IAA6B,IAArBwc,EAAYC,KAKzCD,EAAYvH,IAAIjV,EAAKyN,GAHnBiP,QAAQC,MAAM,+EAA+E7W,MAAM8W,KAAKJ,EAAY1Y,QAAQ,MAIhI,EACAzD,IAAG,CAACoC,EAASzC,IACPsc,GAAWzC,IAAIpX,IACV6Z,GAAWjc,IAAIoC,GAASpC,IAAIL,IAE9B,KAET,MAAA6c,CAAOpa,EAASzC,GACd,IAAKsc,GAAWzC,IAAIpX,GAClB,OAEF,MAAM+Z,EAAcF,GAAWjc,IAAIoC,GACnC+Z,EAAYM,OAAO9c,GAGM,IAArBwc,EAAYC,MACdH,GAAWQ,OAAOra,EAEtB,GAYIsa,GAAiB,gBAOjBC,GAAgBC,IAChBA,GAAYna,OAAOoa,KAAOpa,OAAOoa,IAAIC,SAEvCF,EAAWA,EAAS5O,QAAQ,iBAAiB,CAAC+O,EAAOC,IAAO,IAAIH,IAAIC,OAAOE,QAEtEJ,GA4CHK,GAAuB7a,IAC3BA,EAAQ8a,cAAc,IAAIC,MAAMT,IAAgB,EAE5C,GAAYU,MACXA,GAA4B,iBAAXA,UAGO,IAAlBA,EAAOC,SAChBD,EAASA,EAAO,SAEgB,IAApBA,EAAOE,UAEjBC,GAAaH,GAEb,GAAUA,GACLA,EAAOC,OAASD,EAAO,GAAKA,EAEf,iBAAXA,GAAuBA,EAAO7J,OAAS,EACzCrL,SAAS+C,cAAc0R,GAAcS,IAEvC,KAEHI,GAAYpb,IAChB,IAAK,GAAUA,IAAgD,IAApCA,EAAQqb,iBAAiBlK,OAClD,OAAO,EAET,MAAMmK,EAAgF,YAA7D5V,iBAAiB1F,GAASub,iBAAiB,cAE9DC,EAAgBxb,EAAQyb,QAAQ,uBACtC,IAAKD,EACH,OAAOF,EAET,GAAIE,IAAkBxb,EAAS,CAC7B,MAAM0b,EAAU1b,EAAQyb,QAAQ,WAChC,GAAIC,GAAWA,EAAQlW,aAAegW,EACpC,OAAO,EAET,GAAgB,OAAZE,EACF,OAAO,CAEX,CACA,OAAOJ,CAAgB,EAEnBK,GAAa3b,IACZA,GAAWA,EAAQkb,WAAaU,KAAKC,gBAGtC7b,EAAQ8b,UAAU7W,SAAS,mBAGC,IAArBjF,EAAQ+b,SACV/b,EAAQ+b,SAEV/b,EAAQgc,aAAa,aAAoD,UAArChc,EAAQic,aAAa,aAE5DC,GAAiBlc,IACrB,IAAK8F,SAASC,gBAAgBoW,aAC5B,OAAO,KAIT,GAAmC,mBAAxBnc,EAAQqF,YAA4B,CAC7C,MAAM+W,EAAOpc,EAAQqF,cACrB,OAAO+W,aAAgBtb,WAAasb,EAAO,IAC7C,CACA,OAAIpc,aAAmBc,WACdd,EAIJA,EAAQwF,WAGN0W,GAAelc,EAAQwF,YAFrB,IAEgC,EAErC6W,GAAO,OAUPC,GAAStc,IACbA,EAAQuE,YAAY,EAEhBgY,GAAY,IACZlc,OAAOmc,SAAW1W,SAAS6G,KAAKqP,aAAa,qBACxC3b,OAAOmc,OAET,KAEHC,GAA4B,GAgB5BC,GAAQ,IAAuC,QAAjC5W,SAASC,gBAAgB4W,IACvCC,GAAqBC,IAhBAC,QAiBN,KACjB,MAAMC,EAAIR,KAEV,GAAIQ,EAAG,CACL,MAAMhc,EAAO8b,EAAOG,KACdC,EAAqBF,EAAE7b,GAAGH,GAChCgc,EAAE7b,GAAGH,GAAQ8b,EAAOK,gBACpBH,EAAE7b,GAAGH,GAAMoc,YAAcN,EACzBE,EAAE7b,GAAGH,GAAMqc,WAAa,KACtBL,EAAE7b,GAAGH,GAAQkc,EACNJ,EAAOK,gBAElB,GA5B0B,YAAxBpX,SAASuX,YAENZ,GAA0BtL,QAC7BrL,SAASyF,iBAAiB,oBAAoB,KAC5C,IAAK,MAAMuR,KAAYL,GACrBK,GACF,IAGJL,GAA0BpK,KAAKyK,IAE/BA,GAkBA,EAEEQ,GAAU,CAACC,EAAkB9F,EAAO,GAAI+F,EAAeD,IACxB,mBAArBA,EAAkCA,KAAoB9F,GAAQ+F,EAExEC,GAAyB,CAACX,EAAUY,EAAmBC,GAAoB,KAC/E,IAAKA,EAEH,YADAL,GAAQR,GAGV,MACMc,EA/JiC5d,KACvC,IAAKA,EACH,OAAO,EAIT,IAAI,mBACF6d,EAAkB,gBAClBC,GACEzd,OAAOqF,iBAAiB1F,GAC5B,MAAM+d,EAA0BC,OAAOC,WAAWJ,GAC5CK,EAAuBF,OAAOC,WAAWH,GAG/C,OAAKC,GAA4BG,GAKjCL,EAAqBA,EAAmBlb,MAAM,KAAK,GACnDmb,EAAkBA,EAAgBnb,MAAM,KAAK,GAtDf,KAuDtBqb,OAAOC,WAAWJ,GAAsBG,OAAOC,WAAWH,KANzD,CAMoG,EA0IpFK,CAAiCT,GADlC,EAExB,IAAIU,GAAS,EACb,MAAMC,EAAU,EACdrR,aAEIA,IAAW0Q,IAGfU,GAAS,EACTV,EAAkBjS,oBAAoB6O,GAAgB+D,GACtDf,GAAQR,GAAS,EAEnBY,EAAkBnS,iBAAiB+O,GAAgB+D,GACnDC,YAAW,KACJF,GACHvD,GAAqB6C,EACvB,GACCE,EAAiB,EAYhBW,GAAuB,CAAC1R,EAAM2R,EAAeC,EAAeC,KAChE,MAAMC,EAAa9R,EAAKsE,OACxB,IAAI+H,EAAQrM,EAAKjH,QAAQ4Y,GAIzB,OAAe,IAAXtF,GACMuF,GAAiBC,EAAiB7R,EAAK8R,EAAa,GAAK9R,EAAK,IAExEqM,GAASuF,EAAgB,GAAK,EAC1BC,IACFxF,GAASA,EAAQyF,GAAcA,GAE1B9R,EAAKjK,KAAKC,IAAI,EAAGD,KAAKE,IAAIoW,EAAOyF,EAAa,KAAI,EAerDC,GAAiB,qBACjBC,GAAiB,OACjBC,GAAgB,SAChBC,GAAgB,CAAC,EACvB,IAAIC,GAAW,EACf,MAAMC,GAAe,CACnBC,WAAY,YACZC,WAAY,YAERC,GAAe,IAAIrI,IAAI,CAAC,QAAS,WAAY,UAAW,YAAa,cAAe,aAAc,iBAAkB,YAAa,WAAY,YAAa,cAAe,YAAa,UAAW,WAAY,QAAS,oBAAqB,aAAc,YAAa,WAAY,cAAe,cAAe,cAAe,YAAa,eAAgB,gBAAiB,eAAgB,gBAAiB,aAAc,QAAS,OAAQ,SAAU,QAAS,SAAU,SAAU,UAAW,WAAY,OAAQ,SAAU,eAAgB,SAAU,OAAQ,mBAAoB,mBAAoB,QAAS,QAAS,WAM/lB,SAASsI,GAAarf,EAASsf,GAC7B,OAAOA,GAAO,GAAGA,MAAQN,QAAgBhf,EAAQgf,UAAYA,IAC/D,CACA,SAASO,GAAiBvf,GACxB,MAAMsf,EAAMD,GAAarf,GAGzB,OAFAA,EAAQgf,SAAWM,EACnBP,GAAcO,GAAOP,GAAcO,IAAQ,CAAC,EACrCP,GAAcO,EACvB,CAiCA,SAASE,GAAYC,EAAQC,EAAUC,EAAqB,MAC1D,OAAOliB,OAAOmiB,OAAOH,GAAQ7M,MAAKiN,GAASA,EAAMH,WAAaA,GAAYG,EAAMF,qBAAuBA,GACzG,CACA,SAASG,GAAoBC,EAAmB1B,EAAS2B,GACvD,MAAMC,EAAiC,iBAAZ5B,EAErBqB,EAAWO,EAAcD,EAAqB3B,GAAW2B,EAC/D,IAAIE,EAAYC,GAAaJ,GAI7B,OAHKX,GAAahI,IAAI8I,KACpBA,EAAYH,GAEP,CAACE,EAAaP,EAAUQ,EACjC,CACA,SAASE,GAAWpgB,EAAS+f,EAAmB1B,EAAS2B,EAAoBK,GAC3E,GAAiC,iBAAtBN,IAAmC/f,EAC5C,OAEF,IAAKigB,EAAaP,EAAUQ,GAAaJ,GAAoBC,EAAmB1B,EAAS2B,GAIzF,GAAID,KAAqBd,GAAc,CACrC,MAAMqB,EAAepf,GACZ,SAAU2e,GACf,IAAKA,EAAMU,eAAiBV,EAAMU,gBAAkBV,EAAMW,iBAAmBX,EAAMW,eAAevb,SAAS4a,EAAMU,eAC/G,OAAOrf,EAAGjD,KAAKwiB,KAAMZ,EAEzB,EAEFH,EAAWY,EAAaZ,EAC1B,CACA,MAAMD,EAASF,GAAiBvf,GAC1B0gB,EAAWjB,EAAOS,KAAeT,EAAOS,GAAa,CAAC,GACtDS,EAAmBnB,GAAYkB,EAAUhB,EAAUO,EAAc5B,EAAU,MACjF,GAAIsC,EAEF,YADAA,EAAiBN,OAASM,EAAiBN,QAAUA,GAGvD,MAAMf,EAAMD,GAAaK,EAAUK,EAAkBnU,QAAQgT,GAAgB,KACvE1d,EAAK+e,EA5Db,SAAoCjgB,EAASwa,EAAUtZ,GACrD,OAAO,SAASmd,EAAQwB,GACtB,MAAMe,EAAc5gB,EAAQ6gB,iBAAiBrG,GAC7C,IAAK,IAAI,OACPxN,GACE6S,EAAO7S,GAAUA,IAAWyT,KAAMzT,EAASA,EAAOxH,WACpD,IAAK,MAAMsb,KAAcF,EACvB,GAAIE,IAAe9T,EASnB,OANA+T,GAAWlB,EAAO,CAChBW,eAAgBxT,IAEdqR,EAAQgC,QACVW,GAAaC,IAAIjhB,EAAS6f,EAAMqB,KAAM1G,EAAUtZ,GAE3CA,EAAGigB,MAAMnU,EAAQ,CAAC6S,GAG/B,CACF,CAwC2BuB,CAA2BphB,EAASqe,EAASqB,GAvExE,SAA0B1f,EAASkB,GACjC,OAAO,SAASmd,EAAQwB,GAOtB,OANAkB,GAAWlB,EAAO,CAChBW,eAAgBxgB,IAEdqe,EAAQgC,QACVW,GAAaC,IAAIjhB,EAAS6f,EAAMqB,KAAMhgB,GAEjCA,EAAGigB,MAAMnhB,EAAS,CAAC6f,GAC5B,CACF,CA6DoFwB,CAAiBrhB,EAAS0f,GAC5Gxe,EAAGye,mBAAqBM,EAAc5B,EAAU,KAChDnd,EAAGwe,SAAWA,EACdxe,EAAGmf,OAASA,EACZnf,EAAG8d,SAAWM,EACdoB,EAASpB,GAAOpe,EAChBlB,EAAQuL,iBAAiB2U,EAAWhf,EAAI+e,EAC1C,CACA,SAASqB,GAActhB,EAASyf,EAAQS,EAAW7B,EAASsB,GAC1D,MAAMze,EAAKse,GAAYC,EAAOS,GAAY7B,EAASsB,GAC9Cze,IAGLlB,EAAQyL,oBAAoByU,EAAWhf,EAAIqgB,QAAQ5B,WAC5CF,EAAOS,GAAWhf,EAAG8d,UAC9B,CACA,SAASwC,GAAyBxhB,EAASyf,EAAQS,EAAWuB,GAC5D,MAAMC,EAAoBjC,EAAOS,IAAc,CAAC,EAChD,IAAK,MAAOyB,EAAY9B,KAAUpiB,OAAOmkB,QAAQF,GAC3CC,EAAWE,SAASJ,IACtBH,GAActhB,EAASyf,EAAQS,EAAWL,EAAMH,SAAUG,EAAMF,mBAGtE,CACA,SAASQ,GAAaN,GAGpB,OADAA,EAAQA,EAAMjU,QAAQiT,GAAgB,IAC/BI,GAAaY,IAAUA,CAChC,CACA,MAAMmB,GAAe,CACnB,EAAAc,CAAG9hB,EAAS6f,EAAOxB,EAAS2B,GAC1BI,GAAWpgB,EAAS6f,EAAOxB,EAAS2B,GAAoB,EAC1D,EACA,GAAA+B,CAAI/hB,EAAS6f,EAAOxB,EAAS2B,GAC3BI,GAAWpgB,EAAS6f,EAAOxB,EAAS2B,GAAoB,EAC1D,EACA,GAAAiB,CAAIjhB,EAAS+f,EAAmB1B,EAAS2B,GACvC,GAAiC,iBAAtBD,IAAmC/f,EAC5C,OAEF,MAAOigB,EAAaP,EAAUQ,GAAaJ,GAAoBC,EAAmB1B,EAAS2B,GACrFgC,EAAc9B,IAAcH,EAC5BN,EAASF,GAAiBvf,GAC1B0hB,EAAoBjC,EAAOS,IAAc,CAAC,EAC1C+B,EAAclC,EAAkBmC,WAAW,KACjD,QAAwB,IAAbxC,EAAX,CAQA,GAAIuC,EACF,IAAK,MAAME,KAAgB1kB,OAAO4D,KAAKoe,GACrC+B,GAAyBxhB,EAASyf,EAAQ0C,EAAcpC,EAAkBlN,MAAM,IAGpF,IAAK,MAAOuP,EAAavC,KAAUpiB,OAAOmkB,QAAQF,GAAoB,CACpE,MAAMC,EAAaS,EAAYxW,QAAQkT,GAAe,IACjDkD,IAAejC,EAAkB8B,SAASF,IAC7CL,GAActhB,EAASyf,EAAQS,EAAWL,EAAMH,SAAUG,EAAMF,mBAEpE,CAXA,KAPA,CAEE,IAAKliB,OAAO4D,KAAKqgB,GAAmBvQ,OAClC,OAEFmQ,GAActhB,EAASyf,EAAQS,EAAWR,EAAUO,EAAc5B,EAAU,KAE9E,CAYF,EACA,OAAAgE,CAAQriB,EAAS6f,EAAOpI,GACtB,GAAqB,iBAAVoI,IAAuB7f,EAChC,OAAO,KAET,MAAM+c,EAAIR,KAGV,IAAI+F,EAAc,KACdC,GAAU,EACVC,GAAiB,EACjBC,GAAmB,EAJH5C,IADFM,GAAaN,IAMZ9C,IACjBuF,EAAcvF,EAAEhC,MAAM8E,EAAOpI,GAC7BsF,EAAE/c,GAASqiB,QAAQC,GACnBC,GAAWD,EAAYI,uBACvBF,GAAkBF,EAAYK,gCAC9BF,EAAmBH,EAAYM,sBAEjC,MAAMC,EAAM9B,GAAW,IAAIhG,MAAM8E,EAAO,CACtC0C,UACAO,YAAY,IACVrL,GAUJ,OATIgL,GACFI,EAAIE,iBAEFP,GACFxiB,EAAQ8a,cAAc+H,GAEpBA,EAAIJ,kBAAoBH,GAC1BA,EAAYS,iBAEPF,CACT,GAEF,SAAS9B,GAAWljB,EAAKmlB,EAAO,CAAC,GAC/B,IAAK,MAAOzlB,EAAKa,KAAUX,OAAOmkB,QAAQoB,GACxC,IACEnlB,EAAIN,GAAOa,CACb,CAAE,MAAO6kB,GACPxlB,OAAOC,eAAeG,EAAKN,EAAK,CAC9B2lB,cAAc,EACdtlB,IAAG,IACMQ,GAGb,CAEF,OAAOP,CACT,CASA,SAASslB,GAAc/kB,GACrB,GAAc,SAAVA,EACF,OAAO,EAET,GAAc,UAAVA,EACF,OAAO,EAET,GAAIA,IAAU4f,OAAO5f,GAAOkC,WAC1B,OAAO0d,OAAO5f,GAEhB,GAAc,KAAVA,GAA0B,SAAVA,EAClB,OAAO,KAET,GAAqB,iBAAVA,EACT,OAAOA,EAET,IACE,OAAOglB,KAAKC,MAAMC,mBAAmBllB,GACvC,CAAE,MAAO6kB,GACP,OAAO7kB,CACT,CACF,CACA,SAASmlB,GAAiBhmB,GACxB,OAAOA,EAAIqO,QAAQ,UAAU4X,GAAO,IAAIA,EAAItjB,iBAC9C,CACA,MAAMujB,GAAc,CAClB,gBAAAC,CAAiB1jB,EAASzC,EAAKa,GAC7B4B,EAAQ6B,aAAa,WAAW0hB,GAAiBhmB,KAAQa,EAC3D,EACA,mBAAAulB,CAAoB3jB,EAASzC,GAC3ByC,EAAQ4B,gBAAgB,WAAW2hB,GAAiBhmB,KACtD,EACA,iBAAAqmB,CAAkB5jB,GAChB,IAAKA,EACH,MAAO,CAAC,EAEV,MAAM0B,EAAa,CAAC,EACdmiB,EAASpmB,OAAO4D,KAAKrB,EAAQ8jB,SAASld,QAAOrJ,GAAOA,EAAI2kB,WAAW,QAAU3kB,EAAI2kB,WAAW,cAClG,IAAK,MAAM3kB,KAAOsmB,EAAQ,CACxB,IAAIE,EAAUxmB,EAAIqO,QAAQ,MAAO,IACjCmY,EAAUA,EAAQC,OAAO,GAAG9jB,cAAgB6jB,EAAQlR,MAAM,EAAGkR,EAAQ5S,QACrEzP,EAAWqiB,GAAWZ,GAAcnjB,EAAQ8jB,QAAQvmB,GACtD,CACA,OAAOmE,CACT,EACAuiB,iBAAgB,CAACjkB,EAASzC,IACjB4lB,GAAcnjB,EAAQic,aAAa,WAAWsH,GAAiBhmB,QAgB1E,MAAM2mB,GAEJ,kBAAWC,GACT,MAAO,CAAC,CACV,CACA,sBAAWC,GACT,MAAO,CAAC,CACV,CACA,eAAWpH,GACT,MAAM,IAAIqH,MAAM,sEAClB,CACA,UAAAC,CAAWC,GAIT,OAHAA,EAAS9D,KAAK+D,gBAAgBD,GAC9BA,EAAS9D,KAAKgE,kBAAkBF,GAChC9D,KAAKiE,iBAAiBH,GACfA,CACT,CACA,iBAAAE,CAAkBF,GAChB,OAAOA,CACT,CACA,eAAAC,CAAgBD,EAAQvkB,GACtB,MAAM2kB,EAAa,GAAU3kB,GAAWyjB,GAAYQ,iBAAiBjkB,EAAS,UAAY,CAAC,EAE3F,MAAO,IACFygB,KAAKmE,YAAYT,WACM,iBAAfQ,EAA0BA,EAAa,CAAC,KAC/C,GAAU3kB,GAAWyjB,GAAYG,kBAAkB5jB,GAAW,CAAC,KAC7C,iBAAXukB,EAAsBA,EAAS,CAAC,EAE/C,CACA,gBAAAG,CAAiBH,EAAQM,EAAcpE,KAAKmE,YAAYR,aACtD,IAAK,MAAO7hB,EAAUuiB,KAAkBrnB,OAAOmkB,QAAQiD,GAAc,CACnE,MAAMzmB,EAAQmmB,EAAOhiB,GACfwiB,EAAY,GAAU3mB,GAAS,UAhiBrC4c,OADSA,EAiiB+C5c,GA/hBnD,GAAG4c,IAELvd,OAAOM,UAAUuC,SAASrC,KAAK+c,GAAQL,MAAM,eAAe,GAAGza,cA8hBlE,IAAK,IAAI8kB,OAAOF,GAAehhB,KAAKihB,GAClC,MAAM,IAAIE,UAAU,GAAGxE,KAAKmE,YAAY5H,KAAKkI,0BAA0B3iB,qBAA4BwiB,yBAAiCD,MAExI,CAriBW9J,KAsiBb,EAqBF,MAAMmK,WAAsBjB,GAC1B,WAAAU,CAAY5kB,EAASukB,GACnBa,SACAplB,EAAUmb,GAAWnb,MAIrBygB,KAAK4E,SAAWrlB,EAChBygB,KAAK6E,QAAU7E,KAAK6D,WAAWC,GAC/BzK,GAAKtH,IAAIiO,KAAK4E,SAAU5E,KAAKmE,YAAYW,SAAU9E,MACrD,CAGA,OAAA+E,GACE1L,GAAKM,OAAOqG,KAAK4E,SAAU5E,KAAKmE,YAAYW,UAC5CvE,GAAaC,IAAIR,KAAK4E,SAAU5E,KAAKmE,YAAYa,WACjD,IAAK,MAAMC,KAAgBjoB,OAAOkoB,oBAAoBlF,MACpDA,KAAKiF,GAAgB,IAEzB,CACA,cAAAE,CAAe9I,EAAU9c,EAAS6lB,GAAa,GAC7CpI,GAAuBX,EAAU9c,EAAS6lB,EAC5C,CACA,UAAAvB,CAAWC,GAIT,OAHAA,EAAS9D,KAAK+D,gBAAgBD,EAAQ9D,KAAK4E,UAC3Cd,EAAS9D,KAAKgE,kBAAkBF,GAChC9D,KAAKiE,iBAAiBH,GACfA,CACT,CAGA,kBAAOuB,CAAY9lB,GACjB,OAAO8Z,GAAKlc,IAAIud,GAAWnb,GAAUygB,KAAK8E,SAC5C,CACA,0BAAOQ,CAAoB/lB,EAASukB,EAAS,CAAC,GAC5C,OAAO9D,KAAKqF,YAAY9lB,IAAY,IAAIygB,KAAKzgB,EAA2B,iBAAXukB,EAAsBA,EAAS,KAC9F,CACA,kBAAWyB,GACT,MA5CY,OA6Cd,CACA,mBAAWT,GACT,MAAO,MAAM9E,KAAKzD,MACpB,CACA,oBAAWyI,GACT,MAAO,IAAIhF,KAAK8E,UAClB,CACA,gBAAOU,CAAUllB,GACf,MAAO,GAAGA,IAAO0f,KAAKgF,WACxB,EAUF,MAAMS,GAAclmB,IAClB,IAAIwa,EAAWxa,EAAQic,aAAa,kBACpC,IAAKzB,GAAyB,MAAbA,EAAkB,CACjC,IAAI2L,EAAgBnmB,EAAQic,aAAa,QAMzC,IAAKkK,IAAkBA,EAActE,SAAS,OAASsE,EAAcjE,WAAW,KAC9E,OAAO,KAILiE,EAActE,SAAS,OAASsE,EAAcjE,WAAW,OAC3DiE,EAAgB,IAAIA,EAAcxjB,MAAM,KAAK,MAE/C6X,EAAW2L,GAAmC,MAAlBA,EAAwBA,EAAcC,OAAS,IAC7E,CACA,OAAO5L,EAAWA,EAAS7X,MAAM,KAAKY,KAAI8iB,GAAO9L,GAAc8L,KAAM1iB,KAAK,KAAO,IAAI,EAEjF2iB,GAAiB,CACrB1T,KAAI,CAAC4H,EAAUxa,EAAU8F,SAASC,kBACzB,GAAG3G,UAAUsB,QAAQ3C,UAAU8iB,iBAAiB5iB,KAAK+B,EAASwa,IAEvE+L,QAAO,CAAC/L,EAAUxa,EAAU8F,SAASC,kBAC5BrF,QAAQ3C,UAAU8K,cAAc5K,KAAK+B,EAASwa,GAEvDgM,SAAQ,CAACxmB,EAASwa,IACT,GAAGpb,UAAUY,EAAQwmB,UAAU5f,QAAOzB,GAASA,EAAMshB,QAAQjM,KAEtE,OAAAkM,CAAQ1mB,EAASwa,GACf,MAAMkM,EAAU,GAChB,IAAIC,EAAW3mB,EAAQwF,WAAWiW,QAAQjB,GAC1C,KAAOmM,GACLD,EAAQrU,KAAKsU,GACbA,EAAWA,EAASnhB,WAAWiW,QAAQjB,GAEzC,OAAOkM,CACT,EACA,IAAAE,CAAK5mB,EAASwa,GACZ,IAAIqM,EAAW7mB,EAAQ8mB,uBACvB,KAAOD,GAAU,CACf,GAAIA,EAASJ,QAAQjM,GACnB,MAAO,CAACqM,GAEVA,EAAWA,EAASC,sBACtB,CACA,MAAO,EACT,EAEA,IAAAxhB,CAAKtF,EAASwa,GACZ,IAAIlV,EAAOtF,EAAQ+mB,mBACnB,KAAOzhB,GAAM,CACX,GAAIA,EAAKmhB,QAAQjM,GACf,MAAO,CAAClV,GAEVA,EAAOA,EAAKyhB,kBACd,CACA,MAAO,EACT,EACA,iBAAAC,CAAkBhnB,GAChB,MAAMinB,EAAa,CAAC,IAAK,SAAU,QAAS,WAAY,SAAU,UAAW,aAAc,4BAA4B1jB,KAAIiX,GAAY,GAAGA,2BAAiC7W,KAAK,KAChL,OAAO8c,KAAK7N,KAAKqU,EAAYjnB,GAAS4G,QAAOsgB,IAAOvL,GAAWuL,IAAO9L,GAAU8L,IAClF,EACA,sBAAAC,CAAuBnnB,GACrB,MAAMwa,EAAW0L,GAAYlmB,GAC7B,OAAIwa,GACK8L,GAAeC,QAAQ/L,GAAYA,EAErC,IACT,EACA,sBAAA4M,CAAuBpnB,GACrB,MAAMwa,EAAW0L,GAAYlmB,GAC7B,OAAOwa,EAAW8L,GAAeC,QAAQ/L,GAAY,IACvD,EACA,+BAAA6M,CAAgCrnB,GAC9B,MAAMwa,EAAW0L,GAAYlmB,GAC7B,OAAOwa,EAAW8L,GAAe1T,KAAK4H,GAAY,EACpD,GAUI8M,GAAuB,CAACC,EAAWC,EAAS,UAChD,MAAMC,EAAa,gBAAgBF,EAAU9B,YACvC1kB,EAAOwmB,EAAUvK,KACvBgE,GAAac,GAAGhc,SAAU2hB,EAAY,qBAAqB1mB,OAAU,SAAU8e,GAI7E,GAHI,CAAC,IAAK,QAAQgC,SAASpB,KAAKiH,UAC9B7H,EAAMkD,iBAEJpH,GAAW8E,MACb,OAEF,MAAMzT,EAASsZ,GAAec,uBAAuB3G,OAASA,KAAKhF,QAAQ,IAAI1a,KAC9DwmB,EAAUxB,oBAAoB/Y,GAGtCwa,IACX,GAAE,EAiBEG,GAAc,YACdC,GAAc,QAAQD,KACtBE,GAAe,SAASF,KAQ9B,MAAMG,WAAc3C,GAElB,eAAWnI,GACT,MAfW,OAgBb,CAGA,KAAA+K,GAEE,GADmB/G,GAAaqB,QAAQ5B,KAAK4E,SAAUuC,IACxCnF,iBACb,OAEFhC,KAAK4E,SAASvJ,UAAU1B,OAlBF,QAmBtB,MAAMyL,EAAapF,KAAK4E,SAASvJ,UAAU7W,SApBrB,QAqBtBwb,KAAKmF,gBAAe,IAAMnF,KAAKuH,mBAAmBvH,KAAK4E,SAAUQ,EACnE,CAGA,eAAAmC,GACEvH,KAAK4E,SAASjL,SACd4G,GAAaqB,QAAQ5B,KAAK4E,SAAUwC,IACpCpH,KAAK+E,SACP,CAGA,sBAAOtI,CAAgBqH,GACrB,OAAO9D,KAAKwH,MAAK,WACf,MAAMnd,EAAOgd,GAAM/B,oBAAoBtF,MACvC,GAAsB,iBAAX8D,EAAX,CAGA,QAAqB/K,IAAjB1O,EAAKyZ,IAAyBA,EAAOrC,WAAW,MAAmB,gBAAXqC,EAC1D,MAAM,IAAIU,UAAU,oBAAoBV,MAE1CzZ,EAAKyZ,GAAQ9D,KAJb,CAKF,GACF,EAOF6G,GAAqBQ,GAAO,SAM5BlL,GAAmBkL,IAcnB,MAKMI,GAAyB,4BAO/B,MAAMC,WAAehD,GAEnB,eAAWnI,GACT,MAfW,QAgBb,CAGA,MAAAoL,GAEE3H,KAAK4E,SAASxjB,aAAa,eAAgB4e,KAAK4E,SAASvJ,UAAUsM,OAjB3C,UAkB1B,CAGA,sBAAOlL,CAAgBqH,GACrB,OAAO9D,KAAKwH,MAAK,WACf,MAAMnd,EAAOqd,GAAOpC,oBAAoBtF,MACzB,WAAX8D,GACFzZ,EAAKyZ,IAET,GACF,EAOFvD,GAAac,GAAGhc,SAjCe,2BAiCmBoiB,IAAwBrI,IACxEA,EAAMkD,iBACN,MAAMsF,EAASxI,EAAM7S,OAAOyO,QAAQyM,IACvBC,GAAOpC,oBAAoBsC,GACnCD,QAAQ,IAOfxL,GAAmBuL,IAcnB,MACMG,GAAc,YACdC,GAAmB,aAAaD,KAChCE,GAAkB,YAAYF,KAC9BG,GAAiB,WAAWH,KAC5BI,GAAoB,cAAcJ,KAClCK,GAAkB,YAAYL,KAK9BM,GAAY,CAChBC,YAAa,KACbC,aAAc,KACdC,cAAe,MAEXC,GAAgB,CACpBH,YAAa,kBACbC,aAAc,kBACdC,cAAe,mBAOjB,MAAME,WAAc/E,GAClB,WAAAU,CAAY5kB,EAASukB,GACnBa,QACA3E,KAAK4E,SAAWrlB,EACXA,GAAYipB,GAAMC,gBAGvBzI,KAAK6E,QAAU7E,KAAK6D,WAAWC,GAC/B9D,KAAK0I,QAAU,EACf1I,KAAK2I,sBAAwB7H,QAAQlhB,OAAOgpB,cAC5C5I,KAAK6I,cACP,CAGA,kBAAWnF,GACT,OAAOyE,EACT,CACA,sBAAWxE,GACT,OAAO4E,EACT,CACA,eAAWhM,GACT,MA/CW,OAgDb,CAGA,OAAAwI,GACExE,GAAaC,IAAIR,KAAK4E,SAAUiD,GAClC,CAGA,MAAAiB,CAAO1J,GACAY,KAAK2I,sBAIN3I,KAAK+I,wBAAwB3J,KAC/BY,KAAK0I,QAAUtJ,EAAM4J,SAJrBhJ,KAAK0I,QAAUtJ,EAAM6J,QAAQ,GAAGD,OAMpC,CACA,IAAAE,CAAK9J,GACCY,KAAK+I,wBAAwB3J,KAC/BY,KAAK0I,QAAUtJ,EAAM4J,QAAUhJ,KAAK0I,SAEtC1I,KAAKmJ,eACLtM,GAAQmD,KAAK6E,QAAQuD,YACvB,CACA,KAAAgB,CAAMhK,GACJY,KAAK0I,QAAUtJ,EAAM6J,SAAW7J,EAAM6J,QAAQvY,OAAS,EAAI,EAAI0O,EAAM6J,QAAQ,GAAGD,QAAUhJ,KAAK0I,OACjG,CACA,YAAAS,GACE,MAAME,EAAYlnB,KAAKoC,IAAIyb,KAAK0I,SAChC,GAAIW,GAnEgB,GAoElB,OAEF,MAAM/b,EAAY+b,EAAYrJ,KAAK0I,QACnC1I,KAAK0I,QAAU,EACVpb,GAGLuP,GAAQvP,EAAY,EAAI0S,KAAK6E,QAAQyD,cAAgBtI,KAAK6E,QAAQwD,aACpE,CACA,WAAAQ,GACM7I,KAAK2I,uBACPpI,GAAac,GAAGrB,KAAK4E,SAAUqD,IAAmB7I,GAASY,KAAK8I,OAAO1J,KACvEmB,GAAac,GAAGrB,KAAK4E,SAAUsD,IAAiB9I,GAASY,KAAKkJ,KAAK9J,KACnEY,KAAK4E,SAASvJ,UAAU5E,IAlFG,mBAoF3B8J,GAAac,GAAGrB,KAAK4E,SAAUkD,IAAkB1I,GAASY,KAAK8I,OAAO1J,KACtEmB,GAAac,GAAGrB,KAAK4E,SAAUmD,IAAiB3I,GAASY,KAAKoJ,MAAMhK,KACpEmB,GAAac,GAAGrB,KAAK4E,SAAUoD,IAAgB5I,GAASY,KAAKkJ,KAAK9J,KAEtE,CACA,uBAAA2J,CAAwB3J,GACtB,OAAOY,KAAK2I,wBA3FS,QA2FiBvJ,EAAMkK,aA5FrB,UA4FyDlK,EAAMkK,YACxF,CAGA,kBAAOb,GACL,MAAO,iBAAkBpjB,SAASC,iBAAmB7C,UAAU8mB,eAAiB,CAClF,EAeF,MAEMC,GAAc,eACdC,GAAiB,YACjBC,GAAmB,YACnBC,GAAoB,aAGpBC,GAAa,OACbC,GAAa,OACbC,GAAiB,OACjBC,GAAkB,QAClBC,GAAc,QAAQR,KACtBS,GAAa,OAAOT,KACpBU,GAAkB,UAAUV,KAC5BW,GAAqB,aAAaX,KAClCY,GAAqB,aAAaZ,KAClCa,GAAmB,YAAYb,KAC/Bc,GAAwB,OAAOd,KAAcC,KAC7Cc,GAAyB,QAAQf,KAAcC,KAC/Ce,GAAsB,WACtBC,GAAsB,SAMtBC,GAAkB,UAClBC,GAAgB,iBAChBC,GAAuBF,GAAkBC,GAKzCE,GAAmB,CACvB,CAACnB,IAAmBK,GACpB,CAACJ,IAAoBG,IAEjBgB,GAAY,CAChBC,SAAU,IACVC,UAAU,EACVC,MAAO,QACPC,MAAM,EACNC,OAAO,EACPC,MAAM,GAEFC,GAAgB,CACpBN,SAAU,mBAEVC,SAAU,UACVC,MAAO,mBACPC,KAAM,mBACNC,MAAO,UACPC,KAAM,WAOR,MAAME,WAAiB5G,GACrB,WAAAP,CAAY5kB,EAASukB,GACnBa,MAAMplB,EAASukB,GACf9D,KAAKuL,UAAY,KACjBvL,KAAKwL,eAAiB,KACtBxL,KAAKyL,YAAa,EAClBzL,KAAK0L,aAAe,KACpB1L,KAAK2L,aAAe,KACpB3L,KAAK4L,mBAAqB/F,GAAeC,QArCjB,uBAqC8C9F,KAAK4E,UAC3E5E,KAAK6L,qBACD7L,KAAK6E,QAAQqG,OAASV,IACxBxK,KAAK8L,OAET,CAGA,kBAAWpI,GACT,OAAOoH,EACT,CACA,sBAAWnH,GACT,OAAO0H,EACT,CACA,eAAW9O,GACT,MAnFW,UAoFb,CAGA,IAAA1X,GACEmb,KAAK+L,OAAOnC,GACd,CACA,eAAAoC,IAIO3mB,SAAS4mB,QAAUtR,GAAUqF,KAAK4E,WACrC5E,KAAKnb,MAET,CACA,IAAAshB,GACEnG,KAAK+L,OAAOlC,GACd,CACA,KAAAoB,GACMjL,KAAKyL,YACPrR,GAAqB4F,KAAK4E,UAE5B5E,KAAKkM,gBACP,CACA,KAAAJ,GACE9L,KAAKkM,iBACLlM,KAAKmM,kBACLnM,KAAKuL,UAAYa,aAAY,IAAMpM,KAAKgM,mBAAmBhM,KAAK6E,QAAQkG,SAC1E,CACA,iBAAAsB,GACOrM,KAAK6E,QAAQqG,OAGdlL,KAAKyL,WACPlL,GAAae,IAAItB,KAAK4E,SAAUqF,IAAY,IAAMjK,KAAK8L,UAGzD9L,KAAK8L,QACP,CACA,EAAAQ,CAAG7T,GACD,MAAM8T,EAAQvM,KAAKwM,YACnB,GAAI/T,EAAQ8T,EAAM7b,OAAS,GAAK+H,EAAQ,EACtC,OAEF,GAAIuH,KAAKyL,WAEP,YADAlL,GAAae,IAAItB,KAAK4E,SAAUqF,IAAY,IAAMjK,KAAKsM,GAAG7T,KAG5D,MAAMgU,EAAczM,KAAK0M,cAAc1M,KAAK2M,cAC5C,GAAIF,IAAgBhU,EAClB,OAEF,MAAMtC,EAAQsC,EAAQgU,EAAc7C,GAAaC,GACjD7J,KAAK+L,OAAO5V,EAAOoW,EAAM9T,GAC3B,CACA,OAAAsM,GACM/E,KAAK2L,cACP3L,KAAK2L,aAAa5G,UAEpBJ,MAAMI,SACR,CAGA,iBAAAf,CAAkBF,GAEhB,OADAA,EAAO8I,gBAAkB9I,EAAOiH,SACzBjH,CACT,CACA,kBAAA+H,GACM7L,KAAK6E,QAAQmG,UACfzK,GAAac,GAAGrB,KAAK4E,SAAUsF,IAAiB9K,GAASY,KAAK6M,SAASzN,KAE9C,UAAvBY,KAAK6E,QAAQoG,QACf1K,GAAac,GAAGrB,KAAK4E,SAAUuF,IAAoB,IAAMnK,KAAKiL,UAC9D1K,GAAac,GAAGrB,KAAK4E,SAAUwF,IAAoB,IAAMpK,KAAKqM,uBAE5DrM,KAAK6E,QAAQsG,OAAS3C,GAAMC,eAC9BzI,KAAK8M,yBAET,CACA,uBAAAA,GACE,IAAK,MAAMC,KAAOlH,GAAe1T,KArIX,qBAqImC6N,KAAK4E,UAC5DrE,GAAac,GAAG0L,EAAK1C,IAAkBjL,GAASA,EAAMkD,mBAExD,MAmBM0K,EAAc,CAClB3E,aAAc,IAAMrI,KAAK+L,OAAO/L,KAAKiN,kBAAkBnD,KACvDxB,cAAe,IAAMtI,KAAK+L,OAAO/L,KAAKiN,kBAAkBlD,KACxD3B,YAtBkB,KACS,UAAvBpI,KAAK6E,QAAQoG,QAYjBjL,KAAKiL,QACDjL,KAAK0L,cACPwB,aAAalN,KAAK0L,cAEpB1L,KAAK0L,aAAe7N,YAAW,IAAMmC,KAAKqM,qBAjLjB,IAiL+DrM,KAAK6E,QAAQkG,UAAS,GAOhH/K,KAAK2L,aAAe,IAAInD,GAAMxI,KAAK4E,SAAUoI,EAC/C,CACA,QAAAH,CAASzN,GACP,GAAI,kBAAkB/b,KAAK+b,EAAM7S,OAAO0a,SACtC,OAEF,MAAM3Z,EAAYud,GAAiBzL,EAAMtiB,KACrCwQ,IACF8R,EAAMkD,iBACNtC,KAAK+L,OAAO/L,KAAKiN,kBAAkB3f,IAEvC,CACA,aAAAof,CAAcntB,GACZ,OAAOygB,KAAKwM,YAAYrnB,QAAQ5F,EAClC,CACA,0BAAA4tB,CAA2B1U,GACzB,IAAKuH,KAAK4L,mBACR,OAEF,MAAMwB,EAAkBvH,GAAeC,QAAQ4E,GAAiB1K,KAAK4L,oBACrEwB,EAAgB/R,UAAU1B,OAAO8Q,IACjC2C,EAAgBjsB,gBAAgB,gBAChC,MAAMksB,EAAqBxH,GAAeC,QAAQ,sBAAsBrN,MAAWuH,KAAK4L,oBACpFyB,IACFA,EAAmBhS,UAAU5E,IAAIgU,IACjC4C,EAAmBjsB,aAAa,eAAgB,QAEpD,CACA,eAAA+qB,GACE,MAAM5sB,EAAUygB,KAAKwL,gBAAkBxL,KAAK2M,aAC5C,IAAKptB,EACH,OAEF,MAAM+tB,EAAkB/P,OAAOgQ,SAAShuB,EAAQic,aAAa,oBAAqB,IAClFwE,KAAK6E,QAAQkG,SAAWuC,GAAmBtN,KAAK6E,QAAQ+H,eAC1D,CACA,MAAAb,CAAO5V,EAAO5W,EAAU,MACtB,GAAIygB,KAAKyL,WACP,OAEF,MAAM1N,EAAgBiC,KAAK2M,aACrBa,EAASrX,IAAUyT,GACnB6D,EAAcluB,GAAWue,GAAqBkC,KAAKwM,YAAazO,EAAeyP,EAAQxN,KAAK6E,QAAQuG,MAC1G,GAAIqC,IAAgB1P,EAClB,OAEF,MAAM2P,EAAmB1N,KAAK0M,cAAce,GACtCE,EAAenI,GACZjF,GAAaqB,QAAQ5B,KAAK4E,SAAUY,EAAW,CACpD1F,cAAe2N,EACfngB,UAAW0S,KAAK4N,kBAAkBzX,GAClCuD,KAAMsG,KAAK0M,cAAc3O,GACzBuO,GAAIoB,IAIR,GADmBC,EAAa3D,IACjBhI,iBACb,OAEF,IAAKjE,IAAkB0P,EAGrB,OAEF,MAAMI,EAAY/M,QAAQd,KAAKuL,WAC/BvL,KAAKiL,QACLjL,KAAKyL,YAAa,EAClBzL,KAAKmN,2BAA2BO,GAChC1N,KAAKwL,eAAiBiC,EACtB,MAAMK,EAAuBN,EA3OR,sBADF,oBA6ObO,EAAiBP,EA3OH,qBACA,qBA2OpBC,EAAYpS,UAAU5E,IAAIsX,GAC1BlS,GAAO4R,GACP1P,EAAc1C,UAAU5E,IAAIqX,GAC5BL,EAAYpS,UAAU5E,IAAIqX,GAQ1B9N,KAAKmF,gBAPoB,KACvBsI,EAAYpS,UAAU1B,OAAOmU,EAAsBC,GACnDN,EAAYpS,UAAU5E,IAAIgU,IAC1B1M,EAAc1C,UAAU1B,OAAO8Q,GAAqBsD,EAAgBD,GACpE9N,KAAKyL,YAAa,EAClBkC,EAAa1D,GAAW,GAEYlM,EAAeiC,KAAKgO,eACtDH,GACF7N,KAAK8L,OAET,CACA,WAAAkC,GACE,OAAOhO,KAAK4E,SAASvJ,UAAU7W,SAhQV,QAiQvB,CACA,UAAAmoB,GACE,OAAO9G,GAAeC,QAAQ8E,GAAsB5K,KAAK4E,SAC3D,CACA,SAAA4H,GACE,OAAO3G,GAAe1T,KAAKwY,GAAe3K,KAAK4E,SACjD,CACA,cAAAsH,GACMlM,KAAKuL,YACP0C,cAAcjO,KAAKuL,WACnBvL,KAAKuL,UAAY,KAErB,CACA,iBAAA0B,CAAkB3f,GAChB,OAAI2O,KACK3O,IAAcwc,GAAiBD,GAAaD,GAE9Ctc,IAAcwc,GAAiBF,GAAaC,EACrD,CACA,iBAAA+D,CAAkBzX,GAChB,OAAI8F,KACK9F,IAAU0T,GAAaC,GAAiBC,GAE1C5T,IAAU0T,GAAaE,GAAkBD,EAClD,CAGA,sBAAOrN,CAAgBqH,GACrB,OAAO9D,KAAKwH,MAAK,WACf,MAAMnd,EAAOihB,GAAShG,oBAAoBtF,KAAM8D,GAChD,GAAsB,iBAAXA,GAIX,GAAsB,iBAAXA,EAAqB,CAC9B,QAAqB/K,IAAjB1O,EAAKyZ,IAAyBA,EAAOrC,WAAW,MAAmB,gBAAXqC,EAC1D,MAAM,IAAIU,UAAU,oBAAoBV,MAE1CzZ,EAAKyZ,IACP,OAREzZ,EAAKiiB,GAAGxI,EASZ,GACF,EAOFvD,GAAac,GAAGhc,SAAUklB,GAvSE,uCAuS2C,SAAUnL,GAC/E,MAAM7S,EAASsZ,GAAec,uBAAuB3G,MACrD,IAAKzT,IAAWA,EAAO8O,UAAU7W,SAASgmB,IACxC,OAEFpL,EAAMkD,iBACN,MAAM4L,EAAW5C,GAAShG,oBAAoB/Y,GACxC4hB,EAAanO,KAAKxE,aAAa,oBACrC,OAAI2S,GACFD,EAAS5B,GAAG6B,QACZD,EAAS7B,qBAGyC,SAAhDrJ,GAAYQ,iBAAiBxD,KAAM,UACrCkO,EAASrpB,YACTqpB,EAAS7B,sBAGX6B,EAAS/H,YACT+H,EAAS7B,oBACX,IACA9L,GAAac,GAAGzhB,OAAQ0qB,IAAuB,KAC7C,MAAM8D,EAAYvI,GAAe1T,KA5TR,6BA6TzB,IAAK,MAAM+b,KAAYE,EACrB9C,GAAShG,oBAAoB4I,EAC/B,IAOF/R,GAAmBmP,IAcnB,MAEM+C,GAAc,eAEdC,GAAe,OAAOD,KACtBE,GAAgB,QAAQF,KACxBG,GAAe,OAAOH,KACtBI,GAAiB,SAASJ,KAC1BK,GAAyB,QAAQL,cACjCM,GAAoB,OACpBC,GAAsB,WACtBC,GAAwB,aAExBC,GAA6B,WAAWF,OAAwBA,KAKhEG,GAAyB,8BACzBC,GAAY,CAChBvqB,OAAQ,KACRkjB,QAAQ,GAEJsH,GAAgB,CACpBxqB,OAAQ,iBACRkjB,OAAQ,WAOV,MAAMuH,WAAiBxK,GACrB,WAAAP,CAAY5kB,EAASukB,GACnBa,MAAMplB,EAASukB,GACf9D,KAAKmP,kBAAmB,EACxBnP,KAAKoP,cAAgB,GACrB,MAAMC,EAAaxJ,GAAe1T,KAAK4c,IACvC,IAAK,MAAMO,KAAQD,EAAY,CAC7B,MAAMtV,EAAW8L,GAAea,uBAAuB4I,GACjDC,EAAgB1J,GAAe1T,KAAK4H,GAAU5T,QAAOqpB,GAAgBA,IAAiBxP,KAAK4E,WAChF,OAAb7K,GAAqBwV,EAAc7e,QACrCsP,KAAKoP,cAAcxd,KAAK0d,EAE5B,CACAtP,KAAKyP,sBACAzP,KAAK6E,QAAQpgB,QAChBub,KAAK0P,0BAA0B1P,KAAKoP,cAAepP,KAAK2P,YAEtD3P,KAAK6E,QAAQ8C,QACf3H,KAAK2H,QAET,CAGA,kBAAWjE,GACT,OAAOsL,EACT,CACA,sBAAWrL,GACT,OAAOsL,EACT,CACA,eAAW1S,GACT,MA9DW,UA+Db,CAGA,MAAAoL,GACM3H,KAAK2P,WACP3P,KAAK4P,OAEL5P,KAAK6P,MAET,CACA,IAAAA,GACE,GAAI7P,KAAKmP,kBAAoBnP,KAAK2P,WAChC,OAEF,IAAIG,EAAiB,GAQrB,GALI9P,KAAK6E,QAAQpgB,SACfqrB,EAAiB9P,KAAK+P,uBAhEH,wCAgE4C5pB,QAAO5G,GAAWA,IAAYygB,KAAK4E,WAAU9hB,KAAIvD,GAAW2vB,GAAS5J,oBAAoB/lB,EAAS,CAC/JooB,QAAQ,OAGRmI,EAAepf,QAAUof,EAAe,GAAGX,iBAC7C,OAGF,GADmB5O,GAAaqB,QAAQ5B,KAAK4E,SAAU0J,IACxCtM,iBACb,OAEF,IAAK,MAAMgO,KAAkBF,EAC3BE,EAAeJ,OAEjB,MAAMK,EAAYjQ,KAAKkQ,gBACvBlQ,KAAK4E,SAASvJ,UAAU1B,OAAOiV,IAC/B5O,KAAK4E,SAASvJ,UAAU5E,IAAIoY,IAC5B7O,KAAK4E,SAAS7jB,MAAMkvB,GAAa,EACjCjQ,KAAK0P,0BAA0B1P,KAAKoP,eAAe,GACnDpP,KAAKmP,kBAAmB,EACxB,MAQMgB,EAAa,SADUF,EAAU,GAAGxL,cAAgBwL,EAAU7d,MAAM,KAE1E4N,KAAKmF,gBATY,KACfnF,KAAKmP,kBAAmB,EACxBnP,KAAK4E,SAASvJ,UAAU1B,OAAOkV,IAC/B7O,KAAK4E,SAASvJ,UAAU5E,IAAImY,GAAqBD,IACjD3O,KAAK4E,SAAS7jB,MAAMkvB,GAAa,GACjC1P,GAAaqB,QAAQ5B,KAAK4E,SAAU2J,GAAc,GAItBvO,KAAK4E,UAAU,GAC7C5E,KAAK4E,SAAS7jB,MAAMkvB,GAAa,GAAGjQ,KAAK4E,SAASuL,MACpD,CACA,IAAAP,GACE,GAAI5P,KAAKmP,mBAAqBnP,KAAK2P,WACjC,OAGF,GADmBpP,GAAaqB,QAAQ5B,KAAK4E,SAAU4J,IACxCxM,iBACb,OAEF,MAAMiO,EAAYjQ,KAAKkQ,gBACvBlQ,KAAK4E,SAAS7jB,MAAMkvB,GAAa,GAAGjQ,KAAK4E,SAASthB,wBAAwB2sB,OAC1EpU,GAAOmE,KAAK4E,UACZ5E,KAAK4E,SAASvJ,UAAU5E,IAAIoY,IAC5B7O,KAAK4E,SAASvJ,UAAU1B,OAAOiV,GAAqBD,IACpD,IAAK,MAAM/M,KAAW5B,KAAKoP,cAAe,CACxC,MAAM7vB,EAAUsmB,GAAec,uBAAuB/E,GAClDriB,IAAYygB,KAAK2P,SAASpwB,IAC5BygB,KAAK0P,0BAA0B,CAAC9N,IAAU,EAE9C,CACA5B,KAAKmP,kBAAmB,EAOxBnP,KAAK4E,SAAS7jB,MAAMkvB,GAAa,GACjCjQ,KAAKmF,gBAPY,KACfnF,KAAKmP,kBAAmB,EACxBnP,KAAK4E,SAASvJ,UAAU1B,OAAOkV,IAC/B7O,KAAK4E,SAASvJ,UAAU5E,IAAImY,IAC5BrO,GAAaqB,QAAQ5B,KAAK4E,SAAU6J,GAAe,GAGvBzO,KAAK4E,UAAU,EAC/C,CACA,QAAA+K,CAASpwB,EAAUygB,KAAK4E,UACtB,OAAOrlB,EAAQ8b,UAAU7W,SAASmqB,GACpC,CAGA,iBAAA3K,CAAkBF,GAGhB,OAFAA,EAAO6D,OAAS7G,QAAQgD,EAAO6D,QAC/B7D,EAAOrf,OAASiW,GAAWoJ,EAAOrf,QAC3Bqf,CACT,CACA,aAAAoM,GACE,OAAOlQ,KAAK4E,SAASvJ,UAAU7W,SA3IL,uBAChB,QACC,QA0Ib,CACA,mBAAAirB,GACE,IAAKzP,KAAK6E,QAAQpgB,OAChB,OAEF,MAAMshB,EAAW/F,KAAK+P,uBAAuBhB,IAC7C,IAAK,MAAMxvB,KAAWwmB,EAAU,CAC9B,MAAMqK,EAAWvK,GAAec,uBAAuBpnB,GACnD6wB,GACFpQ,KAAK0P,0BAA0B,CAACnwB,GAAUygB,KAAK2P,SAASS,GAE5D,CACF,CACA,sBAAAL,CAAuBhW,GACrB,MAAMgM,EAAWF,GAAe1T,KAAK2c,GAA4B9O,KAAK6E,QAAQpgB,QAE9E,OAAOohB,GAAe1T,KAAK4H,EAAUiG,KAAK6E,QAAQpgB,QAAQ0B,QAAO5G,IAAYwmB,EAAS3E,SAAS7hB,IACjG,CACA,yBAAAmwB,CAA0BW,EAAcC,GACtC,GAAKD,EAAa3f,OAGlB,IAAK,MAAMnR,KAAW8wB,EACpB9wB,EAAQ8b,UAAUsM,OArKK,aAqKyB2I,GAChD/wB,EAAQ6B,aAAa,gBAAiBkvB,EAE1C,CAGA,sBAAO7T,CAAgBqH,GACrB,MAAMe,EAAU,CAAC,EAIjB,MAHsB,iBAAXf,GAAuB,YAAYzgB,KAAKygB,KACjDe,EAAQ8C,QAAS,GAEZ3H,KAAKwH,MAAK,WACf,MAAMnd,EAAO6kB,GAAS5J,oBAAoBtF,KAAM6E,GAChD,GAAsB,iBAAXf,EAAqB,CAC9B,QAA4B,IAAjBzZ,EAAKyZ,GACd,MAAM,IAAIU,UAAU,oBAAoBV,MAE1CzZ,EAAKyZ,IACP,CACF,GACF,EAOFvD,GAAac,GAAGhc,SAAUqpB,GAAwBK,IAAwB,SAAU3P,IAErD,MAAzBA,EAAM7S,OAAO0a,SAAmB7H,EAAMW,gBAAmD,MAAjCX,EAAMW,eAAekH,UAC/E7H,EAAMkD,iBAER,IAAK,MAAM/iB,KAAWsmB,GAAee,gCAAgC5G,MACnEkP,GAAS5J,oBAAoB/lB,EAAS,CACpCooB,QAAQ,IACPA,QAEP,IAMAxL,GAAmB+S,IAcnB,MAAMqB,GAAS,WAETC,GAAc,eACdC,GAAiB,YAGjBC,GAAiB,UACjBC,GAAmB,YAGnBC,GAAe,OAAOJ,KACtBK,GAAiB,SAASL,KAC1BM,GAAe,OAAON,KACtBO,GAAgB,QAAQP,KACxBQ,GAAyB,QAAQR,KAAcC,KAC/CQ,GAAyB,UAAUT,KAAcC,KACjDS,GAAuB,QAAQV,KAAcC,KAC7CU,GAAoB,OAMpBC,GAAyB,4DACzBC,GAA6B,GAAGD,MAA0BD,KAC1DG,GAAgB,iBAIhBC,GAAgBtV,KAAU,UAAY,YACtCuV,GAAmBvV,KAAU,YAAc,UAC3CwV,GAAmBxV,KAAU,aAAe,eAC5CyV,GAAsBzV,KAAU,eAAiB,aACjD0V,GAAkB1V,KAAU,aAAe,cAC3C2V,GAAiB3V,KAAU,cAAgB,aAG3C4V,GAAY,CAChBC,WAAW,EACX7jB,SAAU,kBACV8jB,QAAS,UACT/pB,OAAQ,CAAC,EAAG,GACZgqB,aAAc,KACd1zB,UAAW,UAEP2zB,GAAgB,CACpBH,UAAW,mBACX7jB,SAAU,mBACV8jB,QAAS,SACT/pB,OAAQ,0BACRgqB,aAAc,yBACd1zB,UAAW,2BAOb,MAAM4zB,WAAiBxN,GACrB,WAAAP,CAAY5kB,EAASukB,GACnBa,MAAMplB,EAASukB,GACf9D,KAAKmS,QAAU,KACfnS,KAAKoS,QAAUpS,KAAK4E,SAAS7f,WAE7Bib,KAAKqS,MAAQxM,GAAehhB,KAAKmb,KAAK4E,SAAU0M,IAAe,IAAMzL,GAAeM,KAAKnG,KAAK4E,SAAU0M,IAAe,IAAMzL,GAAeC,QAAQwL,GAAetR,KAAKoS,SACxKpS,KAAKsS,UAAYtS,KAAKuS,eACxB,CAGA,kBAAW7O,GACT,OAAOmO,EACT,CACA,sBAAWlO,GACT,OAAOsO,EACT,CACA,eAAW1V,GACT,OAAOgU,EACT,CAGA,MAAA5I,GACE,OAAO3H,KAAK2P,WAAa3P,KAAK4P,OAAS5P,KAAK6P,MAC9C,CACA,IAAAA,GACE,GAAI3U,GAAW8E,KAAK4E,WAAa5E,KAAK2P,WACpC,OAEF,MAAM7P,EAAgB,CACpBA,cAAeE,KAAK4E,UAGtB,IADkBrE,GAAaqB,QAAQ5B,KAAK4E,SAAUkM,GAAchR,GACtDkC,iBAAd,CASA,GANAhC,KAAKwS,gBAMD,iBAAkBntB,SAASC,kBAAoB0a,KAAKoS,QAAQpX,QAzExC,eA0EtB,IAAK,MAAMzb,IAAW,GAAGZ,UAAU0G,SAAS6G,KAAK6Z,UAC/CxF,GAAac,GAAG9hB,EAAS,YAAaqc,IAG1CoE,KAAK4E,SAAS6N,QACdzS,KAAK4E,SAASxjB,aAAa,iBAAiB,GAC5C4e,KAAKqS,MAAMhX,UAAU5E,IAAI0a,IACzBnR,KAAK4E,SAASvJ,UAAU5E,IAAI0a,IAC5B5Q,GAAaqB,QAAQ5B,KAAK4E,SAAUmM,GAAejR,EAhBnD,CAiBF,CACA,IAAA8P,GACE,GAAI1U,GAAW8E,KAAK4E,YAAc5E,KAAK2P,WACrC,OAEF,MAAM7P,EAAgB,CACpBA,cAAeE,KAAK4E,UAEtB5E,KAAK0S,cAAc5S,EACrB,CACA,OAAAiF,GACM/E,KAAKmS,SACPnS,KAAKmS,QAAQnZ,UAEf2L,MAAMI,SACR,CACA,MAAAha,GACEiV,KAAKsS,UAAYtS,KAAKuS,gBAClBvS,KAAKmS,SACPnS,KAAKmS,QAAQpnB,QAEjB,CAGA,aAAA2nB,CAAc5S,GAEZ,IADkBS,GAAaqB,QAAQ5B,KAAK4E,SAAUgM,GAAc9Q,GACtDkC,iBAAd,CAMA,GAAI,iBAAkB3c,SAASC,gBAC7B,IAAK,MAAM/F,IAAW,GAAGZ,UAAU0G,SAAS6G,KAAK6Z,UAC/CxF,GAAaC,IAAIjhB,EAAS,YAAaqc,IAGvCoE,KAAKmS,SACPnS,KAAKmS,QAAQnZ,UAEfgH,KAAKqS,MAAMhX,UAAU1B,OAAOwX,IAC5BnR,KAAK4E,SAASvJ,UAAU1B,OAAOwX,IAC/BnR,KAAK4E,SAASxjB,aAAa,gBAAiB,SAC5C4hB,GAAYE,oBAAoBlD,KAAKqS,MAAO,UAC5C9R,GAAaqB,QAAQ5B,KAAK4E,SAAUiM,GAAgB/Q,EAhBpD,CAiBF,CACA,UAAA+D,CAAWC,GAET,GAAgC,iBADhCA,EAASa,MAAMd,WAAWC,IACRxlB,YAA2B,GAAUwlB,EAAOxlB,YAAgE,mBAA3CwlB,EAAOxlB,UAAUgF,sBAElG,MAAM,IAAIkhB,UAAU,GAAG+L,GAAO9L,+GAEhC,OAAOX,CACT,CACA,aAAA0O,GACE,QAAsB,IAAX,EACT,MAAM,IAAIhO,UAAU,gEAEtB,IAAImO,EAAmB3S,KAAK4E,SACG,WAA3B5E,KAAK6E,QAAQvmB,UACfq0B,EAAmB3S,KAAKoS,QACf,GAAUpS,KAAK6E,QAAQvmB,WAChCq0B,EAAmBjY,GAAWsF,KAAK6E,QAAQvmB,WACA,iBAA3B0hB,KAAK6E,QAAQvmB,YAC7Bq0B,EAAmB3S,KAAK6E,QAAQvmB,WAElC,MAAM0zB,EAAehS,KAAK4S,mBAC1B5S,KAAKmS,QAAU,GAAoBQ,EAAkB3S,KAAKqS,MAAOL,EACnE,CACA,QAAArC,GACE,OAAO3P,KAAKqS,MAAMhX,UAAU7W,SAAS2sB,GACvC,CACA,aAAA0B,GACE,MAAMC,EAAiB9S,KAAKoS,QAC5B,GAAIU,EAAezX,UAAU7W,SArKN,WAsKrB,OAAOmtB,GAET,GAAImB,EAAezX,UAAU7W,SAvKJ,aAwKvB,OAAOotB,GAET,GAAIkB,EAAezX,UAAU7W,SAzKA,iBA0K3B,MA5JsB,MA8JxB,GAAIsuB,EAAezX,UAAU7W,SA3KE,mBA4K7B,MA9JyB,SAkK3B,MAAMuuB,EAAkF,QAA1E9tB,iBAAiB+a,KAAKqS,OAAOvX,iBAAiB,iBAAiB6K,OAC7E,OAAImN,EAAezX,UAAU7W,SArLP,UAsLbuuB,EAAQvB,GAAmBD,GAE7BwB,EAAQrB,GAAsBD,EACvC,CACA,aAAAc,GACE,OAAkD,OAA3CvS,KAAK4E,SAAS5J,QAnLD,UAoLtB,CACA,UAAAgY,GACE,MAAM,OACJhrB,GACEgY,KAAK6E,QACT,MAAsB,iBAAX7c,EACFA,EAAO9F,MAAM,KAAKY,KAAInF,GAAS4f,OAAOgQ,SAAS5vB,EAAO,MAEzC,mBAAXqK,EACFirB,GAAcjrB,EAAOirB,EAAYjT,KAAK4E,UAExC5c,CACT,CACA,gBAAA4qB,GACE,MAAMM,EAAwB,CAC5Bx0B,UAAWshB,KAAK6S,gBAChBzc,UAAW,CAAC,CACV9V,KAAM,kBACNmB,QAAS,CACPwM,SAAU+R,KAAK6E,QAAQ5W,WAExB,CACD3N,KAAM,SACNmB,QAAS,CACPuG,OAAQgY,KAAKgT,iBAanB,OAPIhT,KAAKsS,WAAsC,WAAzBtS,KAAK6E,QAAQkN,WACjC/O,GAAYC,iBAAiBjD,KAAKqS,MAAO,SAAU,UACnDa,EAAsB9c,UAAY,CAAC,CACjC9V,KAAM,cACNC,SAAS,KAGN,IACF2yB,KACArW,GAAQmD,KAAK6E,QAAQmN,aAAc,CAACkB,IAE3C,CACA,eAAAC,EAAgB,IACdr2B,EAAG,OACHyP,IAEA,MAAMggB,EAAQ1G,GAAe1T,KAhOF,8DAgO+B6N,KAAKqS,OAAOlsB,QAAO5G,GAAWob,GAAUpb,KAC7FgtB,EAAM7b,QAMXoN,GAAqByO,EAAOhgB,EAAQzP,IAAQ6zB,IAAmBpE,EAAMnL,SAAS7U,IAASkmB,OACzF,CAGA,sBAAOhW,CAAgBqH,GACrB,OAAO9D,KAAKwH,MAAK,WACf,MAAMnd,EAAO6nB,GAAS5M,oBAAoBtF,KAAM8D,GAChD,GAAsB,iBAAXA,EAAX,CAGA,QAA4B,IAAjBzZ,EAAKyZ,GACd,MAAM,IAAIU,UAAU,oBAAoBV,MAE1CzZ,EAAKyZ,IAJL,CAKF,GACF,CACA,iBAAOsP,CAAWhU,GAChB,GA5QuB,IA4QnBA,EAAMwI,QAAgD,UAAfxI,EAAMqB,MA/QnC,QA+QuDrB,EAAMtiB,IACzE,OAEF,MAAMu2B,EAAcxN,GAAe1T,KAAKkf,IACxC,IAAK,MAAM1J,KAAU0L,EAAa,CAChC,MAAMC,EAAUpB,GAAS7M,YAAYsC,GACrC,IAAK2L,IAAyC,IAA9BA,EAAQzO,QAAQiN,UAC9B,SAEF,MAAMyB,EAAenU,EAAMmU,eACrBC,EAAeD,EAAanS,SAASkS,EAAQjB,OACnD,GAAIkB,EAAanS,SAASkS,EAAQ1O,WAA2C,WAA9B0O,EAAQzO,QAAQiN,YAA2B0B,GAA8C,YAA9BF,EAAQzO,QAAQiN,WAA2B0B,EACnJ,SAIF,GAAIF,EAAQjB,MAAM7tB,SAAS4a,EAAM7S,UAA2B,UAAf6S,EAAMqB,MA/RvC,QA+R2DrB,EAAMtiB,KAAqB,qCAAqCuG,KAAK+b,EAAM7S,OAAO0a,UACvJ,SAEF,MAAMnH,EAAgB,CACpBA,cAAewT,EAAQ1O,UAEN,UAAfxF,EAAMqB,OACRX,EAAckH,WAAa5H,GAE7BkU,EAAQZ,cAAc5S,EACxB,CACF,CACA,4BAAO2T,CAAsBrU,GAI3B,MAAMsU,EAAU,kBAAkBrwB,KAAK+b,EAAM7S,OAAO0a,SAC9C0M,EAjTW,WAiTKvU,EAAMtiB,IACtB82B,EAAkB,CAAClD,GAAgBC,IAAkBvP,SAAShC,EAAMtiB,KAC1E,IAAK82B,IAAoBD,EACvB,OAEF,GAAID,IAAYC,EACd,OAEFvU,EAAMkD,iBAGN,MAAMuR,EAAkB7T,KAAKgG,QAAQoL,IAA0BpR,KAAO6F,GAAeM,KAAKnG,KAAMoR,IAAwB,IAAMvL,GAAehhB,KAAKmb,KAAMoR,IAAwB,IAAMvL,GAAeC,QAAQsL,GAAwBhS,EAAMW,eAAehb,YACpPwF,EAAW2nB,GAAS5M,oBAAoBuO,GAC9C,GAAID,EAIF,OAHAxU,EAAM0U,kBACNvpB,EAASslB,YACTtlB,EAAS4oB,gBAAgB/T,GAGvB7U,EAASolB,aAEXvQ,EAAM0U,kBACNvpB,EAASqlB,OACTiE,EAAgBpB,QAEpB,EAOFlS,GAAac,GAAGhc,SAAU4rB,GAAwBG,GAAwBc,GAASuB,uBACnFlT,GAAac,GAAGhc,SAAU4rB,GAAwBK,GAAeY,GAASuB,uBAC1ElT,GAAac,GAAGhc,SAAU2rB,GAAwBkB,GAASkB,YAC3D7S,GAAac,GAAGhc,SAAU6rB,GAAsBgB,GAASkB,YACzD7S,GAAac,GAAGhc,SAAU2rB,GAAwBI,IAAwB,SAAUhS,GAClFA,EAAMkD,iBACN4P,GAAS5M,oBAAoBtF,MAAM2H,QACrC,IAMAxL,GAAmB+V,IAcnB,MAAM6B,GAAS,WAETC,GAAoB,OACpBC,GAAkB,gBAAgBF,KAClCG,GAAY,CAChBC,UAAW,iBACXC,cAAe,KACfhP,YAAY,EACZzK,WAAW,EAEX0Z,YAAa,QAETC,GAAgB,CACpBH,UAAW,SACXC,cAAe,kBACfhP,WAAY,UACZzK,UAAW,UACX0Z,YAAa,oBAOf,MAAME,WAAiB9Q,GACrB,WAAAU,CAAYL,GACVa,QACA3E,KAAK6E,QAAU7E,KAAK6D,WAAWC,GAC/B9D,KAAKwU,aAAc,EACnBxU,KAAK4E,SAAW,IAClB,CAGA,kBAAWlB,GACT,OAAOwQ,EACT,CACA,sBAAWvQ,GACT,OAAO2Q,EACT,CACA,eAAW/X,GACT,OAAOwX,EACT,CAGA,IAAAlE,CAAKxT,GACH,IAAK2D,KAAK6E,QAAQlK,UAEhB,YADAkC,GAAQR,GAGV2D,KAAKyU,UACL,MAAMl1B,EAAUygB,KAAK0U,cACjB1U,KAAK6E,QAAQO,YACfvJ,GAAOtc,GAETA,EAAQ8b,UAAU5E,IAAIud,IACtBhU,KAAK2U,mBAAkB,KACrB9X,GAAQR,EAAS,GAErB,CACA,IAAAuT,CAAKvT,GACE2D,KAAK6E,QAAQlK,WAIlBqF,KAAK0U,cAAcrZ,UAAU1B,OAAOqa,IACpChU,KAAK2U,mBAAkB,KACrB3U,KAAK+E,UACLlI,GAAQR,EAAS,KANjBQ,GAAQR,EAQZ,CACA,OAAA0I,GACO/E,KAAKwU,cAGVjU,GAAaC,IAAIR,KAAK4E,SAAUqP,IAChCjU,KAAK4E,SAASjL,SACdqG,KAAKwU,aAAc,EACrB,CAGA,WAAAE,GACE,IAAK1U,KAAK4E,SAAU,CAClB,MAAMgQ,EAAWvvB,SAASwvB,cAAc,OACxCD,EAAST,UAAYnU,KAAK6E,QAAQsP,UAC9BnU,KAAK6E,QAAQO,YACfwP,EAASvZ,UAAU5E,IApFD,QAsFpBuJ,KAAK4E,SAAWgQ,CAClB,CACA,OAAO5U,KAAK4E,QACd,CACA,iBAAAZ,CAAkBF,GAGhB,OADAA,EAAOuQ,YAAc3Z,GAAWoJ,EAAOuQ,aAChCvQ,CACT,CACA,OAAA2Q,GACE,GAAIzU,KAAKwU,YACP,OAEF,MAAMj1B,EAAUygB,KAAK0U,cACrB1U,KAAK6E,QAAQwP,YAAYS,OAAOv1B,GAChCghB,GAAac,GAAG9hB,EAAS00B,IAAiB,KACxCpX,GAAQmD,KAAK6E,QAAQuP,cAAc,IAErCpU,KAAKwU,aAAc,CACrB,CACA,iBAAAG,CAAkBtY,GAChBW,GAAuBX,EAAU2D,KAAK0U,cAAe1U,KAAK6E,QAAQO,WACpE,EAeF,MAEM2P,GAAc,gBACdC,GAAkB,UAAUD,KAC5BE,GAAoB,cAAcF,KAGlCG,GAAmB,WACnBC,GAAY,CAChBC,WAAW,EACXC,YAAa,MAETC,GAAgB,CACpBF,UAAW,UACXC,YAAa,WAOf,MAAME,WAAkB9R,GACtB,WAAAU,CAAYL,GACVa,QACA3E,KAAK6E,QAAU7E,KAAK6D,WAAWC,GAC/B9D,KAAKwV,WAAY,EACjBxV,KAAKyV,qBAAuB,IAC9B,CAGA,kBAAW/R,GACT,OAAOyR,EACT,CACA,sBAAWxR,GACT,OAAO2R,EACT,CACA,eAAW/Y,GACT,MArCW,WAsCb,CAGA,QAAAmZ,GACM1V,KAAKwV,YAGLxV,KAAK6E,QAAQuQ,WACfpV,KAAK6E,QAAQwQ,YAAY5C,QAE3BlS,GAAaC,IAAInb,SAAU0vB,IAC3BxU,GAAac,GAAGhc,SAAU2vB,IAAiB5V,GAASY,KAAK2V,eAAevW,KACxEmB,GAAac,GAAGhc,SAAU4vB,IAAmB7V,GAASY,KAAK4V,eAAexW,KAC1EY,KAAKwV,WAAY,EACnB,CACA,UAAAK,GACO7V,KAAKwV,YAGVxV,KAAKwV,WAAY,EACjBjV,GAAaC,IAAInb,SAAU0vB,IAC7B,CAGA,cAAAY,CAAevW,GACb,MAAM,YACJiW,GACErV,KAAK6E,QACT,GAAIzF,EAAM7S,SAAWlH,UAAY+Z,EAAM7S,SAAW8oB,GAAeA,EAAY7wB,SAAS4a,EAAM7S,QAC1F,OAEF,MAAM1L,EAAWglB,GAAeU,kBAAkB8O,GAC1B,IAApBx0B,EAAS6P,OACX2kB,EAAY5C,QACHzS,KAAKyV,uBAAyBP,GACvCr0B,EAASA,EAAS6P,OAAS,GAAG+hB,QAE9B5xB,EAAS,GAAG4xB,OAEhB,CACA,cAAAmD,CAAexW,GAzED,QA0ERA,EAAMtiB,MAGVkjB,KAAKyV,qBAAuBrW,EAAM0W,SAAWZ,GA5EzB,UA6EtB,EAeF,MAAMa,GAAyB,oDACzBC,GAA0B,cAC1BC,GAAmB,gBACnBC,GAAkB,eAMxB,MAAMC,GACJ,WAAAhS,GACEnE,KAAK4E,SAAWvf,SAAS6G,IAC3B,CAGA,QAAAkqB,GAEE,MAAMC,EAAgBhxB,SAASC,gBAAgBuC,YAC/C,OAAO1F,KAAKoC,IAAI3E,OAAO02B,WAAaD,EACtC,CACA,IAAAzG,GACE,MAAM/rB,EAAQmc,KAAKoW,WACnBpW,KAAKuW,mBAELvW,KAAKwW,sBAAsBxW,KAAK4E,SAAUqR,IAAkBQ,GAAmBA,EAAkB5yB,IAEjGmc,KAAKwW,sBAAsBT,GAAwBE,IAAkBQ,GAAmBA,EAAkB5yB,IAC1Gmc,KAAKwW,sBAAsBR,GAAyBE,IAAiBO,GAAmBA,EAAkB5yB,GAC5G,CACA,KAAAwO,GACE2N,KAAK0W,wBAAwB1W,KAAK4E,SAAU,YAC5C5E,KAAK0W,wBAAwB1W,KAAK4E,SAAUqR,IAC5CjW,KAAK0W,wBAAwBX,GAAwBE,IACrDjW,KAAK0W,wBAAwBV,GAAyBE,GACxD,CACA,aAAAS,GACE,OAAO3W,KAAKoW,WAAa,CAC3B,CAGA,gBAAAG,GACEvW,KAAK4W,sBAAsB5W,KAAK4E,SAAU,YAC1C5E,KAAK4E,SAAS7jB,MAAM+K,SAAW,QACjC,CACA,qBAAA0qB,CAAsBzc,EAAU8c,EAAexa,GAC7C,MAAMya,EAAiB9W,KAAKoW,WAS5BpW,KAAK+W,2BAA2Bhd,GARHxa,IAC3B,GAAIA,IAAYygB,KAAK4E,UAAYhlB,OAAO02B,WAAa/2B,EAAQsI,YAAcivB,EACzE,OAEF9W,KAAK4W,sBAAsBr3B,EAASs3B,GACpC,MAAMJ,EAAkB72B,OAAOqF,iBAAiB1F,GAASub,iBAAiB+b,GAC1Et3B,EAAQwB,MAAMi2B,YAAYH,EAAe,GAAGxa,EAASkB,OAAOC,WAAWiZ,QAAsB,GAGjG,CACA,qBAAAG,CAAsBr3B,EAASs3B,GAC7B,MAAMI,EAAc13B,EAAQwB,MAAM+Z,iBAAiB+b,GAC/CI,GACFjU,GAAYC,iBAAiB1jB,EAASs3B,EAAeI,EAEzD,CACA,uBAAAP,CAAwB3c,EAAU8c,GAWhC7W,KAAK+W,2BAA2Bhd,GAVHxa,IAC3B,MAAM5B,EAAQqlB,GAAYQ,iBAAiBjkB,EAASs3B,GAEtC,OAAVl5B,GAIJqlB,GAAYE,oBAAoB3jB,EAASs3B,GACzCt3B,EAAQwB,MAAMi2B,YAAYH,EAAel5B,IAJvC4B,EAAQwB,MAAMm2B,eAAeL,EAIgB,GAGnD,CACA,0BAAAE,CAA2Bhd,EAAUod,GACnC,GAAI,GAAUpd,GACZod,EAASpd,QAGX,IAAK,MAAM6L,KAAOC,GAAe1T,KAAK4H,EAAUiG,KAAK4E,UACnDuS,EAASvR,EAEb,EAeF,MAEMwR,GAAc,YAGdC,GAAe,OAAOD,KACtBE,GAAyB,gBAAgBF,KACzCG,GAAiB,SAASH,KAC1BI,GAAe,OAAOJ,KACtBK,GAAgB,QAAQL,KACxBM,GAAiB,SAASN,KAC1BO,GAAsB,gBAAgBP,KACtCQ,GAA0B,oBAAoBR,KAC9CS,GAA0B,kBAAkBT,KAC5CU,GAAyB,QAAQV,cACjCW,GAAkB,aAElBC,GAAoB,OACpBC,GAAoB,eAKpBC,GAAY,CAChBtD,UAAU,EACVnC,OAAO,EACPzH,UAAU,GAENmN,GAAgB,CACpBvD,SAAU,mBACVnC,MAAO,UACPzH,SAAU,WAOZ,MAAMoN,WAAc1T,GAClB,WAAAP,CAAY5kB,EAASukB,GACnBa,MAAMplB,EAASukB,GACf9D,KAAKqY,QAAUxS,GAAeC,QArBV,gBAqBmC9F,KAAK4E,UAC5D5E,KAAKsY,UAAYtY,KAAKuY,sBACtBvY,KAAKwY,WAAaxY,KAAKyY,uBACvBzY,KAAK2P,UAAW,EAChB3P,KAAKmP,kBAAmB,EACxBnP,KAAK0Y,WAAa,IAAIvC,GACtBnW,KAAK6L,oBACP,CAGA,kBAAWnI,GACT,OAAOwU,EACT,CACA,sBAAWvU,GACT,OAAOwU,EACT,CACA,eAAW5b,GACT,MA1DW,OA2Db,CAGA,MAAAoL,CAAO7H,GACL,OAAOE,KAAK2P,SAAW3P,KAAK4P,OAAS5P,KAAK6P,KAAK/P,EACjD,CACA,IAAA+P,CAAK/P,GACCE,KAAK2P,UAAY3P,KAAKmP,kBAGR5O,GAAaqB,QAAQ5B,KAAK4E,SAAU4S,GAAc,CAClE1X,kBAEYkC,mBAGdhC,KAAK2P,UAAW,EAChB3P,KAAKmP,kBAAmB,EACxBnP,KAAK0Y,WAAW9I,OAChBvqB,SAAS6G,KAAKmP,UAAU5E,IAAIshB,IAC5B/X,KAAK2Y,gBACL3Y,KAAKsY,UAAUzI,MAAK,IAAM7P,KAAK4Y,aAAa9Y,KAC9C,CACA,IAAA8P,GACO5P,KAAK2P,WAAY3P,KAAKmP,mBAGT5O,GAAaqB,QAAQ5B,KAAK4E,SAAUyS,IACxCrV,mBAGdhC,KAAK2P,UAAW,EAChB3P,KAAKmP,kBAAmB,EACxBnP,KAAKwY,WAAW3C,aAChB7V,KAAK4E,SAASvJ,UAAU1B,OAAOqe,IAC/BhY,KAAKmF,gBAAe,IAAMnF,KAAK6Y,cAAc7Y,KAAK4E,SAAU5E,KAAKgO,gBACnE,CACA,OAAAjJ,GACExE,GAAaC,IAAI5gB,OAAQw3B,IACzB7W,GAAaC,IAAIR,KAAKqY,QAASjB,IAC/BpX,KAAKsY,UAAUvT,UACf/E,KAAKwY,WAAW3C,aAChBlR,MAAMI,SACR,CACA,YAAA+T,GACE9Y,KAAK2Y,eACP,CAGA,mBAAAJ,GACE,OAAO,IAAIhE,GAAS,CAClB5Z,UAAWmG,QAAQd,KAAK6E,QAAQ+P,UAEhCxP,WAAYpF,KAAKgO,eAErB,CACA,oBAAAyK,GACE,OAAO,IAAIlD,GAAU,CACnBF,YAAarV,KAAK4E,UAEtB,CACA,YAAAgU,CAAa9Y,GAENza,SAAS6G,KAAK1H,SAASwb,KAAK4E,WAC/Bvf,SAAS6G,KAAK4oB,OAAO9U,KAAK4E,UAE5B5E,KAAK4E,SAAS7jB,MAAMgxB,QAAU,QAC9B/R,KAAK4E,SAASzjB,gBAAgB,eAC9B6e,KAAK4E,SAASxjB,aAAa,cAAc,GACzC4e,KAAK4E,SAASxjB,aAAa,OAAQ,UACnC4e,KAAK4E,SAASnZ,UAAY,EAC1B,MAAMstB,EAAYlT,GAAeC,QA7GT,cA6GsC9F,KAAKqY,SAC/DU,IACFA,EAAUttB,UAAY,GAExBoQ,GAAOmE,KAAK4E,UACZ5E,KAAK4E,SAASvJ,UAAU5E,IAAIuhB,IAU5BhY,KAAKmF,gBATsB,KACrBnF,KAAK6E,QAAQ4N,OACfzS,KAAKwY,WAAW9C,WAElB1V,KAAKmP,kBAAmB,EACxB5O,GAAaqB,QAAQ5B,KAAK4E,SAAU6S,GAAe,CACjD3X,iBACA,GAEoCE,KAAKqY,QAASrY,KAAKgO,cAC7D,CACA,kBAAAnC,GACEtL,GAAac,GAAGrB,KAAK4E,SAAUiT,IAAyBzY,IAhJvC,WAiJXA,EAAMtiB,MAGNkjB,KAAK6E,QAAQmG,SACfhL,KAAK4P,OAGP5P,KAAKgZ,6BAA4B,IAEnCzY,GAAac,GAAGzhB,OAAQ83B,IAAgB,KAClC1X,KAAK2P,WAAa3P,KAAKmP,kBACzBnP,KAAK2Y,eACP,IAEFpY,GAAac,GAAGrB,KAAK4E,SAAUgT,IAAyBxY,IAEtDmB,GAAae,IAAItB,KAAK4E,SAAU+S,IAAqBsB,IAC/CjZ,KAAK4E,WAAaxF,EAAM7S,QAAUyT,KAAK4E,WAAaqU,EAAO1sB,SAGjC,WAA1ByT,KAAK6E,QAAQ+P,SAIb5U,KAAK6E,QAAQ+P,UACf5U,KAAK4P,OAJL5P,KAAKgZ,6BAKP,GACA,GAEN,CACA,UAAAH,GACE7Y,KAAK4E,SAAS7jB,MAAMgxB,QAAU,OAC9B/R,KAAK4E,SAASxjB,aAAa,eAAe,GAC1C4e,KAAK4E,SAASzjB,gBAAgB,cAC9B6e,KAAK4E,SAASzjB,gBAAgB,QAC9B6e,KAAKmP,kBAAmB,EACxBnP,KAAKsY,UAAU1I,MAAK,KAClBvqB,SAAS6G,KAAKmP,UAAU1B,OAAOoe,IAC/B/X,KAAKkZ,oBACLlZ,KAAK0Y,WAAWrmB,QAChBkO,GAAaqB,QAAQ5B,KAAK4E,SAAU2S,GAAe,GAEvD,CACA,WAAAvJ,GACE,OAAOhO,KAAK4E,SAASvJ,UAAU7W,SAjLT,OAkLxB,CACA,0BAAAw0B,GAEE,GADkBzY,GAAaqB,QAAQ5B,KAAK4E,SAAU0S,IACxCtV,iBACZ,OAEF,MAAMmX,EAAqBnZ,KAAK4E,SAASvX,aAAehI,SAASC,gBAAgBsC,aAC3EwxB,EAAmBpZ,KAAK4E,SAAS7jB,MAAMiL,UAEpB,WAArBotB,GAAiCpZ,KAAK4E,SAASvJ,UAAU7W,SAASyzB,MAGjEkB,IACHnZ,KAAK4E,SAAS7jB,MAAMiL,UAAY,UAElCgU,KAAK4E,SAASvJ,UAAU5E,IAAIwhB,IAC5BjY,KAAKmF,gBAAe,KAClBnF,KAAK4E,SAASvJ,UAAU1B,OAAOse,IAC/BjY,KAAKmF,gBAAe,KAClBnF,KAAK4E,SAAS7jB,MAAMiL,UAAYotB,CAAgB,GAC/CpZ,KAAKqY,QAAQ,GACfrY,KAAKqY,SACRrY,KAAK4E,SAAS6N,QAChB,CAMA,aAAAkG,GACE,MAAMQ,EAAqBnZ,KAAK4E,SAASvX,aAAehI,SAASC,gBAAgBsC,aAC3EkvB,EAAiB9W,KAAK0Y,WAAWtC,WACjCiD,EAAoBvC,EAAiB,EAC3C,GAAIuC,IAAsBF,EAAoB,CAC5C,MAAMr3B,EAAWma,KAAU,cAAgB,eAC3C+D,KAAK4E,SAAS7jB,MAAMe,GAAY,GAAGg1B,KACrC,CACA,IAAKuC,GAAqBF,EAAoB,CAC5C,MAAMr3B,EAAWma,KAAU,eAAiB,cAC5C+D,KAAK4E,SAAS7jB,MAAMe,GAAY,GAAGg1B,KACrC,CACF,CACA,iBAAAoC,GACElZ,KAAK4E,SAAS7jB,MAAMu4B,YAAc,GAClCtZ,KAAK4E,SAAS7jB,MAAMw4B,aAAe,EACrC,CAGA,sBAAO9c,CAAgBqH,EAAQhE,GAC7B,OAAOE,KAAKwH,MAAK,WACf,MAAMnd,EAAO+tB,GAAM9S,oBAAoBtF,KAAM8D,GAC7C,GAAsB,iBAAXA,EAAX,CAGA,QAA4B,IAAjBzZ,EAAKyZ,GACd,MAAM,IAAIU,UAAU,oBAAoBV,MAE1CzZ,EAAKyZ,GAAQhE,EAJb,CAKF,GACF,EAOFS,GAAac,GAAGhc,SAAUyyB,GA9OK,4BA8O2C,SAAU1Y,GAClF,MAAM7S,EAASsZ,GAAec,uBAAuB3G,MACjD,CAAC,IAAK,QAAQoB,SAASpB,KAAKiH,UAC9B7H,EAAMkD,iBAER/B,GAAae,IAAI/U,EAAQirB,IAAcgC,IACjCA,EAAUxX,kBAIdzB,GAAae,IAAI/U,EAAQgrB,IAAgB,KACnC5c,GAAUqF,OACZA,KAAKyS,OACP,GACA,IAIJ,MAAMgH,EAAc5T,GAAeC,QAnQb,eAoQlB2T,GACFrB,GAAM/S,YAAYoU,GAAa7J,OAEpBwI,GAAM9S,oBAAoB/Y,GAClCob,OAAO3H,KACd,IACA6G,GAAqBuR,IAMrBjc,GAAmBic,IAcnB,MAEMsB,GAAc,gBACdC,GAAiB,YACjBC,GAAwB,OAAOF,KAAcC,KAE7CE,GAAoB,OACpBC,GAAuB,UACvBC,GAAoB,SAEpBC,GAAgB,kBAChBC,GAAe,OAAOP,KACtBQ,GAAgB,QAAQR,KACxBS,GAAe,OAAOT,KACtBU,GAAuB,gBAAgBV,KACvCW,GAAiB,SAASX,KAC1BY,GAAe,SAASZ,KACxBa,GAAyB,QAAQb,KAAcC,KAC/Ca,GAAwB,kBAAkBd,KAE1Ce,GAAY,CAChB7F,UAAU,EACV5J,UAAU,EACVvgB,QAAQ,GAEJiwB,GAAgB,CACpB9F,SAAU,mBACV5J,SAAU,UACVvgB,OAAQ,WAOV,MAAMkwB,WAAkBjW,GACtB,WAAAP,CAAY5kB,EAASukB,GACnBa,MAAMplB,EAASukB,GACf9D,KAAK2P,UAAW,EAChB3P,KAAKsY,UAAYtY,KAAKuY,sBACtBvY,KAAKwY,WAAaxY,KAAKyY,uBACvBzY,KAAK6L,oBACP,CAGA,kBAAWnI,GACT,OAAO+W,EACT,CACA,sBAAW9W,GACT,OAAO+W,EACT,CACA,eAAWne,GACT,MApDW,WAqDb,CAGA,MAAAoL,CAAO7H,GACL,OAAOE,KAAK2P,SAAW3P,KAAK4P,OAAS5P,KAAK6P,KAAK/P,EACjD,CACA,IAAA+P,CAAK/P,GACCE,KAAK2P,UAGSpP,GAAaqB,QAAQ5B,KAAK4E,SAAUqV,GAAc,CAClEna,kBAEYkC,mBAGdhC,KAAK2P,UAAW,EAChB3P,KAAKsY,UAAUzI,OACV7P,KAAK6E,QAAQpa,SAChB,IAAI0rB,IAAkBvG,OAExB5P,KAAK4E,SAASxjB,aAAa,cAAc,GACzC4e,KAAK4E,SAASxjB,aAAa,OAAQ,UACnC4e,KAAK4E,SAASvJ,UAAU5E,IAAIqjB,IAW5B9Z,KAAKmF,gBAVoB,KAClBnF,KAAK6E,QAAQpa,SAAUuV,KAAK6E,QAAQ+P,UACvC5U,KAAKwY,WAAW9C,WAElB1V,KAAK4E,SAASvJ,UAAU5E,IAAIojB,IAC5B7Z,KAAK4E,SAASvJ,UAAU1B,OAAOmgB,IAC/BvZ,GAAaqB,QAAQ5B,KAAK4E,SAAUsV,GAAe,CACjDpa,iBACA,GAEkCE,KAAK4E,UAAU,GACvD,CACA,IAAAgL,GACO5P,KAAK2P,WAGQpP,GAAaqB,QAAQ5B,KAAK4E,SAAUuV,IACxCnY,mBAGdhC,KAAKwY,WAAW3C,aAChB7V,KAAK4E,SAASgW,OACd5a,KAAK2P,UAAW,EAChB3P,KAAK4E,SAASvJ,UAAU5E,IAAIsjB,IAC5B/Z,KAAKsY,UAAU1I,OAUf5P,KAAKmF,gBAToB,KACvBnF,KAAK4E,SAASvJ,UAAU1B,OAAOkgB,GAAmBE,IAClD/Z,KAAK4E,SAASzjB,gBAAgB,cAC9B6e,KAAK4E,SAASzjB,gBAAgB,QACzB6e,KAAK6E,QAAQpa,SAChB,IAAI0rB,IAAkB9jB,QAExBkO,GAAaqB,QAAQ5B,KAAK4E,SAAUyV,GAAe,GAEfra,KAAK4E,UAAU,IACvD,CACA,OAAAG,GACE/E,KAAKsY,UAAUvT,UACf/E,KAAKwY,WAAW3C,aAChBlR,MAAMI,SACR,CAGA,mBAAAwT,GACE,MASM5d,EAAYmG,QAAQd,KAAK6E,QAAQ+P,UACvC,OAAO,IAAIL,GAAS,CAClBJ,UA3HsB,qBA4HtBxZ,YACAyK,YAAY,EACZiP,YAAarU,KAAK4E,SAAS7f,WAC3BqvB,cAAezZ,EAfK,KACU,WAA1BqF,KAAK6E,QAAQ+P,SAIjB5U,KAAK4P,OAHHrP,GAAaqB,QAAQ5B,KAAK4E,SAAUwV,GAG3B,EAUgC,MAE/C,CACA,oBAAA3B,GACE,OAAO,IAAIlD,GAAU,CACnBF,YAAarV,KAAK4E,UAEtB,CACA,kBAAAiH,GACEtL,GAAac,GAAGrB,KAAK4E,SAAU4V,IAAuBpb,IA5IvC,WA6ITA,EAAMtiB,MAGNkjB,KAAK6E,QAAQmG,SACfhL,KAAK4P,OAGPrP,GAAaqB,QAAQ5B,KAAK4E,SAAUwV,IAAqB,GAE7D,CAGA,sBAAO3d,CAAgBqH,GACrB,OAAO9D,KAAKwH,MAAK,WACf,MAAMnd,EAAOswB,GAAUrV,oBAAoBtF,KAAM8D,GACjD,GAAsB,iBAAXA,EAAX,CAGA,QAAqB/K,IAAjB1O,EAAKyZ,IAAyBA,EAAOrC,WAAW,MAAmB,gBAAXqC,EAC1D,MAAM,IAAIU,UAAU,oBAAoBV,MAE1CzZ,EAAKyZ,GAAQ9D,KAJb,CAKF,GACF,EAOFO,GAAac,GAAGhc,SAAUk1B,GA7JK,gCA6J2C,SAAUnb,GAClF,MAAM7S,EAASsZ,GAAec,uBAAuB3G,MAIrD,GAHI,CAAC,IAAK,QAAQoB,SAASpB,KAAKiH,UAC9B7H,EAAMkD,iBAEJpH,GAAW8E,MACb,OAEFO,GAAae,IAAI/U,EAAQ8tB,IAAgB,KAEnC1f,GAAUqF,OACZA,KAAKyS,OACP,IAIF,MAAMgH,EAAc5T,GAAeC,QAAQkU,IACvCP,GAAeA,IAAgBltB,GACjCouB,GAAUtV,YAAYoU,GAAa7J,OAExB+K,GAAUrV,oBAAoB/Y,GACtCob,OAAO3H,KACd,IACAO,GAAac,GAAGzhB,OAAQg6B,IAAuB,KAC7C,IAAK,MAAM7f,KAAY8L,GAAe1T,KAAK6nB,IACzCW,GAAUrV,oBAAoBvL,GAAU8V,MAC1C,IAEFtP,GAAac,GAAGzhB,OAAQ06B,IAAc,KACpC,IAAK,MAAM/6B,KAAWsmB,GAAe1T,KAAK,gDACG,UAAvClN,iBAAiB1F,GAASiC,UAC5Bm5B,GAAUrV,oBAAoB/lB,GAASqwB,MAE3C,IAEF/I,GAAqB8T,IAMrBxe,GAAmBwe,IAUnB,MACME,GAAmB,CAEvB,IAAK,CAAC,QAAS,MAAO,KAAM,OAAQ,OAHP,kBAI7BhqB,EAAG,CAAC,SAAU,OAAQ,QAAS,OAC/BiqB,KAAM,GACNhqB,EAAG,GACHiqB,GAAI,GACJC,IAAK,GACLC,KAAM,GACNC,GAAI,GACJC,IAAK,GACLC,GAAI,GACJC,GAAI,GACJC,GAAI,GACJC,GAAI,GACJC,GAAI,GACJC,GAAI,GACJC,GAAI,GACJC,GAAI,GACJC,GAAI,GACJC,GAAI,GACJxqB,EAAG,GACH0b,IAAK,CAAC,MAAO,SAAU,MAAO,QAAS,QAAS,UAChD+O,GAAI,GACJC,GAAI,GACJC,EAAG,GACHC,IAAK,GACLC,EAAG,GACHC,MAAO,GACPC,KAAM,GACNC,IAAK,GACLC,IAAK,GACLC,OAAQ,GACRC,EAAG,GACHC,GAAI,IAIAC,GAAgB,IAAIpmB,IAAI,CAAC,aAAc,OAAQ,OAAQ,WAAY,WAAY,SAAU,MAAO,eAShGqmB,GAAmB,0DACnBC,GAAmB,CAAC76B,EAAW86B,KACnC,MAAMC,EAAgB/6B,EAAUvC,SAASC,cACzC,OAAIo9B,EAAqBzb,SAAS0b,IAC5BJ,GAAc/lB,IAAImmB,IACbhc,QAAQ6b,GAAiBt5B,KAAKtB,EAAUg7B,YAM5CF,EAAqB12B,QAAO62B,GAAkBA,aAA0BzY,SAAQ9R,MAAKwqB,GAASA,EAAM55B,KAAKy5B,IAAe,EA0C3HI,GAAY,CAChBC,UAAWtC,GACXuC,QAAS,CAAC,EAEVC,WAAY,GACZxwB,MAAM,EACNywB,UAAU,EACVC,WAAY,KACZC,SAAU,eAENC,GAAgB,CACpBN,UAAW,SACXC,QAAS,SACTC,WAAY,oBACZxwB,KAAM,UACNywB,SAAU,UACVC,WAAY,kBACZC,SAAU,UAENE,GAAqB,CACzBC,MAAO,iCACP5jB,SAAU,oBAOZ,MAAM6jB,WAAwBna,GAC5B,WAAAU,CAAYL,GACVa,QACA3E,KAAK6E,QAAU7E,KAAK6D,WAAWC,EACjC,CAGA,kBAAWJ,GACT,OAAOwZ,EACT,CACA,sBAAWvZ,GACT,OAAO8Z,EACT,CACA,eAAWlhB,GACT,MA3CW,iBA4Cb,CAGA,UAAAshB,GACE,OAAO7gC,OAAOmiB,OAAOa,KAAK6E,QAAQuY,SAASt6B,KAAIghB,GAAU9D,KAAK8d,yBAAyBha,KAAS3d,OAAO2a,QACzG,CACA,UAAAid,GACE,OAAO/d,KAAK6d,aAAantB,OAAS,CACpC,CACA,aAAAstB,CAAcZ,GAMZ,OALApd,KAAKie,cAAcb,GACnBpd,KAAK6E,QAAQuY,QAAU,IAClBpd,KAAK6E,QAAQuY,WACbA,GAEEpd,IACT,CACA,MAAAke,GACE,MAAMC,EAAkB94B,SAASwvB,cAAc,OAC/CsJ,EAAgBC,UAAYpe,KAAKqe,eAAere,KAAK6E,QAAQ2Y,UAC7D,IAAK,MAAOzjB,EAAUukB,KAASthC,OAAOmkB,QAAQnB,KAAK6E,QAAQuY,SACzDpd,KAAKue,YAAYJ,EAAiBG,EAAMvkB,GAE1C,MAAMyjB,EAAWW,EAAgBpY,SAAS,GACpCsX,EAAard,KAAK8d,yBAAyB9d,KAAK6E,QAAQwY,YAI9D,OAHIA,GACFG,EAASniB,UAAU5E,OAAO4mB,EAAWn7B,MAAM,MAEtCs7B,CACT,CAGA,gBAAAvZ,CAAiBH,GACfa,MAAMV,iBAAiBH,GACvB9D,KAAKie,cAAcna,EAAOsZ,QAC5B,CACA,aAAAa,CAAcO,GACZ,IAAK,MAAOzkB,EAAUqjB,KAAYpgC,OAAOmkB,QAAQqd,GAC/C7Z,MAAMV,iBAAiB,CACrBlK,WACA4jB,MAAOP,GACNM,GAEP,CACA,WAAAa,CAAYf,EAAUJ,EAASrjB,GAC7B,MAAM0kB,EAAkB5Y,GAAeC,QAAQ/L,EAAUyjB,GACpDiB,KAGLrB,EAAUpd,KAAK8d,yBAAyBV,IAKpC,GAAUA,GACZpd,KAAK0e,sBAAsBhkB,GAAW0iB,GAAUqB,GAG9Cze,KAAK6E,QAAQhY,KACf4xB,EAAgBL,UAAYpe,KAAKqe,eAAejB,GAGlDqB,EAAgBE,YAAcvB,EAX5BqB,EAAgB9kB,SAYpB,CACA,cAAA0kB,CAAeG,GACb,OAAOxe,KAAK6E,QAAQyY,SApJxB,SAAsBsB,EAAYzB,EAAW0B,GAC3C,IAAKD,EAAWluB,OACd,OAAOkuB,EAET,GAAIC,GAAgD,mBAArBA,EAC7B,OAAOA,EAAiBD,GAE1B,MACME,GADY,IAAIl/B,OAAOm/B,WACKC,gBAAgBJ,EAAY,aACxD/9B,EAAW,GAAGlC,UAAUmgC,EAAgB5yB,KAAKkU,iBAAiB,MACpE,IAAK,MAAM7gB,KAAWsB,EAAU,CAC9B,MAAMo+B,EAAc1/B,EAAQC,SAASC,cACrC,IAAKzC,OAAO4D,KAAKu8B,GAAW/b,SAAS6d,GAAc,CACjD1/B,EAAQoa,SACR,QACF,CACA,MAAMulB,EAAgB,GAAGvgC,UAAUY,EAAQ0B,YACrCk+B,EAAoB,GAAGxgC,OAAOw+B,EAAU,MAAQ,GAAIA,EAAU8B,IAAgB,IACpF,IAAK,MAAMl9B,KAAam9B,EACjBtC,GAAiB76B,EAAWo9B,IAC/B5/B,EAAQ4B,gBAAgBY,EAAUvC,SAGxC,CACA,OAAOs/B,EAAgB5yB,KAAKkyB,SAC9B,CA2HmCgB,CAAaZ,EAAKxe,KAAK6E,QAAQsY,UAAWnd,KAAK6E,QAAQ0Y,YAAciB,CACtG,CACA,wBAAAV,CAAyBU,GACvB,OAAO3hB,GAAQ2hB,EAAK,CAACxe,MACvB,CACA,qBAAA0e,CAAsBn/B,EAASk/B,GAC7B,GAAIze,KAAK6E,QAAQhY,KAGf,OAFA4xB,EAAgBL,UAAY,QAC5BK,EAAgB3J,OAAOv1B,GAGzBk/B,EAAgBE,YAAcp/B,EAAQo/B,WACxC,EAeF,MACMU,GAAwB,IAAI/oB,IAAI,CAAC,WAAY,YAAa,eAC1DgpB,GAAoB,OAEpBC,GAAoB,OACpBC,GAAyB,iBACzBC,GAAiB,SACjBC,GAAmB,gBACnBC,GAAgB,QAChBC,GAAgB,QAahBC,GAAgB,CACpBC,KAAM,OACNC,IAAK,MACLC,MAAO/jB,KAAU,OAAS,QAC1BgkB,OAAQ,SACRC,KAAMjkB,KAAU,QAAU,QAEtBkkB,GAAY,CAChBhD,UAAWtC,GACXuF,WAAW,EACXnyB,SAAU,kBACVoyB,WAAW,EACXC,YAAa,GACbC,MAAO,EACPvwB,mBAAoB,CAAC,MAAO,QAAS,SAAU,QAC/CnD,MAAM,EACN7E,OAAQ,CAAC,EAAG,GACZtJ,UAAW,MACXszB,aAAc,KACdsL,UAAU,EACVC,WAAY,KACZxjB,UAAU,EACVyjB,SAAU,+GACVgD,MAAO,GACP5e,QAAS,eAEL6e,GAAgB,CACpBtD,UAAW,SACXiD,UAAW,UACXnyB,SAAU,mBACVoyB,UAAW,2BACXC,YAAa,oBACbC,MAAO,kBACPvwB,mBAAoB,QACpBnD,KAAM,UACN7E,OAAQ,0BACRtJ,UAAW,oBACXszB,aAAc,yBACdsL,SAAU,UACVC,WAAY,kBACZxjB,SAAU,mBACVyjB,SAAU,SACVgD,MAAO,4BACP5e,QAAS,UAOX,MAAM8e,WAAgBhc,GACpB,WAAAP,CAAY5kB,EAASukB,GACnB,QAAsB,IAAX,EACT,MAAM,IAAIU,UAAU,+DAEtBG,MAAMplB,EAASukB,GAGf9D,KAAK2gB,YAAa,EAClB3gB,KAAK4gB,SAAW,EAChB5gB,KAAK6gB,WAAa,KAClB7gB,KAAK8gB,eAAiB,CAAC,EACvB9gB,KAAKmS,QAAU,KACfnS,KAAK+gB,iBAAmB,KACxB/gB,KAAKghB,YAAc,KAGnBhhB,KAAKihB,IAAM,KACXjhB,KAAKkhB,gBACAlhB,KAAK6E,QAAQ9K,UAChBiG,KAAKmhB,WAET,CAGA,kBAAWzd,GACT,OAAOyc,EACT,CACA,sBAAWxc,GACT,OAAO8c,EACT,CACA,eAAWlkB,GACT,MAxGW,SAyGb,CAGA,MAAA6kB,GACEphB,KAAK2gB,YAAa,CACpB,CACA,OAAAU,GACErhB,KAAK2gB,YAAa,CACpB,CACA,aAAAW,GACEthB,KAAK2gB,YAAc3gB,KAAK2gB,UAC1B,CACA,MAAAhZ,GACO3H,KAAK2gB,aAGV3gB,KAAK8gB,eAAeS,OAASvhB,KAAK8gB,eAAeS,MAC7CvhB,KAAK2P,WACP3P,KAAKwhB,SAGPxhB,KAAKyhB,SACP,CACA,OAAA1c,GACEmI,aAAalN,KAAK4gB,UAClBrgB,GAAaC,IAAIR,KAAK4E,SAAS5J,QAAQykB,IAAiBC,GAAkB1f,KAAK0hB,mBAC3E1hB,KAAK4E,SAASpJ,aAAa,2BAC7BwE,KAAK4E,SAASxjB,aAAa,QAAS4e,KAAK4E,SAASpJ,aAAa,2BAEjEwE,KAAK2hB,iBACLhd,MAAMI,SACR,CACA,IAAA8K,GACE,GAAoC,SAAhC7P,KAAK4E,SAAS7jB,MAAMgxB,QACtB,MAAM,IAAInO,MAAM,uCAElB,IAAM5D,KAAK4hB,mBAAoB5hB,KAAK2gB,WAClC,OAEF,MAAMnH,EAAYjZ,GAAaqB,QAAQ5B,KAAK4E,SAAU5E,KAAKmE,YAAYqB,UAlItD,SAoIXqc,GADapmB,GAAeuE,KAAK4E,WACL5E,KAAK4E,SAAS9kB,cAAcwF,iBAAiBd,SAASwb,KAAK4E,UAC7F,GAAI4U,EAAUxX,mBAAqB6f,EACjC,OAIF7hB,KAAK2hB,iBACL,MAAMV,EAAMjhB,KAAK8hB,iBACjB9hB,KAAK4E,SAASxjB,aAAa,mBAAoB6/B,EAAIzlB,aAAa,OAChE,MAAM,UACJ6kB,GACErgB,KAAK6E,QAYT,GAXK7E,KAAK4E,SAAS9kB,cAAcwF,gBAAgBd,SAASwb,KAAKihB,OAC7DZ,EAAUvL,OAAOmM,GACjB1gB,GAAaqB,QAAQ5B,KAAK4E,SAAU5E,KAAKmE,YAAYqB,UAhJpC,cAkJnBxF,KAAKmS,QAAUnS,KAAKwS,cAAcyO,GAClCA,EAAI5lB,UAAU5E,IAAI8oB,IAMd,iBAAkBl6B,SAASC,gBAC7B,IAAK,MAAM/F,IAAW,GAAGZ,UAAU0G,SAAS6G,KAAK6Z,UAC/CxF,GAAac,GAAG9hB,EAAS,YAAaqc,IAU1CoE,KAAKmF,gBAPY,KACf5E,GAAaqB,QAAQ5B,KAAK4E,SAAU5E,KAAKmE,YAAYqB,UAhKrC,WAiKQ,IAApBxF,KAAK6gB,YACP7gB,KAAKwhB,SAEPxhB,KAAK6gB,YAAa,CAAK,GAEK7gB,KAAKihB,IAAKjhB,KAAKgO,cAC/C,CACA,IAAA4B,GACE,GAAK5P,KAAK2P,aAGQpP,GAAaqB,QAAQ5B,KAAK4E,SAAU5E,KAAKmE,YAAYqB,UA/KtD,SAgLHxD,iBAAd,CAQA,GALYhC,KAAK8hB,iBACbzmB,UAAU1B,OAAO4lB,IAIjB,iBAAkBl6B,SAASC,gBAC7B,IAAK,MAAM/F,IAAW,GAAGZ,UAAU0G,SAAS6G,KAAK6Z,UAC/CxF,GAAaC,IAAIjhB,EAAS,YAAaqc,IAG3CoE,KAAK8gB,eAA4B,OAAI,EACrC9gB,KAAK8gB,eAAelB,KAAiB,EACrC5f,KAAK8gB,eAAenB,KAAiB,EACrC3f,KAAK6gB,WAAa,KAYlB7gB,KAAKmF,gBAVY,KACXnF,KAAK+hB,yBAGJ/hB,KAAK6gB,YACR7gB,KAAK2hB,iBAEP3hB,KAAK4E,SAASzjB,gBAAgB,oBAC9Bof,GAAaqB,QAAQ5B,KAAK4E,SAAU5E,KAAKmE,YAAYqB,UAzMpC,WAyM8D,GAEnDxF,KAAKihB,IAAKjhB,KAAKgO,cA1B7C,CA2BF,CACA,MAAAjjB,GACMiV,KAAKmS,SACPnS,KAAKmS,QAAQpnB,QAEjB,CAGA,cAAA62B,GACE,OAAO9gB,QAAQd,KAAKgiB,YACtB,CACA,cAAAF,GAIE,OAHK9hB,KAAKihB,MACRjhB,KAAKihB,IAAMjhB,KAAKiiB,kBAAkBjiB,KAAKghB,aAAehhB,KAAKkiB,2BAEtDliB,KAAKihB,GACd,CACA,iBAAAgB,CAAkB7E,GAChB,MAAM6D,EAAMjhB,KAAKmiB,oBAAoB/E,GAASc,SAG9C,IAAK+C,EACH,OAAO,KAETA,EAAI5lB,UAAU1B,OAAO2lB,GAAmBC,IAExC0B,EAAI5lB,UAAU5E,IAAI,MAAMuJ,KAAKmE,YAAY5H,aACzC,MAAM6lB,EAvuGKC,KACb,GACEA,GAAUlgC,KAAKmgC,MA/BH,IA+BSngC,KAAKogC,gBACnBl9B,SAASm9B,eAAeH,IACjC,OAAOA,CAAM,EAmuGGI,CAAOziB,KAAKmE,YAAY5H,MAAM1c,WAK5C,OAJAohC,EAAI7/B,aAAa,KAAMghC,GACnBpiB,KAAKgO,eACPiT,EAAI5lB,UAAU5E,IAAI6oB,IAEb2B,CACT,CACA,UAAAyB,CAAWtF,GACTpd,KAAKghB,YAAc5D,EACfpd,KAAK2P,aACP3P,KAAK2hB,iBACL3hB,KAAK6P,OAET,CACA,mBAAAsS,CAAoB/E,GAYlB,OAXIpd,KAAK+gB,iBACP/gB,KAAK+gB,iBAAiB/C,cAAcZ,GAEpCpd,KAAK+gB,iBAAmB,IAAInD,GAAgB,IACvC5d,KAAK6E,QAGRuY,UACAC,WAAYrd,KAAK8d,yBAAyB9d,KAAK6E,QAAQyb,eAGpDtgB,KAAK+gB,gBACd,CACA,sBAAAmB,GACE,MAAO,CACL,CAAC1C,IAAyBxf,KAAKgiB,YAEnC,CACA,SAAAA,GACE,OAAOhiB,KAAK8d,yBAAyB9d,KAAK6E,QAAQ2b,QAAUxgB,KAAK4E,SAASpJ,aAAa,yBACzF,CAGA,4BAAAmnB,CAA6BvjB,GAC3B,OAAOY,KAAKmE,YAAYmB,oBAAoBlG,EAAMW,eAAgBC,KAAK4iB,qBACzE,CACA,WAAA5U,GACE,OAAOhO,KAAK6E,QAAQub,WAAapgB,KAAKihB,KAAOjhB,KAAKihB,IAAI5lB,UAAU7W,SAAS86B,GAC3E,CACA,QAAA3P,GACE,OAAO3P,KAAKihB,KAAOjhB,KAAKihB,IAAI5lB,UAAU7W,SAAS+6B,GACjD,CACA,aAAA/M,CAAcyO,GACZ,MAAMviC,EAAYme,GAAQmD,KAAK6E,QAAQnmB,UAAW,CAACshB,KAAMihB,EAAKjhB,KAAK4E,WAC7Die,EAAahD,GAAcnhC,EAAU+lB,eAC3C,OAAO,GAAoBzE,KAAK4E,SAAUqc,EAAKjhB,KAAK4S,iBAAiBiQ,GACvE,CACA,UAAA7P,GACE,MAAM,OACJhrB,GACEgY,KAAK6E,QACT,MAAsB,iBAAX7c,EACFA,EAAO9F,MAAM,KAAKY,KAAInF,GAAS4f,OAAOgQ,SAAS5vB,EAAO,MAEzC,mBAAXqK,EACFirB,GAAcjrB,EAAOirB,EAAYjT,KAAK4E,UAExC5c,CACT,CACA,wBAAA81B,CAAyBU,GACvB,OAAO3hB,GAAQ2hB,EAAK,CAACxe,KAAK4E,UAC5B,CACA,gBAAAgO,CAAiBiQ,GACf,MAAM3P,EAAwB,CAC5Bx0B,UAAWmkC,EACXzsB,UAAW,CAAC,CACV9V,KAAM,OACNmB,QAAS,CACPuO,mBAAoBgQ,KAAK6E,QAAQ7U,qBAElC,CACD1P,KAAM,SACNmB,QAAS,CACPuG,OAAQgY,KAAKgT,eAEd,CACD1yB,KAAM,kBACNmB,QAAS,CACPwM,SAAU+R,KAAK6E,QAAQ5W,WAExB,CACD3N,KAAM,QACNmB,QAAS,CACPlC,QAAS,IAAIygB,KAAKmE,YAAY5H,eAE/B,CACDjc,KAAM,kBACNC,SAAS,EACTC,MAAO,aACPC,GAAI4J,IAGF2V,KAAK8hB,iBAAiB1gC,aAAa,wBAAyBiJ,EAAK1J,MAAMjC,UAAU,KAIvF,MAAO,IACFw0B,KACArW,GAAQmD,KAAK6E,QAAQmN,aAAc,CAACkB,IAE3C,CACA,aAAAgO,GACE,MAAM4B,EAAW9iB,KAAK6E,QAAQjD,QAAQ1f,MAAM,KAC5C,IAAK,MAAM0f,KAAWkhB,EACpB,GAAgB,UAAZlhB,EACFrB,GAAac,GAAGrB,KAAK4E,SAAU5E,KAAKmE,YAAYqB,UAjVlC,SAiV4DxF,KAAK6E,QAAQ9K,UAAUqF,IAC/EY,KAAK2iB,6BAA6BvjB,GAC1CuI,QAAQ,SAEb,GA3VU,WA2VN/F,EAA4B,CACrC,MAAMmhB,EAAUnhB,IAAY+d,GAAgB3f,KAAKmE,YAAYqB,UAnV5C,cAmV0ExF,KAAKmE,YAAYqB,UArV5F,WAsVVwd,EAAWphB,IAAY+d,GAAgB3f,KAAKmE,YAAYqB,UAnV7C,cAmV2ExF,KAAKmE,YAAYqB,UArV5F,YAsVjBjF,GAAac,GAAGrB,KAAK4E,SAAUme,EAAS/iB,KAAK6E,QAAQ9K,UAAUqF,IAC7D,MAAMkU,EAAUtT,KAAK2iB,6BAA6BvjB,GAClDkU,EAAQwN,eAA8B,YAAf1hB,EAAMqB,KAAqBmf,GAAgBD,KAAiB,EACnFrM,EAAQmO,QAAQ,IAElBlhB,GAAac,GAAGrB,KAAK4E,SAAUoe,EAAUhjB,KAAK6E,QAAQ9K,UAAUqF,IAC9D,MAAMkU,EAAUtT,KAAK2iB,6BAA6BvjB,GAClDkU,EAAQwN,eAA8B,aAAf1hB,EAAMqB,KAAsBmf,GAAgBD,IAAiBrM,EAAQ1O,SAASpgB,SAAS4a,EAAMU,eACpHwT,EAAQkO,QAAQ,GAEpB,CAEFxhB,KAAK0hB,kBAAoB,KACnB1hB,KAAK4E,UACP5E,KAAK4P,MACP,EAEFrP,GAAac,GAAGrB,KAAK4E,SAAS5J,QAAQykB,IAAiBC,GAAkB1f,KAAK0hB,kBAChF,CACA,SAAAP,GACE,MAAMX,EAAQxgB,KAAK4E,SAASpJ,aAAa,SACpCglB,IAGAxgB,KAAK4E,SAASpJ,aAAa,eAAkBwE,KAAK4E,SAAS+Z,YAAYhZ,QAC1E3F,KAAK4E,SAASxjB,aAAa,aAAco/B,GAE3CxgB,KAAK4E,SAASxjB,aAAa,yBAA0Bo/B,GACrDxgB,KAAK4E,SAASzjB,gBAAgB,SAChC,CACA,MAAAsgC,GACMzhB,KAAK2P,YAAc3P,KAAK6gB,WAC1B7gB,KAAK6gB,YAAa,GAGpB7gB,KAAK6gB,YAAa,EAClB7gB,KAAKijB,aAAY,KACXjjB,KAAK6gB,YACP7gB,KAAK6P,MACP,GACC7P,KAAK6E,QAAQ0b,MAAM1Q,MACxB,CACA,MAAA2R,GACMxhB,KAAK+hB,yBAGT/hB,KAAK6gB,YAAa,EAClB7gB,KAAKijB,aAAY,KACVjjB,KAAK6gB,YACR7gB,KAAK4P,MACP,GACC5P,KAAK6E,QAAQ0b,MAAM3Q,MACxB,CACA,WAAAqT,CAAYrlB,EAASslB,GACnBhW,aAAalN,KAAK4gB,UAClB5gB,KAAK4gB,SAAW/iB,WAAWD,EAASslB,EACtC,CACA,oBAAAnB,GACE,OAAO/kC,OAAOmiB,OAAOa,KAAK8gB,gBAAgB1f,UAAS,EACrD,CACA,UAAAyC,CAAWC,GACT,MAAMqf,EAAiBngB,GAAYG,kBAAkBnD,KAAK4E,UAC1D,IAAK,MAAMwe,KAAiBpmC,OAAO4D,KAAKuiC,GAClC9D,GAAsB1oB,IAAIysB,WACrBD,EAAeC,GAU1B,OAPAtf,EAAS,IACJqf,KACmB,iBAAXrf,GAAuBA,EAASA,EAAS,CAAC,GAEvDA,EAAS9D,KAAK+D,gBAAgBD,GAC9BA,EAAS9D,KAAKgE,kBAAkBF,GAChC9D,KAAKiE,iBAAiBH,GACfA,CACT,CACA,iBAAAE,CAAkBF,GAchB,OAbAA,EAAOuc,WAAiC,IAArBvc,EAAOuc,UAAsBh7B,SAAS6G,KAAOwO,GAAWoJ,EAAOuc,WACtD,iBAAjBvc,EAAOyc,QAChBzc,EAAOyc,MAAQ,CACb1Q,KAAM/L,EAAOyc,MACb3Q,KAAM9L,EAAOyc,QAGW,iBAAjBzc,EAAO0c,QAChB1c,EAAO0c,MAAQ1c,EAAO0c,MAAM3gC,YAEA,iBAAnBikB,EAAOsZ,UAChBtZ,EAAOsZ,QAAUtZ,EAAOsZ,QAAQv9B,YAE3BikB,CACT,CACA,kBAAA8e,GACE,MAAM9e,EAAS,CAAC,EAChB,IAAK,MAAOhnB,EAAKa,KAAUX,OAAOmkB,QAAQnB,KAAK6E,SACzC7E,KAAKmE,YAAYT,QAAQ5mB,KAASa,IACpCmmB,EAAOhnB,GAAOa,GASlB,OANAmmB,EAAO/J,UAAW,EAClB+J,EAAOlC,QAAU,SAKVkC,CACT,CACA,cAAA6d,GACM3hB,KAAKmS,UACPnS,KAAKmS,QAAQnZ,UACbgH,KAAKmS,QAAU,MAEbnS,KAAKihB,MACPjhB,KAAKihB,IAAItnB,SACTqG,KAAKihB,IAAM,KAEf,CAGA,sBAAOxkB,CAAgBqH,GACrB,OAAO9D,KAAKwH,MAAK,WACf,MAAMnd,EAAOq2B,GAAQpb,oBAAoBtF,KAAM8D,GAC/C,GAAsB,iBAAXA,EAAX,CAGA,QAA4B,IAAjBzZ,EAAKyZ,GACd,MAAM,IAAIU,UAAU,oBAAoBV,MAE1CzZ,EAAKyZ,IAJL,CAKF,GACF,EAOF3H,GAAmBukB,IAcnB,MACM2C,GAAiB,kBACjBC,GAAmB,gBACnBC,GAAY,IACb7C,GAAQhd,QACX0Z,QAAS,GACTp1B,OAAQ,CAAC,EAAG,GACZtJ,UAAW,QACX8+B,SAAU,8IACV5b,QAAS,SAEL4hB,GAAgB,IACjB9C,GAAQ/c,YACXyZ,QAAS,kCAOX,MAAMqG,WAAgB/C,GAEpB,kBAAWhd,GACT,OAAO6f,EACT,CACA,sBAAW5f,GACT,OAAO6f,EACT,CACA,eAAWjnB,GACT,MA7BW,SA8Bb,CAGA,cAAAqlB,GACE,OAAO5hB,KAAKgiB,aAAehiB,KAAK0jB,aAClC,CAGA,sBAAAxB,GACE,MAAO,CACL,CAACmB,IAAiBrjB,KAAKgiB,YACvB,CAACsB,IAAmBtjB,KAAK0jB,cAE7B,CACA,WAAAA,GACE,OAAO1jB,KAAK8d,yBAAyB9d,KAAK6E,QAAQuY,QACpD,CAGA,sBAAO3gB,CAAgBqH,GACrB,OAAO9D,KAAKwH,MAAK,WACf,MAAMnd,EAAOo5B,GAAQne,oBAAoBtF,KAAM8D,GAC/C,GAAsB,iBAAXA,EAAX,CAGA,QAA4B,IAAjBzZ,EAAKyZ,GACd,MAAM,IAAIU,UAAU,oBAAoBV,MAE1CzZ,EAAKyZ,IAJL,CAKF,GACF,EAOF3H,GAAmBsnB,IAcnB,MAEME,GAAc,gBAEdC,GAAiB,WAAWD,KAC5BE,GAAc,QAAQF,KACtBG,GAAwB,OAAOH,cAE/BI,GAAsB,SAEtBC,GAAwB,SAExBC,GAAqB,YAGrBC,GAAsB,GAAGD,mBAA+CA,uBAGxEE,GAAY,CAChBn8B,OAAQ,KAERo8B,WAAY,eACZC,cAAc,EACd93B,OAAQ,KACR+3B,UAAW,CAAC,GAAK,GAAK,IAElBC,GAAgB,CACpBv8B,OAAQ,gBAERo8B,WAAY,SACZC,aAAc,UACd93B,OAAQ,UACR+3B,UAAW,SAOb,MAAME,WAAkB9f,GACtB,WAAAP,CAAY5kB,EAASukB,GACnBa,MAAMplB,EAASukB,GAGf9D,KAAKykB,aAAe,IAAIvzB,IACxB8O,KAAK0kB,oBAAsB,IAAIxzB,IAC/B8O,KAAK2kB,aAA6D,YAA9C1/B,iBAAiB+a,KAAK4E,UAAU5Y,UAA0B,KAAOgU,KAAK4E,SAC1F5E,KAAK4kB,cAAgB,KACrB5kB,KAAK6kB,UAAY,KACjB7kB,KAAK8kB,oBAAsB,CACzBC,gBAAiB,EACjBC,gBAAiB,GAEnBhlB,KAAKilB,SACP,CAGA,kBAAWvhB,GACT,OAAOygB,EACT,CACA,sBAAWxgB,GACT,OAAO4gB,EACT,CACA,eAAWhoB,GACT,MAhEW,WAiEb,CAGA,OAAA0oB,GACEjlB,KAAKklB,mCACLllB,KAAKmlB,2BACDnlB,KAAK6kB,UACP7kB,KAAK6kB,UAAUO,aAEfplB,KAAK6kB,UAAY7kB,KAAKqlB,kBAExB,IAAK,MAAMC,KAAWtlB,KAAK0kB,oBAAoBvlB,SAC7Ca,KAAK6kB,UAAUU,QAAQD,EAE3B,CACA,OAAAvgB,GACE/E,KAAK6kB,UAAUO,aACfzgB,MAAMI,SACR,CAGA,iBAAAf,CAAkBF,GAShB,OAPAA,EAAOvX,OAASmO,GAAWoJ,EAAOvX,SAAWlH,SAAS6G,KAGtD4X,EAAOsgB,WAAatgB,EAAO9b,OAAS,GAAG8b,EAAO9b,oBAAsB8b,EAAOsgB,WAC3C,iBAArBtgB,EAAOwgB,YAChBxgB,EAAOwgB,UAAYxgB,EAAOwgB,UAAUpiC,MAAM,KAAKY,KAAInF,GAAS4f,OAAOC,WAAW7f,MAEzEmmB,CACT,CACA,wBAAAqhB,GACOnlB,KAAK6E,QAAQwf,eAKlB9jB,GAAaC,IAAIR,KAAK6E,QAAQtY,OAAQs3B,IACtCtjB,GAAac,GAAGrB,KAAK6E,QAAQtY,OAAQs3B,GAAaG,IAAuB5kB,IACvE,MAAMomB,EAAoBxlB,KAAK0kB,oBAAoBvnC,IAAIiiB,EAAM7S,OAAOtB,MACpE,GAAIu6B,EAAmB,CACrBpmB,EAAMkD,iBACN,MAAM3G,EAAOqE,KAAK2kB,cAAgB/kC,OAC5BmE,EAASyhC,EAAkBnhC,UAAY2b,KAAK4E,SAASvgB,UAC3D,GAAIsX,EAAK8pB,SAKP,YAJA9pB,EAAK8pB,SAAS,CACZ9jC,IAAKoC,EACL2hC,SAAU,WAMd/pB,EAAKlQ,UAAY1H,CACnB,KAEJ,CACA,eAAAshC,GACE,MAAM5jC,EAAU,CACdka,KAAMqE,KAAK2kB,aACXL,UAAWtkB,KAAK6E,QAAQyf,UACxBF,WAAYpkB,KAAK6E,QAAQuf,YAE3B,OAAO,IAAIuB,sBAAqBxkB,GAAWnB,KAAK4lB,kBAAkBzkB,IAAU1f,EAC9E,CAGA,iBAAAmkC,CAAkBzkB,GAChB,MAAM0kB,EAAgBlI,GAAS3d,KAAKykB,aAAatnC,IAAI,IAAIwgC,EAAMpxB,OAAO4N,MAChEub,EAAWiI,IACf3d,KAAK8kB,oBAAoBC,gBAAkBpH,EAAMpxB,OAAOlI,UACxD2b,KAAK8lB,SAASD,EAAclI,GAAO,EAE/BqH,GAAmBhlB,KAAK2kB,cAAgBt/B,SAASC,iBAAiBmG,UAClEs6B,EAAkBf,GAAmBhlB,KAAK8kB,oBAAoBE,gBACpEhlB,KAAK8kB,oBAAoBE,gBAAkBA,EAC3C,IAAK,MAAMrH,KAASxc,EAAS,CAC3B,IAAKwc,EAAMqI,eAAgB,CACzBhmB,KAAK4kB,cAAgB,KACrB5kB,KAAKimB,kBAAkBJ,EAAclI,IACrC,QACF,CACA,MAAMuI,EAA2BvI,EAAMpxB,OAAOlI,WAAa2b,KAAK8kB,oBAAoBC,gBAEpF,GAAIgB,GAAmBG,GAGrB,GAFAxQ,EAASiI,IAEJqH,EACH,YAMCe,GAAoBG,GACvBxQ,EAASiI,EAEb,CACF,CACA,gCAAAuH,GACEllB,KAAKykB,aAAe,IAAIvzB,IACxB8O,KAAK0kB,oBAAsB,IAAIxzB,IAC/B,MAAMi1B,EAActgB,GAAe1T,KAAK6xB,GAAuBhkB,KAAK6E,QAAQtY,QAC5E,IAAK,MAAM65B,KAAUD,EAAa,CAEhC,IAAKC,EAAOn7B,MAAQiQ,GAAWkrB,GAC7B,SAEF,MAAMZ,EAAoB3f,GAAeC,QAAQugB,UAAUD,EAAOn7B,MAAO+U,KAAK4E,UAG1EjK,GAAU6qB,KACZxlB,KAAKykB,aAAa1yB,IAAIs0B,UAAUD,EAAOn7B,MAAOm7B,GAC9CpmB,KAAK0kB,oBAAoB3yB,IAAIq0B,EAAOn7B,KAAMu6B,GAE9C,CACF,CACA,QAAAM,CAASv5B,GACHyT,KAAK4kB,gBAAkBr4B,IAG3ByT,KAAKimB,kBAAkBjmB,KAAK6E,QAAQtY,QACpCyT,KAAK4kB,cAAgBr4B,EACrBA,EAAO8O,UAAU5E,IAAIstB,IACrB/jB,KAAKsmB,iBAAiB/5B,GACtBgU,GAAaqB,QAAQ5B,KAAK4E,SAAUgf,GAAgB,CAClD9jB,cAAevT,IAEnB,CACA,gBAAA+5B,CAAiB/5B,GAEf,GAAIA,EAAO8O,UAAU7W,SA9LQ,iBA+L3BqhB,GAAeC,QArLc,mBAqLsBvZ,EAAOyO,QAtLtC,cAsLkEK,UAAU5E,IAAIstB,SAGtG,IAAK,MAAMwC,KAAa1gB,GAAeI,QAAQ1Z,EA9LnB,qBAiM1B,IAAK,MAAMxJ,KAAQ8iB,GAAeM,KAAKogB,EAAWrC,IAChDnhC,EAAKsY,UAAU5E,IAAIstB,GAGzB,CACA,iBAAAkC,CAAkBxhC,GAChBA,EAAO4W,UAAU1B,OAAOoqB,IACxB,MAAMyC,EAAc3gB,GAAe1T,KAAK,GAAG6xB,MAAyBD,KAAuBt/B,GAC3F,IAAK,MAAM9E,KAAQ6mC,EACjB7mC,EAAK0b,UAAU1B,OAAOoqB,GAE1B,CAGA,sBAAOtnB,CAAgBqH,GACrB,OAAO9D,KAAKwH,MAAK,WACf,MAAMnd,EAAOm6B,GAAUlf,oBAAoBtF,KAAM8D,GACjD,GAAsB,iBAAXA,EAAX,CAGA,QAAqB/K,IAAjB1O,EAAKyZ,IAAyBA,EAAOrC,WAAW,MAAmB,gBAAXqC,EAC1D,MAAM,IAAIU,UAAU,oBAAoBV,MAE1CzZ,EAAKyZ,IAJL,CAKF,GACF,EAOFvD,GAAac,GAAGzhB,OAAQkkC,IAAuB,KAC7C,IAAK,MAAM2C,KAAO5gB,GAAe1T,KApOT,0BAqOtBqyB,GAAUlf,oBAAoBmhB,EAChC,IAOFtqB,GAAmBqoB,IAcnB,MAEMkC,GAAc,UACdC,GAAe,OAAOD,KACtBE,GAAiB,SAASF,KAC1BG,GAAe,OAAOH,KACtBI,GAAgB,QAAQJ,KACxBK,GAAuB,QAAQL,KAC/BM,GAAgB,UAAUN,KAC1BO,GAAsB,OAAOP,KAC7BQ,GAAiB,YACjBC,GAAkB,aAClBC,GAAe,UACfC,GAAiB,YACjBC,GAAW,OACXC,GAAU,MACVC,GAAoB,SACpBC,GAAoB,OACpBC,GAAoB,OAEpBC,GAA2B,mBAE3BC,GAA+B,QAAQD,MAIvCE,GAAuB,2EACvBC,GAAsB,YAFOF,uBAAiDA,mBAA6CA,OAE/EC,KAC5CE,GAA8B,IAAIP,8BAA6CA,+BAA8CA,4BAMnI,MAAMQ,WAAYtjB,GAChB,WAAAP,CAAY5kB,GACVolB,MAAMplB,GACNygB,KAAKoS,QAAUpS,KAAK4E,SAAS5J,QAdN,uCAelBgF,KAAKoS,UAOVpS,KAAKioB,sBAAsBjoB,KAAKoS,QAASpS,KAAKkoB,gBAC9C3nB,GAAac,GAAGrB,KAAK4E,SAAUoiB,IAAe5nB,GAASY,KAAK6M,SAASzN,KACvE,CAGA,eAAW7C,GACT,MAnDW,KAoDb,CAGA,IAAAsT,GAEE,MAAMsY,EAAYnoB,KAAK4E,SACvB,GAAI5E,KAAKooB,cAAcD,GACrB,OAIF,MAAME,EAASroB,KAAKsoB,iBACdC,EAAYF,EAAS9nB,GAAaqB,QAAQymB,EAAQ1B,GAAc,CACpE7mB,cAAeqoB,IACZ,KACa5nB,GAAaqB,QAAQumB,EAAWtB,GAAc,CAC9D/mB,cAAeuoB,IAEHrmB,kBAAoBumB,GAAaA,EAAUvmB,mBAGzDhC,KAAKwoB,YAAYH,EAAQF,GACzBnoB,KAAKyoB,UAAUN,EAAWE,GAC5B,CAGA,SAAAI,CAAUlpC,EAASmpC,GACZnpC,IAGLA,EAAQ8b,UAAU5E,IAAI+wB,IACtBxnB,KAAKyoB,UAAU5iB,GAAec,uBAAuBpnB,IAcrDygB,KAAKmF,gBAZY,KACsB,QAAjC5lB,EAAQic,aAAa,SAIzBjc,EAAQ4B,gBAAgB,YACxB5B,EAAQ6B,aAAa,iBAAiB,GACtC4e,KAAK2oB,gBAAgBppC,GAAS,GAC9BghB,GAAaqB,QAAQriB,EAASunC,GAAe,CAC3ChnB,cAAe4oB,KAPfnpC,EAAQ8b,UAAU5E,IAAIixB,GAQtB,GAE0BnoC,EAASA,EAAQ8b,UAAU7W,SAASijC,KACpE,CACA,WAAAe,CAAYjpC,EAASmpC,GACdnpC,IAGLA,EAAQ8b,UAAU1B,OAAO6tB,IACzBjoC,EAAQq7B,OACR5a,KAAKwoB,YAAY3iB,GAAec,uBAAuBpnB,IAcvDygB,KAAKmF,gBAZY,KACsB,QAAjC5lB,EAAQic,aAAa,SAIzBjc,EAAQ6B,aAAa,iBAAiB,GACtC7B,EAAQ6B,aAAa,WAAY,MACjC4e,KAAK2oB,gBAAgBppC,GAAS,GAC9BghB,GAAaqB,QAAQriB,EAASqnC,GAAgB,CAC5C9mB,cAAe4oB,KAPfnpC,EAAQ8b,UAAU1B,OAAO+tB,GAQzB,GAE0BnoC,EAASA,EAAQ8b,UAAU7W,SAASijC,KACpE,CACA,QAAA5a,CAASzN,GACP,IAAK,CAAC8nB,GAAgBC,GAAiBC,GAAcC,GAAgBC,GAAUC,IAASnmB,SAAShC,EAAMtiB,KACrG,OAEFsiB,EAAM0U,kBACN1U,EAAMkD,iBACN,MAAMyD,EAAW/F,KAAKkoB,eAAe/hC,QAAO5G,IAAY2b,GAAW3b,KACnE,IAAIqpC,EACJ,GAAI,CAACtB,GAAUC,IAASnmB,SAAShC,EAAMtiB,KACrC8rC,EAAoB7iB,EAAS3G,EAAMtiB,MAAQwqC,GAAW,EAAIvhB,EAASrV,OAAS,OACvE,CACL,MAAM8c,EAAS,CAAC2Z,GAAiBE,IAAgBjmB,SAAShC,EAAMtiB,KAChE8rC,EAAoB9qB,GAAqBiI,EAAU3G,EAAM7S,OAAQihB,GAAQ,EAC3E,CACIob,IACFA,EAAkBnW,MAAM,CACtBoW,eAAe,IAEjBb,GAAI1iB,oBAAoBsjB,GAAmB/Y,OAE/C,CACA,YAAAqY,GAEE,OAAOriB,GAAe1T,KAAK21B,GAAqB9nB,KAAKoS,QACvD,CACA,cAAAkW,GACE,OAAOtoB,KAAKkoB,eAAe/1B,MAAKzN,GAASsb,KAAKooB,cAAc1jC,MAAW,IACzE,CACA,qBAAAujC,CAAsBxjC,EAAQshB,GAC5B/F,KAAK8oB,yBAAyBrkC,EAAQ,OAAQ,WAC9C,IAAK,MAAMC,KAASqhB,EAClB/F,KAAK+oB,6BAA6BrkC,EAEtC,CACA,4BAAAqkC,CAA6BrkC,GAC3BA,EAAQsb,KAAKgpB,iBAAiBtkC,GAC9B,MAAMukC,EAAWjpB,KAAKooB,cAAc1jC,GAC9BwkC,EAAYlpB,KAAKmpB,iBAAiBzkC,GACxCA,EAAMtD,aAAa,gBAAiB6nC,GAChCC,IAAcxkC,GAChBsb,KAAK8oB,yBAAyBI,EAAW,OAAQ,gBAE9CD,GACHvkC,EAAMtD,aAAa,WAAY,MAEjC4e,KAAK8oB,yBAAyBpkC,EAAO,OAAQ,OAG7Csb,KAAKopB,mCAAmC1kC,EAC1C,CACA,kCAAA0kC,CAAmC1kC,GACjC,MAAM6H,EAASsZ,GAAec,uBAAuBjiB,GAChD6H,IAGLyT,KAAK8oB,yBAAyBv8B,EAAQ,OAAQ,YAC1C7H,EAAMyV,IACR6F,KAAK8oB,yBAAyBv8B,EAAQ,kBAAmB,GAAG7H,EAAMyV,MAEtE,CACA,eAAAwuB,CAAgBppC,EAAS8pC,GACvB,MAAMH,EAAYlpB,KAAKmpB,iBAAiB5pC,GACxC,IAAK2pC,EAAU7tB,UAAU7W,SApKN,YAqKjB,OAEF,MAAMmjB,EAAS,CAAC5N,EAAUoa,KACxB,MAAM50B,EAAUsmB,GAAeC,QAAQ/L,EAAUmvB,GAC7C3pC,GACFA,EAAQ8b,UAAUsM,OAAOwM,EAAWkV,EACtC,EAEF1hB,EAAOggB,GAA0BH,IACjC7f,EA5K2B,iBA4KI+f,IAC/BwB,EAAU9nC,aAAa,gBAAiBioC,EAC1C,CACA,wBAAAP,CAAyBvpC,EAASwC,EAAWpE,GACtC4B,EAAQgc,aAAaxZ,IACxBxC,EAAQ6B,aAAaW,EAAWpE,EAEpC,CACA,aAAAyqC,CAAc9Y,GACZ,OAAOA,EAAKjU,UAAU7W,SAASgjC,GACjC,CAGA,gBAAAwB,CAAiB1Z,GACf,OAAOA,EAAKtJ,QAAQ8hB,IAAuBxY,EAAOzJ,GAAeC,QAAQgiB,GAAqBxY,EAChG,CAGA,gBAAA6Z,CAAiB7Z,GACf,OAAOA,EAAKtU,QA5LO,gCA4LoBsU,CACzC,CAGA,sBAAO7S,CAAgBqH,GACrB,OAAO9D,KAAKwH,MAAK,WACf,MAAMnd,EAAO29B,GAAI1iB,oBAAoBtF,MACrC,GAAsB,iBAAX8D,EAAX,CAGA,QAAqB/K,IAAjB1O,EAAKyZ,IAAyBA,EAAOrC,WAAW,MAAmB,gBAAXqC,EAC1D,MAAM,IAAIU,UAAU,oBAAoBV,MAE1CzZ,EAAKyZ,IAJL,CAKF,GACF,EAOFvD,GAAac,GAAGhc,SAAU0hC,GAAsBc,IAAsB,SAAUzoB,GAC1E,CAAC,IAAK,QAAQgC,SAASpB,KAAKiH,UAC9B7H,EAAMkD,iBAEJpH,GAAW8E,OAGfgoB,GAAI1iB,oBAAoBtF,MAAM6P,MAChC,IAKAtP,GAAac,GAAGzhB,OAAQqnC,IAAqB,KAC3C,IAAK,MAAM1nC,KAAWsmB,GAAe1T,KAAK41B,IACxCC,GAAI1iB,oBAAoB/lB,EAC1B,IAMF4c,GAAmB6rB,IAcnB,MAEMhjB,GAAY,YACZskB,GAAkB,YAAYtkB,KAC9BukB,GAAiB,WAAWvkB,KAC5BwkB,GAAgB,UAAUxkB,KAC1BykB,GAAiB,WAAWzkB,KAC5B0kB,GAAa,OAAO1kB,KACpB2kB,GAAe,SAAS3kB,KACxB4kB,GAAa,OAAO5kB,KACpB6kB,GAAc,QAAQ7kB,KAEtB8kB,GAAkB,OAClBC,GAAkB,OAClBC,GAAqB,UACrBrmB,GAAc,CAClByc,UAAW,UACX6J,SAAU,UACV1J,MAAO,UAEH7c,GAAU,CACd0c,WAAW,EACX6J,UAAU,EACV1J,MAAO,KAOT,MAAM2J,WAAcxlB,GAClB,WAAAP,CAAY5kB,EAASukB,GACnBa,MAAMplB,EAASukB,GACf9D,KAAK4gB,SAAW,KAChB5gB,KAAKmqB,sBAAuB,EAC5BnqB,KAAKoqB,yBAA0B,EAC/BpqB,KAAKkhB,eACP,CAGA,kBAAWxd,GACT,OAAOA,EACT,CACA,sBAAWC,GACT,OAAOA,EACT,CACA,eAAWpH,GACT,MA/CS,OAgDX,CAGA,IAAAsT,GACoBtP,GAAaqB,QAAQ5B,KAAK4E,SAAUglB,IACxC5nB,mBAGdhC,KAAKqqB,gBACDrqB,KAAK6E,QAAQub,WACfpgB,KAAK4E,SAASvJ,UAAU5E,IA/CN,QAsDpBuJ,KAAK4E,SAASvJ,UAAU1B,OAAOmwB,IAC/BjuB,GAAOmE,KAAK4E,UACZ5E,KAAK4E,SAASvJ,UAAU5E,IAAIszB,GAAiBC,IAC7ChqB,KAAKmF,gBARY,KACfnF,KAAK4E,SAASvJ,UAAU1B,OAAOqwB,IAC/BzpB,GAAaqB,QAAQ5B,KAAK4E,SAAUilB,IACpC7pB,KAAKsqB,oBAAoB,GAKGtqB,KAAK4E,SAAU5E,KAAK6E,QAAQub,WAC5D,CACA,IAAAxQ,GACO5P,KAAKuqB,YAGQhqB,GAAaqB,QAAQ5B,KAAK4E,SAAU8kB,IACxC1nB,mBAQdhC,KAAK4E,SAASvJ,UAAU5E,IAAIuzB,IAC5BhqB,KAAKmF,gBANY,KACfnF,KAAK4E,SAASvJ,UAAU5E,IAAIqzB,IAC5B9pB,KAAK4E,SAASvJ,UAAU1B,OAAOqwB,GAAoBD,IACnDxpB,GAAaqB,QAAQ5B,KAAK4E,SAAU+kB,GAAa,GAGrB3pB,KAAK4E,SAAU5E,KAAK6E,QAAQub,YAC5D,CACA,OAAArb,GACE/E,KAAKqqB,gBACDrqB,KAAKuqB,WACPvqB,KAAK4E,SAASvJ,UAAU1B,OAAOowB,IAEjCplB,MAAMI,SACR,CACA,OAAAwlB,GACE,OAAOvqB,KAAK4E,SAASvJ,UAAU7W,SAASulC,GAC1C,CAIA,kBAAAO,GACOtqB,KAAK6E,QAAQolB,WAGdjqB,KAAKmqB,sBAAwBnqB,KAAKoqB,0BAGtCpqB,KAAK4gB,SAAW/iB,YAAW,KACzBmC,KAAK4P,MAAM,GACV5P,KAAK6E,QAAQ0b,QAClB,CACA,cAAAiK,CAAeprB,EAAOqrB,GACpB,OAAQrrB,EAAMqB,MACZ,IAAK,YACL,IAAK,WAEDT,KAAKmqB,qBAAuBM,EAC5B,MAEJ,IAAK,UACL,IAAK,WAEDzqB,KAAKoqB,wBAA0BK,EAIrC,GAAIA,EAEF,YADAzqB,KAAKqqB,gBAGP,MAAM5c,EAAcrO,EAAMU,cACtBE,KAAK4E,WAAa6I,GAAezN,KAAK4E,SAASpgB,SAASipB,IAG5DzN,KAAKsqB,oBACP,CACA,aAAApJ,GACE3gB,GAAac,GAAGrB,KAAK4E,SAAU0kB,IAAiBlqB,GAASY,KAAKwqB,eAAeprB,GAAO,KACpFmB,GAAac,GAAGrB,KAAK4E,SAAU2kB,IAAgBnqB,GAASY,KAAKwqB,eAAeprB,GAAO,KACnFmB,GAAac,GAAGrB,KAAK4E,SAAU4kB,IAAepqB,GAASY,KAAKwqB,eAAeprB,GAAO,KAClFmB,GAAac,GAAGrB,KAAK4E,SAAU6kB,IAAgBrqB,GAASY,KAAKwqB,eAAeprB,GAAO,IACrF,CACA,aAAAirB,GACEnd,aAAalN,KAAK4gB,UAClB5gB,KAAK4gB,SAAW,IAClB,CAGA,sBAAOnkB,CAAgBqH,GACrB,OAAO9D,KAAKwH,MAAK,WACf,MAAMnd,EAAO6/B,GAAM5kB,oBAAoBtF,KAAM8D,GAC7C,GAAsB,iBAAXA,EAAqB,CAC9B,QAA4B,IAAjBzZ,EAAKyZ,GACd,MAAM,IAAIU,UAAU,oBAAoBV,MAE1CzZ,EAAKyZ,GAAQ9D,KACf,CACF,GACF,ECr0IK,SAAS0qB,GAAcruB,GACD,WAAvBhX,SAASuX,WAAyBP,IACjChX,SAASyF,iBAAiB,mBAAoBuR,EACrD,CDy0IAwK,GAAqBqjB,IAMrB/tB,GAAmB+tB,IEpyInBQ,IAzCA,WAC2B,GAAGt4B,MAAM5U,KAChC6H,SAAS+a,iBAAiB,+BAETtd,KAAI,SAAU6nC,GAC/B,OAAO,IAAI,GAAkBA,EAAkB,CAC7CpK,MAAO,CAAE1Q,KAAM,IAAKD,KAAM,MAE9B,GACF,IAiCA8a,IA5BA,WACYrlC,SAASm9B,eAAe,mBAC9B13B,iBAAiB,SAAS,WAC5BzF,SAAS6G,KAAKT,UAAY,EAC1BpG,SAASC,gBAAgBmG,UAAY,CACvC,GACF,IAuBAi/B,IArBA,WACE,IAAIE,EAAMvlC,SAASm9B,eAAe,mBAC9BqI,EAASxlC,SACVylC,uBAAuB,aAAa,GACpCxnC,wBACH1D,OAAOkL,iBAAiB,UAAU,WAC5BkV,KAAK+qB,UAAY/qB,KAAKgrB,SAAWhrB,KAAKgrB,QAAUH,EAAOjtC,OACzDgtC,EAAI7pC,MAAMgxB,QAAU,QAEpB6Y,EAAI7pC,MAAMgxB,QAAU,OAEtB/R,KAAK+qB,UAAY/qB,KAAKgrB,OACxB,GACF,IAUAprC,OAAOqrC,UAAY","sources":["webpack://pydata_sphinx_theme/webpack/bootstrap","webpack://pydata_sphinx_theme/webpack/runtime/define property getters","webpack://pydata_sphinx_theme/webpack/runtime/hasOwnProperty shorthand","webpack://pydata_sphinx_theme/webpack/runtime/make namespace object","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/enums.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getNodeName.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getWindow.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/instanceOf.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/modifiers/applyStyles.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/getBasePlacement.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/math.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/userAgent.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/isLayoutViewport.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getBoundingClientRect.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getLayoutRect.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/contains.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getComputedStyle.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/isTableElement.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getDocumentElement.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getParentNode.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getOffsetParent.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/getMainAxisFromPlacement.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/within.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/mergePaddingObject.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/getFreshSideObject.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/expandToHashMap.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/modifiers/arrow.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/getVariation.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/modifiers/computeStyles.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/modifiers/eventListeners.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/getOppositePlacement.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/getOppositeVariationPlacement.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getWindowScroll.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getWindowScrollBarX.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/isScrollParent.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getScrollParent.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/listScrollParents.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/rectToClientRect.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getClippingRect.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getViewportRect.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getDocumentRect.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/computeOffsets.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/detectOverflow.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/modifiers/flip.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/computeAutoPlacement.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/modifiers/hide.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/modifiers/offset.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/modifiers/popperOffsets.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/modifiers/preventOverflow.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/getAltAxis.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getCompositeRect.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getNodeScroll.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/dom-utils/getHTMLElementScroll.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/orderModifiers.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/createPopper.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/debounce.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/utils/mergeByName.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/popper.js","webpack://pydata_sphinx_theme/./node_modules/@popperjs/core/lib/popper-lite.js","webpack://pydata_sphinx_theme/./node_modules/bootstrap/dist/js/bootstrap.esm.js","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/scripts/mixin.js","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/scripts/bootstrap.js"],"sourcesContent":["// The require scope\nvar __webpack_require__ = {};\n\n","// define getter functions for harmony exports\n__webpack_require__.d = (exports, definition) => {\n\tfor(var key in definition) {\n\t\tif(__webpack_require__.o(definition, key) && !__webpack_require__.o(exports, key)) {\n\t\t\tObject.defineProperty(exports, key, { enumerable: true, get: definition[key] });\n\t\t}\n\t}\n};","__webpack_require__.o = (obj, prop) => (Object.prototype.hasOwnProperty.call(obj, prop))","// define __esModule on exports\n__webpack_require__.r = (exports) => {\n\tif(typeof Symbol !== 'undefined' && Symbol.toStringTag) {\n\t\tObject.defineProperty(exports, Symbol.toStringTag, { value: 'Module' });\n\t}\n\tObject.defineProperty(exports, '__esModule', { value: true });\n};","export var top = 'top';\nexport var bottom = 'bottom';\nexport var right = 'right';\nexport var left = 'left';\nexport var auto = 'auto';\nexport var basePlacements = [top, bottom, right, left];\nexport var start = 'start';\nexport var end = 'end';\nexport var clippingParents = 'clippingParents';\nexport var viewport = 'viewport';\nexport var popper = 'popper';\nexport var reference = 'reference';\nexport var variationPlacements = /*#__PURE__*/basePlacements.reduce(function (acc, placement) {\n  return acc.concat([placement + \"-\" + start, placement + \"-\" + end]);\n}, []);\nexport var placements = /*#__PURE__*/[].concat(basePlacements, [auto]).reduce(function (acc, placement) {\n  return acc.concat([placement, placement + \"-\" + start, placement + \"-\" + end]);\n}, []); // modifiers that need to read the DOM\n\nexport var beforeRead = 'beforeRead';\nexport var read = 'read';\nexport var afterRead = 'afterRead'; // pure-logic modifiers\n\nexport var beforeMain = 'beforeMain';\nexport var main = 'main';\nexport var afterMain = 'afterMain'; // modifier with the purpose to write to the DOM (or write into a framework state)\n\nexport var beforeWrite = 'beforeWrite';\nexport var write = 'write';\nexport var afterWrite = 'afterWrite';\nexport var modifierPhases = [beforeRead, read, afterRead, beforeMain, main, afterMain, beforeWrite, write, afterWrite];","export default function getNodeName(element) {\n  return element ? (element.nodeName || '').toLowerCase() : null;\n}","export default function getWindow(node) {\n  if (node == null) {\n    return window;\n  }\n\n  if (node.toString() !== '[object Window]') {\n    var ownerDocument = node.ownerDocument;\n    return ownerDocument ? ownerDocument.defaultView || window : window;\n  }\n\n  return node;\n}","import getWindow from \"./getWindow.js\";\n\nfunction isElement(node) {\n  var OwnElement = getWindow(node).Element;\n  return node instanceof OwnElement || node instanceof Element;\n}\n\nfunction isHTMLElement(node) {\n  var OwnElement = getWindow(node).HTMLElement;\n  return node instanceof OwnElement || node instanceof HTMLElement;\n}\n\nfunction isShadowRoot(node) {\n  // IE 11 has no ShadowRoot\n  if (typeof ShadowRoot === 'undefined') {\n    return false;\n  }\n\n  var OwnElement = getWindow(node).ShadowRoot;\n  return node instanceof OwnElement || node instanceof ShadowRoot;\n}\n\nexport { isElement, isHTMLElement, isShadowRoot };","import getNodeName from \"../dom-utils/getNodeName.js\";\nimport { isHTMLElement } from \"../dom-utils/instanceOf.js\"; // This modifier takes the styles prepared by the `computeStyles` modifier\n// and applies them to the HTMLElements such as popper and arrow\n\nfunction applyStyles(_ref) {\n  var state = _ref.state;\n  Object.keys(state.elements).forEach(function (name) {\n    var style = state.styles[name] || {};\n    var attributes = state.attributes[name] || {};\n    var element = state.elements[name]; // arrow is optional + virtual elements\n\n    if (!isHTMLElement(element) || !getNodeName(element)) {\n      return;\n    } // Flow doesn't support to extend this property, but it's the most\n    // effective way to apply styles to an HTMLElement\n    // $FlowFixMe[cannot-write]\n\n\n    Object.assign(element.style, style);\n    Object.keys(attributes).forEach(function (name) {\n      var value = attributes[name];\n\n      if (value === false) {\n        element.removeAttribute(name);\n      } else {\n        element.setAttribute(name, value === true ? '' : value);\n      }\n    });\n  });\n}\n\nfunction effect(_ref2) {\n  var state = _ref2.state;\n  var initialStyles = {\n    popper: {\n      position: state.options.strategy,\n      left: '0',\n      top: '0',\n      margin: '0'\n    },\n    arrow: {\n      position: 'absolute'\n    },\n    reference: {}\n  };\n  Object.assign(state.elements.popper.style, initialStyles.popper);\n  state.styles = initialStyles;\n\n  if (state.elements.arrow) {\n    Object.assign(state.elements.arrow.style, initialStyles.arrow);\n  }\n\n  return function () {\n    Object.keys(state.elements).forEach(function (name) {\n      var element = state.elements[name];\n      var attributes = state.attributes[name] || {};\n      var styleProperties = Object.keys(state.styles.hasOwnProperty(name) ? state.styles[name] : initialStyles[name]); // Set all values to an empty string to unset them\n\n      var style = styleProperties.reduce(function (style, property) {\n        style[property] = '';\n        return style;\n      }, {}); // arrow is optional + virtual elements\n\n      if (!isHTMLElement(element) || !getNodeName(element)) {\n        return;\n      }\n\n      Object.assign(element.style, style);\n      Object.keys(attributes).forEach(function (attribute) {\n        element.removeAttribute(attribute);\n      });\n    });\n  };\n} // eslint-disable-next-line import/no-unused-modules\n\n\nexport default {\n  name: 'applyStyles',\n  enabled: true,\n  phase: 'write',\n  fn: applyStyles,\n  effect: effect,\n  requires: ['computeStyles']\n};","import { auto } from \"../enums.js\";\nexport default function getBasePlacement(placement) {\n  return placement.split('-')[0];\n}","export var max = Math.max;\nexport var min = Math.min;\nexport var round = Math.round;","export default function getUAString() {\n  var uaData = navigator.userAgentData;\n\n  if (uaData != null && uaData.brands && Array.isArray(uaData.brands)) {\n    return uaData.brands.map(function (item) {\n      return item.brand + \"/\" + item.version;\n    }).join(' ');\n  }\n\n  return navigator.userAgent;\n}","import getUAString from \"../utils/userAgent.js\";\nexport default function isLayoutViewport() {\n  return !/^((?!chrome|android).)*safari/i.test(getUAString());\n}","import { isElement, isHTMLElement } from \"./instanceOf.js\";\nimport { round } from \"../utils/math.js\";\nimport getWindow from \"./getWindow.js\";\nimport isLayoutViewport from \"./isLayoutViewport.js\";\nexport default function getBoundingClientRect(element, includeScale, isFixedStrategy) {\n  if (includeScale === void 0) {\n    includeScale = false;\n  }\n\n  if (isFixedStrategy === void 0) {\n    isFixedStrategy = false;\n  }\n\n  var clientRect = element.getBoundingClientRect();\n  var scaleX = 1;\n  var scaleY = 1;\n\n  if (includeScale && isHTMLElement(element)) {\n    scaleX = element.offsetWidth > 0 ? round(clientRect.width) / element.offsetWidth || 1 : 1;\n    scaleY = element.offsetHeight > 0 ? round(clientRect.height) / element.offsetHeight || 1 : 1;\n  }\n\n  var _ref = isElement(element) ? getWindow(element) : window,\n      visualViewport = _ref.visualViewport;\n\n  var addVisualOffsets = !isLayoutViewport() && isFixedStrategy;\n  var x = (clientRect.left + (addVisualOffsets && visualViewport ? visualViewport.offsetLeft : 0)) / scaleX;\n  var y = (clientRect.top + (addVisualOffsets && visualViewport ? visualViewport.offsetTop : 0)) / scaleY;\n  var width = clientRect.width / scaleX;\n  var height = clientRect.height / scaleY;\n  return {\n    width: width,\n    height: height,\n    top: y,\n    right: x + width,\n    bottom: y + height,\n    left: x,\n    x: x,\n    y: y\n  };\n}","import getBoundingClientRect from \"./getBoundingClientRect.js\"; // Returns the layout rect of an element relative to its offsetParent. Layout\n// means it doesn't take into account transforms.\n\nexport default function getLayoutRect(element) {\n  var clientRect = getBoundingClientRect(element); // Use the clientRect sizes if it's not been transformed.\n  // Fixes https://github.com/popperjs/popper-core/issues/1223\n\n  var width = element.offsetWidth;\n  var height = element.offsetHeight;\n\n  if (Math.abs(clientRect.width - width) <= 1) {\n    width = clientRect.width;\n  }\n\n  if (Math.abs(clientRect.height - height) <= 1) {\n    height = clientRect.height;\n  }\n\n  return {\n    x: element.offsetLeft,\n    y: element.offsetTop,\n    width: width,\n    height: height\n  };\n}","import { isShadowRoot } from \"./instanceOf.js\";\nexport default function contains(parent, child) {\n  var rootNode = child.getRootNode && child.getRootNode(); // First, attempt with faster native method\n\n  if (parent.contains(child)) {\n    return true;\n  } // then fallback to custom implementation with Shadow DOM support\n  else if (rootNode && isShadowRoot(rootNode)) {\n      var next = child;\n\n      do {\n        if (next && parent.isSameNode(next)) {\n          return true;\n        } // $FlowFixMe[prop-missing]: need a better way to handle this...\n\n\n        next = next.parentNode || next.host;\n      } while (next);\n    } // Give up, the result is false\n\n\n  return false;\n}","import getWindow from \"./getWindow.js\";\nexport default function getComputedStyle(element) {\n  return getWindow(element).getComputedStyle(element);\n}","import getNodeName from \"./getNodeName.js\";\nexport default function isTableElement(element) {\n  return ['table', 'td', 'th'].indexOf(getNodeName(element)) >= 0;\n}","import { isElement } from \"./instanceOf.js\";\nexport default function getDocumentElement(element) {\n  // $FlowFixMe[incompatible-return]: assume body is always available\n  return ((isElement(element) ? element.ownerDocument : // $FlowFixMe[prop-missing]\n  element.document) || window.document).documentElement;\n}","import getNodeName from \"./getNodeName.js\";\nimport getDocumentElement from \"./getDocumentElement.js\";\nimport { isShadowRoot } from \"./instanceOf.js\";\nexport default function getParentNode(element) {\n  if (getNodeName(element) === 'html') {\n    return element;\n  }\n\n  return (// this is a quicker (but less type safe) way to save quite some bytes from the bundle\n    // $FlowFixMe[incompatible-return]\n    // $FlowFixMe[prop-missing]\n    element.assignedSlot || // step into the shadow DOM of the parent of a slotted node\n    element.parentNode || ( // DOM Element detected\n    isShadowRoot(element) ? element.host : null) || // ShadowRoot detected\n    // $FlowFixMe[incompatible-call]: HTMLElement is a Node\n    getDocumentElement(element) // fallback\n\n  );\n}","import getWindow from \"./getWindow.js\";\nimport getNodeName from \"./getNodeName.js\";\nimport getComputedStyle from \"./getComputedStyle.js\";\nimport { isHTMLElement, isShadowRoot } from \"./instanceOf.js\";\nimport isTableElement from \"./isTableElement.js\";\nimport getParentNode from \"./getParentNode.js\";\nimport getUAString from \"../utils/userAgent.js\";\n\nfunction getTrueOffsetParent(element) {\n  if (!isHTMLElement(element) || // https://github.com/popperjs/popper-core/issues/837\n  getComputedStyle(element).position === 'fixed') {\n    return null;\n  }\n\n  return element.offsetParent;\n} // `.offsetParent` reports `null` for fixed elements, while absolute elements\n// return the containing block\n\n\nfunction getContainingBlock(element) {\n  var isFirefox = /firefox/i.test(getUAString());\n  var isIE = /Trident/i.test(getUAString());\n\n  if (isIE && isHTMLElement(element)) {\n    // In IE 9, 10 and 11 fixed elements containing block is always established by the viewport\n    var elementCss = getComputedStyle(element);\n\n    if (elementCss.position === 'fixed') {\n      return null;\n    }\n  }\n\n  var currentNode = getParentNode(element);\n\n  if (isShadowRoot(currentNode)) {\n    currentNode = currentNode.host;\n  }\n\n  while (isHTMLElement(currentNode) && ['html', 'body'].indexOf(getNodeName(currentNode)) < 0) {\n    var css = getComputedStyle(currentNode); // This is non-exhaustive but covers the most common CSS properties that\n    // create a containing block.\n    // https://developer.mozilla.org/en-US/docs/Web/CSS/Containing_block#identifying_the_containing_block\n\n    if (css.transform !== 'none' || css.perspective !== 'none' || css.contain === 'paint' || ['transform', 'perspective'].indexOf(css.willChange) !== -1 || isFirefox && css.willChange === 'filter' || isFirefox && css.filter && css.filter !== 'none') {\n      return currentNode;\n    } else {\n      currentNode = currentNode.parentNode;\n    }\n  }\n\n  return null;\n} // Gets the closest ancestor positioned element. Handles some edge cases,\n// such as table ancestors and cross browser bugs.\n\n\nexport default function getOffsetParent(element) {\n  var window = getWindow(element);\n  var offsetParent = getTrueOffsetParent(element);\n\n  while (offsetParent && isTableElement(offsetParent) && getComputedStyle(offsetParent).position === 'static') {\n    offsetParent = getTrueOffsetParent(offsetParent);\n  }\n\n  if (offsetParent && (getNodeName(offsetParent) === 'html' || getNodeName(offsetParent) === 'body' && getComputedStyle(offsetParent).position === 'static')) {\n    return window;\n  }\n\n  return offsetParent || getContainingBlock(element) || window;\n}","export default function getMainAxisFromPlacement(placement) {\n  return ['top', 'bottom'].indexOf(placement) >= 0 ? 'x' : 'y';\n}","import { max as mathMax, min as mathMin } from \"./math.js\";\nexport function within(min, value, max) {\n  return mathMax(min, mathMin(value, max));\n}\nexport function withinMaxClamp(min, value, max) {\n  var v = within(min, value, max);\n  return v > max ? max : v;\n}","import getFreshSideObject from \"./getFreshSideObject.js\";\nexport default function mergePaddingObject(paddingObject) {\n  return Object.assign({}, getFreshSideObject(), paddingObject);\n}","export default function getFreshSideObject() {\n  return {\n    top: 0,\n    right: 0,\n    bottom: 0,\n    left: 0\n  };\n}","export default function expandToHashMap(value, keys) {\n  return keys.reduce(function (hashMap, key) {\n    hashMap[key] = value;\n    return hashMap;\n  }, {});\n}","import getBasePlacement from \"../utils/getBasePlacement.js\";\nimport getLayoutRect from \"../dom-utils/getLayoutRect.js\";\nimport contains from \"../dom-utils/contains.js\";\nimport getOffsetParent from \"../dom-utils/getOffsetParent.js\";\nimport getMainAxisFromPlacement from \"../utils/getMainAxisFromPlacement.js\";\nimport { within } from \"../utils/within.js\";\nimport mergePaddingObject from \"../utils/mergePaddingObject.js\";\nimport expandToHashMap from \"../utils/expandToHashMap.js\";\nimport { left, right, basePlacements, top, bottom } from \"../enums.js\"; // eslint-disable-next-line import/no-unused-modules\n\nvar toPaddingObject = function toPaddingObject(padding, state) {\n  padding = typeof padding === 'function' ? padding(Object.assign({}, state.rects, {\n    placement: state.placement\n  })) : padding;\n  return mergePaddingObject(typeof padding !== 'number' ? padding : expandToHashMap(padding, basePlacements));\n};\n\nfunction arrow(_ref) {\n  var _state$modifiersData$;\n\n  var state = _ref.state,\n      name = _ref.name,\n      options = _ref.options;\n  var arrowElement = state.elements.arrow;\n  var popperOffsets = state.modifiersData.popperOffsets;\n  var basePlacement = getBasePlacement(state.placement);\n  var axis = getMainAxisFromPlacement(basePlacement);\n  var isVertical = [left, right].indexOf(basePlacement) >= 0;\n  var len = isVertical ? 'height' : 'width';\n\n  if (!arrowElement || !popperOffsets) {\n    return;\n  }\n\n  var paddingObject = toPaddingObject(options.padding, state);\n  var arrowRect = getLayoutRect(arrowElement);\n  var minProp = axis === 'y' ? top : left;\n  var maxProp = axis === 'y' ? bottom : right;\n  var endDiff = state.rects.reference[len] + state.rects.reference[axis] - popperOffsets[axis] - state.rects.popper[len];\n  var startDiff = popperOffsets[axis] - state.rects.reference[axis];\n  var arrowOffsetParent = getOffsetParent(arrowElement);\n  var clientSize = arrowOffsetParent ? axis === 'y' ? arrowOffsetParent.clientHeight || 0 : arrowOffsetParent.clientWidth || 0 : 0;\n  var centerToReference = endDiff / 2 - startDiff / 2; // Make sure the arrow doesn't overflow the popper if the center point is\n  // outside of the popper bounds\n\n  var min = paddingObject[minProp];\n  var max = clientSize - arrowRect[len] - paddingObject[maxProp];\n  var center = clientSize / 2 - arrowRect[len] / 2 + centerToReference;\n  var offset = within(min, center, max); // Prevents breaking syntax highlighting...\n\n  var axisProp = axis;\n  state.modifiersData[name] = (_state$modifiersData$ = {}, _state$modifiersData$[axisProp] = offset, _state$modifiersData$.centerOffset = offset - center, _state$modifiersData$);\n}\n\nfunction effect(_ref2) {\n  var state = _ref2.state,\n      options = _ref2.options;\n  var _options$element = options.element,\n      arrowElement = _options$element === void 0 ? '[data-popper-arrow]' : _options$element;\n\n  if (arrowElement == null) {\n    return;\n  } // CSS selector\n\n\n  if (typeof arrowElement === 'string') {\n    arrowElement = state.elements.popper.querySelector(arrowElement);\n\n    if (!arrowElement) {\n      return;\n    }\n  }\n\n  if (!contains(state.elements.popper, arrowElement)) {\n    return;\n  }\n\n  state.elements.arrow = arrowElement;\n} // eslint-disable-next-line import/no-unused-modules\n\n\nexport default {\n  name: 'arrow',\n  enabled: true,\n  phase: 'main',\n  fn: arrow,\n  effect: effect,\n  requires: ['popperOffsets'],\n  requiresIfExists: ['preventOverflow']\n};","export default function getVariation(placement) {\n  return placement.split('-')[1];\n}","import { top, left, right, bottom, end } from \"../enums.js\";\nimport getOffsetParent from \"../dom-utils/getOffsetParent.js\";\nimport getWindow from \"../dom-utils/getWindow.js\";\nimport getDocumentElement from \"../dom-utils/getDocumentElement.js\";\nimport getComputedStyle from \"../dom-utils/getComputedStyle.js\";\nimport getBasePlacement from \"../utils/getBasePlacement.js\";\nimport getVariation from \"../utils/getVariation.js\";\nimport { round } from \"../utils/math.js\"; // eslint-disable-next-line import/no-unused-modules\n\nvar unsetSides = {\n  top: 'auto',\n  right: 'auto',\n  bottom: 'auto',\n  left: 'auto'\n}; // Round the offsets to the nearest suitable subpixel based on the DPR.\n// Zooming can change the DPR, but it seems to report a value that will\n// cleanly divide the values into the appropriate subpixels.\n\nfunction roundOffsetsByDPR(_ref, win) {\n  var x = _ref.x,\n      y = _ref.y;\n  var dpr = win.devicePixelRatio || 1;\n  return {\n    x: round(x * dpr) / dpr || 0,\n    y: round(y * dpr) / dpr || 0\n  };\n}\n\nexport function mapToStyles(_ref2) {\n  var _Object$assign2;\n\n  var popper = _ref2.popper,\n      popperRect = _ref2.popperRect,\n      placement = _ref2.placement,\n      variation = _ref2.variation,\n      offsets = _ref2.offsets,\n      position = _ref2.position,\n      gpuAcceleration = _ref2.gpuAcceleration,\n      adaptive = _ref2.adaptive,\n      roundOffsets = _ref2.roundOffsets,\n      isFixed = _ref2.isFixed;\n  var _offsets$x = offsets.x,\n      x = _offsets$x === void 0 ? 0 : _offsets$x,\n      _offsets$y = offsets.y,\n      y = _offsets$y === void 0 ? 0 : _offsets$y;\n\n  var _ref3 = typeof roundOffsets === 'function' ? roundOffsets({\n    x: x,\n    y: y\n  }) : {\n    x: x,\n    y: y\n  };\n\n  x = _ref3.x;\n  y = _ref3.y;\n  var hasX = offsets.hasOwnProperty('x');\n  var hasY = offsets.hasOwnProperty('y');\n  var sideX = left;\n  var sideY = top;\n  var win = window;\n\n  if (adaptive) {\n    var offsetParent = getOffsetParent(popper);\n    var heightProp = 'clientHeight';\n    var widthProp = 'clientWidth';\n\n    if (offsetParent === getWindow(popper)) {\n      offsetParent = getDocumentElement(popper);\n\n      if (getComputedStyle(offsetParent).position !== 'static' && position === 'absolute') {\n        heightProp = 'scrollHeight';\n        widthProp = 'scrollWidth';\n      }\n    } // $FlowFixMe[incompatible-cast]: force type refinement, we compare offsetParent with window above, but Flow doesn't detect it\n\n\n    offsetParent = offsetParent;\n\n    if (placement === top || (placement === left || placement === right) && variation === end) {\n      sideY = bottom;\n      var offsetY = isFixed && offsetParent === win && win.visualViewport ? win.visualViewport.height : // $FlowFixMe[prop-missing]\n      offsetParent[heightProp];\n      y -= offsetY - popperRect.height;\n      y *= gpuAcceleration ? 1 : -1;\n    }\n\n    if (placement === left || (placement === top || placement === bottom) && variation === end) {\n      sideX = right;\n      var offsetX = isFixed && offsetParent === win && win.visualViewport ? win.visualViewport.width : // $FlowFixMe[prop-missing]\n      offsetParent[widthProp];\n      x -= offsetX - popperRect.width;\n      x *= gpuAcceleration ? 1 : -1;\n    }\n  }\n\n  var commonStyles = Object.assign({\n    position: position\n  }, adaptive && unsetSides);\n\n  var _ref4 = roundOffsets === true ? roundOffsetsByDPR({\n    x: x,\n    y: y\n  }, getWindow(popper)) : {\n    x: x,\n    y: y\n  };\n\n  x = _ref4.x;\n  y = _ref4.y;\n\n  if (gpuAcceleration) {\n    var _Object$assign;\n\n    return Object.assign({}, commonStyles, (_Object$assign = {}, _Object$assign[sideY] = hasY ? '0' : '', _Object$assign[sideX] = hasX ? '0' : '', _Object$assign.transform = (win.devicePixelRatio || 1) <= 1 ? \"translate(\" + x + \"px, \" + y + \"px)\" : \"translate3d(\" + x + \"px, \" + y + \"px, 0)\", _Object$assign));\n  }\n\n  return Object.assign({}, commonStyles, (_Object$assign2 = {}, _Object$assign2[sideY] = hasY ? y + \"px\" : '', _Object$assign2[sideX] = hasX ? x + \"px\" : '', _Object$assign2.transform = '', _Object$assign2));\n}\n\nfunction computeStyles(_ref5) {\n  var state = _ref5.state,\n      options = _ref5.options;\n  var _options$gpuAccelerat = options.gpuAcceleration,\n      gpuAcceleration = _options$gpuAccelerat === void 0 ? true : _options$gpuAccelerat,\n      _options$adaptive = options.adaptive,\n      adaptive = _options$adaptive === void 0 ? true : _options$adaptive,\n      _options$roundOffsets = options.roundOffsets,\n      roundOffsets = _options$roundOffsets === void 0 ? true : _options$roundOffsets;\n  var commonStyles = {\n    placement: getBasePlacement(state.placement),\n    variation: getVariation(state.placement),\n    popper: state.elements.popper,\n    popperRect: state.rects.popper,\n    gpuAcceleration: gpuAcceleration,\n    isFixed: state.options.strategy === 'fixed'\n  };\n\n  if (state.modifiersData.popperOffsets != null) {\n    state.styles.popper = Object.assign({}, state.styles.popper, mapToStyles(Object.assign({}, commonStyles, {\n      offsets: state.modifiersData.popperOffsets,\n      position: state.options.strategy,\n      adaptive: adaptive,\n      roundOffsets: roundOffsets\n    })));\n  }\n\n  if (state.modifiersData.arrow != null) {\n    state.styles.arrow = Object.assign({}, state.styles.arrow, mapToStyles(Object.assign({}, commonStyles, {\n      offsets: state.modifiersData.arrow,\n      position: 'absolute',\n      adaptive: false,\n      roundOffsets: roundOffsets\n    })));\n  }\n\n  state.attributes.popper = Object.assign({}, state.attributes.popper, {\n    'data-popper-placement': state.placement\n  });\n} // eslint-disable-next-line import/no-unused-modules\n\n\nexport default {\n  name: 'computeStyles',\n  enabled: true,\n  phase: 'beforeWrite',\n  fn: computeStyles,\n  data: {}\n};","import getWindow from \"../dom-utils/getWindow.js\"; // eslint-disable-next-line import/no-unused-modules\n\nvar passive = {\n  passive: true\n};\n\nfunction effect(_ref) {\n  var state = _ref.state,\n      instance = _ref.instance,\n      options = _ref.options;\n  var _options$scroll = options.scroll,\n      scroll = _options$scroll === void 0 ? true : _options$scroll,\n      _options$resize = options.resize,\n      resize = _options$resize === void 0 ? true : _options$resize;\n  var window = getWindow(state.elements.popper);\n  var scrollParents = [].concat(state.scrollParents.reference, state.scrollParents.popper);\n\n  if (scroll) {\n    scrollParents.forEach(function (scrollParent) {\n      scrollParent.addEventListener('scroll', instance.update, passive);\n    });\n  }\n\n  if (resize) {\n    window.addEventListener('resize', instance.update, passive);\n  }\n\n  return function () {\n    if (scroll) {\n      scrollParents.forEach(function (scrollParent) {\n        scrollParent.removeEventListener('scroll', instance.update, passive);\n      });\n    }\n\n    if (resize) {\n      window.removeEventListener('resize', instance.update, passive);\n    }\n  };\n} // eslint-disable-next-line import/no-unused-modules\n\n\nexport default {\n  name: 'eventListeners',\n  enabled: true,\n  phase: 'write',\n  fn: function fn() {},\n  effect: effect,\n  data: {}\n};","var hash = {\n  left: 'right',\n  right: 'left',\n  bottom: 'top',\n  top: 'bottom'\n};\nexport default function getOppositePlacement(placement) {\n  return placement.replace(/left|right|bottom|top/g, function (matched) {\n    return hash[matched];\n  });\n}","var hash = {\n  start: 'end',\n  end: 'start'\n};\nexport default function getOppositeVariationPlacement(placement) {\n  return placement.replace(/start|end/g, function (matched) {\n    return hash[matched];\n  });\n}","import getWindow from \"./getWindow.js\";\nexport default function getWindowScroll(node) {\n  var win = getWindow(node);\n  var scrollLeft = win.pageXOffset;\n  var scrollTop = win.pageYOffset;\n  return {\n    scrollLeft: scrollLeft,\n    scrollTop: scrollTop\n  };\n}","import getBoundingClientRect from \"./getBoundingClientRect.js\";\nimport getDocumentElement from \"./getDocumentElement.js\";\nimport getWindowScroll from \"./getWindowScroll.js\";\nexport default function getWindowScrollBarX(element) {\n  // If <html> has a CSS width greater than the viewport, then this will be\n  // incorrect for RTL.\n  // Popper 1 is broken in this case and never had a bug report so let's assume\n  // it's not an issue. I don't think anyone ever specifies width on <html>\n  // anyway.\n  // Browsers where the left scrollbar doesn't cause an issue report `0` for\n  // this (e.g. Edge 2019, IE11, Safari)\n  return getBoundingClientRect(getDocumentElement(element)).left + getWindowScroll(element).scrollLeft;\n}","import getComputedStyle from \"./getComputedStyle.js\";\nexport default function isScrollParent(element) {\n  // Firefox wants us to check `-x` and `-y` variations as well\n  var _getComputedStyle = getComputedStyle(element),\n      overflow = _getComputedStyle.overflow,\n      overflowX = _getComputedStyle.overflowX,\n      overflowY = _getComputedStyle.overflowY;\n\n  return /auto|scroll|overlay|hidden/.test(overflow + overflowY + overflowX);\n}","import getParentNode from \"./getParentNode.js\";\nimport isScrollParent from \"./isScrollParent.js\";\nimport getNodeName from \"./getNodeName.js\";\nimport { isHTMLElement } from \"./instanceOf.js\";\nexport default function getScrollParent(node) {\n  if (['html', 'body', '#document'].indexOf(getNodeName(node)) >= 0) {\n    // $FlowFixMe[incompatible-return]: assume body is always available\n    return node.ownerDocument.body;\n  }\n\n  if (isHTMLElement(node) && isScrollParent(node)) {\n    return node;\n  }\n\n  return getScrollParent(getParentNode(node));\n}","import getScrollParent from \"./getScrollParent.js\";\nimport getParentNode from \"./getParentNode.js\";\nimport getWindow from \"./getWindow.js\";\nimport isScrollParent from \"./isScrollParent.js\";\n/*\ngiven a DOM element, return the list of all scroll parents, up the list of ancesors\nuntil we get to the top window object. This list is what we attach scroll listeners\nto, because if any of these parent elements scroll, we'll need to re-calculate the\nreference element's position.\n*/\n\nexport default function listScrollParents(element, list) {\n  var _element$ownerDocumen;\n\n  if (list === void 0) {\n    list = [];\n  }\n\n  var scrollParent = getScrollParent(element);\n  var isBody = scrollParent === ((_element$ownerDocumen = element.ownerDocument) == null ? void 0 : _element$ownerDocumen.body);\n  var win = getWindow(scrollParent);\n  var target = isBody ? [win].concat(win.visualViewport || [], isScrollParent(scrollParent) ? scrollParent : []) : scrollParent;\n  var updatedList = list.concat(target);\n  return isBody ? updatedList : // $FlowFixMe[incompatible-call]: isBody tells us target will be an HTMLElement here\n  updatedList.concat(listScrollParents(getParentNode(target)));\n}","export default function rectToClientRect(rect) {\n  return Object.assign({}, rect, {\n    left: rect.x,\n    top: rect.y,\n    right: rect.x + rect.width,\n    bottom: rect.y + rect.height\n  });\n}","import { viewport } from \"../enums.js\";\nimport getViewportRect from \"./getViewportRect.js\";\nimport getDocumentRect from \"./getDocumentRect.js\";\nimport listScrollParents from \"./listScrollParents.js\";\nimport getOffsetParent from \"./getOffsetParent.js\";\nimport getDocumentElement from \"./getDocumentElement.js\";\nimport getComputedStyle from \"./getComputedStyle.js\";\nimport { isElement, isHTMLElement } from \"./instanceOf.js\";\nimport getBoundingClientRect from \"./getBoundingClientRect.js\";\nimport getParentNode from \"./getParentNode.js\";\nimport contains from \"./contains.js\";\nimport getNodeName from \"./getNodeName.js\";\nimport rectToClientRect from \"../utils/rectToClientRect.js\";\nimport { max, min } from \"../utils/math.js\";\n\nfunction getInnerBoundingClientRect(element, strategy) {\n  var rect = getBoundingClientRect(element, false, strategy === 'fixed');\n  rect.top = rect.top + element.clientTop;\n  rect.left = rect.left + element.clientLeft;\n  rect.bottom = rect.top + element.clientHeight;\n  rect.right = rect.left + element.clientWidth;\n  rect.width = element.clientWidth;\n  rect.height = element.clientHeight;\n  rect.x = rect.left;\n  rect.y = rect.top;\n  return rect;\n}\n\nfunction getClientRectFromMixedType(element, clippingParent, strategy) {\n  return clippingParent === viewport ? rectToClientRect(getViewportRect(element, strategy)) : isElement(clippingParent) ? getInnerBoundingClientRect(clippingParent, strategy) : rectToClientRect(getDocumentRect(getDocumentElement(element)));\n} // A \"clipping parent\" is an overflowable container with the characteristic of\n// clipping (or hiding) overflowing elements with a position different from\n// `initial`\n\n\nfunction getClippingParents(element) {\n  var clippingParents = listScrollParents(getParentNode(element));\n  var canEscapeClipping = ['absolute', 'fixed'].indexOf(getComputedStyle(element).position) >= 0;\n  var clipperElement = canEscapeClipping && isHTMLElement(element) ? getOffsetParent(element) : element;\n\n  if (!isElement(clipperElement)) {\n    return [];\n  } // $FlowFixMe[incompatible-return]: https://github.com/facebook/flow/issues/1414\n\n\n  return clippingParents.filter(function (clippingParent) {\n    return isElement(clippingParent) && contains(clippingParent, clipperElement) && getNodeName(clippingParent) !== 'body';\n  });\n} // Gets the maximum area that the element is visible in due to any number of\n// clipping parents\n\n\nexport default function getClippingRect(element, boundary, rootBoundary, strategy) {\n  var mainClippingParents = boundary === 'clippingParents' ? getClippingParents(element) : [].concat(boundary);\n  var clippingParents = [].concat(mainClippingParents, [rootBoundary]);\n  var firstClippingParent = clippingParents[0];\n  var clippingRect = clippingParents.reduce(function (accRect, clippingParent) {\n    var rect = getClientRectFromMixedType(element, clippingParent, strategy);\n    accRect.top = max(rect.top, accRect.top);\n    accRect.right = min(rect.right, accRect.right);\n    accRect.bottom = min(rect.bottom, accRect.bottom);\n    accRect.left = max(rect.left, accRect.left);\n    return accRect;\n  }, getClientRectFromMixedType(element, firstClippingParent, strategy));\n  clippingRect.width = clippingRect.right - clippingRect.left;\n  clippingRect.height = clippingRect.bottom - clippingRect.top;\n  clippingRect.x = clippingRect.left;\n  clippingRect.y = clippingRect.top;\n  return clippingRect;\n}","import getWindow from \"./getWindow.js\";\nimport getDocumentElement from \"./getDocumentElement.js\";\nimport getWindowScrollBarX from \"./getWindowScrollBarX.js\";\nimport isLayoutViewport from \"./isLayoutViewport.js\";\nexport default function getViewportRect(element, strategy) {\n  var win = getWindow(element);\n  var html = getDocumentElement(element);\n  var visualViewport = win.visualViewport;\n  var width = html.clientWidth;\n  var height = html.clientHeight;\n  var x = 0;\n  var y = 0;\n\n  if (visualViewport) {\n    width = visualViewport.width;\n    height = visualViewport.height;\n    var layoutViewport = isLayoutViewport();\n\n    if (layoutViewport || !layoutViewport && strategy === 'fixed') {\n      x = visualViewport.offsetLeft;\n      y = visualViewport.offsetTop;\n    }\n  }\n\n  return {\n    width: width,\n    height: height,\n    x: x + getWindowScrollBarX(element),\n    y: y\n  };\n}","import getDocumentElement from \"./getDocumentElement.js\";\nimport getComputedStyle from \"./getComputedStyle.js\";\nimport getWindowScrollBarX from \"./getWindowScrollBarX.js\";\nimport getWindowScroll from \"./getWindowScroll.js\";\nimport { max } from \"../utils/math.js\"; // Gets the entire size of the scrollable document area, even extending outside\n// of the `<html>` and `<body>` rect bounds if horizontally scrollable\n\nexport default function getDocumentRect(element) {\n  var _element$ownerDocumen;\n\n  var html = getDocumentElement(element);\n  var winScroll = getWindowScroll(element);\n  var body = (_element$ownerDocumen = element.ownerDocument) == null ? void 0 : _element$ownerDocumen.body;\n  var width = max(html.scrollWidth, html.clientWidth, body ? body.scrollWidth : 0, body ? body.clientWidth : 0);\n  var height = max(html.scrollHeight, html.clientHeight, body ? body.scrollHeight : 0, body ? body.clientHeight : 0);\n  var x = -winScroll.scrollLeft + getWindowScrollBarX(element);\n  var y = -winScroll.scrollTop;\n\n  if (getComputedStyle(body || html).direction === 'rtl') {\n    x += max(html.clientWidth, body ? body.clientWidth : 0) - width;\n  }\n\n  return {\n    width: width,\n    height: height,\n    x: x,\n    y: y\n  };\n}","import getBasePlacement from \"./getBasePlacement.js\";\nimport getVariation from \"./getVariation.js\";\nimport getMainAxisFromPlacement from \"./getMainAxisFromPlacement.js\";\nimport { top, right, bottom, left, start, end } from \"../enums.js\";\nexport default function computeOffsets(_ref) {\n  var reference = _ref.reference,\n      element = _ref.element,\n      placement = _ref.placement;\n  var basePlacement = placement ? getBasePlacement(placement) : null;\n  var variation = placement ? getVariation(placement) : null;\n  var commonX = reference.x + reference.width / 2 - element.width / 2;\n  var commonY = reference.y + reference.height / 2 - element.height / 2;\n  var offsets;\n\n  switch (basePlacement) {\n    case top:\n      offsets = {\n        x: commonX,\n        y: reference.y - element.height\n      };\n      break;\n\n    case bottom:\n      offsets = {\n        x: commonX,\n        y: reference.y + reference.height\n      };\n      break;\n\n    case right:\n      offsets = {\n        x: reference.x + reference.width,\n        y: commonY\n      };\n      break;\n\n    case left:\n      offsets = {\n        x: reference.x - element.width,\n        y: commonY\n      };\n      break;\n\n    default:\n      offsets = {\n        x: reference.x,\n        y: reference.y\n      };\n  }\n\n  var mainAxis = basePlacement ? getMainAxisFromPlacement(basePlacement) : null;\n\n  if (mainAxis != null) {\n    var len = mainAxis === 'y' ? 'height' : 'width';\n\n    switch (variation) {\n      case start:\n        offsets[mainAxis] = offsets[mainAxis] - (reference[len] / 2 - element[len] / 2);\n        break;\n\n      case end:\n        offsets[mainAxis] = offsets[mainAxis] + (reference[len] / 2 - element[len] / 2);\n        break;\n\n      default:\n    }\n  }\n\n  return offsets;\n}","import getClippingRect from \"../dom-utils/getClippingRect.js\";\nimport getDocumentElement from \"../dom-utils/getDocumentElement.js\";\nimport getBoundingClientRect from \"../dom-utils/getBoundingClientRect.js\";\nimport computeOffsets from \"./computeOffsets.js\";\nimport rectToClientRect from \"./rectToClientRect.js\";\nimport { clippingParents, reference, popper, bottom, top, right, basePlacements, viewport } from \"../enums.js\";\nimport { isElement } from \"../dom-utils/instanceOf.js\";\nimport mergePaddingObject from \"./mergePaddingObject.js\";\nimport expandToHashMap from \"./expandToHashMap.js\"; // eslint-disable-next-line import/no-unused-modules\n\nexport default function detectOverflow(state, options) {\n  if (options === void 0) {\n    options = {};\n  }\n\n  var _options = options,\n      _options$placement = _options.placement,\n      placement = _options$placement === void 0 ? state.placement : _options$placement,\n      _options$strategy = _options.strategy,\n      strategy = _options$strategy === void 0 ? state.strategy : _options$strategy,\n      _options$boundary = _options.boundary,\n      boundary = _options$boundary === void 0 ? clippingParents : _options$boundary,\n      _options$rootBoundary = _options.rootBoundary,\n      rootBoundary = _options$rootBoundary === void 0 ? viewport : _options$rootBoundary,\n      _options$elementConte = _options.elementContext,\n      elementContext = _options$elementConte === void 0 ? popper : _options$elementConte,\n      _options$altBoundary = _options.altBoundary,\n      altBoundary = _options$altBoundary === void 0 ? false : _options$altBoundary,\n      _options$padding = _options.padding,\n      padding = _options$padding === void 0 ? 0 : _options$padding;\n  var paddingObject = mergePaddingObject(typeof padding !== 'number' ? padding : expandToHashMap(padding, basePlacements));\n  var altContext = elementContext === popper ? reference : popper;\n  var popperRect = state.rects.popper;\n  var element = state.elements[altBoundary ? altContext : elementContext];\n  var clippingClientRect = getClippingRect(isElement(element) ? element : element.contextElement || getDocumentElement(state.elements.popper), boundary, rootBoundary, strategy);\n  var referenceClientRect = getBoundingClientRect(state.elements.reference);\n  var popperOffsets = computeOffsets({\n    reference: referenceClientRect,\n    element: popperRect,\n    strategy: 'absolute',\n    placement: placement\n  });\n  var popperClientRect = rectToClientRect(Object.assign({}, popperRect, popperOffsets));\n  var elementClientRect = elementContext === popper ? popperClientRect : referenceClientRect; // positive = overflowing the clipping rect\n  // 0 or negative = within the clipping rect\n\n  var overflowOffsets = {\n    top: clippingClientRect.top - elementClientRect.top + paddingObject.top,\n    bottom: elementClientRect.bottom - clippingClientRect.bottom + paddingObject.bottom,\n    left: clippingClientRect.left - elementClientRect.left + paddingObject.left,\n    right: elementClientRect.right - clippingClientRect.right + paddingObject.right\n  };\n  var offsetData = state.modifiersData.offset; // Offsets can be applied only to the popper element\n\n  if (elementContext === popper && offsetData) {\n    var offset = offsetData[placement];\n    Object.keys(overflowOffsets).forEach(function (key) {\n      var multiply = [right, bottom].indexOf(key) >= 0 ? 1 : -1;\n      var axis = [top, bottom].indexOf(key) >= 0 ? 'y' : 'x';\n      overflowOffsets[key] += offset[axis] * multiply;\n    });\n  }\n\n  return overflowOffsets;\n}","import getOppositePlacement from \"../utils/getOppositePlacement.js\";\nimport getBasePlacement from \"../utils/getBasePlacement.js\";\nimport getOppositeVariationPlacement from \"../utils/getOppositeVariationPlacement.js\";\nimport detectOverflow from \"../utils/detectOverflow.js\";\nimport computeAutoPlacement from \"../utils/computeAutoPlacement.js\";\nimport { bottom, top, start, right, left, auto } from \"../enums.js\";\nimport getVariation from \"../utils/getVariation.js\"; // eslint-disable-next-line import/no-unused-modules\n\nfunction getExpandedFallbackPlacements(placement) {\n  if (getBasePlacement(placement) === auto) {\n    return [];\n  }\n\n  var oppositePlacement = getOppositePlacement(placement);\n  return [getOppositeVariationPlacement(placement), oppositePlacement, getOppositeVariationPlacement(oppositePlacement)];\n}\n\nfunction flip(_ref) {\n  var state = _ref.state,\n      options = _ref.options,\n      name = _ref.name;\n\n  if (state.modifiersData[name]._skip) {\n    return;\n  }\n\n  var _options$mainAxis = options.mainAxis,\n      checkMainAxis = _options$mainAxis === void 0 ? true : _options$mainAxis,\n      _options$altAxis = options.altAxis,\n      checkAltAxis = _options$altAxis === void 0 ? true : _options$altAxis,\n      specifiedFallbackPlacements = options.fallbackPlacements,\n      padding = options.padding,\n      boundary = options.boundary,\n      rootBoundary = options.rootBoundary,\n      altBoundary = options.altBoundary,\n      _options$flipVariatio = options.flipVariations,\n      flipVariations = _options$flipVariatio === void 0 ? true : _options$flipVariatio,\n      allowedAutoPlacements = options.allowedAutoPlacements;\n  var preferredPlacement = state.options.placement;\n  var basePlacement = getBasePlacement(preferredPlacement);\n  var isBasePlacement = basePlacement === preferredPlacement;\n  var fallbackPlacements = specifiedFallbackPlacements || (isBasePlacement || !flipVariations ? [getOppositePlacement(preferredPlacement)] : getExpandedFallbackPlacements(preferredPlacement));\n  var placements = [preferredPlacement].concat(fallbackPlacements).reduce(function (acc, placement) {\n    return acc.concat(getBasePlacement(placement) === auto ? computeAutoPlacement(state, {\n      placement: placement,\n      boundary: boundary,\n      rootBoundary: rootBoundary,\n      padding: padding,\n      flipVariations: flipVariations,\n      allowedAutoPlacements: allowedAutoPlacements\n    }) : placement);\n  }, []);\n  var referenceRect = state.rects.reference;\n  var popperRect = state.rects.popper;\n  var checksMap = new Map();\n  var makeFallbackChecks = true;\n  var firstFittingPlacement = placements[0];\n\n  for (var i = 0; i < placements.length; i++) {\n    var placement = placements[i];\n\n    var _basePlacement = getBasePlacement(placement);\n\n    var isStartVariation = getVariation(placement) === start;\n    var isVertical = [top, bottom].indexOf(_basePlacement) >= 0;\n    var len = isVertical ? 'width' : 'height';\n    var overflow = detectOverflow(state, {\n      placement: placement,\n      boundary: boundary,\n      rootBoundary: rootBoundary,\n      altBoundary: altBoundary,\n      padding: padding\n    });\n    var mainVariationSide = isVertical ? isStartVariation ? right : left : isStartVariation ? bottom : top;\n\n    if (referenceRect[len] > popperRect[len]) {\n      mainVariationSide = getOppositePlacement(mainVariationSide);\n    }\n\n    var altVariationSide = getOppositePlacement(mainVariationSide);\n    var checks = [];\n\n    if (checkMainAxis) {\n      checks.push(overflow[_basePlacement] <= 0);\n    }\n\n    if (checkAltAxis) {\n      checks.push(overflow[mainVariationSide] <= 0, overflow[altVariationSide] <= 0);\n    }\n\n    if (checks.every(function (check) {\n      return check;\n    })) {\n      firstFittingPlacement = placement;\n      makeFallbackChecks = false;\n      break;\n    }\n\n    checksMap.set(placement, checks);\n  }\n\n  if (makeFallbackChecks) {\n    // `2` may be desired in some cases – research later\n    var numberOfChecks = flipVariations ? 3 : 1;\n\n    var _loop = function _loop(_i) {\n      var fittingPlacement = placements.find(function (placement) {\n        var checks = checksMap.get(placement);\n\n        if (checks) {\n          return checks.slice(0, _i).every(function (check) {\n            return check;\n          });\n        }\n      });\n\n      if (fittingPlacement) {\n        firstFittingPlacement = fittingPlacement;\n        return \"break\";\n      }\n    };\n\n    for (var _i = numberOfChecks; _i > 0; _i--) {\n      var _ret = _loop(_i);\n\n      if (_ret === \"break\") break;\n    }\n  }\n\n  if (state.placement !== firstFittingPlacement) {\n    state.modifiersData[name]._skip = true;\n    state.placement = firstFittingPlacement;\n    state.reset = true;\n  }\n} // eslint-disable-next-line import/no-unused-modules\n\n\nexport default {\n  name: 'flip',\n  enabled: true,\n  phase: 'main',\n  fn: flip,\n  requiresIfExists: ['offset'],\n  data: {\n    _skip: false\n  }\n};","import getVariation from \"./getVariation.js\";\nimport { variationPlacements, basePlacements, placements as allPlacements } from \"../enums.js\";\nimport detectOverflow from \"./detectOverflow.js\";\nimport getBasePlacement from \"./getBasePlacement.js\";\nexport default function computeAutoPlacement(state, options) {\n  if (options === void 0) {\n    options = {};\n  }\n\n  var _options = options,\n      placement = _options.placement,\n      boundary = _options.boundary,\n      rootBoundary = _options.rootBoundary,\n      padding = _options.padding,\n      flipVariations = _options.flipVariations,\n      _options$allowedAutoP = _options.allowedAutoPlacements,\n      allowedAutoPlacements = _options$allowedAutoP === void 0 ? allPlacements : _options$allowedAutoP;\n  var variation = getVariation(placement);\n  var placements = variation ? flipVariations ? variationPlacements : variationPlacements.filter(function (placement) {\n    return getVariation(placement) === variation;\n  }) : basePlacements;\n  var allowedPlacements = placements.filter(function (placement) {\n    return allowedAutoPlacements.indexOf(placement) >= 0;\n  });\n\n  if (allowedPlacements.length === 0) {\n    allowedPlacements = placements;\n  } // $FlowFixMe[incompatible-type]: Flow seems to have problems with two array unions...\n\n\n  var overflows = allowedPlacements.reduce(function (acc, placement) {\n    acc[placement] = detectOverflow(state, {\n      placement: placement,\n      boundary: boundary,\n      rootBoundary: rootBoundary,\n      padding: padding\n    })[getBasePlacement(placement)];\n    return acc;\n  }, {});\n  return Object.keys(overflows).sort(function (a, b) {\n    return overflows[a] - overflows[b];\n  });\n}","import { top, bottom, left, right } from \"../enums.js\";\nimport detectOverflow from \"../utils/detectOverflow.js\";\n\nfunction getSideOffsets(overflow, rect, preventedOffsets) {\n  if (preventedOffsets === void 0) {\n    preventedOffsets = {\n      x: 0,\n      y: 0\n    };\n  }\n\n  return {\n    top: overflow.top - rect.height - preventedOffsets.y,\n    right: overflow.right - rect.width + preventedOffsets.x,\n    bottom: overflow.bottom - rect.height + preventedOffsets.y,\n    left: overflow.left - rect.width - preventedOffsets.x\n  };\n}\n\nfunction isAnySideFullyClipped(overflow) {\n  return [top, right, bottom, left].some(function (side) {\n    return overflow[side] >= 0;\n  });\n}\n\nfunction hide(_ref) {\n  var state = _ref.state,\n      name = _ref.name;\n  var referenceRect = state.rects.reference;\n  var popperRect = state.rects.popper;\n  var preventedOffsets = state.modifiersData.preventOverflow;\n  var referenceOverflow = detectOverflow(state, {\n    elementContext: 'reference'\n  });\n  var popperAltOverflow = detectOverflow(state, {\n    altBoundary: true\n  });\n  var referenceClippingOffsets = getSideOffsets(referenceOverflow, referenceRect);\n  var popperEscapeOffsets = getSideOffsets(popperAltOverflow, popperRect, preventedOffsets);\n  var isReferenceHidden = isAnySideFullyClipped(referenceClippingOffsets);\n  var hasPopperEscaped = isAnySideFullyClipped(popperEscapeOffsets);\n  state.modifiersData[name] = {\n    referenceClippingOffsets: referenceClippingOffsets,\n    popperEscapeOffsets: popperEscapeOffsets,\n    isReferenceHidden: isReferenceHidden,\n    hasPopperEscaped: hasPopperEscaped\n  };\n  state.attributes.popper = Object.assign({}, state.attributes.popper, {\n    'data-popper-reference-hidden': isReferenceHidden,\n    'data-popper-escaped': hasPopperEscaped\n  });\n} // eslint-disable-next-line import/no-unused-modules\n\n\nexport default {\n  name: 'hide',\n  enabled: true,\n  phase: 'main',\n  requiresIfExists: ['preventOverflow'],\n  fn: hide\n};","import getBasePlacement from \"../utils/getBasePlacement.js\";\nimport { top, left, right, placements } from \"../enums.js\"; // eslint-disable-next-line import/no-unused-modules\n\nexport function distanceAndSkiddingToXY(placement, rects, offset) {\n  var basePlacement = getBasePlacement(placement);\n  var invertDistance = [left, top].indexOf(basePlacement) >= 0 ? -1 : 1;\n\n  var _ref = typeof offset === 'function' ? offset(Object.assign({}, rects, {\n    placement: placement\n  })) : offset,\n      skidding = _ref[0],\n      distance = _ref[1];\n\n  skidding = skidding || 0;\n  distance = (distance || 0) * invertDistance;\n  return [left, right].indexOf(basePlacement) >= 0 ? {\n    x: distance,\n    y: skidding\n  } : {\n    x: skidding,\n    y: distance\n  };\n}\n\nfunction offset(_ref2) {\n  var state = _ref2.state,\n      options = _ref2.options,\n      name = _ref2.name;\n  var _options$offset = options.offset,\n      offset = _options$offset === void 0 ? [0, 0] : _options$offset;\n  var data = placements.reduce(function (acc, placement) {\n    acc[placement] = distanceAndSkiddingToXY(placement, state.rects, offset);\n    return acc;\n  }, {});\n  var _data$state$placement = data[state.placement],\n      x = _data$state$placement.x,\n      y = _data$state$placement.y;\n\n  if (state.modifiersData.popperOffsets != null) {\n    state.modifiersData.popperOffsets.x += x;\n    state.modifiersData.popperOffsets.y += y;\n  }\n\n  state.modifiersData[name] = data;\n} // eslint-disable-next-line import/no-unused-modules\n\n\nexport default {\n  name: 'offset',\n  enabled: true,\n  phase: 'main',\n  requires: ['popperOffsets'],\n  fn: offset\n};","import computeOffsets from \"../utils/computeOffsets.js\";\n\nfunction popperOffsets(_ref) {\n  var state = _ref.state,\n      name = _ref.name;\n  // Offsets are the actual position the popper needs to have to be\n  // properly positioned near its reference element\n  // This is the most basic placement, and will be adjusted by\n  // the modifiers in the next step\n  state.modifiersData[name] = computeOffsets({\n    reference: state.rects.reference,\n    element: state.rects.popper,\n    strategy: 'absolute',\n    placement: state.placement\n  });\n} // eslint-disable-next-line import/no-unused-modules\n\n\nexport default {\n  name: 'popperOffsets',\n  enabled: true,\n  phase: 'read',\n  fn: popperOffsets,\n  data: {}\n};","import { top, left, right, bottom, start } from \"../enums.js\";\nimport getBasePlacement from \"../utils/getBasePlacement.js\";\nimport getMainAxisFromPlacement from \"../utils/getMainAxisFromPlacement.js\";\nimport getAltAxis from \"../utils/getAltAxis.js\";\nimport { within, withinMaxClamp } from \"../utils/within.js\";\nimport getLayoutRect from \"../dom-utils/getLayoutRect.js\";\nimport getOffsetParent from \"../dom-utils/getOffsetParent.js\";\nimport detectOverflow from \"../utils/detectOverflow.js\";\nimport getVariation from \"../utils/getVariation.js\";\nimport getFreshSideObject from \"../utils/getFreshSideObject.js\";\nimport { min as mathMin, max as mathMax } from \"../utils/math.js\";\n\nfunction preventOverflow(_ref) {\n  var state = _ref.state,\n      options = _ref.options,\n      name = _ref.name;\n  var _options$mainAxis = options.mainAxis,\n      checkMainAxis = _options$mainAxis === void 0 ? true : _options$mainAxis,\n      _options$altAxis = options.altAxis,\n      checkAltAxis = _options$altAxis === void 0 ? false : _options$altAxis,\n      boundary = options.boundary,\n      rootBoundary = options.rootBoundary,\n      altBoundary = options.altBoundary,\n      padding = options.padding,\n      _options$tether = options.tether,\n      tether = _options$tether === void 0 ? true : _options$tether,\n      _options$tetherOffset = options.tetherOffset,\n      tetherOffset = _options$tetherOffset === void 0 ? 0 : _options$tetherOffset;\n  var overflow = detectOverflow(state, {\n    boundary: boundary,\n    rootBoundary: rootBoundary,\n    padding: padding,\n    altBoundary: altBoundary\n  });\n  var basePlacement = getBasePlacement(state.placement);\n  var variation = getVariation(state.placement);\n  var isBasePlacement = !variation;\n  var mainAxis = getMainAxisFromPlacement(basePlacement);\n  var altAxis = getAltAxis(mainAxis);\n  var popperOffsets = state.modifiersData.popperOffsets;\n  var referenceRect = state.rects.reference;\n  var popperRect = state.rects.popper;\n  var tetherOffsetValue = typeof tetherOffset === 'function' ? tetherOffset(Object.assign({}, state.rects, {\n    placement: state.placement\n  })) : tetherOffset;\n  var normalizedTetherOffsetValue = typeof tetherOffsetValue === 'number' ? {\n    mainAxis: tetherOffsetValue,\n    altAxis: tetherOffsetValue\n  } : Object.assign({\n    mainAxis: 0,\n    altAxis: 0\n  }, tetherOffsetValue);\n  var offsetModifierState = state.modifiersData.offset ? state.modifiersData.offset[state.placement] : null;\n  var data = {\n    x: 0,\n    y: 0\n  };\n\n  if (!popperOffsets) {\n    return;\n  }\n\n  if (checkMainAxis) {\n    var _offsetModifierState$;\n\n    var mainSide = mainAxis === 'y' ? top : left;\n    var altSide = mainAxis === 'y' ? bottom : right;\n    var len = mainAxis === 'y' ? 'height' : 'width';\n    var offset = popperOffsets[mainAxis];\n    var min = offset + overflow[mainSide];\n    var max = offset - overflow[altSide];\n    var additive = tether ? -popperRect[len] / 2 : 0;\n    var minLen = variation === start ? referenceRect[len] : popperRect[len];\n    var maxLen = variation === start ? -popperRect[len] : -referenceRect[len]; // We need to include the arrow in the calculation so the arrow doesn't go\n    // outside the reference bounds\n\n    var arrowElement = state.elements.arrow;\n    var arrowRect = tether && arrowElement ? getLayoutRect(arrowElement) : {\n      width: 0,\n      height: 0\n    };\n    var arrowPaddingObject = state.modifiersData['arrow#persistent'] ? state.modifiersData['arrow#persistent'].padding : getFreshSideObject();\n    var arrowPaddingMin = arrowPaddingObject[mainSide];\n    var arrowPaddingMax = arrowPaddingObject[altSide]; // If the reference length is smaller than the arrow length, we don't want\n    // to include its full size in the calculation. If the reference is small\n    // and near the edge of a boundary, the popper can overflow even if the\n    // reference is not overflowing as well (e.g. virtual elements with no\n    // width or height)\n\n    var arrowLen = within(0, referenceRect[len], arrowRect[len]);\n    var minOffset = isBasePlacement ? referenceRect[len] / 2 - additive - arrowLen - arrowPaddingMin - normalizedTetherOffsetValue.mainAxis : minLen - arrowLen - arrowPaddingMin - normalizedTetherOffsetValue.mainAxis;\n    var maxOffset = isBasePlacement ? -referenceRect[len] / 2 + additive + arrowLen + arrowPaddingMax + normalizedTetherOffsetValue.mainAxis : maxLen + arrowLen + arrowPaddingMax + normalizedTetherOffsetValue.mainAxis;\n    var arrowOffsetParent = state.elements.arrow && getOffsetParent(state.elements.arrow);\n    var clientOffset = arrowOffsetParent ? mainAxis === 'y' ? arrowOffsetParent.clientTop || 0 : arrowOffsetParent.clientLeft || 0 : 0;\n    var offsetModifierValue = (_offsetModifierState$ = offsetModifierState == null ? void 0 : offsetModifierState[mainAxis]) != null ? _offsetModifierState$ : 0;\n    var tetherMin = offset + minOffset - offsetModifierValue - clientOffset;\n    var tetherMax = offset + maxOffset - offsetModifierValue;\n    var preventedOffset = within(tether ? mathMin(min, tetherMin) : min, offset, tether ? mathMax(max, tetherMax) : max);\n    popperOffsets[mainAxis] = preventedOffset;\n    data[mainAxis] = preventedOffset - offset;\n  }\n\n  if (checkAltAxis) {\n    var _offsetModifierState$2;\n\n    var _mainSide = mainAxis === 'x' ? top : left;\n\n    var _altSide = mainAxis === 'x' ? bottom : right;\n\n    var _offset = popperOffsets[altAxis];\n\n    var _len = altAxis === 'y' ? 'height' : 'width';\n\n    var _min = _offset + overflow[_mainSide];\n\n    var _max = _offset - overflow[_altSide];\n\n    var isOriginSide = [top, left].indexOf(basePlacement) !== -1;\n\n    var _offsetModifierValue = (_offsetModifierState$2 = offsetModifierState == null ? void 0 : offsetModifierState[altAxis]) != null ? _offsetModifierState$2 : 0;\n\n    var _tetherMin = isOriginSide ? _min : _offset - referenceRect[_len] - popperRect[_len] - _offsetModifierValue + normalizedTetherOffsetValue.altAxis;\n\n    var _tetherMax = isOriginSide ? _offset + referenceRect[_len] + popperRect[_len] - _offsetModifierValue - normalizedTetherOffsetValue.altAxis : _max;\n\n    var _preventedOffset = tether && isOriginSide ? withinMaxClamp(_tetherMin, _offset, _tetherMax) : within(tether ? _tetherMin : _min, _offset, tether ? _tetherMax : _max);\n\n    popperOffsets[altAxis] = _preventedOffset;\n    data[altAxis] = _preventedOffset - _offset;\n  }\n\n  state.modifiersData[name] = data;\n} // eslint-disable-next-line import/no-unused-modules\n\n\nexport default {\n  name: 'preventOverflow',\n  enabled: true,\n  phase: 'main',\n  fn: preventOverflow,\n  requiresIfExists: ['offset']\n};","export default function getAltAxis(axis) {\n  return axis === 'x' ? 'y' : 'x';\n}","import getBoundingClientRect from \"./getBoundingClientRect.js\";\nimport getNodeScroll from \"./getNodeScroll.js\";\nimport getNodeName from \"./getNodeName.js\";\nimport { isHTMLElement } from \"./instanceOf.js\";\nimport getWindowScrollBarX from \"./getWindowScrollBarX.js\";\nimport getDocumentElement from \"./getDocumentElement.js\";\nimport isScrollParent from \"./isScrollParent.js\";\nimport { round } from \"../utils/math.js\";\n\nfunction isElementScaled(element) {\n  var rect = element.getBoundingClientRect();\n  var scaleX = round(rect.width) / element.offsetWidth || 1;\n  var scaleY = round(rect.height) / element.offsetHeight || 1;\n  return scaleX !== 1 || scaleY !== 1;\n} // Returns the composite rect of an element relative to its offsetParent.\n// Composite means it takes into account transforms as well as layout.\n\n\nexport default function getCompositeRect(elementOrVirtualElement, offsetParent, isFixed) {\n  if (isFixed === void 0) {\n    isFixed = false;\n  }\n\n  var isOffsetParentAnElement = isHTMLElement(offsetParent);\n  var offsetParentIsScaled = isHTMLElement(offsetParent) && isElementScaled(offsetParent);\n  var documentElement = getDocumentElement(offsetParent);\n  var rect = getBoundingClientRect(elementOrVirtualElement, offsetParentIsScaled, isFixed);\n  var scroll = {\n    scrollLeft: 0,\n    scrollTop: 0\n  };\n  var offsets = {\n    x: 0,\n    y: 0\n  };\n\n  if (isOffsetParentAnElement || !isOffsetParentAnElement && !isFixed) {\n    if (getNodeName(offsetParent) !== 'body' || // https://github.com/popperjs/popper-core/issues/1078\n    isScrollParent(documentElement)) {\n      scroll = getNodeScroll(offsetParent);\n    }\n\n    if (isHTMLElement(offsetParent)) {\n      offsets = getBoundingClientRect(offsetParent, true);\n      offsets.x += offsetParent.clientLeft;\n      offsets.y += offsetParent.clientTop;\n    } else if (documentElement) {\n      offsets.x = getWindowScrollBarX(documentElement);\n    }\n  }\n\n  return {\n    x: rect.left + scroll.scrollLeft - offsets.x,\n    y: rect.top + scroll.scrollTop - offsets.y,\n    width: rect.width,\n    height: rect.height\n  };\n}","import getWindowScroll from \"./getWindowScroll.js\";\nimport getWindow from \"./getWindow.js\";\nimport { isHTMLElement } from \"./instanceOf.js\";\nimport getHTMLElementScroll from \"./getHTMLElementScroll.js\";\nexport default function getNodeScroll(node) {\n  if (node === getWindow(node) || !isHTMLElement(node)) {\n    return getWindowScroll(node);\n  } else {\n    return getHTMLElementScroll(node);\n  }\n}","export default function getHTMLElementScroll(element) {\n  return {\n    scrollLeft: element.scrollLeft,\n    scrollTop: element.scrollTop\n  };\n}","import { modifierPhases } from \"../enums.js\"; // source: https://stackoverflow.com/questions/49875255\n\nfunction order(modifiers) {\n  var map = new Map();\n  var visited = new Set();\n  var result = [];\n  modifiers.forEach(function (modifier) {\n    map.set(modifier.name, modifier);\n  }); // On visiting object, check for its dependencies and visit them recursively\n\n  function sort(modifier) {\n    visited.add(modifier.name);\n    var requires = [].concat(modifier.requires || [], modifier.requiresIfExists || []);\n    requires.forEach(function (dep) {\n      if (!visited.has(dep)) {\n        var depModifier = map.get(dep);\n\n        if (depModifier) {\n          sort(depModifier);\n        }\n      }\n    });\n    result.push(modifier);\n  }\n\n  modifiers.forEach(function (modifier) {\n    if (!visited.has(modifier.name)) {\n      // check for visited object\n      sort(modifier);\n    }\n  });\n  return result;\n}\n\nexport default function orderModifiers(modifiers) {\n  // order based on dependencies\n  var orderedModifiers = order(modifiers); // order based on phase\n\n  return modifierPhases.reduce(function (acc, phase) {\n    return acc.concat(orderedModifiers.filter(function (modifier) {\n      return modifier.phase === phase;\n    }));\n  }, []);\n}","import getCompositeRect from \"./dom-utils/getCompositeRect.js\";\nimport getLayoutRect from \"./dom-utils/getLayoutRect.js\";\nimport listScrollParents from \"./dom-utils/listScrollParents.js\";\nimport getOffsetParent from \"./dom-utils/getOffsetParent.js\";\nimport orderModifiers from \"./utils/orderModifiers.js\";\nimport debounce from \"./utils/debounce.js\";\nimport mergeByName from \"./utils/mergeByName.js\";\nimport detectOverflow from \"./utils/detectOverflow.js\";\nimport { isElement } from \"./dom-utils/instanceOf.js\";\nvar DEFAULT_OPTIONS = {\n  placement: 'bottom',\n  modifiers: [],\n  strategy: 'absolute'\n};\n\nfunction areValidElements() {\n  for (var _len = arguments.length, args = new Array(_len), _key = 0; _key < _len; _key++) {\n    args[_key] = arguments[_key];\n  }\n\n  return !args.some(function (element) {\n    return !(element && typeof element.getBoundingClientRect === 'function');\n  });\n}\n\nexport function popperGenerator(generatorOptions) {\n  if (generatorOptions === void 0) {\n    generatorOptions = {};\n  }\n\n  var _generatorOptions = generatorOptions,\n      _generatorOptions$def = _generatorOptions.defaultModifiers,\n      defaultModifiers = _generatorOptions$def === void 0 ? [] : _generatorOptions$def,\n      _generatorOptions$def2 = _generatorOptions.defaultOptions,\n      defaultOptions = _generatorOptions$def2 === void 0 ? DEFAULT_OPTIONS : _generatorOptions$def2;\n  return function createPopper(reference, popper, options) {\n    if (options === void 0) {\n      options = defaultOptions;\n    }\n\n    var state = {\n      placement: 'bottom',\n      orderedModifiers: [],\n      options: Object.assign({}, DEFAULT_OPTIONS, defaultOptions),\n      modifiersData: {},\n      elements: {\n        reference: reference,\n        popper: popper\n      },\n      attributes: {},\n      styles: {}\n    };\n    var effectCleanupFns = [];\n    var isDestroyed = false;\n    var instance = {\n      state: state,\n      setOptions: function setOptions(setOptionsAction) {\n        var options = typeof setOptionsAction === 'function' ? setOptionsAction(state.options) : setOptionsAction;\n        cleanupModifierEffects();\n        state.options = Object.assign({}, defaultOptions, state.options, options);\n        state.scrollParents = {\n          reference: isElement(reference) ? listScrollParents(reference) : reference.contextElement ? listScrollParents(reference.contextElement) : [],\n          popper: listScrollParents(popper)\n        }; // Orders the modifiers based on their dependencies and `phase`\n        // properties\n\n        var orderedModifiers = orderModifiers(mergeByName([].concat(defaultModifiers, state.options.modifiers))); // Strip out disabled modifiers\n\n        state.orderedModifiers = orderedModifiers.filter(function (m) {\n          return m.enabled;\n        });\n        runModifierEffects();\n        return instance.update();\n      },\n      // Sync update – it will always be executed, even if not necessary. This\n      // is useful for low frequency updates where sync behavior simplifies the\n      // logic.\n      // For high frequency updates (e.g. `resize` and `scroll` events), always\n      // prefer the async Popper#update method\n      forceUpdate: function forceUpdate() {\n        if (isDestroyed) {\n          return;\n        }\n\n        var _state$elements = state.elements,\n            reference = _state$elements.reference,\n            popper = _state$elements.popper; // Don't proceed if `reference` or `popper` are not valid elements\n        // anymore\n\n        if (!areValidElements(reference, popper)) {\n          return;\n        } // Store the reference and popper rects to be read by modifiers\n\n\n        state.rects = {\n          reference: getCompositeRect(reference, getOffsetParent(popper), state.options.strategy === 'fixed'),\n          popper: getLayoutRect(popper)\n        }; // Modifiers have the ability to reset the current update cycle. The\n        // most common use case for this is the `flip` modifier changing the\n        // placement, which then needs to re-run all the modifiers, because the\n        // logic was previously ran for the previous placement and is therefore\n        // stale/incorrect\n\n        state.reset = false;\n        state.placement = state.options.placement; // On each update cycle, the `modifiersData` property for each modifier\n        // is filled with the initial data specified by the modifier. This means\n        // it doesn't persist and is fresh on each update.\n        // To ensure persistent data, use `${name}#persistent`\n\n        state.orderedModifiers.forEach(function (modifier) {\n          return state.modifiersData[modifier.name] = Object.assign({}, modifier.data);\n        });\n\n        for (var index = 0; index < state.orderedModifiers.length; index++) {\n          if (state.reset === true) {\n            state.reset = false;\n            index = -1;\n            continue;\n          }\n\n          var _state$orderedModifie = state.orderedModifiers[index],\n              fn = _state$orderedModifie.fn,\n              _state$orderedModifie2 = _state$orderedModifie.options,\n              _options = _state$orderedModifie2 === void 0 ? {} : _state$orderedModifie2,\n              name = _state$orderedModifie.name;\n\n          if (typeof fn === 'function') {\n            state = fn({\n              state: state,\n              options: _options,\n              name: name,\n              instance: instance\n            }) || state;\n          }\n        }\n      },\n      // Async and optimistically optimized update – it will not be executed if\n      // not necessary (debounced to run at most once-per-tick)\n      update: debounce(function () {\n        return new Promise(function (resolve) {\n          instance.forceUpdate();\n          resolve(state);\n        });\n      }),\n      destroy: function destroy() {\n        cleanupModifierEffects();\n        isDestroyed = true;\n      }\n    };\n\n    if (!areValidElements(reference, popper)) {\n      return instance;\n    }\n\n    instance.setOptions(options).then(function (state) {\n      if (!isDestroyed && options.onFirstUpdate) {\n        options.onFirstUpdate(state);\n      }\n    }); // Modifiers have the ability to execute arbitrary code before the first\n    // update cycle runs. They will be executed in the same order as the update\n    // cycle. This is useful when a modifier adds some persistent data that\n    // other modifiers need to use, but the modifier is run after the dependent\n    // one.\n\n    function runModifierEffects() {\n      state.orderedModifiers.forEach(function (_ref) {\n        var name = _ref.name,\n            _ref$options = _ref.options,\n            options = _ref$options === void 0 ? {} : _ref$options,\n            effect = _ref.effect;\n\n        if (typeof effect === 'function') {\n          var cleanupFn = effect({\n            state: state,\n            name: name,\n            instance: instance,\n            options: options\n          });\n\n          var noopFn = function noopFn() {};\n\n          effectCleanupFns.push(cleanupFn || noopFn);\n        }\n      });\n    }\n\n    function cleanupModifierEffects() {\n      effectCleanupFns.forEach(function (fn) {\n        return fn();\n      });\n      effectCleanupFns = [];\n    }\n\n    return instance;\n  };\n}\nexport var createPopper = /*#__PURE__*/popperGenerator(); // eslint-disable-next-line import/no-unused-modules\n\nexport { detectOverflow };","export default function debounce(fn) {\n  var pending;\n  return function () {\n    if (!pending) {\n      pending = new Promise(function (resolve) {\n        Promise.resolve().then(function () {\n          pending = undefined;\n          resolve(fn());\n        });\n      });\n    }\n\n    return pending;\n  };\n}","export default function mergeByName(modifiers) {\n  var merged = modifiers.reduce(function (merged, current) {\n    var existing = merged[current.name];\n    merged[current.name] = existing ? Object.assign({}, existing, current, {\n      options: Object.assign({}, existing.options, current.options),\n      data: Object.assign({}, existing.data, current.data)\n    }) : current;\n    return merged;\n  }, {}); // IE11 does not support Object.values\n\n  return Object.keys(merged).map(function (key) {\n    return merged[key];\n  });\n}","import { popperGenerator, detectOverflow } from \"./createPopper.js\";\nimport eventListeners from \"./modifiers/eventListeners.js\";\nimport popperOffsets from \"./modifiers/popperOffsets.js\";\nimport computeStyles from \"./modifiers/computeStyles.js\";\nimport applyStyles from \"./modifiers/applyStyles.js\";\nimport offset from \"./modifiers/offset.js\";\nimport flip from \"./modifiers/flip.js\";\nimport preventOverflow from \"./modifiers/preventOverflow.js\";\nimport arrow from \"./modifiers/arrow.js\";\nimport hide from \"./modifiers/hide.js\";\nvar defaultModifiers = [eventListeners, popperOffsets, computeStyles, applyStyles, offset, flip, preventOverflow, arrow, hide];\nvar createPopper = /*#__PURE__*/popperGenerator({\n  defaultModifiers: defaultModifiers\n}); // eslint-disable-next-line import/no-unused-modules\n\nexport { createPopper, popperGenerator, defaultModifiers, detectOverflow }; // eslint-disable-next-line import/no-unused-modules\n\nexport { createPopper as createPopperLite } from \"./popper-lite.js\"; // eslint-disable-next-line import/no-unused-modules\n\nexport * from \"./modifiers/index.js\";","import { popperGenerator, detectOverflow } from \"./createPopper.js\";\nimport eventListeners from \"./modifiers/eventListeners.js\";\nimport popperOffsets from \"./modifiers/popperOffsets.js\";\nimport computeStyles from \"./modifiers/computeStyles.js\";\nimport applyStyles from \"./modifiers/applyStyles.js\";\nvar defaultModifiers = [eventListeners, popperOffsets, computeStyles, applyStyles];\nvar createPopper = /*#__PURE__*/popperGenerator({\n  defaultModifiers: defaultModifiers\n}); // eslint-disable-next-line import/no-unused-modules\n\nexport { createPopper, popperGenerator, defaultModifiers, detectOverflow };","/*!\n  * Bootstrap v5.3.3 (https://getbootstrap.com/)\n  * Copyright 2011-2024 The Bootstrap Authors (https://github.com/twbs/bootstrap/graphs/contributors)\n  * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n  */\nimport * as Popper from '@popperjs/core';\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap dom/data.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n/**\n * Constants\n */\n\nconst elementMap = new Map();\nconst Data = {\n  set(element, key, instance) {\n    if (!elementMap.has(element)) {\n      elementMap.set(element, new Map());\n    }\n    const instanceMap = elementMap.get(element);\n\n    // make it clear we only want one instance per element\n    // can be removed later when multiple key/instances are fine to be used\n    if (!instanceMap.has(key) && instanceMap.size !== 0) {\n      // eslint-disable-next-line no-console\n      console.error(`Bootstrap doesn't allow more than one instance per element. Bound instance: ${Array.from(instanceMap.keys())[0]}.`);\n      return;\n    }\n    instanceMap.set(key, instance);\n  },\n  get(element, key) {\n    if (elementMap.has(element)) {\n      return elementMap.get(element).get(key) || null;\n    }\n    return null;\n  },\n  remove(element, key) {\n    if (!elementMap.has(element)) {\n      return;\n    }\n    const instanceMap = elementMap.get(element);\n    instanceMap.delete(key);\n\n    // free up element references if there are no instances left for an element\n    if (instanceMap.size === 0) {\n      elementMap.delete(element);\n    }\n  }\n};\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap util/index.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\nconst MAX_UID = 1000000;\nconst MILLISECONDS_MULTIPLIER = 1000;\nconst TRANSITION_END = 'transitionend';\n\n/**\n * Properly escape IDs selectors to handle weird IDs\n * @param {string} selector\n * @returns {string}\n */\nconst parseSelector = selector => {\n  if (selector && window.CSS && window.CSS.escape) {\n    // document.querySelector needs escaping to handle IDs (html5+) containing for instance /\n    selector = selector.replace(/#([^\\s\"#']+)/g, (match, id) => `#${CSS.escape(id)}`);\n  }\n  return selector;\n};\n\n// Shout-out Angus Croll (https://goo.gl/pxwQGp)\nconst toType = object => {\n  if (object === null || object === undefined) {\n    return `${object}`;\n  }\n  return Object.prototype.toString.call(object).match(/\\s([a-z]+)/i)[1].toLowerCase();\n};\n\n/**\n * Public Util API\n */\n\nconst getUID = prefix => {\n  do {\n    prefix += Math.floor(Math.random() * MAX_UID);\n  } while (document.getElementById(prefix));\n  return prefix;\n};\nconst getTransitionDurationFromElement = element => {\n  if (!element) {\n    return 0;\n  }\n\n  // Get transition-duration of the element\n  let {\n    transitionDuration,\n    transitionDelay\n  } = window.getComputedStyle(element);\n  const floatTransitionDuration = Number.parseFloat(transitionDuration);\n  const floatTransitionDelay = Number.parseFloat(transitionDelay);\n\n  // Return 0 if element or transition duration is not found\n  if (!floatTransitionDuration && !floatTransitionDelay) {\n    return 0;\n  }\n\n  // If multiple durations are defined, take the first\n  transitionDuration = transitionDuration.split(',')[0];\n  transitionDelay = transitionDelay.split(',')[0];\n  return (Number.parseFloat(transitionDuration) + Number.parseFloat(transitionDelay)) * MILLISECONDS_MULTIPLIER;\n};\nconst triggerTransitionEnd = element => {\n  element.dispatchEvent(new Event(TRANSITION_END));\n};\nconst isElement = object => {\n  if (!object || typeof object !== 'object') {\n    return false;\n  }\n  if (typeof object.jquery !== 'undefined') {\n    object = object[0];\n  }\n  return typeof object.nodeType !== 'undefined';\n};\nconst getElement = object => {\n  // it's a jQuery object or a node element\n  if (isElement(object)) {\n    return object.jquery ? object[0] : object;\n  }\n  if (typeof object === 'string' && object.length > 0) {\n    return document.querySelector(parseSelector(object));\n  }\n  return null;\n};\nconst isVisible = element => {\n  if (!isElement(element) || element.getClientRects().length === 0) {\n    return false;\n  }\n  const elementIsVisible = getComputedStyle(element).getPropertyValue('visibility') === 'visible';\n  // Handle `details` element as its content may falsie appear visible when it is closed\n  const closedDetails = element.closest('details:not([open])');\n  if (!closedDetails) {\n    return elementIsVisible;\n  }\n  if (closedDetails !== element) {\n    const summary = element.closest('summary');\n    if (summary && summary.parentNode !== closedDetails) {\n      return false;\n    }\n    if (summary === null) {\n      return false;\n    }\n  }\n  return elementIsVisible;\n};\nconst isDisabled = element => {\n  if (!element || element.nodeType !== Node.ELEMENT_NODE) {\n    return true;\n  }\n  if (element.classList.contains('disabled')) {\n    return true;\n  }\n  if (typeof element.disabled !== 'undefined') {\n    return element.disabled;\n  }\n  return element.hasAttribute('disabled') && element.getAttribute('disabled') !== 'false';\n};\nconst findShadowRoot = element => {\n  if (!document.documentElement.attachShadow) {\n    return null;\n  }\n\n  // Can find the shadow root otherwise it'll return the document\n  if (typeof element.getRootNode === 'function') {\n    const root = element.getRootNode();\n    return root instanceof ShadowRoot ? root : null;\n  }\n  if (element instanceof ShadowRoot) {\n    return element;\n  }\n\n  // when we don't find a shadow root\n  if (!element.parentNode) {\n    return null;\n  }\n  return findShadowRoot(element.parentNode);\n};\nconst noop = () => {};\n\n/**\n * Trick to restart an element's animation\n *\n * @param {HTMLElement} element\n * @return void\n *\n * @see https://www.charistheo.io/blog/2021/02/restart-a-css-animation-with-javascript/#restarting-a-css-animation\n */\nconst reflow = element => {\n  element.offsetHeight; // eslint-disable-line no-unused-expressions\n};\nconst getjQuery = () => {\n  if (window.jQuery && !document.body.hasAttribute('data-bs-no-jquery')) {\n    return window.jQuery;\n  }\n  return null;\n};\nconst DOMContentLoadedCallbacks = [];\nconst onDOMContentLoaded = callback => {\n  if (document.readyState === 'loading') {\n    // add listener on the first call when the document is in loading state\n    if (!DOMContentLoadedCallbacks.length) {\n      document.addEventListener('DOMContentLoaded', () => {\n        for (const callback of DOMContentLoadedCallbacks) {\n          callback();\n        }\n      });\n    }\n    DOMContentLoadedCallbacks.push(callback);\n  } else {\n    callback();\n  }\n};\nconst isRTL = () => document.documentElement.dir === 'rtl';\nconst defineJQueryPlugin = plugin => {\n  onDOMContentLoaded(() => {\n    const $ = getjQuery();\n    /* istanbul ignore if */\n    if ($) {\n      const name = plugin.NAME;\n      const JQUERY_NO_CONFLICT = $.fn[name];\n      $.fn[name] = plugin.jQueryInterface;\n      $.fn[name].Constructor = plugin;\n      $.fn[name].noConflict = () => {\n        $.fn[name] = JQUERY_NO_CONFLICT;\n        return plugin.jQueryInterface;\n      };\n    }\n  });\n};\nconst execute = (possibleCallback, args = [], defaultValue = possibleCallback) => {\n  return typeof possibleCallback === 'function' ? possibleCallback(...args) : defaultValue;\n};\nconst executeAfterTransition = (callback, transitionElement, waitForTransition = true) => {\n  if (!waitForTransition) {\n    execute(callback);\n    return;\n  }\n  const durationPadding = 5;\n  const emulatedDuration = getTransitionDurationFromElement(transitionElement) + durationPadding;\n  let called = false;\n  const handler = ({\n    target\n  }) => {\n    if (target !== transitionElement) {\n      return;\n    }\n    called = true;\n    transitionElement.removeEventListener(TRANSITION_END, handler);\n    execute(callback);\n  };\n  transitionElement.addEventListener(TRANSITION_END, handler);\n  setTimeout(() => {\n    if (!called) {\n      triggerTransitionEnd(transitionElement);\n    }\n  }, emulatedDuration);\n};\n\n/**\n * Return the previous/next element of a list.\n *\n * @param {array} list    The list of elements\n * @param activeElement   The active element\n * @param shouldGetNext   Choose to get next or previous element\n * @param isCycleAllowed\n * @return {Element|elem} The proper element\n */\nconst getNextActiveElement = (list, activeElement, shouldGetNext, isCycleAllowed) => {\n  const listLength = list.length;\n  let index = list.indexOf(activeElement);\n\n  // if the element does not exist in the list return an element\n  // depending on the direction and if cycle is allowed\n  if (index === -1) {\n    return !shouldGetNext && isCycleAllowed ? list[listLength - 1] : list[0];\n  }\n  index += shouldGetNext ? 1 : -1;\n  if (isCycleAllowed) {\n    index = (index + listLength) % listLength;\n  }\n  return list[Math.max(0, Math.min(index, listLength - 1))];\n};\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap dom/event-handler.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst namespaceRegex = /[^.]*(?=\\..*)\\.|.*/;\nconst stripNameRegex = /\\..*/;\nconst stripUidRegex = /::\\d+$/;\nconst eventRegistry = {}; // Events storage\nlet uidEvent = 1;\nconst customEvents = {\n  mouseenter: 'mouseover',\n  mouseleave: 'mouseout'\n};\nconst nativeEvents = new Set(['click', 'dblclick', 'mouseup', 'mousedown', 'contextmenu', 'mousewheel', 'DOMMouseScroll', 'mouseover', 'mouseout', 'mousemove', 'selectstart', 'selectend', 'keydown', 'keypress', 'keyup', 'orientationchange', 'touchstart', 'touchmove', 'touchend', 'touchcancel', 'pointerdown', 'pointermove', 'pointerup', 'pointerleave', 'pointercancel', 'gesturestart', 'gesturechange', 'gestureend', 'focus', 'blur', 'change', 'reset', 'select', 'submit', 'focusin', 'focusout', 'load', 'unload', 'beforeunload', 'resize', 'move', 'DOMContentLoaded', 'readystatechange', 'error', 'abort', 'scroll']);\n\n/**\n * Private methods\n */\n\nfunction makeEventUid(element, uid) {\n  return uid && `${uid}::${uidEvent++}` || element.uidEvent || uidEvent++;\n}\nfunction getElementEvents(element) {\n  const uid = makeEventUid(element);\n  element.uidEvent = uid;\n  eventRegistry[uid] = eventRegistry[uid] || {};\n  return eventRegistry[uid];\n}\nfunction bootstrapHandler(element, fn) {\n  return function handler(event) {\n    hydrateObj(event, {\n      delegateTarget: element\n    });\n    if (handler.oneOff) {\n      EventHandler.off(element, event.type, fn);\n    }\n    return fn.apply(element, [event]);\n  };\n}\nfunction bootstrapDelegationHandler(element, selector, fn) {\n  return function handler(event) {\n    const domElements = element.querySelectorAll(selector);\n    for (let {\n      target\n    } = event; target && target !== this; target = target.parentNode) {\n      for (const domElement of domElements) {\n        if (domElement !== target) {\n          continue;\n        }\n        hydrateObj(event, {\n          delegateTarget: target\n        });\n        if (handler.oneOff) {\n          EventHandler.off(element, event.type, selector, fn);\n        }\n        return fn.apply(target, [event]);\n      }\n    }\n  };\n}\nfunction findHandler(events, callable, delegationSelector = null) {\n  return Object.values(events).find(event => event.callable === callable && event.delegationSelector === delegationSelector);\n}\nfunction normalizeParameters(originalTypeEvent, handler, delegationFunction) {\n  const isDelegated = typeof handler === 'string';\n  // TODO: tooltip passes `false` instead of selector, so we need to check\n  const callable = isDelegated ? delegationFunction : handler || delegationFunction;\n  let typeEvent = getTypeEvent(originalTypeEvent);\n  if (!nativeEvents.has(typeEvent)) {\n    typeEvent = originalTypeEvent;\n  }\n  return [isDelegated, callable, typeEvent];\n}\nfunction addHandler(element, originalTypeEvent, handler, delegationFunction, oneOff) {\n  if (typeof originalTypeEvent !== 'string' || !element) {\n    return;\n  }\n  let [isDelegated, callable, typeEvent] = normalizeParameters(originalTypeEvent, handler, delegationFunction);\n\n  // in case of mouseenter or mouseleave wrap the handler within a function that checks for its DOM position\n  // this prevents the handler from being dispatched the same way as mouseover or mouseout does\n  if (originalTypeEvent in customEvents) {\n    const wrapFunction = fn => {\n      return function (event) {\n        if (!event.relatedTarget || event.relatedTarget !== event.delegateTarget && !event.delegateTarget.contains(event.relatedTarget)) {\n          return fn.call(this, event);\n        }\n      };\n    };\n    callable = wrapFunction(callable);\n  }\n  const events = getElementEvents(element);\n  const handlers = events[typeEvent] || (events[typeEvent] = {});\n  const previousFunction = findHandler(handlers, callable, isDelegated ? handler : null);\n  if (previousFunction) {\n    previousFunction.oneOff = previousFunction.oneOff && oneOff;\n    return;\n  }\n  const uid = makeEventUid(callable, originalTypeEvent.replace(namespaceRegex, ''));\n  const fn = isDelegated ? bootstrapDelegationHandler(element, handler, callable) : bootstrapHandler(element, callable);\n  fn.delegationSelector = isDelegated ? handler : null;\n  fn.callable = callable;\n  fn.oneOff = oneOff;\n  fn.uidEvent = uid;\n  handlers[uid] = fn;\n  element.addEventListener(typeEvent, fn, isDelegated);\n}\nfunction removeHandler(element, events, typeEvent, handler, delegationSelector) {\n  const fn = findHandler(events[typeEvent], handler, delegationSelector);\n  if (!fn) {\n    return;\n  }\n  element.removeEventListener(typeEvent, fn, Boolean(delegationSelector));\n  delete events[typeEvent][fn.uidEvent];\n}\nfunction removeNamespacedHandlers(element, events, typeEvent, namespace) {\n  const storeElementEvent = events[typeEvent] || {};\n  for (const [handlerKey, event] of Object.entries(storeElementEvent)) {\n    if (handlerKey.includes(namespace)) {\n      removeHandler(element, events, typeEvent, event.callable, event.delegationSelector);\n    }\n  }\n}\nfunction getTypeEvent(event) {\n  // allow to get the native events from namespaced events ('click.bs.button' --> 'click')\n  event = event.replace(stripNameRegex, '');\n  return customEvents[event] || event;\n}\nconst EventHandler = {\n  on(element, event, handler, delegationFunction) {\n    addHandler(element, event, handler, delegationFunction, false);\n  },\n  one(element, event, handler, delegationFunction) {\n    addHandler(element, event, handler, delegationFunction, true);\n  },\n  off(element, originalTypeEvent, handler, delegationFunction) {\n    if (typeof originalTypeEvent !== 'string' || !element) {\n      return;\n    }\n    const [isDelegated, callable, typeEvent] = normalizeParameters(originalTypeEvent, handler, delegationFunction);\n    const inNamespace = typeEvent !== originalTypeEvent;\n    const events = getElementEvents(element);\n    const storeElementEvent = events[typeEvent] || {};\n    const isNamespace = originalTypeEvent.startsWith('.');\n    if (typeof callable !== 'undefined') {\n      // Simplest case: handler is passed, remove that listener ONLY.\n      if (!Object.keys(storeElementEvent).length) {\n        return;\n      }\n      removeHandler(element, events, typeEvent, callable, isDelegated ? handler : null);\n      return;\n    }\n    if (isNamespace) {\n      for (const elementEvent of Object.keys(events)) {\n        removeNamespacedHandlers(element, events, elementEvent, originalTypeEvent.slice(1));\n      }\n    }\n    for (const [keyHandlers, event] of Object.entries(storeElementEvent)) {\n      const handlerKey = keyHandlers.replace(stripUidRegex, '');\n      if (!inNamespace || originalTypeEvent.includes(handlerKey)) {\n        removeHandler(element, events, typeEvent, event.callable, event.delegationSelector);\n      }\n    }\n  },\n  trigger(element, event, args) {\n    if (typeof event !== 'string' || !element) {\n      return null;\n    }\n    const $ = getjQuery();\n    const typeEvent = getTypeEvent(event);\n    const inNamespace = event !== typeEvent;\n    let jQueryEvent = null;\n    let bubbles = true;\n    let nativeDispatch = true;\n    let defaultPrevented = false;\n    if (inNamespace && $) {\n      jQueryEvent = $.Event(event, args);\n      $(element).trigger(jQueryEvent);\n      bubbles = !jQueryEvent.isPropagationStopped();\n      nativeDispatch = !jQueryEvent.isImmediatePropagationStopped();\n      defaultPrevented = jQueryEvent.isDefaultPrevented();\n    }\n    const evt = hydrateObj(new Event(event, {\n      bubbles,\n      cancelable: true\n    }), args);\n    if (defaultPrevented) {\n      evt.preventDefault();\n    }\n    if (nativeDispatch) {\n      element.dispatchEvent(evt);\n    }\n    if (evt.defaultPrevented && jQueryEvent) {\n      jQueryEvent.preventDefault();\n    }\n    return evt;\n  }\n};\nfunction hydrateObj(obj, meta = {}) {\n  for (const [key, value] of Object.entries(meta)) {\n    try {\n      obj[key] = value;\n    } catch (_unused) {\n      Object.defineProperty(obj, key, {\n        configurable: true,\n        get() {\n          return value;\n        }\n      });\n    }\n  }\n  return obj;\n}\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap dom/manipulator.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\nfunction normalizeData(value) {\n  if (value === 'true') {\n    return true;\n  }\n  if (value === 'false') {\n    return false;\n  }\n  if (value === Number(value).toString()) {\n    return Number(value);\n  }\n  if (value === '' || value === 'null') {\n    return null;\n  }\n  if (typeof value !== 'string') {\n    return value;\n  }\n  try {\n    return JSON.parse(decodeURIComponent(value));\n  } catch (_unused) {\n    return value;\n  }\n}\nfunction normalizeDataKey(key) {\n  return key.replace(/[A-Z]/g, chr => `-${chr.toLowerCase()}`);\n}\nconst Manipulator = {\n  setDataAttribute(element, key, value) {\n    element.setAttribute(`data-bs-${normalizeDataKey(key)}`, value);\n  },\n  removeDataAttribute(element, key) {\n    element.removeAttribute(`data-bs-${normalizeDataKey(key)}`);\n  },\n  getDataAttributes(element) {\n    if (!element) {\n      return {};\n    }\n    const attributes = {};\n    const bsKeys = Object.keys(element.dataset).filter(key => key.startsWith('bs') && !key.startsWith('bsConfig'));\n    for (const key of bsKeys) {\n      let pureKey = key.replace(/^bs/, '');\n      pureKey = pureKey.charAt(0).toLowerCase() + pureKey.slice(1, pureKey.length);\n      attributes[pureKey] = normalizeData(element.dataset[key]);\n    }\n    return attributes;\n  },\n  getDataAttribute(element, key) {\n    return normalizeData(element.getAttribute(`data-bs-${normalizeDataKey(key)}`));\n  }\n};\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap util/config.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Class definition\n */\n\nclass Config {\n  // Getters\n  static get Default() {\n    return {};\n  }\n  static get DefaultType() {\n    return {};\n  }\n  static get NAME() {\n    throw new Error('You have to implement the static method \"NAME\", for each component!');\n  }\n  _getConfig(config) {\n    config = this._mergeConfigObj(config);\n    config = this._configAfterMerge(config);\n    this._typeCheckConfig(config);\n    return config;\n  }\n  _configAfterMerge(config) {\n    return config;\n  }\n  _mergeConfigObj(config, element) {\n    const jsonConfig = isElement(element) ? Manipulator.getDataAttribute(element, 'config') : {}; // try to parse\n\n    return {\n      ...this.constructor.Default,\n      ...(typeof jsonConfig === 'object' ? jsonConfig : {}),\n      ...(isElement(element) ? Manipulator.getDataAttributes(element) : {}),\n      ...(typeof config === 'object' ? config : {})\n    };\n  }\n  _typeCheckConfig(config, configTypes = this.constructor.DefaultType) {\n    for (const [property, expectedTypes] of Object.entries(configTypes)) {\n      const value = config[property];\n      const valueType = isElement(value) ? 'element' : toType(value);\n      if (!new RegExp(expectedTypes).test(valueType)) {\n        throw new TypeError(`${this.constructor.NAME.toUpperCase()}: Option \"${property}\" provided type \"${valueType}\" but expected type \"${expectedTypes}\".`);\n      }\n    }\n  }\n}\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap base-component.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst VERSION = '5.3.3';\n\n/**\n * Class definition\n */\n\nclass BaseComponent extends Config {\n  constructor(element, config) {\n    super();\n    element = getElement(element);\n    if (!element) {\n      return;\n    }\n    this._element = element;\n    this._config = this._getConfig(config);\n    Data.set(this._element, this.constructor.DATA_KEY, this);\n  }\n\n  // Public\n  dispose() {\n    Data.remove(this._element, this.constructor.DATA_KEY);\n    EventHandler.off(this._element, this.constructor.EVENT_KEY);\n    for (const propertyName of Object.getOwnPropertyNames(this)) {\n      this[propertyName] = null;\n    }\n  }\n  _queueCallback(callback, element, isAnimated = true) {\n    executeAfterTransition(callback, element, isAnimated);\n  }\n  _getConfig(config) {\n    config = this._mergeConfigObj(config, this._element);\n    config = this._configAfterMerge(config);\n    this._typeCheckConfig(config);\n    return config;\n  }\n\n  // Static\n  static getInstance(element) {\n    return Data.get(getElement(element), this.DATA_KEY);\n  }\n  static getOrCreateInstance(element, config = {}) {\n    return this.getInstance(element) || new this(element, typeof config === 'object' ? config : null);\n  }\n  static get VERSION() {\n    return VERSION;\n  }\n  static get DATA_KEY() {\n    return `bs.${this.NAME}`;\n  }\n  static get EVENT_KEY() {\n    return `.${this.DATA_KEY}`;\n  }\n  static eventName(name) {\n    return `${name}${this.EVENT_KEY}`;\n  }\n}\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap dom/selector-engine.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\nconst getSelector = element => {\n  let selector = element.getAttribute('data-bs-target');\n  if (!selector || selector === '#') {\n    let hrefAttribute = element.getAttribute('href');\n\n    // The only valid content that could double as a selector are IDs or classes,\n    // so everything starting with `#` or `.`. If a \"real\" URL is used as the selector,\n    // `document.querySelector` will rightfully complain it is invalid.\n    // See https://github.com/twbs/bootstrap/issues/32273\n    if (!hrefAttribute || !hrefAttribute.includes('#') && !hrefAttribute.startsWith('.')) {\n      return null;\n    }\n\n    // Just in case some CMS puts out a full URL with the anchor appended\n    if (hrefAttribute.includes('#') && !hrefAttribute.startsWith('#')) {\n      hrefAttribute = `#${hrefAttribute.split('#')[1]}`;\n    }\n    selector = hrefAttribute && hrefAttribute !== '#' ? hrefAttribute.trim() : null;\n  }\n  return selector ? selector.split(',').map(sel => parseSelector(sel)).join(',') : null;\n};\nconst SelectorEngine = {\n  find(selector, element = document.documentElement) {\n    return [].concat(...Element.prototype.querySelectorAll.call(element, selector));\n  },\n  findOne(selector, element = document.documentElement) {\n    return Element.prototype.querySelector.call(element, selector);\n  },\n  children(element, selector) {\n    return [].concat(...element.children).filter(child => child.matches(selector));\n  },\n  parents(element, selector) {\n    const parents = [];\n    let ancestor = element.parentNode.closest(selector);\n    while (ancestor) {\n      parents.push(ancestor);\n      ancestor = ancestor.parentNode.closest(selector);\n    }\n    return parents;\n  },\n  prev(element, selector) {\n    let previous = element.previousElementSibling;\n    while (previous) {\n      if (previous.matches(selector)) {\n        return [previous];\n      }\n      previous = previous.previousElementSibling;\n    }\n    return [];\n  },\n  // TODO: this is now unused; remove later along with prev()\n  next(element, selector) {\n    let next = element.nextElementSibling;\n    while (next) {\n      if (next.matches(selector)) {\n        return [next];\n      }\n      next = next.nextElementSibling;\n    }\n    return [];\n  },\n  focusableChildren(element) {\n    const focusables = ['a', 'button', 'input', 'textarea', 'select', 'details', '[tabindex]', '[contenteditable=\"true\"]'].map(selector => `${selector}:not([tabindex^=\"-\"])`).join(',');\n    return this.find(focusables, element).filter(el => !isDisabled(el) && isVisible(el));\n  },\n  getSelectorFromElement(element) {\n    const selector = getSelector(element);\n    if (selector) {\n      return SelectorEngine.findOne(selector) ? selector : null;\n    }\n    return null;\n  },\n  getElementFromSelector(element) {\n    const selector = getSelector(element);\n    return selector ? SelectorEngine.findOne(selector) : null;\n  },\n  getMultipleElementsFromSelector(element) {\n    const selector = getSelector(element);\n    return selector ? SelectorEngine.find(selector) : [];\n  }\n};\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap util/component-functions.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\nconst enableDismissTrigger = (component, method = 'hide') => {\n  const clickEvent = `click.dismiss${component.EVENT_KEY}`;\n  const name = component.NAME;\n  EventHandler.on(document, clickEvent, `[data-bs-dismiss=\"${name}\"]`, function (event) {\n    if (['A', 'AREA'].includes(this.tagName)) {\n      event.preventDefault();\n    }\n    if (isDisabled(this)) {\n      return;\n    }\n    const target = SelectorEngine.getElementFromSelector(this) || this.closest(`.${name}`);\n    const instance = component.getOrCreateInstance(target);\n\n    // Method argument is left, for Alert and only, as it doesn't implement the 'hide' method\n    instance[method]();\n  });\n};\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap alert.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$f = 'alert';\nconst DATA_KEY$a = 'bs.alert';\nconst EVENT_KEY$b = `.${DATA_KEY$a}`;\nconst EVENT_CLOSE = `close${EVENT_KEY$b}`;\nconst EVENT_CLOSED = `closed${EVENT_KEY$b}`;\nconst CLASS_NAME_FADE$5 = 'fade';\nconst CLASS_NAME_SHOW$8 = 'show';\n\n/**\n * Class definition\n */\n\nclass Alert extends BaseComponent {\n  // Getters\n  static get NAME() {\n    return NAME$f;\n  }\n\n  // Public\n  close() {\n    const closeEvent = EventHandler.trigger(this._element, EVENT_CLOSE);\n    if (closeEvent.defaultPrevented) {\n      return;\n    }\n    this._element.classList.remove(CLASS_NAME_SHOW$8);\n    const isAnimated = this._element.classList.contains(CLASS_NAME_FADE$5);\n    this._queueCallback(() => this._destroyElement(), this._element, isAnimated);\n  }\n\n  // Private\n  _destroyElement() {\n    this._element.remove();\n    EventHandler.trigger(this._element, EVENT_CLOSED);\n    this.dispose();\n  }\n\n  // Static\n  static jQueryInterface(config) {\n    return this.each(function () {\n      const data = Alert.getOrCreateInstance(this);\n      if (typeof config !== 'string') {\n        return;\n      }\n      if (data[config] === undefined || config.startsWith('_') || config === 'constructor') {\n        throw new TypeError(`No method named \"${config}\"`);\n      }\n      data[config](this);\n    });\n  }\n}\n\n/**\n * Data API implementation\n */\n\nenableDismissTrigger(Alert, 'close');\n\n/**\n * jQuery\n */\n\ndefineJQueryPlugin(Alert);\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap button.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$e = 'button';\nconst DATA_KEY$9 = 'bs.button';\nconst EVENT_KEY$a = `.${DATA_KEY$9}`;\nconst DATA_API_KEY$6 = '.data-api';\nconst CLASS_NAME_ACTIVE$3 = 'active';\nconst SELECTOR_DATA_TOGGLE$5 = '[data-bs-toggle=\"button\"]';\nconst EVENT_CLICK_DATA_API$6 = `click${EVENT_KEY$a}${DATA_API_KEY$6}`;\n\n/**\n * Class definition\n */\n\nclass Button extends BaseComponent {\n  // Getters\n  static get NAME() {\n    return NAME$e;\n  }\n\n  // Public\n  toggle() {\n    // Toggle class and sync the `aria-pressed` attribute with the return value of the `.toggle()` method\n    this._element.setAttribute('aria-pressed', this._element.classList.toggle(CLASS_NAME_ACTIVE$3));\n  }\n\n  // Static\n  static jQueryInterface(config) {\n    return this.each(function () {\n      const data = Button.getOrCreateInstance(this);\n      if (config === 'toggle') {\n        data[config]();\n      }\n    });\n  }\n}\n\n/**\n * Data API implementation\n */\n\nEventHandler.on(document, EVENT_CLICK_DATA_API$6, SELECTOR_DATA_TOGGLE$5, event => {\n  event.preventDefault();\n  const button = event.target.closest(SELECTOR_DATA_TOGGLE$5);\n  const data = Button.getOrCreateInstance(button);\n  data.toggle();\n});\n\n/**\n * jQuery\n */\n\ndefineJQueryPlugin(Button);\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap util/swipe.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$d = 'swipe';\nconst EVENT_KEY$9 = '.bs.swipe';\nconst EVENT_TOUCHSTART = `touchstart${EVENT_KEY$9}`;\nconst EVENT_TOUCHMOVE = `touchmove${EVENT_KEY$9}`;\nconst EVENT_TOUCHEND = `touchend${EVENT_KEY$9}`;\nconst EVENT_POINTERDOWN = `pointerdown${EVENT_KEY$9}`;\nconst EVENT_POINTERUP = `pointerup${EVENT_KEY$9}`;\nconst POINTER_TYPE_TOUCH = 'touch';\nconst POINTER_TYPE_PEN = 'pen';\nconst CLASS_NAME_POINTER_EVENT = 'pointer-event';\nconst SWIPE_THRESHOLD = 40;\nconst Default$c = {\n  endCallback: null,\n  leftCallback: null,\n  rightCallback: null\n};\nconst DefaultType$c = {\n  endCallback: '(function|null)',\n  leftCallback: '(function|null)',\n  rightCallback: '(function|null)'\n};\n\n/**\n * Class definition\n */\n\nclass Swipe extends Config {\n  constructor(element, config) {\n    super();\n    this._element = element;\n    if (!element || !Swipe.isSupported()) {\n      return;\n    }\n    this._config = this._getConfig(config);\n    this._deltaX = 0;\n    this._supportPointerEvents = Boolean(window.PointerEvent);\n    this._initEvents();\n  }\n\n  // Getters\n  static get Default() {\n    return Default$c;\n  }\n  static get DefaultType() {\n    return DefaultType$c;\n  }\n  static get NAME() {\n    return NAME$d;\n  }\n\n  // Public\n  dispose() {\n    EventHandler.off(this._element, EVENT_KEY$9);\n  }\n\n  // Private\n  _start(event) {\n    if (!this._supportPointerEvents) {\n      this._deltaX = event.touches[0].clientX;\n      return;\n    }\n    if (this._eventIsPointerPenTouch(event)) {\n      this._deltaX = event.clientX;\n    }\n  }\n  _end(event) {\n    if (this._eventIsPointerPenTouch(event)) {\n      this._deltaX = event.clientX - this._deltaX;\n    }\n    this._handleSwipe();\n    execute(this._config.endCallback);\n  }\n  _move(event) {\n    this._deltaX = event.touches && event.touches.length > 1 ? 0 : event.touches[0].clientX - this._deltaX;\n  }\n  _handleSwipe() {\n    const absDeltaX = Math.abs(this._deltaX);\n    if (absDeltaX <= SWIPE_THRESHOLD) {\n      return;\n    }\n    const direction = absDeltaX / this._deltaX;\n    this._deltaX = 0;\n    if (!direction) {\n      return;\n    }\n    execute(direction > 0 ? this._config.rightCallback : this._config.leftCallback);\n  }\n  _initEvents() {\n    if (this._supportPointerEvents) {\n      EventHandler.on(this._element, EVENT_POINTERDOWN, event => this._start(event));\n      EventHandler.on(this._element, EVENT_POINTERUP, event => this._end(event));\n      this._element.classList.add(CLASS_NAME_POINTER_EVENT);\n    } else {\n      EventHandler.on(this._element, EVENT_TOUCHSTART, event => this._start(event));\n      EventHandler.on(this._element, EVENT_TOUCHMOVE, event => this._move(event));\n      EventHandler.on(this._element, EVENT_TOUCHEND, event => this._end(event));\n    }\n  }\n  _eventIsPointerPenTouch(event) {\n    return this._supportPointerEvents && (event.pointerType === POINTER_TYPE_PEN || event.pointerType === POINTER_TYPE_TOUCH);\n  }\n\n  // Static\n  static isSupported() {\n    return 'ontouchstart' in document.documentElement || navigator.maxTouchPoints > 0;\n  }\n}\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap carousel.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$c = 'carousel';\nconst DATA_KEY$8 = 'bs.carousel';\nconst EVENT_KEY$8 = `.${DATA_KEY$8}`;\nconst DATA_API_KEY$5 = '.data-api';\nconst ARROW_LEFT_KEY$1 = 'ArrowLeft';\nconst ARROW_RIGHT_KEY$1 = 'ArrowRight';\nconst TOUCHEVENT_COMPAT_WAIT = 500; // Time for mouse compat events to fire after touch\n\nconst ORDER_NEXT = 'next';\nconst ORDER_PREV = 'prev';\nconst DIRECTION_LEFT = 'left';\nconst DIRECTION_RIGHT = 'right';\nconst EVENT_SLIDE = `slide${EVENT_KEY$8}`;\nconst EVENT_SLID = `slid${EVENT_KEY$8}`;\nconst EVENT_KEYDOWN$1 = `keydown${EVENT_KEY$8}`;\nconst EVENT_MOUSEENTER$1 = `mouseenter${EVENT_KEY$8}`;\nconst EVENT_MOUSELEAVE$1 = `mouseleave${EVENT_KEY$8}`;\nconst EVENT_DRAG_START = `dragstart${EVENT_KEY$8}`;\nconst EVENT_LOAD_DATA_API$3 = `load${EVENT_KEY$8}${DATA_API_KEY$5}`;\nconst EVENT_CLICK_DATA_API$5 = `click${EVENT_KEY$8}${DATA_API_KEY$5}`;\nconst CLASS_NAME_CAROUSEL = 'carousel';\nconst CLASS_NAME_ACTIVE$2 = 'active';\nconst CLASS_NAME_SLIDE = 'slide';\nconst CLASS_NAME_END = 'carousel-item-end';\nconst CLASS_NAME_START = 'carousel-item-start';\nconst CLASS_NAME_NEXT = 'carousel-item-next';\nconst CLASS_NAME_PREV = 'carousel-item-prev';\nconst SELECTOR_ACTIVE = '.active';\nconst SELECTOR_ITEM = '.carousel-item';\nconst SELECTOR_ACTIVE_ITEM = SELECTOR_ACTIVE + SELECTOR_ITEM;\nconst SELECTOR_ITEM_IMG = '.carousel-item img';\nconst SELECTOR_INDICATORS = '.carousel-indicators';\nconst SELECTOR_DATA_SLIDE = '[data-bs-slide], [data-bs-slide-to]';\nconst SELECTOR_DATA_RIDE = '[data-bs-ride=\"carousel\"]';\nconst KEY_TO_DIRECTION = {\n  [ARROW_LEFT_KEY$1]: DIRECTION_RIGHT,\n  [ARROW_RIGHT_KEY$1]: DIRECTION_LEFT\n};\nconst Default$b = {\n  interval: 5000,\n  keyboard: true,\n  pause: 'hover',\n  ride: false,\n  touch: true,\n  wrap: true\n};\nconst DefaultType$b = {\n  interval: '(number|boolean)',\n  // TODO:v6 remove boolean support\n  keyboard: 'boolean',\n  pause: '(string|boolean)',\n  ride: '(boolean|string)',\n  touch: 'boolean',\n  wrap: 'boolean'\n};\n\n/**\n * Class definition\n */\n\nclass Carousel extends BaseComponent {\n  constructor(element, config) {\n    super(element, config);\n    this._interval = null;\n    this._activeElement = null;\n    this._isSliding = false;\n    this.touchTimeout = null;\n    this._swipeHelper = null;\n    this._indicatorsElement = SelectorEngine.findOne(SELECTOR_INDICATORS, this._element);\n    this._addEventListeners();\n    if (this._config.ride === CLASS_NAME_CAROUSEL) {\n      this.cycle();\n    }\n  }\n\n  // Getters\n  static get Default() {\n    return Default$b;\n  }\n  static get DefaultType() {\n    return DefaultType$b;\n  }\n  static get NAME() {\n    return NAME$c;\n  }\n\n  // Public\n  next() {\n    this._slide(ORDER_NEXT);\n  }\n  nextWhenVisible() {\n    // FIXME TODO use `document.visibilityState`\n    // Don't call next when the page isn't visible\n    // or the carousel or its parent isn't visible\n    if (!document.hidden && isVisible(this._element)) {\n      this.next();\n    }\n  }\n  prev() {\n    this._slide(ORDER_PREV);\n  }\n  pause() {\n    if (this._isSliding) {\n      triggerTransitionEnd(this._element);\n    }\n    this._clearInterval();\n  }\n  cycle() {\n    this._clearInterval();\n    this._updateInterval();\n    this._interval = setInterval(() => this.nextWhenVisible(), this._config.interval);\n  }\n  _maybeEnableCycle() {\n    if (!this._config.ride) {\n      return;\n    }\n    if (this._isSliding) {\n      EventHandler.one(this._element, EVENT_SLID, () => this.cycle());\n      return;\n    }\n    this.cycle();\n  }\n  to(index) {\n    const items = this._getItems();\n    if (index > items.length - 1 || index < 0) {\n      return;\n    }\n    if (this._isSliding) {\n      EventHandler.one(this._element, EVENT_SLID, () => this.to(index));\n      return;\n    }\n    const activeIndex = this._getItemIndex(this._getActive());\n    if (activeIndex === index) {\n      return;\n    }\n    const order = index > activeIndex ? ORDER_NEXT : ORDER_PREV;\n    this._slide(order, items[index]);\n  }\n  dispose() {\n    if (this._swipeHelper) {\n      this._swipeHelper.dispose();\n    }\n    super.dispose();\n  }\n\n  // Private\n  _configAfterMerge(config) {\n    config.defaultInterval = config.interval;\n    return config;\n  }\n  _addEventListeners() {\n    if (this._config.keyboard) {\n      EventHandler.on(this._element, EVENT_KEYDOWN$1, event => this._keydown(event));\n    }\n    if (this._config.pause === 'hover') {\n      EventHandler.on(this._element, EVENT_MOUSEENTER$1, () => this.pause());\n      EventHandler.on(this._element, EVENT_MOUSELEAVE$1, () => this._maybeEnableCycle());\n    }\n    if (this._config.touch && Swipe.isSupported()) {\n      this._addTouchEventListeners();\n    }\n  }\n  _addTouchEventListeners() {\n    for (const img of SelectorEngine.find(SELECTOR_ITEM_IMG, this._element)) {\n      EventHandler.on(img, EVENT_DRAG_START, event => event.preventDefault());\n    }\n    const endCallBack = () => {\n      if (this._config.pause !== 'hover') {\n        return;\n      }\n\n      // If it's a touch-enabled device, mouseenter/leave are fired as\n      // part of the mouse compatibility events on first tap - the carousel\n      // would stop cycling until user tapped out of it;\n      // here, we listen for touchend, explicitly pause the carousel\n      // (as if it's the second time we tap on it, mouseenter compat event\n      // is NOT fired) and after a timeout (to allow for mouse compatibility\n      // events to fire) we explicitly restart cycling\n\n      this.pause();\n      if (this.touchTimeout) {\n        clearTimeout(this.touchTimeout);\n      }\n      this.touchTimeout = setTimeout(() => this._maybeEnableCycle(), TOUCHEVENT_COMPAT_WAIT + this._config.interval);\n    };\n    const swipeConfig = {\n      leftCallback: () => this._slide(this._directionToOrder(DIRECTION_LEFT)),\n      rightCallback: () => this._slide(this._directionToOrder(DIRECTION_RIGHT)),\n      endCallback: endCallBack\n    };\n    this._swipeHelper = new Swipe(this._element, swipeConfig);\n  }\n  _keydown(event) {\n    if (/input|textarea/i.test(event.target.tagName)) {\n      return;\n    }\n    const direction = KEY_TO_DIRECTION[event.key];\n    if (direction) {\n      event.preventDefault();\n      this._slide(this._directionToOrder(direction));\n    }\n  }\n  _getItemIndex(element) {\n    return this._getItems().indexOf(element);\n  }\n  _setActiveIndicatorElement(index) {\n    if (!this._indicatorsElement) {\n      return;\n    }\n    const activeIndicator = SelectorEngine.findOne(SELECTOR_ACTIVE, this._indicatorsElement);\n    activeIndicator.classList.remove(CLASS_NAME_ACTIVE$2);\n    activeIndicator.removeAttribute('aria-current');\n    const newActiveIndicator = SelectorEngine.findOne(`[data-bs-slide-to=\"${index}\"]`, this._indicatorsElement);\n    if (newActiveIndicator) {\n      newActiveIndicator.classList.add(CLASS_NAME_ACTIVE$2);\n      newActiveIndicator.setAttribute('aria-current', 'true');\n    }\n  }\n  _updateInterval() {\n    const element = this._activeElement || this._getActive();\n    if (!element) {\n      return;\n    }\n    const elementInterval = Number.parseInt(element.getAttribute('data-bs-interval'), 10);\n    this._config.interval = elementInterval || this._config.defaultInterval;\n  }\n  _slide(order, element = null) {\n    if (this._isSliding) {\n      return;\n    }\n    const activeElement = this._getActive();\n    const isNext = order === ORDER_NEXT;\n    const nextElement = element || getNextActiveElement(this._getItems(), activeElement, isNext, this._config.wrap);\n    if (nextElement === activeElement) {\n      return;\n    }\n    const nextElementIndex = this._getItemIndex(nextElement);\n    const triggerEvent = eventName => {\n      return EventHandler.trigger(this._element, eventName, {\n        relatedTarget: nextElement,\n        direction: this._orderToDirection(order),\n        from: this._getItemIndex(activeElement),\n        to: nextElementIndex\n      });\n    };\n    const slideEvent = triggerEvent(EVENT_SLIDE);\n    if (slideEvent.defaultPrevented) {\n      return;\n    }\n    if (!activeElement || !nextElement) {\n      // Some weirdness is happening, so we bail\n      // TODO: change tests that use empty divs to avoid this check\n      return;\n    }\n    const isCycling = Boolean(this._interval);\n    this.pause();\n    this._isSliding = true;\n    this._setActiveIndicatorElement(nextElementIndex);\n    this._activeElement = nextElement;\n    const directionalClassName = isNext ? CLASS_NAME_START : CLASS_NAME_END;\n    const orderClassName = isNext ? CLASS_NAME_NEXT : CLASS_NAME_PREV;\n    nextElement.classList.add(orderClassName);\n    reflow(nextElement);\n    activeElement.classList.add(directionalClassName);\n    nextElement.classList.add(directionalClassName);\n    const completeCallBack = () => {\n      nextElement.classList.remove(directionalClassName, orderClassName);\n      nextElement.classList.add(CLASS_NAME_ACTIVE$2);\n      activeElement.classList.remove(CLASS_NAME_ACTIVE$2, orderClassName, directionalClassName);\n      this._isSliding = false;\n      triggerEvent(EVENT_SLID);\n    };\n    this._queueCallback(completeCallBack, activeElement, this._isAnimated());\n    if (isCycling) {\n      this.cycle();\n    }\n  }\n  _isAnimated() {\n    return this._element.classList.contains(CLASS_NAME_SLIDE);\n  }\n  _getActive() {\n    return SelectorEngine.findOne(SELECTOR_ACTIVE_ITEM, this._element);\n  }\n  _getItems() {\n    return SelectorEngine.find(SELECTOR_ITEM, this._element);\n  }\n  _clearInterval() {\n    if (this._interval) {\n      clearInterval(this._interval);\n      this._interval = null;\n    }\n  }\n  _directionToOrder(direction) {\n    if (isRTL()) {\n      return direction === DIRECTION_LEFT ? ORDER_PREV : ORDER_NEXT;\n    }\n    return direction === DIRECTION_LEFT ? ORDER_NEXT : ORDER_PREV;\n  }\n  _orderToDirection(order) {\n    if (isRTL()) {\n      return order === ORDER_PREV ? DIRECTION_LEFT : DIRECTION_RIGHT;\n    }\n    return order === ORDER_PREV ? DIRECTION_RIGHT : DIRECTION_LEFT;\n  }\n\n  // Static\n  static jQueryInterface(config) {\n    return this.each(function () {\n      const data = Carousel.getOrCreateInstance(this, config);\n      if (typeof config === 'number') {\n        data.to(config);\n        return;\n      }\n      if (typeof config === 'string') {\n        if (data[config] === undefined || config.startsWith('_') || config === 'constructor') {\n          throw new TypeError(`No method named \"${config}\"`);\n        }\n        data[config]();\n      }\n    });\n  }\n}\n\n/**\n * Data API implementation\n */\n\nEventHandler.on(document, EVENT_CLICK_DATA_API$5, SELECTOR_DATA_SLIDE, function (event) {\n  const target = SelectorEngine.getElementFromSelector(this);\n  if (!target || !target.classList.contains(CLASS_NAME_CAROUSEL)) {\n    return;\n  }\n  event.preventDefault();\n  const carousel = Carousel.getOrCreateInstance(target);\n  const slideIndex = this.getAttribute('data-bs-slide-to');\n  if (slideIndex) {\n    carousel.to(slideIndex);\n    carousel._maybeEnableCycle();\n    return;\n  }\n  if (Manipulator.getDataAttribute(this, 'slide') === 'next') {\n    carousel.next();\n    carousel._maybeEnableCycle();\n    return;\n  }\n  carousel.prev();\n  carousel._maybeEnableCycle();\n});\nEventHandler.on(window, EVENT_LOAD_DATA_API$3, () => {\n  const carousels = SelectorEngine.find(SELECTOR_DATA_RIDE);\n  for (const carousel of carousels) {\n    Carousel.getOrCreateInstance(carousel);\n  }\n});\n\n/**\n * jQuery\n */\n\ndefineJQueryPlugin(Carousel);\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap collapse.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$b = 'collapse';\nconst DATA_KEY$7 = 'bs.collapse';\nconst EVENT_KEY$7 = `.${DATA_KEY$7}`;\nconst DATA_API_KEY$4 = '.data-api';\nconst EVENT_SHOW$6 = `show${EVENT_KEY$7}`;\nconst EVENT_SHOWN$6 = `shown${EVENT_KEY$7}`;\nconst EVENT_HIDE$6 = `hide${EVENT_KEY$7}`;\nconst EVENT_HIDDEN$6 = `hidden${EVENT_KEY$7}`;\nconst EVENT_CLICK_DATA_API$4 = `click${EVENT_KEY$7}${DATA_API_KEY$4}`;\nconst CLASS_NAME_SHOW$7 = 'show';\nconst CLASS_NAME_COLLAPSE = 'collapse';\nconst CLASS_NAME_COLLAPSING = 'collapsing';\nconst CLASS_NAME_COLLAPSED = 'collapsed';\nconst CLASS_NAME_DEEPER_CHILDREN = `:scope .${CLASS_NAME_COLLAPSE} .${CLASS_NAME_COLLAPSE}`;\nconst CLASS_NAME_HORIZONTAL = 'collapse-horizontal';\nconst WIDTH = 'width';\nconst HEIGHT = 'height';\nconst SELECTOR_ACTIVES = '.collapse.show, .collapse.collapsing';\nconst SELECTOR_DATA_TOGGLE$4 = '[data-bs-toggle=\"collapse\"]';\nconst Default$a = {\n  parent: null,\n  toggle: true\n};\nconst DefaultType$a = {\n  parent: '(null|element)',\n  toggle: 'boolean'\n};\n\n/**\n * Class definition\n */\n\nclass Collapse extends BaseComponent {\n  constructor(element, config) {\n    super(element, config);\n    this._isTransitioning = false;\n    this._triggerArray = [];\n    const toggleList = SelectorEngine.find(SELECTOR_DATA_TOGGLE$4);\n    for (const elem of toggleList) {\n      const selector = SelectorEngine.getSelectorFromElement(elem);\n      const filterElement = SelectorEngine.find(selector).filter(foundElement => foundElement === this._element);\n      if (selector !== null && filterElement.length) {\n        this._triggerArray.push(elem);\n      }\n    }\n    this._initializeChildren();\n    if (!this._config.parent) {\n      this._addAriaAndCollapsedClass(this._triggerArray, this._isShown());\n    }\n    if (this._config.toggle) {\n      this.toggle();\n    }\n  }\n\n  // Getters\n  static get Default() {\n    return Default$a;\n  }\n  static get DefaultType() {\n    return DefaultType$a;\n  }\n  static get NAME() {\n    return NAME$b;\n  }\n\n  // Public\n  toggle() {\n    if (this._isShown()) {\n      this.hide();\n    } else {\n      this.show();\n    }\n  }\n  show() {\n    if (this._isTransitioning || this._isShown()) {\n      return;\n    }\n    let activeChildren = [];\n\n    // find active children\n    if (this._config.parent) {\n      activeChildren = this._getFirstLevelChildren(SELECTOR_ACTIVES).filter(element => element !== this._element).map(element => Collapse.getOrCreateInstance(element, {\n        toggle: false\n      }));\n    }\n    if (activeChildren.length && activeChildren[0]._isTransitioning) {\n      return;\n    }\n    const startEvent = EventHandler.trigger(this._element, EVENT_SHOW$6);\n    if (startEvent.defaultPrevented) {\n      return;\n    }\n    for (const activeInstance of activeChildren) {\n      activeInstance.hide();\n    }\n    const dimension = this._getDimension();\n    this._element.classList.remove(CLASS_NAME_COLLAPSE);\n    this._element.classList.add(CLASS_NAME_COLLAPSING);\n    this._element.style[dimension] = 0;\n    this._addAriaAndCollapsedClass(this._triggerArray, true);\n    this._isTransitioning = true;\n    const complete = () => {\n      this._isTransitioning = false;\n      this._element.classList.remove(CLASS_NAME_COLLAPSING);\n      this._element.classList.add(CLASS_NAME_COLLAPSE, CLASS_NAME_SHOW$7);\n      this._element.style[dimension] = '';\n      EventHandler.trigger(this._element, EVENT_SHOWN$6);\n    };\n    const capitalizedDimension = dimension[0].toUpperCase() + dimension.slice(1);\n    const scrollSize = `scroll${capitalizedDimension}`;\n    this._queueCallback(complete, this._element, true);\n    this._element.style[dimension] = `${this._element[scrollSize]}px`;\n  }\n  hide() {\n    if (this._isTransitioning || !this._isShown()) {\n      return;\n    }\n    const startEvent = EventHandler.trigger(this._element, EVENT_HIDE$6);\n    if (startEvent.defaultPrevented) {\n      return;\n    }\n    const dimension = this._getDimension();\n    this._element.style[dimension] = `${this._element.getBoundingClientRect()[dimension]}px`;\n    reflow(this._element);\n    this._element.classList.add(CLASS_NAME_COLLAPSING);\n    this._element.classList.remove(CLASS_NAME_COLLAPSE, CLASS_NAME_SHOW$7);\n    for (const trigger of this._triggerArray) {\n      const element = SelectorEngine.getElementFromSelector(trigger);\n      if (element && !this._isShown(element)) {\n        this._addAriaAndCollapsedClass([trigger], false);\n      }\n    }\n    this._isTransitioning = true;\n    const complete = () => {\n      this._isTransitioning = false;\n      this._element.classList.remove(CLASS_NAME_COLLAPSING);\n      this._element.classList.add(CLASS_NAME_COLLAPSE);\n      EventHandler.trigger(this._element, EVENT_HIDDEN$6);\n    };\n    this._element.style[dimension] = '';\n    this._queueCallback(complete, this._element, true);\n  }\n  _isShown(element = this._element) {\n    return element.classList.contains(CLASS_NAME_SHOW$7);\n  }\n\n  // Private\n  _configAfterMerge(config) {\n    config.toggle = Boolean(config.toggle); // Coerce string values\n    config.parent = getElement(config.parent);\n    return config;\n  }\n  _getDimension() {\n    return this._element.classList.contains(CLASS_NAME_HORIZONTAL) ? WIDTH : HEIGHT;\n  }\n  _initializeChildren() {\n    if (!this._config.parent) {\n      return;\n    }\n    const children = this._getFirstLevelChildren(SELECTOR_DATA_TOGGLE$4);\n    for (const element of children) {\n      const selected = SelectorEngine.getElementFromSelector(element);\n      if (selected) {\n        this._addAriaAndCollapsedClass([element], this._isShown(selected));\n      }\n    }\n  }\n  _getFirstLevelChildren(selector) {\n    const children = SelectorEngine.find(CLASS_NAME_DEEPER_CHILDREN, this._config.parent);\n    // remove children if greater depth\n    return SelectorEngine.find(selector, this._config.parent).filter(element => !children.includes(element));\n  }\n  _addAriaAndCollapsedClass(triggerArray, isOpen) {\n    if (!triggerArray.length) {\n      return;\n    }\n    for (const element of triggerArray) {\n      element.classList.toggle(CLASS_NAME_COLLAPSED, !isOpen);\n      element.setAttribute('aria-expanded', isOpen);\n    }\n  }\n\n  // Static\n  static jQueryInterface(config) {\n    const _config = {};\n    if (typeof config === 'string' && /show|hide/.test(config)) {\n      _config.toggle = false;\n    }\n    return this.each(function () {\n      const data = Collapse.getOrCreateInstance(this, _config);\n      if (typeof config === 'string') {\n        if (typeof data[config] === 'undefined') {\n          throw new TypeError(`No method named \"${config}\"`);\n        }\n        data[config]();\n      }\n    });\n  }\n}\n\n/**\n * Data API implementation\n */\n\nEventHandler.on(document, EVENT_CLICK_DATA_API$4, SELECTOR_DATA_TOGGLE$4, function (event) {\n  // preventDefault only for <a> elements (which change the URL) not inside the collapsible element\n  if (event.target.tagName === 'A' || event.delegateTarget && event.delegateTarget.tagName === 'A') {\n    event.preventDefault();\n  }\n  for (const element of SelectorEngine.getMultipleElementsFromSelector(this)) {\n    Collapse.getOrCreateInstance(element, {\n      toggle: false\n    }).toggle();\n  }\n});\n\n/**\n * jQuery\n */\n\ndefineJQueryPlugin(Collapse);\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap dropdown.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$a = 'dropdown';\nconst DATA_KEY$6 = 'bs.dropdown';\nconst EVENT_KEY$6 = `.${DATA_KEY$6}`;\nconst DATA_API_KEY$3 = '.data-api';\nconst ESCAPE_KEY$2 = 'Escape';\nconst TAB_KEY$1 = 'Tab';\nconst ARROW_UP_KEY$1 = 'ArrowUp';\nconst ARROW_DOWN_KEY$1 = 'ArrowDown';\nconst RIGHT_MOUSE_BUTTON = 2; // MouseEvent.button value for the secondary button, usually the right button\n\nconst EVENT_HIDE$5 = `hide${EVENT_KEY$6}`;\nconst EVENT_HIDDEN$5 = `hidden${EVENT_KEY$6}`;\nconst EVENT_SHOW$5 = `show${EVENT_KEY$6}`;\nconst EVENT_SHOWN$5 = `shown${EVENT_KEY$6}`;\nconst EVENT_CLICK_DATA_API$3 = `click${EVENT_KEY$6}${DATA_API_KEY$3}`;\nconst EVENT_KEYDOWN_DATA_API = `keydown${EVENT_KEY$6}${DATA_API_KEY$3}`;\nconst EVENT_KEYUP_DATA_API = `keyup${EVENT_KEY$6}${DATA_API_KEY$3}`;\nconst CLASS_NAME_SHOW$6 = 'show';\nconst CLASS_NAME_DROPUP = 'dropup';\nconst CLASS_NAME_DROPEND = 'dropend';\nconst CLASS_NAME_DROPSTART = 'dropstart';\nconst CLASS_NAME_DROPUP_CENTER = 'dropup-center';\nconst CLASS_NAME_DROPDOWN_CENTER = 'dropdown-center';\nconst SELECTOR_DATA_TOGGLE$3 = '[data-bs-toggle=\"dropdown\"]:not(.disabled):not(:disabled)';\nconst SELECTOR_DATA_TOGGLE_SHOWN = `${SELECTOR_DATA_TOGGLE$3}.${CLASS_NAME_SHOW$6}`;\nconst SELECTOR_MENU = '.dropdown-menu';\nconst SELECTOR_NAVBAR = '.navbar';\nconst SELECTOR_NAVBAR_NAV = '.navbar-nav';\nconst SELECTOR_VISIBLE_ITEMS = '.dropdown-menu .dropdown-item:not(.disabled):not(:disabled)';\nconst PLACEMENT_TOP = isRTL() ? 'top-end' : 'top-start';\nconst PLACEMENT_TOPEND = isRTL() ? 'top-start' : 'top-end';\nconst PLACEMENT_BOTTOM = isRTL() ? 'bottom-end' : 'bottom-start';\nconst PLACEMENT_BOTTOMEND = isRTL() ? 'bottom-start' : 'bottom-end';\nconst PLACEMENT_RIGHT = isRTL() ? 'left-start' : 'right-start';\nconst PLACEMENT_LEFT = isRTL() ? 'right-start' : 'left-start';\nconst PLACEMENT_TOPCENTER = 'top';\nconst PLACEMENT_BOTTOMCENTER = 'bottom';\nconst Default$9 = {\n  autoClose: true,\n  boundary: 'clippingParents',\n  display: 'dynamic',\n  offset: [0, 2],\n  popperConfig: null,\n  reference: 'toggle'\n};\nconst DefaultType$9 = {\n  autoClose: '(boolean|string)',\n  boundary: '(string|element)',\n  display: 'string',\n  offset: '(array|string|function)',\n  popperConfig: '(null|object|function)',\n  reference: '(string|element|object)'\n};\n\n/**\n * Class definition\n */\n\nclass Dropdown extends BaseComponent {\n  constructor(element, config) {\n    super(element, config);\n    this._popper = null;\n    this._parent = this._element.parentNode; // dropdown wrapper\n    // TODO: v6 revert #37011 & change markup https://getbootstrap.com/docs/5.3/forms/input-group/\n    this._menu = SelectorEngine.next(this._element, SELECTOR_MENU)[0] || SelectorEngine.prev(this._element, SELECTOR_MENU)[0] || SelectorEngine.findOne(SELECTOR_MENU, this._parent);\n    this._inNavbar = this._detectNavbar();\n  }\n\n  // Getters\n  static get Default() {\n    return Default$9;\n  }\n  static get DefaultType() {\n    return DefaultType$9;\n  }\n  static get NAME() {\n    return NAME$a;\n  }\n\n  // Public\n  toggle() {\n    return this._isShown() ? this.hide() : this.show();\n  }\n  show() {\n    if (isDisabled(this._element) || this._isShown()) {\n      return;\n    }\n    const relatedTarget = {\n      relatedTarget: this._element\n    };\n    const showEvent = EventHandler.trigger(this._element, EVENT_SHOW$5, relatedTarget);\n    if (showEvent.defaultPrevented) {\n      return;\n    }\n    this._createPopper();\n\n    // If this is a touch-enabled device we add extra\n    // empty mouseover listeners to the body's immediate children;\n    // only needed because of broken event delegation on iOS\n    // https://www.quirksmode.org/blog/archives/2014/02/mouse_event_bub.html\n    if ('ontouchstart' in document.documentElement && !this._parent.closest(SELECTOR_NAVBAR_NAV)) {\n      for (const element of [].concat(...document.body.children)) {\n        EventHandler.on(element, 'mouseover', noop);\n      }\n    }\n    this._element.focus();\n    this._element.setAttribute('aria-expanded', true);\n    this._menu.classList.add(CLASS_NAME_SHOW$6);\n    this._element.classList.add(CLASS_NAME_SHOW$6);\n    EventHandler.trigger(this._element, EVENT_SHOWN$5, relatedTarget);\n  }\n  hide() {\n    if (isDisabled(this._element) || !this._isShown()) {\n      return;\n    }\n    const relatedTarget = {\n      relatedTarget: this._element\n    };\n    this._completeHide(relatedTarget);\n  }\n  dispose() {\n    if (this._popper) {\n      this._popper.destroy();\n    }\n    super.dispose();\n  }\n  update() {\n    this._inNavbar = this._detectNavbar();\n    if (this._popper) {\n      this._popper.update();\n    }\n  }\n\n  // Private\n  _completeHide(relatedTarget) {\n    const hideEvent = EventHandler.trigger(this._element, EVENT_HIDE$5, relatedTarget);\n    if (hideEvent.defaultPrevented) {\n      return;\n    }\n\n    // If this is a touch-enabled device we remove the extra\n    // empty mouseover listeners we added for iOS support\n    if ('ontouchstart' in document.documentElement) {\n      for (const element of [].concat(...document.body.children)) {\n        EventHandler.off(element, 'mouseover', noop);\n      }\n    }\n    if (this._popper) {\n      this._popper.destroy();\n    }\n    this._menu.classList.remove(CLASS_NAME_SHOW$6);\n    this._element.classList.remove(CLASS_NAME_SHOW$6);\n    this._element.setAttribute('aria-expanded', 'false');\n    Manipulator.removeDataAttribute(this._menu, 'popper');\n    EventHandler.trigger(this._element, EVENT_HIDDEN$5, relatedTarget);\n  }\n  _getConfig(config) {\n    config = super._getConfig(config);\n    if (typeof config.reference === 'object' && !isElement(config.reference) && typeof config.reference.getBoundingClientRect !== 'function') {\n      // Popper virtual elements require a getBoundingClientRect method\n      throw new TypeError(`${NAME$a.toUpperCase()}: Option \"reference\" provided type \"object\" without a required \"getBoundingClientRect\" method.`);\n    }\n    return config;\n  }\n  _createPopper() {\n    if (typeof Popper === 'undefined') {\n      throw new TypeError('Bootstrap\\'s dropdowns require Popper (https://popper.js.org)');\n    }\n    let referenceElement = this._element;\n    if (this._config.reference === 'parent') {\n      referenceElement = this._parent;\n    } else if (isElement(this._config.reference)) {\n      referenceElement = getElement(this._config.reference);\n    } else if (typeof this._config.reference === 'object') {\n      referenceElement = this._config.reference;\n    }\n    const popperConfig = this._getPopperConfig();\n    this._popper = Popper.createPopper(referenceElement, this._menu, popperConfig);\n  }\n  _isShown() {\n    return this._menu.classList.contains(CLASS_NAME_SHOW$6);\n  }\n  _getPlacement() {\n    const parentDropdown = this._parent;\n    if (parentDropdown.classList.contains(CLASS_NAME_DROPEND)) {\n      return PLACEMENT_RIGHT;\n    }\n    if (parentDropdown.classList.contains(CLASS_NAME_DROPSTART)) {\n      return PLACEMENT_LEFT;\n    }\n    if (parentDropdown.classList.contains(CLASS_NAME_DROPUP_CENTER)) {\n      return PLACEMENT_TOPCENTER;\n    }\n    if (parentDropdown.classList.contains(CLASS_NAME_DROPDOWN_CENTER)) {\n      return PLACEMENT_BOTTOMCENTER;\n    }\n\n    // We need to trim the value because custom properties can also include spaces\n    const isEnd = getComputedStyle(this._menu).getPropertyValue('--bs-position').trim() === 'end';\n    if (parentDropdown.classList.contains(CLASS_NAME_DROPUP)) {\n      return isEnd ? PLACEMENT_TOPEND : PLACEMENT_TOP;\n    }\n    return isEnd ? PLACEMENT_BOTTOMEND : PLACEMENT_BOTTOM;\n  }\n  _detectNavbar() {\n    return this._element.closest(SELECTOR_NAVBAR) !== null;\n  }\n  _getOffset() {\n    const {\n      offset\n    } = this._config;\n    if (typeof offset === 'string') {\n      return offset.split(',').map(value => Number.parseInt(value, 10));\n    }\n    if (typeof offset === 'function') {\n      return popperData => offset(popperData, this._element);\n    }\n    return offset;\n  }\n  _getPopperConfig() {\n    const defaultBsPopperConfig = {\n      placement: this._getPlacement(),\n      modifiers: [{\n        name: 'preventOverflow',\n        options: {\n          boundary: this._config.boundary\n        }\n      }, {\n        name: 'offset',\n        options: {\n          offset: this._getOffset()\n        }\n      }]\n    };\n\n    // Disable Popper if we have a static display or Dropdown is in Navbar\n    if (this._inNavbar || this._config.display === 'static') {\n      Manipulator.setDataAttribute(this._menu, 'popper', 'static'); // TODO: v6 remove\n      defaultBsPopperConfig.modifiers = [{\n        name: 'applyStyles',\n        enabled: false\n      }];\n    }\n    return {\n      ...defaultBsPopperConfig,\n      ...execute(this._config.popperConfig, [defaultBsPopperConfig])\n    };\n  }\n  _selectMenuItem({\n    key,\n    target\n  }) {\n    const items = SelectorEngine.find(SELECTOR_VISIBLE_ITEMS, this._menu).filter(element => isVisible(element));\n    if (!items.length) {\n      return;\n    }\n\n    // if target isn't included in items (e.g. when expanding the dropdown)\n    // allow cycling to get the last item in case key equals ARROW_UP_KEY\n    getNextActiveElement(items, target, key === ARROW_DOWN_KEY$1, !items.includes(target)).focus();\n  }\n\n  // Static\n  static jQueryInterface(config) {\n    return this.each(function () {\n      const data = Dropdown.getOrCreateInstance(this, config);\n      if (typeof config !== 'string') {\n        return;\n      }\n      if (typeof data[config] === 'undefined') {\n        throw new TypeError(`No method named \"${config}\"`);\n      }\n      data[config]();\n    });\n  }\n  static clearMenus(event) {\n    if (event.button === RIGHT_MOUSE_BUTTON || event.type === 'keyup' && event.key !== TAB_KEY$1) {\n      return;\n    }\n    const openToggles = SelectorEngine.find(SELECTOR_DATA_TOGGLE_SHOWN);\n    for (const toggle of openToggles) {\n      const context = Dropdown.getInstance(toggle);\n      if (!context || context._config.autoClose === false) {\n        continue;\n      }\n      const composedPath = event.composedPath();\n      const isMenuTarget = composedPath.includes(context._menu);\n      if (composedPath.includes(context._element) || context._config.autoClose === 'inside' && !isMenuTarget || context._config.autoClose === 'outside' && isMenuTarget) {\n        continue;\n      }\n\n      // Tab navigation through the dropdown menu or events from contained inputs shouldn't close the menu\n      if (context._menu.contains(event.target) && (event.type === 'keyup' && event.key === TAB_KEY$1 || /input|select|option|textarea|form/i.test(event.target.tagName))) {\n        continue;\n      }\n      const relatedTarget = {\n        relatedTarget: context._element\n      };\n      if (event.type === 'click') {\n        relatedTarget.clickEvent = event;\n      }\n      context._completeHide(relatedTarget);\n    }\n  }\n  static dataApiKeydownHandler(event) {\n    // If not an UP | DOWN | ESCAPE key => not a dropdown command\n    // If input/textarea && if key is other than ESCAPE => not a dropdown command\n\n    const isInput = /input|textarea/i.test(event.target.tagName);\n    const isEscapeEvent = event.key === ESCAPE_KEY$2;\n    const isUpOrDownEvent = [ARROW_UP_KEY$1, ARROW_DOWN_KEY$1].includes(event.key);\n    if (!isUpOrDownEvent && !isEscapeEvent) {\n      return;\n    }\n    if (isInput && !isEscapeEvent) {\n      return;\n    }\n    event.preventDefault();\n\n    // TODO: v6 revert #37011 & change markup https://getbootstrap.com/docs/5.3/forms/input-group/\n    const getToggleButton = this.matches(SELECTOR_DATA_TOGGLE$3) ? this : SelectorEngine.prev(this, SELECTOR_DATA_TOGGLE$3)[0] || SelectorEngine.next(this, SELECTOR_DATA_TOGGLE$3)[0] || SelectorEngine.findOne(SELECTOR_DATA_TOGGLE$3, event.delegateTarget.parentNode);\n    const instance = Dropdown.getOrCreateInstance(getToggleButton);\n    if (isUpOrDownEvent) {\n      event.stopPropagation();\n      instance.show();\n      instance._selectMenuItem(event);\n      return;\n    }\n    if (instance._isShown()) {\n      // else is escape and we check if it is shown\n      event.stopPropagation();\n      instance.hide();\n      getToggleButton.focus();\n    }\n  }\n}\n\n/**\n * Data API implementation\n */\n\nEventHandler.on(document, EVENT_KEYDOWN_DATA_API, SELECTOR_DATA_TOGGLE$3, Dropdown.dataApiKeydownHandler);\nEventHandler.on(document, EVENT_KEYDOWN_DATA_API, SELECTOR_MENU, Dropdown.dataApiKeydownHandler);\nEventHandler.on(document, EVENT_CLICK_DATA_API$3, Dropdown.clearMenus);\nEventHandler.on(document, EVENT_KEYUP_DATA_API, Dropdown.clearMenus);\nEventHandler.on(document, EVENT_CLICK_DATA_API$3, SELECTOR_DATA_TOGGLE$3, function (event) {\n  event.preventDefault();\n  Dropdown.getOrCreateInstance(this).toggle();\n});\n\n/**\n * jQuery\n */\n\ndefineJQueryPlugin(Dropdown);\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap util/backdrop.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$9 = 'backdrop';\nconst CLASS_NAME_FADE$4 = 'fade';\nconst CLASS_NAME_SHOW$5 = 'show';\nconst EVENT_MOUSEDOWN = `mousedown.bs.${NAME$9}`;\nconst Default$8 = {\n  className: 'modal-backdrop',\n  clickCallback: null,\n  isAnimated: false,\n  isVisible: true,\n  // if false, we use the backdrop helper without adding any element to the dom\n  rootElement: 'body' // give the choice to place backdrop under different elements\n};\nconst DefaultType$8 = {\n  className: 'string',\n  clickCallback: '(function|null)',\n  isAnimated: 'boolean',\n  isVisible: 'boolean',\n  rootElement: '(element|string)'\n};\n\n/**\n * Class definition\n */\n\nclass Backdrop extends Config {\n  constructor(config) {\n    super();\n    this._config = this._getConfig(config);\n    this._isAppended = false;\n    this._element = null;\n  }\n\n  // Getters\n  static get Default() {\n    return Default$8;\n  }\n  static get DefaultType() {\n    return DefaultType$8;\n  }\n  static get NAME() {\n    return NAME$9;\n  }\n\n  // Public\n  show(callback) {\n    if (!this._config.isVisible) {\n      execute(callback);\n      return;\n    }\n    this._append();\n    const element = this._getElement();\n    if (this._config.isAnimated) {\n      reflow(element);\n    }\n    element.classList.add(CLASS_NAME_SHOW$5);\n    this._emulateAnimation(() => {\n      execute(callback);\n    });\n  }\n  hide(callback) {\n    if (!this._config.isVisible) {\n      execute(callback);\n      return;\n    }\n    this._getElement().classList.remove(CLASS_NAME_SHOW$5);\n    this._emulateAnimation(() => {\n      this.dispose();\n      execute(callback);\n    });\n  }\n  dispose() {\n    if (!this._isAppended) {\n      return;\n    }\n    EventHandler.off(this._element, EVENT_MOUSEDOWN);\n    this._element.remove();\n    this._isAppended = false;\n  }\n\n  // Private\n  _getElement() {\n    if (!this._element) {\n      const backdrop = document.createElement('div');\n      backdrop.className = this._config.className;\n      if (this._config.isAnimated) {\n        backdrop.classList.add(CLASS_NAME_FADE$4);\n      }\n      this._element = backdrop;\n    }\n    return this._element;\n  }\n  _configAfterMerge(config) {\n    // use getElement() with the default \"body\" to get a fresh Element on each instantiation\n    config.rootElement = getElement(config.rootElement);\n    return config;\n  }\n  _append() {\n    if (this._isAppended) {\n      return;\n    }\n    const element = this._getElement();\n    this._config.rootElement.append(element);\n    EventHandler.on(element, EVENT_MOUSEDOWN, () => {\n      execute(this._config.clickCallback);\n    });\n    this._isAppended = true;\n  }\n  _emulateAnimation(callback) {\n    executeAfterTransition(callback, this._getElement(), this._config.isAnimated);\n  }\n}\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap util/focustrap.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$8 = 'focustrap';\nconst DATA_KEY$5 = 'bs.focustrap';\nconst EVENT_KEY$5 = `.${DATA_KEY$5}`;\nconst EVENT_FOCUSIN$2 = `focusin${EVENT_KEY$5}`;\nconst EVENT_KEYDOWN_TAB = `keydown.tab${EVENT_KEY$5}`;\nconst TAB_KEY = 'Tab';\nconst TAB_NAV_FORWARD = 'forward';\nconst TAB_NAV_BACKWARD = 'backward';\nconst Default$7 = {\n  autofocus: true,\n  trapElement: null // The element to trap focus inside of\n};\nconst DefaultType$7 = {\n  autofocus: 'boolean',\n  trapElement: 'element'\n};\n\n/**\n * Class definition\n */\n\nclass FocusTrap extends Config {\n  constructor(config) {\n    super();\n    this._config = this._getConfig(config);\n    this._isActive = false;\n    this._lastTabNavDirection = null;\n  }\n\n  // Getters\n  static get Default() {\n    return Default$7;\n  }\n  static get DefaultType() {\n    return DefaultType$7;\n  }\n  static get NAME() {\n    return NAME$8;\n  }\n\n  // Public\n  activate() {\n    if (this._isActive) {\n      return;\n    }\n    if (this._config.autofocus) {\n      this._config.trapElement.focus();\n    }\n    EventHandler.off(document, EVENT_KEY$5); // guard against infinite focus loop\n    EventHandler.on(document, EVENT_FOCUSIN$2, event => this._handleFocusin(event));\n    EventHandler.on(document, EVENT_KEYDOWN_TAB, event => this._handleKeydown(event));\n    this._isActive = true;\n  }\n  deactivate() {\n    if (!this._isActive) {\n      return;\n    }\n    this._isActive = false;\n    EventHandler.off(document, EVENT_KEY$5);\n  }\n\n  // Private\n  _handleFocusin(event) {\n    const {\n      trapElement\n    } = this._config;\n    if (event.target === document || event.target === trapElement || trapElement.contains(event.target)) {\n      return;\n    }\n    const elements = SelectorEngine.focusableChildren(trapElement);\n    if (elements.length === 0) {\n      trapElement.focus();\n    } else if (this._lastTabNavDirection === TAB_NAV_BACKWARD) {\n      elements[elements.length - 1].focus();\n    } else {\n      elements[0].focus();\n    }\n  }\n  _handleKeydown(event) {\n    if (event.key !== TAB_KEY) {\n      return;\n    }\n    this._lastTabNavDirection = event.shiftKey ? TAB_NAV_BACKWARD : TAB_NAV_FORWARD;\n  }\n}\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap util/scrollBar.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst SELECTOR_FIXED_CONTENT = '.fixed-top, .fixed-bottom, .is-fixed, .sticky-top';\nconst SELECTOR_STICKY_CONTENT = '.sticky-top';\nconst PROPERTY_PADDING = 'padding-right';\nconst PROPERTY_MARGIN = 'margin-right';\n\n/**\n * Class definition\n */\n\nclass ScrollBarHelper {\n  constructor() {\n    this._element = document.body;\n  }\n\n  // Public\n  getWidth() {\n    // https://developer.mozilla.org/en-US/docs/Web/API/Window/innerWidth#usage_notes\n    const documentWidth = document.documentElement.clientWidth;\n    return Math.abs(window.innerWidth - documentWidth);\n  }\n  hide() {\n    const width = this.getWidth();\n    this._disableOverFlow();\n    // give padding to element to balance the hidden scrollbar width\n    this._setElementAttributes(this._element, PROPERTY_PADDING, calculatedValue => calculatedValue + width);\n    // trick: We adjust positive paddingRight and negative marginRight to sticky-top elements to keep showing fullwidth\n    this._setElementAttributes(SELECTOR_FIXED_CONTENT, PROPERTY_PADDING, calculatedValue => calculatedValue + width);\n    this._setElementAttributes(SELECTOR_STICKY_CONTENT, PROPERTY_MARGIN, calculatedValue => calculatedValue - width);\n  }\n  reset() {\n    this._resetElementAttributes(this._element, 'overflow');\n    this._resetElementAttributes(this._element, PROPERTY_PADDING);\n    this._resetElementAttributes(SELECTOR_FIXED_CONTENT, PROPERTY_PADDING);\n    this._resetElementAttributes(SELECTOR_STICKY_CONTENT, PROPERTY_MARGIN);\n  }\n  isOverflowing() {\n    return this.getWidth() > 0;\n  }\n\n  // Private\n  _disableOverFlow() {\n    this._saveInitialAttribute(this._element, 'overflow');\n    this._element.style.overflow = 'hidden';\n  }\n  _setElementAttributes(selector, styleProperty, callback) {\n    const scrollbarWidth = this.getWidth();\n    const manipulationCallBack = element => {\n      if (element !== this._element && window.innerWidth > element.clientWidth + scrollbarWidth) {\n        return;\n      }\n      this._saveInitialAttribute(element, styleProperty);\n      const calculatedValue = window.getComputedStyle(element).getPropertyValue(styleProperty);\n      element.style.setProperty(styleProperty, `${callback(Number.parseFloat(calculatedValue))}px`);\n    };\n    this._applyManipulationCallback(selector, manipulationCallBack);\n  }\n  _saveInitialAttribute(element, styleProperty) {\n    const actualValue = element.style.getPropertyValue(styleProperty);\n    if (actualValue) {\n      Manipulator.setDataAttribute(element, styleProperty, actualValue);\n    }\n  }\n  _resetElementAttributes(selector, styleProperty) {\n    const manipulationCallBack = element => {\n      const value = Manipulator.getDataAttribute(element, styleProperty);\n      // We only want to remove the property if the value is `null`; the value can also be zero\n      if (value === null) {\n        element.style.removeProperty(styleProperty);\n        return;\n      }\n      Manipulator.removeDataAttribute(element, styleProperty);\n      element.style.setProperty(styleProperty, value);\n    };\n    this._applyManipulationCallback(selector, manipulationCallBack);\n  }\n  _applyManipulationCallback(selector, callBack) {\n    if (isElement(selector)) {\n      callBack(selector);\n      return;\n    }\n    for (const sel of SelectorEngine.find(selector, this._element)) {\n      callBack(sel);\n    }\n  }\n}\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap modal.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$7 = 'modal';\nconst DATA_KEY$4 = 'bs.modal';\nconst EVENT_KEY$4 = `.${DATA_KEY$4}`;\nconst DATA_API_KEY$2 = '.data-api';\nconst ESCAPE_KEY$1 = 'Escape';\nconst EVENT_HIDE$4 = `hide${EVENT_KEY$4}`;\nconst EVENT_HIDE_PREVENTED$1 = `hidePrevented${EVENT_KEY$4}`;\nconst EVENT_HIDDEN$4 = `hidden${EVENT_KEY$4}`;\nconst EVENT_SHOW$4 = `show${EVENT_KEY$4}`;\nconst EVENT_SHOWN$4 = `shown${EVENT_KEY$4}`;\nconst EVENT_RESIZE$1 = `resize${EVENT_KEY$4}`;\nconst EVENT_CLICK_DISMISS = `click.dismiss${EVENT_KEY$4}`;\nconst EVENT_MOUSEDOWN_DISMISS = `mousedown.dismiss${EVENT_KEY$4}`;\nconst EVENT_KEYDOWN_DISMISS$1 = `keydown.dismiss${EVENT_KEY$4}`;\nconst EVENT_CLICK_DATA_API$2 = `click${EVENT_KEY$4}${DATA_API_KEY$2}`;\nconst CLASS_NAME_OPEN = 'modal-open';\nconst CLASS_NAME_FADE$3 = 'fade';\nconst CLASS_NAME_SHOW$4 = 'show';\nconst CLASS_NAME_STATIC = 'modal-static';\nconst OPEN_SELECTOR$1 = '.modal.show';\nconst SELECTOR_DIALOG = '.modal-dialog';\nconst SELECTOR_MODAL_BODY = '.modal-body';\nconst SELECTOR_DATA_TOGGLE$2 = '[data-bs-toggle=\"modal\"]';\nconst Default$6 = {\n  backdrop: true,\n  focus: true,\n  keyboard: true\n};\nconst DefaultType$6 = {\n  backdrop: '(boolean|string)',\n  focus: 'boolean',\n  keyboard: 'boolean'\n};\n\n/**\n * Class definition\n */\n\nclass Modal extends BaseComponent {\n  constructor(element, config) {\n    super(element, config);\n    this._dialog = SelectorEngine.findOne(SELECTOR_DIALOG, this._element);\n    this._backdrop = this._initializeBackDrop();\n    this._focustrap = this._initializeFocusTrap();\n    this._isShown = false;\n    this._isTransitioning = false;\n    this._scrollBar = new ScrollBarHelper();\n    this._addEventListeners();\n  }\n\n  // Getters\n  static get Default() {\n    return Default$6;\n  }\n  static get DefaultType() {\n    return DefaultType$6;\n  }\n  static get NAME() {\n    return NAME$7;\n  }\n\n  // Public\n  toggle(relatedTarget) {\n    return this._isShown ? this.hide() : this.show(relatedTarget);\n  }\n  show(relatedTarget) {\n    if (this._isShown || this._isTransitioning) {\n      return;\n    }\n    const showEvent = EventHandler.trigger(this._element, EVENT_SHOW$4, {\n      relatedTarget\n    });\n    if (showEvent.defaultPrevented) {\n      return;\n    }\n    this._isShown = true;\n    this._isTransitioning = true;\n    this._scrollBar.hide();\n    document.body.classList.add(CLASS_NAME_OPEN);\n    this._adjustDialog();\n    this._backdrop.show(() => this._showElement(relatedTarget));\n  }\n  hide() {\n    if (!this._isShown || this._isTransitioning) {\n      return;\n    }\n    const hideEvent = EventHandler.trigger(this._element, EVENT_HIDE$4);\n    if (hideEvent.defaultPrevented) {\n      return;\n    }\n    this._isShown = false;\n    this._isTransitioning = true;\n    this._focustrap.deactivate();\n    this._element.classList.remove(CLASS_NAME_SHOW$4);\n    this._queueCallback(() => this._hideModal(), this._element, this._isAnimated());\n  }\n  dispose() {\n    EventHandler.off(window, EVENT_KEY$4);\n    EventHandler.off(this._dialog, EVENT_KEY$4);\n    this._backdrop.dispose();\n    this._focustrap.deactivate();\n    super.dispose();\n  }\n  handleUpdate() {\n    this._adjustDialog();\n  }\n\n  // Private\n  _initializeBackDrop() {\n    return new Backdrop({\n      isVisible: Boolean(this._config.backdrop),\n      // 'static' option will be translated to true, and booleans will keep their value,\n      isAnimated: this._isAnimated()\n    });\n  }\n  _initializeFocusTrap() {\n    return new FocusTrap({\n      trapElement: this._element\n    });\n  }\n  _showElement(relatedTarget) {\n    // try to append dynamic modal\n    if (!document.body.contains(this._element)) {\n      document.body.append(this._element);\n    }\n    this._element.style.display = 'block';\n    this._element.removeAttribute('aria-hidden');\n    this._element.setAttribute('aria-modal', true);\n    this._element.setAttribute('role', 'dialog');\n    this._element.scrollTop = 0;\n    const modalBody = SelectorEngine.findOne(SELECTOR_MODAL_BODY, this._dialog);\n    if (modalBody) {\n      modalBody.scrollTop = 0;\n    }\n    reflow(this._element);\n    this._element.classList.add(CLASS_NAME_SHOW$4);\n    const transitionComplete = () => {\n      if (this._config.focus) {\n        this._focustrap.activate();\n      }\n      this._isTransitioning = false;\n      EventHandler.trigger(this._element, EVENT_SHOWN$4, {\n        relatedTarget\n      });\n    };\n    this._queueCallback(transitionComplete, this._dialog, this._isAnimated());\n  }\n  _addEventListeners() {\n    EventHandler.on(this._element, EVENT_KEYDOWN_DISMISS$1, event => {\n      if (event.key !== ESCAPE_KEY$1) {\n        return;\n      }\n      if (this._config.keyboard) {\n        this.hide();\n        return;\n      }\n      this._triggerBackdropTransition();\n    });\n    EventHandler.on(window, EVENT_RESIZE$1, () => {\n      if (this._isShown && !this._isTransitioning) {\n        this._adjustDialog();\n      }\n    });\n    EventHandler.on(this._element, EVENT_MOUSEDOWN_DISMISS, event => {\n      // a bad trick to segregate clicks that may start inside dialog but end outside, and avoid listen to scrollbar clicks\n      EventHandler.one(this._element, EVENT_CLICK_DISMISS, event2 => {\n        if (this._element !== event.target || this._element !== event2.target) {\n          return;\n        }\n        if (this._config.backdrop === 'static') {\n          this._triggerBackdropTransition();\n          return;\n        }\n        if (this._config.backdrop) {\n          this.hide();\n        }\n      });\n    });\n  }\n  _hideModal() {\n    this._element.style.display = 'none';\n    this._element.setAttribute('aria-hidden', true);\n    this._element.removeAttribute('aria-modal');\n    this._element.removeAttribute('role');\n    this._isTransitioning = false;\n    this._backdrop.hide(() => {\n      document.body.classList.remove(CLASS_NAME_OPEN);\n      this._resetAdjustments();\n      this._scrollBar.reset();\n      EventHandler.trigger(this._element, EVENT_HIDDEN$4);\n    });\n  }\n  _isAnimated() {\n    return this._element.classList.contains(CLASS_NAME_FADE$3);\n  }\n  _triggerBackdropTransition() {\n    const hideEvent = EventHandler.trigger(this._element, EVENT_HIDE_PREVENTED$1);\n    if (hideEvent.defaultPrevented) {\n      return;\n    }\n    const isModalOverflowing = this._element.scrollHeight > document.documentElement.clientHeight;\n    const initialOverflowY = this._element.style.overflowY;\n    // return if the following background transition hasn't yet completed\n    if (initialOverflowY === 'hidden' || this._element.classList.contains(CLASS_NAME_STATIC)) {\n      return;\n    }\n    if (!isModalOverflowing) {\n      this._element.style.overflowY = 'hidden';\n    }\n    this._element.classList.add(CLASS_NAME_STATIC);\n    this._queueCallback(() => {\n      this._element.classList.remove(CLASS_NAME_STATIC);\n      this._queueCallback(() => {\n        this._element.style.overflowY = initialOverflowY;\n      }, this._dialog);\n    }, this._dialog);\n    this._element.focus();\n  }\n\n  /**\n   * The following methods are used to handle overflowing modals\n   */\n\n  _adjustDialog() {\n    const isModalOverflowing = this._element.scrollHeight > document.documentElement.clientHeight;\n    const scrollbarWidth = this._scrollBar.getWidth();\n    const isBodyOverflowing = scrollbarWidth > 0;\n    if (isBodyOverflowing && !isModalOverflowing) {\n      const property = isRTL() ? 'paddingLeft' : 'paddingRight';\n      this._element.style[property] = `${scrollbarWidth}px`;\n    }\n    if (!isBodyOverflowing && isModalOverflowing) {\n      const property = isRTL() ? 'paddingRight' : 'paddingLeft';\n      this._element.style[property] = `${scrollbarWidth}px`;\n    }\n  }\n  _resetAdjustments() {\n    this._element.style.paddingLeft = '';\n    this._element.style.paddingRight = '';\n  }\n\n  // Static\n  static jQueryInterface(config, relatedTarget) {\n    return this.each(function () {\n      const data = Modal.getOrCreateInstance(this, config);\n      if (typeof config !== 'string') {\n        return;\n      }\n      if (typeof data[config] === 'undefined') {\n        throw new TypeError(`No method named \"${config}\"`);\n      }\n      data[config](relatedTarget);\n    });\n  }\n}\n\n/**\n * Data API implementation\n */\n\nEventHandler.on(document, EVENT_CLICK_DATA_API$2, SELECTOR_DATA_TOGGLE$2, function (event) {\n  const target = SelectorEngine.getElementFromSelector(this);\n  if (['A', 'AREA'].includes(this.tagName)) {\n    event.preventDefault();\n  }\n  EventHandler.one(target, EVENT_SHOW$4, showEvent => {\n    if (showEvent.defaultPrevented) {\n      // only register focus restorer if modal will actually get shown\n      return;\n    }\n    EventHandler.one(target, EVENT_HIDDEN$4, () => {\n      if (isVisible(this)) {\n        this.focus();\n      }\n    });\n  });\n\n  // avoid conflict when clicking modal toggler while another one is open\n  const alreadyOpen = SelectorEngine.findOne(OPEN_SELECTOR$1);\n  if (alreadyOpen) {\n    Modal.getInstance(alreadyOpen).hide();\n  }\n  const data = Modal.getOrCreateInstance(target);\n  data.toggle(this);\n});\nenableDismissTrigger(Modal);\n\n/**\n * jQuery\n */\n\ndefineJQueryPlugin(Modal);\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap offcanvas.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$6 = 'offcanvas';\nconst DATA_KEY$3 = 'bs.offcanvas';\nconst EVENT_KEY$3 = `.${DATA_KEY$3}`;\nconst DATA_API_KEY$1 = '.data-api';\nconst EVENT_LOAD_DATA_API$2 = `load${EVENT_KEY$3}${DATA_API_KEY$1}`;\nconst ESCAPE_KEY = 'Escape';\nconst CLASS_NAME_SHOW$3 = 'show';\nconst CLASS_NAME_SHOWING$1 = 'showing';\nconst CLASS_NAME_HIDING = 'hiding';\nconst CLASS_NAME_BACKDROP = 'offcanvas-backdrop';\nconst OPEN_SELECTOR = '.offcanvas.show';\nconst EVENT_SHOW$3 = `show${EVENT_KEY$3}`;\nconst EVENT_SHOWN$3 = `shown${EVENT_KEY$3}`;\nconst EVENT_HIDE$3 = `hide${EVENT_KEY$3}`;\nconst EVENT_HIDE_PREVENTED = `hidePrevented${EVENT_KEY$3}`;\nconst EVENT_HIDDEN$3 = `hidden${EVENT_KEY$3}`;\nconst EVENT_RESIZE = `resize${EVENT_KEY$3}`;\nconst EVENT_CLICK_DATA_API$1 = `click${EVENT_KEY$3}${DATA_API_KEY$1}`;\nconst EVENT_KEYDOWN_DISMISS = `keydown.dismiss${EVENT_KEY$3}`;\nconst SELECTOR_DATA_TOGGLE$1 = '[data-bs-toggle=\"offcanvas\"]';\nconst Default$5 = {\n  backdrop: true,\n  keyboard: true,\n  scroll: false\n};\nconst DefaultType$5 = {\n  backdrop: '(boolean|string)',\n  keyboard: 'boolean',\n  scroll: 'boolean'\n};\n\n/**\n * Class definition\n */\n\nclass Offcanvas extends BaseComponent {\n  constructor(element, config) {\n    super(element, config);\n    this._isShown = false;\n    this._backdrop = this._initializeBackDrop();\n    this._focustrap = this._initializeFocusTrap();\n    this._addEventListeners();\n  }\n\n  // Getters\n  static get Default() {\n    return Default$5;\n  }\n  static get DefaultType() {\n    return DefaultType$5;\n  }\n  static get NAME() {\n    return NAME$6;\n  }\n\n  // Public\n  toggle(relatedTarget) {\n    return this._isShown ? this.hide() : this.show(relatedTarget);\n  }\n  show(relatedTarget) {\n    if (this._isShown) {\n      return;\n    }\n    const showEvent = EventHandler.trigger(this._element, EVENT_SHOW$3, {\n      relatedTarget\n    });\n    if (showEvent.defaultPrevented) {\n      return;\n    }\n    this._isShown = true;\n    this._backdrop.show();\n    if (!this._config.scroll) {\n      new ScrollBarHelper().hide();\n    }\n    this._element.setAttribute('aria-modal', true);\n    this._element.setAttribute('role', 'dialog');\n    this._element.classList.add(CLASS_NAME_SHOWING$1);\n    const completeCallBack = () => {\n      if (!this._config.scroll || this._config.backdrop) {\n        this._focustrap.activate();\n      }\n      this._element.classList.add(CLASS_NAME_SHOW$3);\n      this._element.classList.remove(CLASS_NAME_SHOWING$1);\n      EventHandler.trigger(this._element, EVENT_SHOWN$3, {\n        relatedTarget\n      });\n    };\n    this._queueCallback(completeCallBack, this._element, true);\n  }\n  hide() {\n    if (!this._isShown) {\n      return;\n    }\n    const hideEvent = EventHandler.trigger(this._element, EVENT_HIDE$3);\n    if (hideEvent.defaultPrevented) {\n      return;\n    }\n    this._focustrap.deactivate();\n    this._element.blur();\n    this._isShown = false;\n    this._element.classList.add(CLASS_NAME_HIDING);\n    this._backdrop.hide();\n    const completeCallback = () => {\n      this._element.classList.remove(CLASS_NAME_SHOW$3, CLASS_NAME_HIDING);\n      this._element.removeAttribute('aria-modal');\n      this._element.removeAttribute('role');\n      if (!this._config.scroll) {\n        new ScrollBarHelper().reset();\n      }\n      EventHandler.trigger(this._element, EVENT_HIDDEN$3);\n    };\n    this._queueCallback(completeCallback, this._element, true);\n  }\n  dispose() {\n    this._backdrop.dispose();\n    this._focustrap.deactivate();\n    super.dispose();\n  }\n\n  // Private\n  _initializeBackDrop() {\n    const clickCallback = () => {\n      if (this._config.backdrop === 'static') {\n        EventHandler.trigger(this._element, EVENT_HIDE_PREVENTED);\n        return;\n      }\n      this.hide();\n    };\n\n    // 'static' option will be translated to true, and booleans will keep their value\n    const isVisible = Boolean(this._config.backdrop);\n    return new Backdrop({\n      className: CLASS_NAME_BACKDROP,\n      isVisible,\n      isAnimated: true,\n      rootElement: this._element.parentNode,\n      clickCallback: isVisible ? clickCallback : null\n    });\n  }\n  _initializeFocusTrap() {\n    return new FocusTrap({\n      trapElement: this._element\n    });\n  }\n  _addEventListeners() {\n    EventHandler.on(this._element, EVENT_KEYDOWN_DISMISS, event => {\n      if (event.key !== ESCAPE_KEY) {\n        return;\n      }\n      if (this._config.keyboard) {\n        this.hide();\n        return;\n      }\n      EventHandler.trigger(this._element, EVENT_HIDE_PREVENTED);\n    });\n  }\n\n  // Static\n  static jQueryInterface(config) {\n    return this.each(function () {\n      const data = Offcanvas.getOrCreateInstance(this, config);\n      if (typeof config !== 'string') {\n        return;\n      }\n      if (data[config] === undefined || config.startsWith('_') || config === 'constructor') {\n        throw new TypeError(`No method named \"${config}\"`);\n      }\n      data[config](this);\n    });\n  }\n}\n\n/**\n * Data API implementation\n */\n\nEventHandler.on(document, EVENT_CLICK_DATA_API$1, SELECTOR_DATA_TOGGLE$1, function (event) {\n  const target = SelectorEngine.getElementFromSelector(this);\n  if (['A', 'AREA'].includes(this.tagName)) {\n    event.preventDefault();\n  }\n  if (isDisabled(this)) {\n    return;\n  }\n  EventHandler.one(target, EVENT_HIDDEN$3, () => {\n    // focus on trigger when it is closed\n    if (isVisible(this)) {\n      this.focus();\n    }\n  });\n\n  // avoid conflict when clicking a toggler of an offcanvas, while another is open\n  const alreadyOpen = SelectorEngine.findOne(OPEN_SELECTOR);\n  if (alreadyOpen && alreadyOpen !== target) {\n    Offcanvas.getInstance(alreadyOpen).hide();\n  }\n  const data = Offcanvas.getOrCreateInstance(target);\n  data.toggle(this);\n});\nEventHandler.on(window, EVENT_LOAD_DATA_API$2, () => {\n  for (const selector of SelectorEngine.find(OPEN_SELECTOR)) {\n    Offcanvas.getOrCreateInstance(selector).show();\n  }\n});\nEventHandler.on(window, EVENT_RESIZE, () => {\n  for (const element of SelectorEngine.find('[aria-modal][class*=show][class*=offcanvas-]')) {\n    if (getComputedStyle(element).position !== 'fixed') {\n      Offcanvas.getOrCreateInstance(element).hide();\n    }\n  }\n});\nenableDismissTrigger(Offcanvas);\n\n/**\n * jQuery\n */\n\ndefineJQueryPlugin(Offcanvas);\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap util/sanitizer.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n// js-docs-start allow-list\nconst ARIA_ATTRIBUTE_PATTERN = /^aria-[\\w-]*$/i;\nconst DefaultAllowlist = {\n  // Global attributes allowed on any supplied element below.\n  '*': ['class', 'dir', 'id', 'lang', 'role', ARIA_ATTRIBUTE_PATTERN],\n  a: ['target', 'href', 'title', 'rel'],\n  area: [],\n  b: [],\n  br: [],\n  col: [],\n  code: [],\n  dd: [],\n  div: [],\n  dl: [],\n  dt: [],\n  em: [],\n  hr: [],\n  h1: [],\n  h2: [],\n  h3: [],\n  h4: [],\n  h5: [],\n  h6: [],\n  i: [],\n  img: ['src', 'srcset', 'alt', 'title', 'width', 'height'],\n  li: [],\n  ol: [],\n  p: [],\n  pre: [],\n  s: [],\n  small: [],\n  span: [],\n  sub: [],\n  sup: [],\n  strong: [],\n  u: [],\n  ul: []\n};\n// js-docs-end allow-list\n\nconst uriAttributes = new Set(['background', 'cite', 'href', 'itemtype', 'longdesc', 'poster', 'src', 'xlink:href']);\n\n/**\n * A pattern that recognizes URLs that are safe wrt. XSS in URL navigation\n * contexts.\n *\n * Shout-out to Angular https://github.com/angular/angular/blob/15.2.8/packages/core/src/sanitization/url_sanitizer.ts#L38\n */\n// eslint-disable-next-line unicorn/better-regex\nconst SAFE_URL_PATTERN = /^(?!javascript:)(?:[a-z0-9+.-]+:|[^&:/?#]*(?:[/?#]|$))/i;\nconst allowedAttribute = (attribute, allowedAttributeList) => {\n  const attributeName = attribute.nodeName.toLowerCase();\n  if (allowedAttributeList.includes(attributeName)) {\n    if (uriAttributes.has(attributeName)) {\n      return Boolean(SAFE_URL_PATTERN.test(attribute.nodeValue));\n    }\n    return true;\n  }\n\n  // Check if a regular expression validates the attribute.\n  return allowedAttributeList.filter(attributeRegex => attributeRegex instanceof RegExp).some(regex => regex.test(attributeName));\n};\nfunction sanitizeHtml(unsafeHtml, allowList, sanitizeFunction) {\n  if (!unsafeHtml.length) {\n    return unsafeHtml;\n  }\n  if (sanitizeFunction && typeof sanitizeFunction === 'function') {\n    return sanitizeFunction(unsafeHtml);\n  }\n  const domParser = new window.DOMParser();\n  const createdDocument = domParser.parseFromString(unsafeHtml, 'text/html');\n  const elements = [].concat(...createdDocument.body.querySelectorAll('*'));\n  for (const element of elements) {\n    const elementName = element.nodeName.toLowerCase();\n    if (!Object.keys(allowList).includes(elementName)) {\n      element.remove();\n      continue;\n    }\n    const attributeList = [].concat(...element.attributes);\n    const allowedAttributes = [].concat(allowList['*'] || [], allowList[elementName] || []);\n    for (const attribute of attributeList) {\n      if (!allowedAttribute(attribute, allowedAttributes)) {\n        element.removeAttribute(attribute.nodeName);\n      }\n    }\n  }\n  return createdDocument.body.innerHTML;\n}\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap util/template-factory.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$5 = 'TemplateFactory';\nconst Default$4 = {\n  allowList: DefaultAllowlist,\n  content: {},\n  // { selector : text ,  selector2 : text2 , }\n  extraClass: '',\n  html: false,\n  sanitize: true,\n  sanitizeFn: null,\n  template: '<div></div>'\n};\nconst DefaultType$4 = {\n  allowList: 'object',\n  content: 'object',\n  extraClass: '(string|function)',\n  html: 'boolean',\n  sanitize: 'boolean',\n  sanitizeFn: '(null|function)',\n  template: 'string'\n};\nconst DefaultContentType = {\n  entry: '(string|element|function|null)',\n  selector: '(string|element)'\n};\n\n/**\n * Class definition\n */\n\nclass TemplateFactory extends Config {\n  constructor(config) {\n    super();\n    this._config = this._getConfig(config);\n  }\n\n  // Getters\n  static get Default() {\n    return Default$4;\n  }\n  static get DefaultType() {\n    return DefaultType$4;\n  }\n  static get NAME() {\n    return NAME$5;\n  }\n\n  // Public\n  getContent() {\n    return Object.values(this._config.content).map(config => this._resolvePossibleFunction(config)).filter(Boolean);\n  }\n  hasContent() {\n    return this.getContent().length > 0;\n  }\n  changeContent(content) {\n    this._checkContent(content);\n    this._config.content = {\n      ...this._config.content,\n      ...content\n    };\n    return this;\n  }\n  toHtml() {\n    const templateWrapper = document.createElement('div');\n    templateWrapper.innerHTML = this._maybeSanitize(this._config.template);\n    for (const [selector, text] of Object.entries(this._config.content)) {\n      this._setContent(templateWrapper, text, selector);\n    }\n    const template = templateWrapper.children[0];\n    const extraClass = this._resolvePossibleFunction(this._config.extraClass);\n    if (extraClass) {\n      template.classList.add(...extraClass.split(' '));\n    }\n    return template;\n  }\n\n  // Private\n  _typeCheckConfig(config) {\n    super._typeCheckConfig(config);\n    this._checkContent(config.content);\n  }\n  _checkContent(arg) {\n    for (const [selector, content] of Object.entries(arg)) {\n      super._typeCheckConfig({\n        selector,\n        entry: content\n      }, DefaultContentType);\n    }\n  }\n  _setContent(template, content, selector) {\n    const templateElement = SelectorEngine.findOne(selector, template);\n    if (!templateElement) {\n      return;\n    }\n    content = this._resolvePossibleFunction(content);\n    if (!content) {\n      templateElement.remove();\n      return;\n    }\n    if (isElement(content)) {\n      this._putElementInTemplate(getElement(content), templateElement);\n      return;\n    }\n    if (this._config.html) {\n      templateElement.innerHTML = this._maybeSanitize(content);\n      return;\n    }\n    templateElement.textContent = content;\n  }\n  _maybeSanitize(arg) {\n    return this._config.sanitize ? sanitizeHtml(arg, this._config.allowList, this._config.sanitizeFn) : arg;\n  }\n  _resolvePossibleFunction(arg) {\n    return execute(arg, [this]);\n  }\n  _putElementInTemplate(element, templateElement) {\n    if (this._config.html) {\n      templateElement.innerHTML = '';\n      templateElement.append(element);\n      return;\n    }\n    templateElement.textContent = element.textContent;\n  }\n}\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap tooltip.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$4 = 'tooltip';\nconst DISALLOWED_ATTRIBUTES = new Set(['sanitize', 'allowList', 'sanitizeFn']);\nconst CLASS_NAME_FADE$2 = 'fade';\nconst CLASS_NAME_MODAL = 'modal';\nconst CLASS_NAME_SHOW$2 = 'show';\nconst SELECTOR_TOOLTIP_INNER = '.tooltip-inner';\nconst SELECTOR_MODAL = `.${CLASS_NAME_MODAL}`;\nconst EVENT_MODAL_HIDE = 'hide.bs.modal';\nconst TRIGGER_HOVER = 'hover';\nconst TRIGGER_FOCUS = 'focus';\nconst TRIGGER_CLICK = 'click';\nconst TRIGGER_MANUAL = 'manual';\nconst EVENT_HIDE$2 = 'hide';\nconst EVENT_HIDDEN$2 = 'hidden';\nconst EVENT_SHOW$2 = 'show';\nconst EVENT_SHOWN$2 = 'shown';\nconst EVENT_INSERTED = 'inserted';\nconst EVENT_CLICK$1 = 'click';\nconst EVENT_FOCUSIN$1 = 'focusin';\nconst EVENT_FOCUSOUT$1 = 'focusout';\nconst EVENT_MOUSEENTER = 'mouseenter';\nconst EVENT_MOUSELEAVE = 'mouseleave';\nconst AttachmentMap = {\n  AUTO: 'auto',\n  TOP: 'top',\n  RIGHT: isRTL() ? 'left' : 'right',\n  BOTTOM: 'bottom',\n  LEFT: isRTL() ? 'right' : 'left'\n};\nconst Default$3 = {\n  allowList: DefaultAllowlist,\n  animation: true,\n  boundary: 'clippingParents',\n  container: false,\n  customClass: '',\n  delay: 0,\n  fallbackPlacements: ['top', 'right', 'bottom', 'left'],\n  html: false,\n  offset: [0, 6],\n  placement: 'top',\n  popperConfig: null,\n  sanitize: true,\n  sanitizeFn: null,\n  selector: false,\n  template: '<div class=\"tooltip\" role=\"tooltip\">' + '<div class=\"tooltip-arrow\"></div>' + '<div class=\"tooltip-inner\"></div>' + '</div>',\n  title: '',\n  trigger: 'hover focus'\n};\nconst DefaultType$3 = {\n  allowList: 'object',\n  animation: 'boolean',\n  boundary: '(string|element)',\n  container: '(string|element|boolean)',\n  customClass: '(string|function)',\n  delay: '(number|object)',\n  fallbackPlacements: 'array',\n  html: 'boolean',\n  offset: '(array|string|function)',\n  placement: '(string|function)',\n  popperConfig: '(null|object|function)',\n  sanitize: 'boolean',\n  sanitizeFn: '(null|function)',\n  selector: '(string|boolean)',\n  template: 'string',\n  title: '(string|element|function)',\n  trigger: 'string'\n};\n\n/**\n * Class definition\n */\n\nclass Tooltip extends BaseComponent {\n  constructor(element, config) {\n    if (typeof Popper === 'undefined') {\n      throw new TypeError('Bootstrap\\'s tooltips require Popper (https://popper.js.org)');\n    }\n    super(element, config);\n\n    // Private\n    this._isEnabled = true;\n    this._timeout = 0;\n    this._isHovered = null;\n    this._activeTrigger = {};\n    this._popper = null;\n    this._templateFactory = null;\n    this._newContent = null;\n\n    // Protected\n    this.tip = null;\n    this._setListeners();\n    if (!this._config.selector) {\n      this._fixTitle();\n    }\n  }\n\n  // Getters\n  static get Default() {\n    return Default$3;\n  }\n  static get DefaultType() {\n    return DefaultType$3;\n  }\n  static get NAME() {\n    return NAME$4;\n  }\n\n  // Public\n  enable() {\n    this._isEnabled = true;\n  }\n  disable() {\n    this._isEnabled = false;\n  }\n  toggleEnabled() {\n    this._isEnabled = !this._isEnabled;\n  }\n  toggle() {\n    if (!this._isEnabled) {\n      return;\n    }\n    this._activeTrigger.click = !this._activeTrigger.click;\n    if (this._isShown()) {\n      this._leave();\n      return;\n    }\n    this._enter();\n  }\n  dispose() {\n    clearTimeout(this._timeout);\n    EventHandler.off(this._element.closest(SELECTOR_MODAL), EVENT_MODAL_HIDE, this._hideModalHandler);\n    if (this._element.getAttribute('data-bs-original-title')) {\n      this._element.setAttribute('title', this._element.getAttribute('data-bs-original-title'));\n    }\n    this._disposePopper();\n    super.dispose();\n  }\n  show() {\n    if (this._element.style.display === 'none') {\n      throw new Error('Please use show on visible elements');\n    }\n    if (!(this._isWithContent() && this._isEnabled)) {\n      return;\n    }\n    const showEvent = EventHandler.trigger(this._element, this.constructor.eventName(EVENT_SHOW$2));\n    const shadowRoot = findShadowRoot(this._element);\n    const isInTheDom = (shadowRoot || this._element.ownerDocument.documentElement).contains(this._element);\n    if (showEvent.defaultPrevented || !isInTheDom) {\n      return;\n    }\n\n    // TODO: v6 remove this or make it optional\n    this._disposePopper();\n    const tip = this._getTipElement();\n    this._element.setAttribute('aria-describedby', tip.getAttribute('id'));\n    const {\n      container\n    } = this._config;\n    if (!this._element.ownerDocument.documentElement.contains(this.tip)) {\n      container.append(tip);\n      EventHandler.trigger(this._element, this.constructor.eventName(EVENT_INSERTED));\n    }\n    this._popper = this._createPopper(tip);\n    tip.classList.add(CLASS_NAME_SHOW$2);\n\n    // If this is a touch-enabled device we add extra\n    // empty mouseover listeners to the body's immediate children;\n    // only needed because of broken event delegation on iOS\n    // https://www.quirksmode.org/blog/archives/2014/02/mouse_event_bub.html\n    if ('ontouchstart' in document.documentElement) {\n      for (const element of [].concat(...document.body.children)) {\n        EventHandler.on(element, 'mouseover', noop);\n      }\n    }\n    const complete = () => {\n      EventHandler.trigger(this._element, this.constructor.eventName(EVENT_SHOWN$2));\n      if (this._isHovered === false) {\n        this._leave();\n      }\n      this._isHovered = false;\n    };\n    this._queueCallback(complete, this.tip, this._isAnimated());\n  }\n  hide() {\n    if (!this._isShown()) {\n      return;\n    }\n    const hideEvent = EventHandler.trigger(this._element, this.constructor.eventName(EVENT_HIDE$2));\n    if (hideEvent.defaultPrevented) {\n      return;\n    }\n    const tip = this._getTipElement();\n    tip.classList.remove(CLASS_NAME_SHOW$2);\n\n    // If this is a touch-enabled device we remove the extra\n    // empty mouseover listeners we added for iOS support\n    if ('ontouchstart' in document.documentElement) {\n      for (const element of [].concat(...document.body.children)) {\n        EventHandler.off(element, 'mouseover', noop);\n      }\n    }\n    this._activeTrigger[TRIGGER_CLICK] = false;\n    this._activeTrigger[TRIGGER_FOCUS] = false;\n    this._activeTrigger[TRIGGER_HOVER] = false;\n    this._isHovered = null; // it is a trick to support manual triggering\n\n    const complete = () => {\n      if (this._isWithActiveTrigger()) {\n        return;\n      }\n      if (!this._isHovered) {\n        this._disposePopper();\n      }\n      this._element.removeAttribute('aria-describedby');\n      EventHandler.trigger(this._element, this.constructor.eventName(EVENT_HIDDEN$2));\n    };\n    this._queueCallback(complete, this.tip, this._isAnimated());\n  }\n  update() {\n    if (this._popper) {\n      this._popper.update();\n    }\n  }\n\n  // Protected\n  _isWithContent() {\n    return Boolean(this._getTitle());\n  }\n  _getTipElement() {\n    if (!this.tip) {\n      this.tip = this._createTipElement(this._newContent || this._getContentForTemplate());\n    }\n    return this.tip;\n  }\n  _createTipElement(content) {\n    const tip = this._getTemplateFactory(content).toHtml();\n\n    // TODO: remove this check in v6\n    if (!tip) {\n      return null;\n    }\n    tip.classList.remove(CLASS_NAME_FADE$2, CLASS_NAME_SHOW$2);\n    // TODO: v6 the following can be achieved with CSS only\n    tip.classList.add(`bs-${this.constructor.NAME}-auto`);\n    const tipId = getUID(this.constructor.NAME).toString();\n    tip.setAttribute('id', tipId);\n    if (this._isAnimated()) {\n      tip.classList.add(CLASS_NAME_FADE$2);\n    }\n    return tip;\n  }\n  setContent(content) {\n    this._newContent = content;\n    if (this._isShown()) {\n      this._disposePopper();\n      this.show();\n    }\n  }\n  _getTemplateFactory(content) {\n    if (this._templateFactory) {\n      this._templateFactory.changeContent(content);\n    } else {\n      this._templateFactory = new TemplateFactory({\n        ...this._config,\n        // the `content` var has to be after `this._config`\n        // to override config.content in case of popover\n        content,\n        extraClass: this._resolvePossibleFunction(this._config.customClass)\n      });\n    }\n    return this._templateFactory;\n  }\n  _getContentForTemplate() {\n    return {\n      [SELECTOR_TOOLTIP_INNER]: this._getTitle()\n    };\n  }\n  _getTitle() {\n    return this._resolvePossibleFunction(this._config.title) || this._element.getAttribute('data-bs-original-title');\n  }\n\n  // Private\n  _initializeOnDelegatedTarget(event) {\n    return this.constructor.getOrCreateInstance(event.delegateTarget, this._getDelegateConfig());\n  }\n  _isAnimated() {\n    return this._config.animation || this.tip && this.tip.classList.contains(CLASS_NAME_FADE$2);\n  }\n  _isShown() {\n    return this.tip && this.tip.classList.contains(CLASS_NAME_SHOW$2);\n  }\n  _createPopper(tip) {\n    const placement = execute(this._config.placement, [this, tip, this._element]);\n    const attachment = AttachmentMap[placement.toUpperCase()];\n    return Popper.createPopper(this._element, tip, this._getPopperConfig(attachment));\n  }\n  _getOffset() {\n    const {\n      offset\n    } = this._config;\n    if (typeof offset === 'string') {\n      return offset.split(',').map(value => Number.parseInt(value, 10));\n    }\n    if (typeof offset === 'function') {\n      return popperData => offset(popperData, this._element);\n    }\n    return offset;\n  }\n  _resolvePossibleFunction(arg) {\n    return execute(arg, [this._element]);\n  }\n  _getPopperConfig(attachment) {\n    const defaultBsPopperConfig = {\n      placement: attachment,\n      modifiers: [{\n        name: 'flip',\n        options: {\n          fallbackPlacements: this._config.fallbackPlacements\n        }\n      }, {\n        name: 'offset',\n        options: {\n          offset: this._getOffset()\n        }\n      }, {\n        name: 'preventOverflow',\n        options: {\n          boundary: this._config.boundary\n        }\n      }, {\n        name: 'arrow',\n        options: {\n          element: `.${this.constructor.NAME}-arrow`\n        }\n      }, {\n        name: 'preSetPlacement',\n        enabled: true,\n        phase: 'beforeMain',\n        fn: data => {\n          // Pre-set Popper's placement attribute in order to read the arrow sizes properly.\n          // Otherwise, Popper mixes up the width and height dimensions since the initial arrow style is for top placement\n          this._getTipElement().setAttribute('data-popper-placement', data.state.placement);\n        }\n      }]\n    };\n    return {\n      ...defaultBsPopperConfig,\n      ...execute(this._config.popperConfig, [defaultBsPopperConfig])\n    };\n  }\n  _setListeners() {\n    const triggers = this._config.trigger.split(' ');\n    for (const trigger of triggers) {\n      if (trigger === 'click') {\n        EventHandler.on(this._element, this.constructor.eventName(EVENT_CLICK$1), this._config.selector, event => {\n          const context = this._initializeOnDelegatedTarget(event);\n          context.toggle();\n        });\n      } else if (trigger !== TRIGGER_MANUAL) {\n        const eventIn = trigger === TRIGGER_HOVER ? this.constructor.eventName(EVENT_MOUSEENTER) : this.constructor.eventName(EVENT_FOCUSIN$1);\n        const eventOut = trigger === TRIGGER_HOVER ? this.constructor.eventName(EVENT_MOUSELEAVE) : this.constructor.eventName(EVENT_FOCUSOUT$1);\n        EventHandler.on(this._element, eventIn, this._config.selector, event => {\n          const context = this._initializeOnDelegatedTarget(event);\n          context._activeTrigger[event.type === 'focusin' ? TRIGGER_FOCUS : TRIGGER_HOVER] = true;\n          context._enter();\n        });\n        EventHandler.on(this._element, eventOut, this._config.selector, event => {\n          const context = this._initializeOnDelegatedTarget(event);\n          context._activeTrigger[event.type === 'focusout' ? TRIGGER_FOCUS : TRIGGER_HOVER] = context._element.contains(event.relatedTarget);\n          context._leave();\n        });\n      }\n    }\n    this._hideModalHandler = () => {\n      if (this._element) {\n        this.hide();\n      }\n    };\n    EventHandler.on(this._element.closest(SELECTOR_MODAL), EVENT_MODAL_HIDE, this._hideModalHandler);\n  }\n  _fixTitle() {\n    const title = this._element.getAttribute('title');\n    if (!title) {\n      return;\n    }\n    if (!this._element.getAttribute('aria-label') && !this._element.textContent.trim()) {\n      this._element.setAttribute('aria-label', title);\n    }\n    this._element.setAttribute('data-bs-original-title', title); // DO NOT USE IT. Is only for backwards compatibility\n    this._element.removeAttribute('title');\n  }\n  _enter() {\n    if (this._isShown() || this._isHovered) {\n      this._isHovered = true;\n      return;\n    }\n    this._isHovered = true;\n    this._setTimeout(() => {\n      if (this._isHovered) {\n        this.show();\n      }\n    }, this._config.delay.show);\n  }\n  _leave() {\n    if (this._isWithActiveTrigger()) {\n      return;\n    }\n    this._isHovered = false;\n    this._setTimeout(() => {\n      if (!this._isHovered) {\n        this.hide();\n      }\n    }, this._config.delay.hide);\n  }\n  _setTimeout(handler, timeout) {\n    clearTimeout(this._timeout);\n    this._timeout = setTimeout(handler, timeout);\n  }\n  _isWithActiveTrigger() {\n    return Object.values(this._activeTrigger).includes(true);\n  }\n  _getConfig(config) {\n    const dataAttributes = Manipulator.getDataAttributes(this._element);\n    for (const dataAttribute of Object.keys(dataAttributes)) {\n      if (DISALLOWED_ATTRIBUTES.has(dataAttribute)) {\n        delete dataAttributes[dataAttribute];\n      }\n    }\n    config = {\n      ...dataAttributes,\n      ...(typeof config === 'object' && config ? config : {})\n    };\n    config = this._mergeConfigObj(config);\n    config = this._configAfterMerge(config);\n    this._typeCheckConfig(config);\n    return config;\n  }\n  _configAfterMerge(config) {\n    config.container = config.container === false ? document.body : getElement(config.container);\n    if (typeof config.delay === 'number') {\n      config.delay = {\n        show: config.delay,\n        hide: config.delay\n      };\n    }\n    if (typeof config.title === 'number') {\n      config.title = config.title.toString();\n    }\n    if (typeof config.content === 'number') {\n      config.content = config.content.toString();\n    }\n    return config;\n  }\n  _getDelegateConfig() {\n    const config = {};\n    for (const [key, value] of Object.entries(this._config)) {\n      if (this.constructor.Default[key] !== value) {\n        config[key] = value;\n      }\n    }\n    config.selector = false;\n    config.trigger = 'manual';\n\n    // In the future can be replaced with:\n    // const keysWithDifferentValues = Object.entries(this._config).filter(entry => this.constructor.Default[entry[0]] !== this._config[entry[0]])\n    // `Object.fromEntries(keysWithDifferentValues)`\n    return config;\n  }\n  _disposePopper() {\n    if (this._popper) {\n      this._popper.destroy();\n      this._popper = null;\n    }\n    if (this.tip) {\n      this.tip.remove();\n      this.tip = null;\n    }\n  }\n\n  // Static\n  static jQueryInterface(config) {\n    return this.each(function () {\n      const data = Tooltip.getOrCreateInstance(this, config);\n      if (typeof config !== 'string') {\n        return;\n      }\n      if (typeof data[config] === 'undefined') {\n        throw new TypeError(`No method named \"${config}\"`);\n      }\n      data[config]();\n    });\n  }\n}\n\n/**\n * jQuery\n */\n\ndefineJQueryPlugin(Tooltip);\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap popover.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$3 = 'popover';\nconst SELECTOR_TITLE = '.popover-header';\nconst SELECTOR_CONTENT = '.popover-body';\nconst Default$2 = {\n  ...Tooltip.Default,\n  content: '',\n  offset: [0, 8],\n  placement: 'right',\n  template: '<div class=\"popover\" role=\"tooltip\">' + '<div class=\"popover-arrow\"></div>' + '<h3 class=\"popover-header\"></h3>' + '<div class=\"popover-body\"></div>' + '</div>',\n  trigger: 'click'\n};\nconst DefaultType$2 = {\n  ...Tooltip.DefaultType,\n  content: '(null|string|element|function)'\n};\n\n/**\n * Class definition\n */\n\nclass Popover extends Tooltip {\n  // Getters\n  static get Default() {\n    return Default$2;\n  }\n  static get DefaultType() {\n    return DefaultType$2;\n  }\n  static get NAME() {\n    return NAME$3;\n  }\n\n  // Overrides\n  _isWithContent() {\n    return this._getTitle() || this._getContent();\n  }\n\n  // Private\n  _getContentForTemplate() {\n    return {\n      [SELECTOR_TITLE]: this._getTitle(),\n      [SELECTOR_CONTENT]: this._getContent()\n    };\n  }\n  _getContent() {\n    return this._resolvePossibleFunction(this._config.content);\n  }\n\n  // Static\n  static jQueryInterface(config) {\n    return this.each(function () {\n      const data = Popover.getOrCreateInstance(this, config);\n      if (typeof config !== 'string') {\n        return;\n      }\n      if (typeof data[config] === 'undefined') {\n        throw new TypeError(`No method named \"${config}\"`);\n      }\n      data[config]();\n    });\n  }\n}\n\n/**\n * jQuery\n */\n\ndefineJQueryPlugin(Popover);\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap scrollspy.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$2 = 'scrollspy';\nconst DATA_KEY$2 = 'bs.scrollspy';\nconst EVENT_KEY$2 = `.${DATA_KEY$2}`;\nconst DATA_API_KEY = '.data-api';\nconst EVENT_ACTIVATE = `activate${EVENT_KEY$2}`;\nconst EVENT_CLICK = `click${EVENT_KEY$2}`;\nconst EVENT_LOAD_DATA_API$1 = `load${EVENT_KEY$2}${DATA_API_KEY}`;\nconst CLASS_NAME_DROPDOWN_ITEM = 'dropdown-item';\nconst CLASS_NAME_ACTIVE$1 = 'active';\nconst SELECTOR_DATA_SPY = '[data-bs-spy=\"scroll\"]';\nconst SELECTOR_TARGET_LINKS = '[href]';\nconst SELECTOR_NAV_LIST_GROUP = '.nav, .list-group';\nconst SELECTOR_NAV_LINKS = '.nav-link';\nconst SELECTOR_NAV_ITEMS = '.nav-item';\nconst SELECTOR_LIST_ITEMS = '.list-group-item';\nconst SELECTOR_LINK_ITEMS = `${SELECTOR_NAV_LINKS}, ${SELECTOR_NAV_ITEMS} > ${SELECTOR_NAV_LINKS}, ${SELECTOR_LIST_ITEMS}`;\nconst SELECTOR_DROPDOWN = '.dropdown';\nconst SELECTOR_DROPDOWN_TOGGLE$1 = '.dropdown-toggle';\nconst Default$1 = {\n  offset: null,\n  // TODO: v6 @deprecated, keep it for backwards compatibility reasons\n  rootMargin: '0px 0px -25%',\n  smoothScroll: false,\n  target: null,\n  threshold: [0.1, 0.5, 1]\n};\nconst DefaultType$1 = {\n  offset: '(number|null)',\n  // TODO v6 @deprecated, keep it for backwards compatibility reasons\n  rootMargin: 'string',\n  smoothScroll: 'boolean',\n  target: 'element',\n  threshold: 'array'\n};\n\n/**\n * Class definition\n */\n\nclass ScrollSpy extends BaseComponent {\n  constructor(element, config) {\n    super(element, config);\n\n    // this._element is the observablesContainer and config.target the menu links wrapper\n    this._targetLinks = new Map();\n    this._observableSections = new Map();\n    this._rootElement = getComputedStyle(this._element).overflowY === 'visible' ? null : this._element;\n    this._activeTarget = null;\n    this._observer = null;\n    this._previousScrollData = {\n      visibleEntryTop: 0,\n      parentScrollTop: 0\n    };\n    this.refresh(); // initialize\n  }\n\n  // Getters\n  static get Default() {\n    return Default$1;\n  }\n  static get DefaultType() {\n    return DefaultType$1;\n  }\n  static get NAME() {\n    return NAME$2;\n  }\n\n  // Public\n  refresh() {\n    this._initializeTargetsAndObservables();\n    this._maybeEnableSmoothScroll();\n    if (this._observer) {\n      this._observer.disconnect();\n    } else {\n      this._observer = this._getNewObserver();\n    }\n    for (const section of this._observableSections.values()) {\n      this._observer.observe(section);\n    }\n  }\n  dispose() {\n    this._observer.disconnect();\n    super.dispose();\n  }\n\n  // Private\n  _configAfterMerge(config) {\n    // TODO: on v6 target should be given explicitly & remove the {target: 'ss-target'} case\n    config.target = getElement(config.target) || document.body;\n\n    // TODO: v6 Only for backwards compatibility reasons. Use rootMargin only\n    config.rootMargin = config.offset ? `${config.offset}px 0px -30%` : config.rootMargin;\n    if (typeof config.threshold === 'string') {\n      config.threshold = config.threshold.split(',').map(value => Number.parseFloat(value));\n    }\n    return config;\n  }\n  _maybeEnableSmoothScroll() {\n    if (!this._config.smoothScroll) {\n      return;\n    }\n\n    // unregister any previous listeners\n    EventHandler.off(this._config.target, EVENT_CLICK);\n    EventHandler.on(this._config.target, EVENT_CLICK, SELECTOR_TARGET_LINKS, event => {\n      const observableSection = this._observableSections.get(event.target.hash);\n      if (observableSection) {\n        event.preventDefault();\n        const root = this._rootElement || window;\n        const height = observableSection.offsetTop - this._element.offsetTop;\n        if (root.scrollTo) {\n          root.scrollTo({\n            top: height,\n            behavior: 'smooth'\n          });\n          return;\n        }\n\n        // Chrome 60 doesn't support `scrollTo`\n        root.scrollTop = height;\n      }\n    });\n  }\n  _getNewObserver() {\n    const options = {\n      root: this._rootElement,\n      threshold: this._config.threshold,\n      rootMargin: this._config.rootMargin\n    };\n    return new IntersectionObserver(entries => this._observerCallback(entries), options);\n  }\n\n  // The logic of selection\n  _observerCallback(entries) {\n    const targetElement = entry => this._targetLinks.get(`#${entry.target.id}`);\n    const activate = entry => {\n      this._previousScrollData.visibleEntryTop = entry.target.offsetTop;\n      this._process(targetElement(entry));\n    };\n    const parentScrollTop = (this._rootElement || document.documentElement).scrollTop;\n    const userScrollsDown = parentScrollTop >= this._previousScrollData.parentScrollTop;\n    this._previousScrollData.parentScrollTop = parentScrollTop;\n    for (const entry of entries) {\n      if (!entry.isIntersecting) {\n        this._activeTarget = null;\n        this._clearActiveClass(targetElement(entry));\n        continue;\n      }\n      const entryIsLowerThanPrevious = entry.target.offsetTop >= this._previousScrollData.visibleEntryTop;\n      // if we are scrolling down, pick the bigger offsetTop\n      if (userScrollsDown && entryIsLowerThanPrevious) {\n        activate(entry);\n        // if parent isn't scrolled, let's keep the first visible item, breaking the iteration\n        if (!parentScrollTop) {\n          return;\n        }\n        continue;\n      }\n\n      // if we are scrolling up, pick the smallest offsetTop\n      if (!userScrollsDown && !entryIsLowerThanPrevious) {\n        activate(entry);\n      }\n    }\n  }\n  _initializeTargetsAndObservables() {\n    this._targetLinks = new Map();\n    this._observableSections = new Map();\n    const targetLinks = SelectorEngine.find(SELECTOR_TARGET_LINKS, this._config.target);\n    for (const anchor of targetLinks) {\n      // ensure that the anchor has an id and is not disabled\n      if (!anchor.hash || isDisabled(anchor)) {\n        continue;\n      }\n      const observableSection = SelectorEngine.findOne(decodeURI(anchor.hash), this._element);\n\n      // ensure that the observableSection exists & is visible\n      if (isVisible(observableSection)) {\n        this._targetLinks.set(decodeURI(anchor.hash), anchor);\n        this._observableSections.set(anchor.hash, observableSection);\n      }\n    }\n  }\n  _process(target) {\n    if (this._activeTarget === target) {\n      return;\n    }\n    this._clearActiveClass(this._config.target);\n    this._activeTarget = target;\n    target.classList.add(CLASS_NAME_ACTIVE$1);\n    this._activateParents(target);\n    EventHandler.trigger(this._element, EVENT_ACTIVATE, {\n      relatedTarget: target\n    });\n  }\n  _activateParents(target) {\n    // Activate dropdown parents\n    if (target.classList.contains(CLASS_NAME_DROPDOWN_ITEM)) {\n      SelectorEngine.findOne(SELECTOR_DROPDOWN_TOGGLE$1, target.closest(SELECTOR_DROPDOWN)).classList.add(CLASS_NAME_ACTIVE$1);\n      return;\n    }\n    for (const listGroup of SelectorEngine.parents(target, SELECTOR_NAV_LIST_GROUP)) {\n      // Set triggered links parents as active\n      // With both <ul> and <nav> markup a parent is the previous sibling of any nav ancestor\n      for (const item of SelectorEngine.prev(listGroup, SELECTOR_LINK_ITEMS)) {\n        item.classList.add(CLASS_NAME_ACTIVE$1);\n      }\n    }\n  }\n  _clearActiveClass(parent) {\n    parent.classList.remove(CLASS_NAME_ACTIVE$1);\n    const activeNodes = SelectorEngine.find(`${SELECTOR_TARGET_LINKS}.${CLASS_NAME_ACTIVE$1}`, parent);\n    for (const node of activeNodes) {\n      node.classList.remove(CLASS_NAME_ACTIVE$1);\n    }\n  }\n\n  // Static\n  static jQueryInterface(config) {\n    return this.each(function () {\n      const data = ScrollSpy.getOrCreateInstance(this, config);\n      if (typeof config !== 'string') {\n        return;\n      }\n      if (data[config] === undefined || config.startsWith('_') || config === 'constructor') {\n        throw new TypeError(`No method named \"${config}\"`);\n      }\n      data[config]();\n    });\n  }\n}\n\n/**\n * Data API implementation\n */\n\nEventHandler.on(window, EVENT_LOAD_DATA_API$1, () => {\n  for (const spy of SelectorEngine.find(SELECTOR_DATA_SPY)) {\n    ScrollSpy.getOrCreateInstance(spy);\n  }\n});\n\n/**\n * jQuery\n */\n\ndefineJQueryPlugin(ScrollSpy);\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap tab.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME$1 = 'tab';\nconst DATA_KEY$1 = 'bs.tab';\nconst EVENT_KEY$1 = `.${DATA_KEY$1}`;\nconst EVENT_HIDE$1 = `hide${EVENT_KEY$1}`;\nconst EVENT_HIDDEN$1 = `hidden${EVENT_KEY$1}`;\nconst EVENT_SHOW$1 = `show${EVENT_KEY$1}`;\nconst EVENT_SHOWN$1 = `shown${EVENT_KEY$1}`;\nconst EVENT_CLICK_DATA_API = `click${EVENT_KEY$1}`;\nconst EVENT_KEYDOWN = `keydown${EVENT_KEY$1}`;\nconst EVENT_LOAD_DATA_API = `load${EVENT_KEY$1}`;\nconst ARROW_LEFT_KEY = 'ArrowLeft';\nconst ARROW_RIGHT_KEY = 'ArrowRight';\nconst ARROW_UP_KEY = 'ArrowUp';\nconst ARROW_DOWN_KEY = 'ArrowDown';\nconst HOME_KEY = 'Home';\nconst END_KEY = 'End';\nconst CLASS_NAME_ACTIVE = 'active';\nconst CLASS_NAME_FADE$1 = 'fade';\nconst CLASS_NAME_SHOW$1 = 'show';\nconst CLASS_DROPDOWN = 'dropdown';\nconst SELECTOR_DROPDOWN_TOGGLE = '.dropdown-toggle';\nconst SELECTOR_DROPDOWN_MENU = '.dropdown-menu';\nconst NOT_SELECTOR_DROPDOWN_TOGGLE = `:not(${SELECTOR_DROPDOWN_TOGGLE})`;\nconst SELECTOR_TAB_PANEL = '.list-group, .nav, [role=\"tablist\"]';\nconst SELECTOR_OUTER = '.nav-item, .list-group-item';\nconst SELECTOR_INNER = `.nav-link${NOT_SELECTOR_DROPDOWN_TOGGLE}, .list-group-item${NOT_SELECTOR_DROPDOWN_TOGGLE}, [role=\"tab\"]${NOT_SELECTOR_DROPDOWN_TOGGLE}`;\nconst SELECTOR_DATA_TOGGLE = '[data-bs-toggle=\"tab\"], [data-bs-toggle=\"pill\"], [data-bs-toggle=\"list\"]'; // TODO: could only be `tab` in v6\nconst SELECTOR_INNER_ELEM = `${SELECTOR_INNER}, ${SELECTOR_DATA_TOGGLE}`;\nconst SELECTOR_DATA_TOGGLE_ACTIVE = `.${CLASS_NAME_ACTIVE}[data-bs-toggle=\"tab\"], .${CLASS_NAME_ACTIVE}[data-bs-toggle=\"pill\"], .${CLASS_NAME_ACTIVE}[data-bs-toggle=\"list\"]`;\n\n/**\n * Class definition\n */\n\nclass Tab extends BaseComponent {\n  constructor(element) {\n    super(element);\n    this._parent = this._element.closest(SELECTOR_TAB_PANEL);\n    if (!this._parent) {\n      return;\n      // TODO: should throw exception in v6\n      // throw new TypeError(`${element.outerHTML} has not a valid parent ${SELECTOR_INNER_ELEM}`)\n    }\n\n    // Set up initial aria attributes\n    this._setInitialAttributes(this._parent, this._getChildren());\n    EventHandler.on(this._element, EVENT_KEYDOWN, event => this._keydown(event));\n  }\n\n  // Getters\n  static get NAME() {\n    return NAME$1;\n  }\n\n  // Public\n  show() {\n    // Shows this elem and deactivate the active sibling if exists\n    const innerElem = this._element;\n    if (this._elemIsActive(innerElem)) {\n      return;\n    }\n\n    // Search for active tab on same parent to deactivate it\n    const active = this._getActiveElem();\n    const hideEvent = active ? EventHandler.trigger(active, EVENT_HIDE$1, {\n      relatedTarget: innerElem\n    }) : null;\n    const showEvent = EventHandler.trigger(innerElem, EVENT_SHOW$1, {\n      relatedTarget: active\n    });\n    if (showEvent.defaultPrevented || hideEvent && hideEvent.defaultPrevented) {\n      return;\n    }\n    this._deactivate(active, innerElem);\n    this._activate(innerElem, active);\n  }\n\n  // Private\n  _activate(element, relatedElem) {\n    if (!element) {\n      return;\n    }\n    element.classList.add(CLASS_NAME_ACTIVE);\n    this._activate(SelectorEngine.getElementFromSelector(element)); // Search and activate/show the proper section\n\n    const complete = () => {\n      if (element.getAttribute('role') !== 'tab') {\n        element.classList.add(CLASS_NAME_SHOW$1);\n        return;\n      }\n      element.removeAttribute('tabindex');\n      element.setAttribute('aria-selected', true);\n      this._toggleDropDown(element, true);\n      EventHandler.trigger(element, EVENT_SHOWN$1, {\n        relatedTarget: relatedElem\n      });\n    };\n    this._queueCallback(complete, element, element.classList.contains(CLASS_NAME_FADE$1));\n  }\n  _deactivate(element, relatedElem) {\n    if (!element) {\n      return;\n    }\n    element.classList.remove(CLASS_NAME_ACTIVE);\n    element.blur();\n    this._deactivate(SelectorEngine.getElementFromSelector(element)); // Search and deactivate the shown section too\n\n    const complete = () => {\n      if (element.getAttribute('role') !== 'tab') {\n        element.classList.remove(CLASS_NAME_SHOW$1);\n        return;\n      }\n      element.setAttribute('aria-selected', false);\n      element.setAttribute('tabindex', '-1');\n      this._toggleDropDown(element, false);\n      EventHandler.trigger(element, EVENT_HIDDEN$1, {\n        relatedTarget: relatedElem\n      });\n    };\n    this._queueCallback(complete, element, element.classList.contains(CLASS_NAME_FADE$1));\n  }\n  _keydown(event) {\n    if (![ARROW_LEFT_KEY, ARROW_RIGHT_KEY, ARROW_UP_KEY, ARROW_DOWN_KEY, HOME_KEY, END_KEY].includes(event.key)) {\n      return;\n    }\n    event.stopPropagation(); // stopPropagation/preventDefault both added to support up/down keys without scrolling the page\n    event.preventDefault();\n    const children = this._getChildren().filter(element => !isDisabled(element));\n    let nextActiveElement;\n    if ([HOME_KEY, END_KEY].includes(event.key)) {\n      nextActiveElement = children[event.key === HOME_KEY ? 0 : children.length - 1];\n    } else {\n      const isNext = [ARROW_RIGHT_KEY, ARROW_DOWN_KEY].includes(event.key);\n      nextActiveElement = getNextActiveElement(children, event.target, isNext, true);\n    }\n    if (nextActiveElement) {\n      nextActiveElement.focus({\n        preventScroll: true\n      });\n      Tab.getOrCreateInstance(nextActiveElement).show();\n    }\n  }\n  _getChildren() {\n    // collection of inner elements\n    return SelectorEngine.find(SELECTOR_INNER_ELEM, this._parent);\n  }\n  _getActiveElem() {\n    return this._getChildren().find(child => this._elemIsActive(child)) || null;\n  }\n  _setInitialAttributes(parent, children) {\n    this._setAttributeIfNotExists(parent, 'role', 'tablist');\n    for (const child of children) {\n      this._setInitialAttributesOnChild(child);\n    }\n  }\n  _setInitialAttributesOnChild(child) {\n    child = this._getInnerElement(child);\n    const isActive = this._elemIsActive(child);\n    const outerElem = this._getOuterElement(child);\n    child.setAttribute('aria-selected', isActive);\n    if (outerElem !== child) {\n      this._setAttributeIfNotExists(outerElem, 'role', 'presentation');\n    }\n    if (!isActive) {\n      child.setAttribute('tabindex', '-1');\n    }\n    this._setAttributeIfNotExists(child, 'role', 'tab');\n\n    // set attributes to the related panel too\n    this._setInitialAttributesOnTargetPanel(child);\n  }\n  _setInitialAttributesOnTargetPanel(child) {\n    const target = SelectorEngine.getElementFromSelector(child);\n    if (!target) {\n      return;\n    }\n    this._setAttributeIfNotExists(target, 'role', 'tabpanel');\n    if (child.id) {\n      this._setAttributeIfNotExists(target, 'aria-labelledby', `${child.id}`);\n    }\n  }\n  _toggleDropDown(element, open) {\n    const outerElem = this._getOuterElement(element);\n    if (!outerElem.classList.contains(CLASS_DROPDOWN)) {\n      return;\n    }\n    const toggle = (selector, className) => {\n      const element = SelectorEngine.findOne(selector, outerElem);\n      if (element) {\n        element.classList.toggle(className, open);\n      }\n    };\n    toggle(SELECTOR_DROPDOWN_TOGGLE, CLASS_NAME_ACTIVE);\n    toggle(SELECTOR_DROPDOWN_MENU, CLASS_NAME_SHOW$1);\n    outerElem.setAttribute('aria-expanded', open);\n  }\n  _setAttributeIfNotExists(element, attribute, value) {\n    if (!element.hasAttribute(attribute)) {\n      element.setAttribute(attribute, value);\n    }\n  }\n  _elemIsActive(elem) {\n    return elem.classList.contains(CLASS_NAME_ACTIVE);\n  }\n\n  // Try to get the inner element (usually the .nav-link)\n  _getInnerElement(elem) {\n    return elem.matches(SELECTOR_INNER_ELEM) ? elem : SelectorEngine.findOne(SELECTOR_INNER_ELEM, elem);\n  }\n\n  // Try to get the outer element (usually the .nav-item)\n  _getOuterElement(elem) {\n    return elem.closest(SELECTOR_OUTER) || elem;\n  }\n\n  // Static\n  static jQueryInterface(config) {\n    return this.each(function () {\n      const data = Tab.getOrCreateInstance(this);\n      if (typeof config !== 'string') {\n        return;\n      }\n      if (data[config] === undefined || config.startsWith('_') || config === 'constructor') {\n        throw new TypeError(`No method named \"${config}\"`);\n      }\n      data[config]();\n    });\n  }\n}\n\n/**\n * Data API implementation\n */\n\nEventHandler.on(document, EVENT_CLICK_DATA_API, SELECTOR_DATA_TOGGLE, function (event) {\n  if (['A', 'AREA'].includes(this.tagName)) {\n    event.preventDefault();\n  }\n  if (isDisabled(this)) {\n    return;\n  }\n  Tab.getOrCreateInstance(this).show();\n});\n\n/**\n * Initialize on focus\n */\nEventHandler.on(window, EVENT_LOAD_DATA_API, () => {\n  for (const element of SelectorEngine.find(SELECTOR_DATA_TOGGLE_ACTIVE)) {\n    Tab.getOrCreateInstance(element);\n  }\n});\n/**\n * jQuery\n */\n\ndefineJQueryPlugin(Tab);\n\n/**\n * --------------------------------------------------------------------------\n * Bootstrap toast.js\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n * --------------------------------------------------------------------------\n */\n\n\n/**\n * Constants\n */\n\nconst NAME = 'toast';\nconst DATA_KEY = 'bs.toast';\nconst EVENT_KEY = `.${DATA_KEY}`;\nconst EVENT_MOUSEOVER = `mouseover${EVENT_KEY}`;\nconst EVENT_MOUSEOUT = `mouseout${EVENT_KEY}`;\nconst EVENT_FOCUSIN = `focusin${EVENT_KEY}`;\nconst EVENT_FOCUSOUT = `focusout${EVENT_KEY}`;\nconst EVENT_HIDE = `hide${EVENT_KEY}`;\nconst EVENT_HIDDEN = `hidden${EVENT_KEY}`;\nconst EVENT_SHOW = `show${EVENT_KEY}`;\nconst EVENT_SHOWN = `shown${EVENT_KEY}`;\nconst CLASS_NAME_FADE = 'fade';\nconst CLASS_NAME_HIDE = 'hide'; // @deprecated - kept here only for backwards compatibility\nconst CLASS_NAME_SHOW = 'show';\nconst CLASS_NAME_SHOWING = 'showing';\nconst DefaultType = {\n  animation: 'boolean',\n  autohide: 'boolean',\n  delay: 'number'\n};\nconst Default = {\n  animation: true,\n  autohide: true,\n  delay: 5000\n};\n\n/**\n * Class definition\n */\n\nclass Toast extends BaseComponent {\n  constructor(element, config) {\n    super(element, config);\n    this._timeout = null;\n    this._hasMouseInteraction = false;\n    this._hasKeyboardInteraction = false;\n    this._setListeners();\n  }\n\n  // Getters\n  static get Default() {\n    return Default;\n  }\n  static get DefaultType() {\n    return DefaultType;\n  }\n  static get NAME() {\n    return NAME;\n  }\n\n  // Public\n  show() {\n    const showEvent = EventHandler.trigger(this._element, EVENT_SHOW);\n    if (showEvent.defaultPrevented) {\n      return;\n    }\n    this._clearTimeout();\n    if (this._config.animation) {\n      this._element.classList.add(CLASS_NAME_FADE);\n    }\n    const complete = () => {\n      this._element.classList.remove(CLASS_NAME_SHOWING);\n      EventHandler.trigger(this._element, EVENT_SHOWN);\n      this._maybeScheduleHide();\n    };\n    this._element.classList.remove(CLASS_NAME_HIDE); // @deprecated\n    reflow(this._element);\n    this._element.classList.add(CLASS_NAME_SHOW, CLASS_NAME_SHOWING);\n    this._queueCallback(complete, this._element, this._config.animation);\n  }\n  hide() {\n    if (!this.isShown()) {\n      return;\n    }\n    const hideEvent = EventHandler.trigger(this._element, EVENT_HIDE);\n    if (hideEvent.defaultPrevented) {\n      return;\n    }\n    const complete = () => {\n      this._element.classList.add(CLASS_NAME_HIDE); // @deprecated\n      this._element.classList.remove(CLASS_NAME_SHOWING, CLASS_NAME_SHOW);\n      EventHandler.trigger(this._element, EVENT_HIDDEN);\n    };\n    this._element.classList.add(CLASS_NAME_SHOWING);\n    this._queueCallback(complete, this._element, this._config.animation);\n  }\n  dispose() {\n    this._clearTimeout();\n    if (this.isShown()) {\n      this._element.classList.remove(CLASS_NAME_SHOW);\n    }\n    super.dispose();\n  }\n  isShown() {\n    return this._element.classList.contains(CLASS_NAME_SHOW);\n  }\n\n  // Private\n\n  _maybeScheduleHide() {\n    if (!this._config.autohide) {\n      return;\n    }\n    if (this._hasMouseInteraction || this._hasKeyboardInteraction) {\n      return;\n    }\n    this._timeout = setTimeout(() => {\n      this.hide();\n    }, this._config.delay);\n  }\n  _onInteraction(event, isInteracting) {\n    switch (event.type) {\n      case 'mouseover':\n      case 'mouseout':\n        {\n          this._hasMouseInteraction = isInteracting;\n          break;\n        }\n      case 'focusin':\n      case 'focusout':\n        {\n          this._hasKeyboardInteraction = isInteracting;\n          break;\n        }\n    }\n    if (isInteracting) {\n      this._clearTimeout();\n      return;\n    }\n    const nextElement = event.relatedTarget;\n    if (this._element === nextElement || this._element.contains(nextElement)) {\n      return;\n    }\n    this._maybeScheduleHide();\n  }\n  _setListeners() {\n    EventHandler.on(this._element, EVENT_MOUSEOVER, event => this._onInteraction(event, true));\n    EventHandler.on(this._element, EVENT_MOUSEOUT, event => this._onInteraction(event, false));\n    EventHandler.on(this._element, EVENT_FOCUSIN, event => this._onInteraction(event, true));\n    EventHandler.on(this._element, EVENT_FOCUSOUT, event => this._onInteraction(event, false));\n  }\n  _clearTimeout() {\n    clearTimeout(this._timeout);\n    this._timeout = null;\n  }\n\n  // Static\n  static jQueryInterface(config) {\n    return this.each(function () {\n      const data = Toast.getOrCreateInstance(this, config);\n      if (typeof config === 'string') {\n        if (typeof data[config] === 'undefined') {\n          throw new TypeError(`No method named \"${config}\"`);\n        }\n        data[config](this);\n      }\n    });\n  }\n}\n\n/**\n * Data API implementation\n */\n\nenableDismissTrigger(Toast);\n\n/**\n * jQuery\n */\n\ndefineJQueryPlugin(Toast);\n\nexport { Alert, Button, Carousel, Collapse, Dropdown, Modal, Offcanvas, Popover, ScrollSpy, Tab, Toast, Tooltip };\n//# sourceMappingURL=bootstrap.esm.js.map\n","/* define several functions to replace jQuery methods\n * inspired by https://tobiasahlin.com/blog/move-from-jquery-to-vanilla-javascript/\n */\n\n/**\n * Execute a method if DOM has finished loading\n *\n * @param {function} callback the method to execute\n */\nexport function documentReady(callback) {\n  if (document.readyState != \"loading\") callback();\n  else document.addEventListener(\"DOMContentLoaded\", callback);\n}\n","// TODO: @trallard might end up moving this to the main JS file\n// Import and setup functions to control Bootstrap's behavior.\nimport \"@popperjs/core\";\n// Import all of Bootstrap's JS\nimport * as bootstrap from \"bootstrap\";\nimport { documentReady } from \"./mixin\";\n\n/*******************************************************************************\n * Trigger tooltips\n */\n\n/**\n * Add tooltip to each element with the \"tooltip\" data-bs-toogle class\n */\nfunction TriggerTooltip() {\n  var tooltipTriggerList = [].slice.call(\n    document.querySelectorAll('[data-bs-toggle=\"tooltip\"]'),\n  );\n  tooltipTriggerList.map(function (tooltipTriggerEl) {\n    return new bootstrap.Tooltip(tooltipTriggerEl, {\n      delay: { show: 500, hide: 100 },\n    });\n  });\n}\n\n/*******************************************************************************\n * back to top button\n */\nfunction backToTop() {\n  var btn = document.getElementById(\"pst-back-to-top\");\n  btn.addEventListener(\"click\", function () {\n    document.body.scrollTop = 0;\n    document.documentElement.scrollTop = 0;\n  });\n}\n\nfunction showBackToTop() {\n  var btn = document.getElementById(\"pst-back-to-top\");\n  var header = document\n    .getElementsByClassName(\"bd-header\")[0]\n    .getBoundingClientRect();\n  window.addEventListener(\"scroll\", function () {\n    if (this.oldScroll > this.scrollY && this.scrollY > header.bottom) {\n      btn.style.display = \"block\";\n    } else {\n      btn.style.display = \"none\";\n    }\n    this.oldScroll = this.scrollY;\n  });\n}\n\n/*******************************************************************************\n * Call functions after document loading.\n */\n\ndocumentReady(TriggerTooltip);\ndocumentReady(backToTop);\ndocumentReady(showBackToTop);\n\nwindow.bootstrap = bootstrap;\n"],"names":["__webpack_require__","exports","definition","key","o","Object","defineProperty","enumerable","get","obj","prop","prototype","hasOwnProperty","call","Symbol","toStringTag","value","bottom","right","left","auto","basePlacements","start","end","clippingParents","viewport","popper","reference","variationPlacements","reduce","acc","placement","concat","beforeRead","read","afterRead","beforeMain","main","afterMain","beforeWrite","write","afterWrite","modifierPhases","getNodeName","element","nodeName","toLowerCase","getWindow","node","window","toString","ownerDocument","defaultView","isElement","Element","isHTMLElement","HTMLElement","isShadowRoot","ShadowRoot","name","enabled","phase","fn","_ref","state","keys","elements","forEach","style","styles","attributes","assign","removeAttribute","setAttribute","effect","_ref2","initialStyles","position","options","strategy","top","margin","arrow","property","attribute","requires","getBasePlacement","split","Math","max","min","round","getUAString","uaData","navigator","userAgentData","brands","Array","isArray","map","item","brand","version","join","userAgent","isLayoutViewport","test","getBoundingClientRect","includeScale","isFixedStrategy","clientRect","scaleX","scaleY","offsetWidth","width","offsetHeight","height","visualViewport","addVisualOffsets","x","offsetLeft","y","offsetTop","getLayoutRect","abs","contains","parent","child","rootNode","getRootNode","next","isSameNode","parentNode","host","getComputedStyle","isTableElement","indexOf","getDocumentElement","document","documentElement","getParentNode","assignedSlot","getTrueOffsetParent","offsetParent","getOffsetParent","isFirefox","currentNode","css","transform","perspective","contain","willChange","filter","getContainingBlock","getMainAxisFromPlacement","within","mergePaddingObject","paddingObject","expandToHashMap","hashMap","_state$modifiersData$","arrowElement","popperOffsets","modifiersData","basePlacement","axis","len","padding","rects","toPaddingObject","arrowRect","minProp","maxProp","endDiff","startDiff","arrowOffsetParent","clientSize","clientHeight","clientWidth","centerToReference","center","offset","axisProp","centerOffset","_options$element","querySelector","requiresIfExists","getVariation","unsetSides","mapToStyles","_Object$assign2","popperRect","variation","offsets","gpuAcceleration","adaptive","roundOffsets","isFixed","_offsets$x","_offsets$y","_ref3","hasX","hasY","sideX","sideY","win","heightProp","widthProp","_Object$assign","commonStyles","_ref4","dpr","devicePixelRatio","roundOffsetsByDPR","_ref5","_options$gpuAccelerat","_options$adaptive","_options$roundOffsets","data","passive","instance","_options$scroll","scroll","_options$resize","resize","scrollParents","scrollParent","addEventListener","update","removeEventListener","hash","getOppositePlacement","replace","matched","getOppositeVariationPlacement","getWindowScroll","scrollLeft","pageXOffset","scrollTop","pageYOffset","getWindowScrollBarX","isScrollParent","_getComputedStyle","overflow","overflowX","overflowY","getScrollParent","body","listScrollParents","list","_element$ownerDocumen","isBody","target","updatedList","rectToClientRect","rect","getClientRectFromMixedType","clippingParent","html","layoutViewport","getViewportRect","clientTop","clientLeft","getInnerBoundingClientRect","winScroll","scrollWidth","scrollHeight","direction","getDocumentRect","computeOffsets","commonX","commonY","mainAxis","detectOverflow","_options","_options$placement","_options$strategy","_options$boundary","boundary","_options$rootBoundary","rootBoundary","_options$elementConte","elementContext","_options$altBoundary","altBoundary","_options$padding","altContext","clippingClientRect","mainClippingParents","clipperElement","getClippingParents","firstClippingParent","clippingRect","accRect","getClippingRect","contextElement","referenceClientRect","popperClientRect","elementClientRect","overflowOffsets","offsetData","multiply","_skip","_options$mainAxis","checkMainAxis","_options$altAxis","altAxis","checkAltAxis","specifiedFallbackPlacements","fallbackPlacements","_options$flipVariatio","flipVariations","allowedAutoPlacements","preferredPlacement","oppositePlacement","getExpandedFallbackPlacements","placements","_options$allowedAutoP","allowedPlacements","length","overflows","sort","a","b","computeAutoPlacement","referenceRect","checksMap","Map","makeFallbackChecks","firstFittingPlacement","i","_basePlacement","isStartVariation","isVertical","mainVariationSide","altVariationSide","checks","push","every","check","set","_loop","_i","fittingPlacement","find","slice","reset","getSideOffsets","preventedOffsets","isAnySideFullyClipped","some","side","preventOverflow","referenceOverflow","popperAltOverflow","referenceClippingOffsets","popperEscapeOffsets","isReferenceHidden","hasPopperEscaped","_options$offset","invertDistance","skidding","distance","distanceAndSkiddingToXY","_data$state$placement","_options$tether","tether","_options$tetherOffset","tetherOffset","isBasePlacement","tetherOffsetValue","normalizedTetherOffsetValue","offsetModifierState","_offsetModifierState$","mainSide","altSide","additive","minLen","maxLen","arrowPaddingObject","arrowPaddingMin","arrowPaddingMax","arrowLen","minOffset","maxOffset","clientOffset","offsetModifierValue","tetherMax","preventedOffset","_offsetModifierState$2","_mainSide","_altSide","_offset","_len","_min","_max","isOriginSide","_offsetModifierValue","_tetherMin","_tetherMax","_preventedOffset","v","withinMaxClamp","getCompositeRect","elementOrVirtualElement","isOffsetParentAnElement","offsetParentIsScaled","isElementScaled","order","modifiers","visited","Set","result","modifier","add","dep","has","depModifier","DEFAULT_OPTIONS","areValidElements","arguments","args","_key","popperGenerator","generatorOptions","_generatorOptions","_generatorOptions$def","defaultModifiers","_generatorOptions$def2","defaultOptions","pending","orderedModifiers","effectCleanupFns","isDestroyed","setOptions","setOptionsAction","cleanupModifierEffects","merged","orderModifiers","current","existing","m","_ref$options","cleanupFn","forceUpdate","_state$elements","index","_state$orderedModifie","_state$orderedModifie2","Promise","resolve","then","undefined","destroy","onFirstUpdate","createPopper","eventListeners","elementMap","Data","instanceMap","size","console","error","from","remove","delete","TRANSITION_END","parseSelector","selector","CSS","escape","match","id","triggerTransitionEnd","dispatchEvent","Event","object","jquery","nodeType","getElement","isVisible","getClientRects","elementIsVisible","getPropertyValue","closedDetails","closest","summary","isDisabled","Node","ELEMENT_NODE","classList","disabled","hasAttribute","getAttribute","findShadowRoot","attachShadow","root","noop","reflow","getjQuery","jQuery","DOMContentLoadedCallbacks","isRTL","dir","defineJQueryPlugin","plugin","callback","$","NAME","JQUERY_NO_CONFLICT","jQueryInterface","Constructor","noConflict","readyState","execute","possibleCallback","defaultValue","executeAfterTransition","transitionElement","waitForTransition","emulatedDuration","transitionDuration","transitionDelay","floatTransitionDuration","Number","parseFloat","floatTransitionDelay","getTransitionDurationFromElement","called","handler","setTimeout","getNextActiveElement","activeElement","shouldGetNext","isCycleAllowed","listLength","namespaceRegex","stripNameRegex","stripUidRegex","eventRegistry","uidEvent","customEvents","mouseenter","mouseleave","nativeEvents","makeEventUid","uid","getElementEvents","findHandler","events","callable","delegationSelector","values","event","normalizeParameters","originalTypeEvent","delegationFunction","isDelegated","typeEvent","getTypeEvent","addHandler","oneOff","wrapFunction","relatedTarget","delegateTarget","this","handlers","previousFunction","domElements","querySelectorAll","domElement","hydrateObj","EventHandler","off","type","apply","bootstrapDelegationHandler","bootstrapHandler","removeHandler","Boolean","removeNamespacedHandlers","namespace","storeElementEvent","handlerKey","entries","includes","on","one","inNamespace","isNamespace","startsWith","elementEvent","keyHandlers","trigger","jQueryEvent","bubbles","nativeDispatch","defaultPrevented","isPropagationStopped","isImmediatePropagationStopped","isDefaultPrevented","evt","cancelable","preventDefault","meta","_unused","configurable","normalizeData","JSON","parse","decodeURIComponent","normalizeDataKey","chr","Manipulator","setDataAttribute","removeDataAttribute","getDataAttributes","bsKeys","dataset","pureKey","charAt","getDataAttribute","Config","Default","DefaultType","Error","_getConfig","config","_mergeConfigObj","_configAfterMerge","_typeCheckConfig","jsonConfig","constructor","configTypes","expectedTypes","valueType","RegExp","TypeError","toUpperCase","BaseComponent","super","_element","_config","DATA_KEY","dispose","EVENT_KEY","propertyName","getOwnPropertyNames","_queueCallback","isAnimated","getInstance","getOrCreateInstance","VERSION","eventName","getSelector","hrefAttribute","trim","sel","SelectorEngine","findOne","children","matches","parents","ancestor","prev","previous","previousElementSibling","nextElementSibling","focusableChildren","focusables","el","getSelectorFromElement","getElementFromSelector","getMultipleElementsFromSelector","enableDismissTrigger","component","method","clickEvent","tagName","EVENT_KEY$b","EVENT_CLOSE","EVENT_CLOSED","Alert","close","_destroyElement","each","SELECTOR_DATA_TOGGLE$5","Button","toggle","button","EVENT_KEY$9","EVENT_TOUCHSTART","EVENT_TOUCHMOVE","EVENT_TOUCHEND","EVENT_POINTERDOWN","EVENT_POINTERUP","Default$c","endCallback","leftCallback","rightCallback","DefaultType$c","Swipe","isSupported","_deltaX","_supportPointerEvents","PointerEvent","_initEvents","_start","_eventIsPointerPenTouch","clientX","touches","_end","_handleSwipe","_move","absDeltaX","pointerType","maxTouchPoints","EVENT_KEY$8","DATA_API_KEY$5","ARROW_LEFT_KEY$1","ARROW_RIGHT_KEY$1","ORDER_NEXT","ORDER_PREV","DIRECTION_LEFT","DIRECTION_RIGHT","EVENT_SLIDE","EVENT_SLID","EVENT_KEYDOWN$1","EVENT_MOUSEENTER$1","EVENT_MOUSELEAVE$1","EVENT_DRAG_START","EVENT_LOAD_DATA_API$3","EVENT_CLICK_DATA_API$5","CLASS_NAME_CAROUSEL","CLASS_NAME_ACTIVE$2","SELECTOR_ACTIVE","SELECTOR_ITEM","SELECTOR_ACTIVE_ITEM","KEY_TO_DIRECTION","Default$b","interval","keyboard","pause","ride","touch","wrap","DefaultType$b","Carousel","_interval","_activeElement","_isSliding","touchTimeout","_swipeHelper","_indicatorsElement","_addEventListeners","cycle","_slide","nextWhenVisible","hidden","_clearInterval","_updateInterval","setInterval","_maybeEnableCycle","to","items","_getItems","activeIndex","_getItemIndex","_getActive","defaultInterval","_keydown","_addTouchEventListeners","img","swipeConfig","_directionToOrder","clearTimeout","_setActiveIndicatorElement","activeIndicator","newActiveIndicator","elementInterval","parseInt","isNext","nextElement","nextElementIndex","triggerEvent","_orderToDirection","isCycling","directionalClassName","orderClassName","_isAnimated","clearInterval","carousel","slideIndex","carousels","EVENT_KEY$7","EVENT_SHOW$6","EVENT_SHOWN$6","EVENT_HIDE$6","EVENT_HIDDEN$6","EVENT_CLICK_DATA_API$4","CLASS_NAME_SHOW$7","CLASS_NAME_COLLAPSE","CLASS_NAME_COLLAPSING","CLASS_NAME_DEEPER_CHILDREN","SELECTOR_DATA_TOGGLE$4","Default$a","DefaultType$a","Collapse","_isTransitioning","_triggerArray","toggleList","elem","filterElement","foundElement","_initializeChildren","_addAriaAndCollapsedClass","_isShown","hide","show","activeChildren","_getFirstLevelChildren","activeInstance","dimension","_getDimension","scrollSize","selected","triggerArray","isOpen","NAME$a","EVENT_KEY$6","DATA_API_KEY$3","ARROW_UP_KEY$1","ARROW_DOWN_KEY$1","EVENT_HIDE$5","EVENT_HIDDEN$5","EVENT_SHOW$5","EVENT_SHOWN$5","EVENT_CLICK_DATA_API$3","EVENT_KEYDOWN_DATA_API","EVENT_KEYUP_DATA_API","CLASS_NAME_SHOW$6","SELECTOR_DATA_TOGGLE$3","SELECTOR_DATA_TOGGLE_SHOWN","SELECTOR_MENU","PLACEMENT_TOP","PLACEMENT_TOPEND","PLACEMENT_BOTTOM","PLACEMENT_BOTTOMEND","PLACEMENT_RIGHT","PLACEMENT_LEFT","Default$9","autoClose","display","popperConfig","DefaultType$9","Dropdown","_popper","_parent","_menu","_inNavbar","_detectNavbar","_createPopper","focus","_completeHide","referenceElement","_getPopperConfig","_getPlacement","parentDropdown","isEnd","_getOffset","popperData","defaultBsPopperConfig","_selectMenuItem","clearMenus","openToggles","context","composedPath","isMenuTarget","dataApiKeydownHandler","isInput","isEscapeEvent","isUpOrDownEvent","getToggleButton","stopPropagation","NAME$9","CLASS_NAME_SHOW$5","EVENT_MOUSEDOWN","Default$8","className","clickCallback","rootElement","DefaultType$8","Backdrop","_isAppended","_append","_getElement","_emulateAnimation","backdrop","createElement","append","EVENT_KEY$5","EVENT_FOCUSIN$2","EVENT_KEYDOWN_TAB","TAB_NAV_BACKWARD","Default$7","autofocus","trapElement","DefaultType$7","FocusTrap","_isActive","_lastTabNavDirection","activate","_handleFocusin","_handleKeydown","deactivate","shiftKey","SELECTOR_FIXED_CONTENT","SELECTOR_STICKY_CONTENT","PROPERTY_PADDING","PROPERTY_MARGIN","ScrollBarHelper","getWidth","documentWidth","innerWidth","_disableOverFlow","_setElementAttributes","calculatedValue","_resetElementAttributes","isOverflowing","_saveInitialAttribute","styleProperty","scrollbarWidth","_applyManipulationCallback","setProperty","actualValue","removeProperty","callBack","EVENT_KEY$4","EVENT_HIDE$4","EVENT_HIDE_PREVENTED$1","EVENT_HIDDEN$4","EVENT_SHOW$4","EVENT_SHOWN$4","EVENT_RESIZE$1","EVENT_CLICK_DISMISS","EVENT_MOUSEDOWN_DISMISS","EVENT_KEYDOWN_DISMISS$1","EVENT_CLICK_DATA_API$2","CLASS_NAME_OPEN","CLASS_NAME_SHOW$4","CLASS_NAME_STATIC","Default$6","DefaultType$6","Modal","_dialog","_backdrop","_initializeBackDrop","_focustrap","_initializeFocusTrap","_scrollBar","_adjustDialog","_showElement","_hideModal","handleUpdate","modalBody","_triggerBackdropTransition","event2","_resetAdjustments","isModalOverflowing","initialOverflowY","isBodyOverflowing","paddingLeft","paddingRight","showEvent","alreadyOpen","EVENT_KEY$3","DATA_API_KEY$1","EVENT_LOAD_DATA_API$2","CLASS_NAME_SHOW$3","CLASS_NAME_SHOWING$1","CLASS_NAME_HIDING","OPEN_SELECTOR","EVENT_SHOW$3","EVENT_SHOWN$3","EVENT_HIDE$3","EVENT_HIDE_PREVENTED","EVENT_HIDDEN$3","EVENT_RESIZE","EVENT_CLICK_DATA_API$1","EVENT_KEYDOWN_DISMISS","Default$5","DefaultType$5","Offcanvas","blur","DefaultAllowlist","area","br","col","code","dd","div","dl","dt","em","hr","h1","h2","h3","h4","h5","h6","li","ol","p","pre","s","small","span","sub","sup","strong","u","ul","uriAttributes","SAFE_URL_PATTERN","allowedAttribute","allowedAttributeList","attributeName","nodeValue","attributeRegex","regex","Default$4","allowList","content","extraClass","sanitize","sanitizeFn","template","DefaultType$4","DefaultContentType","entry","TemplateFactory","getContent","_resolvePossibleFunction","hasContent","changeContent","_checkContent","toHtml","templateWrapper","innerHTML","_maybeSanitize","text","_setContent","arg","templateElement","_putElementInTemplate","textContent","unsafeHtml","sanitizeFunction","createdDocument","DOMParser","parseFromString","elementName","attributeList","allowedAttributes","sanitizeHtml","DISALLOWED_ATTRIBUTES","CLASS_NAME_FADE$2","CLASS_NAME_SHOW$2","SELECTOR_TOOLTIP_INNER","SELECTOR_MODAL","EVENT_MODAL_HIDE","TRIGGER_HOVER","TRIGGER_FOCUS","AttachmentMap","AUTO","TOP","RIGHT","BOTTOM","LEFT","Default$3","animation","container","customClass","delay","title","DefaultType$3","Tooltip","_isEnabled","_timeout","_isHovered","_activeTrigger","_templateFactory","_newContent","tip","_setListeners","_fixTitle","enable","disable","toggleEnabled","click","_leave","_enter","_hideModalHandler","_disposePopper","_isWithContent","isInTheDom","_getTipElement","_isWithActiveTrigger","_getTitle","_createTipElement","_getContentForTemplate","_getTemplateFactory","tipId","prefix","floor","random","getElementById","getUID","setContent","_initializeOnDelegatedTarget","_getDelegateConfig","attachment","triggers","eventIn","eventOut","_setTimeout","timeout","dataAttributes","dataAttribute","SELECTOR_TITLE","SELECTOR_CONTENT","Default$2","DefaultType$2","Popover","_getContent","EVENT_KEY$2","EVENT_ACTIVATE","EVENT_CLICK","EVENT_LOAD_DATA_API$1","CLASS_NAME_ACTIVE$1","SELECTOR_TARGET_LINKS","SELECTOR_NAV_LINKS","SELECTOR_LINK_ITEMS","Default$1","rootMargin","smoothScroll","threshold","DefaultType$1","ScrollSpy","_targetLinks","_observableSections","_rootElement","_activeTarget","_observer","_previousScrollData","visibleEntryTop","parentScrollTop","refresh","_initializeTargetsAndObservables","_maybeEnableSmoothScroll","disconnect","_getNewObserver","section","observe","observableSection","scrollTo","behavior","IntersectionObserver","_observerCallback","targetElement","_process","userScrollsDown","isIntersecting","_clearActiveClass","entryIsLowerThanPrevious","targetLinks","anchor","decodeURI","_activateParents","listGroup","activeNodes","spy","EVENT_KEY$1","EVENT_HIDE$1","EVENT_HIDDEN$1","EVENT_SHOW$1","EVENT_SHOWN$1","EVENT_CLICK_DATA_API","EVENT_KEYDOWN","EVENT_LOAD_DATA_API","ARROW_LEFT_KEY","ARROW_RIGHT_KEY","ARROW_UP_KEY","ARROW_DOWN_KEY","HOME_KEY","END_KEY","CLASS_NAME_ACTIVE","CLASS_NAME_FADE$1","CLASS_NAME_SHOW$1","SELECTOR_DROPDOWN_TOGGLE","NOT_SELECTOR_DROPDOWN_TOGGLE","SELECTOR_DATA_TOGGLE","SELECTOR_INNER_ELEM","SELECTOR_DATA_TOGGLE_ACTIVE","Tab","_setInitialAttributes","_getChildren","innerElem","_elemIsActive","active","_getActiveElem","hideEvent","_deactivate","_activate","relatedElem","_toggleDropDown","nextActiveElement","preventScroll","_setAttributeIfNotExists","_setInitialAttributesOnChild","_getInnerElement","isActive","outerElem","_getOuterElement","_setInitialAttributesOnTargetPanel","open","EVENT_MOUSEOVER","EVENT_MOUSEOUT","EVENT_FOCUSIN","EVENT_FOCUSOUT","EVENT_HIDE","EVENT_HIDDEN","EVENT_SHOW","EVENT_SHOWN","CLASS_NAME_HIDE","CLASS_NAME_SHOW","CLASS_NAME_SHOWING","autohide","Toast","_hasMouseInteraction","_hasKeyboardInteraction","_clearTimeout","_maybeScheduleHide","isShown","_onInteraction","isInteracting","documentReady","tooltipTriggerEl","btn","header","getElementsByClassName","oldScroll","scrollY","bootstrap"],"sourceRoot":""}
\ No newline at end of file
diff --git a/_static/scripts/fontawesome.js b/_static/scripts/fontawesome.js
new file mode 100644
index 000000000..71955b429
--- /dev/null
+++ b/_static/scripts/fontawesome.js
@@ -0,0 +1,3 @@
+/*! For license information please see fontawesome.js.LICENSE.txt */
+(()=>{var c={654:()=>{!function(){"use strict";var c={},l={};try{"undefined"!=typeof window&&(c=window),"undefined"!=typeof document&&(l=document)}catch(c){}var s=void 0===(v=(c.navigator||{}).userAgent)?"":v,a=c,z=l;function e(c,l){var s,a=Object.keys(c);return Object.getOwnPropertySymbols&&(s=Object.getOwnPropertySymbols(c),l&&(s=s.filter((function(l){return Object.getOwnPropertyDescriptor(c,l).enumerable}))),a.push.apply(a,s)),a}function H(c){for(var l=1;l<arguments.length;l++){var s=null!=arguments[l]?arguments[l]:{};l%2?e(Object(s),!0).forEach((function(l){t(c,l,s[l])})):Object.getOwnPropertyDescriptors?Object.defineProperties(c,Object.getOwnPropertyDescriptors(s)):e(Object(s)).forEach((function(l){Object.defineProperty(c,l,Object.getOwnPropertyDescriptor(s,l))}))}return c}function t(c,l,s){return l in c?Object.defineProperty(c,l,{value:s,enumerable:!0,configurable:!0,writable:!0}):c[l]=s,c}function V(c,l){(null==l||l>c.length)&&(l=c.length);for(var s=0,a=new Array(l);s<l;s++)a[s]=c[s];return a}a.document,z.documentElement&&z.head&&"function"==typeof z.addEventListener&&z.createElement,~s.indexOf("MSIE")||s.indexOf("Trident/");var r="___FONT_AWESOME___",M=function(){try{return!0}catch(c){return!1}}(),h="classic",n="sharp",i=[h,n];function m(c){return new Proxy(c,{get:function(c,l){return l in c?c[l]:c[h]}})}m((t(f={},h,{fa:"solid",fas:"solid","fa-solid":"solid",far:"regular","fa-regular":"regular",fal:"light","fa-light":"light",fat:"thin","fa-thin":"thin",fad:"duotone","fa-duotone":"duotone",fab:"brands","fa-brands":"brands",fak:"kit",fakd:"kit","fa-kit":"kit","fa-kit-duotone":"kit"}),t(f,n,{fa:"solid",fass:"solid","fa-solid":"solid",fasr:"regular","fa-regular":"regular",fasl:"light","fa-light":"light",fast:"thin","fa-thin":"thin"}),f));var o=m((t(C={},h,{solid:"fas",regular:"far",light:"fal",thin:"fat",duotone:"fad",brands:"fab",kit:"fak"}),t(C,n,{solid:"fass",regular:"fasr",light:"fasl",thin:"fast"}),C)),f=(s=(m((t(v={},h,{fab:"fa-brands",fad:"fa-duotone",fak:"fa-kit",fal:"fa-light",far:"fa-regular",fas:"fa-solid",fat:"fa-thin"}),t(v,n,{fass:"fa-solid",fasr:"fa-regular",fasl:"fa-light",fast:"fa-thin"}),v)),m((t(c={},h,{"fa-brands":"fab","fa-duotone":"fad","fa-kit":"fak","fa-light":"fal","fa-regular":"far","fa-solid":"fas","fa-thin":"fat"}),t(c,n,{"fa-solid":"fass","fa-regular":"fasr","fa-light":"fasl","fa-thin":"fast"}),c)),m((t(l={},h,{900:"fas",400:"far",normal:"far",300:"fal",100:"fat"}),t(l,n,{900:"fass",400:"fasr",300:"fasl",100:"fast"}),l)),z=[1,2,3,4,5,6,7,8,9,10]).concat([11,12,13,14,15,16,17,18,19,20]),"duotone-group"),C="swap-opacity",v="primary";c="secondary",l=new Set,Object.keys(o[h]).map(l.add.bind(l)),Object.keys(o[n]).map(l.add.bind(l)),[].concat(i,function(c){if(Array.isArray(c))return V(c)}(l)||function(c){if("undefined"!=typeof Symbol&&null!=c[Symbol.iterator]||null!=c["@@iterator"])return Array.from(c)}(l)||function(c,l){if(c){if("string"==typeof c)return V(c,l);var s=Object.prototype.toString.call(c).slice(8,-1);return"Map"===(s="Object"===s&&c.constructor?c.constructor.name:s)||"Set"===s?Array.from(c):"Arguments"===s||/^(?:Ui|I)nt(?:8|16|32)(?:Clamped)?Array$/.test(s)?V(c,l):void 0}}(l)||function(){throw new TypeError("Invalid attempt to spread non-iterable instance.\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method.")}(),["2xs","xs","sm","lg","xl","2xl","beat","border","fade","beat-fade","bounce","flip-both","flip-horizontal","flip-vertical","flip","fw","inverse","layers-counter","layers-text","layers","li","pull-left","pull-right","pulse","rotate-180","rotate-270","rotate-90","rotate-by","shake","spin-pulse","spin-reverse","spin","stack-1x","stack-2x","stack","ul",f,C,v,c]).concat(z.map((function(c){return"".concat(c,"x")}))).concat(s.map((function(c){return"w-".concat(c)}))),(a=a||{})[r]||(a[r]={}),a[r].styles||(a[r].styles={}),a[r].hooks||(a[r].hooks={}),a[r].shims||(a[r].shims=[]);var L=a[r];function u(c){return Object.keys(c).reduce((function(l,s){var a=c[s];return a.icon?l[a.iconName]=a.icon:l[s]=a,l}),{})}function d(c,l,s){s=void 0!==(a=(2<arguments.length&&void 0!==s?s:{}).skipHooks)&&a;var a=u(l);"function"!=typeof L.hooks.addPack||s?L.styles[c]=H(H({},L.styles[c]||{}),a):L.hooks.addPack(c,u(l)),"fas"===c&&d("fa",l)}var p={monero:[496,512,[],"f3d0","M352 384h108.4C417 455.9 338.1 504 248 504S79 455.9 35.6 384H144V256.2L248 361l104-105v128zM88 336V128l159.4 159.4L408 128v208h74.8c8.5-25.1 13.2-52 13.2-80C496 119 385 8 248 8S0 119 0 256c0 28 4.6 54.9 13.2 80H88z"],hooli:[640,512,[],"f427","M144.5 352l38.3.8c-13.2-4.6-26-10.2-38.3-16.8zm57.7-5.3v5.3l-19.4.8c36.5 12.5 69.9 14.2 94.7 7.2-19.9.2-45.8-2.6-75.3-13.3zm408.9-115.2c15.9 0 28.9-12.9 28.9-28.9s-12.9-24.5-28.9-24.5c-15.9 0-28.9 8.6-28.9 24.5s12.9 28.9 28.9 28.9zm-29 120.5H640V241.5h-57.9zm-73.7 0h57.9V156.7L508.4 184zm-31-119.4c-18.2-18.2-50.4-17.1-50.4-17.1s-32.3-1.1-50.4 17.1c-18.2 18.2-16.8 33.9-16.8 52.6s-1.4 34.3 16.8 52.5 50.4 17.1 50.4 17.1 32.3 1.1 50.4-17.1c18.2-18.2 16.8-33.8 16.8-52.5-.1-18.8 1.3-34.5-16.8-52.6zm-39.8 71.9c0 3.6-1.8 12.5-10.7 12.5s-10.7-8.9-10.7-12.5v-40.4c0-8.7 7.3-10.9 10.7-10.9s10.7 2.1 10.7 10.9zm-106.2-71.9c-18.2-18.2-50.4-17.1-50.4-17.1s-32.2-1.1-50.4 17.1c-1.9 1.9-3.7 3.9-5.3 6-38.2-29.6-72.5-46.5-102.1-61.1v-20.7l-22.5 10.6c-54.4-22.1-89-18.2-97.3.1 0 0-24.9 32.8 61.8 110.8V352h57.9v-28.6c-6.5-4.2-13-8.7-19.4-13.6-14.8-11.2-27.4-21.6-38.4-31.4v-31c13.1 14.7 30.5 31.4 53.4 50.3l4.5 3.6v-29.8c0-6.9 1.7-18.2 10.8-18.2s10.6 6.9 10.6 15V317c18 12.2 37.3 22.1 57.7 29.6v-93.9c0-18.7-13.4-37.4-40.6-37.4-15.8-.1-30.5 8.2-38.5 21.9v-54.3c41.9 20.9 83.9 46.5 99.9 58.3-10.2 14.6-9.3 28.1-9.3 43.7 0 18.7-1.4 34.3 16.8 52.5s50.4 17.1 50.4 17.1 32.3 1.1 50.4-17.1c18.2-18.2 16.7-33.8 16.7-52.5 0-18.5 1.5-34.2-16.7-52.3zM65.2 184v63.3c-48.7-54.5-38.9-76-35.2-79.1 13.5-11.4 37.5-8 64.4 2.1zm226.5 120.5c0 3.6-1.8 12.5-10.7 12.5s-10.7-8.9-10.7-12.5v-40.4c0-8.7 7.3-10.9 10.7-10.9s10.7 2.1 10.7 10.9z"],yelp:[384,512,[],"f1e9","M42.9 240.32l99.62 48.61c19.2 9.4 16.2 37.51-4.5 42.71L30.5 358.45a22.79 22.79 0 0 1-28.21-19.6 197.16 197.16 0 0 1 9-85.32 22.8 22.8 0 0 1 31.61-13.21zm44 239.25a199.45 199.45 0 0 0 79.42 32.11A22.78 22.78 0 0 0 192.94 490l3.9-110.82c.7-21.3-25.5-31.91-39.81-16.1l-74.21 82.4a22.82 22.82 0 0 0 4.09 34.09zm145.34-109.92l58.81 94a22.93 22.93 0 0 0 34 5.5 198.36 198.36 0 0 0 52.71-67.61A23 23 0 0 0 364.17 370l-105.42-34.26c-20.31-6.5-37.81 15.8-26.51 33.91zm148.33-132.23a197.44 197.44 0 0 0-50.41-69.31 22.85 22.85 0 0 0-34 4.4l-62 91.92c-11.9 17.7 4.7 40.61 25.2 34.71L366 268.63a23 23 0 0 0 14.61-31.21zM62.11 30.18a22.86 22.86 0 0 0-9.9 32l104.12 180.44c11.7 20.2 42.61 11.9 42.61-11.4V22.88a22.67 22.67 0 0 0-24.5-22.8 320.37 320.37 0 0 0-112.33 30.1z"],"cc-visa":[576,512,[],"f1f0","M470.1 231.3s7.6 37.2 9.3 45H446c3.3-8.9 16-43.5 16-43.5-.2.3 3.3-9.1 5.3-14.9l2.8 13.4zM576 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h480c26.5 0 48 21.5 48 48zM152.5 331.2L215.7 176h-42.5l-39.3 106-4.3-21.5-14-71.4c-2.3-9.9-9.4-12.7-18.2-13.1H32.7l-.7 3.1c15.8 4 29.9 9.8 42.2 17.1l35.8 135h42.5zm94.4.2L272.1 176h-40.2l-25.1 155.4h40.1zm139.9-50.8c.2-17.7-10.6-31.2-33.7-42.3-14.1-7.1-22.7-11.9-22.7-19.2.2-6.6 7.3-13.4 23.1-13.4 13.1-.3 22.7 2.8 29.9 5.9l3.6 1.7 5.5-33.6c-7.9-3.1-20.5-6.6-36-6.6-39.7 0-67.6 21.2-67.8 51.4-.3 22.3 20 34.7 35.2 42.2 15.5 7.6 20.8 12.6 20.8 19.3-.2 10.4-12.6 15.2-24.1 15.2-16 0-24.6-2.5-37.7-8.3l-5.3-2.5-5.6 34.9c9.4 4.3 26.8 8.1 44.8 8.3 42.2.1 69.7-20.8 70-53zM528 331.4L495.6 176h-31.1c-9.6 0-16.9 2.8-21 12.9l-59.7 142.5H426s6.9-19.2 8.4-23.3H486c1.2 5.5 4.8 23.3 4.8 23.3H528z"],lastfm:[512,512,[],"f202","M225.8 367.1l-18.8-51s-30.5 34-76.2 34c-40.5 0-69.2-35.2-69.2-91.5 0-72.1 36.4-97.9 72.1-97.9 66.5 0 74.8 53.3 100.9 134.9 18.8 56.9 54 102.6 155.4 102.6 72.7 0 122-22.3 122-80.9 0-72.9-62.7-80.6-115-92.1-25.8-5.9-33.4-16.4-33.4-34 0-19.9 15.8-31.7 41.6-31.7 28.2 0 43.4 10.6 45.7 35.8l58.6-7c-4.7-52.8-41.1-74.5-100.9-74.5-52.8 0-104.4 19.9-104.4 83.9 0 39.9 19.4 65.1 68 76.8 44.9 10.6 79.8 13.8 79.8 45.7 0 21.7-21.1 30.5-61 30.5-59.2 0-83.9-31.1-97.9-73.9-32-96.8-43.6-163-161.3-163C45.7 113.8 0 168.3 0 261c0 89.1 45.7 137.2 127.9 137.2 66.2 0 97.9-31.1 97.9-31.1z"],shopware:[512,512,[],"f5b5","M403.5 455.41A246.17 246.17 0 0 1 256 504C118.81 504 8 393 8 256 8 118.81 119 8 256 8a247.39 247.39 0 0 1 165.7 63.5 3.57 3.57 0 0 1-2.86 6.18A418.62 418.62 0 0 0 362.13 74c-129.36 0-222.4 53.47-222.4 155.35 0 109 92.13 145.88 176.83 178.73 33.64 13 65.4 25.36 87 41.59a3.58 3.58 0 0 1 0 5.72zM503 233.09a3.64 3.64 0 0 0-1.27-2.44c-51.76-43-93.62-60.48-144.48-60.48-84.13 0-80.25 52.17-80.25 53.63 0 42.6 52.06 62 112.34 84.49 31.07 11.59 63.19 23.57 92.68 39.93a3.57 3.57 0 0 0 5-1.82A249 249 0 0 0 503 233.09z"],"creative-commons-nc":[496,512,[],"f4e8","M247.6 8C387.4 8 496 115.9 496 256c0 147.2-118.5 248-248.4 248C113.1 504 0 393.2 0 256 0 123.1 104.7 8 247.6 8zM55.8 189.1c-7.4 20.4-11.1 42.7-11.1 66.9 0 110.9 92.1 202.4 203.7 202.4 122.4 0 177.2-101.8 178.5-104.1l-93.4-41.6c-7.7 37.1-41.2 53-68.2 55.4v38.1h-28.8V368c-27.5-.3-52.6-10.2-75.3-29.7l34.1-34.5c31.7 29.4 86.4 31.8 86.4-2.2 0-6.2-2.2-11.2-6.6-15.1-14.2-6-1.8-.1-219.3-97.4zM248.4 52.3c-38.4 0-112.4 8.7-170.5 93l94.8 42.5c10-31.3 40.4-42.9 63.8-44.3v-38.1h28.8v38.1c22.7 1.2 43.4 8.9 62 23L295 199.7c-42.7-29.9-83.5-8-70 11.1 53.4 24.1 43.8 19.8 93 41.6l127.1 56.7c4.1-17.4 6.2-35.1 6.2-53.1 0-57-19.8-105-59.3-143.9-39.3-39.9-87.2-59.8-143.6-59.8z"],aws:[640,512,[],"f375","M180.41 203.01c-.72 22.65 10.6 32.68 10.88 39.05a8.164 8.164 0 0 1-4.1 6.27l-12.8 8.96a10.66 10.66 0 0 1-5.63 1.92c-.43-.02-8.19 1.83-20.48-25.61a78.608 78.608 0 0 1-62.61 29.45c-16.28.89-60.4-9.24-58.13-56.21-1.59-38.28 34.06-62.06 70.93-60.05 7.1.02 21.6.37 46.99 6.27v-15.62c2.69-26.46-14.7-46.99-44.81-43.91-2.4.01-19.4-.5-45.84 10.11-7.36 3.38-8.3 2.82-10.75 2.82-7.41 0-4.36-21.48-2.94-24.2 5.21-6.4 35.86-18.35 65.94-18.18a76.857 76.857 0 0 1 55.69 17.28 70.285 70.285 0 0 1 17.67 52.36l-.01 69.29zM93.99 235.4c32.43-.47 46.16-19.97 49.29-30.47 2.46-10.05 2.05-16.41 2.05-27.4-9.67-2.32-23.59-4.85-39.56-4.87-15.15-1.14-42.82 5.63-41.74 32.26-1.24 16.79 11.12 31.4 29.96 30.48zm170.92 23.05c-7.86.72-11.52-4.86-12.68-10.37l-49.8-164.65c-.97-2.78-1.61-5.65-1.92-8.58a4.61 4.61 0 0 1 3.86-5.25c.24-.04-2.13 0 22.25 0 8.78-.88 11.64 6.03 12.55 10.37l35.72 140.83 33.16-140.83c.53-3.22 2.94-11.07 12.8-10.24h17.16c2.17-.18 11.11-.5 12.68 10.37l33.42 142.63L420.98 80.1c.48-2.18 2.72-11.37 12.68-10.37h19.72c.85-.13 6.15-.81 5.25 8.58-.43 1.85 3.41-10.66-52.75 169.9-1.15 5.51-4.82 11.09-12.68 10.37h-18.69c-10.94 1.15-12.51-9.66-12.68-10.75L328.67 110.7l-32.78 136.99c-.16 1.09-1.73 11.9-12.68 10.75h-18.3zm273.48 5.63c-5.88.01-33.92-.3-57.36-12.29a12.802 12.802 0 0 1-7.81-11.91v-10.75c0-8.45 6.2-6.9 8.83-5.89 10.04 4.06 16.48 7.14 28.81 9.6 36.65 7.53 52.77-2.3 56.72-4.48 13.15-7.81 14.19-25.68 5.25-34.95-10.48-8.79-15.48-9.12-53.13-21-4.64-1.29-43.7-13.61-43.79-52.36-.61-28.24 25.05-56.18 69.52-55.95 12.67-.01 46.43 4.13 55.57 15.62 1.35 2.09 2.02 4.55 1.92 7.04v10.11c0 4.44-1.62 6.66-4.87 6.66-7.71-.86-21.39-11.17-49.16-10.75-6.89-.36-39.89.91-38.41 24.97-.43 18.96 26.61 26.07 29.7 26.89 36.46 10.97 48.65 12.79 63.12 29.58 17.14 22.25 7.9 48.3 4.35 55.44-19.08 37.49-68.42 34.44-69.26 34.42zm40.2 104.86c-70.03 51.72-171.69 79.25-258.49 79.25A469.127 469.127 0 0 1 2.83 327.46c-6.53-5.89-.77-13.96 7.17-9.47a637.37 637.37 0 0 0 316.88 84.12 630.22 630.22 0 0 0 241.59-49.55c11.78-5 21.77 7.8 10.12 16.38zm29.19-33.29c-8.96-11.52-59.28-5.38-81.81-2.69-6.79.77-7.94-5.12-1.79-9.47 40.07-28.17 105.88-20.1 113.44-10.63 7.55 9.47-2.05 75.41-39.56 106.91-5.76 4.87-11.27 2.3-8.71-4.1 8.44-21.25 27.39-68.49 18.43-80.02z"],redhat:[512,512,[],"f7bc","M341.52 285.56c33.65 0 82.34-6.94 82.34-47 .22-6.74.86-1.82-20.88-96.24-4.62-19.15-8.68-27.84-42.31-44.65-26.09-13.34-82.92-35.37-99.73-35.37-15.66 0-20.2 20.17-38.87 20.17-18 0-31.31-15.06-48.12-15.06-16.14 0-26.66 11-34.78 33.62-27.5 77.55-26.28 74.27-26.12 78.27 0 24.8 97.64 106.11 228.47 106.11M429 254.84c4.65 22 4.65 24.35 4.65 27.25 0 37.66-42.33 58.56-98 58.56-125.74.08-235.91-73.65-235.91-122.33a49.55 49.55 0 0 1 4.06-19.72C58.56 200.86 0 208.93 0 260.63c0 84.67 200.63 189 359.49 189 121.79 0 152.51-55.08 152.51-98.58 0-34.21-29.59-73.05-82.93-96.24"],yoast:[448,512,[],"f2b1","M91.3 76h186l-7 18.9h-179c-39.7 0-71.9 31.6-71.9 70.3v205.4c0 35.4 24.9 70.3 84 70.3V460H91.3C41.2 460 0 419.8 0 370.5V165.2C0 115.9 40.7 76 91.3 76zm229.1-56h66.5C243.1 398.1 241.2 418.9 202.2 459.3c-20.8 21.6-49.3 31.7-78.3 32.7v-51.1c49.2-7.7 64.6-49.9 64.6-75.3 0-20.1.6-12.6-82.1-223.2h61.4L218.2 299 320.4 20zM448 161.5V460H234c6.6-9.6 10.7-16.3 12.1-19.4h182.5V161.5c0-32.5-17.1-51.9-48.2-62.9l6.7-17.6c41.7 13.6 60.9 43.1 60.9 80.5z"],cloudflare:[640,512,[],"e07d","M407.906,319.913l-230.8-2.928a4.58,4.58,0,0,1-3.632-1.926,4.648,4.648,0,0,1-.494-4.147,6.143,6.143,0,0,1,5.361-4.076L411.281,303.9c27.631-1.26,57.546-23.574,68.022-50.784l13.286-34.542a7.944,7.944,0,0,0,.524-2.936,7.735,7.735,0,0,0-.164-1.631A151.91,151.91,0,0,0,201.257,198.4,68.12,68.12,0,0,0,94.2,269.59C41.924,271.106,0,313.728,0,366.12a96.054,96.054,0,0,0,1.029,13.958,4.508,4.508,0,0,0,4.445,3.871l426.1.051c.043,0,.08-.019.122-.02a5.606,5.606,0,0,0,5.271-4l3.273-11.265c3.9-13.4,2.448-25.8-4.1-34.9C430.124,325.423,420.09,320.487,407.906,319.913ZM513.856,221.1c-2.141,0-4.271.062-6.391.164a3.771,3.771,0,0,0-3.324,2.653l-9.077,31.193c-3.9,13.4-2.449,25.786,4.1,34.89,6.02,8.4,16.054,13.323,28.238,13.9l49.2,2.939a4.491,4.491,0,0,1,3.51,1.894,4.64,4.64,0,0,1,.514,4.169,6.153,6.153,0,0,1-5.351,4.075l-51.125,2.939c-27.754,1.27-57.669,23.574-68.145,50.784l-3.695,9.606a2.716,2.716,0,0,0,2.427,3.68c.046,0,.088.017.136.017h175.91a4.69,4.69,0,0,0,4.539-3.37,124.807,124.807,0,0,0,4.682-34C640,277.3,583.524,221.1,513.856,221.1Z"],ups:[384,512,[],"f7e0","M103.2 303c-5.2 3.6-32.6 13.1-32.6-19V180H37.9v102.6c0 74.9 80.2 51.1 97.9 39V180h-32.6zM4 74.82v220.9c0 103.7 74.9 135.2 187.7 184.1 112.4-48.9 187.7-80.2 187.7-184.1V74.82c-116.3-61.6-281.8-49.6-375.4 0zm358.1 220.9c0 86.6-53.2 113.6-170.4 165.3-117.5-51.8-170.5-78.7-170.5-165.3v-126.4c102.3-93.8 231.6-100 340.9-89.8zm-209.6-107.4v212.8h32.7v-68.7c24.4 7.3 71.7-2.6 71.7-78.5 0-97.4-80.7-80.92-104.4-65.6zm32.7 117.3v-100.3c8.4-4.2 38.4-12.7 38.4 49.3 0 67.9-36.4 51.8-38.4 51zm79.1-86.4c.1 47.3 51.6 42.5 52.2 70.4.6 23.5-30.4 23-50.8 4.9v30.1c36.2 21.5 81.9 8.1 83.2-33.5 1.7-51.5-54.1-46.6-53.4-73.2.6-20.3 30.6-20.5 48.5-2.2v-28.4c-28.5-22-79.9-9.2-79.7 31.9z"],pixiv:[448,512,[],"e640","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm171.5 84c41 0 76.3 12.9 101.4 35.2l0 0c25.2 22.2 39.8 54.1 39.8 88.8c.1 35.3-16.6 66.3-42.4 87c-25.9 20.8-60.6 32.4-98.8 32.4c-43.5 0-83.8-16.1-83.8-16.1v51.8c7.4 2.2 19.7 7 11.9 14.8H104.8c-7.7-7.8 3.6-12.4 12.1-14.8V175.5C97.1 190.9 87 204.3 81.8 214.2c6 19.4-5.3 18.5-5.3 18.5L56 199.7s72.7-83.7 179.5-83.7zm-3.6 222.9c30 0 56-11.3 73.9-29.2c17.9-18.1 27.9-41.6 28-70.2c-.1-29.3-9.5-54.6-26.7-73.6c-17.2-18.9-42.7-31.3-75.2-31.4c-26.7-.1-59.8 9-80.2 23.7V323.1c18.6 9.3 46.8 15.9 80.2 15.8z"],wpexplorer:[512,512,[],"f2de","M512 256c0 141.2-114.7 256-256 256C114.8 512 0 397.3 0 256S114.7 0 256 0s256 114.7 256 256zm-32 0c0-123.2-100.3-224-224-224C132.5 32 32 132.5 32 256s100.5 224 224 224 224-100.5 224-224zM160.9 124.6l86.9 37.1-37.1 86.9-86.9-37.1 37.1-86.9zm110 169.1l46.6 94h-14.6l-50-100-48.9 100h-14l51.1-106.9-22.3-9.4 6-14 68.6 29.1-6 14.3-16.5-7.1zm-11.8-116.3l68.6 29.4-29.4 68.3L230 246l29.1-68.6zm80.3 42.9l54.6 23.1-23.4 54.3-54.3-23.1 23.1-54.3z"],dyalog:[416,512,[],"f399","M0 32v119.2h64V96h107.2C284.6 96 352 176.2 352 255.9 352 332 293.4 416 171.2 416H0v64h171.2C331.9 480 416 367.3 416 255.9c0-58.7-22.1-113.4-62.3-154.3C308.9 56 245.7 32 171.2 32H0z"],bity:[496,512,[],"f37a","M78.4 67.2C173.8-22 324.5-24 421.5 71c14.3 14.1-6.4 37.1-22.4 21.5-84.8-82.4-215.8-80.3-298.9-3.2-16.3 15.1-36.5-8.3-21.8-22.1zm98.9 418.6c19.3 5.7 29.3-23.6 7.9-30C73 421.9 9.4 306.1 37.7 194.8c5-19.6-24.9-28.1-30.2-7.1-32.1 127.4 41.1 259.8 169.8 298.1zm148.1-2c121.9-40.2 192.9-166.9 164.4-291-4.5-19.7-34.9-13.8-30 7.9 24.2 107.7-37.1 217.9-143.2 253.4-21.2 7-10.4 36 8.8 29.7zm-62.9-79l.2-71.8c0-8.2-6.6-14.8-14.8-14.8-8.2 0-14.8 6.7-14.8 14.8l-.2 71.8c0 8.2 6.6 14.8 14.8 14.8s14.8-6.6 14.8-14.8zm71-269c2.1 90.9 4.7 131.9-85.5 132.5-92.5-.7-86.9-44.3-85.5-132.5 0-21.8-32.5-19.6-32.5 0v71.6c0 69.3 60.7 90.9 118 90.1 57.3.8 118-20.8 118-90.1v-71.6c0-19.6-32.5-21.8-32.5 0z"],stackpath:[448,512,[],"f842","M244.6 232.4c0 8.5-4.26 20.49-21.34 20.49h-19.61v-41.47h19.61c17.13 0 21.34 12.36 21.34 20.98zM448 32v448H0V32zM151.3 287.84c0-21.24-12.12-34.54-46.72-44.85-20.57-7.41-26-10.91-26-18.63s7-14.61 20.41-14.61c14.09 0 20.79 8.45 20.79 18.35h30.7l.19-.57c.5-19.57-15.06-41.65-51.12-41.65-23.37 0-52.55 10.75-52.55 38.29 0 19.4 9.25 31.29 50.74 44.37 17.26 6.15 21.91 10.4 21.91 19.48 0 15.2-19.13 14.23-19.47 14.23-20.4 0-25.65-9.1-25.65-21.9h-30.8l-.18.56c-.68 31.32 28.38 45.22 56.63 45.22 29.98 0 51.12-13.55 51.12-38.29zm125.38-55.63c0-25.3-18.43-45.46-53.42-45.46h-51.78v138.18h32.17v-47.36h19.61c30.25 0 53.42-15.95 53.42-45.36zM297.94 325L347 186.78h-31.09L268 325zm106.52-138.22h-31.09L325.46 325h29.94z"],buysellads:[448,512,[],"f20d","M224 150.7l42.9 160.7h-85.8L224 150.7zM448 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h352c26.5 0 48 21.5 48 48zm-65.3 325.3l-94.5-298.7H159.8L65.3 405.3H156l111.7-91.6 24.2 91.6h90.8z"],"first-order":[448,512,[],"f2b0","M12.9 229.2c.1-.1.2-.3.3-.4 0 .1 0 .3-.1.4h-.2zM224 96.6c-7.1 0-14.6.6-21.4 1.7l3.7 67.4-22-64c-14.3 3.7-27.7 9.4-40 16.6l29.4 61.4-45.1-50.9c-11.4 8.9-21.7 19.1-30.6 30.9l50.6 45.4-61.1-29.7c-7.1 12.3-12.9 25.7-16.6 40l64.3 22.6-68-4c-.9 7.1-1.4 14.6-1.4 22s.6 14.6 1.4 21.7l67.7-4-64 22.6c3.7 14.3 9.4 27.7 16.6 40.3l61.1-29.7L97.7 352c8.9 11.7 19.1 22.3 30.9 30.9l44.9-50.9-29.5 61.4c12.3 7.4 25.7 13.1 40 16.9l22.3-64.6-4 68c7.1 1.1 14.6 1.7 21.7 1.7 7.4 0 14.6-.6 21.7-1.7l-4-68.6 22.6 65.1c14.3-4 27.7-9.4 40-16.9L274.9 332l44.9 50.9c11.7-8.9 22-19.1 30.6-30.9l-50.6-45.1 61.1 29.4c7.1-12.3 12.9-25.7 16.6-40.3l-64-22.3 67.4 4c1.1-7.1 1.4-14.3 1.4-21.7s-.3-14.9-1.4-22l-67.7 4 64-22.3c-3.7-14.3-9.1-28-16.6-40.3l-60.9 29.7 50.6-45.4c-8.9-11.7-19.1-22-30.6-30.9l-45.1 50.9 29.4-61.1c-12.3-7.4-25.7-13.1-40-16.9L241.7 166l4-67.7c-7.1-1.2-14.3-1.7-21.7-1.7zM443.4 128v256L224 512 4.6 384V128L224 0l219.4 128zm-17.1 10.3L224 20.9 21.7 138.3v235.1L224 491.1l202.3-117.7V138.3zM224 37.1l187.7 109.4v218.9L224 474.9 36.3 365.4V146.6L224 37.1zm0 50.9c-92.3 0-166.9 75.1-166.9 168 0 92.6 74.6 167.7 166.9 167.7 92 0 166.9-75.1 166.9-167.7 0-92.9-74.9-168-166.9-168z"],modx:[448,512,[],"f285","M356 241.8l36.7 23.7V480l-133-83.8L356 241.8zM440 75H226.3l-23 37.8 153.5 96.5L440 75zm-89 142.8L55.2 32v214.5l46 29L351 217.8zM97 294.2L8 437h213.7l125-200.5L97 294.2z"],guilded:[448,512,[],"e07e","M443.427,64H4.571c0,103.26,22.192,180.06,43.418,222.358C112.046,414.135,224,448,225.256,448a312.824,312.824,0,0,0,140.55-103.477c25.907-33.923,53.1-87.19,65.916-145.761H171.833c4.14,36.429,22.177,67.946,45.1,86.944h88.589c-17.012,28.213-48.186,54.4-80.456,69.482-31.232-13.259-69.09-46.544-96.548-98.362-26.726-53.833-27.092-105.883-27.092-105.883H437.573A625.91,625.91,0,0,0,443.427,64Z"],vnv:[640,512,[],"f40b","M104.9 352c-34.1 0-46.4-30.4-46.4-30.4L2.6 210.1S-7.8 192 13 192h32.8c10.4 0 13.2 8.7 18.8 18.1l36.7 74.5s5.2 13.1 21.1 13.1 21.1-13.1 21.1-13.1l36.7-74.5c5.6-9.5 8.4-18.1 18.8-18.1h32.8c20.8 0 10.4 18.1 10.4 18.1l-55.8 111.5S174.2 352 140 352h-35.1zm395 0c-34.1 0-46.4-30.4-46.4-30.4l-55.9-111.5S387.2 192 408 192h32.8c10.4 0 13.2 8.7 18.8 18.1l36.7 74.5s5.2 13.1 21.1 13.1 21.1-13.1 21.1-13.1l36.8-74.5c5.6-9.5 8.4-18.1 18.8-18.1H627c20.8 0 10.4 18.1 10.4 18.1l-55.9 111.5S569.3 352 535.1 352h-35.2zM337.6 192c34.1 0 46.4 30.4 46.4 30.4l55.9 111.5s10.4 18.1-10.4 18.1h-32.8c-10.4 0-13.2-8.7-18.8-18.1l-36.7-74.5s-5.2-13.1-21.1-13.1c-15.9 0-21.1 13.1-21.1 13.1l-36.7 74.5c-5.6 9.4-8.4 18.1-18.8 18.1h-32.9c-20.8 0-10.4-18.1-10.4-18.1l55.9-111.5s12.2-30.4 46.4-30.4h35.1z"],"square-js":[448,512,["js-square"],"f3b9","M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM180.9 444.9c-33.7 0-53.2-17.4-63.2-38.5L152 385.7c6.6 11.7 12.6 21.6 27.1 21.6c13.8 0 22.6-5.4 22.6-26.5V237.7h42.1V381.4c0 43.6-25.6 63.5-62.9 63.5zm85.8-43L301 382.1c9 14.7 20.8 25.6 41.5 25.6c17.4 0 28.6-8.7 28.6-20.8c0-14.4-11.4-19.5-30.7-28l-10.5-4.5c-30.4-12.9-50.5-29.2-50.5-63.5c0-31.6 24.1-55.6 61.6-55.6c26.8 0 46 9.3 59.8 33.7L368 290c-7.2-12.9-15-18-27.1-18c-12.3 0-20.1 7.8-20.1 18c0 12.6 7.8 17.7 25.9 25.6l10.5 4.5c35.8 15.3 55.9 31 55.9 66.2c0 37.8-29.8 58.6-69.7 58.6c-39.1 0-64.4-18.6-76.7-43z"],microsoft:[448,512,[],"f3ca","M0 32h214.6v214.6H0V32zm233.4 0H448v214.6H233.4V32zM0 265.4h214.6V480H0V265.4zm233.4 0H448V480H233.4V265.4z"],qq:[448,512,[],"f1d6","M433.754 420.445c-11.526 1.393-44.86-52.741-44.86-52.741 0 31.345-16.136 72.247-51.051 101.786 16.842 5.192 54.843 19.167 45.803 34.421-7.316 12.343-125.51 7.881-159.632 4.037-34.122 3.844-152.316 8.306-159.632-4.037-9.045-15.25 28.918-29.214 45.783-34.415-34.92-29.539-51.059-70.445-51.059-101.792 0 0-33.334 54.134-44.859 52.741-5.37-.65-12.424-29.644 9.347-99.704 10.261-33.024 21.995-60.478 40.144-105.779C60.683 98.063 108.982.006 224 0c113.737.006 163.156 96.133 160.264 214.963 18.118 45.223 29.912 72.85 40.144 105.778 21.768 70.06 14.716 99.053 9.346 99.704z"],orcid:[512,512,[],"f8d2","M294.75 188.19h-45.92V342h47.47c67.62 0 83.12-51.34 83.12-76.91 0-41.64-26.54-76.9-84.67-76.9zM256 8C119 8 8 119 8 256s111 248 248 248 248-111 248-248S393 8 256 8zm-80.79 360.76h-29.84v-207.5h29.84zm-14.92-231.14a19.57 19.57 0 1 1 19.57-19.57 19.64 19.64 0 0 1-19.57 19.57zM300 369h-81V161.26h80.6c76.73 0 110.44 54.83 110.44 103.85C410 318.39 368.38 369 300 369z"],java:[384,512,[],"f4e4","M277.74 312.9c9.8-6.7 23.4-12.5 23.4-12.5s-38.7 7-77.2 10.2c-47.1 3.9-97.7 4.7-123.1 1.3-60.1-8 33-30.1 33-30.1s-36.1-2.4-80.6 19c-52.5 25.4 130 37 224.5 12.1zm-85.4-32.1c-19-42.7-83.1-80.2 0-145.8C296 53.2 242.84 0 242.84 0c21.5 84.5-75.6 110.1-110.7 162.6-23.9 35.9 11.7 74.4 60.2 118.2zm114.6-176.2c.1 0-175.2 43.8-91.5 140.2 24.7 28.4-6.5 54-6.5 54s62.7-32.4 33.9-72.9c-26.9-37.8-47.5-56.6 64.1-121.3zm-6.1 270.5a12.19 12.19 0 0 1-2 2.6c128.3-33.7 81.1-118.9 19.8-97.3a17.33 17.33 0 0 0-8.2 6.3 70.45 70.45 0 0 1 11-3c31-6.5 75.5 41.5-20.6 91.4zM348 437.4s14.5 11.9-15.9 21.2c-57.9 17.5-240.8 22.8-291.6.7-18.3-7.9 16-19 26.8-21.3 11.2-2.4 17.7-2 17.7-2-20.3-14.3-131.3 28.1-56.4 40.2C232.84 509.4 401 461.3 348 437.4zM124.44 396c-78.7 22 47.9 67.4 148.1 24.5a185.89 185.89 0 0 1-28.2-13.8c-44.7 8.5-65.4 9.1-106 4.5-33.5-3.8-13.9-15.2-13.9-15.2zm179.8 97.2c-78.7 14.8-175.8 13.1-233.3 3.6 0-.1 11.8 9.7 72.4 13.6 92.2 5.9 233.8-3.3 237.1-46.9 0 0-6.4 16.5-76.2 29.7zM260.64 353c-59.2 11.4-93.5 11.1-136.8 6.6-33.5-3.5-11.6-19.7-11.6-19.7-86.8 28.8 48.2 61.4 169.5 25.9a60.37 60.37 0 0 1-21.1-12.8z"],invision:[448,512,[],"f7b0","M407.4 32H40.6C18.2 32 0 50.2 0 72.6v366.8C0 461.8 18.2 480 40.6 480h366.8c22.4 0 40.6-18.2 40.6-40.6V72.6c0-22.4-18.2-40.6-40.6-40.6zM176.1 145.6c.4 23.4-22.4 27.3-26.6 27.4-14.9 0-27.1-12-27.1-27 .1-35.2 53.1-35.5 53.7-.4zM332.8 377c-65.6 0-34.1-74-25-106.6 14.1-46.4-45.2-59-59.9.7l-25.8 103.3H177l8.1-32.5c-31.5 51.8-94.6 44.4-94.6-4.3.1-14.3.9-14 23-104.1H81.7l9.7-35.6h76.4c-33.6 133.7-32.6 126.9-32.9 138.2 0 20.9 40.9 13.5 57.4-23.2l19.8-79.4h-32.3l9.7-35.6h68.8l-8.9 40.5c40.5-75.5 127.9-47.8 101.8 38-14.2 51.1-14.6 50.7-14.9 58.8 0 15.5 17.5 22.6 31.8-16.9L386 325c-10.5 36.7-29.4 52-53.2 52z"],"creative-commons-pd-alt":[496,512,[],"f4ed","M247.6 8C104.7 8 0 123.1 0 256c0 138.5 113.6 248 247.6 248C377.5 504 496 403.1 496 256 496 118.1 389.4 8 247.6 8zm.8 450.8c-112.5 0-203.7-93-203.7-202.8 0-105.4 85.5-203.3 203.7-203.3 112.6 0 202.9 89.5 202.8 203.3 0 121.7-99.6 202.8-202.8 202.8zM316.7 186h-53.2v137.2h53.2c21.4 0 70-5.1 70-68.6 0-63.4-48.6-68.6-70-68.6zm.8 108.5h-19.9v-79.7l19.4-.1c3.8 0 35-2.1 35 39.9 0 24.6-10.5 39.9-34.5 39.9zM203.7 186h-68.2v137.3h34.6V279h27c54.1 0 57.1-37.5 57.1-46.5 0-31-16.8-46.5-50.5-46.5zm-4.9 67.3h-29.2v-41.6h28.3c30.9 0 28.8 41.6.9 41.6z"],centercode:[512,512,[],"f380","M329.2 268.6c-3.8 35.2-35.4 60.6-70.6 56.8-35.2-3.8-60.6-35.4-56.8-70.6 3.8-35.2 35.4-60.6 70.6-56.8 35.1 3.8 60.6 35.4 56.8 70.6zm-85.8 235.1C96.7 496-8.2 365.5 10.1 224.3c11.2-86.6 65.8-156.9 139.1-192 161-77.1 349.7 37.4 354.7 216.6 4.1 147-118.4 262.2-260.5 254.8zm179.9-180c27.9-118-160.5-205.9-237.2-234.2-57.5 56.3-69.1 188.6-33.8 344.4 68.8 15.8 169.1-26.4 271-110.2z"],"glide-g":[448,512,[],"f2a6","M407.1 211.2c-3.5-1.4-11.6-3.8-15.4-3.8-37.1 0-62.2 16.8-93.5 34.5l-.9-.9c7-47.3 23.5-91.9 23.5-140.4C320.8 29.1 282.6 0 212.4 0 97.3 0 39 113.7 39 198.4 39 286.3 90.3 335 177.6 335c12 0 11-1 11 3.8-16.9 128.9-90.8 133.1-90.8 94.6 0-39.2 45-58.6 45.5-61-.3-12.2-47-27.6-58.9-27.6-33.9.1-52.4 51.2-52.4 79.3C32 476 64.8 512 117.5 512c77.4 0 134-77.8 151.4-145.4 15.1-60.5 11.2-63.3 19.7-67.6 32.2-16.2 57.5-27 93.8-27 17.8 0 30.5 3.7 58.9 8.4 2.9 0 6.7-2.9 6.7-5.8 0-8-33.4-60.5-40.9-63.4zm-175.3-84.4c-9.3 44.7-18.6 89.6-27.8 134.3-2.3 10.2-13.3 7.8-22 7.8-38.3 0-49-41.8-49-73.1 0-47 18-109.3 61.8-133.4 7-4.1 14.8-6.7 22.6-6.7 18.6 0 20 13.3 20 28.7-.1 14.3-2.7 28.5-5.6 42.4z"],drupal:[448,512,[],"f1a9","M303.973,108.136C268.2,72.459,234.187,38.35,224.047,0c-9.957,38.35-44.25,72.459-80.019,108.136C90.467,161.7,29.716,222.356,29.716,313.436c-2.337,107.3,82.752,196.18,190.053,198.517S415.948,429.2,418.285,321.9q.091-4.231,0-8.464C418.285,222.356,357.534,161.7,303.973,108.136Zm-174.326,223a130.282,130.282,0,0,0-15.211,24.153,4.978,4.978,0,0,1-3.319,2.766h-1.659c-4.333,0-9.219-8.481-9.219-8.481h0c-1.29-2.028-2.489-4.149-3.687-6.361l-.83-1.752c-11.247-25.72-1.475-62.318-1.475-62.318h0a160.585,160.585,0,0,1,23.231-49.873A290.8,290.8,0,0,1,138.5,201.613l9.219,9.219,43.512,44.434a4.979,4.979,0,0,1,0,6.638L145.78,312.33h0Zm96.612,127.311a67.2,67.2,0,0,1-49.781-111.915c14.2-16.871,31.528-33.464,50.334-55.313,22.309,23.785,36.875,40.1,51.164,57.986a28.413,28.413,0,0,1,2.95,4.425,65.905,65.905,0,0,1,11.984,37.981,66.651,66.651,0,0,1-66.466,66.836ZM352.371,351.6h0a7.743,7.743,0,0,1-6.176,5.347H344.9a11.249,11.249,0,0,1-6.269-5.07h0a348.21,348.21,0,0,0-39.456-48.952L281.387,284.49,222.3,223.185a497.888,497.888,0,0,1-35.4-36.322,12.033,12.033,0,0,0-.922-1.382,35.4,35.4,0,0,1-4.7-9.219V174.51a31.346,31.346,0,0,1,9.218-27.656c11.432-11.431,22.955-22.954,33.833-34.939,11.984,13.275,24.8,26,37.428,38.627h0a530.991,530.991,0,0,1,69.6,79.1,147.494,147.494,0,0,1,27.011,83.8A134.109,134.109,0,0,1,352.371,351.6Z"],jxl:[448,512,[],"e67b","M412.2 32H35.8C16 32 0 48 0 67.8V444.2C0 464 16 480 35.8 480H412.2c19.8 0 35.8-16 35.8-35.8V67.8C448 48 432 32 412.2 32zM378.6 333.7c0 40.2-32.6 72.8-72.8 72.8H70.2c0-40.2 32.6-72.8 72.8-72.8H378.6zm0-113.9c0 40.2-32.6 72.8-72.8 72.8H70.2c0-40.2 32.6-72.8 72.8-72.8H378.6zm0-113.9c0 40.2-32.6 72.8-72.8 72.8H70.2c0-40.2 32.6-72.8 72.8-72.8H378.6z"],"hire-a-helper":[512,512,[],"f3b0","M443.1 0H71.9C67.9 37.3 37.4 67.8 0 71.7v371.5c37.4 4.9 66 32.4 71.9 68.8h372.2c3-36.4 32.5-65.8 67.9-69.8V71.7c-36.4-5.9-65-35.3-68.9-71.7zm-37 404.9c-36.3 0-18.8-2-55.1-2-35.8 0-21 2-56.1 2-5.9 0-4.9-8.2 0-9.8 22.8-7.6 22.9-10.2 24.6-12.8 10.4-15.6 5.9-83 5.9-113 0-5.3-6.4-12.8-13.8-12.8H200.4c-7.4 0-13.8 7.5-13.8 12.8 0 30-4.5 97.4 5.9 113 1.7 2.5 1.8 5.2 24.6 12.8 4.9 1.6 6 9.8 0 9.8-35.1 0-20.3-2-56.1-2-36.3 0-18.8 2-55.1 2-7.9 0-5.8-10.8 0-10.8 10.2-3.4 13.5-3.5 21.7-13.8 7.7-12.9 7.9-44.4 7.9-127.8V151.3c0-22.2-12.2-28.3-28.6-32.4-8.8-2.2-4-11.8 1-11.8 36.5 0 20.6 2 57.1 2 32.7 0 16.5-2 49.2-2 3.3 0 8.5 8.3 1 10.8-4.9 1.6-27.6 3.7-27.6 39.3 0 45.6-.2 55.8 1 68.8 0 1.3 2.3 12.8 12.8 12.8h109.2c10.5 0 12.8-11.5 12.8-12.8 1.2-13 1-23.2 1-68.8 0-35.6-22.7-37.7-27.6-39.3-7.5-2.5-2.3-10.8 1-10.8 32.7 0 16.5 2 49.2 2 36.5 0 20.6-2 57.1-2 4.9 0 9.9 9.6 1 11.8-16.4 4.1-28.6 10.3-28.6 32.4v101.2c0 83.4.1 114.9 7.9 127.8 8.2 10.2 11.4 10.4 21.7 13.8 5.8 0 7.8 10.8 0 10.8z"],"creative-commons-by":[496,512,[],"f4e7","M314.9 194.4v101.4h-28.3v120.5h-77.1V295.9h-28.3V194.4c0-4.4 1.6-8.2 4.6-11.3 3.1-3.1 6.9-4.7 11.3-4.7H299c4.1 0 7.8 1.6 11.1 4.7 3.1 3.2 4.8 6.9 4.8 11.3zm-101.5-63.7c0-23.3 11.5-35 34.5-35s34.5 11.7 34.5 35c0 23-11.5 34.5-34.5 34.5s-34.5-11.5-34.5-34.5zM247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3z"],unity:[448,512,[],"e049","M243.583 91.6027L323.695 138.384C326.575 140.026 326.68 144.583 323.695 146.225L228.503 201.854C225.623 203.55 222.22 203.444 219.549 201.854L124.357 146.225C121.425 144.636 121.373 139.973 124.357 138.384L204.417 91.6027V0L0 119.417V358.252L78.3843 312.477V218.914C78.3319 215.576 82.2066 213.192 85.0865 214.993L180.279 270.622C183.159 272.318 184.782 275.338 184.782 278.464V389.669C184.834 393.007 180.959 395.391 178.079 393.589L97.9673 346.808L19.583 392.583L224 512L428.417 392.583L350.033 346.808L269.921 393.589C267.093 395.338 263.114 393.06 263.218 389.669V278.464C263.218 275.126 265.051 272.159 267.721 270.622L362.914 214.993C365.741 213.245 369.72 215.47 369.616 218.914V312.477L448 358.252V119.417L243.583 0V91.6027Z"],whmcs:[448,512,[],"f40d","M448 161v-21.3l-28.5-8.8-2.2-10.4 20.1-20.7L427 80.4l-29 7.5-7.2-7.5 7.5-28.2-19.1-11.6-21.3 21-10.7-3.2-7-26.4h-22.6l-6.2 26.4-12.1 3.2-19.7-21-19.4 11 8.1 27.7-8.1 8.4-28.5-7.5-11 19.1 20.7 21-2.9 10.4-28.5 7.8-.3 21.7 28.8 7.5 2.4 12.1-20.1 19.9 10.4 18.5 29.6-7.5 7.2 8.6-8.1 26.9 19.9 11.6 19.4-20.4 11.6 2.9 6.7 28.5 22.6.3 6.7-28.8 11.6-3.5 20.7 21.6 20.4-12.1-8.8-28 7.8-8.1 28.8 8.8 10.3-20.1-20.9-18.8 2.2-12.1 29.1-7zm-119.2 45.2c-31.3 0-56.8-25.4-56.8-56.8s25.4-56.8 56.8-56.8 56.8 25.4 56.8 56.8c0 31.5-25.4 56.8-56.8 56.8zm72.3 16.4l46.9 14.5V277l-55.1 13.4-4.1 22.7 38.9 35.3-19.2 37.9-54-16.7-14.6 15.2 16.7 52.5-38.3 22.7-38.9-40.5-21.7 6.6-12.6 54-42.4-.5-12.6-53.6-21.7-5.6-36.4 38.4-37.4-21.7 15.2-50.5-13.7-16.1-55.5 14.1-19.7-34.8 37.9-37.4-4.8-22.8-54-14.1.5-40.9L54 219.9l5.7-19.7-38.9-39.4L41.5 125l53.6 14.1 15.2-15.7-15.2-52 36.4-20.7 36.8 39.4L191 84l11.6-52H245l11.6 45.9L234 72l-6.3-1.7-3.3 5.7-11 19.1-3.3 5.6 4.6 4.6 17.2 17.4-.3 1-23.8 6.5-6.2 1.7-.1 6.4-.2 12.9C153.8 161.6 118 204 118 254.7c0 58.3 47.3 105.7 105.7 105.7 50.5 0 92.7-35.4 103.2-82.8l13.2.2 6.9.1 1.6-6.7 5.6-24 1.9-.6 17.1 17.8 4.7 4.9 5.8-3.4 20.4-12.1 5.8-3.5-2-6.5-6.8-21.2z"],rocketchat:[576,512,[],"f3e8","M284.046,224.8a34.114,34.114,0,1,0,34.317,34.113A34.217,34.217,0,0,0,284.046,224.8Zm-110.45,0a34.114,34.114,0,1,0,34.317,34.113A34.217,34.217,0,0,0,173.6,224.8Zm220.923,0a34.114,34.114,0,1,0,34.317,34.113A34.215,34.215,0,0,0,394.519,224.8Zm153.807-55.319c-15.535-24.172-37.31-45.57-64.681-63.618-52.886-34.817-122.374-54-195.666-54a405.975,405.975,0,0,0-72.032,6.357,238.524,238.524,0,0,0-49.51-36.588C99.684-11.7,40.859.711,11.135,11.421A14.291,14.291,0,0,0,5.58,34.782C26.542,56.458,61.222,99.3,52.7,138.252c-33.142,33.9-51.112,74.776-51.112,117.337,0,43.372,17.97,84.248,51.112,118.148,8.526,38.956-26.154,81.816-47.116,103.491a14.284,14.284,0,0,0,5.555,23.34c29.724,10.709,88.549,23.147,155.324-10.2a238.679,238.679,0,0,0,49.51-36.589A405.972,405.972,0,0,0,288,460.14c73.313,0,142.8-19.159,195.667-53.975,27.371-18.049,49.145-39.426,64.679-63.619,17.309-26.923,26.07-55.916,26.07-86.125C574.394,225.4,565.634,196.43,548.326,169.485ZM284.987,409.9a345.65,345.65,0,0,1-89.446-11.5l-20.129,19.393a184.366,184.366,0,0,1-37.138,27.585,145.767,145.767,0,0,1-52.522,14.87c.983-1.771,1.881-3.563,2.842-5.356q30.258-55.68,16.325-100.078c-32.992-25.962-52.778-59.2-52.778-95.4,0-83.1,104.254-150.469,232.846-150.469s232.867,67.373,232.867,150.469C517.854,342.525,413.6,409.9,284.987,409.9Z"],vk:[448,512,[],"f189","M31.4907 63.4907C0 94.9813 0 145.671 0 247.04V264.96C0 366.329 0 417.019 31.4907 448.509C62.9813 480 113.671 480 215.04 480H232.96C334.329 480 385.019 480 416.509 448.509C448 417.019 448 366.329 448 264.96V247.04C448 145.671 448 94.9813 416.509 63.4907C385.019 32 334.329 32 232.96 32H215.04C113.671 32 62.9813 32 31.4907 63.4907ZM75.6 168.267H126.747C128.427 253.76 166.133 289.973 196 297.44V168.267H244.16V242C273.653 238.827 304.64 205.227 315.093 168.267H363.253C359.313 187.435 351.46 205.583 340.186 221.579C328.913 237.574 314.461 251.071 297.733 261.227C316.41 270.499 332.907 283.63 346.132 299.751C359.357 315.873 369.01 334.618 374.453 354.747H321.44C316.555 337.262 306.614 321.61 292.865 309.754C279.117 297.899 262.173 290.368 244.16 288.107V354.747H238.373C136.267 354.747 78.0267 284.747 75.6 168.267Z"],untappd:[640,512,[],"f405","M401.3 49.9c-79.8 160.1-84.6 152.5-87.9 173.2l-5.2 32.8c-1.9 12-6.6 23.5-13.7 33.4L145.6 497.1c-7.6 10.6-20.4 16.2-33.4 14.6-40.3-5-77.8-32.2-95.3-68.5-5.7-11.8-4.5-25.8 3.1-36.4l148.9-207.9c7.1-9.9 16.4-18 27.2-23.7l29.3-15.5c18.5-9.8 9.7-11.9 135.6-138.9 1-4.8 1-7.3 3.6-8 3-.7 6.6-1 6.3-4.6l-.4-4.6c-.2-1.9 1.3-3.6 3.2-3.6 4.5-.1 13.2 1.2 25.6 10 12.3 8.9 16.4 16.8 17.7 21.1.6 1.8-.6 3.7-2.4 4.2l-4.5 1.1c-3.4.9-2.5 4.4-2.3 7.4.1 2.8-2.3 3.6-6.5 6.1zM230.1 36.4c3.4.9 2.5 4.4 2.3 7.4-.2 2.7 2.1 3.5 6.4 6 7.9 15.9 15.3 30.5 22.2 44 .7 1.3 2.3 1.5 3.3.5 11.2-12 24.6-26.2 40.5-42.6 1.3-1.4 1.4-3.5.1-4.9-8-8.2-16.5-16.9-25.6-26.1-1-4.7-1-7.3-3.6-8-3-.8-6.6-1-6.3-4.6.3-3.3 1.4-8.1-2.8-8.2-4.5-.1-13.2 1.1-25.6 10-12.3 8.9-16.4 16.8-17.7 21.1-1.4 4.2 3.6 4.6 6.8 5.4zM620 406.7L471.2 198.8c-13.2-18.5-26.6-23.4-56.4-39.1-11.2-5.9-14.2-10.9-30.5-28.9-1-1.1-2.9-.9-3.6.5-46.3 88.8-47.1 82.8-49 94.8-1.7 10.7-1.3 20 .3 29.8 1.9 12 6.6 23.5 13.7 33.4l148.9 207.9c7.6 10.6 20.2 16.2 33.1 14.7 40.3-4.9 78-32 95.7-68.6 5.4-11.9 4.3-25.9-3.4-36.6z"],mailchimp:[448,512,[],"f59e","M330.61 243.52a36.15 36.15 0 0 1 9.3 0c1.66-3.83 1.95-10.43.45-17.61-2.23-10.67-5.25-17.14-11.48-16.13s-6.47 8.74-4.24 19.42c1.26 6 3.49 11.14 6 14.32zM277.05 252c4.47 2 7.2 3.26 8.28 2.13 1.89-1.94-3.48-9.39-12.12-13.09a31.44 31.44 0 0 0-30.61 3.68c-3 2.18-5.81 5.22-5.41 7.06.85 3.74 10-2.71 22.6-3.48 7-.44 12.8 1.75 17.26 3.71zm-9 5.13c-9.07 1.42-15 6.53-13.47 10.1.9.34 1.17.81 5.21-.81a37 37 0 0 1 18.72-1.95c2.92.34 4.31.52 4.94-.49 1.46-2.22-5.71-8-15.39-6.85zm54.17 17.1c3.38-6.87-10.9-13.93-14.3-7s10.92 13.88 14.32 6.97zm15.66-20.47c-7.66-.13-7.95 15.8-.26 15.93s7.98-15.81.28-15.96zm-218.79 78.9c-1.32.31-6 1.45-8.47-2.35-5.2-8 11.11-20.38 3-35.77-9.1-17.47-27.82-13.54-35.05-5.54-8.71 9.6-8.72 23.54-5 24.08 4.27.57 4.08-6.47 7.38-11.63a12.83 12.83 0 0 1 17.85-3.72c11.59 7.59 1.37 17.76 2.28 28.62 1.39 16.68 18.42 16.37 21.58 9a2.08 2.08 0 0 0-.2-2.33c.03.89.68-1.3-3.35-.39zm299.72-17.07c-3.35-11.73-2.57-9.22-6.78-20.52 2.45-3.67 15.29-24-3.07-43.25-10.4-10.92-33.9-16.54-41.1-18.54-1.5-11.39 4.65-58.7-21.52-83 20.79-21.55 33.76-45.29 33.73-65.65-.06-39.16-48.15-51-107.42-26.47l-12.55 5.33c-.06-.05-22.71-22.27-23.05-22.57C169.5-18-41.77 216.81 25.78 273.85l14.76 12.51a72.49 72.49 0 0 0-4.1 33.5c3.36 33.4 36 60.42 67.53 60.38 57.73 133.06 267.9 133.28 322.29 3 1.74-4.47 9.11-24.61 9.11-42.38s-10.09-25.27-16.53-25.27zm-316 48.16c-22.82-.61-47.46-21.15-49.91-45.51-6.17-61.31 74.26-75.27 84-12.33 4.54 29.64-4.67 58.49-34.12 57.81zM84.3 249.55C69.14 252.5 55.78 261.09 47.6 273c-4.88-4.07-14-12-15.59-15-13.01-24.85 14.24-73 33.3-100.21C112.42 90.56 186.19 39.68 220.36 48.91c5.55 1.57 23.94 22.89 23.94 22.89s-34.15 18.94-65.8 45.35c-42.66 32.85-74.89 80.59-94.2 132.4zM323.18 350.7s-35.74 5.3-69.51-7.07c6.21-20.16 27 6.1 96.4-13.81 15.29-4.38 35.37-13 51-25.35a102.85 102.85 0 0 1 7.12 24.28c3.66-.66 14.25-.52 11.44 18.1-3.29 19.87-11.73 36-25.93 50.84A106.86 106.86 0 0 1 362.55 421a132.45 132.45 0 0 1-20.34 8.58c-53.51 17.48-108.3-1.74-126-43a66.33 66.33 0 0 1-3.55-9.74c-7.53-27.2-1.14-59.83 18.84-80.37 1.23-1.31 2.48-2.85 2.48-4.79a8.45 8.45 0 0 0-1.92-4.54c-7-10.13-31.19-27.4-26.33-60.83 3.5-24 24.49-40.91 44.07-39.91l5 .29c8.48.5 15.89 1.59 22.88 1.88 11.69.5 22.2-1.19 34.64-11.56 4.2-3.5 7.57-6.54 13.26-7.51a17.45 17.45 0 0 1 13.6 2.24c10 6.64 11.4 22.73 11.92 34.49.29 6.72 1.1 23 1.38 27.63.63 10.67 3.43 12.17 9.11 14 3.19 1.05 6.15 1.83 10.51 3.06 13.21 3.71 21 7.48 26 12.31a16.38 16.38 0 0 1 4.74 9.29c1.56 11.37-8.82 25.4-36.31 38.16-46.71 21.68-93.68 14.45-100.48 13.68-20.15-2.71-31.63 23.32-19.55 41.15 22.64 33.41 122.4 20 151.37-21.35.69-1 .12-1.59-.73-1-41.77 28.58-97.06 38.21-128.46 26-4.77-1.85-14.73-6.44-15.94-16.67 43.6 13.49 71 .74 71 .74s2.03-2.79-.56-2.53zm-68.47-5.7zm-83.4-187.5c16.74-19.35 37.36-36.18 55.83-45.63a.73.73 0 0 1 1 1c-1.46 2.66-4.29 8.34-5.19 12.65a.75.75 0 0 0 1.16.79c11.49-7.83 31.48-16.22 49-17.3a.77.77 0 0 1 .52 1.38 41.86 41.86 0 0 0-7.71 7.74.75.75 0 0 0 .59 1.19c12.31.09 29.66 4.4 41 10.74.76.43.22 1.91-.64 1.72-69.55-15.94-123.08 18.53-134.5 26.83a.76.76 0 0 1-1-1.12z"],"css3-alt":[384,512,[],"f38b","M0 32l34.9 395.8L192 480l157.1-52.2L384 32H0zm313.1 80l-4.8 47.3L193 208.6l-.3.1h111.5l-12.8 146.6-98.2 28.7-98.8-29.2-6.4-73.9h48.9l3.2 38.3 52.6 13.3 54.7-15.4 3.7-61.6-166.3-.5v-.1l-.2.1-3.6-46.3L193.1 162l6.5-2.7H76.7L70.9 112h242.2z"],"square-reddit":[448,512,["reddit-square"],"f1a2","M64 32l320 0c35.3 0 64 28.7 64 64l0 320c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64L0 96C0 60.7 28.7 32 64 32zM305.9 166.4c20.6 0 37.3-16.7 37.3-37.3s-16.7-37.3-37.3-37.3c-18 0-33.1 12.8-36.6 29.8c-30.2 3.2-53.8 28.8-53.8 59.9l0 .2c-32.8 1.4-62.8 10.7-86.6 25.5c-8.8-6.8-19.9-10.9-32-10.9c-28.9 0-52.3 23.4-52.3 52.3c0 21 12.3 39 30.1 47.4c1.7 60.7 67.9 109.6 149.3 109.6s147.6-48.9 149.3-109.7c17.7-8.4 29.9-26.4 29.9-47.3c0-28.9-23.4-52.3-52.3-52.3c-12 0-23 4-31.9 10.8c-24-14.9-54.3-24.2-87.5-25.4l0-.1c0-22.2 16.5-40.7 37.9-43.7l0 0c3.9 16.5 18.7 28.7 36.3 28.7zM155 248.1c14.6 0 25.8 15.4 25 34.4s-11.8 25.9-26.5 25.9s-27.5-7.7-26.6-26.7s13.5-33.5 28.1-33.5zm166.4 33.5c.9 19-12 26.7-26.6 26.7s-25.6-6.9-26.5-25.9c-.9-19 10.3-34.4 25-34.4s27.3 14.6 28.1 33.5zm-42.1 49.6c-9 21.5-30.3 36.7-55.1 36.7s-46.1-15.1-55.1-36.7c-1.1-2.6 .7-5.4 3.4-5.7c16.1-1.6 33.5-2.5 51.7-2.5s35.6 .9 51.7 2.5c2.7 .3 4.5 3.1 3.4 5.7z"],"vimeo-v":[448,512,[],"f27d","M447.8 153.6c-2 43.6-32.4 103.3-91.4 179.1-60.9 79.2-112.4 118.8-154.6 118.8-26.1 0-48.2-24.1-66.3-72.3C100.3 250 85.3 174.3 56.2 174.3c-3.4 0-15.1 7.1-35.2 21.1L0 168.2c51.6-45.3 100.9-95.7 131.8-98.5 34.9-3.4 56.3 20.5 64.4 71.5 28.7 181.5 41.4 208.9 93.6 126.7 18.7-29.6 28.8-52.1 30.2-67.6 4.8-45.9-35.8-42.8-63.3-31 22-72.1 64.1-107.1 126.2-105.1 45.8 1.2 67.5 31.1 64.9 89.4z"],contao:[512,512,[],"f26d","M45.4 305c14.4 67.1 26.4 129 68.2 175H34c-18.7 0-34-15.2-34-34V66c0-18.7 15.2-34 34-34h57.7C77.9 44.6 65.6 59.2 54.8 75.6c-45.4 70-27 146.8-9.4 229.4zM478 32h-90.2c21.4 21.4 39.2 49.5 52.7 84.1l-137.1 29.3c-14.9-29-37.8-53.3-82.6-43.9-24.6 5.3-41 19.3-48.3 34.6-8.8 18.7-13.2 39.8 8.2 140.3 21.1 100.2 33.7 117.7 49.5 131.2 12.9 11.1 33.4 17 58.3 11.7 44.5-9.4 55.7-40.7 57.4-73.2l137.4-29.6c3.2 71.5-18.7 125.2-57.4 163.6H478c18.7 0 34-15.2 34-34V66c0-18.8-15.2-34-34-34z"],"square-font-awesome":[448,512,[],"e5ad","M64 32C28.7 32 0 60.7 0 96L0 416c0 35.3 28.7 64 64 64l320 0c35.3 0 64-28.7 64-64l0-320c0-35.3-28.7-64-64-64L64 32zm90 122c0 9.3-4.8 17.4-12.1 22l188.9 0c7.3 0 13.2 5.9 13.2 13.2c0 1.8-.4 3.7-1.1 5.4L312 264l30.9 69.4c.7 1.7 1.1 3.5 1.1 5.4c0 7.3-5.9 13.2-13.2 13.2L144 352l0 32-32 0 0-32 0-176 0-1.5c-6.1-4.8-10-12.2-10-20.5c0-14.4 11.6-26 26-26s26 11.6 26 26z"],deskpro:[480,512,[],"f38f","M205.9 512l31.1-38.4c12.3-.2 25.6-1.4 36.5-6.6 38.9-18.6 38.4-61.9 38.3-63.8-.1-5-.8-4.4-28.9-37.4H362c-.2 50.1-7.3 68.5-10.2 75.7-9.4 23.7-43.9 62.8-95.2 69.4-8.7 1.1-32.8 1.2-50.7 1.1zm200.4-167.7c38.6 0 58.5-13.6 73.7-30.9l-175.5-.3-17.4 31.3 119.2-.1zm-43.6-223.9v168.3h-73.5l-32.7 55.5H250c-52.3 0-58.1-56.5-58.3-58.9-1.2-13.2-21.3-11.6-20.1 1.8 1.4 15.8 8.8 40 26.4 57.1h-91c-25.5 0-110.8-26.8-107-114V16.9C0 .9 9.7.3 15 .1h82c.2 0 .3.1.5.1 4.3-.4 50.1-2.1 50.1 43.7 0 13.3 20.2 13.4 20.2 0 0-18.2-5.5-32.8-15.8-43.7h84.2c108.7-.4 126.5 79.4 126.5 120.2zm-132.5 56l64 29.3c13.3-45.5-42.2-71.7-64-29.3z"],brave:[448,512,[],"e63c","M145.5 0H224h78.5l44.7 50.8s39.3-10.9 57.8 7.6s33.8 34.9 33.8 34.9l-12 29.5 15.3 43.7s-44.9 170.2-50.1 191c-10.4 40.9-17.4 56.8-46.9 77.5s-82.9 56.8-91.6 62.2c-1.9 1.2-3.9 2.5-5.9 3.9c-7.5 5.1-15.8 10.8-23.5 10.8l0 0 0 0c-7.7 0-16.1-5.7-23.5-10.8c-2-1.4-4-2.8-5.9-3.9c-8.7-5.5-62.1-41.5-91.6-62.2s-36.5-36.6-46.9-77.5c-5.3-20.8-50.1-191-50.1-191l15.3-43.7L9.2 93.3s15.3-16.4 33.8-34.9s57.8-7.6 57.8-7.6L145.5 0zM224 407.6l0 0c3.7 0 8.9-4.7 13-8.4c.6-.5 1.2-1.1 1.7-1.5c4.2-3.7 47.8-37.5 51-39.8s5.4-6.5 1.9-8.7c-2.8-1.7-10-5.5-20.3-10.8c-3-1.6-6.3-3.2-9.7-5c-15.4-8-34.5-14.7-37.5-14.7l0 0 0 0c-3 0-22.1 6.8-37.5 14.7c-3.5 1.8-6.7 3.5-9.7 5c-10.3 5.3-17.6 9.1-20.3 10.8c-3.6 2.2-1.4 6.4 1.9 8.7s46.8 36.1 51 39.8c.5 .5 1.1 1 1.7 1.5c4.1 3.7 9.3 8.4 13 8.4l0 0zm0-165.7l0 0c4.7 0 17.6-3 26.4-5l0 0 2-.5c7.8-1.8 7.3-6.3 6.4-13c-.1-.8-.2-1.6-.3-2.4c-.6-6.1-5.8-33.1-9.1-50.3c-1.1-5.8-2-10.5-2.4-12.9c-1.5-8.1-.6-9.4 .7-11.3c.2-.3 .5-.7 .7-1.1c1.4-2.3 16-6.2 27.9-9.5l0 0c2.5-.7 4.8-1.3 6.9-1.9c10.6-3 32.4-.6 44.2 .6c1.8 .2 3.4 .4 4.7 .5c9.6 .9 10.4 2.3 7.2 3.8c-2.3 1.1-16.2 6.3-28.7 10.9l0 0 0 0c-4.7 1.8-9.2 3.5-12.8 4.8c-1.5 .5-3 1.1-4.5 1.7c-12.5 4.6-27.2 10-28.9 19.4c-1.5 8.3 5.2 19.9 11.3 30.3l0 0c1.6 2.8 3.2 5.5 4.6 8.1c6.3 11.9 6.5 13.3 6.1 18.1c-.4 3.9-14.5 12.7-22.4 17.6l0 0c-1.8 1.1-3.3 2.1-4.2 2.7c-.8 .5-2.1 1.4-3.8 2.4c-8.6 5.2-26.3 16-26.3 22.5c0 7.8 24.6 28.1 32.4 33.2s28.9 16.1 37.9 17.8s23-8.5 31.2-23.8c7.7-14.4 1.7-28.5-3.2-40l-.9-2.2c-4.5-10.6 1.9-17 6.2-21.3l0 0c.5-.5 1-1 1.4-1.4L377.7 194c1.3-1.3 2.5-2.6 3.7-3.8l0 0c5.8-5.7 10.8-10.5 10.8-22.8c0-14.9-57.5-84.5-57.5-84.5s-48.5 9.3-55.1 9.3c-5.2 0-15.3-3.5-25.8-7.1l0 0c-2.7-.9-5.4-1.9-8-2.7C232.8 78.1 224 78 224 78l0 0 0 0s-8.7 0-21.8 4.4c-2.7 .9-5.4 1.8-8 2.7l0 0c-10.5 3.6-20.6 7.1-25.8 7.1c-6.5 0-55.1-9.3-55.1-9.3s-57.5 69.6-57.5 84.5c0 12.3 4.9 17.1 10.8 22.8l0 0c1.2 1.2 2.5 2.4 3.7 3.8l43.1 45.8c.4 .5 .9 .9 1.4 1.4l0 0c4.3 4.3 10.6 10.7 6.2 21.3l-.9 2.2c-4.9 11.5-11 25.6-3.2 40c8.2 15.3 22.2 25.5 31.2 23.8s30.1-12.7 37.9-17.8s32.4-25.4 32.4-33.2c0-6.5-17.7-17.3-26.3-22.5c-1.7-1-3.1-1.9-3.8-2.4c-.9-.6-2.4-1.5-4.2-2.7c-7.9-4.9-22-13.7-22.4-17.6c-.4-4.8-.3-6.2 6.1-18.1c1.3-2.5 2.9-5.3 4.6-8.1c6-10.4 12.8-22 11.3-30.3c-1.7-9.4-16.4-14.8-28.9-19.4c-1.6-.6-3.1-1.1-4.5-1.7c-3.6-1.4-8.1-3.1-12.8-4.8l-.1 0c-12.5-4.7-26.4-9.9-28.7-10.9c-3.2-1.5-2.3-2.8 7.2-3.8c1.3-.1 2.9-.3 4.7-.5c11.8-1.3 33.6-3.6 44.2-.6c2.1 .6 4.4 1.2 6.9 1.9c11.9 3.2 26.5 7.2 27.9 9.5c.2 .4 .5 .7 .7 1.1c1.3 1.9 2.2 3.2 .7 11.3c-.4 2.4-1.3 7.1-2.4 12.9c-3.3 17.2-8.5 44.2-9.1 50.3c-.1 .8-.2 1.7-.3 2.4c-.8 6.7-1.4 11.2 6.4 13l2 .5 0 0c8.8 2 21.8 5 26.4 5l0 0z"],sistrix:[448,512,[],"f3ee","M448 449L301.2 300.2c20-27.9 31.9-62.2 31.9-99.2 0-93.1-74.7-168.9-166.5-168.9C74.7 32 0 107.8 0 200.9s74.7 168.9 166.5 168.9c39.8 0 76.3-14.2 105-37.9l146 148.1 30.5-31zM166.5 330.8c-70.6 0-128.1-58.3-128.1-129.9S95.9 71 166.5 71s128.1 58.3 128.1 129.9-57.4 129.9-128.1 129.9z"],"square-instagram":[448,512,["instagram-square"],"e055","M194.4 211.7a53.3 53.3 0 1 0 59.3 88.7 53.3 53.3 0 1 0 -59.3-88.7zm142.3-68.4c-5.2-5.2-11.5-9.3-18.4-12c-18.1-7.1-57.6-6.8-83.1-6.5c-4.1 0-7.9 .1-11.2 .1c-3.3 0-7.2 0-11.4-.1c-25.5-.3-64.8-.7-82.9 6.5c-6.9 2.7-13.1 6.8-18.4 12s-9.3 11.5-12 18.4c-7.1 18.1-6.7 57.7-6.5 83.2c0 4.1 .1 7.9 .1 11.1s0 7-.1 11.1c-.2 25.5-.6 65.1 6.5 83.2c2.7 6.9 6.8 13.1 12 18.4s11.5 9.3 18.4 12c18.1 7.1 57.6 6.8 83.1 6.5c4.1 0 7.9-.1 11.2-.1c3.3 0 7.2 0 11.4 .1c25.5 .3 64.8 .7 82.9-6.5c6.9-2.7 13.1-6.8 18.4-12s9.3-11.5 12-18.4c7.2-18 6.8-57.4 6.5-83c0-4.2-.1-8.1-.1-11.4s0-7.1 .1-11.4c.3-25.5 .7-64.9-6.5-83l0 0c-2.7-6.9-6.8-13.1-12-18.4zm-67.1 44.5A82 82 0 1 1 178.4 324.2a82 82 0 1 1 91.1-136.4zm29.2-1.3c-3.1-2.1-5.6-5.1-7.1-8.6s-1.8-7.3-1.1-11.1s2.6-7.1 5.2-9.8s6.1-4.5 9.8-5.2s7.6-.4 11.1 1.1s6.5 3.9 8.6 7s3.2 6.8 3.2 10.6c0 2.5-.5 5-1.4 7.3s-2.4 4.4-4.1 6.2s-3.9 3.2-6.2 4.2s-4.8 1.5-7.3 1.5l0 0c-3.8 0-7.5-1.1-10.6-3.2zM448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM357 389c-18.7 18.7-41.4 24.6-67 25.9c-26.4 1.5-105.6 1.5-132 0c-25.6-1.3-48.3-7.2-67-25.9s-24.6-41.4-25.8-67c-1.5-26.4-1.5-105.6 0-132c1.3-25.6 7.1-48.3 25.8-67s41.5-24.6 67-25.8c26.4-1.5 105.6-1.5 132 0c25.6 1.3 48.3 7.1 67 25.8s24.6 41.4 25.8 67c1.5 26.3 1.5 105.4 0 131.9c-1.3 25.6-7.1 48.3-25.8 67z"],"battle-net":[512,512,[],"f835","M448.61 225.62c26.87.18 35.57-7.43 38.92-12.37 12.47-16.32-7.06-47.6-52.85-71.33 17.76-33.58 30.11-63.68 36.34-85.3 3.38-11.83 1.09-19 .45-20.25-1.72 10.52-15.85 48.46-48.2 100.05-25-11.22-56.52-20.1-93.77-23.8-8.94-16.94-34.88-63.86-60.48-88.93C252.18 7.14 238.7 1.07 228.18.22h-.05c-13.83-1.55-22.67 5.85-27.4 11-17.2 18.53-24.33 48.87-25 84.07-7.24-12.35-17.17-24.63-28.5-25.93h-.18c-20.66-3.48-38.39 29.22-36 81.29-38.36 1.38-71 5.75-93 11.23-9.9 2.45-16.22 7.27-17.76 9.72 1-.38 22.4-9.22 111.56-9.22 5.22 53 29.75 101.82 26 93.19-9.73 15.4-38.24 62.36-47.31 97.7-5.87 22.88-4.37 37.61.15 47.14 5.57 12.75 16.41 16.72 23.2 18.26 25 5.71 55.38-3.63 86.7-21.14-7.53 12.84-13.9 28.51-9.06 39.34 7.31 19.65 44.49 18.66 88.44-9.45 20.18 32.18 40.07 57.94 55.7 74.12a39.79 39.79 0 0 0 8.75 7.09c5.14 3.21 8.58 3.37 8.58 3.37-8.24-6.75-34-38-62.54-91.78 22.22-16 45.65-38.87 67.47-69.27 122.82 4.6 143.29-24.76 148-31.64 14.67-19.88 3.43-57.44-57.32-93.69zm-77.85 106.22c23.81-37.71 30.34-67.77 29.45-92.33 27.86 17.57 47.18 37.58 49.06 58.83 1.14 12.93-8.1 29.12-78.51 33.5zM216.9 387.69c9.76-6.23 19.53-13.12 29.2-20.49 6.68 13.33 13.6 26.1 20.6 38.19-40.6 21.86-68.84 12.76-49.8-17.7zm215-171.35c-10.29-5.34-21.16-10.34-32.38-15.05a722.459 722.459 0 0 0 22.74-36.9c39.06 24.1 45.9 53.18 9.64 51.95zM279.18 398c-5.51-11.35-11-23.5-16.5-36.44 43.25 1.27 62.42-18.73 63.28-20.41 0 .07-25 15.64-62.53 12.25a718.78 718.78 0 0 0 85.06-84q13.06-15.31 24.93-31.11c-.36-.29-1.54-3-16.51-12-51.7 60.27-102.34 98-132.75 115.92-20.59-11.18-40.84-31.78-55.71-61.49-20-39.92-30-82.39-31.57-116.07 12.3.91 25.27 2.17 38.85 3.88-22.29 36.8-14.39 63-13.47 64.23 0-.07-.95-29.17 20.14-59.57a695.23 695.23 0 0 0 44.67 152.84c.93-.38 1.84.88 18.67-8.25-26.33-74.47-33.76-138.17-34-173.43 20-12.42 48.18-19.8 81.63-17.81 44.57 2.67 86.36 15.25 116.32 30.71q-10.69 15.66-23.33 32.47C365.63 152 339.1 145.84 337.5 146c.11 0 25.9 14.07 41.52 47.22a717.63 717.63 0 0 0-115.34-31.71 646.608 646.608 0 0 0-39.39-6.05c-.07.45-1.81 1.85-2.16 20.33C300 190.28 358.78 215.68 389.36 233c.74 23.55-6.95 51.61-25.41 79.57-24.6 37.31-56.39 67.23-84.77 85.43zm27.4-287c-44.56-1.66-73.58 7.43-94.69 20.67 2-52.3 21.31-76.38 38.21-75.28C267 52.15 305 108.55 306.58 111zm-130.65 3.1c.48 12.11 1.59 24.62 3.21 37.28-14.55-.85-28.74-1.25-42.4-1.26-.08 3.24-.12-51 24.67-49.59h.09c5.76 1.09 10.63 6.88 14.43 13.57zm-28.06 162c20.76 39.7 43.3 60.57 65.25 72.31-46.79 24.76-77.53 20-84.92 4.51-.2-.21-11.13-15.3 19.67-76.81zm210.06 74.8"],"the-red-yeti":[512,512,[],"f69d","M488.23 241.7l20.7 7.1c-9.6-23.9-23.9-37-31.7-44.8l7.1-18.2c.2 0 12.3-27.8-2.5-30.7-.6-11.3-6.6-27-18.4-27-7.6-10.6-17.7-12.3-30.7-5.9a122.2 122.2 0 0 0-25.3 16.5c-5.3-6.4-3 .4-3-29.8-37.1-24.3-45.4-11.7-74.8 3l.5.5a239.36 239.36 0 0 0-68.4-13.3c-5.5-8.7-18.6-19.1-25.1-25.1l24.8 7.1c-5.5-5.5-26.8-12.9-34.2-15.2 18.2-4.1 29.8-20.8 42.5-33-34.9-10.1-67.9-5.9-97.9 11.8l12-44.2L182 0c-31.6 24.2-33 41.9-33.7 45.5-.9-2.4-6.3-19.6-15.2-27a35.12 35.12 0 0 0-.5 25.3c3 8.4 5.9 14.8 8.4 18.9-16-3.3-28.3-4.9-49.2 0h-3.7l33 14.3a194.26 194.26 0 0 0-46.7 67.4l-1.7 8.4 1.7 1.7 7.6-4.7c-3.3 11.6-5.3 19.4-6.6 25.8a200.18 200.18 0 0 0-27.8 40.3c-15 1-31.8 10.8-40.3 14.3l3 3.4 28.8 1c-.5 1-.7 2.2-1.2 3.2-7.3 6.4-39.8 37.7-33 80.7l20.2-22.4c.5 1.7.7 3.4 1.2 5.2 0 25.5.4 89.6 64.9 150.5 43.6 40 96 60.2 157.5 60.2 121.7 0 223-87.3 223-211.5 6.8-9.7-1.2 3 16.7-25.1l13 14.3 2.5-.5A181.84 181.84 0 0 0 495 255a44.74 44.74 0 0 0-6.8-13.3zM398 111.2l-.5 21.9c5.5 18.1 16.9 17.2 22.4 17.2l-3.4-4.7 22.4-5.4a242.44 242.44 0 0 1-27 0c12.8-2.1 33.3-29 43-11.3 3.4 7.6 6.4 17.2 9.3 27.8l1.7-5.9a56.38 56.38 0 0 1-1.7-15.2c5.4.5 8.8 3.4 9.3 10.1.5 6.4 1.7 14.8 3.4 25.3l4.7-11.3c4.6 0 4.5-3.6-2.5 20.7-20.9-8.7-35.1-8.4-46.5-8.4l18.2-16c-25.3 8.2-33 10.8-54.8 20.9-1.1-5.4-5-13.5-16-19.9-3.2 3.8-2.8.9-.7 14.8h-2.5a62.32 62.32 0 0 0-8.4-23.1l4.2-3.4c8.4-7.1 11.8-14.3 10.6-21.9-.5-6.4-5.4-13.5-13.5-20.7 5.6-3.4 15.2-.4 28.3 8.5zm-39.6-10.1c2.7 1.9 11.4 5.4 18.9 17.2 4.2 8.4 4 9.8 3.4 11.1-.5 2.4-.5 4.3-3 7.1-1.7 2.5-5.4 4.7-11.8 7.6-7.6-13-16.5-23.6-27.8-31.2zM91 143.1l1.2-1.7c1.2-2.9 4.2-7.6 9.3-15.2l2.5-3.4-13 12.3 5.4-4.7-10.1 9.3-4.2 1.2c12.3-24.1 23.1-41.3 32.5-50.2 9.3-9.3 16-16 20.2-19.4l-6.4 1.2c-11.3-4.2-19.4-7.1-24.8-8.4 2.5-.5 3.7-.5 3.2-.5 10.3 0 17.5.5 20.9 1.2a52.35 52.35 0 0 0 16 2.5l.5-1.7-8.4-35.8 13.5 29a42.89 42.89 0 0 0 5.9-14.3c1.7-6.4 5.4-13 10.1-19.4s7.6-10.6 9.3-11.3a234.68 234.68 0 0 0-6.4 25.3l-1.7 7.1-.5 4.7 2.5 2.5C190.4 39.9 214 34 239.8 34.5l21.1.5c-11.8 13.5-27.8 21.9-48.5 24.8a201.26 201.26 0 0 1-23.4 2.9l-.2-.5-2.5-1.2a20.75 20.75 0 0 0-14 2c-2.5-.2-4.9-.5-7.1-.7l-2.5 1.7.5 1.2c2 .2 3.9.5 6.2.7l-2 3.4 3.4-.5-10.6 11.3c-4.2 3-5.4 6.4-4.2 9.3l5.4-3.4h1.2a39.4 39.4 0 0 1 25.3-15.2v-3c6.4.5 13 1 19.4 1.2 6.4 0 8.4.5 5.4 1.2a189.6 189.6 0 0 1 20.7 13.5c13.5 10.1 23.6 21.9 30 35.4 8.8 18.2 13.5 37.1 13.5 56.6a141.13 141.13 0 0 1-3 28.3 209.91 209.91 0 0 1-16 46l2.5.5c18.2-19.7 41.9-16 49.2-16l-6.4 5.9 22.4 17.7-1.7 30.7c-5.4-12.3-16.5-21.1-33-27.8 16.5 14.8 23.6 21.1 21.9 20.2-4.8-2.8-3.5-1.9-10.8-3.7 4.1 4.1 17.5 18.8 18.2 20.7l.2.2-.2.2c0 1.8 1.6-1.2-14 22.9-75.2-15.3-106.27-42.7-141.2-63.2l11.8 1.2c-11.8-18.5-15.6-17.7-38.4-26.1L149 225c-8.8-3-18.2-3-28.3.5l7.6-10.6-1.2-1.7c-14.9 4.3-19.8 9.2-22.6 11.3-1.1-5.5-2.8-12.4-12.3-28.8l-1.2 27-13.2-5c1.5-25.2 5.4-50.5 13.2-74.6zm276.5 330c-49.9 25-56.1 22.4-59 23.9-29.8-11.8-50.9-31.7-63.5-58.8l30 16.5c-9.8-9.3-18.3-16.5-38.4-44.3l11.8 23.1-17.7-7.6c14.2 21.1 23.5 51.7 66.6 73.5-120.8 24.2-199-72.1-200.9-74.3a262.57 262.57 0 0 0 35.4 24.8c3.4 1.7 7.1 2.5 10.1 1.2l-16-20.7c9.2 4.2 9.5 4.5 69.1 29-42.5-20.7-73.8-40.8-93.2-60.2-.5 6.4-1.2 10.1-1.2 10.1a80.25 80.25 0 0 1 20.7 26.6c-39-18.9-57.6-47.6-71.3-82.6 49.9 55.1 118.9 37.5 120.5 37.1 34.8 16.4 69.9 23.6 113.9 10.6 3.3 0 20.3 17 25.3 39.1l4.2-3-2.5-23.6c9 9 24.9 22.6 34.4 13-15.6-5.3-23.5-9.5-29.5-31.7 4.6 4.2 7.6 9 27.8 15l1.2-1.2-10.5-14.2c11.7-4.8-3.5 1 32-10.8 4.3 34.3 9 49.2.7 89.5zm115.3-214.4l-2.5.5 3 9.3c-3.5 5.9-23.7 44.3-71.6 79.7-39.5 29.8-76.6 39.1-80.9 40.3l-7.6-7.1-1.2 3 14.3 16-7.1-4.7 3.4 4.2h-1.2l-21.9-13.5 9.3 26.6-19-27.9-1.2 2.5 7.6 29c-6.1-8.2-21-32.6-56.8-39.6l32.5 21.2a214.82 214.82 0 0 1-93.2-6.4c-4.2-1.2-8.9-2.5-13.5-4.2l1.2-3-44.8-22.4 26.1 22.4c-57.7 9.1-113-25.4-126.4-83.4l-2.5-16.4-22.27 22.3c19.5-57.5 25.6-57.9 51.4-70.1-9.1-5.3-1.6-3.3-38.4-9.3 15.8-5.8 33-15.4 73 5.2a18.5 18.5 0 0 1 3.7-1.7c.6-3.2.4-.8 1-11.8 3.9 10 3.6 8.7 3 9.3l1.7.5c12.7-6.5 8.9-4.5 17-8.9l-5.4 13.5 22.3-5.8-8.4 8.4 2.5 2.5c4.5-1.8 30.3 3.4 40.8 16l-23.6-2.5c39.4 23 51.5 54 55.8 69.6l1.7-1.2c-2.8-22.3-12.4-33.9-16-40.1 4.2 5 39.2 34.6 110.4 46-11.3-.5-23.1 5.4-34.9 18.9l46.7-20.2-9.3 21.9c7.6-10.1 14.8-23.6 21.2-39.6v-.5l1.2-3-1.2 16c13.5-41.8 25.3-78.5 35.4-109.7l13.5-27.8v-2l-5.4-4.2h10.1l5.9 4.2 2.5-1.2-3.4-16 12.3 18.9 41.8-20.2-14.8 13 .5 2.9 17.7-.5a184 184 0 0 1 33 4.2l-23.6 2.5-1.2 3 26.6 23.1a254.21 254.21 0 0 1 27 32c-11.2-3.3-10.3-3.4-21.2-3.4l12.3 32.5zm-6.1-71.3l-3.9 13-14.3-11.8zm-254.8 7.1c1.7 10.6 4.7 17.7 8.8 21.9-9.3 6.6-27.5 13.9-46.5 16l.5 1.2a50.22 50.22 0 0 0 24.8-2.5l-7.1 13c4.2-1.7 10.1-7.1 17.7-14.8 11.9-5.5 12.7-5.1 20.2-16-12.7-6.4-15.7-13.7-18.4-18.8zm3.7-102.3c-6.4-3.4-10.6 3-12.3 18.9s2.5 29.5 11.8 39.6 18.2 10.6 26.1 3 3.4-23.6-11.3-47.7a39.57 39.57 0 0 0-14.27-13.8zm-4.7 46.3c5.4 2.2 10.5 1.9 12.3-10.6v-4.7l-1.2.5c-4.3-3.1-2.5-4.5-1.7-6.2l.5-.5c-.9-1.2-5-8.1-12.5 4.7-.5-13.5.5-21.9 3-24.8 1.2-2.5 4.7-1.2 11.3 4.2 6.4 5.4 11.3 16 15.2 32.5 6.5 28-19.8 26.2-26.9 4.9zm-45-5.5c1.6.3 9.3-1.1 9.3-14.8h-.5c-5.4-1.1-2.2-5.5-.7-5.9-1.7-3-3.4-4.2-5.4-4.7-8.1 0-11.6 12.7-8.1 21.2a7.51 7.51 0 0 0 5.43 4.2zM216 82.9l-2.5.5.5 3a48.94 48.94 0 0 1 26.1 5.9c-2.5-5.5-10-14.3-28.3-14.3l.5 2.5zm-71.8 49.4c21.7 16.8 16.5 21.4 46.5 23.6l-2.9-4.7a42.67 42.67 0 0 0 14.8-28.3c1.7-16-1.2-29.5-8.8-41.3l13-7.6a2.26 2.26 0 0 0-.5-1.7 14.21 14.21 0 0 0-13.5 1.7c-12.7 6.7-28 20.9-29 22.4-1.7 1.7-3.4 5.9-5.4 13.5a99.61 99.61 0 0 0-2.9 23.6c-4.7-8-10.5-6.4-19.9-5.9l7.1 7.6c-16.5 0-23.3 15.4-23.6 16 6.8 0 4.6-7.6 30-12.3-4.3-6.3-3.3-5-4.9-6.6zm18.7-18.7c1.2-7.6 3.4-13 6.4-17.2 5.4-6.4 10.6-10.1 16-11.8 4.2-1.7 7.1 1.2 10.1 9.3a72.14 72.14 0 0 1 3 25.3c-.5 9.3-3.4 17.2-8.4 23.1-2.9 3.4-5.4 5.9-6.4 7.6a39.21 39.21 0 0 1-11.3-.5l-7.1-3.4-5.4-6.4c.8-10 1.3-18.8 3.1-26zm42 56.1c-34.8 14.4-34.7 14-36.1 14.3-20.8 4.7-19-24.4-18.9-24.8l5.9-1.2-.5-2.5c-20.2-2.6-31 4.2-32.5 4.9.5.5 3 3.4 5.9 9.3 4.2-6.4 8.8-10.1 15.2-10.6a83.47 83.47 0 0 0 1.7 33.7c.1.5 2.6 17.4 27.5 24.1 11.3 3 27 1.2 48.9-5.4l-9.2.5c-4.2-14.8-6.4-24.8-5.9-29.5 11.3-8.8 21.9-11.3 30.7-7.6h2.5l-11.8-7.6-7.1.5c-5.9 1.2-12.3 4.2-19.4 8.4z"],"square-hacker-news":[448,512,["hacker-news-square"],"f3af","M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM21 229.2c0 0 .1-.1 .1-.2s.1-.2 .2-.2c0 .1 0 .3-.1 .4H21zM239.2 384H207.8V281.3L128 128h37.3c41.5 77.7 48.1 95.8 54.1 112c1.6 4.3 3.1 8.5 5.2 13.6c3.2-7 5.1-11.9 7.1-17.3c5.9-15.3 12.8-33.2 53.5-108.3H320L239.2 283.1V384z"],edge:[512,512,[],"f282","M120.1 37.44C161.1 12.23 207.7-.7753 255 .0016C423 .0016 512 123.8 512 219.5C511.9 252.2 499 283.4 476.1 306.7C453.2 329.9 422.1 343.2 389.4 343.7C314.2 343.7 297.9 320.6 297.9 311.7C297.9 307.9 299.1 305.5 302.7 302.3L303.7 301.1L304.1 299.5C314.6 288 320 273.3 320 257.9C320 179.2 237.8 115.2 136 115.2C98.46 114.9 61.46 124.1 28.48 142.1C55.48 84.58 111.2 44.5 119.8 38.28C120.6 37.73 120.1 37.44 120.1 37.44V37.44zM135.7 355.5C134.3 385.5 140.3 415.5 152.1 442.7C165.7 469.1 184.8 493.7 208.6 512C149.1 500.5 97.11 468.1 59.2 422.7C21.12 376.3 0 318.4 0 257.9C0 206.7 62.4 163.5 136 163.5C172.6 162.9 208.4 174.4 237.8 196.2L234.2 197.4C182.7 215 135.7 288.1 135.7 355.5V355.5zM469.8 400L469.1 400.1C457.3 418.9 443.2 435.2 426.9 449.6C396.1 477.6 358.8 495.1 318.1 499.5C299.5 499.8 281.3 496.3 264.3 488.1C238.7 477.8 217.2 458.1 202.7 435.1C188.3 411.2 181.6 383.4 183.7 355.5C183.1 335.4 189.1 315.2 198.7 297.3C212.6 330.4 236.2 358.6 266.3 378.1C296.4 397.6 331.8 407.6 367.7 406.7C398.7 407 429.8 400 457.9 386.2L459.8 385.3C463.7 383 467.5 381.4 471.4 385.3C475.9 390.2 473.2 394.5 470.2 399.3C470 399.5 469.9 399.8 469.8 400V400z"],threads:[448,512,[],"e618","M331.5 235.7c2.2 .9 4.2 1.9 6.3 2.8c29.2 14.1 50.6 35.2 61.8 61.4c15.7 36.5 17.2 95.8-30.3 143.2c-36.2 36.2-80.3 52.5-142.6 53h-.3c-70.2-.5-124.1-24.1-160.4-70.2c-32.3-41-48.9-98.1-49.5-169.6V256v-.2C17 184.3 33.6 127.2 65.9 86.2C102.2 40.1 156.2 16.5 226.4 16h.3c70.3 .5 124.9 24 162.3 69.9c18.4 22.7 32 50 40.6 81.7l-40.4 10.8c-7.1-25.8-17.8-47.8-32.2-65.4c-29.2-35.8-73-54.2-130.5-54.6c-57 .5-100.1 18.8-128.2 54.4C72.1 146.1 58.5 194.3 58 256c.5 61.7 14.1 109.9 40.3 143.3c28 35.6 71.2 53.9 128.2 54.4c51.4-.4 85.4-12.6 113.7-40.9c32.3-32.2 31.7-71.8 21.4-95.9c-6.1-14.2-17.1-26-31.9-34.9c-3.7 26.9-11.8 48.3-24.7 64.8c-17.1 21.8-41.4 33.6-72.7 35.3c-23.6 1.3-46.3-4.4-63.9-16c-20.8-13.8-33-34.8-34.3-59.3c-2.5-48.3 35.7-83 95.2-86.4c21.1-1.2 40.9-.3 59.2 2.8c-2.4-14.8-7.3-26.6-14.6-35.2c-10-11.7-25.6-17.7-46.2-17.8H227c-16.6 0-39 4.6-53.3 26.3l-34.4-23.6c19.2-29.1 50.3-45.1 87.8-45.1h.8c62.6 .4 99.9 39.5 103.7 107.7l-.2 .2zm-156 68.8c1.3 25.1 28.4 36.8 54.6 35.3c25.6-1.4 54.6-11.4 59.5-73.2c-13.2-2.9-27.8-4.4-43.4-4.4c-4.8 0-9.6 .1-14.4 .4c-42.9 2.4-57.2 23.2-56.2 41.8l-.1 .1z"],napster:[496,512,[],"f3d2","M298.3 373.6c-14.2 13.6-31.3 24.1-50.4 30.5-19-6.4-36.2-16.9-50.3-30.5h100.7zm44-199.6c20-16.9 43.6-29.2 69.6-36.2V299c0 219.4-328 217.6-328 .3V137.7c25.9 6.9 49.6 19.6 69.5 36.4 56.8-40 132.5-39.9 188.9-.1zm-208.8-58.5c64.4-60 164.3-60.1 228.9-.2-7.1 3.5-13.9 7.3-20.6 11.5-58.7-30.5-129.2-30.4-187.9.1-6.3-4-13.9-8.2-20.4-11.4zM43.8 93.2v69.3c-58.4 36.5-58.4 121.1.1 158.3 26.4 245.1 381.7 240.3 407.6 1.5l.3-1.7c58.7-36.3 58.9-121.7.2-158.2V93.2c-17.3.5-34 3-50.1 7.4-82-91.5-225.5-91.5-307.5.1-16.3-4.4-33.1-7-50.6-7.5zM259.2 352s36-.3 61.3-1.5c10.2-.5 21.1-4 25.5-6.5 26.3-15.1 25.4-39.2 26.2-47.4-79.5-.6-99.9-3.9-113 55.4zm-135.5-55.3c.8 8.2-.1 32.3 26.2 47.4 4.4 2.5 15.2 6 25.5 6.5 25.3 1.1 61.3 1.5 61.3 1.5-13.2-59.4-33.7-56.1-113-55.4zm169.1 123.4c-3.2-5.3-6.9-7.3-6.9-7.3-24.8 7.3-52.2 6.9-75.9 0 0 0-2.9 1.5-6.4 6.6-2.8 4.1-3.7 9.6-3.7 9.6 29.1 17.6 67.1 17.6 96.2 0-.1-.1-.3-4-3.3-8.9z"],"square-snapchat":[448,512,["snapchat-square"],"f2ad","M384,32H64A64,64,0,0,0,0,96V416a64,64,0,0,0,64,64H384a64,64,0,0,0,64-64V96A64,64,0,0,0,384,32Zm-3.907,319.309-.083.1a32.364,32.364,0,0,1-8.717,6.823,90.26,90.26,0,0,1-20.586,8.2,12.694,12.694,0,0,0-3.852,1.76c-2.158,1.909-2.1,4.64-4.4,8.55a23.137,23.137,0,0,1-6.84,7.471c-6.707,4.632-14.244,4.923-22.23,5.23-7.214.274-15.39.581-24.729,3.669-3.761,1.245-7.753,3.694-12.377,6.533-11.265,6.9-26.68,16.353-52.3,16.353s-40.925-9.4-52.106-16.279c-4.657-2.888-8.675-5.362-12.543-6.64-9.339-3.08-17.516-3.4-24.729-3.67-7.986-.307-15.523-.6-22.231-5.229a23.085,23.085,0,0,1-6.01-6.11c-3.2-4.632-2.855-7.8-5.254-9.895a13.428,13.428,0,0,0-4.1-1.834,89.986,89.986,0,0,1-20.313-8.127,32.905,32.905,0,0,1-8.3-6.284c-6.583-6.757-8.276-14.776-5.686-21.824,3.436-9.338,11.571-12.111,19.4-16.262,14.776-8.027,26.348-18.055,34.433-29.884a68.236,68.236,0,0,0,5.985-10.567c.789-2.158.772-3.329.241-4.416a7.386,7.386,0,0,0-2.208-2.217c-2.532-1.676-5.113-3.353-6.882-4.5-3.27-2.141-5.868-3.818-7.529-4.98-6.267-4.383-10.65-9.04-13.4-14.245a28.4,28.4,0,0,1-1.369-23.584c4.134-10.924,14.469-17.706,26.978-17.706a37.141,37.141,0,0,1,7.845.83c.689.15,1.37.307,2.042.482-.108-7.43.058-15.357.722-23.119,2.358-27.261,11.912-41.589,21.874-52.994a86.836,86.836,0,0,1,22.28-17.931C188.254,100.383,205.312,96,224,96s35.828,4.383,50.944,13.016a87.169,87.169,0,0,1,22.239,17.9c9.961,11.406,19.516,25.709,21.874,52.995a231.194,231.194,0,0,1,.713,23.118c.673-.174,1.362-.332,2.051-.481a37.131,37.131,0,0,1,7.844-.83c12.5,0,22.82,6.782,26.971,17.706a28.37,28.37,0,0,1-1.4,23.559c-2.74,5.2-7.123,9.861-13.39,14.244-1.668,1.187-4.258,2.864-7.529,4.981-1.835,1.187-4.541,2.947-7.164,4.682a6.856,6.856,0,0,0-1.951,2.034c-.506,1.046-.539,2.191.166,4.208a69.015,69.015,0,0,0,6.085,10.792c8.268,12.1,20.188,22.313,35.454,30.407,1.486.772,2.98,1.5,4.441,2.258.722.332,1.569.763,2.491,1.3,4.9,2.723,9.2,6.01,11.455,12.153C387.821,336.915,386.269,344.7,380.093,351.309Zm-16.719-18.461c-50.313-24.314-58.332-61.918-58.689-64.749-.431-3.379-.921-6.035,2.806-9.472,3.594-3.328,19.541-13.19,23.965-16.278,7.33-5.114,10.534-10.219,8.16-16.495-1.66-4.316-5.686-5.976-9.961-5.976a18.5,18.5,0,0,0-3.993.448c-8.035,1.743-15.838,5.769-20.354,6.857a7.1,7.1,0,0,1-1.66.224c-2.408,0-3.279-1.071-3.088-3.968.564-8.783,1.759-25.925.373-41.937-1.884-22.032-8.99-32.948-17.432-42.6-4.051-4.624-23.135-24.654-59.536-24.654S168.53,134.359,164.479,139c-8.434,9.654-15.531,20.57-17.432,42.6-1.386,16.013-.141,33.147.373,41.937.166,2.756-.68,3.968-3.088,3.968a7.1,7.1,0,0,1-1.66-.224c-4.507-1.087-12.31-5.113-20.346-6.856a18.494,18.494,0,0,0-3.993-.449c-4.25,0-8.3,1.636-9.961,5.977-2.374,6.276.847,11.381,8.168,16.494,4.425,3.088,20.371,12.958,23.966,16.279,3.719,3.437,3.237,6.093,2.805,9.471-.356,2.79-8.384,40.394-58.689,64.749-2.946,1.428-7.96,4.45.88,9.331,13.88,7.628,23.111,6.807,30.3,11.43,6.093,3.927,2.5,12.394,6.923,15.449,5.454,3.76,21.583-.266,42.335,6.6,17.433,5.744,28.116,22.015,58.963,22.015s41.788-16.3,58.938-21.973c20.795-6.865,36.89-2.839,42.336-6.6,4.433-3.055.822-11.522,6.923-15.448,7.181-4.624,16.411-3.8,30.3-11.472C371.36,337.355,366.346,334.333,363.374,332.848Z"],"google-plus-g":[640,512,[],"f0d5","M386.061 228.496c1.834 9.692 3.143 19.384 3.143 31.956C389.204 370.205 315.599 448 204.8 448c-106.084 0-192-85.915-192-192s85.916-192 192-192c51.864 0 95.083 18.859 128.611 50.292l-52.126 50.03c-14.145-13.621-39.028-29.599-76.485-29.599-65.484 0-118.92 54.221-118.92 121.277 0 67.056 53.436 121.277 118.92 121.277 75.961 0 104.513-54.745 108.965-82.773H204.8v-66.009h181.261zm185.406 6.437V179.2h-56.001v55.733h-55.733v56.001h55.733v55.733h56.001v-55.733H627.2v-56.001h-55.733z"],artstation:[512,512,[],"f77a","M2 377.4l43 74.3A51.35 51.35 0 0 0 90.9 480h285.4l-59.2-102.6zM501.8 350L335.6 59.3A51.38 51.38 0 0 0 290.2 32h-88.4l257.3 447.6 40.7-70.5c1.9-3.2 21-29.7 2-59.1zM275 304.5l-115.5-200L44 304.5z"],markdown:[640,512,[],"f60f","M593.8 59.1H46.2C20.7 59.1 0 79.8 0 105.2v301.5c0 25.5 20.7 46.2 46.2 46.2h547.7c25.5 0 46.2-20.7 46.1-46.1V105.2c0-25.4-20.7-46.1-46.2-46.1zM338.5 360.6H277v-120l-61.5 76.9-61.5-76.9v120H92.3V151.4h61.5l61.5 76.9 61.5-76.9h61.5v209.2zm135.3 3.1L381.5 256H443V151.4h61.5V256H566z"],sourcetree:[448,512,[],"f7d3","M427.2 203c0-112.1-90.9-203-203-203C112.1-.2 21.2 90.6 21 202.6A202.86 202.86 0 0 0 161.5 396v101.7a14.3 14.3 0 0 0 14.3 14.3h96.4a14.3 14.3 0 0 0 14.3-14.3V396.1A203.18 203.18 0 0 0 427.2 203zm-271.6 0c0-90.8 137.3-90.8 137.3 0-.1 89.9-137.3 91-137.3 0z"],"google-plus":[512,512,[],"f2b3","M256,8C119.1,8,8,119.1,8,256S119.1,504,256,504,504,392.9,504,256,392.9,8,256,8ZM185.3,380a124,124,0,0,1,0-248c31.3,0,60.1,11,83,32.3l-33.6,32.6c-13.2-12.9-31.3-19.1-49.4-19.1-42.9,0-77.2,35.5-77.2,78.1S142.3,334,185.3,334c32.6,0,64.9-19.1,70.1-53.3H185.3V238.1H302.2a109.2,109.2,0,0,1,1.9,20.7c0,70.8-47.5,121.2-118.8,121.2ZM415.5,273.8v35.5H380V273.8H344.5V238.3H380V202.8h35.5v35.5h35.2v35.5Z"],diaspora:[512,512,[],"f791","M251.64 354.55c-1.4 0-88 119.9-88.7 119.9S76.34 414 76 413.25s86.6-125.7 86.6-127.4c0-2.2-129.6-44-137.6-47.1-1.3-.5 31.4-101.8 31.7-102.1.6-.7 144.4 47 145.5 47 .4 0 .9-.6 1-1.3.4-2 1-148.6 1.7-149.6.8-1.2 104.5-.7 105.1-.3 1.5 1 3.5 156.1 6.1 156.1 1.4 0 138.7-47 139.3-46.3.8.9 31.9 102.2 31.5 102.6-.9.9-140.2 47.1-140.6 48.8-.3 1.4 82.8 122.1 82.5 122.9s-85.5 63.5-86.3 63.5c-1-.2-89-125.5-90.9-125.5z"],foursquare:[368,512,[],"f180","M323.1 3H49.9C12.4 3 0 31.3 0 49.1v433.8c0 20.3 12.1 27.7 18.2 30.1 6.2 2.5 22.8 4.6 32.9-7.1C180 356.5 182.2 354 182.2 354c3.1-3.4 3.4-3.1 6.8-3.1h83.4c35.1 0 40.6-25.2 44.3-39.7l48.6-243C373.8 25.8 363.1 3 323.1 3zm-16.3 73.8l-11.4 59.7c-1.2 6.5-9.5 13.2-16.9 13.2H172.1c-12 0-20.6 8.3-20.6 20.3v13c0 12 8.6 20.6 20.6 20.6h90.4c8.3 0 16.6 9.2 14.8 18.2-1.8 8.9-10.5 53.8-11.4 58.8-.9 4.9-6.8 13.5-16.9 13.5h-73.5c-13.5 0-17.2 1.8-26.5 12.6 0 0-8.9 11.4-89.5 108.3-.9.9-1.8.6-1.8-.3V75.9c0-7.7 6.8-16.6 16.6-16.6h219c8.2 0 15.6 7.7 13.5 17.5z"],"stack-overflow":[384,512,[],"f16c","M290.7 311L95 269.7 86.8 309l195.7 41zm51-87L188.2 95.7l-25.5 30.8 153.5 128.3zm-31.2 39.7L129.2 179l-16.7 36.5L293.7 300zM262 32l-32 24 119.3 160.3 32-24zm20.5 328h-200v39.7h200zm39.7 80H42.7V320h-40v160h359.5V320h-40z"],"github-alt":[480,512,[],"f113","M186.1 328.7c0 20.9-10.9 55.1-36.7 55.1s-36.7-34.2-36.7-55.1 10.9-55.1 36.7-55.1 36.7 34.2 36.7 55.1zM480 278.2c0 31.9-3.2 65.7-17.5 95-37.9 76.6-142.1 74.8-216.7 74.8-75.8 0-186.2 2.7-225.6-74.8-14.6-29-20.2-63.1-20.2-95 0-41.9 13.9-81.5 41.5-113.6-5.2-15.8-7.7-32.4-7.7-48.8 0-21.5 4.9-32.3 14.6-51.8 45.3 0 74.3 9 108.8 36 29-6.9 58.8-10 88.7-10 27 0 54.2 2.9 80.4 9.2 34-26.7 63-35.2 107.8-35.2 9.8 19.5 14.6 30.3 14.6 51.8 0 16.4-2.6 32.7-7.7 48.2 27.5 32.4 39 72.3 39 114.2zm-64.3 50.5c0-43.9-26.7-82.6-73.5-82.6-18.9 0-37 3.4-56 6-14.9 2.3-29.8 3.2-45.1 3.2-15.2 0-30.1-.9-45.1-3.2-18.7-2.6-37-6-56-6-46.8 0-73.5 38.7-73.5 82.6 0 87.8 80.4 101.3 150.4 101.3h48.2c70.3 0 150.6-13.4 150.6-101.3zm-82.6-55.1c-25.8 0-36.7 34.2-36.7 55.1s10.9 55.1 36.7 55.1 36.7-34.2 36.7-55.1-10.9-55.1-36.7-55.1z"],"phoenix-squadron":[512,512,[],"f511","M96 63.38C142.49 27.25 201.55 7.31 260.51 8.81c29.58-.38 59.11 5.37 86.91 15.33-24.13-4.63-49-6.34-73.38-2.45C231.17 27 191 48.84 162.21 80.87c5.67-1 10.78-3.67 16-5.86 18.14-7.87 37.49-13.26 57.23-14.83 19.74-2.13 39.64-.43 59.28 1.92-14.42 2.79-29.12 4.57-43 9.59-34.43 11.07-65.27 33.16-86.3 62.63-13.8 19.71-23.63 42.86-24.67 67.13-.35 16.49 5.22 34.81 19.83 44a53.27 53.27 0 0 0 37.52 6.74c15.45-2.46 30.07-8.64 43.6-16.33 11.52-6.82 22.67-14.55 32-24.25 3.79-3.22 2.53-8.45 2.62-12.79-2.12-.34-4.38-1.11-6.3.3a203 203 0 0 1-35.82 15.37c-20 6.17-42.16 8.46-62.1.78 12.79 1.73 26.06.31 37.74-5.44 20.23-9.72 36.81-25.2 54.44-38.77a526.57 526.57 0 0 1 88.9-55.31c25.71-12 52.94-22.78 81.57-24.12-15.63 13.72-32.15 26.52-46.78 41.38-14.51 14-27.46 29.5-40.11 45.18-3.52 4.6-8.95 6.94-13.58 10.16a150.7 150.7 0 0 0-51.89 60.1c-9.33 19.68-14.5 41.85-11.77 63.65 1.94 13.69 8.71 27.59 20.9 34.91 12.9 8 29.05 8.07 43.48 5.1 32.8-7.45 61.43-28.89 81-55.84 20.44-27.52 30.52-62.2 29.16-96.35-.52-7.5-1.57-15-1.66-22.49 8 19.48 14.82 39.71 16.65 60.83 2 14.28.75 28.76-1.62 42.9-1.91 11-5.67 21.51-7.78 32.43a165 165 0 0 0 39.34-81.07 183.64 183.64 0 0 0-14.21-104.64c20.78 32 32.34 69.58 35.71 107.48.49 12.73.49 25.51 0 38.23A243.21 243.21 0 0 1 482 371.34c-26.12 47.34-68 85.63-117.19 108-78.29 36.23-174.68 31.32-248-14.68A248.34 248.34 0 0 1 25.36 366 238.34 238.34 0 0 1 0 273.08v-31.34C3.93 172 40.87 105.82 96 63.38m222 80.33a79.13 79.13 0 0 0 16-4.48c5-1.77 9.24-5.94 10.32-11.22-8.96 4.99-17.98 9.92-26.32 15.7z"],pagelines:[384,512,[],"f18c","M384 312.7c-55.1 136.7-187.1 54-187.1 54-40.5 81.8-107.4 134.4-184.6 134.7-16.1 0-16.6-24.4 0-24.4 64.4-.3 120.5-42.7 157.2-110.1-41.1 15.9-118.6 27.9-161.6-82.2 109-44.9 159.1 11.2 178.3 45.5 9.9-24.4 17-50.9 21.6-79.7 0 0-139.7 21.9-149.5-98.1 119.1-47.9 152.6 76.7 152.6 76.7 1.6-16.7 3.3-52.6 3.3-53.4 0 0-106.3-73.7-38.1-165.2 124.6 43 61.4 162.4 61.4 162.4.5 1.6.5 23.8 0 33.4 0 0 45.2-89 136.4-57.5-4.2 134-141.9 106.4-141.9 106.4-4.4 27.4-11.2 53.4-20 77.5 0 0 83-91.8 172-20z"],algolia:[512,512,[],"f36c","M256 0C116.1 0 2 112.7 0 252.1C-2 393.6 112.9 510.8 254.5 511.6c43.7 .3 85.9-10.4 123.3-30.7c3.6-2 4.2-7 1.1-9.7l-24-21.2c-4.9-4.3-11.8-5.5-17.8-3c-26.1 11.1-54.5 16.8-83.7 16.4C139 461.9 46.5 366.8 48.3 252.4C50.1 139.5 142.6 48.2 256 48.2H463.7V417.2L345.9 312.5c-3.8-3.4-9.7-2.7-12.7 1.3c-18.9 25-49.7 40.6-83.9 38.2c-47.5-3.3-85.9-41.5-89.5-88.9c-4.2-56.6 40.6-103.9 96.3-103.9c50.4 0 91.9 38.8 96.2 88c.4 4.4 2.4 8.5 5.7 11.4l30.7 27.2c3.5 3.1 9 1.2 9.9-3.4c2.2-11.8 3-24.2 2.1-36.8c-4.9-72-63.3-130-135.4-134.4c-82.7-5.1-151.8 59.5-154 140.6c-2.1 78.9 62.6 147 141.6 148.7c33 .7 63.6-9.6 88.3-27.6L495 509.4c6.6 5.8 17 1.2 17-7.7V9.7c0-5.4-4.4-9.7-9.7-9.7H256z"],"red-river":[448,512,[],"f3e3","M353.2 32H94.8C42.4 32 0 74.4 0 126.8v258.4C0 437.6 42.4 480 94.8 480h258.4c52.4 0 94.8-42.4 94.8-94.8V126.8c0-52.4-42.4-94.8-94.8-94.8zM144.9 200.9v56.3c0 27-21.9 48.9-48.9 48.9V151.9c0-13.2 10.7-23.9 23.9-23.9h154.2c0 27-21.9 48.9-48.9 48.9h-56.3c-12.3-.6-24.6 11.6-24 24zm176.3 72h-56.3c-12.3-.6-24.6 11.6-24 24v56.3c0 27-21.9 48.9-48.9 48.9V247.9c0-13.2 10.7-23.9 23.9-23.9h154.2c0 27-21.9 48.9-48.9 48.9z"],"creative-commons-sa":[496,512,[],"f4ef","M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zM137.7 221c13-83.9 80.5-95.7 108.9-95.7 99.8 0 127.5 82.5 127.5 134.2 0 63.6-41 132.9-128.9 132.9-38.9 0-99.1-20-109.4-97h62.5c1.5 30.1 19.6 45.2 54.5 45.2 23.3 0 58-18.2 58-82.8 0-82.5-49.1-80.6-56.7-80.6-33.1 0-51.7 14.6-55.8 43.8h18.2l-49.2 49.2-49-49.2h19.4z"],safari:[512,512,[],"f267","M274.69,274.69l-37.38-37.38L166,346ZM256,8C119,8,8,119,8,256S119,504,256,504,504,393,504,256,393,8,256,8ZM411.85,182.79l14.78-6.13A8,8,0,0,1,437.08,181h0a8,8,0,0,1-4.33,10.46L418,197.57a8,8,0,0,1-10.45-4.33h0A8,8,0,0,1,411.85,182.79ZM314.43,94l6.12-14.78A8,8,0,0,1,331,74.92h0a8,8,0,0,1,4.33,10.45l-6.13,14.78a8,8,0,0,1-10.45,4.33h0A8,8,0,0,1,314.43,94ZM256,60h0a8,8,0,0,1,8,8V84a8,8,0,0,1-8,8h0a8,8,0,0,1-8-8V68A8,8,0,0,1,256,60ZM181,74.92a8,8,0,0,1,10.46,4.33L197.57,94a8,8,0,1,1-14.78,6.12l-6.13-14.78A8,8,0,0,1,181,74.92Zm-63.58,42.49h0a8,8,0,0,1,11.31,0L140,128.72A8,8,0,0,1,140,140h0a8,8,0,0,1-11.31,0l-11.31-11.31A8,8,0,0,1,117.41,117.41ZM60,256h0a8,8,0,0,1,8-8H84a8,8,0,0,1,8,8h0a8,8,0,0,1-8,8H68A8,8,0,0,1,60,256Zm40.15,73.21-14.78,6.13A8,8,0,0,1,74.92,331h0a8,8,0,0,1,4.33-10.46L94,314.43a8,8,0,0,1,10.45,4.33h0A8,8,0,0,1,100.15,329.21Zm4.33-136h0A8,8,0,0,1,94,197.57l-14.78-6.12A8,8,0,0,1,74.92,181h0a8,8,0,0,1,10.45-4.33l14.78,6.13A8,8,0,0,1,104.48,193.24ZM197.57,418l-6.12,14.78a8,8,0,0,1-14.79-6.12l6.13-14.78A8,8,0,1,1,197.57,418ZM264,444a8,8,0,0,1-8,8h0a8,8,0,0,1-8-8V428a8,8,0,0,1,8-8h0a8,8,0,0,1,8,8Zm67-6.92h0a8,8,0,0,1-10.46-4.33L314.43,418a8,8,0,0,1,4.33-10.45h0a8,8,0,0,1,10.45,4.33l6.13,14.78A8,8,0,0,1,331,437.08Zm63.58-42.49h0a8,8,0,0,1-11.31,0L372,383.28A8,8,0,0,1,372,372h0a8,8,0,0,1,11.31,0l11.31,11.31A8,8,0,0,1,394.59,394.59ZM286.25,286.25,110.34,401.66,225.75,225.75,401.66,110.34ZM437.08,331h0a8,8,0,0,1-10.45,4.33l-14.78-6.13a8,8,0,0,1-4.33-10.45h0A8,8,0,0,1,418,314.43l14.78,6.12A8,8,0,0,1,437.08,331ZM444,264H428a8,8,0,0,1-8-8h0a8,8,0,0,1,8-8h16a8,8,0,0,1,8,8h0A8,8,0,0,1,444,264Z"],google:[488,512,[],"f1a0","M488 261.8C488 403.3 391.1 504 248 504 110.8 504 0 393.2 0 256S110.8 8 248 8c66.8 0 123 24.5 166.3 64.9l-67.5 64.9C258.5 52.6 94.3 116.6 94.3 256c0 86.5 69.1 156.6 153.7 156.6 98.2 0 135-70.4 140.8-106.9H248v-85.3h236.1c2.3 12.7 3.9 24.9 3.9 41.4z"],"square-font-awesome-stroke":[448,512,["font-awesome-alt"],"f35c","M64 64C46.3 64 32 78.3 32 96l0 320c0 17.7 14.3 32 32 32l320 0c17.7 0 32-14.3 32-32l0-320c0-17.7-14.3-32-32-32L64 64zM0 96C0 60.7 28.7 32 64 32l320 0c35.3 0 64 28.7 64 64l0 320c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64L0 96zm154 58c0 9.3-4.8 17.4-12.1 22l188.9 0c7.3 0 13.2 5.9 13.2 13.2c0 1.8-.4 3.7-1.1 5.4L312 264l30.9 69.4c.8 1.7 1.1 3.5 1.1 5.4c0 7.3-5.9 13.2-13.2 13.2L144 352l0 32-32 0 0-32 0-176 0-1.5c-6.1-4.8-10-12.2-10-20.5c0-14.4 11.6-26 26-26s26 11.6 26 26z"],atlassian:[512,512,[],"f77b","M152.2 236.4c-7.7-8.2-19.7-7.7-24.8 2.8L1.6 490.2c-5 10 2.4 21.7 13.4 21.7h175c5.8.1 11-3.2 13.4-8.4 37.9-77.8 15.1-196.3-51.2-267.1zM244.4 8.1c-122.3 193.4-8.5 348.6 65 495.5 2.5 5.1 7.7 8.4 13.4 8.4H497c11.2 0 18.4-11.8 13.4-21.7 0 0-234.5-470.6-240.4-482.3-5.3-10.6-18.8-10.8-25.6.1z"],"linkedin-in":[448,512,[],"f0e1","M100.28 448H7.4V148.9h92.88zM53.79 108.1C24.09 108.1 0 83.5 0 53.8a53.79 53.79 0 0 1 107.58 0c0 29.7-24.1 54.3-53.79 54.3zM447.9 448h-92.68V302.4c0-34.7-.7-79.2-48.29-79.2-48.29 0-55.69 37.7-55.69 76.7V448h-92.78V148.9h89.08v40.8h1.3c12.4-23.5 42.69-48.3 87.88-48.3 94 0 111.28 61.9 111.28 142.3V448z"],"digital-ocean":[512,512,[],"f391","M87 481.8h73.7v-73.6H87zM25.4 346.6v61.6H87v-61.6zm466.2-169.7c-23-74.2-82.4-133.3-156.6-156.6C164.9-32.8 8 93.7 8 255.9h95.8c0-101.8 101-180.5 208.1-141.7 39.7 14.3 71.5 46.1 85.8 85.7 39.1 107-39.7 207.8-141.4 208v.3h-.3V504c162.6 0 288.8-156.8 235.6-327.1zm-235.3 231v-95.3h-95.6v95.6H256v-.3z"],nimblr:[384,512,[],"f5a8","M246.6 299.29c15.57 0 27.15 11.46 27.15 27s-11.62 27-27.15 27c-15.7 0-27.15-11.57-27.15-27s11.55-27 27.15-27zM113 326.25c0-15.61 11.68-27 27.15-27s27.15 11.46 27.15 27-11.47 27-27.15 27c-15.44 0-27.15-11.31-27.15-27M191.76 159C157 159 89.45 178.77 59.25 227L14 0v335.48C14 433.13 93.61 512 191.76 512s177.76-78.95 177.76-176.52S290.13 159 191.76 159zm0 308.12c-73.27 0-132.51-58.9-132.51-131.59s59.24-131.59 132.51-131.59 132.51 58.86 132.51 131.54S265 467.07 191.76 467.07z"],chromecast:[512,512,[],"f838","M447.8,64H64c-23.6,0-42.7,19.1-42.7,42.7v63.9H64v-63.9h383.8v298.6H298.6V448H448c23.6,0,42.7-19.1,42.7-42.7V106.7 C490.7,83.1,471.4,64,447.8,64z M21.3,383.6L21.3,383.6l0,63.9h63.9C85.2,412.2,56.6,383.6,21.3,383.6L21.3,383.6z M21.3,298.6V341 c58.9,0,106.6,48.1,106.6,107h42.7C170.7,365.6,103.7,298.7,21.3,298.6z M213.4,448h42.7c-0.5-129.5-105.3-234.3-234.8-234.6l0,42.4 C127.3,255.6,213.3,342,213.4,448z"],evernote:[384,512,[],"f839","M120.82 132.21c1.6 22.31-17.55 21.59-21.61 21.59-68.93 0-73.64-1-83.58 3.34-.56.22-.74 0-.37-.37L123.79 46.45c.38-.37.6-.22.38.37-4.35 9.99-3.35 15.09-3.35 85.39zm79 308c-14.68-37.08 13-76.93 52.52-76.62 17.49 0 22.6 23.21 7.95 31.42-6.19 3.3-24.95 1.74-25.14 19.2-.05 17.09 19.67 25 31.2 24.89A45.64 45.64 0 0 0 312 393.45v-.08c0-11.63-7.79-47.22-47.54-55.34-7.72-1.54-65-6.35-68.35-50.52-3.74 16.93-17.4 63.49-43.11 69.09-8.74 1.94-69.68 7.64-112.92-36.77 0 0-18.57-15.23-28.23-57.95-3.38-15.75-9.28-39.7-11.14-62 0-18 11.14-30.45 25.07-32.2 81 0 90 2.32 101-7.8 9.82-9.24 7.8-15.5 7.8-102.78 1-8.3 7.79-30.81 53.41-24.14 6 .86 31.91 4.18 37.48 30.64l64.26 11.15c20.43 3.71 70.94 7 80.6 57.94 22.66 121.09 8.91 238.46 7.8 238.46C362.15 485.53 267.06 480 267.06 480c-18.95-.23-54.25-9.4-67.27-39.83zm80.94-204.84c-1 1.92-2.2 6 .85 7 14.09 4.93 39.75 6.84 45.88 5.53 3.11-.25 3.05-4.43 2.48-6.65-3.53-21.85-40.83-26.5-49.24-5.92z"],"hacker-news":[448,512,[],"f1d4","M0 32v448h448V32H0zm21.2 197.2H21c.1-.1.2-.3.3-.4 0 .1 0 .3-.1.4zm218 53.9V384h-31.4V281.3L128 128h37.3c52.5 98.3 49.2 101.2 59.3 125.6 12.3-27 5.8-24.4 60.6-125.6H320l-80.8 155.1z"],"creative-commons-sampling":[496,512,[],"f4f0","M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zm3.6 53.2c2.8-.3 11.5 1 11.5 11.5l6.6 107.2 4.9-59.3c0-6 4.7-10.6 10.6-10.6 5.9 0 10.6 4.7 10.6 10.6 0 2.5-.5-5.7 5.7 81.5l5.8-64.2c.3-2.9 2.9-9.3 10.2-9.3 3.8 0 9.9 2.3 10.6 8.9l11.5 96.5 5.3-12.8c1.8-4.4 5.2-6.6 10.2-6.6h58v21.3h-50.9l-18.2 44.3c-3.9 9.9-19.5 9.1-20.8-3.1l-4-31.9-7.5 92.6c-.3 3-3 9.3-10.2 9.3-3 0-9.8-2.1-10.6-9.3 0-1.9.6 5.8-6.2-77.9l-5.3 72.2c-1.1 4.8-4.8 9.3-10.6 9.3-2.9 0-9.8-2-10.6-9.3 0-1.9.5 6.7-5.8-87.7l-5.8 94.8c0 6.3-3.6 12.4-10.6 12.4-5.2 0-10.6-4.1-10.6-12l-5.8-87.7c-5.8 92.5-5.3 84-5.3 85.9-1.1 4.8-4.8 9.3-10.6 9.3-3 0-9.8-2.1-10.6-9.3 0-.7-.4-1.1-.4-2.6l-6.2-88.6L182 348c-.7 6.5-6.7 9.3-10.6 9.3-5.8 0-9.6-4.1-10.6-8.9L149.7 272c-2 4-3.5 8.4-11.1 8.4H87.2v-21.3H132l13.7-27.9c4.4-9.9 18.2-7.2 19.9 2.7l3.1 20.4 8.4-97.9c0-6 4.8-10.6 10.6-10.6.5 0 10.6-.2 10.6 12.4l4.9 69.1 6.6-92.6c0-10.1 9.5-10.6 10.2-10.6.6 0 10.6.7 10.6 10.6l5.3 80.6 6.2-97.9c.1-1.1-.6-10.3 9.9-11.5z"],adversal:[512,512,[],"f36a","M482.1 32H28.7C5.8 32 0 37.9 0 60.9v390.2C0 474.4 5.8 480 28.7 480h453.4c24.4 0 29.9-5.2 29.9-29.7V62.2c0-24.6-5.4-30.2-29.9-30.2zM178.4 220.3c-27.5-20.2-72.1-8.7-84.2 23.4-4.3 11.1-9.3 9.5-17.5 8.3-9.7-1.5-17.2-3.2-22.5-5.5-28.8-11.4 8.6-55.3 24.9-64.3 41.1-21.4 83.4-22.2 125.3-4.8 40.9 16.8 34.5 59.2 34.5 128.5 2.7 25.8-4.3 58.3 9.3 88.8 1.9 4.4.4 7.9-2.7 10.7-8.4 6.7-39.3 2.2-46.6-7.4-1.9-2.2-1.8-3.6-3.9-6.2-3.6-3.9-7.3-2.2-11.9 1-57.4 36.4-140.3 21.4-147-43.3-3.1-29.3 12.4-57.1 39.6-71 38.2-19.5 112.2-11.8 114-30.9 1.1-10.2-1.9-20.1-11.3-27.3zm286.7 222c0 15.1-11.1 9.9-17.8 9.9H52.4c-7.4 0-18.2 4.8-17.8-10.7.4-13.9 10.5-9.1 17.1-9.1 132.3-.4 264.5-.4 396.8 0 6.8 0 16.6-4.4 16.6 9.9zm3.8-340.5v291c0 5.7-.7 13.9-8.1 13.9-12.4-.4-27.5 7.1-36.1-5.6-5.8-8.7-7.8-4-12.4-1.2-53.4 29.7-128.1 7.1-144.4-85.2-6.1-33.4-.7-67.1 15.7-100 11.8-23.9 56.9-76.1 136.1-30.5v-71c0-26.2-.1-26.2 26-26.2 3.1 0 6.6.4 9.7 0 10.1-.8 13.6 4.4 13.6 14.3-.1.2-.1.3-.1.5zm-51.5 232.3c-19.5 47.6-72.9 43.3-90 5.2-15.1-33.3-15.5-68.2.4-101.5 16.3-34.1 59.7-35.7 81.5-4.8 20.6 28.8 14.9 84.6 8.1 101.1zm-294.8 35.3c-7.5-1.3-33-3.3-33.7-27.8-.4-13.9 7.8-23 19.8-25.8 24.4-5.9 49.3-9.9 73.7-14.7 8.9-2 7.4 4.4 7.8 9.5 1.4 33-26.1 59.2-67.6 58.8z"],"creative-commons":[496,512,[],"f25e","M245.83 214.87l-33.22 17.28c-9.43-19.58-25.24-19.93-27.46-19.93-22.13 0-33.22 14.61-33.22 43.84 0 23.57 9.21 43.84 33.22 43.84 14.47 0 24.65-7.09 30.57-21.26l30.55 15.5c-6.17 11.51-25.69 38.98-65.1 38.98-22.6 0-73.96-10.32-73.96-77.05 0-58.69 43-77.06 72.63-77.06 30.72-.01 52.7 11.95 65.99 35.86zm143.05 0l-32.78 17.28c-9.5-19.77-25.72-19.93-27.9-19.93-22.14 0-33.22 14.61-33.22 43.84 0 23.55 9.23 43.84 33.22 43.84 14.45 0 24.65-7.09 30.54-21.26l31 15.5c-2.1 3.75-21.39 38.98-65.09 38.98-22.69 0-73.96-9.87-73.96-77.05 0-58.67 42.97-77.06 72.63-77.06 30.71-.01 52.58 11.95 65.56 35.86zM247.56 8.05C104.74 8.05 0 123.11 0 256.05c0 138.49 113.6 248 247.56 248 129.93 0 248.44-100.87 248.44-248 0-137.87-106.62-248-248.44-248zm.87 450.81c-112.54 0-203.7-93.04-203.7-202.81 0-105.42 85.43-203.27 203.72-203.27 112.53 0 202.82 89.46 202.82 203.26-.01 121.69-99.68 202.82-202.84 202.82z"],"watchman-monitoring":[512,512,[],"e087","M256,16C123.452,16,16,123.452,16,256S123.452,496,256,496,496,388.548,496,256,388.548,16,256,16ZM121.69,429.122C70.056,388.972,36.741,326.322,36.741,256a218.519,218.519,0,0,1,9.587-64.122l102.9-17.895-.121,10.967-13.943,2.013s-.144,12.5-.144,19.549a12.778,12.778,0,0,0,4.887,10.349l9.468,7.4Zm105.692-283.27,8.48-7.618s6.934-5.38-.143-9.344c-7.188-4.024-39.53-34.5-39.53-34.5-5.348-5.477-8.257-7.347-15.46,0,0,0-32.342,30.474-39.529,34.5-7.078,3.964-.144,9.344-.144,9.344l8.481,7.618-.048,4.369L75.982,131.045c39.644-56.938,105.532-94.3,180.018-94.3A218.754,218.754,0,0,1,420.934,111.77l-193.512,37.7Zm34.063,329.269-33.9-250.857,9.467-7.4a12.778,12.778,0,0,0,4.888-10.349c0-7.044-.144-19.549-.144-19.549l-13.943-2.013-.116-10.474,241.711,31.391A218.872,218.872,0,0,1,475.259,256C475.259,375.074,379.831,472.212,261.445,475.121Z"],fonticons:[448,512,[],"f280","M0 32v448h448V32zm187 140.9c-18.4 0-19 9.9-19 27.4v23.3c0 2.4-3.5 4.4-.6 4.4h67.4l-11.1 37.3H168v112.9c0 5.8-2 6.7 3.2 7.3l43.5 4.1v25.1H84V389l21.3-2c5.2-.6 6.7-2.3 6.7-7.9V267.7c0-2.3-2.9-2.3-5.8-2.3H84V228h28v-21c0-49.6 26.5-70 77.3-70 34.1 0 64.7 8.2 64.7 52.8l-50.7 6.1c.3-18.7-4.4-23-16.3-23zm74.3 241.8v-25.1l20.4-2.6c5.2-.6 7.6-1.7 7.6-7.3V271.8c0-4.1-2.9-6.7-6.7-7.9l-24.2-6.4 6.7-29.5h80.2v151.7c0 5.8-2.6 6.4 2.9 7.3l15.7 2.6v25.1zm80.8-255.5l9 33.2-7.3 7.3-31.2-16.6-31.2 16.6-7.3-7.3 9-33.2-21.8-24.2 3.5-9.6h27.7l15.5-28h9.3l15.5 28h27.7l3.5 9.6z"],weixin:[576,512,[],"f1d7","M385.2 167.6c6.4 0 12.6.3 18.8 1.1C387.4 90.3 303.3 32 207.7 32 100.5 32 13 104.8 13 197.4c0 53.4 29.3 97.5 77.9 131.6l-19.3 58.6 68-34.1c24.4 4.8 43.8 9.7 68.2 9.7 6.2 0 12.1-.3 18.3-.8-4-12.9-6.2-26.6-6.2-40.8-.1-84.9 72.9-154 165.3-154zm-104.5-52.9c14.5 0 24.2 9.7 24.2 24.4 0 14.5-9.7 24.2-24.2 24.2-14.8 0-29.3-9.7-29.3-24.2.1-14.7 14.6-24.4 29.3-24.4zm-136.4 48.6c-14.5 0-29.3-9.7-29.3-24.2 0-14.8 14.8-24.4 29.3-24.4 14.8 0 24.4 9.7 24.4 24.4 0 14.6-9.6 24.2-24.4 24.2zM563 319.4c0-77.9-77.9-141.3-165.4-141.3-92.7 0-165.4 63.4-165.4 141.3S305 460.7 397.6 460.7c19.3 0 38.9-5.1 58.6-9.9l53.4 29.3-14.8-48.6C534 402.1 563 363.2 563 319.4zm-219.1-24.5c-9.7 0-19.3-9.7-19.3-19.6 0-9.7 9.7-19.3 19.3-19.3 14.8 0 24.4 9.7 24.4 19.3 0 10-9.7 19.6-24.4 19.6zm107.1 0c-9.7 0-19.3-9.7-19.3-19.6 0-9.7 9.7-19.3 19.3-19.3 14.5 0 24.4 9.7 24.4 19.3.1 10-9.9 19.6-24.4 19.6z"],shirtsinbulk:[448,512,[],"f214","M100 410.3l30.6 13.4 4.4-9.9-30.6-13.4zm39.4 17.5l30.6 13.4 4.4-9.9-30.6-13.4zm172.1-14l4.4 9.9 30.6-13.4-4.4-9.9zM179.1 445l30.3 13.7 4.4-9.9-30.3-13.4zM60.4 392.8L91 406.2l4.4-9.6-30.6-13.7zm211.4 38.5l4.4 9.9 30.6-13.4-4.4-9.9zm-39.3 17.5l4.4 9.9 30.6-13.7-4.4-9.6zm118.4-52.2l4.4 9.6 30.6-13.4-4.4-9.9zM170 46.6h-33.5v10.5H170zm-47.2 0H89.2v10.5h33.5zm-47.3 0H42.3v10.5h33.3zm141.5 0h-33.2v10.5H217zm94.5 0H278v10.5h33.5zm47.3 0h-33.5v10.5h33.5zm-94.6 0H231v10.5h33.2zm141.5 0h-33.3v10.5h33.3zM52.8 351.1H42v33.5h10.8zm70-215.9H89.2v10.5h33.5zm-70 10.6h22.8v-10.5H42v33.5h10.8zm168.9 228.6c50.5 0 91.3-40.8 91.3-91.3 0-50.2-40.8-91.3-91.3-91.3-50.2 0-91.3 41.1-91.3 91.3 0 50.5 41.1 91.3 91.3 91.3zm-48.2-111.1c0-25.4 29.5-31.8 49.6-31.8 16.9 0 29.2 5.8 44.3 12l-8.8 16.9h-.9c-6.4-9.9-24.8-13.1-35.6-13.1-9 0-29.8 1.8-29.8 14.9 0 21.6 78.5-10.2 78.5 37.9 0 25.4-31.5 31.2-51 31.2-18.1 0-32.4-2.9-47.2-12.2l9-18.4h.9c6.1 12.2 23.6 14.9 35.9 14.9 8.7 0 32.7-1.2 32.7-14.3 0-26.1-77.6 6.3-77.6-38zM52.8 178.4H42V212h10.8zm342.4 206.2H406v-33.5h-10.8zM52.8 307.9H42v33.5h10.8zM0 3.7v406l221.7 98.6L448 409.7V3.7zm418.8 387.1L222 476.5 29.2 390.8V120.7h389.7v270.1zm0-299.3H29.2V32.9h389.7v58.6zm-366 130.1H42v33.5h10.8zm0 43.2H42v33.5h10.8zM170 135.2h-33.5v10.5H170zm225.2 163.1H406v-33.5h-10.8zm0-43.2H406v-33.5h-10.8zM217 135.2h-33.2v10.5H217zM395.2 212H406v-33.5h-10.8zm0 129.5H406V308h-10.8zm-131-206.3H231v10.5h33.2zm47.3 0H278v10.5h33.5zm83.7 33.6H406v-33.5h-33.5v10.5h22.8zm-36.4-33.6h-33.5v10.5h33.5z"],codepen:[512,512,[],"f1cb","M502.285 159.704l-234-156c-7.987-4.915-16.511-4.96-24.571 0l-234 156C3.714 163.703 0 170.847 0 177.989v155.999c0 7.143 3.714 14.286 9.715 18.286l234 156.022c7.987 4.915 16.511 4.96 24.571 0l234-156.022c6-3.999 9.715-11.143 9.715-18.286V177.989c-.001-7.142-3.715-14.286-9.716-18.285zM278 63.131l172.286 114.858-76.857 51.429L278 165.703V63.131zm-44 0v102.572l-95.429 63.715-76.857-51.429L234 63.131zM44 219.132l55.143 36.857L44 292.846v-73.714zm190 229.715L61.714 333.989l76.857-51.429L234 346.275v102.572zm22-140.858l-77.715-52 77.715-52 77.715 52-77.715 52zm22 140.858V346.275l95.429-63.715 76.857 51.429L278 448.847zm190-156.001l-55.143-36.857L468 219.132v73.714z"],"git-alt":[448,512,[],"f841","M439.55 236.05L244 40.45a28.87 28.87 0 0 0-40.81 0l-40.66 40.63 51.52 51.52c27.06-9.14 52.68 16.77 43.39 43.68l49.66 49.66c34.23-11.8 61.18 31 35.47 56.69-26.49 26.49-70.21-2.87-56-37.34L240.22 199v121.85c25.3 12.54 22.26 41.85 9.08 55a34.34 34.34 0 0 1-48.55 0c-17.57-17.6-11.07-46.91 11.25-56v-123c-20.8-8.51-24.6-30.74-18.64-45L142.57 101 8.45 235.14a28.86 28.86 0 0 0 0 40.81l195.61 195.6a28.86 28.86 0 0 0 40.8 0l194.69-194.69a28.86 28.86 0 0 0 0-40.81z"],lyft:[512,512,[],"f3c3","M0 81.1h77.8v208.7c0 33.1 15 52.8 27.2 61-12.7 11.1-51.2 20.9-80.2-2.8C7.8 334 0 310.7 0 289V81.1zm485.9 173.5v-22h23.8v-76.8h-26.1c-10.1-46.3-51.2-80.7-100.3-80.7-56.6 0-102.7 46-102.7 102.7V357c16 2.3 35.4-.3 51.7-14 17.1-14 24.8-37.2 24.8-59v-6.7h38.8v-76.8h-38.8v-23.3c0-34.6 52.2-34.6 52.2 0v77.1c0 56.6 46 102.7 102.7 102.7v-76.5c-14.5 0-26.1-11.7-26.1-25.9zm-294.3-99v113c0 15.4-23.8 15.4-23.8 0v-113H91v132.7c0 23.8 8 54 45 63.9 37 9.8 58.2-10.6 58.2-10.6-2.1 13.4-14.5 23.3-34.9 25.3-15.5 1.6-35.2-3.6-45-7.8v70.3c25.1 7.5 51.5 9.8 77.6 4.7 47.1-9.1 76.8-48.4 76.8-100.8V155.1h-77.1v.5z"],rev:[448,512,[],"f5b2","M289.67 274.89a65.57 65.57 0 1 1-65.56-65.56 65.64 65.64 0 0 1 65.56 65.56zm139.55-5.05h-.13a204.69 204.69 0 0 0-74.32-153l-45.38 26.2a157.07 157.07 0 0 1 71.81 131.84C381.2 361.5 310.73 432 224.11 432S67 361.5 67 274.88c0-81.88 63-149.27 143-156.43v39.12l108.77-62.79L210 32v38.32c-106.7 7.25-191 96-191 204.57 0 111.59 89.12 202.29 200.06 205v.11h210.16V269.84z"],windows:[448,512,[],"f17a","M0 93.7l183.6-25.3v177.4H0V93.7zm0 324.6l183.6 25.3V268.4H0v149.9zm203.8 28L448 480V268.4H203.8v177.9zm0-380.6v180.1H448V32L203.8 65.7z"],"wizards-of-the-coast":[640,512,[],"f730","M219.19 345.69c-1.9 1.38-11.07 8.44-.26 23.57 4.64 6.42 14.11 12.79 21.73 6.55 6.5-4.88 7.35-12.92.26-23.04-5.47-7.76-14.28-12.88-21.73-7.08zm336.75 75.94c-.34 1.7-.55 1.67.79 0 2.09-4.19 4.19-10.21 4.98-19.9 3.14-38.49-40.33-71.49-101.34-78.03-54.73-6.02-124.38 9.17-188.8 60.49l-.26 1.57c2.62 4.98 4.98 10.74 3.4 21.21l.79.26c63.89-58.4 131.19-77.25 184.35-73.85 58.4 3.67 100.03 34.04 100.03 68.08-.01 9.96-2.63 15.72-3.94 20.17zM392.28 240.42c.79 7.07 4.19 10.21 9.17 10.47 5.5.26 9.43-2.62 10.47-6.55.79-3.4 2.09-29.85 2.09-29.85s-11.26 6.55-14.93 10.47c-3.66 3.68-7.33 8.39-6.8 15.46zm-50.02-151.1C137.75 89.32 13.1 226.8.79 241.2c-1.05.52-1.31.79.79 1.31 60.49 16.5 155.81 81.18 196.13 202.16l1.05.26c55.25-69.92 140.88-128.05 236.99-128.05 80.92 0 130.15 42.16 130.15 80.39 0 18.33-6.55 33.52-22.26 46.35 0 .96-.2.79.79.79 14.66-10.74 27.5-28.8 27.5-48.18 0-22.78-12.05-38.23-12.05-38.23 7.07 7.07 10.74 16.24 10.74 16.24 5.76-40.85 26.97-62.32 26.97-62.32-2.36-9.69-6.81-17.81-6.81-17.81 7.59 8.12 14.4 27.5 14.4 41.37 0 10.47-3.4 22.78-12.57 31.95l.26.52c8.12-4.98 16.5-16.76 16.5-37.97 0-15.71-4.71-25.92-4.71-25.92 5.76-5.24 11.26-9.17 15.97-11.78.79 3.4 2.09 9.69 2.36 14.93 0 1.05.79 1.83 1.05 0 .79-5.76-.26-16.24-.26-16.5 6.02-3.14 9.69-4.45 9.69-4.45C617.74 176 489.43 89.32 342.26 89.32zm-99.24 289.62c-11.06 8.99-24.2 4.08-30.64-4.19-7.45-9.58-6.76-24.09 4.19-32.47 14.85-11.35 27.08-.49 31.16 5.5.28.39 12.13 16.57-4.71 31.16zm2.09-136.43l9.43-17.81 11.78 70.96-12.57 6.02-24.62-28.8 14.14-26.71 3.67 4.45-1.83-8.11zm18.59 117.58l-.26-.26c2.05-4.1-2.5-6.61-17.54-31.69-1.31-2.36-3.14-2.88-4.45-2.62l-.26-.52c7.86-5.76 15.45-10.21 25.4-15.71l.52.26c1.31 1.83 2.09 2.88 3.4 4.71l-.26.52c-1.05-.26-2.36-.79-5.24.26-2.09.79-7.86 3.67-12.31 7.59v1.31c1.57 2.36 3.93 6.55 5.76 9.69h.26c10.05-6.28 7.56-4.55 11.52-7.86h.26c.52 1.83.52 1.83 1.83 5.5l-.26.26c-3.06.61-4.65.34-11.52 5.5v.26c9.46 17.02 11.01 16.75 12.57 15.97l.26.26c-2.34 1.59-6.27 4.21-9.68 6.57zm55.26-32.47c-3.14 1.57-6.02 2.88-9.95 4.98l-.26-.26c1.29-2.59 1.16-2.71-11.78-32.47l-.26-.26c-.15 0-8.9 3.65-9.95 7.33h-.52l-1.05-5.76.26-.52c7.29-4.56 25.53-11.64 27.76-12.57l.52.26 3.14 4.98-.26.52c-3.53-1.76-7.35.76-12.31 2.62v.26c12.31 32.01 12.67 30.64 14.66 30.64v.25zm44.77-16.5c-4.19 1.05-5.24 1.31-9.69 2.88l-.26-.26.52-4.45c-1.05-3.4-3.14-11.52-3.67-13.62l-.26-.26c-3.4.79-8.9 2.62-12.83 3.93l-.26.26c.79 2.62 3.14 9.95 4.19 13.88.79 2.36 1.83 2.88 2.88 3.14v.52c-3.67 1.05-7.07 2.62-10.21 3.93l-.26-.26c1.05-1.31 1.05-2.88.26-4.98-1.05-3.14-8.12-23.83-9.17-27.23-.52-1.83-1.57-3.14-2.62-3.14v-.52c3.14-1.05 6.02-2.09 10.74-3.4l.26.26-.26 4.71c1.31 3.93 2.36 7.59 3.14 9.69h.26c3.93-1.31 9.43-2.88 12.83-3.93l.26-.26-2.62-9.43c-.52-1.83-1.05-3.4-2.62-3.93v-.26c4.45-1.05 7.33-1.83 10.74-2.36l.26.26c-1.05 1.31-1.05 2.88-.52 4.45 1.57 6.28 4.71 20.43 6.28 26.45.54 2.62 1.85 3.41 2.63 3.93zm32.21-6.81l-.26.26c-4.71.52-14.14 2.36-22.52 4.19l-.26-.26.79-4.19c-1.57-7.86-3.4-18.59-4.98-26.19-.26-1.83-.79-2.88-2.62-3.67l.79-.52c9.17-1.57 20.16-2.36 24.88-2.62l.26.26c.52 2.36.79 3.14 1.57 5.5l-.26.26c-1.14-1.14-3.34-3.2-16.24-.79l-.26.26c.26 1.57 1.05 6.55 1.57 9.95l.26.26c9.52-1.68 4.76-.06 10.74-2.36h.26c0 1.57-.26 1.83-.26 5.24h-.26c-4.81-1.03-2.15-.9-10.21 0l-.26.26c.26 2.09 1.57 9.43 2.09 12.57l.26.26c1.15.38 14.21-.65 16.24-4.71h.26c-.53 2.38-1.05 4.21-1.58 6.04zm10.74-44.51c-4.45 2.36-8.12 2.88-11 2.88-.25.02-11.41 1.09-17.54-9.95-6.74-10.79-.98-25.2 5.5-31.69 8.8-8.12 23.35-10.1 28.54-17.02 8.03-10.33-13.04-22.31-29.59-5.76l-2.62-2.88 5.24-16.24c25.59-1.57 45.2-3.04 50.02 16.24.79 3.14 0 9.43-.26 12.05 0 2.62-1.83 18.85-2.09 23.04-.52 4.19-.79 18.33-.79 20.69.26 2.36.52 4.19 1.57 5.5 1.57 1.83 5.76 1.83 5.76 1.83l-.79 4.71c-11.82-1.07-10.28-.59-20.43-1.05-3.22-5.15-2.23-3.28-4.19-7.86 0 .01-4.19 3.94-7.33 5.51zm37.18 21.21c-6.35-10.58-19.82-7.16-21.73 5.5-2.63 17.08 14.3 19.79 20.69 10.21l.26.26c-.52 1.83-1.83 6.02-1.83 6.28l-.52.52c-10.3 6.87-28.5-2.5-25.66-18.59 1.94-10.87 14.44-18.93 28.8-9.95l.26.52c0 1.06-.27 3.41-.27 5.25zm5.77-87.73v-6.55c.69 0 19.65 3.28 27.76 7.33l-1.57 17.54s10.21-9.43 15.45-10.74c5.24-1.57 14.93 7.33 14.93 7.33l-11.26 11.26c-12.07-6.35-19.59-.08-20.69.79-5.29 38.72-8.6 42.17 4.45 46.09l-.52 4.71c-17.55-4.29-18.53-4.5-36.92-7.33l.79-4.71c7.25 0 7.48-5.32 7.59-6.81 0 0 4.98-53.16 4.98-55.25-.02-2.87-4.99-3.66-4.99-3.66zm10.99 114.44c-8.12-2.09-14.14-11-10.74-20.69 3.14-9.43 12.31-12.31 18.85-10.21 9.17 2.62 12.83 11.78 10.74 19.38-2.61 8.9-9.42 13.87-18.85 11.52zm42.16 9.69c-2.36-.52-7.07-2.36-8.64-2.88v-.26l1.57-1.83c.59-8.24.59-7.27.26-7.59-4.82-1.81-6.66-2.36-7.07-2.36-1.31 1.83-2.88 4.45-3.67 5.5l-.79 3.4v.26c-1.31-.26-3.93-1.31-6.02-1.57v-.26l2.62-1.83c3.4-4.71 9.95-14.14 13.88-20.16v-2.09l.52-.26c2.09.79 5.5 2.09 7.59 2.88.48.48.18-1.87-1.05 25.14-.24 1.81.02 2.6.8 3.91zm-4.71-89.82c11.25-18.27 30.76-16.19 34.04-3.4L539.7 198c2.34-6.25-2.82-9.9-4.45-11.26l1.83-3.67c12.22 10.37 16.38 13.97 22.52 20.43-25.91 73.07-30.76 80.81-24.62 84.32l-1.83 4.45c-6.37-3.35-8.9-4.42-17.81-8.64l2.09-6.81c-.26-.26-3.93 3.93-9.69 3.67-19.06-1.3-22.89-31.75-9.67-52.9zm29.33 79.34c0-5.71-6.34-7.89-7.86-5.24-1.31 2.09 1.05 4.98 2.88 8.38 1.57 2.62 2.62 6.28 1.05 9.43-2.64 6.34-12.4 5.31-15.45-.79 0-.7-.27.09 1.83-4.71l.79-.26c-.57 5.66 6.06 9.61 8.38 4.98 1.05-2.09-.52-5.5-2.09-8.38-1.57-2.62-3.67-6.28-1.83-9.69 2.72-5.06 11.25-4.47 14.66 2.36v.52l-2.36 3.4zm21.21 13.36c-1.96-3.27-.91-2.14-4.45-4.71h-.26c-2.36 4.19-5.76 10.47-8.64 16.24-1.31 2.36-1.05 3.4-.79 3.93l-.26.26-5.76-4.45.26-.26 2.09-1.31c3.14-5.76 6.55-12.05 9.17-17.02v-.26c-2.64-1.98-1.22-1.51-6.02-1.83v-.26l3.14-3.4h.26c3.67 2.36 9.95 6.81 12.31 8.9l.26.26-1.31 3.91zm27.23-44.26l-2.88-2.88c.79-2.36 1.83-4.98 2.09-7.59.75-9.74-11.52-11.84-11.52-4.98 0 4.98 7.86 19.38 7.86 27.76 0 10.21-5.76 15.71-13.88 16.5-8.38.79-20.16-10.47-20.16-10.47l4.98-14.4 2.88 2.09c-2.97 17.8 17.68 20.37 13.35 5.24-1.06-4.02-18.75-34.2 2.09-38.23 13.62-2.36 23.04 16.5 23.04 16.5l-7.85 10.46zm35.62-10.21c-11-30.38-60.49-127.53-191.95-129.62-53.42-1.05-94.27 15.45-132.76 37.97l85.63-9.17-91.39 20.69 25.14 19.64-3.93-16.5c7.5-1.71 39.15-8.45 66.77-8.9l-22.26 80.39c13.61-.7 18.97-8.98 19.64-22.78l4.98-1.05.26 26.71c-22.46 3.21-37.3 6.69-49.49 9.95l13.09-43.21-61.54-36.66 2.36 8.12 10.21 4.98c6.28 18.59 19.38 56.56 20.43 58.66 1.95 4.28 3.16 5.78 12.05 4.45l1.05 4.98c-16.08 4.86-23.66 7.61-39.02 14.4l-2.36-4.71c4.4-2.94 8.73-3.94 5.5-12.83-23.7-62.5-21.48-58.14-22.78-59.44l2.36-4.45 33.52 67.3c-3.84-11.87 1.68 1.69-32.99-78.82l-41.9 88.51 4.71-13.88-35.88-42.16 27.76 93.48-11.78 8.38C95 228.58 101.05 231.87 93.23 231.52c-5.5-.26-13.62 5.5-13.62 5.5L74.63 231c30.56-23.53 31.62-24.33 58.4-42.68l4.19 7.07s-5.76 4.19-7.86 7.07c-5.9 9.28 1.67 13.28 61.8 75.68l-18.85-58.92 39.8-10.21 25.66 30.64 4.45-12.31-4.98-24.62 13.09-3.4.52 3.14 3.67-10.47-94.27 29.33 11.26-4.98-13.62-42.42 17.28-9.17 30.11 36.14 28.54-13.09c-1.41-7.47-2.47-14.5-4.71-19.64l17.28 13.88 4.71-2.09-59.18-42.68 23.08 11.5c18.98-6.07 25.23-7.47 32.21-9.69l2.62 11c-12.55 12.55 1.43 16.82 6.55 19.38l-13.62-61.01 12.05 28.28c4.19-1.31 7.33-2.09 7.33-2.09l2.62 8.64s-3.14 1.05-6.28 2.09l8.9 20.95 33.78-65.73-20.69 61.01c42.42-24.09 81.44-36.66 131.98-35.88 67.04 1.05 167.33 40.85 199.8 139.83.78 2.1-.01 2.63-.79.27zM203.48 152.43s1.83-.52 4.19-1.31l9.43 7.59c-.4 0-3.44-.25-11.26 2.36l-2.36-8.64zm143.76 38.5c-1.57-.6-26.46-4.81-33.26 20.69l21.73 17.02 11.53-37.71zM318.43 67.07c-58.4 0-106.05 12.05-114.96 14.4v.79c8.38 2.09 14.4 4.19 21.21 11.78l1.57.26c6.55-1.83 48.97-13.88 110.24-13.88 180.16 0 301.67 116.79 301.67 223.37v9.95c0 1.31.79 2.62 1.05.52.52-2.09.79-8.64.79-19.64.26-83.79-96.63-227.55-321.57-227.55zm211.06 169.68c1.31-5.76 0-12.31-7.33-13.09-9.62-1.13-16.14 23.79-17.02 33.52-.79 5.5-1.31 14.93 6.02 14.93 4.68-.01 9.72-.91 18.33-35.36zm-61.53 42.95c-2.62-.79-9.43-.79-12.57 10.47-1.83 6.81.52 13.35 6.02 14.66 3.67 1.05 8.9.52 11.78-10.74 2.62-9.94-1.83-13.61-5.23-14.39zM491 300.65c1.83.52 3.14 1.05 5.76 1.83 0-1.83.52-8.38.79-12.05-1.05 1.31-5.5 8.12-6.55 9.95v.27z"],"square-viadeo":[448,512,["viadeo-square"],"f2aa","M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM118.3 381.2c-68-73.6-19.8-196.1 81.2-196.1c13.3 0 26.6 2.1 39.1 6.7c-4.3 8.4-7.3 17.6-8.4 27.1c-9.7-4.1-20.2-6-30.7-6c-48.8 0-84.6 41.7-84.6 88.9c0 43 28.5 78.7 69.5 85.9c61.5-24 72.9-117.6 72.9-175v0c0-7.3 0-14.8-.6-22.1c-11.2-32.9-26.6-64.6-44.2-94.5c27.1 18.3 41.9 62.5 44.2 94.1v.4c7.7 22.5 11.8 46.2 11.8 70c0 54.1-21.9 99-68.3 128.2l-2.4 .2c50 1 86.2-38.6 86.2-87.2c0-12.2-2.1-24.3-6.9-35.7c9.5-1.9 18.5-5.6 26.4-10.5c15.3 36.6 12.6 87.3-22.8 125.6c-42.4 46.2-120 46.6-162.4 0zM274.6 217.6c21.9-12 49.6-30.7 62.3-53c1.5-3 4.1-8.6 4.5-12c-12.5 27.9-44.2 49.8-73.9 56.7c-4.7-7.3-7.5-15.5-7.5-24.3c0-10.3 5.2-24.1 12.9-31.6c8.3-7.9 18-10.9 27.9-14.1c16-5.1 32.5-10.3 44.5-35.9c32.5 46.2 13.1 130.3-36.3 130.3c-13.3 0-25.1-7.1-34.4-16.1z"],meetup:[512,512,[],"f2e0","M99 414.3c1.1 5.7-2.3 11.1-8 12.3-5.4 1.1-10.9-2.3-12-8-1.1-5.4 2.3-11.1 7.7-12.3 5.4-1.2 11.1 2.3 12.3 8zm143.1 71.4c-6.3 4.6-8 13.4-3.7 20 4.6 6.6 13.4 8.3 20 3.7 6.3-4.6 8-13.4 3.4-20-4.2-6.5-13.1-8.3-19.7-3.7zm-86-462.3c6.3-1.4 10.3-7.7 8.9-14-1.1-6.6-7.4-10.6-13.7-9.1-6.3 1.4-10.3 7.7-9.1 14 1.4 6.6 7.6 10.6 13.9 9.1zM34.4 226.3c-10-6.9-23.7-4.3-30.6 6-6.9 10-4.3 24 5.7 30.9 10 7.1 23.7 4.6 30.6-5.7 6.9-10.4 4.3-24.1-5.7-31.2zm272-170.9c10.6-6.3 13.7-20 7.7-30.3-6.3-10.6-19.7-14-30-7.7s-13.7 20-7.4 30.6c6 10.3 19.4 13.7 29.7 7.4zm-191.1 58c7.7-5.4 9.4-16 4.3-23.7s-15.7-9.4-23.1-4.3c-7.7 5.4-9.4 16-4.3 23.7 5.1 7.8 15.6 9.5 23.1 4.3zm372.3 156c-7.4 1.7-12.3 9.1-10.6 16.9 1.4 7.4 8.9 12.3 16.3 10.6 7.4-1.4 12.3-8.9 10.6-16.6-1.5-7.4-8.9-12.3-16.3-10.9zm39.7-56.8c-1.1-5.7-6.6-9.1-12-8-5.7 1.1-9.1 6.9-8 12.6 1.1 5.4 6.6 9.1 12.3 8 5.4-1.5 9.1-6.9 7.7-12.6zM447 138.9c-8.6 6-10.6 17.7-4.9 26.3 5.7 8.6 17.4 10.6 26 4.9 8.3-6 10.3-17.7 4.6-26.3-5.7-8.7-17.4-10.9-25.7-4.9zm-6.3 139.4c26.3 43.1 15.1 100-26.3 129.1-17.4 12.3-37.1 17.7-56.9 17.1-12 47.1-69.4 64.6-105.1 32.6-1.1.9-2.6 1.7-3.7 2.9-39.1 27.1-92.3 17.4-119.4-22.3-9.7-14.3-14.6-30.6-15.1-46.9-65.4-10.9-90-94-41.1-139.7-28.3-46.9.6-107.4 53.4-114.9C151.6 70 234.1 38.6 290.1 82c67.4-22.3 136.3 29.4 130.9 101.1 41.1 12.6 52.8 66.9 19.7 95.2zm-70 74.3c-3.1-20.6-40.9-4.6-43.1-27.1-3.1-32 43.7-101.1 40-128-3.4-24-19.4-29.1-33.4-29.4-13.4-.3-16.9 2-21.4 4.6-2.9 1.7-6.6 4.9-11.7-.3-6.3-6-11.1-11.7-19.4-12.9-12.3-2-17.7 2-26.6 9.7-3.4 2.9-12 12.9-20 9.1-3.4-1.7-15.4-7.7-24-11.4-16.3-7.1-40 4.6-48.6 20-12.9 22.9-38 113.1-41.7 125.1-8.6 26.6 10.9 48.6 36.9 47.1 11.1-.6 18.3-4.6 25.4-17.4 4-7.4 41.7-107.7 44.6-112.6 2-3.4 8.9-8 14.6-5.1 5.7 3.1 6.9 9.4 6 15.1-1.1 9.7-28 70.9-28.9 77.7-3.4 22.9 26.9 26.6 38.6 4 3.7-7.1 45.7-92.6 49.4-98.3 4.3-6.3 7.4-8.3 11.7-8 3.1 0 8.3.9 7.1 10.9-1.4 9.4-35.1 72.3-38.9 87.7-4.6 20.6 6.6 41.4 24.9 50.6 11.4 5.7 62.5 15.7 58.5-11.1zm5.7 92.3c-10.3 7.4-12.9 22-5.7 32.6 7.1 10.6 21.4 13.1 32 6 10.6-7.4 13.1-22 6-32.6-7.4-10.6-21.7-13.5-32.3-6z"],centos:[448,512,[],"f789","M289.6 97.5l31.6 31.7-76.3 76.5V97.5zm-162.4 31.7l76.3 76.5V97.5h-44.7zm41.5-41.6h44.7v127.9l10.8 10.8 10.8-10.8V87.6h44.7L224.2 32zm26.2 168.1l-10.8-10.8H55.5v-44.8L0 255.7l55.5 55.6v-44.8h128.6l10.8-10.8zm79.3-20.7h107.9v-44.8l-31.6-31.7zm173.3 20.7L392 200.1v44.8H264.3l-10.8 10.8 10.8 10.8H392v44.8l55.5-55.6zM65.4 176.2l32.5-31.7 90.3 90.5h15.3v-15.3l-90.3-90.5 31.6-31.7H65.4zm316.7-78.7h-78.5l31.6 31.7-90.3 90.5V235h15.3l90.3-90.5 31.6 31.7zM203.5 413.9V305.8l-76.3 76.5 31.6 31.7h44.7zM65.4 235h108.8l-76.3-76.5-32.5 31.7zm316.7 100.2l-31.6 31.7-90.3-90.5h-15.3v15.3l90.3 90.5-31.6 31.7h78.5zm0-58.8H274.2l76.3 76.5 31.6-31.7zm-60.9 105.8l-76.3-76.5v108.1h44.7zM97.9 352.9l76.3-76.5H65.4v44.8zm181.8 70.9H235V295.9l-10.8-10.8-10.8 10.8v127.9h-44.7l55.5 55.6zm-166.5-41.6l90.3-90.5v-15.3h-15.3l-90.3 90.5-32.5-31.7v78.7h79.4z"],adn:[496,512,[],"f170","M248 167.5l64.9 98.8H183.1l64.9-98.8zM496 256c0 136.9-111.1 248-248 248S0 392.9 0 256 111.1 8 248 8s248 111.1 248 248zm-99.8 82.7L248 115.5 99.8 338.7h30.4l33.6-51.7h168.6l33.6 51.7h30.2z"],cloudsmith:[512,512,[],"f384","M512 227.6v56.9L284.4 512H227.6L0 284.4V227.6L227.6 0h56.9L512 227.6zm-256 162a133.6 133.6 0 1 0 0-267.1 133.6 133.6 0 1 0 0 267.1z"],opensuse:[640,512,[],"e62b","M471.1 102.7s-.3 18.3-.3 20.3c-9.1-3-74.4-24.1-135.7-26.3c-51.9-1.8-122.8-4.3-223 57.3c-19.4 12.4-73.9 46.1-99.6 109.7C7 277-.1 307 7 335.1c3.3 12.8 8.9 24.9 16.5 35.7c17.4 25 46.6 41.6 78.1 44.4c44.4 3.9 78.1-16 90-53.3c8.2-25.8 0-63.6-31.5-82.9c-25.6-15.7-53.3-12.1-69.2-1.6c-13.9 9.2-21.8 23.5-21.6 39.2c.3 27.8 24.3 42.6 41.5 42.6c5.4 0 10.7-.9 15.8-2.7c6.5-1.8 13.3-6.5 13.3-14.9c0-12.1-11.6-14.8-16.8-13.9c-2.9 .5-4.5 2-11.8 2.4c-2-.2-12-3.1-12-14V316c.2-12.3 13.2-18 25.5-16.9c32.3 2.8 47.7 40.7 28.5 65.7C135 388.5 76.7 388 53.6 344.4c-26-49.2 12.7-111.2 87-98.4c33.2 5.7 83.6 35.5 102.4 104.3h45.9c-5.7-17.6-8.9-68.3 42.7-68.3c56.7 0 63.9 39.9 79.8 68.3H460c-12.8-18.3-21.7-38.7-18.9-55.8c5.6-33.8 39.7-18.4 82.4-17.4c66.5 .4 102.1-27 103.1-28c3.7-3.1 6.5-15.8 7-17.7c1.3-5.1-3.2-2.4-3.2-2.4c-8.7 5.2-30.5 15.2-50.9 15.6c-25.3 .5-76.2-25.4-81.6-28.2c-.3-.4 .1 1.2-11-25.5c88.4 58.3 118.3 40.5 145.2 21.7c.8-.6 4.3-2.9 3.6-5.7c-13.8-48.1-22.4-62.7-34.5-69.6c-37-21.6-125-34.7-129.2-35.3c.1-.1-.9-.3-.9 .7l0 0zm135.6 75.4a37.6 37.6 0 1 1 -75.2-2.6 37.6 37.6 0 1 1 75.2 2.6zm-36.6-27.9a26.3 26.3 0 1 0 -1.7 52.5 26.3 26.3 0 1 0 1.7-52.5zm4.3 28.8c-15.4 0-15.4-15.6 0-15.6s15.4 15.6 0 15.6v0z"],"pied-piper-alt":[576,512,[],"f1a8","M244 246c-3.2-2-6.3-2.9-10.1-2.9-6.6 0-12.6 3.2-19.3 3.7l1.7 4.9zm135.9 197.9c-19 0-64.1 9.5-79.9 19.8l6.9 45.1c35.7 6.1 70.1 3.6 106-9.8-4.8-10-23.5-55.1-33-55.1zM340.8 177c6.6 2.8 11.5 9.2 22.7 22.1 2-1.4 7.5-5.2 7.5-8.6 0-4.9-11.8-13.2-13.2-23 11.2-5.7 25.2-6 37.6-8.9 68.1-16.4 116.3-52.9 146.8-116.7C548.3 29.3 554 16.1 554.6 2l-2 2.6c-28.4 50-33 63.2-81.3 100-31.9 24.4-69.2 40.2-106.6 54.6l-6.3-.3v-21.8c-19.6 1.6-19.7-14.6-31.6-23-18.7 20.6-31.6 40.8-58.9 51.1-12.7 4.8-19.6 10-25.9 21.8 34.9-16.4 91.2-13.5 98.8-10zM555.5 0l-.6 1.1-.3.9.6-.6zm-59.2 382.1c-33.9-56.9-75.3-118.4-150-115.5l-.3-6c-1.1-13.5 32.8 3.2 35.1-31l-14.4 7.2c-19.8-45.7-8.6-54.3-65.5-54.3-14.7 0-26.7 1.7-41.4 4.6 2.9 18.6 2.2 36.7-10.9 50.3l19.5 5.5c-1.7 3.2-2.9 6.3-2.9 9.8 0 21 42.8 2.9 42.8 33.6 0 18.4-36.8 60.1-54.9 60.1-8 0-53.7-50-53.4-60.1l.3-4.6 52.3-11.5c13-2.6 12.3-22.7-2.9-22.7-3.7 0-43.1 9.2-49.4 10.6-2-5.2-7.5-14.1-13.8-14.1-3.2 0-6.3 3.2-9.5 4-9.2 2.6-31 2.9-21.5 20.1L15.9 298.5c-5.5 1.1-8.9 6.3-8.9 11.8 0 6 5.5 10.9 11.5 10.9 8 0 131.3-28.4 147.4-32.2 2.6 3.2 4.6 6.3 7.8 8.6 20.1 14.4 59.8 85.9 76.4 85.9 24.1 0 58-22.4 71.3-41.9 3.2-4.3 6.9-7.5 12.4-6.9.6 13.8-31.6 34.2-33 43.7-1.4 10.2-1 35.2-.3 41.1 26.7 8.1 52-3.6 77.9-2.9 4.3-21 10.6-41.9 9.8-63.5l-.3-9.5c-1.4-34.2-10.9-38.5-34.8-58.6-1.1-1.1-2.6-2.6-3.7-4 2.2-1.4 1.1-1 4.6-1.7 88.5 0 56.3 183.6 111.5 229.9 33.1-15 72.5-27.9 103.5-47.2-29-25.6-52.6-45.7-72.7-79.9zm-196.2 46.1v27.2l11.8-3.4-2.9-23.8zm-68.7-150.4l24.1 61.2 21-13.8-31.3-50.9zm84.4 154.9l2 12.4c9-1.5 58.4-6.6 58.4-14.1 0-1.4-.6-3.2-.9-4.6-26.8 0-36.9 3.8-59.5 6.3z"],"square-dribbble":[448,512,["dribbble-square"],"f397","M165.9 132.5c-38.3 18-66.8 53.3-75.7 95.7c6.1 .1 62.4 .3 126.4-16.7c-22.7-40.2-47.1-74.1-50.7-79zm26.1-9.1c3.8 5.1 28.6 38.9 51 80c48.6-18.3 69.1-45.9 71.6-49.4C281 124.2 235.3 112.9 192 123.4zM277.4 382c-2-12-10-53.8-29.2-103.6c-55.1 18.8-93.8 56.4-108.1 85.6c40.5 31.6 93.3 36.7 137.3 18zM227.8 232.6C159.6 253 93.4 252.2 87.4 252c0 .7 0 1.4 0 2.1s0 1.4 0 2.1c0 35.1 13.3 67.1 35.1 91.4c22.2-37.9 67.1-77.9 116.5-91.8c-3.4-7.8-7.2-15.5-11.1-23.2zm72.5 136.9c30.7-20.7 52.5-53.6 58.6-91.6c-4.6-1.5-42.3-12.7-85.1-5.8c17.9 49.1 25.1 89.1 26.5 97.4zm-34.8-119c45.5-5.7 90.7 3.4 95.2 4.4c-.3-32.3-11.8-61.9-30.9-85.1c-2.9 3.9-25.8 33.2-76.3 53.9c4.8 9.8 8.3 17.8 12 26.8zM384 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64zM224 416a160 160 0 1 1 0-320 160 160 0 1 1 0 320z"],codiepie:[472,512,[],"f284","M422.5 202.9c30.7 0 33.5 53.1-.3 53.1h-10.8v44.3h-26.6v-97.4h37.7zM472 352.6C429.9 444.5 350.4 504 248 504 111 504 0 393 0 256S111 8 248 8c97.4 0 172.8 53.7 218.2 138.4l-186 108.8L472 352.6zm-38.5 12.5l-60.3-30.7c-27.1 44.3-70.4 71.4-122.4 71.4-82.5 0-149.2-66.7-149.2-148.9 0-82.5 66.7-149.2 149.2-149.2 48.4 0 88.9 23.5 116.9 63.4l59.5-34.6c-40.7-62.6-104.7-100-179.2-100-121.2 0-219.5 98.3-219.5 219.5S126.8 475.5 248 475.5c78.6 0 146.5-42.1 185.5-110.4z"],node:[640,512,[],"f419","M316.3 452c-2.1 0-4.2-.6-6.1-1.6L291 439c-2.9-1.6-1.5-2.2-.5-2.5 3.8-1.3 4.6-1.6 8.7-4 .4-.2 1-.1 1.4.1l14.8 8.8c.5.3 1.3.3 1.8 0L375 408c.5-.3.9-.9.9-1.6v-66.7c0-.7-.3-1.3-.9-1.6l-57.8-33.3c-.5-.3-1.2-.3-1.8 0l-57.8 33.3c-.6.3-.9 1-.9 1.6v66.7c0 .6.4 1.2.9 1.5l15.8 9.1c8.6 4.3 13.9-.8 13.9-5.8v-65.9c0-.9.7-1.7 1.7-1.7h7.3c.9 0 1.7.7 1.7 1.7v65.9c0 11.5-6.2 18-17.1 18-3.3 0-6 0-13.3-3.6l-15.2-8.7c-3.7-2.2-6.1-6.2-6.1-10.5v-66.7c0-4.3 2.3-8.4 6.1-10.5l57.8-33.4c3.7-2.1 8.5-2.1 12.1 0l57.8 33.4c3.7 2.2 6.1 6.2 6.1 10.5v66.7c0 4.3-2.3 8.4-6.1 10.5l-57.8 33.4c-1.7 1.1-3.8 1.7-6 1.7zm46.7-65.8c0-12.5-8.4-15.8-26.2-18.2-18-2.4-19.8-3.6-19.8-7.8 0-3.5 1.5-8.1 14.8-8.1 11.9 0 16.3 2.6 18.1 10.6.2.8.8 1.3 1.6 1.3h7.5c.5 0 .9-.2 1.2-.5.3-.4.5-.8.4-1.3-1.2-13.8-10.3-20.2-28.8-20.2-16.5 0-26.3 7-26.3 18.6 0 12.7 9.8 16.1 25.6 17.7 18.9 1.9 20.4 4.6 20.4 8.3 0 6.5-5.2 9.2-17.4 9.2-15.3 0-18.7-3.8-19.8-11.4-.1-.8-.8-1.4-1.7-1.4h-7.5c-.9 0-1.7.7-1.7 1.7 0 9.7 5.3 21.3 30.6 21.3 18.5 0 29-7.2 29-19.8zm54.5-50.1c0 6.1-5 11.1-11.1 11.1s-11.1-5-11.1-11.1c0-6.3 5.2-11.1 11.1-11.1 6-.1 11.1 4.8 11.1 11.1zm-1.8 0c0-5.2-4.2-9.3-9.4-9.3-5.1 0-9.3 4.1-9.3 9.3 0 5.2 4.2 9.4 9.3 9.4 5.2-.1 9.4-4.3 9.4-9.4zm-4.5 6.2h-2.6c-.1-.6-.5-3.8-.5-3.9-.2-.7-.4-1.1-1.3-1.1h-2.2v5h-2.4v-12.5h4.3c1.5 0 4.4 0 4.4 3.3 0 2.3-1.5 2.8-2.4 3.1 1.7.1 1.8 1.2 2.1 2.8.1 1 .3 2.7.6 3.3zm-2.8-8.8c0-1.7-1.2-1.7-1.8-1.7h-2v3.5h1.9c1.6 0 1.9-1.1 1.9-1.8zM137.3 191c0-2.7-1.4-5.1-3.7-6.4l-61.3-35.3c-1-.6-2.2-.9-3.4-1h-.6c-1.2 0-2.3.4-3.4 1L3.7 184.6C1.4 185.9 0 188.4 0 191l.1 95c0 1.3.7 2.5 1.8 3.2 1.1.7 2.5.7 3.7 0L42 268.3c2.3-1.4 3.7-3.8 3.7-6.4v-44.4c0-2.6 1.4-5.1 3.7-6.4l15.5-8.9c1.2-.7 2.4-1 3.7-1 1.3 0 2.6.3 3.7 1l15.5 8.9c2.3 1.3 3.7 3.8 3.7 6.4v44.4c0 2.6 1.4 5.1 3.7 6.4l36.4 20.9c1.1.7 2.6.7 3.7 0 1.1-.6 1.8-1.9 1.8-3.2l.2-95zM472.5 87.3v176.4c0 2.6-1.4 5.1-3.7 6.4l-61.3 35.4c-2.3 1.3-5.1 1.3-7.4 0l-61.3-35.4c-2.3-1.3-3.7-3.8-3.7-6.4v-70.8c0-2.6 1.4-5.1 3.7-6.4l61.3-35.4c2.3-1.3 5.1-1.3 7.4 0l15.3 8.8c1.7 1 3.9-.3 3.9-2.2v-94c0-2.8 3-4.6 5.5-3.2l36.5 20.4c2.3 1.2 3.8 3.7 3.8 6.4zm-46 128.9c0-.7-.4-1.3-.9-1.6l-21-12.2c-.6-.3-1.3-.3-1.9 0l-21 12.2c-.6.3-.9.9-.9 1.6v24.3c0 .7.4 1.3.9 1.6l21 12.1c.6.3 1.3.3 1.8 0l21-12.1c.6-.3.9-.9.9-1.6v-24.3zm209.8-.7c2.3-1.3 3.7-3.8 3.7-6.4V192c0-2.6-1.4-5.1-3.7-6.4l-60.9-35.4c-2.3-1.3-5.1-1.3-7.4 0l-61.3 35.4c-2.3 1.3-3.7 3.8-3.7 6.4v70.8c0 2.7 1.4 5.1 3.7 6.4l60.9 34.7c2.2 1.3 5 1.3 7.3 0l36.8-20.5c2.5-1.4 2.5-5 0-6.4L550 241.6c-1.2-.7-1.9-1.9-1.9-3.2v-22.2c0-1.3.7-2.5 1.9-3.2l19.2-11.1c1.1-.7 2.6-.7 3.7 0l19.2 11.1c1.1.7 1.9 1.9 1.9 3.2v17.4c0 2.8 3.1 4.6 5.6 3.2l36.7-21.3zM559 219c-.4.3-.7.7-.7 1.2v13.6c0 .5.3 1 .7 1.2l11.8 6.8c.4.3 1 .3 1.4 0L584 235c.4-.3.7-.7.7-1.2v-13.6c0-.5-.3-1-.7-1.2l-11.8-6.8c-.4-.3-1-.3-1.4 0L559 219zm-254.2 43.5v-70.4c0-2.6-1.6-5.1-3.9-6.4l-61.1-35.2c-2.1-1.2-5-1.4-7.4 0l-61.1 35.2c-2.3 1.3-3.9 3.7-3.9 6.4v70.4c0 2.8 1.9 5.2 4 6.4l61.2 35.2c2.4 1.4 5.2 1.3 7.4 0l61-35.2c1.8-1 3.1-2.7 3.6-4.7.1-.5.2-1.1.2-1.7zm-74.3-124.9l-.8.5h1.1l-.3-.5zm76.2 130.2l-.4-.7v.9l.4-.2z"],mix:[448,512,[],"f3cb","M0 64v348.9c0 56.2 88 58.1 88 0V174.3c7.9-52.9 88-50.4 88 6.5v175.3c0 57.9 96 58 96 0V240c5.3-54.7 88-52.5 88 4.3v23.8c0 59.9 88 56.6 88 0V64H0z"],steam:[496,512,[],"f1b6","M496 256c0 137-111.2 248-248.4 248-113.8 0-209.6-76.3-239-180.4l95.2 39.3c6.4 32.1 34.9 56.4 68.9 56.4 39.2 0 71.9-32.4 70.2-73.5l84.5-60.2c52.1 1.3 95.8-40.9 95.8-93.5 0-51.6-42-93.5-93.7-93.5s-93.7 42-93.7 93.5v1.2L176.6 279c-15.5-.9-30.7 3.4-43.5 12.1L0 236.1C10.2 108.4 117.1 8 247.6 8 384.8 8 496 119 496 256zM155.7 384.3l-30.5-12.6a52.79 52.79 0 0 0 27.2 25.8c26.9 11.2 57.8-1.6 69-28.4 5.4-13 5.5-27.3.1-40.3-5.4-13-15.5-23.2-28.5-28.6-12.9-5.4-26.7-5.2-38.9-.6l31.5 13c19.8 8.2 29.2 30.9 20.9 50.7-8.3 19.9-31 29.2-50.8 21zm173.8-129.9c-34.4 0-62.4-28-62.4-62.3s28-62.3 62.4-62.3 62.4 28 62.4 62.3-27.9 62.3-62.4 62.3zm.1-15.6c25.9 0 46.9-21 46.9-46.8 0-25.9-21-46.8-46.9-46.8s-46.9 21-46.9 46.8c.1 25.8 21.1 46.8 46.9 46.8z"],"cc-apple-pay":[576,512,[],"f416","M302.2 218.4c0 17.2-10.5 27.1-29 27.1h-24.3v-54.2h24.4c18.4 0 28.9 9.8 28.9 27.1zm47.5 62.6c0 8.3 7.2 13.7 18.5 13.7 14.4 0 25.2-9.1 25.2-21.9v-7.7l-23.5 1.5c-13.3.9-20.2 5.8-20.2 14.4zM576 79v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V79c0-26.5 21.5-48 48-48h480c26.5 0 48 21.5 48 48zM127.8 197.2c8.4.7 16.8-4.2 22.1-10.4 5.2-6.4 8.6-15 7.7-23.7-7.4.3-16.6 4.9-21.9 11.3-4.8 5.5-8.9 14.4-7.9 22.8zm60.6 74.5c-.2-.2-19.6-7.6-19.8-30-.2-18.7 15.3-27.7 16-28.2-8.8-13-22.4-14.4-27.1-14.7-12.2-.7-22.6 6.9-28.4 6.9-5.9 0-14.7-6.6-24.3-6.4-12.5.2-24.2 7.3-30.5 18.6-13.1 22.6-3.4 56 9.3 74.4 6.2 9.1 13.7 19.1 23.5 18.7 9.3-.4 13-6 24.2-6 11.3 0 14.5 6 24.3 5.9 10.2-.2 16.5-9.1 22.8-18.2 6.9-10.4 9.8-20.4 10-21zm135.4-53.4c0-26.6-18.5-44.8-44.9-44.8h-51.2v136.4h21.2v-46.6h29.3c26.8 0 45.6-18.4 45.6-45zm90 23.7c0-19.7-15.8-32.4-40-32.4-22.5 0-39.1 12.9-39.7 30.5h19.1c1.6-8.4 9.4-13.9 20-13.9 13 0 20.2 6 20.2 17.2v7.5l-26.4 1.6c-24.6 1.5-37.9 11.6-37.9 29.1 0 17.7 13.7 29.4 33.4 29.4 13.3 0 25.6-6.7 31.2-17.4h.4V310h19.6v-68zM516 210.9h-21.5l-24.9 80.6h-.4l-24.9-80.6H422l35.9 99.3-1.9 6c-3.2 10.2-8.5 14.2-17.9 14.2-1.7 0-4.9-.2-6.2-.3v16.4c1.2.4 6.5.5 8.1.5 20.7 0 30.4-7.9 38.9-31.8L516 210.9z"],scribd:[384,512,[],"f28a","M42.3 252.7c-16.1-19-24.7-45.9-24.8-79.9 0-100.4 75.2-153.1 167.2-153.1 98.6-1.6 156.8 49 184.3 70.6l-50.5 72.1-37.3-24.6 26.9-38.6c-36.5-24-79.4-36.5-123-35.8-50.7-.8-111.7 27.2-111.7 76.2 0 18.7 11.2 20.7 28.6 15.6 23.3-5.3 41.9.6 55.8 14 26.4 24.3 23.2 67.6-.7 91.9-29.2 29.5-85.2 27.3-114.8-8.4zm317.7 5.9c-15.5-18.8-38.9-29.4-63.2-28.6-38.1-2-71.1 28-70.5 67.2-.7 16.8 6 33 18.4 44.3 14.1 13.9 33 19.7 56.3 14.4 17.4-5.1 28.6-3.1 28.6 15.6 0 4.3-.5 8.5-1.4 12.7-16.7 40.9-59.5 64.4-121.4 64.4-51.9.2-102.4-16.4-144.1-47.3l33.7-39.4-35.6-27.4L0 406.3l15.4 13.8c52.5 46.8 120.4 72.5 190.7 72.2 51.4 0 94.4-10.5 133.6-44.1 57.1-51.4 54.2-149.2 20.3-189.6z"],debian:[448,512,[],"e60b","M380.2 245.6c3-7.6 5.5-14 5.2-24.4l-4.3 9c4.4-13.2 4-27.1 3.6-40.4c-.2-6-.3-11.8 0-17.4l-1.8-.5c-1.5-45.2-40.6-93.1-75.3-109.4c-30-13.8-76.1-16.2-97.3-5.8c1.3-1.1 4.2-2 6.8-2.7l.3-.1c3.3-1 6-1.7 4-2.9c-19.2 1.9-24.9 5.5-31.1 9.4l-.1 0c-4.6 2.9-9.5 6-20.3 8.7c-3.5 3.4 1.7 2 5.8 .9l0 0c4.1-1.1 7.2-1.9-.1 2.4c-3.5 1-6.6 1.3-9.6 1.6l-.1 0c-8.3 .8-15.8 1.6-30.7 17c.8 1.3 3.4-.2 5.3-1.3l.1-.1c2.3-1.4 3.4-2-1.7 4.4c-19.1-2.4-60.3 43.7-69.1 59l4.6 .8c-3.2 8-6.8 14.8-10 20.8c-4.3 8.1-7.9 14.9-8.7 21.3c-.3 5.1-1 11-1.7 17.3l0 0c-.1 1-.2 2-.3 3l-.1 .6c-3 27.3-6.7 60.8 3.9 73l-1.3 13c.6 1.2 1.1 2.3 1.6 3.5c.2 .4 .4 .8 .5 1.1l0 0 0 0 0 0 0 0 0 0 0 0 0 0c1 2.1 2 4.2 3.3 6.2l-3 .2c7 22.1 10.8 22.5 15.1 22.9l0 0c4.4 .4 9.3 .9 18.7 24.2c-2.7-.9-5.5-1.9-9.4-7.2c-.5 4.1 5.8 16.3 13.1 25.8l-3.1 3.6c2.1 3.7 4.8 6.2 7.6 8.8l0 0 0 0c1 .9 2.1 1.9 3.1 2.9c-11.9-6.5 3.2 13.7 11.9 25.2c.8 1.1 1.5 2 2.2 2.9l0 0 0 0 0 0 0 0 0 0c1.4 1.9 2.5 3.4 2.9 4.1l2.4-4.2c-.3 6.1 4.3 13.9 13.1 24.7l7.3-.3c3 6 14 16.7 20.7 17.2l-4.4 5.8c8.1 2.6 10.3 4.3 12.7 6.2c2.6 2.1 5.4 4.3 16.1 8.1l-4.2-7.4c3.5 3 6.2 5.9 8.8 8.7l.1 .1c5.2 5.6 9.9 10.6 19.7 15.3c10.7 3.7 16.6 4.7 22.7 5.8c.3 0 .6 .1 .9 .1c5.4 .8 11.2 1.8 20.8 4.5c-1.1-.1-2.2-.1-3.3-.1h0c-2.3-.1-4.7-.1-7-.1l0 0 0 0 0 0 0 0 0 0 0 0 0 0c-14.4-.2-29.2-.4-42.7-5.2C107.8 480.5 19.5 367.2 26 250.6c-.6-9.9-.3-20.9 0-30.7c.4-13.5 .7-24.8-1.6-28.3l1-3.1c5.3-17.4 11.7-38.2 23.8-62.8l-.1-.2v-.1c.4 .4 3.4 3.4 8.8-5.8c.8-1.8 1.6-3.7 2.4-5.6c.5-1.1 .9-2.2 1.4-3.2c2.5-6.1 5.1-12.3 8.4-17.9l2.6-.6c1.7-10.1 17-23.8 29.8-35.2l1.1-1c5.7-5.1 10.7-9.7 13.6-13.1l.7 4.4c17-15.9 44.6-27.5 65.6-36.4l.5-.2c4.8-2 9.3-3.9 13.3-5.7c-3.4 3.8 2.2 2.7 10 1c4.8-1 10.4-2.1 15.3-2.4l-3.9 2.1c-2.7 1.4-5.4 2.8-8 4.6c8.1-2 11.7-1.4 15.7-.8l.3 0c3.5 .6 7.3 1.2 14.6 .2c-5.6 .8-12.3 3-11.2 3.8c7.9 .9 12.8-.1 17.2-1l.2 0c5.5-1.1 10.3-2 19.3 .9l-1-4.8c7.3 2.6 12.7 4.3 17.5 5.8l.5 .1c10 3 17.6 5.3 34.2 14.1c3.2 .2 5.3-.5 7.4-1.2l.1 0c3.6-1.1 7-2.1 15.2 1.2c.3 .5 .5 1 .7 1.4c.1 .2 .2 .5 .3 .7l0 .1c1 2.6 1.8 4.6 14.6 12.1c1.7-.7-2.7-4.7-6.4-8.2c0 0 0 0-.1-.1c-.2-.1-.3-.3-.5-.4c32.2 17.3 67.3 54.1 78 93.5c-6-11.1-5.2-5.5-4.3 .5c.6 4 1.2 8.1-.2 7.5c4.5 12.1 8.1 24.5 10.4 37.4l-.8-2.9-.1-.3c-3.3-11.9-9.6-34.3-19.9-49.3c-.4 4.3-2.8 3.9-5.2 3.5l-.1 0 0 0c-3.3-.6-6.2-1.1-1.9 12.6c2.6 3.8 3.1 2.4 3.5 1.1l0 0c.5-1.5 .9-2.7 4.7 5.2c.1 4.1 1 8.2 2.1 12.7l0 0 0 0 .1 .6c.1 .3 .1 .5 .2 .8l.1 .6c.6 2.6 1.3 5.4 1.8 8.4c-1.1-.2-2.3-2.2-3.4-4.2c-1.4-2.4-2.8-4.7-3.7-3.2c2.4 11.5 6.5 17.4 8 18.3c-.3 .6-.6 .7-1.1 .7c-.8 0-1.8 .1-1.9 5.3c.7 13.7 3.3 12.5 5.3 11.6l0 0c.6-.3 1.2-.6 1.7-.4c-.6 2.5-1.6 5.1-2.7 7.9c-2.8 7.1-6 15.4-3.4 26.1c-.8-3-2-6-3.1-8.9l-.1-.4c-.2-.5-.4-1-.6-1.5l0 0c-.3-.8-.6-1.6-.9-2.3c-.6 4.4-.3 7.7-.1 10.6c0 .2 0 .5 0 .7c.4 5.3 .7 10-3 19.9c4.3-14.2 3.8-26.9-.2-20.8c1 10.9-3.7 20.4-8 28.9l-.1 .2c-3.6 7.1-6.8 13.5-5.9 19.3l-5.2-7.1c-7.5 10.9-7 13.3-6.5 15.5l0 .1c.5 1.9 1 3.8-3.4 10.8c1.7-2.9 1.3-3.6 1-4.2l0 0c-.4-.8-.7-1.5 1.7-5.1c-1.6 .1-5.5 3.9-10.1 8.5c-3.9 3.9-8.5 8.4-12.8 11.8c-37.5 30.1-82.3 34-125.6 17.8c.2-1-.2-2.1-3.1-4.1c-36.8-28.2-58.5-52.1-50.9-107.5c2.1-1.6 3.6-5.8 5.3-10.8l0 0 0 0 .2-.4 .1-.3 0-.1c2.9-8.4 6.5-18.8 14.3-23.8c7.8-17.3 31.3-33.3 56.4-33.7c25.6-1.4 47.2 13.7 58.1 27.9c-19.8-18.4-52.1-24-79.7-10.4c-28.2 12.7-45 43.8-42.5 74.7c.3-.4 .6-.6 .9-.8l0 0s0 0 0 0c0 0 .1-.1 .1-.1l.1-.1c.6-.5 1.1-.9 1.4-3.3c-.9 60.2 64.8 104.3 112.1 82l.6 1.3c12.7-3.5 15.9-6.5 20.3-10.7l.1-.1 0 0c2.2-2.1 4.7-4.5 8.9-7.3c-.3 .7-1.3 1.7-2.4 2.7c-2.2 2.1-4.6 4.5-1.6 4.6c5-1.3 18.5-13.4 28.5-22.3l0 0 0 0c.6-.5 1.2-1 1.7-1.5c1.5-1.3 2.8-2.5 4-3.6l0 0 .3-.3c1.9-4.2 1.6-5.6 1.3-7l0-.1c-.4-1.6-.8-3.3 2.4-9.6l7.3-3.7c.8-2.1 1.5-4.1 2.2-6c.2-.6 .5-1.2 .7-1.8l-.4-.2zM349.3 34.3l-.2-.1 .2 .1 0 0zM247.8 334.1c-6-3-13.7-8.9-14.8-11.4l-.4 .3c-.3 .6-.5 1.3-.2 2.2c-12.2-5.7-23.4-14.3-32.6-24.9c4.9 7.1 10.1 14.1 17 19.5c-6.9-2.3-15.1-11.8-21.6-19.3l-.1-.1c-4.3-5-7.9-9.1-9.7-9.5c19.8 35.5 80.5 62.3 112.3 49c-14.7 .5-33.4 .3-49.9-5.8zm79.3-119.7l-.1-.2c-.5-1.5-1.1-3.1-1.7-3.4c1.4-5.8 5.4-10.7 4.4 4.6c-1 3.8-1.8 1.5-2.6-1zm-4.2 22.2c-1.3 7.9-5 15.5-10.1 22.5c.2-2-1.2-2.4-2.6-2.8l0 0c-2.9-.8-5.9-1.6 5.6-16.1c-.5 1.9-2.1 4.6-3.7 7.3l0 0 0 0-.3 .4c-3.6 5.9-6.7 11 4 4.3l1-1.8c2.6-4.5 5-8.8 6-13.8h.1zm-55.6 33.9c7.1 .6 14.1 .6 21-1.1c-2.5 2.4-5.2 4.8-8.3 7.2c-11.1-1.7-21.2-6-12.7-6.1zm-92.6 11.6c3.6 7.1 6.4 11.5 9 15.7l.1 .2c2.3 3.7 4.4 7.1 6.8 11.7c-5.1-4.2-8.7-9.5-12.5-15l-.3-.5c-1.4-2.1-2.8-4.2-4.4-6.2l1.2-5.9h.1zm7.5-9.6c1.6 3.3 3.2 6.4 5.7 9.1l2.6 7.7-1.3-2.1c-3.2-5.3-6.3-10.6-8-16.7l.8 1.6 .2 .4zm238.9-41.6c-2.3 17.4-7.7 34.6-16 50.3c7.6-14.9 12.5-30.9 14.8-47.2l1.2-3.1zM35.6 110.6c.4 .8 1.4 .5 2.3 .3c1.9-.5 3.6-.9-.1 7.6c-.5 .3-1 .7-1.5 1l0 0 0 0c-1.4 .9-2.8 1.9-3.9 3c1.9-3.8 3.5-7.4 3.2-11.9zM25.3 152.3c-.7 3.7-1.5 7.9-3.4 13.9c.2-1.9 0-3.5-.2-4.9l0-.1c-.4-3.4-.7-6.3 4.3-12.8c-.3 1.2-.5 2.5-.7 3.8v.1z"],openid:[448,512,[],"f19b","M271.5 432l-68 32C88.5 453.7 0 392.5 0 318.2c0-71.5 82.5-131 191.7-144.3v43c-71.5 12.5-124 53-124 101.3 0 51 58.5 93.3 135.7 103v-340l68-33.2v384zM448 291l-131.3-28.5 36.8-20.7c-19.5-11.5-43.5-20-70-24.8v-43c46.2 5.5 87.7 19.5 120.3 39.3l35-19.8L448 291z"],instalod:[512,512,[],"e081","M153.384,480H387.113L502.554,275.765,204.229,333.211ZM504.726,240.078,387.113,32H155.669L360.23,267.9ZM124.386,48.809,7.274,256,123.236,461.154,225.627,165.561Z"],expeditedssl:[496,512,[],"f23e","M248 43.4C130.6 43.4 35.4 138.6 35.4 256S130.6 468.6 248 468.6 460.6 373.4 460.6 256 365.4 43.4 248 43.4zm-97.4 132.9c0-53.7 43.7-97.4 97.4-97.4s97.4 43.7 97.4 97.4v26.6c0 5-3.9 8.9-8.9 8.9h-17.7c-5 0-8.9-3.9-8.9-8.9v-26.6c0-82.1-124-82.1-124 0v26.6c0 5-3.9 8.9-8.9 8.9h-17.7c-5 0-8.9-3.9-8.9-8.9v-26.6zM389.7 380c0 9.7-8 17.7-17.7 17.7H124c-9.7 0-17.7-8-17.7-17.7V238.3c0-9.7 8-17.7 17.7-17.7h248c9.7 0 17.7 8 17.7 17.7V380zm-248-137.3v132.9c0 2.5-1.9 4.4-4.4 4.4h-8.9c-2.5 0-4.4-1.9-4.4-4.4V242.7c0-2.5 1.9-4.4 4.4-4.4h8.9c2.5 0 4.4 1.9 4.4 4.4zm141.7 48.7c0 13-7.2 24.4-17.7 30.4v31.6c0 5-3.9 8.9-8.9 8.9h-17.7c-5 0-8.9-3.9-8.9-8.9v-31.6c-10.5-6.1-17.7-17.4-17.7-30.4 0-19.7 15.8-35.4 35.4-35.4s35.5 15.8 35.5 35.4zM248 8C111 8 0 119 0 256s111 248 248 248 248-111 248-248S385 8 248 8zm0 478.3C121 486.3 17.7 383 17.7 256S121 25.7 248 25.7 478.3 129 478.3 256 375 486.3 248 486.3z"],sellcast:[448,512,[],"f2da","M353.4 32H94.7C42.6 32 0 74.6 0 126.6v258.7C0 437.4 42.6 480 94.7 480h258.7c52.1 0 94.7-42.6 94.7-94.6V126.6c0-52-42.6-94.6-94.7-94.6zm-50 316.4c-27.9 48.2-89.9 64.9-138.2 37.2-22.9 39.8-54.9 8.6-42.3-13.2l15.7-27.2c5.9-10.3 19.2-13.9 29.5-7.9 18.6 10.8-.1-.1 18.5 10.7 27.6 15.9 63.4 6.3 79.4-21.3 15.9-27.6 6.3-63.4-21.3-79.4-17.8-10.2-.6-.4-18.6-10.6-24.6-14.2-3.4-51.9 21.6-37.5 18.6 10.8-.1-.1 18.5 10.7 48.4 28 65.1 90.3 37.2 138.5zm21.8-208.8c-17 29.5-16.3 28.8-19 31.5-6.5 6.5-16.3 8.7-26.5 3.6-18.6-10.8.1.1-18.5-10.7-27.6-15.9-63.4-6.3-79.4 21.3s-6.3 63.4 21.3 79.4c0 0 18.5 10.6 18.6 10.6 24.6 14.2 3.4 51.9-21.6 37.5-18.6-10.8.1.1-18.5-10.7-48.2-27.8-64.9-90.1-37.1-138.4 27.9-48.2 89.9-64.9 138.2-37.2l4.8-8.4c14.3-24.9 52-3.3 37.7 21.5z"],"square-twitter":[448,512,["twitter-square"],"f081","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM351.3 199.3v0c0 86.7-66 186.6-186.6 186.6c-37.2 0-71.7-10.8-100.7-29.4c5.3 .6 10.4 .8 15.8 .8c30.7 0 58.9-10.4 81.4-28c-28.8-.6-53-19.5-61.3-45.5c10.1 1.5 19.2 1.5 29.6-1.2c-30-6.1-52.5-32.5-52.5-64.4v-.8c8.7 4.9 18.9 7.9 29.6 8.3c-9-6-16.4-14.1-21.5-23.6s-7.8-20.2-7.7-31c0-12.2 3.2-23.4 8.9-33.1c32.3 39.8 80.8 65.8 135.2 68.6c-9.3-44.5 24-80.6 64-80.6c18.9 0 35.9 7.9 47.9 20.7c14.8-2.8 29-8.3 41.6-15.8c-4.9 15.2-15.2 28-28.8 36.1c13.2-1.4 26-5.1 37.8-10.2c-8.9 13.1-20.1 24.7-32.9 34c.2 2.8 .2 5.7 .2 8.5z"],"r-project":[581,512,[],"f4f7","M581 226.6C581 119.1 450.9 32 290.5 32S0 119.1 0 226.6C0 322.4 103.3 402 239.4 418.1V480h99.1v-61.5c24.3-2.7 47.6-7.4 69.4-13.9L448 480h112l-67.4-113.7c54.5-35.4 88.4-84.9 88.4-139.7zm-466.8 14.5c0-73.5 98.9-133 220.8-133s211.9 40.7 211.9 133c0 50.1-26.5 85-70.3 106.4-2.4-1.6-4.7-2.9-6.4-3.7-10.2-5.2-27.8-10.5-27.8-10.5s86.6-6.4 86.6-92.7-90.6-87.9-90.6-87.9h-199V361c-74.1-21.5-125.2-67.1-125.2-119.9zm225.1 38.3v-55.6c57.8 0 87.8-6.8 87.8 27.3 0 36.5-38.2 28.3-87.8 28.3zm-.9 72.5H365c10.8 0 18.9 11.7 24 19.2-16.1 1.9-33 2.8-50.6 2.9v-22.1z"],delicious:[448,512,[],"f1a5","M446.5 68c-.4-1.5-.9-3-1.4-4.5-.9-2.5-2-4.8-3.3-7.1-1.4-2.4-3-4.8-4.7-6.9-2.1-2.5-4.4-4.8-6.9-6.8-1.1-.9-2.2-1.7-3.3-2.5-1.3-.9-2.6-1.7-4-2.4-1.8-1-3.6-1.8-5.5-2.5-1.7-.7-3.5-1.3-5.4-1.7-3.8-1-7.9-1.5-12-1.5H48C21.5 32 0 53.5 0 80v352c0 4.1.5 8.2 1.5 12 2 7.7 5.8 14.6 11 20.3 1 1.1 2.1 2.2 3.3 3.3 5.7 5.2 12.6 9 20.3 11 3.8 1 7.9 1.5 12 1.5h352c26.5 0 48-21.5 48-48V80c-.1-4.1-.6-8.2-1.6-12zM416 432c0 8.8-7.2 16-16 16H224V256H32V80c0-8.8 7.2-16 16-16h176v192h192z"],freebsd:[448,512,[],"f3a4","M303.7 96.2c11.1-11.1 115.5-77 139.2-53.2 23.7 23.7-42.1 128.1-53.2 139.2-11.1 11.1-39.4.9-63.1-22.9-23.8-23.7-34.1-52-22.9-63.1zM109.9 68.1C73.6 47.5 22 24.6 5.6 41.1c-16.6 16.6 7.1 69.4 27.9 105.7 18.5-32.2 44.8-59.3 76.4-78.7zM406.7 174c3.3 11.3 2.7 20.7-2.7 26.1-20.3 20.3-87.5-27-109.3-70.1-18-32.3-11.1-53.4 14.9-48.7 5.7-3.6 12.3-7.6 19.6-11.6-29.8-15.5-63.6-24.3-99.5-24.3-119.1 0-215.6 96.5-215.6 215.6 0 119 96.5 215.6 215.6 215.6S445.3 380.1 445.3 261c0-38.4-10.1-74.5-27.7-105.8-3.9 7-7.6 13.3-10.9 18.8z"],vuejs:[448,512,[],"f41f","M356.9 64.3H280l-56 88.6-48-88.6H0L224 448 448 64.3h-91.1zm-301.2 32h53.8L224 294.5 338.4 96.3h53.8L224 384.5 55.7 96.3z"],accusoft:[640,512,[],"f369","M322.1 252v-1l-51.2-65.8s-12 1.6-25 15.1c-9 9.3-242.1 239.1-243.4 240.9-7 10 1.6 6.8 15.7 1.7.8 0 114.5-36.6 114.5-36.6.5-.6-.1-.1.6-.6-.4-5.1-.8-26.2-1-27.7-.6-5.2 2.2-6.9 7-8.9l92.6-33.8c.6-.8 88.5-81.7 90.2-83.3zm160.1 120.1c13.3 16.1 20.7 13.3 30.8 9.3 3.2-1.2 115.4-47.6 117.8-48.9 8-4.3-1.7-16.7-7.2-23.4-2.1-2.5-205.1-245.6-207.2-248.3-9.7-12.2-14.3-12.9-38.4-12.8-10.2 0-106.8.5-116.5.6-19.2.1-32.9-.3-19.2 16.9C250 75 476.5 365.2 482.2 372.1zm152.7 1.6c-2.3-.3-24.6-4.7-38-7.2 0 0-115 50.4-117.5 51.6-16 7.3-26.9-3.2-36.7-14.6l-57.1-74c-5.4-.9-60.4-9.6-65.3-9.3-3.1.2-9.6.8-14.4 2.9-4.9 2.1-145.2 52.8-150.2 54.7-5.1 2-11.4 3.6-11.1 7.6.2 2.5 2 2.6 4.6 3.5 2.7.8 300.9 67.6 308 69.1 15.6 3.3 38.5 10.5 53.6 1.7 2.1-1.2 123.8-76.4 125.8-77.8 5.4-4 4.3-6.8-1.7-8.2z"],ioxhost:[640,512,[],"f208","M616 160h-67.3C511.2 70.7 422.9 8 320 8 183 8 72 119 72 256c0 16.4 1.6 32.5 4.7 48H24c-13.3 0-24 10.8-24 24 0 13.3 10.7 24 24 24h67.3c37.5 89.3 125.8 152 228.7 152 137 0 248-111 248-248 0-16.4-1.6-32.5-4.7-48H616c13.3 0 24-10.8 24-24 0-13.3-10.7-24-24-24zm-96 96c0 110.5-89.5 200-200 200-75.7 0-141.6-42-175.5-104H424c13.3 0 24-10.8 24-24 0-13.3-10.7-24-24-24H125.8c-3.8-15.4-5.8-31.4-5.8-48 0-110.5 89.5-200 200-200 75.7 0 141.6 42 175.5 104H216c-13.3 0-24 10.8-24 24 0 13.3 10.7 24 24 24h298.2c3.8 15.4 5.8 31.4 5.8 48zm-304-24h208c13.3 0 24 10.7 24 24 0 13.2-10.7 24-24 24H216c-13.3 0-24-10.7-24-24 0-13.2 10.7-24 24-24z"],"fonticons-fi":[384,512,[],"f3a2","M114.4 224h92.4l-15.2 51.2h-76.4V433c0 8-2.8 9.2 4.4 10l59.6 5.6V483H0v-35.2l29.2-2.8c7.2-.8 9.2-3.2 9.2-10.8V278.4c0-3.2-4-3.2-8-3.2H0V224h38.4v-28.8c0-68 36.4-96 106-96 46.8 0 88.8 11.2 88.8 72.4l-69.6 8.4c.4-25.6-6-31.6-22.4-31.6-25.2 0-26 13.6-26 37.6v32c0 3.2-4.8 6-.8 6zM384 483H243.2v-34.4l28-3.6c7.2-.8 10.4-2.4 10.4-10V287c0-5.6-4-9.2-9.2-10.8l-33.2-8.8 9.2-40.4h110v208c0 8-3.6 8.8 4 10l21.6 3.6V483zm-30-347.2l12.4 45.6-10 10-42.8-22.8-42.8 22.8-10-10 12.4-45.6-30-36.4 4.8-10h38L307.2 51H320l21.2 38.4h38l4.8 13.2-30 33.2z"],"app-store":[512,512,[],"f36f","M255.9 120.9l9.1-15.7c5.6-9.8 18.1-13.1 27.9-7.5 9.8 5.6 13.1 18.1 7.5 27.9l-87.5 151.5h63.3c20.5 0 32 24.1 23.1 40.8H113.8c-11.3 0-20.4-9.1-20.4-20.4 0-11.3 9.1-20.4 20.4-20.4h52l66.6-115.4-20.8-36.1c-5.6-9.8-2.3-22.2 7.5-27.9 9.8-5.6 22.2-2.3 27.9 7.5l8.9 15.7zm-78.7 218l-19.6 34c-5.6 9.8-18.1 13.1-27.9 7.5-9.8-5.6-13.1-18.1-7.5-27.9l14.6-25.2c16.4-5.1 29.8-1.2 40.4 11.6zm168.9-61.7h53.1c11.3 0 20.4 9.1 20.4 20.4 0 11.3-9.1 20.4-20.4 20.4h-29.5l19.9 34.5c5.6 9.8 2.3 22.2-7.5 27.9-9.8 5.6-22.2 2.3-27.9-7.5-33.5-58.1-58.7-101.6-75.4-130.6-17.1-29.5-4.9-59.1 7.2-69.1 13.4 23 33.4 57.7 60.1 104zM256 8C119 8 8 119 8 256s111 248 248 248 248-111 248-248S393 8 256 8zm216 248c0 118.7-96.1 216-216 216-118.7 0-216-96.1-216-216 0-118.7 96.1-216 216-216 118.7 0 216 96.1 216 216z"],"cc-mastercard":[576,512,[],"f1f1","M482.9 410.3c0 6.8-4.6 11.7-11.2 11.7-6.8 0-11.2-5.2-11.2-11.7 0-6.5 4.4-11.7 11.2-11.7 6.6 0 11.2 5.2 11.2 11.7zm-310.8-11.7c-7.1 0-11.2 5.2-11.2 11.7 0 6.5 4.1 11.7 11.2 11.7 6.5 0 10.9-4.9 10.9-11.7-.1-6.5-4.4-11.7-10.9-11.7zm117.5-.3c-5.4 0-8.7 3.5-9.5 8.7h19.1c-.9-5.7-4.4-8.7-9.6-8.7zm107.8.3c-6.8 0-10.9 5.2-10.9 11.7 0 6.5 4.1 11.7 10.9 11.7 6.8 0 11.2-4.9 11.2-11.7 0-6.5-4.4-11.7-11.2-11.7zm105.9 26.1c0 .3.3.5.3 1.1 0 .3-.3.5-.3 1.1-.3.3-.3.5-.5.8-.3.3-.5.5-1.1.5-.3.3-.5.3-1.1.3-.3 0-.5 0-1.1-.3-.3 0-.5-.3-.8-.5-.3-.3-.5-.5-.5-.8-.3-.5-.3-.8-.3-1.1 0-.5 0-.8.3-1.1 0-.5.3-.8.5-1.1.3-.3.5-.3.8-.5.5-.3.8-.3 1.1-.3.5 0 .8 0 1.1.3.5.3.8.3 1.1.5s.2.6.5 1.1zm-2.2 1.4c.5 0 .5-.3.8-.3.3-.3.3-.5.3-.8 0-.3 0-.5-.3-.8-.3 0-.5-.3-1.1-.3h-1.6v3.5h.8V426h.3l1.1 1.4h.8l-1.1-1.3zM576 81v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V81c0-26.5 21.5-48 48-48h480c26.5 0 48 21.5 48 48zM64 220.6c0 76.5 62.1 138.5 138.5 138.5 27.2 0 53.9-8.2 76.5-23.1-72.9-59.3-72.4-171.2 0-230.5-22.6-15-49.3-23.1-76.5-23.1-76.4-.1-138.5 62-138.5 138.2zm224 108.8c70.5-55 70.2-162.2 0-217.5-70.2 55.3-70.5 162.6 0 217.5zm-142.3 76.3c0-8.7-5.7-14.4-14.7-14.7-4.6 0-9.5 1.4-12.8 6.5-2.4-4.1-6.5-6.5-12.2-6.5-3.8 0-7.6 1.4-10.6 5.4V392h-8.2v36.7h8.2c0-18.9-2.5-30.2 9-30.2 10.2 0 8.2 10.2 8.2 30.2h7.9c0-18.3-2.5-30.2 9-30.2 10.2 0 8.2 10 8.2 30.2h8.2v-23zm44.9-13.7h-7.9v4.4c-2.7-3.3-6.5-5.4-11.7-5.4-10.3 0-18.2 8.2-18.2 19.3 0 11.2 7.9 19.3 18.2 19.3 5.2 0 9-1.9 11.7-5.4v4.6h7.9V392zm40.5 25.6c0-15-22.9-8.2-22.9-15.2 0-5.7 11.9-4.8 18.5-1.1l3.3-6.5c-9.4-6.1-30.2-6-30.2 8.2 0 14.3 22.9 8.3 22.9 15 0 6.3-13.5 5.8-20.7.8l-3.5 6.3c11.2 7.6 32.6 6 32.6-7.5zm35.4 9.3l-2.2-6.8c-3.8 2.1-12.2 4.4-12.2-4.1v-16.6h13.1V392h-13.1v-11.2h-8.2V392h-7.6v7.3h7.6V416c0 17.6 17.3 14.4 22.6 10.9zm13.3-13.4h27.5c0-16.2-7.4-22.6-17.4-22.6-10.6 0-18.2 7.9-18.2 19.3 0 20.5 22.6 23.9 33.8 14.2l-3.8-6c-7.8 6.4-19.6 5.8-21.9-4.9zm59.1-21.5c-4.6-2-11.6-1.8-15.2 4.4V392h-8.2v36.7h8.2V408c0-11.6 9.5-10.1 12.8-8.4l2.4-7.6zm10.6 18.3c0-11.4 11.6-15.1 20.7-8.4l3.8-6.5c-11.6-9.1-32.7-4.1-32.7 15 0 19.8 22.4 23.8 32.7 15l-3.8-6.5c-9.2 6.5-20.7 2.6-20.7-8.6zm66.7-18.3H408v4.4c-8.3-11-29.9-4.8-29.9 13.9 0 19.2 22.4 24.7 29.9 13.9v4.6h8.2V392zm33.7 0c-2.4-1.2-11-2.9-15.2 4.4V392h-7.9v36.7h7.9V408c0-11 9-10.3 12.8-8.4l2.4-7.6zm40.3-14.9h-7.9v19.3c-8.2-10.9-29.9-5.1-29.9 13.9 0 19.4 22.5 24.6 29.9 13.9v4.6h7.9v-51.7zm7.6-75.1v4.6h.8V302h1.9v-.8h-4.6v.8h1.9zm6.6 123.8c0-.5 0-1.1-.3-1.6-.3-.3-.5-.8-.8-1.1-.3-.3-.8-.5-1.1-.8-.5 0-1.1-.3-1.6-.3-.3 0-.8.3-1.4.3-.5.3-.8.5-1.1.8-.5.3-.8.8-.8 1.1-.3.5-.3 1.1-.3 1.6 0 .3 0 .8.3 1.4 0 .3.3.8.8 1.1.3.3.5.5 1.1.8.5.3 1.1.3 1.4.3.5 0 1.1 0 1.6-.3.3-.3.8-.5 1.1-.8.3-.3.5-.8.8-1.1.3-.6.3-1.1.3-1.4zm3.2-124.7h-1.4l-1.6 3.5-1.6-3.5h-1.4v5.4h.8v-4.1l1.6 3.5h1.1l1.4-3.5v4.1h1.1v-5.4zm4.4-80.5c0-76.2-62.1-138.3-138.5-138.3-27.2 0-53.9 8.2-76.5 23.1 72.1 59.3 73.2 171.5 0 230.5 22.6 15 49.5 23.1 76.5 23.1 76.4.1 138.5-61.9 138.5-138.4z"],"itunes-note":[384,512,[],"f3b5","M381.9 388.2c-6.4 27.4-27.2 42.8-55.1 48-24.5 4.5-44.9 5.6-64.5-10.2-23.9-20.1-24.2-53.4-2.7-74.4 17-16.2 40.9-19.5 76.8-25.8 6-1.1 11.2-2.5 15.6-7.4 6.4-7.2 4.4-4.1 4.4-163.2 0-11.2-5.5-14.3-17-12.3-8.2 1.4-185.7 34.6-185.7 34.6-10.2 2.2-13.4 5.2-13.4 16.7 0 234.7 1.1 223.9-2.5 239.5-4.2 18.2-15.4 31.9-30.2 39.5-16.8 9.3-47.2 13.4-63.4 10.4-43.2-8.1-58.4-58-29.1-86.6 17-16.2 40.9-19.5 76.8-25.8 6-1.1 11.2-2.5 15.6-7.4 10.1-11.5 1.8-256.6 5.2-270.2.8-5.2 3-9.6 7.1-12.9 4.2-3.5 11.8-5.5 13.4-5.5 204-38.2 228.9-43.1 232.4-43.1 11.5-.8 18.1 6 18.1 17.6.2 344.5 1.1 326-1.8 338.5z"],golang:[640,512,[],"e40f","M400.1 194.8C389.2 197.6 380.2 199.1 371 202.4C363.7 204.3 356.3 206.3 347.8 208.5L347.2 208.6C343 209.8 342.6 209.9 338.7 205.4C334 200.1 330.6 196.7 324.1 193.5C304.4 183.9 285.4 186.7 267.7 198.2C246.5 211.9 235.6 232.2 235.9 257.4C236.2 282.4 253.3 302.9 277.1 306.3C299.1 309.1 316.9 301.7 330.9 285.8C333 283.2 334.9 280.5 337 277.5V277.5L337 277.5C337.8 276.5 338.5 275.4 339.3 274.2H279.2C272.7 274.2 271.1 270.2 273.3 264.9C277.3 255.2 284.8 239 289.2 230.9C290.1 229.1 292.3 225.1 296.1 225.1H397.2C401.7 211.7 409 198.2 418.8 185.4C441.5 155.5 468.1 139.9 506 133.4C537.8 127.8 567.7 130.9 594.9 149.3C619.5 166.1 634.7 188.9 638.8 218.8C644.1 260.9 631.9 295.1 602.1 324.4C582.4 345.3 557.2 358.4 528.2 364.3C522.6 365.3 517.1 365.8 511.7 366.3C508.8 366.5 506 366.8 503.2 367.1C474.9 366.5 449 358.4 427.2 339.7C411.9 326.4 401.3 310.1 396.1 291.2C392.4 298.5 388.1 305.6 382.1 312.3C360.5 341.9 331.2 360.3 294.2 365.2C263.6 369.3 235.3 363.4 210.3 344.7C187.3 327.2 174.2 304.2 170.8 275.5C166.7 241.5 176.7 210.1 197.2 184.2C219.4 155.2 248.7 136.8 284.5 130.3C313.8 124.1 341.8 128.4 367.1 145.6C383.6 156.5 395.4 171.4 403.2 189.5C405.1 192.3 403.8 193.9 400.1 194.8zM48.3 200.4C47.05 200.4 46.74 199.8 47.36 198.8L53.91 190.4C54.53 189.5 56.09 188.9 57.34 188.9H168.6C169.8 188.9 170.1 189.8 169.5 190.7L164.2 198.8C163.6 199.8 162 200.7 161.1 200.7L48.3 200.4zM1.246 229.1C0 229.1-.3116 228.4 .3116 227.5L6.855 219.1C7.479 218.2 9.037 217.5 10.28 217.5H152.4C153.6 217.5 154.2 218.5 153.9 219.4L151.4 226.9C151.1 228.1 149.9 228.8 148.6 228.8L1.246 229.1zM75.72 255.9C75.1 256.8 75.41 257.7 76.65 257.7L144.6 258C145.5 258 146.8 257.1 146.8 255.9L147.4 248.4C147.4 247.1 146.8 246.2 145.5 246.2H83.2C81.95 246.2 80.71 247.1 80.08 248.1L75.72 255.9zM577.2 237.9C577 235.3 576.9 233.1 576.5 230.9C570.9 200.1 542.5 182.6 512.9 189.5C483.9 196 465.2 214.4 458.4 243.7C452.8 268 464.6 292.6 487 302.6C504.2 310.1 521.3 309.2 537.8 300.7C562.4 287.1 575.8 268 577.4 241.2C577.3 240 577.3 238.9 577.2 237.9z"],kickstarter:[448,512,["square-kickstarter"],"f3bb","M64 32C28.7 32 0 60.7 0 96L0 416c0 35.3 28.7 64 64 64l320 0c35.3 0 64-28.7 64-64l0-320c0-35.3-28.7-64-64-64L64 32zM320.8 233l-23.3 23.1L320.8 279c24.1 23.9 24.1 63 0 86.9s-63.4 23.9-87.6 0l-8.5-8.4c-11.3 16-29.7 26.5-50.9 26.5c-34.1 0-61.9-27.5-61.9-61.4l0-133.2c0-33.8 27.7-61.4 61.9-61.4c21.1 0 39.6 10.5 50.9 26.5l8.5-8.4c24.1-23.9 63.4-23.9 87.6 0s24.1 63 0 86.9z"],grav:[512,512,[],"f2d6","M301.1 212c4.4 4.4 4.4 11.9 0 16.3l-9.7 9.7c-4.4 4.7-11.9 4.7-16.6 0l-10.5-10.5c-4.4-4.7-4.4-11.9 0-16.6l9.7-9.7c4.4-4.4 11.9-4.4 16.6 0l10.5 10.8zm-30.2-19.7c3-3 3-7.8 0-10.5-2.8-3-7.5-3-10.5 0-2.8 2.8-2.8 7.5 0 10.5 3.1 2.8 7.8 2.8 10.5 0zm-26 5.3c-3 2.8-3 7.5 0 10.2 2.8 3 7.5 3 10.5 0 2.8-2.8 2.8-7.5 0-10.2-3-3-7.7-3-10.5 0zm72.5-13.3c-19.9-14.4-33.8-43.2-11.9-68.1 21.6-24.9 40.7-17.2 59.8.8 11.9 11.3 29.3 24.9 17.2 48.2-12.5 23.5-45.1 33.2-65.1 19.1zm47.7-44.5c-8.9-10-23.3 6.9-15.5 16.1 7.4 9 32.1 2.4 15.5-16.1zM504 256c0 137-111 248-248 248S8 393 8 256 119 8 256 8s248 111 248 248zm-66.2 42.6c2.5-16.1-20.2-16.6-25.2-25.7-13.6-24.1-27.7-36.8-54.5-30.4 11.6-8 23.5-6.1 23.5-6.1.3-6.4 0-13-9.4-24.9 3.9-12.5.3-22.4.3-22.4 15.5-8.6 26.8-24.4 29.1-43.2 3.6-31-18.8-59.2-49.8-62.8-22.1-2.5-43.7 7.7-54.3 25.7-23.2 40.1 1.4 70.9 22.4 81.4-14.4-1.4-34.3-11.9-40.1-34.3-6.6-25.7 2.8-49.8 8.9-61.4 0 0-4.4-5.8-8-8.9 0 0-13.8 0-24.6 5.3 11.9-15.2 25.2-14.4 25.2-14.4 0-6.4-.6-14.9-3.6-21.6-5.4-11-23.8-12.9-31.7 2.8.1-.2.3-.4.4-.5-5 11.9-1.1 55.9 16.9 87.2-2.5 1.4-9.1 6.1-13 10-21.6 9.7-56.2 60.3-56.2 60.3-28.2 10.8-77.2 50.9-70.6 79.7.3 3 1.4 5.5 3 7.5-2.8 2.2-5.5 5-8.3 8.3-11.9 13.8-5.3 35.2 17.7 24.4 15.8-7.2 29.6-20.2 36.3-30.4 0 0-5.5-5-16.3-4.4 27.7-6.6 34.3-9.4 46.2-9.1 8 3.9 8-34.3 8-34.3 0-14.7-2.2-31-11.1-41.5 12.5 12.2 29.1 32.7 28 60.6-.8 18.3-15.2 23-15.2 23-9.1 16.6-43.2 65.9-30.4 106 0 0-9.7-14.9-10.2-22.1-17.4 19.4-46.5 52.3-24.6 64.5 26.6 14.7 108.8-88.6 126.2-142.3 34.6-20.8 55.4-47.3 63.9-65 22 43.5 95.3 94.5 101.1 59z"],weibo:[512,512,[],"f18a","M407 177.6c7.6-24-13.4-46.8-37.4-41.7-22 4.8-28.8-28.1-7.1-32.8 50.1-10.9 92.3 37.1 76.5 84.8-6.8 21.2-38.8 10.8-32-10.3zM214.8 446.7C108.5 446.7 0 395.3 0 310.4c0-44.3 28-95.4 76.3-143.7C176 67 279.5 65.8 249.9 161c-4 13.1 12.3 5.7 12.3 6 79.5-33.6 140.5-16.8 114 51.4-3.7 9.4 1.1 10.9 8.3 13.1 135.7 42.3 34.8 215.2-169.7 215.2zm143.7-146.3c-5.4-55.7-78.5-94-163.4-85.7-84.8 8.6-148.8 60.3-143.4 116s78.5 94 163.4 85.7c84.8-8.6 148.8-60.3 143.4-116zM347.9 35.1c-25.9 5.6-16.8 43.7 8.3 38.3 72.3-15.2 134.8 52.8 111.7 124-7.4 24.2 29.1 37 37.4 12 31.9-99.8-55.1-195.9-157.4-174.3zm-78.5 311c-17.1 38.8-66.8 60-109.1 46.3-40.8-13.1-58-53.4-40.3-89.7 17.7-35.4 63.1-55.4 103.4-45.1 42 10.8 63.1 50.2 46 88.5zm-86.3-30c-12.9-5.4-30 .3-38 12.9-8.3 12.9-4.3 28 8.6 34 13.1 6 30.8.3 39.1-12.9 8-13.1 3.7-28.3-9.7-34zm32.6-13.4c-5.1-1.7-11.4.6-14.3 5.4-2.9 5.1-1.4 10.6 3.7 12.9 5.1 2 11.7-.3 14.6-5.4 2.8-5.2 1.1-10.9-4-12.9z"],uncharted:[448,512,[],"e084","M171.73,232.813A5.381,5.381,0,0,0,176.7,229.5,48.081,48.081,0,0,1,191.6,204.244c1.243-.828,1.657-2.484,1.657-4.141a4.22,4.22,0,0,0-2.071-3.312L74.429,128.473,148.958,85a9.941,9.941,0,0,0,4.968-8.281,9.108,9.108,0,0,0-4.968-8.281L126.6,55.6a9.748,9.748,0,0,0-9.523,0l-100.2,57.966a9.943,9.943,0,0,0-4.969,8.281V236.954a9.109,9.109,0,0,0,4.969,8.281L39.235,258.07a8.829,8.829,0,0,0,4.968,1.242,9.4,9.4,0,0,0,6.625-2.484,10.8,10.8,0,0,0,2.9-7.039V164.5L169.66,232.4A4.5,4.5,0,0,0,171.73,232.813ZM323.272,377.73a12.478,12.478,0,0,0-4.969,1.242l-74.528,43.062V287.882c0-2.9-2.9-5.8-6.211-4.555a53.036,53.036,0,0,1-28.984.414,4.86,4.86,0,0,0-6.21,4.555V421.619l-74.529-43.061a8.83,8.83,0,0,0-4.969-1.242,9.631,9.631,0,0,0-9.523,9.523v26.085a9.107,9.107,0,0,0,4.969,8.281l100.2,57.553A8.829,8.829,0,0,0,223.486,480a11.027,11.027,0,0,0,4.969-1.242l100.2-57.553a9.941,9.941,0,0,0,4.968-8.281V386.839C332.8,382.285,328.24,377.73,323.272,377.73ZM286.007,78a23,23,0,1,0-23-23A23,23,0,0,0,286.007,78Zm63.627-10.086a23,23,0,1,0,23,23A23,23,0,0,0,349.634,67.914ZM412.816,151.6a23,23,0,1,0-23-23A23,23,0,0,0,412.816,151.6Zm-63.182-9.2a23,23,0,1,0,23,23A23,23,0,0,0,349.634,142.4Zm-63.627,83.244a23,23,0,1,0-23-23A23,23,0,0,0,286.007,225.648Zm-62.074,36.358a23,23,0,1,0-23-23A23,23,0,0,0,223.933,262.006Zm188.883-82.358a23,23,0,1,0,23,23A23,23,0,0,0,412.816,179.648Zm0,72.272a23,23,0,1,0,23,23A23,23,0,0,0,412.816,251.92Z"],firstdraft:[384,512,[],"f3a1","M384 192h-64v128H192v128H0v-25.6h166.4v-128h128v-128H384V192zm-25.6 38.4v128h-128v128H64V512h192V384h128V230.4h-25.6zm25.6 192h-89.6V512H320v-64h64v-25.6zM0 0v384h128V256h128V128h128V0H0z"],"square-youtube":[448,512,[61798,"youtube-square"],"f431","M282 256.2l-95.2-54.1V310.3L282 256.2zM384 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64zm14.4 136.1c7.6 28.6 7.6 88.2 7.6 88.2s0 59.6-7.6 88.1c-4.2 15.8-16.5 27.7-32.2 31.9C337.9 384 224 384 224 384s-113.9 0-142.2-7.6c-15.7-4.2-28-16.1-32.2-31.9C42 315.9 42 256.3 42 256.3s0-59.7 7.6-88.2c4.2-15.8 16.5-28.2 32.2-32.4C110.1 128 224 128 224 128s113.9 0 142.2 7.7c15.7 4.2 28 16.6 32.2 32.4z"],"wikipedia-w":[640,512,[],"f266","M640 51.2l-.3 12.2c-28.1.8-45 15.8-55.8 40.3-25 57.8-103.3 240-155.3 358.6H415l-81.9-193.1c-32.5 63.6-68.3 130-99.2 193.1-.3.3-15 0-15-.3C172 352.3 122.8 243.4 75.8 133.4 64.4 106.7 26.4 63.4.2 63.7c0-3.1-.3-10-.3-14.2h161.9v13.9c-19.2 1.1-52.8 13.3-43.3 34.2 21.9 49.7 103.6 240.3 125.6 288.6 15-29.7 57.8-109.2 75.3-142.8-13.9-28.3-58.6-133.9-72.8-160-9.7-17.8-36.1-19.4-55.8-19.7V49.8l142.5.3v13.1c-19.4.6-38.1 7.8-29.4 26.1 18.9 40 30.6 68.1 48.1 104.7 5.6-10.8 34.7-69.4 48.1-100.8 8.9-20.6-3.9-28.6-38.6-29.4.3-3.6 0-10.3.3-13.6 44.4-.3 111.1-.3 123.1-.6v13.6c-22.5.8-45.8 12.8-58.1 31.7l-59.2 122.8c6.4 16.1 63.3 142.8 69.2 156.7L559.2 91.8c-8.6-23.1-36.4-28.1-47.2-28.3V49.6l127.8 1.1.2.5z"],wpressr:[496,512,["rendact"],"f3e4","M248 8C111.03 8 0 119.03 0 256s111.03 248 248 248 248-111.03 248-248S384.97 8 248 8zm171.33 158.6c-15.18 34.51-30.37 69.02-45.63 103.5-2.44 5.51-6.89 8.24-12.97 8.24-23.02-.01-46.03.06-69.05-.05-5.12-.03-8.25 1.89-10.34 6.72-10.19 23.56-20.63 47-30.95 70.5-1.54 3.51-4.06 5.29-7.92 5.29-45.94-.01-91.87-.02-137.81 0-3.13 0-5.63-1.15-7.72-3.45-11.21-12.33-22.46-24.63-33.68-36.94-2.69-2.95-2.79-6.18-1.21-9.73 8.66-19.54 17.27-39.1 25.89-58.66 12.93-29.35 25.89-58.69 38.75-88.08 1.7-3.88 4.28-5.68 8.54-5.65 14.24.1 28.48.02 42.72.05 6.24.01 9.2 4.84 6.66 10.59-13.6 30.77-27.17 61.55-40.74 92.33-5.72 12.99-11.42 25.99-17.09 39-3.91 8.95 7.08 11.97 10.95 5.6.23-.37-1.42 4.18 30.01-67.69 1.36-3.1 3.41-4.4 6.77-4.39 15.21.08 30.43.02 45.64.04 5.56.01 7.91 3.64 5.66 8.75-8.33 18.96-16.71 37.9-24.98 56.89-4.98 11.43 8.08 12.49 11.28 5.33.04-.08 27.89-63.33 32.19-73.16 2.02-4.61 5.44-6.51 10.35-6.5 26.43.05 52.86 0 79.29.05 12.44.02 13.93-13.65 3.9-13.64-25.26.03-50.52.02-75.78.02-6.27 0-7.84-2.47-5.27-8.27 5.78-13.06 11.59-26.11 17.3-39.21 1.73-3.96 4.52-5.79 8.84-5.78 23.09.06 25.98.02 130.78.03 6.08-.01 8.03 2.79 5.62 8.27z"],angellist:[448,512,[],"f209","M347.1 215.4c11.7-32.6 45.4-126.9 45.4-157.1 0-26.6-15.7-48.9-43.7-48.9-44.6 0-84.6 131.7-97.1 163.1C242 144 196.6 0 156.6 0c-31.1 0-45.7 22.9-45.7 51.7 0 35.3 34.2 126.8 46.6 162-6.3-2.3-13.1-4.3-20-4.3-23.4 0-48.3 29.1-48.3 52.6 0 8.9 4.9 21.4 8 29.7-36.9 10-51.1 34.6-51.1 71.7C46 435.6 114.4 512 210.6 512c118 0 191.4-88.6 191.4-202.9 0-43.1-6.9-82-54.9-93.7zM311.7 108c4-12.3 21.1-64.3 37.1-64.3 8.6 0 10.9 8.9 10.9 16 0 19.1-38.6 124.6-47.1 148l-34-6 33.1-93.7zM142.3 48.3c0-11.9 14.5-45.7 46.3 47.1l34.6 100.3c-15.6-1.3-27.7-3-35.4 1.4-10.9-28.8-45.5-119.7-45.5-148.8zM140 244c29.3 0 67.1 94.6 67.1 107.4 0 5.1-4.9 11.4-10.6 11.4-20.9 0-76.9-76.9-76.9-97.7.1-7.7 12.7-21.1 20.4-21.1zm184.3 186.3c-29.1 32-66.3 48.6-109.7 48.6-59.4 0-106.3-32.6-128.9-88.3-17.1-43.4 3.8-68.3 20.6-68.3 11.4 0 54.3 60.3 54.3 73.1 0 4.9-7.7 8.3-11.7 8.3-16.1 0-22.4-15.5-51.1-51.4-29.7 29.7 20.5 86.9 58.3 86.9 26.1 0 43.1-24.2 38-42 3.7 0 8.3.3 11.7-.6 1.1 27.1 9.1 59.4 41.7 61.7 0-.9 2-7.1 2-7.4 0-17.4-10.6-32.6-10.6-50.3 0-28.3 21.7-55.7 43.7-71.7 8-6 17.7-9.7 27.1-13.1 9.7-3.7 20-8 27.4-15.4-1.1-11.2-5.7-21.1-16.9-21.1-27.7 0-120.6 4-120.6-39.7 0-6.7.1-13.1 17.4-13.1 32.3 0 114.3 8 138.3 29.1 18.1 16.1 24.3 113.2-31 174.7zm-98.6-126c9.7 3.1 19.7 4 29.7 6-7.4 5.4-14 12-20.3 19.1-2.8-8.5-6.2-16.8-9.4-25.1z"],"galactic-republic":[496,512,[],"f50c","M248 504C111.25 504 0 392.75 0 256S111.25 8 248 8s248 111.25 248 248-111.25 248-248 248zm0-479.47C120.37 24.53 16.53 128.37 16.53 256S120.37 487.47 248 487.47 479.47 383.63 479.47 256 375.63 24.53 248 24.53zm27.62 21.81v24.62a185.933 185.933 0 0 1 83.57 34.54l17.39-17.36c-28.75-22.06-63.3-36.89-100.96-41.8zm-55.37.07c-37.64 4.94-72.16 19.8-100.88 41.85l17.28 17.36h.08c24.07-17.84 52.55-30.06 83.52-34.67V46.41zm12.25 50.17v82.87c-10.04 2.03-19.42 5.94-27.67 11.42l-58.62-58.59-21.93 21.93 58.67 58.67c-5.47 8.23-9.45 17.59-11.47 27.62h-82.9v31h82.9c2.02 10.02 6.01 19.31 11.47 27.54l-58.67 58.69 21.93 21.93 58.62-58.62a77.873 77.873 0 0 0 27.67 11.47v82.9h31v-82.9c10.05-2.03 19.37-6.06 27.62-11.55l58.67 58.69 21.93-21.93-58.67-58.69c5.46-8.23 9.47-17.52 11.5-27.54h82.87v-31h-82.87c-2.02-10.02-6.03-19.38-11.5-27.62l58.67-58.67-21.93-21.93-58.67 58.67c-8.25-5.49-17.57-9.47-27.62-11.5V96.58h-31zm183.24 30.72l-17.36 17.36a186.337 186.337 0 0 1 34.67 83.67h24.62c-4.95-37.69-19.83-72.29-41.93-101.03zm-335.55.13c-22.06 28.72-36.91 63.26-41.85 100.91h24.65c4.6-30.96 16.76-59.45 34.59-83.52l-17.39-17.39zM38.34 283.67c4.92 37.64 19.75 72.18 41.8 100.9l17.36-17.39c-17.81-24.07-29.92-52.57-34.51-83.52H38.34zm394.7 0c-4.61 30.99-16.8 59.5-34.67 83.6l17.36 17.36c22.08-28.74 36.98-63.29 41.93-100.96h-24.62zM136.66 406.38l-17.36 17.36c28.73 22.09 63.3 36.98 100.96 41.93v-24.64c-30.99-4.63-59.53-16.79-83.6-34.65zm222.53.05c-24.09 17.84-52.58 30.08-83.57 34.67v24.57c37.67-4.92 72.21-19.79 100.96-41.85l-17.31-17.39h-.08z"],"nfc-directional":[512,512,[],"e530","M211.8 488.6C213.4 491.1 213.9 494.2 213.2 497.1C212.6 500 210.8 502.6 208.3 504.2C205.7 505.8 202.7 506.3 199.7 505.7C138.3 491.8 84.1 455.8 47.53 404.5C10.97 353.2-5.395 290.3 1.57 227.7C8.536 165 38.34 107.2 85.29 65.21C132.2 23.2 193-.0131 256 0C257.5 0 258.1 .2931 260.3 .8627C261.7 1.432 262.1 2.267 264 3.319C265.1 4.371 265.9 5.619 266.5 6.993C267 8.367 267.3 9.839 267.3 11.32V112.3L291.8 86.39C292.8 85.31 294 84.44 295.4 83.84C296.7 83.23 298.2 82.9 299.7 82.86C301.2 82.81 302.6 83.06 304 83.59C305.4 84.12 306.7 84.92 307.8 85.94C308.8 86.96 309.7 88.18 310.3 89.54C310.9 90.89 311.3 92.35 311.3 93.84C311.3 95.32 311.1 96.8 310.6 98.18C310 99.57 309.2 100.8 308.2 101.9L264.2 148.5C263.1 149.6 261.9 150.5 260.5 151.1C259 151.7 257.5 152 255.1 152C254.5 152 252.9 151.7 251.5 151.1C250.1 150.5 248.8 149.6 247.8 148.5L203.7 101.9C201.7 99.74 200.6 96.83 200.7 93.84C200.7 90.84 202 87.1 204.2 85.94C206.4 83.88 209.3 82.77 212.3 82.86C215.3 82.94 218.1 84.21 220.2 86.39L244.7 112.4V22.89C188.3 25.64 134.9 48.73 94.23 87.87C53.58 127 28.49 179.6 23.61 235.8C18.73 292 34.38 348.1 67.68 393.7C100.1 439.2 149.7 471.2 204.7 483.6C207.6 484.3 210.2 486.1 211.8 488.6L211.8 488.6zM171.4 126.1C170.6 127.4 169.5 128.5 168.3 129.3C147.8 143.2 131.1 161.9 119.5 183.8C107.9 205.7 101.8 230.1 101.8 254.9C101.8 279.7 107.9 304.1 119.5 325.1C131.1 347.9 147.8 366.6 168.3 380.5C170.8 382.2 172.5 384.8 173 387.8C173.6 390.7 172.1 393.8 171.3 396.2C169.6 398.7 166.1 400.4 164 400.1C161.1 401.5 158 400.9 155.6 399.2C132 383.2 112.8 361.7 99.46 336.5C86.15 311.4 79.19 283.4 79.19 254.9C79.19 226.5 86.15 198.4 99.46 173.3C112.8 148.1 132 126.6 155.6 110.6C156.8 109.8 158.2 109.2 159.6 108.8C161.1 108.5 162.6 108.5 164.1 108.8C165.5 109 166.9 109.6 168.2 110.4C169.5 111.2 170.5 112.3 171.4 113.5C172.2 114.7 172.8 116.1 173.1 117.6C173.4 119.1 173.4 120.6 173.1 122C172.8 123.5 172.3 124.9 171.4 126.1H171.4zM340.9 383.5C341.7 382.3 342.8 381.2 343.1 380.4V380.3C364.4 366.3 381.1 347.6 392.7 325.7C404.2 303.9 410.2 279.5 410.2 254.8C410.2 230.1 404.2 205.7 392.7 183.8C381.1 161.1 364.4 143.3 343.1 129.3C342.8 128.5 341.7 127.4 340.9 126.2C340.1 124.9 339.5 123.5 339.3 122.1C338.1 120.6 339 119.1 339.3 117.7C339.6 116.2 340.2 114.8 341 113.6C341.9 112.4 342.1 111.3 344.2 110.5C345.4 109.7 346.8 109.2 348.3 108.9C349.8 108.6 351.2 108.6 352.7 108.9C354.2 109.2 355.5 109.8 356.8 110.7C380.2 126.7 399.5 148.2 412.7 173.3C426 198.4 432.1 226.4 432.1 254.8C432.1 283.3 426 311.3 412.7 336.4C399.5 361.5 380.2 383 356.8 399C355.5 399.9 354.2 400.5 352.7 400.8C351.2 401.1 349.8 401.1 348.3 400.8C346.8 400.5 345.4 399.1 344.2 399.2C342.1 398.4 341.9 397.3 341 396.1C340.2 394.9 339.6 393.5 339.3 392C339 390.6 338.1 389.1 339.3 387.6C339.5 386.2 340.1 384.8 340.9 383.5V383.5zM312.3 6.307C368.5 19.04 418.7 50.28 455 95.01C485.4 132.6 504.6 178 510.3 226C515.9 274 507.9 322.7 487.1 366.3C466.2 409.9 433.5 446.8 392.6 472.6C351.7 498.3 304.4 512 256 512C254.5 512 253.1 511.7 251.7 511.1C250.3 510.6 249.1 509.7 248 508.7C246.1 507.6 246.1 506.4 245.6 505C245 503.6 244.7 502.2 244.7 500.7V401.5L220.2 427.5C218.1 429.7 215.3 430.1 212.3 431.1C209.3 431.2 206.4 430 204.2 427.1C202 425.9 200.7 423.1 200.7 420.1C200.6 417.1 201.7 414.2 203.7 412L247.8 365.4C249.1 363.2 252.9 362 255.1 362C259.1 362 262 363.2 264.2 365.4L308.2 412C310.3 414.2 311.4 417.1 311.3 420.1C311.2 423.1 309.9 425.9 307.8 427.1C305.6 430 302.7 431.2 299.7 431.1C296.7 430.1 293.8 429.7 291.8 427.5L267.3 401.6V489.1C323.7 486.3 377.1 463.3 417.8 424.1C458.5 384.1 483.6 332.4 488.5 276.2C493.3 219.1 477.7 163.9 444.4 118.3C411.1 72.75 362.4 40.79 307.4 28.36C305.9 28.03 304.6 27.42 303.3 26.57C302.1 25.71 301.1 24.63 300.3 23.37C299.5 22.12 298.1 20.72 298.7 19.26C298.5 17.8 298.5 16.3 298.8 14.85C299.2 13.41 299.8 12.04 300.6 10.82C301.5 9.61 302.6 8.577 303.8 7.784C305.1 6.99 306.5 6.451 307.9 6.198C309.4 5.945 310.9 5.982 312.3 6.307L312.3 6.307zM353.1 256.1C353.1 287.5 335.6 317.2 303.8 339.6C301.7 341.1 299 341.9 296.4 341.6C293.7 341.4 291.2 340.3 289.4 338.4L219.3 268.6C217.1 266.5 215.1 263.6 215.9 260.6C215.9 257.6 217.1 254.7 219.2 252.6C221.4 250.5 224.2 249.3 227.2 249.3C230.2 249.3 233.1 250.5 235.2 252.6L298.3 315.4C319.1 298.3 330.5 277.5 330.5 256.1C330.5 232.2 316.4 209.1 290.8 191C288.3 189.3 286.7 186.7 286.2 183.7C285.7 180.8 286.3 177.7 288.1 175.3C289.8 172.8 292.4 171.2 295.4 170.7C298.3 170.2 301.4 170.8 303.8 172.6C335.6 195 353.1 224.7 353.1 256.1V256.1zM216.7 341.5C213.7 342 210.7 341.3 208.2 339.6C176.5 317.2 158.1 287.5 158.1 256.1C158.1 224.7 176.5 195 208.2 172.6C210.4 171 213.1 170.3 215.7 170.5C218.4 170.8 220.8 171.9 222.7 173.8L292.8 243.6C294.9 245.7 296.1 248.6 296.1 251.6C296.1 254.6 294.1 257.4 292.8 259.6C290.7 261.7 287.8 262.9 284.9 262.9C281.9 262.9 278.1 261.7 276.9 259.6L213.8 196.7C192.9 214 181.6 234.7 181.6 256.1C181.6 279.1 195.7 303.1 221.3 321.1C223.7 322.9 225.4 325.5 225.9 328.5C226.4 331.4 225.7 334.4 224 336.9C222.3 339.3 219.6 341 216.7 341.5L216.7 341.5z"],skype:[448,512,[],"f17e","M424.7 299.8c2.9-14 4.7-28.9 4.7-43.8 0-113.5-91.9-205.3-205.3-205.3-14.9 0-29.7 1.7-43.8 4.7C161.3 40.7 137.7 32 112 32 50.2 32 0 82.2 0 144c0 25.7 8.7 49.3 23.3 68.2-2.9 14-4.7 28.9-4.7 43.8 0 113.5 91.9 205.3 205.3 205.3 14.9 0 29.7-1.7 43.8-4.7 19 14.6 42.6 23.3 68.2 23.3 61.8 0 112-50.2 112-112 .1-25.6-8.6-49.2-23.2-68.1zm-194.6 91.5c-65.6 0-120.5-29.2-120.5-65 0-16 9-30.6 29.5-30.6 31.2 0 34.1 44.9 88.1 44.9 25.7 0 42.3-11.4 42.3-26.3 0-18.7-16-21.6-42-28-62.5-15.4-117.8-22-117.8-87.2 0-59.2 58.6-81.1 109.1-81.1 55.1 0 110.8 21.9 110.8 55.4 0 16.9-11.4 31.8-30.3 31.8-28.3 0-29.2-33.5-75-33.5-25.7 0-42 7-42 22.5 0 19.8 20.8 21.8 69.1 33 41.4 9.3 90.7 26.8 90.7 77.6 0 59.1-57.1 86.5-112 86.5z"],joget:[496,512,[],"f3b7","M378.1 45C337.6 19.9 292.6 8 248.2 8 165 8 83.8 49.9 36.9 125.9c-71.9 116.6-35.6 269.3 81 341.2s269.3 35.6 341.2-80.9c71.9-116.6 35.6-269.4-81-341.2zm51.8 323.2c-40.4 65.5-110.4 101.5-182 101.5-6.8 0-13.6-.4-20.4-1-9-13.6-19.9-33.3-23.7-42.4-5.7-13.7-27.2-45.6 31.2-67.1 51.7-19.1 176.7-16.5 208.8-17.6-4 9-8.6 17.9-13.9 26.6zm-200.8-86.3c-55.5-1.4-81.7-20.8-58.5-48.2s51.1-40.7 68.9-51.2c17.9-10.5 27.3-33.7-23.6-29.7C87.3 161.5 48.6 252.1 37.6 293c-8.8-49.7-.1-102.7 28.5-149.1C128 43.4 259.6 12.2 360.1 74.1c74.8 46.1 111.2 130.9 99.3 212.7-24.9-.5-179.3-3.6-230.3-4.9zm183.8-54.8c-22.7-6-57 11.3-86.7 27.2-29.7 15.8-31.1 8.2-31.1 8.2s40.2-28.1 50.7-34.5 31.9-14 13.4-24.6c-3.2-1.8-6.7-2.7-10.4-2.7-17.8 0-41.5 18.7-67.5 35.6-31.5 20.5-65.3 31.3-65.3 31.3l169.5-1.6 46.5-23.4s3.6-9.5-19.1-15.5z"],fedora:[448,512,[],"f798","M.0413 255.8C.1219 132.2 100.3 32 224 32C347.7 32 448 132.3 448 256C448 379.7 347.8 479.9 224.1 480H50.93C22.84 480 .0832 457.3 .0416 429.2H0V255.8H.0413zM342.6 192.7C342.6 153 307 124.2 269.4 124.2C234.5 124.2 203.6 150.5 199.3 184.1C199.1 187.9 198.9 189.1 198.9 192.6C198.8 213.7 198.9 235.4 198.1 257C199 283.1 199.1 309.1 198.1 333.6C198.1 360.7 178.7 379.1 153.4 379.1C128.1 379.1 107.6 358.9 107.6 333.6C108.1 305.9 130.2 288.3 156.1 287.5H156.3L182.6 287.3V250L156.3 250.2C109.2 249.8 71.72 286.7 70.36 333.6C70.36 379.2 107.9 416.5 153.4 416.5C196.4 416.5 232.1 382.9 236 340.9L236.2 287.4L268.8 287.1C294.1 287.3 293.8 249.3 268.6 249.8L236.2 250.1C236.2 243.7 236.3 237.3 236.3 230.9C236.4 218.2 236.4 205.5 236.2 192.7C236.3 176.2 252 161.5 269.4 161.5C286.9 161.5 305.3 170.2 305.3 192.7C305.3 195.9 305.2 197.8 305 199C303.1 209.5 310.2 219.4 320.7 220.9C331.3 222.4 340.9 214.8 341.9 204.3C342.5 200.1 342.6 196.4 342.6 192.7H342.6z"],"stripe-s":[384,512,[],"f42a","M155.3 154.6c0-22.3 18.6-30.9 48.4-30.9 43.4 0 98.5 13.3 141.9 36.7V26.1C298.3 7.2 251.1 0 203.8 0 88.1 0 11 60.4 11 161.4c0 157.9 216.8 132.3 216.8 200.4 0 26.4-22.9 34.9-54.7 34.9-47.2 0-108.2-19.5-156.1-45.5v128.5a396.09 396.09 0 0 0 156 32.4c118.6 0 200.3-51 200.3-153.6 0-170.2-218-139.7-218-203.9z"],meta:[640,512,[],"e49b","M640 317.9C640 409.2 600.6 466.4 529.7 466.4C467.1 466.4 433.9 431.8 372.8 329.8L341.4 277.2C333.1 264.7 326.9 253 320.2 242.2C300.1 276 273.1 325.2 273.1 325.2C206.1 441.8 168.5 466.4 116.2 466.4C43.42 466.4 0 409.1 0 320.5C0 177.5 79.78 42.4 183.9 42.4C234.1 42.4 277.7 67.08 328.7 131.9C365.8 81.8 406.8 42.4 459.3 42.4C558.4 42.4 640 168.1 640 317.9H640zM287.4 192.2C244.5 130.1 216.5 111.7 183 111.7C121.1 111.7 69.22 217.8 69.22 321.7C69.22 370.2 87.7 397.4 118.8 397.4C149 397.4 167.8 378.4 222 293.6C222 293.6 246.7 254.5 287.4 192.2V192.2zM531.2 397.4C563.4 397.4 578.1 369.9 578.1 322.5C578.1 198.3 523.8 97.08 454.9 97.08C421.7 97.08 393.8 123 360 175.1C369.4 188.9 379.1 204.1 389.3 220.5L426.8 282.9C485.5 377 500.3 397.4 531.2 397.4L531.2 397.4z"],laravel:[512,512,[],"f3bd","M504.4,115.83a5.72,5.72,0,0,0-.28-.68,8.52,8.52,0,0,0-.53-1.25,6,6,0,0,0-.54-.71,9.36,9.36,0,0,0-.72-.94c-.23-.22-.52-.4-.77-.6a8.84,8.84,0,0,0-.9-.68L404.4,55.55a8,8,0,0,0-8,0L300.12,111h0a8.07,8.07,0,0,0-.88.69,7.68,7.68,0,0,0-.78.6,8.23,8.23,0,0,0-.72.93c-.17.24-.39.45-.54.71a9.7,9.7,0,0,0-.52,1.25c-.08.23-.21.44-.28.68a8.08,8.08,0,0,0-.28,2.08V223.18l-80.22,46.19V63.44a7.8,7.8,0,0,0-.28-2.09c-.06-.24-.2-.45-.28-.68a8.35,8.35,0,0,0-.52-1.24c-.14-.26-.37-.47-.54-.72a9.36,9.36,0,0,0-.72-.94,9.46,9.46,0,0,0-.78-.6,9.8,9.8,0,0,0-.88-.68h0L115.61,1.07a8,8,0,0,0-8,0L11.34,56.49h0a6.52,6.52,0,0,0-.88.69,7.81,7.81,0,0,0-.79.6,8.15,8.15,0,0,0-.71.93c-.18.25-.4.46-.55.72a7.88,7.88,0,0,0-.51,1.24,6.46,6.46,0,0,0-.29.67,8.18,8.18,0,0,0-.28,2.1v329.7a8,8,0,0,0,4,6.95l192.5,110.84a8.83,8.83,0,0,0,1.33.54c.21.08.41.2.63.26a7.92,7.92,0,0,0,4.1,0c.2-.05.37-.16.55-.22a8.6,8.6,0,0,0,1.4-.58L404.4,400.09a8,8,0,0,0,4-6.95V287.88l92.24-53.11a8,8,0,0,0,4-7V117.92A8.63,8.63,0,0,0,504.4,115.83ZM111.6,17.28h0l80.19,46.15-80.2,46.18L31.41,63.44Zm88.25,60V278.6l-46.53,26.79-33.69,19.4V123.5l46.53-26.79Zm0,412.78L23.37,388.5V77.32L57.06,96.7l46.52,26.8V338.68a6.94,6.94,0,0,0,.12.9,8,8,0,0,0,.16,1.18h0a5.92,5.92,0,0,0,.38.9,6.38,6.38,0,0,0,.42,1v0a8.54,8.54,0,0,0,.6.78,7.62,7.62,0,0,0,.66.84l0,0c.23.22.52.38.77.58a8.93,8.93,0,0,0,.86.66l0,0,0,0,92.19,52.18Zm8-106.17-80.06-45.32,84.09-48.41,92.26-53.11,80.13,46.13-58.8,33.56Zm184.52,4.57L215.88,490.11V397.8L346.6,323.2l45.77-26.15Zm0-119.13L358.68,250l-46.53-26.79V131.79l33.69,19.4L392.37,178Zm8-105.28-80.2-46.17,80.2-46.16,80.18,46.15Zm8,105.28V178L455,151.19l33.68-19.4v91.39h0Z"],hotjar:[512,512,[],"f3b1","M361.5 0c0 131.6-80.7 176.8-140.2 209.4c-.6 .3-1.1 .6-1.6 .9c-53.8 30.2-88.7 49.8-89.6 122H32C32 200.8 112.7 155.6 172.2 123C227 93.2 262.5 73 262.5 0h98.9zM301 302.6c54.8-29.8 90.3-50 90.3-123h98c0 131.6-80.7 176.7-140.2 209.4c-54.8 29.8-90.3 50-90.3 123h-98c0-131.6 80.7-176.8 140.2-209.4z"],"bluetooth-b":[320,512,[],"f294","M196.48 260.023l92.626-103.333L143.125 0v206.33l-86.111-86.111-31.406 31.405 108.061 108.399L25.608 368.422l31.406 31.405 86.111-86.111L145.84 512l148.552-148.644-97.912-103.333zm40.86-102.996l-49.977 49.978-.338-100.295 50.315 50.317zM187.363 313.04l49.977 49.978-50.315 50.316.338-100.294z"],"square-letterboxd":[448,512,[],"e62e","M384 32c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H384zM105.1 187C66.4 187 35 218.3 35 257s31.4 70 70.1 70c24.8 0 46.5-12.8 59-32.2l.5-.7-.4-.6c-6.5-10.6-10.2-23.1-10.2-36.5c0-13.6 3.9-26.3 10.6-37.1c-12.4-19.8-34.4-32.9-59.5-32.9zM224 187c-24.8 0-46.5 12.8-59 32.2l-.5 .7 .4 .6c6.5 10.6 10.2 23.1 10.2 36.5c0 13.6-3.9 26.3-10.6 37.1C176.9 313.8 198.9 327 224 327c24.8 0 46.5-12.8 59-32.2l.5-.7-.4-.6c-6.5-10.6-10.2-23.1-10.2-36.5c0-13.6 3.9-26.3 10.6-37.1C271.1 200.2 249.1 187 224 187zm118.9 0c-24.8 0-46.5 12.8-59 32.2l-.5 .7 .4 .6c6.5 10.6 10.2 23.1 10.2 36.5c0 13.6-3.9 26.3-10.6 37.1c12.4 19.8 34.4 32.9 59.5 32.9c38.7 0 70.1-31.3 70.1-70s-31.4-70-70.1-70z"],"sticker-mule":[576,512,[],"f3f7","M561.7 199.6c-1.3.3.3 0 0 0zm-6.2-77.4c-7.7-22.3-5.1-7.2-13.4-36.9-1.6-6.5-3.6-14.5-6.2-20-4.4-8.7-4.6-7.5-4.6-9.5 0-5.3 30.7-45.3 19-46.9-5.7-.6-12.2 11.6-20.6 17-8.6 4.2-8 5-10.3 5-2.6 0-5.7-3-6.2-5-2-5.7 1.9-25.9-3.6-25.9-3.6 0-12.3 24.8-17 25.8-5.2 1.3-27.9-11.4-75.1 18-25.3 13.2-86.9 65.2-87 65.3-6.7 4.7-20 4.7-35.5 16-44.4 30.1-109.6 9.4-110.7 9-110.6-26.8-128-15.2-159 11.5-20.8 17.9-23.7 36.5-24.2 38.9-4.2 20.4 5.2 48.3 6.7 64.3 1.8 19.3-2.7 17.7 7.7 98.3.5 1 4.1 0 5.1 1.5 0 8.4-3.8 12.1-4.1 13-1.5 4.5-1.5 10.5 0 16 2.3 8.2 8.2 37.2 8.2 46.9 0 41.8.4 44 2.6 49.4 3.9 10 12.5 9.1 17 12 3.1 3.5-.5 8.5 1 12.5.5 2 3.6 4 6.2 5 9.2 3.6 27 .3 29.9-2.5 1.6-1.5.5-4.5 3.1-5 5.1 0 10.8-.5 14.4-2.5 5.1-2.5 4.1-6 1.5-10.5-.4-.8-7-13.3-9.8-16-2.1-2-5.1-3-7.2-4.5-5.8-4.9-10.3-19.4-10.3-19.5-4.6-19.4-10.3-46.3-4.1-66.8 4.6-17.2 39.5-87.7 39.6-87.8 4.1-6.5 17-11.5 27.3-7 6 1.9 19.3 22 65.4 30.9 47.9 8.7 97.4-2 112.2-2 2.8 2-1.9 13-.5 38.9 0 26.4-.4 13.7-4.1 29.9-2.2 9.7 3.4 23.2-1.5 46.9-1.4 9.8-9.9 32.7-8.2 43.4.5 1 1 2 1.5 3.5.5 4.5 1.5 8.5 4.6 10 7.3 3.6 12-3.5 9.8 11.5-.7 3.1-2.6 12 1.5 15 4.4 3.7 30.6 3.4 36.5.5 2.6-1.5 1.6-4.5 6.4-7.4 1.9-.9 11.3-.4 11.3-6.5.3-1.8-9.2-19.9-9.3-20-2.6-3.5-9.2-4.5-11.3-8-6.9-10.1-1.7-52.6.5-59.4 3-11 5.6-22.4 8.7-32.4 11-42.5 10.3-50.6 16.5-68.3.8-1.8 6.4-23.1 10.3-29.9 9.3-17 21.7-32.4 33.5-47.4 18-22.9 34-46.9 52-69.8 6.1-7 8.2-13.7 18-8 10.8 5.7 21.6 7 31.9 17 14.6 12.8 10.2 18.2 11.8 22.9 1.5 5 7.7 10.5 14.9 9.5 10.4-2 13-2.5 13.4-2.5 2.6-.5 5.7-5 7.2-8 3.1-5.5 7.2-9 7.2-16.5 0-7.7-.4-2.8-20.6-52.9z"],"creative-commons-zero":[496,512,[],"f4f3","M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zm-.4 60.5c-81.9 0-102.5 77.3-102.5 142.8 0 65.5 20.6 142.8 102.5 142.8S350.5 321.5 350.5 256c0-65.5-20.6-142.8-102.5-142.8zm0 53.9c3.3 0 6.4.5 9.2 1.2 5.9 5.1 8.8 12.1 3.1 21.9l-54.5 100.2c-1.7-12.7-1.9-25.1-1.9-34.4 0-28.8 2-88.9 44.1-88.9zm40.8 46.2c2.9 15.4 3.3 31.4 3.3 42.7 0 28.9-2 88.9-44.1 88.9-13.5 0-32.6-7.7-20.1-26.4l60.9-105.2z"],hips:[640,512,[],"f452","M251.6 157.6c0-1.9-.9-2.8-2.8-2.8h-40.9c-1.6 0-2.7 1.4-2.7 2.8v201.8c0 1.4 1.1 2.8 2.7 2.8h40.9c1.9 0 2.8-.9 2.8-2.8zM156.5 168c-16.1-11.8-36.3-17.9-60.3-18-18.1-.1-34.6 3.7-49.8 11.4V80.2c0-1.8-.9-2.7-2.8-2.7H2.7c-1.8 0-2.7.9-2.7 2.7v279.2c0 1.9.9 2.8 2.7 2.8h41c1.9 0 2.8-.9 2.8-2.8V223.3c0-.8-2.8-27 45.8-27 48.5 0 45.8 26.1 45.8 27v122.6c0 9 7.3 16.3 16.4 16.3h27.3c1.8 0 2.7-.9 2.7-2.8V223.3c0-23.4-9.3-41.8-28-55.3zm478.4 110.1c-6.8-15.7-18.4-27-34.9-34.1l-57.6-25.3c-8.6-3.6-9.2-11.2-2.6-16.1 7.4-5.5 44.3-13.9 84 6.8 1.7 1 4-.3 4-2.4v-44.7c0-1.3-.6-2.1-1.9-2.6-17.7-6.6-36.1-9.9-55.1-9.9-26.5 0-45.3 5.8-58.5 15.4-.5.4-28.4 20-22.7 53.7 3.4 19.6 15.8 34.2 37.2 43.6l53.6 23.5c11.6 5.1 15.2 13.3 12.2 21.2-3.7 9.1-13.2 13.6-36.5 13.6-24.3 0-44.7-8.9-58.4-19.1-2.1-1.4-4.4.2-4.4 2.3v34.4c0 10.4 4.9 17.3 14.6 20.7 15.6 5.5 31.6 8.2 48.2 8.2 12.7 0 25.8-1.2 36.3-4.3.7-.3 36-8.9 45.6-45.8 3.5-13.5 2.4-26.5-3.1-39.1zM376.2 149.8c-31.7 0-104.2 20.1-104.2 103.5v183.5c0 .8.6 2.7 2.7 2.7h40.9c1.9 0 2.8-.9 2.8-2.7V348c16.5 12.7 35.8 19.1 57.7 19.1 60.5 0 108.7-48.5 108.7-108.7.1-60.3-48.2-108.6-108.6-108.6zm0 170.9c-17.2 0-31.9-6.1-44-18.2-12.2-12.2-18.2-26.8-18.2-44 0-34.5 27.6-62.2 62.2-62.2 34.5 0 62.2 27.6 62.2 62.2.1 34.3-27.3 62.2-62.2 62.2zM228.3 72.5c-15.9 0-28.8 12.9-28.9 28.9 0 15.6 12.7 28.9 28.9 28.9s28.9-13.1 28.9-28.9c0-16.2-13-28.9-28.9-28.9z"],behance:[576,512,[],"f1b4","M232 237.2c31.8-15.2 48.4-38.2 48.4-74 0-70.6-52.6-87.8-113.3-87.8H0v354.4h171.8c64.4 0 124.9-30.9 124.9-102.9 0-44.5-21.1-77.4-64.7-89.7zM77.9 135.9H151c28.1 0 53.4 7.9 53.4 40.5 0 30.1-19.7 42.2-47.5 42.2h-79v-82.7zm83.3 233.7H77.9V272h84.9c34.3 0 56 14.3 56 50.6 0 35.8-25.9 47-57.6 47zm358.5-240.7H376V94h143.7v34.9zM576 305.2c0-75.9-44.4-139.2-124.9-139.2-78.2 0-131.3 58.8-131.3 135.8 0 79.9 50.3 134.7 131.3 134.7 61.3 0 101-27.6 120.1-86.3H509c-6.7 21.9-34.3 33.5-55.7 33.5-41.3 0-63-24.2-63-65.3h185.1c.3-4.2.6-8.7.6-13.2zM390.4 274c2.3-33.7 24.7-54.8 58.5-54.8 35.4 0 53.2 20.8 56.2 54.8H390.4z"],reddit:[512,512,[],"f1a1","M0 256C0 114.6 114.6 0 256 0S512 114.6 512 256s-114.6 256-256 256L37.1 512c-13.7 0-20.5-16.5-10.9-26.2L75 437C28.7 390.7 0 326.7 0 256zM349.6 153.6c23.6 0 42.7-19.1 42.7-42.7s-19.1-42.7-42.7-42.7c-20.6 0-37.8 14.6-41.8 34c-34.5 3.7-61.4 33-61.4 68.4l0 .2c-37.5 1.6-71.8 12.3-99 29.1c-10.1-7.8-22.8-12.5-36.5-12.5c-33 0-59.8 26.8-59.8 59.8c0 24 14.1 44.6 34.4 54.1c2 69.4 77.6 125.2 170.6 125.2s168.7-55.9 170.6-125.3c20.2-9.6 34.1-30.2 34.1-54c0-33-26.8-59.8-59.8-59.8c-13.7 0-26.3 4.6-36.4 12.4c-27.4-17-62.1-27.7-100-29.1l0-.2c0-25.4 18.9-46.5 43.4-49.9l0 0c4.4 18.8 21.3 32.8 41.5 32.8zM177.1 246.9c16.7 0 29.5 17.6 28.5 39.3s-13.5 29.6-30.3 29.6s-31.4-8.8-30.4-30.5s15.4-38.3 32.1-38.3zm190.1 38.3c1 21.7-13.7 30.5-30.4 30.5s-29.3-7.9-30.3-29.6c-1-21.7 11.8-39.3 28.5-39.3s31.2 16.6 32.1 38.3zm-48.1 56.7c-10.3 24.6-34.6 41.9-63 41.9s-52.7-17.3-63-41.9c-1.2-2.9 .8-6.2 3.9-6.5c18.4-1.9 38.3-2.9 59.1-2.9s40.7 1 59.1 2.9c3.1 .3 5.1 3.6 3.9 6.5z"],discord:[640,512,[],"f392","M524.531,69.836a1.5,1.5,0,0,0-.764-.7A485.065,485.065,0,0,0,404.081,32.03a1.816,1.816,0,0,0-1.923.91,337.461,337.461,0,0,0-14.9,30.6,447.848,447.848,0,0,0-134.426,0,309.541,309.541,0,0,0-15.135-30.6,1.89,1.89,0,0,0-1.924-.91A483.689,483.689,0,0,0,116.085,69.137a1.712,1.712,0,0,0-.788.676C39.068,183.651,18.186,294.69,28.43,404.354a2.016,2.016,0,0,0,.765,1.375A487.666,487.666,0,0,0,176.02,479.918a1.9,1.9,0,0,0,2.063-.676A348.2,348.2,0,0,0,208.12,430.4a1.86,1.86,0,0,0-1.019-2.588,321.173,321.173,0,0,1-45.868-21.853,1.885,1.885,0,0,1-.185-3.126c3.082-2.309,6.166-4.711,9.109-7.137a1.819,1.819,0,0,1,1.9-.256c96.229,43.917,200.41,43.917,295.5,0a1.812,1.812,0,0,1,1.924.233c2.944,2.426,6.027,4.851,9.132,7.16a1.884,1.884,0,0,1-.162,3.126,301.407,301.407,0,0,1-45.89,21.83,1.875,1.875,0,0,0-1,2.611,391.055,391.055,0,0,0,30.014,48.815,1.864,1.864,0,0,0,2.063.7A486.048,486.048,0,0,0,610.7,405.729a1.882,1.882,0,0,0,.765-1.352C623.729,277.594,590.933,167.465,524.531,69.836ZM222.491,337.58c-28.972,0-52.844-26.587-52.844-59.239S193.056,219.1,222.491,219.1c29.665,0,53.306,26.82,52.843,59.239C275.334,310.993,251.924,337.58,222.491,337.58Zm195.38,0c-28.971,0-52.843-26.587-52.843-59.239S388.437,219.1,417.871,219.1c29.667,0,53.307,26.82,52.844,59.239C470.715,310.993,447.538,337.58,417.871,337.58Z"],chrome:[512,512,[],"f268","M0 256C0 209.4 12.47 165.6 34.27 127.1L144.1 318.3C166 357.5 207.9 384 256 384C270.3 384 283.1 381.7 296.8 377.4L220.5 509.6C95.9 492.3 0 385.3 0 256zM365.1 321.6C377.4 302.4 384 279.1 384 256C384 217.8 367.2 183.5 340.7 160H493.4C505.4 189.6 512 222.1 512 256C512 397.4 397.4 511.1 256 512L365.1 321.6zM477.8 128H256C193.1 128 142.3 172.1 130.5 230.7L54.19 98.47C101 38.53 174 0 256 0C350.8 0 433.5 51.48 477.8 128V128zM168 256C168 207.4 207.4 168 256 168C304.6 168 344 207.4 344 256C344 304.6 304.6 344 256 344C207.4 344 168 304.6 168 256z"],"app-store-ios":[448,512,[],"f370","M400 32H48C21.5 32 0 53.5 0 80v352c0 26.5 21.5 48 48 48h352c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48zM127 384.5c-5.5 9.6-17.8 12.8-27.3 7.3-9.6-5.5-12.8-17.8-7.3-27.3l14.3-24.7c16.1-4.9 29.3-1.1 39.6 11.4L127 384.5zm138.9-53.9H84c-11 0-20-9-20-20s9-20 20-20h51l65.4-113.2-20.5-35.4c-5.5-9.6-2.2-21.8 7.3-27.3 9.6-5.5 21.8-2.2 27.3 7.3l8.9 15.4 8.9-15.4c5.5-9.6 17.8-12.8 27.3-7.3 9.6 5.5 12.8 17.8 7.3 27.3l-85.8 148.6h62.1c20.2 0 31.5 23.7 22.7 40zm98.1 0h-29l19.6 33.9c5.5 9.6 2.2 21.8-7.3 27.3-9.6 5.5-21.8 2.2-27.3-7.3-32.9-56.9-57.5-99.7-74-128.1-16.7-29-4.8-58 7.1-67.8 13.1 22.7 32.7 56.7 58.9 102h52c11 0 20 9 20 20 0 11.1-9 20-20 20z"],"cc-discover":[576,512,[],"f1f2","M520.4 196.1c0-7.9-5.5-12.1-15.6-12.1h-4.9v24.9h4.7c10.3 0 15.8-4.4 15.8-12.8zM528 32H48C21.5 32 0 53.5 0 80v352c0 26.5 21.5 48 48 48h480c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48zm-44.1 138.9c22.6 0 52.9-4.1 52.9 24.4 0 12.6-6.6 20.7-18.7 23.2l25.8 34.4h-19.6l-22.2-32.8h-2.2v32.8h-16zm-55.9.1h45.3v14H444v18.2h28.3V217H444v22.2h29.3V253H428zm-68.7 0l21.9 55.2 22.2-55.2h17.5l-35.5 84.2h-8.6l-35-84.2zm-55.9-3c24.7 0 44.6 20 44.6 44.6 0 24.7-20 44.6-44.6 44.6-24.7 0-44.6-20-44.6-44.6 0-24.7 20-44.6 44.6-44.6zm-49.3 6.1v19c-20.1-20.1-46.8-4.7-46.8 19 0 25 27.5 38.5 46.8 19.2v19c-29.7 14.3-63.3-5.7-63.3-38.2 0-31.2 33.1-53 63.3-38zm-97.2 66.3c11.4 0 22.4-15.3-3.3-24.4-15-5.5-20.2-11.4-20.2-22.7 0-23.2 30.6-31.4 49.7-14.3l-8.4 10.8c-10.4-11.6-24.9-6.2-24.9 2.5 0 4.4 2.7 6.9 12.3 10.3 18.2 6.6 23.6 12.5 23.6 25.6 0 29.5-38.8 37.4-56.6 11.3l10.3-9.9c3.7 7.1 9.9 10.8 17.5 10.8zM55.4 253H32v-82h23.4c26.1 0 44.1 17 44.1 41.1 0 18.5-13.2 40.9-44.1 40.9zm67.5 0h-16v-82h16zM544 433c0 8.2-6.8 15-15 15H128c189.6-35.6 382.7-139.2 416-160zM74.1 191.6c-5.2-4.9-11.6-6.6-21.9-6.6H48v54.2h4.2c10.3 0 17-2 21.9-6.4 5.7-5.2 8.9-12.8 8.9-20.7s-3.2-15.5-8.9-20.5z"],wpbeginner:[512,512,[],"f297","M462.799 322.374C519.01 386.682 466.961 480 370.944 480c-39.602 0-78.824-17.687-100.142-50.04-6.887.356-22.702.356-29.59 0C219.848 462.381 180.588 480 141.069 480c-95.49 0-148.348-92.996-91.855-157.626C-29.925 190.523 80.479 32 256.006 32c175.632 0 285.87 158.626 206.793 290.374zm-339.647-82.972h41.529v-58.075h-41.529v58.075zm217.18 86.072v-23.839c-60.506 20.915-132.355 9.198-187.589-33.971l.246 24.897c51.101 46.367 131.746 57.875 187.343 32.913zm-150.753-86.072h166.058v-58.075H189.579v58.075z"],confluence:[512,512,[],"f78d","M2.3 412.2c-4.5 7.6-2.1 17.5 5.5 22.2l105.9 65.2c7.7 4.7 17.7 2.4 22.4-5.3 0-.1.1-.2.1-.2 67.1-112.2 80.5-95.9 280.9-.7 8.1 3.9 17.8.4 21.7-7.7.1-.1.1-.3.2-.4l50.4-114.1c3.6-8.1-.1-17.6-8.1-21.3-22.2-10.4-66.2-31.2-105.9-50.3C127.5 179 44.6 345.3 2.3 412.2zm507.4-312.1c4.5-7.6 2.1-17.5-5.5-22.2L398.4 12.8c-7.5-5-17.6-3.1-22.6 4.4-.2.3-.4.6-.6 1-67.3 112.6-81.1 95.6-280.6.9-8.1-3.9-17.8-.4-21.7 7.7-.1.1-.1.3-.2.4L22.2 141.3c-3.6 8.1.1 17.6 8.1 21.3 22.2 10.4 66.3 31.2 106 50.4 248 120 330.8-45.4 373.4-112.9z"],shoelace:[512,512,[],"e60c","M404.9 331c2.2-1.1 4.4-2.3 6.5-3.7l8.3-4.8c1.5-1.1 4.4-3.4 8.7-6.7l.5-.5c3.4-3.4 7.2-5 11.3-4.9c1.8 0 3.9 .5 6.4 1.5l31-27.5c.9-.7 1.7-1.3 2.6-1.8h.2c3.3-1.9 6-1.8 8.2 .4c3.9 2.1 4.2 5.6 .9 10.6L456.9 322c.2 .5 .4 1 .4 1.5c.5 2.2 .3 4.4-.5 6.6c-.7 1.5-1.8 2.9-3.1 4.2c-1.4 1.4-2.7 2.8-4.2 4.2l-18.8 13.7c-1.7 1.2-3.4 2.3-5.1 3.3c-2.1 1.3-4.3 2.5-6.6 3.6c-1 .4-1.9 .9-2.9 1.3c-5.9 2.5-11.9 4.2-18.2 5c-2.9 24.5-11.3 47.1-25.1 67.8c-17.5 25.7-41.4 45.4-71.8 58.8c-30.2 13.5-63 20.2-98.2 20.2c-48.6-.5-88-11.4-118.2-32.8C49.5 454.4 32 421.5 32 380.3v-5.6c1.2-28.1 9.5-54.6 24.8-79.8c15.1-24.9 37.1-41.7 66.1-50.5c14.9-4.4 29.9-6.6 45-6.6c15.5 0 31.6 2.9 48.1 8.6s35.2 15.5 55.9 29.5L326 312.2c15.1 9.8 28.8 16.5 41.2 20c-2.6-25.1-11.7-46.6-27.3-64.5c-15.7-18.1-35.6-31.3-59.9-39.7l-23.3-8c-21.4-7.5-37.3-14.9-47.7-22.2c-28.2-19.1-43.8-45.2-47-78.5l-.5-9.8c0-32.1 13-58.9 39-80.5C223.5 9.7 251.1 0 283 0c24 0 45.6 6.9 64.7 20.8c19.2 14 30.1 33.8 32.6 59.4l.5 10c0 18.6-4.8 34.5-14.4 47.7c-9.8 13.2-18.5 19.9-26 19.9c-1.6-.1-3.1-.3-4.5-.6l-34 32c-5.5 3-9.2 2.5-11.1-1.6c-1.9-2.2-1.8-4.9 .5-8.2l.2-.2c.5-.7 1.2-1.5 2-2.4l31.6-30c-.4-1.5-.6-3.1-.6-4.8c0-4.1 1.6-7.6 4.9-10.4c13.8-12.4 20.8-26.7 20.8-42.8c0-16-6.1-29.5-18.2-40.4s-28.7-16.5-49.7-16.8c-26.2 0-47.8 7.9-64.7 23.7S192.3 89.9 192.3 112c0 17.8 6.9 33.9 20.6 48.3c13.6 14.2 34.6 25.4 63 33.5c39.8 11.5 70.2 31 91.3 58.3c18.7 24.2 29.1 51.3 31.3 81.4c2.2-.7 4.3-1.5 6.5-2.6zM294.1 178.7c0 1.1 .6 1.6 1.8 1.6c.1 0 9.7-8.9 28.8-26.6c0-2.4-5.1 .9-15.3 10c-10.2 9.2-15.3 14.2-15.3 14.9zm8 6.4c0-1-.5-1.5-1.5-1.5c-1.1 0-2.1 .5-2.9 1.6c-1.9-.1-3.3 .1-4.2 .7c-.4 .2-.5 .5-.5 .7c0 .7 .5 1.3 1.5 1.6h3.3c2.9-1.1 4.4-2.2 4.4-3.3zm22.6-19.9c0-2.8-1.6-2.8-4.9 0c-1.6 1.5-3.6 3.5-6 6.2c-.8 .6-2.6 2.2-5.3 4.9c-2.8 2.9-4.2 4.7-4.2 5.3l.2 1.3c.7 .2 1.2 .4 1.5 .4c.1 0 3.3-2.9 9.5-8.7s9.3-8.9 9.3-9.3zm159.7 120l-30.6 27c1.8 1 3.2 2.4 4 4.2l30.2-27c.2-1.2 .1-2.2-.5-2.9c-.6-.5-1.6-.9-3.1-1.3zm-1.6-.9l-.7-.7-27 21.9 1.6 2 26-23.1zM366.6 363.9c-8-2.1-15.4-4.6-22.2-7.5c-15.3-6.2-34.3-17-57-32.4L250 298.7c-15.8-10.1-30.2-17.6-43.2-22.6c-13.1-4.9-26-7.3-38.6-7.3h-5.5c-32.2 1.7-57.2 13.8-75 36.2c-16.6 20.8-25 45.3-25 73.6c0 31.8 12.8 56.7 38.2 74.7c25.4 18.1 60.2 27.1 104.4 27.1c34.7 0 64-6.2 87.8-18.6c23.7-12.4 42.1-28.8 55.2-49.2c9.8-15.5 15.9-31.8 18.2-48.8z"],mdb:[576,512,[],"f8ca","M17.37 160.41L7 352h43.91l5.59-79.83L84.43 352h44.71l25.54-77.43 4.79 77.43H205l-12.79-191.59H146.7L106 277.74 63.67 160.41zm281 0h-47.9V352h47.9s95 .8 94.2-95.79c-.78-94.21-94.18-95.78-94.18-95.78zm-1.2 146.46V204.78s46 4.27 46.8 50.57-46.78 51.54-46.78 51.54zm238.29-74.24a56.16 56.16 0 0 0 8-38.31c-5.34-35.76-55.08-34.32-55.08-34.32h-51.9v191.58H482s87 4.79 87-63.85c0-43.14-33.52-55.08-33.52-55.08zm-51.9-31.94s13.57-1.59 16 9.59c1.43 6.66-4 12-4 12h-12v-21.57zm-.1 109.46l.1-24.92V267h.08s41.58-4.73 41.19 22.43c-.33 25.65-41.35 20.74-41.35 20.74z"],dochub:[416,512,[],"f394","M397.9 160H256V19.6L397.9 160zM304 192v130c0 66.8-36.5 100.1-113.3 100.1H96V84.8h94.7c12 0 23.1.8 33.1 2.5v-84C212.9 1.1 201.4 0 189.2 0H0v512h189.2C329.7 512 400 447.4 400 318.1V192h-96z"],"accessible-icon":[448,512,[62107],"f368","M423.9 255.8L411 413.1c-3.3 40.7-63.9 35.1-60.6-4.9l10-122.5-41.1 2.3c10.1 20.7 15.8 43.9 15.8 68.5 0 41.2-16.1 78.7-42.3 106.5l-39.3-39.3c57.9-63.7 13.1-167.2-74-167.2-25.9 0-49.5 9.9-67.2 26L73 243.2c22-20.7 50.1-35.1 81.4-40.2l75.3-85.7-42.6-24.8-51.6 46c-30 26.8-70.6-18.5-40.5-45.4l68-60.7c9.8-8.8 24.1-10.2 35.5-3.6 0 0 139.3 80.9 139.5 81.1 16.2 10.1 20.7 36 6.1 52.6L285.7 229l106.1-5.9c18.5-1.1 33.6 14.4 32.1 32.7zm-64.9-154c28.1 0 50.9-22.8 50.9-50.9C409.9 22.8 387.1 0 359 0c-28.1 0-50.9 22.8-50.9 50.9 0 28.1 22.8 50.9 50.9 50.9zM179.6 456.5c-80.6 0-127.4-90.6-82.7-156.1l-39.7-39.7C36.4 287 24 320.3 24 356.4c0 130.7 150.7 201.4 251.4 122.5l-39.7-39.7c-16 10.9-35.3 17.3-56.1 17.3z"],ebay:[640,512,[],"f4f4","M606 189.5l-54.8 109.9-54.9-109.9h-37.5l10.9 20.6c-11.5-19-35.9-26-63.3-26-31.8 0-67.9 8.7-71.5 43.1h33.7c1.4-13.8 15.7-21.8 35-21.8 26 0 41 9.6 41 33v3.4c-12.7 0-28 .1-41.7.4-42.4.9-69.6 10-76.7 34.4 1-5.2 1.5-10.6 1.5-16.2 0-52.1-39.7-76.2-75.4-76.2-21.3 0-43 5.5-58.7 24.2v-80.6h-32.1v169.5c0 10.3-.6 22.9-1.1 33.1h31.5c.7-6.3 1.1-12.9 1.1-19.5 13.6 16.6 35.4 24.9 58.7 24.9 36.9 0 64.9-21.9 73.3-54.2-.5 2.8-.7 5.8-.7 9 0 24.1 21.1 45 60.6 45 26.6 0 45.8-5.7 61.9-25.5 0 6.6.3 13.3 1.1 20.2h29.8c-.7-8.2-1-17.5-1-26.8v-65.6c0-9.3-1.7-17.2-4.8-23.8l61.5 116.1-28.5 54.1h35.9L640 189.5zM243.7 313.8c-29.6 0-50.2-21.5-50.2-53.8 0-32.4 20.6-53.8 50.2-53.8 29.8 0 50.2 21.4 50.2 53.8 0 32.3-20.4 53.8-50.2 53.8zm200.9-47.3c0 30-17.9 48.4-51.6 48.4-25.1 0-35-13.4-35-25.8 0-19.1 18.1-24.4 47.2-25.3 13.1-.5 27.6-.6 39.4-.6zm-411.9 1.6h128.8v-8.5c0-51.7-33.1-75.4-78.4-75.4-56.8 0-83 30.8-83 77.6 0 42.5 25.3 74 82.5 74 31.4 0 68-11.7 74.4-46.1h-33.1c-12 35.8-87.7 36.7-91.2-21.6zm95-21.4H33.3c6.9-56.6 92.1-54.7 94.4 0z"],amazon:[448,512,[],"f270","M257.2 162.7c-48.7 1.8-169.5 15.5-169.5 117.5 0 109.5 138.3 114 183.5 43.2 6.5 10.2 35.4 37.5 45.3 46.8l56.8-56S341 288.9 341 261.4V114.3C341 89 316.5 32 228.7 32 140.7 32 94 87 94 136.3l73.5 6.8c16.3-49.5 54.2-49.5 54.2-49.5 40.7-.1 35.5 29.8 35.5 69.1zm0 86.8c0 80-84.2 68-84.2 17.2 0-47.2 50.5-56.7 84.2-57.8v40.6zm136 163.5c-7.7 10-70 67-174.5 67S34.2 408.5 9.7 379c-6.8-7.7 1-11.3 5.5-8.3C88.5 415.2 203 488.5 387.7 401c7.5-3.7 13.3 2 5.5 12zm39.8 2.2c-6.5 15.8-16 26.8-21.2 31-5.5 4.5-9.5 2.7-6.5-3.8s19.3-46.5 12.7-55c-6.5-8.3-37-4.3-48-3.2-10.8 1-13 2-14-.3-2.3-5.7 21.7-15.5 37.5-17.5 15.7-1.8 41-.8 46 5.7 3.7 5.1 0 27.1-6.5 43.1z"],unsplash:[448,512,[],"e07c","M448,230.17V480H0V230.17H141.13V355.09H306.87V230.17ZM306.87,32H141.13V156.91H306.87Z"],yarn:[496,512,[],"f7e3","M393.9 345.2c-39 9.3-48.4 32.1-104 47.4 0 0-2.7 4-10.4 5.8-13.4 3.3-63.9 6-68.5 6.1-12.4.1-19.9-3.2-22-8.2-6.4-15.3 9.2-22 9.2-22-8.1-5-9-9.9-9.8-8.1-2.4 5.8-3.6 20.1-10.1 26.5-8.8 8.9-25.5 5.9-35.3.8-10.8-5.7.8-19.2.8-19.2s-5.8 3.4-10.5-3.6c-6-9.3-17.1-37.3 11.5-62-1.3-10.1-4.6-53.7 40.6-85.6 0 0-20.6-22.8-12.9-43.3 5-13.4 7-13.3 8.6-13.9 5.7-2.2 11.3-4.6 15.4-9.1 20.6-22.2 46.8-18 46.8-18s12.4-37.8 23.9-30.4c3.5 2.3 16.3 30.6 16.3 30.6s13.6-7.9 15.1-5c8.2 16 9.2 46.5 5.6 65.1-6.1 30.6-21.4 47.1-27.6 57.5-1.4 2.4 16.5 10 27.8 41.3 10.4 28.6 1.1 52.7 2.8 55.3.8 1.4 13.7.8 36.4-13.2 12.8-7.9 28.1-16.9 45.4-17 16.7-.5 17.6 19.2 4.9 22.2zM496 256c0 136.9-111.1 248-248 248S0 392.9 0 256 111.1 8 248 8s248 111.1 248 248zm-79.3 75.2c-1.7-13.6-13.2-23-28-22.8-22 .3-40.5 11.7-52.8 19.2-4.8 3-8.9 5.2-12.4 6.8 3.1-44.5-22.5-73.1-28.7-79.4 7.8-11.3 18.4-27.8 23.4-53.2 4.3-21.7 3-55.5-6.9-74.5-1.6-3.1-7.4-11.2-21-7.4-9.7-20-13-22.1-15.6-23.8-1.1-.7-23.6-16.4-41.4 28-12.2.9-31.3 5.3-47.5 22.8-2 2.2-5.9 3.8-10.1 5.4h.1c-8.4 3-12.3 9.9-16.9 22.3-6.5 17.4.2 34.6 6.8 45.7-17.8 15.9-37 39.8-35.7 82.5-34 36-11.8 73-5.6 79.6-1.6 11.1 3.7 19.4 12 23.8 12.6 6.7 30.3 9.6 43.9 2.8 4.9 5.2 13.8 10.1 30 10.1 6.8 0 58-2.9 72.6-6.5 6.8-1.6 11.5-4.5 14.6-7.1 9.8-3.1 36.8-12.3 62.2-28.7 18-11.7 24.2-14.2 37.6-17.4 12.9-3.2 21-15.1 19.4-28.2z"],"square-steam":[448,512,["steam-square"],"f1b7","M165.6 309.1c18.6 7.7 27.3 28.9 19.6 47.4s-29 27.2-47.6 19.4l-28.5-11.8c5 10.6 13.8 19.4 25.4 24.2c25.2 10.5 54.1-1.4 64.6-26.5c5.1-12.1 5.1-25.5 .1-37.7c-5.1-12.1-14.5-21.6-26.7-26.7c-12.1-5-25-4.8-36.4-.5l29.5 12.2zM448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V240.7l116.6 48.1c12-8.2 26.2-12.1 40.7-11.3l55.4-80.2v-1.1c0-48.2 39.3-87.5 87.6-87.5s87.6 39.3 87.6 87.5c0 49.2-40.9 88.7-89.6 87.5l-79 56.3c1.6 38.5-29.1 68.8-65.7 68.8c-31.8 0-58.5-22.7-64.5-52.7L0 319.2V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM241.9 196.2a58.4 58.4 0 1 0 116.8 0 58.4 58.4 0 1 0 -116.8 0zm14.6-.1a43.9 43.9 0 1 1 87.8 0 43.9 43.9 0 1 1 -87.8 0z"],"500px":[448,512,[],"f26e","M103.3 344.3c-6.5-14.2-6.9-18.3 7.4-23.1 25.6-8 8 9.2 43.2 49.2h.3v-93.9c1.2-50.2 44-92.2 97.7-92.2 53.9 0 97.7 43.5 97.7 96.8 0 63.4-60.8 113.2-128.5 93.3-10.5-4.2-2.1-31.7 8.5-28.6 53 0 89.4-10.1 89.4-64.4 0-61-77.1-89.6-116.9-44.6-23.5 26.4-17.6 42.1-17.6 157.6 50.7 31 118.3 22 160.4-20.1 24.8-24.8 38.5-58 38.5-93 0-35.2-13.8-68.2-38.8-93.3-24.8-24.8-57.8-38.5-93.3-38.5s-68.8 13.8-93.5 38.5c-.3.3-16 16.5-21.2 23.9l-.5.6c-3.3 4.7-6.3 9.1-20.1 6.1-6.9-1.7-14.3-5.8-14.3-11.8V20c0-5 3.9-10.5 10.5-10.5h241.3c8.3 0 8.3 11.6 8.3 15.1 0 3.9 0 15.1-8.3 15.1H130.3v132.9h.3c104.2-109.8 282.8-36 282.8 108.9 0 178.1-244.8 220.3-310.1 62.8zm63.3-260.8c-.5 4.2 4.6 24.5 14.6 20.6C306 56.6 384 144.5 390.6 144.5c4.8 0 22.8-15.3 14.3-22.8-93.2-89-234.5-57-238.3-38.2zM393 414.7C283 524.6 94 475.5 61 310.5c0-12.2-30.4-7.4-28.9 3.3 24 173.4 246 256.9 381.6 121.3 6.9-7.8-12.6-28.4-20.7-20.4zM213.6 306.6c0 4 4.3 7.3 5.5 8.5 3 3 6.1 4.4 8.5 4.4 3.8 0 2.6.2 22.3-19.5 19.6 19.3 19.1 19.5 22.3 19.5 5.4 0 18.5-10.4 10.7-18.2L265.6 284l18.2-18.2c6.3-6.8-10.1-21.8-16.2-15.7L249.7 268c-18.6-18.8-18.4-19.5-21.5-19.5-5 0-18 11.7-12.4 17.3L234 284c-18.1 17.9-20.4 19.2-20.4 22.6z"],"square-vimeo":[448,512,["vimeo-square"],"f194","M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM318.5 309.5C275 366 238.2 394.3 208.1 394.3c-18.7 0-34.4-17.2-47.3-51.6c-3.8-13.9-7.3-26.9-10.5-39c-18.2-68.3-28.6-107.4-46.2-107.4c-2.4 0-10.8 5-25.1 15.1L64 192c7-6.1 13.9-12.4 20.6-18.5c29.1-26.3 55.6-50.3 73.5-51.9c24.9-2.4 40.2 14.6 46 51.1c20.5 129.6 29.6 149.2 66.8 90.5c13.4-21.2 20.6-37.2 21.5-48.3c3.4-32.8-25.6-30.6-45.2-22.2c15.7-51.5 45.8-76.5 90.1-75.1c32.9 1 48.4 22.4 46.5 64c-1.4 31.1-23.2 73.8-65.3 127.9z"],asymmetrik:[576,512,[],"f372","M517.5 309.2c38.8-40 58.1-80 58.5-116.1.8-65.5-59.4-118.2-169.4-135C277.9 38.4 118.1 73.6 0 140.5 52 114 110.6 92.3 170.7 82.3c74.5-20.5 153-25.4 221.3-14.8C544.5 91.3 588.8 195 490.8 299.2c-10.2 10.8-22 21.1-35 30.6L304.9 103.4 114.7 388.9c-65.6-29.4-76.5-90.2-19.1-151.2 20.8-22.2 48.3-41.9 79.5-58.1 20-12.2 39.7-22.6 62-30.7-65.1 20.3-122.7 52.9-161.6 92.9-27.7 28.6-41.4 57.1-41.7 82.9-.5 35.1 23.4 65.1 68.4 83l-34.5 51.7h101.6l22-34.4c22.2 1 45.3 0 68.6-2.7l-22.8 37.1h135.5L340 406.3c18.6-5.3 36.9-11.5 54.5-18.7l45.9 71.8H542L468.6 349c18.5-12.1 35-25.5 48.9-39.8zm-187.6 80.5l-25-40.6-32.7 53.3c-23.4 3.5-46.7 5.1-69.2 4.4l101.9-159.3 78.7 123c-17.2 7.4-35.3 13.9-53.7 19.2z"],"font-awesome":[512,512,[62501,62694,"font-awesome-flag","font-awesome-logo-full"],"f2b4","M91.7 96C106.3 86.8 116 70.5 116 52C116 23.3 92.7 0 64 0S12 23.3 12 52c0 16.7 7.8 31.5 20 41l0 3 0 352 0 64 64 0 0-64 373.6 0c14.6 0 26.4-11.8 26.4-26.4c0-3.7-.8-7.3-2.3-10.7L432 272l61.7-138.9c1.5-3.4 2.3-7 2.3-10.7c0-14.6-11.8-26.4-26.4-26.4L91.7 96z"],gratipay:[496,512,[],"f184","M248 8C111.1 8 0 119.1 0 256s111.1 248 248 248 248-111.1 248-248S384.9 8 248 8zm114.6 226.4l-113 152.7-112.7-152.7c-8.7-11.9-19.1-50.4 13.6-72 28.1-18.1 54.6-4.2 68.5 11.9 15.9 17.9 46.6 16.9 61.7 0 13.9-16.1 40.4-30 68.1-11.9 32.9 21.6 22.6 60 13.8 72z"],apple:[384,512,[],"f179","M318.7 268.7c-.2-36.7 16.4-64.4 50-84.8-18.8-26.9-47.2-41.7-84.7-44.6-35.5-2.8-74.3 20.7-88.5 20.7-15 0-49.4-19.7-76.4-19.7C63.3 141.2 4 184.8 4 273.5q0 39.3 14.4 81.2c12.8 36.7 59 126.7 107.2 125.2 25.2-.6 43-17.9 75.8-17.9 31.8 0 48.3 17.9 76.4 17.9 48.6-.7 90.4-82.5 102.6-119.3-65.2-30.7-61.7-90-61.7-91.9zm-56.6-164.2c27.3-32.4 24.8-61.9 24-72.5-24.1 1.4-52 16.4-67.9 34.9-17.5 19.8-27.8 44.3-25.6 71.9 26.1 2 49.9-11.4 69.5-34.3z"],hive:[512,512,[],"e07f","M260.353,254.878,131.538,33.1a2.208,2.208,0,0,0-3.829.009L.3,254.887A2.234,2.234,0,0,0,.3,257.122L129.116,478.9a2.208,2.208,0,0,0,3.83-.009L260.358,257.113A2.239,2.239,0,0,0,260.353,254.878Zm39.078-25.713a2.19,2.19,0,0,0,1.9,1.111h66.509a2.226,2.226,0,0,0,1.9-3.341L259.115,33.111a2.187,2.187,0,0,0-1.9-1.111H190.707a2.226,2.226,0,0,0-1.9,3.341ZM511.7,254.886,384.9,33.112A2.2,2.2,0,0,0,382.99,32h-66.6a2.226,2.226,0,0,0-1.906,3.34L440.652,256,314.481,476.66a2.226,2.226,0,0,0,1.906,3.34h66.6a2.2,2.2,0,0,0,1.906-1.112L511.7,257.114A2.243,2.243,0,0,0,511.7,254.886ZM366.016,284.917H299.508a2.187,2.187,0,0,0-1.9,1.111l-108.8,190.631a2.226,2.226,0,0,0,1.9,3.341h66.509a2.187,2.187,0,0,0,1.9-1.111l108.8-190.631A2.226,2.226,0,0,0,366.016,284.917Z"],gitkraken:[592,512,[],"f3a6","M565.7 118.1c-2.3-6.1-9.3-9.2-15.3-6.6-5.7 2.4-8.5 8.9-6.3 14.6 10.9 29 16.9 60.5 16.9 93.3 0 134.6-100.3 245.7-230.2 262.7V358.4c7.9-1.5 15.5-3.6 23-6.2v104c106.7-25.9 185.9-122.1 185.9-236.8 0-91.8-50.8-171.8-125.8-213.3-5.7-3.2-13-.9-15.9 5-2.7 5.5-.6 12.2 4.7 15.1 67.9 37.6 113.9 110 113.9 193.2 0 93.3-57.9 173.1-139.8 205.4v-92.2c14.2-4.5 24.9-17.7 24.9-33.5 0-13.1-6.8-24.4-17.3-30.5 8.3-79.5 44.5-58.6 44.5-83.9V170c0-38-87.9-161.8-129-164.7-2.5-.2-5-.2-7.6 0C251.1 8.3 163.2 132 163.2 170v14.8c0 25.3 36.3 4.3 44.5 83.9-10.6 6.1-17.3 17.4-17.3 30.5 0 15.8 10.6 29 24.8 33.5v92.2c-81.9-32.2-139.8-112-139.8-205.4 0-83.1 46-155.5 113.9-193.2 5.4-3 7.4-9.6 4.7-15.1-2.9-5.9-10.1-8.2-15.9-5-75 41.5-125.8 121.5-125.8 213.3 0 114.7 79.2 210.8 185.9 236.8v-104c7.6 2.5 15.1 4.6 23 6.2v123.7C131.4 465.2 31 354.1 31 219.5c0-32.8 6-64.3 16.9-93.3 2.2-5.8-.6-12.2-6.3-14.6-6-2.6-13 .4-15.3 6.6C14.5 149.7 8 183.8 8 219.5c0 155.1 122.6 281.6 276.3 287.8V361.4c6.8.4 15 .5 23.4 0v145.8C461.4 501.1 584 374.6 584 219.5c0-35.7-6.5-69.8-18.3-101.4zM365.9 275.5c13 0 23.7 10.5 23.7 23.7 0 13.1-10.6 23.7-23.7 23.7-13 0-23.7-10.5-23.7-23.7 0-13.1 10.6-23.7 23.7-23.7zm-139.8 47.3c-13.2 0-23.7-10.7-23.7-23.7s10.5-23.7 23.7-23.7c13.1 0 23.7 10.6 23.7 23.7 0 13-10.5 23.7-23.7 23.7z"],keybase:[448,512,[],"f4f5","M286.17 419a18 18 0 1 0 18 18 18 18 0 0 0-18-18zm111.92-147.6c-9.5-14.62-39.37-52.45-87.26-73.71q-9.1-4.06-18.38-7.27a78.43 78.43 0 0 0-47.88-104.13c-12.41-4.1-23.33-6-32.41-5.77-.6-2-1.89-11 9.4-35L198.66 32l-5.48 7.56c-8.69 12.06-16.92 23.55-24.34 34.89a51 51 0 0 0-8.29-1.25c-41.53-2.45-39-2.33-41.06-2.33-50.61 0-50.75 52.12-50.75 45.88l-2.36 36.68c-1.61 27 19.75 50.21 47.63 51.85l8.93.54a214 214 0 0 0-46.29 35.54C14 304.66 14 374 14 429.77v33.64l23.32-29.8a148.6 148.6 0 0 0 14.56 37.56c5.78 10.13 14.87 9.45 19.64 7.33 4.21-1.87 10-6.92 3.75-20.11a178.29 178.29 0 0 1-15.76-53.13l46.82-59.83-24.66 74.11c58.23-42.4 157.38-61.76 236.25-38.59 34.2 10.05 67.45.69 84.74-23.84.72-1 1.2-2.16 1.85-3.22a156.09 156.09 0 0 1 2.8 28.43c0 23.3-3.69 52.93-14.88 81.64-2.52 6.46 1.76 14.5 8.6 15.74 7.42 1.57 15.33-3.1 18.37-11.15C429 443 434 414 434 382.32c0-38.58-13-77.46-35.91-110.92zM142.37 128.58l-15.7-.93-1.39 21.79 13.13.78a93 93 0 0 0 .32 19.57l-22.38-1.34a12.28 12.28 0 0 1-11.76-12.79L107 119c1-12.17 13.87-11.27 13.26-11.32l29.11 1.73a144.35 144.35 0 0 0-7 19.17zm148.42 172.18a10.51 10.51 0 0 1-14.35-1.39l-9.68-11.49-34.42 27a8.09 8.09 0 0 1-11.13-1.08l-15.78-18.64a7.38 7.38 0 0 1 1.34-10.34l34.57-27.18-14.14-16.74-17.09 13.45a7.75 7.75 0 0 1-10.59-1s-3.72-4.42-3.8-4.53a7.38 7.38 0 0 1 1.37-10.34L214 225.19s-18.51-22-18.6-22.14a9.56 9.56 0 0 1 1.74-13.42 10.38 10.38 0 0 1 14.3 1.37l81.09 96.32a9.58 9.58 0 0 1-1.74 13.44zM187.44 419a18 18 0 1 0 18 18 18 18 0 0 0-18-18z"],"apple-pay":[640,512,[],"f415","M116.9 158.5c-7.5 8.9-19.5 15.9-31.5 14.9-1.5-12 4.4-24.8 11.3-32.6 7.5-9.1 20.6-15.6 31.3-16.1 1.2 12.4-3.7 24.7-11.1 33.8m10.9 17.2c-17.4-1-32.3 9.9-40.5 9.9-8.4 0-21-9.4-34.8-9.1-17.9.3-34.5 10.4-43.6 26.5-18.8 32.3-4.9 80 13.3 106.3 8.9 13 19.5 27.3 33.5 26.8 13.3-.5 18.5-8.6 34.5-8.6 16.1 0 20.8 8.6 34.8 8.4 14.5-.3 23.6-13 32.5-26 10.1-14.8 14.3-29.1 14.5-29.9-.3-.3-28-10.9-28.3-42.9-.3-26.8 21.9-39.5 22.9-40.3-12.5-18.6-32-20.6-38.8-21.1m100.4-36.2v194.9h30.3v-66.6h41.9c38.3 0 65.1-26.3 65.1-64.3s-26.4-64-64.1-64h-73.2zm30.3 25.5h34.9c26.3 0 41.3 14 41.3 38.6s-15 38.8-41.4 38.8h-34.8V165zm162.2 170.9c19 0 36.6-9.6 44.6-24.9h.6v23.4h28v-97c0-28.1-22.5-46.3-57.1-46.3-32.1 0-55.9 18.4-56.8 43.6h27.3c2.3-12 13.4-19.9 28.6-19.9 18.5 0 28.9 8.6 28.9 24.5v10.8l-37.8 2.3c-35.1 2.1-54.1 16.5-54.1 41.5.1 25.2 19.7 42 47.8 42zm8.2-23.1c-16.1 0-26.4-7.8-26.4-19.6 0-12.3 9.9-19.4 28.8-20.5l33.6-2.1v11c0 18.2-15.5 31.2-36 31.2zm102.5 74.6c29.5 0 43.4-11.3 55.5-45.4L640 193h-30.8l-35.6 115.1h-.6L537.4 193h-31.6L557 334.9l-2.8 8.6c-4.6 14.6-12.1 20.3-25.5 20.3-2.4 0-7-.3-8.9-.5v23.4c1.8.4 9.3.7 11.6.7z"],padlet:[640,512,[],"e4a0","M297.9 0L298 .001C305.6 .1078 312.4 4.72 315.5 11.78L447.5 320.3L447.8 320.2L448 320.6L445.2 330.6L402.3 488.6C398.6 504.8 382.6 514.9 366.5 511.2L298.1 495.6L229.6 511.2C213.5 514.9 197.5 504.8 193.8 488.6L150.9 330.6L148.2 320.6L148.3 320.2L280.4 11.78C283.4 4.797 290.3 .1837 297.9 .0006L297.9 0zM160.1 322.1L291.1 361.2L298 483.7L305.9 362.2L436.5 322.9L436.7 322.8L305.7 347.9L297.1 27.72L291.9 347.9L160.1 322.1zM426 222.6L520.4 181.6H594.2L437.2 429.2L468.8 320.2L426 222.6zM597.5 181.4L638.9 257.6C642.9 265.1 635 273.5 627.3 269.8L579.7 247.1L597.5 181.4zM127.3 318.5L158.7 430L1.61 154.5C-4.292 144.1 7.128 132.5 17.55 138.3L169.4 222.5L127.3 318.5z"],"amazon-pay":[640,512,[],"f42c","M14 325.3c2.3-4.2 5.2-4.9 9.7-2.5 10.4 5.6 20.6 11.4 31.2 16.7a595.88 595.88 0 0 0 127.4 46.3 616.61 616.61 0 0 0 63.2 11.8 603.33 603.33 0 0 0 95 5.2c17.4-.4 34.8-1.8 52.1-3.8a603.66 603.66 0 0 0 163.3-42.8c2.9-1.2 5.9-2 9.1-1.2 6.7 1.8 9 9 4.1 13.9a70 70 0 0 1-9.6 7.4c-30.7 21.1-64.2 36.4-99.6 47.9a473.31 473.31 0 0 1-75.1 17.6 431 431 0 0 1-53.2 4.8 21.3 21.3 0 0 0-2.5.3H308a21.3 21.3 0 0 0-2.5-.3c-3.6-.2-7.2-.3-10.7-.4a426.3 426.3 0 0 1-50.4-5.3A448.4 448.4 0 0 1 164 420a443.33 443.33 0 0 1-145.6-87c-1.8-1.6-3-3.8-4.4-5.7zM172 65.1l-4.3.6a80.92 80.92 0 0 0-38 15.1c-2.4 1.7-4.6 3.5-7.1 5.4a4.29 4.29 0 0 1-.4-1.4c-.4-2.7-.8-5.5-1.3-8.2-.7-4.6-3-6.6-7.6-6.6h-11.5c-6.9 0-8.2 1.3-8.2 8.2v209.3c0 1 0 2 .1 3 .2 3 2 4.9 4.9 5 7 .1 14.1.1 21.1 0 2.9 0 4.7-2 5-5 .1-1 .1-2 .1-3v-72.4c1.1.9 1.7 1.4 2.2 1.9 17.9 14.9 38.5 19.8 61 15.4 20.4-4 34.6-16.5 43.8-34.9 7-13.9 9.9-28.7 10.3-44.1.5-17.1-1.2-33.9-8.1-49.8-8.5-19.6-22.6-32.5-43.9-36.9-3.2-.7-6.5-1-9.8-1.5-2.8-.1-5.5-.1-8.3-.1zM124.6 107a3.48 3.48 0 0 1 1.7-3.3c13.7-9.5 28.8-14.5 45.6-13.2 14.9 1.1 27.1 8.4 33.5 25.9 3.9 10.7 4.9 21.8 4.9 33 0 10.4-.8 20.6-4 30.6-6.8 21.3-22.4 29.4-42.6 28.5-14-.6-26.2-6-37.4-13.9a3.57 3.57 0 0 1-1.7-3.3c.1-14.1 0-28.1 0-42.2s.1-28 0-42.1zm205.7-41.9c-1 .1-2 .3-2.9.4a148 148 0 0 0-28.9 4.1c-6.1 1.6-12 3.8-17.9 5.8-3.6 1.2-5.4 3.8-5.3 7.7.1 3.3-.1 6.6 0 9.9.1 4.8 2.1 6.1 6.8 4.9 7.8-2 15.6-4.2 23.5-5.7 12.3-2.3 24.7-3.3 37.2-1.4 6.5 1 12.6 2.9 16.8 8.4 3.7 4.8 5.1 10.5 5.3 16.4.3 8.3.2 16.6.3 24.9a7.84 7.84 0 0 1-.2 1.4c-.5-.1-.9 0-1.3-.1a180.56 180.56 0 0 0-32-4.9c-11.3-.6-22.5.1-33.3 3.9-12.9 4.5-23.3 12.3-29.4 24.9-4.7 9.8-5.4 20.2-3.9 30.7 2 14 9 24.8 21.4 31.7 11.9 6.6 24.8 7.4 37.9 5.4 15.1-2.3 28.5-8.7 40.3-18.4a7.36 7.36 0 0 1 1.6-1.1c.6 3.8 1.1 7.4 1.8 11 .6 3.1 2.5 5.1 5.4 5.2 5.4.1 10.9.1 16.3 0a4.84 4.84 0 0 0 4.8-4.7 26.2 26.2 0 0 0 .1-2.8v-106a80 80 0 0 0-.9-12.9c-1.9-12.9-7.4-23.5-19-30.4-6.7-4-14.1-6-21.8-7.1-3.6-.5-7.2-.8-10.8-1.3-3.9.1-7.9.1-11.9.1zm35 127.7a3.33 3.33 0 0 1-1.5 3c-11.2 8.1-23.5 13.5-37.4 14.9-5.7.6-11.4.4-16.8-1.8a20.08 20.08 0 0 1-12.4-13.3 32.9 32.9 0 0 1-.1-19.4c2.5-8.3 8.4-13 16.4-15.6a61.33 61.33 0 0 1 24.8-2.2c8.4.7 16.6 2.3 25 3.4 1.6.2 2.1 1 2.1 2.6-.1 4.8 0 9.5 0 14.3s-.2 9.4-.1 14.1zm259.9 129.4c-1-5-4.8-6.9-9.1-8.3a88.42 88.42 0 0 0-21-3.9 147.32 147.32 0 0 0-39.2 1.9c-14.3 2.7-27.9 7.3-40 15.6a13.75 13.75 0 0 0-3.7 3.5 5.11 5.11 0 0 0-.5 4c.4 1.5 2.1 1.9 3.6 1.8a16.2 16.2 0 0 0 2.2-.1c7.8-.8 15.5-1.7 23.3-2.5 11.4-1.1 22.9-1.8 34.3-.9a71.64 71.64 0 0 1 14.4 2.7c5.1 1.4 7.4 5.2 7.6 10.4.4 8-1.4 15.7-3.5 23.3-4.1 15.4-10 30.3-15.8 45.1a17.6 17.6 0 0 0-1 3c-.5 2.9 1.2 4.8 4.1 4.1a10.56 10.56 0 0 0 4.8-2.5 145.91 145.91 0 0 0 12.7-13.4c12.8-16.4 20.3-35.3 24.7-55.6.8-3.6 1.4-7.3 2.1-10.9v-17.3zM493.1 199q-19.35-53.55-38.7-107.2c-2-5.7-4.2-11.3-6.3-16.9-1.1-2.9-3.2-4.8-6.4-4.8-7.6-.1-15.2-.2-22.9-.1-2.5 0-3.7 2-3.2 4.5a43.1 43.1 0 0 0 1.9 6.1q29.4 72.75 59.1 145.5c1.7 4.1 2.1 7.6.2 11.8-3.3 7.3-5.9 15-9.3 22.3-3 6.5-8 11.4-15.2 13.3a42.13 42.13 0 0 1-15.4 1.1c-2.5-.2-5-.8-7.5-1-3.4-.2-5.1 1.3-5.2 4.8q-.15 5 0 9.9c.1 5.5 2 8 7.4 8.9a108.18 108.18 0 0 0 16.9 2c17.1.4 30.7-6.5 39.5-21.4a131.63 131.63 0 0 0 9.2-18.4q35.55-89.7 70.6-179.6a26.62 26.62 0 0 0 1.6-5.5c.4-2.8-.9-4.4-3.7-4.4-6.6-.1-13.3 0-19.9 0a7.54 7.54 0 0 0-7.7 5.2c-.5 1.4-1.1 2.7-1.6 4.1l-34.8 100c-2.5 7.2-5.1 14.5-7.7 22.2-.4-1.1-.6-1.7-.9-2.4z"],"square-github":[448,512,["github-square"],"f092","M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM265.8 407.7c0-1.8 0-6 .1-11.6c.1-11.4 .1-28.8 .1-43.7c0-15.6-5.2-25.5-11.3-30.7c37-4.1 76-9.2 76-73.1c0-18.2-6.5-27.3-17.1-39c1.7-4.3 7.4-22-1.7-45c-13.9-4.3-45.7 17.9-45.7 17.9c-13.2-3.7-27.5-5.6-41.6-5.6s-28.4 1.9-41.6 5.6c0 0-31.8-22.2-45.7-17.9c-9.1 22.9-3.5 40.6-1.7 45c-10.6 11.7-15.6 20.8-15.6 39c0 63.6 37.3 69 74.3 73.1c-4.8 4.3-9.1 11.7-10.6 22.3c-9.5 4.3-33.8 11.7-48.3-13.9c-9.1-15.8-25.5-17.1-25.5-17.1c-16.2-.2-1.1 10.2-1.1 10.2c10.8 5 18.4 24.2 18.4 24.2c9.7 29.7 56.1 19.7 56.1 19.7c0 9 .1 21.7 .1 30.6c0 4.8 .1 8.6 .1 10c0 4.3-3 9.5-11.5 8C106 393.6 59.8 330.8 59.8 257.4c0-91.8 70.2-161.5 162-161.5s166.2 69.7 166.2 161.5c.1 73.4-44.7 136.3-110.7 158.3c-8.4 1.5-11.5-3.7-11.5-8zm-90.5-54.8c-.2-1.5 1.1-2.8 3-3.2c1.9-.2 3.7 .6 3.9 1.9c.3 1.3-1 2.6-3 3c-1.9 .4-3.7-.4-3.9-1.7zm-9.1 3.2c-2.2 .2-3.7-.9-3.7-2.4c0-1.3 1.5-2.4 3.5-2.4c1.9-.2 3.7 .9 3.7 2.4c0 1.3-1.5 2.4-3.5 2.4zm-14.3-2.2c-1.9-.4-3.2-1.9-2.8-3.2s2.4-1.9 4.1-1.5c2 .6 3.3 2.1 2.8 3.4c-.4 1.3-2.4 1.9-4.1 1.3zm-12.5-7.3c-1.5-1.3-1.9-3.2-.9-4.1c.9-1.1 2.8-.9 4.3 .6c1.3 1.3 1.8 3.3 .9 4.1c-.9 1.1-2.8 .9-4.3-.6zm-8.5-10c-1.1-1.5-1.1-3.2 0-3.9c1.1-.9 2.8-.2 3.7 1.3c1.1 1.5 1.1 3.3 0 4.1c-.9 .6-2.6 0-3.7-1.5zm-6.3-8.8c-1.1-1.3-1.3-2.8-.4-3.5c.9-.9 2.4-.4 3.5 .6c1.1 1.3 1.3 2.8 .4 3.5c-.9 .9-2.4 .4-3.5-.6zm-6-6.4c-1.3-.6-1.9-1.7-1.5-2.6c.4-.6 1.5-.9 2.8-.4c1.3 .7 1.9 1.8 1.5 2.6c-.4 .9-1.7 1.1-2.8 .4z"],stumbleupon:[512,512,[],"f1a4","M502.9 266v69.7c0 62.1-50.3 112.4-112.4 112.4-61.8 0-112.4-49.8-112.4-111.3v-70.2l34.3 16 51.1-15.2V338c0 14.7 12 26.5 26.7 26.5S417 352.7 417 338v-72h85.9zm-224.7-58.2l34.3 16 51.1-15.2V173c0-60.5-51.1-109-112.1-109-60.8 0-112.1 48.2-112.1 108.2v162.4c0 14.9-12 26.7-26.7 26.7S86 349.5 86 334.6V266H0v69.7C0 397.7 50.3 448 112.4 448c61.6 0 112.4-49.5 112.4-110.8V176.9c0-14.7 12-26.7 26.7-26.7s26.7 12 26.7 26.7v30.9z"],fedex:[640,512,[],"f797","M586 284.5l53.3-59.9h-62.4l-21.7 24.8-22.5-24.8H414v-16h56.1v-48.1H318.9V236h-.5c-9.6-11-21.5-14.8-35.4-14.8-28.4 0-49.8 19.4-57.3 44.9-18-59.4-97.4-57.6-121.9-14v-24.2H49v-26.2h60v-41.1H0V345h49v-77.5h48.9c-1.5 5.7-2.3 11.8-2.3 18.2 0 73.1 102.6 91.4 130.2 23.7h-42c-14.7 20.9-45.8 8.9-45.8-14.6h85.5c3.7 30.5 27.4 56.9 60.1 56.9 14.1 0 27-6.9 34.9-18.6h.5V345h212.2l22.1-25 22.3 25H640l-54-60.5zm-446.7-16.6c6.1-26.3 41.7-25.6 46.5 0h-46.5zm153.4 48.9c-34.6 0-34-62.8 0-62.8 32.6 0 34.5 62.8 0 62.8zm167.8 19.1h-94.4V169.4h95v30.2H405v33.9h55.5v28.1h-56.1v44.7h56.1v29.6zm-45.9-39.8v-24.4h56.1v-44l50.7 57-50.7 57v-45.6h-56.1zm138.6 10.3l-26.1 29.5H489l45.6-51.2-45.6-51.2h39.7l26.6 29.3 25.6-29.3h38.5l-45.4 51 46 51.4h-40.5l-26.3-29.5z"],"phoenix-framework":[640,512,[],"f3dc","M212.9 344.3c3.8-.1 22.8-1.4 25.6-2.2-2.4-2.6-43.6-1-68-49.6-4.3-8.6-7.5-17.6-6.4-27.6 2.9-25.5 32.9-30 52-18.5 36 21.6 63.3 91.3 113.7 97.5 37 4.5 84.6-17 108.2-45.4-.6-.1-.8-.2-1-.1-.4.1-.8.2-1.1.3-33.3 12.1-94.3 9.7-134.7-14.8-37.6-22.8-53.1-58.7-51.8-74.6 1.8-21.3 22.9-23.2 35.9-19.6 14.4 3.9 24.4 17.6 38.9 27.4 15.6 10.4 32.9 13.7 51.3 10.3 14.9-2.7 34.4-12.3 36.5-14.5-1.1-.1-1.8-.1-2.5-.2-6.2-.6-12.4-.8-18.5-1.7C279.8 194.5 262.1 47.4 138.5 37.9 94.2 34.5 39.1 46 2.2 72.9c-.8.6-1.5 1.2-2.2 1.8.1.2.1.3.2.5.8 0 1.6-.1 2.4-.2 6.3-1 12.5-.8 18.7.3 23.8 4.3 47.7 23.1 55.9 76.5 5.3 34.3-.7 50.8 8 86.1 19 77.1 91 107.6 127.7 106.4zM75.3 64.9c-.9-1-.9-1.2-1.3-2 12.1-2.6 24.2-4.1 36.6-4.8-1.1 14.7-22.2 21.3-35.3 6.8zm196.9 350.5c-42.8 1.2-92-26.7-123.5-61.4-4.6-5-16.8-20.2-18.6-23.4l.4-.4c6.6 4.1 25.7 18.6 54.8 27 24.2 7 48.1 6.3 71.6-3.3 22.7-9.3 41-.5 43.1 2.9-18.5 3.8-20.1 4.4-24 7.9-5.1 4.4-4.6 11.7 7 17.2 26.2 12.4 63-2.8 97.2 25.4 2.4 2 8.1 7.8 10.1 10.7-.1.2-.3.3-.4.5-4.8-1.5-16.4-7.5-40.2-9.3-24.7-2-46.3 5.3-77.5 6.2zm174.8-252c16.4-5.2 41.3-13.4 66.5-3.3 16.1 6.5 26.2 18.7 32.1 34.6 3.5 9.4 5.1 19.7 5.1 28.7-.2 0-.4 0-.6.1-.2-.4-.4-.9-.5-1.3-5-22-29.9-43.8-67.6-29.9-50.2 18.6-130.4 9.7-176.9-48-.7-.9-2.4-1.7-1.3-3.2.1-.2 2.1.6 3 1.3 18.1 13.4 38.3 21.9 60.3 26.2 30.5 6.1 54.6 2.9 79.9-5.2zm102.7 117.5c-32.4.2-33.8 50.1-103.6 64.4-18.2 3.7-38.7 4.6-44.9 4.2v-.4c2.8-1.5 14.7-2.6 29.7-16.6 7.9-7.3 15.3-15.1 22.8-22.9 19.5-20.2 41.4-42.2 81.9-39 23.1 1.8 29.3 8.2 36.1 12.7.3.2.4.5.7.9-.5 0-.7.1-.9 0-7-2.7-14.3-3.3-21.8-3.3zm-12.3-24.1c-.1.2-.1.4-.2.6-28.9-4.4-48-7.9-68.5 4-17 9.9-31.4 20.5-62 24.4-27.1 3.4-45.1 2.4-66.1-8-.3-.2-.6-.4-1-.6 0-.2.1-.3.1-.5 24.9 3.8 36.4 5.1 55.5-5.8 22.3-12.9 40.1-26.6 71.3-31 29.6-4.1 51.3 2.5 70.9 16.9zM268.6 97.3c-.6-.6-1.1-1.2-2.1-2.3 7.6 0 29.7-1.2 53.4 8.4 19.7 8 32.2 21 50.2 32.9 11.1 7.3 23.4 9.3 36.4 8.1 4.3-.4 8.5-1.2 12.8-1.7.4-.1.9 0 1.5.3-.6.4-1.2.9-1.8 1.2-8.1 4-16.7 6.3-25.6 7.1-26.1 2.6-50.3-3.7-73.4-15.4-19.3-9.9-36.4-22.9-51.4-38.6zM640 335.7c-3.5 3.1-22.7 11.6-42.7 5.3-12.3-3.9-19.5-14.9-31.6-24.1-10-7.6-20.9-7.9-28.1-8.4.6-.8.9-1.2 1.2-1.4 14.8-9.2 30.5-12.2 47.3-6.5 12.5 4.2 19.2 13.5 30.4 24.2 10.8 10.4 21 9.9 23.1 10.5.1-.1.2 0 .4.4zm-212.5 137c2.2 1.2 1.6 1.5 1.5 2-18.5-1.4-33.9-7.6-46.8-22.2-21.8-24.7-41.7-27.9-48.6-29.7.5-.2.8-.4 1.1-.4 13.1.1 26.1.7 38.9 3.9 25.3 6.4 35 25.4 41.6 35.3 3.2 4.8 7.3 8.3 12.3 11.1z"],shopify:[448,512,[],"e057","M388.32,104.1a4.66,4.66,0,0,0-4.4-4c-2,0-37.23-.8-37.23-.8s-21.61-20.82-29.62-28.83V503.2L442.76,472S388.72,106.5,388.32,104.1ZM288.65,70.47a116.67,116.67,0,0,0-7.21-17.61C271,32.85,255.42,22,237,22a15,15,0,0,0-4,.4c-.4-.8-1.2-1.2-1.6-2C223.4,11.63,213,7.63,200.58,8c-24,.8-48,18-67.25,48.83-13.61,21.62-24,48.84-26.82,70.06-27.62,8.4-46.83,14.41-47.23,14.81-14,4.4-14.41,4.8-16,18-1.2,10-38,291.82-38,291.82L307.86,504V65.67a41.66,41.66,0,0,0-4.4.4S297.86,67.67,288.65,70.47ZM233.41,87.69c-16,4.8-33.63,10.4-50.84,15.61,4.8-18.82,14.41-37.63,25.62-50,4.4-4.4,10.41-9.61,17.21-12.81C232.21,54.86,233.81,74.48,233.41,87.69ZM200.58,24.44A27.49,27.49,0,0,1,215,28c-6.4,3.2-12.81,8.41-18.81,14.41-15.21,16.42-26.82,42-31.62,66.45-14.42,4.41-28.83,8.81-42,12.81C131.33,83.28,163.75,25.24,200.58,24.44ZM154.15,244.61c1.6,25.61,69.25,31.22,73.25,91.66,2.8,47.64-25.22,80.06-65.65,82.47-48.83,3.2-75.65-25.62-75.65-25.62l10.4-44s26.82,20.42,48.44,18.82c14-.8,19.22-12.41,18.81-20.42-2-33.62-57.24-31.62-60.84-86.86-3.2-46.44,27.22-93.27,94.47-97.68,26-1.6,39.23,4.81,39.23,4.81L221.4,225.39s-17.21-8-37.63-6.4C154.15,221,153.75,239.8,154.15,244.61ZM249.42,82.88c0-12-1.6-29.22-7.21-43.63,18.42,3.6,27.22,24,31.23,36.43Q262.63,78.68,249.42,82.88Z"],neos:[512,512,[],"f612","M415.44 512h-95.11L212.12 357.46v91.1L125.69 512H28V29.82L68.47 0h108.05l123.74 176.13V63.45L386.69 0h97.69v461.5zM38.77 35.27V496l72-52.88V194l215.5 307.64h84.79l52.35-38.17h-78.27L69 13zm82.54 466.61l80-58.78v-101l-79.76-114.4v220.94L49 501.89h72.34zM80.63 10.77l310.6 442.57h82.37V10.77h-79.75v317.56L170.91 10.77zM311 191.65l72 102.81V15.93l-72 53v122.72z"],"square-threads":[448,512,[],"e619","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM294.2 244.3c19.5 9.3 33.7 23.5 41.2 40.9c10.4 24.3 11.4 63.9-20.2 95.4c-24.2 24.1-53.5 35-95.1 35.3h-.2c-46.8-.3-82.8-16.1-106.9-46.8C91.5 341.8 80.4 303.7 80 256v-.1-.1c.4-47.7 11.5-85.7 33-113.1c24.2-30.7 60.2-46.5 106.9-46.8h.2c46.9 .3 83.3 16 108.2 46.6c12.3 15.1 21.3 33.3 27 54.4l-26.9 7.2c-4.7-17.2-11.9-31.9-21.4-43.6c-19.4-23.9-48.7-36.1-87-36.4c-38 .3-66.8 12.5-85.5 36.2c-17.5 22.3-26.6 54.4-26.9 95.5c.3 41.1 9.4 73.3 26.9 95.5c18.7 23.8 47.4 36 85.5 36.2c34.3-.3 56.9-8.4 75.8-27.3c21.5-21.5 21.1-47.9 14.2-64c-4-9.4-11.4-17.3-21.3-23.3c-2.4 18-7.9 32.2-16.5 43.2c-11.4 14.5-27.7 22.4-48.4 23.5c-15.7 .9-30.8-2.9-42.6-10.7c-13.9-9.2-22-23.2-22.9-39.5c-1.7-32.2 23.8-55.3 63.5-57.6c14.1-.8 27.3-.2 39.5 1.9c-1.6-9.9-4.9-17.7-9.8-23.4c-6.7-7.8-17.1-11.8-30.8-11.9h-.4c-11 0-26 3.1-35.6 17.6l-23-15.8c12.8-19.4 33.6-30.1 58.5-30.1h.6c41.8 .3 66.6 26.3 69.1 71.8c1.4 .6 2.8 1.2 4.2 1.9l.1 .5zm-71.8 67.5c17-.9 36.4-7.6 39.7-48.8c-8.8-1.9-18.6-2.9-29-2.9c-3.2 0-6.4 .1-9.6 .3c-28.6 1.6-38.1 15.5-37.4 27.9c.9 16.7 19 24.5 36.4 23.6l-.1-.1z"],hackerrank:[512,512,[],"f5f7","M477.5 128C463 103.05 285.13 0 256.16 0S49.25 102.79 34.84 128s-14.49 230.8 0 256 192.38 128 221.32 128S463 409.08 477.49 384s14.51-231 .01-256zM316.13 414.22c-4 0-40.91-35.77-38-38.69.87-.87 6.26-1.48 17.55-1.83 0-26.23.59-68.59.94-86.32 0-2-.44-3.43-.44-5.85h-79.93c0 7.1-.46 36.2 1.37 72.88.23 4.54-1.58 6-5.74 5.94-10.13 0-20.27-.11-30.41-.08-4.1 0-5.87-1.53-5.74-6.11.92-33.44 3-84-.15-212.67v-3.17c-9.67-.35-16.38-1-17.26-1.84-2.92-2.92 34.54-38.69 38.49-38.69s41.17 35.78 38.27 38.69c-.87.87-7.9 1.49-16.77 1.84v3.16c-2.42 25.75-2 79.59-2.63 105.39h80.26c0-4.55.39-34.74-1.2-83.64-.1-3.39.95-5.17 4.21-5.2 11.07-.08 22.15-.13 33.23-.06 3.46 0 4.57 1.72 4.5 5.38C333 354.64 336 341.29 336 373.69c8.87.35 16.82 1 17.69 1.84 2.88 2.91-33.62 38.69-37.58 38.69z"],researchgate:[448,512,[],"f4f8","M0 32v448h448V32H0zm262.2 334.4c-6.6 3-33.2 6-50-14.2-9.2-10.6-25.3-33.3-42.2-63.6-8.9 0-14.7 0-21.4-.6v46.4c0 23.5 6 21.2 25.8 23.9v8.1c-6.9-.3-23.1-.8-35.6-.8-13.1 0-26.1.6-33.6.8v-8.1c15.5-2.9 22-1.3 22-23.9V225c0-22.6-6.4-21-22-23.9V193c25.8 1 53.1-.6 70.9-.6 31.7 0 55.9 14.4 55.9 45.6 0 21.1-16.7 42.2-39.2 47.5 13.6 24.2 30 45.6 42.2 58.9 7.2 7.8 17.2 14.7 27.2 14.7v7.3zm22.9-135c-23.3 0-32.2-15.7-32.2-32.2V167c0-12.2 8.8-30.4 34-30.4s30.4 17.9 30.4 17.9l-10.7 7.2s-5.5-12.5-19.7-12.5c-7.9 0-19.7 7.3-19.7 19.7v26.8c0 13.4 6.6 23.3 17.9 23.3 14.1 0 21.5-10.9 21.5-26.8h-17.9v-10.7h30.4c0 20.5 4.7 49.9-34 49.9zm-116.5 44.7c-9.4 0-13.6-.3-20-.8v-69.7c6.4-.6 15-.6 22.5-.6 23.3 0 37.2 12.2 37.2 34.5 0 21.9-15 36.6-39.7 36.6z"],swift:[448,512,[],"f8e1","M448 156.09c0-4.51-.08-9-.2-13.52a196.31 196.31 0 0 0-2.58-29.42 99.62 99.62 0 0 0-9.22-28A94.08 94.08 0 0 0 394.84 44a99.17 99.17 0 0 0-28-9.22 195 195 0 0 0-29.43-2.59c-4.51-.12-9-.17-13.52-.2H124.14c-4.51 0-9 .08-13.52.2-2.45.07-4.91.15-7.37.27a171.68 171.68 0 0 0-22.06 2.32 103.06 103.06 0 0 0-21.21 6.1q-3.46 1.45-6.81 3.12a94.66 94.66 0 0 0-18.39 12.32c-1.88 1.61-3.69 3.28-5.43 5A93.86 93.86 0 0 0 12 85.17a99.45 99.45 0 0 0-9.22 28 196.31 196.31 0 0 0-2.54 29.4c-.13 4.51-.18 9-.21 13.52v199.83c0 4.51.08 9 .21 13.51a196.08 196.08 0 0 0 2.58 29.42 99.3 99.3 0 0 0 9.22 28A94.31 94.31 0 0 0 53.17 468a99.47 99.47 0 0 0 28 9.21 195 195 0 0 0 29.43 2.59c4.5.12 9 .17 13.52.2H323.91c4.51 0 9-.08 13.52-.2a196.59 196.59 0 0 0 29.44-2.59 99.57 99.57 0 0 0 28-9.21A94.22 94.22 0 0 0 436 426.84a99.3 99.3 0 0 0 9.22-28 194.79 194.79 0 0 0 2.59-29.42c.12-4.5.17-9 .2-13.51V172.14c-.01-5.35-.01-10.7-.01-16.05zm-69.88 241c-20-38.93-57.23-29.27-76.31-19.47-1.72 1-3.48 2-5.25 3l-.42.25c-39.5 21-92.53 22.54-145.85-.38A234.64 234.64 0 0 1 45 290.12a230.63 230.63 0 0 0 39.17 23.37c56.36 26.4 113 24.49 153 0-57-43.85-104.6-101-141.09-147.22a197.09 197.09 0 0 1-18.78-25.9c43.7 40 112.7 90.22 137.48 104.12-52.57-55.49-98.89-123.94-96.72-121.74 82.79 83.42 159.18 130.59 159.18 130.59 2.88 1.58 5 2.85 6.73 4a127.44 127.44 0 0 0 4.16-12.47c13.22-48.33-1.66-103.58-35.31-149.2C329.61 141.75 375 229.34 356.4 303.42c-.44 1.73-.95 3.4-1.44 5.09 38.52 47.4 28.04 98.17 23.13 88.59z"],angular:[448,512,[],"f420","M185.7 268.1h76.2l-38.1-91.6-38.1 91.6zM223.8 32L16 106.4l31.8 275.7 176 97.9 176-97.9 31.8-275.7zM354 373.8h-48.6l-26.2-65.4H168.6l-26.2 65.4H93.7L223.8 81.5z"],speakap:[448,512,[],"f3f3","M64 391.78C-15.41 303.59-8 167.42 80.64 87.64s224.8-73 304.21 15.24 72 224.36-16.64 304.14c-18.74 16.87 64 43.09 42 52.26-82.06 34.21-253.91 35-346.23-67.5zm213.31-211.6l38.5-40.86c-9.61-8.89-32-26.83-76.17-27.6-52.33-.91-95.86 28.3-96.77 80-.2 11.33.29 36.72 29.42 54.83 34.46 21.42 86.52 21.51 86 52.26-.37 21.28-26.42 25.81-38.59 25.6-3-.05-30.23-.46-47.61-24.62l-40 42.61c28.16 27 59 32.62 83.49 33.05 10.23.18 96.42.33 97.84-81 .28-15.81-2.07-39.72-28.86-56.59-34.36-21.64-85-19.45-84.43-49.75.41-23.25 31-25.37 37.53-25.26.43 0 26.62.26 39.62 17.37z"],angrycreative:[640,512,[],"f36e","M640 238.2l-3.2 28.2-34.5 2.3-2 18.1 34.5-2.3-3.2 28.2-34.4 2.2-2.3 20.1 34.4-2.2-3 26.1-64.7 4.1 12.7-113.2L527 365.2l-31.9 2-23.8-117.8 30.3-2 13.6 79.4 31.7-82.4 93.1-6.2zM426.8 371.5l28.3-1.8L468 249.6l-28.4 1.9-12.8 120zM162 388.1l-19.4-36-3.5 37.4-28.2 1.7 2.7-29.1c-11 18-32 34.3-56.9 35.8C23.9 399.9-3 377 .3 339.7c2.6-29.3 26.7-62.8 67.5-65.4 37.7-2.4 47.6 23.2 51.3 28.8l2.8-30.8 38.9-2.5c20.1-1.3 38.7 3.7 42.5 23.7l2.6-26.6 64.8-4.2-2.7 27.9-36.4 2.4-1.7 17.9 36.4-2.3-2.7 27.9-36.4 2.3-1.9 19.9 36.3-2.3-2.1 20.8 55-117.2 23.8-1.6L370.4 369l8.9-85.6-22.3 1.4 2.9-27.9 75-4.9-3 28-24.3 1.6-9.7 91.9-58 3.7-4.3-15.6-39.4 2.5-8 16.3-126.2 7.7zm-44.3-70.2l-26.4 1.7C84.6 307.2 76.9 303 65 303.8c-19 1.2-33.3 17.5-34.6 33.3-1.4 16 7.3 32.5 28.7 31.2 12.8-.8 21.3-8.6 28.9-18.9l27-1.7 2.7-29.8zm56.1-7.7c1.2-12.9-7.6-13.6-26.1-12.4l-2.7 28.5c14.2-.9 27.5-2.1 28.8-16.1zm21.1 70.8l5.8-60c-5 13.5-14.7 21.1-27.9 26.6l22.1 33.4zm135.4-45l-7.9-37.8-15.8 39.3 23.7-1.5zm-170.1-74.6l-4.3-17.5-39.6 2.6-8.1 18.2-31.9 2.1 57-121.9 23.9-1.6 30.7 102 9.9-104.7 27-1.8 37.8 63.6 6.5-66.6 28.5-1.9-4 41.2c7.4-13.5 22.9-44.7 63.6-47.5 40.5-2.8 52.4 29.3 53.4 30.3l3.3-32 39.3-2.7c12.7-.9 27.8.3 36.3 9.7l-4.4-11.9 32.2-2.2 12.9 43.2 23-45.7 31-2.2-43.6 78.4-4.8 44.3-28.4 1.9 4.8-44.3-15.8-43c1 22.3-9.2 40.1-32 49.6l25.2 38.8-36.4 2.4-19.2-36.8-4 38.3-28.4 1.9 3.3-31.5c-6.7 9.3-19.7 35.4-59.6 38-26.2 1.7-45.6-10.3-55.4-39.2l-4 40.3-25 1.6-37.6-63.3-6.3 66.2-56.8 3.7zm276.6-82.1c10.2-.7 17.5-2.1 21.6-4.3 4.5-2.4 7-6.4 7.6-12.1.6-5.3-.6-8.8-3.4-10.4-3.6-2.1-10.6-2.8-22.9-2l-2.9 28.8zM327.7 214c5.6 5.9 12.7 8.5 21.3 7.9 4.7-.3 9.1-1.8 13.3-4.1 5.5-3 10.6-8 15.1-14.3l-34.2 2.3 2.4-23.9 63.1-4.3 1.2-12-31.2 2.1c-4.1-3.7-7.8-6.6-11.1-8.1-4-1.7-8.1-2.8-12.2-2.5-8 .5-15.3 3.6-22 9.2-7.7 6.4-12 14.5-12.9 24.4-1.1 9.6 1.4 17.3 7.2 23.3zm-201.3 8.2l23.8-1.6-8.3-37.6-15.5 39.2z"],"y-combinator":[448,512,[],"f23b","M448 32v448H0V32h448zM236 287.5L313.5 142h-32.7L235 233c-4.7 9.3-9 18.3-12.8 26.8L210 233l-45.2-91h-35l76.7 143.8v94.5H236v-92.8z"],empire:[496,512,[],"f1d1","M287.6 54.2c-10.8-2.2-22.1-3.3-33.5-3.6V32.4c78.1 2.2 146.1 44 184.6 106.6l-15.8 9.1c-6.1-9.7-12.7-18.8-20.2-27.1l-18 15.5c-26-29.6-61.4-50.7-101.9-58.4l4.8-23.9zM53.4 322.4l23-7.7c-6.4-18.3-10-38.2-10-58.7s3.3-40.4 9.7-58.7l-22.7-7.7c3.6-10.8 8.3-21.3 13.6-31l-15.8-9.1C34 181 24.1 217.5 24.1 256s10 75 27.1 106.6l15.8-9.1c-5.3-10-9.7-20.3-13.6-31.1zM213.1 434c-40.4-8-75.8-29.1-101.9-58.7l-18 15.8c-7.5-8.6-14.4-17.7-20.2-27.4l-16 9.4c38.5 62.3 106.8 104.3 184.9 106.6v-18.3c-11.3-.3-22.7-1.7-33.5-3.6l4.7-23.8zM93.3 120.9l18 15.5c26-29.6 61.4-50.7 101.9-58.4l-4.7-23.8c10.8-2.2 22.1-3.3 33.5-3.6V32.4C163.9 34.6 95.9 76.4 57.4 139l15.8 9.1c6-9.7 12.6-18.9 20.1-27.2zm309.4 270.2l-18-15.8c-26 29.6-61.4 50.7-101.9 58.7l4.7 23.8c-10.8 1.9-22.1 3.3-33.5 3.6v18.3c78.1-2.2 146.4-44.3 184.9-106.6l-16.1-9.4c-5.7 9.7-12.6 18.8-20.1 27.4zM496 256c0 137-111 248-248 248S0 393 0 256 111 8 248 8s248 111 248 248zm-12.2 0c0-130.1-105.7-235.8-235.8-235.8S12.2 125.9 12.2 256 117.9 491.8 248 491.8 483.8 386.1 483.8 256zm-39-106.6l-15.8 9.1c5.3 9.7 10 20.2 13.6 31l-22.7 7.7c6.4 18.3 9.7 38.2 9.7 58.7s-3.6 40.4-10 58.7l23 7.7c-3.9 10.8-8.3 21-13.6 31l15.8 9.1C462 331 471.9 294.5 471.9 256s-9.9-75-27.1-106.6zm-183 177.7c16.3-3.3 30.4-11.6 40.7-23.5l51.2 44.8c11.9-13.6 21.3-29.3 27.1-46.8l-64.2-22.1c2.5-7.5 3.9-15.2 3.9-23.5s-1.4-16.1-3.9-23.5l64.5-22.1c-6.1-17.4-15.5-33.2-27.4-46.8l-51.2 44.8c-10.2-11.9-24.4-20.5-40.7-23.8l13.3-66.4c-8.6-1.9-17.7-2.8-27.1-2.8-9.4 0-18.5.8-27.1 2.8l13.3 66.4c-16.3 3.3-30.4 11.9-40.7 23.8l-51.2-44.8c-11.9 13.6-21.3 29.3-27.4 46.8l64.5 22.1c-2.5 7.5-3.9 15.2-3.9 23.5s1.4 16.1 3.9 23.5l-64.2 22.1c5.8 17.4 15.2 33.2 27.1 46.8l51.2-44.8c10.2 11.9 24.4 20.2 40.7 23.5l-13.3 66.7c8.6 1.7 17.7 2.8 27.1 2.8 9.4 0 18.5-1.1 27.1-2.8l-13.3-66.7z"],envira:[448,512,[],"f299","M0 32c477.6 0 366.6 317.3 367.1 366.3L448 480h-26l-70.4-71.2c-39 4.2-124.4 34.5-214.4-37C47 300.3 52 214.7 0 32zm79.7 46c-49.7-23.5-5.2 9.2-5.2 9.2 45.2 31.2 66 73.7 90.2 119.9 31.5 60.2 79 139.7 144.2 167.7 65 28 34.2 12.5 6-8.5-28.2-21.2-68.2-87-91-130.2-31.7-60-61-118.6-144.2-158.1z"],"google-scholar":[512,512,[],"e63b","M390.9 298.5c0 0 0 .1 .1 .1c9.2 19.4 14.4 41.1 14.4 64C405.3 445.1 338.5 512 256 512s-149.3-66.9-149.3-149.3c0-22.9 5.2-44.6 14.4-64h0c1.7-3.6 3.6-7.2 5.6-10.7c4.4-7.6 9.4-14.7 15-21.3c27.4-32.6 68.5-53.3 114.4-53.3c33.6 0 64.6 11.1 89.6 29.9c9.1 6.9 17.4 14.7 24.8 23.5c5.6 6.6 10.6 13.8 15 21.3c2 3.4 3.8 7 5.5 10.5zm26.4-18.8c-30.1-58.4-91-98.4-161.3-98.4s-131.2 40-161.3 98.4L0 202.7 256 0 512 202.7l-94.7 77.1z"],"square-gitlab":[448,512,["gitlab-square"],"e5ae","M0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96zm337.5 12.5l44.6 116.4 .4 1.2c5.6 16.8 7.2 35.2 2.3 52.5c-5 17.2-15.4 32.4-29.8 43.3l-.2 .1-68.4 51.2-54.1 40.9c-.5 .2-1.1 .5-1.7 .8c-2 1-4.4 2-6.7 2c-3 0-6.8-1.8-8.3-2.8l-54.2-40.9L93.5 322.3l-.4-.3-.2-.1c-14.3-10.8-24.8-26-29.7-43.3s-4.2-35.7 2.2-52.5l.5-1.2 44.7-116.4c.9-2.3 2.5-4.3 4.5-5.6c1.6-1 3.4-1.6 5.2-1.8c1.3-.7 2.1-.4 3.4 .1c.6 .2 1.2 .5 2 .7c1 .4 1.6 .9 2.4 1.5c.6 .4 1.2 1 2.1 1.5c1.2 1.4 2.2 3 2.7 4.8l29.2 92.2H285l30.2-92.2c.5-1.8 1.4-3.4 2.6-4.8s2.8-2.4 4.5-3.1c1.7-.6 3.6-.9 5.4-.7s3.6 .8 5.2 1.8c2 1.3 3.7 3.3 4.6 5.6z"],studiovinari:[512,512,[],"f3f8","M480.3 187.7l4.2 28v28l-25.1 44.1-39.8 78.4-56.1 67.5-79.1 37.8-17.7 24.5-7.7 12-9.6 4s17.3-63.6 19.4-63.6c2.1 0 20.3.7 20.3.7l66.7-38.6-92.5 26.1-55.9 36.8-22.8 28-6.6 1.4 20.8-73.6 6.9-5.5 20.7 12.9 88.3-45.2 56.8-51.5 14.8-68.4-125.4 23.3 15.2-18.2-173.4-53.3 81.9-10.5-166-122.9L133.5 108 32.2 0l252.9 126.6-31.5-38L378 163 234.7 64l18.7 38.4-49.6-18.1L158.3 0l194.6 122L310 66.2l108 96.4 12-8.9-21-16.4 4.2-37.8L451 89.1l29.2 24.7 11.5 4.2-7 6.2 8.5 12-13.1 7.4-10.3 20.2 10.5 23.9z"],"pied-piper":[480,512,[],"f2ae","M455.93,23.2C429.23,30,387.79,51.69,341.35,90.66A206,206,0,0,0,240,64C125.13,64,32,157.12,32,272s93.13,208,208,208,208-93.13,208-208a207.25,207.25,0,0,0-58.75-144.81,155.35,155.35,0,0,0-17,27.4A176.16,176.16,0,0,1,417.1,272c0,97.66-79.44,177.11-177.09,177.11a175.81,175.81,0,0,1-87.63-23.4c82.94-107.33,150.79-37.77,184.31-226.65,5.79-32.62,28-94.26,126.23-160.18C471,33.45,465.35,20.8,455.93,23.2ZM125,406.4A176.66,176.66,0,0,1,62.9,272C62.9,174.34,142.35,94.9,240,94.9a174,174,0,0,1,76.63,17.75C250.64,174.76,189.77,265.52,125,406.4Z"],wordpress:[512,512,[],"f19a","M61.7 169.4l101.5 278C92.2 413 43.3 340.2 43.3 256c0-30.9 6.6-60.1 18.4-86.6zm337.9 75.9c0-26.3-9.4-44.5-17.5-58.7-10.8-17.5-20.9-32.4-20.9-49.9 0-19.6 14.8-37.8 35.7-37.8.9 0 1.8.1 2.8.2-37.9-34.7-88.3-55.9-143.7-55.9-74.3 0-139.7 38.1-177.8 95.9 5 .2 9.7.3 13.7.3 22.2 0 56.7-2.7 56.7-2.7 11.5-.7 12.8 16.2 1.4 17.5 0 0-11.5 1.3-24.3 2l77.5 230.4L249.8 247l-33.1-90.8c-11.5-.7-22.3-2-22.3-2-11.5-.7-10.1-18.2 1.3-17.5 0 0 35.1 2.7 56 2.7 22.2 0 56.7-2.7 56.7-2.7 11.5-.7 12.8 16.2 1.4 17.5 0 0-11.5 1.3-24.3 2l76.9 228.7 21.2-70.9c9-29.4 16-50.5 16-68.7zm-139.9 29.3l-63.8 185.5c19.1 5.6 39.2 8.7 60.1 8.7 24.8 0 48.5-4.3 70.6-12.1-.6-.9-1.1-1.9-1.5-2.9l-65.4-179.2zm183-120.7c.9 6.8 1.4 14 1.4 21.9 0 21.6-4 45.8-16.2 76.2l-65 187.9C426.2 403 468.7 334.5 468.7 256c0-37-9.4-71.8-26-102.1zM504 256c0 136.8-111.3 248-248 248C119.2 504 8 392.7 8 256 8 119.2 119.2 8 256 8c136.7 0 248 111.2 248 248zm-11.4 0c0-130.5-106.2-236.6-236.6-236.6C125.5 19.4 19.4 125.5 19.4 256S125.6 492.6 256 492.6c130.5 0 236.6-106.1 236.6-236.6z"],"product-hunt":[512,512,[],"f288","M326.3 218.8c0 20.5-16.7 37.2-37.2 37.2h-70.3v-74.4h70.3c20.5 0 37.2 16.7 37.2 37.2zM504 256c0 137-111 248-248 248S8 393 8 256 119 8 256 8s248 111 248 248zm-128.1-37.2c0-47.9-38.9-86.8-86.8-86.8H169.2v248h49.6v-74.4h70.3c47.9 0 86.8-38.9 86.8-86.8z"],firefox:[512,512,[],"f269","M503.52,241.48c-.12-1.56-.24-3.12-.24-4.68v-.12l-.36-4.68v-.12a245.86,245.86,0,0,0-7.32-41.15c0-.12,0-.12-.12-.24l-1.08-4c-.12-.24-.12-.48-.24-.6-.36-1.2-.72-2.52-1.08-3.72-.12-.24-.12-.6-.24-.84-.36-1.2-.72-2.4-1.08-3.48-.12-.36-.24-.6-.36-1-.36-1.2-.72-2.28-1.2-3.48l-.36-1.08c-.36-1.08-.84-2.28-1.2-3.36a8.27,8.27,0,0,0-.36-1c-.48-1.08-.84-2.28-1.32-3.36-.12-.24-.24-.6-.36-.84-.48-1.2-1-2.28-1.44-3.48,0-.12-.12-.24-.12-.36-1.56-3.84-3.24-7.68-5-11.4l-.36-.72c-.48-1-.84-1.8-1.32-2.64-.24-.48-.48-1.08-.72-1.56-.36-.84-.84-1.56-1.2-2.4-.36-.6-.6-1.2-1-1.8s-.84-1.44-1.2-2.28c-.36-.6-.72-1.32-1.08-1.92s-.84-1.44-1.2-2.16a18.07,18.07,0,0,0-1.2-2c-.36-.72-.84-1.32-1.2-2s-.84-1.32-1.2-2-.84-1.32-1.2-1.92-.84-1.44-1.32-2.16a15.63,15.63,0,0,0-1.2-1.8L463.2,119a15.63,15.63,0,0,0-1.2-1.8c-.48-.72-1.08-1.56-1.56-2.28-.36-.48-.72-1.08-1.08-1.56l-1.8-2.52c-.36-.48-.6-.84-1-1.32-1-1.32-1.8-2.52-2.76-3.72a248.76,248.76,0,0,0-23.51-26.64A186.82,186.82,0,0,0,412,62.46c-4-3.48-8.16-6.72-12.48-9.84a162.49,162.49,0,0,0-24.6-15.12c-2.4-1.32-4.8-2.52-7.2-3.72a254,254,0,0,0-55.43-19.56c-1.92-.36-3.84-.84-5.64-1.2h-.12c-1-.12-1.8-.36-2.76-.48a236.35,236.35,0,0,0-38-4H255.14a234.62,234.62,0,0,0-45.48,5c-33.59,7.08-63.23,21.24-82.91,39-1.08,1-1.92,1.68-2.4,2.16l-.48.48H124l-.12.12.12-.12a.12.12,0,0,0,.12-.12l-.12.12a.42.42,0,0,1,.24-.12c14.64-8.76,34.92-16,49.44-19.56l5.88-1.44c.36-.12.84-.12,1.2-.24,1.68-.36,3.36-.72,5.16-1.08.24,0,.6-.12.84-.12C250.94,20.94,319.34,40.14,367,85.61a171.49,171.49,0,0,1,26.88,32.76c30.36,49.2,27.48,111.11,3.84,147.59-34.44,53-111.35,71.27-159,24.84a84.19,84.19,0,0,1-25.56-59,74.05,74.05,0,0,1,6.24-31c1.68-3.84,13.08-25.67,18.24-24.59-13.08-2.76-37.55,2.64-54.71,28.19-15.36,22.92-14.52,58.2-5,83.28a132.85,132.85,0,0,1-12.12-39.24c-12.24-82.55,43.31-153,94.31-170.51-27.48-24-96.47-22.31-147.71,15.36-29.88,22-51.23,53.16-62.51,90.36,1.68-20.88,9.6-52.08,25.8-83.88-17.16,8.88-39,37-49.8,62.88-15.6,37.43-21,82.19-16.08,124.79.36,3.24.72,6.36,1.08,9.6,19.92,117.11,122,206.38,244.78,206.38C392.77,503.42,504,392.19,504,255,503.88,250.48,503.76,245.92,503.52,241.48Z"],linode:[448,512,[],"f2b8","M366.036,186.867l-59.5,36.871-.838,36.871-29.329-19.273-39.384,24.3c2.238,55.211,2.483,59.271,2.51,59.5l-97.2,65.359L127.214,285.748l108.1-62.01L195.09,197.761l-75.417,38.547L98.723,93.015,227.771,43.574,136.432,0,10.737,39.385,38.39,174.3l41.9,32.681L48.445,222.062,69.394,323.457,98.723,351.11,77.774,363.679l16.76,78.769L160.733,512c-10.8-74.842-11.658-78.641-11.725-78.773l77.925-55.3c16.759-12.57,15.083-10.894,15.083-10.894l.838,24.3,33.519,28.491-.838-77.093,46.927-33.519,26.815-18.435-2.514,36.033,25.139,17.6,6.7-74.579,58.657-43.575Z"],goodreads:[448,512,[],"f3a8","M299.9 191.2c5.1 37.3-4.7 79-35.9 100.7-22.3 15.5-52.8 14.1-70.8 5.7-37.1-17.3-49.5-58.6-46.8-97.2 4.3-60.9 40.9-87.9 75.3-87.5 46.9-.2 71.8 31.8 78.2 78.3zM448 88v336c0 30.9-25.1 56-56 56H56c-30.9 0-56-25.1-56-56V88c0-30.9 25.1-56 56-56h336c30.9 0 56 25.1 56 56zM330 313.2s-.1-34-.1-217.3h-29v40.3c-.8.3-1.2-.5-1.6-1.2-9.6-20.7-35.9-46.3-76-46-51.9.4-87.2 31.2-100.6 77.8-4.3 14.9-5.8 30.1-5.5 45.6 1.7 77.9 45.1 117.8 112.4 115.2 28.9-1.1 54.5-17 69-45.2.5-1 1.1-1.9 1.7-2.9.2.1.4.1.6.2.3 3.8.2 30.7.1 34.5-.2 14.8-2 29.5-7.2 43.5-7.8 21-22.3 34.7-44.5 39.5-17.8 3.9-35.6 3.8-53.2-1.2-21.5-6.1-36.5-19-41.1-41.8-.3-1.6-1.3-1.3-2.3-1.3h-26.8c.8 10.6 3.2 20.3 8.5 29.2 24.2 40.5 82.7 48.5 128.2 37.4 49.9-12.3 67.3-54.9 67.4-106.3z"],"square-odnoklassniki":[448,512,["odnoklassniki-square"],"f264","M224 137.1a39.9 39.9 0 1 0 0 79.7 39.9 39.9 0 1 0 0-79.7zM384 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64zM224 95.9A81 81 0 1 1 224 258a81 81 0 1 1 0-162.1zm59.3 168.3c16.8-13.2 29.5-5.5 34.1 3.6c7.8 16-1.1 23.7-21.5 37c-17.1 10.9-40.7 15.2-56.2 16.8l13 12.9 47.7 47.7c17.4 17.9-11 45.8-28.6 28.6c-12-12.2-29.5-29.7-47.7-47.9l0 0-47.7 47.9c-17.7 17.2-46-11-28.4-28.6c3.7-3.7 7.9-7.9 12.5-12.5c10.4-10.4 22.6-22.7 35.2-35.2l12.9-12.9c-15.4-1.6-39.3-5.7-56.6-16.8c-20.3-13.3-29.3-20.9-21.4-37c4.6-9.1 17.3-16.8 34.1-3.6c0 0 22.7 18 59.3 18s59.3-18 59.3-18z"],jsfiddle:[576,512,[],"f1cc","M510.634 237.462c-4.727-2.621-5.664-5.748-6.381-10.776-2.352-16.488-3.539-33.619-9.097-49.095-35.895-99.957-153.99-143.386-246.849-91.646-27.37 15.25-48.971 36.369-65.493 63.903-3.184-1.508-5.458-2.71-7.824-3.686-30.102-12.421-59.049-10.121-85.331 9.167-25.531 18.737-36.422 44.548-32.676 76.408.355 3.025-1.967 7.621-4.514 9.545-39.712 29.992-56.031 78.065-41.902 124.615 13.831 45.569 57.514 79.796 105.608 81.433 30.291 1.031 60.637.546 90.959.539 84.041-.021 168.09.531 252.12-.48 52.664-.634 96.108-36.873 108.212-87.293 11.54-48.074-11.144-97.3-56.832-122.634zm21.107 156.88c-18.23 22.432-42.343 35.253-71.28 35.65-56.874.781-113.767.23-170.652.23 0 .7-163.028.159-163.728.154-43.861-.332-76.739-19.766-95.175-59.995-18.902-41.245-4.004-90.848 34.186-116.106 9.182-6.073 12.505-11.566 10.096-23.136-5.49-26.361 4.453-47.956 26.42-62.981 22.987-15.723 47.422-16.146 72.034-3.083 10.269 5.45 14.607 11.564 22.198-2.527 14.222-26.399 34.557-46.727 60.671-61.294 97.46-54.366 228.37 7.568 230.24 132.697.122 8.15 2.412 12.428 9.848 15.894 57.56 26.829 74.456 96.122 35.142 144.497zm-87.789-80.499c-5.848 31.157-34.622 55.096-66.666 55.095-16.953-.001-32.058-6.545-44.079-17.705-27.697-25.713-71.141-74.98-95.937-93.387-20.056-14.888-41.99-12.333-60.272 3.782-49.996 44.071 15.859 121.775 67.063 77.188 4.548-3.96 7.84-9.543 12.744-12.844 8.184-5.509 20.766-.884 13.168 10.622-17.358 26.284-49.33 38.197-78.863 29.301-28.897-8.704-48.84-35.968-48.626-70.179 1.225-22.485 12.364-43.06 35.414-55.965 22.575-12.638 46.369-13.146 66.991 2.474C295.68 280.7 320.467 323.97 352.185 343.47c24.558 15.099 54.254 7.363 68.823-17.506 28.83-49.209-34.592-105.016-78.868-63.46-3.989 3.744-6.917 8.932-11.41 11.72-10.975 6.811-17.333-4.113-12.809-10.353 20.703-28.554 50.464-40.44 83.271-28.214 31.429 11.714 49.108 44.366 42.76 78.186z"],sith:[448,512,[],"f512","M0 32l69.71 118.75-58.86-11.52 69.84 91.03a146.741 146.741 0 0 0 0 51.45l-69.84 91.03 58.86-11.52L0 480l118.75-69.71-11.52 58.86 91.03-69.84c17.02 3.04 34.47 3.04 51.48 0l91.03 69.84-11.52-58.86L448 480l-69.71-118.78 58.86 11.52-69.84-91.03c3.03-17.01 3.04-34.44 0-51.45l69.84-91.03-58.86 11.52L448 32l-118.75 69.71 11.52-58.9-91.06 69.87c-8.5-1.52-17.1-2.29-25.71-2.29s-17.21.78-25.71 2.29l-91.06-69.87 11.52 58.9L0 32zm224 99.78c31.8 0 63.6 12.12 87.85 36.37 48.5 48.5 48.49 127.21 0 175.7s-127.2 48.46-175.7-.03c-48.5-48.5-48.49-127.21 0-175.7 24.24-24.25 56.05-36.34 87.85-36.34zm0 36.66c-22.42 0-44.83 8.52-61.92 25.61-34.18 34.18-34.19 89.68 0 123.87s89.65 34.18 123.84 0c34.18-34.18 34.19-89.68 0-123.87-17.09-17.09-39.5-25.61-61.92-25.61z"],themeisle:[512,512,[],"f2b2","M208 88.286c0-10 6.286-21.714 17.715-21.714 11.142 0 17.714 11.714 17.714 21.714 0 10.285-6.572 21.714-17.714 21.714C214.286 110 208 98.571 208 88.286zm304 160c0 36.001-11.429 102.286-36.286 129.714-22.858 24.858-87.428 61.143-120.857 70.572l-1.143.286v32.571c0 16.286-12.572 30.571-29.143 30.571-10 0-19.429-5.714-24.572-14.286-5.427 8.572-14.856 14.286-24.856 14.286-10 0-19.429-5.714-24.858-14.286-5.142 8.572-14.571 14.286-24.57 14.286-10.286 0-19.429-5.714-24.858-14.286-5.143 8.572-14.571 14.286-24.571 14.286-18.857 0-29.429-15.714-29.429-32.857-16.286 12.285-35.715 19.428-56.571 19.428-22 0-43.429-8.285-60.286-22.857 10.285-.286 20.571-2.286 30.285-5.714-20.857-5.714-39.428-18.857-52-36.286 21.37 4.645 46.209 1.673 67.143-11.143-22-22-56.571-58.857-68.572-87.428C1.143 321.714 0 303.714 0 289.429c0-49.714 20.286-160 86.286-160 10.571 0 18.857 4.858 23.143 14.857a158.792 158.792 0 0 1 12-15.428c2-2.572 5.714-5.429 7.143-8.286 7.999-12.571 11.714-21.142 21.714-34C182.571 45.428 232 17.143 285.143 17.143c6 0 12 .285 17.714 1.143C313.714 6.571 328.857 0 344.572 0c14.571 0 29.714 6 40 16.286.857.858 1.428 2.286 1.428 3.428 0 3.714-10.285 13.429-12.857 16.286 4.286 1.429 15.714 6.858 15.714 12 0 2.857-2.857 5.143-4.571 7.143 31.429 27.714 49.429 67.143 56.286 108 4.286-5.143 10.285-8.572 17.143-8.572 10.571 0 20.857 7.144 28.571 14.001C507.143 187.143 512 221.714 512 248.286zM188 89.428c0 18.286 12.571 37.143 32.286 37.143 19.714 0 32.285-18.857 32.285-37.143 0-18-12.571-36.857-32.285-36.857-19.715 0-32.286 18.858-32.286 36.857zM237.714 194c0-19.714 3.714-39.143 8.571-58.286-52.039 79.534-13.531 184.571 68.858 184.571 21.428 0 42.571-7.714 60-20 2-7.429 3.714-14.857 3.714-22.572 0-14.286-6.286-21.428-20.572-21.428-4.571 0-9.143.857-13.429 1.714-63.343 12.668-107.142 3.669-107.142-63.999zm-41.142 254.858c0-11.143-8.858-20.857-20.286-20.857-11.429 0-20 9.715-20 20.857v32.571c0 11.143 8.571 21.142 20 21.142 11.428 0 20.286-9.715 20.286-21.142v-32.571zm49.143 0c0-11.143-8.572-20.857-20-20.857-11.429 0-20.286 9.715-20.286 20.857v32.571c0 11.143 8.857 21.142 20.286 21.142 11.428 0 20-10 20-21.142v-32.571zm49.713 0c0-11.143-8.857-20.857-20.285-20.857-11.429 0-20.286 9.715-20.286 20.857v32.571c0 11.143 8.857 21.142 20.286 21.142 11.428 0 20.285-9.715 20.285-21.142v-32.571zm49.715 0c0-11.143-8.857-20.857-20.286-20.857-11.428 0-20.286 9.715-20.286 20.857v32.571c0 11.143 8.858 21.142 20.286 21.142 11.429 0 20.286-10 20.286-21.142v-32.571zM421.714 286c-30.857 59.142-90.285 102.572-158.571 102.572-96.571 0-160.571-84.572-160.571-176.572 0-16.857 2-33.429 6-49.714-20 33.715-29.714 72.572-29.714 111.429 0 60.286 24.857 121.715 71.429 160.857 5.143-9.714 14.857-16.286 26-16.286 10 0 19.428 5.714 24.571 14.286 5.429-8.571 14.571-14.286 24.858-14.286 10 0 19.428 5.714 24.571 14.286 5.429-8.571 14.857-14.286 24.858-14.286 10 0 19.428 5.714 24.857 14.286 5.143-8.571 14.571-14.286 24.572-14.286 10.857 0 20.857 6.572 25.714 16 43.427-36.286 68.569-92 71.426-148.286zm10.572-99.714c0-53.714-34.571-105.714-92.572-105.714-30.285 0-58.571 15.143-78.857 36.857C240.862 183.812 233.41 254 302.286 254c28.805 0 97.357-28.538 84.286 36.857 28.857-26 45.714-65.714 45.714-104.571z"],page4:[496,512,[],"f3d7","M248 504C111 504 0 393 0 256S111 8 248 8c20.9 0 41.3 2.6 60.7 7.5L42.3 392H248v112zm0-143.6V146.8L98.6 360.4H248zm96 31.6v92.7c45.7-19.2 84.5-51.7 111.4-92.7H344zm57.4-138.2l-21.2 8.4 21.2 8.3v-16.7zm-20.3 54.5c-6.7 0-8 6.3-8 12.9v7.7h16.2v-10c0-5.9-2.3-10.6-8.2-10.6zM496 256c0 37.3-8.2 72.7-23 104.4H344V27.3C433.3 64.8 496 153.1 496 256zM360.4 143.6h68.2V96h-13.9v32.6h-13.9V99h-13.9v29.6h-12.7V96h-13.9v47.6zm68.1 185.3H402v-11c0-15.4-5.6-25.2-20.9-25.2-15.4 0-20.7 10.6-20.7 25.9v25.3h68.2v-15zm0-103l-68.2 29.7V268l68.2 29.5v-16.6l-14.4-5.7v-26.5l14.4-5.9v-16.9zm-4.8-68.5h-35.6V184H402v-12.2h11c8.6 15.8 1.3 35.3-18.6 35.3-22.5 0-28.3-25.3-15.5-37.7l-11.6-10.6c-16.2 17.5-12.2 63.9 27.1 63.9 34 0 44.7-35.9 29.3-65.3z"],hashnode:[512,512,[],"e499","M35.19 171.1C-11.72 217.1-11.72 294 35.19 340.9L171.1 476.8C217.1 523.7 294 523.7 340.9 476.8L476.8 340.9C523.7 294 523.7 217.1 476.8 171.1L340.9 35.19C294-11.72 217.1-11.72 171.1 35.19L35.19 171.1zM315.5 315.5C282.6 348.3 229.4 348.3 196.6 315.5C163.7 282.6 163.7 229.4 196.6 196.6C229.4 163.7 282.6 163.7 315.5 196.6C348.3 229.4 348.3 282.6 315.5 315.5z"],react:[512,512,[],"f41b","M418.2 177.2c-5.4-1.8-10.8-3.5-16.2-5.1.9-3.7 1.7-7.4 2.5-11.1 12.3-59.6 4.2-107.5-23.1-123.3-26.3-15.1-69.2.6-112.6 38.4-4.3 3.7-8.5 7.6-12.5 11.5-2.7-2.6-5.5-5.2-8.3-7.7-45.5-40.4-91.1-57.4-118.4-41.5-26.2 15.2-34 60.3-23 116.7 1.1 5.6 2.3 11.1 3.7 16.7-6.4 1.8-12.7 3.8-18.6 5.9C38.3 196.2 0 225.4 0 255.6c0 31.2 40.8 62.5 96.3 81.5 4.5 1.5 9 3 13.6 4.3-1.5 6-2.8 11.9-4 18-10.5 55.5-2.3 99.5 23.9 114.6 27 15.6 72.4-.4 116.6-39.1 3.5-3.1 7-6.3 10.5-9.7 4.4 4.3 9 8.4 13.6 12.4 42.8 36.8 85.1 51.7 111.2 36.6 27-15.6 35.8-62.9 24.4-120.5-.9-4.4-1.9-8.9-3-13.5 3.2-.9 6.3-1.9 9.4-2.9 57.7-19.1 99.5-50 99.5-81.7 0-30.3-39.4-59.7-93.8-78.4zM282.9 92.3c37.2-32.4 71.9-45.1 87.7-36 16.9 9.7 23.4 48.9 12.8 100.4-.7 3.4-1.4 6.7-2.3 10-22.2-5-44.7-8.6-67.3-10.6-13-18.6-27.2-36.4-42.6-53.1 3.9-3.7 7.7-7.2 11.7-10.7zM167.2 307.5c5.1 8.7 10.3 17.4 15.8 25.9-15.6-1.7-31.1-4.2-46.4-7.5 4.4-14.4 9.9-29.3 16.3-44.5 4.6 8.8 9.3 17.5 14.3 26.1zm-30.3-120.3c14.4-3.2 29.7-5.8 45.6-7.8-5.3 8.3-10.5 16.8-15.4 25.4-4.9 8.5-9.7 17.2-14.2 26-6.3-14.9-11.6-29.5-16-43.6zm27.4 68.9c6.6-13.8 13.8-27.3 21.4-40.6s15.8-26.2 24.4-38.9c15-1.1 30.3-1.7 45.9-1.7s31 .6 45.9 1.7c8.5 12.6 16.6 25.5 24.3 38.7s14.9 26.7 21.7 40.4c-6.7 13.8-13.9 27.4-21.6 40.8-7.6 13.3-15.7 26.2-24.2 39-14.9 1.1-30.4 1.6-46.1 1.6s-30.9-.5-45.6-1.4c-8.7-12.7-16.9-25.7-24.6-39s-14.8-26.8-21.5-40.6zm180.6 51.2c5.1-8.8 9.9-17.7 14.6-26.7 6.4 14.5 12 29.2 16.9 44.3-15.5 3.5-31.2 6.2-47 8 5.4-8.4 10.5-17 15.5-25.6zm14.4-76.5c-4.7-8.8-9.5-17.6-14.5-26.2-4.9-8.5-10-16.9-15.3-25.2 16.1 2 31.5 4.7 45.9 8-4.6 14.8-10 29.2-16.1 43.4zM256.2 118.3c10.5 11.4 20.4 23.4 29.6 35.8-19.8-.9-39.7-.9-59.5 0 9.8-12.9 19.9-24.9 29.9-35.8zM140.2 57c16.8-9.8 54.1 4.2 93.4 39 2.5 2.2 5 4.6 7.6 7-15.5 16.7-29.8 34.5-42.9 53.1-22.6 2-45 5.5-67.2 10.4-1.3-5.1-2.4-10.3-3.5-15.5-9.4-48.4-3.2-84.9 12.6-94zm-24.5 263.6c-4.2-1.2-8.3-2.5-12.4-3.9-21.3-6.7-45.5-17.3-63-31.2-10.1-7-16.9-17.8-18.8-29.9 0-18.3 31.6-41.7 77.2-57.6 5.7-2 11.5-3.8 17.3-5.5 6.8 21.7 15 43 24.5 63.6-9.6 20.9-17.9 42.5-24.8 64.5zm116.6 98c-16.5 15.1-35.6 27.1-56.4 35.3-11.1 5.3-23.9 5.8-35.3 1.3-15.9-9.2-22.5-44.5-13.5-92 1.1-5.6 2.3-11.2 3.7-16.7 22.4 4.8 45 8.1 67.9 9.8 13.2 18.7 27.7 36.6 43.2 53.4-3.2 3.1-6.4 6.1-9.6 8.9zm24.5-24.3c-10.2-11-20.4-23.2-30.3-36.3 9.6.4 19.5.6 29.5.6 10.3 0 20.4-.2 30.4-.7-9.2 12.7-19.1 24.8-29.6 36.4zm130.7 30c-.9 12.2-6.9 23.6-16.5 31.3-15.9 9.2-49.8-2.8-86.4-34.2-4.2-3.6-8.4-7.5-12.7-11.5 15.3-16.9 29.4-34.8 42.2-53.6 22.9-1.9 45.7-5.4 68.2-10.5 1 4.1 1.9 8.2 2.7 12.2 4.9 21.6 5.7 44.1 2.5 66.3zm18.2-107.5c-2.8.9-5.6 1.8-8.5 2.6-7-21.8-15.6-43.1-25.5-63.8 9.6-20.4 17.7-41.4 24.5-62.9 5.2 1.5 10.2 3.1 15 4.7 46.6 16 79.3 39.8 79.3 58 0 19.6-34.9 44.9-84.8 61.4zm-149.7-15c25.3 0 45.8-20.5 45.8-45.8s-20.5-45.8-45.8-45.8c-25.3 0-45.8 20.5-45.8 45.8s20.5 45.8 45.8 45.8z"],"cc-paypal":[576,512,[],"f1f4","M186.3 258.2c0 12.2-9.7 21.5-22 21.5-9.2 0-16-5.2-16-15 0-12.2 9.5-22 21.7-22 9.3 0 16.3 5.7 16.3 15.5zM80.5 209.7h-4.7c-1.5 0-3 1-3.2 2.7l-4.3 26.7 8.2-.3c11 0 19.5-1.5 21.5-14.2 2.3-13.4-6.2-14.9-17.5-14.9zm284 0H360c-1.8 0-3 1-3.2 2.7l-4.2 26.7 8-.3c13 0 22-3 22-18-.1-10.6-9.6-11.1-18.1-11.1zM576 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h480c26.5 0 48 21.5 48 48zM128.3 215.4c0-21-16.2-28-34.7-28h-40c-2.5 0-5 2-5.2 4.7L32 294.2c-.3 2 1.2 4 3.2 4h19c2.7 0 5.2-2.9 5.5-5.7l4.5-26.6c1-7.2 13.2-4.7 18-4.7 28.6 0 46.1-17 46.1-45.8zm84.2 8.8h-19c-3.8 0-4 5.5-4.2 8.2-5.8-8.5-14.2-10-23.7-10-24.5 0-43.2 21.5-43.2 45.2 0 19.5 12.2 32.2 31.7 32.2 9 0 20.2-4.9 26.5-11.9-.5 1.5-1 4.7-1 6.2 0 2.3 1 4 3.2 4H200c2.7 0 5-2.9 5.5-5.7l10.2-64.3c.3-1.9-1.2-3.9-3.2-3.9zm40.5 97.9l63.7-92.6c.5-.5.5-1 .5-1.7 0-1.7-1.5-3.5-3.2-3.5h-19.2c-1.7 0-3.5 1-4.5 2.5l-26.5 39-11-37.5c-.8-2.2-3-4-5.5-4h-18.7c-1.7 0-3.2 1.8-3.2 3.5 0 1.2 19.5 56.8 21.2 62.1-2.7 3.8-20.5 28.6-20.5 31.6 0 1.8 1.5 3.2 3.2 3.2h19.2c1.8-.1 3.5-1.1 4.5-2.6zm159.3-106.7c0-21-16.2-28-34.7-28h-39.7c-2.7 0-5.2 2-5.5 4.7l-16.2 102c-.2 2 1.3 4 3.2 4h20.5c2 0 3.5-1.5 4-3.2l4.5-29c1-7.2 13.2-4.7 18-4.7 28.4 0 45.9-17 45.9-45.8zm84.2 8.8h-19c-3.8 0-4 5.5-4.3 8.2-5.5-8.5-14-10-23.7-10-24.5 0-43.2 21.5-43.2 45.2 0 19.5 12.2 32.2 31.7 32.2 9.3 0 20.5-4.9 26.5-11.9-.3 1.5-1 4.7-1 6.2 0 2.3 1 4 3.2 4H484c2.7 0 5-2.9 5.5-5.7l10.2-64.3c.3-1.9-1.2-3.9-3.2-3.9zm47.5-33.3c0-2-1.5-3.5-3.2-3.5h-18.5c-1.5 0-3 1.2-3.2 2.7l-16.2 104-.3.5c0 1.8 1.5 3.5 3.5 3.5h16.5c2.5 0 5-2.9 5.2-5.7L544 191.2v-.3zm-90 51.8c-12.2 0-21.7 9.7-21.7 22 0 9.7 7 15 16.2 15 12 0 21.7-9.2 21.7-21.5.1-9.8-6.9-15.5-16.2-15.5z"],squarespace:[512,512,[],"f5be","M186.12 343.34c-9.65 9.65-9.65 25.29 0 34.94 9.65 9.65 25.29 9.65 34.94 0L378.24 221.1c19.29-19.29 50.57-19.29 69.86 0s19.29 50.57 0 69.86L293.95 445.1c19.27 19.29 50.53 19.31 69.82.04l.04-.04 119.25-119.24c38.59-38.59 38.59-101.14 0-139.72-38.59-38.59-101.15-38.59-139.72 0l-157.22 157.2zm244.53-104.8c-9.65-9.65-25.29-9.65-34.93 0l-157.2 157.18c-19.27 19.29-50.53 19.31-69.82.05l-.05-.05c-9.64-9.64-25.27-9.65-34.92-.01l-.01.01c-9.65 9.64-9.66 25.28-.02 34.93l.02.02c38.58 38.57 101.14 38.57 139.72 0l157.2-157.2c9.65-9.65 9.65-25.29.01-34.93zm-261.99 87.33l157.18-157.18c9.64-9.65 9.64-25.29 0-34.94-9.64-9.64-25.27-9.64-34.91 0L133.72 290.93c-19.28 19.29-50.56 19.3-69.85.01l-.01-.01c-19.29-19.28-19.31-50.54-.03-69.84l.03-.03L218.03 66.89c-19.28-19.29-50.55-19.3-69.85-.02l-.02.02L28.93 186.14c-38.58 38.59-38.58 101.14 0 139.72 38.6 38.59 101.13 38.59 139.73.01zm-87.33-52.4c9.64 9.64 25.27 9.64 34.91 0l157.21-157.19c19.28-19.29 50.55-19.3 69.84-.02l.02.02c9.65 9.65 25.29 9.65 34.93 0 9.65-9.65 9.65-25.29 0-34.93-38.59-38.59-101.13-38.59-139.72 0L81.33 238.54c-9.65 9.64-9.65 25.28-.01 34.93h.01z"],"cc-stripe":[576,512,[],"f1f5","M492.4 220.8c-8.9 0-18.7 6.7-18.7 22.7h36.7c0-16-9.3-22.7-18-22.7zM375 223.4c-8.2 0-13.3 2.9-17 7l.2 52.8c3.5 3.7 8.5 6.7 16.8 6.7 13.1 0 21.9-14.3 21.9-33.4 0-18.6-9-33.2-21.9-33.1zM528 32H48C21.5 32 0 53.5 0 80v352c0 26.5 21.5 48 48 48h480c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48zM122.2 281.1c0 25.6-20.3 40.1-49.9 40.3-12.2 0-25.6-2.4-38.8-8.1v-33.9c12 6.4 27.1 11.3 38.9 11.3 7.9 0 13.6-2.1 13.6-8.7 0-17-54-10.6-54-49.9 0-25.2 19.2-40.2 48-40.2 11.8 0 23.5 1.8 35.3 6.5v33.4c-10.8-5.8-24.5-9.1-35.3-9.1-7.5 0-12.1 2.2-12.1 7.7 0 16 54.3 8.4 54.3 50.7zm68.8-56.6h-27V275c0 20.9 22.5 14.4 27 12.6v28.9c-4.7 2.6-13.3 4.7-24.9 4.7-21.1 0-36.9-15.5-36.9-36.5l.2-113.9 34.7-7.4v30.8H191zm74 2.4c-4.5-1.5-18.7-3.6-27.1 7.4v84.4h-35.5V194.2h30.7l2.2 10.5c8.3-15.3 24.9-12.2 29.6-10.5h.1zm44.1 91.8h-35.7V194.2h35.7zm0-142.9l-35.7 7.6v-28.9l35.7-7.6zm74.1 145.5c-12.4 0-20-5.3-25.1-9l-.1 40.2-35.5 7.5V194.2h31.3l1.8 8.8c4.9-4.5 13.9-11.1 27.8-11.1 24.9 0 48.4 22.5 48.4 63.8 0 45.1-23.2 65.5-48.6 65.6zm160.4-51.5h-69.5c1.6 16.6 13.8 21.5 27.6 21.5 14.1 0 25.2-3 34.9-7.9V312c-9.7 5.3-22.4 9.2-39.4 9.2-34.6 0-58.8-21.7-58.8-64.5 0-36.2 20.5-64.9 54.3-64.9 33.7 0 51.3 28.7 51.3 65.1 0 3.5-.3 10.9-.4 12.9z"],"creative-commons-share":[496,512,[],"f4f2","M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zm101 132.4c7.8 0 13.7 6.1 13.7 13.7v182.5c0 7.7-6.1 13.7-13.7 13.7H214.3c-7.7 0-13.7-6-13.7-13.7v-54h-54c-7.8 0-13.7-6-13.7-13.7V131.1c0-8.2 6.6-12.7 12.4-13.7h136.4c7.7 0 13.7 6 13.7 13.7v54h54zM159.9 300.3h40.7V198.9c0-7.4 5.8-12.6 12-13.7h55.8v-40.3H159.9v155.4zm176.2-88.1H227.6v155.4h108.5V212.2z"],bitcoin:[512,512,[],"f379","M504 256c0 136.967-111.033 248-248 248S8 392.967 8 256 119.033 8 256 8s248 111.033 248 248zm-141.651-35.33c4.937-32.999-20.191-50.739-54.55-62.573l11.146-44.702-27.213-6.781-10.851 43.524c-7.154-1.783-14.502-3.464-21.803-5.13l10.929-43.81-27.198-6.781-11.153 44.686c-5.922-1.349-11.735-2.682-17.377-4.084l.031-.14-37.53-9.37-7.239 29.062s20.191 4.627 19.765 4.913c11.022 2.751 13.014 10.044 12.68 15.825l-12.696 50.925c.76.194 1.744.473 2.829.907-.907-.225-1.876-.473-2.876-.713l-17.796 71.338c-1.349 3.348-4.767 8.37-12.471 6.464.271.395-19.78-4.937-19.78-4.937l-13.51 31.147 35.414 8.827c6.588 1.651 13.045 3.379 19.4 5.006l-11.262 45.213 27.182 6.781 11.153-44.733a1038.209 1038.209 0 0 0 21.687 5.627l-11.115 44.523 27.213 6.781 11.262-45.128c46.404 8.781 81.299 5.239 95.986-36.727 11.836-33.79-.589-53.281-25.004-65.991 17.78-4.098 31.174-15.792 34.747-39.949zm-62.177 87.179c-8.41 33.79-65.308 15.523-83.755 10.943l14.944-59.899c18.446 4.603 77.6 13.717 68.811 48.956zm8.417-87.667c-7.673 30.736-55.031 15.12-70.393 11.292l13.548-54.327c15.363 3.828 64.836 10.973 56.845 43.035z"],keycdn:[512,512,[],"f3ba","M63.8 409.3l60.5-59c32.1 42.8 71.1 66 126.6 67.4 30.5.7 60.3-7 86.4-22.4 5.1 5.3 18.5 19.5 20.9 22-32.2 20.7-69.6 31.1-108.1 30.2-43.3-1.1-84.6-16.7-117.7-44.4.3-.6-38.2 37.5-38.6 37.9 9.5 29.8-13.1 62.4-46.3 62.4C20.7 503.3 0 481.7 0 454.9c0-34.3 33.1-56.6 63.8-45.6zm354.9-252.4c19.1 31.3 29.6 67.4 28.7 104-1.1 44.8-19 87.5-48.6 121 .3.3 23.8 25.2 24.1 25.5 9.6-1.3 19.2 2 25.9 9.1 11.3 12 10.9 30.9-1.1 42.4-12 11.3-30.9 10.9-42.4-1.1-6.7-7-9.4-16.8-7.6-26.3-24.9-26.6-44.4-47.2-44.4-47.2 42.7-34.1 63.3-79.6 64.4-124.2.7-28.9-7.2-57.2-21.1-82.2l22.1-21zM104 53.1c6.7 7 9.4 16.8 7.6 26.3l45.9 48.1c-4.7 3.8-13.3 10.4-22.8 21.3-25.4 28.5-39.6 64.8-40.7 102.9-.7 28.9 6.1 57.2 20 82.4l-22 21.5C72.7 324 63.1 287.9 64.2 250.9c1-44.6 18.3-87.6 47.5-121.1l-25.3-26.4c-9.6 1.3-19.2-2-25.9-9.1-11.3-12-10.9-30.9 1.1-42.4C73.5 40.7 92.2 41 104 53.1zM464.9 8c26 0 47.1 22.4 47.1 48.3S490.9 104 464.9 104c-6.3.1-14-1.1-15.9-1.8l-62.9 59.7c-32.7-43.6-76.7-65.9-126.9-67.2-30.5-.7-60.3 6.8-86.2 22.4l-21.1-22C184.1 74.3 221.5 64 260 64.9c43.3 1.1 84.6 16.7 117.7 44.6l41.1-38.6c-1.5-4.7-2.2-9.6-2.2-14.5C416.5 29.7 438.9 8 464.9 8zM256.7 113.4c5.5 0 10.9.4 16.4 1.1 78.1 9.8 133.4 81.1 123.8 159.1-9.8 78.1-81.1 133.4-159.1 123.8-78.1-9.8-133.4-81.1-123.8-159.2 9.3-72.4 70.1-124.6 142.7-124.8zm-59 119.4c.6 22.7 12.2 41.8 32.4 52.2l-11 51.7h73.7l-11-51.7c20.1-10.9 32.1-29 32.4-52.2-.4-32.8-25.8-57.5-58.3-58.3-32.1.8-57.3 24.8-58.2 58.3zM256 160"],opera:[496,512,[],"f26a","M313.9 32.7c-170.2 0-252.6 223.8-147.5 355.1 36.5 45.4 88.6 75.6 147.5 75.6 36.3 0 70.3-11.1 99.4-30.4-43.8 39.2-101.9 63-165.3 63-3.9 0-8 0-11.9-.3C104.6 489.6 0 381.1 0 248 0 111 111 0 248 0h.8c63.1.3 120.7 24.1 164.4 63.1-29-19.4-63.1-30.4-99.3-30.4zm101.8 397.7c-40.9 24.7-90.7 23.6-132-5.8 56.2-20.5 97.7-91.6 97.7-176.6 0-84.7-41.2-155.8-97.4-176.6 41.8-29.2 91.2-30.3 132.9-5 105.9 98.7 105.5 265.7-1.2 364z"],"itch-io":[512,512,[],"f83a","M71.92 34.77C50.2 47.67 7.4 96.84 7 109.73v21.34c0 27.06 25.29 50.84 48.25 50.84 27.57 0 50.54-22.85 50.54-50 0 27.12 22.18 50 49.76 50s49-22.85 49-50c0 27.12 23.59 50 51.16 50h.5c27.57 0 51.16-22.85 51.16-50 0 27.12 21.47 50 49 50s49.76-22.85 49.76-50c0 27.12 23 50 50.54 50 23 0 48.25-23.78 48.25-50.84v-21.34c-.4-12.9-43.2-62.07-64.92-75C372.56 32.4 325.76 32 256 32S91.14 33.1 71.92 34.77zm132.32 134.39c-22 38.4-77.9 38.71-99.85.25-13.17 23.14-43.17 32.07-56 27.66-3.87 40.15-13.67 237.13 17.73 269.15 80 18.67 302.08 18.12 379.76 0 31.65-32.27 21.32-232 17.75-269.15-12.92 4.44-42.88-4.6-56-27.66-22 38.52-77.85 38.1-99.85-.24-7.1 12.49-23.05 28.94-51.76 28.94a57.54 57.54 0 0 1-51.75-28.94zm-41.58 53.77c16.47 0 31.09 0 49.22 19.78a436.91 436.91 0 0 1 88.18 0C318.22 223 332.85 223 349.31 223c52.33 0 65.22 77.53 83.87 144.45 17.26 62.15-5.52 63.67-33.95 63.73-42.15-1.57-65.49-32.18-65.49-62.79-39.25 6.43-101.93 8.79-155.55 0 0 30.61-23.34 61.22-65.49 62.79-28.42-.06-51.2-1.58-33.94-63.73 18.67-67 31.56-144.45 83.88-144.45zM256 270.79s-44.38 40.77-52.35 55.21l29-1.17v25.32c0 1.55 21.34.16 23.33.16 11.65.54 23.31 1 23.31-.16v-25.28l29 1.17c-8-14.48-52.35-55.24-52.35-55.24z"],umbraco:[510,512,[],"f8e8","M255.35 8C118.36 7.83 7.14 118.72 7 255.68c-.07 137 111 248.2 248 248.27 136.85 0 247.82-110.7 248-247.67S392.34 8.17 255.35 8zm145 266q-1.14 40.68-14 65t-43.51 35q-30.61 10.7-85.45 10.47h-4.6q-54.78.22-85.44-10.47t-43.52-35q-12.85-24.36-14-65a224.81 224.81 0 0 1 0-30.71 418.37 418.37 0 0 1 3.6-43.88c1.88-13.39 3.57-22.58 5.4-32 1-4.88 1.28-6.42 1.82-8.45a5.09 5.09 0 0 1 4.9-3.89h.69l32 5a5.07 5.07 0 0 1 4.16 5 5 5 0 0 1 0 .77l-1.7 8.78q-2.41 13.25-4.84 33.68a380.62 380.62 0 0 0-2.64 42.15q-.28 40.43 8.13 59.83a43.87 43.87 0 0 0 31.31 25.18A243 243 0 0 0 250 340.6h10.25a242.64 242.64 0 0 0 57.27-5.16 43.86 43.86 0 0 0 31.15-25.23q8.53-19.42 8.13-59.78a388 388 0 0 0-2.6-42.15q-2.48-20.38-4.89-33.68l-1.69-8.78a5 5 0 0 1 0-.77 5 5 0 0 1 4.2-5l32-5h.82a5 5 0 0 1 4.9 3.89c.55 2.05.81 3.57 1.83 8.45 1.82 9.62 3.52 18.78 5.39 32a415.71 415.71 0 0 1 3.61 43.88 228.06 228.06 0 0 1-.04 30.73z"],"galactic-senate":[512,512,[],"f50d","M249.86 33.48v26.07C236.28 80.17 226 168.14 225.39 274.9c11.74-15.62 19.13-33.33 19.13-48.24v-16.88c-.03-5.32.75-10.53 2.19-15.65.65-2.14 1.39-4.08 2.62-5.82 1.23-1.75 3.43-3.79 6.68-3.79 3.24 0 5.45 2.05 6.68 3.79 1.23 1.75 1.97 3.68 2.62 5.82 1.44 5.12 2.22 10.33 2.19 15.65v16.88c0 14.91 7.39 32.62 19.13 48.24-.63-106.76-10.91-194.73-24.49-215.35V33.48h-12.28zm-26.34 147.77c-9.52 2.15-18.7 5.19-27.46 9.08 8.9 16.12 9.76 32.64 1.71 37.29-8 4.62-21.85-4.23-31.36-19.82-11.58 8.79-21.88 19.32-30.56 31.09 14.73 9.62 22.89 22.92 18.32 30.66-4.54 7.7-20.03 7.14-35.47-.96-5.78 13.25-9.75 27.51-11.65 42.42 9.68.18 18.67 2.38 26.18 6.04 17.78-.3 32.77-1.96 40.49-4.22 5.55-26.35 23.02-48.23 46.32-59.51.73-25.55 1.88-49.67 3.48-72.07zm64.96 0c1.59 22.4 2.75 46.52 3.47 72.07 23.29 11.28 40.77 33.16 46.32 59.51 7.72 2.26 22.71 3.92 40.49 4.22 7.51-3.66 16.5-5.85 26.18-6.04-1.9-14.91-5.86-29.17-11.65-42.42-15.44 8.1-30.93 8.66-35.47.96-4.57-7.74 3.6-21.05 18.32-30.66-8.68-11.77-18.98-22.3-30.56-31.09-9.51 15.59-23.36 24.44-31.36 19.82-8.05-4.65-7.19-21.16 1.71-37.29a147.49 147.49 0 0 0-27.45-9.08zm-32.48 8.6c-3.23 0-5.86 8.81-6.09 19.93h-.05v16.88c0 41.42-49.01 95.04-93.49 95.04-52 0-122.75-1.45-156.37 29.17v2.51c9.42 17.12 20.58 33.17 33.18 47.97C45.7 380.26 84.77 360.4 141.2 360c45.68 1.02 79.03 20.33 90.76 40.87.01.01-.01.04 0 .05 7.67 2.14 15.85 3.23 24.04 3.21 8.19.02 16.37-1.07 24.04-3.21.01-.01-.01-.04 0-.05 11.74-20.54 45.08-39.85 90.76-40.87 56.43.39 95.49 20.26 108.02 41.35 12.6-14.8 23.76-30.86 33.18-47.97v-2.51c-33.61-30.62-104.37-29.17-156.37-29.17-44.48 0-93.49-53.62-93.49-95.04v-16.88h-.05c-.23-11.12-2.86-19.93-6.09-19.93zm0 96.59c22.42 0 40.6 18.18 40.6 40.6s-18.18 40.65-40.6 40.65-40.6-18.23-40.6-40.65c0-22.42 18.18-40.6 40.6-40.6zm0 7.64c-18.19 0-32.96 14.77-32.96 32.96S237.81 360 256 360s32.96-14.77 32.96-32.96-14.77-32.96-32.96-32.96zm0 6.14c14.81 0 26.82 12.01 26.82 26.82s-12.01 26.82-26.82 26.82-26.82-12.01-26.82-26.82 12.01-26.82 26.82-26.82zm-114.8 66.67c-10.19.07-21.6.36-30.5 1.66.43 4.42 1.51 18.63 7.11 29.76 9.11-2.56 18.36-3.9 27.62-3.9 41.28.94 71.48 34.35 78.26 74.47l.11 4.7c10.4 1.91 21.19 2.94 32.21 2.94 11.03 0 21.81-1.02 32.21-2.94l.11-4.7c6.78-40.12 36.98-73.53 78.26-74.47 9.26 0 18.51 1.34 27.62 3.9 5.6-11.13 6.68-25.34 7.11-29.76-8.9-1.3-20.32-1.58-30.5-1.66-18.76.42-35.19 4.17-48.61 9.67-12.54 16.03-29.16 30.03-49.58 33.07-.09.02-.17.04-.27.05-.05.01-.11.04-.16.05-5.24 1.07-10.63 1.6-16.19 1.6-5.55 0-10.95-.53-16.19-1.6-.05-.01-.11-.04-.16-.05-.1-.02-.17-.04-.27-.05-20.42-3.03-37.03-17.04-49.58-33.07-13.42-5.49-29.86-9.25-48.61-9.67z"],ubuntu:[576,512,[],"f7df","M469.2 75A75.6 75.6 0 1 0 317.9 75a75.6 75.6 0 1 0 151.2 0zM154.2 240.7A75.6 75.6 0 1 0 3 240.7a75.6 75.6 0 1 0 151.2 0zM57 346C75.6 392.9 108 433 150 461.1s91.5 42.6 142 41.7c-14.7-18.6-22.9-41.5-23.2-65.2c-6.8-.9-13.3-2.1-19.5-3.4c-26.8-5.7-51.9-17.3-73.6-34s-39.3-38.1-51.7-62.5c-20.9 9.9-44.5 12.8-67.1 8.2zm395.1 89.8a75.6 75.6 0 1 0 -151.2 0 75.6 75.6 0 1 0 151.2 0zM444 351.6c18.5 14.8 31.6 35.2 37.2 58.2c33.3-41.3 52.6-92.2 54.8-145.2s-12.5-105.4-42.2-149.4c-8.6 21.5-24 39.6-43.8 51.6c15.4 28.6 22.9 60.8 21.9 93.2s-10.7 64-28 91.6zM101.1 135.4c12.4 2.7 24.3 7.5 35.1 14.3c16.6-24.2 38.9-44.1 64.8-58S255.8 70.4 285.2 70c.2-5.9 .9-11.9 2-17.7c3.6-16.7 11.1-32.3 21.8-45.5c-47.7-3.8-95.4 6-137.6 28.5S94.3 91.7 70.8 133.4c2.7-.2 5.3-.3 8-.3c7.5 0 15 .8 22.4 2.3z"],draft2digital:[480,512,[],"f396","M480 398.1l-144-82.2v64.7h-91.3c30.8-35 81.8-95.9 111.8-149.3 35.2-62.6 16.1-123.4-12.8-153.3-4.4-4.6-62.2-62.9-166-41.2-59.1 12.4-89.4 43.4-104.3 67.3-13.1 20.9-17 39.8-18.2 47.7-5.5 33 19.4 67.1 56.7 67.1 31.7 0 57.3-25.7 57.3-57.4 0-27.1-19.7-52.1-48-56.8 1.8-7.3 17.7-21.1 26.3-24.7 41.1-17.3 78 5.2 83.3 33.5 8.3 44.3-37.1 90.4-69.7 127.6C84.5 328.1 18.3 396.8 0 415.9l336-.1V480zM369.9 371l47.1 27.2-47.1 27.2zM134.2 161.4c0 12.4-10 22.4-22.4 22.4s-22.4-10-22.4-22.4 10-22.4 22.4-22.4 22.4 10.1 22.4 22.4zM82.5 380.5c25.6-27.4 97.7-104.7 150.8-169.9 35.1-43.1 40.3-82.4 28.4-112.7-7.4-18.8-17.5-30.2-24.3-35.7 45.3 2.1 68 23.4 82.2 38.3 0 0 42.4 48.2 5.8 113.3-37 65.9-110.9 147.5-128.5 166.7z"],stripe:[640,512,[],"f429","M165 144.7l-43.3 9.2-.2 142.4c0 26.3 19.8 43.3 46.1 43.3 14.6 0 25.3-2.7 31.2-5.9v-33.8c-5.7 2.3-33.7 10.5-33.7-15.7V221h33.7v-37.8h-33.7zm89.1 51.6l-2.7-13.1H213v153.2h44.3V233.3c10.5-13.8 28.2-11.1 33.9-9.3v-40.8c-6-2.1-26.7-6-37.1 13.1zm92.3-72.3l-44.6 9.5v36.2l44.6-9.5zM44.9 228.3c0-6.9 5.8-9.6 15.1-9.7 13.5 0 30.7 4.1 44.2 11.4v-41.8c-14.7-5.8-29.4-8.1-44.1-8.1-36 0-60 18.8-60 50.2 0 49.2 67.5 41.2 67.5 62.4 0 8.2-7.1 10.9-17 10.9-14.7 0-33.7-6.1-48.6-14.2v40c16.5 7.1 33.2 10.1 48.5 10.1 36.9 0 62.3-15.8 62.3-47.8 0-52.9-67.9-43.4-67.9-63.4zM640 261.6c0-45.5-22-81.4-64.2-81.4s-67.9 35.9-67.9 81.1c0 53.5 30.3 78.2 73.5 78.2 21.2 0 37.1-4.8 49.2-11.5v-33.4c-12.1 6.1-26 9.8-43.6 9.8-17.3 0-32.5-6.1-34.5-26.9h86.9c.2-2.3.6-11.6.6-15.9zm-87.9-16.8c0-20 12.3-28.4 23.4-28.4 10.9 0 22.5 8.4 22.5 28.4zm-112.9-64.6c-17.4 0-28.6 8.2-34.8 13.9l-2.3-11H363v204.8l44.4-9.4.1-50.2c6.4 4.7 15.9 11.2 31.4 11.2 31.8 0 60.8-23.2 60.8-79.6.1-51.6-29.3-79.7-60.5-79.7zm-10.6 122.5c-10.4 0-16.6-3.8-20.9-8.4l-.3-66c4.6-5.1 11-8.8 21.2-8.8 16.2 0 27.4 18.2 27.4 41.4.1 23.9-10.9 41.8-27.4 41.8zm-126.7 33.7h44.6V183.2h-44.6z"],houzz:[448,512,[],"f27c","M275.9 330.7H171.3V480H17V32h109.5v104.5l305.1 85.6V480H275.9z"],gg:[512,512,[],"f260","M179.2 230.4l102.4 102.4-102.4 102.4L0 256 179.2 76.8l44.8 44.8-25.6 25.6-19.2-19.2-128 128 128 128 51.5-51.5-77.1-76.5 25.6-25.6zM332.8 76.8L230.4 179.2l102.4 102.4 25.6-25.6-77.1-76.5 51.5-51.5 128 128-128 128-19.2-19.2-25.6 25.6 44.8 44.8L512 256 332.8 76.8z"],dhl:[640,512,[],"f790","M238 301.2h58.7L319 271h-58.7L238 301.2zM0 282.9v6.4h81.8l4.7-6.4H0zM172.9 271c-8.7 0-6-3.6-4.6-5.5 2.8-3.8 7.6-10.4 10.4-14.1 2.8-3.7 2.8-5.9-2.8-5.9h-51l-41.1 55.8h100.1c33.1 0 51.5-22.5 57.2-30.3h-68.2zm317.5-6.9l39.3-53.4h-62.2l-39.3 53.4h62.2zM95.3 271H0v6.4h90.6l4.7-6.4zm111-26.6c-2.8 3.8-7.5 10.4-10.3 14.2-1.4 2-4.1 5.5 4.6 5.5h45.6s7.3-10 13.5-18.4c8.4-11.4.7-35-29.2-35H112.6l-20.4 27.8h111.4c5.6 0 5.5 2.2 2.7 5.9zM0 301.2h73.1l4.7-6.4H0v6.4zm323 0h58.7L404 271h-58.7c-.1 0-22.3 30.2-22.3 30.2zm222 .1h95v-6.4h-90.3l-4.7 6.4zm22.3-30.3l-4.7 6.4H640V271h-72.7zm-13.5 18.3H640v-6.4h-81.5l-4.7 6.4zm-164.2-78.6l-22.5 30.6h-26.2l22.5-30.6h-58.7l-39.3 53.4H409l39.3-53.4h-58.7zm33.5 60.3s-4.3 5.9-6.4 8.7c-7.4 10-.9 21.6 23.2 21.6h94.3l22.3-30.3H423.1z"],"square-pinterest":[448,512,["pinterest-square"],"f0d3","M384 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64h72.6l-2.2-.8c-5.4-48.1-3.1-57.5 15.7-134.7c3.9-16 8.5-35 13.9-57.9c0 0-7.3-14.8-7.3-36.5c0-70.7 75.5-78 75.5-25c0 13.5-5.4 31.1-11.2 49.8c-3.3 10.6-6.6 21.5-9.1 32c-5.7 24.5 12.3 44.4 36.4 44.4c43.7 0 77.2-46 77.2-112.4c0-58.8-42.3-99.9-102.6-99.9C153 139 112 191.4 112 245.6c0 21.1 8.2 43.7 18.3 56c2 2.4 2.3 4.5 1.7 7c-1.1 4.7-3.1 12.9-4.7 19.2c-1 4-1.8 7.3-2.1 8.6c-1.1 4.5-3.5 5.5-8.2 3.3c-30.6-14.3-49.8-59.1-49.8-95.1C67.2 167.1 123.4 96 229.4 96c85.2 0 151.4 60.7 151.4 141.8c0 84.6-53.3 152.7-127.4 152.7c-24.9 0-48.3-12.9-56.3-28.2c0 0-12.3 46.9-15.3 58.4c-5 19.3-17.6 42.9-27.4 59.3H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64z"],xing:[384,512,[],"f168","M162.7 210c-1.8 3.3-25.2 44.4-70.1 123.5-4.9 8.3-10.8 12.5-17.7 12.5H9.8c-7.7 0-12.1-7.5-8.5-14.4l69-121.3c.2 0 .2-.1 0-.3l-43.9-75.6c-4.3-7.8.3-14.1 8.5-14.1H100c7.3 0 13.3 4.1 18 12.2l44.7 77.5zM382.6 46.1l-144 253v.3L330.2 466c3.9 7.1.2 14.1-8.5 14.1h-65.2c-7.6 0-13.6-4-18-12.2l-92.4-168.5c3.3-5.8 51.5-90.8 144.8-255.2 4.6-8.1 10.4-12.2 17.5-12.2h65.7c8 0 12.3 6.7 8.5 14.1z"],blackberry:[512,512,[],"f37b","M166 116.9c0 23.4-16.4 49.1-72.5 49.1H23.4l21-88.8h67.8c42.1 0 53.8 23.3 53.8 39.7zm126.2-39.7h-67.8L205.7 166h70.1c53.8 0 70.1-25.7 70.1-49.1.1-16.4-11.6-39.7-53.7-39.7zM88.8 208.1H21L0 296.9h70.1c56.1 0 72.5-23.4 72.5-49.1 0-16.3-11.7-39.7-53.8-39.7zm180.1 0h-67.8l-18.7 88.8h70.1c53.8 0 70.1-23.4 70.1-49.1 0-16.3-11.7-39.7-53.7-39.7zm189.3-53.8h-67.8l-18.7 88.8h70.1c53.8 0 70.1-23.4 70.1-49.1.1-16.3-11.6-39.7-53.7-39.7zm-28 137.9h-67.8L343.7 381h70.1c56.1 0 70.1-23.4 70.1-49.1 0-16.3-11.6-39.7-53.7-39.7zM240.8 346H173l-18.7 88.8h70.1c56.1 0 70.1-25.7 70.1-49.1.1-16.3-11.6-39.7-53.7-39.7z"],"creative-commons-pd":[496,512,[],"f4ec","M248 8C111 8 0 119.1 0 256c0 137 111 248 248 248s248-111 248-248C496 119.1 385 8 248 8zm0 449.5c-139.2 0-235.8-138-190.2-267.9l78.8 35.1c-2.1 10.5-3.3 21.5-3.3 32.9 0 99 73.9 126.9 120.4 126.9 22.9 0 53.5-6.7 79.4-29.5L297 311.1c-5.5 6.3-17.6 16.7-36.3 16.7-37.8 0-53.7-39.9-53.9-71.9 230.4 102.6 216.5 96.5 217.9 96.8-34.3 62.4-100.6 104.8-176.7 104.8zm194.2-150l-224-100c18.8-34 54.9-30.7 74.7-11l40.4-41.6c-27.1-23.3-58-27.5-78.1-27.5-47.4 0-80.9 20.5-100.7 51.6l-74.9-33.4c36.1-54.9 98.1-91.2 168.5-91.2 111.1 0 201.5 90.4 201.5 201.5 0 18-2.4 35.4-6.8 52-.3-.1-.4-.2-.6-.4z"],playstation:[576,512,[],"f3df","M570.9 372.3c-11.3 14.2-38.8 24.3-38.8 24.3L327 470.2v-54.3l150.9-53.8c17.1-6.1 19.8-14.8 5.8-19.4-13.9-4.6-39.1-3.3-56.2 2.9L327 381.1v-56.4c23.2-7.8 47.1-13.6 75.7-16.8 40.9-4.5 90.9.6 130.2 15.5 44.2 14 49.2 34.7 38 48.9zm-224.4-92.5v-139c0-16.3-3-31.3-18.3-35.6-11.7-3.8-19 7.1-19 23.4v347.9l-93.8-29.8V32c39.9 7.4 98 24.9 129.2 35.4C424.1 94.7 451 128.7 451 205.2c0 74.5-46 102.8-104.5 74.6zM43.2 410.2c-45.4-12.8-53-39.5-32.3-54.8 19.1-14.2 51.7-24.9 51.7-24.9l134.5-47.8v54.5l-96.8 34.6c-17.1 6.1-19.7 14.8-5.8 19.4 13.9 4.6 39.1 3.3 56.2-2.9l46.4-16.9v48.8c-51.6 9.3-101.4 7.3-153.9-10z"],quinscape:[512,512,[],"f459","M313.6 474.6h-1a158.1 158.1 0 0 1 0-316.2c94.9 0 168.2 83.1 157 176.6 4 5.1 8.2 9.6 11.2 15.3 13.4-30.3 20.3-62.4 20.3-97.7C501.1 117.5 391.6 8 256.5 8S12 117.5 12 252.6s109.5 244.6 244.5 244.6a237.36 237.36 0 0 0 70.4-10.1c-5.2-3.5-8.9-8.1-13.3-12.5zm-.1-.1l.4.1zm78.4-168.9a99.2 99.2 0 1 0 99.2 99.2 99.18 99.18 0 0 0-99.2-99.2z"],less:[640,512,[],"f41d","M612.7 219c0-20.5 3.2-32.6 3.2-54.6 0-34.2-12.6-45.2-40.5-45.2h-20.5v24.2h6.3c14.2 0 17.3 4.7 17.3 22.1 0 16.3-1.6 32.6-1.6 51.5 0 24.2 7.9 33.6 23.6 37.3v1.6c-15.8 3.7-23.6 13.1-23.6 37.3 0 18.9 1.6 34.2 1.6 51.5 0 17.9-3.7 22.6-17.3 22.6v.5h-6.3V393h20.5c27.8 0 40.5-11 40.5-45.2 0-22.6-3.2-34.2-3.2-54.6 0-11 6.8-22.6 27.3-23.6v-27.3c-20.5-.7-27.3-12.3-27.3-23.3zm-105.6 32c-15.8-6.3-30.5-10-30.5-20.5 0-7.9 6.3-12.6 17.9-12.6s22.1 4.7 33.6 13.1l21-27.8c-13.1-10-31-20.5-55.2-20.5-35.7 0-59.9 20.5-59.9 49.4 0 25.7 22.6 38.9 41.5 46.2 16.3 6.3 32.1 11.6 32.1 22.1 0 7.9-6.3 13.1-20.5 13.1-13.1 0-26.3-5.3-40.5-16.3l-21 30.5c15.8 13.1 39.9 22.1 59.9 22.1 42 0 64.6-22.1 64.6-51s-22.5-41-43-47.8zm-358.9 59.4c-3.7 0-8.4-3.2-8.4-13.1V119.1H65.2c-28.4 0-41 11-41 45.2 0 22.6 3.2 35.2 3.2 54.6 0 11-6.8 22.6-27.3 23.6v27.3c20.5.5 27.3 12.1 27.3 23.1 0 19.4-3.2 31-3.2 53.6 0 34.2 12.6 45.2 40.5 45.2h20.5v-24.2h-6.3c-13.1 0-17.3-5.3-17.3-22.6s1.6-32.1 1.6-51.5c0-24.2-7.9-33.6-23.6-37.3v-1.6c15.8-3.7 23.6-13.1 23.6-37.3 0-18.9-1.6-34.2-1.6-51.5s3.7-22.1 17.3-22.1H93v150.8c0 32.1 11 53.1 43.1 53.1 10 0 17.9-1.6 23.6-3.7l-5.3-34.2c-3.1.8-4.6.8-6.2.8zM379.9 251c-16.3-6.3-31-10-31-20.5 0-7.9 6.3-12.6 17.9-12.6 11.6 0 22.1 4.7 33.6 13.1l21-27.8c-13.1-10-31-20.5-55.2-20.5-35.7 0-59.9 20.5-59.9 49.4 0 25.7 22.6 38.9 41.5 46.2 16.3 6.3 32.1 11.6 32.1 22.1 0 7.9-6.3 13.1-20.5 13.1-13.1 0-26.3-5.3-40.5-16.3l-20.5 30.5c15.8 13.1 39.9 22.1 59.9 22.1 42 0 64.6-22.1 64.6-51 .1-28.9-22.5-41-43-47.8zm-155-68.8c-38.4 0-75.1 32.1-74.1 82.5 0 52 34.2 82.5 79.3 82.5 18.9 0 39.9-6.8 56.2-17.9l-15.8-27.8c-11.6 6.8-22.6 10-34.2 10-21 0-37.3-10-41.5-34.2H290c.5-3.7 1.6-11 1.6-19.4.6-42.6-22.6-75.7-66.7-75.7zm-30 66.2c3.2-21 15.8-31 30.5-31 18.9 0 26.3 13.1 26.3 31h-56.8z"],"blogger-b":[448,512,[],"f37d","M446.6 222.7c-1.8-8-6.8-15.4-12.5-18.5-1.8-1-13-2.2-25-2.7-20.1-.9-22.3-1.3-28.7-5-10.1-5.9-12.8-12.3-12.9-29.5-.1-33-13.8-63.7-40.9-91.3-19.3-19.7-40.9-33-65.5-40.5-5.9-1.8-19.1-2.4-63.3-2.9-69.4-.8-84.8.6-108.4 10C45.9 59.5 14.7 96.1 3.3 142.9 1.2 151.7.7 165.8.2 246.8c-.6 101.5.1 116.4 6.4 136.5 15.6 49.6 59.9 86.3 104.4 94.3 14.8 2.7 197.3 3.3 216 .8 32.5-4.4 58-17.5 81.9-41.9 17.3-17.7 28.1-36.8 35.2-62.1 4.9-17.6 4.5-142.8 2.5-151.7zm-322.1-63.6c7.8-7.9 10-8.2 58.8-8.2 43.9 0 45.4.1 51.8 3.4 9.3 4.7 13.4 11.3 13.4 21.9 0 9.5-3.8 16.2-12.3 21.6-4.6 2.9-7.3 3.1-50.3 3.3-26.5.2-47.7-.4-50.8-1.2-16.6-4.7-22.8-28.5-10.6-40.8zm191.8 199.8l-14.9 2.4-77.5.9c-68.1.8-87.3-.4-90.9-2-7.1-3.1-13.8-11.7-14.9-19.4-1.1-7.3 2.6-17.3 8.2-22.4 7.1-6.4 10.2-6.6 97.3-6.7 89.6-.1 89.1-.1 97.6 7.8 12.1 11.3 9.5 31.2-4.9 39.4z"],opencart:[640,512,[],"f23d","M423.3 440.7c0 25.3-20.3 45.6-45.6 45.6s-45.8-20.3-45.8-45.6 20.6-45.8 45.8-45.8c25.4 0 45.6 20.5 45.6 45.8zm-253.9-45.8c-25.3 0-45.6 20.6-45.6 45.8s20.3 45.6 45.6 45.6 45.8-20.3 45.8-45.6-20.5-45.8-45.8-45.8zm291.7-270C158.9 124.9 81.9 112.1 0 25.7c34.4 51.7 53.3 148.9 373.1 144.2 333.3-5 130 86.1 70.8 188.9 186.7-166.7 319.4-233.9 17.2-233.9z"],vine:[384,512,[],"f1ca","M384 254.7v52.1c-18.4 4.2-36.9 6.1-52.1 6.1-36.9 77.4-103 143.8-125.1 156.2-14 7.9-27.1 8.4-42.7-.8C137 452 34.2 367.7 0 102.7h74.5C93.2 261.8 139 343.4 189.3 404.5c27.9-27.9 54.8-65.1 75.6-106.9-49.8-25.3-80.1-80.9-80.1-145.6 0-65.6 37.7-115.1 102.2-115.1 114.9 0 106.2 127.9 81.6 181.5 0 0-46.4 9.2-63.5-20.5 3.4-11.3 8.2-30.8 8.2-48.5 0-31.3-11.3-46.6-28.4-46.6-18.2 0-30.8 17.1-30.8 50 .1 79.2 59.4 118.7 129.9 101.9z"],"signal-messenger":[512,512,[],"e663","M194.6 7.5l5.8 23.3C177.7 36.3 156 45.3 136 57.4L123.7 36.8c22-13.3 45.9-23.2 70.9-29.3zm122.9 0l-5.8 23.3C334.3 36.3 356 45.3 376 57.4l12.4-20.6c-22-13.3-46-23.2-71-29.3zM36.8 123.7c-13.3 22-23.2 45.9-29.3 70.9l23.3 5.8C36.3 177.7 45.3 156 57.4 136L36.8 123.7zM24 256c0-11.6 .9-23.3 2.6-34.8L2.9 217.6c-3.8 25.4-3.8 51.3 0 76.7l23.7-3.6C24.9 279.3 24 267.6 24 256zM388.3 475.2L376 454.6c-20 12.1-41.6 21-64.2 26.6l5.8 23.3c24.9-6.2 48.8-16 70.8-29.3zM488 256c0 11.6-.9 23.3-2.6 34.8l23.7 3.6c3.8-25.4 3.8-51.3 0-76.7l-23.7 3.6c1.7 11.5 2.6 23.1 2.6 34.8zm16.5 61.4l-23.3-5.8c-5.6 22.7-14.5 44.3-26.6 64.3l20.6 12.4c13.3-22 23.2-46 29.3-71zm-213.8 168c-23 3.5-46.5 3.5-69.5 0l-3.6 23.7c25.4 3.8 51.3 3.8 76.7 0l-3.6-23.7zm152-91.8c-13.8 18.7-30.4 35.3-49.2 49.1l14.2 19.3c20.7-15.2 39-33.4 54.2-54.1l-19.3-14.4zM393.6 69.2c18.8 13.8 35.3 30.4 49.2 49.2L462.1 104C446.9 83.4 428.6 65.1 408 49.9L393.6 69.2zM69.2 118.4c13.8-18.8 30.4-35.3 49.2-49.2L104 49.9C83.4 65.1 65.1 83.4 49.9 104l19.3 14.4zm406 5.3L454.6 136c12.1 20 21 41.6 26.6 64.2l23.3-5.8c-6.2-24.9-16-48.8-29.3-70.8zm-254-97.1c23-3.5 46.5-3.5 69.5 0l3.6-23.7C268.9-1 243.1-1 217.6 2.9l3.6 23.7zM81.6 468.4L32 480l11.6-49.6L20.2 425 8.6 474.5c-.9 4-.8 8.1 .3 12.1s3.2 7.5 6.1 10.4s6.5 5 10.4 6.1s8.1 1.2 12.1 .3L87 492l-5.4-23.6zM25.2 403.6L48.6 409l8-34.4c-11.7-19.6-20.4-40.8-25.8-63L7.5 317.4c5.2 21.2 13.2 41.7 23.6 60.8l-5.9 25.3zm112 52l-34.4 8 5.4 23.4 25.3-5.9c19.2 10.4 39.6 18.4 60.8 23.6l5.8-23.3c-22.1-5.5-43.3-14.3-62.8-26l-.2 .2zM256 48c-37.2 0-73.6 10-105.6 28.9s-58.4 46-76.3 78.6s-26.9 69.3-25.8 106.4s12 73.3 31.8 104.8L60 452l85.3-20c27.3 17.2 58.2 27.8 90.3 31s64.5-1.1 94.6-12.6s57.2-29.8 79-53.6s37.8-52.2 46.8-83.2s10.5-63.6 4.7-95.3s-19-61.6-38.4-87.4s-44.5-46.7-73.4-61S288.3 48 256 48z"],paypal:[384,512,[],"f1ed","M111.4 295.9c-3.5 19.2-17.4 108.7-21.5 134-.3 1.8-1 2.5-3 2.5H12.3c-7.6 0-13.1-6.6-12.1-13.9L58.8 46.6c1.5-9.6 10.1-16.9 20-16.9 152.3 0 165.1-3.7 204 11.4 60.1 23.3 65.6 79.5 44 140.3-21.5 62.6-72.5 89.5-140.1 90.3-43.4.7-69.5-7-75.3 24.2zM357.1 152c-1.8-1.3-2.5-1.8-3 1.3-2 11.4-5.1 22.5-8.8 33.6-39.9 113.8-150.5 103.9-204.5 103.9-6.1 0-10.1 3.3-10.9 9.4-22.6 140.4-27.1 169.7-27.1 169.7-1 7.1 3.5 12.9 10.6 12.9h63.5c8.6 0 15.7-6.3 17.4-14.9.7-5.4-1.1 6.1 14.4-91.3 4.6-22 14.3-19.7 29.3-19.7 71 0 126.4-28.8 142.9-112.3 6.5-34.8 4.6-71.4-23.8-92.6z"],gitlab:[512,512,[],"f296","M503.5 204.6L502.8 202.8L433.1 21.02C431.7 17.45 429.2 14.43 425.9 12.38C423.5 10.83 420.8 9.865 417.9 9.57C415 9.275 412.2 9.653 409.5 10.68C406.8 11.7 404.4 13.34 402.4 15.46C400.5 17.58 399.1 20.13 398.3 22.9L351.3 166.9H160.8L113.7 22.9C112.9 20.13 111.5 17.59 109.6 15.47C107.6 13.35 105.2 11.72 102.5 10.7C99.86 9.675 96.98 9.295 94.12 9.587C91.26 9.878 88.51 10.83 86.08 12.38C82.84 14.43 80.33 17.45 78.92 21.02L9.267 202.8L8.543 204.6C-1.484 230.8-2.72 259.6 5.023 286.6C12.77 313.5 29.07 337.3 51.47 354.2L51.74 354.4L52.33 354.8L158.3 434.3L210.9 474L242.9 498.2C246.6 500.1 251.2 502.5 255.9 502.5C260.6 502.5 265.2 500.1 268.9 498.2L300.9 474L353.5 434.3L460.2 354.4L460.5 354.1C482.9 337.2 499.2 313.5 506.1 286.6C514.7 259.6 513.5 230.8 503.5 204.6z"],typo3:[448,512,[],"f42b","M178.7 78.4c0-24.7 5.4-32.4 13.9-39.4-69.5 8.5-149.3 34-176.3 66.4-5.4 7.7-9.3 20.8-9.3 37.1C7 246 113.8 480 191.1 480c36.3 0 97.3-59.5 146.7-139-7 2.3-11.6 2.3-18.5 2.3-57.2 0-140.6-198.5-140.6-264.9zM301.5 32c-30.1 0-41.7 5.4-41.7 36.3 0 66.4 53.8 198.5 101.7 198.5 26.3 0 78.8-99.7 78.8-182.3 0-40.9-67-52.5-138.8-52.5z"],"reddit-alien":[512,512,[],"f281","M373 138.6c-25.2 0-46.3-17.5-51.9-41l0 0c-30.6 4.3-54.2 30.7-54.2 62.4l0 .2c47.4 1.8 90.6 15.1 124.9 36.3c12.6-9.7 28.4-15.5 45.5-15.5c41.3 0 74.7 33.4 74.7 74.7c0 29.8-17.4 55.5-42.7 67.5c-2.4 86.8-97 156.6-213.2 156.6S45.5 410.1 43 323.4C17.6 311.5 0 285.7 0 255.7c0-41.3 33.4-74.7 74.7-74.7c17.2 0 33 5.8 45.7 15.6c34-21.1 76.8-34.4 123.7-36.4l0-.3c0-44.3 33.7-80.9 76.8-85.5C325.8 50.2 347.2 32 373 32c29.4 0 53.3 23.9 53.3 53.3s-23.9 53.3-53.3 53.3zM157.5 255.3c-20.9 0-38.9 20.8-40.2 47.9s17.1 38.1 38 38.1s36.6-9.8 37.8-36.9s-14.7-49.1-35.7-49.1zM395 303.1c-1.2-27.1-19.2-47.9-40.2-47.9s-36.9 22-35.7 49.1c1.2 27.1 16.9 36.9 37.8 36.9s39.3-11 38-38.1zm-60.1 70.8c1.5-3.6-1-7.7-4.9-8.1c-23-2.3-47.9-3.6-73.8-3.6s-50.8 1.3-73.8 3.6c-3.9 .4-6.4 4.5-4.9 8.1c12.9 30.8 43.3 52.4 78.7 52.4s65.8-21.6 78.7-52.4z"],yahoo:[512,512,[],"f19e","M223.69,141.06,167,284.23,111,141.06H14.93L120.76,390.19,82.19,480h94.17L317.27,141.06Zm105.4,135.79a58.22,58.22,0,1,0,58.22,58.22A58.22,58.22,0,0,0,329.09,276.85ZM394.65,32l-93,223.47H406.44L499.07,32Z"],dailymotion:[448,512,[],"e052","M298.93,267a48.4,48.4,0,0,0-24.36-6.21q-19.83,0-33.44,13.27t-13.61,33.42q0,21.16,13.28,34.6t33.43,13.44q20.5,0,34.11-13.78T322,307.47A47.13,47.13,0,0,0,315.9,284,44.13,44.13,0,0,0,298.93,267ZM0,32V480H448V32ZM374.71,405.26h-53.1V381.37h-.67q-15.79,26.2-55.78,26.2-27.56,0-48.89-13.1a88.29,88.29,0,0,1-32.94-35.77q-11.6-22.68-11.59-50.89,0-27.56,11.76-50.22a89.9,89.9,0,0,1,32.93-35.78q21.18-13.09,47.72-13.1a80.87,80.87,0,0,1,29.74,5.21q13.28,5.21,25,17V153l55.79-12.09Z"],affiliatetheme:[512,512,[],"f36b","M159.7 237.4C108.4 308.3 43.1 348.2 14 326.6-15.2 304.9 2.8 230 54.2 159.1c51.3-70.9 116.6-110.8 145.7-89.2 29.1 21.6 11.1 96.6-40.2 167.5zm351.2-57.3C437.1 303.5 319 367.8 246.4 323.7c-25-15.2-41.3-41.2-49-73.8-33.6 64.8-92.8 113.8-164.1 133.2 49.8 59.3 124.1 96.9 207 96.9 150 0 271.6-123.1 271.6-274.9.1-8.5-.3-16.8-1-25z"],"pied-piper-pp":[448,512,[],"f1a7","M205.3 174.6c0 21.1-14.2 38.1-31.7 38.1-7.1 0-12.8-1.2-17.2-3.7v-68c4.4-2.7 10.1-4.2 17.2-4.2 17.5 0 31.7 16.9 31.7 37.8zm52.6 67c-7.1 0-12.8 1.5-17.2 4.2v68c4.4 2.5 10.1 3.7 17.2 3.7 17.4 0 31.7-16.9 31.7-37.8 0-21.1-14.3-38.1-31.7-38.1zM448 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h352c26.5 0 48 21.5 48 48zM185 255.1c41 0 74.2-35.6 74.2-79.6 0-44-33.2-79.6-74.2-79.6-12 0-24.1 3.2-34.6 8.8h-45.7V311l51.8-10.1v-50.6c8.6 3.1 18.1 4.8 28.5 4.8zm158.4 25.3c0-44-33.2-79.6-73.9-79.6-3.2 0-6.4.2-9.6.7-3.7 12.5-10.1 23.8-19.2 33.4-13.8 15-32.2 23.8-51.8 24.8V416l51.8-10.1v-50.6c8.6 3.2 18.2 4.7 28.7 4.7 40.8 0 74-35.6 74-79.6z"],bootstrap:[576,512,[],"f836","M333.5,201.4c0-22.1-15.6-34.3-43-34.3h-50.4v71.2h42.5C315.4,238.2,333.5,225,333.5,201.4z M517,188.6 c-9.5-30.9-10.9-68.8-9.8-98.1c1.1-30.5-22.7-58.5-54.7-58.5H123.7c-32.1,0-55.8,28.1-54.7,58.5c1,29.3-0.3,67.2-9.8,98.1 c-9.6,31-25.7,50.6-52.2,53.1v28.5c26.4,2.5,42.6,22.1,52.2,53.1c9.5,30.9,10.9,68.8,9.8,98.1c-1.1,30.5,22.7,58.5,54.7,58.5h328.7 c32.1,0,55.8-28.1,54.7-58.5c-1-29.3,0.3-67.2,9.8-98.1c9.6-31,25.7-50.6,52.1-53.1v-28.5C542.7,239.2,526.5,219.6,517,188.6z M300.2,375.1h-97.9V136.8h97.4c43.3,0,71.7,23.4,71.7,59.4c0,25.3-19.1,47.9-43.5,51.8v1.3c33.2,3.6,55.5,26.6,55.5,58.3 C383.4,349.7,352.1,375.1,300.2,375.1z M290.2,266.4h-50.1v78.4h52.3c34.2,0,52.3-13.7,52.3-39.5 C344.7,279.6,326.1,266.4,290.2,266.4z"],odnoklassniki:[320,512,[],"f263","M275.1 334c-27.4 17.4-65.1 24.3-90 26.9l20.9 20.6 76.3 76.3c27.9 28.6-17.5 73.3-45.7 45.7-19.1-19.4-47.1-47.4-76.3-76.6L84 503.4c-28.2 27.5-73.6-17.6-45.4-45.7 19.4-19.4 47.1-47.4 76.3-76.3l20.6-20.6c-24.6-2.6-62.9-9.1-90.6-26.9-32.6-21-46.9-33.3-34.3-59 7.4-14.6 27.7-26.9 54.6-5.7 0 0 36.3 28.9 94.9 28.9s94.9-28.9 94.9-28.9c26.9-21.1 47.1-8.9 54.6 5.7 12.4 25.7-1.9 38-34.5 59.1zM30.3 129.7C30.3 58 88.6 0 160 0s129.7 58 129.7 129.7c0 71.4-58.3 129.4-129.7 129.4s-129.7-58-129.7-129.4zm66 0c0 35.1 28.6 63.7 63.7 63.7s63.7-28.6 63.7-63.7c0-35.4-28.6-64-63.7-64s-63.7 28.6-63.7 64z"],"nfc-symbol":[576,512,[],"e531","M392.9 32.43C400.6 31.1 408.6 32.89 414.1 37.41C498.2 96.14 544 173.7 544 255.1C544 338.2 498.2 415.9 414.1 474.6C409.3 478.6 402.4 480.5 395.5 479.9C388.5 479.3 382 476.3 377.1 471.4L193.7 288.7C188.1 283.2 185 275.7 184.1 267.8C184.1 260 188.1 252.5 193.6 246.9C199.2 241.4 206.7 238.2 214.5 238.2C222.4 238.2 229.9 241.3 235.4 246.8L400.5 411.2C455.1 366.5 484.8 312 484.8 255.1C484.8 193.5 447.9 132.9 380.9 85.76C374.5 81.24 370.1 74.35 368.8 66.62C367.4 58.89 369.2 50.94 373.8 44.53C378.3 38.12 385.2 33.77 392.9 32.43V32.43zM186.9 479.6C179.2 480.9 171.3 479.1 164.8 474.6C81.67 415.9 35.84 338.2 35.84 255.1C35.84 173.7 81.67 96.14 164.8 37.41C170.5 33.4 177.4 31.53 184.4 32.12C191.3 32.71 197.8 35.72 202.7 40.63L386.1 223.3C391.7 228.8 394.8 236.3 394.8 244.2C394.9 251.1 391.8 259.5 386.2 265.1C380.7 270.6 373.2 273.8 365.3 273.8C357.5 273.8 349.1 270.7 344.4 265.2L179.3 100.7C124.7 145.9 95.03 199.9 95.03 255.1C95.03 318.5 131.9 379.1 198.1 426.2C205.4 430.8 209.7 437.6 211.1 445.4C212.4 453.1 210.6 461.1 206.1 467.5C201.6 473.9 194.7 478.2 186.9 479.6V479.6z"],mintbit:[512,512,[],"e62f","M73.2 512V438.9H365.7V365.7h73.2V219.4H512V0H292.6V73.1H146.3v73.2H73.2V438.9H0V512H73.2zm73.1-219.4h73.2v73.1H146.3V292.6zm73.2-73.1h73.1v73.1H219.4V219.4zm73.1 0V146.3h73.2v73.1H292.6zM365.7 73.1h73.2v73.2H365.7V73.1z"],ethereum:[320,512,[],"f42e","M311.9 260.8L160 353.6 8 260.8 160 0l151.9 260.8zM160 383.4L8 290.6 160 512l152-221.4-152 92.8z"],"speaker-deck":[512,512,[],"f83c","M213.86 296H100a100 100 0 0 1 0-200h132.84a40 40 0 0 1 0 80H98c-26.47 0-26.45 40 0 40h113.82a100 100 0 0 1 0 200H40a40 40 0 0 1 0-80h173.86c26.48 0 26.46-40 0-40zM298 416a120.21 120.21 0 0 0 51.11-80h64.55a19.83 19.83 0 0 0 19.66-20V196a19.83 19.83 0 0 0-19.66-20H296.42a60.77 60.77 0 0 0 0-80h136.93c43.44 0 78.65 35.82 78.65 80v160c0 44.18-35.21 80-78.65 80z"],"creative-commons-nc-eu":[496,512,[],"f4e9","M247.7 8C103.6 8 0 124.8 0 256c0 136.3 111.7 248 247.7 248C377.9 504 496 403.1 496 256 496 117 388.4 8 247.7 8zm.6 450.7c-112 0-203.6-92.5-203.6-202.7 0-23.2 3.7-45.2 10.9-66l65.7 29.1h-4.7v29.5h23.3c0 6.2-.4 3.2-.4 19.5h-22.8v29.5h27c11.4 67 67.2 101.3 124.6 101.3 26.6 0 50.6-7.9 64.8-15.8l-10-46.1c-8.7 4.6-28.2 10.8-47.3 10.8-28.2 0-58.1-10.9-67.3-50.2h90.3l128.3 56.8c-1.5 2.1-56.2 104.3-178.8 104.3zm-16.7-190.6l-.5-.4.9.4h-.4zm77.2-19.5h3.7v-29.5h-70.3l-28.6-12.6c2.5-5.5 5.4-10.5 8.8-14.3 12.9-15.8 31.1-22.4 51.1-22.4 18.3 0 35.3 5.4 46.1 10l11.6-47.3c-15-6.6-37-12.4-62.3-12.4-39 0-72.2 15.8-95.9 42.3-5.3 6.1-9.8 12.9-13.9 20.1l-81.6-36.1c64.6-96.8 157.7-93.6 170.7-93.6 113 0 203 90.2 203 203.4 0 18.7-2.1 36.3-6.3 52.9l-136.1-60.5z"],patreon:[512,512,[],"f3d9","M489.7 153.8c-.1-65.4-51-119-110.7-138.3C304.8-8.5 207-5 136.1 28.4C50.3 68.9 23.3 157.7 22.3 246.2C21.5 319 28.7 510.6 136.9 512c80.3 1 92.3-102.5 129.5-152.3c26.4-35.5 60.5-45.5 102.4-55.9c72-17.8 121.1-74.7 121-150z"],avianex:[512,512,[],"f374","M453.1 32h-312c-38.9 0-76.2 31.2-83.3 69.7L1.2 410.3C-5.9 448.8 19.9 480 58.9 480h312c38.9 0 76.2-31.2 83.3-69.7l56.7-308.5c7-38.6-18.8-69.8-57.8-69.8zm-58.2 347.3l-32 13.5-115.4-110c-14.7 10-29.2 19.5-41.7 27.1l22.1 64.2-17.9 12.7-40.6-61-52.4-48.1 15.7-15.4 58 31.1c9.3-10.5 20.8-22.6 32.8-34.9L203 228.9l-68.8-99.8 18.8-28.9 8.9-4.8L265 207.8l4.9 4.5c19.4-18.8 33.8-32.4 33.8-32.4 7.7-6.5 21.5-2.9 30.7 7.9 9 10.5 10.6 24.7 2.7 31.3-1.8 1.3-15.5 11.4-35.3 25.6l4.5 7.3 94.9 119.4-6.3 7.9z"],ello:[496,512,[],"f5f1","M248 8C111.03 8 0 119.03 0 256s111.03 248 248 248 248-111.03 248-248S384.97 8 248 8zm143.84 285.2C375.31 358.51 315.79 404.8 248 404.8s-127.31-46.29-143.84-111.6c-1.65-7.44 2.48-15.71 9.92-17.36 7.44-1.65 15.71 2.48 17.36 9.92 14.05 52.91 62 90.11 116.56 90.11s102.51-37.2 116.56-90.11c1.65-7.44 9.92-12.4 17.36-9.92 7.44 1.65 12.4 9.92 9.92 17.36z"],gofore:[400,512,[],"f3a7","M324 319.8h-13.2v34.7c-24.5 23.1-56.3 35.8-89.9 35.8-73.2 0-132.4-60.2-132.4-134.4 0-74.1 59.2-134.4 132.4-134.4 35.3 0 68.6 14 93.6 39.4l62.3-63.3C335 55.3 279.7 32 220.7 32 98 32 0 132.6 0 256c0 122.5 97 224 220.7 224 63.2 0 124.5-26.2 171-82.5-2-27.6-13.4-77.7-67.7-77.7zm-12.1-112.5H205.6v89H324c33.5 0 60.5 15.1 76 41.8v-30.6c0-65.2-40.4-100.2-88.1-100.2z"],bimobject:[448,512,[],"f378","M416 32H32C14.4 32 0 46.4 0 64v384c0 17.6 14.4 32 32 32h384c17.6 0 32-14.4 32-32V64c0-17.6-14.4-32-32-32zm-64 257.4c0 49.4-11.4 82.6-103.8 82.6h-16.9c-44.1 0-62.4-14.9-70.4-38.8h-.9V368H96V136h64v74.7h1.1c4.6-30.5 39.7-38.8 69.7-38.8h17.3c92.4 0 103.8 33.1 103.8 82.5v35zm-64-28.9v22.9c0 21.7-3.4 33.8-38.4 33.8h-45.3c-28.9 0-44.1-6.5-44.1-35.7v-19c0-29.3 15.2-35.7 44.1-35.7h45.3c35-.2 38.4 12 38.4 33.7z"],"brave-reverse":[448,512,[],"e63d","M298 0c3 0 5.8 1.3 7.8 3.6l38.1 44c.5-.1 1-.2 1.5-.3c9.2-1.6 18.6-2.2 27.7-1.2c11.6 1.4 21.5 5.4 28.9 12.9c7.7 7.8 15.4 15.8 22.6 23.6c2.5 2.7 4.9 5.2 6.9 7.4c.7 .8 1.4 1.5 1.9 2c3.4 3.7 4.2 8.1 2.7 11.9l-9.8 24.6 13.1 38.1c.7 2 .8 4.1 .2 6.2c-.1 .4-.1 .4-.5 2.1c-.6 2.3-.6 2.3-1.5 5.8c-1.6 6.3-3.5 13.3-5.4 20.9c-5.6 21.6-11.2 43.2-16.4 63.4c-12.9 49.9-21.4 82.7-23.4 90.9c-11.1 44.5-19.9 60-48.3 80.3c-24.9 17.8-76.8 53.6-86.8 60c-1 .6-2 1.3-3.4 2.3c-.5 .4-3.2 2.2-3.9 2.7c-4.9 3.3-8.3 5.5-12.1 7.3c-4.7 2.2-9.3 3.5-13.9 3.5s-9.1-1.2-13.9-3.5c-3.7-1.8-7.2-3.9-12.1-7.3c-.8-.5-3.4-2.4-3.9-2.7c-1.4-1-2.5-1.7-3.4-2.3c-10-6.4-61.9-42.1-86.8-60c-28.4-20.4-37.2-35.8-48.3-80.3c-2-8.2-10.5-41-23.3-90.5c-5.3-20.6-10.9-42.2-16.5-63.8c-2-7.6-3.8-14.6-5.4-20.9c-.9-3.5-.9-3.5-1.5-5.8c-.4-1.7-.4-1.7-.5-2.1c-.5-2-.4-4.2 .2-6.2l13.1-38.1L11.8 104c-1.5-3.8-.7-8.2 2-11.2c1.2-1.3 1.8-2 2.6-2.8c2-2.2 4.4-4.7 6.9-7.4C30.6 74.9 38.3 66.9 46 59.1c7.4-7.5 17.3-11.6 28.9-12.9c9.1-1.1 18.5-.5 27.7 1.2c.5 .1 1 .2 1.5 .3l38.1-44C144.2 1.3 147 0 150 0H298zm-4.7 21.1H154.7L115.6 66.2c-2.6 3-6.7 4.3-10.6 3.2c-.2-.1-.7-.2-1.5-.4c-1.3-.3-2.9-.6-4.5-.9c-7.4-1.3-14.9-1.8-21.7-1C70 68 64.3 70.3 60.7 74c-7.6 7.7-15.2 15.6-22.3 23.3c-1.7 1.8-3.3 3.5-4.8 5.1l8.8 22c1 2.4 1 5 .2 7.5L29.2 170.6c.4 1.4 .5 1.9 1.2 4.8c1.6 6.3 3.5 13.3 5.4 20.9c5.6 21.6 11.2 43.2 16.4 63.4c12.9 50 21.4 82.8 23.4 91C85.7 390.8 92 402 115.8 419c24.6 17.6 76.3 53.2 85.9 59.3c1.2 .8 2.5 1.6 4 2.7c.6 .4 3.2 2.2 3.9 2.7c4 2.8 6.7 4.4 9.2 5.6c2.2 1 3.9 1.5 5.1 1.5s2.9-.5 5.1-1.5c2.5-1.2 5.2-2.8 9.2-5.6c.7-.5 3.3-2.3 3.9-2.7c1.6-1.1 2.8-1.9 4-2.7c9.6-6.1 61.3-41.7 85.9-59.3c23.8-17.1 30.2-28.2 40.1-68.3c2.1-8.3 10.5-41.1 23.3-90.7c5.3-20.6 10.9-42.2 16.5-63.8c2-7.6 3.8-14.6 5.4-20.9c.7-2.9 .9-3.4 1.2-4.8l-13.3-38.8c-.8-2.4-.8-5.1 .2-7.5l8.8-22c-1.5-1.6-3.1-3.3-4.8-5.1c-7.2-7.6-14.7-15.5-22.3-23.3c-3.7-3.7-9.3-6-16.6-6.9c-6.8-.8-14.4-.3-21.7 1c-1.7 .3-3.2 .6-4.5 .9c-.8 .2-1.3 .3-1.5 .4c-3.8 1.1-7.9-.2-10.6-3.2L293.3 21.1zM224 316c2.8 0 20.9 6.5 35.4 14.1s25 13 28.3 15.2s1.3 6.2-1.7 8.4s-44.1 34.6-48.1 38.2s-9.8 9.5-13.8 9.5s-9.8-5.9-13.8-9.5s-45.1-36-48.1-38.2s-5.1-6.2-1.7-8.4s13.9-7.5 28.3-15.2s32.5-14.1 35.4-14.1zm.1-230.7c.7 0 8.8 .2 20.5 4.2c12.3 4.2 25.7 9.4 31.9 9.4s51.9-8.9 51.9-8.9s54.2 66.7 54.2 81s-6.8 18-13.7 25.4s-36.8 39.8-40.7 43.9s-11.9 10.5-7.1 21.8s11.7 25.8 3.9 40.4s-21 24.4-29.4 22.8s-28.4-12.2-35.7-17.1s-30.5-24.3-30.5-31.8s24-20.8 28.4-23.9s24.7-14.8 25.1-19.4s.3-6-5.7-17.4s-16.7-26.7-14.9-36.8s19.1-15.4 31.5-20.2s36.2-13.7 39.2-15.1s2.2-2.7-6.8-3.6s-34.6-4.3-46.1-1.1s-31.2 8.2-32.8 10.9s-3 2.7-1.4 11.8s10.1 52.8 10.9 60.6s2.4 12.9-5.8 14.8s-22.1 5.2-26.8 5.2s-18.6-3.3-26.8-5.2s-6.6-7-5.8-14.8s9.3-51.5 10.9-60.6s.2-9.2-1.4-11.8s-21.3-7.6-32.8-10.9s-37.1 .2-46.1 1.1s-9.8 2.2-6.8 3.6s26.8 10.4 39.2 15.1s29.7 10 31.5 20.2s-9 25.4-14.9 36.8s-6.1 12.8-5.7 17.4s20.6 16.4 25.1 19.4s28.4 16.4 28.4 23.9s-23.2 27-30.5 31.8s-27.2 15.4-35.7 17.1s-21.7-8.2-29.4-22.8s-.8-29.1 3.9-40.4s-3.3-17.7-7.1-21.8s-33.8-36.5-40.7-43.9s-13.7-11.2-13.7-25.4s54.2-81 54.2-81s45.8 8.9 51.9 8.9s19.5-5.2 31.9-9.4s20.6-4.2 20.6-4.2l.1 0z"],"facebook-f":[320,512,[],"f39e","M80 299.3V512H196V299.3h86.5l18-97.8H196V166.9c0-51.7 20.3-71.5 72.7-71.5c16.3 0 29.4 .4 37 1.2V7.9C291.4 4 256.4 0 236.2 0C129.3 0 80 50.5 80 159.4v42.1H14v97.8H80z"],"square-google-plus":[448,512,["google-plus-square"],"f0d4","M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM64 256c0-55.3 44.7-100 100-100c27 0 49.5 9.8 67 26.2l-27.1 26.1c-7.4-7.1-20.3-15.4-39.8-15.4c-34.1 0-61.9 28.2-61.9 63.2c0 34.9 27.8 63.2 61.9 63.2c39.6 0 54.4-28.5 56.8-43.1H164V241.8h94.4c1 5 1.6 10.1 1.6 16.6c0 57.1-38.3 97.6-96 97.6c-55.3 0-100-44.7-100-100zm291 18.2v29H325.8v-29h-29V245h29V216H355v29h29v29.2H355z"],"web-awesome":[640,512,[],"e682","M372.2 52c0 20.9-12.4 39-30.2 47.2L448 192l104.4-20.9c-5.3-7.7-8.4-17.1-8.4-27.1c0-26.5 21.5-48 48-48s48 21.5 48 48c0 26-20.6 47.1-46.4 48L481 442.3c-10.3 23-33.2 37.7-58.4 37.7l-205.2 0c-25.2 0-48-14.8-58.4-37.7L46.4 192C20.6 191.1 0 170 0 144c0-26.5 21.5-48 48-48s48 21.5 48 48c0 10.1-3.1 19.4-8.4 27.1L192 192 298.1 99.1c-17.7-8.3-30-26.3-30-47.1c0-28.7 23.3-52 52-52s52 23.3 52 52z"],mandalorian:[448,512,[],"f50f","M232.27 511.89c-1-3.26-1.69-15.83-1.39-24.58.55-15.89 1-24.72 1.4-28.76.64-6.2 2.87-20.72 3.28-21.38.6-1 .4-27.87-.24-33.13-.31-2.58-.63-11.9-.69-20.73-.13-16.47-.53-20.12-2.73-24.76-1.1-2.32-1.23-3.84-1-11.43a92.38 92.38 0 0 0-.34-12.71c-2-13-3.46-27.7-3.25-33.9s.43-7.15 2.06-9.67c3.05-4.71 6.51-14 8.62-23.27 2.26-9.86 3.88-17.18 4.59-20.74a109.54 109.54 0 0 1 4.42-15.05c2.27-6.25 2.49-15.39.37-15.39-.3 0-1.38 1.22-2.41 2.71s-4.76 4.8-8.29 7.36c-8.37 6.08-11.7 9.39-12.66 12.58s-1 7.23-.16 7.76c.34.21 1.29 2.4 2.11 4.88a28.83 28.83 0 0 1 .72 15.36c-.39 1.77-1 5.47-1.46 8.23s-1 6.46-1.25 8.22a9.85 9.85 0 0 1-1.55 4.26c-1 1-1.14.91-2.05-.53a14.87 14.87 0 0 1-1.44-4.75c-.25-1.74-1.63-7.11-3.08-11.93-3.28-10.9-3.52-16.15-1-21a14.24 14.24 0 0 0 1.67-4.61c0-2.39-2.2-5.32-7.41-9.89-7-6.18-8.63-7.92-10.23-11.3-1.71-3.6-3.06-4.06-4.54-1.54-1.78 3-2.6 9.11-3 22l-.34 12.19 2 2.25c3.21 3.7 12.07 16.45 13.78 19.83 3.41 6.74 4.34 11.69 4.41 23.56s.95 22.75 2 24.71c.36.66.51 1.35.34 1.52s.41 2.09 1.29 4.27a38.14 38.14 0 0 1 2.06 9 91 91 0 0 0 1.71 10.37c2.23 9.56 2.77 14.08 2.39 20.14-.2 3.27-.53 11.07-.73 17.32-1.31 41.76-1.85 58-2 61.21-.12 2-.39 11.51-.6 21.07-.36 16.3-1.3 27.37-2.42 28.65-.64.73-8.07-4.91-12.52-9.49-3.75-3.87-4-4.79-2.83-9.95.7-3 2.26-18.29 3.33-32.62.36-4.78.81-10.5 1-12.71.83-9.37 1.66-20.35 2.61-34.78.56-8.46 1.33-16.44 1.72-17.73s.89-9.89 1.13-19.11l.43-16.77-2.26-4.3c-1.72-3.28-4.87-6.94-13.22-15.34-6-6.07-11.84-12.3-12.91-13.85l-1.95-2.81.75-10.9c1.09-15.71 1.1-48.57 0-59.06l-.89-8.7-3.28-4.52c-5.86-8.08-5.8-7.75-6.22-33.27-.1-6.07-.38-11.5-.63-12.06-.83-1.87-3.05-2.66-8.54-3.05-8.86-.62-11-1.9-23.85-14.55-6.15-6-12.34-12-13.75-13.19-2.81-2.42-2.79-2-.56-9.63l1.35-4.65-1.69-3a32.22 32.22 0 0 0-2.59-4.07c-1.33-1.51-5.5-10.89-6-13.49a4.24 4.24 0 0 1 .87-3.9c2.23-2.86 3.4-5.68 4.45-10.73 2.33-11.19 7.74-26.09 10.6-29.22 3.18-3.47 7.7-1 9.41 5 1.34 4.79 1.37 9.79.1 18.55a101.2 101.2 0 0 0-1 11.11c0 4 .19 4.69 2.25 7.39 3.33 4.37 7.73 7.41 15.2 10.52a18.67 18.67 0 0 1 4.72 2.85c11.17 10.72 18.62 16.18 22.95 16.85 5.18.8 8 4.54 10 13.39 1.31 5.65 4 11.14 5.46 11.14a9.38 9.38 0 0 0 3.33-1.39c2-1.22 2.25-1.73 2.25-4.18a132.88 132.88 0 0 0-2-17.84c-.37-1.66-.78-4.06-.93-5.35s-.61-3.85-1-5.69c-2.55-11.16-3.65-15.46-4.1-16-1.55-2-4.08-10.2-4.93-15.92-1.64-11.11-4-14.23-12.91-17.39A43.15 43.15 0 0 1 165.24 78c-1.15-1-4-3.22-6.35-5.06s-4.41-3.53-4.6-3.76a22.7 22.7 0 0 0-2.69-2c-6.24-4.22-8.84-7-11.26-12l-2.44-5-.22-13-.22-13 6.91-6.55c3.95-3.75 8.48-7.35 10.59-8.43 3.31-1.69 4.45-1.89 11.37-2 8.53-.19 10.12 0 11.66 1.56s1.36 6.4-.29 8.5a6.66 6.66 0 0 0-1.34 2.32c0 .58-2.61 4.91-5.42 9a30.39 30.39 0 0 0-2.37 6.82c20.44 13.39 21.55 3.77 14.07 29L194 66.92c3.11-8.66 6.47-17.26 8.61-26.22.29-7.63-12-4.19-15.4-8.68-2.33-5.93 3.13-14.18 6.06-19.2 1.6-2.34 6.62-4.7 8.82-4.15.88.22 4.16-.35 7.37-1.28a45.3 45.3 0 0 1 7.55-1.68 29.57 29.57 0 0 0 6-1.29c3.65-1.11 4.5-1.17 6.35-.4a29.54 29.54 0 0 0 5.82 1.36 18.18 18.18 0 0 1 6 1.91 22.67 22.67 0 0 0 5 2.17c2.51.68 3 .57 7.05-1.67l4.35-2.4L268.32 5c10.44-.4 10.81-.47 15.26-2.68L288.16 0l2.46 1.43c1.76 1 3.14 2.73 4.85 6 2.36 4.51 2.38 4.58 1.37 7.37-.88 2.44-.89 3.3-.1 6.39a35.76 35.76 0 0 0 2.1 5.91 13.55 13.55 0 0 1 1.31 4c.31 4.33 0 5.3-2.41 6.92-2.17 1.47-7 7.91-7 9.34a14.77 14.77 0 0 1-1.07 3c-5 11.51-6.76 13.56-14.26 17-9.2 4.2-12.3 5.19-16.21 5.19-3.1 0-4 .25-4.54 1.26a18.33 18.33 0 0 1-4.09 3.71 13.62 13.62 0 0 0-4.38 4.78 5.89 5.89 0 0 1-2.49 2.91 6.88 6.88 0 0 0-2.45 1.71 67.62 67.62 0 0 1-7 5.38c-3.33 2.34-6.87 5-7.87 6A7.27 7.27 0 0 1 224 100a5.76 5.76 0 0 0-2.13 1.65c-1.31 1.39-1.49 2.11-1.14 4.6a36.45 36.45 0 0 0 1.42 5.88c1.32 3.8 1.31 7.86 0 10.57s-.89 6.65 1.35 9.59c2 2.63 2.16 4.56.71 8.84a33.45 33.45 0 0 0-1.06 8.91c0 4.88.22 6.28 1.46 8.38s1.82 2.48 3.24 2.32c2-.23 2.3-1.05 4.71-12.12 2.18-10 3.71-11.92 13.76-17.08 2.94-1.51 7.46-4 10-5.44s6.79-3.69 9.37-4.91a40.09 40.09 0 0 0 15.22-11.67c7.11-8.79 10-16.22 12.85-33.3a18.37 18.37 0 0 1 2.86-7.73 20.39 20.39 0 0 0 2.89-7.31c1-5.3 2.85-9.08 5.58-11.51 4.7-4.18 6-1.09 4.59 10.87-.46 3.86-1.1 10.33-1.44 14.38l-.61 7.36 4.45 4.09 4.45 4.09.11 8.42c.06 4.63.47 9.53.92 10.89l.82 2.47-6.43 6.28c-8.54 8.33-12.88 13.93-16.76 21.61-1.77 3.49-3.74 7.11-4.38 8-2.18 3.11-6.46 13-8.76 20.26l-2.29 7.22-7 6.49c-3.83 3.57-8 7.25-9.17 8.17-3.05 2.32-4.26 5.15-4.26 10a14.62 14.62 0 0 0 1.59 7.26 42 42 0 0 1 2.09 4.83 9.28 9.28 0 0 0 1.57 2.89c1.4 1.59 1.92 16.12.83 23.22-.68 4.48-3.63 12-4.7 12-1.79 0-4.06 9.27-5.07 20.74-.18 2-.62 5.94-1 8.7s-1 10-1.35 16.05c-.77 12.22-.19 18.77 2 23.15 3.41 6.69.52 12.69-11 22.84l-4 3.49.07 5.19a40.81 40.81 0 0 0 1.14 8.87c4.61 16 4.73 16.92 4.38 37.13-.46 26.4-.26 40.27.63 44.15a61.31 61.31 0 0 1 1.08 7c.17 2 .66 5.33 1.08 7.36.47 2.26.78 11 .79 22.74v19.06l-1.81 2.63c-2.71 3.91-15.11 13.54-15.49 12.29zm29.53-45.11c-.18-.3-.33-6.87-.33-14.59 0-14.06-.89-27.54-2.26-34.45-.4-2-.81-9.7-.9-17.06-.15-11.93-1.4-24.37-2.64-26.38-.66-1.07-3-17.66-3-21.3 0-4.23 1-6 5.28-9.13s4.86-3.14 5.48-.72c.28 1.1 1.45 5.62 2.6 10 3.93 15.12 4.14 16.27 4.05 21.74-.1 5.78-.13 6.13-1.74 17.73-1 7.07-1.17 12.39-1 28.43.17 19.4-.64 35.73-2 41.27-.71 2.78-2.8 5.48-3.43 4.43zm-71-37.58a101 101 0 0 1-1.73-10.79 100.5 100.5 0 0 0-1.73-10.79 37.53 37.53 0 0 1-1-6.49c-.31-3.19-.91-7.46-1.33-9.48-1-4.79-3.35-19.35-3.42-21.07 0-.74-.34-4.05-.7-7.36-.67-6.21-.84-27.67-.22-28.29 1-1 6.63 2.76 11.33 7.43l5.28 5.25-.45 6.47c-.25 3.56-.6 10.23-.78 14.83s-.49 9.87-.67 11.71-.61 9.36-.94 16.72c-.79 17.41-1.94 31.29-2.65 32a.62.62 0 0 1-1-.14zm-87.18-266.59c21.07 12.79 17.84 14.15 28.49 17.66 13 4.29 18.87 7.13 23.15 16.87C111.6 233.28 86.25 255 78.55 268c-31 52-6 101.59 62.75 87.21-14.18 29.23-78 28.63-98.68-4.9-24.68-39.95-22.09-118.3 61-187.66zm210.79 179c56.66 6.88 82.32-37.74 46.54-89.23 0 0-26.87-29.34-64.28-68 3-15.45 9.49-32.12 30.57-53.82 89.2 63.51 92 141.61 92.46 149.36 4.3 70.64-78.7 91.18-105.29 61.71z"],"first-order-alt":[496,512,[],"f50a","M248 8C111.03 8 0 119.03 0 256s111.03 248 248 248 248-111.03 248-248S384.97 8 248 8zm0 488.21C115.34 496.21 7.79 388.66 7.79 256S115.34 15.79 248 15.79 488.21 123.34 488.21 256 380.66 496.21 248 496.21zm0-459.92C126.66 36.29 28.29 134.66 28.29 256S126.66 475.71 248 475.71 467.71 377.34 467.71 256 369.34 36.29 248 36.29zm0 431.22c-116.81 0-211.51-94.69-211.51-211.51S131.19 44.49 248 44.49 459.51 139.19 459.51 256 364.81 467.51 248 467.51zm186.23-162.98a191.613 191.613 0 0 1-20.13 48.69l-74.13-35.88 61.48 54.82a193.515 193.515 0 0 1-37.2 37.29l-54.8-61.57 35.88 74.27a190.944 190.944 0 0 1-48.63 20.23l-27.29-78.47 4.79 82.93c-8.61 1.18-17.4 1.8-26.33 1.8s-17.72-.62-26.33-1.8l4.76-82.46-27.15 78.03a191.365 191.365 0 0 1-48.65-20.2l35.93-74.34-54.87 61.64a193.85 193.85 0 0 1-37.22-37.28l61.59-54.9-74.26 35.93a191.638 191.638 0 0 1-20.14-48.69l77.84-27.11-82.23 4.76c-1.16-8.57-1.78-17.32-1.78-26.21 0-9 .63-17.84 1.82-26.51l82.38 4.77-77.94-27.16a191.726 191.726 0 0 1 20.23-48.67l74.22 35.92-61.52-54.86a193.85 193.85 0 0 1 37.28-37.22l54.76 61.53-35.83-74.17a191.49 191.49 0 0 1 48.65-20.13l26.87 77.25-4.71-81.61c8.61-1.18 17.39-1.8 26.32-1.8s17.71.62 26.32 1.8l-4.74 82.16 27.05-77.76c17.27 4.5 33.6 11.35 48.63 20.17l-35.82 74.12 54.72-61.47a193.13 193.13 0 0 1 37.24 37.23l-61.45 54.77 74.12-35.86a191.515 191.515 0 0 1 20.2 48.65l-77.81 27.1 82.24-4.75c1.19 8.66 1.82 17.5 1.82 26.49 0 8.88-.61 17.63-1.78 26.19l-82.12-4.75 77.72 27.09z"],osi:[512,512,[],"f41a","M8 266.44C10.3 130.64 105.4 34 221.8 18.34c138.8-18.6 255.6 75.8 278 201.1 21.3 118.8-44 230-151.6 274-9.3 3.8-14.4 1.7-18-7.7q-26.7-69.45-53.4-139c-3.1-8.1-1-13.2 7-16.8 24.2-11 39.3-29.4 43.3-55.8a71.47 71.47 0 0 0-64.5-82.2c-39-3.4-71.8 23.7-77.5 59.7-5.2 33 11.1 63.7 41.9 77.7 9.6 4.4 11.5 8.6 7.8 18.4q-26.85 69.9-53.7 139.9c-2.6 6.9-8.3 9.3-15.5 6.5-52.6-20.3-101.4-61-130.8-119-24.9-49.2-25.2-87.7-26.8-108.7zm20.9-1.9c.4 6.6.6 14.3 1.3 22.1 6.3 71.9 49.6 143.5 131 183.1 3.2 1.5 4.4.8 5.6-2.3q22.35-58.65 45-117.3c1.3-3.3.6-4.8-2.4-6.7-31.6-19.9-47.3-48.5-45.6-86 1-21.6 9.3-40.5 23.8-56.3 30-32.7 77-39.8 115.5-17.6a91.64 91.64 0 0 1 45.2 90.4c-3.6 30.6-19.3 53.9-45.7 69.8-2.7 1.6-3.5 2.9-2.3 6q22.8 58.8 45.2 117.7c1.2 3.1 2.4 3.8 5.6 2.3 35.5-16.6 65.2-40.3 88.1-72 34.8-48.2 49.1-101.9 42.3-161-13.7-117.5-119.4-214.8-255.5-198-106.1 13-195.3 102.5-197.1 225.8z"],"google-wallet":[448,512,[],"f1ee","M156.8 126.8c37.6 60.6 64.2 113.1 84.3 162.5-8.3 33.8-18.8 66.5-31.3 98.3-13.2-52.3-26.5-101.3-56-148.5 6.5-36.4 2.3-73.6 3-112.3zM109.3 200H16.1c-6.5 0-10.5 7.5-6.5 12.7C51.8 267 81.3 330.5 101.3 400h103.5c-16.2-69.7-38.7-133.7-82.5-193.5-3-4-8-6.5-13-6.5zm47.8-88c68.5 108 130 234.5 138.2 368H409c-12-138-68.4-265-143.2-368H157.1zm251.8-68.5c-1.8-6.8-8.2-11.5-15.2-11.5h-88.3c-5.3 0-9 5-7.8 10.3 13.2 46.5 22.3 95.5 26.5 146 48.2 86.2 79.7 178.3 90.6 270.8 15.8-60.5 25.3-133.5 25.3-203 0-73.6-12.1-145.1-31.1-212.6z"],"d-and-d-beyond":[640,512,[],"f6ca","M313.8 241.5c13.8 0 21-10.1 24.8-17.9-1-1.1-5-4.2-7.4-6.6-2.4 4.3-8.2 10.7-13.9 10.7-10.2 0-15.4-14.7-3.2-26.6-.5-.2-4.3-1.8-8 2.4 0-3 1-5.1 2.1-6.6-3.5 1.3-9.8 5.6-11.4 7.9.2-5.8 1.6-7.5.6-9l-.2-.2s-8.5 5.6-9.3 14.7c0 0 1.1-1.6 2.1-1.9.6-.3 1.3 0 .6 1.9-.2.6-5.8 15.7 5.1 26-.6-1.6-1.9-7.6 2.4-1.9-.3.1 5.8 7.1 15.7 7.1zm52.4-21.1c0-4-4.9-4.4-5.6-4.5 2 3.9.9 7.5.2 9 2.5-.4 5.4-1.6 5.4-4.5zm10.3 5.2c0-6.4-6.2-11.4-13.5-10.7 8 1.3 5.6 13.8-5 11.4 3.7-2.6 3.2-9.9-1.3-12.5 1.4 4.2-3 8.2-7.4 4.6-2.4-1.9-8-6.6-10.6-8.6-2.4-2.1-5.5-1-6.6-1.8-1.3-1.1-.5-3.8-2.2-5-1.6-.8-3-.3-4.8-1-1.6-.6-2.7-1.9-2.6-3.5-2.5 4.4 3.4 6.3 4.5 8.5 1 1.9-.8 4.8 4 8.5 14.8 11.6 9.1 8 10.4 18.1.6 4.3 4.2 6.7 6.4 7.4-2.1-1.9-2.9-6.4 0-9.3 0 13.9 19.2 13.3 23.1 6.4-2.4 1.1-7-.2-9-1.9 7.7 1 14.2-4.1 14.6-10.6zm-39.4-18.4c2 .8 1.6.7 6.4 4.5 10.2-24.5 21.7-15.7 22-15.5 2.2-1.9 9.8-3.8 13.8-2.7-2.4-2.7-7.5-6.2-13.3-6.2-4.7 0-7.4 2.2-8 1.3-.8-1.4 3.2-3.4 3.2-3.4-5.4.2-9.6 6.7-11.2 5.9-1.1-.5 1.4-3.7 1.4-3.7-5.1 2.9-9.3 9.1-10.2 13 4.6-5.8 13.8-9.8 19.7-9-10.5.5-19.5 9.7-23.8 15.8zm242.5 51.9c-20.7 0-40 1.3-50.3 2.1l7.4 8.2v77.2l-7.4 8.2c10.4.8 30.9 2.1 51.6 2.1 42.1 0 59.1-20.7 59.1-48.9 0-29.3-23.2-48.9-60.4-48.9zm-15.1 75.6v-53.3c30.1-3.3 46.8 3.8 46.8 26.3 0 25.6-21.4 30.2-46.8 27zM301.6 181c-1-3.4-.2-6.9 1.1-9.4 1 3 2.6 6.4 7.5 9-.5-2.4-.2-5.6.5-8-1.4-5.4 2.1-9.9 6.4-9.9 6.9 0 8.5 8.8 4.7 14.4 2.1 3.2 5.5 5.6 7.7 7.8 3.2-3.7 5.5-9.5 5.5-13.8 0-8.2-5.5-15.9-16.7-16.5-20-.9-20.2 16.6-20 18.9.5 5.2 3.4 7.8 3.3 7.5zm-.4 6c-.5 1.8-7 3.7-10.2 6.9 4.8-1 7-.2 7.8 1.8.5 1.4-.2 3.4-.5 5.6 1.6-1.8 7-5.5 11-6.2-1-.3-3.4-.8-4.3-.8 2.9-3.4 9.3-4.5 12.8-3.7-2.2-.2-6.7 1.1-8.5 2.6 1.6.3 3 .6 4.3 1.1-2.1.8-4.8 3.4-5.8 6.1 7-5 13.1 5.2 7 8.2.8.2 2.7 0 3.5-.5-.3 1.1-1.9 3-3 3.4 2.9 0 7-1.9 8.2-4.6 0 0-1.8.6-2.6-.2s.3-4.3.3-4.3c-2.3 2.9-3.4-1.3-1.3-4.2-1-.3-3.5-.6-4.6-.5 3.2-1.1 10.4-1.8 11.2-.3.6 1.1-1 3.4-1 3.4 4-.5 8.3 1.1 6.7 5.1 2.9-1.4 5.5-5.9 4.8-10.4-.3 1-1.6 2.4-2.9 2.7.2-1.4-1-2.2-1.9-2.6 1.7-9.6-14.6-14.2-14.1-23.9-1 1.3-1.8 5-.8 7.1 2.7 3.2 8.7 6.7 10.1 12.2-2.6-6.4-15.1-11.4-14.6-20.2-1.6 1.6-2.6 7.8-1.3 11 2.4 1.4 4.5 3.8 4.8 6.1-2.2-5.1-11.4-6.1-13.9-12.2-.6 2.2-.3 5 1 6.7 0 0-2.2-.8-7-.6 1.7.6 5.1 3.5 4.8 5.2zm25.9 7.4c-2.7 0-3.5-2.1-4.2-4.3 3.3 1.3 4.2 4.3 4.2 4.3zm38.9 3.7l-1-.6c-1.1-1-2.9-1.4-4.7-1.4-2.9 0-5.8 1.3-7.5 3.4-.8.8-1.4 1.8-2.1 2.6v15.7c3.5 2.6 7.1-2.9 3-7.2 1.5.3 4.6 2.7 5.1 3.2 0 0 2.6-.5 5-.5 2.1 0 3.9.3 5.6 1.1V196c-1.1.5-2.2 1-2.7 1.4zM79.9 305.9c17.2-4.6 16.2-18 16.2-19.9 0-20.6-24.1-25-37-25H3l8.3 8.6v29.5H0l11.4 14.6V346L3 354.6c61.7 0 73.8 1.5 86.4-5.9 6.7-4 9.9-9.8 9.9-17.6 0-5.1 2.6-18.8-19.4-25.2zm-41.3-27.5c20 0 29.6-.8 29.6 9.1v3c0 12.1-19 8.8-29.6 8.8zm0 59.2V315c12.2 0 32.7-2.3 32.7 8.8v4.5h.2c0 11.2-12.5 9.3-32.9 9.3zm101.2-19.3l23.1.2v-.2l14.1-21.2h-37.2v-14.9h52.4l-14.1-21v-.2l-73.5.2 7.4 8.2v77.1l-7.4 8.2h81.2l14.1-21.2-60.1.2zm214.7-60.1c-73.9 0-77.5 99.3-.3 99.3 77.9 0 74.1-99.3.3-99.3zm-.3 77.5c-37.4 0-36.9-55.3.2-55.3 36.8.1 38.8 55.3-.2 55.3zm-91.3-8.3l44.1-66.2h-41.7l6.1 7.2-20.5 37.2h-.3l-21-37.2 6.4-7.2h-44.9l44.1 65.8.2 19.4-7.7 8.2h42.6l-7.2-8.2zm-28.4-151.3c1.6 1.3 2.9 2.4 2.9 6.6v38.8c0 4.2-.8 5.3-2.7 6.4-.1.1-7.5 4.5-7.9 4.6h35.1c10 0 17.4-1.5 26-8.6-.6-5 .2-9.5.8-12 0-.2-1.8 1.4-2.7 3.5 0-5.7 1.6-15.4 9.6-20.5-.1 0-3.7-.8-9 1.1 2-3.1 10-7.9 10.4-7.9-8.2-26-38-22.9-32.2-22.9-30.9 0-32.6.3-39.9-4 .1.8.5 8.2 9.6 14.9zm21.5 5.5c4.6 0 23.1-3.3 23.1 17.3 0 20.7-18.4 17.3-23.1 17.3zm228.9 79.6l7 8.3V312h-.3c-5.4-14.4-42.3-41.5-45.2-50.9h-31.6l7.4 8.5v76.9l-7.2 8.3h39l-7.4-8.2v-47.4h.3c3.7 10.6 44.5 42.9 48.5 55.6h21.3v-85.2l7.4-8.3zm-106.7-96.1c-32.2 0-32.8.2-39.9-4 .1.7.5 8.3 9.6 14.9 3.1 2 2.9 4.3 2.9 9.5 1.8-1.1 3.8-2.2 6.1-3-1.1 1.1-2.7 2.7-3.5 4.5 1-1.1 7.5-5.1 14.6-3.5-1.6.3-4 1.1-6.1 2.9.1 0 2.1-1.1 7.5-.3v-4.3c4.7 0 23.1-3.4 23.1 17.3 0 20.5-18.5 17.3-19.7 17.3 5.7 4.4 5.8 12 2.2 16.3h.3c33.4 0 36.7-27.3 36.7-34 0-3.8-1.1-32-33.8-33.6z"],periscope:[448,512,[],"f3da","M370 63.6C331.4 22.6 280.5 0 226.6 0 111.9 0 18.5 96.2 18.5 214.4c0 75.1 57.8 159.8 82.7 192.7C137.8 455.5 192.6 512 226.6 512c41.6 0 112.9-94.2 120.9-105 24.6-33.1 82-118.3 82-192.6 0-56.5-21.1-110.1-59.5-150.8zM226.6 493.9c-42.5 0-190-167.3-190-279.4 0-107.4 83.9-196.3 190-196.3 100.8 0 184.7 89 184.7 196.3.1 112.1-147.4 279.4-184.7 279.4zM338 206.8c0 59.1-51.1 109.7-110.8 109.7-100.6 0-150.7-108.2-92.9-181.8v.4c0 24.5 20.1 44.4 44.8 44.4 24.7 0 44.8-19.9 44.8-44.4 0-18.2-11.1-33.8-26.9-40.7 76.6-19.2 141 39.3 141 112.4z"],fulcrum:[320,512,[],"f50b","M95.75 164.14l-35.38 43.55L25 164.14l35.38-43.55zM144.23 0l-20.54 198.18L72.72 256l51 57.82L144.23 512V300.89L103.15 256l41.08-44.89zm79.67 164.14l35.38 43.55 35.38-43.55-35.38-43.55zm-48.48 47L216.5 256l-41.08 44.89V512L196 313.82 247 256l-51-57.82L175.42 0z"],cloudscale:[448,512,[],"f383","M318.1 154l-9.4 7.6c-22.5-19.3-51.5-33.6-83.3-33.6C153.8 128 96 188.8 96 260.3c0 6.6.4 13.1 1.4 19.4-2-56 41.8-97.4 92.6-97.4 24.2 0 46.2 9.4 62.6 24.7l-25.2 20.4c-8.3-.9-16.8 1.8-23.1 8.1-11.1 11-11.1 28.9 0 40 11.1 11 28.9 11 40 0 6.3-6.3 9-14.9 8.1-23.1l75.2-88.8c6.3-6.5-3.3-15.9-9.5-9.6zm-83.8 111.5c-5.6 5.5-14.6 5.5-20.2 0-5.6-5.6-5.6-14.6 0-20.2s14.6-5.6 20.2 0 5.6 14.7 0 20.2zM224 32C100.5 32 0 132.5 0 256s100.5 224 224 224 224-100.5 224-224S347.5 32 224 32zm0 384c-88.2 0-160-71.8-160-160S135.8 96 224 96s160 71.8 160 160-71.8 160-160 160z"],forumbee:[448,512,[],"f211","M5.8 309.7C2 292.7 0 275.5 0 258.3 0 135 99.8 35 223.1 35c16.6 0 33.3 2 49.3 5.5C149 87.5 51.9 186 5.8 309.7zm392.9-189.2C385 103 369 87.8 350.9 75.2c-149.6 44.3-266.3 162.1-309.7 312 12.5 18.1 28 35.6 45.2 49 43.1-151.3 161.2-271.7 312.3-315.7zm15.8 252.7c15.2-25.1 25.4-53.7 29.5-82.8-79.4 42.9-145 110.6-187.6 190.3 30-4.4 58.9-15.3 84.6-31.3 35 13.1 70.9 24.3 107 33.6-9.3-36.5-20.4-74.5-33.5-109.8zm29.7-145.5c-2.6-19.5-7.9-38.7-15.8-56.8C290.5 216.7 182 327.5 137.1 466c18.1 7.6 37 12.5 56.6 15.2C240 367.1 330.5 274.4 444.2 227.7z"],mizuni:[496,512,[],"f3cc","M248 8C111 8 0 119.1 0 256c0 137 111 248 248 248s248-111 248-248C496 119.1 385 8 248 8zm-80 351.9c-31.4 10.6-58.8 27.3-80 48.2V136c0-22.1 17.9-40 40-40s40 17.9 40 40v223.9zm120-9.9c-12.9-2-26.2-3.1-39.8-3.1-13.8 0-27.2 1.1-40.2 3.1V136c0-22.1 17.9-40 40-40s40 17.9 40 40v214zm120 57.7c-21.2-20.8-48.6-37.4-80-48V136c0-22.1 17.9-40 40-40s40 17.9 40 40v271.7z"],schlix:[448,512,[],"f3ea","M350.5 157.7l-54.2-46.1 73.4-39 78.3 44.2-97.5 40.9zM192 122.1l45.7-28.2 34.7 34.6-55.4 29-25-35.4zm-65.1 6.6l31.9-22.1L176 135l-36.7 22.5-12.4-28.8zm-23.3 88.2l-8.8-34.8 29.6-18.3 13.1 35.3-33.9 17.8zm-21.2-83.7l23.9-18.1 8.9 24-26.7 18.3-6.1-24.2zM59 206.5l-3.6-28.4 22.3-15.5 6.1 28.7L59 206.5zm-30.6 16.6l20.8-12.8 3.3 33.4-22.9 12-1.2-32.6zM1.4 268l19.2-10.2.4 38.2-21 8.8L1.4 268zm59.1 59.3l-28.3 8.3-1.6-46.8 25.1-10.7 4.8 49.2zM99 263.2l-31.1 13-5.2-40.8L90.1 221l8.9 42.2zM123.2 377l-41.6 5.9-8.1-63.5 35.2-10.8 14.5 68.4zm28.5-139.9l21.2 57.1-46.2 13.6-13.7-54.1 38.7-16.6zm85.7 230.5l-70.9-3.3-24.3-95.8 55.2-8.6 40 107.7zm-84.9-279.7l42.2-22.4 28 45.9-50.8 21.3-19.4-44.8zm41 94.9l61.3-18.7 52.8 86.6-79.8 11.3-34.3-79.2zm51.4-85.6l67.3-28.8 65.5 65.4-88.6 26.2-44.2-62.8z"],"square-xing":[448,512,["xing-square"],"f169","M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM93.8 320.2c-5.5 0-8.7-5.3-6-10.3l49.3-86.7c.1 0 .1-.1 0-.2l-31.4-54c-3-5.6 .2-10.1 6-10.1h46.6c5.2 0 9.5 2.9 12.9 8.7l31.9 55.3c-1.3 2.3-18 31.7-50.1 88.2c-3.5 6.2-7.7 9.1-12.6 9.1H93.8zm163.5-33.4v.2l65.5 119c2.8 5.1 .1 10.1-6 10.1H270.2c-5.5 0-9.7-2.9-12.9-8.7l-66-120.3c1.8-3.2 22.9-40.4 63.3-111.6c11.7-20.7 25.1-44.3 40.1-70.8c3.3-5.8 7.4-8.7 12.5-8.7h46.9c5.7-.1 8.8 4.7 6 10L257.3 286.8z"],bandcamp:[512,512,[],"f2d5","M256,8C119,8,8,119,8,256S119,504,256,504,504,393,504,256,393,8,256,8Zm48.2,326.1h-181L207.9,178h181Z"],wpforms:[448,512,[],"f298","M448 75.2v361.7c0 24.3-19 43.2-43.2 43.2H43.2C19.3 480 0 461.4 0 436.8V75.2C0 51.1 18.8 32 43.2 32h361.7c24 0 43.1 18.8 43.1 43.2zm-37.3 361.6V75.2c0-3-2.6-5.8-5.8-5.8h-9.3L285.3 144 224 94.1 162.8 144 52.5 69.3h-9.3c-3.2 0-5.8 2.8-5.8 5.8v361.7c0 3 2.6 5.8 5.8 5.8h361.7c3.2.1 5.8-2.7 5.8-5.8zM150.2 186v37H76.7v-37h73.5zm0 74.4v37.3H76.7v-37.3h73.5zm11.1-147.3l54-43.7H96.8l64.5 43.7zm210 72.9v37h-196v-37h196zm0 74.4v37.3h-196v-37.3h196zm-84.6-147.3l64.5-43.7H232.8l53.9 43.7zM371.3 335v37.3h-99.4V335h99.4z"],cloudversify:[616,512,[],"f385","M148.6 304c8.2 68.5 67.4 115.5 146 111.3 51.2 43.3 136.8 45.8 186.4-5.6 69.2 1.1 118.5-44.6 131.5-99.5 14.8-62.5-18.2-132.5-92.1-155.1-33-88.1-131.4-101.5-186.5-85-57.3 17.3-84.3 53.2-99.3 109.7-7.8 2.7-26.5 8.9-45 24.1 11.7 0 15.2 8.9 15.2 19.5v20.4c0 10.7-8.7 19.5-19.5 19.5h-20.2c-10.7 0-19.5-6-19.5-16.7V240H98.8C95 240 88 244.3 88 251.9v40.4c0 6.4 5.3 11.8 11.7 11.8h48.9zm227.4 8c-10.7 46.3 21.7 72.4 55.3 86.8C324.1 432.6 259.7 348 296 288c-33.2 21.6-33.7 71.2-29.2 92.9-17.9-12.4-53.8-32.4-57.4-79.8-3-39.9 21.5-75.7 57-93.9C297 191.4 369.9 198.7 400 248c-14.1-48-53.8-70.1-101.8-74.8 30.9-30.7 64.4-50.3 114.2-43.7 69.8 9.3 133.2 82.8 67.7 150.5 35-16.3 48.7-54.4 47.5-76.9l10.5 19.6c11.8 22 15.2 47.6 9.4 72-9.2 39-40.6 68.8-79.7 76.5-32.1 6.3-83.1-5.1-91.8-59.2zM128 208H88.2c-8.9 0-16.2-7.3-16.2-16.2v-39.6c0-8.9 7.3-16.2 16.2-16.2H128c8.9 0 16.2 7.3 16.2 16.2v39.6c0 8.9-7.3 16.2-16.2 16.2zM10.1 168C4.5 168 0 163.5 0 157.9v-27.8c0-5.6 4.5-10.1 10.1-10.1h27.7c5.5 0 10.1 4.5 10.1 10.1v27.8c0 5.6-4.5 10.1-10.1 10.1H10.1zM168 142.7v-21.4c0-5.1 4.2-9.3 9.3-9.3h21.4c5.1 0 9.3 4.2 9.3 9.3v21.4c0 5.1-4.2 9.3-9.3 9.3h-21.4c-5.1 0-9.3-4.2-9.3-9.3zM56 235.5v25c0 6.3-5.1 11.5-11.4 11.5H19.4C13.1 272 8 266.8 8 260.5v-25c0-6.3 5.1-11.5 11.4-11.5h25.1c6.4 0 11.5 5.2 11.5 11.5z"],usps:[576,512,[],"f7e1","M460.3 241.7c25.8-41.3 15.2-48.8-11.7-48.8h-27c-.1 0-1.5-1.4-10.9 8-11.2 5.6-37.9 6.3-37.9 8.7 0 4.5 70.3-3.1 88.1 0 9.5 1.5-1.5 20.4-4.4 32-.5 4.5 2.4 2.3 3.8.1zm-112.1 22.6c64-21.3 97.3-23.9 102-26.2 4.4-2.9-4.4-6.6-26.2-5.8-51.7 2.2-137.6 37.1-172.6 53.9l-30.7-93.3h196.6c-2.7-28.2-152.9-22.6-337.9-22.6L27 415.8c196.4-97.3 258.9-130.3 321.2-151.5zM94.7 96c253.3 53.7 330 65.7 332.1 85.2 36.4 0 45.9 0 52.4 6.6 21.1 19.7-14.6 67.7-14.6 67.7-4.4 2.9-406.4 160.2-406.4 160.2h423.1L549 96z"],megaport:[496,512,[],"f5a3","M214.5 209.6v66.2l33.5 33.5 33.3-33.3v-66.4l-33.4-33.4zM248 8C111 8 0 119 0 256s111 248 248 248 248-111 248-248S385 8 248 8zm145.1 414.4L367 441.6l-26-19.2v-65.5l-33.4-33.4-33.4 33.4v65.5L248 441.6l-26.1-19.2v-65.5l-33.4-33.4-33.5 33.4v65.5l-26.1 19.2-26.1-19.2v-87l59.5-59.5V188l59.5-59.5V52.9l26.1-19.2L274 52.9v75.6l59.5 59.5v87.6l59.7 59.7v87.1z"],magento:[448,512,[],"f3c4","M445.7 127.9V384l-63.4 36.5V164.7L223.8 73.1 65.2 164.7l.4 255.9L2.3 384V128.1L224.2 0l221.5 127.9zM255.6 420.5L224 438.9l-31.8-18.2v-256l-63.3 36.6.1 255.9 94.9 54.9 95.1-54.9v-256l-63.4-36.6v255.9z"],spotify:[496,512,[],"f1bc","M248 8C111.1 8 0 119.1 0 256s111.1 248 248 248 248-111.1 248-248S384.9 8 248 8zm100.7 364.9c-4.2 0-6.8-1.3-10.7-3.6-62.4-37.6-135-39.2-206.7-24.5-3.9 1-9 2.6-11.9 2.6-9.7 0-15.8-7.7-15.8-15.8 0-10.3 6.1-15.2 13.6-16.8 81.9-18.1 165.6-16.5 237 26.2 6.1 3.9 9.7 7.4 9.7 16.5s-7.1 15.4-15.2 15.4zm26.9-65.6c-5.2 0-8.7-2.3-12.3-4.2-62.5-37-155.7-51.9-238.6-29.4-4.8 1.3-7.4 2.6-11.9 2.6-10.7 0-19.4-8.7-19.4-19.4s5.2-17.8 15.5-20.7c27.8-7.8 56.2-13.6 97.8-13.6 64.9 0 127.6 16.1 177 45.5 8.1 4.8 11.3 11 11.3 19.7-.1 10.8-8.5 19.5-19.4 19.5zm31-76.2c-5.2 0-8.4-1.3-12.9-3.9-71.2-42.5-198.5-52.7-280.9-29.7-3.6 1-8.1 2.6-12.9 2.6-13.2 0-23.3-10.3-23.3-23.6 0-13.6 8.4-21.3 17.4-23.9 35.2-10.3 74.6-15.2 117.5-15.2 73 0 149.5 15.2 205.4 47.8 7.8 4.5 12.9 10.7 12.9 22.6 0 13.6-11 23.3-23.2 23.3z"],"optin-monster":[576,512,[],"f23c","M572.6 421.4c5.6-9.5 4.7-15.2-5.4-11.6-3-4.9-7-9.5-11.1-13.8 2.9-9.7-.7-14.2-10.8-9.2-4.6-3.2-10.3-6.5-15.9-9.2 0-15.1-11.6-11.6-17.6-5.7-10.4-1.5-18.7-.3-26.8 5.7.3-6.5.3-13 .3-19.7 12.6 0 40.2-11 45.9-36.2 1.4-6.8 1.6-13.8-.3-21.9-3-13.5-14.3-21.3-25.1-25.7-.8-5.9-7.6-14.3-14.9-15.9s-12.4 4.9-14.1 10.3c-8.5 0-19.2 2.8-21.1 8.4-5.4-.5-11.1-1.4-16.8-1.9 2.7-1.9 5.4-3.5 8.4-4.6 5.4-9.2 14.6-11.4 25.7-11.6V256c19.5-.5 43-5.9 53.8-18.1 12.7-13.8 14.6-37.3 12.4-55.1-2.4-17.3-9.7-37.6-24.6-48.1-8.4-5.9-21.6-.8-22.7 9.5-2.2 19.6 1.2 30-38.6 25.1-10.3-23.8-24.6-44.6-42.7-60C341 49.6 242.9 55.5 166.4 71.7c19.7 4.6 41.1 8.6 59.7 16.5-26.2 2.4-52.7 11.3-76.2 23.2-32.8 17-44 29.9-56.7 42.4 14.9-2.2 28.9-5.1 43.8-3.8-9.7 5.4-18.4 12.2-26.5 20-25.8.9-23.8-5.3-26.2-25.9-1.1-10.5-14.3-15.4-22.7-9.7-28.1 19.9-33.5 79.9-12.2 103.5 10.8 12.2 35.1 17.3 54.9 17.8-.3 1.1-.3 1.9-.3 2.7 10.8.5 19.5 2.7 24.6 11.6 3 1.1 5.7 2.7 8.1 4.6-5.4.5-11.1 1.4-16.5 1.9-3.3-6.6-13.7-8.1-21.1-8.1-1.6-5.7-6.5-12.2-14.1-10.3-6.8 1.9-14.1 10-14.9 15.9-22.5 9.5-30.1 26.8-25.1 47.6 5.3 24.8 33 36.2 45.9 36.2v19.7c-6.6-5-14.3-7.5-26.8-5.7-5.5-5.5-17.3-10.1-17.3 5.7-5.9 2.7-11.4 5.9-15.9 9.2-9.8-4.9-13.6-1.7-11.1 9.2-4.1 4.3-7.8 8.6-11.1 13.8-10.2-3.7-11 2.2-5.4 11.6-1.1 3.5-1.6 7-1.9 10.8-.5 31.6 44.6 64 73.5 65.1 17.3.5 34.6-8.4 43-23.5 113.2 4.9 226.7 4.1 340.2 0 8.1 15.1 25.4 24.3 42.7 23.5 29.2-1.1 74.3-33.5 73.5-65.1.2-3.7-.7-7.2-1.7-10.7zm-73.8-254c1.1-3 2.4-8.4 2.4-14.6 0-5.9 6.8-8.1 14.1-.8 11.1 11.6 14.9 40.5 13.8 51.1-4.1-13.6-13-29-30.3-35.7zm-4.6 6.7c19.5 6.2 28.6 27.6 29.7 48.9-1.1 2.7-3 5.4-4.9 7.6-5.7 5.9-15.4 10-26.2 12.2 4.3-21.3.3-47.3-12.7-63 4.9-.8 10.9-2.4 14.1-5.7zm-24.1 6.8c13.8 11.9 20 39.2 14.1 63.5-4.1.5-8.1.8-11.6.8-1.9-21.9-6.8-44-14.3-64.6 3.7.3 8.1.3 11.8.3zM47.5 203c-1.1-10.5 2.4-39.5 13.8-51.1 7-7.3 14.1-5.1 14.1.8 0 6.2 1.4 11.6 2.4 14.6-17.3 6.8-26.2 22.2-30.3 35.7zm9.7 27.6c-1.9-2.2-3.5-4.9-4.9-7.6 1.4-21.3 10.3-42.7 29.7-48.9 3.2 3.2 9.2 4.9 14.1 5.7-13 15.7-17 41.6-12.7 63-10.8-2.2-20.5-6-26.2-12.2zm47.9 14.6c-4.1 0-8.1-.3-12.7-.8-4.6-18.6-1.9-38.9 5.4-53v.3l12.2-5.1c4.9-1.9 9.7-3.8 14.9-4.9-10.7 19.7-17.4 41.3-19.8 63.5zm184-162.7c41.9 0 76.2 34 76.2 75.9 0 42.2-34.3 76.2-76.2 76.2s-76.2-34-76.2-76.2c0-41.8 34.3-75.9 76.2-75.9zm115.6 174.3c-.3 17.8-7 48.9-23 57-13.2 6.6-6.5-7.5-16.5-58.1 13.3.3 26.6.3 39.5 1.1zm-54-1.6c.8 4.9 3.8 40.3-1.6 41.9-11.6 3.5-40 4.3-51.1-1.1-4.1-3-4.6-35.9-4.3-41.1v.3c18.9-.3 38.1-.3 57 0zM278.3 309c-13 3.5-41.6 4.1-54.6-1.6-6.5-2.7-3.8-42.4-1.9-51.6 19.2-.5 38.4-.5 57.8-.8v.3c1.1 8.3 3.3 51.2-1.3 53.7zm-106.5-51.1c12.2-.8 24.6-1.4 36.8-1.6-2.4 15.4-3 43.5-4.9 52.2-1.1 6.8-4.3 6.8-9.7 4.3-21.9-9.8-27.6-35.2-22.2-54.9zm-35.4 31.3c7.8-1.1 15.7-1.9 23.5-2.7 1.6 6.2 3.8 11.9 7 17.6 10 17 44 35.7 45.1 7 6.2 14.9 40.8 12.2 54.9 10.8 15.7-1.4 23.8-1.4 26.8-14.3 12.4 4.3 30.8 4.1 44 3 11.3-.8 20.8-.5 24.6-8.9 1.1 5.1 1.9 11.6 4.6 16.8 10.8 21.3 37.3 1.4 46.8-31.6 8.6.8 17.6 1.9 26.5 2.7-.4 1.3-3.8 7.3 7.3 11.6-47.6 47-95.7 87.8-163.2 107-63.2-20.8-112.1-59.5-155.9-106.5 9.6-3.4 10.4-8.8 8-12.5zm-21.6 172.5c-3.8 17.8-21.9 29.7-39.7 28.9-19.2-.8-46.5-17-59.2-36.5-2.7-31.1 43.8-61.3 66.2-54.6 14.9 4.3 27.8 30.8 33.5 54 0 3-.3 5.7-.8 8.2zm-8.7-66c-.5-13.5-.5-27-.3-40.5h.3c2.7-1.6 5.7-3.8 7.8-6.5 6.5-1.6 13-5.1 15.1-9.2 3.3-7.1-7-7.5-5.4-12.4 2.7-1.1 5.7-2.2 7.8-3.5 29.2 29.2 58.6 56.5 97.3 77-36.8 11.3-72.4 27.6-105.9 47-1.2-18.6-7.7-35.9-16.7-51.9zm337.6 64.6c-103 3.5-206.2 4.1-309.4 0 0 .3 0 .3-.3.3v-.3h.3c35.1-21.6 72.2-39.2 112.4-50.8 11.6 5.1 23 9.5 34.9 13.2 2.2.8 2.2.8 4.3 0 14.3-4.1 28.4-9.2 42.2-15.4 41.5 11.7 78.8 31.7 115.6 53zm10.5-12.4c-35.9-19.5-73-35.9-111.9-47.6 38.1-20 71.9-47.3 103.5-76.7 2.2 1.4 4.6 2.4 7.6 3.2 0 .8.3 1.9.5 2.4-4.6 2.7-7.8 6.2-5.9 10.3 2.2 3.8 8.6 7.6 15.1 8.9 2.4 2.7 5.1 5.1 8.1 6.8 0 13.8-.3 27.6-.8 41.3l.3-.3c-9.3 15.9-15.5 37-16.5 51.7zm105.9 6.2c-12.7 19.5-40 35.7-59.2 36.5-19.3.9-40.5-13.2-40.5-37 5.7-23.2 18.9-49.7 33.5-54 22.7-6.9 69.2 23.4 66.2 54.5zM372.9 75.2c-3.8-72.1-100.8-79.7-126-23.5 44.6-24.3 90.3-15.7 126 23.5zM74.8 407.1c-15.7 1.6-49.5 25.4-49.5 43.2 0 11.6 15.7 19.5 32.2 14.9 12.2-3.2 31.1-17.6 35.9-27.3 6-11.6-3.7-32.7-18.6-30.8zm215.9-176.2c28.6 0 51.9-21.6 51.9-48.4 0-36.1-40.5-58.1-72.2-44.3 9.5 3 16.5 11.6 16.5 21.6 0 23.3-33.3 32-46.5 11.3-7.3 34.1 19.4 59.8 50.3 59.8zM68 474.1c.5 6.5 12.2 12.7 21.6 9.5 6.8-2.7 14.6-10.5 17.3-16.2 3-7-1.1-20-9.7-18.4-8.9 1.6-29.7 16.7-29.2 25.1zm433.2-67c-14.9-1.9-24.6 19.2-18.9 30.8 4.9 9.7 24.1 24.1 36.2 27.3 16.5 4.6 32.2-3.2 32.2-14.9 0-17.8-33.8-41.6-49.5-43.2zM478.8 449c-8.4-1.6-12.4 11.3-9.5 18.4 2.4 5.7 10.3 13.5 17.3 16.2 9.2 3.2 21.1-3 21.3-9.5.9-8.4-20.2-23.5-29.1-25.1z"],fly:[384,512,[],"f417","M197.8 427.8c12.9 11.7 33.7 33.3 33.2 50.7 0 .8-.1 1.6-.1 2.5-1.8 19.8-18.8 31.1-39.1 31-25-.1-39.9-16.8-38.7-35.8 1-16.2 20.5-36.7 32.4-47.6 2.3-2.1 2.7-2.7 5.6-3.6 3.4 0 3.9.3 6.7 2.8zM331.9 67.3c-16.3-25.7-38.6-40.6-63.3-52.1C243.1 4.5 214-.2 192 0c-44.1 0-71.2 13.2-81.1 17.3C57.3 45.2 26.5 87.2 28 158.6c7.1 82.2 97 176 155.8 233.8 1.7 1.6 4.5 4.5 6.2 5.1l3.3.1c2.1-.7 1.8-.5 3.5-2.1 52.3-49.2 140.7-145.8 155.9-215.7 7-39.2 3.1-72.5-20.8-112.5zM186.8 351.9c-28-51.1-65.2-130.7-69.3-189-3.4-47.5 11.4-131.2 69.3-136.7v325.7zM328.7 180c-16.4 56.8-77.3 128-118.9 170.3C237.6 298.4 275 217 277 158.4c1.6-45.9-9.8-105.8-48-131.4 88.8 18.3 115.5 98.1 99.7 153z"],aviato:[640,512,[],"f421","M107.2 283.5l-19-41.8H36.1l-19 41.8H0l62.2-131.4 62.2 131.4h-17.2zm-45-98.1l-19.6 42.5h39.2l-19.6-42.5zm112.7 102.4l-62.2-131.4h17.1l45.1 96 45.1-96h17l-62.1 131.4zm80.6-4.3V156.4H271v127.1h-15.5zm209.1-115.6v115.6h-17.3V167.9h-41.2v-11.5h99.6v11.5h-41.1zM640 218.8c0 9.2-1.7 17.8-5.1 25.8-3.4 8-8.2 15.1-14.2 21.1-6 6-13.1 10.8-21.1 14.2-8 3.4-16.6 5.1-25.8 5.1s-17.8-1.7-25.8-5.1c-8-3.4-15.1-8.2-21.1-14.2-6-6-10.8-13-14.2-21.1-3.4-8-5.1-16.6-5.1-25.8s1.7-17.8 5.1-25.8c3.4-8 8.2-15.1 14.2-21.1 6-6 13-8.4 21.1-11.9 8-3.4 16.6-5.1 25.8-5.1s17.8 1.7 25.8 5.1c8 3.4 15.1 5.8 21.1 11.9 6 6 10.7 13.1 14.2 21.1 3.4 8 5.1 16.6 5.1 25.8zm-15.5 0c0-7.3-1.3-14-3.9-20.3-2.6-6.3-6.2-11.7-10.8-16.3-4.6-4.6-10-8.2-16.2-10.9-6.2-2.7-12.8-4-19.8-4s-13.6 1.3-19.8 4c-6.2 2.7-11.6 6.3-16.2 10.9-4.6 4.6-8.2 10-10.8 16.3-2.6 6.3-3.9 13.1-3.9 20.3 0 7.3 1.3 14 3.9 20.3 2.6 6.3 6.2 11.7 10.8 16.3 4.6 4.6 10 8.2 16.2 10.9 6.2 2.7 12.8 4 19.8 4s13.6-1.3 19.8-4c6.2-2.7 11.6-6.3 16.2-10.9 4.6-4.6 8.2-10 10.8-16.3 2.6-6.3 3.9-13.1 3.9-20.3zm-94.8 96.7v-6.3l88.9-10-242.9 13.4c.6-2.2 1.1-4.6 1.4-7.2.3-2 .5-4.2.6-6.5l64.8-8.1-64.9 1.9c0-.4-.1-.7-.1-1.1-2.8-17.2-25.5-23.7-25.5-23.7l-1.1-26.3h23.8l19 41.8h17.1L348.6 152l-62.2 131.4h17.1l19-41.8h23.6L345 268s-22.7 6.5-25.5 23.7c-.1.3-.1.7-.1 1.1l-64.9-1.9 64.8 8.1c.1 2.3.3 4.4.6 6.5.3 2.6.8 5 1.4 7.2L78.4 299.2l88.9 10v6.3c-5.9.9-10.5 6-10.5 12.2 0 6.8 5.6 12.4 12.4 12.4 6.8 0 12.4-5.6 12.4-12.4 0-6.2-4.6-11.3-10.5-12.2v-5.8l80.3 9v5.4c-5.7 1.1-9.9 6.2-9.9 12.1 0 6.8 5.6 10.2 12.4 10.2 6.8 0 12.4-3.4 12.4-10.2 0-6-4.3-11-9.9-12.1v-4.9l28.4 3.2v23.7h-5.9V360h5.9v-6.6h5v6.6h5.9v-13.8h-5.9V323l38.3 4.3c8.1 11.4 19 13.6 19 13.6l-.1 6.7-5.1.2-.1 12.1h4.1l.1-5h5.2l.1 5h4.1l-.1-12.1-5.1-.2-.1-6.7s10.9-2.1 19-13.6l38.3-4.3v23.2h-5.9V360h5.9v-6.6h5v6.6h5.9v-13.8h-5.9v-23.7l28.4-3.2v4.9c-5.7 1.1-9.9 6.2-9.9 12.1 0 6.8 5.6 10.2 12.4 10.2 6.8 0 12.4-3.4 12.4-10.2 0-6-4.3-11-9.9-12.1v-5.4l80.3-9v5.8c-5.9.9-10.5 6-10.5 12.2 0 6.8 5.6 12.4 12.4 12.4 6.8 0 12.4-5.6 12.4-12.4-.2-6.3-4.7-11.4-10.7-12.3zm-200.8-87.6l19.6-42.5 19.6 42.5h-17.9l-1.7-40.3-1.7 40.3h-17.9z"],itunes:[448,512,[],"f3b4","M223.6 80.3C129 80.3 52.5 157 52.5 251.5S129 422.8 223.6 422.8s171.2-76.7 171.2-171.2c0-94.6-76.7-171.3-171.2-171.3zm79.4 240c-3.2 13.6-13.5 21.2-27.3 23.8-12.1 2.2-22.2 2.8-31.9-5-11.8-10-12-26.4-1.4-36.8 8.4-8 20.3-9.6 38-12.8 3-.5 5.6-1.2 7.7-3.7 3.2-3.6 2.2-2 2.2-80.8 0-5.6-2.7-7.1-8.4-6.1-4 .7-91.9 17.1-91.9 17.1-5 1.1-6.7 2.6-6.7 8.3 0 116.1.5 110.8-1.2 118.5-2.1 9-7.6 15.8-14.9 19.6-8.3 4.6-23.4 6.6-31.4 5.2-21.4-4-28.9-28.7-14.4-42.9 8.4-8 20.3-9.6 38-12.8 3-.5 5.6-1.2 7.7-3.7 5-5.7.9-127 2.6-133.7.4-2.6 1.5-4.8 3.5-6.4 2.1-1.7 5.8-2.7 6.7-2.7 101-19 113.3-21.4 115.1-21.4 5.7-.4 9 3 9 8.7-.1 170.6.4 161.4-1 167.6zM345.2 32H102.8C45.9 32 0 77.9 0 134.8v242.4C0 434.1 45.9 480 102.8 480h242.4c57 0 102.8-45.9 102.8-102.8V134.8C448 77.9 402.1 32 345.2 32zM223.6 444c-106.3 0-192.5-86.2-192.5-192.5S117.3 59 223.6 59s192.5 86.2 192.5 192.5S329.9 444 223.6 444z"],cuttlefish:[440,512,[],"f38c","M344 305.5c-17.5 31.6-57.4 54.5-96 54.5-56.6 0-104-47.4-104-104s47.4-104 104-104c38.6 0 78.5 22.9 96 54.5 13.7-50.9 41.7-93.3 87-117.8C385.7 39.1 320.5 8 248 8 111 8 0 119 0 256s111 248 248 248c72.5 0 137.7-31.1 183-80.7-45.3-24.5-73.3-66.9-87-117.8z"],blogger:[448,512,[],"f37c","M162.4 196c4.8-4.9 6.2-5.1 36.4-5.1 27.2 0 28.1.1 32.1 2.1 5.8 2.9 8.3 7 8.3 13.6 0 5.9-2.4 10-7.6 13.4-2.8 1.8-4.5 1.9-31.1 2.1-16.4.1-29.5-.2-31.5-.8-10.3-2.9-14.1-17.7-6.6-25.3zm61.4 94.5c-53.9 0-55.8.2-60.2 4.1-3.5 3.1-5.7 9.4-5.1 13.9.7 4.7 4.8 10.1 9.2 12 2.2 1 14.1 1.7 56.3 1.2l47.9-.6 9.2-1.5c9-5.1 10.5-17.4 3.1-24.4-5.3-4.7-5-4.7-60.4-4.7zm223.4 130.1c-3.5 28.4-23 50.4-51.1 57.5-7.2 1.8-9.7 1.9-172.9 1.8-157.8 0-165.9-.1-172-1.8-8.4-2.2-15.6-5.5-22.3-10-5.6-3.8-13.9-11.8-17-16.4-3.8-5.6-8.2-15.3-10-22C.1 423 0 420.3 0 256.3 0 93.2 0 89.7 1.8 82.6 8.1 57.9 27.7 39 53 33.4c7.3-1.6 332.1-1.9 340-.3 21.2 4.3 37.9 17.1 47.6 36.4 7.7 15.3 7-1.5 7.3 180.6.2 115.8 0 164.5-.7 170.5zm-85.4-185.2c-1.1-5-4.2-9.6-7.7-11.5-1.1-.6-8-1.3-15.5-1.7-12.4-.6-13.8-.8-17.8-3.1-6.2-3.6-7.9-7.6-8-18.3 0-20.4-8.5-39.4-25.3-56.5-12-12.2-25.3-20.5-40.6-25.1-3.6-1.1-11.8-1.5-39.2-1.8-42.9-.5-52.5.4-67.1 6.2-27 10.7-46.3 33.4-53.4 62.4-1.3 5.4-1.6 14.2-1.9 64.3-.4 62.8 0 72.1 4 84.5 9.7 30.7 37.1 53.4 64.6 58.4 9.2 1.7 122.2 2.1 133.7.5 20.1-2.7 35.9-10.8 50.7-25.9 10.7-10.9 17.4-22.8 21.8-38.5 3.2-10.9 2.9-88.4 1.7-93.9z"],flickr:[448,512,[],"f16e","M400 32H48C21.5 32 0 53.5 0 80v352c0 26.5 21.5 48 48 48h352c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48zM144.5 319c-35.1 0-63.5-28.4-63.5-63.5s28.4-63.5 63.5-63.5 63.5 28.4 63.5 63.5-28.4 63.5-63.5 63.5zm159 0c-35.1 0-63.5-28.4-63.5-63.5s28.4-63.5 63.5-63.5 63.5 28.4 63.5 63.5-28.4 63.5-63.5 63.5z"],viber:[512,512,[],"f409","M444 49.9C431.3 38.2 379.9.9 265.3.4c0 0-135.1-8.1-200.9 52.3C27.8 89.3 14.9 143 13.5 209.5c-1.4 66.5-3.1 191.1 117 224.9h.1l-.1 51.6s-.8 20.9 13 25.1c16.6 5.2 26.4-10.7 42.3-27.8 8.7-9.4 20.7-23.2 29.8-33.7 82.2 6.9 145.3-8.9 152.5-11.2 16.6-5.4 110.5-17.4 125.7-142 15.8-128.6-7.6-209.8-49.8-246.5zM457.9 287c-12.9 104-89 110.6-103 115.1-6 1.9-61.5 15.7-131.2 11.2 0 0-52 62.7-68.2 79-5.3 5.3-11.1 4.8-11-5.7 0-6.9.4-85.7.4-85.7-.1 0-.1 0 0 0-101.8-28.2-95.8-134.3-94.7-189.8 1.1-55.5 11.6-101 42.6-131.6 55.7-50.5 170.4-43 170.4-43 96.9.4 143.3 29.6 154.1 39.4 35.7 30.6 53.9 103.8 40.6 211.1zm-139-80.8c.4 8.6-12.5 9.2-12.9.6-1.1-22-11.4-32.7-32.6-33.9-8.6-.5-7.8-13.4.7-12.9 27.9 1.5 43.4 17.5 44.8 46.2zm20.3 11.3c1-42.4-25.5-75.6-75.8-79.3-8.5-.6-7.6-13.5.9-12.9 58 4.2 88.9 44.1 87.8 92.5-.1 8.6-13.1 8.2-12.9-.3zm47 13.4c.1 8.6-12.9 8.7-12.9.1-.6-81.5-54.9-125.9-120.8-126.4-8.5-.1-8.5-12.9 0-12.9 73.7.5 133 51.4 133.7 139.2zM374.9 329v.2c-10.8 19-31 40-51.8 33.3l-.2-.3c-21.1-5.9-70.8-31.5-102.2-56.5-16.2-12.8-31-27.9-42.4-42.4-10.3-12.9-20.7-28.2-30.8-46.6-21.3-38.5-26-55.7-26-55.7-6.7-20.8 14.2-41 33.3-51.8h.2c9.2-4.8 18-3.2 23.9 3.9 0 0 12.4 14.8 17.7 22.1 5 6.8 11.7 17.7 15.2 23.8 6.1 10.9 2.3 22-3.7 26.6l-12 9.6c-6.1 4.9-5.3 14-5.3 14s17.8 67.3 84.3 84.3c0 0 9.1.8 14-5.3l9.6-12c4.6-6 15.7-9.8 26.6-3.7 14.7 8.3 33.4 21.2 45.8 32.9 7 5.7 8.6 14.4 3.8 23.6z"],soundcloud:[640,512,[],"f1be","M111.4 256.3l5.8 65-5.8 68.3c-.3 2.5-2.2 4.4-4.4 4.4s-4.2-1.9-4.2-4.4l-5.6-68.3 5.6-65c0-2.2 1.9-4.2 4.2-4.2 2.2 0 4.1 2 4.4 4.2zm21.4-45.6c-2.8 0-4.7 2.2-5 5l-5 105.6 5 68.3c.3 2.8 2.2 5 5 5 2.5 0 4.7-2.2 4.7-5l5.8-68.3-5.8-105.6c0-2.8-2.2-5-4.7-5zm25.5-24.1c-3.1 0-5.3 2.2-5.6 5.3l-4.4 130 4.4 67.8c.3 3.1 2.5 5.3 5.6 5.3 2.8 0 5.3-2.2 5.3-5.3l5.3-67.8-5.3-130c0-3.1-2.5-5.3-5.3-5.3zM7.2 283.2c-1.4 0-2.2 1.1-2.5 2.5L0 321.3l4.7 35c.3 1.4 1.1 2.5 2.5 2.5s2.2-1.1 2.5-2.5l5.6-35-5.6-35.6c-.3-1.4-1.1-2.5-2.5-2.5zm23.6-21.9c-1.4 0-2.5 1.1-2.5 2.5l-6.4 57.5 6.4 56.1c0 1.7 1.1 2.8 2.5 2.8s2.5-1.1 2.8-2.5l7.2-56.4-7.2-57.5c-.3-1.4-1.4-2.5-2.8-2.5zm25.3-11.4c-1.7 0-3.1 1.4-3.3 3.3L47 321.3l5.8 65.8c.3 1.7 1.7 3.1 3.3 3.1 1.7 0 3.1-1.4 3.1-3.1l6.9-65.8-6.9-68.1c0-1.9-1.4-3.3-3.1-3.3zm25.3-2.2c-1.9 0-3.6 1.4-3.6 3.6l-5.8 70 5.8 67.8c0 2.2 1.7 3.6 3.6 3.6s3.6-1.4 3.9-3.6l6.4-67.8-6.4-70c-.3-2.2-2-3.6-3.9-3.6zm241.4-110.9c-1.1-.8-2.8-1.4-4.2-1.4-2.2 0-4.2.8-5.6 1.9-1.9 1.7-3.1 4.2-3.3 6.7v.8l-3.3 176.7 1.7 32.5 1.7 31.7c.3 4.7 4.2 8.6 8.9 8.6s8.6-3.9 8.6-8.6l3.9-64.2-3.9-177.5c-.4-3-2-5.8-4.5-7.2zm-26.7 15.3c-1.4-.8-2.8-1.4-4.4-1.4s-3.1.6-4.4 1.4c-2.2 1.4-3.6 3.9-3.6 6.7l-.3 1.7-2.8 160.8s0 .3 3.1 65.6v.3c0 1.7.6 3.3 1.7 4.7 1.7 1.9 3.9 3.1 6.4 3.1 2.2 0 4.2-1.1 5.6-2.5 1.7-1.4 2.5-3.3 2.5-5.6l.3-6.7 3.1-58.6-3.3-162.8c-.3-2.8-1.7-5.3-3.9-6.7zm-111.4 22.5c-3.1 0-5.8 2.8-5.8 6.1l-4.4 140.6 4.4 67.2c.3 3.3 2.8 5.8 5.8 5.8 3.3 0 5.8-2.5 6.1-5.8l5-67.2-5-140.6c-.2-3.3-2.7-6.1-6.1-6.1zm376.7 62.8c-10.8 0-21.1 2.2-30.6 6.1-6.4-70.8-65.8-126.4-138.3-126.4-17.8 0-35 3.3-50.3 9.4-6.1 2.2-7.8 4.4-7.8 9.2v249.7c0 5 3.9 8.6 8.6 9.2h218.3c43.3 0 78.6-35 78.6-78.3.1-43.6-35.2-78.9-78.5-78.9zm-296.7-60.3c-4.2 0-7.5 3.3-7.8 7.8l-3.3 136.7 3.3 65.6c.3 4.2 3.6 7.5 7.8 7.5 4.2 0 7.5-3.3 7.5-7.5l3.9-65.6-3.9-136.7c-.3-4.5-3.3-7.8-7.5-7.8zm-53.6-7.8c-3.3 0-6.4 3.1-6.4 6.7l-3.9 145.3 3.9 66.9c.3 3.6 3.1 6.4 6.4 6.4 3.6 0 6.4-2.8 6.7-6.4l4.4-66.9-4.4-145.3c-.3-3.6-3.1-6.7-6.7-6.7zm26.7 3.4c-3.9 0-6.9 3.1-6.9 6.9L227 321.3l3.9 66.4c.3 3.9 3.1 6.9 6.9 6.9s6.9-3.1 6.9-6.9l4.2-66.4-4.2-141.7c0-3.9-3-6.9-6.9-6.9z"],digg:[512,512,[],"f1a6","M81.7 172.3H0v174.4h132.7V96h-51v76.3zm0 133.4H50.9v-92.3h30.8v92.3zm297.2-133.4v174.4h81.8v28.5h-81.8V416H512V172.3H378.9zm81.8 133.4h-30.8v-92.3h30.8v92.3zm-235.6 41h82.1v28.5h-82.1V416h133.3V172.3H225.1v174.4zm51.2-133.3h30.8v92.3h-30.8v-92.3zM153.3 96h51.3v51h-51.3V96zm0 76.3h51.3v174.4h-51.3V172.3z"],"tencent-weibo":[384,512,[],"f1d5","M72.3 495.8c1.4 19.9-27.6 22.2-29.7 2.9C31 368.8 73.7 259.2 144 185.5c-15.6-34 9.2-77.1 50.6-77.1 30.3 0 55.1 24.6 55.1 55.1 0 44-49.5 70.8-86.9 45.1-65.7 71.3-101.4 169.8-90.5 287.2zM192 .1C66.1.1-12.3 134.3 43.7 242.4 52.4 259.8 79 246.9 70 229 23.7 136.4 91 29.8 192 29.8c75.4 0 136.9 61.4 136.9 136.9 0 90.8-86.9 153.9-167.7 133.1-19.1-4.1-25.6 24.4-6.6 29.1 110.7 23.2 204-60 204-162.3C358.6 74.7 284 .1 192 .1z"],letterboxd:[640,512,[],"e62d","M521.3 128C586.9 128 640 181.1 640 246.6s-53.1 118.6-118.7 118.6c-42.5 0-79.7-22.3-100.7-55.8c11.4-18.2 18-39.7 18-62.8s-6.6-44.6-18-62.8l0 0 .8-1.2c20.8-32.3 56.8-53.9 97.9-54.6l2 0zM320 128c42.5 0 79.7 22.3 100.7 55.8c-11.4 18.2-18 39.7-18 62.8s6.6 44.6 18 62.8l0 0-.8 1.2c-20.8 32.3-56.8 53.9-97.9 54.6l-2 0c-42.5 0-79.7-22.3-100.7-55.8c11.4-18.2 18-39.7 18-62.8s-6.6-44.6-18-62.8l0 0 .8-1.2c20.8-32.3 56.8-53.9 97.9-54.6l2 0zm-201.3 0c42.5 0 79.7 22.3 100.7 55.8c-11.4 18.2-18 39.7-18 62.8s6.6 44.6 18 62.8l0 0-.8 1.2c-20.8 32.3-56.8 53.9-97.9 54.6l-2 0C53.1 365.1 0 312.1 0 246.6S53.1 128 118.7 128z"],symfony:[512,512,[],"f83d","M256 8C119 8 8 119 8 256s111 248 248 248 248-111 248-248S393 8 256 8zm133.74 143.54c-11.47.41-19.4-6.45-19.77-16.87-.27-9.18 6.68-13.44 6.53-18.85-.23-6.55-10.16-6.82-12.87-6.67-39.78 1.29-48.59 57-58.89 113.85 21.43 3.15 36.65-.72 45.14-6.22 12-7.75-3.34-15.72-1.42-24.56 4-18.16 32.55-19 32 5.3-.36 17.86-25.92 41.81-77.6 35.7-10.76 59.52-18.35 115-58.2 161.72-29 34.46-58.4 39.82-71.58 40.26-24.65.85-41-12.31-41.58-29.84-.56-17 14.45-26.26 24.31-26.59 21.89-.75 30.12 25.67 14.88 34-12.09 9.71.11 12.61 2.05 12.55 10.42-.36 17.34-5.51 22.18-9 24-20 33.24-54.86 45.35-118.35 8.19-49.66 17-78 18.23-82-16.93-12.75-27.08-28.55-49.85-34.72-15.61-4.23-25.12-.63-31.81 7.83-7.92 10-5.29 23 2.37 30.7l12.63 14c15.51 17.93 24 31.87 20.8 50.62-5.06 29.93-40.72 52.9-82.88 39.94-36-11.11-42.7-36.56-38.38-50.62 7.51-24.15 42.36-11.72 34.62 13.6-2.79 8.6-4.92 8.68-6.28 13.07-4.56 14.77 41.85 28.4 51-1.39 4.47-14.52-5.3-21.71-22.25-39.85-28.47-31.75-16-65.49 2.95-79.67C204.23 140.13 251.94 197 262 205.29c37.17-109 100.53-105.46 102.43-105.53 25.16-.81 44.19 10.59 44.83 28.65.25 7.69-4.17 22.59-19.52 23.13z"],maxcdn:[512,512,[],"f136","M461.1 442.7h-97.4L415.6 200c2.3-10.2.9-19.5-4.4-25.7-5-6.1-13.7-9.6-24.2-9.6h-49.3l-59.5 278h-97.4l59.5-278h-83.4l-59.5 278H0l59.5-278-44.6-95.4H387c39.4 0 75.3 16.3 98.3 44.9 23.3 28.6 31.8 67.4 23.6 105.9l-47.8 222.6z"],etsy:[384,512,[],"f2d7","M384 348c-1.75 10.75-13.75 110-15.5 132-117.879-4.299-219.895-4.743-368.5 0v-25.5c45.457-8.948 60.627-8.019 61-35.25 1.793-72.322 3.524-244.143 0-322-1.029-28.46-12.13-26.765-61-36v-25.5c73.886 2.358 255.933 8.551 362.999-3.75-3.5 38.25-7.75 126.5-7.75 126.5H332C320.947 115.665 313.241 68 277.25 68h-137c-10.25 0-10.75 3.5-10.75 9.75V241.5c58 .5 88.5-2.5 88.5-2.5 29.77-.951 27.56-8.502 40.75-65.251h25.75c-4.407 101.351-3.91 61.829-1.75 160.25H257c-9.155-40.086-9.065-61.045-39.501-61.5 0 0-21.5-2-88-2v139c0 26 14.25 38.25 44.25 38.25H263c63.636 0 66.564-24.996 98.751-99.75H384z"],"facebook-messenger":[512,512,[],"f39f","M256.55 8C116.52 8 8 110.34 8 248.57c0 72.3 29.71 134.78 78.07 177.94 8.35 7.51 6.63 11.86 8.05 58.23A19.92 19.92 0 0 0 122 502.31c52.91-23.3 53.59-25.14 62.56-22.7C337.85 521.8 504 423.7 504 248.57 504 110.34 396.59 8 256.55 8zm149.24 185.13l-73 115.57a37.37 37.37 0 0 1-53.91 9.93l-58.08-43.47a15 15 0 0 0-18 0l-78.37 59.44c-10.46 7.93-24.16-4.6-17.11-15.67l73-115.57a37.36 37.36 0 0 1 53.91-9.93l58.06 43.46a15 15 0 0 0 18 0l78.41-59.38c10.44-7.98 24.14 4.54 17.09 15.62z"],audible:[640,512,[],"f373","M640 199.9v54l-320 200L0 254v-54l320 200 320-200.1zm-194.5 72l47.1-29.4c-37.2-55.8-100.7-92.6-172.7-92.6-72 0-135.5 36.7-172.6 92.4h.3c2.5-2.3 5.1-4.5 7.7-6.7 89.7-74.4 219.4-58.1 290.2 36.3zm-220.1 18.8c16.9-11.9 36.5-18.7 57.4-18.7 34.4 0 65.2 18.4 86.4 47.6l45.4-28.4c-20.9-29.9-55.6-49.5-94.8-49.5-38.9 0-73.4 19.4-94.4 49zM103.6 161.1c131.8-104.3 318.2-76.4 417.5 62.1l.7 1 48.8-30.4C517.1 112.1 424.8 58.1 319.9 58.1c-103.5 0-196.6 53.5-250.5 135.6 9.9-10.5 22.7-23.5 34.2-32.6zm467 32.7z"],"think-peaks":[576,512,[],"f731","M465.4 409.4l87.1-150.2-32-.3-55.1 95L259.2 0 23 407.4l32 .3L259.2 55.6zm-355.3-44.1h32.1l117.4-202.5L463 511.9l32.5.1-235.8-404.6z"],bilibili:[512,512,[],"e3d9","M488.6 104.1C505.3 122.2 513 143.8 511.9 169.8V372.2C511.5 398.6 502.7 420.3 485.4 437.3C468.2 454.3 446.3 463.2 419.9 464H92.02C65.57 463.2 43.81 454.2 26.74 436.8C9.682 419.4 .7667 396.5 0 368.2V169.8C.7667 143.8 9.682 122.2 26.74 104.1C43.81 87.75 65.57 78.77 92.02 78H121.4L96.05 52.19C90.3 46.46 87.42 39.19 87.42 30.4C87.42 21.6 90.3 14.34 96.05 8.603C101.8 2.868 109.1 0 117.9 0C126.7 0 134 2.868 139.8 8.603L213.1 78H301.1L375.6 8.603C381.7 2.868 389.2 0 398 0C406.8 0 414.1 2.868 419.9 8.603C425.6 14.34 428.5 21.6 428.5 30.4C428.5 39.19 425.6 46.46 419.9 52.19L394.6 78L423.9 78C450.3 78.77 471.9 87.75 488.6 104.1H488.6zM449.8 173.8C449.4 164.2 446.1 156.4 439.1 150.3C433.9 144.2 425.1 140.9 416.4 140.5H96.05C86.46 140.9 78.6 144.2 72.47 150.3C66.33 156.4 63.07 164.2 62.69 173.8V368.2C62.69 377.4 65.95 385.2 72.47 391.7C78.99 398.2 86.85 401.5 96.05 401.5H416.4C425.6 401.5 433.4 398.2 439.7 391.7C446 385.2 449.4 377.4 449.8 368.2L449.8 173.8zM185.5 216.5C191.8 222.8 195.2 230.6 195.6 239.7V273C195.2 282.2 191.9 289.9 185.8 296.2C179.6 302.5 171.8 305.7 162.2 305.7C152.6 305.7 144.7 302.5 138.6 296.2C132.5 289.9 129.2 282.2 128.8 273V239.7C129.2 230.6 132.6 222.8 138.9 216.5C145.2 210.2 152.1 206.9 162.2 206.5C171.4 206.9 179.2 210.2 185.5 216.5H185.5zM377 216.5C383.3 222.8 386.7 230.6 387.1 239.7V273C386.7 282.2 383.4 289.9 377.3 296.2C371.2 302.5 363.3 305.7 353.7 305.7C344.1 305.7 336.3 302.5 330.1 296.2C323.1 289.9 320.7 282.2 320.4 273V239.7C320.7 230.6 324.1 222.8 330.4 216.5C336.7 210.2 344.5 206.9 353.7 206.5C362.9 206.9 370.7 210.2 377 216.5H377z"],erlang:[640,512,[],"f39d","M87.2 53.5H0v405h100.4c-49.7-52.6-78.8-125.3-78.7-212.1-.1-76.7 24-142.7 65.5-192.9zm238.2 9.7c-45.9.1-85.1 33.5-89.2 83.2h169.9c-1.1-49.7-34.5-83.1-80.7-83.2zm230.7-9.6h.3l-.1-.1zm.3 0c31.4 42.7 48.7 97.5 46.2 162.7.5 6 .5 11.7 0 24.1H230.2c-.2 109.7 38.9 194.9 138.6 195.3 68.5-.3 118-51 151.9-106.1l96.4 48.2c-17.4 30.9-36.5 57.8-57.9 80.8H640v-405z"],"x-twitter":[512,512,[],"e61b","M389.2 48h70.6L305.6 224.2 487 464H345L233.7 318.6 106.5 464H35.8L200.7 275.5 26.8 48H172.4L272.9 180.9 389.2 48zM364.4 421.8h39.1L151.1 88h-42L364.4 421.8z"],"cotton-bureau":[512,512,[],"f89e","M474.31 330.41c-23.66 91.85-94.23 144.59-201.9 148.35V429.6c0-48 26.41-74.39 74.39-74.39 62 0 99.2-37.2 99.2-99.21 0-61.37-36.53-98.28-97.38-99.06-33-69.32-146.5-64.65-177.24 0C110.52 157.72 74 194.63 74 256c0 62.13 37.27 99.41 99.4 99.41 48 0 74.55 26.23 74.55 74.39V479c-134.43-5-211.1-85.07-211.1-223 0-141.82 81.35-223.2 223.2-223.2 114.77 0 189.84 53.2 214.69 148.81H500C473.88 71.51 388.22 8 259.82 8 105 8 12 101.19 12 255.82 12 411.14 105.19 504.34 259.82 504c128.27 0 213.87-63.81 239.67-173.59zM357 182.33c41.37 3.45 64.2 29 64.2 73.67 0 48-26.43 74.41-74.4 74.41-28.61 0-49.33-9.59-61.59-27.33 83.06-16.55 75.59-99.67 71.79-120.75zm-81.68 97.36c-2.46-10.34-16.33-87 56.23-97 2.27 10.09 16.52 87.11-56.26 97zM260 132c28.61 0 49 9.67 61.44 27.61-28.36 5.48-49.36 20.59-61.59 43.45-12.23-22.86-33.23-38-61.6-43.45 12.41-17.69 33.27-27.35 61.57-27.35zm-71.52 50.72c73.17 10.57 58.91 86.81 56.49 97-72.41-9.84-59-86.95-56.25-97zM173.2 330.41c-48 0-74.4-26.4-74.4-74.41 0-44.36 22.86-70 64.22-73.67-6.75 37.2-1.38 106.53 71.65 120.75-12.14 17.63-32.84 27.3-61.14 27.3zm53.21 12.39A80.8 80.8 0 0 0 260 309.25c7.77 14.49 19.33 25.54 33.82 33.55a80.28 80.28 0 0 0-33.58 33.83c-8-14.5-19.07-26.23-33.56-33.83z"],dashcube:[448,512,[],"f210","M326.6 104H110.4c-51.1 0-91.2 43.3-91.2 93.5V427c0 50.5 40.1 85 91.2 85h227.2c51.1 0 91.2-34.5 91.2-85V0L326.6 104zM153.9 416.5c-17.7 0-32.4-15.1-32.4-32.8V240.8c0-17.7 14.7-32.5 32.4-32.5h140.7c17.7 0 32 14.8 32 32.5v123.5l51.1 52.3H153.9z"],"42-group":[640,512,["innosoft"],"e080","M320 96V416C341.011 416 361.818 411.861 381.23 403.821C400.641 395.78 418.28 383.995 433.138 369.138C447.995 354.28 459.78 336.641 467.821 317.23C475.861 297.818 480 277.011 480 256C480 234.989 475.861 214.182 467.821 194.771C459.78 175.359 447.995 157.72 433.138 142.863C418.28 128.005 400.641 116.22 381.23 108.179C361.818 100.139 341.011 96 320 96ZM0 256L160.002 416L320.003 256L160.002 96L0 256ZM480 256C480 277.011 484.138 297.818 492.179 317.23C500.219 336.643 512.005 354.28 526.862 369.138C541.72 383.995 559.357 395.781 578.77 403.821C598.182 411.862 618.989 416 640 416V96C597.565 96 556.869 112.858 526.862 142.863C496.857 172.869 480 213.565 480 256Z"],"stack-exchange":[448,512,[],"f18d","M17.7 332.3h412.7v22c0 37.7-29.3 68-65.3 68h-19L259.3 512v-89.7H83c-36 0-65.3-30.3-65.3-68v-22zm0-23.6h412.7v-85H17.7v85zm0-109.4h412.7v-85H17.7v85zM365 0H83C47 0 17.7 30.3 17.7 67.7V90h412.7V67.7C430.3 30.3 401 0 365 0z"],elementor:[512,512,[],"f430","M.361 256C.361 397 114 511 255 511C397 511 511 397 511 256C511 116 397 2.05 255 2.05C114 2.05 .361 116 .361 256zM192 150V363H149V150H192zM234 150H362V193H234V150zM362 235V278H234V235H362zM234 320H362V363H234V320z"],"square-pied-piper":[448,512,["pied-piper-square"],"e01e","M32 419L0 479.2l.8-328C.8 85.3 54 32 120 32h327.2c-93 28.9-189.9 94.2-253.9 168.6C122.7 282 82.6 338 32 419M448 32S305.2 98.8 261.6 199.1c-23.2 53.6-28.9 118.1-71 158.6-28.9 27.8-69.8 38.2-105.3 56.3-23.2 12-66.4 40.5-84.9 66h328.4c66 0 119.3-53.3 119.3-119.2-.1 0-.1-328.8-.1-328.8z"],"creative-commons-nd":[496,512,[],"f4eb","M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zm94 144.3v42.5H162.1V197h180.3zm0 79.8v42.5H162.1v-42.5h180.3z"],palfed:[576,512,[],"f3d8","M384.9 193.9c0-47.4-55.2-44.2-95.4-29.8-1.3 39.4-2.5 80.7-3 119.8.7 2.8 2.6 6.2 15.1 6.2 36.8 0 83.4-42.8 83.3-96.2zm-194.5 72.2c.2 0 6.5-2.7 11.2-2.7 26.6 0 20.7 44.1-14.4 44.1-21.5 0-37.1-18.1-37.1-43 0-42 42.9-95.6 100.7-126.5 1-12.4 3-22 10.5-28.2 11.2-9 26.6-3.5 29.5 11.1 72.2-22.2 135.2 1 135.2 72 0 77.9-79.3 152.6-140.1 138.2-.1 39.4.9 74.4 2.7 100v.2c.2 3.4.6 12.5-5.3 19.1-9.6 10.6-33.4 10-36.4-22.3-4.1-44.4.2-206.1 1.4-242.5-21.5 15-58.5 50.3-58.5 75.9.2 2.5.4 4 .6 4.6zM8 181.1s-.1 37.4 38.4 37.4h30l22.4 217.2s0 44.3 44.7 44.3h288.9s44.7-.4 44.7-44.3l22.4-217.2h30s38.4 1.2 38.4-37.4c0 0 .1-37.4-38.4-37.4h-30.1c-7.3-25.6-30.2-74.3-119.4-74.3h-28V50.3s-2.7-18.4-21.1-18.4h-85.8s-21.1 0-21.1 18.4v19.1h-28.1s-105 4.2-120.5 74.3h-29S8 142.5 8 181.1z"],superpowers:[448,512,[],"f2dd","M448 32c-83.3 11-166.8 22-250 33-92 12.5-163.3 86.7-169 180-3.3 55.5 18 109.5 57.8 148.2L0 480c83.3-11 166.5-22 249.8-33 91.8-12.5 163.3-86.8 168.7-179.8 3.5-55.5-18-109.5-57.7-148.2L448 32zm-79.7 232.3c-4.2 79.5-74 139.2-152.8 134.5-79.5-4.7-140.7-71-136.3-151 4.5-79.2 74.3-139.3 153-134.5 79.3 4.7 140.5 71 136.1 151z"],resolving:[496,512,[],"f3e7","M281.2 278.2c46-13.3 49.6-23.5 44-43.4L314 195.5c-6.1-20.9-18.4-28.1-71.1-12.8L54.7 236.8l28.6 98.6 197.9-57.2zM248.5 8C131.4 8 33.2 88.7 7.2 197.5l221.9-63.9c34.8-10.2 54.2-11.7 79.3-8.2 36.3 6.1 52.7 25 61.4 55.2l10.7 37.8c8.2 28.1 1 50.6-23.5 73.6-19.4 17.4-31.2 24.5-61.4 33.2L203 351.8l220.4 27.1 9.7 34.2-48.1 13.3-286.8-37.3 23 80.2c36.8 22 80.3 34.7 126.3 34.7 137 0 248.5-111.4 248.5-248.3C497 119.4 385.5 8 248.5 8zM38.3 388.6L0 256.8c0 48.5 14.3 93.4 38.3 131.8z"],xbox:[512,512,[],"f412","M369.9 318.2c44.3 54.3 64.7 98.8 54.4 118.7-7.9 15.1-56.7 44.6-92.6 55.9-29.6 9.3-68.4 13.3-100.4 10.2-38.2-3.7-76.9-17.4-110.1-39C93.3 445.8 87 438.3 87 423.4c0-29.9 32.9-82.3 89.2-142.1 32-33.9 76.5-73.7 81.4-72.6 9.4 2.1 84.3 75.1 112.3 109.5zM188.6 143.8c-29.7-26.9-58.1-53.9-86.4-63.4-15.2-5.1-16.3-4.8-28.7 8.1-29.2 30.4-53.5 79.7-60.3 122.4-5.4 34.2-6.1 43.8-4.2 60.5 5.6 50.5 17.3 85.4 40.5 120.9 9.5 14.6 12.1 17.3 9.3 9.9-4.2-11-.3-37.5 9.5-64 14.3-39 53.9-112.9 120.3-194.4zm311.6 63.5C483.3 127.3 432.7 77 425.6 77c-7.3 0-24.2 6.5-36 13.9-23.3 14.5-41 31.4-64.3 52.8C367.7 197 427.5 283.1 448.2 346c6.8 20.7 9.7 41.1 7.4 52.3-1.7 8.5-1.7 8.5 1.4 4.6 6.1-7.7 19.9-31.3 25.4-43.5 7.4-16.2 15-40.2 18.6-58.7 4.3-22.5 3.9-70.8-.8-93.4zM141.3 43C189 40.5 251 77.5 255.6 78.4c.7.1 10.4-4.2 21.6-9.7 63.9-31.1 94-25.8 107.4-25.2-63.9-39.3-152.7-50-233.9-11.7-23.4 11.1-24 11.9-9.4 11.2z"],"square-web-awesome-stroke":[448,512,[],"e684","M64 64C46.3 64 32 78.3 32 96l0 320c0 17.7 14.3 32 32 32l320 0c17.7 0 32-14.3 32-32l0-320c0-17.7-14.3-32-32-32L64 64zM0 96C0 60.7 28.7 32 64 32l320 0c35.3 0 64 28.7 64 64l0 320c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64L0 96zm235 81.6L288 224l52.2-10.4c-2.6-3.9-4.2-8.5-4.2-13.6c0-13.3 10.7-24 24-24s24 10.7 24 24c0 13-10.3 23.6-23.2 24L304.5 349.1c-5.2 11.5-16.6 18.9-29.2 18.9l-102.6 0c-12.6 0-24-7.4-29.2-18.9L87.2 224C74.3 223.6 64 213 64 200c0-13.3 10.7-24 24-24s24 10.7 24 24c0 5-1.5 9.7-4.2 13.6L160 224l53.1-46.4c-8.9-4.1-15-13.1-15-23.6c0-14.4 11.6-26 26-26s26 11.6 26 26c0 10.5-6.2 19.5-15.1 23.6z"],searchengin:[460,512,[],"f3eb","M220.6 130.3l-67.2 28.2V43.2L98.7 233.5l54.7-24.2v130.3l67.2-209.3zm-83.2-96.7l-1.3 4.7-15.2 52.9C80.6 106.7 52 145.8 52 191.5c0 52.3 34.3 95.9 83.4 105.5v53.6C57.5 340.1 0 272.4 0 191.6c0-80.5 59.8-147.2 137.4-158zm311.4 447.2c-11.2 11.2-23.1 12.3-28.6 10.5-5.4-1.8-27.1-19.9-60.4-44.4-33.3-24.6-33.6-35.7-43-56.7-9.4-20.9-30.4-42.6-57.5-52.4l-9.7-14.7c-24.7 16.9-53 26.9-81.3 28.7l2.1-6.6 15.9-49.5c46.5-11.9 80.9-54 80.9-104.2 0-54.5-38.4-102.1-96-107.1V32.3C254.4 37.4 320 106.8 320 191.6c0 33.6-11.2 64.7-29 90.4l14.6 9.6c9.8 27.1 31.5 48 52.4 57.4s32.2 9.7 56.8 43c24.6 33.2 42.7 54.9 44.5 60.3s.7 17.3-10.5 28.5zm-9.9-17.9c0-4.4-3.6-8-8-8s-8 3.6-8 8 3.6 8 8 8 8-3.6 8-8z"],tiktok:[448,512,[],"e07b","M448,209.91a210.06,210.06,0,0,1-122.77-39.25V349.38A162.55,162.55,0,1,1,185,188.31V278.2a74.62,74.62,0,1,0,52.23,71.18V0l88,0a121.18,121.18,0,0,0,1.86,22.17h0A122.18,122.18,0,0,0,381,102.39a121.43,121.43,0,0,0,67,20.14Z"],"square-facebook":[448,512,["facebook-square"],"f082","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64h98.2V334.2H109.4V256h52.8V222.3c0-87.1 39.4-127.5 125-127.5c16.2 0 44.2 3.2 55.7 6.4V172c-6-.6-16.5-1-29.6-1c-42 0-58.2 15.9-58.2 57.2V256h83.6l-14.4 78.2H255V480H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64z"],renren:[512,512,[],"f18b","M214 169.1c0 110.4-61 205.4-147.6 247.4C30 373.2 8 317.7 8 256.6 8 133.9 97.1 32.2 214 12.5v156.6zM255 504c-42.9 0-83.3-11-118.5-30.4C193.7 437.5 239.9 382.9 255 319c15.5 63.9 61.7 118.5 118.8 154.7C338.7 493 298.3 504 255 504zm190.6-87.5C359 374.5 298 279.6 298 169.1V12.5c116.9 19.7 206 121.4 206 244.1 0 61.1-22 116.6-58.4 159.9z"],linux:[448,512,[],"f17c","M220.8 123.3c1 .5 1.8 1.7 3 1.7 1.1 0 2.8-.4 2.9-1.5.2-1.4-1.9-2.3-3.2-2.9-1.7-.7-3.9-1-5.5-.1-.4.2-.8.7-.6 1.1.3 1.3 2.3 1.1 3.4 1.7zm-21.9 1.7c1.2 0 2-1.2 3-1.7 1.1-.6 3.1-.4 3.5-1.6.2-.4-.2-.9-.6-1.1-1.6-.9-3.8-.6-5.5.1-1.3.6-3.4 1.5-3.2 2.9.1 1 1.8 1.5 2.8 1.4zM420 403.8c-3.6-4-5.3-11.6-7.2-19.7-1.8-8.1-3.9-16.8-10.5-22.4-1.3-1.1-2.6-2.1-4-2.9-1.3-.8-2.7-1.5-4.1-2 9.2-27.3 5.6-54.5-3.7-79.1-11.4-30.1-31.3-56.4-46.5-74.4-17.1-21.5-33.7-41.9-33.4-72C311.1 85.4 315.7.1 234.8 0 132.4-.2 158 103.4 156.9 135.2c-1.7 23.4-6.4 41.8-22.5 64.7-18.9 22.5-45.5 58.8-58.1 96.7-6 17.9-8.8 36.1-6.2 53.3-6.5 5.8-11.4 14.7-16.6 20.2-4.2 4.3-10.3 5.9-17 8.3s-14 6-18.5 14.5c-2.1 3.9-2.8 8.1-2.8 12.4 0 3.9.6 7.9 1.2 11.8 1.2 8.1 2.5 15.7.8 20.8-5.2 14.4-5.9 24.4-2.2 31.7 3.8 7.3 11.4 10.5 20.1 12.3 17.3 3.6 40.8 2.7 59.3 12.5 19.8 10.4 39.9 14.1 55.9 10.4 11.6-2.6 21.1-9.6 25.9-20.2 12.5-.1 26.3-5.4 48.3-6.6 14.9-1.2 33.6 5.3 55.1 4.1.6 2.3 1.4 4.6 2.5 6.7v.1c8.3 16.7 23.8 24.3 40.3 23 16.6-1.3 34.1-11 48.3-27.9 13.6-16.4 36-23.2 50.9-32.2 7.4-4.5 13.4-10.1 13.9-18.3.4-8.2-4.4-17.3-15.5-29.7zM223.7 87.3c9.8-22.2 34.2-21.8 44-.4 6.5 14.2 3.6 30.9-4.3 40.4-1.6-.8-5.9-2.6-12.6-4.9 1.1-1.2 3.1-2.7 3.9-4.6 4.8-11.8-.2-27-9.1-27.3-7.3-.5-13.9 10.8-11.8 23-4.1-2-9.4-3.5-13-4.4-1-6.9-.3-14.6 2.9-21.8zM183 75.8c10.1 0 20.8 14.2 19.1 33.5-3.5 1-7.1 2.5-10.2 4.6 1.2-8.9-3.3-20.1-9.6-19.6-8.4.7-9.8 21.2-1.8 28.1 1 .8 1.9-.2-5.9 5.5-15.6-14.6-10.5-52.1 8.4-52.1zm-13.6 60.7c6.2-4.6 13.6-10 14.1-10.5 4.7-4.4 13.5-14.2 27.9-14.2 7.1 0 15.6 2.3 25.9 8.9 6.3 4.1 11.3 4.4 22.6 9.3 8.4 3.5 13.7 9.7 10.5 18.2-2.6 7.1-11 14.4-22.7 18.1-11.1 3.6-19.8 16-38.2 14.9-3.9-.2-7-1-9.6-2.1-8-3.5-12.2-10.4-20-15-8.6-4.8-13.2-10.4-14.7-15.3-1.4-4.9 0-9 4.2-12.3zm3.3 334c-2.7 35.1-43.9 34.4-75.3 18-29.9-15.8-68.6-6.5-76.5-21.9-2.4-4.7-2.4-12.7 2.6-26.4v-.2c2.4-7.6.6-16-.6-23.9-1.2-7.8-1.8-15 .9-20 3.5-6.7 8.5-9.1 14.8-11.3 10.3-3.7 11.8-3.4 19.6-9.9 5.5-5.7 9.5-12.9 14.3-18 5.1-5.5 10-8.1 17.7-6.9 8.1 1.2 15.1 6.8 21.9 16l19.6 35.6c9.5 19.9 43.1 48.4 41 68.9zm-1.4-25.9c-4.1-6.6-9.6-13.6-14.4-19.6 7.1 0 14.2-2.2 16.7-8.9 2.3-6.2 0-14.9-7.4-24.9-13.5-18.2-38.3-32.5-38.3-32.5-13.5-8.4-21.1-18.7-24.6-29.9s-3-23.3-.3-35.2c5.2-22.9 18.6-45.2 27.2-59.2 2.3-1.7.8 3.2-8.7 20.8-8.5 16.1-24.4 53.3-2.6 82.4.6-20.7 5.5-41.8 13.8-61.5 12-27.4 37.3-74.9 39.3-112.7 1.1.8 4.6 3.2 6.2 4.1 4.6 2.7 8.1 6.7 12.6 10.3 12.4 10 28.5 9.2 42.4 1.2 6.2-3.5 11.2-7.5 15.9-9 9.9-3.1 17.8-8.6 22.3-15 7.7 30.4 25.7 74.3 37.2 95.7 6.1 11.4 18.3 35.5 23.6 64.6 3.3-.1 7 .4 10.9 1.4 13.8-35.7-11.7-74.2-23.3-84.9-4.7-4.6-4.9-6.6-2.6-6.5 12.6 11.2 29.2 33.7 35.2 59 2.8 11.6 3.3 23.7.4 35.7 16.4 6.8 35.9 17.9 30.7 34.8-2.2-.1-3.2 0-4.2 0 3.2-10.1-3.9-17.6-22.8-26.1-19.6-8.6-36-8.6-38.3 12.5-12.1 4.2-18.3 14.7-21.4 27.3-2.8 11.2-3.6 24.7-4.4 39.9-.5 7.7-3.6 18-6.8 29-32.1 22.9-76.7 32.9-114.3 7.2zm257.4-11.5c-.9 16.8-41.2 19.9-63.2 46.5-13.2 15.7-29.4 24.4-43.6 25.5s-26.5-4.8-33.7-19.3c-4.7-11.1-2.4-23.1 1.1-36.3 3.7-14.2 9.2-28.8 9.9-40.6.8-15.2 1.7-28.5 4.2-38.7 2.6-10.3 6.6-17.2 13.7-21.1.3-.2.7-.3 1-.5.8 13.2 7.3 26.6 18.8 29.5 12.6 3.3 30.7-7.5 38.4-16.3 9-.3 15.7-.9 22.6 5.1 9.9 8.5 7.1 30.3 17.1 41.6 10.6 11.6 14 19.5 13.7 24.6zM173.3 148.7c2 1.9 4.7 4.5 8 7.1 6.6 5.2 15.8 10.6 27.3 10.6 11.6 0 22.5-5.9 31.8-10.8 4.9-2.6 10.9-7 14.8-10.4s5.9-6.3 3.1-6.6-2.6 2.6-6 5.1c-4.4 3.2-9.7 7.4-13.9 9.8-7.4 4.2-19.5 10.2-29.9 10.2s-18.7-4.8-24.9-9.7c-3.1-2.5-5.7-5-7.7-6.9-1.5-1.4-1.9-4.6-4.3-4.9-1.4-.1-1.8 3.7 1.7 6.5z"],glide:[448,512,[],"f2a5","M252.8 148.6c0 8.8-1.6 17.7-3.4 26.4-5.8 27.8-11.6 55.8-17.3 83.6-1.4 6.3-8.3 4.9-13.7 4.9-23.8 0-30.5-26-30.5-45.5 0-29.3 11.2-68.1 38.5-83.1 4.3-2.5 9.2-4.2 14.1-4.2 11.4 0 12.3 8.3 12.3 17.9zM448 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h352c26.5 0 48 21.5 48 48zm-64 187c0-5.1-20.8-37.7-25.5-39.5-2.2-.9-7.2-2.3-9.6-2.3-23.1 0-38.7 10.5-58.2 21.5l-.5-.5c4.3-29.4 14.6-57.2 14.6-87.4 0-44.6-23.8-62.7-67.5-62.7-71.7 0-108 70.8-108 123.5 0 54.7 32 85 86.3 85 7.5 0 6.9-.6 6.9 2.3-10.5 80.3-56.5 82.9-56.5 58.9 0-24.4 28-36.5 28.3-38-.2-7.6-29.3-17.2-36.7-17.2-21.1 0-32.7 33-32.7 50.6 0 32.3 20.4 54.7 53.3 54.7 48.2 0 83.4-49.7 94.3-91.7 9.4-37.7 7-39.4 12.3-42.1 20-10.1 35.8-16.8 58.4-16.8 11.1 0 19 2.3 36.7 5.2 1.8.1 4.1-1.7 4.1-3.5z"],linkedin:[448,512,[],"f08c","M416 32H31.9C14.3 32 0 46.5 0 64.3v383.4C0 465.5 14.3 480 31.9 480H416c17.6 0 32-14.5 32-32.3V64.3c0-17.8-14.4-32.3-32-32.3zM135.4 416H69V202.2h66.5V416zm-33.2-243c-21.3 0-38.5-17.3-38.5-38.5S80.9 96 102.2 96c21.2 0 38.5 17.3 38.5 38.5 0 21.3-17.2 38.5-38.5 38.5zm282.1 243h-66.4V312c0-24.8-.5-56.7-34.5-56.7-34.6 0-39.9 27-39.9 54.9V416h-66.4V202.2h63.7v29.2h.9c8.9-16.8 30.6-34.5 62.9-34.5 67.2 0 79.7 44.3 79.7 101.9V416z"],hubspot:[512,512,[],"f3b2","M267.4 211.6c-25.1 23.7-40.8 57.3-40.8 94.6 0 29.3 9.7 56.3 26 78L203.1 434c-4.4-1.6-9.1-2.5-14-2.5-10.8 0-20.9 4.2-28.5 11.8-7.6 7.6-11.8 17.8-11.8 28.6s4.2 20.9 11.8 28.5c7.6 7.6 17.8 11.6 28.5 11.6 10.8 0 20.9-3.9 28.6-11.6 7.6-7.6 11.8-17.8 11.8-28.5 0-4.2-.6-8.2-1.9-12.1l50-50.2c22 16.9 49.4 26.9 79.3 26.9 71.9 0 130-58.3 130-130.2 0-65.2-47.7-119.2-110.2-128.7V116c17.5-7.4 28.2-23.8 28.2-42.9 0-26.1-20.9-47.9-47-47.9S311.2 47 311.2 73.1c0 19.1 10.7 35.5 28.2 42.9v61.2c-15.2 2.1-29.6 6.7-42.7 13.6-27.6-20.9-117.5-85.7-168.9-124.8 1.2-4.4 2-9 2-13.8C129.8 23.4 106.3 0 77.4 0 48.6 0 25.2 23.4 25.2 52.2c0 28.9 23.4 52.3 52.2 52.3 9.8 0 18.9-2.9 26.8-7.6l163.2 114.7zm89.5 163.6c-38.1 0-69-30.9-69-69s30.9-69 69-69 69 30.9 69 69-30.9 69-69 69z"],deploydog:[512,512,[],"f38e","M382.2 136h51.7v239.6h-51.7v-20.7c-19.8 24.8-52.8 24.1-73.8 14.7-26.2-11.7-44.3-38.1-44.3-71.8 0-29.8 14.8-57.9 43.3-70.8 20.2-9.1 52.7-10.6 74.8 12.9V136zm-64.7 161.8c0 18.2 13.6 33.5 33.2 33.5 19.8 0 33.2-16.4 33.2-32.9 0-17.1-13.7-33.2-33.2-33.2-19.6 0-33.2 16.4-33.2 32.6zM188.5 136h51.7v239.6h-51.7v-20.7c-19.8 24.8-52.8 24.1-73.8 14.7-26.2-11.7-44.3-38.1-44.3-71.8 0-29.8 14.8-57.9 43.3-70.8 20.2-9.1 52.7-10.6 74.8 12.9V136zm-64.7 161.8c0 18.2 13.6 33.5 33.2 33.5 19.8 0 33.2-16.4 33.2-32.9 0-17.1-13.7-33.2-33.2-33.2-19.7 0-33.2 16.4-33.2 32.6zM448 96c17.5 0 32 14.4 32 32v256c0 17.5-14.4 32-32 32H64c-17.5 0-32-14.4-32-32V128c0-17.5 14.4-32 32-32h384m0-32H64C28.8 64 0 92.8 0 128v256c0 35.2 28.8 64 64 64h384c35.2 0 64-28.8 64-64V128c0-35.2-28.8-64-64-64z"],twitch:[512,512,[],"f1e8","M391.17,103.47H352.54v109.7h38.63ZM285,103H246.37V212.75H285ZM120.83,0,24.31,91.42V420.58H140.14V512l96.53-91.42h77.25L487.69,256V0ZM449.07,237.75l-77.22,73.12H294.61l-67.6,64v-64H140.14V36.58H449.07Z"],ravelry:[512,512,[],"f2d9","M498.252,234.223c-1.208-10.34-1.7-20.826-3.746-31a310.306,310.306,0,0,0-9.622-36.6,184.068,184.068,0,0,0-30.874-57.5,251.154,251.154,0,0,0-18.818-21.689,237.362,237.362,0,0,0-47.113-36.116A240.8,240.8,0,0,0,331.356,26.65c-11.018-3.1-22.272-5.431-33.515-7.615-6.78-1.314-13.749-1.667-20.627-2.482-.316-.036-.6-.358-.9-.553q-16.143.009-32.288.006c-2.41.389-4.808.925-7.236,1.15a179.331,179.331,0,0,0-34.256,7.1,221.5,221.5,0,0,0-39.768,16.355,281.385,281.385,0,0,0-38.08,24.158c-6.167,4.61-12.268,9.36-17.974,14.518C96.539,88.494,86.34,97.72,76.785,107.555a243.878,243.878,0,0,0-33.648,43.95,206.488,206.488,0,0,0-20.494,44.6,198.2,198.2,0,0,0-7.691,34.759A201.13,201.13,0,0,0,13.4,266.385a299.716,299.716,0,0,0,4.425,40.24,226.865,226.865,0,0,0,16.73,53.3,210.543,210.543,0,0,0,24,39.528,213.589,213.589,0,0,0,26.358,28.416A251.313,251.313,0,0,0,126.7,458.455a287.831,287.831,0,0,0,55.9,25.277,269.5,269.5,0,0,0,40.641,9.835c6.071,1.01,12.275,1.253,18.412,1.873a4.149,4.149,0,0,1,1.19.56h32.289c2.507-.389,5-.937,7.527-1.143,16.336-1.332,32.107-5.335,47.489-10.717A219.992,219.992,0,0,0,379.1,460.322c9.749-6.447,19.395-13.077,28.737-20.1,5.785-4.348,10.988-9.5,16.3-14.457,3.964-3.7,7.764-7.578,11.51-11.5a232.162,232.162,0,0,0,31.427-41.639c9.542-16.045,17.355-32.905,22.3-50.926,2.859-10.413,4.947-21.045,7.017-31.652,1.032-5.279,1.251-10.723,1.87-16.087.036-.317.358-.6.552-.9V236.005A9.757,9.757,0,0,1,498.252,234.223Zm-161.117-1.15s-16.572-2.98-28.47-2.98c-27.2,0-33.57,14.9-33.57,37.04V360.8H201.582V170.062H275.1v31.931c8.924-26.822,26.771-36.189,62.04-36.189Z"],mixer:[512,512,[],"e056","M114.57,76.07a45.71,45.71,0,0,0-67.51-6.41c-17.58,16.18-19,43.52-4.75,62.77l91.78,123L41.76,379.58c-14.23,19.25-13.11,46.59,4.74,62.77A45.71,45.71,0,0,0,114,435.94L242.89,262.7a12.14,12.14,0,0,0,0-14.23ZM470.24,379.58,377.91,255.45l91.78-123c14.22-19.25,12.83-46.59-4.75-62.77a45.71,45.71,0,0,0-67.51,6.41l-128,172.12a12.14,12.14,0,0,0,0,14.23L398,435.94a45.71,45.71,0,0,0,67.51,6.41C483.35,426.17,484.47,398.83,470.24,379.58Z"],"square-lastfm":[448,512,["lastfm-square"],"f203","M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM210.7 280.8c-1.8-5.5-3.4-10.8-5-15.9c-12.9-41.9-21-68.4-58-68.4c-22.4 0-45.1 16.1-45.1 61.2c0 35.2 18 57.2 43.3 57.2c28.6 0 47.6-21.3 47.6-21.3l11.7 31.9s-19.8 19.4-61.2 19.4c-51.3 0-79.9-30.1-79.9-85.8c0-57.9 28.6-92 82.5-92c67.9 0 79.3 35.3 96.4 88.4c1.4 4.4 2.9 8.9 4.4 13.5c8.8 26.8 24.2 46.2 61.2 46.2c24.9 0 38.1-5.5 38.1-19.1c0-17.5-16.9-21.2-40-26.4c-3.2-.7-6.5-1.4-9.9-2.2c-30.4-7.3-42.5-23.1-42.5-48c0-40 32.3-52.4 65.2-52.4c37.4 0 60.1 13.6 63 46.6l-36.7 4.4c-1.5-15.8-11-22.4-28.6-22.4c-16.1 0-26 7.3-26 19.8c0 11 4.8 17.6 20.9 21.3c2.2 .5 4.5 1 6.7 1.4c31.1 6.5 65.1 13.7 65.1 56.1c.1 36.7-30.7 50.6-76.1 50.6c-63.4 0-85.4-28.6-97.1-64.1z"],vimeo:[448,512,[],"f40a","M403.2 32H44.8C20.1 32 0 52.1 0 76.8v358.4C0 459.9 20.1 480 44.8 480h358.4c24.7 0 44.8-20.1 44.8-44.8V76.8c0-24.7-20.1-44.8-44.8-44.8zM377 180.8c-1.4 31.5-23.4 74.7-66 129.4-44 57.2-81.3 85.8-111.7 85.8-18.9 0-34.8-17.4-47.9-52.3-25.5-93.3-36.4-148-57.4-148-2.4 0-10.9 5.1-25.4 15.2l-15.2-19.6c37.3-32.8 72.9-69.2 95.2-71.2 25.2-2.4 40.7 14.8 46.5 51.7 20.7 131.2 29.9 151 67.6 91.6 13.5-21.4 20.8-37.7 21.8-48.9 3.5-33.2-25.9-30.9-45.8-22.4 15.9-52.1 46.3-77.4 91.2-76 33.3.9 49 22.5 47.1 64.7z"],mendeley:[640,512,[],"f7b3","M624.6 325.2c-12.3-12.4-29.7-19.2-48.4-17.2-43.3-1-49.7-34.9-37.5-98.8 22.8-57.5-14.9-131.5-87.4-130.8-77.4.7-81.7 82-130.9 82-48.1 0-54-81.3-130.9-82-72.9-.8-110.1 73.3-87.4 130.8 12.2 63.9 5.8 97.8-37.5 98.8-21.2-2.3-37 6.5-53 22.5-19.9 19.7-19.3 94.8 42.6 102.6 47.1 5.9 81.6-42.9 61.2-87.8-47.3-103.7 185.9-106.1 146.5-8.2-.1.1-.2.2-.3.4-26.8 42.8 6.8 97.4 58.8 95.2 52.1 2.1 85.4-52.6 58.8-95.2-.1-.2-.2-.3-.3-.4-39.4-97.9 193.8-95.5 146.5 8.2-4.6 10-6.7 21.3-5.7 33 4.9 53.4 68.7 74.1 104.9 35.2 17.8-14.8 23.1-65.6 0-88.3zm-303.9-19.1h-.6c-43.4 0-62.8-37.5-62.8-62.8 0-34.7 28.2-62.8 62.8-62.8h.6c34.7 0 62.8 28.1 62.8 62.8 0 25-19.2 62.8-62.8 62.8z"],uniregistry:[384,512,[],"f404","M192 480c39.5 0 76.2-11.8 106.8-32.2H85.3C115.8 468.2 152.5 480 192 480zm-89.1-193.1v-12.4H0v12.4c0 2.5 0 5 .1 7.4h103.1c-.2-2.4-.3-4.9-.3-7.4zm20.5 57H8.5c2.6 8.5 5.8 16.8 9.6 24.8h138.3c-12.9-5.7-24.1-14.2-33-24.8zm-17.7-34.7H1.3c.9 7.6 2.2 15 3.9 22.3h109.7c-4-6.9-7.2-14.4-9.2-22.3zm-2.8-69.3H0v17.3h102.9zm0-173.2H0v4.9h102.9zm0-34.7H0v2.5h102.9zm0 69.3H0v7.4h102.9zm0 104H0v14.8h102.9zm0-69.3H0v9.9h102.9zm0 34.6H0V183h102.9zm166.2 160.9h109.7c1.8-7.3 3.1-14.7 3.9-22.3H278.3c-2.1 7.9-5.2 15.4-9.2 22.3zm12-185.7H384V136H281.1zm0 37.2H384v-12.4H281.1zm0-74.3H384v-7.4H281.1zm0-76.7v2.5H384V32zm-203 410.9h227.7c11.8-8.7 22.7-18.6 32.2-29.7H44.9c9.6 11 21.4 21 33.2 29.7zm203-371.3H384v-4.9H281.1zm0 148.5H384v-14.8H281.1zM38.8 405.7h305.3c6.7-8.5 12.6-17.6 17.8-27.2H23c5.2 9.6 9.2 18.7 15.8 27.2zm188.8-37.1H367c3.7-8 5.8-16.2 8.5-24.8h-115c-8.8 10.7-20.1 19.2-32.9 24.8zm53.5-81.7c0 2.5-.1 5-.4 7.4h103.1c.1-2.5.2-4.9.2-7.4v-12.4H281.1zm0-29.7H384v-17.3H281.1z"],figma:[384,512,[],"f799","M14 95.7924C14 42.8877 56.8878 0 109.793 0H274.161C327.066 0 369.954 42.8877 369.954 95.7924C369.954 129.292 352.758 158.776 326.711 175.897C352.758 193.019 369.954 222.502 369.954 256.002C369.954 308.907 327.066 351.795 274.161 351.795H272.081C247.279 351.795 224.678 342.369 207.666 326.904V415.167C207.666 468.777 163.657 512 110.309 512C57.5361 512 14 469.243 14 416.207C14 382.709 31.1945 353.227 57.2392 336.105C31.1945 318.983 14 289.5 14 256.002C14 222.502 31.196 193.019 57.2425 175.897C31.196 158.776 14 129.292 14 95.7924ZM176.288 191.587H109.793C74.2172 191.587 45.3778 220.427 45.3778 256.002C45.3778 291.44 73.9948 320.194 109.381 320.416C109.518 320.415 109.655 320.415 109.793 320.415H176.288V191.587ZM207.666 256.002C207.666 291.577 236.505 320.417 272.081 320.417H274.161C309.737 320.417 338.576 291.577 338.576 256.002C338.576 220.427 309.737 191.587 274.161 191.587H272.081C236.505 191.587 207.666 220.427 207.666 256.002ZM109.793 351.795C109.655 351.795 109.518 351.794 109.381 351.794C73.9948 352.015 45.3778 380.769 45.3778 416.207C45.3778 451.652 74.6025 480.622 110.309 480.622C146.591 480.622 176.288 451.186 176.288 415.167V351.795H109.793ZM109.793 31.3778C74.2172 31.3778 45.3778 60.2173 45.3778 95.7924C45.3778 131.368 74.2172 160.207 109.793 160.207H176.288V31.3778H109.793ZM207.666 160.207H274.161C309.737 160.207 338.576 131.368 338.576 95.7924C338.576 60.2173 309.737 31.3778 274.161 31.3778H207.666V160.207Z"],"creative-commons-remix":[496,512,[],"f4ee","M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zm161.7 207.7l4.9 2.2v70c-7.2 3.6-63.4 27.5-67.3 28.8-6.5-1.8-113.7-46.8-137.3-56.2l-64.2 26.6-63.3-27.5v-63.8l59.3-24.8c-.7-.7-.4 5-.4-70.4l67.3-29.7L361 178.5v61.6l49.1 20.3zm-70.4 81.5v-43.8h-.4v-1.8l-113.8-46.5V295l113.8 46.9v-.4l.4.4zm7.5-57.6l39.9-16.4-36.8-15.5-39 16.4 35.9 15.5zm52.3 38.1v-43L355.2 298v43.4l44.3-19z"],"cc-amazon-pay":[576,512,[],"f42d","M124.7 201.8c.1-11.8 0-23.5 0-35.3v-35.3c0-1.3.4-2 1.4-2.7 11.5-8 24.1-12.1 38.2-11.1 12.5.9 22.7 7 28.1 21.7 3.3 8.9 4.1 18.2 4.1 27.7 0 8.7-.7 17.3-3.4 25.6-5.7 17.8-18.7 24.7-35.7 23.9-11.7-.5-21.9-5-31.4-11.7-.9-.8-1.4-1.6-1.3-2.8zm154.9 14.6c4.6 1.8 9.3 2 14.1 1.5 11.6-1.2 21.9-5.7 31.3-12.5.9-.6 1.3-1.3 1.3-2.5-.1-3.9 0-7.9 0-11.8 0-4-.1-8 0-12 0-1.4-.4-2-1.8-2.2-7-.9-13.9-2.2-20.9-2.9-7-.6-14-.3-20.8 1.9-6.7 2.2-11.7 6.2-13.7 13.1-1.6 5.4-1.6 10.8.1 16.2 1.6 5.5 5.2 9.2 10.4 11.2zM576 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h480c26.5 0 48 21.5 48 48zm-207.5 23.9c.4 1.7.9 3.4 1.6 5.1 16.5 40.6 32.9 81.3 49.5 121.9 1.4 3.5 1.7 6.4.2 9.9-2.8 6.2-4.9 12.6-7.8 18.7-2.6 5.5-6.7 9.5-12.7 11.2-4.2 1.1-8.5 1.3-12.9.9-2.1-.2-4.2-.7-6.3-.8-2.8-.2-4.2 1.1-4.3 4-.1 2.8-.1 5.6 0 8.3.1 4.6 1.6 6.7 6.2 7.5 4.7.8 9.4 1.6 14.2 1.7 14.3.3 25.7-5.4 33.1-17.9 2.9-4.9 5.6-10.1 7.7-15.4 19.8-50.1 39.5-100.3 59.2-150.5.6-1.5 1.1-3 1.3-4.6.4-2.4-.7-3.6-3.1-3.7-5.6-.1-11.1 0-16.7 0-3.1 0-5.3 1.4-6.4 4.3-.4 1.1-.9 2.3-1.3 3.4l-29.1 83.7c-2.1 6.1-4.2 12.1-6.5 18.6-.4-.9-.6-1.4-.8-1.9-10.8-29.9-21.6-59.9-32.4-89.8-1.7-4.7-3.5-9.5-5.3-14.2-.9-2.5-2.7-4-5.4-4-6.4-.1-12.8-.2-19.2-.1-2.2 0-3.3 1.6-2.8 3.7zM242.4 206c1.7 11.7 7.6 20.8 18 26.6 9.9 5.5 20.7 6.2 31.7 4.6 12.7-1.9 23.9-7.3 33.8-15.5.4-.3.8-.6 1.4-1 .5 3.2.9 6.2 1.5 9.2.5 2.6 2.1 4.3 4.5 4.4 4.6.1 9.1.1 13.7 0 2.3-.1 3.8-1.6 4-3.9.1-.8.1-1.6.1-2.3v-88.8c0-3.6-.2-7.2-.7-10.8-1.6-10.8-6.2-19.7-15.9-25.4-5.6-3.3-11.8-5-18.2-5.9-3-.4-6-.7-9.1-1.1h-10c-.8.1-1.6.3-2.5.3-8.2.4-16.3 1.4-24.2 3.5-5.1 1.3-10 3.2-15 4.9-3 1-4.5 3.2-4.4 6.5.1 2.8-.1 5.6 0 8.3.1 4.1 1.8 5.2 5.7 4.1 6.5-1.7 13.1-3.5 19.7-4.8 10.3-1.9 20.7-2.7 31.1-1.2 5.4.8 10.5 2.4 14.1 7 3.1 4 4.2 8.8 4.4 13.7.3 6.9.2 13.9.3 20.8 0 .4-.1.7-.2 1.2-.4 0-.8 0-1.1-.1-8.8-2.1-17.7-3.6-26.8-4.1-9.5-.5-18.9.1-27.9 3.2-10.8 3.8-19.5 10.3-24.6 20.8-4.1 8.3-4.6 17-3.4 25.8zM98.7 106.9v175.3c0 .8 0 1.7.1 2.5.2 2.5 1.7 4.1 4.1 4.2 5.9.1 11.8.1 17.7 0 2.5 0 4-1.7 4.1-4.1.1-.8.1-1.7.1-2.5v-60.7c.9.7 1.4 1.2 1.9 1.6 15 12.5 32.2 16.6 51.1 12.9 17.1-3.4 28.9-13.9 36.7-29.2 5.8-11.6 8.3-24.1 8.7-37 .5-14.3-1-28.4-6.8-41.7-7.1-16.4-18.9-27.3-36.7-30.9-2.7-.6-5.5-.8-8.2-1.2h-7c-1.2.2-2.4.3-3.6.5-11.7 1.4-22.3 5.8-31.8 12.7-2 1.4-3.9 3-5.9 4.5-.1-.5-.3-.8-.4-1.2-.4-2.3-.7-4.6-1.1-6.9-.6-3.9-2.5-5.5-6.4-5.6h-9.7c-5.9-.1-6.9 1-6.9 6.8zM493.6 339c-2.7-.7-5.1 0-7.6 1-43.9 18.4-89.5 30.2-136.8 35.8-14.5 1.7-29.1 2.8-43.7 3.2-26.6.7-53.2-.8-79.6-4.3-17.8-2.4-35.5-5.7-53-9.9-37-8.9-72.7-21.7-106.7-38.8-8.8-4.4-17.4-9.3-26.1-14-3.8-2.1-6.2-1.5-8.2 2.1v1.7c1.2 1.6 2.2 3.4 3.7 4.8 36 32.2 76.6 56.5 122 72.9 21.9 7.9 44.4 13.7 67.3 17.5 14 2.3 28 3.8 42.2 4.5 3 .1 6 .2 9 .4.7 0 1.4.2 2.1.3h17.7c.7-.1 1.4-.3 2.1-.3 14.9-.4 29.8-1.8 44.6-4 21.4-3.2 42.4-8.1 62.9-14.7 29.6-9.6 57.7-22.4 83.4-40.1 2.8-1.9 5.7-3.8 8-6.2 4.3-4.4 2.3-10.4-3.3-11.9zm50.4-27.7c-.8-4.2-4-5.8-7.6-7-5.7-1.9-11.6-2.8-17.6-3.3-11-.9-22-.4-32.8 1.6-12 2.2-23.4 6.1-33.5 13.1-1.2.8-2.4 1.8-3.1 3-.6.9-.7 2.3-.5 3.4.3 1.3 1.7 1.6 3 1.5.6 0 1.2 0 1.8-.1l19.5-2.1c9.6-.9 19.2-1.5 28.8-.8 4.1.3 8.1 1.2 12 2.2 4.3 1.1 6.2 4.4 6.4 8.7.3 6.7-1.2 13.1-2.9 19.5-3.5 12.9-8.3 25.4-13.3 37.8-.3.8-.7 1.7-.8 2.5-.4 2.5 1 4 3.4 3.5 1.4-.3 3-1.1 4-2.1 3.7-3.6 7.5-7.2 10.6-11.2 10.7-13.8 17-29.6 20.7-46.6.7-3 1.2-6.1 1.7-9.1.2-4.7.2-9.6.2-14.5z"],dropbox:[528,512,[],"f16b","M264.4 116.3l-132 84.3 132 84.3-132 84.3L0 284.1l132.3-84.3L0 116.3 132.3 32l132.1 84.3zM131.6 395.7l132-84.3 132 84.3-132 84.3-132-84.3zm132.8-111.6l132-84.3-132-83.6L395.7 32 528 116.3l-132.3 84.3L528 284.8l-132.3 84.3-131.3-85z"],instagram:[448,512,[],"f16d","M224.1 141c-63.6 0-114.9 51.3-114.9 114.9s51.3 114.9 114.9 114.9S339 319.5 339 255.9 287.7 141 224.1 141zm0 189.6c-41.1 0-74.7-33.5-74.7-74.7s33.5-74.7 74.7-74.7 74.7 33.5 74.7 74.7-33.6 74.7-74.7 74.7zm146.4-194.3c0 14.9-12 26.8-26.8 26.8-14.9 0-26.8-12-26.8-26.8s12-26.8 26.8-26.8 26.8 12 26.8 26.8zm76.1 27.2c-1.7-35.9-9.9-67.7-36.2-93.9-26.2-26.2-58-34.4-93.9-36.2-37-2.1-147.9-2.1-184.9 0-35.8 1.7-67.6 9.9-93.9 36.1s-34.4 58-36.2 93.9c-2.1 37-2.1 147.9 0 184.9 1.7 35.9 9.9 67.7 36.2 93.9s58 34.4 93.9 36.2c37 2.1 147.9 2.1 184.9 0 35.9-1.7 67.7-9.9 93.9-36.2 26.2-26.2 34.4-58 36.2-93.9 2.1-37 2.1-147.8 0-184.8zM398.8 388c-7.8 19.6-22.9 34.7-42.6 42.6-29.5 11.7-99.5 9-132.1 9s-102.7 2.6-132.1-9c-19.6-7.8-34.7-22.9-42.6-42.6-11.7-29.5-9-99.5-9-132.1s-2.6-102.7 9-132.1c7.8-19.6 22.9-34.7 42.6-42.6 29.5-11.7 99.5-9 132.1-9s102.7-2.6 132.1 9c19.6 7.8 34.7 22.9 42.6 42.6 11.7 29.5 9 99.5 9 132.1s2.7 102.7-9 132.1z"],cmplid:[640,512,[],"e360","M226.119,388.165a3.816,3.816,0,0,0-2.294-3.5,3.946,3.946,0,0,0-1.629-.385L72.6,384.3a19.243,19.243,0,0,1-17.924-26.025L81.585,255.692a35.72,35.72,0,0,1,32.373-26H262.525a7.07,7.07,0,0,0,6.392-5.194l10.769-41.131a3.849,3.849,0,0,0-2.237-4.937,3.755,3.755,0,0,0-1.377-.261c-.063,0-.126,0-.189.005H127.38a106.8,106.8,0,0,0-96.99,77.1L3.483,358.824A57.469,57.469,0,0,0,57.314,436q1.43,0,2.86-.072H208.742a7.131,7.131,0,0,0,6.391-5.193L225.839,389.6A3.82,3.82,0,0,0,226.119,388.165ZM306.658,81.2a3.861,3.861,0,0,0,.251-1.367A3.813,3.813,0,0,0,303.079,76c-.064,0-.128,0-.192,0h-41A7.034,7.034,0,0,0,255.5,81.2l-21.347,80.915h51.131ZM180.364,368.249H231.5L263.452,245.69H212.321ZM511.853,79.723a3.809,3.809,0,0,0-3.8-3.661c-.058,0-.137,0-.23.007h-41a7.1,7.1,0,0,0-6.584,5.129L368.91,430.634a3.54,3.54,0,0,0-.262,1.335,3.873,3.873,0,0,0,3.864,3.863c.056,0,.112,0,.169,0h41a7.068,7.068,0,0,0,6.392-5.193L511.533,81.2A3.624,3.624,0,0,0,511.853,79.723ZM324.649,384.47h-41a7.2,7.2,0,0,0-6.392,5.194L266.52,430.8a3.662,3.662,0,0,0-.268,1.374A3.783,3.783,0,0,0,270.023,436c.06,0,.166,0,.3-.012h40.905a7.036,7.036,0,0,0,6.391-5.193l10.769-41.131a3.75,3.75,0,0,0-3.445-5.208c-.108,0-.217,0-.326.014Zm311.324-308.4h-41a7.066,7.066,0,0,0-6.392,5.129l-91.46,349.436a4.073,4.073,0,0,0-.229,1.347,3.872,3.872,0,0,0,3.863,3.851c.056,0,.112,0,.169,0h40.968a7.1,7.1,0,0,0,6.392-5.193L639.68,81.2a3.624,3.624,0,0,0,.32-1.475,3.841,3.841,0,0,0-3.821-3.564c-.068,0-.137,0-.206.006ZM371.562,225.236l10.8-41.1a4.369,4.369,0,0,0,.227-1.388,3.869,3.869,0,0,0-3.861-3.842c-.057,0-.113,0-.169,0h-41.1a7.292,7.292,0,0,0-6.391,5.226l-10.834,41.1a4.417,4.417,0,0,0-.26,1.493c0,.069,0,.138,0,.206a3.776,3.776,0,0,0,3.757,3.507c.076,0,.18,0,.3-.012h41.129A7.034,7.034,0,0,0,371.562,225.236Z"],upwork:[641,512,[],"e641","M494.7 295.6c-50.3 0-83.5-38.9-92.8-53.9c11.9-95.3 46.8-125.4 92.8-125.4c45.5 0 80.9 36.4 80.9 89.7s-35.4 89.7-80.9 89.7zm0-237.8c-81.9 0-127.8 53.4-141 108.4c-14.9-28-25.9-65.5-34.5-100.3H206v141c0 51.1-23.3 89-68.8 89s-71.6-37.8-71.6-89l.5-141H.8v141c0 41.1 13.3 78.4 37.6 105.1c25 27.5 59.2 41.8 98.8 41.8c78.8 0 133.8-60.4 133.8-146.9V112.1c8.2 31.2 27.8 91.1 65.3 143.6l-35 199.4h66.4l23.1-141.3c7.6 6.3 15.7 12 24.2 17c22.2 14 47.7 21.9 73.9 22.8c0 0 4 .2 6.1 .2c81.2 0 145.9-62.9 145.9-147.8s-64.8-148.1-146-148.1z"],facebook:[512,512,[62e3],"f09a","M512 256C512 114.6 397.4 0 256 0S0 114.6 0 256C0 376 82.7 476.8 194.2 504.5V334.2H141.4V256h52.8V222.3c0-87.1 39.4-127.5 125-127.5c16.2 0 44.2 3.2 55.7 6.4V172c-6-.6-16.5-1-29.6-1c-42 0-58.2 15.9-58.2 57.2V256h83.6l-14.4 78.2H287V510.1C413.8 494.8 512 386.9 512 256h0z"],gripfire:[384,512,[],"f3ac","M112.5 301.4c0-73.8 105.1-122.5 105.1-203 0-47.1-34-88-39.1-90.4.4 3.3.6 6.7.6 10C179.1 110.1 32 171.9 32 286.6c0 49.8 32.2 79.2 66.5 108.3 65.1 46.7 78.1 71.4 78.1 86.6 0 10.1-4.8 17-4.8 22.3 13.1-16.7 17.4-31.9 17.5-46.4 0-29.6-21.7-56.3-44.2-86.5-16-22.3-32.6-42.6-32.6-69.5zm205.3-39c-12.1-66.8-78-124.4-94.7-130.9l4 7.2c2.4 5.1 3.4 10.9 3.4 17.1 0 44.7-54.2 111.2-56.6 116.7-2.2 5.1-3.2 10.5-3.2 15.8 0 20.1 15.2 42.1 17.9 42.1 2.4 0 56.6-55.4 58.1-87.7 6.4 11.7 9.1 22.6 9.1 33.4 0 41.2-41.8 96.9-41.8 96.9 0 11.6 31.9 53.2 35.5 53.2 1 0 2.2-1.4 3.2-2.4 37.9-39.3 67.3-85 67.3-136.8 0-8-.7-16.2-2.2-24.6z"],"jedi-order":[448,512,[],"f50e","M398.5 373.6c95.9-122.1 17.2-233.1 17.2-233.1 45.4 85.8-41.4 170.5-41.4 170.5 105-171.5-60.5-271.5-60.5-271.5 96.9 72.7-10.1 190.7-10.1 190.7 85.8 158.4-68.6 230.1-68.6 230.1s-.4-16.9-2.2-85.7c4.3 4.5 34.5 36.2 34.5 36.2l-24.2-47.4 62.6-9.1-62.6-9.1 20.2-55.5-31.4 45.9c-2.2-87.7-7.8-305.1-7.9-306.9v-2.4 1-1 2.4c0 1-5.6 219-7.9 306.9l-31.4-45.9 20.2 55.5-62.6 9.1 62.6 9.1-24.2 47.4 34.5-36.2c-1.8 68.8-2.2 85.7-2.2 85.7s-154.4-71.7-68.6-230.1c0 0-107-118.1-10.1-190.7 0 0-165.5 99.9-60.5 271.5 0 0-86.8-84.8-41.4-170.5 0 0-78.7 111 17.2 233.1 0 0-26.2-16.1-49.4-77.7 0 0 16.9 183.3 222 185.7h4.1c205-2.4 222-185.7 222-185.7-23.6 61.5-49.9 77.7-49.9 77.7z"],uikit:[448,512,[],"f403","M443.9 128v256L218 512 0 384V169.7l87.6 45.1v117l133.5 75.5 135.8-75.5v-151l-101.1-57.6 87.6-53.1L443.9 128zM308.6 49.1L223.8 0l-88.6 54.8 86 47.3 87.4-53z"],"fort-awesome-alt":[512,512,[],"f3a3","M208 237.4h-22.2c-2.1 0-3.7 1.6-3.7 3.7v51.7c0 2.1 1.6 3.7 3.7 3.7H208c2.1 0 3.7-1.6 3.7-3.7v-51.7c0-2.1-1.6-3.7-3.7-3.7zm118.2 0H304c-2.1 0-3.7 1.6-3.7 3.7v51.7c0 2.1 1.6 3.7 3.7 3.7h22.2c2.1 0 3.7-1.6 3.7-3.7v-51.7c-.1-2.1-1.7-3.7-3.7-3.7zm132-125.1c-2.3-3.2-4.6-6.4-7.1-9.5-9.8-12.5-20.8-24-32.8-34.4-4.5-3.9-9.1-7.6-13.9-11.2-1.6-1.2-3.2-2.3-4.8-3.5C372 34.1 340.3 20 306 13c-16.2-3.3-32.9-5-50-5s-33.9 1.7-50 5c-34.3 7.1-66 21.2-93.3 40.8-1.6 1.1-3.2 2.3-4.8 3.5-4.8 3.6-9.4 7.3-13.9 11.2-3 2.6-5.9 5.3-8.8 8s-5.7 5.5-8.4 8.4c-5.5 5.7-10.7 11.8-15.6 18-2.4 3.1-4.8 6.3-7.1 9.5C25.2 153 8.3 202.5 8.3 256c0 2 .1 4 .1 6 .1.7.1 1.3.1 2 .1 1.3.1 2.7.2 4 0 .8.1 1.5.1 2.3 0 1.3.1 2.5.2 3.7.1.8.1 1.6.2 2.4.1 1.1.2 2.3.3 3.5 0 .8.1 1.6.2 2.4.1 1.2.3 2.4.4 3.6.1.8.2 1.5.3 2.3.1 1.3.3 2.6.5 3.9.1.6.2 1.3.3 1.9l.9 5.7c.1.6.2 1.1.3 1.7.3 1.3.5 2.7.8 4 .2.8.3 1.6.5 2.4.2 1 .5 2.1.7 3.2.2.9.4 1.7.6 2.6.2 1 .4 2 .7 3 .2.9.5 1.8.7 2.7.3 1 .5 1.9.8 2.9.3.9.5 1.8.8 2.7.2.9.5 1.9.8 2.8s.5 1.8.8 2.7c.3 1 .6 1.9.9 2.8.6 1.6 1.1 3.3 1.7 4.9.4 1 .7 1.9 1 2.8.3 1 .7 2 1.1 3 .3.8.6 1.5.9 2.3l1.2 3c.3.7.6 1.5.9 2.2.4 1 .9 2 1.3 3l.9 2.1c.5 1 .9 2 1.4 3 .3.7.6 1.3.9 2 .5 1 1 2.1 1.5 3.1.2.6.5 1.1.8 1.7.6 1.1 1.1 2.2 1.7 3.3.1.2.2.3.3.5 2.2 4.1 4.4 8.2 6.8 12.2.2.4.5.8.7 1.2.7 1.1 1.3 2.2 2 3.3.3.5.6.9.9 1.4.6 1.1 1.3 2.1 2 3.2.3.5.6.9.9 1.4.7 1.1 1.4 2.1 2.1 3.2.2.4.5.8.8 1.2.7 1.1 1.5 2.2 2.3 3.3.2.2.3.5.5.7 37.5 51.7 94.4 88.5 160 99.4.9.1 1.7.3 2.6.4 1 .2 2.1.4 3.1.5s1.9.3 2.8.4c1 .2 2 .3 3 .4.9.1 1.9.2 2.9.3s1.9.2 2.9.3 2.1.2 3.1.3c.9.1 1.8.1 2.7.2 1.1.1 2.3.1 3.4.2.8 0 1.7.1 2.5.1 1.3 0 2.6.1 3.9.1.7.1 1.4.1 2.1.1 2 .1 4 .1 6 .1s4-.1 6-.1c.7 0 1.4-.1 2.1-.1 1.3 0 2.6 0 3.9-.1.8 0 1.7-.1 2.5-.1 1.1-.1 2.3-.1 3.4-.2.9 0 1.8-.1 2.7-.2 1-.1 2.1-.2 3.1-.3s1.9-.2 2.9-.3c.9-.1 1.9-.2 2.9-.3s2-.3 3-.4 1.9-.3 2.8-.4c1-.2 2.1-.3 3.1-.5.9-.1 1.7-.3 2.6-.4 65.6-11 122.5-47.7 160.1-102.4.2-.2.3-.5.5-.7.8-1.1 1.5-2.2 2.3-3.3.2-.4.5-.8.8-1.2.7-1.1 1.4-2.1 2.1-3.2.3-.5.6-.9.9-1.4.6-1.1 1.3-2.1 2-3.2.3-.5.6-.9.9-1.4.7-1.1 1.3-2.2 2-3.3.2-.4.5-.8.7-1.2 2.4-4 4.6-8.1 6.8-12.2.1-.2.2-.3.3-.5.6-1.1 1.1-2.2 1.7-3.3.2-.6.5-1.1.8-1.7.5-1 1-2.1 1.5-3.1.3-.7.6-1.3.9-2 .5-1 1-2 1.4-3l.9-2.1c.5-1 .9-2 1.3-3 .3-.7.6-1.5.9-2.2l1.2-3c.3-.8.6-1.5.9-2.3.4-1 .7-2 1.1-3s.7-1.9 1-2.8c.6-1.6 1.2-3.3 1.7-4.9.3-1 .6-1.9.9-2.8s.5-1.8.8-2.7c.2-.9.5-1.9.8-2.8s.6-1.8.8-2.7c.3-1 .5-1.9.8-2.9.2-.9.5-1.8.7-2.7.2-1 .5-2 .7-3 .2-.9.4-1.7.6-2.6.2-1 .5-2.1.7-3.2.2-.8.3-1.6.5-2.4.3-1.3.6-2.7.8-4 .1-.6.2-1.1.3-1.7l.9-5.7c.1-.6.2-1.3.3-1.9.1-1.3.3-2.6.5-3.9.1-.8.2-1.5.3-2.3.1-1.2.3-2.4.4-3.6 0-.8.1-1.6.2-2.4.1-1.1.2-2.3.3-3.5.1-.8.1-1.6.2-2.4.1 1.7.1.5.2-.7 0-.8.1-1.5.1-2.3.1-1.3.2-2.7.2-4 .1-.7.1-1.3.1-2 .1-2 .1-4 .1-6 0-53.5-16.9-103-45.8-143.7zM448 371.5c-9.4 15.5-20.6 29.9-33.6 42.9-20.6 20.6-44.5 36.7-71.2 48-13.9 5.8-28.2 10.3-42.9 13.2v-75.8c0-58.6-88.6-58.6-88.6 0v75.8c-14.7-2.9-29-7.3-42.9-13.2-26.7-11.3-50.6-27.4-71.2-48-13-13-24.2-27.4-33.6-42.9v-71.3c0-2.1 1.6-3.7 3.7-3.7h22.1c2.1 0 3.7 1.6 3.7 3.7V326h29.6V182c0-2.1 1.6-3.7 3.7-3.7h22.1c2.1 0 3.7 1.6 3.7 3.7v25.9h29.5V182c0-2.1 1.6-3.7 3.7-3.7H208c2.1 0 3.7 1.6 3.7 3.7v25.9h29.5V182c0-4.8 6.5-3.7 9.5-3.7V88.1c-4.4-2-7.4-6.7-7.4-11.5 0-16.8 25.4-16.8 25.4 0 0 4.8-3 9.4-7.4 11.5V92c6.3-1.4 12.7-2.3 19.2-2.3 9.4 0 18.4 3.5 26.3 3.5 7.2 0 15.2-3.5 19.4-3.5 2.1 0 3.7 1.6 3.7 3.7v48.4c0 5.6-18.7 6.5-22.4 6.5-8.6 0-16.6-3.5-25.4-3.5-7 0-14.1 1.2-20.8 2.8v30.7c3 0 9.5-1.1 9.5 3.7v25.9h29.5V182c0-2.1 1.6-3.7 3.7-3.7h22.2c2.1 0 3.7 1.6 3.7 3.7v25.9h29.5V182c0-2.1 1.6-3.7 3.7-3.7h22.1c2.1 0 3.7 1.6 3.7 3.7v144h29.5v-25.8c0-2.1 1.6-3.7 3.7-3.7h22.2c2.1 0 3.7 1.6 3.7 3.7z"],phabricator:[496,512,[],"f3db","M323 262.1l-.1-13s21.7-19.8 21.1-21.2l-9.5-20c-.6-1.4-29.5-.5-29.5-.5l-9.4-9.3s.2-28.5-1.2-29.1l-20.1-9.2c-1.4-.6-20.7 21-20.7 21l-13.1-.2s-20.5-21.4-21.9-20.8l-20 8.3c-1.4.5.2 28.9.2 28.9l-9.1 9.1s-29.2-.9-29.7.4l-8.1 19.8c-.6 1.4 21 21 21 21l.1 12.9s-21.7 19.8-21.1 21.2l9.5 20c.6 1.4 29.5.5 29.5.5l9.4 9.3s-.2 31.8 1.2 32.3l20.1 8.3c1.4.6 20.7-23.5 20.7-23.5l13.1.2s20.5 23.8 21.8 23.3l20-7.5c1.4-.6-.2-32.1-.2-32.1l9.1-9.1s29.2.9 29.7-.5l8.1-19.8c.7-1.1-20.9-20.7-20.9-20.7zm-44.9-8.7c.7 17.1-12.8 31.6-30.1 32.4-17.3.8-32.1-12.5-32.8-29.6-.7-17.1 12.8-31.6 30.1-32.3 17.3-.8 32.1 12.5 32.8 29.5zm201.2-37.9l-97-97-.1.1c-75.1-73.3-195.4-72.8-269.8 1.6-50.9 51-27.8 27.9-95.7 95.3-22.3 22.3-22.3 58.7 0 81 69.9 69.4 46.4 46 97.4 97l.1-.1c75.1 73.3 195.4 72.9 269.8-1.6 51-50.9 27.9-27.9 95.3-95.3 22.3-22.3 22.3-58.7 0-81zM140.4 363.8c-59.6-59.5-59.6-156 0-215.5 59.5-59.6 156-59.5 215.6 0 59.5 59.5 59.6 156 0 215.6-59.6 59.5-156 59.4-215.6-.1z"],ussunnah:[482,512,[],"f407","M481.9 268.1A240.9 240.9 0 1 1 .1 268a240.9 240.9 0 1 1 481.9 0zM24.5 268a216.5 216.5 0 1 0 432.9 0A216.5 216.5 0 1 0 24.5 268zm385.9 63.3c-12.7 0-21.6-1.9-26.7-5.9c-5.5-4.3-8.2-12.3-8.2-23.8V205.1c0-6.5-5.2-20.2-15.7-41.2c7 0 17-9.1 30-27.2V284.5c0 11 2.4 19.4 7 25.3c3.7 4.7 10.1 8.9 19 12.6c1.2 .4 2.6 .9 4.1 1.4c2.9 .9 6.3 2.1 10.3 3.5c-1.8 2.7-8.3 4-19.9 4zm-219 0c-1.3 2.4-3.6 5.5-6.8 9.4l-18.5 22.5c-1-6.1-4-13-9.3-20.6s-9.7-11.4-13.4-11.4h-8.3H53.6c3.3-5.3 4.9-8.8 4.9-10.8c0-2-.8-5.3-2.4-9.7c-1.5-4.4-2.4-8.5-2.4-12.4c0-7.4 2.1-13.9 6.3-19.3L80 253.4l-7.1-17.7L89 215.9l6.7 16.8 8-10.3c-1.8 6.4-2.6 12.3-2.6 17.7c0 4.2 2.8 13.3 8.3 27.3l16.2 40.7H135h8 .3c2.8 .4 7.7 5 14.6 13.9c1.8 2.4 4.3 5.8 7.7 10.2c1.4 1.9 2.9 3.9 4.6 6.1c1.3-2.3 2-4.6 2-7.1c0-2-1.3-6.6-4-13.4L163 304.1c-4-10.6-6.1-17.7-6.1-21.3c0-6.3 1.9-12.3 5.8-17.9c.5-.6 1-1.3 1.5-1.9c4.4-5.6 8.8-11.1 13.3-16.5c-1.1 4.6-1.7 8.7-1.7 12c0 3.7 1.7 9.9 5.1 18.8l7.9 20.4c1.9 4.7 3 8.2 3.7 10.3h17.6 8.3l-.9-2.6c-1.4-3.9-4-7-7.7-9.3l15.6-20.1 12.3 32h13.4L245 292.2c-1.5-3.9-4-7-7.7-9.3L253 262.8 270.3 308h13.4l-11.4-29.4c-1.5-3.9-4-7-7.7-9.3l15.6-20L302.6 308h10.3 8.3 7.6c1.5 0 3-1.1 4.5-3.1s2.2-4.1 2.2-6.3V205.1c0-6.5-4.5-20.3-13.7-41.2c5.4 0 14.1-9.1 26.2-27.2V300.2c0 7.2 .6 12 1.7 14.6c1.6 3.4 5.3 6.2 11.1 8.2c-3.9 5.6-8.7 8.5-14.5 8.5H321.1h-8.3H210.5h-19zM93.4 287.3c-2.7-6.7-4-11.7-4-15c-.6 1.2-2.4 3.7-5.4 7.6c-1.4 1.9-2.2 3.7-2.2 5.3c0 2.6 .8 5.7 2.2 9.3l5.6 13.9h0c5 0 9 0 11.9-.1l-8.2-20.9zm13.5-72.4c-3-5.2-7-9.3-11.9-11.9c-3.5-1.9-5.3-4.3-5.3-7.4c0-2.4 4.6-8.6 14-18.3c.2 3.8 1.9 7.6 4.9 11.2c3.1 3.6 4.6 7 4.6 10.1c0 2.6-2.1 8-6.2 16.3zm-27.6 0c-3-5.2-7-9.3-11.9-11.9c-3.5-1.9-5.3-4.3-5.3-7.4c0-2.4 4.6-8.6 14-18.3c.2 3.8 1.9 7.6 4.9 11.2c3.1 3.6 4.6 7 4.6 10.1c0 2.6-2.1 8-6.2 16.3zm87 27.5c-3-5.2-7-9.3-11.9-11.9c-3.5-1.9-5.3-4.3-5.3-7.4c0-2.4 4.6-8.6 14-18.3c.2 3.8 1.9 7.6 4.9 11.2c3.1 3.6 4.6 7 4.6 10.1c0 2.6-2.1 8-6.2 16.3z"],earlybirds:[480,512,[],"f39a","M313.2 47.5c1.2-13 21.3-14 36.6-8.7.9.3 26.2 9.7 19 15.2-27.9-7.4-56.4 18.2-55.6-6.5zm-201 6.9c30.7-8.1 62 20 61.1-7.1-1.3-14.2-23.4-15.3-40.2-9.6-1 .3-28.7 10.5-20.9 16.7zM319.4 160c-8.8 0-16 7.2-16 16s7.2 16 16 16 16-7.2 16-16-7.2-16-16-16zm-159.7 0c-8.8 0-16 7.2-16 16s7.2 16 16 16 16-7.2 16-16-7.2-16-16-16zm318.5 163.2c-9.9 24-40.7 11-63.9-1.2-13.5 69.1-58.1 111.4-126.3 124.2.3.9-2-.1 24 1 33.6 1.4 63.8-3.1 97.4-8-19.8-13.8-11.4-37.1-9.8-38.1 1.4-.9 14.7 1.7 21.6 11.5 8.6-12.5 28.4-14.8 30.2-13.6 1.6 1.1 6.6 20.9-6.9 34.6 4.7-.9 8.2-1.6 9.8-2.1 2.6-.8 17.7 11.3 3.1 13.3-14.3 2.3-22.6 5.1-47.1 10.8-45.9 10.7-85.9 11.8-117.7 12.8l1 11.6c3.8 18.1-23.4 24.3-27.6 6.2.8 17.9-27.1 21.8-28.4-1l-.5 5.3c-.7 18.4-28.4 17.9-28.3-.6-7.5 13.5-28.1 6.8-26.4-8.5l1.2-12.4c-36.7.9-59.7 3.1-61.8 3.1-20.9 0-20.9-31.6 0-31.6 2.4 0 27.7 1.3 63.2 2.8-61.1-15.5-103.7-55-114.9-118.2-25 12.8-57.5 26.8-68.2.8-10.5-25.4 21.5-42.6 66.8-73.4.7-6.6 1.6-13.3 2.7-19.8-14.4-19.6-11.6-36.3-16.1-60.4-16.8 2.4-23.2-9.1-23.6-23.1.3-7.3 2.1-14.9 2.4-15.4 1.1-1.8 10.1-2 12.7-2.6 6-31.7 50.6-33.2 90.9-34.5 19.7-21.8 45.2-41.5 80.9-48.3C203.3 29 215.2 8.5 216.2 8c1.7-.8 21.2 4.3 26.3 23.2 5.2-8.8 18.3-11.4 19.6-10.7 1.1.6 6.4 15-4.9 25.9 40.3 3.5 72.2 24.7 96 50.7 36.1 1.5 71.8 5.9 77.1 34 2.7.6 11.6.8 12.7 2.6.3.5 2.1 8.1 2.4 15.4-.5 13.9-6.8 25.4-23.6 23.1-3.2 17.3-2.7 32.9-8.7 47.7 2.4 11.7 4 23.8 4.8 36.4 37 25.4 70.3 42.5 60.3 66.9zM207.4 159.9c.9-44-37.9-42.2-78.6-40.3-21.7 1-38.9 1.9-45.5 13.9-11.4 20.9 5.9 92.9 23.2 101.2 9.8 4.7 73.4 7.9 86.3-7.1 8.2-9.4 15-49.4 14.6-67.7zm52 58.3c-4.3-12.4-6-30.1-15.3-32.7-2-.5-9-.5-11 0-10 2.8-10.8 22.1-17 37.2 15.4 0 19.3 9.7 23.7 9.7 4.3 0 6.3-11.3 19.6-14.2zm135.7-84.7c-6.6-12.1-24.8-12.9-46.5-13.9-40.2-1.9-78.2-3.8-77.3 40.3-.5 18.3 5 58.3 13.2 67.8 13 14.9 76.6 11.8 86.3 7.1 15.8-7.6 36.5-78.9 24.3-101.3z"],"trade-federation":[496,512,[],"f513","M248 8.8c-137 0-248 111-248 248s111 248 248 248 248-111 248-248-111-248-248-248zm0 482.8c-129.7 0-234.8-105.1-234.8-234.8S118.3 22 248 22s234.8 105.1 234.8 234.8S377.7 491.6 248 491.6zm155.1-328.5v-46.8H209.3V198H54.2l36.7 46h117.7v196.8h48.8V245h83.3v-47h-83.3v-34.8h145.7zm-73.3 45.1v23.9h-82.9v197.4h-26.8V232.1H96.3l-20.1-23.9h143.9v-80.6h171.8V152h-145v56.2zm-161.3-69l-12.4-20.7 2.1 23.8-23.5 5.4 23.3 5.4-2.1 24 12.3-20.5 22.2 9.5-15.7-18.1 15.8-18.1zm-29.6-19.7l9.3-11.5-12.7 5.9-8-12.4 1.7 13.9-14.3 3.8 13.7 2.7-.8 14.7 6.8-12.2 13.8 5.3zm165.4 145.2l-13.1 5.6-7.3-12.2 1.3 14.2-13.9 3.2 13.9 3.2-1.2 14.2 7.3-12.2 13.1 5.5-9.4-10.7zm106.9-77.2l-20.9 9.1-12-19.6 2.2 22.7-22.3 5.4 22.2 4.9-1.8 22.9 11.5-19.6 21.2 8.8-15.1-17zM248 29.9c-125.3 0-226.9 101.6-226.9 226.9S122.7 483.7 248 483.7s226.9-101.6 226.9-226.9S373.3 29.9 248 29.9zM342.6 196v51h-83.3v195.7h-52.7V245.9H89.9l-40-49.9h157.4v-81.6h197.8v50.7H259.4V196zM248 43.2c60.3 0 114.8 25 153.6 65.2H202.5V190H45.1C73.1 104.8 153.4 43.2 248 43.2zm0 427.1c-117.9 0-213.6-95.6-213.6-213.5 0-21.2 3.1-41.8 8.9-61.1L87.1 252h114.7v196.8h64.6V253h83.3v-62.7h-83.2v-19.2h145.6v-50.8c30.8 37 49.3 84.6 49.3 136.5.1 117.9-95.5 213.5-213.4 213.5zM178.8 275l-11-21.4 1.7 24.5-23.7 3.9 23.8 5.9-3.7 23.8 13-20.9 21.5 10.8-15.8-18.8 16.9-17.1z"],autoprefixer:[640,512,[],"f41c","M318.4 16l-161 480h77.5l25.4-81.4h119.5L405 496h77.5L318.4 16zm-40.3 341.9l41.2-130.4h1.5l40.9 130.4h-83.6zM640 405l-10-31.4L462.1 358l19.4 56.5L640 405zm-462.1-47L10 373.7 0 405l158.5 9.4 19.4-56.4z"],whatsapp:[448,512,[],"f232","M380.9 97.1C339 55.1 283.2 32 223.9 32c-122.4 0-222 99.6-222 222 0 39.1 10.2 77.3 29.6 111L0 480l117.7-30.9c32.4 17.7 68.9 27 106.1 27h.1c122.3 0 224.1-99.6 224.1-222 0-59.3-25.2-115-67.1-157zm-157 341.6c-33.2 0-65.7-8.9-94-25.7l-6.7-4-69.8 18.3L72 359.2l-4.4-7c-18.5-29.4-28.2-63.3-28.2-98.2 0-101.7 82.8-184.5 184.6-184.5 49.3 0 95.6 19.2 130.4 54.1 34.8 34.9 56.2 81.2 56.1 130.5 0 101.8-84.9 184.6-186.6 184.6zm101.2-138.2c-5.5-2.8-32.8-16.2-37.9-18-5.1-1.9-8.8-2.8-12.5 2.8-3.7 5.6-14.3 18-17.6 21.8-3.2 3.7-6.5 4.2-12 1.4-32.6-16.3-54-29.1-75.5-66-5.7-9.8 5.7-9.1 16.3-30.3 1.8-3.7.9-6.9-.5-9.7-1.4-2.8-12.5-30.1-17.1-41.2-4.5-10.8-9.1-9.3-12.5-9.5-3.2-.2-6.9-.2-10.6-.2-3.7 0-9.7 1.4-14.8 6.9-5.1 5.6-19.4 19-19.4 46.3 0 27.3 19.9 53.7 22.6 57.4 2.8 3.7 39.1 59.7 94.8 83.8 35.2 15.2 49 16.5 66.6 13.9 10.7-1.6 32.8-13.4 37.4-26.4 4.6-13 4.6-24.1 3.2-26.4-1.3-2.5-5-3.9-10.5-6.6z"],"square-upwork":[448,512,[],"e67c","M56 32l336 0c30.9 0 56 25.1 56 56l0 336c0 30.9-25.1 56-56 56L56 480c-30.9 0-56-25.1-56-56L0 88C0 57.1 25.1 32 56 32zM270.9 274.2c6.6-52.9 25.9-69.5 51.4-69.5c25.3 0 44.9 20.2 44.9 49.7s-19.7 49.7-44.9 49.7c-27.9 0-46.3-21.5-51.4-29.9zm-26.7-41.8c-8.2-15.5-14.3-36.3-19.2-55.6l-29.7 0-33.2 0 0 78.1c0 28.4-12.9 49.4-38.2 49.4s-39.8-20.9-39.8-49.3l.3-78.1-36.2 0 0 78.1c0 22.8 7.4 43.5 20.9 58.2c13.9 15.2 32.8 23.2 54.8 23.2c43.7 0 74.2-33.5 74.2-81.5l0-52.5c4.6 17.3 15.4 50.5 36.2 79.7L215 392.6l36.8 0 12.8-78.4c4.2 3.5 8.7 6.6 13.4 9.4c12.3 7.8 26.4 12.2 40.9 12.6l.1 0c.5 0 1.1 0 1.6 0c.6 0 1.1 0 1.7 0c45.1 0 80.9-34.9 80.9-81.9s-35.9-82.2-80.9-82.2c-45.4 0-70.9 29.7-78.1 60.1z"],slideshare:[512,512,[],"f1e7","M187.7 153.7c-34 0-61.7 25.7-61.7 57.7 0 31.7 27.7 57.7 61.7 57.7s61.7-26 61.7-57.7c0-32-27.7-57.7-61.7-57.7zm143.4 0c-34 0-61.7 25.7-61.7 57.7 0 31.7 27.7 57.7 61.7 57.7 34.3 0 61.7-26 61.7-57.7.1-32-27.4-57.7-61.7-57.7zm156.6 90l-6 4.3V49.7c0-27.4-20.6-49.7-46-49.7H76.6c-25.4 0-46 22.3-46 49.7V248c-2-1.4-4.3-2.9-6.3-4.3-15.1-10.6-25.1 4-16 17.7 18.3 22.6 53.1 50.3 106.3 72C58.3 525.1 252 555.7 248.9 457.5c0-.7.3-56.6.3-96.6 5.1 1.1 9.4 2.3 13.7 3.1 0 39.7.3 92.8.3 93.5-3.1 98.3 190.6 67.7 134.3-124 53.1-21.7 88-49.4 106.3-72 9.1-13.8-.9-28.3-16.1-17.8zm-30.5 19.2c-68.9 37.4-128.3 31.1-160.6 29.7-23.7-.9-32.6 9.1-33.7 24.9-10.3-7.7-18.6-15.5-20.3-17.1-5.1-5.4-13.7-8-27.1-7.7-31.7 1.1-89.7 7.4-157.4-28V72.3c0-34.9 8.9-45.7 40.6-45.7h317.7c30.3 0 40.9 12.9 40.9 45.7v190.6z"],"google-play":[512,512,[],"f3ab","M325.3 234.3L104.6 13l280.8 161.2-60.1 60.1zM47 0C34 6.8 25.3 19.2 25.3 35.3v441.3c0 16.1 8.7 28.5 21.7 35.3l256.6-256L47 0zm425.2 225.6l-58.9-34.1-65.7 64.5 65.7 64.5 60.1-34.1c18-14.3 18-46.5-1.2-60.8zM104.6 499l280.8-161.2-60.1-60.1L104.6 499z"],viadeo:[448,512,[],"f2a9","M276.2 150.5v.7C258.3 98.6 233.6 47.8 205.4 0c43.3 29.2 67 100 70.8 150.5zm32.7 121.7c7.6 18.2 11 37.5 11 57 0 77.7-57.8 141-137.8 139.4l3.8-.3c74.2-46.7 109.3-118.6 109.3-205.1 0-38.1-6.5-75.9-18.9-112 1 11.7 1 23.7 1 35.4 0 91.8-18.1 241.6-116.6 280C95 455.2 49.4 398 49.4 329.2c0-75.6 57.4-142.3 135.4-142.3 16.8 0 33.7 3.1 49.1 9.6 1.7-15.1 6.5-29.9 13.4-43.3-19.9-7.2-41.2-10.7-62.5-10.7-161.5 0-238.7 195.9-129.9 313.7 67.9 74.6 192 73.9 259.8 0 56.6-61.3 60.9-142.4 36.4-201-12.7 8-27.1 13.9-42.2 17zM418.1 11.7c-31 66.5-81.3 47.2-115.8 80.1-12.4 12-20.6 34-20.6 50.5 0 14.1 4.5 27.1 12 38.8 47.4-11 98.3-46 118.2-90.7-.7 5.5-4.8 14.4-7.2 19.2-20.3 35.7-64.6 65.6-99.7 84.9 14.8 14.4 33.7 25.8 55 25.8 79 0 110.1-134.6 58.1-208.6z"],line:[512,512,[],"f3c0","M311 196.8v81.3c0 2.1-1.6 3.7-3.7 3.7h-13c-1.3 0-2.4-.7-3-1.5l-37.3-50.3v48.2c0 2.1-1.6 3.7-3.7 3.7h-13c-2.1 0-3.7-1.6-3.7-3.7V196.9c0-2.1 1.6-3.7 3.7-3.7h12.9c1.1 0 2.4 .6 3 1.6l37.3 50.3V196.9c0-2.1 1.6-3.7 3.7-3.7h13c2.1-.1 3.8 1.6 3.8 3.5zm-93.7-3.7h-13c-2.1 0-3.7 1.6-3.7 3.7v81.3c0 2.1 1.6 3.7 3.7 3.7h13c2.1 0 3.7-1.6 3.7-3.7V196.8c0-1.9-1.6-3.7-3.7-3.7zm-31.4 68.1H150.3V196.8c0-2.1-1.6-3.7-3.7-3.7h-13c-2.1 0-3.7 1.6-3.7 3.7v81.3c0 1 .3 1.8 1 2.5c.7 .6 1.5 1 2.5 1h52.2c2.1 0 3.7-1.6 3.7-3.7v-13c0-1.9-1.6-3.7-3.5-3.7zm193.7-68.1H327.3c-1.9 0-3.7 1.6-3.7 3.7v81.3c0 1.9 1.6 3.7 3.7 3.7h52.2c2.1 0 3.7-1.6 3.7-3.7V265c0-2.1-1.6-3.7-3.7-3.7H344V247.7h35.5c2.1 0 3.7-1.6 3.7-3.7V230.9c0-2.1-1.6-3.7-3.7-3.7H344V213.5h35.5c2.1 0 3.7-1.6 3.7-3.7v-13c-.1-1.9-1.7-3.7-3.7-3.7zM512 93.4V419.4c-.1 51.2-42.1 92.7-93.4 92.6H92.6C41.4 511.9-.1 469.8 0 418.6V92.6C.1 41.4 42.2-.1 93.4 0H419.4c51.2 .1 92.7 42.1 92.6 93.4zM441.6 233.5c0-83.4-83.7-151.3-186.4-151.3s-186.4 67.9-186.4 151.3c0 74.7 66.3 137.4 155.9 149.3c21.8 4.7 19.3 12.7 14.4 42.1c-.8 4.7-3.8 18.4 16.1 10.1s107.3-63.2 146.5-108.2c27-29.7 39.9-59.8 39.9-93.1z"],"google-drive":[512,512,[],"f3aa","M339 314.9L175.4 32h161.2l163.6 282.9H339zm-137.5 23.6L120.9 480h310.5L512 338.5H201.5zM154.1 67.4L0 338.5 80.6 480 237 208.8 154.1 67.4z"],servicestack:[496,512,[],"f3ec","M88 216c81.7 10.2 273.7 102.3 304 232H0c99.5-8.1 184.5-137 88-232zm32-152c32.3 35.6 47.7 83.9 46.4 133.6C249.3 231.3 373.7 321.3 400 448h96C455.3 231.9 222.8 79.5 120 64z"],simplybuilt:[512,512,[],"f215","M481.2 64h-106c-14.5 0-26.6 11.8-26.6 26.3v39.6H163.3V90.3c0-14.5-12-26.3-26.6-26.3h-106C16.1 64 4.3 75.8 4.3 90.3v331.4c0 14.5 11.8 26.3 26.6 26.3h450.4c14.8 0 26.6-11.8 26.6-26.3V90.3c-.2-14.5-12-26.3-26.7-26.3zM149.8 355.8c-36.6 0-66.4-29.7-66.4-66.4 0-36.9 29.7-66.6 66.4-66.6 36.9 0 66.6 29.7 66.6 66.6 0 36.7-29.7 66.4-66.6 66.4zm212.4 0c-36.9 0-66.6-29.7-66.6-66.6 0-36.6 29.7-66.4 66.6-66.4 36.6 0 66.4 29.7 66.4 66.4 0 36.9-29.8 66.6-66.4 66.6z"],bitbucket:[512,512,[61810],"f171","M22.2 32A16 16 0 0 0 6 47.8a26.35 26.35 0 0 0 .2 2.8l67.9 412.1a21.77 21.77 0 0 0 21.3 18.2h325.7a16 16 0 0 0 16-13.4L505 50.7a16 16 0 0 0-13.2-18.3 24.58 24.58 0 0 0-2.8-.2L22.2 32zm285.9 297.8h-104l-28.1-147h157.3l-25.2 147z"],imdb:[448,512,[],"f2d8","M89.5 323.6H53.93V186.2H89.5V323.6zM156.1 250.5L165.2 186.2H211.5V323.6H180.5V230.9L167.1 323.6H145.8L132.8 232.9L132.7 323.6H101.5V186.2H147.6C148.1 194.5 150.4 204.3 151.9 215.6L156.1 250.5zM223.7 323.6V186.2H250.3C267.3 186.2 277.3 187.1 283.3 188.6C289.4 190.3 294 192.8 297.2 196.5C300.3 199.8 302.3 203.1 303 208.5C303.9 212.9 304.4 221.6 304.4 234.7V282.9C304.4 295.2 303.7 303.4 302.5 307.6C301.4 311.7 299.4 315 296.5 317.3C293.7 319.7 290.1 321.4 285.8 322.3C281.6 323.1 275.2 323.6 266.7 323.6H223.7zM259.2 209.7V299.1C264.3 299.1 267.5 298.1 268.6 296.8C269.7 294.8 270.4 289.2 270.4 280.1V226.8C270.4 220.6 270.3 216.6 269.7 214.8C269.4 213 268.5 211.8 267.1 210.1C265.7 210.1 263 209.7 259.2 209.7V209.7zM316.5 323.6V186.2H350.6V230.1C353.5 227.7 356.7 225.2 360.1 223.5C363.7 222 368.9 221.1 372.9 221.1C377.7 221.1 381.8 221.9 385.2 223.3C388.6 224.8 391.2 226.8 393.2 229.5C394.9 232.1 395.9 234.8 396.3 237.3C396.7 239.9 396.1 245.3 396.1 253.5V292.1C396.1 300.3 396.3 306.4 395.3 310.5C394.2 314.5 391.5 318.1 387.5 320.1C383.4 324 378.6 325.4 372.9 325.4C368.9 325.4 363.7 324.5 360.2 322.9C356.7 321.1 353.5 318.4 350.6 314.9L348.5 323.6L316.5 323.6zM361.6 302.9C362.3 301.1 362.6 296.9 362.6 290.4V255C362.6 249.4 362.3 245.5 361.5 243.8C360.8 241.9 357.8 241.1 355.7 241.1C353.7 241.1 352.3 241.9 351.6 243.4C351 244.9 350.6 248.8 350.6 255V291.4C350.6 297.5 351 301.4 351.8 303C352.4 304.7 353.9 305.5 355.9 305.5C358.1 305.5 360.1 304.7 361.6 302.9L361.6 302.9zM418.4 32.04C434.1 33.27 447.1 47.28 447.1 63.92V448.1C447.1 464.5 435.2 478.5 418.9 479.1C418.6 479.1 418.4 480 418.1 480H29.88C29.6 480 29.32 479.1 29.04 479.9C13.31 478.5 1.093 466.1 0 449.7L.0186 61.78C1.081 45.88 13.82 33.09 30.26 31.1H417.7C417.9 31.1 418.2 32.01 418.4 32.04L418.4 32.04zM30.27 41.26C19 42.01 10.02 51.01 9.257 62.4V449.7C9.63 455.1 11.91 460.2 15.7 464C19.48 467.9 24.51 470.3 29.89 470.7H418.1C429.6 469.7 438.7 459.1 438.7 448.1V63.91C438.7 58.17 436.6 52.65 432.7 48.45C428.8 44.24 423.4 41.67 417.7 41.26L30.27 41.26z"],deezer:[576,512,[],"e077","M451.46,244.71H576V172H451.46Zm0-173.89v72.67H576V70.82Zm0,275.06H576V273.2H451.46ZM0,447.09H124.54V374.42H0Zm150.47,0H275V374.42H150.47Zm150.52,0H425.53V374.42H301Zm150.47,0H576V374.42H451.46ZM301,345.88H425.53V273.2H301Zm-150.52,0H275V273.2H150.47Zm0-101.17H275V172H150.47Z"],"raspberry-pi":[407,512,[],"f7bb","M372 232.5l-3.7-6.5c.1-46.4-21.4-65.3-46.5-79.7 7.6-2 15.4-3.6 17.6-13.2 13.1-3.3 15.8-9.4 17.1-15.8 3.4-2.3 14.8-8.7 13.6-19.7 6.4-4.4 10-10.1 8.1-18.1 6.9-7.5 8.7-13.7 5.8-19.4 8.3-10.3 4.6-15.6 1.1-20.9 6.2-11.2.7-23.2-16.6-21.2-6.9-10.1-21.9-7.8-24.2-7.8-2.6-3.2-6-6-16.5-4.7-6.8-6.1-14.4-5-22.3-2.1-9.3-7.3-15.5-1.4-22.6.8C271.6.6 269 5.5 263.5 7.6c-12.3-2.6-16.1 3-22 8.9l-6.9-.1c-18.6 10.8-27.8 32.8-31.1 44.1-3.3-11.3-12.5-33.3-31.1-44.1l-6.9.1c-5.9-5.9-9.7-11.5-22-8.9-5.6-2-8.1-7-19.4-3.4-4.6-1.4-8.9-4.4-13.9-4.3-2.6.1-5.5 1-8.7 3.5-7.9-3-15.5-4-22.3 2.1-10.5-1.3-14 1.4-16.5 4.7-2.3 0-17.3-2.3-24.2 7.8C21.2 16 15.8 28 22 39.2c-3.5 5.4-7.2 10.7 1.1 20.9-2.9 5.7-1.1 11.9 5.8 19.4-1.8 8 1.8 13.7 8.1 18.1-1.2 11 10.2 17.4 13.6 19.7 1.3 6.4 4 12.4 17.1 15.8 2.2 9.5 10 11.2 17.6 13.2-25.1 14.4-46.6 33.3-46.5 79.7l-3.7 6.5c-28.8 17.2-54.7 72.7-14.2 117.7 2.6 14.1 7.1 24.2 11 35.4 5.9 45.2 44.5 66.3 54.6 68.8 14.9 11.2 30.8 21.8 52.2 29.2C159 504.2 181 512 203 512h1c22.1 0 44-7.8 64.2-28.4 21.5-7.4 37.3-18 52.2-29.2 10.2-2.5 48.7-23.6 54.6-68.8 3.9-11.2 8.4-21.3 11-35.4 40.6-45.1 14.7-100.5-14-117.7zm-22.2-8c-1.5 18.7-98.9-65.1-82.1-67.9 45.7-7.5 83.6 19.2 82.1 67.9zm-43 93.1c-24.5 15.8-59.8 5.6-78.8-22.8s-14.6-64.2 9.9-80c24.5-15.8 59.8-5.6 78.8 22.8s14.6 64.2-9.9 80zM238.9 29.3c.8 4.2 1.8 6.8 2.9 7.6 5.4-5.8 9.8-11.7 16.8-17.3 0 3.3-1.7 6.8 2.5 9.4 3.7-5 8.8-9.5 15.5-13.3-3.2 5.6-.6 7.3 1.2 9.6 5.1-4.4 10-8.8 19.4-12.3-2.6 3.1-6.2 6.2-2.4 9.8 5.3-3.3 10.6-6.6 23.1-8.9-2.8 3.1-8.7 6.3-5.1 9.4 6.6-2.5 14-4.4 22.1-5.4-3.9 3.2-7.1 6.3-3.9 8.8 7.1-2.2 16.9-5.1 26.4-2.6l-6 6.1c-.7.8 14.1.6 23.9.8-3.6 5-7.2 9.7-9.3 18.2 1 1 5.8.4 10.4 0-4.7 9.9-12.8 12.3-14.7 16.6 2.9 2.2 6.8 1.6 11.2.1-3.4 6.9-10.4 11.7-16 17.3 1.4 1 3.9 1.6 9.7.9-5.2 5.5-11.4 10.5-18.8 15 1.3 1.5 5.8 1.5 10 1.6-6.7 6.5-15.3 9.9-23.4 14.2 4 2.7 6.9 2.1 10 2.1-5.7 4.7-15.4 7.1-24.4 10 1.7 2.7 3.4 3.4 7.1 4.1-9.5 5.3-23.2 2.9-27 5.6.9 2.7 3.6 4.4 6.7 5.8-15.4.9-57.3-.6-65.4-32.3 15.7-17.3 44.4-37.5 93.7-62.6-38.4 12.8-73 30-102 53.5-34.3-15.9-10.8-55.9 5.8-71.8zm-34.4 114.6c24.2-.3 54.1 17.8 54 34.7-.1 15-21 27.1-53.8 26.9-32.1-.4-53.7-15.2-53.6-29.8 0-11.9 26.2-32.5 53.4-31.8zm-123-12.8c3.7-.7 5.4-1.5 7.1-4.1-9-2.8-18.7-5.3-24.4-10 3.1 0 6 .7 10-2.1-8.1-4.3-16.7-7.7-23.4-14.2 4.2-.1 8.7 0 10-1.6-7.4-4.5-13.6-9.5-18.8-15 5.8.7 8.3.1 9.7-.9-5.6-5.6-12.7-10.4-16-17.3 4.3 1.5 8.3 2 11.2-.1-1.9-4.2-10-6.7-14.7-16.6 4.6.4 9.4 1 10.4 0-2.1-8.5-5.8-13.3-9.3-18.2 9.8-.1 24.6 0 23.9-.8l-6-6.1c9.5-2.5 19.3.4 26.4 2.6 3.2-2.5-.1-5.6-3.9-8.8 8.1 1.1 15.4 2.9 22.1 5.4 3.5-3.1-2.3-6.3-5.1-9.4 12.5 2.3 17.8 5.6 23.1 8.9 3.8-3.6.2-6.7-2.4-9.8 9.4 3.4 14.3 7.9 19.4 12.3 1.7-2.3 4.4-4 1.2-9.6 6.7 3.8 11.8 8.3 15.5 13.3 4.1-2.6 2.5-6.2 2.5-9.4 7 5.6 11.4 11.5 16.8 17.3 1.1-.8 2-3.4 2.9-7.6 16.6 15.9 40.1 55.9 6 71.8-29-23.5-63.6-40.7-102-53.5 49.3 25 78 45.3 93.7 62.6-8 31.8-50 33.2-65.4 32.3 3.1-1.4 5.8-3.2 6.7-5.8-4-2.8-17.6-.4-27.2-5.6zm60.1 24.1c16.8 2.8-80.6 86.5-82.1 67.9-1.5-48.7 36.5-75.5 82.1-67.9zM38.2 342c-23.7-18.8-31.3-73.7 12.6-98.3 26.5-7 9 107.8-12.6 98.3zm91 98.2c-13.3 7.9-45.8 4.7-68.8-27.9-15.5-27.4-13.5-55.2-2.6-63.4 16.3-9.8 41.5 3.4 60.9 25.6 16.9 20 24.6 55.3 10.5 65.7zm-26.4-119.7c-24.5-15.8-28.9-51.6-9.9-80s54.3-38.6 78.8-22.8 28.9 51.6 9.9 80c-19.1 28.4-54.4 38.6-78.8 22.8zM205 496c-29.4 1.2-58.2-23.7-57.8-32.3-.4-12.7 35.8-22.6 59.3-22 23.7-1 55.6 7.5 55.7 18.9.5 11-28.8 35.9-57.2 35.4zm58.9-124.9c.2 29.7-26.2 53.8-58.8 54-32.6.2-59.2-23.8-59.4-53.4v-.6c-.2-29.7 26.2-53.8 58.8-54 32.6-.2 59.2 23.8 59.4 53.4v.6zm82.2 42.7c-25.3 34.6-59.6 35.9-72.3 26.3-13.3-12.4-3.2-50.9 15.1-72 20.9-23.3 43.3-38.5 58.9-26.6 10.5 10.3 16.7 49.1-1.7 72.3zm22.9-73.2c-21.5 9.4-39-105.3-12.6-98.3 43.9 24.7 36.3 79.6 12.6 98.3z"],jira:[496,512,[],"f7b1","M490 241.7C417.1 169 320.6 71.8 248.5 0 83 164.9 6 241.7 6 241.7c-7.9 7.9-7.9 20.7 0 28.7C138.8 402.7 67.8 331.9 248.5 512c379.4-378 15.7-16.7 241.5-241.7 8-7.9 8-20.7 0-28.6zm-241.5 90l-76-75.7 76-75.7 76 75.7-76 75.7z"],docker:[640,512,[],"f395","M349.9 236.3h-66.1v-59.4h66.1v59.4zm0-204.3h-66.1v60.7h66.1V32zm78.2 144.8H362v59.4h66.1v-59.4zm-156.3-72.1h-66.1v60.1h66.1v-60.1zm78.1 0h-66.1v60.1h66.1v-60.1zm276.8 100c-14.4-9.7-47.6-13.2-73.1-8.4-3.3-24-16.7-44.9-41.1-63.7l-14-9.3-9.3 14c-18.4 27.8-23.4 73.6-3.7 103.8-8.7 4.7-25.8 11.1-48.4 10.7H2.4c-8.7 50.8 5.8 116.8 44 162.1 37.1 43.9 92.7 66.2 165.4 66.2 157.4 0 273.9-72.5 328.4-204.2 21.4.4 67.6.1 91.3-45.2 1.5-2.5 6.6-13.2 8.5-17.1l-13.3-8.9zm-511.1-27.9h-66v59.4h66.1v-59.4zm78.1 0h-66.1v59.4h66.1v-59.4zm78.1 0h-66.1v59.4h66.1v-59.4zm-78.1-72.1h-66.1v60.1h66.1v-60.1z"],screenpal:[512,512,[],"e570","M233.5 22.49C233.5 10.07 243.6 0 256 0C268.4 0 278.5 10.07 278.5 22.49C278.5 34.91 268.4 44.98 256 44.98C243.6 44.98 233.5 34.91 233.5 22.49zM313.4 259C313.4 290.7 287.7 316.4 256 316.4C224.3 316.4 198.6 290.7 198.6 259C198.6 227.3 224.3 201.6 256 201.6C287.7 201.6 313.4 227.3 313.4 259zM337.2 350C359.5 330.1 373.7 302.7 377.1 273H496.6C493.1 334.4 466.2 392.2 421.4 434.4C376.7 476.6 317.5 500.2 256 500.2C194.5 500.2 135.3 476.6 90.56 434.4C45.83 392.2 18.94 334.4 15.39 273H135.1C138.5 302.7 152.7 330.1 175 350C197.3 369.9 226.2 380.9 256.1 380.9C285.1 380.9 314.8 369.9 337.2 350zM73.14 140.3C73.54 152.7 63.81 163.1 51.39 163.5C38.97 163.9 28.59 154.2 28.18 141.8C27.78 129.3 37.52 118.9 49.94 118.5C62.35 118.1 72.74 127.9 73.14 140.3zM438.9 141C438.9 128.6 448.9 118.5 461.4 118.5C473.8 118.5 483.8 128.6 483.8 141C483.8 153.5 473.8 163.5 461.4 163.5C448.9 163.5 438.9 153.5 438.9 141zM317.9 95.27C300.6 109.1 278.7 118.1 256 118.1C233.3 118.1 211.4 109.1 194.1 95.27C176.8 80.55 165.3 60.18 161.7 37.78C176.8 31.37 192.5 26.52 208.6 23.31C208.6 35.88 213.6 47.93 222.5 56.82C231.4 65.7 243.4 70.7 256 70.7C268.6 70.7 280.6 65.7 289.5 56.82C298.4 47.93 303.4 35.88 303.4 23.31C319.5 26.52 335.2 31.37 350.3 37.78C346.7 60.18 335.2 80.55 317.9 95.27H317.9zM82.78 231C61.42 238.6 38.06 238.4 16.86 230.4C18.82 214.1 22.46 198.1 27.71 182.5C33.1 185.6 39.05 187.6 45.22 188.5C51.39 189.3 57.67 188.9 63.68 187.3C69.69 185.6 75.33 182.9 80.27 179.1C85.21 175.3 89.36 170.6 92.47 165.2C95.58 159.8 97.61 153.8 98.42 147.7C99.23 141.5 98.83 135.2 97.22 129.2C95.61 123.2 92.83 117.6 89.04 112.6C85.25 107.7 80.53 103.5 75.14 100.4C85.96 88.11 98.01 76.94 111.1 67.07C128.7 81.42 140.6 101.6 144.7 123.9C148.8 146.2 144.8 169.3 133.5 188.9C122.1 208.5 104.1 223.4 82.78 231V231zM429.2 231.1C407.9 223.5 389.9 208.5 378.5 188.9C367.2 169.3 363.3 146.2 367.4 123.9C371.5 101.7 383.4 81.54 400.9 67.19C414 77.04 426.1 88.21 436.9 100.5C426.2 106.9 418.5 117.2 415.4 129.3C412.2 141.3 413.1 154.1 420.2 164.9C426.4 175.7 436.6 183.6 448.6 186.9C460.6 190.2 473.5 188.6 484.3 182.6C489.6 198.1 493.2 214.2 495.2 230.4C473.1 238.5 450.6 238.7 429.2 231.1L429.2 231.1z"],bluetooth:[448,512,[],"f293","M292.6 171.1L249.7 214l-.3-86 43.2 43.1m-43.2 219.8l43.1-43.1-42.9-42.9-.2 86zM416 259.4C416 465 344.1 512 230.9 512S32 465 32 259.4 115.4 0 228.6 0 416 53.9 416 259.4zm-158.5 0l79.4-88.6L211.8 36.5v176.9L138 139.6l-27 26.9 92.7 93-92.7 93 26.9 26.9 73.8-73.8 2.3 170 127.4-127.5-83.9-88.7z"],gitter:[384,512,[],"f426","M66.4 322.5H16V0h50.4v322.5zM166.9 76.1h-50.4V512h50.4V76.1zm100.6 0h-50.4V512h50.4V76.1zM368 76h-50.4v247H368V76z"],"d-and-d":[576,512,[],"f38d","M82.5 98.9c-.6-17.2 2-33.8 12.7-48.2.3 7.4 1.2 14.5 4.2 21.6 5.9-27.5 19.7-49.3 42.3-65.5-1.9 5.9-3.5 11.8-3 17.7 8.7-7.4 18.8-17.8 44.4-22.7 14.7-2.8 29.7-2 42.1 1 38.5 9.3 61 34.3 69.7 72.3 5.3 23.1.7 45-8.3 66.4-5.2 12.4-12 24.4-20.7 35.1-2-1.9-3.9-3.8-5.8-5.6-42.8-40.8-26.8-25.2-37.4-37.4-1.1-1.2-1-2.2-.1-3.6 8.3-13.5 11.8-28.2 10-44-1.1-9.8-4.3-18.9-11.3-26.2-14.5-15.3-39.2-15-53.5.6-11.4 12.5-14.1 27.4-10.9 43.6.2 1.3.4 2.7 0 3.9-3.4 13.7-4.6 27.6-2.5 41.6.1.5.1 1.1.1 1.6 0 .3-.1.5-.2 1.1-21.8-11-36-28.3-43.2-52.2-8.3 17.8-11.1 35.5-6.6 54.1-15.6-15.2-21.3-34.3-22-55.2zm469.6 123.2c-11.6-11.6-25-20.4-40.1-26.6-12.8-5.2-26-7.9-39.9-7.1-10 .6-19.6 3.1-29 6.4-2.5.9-5.1 1.6-7.7 2.2-4.9 1.2-7.3-3.1-4.7-6.8 3.2-4.6 3.4-4.2 15-12 .6-.4 1.2-.8 2.2-1.5h-2.5c-.6 0-1.2.2-1.9.3-19.3 3.3-30.7 15.5-48.9 29.6-10.4 8.1-13.8 3.8-12-.5 1.4-3.5 3.3-6.7 5.1-10 1-1.8 2.3-3.4 3.5-5.1-.2-.2-.5-.3-.7-.5-27 18.3-46.7 42.4-57.7 73.3.3.3.7.6 1 .9.3-.6.5-1.2.9-1.7 10.4-12.1 22.8-21.8 36.6-29.8 18.2-10.6 37.5-18.3 58.7-20.2 4.3-.4 8.7-.1 13.1-.1-1.8.7-3.5.9-5.3 1.1-18.5 2.4-35.5 9-51.5 18.5-30.2 17.9-54.5 42.2-75.1 70.4-.3.4-.4.9-.7 1.3 14.5 5.3 24 17.3 36.1 25.6.2-.1.3-.2.4-.4l1.2-2.7c12.2-26.9 27-52.3 46.7-74.5 16.7-18.8 38-25.3 62.5-20 5.9 1.3 11.4 4.4 17.2 6.8 2.3-1.4 5.1-3.2 8-4.7 8.4-4.3 17.4-7 26.7-9 14.7-3.1 29.5-4.9 44.5-1.3v-.5c-.5-.4-1.2-.8-1.7-1.4zM316.7 397.6c-39.4-33-22.8-19.5-42.7-35.6-.8.9 0-.2-1.9 3-11.2 19.1-25.5 35.3-44 47.6-10.3 6.8-21.5 11.8-34.1 11.8-21.6 0-38.2-9.5-49.4-27.8-12-19.5-13.3-40.7-8.2-62.6 7.8-33.8 30.1-55.2 38.6-64.3-18.7-6.2-33 1.7-46.4 13.9.8-13.9 4.3-26.2 11.8-37.3-24.3 10.6-45.9 25-64.8 43.9-.3-5.8 5.4-43.7 5.6-44.7.3-2.7-.6-5.3-3-7.4-24.2 24.7-44.5 51.8-56.1 84.6 7.4-5.9 14.9-11.4 23.6-16.2-8.3 22.3-19.6 52.8-7.8 101.1 4.6 19 11.9 36.8 24.1 52.3 2.9 3.7 6.3 6.9 9.5 10.3.2-.2.4-.3.6-.5-1.4-7-2.2-14.1-1.5-21.9 2.2 3.2 3.9 6 5.9 8.6 12.6 16 28.7 27.4 47.2 35.6 25 11.3 51.1 13.3 77.9 8.6 54.9-9.7 90.7-48.6 116-98.8 1-1.8.6-2.9-.9-4.2zm172-46.4c-9.5-3.1-22.2-4.2-28.7-2.9 9.9 4 14.1 6.6 18.8 12 12.6 14.4 10.4 34.7-5.4 45.6-11.7 8.1-24.9 10.5-38.9 9.1-1.2-.1-2.3-.4-3-.6 2.8-3.7 6-7 8.1-10.8 9.4-16.8 5.4-42.1-8.7-56.1-2.1-2.1-4.6-3.9-7-5.9-.3 1.3-.1 2.1.1 2.8 4.2 16.6-8.1 32.4-24.8 31.8-7.6-.3-13.9-3.8-19.6-8.5-19.5-16.1-39.1-32.1-58.5-48.3-5.9-4.9-12.5-8.1-20.1-8.7-4.6-.4-9.3-.6-13.9-.9-5.9-.4-8.8-2.8-10.4-8.4-.9-3.4-1.5-6.8-2.2-10.2-1.5-8.1-6.2-13-14.3-14.2-4.4-.7-8.9-1-13.3-1.5-13-1.4-19.8-7.4-22.6-20.3-5 11-1.6 22.4 7.3 29.9 4.5 3.8 9.3 7.3 13.8 11.2 4.6 3.8 7.4 8.7 7.9 14.8.4 4.7.8 9.5 1.8 14.1 2.2 10.6 8.9 18.4 17 25.1 16.5 13.7 33 27.3 49.5 41.1 17.9 15 13.9 32.8 13 56-.9 22.9 12.2 42.9 33.5 51.2 1 .4 2 .6 3.6 1.1-15.7-18.2-10.1-44.1.7-52.3.3 2.2.4 4.3.9 6.4 9.4 44.1 45.4 64.2 85 56.9 16-2.9 30.6-8.9 42.9-19.8 2-1.8 3.7-4.1 5.9-6.5-19.3 4.6-35.8.1-50.9-10.6.7-.3 1.3-.3 1.9-.3 21.3 1.8 40.6-3.4 57-17.4 19.5-16.6 26.6-42.9 17.4-66-8.3-20.1-23.6-32.3-43.8-38.9zM99.4 179.3c-5.3-9.2-13.2-15.6-22.1-21.3 13.7-.5 26.6.2 39.6 3.7-7-12.2-8.5-24.7-5-38.7 5.3 11.9 13.7 20.1 23.6 26.8 19.7 13.2 35.7 19.6 46.7 30.2 3.4 3.3 6.3 7.1 9.6 10.9-.8-2.1-1.4-4.1-2.2-6-5-10.6-13-18.6-22.6-25-1.8-1.2-2.8-2.5-3.4-4.5-3.3-12.5-3-25.1-.7-37.6 1-5.5 2.8-10.9 4.5-16.3.8-2.4 2.3-4.6 4-6.6.6 6.9 0 25.5 19.6 46 10.8 11.3 22.4 21.9 33.9 32.7 9 8.5 18.3 16.7 25.5 26.8 1.1 1.6 2.2 3.3 3.8 4.7-5-13-14.2-24.1-24.2-33.8-9.6-9.3-19.4-18.4-29.2-27.4-3.3-3-4.6-6.7-5.1-10.9-1.2-10.4 0-20.6 4.3-30.2.5-1 1.1-2 1.9-3.3.5 4.2.6 7.9 1.4 11.6 4.8 23.1 20.4 36.3 49.3 63.5 10 9.4 19.3 19.2 25.6 31.6 4.8 9.3 7.3 19 5.7 29.6-.1.6.5 1.7 1.1 2 6.2 2.6 10 6.9 9.7 14.3 7.7-2.6 12.5-8 16.4-14.5 4.2 20.2-9.1 50.3-27.2 58.7.4-4.5 5-23.4-16.5-27.7-6.8-1.3-12.8-1.3-22.9-2.1 4.7-9 10.4-20.6.5-22.4-24.9-4.6-52.8 1.9-57.8 4.6 8.2.4 16.3 1 23.5 3.3-2 6.5-4 12.7-5.8 18.9-1.9 6.5 2.1 14.6 9.3 9.6 1.2-.9 2.3-1.9 3.3-2.7-3.1 17.9-2.9 15.9-2.8 18.3.3 10.2 9.5 7.8 15.7 7.3-2.5 11.8-29.5 27.3-45.4 25.8 7-4.7 12.7-10.3 15.9-17.9-6.5.8-12.9 1.6-19.2 2.4l-.3-.9c4.7-3.4 8-7.8 10.2-13.1 8.7-21.1-3.6-38-25-39.9-9.1-.8-17.8.8-25.9 5.5 6.2-15.6 17.2-26.6 32.6-34.5-15.2-4.3-8.9-2.7-24.6-6.3 14.6-9.3 30.2-13.2 46.5-14.6-5.2-3.2-48.1-3.6-70.2 20.9 7.9 1.4 15.5 2.8 23.2 4.2-23.8 7-44 19.7-62.4 35.6 1.1-4.8 2.7-9.5 3.3-14.3.6-4.5.8-9.2.1-13.6-1.5-9.4-8.9-15.1-19.7-16.3-7.9-.9-15.6.1-23.3 1.3-.9.1-1.7.3-2.9 0 15.8-14.8 36-21.7 53.1-33.5 6-4.5 6.8-8.2 3-14.9zm128.4 26.8c3.3 16 12.6 25.5 23.8 24.3-4.6-11.3-12.1-19.5-23.8-24.3z"],microblog:[448,512,[],"e01a","M399.36,362.23c29.49-34.69,47.1-78.34,47.1-125.79C446.46,123.49,346.86,32,224,32S1.54,123.49,1.54,236.44,101.14,440.87,224,440.87a239.28,239.28,0,0,0,79.44-13.44,7.18,7.18,0,0,1,8.12,2.56c18.58,25.09,47.61,42.74,79.89,49.92a4.42,4.42,0,0,0,5.22-3.43,4.37,4.37,0,0,0-.85-3.62,87,87,0,0,1,3.69-110.69ZM329.52,212.4l-57.3,43.49L293,324.75a6.5,6.5,0,0,1-9.94,7.22L224,290.92,164.94,332a6.51,6.51,0,0,1-9.95-7.22l20.79-68.86-57.3-43.49a6.5,6.5,0,0,1,3.8-11.68l71.88-1.51,23.66-67.92a6.5,6.5,0,0,1,12.28,0l23.66,67.92,71.88,1.51a6.5,6.5,0,0,1,3.88,11.68Z"],"cc-diners-club":[576,512,[],"f24c","M239.7 79.9c-96.9 0-175.8 78.6-175.8 175.8 0 96.9 78.9 175.8 175.8 175.8 97.2 0 175.8-78.9 175.8-175.8 0-97.2-78.6-175.8-175.8-175.8zm-39.9 279.6c-41.7-15.9-71.4-56.4-71.4-103.8s29.7-87.9 71.4-104.1v207.9zm79.8.3V151.6c41.7 16.2 71.4 56.7 71.4 104.1s-29.7 87.9-71.4 104.1zM528 32H48C21.5 32 0 53.5 0 80v352c0 26.5 21.5 48 48 48h480c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48zM329.7 448h-90.3c-106.2 0-193.8-85.5-193.8-190.2C45.6 143.2 133.2 64 239.4 64h90.3c105 0 200.7 79.2 200.7 193.8 0 104.7-95.7 190.2-200.7 190.2z"],"gg-circle":[512,512,[],"f261","M257 8C120 8 9 119 9 256s111 248 248 248 248-111 248-248S394 8 257 8zm-49.5 374.8L81.8 257.1l125.7-125.7 35.2 35.4-24.2 24.2-11.1-11.1-77.2 77.2 77.2 77.2 26.6-26.6-53.1-52.9 24.4-24.4 77.2 77.2-75 75.2zm99-2.2l-35.2-35.2 24.1-24.4 11.1 11.1 77.2-77.2-77.2-77.2-26.5 26.5 53.1 52.9-24.4 24.4-77.2-77.2 75-75L432.2 255 306.5 380.6z"],"pied-piper-hat":[640,512,[],"f4e5","M640 24.9c-80.8 53.6-89.4 92.5-96.4 104.4-6.7 12.2-11.7 60.3-23.3 83.6-11.7 23.6-54.2 42.2-66.1 50-11.7 7.8-28.3 38.1-41.9 64.2-108.1-4.4-167.4 38.8-259.2 93.6 29.4-9.7 43.3-16.7 43.3-16.7 94.2-36 139.3-68.3 281.1-49.2 1.1 0 1.9.6 2.8.8 3.9 2.2 5.3 6.9 3.1 10.8l-53.9 95.8c-2.5 4.7-7.8 7.2-13.1 6.1-126.8-23.8-226.9 17.3-318.9 18.6C24.1 488 0 453.4 0 451.8c0-1.1.6-1.7 1.7-1.7 0 0 38.3 0 103.1-15.3C178.4 294.5 244 245.4 315.4 245.4c0 0 71.7 0 90.6 61.9 22.8-39.7 28.3-49.2 28.3-49.2 5.3-9.4 35-77.2 86.4-141.4 51.5-64 90.4-79.9 119.3-91.8z"],"kickstarter-k":[448,512,[],"f3bc","M356.6 256.2l40.8-40.5c42.2-41.9 42.2-110.3 0-152.1s-111-41.9-153.2 0L229.3 78.4C209.6 50.3 177.1 32 140.2 32C80.5 32 32 80.2 32 139.5V372.5C32 431.9 80.5 480 140.2 480c37.1 0 69.3-18.3 89-46.4l14.9 14.7c42.2 41.9 111 41.9 153.2 0s42.2-110.3 0-152.1l-40.8-40z"],yandex:[256,512,[],"f413","M153.1 315.8L65.7 512H2l96-209.8c-45.1-22.9-75.2-64.4-75.2-141.1C22.7 53.7 90.8 0 171.7 0H254v512h-55.1V315.8h-45.8zm45.8-269.3h-29.4c-44.4 0-87.4 29.4-87.4 114.6 0 82.3 39.4 108.8 87.4 108.8h29.4V46.5z"],readme:[576,512,[],"f4d5","M528.3 46.5H388.5c-48.1 0-89.9 33.3-100.4 80.3-10.6-47-52.3-80.3-100.4-80.3H48c-26.5 0-48 21.5-48 48v245.8c0 26.5 21.5 48 48 48h89.7c102.2 0 132.7 24.4 147.3 75 .7 2.8 5.2 2.8 6 0 14.7-50.6 45.2-75 147.3-75H528c26.5 0 48-21.5 48-48V94.6c0-26.4-21.3-47.9-47.7-48.1zM242 311.9c0 1.9-1.5 3.5-3.5 3.5H78.2c-1.9 0-3.5-1.5-3.5-3.5V289c0-1.9 1.5-3.5 3.5-3.5h160.4c1.9 0 3.5 1.5 3.5 3.5v22.9zm0-60.9c0 1.9-1.5 3.5-3.5 3.5H78.2c-1.9 0-3.5-1.5-3.5-3.5v-22.9c0-1.9 1.5-3.5 3.5-3.5h160.4c1.9 0 3.5 1.5 3.5 3.5V251zm0-60.9c0 1.9-1.5 3.5-3.5 3.5H78.2c-1.9 0-3.5-1.5-3.5-3.5v-22.9c0-1.9 1.5-3.5 3.5-3.5h160.4c1.9 0 3.5 1.5 3.5 3.5v22.9zm259.3 121.7c0 1.9-1.5 3.5-3.5 3.5H337.5c-1.9 0-3.5-1.5-3.5-3.5v-22.9c0-1.9 1.5-3.5 3.5-3.5h160.4c1.9 0 3.5 1.5 3.5 3.5v22.9zm0-60.9c0 1.9-1.5 3.5-3.5 3.5H337.5c-1.9 0-3.5-1.5-3.5-3.5V228c0-1.9 1.5-3.5 3.5-3.5h160.4c1.9 0 3.5 1.5 3.5 3.5v22.9zm0-60.9c0 1.9-1.5 3.5-3.5 3.5H337.5c-1.9 0-3.5-1.5-3.5-3.5v-22.8c0-1.9 1.5-3.5 3.5-3.5h160.4c1.9 0 3.5 1.5 3.5 3.5V190z"],html5:[384,512,[],"f13b","M0 32l34.9 395.8L191.5 480l157.6-52.2L384 32H0zm308.2 127.9H124.4l4.1 49.4h175.6l-13.6 148.4-97.9 27v.3h-1.1l-98.7-27.3-6-75.8h47.7L138 320l53.5 14.5 53.7-14.5 6-62.2H84.3L71.5 112.2h241.1l-4.4 47.7z"],sellsy:[640,512,[],"f213","M539.71 237.308c3.064-12.257 4.29-24.821 4.29-37.384C544 107.382 468.618 32 376.076 32c-77.22 0-144.634 53.012-163.02 127.781-15.322-13.176-34.934-20.53-55.157-20.53-46.271 0-83.962 37.69-83.962 83.961 0 7.354.92 15.015 3.065 22.369-42.9 20.225-70.785 63.738-70.785 111.234C6.216 424.843 61.68 480 129.401 480h381.198c67.72 0 123.184-55.157 123.184-123.184.001-56.384-38.916-106.025-94.073-119.508zM199.88 401.554c0 8.274-7.048 15.321-15.321 15.321H153.61c-8.274 0-15.321-7.048-15.321-15.321V290.626c0-8.273 7.048-15.321 15.321-15.321h30.949c8.274 0 15.321 7.048 15.321 15.321v110.928zm89.477 0c0 8.274-7.048 15.321-15.322 15.321h-30.949c-8.274 0-15.321-7.048-15.321-15.321V270.096c0-8.274 7.048-15.321 15.321-15.321h30.949c8.274 0 15.322 7.048 15.322 15.321v131.458zm89.477 0c0 8.274-7.047 15.321-15.321 15.321h-30.949c-8.274 0-15.322-7.048-15.322-15.321V238.84c0-8.274 7.048-15.321 15.322-15.321h30.949c8.274 0 15.321 7.048 15.321 15.321v162.714zm87.027 0c0 8.274-7.048 15.321-15.322 15.321h-28.497c-8.274 0-15.321-7.048-15.321-15.321V176.941c0-8.579 7.047-15.628 15.321-15.628h28.497c8.274 0 15.322 7.048 15.322 15.628v224.613z"],"square-web-awesome":[448,512,[],"e683","M64 32C28.7 32 0 60.7 0 96L0 416c0 35.3 28.7 64 64 64l320 0c35.3 0 64-28.7 64-64l0-320c0-35.3-28.7-64-64-64L64 32zM235 177.6L288 224l52.2-10.4c-2.6-3.9-4.2-8.5-4.2-13.6c0-13.3 10.7-24 24-24s24 10.7 24 24c0 13-10.3 23.6-23.2 24L304.5 349.1c-5.2 11.5-16.6 18.9-29.2 18.9l-102.6 0c-12.6 0-24-7.4-29.2-18.9L87.2 224C74.3 223.6 64 213 64 200c0-13.3 10.7-24 24-24s24 10.7 24 24c0 5-1.5 9.7-4.2 13.6L160 224l53.1-46.4c-8.9-4.1-15-13.1-15-23.6c0-14.4 11.6-26 26-26s26 11.6 26 26c0 10.5-6.2 19.5-15.1 23.6z"],sass:[640,512,[],"f41e","M301.84 378.92c-.3.6-.6 1.08 0 0zm249.13-87a131.16 131.16 0 0 0-58 13.5c-5.9-11.9-12-22.3-13-30.1-1.2-9.1-2.5-14.5-1.1-25.3s7.7-26.1 7.6-27.2-1.4-6.6-14.3-6.7-24 2.5-25.29 5.9a122.83 122.83 0 0 0-5.3 19.1c-2.3 11.7-25.79 53.5-39.09 75.3-4.4-8.5-8.1-16-8.9-22-1.2-9.1-2.5-14.5-1.1-25.3s7.7-26.1 7.6-27.2-1.4-6.6-14.29-6.7-24 2.5-25.3 5.9-2.7 11.4-5.3 19.1-33.89 77.3-42.08 95.4c-4.2 9.2-7.8 16.6-10.4 21.6-.4.8-.7 1.3-.9 1.7.3-.5.5-1 .5-.8-2.2 4.3-3.5 6.7-3.5 6.7v.1c-1.7 3.2-3.6 6.1-4.5 6.1-.6 0-1.9-8.4.3-19.9 4.7-24.2 15.8-61.8 15.7-63.1-.1-.7 2.1-7.2-7.3-10.7-9.1-3.3-12.4 2.2-13.2 2.2s-1.4 2-1.4 2 10.1-42.4-19.39-42.4c-18.4 0-44 20.2-56.58 38.5-7.9 4.3-25 13.6-43 23.5-6.9 3.8-14 7.7-20.7 11.4-.5-.5-.9-1-1.4-1.5-35.79-38.2-101.87-65.2-99.07-116.5 1-18.7 7.5-67.8 127.07-127.4 98-48.8 176.35-35.4 189.84-5.6 19.4 42.5-41.89 121.6-143.66 133-38.79 4.3-59.18-10.7-64.28-16.3-5.3-5.9-6.1-6.2-8.1-5.1-3.3 1.8-1.2 7 0 10.1 3 7.9 15.5 21.9 36.79 28.9 18.7 6.1 64.18 9.5 119.17-11.8 61.78-23.8 109.87-90.1 95.77-145.6C386.52 18.32 293-.18 204.57 31.22c-52.69 18.7-109.67 48.1-150.66 86.4-48.69 45.6-56.48 85.3-53.28 101.9 11.39 58.9 92.57 97.3 125.06 125.7-1.6.9-3.1 1.7-4.5 2.5-16.29 8.1-78.18 40.5-93.67 74.7-17.5 38.8 2.9 66.6 16.29 70.4 41.79 11.6 84.58-9.3 107.57-43.6s20.2-79.1 9.6-99.5c-.1-.3-.3-.5-.4-.8 4.2-2.5 8.5-5 12.8-7.5 8.29-4.9 16.39-9.4 23.49-13.3-4 10.8-6.9 23.8-8.4 42.6-1.8 22 7.3 50.5 19.1 61.7 5.2 4.9 11.49 5 15.39 5 13.8 0 20-11.4 26.89-25 8.5-16.6 16-35.9 16-35.9s-9.4 52.2 16.3 52.2c9.39 0 18.79-12.1 23-18.3v.1s.2-.4.7-1.2c1-1.5 1.5-2.4 1.5-2.4v-.3c3.8-6.5 12.1-21.4 24.59-46 16.2-31.8 31.69-71.5 31.69-71.5a201.24 201.24 0 0 0 6.2 25.8c2.8 9.5 8.7 19.9 13.4 30-3.8 5.2-6.1 8.2-6.1 8.2a.31.31 0 0 0 .1.2c-3 4-6.4 8.3-9.9 12.5-12.79 15.2-28 32.6-30 37.6-2.4 5.9-1.8 10.3 2.8 13.7 3.4 2.6 9.4 3 15.69 2.5 11.5-.8 19.6-3.6 23.5-5.4a82.2 82.2 0 0 0 20.19-10.6c12.5-9.2 20.1-22.4 19.4-39.8-.4-9.6-3.5-19.2-7.3-28.2 1.1-1.6 2.3-3.3 3.4-5C434.8 301.72 450.1 270 450.1 270a201.24 201.24 0 0 0 6.2 25.8c2.4 8.1 7.09 17 11.39 25.7-18.59 15.1-30.09 32.6-34.09 44.1-7.4 21.3-1.6 30.9 9.3 33.1 4.9 1 11.9-1.3 17.1-3.5a79.46 79.46 0 0 0 21.59-11.1c12.5-9.2 24.59-22.1 23.79-39.6-.3-7.9-2.5-15.8-5.4-23.4 15.7-6.6 36.09-10.2 62.09-7.2 55.68 6.5 66.58 41.3 64.48 55.8s-13.8 22.6-17.7 25-5.1 3.3-4.8 5.1c.5 2.6 2.3 2.5 5.6 1.9 4.6-.8 29.19-11.8 30.29-38.7 1.6-34-31.09-71.4-89-71.1zm-429.18 144.7c-18.39 20.1-44.19 27.7-55.28 21.3C54.61 451 59.31 421.42 82 400c13.8-13 31.59-25 43.39-32.4 2.7-1.6 6.6-4 11.4-6.9.8-.5 1.2-.7 1.2-.7.9-.6 1.9-1.1 2.9-1.7 8.29 30.4.3 57.2-19.1 78.3zm134.36-91.4c-6.4 15.7-19.89 55.7-28.09 53.6-7-1.8-11.3-32.3-1.4-62.3 5-15.1 15.6-33.1 21.9-40.1 10.09-11.3 21.19-14.9 23.79-10.4 3.5 5.9-12.2 49.4-16.2 59.2zm111 53c-2.7 1.4-5.2 2.3-6.4 1.6-.9-.5 1.1-2.4 1.1-2.4s13.9-14.9 19.4-21.7c3.2-4 6.9-8.7 10.89-13.9 0 .5.1 1 .1 1.6-.13 17.9-17.32 30-25.12 34.8zm85.58-19.5c-2-1.4-1.7-6.1 5-20.7 2.6-5.7 8.59-15.3 19-24.5a36.18 36.18 0 0 1 1.9 10.8c-.1 22.5-16.2 30.9-25.89 34.4z"],wirsindhandwerk:[512,512,["wsh"],"e2d0","M50.77161,479.81213h83.36071V367.84741l-83.36071,47.009Zm329.04675,0h82.35022V414.85645l-82.35022-47.009Zm.00568-448V251.568L256.1759,179.1861,134.50378,251.568V31.81213H50.77161V392.60565L256.1759,270.31909,462.16858,392.60565V31.81213Z"],buromobelexperte:[448,512,[],"f37f","M0 32v128h128V32H0zm120 120H8V40h112v112zm40-120v128h128V32H160zm120 120H168V40h112v112zm40-120v128h128V32H320zm120 120H328V40h112v112zM0 192v128h128V192H0zm120 120H8V200h112v112zm40-120v128h128V192H160zm120 120H168V200h112v112zm40-120v128h128V192H320zm120 120H328V200h112v112zM0 352v128h128V352H0zm120 120H8V360h112v112zm40-120v128h128V352H160zm120 120H168V360h112v112zm40-120v128h128V352H320z"],salesforce:[640,512,[],"f83b","M248.89 245.64h-26.35c.69-5.16 3.32-14.12 13.64-14.12 6.75 0 11.97 3.82 12.71 14.12zm136.66-13.88c-.47 0-14.11-1.77-14.11 20s13.63 20 14.11 20c13 0 14.11-13.54 14.11-20 0-21.76-13.66-20-14.11-20zm-243.22 23.76a8.63 8.63 0 0 0-3.29 7.29c0 4.78 2.08 6.05 3.29 7.05 4.7 3.7 15.07 2.12 20.93.95v-16.94c-5.32-1.07-16.73-1.96-20.93 1.65zM640 232c0 87.58-80 154.39-165.36 136.43-18.37 33-70.73 70.75-132.2 41.63-41.16 96.05-177.89 92.18-213.81-5.17C8.91 428.78-50.19 266.52 53.36 205.61 18.61 126.18 76 32 167.67 32a124.24 124.24 0 0 1 98.56 48.7c20.7-21.4 49.4-34.81 81.15-34.81 42.34 0 79 23.52 98.8 58.57C539 63.78 640 132.69 640 232zm-519.55 31.8c0-11.76-11.69-15.17-17.87-17.17-5.27-2.11-13.41-3.51-13.41-8.94 0-9.46 17-6.66 25.17-2.12 0 0 1.17.71 1.64-.47.24-.7 2.36-6.58 2.59-7.29a1.13 1.13 0 0 0-.7-1.41c-12.33-7.63-40.7-8.51-40.7 12.7 0 12.46 11.49 15.44 17.88 17.17 4.72 1.58 13.17 3 13.17 8.7 0 4-3.53 7.06-9.17 7.06a31.76 31.76 0 0 1-19-6.35c-.47-.23-1.42-.71-1.65.71l-2.4 7.47c-.47.94.23 1.18.23 1.41 1.75 1.4 10.3 6.59 22.82 6.59 13.17 0 21.4-7.06 21.4-18.11zm32-42.58c-10.13 0-18.66 3.17-21.4 5.18a1 1 0 0 0-.24 1.41l2.59 7.06a1 1 0 0 0 1.18.7c.65 0 6.8-4 16.93-4 4 0 7.06.71 9.18 2.36 3.6 2.8 3.06 8.29 3.06 10.58-4.79-.3-19.11-3.44-29.41 3.76a16.92 16.92 0 0 0-7.34 14.54c0 5.9 1.51 10.4 6.59 14.35 12.24 8.16 36.28 2 38.1 1.41 1.58-.32 3.53-.66 3.53-1.88v-33.88c.04-4.61.32-21.64-22.78-21.64zM199 200.24a1.11 1.11 0 0 0-1.18-1.18H188a1.11 1.11 0 0 0-1.17 1.18v79a1.11 1.11 0 0 0 1.17 1.18h9.88a1.11 1.11 0 0 0 1.18-1.18zm55.75 28.93c-2.1-2.31-6.79-7.53-17.65-7.53-3.51 0-14.16.23-20.7 8.94-6.35 7.63-6.58 18.11-6.58 21.41 0 3.12.15 14.26 7.06 21.17 2.64 2.91 9.06 8.23 22.81 8.23 10.82 0 16.47-2.35 18.58-3.76.47-.24.71-.71.24-1.88l-2.35-6.83a1.26 1.26 0 0 0-1.41-.7c-2.59.94-6.35 2.82-15.29 2.82-17.42 0-16.85-14.74-16.94-16.7h37.17a1.23 1.23 0 0 0 1.17-.94c-.29 0 2.07-14.7-6.09-24.23zm36.69 52.69c13.17 0 21.41-7.06 21.41-18.11 0-11.76-11.7-15.17-17.88-17.17-4.14-1.66-13.41-3.38-13.41-8.94 0-3.76 3.29-6.35 8.47-6.35a38.11 38.11 0 0 1 16.7 4.23s1.18.71 1.65-.47c.23-.7 2.35-6.58 2.58-7.29a1.13 1.13 0 0 0-.7-1.41c-7.91-4.9-16.74-4.94-20.23-4.94-12 0-20.46 7.29-20.46 17.64 0 12.46 11.48 15.44 17.87 17.17 6.11 2 13.17 3.26 13.17 8.7 0 4-3.52 7.06-9.17 7.06a31.8 31.8 0 0 1-19-6.35 1 1 0 0 0-1.65.71l-2.35 7.52c-.47.94.23 1.18.23 1.41 1.72 1.4 10.33 6.59 22.79 6.59zM357.09 224c0-.71-.24-1.18-1.18-1.18h-11.76c0-.14.94-8.94 4.47-12.47 4.16-4.15 11.76-1.64 12-1.64 1.17.47 1.41 0 1.64-.47l2.83-7.77c.7-.94 0-1.17-.24-1.41-5.09-2-17.35-2.87-24.46 4.24-5.48 5.48-7 13.92-8 19.52h-8.47a1.28 1.28 0 0 0-1.17 1.18l-1.42 7.76c0 .7.24 1.17 1.18 1.17h8.23c-8.51 47.9-8.75 50.21-10.35 55.52-1.08 3.62-3.29 6.9-5.88 7.76-.09 0-3.88 1.68-9.64-.24 0 0-.94-.47-1.41.71-.24.71-2.59 6.82-2.83 7.53s0 1.41.47 1.41c5.11 2 13 1.77 17.88 0 6.28-2.28 9.72-7.89 11.53-12.94 2.75-7.71 2.81-9.79 11.76-59.74h12.23a1.29 1.29 0 0 0 1.18-1.18zm53.39 16c-.56-1.68-5.1-18.11-25.17-18.11-15.25 0-23 10-25.16 18.11-1 3-3.18 14 0 23.52.09.3 4.41 18.12 25.16 18.12 14.95 0 22.9-9.61 25.17-18.12 3.21-9.61 1.01-20.52 0-23.52zm45.4-16.7c-5-1.65-16.62-1.9-22.11 5.41v-4.47a1.11 1.11 0 0 0-1.18-1.17h-9.4a1.11 1.11 0 0 0-1.18 1.17v55.28a1.12 1.12 0 0 0 1.18 1.18h9.64a1.12 1.12 0 0 0 1.18-1.18v-27.77c0-2.91.05-11.37 4.46-15.05 4.9-4.9 12-3.36 13.41-3.06a1.57 1.57 0 0 0 1.41-.94 74 74 0 0 0 3.06-8 1.16 1.16 0 0 0-.47-1.41zm46.81 54.1l-2.12-7.29c-.47-1.18-1.41-.71-1.41-.71-4.23 1.82-10.15 1.89-11.29 1.89-4.64 0-17.17-1.13-17.17-19.76 0-6.23 1.85-19.76 16.47-19.76a34.85 34.85 0 0 1 11.52 1.65s.94.47 1.18-.71c.94-2.59 1.64-4.47 2.59-7.53.23-.94-.47-1.17-.71-1.17-11.59-3.87-22.34-2.53-27.76 0-1.59.74-16.23 6.49-16.23 27.52 0 2.9-.58 30.11 28.94 30.11a44.45 44.45 0 0 0 15.52-2.83 1.3 1.3 0 0 0 .47-1.42zm53.87-39.52c-.8-3-5.37-16.23-22.35-16.23-16 0-23.52 10.11-25.64 18.59a38.58 38.58 0 0 0-1.65 11.76c0 25.87 18.84 29.4 29.88 29.4 10.82 0 16.46-2.35 18.58-3.76.47-.24.71-.71.24-1.88l-2.36-6.83a1.26 1.26 0 0 0-1.41-.7c-2.59.94-6.35 2.82-15.29 2.82-17.42 0-16.85-14.74-16.93-16.7h37.16a1.25 1.25 0 0 0 1.18-.94c-.24-.01.94-7.07-1.41-15.54zm-23.29-6.35c-10.33 0-13 9-13.64 14.12H546c-.88-11.92-7.62-14.13-12.73-14.13z"],"octopus-deploy":[512,512,[],"e082","M455.6,349.2c-45.891-39.09-36.67-77.877-16.095-128.11C475.16,134.04,415.967,34.14,329.93,8.3,237.04-19.6,134.252,24.341,99.677,117.147a180.862,180.862,0,0,0-10.988,73.544c1.733,29.543,14.717,52.97,24.09,80.3,17.2,50.161-28.1,92.743-66.662,117.582-46.806,30.2-36.319,39.857-8.428,41.858,23.378,1.68,44.478-4.548,65.265-15.045,9.2-4.647,40.687-18.931,45.13-28.588C135.9,413.388,111.122,459.5,126.621,488.9c19.1,36.229,67.112-31.77,76.709-45.812,8.591-12.572,42.963-81.279,63.627-46.926,18.865,31.361,8.6,76.391,35.738,104.622,32.854,34.2,51.155-18.312,51.412-44.221.163-16.411-6.1-95.852,29.9-59.944C405.428,418,436.912,467.8,472.568,463.642c38.736-4.516-22.123-67.967-28.262-78.695,5.393,4.279,53.665,34.128,53.818,9.52C498.234,375.678,468.039,359.8,455.6,349.2Z"],medapps:[320,512,[],"f3c6","M118.3 238.4c3.5-12.5 6.9-33.6 13.2-33.6 8.3 1.8 9.6 23.4 18.6 36.6 4.6-23.5 5.3-85.1 14.1-86.7 9-.7 19.7 66.5 22 77.5 9.9 4.1 48.9 6.6 48.9 6.6 1.9 7.3-24 7.6-40 7.8-4.6 14.8-5.4 27.7-11.4 28-4.7.2-8.2-28.8-17.5-49.6l-9.4 65.5c-4.4 13-15.5-22.5-21.9-39.3-3.3-.1-62.4-1.6-47.6-7.8l31-5zM228 448c21.2 0 21.2-32 0-32H92c-21.2 0-21.2 32 0 32h136zm-24 64c21.2 0 21.2-32 0-32h-88c-21.2 0-21.2 32 0 32h88zm34.2-141.5c3.2-18.9 5.2-36.4 11.9-48.8 7.9-14.7 16.1-28.1 24-41 24.6-40.4 45.9-75.2 45.9-125.5C320 69.6 248.2 0 160 0S0 69.6 0 155.2c0 50.2 21.3 85.1 45.9 125.5 7.9 12.9 16 26.3 24 41 6.7 12.5 8.7 29.8 11.9 48.9 3.5 21 36.1 15.7 32.6-5.1-3.6-21.7-5.6-40.7-15.3-58.6C66.5 246.5 33 211.3 33 155.2 33 87.3 90 32 160 32s127 55.3 127 123.2c0 56.1-33.5 91.3-66.1 151.6-9.7 18-11.7 37.4-15.3 58.6-3.4 20.6 29 26.4 32.6 5.1z"],ns8:[640,512,[],"f3d5","M104.324,269.172h26.067V242.994H104.324Zm52.466-26.178-.055-26.178v-.941a39.325,39.325,0,0,0-78.644.941v.166h26.4v-.166a12.98,12.98,0,0,1,25.956,0v26.178Zm52.356,25.846a91.1,91.1,0,0,1-91.1,91.1h-.609a91.1,91.1,0,0,1-91.1-91.1H0v.166A117.33,117.33,0,0,0,117.44,386.28h.775A117.331,117.331,0,0,0,235.49,268.84V242.828H209.146Zm-157.233,0a65.362,65.362,0,0,0,130.723,0H156.292a39.023,39.023,0,0,1-78.035,0V242.883H51.968v-26.62A65.42,65.42,0,0,1,182.8,217.48v25.293h26.344V217.48a91.761,91.761,0,0,0-183.522,0v25.4H51.913Zm418.4-71.173c13.67,0,24.573,6.642,30.052,18.264l.719,1.549,23.245-11.511-.609-1.439c-8.025-19.26-28.5-31.27-53.407-31.27-23.134,0-43.611,11.4-50.972,28.447-.123,26.876-.158,23.9,0,24.85,4.7,11.013,14.555,19.37,28.668,24.241a102.033,102.033,0,0,0,19.813,3.984c5.479.72,10.626,1.384,15.829,3.1,6.364,2.1,10.46,5.257,12.84,9.851v9.851c-3.708,7.527-13.781,12.342-25.791,12.342-14.334,0-25.956-6.918-31.933-19.039l-.72-1.494L415.026,280.9l.553,1.439c7.915,19.426,29.609,32.044,55.289,32.044,23.632,0,44.608-11.4,52.3-28.447l.166-25.9-.166-.664c-4.87-11.014-15.219-19.647-28.944-24.241-7.693-2.712-14.335-3.6-20.7-4.427a83.777,83.777,0,0,1-14.832-2.878c-6.31-1.937-10.4-5.092-12.619-9.63v-8.412C449.45,202.427,458.969,197.667,470.315,197.667ZM287.568,311.344h26.067v-68.4H287.568Zm352.266-53.3c-2.933-6.254-8.3-12.01-15.441-16.714A37.99,37.99,0,0,0,637.4,226l.166-25.347-.166-.664C630.038,184,610.667,173.26,589.25,173.26S548.461,184,541.1,199.992l-.166,25.347.166.664a39.643,39.643,0,0,0,13.006,15.331c-7.2,4.7-12.508,10.46-15.441,16.714l-.166,28.889.166.72c7.582,15.994,27.893,26.731,50.585,26.731s43.057-10.737,50.584-26.731l.166-28.89Zm-73.22-50.806c3.6-6.31,12.563-10.516,22.58-10.516s19.038,4.206,22.636,10.516v13.725c-3.542,6.2-12.563,10.349-22.636,10.349s-19.094-4.15-22.58-10.349Zm47.319,72.169c-3.764,6.641-13.338,10.9-24.683,10.9-11.125,0-20.976-4.372-24.684-10.9V263.25c3.708-6.309,13.5-10.515,24.684-10.515,11.345,0,20.919,4.15,24.683,10.515ZM376.4,265.962l-59.827-89.713h-29v40.623h26.51v.387l62.539,94.085H402.3V176.249H376.4Z"],"pinterest-p":[384,512,[],"f231","M204 6.5C101.4 6.5 0 74.9 0 185.6 0 256 39.6 296 63.6 296c9.9 0 15.6-27.6 15.6-35.4 0-9.3-23.7-29.1-23.7-67.8 0-80.4 61.2-137.4 140.4-137.4 68.1 0 118.5 38.7 118.5 109.8 0 53.1-21.3 152.7-90.3 152.7-24.9 0-46.2-18-46.2-43.8 0-37.8 26.4-74.4 26.4-113.4 0-66.2-93.9-54.2-93.9 25.8 0 16.8 2.1 35.4 9.6 50.7-13.8 59.4-42 147.9-42 209.1 0 18.9 2.7 37.5 4.5 56.4 3.4 3.8 1.7 3.4 6.9 1.5 50.4-69 48.6-82.5 71.4-172.8 12.3 23.4 44.1 36 69.3 36 106.2 0 153.9-103.5 153.9-196.8C384 71.3 298.2 6.5 204 6.5z"],apper:[640,512,[],"f371","M42.1 239.1c22.2 0 29 2.8 33.5 14.6h.8v-22.9c0-11.3-4.8-15.4-17.9-15.4-11.3 0-14.4 2.5-15.1 12.8H4.8c.3-13.9 1.5-19.1 5.8-24.4C17.9 195 29.5 192 56.7 192c33 0 47.1 5 53.9 18.9 2 4.3 4 15.6 4 23.7v76.3H76.3l1.3-19.1h-1c-5.3 15.6-13.6 20.4-35.5 20.4-30.3 0-41.1-10.1-41.1-37.3 0-25.2 12.3-35.8 42.1-35.8zm17.1 48.1c13.1 0 16.9-3 16.9-13.4 0-9.1-4.3-11.6-19.6-11.6-13.1 0-17.9 3-17.9 12.1-.1 10.4 3.7 12.9 20.6 12.9zm77.8-94.9h38.3l-1.5 20.6h.8c9.1-17.1 15.9-20.9 37.5-20.9 14.4 0 24.7 3 31.5 9.1 9.8 8.6 12.8 20.4 12.8 48.1 0 30-3 43.1-12.1 52.9-6.8 7.3-16.4 10.1-33.2 10.1-20.4 0-29.2-5.5-33.8-21.2h-.8v70.3H137v-169zm80.9 60.7c0-27.5-3.3-32.5-20.7-32.5-16.9 0-20.7 5-20.7 28.7 0 28 3.5 33.5 21.2 33.5 16.4 0 20.2-5.6 20.2-29.7zm57.9-60.7h38.3l-1.5 20.6h.8c9.1-17.1 15.9-20.9 37.5-20.9 14.4 0 24.7 3 31.5 9.1 9.8 8.6 12.8 20.4 12.8 48.1 0 30-3 43.1-12.1 52.9-6.8 7.3-16.4 10.1-33.3 10.1-20.4 0-29.2-5.5-33.8-21.2h-.8v70.3h-39.5v-169zm80.9 60.7c0-27.5-3.3-32.5-20.7-32.5-16.9 0-20.7 5-20.7 28.7 0 28 3.5 33.5 21.2 33.5 16.4 0 20.2-5.6 20.2-29.7zm53.8-3.8c0-25.4 3.3-37.8 12.3-45.8 8.8-8.1 22.2-11.3 45.1-11.3 42.8 0 55.7 12.8 55.7 55.7v11.1h-75.3c-.3 2-.3 4-.3 4.8 0 16.9 4.5 21.9 20.1 21.9 13.9 0 17.9-3 17.9-13.9h37.5v2.3c0 9.8-2.5 18.9-6.8 24.7-7.3 9.8-19.6 13.6-44.3 13.6-27.5 0-41.6-3.3-50.6-12.3-8.5-8.5-11.3-21.3-11.3-50.8zm76.4-11.6c-.3-1.8-.3-3.3-.3-3.8 0-12.3-3.3-14.6-19.6-14.6-14.4 0-17.1 3-18.1 15.1l-.3 3.3h38.3zm55.6-45.3h38.3l-1.8 19.9h.7c6.8-14.9 14.4-20.2 29.7-20.2 10.8 0 19.1 3.3 23.4 9.3 5.3 7.3 6.8 14.4 6.8 34 0 1.5 0 5 .2 9.3h-35c.3-1.8.3-3.3.3-4 0-15.4-2-19.4-10.3-19.4-6.3 0-10.8 3.3-13.1 9.3-1 3-1 4.3-1 12.3v68h-38.3V192.3z"],"fort-awesome":[512,512,[],"f286","M489.2 287.9h-27.4c-2.6 0-4.6 2-4.6 4.6v32h-36.6V146.2c0-2.6-2-4.6-4.6-4.6h-27.4c-2.6 0-4.6 2-4.6 4.6v32h-36.6v-32c0-2.6-2-4.6-4.6-4.6h-27.4c-2.6 0-4.6 2-4.6 4.6v32h-36.6v-32c0-6-8-4.6-11.7-4.6v-38c8.3-2 17.1-3.4 25.7-3.4 10.9 0 20.9 4.3 31.4 4.3 4.6 0 27.7-1.1 27.7-8v-60c0-2.6-2-4.6-4.6-4.6-5.1 0-15.1 4.3-24 4.3-9.7 0-20.9-4.3-32.6-4.3-8 0-16 1.1-23.7 2.9v-4.9c5.4-2.6 9.1-8.3 9.1-14.3 0-20.7-31.4-20.8-31.4 0 0 6 3.7 11.7 9.1 14.3v111.7c-3.7 0-11.7-1.4-11.7 4.6v32h-36.6v-32c0-2.6-2-4.6-4.6-4.6h-27.4c-2.6 0-4.6 2-4.6 4.6v32H128v-32c0-2.6-2-4.6-4.6-4.6H96c-2.6 0-4.6 2-4.6 4.6v178.3H54.8v-32c0-2.6-2-4.6-4.6-4.6H22.8c-2.6 0-4.6 2-4.6 4.6V512h182.9v-96c0-72.6 109.7-72.6 109.7 0v96h182.9V292.5c.1-2.6-1.9-4.6-4.5-4.6zm-288.1-4.5c0 2.6-2 4.6-4.6 4.6h-27.4c-2.6 0-4.6-2-4.6-4.6v-64c0-2.6 2-4.6 4.6-4.6h27.4c2.6 0 4.6 2 4.6 4.6v64zm146.4 0c0 2.6-2 4.6-4.6 4.6h-27.4c-2.6 0-4.6-2-4.6-4.6v-64c0-2.6 2-4.6 4.6-4.6h27.4c2.6 0 4.6 2 4.6 4.6v64z"],waze:[512,512,[],"f83f","M502.17 201.67C516.69 287.53 471.23 369.59 389 409.8c13 34.1-12.4 70.2-48.32 70.2a51.68 51.68 0 0 1-51.57-49c-6.44.19-64.2 0-76.33-.64A51.69 51.69 0 0 1 159 479.92c-33.86-1.36-57.95-34.84-47-67.92-37.21-13.11-72.54-34.87-99.62-70.8-13-17.28-.48-41.8 20.84-41.8 46.31 0 32.22-54.17 43.15-110.26C94.8 95.2 193.12 32 288.09 32c102.48 0 197.15 70.67 214.08 169.67zM373.51 388.28c42-19.18 81.33-56.71 96.29-102.14 40.48-123.09-64.15-228-181.71-228-83.45 0-170.32 55.42-186.07 136-9.53 48.91 5 131.35-68.75 131.35C58.21 358.6 91.6 378.11 127 389.54c24.66-21.8 63.87-15.47 79.83 14.34 14.22 1 79.19 1.18 87.9.82a51.69 51.69 0 0 1 78.78-16.42zM205.12 187.13c0-34.74 50.84-34.75 50.84 0s-50.84 34.74-50.84 0zm116.57 0c0-34.74 50.86-34.75 50.86 0s-50.86 34.75-50.86 0zm-122.61 70.69c-3.44-16.94 22.18-22.18 25.62-5.21l.06.28c4.14 21.42 29.85 44 64.12 43.07 35.68-.94 59.25-22.21 64.11-42.77 4.46-16.05 28.6-10.36 25.47 6-5.23 22.18-31.21 62-91.46 62.9-42.55 0-80.88-27.84-87.9-64.25z"],bluesky:[576,512,[],"e671","M407.8 294.7c-3.3-.4-6.7-.8-10-1.3c3.4 .4 6.7 .9 10 1.3zM288 227.1C261.9 176.4 190.9 81.9 124.9 35.3C61.6-9.4 37.5-1.7 21.6 5.5C3.3 13.8 0 41.9 0 58.4S9.1 194 15 213.9c19.5 65.7 89.1 87.9 153.2 80.7c3.3-.5 6.6-.9 10-1.4c-3.3 .5-6.6 1-10 1.4C74.3 308.6-9.1 342.8 100.3 464.5C220.6 589.1 265.1 437.8 288 361.1c22.9 76.7 49.2 222.5 185.6 103.4c102.4-103.4 28.1-156-65.8-169.9c-3.3-.4-6.7-.8-10-1.3c3.4 .4 6.7 .9 10 1.3c64.1 7.1 133.6-15.1 153.2-80.7C566.9 194 576 75 576 58.4s-3.3-44.7-21.6-52.9c-15.8-7.1-40-14.9-103.2 29.8C385.1 81.9 314.1 176.4 288 227.1z"],"cc-jcb":[576,512,[],"f24b","M431.5 244.3V212c41.2 0 38.5.2 38.5.2 7.3 1.3 13.3 7.3 13.3 16 0 8.8-6 14.5-13.3 15.8-1.2.4-3.3.3-38.5.3zm42.8 20.2c-2.8-.7-3.3-.5-42.8-.5v35c39.6 0 40 .2 42.8-.5 7.5-1.5 13.5-8 13.5-17 0-8.7-6-15.5-13.5-17zM576 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h480c26.5 0 48 21.5 48 48zM182 192.3h-57c0 67.1 10.7 109.7-35.8 109.7-19.5 0-38.8-5.7-57.2-14.8v28c30 8.3 68 8.3 68 8.3 97.9 0 82-47.7 82-131.2zm178.5 4.5c-63.4-16-165-14.9-165 59.3 0 77.1 108.2 73.6 165 59.2V287C312.9 311.7 253 309 253 256s59.8-55.6 107.5-31.2v-28zM544 286.5c0-18.5-16.5-30.5-38-32v-.8c19.5-2.7 30.3-15.5 30.3-30.2 0-19-15.7-30-37-31 0 0 6.3-.3-120.3-.3v127.5h122.7c24.3.1 42.3-12.9 42.3-33.2z"],snapchat:[512,512,[62124,"snapchat-ghost"],"f2ab","M496.926,366.6c-3.373-9.176-9.8-14.086-17.112-18.153-1.376-.806-2.641-1.451-3.72-1.947-2.182-1.128-4.414-2.22-6.634-3.373-22.8-12.09-40.609-27.341-52.959-45.42a102.889,102.889,0,0,1-9.089-16.12c-1.054-3.013-1-4.724-.248-6.287a10.221,10.221,0,0,1,2.914-3.038c3.918-2.591,7.96-5.22,10.7-6.993,4.885-3.162,8.754-5.667,11.246-7.44,9.362-6.547,15.909-13.5,20-21.278a42.371,42.371,0,0,0,2.1-35.191c-6.2-16.318-21.613-26.449-40.287-26.449a55.543,55.543,0,0,0-11.718,1.24c-1.029.224-2.059.459-3.063.72.174-11.16-.074-22.94-1.066-34.534-3.522-40.758-17.794-62.123-32.674-79.16A130.167,130.167,0,0,0,332.1,36.443C309.515,23.547,283.91,17,256,17S202.6,23.547,180,36.443a129.735,129.735,0,0,0-33.281,26.783c-14.88,17.038-29.152,38.44-32.673,79.161-.992,11.594-1.24,23.435-1.079,34.533-1-.26-2.021-.5-3.051-.719a55.461,55.461,0,0,0-11.717-1.24c-18.687,0-34.125,10.131-40.3,26.449a42.423,42.423,0,0,0,2.046,35.228c4.105,7.774,10.652,14.731,20.014,21.278,2.48,1.736,6.361,4.24,11.246,7.44,2.641,1.711,6.5,4.216,10.28,6.72a11.054,11.054,0,0,1,3.3,3.311c.794,1.624.818,3.373-.36,6.6a102.02,102.02,0,0,1-8.94,15.785c-12.077,17.669-29.363,32.648-51.434,44.639C32.355,348.608,20.2,352.75,15.069,366.7c-3.868,10.528-1.339,22.506,8.494,32.6a49.137,49.137,0,0,0,12.4,9.387,134.337,134.337,0,0,0,30.342,12.139,20.024,20.024,0,0,1,6.126,2.741c3.583,3.137,3.075,7.861,7.849,14.78a34.468,34.468,0,0,0,8.977,9.127c10.019,6.919,21.278,7.353,33.207,7.811,10.776.41,22.989.881,36.939,5.481,5.778,1.91,11.78,5.605,18.736,9.92C194.842,480.951,217.707,495,255.973,495s61.292-14.123,78.118-24.428c6.907-4.24,12.872-7.9,18.489-9.758,13.949-4.613,26.163-5.072,36.939-5.481,11.928-.459,23.187-.893,33.206-7.812a34.584,34.584,0,0,0,10.218-11.16c3.434-5.84,3.348-9.919,6.572-12.771a18.971,18.971,0,0,1,5.753-2.629A134.893,134.893,0,0,0,476.02,408.71a48.344,48.344,0,0,0,13.019-10.193l.124-.149C498.389,388.5,500.708,376.867,496.926,366.6Zm-34.013,18.277c-20.745,11.458-34.533,10.23-45.259,17.137-9.114,5.865-3.72,18.513-10.342,23.076-8.134,5.617-32.177-.4-63.239,9.858-25.618,8.469-41.961,32.822-88.038,32.822s-62.036-24.3-88.076-32.884c-31-10.255-55.092-4.241-63.239-9.858-6.609-4.563-1.24-17.211-10.341-23.076-10.739-6.907-24.527-5.679-45.26-17.075-13.206-7.291-5.716-11.8-1.314-13.937,75.143-36.381,87.133-92.552,87.666-96.719.645-5.046,1.364-9.014-4.191-14.148-5.369-4.96-29.189-19.7-35.8-24.316-10.937-7.638-15.748-15.264-12.2-24.638,2.48-6.485,8.531-8.928,14.879-8.928a27.643,27.643,0,0,1,5.965.67c12,2.6,23.659,8.617,30.392,10.242a10.749,10.749,0,0,0,2.48.335c3.6,0,4.86-1.811,4.612-5.927-.768-13.132-2.628-38.725-.558-62.644,2.84-32.909,13.442-49.215,26.04-63.636,6.051-6.932,34.484-36.976,88.857-36.976s82.88,29.92,88.931,36.827c12.611,14.421,23.225,30.727,26.04,63.636,2.071,23.919.285,49.525-.558,62.644-.285,4.327,1.017,5.927,4.613,5.927a10.648,10.648,0,0,0,2.48-.335c6.745-1.624,18.4-7.638,30.4-10.242a27.641,27.641,0,0,1,5.964-.67c6.386,0,12.4,2.48,14.88,8.928,3.546,9.374-1.24,17-12.189,24.639-6.609,4.612-30.429,19.343-35.8,24.315-5.568,5.134-4.836,9.1-4.191,14.149.533,4.228,12.511,60.4,87.666,96.718C468.629,373.011,476.119,377.524,462.913,384.877Z"],"fantasy-flight-games":[512,512,[],"f6dc","M256 32.86L32.86 256 256 479.14 479.14 256 256 32.86zM88.34 255.83c1.96-2 11.92-12.3 96.49-97.48 41.45-41.75 86.19-43.77 119.77-18.69 24.63 18.4 62.06 58.9 62.15 59 .68.74 1.07 2.86.58 3.38-11.27 11.84-22.68 23.54-33.5 34.69-34.21-32.31-40.52-38.24-48.51-43.95-17.77-12.69-41.4-10.13-56.98 5.1-2.17 2.13-1.79 3.43.12 5.35 2.94 2.95 28.1 28.33 35.09 35.78-11.95 11.6-23.66 22.97-35.69 34.66-12.02-12.54-24.48-25.53-36.54-38.11-21.39 21.09-41.69 41.11-61.85 60.99zm234.82 101.6c-35.49 35.43-78.09 38.14-106.99 20.47-22.08-13.5-39.38-32.08-72.93-66.84 12.05-12.37 23.79-24.42 35.37-36.31 33.02 31.91 37.06 36.01 44.68 42.09 18.48 14.74 42.52 13.67 59.32-1.8 3.68-3.39 3.69-3.64.14-7.24-10.59-10.73-21.19-21.44-31.77-32.18-1.32-1.34-3.03-2.48-.8-4.69 10.79-10.71 21.48-21.52 32.21-32.29.26-.26.65-.38 1.91-1.07 12.37 12.87 24.92 25.92 37.25 38.75 21.01-20.73 41.24-40.68 61.25-60.42 13.68 13.4 27.13 26.58 40.86 40.03-20.17 20.86-81.68 82.71-100.5 101.5zM256 0L0 256l256 256 256-256L256 0zM16 256L256 16l240 240-240 240L16 256z"],rust:[512,512,[],"e07a","M508.52,249.75,486.7,236.24c-.17-2-.34-3.93-.55-5.88l18.72-17.5a7.35,7.35,0,0,0-2.44-12.25l-24-9c-.54-1.88-1.08-3.78-1.67-5.64l15-20.83a7.35,7.35,0,0,0-4.79-11.54l-25.42-4.15c-.9-1.73-1.79-3.45-2.73-5.15l10.68-23.42a7.35,7.35,0,0,0-6.95-10.39l-25.82.91q-1.79-2.22-3.61-4.4L439,81.84A7.36,7.36,0,0,0,430.16,73L405,78.93q-2.17-1.83-4.4-3.61l.91-25.82a7.35,7.35,0,0,0-10.39-7L367.7,53.23c-1.7-.94-3.43-1.84-5.15-2.73L358.4,25.08a7.35,7.35,0,0,0-11.54-4.79L326,35.26c-1.86-.59-3.75-1.13-5.64-1.67l-9-24a7.35,7.35,0,0,0-12.25-2.44l-17.5,18.72c-1.95-.21-3.91-.38-5.88-.55L262.25,3.48a7.35,7.35,0,0,0-12.5,0L236.24,25.3c-2,.17-3.93.34-5.88.55L212.86,7.13a7.35,7.35,0,0,0-12.25,2.44l-9,24c-1.89.55-3.79,1.08-5.66,1.68l-20.82-15a7.35,7.35,0,0,0-11.54,4.79l-4.15,25.41c-1.73.9-3.45,1.79-5.16,2.73L120.88,42.55a7.35,7.35,0,0,0-10.39,7l.92,25.81c-1.49,1.19-3,2.39-4.42,3.61L81.84,73A7.36,7.36,0,0,0,73,81.84L78.93,107c-1.23,1.45-2.43,2.93-3.62,4.41l-25.81-.91a7.42,7.42,0,0,0-6.37,3.26,7.35,7.35,0,0,0-.57,7.13l10.66,23.41c-.94,1.7-1.83,3.43-2.73,5.16L25.08,153.6a7.35,7.35,0,0,0-4.79,11.54l15,20.82c-.59,1.87-1.13,3.77-1.68,5.66l-24,9a7.35,7.35,0,0,0-2.44,12.25l18.72,17.5c-.21,1.95-.38,3.91-.55,5.88L3.48,249.75a7.35,7.35,0,0,0,0,12.5L25.3,275.76c.17,2,.34,3.92.55,5.87L7.13,299.13a7.35,7.35,0,0,0,2.44,12.25l24,9c.55,1.89,1.08,3.78,1.68,5.65l-15,20.83a7.35,7.35,0,0,0,4.79,11.54l25.42,4.15c.9,1.72,1.79,3.45,2.73,5.14L42.56,391.12a7.35,7.35,0,0,0,.57,7.13,7.13,7.13,0,0,0,6.37,3.26l25.83-.91q1.77,2.22,3.6,4.4L73,430.16A7.36,7.36,0,0,0,81.84,439L107,433.07q2.18,1.83,4.41,3.61l-.92,25.82a7.35,7.35,0,0,0,10.39,6.95l23.43-10.68c1.69.94,3.42,1.83,5.14,2.73l4.15,25.42a7.34,7.34,0,0,0,11.54,4.78l20.83-15c1.86.6,3.76,1.13,5.65,1.68l9,24a7.36,7.36,0,0,0,12.25,2.44l17.5-18.72c1.95.21,3.92.38,5.88.55l13.51,21.82a7.35,7.35,0,0,0,12.5,0l13.51-21.82c2-.17,3.93-.34,5.88-.56l17.5,18.73a7.36,7.36,0,0,0,12.25-2.44l9-24c1.89-.55,3.78-1.08,5.65-1.68l20.82,15a7.34,7.34,0,0,0,11.54-4.78l4.15-25.42c1.72-.9,3.45-1.79,5.15-2.73l23.42,10.68a7.35,7.35,0,0,0,10.39-6.95l-.91-25.82q2.22-1.79,4.4-3.61L430.16,439a7.36,7.36,0,0,0,8.84-8.84L433.07,405q1.83-2.17,3.61-4.4l25.82.91a7.23,7.23,0,0,0,6.37-3.26,7.35,7.35,0,0,0,.58-7.13L458.77,367.7c.94-1.7,1.83-3.43,2.73-5.15l25.42-4.15a7.35,7.35,0,0,0,4.79-11.54l-15-20.83c.59-1.87,1.13-3.76,1.67-5.65l24-9a7.35,7.35,0,0,0,2.44-12.25l-18.72-17.5c.21-1.95.38-3.91.55-5.87l21.82-13.51a7.35,7.35,0,0,0,0-12.5Zm-151,129.08A13.91,13.91,0,0,0,341,389.51l-7.64,35.67A187.51,187.51,0,0,1,177,424.44l-7.64-35.66a13.87,13.87,0,0,0-16.46-10.68l-31.51,6.76a187.38,187.38,0,0,1-16.26-19.21H258.3c1.72,0,2.89-.29,2.89-1.91V309.55c0-1.57-1.17-1.91-2.89-1.91H213.47l.05-34.35H262c4.41,0,23.66,1.28,29.79,25.87,1.91,7.55,6.17,32.14,9.06,40,2.89,8.82,14.6,26.46,27.1,26.46H407a187.3,187.3,0,0,1-17.34,20.09Zm25.77,34.49A15.24,15.24,0,1,1,368,398.08h.44A15.23,15.23,0,0,1,383.24,413.32Zm-225.62-.68a15.24,15.24,0,1,1-15.25-15.25h.45A15.25,15.25,0,0,1,157.62,412.64ZM69.57,234.15l32.83-14.6a13.88,13.88,0,0,0,7.06-18.33L102.69,186h26.56V305.73H75.65A187.65,187.65,0,0,1,69.57,234.15ZM58.31,198.09a15.24,15.24,0,0,1,15.23-15.25H74a15.24,15.24,0,1,1-15.67,15.24Zm155.16,24.49.05-35.32h63.26c3.28,0,23.07,3.77,23.07,18.62,0,12.29-15.19,16.7-27.68,16.7ZM399,306.71c-9.8,1.13-20.63-4.12-22-10.09-5.78-32.49-15.39-39.4-30.57-51.4,18.86-11.95,38.46-29.64,38.46-53.26,0-25.52-17.49-41.59-29.4-49.48-16.76-11-35.28-13.23-40.27-13.23H116.32A187.49,187.49,0,0,1,221.21,70.06l23.47,24.6a13.82,13.82,0,0,0,19.6.44l26.26-25a187.51,187.51,0,0,1,128.37,91.43l-18,40.57A14,14,0,0,0,408,220.43l34.59,15.33a187.12,187.12,0,0,1,.4,32.54H423.71c-1.91,0-2.69,1.27-2.69,3.13v8.82C421,301,409.31,305.58,399,306.71ZM240,60.21A15.24,15.24,0,0,1,255.21,45h.45A15.24,15.24,0,1,1,240,60.21ZM436.84,214a15.24,15.24,0,1,1,0-30.48h.44a15.24,15.24,0,0,1-.44,30.48Z"],wix:[640,512,[],"f5cf","M393.38 131.69c0 13.03 2.08 32.69-28.68 43.83-9.52 3.45-15.95 9.66-15.95 9.66 0-31 4.72-42.22 17.4-48.86 9.75-5.11 27.23-4.63 27.23-4.63zm-115.8 35.54l-34.24 132.66-28.48-108.57c-7.69-31.99-20.81-48.53-48.43-48.53-27.37 0-40.66 16.18-48.43 48.53L89.52 299.89 55.28 167.23C49.73 140.51 23.86 128.96 0 131.96l65.57 247.93s21.63 1.56 32.46-3.96c14.22-7.25 20.98-12.84 29.59-46.57 7.67-30.07 29.11-118.41 31.12-124.7 4.76-14.94 11.09-13.81 15.4 0 1.97 6.3 23.45 94.63 31.12 124.7 8.6 33.73 15.37 39.32 29.59 46.57 10.82 5.52 32.46 3.96 32.46 3.96l65.57-247.93c-24.42-3.07-49.82 8.93-55.3 35.27zm115.78 5.21s-4.1 6.34-13.46 11.57c-6.01 3.36-11.78 5.64-17.97 8.61-15.14 7.26-13.18 13.95-13.18 35.2v152.07s16.55 2.09 27.37-3.43c13.93-7.1 17.13-13.95 17.26-44.78V181.41l-.02.01v-8.98zm163.44 84.08L640 132.78s-35.11-5.98-52.5 9.85c-13.3 12.1-24.41 29.55-54.18 72.47-.47.73-6.25 10.54-13.07 0-29.29-42.23-40.8-60.29-54.18-72.47-17.39-15.83-52.5-9.85-52.5-9.85l83.2 123.74-82.97 123.36s36.57 4.62 53.95-11.21c11.49-10.46 17.58-20.37 52.51-70.72 6.81-10.52 12.57-.77 13.07 0 29.4 42.38 39.23 58.06 53.14 70.72 17.39 15.83 53.32 11.21 53.32 11.21L556.8 256.52z"],"square-behance":[448,512,["behance-square"],"f1b5","M155.3 318.4c17.2 0 31.2-6.1 31.2-25.4c0-19.7-11.7-27.4-30.3-27.5h-46v52.9h45.1zm-5.4-129.6H110.3v44.8H153c15.1 0 25.8-6.6 25.8-22.9c0-17.7-13.7-21.9-28.9-21.9zm129.5 74.8h62.2c-1.7-18.5-11.3-29.7-30.5-29.7c-18.3 0-30.5 11.4-31.7 29.7zM384 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64zM349.5 185H271.7V166.1h77.8V185zM193.7 243.7c23.6 6.7 35 27.5 35 51.6c0 39-32.7 55.7-67.6 55.9H68v-192h90.5c32.9 0 61.4 9.3 61.4 47.5c0 19.3-9 28.8-26.2 37zm118.7-38.6c43.5 0 67.6 34.3 67.6 75.4c0 1.6-.1 3.3-.2 5c0 .8-.1 1.5-.1 2.2H279.5c0 22.2 11.7 35.3 34.1 35.3c11.6 0 26.5-6.2 30.2-18.1h33.7c-10.4 31.9-31.9 46.8-65.1 46.8c-43.8 0-71.1-29.7-71.1-73c0-41.8 28.7-73.6 71.1-73.6z"],supple:[640,512,[],"f3f9","M640 262.5c0 64.1-109 116.1-243.5 116.1-24.8 0-48.6-1.8-71.1-5 7.7.4 15.5.6 23.4.6 134.5 0 243.5-56.9 243.5-127.1 0-29.4-19.1-56.4-51.2-78 60 21.1 98.9 55.1 98.9 93.4zM47.7 227.9c-.1-70.2 108.8-127.3 243.3-127.6 7.9 0 15.6.2 23.3.5-22.5-3.2-46.3-4.9-71-4.9C108.8 96.3-.1 148.5 0 212.6c.1 38.3 39.1 72.3 99.3 93.3-32.3-21.5-51.5-48.6-51.6-78zm60.2 39.9s10.5 13.2 29.3 13.2c17.9 0 28.4-11.5 28.4-25.1 0-28-40.2-25.1-40.2-39.7 0-5.4 5.3-9.1 12.5-9.1 5.7 0 11.3 2.6 11.3 6.6v3.9h14.2v-7.9c0-12.1-15.4-16.8-25.4-16.8-16.5 0-28.5 10.2-28.5 24.1 0 26.6 40.2 25.4 40.2 39.9 0 6.6-5.8 10.1-12.3 10.1-11.9 0-20.7-10.1-20.7-10.1l-8.8 10.9zm120.8-73.6v54.4c0 11.3-7.1 17.8-17.8 17.8-10.7 0-17.8-6.5-17.8-17.7v-54.5h-15.8v55c0 18.9 13.4 31.9 33.7 31.9 20.1 0 33.4-13 33.4-31.9v-55h-15.7zm34.4 85.4h15.8v-29.5h15.5c16 0 27.2-11.5 27.2-28.1s-11.2-27.8-27.2-27.8h-39.1v13.4h7.8v72zm15.8-43v-29.1h12.9c8.7 0 13.7 5.7 13.7 14.4 0 8.9-5.1 14.7-14 14.7h-12.6zm57 43h15.8v-29.5h15.5c16 0 27.2-11.5 27.2-28.1s-11.2-27.8-27.2-27.8h-39.1v13.4h7.8v72zm15.7-43v-29.1h12.9c8.7 0 13.7 5.7 13.7 14.4 0 8.9-5 14.7-14 14.7h-12.6zm57.1 34.8c0 5.8 2.4 8.2 8.2 8.2h37.6c5.8 0 8.2-2.4 8.2-8.2v-13h-14.3v5.2c0 1.7-1 2.6-2.6 2.6h-18.6c-1.7 0-2.6-1-2.6-2.6v-61.2c0-5.7-2.4-8.2-8.2-8.2H401v13.4h5.2c1.7 0 2.6 1 2.6 2.6v61.2zm63.4 0c0 5.8 2.4 8.2 8.2 8.2H519c5.7 0 8.2-2.4 8.2-8.2v-13h-14.3v5.2c0 1.7-1 2.6-2.6 2.6h-19.7c-1.7 0-2.6-1-2.6-2.6v-20.3h27.7v-13.4H488v-22.4h19.2c1.7 0 2.6 1 2.6 2.6v5.2H524v-13c0-5.7-2.5-8.2-8.2-8.2h-51.6v13.4h7.8v63.9zm58.9-76v5.9h1.6v-5.9h2.7v-1.2h-7v1.2h2.7zm5.7-1.2v7.1h1.5v-5.7l2.3 5.7h1.3l2.3-5.7v5.7h1.5v-7.1h-2.3l-2.1 5.1-2.1-5.1h-2.4z"],webflow:[640,512,[],"e65c","M640 64L435.8 463.2H244l85.5-165.5h-3.8C255.1 389.3 149.9 449.5 0 463.2V300.1s95.9-5.7 152.3-64.9H0V64H171.1V204.8l3.8 0L244.9 64H374.3V203.9l3.8 0L450.7 64H640z"],rebel:[512,512,[],"f1d0","M256.5 504C117.2 504 9 387.8 13.2 249.9 16 170.7 56.4 97.7 129.7 49.5c.3 0 1.9-.6 1.1.8-5.8 5.5-111.3 129.8-14.1 226.4 49.8 49.5 90 2.5 90 2.5 38.5-50.1-.6-125.9-.6-125.9-10-24.9-45.7-40.1-45.7-40.1l28.8-31.8c24.4 10.5 43.2 38.7 43.2 38.7.8-29.6-21.9-61.4-21.9-61.4L255.1 8l44.3 50.1c-20.5 28.8-21.9 62.6-21.9 62.6 13.8-23 43.5-39.3 43.5-39.3l28.5 31.8c-27.4 8.9-45.4 39.9-45.4 39.9-15.8 28.5-27.1 89.4.6 127.3 32.4 44.6 87.7-2.8 87.7-2.8 102.7-91.9-10.5-225-10.5-225-6.1-5.5.8-2.8.8-2.8 50.1 36.5 114.6 84.4 116.2 204.8C500.9 400.2 399 504 256.5 504z"],css3:[512,512,[],"f13c","M480 32l-64 368-223.3 80L0 400l19.6-94.8h82l-8 40.6L210 390.2l134.1-44.4 18.8-97.1H29.5l16-82h333.7l10.5-52.7H56.3l16.3-82H480z"],staylinked:[440,512,[],"f3f5","M382.7 292.5l2.7 2.7-170-167.3c-3.5-3.5-9.7-3.7-13.8-.5L144.3 171c-4.2 3.2-4.6 8.7-1.1 12.2l68.1 64.3c3.6 3.5 9.9 3.7 14 .5l.1-.1c4.1-3.2 10.4-3 14 .5l84 81.3c3.6 3.5 3.2 9-.9 12.2l-93.2 74c-4.2 3.3-10.5 3.1-14.2-.4L63.2 268c-3.5-3.5-9.7-3.7-13.9-.5L3.5 302.4c-4.2 3.2-4.7 8.7-1.2 12.2L211 510.7s7.4 6.8 17.3-.8l198-163.9c4-3.2 4.4-8.7.7-12.2zm54.5-83.4L226.7 2.5c-1.5-1.2-8-5.5-16.3 1.1L3.6 165.7c-4.2 3.2-4.8 8.7-1.2 12.2l42.3 41.7 171.7 165.1c3.7 3.5 10.1 3.7 14.3.4l50.2-38.8-.3-.3 7.7-6c4.2-3.2 4.6-8.7.9-12.2l-57.1-54.4c-3.6-3.5-10-3.7-14.2-.5l-.1.1c-4.2 3.2-10.5 3.1-14.2-.4L109 180.8c-3.6-3.5-3.1-8.9 1.1-12.2l92.2-71.5c4.1-3.2 10.3-3 13.9.5l160.4 159c3.7 3.5 10 3.7 14.1.5l45.8-35.8c4.1-3.2 4.4-8.7.7-12.2z"],kaggle:[320,512,[],"f5fa","M304.2 501.5L158.4 320.3 298.2 185c2.6-2.7 1.7-10.5-5.3-10.5h-69.2c-3.5 0-7 1.8-10.5 5.3L80.9 313.5V7.5q0-7.5-7.5-7.5H21.5Q14 0 14 7.5v497q0 7.5 7.5 7.5h51.9q7.5 0 7.5-7.5v-109l30.8-29.3 110.5 140.6c3 3.5 6.5 5.3 10.5 5.3h66.9q5.25 0 6-3z"],"space-awesome":[512,512,[],"e5ac","M96 256H128V512H0V352H32V320H64V288H96V256zM512 352V512H384V256H416V288H448V320H480V352H512zM320 64H352V448H320V416H192V448H160V64H192V32H224V0H288V32H320V64zM288 128H224V192H288V128z"],deviantart:[320,512,[],"f1bd","M320 93.2l-98.2 179.1 7.4 9.5H320v127.7H159.1l-13.5 9.2-43.7 84c-.3 0-8.6 8.6-9.2 9.2H0v-93.2l93.2-179.4-7.4-9.2H0V102.5h156l13.5-9.2 43.7-84c.3 0 8.6-8.6 9.2-9.2H320v93.1z"],cpanel:[640,512,[],"f388","M210.3 220.2c-5.6-24.8-26.9-41.2-51-41.2h-37c-7.1 0-12.5 4.5-14.3 10.9L73.1 320l24.7-.1c6.8 0 12.3-4.5 14.2-10.7l25.8-95.7h19.8c8.4 0 16.2 5.6 18.3 14.8 2.5 10.9-5.9 22.6-18.3 22.6h-10.3c-7 0-12.5 4.6-14.3 10.8l-6.4 23.8h32c37.2 0 58.3-36.2 51.7-65.3zm-156.5 28h18.6c6.9 0 12.4-4.4 14.3-10.9l6.2-23.6h-40C30 213.7 9 227.8 1.7 254.8-7 288.6 18.5 320 52 320h12.4l7.1-26.1c1.2-4.4-2.2-8.3-6.4-8.3H53.8c-24.7 0-24.9-37.4 0-37.4zm247.5-34.8h-77.9l-3.5 13.4c-2.4 9.6 4.5 18.5 14.2 18.5h57.5c4 0 2.4 4.3 2.1 5.3l-8.6 31.8c-.4 1.4-.9 5.3-5.5 5.3h-34.9c-5.3 0-5.3-7.9 0-7.9h21.6c6.8 0 12.3-4.6 14.2-10.8l3.5-13.2h-48.4c-39.2 0-43.6 63.8-.7 63.8l57.5.2c11.2 0 20.6-7.2 23.4-17.8l14-51.8c4.8-19.2-9.7-36.8-28.5-36.8zM633.1 179h-18.9c-4.9 0-9.2 3.2-10.4 7.9L568.2 320c20.7 0 39.8-13.8 44.9-34.5l26.5-98.2c1.2-4.3-2-8.3-6.5-8.3zm-236.3 34.7v.1h-48.3l-26.2 98c-1.2 4.4 2.2 8.3 6.4 8.3h18.9c4.8 0 9.2-3 10.4-7.8l17.2-64H395c12.5 0 21.4 11.8 18.1 23.4l-10.6 40c-1.2 4.3 1.9 8.3 6.4 8.3H428c4.6 0 9.1-2.9 10.3-7.8l8.8-33.1c9-33.1-15.9-65.4-50.3-65.4zm98.3 74.6c-3.6 0-6-3.4-5.1-6.7l8-30c.9-3.9 3.7-6 7.8-6h32.9c2.6 0 4.6 2.4 3.9 5.1l-.7 2.6c-.6 2-1.9 3-3.9 3h-21.6c-7 0-12.6 4.6-14.2 10.8l-3.5 13h53.4c10.5 0 20.3-6.6 23.2-17.6l3.2-12c4.9-19.1-9.3-36.8-28.3-36.8h-47.3c-17.9 0-33.8 12-38.6 29.6l-10.8 40c-5 17.7 8.3 36.7 28.3 36.7h66.7c6.8 0 12.3-4.5 14.2-10.7l5.7-21z"],"goodreads-g":[384,512,[],"f3a9","M42.6 403.3h2.8c12.7 0 25.5 0 38.2.1 1.6 0 3.1-.4 3.6 2.1 7.1 34.9 30 54.6 62.9 63.9 26.9 7.6 54.1 7.8 81.3 1.8 33.8-7.4 56-28.3 68-60.4 8-21.5 10.7-43.8 11-66.5.1-5.8.3-47-.2-52.8l-.9-.3c-.8 1.5-1.7 2.9-2.5 4.4-22.1 43.1-61.3 67.4-105.4 69.1-103 4-169.4-57-172-176.2-.5-23.7 1.8-46.9 8.3-69.7C58.3 47.7 112.3.6 191.6 0c61.3-.4 101.5 38.7 116.2 70.3.5 1.1 1.3 2.3 2.4 1.9V10.6h44.3c0 280.3.1 332.2.1 332.2-.1 78.5-26.7 143.7-103 162.2-69.5 16.9-159 4.8-196-57.2-8-13.5-11.8-28.3-13-44.5zM188.9 36.5c-52.5-.5-108.5 40.7-115 133.8-4.1 59 14.8 122.2 71.5 148.6 27.6 12.9 74.3 15 108.3-8.7 47.6-33.2 62.7-97 54.8-154-9.7-71.1-47.8-120-119.6-119.7z"],"square-git":[448,512,["git-square"],"f1d2","M120.8 335.5c-5.9-.4-12.6-.8-20.2-1.3c-3.3 4.1-6.6 8.4-6.6 13.5c0 18.5 65.5 18.5 65.5-1.5c0-8.3-7.4-8.7-38.8-10.7zm7.8-117.9c-32.3 0-33.7 44.5-.7 44.5c32.5 0 31.7-44.5 .7-44.5zM384 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64zM243.9 172.2c-14.5 0-22.9-8.4-22.9-22.9c0-14.5 8.4-22.3 22.9-22.3c14.7 0 23.1 7.8 23.1 22.3s-8.4 22.9-23.1 22.9zM149.6 195h49.5l0 21.6-23.4 1.8c4.6 5.8 9.4 14 9.4 25.7c0 48.7-57.2 47.2-74.2 42.4l-8.4 13.4c5 .3 9.8 .6 14.3 .8c56.3 3.2 80.5 4.6 80.5 38.5c0 29.2-25.7 45.7-69.9 45.7c-46 0-63.5-11.6-63.5-31.7c0-11.4 5.1-17.5 14-25.9c-8.4-3.5-11.2-9.9-11.2-16.8c0-9.6 7.4-16.3 23-30.6l.2-.2c-12.4-6.1-21.8-19.3-21.8-38.1c0-51.6 56.6-53.3 81.6-46.8zM270.5 303.1l13 1.8 0 20.1H211.1V304.9c2.7-.4 5-.7 6.9-.9c9.9-1.2 10.1-1.3 10.1-6V223.3c0-4.4-.9-4.7-10.1-7.8c-1.9-.7-4.2-1.4-6.9-2.4l2.8-20.6h52.6V298c0 4.1 .2 4.6 4.1 5.1zm106.6-10.4L384 315c-10.9 5.4-26.9 10.2-41.4 10.2c-30.2 0-41.7-12.2-41.7-40.9V217.7c0-.8 0-1.4-.2-1.8c-.8-1.2-4.2-.7-19.6-.7V192.6c22.3-2.5 31.2-13.7 34-41.4h24.2c0 33.3-.6 38 .7 38.6c.3 .1 .7 0 1.3 0h35.8v25.4H339.3v60.7c0 .2 0 .5 0 .9c-.2 6.3-.9 30.4 37.9 15.9z"],"square-tumblr":[448,512,["tumblr-square"],"f174","M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM256.8 416c-75.5 0-91.9-55.5-91.9-87.9v-90H135.2c-3.4 0-6.2-2.8-6.2-6.2V189.4c0-4.5 2.8-8.5 7.1-10c38.8-13.7 50.9-47.5 52.7-73.2c.5-6.9 4.1-10.2 10-10.2h44.3c3.4 0 6.2 2.8 6.2 6.2v72h51.9c3.4 0 6.2 2.8 6.2 6.2v51.1c0 3.4-2.8 6.2-6.2 6.2H249.1V321c0 21.4 14.8 33.5 42.5 22.4c3-1.2 5.6-2 8-1.4c2.2 .5 3.6 2.1 4.6 4.9L318 387.1c1 3.2 2 6.7-.3 9.1c-8.5 9.1-31.2 19.8-60.9 19.8z"],trello:[448,512,[],"f181","M392.3 32H56.1C25.1 32 0 57.1 0 88c-.1 0 0-4 0 336 0 30.9 25.1 56 56 56h336.2c30.8-.2 55.7-25.2 55.7-56V88c.1-30.8-24.8-55.8-55.6-56zM197 371.3c-.2 14.7-12.1 26.6-26.9 26.6H87.4c-14.8.1-26.9-11.8-27-26.6V117.1c0-14.8 12-26.9 26.9-26.9h82.9c14.8 0 26.9 12 26.9 26.9v254.2zm193.1-112c0 14.8-12 26.9-26.9 26.9h-81c-14.8 0-26.9-12-26.9-26.9V117.2c0-14.8 12-26.9 26.8-26.9h81.1c14.8 0 26.9 12 26.9 26.9v142.1z"],"creative-commons-nc-jp":[496,512,[],"f4ea","M247.7 8C103.6 8 0 124.8 0 256c0 136.4 111.8 248 247.7 248C377.9 504 496 403.2 496 256 496 117.2 388.5 8 247.7 8zm.6 450.7c-112 0-203.6-92.5-203.6-202.7 0-21.1 3-41.2 9-60.3l127 56.5h-27.9v38.6h58.1l5.7 11.8v18.7h-63.8V360h63.8v56h61.7v-56h64.2v-35.7l81 36.1c-1.5 2.2-57.1 98.3-175.2 98.3zm87.6-137.3h-57.6v-18.7l2.9-5.6 54.7 24.3zm6.5-51.4v-17.8h-38.6l63-116H301l-43.4 96-23-10.2-39.6-85.7h-65.8l27.3 51-81.9-36.5c27.8-44.1 82.6-98.1 173.7-98.1 112.8 0 203 90 203 203.4 0 21-2.7 40.6-7.9 59l-101-45.1z"],"get-pocket":[448,512,[],"f265","M407.6 64h-367C18.5 64 0 82.5 0 104.6v135.2C0 364.5 99.7 464 224.2 464c124 0 223.8-99.5 223.8-224.2V104.6c0-22.4-17.7-40.6-40.4-40.6zm-162 268.5c-12.4 11.8-31.4 11.1-42.4 0C89.5 223.6 88.3 227.4 88.3 209.3c0-16.9 13.8-30.7 30.7-30.7 17 0 16.1 3.8 105.2 89.3 90.6-86.9 88.6-89.3 105.5-89.3 16.9 0 30.7 13.8 30.7 30.7 0 17.8-2.9 15.7-114.8 123.2z"],perbyte:[448,512,[],"e083","M305.314,284.578H246.6V383.3h58.711q24.423,0,38.193-13.77t13.77-36.11q0-21.826-14.032-35.335T305.314,284.578ZM149.435,128.7H90.724v98.723h58.711q24.42,0,38.19-13.773t13.77-36.107q0-21.826-14.029-35.338T149.435,128.7ZM366.647,32H81.353A81.445,81.445,0,0,0,0,113.352V398.647A81.445,81.445,0,0,0,81.353,480H366.647A81.445,81.445,0,0,0,448,398.647V113.352A81.445,81.445,0,0,0,366.647,32Zm63.635,366.647a63.706,63.706,0,0,1-63.635,63.635H81.353a63.706,63.706,0,0,1-63.635-63.635V113.352A63.706,63.706,0,0,1,81.353,49.718H366.647a63.706,63.706,0,0,1,63.635,63.634ZM305.314,128.7H246.6v98.723h58.711q24.423,0,38.193-13.773t13.77-36.107q0-21.826-14.032-35.338T305.314,128.7Z"],grunt:[384,512,[],"f3ad","M61.3 189.3c-1.1 10 5.2 19.1 5.2 19.1.7-7.5 2.2-12.8 4-16.6.4 10.3 3.2 23.5 12.8 34.1 6.9 7.6 35.6 23.3 54.9 6.1 1 2.4 2.1 5.3 3 8.5 2.9 10.3-2.7 25.3-2.7 25.3s15.1-17.1 13.9-32.5c10.8-.5 21.4-8.4 21.1-19.5 0 0-18.9 10.4-35.5-8.8-9.7-11.2-40.9-42-83.1-31.8 4.3 1 8.9 2.4 13.5 4.1h-.1c-4.2 2-6.5 7.1-7 12zm28.3-1.8c19.5 11 37.4 25.7 44.9 37-5.7 3.3-21.7 10.4-38-1.7-10.3-7.6-9.8-26.2-6.9-35.3zm142.1 45.8c-1.2 15.5 13.9 32.5 13.9 32.5s-5.6-15-2.7-25.3c.9-3.2 2-6 3-8.5 19.3 17.3 48 1.5 54.8-6.1 9.6-10.6 12.3-23.8 12.8-34.1 1.8 3.8 3.4 9.1 4 16.6 0 0 6.4-9.1 5.2-19.1-.6-5-2.9-10-7-11.8h-.1c4.6-1.8 9.2-3.2 13.5-4.1-42.3-10.2-73.4 20.6-83.1 31.8-16.7 19.2-35.5 8.8-35.5 8.8-.2 10.9 10.4 18.9 21.2 19.3zm62.7-45.8c3 9.1 3.4 27.7-7 35.4-16.3 12.1-32.2 5-37.9 1.6 7.5-11.4 25.4-26 44.9-37zM160 418.5h-29.4c-5.5 0-8.2 1.6-9.5 2.9-1.9 2-2.2 4.7-.9 8.1 3.5 9.1 11.4 16.5 13.7 18.6 3.1 2.7 7.5 4.3 11.8 4.3 4.4 0 8.3-1.7 11-4.6 7.5-8.2 11.9-17.1 13-19.8.6-1.5 1.3-4.5-.9-6.8-1.8-1.8-4.7-2.7-8.8-2.7zm189.2-101.2c-2.4 17.9-13 33.8-24.6 43.7-3.1-22.7-3.7-55.5-3.7-62.4 0-14.7 9.5-24.5 12.2-26.1 2.5-1.5 5.4-3 8.3-4.6 18-9.6 40.4-21.6 40.4-43.7 0-16.2-9.3-23.2-15.4-27.8-.8-.6-1.5-1.1-2.2-1.7-2.1-1.7-3.7-3-4.3-4.4-4.4-9.8-3.6-34.2-1.7-37.6.6-.6 16.7-20.9 11.8-39.2-2-7.4-6.9-13.3-14.1-17-5.3-2.7-11.9-4.2-19.5-4.5-.1-2-.5-3.9-.9-5.9-.6-2.6-1.1-5.3-.9-8.1.4-4.7.8-9 2.2-11.3 8.4-13.3 28.8-17.6 29-17.6l12.3-2.4-8.1-9.5c-.1-.2-17.3-17.5-46.3-17.5-7.9 0-16 1.3-24.1 3.9-24.2 7.8-42.9 30.5-49.4 39.3-3.1-1-6.3-1.9-9.6-2.7-4.2-15.8 9-38.5 9-38.5s-13.6-3-33.7 15.2c-2.6-6.5-8.1-20.5-1.8-37.2C184.6 10.1 177.2 26 175 40.4c-7.6-5.4-6.7-23.1-7.2-27.6-7.5.9-29.2 21.9-28.2 48.3-2 .5-3.9 1.1-5.9 1.7-6.5-8.8-25.1-31.5-49.4-39.3-7.9-2.2-16-3.5-23.9-3.5-29 0-46.1 17.3-46.3 17.5L6 46.9l12.3 2.4c.2 0 20.6 4.3 29 17.6 1.4 2.2 1.8 6.6 2.2 11.3.2 2.8-.4 5.5-.9 8.1-.4 1.9-.8 3.9-.9 5.9-7.7.3-14.2 1.8-19.5 4.5-7.2 3.7-12.1 9.6-14.1 17-5 18.2 11.2 38.5 11.8 39.2 1.9 3.4 2.7 27.8-1.7 37.6-.6 1.4-2.2 2.7-4.3 4.4-.7.5-1.4 1.1-2.2 1.7-6.1 4.6-15.4 11.7-15.4 27.8 0 22.1 22.4 34.1 40.4 43.7 3 1.6 5.8 3.1 8.3 4.6 2.7 1.6 12.2 11.4 12.2 26.1 0 6.9-.6 39.7-3.7 62.4-11.6-9.9-22.2-25.9-24.6-43.8 0 0-29.2 22.6-20.6 70.8 5.2 29.5 23.2 46.1 47 54.7 8.8 19.1 29.4 45.7 67.3 49.6C143 504.3 163 512 192.2 512h.2c29.1 0 49.1-7.7 63.6-19.5 37.9-3.9 58.5-30.5 67.3-49.6 23.8-8.7 41.7-25.2 47-54.7 8.2-48.4-21.1-70.9-21.1-70.9zM305.7 37.7c5.6-1.8 11.6-2.7 17.7-2.7 11 0 19.9 3 24.7 5-3.1 1.4-6.4 3.2-9.7 5.3-2.4-.4-5.6-.8-9.2-.8-10.5 0-20.5 3.1-28.7 8.9-12.3 8.7-18 16.9-20.7 22.4-2.2-1.3-4.5-2.5-7.1-3.7-1.6-.8-3.1-1.5-4.7-2.2 6.1-9.1 19.9-26.5 37.7-32.2zm21 18.2c-.8 1-1.6 2.1-2.3 3.2-3.3 5.2-3.9 11.6-4.4 17.8-.5 6.4-1.1 12.5-4.4 17-4.2.8-8.1 1.7-11.5 2.7-2.3-3.1-5.6-7-10.5-11.2 1.4-4.8 5.5-16.1 13.5-22.5 5.6-4.3 12.2-6.7 19.6-7zM45.6 45.3c-3.3-2.2-6.6-4-9.7-5.3 4.8-2 13.7-5 24.7-5 6.1 0 12 .9 17.7 2.7 17.8 5.8 31.6 23.2 37.7 32.1-1.6.7-3.2 1.4-4.8 2.2-2.5 1.2-4.9 2.5-7.1 3.7-2.6-5.4-8.3-13.7-20.7-22.4-8.3-5.8-18.2-8.9-28.8-8.9-3.4.1-6.6.5-9 .9zm44.7 40.1c-4.9 4.2-8.3 8-10.5 11.2-3.4-.9-7.3-1.9-11.5-2.7C65 89.5 64.5 83.4 64 77c-.5-6.2-1.1-12.6-4.4-17.8-.7-1.1-1.5-2.2-2.3-3.2 7.4.3 14 2.6 19.5 7 8 6.3 12.1 17.6 13.5 22.4zM58.1 259.9c-2.7-1.6-5.6-3.1-8.4-4.6-14.9-8-30.2-16.3-30.2-30.5 0-11.1 4.3-14.6 8.9-18.2l.5-.4c.7-.6 1.4-1.2 2.2-1.8-.9 7.2-1.9 13.3-2.7 14.9 0 0 12.1-15 15.7-44.3 1.4-11.5-1.1-34.3-5.1-43 .2 4.9 0 9.8-.3 14.4-.4-.8-.8-1.6-1.3-2.2-3.2-4-11.8-17.5-9.4-26.6.9-3.5 3.1-6 6.7-7.8 3.8-1.9 8.8-2.9 15.1-2.9 12.3 0 25.9 3.7 32.9 6 25.1 8 55.4 30.9 64.1 37.7.2.2.4.3.4.3l5.6 3.9-3.5-5.8c-.2-.3-19.1-31.4-53.2-46.5 2-2.9 7.4-8.1 21.6-15.1 21.4-10.5 46.5-15.8 74.3-15.8 27.9 0 52.9 5.3 74.3 15.8 14.2 6.9 19.6 12.2 21.6 15.1-34 15.1-52.9 46.2-53.1 46.5l-3.5 5.8 5.6-3.9s.2-.1.4-.3c8.7-6.8 39-29.8 64.1-37.7 7-2.2 20.6-6 32.9-6 6.3 0 11.3 1 15.1 2.9 3.5 1.8 5.7 4.4 6.7 7.8 2.5 9.1-6.1 22.6-9.4 26.6-.5.6-.9 1.3-1.3 2.2-.3-4.6-.5-9.5-.3-14.4-4 8.8-6.5 31.5-5.1 43 3.6 29.3 15.7 44.3 15.7 44.3-.8-1.6-1.8-7.7-2.7-14.9.7.6 1.5 1.2 2.2 1.8l.5.4c4.6 3.7 8.9 7.1 8.9 18.2 0 14.2-15.4 22.5-30.2 30.5-2.9 1.5-5.7 3.1-8.4 4.6-8.7 5-18 16.7-19.1 34.2-.9 14.6.9 49.9 3.4 75.9-12.4 4.8-26.7 6.4-39.7 6.8-2-4.1-3.9-8.5-5.5-13.1-.7-2-19.6-51.1-26.4-62.2 5.5 39 17.5 73.7 23.5 89.6-3.5-.5-7.3-.7-11.7-.7h-117c-4.4 0-8.3.3-11.7.7 6-15.9 18.1-50.6 23.5-89.6-6.8 11.2-25.7 60.3-26.4 62.2-1.6 4.6-3.5 9-5.5 13.1-13-.4-27.2-2-39.7-6.8 2.5-26 4.3-61.2 3.4-75.9-.9-17.4-10.3-29.2-19-34.2zM34.8 404.6c-12.1-20-8.7-54.1-3.7-59.1 10.9 34.4 47.2 44.3 74.4 45.4-2.7 4.2-5.2 7.6-7 10l-1.4 1.4c-7.2 7.8-8.6 18.5-4.1 31.8-22.7-.1-46.3-9.8-58.2-29.5zm45.7 43.5c6 1.1 12.2 1.9 18.6 2.4 3.5 8 7.4 15.9 12.3 23.1-14.4-5.9-24.4-16-30.9-25.5zM192 498.2c-60.6-.1-78.3-45.8-84.9-64.7-3.7-10.5-3.4-18.2.9-23.1 2.9-3.3 9.5-7.2 24.6-7.2h118.8c15.1 0 21.8 3.9 24.6 7.2 4.2 4.8 4.5 12.6.9 23.1-6.6 18.8-24.3 64.6-84.9 64.7zm80.6-24.6c4.9-7.2 8.8-15.1 12.3-23.1 6.4-.5 12.6-1.3 18.6-2.4-6.5 9.5-16.5 19.6-30.9 25.5zm76.6-69c-12 19.7-35.6 29.3-58.1 29.7 4.5-13.3 3.1-24.1-4.1-31.8-.4-.5-.9-1-1.4-1.5-1.8-2.4-4.3-5.8-7-10 27.2-1.2 63.5-11 74.4-45.4 5 5 8.4 39.1-3.8 59zM191.9 187.7h.2c12.7-.1 27.2-17.8 27.2-17.8-9.9 6-18.8 8.1-27.3 8.3-8.5-.2-17.4-2.3-27.3-8.3 0 0 14.5 17.6 27.2 17.8zm61.7 230.7h-29.4c-4.2 0-7.2.9-8.9 2.7-2.2 2.3-1.5 5.2-.9 6.7 1 2.6 5.5 11.3 13 19.3 2.7 2.9 6.6 4.5 11 4.5s8.7-1.6 11.8-4.2c2.3-2 10.2-9.2 13.7-18.1 1.3-3.3 1-6-.9-7.9-1.3-1.3-4-2.9-9.4-3z"],weebly:[512,512,[],"f5cc","M425.09 65.83c-39.88 0-73.28 25.73-83.66 64.33-18.16-58.06-65.5-64.33-84.95-64.33-19.78 0-66.8 6.28-85.28 64.33-10.38-38.6-43.45-64.33-83.66-64.33C38.59 65.83 0 99.72 0 143.03c0 28.96 4.18 33.27 77.17 233.48 22.37 60.57 67.77 69.35 92.74 69.35 39.23 0 70.04-19.46 85.93-53.98 15.89 34.83 46.69 54.29 85.93 54.29 24.97 0 70.36-9.1 92.74-69.67 76.55-208.65 77.5-205.58 77.5-227.2.63-48.32-36.01-83.47-86.92-83.47zm26.34 114.81l-65.57 176.44c-7.92 21.49-21.22 37.22-46.24 37.22-23.44 0-37.38-12.41-44.03-33.9l-39.28-117.42h-.95L216.08 360.4c-6.96 21.5-20.9 33.6-44.02 33.6-25.02 0-38.33-15.74-46.24-37.22L60.88 181.55c-5.38-14.83-7.92-23.91-7.92-34.5 0-16.34 15.84-29.36 38.33-29.36 18.69 0 31.99 11.8 36.11 29.05l44.03 139.82h.95l44.66-136.79c6.02-19.67 16.47-32.08 38.96-32.08s32.94 12.11 38.96 32.08l44.66 136.79h.95l44.03-139.82c4.12-17.25 17.42-29.05 36.11-29.05 22.17 0 38.33 13.32 38.33 35.71-.32 7.87-4.12 16.04-7.61 27.24z"],connectdevelop:[576,512,[],"f20e","M550.5 241l-50.089-86.786c1.071-2.142 1.875-4.553 1.875-7.232 0-8.036-6.696-14.733-14.732-15.001l-55.447-95.893c.536-1.607 1.071-3.214 1.071-4.821 0-8.571-6.964-15.268-15.268-15.268-4.821 0-8.839 2.143-11.786 5.625H299.518C296.839 18.143 292.821 16 288 16s-8.839 2.143-11.518 5.625H170.411C167.464 18.143 163.447 16 158.625 16c-8.303 0-15.268 6.696-15.268 15.268 0 1.607.536 3.482 1.072 4.821l-55.983 97.233c-5.356 2.41-9.107 7.5-9.107 13.661 0 .535.268 1.071.268 1.607l-53.304 92.143c-7.232 1.339-12.59 7.5-12.59 15 0 7.232 5.089 13.393 12.054 15l55.179 95.358c-.536 1.607-.804 2.946-.804 4.821 0 7.232 5.089 13.393 12.054 14.732l51.697 89.732c-.536 1.607-1.071 3.482-1.071 5.357 0 8.571 6.964 15.268 15.268 15.268 4.821 0 8.839-2.143 11.518-5.357h106.875C279.161 493.857 283.447 496 288 496s8.839-2.143 11.518-5.357h107.143c2.678 2.946 6.696 4.821 10.982 4.821 8.571 0 15.268-6.964 15.268-15.268 0-1.607-.267-2.946-.803-4.285l51.697-90.268c6.964-1.339 12.054-7.5 12.054-14.732 0-1.607-.268-3.214-.804-4.821l54.911-95.358c6.964-1.339 12.322-7.5 12.322-15-.002-7.232-5.092-13.393-11.788-14.732zM153.535 450.732l-43.66-75.803h43.66v75.803zm0-83.839h-43.66c-.268-1.071-.804-2.142-1.339-3.214l44.999-47.41v50.624zm0-62.411l-50.357 53.304c-1.339-.536-2.679-1.34-4.018-1.607L43.447 259.75c.535-1.339.535-2.679.535-4.018s0-2.41-.268-3.482l51.965-90c2.679-.268 5.357-1.072 7.768-2.679l50.089 51.965v92.946zm0-102.322l-45.803-47.41c1.339-2.143 2.143-4.821 2.143-7.767 0-.268-.268-.804-.268-1.072l43.928-15.804v72.053zm0-80.625l-43.66 15.804 43.66-75.536v59.732zm326.519 39.108l.804 1.339L445.5 329.125l-63.75-67.232 98.036-101.518.268.268zM291.75 355.107l11.518 11.786H280.5l11.25-11.786zm-.268-11.25l-83.303-85.446 79.553-84.375 83.036 87.589-79.286 82.232zm5.357 5.893l79.286-82.232 67.5 71.25-5.892 28.125H313.714l-16.875-17.143zM410.411 44.393c1.071.536 2.142 1.072 3.482 1.34l57.857 100.714v.536c0 2.946.803 5.624 2.143 7.767L376.393 256l-83.035-87.589L410.411 44.393zm-9.107-2.143L287.732 162.518l-57.054-60.268 166.339-60h4.287zm-123.483 0c2.678 2.678 6.16 4.285 10.179 4.285s7.5-1.607 10.179-4.285h75L224.786 95.821 173.893 42.25h103.928zm-116.249 5.625l1.071-2.142a33.834 33.834 0 0 0 2.679-.804l51.161 53.84-54.911 19.821V47.875zm0 79.286l60.803-21.964 59.732 63.214-79.553 84.107-40.982-42.053v-83.304zm0 92.678L198 257.607l-36.428 38.304v-76.072zm0 87.858l42.053-44.464 82.768 85.982-17.143 17.678H161.572v-59.196zm6.964 162.053c-1.607-1.607-3.482-2.678-5.893-3.482l-1.071-1.607v-89.732h99.91l-91.607 94.821h-1.339zm129.911 0c-2.679-2.41-6.428-4.285-10.447-4.285s-7.767 1.875-10.447 4.285h-96.429l91.607-94.821h38.304l91.607 94.821H298.447zm120-11.786l-4.286 7.5c-1.339.268-2.41.803-3.482 1.339l-89.196-91.875h114.376l-17.412 83.036zm12.856-22.232l12.858-60.803h21.964l-34.822 60.803zm34.822-68.839h-20.357l4.553-21.16 17.143 18.214c-.535.803-1.071 1.874-1.339 2.946zm66.161-107.411l-55.447 96.697c-1.339.535-2.679 1.071-4.018 1.874l-20.625-21.964 34.554-163.928 45.803 79.286c-.267 1.339-.803 2.678-.803 4.285 0 1.339.268 2.411.536 3.75z"],leanpub:[576,512,[],"f212","M386.539 111.485l15.096 248.955-10.979-.275c-36.232-.824-71.64 8.783-102.657 27.997-31.016-19.214-66.424-27.997-102.657-27.997-45.564 0-82.07 10.705-123.516 27.723L93.117 129.6c28.546-11.803 61.484-18.115 92.226-18.115 41.173 0 73.836 13.175 102.657 42.544 27.723-28.271 59.013-41.721 98.539-42.544zM569.07 448c-25.526 0-47.485-5.215-70.542-15.645-34.31-15.645-69.993-24.978-107.871-24.978-38.977 0-74.934 12.901-102.657 40.623-27.723-27.723-63.68-40.623-102.657-40.623-37.878 0-73.561 9.333-107.871 24.978C55.239 442.236 32.731 448 8.303 448H6.93L49.475 98.859C88.726 76.626 136.486 64 181.775 64 218.83 64 256.984 71.685 288 93.095 319.016 71.685 357.17 64 394.225 64c45.289 0 93.049 12.626 132.3 34.859L569.07 448zm-43.368-44.741l-34.036-280.246c-30.742-13.999-67.248-21.41-101.009-21.41-38.428 0-74.385 12.077-102.657 38.702-28.272-26.625-64.228-38.702-102.657-38.702-33.761 0-70.267 7.411-101.009 21.41L50.298 403.259c47.211-19.487 82.894-33.486 135.045-33.486 37.604 0 70.817 9.606 102.657 29.644 31.84-20.038 65.052-29.644 102.657-29.644 52.151 0 87.834 13.999 135.045 33.486z"],"black-tie":[448,512,[],"f27e","M0 32v448h448V32H0zm316.5 325.2L224 445.9l-92.5-88.7 64.5-184-64.5-86.6h184.9L252 173.2l64.5 184z"],themeco:[448,512,[],"f5c6","M202.9 8.43c9.9-5.73 26-5.82 35.95-.21L430 115.85c10 5.6 18 19.44 18 30.86V364c0 11.44-8.06 25.29-18 31L238.81 503.74c-9.93 5.66-26 5.57-35.85-.21L17.86 395.12C8 389.34 0 375.38 0 364V146.71c0-11.44 8-25.36 17.91-31.08zm-77.4 199.83c-15.94 0-31.89.14-47.83.14v101.45H96.8V280h28.7c49.71 0 49.56-71.74 0-71.74zm140.14 100.29l-30.73-34.64c37-7.51 34.8-65.23-10.87-65.51-16.09 0-32.17-.14-48.26-.14v101.59h19.13v-33.91h18.41l29.56 33.91h22.76zm-41.59-82.32c23.34 0 23.26 32.46 0 32.46h-29.13v-32.46zm-95.56-1.6c21.18 0 21.11 38.85 0 38.85H96.18v-38.84zm192.65-18.25c-68.46 0-71 105.8 0 105.8 69.48-.01 69.41-105.8 0-105.8zm0 17.39c44.12 0 44.8 70.86 0 70.86s-44.43-70.86 0-70.86z"],python:[448,512,[],"f3e2","M439.8 200.5c-7.7-30.9-22.3-54.2-53.4-54.2h-40.1v47.4c0 36.8-31.2 67.8-66.8 67.8H172.7c-29.2 0-53.4 25-53.4 54.3v101.8c0 29 25.2 46 53.4 54.3 33.8 9.9 66.3 11.7 106.8 0 26.9-7.8 53.4-23.5 53.4-54.3v-40.7H226.2v-13.6h160.2c31.1 0 42.6-21.7 53.4-54.2 11.2-33.5 10.7-65.7 0-108.6zM286.2 404c11.1 0 20.1 9.1 20.1 20.3 0 11.3-9 20.4-20.1 20.4-11 0-20.1-9.2-20.1-20.4.1-11.3 9.1-20.3 20.1-20.3zM167.8 248.1h106.8c29.7 0 53.4-24.5 53.4-54.3V91.9c0-29-24.4-50.7-53.4-55.6-35.8-5.9-74.7-5.6-106.8.1-45.2 8-53.4 24.7-53.4 55.6v40.7h106.9v13.6h-147c-31.1 0-58.3 18.7-66.8 54.2-9.8 40.7-10.2 66.1 0 108.6 7.6 31.6 25.7 54.2 56.8 54.2H101v-48.8c0-35.3 30.5-66.4 66.8-66.4zm-6.7-142.6c-11.1 0-20.1-9.1-20.1-20.3.1-11.3 9-20.4 20.1-20.4 11 0 20.1 9.2 20.1 20.4s-9 20.3-20.1 20.3z"],android:[576,512,[],"f17b","M420.55,301.93a24,24,0,1,1,24-24,24,24,0,0,1-24,24m-265.1,0a24,24,0,1,1,24-24,24,24,0,0,1-24,24m273.7-144.48,47.94-83a10,10,0,1,0-17.27-10h0l-48.54,84.07a301.25,301.25,0,0,0-246.56,0L116.18,64.45a10,10,0,1,0-17.27,10h0l47.94,83C64.53,202.22,8.24,285.55,0,384H576c-8.24-98.45-64.54-181.78-146.85-226.55"],bots:[640,512,[],"e340","M86.344,197.834a51.767,51.767,0,0,0-41.57,20.058V156.018a8.19,8.19,0,0,0-8.19-8.19H8.19A8.19,8.19,0,0,0,0,156.018V333.551a8.189,8.189,0,0,0,8.19,8.189H36.584a8.189,8.189,0,0,0,8.19-8.189v-8.088c11.628,13.373,25.874,19.769,41.573,19.769,34.6,0,61.922-26.164,61.922-73.843C148.266,225.452,121.229,197.834,86.344,197.834ZM71.516,305.691c-9.593,0-21.221-4.942-26.745-12.5V250.164c5.528-7.558,17.152-12.791,26.745-12.791,17.734,0,31.107,13.082,31.107,34.013C102.623,292.609,89.25,305.691,71.516,305.691Zm156.372-59.032a17.4,17.4,0,1,0,17.4,17.4A17.4,17.4,0,0,0,227.888,246.659ZM273.956,156.7V112.039a13.308,13.308,0,1,0-10.237,0V156.7a107.49,107.49,0,1,0,10.237,0Zm85.993,107.367c0,30.531-40.792,55.281-91.112,55.281s-91.111-24.75-91.111-55.281,40.792-55.281,91.111-55.281S359.949,233.532,359.949,264.062Zm-50.163,17.4a17.4,17.4,0,1,0-17.4-17.4h0A17.4,17.4,0,0,0,309.786,281.466ZM580.7,250.455c-14.828-2.617-22.387-3.78-22.387-9.885,0-5.523,7.268-9.884,17.735-9.884a65.56,65.56,0,0,1,34.484,10.1,8.171,8.171,0,0,0,11.288-2.468c.07-.11.138-.221.2-.333l8.611-14.886a8.2,8.2,0,0,0-2.867-11.123,99.863,99.863,0,0,0-52.014-14.138c-38.956,0-60.179,21.514-60.179,46.225,0,36.342,33.725,41.864,57.563,45.642,13.373,2.326,24.13,4.361,24.13,11.048,0,6.4-5.523,10.757-18.9,10.757-13.552,0-30.994-6.222-42.623-13.579a8.206,8.206,0,0,0-11.335,2.491c-.035.054-.069.108-.1.164l-10.2,16.891a8.222,8.222,0,0,0,2.491,11.066c15.224,10.3,37.663,16.692,59.441,16.692,40.409,0,63.957-19.769,63.957-46.515C640,260.63,604.537,254.816,580.7,250.455Zm-95.928,60.787a8.211,8.211,0,0,0-9.521-5.938,23.168,23.168,0,0,1-4.155.387c-7.849,0-12.5-6.106-12.5-14.245V240.28h20.349a8.143,8.143,0,0,0,8.141-8.143V209.466a8.143,8.143,0,0,0-8.141-8.143H458.594V171.091a8.143,8.143,0,0,0-8.143-8.143H422.257a8.143,8.143,0,0,0-8.143,8.143h0v30.232H399a8.143,8.143,0,0,0-8.143,8.143h0v22.671A8.143,8.143,0,0,0,399,240.28h15.115v63.667c0,27.037,15.408,41.282,43.9,41.282,12.183,0,21.383-2.2,27.6-5.446a8.161,8.161,0,0,0,4.145-9.278Z"],"free-code-camp":[576,512,[],"f2c5","M97.22,96.21c10.36-10.65,16-17.12,16-21.9,0-2.76-1.92-5.51-3.83-7.42A14.81,14.81,0,0,0,101,64.05c-8.48,0-20.92,8.79-35.84,25.69C23.68,137,2.51,182.81,3.37,250.34s17.47,117,54.06,161.87C76.22,435.86,90.62,448,100.9,448a13.55,13.55,0,0,0,8.37-3.84c1.91-2.76,3.81-5.63,3.81-8.38,0-5.63-3.86-12.2-13.2-20.55-44.45-42.33-67.32-97-67.48-165C32.25,188.8,54,137.83,97.22,96.21ZM239.47,420.07c.58.37.91.55.91.55Zm93.79.55.17-.13C333.24,420.62,333.17,420.67,333.26,420.62Zm3.13-158.18c-16.24-4.15,50.41-82.89-68.05-177.17,0,0,15.54,49.38-62.83,159.57-74.27,104.35,23.46,168.73,34,175.23-6.73-4.35-47.4-35.7,9.55-128.64,11-18.3,25.53-34.87,43.5-72.16,0,0,15.91,22.45,7.6,71.13C287.7,364,354,342.91,355,343.94c22.75,26.78-17.72,73.51-21.58,76.55,5.49-3.65,117.71-78,33-188.1C360.43,238.4,352.62,266.59,336.39,262.44ZM510.88,89.69C496,72.79,483.52,64,475,64a14.81,14.81,0,0,0-8.39,2.84c-1.91,1.91-3.83,4.66-3.83,7.42,0,4.78,5.6,11.26,16,21.9,43.23,41.61,65,92.59,64.82,154.06-.16,68-23,122.63-67.48,165-9.34,8.35-13.18,14.92-13.2,20.55,0,2.75,1.9,5.62,3.81,8.38A13.61,13.61,0,0,0,475.1,448c10.28,0,24.68-12.13,43.47-35.79,36.59-44.85,53.14-94.38,54.06-161.87S552.32,137,510.88,89.69Z"],hornbill:[512,512,[],"f592","M76.38 370.3a37.8 37.8 0 1 1-32.07-32.42c-78.28-111.35 52-190.53 52-190.53-5.86 43-8.24 91.16-8.24 91.16-67.31 41.49.93 64.06 39.81 72.87a140.38 140.38 0 0 0 131.66 91.94c1.92 0 3.77-.21 5.67-.28l.11 18.86c-99.22 1.39-158.7-29.14-188.94-51.6zm108-327.7A37.57 37.57 0 0 0 181 21.45a37.95 37.95 0 1 0-31.17 54.22c-22.55 29.91-53.83 89.57-52.42 190l21.84-.15c0-.9-.14-1.77-.14-2.68A140.42 140.42 0 0 1 207 132.71c8-37.71 30.7-114.3 73.8-44.29 0 0 48.14 2.38 91.18 8.24 0 0-77.84-128-187.59-54.06zm304.19 134.17a37.94 37.94 0 1 0-53.84-28.7C403 126.13 344.89 99 251.28 100.33l.14 22.5c2.7-.15 5.39-.41 8.14-.41a140.37 140.37 0 0 1 130.49 88.76c39.1 9 105.06 31.58 38.46 72.54 0 0-2.34 48.13-8.21 91.16 0 0 133.45-81.16 49-194.61a37.45 37.45 0 0 0 19.31-3.5zM374.06 436.24c21.43-32.46 46.42-89.69 45.14-179.66l-19.52.14c.08 2.06.3 4.07.3 6.15a140.34 140.34 0 0 1-91.39 131.45c-8.85 38.95-31.44 106.66-72.77 39.49 0 0-48.12-2.34-91.19-8.22 0 0 79.92 131.34 191.9 51a37.5 37.5 0 0 0 3.64 14 37.93 37.93 0 1 0 33.89-54.29z"],js:[448,512,[],"f3b8","M0 32v448h448V32H0zm243.8 349.4c0 43.6-25.6 63.5-62.9 63.5-33.7 0-53.2-17.4-63.2-38.5l34.3-20.7c6.6 11.7 12.6 21.6 27.1 21.6 13.8 0 22.6-5.4 22.6-26.5V237.7h42.1v143.7zm99.6 63.5c-39.1 0-64.4-18.6-76.7-43l34.3-19.8c9 14.7 20.8 25.6 41.5 25.6 17.4 0 28.6-8.7 28.6-20.8 0-14.4-11.4-19.5-30.7-28l-10.5-4.5c-30.4-12.9-50.5-29.2-50.5-63.5 0-31.6 24.1-55.6 61.6-55.6 26.8 0 46 9.3 59.8 33.7L368 290c-7.2-12.9-15-18-27.1-18-12.3 0-20.1 7.8-20.1 18 0 12.6 7.8 17.7 25.9 25.6l10.5 4.5c35.8 15.3 55.9 31 55.9 66.2 0 37.8-29.8 58.6-69.7 58.6z"],ideal:[576,512,[],"e013","M125.61,165.48a49.07,49.07,0,1,0,49.06,49.06A49.08,49.08,0,0,0,125.61,165.48ZM86.15,425.84h78.94V285.32H86.15Zm151.46-211.6c0-20-10-22.53-18.74-22.53H204.82V237.5h14.05C228.62,237.5,237.61,234.69,237.61,214.24Zm201.69,46V168.93h22.75V237.5h33.69C486.5,113.08,388.61,86.19,299.67,86.19H204.84V169h14c25.6,0,41.5,17.35,41.5,45.26,0,28.81-15.52,46-41.5,46h-14V425.88h94.83c144.61,0,194.94-67.16,196.72-165.64Zm-109.75,0H273.3V169h54.43v22.73H296v10.58h30V225H296V237.5h33.51Zm74.66,0-5.16-17.67H369.31l-5.18,17.67H340.47L368,168.92h32.35l27.53,91.34ZM299.65,32H32V480H299.65c161.85,0,251-79.73,251-224.52C550.62,172,518,32,299.65,32Zm0,426.92H53.07V53.07H299.65c142.1,0,229.9,64.61,229.9,202.41C529.55,389.57,448.55,458.92,299.65,458.92Zm83.86-264.85L376,219.88H392.4l-7.52-25.81Z"],git:[512,512,[],"f1d3","M216.29 158.39H137C97 147.9 6.51 150.63 6.51 233.18c0 30.09 15 51.23 35 61-25.1 23-37 33.85-37 49.21 0 11 4.47 21.14 17.89 26.81C8.13 383.61 0 393.35 0 411.65c0 32.11 28.05 50.82 101.63 50.82 70.75 0 111.79-26.42 111.79-73.18 0-58.66-45.16-56.5-151.63-63l13.43-21.55c27.27 7.58 118.7 10 118.7-67.89 0-18.7-7.73-31.71-15-41.07l37.41-2.84zm-63.42 241.9c0 32.06-104.89 32.1-104.89 2.43 0-8.14 5.27-15 10.57-21.54 77.71 5.3 94.32 3.37 94.32 19.11zm-50.81-134.58c-52.8 0-50.46-71.16 1.2-71.16 49.54 0 50.82 71.16-1.2 71.16zm133.3 100.51v-32.1c26.75-3.66 27.24-2 27.24-11V203.61c0-8.5-2.05-7.38-27.24-16.26l4.47-32.92H324v168.71c0 6.51.4 7.32 6.51 8.14l20.73 2.84v32.1zm52.45-244.31c-23.17 0-36.59-13.43-36.59-36.61s13.42-35.77 36.59-35.77c23.58 0 37 12.62 37 35.77s-13.42 36.61-37 36.61zM512 350.46c-17.49 8.53-43.1 16.26-66.28 16.26-48.38 0-66.67-19.5-66.67-65.46V194.75c0-5.42 1.05-4.06-31.71-4.06V154.5c35.78-4.07 50-22 54.47-66.27h38.63c0 65.83-1.34 61.81 3.26 61.81H501v40.65h-60.56v97.15c0 6.92-4.92 51.41 60.57 26.84z"],dev:[448,512,[],"f6cc","M120.12 208.29c-3.88-2.9-7.77-4.35-11.65-4.35H91.03v104.47h17.45c3.88 0 7.77-1.45 11.65-4.35 3.88-2.9 5.82-7.25 5.82-13.06v-69.65c-.01-5.8-1.96-10.16-5.83-13.06zM404.1 32H43.9C19.7 32 .06 51.59 0 75.8v360.4C.06 460.41 19.7 480 43.9 480h360.2c24.21 0 43.84-19.59 43.9-43.8V75.8c-.06-24.21-19.7-43.8-43.9-43.8zM154.2 291.19c0 18.81-11.61 47.31-48.36 47.25h-46.4V172.98h47.38c35.44 0 47.36 28.46 47.37 47.28l.01 70.93zm100.68-88.66H201.6v38.42h32.57v29.57H201.6v38.41h53.29v29.57h-62.18c-11.16.29-20.44-8.53-20.72-19.69V193.7c-.27-11.15 8.56-20.41 19.71-20.69h63.19l-.01 29.52zm103.64 115.29c-13.2 30.75-36.85 24.63-47.44 0l-38.53-144.8h32.57l29.71 113.72 29.57-113.72h32.58l-38.46 144.8z"],sketch:[512,512,[],"f7c6","M27.5 162.2L9 187.1h90.5l6.9-130.7-78.9 105.8zM396.3 45.7L267.7 32l135.7 147.2-7.1-133.5zM112.2 218.3l-11.2-22H9.9L234.8 458zm2-31.2h284l-81.5-88.5L256.3 33zm297.3 9.1L277.6 458l224.8-261.7h-90.9zM415.4 69L406 56.4l.9 17.3 6.1 113.4h90.3zM113.5 93.5l-4.6 85.6L244.7 32 116.1 45.7zm287.7 102.7h-290l42.4 82.9L256.3 480l144.9-283.8z"],"yandex-international":[320,512,[],"f414","M129.5 512V345.9L18.5 48h55.8l81.8 229.7L250.2 0h51.3L180.8 347.8V512h-51.3z"],"cc-amex":[576,512,[],"f1f3","M0 432c0 26.5 21.5 48 48 48H528c26.5 0 48-21.5 48-48v-1.1H514.3l-31.9-35.1-31.9 35.1H246.8V267.1H181L262.7 82.4h78.6l28.1 63.2V82.4h97.2L483.5 130l17-47.6H576V80c0-26.5-21.5-48-48-48H48C21.5 32 0 53.5 0 80V432zm440.4-21.7L482.6 364l42 46.3H576l-68-72.1 68-72.1H525.4l-42 46.7-41.5-46.7H390.5L458 338.6l-67.4 71.6V377.1h-83V354.9h80.9V322.6H307.6V300.2h83V267.1h-122V410.3H440.4zm96.3-72L576 380.2V296.9l-39.3 41.4zm-36.3-92l36.9-100.6V246.3H576V103H515.8l-32.2 89.3L451.7 103H390.5V246.1L327.3 103H276.1L213.7 246.3h43l11.9-28.7h65.9l12 28.7h82.7V146L466 246.3h34.4zM282 185.4l19.5-46.9 19.4 46.9H282z"],uber:[448,512,[],"f402","M414.1 32H33.9C15.2 32 0 47.2 0 65.9V446c0 18.8 15.2 34 33.9 34H414c18.7 0 33.9-15.2 33.9-33.9V65.9C448 47.2 432.8 32 414.1 32zM237.6 391.1C163 398.6 96.4 344.2 88.9 269.6h94.4V290c0 3.7 3 6.8 6.8 6.8H258c3.7 0 6.8-3 6.8-6.8v-67.9c0-3.7-3-6.8-6.8-6.8h-67.9c-3.7 0-6.8 3-6.8 6.8v20.4H88.9c7-69.4 65.4-122.2 135.1-122.2 69.7 0 128.1 52.8 135.1 122.2 7.5 74.5-46.9 141.1-121.5 148.6z"],github:[496,512,[],"f09b","M165.9 397.4c0 2-2.3 3.6-5.2 3.6-3.3.3-5.6-1.3-5.6-3.6 0-2 2.3-3.6 5.2-3.6 3-.3 5.6 1.3 5.6 3.6zm-31.1-4.5c-.7 2 1.3 4.3 4.3 4.9 2.6 1 5.6 0 6.2-2s-1.3-4.3-4.3-5.2c-2.6-.7-5.5.3-6.2 2.3zm44.2-1.7c-2.9.7-4.9 2.6-4.6 4.9.3 2 2.9 3.3 5.9 2.6 2.9-.7 4.9-2.6 4.6-4.6-.3-1.9-3-3.2-5.9-2.9zM244.8 8C106.1 8 0 113.3 0 252c0 110.9 69.8 205.8 169.5 239.2 12.8 2.3 17.3-5.6 17.3-12.1 0-6.2-.3-40.4-.3-61.4 0 0-70 15-84.7-29.8 0 0-11.4-29.1-27.8-36.6 0 0-22.9-15.7 1.6-15.4 0 0 24.9 2 38.6 25.8 21.9 38.6 58.6 27.5 72.9 20.9 2.3-16 8.8-27.1 16-33.7-55.9-6.2-112.3-14.3-112.3-110.5 0-27.5 7.6-41.3 23.6-58.9-2.6-6.5-11.1-33.3 2.6-67.9 20.9-6.5 69 27 69 27 20-5.6 41.5-8.5 62.8-8.5s42.8 2.9 62.8 8.5c0 0 48.1-33.6 69-27 13.7 34.7 5.2 61.4 2.6 67.9 16 17.7 25.8 31.5 25.8 58.9 0 96.5-58.9 104.2-114.8 110.5 9.2 7.9 17 22.9 17 46.4 0 33.7-.3 75.4-.3 83.6 0 6.5 4.6 14.4 17.3 12.1C428.2 457.8 496 362.9 496 252 496 113.3 383.5 8 244.8 8zM97.2 352.9c-1.3 1-1 3.3.7 5.2 1.6 1.6 3.9 2.3 5.2 1 1.3-1 1-3.3-.7-5.2-1.6-1.6-3.9-2.3-5.2-1zm-10.8-8.1c-.7 1.3.3 2.9 2.3 3.9 1.6 1 3.6.7 4.3-.7.7-1.3-.3-2.9-2.3-3.9-2-.6-3.6-.3-4.3.7zm32.4 35.6c-1.6 1.3-1 4.3 1.3 6.2 2.3 2.3 5.2 2.6 6.5 1 1.3-1.3.7-4.3-1.3-6.2-2.2-2.3-5.2-2.6-6.5-1zm-11.4-14.7c-1.6 1-1.6 3.6 0 5.9 1.6 2.3 4.3 3.3 5.6 2.3 1.6-1.3 1.6-3.9 0-6.2-1.4-2.3-4-3.3-5.6-2z"],php:[640,512,[],"f457","M320 104.5c171.4 0 303.2 72.2 303.2 151.5S491.3 407.5 320 407.5c-171.4 0-303.2-72.2-303.2-151.5S148.7 104.5 320 104.5m0-16.8C143.3 87.7 0 163 0 256s143.3 168.3 320 168.3S640 349 640 256 496.7 87.7 320 87.7zM218.2 242.5c-7.9 40.5-35.8 36.3-70.1 36.3l13.7-70.6c38 0 63.8-4.1 56.4 34.3zM97.4 350.3h36.7l8.7-44.8c41.1 0 66.6 3 90.2-19.1 26.1-24 32.9-66.7 14.3-88.1-9.7-11.2-25.3-16.7-46.5-16.7h-70.7L97.4 350.3zm185.7-213.6h36.5l-8.7 44.8c31.5 0 60.7-2.3 74.8 10.7 14.8 13.6 7.7 31-8.3 113.1h-37c15.4-79.4 18.3-86 12.7-92-5.4-5.8-17.7-4.6-47.4-4.6l-18.8 96.6h-36.5l32.7-168.6zM505 242.5c-8 41.1-36.7 36.3-70.1 36.3l13.7-70.6c38.2 0 63.8-4.1 56.4 34.3zM384.2 350.3H421l8.7-44.8c43.2 0 67.1 2.5 90.2-19.1 26.1-24 32.9-66.7 14.3-88.1-9.7-11.2-25.3-16.7-46.5-16.7H417l-32.8 168.7z"],alipay:[448,512,[],"f642","M377.74 32H70.26C31.41 32 0 63.41 0 102.26v307.48C0 448.59 31.41 480 70.26 480h307.48c38.52 0 69.76-31.08 70.26-69.6-45.96-25.62-110.59-60.34-171.6-88.44-32.07 43.97-84.14 81-148.62 81-70.59 0-93.73-45.3-97.04-76.37-3.97-39.01 14.88-81.5 99.52-81.5 35.38 0 79.35 10.25 127.13 24.96 16.53-30.09 26.45-60.34 26.45-60.34h-178.2v-16.7h92.08v-31.24H88.28v-19.01h109.44V92.34h50.92v50.42h109.44v19.01H248.63v31.24h88.77s-15.21 46.62-38.35 90.92c48.93 16.7 100.01 36.04 148.62 52.74V102.26C447.83 63.57 416.43 32 377.74 32zM47.28 322.95c.99 20.17 10.25 53.73 69.93 53.73 52.07 0 92.58-39.68 117.87-72.9-44.63-18.68-84.48-31.41-109.44-31.41-67.45 0-79.35 33.06-78.36 50.58z"],youtube:[576,512,[61802],"f167","M549.655 124.083c-6.281-23.65-24.787-42.276-48.284-48.597C458.781 64 288 64 288 64S117.22 64 74.629 75.486c-23.497 6.322-42.003 24.947-48.284 48.597-11.412 42.867-11.412 132.305-11.412 132.305s0 89.438 11.412 132.305c6.281 23.65 24.787 41.5 48.284 47.821C117.22 448 288 448 288 448s170.78 0 213.371-11.486c23.497-6.321 42.003-24.171 48.284-47.821 11.412-42.867 11.412-132.305 11.412-132.305s0-89.438-11.412-132.305zm-317.51 213.508V175.185l142.739 81.205-142.739 81.201z"],skyatlas:[640,512,[],"f216","M640 329.3c0 65.9-52.5 114.4-117.5 114.4-165.9 0-196.6-249.7-359.7-249.7-146.9 0-147.1 212.2 5.6 212.2 42.5 0 90.9-17.8 125.3-42.5 5.6-4.1 16.9-16.3 22.8-16.3s10.9 5 10.9 10.9c0 7.8-13.1 19.1-18.7 24.1-40.9 35.6-100.3 61.2-154.7 61.2-83.4.1-154-59-154-144.9s67.5-149.1 152.8-149.1c185.3 0 222.5 245.9 361.9 245.9 99.9 0 94.8-139.7 3.4-139.7-17.5 0-35 11.6-46.9 11.6-8.4 0-15.9-7.2-15.9-15.6 0-11.6 5.3-23.7 5.3-36.3 0-66.6-50.9-114.7-116.9-114.7-53.1 0-80 36.9-88.8 36.9-6.2 0-11.2-5-11.2-11.2 0-5.6 4.1-10.3 7.8-14.4 25.3-28.8 64.7-43.7 102.8-43.7 79.4 0 139.1 58.4 139.1 137.8 0 6.9-.3 13.7-1.2 20.6 11.9-3.1 24.1-4.7 35.9-4.7 60.7 0 111.9 45.3 111.9 107.2z"],"firefox-browser":[512,512,[],"e007","M130.22 127.548C130.38 127.558 130.3 127.558 130.22 127.548V127.548ZM481.64 172.898C471.03 147.398 449.56 119.898 432.7 111.168C446.42 138.058 454.37 165.048 457.4 185.168C457.405 185.306 457.422 185.443 457.45 185.578C429.87 116.828 383.098 89.1089 344.9 28.7479C329.908 5.05792 333.976 3.51792 331.82 4.08792L331.7 4.15792C284.99 30.1109 256.365 82.5289 249.12 126.898C232.503 127.771 216.219 131.895 201.19 139.035C199.838 139.649 198.736 140.706 198.066 142.031C197.396 143.356 197.199 144.87 197.506 146.323C197.7 147.162 198.068 147.951 198.586 148.639C199.103 149.327 199.76 149.899 200.512 150.318C201.264 150.737 202.096 150.993 202.954 151.071C203.811 151.148 204.676 151.045 205.491 150.768L206.011 150.558C221.511 143.255 238.408 139.393 255.541 139.238C318.369 138.669 352.698 183.262 363.161 201.528C350.161 192.378 326.811 183.338 304.341 187.248C392.081 231.108 368.541 381.784 246.951 376.448C187.487 373.838 149.881 325.467 146.421 285.648C146.421 285.648 157.671 243.698 227.041 243.698C234.541 243.698 255.971 222.778 256.371 216.698C256.281 214.698 213.836 197.822 197.281 181.518C188.434 172.805 184.229 168.611 180.511 165.458C178.499 163.75 176.392 162.158 174.201 160.688C168.638 141.231 168.399 120.638 173.51 101.058C148.45 112.468 128.96 130.508 114.8 146.428H114.68C105.01 134.178 105.68 93.7779 106.25 85.3479C106.13 84.8179 99.022 89.0159 98.1 89.6579C89.5342 95.7103 81.5528 102.55 74.26 110.088C57.969 126.688 30.128 160.242 18.76 211.318C14.224 231.701 12 255.739 12 263.618C12 398.318 121.21 507.508 255.92 507.508C376.56 507.508 478.939 420.281 496.35 304.888C507.922 228.192 481.64 173.82 481.64 172.898Z"],replyd:[448,512,[],"f3e6","M320 480H128C57.6 480 0 422.4 0 352V160C0 89.6 57.6 32 128 32h192c70.4 0 128 57.6 128 128v192c0 70.4-57.6 128-128 128zM193.4 273.2c-6.1-2-11.6-3.1-16.4-3.1-7.2 0-13.5 1.9-18.9 5.6-5.4 3.7-9.6 9-12.8 15.8h-1.1l-4.2-18.3h-28v138.9h36.1v-89.7c1.5-5.4 4.4-9.8 8.7-13.2 4.3-3.4 9.8-5.1 16.2-5.1 4.6 0 9.8 1 15.6 3.1l4.8-34zm115.2 103.4c-3.2 2.4-7.7 4.8-13.7 7.1-6 2.3-12.8 3.5-20.4 3.5-12.2 0-21.1-3-26.5-8.9-5.5-5.9-8.5-14.7-9-26.4h83.3c.9-4.8 1.6-9.4 2.1-13.9.5-4.4.7-8.6.7-12.5 0-10.7-1.6-19.7-4.7-26.9-3.2-7.2-7.3-13-12.5-17.2-5.2-4.3-11.1-7.3-17.8-9.2-6.7-1.8-13.5-2.8-20.6-2.8-21.1 0-37.5 6.1-49.2 18.3s-17.5 30.5-17.5 55c0 22.8 5.2 40.7 15.6 53.7 10.4 13.1 26.8 19.6 49.2 19.6 10.7 0 20.9-1.5 30.4-4.6 9.5-3.1 17.1-6.8 22.6-11.2l-12-23.6zm-21.8-70.3c3.8 5.4 5.3 13.1 4.6 23.1h-51.7c.9-9.4 3.7-17 8.2-22.6 4.5-5.6 11.5-8.5 21-8.5 8.2-.1 14.1 2.6 17.9 8zm79.9 2.5c4.1 3.9 9.4 5.8 16.1 5.8 7 0 12.6-1.9 16.7-5.8s6.1-9.1 6.1-15.6-2-11.6-6.1-15.4c-4.1-3.8-9.6-5.7-16.7-5.7-6.7 0-12 1.9-16.1 5.7-4.1 3.8-6.1 8.9-6.1 15.4s2 11.7 6.1 15.6zm0 100.5c4.1 3.9 9.4 5.8 16.1 5.8 7 0 12.6-1.9 16.7-5.8s6.1-9.1 6.1-15.6-2-11.6-6.1-15.4c-4.1-3.8-9.6-5.7-16.7-5.7-6.7 0-12 1.9-16.1 5.7-4.1 3.8-6.1 8.9-6.1 15.4 0 6.6 2 11.7 6.1 15.6z"],suse:[640,512,[],"f7d6","M593.1 192.6A10.4 10.4 0 1 1 604.5 210a10.4 10.4 0 1 1 -11.4-17.4zm-47.1 12.2a38.5 38.5 0 1 1 75-17.6 38.5 38.5 0 1 1 -75 17.6zM433.7 336.7c3.2 4.6 5.8 9 7.3 13.4c1 3.1 2.4 7.3 5.5 8.9c.2 .1 .3 .2 .5 .2c5.7 2.1 20.3 1.7 20.3 1.7h26.8c2.3 0 22.4 0 21.9-2.3c-2.4-10.8-14.9-12.7-24.4-18.3c-8.7-5.2-17-11.1-20.8-21.3c-2-5.2-.8-17.4 2.6-21.8c2.5-3.2 6.1-5.3 10-6.2c4.3-.9 8.8-.1 13.1 .3c5.3 .5 10.6 1.5 15.9 2.2c10.3 1.3 20.6 1.9 31 1.6c17.1-.5 34.2-3.2 50.4-8.7c11.3-3.8 22.4-8.9 32-16.1c10.9-8.1 8.1-7.4-3-6.2c-13.3 1.4-26.6 1.6-39.9 .8c-12.4-.7-24.7-2.2-35.9-7.9c-8.8-4.6-16.4-9.1-23.4-16.2c-1-1.1-1.7-4.2 .2-6.2c1.9-1.9 5.8-.8 7 .2c12.2 10.2 30.5 18.6 49.3 19.5c10.2 .5 20.1 .7 30.4 .3c5.1-.2 12.8-.2 17.9-.3c2.6 0 9.8 .7 11.2-2.1c.4-.8 .4-1.8 .3-2.7c-1.5-40.9-4.5-86.9-47.3-106.5c-31.9-14.6-79.7-37.2-99.9-46.6c-4.7-2.2-10.2 1.3-10.2 6.5c0 13.6 .7 33.3 .7 51.1c-9.7-9.9-26-16.1-38.4-21.8c-14.1-6.5-28.7-12-43.5-16.6c-29.8-9.2-60.7-14.9-91.7-18c-35.2-3.5-71-1.8-105.7 5.3C147 115.1 90.8 142.6 48.2 182.7C22.1 207.3 1.6 242.4 .2 277.9c-2 50.3 12.1 77.3 38 105.2c41.3 44.4 130.2 50.6 166.2-2c16.2-23.7 19.7-55.8 8-82c-11.8-26.2-38.8-45.1-67.4-46c-22.2-.7-45.9 10.6-54.5 31.1c-6.5 15.7-2.8 35.1 9 47.3c4.6 4.8 10.9 8.7 17.7 7.1c4-.9 7.4-3.9 8-8c.9-6-4.4-9.9-7.6-14.5c-5.8-8.3-4.7-20.9 2.7-27.9c6.2-6 15.3-7.8 23.9-7.7c8 0 16.2 1.4 23.1 5.5c9.7 5.7 16.2 16.2 18.4 27.2c6.7 33-20.2 59.9-56.6 62c-18.6 1.1-37.6-3.8-52.1-15.5C40.1 329.9 31.1 269.4 73.2 237c40-30.7 90.4-22.8 120.2-6.8c23.8 12.8 41.5 33.6 55 56.7c6.7 11.6 12.5 23.7 17.8 36.1c5.1 11.8 9.9 23.8 20.2 32.5c6.8 5.8 15.2 5.6 24.1 5.6h50.8c6.9 0 5.2-4.6 2.2-7.7c-6.7-6.9-16.4-8.4-25.4-10.9c-20.5-5.6-18.4-32.8-12.7-32.8c18.3 0 18.9 .6 34.9 .3c23.2-.3 30.2-1.7 48.3 5c9.7 3.6 19 13.1 25.1 21.7z"],jenkins:[512,512,[],"f3b6","M487.1 425c-1.4-11.2-19-23.1-28.2-31.9-5.1-5-29-23.1-30.4-29.9-1.4-6.6 9.7-21.5 13.3-28.9 5.1-10.7 8.8-23.7 11.3-32.6 18.8-66.1 20.7-156.9-6.2-211.2-10.2-20.6-38.6-49-56.4-62.5-42-31.7-119.6-35.3-170.1-16.6-14.1 5.2-27.8 9.8-40.1 17.1-33.1 19.4-68.3 32.5-78.1 71.6-24.2 10.8-31.5 41.8-30.3 77.8.2 7 4.1 15.8 2.7 22.4-.7 3.3-5.2 7.6-6.1 9.8-11.6 27.7-2.3 64 11.1 83.7 8.1 11.9 21.5 22.4 39.2 25.2.7 10.6 3.3 19.7 8.2 30.4 3.1 6.8 14.7 19 10.4 27.7-2.2 4.4-21 13.8-27.3 17.6C89 407.2 73.7 415 54.2 429c-12.6 9-32.3 10.2-29.2 31.1 2.1 14.1 10.1 31.6 14.7 45.8.7 2 1.4 4.1 2.1 6h422c4.9-15.3 9.7-30.9 14.6-47.2 3.4-11.4 10.2-27.8 8.7-39.7zM205.9 33.7c1.8-.5 3.4.7 4.9 2.4-.2 5.2-5.4 5.1-8.9 6.8-5.4 6.7-13.4 9.8-20 17.2-6.8 7.5-14.4 27.7-23.4 30-4.5 1.1-9.7-.8-13.6-.5-10.4.7-17.7 6-28.3 7.5 13.6-29.9 56.1-54 89.3-63.4zm-104.8 93.6c13.5-14.9 32.1-24.1 54.8-25.9 11.7 29.7-8.4 65-.9 97.6 2.3 9.9 10.2 25.4-2.4 25.7.3-28.3-34.8-46.3-61.3-29.6-1.8-21.5-4.9-51.7 9.8-67.8zm36.7 200.2c-1-4.1-2.7-12.9-2.3-15.1 1.6-8.7 17.1-12.5 11-24.7-11.3-.1-13.8 10.2-24.1 11.3-26.7 2.6-45.6-35.4-44.4-58.4 1-19.5 17.6-38.2 40.1-35.8 16 1.8 21.4 19.2 24.5 34.7 9.2.5 22.5-.4 26.9-7.6-.6-17.5-8.8-31.6-8.2-47.7 1-30.3 17.5-57.6 4.8-87.4 13.6-30.9 53.5-55.3 83.1-70 36.6-18.3 94.9-3.7 129.3 15.8 19.7 11.1 34.4 32.7 48.3 50.7-19.5-5.8-36.1 4.2-33.1 20.3 16.3-14.9 44.2-.2 52.5 16.4 7.9 15.8 7.8 39.3 9 62.8 2.9 57-10.4 115.9-39.1 157.1-7.7 11-14.1 23-24.9 30.6-26 18.2-65.4 34.7-99.2 23.4-44.7-15-65-44.8-89.5-78.8.7 18.7 13.8 34.1 26.8 48.4 11.3 12.5 25 26.6 39.7 32.4-12.3-2.9-31.1-3.8-36.2 7.2-28.6-1.9-55.1-4.8-68.7-24.2-10.6-15.4-21.4-41.4-26.3-61.4zm222 124.1c4.1-3 11.1-2.9 17.4-3.6-5.4-2.7-13-3.7-19.3-2.2-.1-4.2-2-6.8-3.2-10.2 10.6-3.8 35.5-28.5 49.6-20.3 6.7 3.9 9.5 26.2 10.1 37 .4 9-.8 18-4.5 22.8-18.8-.6-35.8-2.8-50.7-7 .9-6.1-1-12.1.6-16.5zm-17.2-20c-16.8.8-26-1.2-38.3-10.8.2-.8 1.4-.5 1.5-1.4 18 8 40.8-3.3 59-4.9-7.9 5.1-14.6 11.6-22.2 17.1zm-12.1 33.2c-1.6-9.4-3.5-12-2.8-20.2 25-16.6 29.7 28.6 2.8 20.2zM226 438.6c-11.6-.7-48.1-14-38.5-23.7 9.4 6.5 27.5 4.9 41.3 7.3.8 4.4-2.8 10.2-2.8 16.4zM57.7 497.1c-4.3-12.7-9.2-25.1-14.8-36.9 30.8-23.8 65.3-48.9 102.2-63.5 2.8-1.1 23.2 25.4 26.2 27.6 16.5 11.7 37 21 56.2 30.2 1.2 8.8 3.9 20.2 8.7 35.5.7 2.3 1.4 4.7 2.2 7.2H57.7zm240.6 5.7h-.8c.3-.2.5-.4.8-.5v.5zm7.5-5.7c2.1-1.4 4.3-2.8 6.4-4.3 1.1 1.4 2.2 2.8 3.2 4.3h-9.6zm15.1-24.7c-10.8 7.3-20.6 18.3-33.3 25.2-6 3.3-27 11.7-33.4 10.2-3.6-.8-3.9-5.3-5.4-9.5-3.1-9-10.1-23.4-10.8-37-.8-17.2-2.5-46 16-42.4 14.9 2.9 32.3 9.7 43.9 16.1 7.1 3.9 11.1 8.6 21.9 9.5-.1 1.4-.1 2.8-.2 4.3-5.9 3.9-15.3 3.8-21.8 7.1 9.5.4 17 2.7 23.5 5.9-.1 3.4-.3 7-.4 10.6zm53.4 24.7h-14c-.1-3.2-2.8-5.8-6.1-5.8s-5.9 2.6-6.1 5.8h-17.4c-2.8-4.4-5.7-8.6-8.9-12.5 2.1-2.2 4-4.7 6-6.9 9 3.7 14.8-4.9 21.7-4.2 7.9.8 14.2 11.7 25.4 11l-.6 12.6zm8.7 0c.2-4 .4-7.8.6-11.5 15.6-7.3 29 1.3 35.7 11.5H383zm83.4-37c-2.3 11.2-5.8 24-9.9 37.1-.2-.1-.4-.1-.6-.1H428c.6-1.1 1.2-2.2 1.9-3.3-2.6-6.1-9-8.7-10.9-15.5 12.1-22.7 6.5-93.4-24.2-78.5 4.3-6.3 15.6-11.5 20.8-19.3 13 10.4 20.8 20.3 33.2 31.4 6.8 6 20 13.3 21.4 23.1.8 5.5-2.6 18.9-3.8 25.1zM222.2 130.5c5.4-14.9 27.2-34.7 45-32 7.7 1.2 18 8.2 12.2 17.7-30.2-7-45.2 12.6-54.4 33.1-8.1-2-4.9-13.1-2.8-18.8zm184.1 63.1c8.2-3.6 22.4-.7 29.6-5.3-4.2-11.5-10.3-21.4-9.3-37.7.5 0 1 0 1.4.1 6.8 14.2 12.7 29.2 21.4 41.7-5.7 13.5-43.6 25.4-43.1 1.2zm20.4-43zm-117.2 45.7c-6.8-10.9-19-32.5-14.5-45.3 6.5 11.9 8.6 24.4 17.8 33.3 4.1 4 12.2 9 8.2 20.2-.9 2.7-7.8 8.6-11.7 9.7-14.4 4.3-47.9.9-36.6-17.1 11.9.7 27.9 7.8 36.8-.8zm27.3 70c3.8 6.6 1.4 18.7 12.1 20.6 20.2 3.4 43.6-12.3 58.1-17.8 9-15.2-.8-20.7-8.9-30.5-16.6-20-38.8-44.8-38-74.7 6.7-4.9 7.3 7.4 8.2 9.7 8.7 20.3 30.4 46.2 46.3 63.5 3.9 4.3 10.3 8.4 11 11.2 2.1 8.2-5.4 18-4.5 23.5-21.7 13.9-45.8 29.1-81.4 25.6-7.4-6.7-10.3-21.4-2.9-31.1zm-201.3-9.2c-6.8-3.9-8.4-21-16.4-21.4-11.4-.7-9.3 22.2-9.3 35.5-7.8-7.1-9.2-29.1-3.5-40.3-6.6-3.2-9.5 3.6-13.1 5.9 4.7-34.1 49.8-15.8 42.3 20.3zm299.6 28.8c-10.1 19.2-24.4 40.4-54 41-.6-6.2-1.1-15.6 0-19.4 22.7-2.2 36.6-13.7 54-21.6zm-141.9 12.4c18.9 9.9 53.6 11 79.3 10.2 1.4 5.6 1.3 12.6 1.4 19.4-33 1.8-72-6.4-80.7-29.6zm92.2 46.7c-1.7 4.3-5.3 9.3-9.8 11.1-12.1 4.9-45.6 8.7-62.4-.3-10.7-5.7-17.5-18.5-23.4-26-2.8-3.6-16.9-12.9-.2-12.9 13.1 32.7 58 29 95.8 28.1z"],twitter:[512,512,[],"f099","M459.37 151.716c.325 4.548.325 9.097.325 13.645 0 138.72-105.583 298.558-298.558 298.558-59.452 0-114.68-17.219-161.137-47.106 8.447.974 16.568 1.299 25.34 1.299 49.055 0 94.213-16.568 130.274-44.832-46.132-.975-84.792-31.188-98.112-72.772 6.498.974 12.995 1.624 19.818 1.624 9.421 0 18.843-1.3 27.614-3.573-48.081-9.747-84.143-51.98-84.143-102.985v-1.299c13.969 7.797 30.214 12.67 47.431 13.319-28.264-18.843-46.781-51.005-46.781-87.391 0-19.492 5.197-37.36 14.294-52.954 51.655 63.675 129.3 105.258 216.365 109.807-1.624-7.797-2.599-15.918-2.599-24.04 0-57.828 46.782-104.934 104.934-104.934 30.213 0 57.502 12.67 76.67 33.137 23.715-4.548 46.456-13.32 66.599-25.34-7.798 24.366-24.366 44.833-46.132 57.827 21.117-2.273 41.584-8.122 60.426-16.243-14.292 20.791-32.161 39.308-52.628 54.253z"],rockrms:[496,512,[],"f3e9","M248 8C111 8 0 119 0 256s111 248 248 248 248-111 248-248S385 8 248 8zm157.4 419.5h-90l-112-131.3c-17.9-20.4-3.9-56.1 26.6-56.1h75.3l-84.6-99.3-84.3 98.9h-90L193.5 67.2c14.4-18.4 41.3-17.3 54.5 0l157.7 185.1c19 22.8 2 57.2-27.6 56.1-.6 0-74.2.2-74.2.2l101.5 118.9z"],pinterest:[496,512,[],"f0d2","M496 256c0 137-111 248-248 248-25.6 0-50.2-3.9-73.4-11.1 10.1-16.5 25.2-43.5 30.8-65 3-11.6 15.4-59 15.4-59 8.1 15.4 31.7 28.5 56.8 28.5 74.8 0 128.7-68.8 128.7-154.3 0-81.9-66.9-143.2-152.9-143.2-107 0-163.9 71.8-163.9 150.1 0 36.4 19.4 81.7 50.3 96.1 4.7 2.2 7.2 1.2 8.3-3.3.8-3.4 5-20.3 6.9-28.1.6-2.5.3-4.7-1.7-7.1-10.1-12.5-18.3-35.3-18.3-56.6 0-54.7 41.4-107.6 112-107.6 60.9 0 103.6 41.5 103.6 100.9 0 67.1-33.9 113.6-78 113.6-24.3 0-42.6-20.1-36.7-44.8 7-29.5 20.5-61.3 20.5-82.6 0-19-10.2-34.9-31.4-34.9-24.9 0-44.9 25.7-44.9 60.2 0 22 7.4 36.8 7.4 36.8s-24.5 103.8-29 123.2c-5 21.4-3 51.6-.9 71.2C65.4 450.9 0 361.1 0 256 0 119 111 8 248 8s248 111 248 248z"],buffer:[448,512,[],"f837","M427.84 380.67l-196.5 97.82a18.6 18.6 0 0 1-14.67 0L20.16 380.67c-4-2-4-5.28 0-7.29L67.22 350a18.65 18.65 0 0 1 14.69 0l134.76 67a18.51 18.51 0 0 0 14.67 0l134.76-67a18.62 18.62 0 0 1 14.68 0l47.06 23.43c4.05 1.96 4.05 5.24 0 7.24zm0-136.53l-47.06-23.43a18.62 18.62 0 0 0-14.68 0l-134.76 67.08a18.68 18.68 0 0 1-14.67 0L81.91 220.71a18.65 18.65 0 0 0-14.69 0l-47.06 23.43c-4 2-4 5.29 0 7.31l196.51 97.8a18.6 18.6 0 0 0 14.67 0l196.5-97.8c4.05-2.02 4.05-5.3 0-7.31zM20.16 130.42l196.5 90.29a20.08 20.08 0 0 0 14.67 0l196.51-90.29c4-1.86 4-4.89 0-6.74L231.33 33.4a19.88 19.88 0 0 0-14.67 0l-196.5 90.28c-4.05 1.85-4.05 4.88 0 6.74z"],npm:[576,512,[],"f3d4","M288 288h-32v-64h32v64zm288-128v192H288v32H160v-32H0V160h576zm-416 32H32v128h64v-96h32v96h32V192zm160 0H192v160h64v-32h64V192zm224 0H352v128h64v-96h32v96h32v-96h32v96h32V192z"],yammer:[512,512,[],"f840","M500.676,159.486a12.779,12.779,0,0,0-6.4-8.282,13.954,13.954,0,0,0-10.078-1.125L457.8,156.7l-.043-.2-22.3,5.785-1.243.333-.608-2.17A369.037,369.037,0,0,0,347.538,4.289a14.1,14.1,0,0,0-19.784-.463l-102.9,102.747H24.947A24.9,24.9,0,0,0,0,131.417V380.38a24.963,24.963,0,0,0,24.918,24.9H224.986L328.072,508a13.667,13.667,0,0,0,19.327,0c.126-.126.249-.255.37-.385a368.025,368.025,0,0,0,69.577-107.374,403.45,403.45,0,0,0,17.3-50.8v-.028l20.406,5.336.029-.073L483.345,362a20.253,20.253,0,0,0,2.619.5,13.359,13.359,0,0,0,4.139-.072,13.5,13.5,0,0,0,10.515-9.924,415.855,415.855,0,0,0,.058-193.013ZM337.125,24.65l.013.014h-.013Zm-110.2,165.161L174.311,281.1a11.338,11.338,0,0,0-1.489,5.655v46.189a22.04,22.04,0,0,1-22.041,22h-3.4A22.068,22.068,0,0,1,125.3,332.962V287.294a11.532,11.532,0,0,0-1.388-5.51l-51.6-92.2a21.988,21.988,0,0,1,19.264-32.726h3.268a22.059,22.059,0,0,1,19.611,11.916l36.357,70.281,37.515-70.512a22.066,22.066,0,0,1,38.556-.695,21.7,21.7,0,0,1,0,21.967ZM337.145,24.673a348.147,348.147,0,0,1,75.8,141.335l.564,1.952-114.134,29.6V131.417a25.006,25.006,0,0,0-24.947-24.9H255.067Zm60.5,367.305v-.043l-.014.014a347.19,347.19,0,0,1-60.177,95.227l-82.2-81.893h19.177a24.978,24.978,0,0,0,24.947-24.9v-66.2l114.6,29.862A385.191,385.191,0,0,1,397.648,391.978Zm84-52.45.015.014-50.618-13.131L299.379,292.1V219.572l119.746-30.99,4.468-1.157,39.54-10.253,18.511-4.816A393,393,0,0,1,481.644,339.528Z"],btc:[384,512,[],"f15a","M310.204 242.638c27.73-14.18 45.377-39.39 41.28-81.3-5.358-57.351-52.458-76.573-114.85-81.929V0h-48.528v77.203c-12.605 0-25.525.315-38.444.63V0h-48.528v79.409c-17.842.539-38.622.276-97.37 0v51.678c38.314-.678 58.417-3.14 63.023 21.427v217.429c-2.925 19.492-18.524 16.685-53.255 16.071L3.765 443.68c88.481 0 97.37.315 97.37.315V512h48.528v-67.06c13.234.315 26.154.315 38.444.315V512h48.528v-68.005c81.299-4.412 135.647-24.894 142.895-101.467 5.671-61.446-23.32-88.862-69.326-99.89zM150.608 134.553c27.415 0 113.126-8.507 113.126 48.528 0 54.515-85.71 48.212-113.126 48.212v-96.74zm0 251.776V279.821c32.772 0 133.127-9.138 133.127 53.255-.001 60.186-100.355 53.253-133.127 53.253z"],dribbble:[512,512,[],"f17d","M256 8C119.252 8 8 119.252 8 256s111.252 248 248 248 248-111.252 248-248S392.748 8 256 8zm163.97 114.366c29.503 36.046 47.369 81.957 47.835 131.955-6.984-1.477-77.018-15.682-147.502-6.818-5.752-14.041-11.181-26.393-18.617-41.614 78.321-31.977 113.818-77.482 118.284-83.523zM396.421 97.87c-3.81 5.427-35.697 48.286-111.021 76.519-34.712-63.776-73.185-116.168-79.04-124.008 67.176-16.193 137.966 1.27 190.061 47.489zm-230.48-33.25c5.585 7.659 43.438 60.116 78.537 122.509-99.087 26.313-186.36 25.934-195.834 25.809C62.38 147.205 106.678 92.573 165.941 64.62zM44.17 256.323c0-2.166.043-4.322.108-6.473 9.268.19 111.92 1.513 217.706-30.146 6.064 11.868 11.857 23.915 17.174 35.949-76.599 21.575-146.194 83.527-180.531 142.306C64.794 360.405 44.17 310.73 44.17 256.323zm81.807 167.113c22.127-45.233 82.178-103.622 167.579-132.756 29.74 77.283 42.039 142.053 45.189 160.638-68.112 29.013-150.015 21.053-212.768-27.882zm248.38 8.489c-2.171-12.886-13.446-74.897-41.152-151.033 66.38-10.626 124.7 6.768 131.947 9.055-9.442 58.941-43.273 109.844-90.795 141.978z"],"stumbleupon-circle":[496,512,[],"f1a3","M256 8C119 8 8 119 8 256s111 248 248 248 248-111 248-248S393 8 256 8zm0 177.5c-9.8 0-17.8 8-17.8 17.8v106.9c0 40.9-33.9 73.9-74.9 73.9-41.4 0-74.9-33.5-74.9-74.9v-46.5h57.3v45.8c0 10 8 17.8 17.8 17.8s17.8-7.9 17.8-17.8V200.1c0-40 34.2-72.1 74.7-72.1 40.7 0 74.7 32.3 74.7 72.6v23.7l-34.1 10.1-22.9-10.7v-20.6c.1-9.6-7.9-17.6-17.7-17.6zm167.6 123.6c0 41.4-33.5 74.9-74.9 74.9-41.2 0-74.9-33.2-74.9-74.2V263l22.9 10.7 34.1-10.1v47.1c0 9.8 8 17.6 17.8 17.6s17.8-7.9 17.8-17.6v-48h57.3c-.1 45.9-.1 46.4-.1 46.4z"],"internet-explorer":[512,512,[],"f26b","M483.049 159.706c10.855-24.575 21.424-60.438 21.424-87.871 0-72.722-79.641-98.371-209.673-38.577-107.632-7.181-211.221 73.67-237.098 186.457 30.852-34.862 78.271-82.298 121.977-101.158C125.404 166.85 79.128 228.002 43.992 291.725 23.246 329.651 0 390.94 0 436.747c0 98.575 92.854 86.5 180.251 42.006 31.423 15.43 66.559 15.573 101.695 15.573 97.124 0 184.249-54.294 216.814-146.022H377.927c-52.509 88.593-196.819 52.996-196.819-47.436H509.9c6.407-43.581-1.655-95.715-26.851-141.162zM64.559 346.877c17.711 51.15 53.703 95.871 100.266 123.304-88.741 48.94-173.267 29.096-100.266-123.304zm115.977-108.873c2-55.151 50.276-94.871 103.98-94.871 53.418 0 101.981 39.72 103.981 94.871H180.536zm184.536-187.6c21.425-10.287 48.563-22.003 72.558-22.003 31.422 0 54.274 21.717 54.274 53.722 0 20.003-7.427 49.007-14.569 67.867-26.28-42.292-65.986-81.584-112.263-99.586z"],stubber:[448,512,[],"e5c7","M136.5 294.2l58.8 22.9c9.1-36.8 25.4-61.1 55-61.1c49.4 0 71.4 63.6 142.4 63.6c15.6 0 35.9-2.8 55.3-13.3V368c0 61.8-50.4 112-112.3 112H0l41.8-56L0 368l41.7-56L0 256.1l41.8-56L0 144.1 41.8 88 0 32H335.7C397.6 32 448 82.3 448 144.1v51.3c-9.2 36.3-25.9 60.6-55 60.6c-49.6 0-71.6-63.5-142.4-63.5c-35.9 0-95.2 14.6-114.1 101.6h0z"],telegram:[496,512,[62462,"telegram-plane"],"f2c6","M248,8C111.033,8,0,119.033,0,256S111.033,504,248,504,496,392.967,496,256,384.967,8,248,8ZM362.952,176.66c-3.732,39.215-19.881,134.378-28.1,178.3-3.476,18.584-10.322,24.816-16.948,25.425-14.4,1.326-25.338-9.517-39.287-18.661-21.827-14.308-34.158-23.215-55.346-37.177-24.485-16.135-8.612-25,5.342-39.5,3.652-3.793,67.107-61.51,68.335-66.746.153-.655.3-3.1-1.154-4.384s-3.59-.849-5.135-.5q-3.283.746-104.608,69.142-14.845,10.194-26.894,9.934c-8.855-.191-25.888-5.006-38.551-9.123-15.531-5.048-27.875-7.717-26.8-16.291q.84-6.7,18.45-13.7,108.446-47.248,144.628-62.3c68.872-28.647,83.183-33.623,92.511-33.789,2.052-.034,6.639.474,9.61,2.885a10.452,10.452,0,0,1,3.53,6.716A43.765,43.765,0,0,1,362.952,176.66Z"],"old-republic":[496,512,[],"f510","M235.76 10.23c7.5-.31 15-.28 22.5-.09 3.61.14 7.2.4 10.79.73 4.92.27 9.79 1.03 14.67 1.62 2.93.43 5.83.98 8.75 1.46 7.9 1.33 15.67 3.28 23.39 5.4 12.24 3.47 24.19 7.92 35.76 13.21 26.56 12.24 50.94 29.21 71.63 49.88 20.03 20.09 36.72 43.55 48.89 69.19 1.13 2.59 2.44 5.1 3.47 7.74 2.81 6.43 5.39 12.97 7.58 19.63 4.14 12.33 7.34 24.99 9.42 37.83.57 3.14 1.04 6.3 1.4 9.47.55 3.83.94 7.69 1.18 11.56.83 8.34.84 16.73.77 25.1-.07 4.97-.26 9.94-.75 14.89-.24 3.38-.51 6.76-.98 10.12-.39 2.72-.63 5.46-1.11 8.17-.9 5.15-1.7 10.31-2.87 15.41-4.1 18.5-10.3 36.55-18.51 53.63-15.77 32.83-38.83 62.17-67.12 85.12a246.503 246.503 0 0 1-56.91 34.86c-6.21 2.68-12.46 5.25-18.87 7.41-3.51 1.16-7.01 2.38-10.57 3.39-6.62 1.88-13.29 3.64-20.04 5-4.66.91-9.34 1.73-14.03 2.48-5.25.66-10.5 1.44-15.79 1.74-6.69.66-13.41.84-20.12.81-6.82.03-13.65-.12-20.45-.79-3.29-.23-6.57-.5-9.83-.95-2.72-.39-5.46-.63-8.17-1.11-4.12-.72-8.25-1.37-12.35-2.22-4.25-.94-8.49-1.89-12.69-3.02-8.63-2.17-17.08-5.01-25.41-8.13-10.49-4.12-20.79-8.75-30.64-14.25-2.14-1.15-4.28-2.29-6.35-3.57-11.22-6.58-21.86-14.1-31.92-22.34-34.68-28.41-61.41-66.43-76.35-108.7-3.09-8.74-5.71-17.65-7.8-26.68-1.48-6.16-2.52-12.42-3.58-18.66-.4-2.35-.61-4.73-.95-7.09-.6-3.96-.75-7.96-1.17-11.94-.8-9.47-.71-18.99-.51-28.49.14-3.51.34-7.01.7-10.51.31-3.17.46-6.37.92-9.52.41-2.81.65-5.65 1.16-8.44.7-3.94 1.3-7.9 2.12-11.82 3.43-16.52 8.47-32.73 15.26-48.18 1.15-2.92 2.59-5.72 3.86-8.59 8.05-16.71 17.9-32.56 29.49-47.06 20-25.38 45.1-46.68 73.27-62.47 7.5-4.15 15.16-8.05 23.07-11.37 15.82-6.88 32.41-11.95 49.31-15.38 3.51-.67 7.04-1.24 10.56-1.85 2.62-.47 5.28-.7 7.91-1.08 3.53-.53 7.1-.68 10.65-1.04 2.46-.24 4.91-.36 7.36-.51m8.64 24.41c-9.23.1-18.43.99-27.57 2.23-7.3 1.08-14.53 2.6-21.71 4.3-13.91 3.5-27.48 8.34-40.46 14.42-10.46 4.99-20.59 10.7-30.18 17.22-4.18 2.92-8.4 5.8-12.34 9.03-5.08 3.97-9.98 8.17-14.68 12.59-2.51 2.24-4.81 4.7-7.22 7.06-28.22 28.79-48.44 65.39-57.5 104.69-2.04 8.44-3.54 17.02-4.44 25.65-1.1 8.89-1.44 17.85-1.41 26.8.11 7.14.38 14.28 1.22 21.37.62 7.12 1.87 14.16 3.2 21.18 1.07 4.65 2.03 9.32 3.33 13.91 6.29 23.38 16.5 45.7 30.07 65.75 8.64 12.98 18.78 24.93 29.98 35.77 16.28 15.82 35.05 29.04 55.34 39.22 7.28 3.52 14.66 6.87 22.27 9.63 5.04 1.76 10.06 3.57 15.22 4.98 11.26 3.23 22.77 5.6 34.39 7.06 2.91.29 5.81.61 8.72.9 13.82 1.08 27.74 1 41.54-.43 4.45-.6 8.92-.99 13.35-1.78 3.63-.67 7.28-1.25 10.87-2.1 4.13-.98 8.28-1.91 12.36-3.07 26.5-7.34 51.58-19.71 73.58-36.2 15.78-11.82 29.96-25.76 42.12-41.28 3.26-4.02 6.17-8.31 9.13-12.55 3.39-5.06 6.58-10.25 9.6-15.54 2.4-4.44 4.74-8.91 6.95-13.45 5.69-12.05 10.28-24.62 13.75-37.49 2.59-10.01 4.75-20.16 5.9-30.45 1.77-13.47 1.94-27.1 1.29-40.65-.29-3.89-.67-7.77-1-11.66-2.23-19.08-6.79-37.91-13.82-55.8-5.95-15.13-13.53-29.63-22.61-43.13-12.69-18.8-28.24-35.68-45.97-49.83-25.05-20-54.47-34.55-85.65-42.08-7.78-1.93-15.69-3.34-23.63-4.45-3.91-.59-7.85-.82-11.77-1.24-7.39-.57-14.81-.72-22.22-.58zM139.26 83.53c13.3-8.89 28.08-15.38 43.3-20.18-3.17 1.77-6.44 3.38-9.53 5.29-11.21 6.68-21.52 14.9-30.38 24.49-6.8 7.43-12.76 15.73-17.01 24.89-3.29 6.86-5.64 14.19-6.86 21.71-.93 4.85-1.3 9.81-1.17 14.75.13 13.66 4.44 27.08 11.29 38.82 5.92 10.22 13.63 19.33 22.36 27.26 4.85 4.36 10.24 8.09 14.95 12.6 2.26 2.19 4.49 4.42 6.43 6.91 2.62 3.31 4.89 6.99 5.99 11.1.9 3.02.66 6.2.69 9.31.02 4.1-.04 8.2.03 12.3.14 3.54-.02 7.09.11 10.63.08 2.38.02 4.76.05 7.14.16 5.77.06 11.53.15 17.3.11 2.91.02 5.82.13 8.74.03 1.63.13 3.28-.03 4.91-.91.12-1.82.18-2.73.16-10.99 0-21.88-2.63-31.95-6.93-6-2.7-11.81-5.89-17.09-9.83-5.75-4.19-11.09-8.96-15.79-14.31-6.53-7.24-11.98-15.39-16.62-23.95-1.07-2.03-2.24-4.02-3.18-6.12-1.16-2.64-2.62-5.14-3.67-7.82-4.05-9.68-6.57-19.94-8.08-30.31-.49-4.44-1.09-8.88-1.2-13.35-.7-15.73.84-31.55 4.67-46.82 2.12-8.15 4.77-16.18 8.31-23.83 6.32-14.2 15.34-27.18 26.3-38.19 6.28-6.2 13.13-11.84 20.53-16.67zm175.37-20.12c2.74.74 5.41 1.74 8.09 2.68 6.36 2.33 12.68 4.84 18.71 7.96 13.11 6.44 25.31 14.81 35.82 24.97 10.2 9.95 18.74 21.6 25.14 34.34 1.28 2.75 2.64 5.46 3.81 8.26 6.31 15.1 10 31.26 11.23 47.57.41 4.54.44 9.09.45 13.64.07 11.64-1.49 23.25-4.3 34.53-1.97 7.27-4.35 14.49-7.86 21.18-3.18 6.64-6.68 13.16-10.84 19.24-6.94 10.47-15.6 19.87-25.82 27.22-10.48 7.64-22.64 13.02-35.4 15.38-3.51.69-7.08 1.08-10.66 1.21-1.85.06-3.72.16-5.56-.1-.28-2.15 0-4.31-.01-6.46-.03-3.73.14-7.45.1-11.17.19-7.02.02-14.05.21-21.07.03-2.38-.03-4.76.03-7.14.17-5.07-.04-10.14.14-15.21.1-2.99-.24-6.04.51-8.96.66-2.5 1.78-4.86 3.09-7.08 4.46-7.31 11.06-12.96 17.68-18.26 5.38-4.18 10.47-8.77 15.02-13.84 7.68-8.37 14.17-17.88 18.78-28.27 2.5-5.93 4.52-12.1 5.55-18.46.86-4.37 1.06-8.83 1.01-13.27-.02-7.85-1.4-15.65-3.64-23.17-1.75-5.73-4.27-11.18-7.09-16.45-3.87-6.93-8.65-13.31-13.96-19.2-9.94-10.85-21.75-19.94-34.6-27.1-1.85-1.02-3.84-1.82-5.63-2.97zm-100.8 58.45c.98-1.18 1.99-2.33 3.12-3.38-.61.93-1.27 1.81-1.95 2.68-3.1 3.88-5.54 8.31-7.03 13.06-.87 3.27-1.68 6.6-1.73 10-.07 2.52-.08 5.07.32 7.57 1.13 7.63 4.33 14.85 8.77 21.12 2 2.7 4.25 5.27 6.92 7.33 1.62 1.27 3.53 2.09 5.34 3.05 3.11 1.68 6.32 3.23 9.07 5.48 2.67 2.09 4.55 5.33 4.4 8.79-.01 73.67 0 147.34-.01 221.02 0 1.35-.08 2.7.04 4.04.13 1.48.82 2.83 1.47 4.15.86 1.66 1.78 3.34 3.18 4.62.85.77 1.97 1.4 3.15 1.24 1.5-.2 2.66-1.35 3.45-2.57.96-1.51 1.68-3.16 2.28-4.85.76-2.13.44-4.42.54-6.63.14-4.03-.02-8.06.14-12.09.03-5.89.03-11.77.06-17.66.14-3.62.03-7.24.11-10.86.15-4.03-.02-8.06.14-12.09.03-5.99.03-11.98.07-17.97.14-3.62.02-7.24.11-10.86.14-3.93-.02-7.86.14-11.78.03-5.99.03-11.98.06-17.97.16-3.94-.01-7.88.19-11.82.29 1.44.13 2.92.22 4.38.19 3.61.42 7.23.76 10.84.32 3.44.44 6.89.86 10.32.37 3.1.51 6.22.95 9.31.57 4.09.87 8.21 1.54 12.29 1.46 9.04 2.83 18.11 5.09 26.99 1.13 4.82 2.4 9.61 4 14.3 2.54 7.9 5.72 15.67 10.31 22.62 1.73 2.64 3.87 4.98 6.1 7.21.27.25.55.51.88.71.6.25 1.31-.07 1.7-.57.71-.88 1.17-1.94 1.7-2.93 4.05-7.8 8.18-15.56 12.34-23.31.7-1.31 1.44-2.62 2.56-3.61 1.75-1.57 3.84-2.69 5.98-3.63 2.88-1.22 5.9-2.19 9.03-2.42 6.58-.62 13.11.75 19.56 1.85 3.69.58 7.4 1.17 11.13 1.41 3.74.1 7.48.05 11.21-.28 8.55-.92 16.99-2.96 24.94-6.25 5.3-2.24 10.46-4.83 15.31-7.93 11.46-7.21 21.46-16.57 30.04-27.01 1.17-1.42 2.25-2.9 3.46-4.28-1.2 3.24-2.67 6.37-4.16 9.48-1.25 2.9-2.84 5.61-4.27 8.42-5.16 9.63-11.02 18.91-17.75 27.52-4.03 5.21-8.53 10.05-13.33 14.57-6.64 6.05-14.07 11.37-22.43 14.76-8.21 3.37-17.31 4.63-26.09 3.29-3.56-.58-7.01-1.69-10.41-2.88-2.79-.97-5.39-2.38-8.03-3.69-3.43-1.71-6.64-3.81-9.71-6.08 2.71 3.06 5.69 5.86 8.7 8.61 4.27 3.76 8.74 7.31 13.63 10.23 3.98 2.45 8.29 4.4 12.84 5.51 1.46.37 2.96.46 4.45.6-1.25 1.1-2.63 2.04-3.99 2.98-9.61 6.54-20.01 11.86-30.69 16.43-20.86 8.7-43.17 13.97-65.74 15.34-4.66.24-9.32.36-13.98.36-4.98-.11-9.97-.13-14.92-.65-11.2-.76-22.29-2.73-33.17-5.43-10.35-2.71-20.55-6.12-30.3-10.55-8.71-3.86-17.12-8.42-24.99-13.79-1.83-1.31-3.74-2.53-5.37-4.08 6.6-1.19 13.03-3.39 18.99-6.48 5.74-2.86 10.99-6.66 15.63-11.07 2.24-2.19 4.29-4.59 6.19-7.09-3.43 2.13-6.93 4.15-10.62 5.78-4.41 2.16-9.07 3.77-13.81 5.02-5.73 1.52-11.74 1.73-17.61 1.14-8.13-.95-15.86-4.27-22.51-8.98-4.32-2.94-8.22-6.43-11.96-10.06-9.93-10.16-18.2-21.81-25.66-33.86-3.94-6.27-7.53-12.75-11.12-19.22-1.05-2.04-2.15-4.05-3.18-6.1 2.85 2.92 5.57 5.97 8.43 8.88 8.99 8.97 18.56 17.44 29.16 24.48 7.55 4.9 15.67 9.23 24.56 11.03 3.11.73 6.32.47 9.47.81 2.77.28 5.56.2 8.34.3 5.05.06 10.11.04 15.16-.16 3.65-.16 7.27-.66 10.89-1.09 2.07-.25 4.11-.71 6.14-1.2 3.88-.95 8.11-.96 11.83.61 4.76 1.85 8.44 5.64 11.38 9.71 2.16 3.02 4.06 6.22 5.66 9.58 1.16 2.43 2.46 4.79 3.55 7.26 1 2.24 2.15 4.42 3.42 6.52.67 1.02 1.4 2.15 2.62 2.55 1.06-.75 1.71-1.91 2.28-3.03 2.1-4.16 3.42-8.65 4.89-13.05 2.02-6.59 3.78-13.27 5.19-20.02 2.21-9.25 3.25-18.72 4.54-28.13.56-3.98.83-7.99 1.31-11.97.87-10.64 1.9-21.27 2.24-31.94.08-1.86.24-3.71.25-5.57.01-4.35.25-8.69.22-13.03-.01-2.38-.01-4.76 0-7.13.05-5.07-.2-10.14-.22-15.21-.2-6.61-.71-13.2-1.29-19.78-.73-5.88-1.55-11.78-3.12-17.51-2.05-7.75-5.59-15.03-9.8-21.82-3.16-5.07-6.79-9.88-11.09-14.03-3.88-3.86-8.58-7.08-13.94-8.45-1.5-.41-3.06-.45-4.59-.64.07-2.99.7-5.93 1.26-8.85 1.59-7.71 3.8-15.3 6.76-22.6 1.52-4.03 3.41-7.9 5.39-11.72 3.45-6.56 7.62-12.79 12.46-18.46zm31.27 1.7c.35-.06.71-.12 1.07-.19.19 1.79.09 3.58.1 5.37v38.13c-.01 1.74.13 3.49-.15 5.22-.36-.03-.71-.05-1.06-.05-.95-3.75-1.72-7.55-2.62-11.31-.38-1.53-.58-3.09-1.07-4.59-1.7-.24-3.43-.17-5.15-.2-5.06-.01-10.13 0-15.19-.01-1.66-.01-3.32.09-4.98-.03-.03-.39-.26-.91.16-1.18 1.28-.65 2.72-.88 4.06-1.35 3.43-1.14 6.88-2.16 10.31-3.31 1.39-.48 2.9-.72 4.16-1.54.04-.56.02-1.13-.05-1.68-1.23-.55-2.53-.87-3.81-1.28-3.13-1.03-6.29-1.96-9.41-3.02-1.79-.62-3.67-1-5.41-1.79-.03-.37-.07-.73-.11-1.09 5.09-.19 10.2.06 15.3-.12 3.36-.13 6.73.08 10.09-.07.12-.39.26-.77.37-1.16 1.08-4.94 2.33-9.83 3.39-14.75zm5.97-.2c.36.05.72.12 1.08.2.98 3.85 1.73 7.76 2.71 11.61.36 1.42.56 2.88 1.03 4.27 2.53.18 5.07-.01 7.61.05 5.16.12 10.33.12 15.49.07.76-.01 1.52.03 2.28.08-.04.36-.07.72-.1 1.08-1.82.83-3.78 1.25-5.67 1.89-3.73 1.23-7.48 2.39-11.22 3.57-.57.17-1.12.42-1.67.64-.15.55-.18 1.12-.12 1.69.87.48 1.82.81 2.77 1.09 4.88 1.52 9.73 3.14 14.63 4.6.38.13.78.27 1.13.49.4.27.23.79.15 1.18-1.66.13-3.31.03-4.97.04-5.17.01-10.33-.01-15.5.01-1.61.03-3.22-.02-4.82.21-.52 1.67-.72 3.42-1.17 5.11-.94 3.57-1.52 7.24-2.54 10.78-.36.01-.71.02-1.06.06-.29-1.73-.15-3.48-.15-5.22v-38.13c.02-1.78-.08-3.58.11-5.37zM65.05 168.33c1.12-2.15 2.08-4.4 3.37-6.46-1.82 7.56-2.91 15.27-3.62 23-.8 7.71-.85 15.49-.54 23.23 1.05 19.94 5.54 39.83 14.23 57.88 2.99 5.99 6.35 11.83 10.5 17.11 6.12 7.47 12.53 14.76 19.84 21.09 4.8 4.1 9.99 7.78 15.54 10.8 3.27 1.65 6.51 3.39 9.94 4.68 5.01 2.03 10.19 3.61 15.42 4.94 3.83.96 7.78 1.41 11.52 2.71 5 1.57 9.47 4.61 13.03 8.43 4.93 5.23 8.09 11.87 10.2 18.67.99 2.9 1.59 5.91 2.17 8.92.15.75.22 1.52.16 2.29-6.5 2.78-13.26 5.06-20.26 6.18-4.11.78-8.29.99-12.46 1.08-10.25.24-20.47-1.76-30.12-5.12-3.74-1.42-7.49-2.85-11.03-4.72-8.06-3.84-15.64-8.7-22.46-14.46-2.92-2.55-5.83-5.13-8.4-8.03-9.16-9.83-16.3-21.41-21.79-33.65-2.39-5.55-4.61-11.18-6.37-16.96-1.17-3.94-2.36-7.89-3.26-11.91-.75-2.94-1.22-5.95-1.87-8.92-.46-2.14-.69-4.32-1.03-6.48-.85-5.43-1.28-10.93-1.33-16.43.11-6.18.25-12.37 1.07-18.5.4-2.86.67-5.74 1.15-8.6.98-5.7 2.14-11.37 3.71-16.93 3.09-11.65 7.48-22.95 12.69-33.84zm363.73-6.44c1.1 1.66 1.91 3.48 2.78 5.26 2.1 4.45 4.24 8.9 6.02 13.49 7.61 18.76 12.3 38.79 13.04 59.05.02 1.76.07 3.52.11 5.29.13 9.57-1.27 19.09-3.18 28.45-.73 3.59-1.54 7.17-2.58 10.69-4.04 14.72-10 29-18.41 41.78-8.21 12.57-19.01 23.55-31.84 31.41-5.73 3.59-11.79 6.64-18.05 9.19-5.78 2.19-11.71 4.03-17.8 5.11-6.4 1.05-12.91 1.52-19.4 1.23-7.92-.48-15.78-2.07-23.21-4.85-1.94-.8-3.94-1.46-5.84-2.33-.21-1.51.25-2.99.53-4.46 1.16-5.74 3.03-11.36 5.7-16.58 2.37-4.51 5.52-8.65 9.46-11.9 2.43-2.05 5.24-3.61 8.16-4.83 3.58-1.5 7.47-1.97 11.24-2.83 7.23-1.71 14.37-3.93 21.15-7 10.35-4.65 19.71-11.38 27.65-19.46 1.59-1.61 3.23-3.18 4.74-4.87 3.37-3.76 6.71-7.57 9.85-11.53 7.48-10.07 12.82-21.59 16.71-33.48 1.58-5.3 3.21-10.6 4.21-16.05.63-2.87 1.04-5.78 1.52-8.68.87-6.09 1.59-12.22 1.68-18.38.12-6.65.14-13.32-.53-19.94-.73-7.99-1.87-15.96-3.71-23.78z"],odysee:[512,512,[],"e5c6","M406.7 463c-42.3 30.8-94.4 49-150.7 49C144.9 512 50.3 441.2 14.9 342.2c2.4 1.7 5.9 3.6 7.9 4.4c16.3 7.4 40.1-5.4 62.9-28.7c6.9-6.9 14.4-12.4 22.8-17.3c18.3-11.9 37.6-20.8 58.4-27.2c0 0 22.3 34.2 43.1 74.8s-22.3 54-27.2 54c-.3 0-.8 0-1.5-.1c-11-.5-70-3-56 51.1c14.9 57.4 97.5 36.6 139.6 8.9s31.7-118.3 31.7-118.3c41.1-6.4 54 37.1 57.9 59.4c.8 4.6 1.1 9.9 1.4 15.5c1.1 21.2 2.3 45.6 35.3 46.4c5.3 0 10.6-.8 15.5-2zm-95.3-23.7c-2-.5-3.5-2.5-3-5c1-2.5 3-3.5 5-3s3.5 3 3 5s-2.5 3.5-5 3zm-207-95.6c1.5-.5 3.5 1 4 3c0 2-1 4-3 4c-1.5 .5-3.5-1-4-3c-.5-1.5 1-3.5 3-4zM451.8 421C489.3 376.4 512 318.8 512 256c0-67.5-26.1-128.9-68.8-174.7c-.1 23.5-6.1 48.2-16.8 69.2c-11.9 20.3-49 58.9-69.8 78.7c-.7 .3-1.1 .9-1.5 1.4c-.2 .2-.3 .4-.5 .6c-5 6.9-4 16.8 3 21.8c21.3 15.8 56.4 45.6 59.4 72.8c3.5 34.9 27.9 75.6 34.2 86.2l0 0c.8 1.3 1.3 2.1 1.4 2.4c0 2.2-.4 4.3-.8 6.5zM390.7 251c-.5 3 1 5.9 4 6.4s5.9-1 6.4-4s-1-5.9-4-6.4c-3-1-5.9 1-6.4 4zm61.4-60.9l-11.4 5.4-3 12.9-5.4-11.4-12.9-3 11.4-5.4 3-12.9 5.4 11.4 12.9 3zM395.5 41.3c-16.2 8.2-22.1 32.8-29 61.4l0 0c-.3 1.4-.7 2.8-1 4.2c-9.5 38.5-30.6 37.6-41.7 37.2c-1.1 0-2-.1-2.9-.1c-5.1 0-6-4-8.9-17.1c-2.6-12.1-6.9-32-17.9-63.6C271.4-2.5 211.4 13.9 165.9 41.1C110.6 74.2 131.5 143 146.1 190.5c.7 2.2 1.4 4.4 2 6.6c-4 4-13.8 7.5-26 11.9c-12.1 4.3-26.6 9.5-40.3 16.9C47.9 243.9 11.5 274.9 2 288.5C.7 277.8 0 267 0 256C0 114.6 114.6 0 256 0c51.4 0 99.4 15.2 139.5 41.3zM58.9 189.6c-1.5-2-4.5-3-6.4-1.5s-3 4.5-1.5 6.4s4.5 3 6.4 1.5c2.5-1.5 3-4.5 1.5-6.4zM327.3 64.9c2-1.5 5-.5 6.4 1.5c1.5 2.5 1 5.4-1.5 6.4c-2 1.5-5 .5-6.4-1.5s-.5-5 1.5-6.4zM95.1 105c-.5 1.5 .5 3 2 3c1.5 .5 3-.5 3-2c.5-1.5-.5-3-2-3s-3 .5-3 2zm84.7-.5c-3.5-43.1 37.1-54 37.1-54c44.1-15.4 56 5.9 66.4 37.6s3 42.6-38.6 58.9s-61.9-4.5-64.9-42.6zm89.6 14.9h1c2.5 0 5-2 5-5c2-6.9 1-14.4-2-20.8c-1.5-2-4-3.5-6.4-2.5c-3 1-4.5 4-3.5 6.9c2 4.5 3 9.9 1.5 14.9c-.5 3 1.5 5.9 4.5 6.4zm-9.9-41.6c-2 0-4-1-5-3s-2-3.5-3-5c-2-2-2-5.4 0-7.4s5.4-2 7.4 0c2 2.5 3.5 5 5 7.4s.5 5.9-2.5 7.4c-.6 0-1 .2-1.3 .3c-.2 .1-.4 .2-.6 .2z"],"square-whatsapp":[448,512,["whatsapp-square"],"f40c","M92.1 254.6c0 24.9 7 49.2 20.2 70.1l3.1 5-13.3 48.6L152 365.2l4.8 2.9c20.2 12 43.4 18.4 67.1 18.4h.1c72.6 0 133.3-59.1 133.3-131.8c0-35.2-15.2-68.3-40.1-93.2c-25-25-58-38.7-93.2-38.7c-72.7 0-131.8 59.1-131.9 131.8zM274.8 330c-12.6 1.9-22.4 .9-47.5-9.9c-36.8-15.9-61.8-51.5-66.9-58.7c-.4-.6-.7-.9-.8-1.1c-2-2.6-16.2-21.5-16.2-41c0-18.4 9-27.9 13.2-32.3c.3-.3 .5-.5 .7-.8c3.6-4 7.9-5 10.6-5c2.6 0 5.3 0 7.6 .1c.3 0 .5 0 .8 0c2.3 0 5.2 0 8.1 6.8c1.2 2.9 3 7.3 4.9 11.8c3.3 8 6.7 16.3 7.3 17.6c1 2 1.7 4.3 .3 6.9c-3.4 6.8-6.9 10.4-9.3 13c-3.1 3.2-4.5 4.7-2.3 8.6c15.3 26.3 30.6 35.4 53.9 47.1c4 2 6.3 1.7 8.6-1c2.3-2.6 9.9-11.6 12.5-15.5c2.6-4 5.3-3.3 8.9-2s23.1 10.9 27.1 12.9c.8 .4 1.5 .7 2.1 1c2.8 1.4 4.7 2.3 5.5 3.6c.9 1.9 .9 9.9-2.4 19.1c-3.3 9.3-19.1 17.7-26.7 18.8zM448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM148.1 393.9L64 416l22.5-82.2c-13.9-24-21.2-51.3-21.2-79.3C65.4 167.1 136.5 96 223.9 96c42.4 0 82.2 16.5 112.2 46.5c29.9 30 47.9 69.8 47.9 112.2c0 87.4-72.7 158.5-160.1 158.5c-26.6 0-52.7-6.7-75.8-19.3z"],"node-js":[448,512,[],"f3d3","M224 508c-6.7 0-13.5-1.8-19.4-5.2l-61.7-36.5c-9.2-5.2-4.7-7-1.7-8 12.3-4.3 14.8-5.2 27.9-12.7 1.4-.8 3.2-.5 4.6.4l47.4 28.1c1.7 1 4.1 1 5.7 0l184.7-106.6c1.7-1 2.8-3 2.8-5V149.3c0-2.1-1.1-4-2.9-5.1L226.8 37.7c-1.7-1-4-1-5.7 0L36.6 144.3c-1.8 1-2.9 3-2.9 5.1v213.1c0 2 1.1 4 2.9 4.9l50.6 29.2c27.5 13.7 44.3-2.4 44.3-18.7V167.5c0-3 2.4-5.3 5.4-5.3h23.4c2.9 0 5.4 2.3 5.4 5.3V378c0 36.6-20 57.6-54.7 57.6-10.7 0-19.1 0-42.5-11.6l-48.4-27.9C8.1 389.2.7 376.3.7 362.4V149.3c0-13.8 7.4-26.8 19.4-33.7L204.6 9c11.7-6.6 27.2-6.6 38.8 0l184.7 106.7c12 6.9 19.4 19.8 19.4 33.7v213.1c0 13.8-7.4 26.7-19.4 33.7L243.4 502.8c-5.9 3.4-12.6 5.2-19.4 5.2zm149.1-210.1c0-39.9-27-50.5-83.7-58-57.4-7.6-63.2-11.5-63.2-24.9 0-11.1 4.9-25.9 47.4-25.9 37.9 0 51.9 8.2 57.7 33.8.5 2.4 2.7 4.2 5.2 4.2h24c1.5 0 2.9-.6 3.9-1.7s1.5-2.6 1.4-4.1c-3.7-44.1-33-64.6-92.2-64.6-52.7 0-84.1 22.2-84.1 59.5 0 40.4 31.3 51.6 81.8 56.6 60.5 5.9 65.2 14.8 65.2 26.7 0 20.6-16.6 29.4-55.5 29.4-48.9 0-59.6-12.3-63.2-36.6-.4-2.6-2.6-4.5-5.3-4.5h-23.9c-3 0-5.3 2.4-5.3 5.3 0 31.1 16.9 68.2 97.8 68.2 58.4-.1 92-23.2 92-63.4z"],"edge-legacy":[512,512,[],"e078","M25.71,228.16l.35-.48c0,.16,0,.32-.07.48Zm460.58,15.51c0-44-7.76-84.46-28.81-122.4C416.5,47.88,343.91,8,258.89,8,119,7.72,40.62,113.21,26.06,227.68c42.42-61.31,117.07-121.38,220.37-125,0,0,109.67,0,99.42,105H170c6.37-37.39,18.55-59,34.34-78.93-75.05,34.9-121.85,96.1-120.75,188.32.83,71.45,50.13,144.84,120.75,172,83.35,31.84,192.77,7.2,240.13-21.33V363.31C363.6,419.8,173.6,424.23,172.21,295.74H486.29V243.67Z"],slack:[448,512,[62447,"slack-hash"],"f198","M94.12 315.1c0 25.9-21.16 47.06-47.06 47.06S0 341 0 315.1c0-25.9 21.16-47.06 47.06-47.06h47.06v47.06zm23.72 0c0-25.9 21.16-47.06 47.06-47.06s47.06 21.16 47.06 47.06v117.84c0 25.9-21.16 47.06-47.06 47.06s-47.06-21.16-47.06-47.06V315.1zm47.06-188.98c-25.9 0-47.06-21.16-47.06-47.06S139 32 164.9 32s47.06 21.16 47.06 47.06v47.06H164.9zm0 23.72c25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06H47.06C21.16 243.96 0 222.8 0 196.9s21.16-47.06 47.06-47.06H164.9zm188.98 47.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06h-47.06V196.9zm-23.72 0c0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06V79.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06V196.9zM283.1 385.88c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06v-47.06h47.06zm0-23.72c-25.9 0-47.06-21.16-47.06-47.06 0-25.9 21.16-47.06 47.06-47.06h117.84c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06H283.1z"],medrt:[544,512,[],"f3c8","M113.7 256c0 121.8 83.9 222.8 193.5 241.1-18.7 4.5-38.2 6.9-58.2 6.9C111.4 504 0 393 0 256S111.4 8 248.9 8c20.1 0 39.6 2.4 58.2 6.9C197.5 33.2 113.7 134.2 113.7 256m297.4 100.3c-77.7 55.4-179.6 47.5-240.4-14.6 5.5 14.1 12.7 27.7 21.7 40.5 61.6 88.2 182.4 109.3 269.7 47 87.3-62.3 108.1-184.3 46.5-272.6-9-12.9-19.3-24.3-30.5-34.2 37.4 78.8 10.7 178.5-67 233.9m-218.8-244c-1.4 1-2.7 2.1-4 3.1 64.3-17.8 135.9 4 178.9 60.5 35.7 47 42.9 106.6 24.4 158 56.7-56.2 67.6-142.1 22.3-201.8-50-65.5-149.1-74.4-221.6-19.8M296 224c-4.4 0-8-3.6-8-8v-40c0-4.4-3.6-8-8-8h-48c-4.4 0-8 3.6-8 8v40c0 4.4-3.6 8-8 8h-40c-4.4 0-8 3.6-8 8v48c0 4.4 3.6 8 8 8h40c4.4 0 8 3.6 8 8v40c0 4.4 3.6 8 8 8h48c4.4 0 8-3.6 8-8v-40c0-4.4 3.6-8 8-8h40c4.4 0 8-3.6 8-8v-48c0-4.4-3.6-8-8-8h-40z"],usb:[640,512,[],"f287","M641.5 256c0 3.1-1.7 6.1-4.5 7.5L547.9 317c-1.4.8-2.8 1.4-4.5 1.4-1.4 0-3.1-.3-4.5-1.1-2.8-1.7-4.5-4.5-4.5-7.8v-35.6H295.7c25.3 39.6 40.5 106.9 69.6 106.9H392V354c0-5 3.9-8.9 8.9-8.9H490c5 0 8.9 3.9 8.9 8.9v89.1c0 5-3.9 8.9-8.9 8.9h-89.1c-5 0-8.9-3.9-8.9-8.9v-26.7h-26.7c-75.4 0-81.1-142.5-124.7-142.5H140.3c-8.1 30.6-35.9 53.5-69 53.5C32 327.3 0 295.3 0 256s32-71.3 71.3-71.3c33.1 0 61 22.8 69 53.5 39.1 0 43.9 9.5 74.6-60.4C255 88.7 273 95.7 323.8 95.7c7.5-20.9 27-35.6 50.4-35.6 29.5 0 53.5 23.9 53.5 53.5s-23.9 53.5-53.5 53.5c-23.4 0-42.9-14.8-50.4-35.6H294c-29.1 0-44.3 67.4-69.6 106.9h310.1v-35.6c0-3.3 1.7-6.1 4.5-7.8 2.8-1.7 6.4-1.4 8.9.3l89.1 53.5c2.8 1.1 4.5 4.1 4.5 7.2z"],tumblr:[320,512,[],"f173","M309.8 480.3c-13.6 14.5-50 31.7-97.4 31.7-120.8 0-147-88.8-147-140.6v-144H17.9c-5.5 0-10-4.5-10-10v-68c0-7.2 4.5-13.6 11.3-16 62-21.8 81.5-76 84.3-117.1.8-11 6.5-16.3 16.1-16.3h70.9c5.5 0 10 4.5 10 10v115.2h83c5.5 0 10 4.4 10 9.9v81.7c0 5.5-4.5 10-10 10h-83.4V360c0 34.2 23.7 53.6 68 35.8 4.8-1.9 9-3.2 12.7-2.2 3.5.9 5.8 3.4 7.4 7.9l22 64.3c1.8 5 3.3 10.6-.4 14.5z"],vaadin:[448,512,[],"f408","M224.5 140.7c1.5-17.6 4.9-52.7 49.8-52.7h98.6c20.7 0 32.1-7.8 32.1-21.6V54.1c0-12.2 9.3-22.1 21.5-22.1S448 41.9 448 54.1v36.5c0 42.9-21.5 62-66.8 62H280.7c-30.1 0-33 14.7-33 27.1 0 1.3-.1 2.5-.2 3.7-.7 12.3-10.9 22.2-23.4 22.2s-22.7-9.8-23.4-22.2c-.1-1.2-.2-2.4-.2-3.7 0-12.3-3-27.1-33-27.1H66.8c-45.3 0-66.8-19.1-66.8-62V54.1C0 41.9 9.4 32 21.6 32s21.5 9.9 21.5 22.1v12.3C43.1 80.2 54.5 88 75.2 88h98.6c44.8 0 48.3 35.1 49.8 52.7h.9zM224 456c11.5 0 21.4-7 25.7-16.3 1.1-1.8 97.1-169.6 98.2-171.4 11.9-19.6-3.2-44.3-27.2-44.3-13.9 0-23.3 6.4-29.8 20.3L224 362l-66.9-117.7c-6.4-13.9-15.9-20.3-29.8-20.3-24 0-39.1 24.6-27.2 44.3 1.1 1.9 97.1 169.6 98.2 171.4 4.3 9.3 14.2 16.3 25.7 16.3z"],quora:[448,512,[],"f2c4","M440.5 386.7h-29.3c-1.5 13.5-10.5 30.8-33 30.8-20.5 0-35.3-14.2-49.5-35.8 44.2-34.2 74.7-87.5 74.7-153C403.5 111.2 306.8 32 205 32 105.3 32 7.3 111.7 7.3 228.7c0 134.1 131.3 221.6 249 189C276 451.3 302 480 351.5 480c81.8 0 90.8-75.3 89-93.3zM297 329.2C277.5 300 253.3 277 205.5 277c-30.5 0-54.3 10-69 22.8l12.2 24.3c6.2-3 13-4 19.8-4 35.5 0 53.7 30.8 69.2 61.3-10 3-20.7 4.2-32.7 4.2-75 0-107.5-53-107.5-156.7C97.5 124.5 130 71 205 71c76.2 0 108.7 53.5 108.7 157.7.1 41.8-5.4 75.6-16.7 100.5z"],"square-x-twitter":[448,512,[],"e61a","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm297.1 84L257.3 234.6 379.4 396H283.8L209 298.1 123.3 396H75.8l111-126.9L69.7 116h98l67.7 89.5L313.6 116h47.5zM323.3 367.6L153.4 142.9H125.1L296.9 367.6h26.3z"],reacteurope:[576,512,[],"f75d","M250.6 211.74l5.8-4.1 5.8 4.1-2.1-6.8 5.7-4.3-7.1-.1-2.3-6.8-2.3 6.8-7.2.1 5.7 4.3zm63.7 0l5.8-4.1 5.8 4.1-2.1-6.8 5.7-4.3-7.2-.1-2.3-6.8-2.3 6.8-7.2.1 5.7 4.3zm-91.3 50.5h-3.4c-4.8 0-3.8 4-3.8 12.1 0 4.7-2.3 6.1-5.8 6.1s-5.8-1.4-5.8-6.1v-36.6c0-4.7 2.3-6.1 5.8-6.1s5.8 1.4 5.8 6.1c0 7.2-.7 10.5 3.8 10.5h3.4c4.7-.1 3.8-3.9 3.8-12.3 0-9.9-6.7-14.1-16.8-14.1h-.2c-10.1 0-16.8 4.2-16.8 14.1V276c0 10.4 6.7 14.1 16.8 14.1h.2c10.1 0 16.8-3.8 16.8-14.1 0-9.86 1.1-13.76-3.8-13.76zm-80.7 17.4h-14.7v-19.3H139c2.5 0 3.8-1.3 3.8-3.8v-2.1c0-2.5-1.3-3.8-3.8-3.8h-11.4v-18.3H142c2.5 0 3.8-1.3 3.8-3.8v-2.1c0-2.5-1.3-3.8-3.8-3.8h-21.7c-2.4-.1-3.7 1.3-3.7 3.8v59.1c0 2.5 1.3 3.8 3.8 3.8h21.9c2.5 0 3.8-1.3 3.8-3.8v-2.1c0-2.5-1.3-3.8-3.8-3.8zm-42-18.5c4.6-2 7.3-6 7.3-12.4v-11.9c0-10.1-6.7-14.1-16.8-14.1H77.4c-2.5 0-3.8 1.3-3.8 3.8v59.1c0 2.5 1.3 3.8 3.8 3.8h3.4c2.5 0 3.8-1.3 3.8-3.8v-22.9h5.6l7.4 23.5a4.1 4.1 0 0 0 4.3 3.2h3.3c2.8 0 4-1.8 3.2-4.4zm-3.8-14c0 4.8-2.5 6.1-6.1 6.1h-5.8v-20.9h5.8c3.6 0 6.1 1.3 6.1 6.1zM176 226a3.82 3.82 0 0 0-4.2-3.4h-6.9a3.68 3.68 0 0 0-4 3.4l-11 59.2c-.5 2.7.9 4.1 3.4 4.1h3a3.74 3.74 0 0 0 4.1-3.5l1.8-11.3h12.2l1.8 11.3a3.74 3.74 0 0 0 4.1 3.5h3.5c2.6 0 3.9-1.4 3.4-4.1zm-12.3 39.3l4.7-29.7 4.7 29.7zm89.3 20.2v-53.2h7.5c2.5 0 3.8-1.3 3.8-3.8v-2.1c0-2.5-1.3-3.8-3.8-3.8h-25.8c-2.5 0-3.8 1.3-3.8 3.8v2.1c0 2.5 1.3 3.8 3.8 3.8h7.3v53.2c0 2.5 1.3 3.8 3.8 3.8h3.4c2.5.04 3.8-1.3 3.8-3.76zm248-.8h-19.4V258h16.1a1.89 1.89 0 0 0 2-2v-.8a1.89 1.89 0 0 0-2-2h-16.1v-25.8h19.1a1.89 1.89 0 0 0 2-2v-.8a1.77 1.77 0 0 0-2-1.9h-22.2a1.62 1.62 0 0 0-2 1.8v63a1.81 1.81 0 0 0 2 1.9H501a1.81 1.81 0 0 0 2-1.9v-.8a1.84 1.84 0 0 0-2-1.96zm-93.1-62.9h-.8c-10.1 0-15.3 4.7-15.3 14.1V276c0 9.3 5.2 14.1 15.3 14.1h.8c10.1 0 15.3-4.8 15.3-14.1v-40.1c0-9.36-5.2-14.06-15.3-14.06zm10.2 52.4c-.1 8-3 11.1-10.5 11.1s-10.5-3.1-10.5-11.1v-36.6c0-7.9 3-11.1 10.5-11.1s10.5 3.2 10.5 11.1zm-46.5-14.5c6.1-1.6 9.2-6.1 9.2-13.3v-9.7c0-9.4-5.2-14.1-15.3-14.1h-13.7a1.81 1.81 0 0 0-2 1.9v63a1.81 1.81 0 0 0 2 1.9h1.2a1.74 1.74 0 0 0 1.9-1.9v-26.9h11.6l10.4 27.2a2.32 2.32 0 0 0 2.3 1.5h1.5c1.4 0 2-1 1.5-2.3zm-6.4-3.9H355v-28.5h10.2c7.5 0 10.5 3.1 10.5 11.1v6.4c0 7.84-3 11.04-10.5 11.04zm85.9-33.1h-13.7a1.62 1.62 0 0 0-2 1.8v63a1.81 1.81 0 0 0 2 1.9h1.2a1.74 1.74 0 0 0 1.9-1.9v-26.1h10.6c10.1 0 15.3-4.8 15.3-14.1v-10.5c0-9.4-5.2-14.1-15.3-14.1zm10.2 22.8c0 7.9-3 11.1-10.5 11.1h-10.2v-29.2h10.2c7.5-.1 10.5 3.1 10.5 11zM259.5 308l-2.3-6.8-2.3 6.8-7.1.1 5.7 4.3-2.1 6.8 5.8-4.1 5.8 4.1-2.1-6.8 5.7-4.3zm227.6-136.1a364.42 364.42 0 0 0-35.6-11.3c19.6-78 11.6-134.7-22.3-153.9C394.7-12.66 343.3 11 291 61.94q5.1 4.95 10.2 10.2c82.5-80 119.6-53.5 120.9-52.8 22.4 12.7 36 55.8 15.5 137.8a587.83 587.83 0 0 0-84.6-13C281.1 43.64 212.4 2 170.8 2 140 2 127 23 123.2 29.74c-18.1 32-13.3 84.2.1 133.8-70.5 20.3-120.7 54.1-120.3 95 .5 59.6 103.2 87.8 122.1 92.8-20.5 81.9-10.1 135.6 22.3 153.9 28 15.8 75.1 6 138.2-55.2q-5.1-4.95-10.2-10.2c-82.5 80-119.7 53.5-120.9 52.8-22.3-12.6-36-55.6-15.5-137.9 12.4 2.9 41.8 9.5 84.6 13 71.9 100.4 140.6 142 182.1 142 30.8 0 43.8-21 47.6-27.7 18-31.9 13.3-84.1-.1-133.8 152.3-43.8 156.2-130.2 33.9-176.3zM135.9 36.84c2.9-5.1 11.9-20.3 34.9-20.3 36.8 0 98.8 39.6 163.3 126.2a714 714 0 0 0-93.9.9 547.76 547.76 0 0 1 42.2-52.4Q277.3 86 272.2 81a598.25 598.25 0 0 0-50.7 64.2 569.69 569.69 0 0 0-84.4 14.6c-.2-1.4-24.3-82.2-1.2-123zm304.8 438.3c-2.9 5.1-11.8 20.3-34.9 20.3-36.7 0-98.7-39.4-163.3-126.2a695.38 695.38 0 0 0 93.9-.9 547.76 547.76 0 0 1-42.2 52.4q5.1 5.25 10.2 10.2a588.47 588.47 0 0 0 50.7-64.2c47.3-4.7 80.3-13.5 84.4-14.6 22.7 84.4 4.5 117 1.2 123zm9.1-138.6c-3.6-11.9-7.7-24.1-12.4-36.4a12.67 12.67 0 0 1-10.7-5.7l-.1.1a19.61 19.61 0 0 1-5.4 3.6c5.7 14.3 10.6 28.4 14.7 42.2a535.3 535.3 0 0 1-72 13c3.5-5.3 17.2-26.2 32.2-54.2a24.6 24.6 0 0 1-6-3.2c-1.1 1.2-3.6 4.2-10.9 4.2-6.2 11.2-17.4 30.9-33.9 55.2a711.91 711.91 0 0 1-112.4 1c-7.9-11.2-21.5-31.1-36.8-57.8a21 21 0 0 1-3-1.5c-1.9 1.6-3.9 3.2-12.6 3.2 6.3 11.2 17.5 30.7 33.8 54.6a548.81 548.81 0 0 1-72.2-11.7q5.85-21 14.1-42.9c-3.2 0-5.4.2-8.4-1a17.58 17.58 0 0 1-6.9 1c-4.9 13.4-9.1 26.5-12.7 39.4C-31.7 297-12.1 216 126.7 175.64c3.6 11.9 7.7 24.1 12.4 36.4 10.4 0 12.9 3.4 14.4 5.3a12 12 0 0 1 2.3-2.2c-5.8-14.7-10.9-29.2-15.2-43.3 7-1.8 32.4-8.4 72-13-15.9 24.3-26.7 43.9-32.8 55.3a14.22 14.22 0 0 1 6.4 8 23.42 23.42 0 0 1 10.2-8.4c6.5-11.7 17.9-31.9 34.8-56.9a711.72 711.72 0 0 1 112.4-1c31.5 44.6 28.9 48.1 42.5 64.5a21.42 21.42 0 0 1 10.4-7.4c-6.4-11.4-17.6-31-34.3-55.5 40.4 4.1 65 10 72.2 11.7-4 14.4-8.9 29.2-14.6 44.2a20.74 20.74 0 0 1 6.8 4.3l.1.1a12.72 12.72 0 0 1 8.9-5.6c4.9-13.4 9.2-26.6 12.8-39.5a359.71 359.71 0 0 1 34.5 11c106.1 39.9 74 87.9 72.6 90.4-19.8 35.1-80.1 55.2-105.7 62.5zm-114.4-114h-1.2a1.74 1.74 0 0 0-1.9 1.9v49.8c0 7.9-2.6 11.1-10.1 11.1s-10.1-3.1-10.1-11.1v-49.8a1.69 1.69 0 0 0-1.9-1.9H309a1.81 1.81 0 0 0-2 1.9v51.5c0 9.6 5 14.1 15.1 14.1h.4c10.1 0 15.1-4.6 15.1-14.1v-51.5a2 2 0 0 0-2.2-1.9zM321.7 308l-2.3-6.8-2.3 6.8-7.1.1 5.7 4.3-2.1 6.8 5.8-4.1 5.8 4.1-2.1-6.8 5.7-4.3zm-31.1 7.4l-2.3-6.8-2.3 6.8-7.1.1 5.7 4.3-2.1 6.8 5.8-4.1 5.8 4.1-2.1-6.8 5.7-4.3zm5.1-30.8h-19.4v-26.7h16.1a1.89 1.89 0 0 0 2-2v-.8a1.89 1.89 0 0 0-2-2h-16.1v-25.8h19.1a1.89 1.89 0 0 0 2-2v-.8a1.77 1.77 0 0 0-2-1.9h-22.2a1.81 1.81 0 0 0-2 1.9v63a1.81 1.81 0 0 0 2 1.9h22.5a1.77 1.77 0 0 0 2-1.9v-.8a1.83 1.83 0 0 0-2-2.06zm-7.4-99.4L286 192l-7.1.1 5.7 4.3-2.1 6.8 5.8-4.1 5.8 4.1-2.1-6.8 5.7-4.3-7.1-.1z"],medium:[640,512,[62407,"medium-m"],"f23a","M180.5,74.262C80.813,74.262,0,155.633,0,256S80.819,437.738,180.5,437.738,361,356.373,361,256,280.191,74.262,180.5,74.262Zm288.25,10.646c-49.845,0-90.245,76.619-90.245,171.095s40.406,171.1,90.251,171.1,90.251-76.619,90.251-171.1H559C559,161.5,518.6,84.908,468.752,84.908Zm139.506,17.821c-17.526,0-31.735,68.628-31.735,153.274s14.2,153.274,31.735,153.274S640,340.631,640,256C640,171.351,625.785,102.729,608.258,102.729Z"],amilia:[448,512,[],"f36d","M240.1 32c-61.9 0-131.5 16.9-184.2 55.4-5.1 3.1-9.1 9.2-7.2 19.4 1.1 5.1 5.1 27.4 10.2 39.6 4.1 10.2 14.2 10.2 20.3 6.1 32.5-22.3 96.5-47.7 152.3-47.7 57.9 0 58.9 28.4 58.9 73.1v38.5C203 227.7 78.2 251 46.7 264.2 11.2 280.5 16.3 357.7 16.3 376s15.2 104 124.9 104c47.8 0 113.7-20.7 153.3-42.1v25.4c0 3 2.1 8.2 6.1 9.1 3.1 1 50.7 2 59.9 2s62.5.3 66.5-.7c4.1-1 5.1-6.1 5.1-9.1V168c-.1-80.3-57.9-136-192-136zm50.2 348c-21.4 13.2-48.7 24.4-79.1 24.4-52.8 0-58.9-33.5-59-44.7 0-12.2-3-42.7 18.3-52.9 24.3-13.2 75.1-29.4 119.8-33.5z"],mixcloud:[640,512,[],"f289","M212.98 346.566H179.789V195.114L185.973 173.47H175.262L137.127 346.566H76.1069L37.7323 173.47H27.276L33.1913 195.114V346.566H0V165H65.6506L102.248 338.096H110.747L147.329 165H212.98L212.98 346.566ZM544.459 283.589L458.434 345.655V307.534L531.329 255.776L458.434 204.017V165.896L544.459 228.231H553.721L640 165.896V204.017L566.866 255.776L640 307.549V345.655L553.721 283.589H544.459ZM430.157 272.311H248.113V239.255H430.157V272.311Z"],flipboard:[448,512,[],"f44d","M0 32v448h448V32H0zm358.4 179.2h-89.6v89.6h-89.6v89.6H89.6V121.6h268.8v89.6z"],viacoin:[384,512,[],"f237","M384 32h-64l-80.7 192h-94.5L64 32H0l48 112H0v48h68.5l13.8 32H0v48h102.8L192 480l89.2-208H384v-48h-82.3l13.8-32H384v-48h-48l48-112zM192 336l-27-64h54l-27 64z"],"critical-role":[448,512,[],"f6c9","M225.82 0c.26.15 216.57 124.51 217.12 124.72 3 1.18 3.7 3.46 3.7 6.56q-.11 125.17 0 250.36a5.88 5.88 0 0 1-3.38 5.78c-21.37 12-207.86 118.29-218.93 124.58h-3C142 466.34 3.08 386.56 2.93 386.48a3.29 3.29 0 0 1-1.88-3.24c0-.87 0-225.94-.05-253.1a5 5 0 0 1 2.93-4.93C27.19 112.11 213.2 6 224.07 0zM215.4 20.42l-.22-.16Q118.06 75.55 21 130.87c0 .12.08.23.13.35l30.86 11.64c-7.71 6-8.32 6-10.65 5.13-.1 0-24.17-9.28-26.8-10v230.43c.88-1.41 64.07-110.91 64.13-111 1.62-2.82 3-1.92 9.12-1.52 1.4.09 1.48.22.78 1.42-41.19 71.33-36.4 63-67.48 116.94-.81 1.4-.61 1.13 1.25 1.13h186.5c1.44 0 1.69-.23 1.7-1.64v-8.88c0-1.34 2.36-.81-18.37-1-7.46-.07-14.14-3.22-21.38-12.7-7.38-9.66-14.62-19.43-21.85-29.21-2.28-3.08-3.45-2.38-16.76-2.38-1.75 0-1.78 0-1.76 1.82.29 26.21.15 25.27 1 32.66.52 4.37 2.16 4.2 9.69 4.81 3.14.26 3.88 4.08.52 4.92-1.57.39-31.6.51-33.67-.1a2.42 2.42 0 0 1 .3-4.73c3.29-.76 6.16.81 6.66-4.44 1.3-13.66 1.17-9 1.1-79.42 0-10.82-.35-12.58-5.36-13.55-1.22-.24-3.54-.16-4.69-.55-2.88-1-2-4.84 1.77-4.85 33.67 0 46.08-1.07 56.06 4.86 7.74 4.61 12 11.48 12.51 20.4.88 14.59-6.51 22.35-15 32.59a1.46 1.46 0 0 0 0 2.22c2.6 3.25 5 6.63 7.71 9.83 27.56 33.23 24.11 30.54 41.28 33.06.89.13 1-.42 1-1.15v-11c0-1 .32-1.43 1.41-1.26a72.37 72.37 0 0 0 23.58-.3c1.08-.15 1.5.2 1.48 1.33 0 .11.88 26.69.87 26.8-.05 1.52.67 1.62 1.89 1.62h186.71Q386.51 304.6 346 234.33c2.26-.66-.4 0 6.69-1.39 2-.39 2.05-.41 3.11 1.44 7.31 12.64 77.31 134 77.37 134.06V138c-1.72.5-103.3 38.72-105.76 39.68-1.08.42-1.55.2-1.91-.88-.63-1.9-1.34-3.76-2.09-5.62-.32-.79-.09-1.13.65-1.39.1 0 95.53-35.85 103-38.77-65.42-37.57-130.56-75-196-112.6l86.82 150.39-.28.33c-9.57-.9-10.46-1.6-11.8-3.94-1-1.69-73.5-127.71-82-142.16-9.1 14.67-83.56 146.21-85.37 146.32-2.93.17-5.88.08-9.25.08q43.25-74.74 86.18-149zm51.93 129.92a37.68 37.68 0 0 0 5.54-.85c1.69-.3 2.53.2 2.6 1.92 0 .11.07 19.06-.86 20.45s-1.88 1.22-2.6-.19c-5-9.69 6.22-9.66-39.12-12-.7 0-1 .23-1 .93 0 .13 3.72 122 3.73 122.11 0 .89.52 1.2 1.21 1.51a83.92 83.92 0 0 1 8.7 4.05c7.31 4.33 11.38 10.84 12.41 19.31 1.44 11.8-2.77 35.77-32.21 37.14-2.75.13-28.26 1.08-34.14-23.25-4.66-19.26 8.26-32.7 19.89-36.4a2.45 2.45 0 0 0 2-2.66c.1-5.63 3-107.1 3.71-121.35.05-1.08-.62-1.16-1.35-1.15-32.35.52-36.75-.34-40.22 8.52-2.42 6.18-4.14 1.32-3.95.23q1.59-9 3.31-18c.4-2.11 1.43-2.61 3.43-1.86 5.59 2.11 6.72 1.7 37.25 1.92 1.73 0 1.78-.08 1.82-1.85.68-27.49.58-22.59 1-29.55a2.69 2.69 0 0 0-1.63-2.8c-5.6-2.91-8.75-7.55-8.9-13.87-.35-14.81 17.72-21.67 27.38-11.51 6.84 7.19 5.8 18.91-2.45 24.15a4.35 4.35 0 0 0-2.22 4.34c0 .59-.11-4.31 1 30.05 0 .9.43 1.12 1.24 1.11.1 0 23-.09 34.47-.37zM68.27 141.7c19.84-4.51 32.68-.56 52.49 1.69 2.76.31 3.74 1.22 3.62 4-.21 5-1.16 22.33-1.24 23.15a2.65 2.65 0 0 1-1.63 2.34c-4.06 1.7-3.61-4.45-4-7.29-3.13-22.43-73.87-32.7-74.63 25.4-.31 23.92 17 53.63 54.08 50.88 27.24-2 19-20.19 24.84-20.47a2.72 2.72 0 0 1 3 3.36c-1.83 10.85-3.42 18.95-3.45 19.15-1.54 9.17-86.7 22.09-93.35-42.06-2.71-25.85 10.44-53.37 40.27-60.15zm80 87.67h-19.49a2.57 2.57 0 0 1-2.66-1.79c2.38-3.75 5.89.92 5.86-6.14-.08-25.75.21-38 .23-40.1 0-3.42-.53-4.65-3.32-4.94-7-.72-3.11-3.37-1.11-3.38 11.84-.1 22.62-.18 30.05.72 8.77 1.07 16.71 12.63 7.93 22.62-2 2.25-4 4.42-6.14 6.73.95 1.15 6.9 8.82 17.28 19.68 2.66 2.78 6.15 3.51 9.88 3.13a2.21 2.21 0 0 0 2.23-2.12c.3-3.42.26 4.73.45-40.58 0-5.65-.34-6.58-3.23-6.83-3.95-.35-4-2.26-.69-3.37l19.09-.09c.32 0 4.49.53 1 3.38 0 .05-.16 0-.24 0-3.61.26-3.94 1-4 4.62-.27 43.93.07 40.23.41 42.82.11.84.27 2.23 5.1 2.14 2.49 0 3.86 3.37 0 3.4-10.37.08-20.74 0-31.11.07-10.67 0-13.47-6.2-24.21-20.82-1.6-2.18-8.31-2.36-8.2-.37.88 16.47 0 17.78 4 17.67 4.75-.1 4.73 3.57.83 3.55zm275-10.15c-1.21 7.13.17 10.38-5.3 10.34-61.55-.42-47.82-.22-50.72-.31a18.4 18.4 0 0 1-3.63-.73c-2.53-.6 1.48-1.23-.38-5.6-1.43-3.37-2.78-6.78-4.11-10.19a1.94 1.94 0 0 0-2-1.44 138 138 0 0 0-14.58.07 2.23 2.23 0 0 0-1.62 1.06c-1.58 3.62-3.07 7.29-4.51 11-1.27 3.23 7.86 1.32 12.19 2.16 3 .57 4.53 3.72.66 3.73H322.9c-2.92 0-3.09-3.15-.74-3.21a6.3 6.3 0 0 0 5.92-3.47c1.5-3 2.8-6 4.11-9.09 18.18-42.14 17.06-40.17 18.42-41.61a1.83 1.83 0 0 1 3 0c2.93 3.34 18.4 44.71 23.62 51.92 2 2.7 5.74 2 6.36 2 3.61.13 4-1.11 4.13-4.29.09-1.87.08 1.17.07-41.24 0-4.46-2.36-3.74-5.55-4.27-.26 0-2.56-.63-.08-3.06.21-.2-.89-.24 21.7-.15 2.32 0 5.32 2.75-1.21 3.45a2.56 2.56 0 0 0-2.66 2.83c-.07 1.63-.19 38.89.29 41.21a3.06 3.06 0 0 0 3.23 2.43c13.25.43 14.92.44 16-3.41 1.67-5.78 4.13-2.52 3.73-.19zm-104.72 64.37c-4.24 0-4.42-3.39-.61-3.41 35.91-.16 28.11.38 37.19-.65 1.68-.19 2.38.24 2.25 1.89-.26 3.39-.64 6.78-1 10.16-.25 2.16-3.2 2.61-3.4-.15-.38-5.31-2.15-4.45-15.63-5.08-1.58-.07-1.64 0-1.64 1.52V304c0 1.65 0 1.6 1.62 1.47 3.12-.25 10.31.34 15.69-1.52.47-.16 3.3-1.79 3.07 1.76 0 .21-.76 10.35-1.18 11.39-.53 1.29-1.88 1.51-2.58.32-1.17-2 0-5.08-3.71-5.3-15.42-.9-12.91-2.55-12.91 6 0 12.25-.76 16.11 3.89 16.24 16.64.48 14.4 0 16.43-5.71.84-2.37 3.5-1.77 3.18.58-.44 3.21-.85 6.43-1.23 9.64 0 .36-.16 2.4-4.66 2.39-37.16-.08-34.54-.19-35.21-.31-2.72-.51-2.2-3 .22-3.45 1.1-.19 4 .54 4.16-2.56 2.44-56.22-.07-51.34-3.91-51.33zm-.41-109.52c2.46.61 3.13 1.76 2.95 4.65-.33 5.3-.34 9-.55 9.69-.66 2.23-3.15 2.12-3.34-.27-.38-4.81-3.05-7.82-7.57-9.15-26.28-7.73-32.81 15.46-27.17 30.22 5.88 15.41 22 15.92 28.86 13.78 5.92-1.85 5.88-6.5 6.91-7.58 1.23-1.3 2.25-1.84 3.12 1.1 0 .1.57 11.89-6 12.75-1.6.21-19.38 3.69-32.68-3.39-21-11.19-16.74-35.47-6.88-45.33 14-14.06 39.91-7.06 42.32-6.47zM289.8 280.14c3.28 0 3.66 3 .16 3.43-2.61.32-5-.42-5 5.46 0 2-.19 29.05.4 41.45.11 2.29 1.15 3.52 3.44 3.65 22 1.21 14.95-1.65 18.79-6.34 1.83-2.24 2.76.84 2.76 1.08.35 13.62-4 12.39-5.19 12.4l-38.16-.19c-1.93-.23-2.06-3-.42-3.38 2-.48 4.94.4 5.13-2.8 1-15.87.57-44.65.34-47.81-.27-3.77-2.8-3.27-5.68-3.71-2.47-.38-2-3.22.34-3.22 1.45-.02 17.97-.03 23.09-.02zm-31.63-57.79c.07 4.08 2.86 3.46 6 3.58 2.61.1 2.53 3.41-.07 3.43-6.48 0-13.7 0-21.61-.06-3.84 0-3.38-3.35 0-3.37 4.49 0 3.24 1.61 3.41-45.54 0-5.08-3.27-3.54-4.72-4.23-2.58-1.23-1.36-3.09.41-3.15 1.29 0 20.19-.41 21.17.21s1.87 1.65-.42 2.86c-1 .52-3.86-.28-4.15 2.47 0 .21-.82 1.63-.07 43.8zm-36.91 274.27a2.93 2.93 0 0 0 3.26 0c17-9.79 182-103.57 197.42-112.51-.14-.43 11.26-.18-181.52-.27-1.22 0-1.57.37-1.53 1.56 0 .1 1.25 44.51 1.22 50.38a28.33 28.33 0 0 1-1.36 7.71c-.55 1.83.38-.5-13.5 32.23-.73 1.72-1 2.21-2-.08-4.19-10.34-8.28-20.72-12.57-31a23.6 23.6 0 0 1-2-10.79c.16-2.46.8-16.12 1.51-48 0-1.95 0-2-2-2h-183c2.58 1.63 178.32 102.57 196 112.76zm-90.9-188.75c0 2.4.36 2.79 2.76 3 11.54 1.17 21 3.74 25.64-7.32 6-14.46 2.66-34.41-12.48-38.84-2-.59-16-2.76-15.94 1.51.05 8.04.01 11.61.02 41.65zm105.75-15.05c0 2.13 1.07 38.68 1.09 39.13.34 9.94-25.58 5.77-25.23-2.59.08-2 1.37-37.42 1.1-39.43-14.1 7.44-14.42 40.21 6.44 48.8a17.9 17.9 0 0 0 22.39-7.07c4.91-7.76 6.84-29.47-5.43-39a2.53 2.53 0 0 1-.36.12zm-12.28-198c-9.83 0-9.73 14.75-.07 14.87s10.1-14.88.07-14.91zm-80.15 103.83c0 1.8.41 2.4 2.17 2.58 13.62 1.39 12.51-11 12.16-13.36-1.69-11.22-14.38-10.2-14.35-7.81.05 4.5-.03 13.68.02 18.59zm212.32 6.4l-6.1-15.84c-2.16 5.48-4.16 10.57-6.23 15.84z"],sitrox:[448,512,[],"e44a","M212.439 0.00846128V0H448V128H64C64 57.6008 141.755 0.475338 212.439 0.00846128ZM237.256 192V192.007C307.135 192.475 384 249.6 384 320H210.809V319.995C140.915 319.563 64 262.424 64 192H237.256ZM235.565 511.993C306.251 511.521 384 454.399 384 384H0V512H235.565V511.993Z"],discourse:[448,512,[],"f393","M225.9 32C103.3 32 0 130.5 0 252.1 0 256 .1 480 .1 480l225.8-.2c122.7 0 222.1-102.3 222.1-223.9C448 134.3 348.6 32 225.9 32zM224 384c-19.4 0-37.9-4.3-54.4-12.1L88.5 392l22.9-75c-9.8-18.1-15.4-38.9-15.4-61 0-70.7 57.3-128 128-128s128 57.3 128 128-57.3 128-128 128z"],joomla:[448,512,[],"f1aa","M.6 92.1C.6 58.8 27.4 32 60.4 32c30 0 54.5 21.9 59.2 50.2 32.6-7.6 67.1.6 96.5 30l-44.3 44.3c-20.5-20.5-42.6-16.3-55.4-3.5-14.3 14.3-14.3 37.9 0 52.2l99.5 99.5-44 44.3c-87.7-87.2-49.7-49.7-99.8-99.7-26.8-26.5-35-64.8-24.8-98.9C20.4 144.6.6 120.7.6 92.1zm129.5 116.4l44.3 44.3c10-10 89.7-89.7 99.7-99.8 14.3-14.3 37.6-14.3 51.9 0 12.8 12.8 17 35-3.5 55.4l44 44.3c31.2-31.2 38.5-67.6 28.9-101.2 29.2-4.1 51.9-29.2 51.9-59.5 0-33.2-26.8-60.1-59.8-60.1-30.3 0-55.4 22.5-59.5 51.6-33.8-9.9-71.7-1.5-98.3 25.1-18.3 19.1-71.1 71.5-99.6 99.9zm266.3 152.2c8.2-32.7-.9-68.5-26.3-93.9-11.8-12.2 5 4.7-99.5-99.7l-44.3 44.3 99.7 99.7c14.3 14.3 14.3 37.6 0 51.9-12.8 12.8-35 17-55.4-3.5l-44 44.3c27.6 30.2 68 38.8 102.7 28 5.5 27.4 29.7 48.1 58.9 48.1 33 0 59.8-26.8 59.8-60.1 0-30.2-22.5-55-51.6-59.1zm-84.3-53.1l-44-44.3c-87 86.4-50.4 50.4-99.7 99.8-14.3 14.3-37.6 14.3-51.9 0-13.1-13.4-16.9-35.3 3.2-55.4l-44-44.3c-30.2 30.2-38 65.2-29.5 98.3-26.7 6-46.2 29.9-46.2 58.2C0 453.2 26.8 480 59.8 480c28.6 0 52.5-19.8 58.6-46.7 32.7 8.2 68.5-.6 94.2-26 32.1-32 12.2-12.4 99.5-99.7z"],mastodon:[448,512,[],"f4f6","M433 179.11c0-97.2-63.71-125.7-63.71-125.7-62.52-28.7-228.56-28.4-290.48 0 0 0-63.72 28.5-63.72 125.7 0 115.7-6.6 259.4 105.63 289.1 40.51 10.7 75.32 13 103.33 11.4 50.81-2.8 79.32-18.1 79.32-18.1l-1.7-36.9s-36.31 11.4-77.12 10.1c-40.41-1.4-83-4.4-89.63-54a102.54 102.54 0 0 1-.9-13.9c85.63 20.9 158.65 9.1 178.75 6.7 56.12-6.7 105-41.3 111.23-72.9 9.8-49.8 9-121.5 9-121.5zm-75.12 125.2h-46.63v-114.2c0-49.7-64-51.6-64 6.9v62.5h-46.33V197c0-58.5-64-56.6-64-6.9v114.2H90.19c0-122.1-5.2-147.9 18.41-175 25.9-28.9 79.82-30.8 103.83 6.1l11.6 19.5 11.6-19.5c24.11-37.1 78.12-34.8 103.83-6.1 23.71 27.3 18.4 53 18.4 175z"],airbnb:[448,512,[],"f834","M224 373.12c-25.24-31.67-40.08-59.43-45-83.18-22.55-88 112.61-88 90.06 0-5.45 24.25-20.29 52-45 83.18zm138.15 73.23c-42.06 18.31-83.67-10.88-119.3-50.47 103.9-130.07 46.11-200-18.85-200-54.92 0-85.16 46.51-73.28 100.5 6.93 29.19 25.23 62.39 54.43 99.5-32.53 36.05-60.55 52.69-85.15 54.92-50 7.43-89.11-41.06-71.3-91.09 15.1-39.16 111.72-231.18 115.87-241.56 15.75-30.07 25.56-57.4 59.38-57.4 32.34 0 43.4 25.94 60.37 59.87 36 70.62 89.35 177.48 114.84 239.09 13.17 33.07-1.37 71.29-37.01 86.64zm47-136.12C280.27 35.93 273.13 32 224 32c-45.52 0-64.87 31.67-84.66 72.79C33.18 317.1 22.89 347.19 22 349.81-3.22 419.14 48.74 480 111.63 480c21.71 0 60.61-6.06 112.37-62.4 58.68 63.78 101.26 62.4 112.37 62.4 62.89.05 114.85-60.86 89.61-130.19.02-3.89-16.82-38.9-16.82-39.58z"],"wolf-pack-battalion":[512,512,[],"f514","M267.73 471.53l10.56 15.84 5.28-12.32 5.28 7V512c21.06-7.92 21.11-66.86 25.51-97.21 4.62-31.89-.88-92.81 81.37-149.11-8.88-23.61-12-49.43-2.64-80.05C421 189 447 196.21 456.43 239.73l-30.35 8.36c11.15 23 17 46.76 13.2 72.14L412 313.18l-6.16 33.43-18.47-7-8.8 33.39-19.35-7 26.39 21.11 8.8-28.15L419 364.2l7-35.63 26.39 14.52c.25-20 7-58.06-8.8-84.45l26.39 5.28c4-22.07-2.38-39.21-7.92-56.74l22.43 9.68c-.44-25.07-29.94-56.79-61.58-58.5-20.22-1.09-56.74-25.17-54.1-51.9 2-19.87 17.45-42.62 43.11-49.7-44 36.51-9.68 67.3 5.28 73.46 4.4-11.44 17.54-69.08 0-130.2-40.39 22.87-89.65 65.1-93.2 147.79l-58 38.71-3.52 93.25L369.78 220l7 7-17.59 3.52-44 38.71-15.84-5.28-28.1 49.25-3.52 119.64 21.11 15.84-32.55 15.84-32.55-15.84 21.11-15.84-3.52-119.64-28.15-49.26-15.84 5.28-44-38.71-17.58-3.51 7-7 107.33 59.82-3.52-93.25-58.06-38.71C185 65.1 135.77 22.87 95.3 0c-17.54 61.12-4.4 118.76 0 130.2 15-6.16 49.26-36.95 5.28-73.46 25.66 7.08 41.15 29.83 43.11 49.7 2.63 26.74-33.88 50.81-54.1 51.9-31.65 1.72-61.15 33.44-61.59 58.51l22.43-9.68c-5.54 17.53-11.91 34.67-7.92 56.74l26.39-5.28c-15.76 26.39-9.05 64.43-8.8 84.45l26.39-14.52 7 35.63 24.63-5.28 8.8 28.15L153.35 366 134 373l-8.8-33.43-18.47 7-6.16-33.43-27.27 7c-3.82-25.38 2-49.1 13.2-72.14l-30.35-8.36c9.4-43.52 35.47-50.77 63.34-54.1 9.36 30.62 6.24 56.45-2.64 80.05 82.25 56.3 76.75 117.23 81.37 149.11 4.4 30.35 4.45 89.29 25.51 97.21v-29.83l5.28-7 5.28 12.32 10.56-15.84 11.44 21.11 11.43-21.1zm79.17-95L331.06 366c7.47-4.36 13.76-8.42 19.35-12.32-.6 7.22-.27 13.84-3.51 22.84zm28.15-49.26c-.4 10.94-.9 21.66-1.76 31.67-7.85-1.86-15.57-3.8-21.11-7 8.24-7.94 15.55-16.32 22.87-24.68zm24.63 5.28c0-13.43-2.05-24.21-5.28-33.43a235 235 0 0 1-18.47 27.27zm3.52-80.94c19.44 12.81 27.8 33.66 29.91 56.3-12.32-4.53-24.63-9.31-36.95-10.56 5.06-12 6.65-28.14 7-45.74zm-1.76-45.74c.81 14.3 1.84 28.82 1.76 42.23 19.22-8.11 29.78-9.72 44-14.08-10.61-18.96-27.2-25.53-45.76-28.16zM165.68 376.52L181.52 366c-7.47-4.36-13.76-8.42-19.35-12.32.6 7.26.27 13.88 3.51 22.88zm-28.15-49.26c.4 10.94.9 21.66 1.76 31.67 7.85-1.86 15.57-3.8 21.11-7-8.24-7.93-15.55-16.31-22.87-24.67zm-24.64 5.28c0-13.43 2-24.21 5.28-33.43a235 235 0 0 0 18.47 27.27zm-3.52-80.94c-19.44 12.81-27.8 33.66-29.91 56.3 12.32-4.53 24.63-9.31 37-10.56-5-12-6.65-28.14-7-45.74zm1.76-45.74c-.81 14.3-1.84 28.82-1.76 42.23-19.22-8.11-29.78-9.72-44-14.08 10.63-18.95 27.23-25.52 45.76-28.15z"],"buy-n-large":[576,512,[],"f8a6","M288 32C133.27 32 7.79 132.32 7.79 256S133.27 480 288 480s280.21-100.32 280.21-224S442.73 32 288 32zm-85.39 357.19L64.1 390.55l77.25-290.74h133.44c63.15 0 84.93 28.65 78 72.84a60.24 60.24 0 0 1-1.5 6.85 77.39 77.39 0 0 0-17.21-1.93c-42.35 0-76.69 33.88-76.69 75.65 0 37.14 27.14 68 62.93 74.45-18.24 37.16-56.16 60.92-117.71 61.52zM358 207.11h32l-22.16 90.31h-35.41l-11.19-35.63-7.83 35.63h-37.83l26.63-90.31h31.34l15 36.75zm145.86 182.08H306.79L322.63 328a78.8 78.8 0 0 0 11.47.83c42.34 0 76.69-33.87 76.69-75.65 0-32.65-21-60.46-50.38-71.06l21.33-82.35h92.5l-53.05 205.36h103.87zM211.7 269.39H187l-13.8 56.47h24.7c16.14 0 32.11-3.18 37.94-26.65 5.56-22.31-7.99-29.82-24.14-29.82zM233 170h-21.34L200 217.71h21.37c18 0 35.38-14.64 39.21-30.14C265.23 168.71 251.07 170 233 170z"],gulp:[256,512,[],"f3ae","M209.8 391.1l-14.1 24.6-4.6 80.2c0 8.9-28.3 16.1-63.1 16.1s-63.1-7.2-63.1-16.1l-5.8-79.4-14.9-25.4c41.2 17.3 126 16.7 165.6 0zm-196-253.3l13.6 125.5c5.9-20 20.8-47 40-55.2 6.3-2.7 12.7-2.7 18.7.9 5.2 3 9.6 9.3 10.1 11.8 1.2 6.5-2 9.1-4.5 9.1-3 0-5.3-4.6-6.8-7.3-4.1-7.3-10.3-7.6-16.9-2.8-6.9 5-12.9 13.4-17.1 20.7-5.1 8.8-9.4 18.5-12 28.2-1.5 5.6-2.9 14.6-.6 19.9 1 2.2 2.5 3.6 4.9 3.6 5 0 12.3-6.6 15.8-10.1 4.5-4.5 10.3-11.5 12.5-16l5.2-15.5c2.6-6.8 9.9-5.6 9.9 0 0 10.2-3.7 13.6-10 34.7-5.8 19.5-7.6 25.8-7.6 25.8-.7 2.8-3.4 7.5-6.3 7.5-1.2 0-2.1-.4-2.6-1.2-1-1.4-.9-5.3-.8-6.3.2-3.2 6.3-22.2 7.3-25.2-2 2.2-4.1 4.4-6.4 6.6-5.4 5.1-14.1 11.8-21.5 11.8-3.4 0-5.6-.9-7.7-2.4l7.6 79.6c2 5 39.2 17.1 88.2 17.1 49.1 0 86.3-12.2 88.2-17.1l10.9-94.6c-5.7 5.2-12.3 11.6-19.6 14.8-5.4 2.3-17.4 3.8-17.4-5.7 0-5.2 9.1-14.8 14.4-21.5 1.4-1.7 4.7-5.9 4.7-8.1 0-2.9-6-2.2-11.7 2.5-3.2 2.7-6.2 6.3-8.7 9.7-4.3 6-6.6 11.2-8.5 15.5-6.2 14.2-4.1 8.6-9.1 22-5 13.3-4.2 11.8-5.2 14-.9 1.9-2.2 3.5-4 4.5-1.9 1-4.5.9-6.1-.3-.9-.6-1.3-1.9-1.3-3.7 0-.9.1-1.8.3-2.7 1.5-6.1 7.8-18.1 15-34.3 1.6-3.7 1-2.6.8-2.3-6.2 6-10.9 8.9-14.4 10.5-5.8 2.6-13 2.6-14.5-4.1-.1-.4-.1-.8-.2-1.2-11.8 9.2-24.3 11.7-20-8.1-4.6 8.2-12.6 14.9-22.4 14.9-4.1 0-7.1-1.4-8.6-5.1-2.3-5.5 1.3-14.9 4.6-23.8 1.7-4.5 4-9.9 7.1-16.2 1.6-3.4 4.2-5.4 7.6-4.5.6.2 1.1.4 1.6.7 2.6 1.8 1.6 4.5.3 7.2-3.8 7.5-7.1 13-9.3 20.8-.9 3.3-2 9 1.5 9 2.4 0 4.7-.8 6.9-2.4 4.6-3.4 8.3-8.5 11.1-13.5 2-3.6 4.4-8.3 5.6-12.3.5-1.7 1.1-3.3 1.8-4.8 1.1-2.5 2.6-5.1 5.2-5.1 1.3 0 2.4.5 3.2 1.5 1.7 2.2 1.3 4.5.4 6.9-2 5.6-4.7 10.6-6.9 16.7-1.3 3.5-2.7 8-2.7 11.7 0 3.4 3.7 2.6 6.8 1.2 2.4-1.1 4.8-2.8 6.8-4.5 1.2-4.9.9-3.8 26.4-68.2 1.3-3.3 3.7-4.7 6.1-4.7 1.2 0 2.2.4 3.2 1.1 1.7 1.3 1.7 4.1 1 6.2-.7 1.9-.6 1.3-4.5 10.5-5.2 12.1-8.6 20.8-13.2 31.9-1.9 4.6-7.7 18.9-8.7 22.3-.6 2.2-1.3 5.8 1 5.8 5.4 0 19.3-13.1 23.1-17 .2-.3.5-.4.9-.6.6-1.9 1.2-3.7 1.7-5.5 1.4-3.8 2.7-8.2 5.3-11.3.8-1 1.7-1.6 2.7-1.6 2.8 0 4.2 1.2 4.2 4 0 1.1-.7 5.1-1.1 6.2 1.4-1.5 2.9-3 4.5-4.5 15-13.9 25.7-6.8 25.7.2 0 7.4-8.9 17.7-13.8 23.4-1.6 1.9-4.9 5.4-5 6.4 0 1.3.9 1.8 2.2 1.8 2 0 6.4-3.5 8-4.7 5-3.9 11.8-9.9 16.6-14.1l14.8-136.8c-30.5 17.1-197.6 17.2-228.3.2zm229.7-8.5c0 21-231.2 21-231.2 0 0-8.8 51.8-15.9 115.6-15.9 9 0 17.8.1 26.3.4l12.6-48.7L228.1.6c1.4-1.4 5.8-.2 9.9 3.5s6.6 7.9 5.3 9.3l-.1.1L185.9 74l-10 40.7c39.9 2.6 67.6 8.1 67.6 14.6zm-69.4 4.6c0-.8-.9-1.5-2.5-2.1l-.2.8c0 1.3-5 2.4-11.1 2.4s-11.1-1.1-11.1-2.4c0-.1 0-.2.1-.3l.2-.7c-1.8.6-3 1.4-3 2.3 0 2.1 6.2 3.7 13.7 3.7 7.7.1 13.9-1.6 13.9-3.7z"],"creative-commons-sampling-plus":[496,512,[],"f4f1","M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zm107 205.6c-4.7 0-9 2.8-10.7 7.2l-4 9.5-11-92.8c-1.7-13.9-22-13.4-23.1.4l-4.3 51.4-5.2-68.8c-1.1-14.3-22.1-14.2-23.2 0l-3.5 44.9-5.9-94.3c-.9-14.5-22.3-14.4-23.2 0l-5.1 83.7-4.3-66.3c-.9-14.4-22.2-14.4-23.2 0l-5.3 80.2-4.1-57c-1.1-14.3-22-14.3-23.2-.2l-7.7 89.8-1.8-12.2c-1.7-11.4-17.1-13.6-22-3.3l-13.2 27.7H87.5v23.2h51.3c4.4 0 8.4-2.5 10.4-6.4l10.7 73.1c2 13.5 21.9 13 23.1-.7l3.8-43.6 5.7 78.3c1.1 14.4 22.3 14.2 23.2-.1l4.6-70.4 4.8 73.3c.9 14.4 22.3 14.4 23.2-.1l4.9-80.5 4.5 71.8c.9 14.3 22.1 14.5 23.2.2l4.6-58.6 4.9 64.4c1.1 14.3 22 14.2 23.1.1l6.8-83 2.7 22.3c1.4 11.8 17.7 14.1 22.3 3.1l18-43.4h50.5V258l-58.4.3zm-78 5.2h-21.9v21.9c0 4.1-3.3 7.5-7.5 7.5-4.1 0-7.5-3.3-7.5-7.5v-21.9h-21.9c-4.1 0-7.5-3.3-7.5-7.5 0-4.1 3.4-7.5 7.5-7.5h21.9v-21.9c0-4.1 3.4-7.5 7.5-7.5s7.5 3.3 7.5 7.5v21.9h21.9c4.1 0 7.5 3.3 7.5 7.5 0 4.1-3.4 7.5-7.5 7.5z"],strava:[384,512,[],"f428","M158.4 0L7 292h89.2l62.2-116.1L220.1 292h88.5zm150.2 292l-43.9 88.2-44.6-88.2h-67.6l112.2 220 111.5-220z"],ember:[640,512,[],"f423","M639.9 254.6c-1.1-10.7-10.7-6.8-10.7-6.8s-15.6 12.1-29.3 10.7c-13.7-1.3-9.4-32-9.4-32s3-28.1-5.1-30.4c-8.1-2.4-18 7.3-18 7.3s-12.4 13.7-18.3 31.2l-1.6.5s1.9-30.6-.3-37.6c-1.6-3.5-16.4-3.2-18.8 3s-14.2 49.2-15 67.2c0 0-23.1 19.6-43.3 22.8s-25-9.4-25-9.4 54.8-15.3 52.9-59.1-44.2-27.6-49-24c-4.6 3.5-29.4 18.4-36.6 59.7-.2 1.4-.7 7.5-.7 7.5s-21.2 14.2-33 18c0 0 33-55.6-7.3-80.9-11.4-6.8-21.3-.5-27.2 5.3 13.6-17.3 46.4-64.2 36.9-105.2-5.8-24.4-18-27.1-29.2-23.1-17 6.7-23.5 16.7-23.5 16.7s-22 32-27.1 79.5-12.6 105.1-12.6 105.1-10.5 10.2-20.2 10.7-5.4-28.7-5.4-28.7 7.5-44.6 7-52.1-1.1-11.6-9.9-14.2c-8.9-2.7-18.5 8.6-18.5 8.6s-25.5 38.7-27.7 44.6l-1.3 2.4-1.3-1.6s18-52.7.8-53.5-28.5 18.8-28.5 18.8-19.6 32.8-20.4 36.5l-1.3-1.6s8.1-38.2 6.4-47.6c-1.6-9.4-10.5-7.5-10.5-7.5s-11.3-1.3-14.2 5.9-13.7 55.3-15 70.7c0 0-28.2 20.2-46.8 20.4-18.5.3-16.7-11.8-16.7-11.8s68-23.3 49.4-69.2c-8.3-11.8-18-15.5-31.7-15.3-13.7.3-30.3 8.6-41.3 33.3-5.3 11.8-6.8 23-7.8 31.5 0 0-12.3 2.4-18.8-2.9s-10 0-10 0-11.2 14-.1 18.3 28.1 6.1 28.1 6.1c1.6 7.5 6.2 19.5 19.6 29.7 20.2 15.3 58.8-1.3 58.8-1.3l15.9-8.8s.5 14.6 12.1 16.7 16.4 1 36.5-47.9c11.8-25 12.6-23.6 12.6-23.6l1.3-.3s-9.1 46.8-5.6 59.7C187.7 319.4 203 318 203 318s8.3 2.4 15-21.2 19.6-49.9 19.6-49.9h1.6s-5.6 48.1 3 63.7 30.9 5.3 30.9 5.3 15.6-7.8 18-10.2c0 0 18.5 15.8 44.6 12.9 58.3-11.5 79.1-25.9 79.1-25.9s10 24.4 41.1 26.7c35.5 2.7 54.8-18.6 54.8-18.6s-.3 13.5 12.1 18.6 20.7-22.8 20.7-22.8l20.7-57.2h1.9s1.1 37.3 21.5 43.2 47-13.7 47-13.7 6.4-3.5 5.3-14.3zm-578 5.3c.8-32 21.8-45.9 29-39 7.3 7 4.6 22-9.1 31.4-13.7 9.5-19.9 7.6-19.9 7.6zm272.8-123.8s19.1-49.7 23.6-25.5-40 96.2-40 96.2c.5-16.2 16.4-70.7 16.4-70.7zm22.8 138.4c-12.6 33-43.3 19.6-43.3 19.6s-3.5-11.8 6.4-44.9 33.3-20.2 33.3-20.2 16.2 12.4 3.6 45.5zm84.6-14.6s-3-10.5 8.1-30.6c11-20.2 19.6-9.1 19.6-9.1s9.4 10.2-1.3 25.5-26.4 14.2-26.4 14.2z"],"canadian-maple-leaf":[512,512,[],"f785","M383.8 351.7c2.5-2.5 105.2-92.4 105.2-92.4l-17.5-7.5c-10-4.9-7.4-11.5-5-17.4 2.4-7.6 20.1-67.3 20.1-67.3s-47.7 10-57.7 12.5c-7.5 2.4-10-2.5-12.5-7.5s-15-32.4-15-32.4-52.6 59.9-55.1 62.3c-10 7.5-20.1 0-17.6-10 0-10 27.6-129.6 27.6-129.6s-30.1 17.4-40.1 22.4c-7.5 5-12.6 5-17.6-5C293.5 72.3 255.9 0 255.9 0s-37.5 72.3-42.5 79.8c-5 10-10 10-17.6 5-10-5-40.1-22.4-40.1-22.4S183.3 182 183.3 192c2.5 10-7.5 17.5-17.6 10-2.5-2.5-55.1-62.3-55.1-62.3S98.1 167 95.6 172s-5 9.9-12.5 7.5C73 177 25.4 167 25.4 167s17.6 59.7 20.1 67.3c2.4 6 5 12.5-5 17.4L23 259.3s102.6 89.9 105.2 92.4c5.1 5 10 7.5 5.1 22.5-5.1 15-10.1 35.1-10.1 35.1s95.2-20.1 105.3-22.6c8.7-.9 18.3 2.5 18.3 12.5S241 512 241 512h30s-5.8-102.7-5.8-112.8 9.5-13.4 18.4-12.5c10 2.5 105.2 22.6 105.2 22.6s-5-20.1-10-35.1 0-17.5 5-22.5z"],teamspeak:[576,512,[],"f4f9","M152.8 37.2c-32.2 38.1-56.1 82.6-69.9 130.5c0 .2-.1 .3-.1 .5C43.5 184.4 16 223 16 268c0 59.6 48.4 108 108 108s108-48.4 108-108c0-53.5-38.9-97.9-90-106.5c15.7-41.8 40.4-79.6 72.3-110.7c1.8-1.6 4-2.6 6.3-3.1c37.2-11.5 76.7-13.3 114.8-5.2C454.7 67.6 534 180.7 517.1 301.3c-8.4 62.6-38.6 112.7-87.7 151.4c-50.1 39.7-107.5 54.3-170.2 52.2l-24-1c12.4 2.8 25 4.9 37.6 6.3c40.7 4.2 81.4 2.1 120.1-12.5c94-35.5 149.3-102.3 162.9-202.5c4.8-52.6-5.8-105.4-30.8-152C454.6 11.3 290.8-38.4 159 32c-2.4 1.4-4.5 3.1-6.3 5.2zM309.4 433.9c-2.1 11.5-4.2 21.9-14.6 31.3c53.2-1 123.2-29.2 161.8-97.1c39.7-69.9 37.6-139.9-6.3-207.8C413.8 105 360.5 77.9 293.7 73.7c1.5 2.3 3.2 4.4 5.2 6.3l5.2 6.3c25.1 31.3 37.6 67.9 42.8 107.5c2.1 15.7-1 30.3-13.6 41.8c-4.2 3.1-5.2 6.3-4.2 10.4l7.3 17.7L365.7 318c5.2 11.5 4.2 19.8-6.3 28.2c-3.2 2.5-6.7 4.6-10.4 6.3l-18.8 8.4 3.1 13.6c3.1 6.3 1 12.5-3.1 17.7c-2.5 2.4-3.8 5.9-3.1 9.4c2.1 11.5-2.1 19.8-12.5 25.1c-2.1 1-4.2 5.2-5.2 7.3zm-133.6-3.1c16.7 11.5 34.5 20.9 53.2 26.1c24 5.2 41.8-6.3 44.9-30.3c1-8.4 5.2-14.6 12.5-17.7c7.3-4.2 8.4-7.3 2.1-13.6l-9.4-8.4 13.6-4.2c6.3-2.1 7.3-5.2 5.2-11.5c-1.4-3-2.4-6.2-3.1-9.4c-3.1-14.6-2.1-15.7 11.5-18.8c8.4-3.1 15.7-6.3 21.9-12.5c3.1-2.1 3.1-4.2 1-8.4l-16.7-30.3c-1-1.9-2.1-3.8-3.1-5.7c-6.4-11.7-13-23.6-15.7-37.1c-2.1-9.4-1-17.7 8.4-24c5.2-4.2 8.4-9.4 8.4-16.7c-.4-10.1-1.5-20.3-3.1-30.3c-6.3-37.6-23-68.9-51.2-95c-5.2-4.2-9.4-6.3-16.7-4.2L203.9 91.5c2 1.2 4 2.4 6 3.6l0 0c6.3 3.7 12.2 7.3 17 12.1c30.3 26.1 41.8 61.6 45.9 100.2c1 8.4 0 16.7-7.3 21.9c-8.4 5.2-10.4 12.5-7.3 20.9c4.9 13.2 10.4 26 16.7 38.6L291.6 318c-6.3 8.4-13.6 11.5-21.9 14.6c-12.5 3.1-14.6 7.3-10.4 20.9c.6 1.5 1.4 2.8 2.1 4.2c2.1 5.2 1 8.4-4.2 10.4l-12.5 3.1 5.2 4.2 4.2 4.2c4.2 5.2 4.2 8.4-2.1 10.4c-7.3 4.2-11.5 9.4-11.5 17.7c0 12.5-7.3 19.8-18.8 24c-3.8 1-7.6 1.5-11.5 1l-34.5-2.1z"],pushed:[432,512,[],"f3e1","M407 111.9l-98.5-9 14-33.4c10.4-23.5-10.8-40.4-28.7-37L22.5 76.9c-15.1 2.7-26 18.3-21.4 36.6l105.1 348.3c6.5 21.3 36.7 24.2 47.7 7l35.3-80.8 235.2-231.3c16.4-16.8 4.3-42.9-17.4-44.8zM297.6 53.6c5.1-.7 7.5 2.5 5.2 7.4L286 100.9 108.6 84.6l189-31zM22.7 107.9c-3.1-5.1 1-10 6.1-9.1l248.7 22.7-96.9 230.7L22.7 107.9zM136 456.4c-2.6 4-7.9 3.1-9.4-1.2L43.5 179.7l127.7 197.6c-7 15-35.2 79.1-35.2 79.1zm272.8-314.5L210.1 337.3l89.7-213.7 106.4 9.7c4 1.1 5.7 5.3 2.6 8.6z"],"wordpress-simple":[512,512,[],"f411","M256 8C119.3 8 8 119.2 8 256c0 136.7 111.3 248 248 248s248-111.3 248-248C504 119.2 392.7 8 256 8zM33 256c0-32.3 6.9-63 19.3-90.7l106.4 291.4C84.3 420.5 33 344.2 33 256zm223 223c-21.9 0-43-3.2-63-9.1l66.9-194.4 68.5 187.8c.5 1.1 1 2.1 1.6 3.1-23.1 8.1-48 12.6-74 12.6zm30.7-327.5c13.4-.7 25.5-2.1 25.5-2.1 12-1.4 10.6-19.1-1.4-18.4 0 0-36.1 2.8-59.4 2.8-21.9 0-58.7-2.8-58.7-2.8-12-.7-13.4 17.7-1.4 18.4 0 0 11.4 1.4 23.4 2.1l34.7 95.2L200.6 393l-81.2-241.5c13.4-.7 25.5-2.1 25.5-2.1 12-1.4 10.6-19.1-1.4-18.4 0 0-36.1 2.8-59.4 2.8-4.2 0-9.1-.1-14.4-.3C109.6 73 178.1 33 256 33c58 0 110.9 22.2 150.6 58.5-1-.1-1.9-.2-2.9-.2-21.9 0-37.4 19.1-37.4 39.6 0 18.4 10.6 33.9 21.9 52.3 8.5 14.8 18.4 33.9 18.4 61.5 0 19.1-7.3 41.2-17 72.1l-22.2 74.3-80.7-239.6zm81.4 297.2l68.1-196.9c12.7-31.8 17-57.2 17-79.9 0-8.2-.5-15.8-1.5-22.9 17.4 31.8 27.3 68.2 27.3 107 0 82.3-44.6 154.1-110.9 192.7z"],nutritionix:[400,512,[],"f3d6","M88 8.1S221.4-.1 209 112.5c0 0 19.1-74.9 103-40.6 0 0-17.7 74-88 56 0 0 14.6-54.6 66.1-56.6 0 0-39.9-10.3-82.1 48.8 0 0-19.8-94.5-93.6-99.7 0 0 75.2 19.4 77.6 107.5 0 .1-106.4 7-104-119.8zm312 315.6c0 48.5-9.7 95.3-32 132.3-42.2 30.9-105 48-168 48-62.9 0-125.8-17.1-168-48C9.7 419 0 372.2 0 323.7 0 275.3 17.7 229 40 192c42.2-30.9 97.1-48.6 160-48.6 63 0 117.8 17.6 160 48.6 22.3 37 40 83.3 40 131.7zM120 428c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zM192 428c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zM264 428c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zM336 428c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm24-39.6c-4.8-22.3-7.4-36.9-16-56-38.8-19.9-90.5-32-144-32S94.8 180.1 56 200c-8.8 19.5-11.2 33.9-16 56 42.2-7.9 98.7-14.8 160-14.8s117.8 6.9 160 14.8z"],wodu:[640,512,[],"e088","M178.414 339.706H141.1L112.166 223.475h-.478L83.228 339.706H45.2L0 168.946H37.548L64.574 285.177h.478L94.707 168.946h35.157l29.178 117.667h.479L187.5 168.946h36.831zM271.4 212.713c38.984 0 64.1 25.828 64.1 65.291 0 39.222-25.111 65.05-64.1 65.05-38.743 0-63.855-25.828-63.855-65.05C207.547 238.541 232.659 212.713 271.4 212.713zm0 104.753c23.2 0 30.133-19.852 30.133-39.462 0-19.852-6.934-39.7-30.133-39.7-27.7 0-29.894 19.85-29.894 39.7C241.508 297.614 248.443 317.466 271.4 317.466zM435.084 323.922h-.478c-7.893 13.392-21.765 19.132-37.548 19.132-37.31 0-55.485-32.045-55.485-66.246 0-33.243 18.415-64.095 54.767-64.095 14.589 0 28.938 6.218 36.831 18.416h.24V168.946h33.96v170.76H435.084zM405.428 238.3c-22.24 0-29.894 19.134-29.894 39.463 0 19.371 8.848 39.7 29.894 39.7 22.482 0 29.178-19.613 29.178-39.94C434.606 257.436 427.432 238.3 405.428 238.3zM592.96 339.706H560.673V322.487h-.718c-8.609 13.87-23.436 20.567-37.786 20.567-36.113 0-45.2-20.328-45.2-50.941V216.061h33.959V285.9c0 20.329 5.979 30.372 21.765 30.372 18.415 0 26.306-10.283 26.306-35.393V216.061H592.96zM602.453 302.876H640v36.83H602.453z"],"google-pay":[640,512,[],"e079","M105.72,215v41.25h57.1a49.66,49.66,0,0,1-21.14,32.6c-9.54,6.55-21.72,10.28-36,10.28-27.6,0-50.93-18.91-59.3-44.22a65.61,65.61,0,0,1,0-41l0,0c8.37-25.46,31.7-44.37,59.3-44.37a56.43,56.43,0,0,1,40.51,16.08L176.47,155a101.24,101.24,0,0,0-70.75-27.84,105.55,105.55,0,0,0-94.38,59.11,107.64,107.64,0,0,0,0,96.18v.15a105.41,105.41,0,0,0,94.38,59c28.47,0,52.55-9.53,70-25.91,20-18.61,31.41-46.15,31.41-78.91A133.76,133.76,0,0,0,205.38,215Zm389.41-4c-10.13-9.38-23.93-14.14-41.39-14.14-22.46,0-39.34,8.34-50.5,24.86l20.85,13.26q11.45-17,31.26-17a34.05,34.05,0,0,1,22.75,8.79A28.14,28.14,0,0,1,487.79,248v5.51c-9.1-5.07-20.55-7.75-34.64-7.75-16.44,0-29.65,3.88-39.49,11.77s-14.82,18.31-14.82,31.56a39.74,39.74,0,0,0,13.94,31.27c9.25,8.34,21,12.51,34.79,12.51,16.29,0,29.21-7.3,39-21.89h1v17.72h22.61V250C510.25,233.45,505.26,220.34,495.13,211ZM475.9,300.3a37.32,37.32,0,0,1-26.57,11.16A28.61,28.61,0,0,1,431,305.21a19.41,19.41,0,0,1-7.77-15.63c0-7,3.22-12.81,9.54-17.42s14.53-7,24.07-7C470,265,480.3,268,487.64,273.94,487.64,284.07,483.68,292.85,475.9,300.3Zm-93.65-142A55.71,55.71,0,0,0,341.74,142H279.07V328.74H302.7V253.1h39c16,0,29.5-5.36,40.51-15.93.88-.89,1.76-1.79,2.65-2.68A54.45,54.45,0,0,0,382.25,158.26Zm-16.58,62.23a30.65,30.65,0,0,1-23.34,9.68H302.7V165h39.63a32,32,0,0,1,22.6,9.23A33.18,33.18,0,0,1,365.67,220.49ZM614.31,201,577.77,292.7h-.45L539.9,201H514.21L566,320.55l-29.35,64.32H561L640,201Z"],intercom:[448,512,[],"f7af","M392 32H56C25.1 32 0 57.1 0 88v336c0 30.9 25.1 56 56 56h336c30.9 0 56-25.1 56-56V88c0-30.9-25.1-56-56-56zm-108.3 82.1c0-19.8 29.9-19.8 29.9 0v199.5c0 19.8-29.9 19.8-29.9 0V114.1zm-74.6-7.5c0-19.8 29.9-19.8 29.9 0v216.5c0 19.8-29.9 19.8-29.9 0V106.6zm-74.7 7.5c0-19.8 29.9-19.8 29.9 0v199.5c0 19.8-29.9 19.8-29.9 0V114.1zM59.7 144c0-19.8 29.9-19.8 29.9 0v134.3c0 19.8-29.9 19.8-29.9 0V144zm323.4 227.8c-72.8 63-241.7 65.4-318.1 0-15-12.8 4.4-35.5 19.4-22.7 65.9 55.3 216.1 53.9 279.3 0 14.9-12.9 34.3 9.8 19.4 22.7zm5.2-93.5c0 19.8-29.9 19.8-29.9 0V144c0-19.8 29.9-19.8 29.9 0v134.3z"],zhihu:[640,512,[],"f63f","M170.54 148.13v217.54l23.43.01 7.71 26.37 42.01-26.37h49.53V148.13H170.54zm97.75 193.93h-27.94l-27.9 17.51-5.08-17.47-11.9-.04V171.75h72.82v170.31zm-118.46-94.39H97.5c1.74-27.1 2.2-51.59 2.2-73.46h51.16s1.97-22.56-8.58-22.31h-88.5c3.49-13.12 7.87-26.66 13.12-40.67 0 0-24.07 0-32.27 21.57-3.39 8.9-13.21 43.14-30.7 78.12 5.89-.64 25.37-1.18 36.84-22.21 2.11-5.89 2.51-6.66 5.14-14.53h28.87c0 10.5-1.2 66.88-1.68 73.44H20.83c-11.74 0-15.56 23.62-15.56 23.62h65.58C66.45 321.1 42.83 363.12 0 396.34c20.49 5.85 40.91-.93 51-9.9 0 0 22.98-20.9 35.59-69.25l53.96 64.94s7.91-26.89-1.24-39.99c-7.58-8.92-28.06-33.06-36.79-41.81L87.9 311.95c4.36-13.98 6.99-27.55 7.87-40.67h61.65s-.09-23.62-7.59-23.62v.01zm412.02-1.6c20.83-25.64 44.98-58.57 44.98-58.57s-18.65-14.8-27.38-4.06c-6 8.15-36.83 48.2-36.83 48.2l19.23 14.43zm-150.09-59.09c-9.01-8.25-25.91 2.13-25.91 2.13s39.52 55.04 41.12 57.45l19.46-13.73s-25.67-37.61-34.66-45.86h-.01zM640 258.35c-19.78 0-130.91.93-131.06.93v-101c4.81 0 12.42-.4 22.85-1.2 40.88-2.41 70.13-4 87.77-4.81 0 0 12.22-27.19-.59-33.44-3.07-1.18-23.17 4.58-23.17 4.58s-165.22 16.49-232.36 18.05c1.6 8.82 7.62 17.08 15.78 19.55 13.31 3.48 22.69 1.7 49.15.89 24.83-1.6 43.68-2.43 56.51-2.43v99.81H351.41s2.82 22.31 25.51 22.85h107.94v70.92c0 13.97-11.19 21.99-24.48 21.12-14.08.11-26.08-1.15-41.69-1.81 1.99 3.97 6.33 14.39 19.31 21.84 9.88 4.81 16.17 6.57 26.02 6.57 29.56 0 45.67-17.28 44.89-45.31v-73.32h122.36c9.68 0 8.7-23.78 8.7-23.78l.03-.01z"],korvue:[446,512,[],"f42f","M386.5 34h-327C26.8 34 0 60.8 0 93.5v327.1C0 453.2 26.8 480 59.5 480h327.1c33 0 59.5-26.8 59.5-59.5v-327C446 60.8 419.2 34 386.5 34zM87.1 120.8h96v116l61.8-116h110.9l-81.2 132H87.1v-132zm161.8 272.1l-65.7-113.6v113.6h-96V262.1h191.5l88.6 130.8H248.9z"],pix:[512,512,[],"e43a","M242.4 292.5C247.8 287.1 257.1 287.1 262.5 292.5L339.5 369.5C353.7 383.7 372.6 391.5 392.6 391.5H407.7L310.6 488.6C280.3 518.1 231.1 518.1 200.8 488.6L103.3 391.2H112.6C132.6 391.2 151.5 383.4 165.7 369.2L242.4 292.5zM262.5 218.9C256.1 224.4 247.9 224.5 242.4 218.9L165.7 142.2C151.5 127.1 132.6 120.2 112.6 120.2H103.3L200.7 22.76C231.1-7.586 280.3-7.586 310.6 22.76L407.8 119.9H392.6C372.6 119.9 353.7 127.7 339.5 141.9L262.5 218.9zM112.6 142.7C126.4 142.7 139.1 148.3 149.7 158.1L226.4 234.8C233.6 241.1 243 245.6 252.5 245.6C261.9 245.6 271.3 241.1 278.5 234.8L355.5 157.8C365.3 148.1 378.8 142.5 392.6 142.5H430.3L488.6 200.8C518.9 231.1 518.9 280.3 488.6 310.6L430.3 368.9H392.6C378.8 368.9 365.3 363.3 355.5 353.5L278.5 276.5C264.6 262.6 240.3 262.6 226.4 276.6L149.7 353.2C139.1 363 126.4 368.6 112.6 368.6H80.78L22.76 310.6C-7.586 280.3-7.586 231.1 22.76 200.8L80.78 142.7H112.6z"],"steam-symbol":[448,512,[],"f3f6","M395.5 177.5c0 33.8-27.5 61-61 61-33.8 0-61-27.3-61-61s27.3-61 61-61c33.5 0 61 27.2 61 61zm52.5.2c0 63-51 113.8-113.7 113.8L225 371.3c-4 43-40.5 76.8-84.5 76.8-40.5 0-74.7-28.8-83-67L0 358V250.7L97.2 290c15.1-9.2 32.2-13.3 52-11.5l71-101.7c.5-62.3 51.5-112.8 114-112.8C397 64 448 115 448 177.7zM203 363c0-34.7-27.8-62.5-62.5-62.5-4.5 0-9 .5-13.5 1.5l26 10.5c25.5 10.2 38 39 27.7 64.5-10.2 25.5-39.2 38-64.7 27.5-10.2-4-20.5-8.3-30.7-12.2 10.5 19.7 31.2 33.2 55.2 33.2 34.7 0 62.5-27.8 62.5-62.5zm207.5-185.3c0-42-34.3-76.2-76.2-76.2-42.3 0-76.5 34.2-76.5 76.2 0 42.2 34.3 76.2 76.5 76.2 41.9.1 76.2-33.9 76.2-76.2z"]};!function(c){try{for(var l=arguments.length,s=new Array(1<l?l-1:0),a=1;a<l;a++)s[a-1]=arguments[a];c.apply(void 0,s)}catch(c){if(!M)throw c}}((function(){d("fab",p),d("fa-brands",p)}))}(),function(){"use strict";var c={},l={};try{"undefined"!=typeof window&&(c=window),"undefined"!=typeof document&&(l=document)}catch(c){}var s=void 0===(v=(c.navigator||{}).userAgent)?"":v,a=c,z=l;function e(c,l){var s,a=Object.keys(c);return Object.getOwnPropertySymbols&&(s=Object.getOwnPropertySymbols(c),l&&(s=s.filter((function(l){return Object.getOwnPropertyDescriptor(c,l).enumerable}))),a.push.apply(a,s)),a}function H(c){for(var l=1;l<arguments.length;l++){var s=null!=arguments[l]?arguments[l]:{};l%2?e(Object(s),!0).forEach((function(l){t(c,l,s[l])})):Object.getOwnPropertyDescriptors?Object.defineProperties(c,Object.getOwnPropertyDescriptors(s)):e(Object(s)).forEach((function(l){Object.defineProperty(c,l,Object.getOwnPropertyDescriptor(s,l))}))}return c}function t(c,l,s){return l in c?Object.defineProperty(c,l,{value:s,enumerable:!0,configurable:!0,writable:!0}):c[l]=s,c}function V(c,l){(null==l||l>c.length)&&(l=c.length);for(var s=0,a=new Array(l);s<l;s++)a[s]=c[s];return a}a.document,z.documentElement&&z.head&&"function"==typeof z.addEventListener&&z.createElement,~s.indexOf("MSIE")||s.indexOf("Trident/");var r="___FONT_AWESOME___",M=function(){try{return!0}catch(c){return!1}}(),h="classic",n="sharp",i=[h,n];function m(c){return new Proxy(c,{get:function(c,l){return l in c?c[l]:c[h]}})}m((t(f={},h,{fa:"solid",fas:"solid","fa-solid":"solid",far:"regular","fa-regular":"regular",fal:"light","fa-light":"light",fat:"thin","fa-thin":"thin",fad:"duotone","fa-duotone":"duotone",fab:"brands","fa-brands":"brands",fak:"kit",fakd:"kit","fa-kit":"kit","fa-kit-duotone":"kit"}),t(f,n,{fa:"solid",fass:"solid","fa-solid":"solid",fasr:"regular","fa-regular":"regular",fasl:"light","fa-light":"light",fast:"thin","fa-thin":"thin"}),f));var o=m((t(C={},h,{solid:"fas",regular:"far",light:"fal",thin:"fat",duotone:"fad",brands:"fab",kit:"fak"}),t(C,n,{solid:"fass",regular:"fasr",light:"fasl",thin:"fast"}),C)),f=(s=(m((t(v={},h,{fab:"fa-brands",fad:"fa-duotone",fak:"fa-kit",fal:"fa-light",far:"fa-regular",fas:"fa-solid",fat:"fa-thin"}),t(v,n,{fass:"fa-solid",fasr:"fa-regular",fasl:"fa-light",fast:"fa-thin"}),v)),m((t(c={},h,{"fa-brands":"fab","fa-duotone":"fad","fa-kit":"fak","fa-light":"fal","fa-regular":"far","fa-solid":"fas","fa-thin":"fat"}),t(c,n,{"fa-solid":"fass","fa-regular":"fasr","fa-light":"fasl","fa-thin":"fast"}),c)),m((t(l={},h,{900:"fas",400:"far",normal:"far",300:"fal",100:"fat"}),t(l,n,{900:"fass",400:"fasr",300:"fasl",100:"fast"}),l)),z=[1,2,3,4,5,6,7,8,9,10]).concat([11,12,13,14,15,16,17,18,19,20]),"duotone-group"),C="swap-opacity",v="primary";c="secondary",l=new Set,Object.keys(o[h]).map(l.add.bind(l)),Object.keys(o[n]).map(l.add.bind(l)),[].concat(i,function(c){if(Array.isArray(c))return V(c)}(l)||function(c){if("undefined"!=typeof Symbol&&null!=c[Symbol.iterator]||null!=c["@@iterator"])return Array.from(c)}(l)||function(c,l){if(c){if("string"==typeof c)return V(c,l);var s=Object.prototype.toString.call(c).slice(8,-1);return"Map"===(s="Object"===s&&c.constructor?c.constructor.name:s)||"Set"===s?Array.from(c):"Arguments"===s||/^(?:Ui|I)nt(?:8|16|32)(?:Clamped)?Array$/.test(s)?V(c,l):void 0}}(l)||function(){throw new TypeError("Invalid attempt to spread non-iterable instance.\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method.")}(),["2xs","xs","sm","lg","xl","2xl","beat","border","fade","beat-fade","bounce","flip-both","flip-horizontal","flip-vertical","flip","fw","inverse","layers-counter","layers-text","layers","li","pull-left","pull-right","pulse","rotate-180","rotate-270","rotate-90","rotate-by","shake","spin-pulse","spin-reverse","spin","stack-1x","stack-2x","stack","ul",f,C,v,c]).concat(z.map((function(c){return"".concat(c,"x")}))).concat(s.map((function(c){return"w-".concat(c)}))),(a=a||{})[r]||(a[r]={}),a[r].styles||(a[r].styles={}),a[r].hooks||(a[r].hooks={}),a[r].shims||(a[r].shims=[]);var L=a[r];function u(c){return Object.keys(c).reduce((function(l,s){var a=c[s];return a.icon?l[a.iconName]=a.icon:l[s]=a,l}),{})}function d(c,l,s){s=void 0!==(a=(2<arguments.length&&void 0!==s?s:{}).skipHooks)&&a;var a=u(l);"function"!=typeof L.hooks.addPack||s?L.styles[c]=H(H({},L.styles[c]||{}),a):L.hooks.addPack(c,u(l)),"fas"===c&&d("fa",l)}var p={"trash-can":[448,512,[61460,"trash-alt"],"f2ed","M170.5 51.6L151.5 80h145l-19-28.4c-1.5-2.2-4-3.6-6.7-3.6H177.1c-2.7 0-5.2 1.3-6.7 3.6zm147-26.6L354.2 80H368h48 8c13.3 0 24 10.7 24 24s-10.7 24-24 24h-8V432c0 44.2-35.8 80-80 80H112c-44.2 0-80-35.8-80-80V128H24c-13.3 0-24-10.7-24-24S10.7 80 24 80h8H80 93.8l36.7-55.1C140.9 9.4 158.4 0 177.1 0h93.7c18.7 0 36.2 9.4 46.6 24.9zM80 128V432c0 17.7 14.3 32 32 32H336c17.7 0 32-14.3 32-32V128H80zm80 64V400c0 8.8-7.2 16-16 16s-16-7.2-16-16V192c0-8.8 7.2-16 16-16s16 7.2 16 16zm80 0V400c0 8.8-7.2 16-16 16s-16-7.2-16-16V192c0-8.8 7.2-16 16-16s16 7.2 16 16zm80 0V400c0 8.8-7.2 16-16 16s-16-7.2-16-16V192c0-8.8 7.2-16 16-16s16 7.2 16 16z"],message:[512,512,["comment-alt"],"f27a","M160 368c26.5 0 48 21.5 48 48v16l72.5-54.4c8.3-6.2 18.4-9.6 28.8-9.6H448c8.8 0 16-7.2 16-16V64c0-8.8-7.2-16-16-16H64c-8.8 0-16 7.2-16 16V352c0 8.8 7.2 16 16 16h96zm48 124l-.2 .2-5.1 3.8-17.1 12.8c-4.8 3.6-11.3 4.2-16.8 1.5s-8.8-8.2-8.8-14.3V474.7v-6.4V468v-4V416H112 64c-35.3 0-64-28.7-64-64V64C0 28.7 28.7 0 64 0H448c35.3 0 64 28.7 64 64V352c0 35.3-28.7 64-64 64H309.3L208 492z"],"file-lines":[384,512,[128441,128462,61686,"file-alt","file-text"],"f15c","M64 464c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16H224v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm56 256c-13.3 0-24 10.7-24 24s10.7 24 24 24H264c13.3 0 24-10.7 24-24s-10.7-24-24-24H120zm0 96c-13.3 0-24 10.7-24 24s10.7 24 24 24H264c13.3 0 24-10.7 24-24s-10.7-24-24-24H120z"],"calendar-days":[448,512,["calendar-alt"],"f073","M152 24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H64C28.7 64 0 92.7 0 128v16 48V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V192 144 128c0-35.3-28.7-64-64-64H344V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H152V24zM48 192h80v56H48V192zm0 104h80v64H48V296zm128 0h96v64H176V296zm144 0h80v64H320V296zm80-48H320V192h80v56zm0 160v40c0 8.8-7.2 16-16 16H320V408h80zm-128 0v56H176V408h96zm-144 0v56H64c-8.8 0-16-7.2-16-16V408h80zM272 248H176V192h96v56z"],"hand-point-right":[512,512,[],"f0a4","M448 128l-177.6 0c1 5.2 1.6 10.5 1.6 16l0 16 32 0 144 0c8.8 0 16-7.2 16-16s-7.2-16-16-16zM224 144c0-17.7-14.3-32-32-32c0 0 0 0 0 0l-24 0c-66.3 0-120 53.7-120 120l0 48c0 52.5 33.7 97.1 80.7 113.4c-.5-3.1-.7-6.2-.7-9.4c0-20 9.2-37.9 23.6-49.7c-4.9-9-7.6-19.4-7.6-30.3c0-15.1 5.3-29 14-40c-8.8-11-14-24.9-14-40l0-40c0-13.3 10.7-24 24-24s24 10.7 24 24l0 40c0 8.8 7.2 16 16 16s16-7.2 16-16l0-40 0-40zM192 64s0 0 0 0c18 0 34.6 6 48 16l208 0c35.3 0 64 28.7 64 64s-28.7 64-64 64l-82 0c1.3 5.1 2 10.5 2 16c0 25.3-14.7 47.2-36 57.6c2.6 7 4 14.5 4 22.4c0 20-9.2 37.9-23.6 49.7c4.9 9 7.6 19.4 7.6 30.3c0 35.3-28.7 64-64 64l-64 0-24 0C75.2 448 0 372.8 0 280l0-48C0 139.2 75.2 64 168 64l24 0zm64 336c8.8 0 16-7.2 16-16s-7.2-16-16-16l-48 0-16 0c-8.8 0-16 7.2-16 16s7.2 16 16 16l64 0zm16-176c0 5.5-.7 10.9-2 16l2 0 32 0c8.8 0 16-7.2 16-16s-7.2-16-16-16l-32 0 0 16zm-24 64l-40 0c-8.8 0-16 7.2-16 16s7.2 16 16 16l48 0 16 0c8.8 0 16-7.2 16-16s-7.2-16-16-16l-24 0z"],"face-smile-beam":[512,512,[128522,"smile-beam"],"f5b8","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm177.6 62.1C192.8 334.5 218.8 352 256 352s63.2-17.5 78.4-33.9c9-9.7 24.2-10.4 33.9-1.4s10.4 24.2 1.4 33.9c-22 23.8-60 49.4-113.6 49.4s-91.7-25.5-113.6-49.4c-9-9.7-8.4-24.9 1.4-33.9s24.9-8.4 33.9 1.4zm40-89.3l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z"],"face-grin-stars":[512,512,[129321,"grin-stars"],"f587","M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM183.2 132.6c-1.3-2.8-4.1-4.6-7.2-4.6s-5.9 1.8-7.2 4.6l-16.6 34.7-38.1 5c-3.1 .4-5.6 2.5-6.6 5.5s-.1 6.2 2.1 8.3l27.9 26.5-7 37.8c-.6 3 .7 6.1 3.2 7.9s5.8 2 8.5 .6L176 240.5l33.8 18.3c2.7 1.5 6 1.3 8.5-.6s3.7-4.9 3.2-7.9l-7-37.8L242.4 186c2.2-2.1 3.1-5.3 2.1-8.3s-3.5-5.1-6.6-5.5l-38.1-5-16.6-34.7zm160 0c-1.3-2.8-4.1-4.6-7.2-4.6s-5.9 1.8-7.2 4.6l-16.6 34.7-38.1 5c-3.1 .4-5.6 2.5-6.6 5.5s-.1 6.2 2.1 8.3l27.9 26.5-7 37.8c-.6 3 .7 6.1 3.2 7.9s5.8 2 8.5 .6L336 240.5l33.8 18.3c2.7 1.5 6 1.3 8.5-.6s3.7-4.9 3.2-7.9l-7-37.8L402.4 186c2.2-2.1 3.1-5.3 2.1-8.3s-3.5-5.1-6.6-5.5l-38.1-5-16.6-34.7zm6.3 175.8c-28.9 6.8-60.5 10.5-93.6 10.5s-64.7-3.7-93.6-10.5c-18.7-4.4-35.9 12-25.5 28.1c24.6 38.1 68.7 63.5 119.1 63.5s94.5-25.4 119.1-63.5c10.4-16.1-6.8-32.5-25.5-28.1z"],"address-book":[512,512,[62138,"contact-book"],"f2b9","M384 48c8.8 0 16 7.2 16 16V448c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16H384zM96 0C60.7 0 32 28.7 32 64V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H96zM240 256a64 64 0 1 0 0-128 64 64 0 1 0 0 128zm-32 32c-44.2 0-80 35.8-80 80c0 8.8 7.2 16 16 16H336c8.8 0 16-7.2 16-16c0-44.2-35.8-80-80-80H208zM512 80c0-8.8-7.2-16-16-16s-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V80zM496 192c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V208c0-8.8-7.2-16-16-16zm16 144c0-8.8-7.2-16-16-16s-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V336z"],comments:[640,512,[128490,61670],"f086","M88.2 309.1c9.8-18.3 6.8-40.8-7.5-55.8C59.4 230.9 48 204 48 176c0-63.5 63.8-128 160-128s160 64.5 160 128s-63.8 128-160 128c-13.1 0-25.8-1.3-37.8-3.6c-10.4-2-21.2-.6-30.7 4.2c-4.1 2.1-8.3 4.1-12.6 6c-16 7.2-32.9 13.5-49.9 18c2.8-4.6 5.4-9.1 7.9-13.6c1.1-1.9 2.2-3.9 3.2-5.9zM0 176c0 41.8 17.2 80.1 45.9 110.3c-.9 1.7-1.9 3.5-2.8 5.1c-10.3 18.4-22.3 36.5-36.6 52.1c-6.6 7-8.3 17.2-4.6 25.9C5.8 378.3 14.4 384 24 384c43 0 86.5-13.3 122.7-29.7c4.8-2.2 9.6-4.5 14.2-6.8c15.1 3 30.9 4.5 47.1 4.5c114.9 0 208-78.8 208-176S322.9 0 208 0S0 78.8 0 176zM432 480c16.2 0 31.9-1.6 47.1-4.5c4.6 2.3 9.4 4.6 14.2 6.8C529.5 498.7 573 512 616 512c9.6 0 18.2-5.7 22-14.5c3.8-8.8 2-19-4.6-25.9c-14.2-15.6-26.2-33.7-36.6-52.1c-.9-1.7-1.9-3.4-2.8-5.1C622.8 384.1 640 345.8 640 304c0-94.4-87.9-171.5-198.2-175.8c4.1 15.2 6.2 31.2 6.2 47.8l0 .6c87.2 6.7 144 67.5 144 127.4c0 28-11.4 54.9-32.7 77.2c-14.3 15-17.3 37.6-7.5 55.8c1.1 2 2.2 4 3.2 5.9c2.5 4.5 5.2 9 7.9 13.6c-17-4.5-33.9-10.7-49.9-18c-4.3-1.9-8.5-3.9-12.6-6c-9.5-4.8-20.3-6.2-30.7-4.2c-12.1 2.4-24.7 3.6-37.8 3.6c-61.7 0-110-26.5-136.8-62.3c-16 5.4-32.8 9.4-50 11.8C279 439.8 350 480 432 480z"],paste:[512,512,["file-clipboard"],"f0ea","M104.6 48H64C28.7 48 0 76.7 0 112V384c0 35.3 28.7 64 64 64h96V400H64c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16H80c0 17.7 14.3 32 32 32h72.4C202 108.4 227.6 96 256 96h62c-7.1-27.6-32.2-48-62-48H215.4C211.6 20.9 188.2 0 160 0s-51.6 20.9-55.4 48zM144 56a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zM448 464H256c-8.8 0-16-7.2-16-16V192c0-8.8 7.2-16 16-16l140.1 0L464 243.9V448c0 8.8-7.2 16-16 16zM256 512H448c35.3 0 64-28.7 64-64V243.9c0-12.7-5.1-24.9-14.1-33.9l-67.9-67.9c-9-9-21.2-14.1-33.9-14.1H256c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64z"],"face-grin-tongue-squint":[512,512,[128541,"grin-tongue-squint"],"f58a","M464 256c0-114.9-93.1-208-208-208S48 141.1 48 256c0 81.7 47.1 152.4 115.7 186.4c-2.4-8.4-3.7-17.3-3.7-26.4V392.7c-24-17.5-43.1-41.4-54.8-69.2c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19c12.3-3.8 24.3 6.9 19.3 18.7c-11.8 28-31.1 52-55.4 69.6V416c0 9.2-1.3 18-3.7 26.4C416.9 408.4 464 337.7 464 256zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm116-98.9c0-9 9.6-14.7 17.5-10.5l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6zm262.5-10.5c7.9-4.2 17.5 1.5 17.5 10.5c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9zM320 416V378.6c0-14.7-11.9-26.6-26.6-26.6h-2c-11.3 0-21.1 7.9-23.6 18.9c-2.8 12.6-20.8 12.6-23.6 0c-2.5-11.1-12.3-18.9-23.6-18.9h-2c-14.7 0-26.6 11.9-26.6 26.6V416c0 35.3 28.7 64 64 64s64-28.7 64-64z"],"face-flushed":[512,512,[128563,"flushed"],"f579","M464 256A208 208 0 1 1 48 256a208 208 0 1 1 416 0zM256 0a256 256 0 1 0 0 512A256 256 0 1 0 256 0zM160.4 248a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm216-24a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zM192 336c-13.3 0-24 10.7-24 24s10.7 24 24 24H320c13.3 0 24-10.7 24-24s-10.7-24-24-24H192zM160 176a48 48 0 1 1 0 96 48 48 0 1 1 0-96zm0 128a80 80 0 1 0 0-160 80 80 0 1 0 0 160zm144-80a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm128 0a80 80 0 1 0 -160 0 80 80 0 1 0 160 0z"],"square-caret-right":[448,512,["caret-square-right"],"f152","M400 96c0-8.8-7.2-16-16-16L64 80c-8.8 0-16 7.2-16 16l0 320c0 8.8 7.2 16 16 16l320 0c8.8 0 16-7.2 16-16l0-320zM384 32c35.3 0 64 28.7 64 64l0 320c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64L0 96C0 60.7 28.7 32 64 32l320 0zM320 256c0 6.7-2.8 13-7.7 17.6l-112 104c-7 6.5-17.2 8.2-25.9 4.4s-14.4-12.5-14.4-22l0-208c0-9.5 5.7-18.2 14.4-22s18.9-2.1 25.9 4.4l112 104c4.9 4.5 7.7 10.9 7.7 17.6z"],"square-minus":[448,512,[61767,"minus-square"],"f146","M64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V96c0-8.8-7.2-16-16-16H64zM0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM152 232H296c13.3 0 24 10.7 24 24s-10.7 24-24 24H152c-13.3 0-24-10.7-24-24s10.7-24 24-24z"],compass:[512,512,[129517],"f14e","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm306.7 69.1L162.4 380.6c-19.4 7.5-38.5-11.6-31-31l55.5-144.3c3.3-8.5 9.9-15.1 18.4-18.4l144.3-55.5c19.4-7.5 38.5 11.6 31 31L325.1 306.7c-3.2 8.5-9.9 15.1-18.4 18.4zM288 256a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z"],"square-caret-down":[448,512,["caret-square-down"],"f150","M384 432c8.8 0 16-7.2 16-16l0-320c0-8.8-7.2-16-16-16L64 80c-8.8 0-16 7.2-16 16l0 320c0 8.8 7.2 16 16 16l320 0zm64-16c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64L0 96C0 60.7 28.7 32 64 32l320 0c35.3 0 64 28.7 64 64l0 320zM224 352c-6.7 0-13-2.8-17.6-7.7l-104-112c-6.5-7-8.2-17.2-4.4-25.9s12.5-14.4 22-14.4l208 0c9.5 0 18.2 5.7 22 14.4s2.1 18.9-4.4 25.9l-104 112c-4.5 4.9-10.9 7.7-17.6 7.7z"],"face-kiss-beam":[512,512,[128537,"kiss-beam"],"f597","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm304.7 41.7c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4c-2.7 1.5-5.7 3-8.7 4.3c3.1 1.3 6 2.7 8.7 4.3c6.6 3.7 12.5 8.2 16.7 13.4c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4C274.7 427.1 257.4 432 240 432c-3.6 0-6.8-2.5-7.7-6s.6-7.2 3.8-9l0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1c-.4-.2-.7-.4-.9-.5l-.2-.1 0 0 0 0 0 0c-2.5-1.4-4.1-4.1-4.1-7s1.6-5.6 4.1-7l0 0 0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1c-.4-.2-.7-.4-.9-.5l-.2-.1 0 0 0 0 0 0c-3.2-1.8-4.7-5.5-3.8-9s4.1-6 7.7-6c17.4 0 34.7 4.9 47.9 12.3c6.6 3.7 12.5 8.2 16.7 13.4zm-87.1-68.9l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z"],lightbulb:[384,512,[128161],"f0eb","M297.2 248.9C311.6 228.3 320 203.2 320 176c0-70.7-57.3-128-128-128S64 105.3 64 176c0 27.2 8.4 52.3 22.8 72.9c3.7 5.3 8.1 11.3 12.8 17.7l0 0c12.9 17.7 28.3 38.9 39.8 59.8c10.4 19 15.7 38.8 18.3 57.5H109c-2.2-12-5.9-23.7-11.8-34.5c-9.9-18-22.2-34.9-34.5-51.8l0 0 0 0c-5.2-7.1-10.4-14.2-15.4-21.4C27.6 247.9 16 213.3 16 176C16 78.8 94.8 0 192 0s176 78.8 176 176c0 37.3-11.6 71.9-31.4 100.3c-5 7.2-10.2 14.3-15.4 21.4l0 0 0 0c-12.3 16.8-24.6 33.7-34.5 51.8c-5.9 10.8-9.6 22.5-11.8 34.5H226.4c2.6-18.7 7.9-38.6 18.3-57.5c11.5-20.9 26.9-42.1 39.8-59.8l0 0 0 0 0 0c4.7-6.4 9-12.4 12.7-17.7zM192 128c-26.5 0-48 21.5-48 48c0 8.8-7.2 16-16 16s-16-7.2-16-16c0-44.2 35.8-80 80-80c8.8 0 16 7.2 16 16s-7.2 16-16 16zm0 384c-44.2 0-80-35.8-80-80V416H272v16c0 44.2-35.8 80-80 80z"],flag:[448,512,[127988,61725],"f024","M48 24C48 10.7 37.3 0 24 0S0 10.7 0 24V64 350.5 400v88c0 13.3 10.7 24 24 24s24-10.7 24-24V388l80.3-20.1c41.1-10.3 84.6-5.5 122.5 13.4c44.2 22.1 95.5 24.8 141.7 7.4l34.7-13c12.5-4.7 20.8-16.6 20.8-30V66.1c0-23-24.2-38-44.8-27.7l-9.6 4.8c-46.3 23.2-100.8 23.2-147.1 0c-35.1-17.6-75.4-22-113.5-12.5L48 52V24zm0 77.5l96.6-24.2c27-6.7 55.5-3.6 80.4 8.8c54.9 27.4 118.7 29.7 175 6.8V334.7l-24.4 9.1c-33.7 12.6-71.2 10.7-103.4-5.4c-48.2-24.1-103.3-30.1-155.6-17.1L48 338.5v-237z"],"square-check":[448,512,[9745,9989,61510,"check-square"],"f14a","M64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V96c0-8.8-7.2-16-16-16H64zM0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM337 209L209 337c-9.4 9.4-24.6 9.4-33.9 0l-64-64c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47L303 175c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z"],"circle-dot":[512,512,[128280,"dot-circle"],"f192","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm256-96a96 96 0 1 1 0 192 96 96 0 1 1 0-192z"],"face-dizzy":[512,512,["dizzy"],"f567","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm256 32a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM103 135c9.4-9.4 24.6-9.4 33.9 0l23 23 23-23c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-23 23 23 23c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-23-23-23 23c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l23-23-23-23c-9.4-9.4-9.4-24.6 0-33.9zm192 0c9.4-9.4 24.6-9.4 33.9 0l23 23 23-23c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-23 23 23 23c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-23-23-23 23c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l23-23-23-23c-9.4-9.4-9.4-24.6 0-33.9z"],futbol:[512,512,[9917,"futbol-ball","soccer-ball"],"f1e3","M435.4 361.3l-89.7-6c-5.2-.3-10.3 1.1-14.5 4.2s-7.2 7.4-8.4 12.5l-22 87.2c-14.4 3.2-29.4 4.8-44.8 4.8s-30.3-1.7-44.8-4.8l-22-87.2c-1.3-5-4.3-9.4-8.4-12.5s-9.3-4.5-14.5-4.2l-89.7 6C61.7 335.9 51.9 307 49 276.2L125 228.3c4.4-2.8 7.6-7 9.2-11.9s1.4-10.2-.5-15L100.4 118c19.9-22.4 44.6-40.5 72.4-52.7l69.1 57.6c4 3.3 9 5.1 14.1 5.1s10.2-1.8 14.1-5.1l69.1-57.6c27.8 12.2 52.5 30.3 72.4 52.7l-33.4 83.4c-1.9 4.8-2.1 10.1-.5 15s4.9 9.1 9.2 11.9L463 276.2c-3 30.8-12.7 59.7-27.6 85.1zM256 48l.9 0h-1.8l.9 0zM56.7 196.2c.9-3 1.9-6.1 2.9-9.1l-2.9 9.1zM132 423l3.8 2.7c-1.3-.9-2.5-1.8-3.8-2.7zm248.1-.1c-1.3 1-2.7 2-4 2.9l4-2.9zm75.2-226.6l-3-9.2c1.1 3 2.1 6.1 3 9.2zM256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm14.1-325.7c-8.4-6.1-19.8-6.1-28.2 0L194 221c-8.4 6.1-11.9 16.9-8.7 26.8l18.3 56.3c3.2 9.9 12.4 16.6 22.8 16.6h59.2c10.4 0 19.6-6.7 22.8-16.6l18.3-56.3c3.2-9.9-.3-20.7-8.7-26.8l-47.9-34.8z"],"pen-to-square":[512,512,["edit"],"f044","M441 58.9L453.1 71c9.4 9.4 9.4 24.6 0 33.9L424 134.1 377.9 88 407 58.9c9.4-9.4 24.6-9.4 33.9 0zM209.8 256.2L344 121.9 390.1 168 255.8 302.2c-2.9 2.9-6.5 5-10.4 6.1l-58.5 16.7 16.7-58.5c1.1-3.9 3.2-7.5 6.1-10.4zM373.1 25L175.8 222.2c-8.7 8.7-15 19.4-18.3 31.1l-28.6 100c-2.4 8.4-.1 17.4 6.1 23.6s15.2 8.5 23.6 6.1l100-28.6c11.8-3.4 22.5-9.7 31.1-18.3L487 138.9c28.1-28.1 28.1-73.7 0-101.8L474.9 25C446.8-3.1 401.2-3.1 373.1 25zM88 64C39.4 64 0 103.4 0 152V424c0 48.6 39.4 88 88 88H360c48.6 0 88-39.4 88-88V312c0-13.3-10.7-24-24-24s-24 10.7-24 24V424c0 22.1-17.9 40-40 40H88c-22.1 0-40-17.9-40-40V152c0-22.1 17.9-40 40-40H200c13.3 0 24-10.7 24-24s-10.7-24-24-24H88z"],"hourglass-half":[384,512,["hourglass-2"],"f252","M0 24C0 10.7 10.7 0 24 0H360c13.3 0 24 10.7 24 24s-10.7 24-24 24h-8V67c0 40.3-16 79-44.5 107.5L225.9 256l81.5 81.5C336 366 352 404.7 352 445v19h8c13.3 0 24 10.7 24 24s-10.7 24-24 24H24c-13.3 0-24-10.7-24-24s10.7-24 24-24h8V445c0-40.3 16-79 44.5-107.5L158.1 256 76.5 174.5C48 146 32 107.3 32 67V48H24C10.7 48 0 37.3 0 24zM110.5 371.5c-3.9 3.9-7.5 8.1-10.7 12.5H284.2c-3.2-4.4-6.8-8.6-10.7-12.5L192 289.9l-81.5 81.5zM284.2 128C297 110.4 304 89 304 67V48H80V67c0 22.1 7 43.4 19.8 61H284.2z"],"eye-slash":[640,512,[],"f070","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L525.6 386.7c39.6-40.6 66.4-86.1 79.9-118.4c3.3-7.9 3.3-16.7 0-24.6c-14.9-35.7-46.2-87.7-93-131.1C465.5 68.8 400.8 32 320 32c-68.2 0-125 26.3-169.3 60.8L38.8 5.1zm151 118.3C226 97.7 269.5 80 320 80c65.2 0 118.8 29.6 159.9 67.7C518.4 183.5 545 226 558.6 256c-12.6 28-36.6 66.8-70.9 100.9l-53.8-42.2c9.1-17.6 14.2-37.5 14.2-58.7c0-70.7-57.3-128-128-128c-32.2 0-61.7 11.9-84.2 31.5l-46.1-36.1zM394.9 284.2l-81.5-63.9c4.2-8.5 6.6-18.2 6.6-28.3c0-5.5-.7-10.9-2-16c.7 0 1.3 0 2 0c44.2 0 80 35.8 80 80c0 9.9-1.8 19.4-5.1 28.2zm51.3 163.3l-41.9-33C378.8 425.4 350.7 432 320 432c-65.2 0-118.8-29.6-159.9-67.7C121.6 328.5 95 286 81.4 256c8.3-18.4 21.5-41.5 39.4-64.8L83.1 161.5C60.3 191.2 44 220.8 34.5 243.7c-3.3 7.9-3.3 16.7 0 24.6c14.9 35.7 46.2 87.7 93 131.1C174.5 443.2 239.2 480 320 480c47.8 0 89.9-12.9 126.2-32.5zm-88-69.3L302 334c-23.5-5.4-43.1-21.2-53.7-42.3l-56.1-44.2c-.2 2.8-.3 5.6-.3 8.5c0 70.7 57.3 128 128 128c13.3 0 26.1-2 38.2-5.8z"],hand:[512,512,[129306,9995,"hand-paper"],"f256","M256 0c-25.3 0-47.2 14.7-57.6 36c-7-2.6-14.5-4-22.4-4c-35.3 0-64 28.7-64 64V261.5l-2.7-2.7c-25-25-65.5-25-90.5 0s-25 65.5 0 90.5L106.5 437c48 48 113.1 75 181 75H296h8c1.5 0 3-.1 4.5-.4c91.7-6.2 165-79.4 171.1-171.1c.3-1.5 .4-3 .4-4.5V160c0-35.3-28.7-64-64-64c-5.5 0-10.9 .7-16 2V96c0-35.3-28.7-64-64-64c-7.9 0-15.4 1.4-22.4 4C303.2 14.7 281.3 0 256 0zM240 96.1c0 0 0-.1 0-.1V64c0-8.8 7.2-16 16-16s16 7.2 16 16V95.9c0 0 0 .1 0 .1V232c0 13.3 10.7 24 24 24s24-10.7 24-24V96c0 0 0 0 0-.1c0-8.8 7.2-16 16-16s16 7.2 16 16v55.9c0 0 0 .1 0 .1v80c0 13.3 10.7 24 24 24s24-10.7 24-24V160.1c0 0 0-.1 0-.1c0-8.8 7.2-16 16-16s16 7.2 16 16V332.9c-.1 .6-.1 1.3-.2 1.9c-3.4 69.7-59.3 125.6-129 129c-.6 0-1.3 .1-1.9 .2H296h-8.5c-55.2 0-108.1-21.9-147.1-60.9L52.7 315.3c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L119 336.4c6.9 6.9 17.2 8.9 26.2 5.2s14.8-12.5 14.8-22.2V96c0-8.8 7.2-16 16-16c8.8 0 16 7.1 16 15.9V232c0 13.3 10.7 24 24 24s24-10.7 24-24V96.1z"],"hand-spock":[576,512,[128406],"f259","M170.2 80.8C161 47 180.8 12 214.6 2.4c34-9.6 69.4 10.2 79 44.2l30.3 107.1L337.1 84c6.6-34.7 40.1-57.5 74.8-50.9c31.4 6 53 33.9 52 64.9c10-2.6 20.8-2.8 31.5-.1c34.3 8.6 55.1 43.3 46.6 77.6L486.7 397.2C469.8 464.7 409.2 512 339.6 512c-11.2 0-22.5 0-33.7 0c-56.9 0-112.2-19-157.2-53.9l-92-71.6c-27.9-21.7-32.9-61.9-11.2-89.8s61.9-32.9 89.8-11.2l17 13.2L100.5 167.5c-13-32.9 3.2-70.1 36-83c11.1-4.4 22.7-5.4 33.7-3.7zm77.1-21.2c-2.4-8.5-11.2-13.4-19.7-11s-13.4 11.2-11 19.7l54.8 182.4c3.5 12.3-3.3 25.2-15.4 29.3s-25.3-2-30-13.9L174.9 138.1c-3.2-8.2-12.5-12.3-20.8-9s-12.3 12.5-9 20.8l73.3 185.6c12 30.3-23.7 57-49.4 37l-63.1-49.1c-7-5.4-17-4.2-22.5 2.8s-4.2 17 2.8 22.5l92 71.6c36.5 28.4 81.4 43.8 127.7 43.8c11.2 0 22.5 0 33.7 0c47.5 0 89-32.4 100.5-78.5l55.4-221.6c2.1-8.6-3.1-17.3-11.6-19.4s-17.3 3.1-19.4 11.6l-26 104C435.6 271.8 425 280 413 280c-16.5 0-28.9-15-25.8-31.2L415.7 99c1.7-8.7-4-17.1-12.7-18.7s-17.1 4-18.7 12.7L352.5 260c-2.2 11.6-12.4 20-24.2 20c-11 0-20.7-7.3-23.7-17.9L247.4 59.6z"],"face-kiss":[512,512,[128535,"kiss"],"f596","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm304.7 25.7c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4c-2.7 1.5-5.7 3-8.7 4.3c3.1 1.3 6 2.7 8.7 4.3c6.6 3.7 12.5 8.2 16.7 13.4c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4C274.7 411.1 257.4 416 240 416c-3.6 0-6.8-2.5-7.7-6s.6-7.2 3.8-9l0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1l-.8-.5-.1-.1-.2-.1 0 0 0 0 0 0c-2.5-1.4-4.1-4.1-4.1-7s1.6-5.6 4.1-7l0 0 0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1c-.4-.2-.7-.4-.9-.5l-.2-.1 0 0 0 0 0 0c-3.2-1.8-4.7-5.5-3.8-9s4.1-6 7.7-6c17.4 0 34.7 4.9 47.9 12.3c6.6 3.7 12.5 8.2 16.7 13.4zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"face-grin-tongue":[512,512,[128539,"grin-tongue"],"f589","M464 256c0-114.9-93.1-208-208-208S48 141.1 48 256c0 81.7 47.1 152.4 115.7 186.4c-2.4-8.4-3.7-17.3-3.7-26.4V363.6c-8.9-8-16.7-17.1-23.1-27.1c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5c18.7-4.4 35.9 12 25.5 28.1c-6.4 9.9-14.2 19-23 27V416c0 9.2-1.3 18-3.7 26.4C416.9 408.4 464 337.7 464 256zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm176.4-80a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM320 416V378.6c0-14.7-11.9-26.6-26.6-26.6h-2c-11.3 0-21.1 7.9-23.6 18.9c-2.8 12.6-20.8 12.6-23.6 0c-2.5-11.1-12.3-18.9-23.6-18.9h-2c-14.7 0-26.6 11.9-26.6 26.6V416c0 35.3 28.7 64 64 64s64-28.7 64-64z"],"chess-bishop":[320,512,[9821],"f43a","M104 0C90.7 0 80 10.7 80 24c0 11.2 7.6 20.6 18 23.2c-7.8 8-16.1 17-24.4 27C38.2 116.7 0 178.8 0 250.9c0 44.8 24.6 72.2 48 87.8V352H96V325c0-9-5-17.2-13-21.3c-18-9.3-35-24.7-35-52.7c0-55.5 29.8-106.8 62.4-145.9c16-19.2 32.1-34.8 44.2-45.5c1.9-1.7 3.7-3.2 5.3-4.6c1.7 1.4 3.4 3 5.3 4.6c12.1 10.7 28.2 26.3 44.2 45.5c5.3 6.3 10.5 13 15.5 20L159 191c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l57.8-57.8c12.8 25.9 21.2 54.3 21.2 83.8c0 28-17 43.4-35 52.7c-8 4.1-13 12.3-13 21.3v27h48V338.7c23.4-15.6 48-42.9 48-87.8c0-72.1-38.2-134.2-73.6-176.7c-8.3-9.9-16.6-19-24.4-27c10.3-2.7 18-12.1 18-23.2c0-13.3-10.7-24-24-24H160 104zM52.7 464l16.6-32H250.8l16.6 32H52.7zm207.9-80H59.5c-12 0-22.9 6.7-28.4 17.3L4.6 452.5c-3 5.8-4.6 12.2-4.6 18.7C0 493.8 18.2 512 40.8 512H279.2c22.5 0 40.8-18.2 40.8-40.8c0-6.5-1.6-12.9-4.6-18.7l-26.5-51.2c-5.5-10.6-16.5-17.3-28.4-17.3z"],"face-grin-wink":[512,512,["grin-wink"],"f58c","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm349.5 52.4c18.7-4.4 35.9 12 25.5 28.1C350.4 374.6 306.3 400 255.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm165.8 21.7c-7.6 8.1-20.2 8.5-28.3 .9s-8.5-20.2-.9-28.3c14.5-15.5 35.2-22.3 54.6-22.3s40.1 6.8 54.6 22.3c7.6 8.1 7.1 20.7-.9 28.3s-20.7 7.1-28.3-.9c-5.5-5.8-14.8-9.7-25.4-9.7s-19.9 3.8-25.4 9.7z"],"face-grin-wide":[512,512,[128515,"grin-alt"],"f581","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm349.5 52.4c18.7-4.4 35.9 12 25.5 28.1C350.4 374.6 306.3 400 255.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5zM224 192c0 35.3-14.3 64-32 64s-32-28.7-32-64s14.3-64 32-64s32 28.7 32 64zm96 64c-17.7 0-32-28.7-32-64s14.3-64 32-64s32 28.7 32 64s-14.3 64-32 64z"],"face-frown-open":[512,512,[128550,"frown-open"],"f57a","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM182.4 382.5c-12.4 5.2-26.5-4.1-21.1-16.4c16-36.6 52.4-62.1 94.8-62.1s78.8 25.6 94.8 62.1c5.4 12.3-8.7 21.6-21.1 16.4c-22.4-9.5-47.4-14.8-73.7-14.8s-51.3 5.3-73.7 14.8zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"hand-point-up":[384,512,[9757],"f0a6","M64 64V241.6c5.2-1 10.5-1.6 16-1.6H96V208 64c0-8.8-7.2-16-16-16s-16 7.2-16 16zM80 288c-17.7 0-32 14.3-32 32c0 0 0 0 0 0v24c0 66.3 53.7 120 120 120h48c52.5 0 97.1-33.7 113.4-80.7c-3.1 .5-6.2 .7-9.4 .7c-20 0-37.9-9.2-49.7-23.6c-9 4.9-19.4 7.6-30.3 7.6c-15.1 0-29-5.3-40-14c-11 8.8-24.9 14-40 14H120c-13.3 0-24-10.7-24-24s10.7-24 24-24h40c8.8 0 16-7.2 16-16s-7.2-16-16-16H120 80zM0 320s0 0 0 0c0-18 6-34.6 16-48V64C16 28.7 44.7 0 80 0s64 28.7 64 64v82c5.1-1.3 10.5-2 16-2c25.3 0 47.2 14.7 57.6 36c7-2.6 14.5-4 22.4-4c20 0 37.9 9.2 49.7 23.6c9-4.9 19.4-7.6 30.3-7.6c35.3 0 64 28.7 64 64v64 24c0 92.8-75.2 168-168 168H168C75.2 512 0 436.8 0 344V320zm336-64c0-8.8-7.2-16-16-16s-16 7.2-16 16v48 16c0 8.8 7.2 16 16 16s16-7.2 16-16V256zM160 240c5.5 0 10.9 .7 16 2v-2V208c0-8.8-7.2-16-16-16s-16 7.2-16 16v32h16zm64 24v40c0 8.8 7.2 16 16 16s16-7.2 16-16V256 240c0-8.8-7.2-16-16-16s-16 7.2-16 16v24z"],bookmark:[384,512,[128278,61591],"f02e","M0 48C0 21.5 21.5 0 48 0l0 48V441.4l130.1-92.9c8.3-6 19.6-6 27.9 0L336 441.4V48H48V0H336c26.5 0 48 21.5 48 48V488c0 9-5 17.2-13 21.3s-17.6 3.4-24.9-1.8L192 397.5 37.9 507.5c-7.3 5.2-16.9 5.9-24.9 1.8S0 497 0 488V48z"],"hand-point-down":[384,512,[],"f0a7","M64 448l0-177.6c5.2 1 10.5 1.6 16 1.6l16 0 0 32 0 144c0 8.8-7.2 16-16 16s-16-7.2-16-16zM80 224c-17.7 0-32-14.3-32-32c0 0 0 0 0 0l0-24c0-66.3 53.7-120 120-120l48 0c52.5 0 97.1 33.7 113.4 80.7c-3.1-.5-6.2-.7-9.4-.7c-20 0-37.9 9.2-49.7 23.6c-9-4.9-19.4-7.6-30.3-7.6c-15.1 0-29 5.3-40 14c-11-8.8-24.9-14-40-14l-40 0c-13.3 0-24 10.7-24 24s10.7 24 24 24l40 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-40 0-40 0zM0 192s0 0 0 0c0 18 6 34.6 16 48l0 208c0 35.3 28.7 64 64 64s64-28.7 64-64l0-82c5.1 1.3 10.5 2 16 2c25.3 0 47.2-14.7 57.6-36c7 2.6 14.5 4 22.4 4c20 0 37.9-9.2 49.7-23.6c9 4.9 19.4 7.6 30.3 7.6c35.3 0 64-28.7 64-64l0-64 0-24C384 75.2 308.8 0 216 0L168 0C75.2 0 0 75.2 0 168l0 24zm336 64c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-48 0-16c0-8.8 7.2-16 16-16s16 7.2 16 16l0 64zM160 272c5.5 0 10.9-.7 16-2l0 2 0 32c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-32 16 0zm64-24l0-40c0-8.8 7.2-16 16-16s16 7.2 16 16l0 48 0 16c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-24z"],folder:[512,512,[128193,128447,61716,"folder-blank"],"f07b","M0 96C0 60.7 28.7 32 64 32H196.1c19.1 0 37.4 7.6 50.9 21.1L289.9 96H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H448c8.8 0 16-7.2 16-16V160c0-8.8-7.2-16-16-16H286.6c-10.6 0-20.8-4.2-28.3-11.7L213.1 87c-4.5-4.5-10.6-7-17-7H64z"],user:[448,512,[128100,62144],"f007","M304 128a80 80 0 1 0 -160 0 80 80 0 1 0 160 0zM96 128a128 128 0 1 1 256 0A128 128 0 1 1 96 128zM49.3 464H398.7c-8.9-63.3-63.3-112-129-112H178.3c-65.7 0-120.1 48.7-129 112zM0 482.3C0 383.8 79.8 304 178.3 304h91.4C368.2 304 448 383.8 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3z"],"square-caret-left":[448,512,["caret-square-left"],"f191","M48 416c0 8.8 7.2 16 16 16l320 0c8.8 0 16-7.2 16-16l0-320c0-8.8-7.2-16-16-16L64 80c-8.8 0-16 7.2-16 16l0 320zm16 64c-35.3 0-64-28.7-64-64L0 96C0 60.7 28.7 32 64 32l320 0c35.3 0 64 28.7 64 64l0 320c0 35.3-28.7 64-64 64L64 480zm64-224c0-6.7 2.8-13 7.7-17.6l112-104c7-6.5 17.2-8.2 25.9-4.4s14.4 12.5 14.4 22l0 208c0 9.5-5.7 18.2-14.4 22s-18.9 2.1-25.9-4.4l-112-104c-4.9-4.5-7.7-10.9-7.7-17.6z"],star:[576,512,[11088,61446],"f005","M287.9 0c9.2 0 17.6 5.2 21.6 13.5l68.6 141.3 153.2 22.6c9 1.3 16.5 7.6 19.3 16.3s.5 18.1-5.9 24.5L433.6 328.4l26.2 155.6c1.5 9-2.2 18.1-9.7 23.5s-17.3 6-25.3 1.7l-137-73.2L151 509.1c-8.1 4.3-17.9 3.7-25.3-1.7s-11.2-14.5-9.7-23.5l26.2-155.6L31.1 218.2c-6.5-6.4-8.7-15.9-5.9-24.5s10.3-14.9 19.3-16.3l153.2-22.6L266.3 13.5C270.4 5.2 278.7 0 287.9 0zm0 79L235.4 187.2c-3.5 7.1-10.2 12.1-18.1 13.3L99 217.9 184.9 303c5.5 5.5 8.1 13.3 6.8 21L171.4 443.7l105.2-56.2c7.1-3.8 15.6-3.8 22.6 0l105.2 56.2L384.2 324.1c-1.3-7.7 1.2-15.5 6.8-21l85.9-85.1L358.6 200.5c-7.8-1.2-14.6-6.1-18.1-13.3L287.9 79z"],"chess-knight":[448,512,[9822],"f441","M226.6 48H117.3l17.1 12.8c6 4.5 9.6 11.6 9.6 19.2s-3.6 14.7-9.6 19.2l-6.5 4.9c-10 7.5-16 19.3-16 31.9l-.3 91c0 10.2 4.9 19.9 13.2 25.8l1.9 1.3c9.9 7.1 23.3 7 33.2-.1l49.9-36.3c10.7-7.8 25.7-5.4 33.5 5.3s5.4 25.7-5.3 33.5l-49.9 36.3-53.8 39.1c-7.3 5.3-13 12.2-16.9 20.1H66.8c5.3-22.1 17.8-41.9 35.9-56.3c-1.3-.8-2.6-1.7-3.8-2.6L97 291.8c-21-15-33.4-39.2-33.3-65l.3-91c.1-19.8 6.7-38.7 18.6-53.9l-.4-.3C70.7 73 64 59.6 64 45.3C64 20.3 84.3 0 109.3 0H226.6C331.2 0 416 84.8 416 189.4c0 11.1-1 22.2-2.9 33.2L390.1 352H341.3l24.5-137.8c1.5-8.2 2.2-16.5 2.2-24.8C368 111.3 304.7 48 226.6 48zM85.2 432L68.7 464H379.3l-16.6-32H85.2zm315.7-30.7l26.5 51.2c3 5.8 4.6 12.2 4.6 18.7c0 22.5-18.2 40.8-40.8 40.8H56.8C34.2 512 16 493.8 16 471.2c0-6.5 1.6-12.9 4.6-18.7l26.5-51.2C52.5 390.7 63.5 384 75.5 384h297c12 0 22.9 6.7 28.4 17.3zM172 128a20 20 0 1 1 0 40 20 20 0 1 1 0-40z"],"face-laugh-squint":[512,512,["laugh-squint"],"f59b","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm130.7 57.9c-4.2-13.6 7.1-25.9 21.3-25.9H364.5c14.2 0 25.5 12.4 21.3 25.9C369 368.4 318.2 408 258.2 408s-110.8-39.6-127.5-94.1zm2.8-183.3l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6c0-9 9.6-14.7 17.5-10.5zM396 141.1c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9c7.9-4.2 17.5 1.5 17.5 10.5z"],"face-laugh":[512,512,["laugh"],"f599","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm130.7 57.9c-4.2-13.6 7.1-25.9 21.3-25.9H364.5c14.2 0 25.5 12.4 21.3 25.9C369 368.4 318.2 408 258.2 408s-110.8-39.6-127.5-94.1zM144.4 192a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"folder-open":[576,512,[128194,128449,61717],"f07c","M384 480h48c11.4 0 21.9-6 27.6-15.9l112-192c5.8-9.9 5.8-22.1 .1-32.1S555.5 224 544 224H144c-11.4 0-21.9 6-27.6 15.9L48 357.1V96c0-8.8 7.2-16 16-16H181.5c4.2 0 8.3 1.7 11.3 4.7l26.5 26.5c21 21 49.5 32.8 79.2 32.8H416c8.8 0 16 7.2 16 16v32h48V160c0-35.3-28.7-64-64-64H298.5c-17 0-33.3-6.7-45.3-18.7L226.7 50.7c-12-12-28.3-18.7-45.3-18.7H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H87.7 384z"],clipboard:[384,512,[128203],"f328","M280 64h40c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128C0 92.7 28.7 64 64 64h40 9.6C121 27.5 153.3 0 192 0s71 27.5 78.4 64H280zM64 112c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16H320c8.8 0 16-7.2 16-16V128c0-8.8-7.2-16-16-16H304v24c0 13.3-10.7 24-24 24H192 104c-13.3 0-24-10.7-24-24V112H64zm128-8a24 24 0 1 0 0-48 24 24 0 1 0 0 48z"],"chess-queen":[512,512,[9819],"f445","M256 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-95.2-8c-18.1 0-31.3 12.8-35.6 26.9c-8 26.2-32.4 45.2-61.2 45.2c-10 0-19.4-2.3-27.7-6.3c-7.6-3.7-16.7-3.3-24 1.2C.7 162.1-3.1 177.1 3.7 188.9L97.6 352H153l-83-144.1c40.5-2.2 75.3-25.9 93.1-59.8c22 26.8 55.4 43.9 92.8 43.9s70.8-17.1 92.8-43.9c17.8 34 52.6 57.7 93.1 59.8L359 352h55.4l93.9-163.1c6.8-11.7 3-26.7-8.6-33.8c-7.3-4.5-16.4-4.9-24-1.2c-8.4 4-17.7 6.3-27.7 6.3c-28.8 0-53.2-19-61.2-45.2C382.5 100.8 369.3 88 351.2 88c-14.5 0-26.3 8.5-32.4 19.3c-12.4 22-35.9 36.7-62.8 36.7s-50.4-14.8-62.8-36.7C187.1 96.5 175.4 88 160.8 88zM133.2 432H378.8l16.6 32H116.7l16.6-32zm283.7-30.7c-5.5-10.6-16.5-17.3-28.4-17.3h-265c-12 0-22.9 6.7-28.4 17.3L68.6 452.5c-3 5.8-4.6 12.2-4.6 18.7c0 22.5 18.2 40.8 40.8 40.8H407.2c22.5 0 40.8-18.2 40.8-40.8c0-6.5-1.6-12.9-4.6-18.7l-26.5-51.2z"],"hand-back-fist":[448,512,["hand-rock"],"f255","M144 64c0-8.8 7.2-16 16-16s16 7.2 16 16c0 9.1 5.1 17.4 13.3 21.5s17.9 3.2 25.1-2.3c2.7-2 6-3.2 9.6-3.2c8.8 0 16 7.2 16 16c0 9.1 5.1 17.4 13.3 21.5s17.9 3.2 25.1-2.3c2.7-2 6-3.2 9.6-3.2c8.8 0 16 7.2 16 16c0 9.1 5.1 17.4 13.3 21.5s17.9 3.2 25.1-2.3c2.7-2 6-3.2 9.6-3.2c8.8 0 16 7.2 16 16V264c0 31.3-20 58-48 67.9c-9.6 3.4-16 12.5-16 22.6V488c0 13.3 10.7 24 24 24s24-10.7 24-24V370.2c38-20.1 64-60.1 64-106.2V160c0-35.3-28.7-64-64-64c-2.8 0-5.6 .2-8.3 .5C332.8 77.1 311.9 64 288 64c-2.8 0-5.6 .2-8.3 .5C268.8 45.1 247.9 32 224 32c-2.8 0-5.6 .2-8.3 .5C204.8 13.1 183.9 0 160 0C124.7 0 96 28.7 96 64v64.3c-11.7 7.4-22.5 16.4-32 26.9l17.8 16.1L64 155.2l-9.4 10.5C40 181.8 32 202.8 32 224.6v12.8c0 49.6 24.2 96.1 64.8 124.5l13.8-19.7L96.8 361.9l8.9 6.2c6.9 4.8 14.4 8.6 22.3 11.3V488c0 13.3 10.7 24 24 24s24-10.7 24-24V359.9c0-12.6-9.8-23.1-22.4-23.9c-7.3-.5-14.3-2.9-20.3-7.1l-13.1 18.7 13.1-18.7-8.9-6.2C96.6 303.1 80 271.3 80 237.4V224.6c0-9.9 3.7-19.4 10.3-26.8l9.4-10.5c3.8-4.2 7.9-8.1 12.3-11.6V208c0 8.8 7.2 16 16 16s16-7.2 16-16V142.3 128 64z"],"square-caret-up":[448,512,["caret-square-up"],"f151","M64 80c-8.8 0-16 7.2-16 16l0 320c0 8.8 7.2 16 16 16l320 0c8.8 0 16-7.2 16-16l0-320c0-8.8-7.2-16-16-16L64 80zM0 96C0 60.7 28.7 32 64 32l320 0c35.3 0 64 28.7 64 64l0 320c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64L0 96zm224 64c6.7 0 13 2.8 17.6 7.7l104 112c6.5 7 8.2 17.2 4.4 25.9s-12.5 14.4-22 14.4l-208 0c-9.5 0-18.2-5.7-22-14.4s-2.1-18.9 4.4-25.9l104-112c4.5-4.9 10.9-7.7 17.6-7.7z"],"chart-bar":[512,512,["bar-chart"],"f080","M24 32c13.3 0 24 10.7 24 24V408c0 13.3 10.7 24 24 24H488c13.3 0 24 10.7 24 24s-10.7 24-24 24H72c-39.8 0-72-32.2-72-72V56C0 42.7 10.7 32 24 32zM128 136c0-13.3 10.7-24 24-24l208 0c13.3 0 24 10.7 24 24s-10.7 24-24 24l-208 0c-13.3 0-24-10.7-24-24zm24 72H296c13.3 0 24 10.7 24 24s-10.7 24-24 24H152c-13.3 0-24-10.7-24-24s10.7-24 24-24zm0 96H424c13.3 0 24 10.7 24 24s-10.7 24-24 24H152c-13.3 0-24-10.7-24-24s10.7-24 24-24z"],"window-restore":[512,512,[],"f2d2","M432 48H208c-17.7 0-32 14.3-32 32V96H128V80c0-44.2 35.8-80 80-80H432c44.2 0 80 35.8 80 80V304c0 44.2-35.8 80-80 80H416V336h16c17.7 0 32-14.3 32-32V80c0-17.7-14.3-32-32-32zM48 448c0 8.8 7.2 16 16 16H320c8.8 0 16-7.2 16-16V256H48V448zM64 128H320c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V192c0-35.3 28.7-64 64-64z"],"square-plus":[448,512,[61846,"plus-square"],"f0fe","M64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V96c0-8.8-7.2-16-16-16H64zM0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM200 344V280H136c-13.3 0-24-10.7-24-24s10.7-24 24-24h64V168c0-13.3 10.7-24 24-24s24 10.7 24 24v64h64c13.3 0 24 10.7 24 24s-10.7 24-24 24H248v64c0 13.3-10.7 24-24 24s-24-10.7-24-24z"],image:[512,512,[],"f03e","M448 80c8.8 0 16 7.2 16 16V415.8l-5-6.5-136-176c-4.5-5.9-11.6-9.3-19-9.3s-14.4 3.4-19 9.3L202 340.7l-30.5-42.7C167 291.7 159.8 288 152 288s-15 3.7-19.5 10.1l-80 112L48 416.3l0-.3V96c0-8.8 7.2-16 16-16H448zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm80 192a48 48 0 1 0 0-96 48 48 0 1 0 0 96z"],"folder-closed":[512,512,[],"e185","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H289.9L247 53.1C233.5 39.6 215.2 32 196.1 32H64zM48 96c0-8.8 7.2-16 16-16H196.1c6.4 0 12.5 2.5 17 7l45.3 45.3c7.5 7.5 17.7 11.7 28.3 11.7H448c8.8 0 16 7.2 16 16v32H48V96zm0 144H464V416c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V240z"],lemon:[448,512,[127819],"f094","M368 80c-3.2 0-6.2 .4-8.9 1.3C340 86.8 313 91.9 284.8 84.6C227.4 69.7 160.2 92 110.1 142.1S37.7 259.4 52.6 316.8c7.3 28.2 2.2 55.2-3.3 74.3c-.8 2.8-1.3 5.8-1.3 8.9c0 17.7 14.3 32 32 32c3.2 0 6.2-.4 8.9-1.3c19.1-5.5 46.1-10.7 74.3-3.3c57.4 14.9 124.6-7.4 174.7-57.5s72.4-117.3 57.5-174.7c-7.3-28.2-2.2-55.2 3.3-74.3c.8-2.8 1.3-5.8 1.3-8.9c0-17.7-14.3-32-32-32zm0-48c44.2 0 80 35.8 80 80c0 7.7-1.1 15.2-3.1 22.3c-4.6 15.8-7.1 32.9-3 48.9c20.1 77.6-10.9 161.5-70 220.7s-143.1 90.2-220.7 70c-16-4.1-33-1.6-48.9 3c-7.1 2-14.6 3.1-22.3 3.1c-44.2 0-80-35.8-80-80c0-7.7 1.1-15.2 3.1-22.3c4.6-15.8 7.1-32.9 3-48.9C-14 251.3 17 167.3 76.2 108.2S219.3 18 296.8 38.1c16 4.1 33 1.6 48.9-3c7.1-2 14.6-3.1 22.3-3.1zM246.7 167c-52 15.2-96.5 59.7-111.7 111.7c-3.7 12.7-17.1 20-29.8 16.3s-20-17.1-16.3-29.8c19.8-67.7 76.6-124.5 144.3-144.3c12.7-3.7 26.1 3.6 29.8 16.3s-3.6 26.1-16.3 29.8z"],handshake:[640,512,[],"f2b5","M272.2 64.6l-51.1 51.1c-15.3 4.2-29.5 11.9-41.5 22.5L153 161.9C142.8 171 129.5 176 115.8 176H96V304c20.4 .6 39.8 8.9 54.3 23.4l35.6 35.6 7 7 0 0L219.9 397c6.2 6.2 16.4 6.2 22.6 0c1.7-1.7 3-3.7 3.7-5.8c2.8-7.7 9.3-13.5 17.3-15.3s16.4 .6 22.2 6.5L296.5 393c11.6 11.6 30.4 11.6 41.9 0c5.4-5.4 8.3-12.3 8.6-19.4c.4-8.8 5.6-16.6 13.6-20.4s17.3-3 24.4 2.1c9.4 6.7 22.5 5.8 30.9-2.6c9.4-9.4 9.4-24.6 0-33.9L340.1 243l-35.8 33c-27.3 25.2-69.2 25.6-97 .9c-31.7-28.2-32.4-77.4-1.6-106.5l70.1-66.2C303.2 78.4 339.4 64 377.1 64c36.1 0 71 13.3 97.9 37.2L505.1 128H544h40 40c8.8 0 16 7.2 16 16V352c0 17.7-14.3 32-32 32H576c-11.8 0-22.2-6.4-27.7-16H463.4c-3.4 6.7-7.9 13.1-13.5 18.7c-17.1 17.1-40.8 23.8-63 20.1c-3.6 7.3-8.5 14.1-14.6 20.2c-27.3 27.3-70 30-100.4 8.1c-25.1 20.8-62.5 19.5-86-4.1L159 404l-7-7-35.6-35.6c-5.5-5.5-12.7-8.7-20.4-9.3C96 369.7 81.6 384 64 384H32c-17.7 0-32-14.3-32-32V144c0-8.8 7.2-16 16-16H56 96h19.8c2 0 3.9-.7 5.3-2l26.5-23.6C175.5 77.7 211.4 64 248.7 64H259c4.4 0 8.9 .2 13.2 .6zM544 320V176H496c-5.9 0-11.6-2.2-15.9-6.1l-36.9-32.8c-18.2-16.2-41.7-25.1-66.1-25.1c-25.4 0-49.8 9.7-68.3 27.1l-70.1 66.2c-10.3 9.8-10.1 26.3 .5 35.7c9.3 8.3 23.4 8.1 32.5-.3l71.9-66.4c9.7-9 24.9-8.4 33.9 1.4s8.4 24.9-1.4 33.9l-.8 .8 74.4 74.4c10 10 16.5 22.3 19.4 35.1H544zM64 336a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm528 16a16 16 0 1 0 0-32 16 16 0 1 0 0 32z"],gem:[512,512,[128142],"f3a5","M168.5 72L256 165l87.5-93h-175zM383.9 99.1L311.5 176h129L383.9 99.1zm50 124.9H256 78.1L256 420.3 433.9 224zM71.5 176h129L128.1 99.1 71.5 176zm434.3 40.1l-232 256c-4.5 5-11 7.9-17.8 7.9s-13.2-2.9-17.8-7.9l-232-256c-7.7-8.5-8.3-21.2-1.5-30.4l112-152c4.5-6.1 11.7-9.8 19.3-9.8H376c7.6 0 14.8 3.6 19.3 9.8l112 152c6.8 9.2 6.1 21.9-1.5 30.4z"],"circle-play":[512,512,[61469,"play-circle"],"f144","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM188.3 147.1c7.6-4.2 16.8-4.1 24.3 .5l144 88c7.1 4.4 11.5 12.1 11.5 20.5s-4.4 16.1-11.5 20.5l-144 88c-7.4 4.5-16.7 4.7-24.3 .5s-12.3-12.2-12.3-20.9V168c0-8.7 4.7-16.7 12.3-20.9z"],"circle-check":[512,512,[61533,"check-circle"],"f058","M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM369 209c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-111 111-47-47c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l64 64c9.4 9.4 24.6 9.4 33.9 0L369 209z"],"circle-stop":[512,512,[62094,"stop-circle"],"f28d","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm192-96H320c17.7 0 32 14.3 32 32V320c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32V192c0-17.7 14.3-32 32-32z"],"id-badge":[384,512,[],"f2c1","M256 48V64c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32V48H64c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16H320c8.8 0 16-7.2 16-16V64c0-8.8-7.2-16-16-16H256zM0 64C0 28.7 28.7 0 64 0H320c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM160 320h64c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80zm-32-96a64 64 0 1 1 128 0 64 64 0 1 1 -128 0z"],"face-laugh-beam":[512,512,[128513,"laugh-beam"],"f59a","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm130.7 57.9c-4.2-13.6 7.1-25.9 21.3-25.9H364.5c14.2 0 25.5 12.4 21.3 25.9C369 368.4 318.2 408 258.2 408s-110.8-39.6-127.5-94.1zm86.9-85.1l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z"],registered:[512,512,[174],"f25d","M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM160 152V264v96c0 13.3 10.7 24 24 24s24-10.7 24-24V288h60.9l37.2 81.9c5.5 12.1 19.7 17.4 31.8 11.9s17.4-19.7 11.9-31.8L315.7 275c21.8-14.3 36.3-39 36.3-67c0-44.2-35.8-80-80-80H184c-13.3 0-24 10.7-24 24zm48 88V176h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H208z"],"address-card":[576,512,[62140,"contact-card","vcard"],"f2bb","M512 80c8.8 0 16 7.2 16 16V416c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V96c0-8.8 7.2-16 16-16H512zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM208 256a64 64 0 1 0 0-128 64 64 0 1 0 0 128zm-32 32c-44.2 0-80 35.8-80 80c0 8.8 7.2 16 16 16H304c8.8 0 16-7.2 16-16c0-44.2-35.8-80-80-80H176zM376 144c-13.3 0-24 10.7-24 24s10.7 24 24 24h80c13.3 0 24-10.7 24-24s-10.7-24-24-24H376zm0 96c-13.3 0-24 10.7-24 24s10.7 24 24 24h80c13.3 0 24-10.7 24-24s-10.7-24-24-24H376z"],"face-tired":[512,512,[128555,"tired"],"f5c8","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm176.5 64.3C196.1 302.1 223.8 288 256 288s59.9 14.1 79.5 32.3C354.5 338.1 368 362 368 384c0 5.4-2.7 10.4-7.2 13.4s-10.2 3.4-15.2 1.3l-17.2-7.5c-22.8-10-47.5-15.1-72.4-15.1s-49.6 5.2-72.4 15.1l-17.2 7.5c-4.9 2.2-10.7 1.7-15.2-1.3s-7.2-8-7.2-13.4c0-22 13.5-45.9 32.5-63.7zm-43-173.6l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6c0-9 9.6-14.7 17.5-10.5zM396 157.1c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9c7.9-4.2 17.5 1.5 17.5 10.5z"],"font-awesome":[512,512,[62501,62694,"font-awesome-flag","font-awesome-logo-full"],"f2b4","M91.7 96C106.3 86.8 116 70.5 116 52C116 23.3 92.7 0 64 0S12 23.3 12 52c0 16.7 7.8 31.5 20 41l0 3 0 48 0 256 0 48 0 64 48 0 0-64 389.6 0c14.6 0 26.4-11.8 26.4-26.4c0-3.7-.8-7.3-2.3-10.7L432 272l61.7-138.9c1.5-3.4 2.3-7 2.3-10.7c0-14.6-11.8-26.4-26.4-26.4L91.7 96zM80 400l0-256 356.4 0L388.1 252.5c-5.5 12.4-5.5 26.6 0 39L436.4 400 80 400z"],"face-smile-wink":[512,512,[128521,"smile-wink"],"f4da","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm177.6 62.1C192.8 334.5 218.8 352 256 352s63.2-17.5 78.4-33.9c9-9.7 24.2-10.4 33.9-1.4s10.4 24.2 1.4 33.9c-22 23.8-60 49.4-113.6 49.4s-91.7-25.5-113.6-49.4c-9-9.7-8.4-24.9 1.4-33.9s24.9-8.4 33.9 1.4zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm165.8 21.7c-7.6 8.1-20.2 8.5-28.3 .9s-8.5-20.2-.9-28.3c14.5-15.5 35.2-22.3 54.6-22.3s40.1 6.8 54.6 22.3c7.6 8.1 7.1 20.7-.9 28.3s-20.7 7.1-28.3-.9c-5.5-5.8-14.8-9.7-25.4-9.7s-19.9 3.8-25.4 9.7z"],"file-word":[384,512,[],"f1c2","M48 448V64c0-8.8 7.2-16 16-16H224v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm55 241.1c-3.8-12.7-17.2-19.9-29.9-16.1s-19.9 17.2-16.1 29.9l48 160c3 10.2 12.4 17.1 23 17.1s19.9-7 23-17.1l25-83.4 25 83.4c3 10.2 12.4 17.1 23 17.1s19.9-7 23-17.1l48-160c3.8-12.7-3.4-26.1-16.1-29.9s-26.1 3.4-29.9 16.1l-25 83.4-25-83.4c-3-10.2-12.4-17.1-23-17.1s-19.9 7-23 17.1l-25 83.4-25-83.4z"],"file-powerpoint":[384,512,[],"f1c4","M64 464c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16H224v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm72 208c-13.3 0-24 10.7-24 24V336v56c0 13.3 10.7 24 24 24s24-10.7 24-24V360h44c42 0 76-34 76-76s-34-76-76-76H136zm68 104H160V256h44c15.5 0 28 12.5 28 28s-12.5 28-28 28z"],"envelope-open":[512,512,[62135],"f2b6","M255.4 48.2c.2-.1 .4-.2 .6-.2s.4 .1 .6 .2L460.6 194c2.1 1.5 3.4 3.9 3.4 6.5v13.6L291.5 355.7c-20.7 17-50.4 17-71.1 0L48 214.1V200.5c0-2.6 1.2-5 3.4-6.5L255.4 48.2zM48 276.2L190 392.8c38.4 31.5 93.7 31.5 132 0L464 276.2V456c0 4.4-3.6 8-8 8H56c-4.4 0-8-3.6-8-8V276.2zM256 0c-10.2 0-20.2 3.2-28.5 9.1L23.5 154.9C8.7 165.4 0 182.4 0 200.5V456c0 30.9 25.1 56 56 56H456c30.9 0 56-25.1 56-56V200.5c0-18.1-8.7-35.1-23.4-45.6L284.5 9.1C276.2 3.2 266.2 0 256 0z"],"file-zipper":[384,512,["file-archive"],"f1c6","M64 464c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16h48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16h48v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm48 112c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H128c-8.8 0-16 7.2-16 16zm0 64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H128c-8.8 0-16 7.2-16 16zm-6.3 71.8L82.1 335.9c-1.4 5.4-2.1 10.9-2.1 16.4c0 35.2 28.8 63.7 64 63.7s64-28.5 64-63.7c0-5.5-.7-11.1-2.1-16.4l-23.5-88.2c-3.7-14-16.4-23.8-30.9-23.8H136.6c-14.5 0-27.2 9.7-30.9 23.8zM128 336h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H128c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],square:[448,512,[9632,9723,9724,61590],"f0c8","M384 80c8.8 0 16 7.2 16 16V416c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V96c0-8.8 7.2-16 16-16H384zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64z"],snowflake:[448,512,[10052,10054],"f2dc","M224 0c13.3 0 24 10.7 24 24V70.1l23-23c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-57 57v76.5l66.2-38.2 20.9-77.8c3.4-12.8 16.6-20.4 29.4-17s20.4 16.6 17 29.4L373 142.2l37.1-21.4c11.5-6.6 26.2-2.7 32.8 8.8s2.7 26.2-8.8 32.8L397 183.8l31.5 8.4c12.8 3.4 20.4 16.6 17 29.4s-16.6 20.4-29.4 17l-77.8-20.9L272 256l66.2 38.2 77.8-20.9c12.8-3.4 26 4.2 29.4 17s-4.2 26-17 29.4L397 328.2l37.1 21.4c11.5 6.6 15.4 21.3 8.8 32.8s-21.3 15.4-32.8 8.8L373 369.8l8.4 31.5c3.4 12.8-4.2 26-17 29.4s-26-4.2-29.4-17l-20.9-77.8L248 297.6v76.5l57 57c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-23-23V488c0 13.3-10.7 24-24 24s-24-10.7-24-24V441.9l-23 23c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l57-57V297.6l-66.2 38.2-20.9 77.8c-3.4 12.8-16.6 20.4-29.4 17s-20.4-16.6-17-29.4L75 369.8 37.9 391.2c-11.5 6.6-26.2 2.7-32.8-8.8s-2.7-26.2 8.8-32.8L51 328.2l-31.5-8.4c-12.8-3.4-20.4-16.6-17-29.4s16.6-20.4 29.4-17l77.8 20.9L176 256l-66.2-38.2L31.9 238.6c-12.8 3.4-26-4.2-29.4-17s4.2-26 17-29.4L51 183.8 13.9 162.4c-11.5-6.6-15.4-21.3-8.8-32.8s21.3-15.4 32.8-8.8L75 142.2l-8.4-31.5c-3.4-12.8 4.2-26 17-29.4s26 4.2 29.4 17l20.9 77.8L200 214.4V137.9L143 81c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l23 23V24c0-13.3 10.7-24 24-24z"],newspaper:[512,512,[128240],"f1ea","M168 80c-13.3 0-24 10.7-24 24V408c0 8.4-1.4 16.5-4.1 24H440c13.3 0 24-10.7 24-24V104c0-13.3-10.7-24-24-24H168zM72 480c-39.8 0-72-32.2-72-72V112C0 98.7 10.7 88 24 88s24 10.7 24 24V408c0 13.3 10.7 24 24 24s24-10.7 24-24V104c0-39.8 32.2-72 72-72H440c39.8 0 72 32.2 72 72V408c0 39.8-32.2 72-72 72H72zM176 136c0-13.3 10.7-24 24-24h96c13.3 0 24 10.7 24 24v80c0 13.3-10.7 24-24 24H200c-13.3 0-24-10.7-24-24V136zm200-24h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H376c-13.3 0-24-10.7-24-24s10.7-24 24-24zm0 80h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H376c-13.3 0-24-10.7-24-24s10.7-24 24-24zM200 272H408c13.3 0 24 10.7 24 24s-10.7 24-24 24H200c-13.3 0-24-10.7-24-24s10.7-24 24-24zm0 80H408c13.3 0 24 10.7 24 24s-10.7 24-24 24H200c-13.3 0-24-10.7-24-24s10.7-24 24-24z"],"face-kiss-wink-heart":[512,512,[128536,"kiss-wink-heart"],"f598","M338.9 446.8c-25.4 11-53.4 17.2-82.9 17.2C141.1 464 48 370.9 48 256S141.1 48 256 48s208 93.1 208 208c0 22.4-3.5 43.9-10.1 64.1c3.1 4.5 5.7 9.4 7.8 14.6c12.7-1.6 25.1 .4 36.2 5c9.1-26.2 14-54.4 14-83.7C512 114.6 397.4 0 256 0S0 114.6 0 256S114.6 512 256 512c35.4 0 69.1-7.2 99.7-20.2c-4.8-5.5-8.5-12.2-10.4-19.7l-6.5-25.3zM296 316c0-6.9-3.1-13.2-7.3-18.3c-4.3-5.2-10.1-9.7-16.7-13.4C258.7 276.9 241.4 272 224 272c-3.6 0-6.8 2.5-7.7 6s.6 7.2 3.8 9l0 0 0 0 0 0 .2 .1c.2 .1 .5 .3 .9 .5c.8 .5 2 1.2 3.4 2.1c2.8 1.9 6.5 4.5 10.2 7.6c3.7 3.1 7.2 6.6 9.6 10.1c2.5 3.5 3.5 6.4 3.5 8.6s-1 5-3.5 8.6c-2.5 3.5-5.9 6.9-9.6 10.1c-3.7 3.1-7.4 5.7-10.2 7.6c-1.4 .9-2.6 1.6-3.4 2.1c-.4 .2-.7 .4-.9 .5l-.2 .1 0 0 0 0 0 0 0 0 0 0c-2.5 1.4-4.1 4.1-4.1 7s1.6 5.6 4.1 7l0 0 0 0 0 0 .2 .1c.2 .1 .5 .3 .9 .5c.8 .5 2 1.2 3.4 2.1c2.8 1.9 6.5 4.5 10.2 7.6c3.7 3.1 7.2 6.6 9.6 10.1c2.5 3.5 3.5 6.4 3.5 8.6s-1 5-3.5 8.6c-2.5 3.5-5.9 6.9-9.6 10.1c-3.7 3.1-7.4 5.7-10.2 7.6c-1.4 .9-2.6 1.6-3.4 2.1c-.4 .2-.7 .4-.9 .5l-.2 .1 0 0 0 0 0 0 0 0c-3.2 1.8-4.7 5.5-3.8 9s4.1 6 7.7 6c17.4 0 34.7-4.9 47.9-12.3c6.6-3.7 12.5-8.2 16.7-13.4c4.3-5.1 7.3-11.4 7.3-18.3s-3.1-13.2-7.3-18.3c-4.3-5.2-10.1-9.7-16.7-13.4c-2.7-1.5-5.7-3-8.7-4.3c3.1-1.3 6-2.7 8.7-4.3c6.6-3.7 12.5-8.2 16.7-13.4c4.3-5.1 7.3-11.4 7.3-18.3zM176.4 240a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm159.3-20c10.6 0 19.9 3.8 25.4 9.7c7.6 8.1 20.2 8.5 28.3 .9s8.5-20.2 .9-28.3C375.7 186.8 355 180 335.6 180s-40.1 6.8-54.6 22.3c-7.6 8.1-7.1 20.7 .9 28.3s20.7 7.1 28.3-.9c5.5-5.8 14.8-9.7 25.4-9.7zM434 352.3c-6-23.2-28.8-37-51.1-30.8s-35.4 30.1-29.5 53.4l22.9 89.3c2.2 8.7 11.2 13.9 19.8 11.4l84.9-23.8c22.2-6.2 35.4-30.1 29.5-53.4s-28.8-37-51.1-30.8l-20.2 5.6-5.4-21z"],"star-half-stroke":[576,512,["star-half-alt"],"f5c0","M309.5 13.5C305.5 5.2 297.1 0 287.9 0s-17.6 5.2-21.6 13.5L197.7 154.8 44.5 177.5c-9 1.3-16.5 7.6-19.3 16.3s-.5 18.1 5.9 24.5L142.2 328.4 116 483.9c-1.5 9 2.2 18.1 9.7 23.5s17.3 6 25.3 1.7l137-73.2 137 73.2c8.1 4.3 17.9 3.7 25.3-1.7s11.2-14.5 9.7-23.5L433.6 328.4 544.8 218.2c6.5-6.4 8.7-15.9 5.9-24.5s-10.3-14.9-19.3-16.3L378.1 154.8 309.5 13.5zM288 384.7V79.1l52.5 108.1c3.5 7.1 10.2 12.1 18.1 13.3l118.3 17.5L391 303c-5.5 5.5-8.1 13.3-6.8 21l20.2 119.6L299.2 387.5c-3.5-1.9-7.4-2.8-11.2-2.8z"],"file-excel":[384,512,[],"f1c3","M48 448V64c0-8.8 7.2-16 16-16H224v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm90.9 233.3c-8.1-10.5-23.2-12.3-33.7-4.2s-12.3 23.2-4.2 33.7L161.6 320l-44.5 57.3c-8.1 10.5-6.3 25.5 4.2 33.7s25.5 6.3 33.7-4.2L192 359.1l37.1 47.6c8.1 10.5 23.2 12.3 33.7 4.2s12.3-23.2 4.2-33.7L222.4 320l44.5-57.3c8.1-10.5 6.3-25.5-4.2-33.7s-25.5-6.3-33.7 4.2L192 280.9l-37.1-47.6z"],"face-grin-beam":[512,512,[128516,"grin-beam"],"f582","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm349.5 52.4c18.7-4.4 35.9 12 25.5 28.1C350.4 374.6 306.3 400 255.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5zM217.6 228.8l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z"],"object-ungroup":[640,512,[],"f248","M48.2 66.8c-.1-.8-.2-1.7-.2-2.5c0-.1 0-.1 0-.2c0-8.8 7.2-16 16-16c.9 0 1.9 .1 2.8 .2C74.3 49.5 80 56.1 80 64c0 8.8-7.2 16-16 16c-7.9 0-14.5-5.7-15.8-13.2zM0 64c0 26.9 16.5 49.9 40 59.3V228.7C16.5 238.1 0 261.1 0 288c0 35.3 28.7 64 64 64c26.9 0 49.9-16.5 59.3-40H324.7c9.5 23.5 32.5 40 59.3 40c35.3 0 64-28.7 64-64c0-26.9-16.5-49.9-40-59.3V123.3c23.5-9.5 40-32.5 40-59.3c0-35.3-28.7-64-64-64c-26.9 0-49.9 16.5-59.3 40H123.3C113.9 16.5 90.9 0 64 0C28.7 0 0 28.7 0 64zm368 0a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zM324.7 88c6.5 16 19.3 28.9 35.3 35.3V228.7c-16 6.5-28.9 19.3-35.3 35.3H123.3c-6.5-16-19.3-28.9-35.3-35.3V123.3c16-6.5 28.9-19.3 35.3-35.3H324.7zM384 272a16 16 0 1 1 0 32 16 16 0 1 1 0-32zM80 288c0 7.9-5.7 14.5-13.2 15.8c-.8 .1-1.7 .2-2.5 .2l-.2 0c-8.8 0-16-7.2-16-16c0-.9 .1-1.9 .2-2.8C49.5 277.7 56.1 272 64 272c8.8 0 16 7.2 16 16zm391.3-40h45.4c6.5 16 19.3 28.9 35.3 35.3V388.7c-16 6.5-28.9 19.3-35.3 35.3H315.3c-6.5-16-19.3-28.9-35.3-35.3V352H232v36.7c-23.5 9.5-40 32.5-40 59.3c0 35.3 28.7 64 64 64c26.9 0 49.9-16.5 59.3-40H516.7c9.5 23.5 32.5 40 59.3 40c35.3 0 64-28.7 64-64c0-26.9-16.5-49.9-40-59.3V283.3c23.5-9.5 40-32.5 40-59.3c0-35.3-28.7-64-64-64c-26.9 0-49.9 16.5-59.3 40H448v16.4c9.8 8.8 17.8 19.5 23.3 31.6zm88.9-26.7a16 16 0 1 1 31.5 5.5 16 16 0 1 1 -31.5-5.5zM271.8 450.7a16 16 0 1 1 -31.5-5.5 16 16 0 1 1 31.5 5.5zm301.5 13c-7.5-1.3-13.2-7.9-13.2-15.8c0-8.8 7.2-16 16-16c7.9 0 14.5 5.7 15.8 13.2l0 .1c.1 .9 .2 1.8 .2 2.7c0 8.8-7.2 16-16 16c-.9 0-1.9-.1-2.8-.2z"],"circle-right":[512,512,[61838,"arrow-alt-circle-right"],"f35a","M464 256A208 208 0 1 1 48 256a208 208 0 1 1 416 0zM0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zM294.6 135.1c-4.2-4.5-10.1-7.1-16.3-7.1C266 128 256 138 256 150.3V208H160c-17.7 0-32 14.3-32 32v32c0 17.7 14.3 32 32 32h96v57.7c0 12.3 10 22.3 22.3 22.3c6.2 0 12.1-2.6 16.3-7.1l99.9-107.1c3.5-3.8 5.5-8.7 5.5-13.8s-2-10.1-5.5-13.8L294.6 135.1z"],"face-rolling-eyes":[512,512,[128580,"meh-rolling-eyes"],"f5a5","M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM168 376c0 13.3 10.7 24 24 24H320c13.3 0 24-10.7 24-24s-10.7-24-24-24H192c-13.3 0-24 10.7-24 24zm-8-104c-26.5 0-48-21.5-48-48c0-14.3 6.3-27.2 16.2-36c-.2 1.3-.2 2.6-.2 4c0 17.7 14.3 32 32 32s32-14.3 32-32c0-1.4-.1-2.7-.2-4c10 8.8 16.2 21.7 16.2 36c0 26.5-21.5 48-48 48zm0 32a80 80 0 1 0 0-160 80 80 0 1 0 0 160zm192-32c-26.5 0-48-21.5-48-48c0-14.3 6.3-27.2 16.2-36c-.2 1.3-.2 2.6-.2 4c0 17.7 14.3 32 32 32s32-14.3 32-32c0-1.4-.1-2.7-.2-4c10 8.8 16.2 21.7 16.2 36c0 26.5-21.5 48-48 48zm0 32a80 80 0 1 0 0-160 80 80 0 1 0 0 160z"],"object-group":[576,512,[],"f247","M48 115.8C38.2 107 32 94.2 32 80c0-26.5 21.5-48 48-48c14.2 0 27 6.2 35.8 16H460.2c8.8-9.8 21.6-16 35.8-16c26.5 0 48 21.5 48 48c0 14.2-6.2 27-16 35.8V396.2c9.8 8.8 16 21.6 16 35.8c0 26.5-21.5 48-48 48c-14.2 0-27-6.2-35.8-16H115.8c-8.8 9.8-21.6 16-35.8 16c-26.5 0-48-21.5-48-48c0-14.2 6.2-27 16-35.8V115.8zM125.3 96c-4.8 13.6-15.6 24.4-29.3 29.3V386.7c13.6 4.8 24.4 15.6 29.3 29.3H450.7c4.8-13.6 15.6-24.4 29.3-29.3V125.3c-13.6-4.8-24.4-15.6-29.3-29.3H125.3zm2.7 64c0-17.7 14.3-32 32-32H288c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32V160zM256 320h32c35.3 0 64-28.7 64-64V224h64c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H288c-17.7 0-32-14.3-32-32V320z"],heart:[512,512,[128153,128154,128155,128156,128420,129293,129294,129505,9829,10084,61578],"f004","M225.8 468.2l-2.5-2.3L48.1 303.2C17.4 274.7 0 234.7 0 192.8v-3.3c0-70.4 50-130.8 119.2-144C158.6 37.9 198.9 47 231 69.6c9 6.4 17.4 13.8 25 22.3c4.2-4.8 8.7-9.2 13.5-13.3c3.7-3.2 7.5-6.2 11.5-9c0 0 0 0 0 0C313.1 47 353.4 37.9 392.8 45.4C462 58.6 512 119.1 512 189.5v3.3c0 41.9-17.4 81.9-48.1 110.4L288.7 465.9l-2.5 2.3c-8.2 7.6-19 11.9-30.2 11.9s-22-4.2-30.2-11.9zM239.1 145c-.4-.3-.7-.7-1-1.1l-17.8-20c0 0-.1-.1-.1-.1c0 0 0 0 0 0c-23.1-25.9-58-37.7-92-31.2C81.6 101.5 48 142.1 48 189.5v3.3c0 28.5 11.9 55.8 32.8 75.2L256 430.7 431.2 268c20.9-19.4 32.8-46.7 32.8-75.2v-3.3c0-47.3-33.6-88-80.1-96.9c-34-6.5-69 5.4-92 31.2c0 0 0 0-.1 .1s0 0-.1 .1l-17.8 20c-.3 .4-.7 .7-1 1.1c-4.5 4.5-10.6 7-16.9 7s-12.4-2.5-16.9-7z"],"face-surprise":[512,512,[128558,"surprise"],"f5c2","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm176.4-80a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM256 288a64 64 0 1 1 0 128 64 64 0 1 1 0-128z"],"circle-pause":[512,512,[62092,"pause-circle"],"f28b","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm224-72V328c0 13.3-10.7 24-24 24s-24-10.7-24-24V184c0-13.3 10.7-24 24-24s24 10.7 24 24zm112 0V328c0 13.3-10.7 24-24 24s-24-10.7-24-24V184c0-13.3 10.7-24 24-24s24 10.7 24 24z"],circle:[512,512,[128308,128309,128992,128993,128994,128995,128996,9679,9898,9899,11044,61708,61915],"f111","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256z"],"circle-up":[512,512,[61467,"arrow-alt-circle-up"],"f35b","M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM135.1 217.4c-4.5 4.2-7.1 10.1-7.1 16.3c0 12.3 10 22.3 22.3 22.3H208v96c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V256h57.7c12.3 0 22.3-10 22.3-22.3c0-6.2-2.6-12.1-7.1-16.3L269.8 117.5c-3.8-3.5-8.7-5.5-13.8-5.5s-10.1 2-13.8 5.5L135.1 217.4z"],"file-audio":[384,512,[],"f1c7","M64 464H320c8.8 0 16-7.2 16-16V160H256c-17.7 0-32-14.3-32-32V48H64c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16zM0 64C0 28.7 28.7 0 64 0H229.5c17 0 33.3 6.7 45.3 18.7l90.5 90.5c12 12 18.7 28.3 18.7 45.3V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM192 272V400c0 6.5-3.9 12.3-9.9 14.8s-12.9 1.1-17.4-3.5L129.4 376H112c-8.8 0-16-7.2-16-16V312c0-8.8 7.2-16 16-16h17.4l35.3-35.3c4.6-4.6 11.5-5.9 17.4-3.5s9.9 8.3 9.9 14.8zm85.8-4c11.6 20 18.2 43.3 18.2 68s-6.6 48-18.2 68c-6.6 11.5-21.3 15.4-32.8 8.8s-15.4-21.3-8.8-32.8c7.5-12.9 11.8-27.9 11.8-44s-4.3-31.1-11.8-44c-6.6-11.5-2.7-26.2 8.8-32.8s26.2-2.7 32.8 8.8z"],"file-image":[384,512,[128443],"f1c5","M64 464c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16H224v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm96 256a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm69.2 46.9c-3-4.3-7.9-6.9-13.2-6.9s-10.2 2.6-13.2 6.9l-41.3 59.7-11.9-19.1c-2.9-4.7-8.1-7.5-13.6-7.5s-10.6 2.8-13.6 7.5l-40 64c-3.1 4.9-3.2 11.1-.4 16.2s8.2 8.2 14 8.2h48 32 40 72c6 0 11.4-3.3 14.2-8.6s2.4-11.6-1-16.5l-72-104z"],"circle-question":[512,512,[62108,"question-circle"],"f059","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm169.8-90.7c7.9-22.3 29.1-37.3 52.8-37.3h58.3c34.9 0 63.1 28.3 63.1 63.1c0 22.6-12.1 43.5-31.7 54.8L280 264.4c-.2 13-10.9 23.6-24 23.6c-13.3 0-24-10.7-24-24V250.5c0-8.6 4.6-16.5 12.1-20.8l44.3-25.4c4.7-2.7 7.6-7.7 7.6-13.1c0-8.4-6.8-15.1-15.1-15.1H222.6c-3.4 0-6.4 2.1-7.5 5.3l-.4 1.2c-4.4 12.5-18.2 19-30.6 14.6s-19-18.2-14.6-30.6l.4-1.2zM224 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],"face-meh-blank":[512,512,[128566,"meh-blank"],"f5a4","M256 48a208 208 0 1 0 0 416 208 208 0 1 0 0-416zM512 256A256 256 0 1 1 0 256a256 256 0 1 1 512 0zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],eye:[576,512,[128065],"f06e","M288 80c-65.2 0-118.8 29.6-159.9 67.7C89.6 183.5 63 226 49.4 256c13.6 30 40.2 72.5 78.6 108.3C169.2 402.4 222.8 432 288 432s118.8-29.6 159.9-67.7C486.4 328.5 513 286 526.6 256c-13.6-30-40.2-72.5-78.6-108.3C406.8 109.6 353.2 80 288 80zM95.4 112.6C142.5 68.8 207.2 32 288 32s145.5 36.8 192.6 80.6c46.8 43.5 78.1 95.4 93 131.1c3.3 7.9 3.3 16.7 0 24.6c-14.9 35.7-46.2 87.7-93 131.1C433.5 443.2 368.8 480 288 480s-145.5-36.8-192.6-80.6C48.6 356 17.3 304 2.5 268.3c-3.3-7.9-3.3-16.7 0-24.6C17.3 208 48.6 156 95.4 112.6zM288 336c44.2 0 80-35.8 80-80s-35.8-80-80-80c-.7 0-1.3 0-2 0c1.3 5.1 2 10.5 2 16c0 35.3-28.7 64-64 64c-5.5 0-10.9-.7-16-2c0 .7 0 1.3 0 2c0 44.2 35.8 80 80 80zm0-208a128 128 0 1 1 0 256 128 128 0 1 1 0-256z"],"face-sad-cry":[512,512,[128557,"sad-cry"],"f5b3","M400 406.1V288c0-13.3-10.7-24-24-24s-24 10.7-24 24V440.6c-28.7 15-61.4 23.4-96 23.4s-67.3-8.5-96-23.4V288c0-13.3-10.7-24-24-24s-24 10.7-24 24V406.1C72.6 368.2 48 315 48 256C48 141.1 141.1 48 256 48s208 93.1 208 208c0 59-24.6 112.2-64 150.1zM256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM159.6 220c10.6 0 19.9 3.8 25.4 9.7c7.6 8.1 20.2 8.5 28.3 .9s8.5-20.2 .9-28.3C199.7 186.8 179 180 159.6 180s-40.1 6.8-54.6 22.3c-7.6 8.1-7.1 20.7 .9 28.3s20.7 7.1 28.3-.9c5.5-5.8 14.8-9.7 25.4-9.7zm166.6 9.7c5.5-5.8 14.8-9.7 25.4-9.7s19.9 3.8 25.4 9.7c7.6 8.1 20.2 8.5 28.3 .9s8.5-20.2 .9-28.3C391.7 186.8 371 180 351.6 180s-40.1 6.8-54.6 22.3c-7.6 8.1-7.1 20.7 .9 28.3s20.7 7.1 28.3-.9zM208 320v32c0 26.5 21.5 48 48 48s48-21.5 48-48V320c0-26.5-21.5-48-48-48s-48 21.5-48 48z"],"file-code":[384,512,[],"f1c9","M64 464c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16H224v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm97 289c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L79 303c-9.4 9.4-9.4 24.6 0 33.9l48 48c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-31-31 31-31zM257 255c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l31 31-31 31c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l48-48c9.4-9.4 9.4-24.6 0-33.9l-48-48z"],"window-maximize":[512,512,[128470],"f2d0","M.3 89.5C.1 91.6 0 93.8 0 96V224 416c0 35.3 28.7 64 64 64l384 0c35.3 0 64-28.7 64-64V224 96c0-35.3-28.7-64-64-64H64c-2.2 0-4.4 .1-6.5 .3c-9.2 .9-17.8 3.8-25.5 8.2C21.8 46.5 13.4 55.1 7.7 65.5c-3.9 7.3-6.5 15.4-7.4 24zM48 224H464l0 192c0 8.8-7.2 16-16 16L64 432c-8.8 0-16-7.2-16-16l0-192z"],"face-frown":[512,512,[9785,"frown"],"f119","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM174.6 384.1c-4.5 12.5-18.2 18.9-30.7 14.4s-18.9-18.2-14.4-30.7C146.9 319.4 198.9 288 256 288s109.1 31.4 126.6 79.9c4.5 12.5-2 26.2-14.4 30.7s-26.2-2-30.7-14.4C328.2 358.5 297.2 336 256 336s-72.2 22.5-81.4 48.1zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"floppy-disk":[448,512,[128190,128426,"save"],"f0c7","M48 96V416c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V170.5c0-4.2-1.7-8.3-4.7-11.3l33.9-33.9c12 12 18.7 28.3 18.7 45.3V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H309.5c17 0 33.3 6.7 45.3 18.7l74.5 74.5-33.9 33.9L320.8 84.7c-.3-.3-.5-.5-.8-.8V184c0 13.3-10.7 24-24 24H104c-13.3 0-24-10.7-24-24V80H64c-8.8 0-16 7.2-16 16zm80-16v80H272V80H128zm32 240a64 64 0 1 1 128 0 64 64 0 1 1 -128 0z"],"comment-dots":[512,512,[128172,62075,"commenting"],"f4ad","M168.2 384.9c-15-5.4-31.7-3.1-44.6 6.4c-8.2 6-22.3 14.8-39.4 22.7c5.6-14.7 9.9-31.3 11.3-49.4c1-12.9-3.3-25.7-11.8-35.5C60.4 302.8 48 272 48 240c0-79.5 83.3-160 208-160s208 80.5 208 160s-83.3 160-208 160c-31.6 0-61.3-5.5-87.8-15.1zM26.3 423.8c-1.6 2.7-3.3 5.4-5.1 8.1l-.3 .5c-1.6 2.3-3.2 4.6-4.8 6.9c-3.5 4.7-7.3 9.3-11.3 13.5c-4.6 4.6-5.9 11.4-3.4 17.4c2.5 6 8.3 9.9 14.8 9.9c5.1 0 10.2-.3 15.3-.8l.7-.1c4.4-.5 8.8-1.1 13.2-1.9c.8-.1 1.6-.3 2.4-.5c17.8-3.5 34.9-9.5 50.1-16.1c22.9-10 42.4-21.9 54.3-30.6c31.8 11.5 67 17.9 104.1 17.9c141.4 0 256-93.1 256-208S397.4 32 256 32S0 125.1 0 240c0 45.1 17.7 86.8 47.7 120.9c-1.9 24.5-11.4 46.3-21.4 62.9zM144 272a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm144-32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm80 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"face-grin-squint":[512,512,[128518,"grin-squint"],"f585","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm349.5 52.4c18.7-4.4 35.9 12 25.5 28.1C350.4 374.6 306.3 400 255.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5zm-216-161.7l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6c0-9 9.6-14.7 17.5-10.5zM396 157.1c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9c7.9-4.2 17.5 1.5 17.5 10.5z"],"hand-pointer":[448,512,[],"f25a","M160 64c0-8.8 7.2-16 16-16s16 7.2 16 16V200c0 10.3 6.6 19.5 16.4 22.8s20.6-.1 26.8-8.3c3-3.9 7.6-6.4 12.8-6.4c8.8 0 16 7.2 16 16c0 10.3 6.6 19.5 16.4 22.8s20.6-.1 26.8-8.3c3-3.9 7.6-6.4 12.8-6.4c7.8 0 14.3 5.6 15.7 13c1.6 8.2 7.3 15.1 15.1 18s16.7 1.6 23.3-3.6c2.7-2.1 6.1-3.4 9.9-3.4c8.8 0 16 7.2 16 16l0 16V392c0 39.8-32.2 72-72 72H272 212.3h-.9c-37.4 0-72.4-18.7-93.2-49.9L50.7 312.9c-4.9-7.4-2.9-17.3 4.4-22.2s17.3-2.9 22.2 4.4L116 353.2c5.9 8.8 16.8 12.7 26.9 9.7s17-12.4 17-23V320 64zM176 0c-35.3 0-64 28.7-64 64V261.7C91.2 238 55.5 232.8 28.5 250.7C-.9 270.4-8.9 310.1 10.8 339.5L78.3 440.8c29.7 44.5 79.6 71.2 133.1 71.2h.9H272h56c66.3 0 120-53.7 120-120V288l0-16c0-35.3-28.7-64-64-64c-4.5 0-8.8 .5-13 1.3c-11.7-15.4-30.2-25.3-51-25.3c-6.9 0-13.5 1.1-19.7 3.1C288.7 170.7 269.6 160 248 160c-2.7 0-5.4 .2-8 .5V64c0-35.3-28.7-64-64-64zm48 304c0-8.8-7.2-16-16-16s-16 7.2-16 16v96c0 8.8 7.2 16 16 16s16-7.2 16-16V304zm48-16c-8.8 0-16 7.2-16 16v96c0 8.8 7.2 16 16 16s16-7.2 16-16V304c0-8.8-7.2-16-16-16zm80 16c0-8.8-7.2-16-16-16s-16 7.2-16 16v96c0 8.8 7.2 16 16 16s16-7.2 16-16V304z"],"hand-scissors":[512,512,[],"f257","M.2 276.3c-1.2-35.3 26.4-65 61.7-66.2l3.3-.1L57 208.1C22.5 200.5 .7 166.3 8.3 131.8S50.2 75.5 84.7 83.2l173 38.3c2.3-2.9 4.7-5.7 7.1-8.5l18.4-20.3C299.9 74.5 323.5 64 348.3 64l10.2 0c54.1 0 104.1 28.7 131.3 75.4l1.5 2.6c13.6 23.2 20.7 49.7 20.7 76.6L512 344c0 66.3-53.7 120-120 120l-8 0-96 0c-35.3 0-64-28.7-64-64c0-2.8 .2-5.6 .5-8.3c-19.4-11-32.5-31.8-32.5-55.7c0-.8 0-1.6 0-2.4L66.4 338c-35.3 1.2-65-26.4-66.2-61.7zm63.4-18.2c-8.8 .3-15.7 7.7-15.4 16.5s7.7 15.7 16.5 15.4l161.5-5.6c9.8-.3 18.7 5.3 22.7 14.2s2.2 19.3-4.5 26.4c-2.8 2.9-4.4 6.7-4.4 11c0 8.8 7.2 16 16 16c9.1 0 17.4 5.1 21.5 13.3s3.2 17.9-2.3 25.1c-2 2.7-3.2 6-3.2 9.6c0 8.8 7.2 16 16 16l96 0 8 0c39.8 0 72-32.2 72-72l0-125.4c0-18.4-4.9-36.5-14.2-52.4l-1.5-2.6c-18.6-32-52.8-51.6-89.8-51.6l-10.2 0c-11.3 0-22 4.8-29.6 13.1l-17.5-15.9 17.5 15.9-18.4 20.3c-.6 .6-1.1 1.3-1.7 1.9l57 13.2c8.6 2 14 10.6 12 19.2s-10.6 14-19.2 12l-85.6-19.7L74.3 130c-8.6-1.9-17.2 3.5-19.1 12.2s3.5 17.2 12.2 19.1l187.5 41.6c10.2 2.3 17.8 10.9 18.7 21.4l.1 1c.6 6.6-1.5 13.1-5.8 18.1s-10.6 7.9-17.2 8.2L63.6 258.1z"],"face-grin-tears":[640,512,[128514,"grin-tears"],"f588","M516.1 325.5c1 3 2.1 6 3.3 8.9c3.3 8.1 8.4 18.5 16.5 26.6c3.9 3.9 8.2 7.4 12.7 10.3C506.4 454.8 419.9 512 320 512s-186.4-57.2-228.6-140.6c4.5-2.9 8.7-6.3 12.7-10.3c8.1-8.1 13.2-18.6 16.5-26.6c1.2-2.9 2.3-5.9 3.3-8.9C152.5 406.2 229.5 464 320 464s167.5-57.8 196.1-138.5zM320 48c-101.4 0-185.8 72.5-204.3 168.5c-6.7-3.1-14.3-4.3-22.3-3.1c-6.8 .9-16.2 2.4-26.6 4.4C85.3 94.5 191.6 0 320 0S554.7 94.5 573.2 217.7c-10.3-2-19.8-3.5-26.6-4.4c-8-1.2-15.7 .1-22.3 3.1C505.8 120.5 421.4 48 320 48zM78.5 341.1C60 356.7 32 355.5 14.3 337.7c-18.7-18.7-19.1-48.8-.7-67.2c8.6-8.6 30.1-15.1 50.5-19.6c13-2.8 25.5-4.8 33.9-6c5.4-.8 9.9 3.7 9 9c-3.1 21.5-11.4 70.2-25.5 84.4c-.9 1-1.9 1.8-2.9 2.7zm483 0c-.8-.6-1.5-1.3-2.3-2c-.2-.2-.5-.4-.7-.7c-14.1-14.1-22.5-62.9-25.5-84.4c-.8-5.4 3.7-9.9 9-9c1 .1 2.2 .3 3.3 .5c8.2 1.2 19.2 3 30.6 5.5c20.4 4.4 41.9 10.9 50.5 19.6c18.4 18.4 18 48.5-.7 67.2c-17.7 17.7-45.7 19-64.2 3.4zM439 336.5C414.4 374.6 370.3 400 319.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5c18.7-4.4 35.9 12 25.5 28.1zM281.6 228.8l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0zm160 0l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0z"],"calendar-xmark":[448,512,["calendar-times"],"f273","M128 0c13.3 0 24 10.7 24 24V64H296V24c0-13.3 10.7-24 24-24s24 10.7 24 24V64h40c35.3 0 64 28.7 64 64v16 48V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V192 144 128C0 92.7 28.7 64 64 64h40V24c0-13.3 10.7-24 24-24zM400 192H48V448c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V192zm-95 89l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z"],"file-video":[384,512,[],"f1c8","M320 464c8.8 0 16-7.2 16-16V160H256c-17.7 0-32-14.3-32-32V48H64c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16H320zM0 64C0 28.7 28.7 0 64 0H229.5c17 0 33.3 6.7 45.3 18.7l90.5 90.5c12 12 18.7 28.3 18.7 45.3V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM80 288c0-17.7 14.3-32 32-32h96c17.7 0 32 14.3 32 32v16l44.9-29.9c2-1.3 4.4-2.1 6.8-2.1c6.8 0 12.3 5.5 12.3 12.3V387.7c0 6.8-5.5 12.3-12.3 12.3c-2.4 0-4.8-.7-6.8-2.1L240 368v16c0 17.7-14.3 32-32 32H112c-17.7 0-32-14.3-32-32V288z"],"file-pdf":[512,512,[],"f1c1","M64 464l48 0 0 48-48 0c-35.3 0-64-28.7-64-64L0 64C0 28.7 28.7 0 64 0L229.5 0c17 0 33.3 6.7 45.3 18.7l90.5 90.5c12 12 18.7 28.3 18.7 45.3L384 304l-48 0 0-144-80 0c-17.7 0-32-14.3-32-32l0-80L64 48c-8.8 0-16 7.2-16 16l0 384c0 8.8 7.2 16 16 16zM176 352l32 0c30.9 0 56 25.1 56 56s-25.1 56-56 56l-16 0 0 32c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-48 0-80c0-8.8 7.2-16 16-16zm32 80c13.3 0 24-10.7 24-24s-10.7-24-24-24l-16 0 0 48 16 0zm96-80l32 0c26.5 0 48 21.5 48 48l0 64c0 26.5-21.5 48-48 48l-32 0c-8.8 0-16-7.2-16-16l0-128c0-8.8 7.2-16 16-16zm32 128c8.8 0 16-7.2 16-16l0-64c0-8.8-7.2-16-16-16l-16 0 0 96 16 0zm80-112c0-8.8 7.2-16 16-16l48 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-32 0 0 32 32 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-32 0 0 48c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-64 0-64z"],comment:[512,512,[128489,61669],"f075","M123.6 391.3c12.9-9.4 29.6-11.8 44.6-6.4c26.5 9.6 56.2 15.1 87.8 15.1c124.7 0 208-80.5 208-160s-83.3-160-208-160S48 160.5 48 240c0 32 12.4 62.8 35.7 89.2c8.6 9.7 12.8 22.5 11.8 35.5c-1.4 18.1-5.7 34.7-11.3 49.4c17-7.9 31.1-16.7 39.4-22.7zM21.2 431.9c1.8-2.7 3.5-5.4 5.1-8.1c10-16.6 19.5-38.4 21.4-62.9C17.7 326.8 0 285.1 0 240C0 125.1 114.6 32 256 32s256 93.1 256 208s-114.6 208-256 208c-37.1 0-72.3-6.4-104.1-17.9c-11.9 8.7-31.3 20.6-54.3 30.6c-15.1 6.6-32.3 12.6-50.1 16.1c-.8 .2-1.6 .3-2.4 .5c-4.4 .8-8.7 1.5-13.2 1.9c-.2 0-.5 .1-.7 .1c-5.1 .5-10.2 .8-15.3 .8c-6.5 0-12.3-3.9-14.8-9.9c-2.5-6-1.1-12.8 3.4-17.4c4.1-4.2 7.8-8.7 11.3-13.5c1.7-2.3 3.3-4.6 4.8-6.9c.1-.2 .2-.3 .3-.5z"],envelope:[512,512,[128386,9993,61443],"f0e0","M64 112c-8.8 0-16 7.2-16 16v22.1L220.5 291.7c20.7 17 50.4 17 71.1 0L464 150.1V128c0-8.8-7.2-16-16-16H64zM48 212.2V384c0 8.8 7.2 16 16 16H448c8.8 0 16-7.2 16-16V212.2L322 328.8c-38.4 31.5-93.7 31.5-132 0L48 212.2zM0 128C0 92.7 28.7 64 64 64H448c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128z"],hourglass:[384,512,[9203,62032,"hourglass-empty"],"f254","M24 0C10.7 0 0 10.7 0 24S10.7 48 24 48h8V67c0 40.3 16 79 44.5 107.5L158.1 256 76.5 337.5C48 366 32 404.7 32 445v19H24c-13.3 0-24 10.7-24 24s10.7 24 24 24H360c13.3 0 24-10.7 24-24s-10.7-24-24-24h-8V445c0-40.3-16-79-44.5-107.5L225.9 256l81.5-81.5C336 146 352 107.3 352 67V48h8c13.3 0 24-10.7 24-24s-10.7-24-24-24H24zM192 289.9l81.5 81.5C293 391 304 417.4 304 445v19H80V445c0-27.6 11-54 30.5-73.5L192 289.9zm0-67.9l-81.5-81.5C91 121 80 94.6 80 67V48H304V67c0 27.6-11 54-30.5 73.5L192 222.1z"],"calendar-check":[448,512,[],"f274","M128 0c13.3 0 24 10.7 24 24V64H296V24c0-13.3 10.7-24 24-24s24 10.7 24 24V64h40c35.3 0 64 28.7 64 64v16 48V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V192 144 128C0 92.7 28.7 64 64 64h40V24c0-13.3 10.7-24 24-24zM400 192H48V448c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V192zM329 297L217 409c-9.4 9.4-24.6 9.4-33.9 0l-64-64c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47 95-95c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z"],"hard-drive":[512,512,[128436,"hdd"],"f0a0","M64 80c-8.8 0-16 7.2-16 16V258c5.1-1.3 10.5-2 16-2H448c5.5 0 10.9 .7 16 2V96c0-8.8-7.2-16-16-16H64zM48 320v96c0 8.8 7.2 16 16 16H448c8.8 0 16-7.2 16-16V320c0-8.8-7.2-16-16-16H64c-8.8 0-16 7.2-16 16zM0 320V96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V320v96c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V320zm280 48a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm120-24a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],"face-grin-squint-tears":[512,512,[129315,"grin-squint-tears"],"f586","M426.8 14.2C446-5 477.5-4.6 497.1 14.9s20 51 .7 70.3c-14.8 14.8-65.7 23.6-88.3 26.7c-5.6 .9-10.3-3.9-9.5-9.5C403.3 79.9 412 29 426.8 14.2zM75 75C158.2-8.3 284.5-22.2 382.2 33.2c-1.5 4.8-2.9 9.6-4.1 14.3c-3.1 12.2-5.5 24.6-7.3 35c-80.8-53.6-190.7-44.8-261.9 26.4C37.7 180.1 28.9 290 82.5 370.8c-10.5 1.8-22.9 4.2-35 7.3c-4.7 1.2-9.5 2.5-14.3 4.1C-22.2 284.5-8.2 158.2 75 75zm389.6 58.9c4.7-1.2 9.5-2.5 14.3-4.1C534.2 227.5 520.2 353.8 437 437c-83.2 83.2-209.5 97.2-307.2 41.8c1.5-4.8 2.8-9.6 4-14.3c3.1-12.2 5.5-24.6 7.3-35c80.8 53.6 190.7 44.8 261.9-26.4c71.2-71.2 80-181.1 26.4-261.9c10.5-1.8 22.9-4.2 35-7.3zm-105.4 93c10.1-16.3 33.9-16.9 37.9 1.9c9.5 44.4-3.7 93.5-39.3 129.1s-84.8 48.8-129.1 39.3c-18.7-4-18.2-27.8-1.9-37.9c25.2-15.7 50.2-35.4 73.6-58.8s43.1-48.4 58.8-73.6zM92 265.3l97.4-29.7c11.6-3.5 22.5 7.3 19 19l-29.7 97.4c-2.6 8.6-13.4 11.3-19.8 4.9c-2-2-3.2-4.6-3.4-7.3l-5.1-56.1-56.1-5.1c-2.8-.3-5.4-1.5-7.3-3.4c-6.3-6.3-3.6-17.2 4.9-19.8zm193-178.2c2 2 3.2 4.6 3.4 7.3l5.1 56.1 56.1 5.1c2.8 .3 5.4 1.5 7.3 3.4c6.3 6.3 3.6 17.2-4.9 19.8l-97.4 29.7c-11.6 3.5-22.5-7.3-19-19L265.3 92c2.6-8.6 13.4-11.3 19.8-4.9zM14.9 497.1c-19.6-19.6-20-51-.7-70.3C29 412 79.8 403.2 102.4 400.1c5.6-.9 10.3 3.9 9.5 9.5c-3.2 22.5-11.9 73.5-26.7 88.3C66 517 34.5 516.6 14.9 497.1z"],"rectangle-list":[576,512,["list-alt"],"f022","M64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H512c8.8 0 16-7.2 16-16V96c0-8.8-7.2-16-16-16H64zM0 96C0 60.7 28.7 32 64 32H512c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm96 64a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm104 0c0-13.3 10.7-24 24-24H448c13.3 0 24 10.7 24 24s-10.7 24-24 24H224c-13.3 0-24-10.7-24-24zm0 96c0-13.3 10.7-24 24-24H448c13.3 0 24 10.7 24 24s-10.7 24-24 24H224c-13.3 0-24-10.7-24-24zm0 96c0-13.3 10.7-24 24-24H448c13.3 0 24 10.7 24 24s-10.7 24-24 24H224c-13.3 0-24-10.7-24-24zm-72-64a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM96 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],"calendar-plus":[448,512,[],"f271","M152 24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H64C28.7 64 0 92.7 0 128v16 48V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V192 144 128c0-35.3-28.7-64-64-64H344V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H152V24zM48 192H400V448c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V192zm176 40c-13.3 0-24 10.7-24 24v48H152c-13.3 0-24 10.7-24 24s10.7 24 24 24h48v48c0 13.3 10.7 24 24 24s24-10.7 24-24V352h48c13.3 0 24-10.7 24-24s-10.7-24-24-24H248V256c0-13.3-10.7-24-24-24z"],"circle-left":[512,512,[61840,"arrow-alt-circle-left"],"f359","M48 256a208 208 0 1 1 416 0A208 208 0 1 1 48 256zm464 0A256 256 0 1 0 0 256a256 256 0 1 0 512 0zM217.4 376.9c4.2 4.5 10.1 7.1 16.3 7.1c12.3 0 22.3-10 22.3-22.3V304h96c17.7 0 32-14.3 32-32V240c0-17.7-14.3-32-32-32H256V150.3c0-12.3-10-22.3-22.3-22.3c-6.2 0-12.1 2.6-16.3 7.1L117.5 242.2c-3.5 3.8-5.5 8.7-5.5 13.8s2 10.1 5.5 13.8l99.9 107.1z"],"money-bill-1":[576,512,["money-bill-alt"],"f3d1","M112 112c0 35.3-28.7 64-64 64V336c35.3 0 64 28.7 64 64H464c0-35.3 28.7-64 64-64V176c-35.3 0-64-28.7-64-64H112zM0 128C0 92.7 28.7 64 64 64H512c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128zM176 256a112 112 0 1 1 224 0 112 112 0 1 1 -224 0zm80-48c0 8.8 7.2 16 16 16v64h-8c-8.8 0-16 7.2-16 16s7.2 16 16 16h24 24c8.8 0 16-7.2 16-16s-7.2-16-16-16h-8V208c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16z"],clock:[512,512,[128339,"clock-four"],"f017","M464 256A208 208 0 1 1 48 256a208 208 0 1 1 416 0zM0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zM232 120V256c0 8 4 15.5 10.7 20l96 64c11 7.4 25.9 4.4 33.3-6.7s4.4-25.9-6.7-33.3L280 243.2V120c0-13.3-10.7-24-24-24s-24 10.7-24 24z"],keyboard:[576,512,[9e3],"f11c","M64 112c-8.8 0-16 7.2-16 16V384c0 8.8 7.2 16 16 16H512c8.8 0 16-7.2 16-16V128c0-8.8-7.2-16-16-16H64zM0 128C0 92.7 28.7 64 64 64H512c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128zM176 320H400c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zm-72-72c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H120c-8.8 0-16-7.2-16-16V248zm16-96h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H120c-8.8 0-16-7.2-16-16V168c0-8.8 7.2-16 16-16zm64 96c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H200c-8.8 0-16-7.2-16-16V248zm16-96h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H200c-8.8 0-16-7.2-16-16V168c0-8.8 7.2-16 16-16zm64 96c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H280c-8.8 0-16-7.2-16-16V248zm16-96h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H280c-8.8 0-16-7.2-16-16V168c0-8.8 7.2-16 16-16zm64 96c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H360c-8.8 0-16-7.2-16-16V248zm16-96h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H360c-8.8 0-16-7.2-16-16V168c0-8.8 7.2-16 16-16zm64 96c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H440c-8.8 0-16-7.2-16-16V248zm16-96h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H440c-8.8 0-16-7.2-16-16V168c0-8.8 7.2-16 16-16z"],"closed-captioning":[576,512,[],"f20a","M512 80c8.8 0 16 7.2 16 16V416c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V96c0-8.8 7.2-16 16-16H512zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM200 208c14.2 0 27 6.1 35.8 16c8.8 9.9 24 10.7 33.9 1.9s10.7-24 1.9-33.9c-17.5-19.6-43.1-32-71.5-32c-53 0-96 43-96 96s43 96 96 96c28.4 0 54-12.4 71.5-32c8.8-9.9 8-25-1.9-33.9s-25-8-33.9 1.9c-8.8 9.9-21.6 16-35.8 16c-26.5 0-48-21.5-48-48s21.5-48 48-48zm144 48c0-26.5 21.5-48 48-48c14.2 0 27 6.1 35.8 16c8.8 9.9 24 10.7 33.9 1.9s10.7-24 1.9-33.9c-17.5-19.6-43.1-32-71.5-32c-53 0-96 43-96 96s43 96 96 96c28.4 0 54-12.4 71.5-32c8.8-9.9 8-25-1.9-33.9s-25-8-33.9 1.9c-8.8 9.9-21.6 16-35.8 16c-26.5 0-48-21.5-48-48z"],images:[576,512,[],"f302","M160 80H512c8.8 0 16 7.2 16 16V320c0 8.8-7.2 16-16 16H490.8L388.1 178.9c-4.4-6.8-12-10.9-20.1-10.9s-15.7 4.1-20.1 10.9l-52.2 79.8-12.4-16.9c-4.5-6.2-11.7-9.8-19.4-9.8s-14.8 3.6-19.4 9.8L175.6 336H160c-8.8 0-16-7.2-16-16V96c0-8.8 7.2-16 16-16zM96 96V320c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H160c-35.3 0-64 28.7-64 64zM48 120c0-13.3-10.7-24-24-24S0 106.7 0 120V344c0 75.1 60.9 136 136 136H456c13.3 0 24-10.7 24-24s-10.7-24-24-24H136c-48.6 0-88-39.4-88-88V120zm208 24a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z"],"face-grin":[512,512,[128512,"grin"],"f580","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm349.5 52.4c18.7-4.4 35.9 12 25.5 28.1C350.4 374.6 306.3 400 255.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"face-meh":[512,512,[128528,"meh"],"f11a","M464 256A208 208 0 1 1 48 256a208 208 0 1 1 416 0zM256 0a256 256 0 1 0 0 512A256 256 0 1 0 256 0zM176.4 240a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm192-32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM184 328c-13.3 0-24 10.7-24 24s10.7 24 24 24H328c13.3 0 24-10.7 24-24s-10.7-24-24-24H184z"],"id-card":[576,512,[62147,"drivers-license"],"f2c2","M528 160V416c0 8.8-7.2 16-16 16H320c0-44.2-35.8-80-80-80H176c-44.2 0-80 35.8-80 80H64c-8.8 0-16-7.2-16-16V160H528zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM272 256a64 64 0 1 0 -128 0 64 64 0 1 0 128 0zm104-48c-13.3 0-24 10.7-24 24s10.7 24 24 24h80c13.3 0 24-10.7 24-24s-10.7-24-24-24H376zm0 96c-13.3 0-24 10.7-24 24s10.7 24 24 24h80c13.3 0 24-10.7 24-24s-10.7-24-24-24H376z"],sun:[512,512,[9728],"f185","M375.7 19.7c-1.5-8-6.9-14.7-14.4-17.8s-16.1-2.2-22.8 2.4L256 61.1 173.5 4.2c-6.7-4.6-15.3-5.5-22.8-2.4s-12.9 9.8-14.4 17.8l-18.1 98.5L19.7 136.3c-8 1.5-14.7 6.9-17.8 14.4s-2.2 16.1 2.4 22.8L61.1 256 4.2 338.5c-4.6 6.7-5.5 15.3-2.4 22.8s9.8 13 17.8 14.4l98.5 18.1 18.1 98.5c1.5 8 6.9 14.7 14.4 17.8s16.1 2.2 22.8-2.4L256 450.9l82.5 56.9c6.7 4.6 15.3 5.5 22.8 2.4s12.9-9.8 14.4-17.8l18.1-98.5 98.5-18.1c8-1.5 14.7-6.9 17.8-14.4s2.2-16.1-2.4-22.8L450.9 256l56.9-82.5c4.6-6.7 5.5-15.3 2.4-22.8s-9.8-12.9-17.8-14.4l-98.5-18.1L375.7 19.7zM269.6 110l65.6-45.2 14.4 78.3c1.8 9.8 9.5 17.5 19.3 19.3l78.3 14.4L402 242.4c-5.7 8.2-5.7 19 0 27.2l45.2 65.6-78.3 14.4c-9.8 1.8-17.5 9.5-19.3 19.3l-14.4 78.3L269.6 402c-8.2-5.7-19-5.7-27.2 0l-65.6 45.2-14.4-78.3c-1.8-9.8-9.5-17.5-19.3-19.3L64.8 335.2 110 269.6c5.7-8.2 5.7-19 0-27.2L64.8 176.8l78.3-14.4c9.8-1.8 17.5-9.5 19.3-19.3l14.4-78.3L242.4 110c8.2 5.7 19 5.7 27.2 0zM256 368a112 112 0 1 0 0-224 112 112 0 1 0 0 224zM192 256a64 64 0 1 1 128 0 64 64 0 1 1 -128 0z"],"face-laugh-wink":[512,512,["laugh-wink"],"f59c","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm130.7 57.9c-4.2-13.6 7.1-25.9 21.3-25.9H364.5c14.2 0 25.5 12.4 21.3 25.9C369 368.4 318.2 408 258.2 408s-110.8-39.6-127.5-94.1zM144.4 192a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm165.8 21.7c-7.6 8.1-20.2 8.5-28.3 .9s-8.5-20.2-.9-28.3c14.5-15.5 35.2-22.3 54.6-22.3s40.1 6.8 54.6 22.3c7.6 8.1 7.1 20.7-.9 28.3s-20.7 7.1-28.3-.9c-5.5-5.8-14.8-9.7-25.4-9.7s-19.9 3.8-25.4 9.7z"],"circle-down":[512,512,[61466,"arrow-alt-circle-down"],"f358","M256 464a208 208 0 1 1 0-416 208 208 0 1 1 0 416zM256 0a256 256 0 1 0 0 512A256 256 0 1 0 256 0zM376.9 294.6c4.5-4.2 7.1-10.1 7.1-16.3c0-12.3-10-22.3-22.3-22.3H304V160c0-17.7-14.3-32-32-32l-32 0c-17.7 0-32 14.3-32 32v96H150.3C138 256 128 266 128 278.3c0 6.2 2.6 12.1 7.1 16.3l107.1 99.9c3.8 3.5 8.7 5.5 13.8 5.5s10.1-2 13.8-5.5l107.1-99.9z"],"thumbs-down":[512,512,[128078,61576],"f165","M323.8 477.2c-38.2 10.9-78.1-11.2-89-49.4l-5.7-20c-3.7-13-10.4-25-19.5-35l-51.3-56.4c-8.9-9.8-8.2-25 1.6-33.9s25-8.2 33.9 1.6l51.3 56.4c14.1 15.5 24.4 34 30.1 54.1l5.7 20c3.6 12.7 16.9 20.1 29.7 16.5s20.1-16.9 16.5-29.7l-5.7-20c-5.7-19.9-14.7-38.7-26.6-55.5c-5.2-7.3-5.8-16.9-1.7-24.9s12.3-13 21.3-13L448 288c8.8 0 16-7.2 16-16c0-6.8-4.3-12.7-10.4-15c-7.4-2.8-13-9-14.9-16.7s.1-15.8 5.3-21.7c2.5-2.8 4-6.5 4-10.6c0-7.8-5.6-14.3-13-15.7c-8.2-1.6-15.1-7.3-18-15.2s-1.6-16.7 3.6-23.3c2.1-2.7 3.4-6.1 3.4-9.9c0-6.7-4.2-12.6-10.2-14.9c-11.5-4.5-17.7-16.9-14.4-28.8c.4-1.3 .6-2.8 .6-4.3c0-8.8-7.2-16-16-16H286.5c-12.6 0-25 3.7-35.5 10.7l-61.7 41.1c-11 7.4-25.9 4.4-33.3-6.7s-4.4-25.9 6.7-33.3l61.7-41.1c18.4-12.3 40-18.8 62.1-18.8H384c34.7 0 62.9 27.6 64 62c14.6 11.7 24 29.7 24 50c0 4.5-.5 8.8-1.3 13c15.4 11.7 25.3 30.2 25.3 51c0 6.5-1 12.8-2.8 18.7C504.8 238.3 512 254.3 512 272c0 35.3-28.6 64-64 64l-92.3 0c4.7 10.4 8.7 21.2 11.8 32.2l5.7 20c10.9 38.2-11.2 78.1-49.4 89zM32 384c-17.7 0-32-14.3-32-32V128c0-17.7 14.3-32 32-32H96c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32H32z"],"chess-pawn":[320,512,[9823],"f443","M232 152A72 72 0 1 0 88 152a72 72 0 1 0 144 0zm24 120H243.4l10.7 80H205.7L195 272H160 125l-10.7 80H65.9l10.7-80H64c-13.3 0-24-10.7-24-24s10.7-24 24-24c-15.1-20.1-24-45-24-72C40 85.7 93.7 32 160 32s120 53.7 120 120c0 27-8.9 51.9-24 72c13.3 0 24 10.7 24 24s-10.7 24-24 24zM52.7 464H267.3l-16.6-32H69.2L52.7 464zm207.9-80c12 0 22.9 6.7 28.4 17.3l26.5 51.2c3 5.8 4.6 12.2 4.6 18.7c0 22.5-18.2 40.8-40.8 40.8H40.8C18.2 512 0 493.8 0 471.2c0-6.5 1.6-12.9 4.6-18.7l26.5-51.2C36.5 390.7 47.5 384 59.5 384h201z"],"credit-card":[576,512,[128179,62083,"credit-card-alt"],"f09d","M512 80c8.8 0 16 7.2 16 16v32H48V96c0-8.8 7.2-16 16-16H512zm16 144V416c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V224H528zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm56 304c-13.3 0-24 10.7-24 24s10.7 24 24 24h48c13.3 0 24-10.7 24-24s-10.7-24-24-24H120zm128 0c-13.3 0-24 10.7-24 24s10.7 24 24 24H360c13.3 0 24-10.7 24-24s-10.7-24-24-24H248z"],bell:[448,512,[128276,61602],"f0f3","M224 0c-17.7 0-32 14.3-32 32V51.2C119 66 64 130.6 64 208v25.4c0 45.4-15.5 89.5-43.8 124.9L5.3 377c-5.8 7.2-6.9 17.1-2.9 25.4S14.8 416 24 416H424c9.2 0 17.6-5.3 21.6-13.6s2.9-18.2-2.9-25.4l-14.9-18.6C399.5 322.9 384 278.8 384 233.4V208c0-77.4-55-142-128-156.8V32c0-17.7-14.3-32-32-32zm0 96c61.9 0 112 50.1 112 112v25.4c0 47.9 13.9 94.6 39.7 134.6H72.3C98.1 328 112 281.3 112 233.4V208c0-61.9 50.1-112 112-112zm64 352H224 160c0 17 6.7 33.3 18.7 45.3s28.3 18.7 45.3 18.7s33.3-6.7 45.3-18.7s18.7-28.3 18.7-45.3z"],file:[384,512,[128196,128459,61462],"f15b","M320 464c8.8 0 16-7.2 16-16V160H256c-17.7 0-32-14.3-32-32V48H64c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16H320zM0 64C0 28.7 28.7 0 64 0H229.5c17 0 33.3 6.7 45.3 18.7l90.5 90.5c12 12 18.7 28.3 18.7 45.3V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64z"],hospital:[640,512,[127973,62589,"hospital-alt","hospital-wide"],"f0f8","M232 0c-39.8 0-72 32.2-72 72v8H72C32.2 80 0 112.2 0 152V440c0 39.8 32.2 72 72 72h.2 .2 .2 .2 .2H73h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2H75h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2H77h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2H79h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2H82h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2H85h.2 .2 .2 .2H86h.2 .2 .2 .2H87h.2 .2 .2 .2H88h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2H98h.2 .2 .2 .2H99h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2v0H456h8v0H568c39.8 0 72-32.2 72-72V152c0-39.8-32.2-72-72-72H480V72c0-39.8-32.2-72-72-72H232zM480 128h88c13.3 0 24 10.7 24 24v40H536c-13.3 0-24 10.7-24 24s10.7 24 24 24h56v48H536c-13.3 0-24 10.7-24 24s10.7 24 24 24h56V440c0 13.3-10.7 24-24 24H480V336 128zM72 128h88V464h-.1-.2-.2-.2H159h-.2-.2-.2H158h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H154h-.2-.2-.2H153h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H150h-.2-.2-.2H149h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H146h-.2-.2-.2H145h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H142h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H139h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H136h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H133h-.2-.2-.2-.2-.2-.2-.2-.2H131h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H128h-.2-.2-.2-.2-.2-.2-.2-.2H126h-.2-.2-.2-.2-.2-.2-.2-.2H124h-.2-.2-.2-.2-.2-.2-.2-.2H122h-.2-.2-.2-.2-.2-.2-.2-.2H120h-.2-.2-.2-.2-.2-.2-.2-.2H118h-.2-.2-.2-.2-.2-.2-.2-.2H116h-.2-.2-.2-.2-.2-.2-.2-.2H114h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H111h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H108h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H105h-.2-.2-.2-.2H104h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H100h-.2-.2-.2-.2H99h-.2-.2-.2-.2H98h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H88h-.2-.2-.2-.2H87h-.2-.2-.2-.2H86h-.2-.2-.2-.2H85h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H82h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H79h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H77h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H75h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H73h-.2-.2-.2-.2-.2H72c-13.2 0-24-10.7-24-24V336h56c13.3 0 24-10.7 24-24s-10.7-24-24-24H48V240h56c13.3 0 24-10.7 24-24s-10.7-24-24-24H48V152c0-13.3 10.7-24 24-24zM208 72c0-13.3 10.7-24 24-24H408c13.3 0 24 10.7 24 24V336 464H368V400c0-26.5-21.5-48-48-48s-48 21.5-48 48v64H208V72zm88 24v24H272c-8.8 0-16 7.2-16 16v16c0 8.8 7.2 16 16 16h24v24c0 8.8 7.2 16 16 16h16c8.8 0 16-7.2 16-16V168h24c8.8 0 16-7.2 16-16V136c0-8.8-7.2-16-16-16H344V96c0-8.8-7.2-16-16-16H312c-8.8 0-16 7.2-16 16z"],"chess-rook":[448,512,[9820],"f447","M80 80V192c0 2.5 1.2 4.9 3.2 6.4l51.2 38.4c6.8 5.1 10.4 13.4 9.5 21.9L133.5 352H85.2l9.4-85L54.4 236.8C40.3 226.2 32 209.6 32 192V72c0-22.1 17.9-40 40-40H376c22.1 0 40 17.9 40 40V192c0 17.6-8.3 34.2-22.4 44.8L353.4 267l9.4 85H314.5l-10.4-93.3c-.9-8.4 2.7-16.8 9.5-21.9l51.2-38.4c2-1.5 3.2-3.9 3.2-6.4V80H304v24c0 13.3-10.7 24-24 24s-24-10.7-24-24V80H192v24c0 13.3-10.7 24-24 24s-24-10.7-24-24V80H80zm4.7 384H363.3l-16.6-32H101.2L84.7 464zm271.9-80c12 0 22.9 6.7 28.4 17.3l26.5 51.2c3 5.8 4.6 12.2 4.6 18.7c0 22.5-18.2 40.8-40.8 40.8H72.8C50.2 512 32 493.8 32 471.2c0-6.5 1.6-12.9 4.6-18.7l26.5-51.2C68.5 390.7 79.5 384 91.5 384h265zM208 288c-8.8 0-16-7.2-16-16V224c0-17.7 14.3-32 32-32s32 14.3 32 32v48c0 8.8-7.2 16-16 16H208z"],"star-half":[576,512,[61731],"f089","M293.3 .6c10.9 2.5 18.6 12.2 18.6 23.4V408.7c0 8.9-4.9 17-12.7 21.2L151 509.1c-8.1 4.3-17.9 3.7-25.3-1.7s-11.2-14.5-9.7-23.5l26.2-155.6L31.1 218.3c-6.5-6.4-8.7-15.9-5.9-24.5s10.3-14.9 19.3-16.3l153.2-22.6L266.3 13.5c4.9-10.1 16.1-15.4 27-12.9zM263.9 128.4l-28.6 58.8c-3.5 7.1-10.2 12.1-18.1 13.3L99 217.9 184.9 303c5.5 5.5 8.1 13.3 6.8 21L171.4 443.7l92.5-49.4V128.4z"],"chess-king":[448,512,[9818],"f43f","M248 24c0-13.3-10.7-24-24-24s-24 10.7-24 24V56H168c-13.3 0-24 10.7-24 24s10.7 24 24 24h32v40H59.6C26.7 144 0 170.7 0 203.6c0 8.2 1.7 16.3 4.9 23.8L59.1 352h52.3L49 208.2c-.6-1.5-1-3-1-4.6c0-6.4 5.2-11.6 11.6-11.6H224 388.4c6.4 0 11.6 5.2 11.6 11.6c0 1.6-.3 3.2-1 4.6L336.5 352h52.3l54.2-124.6c3.3-7.5 4.9-15.6 4.9-23.8c0-32.9-26.7-59.6-59.6-59.6H248V104h32c13.3 0 24-10.7 24-24s-10.7-24-24-24H248V24zM101.2 432H346.8l16.6 32H84.7l16.6-32zm283.7-30.7c-5.5-10.6-16.5-17.3-28.4-17.3H91.5c-12 0-22.9 6.7-28.4 17.3L36.6 452.5c-3 5.8-4.6 12.2-4.6 18.7C32 493.8 50.2 512 72.8 512H375.2c22.5 0 40.8-18.2 40.8-40.8c0-6.5-1.6-12.9-4.6-18.7l-26.5-51.2z"],"circle-user":[512,512,[62142,"user-circle"],"f2bd","M406.5 399.6C387.4 352.9 341.5 320 288 320H224c-53.5 0-99.4 32.9-118.5 79.6C69.9 362.2 48 311.7 48 256C48 141.1 141.1 48 256 48s208 93.1 208 208c0 55.7-21.9 106.2-57.5 143.6zm-40.1 32.7C334.4 452.4 296.6 464 256 464s-78.4-11.6-110.5-31.7c7.3-36.7 39.7-64.3 78.5-64.3h64c38.8 0 71.2 27.6 78.5 64.3zM256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm0-272a40 40 0 1 1 0-80 40 40 0 1 1 0 80zm-88-40a88 88 0 1 0 176 0 88 88 0 1 0 -176 0z"],copy:[448,512,[],"f0c5","M384 336H192c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16l140.1 0L400 115.9V320c0 8.8-7.2 16-16 16zM192 384H384c35.3 0 64-28.7 64-64V115.9c0-12.7-5.1-24.9-14.1-33.9L366.1 14.1c-9-9-21.2-14.1-33.9-14.1H192c-35.3 0-64 28.7-64 64V320c0 35.3 28.7 64 64 64zM64 128c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64H256c35.3 0 64-28.7 64-64V416H272v32c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V192c0-8.8 7.2-16 16-16H96V128H64z"],"share-from-square":[576,512,[61509,"share-square"],"f14d","M400 255.4V240 208c0-8.8-7.2-16-16-16H352 336 289.5c-50.9 0-93.9 33.5-108.3 79.6c-3.3-9.4-5.2-19.8-5.2-31.6c0-61.9 50.1-112 112-112h48 16 32c8.8 0 16-7.2 16-16V80 64.6L506 160 400 255.4zM336 240h16v48c0 17.7 14.3 32 32 32h3.7c7.9 0 15.5-2.9 21.4-8.2l139-125.1c7.6-6.8 11.9-16.5 11.9-26.7s-4.3-19.9-11.9-26.7L409.9 8.9C403.5 3.2 395.3 0 386.7 0C367.5 0 352 15.5 352 34.7V80H336 304 288c-88.4 0-160 71.6-160 160c0 60.4 34.6 99.1 63.9 120.9c5.9 4.4 11.5 8.1 16.7 11.2c4.4 2.7 8.5 4.9 11.9 6.6c3.4 1.7 6.2 3 8.2 3.9c2.2 1 4.6 1.4 7.1 1.4h2.5c9.8 0 17.8-8 17.8-17.8c0-7.8-5.3-14.7-11.6-19.5l0 0c-.4-.3-.7-.5-1.1-.8c-1.7-1.1-3.4-2.5-5-4.1c-.8-.8-1.7-1.6-2.5-2.6s-1.6-1.9-2.4-2.9c-1.8-2.5-3.5-5.3-5-8.5c-2.6-6-4.3-13.3-4.3-22.4c0-36.1 29.3-65.5 65.5-65.5H304h32zM72 32C32.2 32 0 64.2 0 104V440c0 39.8 32.2 72 72 72H408c39.8 0 72-32.2 72-72V376c0-13.3-10.7-24-24-24s-24 10.7-24 24v64c0 13.3-10.7 24-24 24H72c-13.3 0-24-10.7-24-24V104c0-13.3 10.7-24 24-24h64c13.3 0 24-10.7 24-24s-10.7-24-24-24H72z"],copyright:[512,512,[169],"f1f9","M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM199.4 312.6c-31.2-31.2-31.2-81.9 0-113.1s81.9-31.2 113.1 0c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9c-50-50-131-50-181 0s-50 131 0 181s131 50 181 0c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0c-31.2 31.2-81.9 31.2-113.1 0z"],map:[576,512,[128506,62072],"f279","M565.6 36.2C572.1 40.7 576 48.1 576 56V392c0 10-6.2 18.9-15.5 22.4l-168 64c-5.2 2-10.9 2.1-16.1 .3L192.5 417.5l-160 61c-7.4 2.8-15.7 1.8-22.2-2.7S0 463.9 0 456V120c0-10 6.1-18.9 15.5-22.4l168-64c5.2-2 10.9-2.1 16.1-.3L383.5 94.5l160-61c7.4-2.8 15.7-1.8 22.2 2.7zM48 136.5V421.2l120-45.7V90.8L48 136.5zM360 422.7V137.3l-144-48V374.7l144 48zm48-1.5l120-45.7V90.8L408 136.5V421.2z"],"bell-slash":[640,512,[128277,61943],"f1f6","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L542.6 400c2.7-7.8 1.3-16.5-3.9-23l-14.9-18.6C495.5 322.9 480 278.8 480 233.4V208c0-77.4-55-142-128-156.8V32c0-17.7-14.3-32-32-32s-32 14.3-32 32V51.2c-42.6 8.6-79 34.2-102 69.3L38.8 5.1zM224 150.3C243.6 117.7 279.3 96 320 96c61.9 0 112 50.1 112 112v25.4c0 32.7 6.4 64.8 18.7 94.5L224 150.3zM406.2 416l-60.9-48H168.3c21.2-32.8 34.4-70.3 38.4-109.1L160 222.1v11.4c0 45.4-15.5 89.5-43.8 124.9L101.3 377c-5.8 7.2-6.9 17.1-2.9 25.4s12.4 13.6 21.6 13.6H406.2zM384 448H320 256c0 17 6.7 33.3 18.7 45.3s28.3 18.7 45.3 18.7s33.3-6.7 45.3-18.7s18.7-28.3 18.7-45.3z"],"hand-lizard":[512,512,[],"f258","M72 112c-13.3 0-24 10.7-24 24s10.7 24 24 24H240c35.3 0 64 28.7 64 64s-28.7 64-64 64H136c-13.3 0-24 10.7-24 24s10.7 24 24 24H288c4.5 0 8.9 1.3 12.7 3.6l64 40c7 4.4 11.3 12.1 11.3 20.4v24c0 13.3-10.7 24-24 24s-24-10.7-24-24V413.3L281.1 384H136c-39.8 0-72-32.2-72-72s32.2-72 72-72H240c8.8 0 16-7.2 16-16s-7.2-16-16-16H72c-39.8 0-72-32.2-72-72S32.2 64 72 64H281.6c46.7 0 90.9 21.5 119.7 58.3l78.4 100.1c20.9 26.7 32.3 59.7 32.3 93.7V424c0 13.3-10.7 24-24 24s-24-10.7-24-24V316.1c0-23.2-7.8-45.8-22.1-64.1L363.5 151.9c-19.7-25.2-49.9-39.9-81.9-39.9H72z"],"face-smile":[512,512,[128578,"smile"],"f118","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm177.6 62.1C192.8 334.5 218.8 352 256 352s63.2-17.5 78.4-33.9c9-9.7 24.2-10.4 33.9-1.4s10.4 24.2 1.4 33.9c-22 23.8-60 49.4-113.6 49.4s-91.7-25.5-113.6-49.4c-9-9.7-8.4-24.9 1.4-33.9s24.9-8.4 33.9 1.4zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"hand-peace":[512,512,[9996],"f25b","M250.8 1.4c-35.2-3.7-66.6 21.8-70.3 57L174 119 156.7 69.6C145 36.3 108.4 18.8 75.1 30.5S24.2 78.8 35.9 112.1L88.7 262.2C73.5 276.7 64 297.3 64 320v0 24c0 92.8 75.2 168 168 168h48c92.8 0 168-75.2 168-168V272 256 224c0-35.3-28.7-64-64-64c-7.9 0-15.4 1.4-22.4 4c-10.4-21.3-32.3-36-57.6-36c-.7 0-1.5 0-2.2 0l5.9-56.3c3.7-35.2-21.8-66.6-57-70.3zm-.2 155.4C243.9 166.9 240 179 240 192v48c0 .7 0 1.4 0 2c-5.1-1.3-10.5-2-16-2h-7.4l-5.4-15.3 17-161.3c.9-8.8 8.8-15.2 17.6-14.2s15.2 8.8 14.2 17.6l-9.5 90.1zM111.4 85.6L165.7 240H144c-4 0-8 .3-11.9 .9L81.2 96.2c-2.9-8.3 1.5-17.5 9.8-20.4s17.5 1.5 20.4 9.8zM288 192c0-8.8 7.2-16 16-16s16 7.2 16 16v32 16c0 8.8-7.2 16-16 16s-16-7.2-16-16V192zm38.4 108c10.4 21.3 32.3 36 57.6 36c5.5 0 10.9-.7 16-2v10c0 66.3-53.7 120-120 120H232c-66.3 0-120-53.7-120-120l0-24 0 0c0-17.7 14.3-32 32-32h80c8.8 0 16 7.2 16 16s-7.2 16-16 16H184c-13.3 0-24 10.7-24 24s10.7 24 24 24h40c35.3 0 64-28.7 64-64c0-.7 0-1.4 0-2c5.1 1.3 10.5 2 16 2c7.9 0 15.4-1.4 22.4-4zM400 272c0 8.8-7.2 16-16 16s-16-7.2-16-16V240 224c0-8.8 7.2-16 16-16s16 7.2 16 16v32 16z"],"face-grin-hearts":[512,512,[128525,"grin-hearts"],"f584","M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm349.5 52.4c18.7-4.4 35.9 12 25.5 28.1C350.4 374.6 306.3 400 255.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5zM215.3 137.1c17.8 4.8 28.4 23.1 23.6 40.8l-17.4 65c-2.3 8.5-11.1 13.6-19.6 11.3l-65.1-17.4c-17.8-4.8-28.4-23.1-23.6-40.8s23.1-28.4 40.8-23.6l16.1 4.3 4.3-16.1c4.8-17.8 23.1-28.4 40.8-23.6zm122.3 23.6l4.3 16.1 16.1-4.3c17.8-4.8 36.1 5.8 40.8 23.6s-5.8 36.1-23.6 40.8l-65.1 17.4c-8.5 2.3-17.3-2.8-19.6-11.3l-17.4-65c-4.8-17.8 5.8-36.1 23.6-40.8s36.1 5.8 40.9 23.6z"],building:[384,512,[127970,61687],"f1ad","M64 48c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16h80V400c0-26.5 21.5-48 48-48s48 21.5 48 48v64h80c8.8 0 16-7.2 16-16V64c0-8.8-7.2-16-16-16H64zM0 64C0 28.7 28.7 0 64 0H320c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zm88 40c0-8.8 7.2-16 16-16h48c8.8 0 16 7.2 16 16v48c0 8.8-7.2 16-16 16H104c-8.8 0-16-7.2-16-16V104zM232 88h48c8.8 0 16 7.2 16 16v48c0 8.8-7.2 16-16 16H232c-8.8 0-16-7.2-16-16V104c0-8.8 7.2-16 16-16zM88 232c0-8.8 7.2-16 16-16h48c8.8 0 16 7.2 16 16v48c0 8.8-7.2 16-16 16H104c-8.8 0-16-7.2-16-16V232zm144-16h48c8.8 0 16 7.2 16 16v48c0 8.8-7.2 16-16 16H232c-8.8 0-16-7.2-16-16V232c0-8.8 7.2-16 16-16z"],"face-grin-beam-sweat":[512,512,[128517,"grin-beam-sweat"],"f583","M476.8 126.3C497.1 120.8 512 102.7 512 81c0-20-28.6-60.4-41.6-77.7c-3.2-4.4-9.6-4.4-12.8 0c-9.5 12.6-27.1 37.2-36 57.5c-.3 .7-.6 1.4-.9 2.1C417.8 69.7 416 76 416 81c0 26 21.5 47 48 47c4.4 0 8.7-.6 12.8-1.7zM395.4 41.2C355.3 15.2 307.4 0 256 0C114.6 0 0 114.6 0 256S114.6 512 256 512s256-114.6 256-256c0-35.8-7.3-69.9-20.6-100.8c-8.6 3.1-17.8 4.8-27.4 4.8c-8.9 0-17.6-1.5-25.7-4.2C454.7 185.5 464 219.7 464 256c0 114.9-93.1 208-208 208S48 370.9 48 256S141.1 48 256 48c48.7 0 93.4 16.7 128.9 44.7c-.6-3.8-.9-7.7-.9-11.7c0-11.4 3.8-22.4 7.1-30.5c1.3-3.1 2.7-6.2 4.3-9.3zM375 336.5c10.4-16.1-6.8-32.5-25.5-28.1c-28.9 6.8-60.5 10.5-93.6 10.5s-64.7-3.7-93.6-10.5c-18.7-4.4-35.9 12-25.5 28.1c24.6 38.1 68.7 63.5 119.1 63.5s94.5-25.4 119.1-63.5zM217.6 228.8l0 0 0 0 0 0c2.1 2.8 5.7 3.9 8.9 2.8s5.5-4.1 5.5-7.6c0-17.9-6.7-35.6-16.6-48.8c-9.8-13-23.9-23.2-39.4-23.2s-29.6 10.2-39.4 23.2C126.7 188.4 120 206.1 120 224c0 3.4 2.2 6.5 5.5 7.6s6.9 0 8.9-2.8l0 0 0 0 0 0 .2-.2c.2-.2 .4-.5 .7-.9c.6-.8 1.6-2 2.8-3.4c2.5-2.8 6-6.6 10.2-10.3c8.8-7.8 18.8-14 27.7-14s18.9 6.2 27.7 14c4.2 3.7 7.7 7.5 10.2 10.3c1.2 1.4 2.2 2.6 2.8 3.4c.3 .4 .6 .7 .7 .9l.2 .2 0 0zm160 0l0 0 0 0c2.1 2.8 5.7 3.9 8.9 2.8s5.5-4.1 5.5-7.6c0-17.9-6.7-35.6-16.6-48.8c-9.8-13-23.9-23.2-39.4-23.2s-29.6 10.2-39.4 23.2C286.7 188.4 280 206.1 280 224c0 3.4 2.2 6.5 5.5 7.6s6.9 0 8.9-2.8l0 0 0 0 0 0 .2-.2c.2-.2 .4-.5 .7-.9c.6-.8 1.6-2 2.8-3.4c2.5-2.8 6-6.6 10.2-10.3c8.8-7.8 18.8-14 27.7-14s18.9 6.2 27.7 14c4.2 3.7 7.7 7.5 10.2 10.3c1.2 1.4 2.2 2.6 2.8 3.4c.3 .4 .6 .7 .7 .9l.2 .2 0 0 0 0z"],moon:[384,512,[127769,9214],"f186","M144.7 98.7c-21 34.1-33.1 74.3-33.1 117.3c0 98 62.8 181.4 150.4 211.7c-12.4 2.8-25.3 4.3-38.6 4.3C126.6 432 48 353.3 48 256c0-68.9 39.4-128.4 96.8-157.3zm62.1-66C91.1 41.2 0 137.9 0 256C0 379.7 100 480 223.5 480c47.8 0 92-15 128.4-40.6c1.9-1.3 3.7-2.7 5.5-4c4.8-3.6 9.4-7.4 13.9-11.4c2.7-2.4 5.3-4.8 7.9-7.3c5-4.9 6.3-12.5 3.1-18.7s-10.1-9.7-17-8.5c-3.7 .6-7.4 1.2-11.1 1.6c-5 .5-10.1 .9-15.3 1c-1.2 0-2.5 0-3.7 0c-.1 0-.2 0-.3 0c-96.8-.2-175.2-78.9-175.2-176c0-54.8 24.9-103.7 64.1-136c1-.9 2.1-1.7 3.2-2.6c4-3.2 8.2-6.2 12.5-9c3.1-2 6.3-4 9.6-5.8c6.1-3.5 9.2-10.5 7.7-17.3s-7.3-11.9-14.3-12.5c-3.6-.3-7.1-.5-10.7-.6c-2.7-.1-5.5-.1-8.2-.1c-3.3 0-6.5 .1-9.8 .2c-2.3 .1-4.6 .2-6.9 .4z"],calendar:[448,512,[128197,128198],"f133","M152 24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H64C28.7 64 0 92.7 0 128v16 48V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V192 144 128c0-35.3-28.7-64-64-64H344V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H152V24zM48 192H400V448c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V192z"],"face-grin-tongue-wink":[512,512,[128540,"grin-tongue-wink"],"f58b","M348.3 442.4c2.4-8.4 3.7-17.3 3.7-26.4V363.5c8.8-8 16.6-17.1 23-27c10.4-16.1-6.8-32.5-25.5-28.1c-28.9 6.8-60.5 10.5-93.6 10.5s-64.7-3.7-93.6-10.5c-18.7-4.4-35.9 12-25.5 28.1c6.5 10 14.3 19.1 23.1 27.1V416c0 9.2 1.3 18 3.7 26.4C95.1 408.4 48 337.7 48 256C48 141.1 141.1 48 256 48s208 93.1 208 208c0 81.7-47.1 152.4-115.7 186.4zM256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM159.6 220c10.6 0 19.9 3.8 25.4 9.7c7.6 8.1 20.2 8.5 28.3 .9s8.5-20.2 .9-28.3C199.7 186.8 179 180 159.6 180s-40.1 6.8-54.6 22.3c-7.6 8.1-7.1 20.7 .9 28.3s20.7 7.1 28.3-.9c5.5-5.8 14.8-9.7 25.4-9.7zm176.7 12a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm-.4-72a48 48 0 1 1 0 96 48 48 0 1 1 0-96zm0 128a80 80 0 1 0 0-160 80 80 0 1 0 0 160zM320 416c0 35.3-28.7 64-64 64s-64-28.7-64-64V378.6c0-14.7 11.9-26.6 26.6-26.6h2c11.3 0 21.1 7.9 23.6 18.9c2.8 12.6 20.8 12.6 23.6 0c2.5-11.1 12.3-18.9 23.6-18.9h2c14.7 0 26.6 11.9 26.6 26.6V416z"],clone:[512,512,[],"f24d","M64 464H288c8.8 0 16-7.2 16-16V384h48v64c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V224c0-35.3 28.7-64 64-64h64v48H64c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16zM224 304H448c8.8 0 16-7.2 16-16V64c0-8.8-7.2-16-16-16H224c-8.8 0-16 7.2-16 16V288c0 8.8 7.2 16 16 16zm-64-16V64c0-35.3 28.7-64 64-64H448c35.3 0 64 28.7 64 64V288c0 35.3-28.7 64-64 64H224c-35.3 0-64-28.7-64-64z"],"face-angry":[512,512,[128544,"angry"],"f556","M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm72.4-118.5c9.7-9 10.2-24.2 1.2-33.9C315.3 344.3 290.6 328 256 328s-59.3 16.3-73.5 31.6c-9 9.7-8.5 24.9 1.2 33.9s24.9 8.5 33.9-1.2c7.4-7.9 20-16.4 38.5-16.4s31.1 8.5 38.5 16.4c9 9.7 24.2 10.2 33.9 1.2zM176.4 272c17.7 0 32-14.3 32-32c0-1.5-.1-3-.3-4.4l10.9 3.6c8.4 2.8 17.4-1.7 20.2-10.1s-1.7-17.4-10.1-20.2l-96-32c-8.4-2.8-17.4 1.7-20.2 10.1s1.7 17.4 10.1 20.2l30.7 10.2c-5.8 5.8-9.3 13.8-9.3 22.6c0 17.7 14.3 32 32 32zm192-32c0-8.9-3.6-17-9.5-22.8l30.2-10.1c8.4-2.8 12.9-11.9 10.1-20.2s-11.9-12.9-20.2-10.1l-96 32c-8.4 2.8-12.9 11.9-10.1 20.2s11.9 12.9 20.2 10.1l11.7-3.9c-.2 1.5-.3 3.1-.3 4.7c0 17.7 14.3 32 32 32s32-14.3 32-32z"],"rectangle-xmark":[512,512,[62164,"rectangle-times","times-rectangle","window-close"],"f410","M64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H448c8.8 0 16-7.2 16-16V96c0-8.8-7.2-16-16-16H64zM0 96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm175 79c9.4-9.4 24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9z"],"paper-plane":[512,512,[61913],"f1d8","M16.1 260.2c-22.6 12.9-20.5 47.3 3.6 57.3L160 376V479.3c0 18.1 14.6 32.7 32.7 32.7c9.7 0 18.9-4.3 25.1-11.8l62-74.3 123.9 51.6c18.9 7.9 40.8-4.5 43.9-24.7l64-416c1.9-12.1-3.4-24.3-13.5-31.2s-23.3-7.5-34-1.4l-448 256zm52.1 25.5L409.7 90.6 190.1 336l1.2 1L68.2 285.7zM403.3 425.4L236.7 355.9 450.8 116.6 403.3 425.4z"],"life-ring":[512,512,[],"f1cd","M385.1 419.1C349.7 447.2 304.8 464 256 464s-93.7-16.8-129.1-44.9l80.4-80.4c14.3 8.4 31 13.3 48.8 13.3s34.5-4.8 48.8-13.3l80.4 80.4zm68.1 .2C489.9 374.9 512 318.1 512 256s-22.1-118.9-58.8-163.3L465 81c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L419.3 58.8C374.9 22.1 318.1 0 256 0S137.1 22.1 92.7 58.8L81 47c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9L58.8 92.7C22.1 137.1 0 193.9 0 256s22.1 118.9 58.8 163.3L47 431c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l11.8-11.8C137.1 489.9 193.9 512 256 512s118.9-22.1 163.3-58.8L431 465c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-11.8-11.8zm-34.1-34.1l-80.4-80.4c8.4-14.3 13.3-31 13.3-48.8s-4.8-34.5-13.3-48.8l80.4-80.4C447.2 162.3 464 207.2 464 256s-16.8 93.7-44.9 129.1zM385.1 92.9l-80.4 80.4c-14.3-8.4-31-13.3-48.8-13.3s-34.5 4.8-48.8 13.3L126.9 92.9C162.3 64.8 207.2 48 256 48s93.7 16.8 129.1 44.9zM173.3 304.8L92.9 385.1C64.8 349.7 48 304.8 48 256s16.8-93.7 44.9-129.1l80.4 80.4c-8.4 14.3-13.3 31-13.3 48.8s4.8 34.5 13.3 48.8zM208 256a48 48 0 1 1 96 0 48 48 0 1 1 -96 0z"],"face-grimace":[512,512,[128556,"grimace"],"f57f","M256 48a208 208 0 1 0 0 416 208 208 0 1 0 0-416zM512 256A256 256 0 1 1 0 256a256 256 0 1 1 512 0zM168 320c-13.3 0-24 10.7-24 24s10.7 24 24 24h8V320h-8zm40 48h32V320H208v48zm96 0V320H272v48h32zm32 0h8c13.3 0 24-10.7 24-24s-10.7-24-24-24h-8v48zM168 288H344c30.9 0 56 25.1 56 56s-25.1 56-56 56H168c-30.9 0-56-25.1-56-56s25.1-56 56-56zm-23.6-80a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"calendar-minus":[448,512,[],"f272","M128 0c13.3 0 24 10.7 24 24V64H296V24c0-13.3 10.7-24 24-24s24 10.7 24 24V64h40c35.3 0 64 28.7 64 64v16 48V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V192 144 128C0 92.7 28.7 64 64 64h40V24c0-13.3 10.7-24 24-24zM400 192H48V448c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V192zM296 352H152c-13.3 0-24-10.7-24-24s10.7-24 24-24H296c13.3 0 24 10.7 24 24s-10.7 24-24 24z"],"circle-xmark":[512,512,[61532,"times-circle","xmark-circle"],"f057","M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM175 175c-9.4 9.4-9.4 24.6 0 33.9l47 47-47 47c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l47-47 47 47c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-47-47 47-47c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-47 47-47-47c-9.4-9.4-24.6-9.4-33.9 0z"],"thumbs-up":[512,512,[128077,61575],"f164","M323.8 34.8c-38.2-10.9-78.1 11.2-89 49.4l-5.7 20c-3.7 13-10.4 25-19.5 35l-51.3 56.4c-8.9 9.8-8.2 25 1.6 33.9s25 8.2 33.9-1.6l51.3-56.4c14.1-15.5 24.4-34 30.1-54.1l5.7-20c3.6-12.7 16.9-20.1 29.7-16.5s20.1 16.9 16.5 29.7l-5.7 20c-5.7 19.9-14.7 38.7-26.6 55.5c-5.2 7.3-5.8 16.9-1.7 24.9s12.3 13 21.3 13L448 224c8.8 0 16 7.2 16 16c0 6.8-4.3 12.7-10.4 15c-7.4 2.8-13 9-14.9 16.7s.1 15.8 5.3 21.7c2.5 2.8 4 6.5 4 10.6c0 7.8-5.6 14.3-13 15.7c-8.2 1.6-15.1 7.3-18 15.2s-1.6 16.7 3.6 23.3c2.1 2.7 3.4 6.1 3.4 9.9c0 6.7-4.2 12.6-10.2 14.9c-11.5 4.5-17.7 16.9-14.4 28.8c.4 1.3 .6 2.8 .6 4.3c0 8.8-7.2 16-16 16H286.5c-12.6 0-25-3.7-35.5-10.7l-61.7-41.1c-11-7.4-25.9-4.4-33.3 6.7s-4.4 25.9 6.7 33.3l61.7 41.1c18.4 12.3 40 18.8 62.1 18.8H384c34.7 0 62.9-27.6 64-62c14.6-11.7 24-29.7 24-50c0-4.5-.5-8.8-1.3-13c15.4-11.7 25.3-30.2 25.3-51c0-6.5-1-12.8-2.8-18.7C504.8 273.7 512 257.7 512 240c0-35.3-28.6-64-64-64l-92.3 0c4.7-10.4 8.7-21.2 11.8-32.2l5.7-20c10.9-38.2-11.2-78.1-49.4-89zM32 192c-17.7 0-32 14.3-32 32V448c0 17.7 14.3 32 32 32H96c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32H32z"],"window-minimize":[512,512,[128469],"f2d1","M24 432c-13.3 0-24 10.7-24 24s10.7 24 24 24H488c13.3 0 24-10.7 24-24s-10.7-24-24-24H24z"],"square-full":[512,512,[128997,128998,128999,129e3,129001,129002,129003,11035,11036],"f45c","M464 48V464H48V48H464zM48 0H0V48 464v48H48 464h48V464 48 0H464 48z"],"note-sticky":[448,512,[62026,"sticky-note"],"f249","M64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H288V352c0-17.7 14.3-32 32-32h80V96c0-8.8-7.2-16-16-16H64zM288 480H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V320v5.5c0 17-6.7 33.3-18.7 45.3l-90.5 90.5c-12 12-28.3 18.7-45.3 18.7H288z"],"face-sad-tear":[512,512,[128546,"sad-tear"],"f5b4","M175.9 448c-35-.1-65.5-22.6-76-54.6C67.6 356.8 48 308.7 48 256C48 141.1 141.1 48 256 48s208 93.1 208 208s-93.1 208-208 208c-28.4 0-55.5-5.7-80.1-16zM0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zM128 369c0 26 21.5 47 48 47s48-21 48-47c0-20-28.4-60.4-41.6-77.7c-3.2-4.4-9.6-4.4-12.8 0C156.6 308.6 128 349 128 369zm128-65c-13.3 0-24 10.7-24 24s10.7 24 24 24c30.7 0 58.7 11.5 80 30.6c9.9 8.8 25 8 33.9-1.9s8-25-1.9-33.9C338.3 320.2 299 304 256 304zm47.6-96a32 32 0 1 0 64 0 32 32 0 1 0 -64 0zm-128 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"hand-point-left":[512,512,[],"f0a5","M64 128l177.6 0c-1 5.2-1.6 10.5-1.6 16l0 16-32 0L64 160c-8.8 0-16-7.2-16-16s7.2-16 16-16zm224 16c0-17.7 14.3-32 32-32c0 0 0 0 0 0l24 0c66.3 0 120 53.7 120 120l0 48c0 52.5-33.7 97.1-80.7 113.4c.5-3.1 .7-6.2 .7-9.4c0-20-9.2-37.9-23.6-49.7c4.9-9 7.6-19.4 7.6-30.3c0-15.1-5.3-29-14-40c8.8-11 14-24.9 14-40l0-40c0-13.3-10.7-24-24-24s-24 10.7-24 24l0 40c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-40 0-40zm32-80s0 0 0 0c-18 0-34.6 6-48 16L64 80C28.7 80 0 108.7 0 144s28.7 64 64 64l82 0c-1.3 5.1-2 10.5-2 16c0 25.3 14.7 47.2 36 57.6c-2.6 7-4 14.5-4 22.4c0 20 9.2 37.9 23.6 49.7c-4.9 9-7.6 19.4-7.6 30.3c0 35.3 28.7 64 64 64l64 0 24 0c92.8 0 168-75.2 168-168l0-48c0-92.8-75.2-168-168-168l-24 0zM256 400c-8.8 0-16-7.2-16-16s7.2-16 16-16l48 0 16 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-64 0zM240 224c0 5.5 .7 10.9 2 16l-2 0-32 0c-8.8 0-16-7.2-16-16s7.2-16 16-16l32 0 0 16zm24 64l40 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-48 0-16 0c-8.8 0-16-7.2-16-16s7.2-16 16-16l24 0z"]};!function(c){try{for(var l=arguments.length,s=new Array(1<l?l-1:0),a=1;a<l;a++)s[a-1]=arguments[a];c.apply(void 0,s)}catch(c){if(!M)throw c}}((function(){d("far",p),d("fa-regular",p)}))}(),function(){"use strict";var c={},l={};try{"undefined"!=typeof window&&(c=window),"undefined"!=typeof document&&(l=document)}catch(c){}var s=void 0===(v=(c.navigator||{}).userAgent)?"":v,a=c,z=l;function e(c,l){var s,a=Object.keys(c);return Object.getOwnPropertySymbols&&(s=Object.getOwnPropertySymbols(c),l&&(s=s.filter((function(l){return Object.getOwnPropertyDescriptor(c,l).enumerable}))),a.push.apply(a,s)),a}function H(c){for(var l=1;l<arguments.length;l++){var s=null!=arguments[l]?arguments[l]:{};l%2?e(Object(s),!0).forEach((function(l){t(c,l,s[l])})):Object.getOwnPropertyDescriptors?Object.defineProperties(c,Object.getOwnPropertyDescriptors(s)):e(Object(s)).forEach((function(l){Object.defineProperty(c,l,Object.getOwnPropertyDescriptor(s,l))}))}return c}function t(c,l,s){return l in c?Object.defineProperty(c,l,{value:s,enumerable:!0,configurable:!0,writable:!0}):c[l]=s,c}function V(c,l){(null==l||l>c.length)&&(l=c.length);for(var s=0,a=new Array(l);s<l;s++)a[s]=c[s];return a}a.document,z.documentElement&&z.head&&"function"==typeof z.addEventListener&&z.createElement,~s.indexOf("MSIE")||s.indexOf("Trident/");var r="___FONT_AWESOME___",M=function(){try{return!0}catch(c){return!1}}(),h="classic",n="sharp",i=[h,n];function m(c){return new Proxy(c,{get:function(c,l){return l in c?c[l]:c[h]}})}m((t(f={},h,{fa:"solid",fas:"solid","fa-solid":"solid",far:"regular","fa-regular":"regular",fal:"light","fa-light":"light",fat:"thin","fa-thin":"thin",fad:"duotone","fa-duotone":"duotone",fab:"brands","fa-brands":"brands",fak:"kit",fakd:"kit","fa-kit":"kit","fa-kit-duotone":"kit"}),t(f,n,{fa:"solid",fass:"solid","fa-solid":"solid",fasr:"regular","fa-regular":"regular",fasl:"light","fa-light":"light",fast:"thin","fa-thin":"thin"}),f));var o=m((t(C={},h,{solid:"fas",regular:"far",light:"fal",thin:"fat",duotone:"fad",brands:"fab",kit:"fak"}),t(C,n,{solid:"fass",regular:"fasr",light:"fasl",thin:"fast"}),C)),f=(s=(m((t(v={},h,{fab:"fa-brands",fad:"fa-duotone",fak:"fa-kit",fal:"fa-light",far:"fa-regular",fas:"fa-solid",fat:"fa-thin"}),t(v,n,{fass:"fa-solid",fasr:"fa-regular",fasl:"fa-light",fast:"fa-thin"}),v)),m((t(c={},h,{"fa-brands":"fab","fa-duotone":"fad","fa-kit":"fak","fa-light":"fal","fa-regular":"far","fa-solid":"fas","fa-thin":"fat"}),t(c,n,{"fa-solid":"fass","fa-regular":"fasr","fa-light":"fasl","fa-thin":"fast"}),c)),m((t(l={},h,{900:"fas",400:"far",normal:"far",300:"fal",100:"fat"}),t(l,n,{900:"fass",400:"fasr",300:"fasl",100:"fast"}),l)),z=[1,2,3,4,5,6,7,8,9,10]).concat([11,12,13,14,15,16,17,18,19,20]),"duotone-group"),C="swap-opacity",v="primary";c="secondary",l=new Set,Object.keys(o[h]).map(l.add.bind(l)),Object.keys(o[n]).map(l.add.bind(l)),[].concat(i,function(c){if(Array.isArray(c))return V(c)}(l)||function(c){if("undefined"!=typeof Symbol&&null!=c[Symbol.iterator]||null!=c["@@iterator"])return Array.from(c)}(l)||function(c,l){if(c){if("string"==typeof c)return V(c,l);var s=Object.prototype.toString.call(c).slice(8,-1);return"Map"===(s="Object"===s&&c.constructor?c.constructor.name:s)||"Set"===s?Array.from(c):"Arguments"===s||/^(?:Ui|I)nt(?:8|16|32)(?:Clamped)?Array$/.test(s)?V(c,l):void 0}}(l)||function(){throw new TypeError("Invalid attempt to spread non-iterable instance.\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method.")}(),["2xs","xs","sm","lg","xl","2xl","beat","border","fade","beat-fade","bounce","flip-both","flip-horizontal","flip-vertical","flip","fw","inverse","layers-counter","layers-text","layers","li","pull-left","pull-right","pulse","rotate-180","rotate-270","rotate-90","rotate-by","shake","spin-pulse","spin-reverse","spin","stack-1x","stack-2x","stack","ul",f,C,v,c]).concat(z.map((function(c){return"".concat(c,"x")}))).concat(s.map((function(c){return"w-".concat(c)}))),(a=a||{})[r]||(a[r]={}),a[r].styles||(a[r].styles={}),a[r].hooks||(a[r].hooks={}),a[r].shims||(a[r].shims=[]);var L=a[r];function u(c){return Object.keys(c).reduce((function(l,s){var a=c[s];return a.icon?l[a.iconName]=a.icon:l[s]=a,l}),{})}function d(c,l,s){s=void 0!==(a=(2<arguments.length&&void 0!==s?s:{}).skipHooks)&&a;var a=u(l);"function"!=typeof L.hooks.addPack||s?L.styles[c]=H(H({},L.styles[c]||{}),a):L.hooks.addPack(c,u(l)),"fas"===c&&d("fa",l)}var p={0:[320,512,[],"30","M0 192C0 103.6 71.6 32 160 32s160 71.6 160 160V320c0 88.4-71.6 160-160 160S0 408.4 0 320V192zM160 96c-53 0-96 43-96 96V320c0 53 43 96 96 96s96-43 96-96V192c0-53-43-96-96-96z"],1:[256,512,[],"31","M160 64c0-11.8-6.5-22.6-16.9-28.2s-23-5-32.8 1.6l-96 64C-.5 111.2-4.4 131 5.4 145.8s29.7 18.7 44.4 8.9L96 123.8V416H32c-17.7 0-32 14.3-32 32s14.3 32 32 32h96 96c17.7 0 32-14.3 32-32s-14.3-32-32-32H160V64z"],2:[320,512,[],"32","M142.9 96c-21.5 0-42.2 8.5-57.4 23.8L54.6 150.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L40.2 74.5C67.5 47.3 104.4 32 142.9 32C223 32 288 97 288 177.1c0 38.5-15.3 75.4-42.5 102.6L109.3 416H288c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-12.9 0-24.6-7.8-29.6-19.8s-2.2-25.7 6.9-34.9L200.2 234.5c15.2-15.2 23.8-35.9 23.8-57.4c0-44.8-36.3-81.1-81.1-81.1z"],3:[320,512,[],"33","M0 64C0 46.3 14.3 32 32 32H272c13.2 0 25 8.1 29.8 20.4s1.5 26.3-8.2 35.2L162.3 208H184c75.1 0 136 60.9 136 136s-60.9 136-136 136H105.4C63 480 24.2 456 5.3 418.1l-1.9-3.8c-7.9-15.8-1.5-35 14.3-42.9s35-1.5 42.9 14.3l1.9 3.8c8.1 16.3 24.8 26.5 42.9 26.5H184c39.8 0 72-32.2 72-72s-32.2-72-72-72H80c-13.2 0-25-8.1-29.8-20.4s-1.5-26.3 8.2-35.2L189.7 96H32C14.3 96 0 81.7 0 64z"],4:[384,512,[],"34","M189 77.6c7.5-16 .7-35.1-15.3-42.6s-35.1-.7-42.6 15.3L3 322.4c-4.7 9.9-3.9 21.5 1.9 30.8S21 368 32 368H256v80c0 17.7 14.3 32 32 32s32-14.3 32-32V368h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H320V160c0-17.7-14.3-32-32-32s-32 14.3-32 32V304H82.4L189 77.6z"],5:[320,512,[],"35","M32.5 58.3C35.3 43.1 48.5 32 64 32H256c17.7 0 32 14.3 32 32s-14.3 32-32 32H90.7L70.3 208H184c75.1 0 136 60.9 136 136s-60.9 136-136 136H100.5c-39.4 0-75.4-22.3-93-57.5l-4.1-8.2c-7.9-15.8-1.5-35 14.3-42.9s35-1.5 42.9 14.3l4.1 8.2c6.8 13.6 20.6 22.1 35.8 22.1H184c39.8 0 72-32.2 72-72s-32.2-72-72-72H32c-9.5 0-18.5-4.2-24.6-11.5s-8.6-16.9-6.9-26.2l32-176z"],6:[320,512,[],"36","M232.4 84.7c11.4-13.5 9.7-33.7-3.8-45.1s-33.7-9.7-45.1 3.8L38.6 214.7C14.7 242.9 1.1 278.4 .1 315.2c0 1.4-.1 2.9-.1 4.3c0 .2 0 .3 0 .5c0 88.4 71.6 160 160 160s160-71.6 160-160c0-85.5-67.1-155.4-151.5-159.8l63.9-75.6zM256 320A96 96 0 1 1 64 320a96 96 0 1 1 192 0z"],7:[320,512,[],"37","M0 64C0 46.3 14.3 32 32 32H288c11.5 0 22 6.1 27.7 16.1s5.7 22.2-.1 32.1l-224 384c-8.9 15.3-28.5 20.4-43.8 11.5s-20.4-28.5-11.5-43.8L232.3 96H32C14.3 96 0 81.7 0 64z"],8:[320,512,[],"38","M304 160c0-70.7-57.3-128-128-128H144C73.3 32 16 89.3 16 160c0 34.6 13.7 66 36 89C20.5 272.3 0 309.8 0 352c0 70.7 57.3 128 128 128h64c70.7 0 128-57.3 128-128c0-42.2-20.5-79.7-52-103c22.3-23 36-54.4 36-89zM176.1 288H192c35.3 0 64 28.7 64 64s-28.7 64-64 64H128c-35.3 0-64-28.7-64-64s28.7-64 64-64h15.9c0 0 .1 0 .1 0h32c0 0 .1 0 .1 0zm0-64c0 0 0 0 0 0H144c0 0 0 0 0 0c-35.3 0-64-28.7-64-64c0-35.3 28.7-64 64-64h32c35.3 0 64 28.7 64 64c0 35.3-28.6 64-64 64z"],9:[320,512,[],"39","M64 192a96 96 0 1 0 192 0A96 96 0 1 0 64 192zm87.5 159.8C67.1 347.4 0 277.5 0 192C0 103.6 71.6 32 160 32s160 71.6 160 160c0 2.6-.1 5.3-.2 7.9c-1.7 35.7-15.2 70-38.4 97.4l-145 171.4c-11.4 13.5-31.6 15.2-45.1 3.8s-15.2-31.6-3.8-45.1l63.9-75.6z"],"fill-drip":[576,512,[],"f576","M41.4 9.4C53.9-3.1 74.1-3.1 86.6 9.4L168 90.7l53.1-53.1c28.1-28.1 73.7-28.1 101.8 0L474.3 189.1c28.1 28.1 28.1 73.7 0 101.8L283.9 481.4c-37.5 37.5-98.3 37.5-135.8 0L30.6 363.9c-37.5-37.5-37.5-98.3 0-135.8L122.7 136 41.4 54.6c-12.5-12.5-12.5-32.8 0-45.3zm176 221.3L168 181.3 75.9 273.4c-4.2 4.2-7 9.3-8.4 14.6H386.7l42.3-42.3c3.1-3.1 3.1-8.2 0-11.3L277.7 82.9c-3.1-3.1-8.2-3.1-11.3 0L213.3 136l49.4 49.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0zM512 512c-35.3 0-64-28.7-64-64c0-25.2 32.6-79.6 51.2-108.7c6-9.4 19.5-9.4 25.5 0C543.4 368.4 576 422.8 576 448c0 35.3-28.7 64-64 64z"],"arrows-to-circle":[640,512,[],"e4bd","M9.4 9.4C21.9-3.1 42.1-3.1 54.6 9.4L160 114.7V96c0-17.7 14.3-32 32-32s32 14.3 32 32v96c0 4.3-.9 8.5-2.4 12.2c-1.6 3.7-3.8 7.3-6.9 10.3l-.1 .1c-3.1 3-6.6 5.3-10.3 6.9c-3.8 1.6-7.9 2.4-12.2 2.4H96c-17.7 0-32-14.3-32-32s14.3-32 32-32h18.7L9.4 54.6C-3.1 42.1-3.1 21.9 9.4 9.4zM256 256a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM114.7 352H96c-17.7 0-32-14.3-32-32s14.3-32 32-32h96 0l.1 0c8.8 0 16.7 3.6 22.5 9.3l.1 .1c3 3.1 5.3 6.6 6.9 10.3c1.6 3.8 2.4 7.9 2.4 12.2v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V397.3L54.6 502.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L114.7 352zM416 96c0-17.7 14.3-32 32-32s32 14.3 32 32v18.7L585.4 9.4c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3L525.3 160H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H448c-8.8 0-16.8-3.6-22.6-9.3l-.1-.1c-3-3.1-5.3-6.6-6.9-10.3s-2.4-7.8-2.4-12.2l0-.1v0V96zM525.3 352L630.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L480 397.3V416c0 17.7-14.3 32-32 32s-32-14.3-32-32V320v0c0 0 0-.1 0-.1c0-4.3 .9-8.4 2.4-12.2c1.6-3.8 3.9-7.3 6.9-10.4c5.8-5.8 13.7-9.3 22.5-9.4c0 0 .1 0 .1 0h0 96c17.7 0 32 14.3 32 32s-14.3 32-32 32H525.3z"],"circle-chevron-right":[512,512,["chevron-circle-right"],"f138","M0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zM241 377c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l87-87-87-87c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0L345 239c9.4 9.4 9.4 24.6 0 33.9L241 377z"],at:[512,512,[61946],"40","M256 64C150 64 64 150 64 256s86 192 192 192c17.7 0 32 14.3 32 32s-14.3 32-32 32C114.6 512 0 397.4 0 256S114.6 0 256 0S512 114.6 512 256v32c0 53-43 96-96 96c-29.3 0-55.6-13.2-73.2-33.9C320 371.1 289.5 384 256 384c-70.7 0-128-57.3-128-128s57.3-128 128-128c27.9 0 53.7 8.9 74.7 24.1c5.7-5 13.1-8.1 21.3-8.1c17.7 0 32 14.3 32 32v80 32c0 17.7 14.3 32 32 32s32-14.3 32-32V256c0-106-86-192-192-192zm64 192a64 64 0 1 0 -128 0 64 64 0 1 0 128 0z"],"trash-can":[448,512,[61460,"trash-alt"],"f2ed","M135.2 17.7C140.6 6.8 151.7 0 163.8 0H284.2c12.1 0 23.2 6.8 28.6 17.7L320 32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 96 0 81.7 0 64S14.3 32 32 32h96l7.2-14.3zM32 128H416V448c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V128zm96 64c-8.8 0-16 7.2-16 16V432c0 8.8 7.2 16 16 16s16-7.2 16-16V208c0-8.8-7.2-16-16-16zm96 0c-8.8 0-16 7.2-16 16V432c0 8.8 7.2 16 16 16s16-7.2 16-16V208c0-8.8-7.2-16-16-16zm96 0c-8.8 0-16 7.2-16 16V432c0 8.8 7.2 16 16 16s16-7.2 16-16V208c0-8.8-7.2-16-16-16z"],"text-height":[576,512,[],"f034","M64 128V96h64l0 320H96c-17.7 0-32 14.3-32 32s14.3 32 32 32H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H192l0-320h64v32c0 17.7 14.3 32 32 32s32-14.3 32-32V80c0-26.5-21.5-48-48-48H160 48C21.5 32 0 53.5 0 80v48c0 17.7 14.3 32 32 32s32-14.3 32-32zM502.6 41.4c-12.5-12.5-32.8-12.5-45.3 0l-64 64c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8h32V352H416c-12.9 0-24.6 7.8-29.6 19.8s-2.2 25.7 6.9 34.9l64 64c12.5 12.5 32.8 12.5 45.3 0l64-64c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8H512V160h32c12.9 0 24.6-7.8 29.6-19.8s2.2-25.7-6.9-34.9l-64-64z"],"user-xmark":[640,512,["user-times"],"f235","M96 128a128 128 0 1 1 256 0A128 128 0 1 1 96 128zM0 482.3C0 383.8 79.8 304 178.3 304h91.4C368.2 304 448 383.8 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3zM471 143c9.4-9.4 24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9z"],stethoscope:[576,512,[129658],"f0f1","M142.4 21.9c5.6 16.8-3.5 34.9-20.2 40.5L96 71.1V192c0 53 43 96 96 96s96-43 96-96V71.1l-26.1-8.7c-16.8-5.6-25.8-23.7-20.2-40.5s23.7-25.8 40.5-20.2l26.1 8.7C334.4 19.1 352 43.5 352 71.1V192c0 77.2-54.6 141.6-127.3 156.7C231 404.6 278.4 448 336 448c61.9 0 112-50.1 112-112V265.3c-28.3-12.3-48-40.5-48-73.3c0-44.2 35.8-80 80-80s80 35.8 80 80c0 32.8-19.7 61-48 73.3V336c0 97.2-78.8 176-176 176c-92.9 0-168.9-71.9-175.5-163.1C87.2 334.2 32 269.6 32 192V71.1c0-27.5 17.6-52 43.8-60.7l26.1-8.7c16.8-5.6 34.9 3.5 40.5 20.2zM480 224a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],message:[512,512,["comment-alt"],"f27a","M64 0C28.7 0 0 28.7 0 64V352c0 35.3 28.7 64 64 64h96v80c0 6.1 3.4 11.6 8.8 14.3s11.9 2.1 16.8-1.5L309.3 416H448c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64z"],info:[192,512,[],"f129","M48 80a48 48 0 1 1 96 0A48 48 0 1 1 48 80zM0 224c0-17.7 14.3-32 32-32H96c17.7 0 32 14.3 32 32V448h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H64V256H32c-17.7 0-32-14.3-32-32z"],"down-left-and-up-right-to-center":[512,512,["compress-alt"],"f422","M439 7c9.4-9.4 24.6-9.4 33.9 0l32 32c9.4 9.4 9.4 24.6 0 33.9l-87 87 39 39c6.9 6.9 8.9 17.2 5.2 26.2s-12.5 14.8-22.2 14.8H296c-13.3 0-24-10.7-24-24V72c0-9.7 5.8-18.5 14.8-22.2s19.3-1.7 26.2 5.2l39 39L439 7zM72 272H216c13.3 0 24 10.7 24 24V440c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-39-39L73 505c-9.4 9.4-24.6 9.4-33.9 0L7 473c-9.4-9.4-9.4-24.6 0-33.9l87-87L55 313c-6.9-6.9-8.9-17.2-5.2-26.2s12.5-14.8 22.2-14.8z"],explosion:[576,512,[],"e4e9","M499.6 11.3c6.7-10.7 20.5-14.5 31.7-8.5s15.8 19.5 10.6 31L404.8 338.6c2.2 2.3 4.3 4.7 6.3 7.1l97.2-54.7c10.5-5.9 23.6-3.1 30.9 6.4s6.3 23-2.2 31.5l-87 87H378.5c-13.2-37.3-48.7-64-90.5-64s-77.4 26.7-90.5 64H117.8L42.3 363.7c-9.7-6.7-13.1-19.6-7.9-30.3s17.4-15.9 28.7-12.4l97.2 30.4c3-3.9 6.1-7.7 9.4-11.3L107.4 236.3c-6.1-10.1-3.9-23.1 5.1-30.7s22.2-7.5 31.1 .1L246 293.6c1.5-.4 3-.8 4.5-1.1l13.6-142.7c1.2-12.3 11.5-21.7 23.9-21.7s22.7 9.4 23.9 21.7l13.5 141.9L499.6 11.3zM64 448v0H512v0h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H64zM288 0c13.3 0 24 10.7 24 24V72c0 13.3-10.7 24-24 24s-24-10.7-24-24V24c0-13.3 10.7-24 24-24z"],"file-lines":[384,512,[128441,128462,61686,"file-alt","file-text"],"f15c","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM112 256H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],"wave-square":[640,512,[],"f83e","M128 64c0-17.7 14.3-32 32-32H320c17.7 0 32 14.3 32 32V416h96V256c0-17.7 14.3-32 32-32H608c17.7 0 32 14.3 32 32s-14.3 32-32 32H512V448c0 17.7-14.3 32-32 32H320c-17.7 0-32-14.3-32-32V96H192V256c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32h96V64z"],ring:[512,512,[],"f70b","M64 208c0 7.8 4.4 18.7 17.1 30.3C126.5 214.1 188.9 200 256 200s129.5 14.1 174.9 38.3C443.6 226.7 448 215.8 448 208c0-12.3-10.8-32-47.9-50.6C364.9 139.8 314 128 256 128s-108.9 11.8-144.1 29.4C74.8 176 64 195.7 64 208zm192 40c-47 0-89.3 7.6-122.9 19.7C166.3 280.2 208.8 288 256 288s89.7-7.8 122.9-20.3C345.3 255.6 303 248 256 248zM0 208c0-49.6 39.4-85.8 83.3-107.8C129.1 77.3 190.3 64 256 64s126.9 13.3 172.7 36.2c43.9 22 83.3 58.2 83.3 107.8v96c0 49.6-39.4 85.8-83.3 107.8C382.9 434.7 321.7 448 256 448s-126.9-13.3-172.7-36.2C39.4 389.8 0 353.6 0 304V208z"],"building-un":[384,512,[],"e4d9","M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM237.3 71.1l34.7 52V80c0-8.8 7.2-16 16-16s16 7.2 16 16v96c0 7.1-4.6 13.3-11.4 15.3s-14-.6-17.9-6.4l-34.7-52V176c0 8.8-7.2 16-16 16s-16-7.2-16-16V80c0-7.1 4.6-13.3 11.4-15.3s14 .6 17.9 6.4zM112 80v64c0 8.8 7.2 16 16 16s16-7.2 16-16V80c0-8.8 7.2-16 16-16s16 7.2 16 16v64c0 26.5-21.5 48-48 48s-48-21.5-48-48V80c0-8.8 7.2-16 16-16s16 7.2 16 16z"],"dice-three":[448,512,[9858],"f527","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm64 96a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm64 128a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm128 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"calendar-days":[448,512,["calendar-alt"],"f073","M128 0c17.7 0 32 14.3 32 32V64H288V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64h48c26.5 0 48 21.5 48 48v48H0V112C0 85.5 21.5 64 48 64H96V32c0-17.7 14.3-32 32-32zM0 192H448V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V192zm64 80v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H80c-8.8 0-16 7.2-16 16zm128 0v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H208c-8.8 0-16 7.2-16 16zm144-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H336zM64 400v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V400c0-8.8-7.2-16-16-16H80c-8.8 0-16 7.2-16 16zm144-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V400c0-8.8-7.2-16-16-16H208zm112 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V400c0-8.8-7.2-16-16-16H336c-8.8 0-16 7.2-16 16z"],"anchor-circle-check":[640,512,[],"e4aa","M320 96a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm21.1 80C367 158.8 384 129.4 384 96c0-53-43-96-96-96s-96 43-96 96c0 33.4 17 62.8 42.9 80H224c-17.7 0-32 14.3-32 32s14.3 32 32 32h32V448H208c-53 0-96-43-96-96v-6.1l7 7c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L97 263c-9.4-9.4-24.6-9.4-33.9 0L7 319c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l7-7V352c0 88.4 71.6 160 160 160h80 80c8.2 0 16.3-.6 24.2-1.8c-22.2-16.2-40.4-37.5-53-62.2H320V368 240h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H341.1zM640 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z"],"building-circle-arrow-right":[640,512,[],"e4d1","M0 48C0 21.5 21.5 0 48 0H336c26.5 0 48 21.5 48 48V232.2c-39.1 32.3-64 81.1-64 135.8c0 49.5 20.4 94.2 53.3 126.2C364.5 505.1 351.1 512 336 512H240V432c0-26.5-21.5-48-48-48s-48 21.5-48 48v80H48c-26.5 0-48-21.5-48-48V48zM80 224c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H80zm80 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H176c-8.8 0-16 7.2-16 16zm112-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H272zM64 112v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H80c-8.8 0-16 7.2-16 16zM176 96c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H176zm80 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16zm96 256a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm140.7-67.3c-6.2 6.2-6.2 16.4 0 22.6L521.4 352H432c-8.8 0-16 7.2-16 16s7.2 16 16 16h89.4l-28.7 28.7c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0l56-56c6.2-6.2 6.2-16.4 0-22.6l-56-56c-6.2-6.2-16.4-6.2-22.6 0z"],volleyball:[512,512,[127952,"volleyball-ball"],"f45f","M511.8 267.4c-26.1 8.7-53.4 13.8-81 15.1c9.2-105.3-31.5-204.2-103.2-272.4C434.1 41.1 512 139.5 512 256c0 3.8-.1 7.6-.2 11.4zm-3.9 34.7c-5.8 32-17.6 62-34.2 88.7c-97.5 48.5-217.7 42.6-311.9-24.5c23.7-36.2 55.4-67.7 94.5-91.8c79.9 43.2 170.1 50.8 251.6 27.6zm-236-55.5c-2.5-90.9-41.1-172.7-101.9-231.7C196.8 5.2 225.8 0 256 0c2.7 0 5.3 0 7.9 .1c90.8 60.2 145.7 167.2 134.7 282.3c-43.1-2.4-86.4-14.1-126.8-35.9zM138 28.8c20.6 18.3 38.7 39.4 53.7 62.6C95.9 136.1 30.6 220.8 7.3 316.9C2.5 297.4 0 277 0 256C0 157.2 56 71.5 138 28.8zm69.6 90.5c19.5 38.6 31 81.9 32.3 127.7C162.5 294.6 110.9 368.9 90.2 451C66 430.4 45.6 405.4 30.4 377.2c6.7-108.7 71.9-209.9 177.1-257.9zM256 512c-50.7 0-98-14.7-137.8-40.2c5.6-27 14.8-53.1 27.4-77.7C232.2 454.6 338.1 468.8 433 441c-46 44-108.3 71-177 71z"],"arrows-up-to-line":[576,512,[],"e4c2","M32 96l512 0c17.7 0 32-14.3 32-32s-14.3-32-32-32L32 32C14.3 32 0 46.3 0 64S14.3 96 32 96zM9.4 233.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L96 237.3 96 448c0 17.7 14.3 32 32 32s32-14.3 32-32l0-210.7 41.4 41.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-96-96c-12.5-12.5-32.8-12.5-45.3 0l-96 96zm320 45.3c12.5 12.5 32.8 12.5 45.3 0L416 237.3 416 448c0 17.7 14.3 32 32 32s32-14.3 32-32l0-210.7 41.4 41.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-96-96c-12.5-12.5-32.8-12.5-45.3 0l-96 96c-12.5 12.5-12.5 32.8 0 45.3z"],"sort-down":[320,512,["sort-desc"],"f0dd","M182.6 470.6c-12.5 12.5-32.8 12.5-45.3 0l-128-128c-9.2-9.2-11.9-22.9-6.9-34.9s16.6-19.8 29.6-19.8H288c12.9 0 24.6 7.8 29.6 19.8s2.2 25.7-6.9 34.9l-128 128z"],"circle-minus":[512,512,["minus-circle"],"f056","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM184 232H328c13.3 0 24 10.7 24 24s-10.7 24-24 24H184c-13.3 0-24-10.7-24-24s10.7-24 24-24z"],"door-open":[576,512,[],"f52b","M320 32c0-9.9-4.5-19.2-12.3-25.2S289.8-1.4 280.2 1l-179.9 45C79 51.3 64 70.5 64 92.5V448H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H96 288h32V480 32zM256 256c0 17.7-10.7 32-24 32s-24-14.3-24-32s10.7-32 24-32s24 14.3 24 32zm96-128h96V480c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H512V128c0-35.3-28.7-64-64-64H352v64z"],"right-from-bracket":[512,512,["sign-out-alt"],"f2f5","M377.9 105.9L500.7 228.7c7.2 7.2 11.3 17.1 11.3 27.3s-4.1 20.1-11.3 27.3L377.9 406.1c-6.4 6.4-15 9.9-24 9.9c-18.7 0-33.9-15.2-33.9-33.9l0-62.1-128 0c-17.7 0-32-14.3-32-32l0-64c0-17.7 14.3-32 32-32l128 0 0-62.1c0-18.7 15.2-33.9 33.9-33.9c9 0 17.6 3.6 24 9.9zM160 96L96 96c-17.7 0-32 14.3-32 32l0 256c0 17.7 14.3 32 32 32l64 0c17.7 0 32 14.3 32 32s-14.3 32-32 32l-64 0c-53 0-96-43-96-96L0 128C0 75 43 32 96 32l64 0c17.7 0 32 14.3 32 32s-14.3 32-32 32z"],atom:[512,512,[9883],"f5d2","M256 398.8c-11.8 5.1-23.4 9.7-34.9 13.5c16.7 33.8 31 35.7 34.9 35.7s18.1-1.9 34.9-35.7c-11.4-3.9-23.1-8.4-34.9-13.5zM446 256c33 45.2 44.3 90.9 23.6 128c-20.2 36.3-62.5 49.3-115.2 43.2c-22 52.1-55.6 84.8-98.4 84.8s-76.4-32.7-98.4-84.8c-52.7 6.1-95-6.8-115.2-43.2C21.7 346.9 33 301.2 66 256c-33-45.2-44.3-90.9-23.6-128c20.2-36.3 62.5-49.3 115.2-43.2C179.6 32.7 213.2 0 256 0s76.4 32.7 98.4 84.8c52.7-6.1 95 6.8 115.2 43.2c20.7 37.1 9.4 82.8-23.6 128zm-65.8 67.4c-1.7 14.2-3.9 28-6.7 41.2c31.8 1.4 38.6-8.7 40.2-11.7c2.3-4.2 7-17.9-11.9-48.1c-6.8 6.3-14 12.5-21.6 18.6zm-6.7-175.9c2.8 13.1 5 26.9 6.7 41.2c7.6 6.1 14.8 12.3 21.6 18.6c18.9-30.2 14.2-44 11.9-48.1c-1.6-2.9-8.4-13-40.2-11.7zM290.9 99.7C274.1 65.9 259.9 64 256 64s-18.1 1.9-34.9 35.7c11.4 3.9 23.1 8.4 34.9 13.5c11.8-5.1 23.4-9.7 34.9-13.5zm-159 88.9c1.7-14.3 3.9-28 6.7-41.2c-31.8-1.4-38.6 8.7-40.2 11.7c-2.3 4.2-7 17.9 11.9 48.1c6.8-6.3 14-12.5 21.6-18.6zM110.2 304.8C91.4 335 96 348.7 98.3 352.9c1.6 2.9 8.4 13 40.2 11.7c-2.8-13.1-5-26.9-6.7-41.2c-7.6-6.1-14.8-12.3-21.6-18.6zM336 256a80 80 0 1 0 -160 0 80 80 0 1 0 160 0zm-80-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],soap:[512,512,[129532],"e06e","M208 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM320 256a64 64 0 1 0 0-128 64 64 0 1 0 0 128zM416 32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm0 160c0 27.6-11.7 52.5-30.4 70.1C422.1 275.7 448 310.8 448 352c0 53-43 96-96 96H160c-53 0-96-43-96-96s43-96 96-96h88.4c-15.2-17-24.4-39.4-24.4-64H96c-53 0-96 43-96 96V416c0 53 43 96 96 96H416c53 0 96-43 96-96V288c0-53-43-96-96-96zM160 288c-35.3 0-64 28.7-64 64s28.7 64 64 64H352c35.3 0 64-28.7 64-64s-28.7-64-64-64H320 160z"],icons:[512,512,["heart-music-camera-bolt"],"f86d","M500.3 7.3C507.7 13.3 512 22.4 512 32V176c0 26.5-28.7 48-64 48s-64-21.5-64-48s28.7-48 64-48V71L352 90.2V208c0 26.5-28.7 48-64 48s-64-21.5-64-48s28.7-48 64-48V64c0-15.3 10.8-28.4 25.7-31.4l160-32c9.4-1.9 19.1 .6 26.6 6.6zM74.7 304l11.8-17.8c5.9-8.9 15.9-14.2 26.6-14.2h61.7c10.7 0 20.7 5.3 26.6 14.2L213.3 304H240c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V352c0-26.5 21.5-48 48-48H74.7zM192 408a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM478.7 278.3L440.3 368H496c6.7 0 12.6 4.1 15 10.4s.6 13.3-4.4 17.7l-128 112c-5.6 4.9-13.9 5.3-19.9 .9s-8.2-12.4-5.3-19.2L391.7 400H336c-6.7 0-12.6-4.1-15-10.4s-.6-13.3 4.4-17.7l128-112c5.6-4.9 13.9-5.3 19.9-.9s8.2 12.4 5.3 19.2zm-339-59.2c-6.5 6.5-17 6.5-23 0L19.9 119.2c-28-29-26.5-76.9 5-103.9c27-23.5 68.4-19 93.4 6.5l10 10.5 9.5-10.5c25-25.5 65.9-30 93.9-6.5c31 27 32.5 74.9 4.5 103.9l-96.4 99.9z"],"microphone-lines-slash":[640,512,["microphone-alt-slash"],"f539","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L472.1 344.7c15.2-26 23.9-56.3 23.9-88.7V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v24 16c0 21.2-5.1 41.1-14.2 58.7L416 300.8V256H358.9l-34.5-27c2.9-3.1 7-5 11.6-5h80V192H336c-8.8 0-16-7.2-16-16s7.2-16 16-16h80V128H336c-8.8 0-16-7.2-16-16s7.2-16 16-16h80c0-53-43-96-96-96s-96 43-96 96v54.3L38.8 5.1zM358.2 378.2C346.1 382 333.3 384 320 384c-70.7 0-128-57.3-128-128v-8.7L144.7 210c-.5 1.9-.7 3.9-.7 6v40c0 89.1 66.2 162.7 152 174.4V464H248c-13.3 0-24 10.7-24 24s10.7 24 24 24h72 72c13.3 0 24-10.7 24-24s-10.7-24-24-24H344V430.4c20.4-2.8 39.7-9.1 57.3-18.2l-43.1-33.9z"],"bridge-circle-check":[640,512,[],"e4c9","M64 32C46.3 32 32 46.3 32 64s14.3 32 32 32h40v64H32V288c53 0 96 43 96 96v64c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V384c0-53 43-96 96-96c6.3 0 12.4 .6 18.3 1.7C367.1 231.8 426.9 192 496 192c42.5 0 81.6 15.1 112 40.2V160H536V96h40c17.7 0 32-14.3 32-32s-14.3-32-32-32H64zM488 96v64H408V96h80zM360 96v64H280V96h80zM232 96v64H152V96h80zM640 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z"],"pump-medical":[448,512,[],"e06a","M128 32v96H256V96h60.1c4.2 0 8.3 1.7 11.3 4.7l33.9 33.9c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L372.7 55.4c-15-15-35.4-23.4-56.6-23.4H256c0-17.7-14.3-32-32-32H160c-17.7 0-32 14.3-32 32zM117.4 160c-33.3 0-61 25.5-63.8 58.7L35 442.7C31.9 480 61.3 512 98.8 512H285.2c37.4 0 66.9-32 63.8-69.3l-18.7-224c-2.8-33.2-30.5-58.7-63.8-58.7H117.4zM216 280v32h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H216v32c0 13.3-10.7 24-24 24s-24-10.7-24-24V360H136c-13.3 0-24-10.7-24-24s10.7-24 24-24h32V280c0-13.3 10.7-24 24-24s24 10.7 24 24z"],fingerprint:[512,512,[],"f577","M48 256C48 141.1 141.1 48 256 48c63.1 0 119.6 28.1 157.8 72.5c8.6 10.1 23.8 11.2 33.8 2.6s11.2-23.8 2.6-33.8C403.3 34.6 333.7 0 256 0C114.6 0 0 114.6 0 256v40c0 13.3 10.7 24 24 24s24-10.7 24-24V256zm458.5-52.9c-2.7-13-15.5-21.3-28.4-18.5s-21.3 15.5-18.5 28.4c2.9 13.9 4.5 28.3 4.5 43.1v40c0 13.3 10.7 24 24 24s24-10.7 24-24V256c0-18.1-1.9-35.8-5.5-52.9zM256 80c-19 0-37.4 3-54.5 8.6c-15.2 5-18.7 23.7-8.3 35.9c7.1 8.3 18.8 10.8 29.4 7.9c10.6-2.9 21.8-4.4 33.4-4.4c70.7 0 128 57.3 128 128v24.9c0 25.2-1.5 50.3-4.4 75.3c-1.7 14.6 9.4 27.8 24.2 27.8c11.8 0 21.9-8.6 23.3-20.3c3.3-27.4 5-55 5-82.7V256c0-97.2-78.8-176-176-176zM150.7 148.7c-9.1-10.6-25.3-11.4-33.9-.4C93.7 178 80 215.4 80 256v24.9c0 24.2-2.6 48.4-7.8 71.9C68.8 368.4 80.1 384 96.1 384c10.5 0 19.9-7 22.2-17.3c6.4-28.1 9.7-56.8 9.7-85.8V256c0-27.2 8.5-52.4 22.9-73.1c7.2-10.4 8-24.6-.2-34.2zM256 160c-53 0-96 43-96 96v24.9c0 35.9-4.6 71.5-13.8 106.1c-3.8 14.3 6.7 29 21.5 29c9.5 0 17.9-6.2 20.4-15.4c10.5-39 15.9-79.2 15.9-119.7V256c0-28.7 23.3-52 52-52s52 23.3 52 52v24.9c0 36.3-3.5 72.4-10.4 107.9c-2.7 13.9 7.7 27.2 21.8 27.2c10.2 0 19-7 21-17c7.7-38.8 11.6-78.3 11.6-118.1V256c0-53-43-96-96-96zm24 96c0-13.3-10.7-24-24-24s-24 10.7-24 24v24.9c0 59.9-11 119.3-32.5 175.2l-5.9 15.3c-4.8 12.4 1.4 26.3 13.8 31s26.3-1.4 31-13.8l5.9-15.3C267.9 411.9 280 346.7 280 280.9V256z"],"hand-point-right":[512,512,[],"f0a4","M480 96c17.7 0 32 14.3 32 32s-14.3 32-32 32l-208 0 0-64 208 0zM320 288c17.7 0 32 14.3 32 32s-14.3 32-32 32l-64 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l64 0zm64-64c0 17.7-14.3 32-32 32l-48 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l48 0c17.7 0 32 14.3 32 32zM288 384c17.7 0 32 14.3 32 32s-14.3 32-32 32l-64 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l64 0zm-88-96l.6 0c-5.4 9.4-8.6 20.3-8.6 32c0 13.2 4 25.4 10.8 35.6C177.9 364.3 160 388.1 160 416c0 11.7 3.1 22.6 8.6 32l-8.6 0C71.6 448 0 376.4 0 288l0-61.7c0-42.4 16.9-83.1 46.9-113.1l11.6-11.6C82.5 77.5 115.1 64 149 64l27 0c35.3 0 64 28.7 64 64l0 88c0 22.1-17.9 40-40 40s-40-17.9-40-40l0-56c0-8.8-7.2-16-16-16s-16 7.2-16 16l0 56c0 39.8 32.2 72 72 72z"],"magnifying-glass-location":[512,512,["search-location"],"f689","M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zM288 176c0-44.2-35.8-80-80-80s-80 35.8-80 80c0 48.8 46.5 111.6 68.6 138.6c6 7.3 16.8 7.3 22.7 0c22.1-27 68.6-89.8 68.6-138.6zm-112 0a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],"forward-step":[320,512,["step-forward"],"f051","M52.5 440.6c-9.5 7.9-22.8 9.7-34.1 4.4S0 428.4 0 416V96C0 83.6 7.2 72.3 18.4 67s24.5-3.6 34.1 4.4l192 160L256 241V96c0-17.7 14.3-32 32-32s32 14.3 32 32V416c0 17.7-14.3 32-32 32s-32-14.3-32-32V271l-11.5 9.6-192 160z"],"face-smile-beam":[512,512,[128522,"smile-beam"],"f5b8","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM164.1 325.5C182 346.2 212.6 368 256 368s74-21.8 91.9-42.5c5.8-6.7 15.9-7.4 22.6-1.6s7.4 15.9 1.6 22.6C349.8 372.1 311.1 400 256 400s-93.8-27.9-116.1-53.5c-5.8-6.7-5.1-16.8 1.6-22.6s16.8-5.1 22.6 1.6zm53.5-96.7l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z"],"flag-checkered":[448,512,[127937],"f11e","M32 0C49.7 0 64 14.3 64 32V48l69-17.2c38.1-9.5 78.3-5.1 113.5 12.5c46.3 23.2 100.8 23.2 147.1 0l9.6-4.8C423.8 28.1 448 43.1 448 66.1V345.8c0 13.3-8.3 25.3-20.8 30l-34.7 13c-46.2 17.3-97.6 14.6-141.7-7.4c-37.9-19-81.3-23.7-122.5-13.4L64 384v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V400 334 64 32C0 14.3 14.3 0 32 0zM64 187.1l64-13.9v65.5L64 252.6V318l48.8-12.2c5.1-1.3 10.1-2.4 15.2-3.3V238.7l38.9-8.4c8.3-1.8 16.7-2.5 25.1-2.1l0-64c13.6 .4 27.2 2.6 40.4 6.4l23.6 6.9v66.7l-41.7-12.3c-7.3-2.1-14.8-3.4-22.3-3.8v71.4c21.8 1.9 43.3 6.7 64 14.4V244.2l22.7 6.7c13.5 4 27.3 6.4 41.3 7.4V194c-7.8-.8-15.6-2.3-23.2-4.5l-40.8-12v-62c-13-3.8-25.8-8.8-38.2-15c-8.2-4.1-16.9-7-25.8-8.8v72.4c-13-.4-26 .8-38.7 3.6L128 173.2V98L64 114v73.1zM320 335.7c16.8 1.5 33.9-.7 50-6.8l14-5.2V251.9l-7.9 1.8c-18.4 4.3-37.3 5.7-56.1 4.5v77.4zm64-149.4V115.4c-20.9 6.1-42.4 9.1-64 9.1V194c13.9 1.4 28 .5 41.7-2.6l22.3-5.2z"],football:[512,512,[127944,"football-ball"],"f44e","M247.5 25.4c-13.5 3.3-26.4 7.2-38.6 11.7C142.9 61.6 96.7 103.6 66 153.6c-18.3 29.8-30.9 62.3-39.2 95.4L264.5 486.6c13.5-3.3 26.4-7.2 38.6-11.7c66-24.5 112.2-66.5 142.9-116.5c18.3-29.8 30.9-62.3 39.1-95.3L247.5 25.4zM495.2 205.3c6.1-56.8 1.4-112.2-7.7-156.4c-2.7-12.9-13-22.9-26.1-25.1c-58.2-9.7-109.9-12-155.6-7.9L495.2 205.3zM206.1 496L16.8 306.7c-6.1 56.8-1.4 112.2 7.7 156.4c2.7 12.9 13 22.9 26.1 25.1c58.2 9.7 109.9 12 155.6 7.9zm54.6-331.3c6.2-6.2 16.4-6.2 22.6 0l64 64c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0l-64-64c-6.2-6.2-6.2-16.4 0-22.6zm-48 48c6.2-6.2 16.4-6.2 22.6 0l64 64c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0l-64-64c-6.2-6.2-6.2-16.4 0-22.6zm-48 48c6.2-6.2 16.4-6.2 22.6 0l64 64c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0l-64-64c-6.2-6.2-6.2-16.4 0-22.6z"],"school-circle-exclamation":[640,512,[],"e56c","M337.8 5.4C327-1.8 313-1.8 302.2 5.4L166.3 96H48C21.5 96 0 117.5 0 144V464c0 26.5 21.5 48 48 48H320v0H256V416c0-35.3 28.7-64 64-64l.3 0h.5c3.4-37.7 18.7-72.1 42.2-99.1C350.2 260 335.6 264 320 264c-48.6 0-88-39.4-88-88s39.4-88 88-88s88 39.4 88 88c0 18.3-5.6 35.3-15.1 49.4c29-21 64.6-33.4 103.1-33.4c59.5 0 112.1 29.6 144 74.8V144c0-26.5-21.5-48-48-48H473.7L337.8 5.4zM96 192h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V208c0-8.8 7.2-16 16-16zm0 128h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zM320 128c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H336V144c0-8.8-7.2-16-16-16zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z"],crop:[512,512,[],"f125","M448 109.3l54.6-54.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L402.7 64 160 64v64l178.7 0L128 338.7V32c0-17.7-14.3-32-32-32S64 14.3 64 32V64H32C14.3 64 0 78.3 0 96s14.3 32 32 32H64V384c0 35.3 28.7 64 64 64H352V384H173.3L384 173.3 384 480c0 17.7 14.3 32 32 32s32-14.3 32-32V448h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H448l0-274.7z"],"angles-down":[448,512,["angle-double-down"],"f103","M246.6 470.6c-12.5 12.5-32.8 12.5-45.3 0l-160-160c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0L224 402.7 361.4 265.4c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3l-160 160zm160-352l-160 160c-12.5 12.5-32.8 12.5-45.3 0l-160-160c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0L224 210.7 361.4 73.4c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3z"],"users-rectangle":[640,512,[],"e594","M96 0C43 0 0 43 0 96V416c0 53 43 96 96 96H544c53 0 96-43 96-96V96c0-53-43-96-96-96H96zM64 96c0-17.7 14.3-32 32-32H544c17.7 0 32 14.3 32 32V416c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V96zm159.8 80a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM96 309.3c0 14.7 11.9 26.7 26.7 26.7h56.1c8-34.1 32.8-61.7 65.2-73.6c-7.5-4.1-16.2-6.4-25.3-6.4H149.3C119.9 256 96 279.9 96 309.3zM461.2 336h56.1c14.7 0 26.7-11.9 26.7-26.7c0-29.5-23.9-53.3-53.3-53.3H421.3c-9.2 0-17.8 2.3-25.3 6.4c32.4 11.9 57.2 39.5 65.2 73.6zM372 289c-3.9-.7-7.9-1-12-1H280c-4.1 0-8.1 .3-12 1c-26 4.4-47.3 22.7-55.9 47c-2.7 7.5-4.1 15.6-4.1 24c0 13.3 10.7 24 24 24H408c13.3 0 24-10.7 24-24c0-8.4-1.4-16.5-4.1-24c-8.6-24.3-29.9-42.6-55.9-47zM512 176a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM320 256a64 64 0 1 0 0-128 64 64 0 1 0 0 128z"],"people-roof":[640,512,[],"e537","M335.5 4l288 160c15.4 8.6 21 28.1 12.4 43.5s-28.1 21-43.5 12.4L320 68.6 47.5 220c-15.4 8.6-34.9 3-43.5-12.4s-3-34.9 12.4-43.5L304.5 4c9.7-5.4 21.4-5.4 31.1 0zM320 160a40 40 0 1 1 0 80 40 40 0 1 1 0-80zM144 256a40 40 0 1 1 0 80 40 40 0 1 1 0-80zm312 40a40 40 0 1 1 80 0 40 40 0 1 1 -80 0zM226.9 491.4L200 441.5V480c0 17.7-14.3 32-32 32H120c-17.7 0-32-14.3-32-32V441.5L61.1 491.4c-6.3 11.7-20.8 16-32.5 9.8s-16-20.8-9.8-32.5l37.9-70.3c15.3-28.5 45.1-46.3 77.5-46.3h19.5c16.3 0 31.9 4.5 45.4 12.6l33.6-62.3c15.3-28.5 45.1-46.3 77.5-46.3h19.5c32.4 0 62.1 17.8 77.5 46.3l33.6 62.3c13.5-8.1 29.1-12.6 45.4-12.6h19.5c32.4 0 62.1 17.8 77.5 46.3l37.9 70.3c6.3 11.7 1.9 26.2-9.8 32.5s-26.2 1.9-32.5-9.8L552 441.5V480c0 17.7-14.3 32-32 32H472c-17.7 0-32-14.3-32-32V441.5l-26.9 49.9c-6.3 11.7-20.8 16-32.5 9.8s-16-20.8-9.8-32.5l36.3-67.5c-1.7-1.7-3.2-3.6-4.3-5.8L376 345.5V400c0 17.7-14.3 32-32 32H296c-17.7 0-32-14.3-32-32V345.5l-26.9 49.9c-1.2 2.2-2.6 4.1-4.3 5.8l36.3 67.5c6.3 11.7 1.9 26.2-9.8 32.5s-26.2 1.9-32.5-9.8z"],"people-line":[640,512,[],"e534","M360 72a40 40 0 1 0 -80 0 40 40 0 1 0 80 0zM144 208a40 40 0 1 0 0-80 40 40 0 1 0 0 80zM32 416c-17.7 0-32 14.3-32 32s14.3 32 32 32H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H32zM496 208a40 40 0 1 0 0-80 40 40 0 1 0 0 80zM200 313.5l26.9 49.9c6.3 11.7 20.8 16 32.5 9.8s16-20.8 9.8-32.5l-36.3-67.5c1.7-1.7 3.2-3.6 4.3-5.8L264 217.5V272c0 17.7 14.3 32 32 32h48c17.7 0 32-14.3 32-32V217.5l26.9 49.9c1.2 2.2 2.6 4.1 4.3 5.8l-36.3 67.5c-6.3 11.7-1.9 26.2 9.8 32.5s26.2 1.9 32.5-9.8L440 313.5V352c0 17.7 14.3 32 32 32h48c17.7 0 32-14.3 32-32V313.5l26.9 49.9c6.3 11.7 20.8 16 32.5 9.8s16-20.8 9.8-32.5l-37.9-70.3c-15.3-28.5-45.1-46.3-77.5-46.3H486.2c-16.3 0-31.9 4.5-45.4 12.6l-33.6-62.3c-15.3-28.5-45.1-46.3-77.5-46.3H310.2c-32.4 0-62.1 17.8-77.5 46.3l-33.6 62.3c-13.5-8.1-29.1-12.6-45.4-12.6H134.2c-32.4 0-62.1 17.8-77.5 46.3L18.9 340.6c-6.3 11.7-1.9 26.2 9.8 32.5s26.2 1.9 32.5-9.8L88 313.5V352c0 17.7 14.3 32 32 32h48c17.7 0 32-14.3 32-32V313.5z"],"beer-mug-empty":[512,512,["beer"],"f0fc","M32 64c0-17.7 14.3-32 32-32H352c17.7 0 32 14.3 32 32V96h51.2c42.4 0 76.8 34.4 76.8 76.8V274.9c0 30.4-17.9 57.9-45.6 70.2L384 381.7V416c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V64zM384 311.6l56.4-25.1c4.6-2.1 7.6-6.6 7.6-11.7V172.8c0-7.1-5.7-12.8-12.8-12.8H384V311.6zM160 144c0-8.8-7.2-16-16-16s-16 7.2-16 16V368c0 8.8 7.2 16 16 16s16-7.2 16-16V144zm64 0c0-8.8-7.2-16-16-16s-16 7.2-16 16V368c0 8.8 7.2 16 16 16s16-7.2 16-16V144zm64 0c0-8.8-7.2-16-16-16s-16 7.2-16 16V368c0 8.8 7.2 16 16 16s16-7.2 16-16V144z"],"diagram-predecessor":[512,512,[],"e477","M448 416l0-64L64 352l0 64 384 0zm0 64L64 480c-35.3 0-64-28.7-64-64l0-64c0-35.3 28.7-64 64-64l384 0c35.3 0 64 28.7 64 64l0 64c0 35.3-28.7 64-64 64zM288 160c0 35.3-28.7 64-64 64L64 224c-35.3 0-64-28.7-64-64L0 96C0 60.7 28.7 32 64 32l144 0 16 0 144 0c44.2 0 80 35.8 80 80l0 16 38.1 0c21.4 0 32.1 25.9 17 41L433 239c-9.4 9.4-24.6 9.4-33.9 0L329 169c-15.1-15.1-4.4-41 17-41l38.1 0 0-16c0-8.8-7.2-16-16-16l-80 0 0 64z"],"arrow-up-long":[384,512,["long-arrow-up"],"f176","M214.6 9.4c-12.5-12.5-32.8-12.5-45.3 0l-128 128c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L160 109.3V480c0 17.7 14.3 32 32 32s32-14.3 32-32V109.3l73.4 73.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-128-128z"],"fire-flame-simple":[384,512,["burn"],"f46a","M372.5 256.5l-.7-1.9C337.8 160.8 282 76.5 209.1 8.5l-3.3-3C202.1 2 197.1 0 192 0s-10.1 2-13.8 5.5l-3.3 3C102 76.5 46.2 160.8 12.2 254.6l-.7 1.9C3.9 277.3 0 299.4 0 321.6C0 426.7 86.8 512 192 512s192-85.3 192-190.4c0-22.2-3.9-44.2-11.5-65.1zm-90.8 49.5c4.1 9.3 6.2 19.4 6.2 29.5c0 53-43 96.5-96 96.5s-96-43.5-96-96.5c0-10.1 2.1-20.3 6.2-29.5l1.9-4.3c15.8-35.4 37.9-67.7 65.3-95.1l8.9-8.9c3.6-3.6 8.5-5.6 13.6-5.6s10 2 13.6 5.6l8.9 8.9c27.4 27.4 49.6 59.7 65.3 95.1l1.9 4.3z"],person:[320,512,[129485,"male"],"f183","M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l58.3 97c9.1 15.1 4.2 34.8-10.9 43.9s-34.8 4.2-43.9-10.9L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152z"],laptop:[640,512,[128187],"f109","M128 32C92.7 32 64 60.7 64 96V352h64V96H512V352h64V96c0-35.3-28.7-64-64-64H128zM19.2 384C8.6 384 0 392.6 0 403.2C0 445.6 34.4 480 76.8 480H563.2c42.4 0 76.8-34.4 76.8-76.8c0-10.6-8.6-19.2-19.2-19.2H19.2z"],"file-csv":[512,512,[],"f6dd","M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384V304H176c-35.3 0-64 28.7-64 64V512H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zM200 352h16c22.1 0 40 17.9 40 40v8c0 8.8-7.2 16-16 16s-16-7.2-16-16v-8c0-4.4-3.6-8-8-8H200c-4.4 0-8 3.6-8 8v80c0 4.4 3.6 8 8 8h16c4.4 0 8-3.6 8-8v-8c0-8.8 7.2-16 16-16s16 7.2 16 16v8c0 22.1-17.9 40-40 40H200c-22.1 0-40-17.9-40-40V392c0-22.1 17.9-40 40-40zm133.1 0H368c8.8 0 16 7.2 16 16s-7.2 16-16 16H333.1c-7.2 0-13.1 5.9-13.1 13.1c0 5.2 3 9.9 7.8 12l37.4 16.6c16.3 7.2 26.8 23.4 26.8 41.2c0 24.9-20.2 45.1-45.1 45.1H304c-8.8 0-16-7.2-16-16s7.2-16 16-16h42.9c7.2 0 13.1-5.9 13.1-13.1c0-5.2-3-9.9-7.8-12l-37.4-16.6c-16.3-7.2-26.8-23.4-26.8-41.2c0-24.9 20.2-45.1 45.1-45.1zm98.9 0c8.8 0 16 7.2 16 16v31.6c0 23 5.5 45.6 16 66c10.5-20.3 16-42.9 16-66V368c0-8.8 7.2-16 16-16s16 7.2 16 16v31.6c0 34.7-10.3 68.7-29.6 97.6l-5.1 7.7c-3 4.5-8 7.1-13.3 7.1s-10.3-2.7-13.3-7.1l-5.1-7.7c-19.3-28.9-29.6-62.9-29.6-97.6V368c0-8.8 7.2-16 16-16z"],menorah:[640,512,[],"f676","M20.8 7.4C22.8 2.9 27.1 0 32 0s9.2 2.9 11.2 7.4L61.3 49.7c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32S0 81.7 0 64V62.8c0-4.5 .9-8.9 2.7-13.1L20.8 7.4zm96 0C118.8 2.9 123.1 0 128 0s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32s-32-14.3-32-32V62.8c0-4.5 .9-8.9 2.7-13.1L116.8 7.4zm77.8 42.4L212.8 7.4C214.8 2.9 219.1 0 224 0s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32s-32-14.3-32-32V62.8c0-4.5 .9-8.9 2.7-13.1zM308.8 7.4C310.8 2.9 315.1 0 320 0s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32s-32-14.3-32-32V62.8c0-4.5 .9-8.9 2.7-13.1L308.8 7.4zm77.8 42.4L404.8 7.4C406.8 2.9 411.1 0 416 0s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32s-32-14.3-32-32V62.8c0-4.5 .9-8.9 2.7-13.1zM500.8 7.4C502.8 2.9 507.1 0 512 0s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32s-32-14.3-32-32V62.8c0-4.5 .9-8.9 2.7-13.1L500.8 7.4zm77.8 42.4L596.8 7.4C598.8 2.9 603.1 0 608 0s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32s-32-14.3-32-32V62.8c0-4.5 .9-8.9 2.7-13.1zM32 128c17.7 0 32 14.3 32 32V288c0 17.7 14.3 32 32 32H288V160c0-17.7 14.3-32 32-32s32 14.3 32 32V320H544c17.7 0 32-14.3 32-32V160c0-17.7 14.3-32 32-32s32 14.3 32 32V288c0 53-43 96-96 96H352v64H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H320 160c-17.7 0-32-14.3-32-32s14.3-32 32-32H288V384H96c-53 0-96-43-96-96V160c0-17.7 14.3-32 32-32zm96 0c17.7 0 32 14.3 32 32v96 32H96V256 160c0-17.7 14.3-32 32-32zm96 0c17.7 0 32 14.3 32 32v96 32H192V256 160c0-17.7 14.3-32 32-32zm192 0c17.7 0 32 14.3 32 32v96 32H384V256 160c0-17.7 14.3-32 32-32zm96 0c17.7 0 32 14.3 32 32v96 32H480V256 160c0-17.7 14.3-32 32-32z"],"truck-plane":[640,512,[],"e58f","M200 0c-30.6 0-56 54.7-56 86.1V192.5L7.8 274.3C2.9 277.2 0 282.4 0 288v64c0 5.1 2.4 9.8 6.4 12.8s9.3 3.9 14.1 2.5l123.4-37v81.2l-50 40c-3.8 3-6 7.6-6 12.5v32c0 5.1 2.5 10 6.6 13s9.5 3.8 14.4 2.2L200 480.9 290.4 511c-1.6-4.7-2.4-9.8-2.4-15V463.4c-18.2-10.5-30.7-29.7-31.9-51.8l-.1-.1V408 325.5 184l0-1.1 0 0V86.1C256 54.7 231.5 0 200 0zm88 176V400c0 20.9 13.4 38.7 32 45.3V488c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V448H544v40c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V445.3c18.6-6.6 32-24.4 32-45.3V176c0-26.5-21.5-48-48-48H336c-26.5 0-48 21.5-48 48zm79.8 78.7c3.3-8.7 11.2-14.7 20.5-14.7H539.7c9.2 0 17.2 6 20.5 14.7L576 304H352l15.8-49.3zM568 352a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM336 376a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z"],"record-vinyl":[512,512,[],"f8d9","M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm256-96a96 96 0 1 1 0 192 96 96 0 1 1 0-192zm0 224a128 128 0 1 0 0-256 128 128 0 1 0 0 256zm0-96a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"face-grin-stars":[512,512,[129321,"grin-stars"],"f587","M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm407.4 75.5c5-11.8-7-22.5-19.3-18.7c-39.7 12.2-84.5 19-131.8 19s-92.1-6.8-131.8-19c-12.3-3.8-24.3 6.9-19.3 18.7c25 59.1 83.2 100.5 151.1 100.5s126.2-41.4 151.1-100.5zM160 120c-3.1 0-5.9 1.8-7.2 4.6l-16.6 34.7-38.1 5c-3.1 .4-5.6 2.5-6.6 5.5s-.1 6.2 2.1 8.3l27.9 26.5-7 37.8c-.6 3 .7 6.1 3.2 7.9s5.8 2 8.5 .6L160 232.5l33.8 18.3c2.7 1.5 6 1.3 8.5-.6s3.7-4.9 3.2-7.9l-7-37.8L226.4 178c2.2-2.1 3.1-5.3 2.1-8.3s-3.5-5.1-6.6-5.5l-38.1-5-16.6-34.7c-1.3-2.8-4.1-4.6-7.2-4.6zm192 0c-3.1 0-5.9 1.8-7.2 4.6l-16.6 34.7-38.1 5c-3.1 .4-5.6 2.5-6.6 5.5s-.1 6.2 2.1 8.3l27.9 26.5-7 37.8c-.6 3 .7 6.1 3.2 7.9s5.8 2 8.5 .6L352 232.5l33.8 18.3c2.7 1.5 6 1.3 8.5-.6s3.7-4.9 3.2-7.9l-7-37.8L418.4 178c2.2-2.1 3.1-5.3 2.1-8.3s-3.5-5.1-6.6-5.5l-38.1-5-16.6-34.7c-1.3-2.8-4.1-4.6-7.2-4.6z"],bong:[448,512,[],"f55c","M160 208.5c0 29.1-15.6 53.9-37.2 67.8c-17.2 11.1-31.5 26.1-41.7 43.7H302.9c-10.2-17.6-24.5-32.6-41.7-43.7c-21.6-13.9-37.2-38.7-37.2-67.8V64H160V208.5zM288 64V208.5c0 5.7 3.1 10.9 7.9 14c11.2 7.2 21.5 15.5 30.9 24.8L366.1 208l-7-7c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l24 24 24 24c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-7-7-43.3 43.3C374 314.1 384 347.9 384 384c0 39.4-11.9 76.1-32.2 106.5c-9.6 14.4-26.5 21.5-43.8 21.5H76.1c-17.3 0-34.2-7.1-43.8-21.5C11.9 460.1 0 423.4 0 384c0-67.8 35.1-127.3 88.1-161.5c4.8-3.1 7.9-8.3 7.9-14V64C78.3 64 64 49.7 64 32S78.3 0 96 0h16H272h16c17.7 0 32 14.3 32 32s-14.3 32-32 32z"],"spaghetti-monster-flying":[640,512,["pastafarianism"],"f67b","M208 64a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm48 0c0 16.2-6 31.1-16 42.3l15.6 31.2c18.7-6 39.9-9.5 64.4-9.5s45.8 3.5 64.4 9.5L400 106.3C390 95.1 384 80.2 384 64c0-35.3 28.7-64 64-64s64 28.7 64 64s-28.7 64-64 64c-1.7 0-3.4-.1-5.1-.2L427.8 158c21.1 13.6 37.7 30.2 51.4 46.4c7.1 8.3 13.5 16.6 19.3 24l1.4 1.8c6.3 8.1 11.6 14.8 16.7 20.4C527.3 262.3 532.7 264 536 264c2.5 0 4.3-.6 7.1-3.3c3.7-3.5 7.1-8.8 12.5-17.4l.6-.9c4.6-7.4 11-17.6 19.4-25.7c9.7-9.3 22.9-16.7 40.4-16.7c13.3 0 24 10.7 24 24s-10.7 24-24 24c-2.5 0-4.3 .6-7.1 3.3c-3.7 3.5-7.1 8.8-12.5 17.4l-.6 .9c-4.6 7.4-11 17.6-19.4 25.7c-9.7 9.3-22.9 16.7-40.4 16.7c-18.5 0-32.9-8.5-44.3-18.6c-3.1 4-6.6 8.3-10.5 12.7c1.4 4.3 2.8 8.5 4 12.5c.9 3 1.8 5.8 2.6 8.6c3 9.8 5.5 18.2 8.6 25.9c3.9 9.8 7.4 15.4 10.8 18.5c2.6 2.4 5.9 4.3 12.8 4.3c8.7 0 16.9-4.2 33.7-13.2c15-8 35.7-18.8 62.3-18.8c13.3 0 24 10.7 24 24s-10.7 24-24 24c-13.4 0-24.7 5.2-39.7 13.2c-1 .6-2.1 1.1-3.2 1.7C559.9 414 541.4 424 520 424c-18.4 0-33.6-6.1-45.5-17.2c-11.1-10.3-17.9-23.7-22.7-36c-3.6-9-6.7-19.1-9.5-28.5c-16.4 12.3-36.1 23.6-58.9 31.3c3.6 10.8 8.4 23.5 14.4 36.2c7.5 15.9 16.2 30.4 25.8 40.5C433 460.5 441.2 464 448 464c13.3 0 24 10.7 24 24s-10.7 24-24 24c-25.2 0-45-13.5-59.5-28.8c-14.5-15.4-25.7-34.9-34.2-53c-8-17-14.1-33.8-18.3-46.9c-5.2 .4-10.6 .6-16 .6s-10.8-.2-16-.6c-4.2 13-10.3 29.9-18.3 46.9c-8.5 18.1-19.8 37.6-34.2 53C237 498.5 217.2 512 192 512c-13.3 0-24-10.7-24-24s10.7-24 24-24c6.8 0 15-3.5 24.5-13.7c9.5-10.1 18.3-24.6 25.8-40.5c5.9-12.6 10.7-25.4 14.4-36.2c-22.8-7.7-42.5-19-58.9-31.3c-2.9 9.4-6 19.5-9.5 28.5c-4.8 12.2-11.6 25.6-22.7 36C153.6 417.9 138.4 424 120 424c-21.4 0-39.9-10-53.1-17.1l0 0c-1.1-.6-2.2-1.2-3.2-1.7c-15-8-26.3-13.2-39.7-13.2c-13.3 0-24-10.7-24-24s10.7-24 24-24c26.6 0 47.3 10.8 62.3 18.8c16.8 9 25 13.2 33.7 13.2c6.8 0 10.2-1.9 12.8-4.3c3.4-3.2 7-8.8 10.8-18.5c3-7.7 5.6-16.1 8.6-25.9c.8-2.7 1.7-5.6 2.6-8.6c1.2-4 2.6-8.2 4-12.5c-3.9-4.5-7.4-8.8-10.5-12.7C136.9 303.5 122.5 312 104 312c-17.5 0-30.7-7.4-40.4-16.7c-8.4-8.1-14.8-18.3-19.4-25.7l-.6-.9c-5.4-8.6-8.8-13.9-12.5-17.4c-2.8-2.7-4.6-3.3-7.1-3.3c-13.3 0-24-10.7-24-24s10.7-24 24-24c17.5 0 30.7 7.4 40.4 16.7c8.4 8.1 14.8 18.3 19.4 25.7l.6 .9c5.4 8.6 8.8 13.9 12.5 17.4c2.8 2.7 4.6 3.3 7.1 3.3c3.3 0 8.7-1.7 19.4-13.4c5.1-5.6 10.4-12.3 16.7-20.4l1.4-1.8c5.8-7.4 12.2-15.7 19.3-24c13.8-16.2 30.3-32.8 51.4-46.4l-15.1-30.2c-1.7 .1-3.4 .2-5.1 .2c-35.3 0-64-28.7-64-64s28.7-64 64-64s64 28.7 64 64zm208 0a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z"],"arrow-down-up-across-line":[576,512,[],"e4af","M137.4 502.6c12.5 12.5 32.8 12.5 45.3 0l96-96c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L192 402.7V288H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H448V109.3l41.4 41.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-96-96c-12.5-12.5-32.8-12.5-45.3 0l-96 96c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L384 109.3V224H192 128 32c-17.7 0-32 14.3-32 32s14.3 32 32 32h96V402.7L86.6 361.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l96 96zM128 192h64V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V192zM448 320H384V448c0 17.7 14.3 32 32 32s32-14.3 32-32V320z"],spoon:[512,512,[129348,61873,"utensil-spoon"],"f2e5","M245.8 220.9c-14.5-17.6-21.8-39.2-21.8-60.8C224 80 320 0 416 0c53 0 96 43 96 96c0 96-80 192-160.2 192c-21.6 0-43.2-7.3-60.8-21.8L54.6 502.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L245.8 220.9z"],"jar-wheat":[320,512,[],"e517","M32 32C32 14.3 46.3 0 64 0H256c17.7 0 32 14.3 32 32s-14.3 32-32 32H64C46.3 64 32 49.7 32 32zM0 160c0-35.3 28.7-64 64-64H256c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V160zm112 0H69.8c-3.2 0-5.8 2.6-5.8 5.8C64 198 90 224 122.2 224H144h32 21.8c32.1 0 58.2-26 58.2-58.2c0-3.2-2.6-5.8-5.8-5.8H208c-19.1 0-36.3 8.4-48 21.7c-11.7-13.3-28.9-21.7-48-21.7zm48 117.7c-11.7-13.3-28.9-21.7-48-21.7H69.8c-3.2 0-5.8 2.6-5.8 5.8C64 294 90 320 122.2 320H144h32 21.8c32.1 0 58.2-26 58.2-58.2c0-3.2-2.6-5.8-5.8-5.8H208c-19.1 0-36.3 8.4-48 21.7zM112 352H69.8c-3.2 0-5.8 2.6-5.8 5.8C64 390 90 416 122.2 416H144v32c0 8.8 7.2 16 16 16s16-7.2 16-16V416h21.8c32.1 0 58.2-26 58.2-58.2c0-3.2-2.6-5.8-5.8-5.8H208c-19.1 0-36.3 8.4-48 21.7c-11.7-13.3-28.9-21.7-48-21.7z"],"envelopes-bulk":[640,512,["mail-bulk"],"f674","M128 0C110.3 0 96 14.3 96 32V224h96V192c0-35.3 28.7-64 64-64H480V32c0-17.7-14.3-32-32-32H128zM256 160c-17.7 0-32 14.3-32 32v32h96c35.3 0 64 28.7 64 64V416H576c17.7 0 32-14.3 32-32V192c0-17.7-14.3-32-32-32H256zm240 64h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H496c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zM64 256c-17.7 0-32 14.3-32 32v13L187.1 415.9c1.4 1 3.1 1.6 4.9 1.6s3.5-.6 4.9-1.6L352 301V288c0-17.7-14.3-32-32-32H64zm288 84.8L216 441.6c-6.9 5.1-15.3 7.9-24 7.9s-17-2.8-24-7.9L32 340.8V480c0 17.7 14.3 32 32 32H320c17.7 0 32-14.3 32-32V340.8z"],"file-circle-exclamation":[576,512,[],"e4eb","M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v38.6C310.1 219.5 256 287.4 256 368c0 59.1 29.1 111.3 73.7 143.3c-3.2 .5-6.4 .7-9.7 .7H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zm48 96a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm0 240a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm0-192c-8.8 0-16 7.2-16 16v80c0 8.8 7.2 16 16 16s16-7.2 16-16V288c0-8.8-7.2-16-16-16z"],"circle-h":[512,512,[9405,"hospital-symbol"],"f47e","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM368 152V256 360c0 13.3-10.7 24-24 24s-24-10.7-24-24V280H192l0 80c0 13.3-10.7 24-24 24s-24-10.7-24-24l0-208c0-13.3 10.7-24 24-24s24 10.7 24 24v80H320V152c0-13.3 10.7-24 24-24s24 10.7 24 24z"],pager:[512,512,[128223],"f815","M0 128C0 92.7 28.7 64 64 64H448c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128zm64 32v64c0 17.7 14.3 32 32 32H416c17.7 0 32-14.3 32-32V160c0-17.7-14.3-32-32-32H96c-17.7 0-32 14.3-32 32zM80 320c-13.3 0-24 10.7-24 24s10.7 24 24 24h56c13.3 0 24-10.7 24-24s-10.7-24-24-24H80zm136 0c-13.3 0-24 10.7-24 24s10.7 24 24 24h48c13.3 0 24-10.7 24-24s-10.7-24-24-24H216z"],"address-book":[512,512,[62138,"contact-book"],"f2b9","M96 0C60.7 0 32 28.7 32 64V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H96zM208 288h64c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16H144c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80zm-32-96a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM512 80c0-8.8-7.2-16-16-16s-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V80zM496 192c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V208c0-8.8-7.2-16-16-16zm16 144c0-8.8-7.2-16-16-16s-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V336z"],strikethrough:[512,512,[],"f0cc","M161.3 144c3.2-17.2 14-30.1 33.7-38.6c21.1-9 51.8-12.3 88.6-6.5c11.9 1.9 48.8 9.1 60.1 12c17.1 4.5 34.6-5.6 39.2-22.7s-5.6-34.6-22.7-39.2c-14.3-3.8-53.6-11.4-66.6-13.4c-44.7-7-88.3-4.2-123.7 10.9c-36.5 15.6-64.4 44.8-71.8 87.3c-.1 .6-.2 1.1-.2 1.7c-2.8 23.9 .5 45.6 10.1 64.6c4.5 9 10.2 16.9 16.7 23.9H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H270.1c-.1 0-.3-.1-.4-.1l-1.1-.3c-36-10.8-65.2-19.6-85.2-33.1c-9.3-6.3-15-12.6-18.2-19.1c-3.1-6.1-5.2-14.6-3.8-27.4zM348.9 337.2c2.7 6.5 4.4 15.8 1.9 30.1c-3 17.6-13.8 30.8-33.9 39.4c-21.1 9-51.7 12.3-88.5 6.5c-18-2.9-49.1-13.5-74.4-22.1c-5.6-1.9-11-3.7-15.9-5.4c-16.8-5.6-34.9 3.5-40.5 20.3s3.5 34.9 20.3 40.5c3.6 1.2 7.9 2.7 12.7 4.3l0 0 0 0c24.9 8.5 63.6 21.7 87.6 25.6l0 0 .2 0c44.7 7 88.3 4.2 123.7-10.9c36.5-15.6 64.4-44.8 71.8-87.3c3.6-21 2.7-40.4-3.1-58.1H335.1c7 5.6 11.4 11.2 13.9 17.2z"],k:[320,512,[107],"4b","M311 86.3c12.3-12.7 12-32.9-.7-45.2s-32.9-12-45.2 .7l-155.2 160L64 249V64c0-17.7-14.3-32-32-32S0 46.3 0 64V328 448c0 17.7 14.3 32 32 32s32-14.3 32-32V341l64.7-66.7 133 192c10.1 14.5 30 18.1 44.5 8.1s18.1-30 8.1-44.5L174.1 227.4 311 86.3z"],"landmark-flag":[512,512,[],"e51c","M272 0h80c8.8 0 16 7.2 16 16V80c0 8.8-7.2 16-16 16H272v32H464c17.7 0 32 14.3 32 32s-14.3 32-32 32H48c-17.7 0-32-14.3-32-32s14.3-32 32-32H240V16c0-8.8 7.2-16 16-16h16zM64 224h64V416h40V224h64V416h48V224h64V416h40V224h64V420.3c.6 .3 1.2 .7 1.8 1.1l48 32c11.7 7.8 17 22.4 12.9 35.9S494.1 512 480 512H32c-14.1 0-26.5-9.2-30.6-22.7s1.1-28.1 12.9-35.9l48-32c.6-.4 1.2-.7 1.8-1.1V224z"],pencil:[512,512,[9999,61504,"pencil-alt"],"f303","M410.3 231l11.3-11.3-33.9-33.9-62.1-62.1L291.7 89.8l-11.3 11.3-22.6 22.6L58.6 322.9c-10.4 10.4-18 23.3-22.2 37.4L1 480.7c-2.5 8.4-.2 17.5 6.1 23.7s15.3 8.5 23.7 6.1l120.3-35.4c14.1-4.2 27-11.8 37.4-22.2L387.7 253.7 410.3 231zM160 399.4l-9.1 22.7c-4 3.1-8.5 5.4-13.3 6.9L59.4 452l23-78.1c1.4-4.9 3.8-9.4 6.9-13.3l22.7-9.1v32c0 8.8 7.2 16 16 16h32zM362.7 18.7L348.3 33.2 325.7 55.8 314.3 67.1l33.9 33.9 62.1 62.1 33.9 33.9 11.3-11.3 22.6-22.6 14.5-14.5c25-25 25-65.5 0-90.5L453.3 18.7c-25-25-65.5-25-90.5 0zm-47.4 168l-144 144c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6l144-144c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z"],backward:[512,512,[9194],"f04a","M459.5 440.6c9.5 7.9 22.8 9.7 34.1 4.4s18.4-16.6 18.4-29V96c0-12.4-7.2-23.7-18.4-29s-24.5-3.6-34.1 4.4L288 214.3V256v41.7L459.5 440.6zM256 352V256 128 96c0-12.4-7.2-23.7-18.4-29s-24.5-3.6-34.1 4.4l-192 160C4.2 237.5 0 246.5 0 256s4.2 18.5 11.5 24.6l192 160c9.5 7.9 22.8 9.7 34.1 4.4s18.4-16.6 18.4-29V352z"],"caret-right":[256,512,[],"f0da","M246.6 278.6c12.5-12.5 12.5-32.8 0-45.3l-128-128c-9.2-9.2-22.9-11.9-34.9-6.9s-19.8 16.6-19.8 29.6l0 256c0 12.9 7.8 24.6 19.8 29.6s25.7 2.2 34.9-6.9l128-128z"],comments:[640,512,[128490,61670],"f086","M208 352c114.9 0 208-78.8 208-176S322.9 0 208 0S0 78.8 0 176c0 38.6 14.7 74.3 39.6 103.4c-3.5 9.4-8.7 17.7-14.2 24.7c-4.8 6.2-9.7 11-13.3 14.3c-1.8 1.6-3.3 2.9-4.3 3.7c-.5 .4-.9 .7-1.1 .8l-.2 .2 0 0 0 0C1 327.2-1.4 334.4 .8 340.9S9.1 352 16 352c21.8 0 43.8-5.6 62.1-12.5c9.2-3.5 17.8-7.4 25.3-11.4C134.1 343.3 169.8 352 208 352zM448 176c0 112.3-99.1 196.9-216.5 207C255.8 457.4 336.4 512 432 512c38.2 0 73.9-8.7 104.7-23.9c7.5 4 16 7.9 25.2 11.4c18.3 6.9 40.3 12.5 62.1 12.5c6.9 0 13.1-4.5 15.2-11.1c2.1-6.6-.2-13.8-5.8-17.9l0 0 0 0-.2-.2c-.2-.2-.6-.4-1.1-.8c-1-.8-2.5-2-4.3-3.7c-3.6-3.3-8.5-8.1-13.3-14.3c-5.5-7-10.7-15.4-14.2-24.7c24.9-29 39.6-64.7 39.6-103.4c0-92.8-84.9-168.9-192.6-175.5c.4 5.1 .6 10.3 .6 15.5z"],paste:[512,512,["file-clipboard"],"f0ea","M160 0c-23.7 0-44.4 12.9-55.4 32H48C21.5 32 0 53.5 0 80V400c0 26.5 21.5 48 48 48H192V176c0-44.2 35.8-80 80-80h48V80c0-26.5-21.5-48-48-48H215.4C204.4 12.9 183.7 0 160 0zM272 128c-26.5 0-48 21.5-48 48V448v16c0 26.5 21.5 48 48 48H464c26.5 0 48-21.5 48-48V243.9c0-12.7-5.1-24.9-14.1-33.9l-67.9-67.9c-9-9-21.2-14.1-33.9-14.1H320 272zM160 40a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],"code-pull-request":[512,512,[],"e13c","M305.8 2.1C314.4 5.9 320 14.5 320 24V64h16c70.7 0 128 57.3 128 128V358.7c28.3 12.3 48 40.5 48 73.3c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-32.8 19.7-61 48-73.3V192c0-35.3-28.7-64-64-64H320v40c0 9.5-5.6 18.1-14.2 21.9s-18.8 2.3-25.8-4.1l-80-72c-5.1-4.6-7.9-11-7.9-17.8s2.9-13.3 7.9-17.8l80-72c7-6.3 17.2-7.9 25.8-4.1zM104 80A24 24 0 1 0 56 80a24 24 0 1 0 48 0zm8 73.3V358.7c28.3 12.3 48 40.5 48 73.3c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-32.8 19.7-61 48-73.3V153.3C19.7 141 0 112.8 0 80C0 35.8 35.8 0 80 0s80 35.8 80 80c0 32.8-19.7 61-48 73.3zM104 432a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zm328 24a24 24 0 1 0 0-48 24 24 0 1 0 0 48z"],"clipboard-list":[384,512,[],"f46d","M192 0c-41.8 0-77.4 26.7-90.5 64H64C28.7 64 0 92.7 0 128V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H282.5C269.4 26.7 233.8 0 192 0zm0 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM72 272a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm104-16H304c8.8 0 16 7.2 16 16s-7.2 16-16 16H176c-8.8 0-16-7.2-16-16s7.2-16 16-16zM72 368a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm88 0c0-8.8 7.2-16 16-16H304c8.8 0 16 7.2 16 16s-7.2 16-16 16H176c-8.8 0-16-7.2-16-16z"],"truck-ramp-box":[640,512,["truck-loading"],"f4de","M640 0V400c0 61.9-50.1 112-112 112c-61 0-110.5-48.7-112-109.3L48.4 502.9c-17.1 4.6-34.6-5.4-39.3-22.5s5.4-34.6 22.5-39.3L352 353.8V64c0-35.3 28.7-64 64-64H640zM576 400a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM23.1 207.7c-4.6-17.1 5.6-34.6 22.6-39.2l46.4-12.4 20.7 77.3c2.3 8.5 11.1 13.6 19.6 11.3l30.9-8.3c8.5-2.3 13.6-11.1 11.3-19.6l-20.7-77.3 46.4-12.4c17.1-4.6 34.6 5.6 39.2 22.6l41.4 154.5c4.6 17.1-5.6 34.6-22.6 39.2L103.7 384.9c-17.1 4.6-34.6-5.6-39.2-22.6L23.1 207.7z"],"user-check":[640,512,[],"f4fc","M96 128a128 128 0 1 1 256 0A128 128 0 1 1 96 128zM0 482.3C0 383.8 79.8 304 178.3 304h91.4C368.2 304 448 383.8 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3zM625 177L497 305c-9.4 9.4-24.6 9.4-33.9 0l-64-64c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47L591 143c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z"],"vial-virus":[512,512,[],"e597","M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96V384c0 53 43 96 96 96c28.6 0 54.2-12.5 71.8-32.3c.1-14.2 5.6-28.3 16.4-39.1c.2-.2 .1-.6-.2-.6c-30.9 0-56-25.1-56-56s25.1-56 56-56c.3 0 .4-.4 .2-.6c-21.9-21.9-21.9-57.3 0-79.2c2.4-2.4 5-4.6 7.8-6.5V96c17.7 0 32-14.3 32-32s-14.3-32-32-32H160 96 32zM96 192V96h64v96H96zM216 376c28.8 0 43.2 34.8 22.9 55.2c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0c20.4-20.4 55.2-5.9 55.2 22.9c0 13.3 10.7 24 24 24s24-10.7 24-24c0-28.8 34.8-43.2 55.2-22.9c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9C444.8 410.8 459.2 376 488 376c13.3 0 24-10.7 24-24s-10.7-24-24-24c-28.8 0-43.2-34.8-22.9-55.2c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0C410.8 259.2 376 244.8 376 216c0-13.3-10.7-24-24-24s-24 10.7-24 24c0 28.8-34.8 43.2-55.2 22.9c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9c20.4 20.4 5.9 55.2-22.9 55.2c-13.3 0-24 10.7-24 24s10.7 24 24 24zm104-88a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm40 96a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z"],"sheet-plastic":[384,512,[],"e571","M0 448c0 35.3 28.7 64 64 64H224V384c0-17.7 14.3-32 32-32H384V64c0-35.3-28.7-64-64-64H64C28.7 0 0 28.7 0 64V448zM171.3 75.3l-96 96c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6l96-96c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6zm96 32l-160 160c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6l160-160c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6zM384 384H256V512L384 384z"],blog:[512,512,[],"f781","M192 32c0 17.7 14.3 32 32 32c123.7 0 224 100.3 224 224c0 17.7 14.3 32 32 32s32-14.3 32-32C512 128.9 383.1 0 224 0c-17.7 0-32 14.3-32 32zm0 96c0 17.7 14.3 32 32 32c70.7 0 128 57.3 128 128c0 17.7 14.3 32 32 32s32-14.3 32-32c0-106-86-192-192-192c-17.7 0-32 14.3-32 32zM96 144c0-26.5-21.5-48-48-48S0 117.5 0 144V368c0 79.5 64.5 144 144 144s144-64.5 144-144s-64.5-144-144-144H128v96h16c26.5 0 48 21.5 48 48s-21.5 48-48 48s-48-21.5-48-48V144z"],"user-ninja":[448,512,[129399],"f504","M224 256c-57.2 0-105.6-37.5-122-89.3c-1.1 1.3-2.2 2.6-3.5 3.8c-15.8 15.8-38.8 20.7-53.6 22.1c-8.1 .8-14.6-5.7-13.8-13.8c1.4-14.7 6.3-37.8 22.1-53.6c5.8-5.8 12.6-10.1 19.6-13.4c-7-3.2-13.8-7.6-19.6-13.4C37.4 82.7 32.6 59.7 31.1 44.9c-.8-8.1 5.7-14.6 13.8-13.8c14.7 1.4 37.8 6.3 53.6 22.1c4.8 4.8 8.7 10.4 11.7 16.1C131.4 28.2 174.4 0 224 0c70.7 0 128 57.3 128 128s-57.3 128-128 128zM0 482.3C0 399.5 56.4 330 132.8 309.9c6-1.6 12.2 .9 15.9 5.8l62.5 83.3c6.4 8.5 19.2 8.5 25.6 0l62.5-83.3c3.7-4.9 9.9-7.4 15.9-5.8C391.6 330 448 399.5 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3zM160 96c-8.8 0-16 7.2-16 16s7.2 16 16 16H288c8.8 0 16-7.2 16-16s-7.2-16-16-16H160z"],"person-arrow-up-from-line":[640,512,[],"e539","M192 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-8 352V352h16v96H184zm-64 0H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H152h80H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H264V256.9l28.6 47.5c9.1 15.1 28.8 20 43.9 10.9s20-28.8 10.9-43.9l-58.3-97c-17.4-28.9-48.6-46.6-82.3-46.6H177.1c-33.7 0-64.9 17.7-82.3 46.6l-58.3 97c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9L120 256.9V448zM598.6 121.4l-80-80c-12.5-12.5-32.8-12.5-45.3 0l-80 80c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L464 141.3 464 384c0 17.7 14.3 32 32 32s32-14.3 32-32V141.3l25.4 25.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3z"],"scroll-torah":[640,512,["torah"],"f6a0","M96 480V32C96 14.3 74.5 0 48 0S0 14.3 0 32V480c0 17.7 21.5 32 48 32s48-14.3 48-32zM512 32H128V480H512V32zM592 0c-26.5 0-48 14.3-48 32V480c0 17.7 21.5 32 48 32s48-14.3 48-32V32c0-17.7-21.5-32-48-32zM196 313.7c0-3.2 .9-6.4 2.5-9.2L226.7 256l-28.3-48.5c-1.6-2.8-2.5-6-2.5-9.2c0-10.1 8.2-18.3 18.3-18.3H271l31.4-53.9c3.6-6.3 10.3-10.1 17.6-10.1s13.9 3.8 17.6 10.1L369 180h56.7c10.1 0 18.3 8.2 18.3 18.3c0 3.2-.9 6.4-2.5 9.2L413.3 256l28.3 48.5c1.6 2.8 2.5 6 2.5 9.2c0 10.1-8.2 18.3-18.3 18.3H369l-31.4 53.9c-3.6 6.3-10.3 10.1-17.6 10.1s-13.9-3.8-17.6-10.1L271 332H214.3c-10.1 0-18.3-8.2-18.3-18.3zm124 54.7L341.2 332H298.8L320 368.4zM254.5 256l30.3 52h70.4l30.3-52-30.3-52H284.8l-30.3 52zm144.9 23.8L383 308h32.8l-16.4-28.2zM415.8 204H383l16.4 28.2L415.8 204zM320 143.6L298.8 180h42.4L320 143.6zM224.2 204l16.4 28.2L257 204H224.2zM257 308l-16.4-28.2L224.2 308H257z"],"broom-ball":[576,512,["quidditch","quidditch-broom-ball"],"f458","M566.6 9.4c12.5 12.5 12.5 32.8 0 45.3l-192 192 34.7 34.7c4.2 4.2 6.6 10 6.6 16c0 12.5-10.1 22.6-22.6 22.6H364.3L256 211.7V182.6c0-12.5 10.1-22.6 22.6-22.6c6 0 11.8 2.4 16 6.6l34.7 34.7 192-192c12.5-12.5 32.8-12.5 45.3 0zm-344 225.5L341.1 353.4c3.7 42.7-11.7 85.2-42.3 115.8C271.4 496.6 234.2 512 195.5 512L22.1 512C9.9 512 0 502.1 0 489.9c0-6.3 2.7-12.3 7.3-16.5L133.7 359.7c4.2-3.7-.4-10.4-5.4-7.9L77.2 377.4c-6.1 3-13.2-1.4-13.2-8.2c0-31.5 12.5-61.7 34.8-84l8-8c30.6-30.6 73.1-45.9 115.8-42.3zM464 352a80 80 0 1 1 0 160 80 80 0 1 1 0-160z"],"toggle-off":[576,512,[],"f204","M384 128c70.7 0 128 57.3 128 128s-57.3 128-128 128H192c-70.7 0-128-57.3-128-128s57.3-128 128-128H384zM576 256c0-106-86-192-192-192H192C86 64 0 150 0 256S86 448 192 448H384c106 0 192-86 192-192zM192 352a96 96 0 1 0 0-192 96 96 0 1 0 0 192z"],"box-archive":[512,512,["archive"],"f187","M32 32H480c17.7 0 32 14.3 32 32V96c0 17.7-14.3 32-32 32H32C14.3 128 0 113.7 0 96V64C0 46.3 14.3 32 32 32zm0 128H480V416c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V160zm128 80c0 8.8 7.2 16 16 16H336c8.8 0 16-7.2 16-16s-7.2-16-16-16H176c-8.8 0-16 7.2-16 16z"],"person-drowning":[576,512,[],"e545","M192 64c0-17.7-14.3-32-32-32s-32 14.3-32 32V96.2c0 54.1 23.5 104 62.2 138.3l-21 146.7c7.8 2.1 15.5 3.3 22.8 3.3c21.1 0 42-8.5 59.2-20.3c22.1-15.5 51.6-15.5 73.7 0c12.4 8.5 26.1 14.8 39.7 18l17.7-97.6c10.7-1.2 21.3-3.1 31.9-5.5l105-23.9c17.2-3.9 28-21.1 24.1-38.3s-21.1-28-38.3-24.1L400 216.6c-41 9.3-83.7 7.5-123.7-5.2c-50.2-16-84.3-62.6-84.3-115.3V64zM320 192a64 64 0 1 0 0-128 64 64 0 1 0 0 128zM306.5 389.9c-11.1-7.9-25.9-7.9-37 0C247 405.4 219.5 416 192 416c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C126.5 469.7 159 480 192 480c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.4-27.3-10.1-39.2-1.7l0 0C439.4 405.2 410.9 416 384 416c-27.5 0-55-10.6-77.5-26.1z"],"arrow-down-9-1":[576,512,["sort-numeric-desc","sort-numeric-down-alt"],"f886","M160 480c9 0 17.5-3.8 23.6-10.4l88-96c11.9-13 11.1-33.3-2-45.2s-33.3-11.1-45.2 2L192 365.7V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V365.7L95.6 330.4c-11.9-13-32.2-13.9-45.2-2s-13.9 32.2-2 45.2l88 96C142.5 476.2 151 480 160 480zM450.7 294c-8.3-6-19.1-7.7-28.8-4.4l-48 16c-16.8 5.6-25.8 23.7-20.2 40.5s23.7 25.8 40.5 20.2l5.9-2V416H384c-17.7 0-32 14.3-32 32s14.3 32 32 32h48 48c17.7 0 32-14.3 32-32s-14.3-32-32-32H464V320c0-10.3-4.9-19.9-13.3-26zM418.3 91a32 32 0 1 1 27.4 57.9A32 32 0 1 1 418.3 91zM405.1 203.8l-6.8 9.2c-10.5 14.2-7.5 34.2 6.7 44.8s34.2 7.5 44.8-6.7l48.8-65.8c14-18.9 21.5-41.7 21.5-65.2c0-48.6-39.4-88-88-88s-88 39.4-88 88c0 39.2 25.6 72.4 61.1 83.8z"],"face-grin-tongue-squint":[512,512,[128541,"grin-tongue-squint"],"f58a","M0 256C0 368.9 73.1 464.7 174.5 498.8C165.3 484 160 466.6 160 448V400.7c-24-17.5-43.1-41.4-54.8-69.2c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19c12.3-3.8 24.3 6.9 19.3 18.7c-11.8 28-31.1 52-55.4 69.6V448c0 18.6-5.3 36-14.5 50.8C438.9 464.7 512 368.9 512 256C512 114.6 397.4 0 256 0S0 114.6 0 256zM116 141.1c0-9 9.6-14.7 17.5-10.5l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6zm262.5-10.5c7.9-4.2 17.5 1.5 17.5 10.5c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9zM320 448V402.6c0-14.7-11.9-26.6-26.6-26.6h-2c-11.3 0-21.1 7.9-23.6 18.9c-2.8 12.6-20.8 12.6-23.6 0c-2.5-11.1-12.3-18.9-23.6-18.9h-2c-14.7 0-26.6 11.9-26.6 26.6V448c0 35.3 28.7 64 64 64s64-28.7 64-64z"],"spray-can":[512,512,[],"f5bd","M128 0h64c17.7 0 32 14.3 32 32v96H96V32c0-17.7 14.3-32 32-32zM0 256c0-53 43-96 96-96H224c53 0 96 43 96 96V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V256zm240 80A80 80 0 1 0 80 336a80 80 0 1 0 160 0zM256 64a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM384 32a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm64 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm32 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM448 256a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM384 128a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"truck-monster":[640,512,[],"f63b","M288 64v64H416L368 64H288zM419.2 25.6L496 128h80c17.7 0 32 14.3 32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32c-29.2-38.9-75.7-64-128-64s-98.8 25.1-128 64H288c-29.2-38.9-75.7-64-128-64s-98.8 25.1-128 64c-17.7 0-32-14.3-32-32s14.3-32 32-32V160c0-17.7 14.3-32 32-32H224V48c0-26.5 21.5-48 48-48h96c20.1 0 39.1 9.5 51.2 25.6zM152 256h16c12.1 0 22.1 8.9 23.8 20.6c7.6 2.2 14.9 5.3 21.7 9c9.4-7 22.8-6.3 31.3 2.3l11.3 11.3c8.6 8.6 9.3 21.9 2.3 31.3c3.7 6.8 6.8 14.1 9 21.7c11.6 1.7 20.6 11.7 20.6 23.8v16c0 12.1-8.9 22.1-20.6 23.8c-2.2 7.6-5.3 14.9-9 21.7c7 9.4 6.3 22.8-2.3 31.3l-11.3 11.3c-8.6 8.6-21.9 9.3-31.3 2.2c-6.8 3.7-14.1 6.8-21.7 9C190.1 503.1 180.1 512 168 512H152c-12.1 0-22.1-8.9-23.8-20.6c-7.6-2.2-14.9-5.3-21.7-9c-9.4 7.1-22.8 6.3-31.3-2.2L63.8 468.9c-8.6-8.6-9.3-21.9-2.3-31.3c-3.7-6.9-6.8-14.1-9-21.8C40.9 414.1 32 404.1 32 392V376c0-12.1 8.9-22.1 20.6-23.8c2.2-7.6 5.3-14.9 9-21.8c-7-9.4-6.3-22.8 2.3-31.3l11.3-11.3c8.6-8.6 21.9-9.3 31.3-2.3c6.8-3.7 14.1-6.8 21.7-9c1.7-11.6 11.7-20.6 23.8-20.6zm8 176a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM448.2 276.6c1.7-11.6 11.7-20.6 23.8-20.6h16c12.1 0 22.1 8.9 23.8 20.6c7.6 2.2 14.9 5.3 21.8 9c9.4-7 22.8-6.3 31.3 2.3l11.3 11.3c8.6 8.6 9.3 21.9 2.2 31.3c3.7 6.8 6.8 14.1 9 21.7c11.6 1.7 20.6 11.7 20.6 23.8v16c0 12.1-8.9 22.1-20.6 23.8c-2.2 7.6-5.3 14.9-9 21.7c7 9.4 6.3 22.8-2.2 31.3l-11.3 11.3c-8.6 8.6-21.9 9.3-31.3 2.2c-6.9 3.7-14.1 6.8-21.8 9C510.1 503.1 500.1 512 488 512H472c-12.1 0-22.1-8.9-23.8-20.6c-7.6-2.2-14.9-5.3-21.7-9c-9.4 7.1-22.8 6.3-31.3-2.2l-11.3-11.3c-8.6-8.6-9.3-21.9-2.2-31.3c-3.7-6.9-6.8-14.1-9-21.8C360.9 414.1 352 404.1 352 392V376c0-12.1 8.9-22.1 20.6-23.8c2.2-7.6 5.3-14.9 9-21.8c-7-9.4-6.3-22.8 2.2-31.3l11.3-11.3c8.6-8.6 21.9-9.3 31.3-2.3c6.8-3.7 14.1-6.8 21.7-9zM528 384a48 48 0 1 0 -96 0 48 48 0 1 0 96 0z"],w:[576,512,[119],"57","M20.8 34c16.5-6.2 35 2.2 41.2 18.7l110.2 294L257.3 55c4-13.7 16.5-23 30.7-23s26.7 9.4 30.7 23l85.1 291.7L514 52.8c6.2-16.5 24.6-24.9 41.2-18.7s24.9 24.7 18.7 41.2l-144 384c-4.8 12.9-17.4 21.3-31.2 20.7s-25.7-9.8-29.5-23L288 178.3 206.7 457c-3.9 13.2-15.8 22.5-29.5 23s-26.3-7.8-31.2-20.7L2 75.2C-4.2 58.7 4.2 40.2 20.8 34z"],"earth-africa":[512,512,[127757,"globe-africa"],"f57c","M177.8 63.2l10 17.4c2.8 4.8 4.2 10.3 4.2 15.9v41.4c0 3.9 1.6 7.7 4.3 10.4c6.2 6.2 16.5 5.7 22-1.2l13.6-17c4.7-5.9 12.9-7.7 19.6-4.3l15.2 7.6c3.4 1.7 7.2 2.6 11 2.6c6.5 0 12.8-2.6 17.4-7.2l3.9-3.9c2.9-2.9 7.3-3.6 11-1.8l29.2 14.6c7.8 3.9 12.6 11.8 12.6 20.5c0 10.5-7.1 19.6-17.3 22.2l-35.4 8.8c-7.4 1.8-15.1 1.5-22.4-.9l-32-10.7c-3.3-1.1-6.7-1.7-10.2-1.7c-7 0-13.8 2.3-19.4 6.5L176 212c-10.1 7.6-16 19.4-16 32v28c0 26.5 21.5 48 48 48h32c8.8 0 16 7.2 16 16v48c0 17.7 14.3 32 32 32c10.1 0 19.6-4.7 25.6-12.8l25.6-34.1c8.3-11.1 12.8-24.6 12.8-38.4V318.6c0-3.9 2.6-7.3 6.4-8.2l5.3-1.3c11.9-3 20.3-13.7 20.3-26c0-7.1-2.8-13.9-7.8-18.9l-33.5-33.5c-3.7-3.7-3.7-9.7 0-13.4c5.7-5.7 14.1-7.7 21.8-5.1l14.1 4.7c12.3 4.1 25.7-1.5 31.5-13c3.5-7 11.2-10.8 18.9-9.2l27.4 5.5C432 112.4 351.5 48 256 48c-27.7 0-54 5.4-78.2 15.2zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256z"],rainbow:[640,512,[127752],"f75b","M320 96C178.6 96 64 210.6 64 352v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352C0 175.3 143.3 32 320 32s320 143.3 320 320v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352C576 210.6 461.4 96 320 96zm0 192c-35.3 0-64 28.7-64 64v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352c0-70.7 57.3-128 128-128s128 57.3 128 128v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352c0-35.3-28.7-64-64-64zM160 352v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352c0-123.7 100.3-224 224-224s224 100.3 224 224v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352c0-88.4-71.6-160-160-160s-160 71.6-160 160z"],"circle-notch":[512,512,[],"f1ce","M222.7 32.1c5 16.9-4.6 34.8-21.5 39.8C121.8 95.6 64 169.1 64 256c0 106 86 192 192 192s192-86 192-192c0-86.9-57.8-160.4-137.1-184.1c-16.9-5-26.6-22.9-21.5-39.8s22.9-26.6 39.8-21.5C434.9 42.1 512 140 512 256c0 141.4-114.6 256-256 256S0 397.4 0 256C0 140 77.1 42.1 182.9 10.6c16.9-5 34.8 4.6 39.8 21.5z"],"tablet-screen-button":[448,512,["tablet-alt"],"f3fa","M0 64C0 28.7 28.7 0 64 0H384c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM256 448a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM384 64H64V384H384V64z"],paw:[512,512,[],"f1b0","M226.5 92.9c14.3 42.9-.3 86.2-32.6 96.8s-70.1-15.6-84.4-58.5s.3-86.2 32.6-96.8s70.1 15.6 84.4 58.5zM100.4 198.6c18.9 32.4 14.3 70.1-10.2 84.1s-59.7-.9-78.5-33.3S-2.7 179.3 21.8 165.3s59.7 .9 78.5 33.3zM69.2 401.2C121.6 259.9 214.7 224 256 224s134.4 35.9 186.8 177.2c3.6 9.7 5.2 20.1 5.2 30.5v1.6c0 25.8-20.9 46.7-46.7 46.7c-11.5 0-22.9-1.4-34-4.2l-88-22c-15.3-3.8-31.3-3.8-46.6 0l-88 22c-11.1 2.8-22.5 4.2-34 4.2C84.9 480 64 459.1 64 433.3v-1.6c0-10.4 1.6-20.8 5.2-30.5zM421.8 282.7c-24.5-14-29.1-51.7-10.2-84.1s54-47.3 78.5-33.3s29.1 51.7 10.2 84.1s-54 47.3-78.5 33.3zM310.1 189.7c-32.3-10.6-46.9-53.9-32.6-96.8s52.1-69.1 84.4-58.5s46.9 53.9 32.6 96.8s-52.1 69.1-84.4 58.5z"],cloud:[640,512,[9729],"f0c2","M0 336c0 79.5 64.5 144 144 144H512c70.7 0 128-57.3 128-128c0-61.9-44-113.6-102.4-125.4c4.1-10.7 6.4-22.4 6.4-34.6c0-53-43-96-96-96c-19.7 0-38.1 6-53.3 16.2C367 64.2 315.3 32 256 32C167.6 32 96 103.6 96 192c0 2.7 .1 5.4 .2 8.1C40.2 219.8 0 273.2 0 336z"],"trowel-bricks":[512,512,[],"e58a","M240.8 4.8C250.3 10.6 256 20.9 256 32v72h89c3.6-13.8 16.1-24 31-24h88c26.5 0 48 21.5 48 48s-21.5 48-48 48H376c-14.9 0-27.4-10.2-31-24H256v72c0 11.1-5.7 21.4-15.2 27.2s-21.2 6.4-31.1 1.4l-192-96C6.8 151.2 0 140.1 0 128s6.8-23.2 17.7-28.6l192-96c9.9-5 21.7-4.4 31.1 1.4zM288 256c0-17.7 14.3-32 32-32H480c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H320c-17.7 0-32-14.3-32-32V256zM32 384h96c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32V416c0-17.7 14.3-32 32-32zm192 0H480c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H224c-17.7 0-32-14.3-32-32V416c0-17.7 14.3-32 32-32z"],"face-flushed":[512,512,[128563,"flushed"],"f579","M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM176 384c0 8.8 7.2 16 16 16H320c8.8 0 16-7.2 16-16s-7.2-16-16-16H192c-8.8 0-16 7.2-16 16zm-16-88a72 72 0 1 0 0-144 72 72 0 1 0 0 144zm264-72a72 72 0 1 0 -144 0 72 72 0 1 0 144 0zm-288 0a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm192 0a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z"],"hospital-user":[576,512,[],"f80d","M48 0C21.5 0 0 21.5 0 48V256H144c8.8 0 16 7.2 16 16s-7.2 16-16 16H0v64H144c8.8 0 16 7.2 16 16s-7.2 16-16 16H0v80c0 26.5 21.5 48 48 48H265.9c-6.3-10.2-9.9-22.2-9.9-35.1c0-46.9 25.8-87.8 64-109.2V271.8 48c0-26.5-21.5-48-48-48H48zM152 64h16c8.8 0 16 7.2 16 16v24h24c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H184v24c0 8.8-7.2 16-16 16H152c-8.8 0-16-7.2-16-16V152H112c-8.8 0-16-7.2-16-16V120c0-8.8 7.2-16 16-16h24V80c0-8.8 7.2-16 16-16zM512 272a80 80 0 1 0 -160 0 80 80 0 1 0 160 0zM288 477.1c0 19.3 15.6 34.9 34.9 34.9H541.1c19.3 0 34.9-15.6 34.9-34.9c0-51.4-41.7-93.1-93.1-93.1H381.1c-51.4 0-93.1 41.7-93.1 93.1z"],"tent-arrow-left-right":[576,512,[],"e57f","M488.1 6.2c-9.9-8.9-25-8.1-33.9 1.8s-8.1 25 1.8 33.9L489.5 72 86.5 72l33.5-30.2c9.9-8.9 10.7-24 1.8-33.9S97.8-2.7 87.9 6.2l-80 72C2.9 82.7 0 89.2 0 96s2.9 13.3 7.9 17.8l80 72c9.9 8.9 25 8.1 33.9-1.8s8.1-25-1.8-33.9L86.5 120l402.9 0-33.5 30.2c-9.9 8.9-10.7 24-1.8 33.9s24 10.7 33.9 1.8l80-72c5.1-4.6 7.9-11 7.9-17.8s-2.9-13.3-7.9-17.8l-80-72zM307.4 166.5c-11.5-8.7-27.3-8.7-38.8 0l-168 128c-6.6 5-11 12.5-12.3 20.7l-24 160c-1.4 9.2 1.3 18.6 7.4 25.6S86.7 512 96 512H240h16c17.7 0 32-14.3 32-32V361.9c0-5.5 4.4-9.9 9.9-9.9c3.7 0 7.2 2.1 8.8 5.5l68.4 136.8c5.4 10.8 16.5 17.7 28.6 17.7H464h16c9.3 0 18.2-4.1 24.2-11.1s8.8-16.4 7.4-25.6l-24-160c-1.2-8.2-5.6-15.7-12.3-20.7l-168-128z"],gavel:[512,512,["legal"],"f0e3","M318.6 9.4c-12.5-12.5-32.8-12.5-45.3 0l-120 120c-12.5 12.5-12.5 32.8 0 45.3l16 16c12.5 12.5 32.8 12.5 45.3 0l4-4L325.4 293.4l-4 4c-12.5 12.5-12.5 32.8 0 45.3l16 16c12.5 12.5 32.8 12.5 45.3 0l120-120c12.5-12.5 12.5-32.8 0-45.3l-16-16c-12.5-12.5-32.8-12.5-45.3 0l-4 4L330.6 74.6l4-4c12.5-12.5 12.5-32.8 0-45.3l-16-16zm-152 288c-12.5-12.5-32.8-12.5-45.3 0l-112 112c-12.5 12.5-12.5 32.8 0 45.3l48 48c12.5 12.5 32.8 12.5 45.3 0l112-112c12.5-12.5 12.5-32.8 0-45.3l-1.4-1.4L272 285.3 226.7 240 168 298.7l-1.4-1.4z"],binoculars:[512,512,[],"f1e5","M128 32h32c17.7 0 32 14.3 32 32V96H96V64c0-17.7 14.3-32 32-32zm64 96V448c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32V388.9c0-34.6 9.4-68.6 27.2-98.3C40.9 267.8 49.7 242.4 53 216L60.5 156c2-16 15.6-28 31.8-28H192zm227.8 0c16.1 0 29.8 12 31.8 28L459 216c3.3 26.4 12.1 51.8 25.8 74.6c17.8 29.7 27.2 63.7 27.2 98.3V448c0 17.7-14.3 32-32 32H352c-17.7 0-32-14.3-32-32V128h99.8zM320 64c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V96H320V64zm-32 64V288H224V128h64z"],"microphone-slash":[640,512,[],"f131","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L472.1 344.7c15.2-26 23.9-56.3 23.9-88.7V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v40c0 21.2-5.1 41.1-14.2 58.7L416 300.8V96c0-53-43-96-96-96s-96 43-96 96v54.3L38.8 5.1zm362.5 407l-43.1-33.9C346.1 382 333.3 384 320 384c-70.7 0-128-57.3-128-128v-8.7L144.7 210c-.5 1.9-.7 3.9-.7 6v40c0 89.1 66.2 162.7 152 174.4V464H248c-13.3 0-24 10.7-24 24s10.7 24 24 24h72 72c13.3 0 24-10.7 24-24s-10.7-24-24-24H344V430.4c20.4-2.8 39.7-9.1 57.3-18.2z"],"box-tissue":[512,512,[],"e05b","M92.5 0H208c40 0 52 24 64 48s24 48 64 48h85.2C436 96 448 108 448 122.8c0 3.4-.7 6.8-1.9 10L409.6 224 384 288H128l-16-64L64.9 35.4c-.6-2.3-.9-4.6-.9-6.9C64 12.8 76.8 0 92.5 0zM79 224l16 64H80c-8.8 0-16 7.2-16 16s7.2 16 16 16h48H384h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H418.5l25.6-64H464c26.5 0 48 21.5 48 48V384H0V272c0-26.5 21.5-48 48-48H79zM0 416H512v48c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V416z"],motorcycle:[640,512,[127949],"f21c","M280 32c-13.3 0-24 10.7-24 24s10.7 24 24 24h57.7l16.4 30.3L256 192l-45.3-45.3c-12-12-28.3-18.7-45.3-18.7H64c-17.7 0-32 14.3-32 32v32h96c88.4 0 160 71.6 160 160c0 11-1.1 21.7-3.2 32h70.4c-2.1-10.3-3.2-21-3.2-32c0-52.2 25-98.6 63.7-127.8l15.4 28.6C402.4 276.3 384 312 384 352c0 70.7 57.3 128 128 128s128-57.3 128-128s-57.3-128-128-128c-13.5 0-26.5 2.1-38.7 6L418.2 128H480c17.7 0 32-14.3 32-32V64c0-17.7-14.3-32-32-32H459.6c-7.5 0-14.7 2.6-20.5 7.4L391.7 78.9l-14-26c-7-12.9-20.5-21-35.2-21H280zM462.7 311.2l28.2 52.2c6.3 11.7 20.9 16 32.5 9.7s16-20.9 9.7-32.5l-28.2-52.2c2.3-.3 4.7-.4 7.1-.4c35.3 0 64 28.7 64 64s-28.7 64-64 64s-64-28.7-64-64c0-15.5 5.5-29.7 14.7-40.8zM187.3 376c-9.5 23.5-32.5 40-59.3 40c-35.3 0-64-28.7-64-64s28.7-64 64-64c26.9 0 49.9 16.5 59.3 40h66.4C242.5 268.8 190.5 224 128 224C57.3 224 0 281.3 0 352s57.3 128 128 128c62.5 0 114.5-44.8 125.8-104H187.3zM128 384a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"bell-concierge":[512,512,[128718,"concierge-bell"],"f562","M216 64c-13.3 0-24 10.7-24 24s10.7 24 24 24h16v33.3C119.6 157.2 32 252.4 32 368H480c0-115.6-87.6-210.8-200-222.7V112h16c13.3 0 24-10.7 24-24s-10.7-24-24-24H256 216zM24 400c-13.3 0-24 10.7-24 24s10.7 24 24 24H488c13.3 0 24-10.7 24-24s-10.7-24-24-24H24z"],"pen-ruler":[512,512,["pencil-ruler"],"f5ae","M469.3 19.3l23.4 23.4c25 25 25 65.5 0 90.5l-56.4 56.4L322.3 75.7l56.4-56.4c25-25 65.5-25 90.5 0zM44.9 353.2L299.7 98.3 413.7 212.3 158.8 467.1c-6.7 6.7-15.1 11.6-24.2 14.2l-104 29.7c-8.4 2.4-17.4 .1-23.6-6.1s-8.5-15.2-6.1-23.6l29.7-104c2.6-9.2 7.5-17.5 14.2-24.2zM249.4 103.4L103.4 249.4 16 161.9c-18.7-18.7-18.7-49.1 0-67.9L94.1 16c18.7-18.7 49.1-18.7 67.9 0l19.8 19.8c-.3 .3-.7 .6-1 .9l-64 64c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0l64-64c.3-.3 .6-.7 .9-1l45.1 45.1zM408.6 262.6l45.1 45.1c-.3 .3-.7 .6-1 .9l-64 64c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0l64-64c.3-.3 .6-.7 .9-1L496 350.1c18.7 18.7 18.7 49.1 0 67.9L417.9 496c-18.7 18.7-49.1 18.7-67.9 0l-87.4-87.4L408.6 262.6z"],"people-arrows":[640,512,["people-arrows-left-right"],"e068","M64 64a64 64 0 1 1 128 0A64 64 0 1 1 64 64zM25.9 233.4C29.3 191.9 64 160 105.6 160h44.8c27 0 51 13.4 65.5 34.1c-2.7 1.9-5.2 4-7.5 6.3l-64 64c-21.9 21.9-21.9 57.3 0 79.2L192 391.2V464c0 26.5-21.5 48-48 48H112c-26.5 0-48-21.5-48-48V348.3c-26.5-9.5-44.7-35.8-42.2-65.6l4.1-49.3zM448 64a64 64 0 1 1 128 0A64 64 0 1 1 448 64zM431.6 200.4c-2.3-2.3-4.9-4.4-7.5-6.3c14.5-20.7 38.6-34.1 65.5-34.1h44.8c41.6 0 76.3 31.9 79.7 73.4l4.1 49.3c2.5 29.8-15.7 56.1-42.2 65.6V464c0 26.5-21.5 48-48 48H496c-26.5 0-48-21.5-48-48V391.2l47.6-47.6c21.9-21.9 21.9-57.3 0-79.2l-64-64zM272 240v32h96V240c0-9.7 5.8-18.5 14.8-22.2s19.3-1.7 26.2 5.2l64 64c9.4 9.4 9.4 24.6 0 33.9l-64 64c-6.9 6.9-17.2 8.9-26.2 5.2s-14.8-12.5-14.8-22.2V336H272v32c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-64-64c-9.4-9.4-9.4-24.6 0-33.9l64-64c6.9-6.9 17.2-8.9 26.2-5.2s14.8 12.5 14.8 22.2z"],"mars-and-venus-burst":[640,512,[],"e523","M504 0c-9.7 0-18.5 5.8-22.2 14.8s-1.7 19.3 5.2 26.2l39 39-22.2 22.2C475.9 78.4 439.6 64 400 64c-88.4 0-160 71.6-160 160c0 80.2 59 146.6 136 158.2V408H352c-13.3 0-24 10.7-24 24s10.7 24 24 24h24v32c0 13.3 10.7 24 24 24s24-10.7 24-24V456h24c13.3 0 24-10.7 24-24s-10.7-24-24-24H424V382.2c77-11.6 136-78 136-158.2c0-31.4-9-60.7-24.7-85.4L560 113.9l39 39c6.9 6.9 17.2 8.9 26.2 5.2s14.8-12.5 14.8-22.2V24c0-13.3-10.7-24-24-24H504zM400 128a96 96 0 1 1 0 192 96 96 0 1 1 0-192zM190.9 18.1C188.4 12 182.6 8 176 8s-12.4 4-14.9 10.1l-29.4 74L55.6 68.9c-6.3-1.9-13.1 .2-17.2 5.3s-4.6 12.2-1.4 17.9l39.5 69.1L10.9 206.4c-5.4 3.7-8 10.3-6.5 16.7s6.7 11.2 13.1 12.2l78.7 12.2L90.6 327c-.5 6.5 3.1 12.7 9 15.5s12.9 1.8 17.8-2.6L176 286.1l58.6 53.9c4.1 3.8 9.9 5.1 15.2 3.6C223.6 310.8 208 269.2 208 224c0-60.8 28.3-115 72.4-150.2L220.3 92.1l-29.4-74z"],"square-caret-right":[448,512,["caret-square-right"],"f152","M448 96c0-35.3-28.7-64-64-64L64 32C28.7 32 0 60.7 0 96L0 416c0 35.3 28.7 64 64 64l320 0c35.3 0 64-28.7 64-64l0-320zM320 256c0 6.7-2.8 13-7.7 17.6l-112 104c-7 6.5-17.2 8.2-25.9 4.4s-14.4-12.5-14.4-22l0-208c0-9.5 5.7-18.2 14.4-22s18.9-2.1 25.9 4.4l112 104c4.9 4.5 7.7 10.9 7.7 17.6z"],scissors:[512,512,[9984,9986,9988,"cut"],"f0c4","M256 192l-39.5-39.5c4.9-12.6 7.5-26.2 7.5-40.5C224 50.1 173.9 0 112 0S0 50.1 0 112s50.1 112 112 112c14.3 0 27.9-2.7 40.5-7.5L192 256l-39.5 39.5c-12.6-4.9-26.2-7.5-40.5-7.5C50.1 288 0 338.1 0 400s50.1 112 112 112s112-50.1 112-112c0-14.3-2.7-27.9-7.5-40.5L499.2 76.8c7.1-7.1 7.1-18.5 0-25.6c-28.3-28.3-74.1-28.3-102.4 0L256 192zm22.6 150.6L396.8 460.8c28.3 28.3 74.1 28.3 102.4 0c7.1-7.1 7.1-18.5 0-25.6L342.6 278.6l-64 64zM64 112a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm48 240a48 48 0 1 1 0 96 48 48 0 1 1 0-96z"],"sun-plant-wilt":[640,512,[],"e57a","M160 0c-6.3 0-12 3.7-14.6 9.5L120.6 64.9 63.9 43.2c-5.9-2.3-12.6-.8-17 3.6s-5.9 11.1-3.6 17l21.7 56.7L9.5 145.4C3.7 148 0 153.7 0 160s3.7 12 9.5 14.6l55.4 24.8L43.2 256.1c-2.3 5.9-.8 12.6 3.6 17s11.1 5.9 17 3.6l56.7-21.7 24.8 55.4c2.6 5.8 8.3 9.5 14.6 9.5s12-3.7 14.6-9.5l24.8-55.4 56.7 21.7c5.9 2.3 12.6 .8 17-3.6s5.9-11.1 3.6-17l-21.7-56.7 55.4-24.8c5.8-2.6 9.5-8.3 9.5-14.6s-3.7-12-9.5-14.6l-55.4-24.8 21.7-56.7c2.3-5.9 .8-12.6-3.6-17s-11.1-5.9-17-3.6L199.4 64.9 174.6 9.5C172 3.7 166.3 0 160 0zm0 96a64 64 0 1 1 0 128 64 64 0 1 1 0-128zm32 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm312 16c0-17.7 14.3-32 32-32s32 14.3 32 32v53.4c-14.8 7.7-24 23.1-24 44.6c0 16.8 16 44 37.4 67.2c5.8 6.2 15.5 6.2 21.2 0C624 318 640 290.7 640 274c0-21.5-9.2-37-24-44.6V176c0-44.2-35.8-80-80-80s-80 35.8-80 80v22.7c-9.8-4.3-20.6-6.7-32-6.7c-44.2 0-80 35.8-80 80v21.4c-14.8 7.7-24 23.1-24 44.6c0 16.8 16 44 37.4 67.2c5.8 6.2 15.5 6.2 21.2 0C400 382 416 354.7 416 338c0-21.5-9.2-37-24-44.6V272c0-17.7 14.3-32 32-32s32 14.3 32 32v8V448H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H504V280v-8V176z"],"toilets-portable":[576,512,[],"e584","M32 0H224c17.7 0 32 14.3 32 32V64H0V32C0 14.3 14.3 0 32 0zM0 96H24 232h24v24V488c0 13.3-10.7 24-24 24s-24-10.7-24-24v-8H48v8c0 13.3-10.7 24-24 24s-24-10.7-24-24V120 96zM192 224c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V240c0-8.8-7.2-16-16-16zM352 0H544c17.7 0 32 14.3 32 32V64H320V32c0-17.7 14.3-32 32-32zM320 96h24H552h24v24V488c0 13.3-10.7 24-24 24s-24-10.7-24-24v-8H368v8c0 13.3-10.7 24-24 24s-24-10.7-24-24V120 96zM512 224c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V240c0-8.8-7.2-16-16-16z"],"hockey-puck":[512,512,[],"f453","M256 256C114.6 256 0 213 0 160s114.6-96 256-96s256 43 256 96s-114.6 96-256 96zm192.3 1.8c24.7-9.3 46.9-21 63.7-35.6V352c0 53-114.6 96-256 96S0 405 0 352V222.3c16.8 14.6 39 26.3 63.7 35.6C114.5 276.9 182.5 288 256 288s141.5-11.1 192.3-30.2z"],table:[512,512,[],"f0ce","M64 256V160H224v96H64zm0 64H224v96H64V320zm224 96V320H448v96H288zM448 256H288V160H448v96zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64z"],"magnifying-glass-arrow-right":[512,512,[],"e521","M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zM241 119c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l31 31H120c-13.3 0-24 10.7-24 24s10.7 24 24 24H238.1l-31 31c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l72-72c9.4-9.4 9.4-24.6 0-33.9l-72-72z"],"tachograph-digital":[640,512,["digital-tachograph"],"f566","M64 64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H576c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H64zm32 64H320c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V160c0-17.7 14.3-32 32-32zM64 368c0-8.8 7.2-16 16-16H336c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16zm320 0c0-8.8 7.2-16 16-16H560c8.8 0 16 7.2 16 16s-7.2 16-16 16H400c-8.8 0-16-7.2-16-16zM80 288a16 16 0 1 1 0 32 16 16 0 1 1 0-32zm48 16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zm80-16a16 16 0 1 1 0 32 16 16 0 1 1 0-32zm48 16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zm80-16a16 16 0 1 1 0 32 16 16 0 1 1 0-32z"],"users-slash":[640,512,[],"e073","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L440.6 320H618.7c11.8 0 21.3-9.6 21.3-21.3C640 239.8 592.2 192 533.3 192H490.7c-15.9 0-31 3.5-44.6 9.7c1.3 7.2 1.9 14.7 1.9 22.3c0 30.2-10.5 58-28 79.9l-25.2-19.7C408.1 267.7 416 246.8 416 224c0-53-43-96-96-96c-31.1 0-58.7 14.8-76.3 37.7l-40.6-31.8c13-14.2 20.9-33.1 20.9-53.9c0-44.2-35.8-80-80-80C116.3 0 91.9 14.1 77.5 35.5L38.8 5.1zM106.7 192C47.8 192 0 239.8 0 298.7C0 310.4 9.6 320 21.3 320H234.7c.2 0 .4 0 .7 0c-20.6-18.2-35.2-42.8-40.8-70.8L121.8 192H106.7zM261.3 352C187.7 352 128 411.7 128 485.3c0 14.7 11.9 26.7 26.7 26.7H485.3c10.5 0 19.5-6 23.9-14.8L324.9 352H261.3zM512 160A80 80 0 1 0 512 0a80 80 0 1 0 0 160z"],clover:[448,512,[],"e139","M216.6 49.9C205.1 38.5 189.5 32 173.3 32C139.4 32 112 59.4 112 93.3v4.9c0 12 3.3 23.7 9.4 34l18.8 31.3c1.1 1.8 1.2 3.1 1 4.2c-.2 1.2-.8 2.5-2 3.6s-2.4 1.8-3.6 2c-1 .2-2.4 .1-4.2-1l-31.3-18.8c-10.3-6.2-22-9.4-34-9.4H61.3C27.4 144 0 171.4 0 205.3c0 16.2 6.5 31.8 17.9 43.3l1.2 1.2c3.4 3.4 3.4 9 0 12.4l-1.2 1.2C6.5 274.9 0 290.5 0 306.7C0 340.6 27.4 368 61.3 368h4.9c12 0 23.7-3.3 34-9.4l31.3-18.8c1.8-1.1 3.1-1.2 4.2-1c1.2 .2 2.5 .8 3.6 2s1.8 2.4 2 3.6c.2 1 .1 2.4-1 4.2l-18.8 31.3c-6.2 10.3-9.4 22-9.4 34v4.9c0 33.8 27.4 61.3 61.3 61.3c16.2 0 31.8-6.5 43.3-17.9l1.2-1.2c3.4-3.4 9-3.4 12.4 0l1.2 1.2c11.5 11.5 27.1 17.9 43.3 17.9c33.8 0 61.3-27.4 61.3-61.3v-4.9c0-12-3.3-23.7-9.4-34l-18.8-31.3c-1.1-1.8-1.2-3.1-1-4.2c.2-1.2 .8-2.5 2-3.6s2.4-1.8 3.6-2c1-.2 2.4-.1 4.2 1l31.3 18.8c10.3 6.2 22 9.4 34 9.4h4.9c33.8 0 61.3-27.4 61.3-61.3c0-16.2-6.5-31.8-17.9-43.3l-1.2-1.2c-3.4-3.4-3.4-9 0-12.4l1.2-1.2c11.5-11.5 17.9-27.1 17.9-43.3c0-33.8-27.4-61.3-61.3-61.3h-4.9c-12 0-23.7 3.3-34 9.4l-31.3 18.8c-1.8 1.1-3.1 1.2-4.2 1c-1.2-.2-2.5-.8-3.6-2s-1.8-2.4-2-3.6c-.2-1-.1-2.4 1-4.2l18.8-31.3c6.2-10.3 9.4-22 9.4-34V93.3C336 59.4 308.6 32 274.7 32c-16.2 0-31.8 6.5-43.3 17.9l-1.2 1.2c-3.4 3.4-9 3.4-12.4 0l-1.2-1.2z"],reply:[512,512,[61714,"mail-reply"],"f3e5","M205 34.8c11.5 5.1 19 16.6 19 29.2v64H336c97.2 0 176 78.8 176 176c0 113.3-81.5 163.9-100.2 174.1c-2.5 1.4-5.3 1.9-8.1 1.9c-10.9 0-19.7-8.9-19.7-19.7c0-7.5 4.3-14.4 9.8-19.5c9.4-8.8 22.2-26.4 22.2-56.7c0-53-43-96-96-96H224v64c0 12.6-7.4 24.1-19 29.2s-25 3-34.4-5.4l-160-144C3.9 225.7 0 217.1 0 208s3.9-17.7 10.6-23.8l160-144c9.4-8.5 22.9-10.6 34.4-5.4z"],"star-and-crescent":[512,512,[9770],"f699","M0 256C0 114.6 114.6 0 256 0c33 0 64.6 6.3 93.6 17.7c7.4 2.9 11.5 10.7 9.8 18.4s-8.8 13-16.7 12.4c-4.8-.3-9.7-.5-14.6-.5c-114.9 0-208 93.1-208 208s93.1 208 208 208c4.9 0 9.8-.2 14.6-.5c7.9-.5 15 4.7 16.7 12.4s-2.4 15.5-9.8 18.4C320.6 505.7 289 512 256 512C114.6 512 0 397.4 0 256zM375.4 137.4c3.5-7.1 13.7-7.1 17.2 0l31.5 63.8c1.4 2.8 4.1 4.8 7.2 5.3l70.4 10.2c7.9 1.1 11 10.8 5.3 16.4l-50.9 49.6c-2.3 2.2-3.3 5.4-2.8 8.5l12 70.1c1.3 7.8-6.9 13.8-13.9 10.1l-63-33.1c-2.8-1.5-6.1-1.5-8.9 0l-63 33.1c-7 3.7-15.3-2.3-13.9-10.1l12-70.1c.5-3.1-.5-6.3-2.8-8.5L261 233.1c-5.7-5.6-2.6-15.2 5.3-16.4l70.4-10.2c3.1-.5 5.8-2.4 7.2-5.3l31.5-63.8z"],"house-fire":[640,512,[],"e50c","M288 350.1l0 1.9H256c-17.7 0-32 14.3-32 32v64 24c0 22.1-17.9 40-40 40H160 128.1c-1.5 0-3-.1-4.5-.2c-1.2 .1-2.4 .2-3.6 .2H104c-22.1 0-40-17.9-40-40V360c0-.9 0-1.9 .1-2.8V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L447.3 128.1c-12.3-1-25 3-34.8 11.7c-35.4 31.6-65.6 67.7-87.3 102.8C304.3 276.5 288 314.9 288 350.1zM480 512c-88.4 0-160-71.6-160-160c0-76.7 62.5-144.7 107.2-179.4c5-3.9 10.9-5.8 16.8-5.8c7.9-.1 16 3.1 22 9.2l46 46 11.3-11.3c11.7-11.7 30.6-12.7 42.3-1C624.5 268 640 320.2 640 352c0 88.4-71.6 160-160 160zm64-111.8c0-36.5-37-73-54.8-88.4c-5.4-4.7-13.1-4.7-18.5 0C453 327.1 416 363.6 416 400.2c0 35.3 28.7 64 64 64s64-28.7 64-64z"],"square-minus":[448,512,[61767,"minus-square"],"f146","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm88 200H296c13.3 0 24 10.7 24 24s-10.7 24-24 24H152c-13.3 0-24-10.7-24-24s10.7-24 24-24z"],helicopter:[640,512,[128641],"f533","M128 32c0-17.7 14.3-32 32-32H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H384v64h32c88.4 0 160 71.6 160 160v64c0 17.7-14.3 32-32 32H384 320c-20.1 0-39.1-9.5-51.2-25.6l-71.4-95.2c-3.5-4.7-8.3-8.3-13.7-10.5L47.2 198.1c-9.5-3.8-16.7-12-19.2-22L5 83.9C2.4 73.8 10.1 64 20.5 64H48c10.1 0 19.6 4.7 25.6 12.8L112 128H320V64H160c-17.7 0-32-14.3-32-32zM384 320H512V288c0-53-43-96-96-96H384V320zM630.6 425.4c12.5 12.5 12.5 32.8 0 45.3l-3.9 3.9c-24 24-56.6 37.5-90.5 37.5H256c-17.7 0-32-14.3-32-32s14.3-32 32-32H536.2c17 0 33.3-6.7 45.3-18.7l3.9-3.9c12.5-12.5 32.8-12.5 45.3 0z"],compass:[512,512,[129517],"f14e","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm50.7-186.9L162.4 380.6c-19.4 7.5-38.5-11.6-31-31l55.5-144.3c3.3-8.5 9.9-15.1 18.4-18.4l144.3-55.5c19.4-7.5 38.5 11.6 31 31L325.1 306.7c-3.2 8.5-9.9 15.1-18.4 18.4zM288 256a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z"],"square-caret-down":[448,512,["caret-square-down"],"f150","M384 480c35.3 0 64-28.7 64-64l0-320c0-35.3-28.7-64-64-64L64 32C28.7 32 0 60.7 0 96L0 416c0 35.3 28.7 64 64 64l320 0zM224 352c-6.7 0-13-2.8-17.6-7.7l-104-112c-6.5-7-8.2-17.2-4.4-25.9s12.5-14.4 22-14.4l208 0c9.5 0 18.2 5.7 22 14.4s2.1 18.9-4.4 25.9l-104 112c-4.5 4.9-10.9 7.7-17.6 7.7z"],"file-circle-question":[576,512,[],"e4ef","M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v38.6C310.1 219.5 256 287.4 256 368c0 59.1 29.1 111.3 73.7 143.3c-3.2 .5-6.4 .7-9.7 .7H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zm48 96a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm0 240a24 24 0 1 0 0-48 24 24 0 1 0 0 48zM368 321.6V328c0 8.8 7.2 16 16 16s16-7.2 16-16v-6.4c0-5.3 4.3-9.6 9.6-9.6h40.5c7.7 0 13.9 6.2 13.9 13.9c0 5.2-2.9 9.9-7.4 12.3l-32 16.8c-5.3 2.8-8.6 8.2-8.6 14.2V384c0 8.8 7.2 16 16 16s16-7.2 16-16v-5.1l23.5-12.3c15.1-7.9 24.5-23.6 24.5-40.6c0-25.4-20.6-45.9-45.9-45.9H409.6c-23 0-41.6 18.6-41.6 41.6z"],"laptop-code":[640,512,[],"f5fc","M64 96c0-35.3 28.7-64 64-64H512c35.3 0 64 28.7 64 64V352H512V96H128V352H64V96zM0 403.2C0 392.6 8.6 384 19.2 384H620.8c10.6 0 19.2 8.6 19.2 19.2c0 42.4-34.4 76.8-76.8 76.8H76.8C34.4 480 0 445.6 0 403.2zM281 209l-31 31 31 31c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-48-48c-9.4-9.4-9.4-24.6 0-33.9l48-48c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9zM393 175l48 48c9.4 9.4 9.4 24.6 0 33.9l-48 48c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l31-31-31-31c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0z"],swatchbook:[512,512,[],"f5c3","M0 32C0 14.3 14.3 0 32 0H160c17.7 0 32 14.3 32 32V416c0 53-43 96-96 96s-96-43-96-96V32zM223.6 425.9c.3-3.3 .4-6.6 .4-9.9V154l75.4-75.4c12.5-12.5 32.8-12.5 45.3 0l90.5 90.5c12.5 12.5 12.5 32.8 0 45.3L223.6 425.9zM182.8 512l192-192H480c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H182.8zM128 64H64v64h64V64zM64 192v64h64V192H64zM96 440a24 24 0 1 0 0-48 24 24 0 1 0 0 48z"],"prescription-bottle":[384,512,[],"f485","M0 32C0 14.3 14.3 0 32 0H352c17.7 0 32 14.3 32 32V64c0 17.7-14.3 32-32 32H32C14.3 96 0 81.7 0 64V32zm32 96H352V448c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V416H144c8.8 0 16-7.2 16-16s-7.2-16-16-16H32V320H144c8.8 0 16-7.2 16-16s-7.2-16-16-16H32V224H144c8.8 0 16-7.2 16-16s-7.2-16-16-16H32V128z"],bars:[448,512,["navicon"],"f0c9","M0 96C0 78.3 14.3 64 32 64H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 128 0 113.7 0 96zM0 256c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32zM448 416c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H416c17.7 0 32 14.3 32 32z"],"people-group":[640,512,[],"e533","M72 88a56 56 0 1 1 112 0A56 56 0 1 1 72 88zM64 245.7C54 256.9 48 271.8 48 288s6 31.1 16 42.3V245.7zm144.4-49.3C178.7 222.7 160 261.2 160 304c0 34.3 12 65.8 32 90.5V416c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V389.2C26.2 371.2 0 332.7 0 288c0-61.9 50.1-112 112-112h32c24 0 46.2 7.5 64.4 20.3zM448 416V394.5c20-24.7 32-56.2 32-90.5c0-42.8-18.7-81.3-48.4-107.7C449.8 183.5 472 176 496 176h32c61.9 0 112 50.1 112 112c0 44.7-26.2 83.2-64 101.2V416c0 17.7-14.3 32-32 32H480c-17.7 0-32-14.3-32-32zm8-328a56 56 0 1 1 112 0A56 56 0 1 1 456 88zM576 245.7v84.7c10-11.3 16-26.1 16-42.3s-6-31.1-16-42.3zM320 32a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM240 304c0 16.2 6 31 16 42.3V261.7c-10 11.3-16 26.1-16 42.3zm144-42.3v84.7c10-11.3 16-26.1 16-42.3s-6-31.1-16-42.3zM448 304c0 44.7-26.2 83.2-64 101.2V448c0 17.7-14.3 32-32 32H288c-17.7 0-32-14.3-32-32V405.2c-37.8-18-64-56.5-64-101.2c0-61.9 50.1-112 112-112h32c61.9 0 112 50.1 112 112z"],"hourglass-end":[384,512,[8987,"hourglass-3"],"f253","M32 0C14.3 0 0 14.3 0 32S14.3 64 32 64V75c0 42.4 16.9 83.1 46.9 113.1L146.7 256 78.9 323.9C48.9 353.9 32 394.6 32 437v11c-17.7 0-32 14.3-32 32s14.3 32 32 32H64 320h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V437c0-42.4-16.9-83.1-46.9-113.1L237.3 256l67.9-67.9c30-30 46.9-70.7 46.9-113.1V64c17.7 0 32-14.3 32-32s-14.3-32-32-32H320 64 32zM96 75V64H288V75c0 25.5-10.1 49.9-28.1 67.9L192 210.7l-67.9-67.9C106.1 124.9 96 100.4 96 75z"],"heart-crack":[512,512,[128148,"heart-broken"],"f7a9","M119.4 44.1c23.3-3.9 46.8-1.9 68.6 5.3l49.8 77.5-75.4 75.4c-1.5 1.5-2.4 3.6-2.3 5.8s1 4.2 2.6 5.7l112 104c2.9 2.7 7.4 2.9 10.5 .3s3.8-7 1.7-10.4l-60.4-98.1 90.7-75.6c2.6-2.1 3.5-5.7 2.4-8.8L296.8 61.8c28.5-16.7 62.4-23.2 95.7-17.6C461.5 55.6 512 115.2 512 185.1v5.8c0 41.5-17.2 81.2-47.6 109.5L283.7 469.1c-7.5 7-17.4 10.9-27.7 10.9s-20.2-3.9-27.7-10.9L47.6 300.4C17.2 272.1 0 232.4 0 190.9v-5.8c0-69.9 50.5-129.5 119.4-141z"],"square-up-right":[448,512,[8599,"external-link-square-alt"],"f360","M384 32c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H384zM320 313.4V176c0-8.8-7.2-16-16-16H166.6c-12.5 0-22.6 10.1-22.6 22.6c0 6 2.4 11.8 6.6 16L184 232l-66.3 66.3C114 302 112 306.9 112 312s2 10 5.7 13.7l36.7 36.7c3.6 3.6 8.5 5.7 13.7 5.7s10-2 13.7-5.7L248 296l33.4 33.4c4.2 4.2 10 6.6 16 6.6c12.5 0 22.6-10.1 22.6-22.6z"],"face-kiss-beam":[512,512,[128537,"kiss-beam"],"f597","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm48.7-198.3c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4c-2.7 1.5-5.7 3-8.7 4.3c3.1 1.3 6 2.7 8.7 4.3c6.6 3.7 12.5 8.2 16.7 13.4c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4C274.7 443.1 257.4 448 240 448c-3.6 0-6.8-2.5-7.7-6s.6-7.2 3.8-9l0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1c-.4-.2-.7-.4-.9-.5l-.2-.1 0 0 0 0 0 0c-2.5-1.4-4.1-4.1-4.1-7s1.6-5.6 4.1-7l0 0 0 0 0 0 0 0 0 0 .2-.1 .3-.2 .6-.4c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1l-.4-.3-.5-.3-.2-.1 0 0 0 0 0 0c-3.2-1.8-4.7-5.5-3.8-9s4.1-6 7.7-6c17.4 0 34.7 4.9 47.9 12.3c6.6 3.7 12.5 8.2 16.7 13.4zm-87.1-84.9l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z"],film:[512,512,[127902],"f008","M0 96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM48 368v32c0 8.8 7.2 16 16 16H96c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H64c-8.8 0-16 7.2-16 16zm368-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H416zM48 240v32c0 8.8 7.2 16 16 16H96c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H64c-8.8 0-16 7.2-16 16zm368-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H416zM48 112v32c0 8.8 7.2 16 16 16H96c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H64c-8.8 0-16 7.2-16 16zM416 96c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H416zM160 128v64c0 17.7 14.3 32 32 32H320c17.7 0 32-14.3 32-32V128c0-17.7-14.3-32-32-32H192c-17.7 0-32 14.3-32 32zm32 160c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H320c17.7 0 32-14.3 32-32V320c0-17.7-14.3-32-32-32H192z"],"ruler-horizontal":[640,512,[],"f547","M0 336c0 26.5 21.5 48 48 48l544 0c26.5 0 48-21.5 48-48l0-160c0-26.5-21.5-48-48-48l-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0c-26.5 0-48 21.5-48 48L0 336z"],"people-robbery":[576,512,[],"e536","M488.2 59.1C478.1 99.6 441.7 128 400 128s-78.1-28.4-88.2-68.9L303 24.2C298.8 7.1 281.4-3.3 264.2 1S236.7 22.6 241 39.8l8.7 34.9c11 44 40.2 79.6 78.3 99.6V480c0 17.7 14.3 32 32 32s32-14.3 32-32V352h16V480c0 17.7 14.3 32 32 32s32-14.3 32-32V174.3c38.1-20 67.3-55.6 78.3-99.6L559 39.8c4.3-17.1-6.1-34.5-23.3-38.8S501.2 7.1 497 24.2l-8.7 34.9zM400 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM80 96A48 48 0 1 0 80 0a48 48 0 1 0 0 96zm-8 32c-35.3 0-64 28.7-64 64v96l0 .6V480c0 17.7 14.3 32 32 32s32-14.3 32-32V352H88V480c0 17.7 14.3 32 32 32s32-14.3 32-32V252.7l13 20.5c5.9 9.2 16.1 14.9 27 14.9h48c17.7 0 32-14.3 32-32s-14.3-32-32-32H209.6l-37.4-58.9C157.6 142 132.1 128 104.7 128H72z"],lightbulb:[384,512,[128161],"f0eb","M272 384c9.6-31.9 29.5-59.1 49.2-86.2l0 0c5.2-7.1 10.4-14.2 15.4-21.4c19.8-28.5 31.4-63 31.4-100.3C368 78.8 289.2 0 192 0S16 78.8 16 176c0 37.3 11.6 71.9 31.4 100.3c5 7.2 10.2 14.3 15.4 21.4l0 0c19.8 27.1 39.7 54.4 49.2 86.2H272zM192 512c44.2 0 80-35.8 80-80V416H112v16c0 44.2 35.8 80 80 80zM112 176c0 8.8-7.2 16-16 16s-16-7.2-16-16c0-61.9 50.1-112 112-112c8.8 0 16 7.2 16 16s-7.2 16-16 16c-44.2 0-80 35.8-80 80z"],"caret-left":[256,512,[],"f0d9","M9.4 278.6c-12.5-12.5-12.5-32.8 0-45.3l128-128c9.2-9.2 22.9-11.9 34.9-6.9s19.8 16.6 19.8 29.6l0 256c0 12.9-7.8 24.6-19.8 29.6s-25.7 2.2-34.9-6.9l-128-128z"],"circle-exclamation":[512,512,["exclamation-circle"],"f06a","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm0-384c13.3 0 24 10.7 24 24V264c0 13.3-10.7 24-24 24s-24-10.7-24-24V152c0-13.3 10.7-24 24-24zM224 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],"school-circle-xmark":[640,512,[],"e56d","M337.8 5.4C327-1.8 313-1.8 302.2 5.4L166.3 96H48C21.5 96 0 117.5 0 144V464c0 26.5 21.5 48 48 48H320v0H256V416c0-35.3 28.7-64 64-64l.3 0h.5c3.4-37.7 18.7-72.1 42.2-99.1C350.2 260 335.6 264 320 264c-48.6 0-88-39.4-88-88s39.4-88 88-88s88 39.4 88 88c0 18.3-5.6 35.3-15.1 49.4c29-21 64.6-33.4 103.1-33.4c59.5 0 112.1 29.6 144 74.8V144c0-26.5-21.5-48-48-48H473.7L337.8 5.4zM96 192h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V208c0-8.8 7.2-16 16-16zm0 128h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zM320 128c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H336V144c0-8.8-7.2-16-16-16zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm22.6-144l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L496 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L473.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L496 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6L518.6 368z"],"arrow-right-from-bracket":[512,512,["sign-out"],"f08b","M502.6 278.6c12.5-12.5 12.5-32.8 0-45.3l-128-128c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L402.7 224 192 224c-17.7 0-32 14.3-32 32s14.3 32 32 32l210.7 0-73.4 73.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l128-128zM160 96c17.7 0 32-14.3 32-32s-14.3-32-32-32L96 32C43 32 0 75 0 128L0 384c0 53 43 96 96 96l64 0c17.7 0 32-14.3 32-32s-14.3-32-32-32l-64 0c-17.7 0-32-14.3-32-32l0-256c0-17.7 14.3-32 32-32l64 0z"],"circle-chevron-down":[512,512,["chevron-circle-down"],"f13a","M256 0a256 256 0 1 0 0 512A256 256 0 1 0 256 0zM135 241c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l87 87 87-87c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9L273 345c-9.4 9.4-24.6 9.4-33.9 0L135 241z"],"unlock-keyhole":[448,512,["unlock-alt"],"f13e","M224 64c-44.2 0-80 35.8-80 80v48H384c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V256c0-35.3 28.7-64 64-64H80V144C80 64.5 144.5 0 224 0c57.5 0 107 33.7 130.1 82.3c7.6 16 .8 35.1-15.2 42.6s-35.1 .8-42.6-15.2C283.4 82.6 255.9 64 224 64zm32 320c17.7 0 32-14.3 32-32s-14.3-32-32-32H192c-17.7 0-32 14.3-32 32s14.3 32 32 32h64z"],"cloud-showers-heavy":[512,512,[],"f740","M96 320c-53 0-96-43-96-96c0-42.5 27.6-78.6 65.9-91.2C64.7 126.1 64 119.1 64 112C64 50.1 114.1 0 176 0c43.1 0 80.5 24.3 99.2 60c14.7-17.1 36.5-28 60.8-28c44.2 0 80 35.8 80 80c0 5.5-.6 10.8-1.6 16c.5 0 1.1 0 1.6 0c53 0 96 43 96 96s-43 96-96 96H96zM81.5 353.9c12.2 5.2 17.8 19.3 12.6 31.5l-48 112c-5.2 12.2-19.3 17.8-31.5 12.6S-3.3 490.7 1.9 478.5l48-112c5.2-12.2 19.3-17.8 31.5-12.6zm120 0c12.2 5.2 17.8 19.3 12.6 31.5l-48 112c-5.2 12.2-19.3 17.8-31.5 12.6s-17.8-19.3-12.6-31.5l48-112c5.2-12.2 19.3-17.8 31.5-12.6zm244.6 31.5l-48 112c-5.2 12.2-19.3 17.8-31.5 12.6s-17.8-19.3-12.6-31.5l48-112c5.2-12.2 19.3-17.8 31.5-12.6s17.8 19.3 12.6 31.5zM313.5 353.9c12.2 5.2 17.8 19.3 12.6 31.5l-48 112c-5.2 12.2-19.3 17.8-31.5 12.6s-17.8-19.3-12.6-31.5l48-112c5.2-12.2 19.3-17.8 31.5-12.6z"],"headphones-simple":[512,512,["headphones-alt"],"f58f","M256 80C141.1 80 48 173.1 48 288V392c0 13.3-10.7 24-24 24s-24-10.7-24-24V288C0 146.6 114.6 32 256 32s256 114.6 256 256V392c0 13.3-10.7 24-24 24s-24-10.7-24-24V288c0-114.9-93.1-208-208-208zM80 352c0-35.3 28.7-64 64-64h16c17.7 0 32 14.3 32 32V448c0 17.7-14.3 32-32 32H144c-35.3 0-64-28.7-64-64V352zm288-64c35.3 0 64 28.7 64 64v64c0 35.3-28.7 64-64 64H352c-17.7 0-32-14.3-32-32V320c0-17.7 14.3-32 32-32h16z"],sitemap:[576,512,[],"f0e8","M208 80c0-26.5 21.5-48 48-48h64c26.5 0 48 21.5 48 48v64c0 26.5-21.5 48-48 48h-8v40H464c30.9 0 56 25.1 56 56v32h8c26.5 0 48 21.5 48 48v64c0 26.5-21.5 48-48 48H464c-26.5 0-48-21.5-48-48V368c0-26.5 21.5-48 48-48h8V288c0-4.4-3.6-8-8-8H312v40h8c26.5 0 48 21.5 48 48v64c0 26.5-21.5 48-48 48H256c-26.5 0-48-21.5-48-48V368c0-26.5 21.5-48 48-48h8V280H112c-4.4 0-8 3.6-8 8v32h8c26.5 0 48 21.5 48 48v64c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V368c0-26.5 21.5-48 48-48h8V288c0-30.9 25.1-56 56-56H264V192h-8c-26.5 0-48-21.5-48-48V80z"],"circle-dollar-to-slot":[512,512,["donate"],"f4b9","M326.7 403.7c-22.1 8-45.9 12.3-70.7 12.3s-48.7-4.4-70.7-12.3c-.3-.1-.5-.2-.8-.3c-30-11-56.8-28.7-78.6-51.4C70 314.6 48 263.9 48 208C48 93.1 141.1 0 256 0S464 93.1 464 208c0 55.9-22 106.6-57.9 144c-1 1-2 2.1-3 3.1c-21.4 21.4-47.4 38.1-76.3 48.6zM256 84c-11 0-20 9-20 20v14c-7.6 1.7-15.2 4.4-22.2 8.5c-13.9 8.3-25.9 22.8-25.8 43.9c.1 20.3 12 33.1 24.7 40.7c11 6.6 24.7 10.8 35.6 14l1.7 .5c12.6 3.8 21.8 6.8 28 10.7c5.1 3.2 5.8 5.4 5.9 8.2c.1 5-1.8 8-5.9 10.5c-5 3.1-12.9 5-21.4 4.7c-11.1-.4-21.5-3.9-35.1-8.5c-2.3-.8-4.7-1.6-7.2-2.4c-10.5-3.5-21.8 2.2-25.3 12.6s2.2 21.8 12.6 25.3c1.9 .6 4 1.3 6.1 2.1l0 0 0 0c8.3 2.9 17.9 6.2 28.2 8.4V312c0 11 9 20 20 20s20-9 20-20V298.2c8-1.7 16-4.5 23.2-9c14.3-8.9 25.1-24.1 24.8-45c-.3-20.3-11.7-33.4-24.6-41.6c-11.5-7.2-25.9-11.6-37.1-15l-.7-.2c-12.8-3.9-21.9-6.7-28.3-10.5c-5.2-3.1-5.3-4.9-5.3-6.7c0-3.7 1.4-6.5 6.2-9.3c5.4-3.2 13.6-5.1 21.5-5c9.6 .1 20.2 2.2 31.2 5.2c10.7 2.8 21.6-3.5 24.5-14.2s-3.5-21.6-14.2-24.5c-6.5-1.7-13.7-3.4-21.1-4.7V104c0-11-9-20-20-20zM48 352H64c19.5 25.9 44 47.7 72.2 64H64v32H256 448V416H375.8c28.2-16.3 52.8-38.1 72.2-64h16c26.5 0 48 21.5 48 48v64c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V400c0-26.5 21.5-48 48-48z"],memory:[576,512,[],"f538","M64 64C28.7 64 0 92.7 0 128v7.4c0 6.8 4.4 12.6 10.1 16.3C23.3 160.3 32 175.1 32 192s-8.7 31.7-21.9 40.3C4.4 236 0 241.8 0 248.6V320H576V248.6c0-6.8-4.4-12.6-10.1-16.3C552.7 223.7 544 208.9 544 192s8.7-31.7 21.9-40.3c5.7-3.7 10.1-9.5 10.1-16.3V128c0-35.3-28.7-64-64-64H64zM576 352H0v64c0 17.7 14.3 32 32 32H80V416c0-8.8 7.2-16 16-16s16 7.2 16 16v32h96V416c0-8.8 7.2-16 16-16s16 7.2 16 16v32h96V416c0-8.8 7.2-16 16-16s16 7.2 16 16v32h96V416c0-8.8 7.2-16 16-16s16 7.2 16 16v32h48c17.7 0 32-14.3 32-32V352zM192 160v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V160c0-17.7 14.3-32 32-32s32 14.3 32 32zm128 0v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V160c0-17.7 14.3-32 32-32s32 14.3 32 32zm128 0v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V160c0-17.7 14.3-32 32-32s32 14.3 32 32z"],"road-spikes":[640,512,[],"e568","M64 116.8c0-15.8 20.5-22 29.3-8.9L192 256V116.8c0-15.8 20.5-22 29.3-8.9L320 256V116.8c0-15.8 20.5-22 29.3-8.9L448 256V116.8c0-15.8 20.5-22 29.3-8.9L606.8 302.2c14.2 21.3-1.1 49.7-26.6 49.7H512 448 384 320 256 192 64V116.8zM32 384H608c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],"fire-burner":[640,512,[],"e4f1","M345.7 48.3L358 34.5c5.4-6.1 13.3-8.8 20.9-8.9c7.2 0 14.3 2.6 19.9 7.8c19.7 18.3 39.8 43.2 55 70.6C469 131.2 480 162.2 480 192.2C480 280.8 408.7 352 320 352c-89.6 0-160-71.3-160-159.8c0-37.3 16-73.4 36.8-104.5c20.9-31.3 47.5-59 70.9-80.2C273.4 2.3 280.7-.2 288 0c14.1 .3 23.8 11.4 32.7 21.6c0 0 0 0 0 0c2 2.3 4 4.6 6 6.7l19 19.9zM384 240.2c0-36.5-37-73-54.8-88.4c-5.4-4.7-13.1-4.7-18.5 0C293 167.1 256 203.6 256 240.2c0 35.3 28.7 64 64 64s64-28.7 64-64zM32 288c0-17.7 14.3-32 32-32l32 0c17.7 0 32 14.3 32 32s-14.3 32-32 32l0 64 448 0 0-64c-17.7 0-32-14.3-32-32s14.3-32 32-32l32 0c17.7 0 32 14.3 32 32l0 96c17.7 0 32 14.3 32 32l0 64c0 17.7-14.3 32-32 32L32 512c-17.7 0-32-14.3-32-32l0-64c0-17.7 14.3-32 32-32l0-96zM320 480a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm160-32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM192 480a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],flag:[448,512,[127988,61725],"f024","M64 32C64 14.3 49.7 0 32 0S0 14.3 0 32V64 368 480c0 17.7 14.3 32 32 32s32-14.3 32-32V352l64.3-16.1c41.1-10.3 84.6-5.5 122.5 13.4c44.2 22.1 95.5 24.8 141.7 7.4l34.7-13c12.5-4.7 20.8-16.6 20.8-30V66.1c0-23-24.2-38-44.8-27.7l-9.6 4.8c-46.3 23.2-100.8 23.2-147.1 0c-35.1-17.6-75.4-22-113.5-12.5L64 48V32z"],hanukiah:[640,512,[128334],"f6e6","M314.2 3.3C309.1 12.1 296 36.6 296 56c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7C324.6 1.2 322.4 0 320 0s-4.6 1.2-5.8 3.3zm-288 48C21.1 60.1 8 84.6 8 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7C36.6 49.2 34.4 48 32 48s-4.6 1.2-5.8 3.3zM88 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3C101.1 60.1 88 84.6 88 104zm82.2-52.7C165.1 60.1 152 84.6 152 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3zM216 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3C229.1 60.1 216 84.6 216 104zM394.2 51.3C389.1 60.1 376 84.6 376 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3zM440 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3C453.1 60.1 440 84.6 440 104zm82.2-52.7C517.1 60.1 504 84.6 504 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3zM584 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3C597.1 60.1 584 84.6 584 104zM112 160c-8.8 0-16 7.2-16 16v96 16h32V272 176c0-8.8-7.2-16-16-16zm64 0c-8.8 0-16 7.2-16 16v96 16h32V272 176c0-8.8-7.2-16-16-16zm64 0c-8.8 0-16 7.2-16 16v96 16h32V272 176c0-8.8-7.2-16-16-16zm160 0c-8.8 0-16 7.2-16 16v96 16h32V272 176c0-8.8-7.2-16-16-16zm64 0c-8.8 0-16 7.2-16 16v96 16h32V272 176c0-8.8-7.2-16-16-16zm64 0c-8.8 0-16 7.2-16 16v96 16h32V272 176c0-8.8-7.2-16-16-16zM352 144c0-17.7-14.3-32-32-32s-32 14.3-32 32V320H96c-17.7 0-32-14.3-32-32V192c0-17.7-14.3-32-32-32s-32 14.3-32 32v96c0 53 43 96 96 96H288v64H160c-17.7 0-32 14.3-32 32s14.3 32 32 32H320 480c17.7 0 32-14.3 32-32s-14.3-32-32-32H352V384H544c53 0 96-43 96-96V192c0-17.7-14.3-32-32-32s-32 14.3-32 32v96c0 17.7-14.3 32-32 32H352V144z"],feather:[512,512,[129718],"f52d","M278.5 215.6L23 471c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l57-57h68c49.7 0 97.9-14.4 139-41c11.1-7.2 5.5-23-7.8-23c-5.1 0-9.2-4.1-9.2-9.2c0-4.1 2.7-7.6 6.5-8.8l81-24.3c2.5-.8 4.8-2.1 6.7-4l22.4-22.4c10.1-10.1 2.9-27.3-11.3-27.3l-32.2 0c-5.1 0-9.2-4.1-9.2-9.2c0-4.1 2.7-7.6 6.5-8.8l112-33.6c4-1.2 7.4-3.9 9.3-7.7C506.4 207.6 512 184.1 512 160c0-41-16.3-80.3-45.3-109.3l-5.5-5.5C432.3 16.3 393 0 352 0s-80.3 16.3-109.3 45.3L139 149C91 197 64 262.1 64 330v55.3L253.6 195.8c6.2-6.2 16.4-6.2 22.6 0c5.4 5.4 6.1 13.6 2.2 19.8z"],"volume-low":[448,512,[128264,"volume-down"],"f027","M301.1 34.8C312.6 40 320 51.4 320 64V448c0 12.6-7.4 24-18.9 29.2s-25 3.1-34.4-5.3L131.8 352H64c-35.3 0-64-28.7-64-64V224c0-35.3 28.7-64 64-64h67.8L266.7 40.1c9.4-8.4 22.9-10.4 34.4-5.3zM412.6 181.5C434.1 199.1 448 225.9 448 256s-13.9 56.9-35.4 74.5c-10.3 8.4-25.4 6.8-33.8-3.5s-6.8-25.4 3.5-33.8C393.1 284.4 400 271 400 256s-6.9-28.4-17.7-37.3c-10.3-8.4-11.8-23.5-3.5-33.8s23.5-11.8 33.8-3.5z"],"comment-slash":[640,512,[],"f4b3","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L512.9 376.7C552.2 340.2 576 292.3 576 240C576 125.1 461.4 32 320 32c-67.7 0-129.3 21.4-175.1 56.3L38.8 5.1zm385.2 425L82.9 161.3C70.7 185.6 64 212.2 64 240c0 45.1 17.7 86.8 47.7 120.9c-1.9 24.5-11.4 46.3-21.4 62.9c-5.5 9.2-11.1 16.6-15.2 21.6c-2.1 2.5-3.7 4.4-4.9 5.7c-.6 .6-1 1.1-1.3 1.4l-.3 .3 0 0 0 0 0 0 0 0c-4.6 4.6-5.9 11.4-3.4 17.4c2.5 6 8.3 9.9 14.8 9.9c28.7 0 57.6-8.9 81.6-19.3c22.9-10 42.4-21.9 54.3-30.6c31.8 11.5 67 17.9 104.1 17.9c37 0 72.3-6.4 104.1-17.9z"],"cloud-sun-rain":[640,512,[127782],"f743","M294.2 1.2c5.1 2.1 8.7 6.7 9.6 12.1l10.4 62.4c-23.3 10.8-42.9 28.4-56 50.3c-14.6-9-31.8-14.1-50.2-14.1c-53 0-96 43-96 96c0 35.5 19.3 66.6 48 83.2c.8 31.8 13.2 60.7 33.1 82.7l-56 39.2c-4.5 3.2-10.3 3.8-15.4 1.6s-8.7-6.7-9.6-12.1L98.1 317.9 13.4 303.8c-5.4-.9-10-4.5-12.1-9.6s-1.5-10.9 1.6-15.4L52.5 208 2.9 137.2c-3.2-4.5-3.8-10.3-1.6-15.4s6.7-8.7 12.1-9.6L98.1 98.1l14.1-84.7c.9-5.4 4.5-10 9.6-12.1s10.9-1.5 15.4 1.6L208 52.5 278.8 2.9c4.5-3.2 10.3-3.8 15.4-1.6zM208 144c13.8 0 26.7 4.4 37.1 11.9c-1.2 4.1-2.2 8.3-3 12.6c-37.9 14.6-67.2 46.6-77.8 86.4C151.8 243.1 144 226.5 144 208c0-35.3 28.7-64 64-64zm69.4 276c11 7.4 14 22.3 6.7 33.3l-32 48c-7.4 11-22.3 14-33.3 6.7s-14-22.3-6.7-33.3l32-48c7.4-11 22.3-14 33.3-6.7zm96 0c11 7.4 14 22.3 6.7 33.3l-32 48c-7.4 11-22.3 14-33.3 6.7s-14-22.3-6.7-33.3l32-48c7.4-11 22.3-14 33.3-6.7zm96 0c11 7.4 14 22.3 6.7 33.3l-32 48c-7.4 11-22.3 14-33.3 6.7s-14-22.3-6.7-33.3l32-48c7.4-11 22.3-14 33.3-6.7zm96 0c11 7.4 14 22.3 6.7 33.3l-32 48c-7.4 11-22.3 14-33.3 6.7s-14-22.3-6.7-33.3l32-48c7.4-11 22.3-14 33.3-6.7zm74.5-116.1c0 44.2-35.8 80-80 80H288c-53 0-96-43-96-96c0-47.6 34.6-87 80-94.6l0-1.3c0-53 43-96 96-96c34.9 0 65.4 18.6 82.2 46.4c13-9.1 28.8-14.4 45.8-14.4c44.2 0 80 35.8 80 80c0 5.9-.6 11.7-1.9 17.2c37.4 6.7 65.8 39.4 65.8 78.7z"],compress:[448,512,[],"f066","M160 64c0-17.7-14.3-32-32-32s-32 14.3-32 32v64H32c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c17.7 0 32-14.3 32-32V64zM32 320c-17.7 0-32 14.3-32 32s14.3 32 32 32H96v64c0 17.7 14.3 32 32 32s32-14.3 32-32V352c0-17.7-14.3-32-32-32H32zM352 64c0-17.7-14.3-32-32-32s-32 14.3-32 32v96c0 17.7 14.3 32 32 32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H352V64zM320 320c-17.7 0-32 14.3-32 32v96c0 17.7 14.3 32 32 32s32-14.3 32-32V384h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H320z"],"wheat-awn":[512,512,["wheat-alt"],"e2cd","M505 41c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L383 95c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l88-88zM305.5 27.3c-6.2-6.2-16.4-6.2-22.6 0L271.5 38.6c-37.5 37.5-37.5 98.3 0 135.8l10.4 10.4-30.5 30.5c-3.4-27.3-15.5-53.8-36.5-74.8l-11.3-11.3c-6.2-6.2-16.4-6.2-22.6 0l-11.3 11.3c-37.5 37.5-37.5 98.3 0 135.8l10.4 10.4-30.5 30.5c-3.4-27.3-15.5-53.8-36.5-74.8L101.8 231c-6.2-6.2-16.4-6.2-22.6 0L67.9 242.3c-37.5 37.5-37.5 98.3 0 135.8l10.4 10.4L9.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l68.9-68.9 12.2 12.2c37.5 37.5 98.3 37.5 135.8 0l11.3-11.3c6.2-6.2 6.2-16.4 0-22.6l-11.3-11.3c-21.8-21.8-49.6-34.1-78.1-36.9l31.9-31.9 12.2 12.2c37.5 37.5 98.3 37.5 135.8 0l11.3-11.3c6.2-6.2 6.2-16.4 0-22.6l-11.3-11.3c-21.8-21.8-49.6-34.1-78.1-36.9l31.9-31.9 12.2 12.2c37.5 37.5 98.3 37.5 135.8 0L486.5 231c6.2-6.2 6.2-16.4 0-22.6L475.2 197c-5.2-5.2-10.6-9.8-16.4-13.9L505 137c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-59.4 59.4c-20.6-4.4-42-3.7-62.3 2.1c6.1-21.3 6.6-43.8 1.4-65.3L409 41c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L329.1 52.9c-3.7-5-7.8-9.8-12.4-14.3L305.5 27.3z"],ankh:[320,512,[9765],"f644","M96 128c0-35.3 28.7-64 64-64s64 28.7 64 64c0 41.6-20.7 76.6-46.6 104.1c-5.9 6.2-11.8 11.8-17.4 16.7c-5.6-4.9-11.5-10.5-17.4-16.7C116.7 204.6 96 169.6 96 128zM160 0C89.3 0 32 57.3 32 128c0 52.4 21.5 95.5 46.8 128H32c-17.7 0-32 14.3-32 32s14.3 32 32 32h96V480c0 17.7 14.3 32 32 32s32-14.3 32-32V320h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H241.2c25.4-32.5 46.8-75.6 46.8-128C288 57.3 230.7 0 160 0z"],"hands-holding-child":[640,512,[],"e4fa","M320 0a40 40 0 1 1 0 80 40 40 0 1 1 0-80zm44.7 164.3L375.8 253c1.6 13.2-7.7 25.1-20.8 26.8s-25.1-7.7-26.8-20.8l-4.4-35h-7.6l-4.4 35c-1.6 13.2-13.6 22.5-26.8 20.8s-22.5-13.6-20.8-26.8l11.1-88.8L255.5 181c-10.1 8.6-25.3 7.3-33.8-2.8s-7.3-25.3 2.8-33.8l27.9-23.6C271.3 104.8 295.3 96 320 96s48.7 8.8 67.6 24.7l27.9 23.6c10.1 8.6 11.4 23.7 2.8 33.8s-23.7 11.4-33.8 2.8l-19.8-16.7zM40 64c22.1 0 40 17.9 40 40v40 80 40.2c0 17 6.7 33.3 18.7 45.3l51.1 51.1c8.3 8.3 21.3 9.6 31 3.1c12.9-8.6 14.7-26.9 3.7-37.8l-15.2-15.2-32-32c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0l32 32 15.2 15.2 0 0 25.3 25.3c21 21 32.8 49.5 32.8 79.2V464c0 26.5-21.5 48-48 48H173.3c-17 0-33.3-6.7-45.3-18.7L28.1 393.4C10.1 375.4 0 351 0 325.5V224 160 104C0 81.9 17.9 64 40 64zm560 0c22.1 0 40 17.9 40 40v56 64V325.5c0 25.5-10.1 49.9-28.1 67.9L512 493.3c-12 12-28.3 18.7-45.3 18.7H400c-26.5 0-48-21.5-48-48V385.1c0-29.7 11.8-58.2 32.8-79.2l25.3-25.3 0 0 15.2-15.2 32-32c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3l-32 32-15.2 15.2c-11 11-9.2 29.2 3.7 37.8c9.7 6.5 22.7 5.2 31-3.1l51.1-51.1c12-12 18.7-28.3 18.7-45.3V224 144 104c0-22.1 17.9-40 40-40z"],asterisk:[384,512,[10033,61545],"2a","M192 32c17.7 0 32 14.3 32 32V199.5l111.5-66.9c15.2-9.1 34.8-4.2 43.9 11s4.2 34.8-11 43.9L254.2 256l114.3 68.6c15.2 9.1 20.1 28.7 11 43.9s-28.7 20.1-43.9 11L224 312.5V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V312.5L48.5 379.4c-15.2 9.1-34.8 4.2-43.9-11s-4.2-34.8 11-43.9L129.8 256 15.5 187.4c-15.2-9.1-20.1-28.7-11-43.9s28.7-20.1 43.9-11L160 199.5V64c0-17.7 14.3-32 32-32z"],"square-check":[448,512,[9745,9989,61510,"check-square"],"f14a","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM337 209L209 337c-9.4 9.4-24.6 9.4-33.9 0l-64-64c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47L303 175c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z"],"peseta-sign":[384,512,[],"e221","M64 32C46.3 32 32 46.3 32 64v96c-17.7 0-32 14.3-32 32s14.3 32 32 32l0 96V448c0 17.7 14.3 32 32 32s32-14.3 32-32V352h96c77.4 0 142-55 156.8-128H352c17.7 0 32-14.3 32-32s-14.3-32-32-32h-3.2C334 87 269.4 32 192 32H64zM282.5 160H96V96h96c41.8 0 77.4 26.7 90.5 64zM96 224H282.5c-13.2 37.3-48.7 64-90.5 64H96V224z"],heading:[448,512,["header"],"f1dc","M0 64C0 46.3 14.3 32 32 32H80h48c17.7 0 32 14.3 32 32s-14.3 32-32 32H112V208H336V96H320c-17.7 0-32-14.3-32-32s14.3-32 32-32h48 48c17.7 0 32 14.3 32 32s-14.3 32-32 32H400V240 416h16c17.7 0 32 14.3 32 32s-14.3 32-32 32H368 320c-17.7 0-32-14.3-32-32s14.3-32 32-32h16V272H112V416h16c17.7 0 32 14.3 32 32s-14.3 32-32 32H80 32c-17.7 0-32-14.3-32-32s14.3-32 32-32H48V240 96H32C14.3 96 0 81.7 0 64z"],ghost:[384,512,[128123],"f6e2","M40.1 467.1l-11.2 9c-3.2 2.5-7.1 3.9-11.1 3.9C8 480 0 472 0 462.2V192C0 86 86 0 192 0S384 86 384 192V462.2c0 9.8-8 17.8-17.8 17.8c-4 0-7.9-1.4-11.1-3.9l-11.2-9c-13.4-10.7-32.8-9-44.1 3.9L269.3 506c-3.3 3.8-8.2 6-13.3 6s-9.9-2.2-13.3-6l-26.6-30.5c-12.7-14.6-35.4-14.6-48.2 0L141.3 506c-3.3 3.8-8.2 6-13.3 6s-9.9-2.2-13.3-6L84.2 471c-11.3-12.9-30.7-14.6-44.1-3.9zM160 192a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm96 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],list:[512,512,["list-squares"],"f03a","M40 48C26.7 48 16 58.7 16 72v48c0 13.3 10.7 24 24 24H88c13.3 0 24-10.7 24-24V72c0-13.3-10.7-24-24-24H40zM192 64c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H192zm0 160c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H192zm0 160c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H192zM16 232v48c0 13.3 10.7 24 24 24H88c13.3 0 24-10.7 24-24V232c0-13.3-10.7-24-24-24H40c-13.3 0-24 10.7-24 24zM40 368c-13.3 0-24 10.7-24 24v48c0 13.3 10.7 24 24 24H88c13.3 0 24-10.7 24-24V392c0-13.3-10.7-24-24-24H40z"],"square-phone-flip":[448,512,["phone-square-alt"],"f87b","M384 32c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H384zm-90.7 96.7c-9.7-2.6-19.9 2.3-23.7 11.6l-20 48c-3.4 8.2-1 17.6 5.8 23.2L280 231.7c-16.6 35.2-45.1 63.7-80.3 80.3l-20.2-24.7c-5.6-6.8-15-9.2-23.2-5.8l-48 20c-9.3 3.9-14.2 14-11.6 23.7l12 44C111.1 378 119 384 128 384c123.7 0 224-100.3 224-224c0-9-6-16.9-14.7-19.3l-44-12z"],"cart-plus":[576,512,[],"f217","M0 24C0 10.7 10.7 0 24 0H69.5c22 0 41.5 12.8 50.6 32h411c26.3 0 45.5 25 38.6 50.4l-41 152.3c-8.5 31.4-37 53.3-69.5 53.3H170.7l5.4 28.5c2.2 11.3 12.1 19.5 23.6 19.5H488c13.3 0 24 10.7 24 24s-10.7 24-24 24H199.7c-34.6 0-64.3-24.6-70.7-58.5L77.4 54.5c-.7-3.8-4-6.5-7.9-6.5H24C10.7 48 0 37.3 0 24zM128 464a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm336-48a48 48 0 1 1 0 96 48 48 0 1 1 0-96zM252 160c0 11 9 20 20 20h44v44c0 11 9 20 20 20s20-9 20-20V180h44c11 0 20-9 20-20s-9-20-20-20H356V96c0-11-9-20-20-20s-20 9-20 20v44H272c-11 0-20 9-20 20z"],gamepad:[640,512,[],"f11b","M192 64C86 64 0 150 0 256S86 448 192 448H448c106 0 192-86 192-192s-86-192-192-192H192zM496 168a40 40 0 1 1 0 80 40 40 0 1 1 0-80zM392 304a40 40 0 1 1 80 0 40 40 0 1 1 -80 0zM168 200c0-13.3 10.7-24 24-24s24 10.7 24 24v32h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H216v32c0 13.3-10.7 24-24 24s-24-10.7-24-24V280H136c-13.3 0-24-10.7-24-24s10.7-24 24-24h32V200z"],"circle-dot":[512,512,[128280,"dot-circle"],"f192","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm0-352a96 96 0 1 1 0 192 96 96 0 1 1 0-192z"],"face-dizzy":[512,512,["dizzy"],"f567","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm0-224a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM100.7 132.7c6.2-6.2 16.4-6.2 22.6 0L160 169.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6L182.6 192l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L160 214.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L137.4 192l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6zm192 0c6.2-6.2 16.4-6.2 22.6 0L352 169.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6L374.6 192l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L352 214.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L329.4 192l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6z"],egg:[384,512,[129370],"f7fb","M192 496C86 496 0 394 0 288C0 176 64 16 192 16s192 160 192 272c0 106-86 208-192 208zM154.8 134c6.5-6 7-16.1 1-22.6s-16.1-7-22.6-1c-23.9 21.8-41.1 52.7-52.3 84.2C69.7 226.1 64 259.7 64 288c0 8.8 7.2 16 16 16s16-7.2 16-16c0-24.5 5-54.4 15.1-82.8c10.1-28.5 25-54.1 43.7-71.2z"],"house-medical-circle-xmark":[640,512,[],"e513","M320 368c0 59.5 29.5 112.1 74.8 144H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L522.1 193.9c-8.5-1.3-17.3-1.9-26.1-1.9c-54.7 0-103.5 24.9-135.8 64H320V208c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16v48H208c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16zM496 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm22.6 144l36.7-36.7c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0L496 345.4l-36.7-36.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6L473.4 368l-36.7 36.7c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L496 390.6l36.7 36.7c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6L518.6 368z"],campground:[576,512,[9978],"f6bb","M377 52c11-13.8 8.8-33.9-5-45s-33.9-8.8-45 5L288 60.8 249 12c-11-13.8-31.2-16-45-5s-16 31.2-5 45l48 60L12.3 405.4C4.3 415.4 0 427.7 0 440.4V464c0 26.5 21.5 48 48 48H288 528c26.5 0 48-21.5 48-48V440.4c0-12.7-4.3-25.1-12.3-35L329 112l48-60zM288 448H168.5L288 291.7 407.5 448H288z"],"folder-plus":[512,512,[],"f65e","M512 416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H192c20.1 0 39.1 9.5 51.2 25.6l19.2 25.6c6 8.1 15.5 12.8 25.6 12.8H448c35.3 0 64 28.7 64 64V416zM232 376c0 13.3 10.7 24 24 24s24-10.7 24-24V312h64c13.3 0 24-10.7 24-24s-10.7-24-24-24H280V200c0-13.3-10.7-24-24-24s-24 10.7-24 24v64H168c-13.3 0-24 10.7-24 24s10.7 24 24 24h64v64z"],futbol:[512,512,[9917,"futbol-ball","soccer-ball"],"f1e3","M417.3 360.1l-71.6-4.8c-5.2-.3-10.3 1.1-14.5 4.2s-7.2 7.4-8.4 12.5l-17.6 69.6C289.5 445.8 273 448 256 448s-33.5-2.2-49.2-6.4L189.2 372c-1.3-5-4.3-9.4-8.4-12.5s-9.3-4.5-14.5-4.2l-71.6 4.8c-17.6-27.2-28.5-59.2-30.4-93.6L125 228.3c4.4-2.8 7.6-7 9.2-11.9s1.4-10.2-.5-15l-26.7-66.6C128 109.2 155.3 89 186.7 76.9l55.2 46c4 3.3 9 5.1 14.1 5.1s10.2-1.8 14.1-5.1l55.2-46c31.3 12.1 58.7 32.3 79.6 57.9l-26.7 66.6c-1.9 4.8-2.1 10.1-.5 15s4.9 9.1 9.2 11.9l60.7 38.2c-1.9 34.4-12.8 66.4-30.4 93.6zM256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm14.1-325.7c-8.4-6.1-19.8-6.1-28.2 0L194 221c-8.4 6.1-11.9 16.9-8.7 26.8l18.3 56.3c3.2 9.9 12.4 16.6 22.8 16.6h59.2c10.4 0 19.6-6.7 22.8-16.6l18.3-56.3c3.2-9.9-.3-20.7-8.7-26.8l-47.9-34.8z"],paintbrush:[576,512,[128396,"paint-brush"],"f1fc","M339.3 367.1c27.3-3.9 51.9-19.4 67.2-42.9L568.2 74.1c12.6-19.5 9.4-45.3-7.6-61.2S517.7-4.4 499.1 9.6L262.4 187.2c-24 18-38.2 46.1-38.4 76.1L339.3 367.1zm-19.6 25.4l-116-104.4C143.9 290.3 96 339.6 96 400c0 3.9 .2 7.8 .6 11.6C98.4 429.1 86.4 448 68.8 448H64c-17.7 0-32 14.3-32 32s14.3 32 32 32H208c61.9 0 112-50.1 112-112c0-2.5-.1-5-.2-7.5z"],lock:[448,512,[128274],"f023","M144 144v48H304V144c0-44.2-35.8-80-80-80s-80 35.8-80 80zM80 192V144C80 64.5 144.5 0 224 0s144 64.5 144 144v48h16c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V256c0-35.3 28.7-64 64-64H80z"],"gas-pump":[512,512,[9981],"f52f","M32 64C32 28.7 60.7 0 96 0H256c35.3 0 64 28.7 64 64V256h8c48.6 0 88 39.4 88 88v32c0 13.3 10.7 24 24 24s24-10.7 24-24V222c-27.6-7.1-48-32.2-48-62V96L384 64c-8.8-8.8-8.8-23.2 0-32s23.2-8.8 32 0l77.3 77.3c12 12 18.7 28.3 18.7 45.3V168v24 32V376c0 39.8-32.2 72-72 72s-72-32.2-72-72V344c0-22.1-17.9-40-40-40h-8V448c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32V64zM96 80v96c0 8.8 7.2 16 16 16H240c8.8 0 16-7.2 16-16V80c0-8.8-7.2-16-16-16H112c-8.8 0-16 7.2-16 16z"],"hot-tub-person":[512,512,["hot-tub"],"f593","M272 24c0-13.3-10.7-24-24-24s-24 10.7-24 24v5.2c0 34 14.4 66.4 39.7 89.2l16.4 14.8c15.2 13.7 23.8 33.1 23.8 53.5V200c0 13.3 10.7 24 24 24s24-10.7 24-24V186.8c0-34-14.4-66.4-39.7-89.2L295.8 82.8C280.7 69.1 272 49.7 272 29.2V24zM0 320v16V448c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V320c0-35.3-28.7-64-64-64H277.3c-13.8 0-27.3-4.5-38.4-12.8l-85.3-64C137 166.7 116.8 160 96 160c-53 0-96 43-96 96v64zm128 16v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V336c0-8.8 7.2-16 16-16s16 7.2 16 16zm80-16c8.8 0 16 7.2 16 16v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V336c0-8.8 7.2-16 16-16zm112 16v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V336c0-8.8 7.2-16 16-16s16 7.2 16 16zm80-16c8.8 0 16 7.2 16 16v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V336c0-8.8 7.2-16 16-16zM360 0c-13.3 0-24 10.7-24 24v5.2c0 34 14.4 66.4 39.7 89.2l16.4 14.8c15.2 13.7 23.8 33.1 23.8 53.5V200c0 13.3 10.7 24 24 24s24-10.7 24-24V186.8c0-34-14.4-66.4-39.7-89.2L407.8 82.8C392.7 69.1 384 49.7 384 29.2V24c0-13.3-10.7-24-24-24zM64 128A64 64 0 1 0 64 0a64 64 0 1 0 0 128z"],"map-location":[576,512,["map-marked"],"f59f","M302.8 312C334.9 271.9 408 174.6 408 120C408 53.7 354.3 0 288 0S168 53.7 168 120c0 54.6 73.1 151.9 105.2 192c7.7 9.6 22 9.6 29.6 0zM416 503l144.9-58c9.1-3.6 15.1-12.5 15.1-22.3V152c0-17-17.1-28.6-32.9-22.3l-116 46.4c-.5 1.2-1 2.5-1.5 3.7c-2.9 6.8-6.1 13.7-9.6 20.6V503zM15.1 187.3C6 191 0 199.8 0 209.6V480.4c0 17 17.1 28.6 32.9 22.3L160 451.8V200.4c-3.5-6.9-6.7-13.8-9.6-20.6c-5.6-13.2-10.4-27.4-12.8-41.5l-122.6 49zM384 255c-20.5 31.3-42.3 59.6-56.2 77c-20.5 25.6-59.1 25.6-79.6 0c-13.9-17.4-35.7-45.7-56.2-77V449.4l192 54.9V255z"],"house-flood-water":[576,512,[],"e50e","M306.8 6.1C295.6-2 280.4-2 269.2 6.1l-176 128c-11.2 8.2-15.9 22.6-11.6 35.8S98.1 192 112 192h16v73c1.7 1 3.3 2 4.9 3.1c18 12.4 40.1 20.3 59.2 20.3c21.1 0 42-8.5 59.2-20.3c22.1-15.5 51.6-15.5 73.7 0c18.4 12.7 39.6 20.3 59.2 20.3c19 0 41.2-7.9 59.2-20.3c1.5-1 3-2 4.5-2.9l-.3-73.2H464c13.9 0 26.1-8.9 30.4-22.1s-.4-27.6-11.6-35.8l-176-128zM269.5 309.9C247 325.4 219.5 336 192 336c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C126.5 389.7 159 400 192 400c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.5-27.3-10.1-39.2-1.7l0 0C439.4 325.2 410.9 336 384 336c-27.5 0-55-10.6-77.5-26.1c-11.1-7.9-25.9-7.9-37 0zM384 448c-27.5 0-55-10.6-77.5-26.1c-11.1-7.9-25.9-7.9-37 0C247 437.4 219.5 448 192 448c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C126.5 501.7 159 512 192 512c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.4-27.3-10.1-39.2-1.7l0 0C439.4 437.2 410.9 448 384 448z"],tree:[448,512,[127794],"f1bb","M210.6 5.9L62 169.4c-3.9 4.2-6 9.8-6 15.5C56 197.7 66.3 208 79.1 208H104L30.6 281.4c-4.2 4.2-6.6 10-6.6 16C24 309.9 34.1 320 46.6 320H80L5.4 409.5C1.9 413.7 0 419 0 424.5c0 13 10.5 23.5 23.5 23.5H192v32c0 17.7 14.3 32 32 32s32-14.3 32-32V448H424.5c13 0 23.5-10.5 23.5-23.5c0-5.5-1.9-10.8-5.4-15L368 320h33.4c12.5 0 22.6-10.1 22.6-22.6c0-6-2.4-11.8-6.6-16L344 208h24.9c12.7 0 23.1-10.3 23.1-23.1c0-5.7-2.1-11.3-6-15.5L237.4 5.9C234 2.1 229.1 0 224 0s-10 2.1-13.4 5.9z"],"bridge-lock":[640,512,[],"e4cc","M32 64c0-17.7 14.3-32 32-32H576c17.7 0 32 14.3 32 32s-14.3 32-32 32H536v64h-8c-61.9 0-112 50.1-112 112v24.6c-9.9 5.8-18.2 14.1-23.8 24.1c-17.6-20-43.4-32.7-72.2-32.7c-53 0-96 43-96 96v64c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32V384c0-53-43-96-96-96V160h72V96H64C46.3 96 32 81.7 32 64zM408 96v64h80V96H408zm-48 64V96H280v64h80zM152 96v64h80V96H152zM528 240c-17.7 0-32 14.3-32 32v48h64V272c0-17.7-14.3-32-32-32zm-80 32c0-44.2 35.8-80 80-80s80 35.8 80 80v48c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H448c-17.7 0-32-14.3-32-32V352c0-17.7 14.3-32 32-32V272z"],"sack-dollar":[512,512,[128176],"f81d","M320 96H192L144.6 24.9C137.5 14.2 145.1 0 157.9 0H354.1c12.8 0 20.4 14.2 13.3 24.9L320 96zM192 128H320c3.8 2.5 8.1 5.3 13 8.4C389.7 172.7 512 250.9 512 416c0 53-43 96-96 96H96c-53 0-96-43-96-96C0 250.9 122.3 172.7 179 136.4l0 0 0 0c4.8-3.1 9.2-5.9 13-8.4zm84 88c0-11-9-20-20-20s-20 9-20 20v14c-7.6 1.7-15.2 4.4-22.2 8.5c-13.9 8.3-25.9 22.8-25.8 43.9c.1 20.3 12 33.1 24.7 40.7c11 6.6 24.7 10.8 35.6 14l1.7 .5c12.6 3.8 21.8 6.8 28 10.7c5.1 3.2 5.8 5.4 5.9 8.2c.1 5-1.8 8-5.9 10.5c-5 3.1-12.9 5-21.4 4.7c-11.1-.4-21.5-3.9-35.1-8.5c-2.3-.8-4.7-1.6-7.2-2.4c-10.5-3.5-21.8 2.2-25.3 12.6s2.2 21.8 12.6 25.3c1.9 .6 4 1.3 6.1 2.1l0 0 0 0c8.3 2.9 17.9 6.2 28.2 8.4V424c0 11 9 20 20 20s20-9 20-20V410.2c8-1.7 16-4.5 23.2-9c14.3-8.9 25.1-24.1 24.8-45c-.3-20.3-11.7-33.4-24.6-41.6c-11.5-7.2-25.9-11.6-37.1-15l0 0-.7-.2c-12.8-3.9-21.9-6.7-28.3-10.5c-5.2-3.1-5.3-4.9-5.3-6.7c0-3.7 1.4-6.5 6.2-9.3c5.4-3.2 13.6-5.1 21.5-5c9.6 .1 20.2 2.2 31.2 5.2c10.7 2.8 21.6-3.5 24.5-14.2s-3.5-21.6-14.2-24.5c-6.5-1.7-13.7-3.4-21.1-4.7V216z"],"pen-to-square":[512,512,["edit"],"f044","M471.6 21.7c-21.9-21.9-57.3-21.9-79.2 0L362.3 51.7l97.9 97.9 30.1-30.1c21.9-21.9 21.9-57.3 0-79.2L471.6 21.7zm-299.2 220c-6.1 6.1-10.8 13.6-13.5 21.9l-29.6 88.8c-2.9 8.6-.6 18.1 5.8 24.6s15.9 8.7 24.6 5.8l88.8-29.6c8.2-2.7 15.7-7.4 21.9-13.5L437.7 172.3 339.7 74.3 172.4 241.7zM96 64C43 64 0 107 0 160V416c0 53 43 96 96 96H352c53 0 96-43 96-96V320c0-17.7-14.3-32-32-32s-32 14.3-32 32v96c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V160c0-17.7 14.3-32 32-32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H96z"],"car-side":[640,512,[128663],"f5e4","M171.3 96H224v96H111.3l30.4-75.9C146.5 104 158.2 96 171.3 96zM272 192V96h81.2c9.7 0 18.9 4.4 25 12l67.2 84H272zm256.2 1L428.2 68c-18.2-22.8-45.8-36-75-36H171.3c-39.3 0-74.6 23.9-89.1 60.3L40.6 196.4C16.8 205.8 0 228.9 0 256V368c0 17.7 14.3 32 32 32H65.3c7.6 45.4 47.1 80 94.7 80s87.1-34.6 94.7-80H385.3c7.6 45.4 47.1 80 94.7 80s87.1-34.6 94.7-80H608c17.7 0 32-14.3 32-32V320c0-65.2-48.8-119-111.8-127zM434.7 368a48 48 0 1 1 90.5 32 48 48 0 1 1 -90.5-32zM160 336a48 48 0 1 1 0 96 48 48 0 1 1 0-96z"],"share-nodes":[448,512,["share-alt"],"f1e0","M352 224c53 0 96-43 96-96s-43-96-96-96s-96 43-96 96c0 4 .2 8 .7 11.9l-94.1 47C145.4 170.2 121.9 160 96 160c-53 0-96 43-96 96s43 96 96 96c25.9 0 49.4-10.2 66.6-26.9l94.1 47c-.5 3.9-.7 7.8-.7 11.9c0 53 43 96 96 96s96-43 96-96s-43-96-96-96c-25.9 0-49.4 10.2-66.6 26.9l-94.1-47c.5-3.9 .7-7.8 .7-11.9s-.2-8-.7-11.9l94.1-47C302.6 213.8 326.1 224 352 224z"],"heart-circle-minus":[576,512,[],"e4ff","M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9l2.6-2.4C267.2 438.6 256 404.6 256 368c0-97.2 78.8-176 176-176c28.3 0 55 6.7 78.7 18.5c.9-6.5 1.3-13 1.3-19.6v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5zM576 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-64 0c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16s7.2-16 16-16H496c8.8 0 16 7.2 16 16z"],"hourglass-half":[384,512,["hourglass-2"],"f252","M32 0C14.3 0 0 14.3 0 32S14.3 64 32 64V75c0 42.4 16.9 83.1 46.9 113.1L146.7 256 78.9 323.9C48.9 353.9 32 394.6 32 437v11c-17.7 0-32 14.3-32 32s14.3 32 32 32H64 320h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V437c0-42.4-16.9-83.1-46.9-113.1L237.3 256l67.9-67.9c30-30 46.9-70.7 46.9-113.1V64c17.7 0 32-14.3 32-32s-14.3-32-32-32H320 64 32zM96 75V64H288V75c0 19-5.6 37.4-16 53H112c-10.3-15.6-16-34-16-53zm16 309c3.5-5.3 7.6-10.3 12.1-14.9L192 301.3l67.9 67.9c4.6 4.6 8.6 9.6 12.1 14.9H112z"],microscope:[512,512,[128300],"f610","M160 32c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32c17.7 0 32 14.3 32 32V288c0 17.7-14.3 32-32 32c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32c-17.7 0-32-14.3-32-32V64c0-17.7 14.3-32 32-32zM32 448H320c70.7 0 128-57.3 128-128s-57.3-128-128-128V128c106 0 192 86 192 192c0 49.2-18.5 94-48.9 128H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H320 32c-17.7 0-32-14.3-32-32s14.3-32 32-32zm80-64H304c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],sink:[512,512,[],"e06d","M288 96c0-17.7 14.3-32 32-32s32 14.3 32 32s14.3 32 32 32s32-14.3 32-32c0-53-43-96-96-96s-96 43-96 96V288H160V264c0-30.9-25.1-56-56-56H56c-13.3 0-24 10.7-24 24s10.7 24 24 24h48c4.4 0 8 3.6 8 8v24H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H256 480c17.7 0 32-14.3 32-32s-14.3-32-32-32H400V264c0-4.4 3.6-8 8-8h56c13.3 0 24-10.7 24-24s-10.7-24-24-24H408c-30.9 0-56 25.1-56 56v24H288V96zM480 416V384H32v32c0 53 43 96 96 96H384c53 0 96-43 96-96z"],"bag-shopping":[448,512,["shopping-bag"],"f290","M160 112c0-35.3 28.7-64 64-64s64 28.7 64 64v48H160V112zm-48 48H48c-26.5 0-48 21.5-48 48V416c0 53 43 96 96 96H352c53 0 96-43 96-96V208c0-26.5-21.5-48-48-48H336V112C336 50.1 285.9 0 224 0S112 50.1 112 112v48zm24 48a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm152 24a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z"],"arrow-down-z-a":[576,512,["sort-alpha-desc","sort-alpha-down-alt"],"f881","M183.6 469.6C177.5 476.2 169 480 160 480s-17.5-3.8-23.6-10.4l-88-96c-11.9-13-11.1-33.3 2-45.2s33.3-11.1 45.2 2L128 365.7V64c0-17.7 14.3-32 32-32s32 14.3 32 32V365.7l32.4-35.4c11.9-13 32.2-13.9 45.2-2s13.9 32.2 2 45.2l-88 96zM320 64c0-17.7 14.3-32 32-32H480c12.9 0 24.6 7.8 29.6 19.8s2.2 25.7-6.9 34.9L429.3 160H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H352c-12.9 0-24.6-7.8-29.6-19.8s-2.2-25.7 6.9-34.9L402.7 96H352c-17.7 0-32-14.3-32-32zm96 192c12.1 0 23.2 6.8 28.6 17.7l64 128 16 32c7.9 15.8 1.5 35-14.3 42.9s-35 1.5-42.9-14.3L460.2 448H371.8l-7.2 14.3c-7.9 15.8-27.1 22.2-42.9 14.3s-22.2-27.1-14.3-42.9l16-32 64-128c5.4-10.8 16.5-17.7 28.6-17.7zM395.8 400h40.4L416 359.6 395.8 400z"],mitten:[448,512,[],"f7b5","M352 384H64L5.4 178.9C1.8 166.4 0 153.4 0 140.3C0 62.8 62.8 0 140.3 0h3.4c66 0 123.5 44.9 139.5 108.9l31.4 125.8 17.6-20.1C344.8 200.2 362.9 192 382 192h2.8c34.9 0 63.3 28.3 63.3 63.3c0 15.9-6 31.2-16.8 42.9L352 384zM32 448c0-17.7 14.3-32 32-32H352c17.7 0 32 14.3 32 32v32c0 17.7-14.3 32-32 32H64c-17.7 0-32-14.3-32-32V448z"],"person-rays":[512,512,[],"e54d","M208 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9l-28.6 47.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l58.3 97c9.1 15.1 4.2 34.8-10.9 43.9s-34.8 4.2-43.9-10.9L328 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H248zM7 7C16.4-2.3 31.6-2.3 41 7l80 80c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L7 41C-2.3 31.6-2.3 16.4 7 7zM471 7c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-80 80c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9L471 7zM7 505c-9.4-9.4-9.4-24.6 0-33.9l80-80c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9L41 505c-9.4 9.4-24.6 9.4-33.9 0zm464 0l-80-80c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l80 80c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0z"],users:[640,512,[],"f0c0","M144 0a80 80 0 1 1 0 160A80 80 0 1 1 144 0zM512 0a80 80 0 1 1 0 160A80 80 0 1 1 512 0zM0 298.7C0 239.8 47.8 192 106.7 192h42.7c15.9 0 31 3.5 44.6 9.7c-1.3 7.2-1.9 14.7-1.9 22.3c0 38.2 16.8 72.5 43.3 96c-.2 0-.4 0-.7 0H21.3C9.6 320 0 310.4 0 298.7zM405.3 320c-.2 0-.4 0-.7 0c26.6-23.5 43.3-57.8 43.3-96c0-7.6-.7-15-1.9-22.3c13.6-6.3 28.7-9.7 44.6-9.7h42.7C592.2 192 640 239.8 640 298.7c0 11.8-9.6 21.3-21.3 21.3H405.3zM224 224a96 96 0 1 1 192 0 96 96 0 1 1 -192 0zM128 485.3C128 411.7 187.7 352 261.3 352H378.7C452.3 352 512 411.7 512 485.3c0 14.7-11.9 26.7-26.7 26.7H154.7c-14.7 0-26.7-11.9-26.7-26.7z"],"eye-slash":[640,512,[],"f070","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L525.6 386.7c39.6-40.6 66.4-86.1 79.9-118.4c3.3-7.9 3.3-16.7 0-24.6c-14.9-35.7-46.2-87.7-93-131.1C465.5 68.8 400.8 32 320 32c-68.2 0-125 26.3-169.3 60.8L38.8 5.1zM223.1 149.5C248.6 126.2 282.7 112 320 112c79.5 0 144 64.5 144 144c0 24.9-6.3 48.3-17.4 68.7L408 294.5c8.4-19.3 10.6-41.4 4.8-63.3c-11.1-41.5-47.8-69.4-88.6-71.1c-5.8-.2-9.2 6.1-7.4 11.7c2.1 6.4 3.3 13.2 3.3 20.3c0 10.2-2.4 19.8-6.6 28.3l-90.3-70.8zM373 389.9c-16.4 6.5-34.3 10.1-53 10.1c-79.5 0-144-64.5-144-144c0-6.9 .5-13.6 1.4-20.2L83.1 161.5C60.3 191.2 44 220.8 34.5 243.7c-3.3 7.9-3.3 16.7 0 24.6c14.9 35.7 46.2 87.7 93 131.1C174.5 443.2 239.2 480 320 480c47.8 0 89.9-12.9 126.2-32.5L373 389.9z"],"flask-vial":[640,512,[],"e4f3","M175 389.4c-9.8 16-15 34.3-15 53.1c-10 3.5-20.8 5.5-32 5.5c-53 0-96-43-96-96V64C14.3 64 0 49.7 0 32S14.3 0 32 0H96h64 64c17.7 0 32 14.3 32 32s-14.3 32-32 32V309.9l-49 79.6zM96 64v96h64V64H96zM352 0H480h32c17.7 0 32 14.3 32 32s-14.3 32-32 32V214.9L629.7 406.2c6.7 10.9 10.3 23.5 10.3 36.4c0 38.3-31.1 69.4-69.4 69.4H261.4c-38.3 0-69.4-31.1-69.4-69.4c0-12.8 3.6-25.4 10.3-36.4L320 214.9V64c-17.7 0-32-14.3-32-32s14.3-32 32-32h32zm32 64V224c0 5.9-1.6 11.7-4.7 16.8L330.5 320h171l-48.8-79.2c-3.1-5-4.7-10.8-4.7-16.8V64H384z"],hand:[512,512,[129306,9995,"hand-paper"],"f256","M288 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V336c0 1.5 0 3.1 .1 4.6L67.6 283c-16-15.2-41.3-14.6-56.6 1.4s-14.6 41.3 1.4 56.6L124.8 448c43.1 41.1 100.4 64 160 64H304c97.2 0 176-78.8 176-176V128c0-17.7-14.3-32-32-32s-32 14.3-32 32V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V32z"],om:[512,512,[128329],"f679","M379.3 4.7c-6.2-6.2-16.4-6.2-22.6 0l-16 16c-6.2 6.2-6.2 16.4 0 22.6l16 16c6.2 6.2 16.4 6.2 22.6 0l16-16c6.2-6.2 6.2-16.4 0-22.6l-16-16zM281 66.7c-2.2-1.5-4.9-2.5-7.7-2.7c-.6 0-1.3-.1-1.9 0c-3.9 .2-7.4 1.7-10.1 4.2c-.9 .8-1.6 1.7-2.3 2.6c-1.7 2.4-2.7 5.3-2.9 8.5c0 .7 0 1.4 0 2.1c.2 2.2 .9 4.3 1.9 6.2l.3 .6c.3 .6 .8 1.4 1.4 2.4c1.2 2 2.9 4.8 5.1 8.2c4.4 6.7 11.1 15.5 20 24.4C302.4 141.1 330.3 160 368 160c31.2 0 56.6-10.4 73.9-20.2c8.7-5 15.6-9.9 20.4-13.8c2.4-1.9 4.3-3.6 5.7-4.9c.7-.6 1.3-1.2 1.7-1.6l.6-.5 .1-.1 .1-.1 0 0 0 0c5.9-5.8 9.5-13.9 9.5-22.8c0-17.7-14.3-32-32-32c-8.7 0-16.7 3.5-22.4 9.2c-.1 .1-.2 .2-.5 .4c-.5 .5-1.5 1.3-2.8 2.4c-2.7 2.2-6.8 5.2-12.1 8.2C399.4 90.4 384.8 96 368 96c-20.8 0-42.4-7-59.5-14.6c-8.4-3.7-15.4-7.5-20.3-10.3c-2.4-1.4-4.3-2.5-5.6-3.3c-.6-.4-1.1-.7-1.4-.9l-.3-.2zM115.2 169.6c8-6 17.9-9.6 28.8-9.6c26.5 0 48 21.5 48 48s-21.5 48-48 48H109.8c-7.6 0-13.8 6.2-13.8 13.8c0 1.5 .2 2.9 .7 4.4l8 24c4.4 13.1 16.6 21.9 30.4 21.9H144h16c35.3 0 64 28.7 64 64s-28.7 64-64 64c-50.8 0-82.7-21.5-102.2-42.8c-9.9-10.8-16.6-21.6-20.9-29.7c-2.1-4-3.6-7.3-4.5-9.6c-.5-1.1-.8-2-1-2.5l-.2-.5c-.3-.9-.7-1.8-1.1-2.6c-1.2-2.2-2.8-4-4.7-5.4c-1.9-1.4-4.1-2.3-6.5-2.8c-1.4-.3-2.9-.3-4.4-.2c-2.5 .2-4.8 1-6.8 2.3c-1.1 .7-2.2 1.5-3.1 2.5c-2.4 2.5-4.1 5.8-4.5 9.5c-.1 .6-.1 1.1-.1 1.7c0 0 0 0 0 0c0 .8 .1 1.7 .2 2.5l0 .1c0 .3 .1 .8 .2 1.3c.2 1.1 .4 2.7 .8 4.6c.8 3.9 2 9.4 3.9 15.9c3.8 13 10.3 30.4 21.3 48C48.7 476.2 89.4 512 160 512c70.7 0 128-57.3 128-128c0-23.3-6.2-45.2-17.1-64h22.6c25.5 0 49.9-10.1 67.9-28.1l26.5-26.5c6-6 14.1-9.4 22.6-9.4H416c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32c-25.7 0-41.4-12.5-51.2-25.6c-5-6.7-8.4-13.4-10.5-18.6c-1.1-2.5-1.8-4.6-2.2-6c-.2-.7-.4-1.2-.5-1.5l-.1-.2c-.3-1.3-.8-2.6-1.5-3.8c-1.1-2-2.6-3.8-4.4-5.1c-2.7-2-6-3.2-9.6-3.2l-.2 0c-8 .1-14.6 6.1-15.6 13.9l0 0c0 .3-.1 .6-.2 1.1c-.1 .9-.3 2.1-.4 3.6c-.3 3-.6 7.3-.6 12.4c0 10.1 1.1 23.9 5.8 38.1c4.8 14.3 13.4 29.3 28.6 40.7C368.7 473.3 389.3 480 416 480c53 0 96-43 96-96V288c0-53-43-96-96-96h-5.5c-25.5 0-49.9 10.1-67.9 28.1l-26.5 26.5c-6 6-14.1 9.4-22.6 9.4H245.2c6.9-14.5 10.8-30.8 10.8-48c0-61.9-50.1-112-112-112c-25.2 0-48.5 8.3-67.2 22.4c-14.1 10.6-17 30.7-6.4 44.8s30.7 17 44.8 6.4z"],worm:[512,512,[],"e599","M256 96c0-53 43-96 96-96h38.4C439.9 0 480 40.1 480 89.6V176v16V376c0 75.1-60.9 136-136 136s-136-60.9-136-136V296c0-22.1-17.9-40-40-40s-40 17.9-40 40V464c0 26.5-21.5 48-48 48s-48-21.5-48-48V296c0-75.1 60.9-136 136-136s136 60.9 136 136v80c0 22.1 17.9 40 40 40s40-17.9 40-40V192H352c-53 0-96-43-96-96zm144-8a24 24 0 1 0 -48 0 24 24 0 1 0 48 0z"],"house-circle-xmark":[640,512,[],"e50b","M320.7 352c8.1-89.7 83.5-160 175.3-160c8.9 0 17.6 .7 26.1 1.9L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32v69.7c-.1 .9-.1 1.8-.1 2.8V472c0 22.1 17.9 40 40 40h16c1.2 0 2.4-.1 3.6-.2c1.5 .1 3 .2 4.5 .2H160h24c22.1 0 40-17.9 40-40V448 384c0-17.7 14.3-32 32-32h64l.7 0zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L518.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L496 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L473.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L496 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z"],plug:[384,512,[128268],"f1e6","M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8C297 398 352 333.4 352 256V224c17.7 0 32-14.3 32-32s-14.3-32-32-32H32z"],"chevron-up":[512,512,[],"f077","M233.4 105.4c12.5-12.5 32.8-12.5 45.3 0l192 192c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L256 173.3 86.6 342.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l192-192z"],"hand-spock":[576,512,[128406],"f259","M246.9 23.7C242.3 6.6 224.8-3.5 207.7 1.1s-27.2 22.1-22.6 39.2L238 237.8c2.5 9.2-4.5 18.2-14 18.2c-6.4 0-12-4.2-13.9-10.3L166.6 102.7c-5.1-16.9-23-26.4-39.9-21.3s-26.4 23-21.3 39.9l62.8 206.4c2.4 7.9-7.2 13.8-13.2 8.1L99.6 283c-16-15.2-41.3-14.6-56.6 1.4s-14.6 41.3 1.4 56.6L156.8 448c43.1 41.1 100.4 64 160 64h10.9 8.2c.1 0 .1-.1 .1-.1s.1-.1 .1-.1c58.3-3.5 108.6-43.2 125.3-99.7l81.2-275c5-16.9-4.7-34.7-21.6-39.8s-34.7 4.7-39.8 21.6L443.5 247.1c-1.6 5.3-6.4 8.9-12 8.9c-7.9 0-13.8-7.3-12.2-15.1l36-170.3c3.7-17.3-7.4-34.3-24.7-37.9s-34.3 7.4-37.9 24.7L355.1 235.1c-2.6 12.2-13.3 20.9-25.8 20.9c-11.9 0-22.4-8-25.4-19.5l-57-212.8z"],stopwatch:[448,512,[9201],"f2f2","M176 0c-17.7 0-32 14.3-32 32s14.3 32 32 32h16V98.4C92.3 113.8 16 200 16 304c0 114.9 93.1 208 208 208s208-93.1 208-208c0-41.8-12.3-80.7-33.5-113.2l24.1-24.1c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L355.7 143c-28.1-23-62.2-38.8-99.7-44.6V64h16c17.7 0 32-14.3 32-32s-14.3-32-32-32H224 176zm72 192V320c0 13.3-10.7 24-24 24s-24-10.7-24-24V192c0-13.3 10.7-24 24-24s24 10.7 24 24z"],"face-kiss":[512,512,[128535,"kiss"],"f596","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm48.7-198.3c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4c-2.7 1.5-5.7 3-8.7 4.3c3.1 1.3 6 2.7 8.7 4.3c6.6 3.7 12.5 8.2 16.7 13.4c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4C274.7 443.1 257.4 448 240 448c-3.6 0-6.8-2.5-7.7-6s.6-7.2 3.8-9l0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1c-.4-.2-.7-.4-.9-.5l-.2-.1 0 0 0 0 0 0c-2.5-1.4-4.1-4.1-4.1-7s1.6-5.6 4.1-7l0 0 0 0 0 0 0 0 0 0 .2-.1 .3-.2 .6-.4c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1l-.4-.3-.5-.3-.2-.1 0 0 0 0 0 0c-3.2-1.8-4.7-5.5-3.8-9s4.1-6 7.7-6c17.4 0 34.7 4.9 47.9 12.3c6.6 3.7 12.5 8.2 16.7 13.4zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"bridge-circle-xmark":[640,512,[],"e4cb","M64 32C46.3 32 32 46.3 32 64s14.3 32 32 32h40v64H32V288c53 0 96 43 96 96v64c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V384c0-53 43-96 96-96c6.3 0 12.4 .6 18.3 1.7C367.1 231.8 426.9 192 496 192c42.5 0 81.6 15.1 112 40.2V160H536V96h40c17.7 0 32-14.3 32-32s-14.3-32-32-32H64zM488 96v64H408V96h80zM360 96v64H280V96h80zM232 96v64H152V96h80zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L518.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L496 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L473.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L496 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z"],"face-grin-tongue":[512,512,[128539,"grin-tongue"],"f589","M0 256C0 368.9 73.1 464.7 174.5 498.8C165.3 484 160 466.6 160 448V400.7c-24-17.5-43.1-41.4-54.8-69.2c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19c12.3-3.8 24.3 6.9 19.3 18.7c-11.8 28-31.1 52-55.4 69.6V448c0 18.6-5.3 36-14.5 50.8C438.9 464.7 512 368.9 512 256C512 114.6 397.4 0 256 0S0 114.6 0 256zm176.4-80a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM320 448V402.6c0-14.7-11.9-26.6-26.6-26.6h-2c-11.3 0-21.1 7.9-23.6 18.9c-2.8 12.6-20.8 12.6-23.6 0c-2.5-11.1-12.3-18.9-23.6-18.9h-2c-14.7 0-26.6 11.9-26.6 26.6V448c0 35.3 28.7 64 64 64s64-28.7 64-64z"],"chess-bishop":[320,512,[9821],"f43a","M128 0C110.3 0 96 14.3 96 32c0 16.1 11.9 29.4 27.4 31.7C78.4 106.8 8 190 8 288c0 47.4 30.8 72.3 56 84.7V400H256V372.7c25.2-12.5 56-37.4 56-84.7c0-37.3-10.2-72.4-25.3-104.1l-99.4 99.4c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L270.8 154.6c-23.2-38.1-51.8-69.5-74.2-90.9C212.1 61.4 224 48.1 224 32c0-17.7-14.3-32-32-32H128zM48 432L6.6 473.4c-4.2 4.2-6.6 10-6.6 16C0 501.9 10.1 512 22.6 512H297.4c12.5 0 22.6-10.1 22.6-22.6c0-6-2.4-11.8-6.6-16L272 432H48z"],"face-grin-wink":[512,512,["grin-wink"],"f58c","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM388.1 312.8c12.3-3.8 24.3 6.9 19.3 18.7C382.4 390.6 324.2 432 256.3 432s-126.2-41.4-151.1-100.5c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19zm-16.9-79.2c-17.6-23.5-52.8-23.5-70.4 0c-5.3 7.1-15.3 8.5-22.4 3.2s-8.5-15.3-3.2-22.4c30.4-40.5 91.2-40.5 121.6 0c5.3 7.1 3.9 17.1-3.2 22.4s-17.1 3.9-22.4-3.2zM176.4 176a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"ear-deaf":[512,512,["deaf","deafness","hard-of-hearing"],"f2a4","M502.6 54.6l-40 40c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l40-40c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3zm-320 320l-128 128c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l128-128c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3zM240 128c-57.6 0-105.1 43.6-111.3 99.5c-1.9 17.6-17.8 30.2-35.3 28.3s-30.2-17.8-28.3-35.3C74.8 132.5 149.4 64 240 64c97.2 0 176 78.8 176 176c0 46-17.7 87.9-46.6 119.3c-12 13-17.4 24.8-17.4 34.7V400c0 61.9-50.1 112-112 112c-17.7 0-32-14.3-32-32s14.3-32 32-32c26.5 0 48-21.5 48-48v-6.1c0-32.9 17.4-59.6 34.4-78c18.4-20 29.6-46.6 29.6-75.9c0-61.9-50.1-112-112-112zm0 80c-17.7 0-32 14.3-32 32c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-44.2 35.8-80 80-80s80 35.8 80 80c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-17.7-14.3-32-32-32z"],"road-circle-check":[640,512,[],"e564","M213.2 32H288V96c0 17.7 14.3 32 32 32s32-14.3 32-32V32h74.8c27.1 0 51.3 17.1 60.3 42.6l42.7 120.6c-10.9-2.1-22.2-3.2-33.8-3.2c-59.5 0-112.1 29.6-144 74.8V224c0-17.7-14.3-32-32-32s-32 14.3-32 32v64c0 17.7 14.3 32 32 32c2.3 0 4.6-.3 6.8-.7c-4.5 15.5-6.8 31.8-6.8 48.7c0 5.4 .2 10.7 .7 16l-.7 0c-17.7 0-32 14.3-32 32v64H86.6C56.5 480 32 455.5 32 425.4c0-6.2 1.1-12.4 3.1-18.2L152.9 74.6C162 49.1 186.1 32 213.2 32zM352 368a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm211.3-43.3c-6.2-6.2-16.4-6.2-22.6 0L480 385.4l-28.7-28.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l40 40c6.2 6.2 16.4 6.2 22.6 0l72-72c6.2-6.2 6.2-16.4 0-22.6z"],"dice-five":[448,512,[9860],"f523","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm64 96a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM96 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM224 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm64-64a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm32 160a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"square-rss":[448,512,["rss-square"],"f143","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM96 136c0-13.3 10.7-24 24-24c137 0 248 111 248 248c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-110.5-89.5-200-200-200c-13.3 0-24-10.7-24-24zm0 96c0-13.3 10.7-24 24-24c83.9 0 152 68.1 152 152c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-57.4-46.6-104-104-104c-13.3 0-24-10.7-24-24zm0 120a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],"land-mine-on":[640,512,[],"e51b","M344 24V168c0 13.3-10.7 24-24 24s-24-10.7-24-24V24c0-13.3 10.7-24 24-24s24 10.7 24 24zM192 320c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32v32H192V320zm-77.3 90.5c8.1-16.3 24.8-26.5 42.9-26.5H482.3c18.2 0 34.8 10.3 42.9 26.5l27.6 55.2C563.5 487 548 512 524.2 512H115.8c-23.8 0-39.3-25-28.6-46.3l27.6-55.2zM36.3 138.3c7.5-10.9 22.5-13.6 33.4-6.1l104 72c10.9 7.5 13.6 22.5 6.1 33.4s-22.5 13.6-33.4 6.1l-104-72c-10.9-7.5-13.6-22.5-6.1-33.4zm534.1-6.1c10.9-7.5 25.8-4.8 33.4 6.1s4.8 25.8-6.1 33.4l-104 72c-10.9 7.5-25.8 4.8-33.4-6.1s-4.8-25.8 6.1-33.4l104-72z"],"i-cursor":[256,512,[],"f246","M.1 29.3C-1.4 47 11.7 62.4 29.3 63.9l8 .7C70.5 67.3 96 95 96 128.3V224H64c-17.7 0-32 14.3-32 32s14.3 32 32 32H96v95.7c0 33.3-25.5 61-58.7 63.8l-8 .7C11.7 449.6-1.4 465 .1 482.7s16.9 30.7 34.5 29.2l8-.7c34.1-2.8 64.2-18.9 85.4-42.9c21.2 24 51.2 40.1 85.4 42.9l8 .7c17.6 1.5 33.1-11.6 34.5-29.2s-11.6-33.1-29.2-34.5l-8-.7C185.5 444.7 160 417 160 383.7V288h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H160V128.3c0-33.3 25.5-61 58.7-63.8l8-.7c17.6-1.5 30.7-16.9 29.2-34.5S239-1.4 221.3 .1l-8 .7C179.2 3.6 149.2 19.7 128 43.7c-21.2-24-51.2-40-85.4-42.9l-8-.7C17-1.4 1.6 11.7 .1 29.3z"],stamp:[512,512,[],"f5bf","M312 201.8c0-17.4 9.2-33.2 19.9-47C344.5 138.5 352 118.1 352 96c0-53-43-96-96-96s-96 43-96 96c0 22.1 7.5 42.5 20.1 58.8c10.7 13.8 19.9 29.6 19.9 47c0 29.9-24.3 54.2-54.2 54.2H112C50.1 256 0 306.1 0 368c0 20.9 13.4 38.7 32 45.3V464c0 26.5 21.5 48 48 48H432c26.5 0 48-21.5 48-48V413.3c18.6-6.6 32-24.4 32-45.3c0-61.9-50.1-112-112-112H366.2c-29.9 0-54.2-24.3-54.2-54.2zM416 416v32H96V416H416z"],stairs:[576,512,[],"e289","M384 64c0-17.7 14.3-32 32-32H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H448v96c0 17.7-14.3 32-32 32H320v96c0 17.7-14.3 32-32 32H192v96c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32h96V320c0-17.7 14.3-32 32-32h96V192c0-17.7 14.3-32 32-32h96V64z"],i:[320,512,[105],"49","M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96h96V416H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H192V96h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H160 32z"],"hryvnia-sign":[384,512,[8372,"hryvnia"],"f6f2","M121.9 116.2C138.3 103.1 158.7 96 179.6 96H223c27.1 0 49 21.9 49 49c0 11.5-4 22.4-11.1 31H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H155.5l-50.6 28.9c-1.7 1-3.4 2-5.1 3.1H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H52.3c-2.8 9.9-4.3 20.4-4.3 31c0 62.4 50.6 113 113 113h43.4c35.5 0 70-12.1 97.7-34.3L308 441c13.8-11 16-31.2 5-45s-31.2-16-45-5l-5.9 4.7c-16.4 13.1-36.7 20.2-57.7 20.2H161c-27.1 0-49-21.9-49-49c0-11.5 4-22.4 11.1-31H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H228.5l50.6-28.9c1.7-1 3.4-2 5.1-3.1H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H331.7c2.8-10 4.3-20.4 4.3-31c0-62.4-50.6-113-113-113H179.6c-35.5 0-70 12.1-97.7 34.3L76 71c-13.8 11-16 31.2-5 45s31.2 16 45 5l5.9-4.7z"],pills:[576,512,[],"f484","M112 96c-26.5 0-48 21.5-48 48V256h96V144c0-26.5-21.5-48-48-48zM0 144C0 82.1 50.1 32 112 32s112 50.1 112 112V368c0 61.9-50.1 112-112 112S0 429.9 0 368V144zM554.9 399.4c-7.1 12.3-23.7 13.1-33.8 3.1L333.5 214.9c-10-10-9.3-26.7 3.1-33.8C360 167.7 387.1 160 416 160c88.4 0 160 71.6 160 160c0 28.9-7.7 56-21.1 79.4zm-59.5 59.5C472 472.3 444.9 480 416 480c-88.4 0-160-71.6-160-160c0-28.9 7.7-56 21.1-79.4c7.1-12.3 23.7-13.1 33.8-3.1L498.5 425.1c10 10 9.3 26.7-3.1 33.8z"],"face-grin-wide":[512,512,[128515,"grin-alt"],"f581","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM388.1 312.8c12.3-3.8 24.3 6.9 19.3 18.7C382.4 390.6 324.2 432 256.3 432s-126.2-41.4-151.1-100.5c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19zM208 192c0 35.3-14.3 64-32 64s-32-28.7-32-64s14.3-64 32-64s32 28.7 32 64zm128 64c-17.7 0-32-28.7-32-64s14.3-64 32-64s32 28.7 32 64s-14.3 64-32 64z"],tooth:[448,512,[129463],"f5c9","M186.1 52.1C169.3 39.1 148.7 32 127.5 32C74.7 32 32 74.7 32 127.5v6.2c0 15.8 3.7 31.3 10.7 45.5l23.5 47.1c4.5 8.9 7.6 18.4 9.4 28.2l36.7 205.8c2 11.2 11.6 19.4 22.9 19.8s21.4-7.4 24-18.4l28.9-121.3C192.2 323.7 207 312 224 312s31.8 11.7 35.8 28.3l28.9 121.3c2.6 11.1 12.7 18.8 24 18.4s20.9-8.6 22.9-19.8l36.7-205.8c1.8-9.8 4.9-19.3 9.4-28.2l23.5-47.1c7.1-14.1 10.7-29.7 10.7-45.5v-2.1c0-55-44.6-99.6-99.6-99.6c-24.1 0-47.4 8.8-65.6 24.6l-3.2 2.8 19.5 15.2c7 5.4 8.2 15.5 2.8 22.5s-15.5 8.2-22.5 2.8l-24.4-19-37-28.8z"],v:[384,512,[118],"56","M19.7 34.5c16.3-6.8 35 .9 41.8 17.2L192 364.8 322.5 51.7c6.8-16.3 25.5-24 41.8-17.2s24 25.5 17.2 41.8l-160 384c-5 11.9-16.6 19.7-29.5 19.7s-24.6-7.8-29.5-19.7L2.5 76.3c-6.8-16.3 .9-35 17.2-41.8z"],"bangladeshi-taka-sign":[384,512,[],"e2e6","M36 32.2C18.4 30.1 2.4 42.5 .2 60S10.5 93.6 28 95.8l7.9 1c16 2 28 15.6 28 31.8V160H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64V384c0 53 43 96 96 96h32c106 0 192-86 192-192V256c0-53-43-96-96-96H272c-17.7 0-32 14.3-32 32s14.3 32 32 32h16c17.7 0 32 14.3 32 32v32c0 70.7-57.3 128-128 128H160c-17.7 0-32-14.3-32-32V224h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H128V128.5c0-48.4-36.1-89.3-84.1-95.3l-7.9-1z"],bicycle:[640,512,[128690],"f206","M312 32c-13.3 0-24 10.7-24 24s10.7 24 24 24h25.7l34.6 64H222.9l-27.4-38C191 99.7 183.7 96 176 96H120c-13.3 0-24 10.7-24 24s10.7 24 24 24h43.7l22.1 30.7-26.6 53.1c-10-2.5-20.5-3.8-31.2-3.8C57.3 224 0 281.3 0 352s57.3 128 128 128c65.3 0 119.1-48.9 127-112h49c8.5 0 16.3-4.5 20.7-11.8l84.8-143.5 21.7 40.1C402.4 276.3 384 312 384 352c0 70.7 57.3 128 128 128s128-57.3 128-128s-57.3-128-128-128c-13.5 0-26.5 2.1-38.7 6L375.4 48.8C369.8 38.4 359 32 347.2 32H312zM458.6 303.7l32.3 59.7c6.3 11.7 20.9 16 32.5 9.7s16-20.9 9.7-32.5l-32.3-59.7c3.6-.6 7.4-.9 11.2-.9c39.8 0 72 32.2 72 72s-32.2 72-72 72s-72-32.2-72-72c0-18.6 7-35.5 18.6-48.3zM133.2 368h65c-7.3 32.1-36 56-70.2 56c-39.8 0-72-32.2-72-72s32.2-72 72-72c1.7 0 3.4 .1 5.1 .2l-24.2 48.5c-9 18.1 4.1 39.4 24.3 39.4zm33.7-48l50.7-101.3 72.9 101.2-.1 .1H166.8zm90.6-128H365.9L317 274.8 257.4 192z"],"staff-snake":[384,512,["rod-asclepius","rod-snake","staff-aesculapius"],"e579","M222.6 43.2l-.1 4.8H288c53 0 96 43 96 96s-43 96-96 96H248V160h40c8.8 0 16-7.2 16-16s-7.2-16-16-16H248 220l-4.5 144H256c53 0 96 43 96 96s-43 96-96 96H240V384h16c8.8 0 16-7.2 16-16s-7.2-16-16-16H213l-3.1 99.5L208.5 495l0 1c-.3 8.9-7.6 16-16.5 16s-16.2-7.1-16.5-16l0-1-1-31H136c-22.1 0-40-17.9-40-40s17.9-40 40-40h36l-1-32H152c-53 0-96-43-96-96c0-47.6 34.6-87.1 80-94.7V256c0 8.8 7.2 16 16 16h16.5L164 128H136 122.6c-9 18.9-28.3 32-50.6 32H56c-30.9 0-56-25.1-56-56S25.1 48 56 48h8 8 89.5l-.1-4.8L161 32c0-.7 0-1.3 0-1.9c.5-16.6 14.1-30 31-30s30.5 13.4 31 30c0 .6 0 1.3 0 1.9l-.4 11.2zM64 112a16 16 0 1 0 0-32 16 16 0 1 0 0 32z"],"head-side-cough-slash":[640,512,[],"e062","M448 325.8l44 34.5c8.1 1.4 14.8 6.8 18 14.1L552.9 408c10.6 .4 19.5 7.6 22.2 17.4l39.1 30.6c.6 0 1.2-.1 1.8-.1c11.1 0 20.4 7.5 23.2 17.8h-3.9c6.2 8.5 6.4 20.4-.4 29c-8.2 10.4-23.3 12.3-33.7 4.1L9.2 42.9C-1.2 34.7-3.1 19.6 5.1 9.2S28.4-3.1 38.8 5.1L89.6 44.9C127 16.7 173.5 0 224 0h24c95.2 0 181.2 69.3 197.3 160.2c2.3 13 6.8 25.7 15.1 36l42 52.6c6.2 7.8 9.6 17.4 9.6 27.4c0 24.2-19.6 43.8-43.8 43.8H448v0 5.8zM0 224.2c0-38.7 9.8-75.1 27.1-106.9L341.8 365.3l-2.5 .3c-11 1.4-19.2 10.7-19.2 21.8c0 11.6 9 21.2 20.6 21.9l62 3.9 43 33.9C439.3 466.2 421.2 480 400 480H320v8c0 13.3-10.7 24-24 24H256v0H96c-17.7 0-32-14.3-32-32V407.3c0-16.7-6.9-32.5-17.1-45.8C16.6 322.4 0 274.1 0 224.2zM616 360a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm-64-48a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm40-24a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z"],"truck-medical":[640,512,[128657,"ambulance"],"f0f9","M0 48C0 21.5 21.5 0 48 0H368c26.5 0 48 21.5 48 48V96h50.7c17 0 33.3 6.7 45.3 18.7L589.3 192c12 12 18.7 28.3 18.7 45.3V256v32 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H576c0 53-43 96-96 96s-96-43-96-96H256c0 53-43 96-96 96s-96-43-96-96H48c-26.5 0-48-21.5-48-48V48zM416 256H544V237.3L466.7 160H416v96zM160 464a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm368-48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM176 80v48l-48 0c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V192h48c8.8 0 16-7.2 16-16V144c0-8.8-7.2-16-16-16H240V80c0-8.8-7.2-16-16-16H192c-8.8 0-16 7.2-16 16z"],"wheat-awn-circle-exclamation":[640,512,[],"e598","M505 41c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L383 95c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l88-88zM305.5 27.3c-6.2-6.2-16.4-6.2-22.6 0L271.5 38.6c-37.5 37.5-37.5 98.3 0 135.8l10.4 10.4-30.5 30.5c-3.4-27.3-15.5-53.8-36.5-74.8l-11.3-11.3c-6.2-6.2-16.4-6.2-22.6 0l-11.3 11.3c-37.5 37.5-37.5 98.3 0 135.8l10.4 10.4-30.5 30.5c-3.4-27.3-15.5-53.8-36.5-74.8L101.8 231c-6.2-6.2-16.4-6.2-22.6 0L67.9 242.3c-37.5 37.5-37.5 98.3 0 135.8l10.4 10.4L9.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l68.9-68.9 12.2 12.2c37.5 37.5 98.3 37.5 135.8 0l11.3-11.3c6.2-6.2 6.2-16.4 0-22.6l-11.3-11.3c-21.8-21.8-49.6-34.1-78.1-36.9l31.9-31.9 12.2 12.2c22.5 22.5 53.3 31.5 82.4 27c0-1 0-2.1 0-3.1c0-33.1 9.1-64.1 25-90.6c-15.5-8.7-32.5-13.8-49.8-15.5l31.9-31.9 12.2 12.2c6 6 12.6 11.1 19.7 15.2c27.5-34 67.3-57.5 112.6-63.8c-4.1-3.8-8.4-7.3-12.9-10.5L505 137c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-59.4 59.4c-20.6-4.4-42-3.7-62.3 2.1c6.1-21.3 6.6-43.8 1.4-65.3L409 41c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L329.1 52.9c-3.7-5-7.8-9.8-12.4-14.3L305.5 27.3zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z"],snowman:[512,512,[9731,9924],"f7d0","M341.1 140.6c-2 3.9-1.6 8.6 1.2 12c7 8.5 12.9 18.1 17.2 28.4L408 160.2V120c0-13.3 10.7-24 24-24s24 10.7 24 24v19.6l22.5-9.7c12.2-5.2 26.3 .4 31.5 12.6s-.4 26.3-12.6 31.5l-56 24-73.6 31.5c-.5 9.5-2.1 18.6-4.8 27.3c-1.2 3.8-.1 8 2.8 10.8C396.7 296.9 416 338.2 416 384c0 44.7-18.3 85-47.8 114.1c-9.9 9.7-23.7 13.9-37.5 13.9H181.3c-13.9 0-27.7-4.2-37.5-13.9C114.3 469 96 428.7 96 384c0-45.8 19.3-87.1 50.1-116.3c2.9-2.8 4-6.9 2.8-10.8c-2.7-8.7-4.3-17.9-4.8-27.3L70.5 198.1l-56-24C2.4 168.8-3.3 154.7 1.9 142.5s19.3-17.8 31.5-12.6L56 139.6V120c0-13.3 10.7-24 24-24s24 10.7 24 24v40.2L152.6 181c4.3-10.3 10.1-19.9 17.2-28.4c2.8-3.4 3.3-8.1 1.2-12C164 127.2 160 112.1 160 96c0-53 43-96 96-96s96 43 96 96c0 16.1-4 31.2-10.9 44.6zM224 96a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm48 128a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm-16 80a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm16 48a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zM288 96a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm-48 24v3.2c0 3.2 .8 6.3 2.3 9l9 16.9c.9 1.7 2.7 2.8 4.7 2.8s3.8-1.1 4.7-2.8l9-16.9c1.5-2.8 2.3-5.9 2.3-9V120c0-8.8-7.2-16-16-16s-16 7.2-16 16z"],"mortar-pestle":[512,512,[],"f5a7","M504.3 11.1C493.3-1.6 474.5-3.7 461 6.2L252.3 160H397.3L502.6 54.6c11.8-11.8 12.6-30.8 1.6-43.5zM32 192c-17.7 0-32 14.3-32 32s14.3 32 32 32c0 82.5 43.4 147.7 123.9 176.2c-11.1 13.9-19.4 30.3-23.9 48.1C127.6 497.4 142.3 512 160 512H352c17.7 0 32.4-14.6 28.1-31.7c-4.5-17.8-12.8-34.1-23.9-48.1C436.6 403.7 480 338.5 480 256c17.7 0 32-14.3 32-32s-14.3-32-32-32H32z"],"road-barrier":[640,512,[],"e562","M32 32C14.3 32 0 46.3 0 64V448c0 17.7 14.3 32 32 32s32-14.3 32-32V266.3L149.2 96H64V64c0-17.7-14.3-32-32-32zM405.2 96H330.8l-5.4 10.7L234.8 288h74.3l5.4-10.7L405.2 96zM362.8 288h74.3l5.4-10.7L533.2 96H458.8l-5.4 10.7L362.8 288zM202.8 96l-5.4 10.7L106.8 288h74.3l5.4-10.7L277.2 96H202.8zm288 192H576V448c0 17.7 14.3 32 32 32s32-14.3 32-32V64c0-17.7-14.3-32-32-32s-32 14.3-32 32v53.7L490.8 288z"],school:[640,512,[127979],"f549","M337.8 5.4C327-1.8 313-1.8 302.2 5.4L166.3 96H48C21.5 96 0 117.5 0 144V464c0 26.5 21.5 48 48 48H256V416c0-35.3 28.7-64 64-64s64 28.7 64 64v96H592c26.5 0 48-21.5 48-48V144c0-26.5-21.5-48-48-48H473.7L337.8 5.4zM96 192h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V208c0-8.8 7.2-16 16-16zm400 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H512c-8.8 0-16-7.2-16-16V208zM96 320h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zm400 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H512c-8.8 0-16-7.2-16-16V336zM232 176a88 88 0 1 1 176 0 88 88 0 1 1 -176 0zm88-48c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H336V144c0-8.8-7.2-16-16-16z"],igloo:[576,512,[],"f7ae","M320 33.8V160H48.5C100.2 82.8 188.1 32 288 32c10.8 0 21.5 .6 32 1.8zM352 160V39.1C424.9 55.7 487.2 99.8 527.5 160H352zM29.9 192H96V320H0c0-46 10.8-89.4 29.9-128zM192 320H128V192H448V320H384v32H576v80c0 26.5-21.5 48-48 48H352V352c0-35.3-28.7-64-64-64s-64 28.7-64 64V480H48c-26.5 0-48-21.5-48-48V352H192V320zm288 0V192h66.1c19.2 38.6 29.9 82 29.9 128H480z"],joint:[640,512,[],"f595","M448 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V43c0 55.2 21.9 108.1 60.9 147.1l21 21c9 9 14.1 21.2 14.1 33.9v11c0 17.7 14.3 32 32 32s32-14.3 32-32V245c0-29.7-11.8-58.2-32.8-79.2l-21-21C463.2 117.8 448 81.2 448 43V32zM576 256c0 17.7 14.3 32 32 32s32-14.3 32-32V245c0-55.2-21.9-108.1-60.9-147.1l-21-21c-9-9-14.1-21.2-14.1-33.9V32c0-17.7-14.3-32-32-32s-32 14.3-32 32V43c0 29.7 11.8 58.2 32.8 79.2l21 21c27 27 42.2 63.6 42.2 101.8v11zM229.8 360c-4.7-2.3-10-2.7-15.2-2c-37.8 5.6-75.2 14.3-106.9 22.8C81.3 388 58.3 395.1 42 400.4c-8.2 2.7-14.7 4.9-19.2 6.5c-2.3 .8-4 1.4-5.2 1.8l-1.3 .5C6.8 412.5 0 421.4 0 432s6.8 19.5 16.3 22.7l1.3 .5c1.2 .4 3 1.1 5.2 1.8c4.5 1.6 11 3.8 19.2 6.5c16.3 5.4 39.2 12.5 65.7 19.6C160.3 497.3 228.8 512 288 512h67.3c4.1 0 6.3-5.1 3.6-8.3L256.5 380.8c-7.4-8.9-16.5-15.9-26.7-20.8zM445 512h19 51.3c4.1 0 6.3-5.1 3.6-8.3L416.5 380.8C401.3 362.5 378.8 352 355 352H336 288c-1.1 0-2.3 0-3.4 0c-4.1 0-6.2 5.1-3.5 8.3L383.5 483.2C398.7 501.5 421.2 512 445 512zm-3.9-151.7L543.5 483.2c14.6 17.5 35.9 27.9 58.6 28.7c21.1-1.1 37.9-18.6 37.9-39.9V392c0-22.1-17.9-40-40-40H444.7c-4.1 0-6.3 5.1-3.6 8.3z"],"angle-right":[320,512,[8250],"f105","M278.6 233.4c12.5 12.5 12.5 32.8 0 45.3l-160 160c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L210.7 256 73.4 118.6c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0l160 160z"],horse:[576,512,[128014],"f6f0","M448 238.1V160h16l9.8 19.6c12.5 25.1 42.2 36.4 68.3 26c20.5-8.2 33.9-28 33.9-50.1V80c0-19.1-8.4-36.3-21.7-48H560c8.8 0 16-7.2 16-16s-7.2-16-16-16H480 448C377.3 0 320 57.3 320 128H224 203.2 148.8c-30.7 0-57.6 16.3-72.5 40.8C33.2 174.5 0 211.4 0 256v56c0 13.3 10.7 24 24 24s24-10.7 24-24V256c0-13.4 6.6-25.2 16.7-32.5c1.6 13 6.3 25.4 13.6 36.4l28.2 42.4c8.3 12.4 6.4 28.7-1.2 41.6c-16.5 28-20.6 62.2-10 93.9l17.5 52.4c4.4 13.1 16.6 21.9 30.4 21.9h33.7c21.8 0 37.3-21.4 30.4-42.1l-20.8-62.5c-2.1-6.4-.5-13.4 4.3-18.2l12.7-12.7c13.2-13.2 20.6-31.1 20.6-49.7c0-2.3-.1-4.6-.3-6.9l84 24c4.1 1.2 8.2 2.1 12.3 2.8V480c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V315.7c19.2-19.2 31.5-45.7 32-75.7h0v-1.9zM496 64a16 16 0 1 1 0 32 16 16 0 1 1 0-32z"],q:[448,512,[113],"51","M64 256c0 88.4 71.6 160 160 160c28.9 0 56-7.7 79.4-21.1l-72-86.4c-11.3-13.6-9.5-33.8 4.1-45.1s33.8-9.5 45.1 4.1l70.9 85.1C371.9 325.8 384 292.3 384 256c0-88.4-71.6-160-160-160S64 167.6 64 256zM344.9 444.6C310 467 268.5 480 224 480C100.3 480 0 379.7 0 256S100.3 32 224 32s224 100.3 224 224c0 56.1-20.6 107.4-54.7 146.7l47.3 56.8c11.3 13.6 9.5 33.8-4.1 45.1s-33.8 9.5-45.1-4.1l-46.6-55.9z"],g:[448,512,[103],"47","M224 96C135.6 96 64 167.6 64 256s71.6 160 160 160c77.4 0 142-55 156.8-128H256c-17.7 0-32-14.3-32-32s14.3-32 32-32H400c25.8 0 49.6 21.4 47.2 50.6C437.8 389.6 341.4 480 224 480C100.3 480 0 379.7 0 256S100.3 32 224 32c57.4 0 109.7 21.6 149.3 57c13.2 11.8 14.3 32 2.5 45.2s-32 14.3-45.2 2.5C302.3 111.4 265 96 224 96z"],"notes-medical":[512,512,[],"f481","M96 352V96c0-35.3 28.7-64 64-64H416c35.3 0 64 28.7 64 64V293.5c0 17-6.7 33.3-18.7 45.3l-58.5 58.5c-12 12-28.3 18.7-45.3 18.7H160c-35.3 0-64-28.7-64-64zM272 128c-8.8 0-16 7.2-16 16v48H208c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V256h48c8.8 0 16-7.2 16-16V208c0-8.8-7.2-16-16-16H320V144c0-8.8-7.2-16-16-16H272zm24 336c13.3 0 24 10.7 24 24s-10.7 24-24 24H136C60.9 512 0 451.1 0 376V152c0-13.3 10.7-24 24-24s24 10.7 24 24l0 224c0 48.6 39.4 88 88 88H296z"],"temperature-half":[320,512,[127777,"temperature-2","thermometer-2","thermometer-half"],"f2c9","M160 64c-26.5 0-48 21.5-48 48V276.5c0 17.3-7.1 31.9-15.3 42.5C86.2 332.6 80 349.5 80 368c0 44.2 35.8 80 80 80s80-35.8 80-80c0-18.5-6.2-35.4-16.7-48.9c-8.2-10.6-15.3-25.2-15.3-42.5V112c0-26.5-21.5-48-48-48zM48 112C48 50.2 98.1 0 160 0s112 50.1 112 112V276.5c0 .1 .1 .3 .2 .6c.2 .6 .8 1.6 1.7 2.8c18.9 24.4 30.1 55 30.1 88.1c0 79.5-64.5 144-144 144S16 447.5 16 368c0-33.2 11.2-63.8 30.1-88.1c.9-1.2 1.5-2.2 1.7-2.8c.1-.3 .2-.5 .2-.6V112zM208 368c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-20.9 13.4-38.7 32-45.3V208c0-8.8 7.2-16 16-16s16 7.2 16 16V322.7c18.6 6.6 32 24.4 32 45.3z"],"dong-sign":[384,512,[],"e169","M288 32c-17.7 0-32 14.3-32 32l-32 0c-17.7 0-32 14.3-32 32s14.3 32 32 32h32v49.1c-18.8-10.9-40.7-17.1-64-17.1c-70.7 0-128 57.3-128 128s57.3 128 128 128c24.5 0 47.4-6.9 66.8-18.8c5 11.1 16.2 18.8 29.2 18.8c17.7 0 32-14.3 32-32V288 128c17.7 0 32-14.3 32-32s-14.3-32-32-32c0-17.7-14.3-32-32-32zM128 288a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM32 448c-17.7 0-32 14.3-32 32s14.3 32 32 32H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H32z"],capsules:[576,512,[],"f46b","M64 144c0-26.5 21.5-48 48-48s48 21.5 48 48V256H64V144zM0 144V368c0 61.9 50.1 112 112 112s112-50.1 112-112V189.6c1.8 19.1 8.2 38 19.8 54.8L372.3 431.7c35.5 51.7 105.3 64.3 156 28.1s63-107.5 27.5-159.2L427.3 113.3C391.8 61.5 321.9 49 271.3 85.2c-28 20-44.3 50.8-47.3 83V144c0-61.9-50.1-112-112-112S0 82.1 0 144zm296.6 64.2c-16-23.3-10-55.3 11.9-71c21.2-15.1 50.5-10.3 66 12.2l67 97.6L361.6 303l-65-94.8zM491 407.7c-.8 .6-1.6 1.1-2.4 1.6l4-2.8c-.5 .4-1 .8-1.6 1.2z"],"poo-storm":[448,512,["poo-bolt"],"f75a","M236.9 .2c-5.5-.7-11 1.4-14.5 5.7s-4.6 10.1-2.8 15.3c2.8 8.2 4.3 16.9 4.3 26.1c0 21.7-8.5 37.2-21.9 47.6c-13.8 10.8-34 17-57.8 17H128c-35.3 0-64 28.7-64 64c0 12.2 3.4 23.5 9.3 33.2C31.7 216.2 0 252.4 0 296c0 40.9 28 75.4 65.8 85.2c-5.3-18.5 1-38.5 16.2-50.7l160-128c17.6-14.1 42.6-14 60.2 .2s22.8 38.6 12.8 58.8L285.7 320H304c20.4 0 38.5 12.9 45.3 32.1c3.7 10.6 3.5 21.8 0 31.9H360c48.6 0 88-39.4 88-88c0-43.6-31.7-79.8-73.3-86.8c5.9-9.7 9.3-21.1 9.3-33.2c0-35.3-28.7-64-64-64h-1.4c.9-5.4 1.4-10.9 1.4-16.6c0-48.7-36.1-88.9-83.1-95.2zm45.1 227.4c-5.8-4.7-14.2-4.7-20.1-.1l-160 128c-5.3 4.2-7.4 11.4-5.1 17.8s8.3 10.7 15.1 10.7h70.1L129.7 488.8c-3.4 6.7-1.6 14.9 4.3 19.6s14.2 4.7 20.1 .1l160-128c5.3-4.2 7.4-11.4 5.1-17.8s-8.3-10.7-15.1-10.7H233.9l52.4-104.8c3.4-6.7 1.6-14.9-4.3-19.6z"],"face-frown-open":[512,512,[128550,"frown-open"],"f57a","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM176.4 176a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm-122 174.5c-12.4 5.2-26.5-4.1-21.1-16.4c16-36.6 52.4-62.1 94.8-62.1s78.8 25.6 94.8 62.1c5.4 12.3-8.7 21.6-21.1 16.4c-22.4-9.5-47.4-14.8-73.7-14.8s-51.3 5.3-73.7 14.8z"],"hand-point-up":[384,512,[9757],"f0a6","M32 32C32 14.3 46.3 0 64 0S96 14.3 96 32V240H32V32zM224 192c0-17.7 14.3-32 32-32s32 14.3 32 32v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V192zm-64-64c17.7 0 32 14.3 32 32v48c0 17.7-14.3 32-32 32s-32-14.3-32-32V160c0-17.7 14.3-32 32-32zm160 96c0-17.7 14.3-32 32-32s32 14.3 32 32v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V224zm-96 88l0-.6c9.4 5.4 20.3 8.6 32 8.6c13.2 0 25.4-4 35.6-10.8c8.7 24.9 32.5 42.8 60.4 42.8c11.7 0 22.6-3.1 32-8.6V352c0 88.4-71.6 160-160 160H162.3c-42.4 0-83.1-16.9-113.1-46.9L37.5 453.5C13.5 429.5 0 396.9 0 363V336c0-35.3 28.7-64 64-64h88c22.1 0 40 17.9 40 40s-17.9 40-40 40H96c-8.8 0-16 7.2-16 16s7.2 16 16 16h56c39.8 0 72-32.2 72-72z"],"money-bill":[576,512,[],"f0d6","M64 64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H64zm64 320H64V320c35.3 0 64 28.7 64 64zM64 192V128h64c0 35.3-28.7 64-64 64zM448 384c0-35.3 28.7-64 64-64v64H448zm64-192c-35.3 0-64-28.7-64-64h64v64zM288 160a96 96 0 1 1 0 192 96 96 0 1 1 0-192z"],bookmark:[384,512,[128278,61591],"f02e","M0 48V487.7C0 501.1 10.9 512 24.3 512c5 0 9.9-1.5 14-4.4L192 400 345.7 507.6c4.1 2.9 9 4.4 14 4.4c13.4 0 24.3-10.9 24.3-24.3V48c0-26.5-21.5-48-48-48H48C21.5 0 0 21.5 0 48z"],"align-justify":[448,512,[],"f039","M448 64c0-17.7-14.3-32-32-32H32C14.3 32 0 46.3 0 64S14.3 96 32 96H416c17.7 0 32-14.3 32-32zm0 256c0-17.7-14.3-32-32-32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32zM0 192c0 17.7 14.3 32 32 32H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H32c-17.7 0-32 14.3-32 32zM448 448c0-17.7-14.3-32-32-32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32z"],"umbrella-beach":[576,512,[127958],"f5ca","M346.3 271.8l-60.1-21.9L214 448H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H282.1l64.1-176.2zm121.1-.2l-3.3 9.1 67.7 24.6c18.1 6.6 38-4.2 39.6-23.4c6.5-78.5-23.9-155.5-80.8-208.5c2 8 3.2 16.3 3.4 24.8l.2 6c1.8 57-7.3 113.8-26.8 167.4zM462 99.1c-1.1-34.4-22.5-64.8-54.4-77.4c-.9-.4-1.9-.7-2.8-1.1c-33-11.7-69.8-2.4-93.1 23.8l-4 4.5C272.4 88.3 245 134.2 226.8 184l-3.3 9.1L434 269.7l3.3-9.1c18.1-49.8 26.6-102.5 24.9-155.5l-.2-6zM107.2 112.9c-11.1 15.7-2.8 36.8 15.3 43.4l71 25.8 3.3-9.1c19.5-53.6 49.1-103 87.1-145.5l4-4.5c6.2-6.9 13.1-13 20.5-18.2c-79.6 2.5-154.7 42.2-201.2 108z"],"helmet-un":[512,512,[],"e503","M479.5 224C471.2 98.9 367.2 0 240 0C107.5 0 0 107.5 0 240v56.3C0 344.8 39.2 384 87.7 384H200h14.9L343.5 505.4c4.5 4.2 10.4 6.6 16.5 6.6h96c13.3 0 24-10.7 24-24s-10.7-24-24-24H369.5l-1.5-1.5V288h80 32c17.7 0 32-14.3 32-32s-14.3-32-32-32h-.5zM320 417.2l-78-73.7L274.4 288H320V417.2zM285.3 103.1l34.7 52V112c0-8.8 7.2-16 16-16s16 7.2 16 16v96c0 7.1-4.6 13.3-11.4 15.3s-14-.6-17.9-6.4l-34.7-52V208c0 8.8-7.2 16-16 16s-16-7.2-16-16V112c0-7.1 4.6-13.3 11.4-15.3s14 .6 17.9 6.4zM160 112v64c0 8.8 7.2 16 16 16s16-7.2 16-16V112c0-8.8 7.2-16 16-16s16 7.2 16 16v64c0 26.5-21.5 48-48 48s-48-21.5-48-48V112c0-8.8 7.2-16 16-16s16 7.2 16 16z"],bullseye:[512,512,[],"f140","M448 256A192 192 0 1 0 64 256a192 192 0 1 0 384 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm256 80a80 80 0 1 0 0-160 80 80 0 1 0 0 160zm0-224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zM224 256a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],bacon:[576,512,[129363],"f7e5","M439.2 1.2c11.2-3.2 23.2-.1 31.4 8.1L518 56.7l-26.5 7.9c-58 16.6-98.1 39.6-129.6 67.4c-31.2 27.5-53.2 59.1-75.1 90.9l-2.3 3.3C241.6 288.7 195 356.6 72.8 417.7L37.9 435.2 9.4 406.6c-7.3-7.3-10.6-17.6-9-27.8s8.1-18.9 17.3-23.5C136.1 296.2 180.9 231 223.3 169.3l2.3-3.4c21.8-31.8 44.9-64.9 77.7-93.9c33.4-29.5 75.8-53.6 135.9-70.8zM61.8 459l25.4-12.7c129.5-64.7 179.9-138.1 223.8-202l2.2-3.3c22.1-32.1 42.1-60.5 69.9-85.1c27.5-24.3 63.4-45.2 117.3-60.6l0 0 .2-.1 43.1-12.9 23 23c8 8 11.2 19.7 8.3 30.7s-11.3 19.6-22.2 22.7c-51.9 14.8-85.6 34.7-111.1 57.2c-26.1 23-45.1 49.9-67.3 82.1l-2.2 3.2C327.8 365.9 275.5 442 142.3 508.6c-12.3 6.2-27.2 3.7-36.9-6L61.8 459z"],"hand-point-down":[384,512,[],"f0a7","M32 480c0 17.7 14.3 32 32 32s32-14.3 32-32V272H32V480zM224 320c0 17.7 14.3 32 32 32s32-14.3 32-32V256c0-17.7-14.3-32-32-32s-32 14.3-32 32v64zm-64 64c17.7 0 32-14.3 32-32V304c0-17.7-14.3-32-32-32s-32 14.3-32 32v48c0 17.7 14.3 32 32 32zm160-96c0 17.7 14.3 32 32 32s32-14.3 32-32V224c0-17.7-14.3-32-32-32s-32 14.3-32 32v64zm-96-88l0 .6c9.4-5.4 20.3-8.6 32-8.6c13.2 0 25.4 4 35.6 10.8c8.7-24.9 32.5-42.8 60.4-42.8c11.7 0 22.6 3.1 32 8.6V160C384 71.6 312.4 0 224 0H162.3C119.8 0 79.1 16.9 49.1 46.9L37.5 58.5C13.5 82.5 0 115.1 0 149v27c0 35.3 28.7 64 64 64h88c22.1 0 40-17.9 40-40s-17.9-40-40-40H96c-8.8 0-16-7.2-16-16s7.2-16 16-16h56c39.8 0 72 32.2 72 72z"],"arrow-up-from-bracket":[448,512,[],"e09a","M246.6 9.4c-12.5-12.5-32.8-12.5-45.3 0l-128 128c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L192 109.3 192 320c0 17.7 14.3 32 32 32s32-14.3 32-32l0-210.7 73.4 73.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-128-128zM64 352c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 64c0 53 43 96 96 96l256 0c53 0 96-43 96-96l0-64c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 64c0 17.7-14.3 32-32 32L96 448c-17.7 0-32-14.3-32-32l0-64z"],folder:[512,512,[128193,128447,61716,"folder-blank"],"f07b","M64 480H448c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H288c-10.1 0-19.6-4.7-25.6-12.8L243.2 57.6C231.1 41.5 212.1 32 192 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64z"],"file-waveform":[448,512,["file-medical-alt"],"f478","M96 0C60.7 0 32 28.7 32 64V288H144c6.1 0 11.6 3.4 14.3 8.8L176 332.2l49.7-99.4c2.7-5.4 8.3-8.8 14.3-8.8s11.6 3.4 14.3 8.8L281.9 288H352c8.8 0 16 7.2 16 16s-7.2 16-16 16H272c-6.1 0-11.6-3.4-14.3-8.8L240 275.8l-49.7 99.4c-2.7 5.4-8.3 8.8-14.3 8.8s-11.6-3.4-14.3-8.8L134.1 320H32V448c0 35.3 28.7 64 64 64H352c35.3 0 64-28.7 64-64V160H288c-17.7 0-32-14.3-32-32V0H96zM288 0V128H416L288 0z"],radiation:[512,512,[],"f7b9","M216 186.7c-23.9 13.8-40 39.7-40 69.3L32 256C14.3 256-.2 241.6 2 224.1C10.7 154 47.8 92.7 101.3 52c14.1-10.7 33.8-5.3 42.7 10l72 124.7zM256 336c14.6 0 28.2-3.9 40-10.7l72 124.8c8.8 15.3 3.7 35.1-12.6 41.9c-30.6 12.9-64.2 20-99.4 20s-68.9-7.1-99.4-20c-16.3-6.9-21.4-26.6-12.6-41.9l72-124.8c11.8 6.8 25.4 10.7 40 10.7zm224-80l-144 0c0-29.6-16.1-55.5-40-69.3L368 62c8.8-15.3 28.6-20.7 42.7-10c53.6 40.7 90.6 102 99.4 172.1c2.2 17.5-12.4 31.9-30 31.9zM256 208a48 48 0 1 1 0 96 48 48 0 1 1 0-96z"],"chart-simple":[448,512,[],"e473","M160 80c0-26.5 21.5-48 48-48h32c26.5 0 48 21.5 48 48V432c0 26.5-21.5 48-48 48H208c-26.5 0-48-21.5-48-48V80zM0 272c0-26.5 21.5-48 48-48H80c26.5 0 48 21.5 48 48V432c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V272zM368 96h32c26.5 0 48 21.5 48 48V432c0 26.5-21.5 48-48 48H368c-26.5 0-48-21.5-48-48V144c0-26.5 21.5-48 48-48z"],"mars-stroke":[512,512,[9894],"f229","M376 0c-9.7 0-18.5 5.8-22.2 14.8s-1.7 19.3 5.2 26.2l33.4 33.4L370.3 96.4 345 71c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l25.4 25.4L307.8 159c-28.4-19.5-62.7-31-99.8-31c-97.2 0-176 78.8-176 176s78.8 176 176 176s176-78.8 176-176c0-37-11.4-71.4-31-99.8l28.6-28.6L407 201c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-25.4-25.4 22.1-22.1L471 153c6.9 6.9 17.2 8.9 26.2 5.2s14.8-12.5 14.8-22.2V24c0-13.3-10.7-24-24-24H376zm88 48h0v0l0 0zM96 304a112 112 0 1 1 224 0A112 112 0 1 1 96 304z"],vial:[512,512,[129514],"f492","M342.6 9.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l9.4 9.4L28.1 342.6C10.1 360.6 0 385 0 410.5V416c0 53 43 96 96 96h5.5c25.5 0 49.9-10.1 67.9-28.1L448 205.3l9.4 9.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-32-32-96-96-32-32zM205.3 256L352 109.3 402.7 160l-96 96H205.3z"],gauge:[512,512,["dashboard","gauge-med","tachometer-alt-average"],"f624","M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm320 96c0-26.9-16.5-49.9-40-59.3V88c0-13.3-10.7-24-24-24s-24 10.7-24 24V292.7c-23.5 9.5-40 32.5-40 59.3c0 35.3 28.7 64 64 64s64-28.7 64-64zM144 176a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm-16 80a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm288 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM400 144a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z"],"wand-magic-sparkles":[576,512,["magic-wand-sparkles"],"e2ca","M234.7 42.7L197 56.8c-3 1.1-5 4-5 7.2s2 6.1 5 7.2l37.7 14.1L248.8 123c1.1 3 4 5 7.2 5s6.1-2 7.2-5l14.1-37.7L315 71.2c3-1.1 5-4 5-7.2s-2-6.1-5-7.2L277.3 42.7 263.2 5c-1.1-3-4-5-7.2-5s-6.1 2-7.2 5L234.7 42.7zM46.1 395.4c-18.7 18.7-18.7 49.1 0 67.9l34.6 34.6c18.7 18.7 49.1 18.7 67.9 0L529.9 116.5c18.7-18.7 18.7-49.1 0-67.9L495.3 14.1c-18.7-18.7-49.1-18.7-67.9 0L46.1 395.4zM484.6 82.6l-105 105-23.3-23.3 105-105 23.3 23.3zM7.5 117.2C3 118.9 0 123.2 0 128s3 9.1 7.5 10.8L64 160l21.2 56.5c1.7 4.5 6 7.5 10.8 7.5s9.1-3 10.8-7.5L128 160l56.5-21.2c4.5-1.7 7.5-6 7.5-10.8s-3-9.1-7.5-10.8L128 96 106.8 39.5C105.1 35 100.8 32 96 32s-9.1 3-10.8 7.5L64 96 7.5 117.2zm352 256c-4.5 1.7-7.5 6-7.5 10.8s3 9.1 7.5 10.8L416 416l21.2 56.5c1.7 4.5 6 7.5 10.8 7.5s9.1-3 10.8-7.5L480 416l56.5-21.2c4.5-1.7 7.5-6 7.5-10.8s-3-9.1-7.5-10.8L480 352l-21.2-56.5c-1.7-4.5-6-7.5-10.8-7.5s-9.1 3-10.8 7.5L416 352l-56.5 21.2z"],e:[320,512,[101],"45","M64 32C28.7 32 0 60.7 0 96V256 416c0 35.3 28.7 64 64 64H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H64V288H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H64V96H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H64z"],"pen-clip":[512,512,["pen-alt"],"f305","M453.3 19.3l39.4 39.4c25 25 25 65.5 0 90.5l-52.1 52.1 0 0-1-1 0 0-16-16-96-96-17-17 52.1-52.1c25-25 65.5-25 90.5 0zM241 114.9c-9.4-9.4-24.6-9.4-33.9 0L105 217c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9L173.1 81c28.1-28.1 73.7-28.1 101.8 0L288 94.1l17 17 96 96 16 16 1 1-17 17L229.5 412.5c-48 48-109.2 80.8-175.8 94.1l-25 5c-7.9 1.6-16-.9-21.7-6.6s-8.1-13.8-6.6-21.7l5-25c13.3-66.6 46.1-127.8 94.1-175.8L254.1 128 241 114.9z"],"bridge-circle-exclamation":[640,512,[],"e4ca","M64 32C46.3 32 32 46.3 32 64s14.3 32 32 32h40v64H32V288c53 0 96 43 96 96v64c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V384c0-53 43-96 96-96c6.3 0 12.4 .6 18.3 1.7C367.1 231.8 426.9 192 496 192c42.5 0 81.6 15.1 112 40.2V160H536V96h40c17.7 0 32-14.3 32-32s-14.3-32-32-32H64zM488 96v64H408V96h80zM360 96v64H280V96h80zM232 96v64H152V96h80zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z"],user:[448,512,[128100,62144],"f007","M224 256A128 128 0 1 0 224 0a128 128 0 1 0 0 256zm-45.7 48C79.8 304 0 383.8 0 482.3C0 498.7 13.3 512 29.7 512H418.3c16.4 0 29.7-13.3 29.7-29.7C448 383.8 368.2 304 269.7 304H178.3z"],"school-circle-check":[640,512,[],"e56b","M337.8 5.4C327-1.8 313-1.8 302.2 5.4L166.3 96H48C21.5 96 0 117.5 0 144V464c0 26.5 21.5 48 48 48H320v0H256V416c0-35.3 28.7-64 64-64l.3 0h.5c3.4-37.7 18.7-72.1 42.2-99.1C350.2 260 335.6 264 320 264c-48.6 0-88-39.4-88-88s39.4-88 88-88s88 39.4 88 88c0 18.3-5.6 35.3-15.1 49.4c29-21 64.6-33.4 103.1-33.4c59.5 0 112.1 29.6 144 74.8V144c0-26.5-21.5-48-48-48H473.7L337.8 5.4zM96 192h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V208c0-8.8 7.2-16 16-16zm0 128h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zM320 128c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H336V144c0-8.8-7.2-16-16-16zM640 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-99.3-43.3c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 385.4l60.7-60.7z"],dumpster:[576,512,[],"f793","M49.7 32c-10.5 0-19.8 6.9-22.9 16.9L.9 133c-.6 2-.9 4.1-.9 6.1C0 150.7 9.3 160 20.9 160h94L140.5 32H49.7zM272 160V32H173.1L147.5 160H272zm32 0H428.5L402.9 32H304V160zm157.1 0h94c11.5 0 20.9-9.3 20.9-20.9c0-2.1-.3-4.1-.9-6.1L549.2 48.9C546.1 38.9 536.8 32 526.3 32H435.5l25.6 128zM32 192l4 32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H44L64 448c0 17.7 14.3 32 32 32s32-14.3 32-32H448c0 17.7 14.3 32 32 32s32-14.3 32-32l20-160h12c17.7 0 32-14.3 32-32s-14.3-32-32-32h-4l4-32H32z"],"van-shuttle":[640,512,[128656,"shuttle-van"],"f5b6","M64 104v88h96V96H72c-4.4 0-8 3.6-8 8zm482 88L465.1 96H384v96H546zm-226 0V96H224v96h96zM592 384H576c0 53-43 96-96 96s-96-43-96-96H256c0 53-43 96-96 96s-96-43-96-96H48c-26.5 0-48-21.5-48-48V104C0 64.2 32.2 32 72 32H192 352 465.1c18.9 0 36.8 8.3 49 22.8L625 186.5c9.7 11.5 15 26.1 15 41.2V336c0 26.5-21.5 48-48 48zm-64 0a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM160 432a48 48 0 1 0 0-96 48 48 0 1 0 0 96z"],"building-user":[640,512,[],"e4da","M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h89.9c-6.3-10.2-9.9-22.2-9.9-35.1c0-46.9 25.8-87.8 64-109.2V271.8 48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zM576 272a80 80 0 1 0 -160 0 80 80 0 1 0 160 0zM352 477.1c0 19.3 15.6 34.9 34.9 34.9H605.1c19.3 0 34.9-15.6 34.9-34.9c0-51.4-41.7-93.1-93.1-93.1H445.1c-51.4 0-93.1 41.7-93.1 93.1z"],"square-caret-left":[448,512,["caret-square-left"],"f191","M0 416c0 35.3 28.7 64 64 64l320 0c35.3 0 64-28.7 64-64l0-320c0-35.3-28.7-64-64-64L64 32C28.7 32 0 60.7 0 96L0 416zM128 256c0-6.7 2.8-13 7.7-17.6l112-104c7-6.5 17.2-8.2 25.9-4.4s14.4 12.5 14.4 22l0 208c0 9.5-5.7 18.2-14.4 22s-18.9 2.1-25.9-4.4l-112-104c-4.9-4.5-7.7-10.9-7.7-17.6z"],highlighter:[576,512,[],"f591","M315 315l158.4-215L444.1 70.6 229 229 315 315zm-187 5l0 0V248.3c0-15.3 7.2-29.6 19.5-38.6L420.6 8.4C428 2.9 437 0 446.2 0c11.4 0 22.4 4.5 30.5 12.6l54.8 54.8c8.1 8.1 12.6 19 12.6 30.5c0 9.2-2.9 18.2-8.4 25.6L334.4 396.5c-9 12.3-23.4 19.5-38.6 19.5H224l-25.4 25.4c-12.5 12.5-32.8 12.5-45.3 0l-50.7-50.7c-12.5-12.5-12.5-32.8 0-45.3L128 320zM7 466.3l63-63 70.6 70.6-31 31c-4.5 4.5-10.6 7-17 7H24c-13.3 0-24-10.7-24-24v-4.7c0-6.4 2.5-12.5 7-17z"],key:[512,512,[128273],"f084","M336 352c97.2 0 176-78.8 176-176S433.2 0 336 0S160 78.8 160 176c0 18.7 2.9 36.8 8.3 53.7L7 391c-4.5 4.5-7 10.6-7 17v80c0 13.3 10.7 24 24 24h80c13.3 0 24-10.7 24-24V448h40c13.3 0 24-10.7 24-24V384h40c6.4 0 12.5-2.5 17-7l33.3-33.3c16.9 5.4 35 8.3 53.7 8.3zM376 96a40 40 0 1 1 0 80 40 40 0 1 1 0-80z"],bullhorn:[512,512,[128226,128363],"f0a1","M480 32c0-12.9-7.8-24.6-19.8-29.6s-25.7-2.2-34.9 6.9L381.7 53c-48 48-113.1 75-181 75H192 160 64c-35.3 0-64 28.7-64 64v96c0 35.3 28.7 64 64 64l0 128c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V352l8.7 0c67.9 0 133 27 181 75l43.6 43.6c9.2 9.2 22.9 11.9 34.9 6.9s19.8-16.6 19.8-29.6V300.4c18.6-8.8 32-32.5 32-60.4s-13.4-51.6-32-60.4V32zm-64 76.7V240 371.3C357.2 317.8 280.5 288 200.7 288H192V192h8.7c79.8 0 156.5-29.8 215.3-83.3z"],globe:[512,512,[127760],"f0ac","M352 256c0 22.2-1.2 43.6-3.3 64H163.3c-2.2-20.4-3.3-41.8-3.3-64s1.2-43.6 3.3-64H348.7c2.2 20.4 3.3 41.8 3.3 64zm28.8-64H503.9c5.3 20.5 8.1 41.9 8.1 64s-2.8 43.5-8.1 64H380.8c2.1-20.6 3.2-42 3.2-64s-1.1-43.4-3.2-64zm112.6-32H376.7c-10-63.9-29.8-117.4-55.3-151.6c78.3 20.7 142 77.5 171.9 151.6zm-149.1 0H167.7c6.1-36.4 15.5-68.6 27-94.7c10.5-23.6 22.2-40.7 33.5-51.5C239.4 3.2 248.7 0 256 0s16.6 3.2 27.8 13.8c11.3 10.8 23 27.9 33.5 51.5c11.6 26 20.9 58.2 27 94.7zm-209 0H18.6C48.6 85.9 112.2 29.1 190.6 8.4C165.1 42.6 145.3 96.1 135.3 160zM8.1 192H131.2c-2.1 20.6-3.2 42-3.2 64s1.1 43.4 3.2 64H8.1C2.8 299.5 0 278.1 0 256s2.8-43.5 8.1-64zM194.7 446.6c-11.6-26-20.9-58.2-27-94.6H344.3c-6.1 36.4-15.5 68.6-27 94.6c-10.5 23.6-22.2 40.7-33.5 51.5C272.6 508.8 263.3 512 256 512s-16.6-3.2-27.8-13.8c-11.3-10.8-23-27.9-33.5-51.5zM135.3 352c10 63.9 29.8 117.4 55.3 151.6C112.2 482.9 48.6 426.1 18.6 352H135.3zm358.1 0c-30 74.1-93.6 130.9-171.9 151.6c25.5-34.2 45.2-87.7 55.3-151.6H493.4z"],synagogue:[640,512,[128333],"f69b","M309.8 3.7c5.9-4.9 14.6-4.9 20.5 0l121 100.8C469.5 119.7 480 142.2 480 166V280.1 512H464 352V416c0-17.7-14.3-32-32-32s-32 14.3-32 32v96H176 160V280.1 166c0-23.7 10.5-46.3 28.8-61.5L309.8 3.7zM512 512V244.5l28.1-31.2c3-3.4 7.4-5.3 11.9-5.3s8.9 1.9 11.9 5.3l63.8 70.9c7.9 8.8 12.3 20.3 12.3 32.1V448c0 35.3-28.7 64-64 64H512zM128 244.5V512H64c-35.3 0-64-28.7-64-64V316.3c0-11.9 4.4-23.3 12.3-32.1l63.8-70.9c3-3.4 7.4-5.3 11.9-5.3s8.9 1.9 11.9 5.3L128 244.5zM327 124.3c-3.1-5.4-10.9-5.4-13.9 0l-15.9 28.1-32.3-.3c-6.2-.1-10.1 6.7-7 12.1L274.3 192l-16.4 27.8c-3.2 5.4 .7 12.1 7 12.1l32.3-.3L313 259.7c3.1 5.4 10.9 5.4 13.9 0l15.9-28.1 32.3 .3c6.2 .1 10.1-6.7 7-12.1L365.7 192l16.4-27.8c3.2-5.4-.7-12.1-7-12.1l-32.3 .3L327 124.3z"],"person-half-dress":[320,512,[],"e548","M160 0a48 48 0 1 1 0 96 48 48 0 1 1 0-96zm8 352V128h6.9c33.7 0 64.9 17.7 82.3 46.6l58.3 97c9.1 15.1 4.2 34.8-10.9 43.9s-34.8 4.2-43.9-10.9L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352h0zM58.2 182.3c19.9-33.1 55.3-53.5 93.8-54.3V384h0v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384H70.2c-10.9 0-18.6-10.7-15.2-21.1L93.3 248.1 59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l53.6-89.2z"],"road-bridge":[640,512,[],"e563","M352 0H608c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H352c-17.7 0-32-14.3-32-32V32c0-17.7 14.3-32 32-32zM480 200c-13.3 0-24 10.7-24 24v64c0 13.3 10.7 24 24 24s24-10.7 24-24V224c0-13.3-10.7-24-24-24zm24 184c0-13.3-10.7-24-24-24s-24 10.7-24 24v64c0 13.3 10.7 24 24 24s24-10.7 24-24V384zM480 40c-13.3 0-24 10.7-24 24v64c0 13.3 10.7 24 24 24s24-10.7 24-24V64c0-13.3-10.7-24-24-24zM32 96H288v64H248v64h40v96c-53 0-96 43-96 96v64c0 17.7-14.3 32-32 32H128c-17.7 0-32-14.3-32-32V416c0-53-43-96-96-96V224H72V160H32c-17.7 0-32-14.3-32-32s14.3-32 32-32zm168 64H120v64h80V160z"],"location-arrow":[448,512,[],"f124","M429.6 92.1c4.9-11.9 2.1-25.6-7-34.7s-22.8-11.9-34.7-7l-352 144c-14.2 5.8-22.2 20.8-19.3 35.8s16.1 25.8 31.4 25.8H224V432c0 15.3 10.8 28.4 25.8 31.4s30-5.1 35.8-19.3l144-352z"],c:[384,512,[99],"43","M329.1 142.9c-62.5-62.5-155.8-62.5-218.3 0s-62.5 163.8 0 226.3s155.8 62.5 218.3 0c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3c-87.5 87.5-221.3 87.5-308.8 0s-87.5-229.3 0-316.8s221.3-87.5 308.8 0c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0z"],"tablet-button":[448,512,[],"f10a","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zM224 400a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"building-lock":[576,512,[],"e4d6","M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h88.6c-5.4-9.4-8.6-20.3-8.6-32V352c0-23.7 12.9-44.4 32-55.4V272c0-30.5 12.2-58.2 32-78.4V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zM464 240c17.7 0 32 14.3 32 32v48H432V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H544c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z"],"pizza-slice":[512,512,[],"f818","M169.7 .9c-22.8-1.6-41.9 14-47.5 34.7L110.4 80c.5 0 1.1 0 1.6 0c176.7 0 320 143.3 320 320c0 .5 0 1.1 0 1.6l44.4-11.8c20.8-5.5 36.3-24.7 34.7-47.5C498.5 159.5 352.5 13.5 169.7 .9zM399.8 410.2c.1-3.4 .2-6.8 .2-10.2c0-159.1-128.9-288-288-288c-3.4 0-6.8 .1-10.2 .2L.5 491.9c-1.5 5.5 .1 11.4 4.1 15.4s9.9 5.6 15.4 4.1L399.8 410.2zM176 208a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm64 128a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM96 384a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],"money-bill-wave":[576,512,[],"f53a","M0 112.5V422.3c0 18 10.1 35 27 41.3c87 32.5 174 10.3 261-11.9c79.8-20.3 159.6-40.7 239.3-18.9c23 6.3 48.7-9.5 48.7-33.4V89.7c0-18-10.1-35-27-41.3C462 15.9 375 38.1 288 60.3C208.2 80.6 128.4 100.9 48.7 79.1C25.6 72.8 0 88.6 0 112.5zM288 352c-44.2 0-80-43-80-96s35.8-96 80-96s80 43 80 96s-35.8 96-80 96zM64 352c35.3 0 64 28.7 64 64H64V352zm64-208c0 35.3-28.7 64-64 64V144h64zM512 304v64H448c0-35.3 28.7-64 64-64zM448 96h64v64c-35.3 0-64-28.7-64-64z"],"chart-area":[512,512,["area-chart"],"f1fe","M64 64c0-17.7-14.3-32-32-32S0 46.3 0 64V400c0 44.2 35.8 80 80 80H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H80c-8.8 0-16-7.2-16-16V64zm96 288H448c17.7 0 32-14.3 32-32V251.8c0-7.6-2.7-15-7.7-20.8l-65.8-76.8c-12.1-14.2-33.7-15-46.9-1.8l-21 21c-10 10-26.4 9.2-35.4-1.6l-39.2-47c-12.6-15.1-35.7-15.4-48.7-.6L135.9 215c-5.1 5.8-7.9 13.3-7.9 21.1v84c0 17.7 14.3 32 32 32z"],"house-flag":[640,512,[],"e50d","M480 0c-17.7 0-32 14.3-32 32V192 512h64V192H624c8.8 0 16-7.2 16-16V48c0-8.8-7.2-16-16-16H512c0-17.7-14.3-32-32-32zM416 159L276.8 39.7c-12-10.3-29.7-10.3-41.7 0l-224 192C1 240.4-2.7 254.5 2 267.1S18.6 288 32 288H64V480c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V384c0-17.7 14.3-32 32-32h64c17.7 0 32 14.3 32 32v96c0 17.7 14.3 32 32 32h64.7l.2 0h-1V159z"],"person-circle-minus":[576,512,[],"e540","M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l44.9 74.7c-16.1 17.6-28.6 38.5-36.6 61.5c-1.9-1.8-3.5-3.9-4.9-6.3L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152zm136 16a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm224 0c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16s7.2 16 16 16H496c8.8 0 16-7.2 16-16z"],ban:[512,512,[128683,"cancel"],"f05e","M367.2 412.5L99.5 144.8C77.1 176.1 64 214.5 64 256c0 106 86 192 192 192c41.5 0 79.9-13.1 111.2-35.5zm45.3-45.3C434.9 335.9 448 297.5 448 256c0-106-86-192-192-192c-41.5 0-79.9 13.1-111.2 35.5L412.5 367.2zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256z"],"camera-rotate":[640,512,[],"e0d8","M213.1 64.8L202.7 96H128c-35.3 0-64 28.7-64 64V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H437.3L426.9 64.8C420.4 45.2 402.1 32 381.4 32H258.6c-20.7 0-39 13.2-45.5 32.8zM448 256c0 8.8-7.2 16-16 16H355.3c-6.2 0-11.3-5.1-11.3-11.3c0-3 1.2-5.9 3.3-8L371 229c-13.6-13.4-31.9-21-51-21c-19.2 0-37.7 7.6-51.3 21.3L249 249c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l19.7-19.7C257.4 172.7 288 160 320 160c31.8 0 62.4 12.6 85 35l23.7-23.7c2.1-2.1 5-3.3 8-3.3c6.2 0 11.3 5.1 11.3 11.3V256zM192 320c0-8.8 7.2-16 16-16h76.7c6.2 0 11.3 5.1 11.3 11.3c0 3-1.2 5.9-3.3 8L269 347c13.6 13.4 31.9 21 51 21c19.2 0 37.7-7.6 51.3-21.3L391 327c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-19.7 19.7C382.6 403.3 352 416 320 416c-31.8 0-62.4-12.6-85-35l-23.7 23.7c-2.1 2.1-5 3.3-8 3.3c-6.2 0-11.3-5.1-11.3-11.3V320z"],"spray-can-sparkles":[512,512,["air-freshener"],"f5d0","M96 32v96H224V32c0-17.7-14.3-32-32-32H128C110.3 0 96 14.3 96 32zm0 128c-53 0-96 43-96 96V464c0 26.5 21.5 48 48 48H272c26.5 0 48-21.5 48-48V256c0-53-43-96-96-96H96zm64 96a80 80 0 1 1 0 160 80 80 0 1 1 0-160zM384 48c0-1.4-1-3-2.2-3.6L352 32 339.6 2.2C339 1 337.4 0 336 0s-3 1-3.6 2.2L320 32 290.2 44.4C289 45 288 46.6 288 48c0 1.4 1 3 2.2 3.6L320 64l12.4 29.8C333 95 334.6 96 336 96s3-1 3.6-2.2L352 64l29.8-12.4C383 51 384 49.4 384 48zm76.4 45.8C461 95 462.6 96 464 96s3-1 3.6-2.2L480 64l29.8-12.4C511 51 512 49.4 512 48c0-1.4-1-3-2.2-3.6L480 32 467.6 2.2C467 1 465.4 0 464 0s-3 1-3.6 2.2L448 32 418.2 44.4C417 45 416 46.6 416 48c0 1.4 1 3 2.2 3.6L448 64l12.4 29.8zm7.2 100.4c-.6-1.2-2.2-2.2-3.6-2.2s-3 1-3.6 2.2L448 224l-29.8 12.4c-1.2 .6-2.2 2.2-2.2 3.6c0 1.4 1 3 2.2 3.6L448 256l12.4 29.8c.6 1.2 2.2 2.2 3.6 2.2s3-1 3.6-2.2L480 256l29.8-12.4c1.2-.6 2.2-2.2 2.2-3.6c0-1.4-1-3-2.2-3.6L480 224l-12.4-29.8zM448 144c0-1.4-1-3-2.2-3.6L416 128 403.6 98.2C403 97 401.4 96 400 96s-3 1-3.6 2.2L384 128l-29.8 12.4c-1.2 .6-2.2 2.2-2.2 3.6c0 1.4 1 3 2.2 3.6L384 160l12.4 29.8c.6 1.2 2.2 2.2 3.6 2.2s3-1 3.6-2.2L416 160l29.8-12.4c1.2-.6 2.2-2.2 2.2-3.6z"],star:[576,512,[11088,61446],"f005","M316.9 18C311.6 7 300.4 0 288.1 0s-23.4 7-28.8 18L195 150.3 51.4 171.5c-12 1.8-22 10.2-25.7 21.7s-.7 24.2 7.9 32.7L137.8 329 113.2 474.7c-2 12 3 24.2 12.9 31.3s23 8 33.8 2.3l128.3-68.5 128.3 68.5c10.8 5.7 23.9 4.9 33.8-2.3s14.9-19.3 12.9-31.3L438.5 329 542.7 225.9c8.6-8.5 11.7-21.2 7.9-32.7s-13.7-19.9-25.7-21.7L381.2 150.3 316.9 18z"],repeat:[512,512,[128257],"f363","M0 224c0 17.7 14.3 32 32 32s32-14.3 32-32c0-53 43-96 96-96H320v32c0 12.9 7.8 24.6 19.8 29.6s25.7 2.2 34.9-6.9l64-64c12.5-12.5 12.5-32.8 0-45.3l-64-64c-9.2-9.2-22.9-11.9-34.9-6.9S320 19.1 320 32V64H160C71.6 64 0 135.6 0 224zm512 64c0-17.7-14.3-32-32-32s-32 14.3-32 32c0 53-43 96-96 96H192V352c0-12.9-7.8-24.6-19.8-29.6s-25.7-2.2-34.9 6.9l-64 64c-12.5 12.5-12.5 32.8 0 45.3l64 64c9.2 9.2 22.9 11.9 34.9 6.9s19.8-16.6 19.8-29.6V448H352c88.4 0 160-71.6 160-160z"],cross:[384,512,[128327,10013],"f654","M176 0c-26.5 0-48 21.5-48 48v80H48c-26.5 0-48 21.5-48 48v32c0 26.5 21.5 48 48 48h80V464c0 26.5 21.5 48 48 48h32c26.5 0 48-21.5 48-48V256h80c26.5 0 48-21.5 48-48V176c0-26.5-21.5-48-48-48H256V48c0-26.5-21.5-48-48-48H176z"],box:[448,512,[128230],"f466","M50.7 58.5L0 160H208V32H93.7C75.5 32 58.9 42.3 50.7 58.5zM240 160H448L397.3 58.5C389.1 42.3 372.5 32 354.3 32H240V160zm208 32H0V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V192z"],"venus-mars":[640,512,[9892],"f228","M176 288a112 112 0 1 0 0-224 112 112 0 1 0 0 224zM352 176c0 86.3-62.1 158.1-144 173.1V384h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H208v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H112c-17.7 0-32-14.3-32-32s14.3-32 32-32h32V349.1C62.1 334.1 0 262.3 0 176C0 78.8 78.8 0 176 0s176 78.8 176 176zM271.9 360.6c19.3-10.1 36.9-23.1 52.1-38.4c20 18.5 46.7 29.8 76.1 29.8c61.9 0 112-50.1 112-112s-50.1-112-112-112c-7.2 0-14.3 .7-21.1 2c-4.9-21.5-13-41.7-24-60.2C369.3 66 384.4 64 400 64c37 0 71.4 11.4 99.8 31l20.6-20.6L487 41c-6.9-6.9-8.9-17.2-5.2-26.2S494.3 0 504 0H616c13.3 0 24 10.7 24 24V136c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-33.4-33.4L545 140.2c19.5 28.4 31 62.7 31 99.8c0 97.2-78.8 176-176 176c-50.5 0-96-21.3-128.1-55.4z"],"arrow-pointer":[320,512,["mouse-pointer"],"f245","M0 55.2V426c0 12.2 9.9 22 22 22c6.3 0 12.4-2.7 16.6-7.5L121.2 346l58.1 116.3c7.9 15.8 27.1 22.2 42.9 14.3s22.2-27.1 14.3-42.9L179.8 320H297.9c12.2 0 22.1-9.9 22.1-22.1c0-6.3-2.7-12.3-7.4-16.5L38.6 37.9C34.3 34.1 28.9 32 23.2 32C10.4 32 0 42.4 0 55.2z"],maximize:[512,512,["expand-arrows-alt"],"f31e","M200 32H56C42.7 32 32 42.7 32 56V200c0 9.7 5.8 18.5 14.8 22.2s19.3 1.7 26.2-5.2l40-40 79 79-79 79L73 295c-6.9-6.9-17.2-8.9-26.2-5.2S32 302.3 32 312V456c0 13.3 10.7 24 24 24H200c9.7 0 18.5-5.8 22.2-14.8s1.7-19.3-5.2-26.2l-40-40 79-79 79 79-40 40c-6.9 6.9-8.9 17.2-5.2 26.2s12.5 14.8 22.2 14.8H456c13.3 0 24-10.7 24-24V312c0-9.7-5.8-18.5-14.8-22.2s-19.3-1.7-26.2 5.2l-40 40-79-79 79-79 40 40c6.9 6.9 17.2 8.9 26.2 5.2s14.8-12.5 14.8-22.2V56c0-13.3-10.7-24-24-24H312c-9.7 0-18.5 5.8-22.2 14.8s-1.7 19.3 5.2 26.2l40 40-79 79-79-79 40-40c6.9-6.9 8.9-17.2 5.2-26.2S209.7 32 200 32z"],"charging-station":[576,512,[],"f5e7","M96 0C60.7 0 32 28.7 32 64V448c-17.7 0-32 14.3-32 32s14.3 32 32 32H320c17.7 0 32-14.3 32-32s-14.3-32-32-32V304h16c22.1 0 40 17.9 40 40v32c0 39.8 32.2 72 72 72s72-32.2 72-72V252.3c32.5-10.2 56-40.5 56-76.3V144c0-8.8-7.2-16-16-16H544V80c0-8.8-7.2-16-16-16s-16 7.2-16 16v48H480V80c0-8.8-7.2-16-16-16s-16 7.2-16 16v48H432c-8.8 0-16 7.2-16 16v32c0 35.8 23.5 66.1 56 76.3V376c0 13.3-10.7 24-24 24s-24-10.7-24-24V344c0-48.6-39.4-88-88-88H320V64c0-35.3-28.7-64-64-64H96zM216.9 82.7c6 4 8.5 11.5 6.3 18.3l-25 74.9H256c6.7 0 12.7 4.2 15 10.4s.5 13.3-4.6 17.7l-112 96c-5.5 4.7-13.4 5.1-19.3 1.1s-8.5-11.5-6.3-18.3l25-74.9H96c-6.7 0-12.7-4.2-15-10.4s-.5-13.3 4.6-17.7l112-96c5.5-4.7 13.4-5.1 19.3-1.1z"],shapes:[512,512,["triangle-circle-square"],"f61f","M315.4 15.5C309.7 5.9 299.2 0 288 0s-21.7 5.9-27.4 15.5l-96 160c-5.9 9.9-6.1 22.2-.4 32.2s16.3 16.2 27.8 16.2H384c11.5 0 22.2-6.2 27.8-16.2s5.5-22.3-.4-32.2l-96-160zM288 312V456c0 22.1 17.9 40 40 40H472c22.1 0 40-17.9 40-40V312c0-22.1-17.9-40-40-40H328c-22.1 0-40 17.9-40 40zM128 512a128 128 0 1 0 0-256 128 128 0 1 0 0 256z"],shuffle:[512,512,[128256,"random"],"f074","M403.8 34.4c12-5 25.7-2.2 34.9 6.9l64 64c6 6 9.4 14.1 9.4 22.6s-3.4 16.6-9.4 22.6l-64 64c-9.2 9.2-22.9 11.9-34.9 6.9s-19.8-16.6-19.8-29.6V160H352c-10.1 0-19.6 4.7-25.6 12.8L284 229.3 244 176l31.2-41.6C293.3 110.2 321.8 96 352 96h32V64c0-12.9 7.8-24.6 19.8-29.6zM164 282.7L204 336l-31.2 41.6C154.7 401.8 126.2 416 96 416H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H96c10.1 0 19.6-4.7 25.6-12.8L164 282.7zm274.6 188c-9.2 9.2-22.9 11.9-34.9 6.9s-19.8-16.6-19.8-29.6V416H352c-30.2 0-58.7-14.2-76.8-38.4L121.6 172.8c-6-8.1-15.5-12.8-25.6-12.8H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H96c30.2 0 58.7 14.2 76.8 38.4L326.4 339.2c6 8.1 15.5 12.8 25.6 12.8h32V320c0-12.9 7.8-24.6 19.8-29.6s25.7-2.2 34.9 6.9l64 64c6 6 9.4 14.1 9.4 22.6s-3.4 16.6-9.4 22.6l-64 64z"],"person-running":[448,512,[127939,"running"],"f70c","M320 48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM125.7 175.5c9.9-9.9 23.4-15.5 37.5-15.5c1.9 0 3.8 .1 5.6 .3L137.6 254c-9.3 28 1.7 58.8 26.8 74.5l86.2 53.9-25.4 88.8c-4.9 17 5 34.7 22 39.6s34.7-5 39.6-22l28.7-100.4c5.9-20.6-2.6-42.6-20.7-53.9L238 299l30.9-82.4 5.1 12.3C289 264.7 323.9 288 362.7 288H384c17.7 0 32-14.3 32-32s-14.3-32-32-32H362.7c-12.9 0-24.6-7.8-29.5-19.7l-6.3-15c-14.6-35.1-44.1-61.9-80.5-73.1l-48.7-15c-11.1-3.4-22.7-5.2-34.4-5.2c-31 0-60.8 12.3-82.7 34.3L57.4 153.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l23.1-23.1zM91.2 352H32c-17.7 0-32 14.3-32 32s14.3 32 32 32h69.6c19 0 36.2-11.2 43.9-28.5L157 361.6l-9.5-6c-17.5-10.9-30.5-26.8-37.9-44.9L91.2 352z"],"mobile-retro":[320,512,[],"e527","M0 64C0 28.7 28.7 0 64 0H256c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zm64 96v64c0 17.7 14.3 32 32 32H224c17.7 0 32-14.3 32-32V160c0-17.7-14.3-32-32-32H96c-17.7 0-32 14.3-32 32zM80 352a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm24 56a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zm56-56a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm24 56a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zm56-56a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm24 56a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zM128 48c-8.8 0-16 7.2-16 16s7.2 16 16 16h64c8.8 0 16-7.2 16-16s-7.2-16-16-16H128z"],"grip-lines-vertical":[192,512,[],"f7a5","M64 64c0-17.7-14.3-32-32-32S0 46.3 0 64V448c0 17.7 14.3 32 32 32s32-14.3 32-32V64zm128 0c0-17.7-14.3-32-32-32s-32 14.3-32 32V448c0 17.7 14.3 32 32 32s32-14.3 32-32V64z"],spider:[512,512,[128375],"f717","M158.4 32.6c4.8-12.4-1.4-26.3-13.8-31s-26.3 1.4-31 13.8L81.1 100c-7.9 20.7-3 44.1 12.7 59.7l57.4 57.4L70.8 190.3c-2.4-.8-4.3-2.7-5.1-5.1L46.8 128.4C42.6 115.8 29 109 16.4 113.2S-3 131 1.2 143.6l18.9 56.8c5.6 16.7 18.7 29.8 35.4 35.4L116.1 256 55.6 276.2c-16.7 5.6-29.8 18.7-35.4 35.4L1.2 368.4C-3 381 3.8 394.6 16.4 398.8s26.2-2.6 30.4-15.2l18.9-56.8c.8-2.4 2.7-4.3 5.1-5.1l80.4-26.8L93.7 352.3C78.1 368 73.1 391.4 81.1 412l32.5 84.6c4.8 12.4 18.6 18.5 31 13.8s18.5-18.6 13.8-31l-32.5-84.6c-1.1-3-.4-6.3 1.8-8.5L160 353.9c1 52.1 43.6 94.1 96 94.1s95-41.9 96-94.1l32.3 32.3c2.2 2.2 2.9 5.6 1.8 8.5l-32.5 84.6c-4.8 12.4 1.4 26.3 13.8 31s26.3-1.4 31-13.8L430.9 412c7.9-20.7 3-44.1-12.7-59.7l-57.4-57.4 80.4 26.8c2.4 .8 4.3 2.7 5.1 5.1l18.9 56.8c4.2 12.6 17.8 19.4 30.4 15.2s19.4-17.8 15.2-30.4l-18.9-56.8c-5.6-16.7-18.7-29.8-35.4-35.4L395.9 256l60.5-20.2c16.7-5.6 29.8-18.7 35.4-35.4l18.9-56.8c4.2-12.6-2.6-26.2-15.2-30.4s-26.2 2.6-30.4 15.2l-18.9 56.8c-.8 2.4-2.7 4.3-5.1 5.1l-80.4 26.8 57.4-57.4c15.6-15.6 20.6-39 12.7-59.7L398.4 15.4C393.6 3 379.8-3.2 367.4 1.6s-18.5 18.6-13.8 31l32.5 84.6c1.1 3 .4 6.3-1.8 8.5L336 174.1V160c0-31.8-18.6-59.3-45.5-72.2c-9.1-4.4-18.5 3.3-18.5 13.4V112c0 8.8-7.2 16-16 16s-16-7.2-16-16V101.2c0-10.1-9.4-17.7-18.5-13.4C194.6 100.7 176 128.2 176 160v14.1l-48.3-48.3c-2.2-2.2-2.9-5.6-1.8-8.5l32.5-84.6z"],"hands-bound":[640,512,[],"e4f9","M96 32C96 14.3 81.7 0 64 0S32 14.3 32 32V96v59.1 .7V192v21.9c0 14.2 5.1 27.9 14.3 38.7L131.6 352H128c-13.3 0-24 10.7-24 24s10.7 24 24 24h32H288h64H480h32c13.3 0 24-10.7 24-24s-10.7-24-24-24h-3.6l85.3-99.5c9.2-10.8 14.3-24.5 14.3-38.7V192 155.8v-.7V96 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V96v48.8l-69.3 92.4c-5.7 7.6-16.1 9.6-24.2 4.8c-9.7-5.7-12.1-18.7-5.1-27.5L473 180c10.8-13.5 8.9-33.3-4.4-44.5s-33-9.8-44.5 3.2l-46.7 52.5C361 209.7 352 233.4 352 258.1V320v32H288V320 258.1c0-24.6-9-48.4-25.4-66.8l-46.7-52.5c-11.5-13-31.3-14.4-44.5-3.2s-15.2 30.9-4.4 44.5l27.6 34.5c7 8.8 4.7 21.8-5.1 27.5c-8.1 4.8-18.6 2.7-24.2-4.8L96 144.8V96 32zm64 448v32H288V480h64v32H480V480h32c13.3 0 24-10.7 24-24s-10.7-24-24-24H480 352 288 160 128c-13.3 0-24 10.7-24 24s10.7 24 24 24h32z"],"file-invoice-dollar":[384,512,[],"f571","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM64 80c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16zm0 64c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16zm128 72c8.8 0 16 7.2 16 16v17.3c8.5 1.2 16.7 3.1 24.1 5.1c8.5 2.3 13.6 11 11.3 19.6s-11 13.6-19.6 11.3c-11.1-3-22-5.2-32.1-5.3c-8.4-.1-17.4 1.8-23.6 5.5c-5.7 3.4-8.1 7.3-8.1 12.8c0 3.7 1.3 6.5 7.3 10.1c6.9 4.1 16.6 7.1 29.2 10.9l.5 .1 0 0 0 0c11.3 3.4 25.3 7.6 36.3 14.6c12.1 7.6 22.4 19.7 22.7 38.2c.3 19.3-9.6 33.3-22.9 41.6c-7.7 4.8-16.4 7.6-25.1 9.1V440c0 8.8-7.2 16-16 16s-16-7.2-16-16V422.2c-11.2-2.1-21.7-5.7-30.9-8.9l0 0 0 0c-2.1-.7-4.2-1.4-6.2-2.1c-8.4-2.8-12.9-11.9-10.1-20.2s11.9-12.9 20.2-10.1c2.5 .8 4.8 1.6 7.1 2.4l0 0 0 0 0 0c13.6 4.6 24.6 8.4 36.3 8.7c9.1 .3 17.9-1.7 23.7-5.3c5.1-3.2 7.9-7.3 7.8-14c-.1-4.6-1.8-7.8-7.7-11.6c-6.8-4.3-16.5-7.4-29-11.2l-1.6-.5 0 0c-11-3.3-24.3-7.3-34.8-13.7c-12-7.2-22.6-18.9-22.7-37.3c-.1-19.4 10.8-32.8 23.8-40.5c7.5-4.4 15.8-7.2 24.1-8.7V232c0-8.8 7.2-16 16-16z"],"plane-circle-exclamation":[640,512,[],"e556","M256 0c-35 0-64 59.5-64 93.7v84.6L8.1 283.4c-5 2.8-8.1 8.2-8.1 13.9v65.5c0 10.6 10.2 18.3 20.4 15.4l171.6-49 0 70.9-57.6 43.2c-4 3-6.4 7.8-6.4 12.8v42c0 7.8 6.3 14 14 14c1.3 0 2.6-.2 3.9-.5L256 480l110.1 31.5c1.3 .4 2.6 .5 3.9 .5c6 0 11.1-3.7 13.1-9C344.5 470.7 320 422.2 320 368c0-60.6 30.6-114 77.1-145.6L320 178.3V93.7C320 59.5 292 0 256 0zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z"],"x-ray":[512,512,[],"f497","M0 64C0 46.3 14.3 32 32 32H480c17.7 0 32 14.3 32 32s-14.3 32-32 32V416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32V96C14.3 96 0 81.7 0 64zM256 96c-8.8 0-16 7.2-16 16v32H160c-8.8 0-16 7.2-16 16s7.2 16 16 16h80v48H128c-8.8 0-16 7.2-16 16s7.2 16 16 16H240v70.6L189.1 307c-5.2-2-10.6-3-16.2-3h-2.1c-23.6 0-42.8 19.2-42.8 42.8c0 9.6 3.2 18.9 9.1 26.4l18.2 23.2c9.7 12.4 24.6 19.6 40.3 19.6H316.4c15.7 0 30.6-7.2 40.3-19.6l18.2-23.2c5.9-7.5 9.1-16.8 9.1-26.4c0-23.6-19.2-42.8-42.8-42.8H339c-5.5 0-11 1-16.2 3L272 326.6V256H384c8.8 0 16-7.2 16-16s-7.2-16-16-16H272V176h80c8.8 0 16-7.2 16-16s-7.2-16-16-16H272V112c0-8.8-7.2-16-16-16zM208 352a16 16 0 1 1 0 32 16 16 0 1 1 0-32zm80 16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0z"],"spell-check":[576,512,[],"f891","M112 0C99.1 0 87.4 7.8 82.5 19.7l-66.7 160-13.3 32c-6.8 16.3 .9 35 17.2 41.8s35-.9 41.8-17.2L66.7 224h90.7l5.1 12.3c6.8 16.3 25.5 24 41.8 17.2s24-25.5 17.2-41.8l-13.3-32-66.7-160C136.6 7.8 124.9 0 112 0zm18.7 160H93.3L112 115.2 130.7 160zM256 32v96 96c0 17.7 14.3 32 32 32h80c44.2 0 80-35.8 80-80c0-23.1-9.8-43.8-25.4-58.4c6-11.2 9.4-24 9.4-37.6c0-44.2-35.8-80-80-80H288c-17.7 0-32 14.3-32 32zm96 64H320V64h32c8.8 0 16 7.2 16 16s-7.2 16-16 16zm-32 64h32 16c8.8 0 16 7.2 16 16s-7.2 16-16 16H320V160zM566.6 310.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L352 434.7l-73.4-73.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l96 96c12.5 12.5 32.8 12.5 45.3 0l192-192z"],slash:[640,512,[],"f715","M5.1 9.2C13.3-1.2 28.4-3.1 38.8 5.1l592 464c10.4 8.2 12.3 23.3 4.1 33.7s-23.3 12.3-33.7 4.1L9.2 42.9C-1.2 34.7-3.1 19.6 5.1 9.2z"],"computer-mouse":[384,512,[128433,"mouse"],"f8cc","M0 192H176V0H160C71.6 0 0 71.6 0 160v32zm0 32V352c0 88.4 71.6 160 160 160h64c88.4 0 160-71.6 160-160V224H192 0zm384-32V160C384 71.6 312.4 0 224 0H208V192H384z"],"arrow-right-to-bracket":[512,512,["sign-in"],"f090","M352 96l64 0c17.7 0 32 14.3 32 32l0 256c0 17.7-14.3 32-32 32l-64 0c-17.7 0-32 14.3-32 32s14.3 32 32 32l64 0c53 0 96-43 96-96l0-256c0-53-43-96-96-96l-64 0c-17.7 0-32 14.3-32 32s14.3 32 32 32zm-9.4 182.6c12.5-12.5 12.5-32.8 0-45.3l-128-128c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L242.7 224 32 224c-17.7 0-32 14.3-32 32s14.3 32 32 32l210.7 0-73.4 73.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l128-128z"],"shop-slash":[640,512,["store-alt-slash"],"e070","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-54.8-43V224H512V376L384 275.7V224H320v1.5L277.2 192H603.2c20.3 0 36.8-16.5 36.8-36.8c0-7.3-2.2-14.4-6.2-20.4L558.2 21.4C549.3 8 534.4 0 518.3 0H121.7c-16 0-31 8-39.9 21.4L74.1 32.8 38.8 5.1zM36.8 192h85L21 112.5 6.2 134.7c-4 6.1-6.2 13.2-6.2 20.4C0 175.5 16.5 192 36.8 192zM320 384H128V224H64V384v80c0 26.5 21.5 48 48 48H336c26.5 0 48-21.5 48-48V398.5l-64-50.4V384zM544 512l-.3 0h.6l-.3 0z"],server:[512,512,[],"f233","M64 32C28.7 32 0 60.7 0 96v64c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm280 72a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm48 24a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zM64 288c-35.3 0-64 28.7-64 64v64c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V352c0-35.3-28.7-64-64-64H64zm280 72a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm56 24a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z"],"virus-covid-slash":[640,512,[],"e4a9","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L472.1 344.7c11.4-19.5 19.1-41.4 22.3-64.7H528v16c0 13.3 10.7 24 24 24s24-10.7 24-24V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v16H494.4c-4.2-30.7-16.3-58.8-34.1-82.3L484 125.9l11.3 11.3c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L472.7 46.7c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9L450.1 92l-23.8 23.8C402.8 97.9 374.7 85.8 344 81.6V48h16c13.3 0 24-10.7 24-24s-10.7-24-24-24H280c-13.3 0-24 10.7-24 24s10.7 24 24 24h16V81.6c-30.7 4.2-58.8 16.3-82.3 34.1L189.9 92l11.3-11.3c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L134.1 79.8 38.8 5.1zm362.5 407L149.2 213.5c-1.5 6-2.7 12.2-3.5 18.5H112V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v80c0 13.3 10.7 24 24 24s24-10.7 24-24V280h33.6c4.2 30.7 16.3 58.8 34.1 82.3L156 386.1l-11.3-11.3c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l56.6 56.6c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L189.9 420l23.8-23.8c23.5 17.9 51.7 29.9 82.3 34.1V464H280c-13.3 0-24 10.7-24 24s10.7 24 24 24h80c13.3 0 24-10.7 24-24s-10.7-24-24-24H344V430.4c20.4-2.8 39.7-9.1 57.3-18.2z"],"shop-lock":[640,512,[],"e4a5","M36.8 192H449.6c20.2-19.8 47.9-32 78.4-32c30.5 0 58.1 12.2 78.3 31.9c18.9-1.6 33.7-17.4 33.7-36.7c0-7.3-2.2-14.4-6.2-20.4L558.2 21.4C549.3 8 534.4 0 518.3 0H121.7c-16 0-31 8-39.9 21.4L6.2 134.7c-4 6.1-6.2 13.2-6.2 20.4C0 175.5 16.5 192 36.8 192zM384 224H320V384H128V224H64V384v80c0 26.5 21.5 48 48 48H336c26.5 0 48-21.5 48-48V384 352 224zm144 16c17.7 0 32 14.3 32 32v48H496V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z"],"hourglass-start":[384,512,["hourglass-1"],"f251","M32 0C14.3 0 0 14.3 0 32S14.3 64 32 64V75c0 42.4 16.9 83.1 46.9 113.1L146.7 256 78.9 323.9C48.9 353.9 32 394.6 32 437v11c-17.7 0-32 14.3-32 32s14.3 32 32 32H64 320h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V437c0-42.4-16.9-83.1-46.9-113.1L237.3 256l67.9-67.9c30-30 46.9-70.7 46.9-113.1V64c17.7 0 32-14.3 32-32s-14.3-32-32-32H320 64 32zM288 437v11H96V437c0-25.5 10.1-49.9 28.1-67.9L192 301.3l67.9 67.9c18 18 28.1 42.4 28.1 67.9z"],"blender-phone":[576,512,[],"f6b6","M224 352L196.8 52.3C194.2 24.2 216.3 0 244.6 0H534.1c21.1 0 36.4 20.1 30.9 40.4L558.5 64H400c-8.8 0-16 7.2-16 16s7.2 16 16 16H549.8l-17.5 64H400c-8.8 0-16 7.2-16 16s7.2 16 16 16H523.6l-17.5 64H400c-8.8 0-16 7.2-16 16s7.2 16 16 16h97.5L480 352H224zm-16 32H496c26.5 0 48 21.5 48 48v32c0 26.5-21.5 48-48 48H208c-26.5 0-48-21.5-48-48V432c0-26.5 21.5-48 48-48zm144 96a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM147.5 30.7c10.8 6.7 15.3 21 10.6 33.4l-22 57.8c-4.2 10.9-14.5 17.6-25.3 16.4l-33.3-3.6c-13.6 42.2-13.6 88.4 0 130.7l33.3-3.6c10.9-1.2 21.2 5.5 25.3 16.4l22 57.8c4.7 12.4 .2 26.7-10.6 33.4l-44 27.2c-9.7 6-21.9 4.2-29.8-4.3C-24.6 286-24.6 114 73.7 7.8C81.6-.7 93.8-2.5 103.5 3.5l44 27.2z"],"building-wheat":[640,512,[],"e4db","M0 48C0 21.5 21.5 0 48 0H336c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H240V432c0-26.5-21.5-48-48-48s-48 21.5-48 48v80H48c-26.5 0-48-21.5-48-48V48zM80 224c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H80zm80 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H176c-8.8 0-16 7.2-16 16zm112-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H272zM64 112v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H80c-8.8 0-16 7.2-16 16zM176 96c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H176zm80 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16zm384 80v16c0 44.2-35.8 80-80 80H544V272c0-44.2 35.8-80 80-80h16zm0 128c0 44.2-35.8 80-80 80H544V384c0-44.2 35.8-80 80-80h16v16zm0 112c0 44.2-35.8 80-80 80H544V496c0-44.2 35.8-80 80-80h16v16zM512 496v16H496c-44.2 0-80-35.8-80-80V416h16c44.2 0 80 35.8 80 80zm0-96H496c-44.2 0-80-35.8-80-80V304h16c44.2 0 80 35.8 80 80v16zm0-128v16H496c-44.2 0-80-35.8-80-80V192h16c44.2 0 80 35.8 80 80zM528 32c13.3 0 24 10.7 24 24V160c0 13.3-10.7 24-24 24s-24-10.7-24-24V56c0-13.3 10.7-24 24-24zm96 64v32c0 13.3-10.7 24-24 24s-24-10.7-24-24V96c0-13.3 10.7-24 24-24s24 10.7 24 24zM456 72c13.3 0 24 10.7 24 24v32c0 13.3-10.7 24-24 24s-24-10.7-24-24V96c0-13.3 10.7-24 24-24z"],"person-breastfeeding":[448,512,[],"e53a","M224 0a80 80 0 1 1 0 160A80 80 0 1 1 224 0zM436.8 382.8L373.5 462c-16.6 20.7-46.8 24.1-67.5 7.5c-17.6-14.1-22.7-38.1-13.5-57.7l-.8-.1c-38.9-5.6-74.3-25.1-99.7-54.8V320c0-17.7-14.3-32-32-32s-32 14.3-32 32v48c0 .8 0 1.6 .1 2.4l101.4 50.7c23.7 11.9 33.3 40.7 21.5 64.4s-40.7 33.3-64.4 21.5L27.2 427.3c-1.1-.5-2.2-1.1-3.3-1.7c-4.9-2.8-9.2-6.4-12.6-10.6c-4.6-5.4-7.8-11.7-9.6-18.4c-3.3-12-1.9-25.2 4.8-36.6c.6-1.1 1.3-2.2 2-3.2L75.6 256.1c26.7-40.1 71.7-64.1 119.8-64.1h75.2c46.5 0 90.1 22.5 117.2 60.3l50.7 70.9c2.2 3 4 6.1 5.5 9.4c2.9 6.7 4.3 13.8 4 20.8c-.3 10.6-4.2 21-11.2 29.4zM320 332a44 44 0 1 0 -88 0 44 44 0 1 0 88 0z"],"right-to-bracket":[512,512,["sign-in-alt"],"f2f6","M217.9 105.9L340.7 228.7c7.2 7.2 11.3 17.1 11.3 27.3s-4.1 20.1-11.3 27.3L217.9 406.1c-6.4 6.4-15 9.9-24 9.9c-18.7 0-33.9-15.2-33.9-33.9l0-62.1L32 320c-17.7 0-32-14.3-32-32l0-64c0-17.7 14.3-32 32-32l128 0 0-62.1c0-18.7 15.2-33.9 33.9-33.9c9 0 17.6 3.6 24 9.9zM352 416l64 0c17.7 0 32-14.3 32-32l0-256c0-17.7-14.3-32-32-32l-64 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l64 0c53 0 96 43 96 96l0 256c0 53-43 96-96 96l-64 0c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],venus:[384,512,[9792],"f221","M80 176a112 112 0 1 1 224 0A112 112 0 1 1 80 176zM224 349.1c81.9-15 144-86.8 144-173.1C368 78.8 289.2 0 192 0S16 78.8 16 176c0 86.3 62.1 158.1 144 173.1V384H128c-17.7 0-32 14.3-32 32s14.3 32 32 32h32v32c0 17.7 14.3 32 32 32s32-14.3 32-32V448h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H224V349.1z"],passport:[448,512,[],"f5ab","M0 64C0 28.7 28.7 0 64 0H384c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM183 278.8c-27.9-13.2-48.4-39.4-53.7-70.8h39.1c1.6 30.4 7.7 53.8 14.6 70.8zm41.3 9.2l-.3 0-.3 0c-2.4-3.5-5.7-8.9-9.1-16.5c-6-13.6-12.4-34.3-14.2-63.5h47.1c-1.8 29.2-8.1 49.9-14.2 63.5c-3.4 7.6-6.7 13-9.1 16.5zm40.7-9.2c6.8-17.1 12.9-40.4 14.6-70.8h39.1c-5.3 31.4-25.8 57.6-53.7 70.8zM279.6 176c-1.6-30.4-7.7-53.8-14.6-70.8c27.9 13.2 48.4 39.4 53.7 70.8H279.6zM223.7 96l.3 0 .3 0c2.4 3.5 5.7 8.9 9.1 16.5c6 13.6 12.4 34.3 14.2 63.5H200.5c1.8-29.2 8.1-49.9 14.2-63.5c3.4-7.6 6.7-13 9.1-16.5zM183 105.2c-6.8 17.1-12.9 40.4-14.6 70.8H129.3c5.3-31.4 25.8-57.6 53.7-70.8zM352 192A128 128 0 1 0 96 192a128 128 0 1 0 256 0zM112 384c-8.8 0-16 7.2-16 16s7.2 16 16 16H336c8.8 0 16-7.2 16-16s-7.2-16-16-16H112z"],"heart-pulse":[512,512,["heartbeat"],"f21e","M228.3 469.1L47.6 300.4c-4.2-3.9-8.2-8.1-11.9-12.4h87c22.6 0 43-13.6 51.7-34.5l10.5-25.2 49.3 109.5c3.8 8.5 12.1 14 21.4 14.1s17.8-5 22-13.3L320 253.7l1.7 3.4c9.5 19 28.9 31 50.1 31H476.3c-3.7 4.3-7.7 8.5-11.9 12.4L283.7 469.1c-7.5 7-17.4 10.9-27.7 10.9s-20.2-3.9-27.7-10.9zM503.7 240h-132c-3 0-5.8-1.7-7.2-4.4l-23.2-46.3c-4.1-8.1-12.4-13.3-21.5-13.3s-17.4 5.1-21.5 13.3l-41.4 82.8L205.9 158.2c-3.9-8.7-12.7-14.3-22.2-14.1s-18.1 5.9-21.8 14.8l-31.8 76.3c-1.2 3-4.2 4.9-7.4 4.9H16c-2.6 0-5 .4-7.3 1.1C3 225.2 0 208.2 0 190.9v-5.8c0-69.9 50.5-129.5 119.4-141C165 36.5 211.4 51.4 244 84l12 12 12-12c32.6-32.6 79-47.5 124.6-39.9C461.5 55.6 512 115.2 512 185.1v5.8c0 16.9-2.8 33.5-8.3 49.1z"],"people-carry-box":[640,512,["people-carry"],"f4ce","M80 48a48 48 0 1 1 96 0A48 48 0 1 1 80 48zm64 193.7v65.1l51 51c7.1 7.1 11.8 16.2 13.4 26.1l15.2 90.9c2.9 17.4-8.9 33.9-26.3 36.8s-33.9-8.9-36.8-26.3l-14.3-85.9L66.8 320C54.8 308 48 291.7 48 274.7V186.6c0-32.4 26.2-58.6 58.6-58.6c24.1 0 46.5 12 59.9 32l47.4 71.1 10.1 5V160c0-17.7 14.3-32 32-32H384c17.7 0 32 14.3 32 32v76.2l10.1-5L473.5 160c13.3-20 35.8-32 59.9-32c32.4 0 58.6 26.2 58.6 58.6v88.1c0 17-6.7 33.3-18.7 45.3l-79.4 79.4-14.3 85.9c-2.9 17.4-19.4 29.2-36.8 26.3s-29.2-19.4-26.3-36.8l15.2-90.9c1.6-9.9 6.3-19 13.4-26.1l51-51V241.7l-19 28.5c-4.6 7-11 12.6-18.5 16.3l-59.6 29.8c-2.4 1.3-4.9 2.2-7.6 2.8c-2.6 .6-5.3 .9-7.9 .8H256.7c-2.5 .1-5-.2-7.5-.7c-2.9-.6-5.6-1.6-8.1-3l-59.5-29.8c-7.5-3.7-13.8-9.4-18.5-16.3l-19-28.5zM2.3 468.1L50.1 348.6l49.2 49.2-37.6 94c-6.6 16.4-25.2 24.4-41.6 17.8S-4.3 484.5 2.3 468.1zM512 0a48 48 0 1 1 0 96 48 48 0 1 1 0-96zm77.9 348.6l47.8 119.5c6.6 16.4-1.4 35-17.8 41.6s-35-1.4-41.6-17.8l-37.6-94 49.2-49.2z"],"temperature-high":[512,512,[],"f769","M416 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm0 128A96 96 0 1 0 416 0a96 96 0 1 0 0 192zM96 112c0-26.5 21.5-48 48-48s48 21.5 48 48V276.5c0 17.3 7.1 31.9 15.3 42.5C217.8 332.6 224 349.5 224 368c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-18.5 6.2-35.4 16.7-48.9C88.9 308.4 96 293.8 96 276.5V112zM144 0C82.1 0 32 50.2 32 112V276.5c0 .1-.1 .3-.2 .6c-.2 .6-.8 1.6-1.7 2.8C11.2 304.2 0 334.8 0 368c0 79.5 64.5 144 144 144s144-64.5 144-144c0-33.2-11.2-63.8-30.1-88.1c-.9-1.2-1.5-2.2-1.7-2.8c-.1-.3-.2-.5-.2-.6V112C256 50.2 205.9 0 144 0zm0 416c26.5 0 48-21.5 48-48c0-20.9-13.4-38.7-32-45.3V112c0-8.8-7.2-16-16-16s-16 7.2-16 16V322.7c-18.6 6.6-32 24.4-32 45.3c0 26.5 21.5 48 48 48z"],microchip:[512,512,[],"f2db","M176 24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64c-35.3 0-64 28.7-64 64H24c-13.3 0-24 10.7-24 24s10.7 24 24 24H64v56H24c-13.3 0-24 10.7-24 24s10.7 24 24 24H64v56H24c-13.3 0-24 10.7-24 24s10.7 24 24 24H64c0 35.3 28.7 64 64 64v40c0 13.3 10.7 24 24 24s24-10.7 24-24V448h56v40c0 13.3 10.7 24 24 24s24-10.7 24-24V448h56v40c0 13.3 10.7 24 24 24s24-10.7 24-24V448c35.3 0 64-28.7 64-64h40c13.3 0 24-10.7 24-24s-10.7-24-24-24H448V280h40c13.3 0 24-10.7 24-24s-10.7-24-24-24H448V176h40c13.3 0 24-10.7 24-24s-10.7-24-24-24H448c0-35.3-28.7-64-64-64V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H280V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H176V24zM160 128H352c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32V160c0-17.7 14.3-32 32-32zm192 32H160V352H352V160z"],crown:[576,512,[128081],"f521","M309 106c11.4-7 19-19.7 19-34c0-22.1-17.9-40-40-40s-40 17.9-40 40c0 14.4 7.6 27 19 34L209.7 220.6c-9.1 18.2-32.7 23.4-48.6 10.7L72 160c5-6.7 8-15 8-24c0-22.1-17.9-40-40-40S0 113.9 0 136s17.9 40 40 40c.2 0 .5 0 .7 0L86.4 427.4c5.5 30.4 32 52.6 63 52.6H426.6c30.9 0 57.4-22.1 63-52.6L535.3 176c.2 0 .5 0 .7 0c22.1 0 40-17.9 40-40s-17.9-40-40-40s-40 17.9-40 40c0 9 3 17.3 8 24l-89.1 71.3c-15.9 12.7-39.5 7.5-48.6-10.7L309 106z"],"weight-hanging":[512,512,[],"f5cd","M224 96a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm122.5 32c3.5-10 5.5-20.8 5.5-32c0-53-43-96-96-96s-96 43-96 96c0 11.2 1.9 22 5.5 32H120c-22 0-41.2 15-46.6 36.4l-72 288c-3.6 14.3-.4 29.5 8.7 41.2S33.2 512 48 512H464c14.8 0 28.7-6.8 37.8-18.5s12.3-26.8 8.7-41.2l-72-288C433.2 143 414 128 392 128H346.5z"],"xmarks-lines":[640,512,[],"e59a","M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H32zm0 384c-17.7 0-32 14.3-32 32s14.3 32 32 32H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H32zM7 167c-9.4 9.4-9.4 24.6 0 33.9l55 55L7 311c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l55-55 55 55c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-55-55 55-55c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-55 55L41 167c-9.4-9.4-24.6-9.4-33.9 0zM265 167c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l55 55-55 55c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l55-55 55 55c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-55-55 55-55c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-55 55-55-55zM455 167c-9.4 9.4-9.4 24.6 0 33.9l55 55-55 55c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l55-55 55 55c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-55-55 55-55c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-55 55-55-55c-9.4-9.4-24.6-9.4-33.9 0z"],"file-prescription":[384,512,[],"f572","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM104 196h72c33.1 0 60 26.9 60 60c0 25.5-15.9 47.2-38.3 55.9l43 40.3 33.8-31c8.1-7.5 20.8-6.9 28.3 1.2s6.9 20.8-1.2 28.3L270 379.7l31.7 29.7c8.1 7.6 8.5 20.2 .9 28.3s-20.2 8.5-28.3 .9l-33.9-31.8-34.9 32c-8.1 7.5-20.8 6.9-28.3-1.2s-6.9-20.8 1.2-28.3l32.6-29.9-64.8-60.8c-.9-.8-1.6-1.7-2.3-2.6H124v44c0 11-9 20-20 20s-20-9-20-20V296 216c0-11 9-20 20-20zm72 80c11 0 20-9 20-20s-9-20-20-20H124v40h52z"],"weight-scale":[512,512,["weight"],"f496","M128 176a128 128 0 1 1 256 0 128 128 0 1 1 -256 0zM391.8 64C359.5 24.9 310.7 0 256 0S152.5 24.9 120.2 64H64C28.7 64 0 92.7 0 128V448c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H391.8zM296 224c0-10.6-4.1-20.2-10.9-27.4l33.6-78.3c3.5-8.1-.3-17.5-8.4-21s-17.5 .3-21 8.4L255.7 184c-22 .1-39.7 18-39.7 40c0 22.1 17.9 40 40 40s40-17.9 40-40z"],"user-group":[640,512,[128101,"user-friends"],"f500","M96 128a128 128 0 1 1 256 0A128 128 0 1 1 96 128zM0 482.3C0 383.8 79.8 304 178.3 304h91.4C368.2 304 448 383.8 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3zM609.3 512H471.4c5.4-9.4 8.6-20.3 8.6-32v-8c0-60.7-27.1-115.2-69.8-151.8c2.4-.1 4.7-.2 7.1-.2h61.4C567.8 320 640 392.2 640 481.3c0 17-13.8 30.7-30.7 30.7zM432 256c-31 0-59-12.6-79.3-32.9C372.4 196.5 384 163.6 384 128c0-26.8-6.6-52.1-18.3-74.3C384.3 40.1 407.2 32 432 32c61.9 0 112 50.1 112 112s-50.1 112-112 112z"],"arrow-up-a-z":[576,512,["sort-alpha-up"],"f15e","M183.6 42.4C177.5 35.8 169 32 160 32s-17.5 3.8-23.6 10.4l-88 96c-11.9 13-11.1 33.3 2 45.2s33.3 11.1 45.2-2L128 146.3V448c0 17.7 14.3 32 32 32s32-14.3 32-32V146.3l32.4 35.4c11.9 13 32.2 13.9 45.2 2s13.9-32.2 2-45.2l-88-96zM320 320c0 17.7 14.3 32 32 32h50.7l-73.4 73.4c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H429.3l73.4-73.4c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8H352c-17.7 0-32 14.3-32 32zM416 32c-12.1 0-23.2 6.8-28.6 17.7l-64 128-16 32c-7.9 15.8-1.5 35 14.3 42.9s35 1.5 42.9-14.3l7.2-14.3h88.4l7.2 14.3c7.9 15.8 27.1 22.2 42.9 14.3s22.2-27.1 14.3-42.9l-16-32-64-128C439.2 38.8 428.1 32 416 32zM395.8 176L416 135.6 436.2 176H395.8z"],"chess-knight":[448,512,[9822],"f441","M96 48L82.7 61.3C70.7 73.3 64 89.5 64 106.5V238.9c0 10.7 5.3 20.7 14.2 26.6l10.6 7c14.3 9.6 32.7 10.7 48.1 3l3.2-1.6c2.6-1.3 5-2.8 7.3-4.5l49.4-37c6.6-5 15.7-5 22.3 0c10.2 7.7 9.9 23.1-.7 30.3L90.4 350C73.9 361.3 64 380 64 400H384l28.9-159c2.1-11.3 3.1-22.8 3.1-34.3V192C416 86 330 0 224 0H83.8C72.9 0 64 8.9 64 19.8c0 7.5 4.2 14.3 10.9 17.7L96 48zm24 68a20 20 0 1 1 40 0 20 20 0 1 1 -40 0zM22.6 473.4c-4.2 4.2-6.6 10-6.6 16C16 501.9 26.1 512 38.6 512H409.4c12.5 0 22.6-10.1 22.6-22.6c0-6-2.4-11.8-6.6-16L384 432H64L22.6 473.4z"],"face-laugh-squint":[512,512,["laugh-squint"],"f59b","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM96.8 314.1c-3.8-13.7 7.4-26.1 21.6-26.1H393.6c14.2 0 25.5 12.4 21.6 26.1C396.2 382 332.1 432 256 432s-140.2-50-159.2-117.9zm36.7-199.4l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6c0-9 9.6-14.7 17.5-10.5zM396 125.1c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9c7.9-4.2 17.5 1.5 17.5 10.5z"],wheelchair:[512,512,[],"f193","M192 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM120.5 247.2c12.4-4.7 18.7-18.5 14-30.9s-18.5-18.7-30.9-14C43.1 225.1 0 283.5 0 352c0 88.4 71.6 160 160 160c61.2 0 114.3-34.3 141.2-84.7c6.2-11.7 1.8-26.2-9.9-32.5s-26.2-1.8-32.5 9.9C240 440 202.8 464 160 464C98.1 464 48 413.9 48 352c0-47.9 30.1-88.8 72.5-104.8zM259.8 176l-1.9-9.7c-4.5-22.3-24-38.3-46.8-38.3c-30.1 0-52.7 27.5-46.8 57l23.1 115.5c6 29.9 32.2 51.4 62.8 51.4h5.1c.4 0 .8 0 1.3 0h94.1c6.7 0 12.6 4.1 15 10.4L402 459.2c6 16.1 23.8 24.6 40.1 19.1l48-16c16.8-5.6 25.8-23.7 20.2-40.5s-23.7-25.8-40.5-20.2l-18.7 6.2-25.5-68c-11.7-31.2-41.6-51.9-74.9-51.9H282.2l-9.6-48H336c17.7 0 32-14.3 32-32s-14.3-32-32-32H259.8z"],"circle-arrow-up":[512,512,["arrow-circle-up"],"f0aa","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM385 231c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-71-71V376c0 13.3-10.7 24-24 24s-24-10.7-24-24V193.9l-71 71c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9L239 119c9.4-9.4 24.6-9.4 33.9 0L385 231z"],"toggle-on":[576,512,[],"f205","M192 64C86 64 0 150 0 256S86 448 192 448H384c106 0 192-86 192-192s-86-192-192-192H192zm192 96a96 96 0 1 1 0 192 96 96 0 1 1 0-192z"],"person-walking":[320,512,[128694,"walking"],"f554","M160 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM126.5 199.3c-1 .4-1.9 .8-2.9 1.2l-8 3.5c-16.4 7.3-29 21.2-34.7 38.2l-2.6 7.8c-5.6 16.8-23.7 25.8-40.5 20.2s-25.8-23.7-20.2-40.5l2.6-7.8c11.4-34.1 36.6-61.9 69.4-76.5l8-3.5c20.8-9.2 43.3-14 66.1-14c44.6 0 84.8 26.8 101.9 67.9L281 232.7l21.4 10.7c15.8 7.9 22.2 27.1 14.3 42.9s-27.1 22.2-42.9 14.3L247 287.3c-10.3-5.2-18.4-13.8-22.8-24.5l-9.6-23-19.3 65.5 49.5 54c5.4 5.9 9.2 13 11.2 20.8l23 92.1c4.3 17.1-6.1 34.5-23.3 38.8s-34.5-6.1-38.8-23.3l-22-88.1-70.7-77.1c-14.8-16.1-20.3-38.6-14.7-59.7l16.9-63.5zM68.7 398l25-62.4c2.1 3 4.5 5.8 7 8.6l40.7 44.4-14.5 36.2c-2.4 6-6 11.5-10.6 16.1L54.6 502.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L68.7 398z"],l:[320,512,[108],"4c","M64 32c17.7 0 32 14.3 32 32V416H288c17.7 0 32 14.3 32 32s-14.3 32-32 32H64c-17.7 0-32-14.3-32-32V64c0-17.7 14.3-32 32-32z"],fire:[448,512,[128293],"f06d","M159.3 5.4c7.8-7.3 19.9-7.2 27.7 .1c27.6 25.9 53.5 53.8 77.7 84c11-14.4 23.5-30.1 37-42.9c7.9-7.4 20.1-7.4 28 .1c34.6 33 63.9 76.6 84.5 118c20.3 40.8 33.8 82.5 33.8 111.9C448 404.2 348.2 512 224 512C98.4 512 0 404.1 0 276.5c0-38.4 17.8-85.3 45.4-131.7C73.3 97.7 112.7 48.6 159.3 5.4zM225.7 416c25.3 0 47.7-7 68.8-21c42.1-29.4 53.4-88.2 28.1-134.4c-4.5-9-16-9.6-22.5-2l-25.2 29.3c-6.6 7.6-18.5 7.4-24.7-.5c-16.5-21-46-58.5-62.8-79.8c-6.3-8-18.3-8.1-24.7-.1c-33.8 42.5-50.8 69.3-50.8 99.4C112 375.4 162.6 416 225.7 416z"],"bed-pulse":[640,512,["procedures"],"f487","M483.2 9.6L524 64h92c13.3 0 24 10.7 24 24s-10.7 24-24 24H512c-7.6 0-14.7-3.6-19.2-9.6L468.7 70.3l-47 99.9c-3.7 7.8-11.3 13.1-19.9 13.7s-16.9-3.4-21.7-10.6L339.2 112H216c-13.3 0-24-10.7-24-24s10.7-24 24-24H352c8 0 15.5 4 20 10.7l24.4 36.6 45.9-97.5C445.9 6.2 453.2 1 461.6 .1s16.6 2.7 21.6 9.5zM320 160h12.7l20.7 31.1c11.2 16.8 30.6 26.3 50.7 24.8s37.9-13.7 46.5-32L461.9 160H544c53 0 96 43 96 96V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H352 320 64v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V96C0 78.3 14.3 64 32 64s32 14.3 32 32V352H288V192c0-17.7 14.3-32 32-32zm-144 0a80 80 0 1 1 0 160 80 80 0 1 1 0-160z"],"shuttle-space":[640,512,["space-shuttle"],"f197","M130 480c40.6 0 80.4-11 115.2-31.9L352 384l-224 0 0 96h2zM352 128L245.2 63.9C210.4 43 170.6 32 130 32h-2v96l224 0zM96 128l0-96H80C53.5 32 32 53.5 32 80v48h8c-22.1 0-40 17.9-40 40v16V328v16c0 22.1 17.9 40 40 40H32v48c0 26.5 21.5 48 48 48H96l0-96h8c26.2 0 49.4-12.6 64-32H456c69.3 0 135-22.7 179.2-81.6c6.4-8.5 6.4-20.3 0-28.8C591 182.7 525.3 160 456 160H168c-14.6-19.4-37.8-32-64-32l-8 0zM512 243.6v24.9c0 19.6-15.9 35.6-35.6 35.6c-2.5 0-4.4-2-4.4-4.4V212.4c0-2.5 2-4.4 4.4-4.4c19.6 0 35.6 15.9 35.6 35.6z"],"face-laugh":[512,512,["laugh"],"f599","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM96.8 314.1c-3.8-13.7 7.4-26.1 21.6-26.1H393.6c14.2 0 25.5 12.4 21.6 26.1C396.2 382 332.1 432 256 432s-140.2-50-159.2-117.9zM144.4 192a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"folder-open":[576,512,[128194,128449,61717],"f07c","M88.7 223.8L0 375.8V96C0 60.7 28.7 32 64 32H181.5c17 0 33.3 6.7 45.3 18.7l26.5 26.5c12 12 28.3 18.7 45.3 18.7H416c35.3 0 64 28.7 64 64v32H144c-22.8 0-43.8 12.1-55.3 31.8zm27.6 16.1C122.1 230 132.6 224 144 224H544c11.5 0 22 6.1 27.7 16.1s5.7 22.2-.1 32.1l-112 192C453.9 474 443.4 480 432 480H32c-11.5 0-22-6.1-27.7-16.1s-5.7-22.2 .1-32.1l112-192z"],"heart-circle-plus":[576,512,[],"e500","M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9l2.6-2.4C267.2 438.6 256 404.6 256 368c0-97.2 78.8-176 176-176c28.3 0 55 6.7 78.7 18.5c.9-6.5 1.3-13 1.3-19.6v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm16-208v48h48c8.8 0 16 7.2 16 16s-7.2 16-16 16H448v48c0 8.8-7.2 16-16 16s-16-7.2-16-16V384H368c-8.8 0-16-7.2-16-16s7.2-16 16-16h48V304c0-8.8 7.2-16 16-16s16 7.2 16 16z"],"code-fork":[448,512,[],"e13b","M80 104a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm80-24c0 32.8-19.7 61-48 73.3V192c0 17.7 14.3 32 32 32H304c17.7 0 32-14.3 32-32V153.3C307.7 141 288 112.8 288 80c0-44.2 35.8-80 80-80s80 35.8 80 80c0 32.8-19.7 61-48 73.3V192c0 53-43 96-96 96H256v70.7c28.3 12.3 48 40.5 48 73.3c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-32.8 19.7-61 48-73.3V288H144c-53 0-96-43-96-96V153.3C19.7 141 0 112.8 0 80C0 35.8 35.8 0 80 0s80 35.8 80 80zm208 24a24 24 0 1 0 0-48 24 24 0 1 0 0 48zM248 432a24 24 0 1 0 -48 0 24 24 0 1 0 48 0z"],city:[640,512,[127961],"f64f","M480 48c0-26.5-21.5-48-48-48H336c-26.5 0-48 21.5-48 48V96H224V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V96H112V24c0-13.3-10.7-24-24-24S64 10.7 64 24V96H48C21.5 96 0 117.5 0 144v96V464c0 26.5 21.5 48 48 48H304h32 96H592c26.5 0 48-21.5 48-48V240c0-26.5-21.5-48-48-48H480V48zm96 320v32c0 8.8-7.2 16-16 16H528c-8.8 0-16-7.2-16-16V368c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16zM240 416H208c-8.8 0-16-7.2-16-16V368c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16zM128 400c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V368c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32zM560 256c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H528c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h32zM256 176v32c0 8.8-7.2 16-16 16H208c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16zM112 160c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16h32zM256 304c0 8.8-7.2 16-16 16H208c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32zM112 320H80c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16zm304-48v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16zM400 64c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V80c0-8.8 7.2-16 16-16h32zm16 112v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16z"],"microphone-lines":[384,512,[127897,"microphone-alt"],"f3c9","M96 96V256c0 53 43 96 96 96s96-43 96-96H208c-8.8 0-16-7.2-16-16s7.2-16 16-16h80V192H208c-8.8 0-16-7.2-16-16s7.2-16 16-16h80V128H208c-8.8 0-16-7.2-16-16s7.2-16 16-16h80c0-53-43-96-96-96S96 43 96 96zM320 240v16c0 70.7-57.3 128-128 128s-128-57.3-128-128V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v40c0 89.1 66.2 162.7 152 174.4V464H120c-13.3 0-24 10.7-24 24s10.7 24 24 24h72 72c13.3 0 24-10.7 24-24s-10.7-24-24-24H216V430.4c85.8-11.7 152-85.3 152-174.4V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v24z"],"pepper-hot":[512,512,[127798],"f816","M428.3 3c11.6-6.4 26.2-2.3 32.6 9.3l4.8 8.7c19.3 34.7 19.8 75.7 3.4 110C495.8 159.6 512 197.9 512 240c0 18.5-3.1 36.3-8.9 52.8c-6.1 17.3-28.5 16.3-36.8-.1l-11.7-23.4c-4.1-8.1-12.4-13.3-21.5-13.3H360c-13.3 0-24-10.7-24-24V152c0-13.3-10.7-24-24-24l-17.1 0c-21.3 0-30-23.9-10.8-32.9C304.7 85.4 327.7 80 352 80c28.3 0 54.8 7.3 77.8 20.2c5.5-18.2 3.7-38.4-6-55.8L419 35.7c-6.4-11.6-2.3-26.2 9.3-32.6zM171.2 345.5L264 160l40 0v80c0 26.5 21.5 48 48 48h76.2l23.9 47.8C372.3 443.9 244.3 512 103.2 512H44.4C19.9 512 0 492.1 0 467.6c0-20.8 14.5-38.8 34.8-43.3l49.8-11.1c37.6-8.4 69.5-33.2 86.7-67.7z"],unlock:[448,512,[128275],"f09c","M144 144c0-44.2 35.8-80 80-80c31.9 0 59.4 18.6 72.3 45.7c7.6 16 26.7 22.8 42.6 15.2s22.8-26.7 15.2-42.6C331 33.7 281.5 0 224 0C144.5 0 80 64.5 80 144v48H64c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V256c0-35.3-28.7-64-64-64H144V144z"],"colon-sign":[384,512,[],"e140","M255 39.8c4.3-17.1-6.1-34.5-23.3-38.8S197.2 7.1 193 24.2L181.9 68.6C96.1 87.8 32 164.4 32 256c0 58.1 25.8 110.2 66.7 145.4L81 472.2c-4.3 17.1 6.1 34.5 23.3 38.8s34.5-6.1 38.8-23.3l13-52.1c9 3.4 18.4 6.2 28 8.2L177 472.2c-4.3 17.1 6.1 34.5 23.3 38.8s34.5-6.1 38.8-23.3l10.4-41.4c33.4-4.4 64.1-17.4 89.8-36.7c14.1-10.6 17-30.7 6.4-44.8s-30.7-17-44.8-6.4c-10.2 7.7-21.7 13.9-34 18.3L321 160c9.4-.3 18.5-4.7 24.6-12.8c10.6-14.1 7.8-34.2-6.4-44.8c-1.1-.8-2.2-1.6-3.3-2.4L351 39.8c4.3-17.1-6.1-34.5-23.3-38.8S293.2 7.1 289 24.2L277.2 71.5c-9.3-2.7-18.8-4.6-28.6-5.9L255 39.8zM163.2 143.3L117.3 326.8C103.9 306.5 96 282.2 96 256c0-48.7 27.2-91 67.2-112.7zm8.6 229.5l61.1-244.6c9.9 .7 19.5 2.5 28.7 5.3l-62 248.1c-9.7-1.9-19-4.8-27.8-8.8z"],headset:[512,512,[],"f590","M256 48C141.1 48 48 141.1 48 256v40c0 13.3-10.7 24-24 24s-24-10.7-24-24V256C0 114.6 114.6 0 256 0S512 114.6 512 256V400.1c0 48.6-39.4 88-88.1 88L313.6 488c-8.3 14.3-23.8 24-41.6 24H240c-26.5 0-48-21.5-48-48s21.5-48 48-48h32c17.8 0 33.3 9.7 41.6 24l110.4 .1c22.1 0 40-17.9 40-40V256c0-114.9-93.1-208-208-208zM144 208h16c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32H144c-35.3 0-64-28.7-64-64V272c0-35.3 28.7-64 64-64zm224 0c35.3 0 64 28.7 64 64v48c0 35.3-28.7 64-64 64H352c-17.7 0-32-14.3-32-32V240c0-17.7 14.3-32 32-32h16z"],"store-slash":[640,512,[],"e071","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-86.8-68V384 252.6c-4 1-8 1.8-12.3 2.3l-.1 0c-5.3 .7-10.7 1.1-16.2 1.1c-12.4 0-24.3-1.9-35.4-5.3V350.9L301.2 210.7c7-4.4 13.3-9.7 18.8-15.7c15.9 17.6 39.1 29 65.2 29c26.2 0 49.3-11.4 65.2-29c16 17.6 39.1 29 65.2 29c4.1 0 8.1-.3 12.1-.8c55.5-7.4 81.8-72.5 52.1-119.4L522.3 13.1C517.2 5 508.1 0 498.4 0H141.6c-9.7 0-18.8 5-23.9 13.1l-22.7 36L38.8 5.1zM514.9 501.7L365.5 384H160V250.6c-11.2 3.5-23.2 5.4-35.6 5.4c-5.5 0-11-.4-16.3-1.1l-.1 0c-4.1-.6-8.1-1.3-12-2.3V384v64c0 35.3 28.7 64 64 64H480c12.9 0 24.8-3.8 34.9-10.3zM155.3 218.4L48.9 134.5c-6.1 40.6 19.5 82.8 63.3 88.7c4 .5 8.1 .8 12.1 .8c11 0 21.4-2 31-5.6z"],"road-circle-xmark":[640,512,[],"e566","M213.2 32H288V96c0 17.7 14.3 32 32 32s32-14.3 32-32V32h74.8c27.1 0 51.3 17.1 60.3 42.6l42.7 120.6c-10.9-2.1-22.2-3.2-33.8-3.2c-59.5 0-112.1 29.6-144 74.8V224c0-17.7-14.3-32-32-32s-32 14.3-32 32v64c0 17.7 14.3 32 32 32c2.3 0 4.6-.3 6.8-.7c-4.5 15.5-6.8 31.8-6.8 48.7c0 5.4 .2 10.7 .7 16l-.7 0c-17.7 0-32 14.3-32 32v64H86.6C56.5 480 32 455.5 32 425.4c0-6.2 1.1-12.4 3.1-18.2L152.9 74.6C162 49.1 186.1 32 213.2 32zM496 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm22.6 144l36.7-36.7c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0L496 345.4l-36.7-36.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6L473.4 368l-36.7 36.7c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L496 390.6l36.7 36.7c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6L518.6 368z"],"user-minus":[640,512,[],"f503","M96 128a128 128 0 1 1 256 0A128 128 0 1 1 96 128zM0 482.3C0 383.8 79.8 304 178.3 304h91.4C368.2 304 448 383.8 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3zM472 200H616c13.3 0 24 10.7 24 24s-10.7 24-24 24H472c-13.3 0-24-10.7-24-24s10.7-24 24-24z"],"mars-stroke-up":[320,512,[9896,"mars-stroke-v"],"f22a","M148.7 4.7c6.2-6.2 16.4-6.2 22.6 0l64 64c4.6 4.6 5.9 11.5 3.5 17.4s-8.3 9.9-14.8 9.9H184v24h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H184v24c0 .6 0 1.2-.1 1.8c77 11.6 136.1 78 136.1 158.2c0 88.4-71.6 160-160 160S0 440.4 0 352c0-80.2 59.1-146.7 136.1-158.2c0-.6-.1-1.2-.1-1.8V168H104c-13.3 0-24-10.7-24-24s10.7-24 24-24h32V96H96c-6.5 0-12.3-3.9-14.8-9.9s-1.1-12.9 3.5-17.4l64-64zM256 352A96 96 0 1 0 64 352a96 96 0 1 0 192 0z"],"champagne-glasses":[640,512,[129346,"glass-cheers"],"f79f","M155.6 17.3C163 3 179.9-3.6 195 1.9L320 47.5l125-45.6c15.1-5.5 32 1.1 39.4 15.4l78.8 152.9c28.8 55.8 10.3 122.3-38.5 156.6L556.1 413l41-15c16.6-6 35 2.5 41 19.1s-2.5 35-19.1 41l-71.1 25.9L476.8 510c-16.6 6.1-35-2.5-41-19.1s2.5-35 19.1-41l41-15-31.3-86.2c-59.4 5.2-116.2-34-130-95.2L320 188.8l-14.6 64.7c-13.8 61.3-70.6 100.4-130 95.2l-31.3 86.2 41 15c16.6 6 25.2 24.4 19.1 41s-24.4 25.2-41 19.1L92.2 484.1 21.1 458.2c-16.6-6.1-25.2-24.4-19.1-41s24.4-25.2 41-19.1l41 15 31.3-86.2C66.5 292.5 48.1 226 76.9 170.2L155.6 17.3zm44 54.4l-27.2 52.8L261.6 157l13.1-57.9L199.6 71.7zm240.9 0L365.4 99.1 378.5 157l89.2-32.5L440.5 71.7z"],clipboard:[384,512,[128203],"f328","M192 0c-41.8 0-77.4 26.7-90.5 64H64C28.7 64 0 92.7 0 128V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H282.5C269.4 26.7 233.8 0 192 0zm0 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM112 192H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],"house-circle-exclamation":[640,512,[],"e50a","M320.7 352c8.1-89.7 83.5-160 175.3-160c8.9 0 17.6 .7 26.1 1.9L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32v69.7c-.1 .9-.1 1.8-.1 2.8V472c0 22.1 17.9 40 40 40h16c1.2 0 2.4-.1 3.6-.2c1.5 .1 3 .2 4.5 .2H160h24c22.1 0 40-17.9 40-40V448 384c0-17.7 14.3-32 32-32h64l.7 0zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z"],"file-arrow-up":[384,512,["file-upload"],"f574","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM216 408c0 13.3-10.7 24-24 24s-24-10.7-24-24V305.9l-31 31c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l72-72c9.4-9.4 24.6-9.4 33.9 0l72 72c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-31-31V408z"],wifi:[640,512,["wifi-3","wifi-strong"],"f1eb","M54.2 202.9C123.2 136.7 216.8 96 320 96s196.8 40.7 265.8 106.9c12.8 12.2 33 11.8 45.2-.9s11.8-33-.9-45.2C549.7 79.5 440.4 32 320 32S90.3 79.5 9.8 156.7C-2.9 169-3.3 189.2 8.9 202s32.5 13.2 45.2 .9zM320 256c56.8 0 108.6 21.1 148.2 56c13.3 11.7 33.5 10.4 45.2-2.8s10.4-33.5-2.8-45.2C459.8 219.2 393 192 320 192s-139.8 27.2-190.5 72c-13.3 11.7-14.5 31.9-2.8 45.2s31.9 14.5 45.2 2.8c39.5-34.9 91.3-56 148.2-56zm64 160a64 64 0 1 0 -128 0 64 64 0 1 0 128 0z"],bath:[512,512,[128705,"bathtub"],"f2cd","M96 77.3c0-7.3 5.9-13.3 13.3-13.3c3.5 0 6.9 1.4 9.4 3.9l14.9 14.9C130 91.8 128 101.7 128 112c0 19.9 7.2 38 19.2 52c-5.3 9.2-4 21.1 3.8 29c9.4 9.4 24.6 9.4 33.9 0L289 89c9.4-9.4 9.4-24.6 0-33.9c-7.9-7.9-19.8-9.1-29-3.8C246 39.2 227.9 32 208 32c-10.3 0-20.2 2-29.2 5.5L163.9 22.6C149.4 8.1 129.7 0 109.3 0C66.6 0 32 34.6 32 77.3V256c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H96V77.3zM32 352v16c0 28.4 12.4 54 32 71.6V480c0 17.7 14.3 32 32 32s32-14.3 32-32V464H384v16c0 17.7 14.3 32 32 32s32-14.3 32-32V439.6c19.6-17.6 32-43.1 32-71.6V352H32z"],underline:[448,512,[],"f0cd","M16 64c0-17.7 14.3-32 32-32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H128V224c0 53 43 96 96 96s96-43 96-96V96H304c-17.7 0-32-14.3-32-32s14.3-32 32-32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H384V224c0 88.4-71.6 160-160 160s-160-71.6-160-160V96H48C30.3 96 16 81.7 16 64zM0 448c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32z"],"user-pen":[640,512,["user-edit"],"f4ff","M224 256A128 128 0 1 0 224 0a128 128 0 1 0 0 256zm-45.7 48C79.8 304 0 383.8 0 482.3C0 498.7 13.3 512 29.7 512H322.8c-3.1-8.8-3.7-18.4-1.4-27.8l15-60.1c2.8-11.3 8.6-21.5 16.8-29.7l40.3-40.3c-32.1-31-75.7-50.1-123.9-50.1H178.3zm435.5-68.3c-15.6-15.6-40.9-15.6-56.6 0l-29.4 29.4 71 71 29.4-29.4c15.6-15.6 15.6-40.9 0-56.6l-14.4-14.4zM375.9 417c-4.1 4.1-7 9.2-8.4 14.9l-15 60.1c-1.4 5.5 .2 11.2 4.2 15.2s9.7 5.6 15.2 4.2l60.1-15c5.6-1.4 10.8-4.3 14.9-8.4L576.1 358.7l-71-71L375.9 417z"],signature:[640,512,[],"f5b7","M192 128c0-17.7 14.3-32 32-32s32 14.3 32 32v7.8c0 27.7-2.4 55.3-7.1 82.5l-84.4 25.3c-40.6 12.2-68.4 49.6-68.4 92v71.9c0 40 32.5 72.5 72.5 72.5c26 0 50-13.9 62.9-36.5l13.9-24.3c26.8-47 46.5-97.7 58.4-150.5l94.4-28.3-12.5 37.5c-3.3 9.8-1.6 20.5 4.4 28.8s15.7 13.3 26 13.3H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H460.4l18-53.9c3.8-11.3 .9-23.8-7.4-32.4s-20.7-11.8-32.2-8.4L316.4 198.1c2.4-20.7 3.6-41.4 3.6-62.3V128c0-53-43-96-96-96s-96 43-96 96v32c0 17.7 14.3 32 32 32s32-14.3 32-32V128zm-9.2 177l49-14.7c-10.4 33.8-24.5 66.4-42.1 97.2l-13.9 24.3c-1.5 2.6-4.3 4.3-7.4 4.3c-4.7 0-8.5-3.8-8.5-8.5V335.6c0-14.1 9.3-26.6 22.8-30.7zM24 368c-13.3 0-24 10.7-24 24s10.7 24 24 24H64.3c-.2-2.8-.3-5.6-.3-8.5V368H24zm592 48c13.3 0 24-10.7 24-24s-10.7-24-24-24H305.9c-6.7 16.3-14.2 32.3-22.3 48H616z"],stroopwafel:[512,512,[],"f551","M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM312.6 63.7c-6.2-6.2-16.4-6.2-22.6 0L256 97.6 222.1 63.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l33.9 33.9-45.3 45.3-56.6-56.6c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l56.6 56.6-45.3 45.3L86.3 199.4c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6L97.6 256 63.7 289.9c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0l33.9-33.9 45.3 45.3-56.6 56.6c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0l56.6-56.6 45.3 45.3-33.9 33.9c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L256 414.4l33.9 33.9c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-33.9-33.9 45.3-45.3 56.6 56.6c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-56.6-56.6 45.3-45.3 33.9 33.9c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6L414.4 256l33.9-33.9c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0l-33.9 33.9-45.3-45.3 56.6-56.6c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0l-56.6 56.6-45.3-45.3 33.9-33.9c6.2-6.2 6.2-16.4 0-22.6zM142.9 256l45.3-45.3L233.4 256l-45.3 45.3L142.9 256zm67.9 67.9L256 278.6l45.3 45.3L256 369.1l-45.3-45.3zM278.6 256l45.3-45.3L369.1 256l-45.3 45.3L278.6 256zm22.6-67.9L256 233.4l-45.3-45.3L256 142.9l45.3 45.3z"],bold:[384,512,[],"f032","M0 64C0 46.3 14.3 32 32 32H80 96 224c70.7 0 128 57.3 128 128c0 31.3-11.3 60.1-30 82.3c37.1 22.4 62 63.1 62 109.7c0 70.7-57.3 128-128 128H96 80 32c-17.7 0-32-14.3-32-32s14.3-32 32-32H48V256 96H32C14.3 96 0 81.7 0 64zM224 224c35.3 0 64-28.7 64-64s-28.7-64-64-64H112V224H224zM112 288V416H256c35.3 0 64-28.7 64-64s-28.7-64-64-64H224 112z"],"anchor-lock":[640,512,[],"e4ad","M320 96a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm21.1 80C367 158.8 384 129.4 384 96c0-53-43-96-96-96s-96 43-96 96c0 33.4 17 62.8 42.9 80H224c-17.7 0-32 14.3-32 32s14.3 32 32 32h32V448H208c-53 0-96-43-96-96v-6.1l7 7c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L97 263c-9.4-9.4-24.6-9.4-33.9 0L7 319c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l7-7V352c0 88.4 71.6 160 160 160h80 80c8 0 15.9-.6 23.6-1.7c-4.8-9-7.6-19.3-7.6-30.3V446.7c-5.2 .9-10.5 1.3-16 1.3H320V240h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H341.1zM528 240c17.7 0 32 14.3 32 32v48H496V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z"],"building-ngo":[384,512,[],"e4d7","M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM168 64h48c8.8 0 16 7.2 16 16s-7.2 16-16 16H184v64h16V144c0-8.8 7.2-16 16-16s16 7.2 16 16v24c0 13.3-10.7 24-24 24H176c-13.3 0-24-10.7-24-24V80c0-8.8 7.2-16 16-16zM304 96c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16s16-7.2 16-16V112c0-8.8-7.2-16-16-16zm-48 16c0-26.5 21.5-48 48-48s48 21.5 48 48v32c0 26.5-21.5 48-48 48s-48-21.5-48-48V112zM61.3 71.1l34.7 52V80c0-8.8 7.2-16 16-16s16 7.2 16 16v96c0 7.1-4.6 13.3-11.4 15.3s-14-.6-17.9-6.4L64 132.8V176c0 8.8-7.2 16-16 16s-16-7.2-16-16V80c0-7.1 4.6-13.3 11.4-15.3s14 .6 17.9 6.4z"],"manat-sign":[384,512,[],"e1d5","M192 32c-17.7 0-32 14.3-32 32V98.7C69.2 113.9 0 192.9 0 288V448c0 17.7 14.3 32 32 32s32-14.3 32-32V288c0-59.6 40.8-109.8 96-124V448c0 17.7 14.3 32 32 32s32-14.3 32-32V164c55.2 14.2 96 64.3 96 124V448c0 17.7 14.3 32 32 32s32-14.3 32-32V288c0-95.1-69.2-174.1-160-189.3V64c0-17.7-14.3-32-32-32z"],"not-equal":[448,512,[],"f53e","M369.8 37.4c14.7 9.8 18.7 29.7 8.9 44.4L337.1 144H400c17.7 0 32 14.3 32 32s-14.3 32-32 32H294.5l-64 96H400c17.7 0 32 14.3 32 32s-14.3 32-32 32H187.8l-65.2 97.7c-9.8 14.7-29.7 18.7-44.4 8.9s-18.7-29.7-8.9-44.4L110.9 368H48c-17.7 0-32-14.3-32-32s14.3-32 32-32H153.5l64-96H48c-17.7 0-32-14.3-32-32s14.3-32 32-32H260.2l65.2-97.7c9.8-14.7 29.7-18.7 44.4-8.9z"],"border-top-left":[448,512,["border-style"],"f853","M0 448c0 17.7 14.3 32 32 32s32-14.3 32-32l0-336c0-8.8 7.2-16 16-16l336 0c17.7 0 32-14.3 32-32s-14.3-32-32-32L80 32C35.8 32 0 67.8 0 112L0 448zm160 0a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm192 0a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm-96 0a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm192 0a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM416 288a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm0 32a32 32 0 1 0 0 64 32 32 0 1 0 0-64zm0-128a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"map-location-dot":[576,512,["map-marked-alt"],"f5a0","M408 120c0 54.6-73.1 151.9-105.2 192c-7.7 9.6-22 9.6-29.6 0C241.1 271.9 168 174.6 168 120C168 53.7 221.7 0 288 0s120 53.7 120 120zm8 80.4c3.5-6.9 6.7-13.8 9.6-20.6c.5-1.2 1-2.5 1.5-3.7l116-46.4C558.9 123.4 576 135 576 152V422.8c0 9.8-6 18.6-15.1 22.3L416 503V200.4zM137.6 138.3c2.4 14.1 7.2 28.3 12.8 41.5c2.9 6.8 6.1 13.7 9.6 20.6V451.8L32.9 502.7C17.1 509 0 497.4 0 480.4V209.6c0-9.8 6-18.6 15.1-22.3l122.6-49zM327.8 332c13.9-17.4 35.7-45.7 56.2-77V504.3L192 449.4V255c20.5 31.3 42.3 59.6 56.2 77c20.5 25.6 59.1 25.6 79.6 0zM288 152a40 40 0 1 0 0-80 40 40 0 1 0 0 80z"],jedi:[576,512,[],"f669","M246 315.7l-21.2-31.9c-2.1-3.2-1.7-7.4 1-10.1s6.9-3.1 10.1-1l29.5 19.7c2.1 1.4 4.9 0 5-2.6L279.7 8c.1-4.5 3.8-8 8.3-8s8.1 3.5 8.3 8l9.4 281.9c.1 2.5 2.9 3.9 5 2.6l29.5-19.7c3.2-2.1 7.4-1.7 10.1 1s3.1 6.9 1 10.1L330 315.7c-1.3 1.9-.2 4.5 2 4.9l37.6 7.5c3.7 .7 6.4 4 6.4 7.8s-2.7 7.1-6.4 7.8L332 351.4c-2.2 .4-3.3 3-2 4.9l21.2 31.9c2.1 3.2 1.7 7.4-1 10.1s-6.9 3.1-10.1 1l-26.3-17.6c-2.2-1.4-5.1 .2-5 2.8l2.1 61.5C370.6 435.2 416 382.9 416 320c0-37-15.7-70.4-40.8-93.7c-7-6.5-6.5-18.6 1-24.4C410.1 175.5 432 134.3 432 88c0-16.8-2.9-33-8.2-48c-4.6-13 10.2-30 21.4-22c53.5 38 92.7 94.8 107.8 160.7c.5 2.1-.2 4.3-1.7 5.9l-28.4 28.4c-4 4-1.2 10.9 4.5 10.9h26c3.4 0 6.2 2.6 6.3 6c.1 3.3 .2 6.6 .2 10c0 17.5-1.7 34.7-4.8 51.3c-.2 1.2-.9 2.4-1.7 3.3l-46.5 46.5c-4 4-1.2 10.9 4.5 10.9H526c4.6 0 7.7 4.8 5.7 9C487.2 450.5 394.8 512 288 512S88.8 450.5 44.3 361c-2.1-4.2 1-9 5.7-9H64.5c5.7 0 8.6-6.9 4.5-10.9L22.6 294.6c-.9-.9-1.5-2-1.7-3.3C17.7 274.7 16 257.5 16 240c0-3.3 .1-6.7 .2-10c.1-3.4 2.9-6 6.3-6h26c5.7 0 8.6-6.9 4.5-10.9L24.6 184.6c-1.5-1.5-2.2-3.8-1.7-5.9C38.1 112.8 77.3 56 130.8 18c11.3-8 26 8.9 21.4 22c-5.3 15-8.2 31.2-8.2 48c0 46.3 21.9 87.5 55.8 113.9c7.5 5.8 8 17.9 1 24.4C175.7 249.6 160 283 160 320c0 62.9 45.4 115.2 105.1 126l2.1-61.5c.1-2.6-2.8-4.2-5-2.8l-26.3 17.6c-3.2 2.1-7.4 1.7-10.1-1s-3.1-6.9-1-10.1L246 356.3c1.3-1.9 .2-4.5-2-4.9l-37.6-7.5c-3.7-.7-6.4-4-6.4-7.8s2.7-7.1 6.4-7.8l37.6-7.5c2.2-.4 3.3-3 2-4.9z"],"square-poll-vertical":[448,512,["poll"],"f681","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm64 192c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V256c0-17.7 14.3-32 32-32zm64-64c0-17.7 14.3-32 32-32s32 14.3 32 32V352c0 17.7-14.3 32-32 32s-32-14.3-32-32V160zM320 288c17.7 0 32 14.3 32 32v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V320c0-17.7 14.3-32 32-32z"],"mug-hot":[512,512,[9749],"f7b6","M88 0C74.7 0 64 10.7 64 24c0 38.9 23.4 59.4 39.1 73.1l1.1 1C120.5 112.3 128 119.9 128 136c0 13.3 10.7 24 24 24s24-10.7 24-24c0-38.9-23.4-59.4-39.1-73.1l-1.1-1C119.5 47.7 112 40.1 112 24c0-13.3-10.7-24-24-24zM32 192c-17.7 0-32 14.3-32 32V416c0 53 43 96 96 96H288c53 0 96-43 96-96h16c61.9 0 112-50.1 112-112s-50.1-112-112-112H352 32zm352 64h16c26.5 0 48 21.5 48 48s-21.5 48-48 48H384V256zM224 24c0-13.3-10.7-24-24-24s-24 10.7-24 24c0 38.9 23.4 59.4 39.1 73.1l1.1 1C232.5 112.3 240 119.9 240 136c0 13.3 10.7 24 24 24s24-10.7 24-24c0-38.9-23.4-59.4-39.1-73.1l-1.1-1C231.5 47.7 224 40.1 224 24z"],"car-battery":[512,512,["battery-car"],"f5df","M80 96c0-17.7 14.3-32 32-32h64c17.7 0 32 14.3 32 32l96 0c0-17.7 14.3-32 32-32h64c17.7 0 32 14.3 32 32h16c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V160c0-35.3 28.7-64 64-64l16 0zm304 96c0-8.8-7.2-16-16-16s-16 7.2-16 16v32H320c-8.8 0-16 7.2-16 16s7.2 16 16 16h32v32c0 8.8 7.2 16 16 16s16-7.2 16-16V256h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H384V192zM80 240c0 8.8 7.2 16 16 16h96c8.8 0 16-7.2 16-16s-7.2-16-16-16H96c-8.8 0-16 7.2-16 16z"],gift:[512,512,[127873],"f06b","M190.5 68.8L225.3 128H224 152c-22.1 0-40-17.9-40-40s17.9-40 40-40h2.2c14.9 0 28.8 7.9 36.3 20.8zM64 88c0 14.4 3.5 28 9.6 40H32c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H480c17.7 0 32-14.3 32-32V160c0-17.7-14.3-32-32-32H438.4c6.1-12 9.6-25.6 9.6-40c0-48.6-39.4-88-88-88h-2.2c-31.9 0-61.5 16.9-77.7 44.4L256 85.5l-24.1-41C215.7 16.9 186.1 0 154.2 0H152C103.4 0 64 39.4 64 88zm336 0c0 22.1-17.9 40-40 40H288h-1.3l34.8-59.2C329.1 55.9 342.9 48 357.8 48H360c22.1 0 40 17.9 40 40zM32 288V464c0 26.5 21.5 48 48 48H224V288H32zM288 512H432c26.5 0 48-21.5 48-48V288H288V512z"],"dice-two":[448,512,[9857],"f528","M0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM352 352a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM128 192a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"chess-queen":[512,512,[9819],"f445","M256 0a56 56 0 1 1 0 112A56 56 0 1 1 256 0zM134.1 143.8c3.3-13 15-23.8 30.2-23.8c12.3 0 22.6 7.2 27.7 17c12 23.2 36.2 39 64 39s52-15.8 64-39c5.1-9.8 15.4-17 27.7-17c15.3 0 27 10.8 30.2 23.8c7 27.8 32.2 48.3 62.1 48.3c10.8 0 21-2.7 29.8-7.4c8.4-4.4 18.9-4.5 27.6 .9c13 8 17.1 25 9.2 38L399.7 400H384 343.6 168.4 128 112.3L5.4 223.6c-7.9-13-3.8-30 9.2-38c8.7-5.3 19.2-5.3 27.6-.9c8.9 4.7 19 7.4 29.8 7.4c29.9 0 55.1-20.5 62.1-48.3zM256 224l0 0 0 0h0zM112 432H400l41.4 41.4c4.2 4.2 6.6 10 6.6 16c0 12.5-10.1 22.6-22.6 22.6H86.6C74.1 512 64 501.9 64 489.4c0-6 2.4-11.8 6.6-16L112 432z"],glasses:[576,512,[],"f530","M118.6 80c-11.5 0-21.4 7.9-24 19.1L57 260.3c20.5-6.2 48.3-12.3 78.7-12.3c32.3 0 61.8 6.9 82.8 13.5c10.6 3.3 19.3 6.7 25.4 9.2c3.1 1.3 5.5 2.4 7.3 3.2c.9 .4 1.6 .7 2.1 1l.6 .3 .2 .1 .1 0 0 0 0 0s0 0-6.3 12.7h0l6.3-12.7c5.8 2.9 10.4 7.3 13.5 12.7h40.6c3.1-5.3 7.7-9.8 13.5-12.7l6.3 12.7h0c-6.3-12.7-6.3-12.7-6.3-12.7l0 0 0 0 .1 0 .2-.1 .6-.3c.5-.2 1.2-.6 2.1-1c1.8-.8 4.2-1.9 7.3-3.2c6.1-2.6 14.8-5.9 25.4-9.2c21-6.6 50.4-13.5 82.8-13.5c30.4 0 58.2 6.1 78.7 12.3L481.4 99.1c-2.6-11.2-12.6-19.1-24-19.1c-3.1 0-6.2 .6-9.2 1.8L416.9 94.3c-12.3 4.9-26.3-1.1-31.2-13.4s1.1-26.3 13.4-31.2l31.3-12.5c8.6-3.4 17.7-5.2 27-5.2c33.8 0 63.1 23.3 70.8 56.2l43.9 188c1.7 7.3 2.9 14.7 3.5 22.1c.3 1.9 .5 3.8 .5 5.7v6.7V352v16c0 61.9-50.1 112-112 112H419.7c-59.4 0-108.5-46.4-111.8-105.8L306.6 352H269.4l-1.2 22.2C264.9 433.6 215.8 480 156.3 480H112C50.1 480 0 429.9 0 368V352 310.7 304c0-1.9 .2-3.8 .5-5.7c.6-7.4 1.8-14.8 3.5-22.1l43.9-188C55.5 55.3 84.8 32 118.6 32c9.2 0 18.4 1.8 27 5.2l31.3 12.5c12.3 4.9 18.3 18.9 13.4 31.2s-18.9 18.3-31.2 13.4L127.8 81.8c-2.9-1.2-6-1.8-9.2-1.8zM64 325.4V368c0 26.5 21.5 48 48 48h44.3c25.5 0 46.5-19.9 47.9-45.3l2.5-45.6c-2.3-.8-4.9-1.7-7.5-2.5c-17.2-5.4-39.9-10.5-63.6-10.5c-23.7 0-46.2 5.1-63.2 10.5c-3.1 1-5.9 1.9-8.5 2.9zM512 368V325.4c-2.6-.9-5.5-1.9-8.5-2.9c-17-5.4-39.5-10.5-63.2-10.5c-23.7 0-46.4 5.1-63.6 10.5c-2.7 .8-5.2 1.7-7.5 2.5l2.5 45.6c1.4 25.4 22.5 45.3 47.9 45.3H464c26.5 0 48-21.5 48-48z"],"chess-board":[448,512,[],"f43c","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm64 64v64h64V96h64v64h64V96h64v64H320v64h64v64H320v64h64v64H320V352H256v64H192V352H128v64H64V352h64V288H64V224h64V160H64V96h64zm64 128h64V160H192v64zm0 64V224H128v64h64zm64 0H192v64h64V288zm0 0h64V224H256v64z"],"building-circle-check":[640,512,[],"e4d2","M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c15.1 0 28.5-6.9 37.3-17.8C340.4 462.2 320 417.5 320 368c0-54.7 24.9-103.5 64-135.8V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zM640 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z"],"person-chalkboard":[640,512,[],"e53d","M192 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-8 384V352h16V480c0 17.7 14.3 32 32 32s32-14.3 32-32V192h56 64 16c17.7 0 32-14.3 32-32s-14.3-32-32-32H384V64H576V256H384V224H320v48c0 26.5 21.5 48 48 48H592c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H368c-26.5 0-48 21.5-48 48v80H243.1 177.1c-33.7 0-64.9 17.7-82.3 46.6l-58.3 97c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9L120 256.9V480c0 17.7 14.3 32 32 32s32-14.3 32-32z"],"mars-stroke-right":[640,512,[9897,"mars-stroke-h"],"f22b","M208 368a112 112 0 1 0 0-224 112 112 0 1 0 0 224zm174.4-88C370.7 365.8 297.1 432 208 432c-97.2 0-176-78.8-176-176s78.8-176 176-176c89.1 0 162.7 66.2 174.4 152H416V176c0-13.3 10.7-24 24-24s24 10.7 24 24v56h32V176c0-9.7 5.8-18.5 14.8-22.2s19.3-1.7 26.2 5.2l80 80c9.4 9.4 9.4 24.6 0 33.9l-80 80c-6.9 6.9-17.2 8.9-26.2 5.2s-14.8-12.5-14.8-22.2V280H464v56c0 13.3-10.7 24-24 24s-24-10.7-24-24V280H382.4z"],"hand-back-fist":[448,512,["hand-rock"],"f255","M144 0C117.5 0 96 21.5 96 48V96v28.5V176c0 8.8-7.2 16-16 16s-16-7.2-16-16V149.3l-9 7.5C40.4 169 32 187 32 206V244c0 38 16.9 74 46.1 98.3L128 384v96c0 17.7 14.3 32 32 32H320c17.7 0 32-14.3 32-32V374.7c46.9-19 80-65 80-118.7V176 160 144c0-26.5-21.5-48-48-48c-12.4 0-23.6 4.7-32.1 12.3C350 83.5 329.3 64 304 64c-12.4 0-23.6 4.7-32.1 12.3C270 51.5 249.3 32 224 32c-12.4 0-23.6 4.7-32.1 12.3C190 19.5 169.3 0 144 0z"],"square-caret-up":[448,512,["caret-square-up"],"f151","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM224 160c6.7 0 13 2.8 17.6 7.7l104 112c6.5 7 8.2 17.2 4.4 25.9s-12.5 14.4-22 14.4H120c-9.5 0-18.2-5.7-22-14.4s-2.1-18.9 4.4-25.9l104-112c4.5-4.9 10.9-7.7 17.6-7.7z"],"cloud-showers-water":[576,512,[],"e4e4","M224 0c38.6 0 71.9 22.8 87.2 55.7C325.7 41.1 345.8 32 368 32c38.7 0 71 27.5 78.4 64H448c35.3 0 64 28.7 64 64s-28.7 64-64 64H128c-35.3 0-64-28.7-64-64s28.7-64 64-64c0-53 43-96 96-96zM140.6 292.3l-48 80c-6.8 11.4-21.6 15-32.9 8.2s-15.1-21.6-8.2-32.9l48-80c6.8-11.4 21.6-15.1 32.9-8.2s15.1 21.6 8.2 32.9zm327.8-32.9c11.4 6.8 15 21.6 8.2 32.9l-48 80c-6.8 11.4-21.6 15-32.9 8.2s-15-21.6-8.2-32.9l48-80c6.8-11.4 21.6-15.1 32.9-8.2zM252.6 292.3l-48 80c-6.8 11.4-21.6 15-32.9 8.2s-15.1-21.6-8.2-32.9l48-80c6.8-11.4 21.6-15.1 32.9-8.2s15.1 21.6 8.2 32.9zm103.8-32.9c11.4 6.8 15 21.6 8.2 32.9l-48 80c-6.8 11.4-21.6 15-32.9 8.2s-15.1-21.6-8.2-32.9l48-80c6.8-11.4 21.6-15.1 32.9-8.2zM306.5 421.9C329 437.4 356.5 448 384 448c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 501.7 417 512 384 512c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 437.2 165.1 448 192 448c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z"],"chart-bar":[512,512,["bar-chart"],"f080","M32 32c17.7 0 32 14.3 32 32V400c0 8.8 7.2 16 16 16H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H80c-44.2 0-80-35.8-80-80V64C0 46.3 14.3 32 32 32zm96 96c0-17.7 14.3-32 32-32l192 0c17.7 0 32 14.3 32 32s-14.3 32-32 32l-192 0c-17.7 0-32-14.3-32-32zm32 64H288c17.7 0 32 14.3 32 32s-14.3 32-32 32H160c-17.7 0-32-14.3-32-32s14.3-32 32-32zm0 96H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H160c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],"hands-bubbles":[576,512,["hands-wash"],"e05e","M416 64a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm96 128a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM160 464a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM32 160l.1 72.6c.1 52.2 24 101 64 133.1c-.1-1.9-.1-3.8-.1-5.7v-8c0-71.8 37-138.6 97.9-176.7l60.2-37.6c8.6-5.4 17.9-8.4 27.3-9.4l45.9-79.5c6.6-11.5 2.7-26.2-8.8-32.8s-26.2-2.7-32.8 8.8l-78 135.1c-3.3 5.7-10.7 7.7-16.4 4.4s-7.7-10.7-4.4-16.4l62-107.4c6.6-11.5 2.7-26.2-8.8-32.8S214 5 207.4 16.5l-68 117.8 0 0 0 0-43.3 75L96 160c0-17.7-14.4-32-32-32s-32 14.4-32 32zM332.1 88.5L307.5 131c13.9 4.5 26.4 13.7 34.7 27c.9 1.5 1.8 2.9 2.5 4.4l28.9-50c6.6-11.5 2.7-26.2-8.8-32.8s-26.2-2.7-32.8 8.8zm46.4 63.7l-26.8 46.4c-.6 6-2.1 11.8-4.3 17.4H352h13.3l0 0H397l23-39.8c6.6-11.5 2.7-26.2-8.8-32.8s-26.2-2.7-32.8 8.8zM315.1 175c-9.4-15-29.1-19.5-44.1-10.2l-60.2 37.6C159.3 234.7 128 291.2 128 352v8c0 8.9 .8 17.6 2.2 26.1c35.4 8.2 61.8 40 61.8 77.9c0 6.3-.7 12.5-2.1 18.4C215.1 501 246.3 512 280 512H456c13.3 0 24-10.7 24-24s-10.7-24-24-24H364c-6.6 0-12-5.4-12-12s5.4-12 12-12H488c13.3 0 24-10.7 24-24s-10.7-24-24-24H364c-6.6 0-12-5.4-12-12s5.4-12 12-12H520c13.3 0 24-10.7 24-24s-10.7-24-24-24H364c-6.6 0-12-5.4-12-12s5.4-12 12-12H488c13.3 0 24-10.7 24-24s-10.7-24-24-24H352l0 0 0 0H258.8L305 219.1c15-9.4 19.5-29.1 10.2-44.1z"],"less-than-equal":[448,512,[],"f537","M395.9 93.7c16.4-6.6 24.4-25.2 17.8-41.6s-25.2-24.4-41.6-17.8l-320 128C40 167.1 32 178.9 32 192s8 24.9 20.1 29.7l320 128c16.4 6.6 35-1.4 41.6-17.8s-1.4-35-17.8-41.6L150.2 192 395.9 93.7zM32 416c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H32z"],train:[448,512,[128646],"f238","M96 0C43 0 0 43 0 96V352c0 48 35.2 87.7 81.1 94.9l-46 46C28.1 499.9 33.1 512 43 512H82.7c8.5 0 16.6-3.4 22.6-9.4L160 448H288l54.6 54.6c6 6 14.1 9.4 22.6 9.4H405c10 0 15-12.1 7.9-19.1l-46-46c46-7.1 81.1-46.9 81.1-94.9V96c0-53-43-96-96-96H96zM64 96c0-17.7 14.3-32 32-32H352c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V96zM224 288a48 48 0 1 1 0 96 48 48 0 1 1 0-96z"],"eye-low-vision":[640,512,["low-vision"],"f2a8","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L525.6 386.7c39.6-40.6 66.4-86.1 79.9-118.4c3.3-7.9 3.3-16.7 0-24.6c-14.9-35.7-46.2-87.7-93-131.1C465.5 68.8 400.8 32 320 32c-68.2 0-125 26.3-169.3 60.8L38.8 5.1zM223 149.5c48.6-44.3 123-50.8 179.3-11.7c60.8 42.4 78.9 123.2 44.2 186.9L408 294.5c8.4-19.3 10.6-41.4 4.8-63.3c-11.1-41.5-47.8-69.4-88.6-71.1c-5.8-.2-9.2 6.1-7.4 11.7c2.1 6.4 3.3 13.2 3.3 20.3c0 10.2-2.4 19.8-6.6 28.3L223 149.5zm-139.9 12c-11 14.4-20.5 28.7-28.4 42.2l339 265.7c18.7-5.5 36.2-13 52.6-21.8L83.1 161.5zm-50 86.3c-1.8 6.8-1.3 14 1.4 20.5c14.9 35.7 46.2 87.7 93 131.1C174.5 443.2 239.2 480 320 480c3.1 0 6.1-.1 9.2-.2L33.1 247.8z"],crow:[640,512,[],"f520","M456 0c-48.6 0-88 39.4-88 88v29.2L12.5 390.6c-14 10.8-16.6 30.9-5.9 44.9s30.9 16.6 44.9 5.9L126.1 384H259.2l46.6 113.1c5 12.3 19.1 18.1 31.3 13.1s18.1-19.1 13.1-31.3L311.1 384H352c1.1 0 2.1 0 3.2 0l46.6 113.2c5 12.3 19.1 18.1 31.3 13.1s18.1-19.1 13.1-31.3l-42-102C484.9 354.1 544 280 544 192V128v-8l80.5-20.1c8.6-2.1 13.8-10.8 11.6-19.4C629 52 603.4 32 574 32H523.9C507.7 12.5 483.3 0 456 0zm0 64a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],sailboat:[576,512,[],"e445","M256 16c0-7 4.5-13.2 11.2-15.3s13.9 .4 17.9 6.1l224 320c3.4 4.9 3.8 11.3 1.1 16.6s-8.2 8.6-14.2 8.6H272c-8.8 0-16-7.2-16-16V16zM212.1 96.5c7 1.9 11.9 8.2 11.9 15.5V336c0 8.8-7.2 16-16 16H80c-5.7 0-11-3-13.8-8s-2.9-11-.1-16l128-224c3.6-6.3 11-9.4 18-7.5zM5.7 404.3C2.8 394.1 10.5 384 21.1 384H554.9c10.6 0 18.3 10.1 15.4 20.3l-4 14.3C550.7 473.9 500.4 512 443 512H133C75.6 512 25.3 473.9 9.7 418.7l-4-14.3z"],"window-restore":[512,512,[],"f2d2","M432 64H208c-8.8 0-16 7.2-16 16V96H128V80c0-44.2 35.8-80 80-80H432c44.2 0 80 35.8 80 80V304c0 44.2-35.8 80-80 80H416V320h16c8.8 0 16-7.2 16-16V80c0-8.8-7.2-16-16-16zM0 192c0-35.3 28.7-64 64-64H320c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V192zm64 32c0 17.7 14.3 32 32 32H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H96c-17.7 0-32 14.3-32 32z"],"square-plus":[448,512,[61846,"plus-square"],"f0fe","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM200 344V280H136c-13.3 0-24-10.7-24-24s10.7-24 24-24h64V168c0-13.3 10.7-24 24-24s24 10.7 24 24v64h64c13.3 0 24 10.7 24 24s-10.7 24-24 24H248v64c0 13.3-10.7 24-24 24s-24-10.7-24-24z"],"torii-gate":[512,512,[9961],"f6a1","M0 80c0 26.5 21.5 48 48 48H64v64h64V128h96v64h64V128h96v64h64V128h16c26.5 0 48-21.5 48-48V13.4C512 6 506 0 498.6 0c-1.7 0-3.4 .3-5 1l-49 19.6C425.7 28.1 405.5 32 385.2 32H126.8c-20.4 0-40.5-3.9-59.4-11.4L18.4 1c-1.6-.6-3.3-1-5-1C6 0 0 6 0 13.4V80zM64 288V480c0 17.7 14.3 32 32 32s32-14.3 32-32V288H384V480c0 17.7 14.3 32 32 32s32-14.3 32-32V288h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64z"],frog:[576,512,[],"f52e","M368 32c41.7 0 75.9 31.8 79.7 72.5l85.6 26.3c25.4 7.8 42.8 31.3 42.8 57.9c0 21.8-11.7 41.9-30.7 52.7L400.8 323.5 493.3 416H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H480c-8.5 0-16.6-3.4-22.6-9.4L346.9 360.2c11.7-36 3.2-77.1-25.4-105.7c-40.6-40.6-106.3-40.6-146.9-.1L101 324.4c-6.4 6.1-6.7 16.2-.6 22.6s16.2 6.6 22.6 .6l73.8-70.2 .1-.1 .1-.1c3.5-3.5 7.3-6.6 11.3-9.2c27.9-18.5 65.9-15.4 90.5 9.2c24.7 24.7 27.7 62.9 9 90.9c-2.6 3.8-5.6 7.5-9 10.9L261.8 416H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H64c-35.3 0-64-28.7-64-64C0 249.6 127 112.9 289.3 97.5C296.2 60.2 328.8 32 368 32zm0 104a24 24 0 1 0 0-48 24 24 0 1 0 0 48z"],bucket:[448,512,[],"e4cf","M96 152v8H48v-8C48 68.1 116.1 0 200 0h48c83.9 0 152 68.1 152 152v8H352v-8c0-57.4-46.6-104-104-104H200C142.6 48 96 94.6 96 152zM0 224c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32h-5.1L388.5 469c-2.6 24.4-23.2 43-47.7 43H107.2c-24.6 0-45.2-18.5-47.7-43L37.1 256H32c-17.7 0-32-14.3-32-32z"],image:[512,512,[],"f03e","M0 96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM323.8 202.5c-4.5-6.6-11.9-10.5-19.8-10.5s-15.4 3.9-19.8 10.5l-87 127.6L170.7 297c-4.6-5.7-11.5-9-18.7-9s-14.2 3.3-18.7 9l-64 80c-5.8 7.2-6.9 17.1-2.9 25.4s12.4 13.6 21.6 13.6h96 32H424c8.9 0 17.1-4.9 21.2-12.8s3.6-17.4-1.4-24.7l-120-176zM112 192a48 48 0 1 0 0-96 48 48 0 1 0 0 96z"],microphone:[384,512,[],"f130","M192 0C139 0 96 43 96 96V256c0 53 43 96 96 96s96-43 96-96V96c0-53-43-96-96-96zM64 216c0-13.3-10.7-24-24-24s-24 10.7-24 24v40c0 89.1 66.2 162.7 152 174.4V464H120c-13.3 0-24 10.7-24 24s10.7 24 24 24h72 72c13.3 0 24-10.7 24-24s-10.7-24-24-24H216V430.4c85.8-11.7 152-85.3 152-174.4V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v40c0 70.7-57.3 128-128 128s-128-57.3-128-128V216z"],cow:[640,512,[128004],"f6c8","M96 224v32V416c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V327.8c9.9 6.6 20.6 12 32 16.1V368c0 8.8 7.2 16 16 16s16-7.2 16-16V351.1c5.3 .6 10.6 .9 16 .9s10.7-.3 16-.9V368c0 8.8 7.2 16 16 16s16-7.2 16-16V343.8c11.4-4 22.1-9.4 32-16.1V416c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V256l32 32v49.5c0 9.5 2.8 18.7 8.1 26.6L530 427c8.8 13.1 23.5 21 39.3 21c22.5 0 41.9-15.9 46.3-38l20.3-101.6c2.6-13-.3-26.5-8-37.3l-3.9-5.5V184c0-13.3-10.7-24-24-24s-24 10.7-24 24v14.4l-52.9-74.1C496 86.5 452.4 64 405.9 64H272 256 192 144C77.7 64 24 117.7 24 184v54C9.4 249.8 0 267.8 0 288v17.6c0 8 6.4 14.4 14.4 14.4C46.2 320 72 294.2 72 262.4V256 224 184c0-24.3 12.1-45.8 30.5-58.9C98.3 135.9 96 147.7 96 160v64zM560 336a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zM166.6 166.6c-4.2-4.2-6.6-10-6.6-16c0-12.5 10.1-22.6 22.6-22.6H361.4c12.5 0 22.6 10.1 22.6 22.6c0 6-2.4 11.8-6.6 16l-23.4 23.4C332.2 211.8 302.7 224 272 224s-60.2-12.2-81.9-33.9l-23.4-23.4z"],"caret-up":[320,512,[],"f0d8","M182.6 137.4c-12.5-12.5-32.8-12.5-45.3 0l-128 128c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8H288c12.9 0 24.6-7.8 29.6-19.8s2.2-25.7-6.9-34.9l-128-128z"],screwdriver:[512,512,[129691],"f54a","M465 7c-8.5-8.5-22-9.4-31.6-2.1l-104 80c-5.9 4.5-9.4 11.6-9.4 19v54.1l-85.6 85.6c6.7 4.2 13 9.3 18.8 15.1s10.9 12.2 15.1 18.8L353.9 192H408c7.5 0 14.5-3.5 19-9.4l80-104c7.4-9.6 6.5-23.1-2.1-31.6L465 7zM121.4 281.4l-112 112c-12.5 12.5-12.5 32.8 0 45.3l64 64c12.5 12.5 32.8 12.5 45.3 0l112-112c30.2-30.2 30.2-79.1 0-109.3s-79.1-30.2-109.3 0z"],"folder-closed":[512,512,[],"e185","M448 480H64c-35.3 0-64-28.7-64-64V192H512V416c0 35.3-28.7 64-64 64zm64-320H0V96C0 60.7 28.7 32 64 32H192c20.1 0 39.1 9.5 51.2 25.6l19.2 25.6c6 8.1 15.5 12.8 25.6 12.8H448c35.3 0 64 28.7 64 64z"],"house-tsunami":[576,512,[],"e515","M80.8 136.5C104.9 93.8 152.6 64 209 64c16.9 0 33.1 2.7 48.2 7.7c16.8 5.5 34.9-3.6 40.4-20.4s-3.6-34.9-20.4-40.4C255.8 3.8 232.8 0 209 0C95.2 0 0 88 0 200c0 91.6 53.5 172.1 142.2 194.1c13.4 3.8 27.5 5.9 42.2 5.9c.7 0 1.4 0 2.1-.1c1.8 0 3.7 .1 5.5 .1l0 0c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.5-27.3-10.1-39.2-1.7l0 0C439.4 325.2 410.9 336 384 336c-27.5 0-55-10.6-77.5-26.1c-11.1-7.9-25.9-7.9-37 0c-22.4 15.5-49.9 26.1-77.4 26.1c0 0-.1 0-.1 0c-12.4 0-24-1.5-34.9-4.3C121.6 320.2 96 287 96 248c0-48.5 39.5-88 88.4-88c13.5 0 26.1 3 37.5 8.3c16 7.5 35.1 .6 42.5-15.5s.6-35.1-15.5-42.5C229.3 101.1 207.4 96 184.4 96c-40 0-76.4 15.4-103.6 40.5zm252-18.1c-8.1 6-12.8 15.5-12.8 25.6V265c1.6 1 3.3 2 4.8 3.1c18.4 12.7 39.6 20.3 59.2 20.3c19 0 41.2-7.9 59.2-20.3c23.8-16.7 55.8-15.3 78.1 3.4c10.6 8.8 24.2 15.6 37.3 18.6c5.8 1.4 11.2 3.4 16.2 6.2c.7-2.7 1.1-5.5 1.1-8.4l-.4-144c0-10-4.7-19.4-12.7-25.5l-95.5-72c-11.4-8.6-27.1-8.6-38.5 0l-96 72zM384 448c-27.5 0-55-10.6-77.5-26.1c-11.1-7.9-25.9-7.9-37 0C247 437.4 219.5 448 192 448c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C126.5 501.7 159 512 192 512c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.4-27.3-10.1-39.2-1.7l0 0C439.4 437.2 410.9 448 384 448z"],"square-nfi":[448,512,[],"e576","M0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm75.7 64.6C68.8 162.5 64 168.8 64 176V336c0 8.8 7.2 16 16 16s16-7.2 16-16V233.8l66.3 110.5c3.7 6.2 11.1 9.1 18 7.2s11.7-8.2 11.7-15.4V176c0-8.8-7.2-16-16-16s-16 7.2-16 16V278.2L93.7 167.8c-3.7-6.2-11.1-9.1-18-7.2zM224 176v64 96c0 8.8 7.2 16 16 16s16-7.2 16-16V256h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H256V192h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H240c-8.8 0-16 7.2-16 16zm160 0c0-8.8-7.2-16-16-16s-16 7.2-16 16V336c0 8.8 7.2 16 16 16s16-7.2 16-16V176z"],"arrow-up-from-ground-water":[576,512,[],"e4b5","M288 352c17.7 0 32-14.3 32-32V109.3l25.4 25.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-80-80c-12.5-12.5-32.8-12.5-45.3 0l-80 80c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L256 109.3V320c0 17.7 14.3 32 32 32zm-18.5 69.9C247 437.4 219.5 448 192 448c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C126.5 501.7 159 512 192 512c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.4-27.3-10.1-39.2-1.7l0 0C439.4 437.2 410.9 448 384 448c-27.5 0-55-10.6-77.5-26.1c-11.1-7.9-25.9-7.9-37 0zM192 192H48c-26.5 0-48 21.5-48 48V425c5.3-3.1 11.2-5.4 17.5-6.9c13.1-3.1 26.7-9.8 37.3-18.6c22.2-18.7 54.3-20.1 78.1-3.4c18 12.4 40.1 20.3 59.1 20.3V192zm384 48c0-26.5-21.5-48-48-48H384V416.5h0c19 0 41.2-7.9 59.2-20.3c23.8-16.7 55.8-15.3 78.1 3.4c10.6 8.8 24.2 15.6 37.3 18.6c6.3 1.5 12.1 3.8 17.5 6.9V240z"],"martini-glass":[512,512,[127864,"glass-martini-alt"],"f57b","M32 0C19.1 0 7.4 7.8 2.4 19.8s-2.2 25.7 6.9 34.9L224 269.3V448H160c-17.7 0-32 14.3-32 32s14.3 32 32 32h96 96c17.7 0 32-14.3 32-32s-14.3-32-32-32H288V269.3L502.6 54.6c9.2-9.2 11.9-22.9 6.9-34.9S492.9 0 480 0H32zM173.3 128l-64-64H402.7l-64 64H173.3z"],"rotate-left":[512,512,["rotate-back","rotate-backward","undo-alt"],"f2ea","M48.5 224H40c-13.3 0-24-10.7-24-24V72c0-9.7 5.8-18.5 14.8-22.2s19.3-1.7 26.2 5.2L98.6 96.6c87.6-86.5 228.7-86.2 315.8 1c87.5 87.5 87.5 229.3 0 316.8s-229.3 87.5-316.8 0c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0c62.5 62.5 163.8 62.5 226.3 0s62.5-163.8 0-226.3c-62.2-62.2-162.7-62.5-225.3-1L185 183c6.9 6.9 8.9 17.2 5.2 26.2s-12.5 14.8-22.2 14.8H48.5z"],"table-columns":[512,512,["columns"],"f0db","M0 96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm64 64V416H224V160H64zm384 0H288V416H448V160z"],lemon:[448,512,[127819],"f094","M448 96c0-35.3-28.7-64-64-64c-6.6 0-13 1-19 2.9c-22.5 7-48.1 14.9-71 9c-75.2-19.1-156.4 11-213.7 68.3S-7.2 250.8 11.9 326c5.8 22.9-2 48.4-9 71C1 403 0 409.4 0 416c0 35.3 28.7 64 64 64c6.6 0 13-1 19.1-2.9c22.5-7 48.1-14.9 71-9c75.2 19.1 156.4-11 213.7-68.3s87.5-138.5 68.3-213.7c-5.8-22.9 2-48.4 9-71c1.9-6 2.9-12.4 2.9-19.1zM212.5 127.4c-54.6 16-101.1 62.5-117.1 117.1C92.9 253 84 257.8 75.5 255.4S62.2 244 64.6 235.5c19.1-65.1 73.7-119.8 138.9-138.9c8.5-2.5 17.4 2.4 19.9 10.9s-2.4 17.4-10.9 19.9z"],"head-side-mask":[576,512,[],"e063","M32 224.2c0-22.2 3.2-43.6 9.2-63.9L262.2 321c-4 9.5-6.2 20-6.2 31V512H128c-17.7 0-32-14.3-32-32V407.3c0-16.7-6.9-32.5-17.1-45.8C48.6 322.4 32 274.1 32 224.2zm248.3 70.4L53 129.3C88.7 53 166.2 0 256 0h24c95.2 0 181.2 69.3 197.3 160.2c2.3 13 6.8 25.7 15.1 36l42 52.6c5.4 6.7 8.6 14.8 9.4 23.2H336c-21.7 0-41.3 8.6-55.7 22.6zM336 304H534l0 0h10l-19.7 64H368c-8.8 0-16 7.2-16 16s7.2 16 16 16H514.5l-9.8 32H368c-8.8 0-16 7.2-16 16s7.2 16 16 16H494.8l-.9 2.8c-8.3 26.9-33.1 45.2-61.2 45.2H288V352c0-14 6-26.7 15.6-35.4c0 0 0 0 0 0c8.5-7.8 19.9-12.6 32.4-12.6zm48-80a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],handshake:[640,512,[],"f2b5","M323.4 85.2l-96.8 78.4c-16.1 13-19.2 36.4-7 53.1c12.9 17.8 38 21.3 55.3 7.8l99.3-77.2c7-5.4 17-4.2 22.5 2.8s4.2 17-2.8 22.5l-20.9 16.2L512 316.8V128h-.7l-3.9-2.5L434.8 79c-15.3-9.8-33.2-15-51.4-15c-21.8 0-43 7.5-60 21.2zm22.8 124.4l-51.7 40.2C263 274.4 217.3 268 193.7 235.6c-22.2-30.5-16.6-73.1 12.7-96.8l83.2-67.3c-11.6-4.9-24.1-7.4-36.8-7.4C234 64 215.7 69.6 200 80l-72 48V352h28.2l91.4 83.4c19.6 17.9 49.9 16.5 67.8-3.1c5.5-6.1 9.2-13.2 11.1-20.6l17 15.6c19.5 17.9 49.9 16.6 67.8-2.9c4.5-4.9 7.8-10.6 9.9-16.5c19.4 13 45.8 10.3 62.1-7.5c17.9-19.5 16.6-49.9-2.9-67.8l-134.2-123zM16 128c-8.8 0-16 7.2-16 16V352c0 17.7 14.3 32 32 32H64c17.7 0 32-14.3 32-32V128H16zM48 320a16 16 0 1 1 0 32 16 16 0 1 1 0-32zM544 128V352c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V144c0-8.8-7.2-16-16-16H544zm32 208a16 16 0 1 1 32 0 16 16 0 1 1 -32 0z"],gem:[512,512,[128142],"f3a5","M116.7 33.8c4.5-6.1 11.7-9.8 19.3-9.8H376c7.6 0 14.8 3.6 19.3 9.8l112 152c6.8 9.2 6.1 21.9-1.5 30.4l-232 256c-4.5 5-11 7.9-17.8 7.9s-13.2-2.9-17.8-7.9l-232-256c-7.7-8.5-8.3-21.2-1.5-30.4l112-152zm38.5 39.8c-3.3 2.5-4.2 7-2.1 10.5l57.4 95.6L63.3 192c-4.1 .3-7.3 3.8-7.3 8s3.2 7.6 7.3 8l192 16c.4 0 .9 0 1.3 0l192-16c4.1-.3 7.3-3.8 7.3-8s-3.2-7.6-7.3-8L301.5 179.8l57.4-95.6c2.1-3.5 1.2-8.1-2.1-10.5s-7.9-2-10.7 1L256 172.2 165.9 74.6c-2.8-3-7.4-3.4-10.7-1z"],dolly:[576,512,["dolly-box"],"f472","M0 32C0 14.3 14.3 0 32 0h72.9c27.5 0 52 17.6 60.7 43.8L257.7 320c30.1 .5 56.8 14.9 74 37l202.1-67.4c16.8-5.6 34.9 3.5 40.5 20.2s-3.5 34.9-20.2 40.5L352 417.7c-.9 52.2-43.5 94.3-96 94.3c-53 0-96-43-96-96c0-30.8 14.5-58.2 37-75.8L104.9 64H32C14.3 64 0 49.7 0 32zM244.8 134.5c-5.5-16.8 3.7-34.9 20.5-40.3L311 79.4l19.8 60.9 60.9-19.8L371.8 59.6l45.7-14.8c16.8-5.5 34.9 3.7 40.3 20.5l49.4 152.2c5.5 16.8-3.7 34.9-20.5 40.3L334.5 307.2c-16.8 5.5-34.9-3.7-40.3-20.5L244.8 134.5z"],smoking:[640,512,[128684],"f48d","M448 32V43c0 38.2 15.2 74.8 42.2 101.8l21 21c21 21 32.8 49.5 32.8 79.2v11c0 17.7-14.3 32-32 32s-32-14.3-32-32V245c0-12.7-5.1-24.9-14.1-33.9l-21-21C405.9 151.1 384 98.1 384 43V32c0-17.7 14.3-32 32-32s32 14.3 32 32zM576 256V245c0-38.2-15.2-74.8-42.2-101.8l-21-21c-21-21-32.8-49.5-32.8-79.2V32c0-17.7 14.3-32 32-32s32 14.3 32 32V43c0 12.7 5.1 24.9 14.1 33.9l21 21c39 39 60.9 91.9 60.9 147.1v11c0 17.7-14.3 32-32 32s-32-14.3-32-32zM0 416c0-35.3 28.7-64 64-64H416c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H64c-35.3 0-64-28.7-64-64V416zm224 0v32H384V416H224zm288-64c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384c0-17.7 14.3-32 32-32zm96 0c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384c0-17.7 14.3-32 32-32z"],minimize:[512,512,["compress-arrows-alt"],"f78c","M456 224H312c-13.3 0-24-10.7-24-24V56c0-9.7 5.8-18.5 14.8-22.2s19.3-1.7 26.2 5.2l40 40L442.3 5.7C446 2 450.9 0 456 0s10 2 13.7 5.7l36.7 36.7C510 46 512 50.9 512 56s-2 10-5.7 13.7L433 143l40 40c6.9 6.9 8.9 17.2 5.2 26.2s-12.5 14.8-22.2 14.8zm0 64c9.7 0 18.5 5.8 22.2 14.8s1.7 19.3-5.2 26.2l-40 40 73.4 73.4c3.6 3.6 5.7 8.5 5.7 13.7s-2 10-5.7 13.7l-36.7 36.7C466 510 461.1 512 456 512s-10-2-13.7-5.7L369 433l-40 40c-6.9 6.9-17.2 8.9-26.2 5.2s-14.8-12.5-14.8-22.2V312c0-13.3 10.7-24 24-24H456zm-256 0c13.3 0 24 10.7 24 24V456c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-40-40L69.7 506.3C66 510 61.1 512 56 512s-10-2-13.7-5.7L5.7 469.7C2 466 0 461.1 0 456s2-10 5.7-13.7L79 369 39 329c-6.9-6.9-8.9-17.2-5.2-26.2s12.5-14.8 22.2-14.8H200zM56 224c-9.7 0-18.5-5.8-22.2-14.8s-1.7-19.3 5.2-26.2l40-40L5.7 69.7C2 66 0 61.1 0 56s2-10 5.7-13.7L42.3 5.7C46 2 50.9 0 56 0s10 2 13.7 5.7L143 79l40-40c6.9-6.9 17.2-8.9 26.2-5.2s14.8 12.5 14.8 22.2V200c0 13.3-10.7 24-24 24H56z"],monument:[384,512,[],"f5a6","M180.7 4.7c6.2-6.2 16.4-6.2 22.6 0l80 80c2.5 2.5 4.1 5.8 4.6 9.3l40.2 322H55.9L96.1 94c.4-3.5 2-6.8 4.6-9.3l80-80zM152 272c-13.3 0-24 10.7-24 24s10.7 24 24 24h80c13.3 0 24-10.7 24-24s-10.7-24-24-24H152zM32 448H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],snowplow:[640,512,[],"f7d2","M298.9 64l68.6 160H256l-64-64V64H298.9zM445.1 242.7l-87.4-204C347.6 15.3 324.5 0 298.9 0H176c-26.5 0-48 21.5-48 48V160H96c-17.7 0-32 14.3-32 32V298.8C26.2 316.8 0 355.3 0 400c0 61.9 50.1 112 112 112H368c61.9 0 112-50.1 112-112c0-17.2-3.9-33.5-10.8-48H512v50.7c0 17 6.7 33.3 18.7 45.3l54.6 54.6c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L576 402.7V320 235.2L633 164c11-13.8 8.8-33.9-5-45s-33.9-8.8-45 5l-57 71.2c-9.1 11.3-14 25.4-14 40V288H448V256.7c.1-2.4-.2-4.8-.6-7.1s-1.2-4.7-2.2-6.8zM368 352c26.5 0 48 21.5 48 48s-21.5 48-48 48H112c-26.5 0-48-21.5-48-48s21.5-48 48-48H368zM144 400a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zm216 24a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm-56-24a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zM200 424a24 24 0 1 0 0-48 24 24 0 1 0 0 48z"],"angles-right":[512,512,[187,"angle-double-right"],"f101","M470.6 278.6c12.5-12.5 12.5-32.8 0-45.3l-160-160c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L402.7 256 265.4 393.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l160-160zm-352 160l160-160c12.5-12.5 12.5-32.8 0-45.3l-160-160c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L210.7 256 73.4 393.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0z"],cannabis:[512,512,[],"f55f","M256 0c5.3 0 10.3 2.7 13.3 7.1c15.8 23.5 36.7 63.7 49.2 109c7.2 26.4 11.8 55.2 10.4 84c11.5-8.8 23.7-16.7 35.8-23.6c41-23.3 84.4-36.9 112.2-42.5c5.2-1 10.7 .6 14.4 4.4s5.4 9.2 4.4 14.5c-5.6 27.7-19.3 70.9-42.7 111.7c-9.1 15.9-19.9 31.7-32.4 46.3c27.8 6.6 52.4 17.3 67.2 25.5c5.1 2.8 8.2 8.2 8.2 14s-3.2 11.2-8.2 14c-15.2 8.4-40.9 19.5-69.8 26.1c-20.2 4.6-42.9 7.2-65.2 4.6l8.3 33.1c1.5 6.1-.6 12.4-5.5 16.4s-11.6 4.6-17.2 1.9L280 417.2V488c0 13.3-10.7 24-24 24s-24-10.7-24-24V417.2l-58.5 29.1c-5.6 2.8-12.3 2.1-17.2-1.9s-7-10.3-5.5-16.4l8.3-33.1c-22.2 2.6-45 0-65.2-4.6c-28.9-6.6-54.6-17.6-69.8-26.1c-5.1-2.8-8.2-8.2-8.2-14s3.2-11.2 8.2-14c14.8-8.2 39.4-18.8 67.2-25.5C78.9 296.3 68.1 280.5 59 264.6c-23.4-40.8-37.1-84-42.7-111.7c-1.1-5.2 .6-10.7 4.4-14.5s9.2-5.4 14.4-4.4c27.9 5.5 71.2 19.2 112.2 42.5c12.1 6.9 24.3 14.7 35.8 23.6c-1.4-28.7 3.1-57.6 10.4-84c12.5-45.3 33.4-85.5 49.2-109c3-4.4 8-7.1 13.3-7.1z"],"circle-play":[512,512,[61469,"play-circle"],"f144","M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM188.3 147.1c-7.6 4.2-12.3 12.3-12.3 20.9V344c0 8.7 4.7 16.7 12.3 20.9s16.8 4.1 24.3-.5l144-88c7.1-4.4 11.5-12.1 11.5-20.5s-4.4-16.1-11.5-20.5l-144-88c-7.4-4.5-16.7-4.7-24.3-.5z"],tablets:[640,512,[],"f490","M614.3 247c16.3-25 25.7-54.9 25.7-87C640 71.6 568.4 0 480 0c-32.1 0-61.9 9.4-87 25.7c-7.9 5.2-8.5 16.2-1.8 22.9L591.4 248.8c6.7 6.7 17.8 6.2 22.9-1.8zM567 294.3c7.9-5.2 8.5-16.2 1.8-22.9L368.6 71.2c-6.7-6.7-17.8-6.2-22.9 1.8c-16.3 25-25.7 54.9-25.7 87c0 88.4 71.6 160 160 160c32.1 0 61.9-9.4 87-25.7zM301.5 368H18.5c-9.5 0-16.9 8.2-15 17.5C18.9 457.8 83.1 512 160 512s141.1-54.2 156.5-126.5c2-9.3-5.5-17.5-15-17.5zm0-32c9.5 0 16.9-8.2 15-17.5C301.1 246.2 236.9 192 160 192S18.9 246.2 3.5 318.5c-2 9.3 5.5 17.5 15 17.5H301.5z"],ethernet:[512,512,[],"f796","M0 224V416c0 17.7 14.3 32 32 32H96V336c0-8.8 7.2-16 16-16s16 7.2 16 16V448h64V336c0-8.8 7.2-16 16-16s16 7.2 16 16V448h64V336c0-8.8 7.2-16 16-16s16 7.2 16 16V448h64V336c0-8.8 7.2-16 16-16s16 7.2 16 16V448h64c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32H448V160c0-17.7-14.3-32-32-32H384V96c0-17.7-14.3-32-32-32H160c-17.7 0-32 14.3-32 32v32H96c-17.7 0-32 14.3-32 32v32H32c-17.7 0-32 14.3-32 32z"],"euro-sign":[320,512,[8364,"eur","euro"],"f153","M48.1 240c-.1 2.7-.1 5.3-.1 8v16c0 2.7 0 5.3 .1 8H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H60.3C89.9 419.9 170 480 264 480h24c17.7 0 32-14.3 32-32s-14.3-32-32-32H264c-57.9 0-108.2-32.4-133.9-80H256c17.7 0 32-14.3 32-32s-14.3-32-32-32H112.2c-.1-2.6-.2-5.3-.2-8V248c0-2.7 .1-5.4 .2-8H256c17.7 0 32-14.3 32-32s-14.3-32-32-32H130.1c25.7-47.6 76-80 133.9-80h24c17.7 0 32-14.3 32-32s-14.3-32-32-32H264C170 32 89.9 92.1 60.3 176H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H48.1z"],chair:[448,512,[129681],"f6c0","M248 48V256h48V58.7c23.9 13.8 40 39.7 40 69.3V256h48V128C384 57.3 326.7 0 256 0H192C121.3 0 64 57.3 64 128V256h48V128c0-29.6 16.1-55.5 40-69.3V256h48V48h48zM48 288c-12.1 0-23.2 6.8-28.6 17.7l-16 32c-5 9.9-4.4 21.7 1.4 31.1S20.9 384 32 384l0 96c0 17.7 14.3 32 32 32s32-14.3 32-32V384H352v96c0 17.7 14.3 32 32 32s32-14.3 32-32V384c11.1 0 21.4-5.7 27.2-15.2s6.4-21.2 1.4-31.1l-16-32C423.2 294.8 412.1 288 400 288H48z"],"circle-check":[512,512,[61533,"check-circle"],"f058","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM369 209L241 337c-9.4 9.4-24.6 9.4-33.9 0l-64-64c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47L335 175c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z"],"circle-stop":[512,512,[62094,"stop-circle"],"f28d","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM192 160H320c17.7 0 32 14.3 32 32V320c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32V192c0-17.7 14.3-32 32-32z"],"compass-drafting":[512,512,["drafting-compass"],"f568","M352 96c0 14.3-3.1 27.9-8.8 40.2L396 227.4c-23.7 25.3-54.2 44.1-88.5 53.6L256 192h0 0l-68 117.5c21.5 6.8 44.3 10.5 68.1 10.5c70.7 0 133.8-32.7 174.9-84c11.1-13.8 31.2-16 45-5s16 31.2 5 45C428.1 341.8 347 384 256 384c-35.4 0-69.4-6.4-100.7-18.1L98.7 463.7C94 471.8 87 478.4 78.6 482.6L23.2 510.3c-5 2.5-10.9 2.2-15.6-.7S0 501.5 0 496V440.6c0-8.4 2.2-16.7 6.5-24.1l60-103.7C53.7 301.6 41.8 289.3 31.2 276c-11.1-13.8-8.8-33.9 5-45s33.9-8.8 45 5c5.7 7.1 11.8 13.8 18.2 20.1l69.4-119.9c-5.6-12.2-8.8-25.8-8.8-40.2c0-53 43-96 96-96s96 43 96 96zm21 297.9c32.6-12.8 62.5-30.8 88.9-52.9l43.7 75.5c4.2 7.3 6.5 15.6 6.5 24.1V496c0 5.5-2.9 10.7-7.6 13.6s-10.6 3.2-15.6 .7l-55.4-27.7c-8.4-4.2-15.4-10.8-20.1-18.9L373 393.9zM256 128a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"plate-wheat":[512,512,[],"e55a","M176 32c44.2 0 80 35.8 80 80v16c0 8.8-7.2 16-16 16c-44.2 0-80-35.8-80-80V48c0-8.8 7.2-16 16-16zM56 64h48c13.3 0 24 10.7 24 24s-10.7 24-24 24H56c-13.3 0-24-10.7-24-24s10.7-24 24-24zM24 136H136c13.3 0 24 10.7 24 24s-10.7 24-24 24H24c-13.3 0-24-10.7-24-24s10.7-24 24-24zm8 96c0-13.3 10.7-24 24-24h48c13.3 0 24 10.7 24 24s-10.7 24-24 24H56c-13.3 0-24-10.7-24-24zM272 48c0-8.8 7.2-16 16-16c44.2 0 80 35.8 80 80v16c0 8.8-7.2 16-16 16c-44.2 0-80-35.8-80-80V48zM400 32c44.2 0 80 35.8 80 80v16c0 8.8-7.2 16-16 16c-44.2 0-80-35.8-80-80V48c0-8.8 7.2-16 16-16zm80 160v16c0 44.2-35.8 80-80 80c-8.8 0-16-7.2-16-16V256c0-44.2 35.8-80 80-80c8.8 0 16 7.2 16 16zM352 176c8.8 0 16 7.2 16 16v16c0 44.2-35.8 80-80 80c-8.8 0-16-7.2-16-16V256c0-44.2 35.8-80 80-80zm-96 16v16c0 44.2-35.8 80-80 80c-8.8 0-16-7.2-16-16V256c0-44.2 35.8-80 80-80c8.8 0 16 7.2 16 16zM3.5 347.6C1.6 332.9 13 320 27.8 320H484.2c14.8 0 26.2 12.9 24.4 27.6C502.3 397.8 464.2 437 416 446v2c0 17.7-14.3 32-32 32H128c-17.7 0-32-14.3-32-32v-2c-48.2-9-86.3-48.2-92.5-98.4z"],icicles:[512,512,[],"f7ad","M75.8 304.8L1 35.7c-.7-2.5-1-5-1-7.5C0 12.6 12.6 0 28.2 0H482.4C498.8 0 512 13.2 512 29.6c0 1.6-.1 3.3-.4 4.9L434.6 496.1c-1.5 9.2-9.5 15.9-18.8 15.9c-9.2 0-17.1-6.6-18.7-15.6L336 160 307.2 303.9c-1.9 9.3-10.1 16.1-19.6 16.1c-9.2 0-17.2-6.2-19.4-15.1L240 192 210.6 368.2c-1.5 9.1-9.4 15.8-18.6 15.8s-17.1-6.7-18.6-15.8L144 192 115.9 304.3c-2.3 9.2-10.6 15.7-20.1 15.7c-9.3 0-17.5-6.2-20-15.2z"],"person-shelter":[512,512,[],"e54f","M271.9 4.2c-9.8-5.6-21.9-5.6-31.8 0l-224 128C6.2 137.9 0 148.5 0 160V480c0 17.7 14.3 32 32 32s32-14.3 32-32V178.6L256 68.9 448 178.6V480c0 17.7 14.3 32 32 32s32-14.3 32-32V160c0-11.5-6.2-22.1-16.1-27.8l-224-128zM256 208a40 40 0 1 0 0-80 40 40 0 1 0 0 80zm-8 280V400h16v88c0 13.3 10.7 24 24 24s24-10.7 24-24V313.5l26.9 49.9c6.3 11.7 20.8 16 32.5 9.8s16-20.8 9.8-32.5l-37.9-70.3c-15.3-28.5-45.1-46.3-77.5-46.3H246.2c-32.4 0-62.1 17.8-77.5 46.3l-37.9 70.3c-6.3 11.7-1.9 26.2 9.8 32.5s26.2 1.9 32.5-9.8L200 313.5V488c0 13.3 10.7 24 24 24s24-10.7 24-24z"],neuter:[384,512,[9906],"f22c","M80 176a112 112 0 1 1 224 0A112 112 0 1 1 80 176zM224 349.1c81.9-15 144-86.8 144-173.1C368 78.8 289.2 0 192 0S16 78.8 16 176c0 86.3 62.1 158.1 144 173.1V480c0 17.7 14.3 32 32 32s32-14.3 32-32V349.1z"],"id-badge":[384,512,[],"f2c1","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zm96 320h64c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80zm-32-96a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM144 64h96c8.8 0 16 7.2 16 16s-7.2 16-16 16H144c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],marker:[512,512,[],"f5a1","M481 31C445.1-4.8 386.9-4.8 351 31l-15 15L322.9 33C294.8 4.9 249.2 4.9 221.1 33L135 119c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0L255 66.9c9.4-9.4 24.6-9.4 33.9 0L302.1 80 186.3 195.7 316.3 325.7 481 161c35.9-35.9 35.9-94.1 0-129.9zM293.7 348.3L163.7 218.3 99.5 282.5c-48 48-80.8 109.2-94.1 175.8l-5 25c-1.6 7.9 .9 16 6.6 21.7s13.8 8.1 21.7 6.6l25-5c66.6-13.3 127.8-46.1 175.8-94.1l64.2-64.2z"],"face-laugh-beam":[512,512,[128513,"laugh-beam"],"f59a","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM96.8 314.1c-3.8-13.7 7.4-26.1 21.6-26.1H393.6c14.2 0 25.5 12.4 21.6 26.1C396.2 382 332.1 432 256 432s-140.2-50-159.2-117.9zM217.6 212.8l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z"],"helicopter-symbol":[512,512,[],"e502","M445.3 224H510C495.6 108.2 403.8 16.4 288 2V66.7C368.4 80.1 431.9 143.6 445.3 224zM510 288H445.3C431.9 368.4 368.4 431.9 288 445.4V510c115.8-14.4 207.6-106.2 222-222zM2 288C16.4 403.8 108.2 495.6 224 510V445.4C143.6 431.9 80.1 368.4 66.7 288H2zm0-64H66.7C80.1 143.6 143.6 80.1 224 66.7V2C108.2 16.4 16.4 108.2 2 224zm206-64c0-17.7-14.3-32-32-32s-32 14.3-32 32V352c0 17.7 14.3 32 32 32s32-14.3 32-32V288h96v64c0 17.7 14.3 32 32 32s32-14.3 32-32V160c0-17.7-14.3-32-32-32s-32 14.3-32 32v64H208V160z"],"universal-access":[512,512,[],"f29a","M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm161.5-86.1c-12.2-5.2-26.3 .4-31.5 12.6s.4 26.3 12.6 31.5l11.9 5.1c17.3 7.4 35.2 12.9 53.6 16.3v50.1c0 4.3-.7 8.6-2.1 12.6l-28.7 86.1c-4.2 12.6 2.6 26.2 15.2 30.4s26.2-2.6 30.4-15.2l24.4-73.2c1.3-3.8 4.8-6.4 8.8-6.4s7.6 2.6 8.8 6.4l24.4 73.2c4.2 12.6 17.8 19.4 30.4 15.2s19.4-17.8 15.2-30.4l-28.7-86.1c-1.4-4.1-2.1-8.3-2.1-12.6V235.5c18.4-3.5 36.3-8.9 53.6-16.3l11.9-5.1c12.2-5.2 17.8-19.3 12.6-31.5s-19.3-17.8-31.5-12.6L338.7 175c-26.1 11.2-54.2 17-82.7 17s-56.5-5.8-82.7-17l-11.9-5.1zM256 160a40 40 0 1 0 0-80 40 40 0 1 0 0 80z"],"circle-chevron-up":[512,512,["chevron-circle-up"],"f139","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM377 271c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-87-87-87 87c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9L239 167c9.4-9.4 24.6-9.4 33.9 0L377 271z"],"lari-sign":[384,512,[],"e1c8","M144 32c17.7 0 32 14.3 32 32V96.7c5.3-.4 10.6-.7 16-.7s10.7 .2 16 .7V64c0-17.7 14.3-32 32-32s32 14.3 32 32v49.4c54.9 25.2 95.8 75.5 108.2 136.2c3.5 17.3-7.7 34.2-25 37.7s-34.2-7.7-37.7-25c-6.1-29.9-22.5-55.9-45.4-74.3V256c0 17.7-14.3 32-32 32s-32-14.3-32-32V161c-5.2-.7-10.6-1-16-1s-10.8 .3-16 1v95c0 17.7-14.3 32-32 32s-32-14.3-32-32V188.1C82.7 211.5 64 247.6 64 288c0 70.7 57.3 128 128 128H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H192 32c-17.7 0-32-14.3-32-32s14.3-32 32-32H48.9C18.5 382 0 337.2 0 288c0-77.5 45.9-144.3 112-174.6V64c0-17.7 14.3-32 32-32z"],volcano:[512,512,[127755],"f770","M160 144c-35.3 0-64-28.7-64-64s28.7-64 64-64c15.7 0 30 5.6 41.2 15C212.4 12.4 232.7 0 256 0s43.6 12.4 54.8 31C322 21.6 336.3 16 352 16c35.3 0 64 28.7 64 64s-28.7 64-64 64c-14.7 0-28.3-5-39.1-13.3l-32 48C275.3 187 266 192 256 192s-19.3-5-24.9-13.3l-32-48C188.3 139 174.7 144 160 144zM144 352l48.4-24.2c10.2-5.1 21.6-7.8 33-7.8c19.6 0 38.4 7.8 52.2 21.6l32.5 32.5c6.3 6.3 14.9 9.9 23.8 9.9c11.3 0 21.8-5.6 28-15l9.7-14.6-58.9-66.3c-9.1-10.2-22.2-16.1-35.9-16.1H235.1c-13.7 0-26.8 5.9-35.9 16.1l-59.9 67.4L144 352zm19.4-95.8c18.2-20.5 44.3-32.2 71.8-32.2h41.8c27.4 0 53.5 11.7 71.8 32.2l150.2 169c8.5 9.5 13.2 21.9 13.2 34.7c0 28.8-23.4 52.2-52.2 52.2H52.2C23.4 512 0 488.6 0 459.8c0-12.8 4.7-25.1 13.2-34.7l150.2-169z"],"person-walking-dashed-line-arrow-right":[640,512,[],"e553","M208 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM123.7 200.5c1-.4 1.9-.8 2.9-1.2l-16.9 63.5c-5.6 21.1-.1 43.6 14.7 59.7l70.7 77.1 22 88.1c4.3 17.1 21.7 27.6 38.8 23.3s27.6-21.7 23.3-38.8l-23-92.1c-1.9-7.8-5.8-14.9-11.2-20.8l-49.5-54 19.3-65.5 9.6 23c4.4 10.6 12.5 19.3 22.8 24.5l26.7 13.3c15.8 7.9 35 1.5 42.9-14.3s1.5-35-14.3-42.9L281 232.7l-15.3-36.8C248.5 154.8 208.3 128 163.7 128c-22.8 0-45.3 4.8-66.1 14l-8 3.5c-32.9 14.6-58.1 42.4-69.4 76.5l-2.6 7.8c-5.6 16.8 3.5 34.9 20.2 40.5s34.9-3.5 40.5-20.2l2.6-7.8c5.7-17.1 18.3-30.9 34.7-38.2l8-3.5zm-30 135.1L68.7 398 9.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L116.3 441c4.6-4.6 8.2-10.1 10.6-16.1l14.5-36.2-40.7-44.4c-2.5-2.7-4.8-5.6-7-8.6zM550.6 153.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L530.7 224H384c-17.7 0-32 14.3-32 32s14.3 32 32 32H530.7l-25.4 25.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l80-80c12.5-12.5 12.5-32.8 0-45.3l-80-80zM392 0c-13.3 0-24 10.7-24 24V72c0 13.3 10.7 24 24 24s24-10.7 24-24V24c0-13.3-10.7-24-24-24zm24 152c0-13.3-10.7-24-24-24s-24 10.7-24 24v16c0 13.3 10.7 24 24 24s24-10.7 24-24V152zM392 320c-13.3 0-24 10.7-24 24v16c0 13.3 10.7 24 24 24s24-10.7 24-24V344c0-13.3-10.7-24-24-24zm24 120c0-13.3-10.7-24-24-24s-24 10.7-24 24v48c0 13.3 10.7 24 24 24s24-10.7 24-24V440z"],"sterling-sign":[320,512,[163,"gbp","pound-sign"],"f154","M112 160.4c0-35.5 28.8-64.4 64.4-64.4c6.9 0 13.8 1.1 20.4 3.3l81.2 27.1c16.8 5.6 34.9-3.5 40.5-20.2s-3.5-34.9-20.2-40.5L217 38.6c-13.1-4.4-26.8-6.6-40.6-6.6C105.5 32 48 89.5 48 160.4V224H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H48v44.5c0 17.4-4.7 34.5-13.7 49.4L4.6 431.5c-5.9 9.9-6.1 22.2-.4 32.2S20.5 480 32 480H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H88.5l.7-1.1C104.1 390 112 361.5 112 332.5V288H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H112V160.4z"],viruses:[640,512,[],"e076","M192 0c13.3 0 24 10.7 24 24V37.5c0 35.6 43.1 53.5 68.3 28.3l9.5-9.5c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-9.5 9.5C293 124.9 310.9 168 346.5 168H360c13.3 0 24 10.7 24 24s-10.7 24-24 24H346.5c-35.6 0-53.5 43.1-28.3 68.3l9.5 9.5c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-9.5-9.5C259.1 293 216 310.9 216 346.5V360c0 13.3-10.7 24-24 24s-24-10.7-24-24V346.5c0-35.6-43.1-53.5-68.3-28.3l-9.5 9.5c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l9.5-9.5C91 259.1 73.1 216 37.5 216H24c-13.3 0-24-10.7-24-24s10.7-24 24-24H37.5c35.6 0 53.5-43.1 28.3-68.3l-9.5-9.5c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l9.5 9.5C124.9 91 168 73.1 168 37.5V24c0-13.3 10.7-24 24-24zm48 224a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm-48-64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm320 80c0 33 39.9 49.5 63.2 26.2c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6C574.5 312.1 591 352 624 352c8.8 0 16 7.2 16 16s-7.2 16-16 16c-33 0-49.5 39.9-26.2 63.2c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0C551.9 446.5 512 463 512 496c0 8.8-7.2 16-16 16s-16-7.2-16-16c0-33-39.9-49.5-63.2-26.2c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6C417.5 423.9 401 384 368 384c-8.8 0-16-7.2-16-16s7.2-16 16-16c33 0 49.5-39.9 26.2-63.2c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0C440.1 289.5 480 273 480 240c0-8.8 7.2-16 16-16s16 7.2 16 16zm0 112a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z"],"square-person-confined":[448,512,[],"e577","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm96 112a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm80 104c0-30.9 25.1-56 56-56s56 25.1 56 56V350.1c0 36.4-29.5 65.9-65.9 65.9c-17.5 0-34.3-6.9-46.6-19.3L184.8 342l-28.1 56.3c-7.9 15.8-27.1 22.2-42.9 14.3s-22.2-27.1-14.3-42.9l48-96c4.6-9.2 13.3-15.6 23.5-17.3s20.5 1.7 27.8 9L240 306.7V248z"],"user-tie":[448,512,[],"f508","M96 128a128 128 0 1 0 256 0A128 128 0 1 0 96 128zm94.5 200.2l18.6 31L175.8 483.1l-36-146.9c-2-8.1-9.8-13.4-17.9-11.3C51.9 342.4 0 405.8 0 481.3c0 17 13.8 30.7 30.7 30.7H162.5c0 0 0 0 .1 0H168 280h5.5c0 0 0 0 .1 0H417.3c17 0 30.7-13.8 30.7-30.7c0-75.5-51.9-138.9-121.9-156.4c-8.1-2-15.9 3.3-17.9 11.3l-36 146.9L238.9 359.2l18.6-31c6.4-10.7-1.3-24.2-13.7-24.2H224 204.3c-12.4 0-20.1 13.6-13.7 24.2z"],"arrow-down-long":[384,512,["long-arrow-down"],"f175","M169.4 502.6c12.5 12.5 32.8 12.5 45.3 0l128-128c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L224 402.7 224 32c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 370.7L86.6 329.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l128 128z"],"tent-arrow-down-to-line":[640,512,[],"e57e","M241.8 111.9c8.9 9.9 8.1 25-1.8 33.9l-80 72c-9.1 8.2-23 8.2-32.1 0l-80-72c-9.9-8.9-10.7-24-1.8-33.9s24-10.7 33.9-1.8l39.9 36L120 24c0-13.3 10.7-24 24-24s24 10.7 24 24l0 122.1 39.9-36c9.9-8.9 25-8.1 33.9 1.8zm122.8 22.6c11.5-8.7 27.3-8.7 38.8 0l168 128c6.6 5 11 12.5 12.3 20.7l24 160 .7 4.7c17.5 .2 31.6 14.4 31.6 32c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H159.6l.7-4.7 24-160c1.2-8.2 5.6-15.7 12.3-20.7l168-128zM384 448h80L402.7 325.5c-1.7-3.4-5.1-5.5-8.8-5.5c-5.5 0-9.9 4.4-9.9 9.9V448z"],certificate:[512,512,[],"f0a3","M211 7.3C205 1 196-1.4 187.6 .8s-14.9 8.9-17.1 17.3L154.7 80.6l-62-17.5c-8.4-2.4-17.4 0-23.5 6.1s-8.5 15.1-6.1 23.5l17.5 62L18.1 170.6c-8.4 2.1-15 8.7-17.3 17.1S1 205 7.3 211l46.2 45L7.3 301C1 307-1.4 316 .8 324.4s8.9 14.9 17.3 17.1l62.5 15.8-17.5 62c-2.4 8.4 0 17.4 6.1 23.5s15.1 8.5 23.5 6.1l62-17.5 15.8 62.5c2.1 8.4 8.7 15 17.1 17.3s17.3-.2 23.4-6.4l45-46.2 45 46.2c6.1 6.2 15 8.7 23.4 6.4s14.9-8.9 17.1-17.3l15.8-62.5 62 17.5c8.4 2.4 17.4 0 23.5-6.1s8.5-15.1 6.1-23.5l-17.5-62 62.5-15.8c8.4-2.1 15-8.7 17.3-17.1s-.2-17.4-6.4-23.4l-46.2-45 46.2-45c6.2-6.1 8.7-15 6.4-23.4s-8.9-14.9-17.3-17.1l-62.5-15.8 17.5-62c2.4-8.4 0-17.4-6.1-23.5s-15.1-8.5-23.5-6.1l-62 17.5L341.4 18.1c-2.1-8.4-8.7-15-17.1-17.3S307 1 301 7.3L256 53.5 211 7.3z"],"reply-all":[576,512,["mail-reply-all"],"f122","M209.4 39.5c-9.1-9.6-24.3-10-33.9-.9L33.8 173.2c-19.9 18.9-19.9 50.7 0 69.6L175.5 377.4c9.6 9.1 24.8 8.7 33.9-.9s8.7-24.8-.9-33.9L66.8 208 208.5 73.4c9.6-9.1 10-24.3 .9-33.9zM352 64c0-12.6-7.4-24.1-19-29.2s-25-3-34.4 5.4l-160 144c-6.7 6.1-10.6 14.7-10.6 23.8s3.9 17.7 10.6 23.8l160 144c9.4 8.5 22.9 10.6 34.4 5.4s19-16.6 19-29.2V288h32c53 0 96 43 96 96c0 30.4-12.8 47.9-22.2 56.7c-5.5 5.1-9.8 12-9.8 19.5c0 10.9 8.8 19.7 19.7 19.7c2.8 0 5.6-.6 8.1-1.9C494.5 467.9 576 417.3 576 304c0-97.2-78.8-176-176-176H352V64z"],suitcase:[512,512,[129523],"f0f2","M176 56V96H336V56c0-4.4-3.6-8-8-8H184c-4.4 0-8 3.6-8 8zM128 96V56c0-30.9 25.1-56 56-56H328c30.9 0 56 25.1 56 56V96v32V480H128V128 96zM64 96H96V480H64c-35.3 0-64-28.7-64-64V160c0-35.3 28.7-64 64-64zM448 480H416V96h32c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64z"],"person-skating":[448,512,["skating"],"f7c5","M352 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM128 128c0-17.7 14.3-32 32-32H319.4c43.6 0 64.6 53.4 32.8 83.1l-74.4 69.4 60.2 60.2c9 9 14.1 21.2 14.1 33.9V416c0 17.7-14.3 32-32 32s-32-14.3-32-32V349.3l-77.9-77.8c-26.6-26.6-24.6-70.3 4.3-94.4l20.4-17H160c-17.7 0-32-14.3-32-32zM81.4 353.4l86.9-86.9c4.6 10 11 19.3 19.3 27.5l21.8 21.8-82.7 82.7c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3zm322.5 95.1c8.6 2.1 13.8 10.8 11.6 19.4l-.4 1.7c-6.2 24.9-28.6 42.4-54.3 42.4H272c-8.8 0-16-7.2-16-16s7.2-16 16-16h88.8c11 0 20.6-7.5 23.3-18.2l.4-1.7c2.1-8.6 10.8-13.8 19.4-11.6zM135.2 478.3l-6.2 3.1c-21.6 10.8-47.6 6.6-64.6-10.5L4.7 411.3c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0l59.6 59.6c7.3 7.3 18.5 9.1 27.7 4.5l6.2-3.1c7.9-4 17.5-.7 21.5 7.2s.7 17.5-7.2 21.5z"],"filter-circle-dollar":[576,512,["funnel-dollar"],"f662","M3.9 22.9C10.5 8.9 24.5 0 40 0H472c15.5 0 29.5 8.9 36.1 22.9s4.6 30.5-5.2 42.5L396.4 195.6C316.2 212.1 256 283 256 368c0 27.4 6.3 53.4 17.5 76.5c-1.6-.8-3.2-1.8-4.7-2.9l-64-48c-8.1-6-12.8-15.5-12.8-25.6V288.9L9 65.3C-.7 53.4-2.8 36.8 3.9 22.9zM288 368a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm120.8-32.6c.6-.9 1.8-2.1 4.2-3.4c5.1-2.7 12.5-4.1 18.7-4c8.2 .1 17.1 1.8 26.4 4.1c8.6 2.1 17.3-3.1 19.4-11.7s-3.1-17.3-11.7-19.4c-5.6-1.4-11.6-2.7-17.9-3.7V288c0-8.8-7.2-16-16-16s-16 7.2-16 16v9.5c-6.1 1.2-12.3 3.2-18 6.3c-11.8 6.3-23 18.4-21.8 37.2c1 16 11.7 25.3 21.6 30.7c8.8 4.7 19.7 7.8 28.6 10.3l1.8 .5c10.3 2.9 17.9 5.2 23.2 8.3c4.5 2.7 4.7 4.2 4.7 5.6c.1 2.4-.5 3.7-1 4.5c-.6 1-1.8 2.2-4 3.3c-4.7 2.5-11.8 3.8-18.5 3.6c-9.5-.3-18.5-3.1-29.9-6.8c-1.9-.6-3.8-1.2-5.8-1.8c-8.4-2.6-17.4 2.1-20 10.5s2.1 17.4 10.5 20c1.6 .5 3.3 1 5 1.6l0 0 0 0c7 2.3 15.1 4.8 23.7 6.6v11.4c0 8.8 7.2 16 16 16s16-7.2 16-16V438.7c6.2-1.1 12.5-3.1 18.3-6.2c12.1-6.5 22.3-18.7 21.7-36.9c-.5-16.2-10.3-26.3-20.5-32.3c-9.4-5.6-21.2-8.9-30.5-11.5l-.2 0c-10.4-2.9-18.3-5.2-23.9-8.2c-4.8-2.6-4.8-4-4.8-4.5l0-.1c-.1-1.9 .3-2.9 .8-3.6z"],"camera-retro":[512,512,[128247],"f083","M220.6 121.2L271.1 96 448 96v96H333.2c-21.9-15.1-48.5-24-77.2-24s-55.2 8.9-77.2 24H64V128H192c9.9 0 19.7-2.3 28.6-6.8zM0 128V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H271.1c-9.9 0-19.7 2.3-28.6 6.8L192 64H160V48c0-8.8-7.2-16-16-16H80c-8.8 0-16 7.2-16 16l0 16C28.7 64 0 92.7 0 128zM168 304a88 88 0 1 1 176 0 88 88 0 1 1 -176 0z"],"circle-arrow-down":[512,512,["arrow-circle-down"],"f0ab","M256 0a256 256 0 1 0 0 512A256 256 0 1 0 256 0zM127 281c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l71 71L232 136c0-13.3 10.7-24 24-24s24 10.7 24 24l0 182.1 71-71c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9L273 393c-9.4 9.4-24.6 9.4-33.9 0L127 281z"],"file-import":[512,512,["arrow-right-to-file"],"f56f","M128 64c0-35.3 28.7-64 64-64H352V128c0 17.7 14.3 32 32 32H512V448c0 35.3-28.7 64-64 64H192c-35.3 0-64-28.7-64-64V336H302.1l-39 39c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l80-80c9.4-9.4 9.4-24.6 0-33.9l-80-80c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l39 39H128V64zm0 224v48H24c-13.3 0-24-10.7-24-24s10.7-24 24-24H128zM512 128H384V0L512 128z"],"square-arrow-up-right":[448,512,["external-link-square"],"f14c","M384 32c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H384zM160 144c-13.3 0-24 10.7-24 24s10.7 24 24 24h94.1L119 327c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l135-135V328c0 13.3 10.7 24 24 24s24-10.7 24-24V168c0-13.3-10.7-24-24-24H160z"],"box-open":[640,512,[],"f49e","M58.9 42.1c3-6.1 9.6-9.6 16.3-8.7L320 64 564.8 33.4c6.7-.8 13.3 2.7 16.3 8.7l41.7 83.4c9 17.9-.6 39.6-19.8 45.1L439.6 217.3c-13.9 4-28.8-1.9-36.2-14.3L320 64 236.6 203c-7.4 12.4-22.3 18.3-36.2 14.3L37.1 170.6c-19.3-5.5-28.8-27.2-19.8-45.1L58.9 42.1zM321.1 128l54.9 91.4c14.9 24.8 44.6 36.6 72.5 28.6L576 211.6v167c0 22-15 41.2-36.4 46.6l-204.1 51c-10.2 2.6-20.9 2.6-31 0l-204.1-51C79 419.7 64 400.5 64 378.5v-167L191.6 248c27.8 8 57.6-3.8 72.5-28.6L318.9 128h2.2z"],scroll:[576,512,[128220],"f70e","M0 80v48c0 17.7 14.3 32 32 32H48 96V80c0-26.5-21.5-48-48-48S0 53.5 0 80zM112 32c10 13.4 16 30 16 48V384c0 35.3 28.7 64 64 64s64-28.7 64-64v-5.3c0-32.4 26.3-58.7 58.7-58.7H480V128c0-53-43-96-96-96H112zM464 480c61.9 0 112-50.1 112-112c0-8.8-7.2-16-16-16H314.7c-14.7 0-26.7 11.9-26.7 26.7V384c0 53-43 96-96 96H368h96z"],spa:[576,512,[],"f5bb","M183.1 235.3c33.7 20.7 62.9 48.1 85.8 80.5c7 9.9 13.4 20.3 19.1 31c5.7-10.8 12.1-21.1 19.1-31c22.9-32.4 52.1-59.8 85.8-80.5C437.6 207.8 490.1 192 546 192h9.9c11.1 0 20.1 9 20.1 20.1C576 360.1 456.1 480 308.1 480H288 267.9C119.9 480 0 360.1 0 212.1C0 201 9 192 20.1 192H30c55.9 0 108.4 15.8 153.1 43.3zM301.5 37.6c15.7 16.9 61.1 71.8 84.4 164.6c-38 21.6-71.4 50.8-97.9 85.6c-26.5-34.8-59.9-63.9-97.9-85.6c23.2-92.8 68.6-147.7 84.4-164.6C278 33.9 282.9 32 288 32s10 1.9 13.5 5.6z"],"location-pin-lock":[512,512,[],"e51f","M215.7 499.2c11-13.8 25.1-31.7 40.3-52.3V352c0-23.7 12.9-44.4 32-55.4V272c0-55.6 40.5-101.7 93.6-110.5C367 70 287.7 0 192 0C86 0 0 86 0 192c0 87.4 117 243 168.3 307.2c12.3 15.3 35.1 15.3 47.4 0zM192 128a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM400 240c17.7 0 32 14.3 32 32v48H368V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H480c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z"],pause:[320,512,[9208],"f04c","M48 64C21.5 64 0 85.5 0 112V400c0 26.5 21.5 48 48 48H80c26.5 0 48-21.5 48-48V112c0-26.5-21.5-48-48-48H48zm192 0c-26.5 0-48 21.5-48 48V400c0 26.5 21.5 48 48 48h32c26.5 0 48-21.5 48-48V112c0-26.5-21.5-48-48-48H240z"],"hill-avalanche":[576,512,[],"e507","M439.7 401.9c34.2 23.1 81.1 19.5 111.4-10.8c34.4-34.4 34.4-90.1 0-124.4c-27.8-27.8-69.5-33.1-102.6-16c-11.8 6.1-16.4 20.6-10.3 32.3s20.6 16.4 32.3 10.3c15.1-7.8 34-5.3 46.6 7.3c15.6 15.6 15.6 40.9 0 56.6s-40.9 15.6-56.6 0l-81.7-81.7C401.2 261.3 416 236.4 416 208c0-33.9-21.1-62.9-50.9-74.5c1.9-6.8 2.9-14 2.9-21.5c0-44.2-35.8-80-80-80c-27.3 0-51.5 13.7-65.9 34.6C216.3 46.6 197.9 32 176 32c-26.5 0-48 21.5-48 48c0 4 .5 7.9 1.4 11.6L439.7 401.9zM480 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm0 128a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM68.3 87C43.1 61.8 0 79.7 0 115.3V432c0 44.2 35.8 80 80 80H396.7c35.6 0 53.5-43.1 28.3-68.3L68.3 87z"],"temperature-empty":[320,512,["temperature-0","thermometer-0","thermometer-empty"],"f2cb","M112 112c0-26.5 21.5-48 48-48s48 21.5 48 48V276.5c0 17.3 7.1 31.9 15.3 42.5C233.8 332.6 240 349.5 240 368c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-18.5 6.2-35.4 16.7-48.9c8.2-10.6 15.3-25.2 15.3-42.5V112zM160 0C98.1 0 48 50.2 48 112V276.5c0 .1-.1 .3-.2 .6c-.2 .6-.8 1.6-1.7 2.8C27.2 304.2 16 334.8 16 368c0 79.5 64.5 144 144 144s144-64.5 144-144c0-33.2-11.2-63.8-30.1-88.1c-.9-1.2-1.5-2.2-1.7-2.8c-.1-.3-.2-.5-.2-.6V112C272 50.2 221.9 0 160 0zm0 416a48 48 0 1 0 0-96 48 48 0 1 0 0 96z"],bomb:[512,512,[128163],"f1e2","M459.1 52.4L442.6 6.5C440.7 2.6 436.5 0 432.1 0s-8.5 2.6-10.4 6.5L405.2 52.4l-46 16.8c-4.3 1.6-7.3 5.9-7.2 10.4c0 4.5 3 8.7 7.2 10.2l45.7 16.8 16.8 45.8c1.5 4.4 5.8 7.5 10.4 7.5s8.9-3.1 10.4-7.5l16.5-45.8 45.7-16.8c4.2-1.5 7.2-5.7 7.2-10.2c0-4.6-3-8.9-7.2-10.4L459.1 52.4zm-132.4 53c-12.5-12.5-32.8-12.5-45.3 0l-2.9 2.9C256.5 100.3 232.7 96 208 96C93.1 96 0 189.1 0 304S93.1 512 208 512s208-93.1 208-208c0-24.7-4.3-48.5-12.2-70.5l2.9-2.9c12.5-12.5 12.5-32.8 0-45.3l-80-80zM200 192c-57.4 0-104 46.6-104 104v8c0 8.8-7.2 16-16 16s-16-7.2-16-16v-8c0-75.1 60.9-136 136-136h8c8.8 0 16 7.2 16 16s-7.2 16-16 16h-8z"],registered:[512,512,[174],"f25d","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM160 152c0-13.3 10.7-24 24-24h88c44.2 0 80 35.8 80 80c0 28-14.4 52.7-36.3 67l34.1 75.1c5.5 12.1 .1 26.3-11.9 31.8s-26.3 .1-31.8-11.9L268.9 288H208v72c0 13.3-10.7 24-24 24s-24-10.7-24-24V264 152zm48 88h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H208v64z"],"address-card":[576,512,[62140,"contact-card","vcard"],"f2bb","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm80 256h64c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80zm-32-96a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zm256-32H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H368c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H368c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H368c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],"scale-unbalanced-flip":[640,512,["balance-scale-right"],"f516","M117.9 62.4c-16.8-5.6-25.8-23.7-20.2-40.5s23.7-25.8 40.5-20.2l113 37.7C265 15.8 290.7 0 320 0c44.2 0 80 35.8 80 80c0 3-.2 5.9-.5 8.8l122.6 40.9c16.8 5.6 25.8 23.7 20.2 40.5s-23.7 25.8-40.5 20.2L366.4 145.2c-4.5 3.2-9.3 5.9-14.4 8.2V480c0 17.7-14.3 32-32 32H128c-17.7 0-32-14.3-32-32s14.3-32 32-32H288V153.3c-21-9.2-37.2-27-44.2-49l-125.9-42zM200.4 288L128 163.8 55.6 288H200.4zM128 384C65.1 384 12.8 350 2 305.1c-2.6-11 1-22.3 6.7-32.1l95.2-163.2c5-8.6 14.2-13.8 24.1-13.8s19.1 5.3 24.1 13.8l95.2 163.2c5.7 9.8 9.3 21.1 6.7 32.1C243.2 350 190.9 384 128 384zm382.8-92.2L438.4 416H583.3L510.8 291.8zm126 141.3C626 478 573.7 512 510.8 512s-115.2-34-126-78.9c-2.6-11 1-22.3 6.7-32.1l95.2-163.2c5-8.6 14.2-13.8 24.1-13.8s19.1 5.3 24.1 13.8l95.2 163.2c5.7 9.8 9.3 21.1 6.7 32.1z"],subscript:[512,512,[],"f12c","M32 64C14.3 64 0 78.3 0 96s14.3 32 32 32H47.3l89.6 128L47.3 384H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64c10.4 0 20.2-5.1 26.2-13.6L176 311.8l85.8 122.6c6 8.6 15.8 13.6 26.2 13.6h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H304.7L215.1 256l89.6-128H320c17.7 0 32-14.3 32-32s-14.3-32-32-32H288c-10.4 0-20.2 5.1-26.2 13.6L176 200.2 90.2 77.6C84.2 69.1 74.4 64 64 64H32zM480 320c0-11.1-5.7-21.4-15.2-27.2s-21.2-6.4-31.1-1.4l-32 16c-15.8 7.9-22.2 27.1-14.3 42.9C393 361.5 404.3 368 416 368v80c-17.7 0-32 14.3-32 32s14.3 32 32 32h32 32c17.7 0 32-14.3 32-32s-14.3-32-32-32V320z"],"diamond-turn-right":[512,512,["directions"],"f5eb","M227.7 11.7c15.6-15.6 40.9-15.6 56.6 0l216 216c15.6 15.6 15.6 40.9 0 56.6l-216 216c-15.6 15.6-40.9 15.6-56.6 0l-216-216c-15.6-15.6-15.6-40.9 0-56.6l216-216zm87.6 137c-4.6-4.6-11.5-5.9-17.4-3.5s-9.9 8.3-9.9 14.8v56H224c-35.3 0-64 28.7-64 64v48c0 13.3 10.7 24 24 24s24-10.7 24-24V280c0-8.8 7.2-16 16-16h64v56c0 6.5 3.9 12.3 9.9 14.8s12.9 1.1 17.4-3.5l80-80c6.2-6.2 6.2-16.4 0-22.6l-80-80z"],burst:[512,512,[],"e4dc","M37.6 4.2C28-2.3 15.2-1.1 7 7s-9.4 21-2.8 30.5l112 163.3L16.6 233.2C6.7 236.4 0 245.6 0 256s6.7 19.6 16.6 22.8l103.1 33.4L66.8 412.8c-4.9 9.3-3.2 20.7 4.3 28.1s18.8 9.2 28.1 4.3l100.6-52.9 33.4 103.1c3.2 9.9 12.4 16.6 22.8 16.6s19.6-6.7 22.8-16.6l33.4-103.1 100.6 52.9c9.3 4.9 20.7 3.2 28.1-4.3s9.2-18.8 4.3-28.1L392.3 312.2l103.1-33.4c9.9-3.2 16.6-12.4 16.6-22.8s-6.7-19.6-16.6-22.8L388.9 198.7l25.7-70.4c3.2-8.8 1-18.6-5.6-25.2s-16.4-8.8-25.2-5.6l-70.4 25.7L278.8 16.6C275.6 6.7 266.4 0 256 0s-19.6 6.7-22.8 16.6l-32.3 99.6L37.6 4.2z"],"house-laptop":[640,512,["laptop-house"],"e066","M218.3 8.5c12.3-11.3 31.2-11.3 43.4 0l208 192c6.7 6.2 10.3 14.8 10.3 23.5H336c-19.1 0-36.3 8.4-48 21.7V208c0-8.8-7.2-16-16-16H208c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h64V416H112c-26.5 0-48-21.5-48-48V256H32c-13.2 0-25-8.1-29.8-20.3s-1.6-26.2 8.1-35.2l208-192zM352 304V448H544V304H352zm-48-16c0-17.7 14.3-32 32-32H560c17.7 0 32 14.3 32 32V448h32c8.8 0 16 7.2 16 16c0 26.5-21.5 48-48 48H544 352 304c-26.5 0-48-21.5-48-48c0-8.8 7.2-16 16-16h32V288z"],"face-tired":[512,512,[128555,"tired"],"f5c8","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM164.7 328.7c22-22 53.9-40.7 91.3-40.7s69.3 18.7 91.3 40.7c11.1 11.1 20.1 23.4 26.4 35.4c6.2 11.7 10.3 24.4 10.3 35.9c0 5.2-2.6 10.2-6.9 13.2s-9.8 3.7-14.7 1.8l-20.5-7.7c-26.9-10.1-55.5-15.3-84.3-15.3h-3.2c-28.8 0-57.3 5.2-84.3 15.3L149.6 415c-4.9 1.8-10.4 1.2-14.7-1.8s-6.9-7.9-6.9-13.2c0-11.6 4.2-24.2 10.3-35.9c6.3-12 15.3-24.3 26.4-35.4zm-31.2-182l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6c0-9 9.6-14.7 17.5-10.5zM396 157.1c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9c7.9-4.2 17.5 1.5 17.5 10.5z"],"money-bills":[640,512,[],"e1f3","M96 96V320c0 35.3 28.7 64 64 64H576c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H160c-35.3 0-64 28.7-64 64zm64 160c35.3 0 64 28.7 64 64H160V256zM224 96c0 35.3-28.7 64-64 64V96h64zM576 256v64H512c0-35.3 28.7-64 64-64zM512 96h64v64c-35.3 0-64-28.7-64-64zM288 208a80 80 0 1 1 160 0 80 80 0 1 1 -160 0zM48 120c0-13.3-10.7-24-24-24S0 106.7 0 120V360c0 66.3 53.7 120 120 120H520c13.3 0 24-10.7 24-24s-10.7-24-24-24H120c-39.8 0-72-32.2-72-72V120z"],smog:[640,512,[],"f75f","M32 144c0 79.5 64.5 144 144 144H299.3c22.6 19.9 52.2 32 84.7 32s62.1-12.1 84.7-32H496c61.9 0 112-50.1 112-112s-50.1-112-112-112c-10.7 0-21 1.5-30.8 4.3C443.8 27.7 401.1 0 352 0c-32.6 0-62.4 12.2-85.1 32.3C242.1 12.1 210.5 0 176 0C96.5 0 32 64.5 32 144zM616 368H280c-13.3 0-24 10.7-24 24s10.7 24 24 24H616c13.3 0 24-10.7 24-24s-10.7-24-24-24zm-64 96H440c-13.3 0-24 10.7-24 24s10.7 24 24 24H552c13.3 0 24-10.7 24-24s-10.7-24-24-24zm-192 0H24c-13.3 0-24 10.7-24 24s10.7 24 24 24H360c13.3 0 24-10.7 24-24s-10.7-24-24-24zM224 392c0-13.3-10.7-24-24-24H96c-13.3 0-24 10.7-24 24s10.7 24 24 24H200c13.3 0 24-10.7 24-24z"],crutch:[512,512,[],"f7f7","M297.4 9.4c-12.5 12.5-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-160-160c-12.5-12.5-32.8-12.5-45.3 0zm-96 144l-34.8 34.8c-12.9 12.9-21.9 29.2-25.8 47.1L116.8 342.9c-1.3 5.9-4.3 11.4-8.6 15.7L9.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l98.8-98.8c4.3-4.3 9.7-7.3 15.7-8.6l107.6-23.9c17.8-4 34.1-12.9 47.1-25.8l34.7-34.7c0 0 .1-.1 .1-.1s.1-.1 .1-.1l74.6-74.6-45.3-45.3L336 242.7 269.3 176l52.1-52.1L276.1 78.6l-74.7 74.7zM224 221.3L290.7 288l-12.2 12.2c-4.3 4.3-9.7 7.3-15.7 8.6l-76.7 17 17-76.7c1.3-5.9 4.3-11.4 8.6-15.7L224 221.3z"],"font-awesome":[512,512,[62501,62694,"font-awesome-flag","font-awesome-logo-full"],"f2b4","M91.7 96C106.3 86.8 116 70.5 116 52C116 23.3 92.7 0 64 0S12 23.3 12 52c0 16.7 7.8 31.5 20 41l0 3 0 352 0 64 64 0 0-64 373.6 0c14.6 0 26.4-11.8 26.4-26.4c0-3.7-.8-7.3-2.3-10.7L432 272l61.7-138.9c1.5-3.4 2.3-7 2.3-10.7c0-14.6-11.8-26.4-26.4-26.4L91.7 96z"],"cloud-arrow-up":[640,512,[62338,"cloud-upload","cloud-upload-alt"],"f0ee","M144 480C64.5 480 0 415.5 0 336c0-62.8 40.2-116.2 96.2-135.9c-.1-2.7-.2-5.4-.2-8.1c0-88.4 71.6-160 160-160c59.3 0 111 32.2 138.7 80.2C409.9 102 428.3 96 448 96c53 0 96 43 96 96c0 12.2-2.3 23.8-6.4 34.6C596 238.4 640 290.1 640 352c0 70.7-57.3 128-128 128H144zm79-217c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l39-39V392c0 13.3 10.7 24 24 24s24-10.7 24-24V257.9l39 39c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-80-80c-9.4-9.4-24.6-9.4-33.9 0l-80 80z"],palette:[512,512,[127912],"f53f","M512 256c0 .9 0 1.8 0 2.7c-.4 36.5-33.6 61.3-70.1 61.3H344c-26.5 0-48 21.5-48 48c0 3.4 .4 6.7 1 9.9c2.1 10.2 6.5 20 10.8 29.9c6.1 13.8 12.1 27.5 12.1 42c0 31.8-21.6 60.7-53.4 62c-3.5 .1-7 .2-10.6 .2C114.6 512 0 397.4 0 256S114.6 0 256 0S512 114.6 512 256zM128 288a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm0-96a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM288 96a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm96 96a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"arrows-turn-right":[448,512,[],"e4c0","M297.4 9.4c12.5-12.5 32.8-12.5 45.3 0l96 96c12.5 12.5 12.5 32.8 0 45.3l-96 96c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L338.7 160H128c-35.3 0-64 28.7-64 64v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V224C0 153.3 57.3 96 128 96H338.7L297.4 54.6c-12.5-12.5-12.5-32.8 0-45.3zm-96 256c12.5-12.5 32.8-12.5 45.3 0l96 96c12.5 12.5 12.5 32.8 0 45.3l-96 96c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L242.7 416H96c-17.7 0-32 14.3-32 32v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448c0-53 43-96 96-96H242.7l-41.4-41.4c-12.5-12.5-12.5-32.8 0-45.3z"],vest:[448,512,[],"e085","M207.1 237.4L151.2 69.7C168.6 79.7 192.6 88 224 88s55.4-8.3 72.8-18.3L226.5 280.6c-1.6 4.9-2.5 10-2.5 15.2V464c0 26.5 21.5 48 48 48H400c26.5 0 48-21.5 48-48V270.5c0-9.5-2.8-18.7-8.1-26.6l-47.9-71.8c-5.3-7.9-8.1-17.1-8.1-26.6V128 54.3 48c0-26.5-21.5-48-48-48h-4.5c-.2 0-.4 0-.6 0c-.4 0-.8 0-1.2 0C311 0 295.7 9.7 285.7 18.8C276.4 27.2 257.2 40 224 40s-52.4-12.8-61.7-21.2C152.3 9.7 137 0 118.3 0c-.4 0-.8 0-1.2 0c-.2 0-.4 0-.6 0H112C85.5 0 64 21.5 64 48v6.3V128v17.5c0 9.5-2.8 18.7-8.1 26.6L8.1 243.9C2.8 251.8 0 261.1 0 270.5V464c0 26.5 21.5 48 48 48H176c9.9 0 19-3 26.7-8.1C195.9 492.2 192 478.5 192 464V295.8c0-8.6 1.4-17.1 4.1-25.3l11-33.1zM347.3 356.7l48 48c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0l-48-48c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0zm-294.6 48l48-48c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6l-48 48c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6z"],ferry:[576,512,[],"e4ea","M224 0H352c17.7 0 32 14.3 32 32h75.1c20.6 0 31.6 24.3 18.1 39.8L456 96H120L98.8 71.8C85.3 56.3 96.3 32 116.9 32H192c0-17.7 14.3-32 32-32zM96 128H480c17.7 0 32 14.3 32 32V283.5c0 13.3-4.2 26.3-11.9 37.2l-51.4 71.9c-1.9 1.1-3.7 2.2-5.5 3.5c-15.5 10.7-34 18-51 19.9H375.6c-17.1-1.8-35-9-50.8-19.9c-22.1-15.5-51.6-15.5-73.7 0c-14.8 10.2-32.5 18-50.6 19.9H183.9c-17-1.8-35.6-9.2-51-19.9c-1.8-1.3-3.7-2.4-5.6-3.5L75.9 320.7C68.2 309.8 64 296.8 64 283.5V160c0-17.7 14.3-32 32-32zm32 64v96H448V192H128zM306.5 421.9C329 437.4 356.5 448 384 448c26.9 0 55.3-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 501.7 417 512 384 512c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 437.2 165.1 448 192 448c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z"],"arrows-down-to-people":[640,512,[],"e4b9","M144 0c-13.3 0-24 10.7-24 24V142.1L97 119c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l64 64c9.4 9.4 24.6 9.4 33.9 0l64-64c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-23 23V24c0-13.3-10.7-24-24-24zM360 200a40 40 0 1 0 -80 0 40 40 0 1 0 80 0zM184 296a40 40 0 1 0 -80 0 40 40 0 1 0 80 0zm312 40a40 40 0 1 0 0-80 40 40 0 1 0 0 80zM200 441.5l26.9 49.9c6.3 11.7 20.8 16 32.5 9.8s16-20.8 9.8-32.5l-36.3-67.5c1.7-1.7 3.2-3.6 4.3-5.8L264 345.5V400c0 17.7 14.3 32 32 32h48c17.7 0 32-14.3 32-32V345.5l26.9 49.9c1.2 2.2 2.6 4.1 4.3 5.8l-36.3 67.5c-6.3 11.7-1.9 26.2 9.8 32.5s26.2 1.9 32.5-9.8L440 441.5V480c0 17.7 14.3 32 32 32h48c17.7 0 32-14.3 32-32V441.5l26.9 49.9c6.3 11.7 20.8 16 32.5 9.8s16-20.8 9.8-32.5l-37.9-70.3c-15.3-28.5-45.1-46.3-77.5-46.3H486.2c-16.3 0-31.9 4.5-45.4 12.6l-33.6-62.3c-15.3-28.5-45.1-46.3-77.5-46.3H310.2c-32.4 0-62.1 17.8-77.5 46.3l-33.6 62.3c-13.5-8.1-29.1-12.6-45.4-12.6H134.2c-32.4 0-62.1 17.8-77.5 46.3L18.9 468.6c-6.3 11.7-1.9 26.2 9.8 32.5s26.2 1.9 32.5-9.8L88 441.5V480c0 17.7 14.3 32 32 32h48c17.7 0 32-14.3 32-32V441.5zM415 153l64 64c9.4 9.4 24.6 9.4 33.9 0l64-64c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-23 23V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V142.1l-23-23c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9z"],seedling:[512,512,[127793,"sprout"],"f4d8","M512 32c0 113.6-84.6 207.5-194.2 222c-7.1-53.4-30.6-101.6-65.3-139.3C290.8 46.3 364 0 448 0h32c17.7 0 32 14.3 32 32zM0 96C0 78.3 14.3 64 32 64H64c123.7 0 224 100.3 224 224v32V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V320C100.3 320 0 219.7 0 96z"],"left-right":[512,512,[8596,"arrows-alt-h"],"f337","M504.3 273.6c4.9-4.5 7.7-10.9 7.7-17.6s-2.8-13-7.7-17.6l-112-104c-7-6.5-17.2-8.2-25.9-4.4s-14.4 12.5-14.4 22l0 56-192 0 0-56c0-9.5-5.7-18.2-14.4-22s-18.9-2.1-25.9 4.4l-112 104C2.8 243 0 249.3 0 256s2.8 13 7.7 17.6l112 104c7 6.5 17.2 8.2 25.9 4.4s14.4-12.5 14.4-22l0-56 192 0 0 56c0 9.5 5.7 18.2 14.4 22s18.9 2.1 25.9-4.4l112-104z"],"boxes-packing":[640,512,[],"e4c7","M256 48c0-26.5 21.5-48 48-48H592c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H381.3c1.8-5 2.7-10.4 2.7-16V253.3c18.6-6.6 32-24.4 32-45.3V176c0-26.5-21.5-48-48-48H256V48zM571.3 347.3c6.2-6.2 6.2-16.4 0-22.6l-64-64c-6.2-6.2-16.4-6.2-22.6 0l-64 64c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L480 310.6V432c0 8.8 7.2 16 16 16s16-7.2 16-16V310.6l36.7 36.7c6.2 6.2 16.4 6.2 22.6 0zM0 176c0-8.8 7.2-16 16-16H368c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H16c-8.8 0-16-7.2-16-16V176zm352 80V480c0 17.7-14.3 32-32 32H64c-17.7 0-32-14.3-32-32V256H352zM144 320c-8.8 0-16 7.2-16 16s7.2 16 16 16h96c8.8 0 16-7.2 16-16s-7.2-16-16-16H144z"],"circle-arrow-left":[512,512,["arrow-circle-left"],"f0a8","M512 256A256 256 0 1 0 0 256a256 256 0 1 0 512 0zM231 127c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-71 71L376 232c13.3 0 24 10.7 24 24s-10.7 24-24 24l-182.1 0 71 71c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L119 273c-9.4-9.4-9.4-24.6 0-33.9L231 127z"],"group-arrows-rotate":[512,512,[],"e4f6","M201.1 71.9c16.9-5 26.6-22.9 21.5-39.8s-22.9-26.6-39.8-21.5c-21.5 6.4-41.8 15.5-60.6 27C114.3 34 105.4 32 96 32C60.7 32 32 60.7 32 96c0 9.4 2 18.3 5.6 26.3c-11.5 18.7-20.6 39-27 60.6c-5 16.9 4.6 34.8 21.5 39.8s34.8-4.6 39.8-21.5c4.3-14.6 10.4-28.5 17.9-41.4c2 .2 4.1 .3 6.1 .3c35.3 0 64-28.7 64-64c0-2.1-.1-4.1-.3-6.1c12.9-7.5 26.8-13.6 41.4-17.9zm128-61.3c-16.9-5-34.8 4.6-39.8 21.5s4.6 34.8 21.5 39.8c14.6 4.3 28.5 10.4 41.4 17.9c-.2 2-.3 4.1-.3 6.1c0 35.3 28.7 64 64 64c2.1 0 4.1-.1 6.2-.3c7.5 12.9 13.6 26.8 17.9 41.4c5 16.9 22.9 26.6 39.8 21.5s26.6-22.9 21.5-39.8c-6.4-21.5-15.5-41.8-27-60.6c3.6-8 5.6-16.9 5.6-26.3c0-35.3-28.7-64-64-64c-9.4 0-18.3 2-26.3 5.6c-18.7-11.5-39-20.6-60.6-27zM71.9 310.9c-5-16.9-22.9-26.6-39.8-21.5s-26.6 22.9-21.5 39.8c6.4 21.5 15.5 41.8 27 60.6C34 397.7 32 406.6 32 416c0 35.3 28.7 64 64 64c9.4 0 18.3-2 26.3-5.6c18.7 11.5 39 20.6 60.6 27c16.9 5 34.8-4.6 39.8-21.5s-4.6-34.8-21.5-39.8c-14.6-4.3-28.5-10.4-41.4-17.9c.2-2 .3-4.1 .3-6.2c0-35.3-28.7-64-64-64c-2.1 0-4.1 .1-6.2 .3c-7.5-12.9-13.6-26.8-17.9-41.4zm429.4 18.3c5-16.9-4.6-34.8-21.5-39.8s-34.8 4.6-39.8 21.5c-4.3 14.6-10.4 28.5-17.9 41.4c-2-.2-4.1-.3-6.2-.3c-35.3 0-64 28.7-64 64c0 2.1 .1 4.1 .3 6.2c-12.9 7.5-26.8 13.6-41.4 17.9c-16.9 5-26.6 22.9-21.5 39.8s22.9 26.6 39.8 21.5c21.5-6.4 41.8-15.5 60.6-27c8 3.6 16.9 5.6 26.3 5.6c35.3 0 64-28.7 64-64c0-9.4-2-18.3-5.6-26.3c11.5-18.7 20.6-39 27-60.6zM192.8 256.8c0-15.6 5.6-29.9 14.9-41.1L223 231c6.6 6.6 17.8 1.9 17.8-7.4V163.2c0-5.7-4.7-10.4-10.4-10.4H169.9c-9.3 0-13.9 11.2-7.4 17.8l11.2 11.2c-17.9 19.8-28.9 46.2-28.9 75.1c0 43.6 24.9 81.3 61.1 99.8c11.8 6 26.3 1.4 32.3-10.4s1.4-26.3-10.4-32.3c-20.8-10.6-34.9-32.2-34.9-57zm93.1-58.6c20.8 10.6 34.9 32.2 34.9 57c0 15.6-5.6 29.9-14.9 41.1L290.6 281c-6.6-6.6-17.8-1.9-17.8 7.4v60.5c0 5.7 4.7 10.4 10.4 10.4h60.5c9.3 0 13.9-11.2 7.4-17.8l-11.2-11.2c17.9-19.8 28.9-46.2 28.9-75.1c0-43.6-24.9-81.3-61.1-99.8c-11.8-6-26.3-1.4-32.3 10.4s-1.4 26.3 10.4 32.3z"],"bowl-food":[512,512,[],"e4c6","M0 192c0-35.3 28.7-64 64-64c.5 0 1.1 0 1.6 0C73 91.5 105.3 64 144 64c15 0 29 4.1 40.9 11.2C198.2 49.6 225.1 32 256 32s57.8 17.6 71.1 43.2C339 68.1 353 64 368 64c38.7 0 71 27.5 78.4 64c.5 0 1.1 0 1.6 0c35.3 0 64 28.7 64 64c0 11.7-3.1 22.6-8.6 32H8.6C3.1 214.6 0 203.7 0 192zm0 91.4C0 268.3 12.3 256 27.4 256H484.6c15.1 0 27.4 12.3 27.4 27.4c0 70.5-44.4 130.7-106.7 154.1L403.5 452c-2 16-15.6 28-31.8 28H140.2c-16.1 0-29.8-12-31.8-28l-1.8-14.4C44.4 414.1 0 353.9 0 283.4z"],"candy-cane":[512,512,[],"f786","M348.8 131.5c3.7-2.3 7.9-3.5 12.2-3.5c12.7 0 23 10.3 23 23v5.6c0 9.9-5.1 19.1-13.5 24.3L30.1 393.7C.1 412.5-9 451.9 9.7 481.9s58.2 39.1 88.2 20.4L438.4 289.5c45.8-28.6 73.6-78.8 73.6-132.8V151C512 67.6 444.4 0 361 0c-28.3 0-56 8-80.1 23L254.1 39.7c-30 18.7-39.1 58.2-20.4 88.2s58.2 39.1 88.2 20.4l26.8-16.8zM298.4 49.8c9.2-5.7 19.1-10.1 29.4-13.1L348 97.5c-5.7 1.4-11.2 3.7-16.3 6.8l-12.6 7.9L298.4 49.8zm88.5 52.7l46.2-46.2c8.5 6.5 16.1 14.1 22.6 22.6l-46.2 46.2c-5.1-9.6-13-17.5-22.6-22.6zm28.9 59.3l61.6 20.5c-2.2 10.5-5.8 20.7-10.5 30.2l-62-20.7c6.2-8.8 10.1-19.1 11-30.1zm-86.1 82.5l60.4 37.7-30.2 18.9-60.4-37.7 30.2-18.9zm-107.2 67l60.4 37.7-30.2 18.9-60.4-37.7 30.2-18.9zM119.3 375.7l60.4 37.7-30.2 18.9L89.1 394.6l30.2-18.9z"],"arrow-down-wide-short":[576,512,["sort-amount-asc","sort-amount-down"],"f160","M151.6 469.6C145.5 476.2 137 480 128 480s-17.5-3.8-23.6-10.4l-88-96c-11.9-13-11.1-33.3 2-45.2s33.3-11.1 45.2 2L96 365.7V64c0-17.7 14.3-32 32-32s32 14.3 32 32V365.7l32.4-35.4c11.9-13 32.2-13.9 45.2-2s13.9 32.2 2 45.2l-88 96zM320 480c-17.7 0-32-14.3-32-32s14.3-32 32-32h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H320zm0-128c-17.7 0-32-14.3-32-32s14.3-32 32-32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H320zm0-128c-17.7 0-32-14.3-32-32s14.3-32 32-32H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H320zm0-128c-17.7 0-32-14.3-32-32s14.3-32 32-32H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H320z"],"cloud-bolt":[512,512,[127785,"thunderstorm"],"f76c","M0 224c0 53 43 96 96 96h47.2L290 202.5c17.6-14.1 42.6-14 60.2 .2s22.8 38.6 12.8 58.8L333.7 320H352h64c53 0 96-43 96-96s-43-96-96-96c-.5 0-1.1 0-1.6 0c1.1-5.2 1.6-10.5 1.6-16c0-44.2-35.8-80-80-80c-24.3 0-46.1 10.9-60.8 28C256.5 24.3 219.1 0 176 0C114.1 0 64 50.1 64 112c0 7.1 .7 14.1 1.9 20.8C27.6 145.4 0 181.5 0 224zm330.1 3.6c-5.8-4.7-14.2-4.7-20.1-.1l-160 128c-5.3 4.2-7.4 11.4-5.1 17.8s8.3 10.7 15.1 10.7h70.1L177.7 488.8c-3.4 6.7-1.6 14.9 4.3 19.6s14.2 4.7 20.1 .1l160-128c5.3-4.2 7.4-11.4 5.1-17.8s-8.3-10.7-15.1-10.7H281.9l52.4-104.8c3.4-6.7 1.6-14.9-4.2-19.6z"],"text-slash":[640,512,["remove-format"],"f87d","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L355.7 253.5 400.2 96H503L497 120.2c-4.3 17.1 6.1 34.5 23.3 38.8s34.5-6.1 38.8-23.3l11-44.1C577.6 61.3 554.7 32 523.5 32H376.1h-.3H204.5c-22 0-41.2 15-46.6 36.4l-6.3 25.2L38.8 5.1zm168 131.7c.1-.3 .2-.7 .3-1L217 96H333.7L301.3 210.8l-94.5-74.1zM327.3 353.9L272.9 311 243.3 416H192c-17.7 0-32 14.3-32 32s14.3 32 32 32H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H309.8l17.6-62.1z"],"face-smile-wink":[512,512,[128521,"smile-wink"],"f4da","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM164.1 325.5C182 346.2 212.6 368 256 368s74-21.8 91.9-42.5c5.8-6.7 15.9-7.4 22.6-1.6s7.4 15.9 1.6 22.6C349.8 372.1 311.1 400 256 400s-93.8-27.9-116.1-53.5c-5.8-6.7-5.1-16.8 1.6-22.6s16.8-5.1 22.6 1.6zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm156.4 25.6c-5.3 7.1-15.3 8.5-22.4 3.2s-8.5-15.3-3.2-22.4c30.4-40.5 91.2-40.5 121.6 0c5.3 7.1 3.9 17.1-3.2 22.4s-17.1 3.9-22.4-3.2c-17.6-23.5-52.8-23.5-70.4 0z"],"file-word":[384,512,[],"f1c2","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM111 257.1l26.8 89.2 31.6-90.3c3.4-9.6 12.5-16.1 22.7-16.1s19.3 6.4 22.7 16.1l31.6 90.3L273 257.1c3.8-12.7 17.2-19.9 29.9-16.1s19.9 17.2 16.1 29.9l-48 160c-3 10-12 16.9-22.4 17.1s-19.8-6.2-23.2-16.1L192 336.6l-33.3 95.3c-3.4 9.8-12.8 16.3-23.2 16.1s-19.5-7.1-22.4-17.1l-48-160c-3.8-12.7 3.4-26.1 16.1-29.9s26.1 3.4 29.9 16.1z"],"file-powerpoint":[384,512,[],"f1c4","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM136 240h68c42 0 76 34 76 76s-34 76-76 76H160v32c0 13.3-10.7 24-24 24s-24-10.7-24-24V368 264c0-13.3 10.7-24 24-24zm68 104c15.5 0 28-12.5 28-28s-12.5-28-28-28H160v56h44z"],"arrows-left-right":[512,512,["arrows-h"],"f07e","M406.6 374.6l96-96c12.5-12.5 12.5-32.8 0-45.3l-96-96c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L402.7 224l-293.5 0 41.4-41.4c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-96 96c-12.5 12.5-12.5 32.8 0 45.3l96 96c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L109.3 288l293.5 0-41.4 41.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0z"],"house-lock":[640,512,[],"e510","M384 480c0 11.7 3.1 22.6 8.6 32H392c-22.1 0-40-17.9-40-40V448 384c0-17.7-14.3-32-32-32H256c-17.7 0-32 14.3-32 32v64 24c0 22.1-17.9 40-40 40H160 128.1c-1.5 0-3-.1-4.5-.2c-1.2 .1-2.4 .2-3.6 .2H104c-22.1 0-40-17.9-40-40V360c0-.9 0-1.9 .1-2.8V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L490.7 166.3C447.2 181.7 416 223.2 416 272v24.6c-19.1 11.1-32 31.7-32 55.4V480zM528 240c-17.7 0-32 14.3-32 32v48h64V272c0-17.7-14.3-32-32-32zm-80 32c0-44.2 35.8-80 80-80s80 35.8 80 80v48c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H448c-17.7 0-32-14.3-32-32V352c0-17.7 14.3-32 32-32V272z"],"cloud-arrow-down":[640,512,[62337,"cloud-download","cloud-download-alt"],"f0ed","M144 480C64.5 480 0 415.5 0 336c0-62.8 40.2-116.2 96.2-135.9c-.1-2.7-.2-5.4-.2-8.1c0-88.4 71.6-160 160-160c59.3 0 111 32.2 138.7 80.2C409.9 102 428.3 96 448 96c53 0 96 43 96 96c0 12.2-2.3 23.8-6.4 34.6C596 238.4 640 290.1 640 352c0 70.7-57.3 128-128 128H144zm79-167l80 80c9.4 9.4 24.6 9.4 33.9 0l80-80c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-39 39V184c0-13.3-10.7-24-24-24s-24 10.7-24 24V318.1l-39-39c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9z"],children:[640,512,[],"e4e1","M160 0a64 64 0 1 1 0 128A64 64 0 1 1 160 0zM88 480V400H70.2c-10.9 0-18.6-10.7-15.2-21.1l31.1-93.4L57.5 323.3c-10.7 14.1-30.8 16.8-44.8 6.2s-16.8-30.7-6.2-44.8L65.4 207c22.4-29.6 57.5-47 94.6-47s72.2 17.4 94.6 47l58.9 77.7c10.7 14.1 7.9 34.2-6.2 44.8s-34.2 7.9-44.8-6.2l-28.6-37.8L265 378.9c3.5 10.4-4.3 21.1-15.2 21.1H232v80c0 17.7-14.3 32-32 32s-32-14.3-32-32V400H152v80c0 17.7-14.3 32-32 32s-32-14.3-32-32zM480 0a64 64 0 1 1 0 128A64 64 0 1 1 480 0zm-8 384v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V300.5L395.1 321c-9.4 15-29.2 19.4-44.1 10s-19.4-29.2-10-44.1l51.7-82.1c17.6-27.9 48.3-44.9 81.2-44.9h12.3c33 0 63.7 16.9 81.2 44.9L619.1 287c9.4 15 4.9 34.7-10 44.1s-34.7 4.9-44.1-10L552 300.5V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V384H472z"],chalkboard:[576,512,["blackboard"],"f51b","M96 32C60.7 32 32 60.7 32 96V384H96V96l384 0V384h64V96c0-35.3-28.7-64-64-64H96zM224 384v32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H416V384c0-17.7-14.3-32-32-32H256c-17.7 0-32 14.3-32 32z"],"user-large-slash":[640,512,["user-alt-slash"],"f4fa","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L381.9 274c48.5-23.2 82.1-72.7 82.1-130C464 64.5 399.5 0 320 0C250.4 0 192.4 49.3 178.9 114.9L38.8 5.1zM545.5 512H528L284.3 320h-59C136.2 320 64 392.2 64 481.3c0 17 13.8 30.7 30.7 30.7H545.3l.3 0z"],"envelope-open":[512,512,[62135],"f2b6","M64 208.1L256 65.9 448 208.1v47.4L289.5 373c-9.7 7.2-21.4 11-33.5 11s-23.8-3.9-33.5-11L64 255.5V208.1zM256 0c-12.1 0-23.8 3.9-33.5 11L25.9 156.7C9.6 168.8 0 187.8 0 208.1V448c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V208.1c0-20.3-9.6-39.4-25.9-51.4L289.5 11C279.8 3.9 268.1 0 256 0z"],"handshake-simple-slash":[640,512,["handshake-alt-slash"],"e05f","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-135-105.8c-1.1-11.3-6.3-22.3-15.3-30.7l-134.2-123-23.4 18.2-26-20.3 77.2-60.1c7-5.4 17-4.2 22.5 2.8s4.2 17-2.8 22.5l-20.9 16.2L550.2 352H592c26.5 0 48-21.5 48-48V176c0-26.5-21.5-48-48-48H516h-4-.7l-3.9-2.5L434.8 79c-15.3-9.8-33.2-15-51.4-15c-21.8 0-43 7.5-60 21.2l-89.7 72.6-25.8-20.3 81.8-66.2c-11.6-4.9-24.1-7.4-36.8-7.4C234 64 215.7 69.6 200 80l-35.5 23.7L38.8 5.1zM413.6 421.9L41.2 128.5C17.9 131.8 0 151.8 0 176V304c0 26.5 21.5 48 48 48H156.2l91.4 83.4c19.6 17.9 49.9 16.5 67.8-3.1c5.5-6.1 9.2-13.2 11.1-20.6l17 15.6c19.5 17.9 49.9 16.6 67.8-2.9c.8-.8 1.5-1.7 2.2-2.6z"],"mattress-pillow":[640,512,[],"e525","M256 64H64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H256V64zm32 384H576c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H288V448zM64 160c0-17.7 14.3-32 32-32h64c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V160z"],"guarani-sign":[384,512,[],"e19a","M192 0c-17.7 0-32 14.3-32 32V66.7C69.2 81.9 0 160.9 0 256s69.2 174.1 160 189.3V480c0 17.7 14.3 32 32 32s32-14.3 32-32V445.3c90.8-15.2 160-94.2 160-189.3c0-17.7-14.3-32-32-32H224V132c22.1 5.7 41.8 17.1 57.6 32.6c12.6 12.4 32.9 12.2 45.3-.4s12.2-32.9-.5-45.3C299 92 263.5 73.3 224 66.7V32c0-17.7-14.3-32-32-32zM160 132V380c-55.2-14.2-96-64.3-96-124s40.8-109.8 96-124zM224 380V288h92c-11.6 45-47 80.4-92 92z"],"arrows-rotate":[512,512,[128472,"refresh","sync"],"f021","M105.1 202.6c7.7-21.8 20.2-42.3 37.8-59.8c62.5-62.5 163.8-62.5 226.3 0L386.3 160H352c-17.7 0-32 14.3-32 32s14.3 32 32 32H463.5c0 0 0 0 0 0h.4c17.7 0 32-14.3 32-32V80c0-17.7-14.3-32-32-32s-32 14.3-32 32v35.2L414.4 97.6c-87.5-87.5-229.3-87.5-316.8 0C73.2 122 55.6 150.7 44.8 181.4c-5.9 16.7 2.9 34.9 19.5 40.8s34.9-2.9 40.8-19.5zM39 289.3c-5 1.5-9.8 4.2-13.7 8.2c-4 4-6.7 8.8-8.1 14c-.3 1.2-.6 2.5-.8 3.8c-.3 1.7-.4 3.4-.4 5.1V432c0 17.7 14.3 32 32 32s32-14.3 32-32V396.9l17.6 17.5 0 0c87.5 87.4 229.3 87.4 316.7 0c24.4-24.4 42.1-53.1 52.9-83.7c5.9-16.7-2.9-34.9-19.5-40.8s-34.9 2.9-40.8 19.5c-7.7 21.8-20.2 42.3-37.8 59.8c-62.5 62.5-163.8 62.5-226.3 0l-.1-.1L125.6 352H160c17.7 0 32-14.3 32-32s-14.3-32-32-32H48.4c-1.6 0-3.2 .1-4.8 .3s-3.1 .5-4.6 1z"],"fire-extinguisher":[512,512,[129519],"f134","M500.3 7.3C507.7 13.3 512 22.4 512 32v96c0 9.6-4.3 18.7-11.7 24.7s-17.2 8.5-26.6 6.6l-160-32C301.5 124.9 292 115.7 289 104H224v34.8c37.8 18 64 56.5 64 101.2V384H64V240c0-44.7 26.2-83.2 64-101.2V110c-36.2 11.1-66 36.9-82.3 70.5c-5.8 11.9-20.2 16.9-32.1 11.1S-3.3 171.4 2.5 159.5C26.7 109.8 72.7 72.6 128 60.4V32c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V56h65c3-11.7 12.5-20.9 24.7-23.4l160-32c9.4-1.9 19.1 .6 26.6 6.6zM288 416v32c0 35.3-28.7 64-64 64H128c-35.3 0-64-28.7-64-64V416H288zM176 96a16 16 0 1 0 0-32 16 16 0 1 0 0 32z"],"cruzeiro-sign":[448,512,[],"e152","M96 256c0-88.4 71.6-160 160-160c41 0 78.3 15.4 106.7 40.7c13.2 11.8 33.4 10.7 45.2-2.5s10.7-33.4-2.5-45.2c-39.6-35.5-92-57-149.3-57C132.3 32 32 132.3 32 256s100.3 224 224 224c57.4 0 109.7-21.6 149.3-57c13.2-11.8 14.3-32 2.5-45.2s-32-14.3-45.2-2.5C334.3 400.6 297 416 256 416V320v-8.7c0-12.8 10.4-23.3 23.3-23.3c4.6 0 9.1 1.4 12.9 3.9l10.1 6.7c14.7 9.8 34.6 5.8 44.4-8.9s5.8-34.6-8.9-44.4l-10.1-6.7c-14.3-9.6-31.2-14.7-48.4-14.7c-12.4 0-24.2 2.6-34.9 7.3c-5.5-4.5-12.6-7.3-20.3-7.3c-17.7 0-32 14.3-32 32v55.3V320v82.7C135.5 378 96 321.6 96 256z"],"greater-than-equal":[448,512,[],"f532","M52.1 93.7C35.7 87.1 27.7 68.5 34.3 52.1s25.2-24.4 41.6-17.8l320 128C408 167.1 416 178.9 416 192s-8 24.9-20.1 29.7l-320 128c-16.4 6.6-35-1.4-41.6-17.8s1.4-35 17.8-41.6L297.8 192 52.1 93.7zM416 416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H416z"],"shield-halved":[512,512,["shield-alt"],"f3ed","M256 0c4.6 0 9.2 1 13.4 2.9L457.7 82.8c22 9.3 38.4 31 38.3 57.2c-.5 99.2-41.3 280.7-213.6 363.2c-16.7 8-36.1 8-52.8 0C57.3 420.7 16.5 239.2 16 140c-.1-26.2 16.3-47.9 38.3-57.2L242.7 2.9C246.8 1 251.4 0 256 0zm0 66.8V444.8C394 378 431.1 230.1 432 141.4L256 66.8l0 0z"],"book-atlas":[448,512,["atlas"],"f558","M0 96C0 43 43 0 96 0H384h32c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384 96c-53 0-96-43-96-96V96zM64 416c0 17.7 14.3 32 32 32H352V384H96c-17.7 0-32 14.3-32 32zM247.4 283.8c-3.7 3.7-6.2 4.2-7.4 4.2s-3.7-.5-7.4-4.2c-3.8-3.7-8-10-11.8-18.9c-6.2-14.5-10.8-34.3-12.2-56.9h63c-1.5 22.6-6 42.4-12.2 56.9c-3.8 8.9-8 15.2-11.8 18.9zm42.7-9.9c7.3-18.3 12-41.1 13.4-65.9h31.1c-4.7 27.9-21.4 51.7-44.5 65.9zm0-163.8c23.2 14.2 39.9 38 44.5 65.9H303.5c-1.4-24.7-6.1-47.5-13.4-65.9zM368 192a128 128 0 1 0 -256 0 128 128 0 1 0 256 0zM145.3 208h31.1c1.4 24.7 6.1 47.5 13.4 65.9c-23.2-14.2-39.9-38-44.5-65.9zm31.1-32H145.3c4.7-27.9 21.4-51.7 44.5-65.9c-7.3 18.3-12 41.1-13.4 65.9zm56.1-75.8c3.7-3.7 6.2-4.2 7.4-4.2s3.7 .5 7.4 4.2c3.8 3.7 8 10 11.8 18.9c6.2 14.5 10.8 34.3 12.2 56.9h-63c1.5-22.6 6-42.4 12.2-56.9c3.8-8.9 8-15.2 11.8-18.9z"],virus:[512,512,[],"e074","M288 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V43.5c0 49.9-60.3 74.9-95.6 39.6L120.2 75C107.7 62.5 87.5 62.5 75 75s-12.5 32.8 0 45.3l8.2 8.2C118.4 163.7 93.4 224 43.5 224H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H43.5c49.9 0 74.9 60.3 39.6 95.6L75 391.8c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l8.2-8.2c35.3-35.3 95.6-10.3 95.6 39.6V480c0 17.7 14.3 32 32 32s32-14.3 32-32V468.5c0-49.9 60.3-74.9 95.6-39.6l8.2 8.2c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-8.2-8.2c-35.3-35.3-10.3-95.6 39.6-95.6H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H468.5c-49.9 0-74.9-60.3-39.6-95.6l8.2-8.2c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-8.2 8.2C348.3 118.4 288 93.4 288 43.5V32zM176 224a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm128 56a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],"envelope-circle-check":[640,512,[],"e4e8","M48 64C21.5 64 0 85.5 0 112c0 15.1 7.1 29.3 19.2 38.4L236.8 313.6c11.4 8.5 27 8.5 38.4 0l57.4-43c23.9-59.8 79.7-103.3 146.3-109.8l13.9-10.4c12.1-9.1 19.2-23.3 19.2-38.4c0-26.5-21.5-48-48-48H48zM294.4 339.2c-22.8 17.1-54 17.1-76.8 0L0 176V384c0 35.3 28.7 64 64 64H360.2C335.1 417.6 320 378.5 320 336c0-5.6 .3-11.1 .8-16.6l-26.4 19.8zM640 336a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 353.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z"],"layer-group":[576,512,[],"f5fd","M264.5 5.2c14.9-6.9 32.1-6.9 47 0l218.6 101c8.5 3.9 13.9 12.4 13.9 21.8s-5.4 17.9-13.9 21.8l-218.6 101c-14.9 6.9-32.1 6.9-47 0L45.9 149.8C37.4 145.8 32 137.3 32 128s5.4-17.9 13.9-21.8L264.5 5.2zM476.9 209.6l53.2 24.6c8.5 3.9 13.9 12.4 13.9 21.8s-5.4 17.9-13.9 21.8l-218.6 101c-14.9 6.9-32.1 6.9-47 0L45.9 277.8C37.4 273.8 32 265.3 32 256s5.4-17.9 13.9-21.8l53.2-24.6 152 70.2c23.4 10.8 50.4 10.8 73.8 0l152-70.2zm-152 198.2l152-70.2 53.2 24.6c8.5 3.9 13.9 12.4 13.9 21.8s-5.4 17.9-13.9 21.8l-218.6 101c-14.9 6.9-32.1 6.9-47 0L45.9 405.8C37.4 401.8 32 393.3 32 384s5.4-17.9 13.9-21.8l53.2-24.6 152 70.2c23.4 10.8 50.4 10.8 73.8 0z"],"arrows-to-dot":[512,512,[],"e4be","M256 0c17.7 0 32 14.3 32 32V64h32c12.9 0 24.6 7.8 29.6 19.8s2.2 25.7-6.9 34.9l-64 64c-12.5 12.5-32.8 12.5-45.3 0l-64-64c-9.2-9.2-11.9-22.9-6.9-34.9s16.6-19.8 29.6-19.8h32V32c0-17.7 14.3-32 32-32zM169.4 393.4l64-64c12.5-12.5 32.8-12.5 45.3 0l64 64c9.2 9.2 11.9 22.9 6.9 34.9s-16.6 19.8-29.6 19.8H288v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H192c-12.9 0-24.6-7.8-29.6-19.8s-2.2-25.7 6.9-34.9zM32 224H64V192c0-12.9 7.8-24.6 19.8-29.6s25.7-2.2 34.9 6.9l64 64c12.5 12.5 12.5 32.8 0 45.3l-64 64c-9.2 9.2-22.9 11.9-34.9 6.9s-19.8-16.6-19.8-29.6V288H32c-17.7 0-32-14.3-32-32s14.3-32 32-32zm297.4 54.6c-12.5-12.5-12.5-32.8 0-45.3l64-64c9.2-9.2 22.9-11.9 34.9-6.9s19.8 16.6 19.8 29.6v32h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H448v32c0 12.9-7.8 24.6-19.8 29.6s-25.7 2.2-34.9-6.9l-64-64zM256 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],archway:[512,512,[],"f557","M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H32zm0 384c-17.7 0-32 14.3-32 32s14.3 32 32 32H96h64V352c0-53 43-96 96-96s96 43 96 96V480h64 64c17.7 0 32-14.3 32-32s-14.3-32-32-32V128H32V416z"],"heart-circle-check":[576,512,[],"e4fd","M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9l2.6-2.4C267.2 438.6 256 404.6 256 368c0-97.2 78.8-176 176-176c28.3 0 55 6.7 78.7 18.5c.9-6.5 1.3-13 1.3-19.6v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5zM576 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L416 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z"],"house-chimney-crack":[576,512,["house-damage"],"f6f1","M575.8 255.5c0 18-15 32.1-32 32.1h-32l.7 160.2c.2 35.5-28.5 64.3-64 64.3H326.4L288 448l80.8-67.3c7.8-6.5 7.6-18.6-.4-24.9L250.6 263.2c-14.6-11.5-33.8 7-22.8 22L288 368l-85.5 71.2c-6.1 5-7.5 13.8-3.5 20.5L230.4 512H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L416 100.7V64c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V185l52.8 46.4c8 7 12 15 11 24z"],"file-zipper":[384,512,["file-archive"],"f1c6","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM96 48c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16zm0 64c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16zm0 64c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16zm-6.3 71.8c3.7-14 16.4-23.8 30.9-23.8h14.8c14.5 0 27.2 9.7 30.9 23.8l23.5 88.2c1.4 5.4 2.1 10.9 2.1 16.4c0 35.2-28.8 63.7-64 63.7s-64-28.5-64-63.7c0-5.5 .7-11.1 2.1-16.4l23.5-88.2zM112 336c-8.8 0-16 7.2-16 16s7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H112z"],square:[448,512,[9632,9723,9724,61590],"f0c8","M0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96z"],"martini-glass-empty":[512,512,["glass-martini"],"f000","M32 0C19.1 0 7.4 7.8 2.4 19.8s-2.2 25.7 6.9 34.9L224 269.3V448H160c-17.7 0-32 14.3-32 32s14.3 32 32 32h96 96c17.7 0 32-14.3 32-32s-14.3-32-32-32H288V269.3L502.6 54.6c9.2-9.2 11.9-22.9 6.9-34.9S492.9 0 480 0H32zM256 210.7L109.3 64H402.7L256 210.7z"],couch:[640,512,[],"f4b8","M64 160C64 89.3 121.3 32 192 32H448c70.7 0 128 57.3 128 128v33.6c-36.5 7.4-64 39.7-64 78.4v48H128V272c0-38.7-27.5-71-64-78.4V160zM544 272c0-20.9 13.4-38.7 32-45.3c5-1.8 10.4-2.7 16-2.7c26.5 0 48 21.5 48 48V448c0 17.7-14.3 32-32 32H576c-17.7 0-32-14.3-32-32H96c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32V272c0-26.5 21.5-48 48-48c5.6 0 11 1 16 2.7c18.6 6.6 32 24.4 32 45.3v48 32h32H512h32V320 272z"],"cedi-sign":[384,512,[],"e0df","M256 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V66.7C101.2 81.9 32 160.9 32 256s69.2 174.1 160 189.3V480c0 17.7 14.3 32 32 32s32-14.3 32-32V445.3c30.9-5.2 59.2-17.7 83.2-35.8c14.1-10.6 17-30.7 6.4-44.8s-30.7-17-44.8-6.4c-13.2 9.9-28.3 17.3-44.8 21.6V132c16.4 4.2 31.6 11.6 44.8 21.6c14.1 10.6 34.2 7.8 44.8-6.4s7.8-34.2-6.4-44.8c-24-18-52.4-30.6-83.2-35.8V32zM192 132V380c-55.2-14.2-96-64.3-96-124s40.8-109.8 96-124z"],italic:[384,512,[],"f033","M128 64c0-17.7 14.3-32 32-32H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H293.3L160 416h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H90.7L224 96H160c-17.7 0-32-14.3-32-32z"],"table-cells-column-lock":[640,512,[],"e678","M0 96C0 60.7 28.7 32 64 32l384 0c35.3 0 64 28.7 64 64l0 65.1c-37.8 5.4-69.4 29.6-85.2 62.9L360 224l0 64 56 0 0 8.6c-19.1 11.1-32 31.7-32 55.4l-24 0 0 64 24 0 0 64L64 480c-35.3 0-64-28.7-64-64L0 96zm208 0l0 64 88 0 0-64-88 0zm240 0l-88 0 0 64 88 0 0-64zM208 224l0 64 88 0 0-64-88 0zm0 128l0 64 88 0 0-64-88 0zM528 240c-17.7 0-32 14.3-32 32l0 48 64 0 0-48c0-17.7-14.3-32-32-32zm-80 32c0-44.2 35.8-80 80-80s80 35.8 80 80l0 48c17.7 0 32 14.3 32 32l0 128c0 17.7-14.3 32-32 32l-160 0c-17.7 0-32-14.3-32-32l0-128c0-17.7 14.3-32 32-32l0-48z"],church:[640,512,[9962],"f51d","M344 24c0-13.3-10.7-24-24-24s-24 10.7-24 24V48H264c-13.3 0-24 10.7-24 24s10.7 24 24 24h32v46.4L183.3 210c-14.5 8.7-23.3 24.3-23.3 41.2V512h96V416c0-35.3 28.7-64 64-64s64 28.7 64 64v96h96V251.2c0-16.9-8.8-32.5-23.3-41.2L344 142.4V96h32c13.3 0 24-10.7 24-24s-10.7-24-24-24H344V24zM24.9 330.3C9.5 338.8 0 354.9 0 372.4V464c0 26.5 21.5 48 48 48h80V273.6L24.9 330.3zM592 512c26.5 0 48-21.5 48-48V372.4c0-17.5-9.5-33.6-24.9-42.1L512 273.6V512h80z"],"comments-dollar":[640,512,[],"f653","M416 176c0 97.2-93.1 176-208 176c-38.2 0-73.9-8.7-104.7-23.9c-7.5 4-16 7.9-25.2 11.4C59.8 346.4 37.8 352 16 352c-6.9 0-13.1-4.5-15.2-11.1s.2-13.8 5.8-17.9l0 0 0 0 .2-.2c.2-.2 .6-.4 1.1-.8c1-.8 2.5-2 4.3-3.7c3.6-3.3 8.5-8.1 13.3-14.3c5.5-7 10.7-15.4 14.2-24.7C14.7 250.3 0 214.6 0 176C0 78.8 93.1 0 208 0S416 78.8 416 176zM231.5 383C348.9 372.9 448 288.3 448 176c0-5.2-.2-10.4-.6-15.5C555.1 167.1 640 243.2 640 336c0 38.6-14.7 74.3-39.6 103.4c3.5 9.4 8.7 17.7 14.2 24.7c4.8 6.2 9.7 11 13.3 14.3c1.8 1.6 3.3 2.9 4.3 3.7c.5 .4 .9 .7 1.1 .8l.2 .2 0 0 0 0c5.6 4.1 7.9 11.3 5.8 17.9c-2.1 6.6-8.3 11.1-15.2 11.1c-21.8 0-43.8-5.6-62.1-12.5c-9.2-3.5-17.8-7.4-25.2-11.4C505.9 503.3 470.2 512 432 512c-95.6 0-176.2-54.6-200.5-129zM228 72c0-11-9-20-20-20s-20 9-20 20V86c-7.6 1.7-15.2 4.4-22.2 8.5c-13.9 8.3-25.9 22.8-25.8 43.9c.1 20.3 12 33.1 24.7 40.7c11 6.6 24.7 10.8 35.6 14l1.7 .5c12.6 3.8 21.8 6.8 28 10.7c5.1 3.2 5.8 5.4 5.9 8.2c.1 5-1.8 8-5.9 10.5c-5 3.1-12.9 5-21.4 4.7c-11.1-.4-21.5-3.9-35.1-8.5c-2.3-.8-4.7-1.6-7.2-2.4c-10.5-3.5-21.8 2.2-25.3 12.6s2.2 21.8 12.6 25.3c1.9 .6 4 1.3 6.1 2.1l0 0 0 0c8.3 2.9 17.9 6.2 28.2 8.4V280c0 11 9 20 20 20s20-9 20-20V266.2c8-1.7 16-4.5 23.2-9c14.3-8.9 25.1-24.1 24.8-45c-.3-20.3-11.7-33.4-24.6-41.6c-11.5-7.2-25.9-11.6-37.1-15l-.7-.2c-12.8-3.9-21.9-6.7-28.3-10.5c-5.2-3.1-5.3-4.9-5.3-6.7c0-3.7 1.4-6.5 6.2-9.3c5.4-3.2 13.6-5.1 21.5-5c9.6 .1 20.2 2.2 31.2 5.2c10.7 2.8 21.6-3.5 24.5-14.2s-3.5-21.6-14.2-24.5c-6.5-1.7-13.7-3.4-21.1-4.7V72z"],democrat:[640,512,[],"f747","M64 32c0-8.9 3.8-20.9 6.2-27.3C71.2 1.8 74 0 77 0c1.9 0 3.8 .7 5.2 2.1L128 45.7 173.8 2.1C175.2 .7 177.1 0 179 0c3 0 5.8 1.8 6.8 4.7c2.4 6.5 6.2 18.4 6.2 27.3c0 26.5-21.9 42-29.5 46.6l76.2 72.6c6 5.7 13.9 8.8 22.1 8.8H480l32 0c40.3 0 78.2 19 102.4 51.2l19.2 25.6c10.6 14.1 7.7 34.2-6.4 44.8s-34.2 7.7-44.8-6.4l-19.2-25.6c-5.3-7-11.8-12.8-19.2-17V320H192l-40.4-94.3c-3.9-9.2-15.3-12.6-23.6-7l-42.1 28c-9.1 6.1-19.7 9.3-30.7 9.3h-2C23.9 256 0 232.1 0 202.7c0-12.1 4.1-23.8 11.7-33.3L87.6 74.6C78.1 67.4 64 53.2 64 32zM448 352h96v64 64c0 17.7-14.3 32-32 32H480c-17.7 0-32-14.3-32-32V416H288v64c0 17.7-14.3 32-32 32H224c-17.7 0-32-14.3-32-32V416 352h96H448zM260.9 210.9c-.9-1.8-2.8-2.9-4.8-2.9s-3.9 1.1-4.8 2.9l-10.5 20.5-23.5 3.3c-2 .3-3.7 1.6-4.3 3.5s-.1 3.9 1.3 5.3l17 16-4 22.6c-.3 1.9 .5 3.9 2.1 5s3.8 1.3 5.6 .4l21-10.7 21 10.7c1.8 .9 4 .8 5.6-.4s2.5-3.1 2.1-5l-4-22.6 17-16c1.5-1.4 2-3.4 1.3-5.3s-2.3-3.2-4.3-3.5l-23.5-3.3-10.5-20.5zM368.1 208c-2 0-3.9 1.1-4.8 2.9l-10.5 20.5-23.5 3.3c-2 .3-3.7 1.6-4.3 3.5s-.1 3.9 1.3 5.3l17 16-4 22.6c-.3 1.9 .5 3.9 2.1 5s3.8 1.3 5.6 .4l21-10.7 21 10.7c1.8 .9 4 .8 5.6-.4s2.5-3.1 2.1-5l-4-22.6 17-16c1.5-1.4 2-3.4 1.4-5.3s-2.3-3.2-4.3-3.5l-23.5-3.3-10.5-20.5c-.9-1.8-2.8-2.9-4.8-2.9zm116.8 2.9c-.9-1.8-2.8-2.9-4.8-2.9s-3.9 1.1-4.8 2.9l-10.5 20.5-23.5 3.3c-2 .3-3.7 1.6-4.3 3.5s-.1 3.9 1.3 5.3l17 16-4 22.6c-.3 1.9 .5 3.9 2.1 5s3.8 1.3 5.6 .4l21-10.7 21 10.7c1.8 .9 4 .8 5.6-.4s2.5-3.1 2.1-5l-4-22.6 17-16c1.5-1.4 2-3.4 1.4-5.3s-2.3-3.2-4.3-3.5l-23.5-3.3-10.5-20.5z"],z:[384,512,[122],"5a","M0 64C0 46.3 14.3 32 32 32H352c12.4 0 23.7 7.2 29 18.4s3.6 24.5-4.4 34.1L100.3 416H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-12.4 0-23.7-7.2-29-18.4s-3.6-24.5 4.4-34.1L283.7 96H32C14.3 96 0 81.7 0 64z"],"person-skiing":[512,512,[9975,"skiing"],"f7c9","M380.7 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM2.7 268.9c6.1-11.8 20.6-16.3 32.4-10.2L232.7 361.3l46.2-69.2-75.1-75.1c-14.6-14.6-20.4-33.9-18.4-52.1l108.8 52 39.3 39.3c16.2 16.2 18.7 41.5 6 60.6L289.8 391l128.7 66.8c13.6 7.1 29.8 7.2 43.6 .3l15.2-7.6c11.9-5.9 26.3-1.1 32.2 10.7s1.1 26.3-10.7 32.2l-15.2 7.6c-27.5 13.7-59.9 13.5-87.2-.7L12.9 301.3C1.2 295.2-3.4 280.7 2.7 268.9zM118.9 65.6L137 74.2l8.7-17.4c4-7.9 13.6-11.1 21.5-7.2s11.1 13.6 7.2 21.5l-8.5 16.9 54.7 26.2c1.5-.7 3.1-1.4 4.7-2.1l83.4-33.4c34.2-13.7 72.8 4.2 84.5 39.2l17.1 51.2 52.1 26.1c15.8 7.9 22.2 27.1 14.3 42.9s-27.1 22.2-42.9 14.3l-58.1-29c-11.4-5.7-20-15.7-24.1-27.8l-5.8-17.3-27.3 12.1-6.8 3-6.7-3.2L151.5 116.7l-9.2 18.4c-4 7.9-13.6 11.1-21.5 7.2s-11.1-13.6-7.2-21.5l9-18-17.6-8.4c-8-3.8-11.3-13.4-7.5-21.3s13.4-11.3 21.3-7.5z"],"road-lock":[640,512,[],"e567","M288 32H213.2c-27.1 0-51.3 17.1-60.3 42.6L35.1 407.2c-2.1 5.9-3.1 12-3.1 18.2C32 455.5 56.5 480 86.6 480H288V416c0-17.7 14.3-32 32-32s32 14.3 32 32v64h32V352c0-23.7 12.9-44.4 32-55.4V272c0-58.3 44.6-106.2 101.5-111.5L487.1 74.6C478 49.1 453.9 32 426.8 32H352V96c0 17.7-14.3 32-32 32s-32-14.3-32-32V32zm64 192v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V224c0-17.7 14.3-32 32-32s32 14.3 32 32zm176 16c17.7 0 32 14.3 32 32v48H496V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z"],a:[384,512,[97],"41","M221.5 51.7C216.6 39.8 204.9 32 192 32s-24.6 7.8-29.5 19.7l-120 288-40 96c-6.8 16.3 .9 35 17.2 41.8s35-.9 41.8-17.2L93.3 384H290.7l31.8 76.3c6.8 16.3 25.5 24 41.8 17.2s24-25.5 17.2-41.8l-40-96-120-288zM264 320H120l72-172.8L264 320z"],"temperature-arrow-down":[576,512,["temperature-down"],"e03f","M128 112c0-26.5 21.5-48 48-48s48 21.5 48 48V276.5c0 17.3 7.1 31.9 15.3 42.5C249.8 332.6 256 349.5 256 368c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-18.5 6.2-35.4 16.7-48.9c8.2-10.6 15.3-25.2 15.3-42.5V112zM176 0C114.1 0 64 50.1 64 112V276.4c0 .1-.1 .3-.2 .6c-.2 .6-.8 1.6-1.7 2.8C43.2 304.2 32 334.8 32 368c0 79.5 64.5 144 144 144s144-64.5 144-144c0-33.2-11.2-63.8-30.1-88.1c-.9-1.2-1.5-2.2-1.7-2.8c-.1-.3-.2-.5-.2-.6V112C288 50.1 237.9 0 176 0zm0 416c26.5 0 48-21.5 48-48c0-20.9-13.4-38.7-32-45.3V272c0-8.8-7.2-16-16-16s-16 7.2-16 16v50.7c-18.6 6.6-32 24.4-32 45.3c0 26.5 21.5 48 48 48zm336-64H480V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V352H384c-12.9 0-24.6 7.8-29.6 19.8s-2.2 25.7 6.9 34.9l64 64c6 6 14.1 9.4 22.6 9.4s16.6-3.4 22.6-9.4l64-64c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8z"],"feather-pointed":[512,512,["feather-alt"],"f56b","M278.5 215.6L23 471c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l74.8-74.8c7.4 4.6 15.3 8.2 23.8 10.5C200.3 452.8 270 454.5 338 409.4c12.2-8.1 5.8-25.4-8.8-25.4l-16.1 0c-5.1 0-9.2-4.1-9.2-9.2c0-4.1 2.7-7.6 6.5-8.8l97.7-29.3c3.4-1 6.4-3.1 8.4-6.1c4.4-6.4 8.6-12.9 12.6-19.6c6.2-10.3-1.5-23-13.5-23l-38.6 0c-5.1 0-9.2-4.1-9.2-9.2c0-4.1 2.7-7.6 6.5-8.8l80.9-24.3c4.6-1.4 8.4-4.8 10.2-9.3C494.5 163 507.8 86.1 511.9 36.8c.8-9.9-3-19.6-10-26.6s-16.7-10.8-26.6-10C391.5 7 228.5 40.5 137.4 131.6C57.3 211.7 56.7 302.3 71.3 356.4c2.1 7.9 12 9.6 17.8 3.8L253.6 195.8c6.2-6.2 16.4-6.2 22.6 0c5.4 5.4 6.1 13.6 2.2 19.8z"],p:[320,512,[112],"50","M0 96C0 60.7 28.7 32 64 32h96c88.4 0 160 71.6 160 160s-71.6 160-160 160H64v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V320 96zM64 288h96c53 0 96-43 96-96s-43-96-96-96H64V288z"],snowflake:[448,512,[10052,10054],"f2dc","M224 0c17.7 0 32 14.3 32 32V62.1l15-15c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-49 49v70.3l61.4-35.8 17.7-66.1c3.4-12.8 16.6-20.4 29.4-17s20.4 16.6 17 29.4l-5.2 19.3 23.6-13.8c15.3-8.9 34.9-3.7 43.8 11.5s3.8 34.9-11.5 43.8l-25.3 14.8 21.7 5.8c12.8 3.4 20.4 16.6 17 29.4s-16.6 20.4-29.4 17l-67.7-18.1L287.5 256l60.9 35.5 67.7-18.1c12.8-3.4 26 4.2 29.4 17s-4.2 26-17 29.4l-21.7 5.8 25.3 14.8c15.3 8.9 20.4 28.5 11.5 43.8s-28.5 20.4-43.8 11.5l-23.6-13.8 5.2 19.3c3.4 12.8-4.2 26-17 29.4s-26-4.2-29.4-17l-17.7-66.1L256 311.7v70.3l49 49c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-15-15V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V449.9l-15 15c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l49-49V311.7l-61.4 35.8-17.7 66.1c-3.4 12.8-16.6 20.4-29.4 17s-20.4-16.6-17-29.4l5.2-19.3L48.1 395.6c-15.3 8.9-34.9 3.7-43.8-11.5s-3.7-34.9 11.5-43.8l25.3-14.8-21.7-5.8c-12.8-3.4-20.4-16.6-17-29.4s16.6-20.4 29.4-17l67.7 18.1L160.5 256 99.6 220.5 31.9 238.6c-12.8 3.4-26-4.2-29.4-17s4.2-26 17-29.4l21.7-5.8L15.9 171.6C.6 162.7-4.5 143.1 4.4 127.9s28.5-20.4 43.8-11.5l23.6 13.8-5.2-19.3c-3.4-12.8 4.2-26 17-29.4s26 4.2 29.4 17l17.7 66.1L192 200.3V129.9L143 81c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l15 15V32c0-17.7 14.3-32 32-32z"],newspaper:[512,512,[128240],"f1ea","M96 96c0-35.3 28.7-64 64-64H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H80c-44.2 0-80-35.8-80-80V128c0-17.7 14.3-32 32-32s32 14.3 32 32V400c0 8.8 7.2 16 16 16s16-7.2 16-16V96zm64 24v80c0 13.3 10.7 24 24 24H296c13.3 0 24-10.7 24-24V120c0-13.3-10.7-24-24-24H184c-13.3 0-24 10.7-24 24zm208-8c0 8.8 7.2 16 16 16h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H384c-8.8 0-16 7.2-16 16zm0 96c0 8.8 7.2 16 16 16h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H384c-8.8 0-16 7.2-16 16zM160 304c0 8.8 7.2 16 16 16H432c8.8 0 16-7.2 16-16s-7.2-16-16-16H176c-8.8 0-16 7.2-16 16zm0 96c0 8.8 7.2 16 16 16H432c8.8 0 16-7.2 16-16s-7.2-16-16-16H176c-8.8 0-16 7.2-16 16z"],"rectangle-ad":[576,512,["ad"],"f641","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM229.5 173.3l72 144c5.9 11.9 1.1 26.3-10.7 32.2s-26.3 1.1-32.2-10.7L253.2 328H162.8l-5.4 10.7c-5.9 11.9-20.3 16.7-32.2 10.7s-16.7-20.3-10.7-32.2l72-144c4.1-8.1 12.4-13.3 21.5-13.3s17.4 5.1 21.5 13.3zM208 237.7L186.8 280h42.3L208 237.7zM392 256a24 24 0 1 0 0 48 24 24 0 1 0 0-48zm24-43.9V184c0-13.3 10.7-24 24-24s24 10.7 24 24v96 48c0 13.3-10.7 24-24 24c-6.6 0-12.6-2.7-17-7c-9.4 4.5-19.9 7-31 7c-39.8 0-72-32.2-72-72s32.2-72 72-72c8.4 0 16.5 1.4 24 4.1z"],"circle-arrow-right":[512,512,["arrow-circle-right"],"f0a9","M0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zM281 385c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l71-71L136 280c-13.3 0-24-10.7-24-24s10.7-24 24-24l182.1 0-71-71c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0L393 239c9.4 9.4 9.4 24.6 0 33.9L281 385z"],"filter-circle-xmark":[576,512,[],"e17b","M3.9 22.9C10.5 8.9 24.5 0 40 0H472c15.5 0 29.5 8.9 36.1 22.9s4.6 30.5-5.2 42.5L396.4 195.6C316.2 212.1 256 283 256 368c0 27.4 6.3 53.4 17.5 76.5c-1.6-.8-3.2-1.8-4.7-2.9l-64-48c-8.1-6-12.8-15.5-12.8-25.6V288.9L9 65.3C-.7 53.4-2.8 36.8 3.9 22.9zM432 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm59.3 107.3c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0L432 345.4l-36.7-36.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6L409.4 368l-36.7 36.7c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L432 390.6l36.7 36.7c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6L454.6 368l36.7-36.7z"],locust:[576,512,[],"e520","M312 32c-13.3 0-24 10.7-24 24s10.7 24 24 24h16c98.7 0 180.6 71.4 197 165.4c-9-3.5-18.8-5.4-29-5.4H431.8l-41.8-97.5c-3.4-7.9-10.8-13.4-19.3-14.4s-17 2.7-22.1 9.6l-40.9 55.5-21.7-50.7c-3.3-7.8-10.5-13.2-18.9-14.3s-16.7 2.3-22 8.9l-240 304c-8.2 10.4-6.4 25.5 4 33.7s25.5 6.4 33.7-4l79.4-100.5 43 16.4-40.5 55c-7.9 10.7-5.6 25.7 5.1 33.6s25.7 5.6 33.6-5.1L215.1 400h74.5l-29.3 42.3c-7.5 10.9-4.8 25.8 6.1 33.4s25.8 4.8 33.4-6.1L348 400h80.4l38.8 67.9c6.6 11.5 21.2 15.5 32.7 8.9s15.5-21.2 8.9-32.7L483.6 400H496c44.1 0 79.8-35.7 80-79.7c0-.1 0-.2 0-.3V280C576 143 465 32 328 32H312zm50.5 168l17.1 40H333l29.5-40zm-87.7 38.1l-1.4 1.9H225.1l32.7-41.5 16.9 39.5zM88.8 240C57.4 240 32 265.4 32 296.8c0 15.5 6.3 30 16.9 40.4L126.7 240H88.8zM496 288a16 16 0 1 1 0 32 16 16 0 1 1 0-32z"],sort:[320,512,["unsorted"],"f0dc","M137.4 41.4c12.5-12.5 32.8-12.5 45.3 0l128 128c9.2 9.2 11.9 22.9 6.9 34.9s-16.6 19.8-29.6 19.8H32c-12.9 0-24.6-7.8-29.6-19.8s-2.2-25.7 6.9-34.9l128-128zm0 429.3l-128-128c-9.2-9.2-11.9-22.9-6.9-34.9s16.6-19.8 29.6-19.8H288c12.9 0 24.6 7.8 29.6 19.8s2.2 25.7-6.9 34.9l-128 128c-12.5 12.5-32.8 12.5-45.3 0z"],"list-ol":[512,512,["list-1-2","list-numeric"],"f0cb","M24 56c0-13.3 10.7-24 24-24H80c13.3 0 24 10.7 24 24V176h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H40c-13.3 0-24-10.7-24-24s10.7-24 24-24H56V80H48C34.7 80 24 69.3 24 56zM86.7 341.2c-6.5-7.4-18.3-6.9-24 1.2L51.5 357.9c-7.7 10.8-22.7 13.3-33.5 5.6s-13.3-22.7-5.6-33.5l11.1-15.6c23.7-33.2 72.3-35.6 99.2-4.9c21.3 24.4 20.8 60.9-1.1 84.7L86.8 432H120c13.3 0 24 10.7 24 24s-10.7 24-24 24H32c-9.5 0-18.2-5.6-22-14.4s-2.1-18.9 4.3-25.9l72-78c5.3-5.8 5.4-14.6 .3-20.5zM224 64H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32s14.3-32 32-32zm0 160H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32s14.3-32 32-32zm0 160H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],"person-dress-burst":[640,512,[],"e544","M528 48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM390.2 384H408v96c0 17.7 14.3 32 32 32s32-14.3 32-32V384h16v96c0 17.7 14.3 32 32 32s32-14.3 32-32V384h17.8c10.9 0 18.6-10.7 15.2-21.1L546.7 248.1l33.9 56.3c9.1 15.1 28.8 20 43.9 10.9s20-28.8 10.9-43.9l-53.6-89.2c-20.2-33.7-56.7-54.3-96-54.3H474.2c-39.3 0-75.7 20.6-96 54.3l-53.6 89.2c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9l33.9-56.3L375 362.9c-3.5 10.4 4.3 21.1 15.2 21.1zM190.9 18.1C188.4 12 182.6 8 176 8s-12.4 4-14.9 10.1l-29.4 74L55.6 68.9c-6.3-1.9-13.1 .2-17.2 5.3s-4.6 12.2-1.4 17.9l39.5 69.1L10.9 206.4c-5.4 3.7-8 10.3-6.5 16.7s6.7 11.2 13.1 12.2l78.7 12.2L90.6 327c-.5 6.5 3.1 12.7 9 15.5s12.9 1.8 17.8-2.6L176 286.1l58.6 53.9c4.8 4.4 11.9 5.5 17.8 2.6s9.5-9 9-15.5l-5.6-79.4 50.5-7.8 24.4-40.5-55.2-38L315 92.2c3.3-5.7 2.7-12.8-1.4-17.9s-10.9-7.2-17.2-5.3L220.3 92.1l-29.4-74z"],"money-check-dollar":[576,512,["money-check-alt"],"f53d","M64 64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H64zM272 192H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H272c-8.8 0-16-7.2-16-16s7.2-16 16-16zM256 304c0-8.8 7.2-16 16-16H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H272c-8.8 0-16-7.2-16-16zM164 152v13.9c7.5 1.2 14.6 2.9 21.1 4.7c10.7 2.8 17 13.8 14.2 24.5s-13.8 17-24.5 14.2c-11-2.9-21.6-5-31.2-5.2c-7.9-.1-16 1.8-21.5 5c-4.8 2.8-6.2 5.6-6.2 9.3c0 1.8 .1 3.5 5.3 6.7c6.3 3.8 15.5 6.7 28.3 10.5l.7 .2c11.2 3.4 25.6 7.7 37.1 15c12.9 8.1 24.3 21.3 24.6 41.6c.3 20.9-10.5 36.1-24.8 45c-7.2 4.5-15.2 7.3-23.2 9V360c0 11-9 20-20 20s-20-9-20-20V345.4c-10.3-2.2-20-5.5-28.2-8.4l0 0 0 0c-2.1-.7-4.1-1.4-6.1-2.1c-10.5-3.5-16.1-14.8-12.6-25.3s14.8-16.1 25.3-12.6c2.5 .8 4.9 1.7 7.2 2.4c13.6 4.6 24 8.1 35.1 8.5c8.6 .3 16.5-1.6 21.4-4.7c4.1-2.5 6-5.5 5.9-10.5c0-2.9-.8-5-5.9-8.2c-6.3-4-15.4-6.9-28-10.7l-1.7-.5c-10.9-3.3-24.6-7.4-35.6-14c-12.7-7.7-24.6-20.5-24.7-40.7c-.1-21.1 11.8-35.7 25.8-43.9c6.9-4.1 14.5-6.8 22.2-8.5V152c0-11 9-20 20-20s20 9 20 20z"],"vector-square":[448,512,[],"f5cb","M368 80h32v32H368V80zM352 32c-17.7 0-32 14.3-32 32H128c0-17.7-14.3-32-32-32H32C14.3 32 0 46.3 0 64v64c0 17.7 14.3 32 32 32V352c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H96c17.7 0 32-14.3 32-32H320c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V384c0-17.7-14.3-32-32-32V160c17.7 0 32-14.3 32-32V64c0-17.7-14.3-32-32-32H352zM96 160c17.7 0 32-14.3 32-32H320c0 17.7 14.3 32 32 32V352c-17.7 0-32 14.3-32 32H128c0-17.7-14.3-32-32-32V160zM48 400H80v32H48V400zm320 32V400h32v32H368zM48 112V80H80v32H48z"],"bread-slice":[512,512,[],"f7ec","M256 32C192 32 0 64 0 192c0 35.3 28.7 64 64 64V432c0 26.5 21.5 48 48 48H400c26.5 0 48-21.5 48-48V256c35.3 0 64-28.7 64-64C512 64 320 32 256 32z"],language:[640,512,[],"f1ab","M0 128C0 92.7 28.7 64 64 64H256h48 16H576c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H320 304 256 64c-35.3 0-64-28.7-64-64V128zm320 0V384H576V128H320zM178.3 175.9c-3.2-7.2-10.4-11.9-18.3-11.9s-15.1 4.7-18.3 11.9l-64 144c-4.5 10.1 .1 21.9 10.2 26.4s21.9-.1 26.4-10.2l8.9-20.1h73.6l8.9 20.1c4.5 10.1 16.3 14.6 26.4 10.2s14.6-16.3 10.2-26.4l-64-144zM160 233.2L179 276H141l19-42.8zM448 164c11 0 20 9 20 20v4h44 16c11 0 20 9 20 20s-9 20-20 20h-2l-1.6 4.5c-8.9 24.4-22.4 46.6-39.6 65.4c.9 .6 1.8 1.1 2.7 1.6l18.9 11.3c9.5 5.7 12.5 18 6.9 27.4s-18 12.5-27.4 6.9l-18.9-11.3c-4.5-2.7-8.8-5.5-13.1-8.5c-10.6 7.5-21.9 14-34 19.4l-3.6 1.6c-10.1 4.5-21.9-.1-26.4-10.2s.1-21.9 10.2-26.4l3.6-1.6c6.4-2.9 12.6-6.1 18.5-9.8l-12.2-12.2c-7.8-7.8-7.8-20.5 0-28.3s20.5-7.8 28.3 0l14.6 14.6 .5 .5c12.4-13.1 22.5-28.3 29.8-45H448 376c-11 0-20-9-20-20s9-20 20-20h52v-4c0-11 9-20 20-20z"],"face-kiss-wink-heart":[512,512,[128536,"kiss-wink-heart"],"f598","M498 339.7c9.1-26.2 14-54.4 14-83.7C512 114.6 397.4 0 256 0S0 114.6 0 256S114.6 512 256 512c35.4 0 69.1-7.2 99.7-20.2c-4.8-5.5-8.5-12.2-10.4-19.7l-22.9-89.3c-10-39 11.8-80.9 51.8-92.1c37.2-10.4 73.8 10.1 87.5 44c12.7-1.6 25.1 .4 36.2 5zM296 332c0 6.9-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4c-2.7 1.5-5.7 3-8.7 4.3c3.1 1.3 6 2.7 8.7 4.3c6.6 3.7 12.5 8.2 16.7 13.4c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4C258.7 443.1 241.4 448 224 448c-3.6 0-6.8-2.5-7.7-6s.6-7.2 3.8-9l0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1l-.6-.4-.3-.2-.2-.1 0 0 0 0 0 0c-2.5-1.4-4.1-4.1-4.1-7s1.6-5.6 4.1-7l0 0 0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1c-.4-.2-.7-.4-.9-.5l-.2-.1 0 0 0 0 0 0c-3.2-1.8-4.7-5.5-3.8-9s4.1-6 7.7-6c17.4 0 34.7 4.9 47.9 12.3c6.6 3.7 12.5 8.2 16.7 13.4c4.3 5.1 7.3 11.4 7.3 18.3zM176.4 176a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm194.8 57.6c-17.6-23.5-52.8-23.5-70.4 0c-5.3 7.1-15.3 8.5-22.4 3.2s-8.5-15.3-3.2-22.4c30.4-40.5 91.2-40.5 121.6 0c5.3 7.1 3.9 17.1-3.2 22.4s-17.1 3.9-22.4-3.2zM434 352.3c-6-23.2-28.8-37-51.1-30.8s-35.4 30.1-29.5 53.4l22.9 89.3c2.2 8.7 11.2 13.9 19.8 11.4l84.9-23.8c22.2-6.2 35.4-30.1 29.5-53.4s-28.8-37-51.1-30.8l-20.2 5.6-5.4-21z"],filter:[512,512,[],"f0b0","M3.9 54.9C10.5 40.9 24.5 32 40 32H472c15.5 0 29.5 8.9 36.1 22.9s4.6 30.5-5.2 42.5L320 320.9V448c0 12.1-6.8 23.2-17.7 28.6s-23.8 4.3-33.5-3l-64-48c-8.1-6-12.8-15.5-12.8-25.6V320.9L9 97.3C-.7 85.4-2.8 68.8 3.9 54.9z"],question:[320,512,[10067,10068,61736],"3f","M80 160c0-35.3 28.7-64 64-64h32c35.3 0 64 28.7 64 64v3.6c0 21.8-11.1 42.1-29.4 53.8l-42.2 27.1c-25.2 16.2-40.4 44.1-40.4 74V320c0 17.7 14.3 32 32 32s32-14.3 32-32v-1.4c0-8.2 4.2-15.8 11-20.2l42.2-27.1c36.6-23.6 58.8-64.1 58.8-107.7V160c0-70.7-57.3-128-128-128H144C73.3 32 16 89.3 16 160c0 17.7 14.3 32 32 32s32-14.3 32-32zm80 320a40 40 0 1 0 0-80 40 40 0 1 0 0 80z"],"file-signature":[576,512,[],"f573","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V428.7c-2.7 1.1-5.4 2-8.2 2.7l-60.1 15c-3 .7-6 1.2-9 1.4c-.9 .1-1.8 .2-2.7 .2H240c-6.1 0-11.6-3.4-14.3-8.8l-8.8-17.7c-1.7-3.4-5.1-5.5-8.8-5.5s-7.2 2.1-8.8 5.5l-8.8 17.7c-2.9 5.9-9.2 9.4-15.7 8.8s-12.1-5.1-13.9-11.3L144 381l-9.8 32.8c-6.1 20.3-24.8 34.2-46 34.2H80c-8.8 0-16-7.2-16-16s7.2-16 16-16h8.2c7.1 0 13.3-4.6 15.3-11.4l14.9-49.5c3.4-11.3 13.8-19.1 25.6-19.1s22.2 7.8 25.6 19.1l11.6 38.6c7.4-6.2 16.8-9.7 26.8-9.7c15.9 0 30.4 9 37.5 23.2l4.4 8.8h8.9c-3.1-8.8-3.7-18.4-1.4-27.8l15-60.1c2.8-11.3 8.6-21.5 16.8-29.7L384 203.6V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM549.8 139.7c-15.6-15.6-40.9-15.6-56.6 0l-29.4 29.4 71 71 29.4-29.4c15.6-15.6 15.6-40.9 0-56.6l-14.4-14.4zM311.9 321c-4.1 4.1-7 9.2-8.4 14.9l-15 60.1c-1.4 5.5 .2 11.2 4.2 15.2s9.7 5.6 15.2 4.2l60.1-15c5.6-1.4 10.8-4.3 14.9-8.4L512.1 262.7l-71-71L311.9 321z"],"up-down-left-right":[512,512,["arrows-alt"],"f0b2","M278.6 9.4c-12.5-12.5-32.8-12.5-45.3 0l-64 64c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8h32v96H128V192c0-12.9-7.8-24.6-19.8-29.6s-25.7-2.2-34.9 6.9l-64 64c-12.5 12.5-12.5 32.8 0 45.3l64 64c9.2 9.2 22.9 11.9 34.9 6.9s19.8-16.6 19.8-29.6V288h96v96H192c-12.9 0-24.6 7.8-29.6 19.8s-2.2 25.7 6.9 34.9l64 64c12.5 12.5 32.8 12.5 45.3 0l64-64c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8H288V288h96v32c0 12.9 7.8 24.6 19.8 29.6s25.7 2.2 34.9-6.9l64-64c12.5-12.5 12.5-32.8 0-45.3l-64-64c-9.2-9.2-22.9-11.9-34.9-6.9s-19.8 16.6-19.8 29.6v32H288V128h32c12.9 0 24.6-7.8 29.6-19.8s2.2-25.7-6.9-34.9l-64-64z"],"house-chimney-user":[576,512,[],"e065","M543.8 287.6c17 0 32-14 32-32.1c1-9-3-17-11-24L512 185V64c0-17.7-14.3-32-32-32H448c-17.7 0-32 14.3-32 32v36.7L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32V448c0 35.3 28.7 64 64 64H448.5c35.5 0 64.2-28.8 64-64.3l-.7-160.2h32zM288 160a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM176 400c0-44.2 35.8-80 80-80h64c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16H192c-8.8 0-16-7.2-16-16z"],"hand-holding-heart":[576,512,[],"f4be","M163.9 136.9c-29.4-29.8-29.4-78.2 0-108s77-29.8 106.4 0l17.7 18 17.7-18c29.4-29.8 77-29.8 106.4 0s29.4 78.2 0 108L310.5 240.1c-6.2 6.3-14.3 9.4-22.5 9.4s-16.3-3.1-22.5-9.4L163.9 136.9zM568.2 336.3c13.1 17.8 9.3 42.8-8.5 55.9L433.1 485.5c-23.4 17.2-51.6 26.5-80.7 26.5H192 32c-17.7 0-32-14.3-32-32V416c0-17.7 14.3-32 32-32H68.8l44.9-36c22.7-18.2 50.9-28 80-28H272h16 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H288 272c-8.8 0-16 7.2-16 16s7.2 16 16 16H392.6l119.7-88.2c17.8-13.1 42.8-9.3 55.9 8.5zM193.6 384l0 0-.9 0c.3 0 .6 0 .9 0z"],"puzzle-piece":[512,512,[129513],"f12e","M192 104.8c0-9.2-5.8-17.3-13.2-22.8C167.2 73.3 160 61.3 160 48c0-26.5 28.7-48 64-48s64 21.5 64 48c0 13.3-7.2 25.3-18.8 34c-7.4 5.5-13.2 13.6-13.2 22.8c0 12.8 10.4 23.2 23.2 23.2H336c26.5 0 48 21.5 48 48v56.8c0 12.8 10.4 23.2 23.2 23.2c9.2 0 17.3-5.8 22.8-13.2c8.7-11.6 20.7-18.8 34-18.8c26.5 0 48 28.7 48 64s-21.5 64-48 64c-13.3 0-25.3-7.2-34-18.8c-5.5-7.4-13.6-13.2-22.8-13.2c-12.8 0-23.2 10.4-23.2 23.2V464c0 26.5-21.5 48-48 48H279.2c-12.8 0-23.2-10.4-23.2-23.2c0-9.2 5.8-17.3 13.2-22.8c11.6-8.7 18.8-20.7 18.8-34c0-26.5-28.7-48-64-48s-64 21.5-64 48c0 13.3 7.2 25.3 18.8 34c7.4 5.5 13.2 13.6 13.2 22.8c0 12.8-10.4 23.2-23.2 23.2H48c-26.5 0-48-21.5-48-48V343.2C0 330.4 10.4 320 23.2 320c9.2 0 17.3 5.8 22.8 13.2C54.7 344.8 66.7 352 80 352c26.5 0 48-28.7 48-64s-21.5-64-48-64c-13.3 0-25.3 7.2-34 18.8C40.5 250.2 32.4 256 23.2 256C10.4 256 0 245.6 0 232.8V176c0-26.5 21.5-48 48-48H168.8c12.8 0 23.2-10.4 23.2-23.2z"],"money-check":[576,512,[],"f53c","M64 64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H64zm48 160H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zM96 336c0-8.8 7.2-16 16-16H464c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16zM376 160h80c13.3 0 24 10.7 24 24v48c0 13.3-10.7 24-24 24H376c-13.3 0-24-10.7-24-24V184c0-13.3 10.7-24 24-24z"],"star-half-stroke":[576,512,["star-half-alt"],"f5c0","M288 376.4l.1-.1 26.4 14.1 85.2 45.5-16.5-97.6-4.8-28.7 20.7-20.5 70.1-69.3-96.1-14.2-29.3-4.3-12.9-26.6L288.1 86.9l-.1 .3V376.4zm175.1 98.3c2 12-3 24.2-12.9 31.3s-23 8-33.8 2.3L288.1 439.8 159.8 508.3C149 514 135.9 513.1 126 506s-14.9-19.3-12.9-31.3L137.8 329 33.6 225.9c-8.6-8.5-11.7-21.2-7.9-32.7s13.7-19.9 25.7-21.7L195 150.3 259.4 18c5.4-11 16.5-18 28.8-18s23.4 7 28.8 18l64.3 132.3 143.6 21.2c12 1.8 22 10.2 25.7 21.7s.7 24.2-7.9 32.7L438.5 329l24.6 145.7z"],code:[640,512,[],"f121","M392.8 1.2c-17-4.9-34.7 5-39.6 22l-128 448c-4.9 17 5 34.7 22 39.6s34.7-5 39.6-22l128-448c4.9-17-5-34.7-22-39.6zm80.6 120.1c-12.5 12.5-12.5 32.8 0 45.3L562.7 256l-89.4 89.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l112-112c12.5-12.5 12.5-32.8 0-45.3l-112-112c-12.5-12.5-32.8-12.5-45.3 0zm-306.7 0c-12.5-12.5-32.8-12.5-45.3 0l-112 112c-12.5 12.5-12.5 32.8 0 45.3l112 112c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L77.3 256l89.4-89.4c12.5-12.5 12.5-32.8 0-45.3z"],"whiskey-glass":[512,512,[129347,"glass-whiskey"],"f7a0","M32 32c-9.3 0-18.1 4-24.2 11.1S-1 59.4 .3 68.6l50 342.9c5.7 39.3 39.4 68.5 79.2 68.5h253c39.7 0 73.4-29.1 79.2-68.5l50-342.9c1.3-9.2-1.4-18.5-7.5-25.5S489.3 32 480 32H32zM87.7 224L69 96H443L424.3 224H87.7z"],"building-circle-exclamation":[640,512,[],"e4d3","M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c15.1 0 28.5-6.9 37.3-17.8C340.4 462.2 320 417.5 320 368c0-54.7 24.9-103.5 64-135.8V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z"],"magnifying-glass-chart":[512,512,[],"e522","M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zm-312 8v64c0 13.3 10.7 24 24 24s24-10.7 24-24l0-64c0-13.3-10.7-24-24-24s-24 10.7-24 24zm80-96V280c0 13.3 10.7 24 24 24s24-10.7 24-24V120c0-13.3-10.7-24-24-24s-24 10.7-24 24zm80 64v96c0 13.3 10.7 24 24 24s24-10.7 24-24V184c0-13.3-10.7-24-24-24s-24 10.7-24 24z"],"arrow-up-right-from-square":[512,512,["external-link"],"f08e","M320 0c-17.7 0-32 14.3-32 32s14.3 32 32 32h82.7L201.4 265.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L448 109.3V192c0 17.7 14.3 32 32 32s32-14.3 32-32V32c0-17.7-14.3-32-32-32H320zM80 32C35.8 32 0 67.8 0 112V432c0 44.2 35.8 80 80 80H400c44.2 0 80-35.8 80-80V320c0-17.7-14.3-32-32-32s-32 14.3-32 32V432c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16H192c17.7 0 32-14.3 32-32s-14.3-32-32-32H80z"],"cubes-stacked":[448,512,[],"e4e6","M192 64v64c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V64c0-17.7-14.3-32-32-32H224c-17.7 0-32 14.3-32 32zM82.7 207c-15.3 8.8-20.5 28.4-11.7 43.7l32 55.4c8.8 15.3 28.4 20.5 43.7 11.7l55.4-32c15.3-8.8 20.5-28.4 11.7-43.7l-32-55.4c-8.8-15.3-28.4-20.5-43.7-11.7L82.7 207zM288 192c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32H288zm64 160c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V384c0-17.7-14.3-32-32-32H352zM160 384v64c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V384c0-17.7-14.3-32-32-32H192c-17.7 0-32 14.3-32 32zM32 352c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H96c17.7 0 32-14.3 32-32V384c0-17.7-14.3-32-32-32H32z"],"won-sign":[512,512,[8361,"krw","won"],"f159","M62.4 53.9C56.8 37.1 38.6 28.1 21.9 33.6S-3.9 57.4 1.6 74.1L51.6 224H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H72.9l56.7 170.1c4.5 13.5 17.4 22.4 31.6 21.9s26.4-10.4 29.8-24.2L233 288h46L321 455.8c3.4 13.8 15.6 23.7 29.8 24.2s27.1-8.4 31.6-21.9L439.1 288H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H460.4l50-149.9c5.6-16.8-3.5-34.9-20.2-40.5s-34.9 3.5-40.5 20.2L392.9 224H329L287 56.2C283.5 42 270.7 32 256 32s-27.5 10-31 24.2L183 224h-64L62.4 53.9zm78 234.1H167l-11.4 45.6L140.4 288zM249 224l7-28.1 7 28.1H249zm96 64h26.6l-15.2 45.6L345 288z"],"virus-covid":[512,512,[],"e4a8","M192 24c0-13.3 10.7-24 24-24h80c13.3 0 24 10.7 24 24s-10.7 24-24 24H280V81.6c30.7 4.2 58.8 16.3 82.3 34.1L386.1 92 374.8 80.6c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l56.6 56.6c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L420 125.9l-23.8 23.8c17.9 23.5 29.9 51.7 34.1 82.3H464V216c0-13.3 10.7-24 24-24s24 10.7 24 24v80c0 13.3-10.7 24-24 24s-24-10.7-24-24V280H430.4c-4.2 30.7-16.3 58.8-34.1 82.3L420 386.1l11.3-11.3c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-56.6 56.6c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9L386.1 420l-23.8-23.8c-23.5 17.9-51.7 29.9-82.3 34.1V464h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H216c-13.3 0-24-10.7-24-24s10.7-24 24-24h16V430.4c-30.7-4.2-58.8-16.3-82.3-34.1L125.9 420l11.3 11.3c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L46.7 408.7c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0L92 386.1l23.8-23.8C97.9 338.8 85.8 310.7 81.6 280H48v16c0 13.3-10.7 24-24 24s-24-10.7-24-24V216c0-13.3 10.7-24 24-24s24 10.7 24 24v16H81.6c4.2-30.7 16.3-58.8 34.1-82.3L92 125.9 80.6 137.2c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l56.6-56.6c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9L125.9 92l23.8 23.8c23.5-17.9 51.7-29.9 82.3-34.1V48H216c-13.3 0-24-10.7-24-24zm48 200a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zm64 104a24 24 0 1 0 0-48 24 24 0 1 0 0 48z"],"austral-sign":[448,512,[],"e0a9","M253.5 51.7C248.6 39.8 236.9 32 224 32s-24.6 7.8-29.5 19.7L122.7 224H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H96L82.7 320H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H56L34.5 435.7c-6.8 16.3 .9 35 17.2 41.8s35-.9 41.8-17.2L125.3 384H322.7l31.8 76.3c6.8 16.3 25.5 24 41.8 17.2s24-25.5 17.2-41.8L392 384h24c17.7 0 32-14.3 32-32s-14.3-32-32-32H365.3L352 288h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H325.3L253.5 51.7zM256 224H192l32-76.8L256 224zm-90.7 64H282.7L296 320H152l13.3-32z"],f:[320,512,[102],"46","M64 32C28.7 32 0 60.7 0 96V256 448c0 17.7 14.3 32 32 32s32-14.3 32-32V288H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H64V96H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H64z"],leaf:[512,512,[],"f06c","M272 96c-78.6 0-145.1 51.5-167.7 122.5c33.6-17 71.5-26.5 111.7-26.5h88c8.8 0 16 7.2 16 16s-7.2 16-16 16H288 216s0 0 0 0c-16.6 0-32.7 1.9-48.3 5.4c-25.9 5.9-49.9 16.4-71.4 30.7c0 0 0 0 0 0C38.3 298.8 0 364.9 0 440v16c0 13.3 10.7 24 24 24s24-10.7 24-24V440c0-48.7 20.7-92.5 53.8-123.2C121.6 392.3 190.3 448 272 448l1 0c132.1-.7 239-130.9 239-291.4c0-42.6-7.5-83.1-21.1-119.6c-2.6-6.9-12.7-6.6-16.2-.1C455.9 72.1 418.7 96 376 96L272 96z"],road:[576,512,[128739],"f018","M256 32H181.2c-27.1 0-51.3 17.1-60.3 42.6L3.1 407.2C1.1 413 0 419.2 0 425.4C0 455.5 24.5 480 54.6 480H256V416c0-17.7 14.3-32 32-32s32 14.3 32 32v64H521.4c30.2 0 54.6-24.5 54.6-54.6c0-6.2-1.1-12.4-3.1-18.2L455.1 74.6C446 49.1 421.9 32 394.8 32H320V96c0 17.7-14.3 32-32 32s-32-14.3-32-32V32zm64 192v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V224c0-17.7 14.3-32 32-32s32 14.3 32 32z"],taxi:[512,512,[128662,"cab"],"f1ba","M192 0c-17.7 0-32 14.3-32 32V64c0 .1 0 .1 0 .2c-38.6 2.2-72.3 27.3-85.2 64.1L39.6 228.8C16.4 238.4 0 261.3 0 288V432v48c0 17.7 14.3 32 32 32H64c17.7 0 32-14.3 32-32V432H416v48c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V432 288c0-26.7-16.4-49.6-39.6-59.2L437.2 128.3c-12.9-36.8-46.6-62-85.2-64.1c0-.1 0-.1 0-.2V32c0-17.7-14.3-32-32-32H192zM165.4 128H346.6c13.6 0 25.7 8.6 30.2 21.4L402.9 224H109.1l26.1-74.6c4.5-12.8 16.6-21.4 30.2-21.4zM96 288a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm288 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],"person-circle-plus":[576,512,[],"e541","M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l44.9 74.7c-16.1 17.6-28.6 38.5-36.6 61.5c-1.9-1.8-3.5-3.9-4.9-6.3L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152zM432 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm16 80c0-8.8-7.2-16-16-16s-16 7.2-16 16v48H368c-8.8 0-16 7.2-16 16s7.2 16 16 16h48v48c0 8.8 7.2 16 16 16s16-7.2 16-16V384h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H448V304z"],"chart-pie":[576,512,["pie-chart"],"f200","M304 240V16.6c0-9 7-16.6 16-16.6C443.7 0 544 100.3 544 224c0 9-7.6 16-16.6 16H304zM32 272C32 150.7 122.1 50.3 239 34.3c9.2-1.3 17 6.1 17 15.4V288L412.5 444.5c6.7 6.7 6.2 17.7-1.5 23.1C371.8 495.6 323.8 512 272 512C139.5 512 32 404.6 32 272zm526.4 16c9.3 0 16.6 7.8 15.4 17c-7.7 55.9-34.6 105.6-73.9 142.3c-6 5.6-15.4 5.2-21.2-.7L320 288H558.4z"],"bolt-lightning":[384,512,[],"e0b7","M0 256L28.5 28c2-16 15.6-28 31.8-28H228.9c15 0 27.1 12.1 27.1 27.1c0 3.2-.6 6.5-1.7 9.5L208 160H347.3c20.2 0 36.7 16.4 36.7 36.7c0 7.4-2.2 14.6-6.4 20.7l-192.2 281c-5.9 8.6-15.6 13.7-25.9 13.7h-2.9c-15.7 0-28.5-12.8-28.5-28.5c0-2.3 .3-4.6 .9-6.9L176 288H32c-17.7 0-32-14.3-32-32z"],"sack-xmark":[512,512,[],"e56a","M192 96H320l47.4-71.1C374.5 14.2 366.9 0 354.1 0H157.9c-12.8 0-20.4 14.2-13.3 24.9L192 96zm128 32H192c-3.8 2.5-8.1 5.3-13 8.4l0 0 0 0C122.3 172.7 0 250.9 0 416c0 53 43 96 96 96H416c53 0 96-43 96-96c0-165.1-122.3-243.3-179-279.6c-4.8-3.1-9.2-5.9-13-8.4zM289.9 336l47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47z"],"file-excel":[384,512,[],"f1c3","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM155.7 250.2L192 302.1l36.3-51.9c7.6-10.9 22.6-13.5 33.4-5.9s13.5 22.6 5.9 33.4L221.3 344l46.4 66.2c7.6 10.9 5 25.8-5.9 33.4s-25.8 5-33.4-5.9L192 385.8l-36.3 51.9c-7.6 10.9-22.6 13.5-33.4 5.9s-13.5-22.6-5.9-33.4L162.7 344l-46.4-66.2c-7.6-10.9-5-25.8 5.9-33.4s25.8-5 33.4 5.9z"],"file-contract":[384,512,[],"f56c","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM80 64h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16s7.2-16 16-16zm54.2 253.8c-6.1 20.3-24.8 34.2-46 34.2H80c-8.8 0-16-7.2-16-16s7.2-16 16-16h8.2c7.1 0 13.3-4.6 15.3-11.4l14.9-49.5c3.4-11.3 13.8-19.1 25.6-19.1s22.2 7.7 25.6 19.1l11.6 38.6c7.4-6.2 16.8-9.7 26.8-9.7c15.9 0 30.4 9 37.5 23.2l4.4 8.8H304c8.8 0 16 7.2 16 16s-7.2 16-16 16H240c-6.1 0-11.6-3.4-14.3-8.8l-8.8-17.7c-1.7-3.4-5.1-5.5-8.8-5.5s-7.2 2.1-8.8 5.5l-8.8 17.7c-2.9 5.9-9.2 9.4-15.7 8.8s-12.1-5.1-13.9-11.3L144 349l-9.8 32.8z"],"fish-fins":[576,512,[],"e4f2","M275.2 38.4c-10.6-8-25-8.5-36.3-1.5S222 57.3 224.6 70.3l9.7 48.6c-19.4 9-36.9 19.9-52.4 31.5c-15.3 11.5-29 23.9-40.7 36.3L48.1 132.4c-12.5-7.3-28.4-5.3-38.6 4.9S-3 163.3 4.2 175.9L50 256 4.2 336.1c-7.2 12.6-5 28.4 5.3 38.6s26.1 12.2 38.6 4.9l93.1-54.3c11.8 12.3 25.4 24.8 40.7 36.3c15.5 11.6 33 22.5 52.4 31.5l-9.7 48.6c-2.6 13 3.1 26.3 14.3 33.3s25.6 6.5 36.3-1.5l77.6-58.2c54.9-4 101.5-27 137.2-53.8c39.2-29.4 67.2-64.7 81.6-89.5c5.8-9.9 5.8-22.2 0-32.1c-14.4-24.8-42.5-60.1-81.6-89.5c-35.8-26.8-82.3-49.8-137.2-53.8L275.2 38.4zM384 256a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],"building-flag":[640,512,[],"e4d5","M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zM448 0c-17.7 0-32 14.3-32 32V512h64V192H624c8.8 0 16-7.2 16-16V48c0-8.8-7.2-16-16-16H480c0-17.7-14.3-32-32-32z"],"face-grin-beam":[512,512,[128516,"grin-beam"],"f582","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM388.1 312.8c12.3-3.8 24.3 6.9 19.3 18.7C382.4 390.6 324.2 432 256.3 432s-126.2-41.4-151.1-100.5c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19zm-170.5-84l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z"],"object-ungroup":[640,512,[],"f248","M32 119.4C12.9 108.4 0 87.7 0 64C0 28.7 28.7 0 64 0c23.7 0 44.4 12.9 55.4 32H328.6C339.6 12.9 360.3 0 384 0c35.3 0 64 28.7 64 64c0 23.7-12.9 44.4-32 55.4V232.6c19.1 11.1 32 31.7 32 55.4c0 35.3-28.7 64-64 64c-23.7 0-44.4-12.9-55.4-32H119.4c-11.1 19.1-31.7 32-55.4 32c-35.3 0-64-28.7-64-64c0-23.7 12.9-44.4 32-55.4V119.4zM119.4 96c-5.6 9.7-13.7 17.8-23.4 23.4V232.6c9.7 5.6 17.8 13.7 23.4 23.4H328.6c5.6-9.7 13.7-17.8 23.4-23.4V119.4c-9.7-5.6-17.8-13.7-23.4-23.4H119.4zm192 384c-11.1 19.1-31.7 32-55.4 32c-35.3 0-64-28.7-64-64c0-23.7 12.9-44.4 32-55.4V352h64v40.6c9.7 5.6 17.8 13.7 23.4 23.4H520.6c5.6-9.7 13.7-17.8 23.4-23.4V279.4c-9.7-5.6-17.8-13.7-23.4-23.4h-46c-5.4-15.4-14.6-28.9-26.5-39.6V192h72.6c11.1-19.1 31.7-32 55.4-32c35.3 0 64 28.7 64 64c0 23.7-12.9 44.4-32 55.4V392.6c19.1 11.1 32 31.7 32 55.4c0 35.3-28.7 64-64 64c-23.7 0-44.4-12.9-55.4-32H311.4z"],poop:[512,512,[],"f619","M254.4 6.6c3.5-4.3 9-6.5 14.5-5.7C315.8 7.2 352 47.4 352 96c0 11.2-1.9 22-5.5 32H352c35.3 0 64 28.7 64 64c0 19.1-8.4 36.3-21.7 48H408c39.8 0 72 32.2 72 72c0 23.2-11 43.8-28 57c34.1 5.7 60 35.3 60 71c0 39.8-32.2 72-72 72H72c-39.8 0-72-32.2-72-72c0-35.7 25.9-65.3 60-71c-17-13.2-28-33.8-28-57c0-39.8 32.2-72 72-72h13.7C104.4 228.3 96 211.1 96 192c0-35.3 28.7-64 64-64h16.2c44.1-.1 79.8-35.9 79.8-80c0-9.2-1.5-17.9-4.3-26.1c-1.8-5.2-.8-11.1 2.8-15.4z"],"location-pin":[384,512,["map-marker"],"f041","M384 192c0 87.4-117 243-168.3 307.2c-12.3 15.3-35.1 15.3-47.4 0C117 435 0 279.4 0 192C0 86 86 0 192 0S384 86 384 192z"],kaaba:[576,512,[128331],"f66b","M60 120l228 71.2L516 120 288 48.8 60 120zM278.5 1.5c6.2-1.9 12.9-1.9 19.1 0l256 80C566.9 85.6 576 98 576 112v16 0 21.2L292.8 237.7c-3.1 1-6.4 1-9.5 0L0 149.2V128 112C0 98 9.1 85.6 22.5 81.5l256-80zm23.9 266.8L576 182.8v46.5l-52.8 16.5c-8.4 2.6-13.1 11.6-10.5 20s11.6 13.1 20 10.5L576 262.8V400c0 14-9.1 26.4-22.5 30.5l-256 80c-6.2 1.9-12.9 1.9-19.1 0l-256-80C9.1 426.4 0 414 0 400V262.8l43.2 13.5c8.4 2.6 17.4-2.1 20-10.5s-2.1-17.4-10.5-20L0 229.2V182.8l273.7 85.5c9.3 2.9 19.3 2.9 28.6 0zm-185.5-2.6c-8.4-2.6-17.4 2.1-20 10.5s2.1 17.4 10.5 20l64 20c8.4 2.6 17.4-2.1 20-10.5s-2.1-17.4-10.5-20l-64-20zm352 30.5c8.4-2.6 13.1-11.6 10.5-20s-11.6-13.1-20-10.5l-64 20c-8.4 2.6-13.1 11.6-10.5 20s11.6 13.1 20 10.5l64-20zm-224 9.5c-8.4-2.6-17.4 2.1-20 10.5s2.1 17.4 10.5 20l38.5 12c9.3 2.9 19.3 2.9 28.6 0l38.5-12c8.4-2.6 13.1-11.6 10.5-20s-11.6-13.1-20-10.5l-38.5 12c-3.1 1-6.4 1-9.5 0l-38.5-12z"],"toilet-paper":[640,512,[129531],"f71e","M444.2 0C397.2 49.6 384 126.5 384 192c0 158.8-27.3 247-42.7 283.9c-10 24-33.2 36.1-55.4 36.1H48c-11.5 0-22.2-6.2-27.8-16.2s-5.6-22.3 .4-32.2c9.8-17.7 15.4-38.2 20.5-57.7C52.3 362.8 64 293.5 64 192C64 86 107 0 160 0H444.2zM512 384c-53 0-96-86-96-192S459 0 512 0s96 86 96 192s-43 192-96 192zm0-128c17.7 0 32-28.7 32-64s-14.3-64-32-64s-32 28.7-32 64s14.3 64 32 64zM144 208a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm64 0a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm48 16a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm80-16a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z"],"helmet-safety":[576,512,["hard-hat","hat-hard"],"f807","M256 32c-17.7 0-32 14.3-32 32v2.3 99.6c0 5.6-4.5 10.1-10.1 10.1c-3.6 0-7-1.9-8.8-5.1L157.1 87C83 123.5 32 199.8 32 288v64H544l0-66.4c-.9-87.2-51.7-162.4-125.1-198.6l-48 83.9c-1.8 3.2-5.2 5.1-8.8 5.1c-5.6 0-10.1-4.5-10.1-10.1V66.3 64c0-17.7-14.3-32-32-32H256zM16.6 384C7.4 384 0 391.4 0 400.6c0 4.7 2 9.2 5.8 11.9C27.5 428.4 111.8 480 288 480s260.5-51.6 282.2-67.5c3.8-2.8 5.8-7.2 5.8-11.9c0-9.2-7.4-16.6-16.6-16.6H16.6z"],eject:[448,512,[9167],"f052","M224 32c13.5 0 26.3 5.6 35.4 15.6l176 192c12.9 14 16.2 34.3 8.6 51.8S419 320 400 320H48c-19 0-36.3-11.2-43.9-28.7s-4.3-37.7 8.6-51.8l176-192C197.7 37.6 210.5 32 224 32zM0 432c0-26.5 21.5-48 48-48H400c26.5 0 48 21.5 48 48s-21.5 48-48 48H48c-26.5 0-48-21.5-48-48z"],"circle-right":[512,512,[61838,"arrow-alt-circle-right"],"f35a","M0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zM294.6 135.1l99.9 107.1c3.5 3.8 5.5 8.7 5.5 13.8s-2 10.1-5.5 13.8L294.6 376.9c-4.2 4.5-10.1 7.1-16.3 7.1C266 384 256 374 256 361.7l0-57.7-96 0c-17.7 0-32-14.3-32-32l0-32c0-17.7 14.3-32 32-32l96 0 0-57.7c0-12.3 10-22.3 22.3-22.3c6.2 0 12.1 2.6 16.3 7.1z"],"plane-circle-check":[640,512,[],"e555","M256 0c-35 0-64 59.5-64 93.7v84.6L8.1 283.4c-5 2.8-8.1 8.2-8.1 13.9v65.5c0 10.6 10.2 18.3 20.4 15.4l171.6-49 0 70.9-57.6 43.2c-4 3-6.4 7.8-6.4 12.8v42c0 7.8 6.3 14 14 14c1.3 0 2.6-.2 3.9-.5L256 480l110.1 31.5c1.3 .4 2.6 .5 3.9 .5c6 0 11.1-3.7 13.1-9C344.5 470.7 320 422.2 320 368c0-60.6 30.6-114 77.1-145.6L320 178.3V93.7C320 59.5 292 0 256 0zM640 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z"],"face-rolling-eyes":[512,512,[128580,"meh-rolling-eyes"],"f5a5","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM192 368H320c8.8 0 16 7.2 16 16s-7.2 16-16 16H192c-8.8 0-16-7.2-16-16s7.2-16 16-16zm32-144c0 35.3-28.7 64-64 64s-64-28.7-64-64c0-26 15.5-48.4 37.8-58.4c-3.7 5.2-5.8 11.6-5.8 18.4c0 17.7 14.3 32 32 32s32-14.3 32-32c0-6.9-2.2-13.2-5.8-18.4C208.5 175.6 224 198 224 224zm128 64c-35.3 0-64-28.7-64-64c0-26 15.5-48.4 37.8-58.4c-3.7 5.2-5.8 11.6-5.8 18.4c0 17.7 14.3 32 32 32s32-14.3 32-32c0-6.9-2.2-13.2-5.8-18.4C400.5 175.6 416 198 416 224c0 35.3-28.7 64-64 64z"],"object-group":[576,512,[],"f247","M32 119.4C12.9 108.4 0 87.7 0 64C0 28.7 28.7 0 64 0c23.7 0 44.4 12.9 55.4 32H456.6C467.6 12.9 488.3 0 512 0c35.3 0 64 28.7 64 64c0 23.7-12.9 44.4-32 55.4V392.6c19.1 11.1 32 31.7 32 55.4c0 35.3-28.7 64-64 64c-23.7 0-44.4-12.9-55.4-32H119.4c-11.1 19.1-31.7 32-55.4 32c-35.3 0-64-28.7-64-64c0-23.7 12.9-44.4 32-55.4V119.4zM456.6 96H119.4c-5.6 9.7-13.7 17.8-23.4 23.4V392.6c9.7 5.6 17.8 13.7 23.4 23.4H456.6c5.6-9.7 13.7-17.8 23.4-23.4V119.4c-9.7-5.6-17.8-13.7-23.4-23.4zM128 160c0-17.7 14.3-32 32-32H288c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32V160zM256 320h32c35.3 0 64-28.7 64-64V224h64c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H288c-17.7 0-32-14.3-32-32V320z"],"chart-line":[512,512,["line-chart"],"f201","M64 64c0-17.7-14.3-32-32-32S0 46.3 0 64V400c0 44.2 35.8 80 80 80H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H80c-8.8 0-16-7.2-16-16V64zm406.6 86.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L320 210.7l-57.4-57.4c-12.5-12.5-32.8-12.5-45.3 0l-112 112c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L240 221.3l57.4 57.4c12.5 12.5 32.8 12.5 45.3 0l128-128z"],"mask-ventilator":[640,512,[],"e524","M159.1 176C139.4 219.2 128 264.7 128 300.8c0 15.9 2.2 31.4 6.3 46l-31.8-7.9C70.5 330.9 48 302.1 48 269V184c0-4.4 3.6-8 8-8H159.1zm26-48H56c-30.9 0-56 25.1-56 56v85c0 55.1 37.5 103.1 90.9 116.4l71.3 17.8c22.7 30.5 55.4 54.1 93.8 66.6V393.3c-19.7-16.4-32-40.3-32-66.9c0-49.5 43-134.4 96-134.4c52.5 0 96 84.9 96 134.4c0 26.7-12.4 50.4-32 66.8v76.6c38-12.6 70.6-36 93.5-66.4l71.6-17.9C602.5 372.1 640 324.1 640 269V184c0-30.9-25.1-56-56-56H454.5C419.7 73.8 372.1 32 320 32c-52.6 0-100.2 41.8-134.9 96zm295.6 48H584c4.4 0 8 3.6 8 8v85c0 33-22.5 61.8-54.5 69.9l-31.8 8c4.2-14.7 6.4-30.1 6.4-46.1c0-36.1-11.6-81.6-31.3-124.8zM288 320V512h64V320c0-17.7-14.3-32-32-32s-32 14.3-32 32z"],"arrow-right":[448,512,[8594],"f061","M438.6 278.6c12.5-12.5 12.5-32.8 0-45.3l-160-160c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L338.8 224 32 224c-17.7 0-32 14.3-32 32s14.3 32 32 32l306.7 0L233.4 393.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l160-160z"],"signs-post":[512,512,["map-signs"],"f277","M224 32H64C46.3 32 32 46.3 32 64v64c0 17.7 14.3 32 32 32H441.4c4.2 0 8.3-1.7 11.3-4.7l48-48c6.2-6.2 6.2-16.4 0-22.6l-48-48c-3-3-7.1-4.7-11.3-4.7H288c0-17.7-14.3-32-32-32s-32 14.3-32 32zM480 256c0-17.7-14.3-32-32-32H288V192H224v32H70.6c-4.2 0-8.3 1.7-11.3 4.7l-48 48c-6.2 6.2-6.2 16.4 0 22.6l48 48c3 3 7.1 4.7 11.3 4.7H448c17.7 0 32-14.3 32-32V256zM288 480V384H224v96c0 17.7 14.3 32 32 32s32-14.3 32-32z"],"cash-register":[512,512,[],"f788","M64 0C46.3 0 32 14.3 32 32V96c0 17.7 14.3 32 32 32h80v32H87c-31.6 0-58.5 23.1-63.3 54.4L1.1 364.1C.4 368.8 0 373.6 0 378.4V448c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V378.4c0-4.8-.4-9.6-1.1-14.4L488.2 214.4C483.5 183.1 456.6 160 425 160H208V128h80c17.7 0 32-14.3 32-32V32c0-17.7-14.3-32-32-32H64zM96 48H256c8.8 0 16 7.2 16 16s-7.2 16-16 16H96c-8.8 0-16-7.2-16-16s7.2-16 16-16zM64 432c0-8.8 7.2-16 16-16H432c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16zm48-168a24 24 0 1 1 0-48 24 24 0 1 1 0 48zm120-24a24 24 0 1 1 -48 0 24 24 0 1 1 48 0zM160 344a24 24 0 1 1 0-48 24 24 0 1 1 0 48zM328 240a24 24 0 1 1 -48 0 24 24 0 1 1 48 0zM256 344a24 24 0 1 1 0-48 24 24 0 1 1 0 48zM424 240a24 24 0 1 1 -48 0 24 24 0 1 1 48 0zM352 344a24 24 0 1 1 0-48 24 24 0 1 1 0 48z"],"person-circle-question":[576,512,[],"e542","M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l44.9 74.7c-16.1 17.6-28.6 38.5-36.6 61.5c-1.9-1.8-3.5-3.9-4.9-6.3L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152zM432 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm0 240a24 24 0 1 0 0-48 24 24 0 1 0 0 48zM368 321.6V328c0 8.8 7.2 16 16 16s16-7.2 16-16v-6.4c0-5.3 4.3-9.6 9.6-9.6h40.5c7.7 0 13.9 6.2 13.9 13.9c0 5.2-2.9 9.9-7.4 12.3l-32 16.8c-5.3 2.8-8.6 8.2-8.6 14.2V384c0 8.8 7.2 16 16 16s16-7.2 16-16v-5.1l23.5-12.3c15.1-7.9 24.5-23.6 24.5-40.6c0-25.4-20.6-45.9-45.9-45.9H409.6c-23 0-41.6 18.6-41.6 41.6z"],h:[384,512,[104],"48","M320 256l0 192c0 17.7 14.3 32 32 32s32-14.3 32-32l0-224V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V192L64 192 64 64c0-17.7-14.3-32-32-32S0 46.3 0 64V448c0 17.7 14.3 32 32 32s32-14.3 32-32l0-192 256 0z"],tarp:[576,512,[],"e57b","M576 128c0-35.3-28.7-64-64-64H64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64l352 0 0-128c0-17.7 14.3-32 32-32H576V128zM448 448L576 320H448l0 128zM96 128a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"screwdriver-wrench":[512,512,["tools"],"f7d9","M78.6 5C69.1-2.4 55.6-1.5 47 7L7 47c-8.5 8.5-9.4 22-2.1 31.6l80 104c4.5 5.9 11.6 9.4 19 9.4h54.1l109 109c-14.7 29-10 65.4 14.3 89.6l112 112c12.5 12.5 32.8 12.5 45.3 0l64-64c12.5-12.5 12.5-32.8 0-45.3l-112-112c-24.2-24.2-60.6-29-89.6-14.3l-109-109V104c0-7.5-3.5-14.5-9.4-19L78.6 5zM19.9 396.1C7.2 408.8 0 426.1 0 444.1C0 481.6 30.4 512 67.9 512c18 0 35.3-7.2 48-19.9L233.7 374.3c-7.8-20.9-9-43.6-3.6-65.1l-61.7-61.7L19.9 396.1zM512 144c0-10.5-1.1-20.7-3.2-30.5c-2.4-11.2-16.1-14.1-24.2-6l-63.9 63.9c-3 3-7.1 4.7-11.3 4.7H352c-8.8 0-16-7.2-16-16V102.6c0-4.2 1.7-8.3 4.7-11.3l63.9-63.9c8.1-8.1 5.2-21.8-6-24.2C388.7 1.1 378.5 0 368 0C288.5 0 224 64.5 224 144l0 .8 85.3 85.3c36-9.1 75.8 .5 104 28.7L429 274.5c49-23 83-72.8 83-130.5zM56 432a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z"],"arrows-to-eye":[640,512,[],"e4bf","M15 15C24.4 5.7 39.6 5.7 49 15l63 63V40c0-13.3 10.7-24 24-24s24 10.7 24 24v96c0 13.3-10.7 24-24 24H40c-13.3 0-24-10.7-24-24s10.7-24 24-24H78.1L15 49C5.7 39.6 5.7 24.4 15 15zM133.5 243.9C158.6 193.6 222.7 112 320 112s161.4 81.6 186.5 131.9c3.8 7.6 3.8 16.5 0 24.2C481.4 318.4 417.3 400 320 400s-161.4-81.6-186.5-131.9c-3.8-7.6-3.8-16.5 0-24.2zM320 320a64 64 0 1 0 0-128 64 64 0 1 0 0 128zM591 15c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-63 63H600c13.3 0 24 10.7 24 24s-10.7 24-24 24H504c-13.3 0-24-10.7-24-24V40c0-13.3 10.7-24 24-24s24 10.7 24 24V78.1l63-63zM15 497c-9.4-9.4-9.4-24.6 0-33.9l63-63H40c-13.3 0-24-10.7-24-24s10.7-24 24-24h96c13.3 0 24 10.7 24 24v96c0 13.3-10.7 24-24 24s-24-10.7-24-24V433.9L49 497c-9.4 9.4-24.6 9.4-33.9 0zm576 0l-63-63V472c0 13.3-10.7 24-24 24s-24-10.7-24-24V376c0-13.3 10.7-24 24-24h96c13.3 0 24 10.7 24 24s-10.7 24-24 24H561.9l63 63c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0z"],"plug-circle-bolt":[576,512,[],"e55b","M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8c12.3-2.5 24.1-6.4 35.1-11.5c-2.1-10.8-3.1-21.9-3.1-33.3c0-80.3 53.8-148 127.3-169.2c.5-2.2 .7-4.5 .7-6.8c0-17.7-14.3-32-32-32H32zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm47.9-225c4.3 3.7 5.4 9.9 2.6 14.9L452.4 356H488c5.2 0 9.8 3.3 11.4 8.2s-.1 10.3-4.2 13.4l-96 72c-4.5 3.4-10.8 3.2-15.1-.6s-5.4-9.9-2.6-14.9L411.6 380H376c-5.2 0-9.8-3.3-11.4-8.2s.1-10.3 4.2-13.4l96-72c4.5-3.4 10.8-3.2 15.1 .6z"],heart:[512,512,[128153,128154,128155,128156,128420,129293,129294,129505,9829,10084,61578],"f004","M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9L464.4 300.4c30.4-28.3 47.6-68 47.6-109.5v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5z"],"mars-and-venus":[512,512,[9893],"f224","M337.8 14.8C341.5 5.8 350.3 0 360 0H472c13.3 0 24 10.7 24 24V136c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-39-39-24.7 24.7C407 163.3 416 192.6 416 224c0 80.2-59 146.6-136 158.2V408h24c13.3 0 24 10.7 24 24s-10.7 24-24 24H280v32c0 13.3-10.7 24-24 24s-24-10.7-24-24V456H208c-13.3 0-24-10.7-24-24s10.7-24 24-24h24V382.2C155 370.6 96 304.2 96 224c0-88.4 71.6-160 160-160c39.6 0 75.9 14.4 103.8 38.2L382.1 80 343 41c-6.9-6.9-8.9-17.2-5.2-26.2zM448 48l0 0h0v0zM352 224a96 96 0 1 0 -192 0 96 96 0 1 0 192 0z"],"house-user":[576,512,["home-user"],"e1b0","M575.8 255.5c0 18-15 32.1-32 32.1h-32l.7 160.2c.2 35.5-28.5 64.3-64 64.3H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L564.8 231.5c8 7 12 15 11 24zM352 224a64 64 0 1 0 -128 0 64 64 0 1 0 128 0zm-96 96c-44.2 0-80 35.8-80 80c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16c0-44.2-35.8-80-80-80H256z"],"dumpster-fire":[640,512,[],"f794","M49.7 32l90.8 0L114.9 160l-94 0C9.3 160 0 150.7 0 139.1c0-2.1 .3-4.1 .9-6.1L26.8 48.9C29.9 38.9 39.2 32 49.7 32zM272 160l-124.5 0L173.1 32 272 32l0 128zm32 0l0-128 98.9 0 14.4 72.1c-1.7 1.1-3.3 2.4-4.8 3.8c-18.4 16.4-35.4 34-50.5 52.1l-58 0zm209.9-23.7c-1.7 1.6-3.4 3.2-5 4.8C498 129.6 486.7 118.6 475 108c-7.6-6.9-17-10.8-26.6-11.8L435.5 32l90.8 0c10.5 0 19.8 6.9 22.9 16.9L575.1 133c.2 .7 .4 1.4 .5 2.1c-17.8-15-44.3-14.6-61.7 1.2zM325.2 210.7C304.3 244.5 288 282.9 288 318.1c0 49.3 18.6 95.2 49.6 129.9L128 448c0 17.7-14.3 32-32 32s-32-14.3-32-32L44 288l-12 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l4 0-4-32 305.5 0c-4.4 6.2-8.5 12.5-12.3 18.7zm180.6-34.5L518 162.5c5.4-6.1 13.3-8.8 20.9-8.9c7.2 0 14.3 2.6 19.9 7.8c19.7 18.3 39.8 43.2 55 70.6C629 259.2 640 290.2 640 320.2C640 408.8 568.7 480 480 480c-89.6 0-160-71.3-160-159.8c0-37.3 16-73.4 36.8-104.5c20.9-31.3 47.5-59 70.9-80.2c5.7-5.2 13.1-7.7 20.3-7.5c14.1 .3 23.8 11.4 32.7 21.6c0 0 0 0 0 0c2 2.3 4 4.6 6 6.7l19 19.9zM544 368.2c0-36.5-37-73-54.8-88.4c-5.4-4.7-13.1-4.7-18.5 0C453 295.1 416 331.6 416 368.2c0 35.3 28.7 64 64 64s64-28.7 64-64z"],"house-crack":[576,512,[],"e3b1","M543.8 287.6c17 0 32-14 32-32.1c1-9-3-17-11-24L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32V448c0 35.3 28.7 64 64 64H230.4l-31.3-52.2c-4.1-6.8-2.6-15.5 3.5-20.5L288 368l-60.2-82.8c-10.9-15 8.2-33.5 22.8-22l117.9 92.6c8 6.3 8.2 18.4 .4 24.9L288 448l38.4 64H448.5c35.5 0 64.2-28.8 64-64.3l-.7-160.2h32z"],"martini-glass-citrus":[576,512,["cocktail"],"f561","M432 240c53 0 96-43 96-96s-43-96-96-96c-35.5 0-66.6 19.3-83.2 48H296.2C316 40.1 369.3 0 432 0c79.5 0 144 64.5 144 144s-64.5 144-144 144c-27.7 0-53.5-7.8-75.5-21.3l35.4-35.4c12.2 5.6 25.8 8.7 40.1 8.7zM1.8 142.8C5.5 133.8 14.3 128 24 128H392c9.7 0 18.5 5.8 22.2 14.8s1.7 19.3-5.2 26.2l-177 177V464h64c13.3 0 24 10.7 24 24s-10.7 24-24 24H208 120c-13.3 0-24-10.7-24-24s10.7-24 24-24h64V345.9L7 169c-6.9-6.9-8.9-17.2-5.2-26.2z"],"face-surprise":[512,512,[128558,"surprise"],"f5c2","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM176.4 176a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM256 288a64 64 0 1 1 0 128 64 64 0 1 1 0-128z"],"bottle-water":[320,512,[],"e4c5","M120 0l80 0c13.3 0 24 10.7 24 24l0 40L96 64l0-40c0-13.3 10.7-24 24-24zM32 167.5c0-19.5 10-37.6 26.6-47.9l15.8-9.9C88.7 100.7 105.2 96 122.1 96l75.8 0c16.9 0 33.4 4.7 47.7 13.7l15.8 9.9C278 129.9 288 148 288 167.5c0 17-7.5 32.3-19.4 42.6C280.6 221.7 288 238 288 256c0 19.1-8.4 36.3-21.7 48c13.3 11.7 21.7 28.9 21.7 48s-8.4 36.3-21.7 48c13.3 11.7 21.7 28.9 21.7 48c0 35.3-28.7 64-64 64L96 512c-35.3 0-64-28.7-64-64c0-19.1 8.4-36.3 21.7-48C40.4 388.3 32 371.1 32 352s8.4-36.3 21.7-48C40.4 292.3 32 275.1 32 256c0-18 7.4-34.3 19.4-45.9C39.5 199.7 32 184.5 32 167.5zM96 240c0 8.8 7.2 16 16 16l96 0c8.8 0 16-7.2 16-16s-7.2-16-16-16l-96 0c-8.8 0-16 7.2-16 16zm16 112c-8.8 0-16 7.2-16 16s7.2 16 16 16l96 0c8.8 0 16-7.2 16-16s-7.2-16-16-16l-96 0z"],"circle-pause":[512,512,[62092,"pause-circle"],"f28b","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM224 192V320c0 17.7-14.3 32-32 32s-32-14.3-32-32V192c0-17.7 14.3-32 32-32s32 14.3 32 32zm128 0V320c0 17.7-14.3 32-32 32s-32-14.3-32-32V192c0-17.7 14.3-32 32-32s32 14.3 32 32z"],"toilet-paper-slash":[640,512,[],"e072","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-109.7-86C569.9 374 608 291.9 608 192C608 86 565 0 512 0s-96 86-96 192c0 49.1 9.2 93.9 24.4 127.9l-59-46.2c1.6-24.8 2.6-52 2.6-81.6c0-65.5 13.2-142.4 60.2-192H160c-24.8 0-47.4 18.8-64.4 49.6L38.8 5.1zM367.3 385.4L66.5 148.4C64.9 162.4 64 177 64 192c0 101.5-11.7 170.8-23 213.9c-5.1 19.4-10.7 39.9-20.5 57.7c-5.9 9.9-6.1 22.1-.4 32.2S36.5 512 48 512H285.9c22.3 0 45.4-12.1 55.4-36.1c7.4-17.7 17.5-47.2 26-90.6zM544 192c0 35.3-14.3 64-32 64s-32-28.7-32-64s14.3-64 32-64s32 28.7 32 64z"],"apple-whole":[448,512,[127822,127823,"apple-alt"],"f5d1","M224 112c-8.8 0-16-7.2-16-16V80c0-44.2 35.8-80 80-80h16c8.8 0 16 7.2 16 16V32c0 44.2-35.8 80-80 80H224zM0 288c0-76.3 35.7-160 112-160c27.3 0 59.7 10.3 82.7 19.3c18.8 7.3 39.9 7.3 58.7 0c22.9-8.9 55.4-19.3 82.7-19.3c76.3 0 112 83.7 112 160c0 128-80 224-160 224c-16.5 0-38.1-6.6-51.5-11.3c-8.1-2.8-16.9-2.8-25 0c-13.4 4.7-35 11.3-51.5 11.3C80 512 0 416 0 288z"],"kitchen-set":[576,512,[],"e51a","M240 144A96 96 0 1 0 48 144a96 96 0 1 0 192 0zm44.4 32C269.9 240.1 212.5 288 144 288C64.5 288 0 223.5 0 144S64.5 0 144 0c68.5 0 125.9 47.9 140.4 112h71.8c8.8-9.8 21.6-16 35.8-16H496c26.5 0 48 21.5 48 48s-21.5 48-48 48H392c-14.2 0-27-6.2-35.8-16H284.4zM144 80a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM400 240c13.3 0 24 10.7 24 24v8h96c13.3 0 24 10.7 24 24s-10.7 24-24 24H280c-13.3 0-24-10.7-24-24s10.7-24 24-24h96v-8c0-13.3 10.7-24 24-24zM288 464V352H512V464c0 26.5-21.5 48-48 48H336c-26.5 0-48-21.5-48-48zM48 320h80 16 32c26.5 0 48 21.5 48 48s-21.5 48-48 48H160c0 17.7-14.3 32-32 32H64c-17.7 0-32-14.3-32-32V336c0-8.8 7.2-16 16-16zm128 64c8.8 0 16-7.2 16-16s-7.2-16-16-16H160v32h16zM24 464H200c13.3 0 24 10.7 24 24s-10.7 24-24 24H24c-13.3 0-24-10.7-24-24s10.7-24 24-24z"],r:[320,512,[114],"52","M64 32C28.7 32 0 60.7 0 96V288 448c0 17.7 14.3 32 32 32s32-14.3 32-32V320h95.3L261.8 466.4c10.1 14.5 30.1 18 44.6 7.9s18-30.1 7.9-44.6L230.1 309.5C282.8 288.1 320 236.4 320 176c0-79.5-64.5-144-144-144H64zM176 256H64V96H176c44.2 0 80 35.8 80 80s-35.8 80-80 80z"],"temperature-quarter":[320,512,["temperature-1","thermometer-1","thermometer-quarter"],"f2ca","M160 64c-26.5 0-48 21.5-48 48V276.5c0 17.3-7.1 31.9-15.3 42.5C86.2 332.6 80 349.5 80 368c0 44.2 35.8 80 80 80s80-35.8 80-80c0-18.5-6.2-35.4-16.7-48.9c-8.2-10.6-15.3-25.2-15.3-42.5V112c0-26.5-21.5-48-48-48zM48 112C48 50.2 98.1 0 160 0s112 50.1 112 112V276.5c0 .1 .1 .3 .2 .6c.2 .6 .8 1.6 1.7 2.8c18.9 24.4 30.1 55 30.1 88.1c0 79.5-64.5 144-144 144S16 447.5 16 368c0-33.2 11.2-63.8 30.1-88.1c.9-1.2 1.5-2.2 1.7-2.8c.1-.3 .2-.5 .2-.6V112zM208 368c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-20.9 13.4-38.7 32-45.3V272c0-8.8 7.2-16 16-16s16 7.2 16 16v50.7c18.6 6.6 32 24.4 32 45.3z"],cube:[512,512,[],"f1b2","M234.5 5.7c13.9-5 29.1-5 43.1 0l192 68.6C495 83.4 512 107.5 512 134.6V377.4c0 27-17 51.2-42.5 60.3l-192 68.6c-13.9 5-29.1 5-43.1 0l-192-68.6C17 428.6 0 404.5 0 377.4V134.6c0-27 17-51.2 42.5-60.3l192-68.6zM256 66L82.3 128 256 190l173.7-62L256 66zm32 368.6l160-57.1v-188L288 246.6v188z"],"bitcoin-sign":[320,512,[],"e0b4","M48 32C48 14.3 62.3 0 80 0s32 14.3 32 32V64h32V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64c0 1.5-.1 3.1-.3 4.5C254.1 82.2 288 125.1 288 176c0 24.2-7.7 46.6-20.7 64.9c31.7 19.8 52.7 55 52.7 95.1c0 61.9-50.1 112-112 112v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H112v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H41.7C18.7 448 0 429.3 0 406.3V288 265.7 224 101.6C0 80.8 16.8 64 37.6 64H48V32zM64 224H176c26.5 0 48-21.5 48-48s-21.5-48-48-48H64v96zm112 64H64v96H208c26.5 0 48-21.5 48-48s-21.5-48-48-48H176z"],"shield-dog":[512,512,[],"e573","M269.4 2.9C265.2 1 260.7 0 256 0s-9.2 1-13.4 2.9L54.3 82.8c-22 9.3-38.4 31-38.3 57.2c.5 99.2 41.3 280.7 213.6 363.2c16.7 8 36.1 8 52.8 0C454.7 420.7 495.5 239.2 496 140c.1-26.2-16.3-47.9-38.3-57.2L269.4 2.9zM160.9 286.2c4.8 1.2 9.9 1.8 15.1 1.8c35.3 0 64-28.7 64-64V160h44.2c12.1 0 23.2 6.8 28.6 17.7L320 192h64c8.8 0 16 7.2 16 16v32c0 44.2-35.8 80-80 80H272v50.7c0 7.3-5.9 13.3-13.3 13.3c-1.8 0-3.6-.4-5.2-1.1l-98.7-42.3c-6.6-2.8-10.8-9.3-10.8-16.4c0-2.8 .6-5.5 1.9-8l15-30zM160 160h40 8v32 32c0 17.7-14.3 32-32 32s-32-14.3-32-32V176c0-8.8 7.2-16 16-16zm128 48a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z"],"solar-panel":[640,512,[],"f5ba","M122.2 0C91.7 0 65.5 21.5 59.5 51.4L8.3 307.4C.4 347 30.6 384 71 384H288v64H224c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H352V384H569c40.4 0 70.7-36.9 62.8-76.6l-51.2-256C574.5 21.5 548.3 0 517.8 0H122.2zM260.9 64H379.1l10.4 104h-139L260.9 64zM202.3 168H101.4L122.2 64h90.4L202.3 168zM91.8 216H197.5L187.1 320H71L91.8 216zm153.9 0H394.3l10.4 104-169.4 0 10.4-104zm196.8 0H548.2L569 320h-116L442.5 216zm96-48H437.7L427.3 64h90.4l31.4-6.3L517.8 64l20.8 104z"],"lock-open":[576,512,[],"f3c1","M352 144c0-44.2 35.8-80 80-80s80 35.8 80 80v48c0 17.7 14.3 32 32 32s32-14.3 32-32V144C576 64.5 511.5 0 432 0S288 64.5 288 144v48H64c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V256c0-35.3-28.7-64-64-64H352V144z"],elevator:[512,512,[],"e16d","M132.7 4.7l-64 64c-4.6 4.6-5.9 11.5-3.5 17.4s8.3 9.9 14.8 9.9H208c6.5 0 12.3-3.9 14.8-9.9s1.1-12.9-3.5-17.4l-64-64c-6.2-6.2-16.4-6.2-22.6 0zM64 128c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V192c0-35.3-28.7-64-64-64H64zm96 96a48 48 0 1 1 0 96 48 48 0 1 1 0-96zM80 400c0-26.5 21.5-48 48-48h64c26.5 0 48 21.5 48 48v16c0 17.7-14.3 32-32 32H112c-17.7 0-32-14.3-32-32V400zm192 0c0-26.5 21.5-48 48-48h64c26.5 0 48 21.5 48 48v16c0 17.7-14.3 32-32 32H304c-17.7 0-32-14.3-32-32V400zm32-128a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM356.7 91.3c6.2 6.2 16.4 6.2 22.6 0l64-64c4.6-4.6 5.9-11.5 3.5-17.4S438.5 0 432 0H304c-6.5 0-12.3 3.9-14.8 9.9s-1.1 12.9 3.5 17.4l64 64z"],"money-bill-transfer":[640,512,[],"e528","M535 41c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l64 64c4.5 4.5 7 10.6 7 17s-2.5 12.5-7 17l-64 64c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l23-23L384 112c-13.3 0-24-10.7-24-24s10.7-24 24-24l174.1 0L535 41zM105 377l-23 23L256 400c13.3 0 24 10.7 24 24s-10.7 24-24 24L81.9 448l23 23c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L7 441c-4.5-4.5-7-10.6-7-17s2.5-12.5 7-17l64-64c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9zM96 64H337.9c-3.7 7.2-5.9 15.3-5.9 24c0 28.7 23.3 52 52 52l117.4 0c-4 17 .6 35.5 13.8 48.8c20.3 20.3 53.2 20.3 73.5 0L608 169.5V384c0 35.3-28.7 64-64 64H302.1c3.7-7.2 5.9-15.3 5.9-24c0-28.7-23.3-52-52-52l-117.4 0c4-17-.6-35.5-13.8-48.8c-20.3-20.3-53.2-20.3-73.5 0L32 342.5V128c0-35.3 28.7-64 64-64zm64 64H96v64c35.3 0 64-28.7 64-64zM544 320c-35.3 0-64 28.7-64 64h64V320zM320 352a96 96 0 1 0 0-192 96 96 0 1 0 0 192z"],"money-bill-trend-up":[512,512,[],"e529","M470.7 9.4c3 3.1 5.3 6.6 6.9 10.3s2.4 7.8 2.4 12.2l0 .1v0 96c0 17.7-14.3 32-32 32s-32-14.3-32-32V109.3L310.6 214.6c-11.8 11.8-30.8 12.6-43.5 1.7L176 138.1 84.8 216.3c-13.4 11.5-33.6 9.9-45.1-3.5s-9.9-33.6 3.5-45.1l112-96c12-10.3 29.7-10.3 41.7 0l89.5 76.7L370.7 64H352c-17.7 0-32-14.3-32-32s14.3-32 32-32h96 0c8.8 0 16.8 3.6 22.6 9.3l.1 .1zM0 304c0-26.5 21.5-48 48-48H464c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V304zM48 416v48H96c0-26.5-21.5-48-48-48zM96 304H48v48c26.5 0 48-21.5 48-48zM464 416c-26.5 0-48 21.5-48 48h48V416zM416 304c0 26.5 21.5 48 48 48V304H416zm-96 80a64 64 0 1 0 -128 0 64 64 0 1 0 128 0z"],"house-flood-water-circle-arrow-right":[640,512,[],"e50f","M288 144A144 144 0 1 0 0 144a144 144 0 1 0 288 0zM140.7 76.7c6.2-6.2 16.4-6.2 22.6 0l56 56c6.2 6.2 6.2 16.4 0 22.6l-56 56c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L169.4 160H80c-8.8 0-16-7.2-16-16s7.2-16 16-16h89.4L140.7 99.3c-6.2-6.2-6.2-16.4 0-22.6zM320 144c0 57.3-27.4 108.2-69.8 140.3c11.8-3.6 23-9.4 33-16.2c22.1-15.5 51.6-15.5 73.7 0c18.4 12.7 39.6 20.3 59.2 20.3c19 0 41.2-7.9 59.2-20.3c23.8-16.7 55.8-15.4 78.1 3.4c2.1 1.7 4.2 3.3 6.5 4.9l-.3-84.4H576c13.9 0 26.1-8.9 30.4-22.1s-.4-27.6-11.6-35.8l-176-128C407.6-2 392.4-2 381.2 6.1L301 64.4c12.1 23.9 19 50.9 19 79.6zm18.5 165.9c-11.1-7.9-25.9-7.9-37 0C279 325.4 251.5 336 224 336c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C158.5 389.7 191 400 224 400c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.6 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.5-27.3-10.1-39.2-1.7l0 0C471.4 325.2 442.9 336 416 336c-27.5 0-55-10.6-77.5-26.1zm0 112c-11.1-7.9-25.9-7.9-37 0C279 437.4 251.5 448 224 448c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C158.5 501.7 191 512 224 512c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.6 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.4-27.3-10.1-39.2-1.7l0 0C471.4 437.2 442.9 448 416 448c-27.5 0-55-10.6-77.5-26.1z"],"square-poll-horizontal":[448,512,["poll-h"],"f682","M448 96c0-35.3-28.7-64-64-64L64 32C28.7 32 0 60.7 0 96L0 416c0 35.3 28.7 64 64 64l320 0c35.3 0 64-28.7 64-64l0-320zM256 160c0 17.7-14.3 32-32 32l-96 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l96 0c17.7 0 32 14.3 32 32zm64 64c17.7 0 32 14.3 32 32s-14.3 32-32 32l-192 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l192 0zM192 352c0 17.7-14.3 32-32 32l-32 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l32 0c17.7 0 32 14.3 32 32z"],circle:[512,512,[128308,128309,128992,128993,128994,128995,128996,9679,9898,9899,11044,61708,61915],"f111","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512z"],"backward-fast":[512,512,[9198,"fast-backward"],"f049","M493.6 445c-11.2 5.3-24.5 3.6-34.1-4.4L288 297.7V416c0 12.4-7.2 23.7-18.4 29s-24.5 3.6-34.1-4.4L64 297.7V416c0 17.7-14.3 32-32 32s-32-14.3-32-32V96C0 78.3 14.3 64 32 64s32 14.3 32 32V214.3L235.5 71.4c9.5-7.9 22.8-9.7 34.1-4.4S288 83.6 288 96V214.3L459.5 71.4c9.5-7.9 22.8-9.7 34.1-4.4S512 83.6 512 96V416c0 12.4-7.2 23.7-18.4 29z"],recycle:[512,512,[9842,9850,9851],"f1b8","M174.7 45.1C192.2 17 223 0 256 0s63.8 17 81.3 45.1l38.6 61.7 27-15.6c8.4-4.9 18.9-4.2 26.6 1.7s11.1 15.9 8.6 25.3l-23.4 87.4c-3.4 12.8-16.6 20.4-29.4 17l-87.4-23.4c-9.4-2.5-16.3-10.4-17.6-20s3.4-19.1 11.8-23.9l28.4-16.4L283 79c-5.8-9.3-16-15-27-15s-21.2 5.7-27 15l-17.5 28c-9.2 14.8-28.6 19.5-43.6 10.5c-15.3-9.2-20.2-29.2-10.7-44.4l17.5-28zM429.5 251.9c15-9 34.4-4.3 43.6 10.5l24.4 39.1c9.4 15.1 14.4 32.4 14.6 50.2c.3 53.1-42.7 96.4-95.8 96.4L320 448v32c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-64-64c-9.4-9.4-9.4-24.6 0-33.9l64-64c6.9-6.9 17.2-8.9 26.2-5.2s14.8 12.5 14.8 22.2v32l96.2 0c17.6 0 31.9-14.4 31.8-32c0-5.9-1.7-11.7-4.8-16.7l-24.4-39.1c-9.5-15.2-4.7-35.2 10.7-44.4zm-364.6-31L36 204.2c-8.4-4.9-13.1-14.3-11.8-23.9s8.2-17.5 17.6-20l87.4-23.4c12.8-3.4 26 4.2 29.4 17L182 241.2c2.5 9.4-.9 19.3-8.6 25.3s-18.2 6.6-26.6 1.7l-26.5-15.3L68.8 335.3c-3.1 5-4.8 10.8-4.8 16.7c-.1 17.6 14.2 32 31.8 32l32.2 0c17.7 0 32 14.3 32 32s-14.3 32-32 32l-32.2 0C42.7 448-.3 404.8 0 351.6c.1-17.8 5.1-35.1 14.6-50.2l50.3-80.5z"],"user-astronaut":[448,512,[],"f4fb","M370.7 96.1C346.1 39.5 289.7 0 224 0S101.9 39.5 77.3 96.1C60.9 97.5 48 111.2 48 128v64c0 16.8 12.9 30.5 29.3 31.9C101.9 280.5 158.3 320 224 320s122.1-39.5 146.7-96.1c16.4-1.4 29.3-15.1 29.3-31.9V128c0-16.8-12.9-30.5-29.3-31.9zM336 144v16c0 53-43 96-96 96H208c-53 0-96-43-96-96V144c0-26.5 21.5-48 48-48H288c26.5 0 48 21.5 48 48zM189.3 162.7l-6-21.2c-.9-3.3-3.9-5.5-7.3-5.5s-6.4 2.2-7.3 5.5l-6 21.2-21.2 6c-3.3 .9-5.5 3.9-5.5 7.3s2.2 6.4 5.5 7.3l21.2 6 6 21.2c.9 3.3 3.9 5.5 7.3 5.5s6.4-2.2 7.3-5.5l6-21.2 21.2-6c3.3-.9 5.5-3.9 5.5-7.3s-2.2-6.4-5.5-7.3l-21.2-6zM112.7 316.5C46.7 342.6 0 407 0 482.3C0 498.7 13.3 512 29.7 512H128V448c0-17.7 14.3-32 32-32H288c17.7 0 32 14.3 32 32v64l98.3 0c16.4 0 29.7-13.3 29.7-29.7c0-75.3-46.7-139.7-112.7-165.8C303.9 338.8 265.5 352 224 352s-79.9-13.2-111.3-35.5zM176 448c-8.8 0-16 7.2-16 16v48h32V464c0-8.8-7.2-16-16-16zm96 32a16 16 0 1 0 0-32 16 16 0 1 0 0 32z"],"plane-slash":[640,512,[],"e069","M440.6 320h73.8c34.2 0 93.7-28 93.7-64c0-35-59.5-64-93.7-64l-116.6 0L297.2 16.1C291.5 6.2 280.9 0 269.4 0H213.2c-10.6 0-18.3 10.2-15.4 20.4l40.3 140.9L38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L440.6 320zm-78.1 61.6L41.5 128.7C36 130.6 32 135.9 32 142c0 1.3 .2 2.6 .5 3.9L64 256 32.5 366.1c-.4 1.3-.5 2.6-.5 3.9c0 7.8 6.3 14 14 14H88c5 0 9.8-2.4 12.8-6.4L144 320H246.9l-49 171.6c-2.9 10.2 4.8 20.4 15.4 20.4l56.2 0c11.5 0 22.1-6.2 27.8-16.1l65.3-114.3z"],trademark:[640,512,[8482],"f25c","M345.6 108.8c-8.3-11-22.7-15.5-35.7-11.2S288 114.2 288 128V384c0 17.7 14.3 32 32 32s32-14.3 32-32V224l86.4 115.2c6 8.1 15.5 12.8 25.6 12.8s19.6-4.7 25.6-12.8L576 224V384c0 17.7 14.3 32 32 32s32-14.3 32-32V128c0-13.8-8.8-26-21.9-30.4s-27.5 .1-35.7 11.2L464 266.7 345.6 108.8zM0 128c0 17.7 14.3 32 32 32H96V384c0 17.7 14.3 32 32 32s32-14.3 32-32V160h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H32C14.3 96 0 110.3 0 128z"],basketball:[512,512,[127936,"basketball-ball"],"f434","M86.6 64l85.2 85.2C194.5 121.7 208 86.4 208 48c0-14.7-2-28.9-5.7-42.4C158.6 15 119 35.5 86.6 64zM64 86.6C35.5 119 15 158.6 5.6 202.3C19.1 206 33.3 208 48 208c38.4 0 73.7-13.5 101.3-36.1L64 86.6zM256 0c-7.3 0-14.6 .3-21.8 .9C238 16 240 31.8 240 48c0 47.3-17.1 90.5-45.4 124L256 233.4 425.4 64C380.2 24.2 320.9 0 256 0zM48 240c-16.2 0-32-2-47.1-5.8C.3 241.4 0 248.7 0 256c0 64.9 24.2 124.2 64 169.4L233.4 256 172 194.6C138.5 222.9 95.3 240 48 240zm463.1 37.8c.6-7.2 .9-14.5 .9-21.8c0-64.9-24.2-124.2-64-169.4L278.6 256 340 317.4c33.4-28.3 76.7-45.4 124-45.4c16.2 0 32 2 47.1 5.8zm-4.7 31.9C492.9 306 478.7 304 464 304c-38.4 0-73.7 13.5-101.3 36.1L448 425.4c28.5-32.3 49.1-71.9 58.4-115.7zM340.1 362.7C317.5 390.3 304 425.6 304 464c0 14.7 2 28.9 5.7 42.4C353.4 497 393 476.5 425.4 448l-85.2-85.2zM317.4 340L256 278.6 86.6 448c45.1 39.8 104.4 64 169.4 64c7.3 0 14.6-.3 21.8-.9C274 496 272 480.2 272 464c0-47.3 17.1-90.5 45.4-124z"],"satellite-dish":[512,512,[128225],"f7c0","M192 32c0-17.7 14.3-32 32-32C383.1 0 512 128.9 512 288c0 17.7-14.3 32-32 32s-32-14.3-32-32C448 164.3 347.7 64 224 64c-17.7 0-32-14.3-32-32zM60.6 220.6L164.7 324.7l28.4-28.4c-.7-2.6-1.1-5.4-1.1-8.3c0-17.7 14.3-32 32-32s32 14.3 32 32s-14.3 32-32 32c-2.9 0-5.6-.4-8.3-1.1l-28.4 28.4L291.4 451.4c14.5 14.5 11.8 38.8-7.3 46.3C260.5 506.9 234.9 512 208 512C93.1 512 0 418.9 0 304c0-26.9 5.1-52.5 14.4-76.1c7.5-19 31.8-21.8 46.3-7.3zM224 96c106 0 192 86 192 192c0 17.7-14.3 32-32 32s-32-14.3-32-32c0-70.7-57.3-128-128-128c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],"circle-up":[512,512,[61467,"arrow-alt-circle-up"],"f35b","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM135.1 217.4l107.1-99.9c3.8-3.5 8.7-5.5 13.8-5.5s10.1 2 13.8 5.5l107.1 99.9c4.5 4.2 7.1 10.1 7.1 16.3c0 12.3-10 22.3-22.3 22.3H304v96c0 17.7-14.3 32-32 32H240c-17.7 0-32-14.3-32-32V256H150.3C138 256 128 246 128 233.7c0-6.2 2.6-12.1 7.1-16.3z"],"mobile-screen-button":[384,512,["mobile-alt"],"f3cd","M16 64C16 28.7 44.7 0 80 0H304c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H80c-35.3 0-64-28.7-64-64V64zM224 448a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM304 64H80V384H304V64z"],"volume-high":[640,512,[128266,"volume-up"],"f028","M533.6 32.5C598.5 85.2 640 165.8 640 256s-41.5 170.7-106.4 223.5c-10.3 8.4-25.4 6.8-33.8-3.5s-6.8-25.4 3.5-33.8C557.5 398.2 592 331.2 592 256s-34.5-142.2-88.7-186.3c-10.3-8.4-11.8-23.5-3.5-33.8s23.5-11.8 33.8-3.5zM473.1 107c43.2 35.2 70.9 88.9 70.9 149s-27.7 113.8-70.9 149c-10.3 8.4-25.4 6.8-33.8-3.5s-6.8-25.4 3.5-33.8C475.3 341.3 496 301.1 496 256s-20.7-85.3-53.2-111.8c-10.3-8.4-11.8-23.5-3.5-33.8s23.5-11.8 33.8-3.5zm-60.5 74.5C434.1 199.1 448 225.9 448 256s-13.9 56.9-35.4 74.5c-10.3 8.4-25.4 6.8-33.8-3.5s-6.8-25.4 3.5-33.8C393.1 284.4 400 271 400 256s-6.9-28.4-17.7-37.3c-10.3-8.4-11.8-23.5-3.5-33.8s23.5-11.8 33.8-3.5zM301.1 34.8C312.6 40 320 51.4 320 64V448c0 12.6-7.4 24-18.9 29.2s-25 3.1-34.4-5.3L131.8 352H64c-35.3 0-64-28.7-64-64V224c0-35.3 28.7-64 64-64h67.8L266.7 40.1c9.4-8.4 22.9-10.4 34.4-5.3z"],"users-rays":[640,512,[],"e593","M41 7C31.6-2.3 16.4-2.3 7 7S-2.3 31.6 7 41l72 72c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L41 7zM599 7L527 79c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l72-72c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0zM7 505c9.4 9.4 24.6 9.4 33.9 0l72-72c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L7 471c-9.4 9.4-9.4 24.6 0 33.9zm592 0c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-72-72c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l72 72zM320 256a64 64 0 1 0 0-128 64 64 0 1 0 0 128zM212.1 336c-2.7 7.5-4.1 15.6-4.1 24c0 13.3 10.7 24 24 24H408c13.3 0 24-10.7 24-24c0-8.4-1.4-16.5-4.1-24c-.5-1.4-1-2.7-1.6-4c-9.4-22.3-29.8-38.9-54.3-43c-3.9-.7-7.9-1-12-1H280c-4.1 0-8.1 .3-12 1c-.8 .1-1.7 .3-2.5 .5c-24.9 5.1-45.1 23-53.4 46.5zM175.8 224a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-26.5 32C119.9 256 96 279.9 96 309.3c0 14.7 11.9 26.7 26.7 26.7h56.1c8-34.1 32.8-61.7 65.2-73.6c-7.5-4.1-16.2-6.4-25.3-6.4H149.3zm368 80c14.7 0 26.7-11.9 26.7-26.7c0-29.5-23.9-53.3-53.3-53.3H421.3c-9.2 0-17.8 2.3-25.3 6.4c32.4 11.9 57.2 39.5 65.2 73.6h56.1zM464 224a48 48 0 1 0 0-96 48 48 0 1 0 0 96z"],wallet:[512,512,[],"f555","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V192c0-35.3-28.7-64-64-64H80c-8.8 0-16-7.2-16-16s7.2-16 16-16H448c17.7 0 32-14.3 32-32s-14.3-32-32-32H64zM416 272a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"clipboard-check":[384,512,[],"f46c","M192 0c-41.8 0-77.4 26.7-90.5 64H64C28.7 64 0 92.7 0 128V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H282.5C269.4 26.7 233.8 0 192 0zm0 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM305 273L177 401c-9.4 9.4-24.6 9.4-33.9 0L79 337c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47L271 239c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z"],"file-audio":[384,512,[],"f1c7","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zm2 226.3c37.1 22.4 62 63.1 62 109.7s-24.9 87.3-62 109.7c-7.6 4.6-17.4 2.1-22-5.4s-2.1-17.4 5.4-22C269.4 401.5 288 370.9 288 336s-18.6-65.5-46.5-82.3c-7.6-4.6-10-14.4-5.4-22s14.4-10 22-5.4zm-91.9 30.9c6 2.5 9.9 8.3 9.9 14.8V400c0 6.5-3.9 12.3-9.9 14.8s-12.9 1.1-17.4-3.5L113.4 376H80c-8.8 0-16-7.2-16-16V312c0-8.8 7.2-16 16-16h33.4l35.3-35.3c4.6-4.6 11.5-5.9 17.4-3.5zm51 34.9c6.6-5.9 16.7-5.3 22.6 1.3C249.8 304.6 256 319.6 256 336s-6.2 31.4-16.3 42.7c-5.9 6.6-16 7.1-22.6 1.3s-7.1-16-1.3-22.6c5.1-5.7 8.1-13.1 8.1-21.3s-3.1-15.7-8.1-21.3c-5.9-6.6-5.3-16.7 1.3-22.6z"],burger:[512,512,["hamburger"],"f805","M61.1 224C45 224 32 211 32 194.9c0-1.9 .2-3.7 .6-5.6C37.9 168.3 78.8 32 256 32s218.1 136.3 223.4 157.3c.5 1.9 .6 3.7 .6 5.6c0 16.1-13 29.1-29.1 29.1H61.1zM144 128a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm240 16a16 16 0 1 0 0-32 16 16 0 1 0 0 32zM272 96a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zM16 304c0-26.5 21.5-48 48-48H448c26.5 0 48 21.5 48 48s-21.5 48-48 48H64c-26.5 0-48-21.5-48-48zm16 96c0-8.8 7.2-16 16-16H464c8.8 0 16 7.2 16 16v16c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V400z"],wrench:[512,512,[128295],"f0ad","M352 320c88.4 0 160-71.6 160-160c0-15.3-2.2-30.1-6.2-44.2c-3.1-10.8-16.4-13.2-24.3-5.3l-76.8 76.8c-3 3-7.1 4.7-11.3 4.7H336c-8.8 0-16-7.2-16-16V118.6c0-4.2 1.7-8.3 4.7-11.3l76.8-76.8c7.9-7.9 5.4-21.2-5.3-24.3C382.1 2.2 367.3 0 352 0C263.6 0 192 71.6 192 160c0 19.1 3.4 37.5 9.5 54.5L19.9 396.1C7.2 408.8 0 426.1 0 444.1C0 481.6 30.4 512 67.9 512c18 0 35.3-7.2 48-19.9L297.5 310.5c17 6.2 35.4 9.5 54.5 9.5zM80 408a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],bugs:[576,512,[],"e4d0","M164.5 107.4l33.4-73.5c5.5-12.1 .1-26.3-11.9-31.8s-26.3-.1-31.8 11.9L128 71.7 101.9 14.1C96.4 2 82.1-3.3 70.1 2.1S52.7 21.9 58.1 33.9l33.4 73.5c-10.2 7.1-18.2 17-22.9 28.6h-17l-4.1-20.7c-2.6-13-15.2-21.4-28.2-18.8S-2.1 111.7 .5 124.7l8 40C10.7 175.9 20.6 184 32 184H64v23.3l-37.8 9.5c-9.5 2.4-16.6 10.2-17.9 19.9l-8 56c-1.9 13.1 7.2 25.3 20.4 27.2s25.3-7.2 27.2-20.4l5.7-40 18.4-4.6C82.7 274.6 103.8 288 128 288s45.3-13.4 56.1-33.2l18.4 4.6 5.7 40c1.9 13.1 14 22.2 27.2 20.4s22.2-14 20.4-27.2l-8-56c-1.4-9.7-8.5-17.5-17.9-19.9L192 207.3V184h32c11.4 0 21.3-8.1 23.5-19.3l8-40c2.6-13-5.8-25.6-18.8-28.2s-25.6 5.8-28.2 18.8L204.3 136h-17c-4.7-11.6-12.7-21.5-22.9-28.6zM496 286.5l65.6-47c10.8-7.7 13.3-22.7 5.6-33.5s-22.7-13.3-33.5-5.6l-51.4 36.8 6.1-62.9c1.3-13.2-8.4-24.9-21.6-26.2s-24.9 8.4-26.2 21.6L432.8 250c-12.3 1-24.2 5.6-34.1 13.3L384 254.8l6.8-20c4.2-12.6-2.5-26.2-15-30.4s-26.2 2.5-30.4 15l-13.1 38.6c-3.7 10.8 .8 22.8 10.7 28.5l27.7 16L359 322.7 321.5 312c-9.4-2.7-19.5 .6-25.5 8.3l-34.9 44.5c-8.2 10.4-6.4 25.5 4.1 33.7s25.5 6.4 33.7-4.1l25-31.8 18.2 5.2c-.5 22.6 11 44.7 32 56.8s45.9 11 65.2-.7l13.6 13.2-15.1 37.5c-4.9 12.3 1 26.3 13.3 31.2s26.3-1 31.2-13.3L503.5 440c3.6-9.1 1.4-19.4-5.6-26.2l-28-27.1 11.6-20.1 27.7 16c9.9 5.7 22.5 3.7 30-4.9L566.2 347c8.7-10 7.8-25.1-2.2-33.9s-25.1-7.8-33.9 2.2l-13.9 15.9-14.7-8.5c1.7-12.4-.2-25-5.5-36.2z"],"rupee-sign":[448,512,[8360,"rupee"],"f156","M0 64C0 46.3 14.3 32 32 32h80c79.5 0 144 64.5 144 144c0 58.8-35.2 109.3-85.7 131.7l51.4 128.4c6.6 16.4-1.4 35-17.8 41.6s-35-1.4-41.6-17.8L106.3 320H64V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V288 64zM64 256h48c44.2 0 80-35.8 80-80s-35.8-80-80-80H64V256zm256.5 16.4c-.9 6 0 8.7 .4 9.8c.4 1.1 1.4 2.6 4.2 4.9c7.2 5.7 18.7 10 37.9 16.8l1.3 .5c16 5.6 38.7 13.6 55.7 28.1c9.5 8.1 17.9 18.6 23.1 32.3c5.1 13.7 6.1 28.5 3.8 44c-4.2 28.1-20.5 49.3-43.8 60.9c-22.1 11-48.1 12.5-73.2 8l-.2 0 0 0c-9.3-1.8-20.5-5.7-29.3-9c-6-2.3-12.6-4.9-17.7-6.9l0 0c-2.5-1-4.6-1.8-6.3-2.5c-16.5-6.4-24.6-25-18.2-41.4s24.9-24.6 41.4-18.2c2.6 1 5.2 2 7.9 3.1l0 0c4.8 1.9 9.8 3.9 15.4 6c8.8 3.3 15.3 5.4 18.7 6c15.7 2.8 26.7 .8 32.9-2.3c5-2.5 8-6 9.1-13c1-6.9 .2-10.5-.5-12.3c-.6-1.7-1.8-3.6-4.5-5.9c-6.9-5.8-18.2-10.4-36.9-17l-3-1.1c-15.5-5.4-37-13-53.3-25.9c-9.5-7.5-18.3-17.6-23.7-31c-5.5-13.4-6.6-28-4.4-43.2c8.4-57.1 67-78 116.9-68.9c6.9 1.3 27.3 5.8 35.4 8.4c16.9 5.2 26.3 23.2 21.1 40.1s-23.2 26.3-40.1 21.1c-4.7-1.4-22.3-5.5-27.9-6.5c-14.6-2.7-25.8-.4-32.6 3.2c-6.3 3.3-8.9 7.6-9.5 12z"],"file-image":[384,512,[128443],"f1c5","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM64 256a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm152 32c5.3 0 10.2 2.6 13.2 6.9l88 128c3.4 4.9 3.7 11.3 1 16.5s-8.2 8.6-14.2 8.6H216 176 128 80c-5.8 0-11.1-3.1-13.9-8.1s-2.8-11.2 .2-16.1l48-80c2.9-4.8 8.1-7.8 13.7-7.8s10.8 2.9 13.7 7.8l12.8 21.4 48.3-70.2c3-4.3 7.9-6.9 13.2-6.9z"],"circle-question":[512,512,[62108,"question-circle"],"f059","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM169.8 165.3c7.9-22.3 29.1-37.3 52.8-37.3h58.3c34.9 0 63.1 28.3 63.1 63.1c0 22.6-12.1 43.5-31.7 54.8L280 264.4c-.2 13-10.9 23.6-24 23.6c-13.3 0-24-10.7-24-24V250.5c0-8.6 4.6-16.5 12.1-20.8l44.3-25.4c4.7-2.7 7.6-7.7 7.6-13.1c0-8.4-6.8-15.1-15.1-15.1H222.6c-3.4 0-6.4 2.1-7.5 5.3l-.4 1.2c-4.4 12.5-18.2 19-30.6 14.6s-19-18.2-14.6-30.6l.4-1.2zM224 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],"plane-departure":[640,512,[128747],"f5b0","M381 114.9L186.1 41.8c-16.7-6.2-35.2-5.3-51.1 2.7L89.1 67.4C78 73 77.2 88.5 87.6 95.2l146.9 94.5L136 240 77.8 214.1c-8.7-3.9-18.8-3.7-27.3 .6L18.3 230.8c-9.3 4.7-11.8 16.8-5 24.7l73.1 85.3c6.1 7.1 15 11.2 24.3 11.2H248.4c5 0 9.9-1.2 14.3-3.4L535.6 212.2c46.5-23.3 82.5-63.3 100.8-112C645.9 75 627.2 48 600.2 48H542.8c-20.2 0-40.2 4.8-58.2 14L381 114.9zM0 480c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H32c-17.7 0-32 14.3-32 32z"],"handshake-slash":[640,512,[],"e060","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-135-105.8c-1.1-11.3-6.3-22.3-15.3-30.7l-134.2-123-23.4 18.2-26-20.3 77.2-60.1c7-5.4 17-4.2 22.5 2.8s4.2 17-2.8 22.5l-20.9 16.2L512 316.8V128h-.7l-3.9-2.5L434.8 79c-15.3-9.8-33.2-15-51.4-15c-21.8 0-43 7.5-60 21.2l-89.7 72.6-25.8-20.3 81.8-66.2c-11.6-4.9-24.1-7.4-36.8-7.4C234 64 215.7 69.6 200 80l-35.5 23.7L38.8 5.1zM413.6 421.9L128 196.8V352h28.2l91.4 83.4c19.6 17.9 49.9 16.5 67.8-3.1c5.5-6.1 9.2-13.2 11.1-20.6l17 15.6c19.5 17.9 49.9 16.6 67.8-2.9c.8-.8 1.5-1.7 2.2-2.6zM96 171.6L40.6 128H16c-8.8 0-16 7.2-16 16V352c0 17.7 14.3 32 32 32H64c17.7 0 32-14.3 32-32V171.6zM48 320a16 16 0 1 1 0 32 16 16 0 1 1 0-32zM544 128V352c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V144c0-8.8-7.2-16-16-16H544zm32 208a16 16 0 1 1 32 0 16 16 0 1 1 -32 0z"],"book-bookmark":[448,512,[],"e0bb","M0 96C0 43 43 0 96 0h96V190.7c0 13.4 15.5 20.9 26 12.5L272 160l54 43.2c10.5 8.4 26 .9 26-12.5V0h32 32c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384 96c-53 0-96-43-96-96V96zM64 416c0 17.7 14.3 32 32 32H352V384H96c-17.7 0-32 14.3-32 32z"],"code-branch":[448,512,[],"f126","M80 104a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm80-24c0 32.8-19.7 61-48 73.3v87.8c18.8-10.9 40.7-17.1 64-17.1h96c35.3 0 64-28.7 64-64v-6.7C307.7 141 288 112.8 288 80c0-44.2 35.8-80 80-80s80 35.8 80 80c0 32.8-19.7 61-48 73.3V160c0 70.7-57.3 128-128 128H176c-35.3 0-64 28.7-64 64v6.7c28.3 12.3 48 40.5 48 73.3c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-32.8 19.7-61 48-73.3V352 153.3C19.7 141 0 112.8 0 80C0 35.8 35.8 0 80 0s80 35.8 80 80zm232 0a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zM80 456a24 24 0 1 0 0-48 24 24 0 1 0 0 48z"],"hat-cowboy":[640,512,[],"f8c0","M320 64c14.4 0 22.3-7 30.8-14.4C360.4 41.1 370.7 32 392 32c49.3 0 84.4 152.2 97.9 221.9C447.8 272.1 390.9 288 320 288s-127.8-15.9-169.9-34.1C163.6 184.2 198.7 32 248 32c21.3 0 31.6 9.1 41.2 17.6C297.7 57 305.6 64 320 64zM111.1 270.7c47.2 24.5 117.5 49.3 209 49.3s161.8-24.8 208.9-49.3c24.8-12.9 49.8-28.3 70.1-47.7c7.9-7.9 20.2-9.2 29.6-3.3c9.5 5.9 13.5 17.9 9.9 28.5c-13.5 37.7-38.4 72.3-66.1 100.6C523.7 398.9 443.6 448 320 448s-203.6-49.1-252.5-99.2C39.8 320.4 14.9 285.8 1.4 248.1c-3.6-10.6 .4-22.6 9.9-28.5c9.5-5.9 21.7-4.5 29.6 3.3c20.4 19.4 45.3 34.8 70.1 47.7z"],bridge:[576,512,[],"e4c8","M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96H72v64H0V288c53 0 96 43 96 96v64c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V384c0-53 43-96 96-96s96 43 96 96v64c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V384c0-53 43-96 96-96V160H504V96h40c17.7 0 32-14.3 32-32s-14.3-32-32-32H32zM456 96v64H376V96h80zM328 96v64H248V96h80zM200 96v64H120V96h80z"],"phone-flip":[512,512,[128381,"phone-alt"],"f879","M347.1 24.6c7.7-18.6 28-28.5 47.4-23.2l88 24C499.9 30.2 512 46 512 64c0 247.4-200.6 448-448 448c-18 0-33.8-12.1-38.6-29.5l-24-88c-5.3-19.4 4.6-39.7 23.2-47.4l96-40c16.3-6.8 35.2-2.1 46.3 11.6L207.3 368c70.4-33.3 127.4-90.3 160.7-160.7L318.7 167c-13.7-11.2-18.4-30-11.6-46.3l40-96z"],"truck-front":[512,512,[],"e2b7","M0 80C0 35.8 35.8 0 80 0H432c44.2 0 80 35.8 80 80V368c0 26.2-12.6 49.4-32 64v48c0 17.7-14.3 32-32 32H416c-17.7 0-32-14.3-32-32V448H128v32c0 17.7-14.3 32-32 32H64c-17.7 0-32-14.3-32-32V432C12.6 417.4 0 394.2 0 368V80zm129.9 72.2L112 224H400l-17.9-71.8C378.5 138 365.7 128 351 128H161c-14.7 0-27.5 10-31 24.2zM128 320a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm288 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],cat:[576,512,[128008],"f6be","M320 192h17.1c22.1 38.3 63.5 64 110.9 64c11 0 21.8-1.4 32-4v4 32V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V339.2L280 448h56c17.7 0 32 14.3 32 32s-14.3 32-32 32H192c-53 0-96-43-96-96V192.5c0-16.1-12-29.8-28-31.8l-7.9-1c-17.5-2.2-30-18.2-27.8-35.7s18.2-30 35.7-27.8l7.9 1c48 6 84.1 46.8 84.1 95.3v85.3c34.4-51.7 93.2-85.8 160-85.8zm160 26.5v0c-10 3.5-20.8 5.5-32 5.5c-28.4 0-54-12.4-71.6-32h0c-3.7-4.1-7-8.5-9.9-13.2C357.3 164 352 146.6 352 128v0V32 12 10.7C352 4.8 356.7 .1 362.6 0h.2c3.3 0 6.4 1.6 8.4 4.2l0 .1L384 21.3l27.2 36.3L416 64h64l4.8-6.4L512 21.3 524.8 4.3l0-.1c2-2.6 5.1-4.2 8.4-4.2h.2C539.3 .1 544 4.8 544 10.7V12 32v96c0 17.3-4.6 33.6-12.6 47.6c-11.3 19.8-29.6 35.2-51.4 42.9zM432 128a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm48 16a16 16 0 1 0 0-32 16 16 0 1 0 0 32z"],"anchor-circle-exclamation":[640,512,[],"e4ab","M320 96a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm21.1 80C367 158.8 384 129.4 384 96c0-53-43-96-96-96s-96 43-96 96c0 33.4 17 62.8 42.9 80H224c-17.7 0-32 14.3-32 32s14.3 32 32 32h32V448H208c-53 0-96-43-96-96v-6.1l7 7c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L97 263c-9.4-9.4-24.6-9.4-33.9 0L7 319c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l7-7V352c0 88.4 71.6 160 160 160h80 80c8.2 0 16.3-.6 24.2-1.8c-22.2-16.2-40.4-37.5-53-62.2H320V368 240h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H341.1zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z"],"truck-field":[640,512,[],"e58d","M32 96c0-35.3 28.7-64 64-64H320c23.7 0 44.4 12.9 55.4 32h51.8c25.3 0 48.2 14.9 58.5 38l52.8 118.8c.5 1.1 .9 2.1 1.3 3.2H544c35.3 0 64 28.7 64 64v32c17.7 0 32 14.3 32 32s-14.3 32-32 32H576c0 53-43 96-96 96s-96-43-96-96H256c0 53-43 96-96 96s-96-43-96-96H32c-17.7 0-32-14.3-32-32s14.3-32 32-32V288c-17.7 0-32-14.3-32-32V160c0-17.7 14.3-32 32-32V96zM384 224h85.9l-42.7-96H384v96zM160 432a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm368-48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0z"],route:[512,512,[],"f4d7","M512 96c0 50.2-59.1 125.1-84.6 155c-3.8 4.4-9.4 6.1-14.5 5H320c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c53 0 96 43 96 96s-43 96-96 96H139.6c8.7-9.9 19.3-22.6 30-36.8c6.3-8.4 12.8-17.6 19-27.2H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H320c-53 0-96-43-96-96s43-96 96-96h39.8c-21-31.5-39.8-67.7-39.8-96c0-53 43-96 96-96s96 43 96 96zM117.1 489.1c-3.8 4.3-7.2 8.1-10.1 11.3l-1.8 2-.2-.2c-6 4.6-14.6 4-20-1.8C59.8 473 0 402.5 0 352c0-53 43-96 96-96s96 43 96 96c0 30-21.1 67-43.5 97.9c-10.7 14.7-21.7 28-30.8 38.5l-.6 .7zM128 352a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM416 128a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"clipboard-question":[384,512,[],"e4e3","M192 0c-41.8 0-77.4 26.7-90.5 64H64C28.7 64 0 92.7 0 128V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H282.5C269.4 26.7 233.8 0 192 0zm0 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM105.8 229.3c7.9-22.3 29.1-37.3 52.8-37.3h58.3c34.9 0 63.1 28.3 63.1 63.1c0 22.6-12.1 43.5-31.7 54.8L216 328.4c-.2 13-10.9 23.6-24 23.6c-13.3 0-24-10.7-24-24V314.5c0-8.6 4.6-16.5 12.1-20.8l44.3-25.4c4.7-2.7 7.6-7.7 7.6-13.1c0-8.4-6.8-15.1-15.1-15.1H158.6c-3.4 0-6.4 2.1-7.5 5.3l-.4 1.2c-4.4 12.5-18.2 19-30.6 14.6s-19-18.2-14.6-30.6l.4-1.2zM160 416a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],panorama:[640,512,[],"e209","M45.6 32C20.4 32 0 52.4 0 77.6V434.4C0 459.6 20.4 480 45.6 480c5.1 0 10-.8 14.7-2.4C74.6 472.8 177.6 440 320 440s245.4 32.8 259.6 37.6c4.7 1.6 9.7 2.4 14.7 2.4c25.2 0 45.6-20.4 45.6-45.6V77.6C640 52.4 619.6 32 594.4 32c-5 0-10 .8-14.7 2.4C565.4 39.2 462.4 72 320 72S74.6 39.2 60.4 34.4C55.6 32.8 50.7 32 45.6 32zM96 160a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm272 0c7.9 0 15.4 3.9 19.8 10.5L512.3 353c5.4 8 5.6 18.4 .4 26.5s-14.7 12.3-24.2 10.7C442.7 382.4 385.2 376 320 376c-65.6 0-123.4 6.5-169.3 14.4c-9.8 1.7-19.7-2.9-24.7-11.5s-4.3-19.4 1.9-27.2L197.3 265c4.6-5.7 11.4-9 18.7-9s14.2 3.3 18.7 9l26.4 33.1 87-127.6c4.5-6.6 11.9-10.5 19.8-10.5z"],"comment-medical":[512,512,[],"f7f5","M256 448c141.4 0 256-93.1 256-208S397.4 32 256 32S0 125.1 0 240c0 45.1 17.7 86.8 47.7 120.9c-1.9 24.5-11.4 46.3-21.4 62.9c-5.5 9.2-11.1 16.6-15.2 21.6c-2.1 2.5-3.7 4.4-4.9 5.7c-.6 .6-1 1.1-1.3 1.4l-.3 .3 0 0 0 0 0 0 0 0c-4.6 4.6-5.9 11.4-3.4 17.4c2.5 6 8.3 9.9 14.8 9.9c28.7 0 57.6-8.9 81.6-19.3c22.9-10 42.4-21.9 54.3-30.6c31.8 11.5 67 17.9 104.1 17.9zM224 160c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H288v48c0 8.8-7.2 16-16 16H240c-8.8 0-16-7.2-16-16V272H176c-8.8 0-16-7.2-16-16V224c0-8.8 7.2-16 16-16h48V160z"],"teeth-open":[576,512,[],"f62f","M96 32C43 32 0 75 0 128v64c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128c0-53-43-96-96-96H96zM224 96c26.5 0 48 21.5 48 48v56c0 13.3-10.7 24-24 24H200c-13.3 0-24-10.7-24-24V144c0-26.5 21.5-48 48-48zm80 48c0-26.5 21.5-48 48-48s48 21.5 48 48v56c0 13.3-10.7 24-24 24H328c-13.3 0-24-10.7-24-24V144zM96 128c26.5 0 48 21.5 48 48v24c0 13.3-10.7 24-24 24H72c-13.3 0-24-10.7-24-24V176c0-26.5 21.5-48 48-48zm336 48c0-26.5 21.5-48 48-48s48 21.5 48 48v24c0 13.3-10.7 24-24 24H456c-13.3 0-24-10.7-24-24V176zM96 480H480c53 0 96-43 96-96V352c0-35.3-28.7-64-64-64H64c-35.3 0-64 28.7-64 64v32c0 53 43 96 96 96zm0-64c-26.5 0-48-21.5-48-48V344c0-13.3 10.7-24 24-24h48c13.3 0 24 10.7 24 24v24c0 26.5-21.5 48-48 48zm80-48V344c0-13.3 10.7-24 24-24h48c13.3 0 24 10.7 24 24v24c0 26.5-21.5 48-48 48s-48-21.5-48-48zm176 48c-26.5 0-48-21.5-48-48V344c0-13.3 10.7-24 24-24h48c13.3 0 24 10.7 24 24v24c0 26.5-21.5 48-48 48zm80-48V344c0-13.3 10.7-24 24-24h48c13.3 0 24 10.7 24 24v24c0 26.5-21.5 48-48 48s-48-21.5-48-48z"],"file-circle-minus":[576,512,[],"e4ed","M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v38.6C310.1 219.5 256 287.4 256 368c0 59.1 29.1 111.3 73.7 143.3c-3.2 .5-6.4 .7-9.7 .7H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zM288 368a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm224 0c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16s7.2 16 16 16H496c8.8 0 16-7.2 16-16z"],tags:[512,512,[],"f02c","M345 39.1L472.8 168.4c52.4 53 52.4 138.2 0 191.2L360.8 472.9c-9.3 9.4-24.5 9.5-33.9 .2s-9.5-24.5-.2-33.9L438.6 325.9c33.9-34.3 33.9-89.4 0-123.7L310.9 72.9c-9.3-9.4-9.2-24.6 .2-33.9s24.6-9.2 33.9 .2zM0 229.5V80C0 53.5 21.5 32 48 32H197.5c17 0 33.3 6.7 45.3 18.7l168 168c25 25 25 65.5 0 90.5L277.3 442.7c-25 25-65.5 25-90.5 0l-168-168C6.7 262.7 0 246.5 0 229.5zM144 144a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z"],"wine-glass":[320,512,[127863],"f4e3","M32.1 29.3C33.5 12.8 47.4 0 64 0H256c16.6 0 30.5 12.8 31.9 29.3l14 168.4c6 72-42.5 135.2-109.9 150.6V448h48c17.7 0 32 14.3 32 32s-14.3 32-32 32H160 80c-17.7 0-32-14.3-32-32s14.3-32 32-32h48V348.4C60.6 333 12.1 269.8 18.1 197.8l14-168.4zm56 98.7H231.9l-5.3-64H93.4l-5.3 64z"],"forward-fast":[512,512,[9197,"fast-forward"],"f050","M18.4 445c11.2 5.3 24.5 3.6 34.1-4.4L224 297.7V416c0 12.4 7.2 23.7 18.4 29s24.5 3.6 34.1-4.4L448 297.7V416c0 17.7 14.3 32 32 32s32-14.3 32-32V96c0-17.7-14.3-32-32-32s-32 14.3-32 32V214.3L276.5 71.4c-9.5-7.9-22.8-9.7-34.1-4.4S224 83.6 224 96V214.3L52.5 71.4c-9.5-7.9-22.8-9.7-34.1-4.4S0 83.6 0 96V416c0 12.4 7.2 23.7 18.4 29z"],"face-meh-blank":[512,512,[128566,"meh-blank"],"f5a4","M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm208.4-48a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm128 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"square-parking":[448,512,[127359,"parking"],"f540","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM192 256h48c17.7 0 32-14.3 32-32s-14.3-32-32-32H192v64zm48 64H192v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V288 168c0-22.1 17.9-40 40-40h72c53 0 96 43 96 96s-43 96-96 96z"],"house-signal":[576,512,[],"e012","M357.7 8.5c-12.3-11.3-31.2-11.3-43.4 0l-208 192c-9.4 8.6-12.7 22-8.5 34c87.1 25.3 155.6 94.2 180.3 181.6H464c26.5 0 48-21.5 48-48V256h32c13.2 0 25-8.1 29.8-20.3s1.6-26.2-8.1-35.2l-208-192zM288 208c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H304c-8.8 0-16-7.2-16-16V208zM24 256c-13.3 0-24 10.7-24 24s10.7 24 24 24c101.6 0 184 82.4 184 184c0 13.3 10.7 24 24 24s24-10.7 24-24c0-128.1-103.9-232-232-232zm8 256a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM0 376c0 13.3 10.7 24 24 24c48.6 0 88 39.4 88 88c0 13.3 10.7 24 24 24s24-10.7 24-24c0-75.1-60.9-136-136-136c-13.3 0-24 10.7-24 24z"],"bars-progress":[512,512,["tasks-alt"],"f828","M448 160H320V128H448v32zM48 64C21.5 64 0 85.5 0 112v64c0 26.5 21.5 48 48 48H464c26.5 0 48-21.5 48-48V112c0-26.5-21.5-48-48-48H48zM448 352v32H192V352H448zM48 288c-26.5 0-48 21.5-48 48v64c0 26.5 21.5 48 48 48H464c26.5 0 48-21.5 48-48V336c0-26.5-21.5-48-48-48H48z"],"faucet-drip":[512,512,[128688],"e006","M224 0c17.7 0 32 14.3 32 32V44l96-12c17.7 0 32 14.3 32 32s-14.3 32-32 32L256 84l-31-3.9-1-.1-1 .1L192 84 96 96C78.3 96 64 81.7 64 64s14.3-32 32-32l96 12V32c0-17.7 14.3-32 32-32zM0 224c0-17.7 14.3-32 32-32h96l22.6-22.6c6-6 14.1-9.4 22.6-9.4H192V116.2l32-4 32 4V160h18.7c8.5 0 16.6 3.4 22.6 9.4L320 192h32c88.4 0 160 71.6 160 160c0 17.7-14.3 32-32 32H416c-17.7 0-32-14.3-32-32s-14.3-32-32-32H315.9c-20.2 29-53.9 48-91.9 48s-71.7-19-91.9-48H32c-17.7 0-32-14.3-32-32V224zM436.8 423.4c1.9-4.5 6.3-7.4 11.2-7.4s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V480c0 17.7-14.3 32-32 32s-32-14.3-32-32v-1.2c0-4.5 .9-8.9 2.7-13.1l18.2-42.4z"],"cart-flatbed":[640,512,["dolly-flatbed"],"f474","M32 0C14.3 0 0 14.3 0 32S14.3 64 32 64H48c8.8 0 16 7.2 16 16V368c0 44.2 35.8 80 80 80h18.7c-1.8 5-2.7 10.4-2.7 16c0 26.5 21.5 48 48 48s48-21.5 48-48c0-5.6-1-11-2.7-16H450.7c-1.8 5-2.7 10.4-2.7 16c0 26.5 21.5 48 48 48s48-21.5 48-48c0-5.6-1-11-2.7-16H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H144c-8.8 0-16-7.2-16-16V80C128 35.8 92.2 0 48 0H32zM192 80V272c0 26.5 21.5 48 48 48H560c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48H464V176c0 5.9-3.2 11.3-8.5 14.1s-11.5 2.5-16.4-.8L400 163.2l-39.1 26.1c-4.9 3.3-11.2 3.6-16.4 .8s-8.5-8.2-8.5-14.1V32H240c-26.5 0-48 21.5-48 48z"],"ban-smoking":[512,512,[128685,"smoking-ban"],"f54d","M99.5 144.8L178.7 224l96 96 92.5 92.5C335.9 434.9 297.5 448 256 448C150 448 64 362 64 256c0-41.5 13.1-79.9 35.5-111.2zM333.3 288l-32-32H384v32H333.3zm32 32H400c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H269.3L144.8 99.5C176.1 77.1 214.5 64 256 64c106 0 192 86 192 192c0 41.5-13.1 79.9-35.5 111.2L365.3 320zM256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM272 96c-8.8 0-16 7.2-16 16c0 26.5 21.5 48 48 48h32c8.8 0 16 7.2 16 16s7.2 16 16 16s16-7.2 16-16c0-26.5-21.5-48-48-48H304c-8.8 0-16-7.2-16-16s-7.2-16-16-16zM229.5 320l-96-96H112c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16H229.5z"],terminal:[576,512,[],"f120","M9.4 86.6C-3.1 74.1-3.1 53.9 9.4 41.4s32.8-12.5 45.3 0l192 192c12.5 12.5 12.5 32.8 0 45.3l-192 192c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L178.7 256 9.4 86.6zM256 416H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H256c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],"mobile-button":[384,512,[],"f10b","M80 0C44.7 0 16 28.7 16 64V448c0 35.3 28.7 64 64 64H304c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H80zM192 400a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"house-medical-flag":[640,512,[],"e514","M480 0c17.7 0 32 14.3 32 32H624c8.8 0 16 7.2 16 16V176c0 8.8-7.2 16-16 16H512V512H448V192 32c0-17.7 14.3-32 32-32zM276.8 39.7L416 159V512h1l-.2 0H96c-17.7 0-32-14.3-32-32V288H32c-13.4 0-25.4-8.3-30-20.9s-1-26.7 9.2-35.4l224-192c12-10.3 29.7-10.3 41.7 0zM224 208v48H176c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V320h48c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H288V208c0-8.8-7.2-16-16-16H240c-8.8 0-16 7.2-16 16z"],"basket-shopping":[576,512,["shopping-basket"],"f291","M253.3 35.1c6.1-11.8 1.5-26.3-10.2-32.4s-26.3-1.5-32.4 10.2L117.6 192H32c-17.7 0-32 14.3-32 32s14.3 32 32 32L83.9 463.5C91 492 116.6 512 146 512H430c29.4 0 55-20 62.1-48.5L544 256c17.7 0 32-14.3 32-32s-14.3-32-32-32H458.4L365.3 12.9C359.2 1.2 344.7-3.4 332.9 2.7s-16.3 20.6-10.2 32.4L404.3 192H171.7L253.3 35.1zM192 304v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V304c0-8.8 7.2-16 16-16s16 7.2 16 16zm96-16c8.8 0 16 7.2 16 16v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V304c0-8.8 7.2-16 16-16zm128 16v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V304c0-8.8 7.2-16 16-16s16 7.2 16 16z"],tape:[576,512,[],"f4db","M380.8 416c41.5-40.7 67.2-97.3 67.2-160C448 132.3 347.7 32 224 32S0 132.3 0 256S100.3 480 224 480H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H380.8zM224 160a96 96 0 1 1 0 192 96 96 0 1 1 0-192zm64 96a64 64 0 1 0 -128 0 64 64 0 1 0 128 0z"],"bus-simple":[448,512,["bus-alt"],"f55e","M224 0C348.8 0 448 35.2 448 80V96 416c0 17.7-14.3 32-32 32v32c0 17.7-14.3 32-32 32H352c-17.7 0-32-14.3-32-32V448H128v32c0 17.7-14.3 32-32 32H64c-17.7 0-32-14.3-32-32l0-32c-17.7 0-32-14.3-32-32V96 80C0 35.2 99.2 0 224 0zM64 128V256c0 17.7 14.3 32 32 32H352c17.7 0 32-14.3 32-32V128c0-17.7-14.3-32-32-32H96c-17.7 0-32 14.3-32 32zM80 400a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm288 0a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],eye:[576,512,[128065],"f06e","M288 32c-80.8 0-145.5 36.8-192.6 80.6C48.6 156 17.3 208 2.5 243.7c-3.3 7.9-3.3 16.7 0 24.6C17.3 304 48.6 356 95.4 399.4C142.5 443.2 207.2 480 288 480s145.5-36.8 192.6-80.6c46.8-43.5 78.1-95.4 93-131.1c3.3-7.9 3.3-16.7 0-24.6c-14.9-35.7-46.2-87.7-93-131.1C433.5 68.8 368.8 32 288 32zM144 256a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm144-64c0 35.3-28.7 64-64 64c-7.1 0-13.9-1.2-20.3-3.3c-5.5-1.8-11.9 1.6-11.7 7.4c.3 6.9 1.3 13.8 3.2 20.7c13.7 51.2 66.4 81.6 117.6 67.9s81.6-66.4 67.9-117.6c-11.1-41.5-47.8-69.4-88.6-71.1c-5.8-.2-9.2 6.1-7.4 11.7c2.1 6.4 3.3 13.2 3.3 20.3z"],"face-sad-cry":[512,512,[128557,"sad-cry"],"f5b3","M352 493.4c-29.6 12-62.1 18.6-96 18.6s-66.4-6.6-96-18.6V288c0-8.8-7.2-16-16-16s-16 7.2-16 16V477.8C51.5 433.5 0 350.8 0 256C0 114.6 114.6 0 256 0S512 114.6 512 256c0 94.8-51.5 177.5-128 221.8V288c0-8.8-7.2-16-16-16s-16 7.2-16 16V493.4zM195.2 233.6c5.3 7.1 15.3 8.5 22.4 3.2s8.5-15.3 3.2-22.4c-30.4-40.5-91.2-40.5-121.6 0c-5.3 7.1-3.9 17.1 3.2 22.4s17.1 3.9 22.4-3.2c17.6-23.5 52.8-23.5 70.4 0zm121.6 0c17.6-23.5 52.8-23.5 70.4 0c5.3 7.1 15.3 8.5 22.4 3.2s8.5-15.3 3.2-22.4c-30.4-40.5-91.2-40.5-121.6 0c-5.3 7.1-3.9 17.1 3.2 22.4s17.1 3.9 22.4-3.2zM208 336v32c0 26.5 21.5 48 48 48s48-21.5 48-48V336c0-26.5-21.5-48-48-48s-48 21.5-48 48z"],"audio-description":[576,512,[],"f29e","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM213.5 173.3l72 144c5.9 11.9 1.1 26.3-10.7 32.2s-26.3 1.1-32.2-10.7l-9.4-18.9H150.9l-9.4 18.9c-5.9 11.9-20.3 16.7-32.2 10.7s-16.7-20.3-10.7-32.2l72-144c4.1-8.1 12.4-13.3 21.5-13.3s17.4 5.1 21.5 13.3zm-.4 106.6L192 237.7l-21.1 42.2h42.2zM304 184c0-13.3 10.7-24 24-24h56c53 0 96 43 96 96s-43 96-96 96H328c-13.3 0-24-10.7-24-24V184zm48 24v96h32c26.5 0 48-21.5 48-48s-21.5-48-48-48H352z"],"person-military-to-person":[512,512,[],"e54c","M71 12.5c-8.6 1-15 8.2-15 16.8c0 9.3 7.5 16.8 16.7 16.9H184.1c8.8-.1 15.9-7.2 15.9-16V16c0-9.5-8.3-17-17.8-15.9L71 12.5zM189.5 78.1H66.5C64.9 83.8 64 89.8 64 96c0 35.3 28.7 64 64 64s64-28.7 64-64c0-6.2-.9-12.2-2.5-17.9zM32 256v32c0 17.7 14.3 32 32 32H192c1.8 0 3.5-.1 5.2-.4L53 208.6C40.1 220.3 32 237.2 32 256zm190.2 42.5c1.1-3.3 1.8-6.8 1.8-10.5V256c0-35.3-28.7-64-64-64H96c-3.7 0-7.4 .3-10.9 .9L222.2 298.5zM384 160a64 64 0 1 0 0-128 64 64 0 1 0 0 128zm-32 32c-35.3 0-64 28.7-64 64v32c0 17.7 14.3 32 32 32H448c17.7 0 32-14.3 32-32V256c0-35.3-28.7-64-64-64H352zM215.8 450.1c5.2-4.6 8.2-11.1 8.2-18.1s-3-13.5-8.2-18.1l-64-56c-7.1-6.2-17.1-7.7-25.7-3.8S112 366.6 112 376v32l-88 0c-13.3 0-24 10.7-24 24s10.7 24 24 24l88 0v32c0 9.4 5.5 18 14.1 21.9s18.6 2.4 25.7-3.8l64-56zM288 431.9c0 6.9 2.9 13.5 8.1 18.1l64 56.4c7.1 6.2 17.1 7.8 25.7 3.9s14.1-12.4 14.1-21.9l0-32.4 88 0c13.3 0 24-10.7 24-24s-10.7-24-24-24l-88 0 0-32c0-9.4-5.5-18-14.1-21.9s-18.6-2.4-25.7 3.8l-64 56c-5.2 4.5-8.2 11.1-8.2 18z"],"file-shield":[576,512,[],"e4f0","M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v47l-92.8 37.1c-21.3 8.5-35.2 29.1-35.2 52c0 56.6 18.9 148 94.2 208.3c-9 4.8-19.3 7.6-30.2 7.6H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zm39.1 97.7c5.7-2.3 12.1-2.3 17.8 0l120 48C570 277.4 576 286.2 576 296c0 63.3-25.9 168.8-134.8 214.2c-5.9 2.5-12.6 2.5-18.5 0C313.9 464.8 288 359.3 288 296c0-9.8 6-18.6 15.1-22.3l120-48zM527.4 312L432 273.8V461.7c68.2-33 91.5-99 95.4-149.7z"],"user-slash":[640,512,[],"f506","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L353.3 251.6C407.9 237 448 187.2 448 128C448 57.3 390.7 0 320 0C250.2 0 193.5 55.8 192 125.2L38.8 5.1zM264.3 304.3C170.5 309.4 96 387.2 96 482.3c0 16.4 13.3 29.7 29.7 29.7H514.3c3.9 0 7.6-.7 11-2.1l-261-205.6z"],pen:[512,512,[128394],"f304","M362.7 19.3L314.3 67.7 444.3 197.7l48.4-48.4c25-25 25-65.5 0-90.5L453.3 19.3c-25-25-65.5-25-90.5 0zm-71 71L58.6 323.5c-10.4 10.4-18 23.3-22.2 37.4L1 481.2C-1.5 489.7 .8 498.8 7 505s15.3 8.5 23.7 6.1l120.3-35.4c14.1-4.2 27-11.8 37.4-22.2L421.7 220.3 291.7 90.3z"],"tower-observation":[512,512,[],"e586","M241.7 3.4c9-4.5 19.6-4.5 28.6 0l160 80c15.8 7.9 22.2 27.1 14.3 42.9C439 137.5 427.7 144 416 144v80c0 17.7-14.3 32-32 32h-4.9l32 192H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H384.5c-.4 0-.8 0-1.1 0H128.6c-.4 0-.8 0-1.1 0H32c-17.7 0-32-14.3-32-32s14.3-32 32-32h68.9l32-192H128c-17.7 0-32-14.3-32-32V144c-11.7 0-23-6.5-28.6-17.7c-7.9-15.8-1.5-35 14.3-42.9l160-80zM314.5 448L256 399.2 197.5 448h117zM197.8 256l-4.7 28.3L256 336.8l62.9-52.5L314.2 256H197.8zm-13.9 83.2l-11.2 67L218.5 368l-34.6-28.8zM293.5 368l45.8 38.1-11.2-67L293.5 368zM176 128c-8.8 0-16 7.2-16 16s7.2 16 16 16H336c8.8 0 16-7.2 16-16s-7.2-16-16-16H176z"],"file-code":[384,512,[],"f1c9","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM153 289l-31 31 31 31c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L71 337c-9.4-9.4-9.4-24.6 0-33.9l48-48c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9zM265 255l48 48c9.4 9.4 9.4 24.6 0 33.9l-48 48c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l31-31-31-31c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0z"],signal:[640,512,[128246,"signal-5","signal-perfect"],"f012","M576 0c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V32c0-17.7 14.3-32 32-32zM448 96c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V128c0-17.7 14.3-32 32-32zM352 224V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V224c0-17.7 14.3-32 32-32s32 14.3 32 32zM192 288c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V320c0-17.7 14.3-32 32-32zM96 416v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V416c0-17.7 14.3-32 32-32s32 14.3 32 32z"],bus:[576,512,[128653],"f207","M288 0C422.4 0 512 35.2 512 80V96l0 32c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32l0 160c0 17.7-14.3 32-32 32v32c0 17.7-14.3 32-32 32H416c-17.7 0-32-14.3-32-32V448H192v32c0 17.7-14.3 32-32 32H128c-17.7 0-32-14.3-32-32l0-32c-17.7 0-32-14.3-32-32l0-160c-17.7 0-32-14.3-32-32V160c0-17.7 14.3-32 32-32h0V96h0V80C64 35.2 153.6 0 288 0zM128 160v96c0 17.7 14.3 32 32 32H272V128H160c-17.7 0-32 14.3-32 32zM304 288H416c17.7 0 32-14.3 32-32V160c0-17.7-14.3-32-32-32H304V288zM144 400a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm288 0a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM384 80c0-8.8-7.2-16-16-16H208c-8.8 0-16 7.2-16 16s7.2 16 16 16H368c8.8 0 16-7.2 16-16z"],"heart-circle-xmark":[576,512,[],"e501","M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9l2.6-2.4C267.2 438.6 256 404.6 256 368c0-97.2 78.8-176 176-176c28.3 0 55 6.7 78.7 18.5c.9-6.5 1.3-13 1.3-19.6v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L454.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L432 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L409.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L432 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z"],"house-chimney":[576,512,[63499,"home-lg"],"e3af","M543.8 287.6c17 0 32-14 32-32.1c1-9-3-17-11-24L512 185V64c0-17.7-14.3-32-32-32H448c-17.7 0-32 14.3-32 32v36.7L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32v69.7c-.1 .9-.1 1.8-.1 2.8V472c0 22.1 17.9 40 40 40h16c1.2 0 2.4-.1 3.6-.2c1.5 .1 3 .2 4.5 .2H160h24c22.1 0 40-17.9 40-40V448 384c0-17.7 14.3-32 32-32h64c17.7 0 32 14.3 32 32v64 24c0 22.1 17.9 40 40 40h24 32.5c1.4 0 2.8 0 4.2-.1c1.1 .1 2.2 .1 3.3 .1h16c22.1 0 40-17.9 40-40V455.8c.3-2.6 .5-5.3 .5-8.1l-.7-160.2h32z"],"window-maximize":[512,512,[128470],"f2d0","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM96 96H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H96c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],"face-frown":[512,512,[9785,"frown"],"f119","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM159.3 388.7c-2.6 8.4-11.6 13.2-20 10.5s-13.2-11.6-10.5-20C145.2 326.1 196.3 288 256 288s110.8 38.1 127.3 91.3c2.6 8.4-2.1 17.4-10.5 20s-17.4-2.1-20-10.5C340.5 349.4 302.1 320 256 320s-84.5 29.4-96.7 68.7zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],prescription:[448,512,[],"f5b1","M32 0C14.3 0 0 14.3 0 32V192v96c0 17.7 14.3 32 32 32s32-14.3 32-32V224h50.7l128 128L137.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L288 397.3 393.4 502.6c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L333.3 352 438.6 246.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L288 306.7l-85.8-85.8C251.4 209.1 288 164.8 288 112C288 50.1 237.9 0 176 0H32zM176 160H64V64H176c26.5 0 48 21.5 48 48s-21.5 48-48 48z"],shop:[640,512,["store-alt"],"f54f","M36.8 192H603.2c20.3 0 36.8-16.5 36.8-36.8c0-7.3-2.2-14.4-6.2-20.4L558.2 21.4C549.3 8 534.4 0 518.3 0H121.7c-16 0-31 8-39.9 21.4L6.2 134.7c-4 6.1-6.2 13.2-6.2 20.4C0 175.5 16.5 192 36.8 192zM64 224V384v80c0 26.5 21.5 48 48 48H336c26.5 0 48-21.5 48-48V384 224H320V384H128V224H64zm448 0V480c0 17.7 14.3 32 32 32s32-14.3 32-32V224H512z"],"floppy-disk":[448,512,[128190,128426,"save"],"f0c7","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V173.3c0-17-6.7-33.3-18.7-45.3L352 50.7C340 38.7 323.7 32 306.7 32H64zm0 96c0-17.7 14.3-32 32-32H288c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V128zM224 288a64 64 0 1 1 0 128 64 64 0 1 1 0-128z"],vihara:[640,512,[],"f6a7","M281 22L305.8 4.7c1.3-.9 2.7-1.8 4.1-2.4C313.1 .7 316.6 0 320 0s6.9 .7 10.1 2.2c1.4 .7 2.8 1.5 4.1 2.4L359 22C393 45.8 430.8 63.5 470.8 74.4l23 6.3c1.8 .5 3.6 1.1 5.2 2c3.2 1.7 5.9 4 8.1 6.8c3.8 4.9 5.6 11.3 4.7 17.8c-.4 2.8-1.2 5.4-2.5 7.8c-1.7 3.2-4 5.9-6.8 8.1c-4.3 3.2-9.6 5.1-15.1 4.9H480v56.1l6.4 5.1 5.2 4.1c21.1 16.7 45 29.6 70.5 38.1l28.9 9.6c1.6 .5 3.2 1.2 4.6 2c3.1 1.7 5.8 4.1 7.8 6.9s3.5 6.1 4.1 9.6c.5 2.7 .6 5.5 .1 8.3s-1.4 5.4-2.7 7.8c-1.7 3.1-4.1 5.8-6.9 7.8s-6.1 3.5-9.6 4.1c-1.6 .3-3.3 .4-5 .4H544v65.9c20.5 22.8 47.4 39.2 77.4 46.7C632 403 640 412.6 640 424c0 13.3-10.7 24-24 24H576v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H352v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H128v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H24c-13.3 0-24-10.7-24-24c0-11.4 8-21 18.6-23.4c30-7.6 56.9-23.9 77.4-46.7V288H56.6c-1.7 0-3.4-.1-5-.4c-3.5-.7-6.8-2.1-9.6-4.1s-5.2-4.7-7-7.8c-1.3-2.4-2.3-5-2.7-7.8s-.4-5.6 .1-8.3c.7-3.5 2.1-6.8 4.1-9.6s4.7-5.2 7.8-6.9c1.4-.8 3-1.5 4.6-2l28.9-9.6c25.5-8.5 49.4-21.4 70.5-38.1l5.2-4.1 6.4-5.1V176 128h-7.5c-5.5 .1-10.8-1.7-15.1-4.9c-2.8-2.1-5.1-4.8-6.8-8.1c-1.2-2.4-2.1-5-2.5-7.8c-.9-6.5 .9-12.8 4.7-17.8c2.1-2.8 4.8-5.1 8.1-6.8c1.6-.8 3.4-1.5 5.2-2l23-6.3C209.2 63.5 247 45.8 281 22zM416 128H320 224v64h72 48 72V128zM160 288v64H296h24 24H480V288H344 320h0H296 160z"],"scale-unbalanced":[640,512,["balance-scale-left"],"f515","M522.1 62.4c16.8-5.6 25.8-23.7 20.2-40.5S518.6-3.9 501.9 1.6l-113 37.7C375 15.8 349.3 0 320 0c-44.2 0-80 35.8-80 80c0 3 .2 5.9 .5 8.8L117.9 129.6c-16.8 5.6-25.8 23.7-20.2 40.5s23.7 25.8 40.5 20.2l135.5-45.2c4.5 3.2 9.3 5.9 14.4 8.2V480c0 17.7 14.3 32 32 32H512c17.7 0 32-14.3 32-32s-14.3-32-32-32H352V153.3c21-9.2 37.2-27 44.2-49l125.9-42zM439.6 288L512 163.8 584.4 288H439.6zM512 384c62.9 0 115.2-34 126-78.9c2.6-11-1-22.3-6.7-32.1L536.1 109.8c-5-8.6-14.2-13.8-24.1-13.8s-19.1 5.3-24.1 13.8L392.7 273.1c-5.7 9.8-9.3 21.1-6.7 32.1C396.8 350 449.1 384 512 384zM129.2 291.8L201.6 416H56.7l72.4-124.2zM3.2 433.1C14 478 66.3 512 129.2 512s115.2-34 126-78.9c2.6-11-1-22.3-6.7-32.1L153.2 237.8c-5-8.6-14.2-13.8-24.1-13.8s-19.1 5.3-24.1 13.8L9.9 401.1c-5.7 9.8-9.3 21.1-6.7 32.1z"],"sort-up":[320,512,["sort-asc"],"f0de","M182.6 41.4c-12.5-12.5-32.8-12.5-45.3 0l-128 128c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8H288c12.9 0 24.6-7.8 29.6-19.8s2.2-25.7-6.9-34.9l-128-128z"],"comment-dots":[512,512,[128172,62075,"commenting"],"f4ad","M256 448c141.4 0 256-93.1 256-208S397.4 32 256 32S0 125.1 0 240c0 45.1 17.7 86.8 47.7 120.9c-1.9 24.5-11.4 46.3-21.4 62.9c-5.5 9.2-11.1 16.6-15.2 21.6c-2.1 2.5-3.7 4.4-4.9 5.7c-.6 .6-1 1.1-1.3 1.4l-.3 .3 0 0 0 0 0 0 0 0c-4.6 4.6-5.9 11.4-3.4 17.4c2.5 6 8.3 9.9 14.8 9.9c28.7 0 57.6-8.9 81.6-19.3c22.9-10 42.4-21.9 54.3-30.6c31.8 11.5 67 17.9 104.1 17.9zM128 208a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 0a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm96 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],"plant-wilt":[512,512,[],"e5aa","M288 120c0-30.9 25.1-56 56-56s56 25.1 56 56v13c-29.3 10-48 34.5-48 70.1c0 27.9 25.3 74.8 66 111.6c3.8 3.5 8.9 5.3 14 5.3s10.2-1.8 14-5.3c40.7-36.8 66-83.7 66-111.6c0-35.6-18.7-60.2-48-70.1V120C464 53.7 410.3 0 344 0S224 53.7 224 120v21.8C207.3 133 188.2 128 168 128c-66.3 0-120 53.7-120 120v13c-29.3 10-48 34.5-48 70.1C0 359 25.3 405.9 66 442.7c3.8 3.5 8.9 5.3 14 5.3s10.2-1.8 14-5.3c40.7-36.8 66-83.7 66-111.6c0-35.6-18.7-60.2-48-70.1V248c0-30.9 25.1-56 56-56s56 25.1 56 56v32V480c0 17.7 14.3 32 32 32s32-14.3 32-32V280 248 120z"],diamond:[512,512,[9830],"f219","M284.3 11.7c-15.6-15.6-40.9-15.6-56.6 0l-216 216c-15.6 15.6-15.6 40.9 0 56.6l216 216c15.6 15.6 40.9 15.6 56.6 0l216-216c15.6-15.6 15.6-40.9 0-56.6l-216-216z"],"face-grin-squint":[512,512,[128518,"grin-squint"],"f585","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM388.1 312.8c12.3-3.8 24.3 6.9 19.3 18.7C382.4 390.6 324.2 432 256.3 432s-126.2-41.4-151.1-100.5c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19zM133.5 146.7l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6c0-9 9.6-14.7 17.5-10.5zM396 157.1c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9c7.9-4.2 17.5 1.5 17.5 10.5z"],"hand-holding-dollar":[576,512,["hand-holding-usd"],"f4c0","M312 24V34.5c6.4 1.2 12.6 2.7 18.2 4.2c12.8 3.4 20.4 16.6 17 29.4s-16.6 20.4-29.4 17c-10.9-2.9-21.1-4.9-30.2-5c-7.3-.1-14.7 1.7-19.4 4.4c-2.1 1.3-3.1 2.4-3.5 3c-.3 .5-.7 1.2-.7 2.8c0 .3 0 .5 0 .6c.2 .2 .9 1.2 3.3 2.6c5.8 3.5 14.4 6.2 27.4 10.1l.9 .3c11.1 3.3 25.9 7.8 37.9 15.3c13.7 8.6 26.1 22.9 26.4 44.9c.3 22.5-11.4 38.9-26.7 48.5c-6.7 4.1-13.9 7-21.3 8.8V232c0 13.3-10.7 24-24 24s-24-10.7-24-24V220.6c-9.5-2.3-18.2-5.3-25.6-7.8c-2.1-.7-4.1-1.4-6-2c-12.6-4.2-19.4-17.8-15.2-30.4s17.8-19.4 30.4-15.2c2.6 .9 5 1.7 7.3 2.5c13.6 4.6 23.4 7.9 33.9 8.3c8 .3 15.1-1.6 19.2-4.1c1.9-1.2 2.8-2.2 3.2-2.9c.4-.6 .9-1.8 .8-4.1l0-.2c0-1 0-2.1-4-4.6c-5.7-3.6-14.3-6.4-27.1-10.3l-1.9-.6c-10.8-3.2-25-7.5-36.4-14.4c-13.5-8.1-26.5-22-26.6-44.1c-.1-22.9 12.9-38.6 27.7-47.4c6.4-3.8 13.3-6.4 20.2-8.2V24c0-13.3 10.7-24 24-24s24 10.7 24 24zM568.2 336.3c13.1 17.8 9.3 42.8-8.5 55.9L433.1 485.5c-23.4 17.2-51.6 26.5-80.7 26.5H192 32c-17.7 0-32-14.3-32-32V416c0-17.7 14.3-32 32-32H68.8l44.9-36c22.7-18.2 50.9-28 80-28H272h16 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H288 272c-8.8 0-16 7.2-16 16s7.2 16 16 16H392.6l119.7-88.2c17.8-13.1 42.8-9.3 55.9 8.5zM193.6 384l0 0-.9 0c.3 0 .6 0 .9 0z"],bacterium:[512,512,[],"e05a","M423.1 30.6c3.6-12.7-3.7-26-16.5-29.7s-26 3.7-29.7 16.5l-4.2 14.7c-9.8-.4-19.9 .5-29.9 2.8c-12.1 2.8-23.7 5.9-34.9 9.4l-5.9-13.7c-5.2-12.2-19.3-17.8-31.5-12.6s-17.8 19.3-12.6 31.5l4.9 11.3c-22 9.4-42 20.1-60.2 31.8L196 82.7c-7.4-11-22.3-14-33.3-6.7s-14 22.3-6.7 33.3l7.8 11.6c-18 15-33.7 30.8-47.3 47.1L103 157.3c-10.4-8.3-25.5-6.6-33.7 3.7s-6.6 25.5 3.7 33.7l15 12c-2.1 3.2-4.1 6.5-6 9.7c-9.4 15.7-17 31-23.2 45.3l-9.9-3.9c-12.3-4.9-26.3 1.1-31.2 13.4s1.1 26.3 13.4 31.2l11.6 4.6c-.3 1.1-.6 2.1-.9 3.1c-3.5 12.5-5.7 23.2-7.1 31.3c-.7 4.1-1.2 7.5-1.6 10.3c-.2 1.4-.3 2.6-.4 3.6l-.1 1.4-.1 .6 0 .3 0 .1c0 0 0 .1 39.2 3.7l0 0-39.2-3.6c-.5 5-.6 10-.4 14.9l-14.7 4.2C4.7 380.6-2.7 393.8 .9 406.6s16.9 20.1 29.7 16.5l13.8-3.9c10.6 20.7 27.6 37.8 48.5 48.5l-3.9 13.7c-3.6 12.7 3.7 26 16.5 29.7s26-3.7 29.7-16.5l4.2-14.7c23.8 1 46.3-5.5 65.1-17.6L215 473c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-10.6-10.6c9.1-14.1 15.1-30.5 17-48.3l.1-.8c.3-1.7 1-5.1 2.3-9.8l.2-.8 12.6 5.4c12.2 5.2 26.3-.4 31.5-12.6s-.4-26.3-12.6-31.5l-11.3-4.8c9.9-14.9 24.9-31.6 48.6-46l2.1 7.5c3.6 12.7 16.9 20.1 29.7 16.5s20.1-16.9 16.5-29.7L371 259.2c6.9-2.2 14.3-4.3 22.2-6.1c12.9-3 24.7-8 35.2-14.8L439 249c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-10.6-10.6c12.2-19 18.6-41.6 17.6-65.1l14.7-4.2c12.7-3.6 20.1-16.9 16.5-29.7s-16.9-20.1-29.7-16.5l-13.7 3.9c-10.8-21.2-28-38-48.5-48.5l3.9-13.8zM92.1 363.3l0 0L144 368l-51.9-4.7zM112 320a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM240 184a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],"hand-pointer":[448,512,[],"f25a","M128 40c0-22.1 17.9-40 40-40s40 17.9 40 40V188.2c8.5-7.6 19.7-12.2 32-12.2c20.6 0 38.2 13 45 31.2c8.8-9.3 21.2-15.2 35-15.2c25.3 0 46 19.5 47.9 44.3c8.5-7.7 19.8-12.3 32.1-12.3c26.5 0 48 21.5 48 48v48 16 48c0 70.7-57.3 128-128 128l-16 0H240l-.1 0h-5.2c-5 0-9.9-.3-14.7-1c-55.3-5.6-106.2-34-140-79L8 336c-13.3-17.7-9.7-42.7 8-56s42.7-9.7 56 8l56 74.7V40zM240 304c0-8.8-7.2-16-16-16s-16 7.2-16 16v96c0 8.8 7.2 16 16 16s16-7.2 16-16V304zm48-16c-8.8 0-16 7.2-16 16v96c0 8.8 7.2 16 16 16s16-7.2 16-16V304c0-8.8-7.2-16-16-16zm80 16c0-8.8-7.2-16-16-16s-16 7.2-16 16v96c0 8.8 7.2 16 16 16s16-7.2 16-16V304z"],"drum-steelpan":[576,512,[],"f56a","M288 32c159.1 0 288 48 288 128V352c0 80-128.9 128-288 128S0 432 0 352V160C0 80 128.9 32 288 32zM528 160c0-9.9-8-29.9-55-49.8c-18.6-7.9-40.9-14.4-66-19.4l-27.8 43.6c-7.3 11.5-11.2 24.8-11.2 38.4c0 17.5 6.4 34.4 18.1 47.5l9.8 11c29.8-5.2 55.9-12.5 77.2-21.5c47.1-19.9 55-39.9 55-49.8zM349.2 237.3c-8-26.2-32.4-45.3-61.2-45.3s-53.3 19.1-61.2 45.3c19.4 1.7 39.9 2.7 61.2 2.7s41.8-.9 61.2-2.7zM169 90.8c-25.2 5-47.4 11.6-66 19.4C56 130.1 48 150.1 48 160s8 29.9 55 49.8c21.3 9 47.4 16.3 77.2 21.5l9.8-11c11.6-13.1 18.1-30 18.1-47.5c0-13.6-3.9-26.9-11.2-38.4L169 90.8zm56.3-8C224.5 87 224 91.5 224 96c0 35.3 28.7 64 64 64s64-28.7 64-64c0-4.5-.5-9-1.4-13.2C330.8 81 309.8 80 288 80s-42.8 1-62.6 2.8z"],"hand-scissors":[512,512,[],"f257","M40 208c-22.1 0-40 17.9-40 40s17.9 40 40 40l180.2 0c-7.6 8.5-12.2 19.7-12.2 32c0 25.3 19.5 46 44.3 47.9c-7.7 8.5-12.3 19.8-12.3 32.1c0 26.5 21.5 48 48 48l32 0 64 0c70.7 0 128-57.3 128-128l0-113.1c0-40.2-16-78.8-44.4-107.3C444.8 76.8 413.9 64 381.7 64L336 64c-21.3 0-39.3 13.9-45.6 33.1l74.5 23.7c8.4 2.7 13.1 11.7 10.4 20.1s-11.7 13.1-20.1 10.4L288 129.9l0 .1L84 65.8C62.9 59.2 40.5 70.9 33.8 92s5.1 43.5 26.2 50.2L269.5 208 40 208z"],"hands-praying":[640,512,["praying-hands"],"f684","M351.2 4.8c3.2-2 6.6-3.3 10-4.1c4.7-1 9.6-.9 14.1 .1c7.7 1.8 14.8 6.5 19.4 13.6L514.6 194.2c8.8 13.1 13.4 28.6 13.4 44.4v73.5c0 6.9 4.4 13 10.9 15.2l79.2 26.4C631.2 358 640 370.2 640 384v96c0 9.9-4.6 19.3-12.5 25.4s-18.1 8.1-27.7 5.5L431 465.9c-56-14.9-95-65.7-95-123.7V224c0-17.7 14.3-32 32-32s32 14.3 32 32v80c0 8.8 7.2 16 16 16s16-7.2 16-16V219.1c0-7-1.8-13.8-5.3-19.8L340.3 48.1c-1.7-3-2.9-6.1-3.6-9.3c-1-4.7-1-9.6 .1-14.1c1.9-8 6.8-15.2 14.3-19.9zm-62.4 0c7.5 4.6 12.4 11.9 14.3 19.9c1.1 4.6 1.2 9.4 .1 14.1c-.7 3.2-1.9 6.3-3.6 9.3L213.3 199.3c-3.5 6-5.3 12.9-5.3 19.8V304c0 8.8 7.2 16 16 16s16-7.2 16-16V224c0-17.7 14.3-32 32-32s32 14.3 32 32V342.3c0 58-39 108.7-95 123.7l-168.7 45c-9.6 2.6-19.9 .5-27.7-5.5S0 490 0 480V384c0-13.8 8.8-26 21.9-30.4l79.2-26.4c6.5-2.2 10.9-8.3 10.9-15.2V238.5c0-15.8 4.7-31.2 13.4-44.4L245.2 14.5c4.6-7.1 11.7-11.8 19.4-13.6c4.6-1.1 9.4-1.2 14.1-.1c3.5 .8 6.9 2.1 10 4.1z"],"arrow-rotate-right":[512,512,[8635,"arrow-right-rotate","arrow-rotate-forward","redo"],"f01e","M386.3 160H336c-17.7 0-32 14.3-32 32s14.3 32 32 32H464c17.7 0 32-14.3 32-32V64c0-17.7-14.3-32-32-32s-32 14.3-32 32v51.2L414.4 97.6c-87.5-87.5-229.3-87.5-316.8 0s-87.5 229.3 0 316.8s229.3 87.5 316.8 0c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0c-62.5 62.5-163.8 62.5-226.3 0s-62.5-163.8 0-226.3s163.8-62.5 226.3 0L386.3 160z"],biohazard:[576,512,[9763],"f780","M173.2 0c-1.8 0-3.5 .7-4.8 2C138.5 32.3 120 74 120 120c0 26.2 6 50.9 16.6 73c-22 2.4-43.8 9.1-64.2 20.5C37.9 232.8 13.3 262.4 .4 296c-.7 1.7-.5 3.7 .5 5.2c2.2 3.7 7.4 4.3 10.6 1.3C64.2 254.3 158 245.1 205 324s-8.1 153.1-77.6 173.2c-4.2 1.2-6.3 5.9-4.1 9.6c1 1.6 2.6 2.7 4.5 3c36.5 5.9 75.2 .1 109.7-19.2c20.4-11.4 37.4-26.5 50.5-43.8c13.1 17.3 30.1 32.4 50.5 43.8c34.5 19.3 73.3 25.2 109.7 19.2c1.9-.3 3.5-1.4 4.5-3c2.2-3.7 .1-8.4-4.1-9.6C379.1 477.1 324 403 371 324s140.7-69.8 193.5-21.4c3.2 2.9 8.4 2.3 10.6-1.3c1-1.6 1.1-3.5 .5-5.2c-12.9-33.6-37.5-63.2-72.1-82.5c-20.4-11.4-42.2-18.1-64.2-20.5C450 170.9 456 146.2 456 120c0-46-18.5-87.7-48.4-118c-1.3-1.3-3-2-4.8-2c-5 0-8.4 5.2-6.7 9.9C421.7 80.5 385.6 176 288 176S154.3 80.5 179.9 9.9c1.7-4.7-1.6-9.9-6.7-9.9zM240 272a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM181.7 417.6c6.3-11.8 9.8-25.1 8.6-39.8c-19.5-18-34-41.4-41.2-67.8c-12.5-8.1-26.2-11.8-40-12.4c-9-.4-18.1 .6-27.1 2.7c7.8 57.1 38.7 106.8 82.9 139.4c6.8-6.7 12.6-14.1 16.8-22.1zM288 64c-28.8 0-56.3 5.9-81.2 16.5c2 8.3 5 16.2 9 23.5c6.8 12.4 16.7 23.1 30.1 30.3c13.3-4.1 27.5-6.3 42.2-6.3s28.8 2.2 42.2 6.3c13.4-7.2 23.3-17.9 30.1-30.3c4-7.3 7-15.2 9-23.5C344.3 69.9 316.8 64 288 64zM426.9 310c-7.2 26.4-21.7 49.7-41.2 67.8c-1.2 14.7 2.2 28.1 8.6 39.8c4.3 8 10 15.4 16.8 22.1c44.3-32.6 75.2-82.3 82.9-139.4c-9-2.2-18.1-3.1-27.1-2.7c-13.8 .6-27.5 4.4-40 12.4z"],"location-crosshairs":[512,512,["location"],"f601","M256 0c17.7 0 32 14.3 32 32V66.7C368.4 80.1 431.9 143.6 445.3 224H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H445.3C431.9 368.4 368.4 431.9 288 445.3V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V445.3C143.6 431.9 80.1 368.4 66.7 288H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H66.7C80.1 143.6 143.6 80.1 224 66.7V32c0-17.7 14.3-32 32-32zM128 256a128 128 0 1 0 256 0 128 128 0 1 0 -256 0zm128-80a80 80 0 1 1 0 160 80 80 0 1 1 0-160z"],"mars-double":[640,512,[9891],"f227","M312 32c-9.7 0-18.5 5.8-22.2 14.8s-1.7 19.3 5.2 26.2l33.4 33.4L275.8 159c-28.4-19.5-62.7-31-99.8-31C78.8 128 0 206.8 0 304s78.8 176 176 176s176-78.8 176-176c0-37-11.4-71.4-31-99.8l52.6-52.6L407 185c6.9 6.9 17.2 8.9 26.2 5.2s14.8-12.5 14.8-22.2V56c0-13.3-10.7-24-24-24H312zm88 48h0v0l0 0zM64 304a112 112 0 1 1 224 0A112 112 0 1 1 64 304zM368 480c97.2 0 176-78.8 176-176c0-37-11.4-71.4-31-99.8l52.6-52.6L599 185c6.9 6.9 17.2 8.9 26.2 5.2s14.8-12.5 14.8-22.2V56c0-13.3-10.7-24-24-24H504c-9.7 0-18.5 5.8-22.2 14.8c-1.2 2.9-1.8 6-1.8 9l0 .2v.2c0 6.2 2.5 12.2 7 16.8l33.4 33.4L480 146.7V168c0 22.6-13.6 43.1-34.6 51.7c-.8 .3-1.7 .7-2.5 1C465.7 241.2 480 270.9 480 304c0 61.9-50.1 112-112 112c-5.4 0-10.8-.4-16-1.1c-12.9 20.4-29.1 38.3-48.1 53.1c19.8 7.8 41.4 12 64 12z"],"child-dress":[320,512,[],"e59c","M224 64A64 64 0 1 0 96 64a64 64 0 1 0 128 0zM88 400v80c0 17.7 14.3 32 32 32s32-14.3 32-32V400h16v80c0 17.7 14.3 32 32 32s32-14.3 32-32V400h17.8c10.9 0 18.6-10.7 15.2-21.1l-31.1-93.4 28.6 37.8c10.7 14.1 30.8 16.8 44.8 6.2s16.8-30.7 6.2-44.8L254.6 207c-22.4-29.6-57.5-47-94.6-47s-72.2 17.4-94.6 47L6.5 284.7c-10.7 14.1-7.9 34.2 6.2 44.8s34.2 7.9 44.8-6.2l28.7-37.8L55 378.9C51.6 389.3 59.3 400 70.2 400H88z"],"users-between-lines":[640,512,[],"e591","M0 24C0 10.7 10.7 0 24 0H616c13.3 0 24 10.7 24 24s-10.7 24-24 24H24C10.7 48 0 37.3 0 24zM0 488c0-13.3 10.7-24 24-24H616c13.3 0 24 10.7 24 24s-10.7 24-24 24H24c-13.3 0-24-10.7-24-24zM83.2 160a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM32 320c0-35.3 28.7-64 64-64h96c12.2 0 23.7 3.4 33.4 9.4c-37.2 15.1-65.6 47.2-75.8 86.6H64c-17.7 0-32-14.3-32-32zm461.6 32c-10.3-40.1-39.6-72.6-77.7-87.4c9.4-5.5 20.4-8.6 32.1-8.6h96c35.3 0 64 28.7 64 64c0 17.7-14.3 32-32 32H493.6zM391.2 290.4c32.1 7.4 58.1 30.9 68.9 61.6c3.5 10 5.5 20.8 5.5 32c0 17.7-14.3 32-32 32h-224c-17.7 0-32-14.3-32-32c0-11.2 1.9-22 5.5-32c10.5-29.7 35.3-52.8 66.1-60.9c7.8-2.1 16-3.1 24.5-3.1h96c7.4 0 14.7 .8 21.6 2.4zm44-130.4a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM321.6 96a80 80 0 1 1 0 160 80 80 0 1 1 0-160z"],"lungs-virus":[640,512,[],"e067","M320 0c17.7 0 32 14.3 32 32V156.2c-8.5-7.6-19.7-12.2-32-12.2s-23.5 4.6-32 12.2V32c0-17.7 14.3-32 32-32zM444.5 195.5c-16.4-16.4-41.8-18.5-60.5-6.1V165.3C384 127 415 96 453.3 96c21.7 0 42.8 10.2 55.8 28.8c15.4 22.1 44.3 65.4 71 116.9c26.5 50.9 52.4 112.5 59.6 170.3c.2 1.3 .2 2.6 .2 4v7c0 49.1-39.8 89-89 89c-7.3 0-14.5-.9-21.6-2.7l-72.7-18.2c-20.9-5.2-38.7-17.1-51.5-32.9c14 1.5 28.5-3 39.2-13.8l-22.6-22.6 22.6 22.6c18.7-18.7 18.7-49.1 0-67.9c-1.1-1.1-1.4-2-1.5-2.5c-.1-.8-.1-1.8 .4-2.9s1.2-1.9 1.8-2.3c.5-.3 1.3-.8 2.9-.8c26.5 0 48-21.5 48-48s-21.5-48-48-48c-1.6 0-2.4-.4-2.9-.8c-.6-.4-1.3-1.2-1.8-2.3s-.5-2.2-.4-2.9c.1-.6 .4-1.4 1.5-2.5c18.7-18.7 18.7-49.1 0-67.9zM421.8 421.8c-6.2 6.2-16.4 6.2-22.6 0C375.9 398.5 336 415 336 448c0 8.8-7.2 16-16 16s-16-7.2-16-16c0-33-39.9-49.5-63.2-26.2c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6C241.5 375.9 225 336 192 336c-8.8 0-16-7.2-16-16s7.2-16 16-16c33 0 49.5-39.9 26.2-63.2c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0C264.1 241.5 304 225 304 192c0-8.8 7.2-16 16-16s16 7.2 16 16c0 33 39.9 49.5 63.2 26.2c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6C398.5 264.1 415 304 448 304c8.8 0 16 7.2 16 16s-7.2 16-16 16c-33 0-49.5 39.9-26.2 63.2c6.2 6.2 6.2 16.4 0 22.6zM183.3 491.2l-72.7 18.2c-7.1 1.8-14.3 2.7-21.6 2.7c-49.1 0-89-39.8-89-89v-7c0-1.3 .1-2.7 .2-4c7.2-57.9 33.1-119.4 59.6-170.3c26.8-51.5 55.6-94.8 71-116.9c13-18.6 34-28.8 55.8-28.8C225 96 256 127 256 165.3v24.1c-18.6-12.4-44-10.3-60.5 6.1c-18.7 18.7-18.7 49.1 0 67.9c1.1 1.1 1.4 2 1.5 2.5c.1 .8 .1 1.8-.4 2.9s-1.2 1.9-1.8 2.3c-.5 .3-1.3 .8-2.9 .8c-26.5 0-48 21.5-48 48s21.5 48 48 48c1.6 0 2.4 .4 2.9 .8c.6 .4 1.3 1.2 1.8 2.3s.5 2.2 .4 2.9c-.1 .6-.4 1.4-1.5 2.5c-18.7 18.7-18.7 49.1 0 67.9c10.7 10.7 25.3 15.3 39.2 13.8c-12.8 15.9-30.6 27.7-51.5 32.9zM296 320a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm72 32a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z"],"face-grin-tears":[640,512,[128514,"grin-tears"],"f588","M548.6 371.4C506.4 454.8 419.9 512 320 512s-186.4-57.2-228.6-140.6c4.5-2.9 8.7-6.3 12.7-10.3c8.1-8.1 13.2-18.6 16.5-26.6c3.6-8.8 6.5-18.4 8.8-27.5c4.6-18.2 7.7-37 9.3-48.2c3.9-26.5-18.8-49.2-45.2-45.4c-6.8 .9-16.2 2.4-26.6 4.4C85.3 94.5 191.6 0 320 0S554.7 94.5 573.2 217.7c-10.3-2-19.8-3.5-26.6-4.4c-26.5-3.9-49.2 18.8-45.2 45.4c1.6 11.3 4.6 30 9.3 48.2c2.3 9.1 5.2 18.8 8.8 27.5c3.3 8.1 8.4 18.5 16.5 26.6c3.9 3.9 8.2 7.4 12.7 10.3zM107 254.1c-3.1 21.5-11.4 70.2-25.5 84.4c-.9 1-1.9 1.8-2.9 2.7C60 356.7 32 355.5 14.3 337.7c-18.7-18.7-19.1-48.8-.7-67.2c8.6-8.6 30.1-15.1 50.5-19.6c13-2.8 25.5-4.8 33.9-6c5.4-.8 9.9 3.7 9 9zm454.5 87.1c-.8-.6-1.5-1.3-2.3-2c-.2-.2-.5-.4-.7-.7c-14.1-14.1-22.5-62.9-25.5-84.4c-.8-5.4 3.7-9.9 9-9c1 .1 2.2 .3 3.3 .5c8.2 1.2 19.2 3 30.6 5.5c20.4 4.4 41.9 10.9 50.5 19.6c18.4 18.4 18 48.5-.7 67.2c-17.7 17.7-45.7 19-64.2 3.4zm-90.1-9.7c5-11.8-7-22.5-19.3-18.7c-39.7 12.2-84.5 19-131.8 19s-92.1-6.8-131.8-19c-12.3-3.8-24.3 6.9-19.3 18.7c25 59.1 83.2 100.5 151.1 100.5s126.2-41.4 151.1-100.5zM281.6 228.8l0 0 0 0 0 0c2.1 2.8 5.7 3.9 8.9 2.8s5.5-4.1 5.5-7.6c0-17.9-6.7-35.6-16.6-48.8c-9.8-13-23.9-23.2-39.4-23.2s-29.6 10.2-39.4 23.2C190.7 188.4 184 206.1 184 224c0 3.4 2.2 6.5 5.5 7.6s6.9 0 8.9-2.8l0 0 0 0 0 0 .2-.2c.2-.2 .4-.5 .7-.9c.6-.8 1.6-2 2.8-3.4c2.5-2.8 6-6.6 10.2-10.3c8.8-7.8 18.8-14 27.7-14s18.9 6.2 27.7 14c4.2 3.7 7.7 7.5 10.2 10.3c1.2 1.4 2.2 2.6 2.8 3.4c.3 .4 .6 .7 .7 .9l.2 .2 0 0zm160 0l0 0 0 0c2.1 2.8 5.7 3.9 8.9 2.8s5.5-4.1 5.5-7.6c0-17.9-6.7-35.6-16.6-48.8c-9.8-13-23.9-23.2-39.4-23.2s-29.6 10.2-39.4 23.2C350.7 188.4 344 206.1 344 224c0 3.4 2.2 6.5 5.5 7.6s6.9 0 8.9-2.8l0 0 0 0 0 0 .2-.2c.2-.2 .4-.5 .7-.9c.6-.8 1.6-2 2.8-3.4c2.5-2.8 6-6.6 10.2-10.3c8.8-7.8 18.8-14 27.7-14s18.9 6.2 27.7 14c4.2 3.7 7.7 7.5 10.2 10.3c1.2 1.4 2.2 2.6 2.8 3.4c.3 .4 .6 .7 .7 .9l.2 .2 0 0 0 0z"],phone:[512,512,[128222,128379],"f095","M164.9 24.6c-7.7-18.6-28-28.5-47.4-23.2l-88 24C12.1 30.2 0 46 0 64C0 311.4 200.6 512 448 512c18 0 33.8-12.1 38.6-29.5l24-88c5.3-19.4-4.6-39.7-23.2-47.4l-96-40c-16.3-6.8-35.2-2.1-46.3 11.6L304.7 368C234.3 334.7 177.3 277.7 144 207.3L193.3 167c13.7-11.2 18.4-30 11.6-46.3l-40-96z"],"calendar-xmark":[448,512,["calendar-times"],"f273","M128 0c17.7 0 32 14.3 32 32V64H288V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64h48c26.5 0 48 21.5 48 48v48H0V112C0 85.5 21.5 64 48 64H96V32c0-17.7 14.3-32 32-32zM0 192H448V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V192zM305 305c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-47 47-47-47c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l47 47-47 47c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l47-47 47 47c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-47-47 47-47z"],"child-reaching":[384,512,[],"e59d","M256 64A64 64 0 1 0 128 64a64 64 0 1 0 128 0zM152.9 169.3c-23.7-8.4-44.5-24.3-58.8-45.8L74.6 94.2C64.8 79.5 45 75.6 30.2 85.4s-18.7 29.7-8.9 44.4L40.9 159c18.1 27.1 42.8 48.4 71.1 62.4V480c0 17.7 14.3 32 32 32s32-14.3 32-32V384h32v96c0 17.7 14.3 32 32 32s32-14.3 32-32V221.6c29.1-14.2 54.4-36.2 72.7-64.2l18.2-27.9c9.6-14.8 5.4-34.6-9.4-44.3s-34.6-5.5-44.3 9.4L291 122.4c-21.8 33.4-58.9 53.6-98.8 53.6c-12.6 0-24.9-2-36.6-5.8c-.9-.3-1.8-.7-2.7-.9z"],"head-side-virus":[512,512,[],"e064","M0 224.2C0 100.6 100.2 0 224 0h24c95.2 0 181.2 69.3 197.3 160.2c2.3 13 6.8 25.7 15.1 36l42 52.6c6.2 7.8 9.6 17.4 9.6 27.4c0 24.2-19.6 43.8-43.8 43.8H448v64c0 35.3-28.7 64-64 64H320v32c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V407.3c0-16.7-6.9-32.5-17.1-45.8C16.6 322.4 0 274.1 0 224.2zM224 64c-8.8 0-16 7.2-16 16c0 33-39.9 49.5-63.2 26.2c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6C145.5 152.1 129 192 96 192c-8.8 0-16 7.2-16 16s7.2 16 16 16c33 0 49.5 39.9 26.2 63.2c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0C168.1 286.5 208 303 208 336c0 8.8 7.2 16 16 16s16-7.2 16-16c0-33 39.9-49.5 63.2-26.2c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6C302.5 263.9 319 224 352 224c8.8 0 16-7.2 16-16s-7.2-16-16-16c-33 0-49.5-39.9-26.2-63.2c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0C279.9 129.5 240 113 240 80c0-8.8-7.2-16-16-16zm-24 96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm40 80a16 16 0 1 1 32 0 16 16 0 1 1 -32 0z"],"user-gear":[640,512,["user-cog"],"f4fe","M224 0a128 128 0 1 1 0 256A128 128 0 1 1 224 0zM178.3 304h91.4c11.8 0 23.4 1.2 34.5 3.3c-2.1 18.5 7.4 35.6 21.8 44.8c-16.6 10.6-26.7 31.6-20 53.3c4 12.9 9.4 25.5 16.4 37.6s15.2 23.1 24.4 33c15.7 16.9 39.6 18.4 57.2 8.7v.9c0 9.2 2.7 18.5 7.9 26.3H29.7C13.3 512 0 498.7 0 482.3C0 383.8 79.8 304 178.3 304zM436 218.2c0-7 4.5-13.3 11.3-14.8c10.5-2.4 21.5-3.7 32.7-3.7s22.2 1.3 32.7 3.7c6.8 1.5 11.3 7.8 11.3 14.8v17.7c0 7.8 4.8 14.8 11.6 18.7c6.8 3.9 15.1 4.5 21.8 .6l13.8-7.9c6.1-3.5 13.7-2.7 18.5 2.4c7.6 8.1 14.3 17.2 20.1 27.2s10.3 20.4 13.5 31c2.1 6.7-1.1 13.7-7.2 17.2l-14.4 8.3c-6.5 3.7-10 10.9-10 18.4s3.5 14.7 10 18.4l14.4 8.3c6.1 3.5 9.2 10.5 7.2 17.2c-3.3 10.6-7.8 21-13.5 31s-12.5 19.1-20.1 27.2c-4.8 5.1-12.5 5.9-18.5 2.4l-13.8-7.9c-6.7-3.9-15.1-3.3-21.8 .6c-6.8 3.9-11.6 10.9-11.6 18.7v17.7c0 7-4.5 13.3-11.3 14.8c-10.5 2.4-21.5 3.7-32.7 3.7s-22.2-1.3-32.7-3.7c-6.8-1.5-11.3-7.8-11.3-14.8V467.8c0-7.9-4.9-14.9-11.7-18.9c-6.8-3.9-15.2-4.5-22-.6l-13.5 7.8c-6.1 3.5-13.7 2.7-18.5-2.4c-7.6-8.1-14.3-17.2-20.1-27.2s-10.3-20.4-13.5-31c-2.1-6.7 1.1-13.7 7.2-17.2l14-8.1c6.5-3.8 10.1-11.1 10.1-18.6s-3.5-14.8-10.1-18.6l-14-8.1c-6.1-3.5-9.2-10.5-7.2-17.2c3.3-10.6 7.7-21 13.5-31s12.5-19.1 20.1-27.2c4.8-5.1 12.4-5.9 18.5-2.4l13.6 7.8c6.8 3.9 15.2 3.3 22-.6c6.9-3.9 11.7-11 11.7-18.9V218.2zm92.1 133.5a48.1 48.1 0 1 0 -96.1 0 48.1 48.1 0 1 0 96.1 0z"],"arrow-up-1-9":[576,512,["sort-numeric-up"],"f163","M450.7 38c8.3 6 13.3 15.7 13.3 26v96h16c17.7 0 32 14.3 32 32s-14.3 32-32 32H432 384c-17.7 0-32-14.3-32-32s14.3-32 32-32h16V108.4l-5.9 2c-16.8 5.6-34.9-3.5-40.5-20.2s3.5-34.9 20.2-40.5l48-16c9.8-3.3 20.5-1.6 28.8 4.4zM160 32c9 0 17.5 3.8 23.6 10.4l88 96c11.9 13 11.1 33.3-2 45.2s-33.3 11.1-45.2-2L192 146.3V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V146.3L95.6 181.6c-11.9 13-32.2 13.9-45.2 2s-13.9-32.2-2-45.2l88-96C142.5 35.8 151 32 160 32zM445.7 364.9A32 32 0 1 0 418.3 307a32 32 0 1 0 27.4 57.9zm-40.7 54.9C369.6 408.4 344 375.2 344 336c0-48.6 39.4-88 88-88s88 39.4 88 88c0 23.5-7.5 46.3-21.5 65.2L449.7 467c-10.5 14.2-30.6 17.2-44.8 6.7s-17.2-30.6-6.7-44.8l6.8-9.2z"],"door-closed":[576,512,[128682],"f52a","M96 64c0-35.3 28.7-64 64-64H416c35.3 0 64 28.7 64 64V448h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H432 144 32c-17.7 0-32-14.3-32-32s14.3-32 32-32H96V64zM384 288a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"shield-virus":[512,512,[],"e06c","M269.4 2.9C265.2 1 260.7 0 256 0s-9.2 1-13.4 2.9L54.3 82.8c-22 9.3-38.4 31-38.3 57.2c.5 99.2 41.3 280.7 213.6 363.2c16.7 8 36.1 8 52.8 0C454.7 420.7 495.5 239.2 496 140c.1-26.2-16.3-47.9-38.3-57.2L269.4 2.9zM256 112c8.8 0 16 7.2 16 16c0 33 39.9 49.5 63.2 26.2c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6C334.5 200.1 351 240 384 240c8.8 0 16 7.2 16 16s-7.2 16-16 16c-33 0-49.5 39.9-26.2 63.2c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0C311.9 334.5 272 351 272 384c0 8.8-7.2 16-16 16s-16-7.2-16-16c0-33-39.9-49.5-63.2-26.2c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6C177.5 311.9 161 272 128 272c-8.8 0-16-7.2-16-16s7.2-16 16-16c33 0 49.5-39.9 26.2-63.2c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0C200.1 177.5 240 161 240 128c0-8.8 7.2-16 16-16zM232 256a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm72 32a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z"],"dice-six":[448,512,[9861],"f526","M0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm160 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM128 288a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm32 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM320 192a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm32 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM320 384a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"mosquito-net":[640,512,[],"e52c","M168.8 462.3c-7.9-4-11.1-13.6-7.2-21.5L192 380.2l0-44.2c0-4.2 1.7-8.3 4.7-11.3L256 265.4V242.2L139.2 344C87.8 395.3 0 358.9 0 286.3c0-41.1 30.6-75.8 71.4-80.9l159.9-23.9-49.6-41.3c-5.1-4.2-7-11.1-4.9-17.4l13.9-41.7-29-58.1c-4-7.9-.7-17.5 7.2-21.5s17.5-.7 21.5 7.2l32 64c1.9 3.8 2.2 8.2 .9 12.2l-12.5 37.6L256 160.5V137.9c0-14.9 10.1-27.3 23.8-31V63.7c0-4.5 3.7-8.2 8.2-8.2s8.2 3.7 8.2 8.2V107c13.7 3.6 23.8 16.1 23.8 31v22.6l45.4-37.8L352.8 85.1c-1.3-4-1-8.4 .9-12.2l32-64c4-7.9 13.6-11.1 21.5-7.2s11.1 13.6 7.2 21.5l-29 58.1 13.9 41.7c2.1 6.2 .1 13.1-4.9 17.4l-49.6 41.3 159.9 23.9c22.5 2.8 41.8 14.6 54.7 31.4c-2.7 2.6-5.2 5.4-7.3 8.6c-8.6-12.9-23.3-21.5-40-21.5s-31.4 8.5-40 21.5c-8.6-12.9-23.3-21.5-40-21.5c-21.7 0-40 14.3-45.9 34.1c-10.7 3.2-19.8 10.1-25.9 19.2l-40.2-35v23.1l32.4 32.4c-.3 2-.4 4.1-.4 6.2c0 16.7 8.5 31.4 21.5 40c-4 2.6-7.5 5.9-10.6 9.5L320 310.6v50c0 17.7-14.3 32-32 32s-32-14.3-32-32v-50l-32 32 0 41.4c0 2.5-.6 4.9-1.7 7.2l-32 64c-4 7.9-13.6 11.1-21.5 7.2zM512 256c8.8 0 16 7.2 16 16v16h48V272c0-8.8 7.2-16 16-16s16 7.2 16 16v16h16c8.8 0 16 7.2 16 16s-7.2 16-16 16H608v48h16c8.8 0 16 7.2 16 16s-7.2 16-16 16H608v48h16c8.8 0 16 7.2 16 16s-7.2 16-16 16H608v16c0 8.8-7.2 16-16 16s-16-7.2-16-16V480H528v16c0 8.8-7.2 16-16 16s-16-7.2-16-16V480H448v16c0 8.8-7.2 16-16 16s-16-7.2-16-16V480H400c-8.8 0-16-7.2-16-16s7.2-16 16-16h16V400H400c-8.8 0-16-7.2-16-16s7.2-16 16-16h16V320H400c-8.8 0-16-7.2-16-16s7.2-16 16-16h16V272c0-8.8 7.2-16 16-16s16 7.2 16 16v16h48V272c0-8.8 7.2-16 16-16zm16 112h48V320H528v48zm0 80h48V400H528v48zM448 320v48h48V320H448zm0 80v48h48V400H448z"],"bridge-water":[576,512,[],"e4ce","M0 96C0 78.3 14.3 64 32 64H544c17.7 0 32 14.3 32 32v35.6c0 15.7-12.7 28.4-28.4 28.4c-37.3 0-67.6 30.2-67.6 67.6V352.5c-12.9 0-25.8 3.9-36.8 11.7c-18 12.4-40.1 20.3-59.2 20.3h0l0-.5V256c0-53-43-96-96-96s-96 43-96 96V384l0 .5c-19 0-41.2-7.9-59.1-20.3c-11.1-7.8-24-11.7-36.9-11.7V227.6C96 190.2 65.8 160 28.4 160C12.7 160 0 147.3 0 131.6V96zM306.5 389.9C329 405.4 356.5 416 384 416c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 469.7 417 480 384 480c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 405.2 165.1 416 192 416c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z"],"person-booth":[576,512,[],"f756","M256 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V192h64V32zm320 0c0-17.7-14.3-32-32-32s-32 14.3-32 32V480c0 17.7 14.3 32 32 32s32-14.3 32-32V32zM224 512c17.7 0 32-14.3 32-32V320H192V480c0 17.7 14.3 32 32 32zM320 0c-9.3 0-18.1 4-24.2 11s-8.8 16.3-7.5 25.5l31.2 218.6L288.6 409.7c-3.5 17.3 7.8 34.2 25.1 37.7s34.2-7.8 37.7-25.1l.7-3.6c1.3 16.4 15.1 29.4 31.9 29.4c17.7 0 32-14.3 32-32c0 17.7 14.3 32 32 32s32-14.3 32-32V32c0-17.7-14.3-32-32-32H320zM112 80A48 48 0 1 0 16 80a48 48 0 1 0 96 0zm0 261.3V269.3l4.7 4.7c9 9 21.2 14.1 33.9 14.1H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H157.3l-41.6-41.6c-14.3-14.3-33.8-22.4-54-22.4C27.6 160 0 187.6 0 221.6v55.7l0 .9V480c0 17.7 14.3 32 32 32s32-14.3 32-32V384l32 42.7V480c0 17.7 14.3 32 32 32s32-14.3 32-32V421.3c0-10.4-3.4-20.5-9.6-28.8L112 341.3z"],"text-width":[448,512,[],"f035","M64 128V96H192l0 128H176c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H256l0-128H384v32c0 17.7 14.3 32 32 32s32-14.3 32-32V80c0-26.5-21.5-48-48-48H224 48C21.5 32 0 53.5 0 80v48c0 17.7 14.3 32 32 32s32-14.3 32-32zM9.4 361.4c-12.5 12.5-12.5 32.8 0 45.3l64 64c9.2 9.2 22.9 11.9 34.9 6.9s19.8-16.6 19.8-29.6V416H320v32c0 12.9 7.8 24.6 19.8 29.6s25.7 2.2 34.9-6.9l64-64c12.5-12.5 12.5-32.8 0-45.3l-64-64c-9.2-9.2-22.9-11.9-34.9-6.9s-19.8 16.6-19.8 29.6v32H128V320c0-12.9-7.8-24.6-19.8-29.6s-25.7-2.2-34.9 6.9l-64 64z"],"hat-wizard":[512,512,[],"f6e8","M64 416L168.6 180.7c15.3-34.4 40.3-63.5 72-83.7l146.9-94c3-1.9 6.5-2.9 10-2.9C407.7 0 416 8.3 416 18.6v1.6c0 2.6-.5 5.1-1.4 7.5L354.8 176.9c-1.9 4.7-2.8 9.7-2.8 14.7c0 5.5 1.2 11 3.4 16.1L448 416H240.9l11.8-35.4 40.4-13.5c6.5-2.2 10.9-8.3 10.9-15.2s-4.4-13-10.9-15.2l-40.4-13.5-13.5-40.4C237 276.4 230.9 272 224 272s-13 4.4-15.2 10.9l-13.5 40.4-40.4 13.5C148.4 339 144 345.1 144 352s4.4 13 10.9 15.2l40.4 13.5L207.1 416H64zM279.6 141.5c-1.1-3.3-4.1-5.5-7.6-5.5s-6.5 2.2-7.6 5.5l-6.7 20.2-20.2 6.7c-3.3 1.1-5.5 4.1-5.5 7.6s2.2 6.5 5.5 7.6l20.2 6.7 6.7 20.2c1.1 3.3 4.1 5.5 7.6 5.5s6.5-2.2 7.6-5.5l6.7-20.2 20.2-6.7c3.3-1.1 5.5-4.1 5.5-7.6s-2.2-6.5-5.5-7.6l-20.2-6.7-6.7-20.2zM32 448H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],"pen-fancy":[512,512,[128395,10002],"f5ac","M373.5 27.1C388.5 9.9 410.2 0 433 0c43.6 0 79 35.4 79 79c0 22.8-9.9 44.6-27.1 59.6L277.7 319l-10.3-10.3-64-64L193 234.3 373.5 27.1zM170.3 256.9l10.4 10.4 64 64 10.4 10.4-19.2 83.4c-3.9 17.1-16.9 30.7-33.8 35.4L24.4 510.3l95.4-95.4c2.6 .7 5.4 1.1 8.3 1.1c17.7 0 32-14.3 32-32s-14.3-32-32-32s-32 14.3-32 32c0 2.9 .4 5.6 1.1 8.3L1.7 487.6 51.5 310c4.7-16.9 18.3-29.9 35.4-33.8l83.4-19.2z"],"person-digging":[576,512,["digging"],"f85e","M208 64a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM9.8 214.8c5.1-12.2 19.1-18 31.4-12.9L60.7 210l22.9-38.1C99.9 144.6 129.3 128 161 128c51.4 0 97 32.9 113.3 81.7l34.6 103.7 79.3 33.1 34.2-45.6c6.4-8.5 16.6-13.3 27.2-12.8s20.3 6.4 25.8 15.5l96 160c5.9 9.9 6.1 22.2 .4 32.2s-16.3 16.2-27.8 16.2H288c-11.1 0-21.4-5.7-27.2-15.2s-6.4-21.2-1.4-31.1l16-32c5.4-10.8 16.5-17.7 28.6-17.7h32l22.5-30L22.8 246.2c-12.2-5.1-18-19.1-12.9-31.4zm82.8 91.8l112 48c11.8 5 19.4 16.6 19.4 29.4v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V405.1l-60.6-26-37 111c-5.6 16.8-23.7 25.8-40.5 20.2S-3.9 486.6 1.6 469.9l48-144 11-33 32 13.7z"],trash:[448,512,[],"f1f8","M135.2 17.7L128 32H32C14.3 32 0 46.3 0 64S14.3 96 32 96H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H320l-7.2-14.3C307.4 6.8 296.3 0 284.2 0H163.8c-12.1 0-23.2 6.8-28.6 17.7zM416 128H32L53.2 467c1.6 25.3 22.6 45 47.9 45H346.9c25.3 0 46.3-19.7 47.9-45L416 128z"],"gauge-simple":[512,512,["gauge-simple-med","tachometer-average"],"f629","M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm320 96c0-26.9-16.5-49.9-40-59.3V88c0-13.3-10.7-24-24-24s-24 10.7-24 24V292.7c-23.5 9.5-40 32.5-40 59.3c0 35.3 28.7 64 64 64s64-28.7 64-64z"],"book-medical":[448,512,[],"f7e6","M0 96C0 43 43 0 96 0H384h32c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384 96c-53 0-96-43-96-96V96zM64 416c0 17.7 14.3 32 32 32H352V384H96c-17.7 0-32 14.3-32 32zM208 112v48H160c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V224h48c8.8 0 16-7.2 16-16V176c0-8.8-7.2-16-16-16H272V112c0-8.8-7.2-16-16-16H224c-8.8 0-16 7.2-16 16z"],poo:[512,512,[128169],"f2fe","M268.9 .9c-5.5-.7-11 1.4-14.5 5.7s-4.6 10.1-2.8 15.4c2.8 8.2 4.3 16.9 4.3 26.1c0 44.1-35.7 79.9-79.8 80H160c-35.3 0-64 28.7-64 64c0 19.1 8.4 36.3 21.7 48H104c-39.8 0-72 32.2-72 72c0 23.2 11 43.8 28 57c-34.1 5.7-60 35.3-60 71c0 39.8 32.2 72 72 72H440c39.8 0 72-32.2 72-72c0-35.7-25.9-65.3-60-71c17-13.2 28-33.8 28-57c0-39.8-32.2-72-72-72H394.3c13.3-11.7 21.7-28.9 21.7-48c0-35.3-28.7-64-64-64h-5.5c3.5-10 5.5-20.8 5.5-32c0-48.6-36.2-88.8-83.1-95.1zM192 256a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm96 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm64 108.3c0 2.4-.7 4.8-2.2 6.7c-8.2 10.5-39.5 45-93.8 45s-85.6-34.6-93.8-45c-1.5-1.9-2.2-4.3-2.2-6.7c0-6.8 5.5-12.3 12.3-12.3H339.7c6.8 0 12.3 5.5 12.3 12.3z"],"quote-right":[448,512,[8221,"quote-right-alt"],"f10e","M448 296c0 66.3-53.7 120-120 120h-8c-17.7 0-32-14.3-32-32s14.3-32 32-32h8c30.9 0 56-25.1 56-56v-8H320c-35.3 0-64-28.7-64-64V160c0-35.3 28.7-64 64-64h64c35.3 0 64 28.7 64 64v32 32 72zm-256 0c0 66.3-53.7 120-120 120H64c-17.7 0-32-14.3-32-32s14.3-32 32-32h8c30.9 0 56-25.1 56-56v-8H64c-35.3 0-64-28.7-64-64V160c0-35.3 28.7-64 64-64h64c35.3 0 64 28.7 64 64v32 32 72z"],shirt:[640,512,[128085,"t-shirt","tshirt"],"f553","M211.8 0c7.8 0 14.3 5.7 16.7 13.2C240.8 51.9 277.1 80 320 80s79.2-28.1 91.5-66.8C413.9 5.7 420.4 0 428.2 0h12.6c22.5 0 44.2 7.9 61.5 22.3L628.5 127.4c6.6 5.5 10.7 13.5 11.4 22.1s-2.1 17.1-7.8 23.6l-56 64c-11.4 13.1-31.2 14.6-44.6 3.5L480 197.7V448c0 35.3-28.7 64-64 64H224c-35.3 0-64-28.7-64-64V197.7l-51.5 42.9c-13.3 11.1-33.1 9.6-44.6-3.5l-56-64c-5.7-6.5-8.5-15-7.8-23.6s4.8-16.6 11.4-22.1L137.7 22.3C155 7.9 176.7 0 199.2 0h12.6z"],cubes:[576,512,[],"f1b3","M290.8 48.6l78.4 29.7L288 109.5 206.8 78.3l78.4-29.7c1.8-.7 3.8-.7 5.7 0zM136 92.5V204.7c-1.3 .4-2.6 .8-3.9 1.3l-96 36.4C14.4 250.6 0 271.5 0 294.7V413.9c0 22.2 13.1 42.3 33.5 51.3l96 42.2c14.4 6.3 30.7 6.3 45.1 0L288 457.5l113.5 49.9c14.4 6.3 30.7 6.3 45.1 0l96-42.2c20.3-8.9 33.5-29.1 33.5-51.3V294.7c0-23.3-14.4-44.1-36.1-52.4l-96-36.4c-1.3-.5-2.6-.9-3.9-1.3V92.5c0-23.3-14.4-44.1-36.1-52.4l-96-36.4c-12.8-4.8-26.9-4.8-39.7 0l-96 36.4C150.4 48.4 136 69.3 136 92.5zM392 210.6l-82.4 31.2V152.6L392 121v89.6zM154.8 250.9l78.4 29.7L152 311.7 70.8 280.6l78.4-29.7c1.8-.7 3.8-.7 5.7 0zm18.8 204.4V354.8L256 323.2v95.9l-82.4 36.2zM421.2 250.9c1.8-.7 3.8-.7 5.7 0l78.4 29.7L424 311.7l-81.2-31.1 78.4-29.7zM523.2 421.2l-77.6 34.1V354.8L528 323.2v90.7c0 3.2-1.9 6-4.8 7.3z"],divide:[448,512,[10135,247],"f529","M272 96a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zm0 320a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM400 288c17.7 0 32-14.3 32-32s-14.3-32-32-32H48c-17.7 0-32 14.3-32 32s14.3 32 32 32H400z"],"tenge-sign":[384,512,[8376,"tenge"],"f7d7","M0 64C0 46.3 14.3 32 32 32H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 96 0 81.7 0 64zM0 192c0-17.7 14.3-32 32-32H192 352c17.7 0 32 14.3 32 32s-14.3 32-32 32H224V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V224H32c-17.7 0-32-14.3-32-32z"],headphones:[512,512,[127911],"f025","M256 80C149.9 80 62.4 159.4 49.6 262c9.4-3.8 19.6-6 30.4-6c26.5 0 48 21.5 48 48V432c0 26.5-21.5 48-48 48c-44.2 0-80-35.8-80-80V384 336 288C0 146.6 114.6 32 256 32s256 114.6 256 256v48 48 16c0 44.2-35.8 80-80 80c-26.5 0-48-21.5-48-48V304c0-26.5 21.5-48 48-48c10.8 0 21 2.1 30.4 6C449.6 159.4 362.1 80 256 80z"],"hands-holding":[640,512,[],"f4c2","M80 104c0-22.1-17.9-40-40-40S0 81.9 0 104v56 64V325.5c0 25.5 10.1 49.9 28.1 67.9L128 493.3c12 12 28.3 18.7 45.3 18.7H240c26.5 0 48-21.5 48-48V385.1c0-29.7-11.8-58.2-32.8-79.2l-25.3-25.3 0 0-15.2-15.2-32-32c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l32 32 15.2 15.2c11 11 9.2 29.2-3.7 37.8c-9.7 6.5-22.7 5.2-31-3.1L98.7 309.5c-12-12-18.7-28.3-18.7-45.3V224 144 104zm480 0v40 80 40.2c0 17-6.7 33.3-18.7 45.3l-51.1 51.1c-8.3 8.3-21.3 9.6-31 3.1c-12.9-8.6-14.7-26.9-3.7-37.8l15.2-15.2 32-32c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-32 32-15.2 15.2 0 0-25.3 25.3c-21 21-32.8 49.5-32.8 79.2V464c0 26.5 21.5 48 48 48h66.7c17 0 33.3-6.7 45.3-18.7l99.9-99.9c18-18 28.1-42.4 28.1-67.9V224 160 104c0-22.1-17.9-40-40-40s-40 17.9-40 40z"],"hands-clapping":[512,512,[],"e1a8","M336 16V80c0 8.8-7.2 16-16 16s-16-7.2-16-16V16c0-8.8 7.2-16 16-16s16 7.2 16 16zm-98.7 7.1l32 48c4.9 7.4 2.9 17.3-4.4 22.2s-17.3 2.9-22.2-4.4l-32-48c-4.9-7.4-2.9-17.3 4.4-22.2s17.3-2.9 22.2 4.4zM135 119c9.4-9.4 24.6-9.4 33.9 0L292.7 242.7c10.1 10.1 27.3 2.9 27.3-11.3V192c0-17.7 14.3-32 32-32s32 14.3 32 32V345.6c0 57.1-30 110-78.9 139.4c-64 38.4-145.8 28.3-198.5-24.4L7 361c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l53 53c6.1 6.1 16 6.1 22.1 0s6.1-16 0-22.1L23 265c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l93 93c6.1 6.1 16 6.1 22.1 0s6.1-16 0-22.1L55 185c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l117 117c6.1 6.1 16 6.1 22.1 0s6.1-16 0-22.1l-93-93c-9.4-9.4-9.4-24.6 0-33.9zM433.1 484.9c-24.2 14.5-50.9 22.1-77.7 23.1c48.1-39.6 76.6-99 76.6-162.4l0-98.1c8.2-.1 16-6.4 16-16V192c0-17.7 14.3-32 32-32s32 14.3 32 32V345.6c0 57.1-30 110-78.9 139.4zM424.9 18.7c7.4 4.9 9.3 14.8 4.4 22.2l-32 48c-4.9 7.4-14.8 9.3-22.2 4.4s-9.3-14.8-4.4-22.2l32-48c4.9-7.4 14.8-9.3 22.2-4.4z"],republican:[640,512,[],"f75e","M0 192C0 103.6 71.6 32 160 32H384c88.4 0 160 71.6 160 160v64H0V192zm415.9-64c-2.4 0-4.7 1.3-5.7 3.4l-12.6 24.6-28.2 4c-2.4 .3-4.4 2-5.2 4.2s-.1 4.7 1.6 6.3l20.4 19.2-4.8 27.1c-.4 2.3 .6 4.7 2.5 6s4.6 1.6 6.7 .5l25.2-12.8 25.2 12.8c2.2 1.1 4.8 .9 6.7-.5s3-3.7 2.5-6l-4.8-27.1L466 170.5c1.7-1.6 2.4-4.1 1.6-6.3s-2.8-3.9-5.2-4.2l-28.2-4-12.6-24.6c-1.1-2.1-3.3-3.4-5.7-3.4zm-138.3 3.4c-1.1-2.1-3.3-3.4-5.7-3.4s-4.7 1.3-5.7 3.4l-12.6 24.6-28.2 4c-2.4 .3-4.4 2-5.2 4.2s-.1 4.7 1.6 6.3l20.4 19.2-4.8 27.1c-.4 2.3 .6 4.7 2.5 6s4.6 1.6 6.7 .5l25.2-12.8 25.2 12.8c2.2 1.1 4.8 .9 6.7-.5s3-3.7 2.5-6l-4.8-27.1L322 170.5c1.7-1.6 2.4-4.1 1.6-6.3s-2.8-3.9-5.2-4.2l-28.2-4-12.6-24.6zM127.9 128c-2.4 0-4.7 1.3-5.7 3.4l-12.6 24.6-28.2 4c-2.4 .3-4.4 2-5.2 4.2s-.1 4.7 1.6 6.3l20.4 19.2-4.8 27.1c-.4 2.3 .6 4.7 2.5 6s4.6 1.6 6.7 .5l25.2-12.8 25.2 12.8c2.2 1.1 4.8 .9 6.7-.5s3-3.7 2.5-6l-4.8-27.1L178 170.5c1.7-1.6 2.4-4.1 1.6-6.3s-2.8-3.9-5.2-4.2l-28.2-4-12.6-24.6c-1.1-2.1-3.3-3.4-5.7-3.4zm.1 160H320h96 32 64 32v32 80c0 8.8 7.2 16 16 16s16-7.2 16-16V352c0-17.7 14.3-32 32-32s32 14.3 32 32v48c0 44.2-35.8 80-80 80s-80-35.8-80-80V352H448v32 64c0 17.7-14.3 32-32 32H352c-17.7 0-32-14.3-32-32V384H128v64c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32V384 288H128z"],"arrow-left":[448,512,[8592],"f060","M9.4 233.4c-12.5 12.5-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L109.2 288 416 288c17.7 0 32-14.3 32-32s-14.3-32-32-32l-306.7 0L214.6 118.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-160 160z"],"person-circle-xmark":[576,512,[],"e543","M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l44.9 74.7c-16.1 17.6-28.6 38.5-36.6 61.5c-1.9-1.8-3.5-3.9-4.9-6.3L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152zM432 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm59.3 107.3c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0L432 345.4l-36.7-36.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6L409.4 368l-36.7 36.7c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L432 390.6l36.7 36.7c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6L454.6 368l36.7-36.7z"],ruler:[512,512,[128207],"f545","M177.9 494.1c-18.7 18.7-49.1 18.7-67.9 0L17.9 401.9c-18.7-18.7-18.7-49.1 0-67.9l50.7-50.7 48 48c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-48-48 41.4-41.4 48 48c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-48-48 41.4-41.4 48 48c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-48-48 41.4-41.4 48 48c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-48-48 50.7-50.7c18.7-18.7 49.1-18.7 67.9 0l92.1 92.1c18.7 18.7 18.7 49.1 0 67.9L177.9 494.1z"],"align-left":[448,512,[],"f036","M288 64c0 17.7-14.3 32-32 32H32C14.3 96 0 81.7 0 64S14.3 32 32 32H256c17.7 0 32 14.3 32 32zm0 256c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H256c17.7 0 32 14.3 32 32zM0 192c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32zM448 448c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H416c17.7 0 32 14.3 32 32z"],"dice-d6":[448,512,[],"f6d1","M201 10.3c14.3-7.8 31.6-7.8 46 0L422.3 106c5.1 2.8 8.3 8.2 8.3 14s-3.2 11.2-8.3 14L231.7 238c-4.8 2.6-10.5 2.6-15.3 0L25.7 134c-5.1-2.8-8.3-8.2-8.3-14s3.2-11.2 8.3-14L201 10.3zM23.7 170l176 96c5.1 2.8 8.3 8.2 8.3 14V496c0 5.6-3 10.9-7.8 13.8s-10.9 3-15.8 .3L25 423.1C9.6 414.7 0 398.6 0 381V184c0-5.6 3-10.9 7.8-13.8s10.9-3 15.8-.3zm400.7 0c5-2.7 11-2.6 15.8 .3s7.8 8.1 7.8 13.8V381c0 17.6-9.6 33.7-25 42.1L263.7 510c-5 2.7-11 2.6-15.8-.3s-7.8-8.1-7.8-13.8V280c0-5.9 3.2-11.2 8.3-14l176-96z"],restroom:[640,512,[],"f7bd","M80 48a48 48 0 1 1 96 0A48 48 0 1 1 80 48zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V325.2c-8.1 9.2-21.1 13.2-33.5 9.4c-16.9-5.3-26.3-23.2-21-40.1l30.9-99.1C44.9 155.3 82 128 124 128h8c42 0 79.1 27.3 91.6 67.4l30.9 99.1c5.3 16.9-4.1 34.8-21 40.1c-12.4 3.9-25.4-.2-33.5-9.4V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H120zM320 0c13.3 0 24 10.7 24 24V488c0 13.3-10.7 24-24 24s-24-10.7-24-24V24c0-13.3 10.7-24 24-24zM464 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM440 480V384H422.2c-10.9 0-18.6-10.7-15.2-21.1l9-26.9c-3.2 0-6.4-.5-9.5-1.5c-16.9-5.3-26.3-23.2-21-40.1l29.7-95.2C428.4 156.9 467.6 128 512 128s83.6 28.9 96.8 71.2l29.7 95.2c5.3 16.9-4.1 34.8-21 40.1c-3.2 1-6.4 1.5-9.5 1.5l9 26.9c3.5 10.4-4.3 21.1-15.2 21.1H584v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384H504v96c0 17.7-14.3 32-32 32s-32-14.3-32-32z"],j:[320,512,[106],"4a","M288 32c17.7 0 32 14.3 32 32V320c0 88.4-71.6 160-160 160S0 408.4 0 320V288c0-17.7 14.3-32 32-32s32 14.3 32 32v32c0 53 43 96 96 96s96-43 96-96V64c0-17.7 14.3-32 32-32z"],"users-viewfinder":[640,512,[],"e595","M48 48h88c13.3 0 24-10.7 24-24s-10.7-24-24-24H32C14.3 0 0 14.3 0 32V136c0 13.3 10.7 24 24 24s24-10.7 24-24V48zM175.8 224a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-26.5 32C119.9 256 96 279.9 96 309.3c0 14.7 11.9 26.7 26.7 26.7h56.1c8-34.1 32.8-61.7 65.2-73.6c-7.5-4.1-16.2-6.4-25.3-6.4H149.3zm368 80c14.7 0 26.7-11.9 26.7-26.7c0-29.5-23.9-53.3-53.3-53.3H421.3c-9.2 0-17.8 2.3-25.3 6.4c32.4 11.9 57.2 39.5 65.2 73.6h56.1zm-89.4 0c-8.6-24.3-29.9-42.6-55.9-47c-3.9-.7-7.9-1-12-1H280c-4.1 0-8.1 .3-12 1c-26 4.4-47.3 22.7-55.9 47c-2.7 7.5-4.1 15.6-4.1 24c0 13.3 10.7 24 24 24H408c13.3 0 24-10.7 24-24c0-8.4-1.4-16.5-4.1-24zM464 224a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-80-32a64 64 0 1 0 -128 0 64 64 0 1 0 128 0zM504 48h88v88c0 13.3 10.7 24 24 24s24-10.7 24-24V32c0-17.7-14.3-32-32-32H504c-13.3 0-24 10.7-24 24s10.7 24 24 24zM48 464V376c0-13.3-10.7-24-24-24s-24 10.7-24 24V480c0 17.7 14.3 32 32 32H136c13.3 0 24-10.7 24-24s-10.7-24-24-24H48zm456 0c-13.3 0-24 10.7-24 24s10.7 24 24 24H608c17.7 0 32-14.3 32-32V376c0-13.3-10.7-24-24-24s-24 10.7-24 24v88H504z"],"file-video":[384,512,[],"f1c8","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM64 288c0-17.7 14.3-32 32-32h96c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V288zM300.9 397.9L256 368V304l44.9-29.9c2-1.3 4.4-2.1 6.8-2.1c6.8 0 12.3 5.5 12.3 12.3V387.7c0 6.8-5.5 12.3-12.3 12.3c-2.4 0-4.8-.7-6.8-2.1z"],"up-right-from-square":[512,512,["external-link-alt"],"f35d","M352 0c-12.9 0-24.6 7.8-29.6 19.8s-2.2 25.7 6.9 34.9L370.7 96 201.4 265.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L416 141.3l41.4 41.4c9.2 9.2 22.9 11.9 34.9 6.9s19.8-16.6 19.8-29.6V32c0-17.7-14.3-32-32-32H352zM80 32C35.8 32 0 67.8 0 112V432c0 44.2 35.8 80 80 80H400c44.2 0 80-35.8 80-80V320c0-17.7-14.3-32-32-32s-32 14.3-32 32V432c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16H192c17.7 0 32-14.3 32-32s-14.3-32-32-32H80z"],"table-cells":[512,512,["th"],"f00a","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm88 64v64H64V96h88zm56 0h88v64H208V96zm240 0v64H360V96h88zM64 224h88v64H64V224zm232 0v64H208V224h88zm64 0h88v64H360V224zM152 352v64H64V352h88zm56 0h88v64H208V352zm240 0v64H360V352h88z"],"file-pdf":[512,512,[],"f1c1","M0 64C0 28.7 28.7 0 64 0L224 0l0 128c0 17.7 14.3 32 32 32l128 0 0 144-208 0c-35.3 0-64 28.7-64 64l0 144-48 0c-35.3 0-64-28.7-64-64L0 64zm384 64l-128 0L256 0 384 128zM176 352l32 0c30.9 0 56 25.1 56 56s-25.1 56-56 56l-16 0 0 32c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-48 0-80c0-8.8 7.2-16 16-16zm32 80c13.3 0 24-10.7 24-24s-10.7-24-24-24l-16 0 0 48 16 0zm96-80l32 0c26.5 0 48 21.5 48 48l0 64c0 26.5-21.5 48-48 48l-32 0c-8.8 0-16-7.2-16-16l0-128c0-8.8 7.2-16 16-16zm32 128c8.8 0 16-7.2 16-16l0-64c0-8.8-7.2-16-16-16l-16 0 0 96 16 0zm80-112c0-8.8 7.2-16 16-16l48 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-32 0 0 32 32 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-32 0 0 48c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-64 0-64z"],"book-bible":[448,512,["bible"],"f647","M96 0C43 0 0 43 0 96V416c0 53 43 96 96 96H384h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V384c17.7 0 32-14.3 32-32V32c0-17.7-14.3-32-32-32H384 96zm0 384H352v64H96c-17.7 0-32-14.3-32-32s14.3-32 32-32zM208 80c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272V304c0 8.8-7.2 16-16 16H224c-8.8 0-16-7.2-16-16V192H160c-8.8 0-16-7.2-16-16V144c0-8.8 7.2-16 16-16h48V80z"],o:[448,512,[111],"4f","M224 96a160 160 0 1 0 0 320 160 160 0 1 0 0-320zM448 256A224 224 0 1 1 0 256a224 224 0 1 1 448 0z"],"suitcase-medical":[512,512,["medkit"],"f0fa","M184 48H328c4.4 0 8 3.6 8 8V96H176V56c0-4.4 3.6-8 8-8zm-56 8V96v32V480H384V128 96 56c0-30.9-25.1-56-56-56H184c-30.9 0-56 25.1-56 56zM96 96H64C28.7 96 0 124.7 0 160V416c0 35.3 28.7 64 64 64H96V96zM416 480h32c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H416V480zM224 208c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H288v48c0 8.8-7.2 16-16 16H240c-8.8 0-16-7.2-16-16V320H176c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h48V208z"],"user-secret":[448,512,[128373],"f21b","M224 16c-6.7 0-10.8-2.8-15.5-6.1C201.9 5.4 194 0 176 0c-30.5 0-52 43.7-66 89.4C62.7 98.1 32 112.2 32 128c0 14.3 25 27.1 64.6 35.9c-.4 4-.6 8-.6 12.1c0 17 3.3 33.2 9.3 48H45.4C38 224 32 230 32 237.4c0 1.7 .3 3.4 1 5l38.8 96.9C28.2 371.8 0 423.8 0 482.3C0 498.7 13.3 512 29.7 512H418.3c16.4 0 29.7-13.3 29.7-29.7c0-58.5-28.2-110.4-71.7-143L415 242.4c.6-1.6 1-3.3 1-5c0-7.4-6-13.4-13.4-13.4H342.7c6-14.8 9.3-31 9.3-48c0-4.1-.2-8.1-.6-12.1C391 155.1 416 142.3 416 128c0-15.8-30.7-29.9-78-38.6C324 43.7 302.5 0 272 0c-18 0-25.9 5.4-32.5 9.9c-4.8 3.3-8.8 6.1-15.5 6.1zm56 208H267.6c-16.5 0-31.1-10.6-36.3-26.2c-2.3-7-12.2-7-14.5 0c-5.2 15.6-19.9 26.2-36.3 26.2H168c-22.1 0-40-17.9-40-40V169.6c28.2 4.1 61 6.4 96 6.4s67.8-2.3 96-6.4V184c0 22.1-17.9 40-40 40zm-88 96l16 32L176 480 128 288l64 32zm128-32L272 480 240 352l16-32 64-32z"],otter:[640,512,[129446],"f700","M181.5 197.1l12.9 6.4c5.9 3 12.4 4.5 19.1 4.5c23.5 0 42.6-19.1 42.6-42.6V144c0-35.3-28.7-64-64-64H128c-35.3 0-64 28.7-64 64v21.4c0 23.5 19.1 42.6 42.6 42.6c6.6 0 13.1-1.5 19.1-4.5l12.9-6.4 8.4-4.2L135.1 185c-4.5-3-7.1-8-7.1-13.3V168c0-13.3 10.7-24 24-24h16c13.3 0 24 10.7 24 24v3.7c0 5.3-2.7 10.3-7.1 13.3l-11.8 7.9 8.4 4.2zm-8.6 49.4L160 240l-12.9 6.4c-12.6 6.3-26.5 9.6-40.5 9.6c-3.6 0-7.1-.2-10.6-.6v.6c0 35.3 28.7 64 64 64h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384V336 320c0-23.7 12.9-44.4 32-55.4c9.4-5.4 20.3-8.6 32-8.6V240c0-26.5 21.5-48 48-48c8.8 0 16 7.2 16 16v32 16 48c0 8.8 7.2 16 16 16s16-7.2 16-16V204.3c0-48.2-30.8-91-76.6-106.3l-8.5-2.8c-8-2.7-12.6-11.1-10.4-19.3s10.3-13.2 18.6-11.6l19.9 4C576 86.1 640 164.2 640 254.9l0 1.1h0c0 123.7-100.3 224-224 224h-1.1H256h-.6C132 480 32 380 32 256.6V256 216.8c-10.1-14.6-16-32.3-16-51.4V144l0-1.4C6.7 139.3 0 130.5 0 120c0-13.3 10.7-24 24-24h2.8C44.8 58.2 83.3 32 128 32h64c44.7 0 83.2 26.2 101.2 64H296c13.3 0 24 10.7 24 24c0 10.5-6.7 19.3-16 22.6l0 1.4v21.4c0 1.4 0 2.8-.1 4.3c12-6.2 25.7-9.6 40.1-9.6h8c17.7 0 32 14.3 32 32s-14.3 32-32 32h-8c-13.3 0-24 10.7-24 24v8h56.4c-15.2 17-24.4 39.4-24.4 64H320c-42.3 0-78.2-27.4-91-65.3c-5.1 .9-10.3 1.3-15.6 1.3c-14.1 0-27.9-3.3-40.5-9.6zM96 128a16 16 0 1 1 0 32 16 16 0 1 1 0-32zm112 16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0z"],"person-dress":[320,512,["female"],"f182","M160 0a48 48 0 1 1 0 96 48 48 0 1 1 0-96zM88 384H70.2c-10.9 0-18.6-10.7-15.2-21.1L93.3 248.1 59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l53.6-89.2c20.3-33.7 56.7-54.3 96-54.3h11.6c39.3 0 75.7 20.6 96 54.3l53.6 89.2c9.1 15.1 4.2 34.8-10.9 43.9s-34.8 4.2-43.9-10.9l-33.9-56.3L265 362.9c3.5 10.4-4.3 21.1-15.2 21.1H232v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384H152v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384z"],"comment-dollar":[512,512,[],"f651","M256 448c141.4 0 256-93.1 256-208S397.4 32 256 32S0 125.1 0 240c0 45.1 17.7 86.8 47.7 120.9c-1.9 24.5-11.4 46.3-21.4 62.9c-5.5 9.2-11.1 16.6-15.2 21.6c-2.1 2.5-3.7 4.4-4.9 5.7c-.6 .6-1 1.1-1.3 1.4l-.3 .3 0 0 0 0 0 0 0 0c-4.6 4.6-5.9 11.4-3.4 17.4c2.5 6 8.3 9.9 14.8 9.9c28.7 0 57.6-8.9 81.6-19.3c22.9-10 42.4-21.9 54.3-30.6c31.8 11.5 67 17.9 104.1 17.9zm20-312v13.9c7.5 1.2 14.6 2.9 21.1 4.7c10.7 2.8 17 13.8 14.2 24.5s-13.8 17-24.5 14.2c-11-2.9-21.6-5-31.2-5.2c-7.9-.1-16 1.8-21.5 5c-4.8 2.8-6.2 5.6-6.2 9.3c0 1.8 .1 3.5 5.3 6.7c6.3 3.8 15.5 6.7 28.3 10.5l.7 .2c11.2 3.4 25.6 7.7 37.1 15c12.9 8.1 24.3 21.3 24.6 41.6c.3 20.9-10.5 36.1-24.8 45c-7.2 4.5-15.2 7.3-23.2 9V344c0 11-9 20-20 20s-20-9-20-20V329.4c-10.3-2.2-20-5.5-28.2-8.4l0 0 0 0c-2.1-.7-4.1-1.4-6.1-2.1c-10.5-3.5-16.1-14.8-12.6-25.3s14.8-16.1 25.3-12.6c2.5 .8 4.9 1.7 7.2 2.4c13.6 4.6 24 8.1 35.1 8.5c8.6 .3 16.5-1.6 21.4-4.7c4.1-2.5 6-5.5 5.9-10.5c0-2.9-.8-5-5.9-8.2c-6.3-4-15.4-6.9-28-10.7l-1.7-.5c-10.9-3.3-24.6-7.4-35.6-14c-12.7-7.7-24.6-20.5-24.7-40.7c-.1-21.1 11.8-35.7 25.8-43.9c6.9-4.1 14.5-6.8 22.2-8.5V136c0-11 9-20 20-20s20 9 20 20z"],"business-time":[640,512,["briefcase-clock"],"f64a","M184 48H328c4.4 0 8 3.6 8 8V96H176V56c0-4.4 3.6-8 8-8zm-56 8V96H64C28.7 96 0 124.7 0 160v96H192 352h8.2c32.3-39.1 81.1-64 135.8-64c5.4 0 10.7 .2 16 .7V160c0-35.3-28.7-64-64-64H384V56c0-30.9-25.1-56-56-56H184c-30.9 0-56 25.1-56 56zM320 352H224c-17.7 0-32-14.3-32-32V288H0V416c0 35.3 28.7 64 64 64H360.2C335.1 449.6 320 410.5 320 368c0-5.4 .2-10.7 .7-16l-.7 0zm320 16a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zM496 288c8.8 0 16 7.2 16 16v48h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H496c-8.8 0-16-7.2-16-16V304c0-8.8 7.2-16 16-16z"],"table-cells-large":[512,512,["th-large"],"f009","M448 96V224H288V96H448zm0 192V416H288V288H448zM224 224H64V96H224V224zM64 288H224V416H64V288zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64z"],"book-tanakh":[448,512,["tanakh"],"f827","M352 0c53 0 96 43 96 96V416c0 53-43 96-96 96H64 32c-17.7 0-32-14.3-32-32s14.3-32 32-32V384c-17.7 0-32-14.3-32-32V32C0 14.3 14.3 0 32 0H64 352zm0 384H96v64H352c17.7 0 32-14.3 32-32s-14.3-32-32-32zM138.7 208l13.9 24H124.9l13.9-24zm-13.9-24L97.1 232c-6.2 10.7 1.5 24 13.9 24h55.4l27.7 48c6.2 10.7 21.6 10.7 27.7 0l27.7-48H305c12.3 0 20-13.3 13.9-24l-27.7-48 27.7-48c6.2-10.7-1.5-24-13.9-24H249.6L221.9 64c-6.2-10.7-21.6-10.7-27.7 0l-27.7 48H111c-12.3 0-20 13.3-13.9 24l27.7 48zm27.7 0l27.7-48h55.4l27.7 48-27.7 48H180.3l-27.7-48zm0-48l-13.9 24-13.9-24h27.7zm41.6-24L208 88l13.9 24H194.1zm69.3 24h27.7l-13.9 24-13.9-24zm13.9 72l13.9 24H263.4l13.9-24zm-55.4 48L208 280l-13.9-24h27.7z"],"phone-volume":[512,512,["volume-control-phone"],"f2a0","M280 0C408.1 0 512 103.9 512 232c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-101.6-82.4-184-184-184c-13.3 0-24-10.7-24-24s10.7-24 24-24zm8 192a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm-32-72c0-13.3 10.7-24 24-24c75.1 0 136 60.9 136 136c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-48.6-39.4-88-88-88c-13.3 0-24-10.7-24-24zM117.5 1.4c19.4-5.3 39.7 4.6 47.4 23.2l40 96c6.8 16.3 2.1 35.2-11.6 46.3L144 207.3c33.3 70.4 90.3 127.4 160.7 160.7L345 318.7c11.2-13.7 30-18.4 46.3-11.6l96 40c18.6 7.7 28.5 28 23.2 47.4l-24 88C481.8 499.9 466 512 448 512C200.6 512 0 311.4 0 64C0 46 12.1 30.2 29.5 25.4l88-24z"],"hat-cowboy-side":[640,512,[],"f8c1","M152.7 135.9l-10.4 57.2c6.8-.7 13.6-1.1 20.5-1.1h10.7c39.4 0 77.8 12.1 110.1 34.7L562.4 421.8l35.1 24.6c24.4-6 42.5-28.1 42.5-54.4c0-75.8-94.7-126.6-134.6-144.7L474 83.9C468.2 53.8 441.8 32 411.1 32h-2.7c-5.6 0-11.1 .7-16.5 2.2L199.2 85.5c-23.9 6.4-42 26-46.5 50.4zM0 384c0 35.3 28.7 64 64 64H544L265.3 252.9c-26.9-18.8-58.9-28.9-91.8-28.9H162.9c-60.6 0-116 34.2-143.1 88.4L13.5 325C4.6 342.7 0 362.3 0 382.2V384z"],"clipboard-user":[384,512,[],"f7f3","M192 0c-41.8 0-77.4 26.7-90.5 64H64C28.7 64 0 92.7 0 128V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H282.5C269.4 26.7 233.8 0 192 0zm0 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM128 256a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM80 432c0-44.2 35.8-80 80-80h64c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16z"],child:[320,512,[],"f1ae","M96 64a64 64 0 1 1 128 0A64 64 0 1 1 96 64zm48 320v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V287.8L59.1 321c-9.4 15-29.2 19.4-44.1 10S-4.5 301.9 4.9 287l39.9-63.3C69.7 184 113.2 160 160 160s90.3 24 115.2 63.6L315.1 287c9.4 15 4.9 34.7-10 44.1s-34.7 4.9-44.1-10L240 287.8V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V384H144z"],"lira-sign":[320,512,[8356],"f195","M112 160.4c0-35.5 28.8-64.4 64.4-64.4c6.9 0 13.8 1.1 20.4 3.3l81.2 27.1c16.8 5.6 34.9-3.5 40.5-20.2s-3.5-34.9-20.2-40.5L217 38.6c-13.1-4.4-26.8-6.6-40.6-6.6C105.5 32 48 89.5 48 160.4V192H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H48v32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H46c-2.2 10.5-6.1 20.6-11.7 29.9L4.6 431.5c-5.9 9.9-6.1 22.2-.4 32.2S20.5 480 32 480H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H88.5l.7-1.1c11.6-19.3 18.9-40.7 21.6-62.9H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H112V256H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H112V160.4z"],satellite:[512,512,[128752],"f7bf","M233 7c-9.4-9.4-24.6-9.4-33.9 0l-96 96c-9.4 9.4-9.4 24.6 0 33.9l89.4 89.4-15.5 15.5C152.3 230.4 124.9 224 96 224c-31.7 0-61.5 7.7-87.8 21.2c-9 4.7-10.3 16.7-3.1 23.8L112.7 376.7 96.3 393.1c-2.6-.7-5.4-1.1-8.3-1.1c-17.7 0-32 14.3-32 32s14.3 32 32 32s32-14.3 32-32c0-2.9-.4-5.6-1.1-8.3l16.4-16.4L242.9 506.9c7.2 7.2 19.2 5.9 23.8-3.1C280.3 477.5 288 447.7 288 416c0-28.9-6.4-56.3-17.8-80.9l15.5-15.5L375 409c9.4 9.4 24.6 9.4 33.9 0l96-96c9.4-9.4 9.4-24.6 0-33.9l-89.4-89.4 55-55c12.5-12.5 12.5-32.8 0-45.3l-48-48c-12.5-12.5-32.8-12.5-45.3 0l-55 55L233 7zm159 351l-72.4-72.4 62.1-62.1L454.1 296 392 358.1zM226.3 192.4L153.9 120 216 57.9l72.4 72.4-62.1 62.1z"],"plane-lock":[640,512,[],"e558","M192 93.7C192 59.5 221 0 256 0c36 0 64 59.5 64 93.7v84.6l101.8 58.2C418 247.6 416 259.6 416 272v24.6c-17.9 10.4-30.3 29.1-31.8 50.9L320 329.1V400l57.6 43.2c4 3 6.4 7.8 6.4 12.8v24 18c0 7.8-6.3 14-14 14c-1.3 0-2.6-.2-3.9-.5L256 480 145.9 511.5c-1.3 .4-2.6 .5-3.9 .5c-7.8 0-14-6.3-14-14V456c0-5 2.4-9.8 6.4-12.8L192 400l0-70.9-171.6 49C10.2 381.1 0 373.4 0 362.8V297.3c0-5.7 3.1-11 8.1-13.9L192 178.3V93.7zM528 240c-17.7 0-32 14.3-32 32v48h64V272c0-17.7-14.3-32-32-32zm-80 32c0-44.2 35.8-80 80-80s80 35.8 80 80v48c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H448c-17.7 0-32-14.3-32-32V352c0-17.7 14.3-32 32-32V272z"],tag:[448,512,[127991],"f02b","M0 80V229.5c0 17 6.7 33.3 18.7 45.3l176 176c25 25 65.5 25 90.5 0L418.7 317.3c25-25 25-65.5 0-90.5l-176-176c-12-12-28.3-18.7-45.3-18.7H48C21.5 32 0 53.5 0 80zm112 32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],comment:[512,512,[128489,61669],"f075","M512 240c0 114.9-114.6 208-256 208c-37.1 0-72.3-6.4-104.1-17.9c-11.9 8.7-31.3 20.6-54.3 30.6C73.6 471.1 44.7 480 16 480c-6.5 0-12.3-3.9-14.8-9.9c-2.5-6-1.1-12.8 3.4-17.4l0 0 0 0 0 0 0 0 .3-.3c.3-.3 .7-.7 1.3-1.4c1.1-1.2 2.8-3.1 4.9-5.7c4.1-5 9.6-12.4 15.2-21.6c10-16.6 19.5-38.4 21.4-62.9C17.7 326.8 0 285.1 0 240C0 125.1 114.6 32 256 32s256 93.1 256 208z"],"cake-candles":[448,512,[127874,"birthday-cake","cake"],"f1fd","M86.4 5.5L61.8 47.6C58 54.1 56 61.6 56 69.2V72c0 22.1 17.9 40 40 40s40-17.9 40-40V69.2c0-7.6-2-15-5.8-21.6L105.6 5.5C103.6 2.1 100 0 96 0s-7.6 2.1-9.6 5.5zm128 0L189.8 47.6c-3.8 6.5-5.8 14-5.8 21.6V72c0 22.1 17.9 40 40 40s40-17.9 40-40V69.2c0-7.6-2-15-5.8-21.6L233.6 5.5C231.6 2.1 228 0 224 0s-7.6 2.1-9.6 5.5zM317.8 47.6c-3.8 6.5-5.8 14-5.8 21.6V72c0 22.1 17.9 40 40 40s40-17.9 40-40V69.2c0-7.6-2-15-5.8-21.6L361.6 5.5C359.6 2.1 356 0 352 0s-7.6 2.1-9.6 5.5L317.8 47.6zM128 176c0-17.7-14.3-32-32-32s-32 14.3-32 32v48c-35.3 0-64 28.7-64 64v71c8.3 5.2 18.1 9 28.8 9c13.5 0 27.2-6.1 38.4-13.4c5.4-3.5 9.9-7.1 13-9.7c1.5-1.3 2.7-2.4 3.5-3.1c.4-.4 .7-.6 .8-.8l.1-.1 0 0 0 0s0 0 0 0s0 0 0 0c3.1-3.2 7.4-4.9 11.9-4.8s8.6 2.1 11.6 5.4l0 0 0 0 .1 .1c.1 .1 .4 .4 .7 .7c.7 .7 1.7 1.7 3.1 3c2.8 2.6 6.8 6.1 11.8 9.5c10.2 7.1 23 13.1 36.3 13.1s26.1-6 36.3-13.1c5-3.5 9-6.9 11.8-9.5c1.4-1.3 2.4-2.3 3.1-3c.3-.3 .6-.6 .7-.7l.1-.1c3-3.5 7.4-5.4 12-5.4s9 2 12 5.4l.1 .1c.1 .1 .4 .4 .7 .7c.7 .7 1.7 1.7 3.1 3c2.8 2.6 6.8 6.1 11.8 9.5c10.2 7.1 23 13.1 36.3 13.1s26.1-6 36.3-13.1c5-3.5 9-6.9 11.8-9.5c1.4-1.3 2.4-2.3 3.1-3c.3-.3 .6-.6 .7-.7l.1-.1c2.9-3.4 7.1-5.3 11.6-5.4s8.7 1.6 11.9 4.8l0 0 0 0 0 0 .1 .1c.2 .2 .4 .4 .8 .8c.8 .7 1.9 1.8 3.5 3.1c3.1 2.6 7.5 6.2 13 9.7c11.2 7.3 24.9 13.4 38.4 13.4c10.7 0 20.5-3.9 28.8-9V288c0-35.3-28.7-64-64-64V176c0-17.7-14.3-32-32-32s-32 14.3-32 32v48H256V176c0-17.7-14.3-32-32-32s-32 14.3-32 32v48H128V176zM448 394.6c-8.5 3.3-18.2 5.4-28.8 5.4c-22.5 0-42.4-9.9-55.8-18.6c-4.1-2.7-7.8-5.4-10.9-7.8c-2.8 2.4-6.1 5-9.8 7.5C329.8 390 310.6 400 288 400s-41.8-10-54.6-18.9c-3.5-2.4-6.7-4.9-9.4-7.2c-2.7 2.3-5.9 4.7-9.4 7.2C201.8 390 182.6 400 160 400s-41.8-10-54.6-18.9c-3.7-2.6-7-5.2-9.8-7.5c-3.1 2.4-6.8 5.1-10.9 7.8C71.2 390.1 51.3 400 28.8 400c-10.6 0-20.3-2.2-28.8-5.4V480c0 17.7 14.3 32 32 32H416c17.7 0 32-14.3 32-32V394.6z"],envelope:[512,512,[128386,9993,61443],"f0e0","M48 64C21.5 64 0 85.5 0 112c0 15.1 7.1 29.3 19.2 38.4L236.8 313.6c11.4 8.5 27 8.5 38.4 0L492.8 150.4c12.1-9.1 19.2-23.3 19.2-38.4c0-26.5-21.5-48-48-48H48zM0 176V384c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V176L294.4 339.2c-22.8 17.1-54 17.1-76.8 0L0 176z"],"angles-up":[448,512,["angle-double-up"],"f102","M246.6 41.4c-12.5-12.5-32.8-12.5-45.3 0l-160 160c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L224 109.3 361.4 246.6c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-160-160zm160 352l-160-160c-12.5-12.5-32.8-12.5-45.3 0l-160 160c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L224 301.3 361.4 438.6c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3z"],paperclip:[448,512,[128206],"f0c6","M364.2 83.8c-24.4-24.4-64-24.4-88.4 0l-184 184c-42.1 42.1-42.1 110.3 0 152.4s110.3 42.1 152.4 0l152-152c10.9-10.9 28.7-10.9 39.6 0s10.9 28.7 0 39.6l-152 152c-64 64-167.6 64-231.6 0s-64-167.6 0-231.6l184-184c46.3-46.3 121.3-46.3 167.6 0s46.3 121.3 0 167.6l-176 176c-28.6 28.6-75 28.6-103.6 0s-28.6-75 0-103.6l144-144c10.9-10.9 28.7-10.9 39.6 0s10.9 28.7 0 39.6l-144 144c-6.7 6.7-6.7 17.7 0 24.4s17.7 6.7 24.4 0l176-176c24.4-24.4 24.4-64 0-88.4z"],"arrow-right-to-city":[640,512,[],"e4b3","M288 48c0-26.5 21.5-48 48-48h96c26.5 0 48 21.5 48 48V192h40V120c0-13.3 10.7-24 24-24s24 10.7 24 24v72h24c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H432 336c-26.5 0-48-21.5-48-48V48zm64 32v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V80c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16zm16 80c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V176c0-8.8-7.2-16-16-16H368zM352 272v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16zm176-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H528zM512 368v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H528c-8.8 0-16 7.2-16 16zM166.6 153.4l80 80c12.5 12.5 12.5 32.8 0 45.3l-80 80c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L146.7 288H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H146.7l-25.4-25.4c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0z"],ribbon:[448,512,[127895],"f4d6","M333.2 322.8l0 0-133.9-146 0 0L146 118.6c7.8-5.1 37-22.6 78-22.6s70.2 17.4 78 22.6L245.7 180l85.6 93.4 27.4-29.8c16.3-17.7 25.3-40.9 25.3-65V149.1c0-19-5.6-37.5-16.1-53.3L327.8 35.6C312.9 13.4 287.9 0 261.2 0h-76c-25.8 0-50.1 12.5-65.1 33.5L81.9 87C70.3 103.2 64 122.8 64 142.8V164c0 23.2 8.4 45.6 23.6 63.1l56 64.2 0 0 83.3 95.6 0 0 91.8 105.3c10 11.5 26.8 14.3 40 6.8l54.5-31.1c17.8-10.2 21.6-34.3 7.7-49.4l-87.7-95.7zM205.2 410.6l-83.3-95.6L27.1 418.5c-13.9 15.1-10.1 39.2 7.7 49.4l55.1 31.5c13 7.4 29.3 4.9 39.4-6.1l75.9-82.6z"],lungs:[640,512,[129729],"f604","M320 0c17.7 0 32 14.3 32 32V164.1c0 16.4 8.4 31.7 22.2 40.5l9.8 6.2V165.3C384 127 415 96 453.3 96c21.7 0 42.8 10.2 55.8 28.8c15.4 22.1 44.3 65.4 71 116.9c26.5 50.9 52.4 112.5 59.6 170.3c.2 1.3 .2 2.6 .2 4v7c0 49.1-39.8 89-89 89c-7.3 0-14.5-.9-21.6-2.7l-72.7-18.2C414 480.5 384 442.1 384 398V325l90.5 57.6c7.5 4.7 17.3 2.5 22.1-4.9s2.5-17.3-4.9-22.1L384 287.1v-.4l-44.1-28.1c-7.3-4.6-13.9-10.1-19.9-16.1c-5.9 6-12.6 11.5-19.9 16.1L256 286.7 161.2 347l-13.5 8.6c0 0 0 0-.1 0c-7.4 4.8-9.6 14.6-4.8 22.1c4.7 7.5 14.6 9.7 22.1 4.9l91.1-58V398c0 44.1-30 82.5-72.7 93.1l-72.7 18.2c-7.1 1.8-14.3 2.7-21.6 2.7c-49.1 0-89-39.8-89-89v-7c0-1.3 .1-2.7 .2-4c7.2-57.9 33.1-119.4 59.6-170.3c26.8-51.5 55.6-94.8 71-116.9c13-18.6 34-28.8 55.8-28.8C225 96 256 127 256 165.3v45.5l9.8-6.2c13.8-8.8 22.2-24.1 22.2-40.5V32c0-17.7 14.3-32 32-32z"],"arrow-up-9-1":[576,512,["sort-numeric-up-alt"],"f887","M160 32c9 0 17.5 3.8 23.6 10.4l88 96c11.9 13 11.1 33.3-2 45.2s-33.3 11.1-45.2-2L192 146.3V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V146.3L95.6 181.6c-11.9 13-32.2 13.9-45.2 2s-13.9-32.2-2-45.2l88-96C142.5 35.8 151 32 160 32zM450.7 294c8.3 6 13.3 15.7 13.3 26v96h16c17.7 0 32 14.3 32 32s-14.3 32-32 32H432 384c-17.7 0-32-14.3-32-32s14.3-32 32-32h16V364.4l-5.9 2c-16.8 5.6-34.9-3.5-40.5-20.2s3.5-34.9 20.2-40.5l48-16c9.8-3.3 20.5-1.6 28.8 4.4zm-5-145.1A32 32 0 1 0 418.3 91a32 32 0 1 0 27.4 57.9zm-40.7 54.9C369.6 192.4 344 159.2 344 120c0-48.6 39.4-88 88-88s88 39.4 88 88c0 23.5-7.5 46.3-21.5 65.2L449.7 251c-10.5 14.2-30.6 17.2-44.8 6.7s-17.2-30.6-6.7-44.8l6.8-9.2z"],"litecoin-sign":[384,512,[],"e1d3","M128 64c0-17.7-14.3-32-32-32S64 46.3 64 64V213.6L23.2 225.2c-17 4.9-26.8 22.6-22 39.6s22.6 26.8 39.6 22L64 280.1V448c0 17.7 14.3 32 32 32H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H128V261.9l136.8-39.1c17-4.9 26.8-22.6 22-39.6s-22.6-26.8-39.6-22L128 195.3V64z"],"border-none":[448,512,[],"f850","M32 480a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm96-64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm0-384a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm0 256a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM320 416a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm0-320a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm0 128a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM224 480a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm0-448a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm0 256a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM416 416a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm0-384a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM32 96a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM416 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM32 288a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm192 32a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm192 64a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM32 320a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM416 192a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM32 128a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm192 64a32 32 0 1 1 0-64 32 32 0 1 1 0 64z"],"circle-nodes":[512,512,[],"e4e2","M418.4 157.9c35.3-8.3 61.6-40 61.6-77.9c0-44.2-35.8-80-80-80c-43.4 0-78.7 34.5-80 77.5L136.2 151.1C121.7 136.8 101.9 128 80 128c-44.2 0-80 35.8-80 80s35.8 80 80 80c12.2 0 23.8-2.7 34.1-7.6L259.7 407.8c-2.4 7.6-3.7 15.8-3.7 24.2c0 44.2 35.8 80 80 80s80-35.8 80-80c0-27.7-14-52.1-35.4-66.4l37.8-207.7zM156.3 232.2c2.2-6.9 3.5-14.2 3.7-21.7l183.8-73.5c3.6 3.5 7.4 6.7 11.6 9.5L317.6 354.1c-5.5 1.3-10.8 3.1-15.8 5.5L156.3 232.2z"],"parachute-box":[512,512,[],"f4cd","M383.5 192c.3-5.3 .5-10.6 .5-16c0-51-15.9-96-40.2-127.6C319.5 16.9 288.2 0 256 0s-63.5 16.9-87.8 48.4C143.9 80 128 125 128 176c0 5.4 .2 10.7 .5 16H240V320H208c-7 0-13.7 1.5-19.7 4.2L68.2 192H96.5c-.3-5.3-.5-10.6-.5-16c0-64 22.2-121.2 57.1-159.3C62 49.3 18.6 122.6 4.2 173.6C1.5 183.1 9 192 18.9 192h6L165.2 346.3c-3.3 6.5-5.2 13.9-5.2 21.7v96c0 26.5 21.5 48 48 48h96c26.5 0 48-21.5 48-48V368c0-7.8-1.9-15.2-5.2-21.7L487.1 192h6c9.9 0 17.4-8.9 14.7-18.4C493.4 122.6 450 49.3 358.9 16.7C393.8 54.8 416 112.1 416 176c0 5.4-.2 10.7-.5 16h28.3L323.7 324.2c-6-2.7-12.7-4.2-19.7-4.2H272V192H383.5z"],indent:[448,512,[],"f03c","M0 64C0 46.3 14.3 32 32 32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 96 0 81.7 0 64zM192 192c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32zm32 96H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32s14.3-32 32-32zM0 448c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32zM127.8 268.6L25.8 347.9C15.3 356.1 0 348.6 0 335.3V176.7c0-13.3 15.3-20.8 25.8-12.6l101.9 79.3c8.2 6.4 8.2 18.9 0 25.3z"],"truck-field-un":[640,512,[],"e58e","M96 32C60.7 32 32 60.7 32 96v32c-17.7 0-32 14.3-32 32v96c0 17.7 14.3 32 32 32v32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64c0 53 43 96 96 96s96-43 96-96H384c0 53 43 96 96 96s96-43 96-96h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V288c0-35.3-28.7-64-64-64h-4.2c-.4-1.1-.9-2.1-1.3-3.2L485.7 102c-10.3-23.1-33.2-38-58.5-38H375.4C364.4 44.9 343.7 32 320 32H96zm288 96h43.2l42.7 96H384V128zM112 384a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm368-48a48 48 0 1 1 0 96 48 48 0 1 1 0-96zM253.3 135.1l34.7 52V144c0-8.8 7.2-16 16-16s16 7.2 16 16v96c0 7.1-4.6 13.3-11.4 15.3s-14-.6-17.9-6.4l-34.7-52V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V144c0-7.1 4.6-13.3 11.4-15.3s14 .6 17.9 6.4zM128 144v64c0 8.8 7.2 16 16 16s16-7.2 16-16V144c0-8.8 7.2-16 16-16s16 7.2 16 16v64c0 26.5-21.5 48-48 48s-48-21.5-48-48V144c0-8.8 7.2-16 16-16s16 7.2 16 16z"],hourglass:[384,512,[9203,62032,"hourglass-empty"],"f254","M0 32C0 14.3 14.3 0 32 0H64 320h32c17.7 0 32 14.3 32 32s-14.3 32-32 32V75c0 42.4-16.9 83.1-46.9 113.1L237.3 256l67.9 67.9c30 30 46.9 70.7 46.9 113.1v11c17.7 0 32 14.3 32 32s-14.3 32-32 32H320 64 32c-17.7 0-32-14.3-32-32s14.3-32 32-32V437c0-42.4 16.9-83.1 46.9-113.1L146.7 256 78.9 188.1C48.9 158.1 32 117.4 32 75V64C14.3 64 0 49.7 0 32zM96 64V75c0 25.5 10.1 49.9 28.1 67.9L192 210.7l67.9-67.9c18-18 28.1-42.4 28.1-67.9V64H96zm0 384H288V437c0-25.5-10.1-49.9-28.1-67.9L192 301.3l-67.9 67.9c-18 18-28.1 42.4-28.1 67.9v11z"],mountain:[512,512,[127956],"f6fc","M256 32c12.5 0 24.1 6.4 30.8 17L503.4 394.4c5.6 8.9 8.6 19.2 8.6 29.7c0 30.9-25 55.9-55.9 55.9H55.9C25 480 0 455 0 424.1c0-10.5 3-20.8 8.6-29.7L225.2 49c6.6-10.6 18.3-17 30.8-17zm65 192L256 120.4 176.9 246.5l18.3 24.4c6.4 8.5 19.2 8.5 25.6 0l25.6-34.1c6-8.1 15.5-12.8 25.6-12.8h49z"],"user-doctor":[448,512,["user-md"],"f0f0","M224 256A128 128 0 1 0 224 0a128 128 0 1 0 0 256zm-96 55.2C54 332.9 0 401.3 0 482.3C0 498.7 13.3 512 29.7 512H418.3c16.4 0 29.7-13.3 29.7-29.7c0-81-54-149.4-128-171.1V362c27.6 7.1 48 32.2 48 62v40c0 8.8-7.2 16-16 16H336c-8.8 0-16-7.2-16-16s7.2-16 16-16V424c0-17.7-14.3-32-32-32s-32 14.3-32 32v24c8.8 0 16 7.2 16 16s-7.2 16-16 16H256c-8.8 0-16-7.2-16-16V424c0-29.8 20.4-54.9 48-62V304.9c-6-.6-12.1-.9-18.3-.9H178.3c-6.2 0-12.3 .3-18.3 .9v65.4c23.1 6.9 40 28.3 40 53.7c0 30.9-25.1 56-56 56s-56-25.1-56-56c0-25.4 16.9-46.8 40-53.7V311.2zM144 448a24 24 0 1 0 0-48 24 24 0 1 0 0 48z"],"circle-info":[512,512,["info-circle"],"f05a","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM216 336h24V272H216c-13.3 0-24-10.7-24-24s10.7-24 24-24h48c13.3 0 24 10.7 24 24v88h8c13.3 0 24 10.7 24 24s-10.7 24-24 24H216c-13.3 0-24-10.7-24-24s10.7-24 24-24zm40-208a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"cloud-meatball":[512,512,[],"f73b","M0 224c0 53 43 96 96 96h44.7c9.5-23.5 32.5-40 59.3-40c2 0 3.9 .1 5.8 .3C217.6 265.5 235.7 256 256 256s38.4 9.5 50.2 24.3c1.9-.2 3.9-.3 5.8-.3c26.9 0 49.9 16.5 59.3 40H416c53 0 96-43 96-96s-43-96-96-96c-.5 0-1.1 0-1.6 0c1.1-5.2 1.6-10.5 1.6-16c0-44.2-35.8-80-80-80c-24.3 0-46.1 10.9-60.8 28C256.5 24.3 219.1 0 176 0C114.1 0 64 50.1 64 112c0 7.1 .7 14.1 1.9 20.8C27.6 145.4 0 181.5 0 224zm288 96c0-17.7-14.3-32-32-32s-32 14.3-32 32c0 1 .1 2.1 .1 3.1c-.7-.8-1.4-1.6-2.1-2.3c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3c.7 .7 1.5 1.4 2.3 2.1c-1-.1-2.1-.1-3.1-.1c-17.7 0-32 14.3-32 32s14.3 32 32 32c1 0 2.1-.1 3.1-.1c-.8 .7-1.6 1.3-2.3 2.1c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0c.7-.7 1.4-1.5 2.1-2.3c-.1 1-.1 2.1-.1 3.1c0 17.7 14.3 32 32 32s32-14.3 32-32c0-1-.1-2.1-.1-3.1c.7 .8 1.3 1.6 2.1 2.3c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3c-.7-.7-1.5-1.4-2.3-2.1c1 .1 2.1 .1 3.1 .1c17.7 0 32-14.3 32-32s-14.3-32-32-32c-1 0-2.1 .1-3.1 .1c.8-.7 1.6-1.3 2.3-2.1c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0c-.7 .7-1.4 1.5-2.1 2.3c.1-1 .1-2.1 .1-3.1zM48 448a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm416 0a48 48 0 1 0 0-96 48 48 0 1 0 0 96z"],camera:[512,512,[62258,"camera-alt"],"f030","M149.1 64.8L138.7 96H64C28.7 96 0 124.7 0 160V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H373.3L362.9 64.8C356.4 45.2 338.1 32 317.4 32H194.6c-20.7 0-39 13.2-45.5 32.8zM256 192a96 96 0 1 1 0 192 96 96 0 1 1 0-192z"],"square-virus":[448,512,[],"e578","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM223.8 93.7c13.3 0 24 10.7 24 24c0 29.3 35.4 43.9 56.1 23.2c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9c-20.7 20.7-6 56.1 23.2 56.1c13.3 0 24 10.7 24 24s-10.7 24-24 24c-29.3 0-43.9 35.4-23.2 56.1c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0c-20.7-20.7-56.1-6-56.1 23.2c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-29.3-35.4-43.9-56.1-23.2c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9c20.7-20.7 6-56.1-23.2-56.1c-13.3 0-24-10.7-24-24s10.7-24 24-24c29.3 0 43.9-35.4 23.2-56.1c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0c20.7 20.7 56.1 6 56.1-23.2c0-13.3 10.7-24 24-24zM192 256a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm88 32a24 24 0 1 0 -48 0 24 24 0 1 0 48 0z"],meteor:[512,512,[9732],"f753","M493.7 .9L299.4 75.6l2.3-29.3c1-12.8-12.8-21.5-24-15.1L101.3 133.4C38.6 169.7 0 236.6 0 309C0 421.1 90.9 512 203 512c72.4 0 139.4-38.6 175.7-101.3L480.8 234.3c6.5-11.1-2.2-25-15.1-24l-29.3 2.3L511.1 18.3c.6-1.5 .9-3.2 .9-4.8C512 6 506 0 498.5 0c-1.7 0-3.3 .3-4.8 .9zM192 192a128 128 0 1 1 0 256 128 128 0 1 1 0-256zm0 96a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm16 96a16 16 0 1 0 0-32 16 16 0 1 0 0 32z"],"car-on":[512,512,[],"e4dd","M280 24c0-13.3-10.7-24-24-24s-24 10.7-24 24v80c0 13.3 10.7 24 24 24s24-10.7 24-24V24zM185.8 224H326.2c6.8 0 12.8 4.3 15.1 10.6L360.3 288H151.7l19.1-53.4c2.3-6.4 8.3-10.6 15.1-10.6zm-75.3-10.9L82.2 292.4C62.1 300.9 48 320.8 48 344v40 64 32c0 17.7 14.3 32 32 32H96c17.7 0 32-14.3 32-32V448H384v32c0 17.7 14.3 32 32 32h16c17.7 0 32-14.3 32-32V448 384 344c0-23.2-14.1-43.1-34.2-51.6l-28.3-79.3C390.1 181.3 360 160 326.2 160H185.8c-33.8 0-64 21.3-75.3 53.1zM128 344a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm232 24a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zM39 39c-9.4 9.4-9.4 24.6 0 33.9l48 48c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L73 39c-9.4-9.4-24.6-9.4-33.9 0zm400 0L391 87c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l48-48c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0z"],sleigh:[640,512,[],"f7cc","M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96V256c0 53 43 96 96 96v32h64V352H384v32h64V352c53 0 96-43 96-96V160c17.7 0 32-14.3 32-32s-14.3-32-32-32H512 480c-17.7 0-32 14.3-32 32v41.3c0 30.2-24.5 54.7-54.7 54.7c-75.5 0-145.6-38.9-185.6-102.9l-4.3-6.9C174.2 67.6 125 37.6 70.7 32.7c-2.2-.5-4.4-.7-6.7-.7H55 32zM640 384c0-17.7-14.3-32-32-32s-32 14.3-32 32v8c0 13.3-10.7 24-24 24H64c-17.7 0-32 14.3-32 32s14.3 32 32 32H552c48.6 0 88-39.4 88-88v-8z"],"arrow-down-1-9":[576,512,["sort-numeric-asc","sort-numeric-down"],"f162","M450.7 38c-8.3-6-19.1-7.7-28.8-4.4l-48 16c-16.8 5.6-25.8 23.7-20.2 40.5s23.7 25.8 40.5 20.2l5.9-2V160H384c-17.7 0-32 14.3-32 32s14.3 32 32 32h48 48c17.7 0 32-14.3 32-32s-14.3-32-32-32H464V64c0-10.3-4.9-19.9-13.3-26zM160 480c9 0 17.5-3.8 23.6-10.4l88-96c11.9-13 11.1-33.3-2-45.2s-33.3-11.1-45.2 2L192 365.7V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V365.7L95.6 330.4c-11.9-13-32.2-13.9-45.2-2s-13.9 32.2-2 45.2l88 96C142.5 476.2 151 480 160 480zM418.3 307a32 32 0 1 1 27.4 57.9A32 32 0 1 1 418.3 307zM405.1 419.8l-6.8 9.2c-10.5 14.2-7.5 34.2 6.7 44.8s34.2 7.5 44.8-6.7l48.8-65.8c14-18.9 21.5-41.7 21.5-65.2c0-48.6-39.4-88-88-88s-88 39.4-88 88c0 39.2 25.6 72.4 61.1 83.8z"],"hand-holding-droplet":[576,512,["hand-holding-water"],"f4c1","M275.5 6.6C278.3 2.5 283 0 288 0s9.7 2.5 12.5 6.6L366.8 103C378 119.3 384 138.6 384 158.3V160c0 53-43 96-96 96s-96-43-96-96v-1.7c0-19.8 6-39 17.2-55.3L275.5 6.6zM568.2 336.3c13.1 17.8 9.3 42.8-8.5 55.9L433.1 485.5c-23.4 17.2-51.6 26.5-80.7 26.5H192 32c-17.7 0-32-14.3-32-32V416c0-17.7 14.3-32 32-32H68.8l44.9-36c22.7-18.2 50.9-28 80-28H272h16 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H288 272c-8.8 0-16 7.2-16 16s7.2 16 16 16H392.6l119.7-88.2c17.8-13.1 42.8-9.3 55.9 8.5zM193.6 384l0 0-.9 0c.3 0 .6 0 .9 0z"],water:[576,512,[],"f773","M269.5 69.9c11.1-7.9 25.9-7.9 37 0C329 85.4 356.5 96 384 96c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 149.7 417 160 384 160c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4C42.8 92.6 61 83.5 75.3 71.6c11.1-9.5 27.3-10.1 39.2-1.7l0 0C136.7 85.2 165.1 96 192 96c27.5 0 55-10.6 77.5-26.1zm37 288C329 373.4 356.5 384 384 384c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 437.7 417 448 384 448c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 373.2 165.1 384 192 384c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0zm0-144C329 229.4 356.5 240 384 240c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 293.7 417 304 384 304c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.5 27.3-10.1 39.2-1.7l0 0C136.7 229.2 165.1 240 192 240c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z"],"calendar-check":[448,512,[],"f274","M128 0c17.7 0 32 14.3 32 32V64H288V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64h48c26.5 0 48 21.5 48 48v48H0V112C0 85.5 21.5 64 48 64H96V32c0-17.7 14.3-32 32-32zM0 192H448V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V192zM329 305c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-95 95-47-47c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l64 64c9.4 9.4 24.6 9.4 33.9 0L329 305z"],braille:[640,512,[],"f2a1","M0 96a64 64 0 1 1 128 0A64 64 0 1 1 0 96zM224 272a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm0-80a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM80 416a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zM0 416a64 64 0 1 1 128 0A64 64 0 1 1 0 416zm240 0a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm-80 0a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM64 192a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM224 32a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM352 96a64 64 0 1 1 128 0A64 64 0 1 1 352 96zm240 0a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm-80 0a64 64 0 1 1 128 0A64 64 0 1 1 512 96zm64 176a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm0-80a64 64 0 1 1 0 128 64 64 0 1 1 0-128zm16 224a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm-80 0a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM416 272a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm0-80a64 64 0 1 1 0 128 64 64 0 1 1 0-128zm16 224a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm-80 0a64 64 0 1 1 128 0 64 64 0 1 1 -128 0z"],"prescription-bottle-medical":[384,512,["prescription-bottle-alt"],"f486","M0 32C0 14.3 14.3 0 32 0H352c17.7 0 32 14.3 32 32V64c0 17.7-14.3 32-32 32H32C14.3 96 0 81.7 0 64V32zm32 96H352V448c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V128zM160 240v48H112c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V352h48c8.8 0 16-7.2 16-16V304c0-8.8-7.2-16-16-16H224V240c0-8.8-7.2-16-16-16H176c-8.8 0-16 7.2-16 16z"],landmark:[512,512,[127963],"f66f","M240.1 4.2c9.8-5.6 21.9-5.6 31.8 0l171.8 98.1L448 104l0 .9 47.9 27.4c12.6 7.2 18.8 22 15.1 36s-16.4 23.8-30.9 23.8H32c-14.5 0-27.2-9.8-30.9-23.8s2.5-28.8 15.1-36L64 104.9V104l4.4-1.6L240.1 4.2zM64 224h64V416h40V224h64V416h48V224h64V416h40V224h64V420.3c.6 .3 1.2 .7 1.8 1.1l48 32c11.7 7.8 17 22.4 12.9 35.9S494.1 512 480 512H32c-14.1 0-26.5-9.2-30.6-22.7s1.1-28.1 12.9-35.9l48-32c.6-.4 1.2-.7 1.8-1.1V224z"],truck:[640,512,[128666,9951],"f0d1","M48 0C21.5 0 0 21.5 0 48V368c0 26.5 21.5 48 48 48H64c0 53 43 96 96 96s96-43 96-96H384c0 53 43 96 96 96s96-43 96-96h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V288 256 237.3c0-17-6.7-33.3-18.7-45.3L512 114.7c-12-12-28.3-18.7-45.3-18.7H416V48c0-26.5-21.5-48-48-48H48zM416 160h50.7L544 237.3V256H416V160zM112 416a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm368-48a48 48 0 1 1 0 96 48 48 0 1 1 0-96z"],crosshairs:[512,512,[],"f05b","M256 0c17.7 0 32 14.3 32 32V42.4c93.7 13.9 167.7 88 181.6 181.6H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H469.6c-13.9 93.7-88 167.7-181.6 181.6V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V469.6C130.3 455.7 56.3 381.7 42.4 288H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H42.4C56.3 130.3 130.3 56.3 224 42.4V32c0-17.7 14.3-32 32-32zM107.4 288c12.5 58.3 58.4 104.1 116.6 116.6V384c0-17.7 14.3-32 32-32s32 14.3 32 32v20.6c58.3-12.5 104.1-58.4 116.6-116.6H384c-17.7 0-32-14.3-32-32s14.3-32 32-32h20.6C392.1 165.7 346.3 119.9 288 107.4V128c0 17.7-14.3 32-32 32s-32-14.3-32-32V107.4C165.7 119.9 119.9 165.7 107.4 224H128c17.7 0 32 14.3 32 32s-14.3 32-32 32H107.4zM256 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"person-cane":[448,512,[],"e53c","M272 48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zm-8 187.3l47.4 57.1c11.3 13.6 31.5 15.5 45.1 4.2s15.5-31.5 4.2-45.1l-73.7-88.9c-18.2-22-45.3-34.7-73.9-34.7H177.1c-33.7 0-64.9 17.7-82.3 46.6l-58.3 97c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9L120 256.9V480c0 17.7 14.3 32 32 32s32-14.3 32-32V352h16V480c0 17.7 14.3 32 32 32s32-14.3 32-32V235.3zM352 376c0-4.4 3.6-8 8-8s8 3.6 8 8V488c0 13.3 10.7 24 24 24s24-10.7 24-24V376c0-30.9-25.1-56-56-56s-56 25.1-56 56v8c0 13.3 10.7 24 24 24s24-10.7 24-24v-8z"],tent:[576,512,[],"e57d","M269.4 6C280.5-2 295.5-2 306.6 6l224 160c7.4 5.3 12.2 13.5 13.2 22.5l32 288c1 9-1.9 18.1-8 24.9s-14.7 10.7-23.8 10.7H464 435.8c-12.1 0-23.2-6.8-28.6-17.7L306.7 293.5c-1.7-3.4-5.1-5.5-8.8-5.5c-5.5 0-9.9 4.4-9.9 9.9V480c0 17.7-14.3 32-32 32H240 32c-9.1 0-17.8-3.9-23.8-10.7s-9-15.8-8-24.9l32-288c1-9 5.8-17.2 13.2-22.5L269.4 6z"],"vest-patches":[448,512,[],"e086","M151.2 69.7l55.9 167.7-11 33.1c-2.7 8.2-4.1 16.7-4.1 25.3V464c0 14.5 3.9 28.2 10.7 39.9C195 509 185.9 512 176 512H48c-26.5 0-48-21.5-48-48V270.5c0-9.5 2.8-18.7 8.1-26.6l47.9-71.8c5.3-7.9 8.1-17.1 8.1-26.6V128 54.3 48C64 21.5 85.5 0 112 0h4.5c.2 0 .4 0 .6 0c.4 0 .8 0 1.2 0c18.8 0 34.1 9.7 44.1 18.8C171.6 27.2 190.8 40 224 40s52.4-12.8 61.7-21.2C295.7 9.7 311 0 329.7 0c.4 0 .8 0 1.2 0c.2 0 .4 0 .6 0H336c26.5 0 48 21.5 48 48v6.3V128v17.5c0 9.5 2.8 18.7 8.1 26.6l47.9 71.8c5.3 7.9 8.1 17.1 8.1 26.6V464c0 26.5-21.5 48-48 48H272c-26.5 0-48-21.5-48-48V295.8c0-5.2 .8-10.3 2.5-15.2L296.8 69.7C279.4 79.7 255.4 88 224 88s-55.4-8.3-72.8-18.3zM96 456a40 40 0 1 0 0-80 40 40 0 1 0 0 80zM63.5 255.5c-4.7 4.7-4.7 12.3 0 17L79 288 63.5 303.5c-4.7 4.7-4.7 12.3 0 17s12.3 4.7 17 0L96 305l15.5 15.5c4.7 4.7 12.3 4.7 17 0s4.7-12.3 0-17L113 288l15.5-15.5c4.7-4.7 4.7-12.3 0-17s-12.3-4.7-17 0L96 271 80.5 255.5c-4.7-4.7-12.3-4.7-17 0zM304 280v8 32c0 8.8 7.2 16 16 16h32 8c13.3 0 24-10.7 24-24s-10.7-24-24-24h-8v-8c0-13.3-10.7-24-24-24s-24 10.7-24 24z"],"check-double":[448,512,[],"f560","M342.6 86.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L160 178.7l-57.4-57.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l80 80c12.5 12.5 32.8 12.5 45.3 0l160-160zm96 128c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L160 402.7 54.6 297.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l128 128c12.5 12.5 32.8 12.5 45.3 0l256-256z"],"arrow-down-a-z":[576,512,["sort-alpha-asc","sort-alpha-down"],"f15d","M183.6 469.6C177.5 476.2 169 480 160 480s-17.5-3.8-23.6-10.4l-88-96c-11.9-13-11.1-33.3 2-45.2s33.3-11.1 45.2 2L128 365.7V64c0-17.7 14.3-32 32-32s32 14.3 32 32V365.7l32.4-35.4c11.9-13 32.2-13.9 45.2-2s13.9 32.2 2 45.2l-88 96zM320 320c0-17.7 14.3-32 32-32H480c12.9 0 24.6 7.8 29.6 19.8s2.2 25.7-6.9 34.9L429.3 416H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H352c-12.9 0-24.6-7.8-29.6-19.8s-2.2-25.7 6.9-34.9L402.7 352H352c-17.7 0-32-14.3-32-32zM416 32c12.1 0 23.2 6.8 28.6 17.7l64 128 16 32c7.9 15.8 1.5 35-14.3 42.9s-35 1.5-42.9-14.3L460.2 224H371.8l-7.2 14.3c-7.9 15.8-27.1 22.2-42.9 14.3s-22.2-27.1-14.3-42.9l16-32 64-128C392.8 38.8 403.9 32 416 32zM395.8 176h40.4L416 135.6 395.8 176z"],"money-bill-wheat":[512,512,[],"e52a","M176 0c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16c-44.2 0-80-35.8-80-80c0-8.8 7.2-16 16-16zM56 16h48c13.3 0 24 10.7 24 24s-10.7 24-24 24H56C42.7 64 32 53.3 32 40s10.7-24 24-24zM24 88H136c13.3 0 24 10.7 24 24s-10.7 24-24 24H24c-13.3 0-24-10.7-24-24S10.7 88 24 88zm8 96c0-13.3 10.7-24 24-24h48c13.3 0 24 10.7 24 24s-10.7 24-24 24H56c-13.3 0-24-10.7-24-24zM272 16c0-8.8 7.2-16 16-16c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16c-44.2 0-80-35.8-80-80zM400 0c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16c-44.2 0-80-35.8-80-80c0-8.8 7.2-16 16-16zm80 144c0 44.2-35.8 80-80 80c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80c8.8 0 16 7.2 16 16zM352 128c8.8 0 16 7.2 16 16c0 44.2-35.8 80-80 80c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80zm-96 16c0 44.2-35.8 80-80 80c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80c8.8 0 16 7.2 16 16zM0 304c0-26.5 21.5-48 48-48H464c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V304zM48 416v48H96c0-26.5-21.5-48-48-48zM96 304H48v48c26.5 0 48-21.5 48-48zM464 416c-26.5 0-48 21.5-48 48h48V416zM416 304c0 26.5 21.5 48 48 48V304H416zm-96 80a64 64 0 1 0 -128 0 64 64 0 1 0 128 0z"],cookie:[512,512,[127850],"f563","M247.2 17c-22.1-3.1-44.6 .9-64.4 11.4l-74 39.5C89.1 78.4 73.2 94.9 63.4 115L26.7 190.6c-9.8 20.1-13 42.9-9.1 64.9l14.5 82.8c3.9 22.1 14.6 42.3 30.7 57.9l60.3 58.4c16.1 15.6 36.6 25.6 58.7 28.7l83 11.7c22.1 3.1 44.6-.9 64.4-11.4l74-39.5c19.7-10.5 35.6-27 45.4-47.2l36.7-75.5c9.8-20.1 13-42.9 9.1-64.9l-14.6-82.8c-3.9-22.1-14.6-42.3-30.7-57.9L388.9 57.5c-16.1-15.6-36.6-25.6-58.7-28.7L247.2 17zM208 144a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM144 336a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm224-64a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"arrow-rotate-left":[512,512,[8634,"arrow-left-rotate","arrow-rotate-back","arrow-rotate-backward","undo"],"f0e2","M125.7 160H176c17.7 0 32 14.3 32 32s-14.3 32-32 32H48c-17.7 0-32-14.3-32-32V64c0-17.7 14.3-32 32-32s32 14.3 32 32v51.2L97.6 97.6c87.5-87.5 229.3-87.5 316.8 0s87.5 229.3 0 316.8s-229.3 87.5-316.8 0c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0c62.5 62.5 163.8 62.5 226.3 0s62.5-163.8 0-226.3s-163.8-62.5-226.3 0L125.7 160z"],"hard-drive":[512,512,[128436,"hdd"],"f0a0","M0 96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V280.4c-17-15.2-39.4-24.4-64-24.4H64c-24.6 0-47 9.2-64 24.4V96zM64 288H448c35.3 0 64 28.7 64 64v64c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V352c0-35.3 28.7-64 64-64zM320 416a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm128-32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z"],"face-grin-squint-tears":[512,512,[129315,"grin-squint-tears"],"f586","M426.8 14.2C446-5 477.5-4.6 497.1 14.9s20 51 .7 70.3c-6.8 6.8-21.4 12.4-37.4 16.7c-16.3 4.4-34.1 7.5-46.3 9.3c-1.6 .2-3.1 .5-4.6 .6c-5.6 .9-10.3-3.9-9.5-9.5c1.6-11.2 4.6-29.6 9-47c.3-1.3 .7-2.6 1-3.9c4.3-15.9 9.8-30.5 16.7-37.4zm-44.7 19c-1.5 4.8-2.9 9.6-4.1 14.3c-4.8 18.9-8 38.5-9.7 50.3c-4 26.8 18.9 49.7 45.7 45.8c11.9-1.6 31.5-4.8 50.4-9.7c4.7-1.2 9.5-2.5 14.3-4.1C534.2 227.5 520.2 353.8 437 437c-83.2 83.2-209.5 97.2-307.2 41.8c1.5-4.8 2.8-9.6 4-14.3c4.8-18.9 8-38.5 9.7-50.3c4-26.8-18.9-49.7-45.7-45.8c-11.9 1.6-31.5 4.8-50.4 9.7c-4.7 1.2-9.5 2.5-14.3 4.1C-22.2 284.5-8.2 158.2 75 75C158.2-8.3 284.5-22.2 382.2 33.2zM51.5 410.1c18.5-5 38.8-8.3 50.9-10c5.6-.9 10.3 3.9 9.5 9.5c-1.7 12.1-5 32.4-10 50.9C97.6 476.4 92 491 85.2 497.8C66 517 34.5 516.6 14.9 497.1s-20-51-.7-70.3c6.8-6.8 21.4-12.4 37.4-16.7zM416.4 202.3c-4.8-11.9-20.9-10.9-26.9 .4c-19.4 36.7-46.3 73.2-79.8 106.7s-70 60.3-106.7 79.8c-11.3 6-12.3 22.1-.4 26.9c59.4 24.1 129.9 12.2 177.9-35.8s59.9-118.5 35.8-177.9zM87.1 285.1c2 2 4.6 3.2 7.3 3.4l56.1 5.1 5.1 56.1c.3 2.8 1.5 5.4 3.4 7.3c6.3 6.3 17.2 3.6 19.8-4.9l29.7-97.4c3.5-11.6-7.3-22.5-19-19L92 265.3c-8.6 2.6-11.3 13.4-4.9 19.8zM265.3 92l-29.7 97.4c-3.5 11.6 7.3 22.5 19 19l97.4-29.7c8.6-2.6 11.3-13.4 4.9-19.8c-2-2-4.6-3.2-7.3-3.4l-56.1-5.1-5.1-56.1c-.3-2.8-1.5-5.4-3.4-7.3c-6.3-6.3-17.2-3.6-19.8 4.9z"],dumbbell:[640,512,[],"f44b","M96 64c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V224v64V448c0 17.7-14.3 32-32 32H128c-17.7 0-32-14.3-32-32V384H64c-17.7 0-32-14.3-32-32V288c-17.7 0-32-14.3-32-32s14.3-32 32-32V160c0-17.7 14.3-32 32-32H96V64zm448 0v64h32c17.7 0 32 14.3 32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32v64c0 17.7-14.3 32-32 32H544v64c0 17.7-14.3 32-32 32H480c-17.7 0-32-14.3-32-32V288 224 64c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32zM416 224v64H224V224H416z"],"rectangle-list":[576,512,["list-alt"],"f022","M0 96C0 60.7 28.7 32 64 32H512c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM128 288a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm32-128a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM128 384a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm96-248c-13.3 0-24 10.7-24 24s10.7 24 24 24H448c13.3 0 24-10.7 24-24s-10.7-24-24-24H224zm0 96c-13.3 0-24 10.7-24 24s10.7 24 24 24H448c13.3 0 24-10.7 24-24s-10.7-24-24-24H224zm0 96c-13.3 0-24 10.7-24 24s10.7 24 24 24H448c13.3 0 24-10.7 24-24s-10.7-24-24-24H224z"],"tarp-droplet":[576,512,[],"e57c","M288 160c-35.3 0-64-26.9-64-60c0-24 33.7-70.1 52.2-93.5c6.1-7.7 17.5-7.7 23.6 0C318.3 29.9 352 76 352 100c0 33.1-28.7 60-64 60zM64 128H197.5c13.2 37.3 48.7 64 90.5 64s77.4-26.7 90.5-64H512c35.3 0 64 28.7 64 64V352H448c-17.7 0-32 14.3-32 32l0 128L64 512c-35.3 0-64-28.7-64-64V192c0-35.3 28.7-64 64-64zM448 512l0-128H576L448 512zM96 256a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"house-medical-circle-check":[640,512,[],"e511","M320 368c0 59.5 29.5 112.1 74.8 144H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L522.1 193.9c-8.5-1.3-17.3-1.9-26.1-1.9c-54.7 0-103.5 24.9-135.8 64H320V208c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16v48H208c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16zm32 0a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm211.3-43.3c-6.2-6.2-16.4-6.2-22.6 0L480 385.4l-28.7-28.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l40 40c6.2 6.2 16.4 6.2 22.6 0l72-72c6.2-6.2 6.2-16.4 0-22.6z"],"person-skiing-nordic":[576,512,["skiing-nordic"],"f7ca","M336 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM227.2 160c1.9 0 3.8 .1 5.6 .3L201.6 254c-9.3 28 1.7 58.8 26.8 74.5l86.2 53.9L291.3 464H202.8l41.1-88.1-32.4-20.3c-7.8-4.9-14.7-10.7-20.6-17.3L132.2 464H99.7l54.2-257.6c4.6-1.5 9-4.1 12.7-7.8l23.1-23.1c9.9-9.9 23.4-15.5 37.5-15.5zM121.4 198.6c.4 .4 .8 .8 1.3 1.2L67 464H24c-13.3 0-24 10.7-24 24s10.7 24 24 24H159.3c.4 0 .9 0 1.3 0H319.3c.5 0 1 0 1.4 0H504c39.8 0 72-32.2 72-72v-8c0-13.3-10.7-24-24-24s-24 10.7-24 24v8c0 13.3-10.7 24-24 24H434.6l27.6-179.3c10.5-5.2 17.8-16.1 17.8-28.7c0-17.7-14.3-32-32-32H426.7c-12.9 0-24.6-7.8-29.5-19.7l-6.3-15c-14.6-35.1-44.1-61.9-80.5-73.1l-48.7-15c-11.1-3.4-22.7-5.2-34.4-5.2c-31 0-60.8 12.3-82.7 34.3l-23.1 23.1c-12.5 12.5-12.5 32.8 0 45.3zm308 89.4L402.3 464H357.8l21.6-75.6c5.9-20.6-2.6-42.6-20.7-53.9L302 299l30.9-82.4 5.1 12.3C353 264.7 387.9 288 426.7 288h2.7z"],"calendar-plus":[448,512,[],"f271","M96 32V64H48C21.5 64 0 85.5 0 112v48H448V112c0-26.5-21.5-48-48-48H352V32c0-17.7-14.3-32-32-32s-32 14.3-32 32V64H160V32c0-17.7-14.3-32-32-32S96 14.3 96 32zM448 192H0V464c0 26.5 21.5 48 48 48H400c26.5 0 48-21.5 48-48V192zM224 248c13.3 0 24 10.7 24 24v56h56c13.3 0 24 10.7 24 24s-10.7 24-24 24H248v56c0 13.3-10.7 24-24 24s-24-10.7-24-24V376H144c-13.3 0-24-10.7-24-24s10.7-24 24-24h56V272c0-13.3 10.7-24 24-24z"],"plane-arrival":[640,512,[128748],"f5af","M.3 166.9L0 68C0 57.7 9.5 50.1 19.5 52.3l35.6 7.9c10.6 2.3 19.2 9.9 23 20L96 128l127.3 37.6L181.8 20.4C178.9 10.2 186.6 0 197.2 0h40.1c11.6 0 22.2 6.2 27.9 16.3l109 193.8 107.2 31.7c15.9 4.7 30.8 12.5 43.7 22.8l34.4 27.6c24 19.2 18.1 57.3-10.7 68.2c-41.2 15.6-86.2 18.1-128.8 7L121.7 289.8c-11.1-2.9-21.2-8.7-29.3-16.9L9.5 189.4c-5.9-6-9.3-14.1-9.3-22.5zM32 448H608c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32zm96-80a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm128-16a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"circle-left":[512,512,[61840,"arrow-alt-circle-left"],"f359","M512 256A256 256 0 1 0 0 256a256 256 0 1 0 512 0zM217.4 376.9L117.5 269.8c-3.5-3.8-5.5-8.7-5.5-13.8s2-10.1 5.5-13.8l99.9-107.1c4.2-4.5 10.1-7.1 16.3-7.1c12.3 0 22.3 10 22.3 22.3l0 57.7 96 0c17.7 0 32 14.3 32 32l0 32c0 17.7-14.3 32-32 32l-96 0 0 57.7c0 12.3-10 22.3-22.3 22.3c-6.2 0-12.1-2.6-16.3-7.1z"],"train-subway":[448,512,["subway"],"f239","M96 0C43 0 0 43 0 96V352c0 48 35.2 87.7 81.1 94.9l-46 46C28.1 499.9 33.1 512 43 512H82.7c8.5 0 16.6-3.4 22.6-9.4L160 448H288l54.6 54.6c6 6 14.1 9.4 22.6 9.4H405c10 0 15-12.1 7.9-19.1l-46-46c46-7.1 81.1-46.9 81.1-94.9V96c0-53-43-96-96-96H96zM64 128c0-17.7 14.3-32 32-32h80c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V128zM272 96h80c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H272c-17.7 0-32-14.3-32-32V128c0-17.7 14.3-32 32-32zM64 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm288-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"chart-gantt":[512,512,[],"e0e4","M32 32c17.7 0 32 14.3 32 32V400c0 8.8 7.2 16 16 16H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H80c-44.2 0-80-35.8-80-80V64C0 46.3 14.3 32 32 32zm96 96c0-17.7 14.3-32 32-32l96 0c17.7 0 32 14.3 32 32s-14.3 32-32 32H160c-17.7 0-32-14.3-32-32zm96 64H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32s14.3-32 32-32zm160 96h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],"indian-rupee-sign":[320,512,["indian-rupee","inr"],"e1bc","M0 64C0 46.3 14.3 32 32 32H96h16H288c17.7 0 32 14.3 32 32s-14.3 32-32 32H231.8c9.6 14.4 16.7 30.6 20.7 48H288c17.7 0 32 14.3 32 32s-14.3 32-32 32H252.4c-13.2 58.3-61.9 103.2-122.2 110.9L274.6 422c14.4 10.3 17.7 30.3 7.4 44.6s-30.3 17.7-44.6 7.4L13.4 314C2.1 306-2.7 291.5 1.5 278.2S18.1 256 32 256h80c32.8 0 61-19.7 73.3-48H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H185.3C173 115.7 144.8 96 112 96H96 32C14.3 96 0 81.7 0 64z"],"crop-simple":[512,512,["crop-alt"],"f565","M128 32c0-17.7-14.3-32-32-32S64 14.3 64 32V64H32C14.3 64 0 78.3 0 96s14.3 32 32 32H64V384c0 35.3 28.7 64 64 64H352V384H128V32zM384 480c0 17.7 14.3 32 32 32s32-14.3 32-32V448h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H448l0-256c0-35.3-28.7-64-64-64L160 64v64l224 0 0 352z"],"money-bill-1":[576,512,["money-bill-alt"],"f3d1","M64 64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H64zm64 320H64V320c35.3 0 64 28.7 64 64zM64 192V128h64c0 35.3-28.7 64-64 64zM448 384c0-35.3 28.7-64 64-64v64H448zm64-192c-35.3 0-64-28.7-64-64h64v64zM176 256a112 112 0 1 1 224 0 112 112 0 1 1 -224 0zm76-48c0 9.7 6.9 17.7 16 19.6V276h-4c-11 0-20 9-20 20s9 20 20 20h24 24c11 0 20-9 20-20s-9-20-20-20h-4V208c0-11-9-20-20-20H272c-11 0-20 9-20 20z"],"left-long":[512,512,["long-arrow-alt-left"],"f30a","M177.5 414c-8.8 3.8-19 2-26-4.6l-144-136C2.7 268.9 0 262.6 0 256s2.7-12.9 7.5-17.4l144-136c7-6.6 17.2-8.4 26-4.6s14.5 12.5 14.5 22l0 72 288 0c17.7 0 32 14.3 32 32l0 64c0 17.7-14.3 32-32 32l-288 0 0 72c0 9.6-5.7 18.2-14.5 22z"],dna:[448,512,[129516],"f471","M416 0c17.7 0 32 14.3 32 32c0 59.8-30.3 107.5-69.4 146.6c-28 28-62.5 53.5-97.3 77.4l-2.5 1.7c-11.9 8.1-23.8 16.1-35.5 23.9l0 0 0 0 0 0-1.6 1c-6 4-11.9 7.9-17.8 11.9c-20.9 14-40.8 27.7-59.3 41.5H283.3c-9.8-7.4-20.1-14.7-30.7-22.1l7-4.7 3-2c15.1-10.1 30.9-20.6 46.7-31.6c25 18.1 48.9 37.3 69.4 57.7C417.7 372.5 448 420.2 448 480c0 17.7-14.3 32-32 32s-32-14.3-32-32H64c0 17.7-14.3 32-32 32s-32-14.3-32-32c0-59.8 30.3-107.5 69.4-146.6c28-28 62.5-53.5 97.3-77.4c-34.8-23.9-69.3-49.3-97.3-77.4C30.3 139.5 0 91.8 0 32C0 14.3 14.3 0 32 0S64 14.3 64 32H384c0-17.7 14.3-32 32-32zM338.6 384H109.4c-10.1 10.6-18.6 21.3-25.5 32H364.1c-6.8-10.7-15.3-21.4-25.5-32zM109.4 128H338.6c10.1-10.7 18.6-21.3 25.5-32H83.9c6.8 10.7 15.3 21.3 25.5 32zm55.4 48c18.4 13.8 38.4 27.5 59.3 41.5c20.9-14 40.8-27.7 59.3-41.5H164.7z"],"virus-slash":[640,512,[],"e075","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-154.3-121c-2-30.1 20.8-60.1 56-60.1H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H532.5c-49.9 0-74.9-60.3-39.6-95.6l8.2-8.2c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-8.2 8.2C412.3 118.4 352 93.4 352 43.5V32c0-17.7-14.3-32-32-32s-32 14.3-32 32V43.5c0 49.9-60.3 74.9-95.6 39.6L184.2 75c-12.5-12.5-32.8-12.5-45.3 0c-1.6 1.6-3.1 3.4-4.3 5.3L38.8 5.1zm225.8 177c6.9-3.9 14.9-6.1 23.4-6.1c26.5 0 48 21.5 48 48c0 4.4-.6 8.7-1.7 12.7l-69.7-54.6zM402 412.7L144.7 210c-9.5 8.5-22.2 14-37.2 14H96c-17.7 0-32 14.3-32 32s14.3 32 32 32h11.5c49.9 0 74.9 60.3 39.6 95.6l-8.2 8.2c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l8.2-8.2c35.3-35.3 95.6-10.3 95.6 39.6V480c0 17.7 14.3 32 32 32s32-14.3 32-32V468.5c0-31.2 23.6-52.7 50-55.7z"],minus:[448,512,[8211,8722,10134,"subtract"],"f068","M432 256c0 17.7-14.3 32-32 32L48 288c-17.7 0-32-14.3-32-32s14.3-32 32-32l352 0c17.7 0 32 14.3 32 32z"],chess:[512,512,[],"f439","M144 16c0-8.8-7.2-16-16-16s-16 7.2-16 16V32H96c-8.8 0-16 7.2-16 16s7.2 16 16 16h16V96H60.2C49.1 96 40 105.1 40 116.2c0 2.5 .5 4.9 1.3 7.3L73.8 208H72c-13.3 0-24 10.7-24 24s10.7 24 24 24h4L60 384H196L180 256h4c13.3 0 24-10.7 24-24s-10.7-24-24-24h-1.8l32.5-84.5c.9-2.3 1.3-4.8 1.3-7.3c0-11.2-9.1-20.2-20.2-20.2H144V64h16c8.8 0 16-7.2 16-16s-7.2-16-16-16H144V16zM48 416L4.8 473.6C1.7 477.8 0 482.8 0 488c0 13.3 10.7 24 24 24H232c13.3 0 24-10.7 24-24c0-5.2-1.7-10.2-4.8-14.4L208 416H48zm288 0l-43.2 57.6c-3.1 4.2-4.8 9.2-4.8 14.4c0 13.3 10.7 24 24 24H488c13.3 0 24-10.7 24-24c0-5.2-1.7-10.2-4.8-14.4L464 416H336zM304 208v51.9c0 7.8 2.8 15.3 8 21.1L339.2 312 337 384H462.5l-3.3-72 28.3-30.8c5.4-5.9 8.5-13.6 8.5-21.7V208c0-8.8-7.2-16-16-16H464c-8.8 0-16 7.2-16 16v16H424V208c0-8.8-7.2-16-16-16H392c-8.8 0-16 7.2-16 16v16H352V208c0-8.8-7.2-16-16-16H320c-8.8 0-16 7.2-16 16zm80 96c0-8.8 7.2-16 16-16s16 7.2 16 16v32H384V304z"],"arrow-left-long":[512,512,["long-arrow-left"],"f177","M9.4 233.4c-12.5 12.5-12.5 32.8 0 45.3l128 128c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L109.3 288 480 288c17.7 0 32-14.3 32-32s-14.3-32-32-32l-370.7 0 73.4-73.4c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-128 128z"],"plug-circle-check":[576,512,[],"e55c","M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8c12.3-2.5 24.1-6.4 35.1-11.5c-2.1-10.8-3.1-21.9-3.1-33.3c0-80.3 53.8-148 127.3-169.2c.5-2.2 .7-4.5 .7-6.8c0-17.7-14.3-32-32-32H32zM576 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L416 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z"],"street-view":[512,512,[],"f21d","M320 64A64 64 0 1 0 192 64a64 64 0 1 0 128 0zm-96 96c-35.3 0-64 28.7-64 64v48c0 17.7 14.3 32 32 32h1.8l11.1 99.5c1.8 16.2 15.5 28.5 31.8 28.5h38.7c16.3 0 30-12.3 31.8-28.5L318.2 304H320c17.7 0 32-14.3 32-32V224c0-35.3-28.7-64-64-64H224zM132.3 394.2c13-2.4 21.7-14.9 19.3-27.9s-14.9-21.7-27.9-19.3c-32.4 5.9-60.9 14.2-82 24.8c-10.5 5.3-20.3 11.7-27.8 19.6C6.4 399.5 0 410.5 0 424c0 21.4 15.5 36.1 29.1 45c14.7 9.6 34.3 17.3 56.4 23.4C130.2 504.7 190.4 512 256 512s125.8-7.3 170.4-19.6c22.1-6.1 41.8-13.8 56.4-23.4c13.7-8.9 29.1-23.6 29.1-45c0-13.5-6.4-24.5-14-32.6c-7.5-7.9-17.3-14.3-27.8-19.6c-21-10.6-49.5-18.9-82-24.8c-13-2.4-25.5 6.3-27.9 19.3s6.3 25.5 19.3 27.9c30.2 5.5 53.7 12.8 69 20.5c3.2 1.6 5.8 3.1 7.9 4.5c3.6 2.4 3.6 7.2 0 9.6c-8.8 5.7-23.1 11.8-43 17.3C374.3 457 318.5 464 256 464s-118.3-7-157.7-17.9c-19.9-5.5-34.2-11.6-43-17.3c-3.6-2.4-3.6-7.2 0-9.6c2.1-1.4 4.8-2.9 7.9-4.5c15.3-7.7 38.8-14.9 69-20.5z"],"franc-sign":[320,512,[],"e18f","M80 32C62.3 32 48 46.3 48 64V224v96H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H48v64c0 17.7 14.3 32 32 32s32-14.3 32-32V384h80c17.7 0 32-14.3 32-32s-14.3-32-32-32H112V256H256c17.7 0 32-14.3 32-32s-14.3-32-32-32H112V96H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H80z"],"volume-off":[320,512,[],"f026","M320 64c0-12.6-7.4-24-18.9-29.2s-25-3.1-34.4 5.3L131.8 160H64c-35.3 0-64 28.7-64 64v64c0 35.3 28.7 64 64 64h67.8L266.7 471.9c9.4 8.4 22.9 10.4 34.4 5.3S320 460.6 320 448V64z"],"hands-asl-interpreting":[640,512,["american-sign-language-interpreting","asl-interpreting","hands-american-sign-language-interpreting"],"f2a3","M156.6 46.3c7.9-15.8 1.5-35-14.3-42.9s-35-1.5-42.9 14.3L13.5 189.4C4.6 207.2 0 226.8 0 246.7V256c0 70.7 57.3 128 128 128h72 8v-.3c35.2-2.7 65.4-22.8 82.1-51.7c8.8-15.3 3.6-34.9-11.7-43.7s-34.9-3.6-43.7 11.7c-7 12-19.9 20-34.7 20c-22.1 0-40-17.9-40-40s17.9-40 40-40c14.8 0 27.7 8 34.7 20c8.8 15.3 28.4 20.5 43.7 11.7s20.5-28.4 11.7-43.7c-12.8-22.1-33.6-39.1-58.4-47.1l80.8-22c17-4.6 27.1-22.2 22.5-39.3s-22.2-27.1-39.3-22.5L194.9 124.6l81.6-68c13.6-11.3 15.4-31.5 4.1-45.1S249.1-3.9 235.5 7.4L133.6 92.3l23-46zM483.4 465.7c-7.9 15.8-1.5 35 14.3 42.9s35 1.5 42.9-14.3l85.9-171.7c8.9-17.8 13.5-37.4 13.5-57.2V256c0-70.7-57.3-128-128-128H440h-8v.3c-35.2 2.7-65.4 22.8-82.1 51.7c-8.9 15.3-3.6 34.9 11.7 43.7s34.9 3.6 43.7-11.7c7-12 19.9-20 34.7-20c22.1 0 40 17.9 40 40s-17.9 40-40 40c-14.8 0-27.7-8-34.7-20c-8.9-15.3-28.4-20.5-43.7-11.7s-20.5 28.4-11.7 43.7c12.8 22.1 33.6 39.1 58.4 47.1l-80.8 22c-17.1 4.7-27.1 22.2-22.5 39.3s22.2 27.1 39.3 22.5l100.7-27.5-81.6 68c-13.6 11.3-15.4 31.5-4.1 45.1s31.5 15.4 45.1 4.1l101.9-84.9-23 46z"],gear:[512,512,[9881,"cog"],"f013","M495.9 166.6c3.2 8.7 .5 18.4-6.4 24.6l-43.3 39.4c1.1 8.3 1.7 16.8 1.7 25.4s-.6 17.1-1.7 25.4l43.3 39.4c6.9 6.2 9.6 15.9 6.4 24.6c-4.4 11.9-9.7 23.3-15.8 34.3l-4.7 8.1c-6.6 11-14 21.4-22.1 31.2c-5.9 7.2-15.7 9.6-24.5 6.8l-55.7-17.7c-13.4 10.3-28.2 18.9-44 25.4l-12.5 57.1c-2 9.1-9 16.3-18.2 17.8c-13.8 2.3-28 3.5-42.5 3.5s-28.7-1.2-42.5-3.5c-9.2-1.5-16.2-8.7-18.2-17.8l-12.5-57.1c-15.8-6.5-30.6-15.1-44-25.4L83.1 425.9c-8.8 2.8-18.6 .3-24.5-6.8c-8.1-9.8-15.5-20.2-22.1-31.2l-4.7-8.1c-6.1-11-11.4-22.4-15.8-34.3c-3.2-8.7-.5-18.4 6.4-24.6l43.3-39.4C64.6 273.1 64 264.6 64 256s.6-17.1 1.7-25.4L22.4 191.2c-6.9-6.2-9.6-15.9-6.4-24.6c4.4-11.9 9.7-23.3 15.8-34.3l4.7-8.1c6.6-11 14-21.4 22.1-31.2c5.9-7.2 15.7-9.6 24.5-6.8l55.7 17.7c13.4-10.3 28.2-18.9 44-25.4l12.5-57.1c2-9.1 9-16.3 18.2-17.8C227.3 1.2 241.5 0 256 0s28.7 1.2 42.5 3.5c9.2 1.5 16.2 8.7 18.2 17.8l12.5 57.1c15.8 6.5 30.6 15.1 44 25.4l55.7-17.7c8.8-2.8 18.6-.3 24.5 6.8c8.1 9.8 15.5 20.2 22.1 31.2l4.7 8.1c6.1 11 11.4 22.4 15.8 34.3zM256 336a80 80 0 1 0 0-160 80 80 0 1 0 0 160z"],"droplet-slash":[640,512,["tint-slash"],"f5c7","M320 512c53.2 0 101.4-21.6 136.1-56.6l-298.3-235C140 257.1 128 292.3 128 320c0 106 86 192 192 192zM505.2 370.7c4.4-16.1 6.8-33.1 6.8-50.7c0-91.2-130.2-262.3-166.6-308.3C339.4 4.2 330.5 0 320.9 0h-1.8c-9.6 0-18.5 4.2-24.5 11.7C277.8 33 240.7 81.3 205.8 136L38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L505.2 370.7zM224 336c0 44.2 35.8 80 80 80c8.8 0 16 7.2 16 16s-7.2 16-16 16c-61.9 0-112-50.1-112-112c0-8.8 7.2-16 16-16s16 7.2 16 16z"],mosque:[640,512,[128332],"f678","M400 0c5 0 9.8 2.4 12.8 6.4c34.7 46.3 78.1 74.9 133.5 111.5l0 0 0 0c5.2 3.4 10.5 7 16 10.6c28.9 19.2 45.7 51.7 45.7 86.1c0 28.6-11.3 54.5-29.8 73.4H221.8c-18.4-19-29.8-44.9-29.8-73.4c0-34.4 16.7-66.9 45.7-86.1c5.4-3.6 10.8-7.1 16-10.6l0 0 0 0C309.1 81.3 352.5 52.7 387.2 6.4c3-4 7.8-6.4 12.8-6.4zM288 512V440c0-13.3-10.7-24-24-24s-24 10.7-24 24v72H192c-17.7 0-32-14.3-32-32V352c0-17.7 14.3-32 32-32H608c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H560V440c0-13.3-10.7-24-24-24s-24 10.7-24 24v72H448V454c0-19-8.4-37-23-49.2L400 384l-25 20.8C360.4 417 352 435 352 454v58H288zM70.4 5.2c5.7-4.3 13.5-4.3 19.2 0l16 12C139.8 42.9 160 83.2 160 126v2H0v-2C0 83.2 20.2 42.9 54.4 17.2l16-12zM0 160H160V296.6c-19.1 11.1-32 31.7-32 55.4V480c0 9.6 2.1 18.6 5.8 26.8c-6.6 3.4-14 5.2-21.8 5.2H48c-26.5 0-48-21.5-48-48V176 160z"],mosquito:[640,512,[],"e52b","M463.7 505.9c9.8-8.9 10.7-24.3 2.1-34.3l-42.1-49 0-54.7c0-5.5-1.8-10.8-5.1-15.1L352 266.3l0-.3L485.4 387.8C542.4 447.6 640 405.2 640 320.6c0-47.9-34-88.3-79.4-94.2l-153-23.9 40.8-40.9c7.8-7.8 9.4-20.1 3.9-29.8L428.5 90.1l38.2-50.9c8-10.6 6.1-25.9-4.3-34.1s-25.2-6.3-33.2 4.4l-48 63.9c-5.9 7.9-6.6 18.6-1.7 27.2L402.2 140 352 190.3l0-38.2c0-14.9-10.2-27.4-24-31l0-57.2c0-4.4-3.6-8-8-8s-8 3.6-8 8l0 57.2c-13.8 3.6-24 16.1-24 31l0 38.1L237.8 140l22.6-39.5c4.9-8.6 4.2-19.3-1.7-27.2l-48-63.9c-8-10.6-22.8-12.6-33.2-4.4s-12.2 23.5-4.3 34.1l38.2 50.9-23.9 41.7c-5.5 9.7-3.9 22 3.9 29.8l40.8 40.9-153 23.9C34 232.3 0 272.7 0 320.6c0 84.6 97.6 127 154.6 67.1L288 266l0 .3-66.5 86.4c-3.3 4.3-5.1 9.6-5.1 15.1l0 54.7-42.1 49c-8.6 10.1-7.7 25.5 2.1 34.3s24.7 7.9 33.4-2.1l48-55.9c3.8-4.4 5.9-10.2 5.9-16.1l0-55.4L288 344.7l0 63.1c0 17.7 14.3 32 32 32s32-14.3 32-32l0-63.1 24.3 31.6 0 55.4c0 5.9 2.1 11.7 5.9 16.1l48 55.9c8.6 10.1 23.6 11 33.4 2.1z"],"star-of-david":[512,512,[10017],"f69a","M404.2 309.5L383.1 344h42.3l-21.1-34.5zM371.4 256l-54-88H194.6l-54 88 54 88H317.4l54-88zm65.7 0l53.4 87c3.6 5.9 5.5 12.7 5.5 19.6c0 20.7-16.8 37.4-37.4 37.4H348.7l-56.2 91.5C284.8 504.3 270.9 512 256 512s-28.8-7.7-36.6-20.5L163.3 400H53.4C32.8 400 16 383.2 16 362.6c0-6.9 1.9-13.7 5.5-19.6l53.4-87L21.5 169c-3.6-5.9-5.5-12.7-5.5-19.6C16 128.8 32.8 112 53.4 112H163.3l56.2-91.5C227.2 7.7 241.1 0 256 0s28.8 7.7 36.6 20.5L348.7 112H458.6c20.7 0 37.4 16.8 37.4 37.4c0 6.9-1.9 13.7-5.5 19.6l-53.4 87zm-54-88l21.1 34.5L425.4 168H383.1zM283 112L256 68l-27 44h54zM128.9 168H86.6l21.1 34.5L128.9 168zM107.8 309.5L86.6 344h42.3l-21.1-34.5zM229 400l27 44 27-44H229z"],"person-military-rifle":[512,512,[],"e54b","M160 39c0-13 10-23.8 22.9-24.9L334.7 1.4C344 .7 352 8 352 17.4V48c0 8.8-7.2 16-16 16H185c-13.8 0-25-11.2-25-25zm17.6 57H334.4c1 5.2 1.6 10.5 1.6 16c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-5.5 .6-10.8 1.6-16zm228 364.3L352 369.7V480c0 1.3-.1 2.5-.2 3.8L177.5 234.9c16.6-7.1 34.6-10.9 53.3-10.9h50.4c15.9 0 31.3 2.8 45.8 7.9L421.9 67.7c-7.7-4.4-10.3-14.2-5.9-21.9s14.2-10.3 21.9-5.9l13.9 8 13.9 8c7.7 4.4 10.3 14.2 5.9 21.9L416 173.9l1.6 .9c15.3 8.8 20.6 28.4 11.7 43.7L392.6 282c2 2.8 3.9 5.8 5.7 8.8l76.1 128.8c11.2 19 4.9 43.5-14.1 54.8s-43.5 4.9-54.8-14.1zM320 512H192c-17.7 0-32-14.3-32-32V369.7l-53.6 90.6c-11.2 19-35.8 25.3-54.8 14.1s-25.3-35.8-14.1-54.8l76.1-128.8c9.4-15.8 21.7-29.3 36-40L331.1 510c-3.5 1.3-7.2 2-11.1 2zM296 320a24 24 0 1 0 0-48 24 24 0 1 0 0 48z"],"cart-shopping":[576,512,[128722,"shopping-cart"],"f07a","M0 24C0 10.7 10.7 0 24 0H69.5c22 0 41.5 12.8 50.6 32h411c26.3 0 45.5 25 38.6 50.4l-41 152.3c-8.5 31.4-37 53.3-69.5 53.3H170.7l5.4 28.5c2.2 11.3 12.1 19.5 23.6 19.5H488c13.3 0 24 10.7 24 24s-10.7 24-24 24H199.7c-34.6 0-64.3-24.6-70.7-58.5L77.4 54.5c-.7-3.8-4-6.5-7.9-6.5H24C10.7 48 0 37.3 0 24zM128 464a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm336-48a48 48 0 1 1 0 96 48 48 0 1 1 0-96z"],vials:[512,512,[],"f493","M0 64C0 46.3 14.3 32 32 32H88h48 56c17.7 0 32 14.3 32 32s-14.3 32-32 32V400c0 44.2-35.8 80-80 80s-80-35.8-80-80V96C14.3 96 0 81.7 0 64zM136 96H88V256h48V96zM288 64c0-17.7 14.3-32 32-32h56 48 56c17.7 0 32 14.3 32 32s-14.3 32-32 32V400c0 44.2-35.8 80-80 80s-80-35.8-80-80V96c-17.7 0-32-14.3-32-32zM424 96H376V256h48V96z"],"plug-circle-plus":[576,512,[],"e55f","M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8c12.3-2.5 24.1-6.4 35.1-11.5c-2.1-10.8-3.1-21.9-3.1-33.3c0-80.3 53.8-148 127.3-169.2c.5-2.2 .7-4.5 .7-6.8c0-17.7-14.3-32-32-32H32zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm16-208v48h48c8.8 0 16 7.2 16 16s-7.2 16-16 16H448v48c0 8.8-7.2 16-16 16s-16-7.2-16-16V384H368c-8.8 0-16-7.2-16-16s7.2-16 16-16h48V304c0-8.8 7.2-16 16-16s16 7.2 16 16z"],"place-of-worship":[640,512,[],"f67f","M224 109.3V217.6L183.3 242c-14.5 8.7-23.3 24.3-23.3 41.2V512h96V416c0-35.3 28.7-64 64-64s64 28.7 64 64v96h96V283.2c0-16.9-8.8-32.5-23.3-41.2L416 217.6V109.3c0-8.5-3.4-16.6-9.4-22.6L331.3 11.3c-6.2-6.2-16.4-6.2-22.6 0L233.4 86.6c-6 6-9.4 14.1-9.4 22.6zM24.9 330.3C9.5 338.8 0 354.9 0 372.4V464c0 26.5 21.5 48 48 48h80V273.6L24.9 330.3zM592 512c26.5 0 48-21.5 48-48V372.4c0-17.5-9.5-33.6-24.9-42.1L512 273.6V512h80z"],"grip-vertical":[320,512,[],"f58e","M40 352l48 0c22.1 0 40 17.9 40 40l0 48c0 22.1-17.9 40-40 40l-48 0c-22.1 0-40-17.9-40-40l0-48c0-22.1 17.9-40 40-40zm192 0l48 0c22.1 0 40 17.9 40 40l0 48c0 22.1-17.9 40-40 40l-48 0c-22.1 0-40-17.9-40-40l0-48c0-22.1 17.9-40 40-40zM40 320c-22.1 0-40-17.9-40-40l0-48c0-22.1 17.9-40 40-40l48 0c22.1 0 40 17.9 40 40l0 48c0 22.1-17.9 40-40 40l-48 0zM232 192l48 0c22.1 0 40 17.9 40 40l0 48c0 22.1-17.9 40-40 40l-48 0c-22.1 0-40-17.9-40-40l0-48c0-22.1 17.9-40 40-40zM40 160c-22.1 0-40-17.9-40-40L0 72C0 49.9 17.9 32 40 32l48 0c22.1 0 40 17.9 40 40l0 48c0 22.1-17.9 40-40 40l-48 0zM232 32l48 0c22.1 0 40 17.9 40 40l0 48c0 22.1-17.9 40-40 40l-48 0c-22.1 0-40-17.9-40-40l0-48c0-22.1 17.9-40 40-40z"],"arrow-turn-up":[384,512,["level-up"],"f148","M32 448c-17.7 0-32 14.3-32 32s14.3 32 32 32l96 0c53 0 96-43 96-96l0-306.7 73.4 73.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-128-128c-12.5-12.5-32.8-12.5-45.3 0l-128 128c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L160 109.3 160 416c0 17.7-14.3 32-32 32l-96 0z"],u:[384,512,[117],"55","M32 32c17.7 0 32 14.3 32 32V288c0 70.7 57.3 128 128 128s128-57.3 128-128V64c0-17.7 14.3-32 32-32s32 14.3 32 32V288c0 106-86 192-192 192S0 394 0 288V64C0 46.3 14.3 32 32 32z"],"square-root-variable":[576,512,["square-root-alt"],"f698","M282.6 78.1c8-27.3 33-46.1 61.4-46.1H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H344L238.7 457c-3.6 12.3-14.1 21.2-26.8 22.8s-25.1-4.6-31.5-15.6L77.6 288H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H77.6c22.8 0 43.8 12.1 55.3 31.8l65.2 111.8L282.6 78.1zM393.4 233.4c12.5-12.5 32.8-12.5 45.3 0L480 274.7l41.4-41.4c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3L525.3 320l41.4 41.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L480 365.3l-41.4 41.4c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L434.7 320l-41.4-41.4c-12.5-12.5-12.5-32.8 0-45.3z"],clock:[512,512,[128339,"clock-four"],"f017","M256 0a256 256 0 1 1 0 512A256 256 0 1 1 256 0zM232 120V256c0 8 4 15.5 10.7 20l96 64c11 7.4 25.9 4.4 33.3-6.7s4.4-25.9-6.7-33.3L280 243.2V120c0-13.3-10.7-24-24-24s-24 10.7-24 24z"],"backward-step":[320,512,["step-backward"],"f048","M267.5 440.6c9.5 7.9 22.8 9.7 34.1 4.4s18.4-16.6 18.4-29V96c0-12.4-7.2-23.7-18.4-29s-24.5-3.6-34.1 4.4l-192 160L64 241V96c0-17.7-14.3-32-32-32S0 78.3 0 96V416c0 17.7 14.3 32 32 32s32-14.3 32-32V271l11.5 9.6 192 160z"],pallet:[640,512,[],"f482","M32 320c-17.7 0-32 14.3-32 32s14.3 32 32 32H64v64H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H96 320 544h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H576V384h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H544 320 96 32zm96 64H288v64H128V384zm224 0H512v64H352V384z"],faucet:[512,512,[],"e005","M192 96v12L96 96c-17.7 0-32 14.3-32 32s14.3 32 32 32l96-12 31-3.9 1-.1 1 .1 31 3.9 96 12c17.7 0 32-14.3 32-32s-14.3-32-32-32l-96 12V96c0-17.7-14.3-32-32-32s-32 14.3-32 32zM32 256c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H132.1c20.2 29 53.9 48 91.9 48s71.7-19 91.9-48H352c17.7 0 32 14.3 32 32s14.3 32 32 32h64c17.7 0 32-14.3 32-32c0-88.4-71.6-160-160-160H320l-22.6-22.6c-6-6-14.1-9.4-22.6-9.4H256V180.2l-32-4-32 4V224H173.3c-8.5 0-16.6 3.4-22.6 9.4L128 256H32z"],"baseball-bat-ball":[512,512,[],"f432","M424 0c-12.4 0-24.2 4.9-33 13.7L233.5 171.2c-10.5 10.5-19.8 22.1-27.7 34.6L132.7 321.6c-7.3 11.5-15.8 22.2-25.5 31.9L69.9 390.7l51.3 51.3 37.3-37.3c9.6-9.6 20.3-18.2 31.9-25.5l115.8-73.1c12.5-7.9 24.1-17.2 34.6-27.7L498.3 121c8.7-8.7 13.7-20.6 13.7-33s-4.9-24.2-13.7-33L457 13.7C448.2 4.9 436.4 0 424 0zm88 432a80 80 0 1 0 -160 0 80 80 0 1 0 160 0zM15 399c-9.4 9.4-9.4 24.6 0 33.9l64 64c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L49 399c-9.4-9.4-24.6-9.4-33.9 0z"],s:[320,512,[115],"53","M99.1 105.4C79 114 68.2 127.2 65.2 144.8c-2.4 14.1-.7 23.2 2 29.4c2.8 6.3 7.9 12.4 16.7 18.6c19.2 13.4 48.3 22.1 84.9 32.5c1 .3 1.9 .6 2.9 .8c32.7 9.3 72 20.6 100.9 40.7c15.7 10.9 29.9 25.5 38.6 45.1c8.8 19.8 10.8 42 6.6 66.3c-7.3 42.5-35.3 71.7-71.8 87.3c-35.4 15.2-79.1 17.9-123.7 10.9l-.2 0 0 0c-24-3.9-62.7-17.1-87.6-25.6c-4.8-1.7-9.2-3.1-12.8-4.3C5.1 440.8-3.9 422.7 1.6 405.9s23.7-25.8 40.5-20.3c4.9 1.6 10.2 3.4 15.9 5.4c25.4 8.6 56.4 19.2 74.4 22.1c36.8 5.7 67.5 2.5 88.5-6.5c20.1-8.6 30.8-21.8 33.9-39.4c2.4-14.1 .7-23.2-2-29.4c-2.8-6.3-7.9-12.4-16.7-18.6c-19.2-13.4-48.3-22.1-84.9-32.5c-1-.3-1.9-.6-2.9-.8c-32.7-9.3-72-20.6-100.9-40.7c-15.7-10.9-29.9-25.5-38.6-45.1c-8.8-19.8-10.8-42-6.6-66.3l31.5 5.5L2.1 133.9C9.4 91.4 37.4 62.2 73.9 46.6c35.4-15.2 79.1-17.9 123.7-10.9c13 2 52.4 9.6 66.6 13.4c17.1 4.5 27.2 22.1 22.7 39.2s-22.1 27.2-39.2 22.7c-11.2-3-48.1-10.2-60.1-12l4.9-31.5-4.9 31.5c-36.9-5.8-67.5-2.5-88.6 6.5z"],timeline:[640,512,[],"e29c","M128 72a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm32 97.3c28.3-12.3 48-40.5 48-73.3c0-44.2-35.8-80-80-80S48 51.8 48 96c0 32.8 19.7 61 48 73.3V224H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H288v54.7c-28.3 12.3-48 40.5-48 73.3c0 44.2 35.8 80 80 80s80-35.8 80-80c0-32.8-19.7-61-48-73.3V288H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H544V169.3c28.3-12.3 48-40.5 48-73.3c0-44.2-35.8-80-80-80s-80 35.8-80 80c0 32.8 19.7 61 48 73.3V224H160V169.3zM488 96a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zM320 392a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],keyboard:[576,512,[9e3],"f11c","M64 64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H64zm16 64h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V144c0-8.8 7.2-16 16-16zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm16 80h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zm80-176c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V144zm16 80h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zM160 336c0-8.8 7.2-16 16-16H400c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V336zM272 128h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V144c0-8.8 7.2-16 16-16zM256 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM368 128h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V144c0-8.8 7.2-16 16-16zM352 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V240zM464 128h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H464c-8.8 0-16-7.2-16-16V144c0-8.8 7.2-16 16-16zM448 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H464c-8.8 0-16-7.2-16-16V240zm16 80h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H464c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16z"],"caret-down":[320,512,[],"f0d7","M137.4 374.6c12.5 12.5 32.8 12.5 45.3 0l128-128c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8L32 192c-12.9 0-24.6 7.8-29.6 19.8s-2.2 25.7 6.9 34.9l128 128z"],"house-chimney-medical":[576,512,["clinic-medical"],"f7f2","M575.8 255.5c0 18-15 32.1-32 32.1h-32l.7 160.2c.2 35.5-28.5 64.3-64 64.3H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L416 100.7V64c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V185l52.8 46.4c8 7 12 15 11 24zM272 192c-8.8 0-16 7.2-16 16v48H208c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V320h48c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H320V208c0-8.8-7.2-16-16-16H272z"],"temperature-three-quarters":[320,512,["temperature-3","thermometer-3","thermometer-three-quarters"],"f2c8","M160 64c-26.5 0-48 21.5-48 48V276.5c0 17.3-7.1 31.9-15.3 42.5C86.2 332.6 80 349.5 80 368c0 44.2 35.8 80 80 80s80-35.8 80-80c0-18.5-6.2-35.4-16.7-48.9c-8.2-10.6-15.3-25.2-15.3-42.5V112c0-26.5-21.5-48-48-48zM48 112C48 50.2 98.1 0 160 0s112 50.1 112 112V276.5c0 .1 .1 .3 .2 .6c.2 .6 .8 1.6 1.7 2.8c18.9 24.4 30.1 55 30.1 88.1c0 79.5-64.5 144-144 144S16 447.5 16 368c0-33.2 11.2-63.8 30.1-88.1c.9-1.2 1.5-2.2 1.7-2.8c.1-.3 .2-.5 .2-.6V112zM208 368c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-20.9 13.4-38.7 32-45.3V144c0-8.8 7.2-16 16-16s16 7.2 16 16V322.7c18.6 6.6 32 24.4 32 45.3z"],"mobile-screen":[384,512,["mobile-android-alt"],"f3cf","M16 64C16 28.7 44.7 0 80 0H304c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H80c-35.3 0-64-28.7-64-64V64zM144 448c0 8.8 7.2 16 16 16h64c8.8 0 16-7.2 16-16s-7.2-16-16-16H160c-8.8 0-16 7.2-16 16zM304 64H80V384H304V64z"],"plane-up":[512,512,[],"e22d","M192 93.7C192 59.5 221 0 256 0c36 0 64 59.5 64 93.7l0 66.3L497.8 278.5c8.9 5.9 14.2 15.9 14.2 26.6v56.7c0 10.9-10.7 18.6-21.1 15.2L320 320v80l57.6 43.2c4 3 6.4 7.8 6.4 12.8v42c0 7.8-6.3 14-14 14c-1.3 0-2.6-.2-3.9-.5L256 480 145.9 511.5c-1.3 .4-2.6 .5-3.9 .5c-7.8 0-14-6.3-14-14V456c0-5 2.4-9.8 6.4-12.8L192 400V320L21.1 377C10.7 380.4 0 372.7 0 361.8V305.1c0-10.7 5.3-20.7 14.2-26.6L192 160V93.7z"],"piggy-bank":[576,512,[],"f4d3","M400 96l0 .7c-5.3-.4-10.6-.7-16-.7H256c-16.5 0-32.5 2.1-47.8 6c-.1-2-.2-4-.2-6c0-53 43-96 96-96s96 43 96 96zm-16 32c3.5 0 7 .1 10.4 .3c4.2 .3 8.4 .7 12.6 1.3C424.6 109.1 450.8 96 480 96h11.5c10.4 0 18 9.8 15.5 19.9l-13.8 55.2c15.8 14.8 28.7 32.8 37.5 52.9H544c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H512c-9.1 12.1-19.9 22.9-32 32v64c0 17.7-14.3 32-32 32H416c-17.7 0-32-14.3-32-32V448H256v32c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32V416c-34.9-26.2-58.7-66.3-63.2-112H68c-37.6 0-68-30.4-68-68s30.4-68 68-68h4c13.3 0 24 10.7 24 24s-10.7 24-24 24H68c-11 0-20 9-20 20s9 20 20 20H99.2c12.1-59.8 57.7-107.5 116.3-122.8c12.9-3.4 26.5-5.2 40.5-5.2H384zm64 136a24 24 0 1 0 -48 0 24 24 0 1 0 48 0z"],"battery-half":[576,512,["battery-3"],"f242","M464 160c8.8 0 16 7.2 16 16V336c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16H464zM80 96C35.8 96 0 131.8 0 176V336c0 44.2 35.8 80 80 80H464c44.2 0 80-35.8 80-80V320c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32V176c0-44.2-35.8-80-80-80H80zm208 96H96V320H288V192z"],"mountain-city":[640,512,[],"e52e","M336 0c-26.5 0-48 21.5-48 48v92.1l71.4 118.4c2.5-1.6 5.4-2.5 8.6-2.5h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16h-3.5l73.8 122.4c12.4 20.6 12.9 46.3 1.2 67.3c-.4 .8-.9 1.6-1.4 2.3H592c26.5 0 48-21.5 48-48V240c0-26.5-21.5-48-48-48H568V120c0-13.3-10.7-24-24-24s-24 10.7-24 24v72H480V48c0-26.5-21.5-48-48-48H336zm32 64h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V80c0-8.8 7.2-16 16-16zM352 176c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V176zm160 96c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H528c-8.8 0-16-7.2-16-16V272zm16 80h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H528c-8.8 0-16-7.2-16-16V368c0-8.8 7.2-16 16-16zM224 188.9L283.8 288H223l-48 64-24.6-41.2L224 188.9zm29.4-44.2C247.1 134.3 236 128 224 128s-23.1 6.3-29.4 16.7L5.1 458.9c-6.5 10.8-6.7 24.3-.7 35.3S22 512 34.5 512H413.5c12.5 0 24-6.8 30.1-17.8s5.8-24.5-.7-35.3L253.4 144.7z"],coins:[512,512,[],"f51e","M512 80c0 18-14.3 34.6-38.4 48c-29.1 16.1-72.5 27.5-122.3 30.9c-3.7-1.8-7.4-3.5-11.3-5C300.6 137.4 248.2 128 192 128c-8.3 0-16.4 .2-24.5 .6l-1.1-.6C142.3 114.6 128 98 128 80c0-44.2 86-80 192-80S512 35.8 512 80zM160.7 161.1c10.2-.7 20.7-1.1 31.3-1.1c62.2 0 117.4 12.3 152.5 31.4C369.3 204.9 384 221.7 384 240c0 4-.7 7.9-2.1 11.7c-4.6 13.2-17 25.3-35 35.5c0 0 0 0 0 0c-.1 .1-.3 .1-.4 .2l0 0 0 0c-.3 .2-.6 .3-.9 .5c-35 19.4-90.8 32-153.6 32c-59.6 0-112.9-11.3-148.2-29.1c-1.9-.9-3.7-1.9-5.5-2.9C14.3 274.6 0 258 0 240c0-34.8 53.4-64.5 128-75.4c10.5-1.5 21.4-2.7 32.7-3.5zM416 240c0-21.9-10.6-39.9-24.1-53.4c28.3-4.4 54.2-11.4 76.2-20.5c16.3-6.8 31.5-15.2 43.9-25.5V176c0 19.3-16.5 37.1-43.8 50.9c-14.6 7.4-32.4 13.7-52.4 18.5c.1-1.8 .2-3.5 .2-5.3zm-32 96c0 18-14.3 34.6-38.4 48c-1.8 1-3.6 1.9-5.5 2.9C304.9 404.7 251.6 416 192 416c-62.8 0-118.6-12.6-153.6-32C14.3 370.6 0 354 0 336V300.6c12.5 10.3 27.6 18.7 43.9 25.5C83.4 342.6 135.8 352 192 352s108.6-9.4 148.1-25.9c7.8-3.2 15.3-6.9 22.4-10.9c6.1-3.4 11.8-7.2 17.2-11.2c1.5-1.1 2.9-2.3 4.3-3.4V304v5.7V336zm32 0V304 278.1c19-4.2 36.5-9.5 52.1-16c16.3-6.8 31.5-15.2 43.9-25.5V272c0 10.5-5 21-14.9 30.9c-16.3 16.3-45 29.7-81.3 38.4c.1-1.7 .2-3.5 .2-5.3zM192 448c56.2 0 108.6-9.4 148.1-25.9c16.3-6.8 31.5-15.2 43.9-25.5V432c0 44.2-86 80-192 80S0 476.2 0 432V396.6c12.5 10.3 27.6 18.7 43.9 25.5C83.4 438.6 135.8 448 192 448z"],khanda:[512,512,[9772],"f66d","M245.8 3.7c5.9-4.9 14.6-4.9 20.5 0l48 40c5.9 4.9 7.5 13.2 3.8 19.9l0 0 0 0 0 0 0 0-.1 .1-.3 .6c-.3 .5-.7 1.3-1.2 2.3c-1 2-2.6 5-4.4 8.6c-.5 .9-.9 1.9-1.4 2.9C344.9 97.4 368 134 368 176s-23.1 78.6-57.3 97.8c.5 1 1 2 1.4 2.9c1.8 3.7 3.3 6.6 4.4 8.6c.5 1 .9 1.8 1.2 2.3l.3 .6 .1 .1 0 0 0 0c3.6 6.7 2 15-3.8 19.9L272 343.5v19.8l35.6-24.5 41.1-28.2c42.8-29.4 68.4-78 68.4-130c0-31.1-9.2-61.6-26.5-87.5l-2.8-4.2c-4-6-3.5-14 1.3-19.5s12.7-7 19.2-3.7L401.1 80c7.2-14.3 7.2-14.3 7.2-14.3l0 0 0 0 .1 0 .3 .2 1 .5c.8 .4 2 1.1 3.5 1.9c2.9 1.7 7 4.1 11.8 7.3c9.6 6.4 22.5 16.1 35.4 29c25.7 25.7 52.7 65.6 52.7 119.3c0 53.1-26.4 100.5-51.2 133.6c-12.6 16.7-25.1 30.3-34.5 39.7c-4.7 4.7-8.7 8.4-11.5 10.9c-1.4 1.3-2.5 2.2-3.3 2.9l-.9 .8-.3 .2-.1 .1 0 0 0 0s0 0-10.2-12.3l10.2 12.3c-5.1 4.3-12.4 4.9-18.2 1.6l-75.6-43-32.7 22.5 45.5 31.3c1.8-.4 3.7-.7 5.7-.7c13.3 0 24 10.7 24 24s-10.7 24-24 24c-12.2 0-22.3-9.1-23.8-21L272 423.4v28.9c9.6 5.5 16 15.9 16 27.7c0 17.7-14.3 32-32 32s-32-14.3-32-32c0-11.8 6.4-22.2 16-27.7V424.1l-40.3 27.7C197.8 463.3 187.9 472 176 472c-13.3 0-24-10.7-24-24s10.7-24 24-24c2.2 0 4.4 .3 6.5 .9l45.8-31.5-32.7-22.5-75.6 43c-5.8 3.3-13 2.7-18.2-1.6L112 400c-10.2 12.3-10.2 12.3-10.3 12.3l0 0 0 0-.1-.1-.3-.2-.9-.8c-.8-.7-1.9-1.7-3.3-2.9c-2.8-2.5-6.7-6.2-11.5-10.9c-9.4-9.4-21.9-23-34.5-39.7C26.4 324.5 0 277.1 0 224c0-53.7 26.9-93.6 52.7-119.3c12.9-12.9 25.8-22.6 35.4-29C93 72.5 97 70 99.9 68.4c1.5-.8 2.6-1.5 3.5-1.9l1-.5 .3-.2 .1 0 0 0 0 0s0 0 7.2 14.3l-7.2-14.3c6.5-3.2 14.3-1.7 19.2 3.7s5.3 13.4 1.3 19.5l-2.8 4.2C105.2 119 96 149.5 96 180.6c0 51.9 25.6 100.6 68.4 130l41.1 28.2L240 362.6V343.5l-42.2-35.2c-5.9-4.9-7.5-13.2-3.8-19.9l0 0 0 0 0 0 .1-.1 .3-.6c.3-.5 .7-1.3 1.2-2.3c1-2 2.6-5 4.4-8.6c.5-.9 .9-1.9 1.4-2.9C167.1 254.6 144 218 144 176s23.1-78.6 57.3-97.8c-.5-1-1-2-1.4-2.9c-1.8-3.7-3.3-6.6-4.4-8.6c-.5-1-.9-1.8-1.2-2.3l-.3-.6-.1-.1 0 0 0 0 0 0c-3.6-6.7-2-15 3.8-19.9l48-40zM220.2 122.9c-17 11.5-28.2 31-28.2 53.1s11.2 41.6 28.2 53.1C227 210.2 232 190.9 232 176s-5-34.2-11.8-53.1zm71.5 106.2c17-11.5 28.2-31 28.2-53.1s-11.2-41.6-28.2-53.1C285 141.8 280 161.1 280 176s5 34.2 11.8 53.1z"],sliders:[512,512,["sliders-h"],"f1de","M0 416c0 17.7 14.3 32 32 32l54.7 0c12.3 28.3 40.5 48 73.3 48s61-19.7 73.3-48L480 448c17.7 0 32-14.3 32-32s-14.3-32-32-32l-246.7 0c-12.3-28.3-40.5-48-73.3-48s-61 19.7-73.3 48L32 384c-17.7 0-32 14.3-32 32zm128 0a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM320 256a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm32-80c-32.8 0-61 19.7-73.3 48L32 224c-17.7 0-32 14.3-32 32s14.3 32 32 32l246.7 0c12.3 28.3 40.5 48 73.3 48s61-19.7 73.3-48l54.7 0c17.7 0 32-14.3 32-32s-14.3-32-32-32l-54.7 0c-12.3-28.3-40.5-48-73.3-48zM192 128a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm73.3-64C253 35.7 224.8 16 192 16s-61 19.7-73.3 48L32 64C14.3 64 0 78.3 0 96s14.3 32 32 32l86.7 0c12.3 28.3 40.5 48 73.3 48s61-19.7 73.3-48L480 128c17.7 0 32-14.3 32-32s-14.3-32-32-32L265.3 64z"],"folder-tree":[576,512,[],"f802","M64 32C64 14.3 49.7 0 32 0S0 14.3 0 32v96V384c0 35.3 28.7 64 64 64H256V384H64V160H256V96H64V32zM288 192c0 17.7 14.3 32 32 32H544c17.7 0 32-14.3 32-32V64c0-17.7-14.3-32-32-32H445.3c-8.5 0-16.6-3.4-22.6-9.4L409.4 9.4c-6-6-14.1-9.4-22.6-9.4H320c-17.7 0-32 14.3-32 32V192zm0 288c0 17.7 14.3 32 32 32H544c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32H445.3c-8.5 0-16.6-3.4-22.6-9.4l-13.3-13.3c-6-6-14.1-9.4-22.6-9.4H320c-17.7 0-32 14.3-32 32V480z"],"network-wired":[640,512,[],"f6ff","M256 64H384v64H256V64zM240 0c-26.5 0-48 21.5-48 48v96c0 26.5 21.5 48 48 48h48v32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32h96v32H80c-26.5 0-48 21.5-48 48v96c0 26.5 21.5 48 48 48H240c26.5 0 48-21.5 48-48V368c0-26.5-21.5-48-48-48H192V288H448v32H400c-26.5 0-48 21.5-48 48v96c0 26.5 21.5 48 48 48H560c26.5 0 48-21.5 48-48V368c0-26.5-21.5-48-48-48H512V288h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H352V192h48c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H240zM96 448V384H224v64H96zm320-64H544v64H416V384z"],"map-pin":[320,512,[128205],"f276","M16 144a144 144 0 1 1 288 0A144 144 0 1 1 16 144zM160 80c8.8 0 16-7.2 16-16s-7.2-16-16-16c-53 0-96 43-96 96c0 8.8 7.2 16 16 16s16-7.2 16-16c0-35.3 28.7-64 64-64zM128 480V317.1c10.4 1.9 21.1 2.9 32 2.9s21.6-1 32-2.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32z"],hamsa:[512,512,[],"f665","M34.6 288H80c8.8 0 16-7.2 16-16V72c0-22.1 17.9-40 40-40s40 17.9 40 40V204c0 11 9 20 20 20s20-9 20-20V40c0-22.1 17.9-40 40-40s40 17.9 40 40V204c0 11 9 20 20 20s20-9 20-20V72c0-22.1 17.9-40 40-40s40 17.9 40 40V272c0 8.8 7.2 16 16 16h45.4c19.1 0 34.6 15.5 34.6 34.6c0 8.6-3.2 16.9-9 23.3L416.6 441c-41.1 45.2-99.4 71-160.6 71s-119.4-25.8-160.6-71L9 345.9c-5.8-6.4-9-14.7-9-23.3C0 303.5 15.5 288 34.6 288zM256 288c-38.4 0-76.8 35.8-90.6 50.2c-3.6 3.7-5.4 8.7-5.4 13.8s1.8 10.1 5.4 13.8C179.2 380.2 217.6 416 256 416s76.8-35.8 90.6-50.2c3.6-3.7 5.4-8.7 5.4-13.8s-1.8-10.1-5.4-13.8C332.8 323.8 294.4 288 256 288zm0 32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"cent-sign":[384,512,[],"e3f5","M224 0c17.7 0 32 14.3 32 32V66.7c30.9 5.2 59.2 17.7 83.2 35.8c14.1 10.6 17 30.7 6.4 44.8s-30.7 17-44.8 6.4C279.4 137.5 252.9 128 224 128c-70.7 0-128 57.3-128 128s57.3 128 128 128c28.9 0 55.4-9.5 76.8-25.6c14.1-10.6 34.2-7.8 44.8 6.4s7.8 34.2-6.4 44.8c-24 18-52.4 30.6-83.2 35.8V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V445.3C101.2 430.1 32 351.1 32 256s69.2-174.1 160-189.3V32c0-17.7 14.3-32 32-32z"],flask:[448,512,[],"f0c3","M288 0H160 128C110.3 0 96 14.3 96 32s14.3 32 32 32V196.8c0 11.8-3.3 23.5-9.5 33.5L10.3 406.2C3.6 417.2 0 429.7 0 442.6C0 480.9 31.1 512 69.4 512H378.6c38.3 0 69.4-31.1 69.4-69.4c0-12.8-3.6-25.4-10.3-36.4L329.5 230.4c-6.2-10.1-9.5-21.7-9.5-33.5V64c17.7 0 32-14.3 32-32s-14.3-32-32-32H288zM192 196.8V64h64V196.8c0 23.7 6.6 46.9 19 67.1L309.5 320h-171L173 263.9c12.4-20.2 19-43.4 19-67.1z"],"person-pregnant":[384,512,[],"e31e","M192 0a48 48 0 1 1 0 96 48 48 0 1 1 0-96zM120 383c-13.8-3.6-24-16.1-24-31V296.9l-4.6 7.6c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c15-24.9 40.3-41.5 68.7-45.6c4.1-.6 8.2-1 12.5-1h1.1 12.5H192c1.4 0 2.8 .1 4.1 .3c35.7 2.9 65.4 29.3 72.1 65l6.1 32.5c44.3 8.6 77.7 47.5 77.7 94.3v32c0 17.7-14.3 32-32 32H304 264v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384h-8-8v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V383z"],"wand-sparkles":[512,512,[],"f72b","M464 6.1c9.5-8.5 24-8.1 33 .9l8 8c9 9 9.4 23.5 .9 33l-85.8 95.9c-2.6 2.9-4.1 6.7-4.1 10.7V176c0 8.8-7.2 16-16 16H384.2c-4.6 0-8.9 1.9-11.9 5.3L100.7 500.9C94.3 508 85.3 512 75.8 512c-8.8 0-17.3-3.5-23.5-9.8L9.7 459.7C3.5 453.4 0 445 0 436.2c0-9.5 4-18.5 11.1-24.8l111.6-99.8c3.4-3 5.3-7.4 5.3-11.9V272c0-8.8 7.2-16 16-16h34.6c3.9 0 7.7-1.5 10.7-4.1L464 6.1zM432 288c3.6 0 6.7 2.4 7.7 5.8l14.8 51.7 51.7 14.8c3.4 1 5.8 4.1 5.8 7.7s-2.4 6.7-5.8 7.7l-51.7 14.8-14.8 51.7c-1 3.4-4.1 5.8-7.7 5.8s-6.7-2.4-7.7-5.8l-14.8-51.7-51.7-14.8c-3.4-1-5.8-4.1-5.8-7.7s2.4-6.7 5.8-7.7l51.7-14.8 14.8-51.7c1-3.4 4.1-5.8 7.7-5.8zM87.7 69.8l14.8 51.7 51.7 14.8c3.4 1 5.8 4.1 5.8 7.7s-2.4 6.7-5.8 7.7l-51.7 14.8L87.7 218.2c-1 3.4-4.1 5.8-7.7 5.8s-6.7-2.4-7.7-5.8L57.5 166.5 5.8 151.7c-3.4-1-5.8-4.1-5.8-7.7s2.4-6.7 5.8-7.7l51.7-14.8L72.3 69.8c1-3.4 4.1-5.8 7.7-5.8s6.7 2.4 7.7 5.8zM208 0c3.7 0 6.9 2.5 7.8 6.1l6.8 27.3 27.3 6.8c3.6 .9 6.1 4.1 6.1 7.8s-2.5 6.9-6.1 7.8l-27.3 6.8-6.8 27.3c-.9 3.6-4.1 6.1-7.8 6.1s-6.9-2.5-7.8-6.1l-6.8-27.3-27.3-6.8c-3.6-.9-6.1-4.1-6.1-7.8s2.5-6.9 6.1-7.8l27.3-6.8 6.8-27.3c.9-3.6 4.1-6.1 7.8-6.1z"],"ellipsis-vertical":[128,512,["ellipsis-v"],"f142","M64 360a56 56 0 1 0 0 112 56 56 0 1 0 0-112zm0-160a56 56 0 1 0 0 112 56 56 0 1 0 0-112zM120 96A56 56 0 1 0 8 96a56 56 0 1 0 112 0z"],ticket:[576,512,[127903],"f145","M64 64C28.7 64 0 92.7 0 128v64c0 8.8 7.4 15.7 15.7 18.6C34.5 217.1 48 235 48 256s-13.5 38.9-32.3 45.4C7.4 304.3 0 311.2 0 320v64c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V320c0-8.8-7.4-15.7-15.7-18.6C541.5 294.9 528 277 528 256s13.5-38.9 32.3-45.4c8.3-2.9 15.7-9.8 15.7-18.6V128c0-35.3-28.7-64-64-64H64zm64 112l0 160c0 8.8 7.2 16 16 16H432c8.8 0 16-7.2 16-16V176c0-8.8-7.2-16-16-16H144c-8.8 0-16 7.2-16 16zM96 160c0-17.7 14.3-32 32-32H448c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32H128c-17.7 0-32-14.3-32-32V160z"],"power-off":[512,512,[9211],"f011","M288 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V256c0 17.7 14.3 32 32 32s32-14.3 32-32V32zM143.5 120.6c13.6-11.3 15.4-31.5 4.1-45.1s-31.5-15.4-45.1-4.1C49.7 115.4 16 181.8 16 256c0 132.5 107.5 240 240 240s240-107.5 240-240c0-74.2-33.8-140.6-86.6-184.6c-13.6-11.3-33.8-9.4-45.1 4.1s-9.4 33.8 4.1 45.1c38.9 32.3 63.5 81 63.5 135.4c0 97.2-78.8 176-176 176s-176-78.8-176-176c0-54.4 24.7-103.1 63.5-135.4z"],"right-long":[512,512,["long-arrow-alt-right"],"f30b","M334.5 414c8.8 3.8 19 2 26-4.6l144-136c4.8-4.5 7.5-10.8 7.5-17.4s-2.7-12.9-7.5-17.4l-144-136c-7-6.6-17.2-8.4-26-4.6s-14.5 12.5-14.5 22l0 72L32 192c-17.7 0-32 14.3-32 32l0 64c0 17.7 14.3 32 32 32l288 0 0 72c0 9.6 5.7 18.2 14.5 22z"],"flag-usa":[448,512,[],"f74d","M32 0C49.7 0 64 14.3 64 32V48l69-17.2c38.1-9.5 78.3-5.1 113.5 12.5c46.3 23.2 100.8 23.2 147.1 0l9.6-4.8C423.8 28.1 448 43.1 448 66.1v36.1l-44.7 16.2c-42.8 15.6-90 13.9-131.6-4.6l-16.1-7.2c-20.3-9-41.8-14.7-63.6-16.9v32.2c17.4 2.1 34.4 6.7 50.6 13.9l16.1 7.2c49.2 21.9 105 23.8 155.6 5.4L448 136.3v62l-44.7 16.2c-42.8 15.6-90 13.9-131.6-4.6l-16.1-7.2c-40.2-17.9-85-22.5-128.1-13.3L64 203.1v32.7l70.2-15.1c36.4-7.8 74.3-3.9 108.4 11.3l16.1 7.2c49.2 21.9 105 23.8 155.6 5.4L448 232.3v62l-44.7 16.2c-42.8 15.6-90 13.9-131.6-4.6l-16.1-7.2c-40.2-17.9-85-22.5-128.1-13.3L64 299.1v32.7l70.2-15.1c36.4-7.8 74.3-3.9 108.4 11.3l16.1 7.2c49.2 21.9 105 23.8 155.6 5.4L448 328.3v33.5c0 13.3-8.3 25.3-20.8 30l-34.7 13c-46.2 17.3-97.6 14.6-141.7-7.4c-37.9-19-81.3-23.7-122.5-13.4L64 400v80c0 17.7-14.3 32-32 32s-32-14.3-32-32V416 345.5 312.8 249.5 216.8 153.5 120.8 64 32C0 14.3 14.3 0 32 0zm80 96A16 16 0 1 0 80 96a16 16 0 1 0 32 0zm32 0a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm-32 48a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm32 0a16 16 0 1 0 0-32 16 16 0 1 0 0 32z"],"laptop-file":[640,512,[],"e51d","M128 0C92.7 0 64 28.7 64 64V288H19.2C8.6 288 0 296.6 0 307.2C0 349.6 34.4 384 76.8 384H320V288H128V64H448V96h64V64c0-35.3-28.7-64-64-64H128zM512 128H400c-26.5 0-48 21.5-48 48V464c0 26.5 21.5 48 48 48H592c26.5 0 48-21.5 48-48V256H544c-17.7 0-32-14.3-32-32V128zm32 0v96h96l-96-96z"],tty:[512,512,["teletype"],"f1e4","M38.3 241.3L15.1 200.6c-9.2-16.2-8.4-36.5 4.5-50C61.4 106.8 144.7 48 256 48s194.6 58.8 236.4 102.6c12.9 13.5 13.7 33.8 4.5 50l-23.1 40.7c-7.5 13.2-23.3 19.3-37.8 14.6l-81.1-26.6c-13.1-4.3-22-16.6-22-30.4V144c-49.6-18.1-104-18.1-153.6 0v54.8c0 13.8-8.9 26.1-22 30.4L76.1 255.8c-14.5 4.7-30.3-1.4-37.8-14.6zM32 336c0-8.8 7.2-16 16-16H80c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H48c-8.8 0-16-7.2-16-16V336zm0 96c0-8.8 7.2-16 16-16H80c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H48c-8.8 0-16-7.2-16-16V432zM144 320h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H144c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H240c-8.8 0-16-7.2-16-16V336zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H336c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H432c-8.8 0-16-7.2-16-16V336zm16 80h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H432c-8.8 0-16-7.2-16-16V432c0-8.8 7.2-16 16-16zM128 432c0-8.8 7.2-16 16-16H368c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H144c-8.8 0-16-7.2-16-16V432z"],"diagram-next":[512,512,[],"e476","M512 160c0 35.3-28.7 64-64 64H280v64h46.1c21.4 0 32.1 25.9 17 41L273 399c-9.4 9.4-24.6 9.4-33.9 0L169 329c-15.1-15.1-4.4-41 17-41H232V224H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64v64zM448 416V352H365.3l.4-.4c18.4-18.4 20.4-43.7 11-63.6l71.3 0c35.3 0 64 28.7 64 64v64c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64V352c0-35.3 28.7-64 64-64l71.3 0c-9.4 19.9-7.4 45.2 11 63.6l.4 .4H64v64H210.7l5.7 5.7c21.9 21.9 57.3 21.9 79.2 0l5.7-5.7H448z"],"person-rifle":[576,512,[],"e54e","M265.2 192c25.4 0 49.8 7.1 70.8 19.9V512H144V337.7L90.4 428.3c-11.2 19-35.8 25.3-54.8 14.1s-25.3-35.8-14.1-54.8L97.7 258.8c24.5-41.4 69-66.8 117.1-66.8h50.4zM160 80a80 80 0 1 1 160 0A80 80 0 1 1 160 80zM448 0c8.8 0 16 7.2 16 16V132.3c9.6 5.5 16 15.9 16 27.7V269.3l16-5.3V208c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16v84.5c0 6.9-4.4 13-10.9 15.2L480 325.3V352h48c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H484l23 92.1c2.5 10.1-5.1 19.9-15.5 19.9H432c-8.8 0-16-7.2-16-16V400H400c-17.7 0-32-14.3-32-32V224c0-17.7 14.3-32 32-32V160c0-11.8 6.4-22.2 16-27.7V32c-8.8 0-16-7.2-16-16s7.2-16 16-16h16 16z"],"house-medical-circle-exclamation":[640,512,[],"e512","M320 368c0 59.5 29.5 112.1 74.8 144H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L522.1 193.9c-8.5-1.3-17.3-1.9-26.1-1.9c-54.7 0-103.5 24.9-135.8 64H320V208c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16v48H208c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16zM496 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm0 240a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm0-192c-8.8 0-16 7.2-16 16v80c0 8.8 7.2 16 16 16s16-7.2 16-16V288c0-8.8-7.2-16-16-16z"],"closed-captioning":[576,512,[],"f20a","M0 96C0 60.7 28.7 32 64 32H512c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM200 208c14.2 0 27 6.1 35.8 16c8.8 9.9 24 10.7 33.9 1.9s10.7-24 1.9-33.9c-17.5-19.6-43.1-32-71.5-32c-53 0-96 43-96 96s43 96 96 96c28.4 0 54-12.4 71.5-32c8.8-9.9 8-25-1.9-33.9s-25-8-33.9 1.9c-8.8 9.9-21.6 16-35.8 16c-26.5 0-48-21.5-48-48s21.5-48 48-48zm144 48c0-26.5 21.5-48 48-48c14.2 0 27 6.1 35.8 16c8.8 9.9 24 10.7 33.9 1.9s10.7-24 1.9-33.9c-17.5-19.6-43.1-32-71.5-32c-53 0-96 43-96 96s43 96 96 96c28.4 0 54-12.4 71.5-32c8.8-9.9 8-25-1.9-33.9s-25-8-33.9 1.9c-8.8 9.9-21.6 16-35.8 16c-26.5 0-48-21.5-48-48z"],"person-hiking":[384,512,["hiking"],"f6ec","M192 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm51.3 182.7L224.2 307l49.7 49.7c9 9 14.1 21.2 14.1 33.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V397.3l-73.9-73.9c-15.8-15.8-22.2-38.6-16.9-60.3l20.4-84c8.3-34.1 42.7-54.9 76.7-46.4c19 4.8 35.6 16.4 46.4 32.7L305.1 208H336V184c0-13.3 10.7-24 24-24s24 10.7 24 24v55.8c0 .1 0 .2 0 .2s0 .2 0 .2V488c0 13.3-10.7 24-24 24s-24-10.7-24-24V272H296.6c-16 0-31-8-39.9-21.4l-13.3-20zM81.1 471.9L117.3 334c3 4.2 6.4 8.2 10.1 11.9l41.9 41.9L142.9 488.1c-4.5 17.1-22 27.3-39.1 22.8s-27.3-22-22.8-39.1zm55.5-346L101.4 266.5c-3 12.1-14.9 19.9-27.2 17.9l-47.9-8c-14-2.3-22.9-16.3-19.2-30L31.9 155c9.5-34.8 41.1-59 77.2-59h4.2c15.6 0 27.1 14.7 23.3 29.8z"],"venus-double":[640,512,[9890],"f226","M192 288a112 112 0 1 0 0-224 112 112 0 1 0 0 224zM368 176c0 86.3-62.1 158.1-144 173.1V384h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H224v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H128c-17.7 0-32-14.3-32-32s14.3-32 32-32h32V349.1C78.1 334.1 16 262.3 16 176C16 78.8 94.8 0 192 0s176 78.8 176 176zM344 318c14.6-15.6 26.8-33.4 36-53c18.8 14.4 42.4 23 68 23c61.9 0 112-50.1 112-112s-50.1-112-112-112c-25.6 0-49.1 8.6-68 23c-9.3-19.5-21.5-37.4-36-53C373.1 12.6 409.1 0 448 0c97.2 0 176 78.8 176 176c0 86.3-62.1 158.1-144 173.1V384h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H480v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H384c-17.7 0-32-14.3-32-32s14.3-32 32-32h32V349.1c-26.6-4.9-51.1-15.7-72-31.1z"],images:[576,512,[],"f302","M160 32c-35.3 0-64 28.7-64 64V320c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H160zM396 138.7l96 144c4.9 7.4 5.4 16.8 1.2 24.6S480.9 320 472 320H328 280 200c-9.2 0-17.6-5.3-21.6-13.6s-2.9-18.2 2.9-25.4l64-80c4.6-5.7 11.4-9 18.7-9s14.2 3.3 18.7 9l17.3 21.6 56-84C360.5 132 368 128 376 128s15.5 4 20 10.7zM192 128a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM48 120c0-13.3-10.7-24-24-24S0 106.7 0 120V344c0 75.1 60.9 136 136 136H456c13.3 0 24-10.7 24-24s-10.7-24-24-24H136c-48.6 0-88-39.4-88-88V120z"],calculator:[384,512,[128425],"f1ec","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zM96 64H288c17.7 0 32 14.3 32 32v32c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V96c0-17.7 14.3-32 32-32zm32 160a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zM96 352a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM64 416c0-17.7 14.3-32 32-32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H96c-17.7 0-32-14.3-32-32zM192 256a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm32 64a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm64-64a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm32 64a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zM288 448a32 32 0 1 1 0-64 32 32 0 1 1 0 64z"],"people-pulling":[576,512,[],"e535","M80 96A48 48 0 1 0 80 0a48 48 0 1 0 0 96zM64 128c-35.3 0-64 28.7-64 64V320c0 17.7 14.3 32 32 32c9.8 0 18.5-4.4 24.4-11.2L80.4 485.3c2.9 17.4 19.4 29.2 36.8 26.3s29.2-19.4 26.3-36.8L123.1 352h15.7l30 134.9c3.8 17.3 20.9 28.1 38.2 24.3s28.1-20.9 24.3-38.2l-57.3-258 116.3 53.8c.5 .3 1.1 .5 1.6 .7c8.6 3.6 18 3.1 25.9-.7c3.4-1.6 6.6-3.9 9.3-6.7c3.1-3.2 5.5-7 7.1-11.4c.1-.3 .2-.7 .3-1l2.5-7.5c5.7-17.1 18.3-30.9 34.7-38.2l8-3.5c1-.4 1.9-.8 2.9-1.2l-16.9 63.5c-5.6 21.1-.1 43.6 14.7 59.7l70.7 77.1 22 88.1c4.3 17.1 21.7 27.6 38.8 23.3s27.6-21.7 23.3-38.8l-23-92.1c-1.9-7.8-5.8-14.9-11.2-20.8l-49.5-54 19.3-65.5 9.6 23c4.4 10.6 12.5 19.3 22.8 24.5l26.7 13.3c15.8 7.9 35 1.5 42.9-14.3s1.5-35-14.3-42.9L537 232.7l-15.3-36.8C504.5 154.8 464.3 128 419.7 128c-22.8 0-45.3 4.8-66.1 14l-8 3.5c-24.4 10.9-44.6 29-58.1 51.6L157.3 136.9C144.7 131 130.9 128 117 128H64zM464 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM349.7 335.6l-25 62.4-59.4 59.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L372.3 441c4.6-4.6 8.2-10.1 10.6-16.1l14.5-36.2-40.7-44.4c-2.5-2.7-4.8-5.6-7-8.6z"],n:[384,512,[110],"4e","M21.1 33.9c12.7-4.6 26.9-.7 35.5 9.6L320 359.6V64c0-17.7 14.3-32 32-32s32 14.3 32 32V448c0 13.5-8.4 25.5-21.1 30.1s-26.9 .7-35.5-9.6L64 152.4V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V64C0 50.5 8.4 38.5 21.1 33.9z"],"cable-car":[512,512,[128673,57551,"tram"],"f7da","M288 0a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM160 56a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM32 288c0-35.3 28.7-64 64-64H232V157.5l-203.1 42c-13 2.7-25.7-5.7-28.4-18.6s5.7-25.7 18.6-28.4l232-48 232-48c13-2.7 25.7 5.7 28.4 18.6s-5.7 25.7-18.6 28.4L280 147.5V224H416c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V288zm64 0c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h64c8.8 0 16-7.2 16-16V304c0-8.8-7.2-16-16-16H96zm112 16v64c0 8.8 7.2 16 16 16h64c8.8 0 16-7.2 16-16V304c0-8.8-7.2-16-16-16H224c-8.8 0-16 7.2-16 16zm144-16c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h64c8.8 0 16-7.2 16-16V304c0-8.8-7.2-16-16-16H352z"],"cloud-rain":[512,512,[127783,9926],"f73d","M96 320c-53 0-96-43-96-96c0-42.5 27.6-78.6 65.9-91.2C64.7 126.1 64 119.1 64 112C64 50.1 114.1 0 176 0c43.1 0 80.5 24.3 99.2 60c14.7-17.1 36.5-28 60.8-28c44.2 0 80 35.8 80 80c0 5.5-.6 10.8-1.6 16c.5 0 1.1 0 1.6 0c53 0 96 43 96 96s-43 96-96 96H96zm-6.8 52c1.3-2.5 3.9-4 6.8-4s5.4 1.5 6.8 4l35.1 64.6c4.1 7.5 6.2 15.8 6.2 24.3v3c0 26.5-21.5 48-48 48s-48-21.5-48-48v-3c0-8.5 2.1-16.9 6.2-24.3L89.2 372zm160 0c1.3-2.5 3.9-4 6.8-4s5.4 1.5 6.8 4l35.1 64.6c4.1 7.5 6.2 15.8 6.2 24.3v3c0 26.5-21.5 48-48 48s-48-21.5-48-48v-3c0-8.5 2.1-16.9 6.2-24.3L249.2 372zm124.9 64.6L409.2 372c1.3-2.5 3.9-4 6.8-4s5.4 1.5 6.8 4l35.1 64.6c4.1 7.5 6.2 15.8 6.2 24.3v3c0 26.5-21.5 48-48 48s-48-21.5-48-48v-3c0-8.5 2.1-16.9 6.2-24.3z"],"building-circle-xmark":[640,512,[],"e4d4","M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c15.1 0 28.5-6.9 37.3-17.8C340.4 462.2 320 417.5 320 368c0-54.7 24.9-103.5 64-135.8V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L518.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L496 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L473.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L496 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z"],ship:[576,512,[128674],"f21a","M192 32c0-17.7 14.3-32 32-32H352c17.7 0 32 14.3 32 32V64h48c26.5 0 48 21.5 48 48V240l44.4 14.8c23.1 7.7 29.5 37.5 11.5 53.9l-101 92.6c-16.2 9.4-34.7 15.1-50.9 15.1c-19.6 0-40.8-7.7-59.2-20.3c-22.1-15.5-51.6-15.5-73.7 0c-17.1 11.8-38 20.3-59.2 20.3c-16.2 0-34.7-5.7-50.9-15.1l-101-92.6c-18-16.5-11.6-46.2 11.5-53.9L96 240V112c0-26.5 21.5-48 48-48h48V32zM160 218.7l107.8-35.9c13.1-4.4 27.3-4.4 40.5 0L416 218.7V128H160v90.7zM306.5 421.9C329 437.4 356.5 448 384 448c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 501.7 417 512 384 512c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 437.2 165.1 448 192 448c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z"],"arrows-down-to-line":[576,512,[],"e4b8","M544 416L32 416c-17.7 0-32 14.3-32 32s14.3 32 32 32l512 0c17.7 0 32-14.3 32-32s-14.3-32-32-32zm22.6-137.4c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L480 274.7 480 64c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 210.7-41.4-41.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l96 96c12.5 12.5 32.8 12.5 45.3 0l96-96zm-320-45.3c-12.5-12.5-32.8-12.5-45.3 0L160 274.7 160 64c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 210.7L54.6 233.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l96 96c12.5 12.5 32.8 12.5 45.3 0l96-96c12.5-12.5 12.5-32.8 0-45.3z"],download:[512,512,[],"f019","M288 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V274.7l-73.4-73.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l128 128c12.5 12.5 32.8 12.5 45.3 0l128-128c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L288 274.7V32zM64 352c-35.3 0-64 28.7-64 64v32c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V416c0-35.3-28.7-64-64-64H346.5l-45.3 45.3c-25 25-65.5 25-90.5 0L165.5 352H64zm368 56a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],"face-grin":[512,512,[128512,"grin"],"f580","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM388.1 312.8c12.3-3.8 24.3 6.9 19.3 18.7C382.4 390.6 324.2 432 256.3 432s-126.2-41.4-151.1-100.5c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"delete-left":[576,512,[9003,"backspace"],"f55a","M576 128c0-35.3-28.7-64-64-64H205.3c-17 0-33.3 6.7-45.3 18.7L9.4 233.4c-6 6-9.4 14.1-9.4 22.6s3.4 16.6 9.4 22.6L160 429.3c12 12 28.3 18.7 45.3 18.7H512c35.3 0 64-28.7 64-64V128zM271 175c9.4-9.4 24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9z"],"eye-dropper":[512,512,["eye-dropper-empty","eyedropper"],"f1fb","M341.6 29.2L240.1 130.8l-9.4-9.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-9.4-9.4L482.8 170.4c39-39 39-102.2 0-141.1s-102.2-39-141.1 0zM55.4 323.3c-15 15-23.4 35.4-23.4 56.6v42.4L5.4 462.2c-8.5 12.7-6.8 29.6 4 40.4s27.7 12.5 40.4 4L89.7 480h42.4c21.2 0 41.6-8.4 56.6-23.4L309.4 335.9l-45.3-45.3L143.4 411.3c-3 3-7.1 4.7-11.3 4.7H96V379.9c0-4.2 1.7-8.3 4.7-11.3L221.4 247.9l-45.3-45.3L55.4 323.3z"],"file-circle-check":[576,512,[],"e5a0","M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v38.6C310.1 219.5 256 287.4 256 368c0 59.1 29.1 111.3 73.7 143.3c-3.2 .5-6.4 .7-9.7 .7H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zM288 368a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm211.3-43.3c-6.2-6.2-16.4-6.2-22.6 0L416 385.4l-28.7-28.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l40 40c6.2 6.2 16.4 6.2 22.6 0l72-72c6.2-6.2 6.2-16.4 0-22.6z"],forward:[512,512,[9193],"f04e","M52.5 440.6c-9.5 7.9-22.8 9.7-34.1 4.4S0 428.4 0 416V96C0 83.6 7.2 72.3 18.4 67s24.5-3.6 34.1 4.4L224 214.3V256v41.7L52.5 440.6zM256 352V256 128 96c0-12.4 7.2-23.7 18.4-29s24.5-3.6 34.1 4.4l192 160c7.3 6.1 11.5 15.1 11.5 24.6s-4.2 18.5-11.5 24.6l-192 160c-9.5 7.9-22.8 9.7-34.1 4.4s-18.4-16.6-18.4-29V352z"],mobile:[384,512,[128241,"mobile-android","mobile-phone"],"f3ce","M80 0C44.7 0 16 28.7 16 64V448c0 35.3 28.7 64 64 64H304c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H80zm80 432h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H160c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],"face-meh":[512,512,[128528,"meh"],"f11a","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM176.4 176a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM160 336H352c8.8 0 16 7.2 16 16s-7.2 16-16 16H160c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],"align-center":[448,512,[],"f037","M352 64c0-17.7-14.3-32-32-32H128c-17.7 0-32 14.3-32 32s14.3 32 32 32H320c17.7 0 32-14.3 32-32zm96 128c0-17.7-14.3-32-32-32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32zM0 448c0 17.7 14.3 32 32 32H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H32c-17.7 0-32 14.3-32 32zM352 320c0-17.7-14.3-32-32-32H128c-17.7 0-32 14.3-32 32s14.3 32 32 32H320c17.7 0 32-14.3 32-32z"],"book-skull":[448,512,["book-dead"],"f6b7","M0 96C0 43 43 0 96 0H384h32c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384 96c-53 0-96-43-96-96V96zM64 416c0 17.7 14.3 32 32 32H352V384H96c-17.7 0-32 14.3-32 32zM320 112c0-35.3-35.8-64-80-64s-80 28.7-80 64c0 20.9 12.6 39.5 32 51.2V176c0 8.8 7.2 16 16 16h64c8.8 0 16-7.2 16-16V163.2c19.4-11.7 32-30.3 32-51.2zM208 96a16 16 0 1 1 0 32 16 16 0 1 1 0-32zm48 16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zM134.3 209.3c-8.1-3.5-17.5 .3-21 8.4s.3 17.5 8.4 21L199.4 272l-77.7 33.3c-8.1 3.5-11.9 12.9-8.4 21s12.9 11.9 21 8.4L240 289.4l105.7 45.3c8.1 3.5 17.5-.3 21-8.4s-.3-17.5-8.4-21L280.6 272l77.7-33.3c8.1-3.5 11.9-12.9 8.4-21s-12.9-11.9-21-8.4L240 254.6 134.3 209.3z"],"id-card":[576,512,[62147,"drivers-license"],"f2c2","M0 96l576 0c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96zm0 32V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128H0zM64 405.3c0-29.5 23.9-53.3 53.3-53.3H234.7c29.5 0 53.3 23.9 53.3 53.3c0 5.9-4.8 10.7-10.7 10.7H74.7c-5.9 0-10.7-4.8-10.7-10.7zM176 192a64 64 0 1 1 0 128 64 64 0 1 1 0-128zm176 16c0-8.8 7.2-16 16-16H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H368c-8.8 0-16-7.2-16-16zm0 64c0-8.8 7.2-16 16-16H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H368c-8.8 0-16-7.2-16-16zm0 64c0-8.8 7.2-16 16-16H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H368c-8.8 0-16-7.2-16-16z"],outdent:[448,512,["dedent"],"f03b","M0 64C0 46.3 14.3 32 32 32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 96 0 81.7 0 64zM192 192c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32zm32 96H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32s14.3-32 32-32zM0 448c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32zM.2 268.6c-8.2-6.4-8.2-18.9 0-25.3l101.9-79.3c10.5-8.2 25.8-.7 25.8 12.6V335.3c0 13.3-15.3 20.8-25.8 12.6L.2 268.6z"],"heart-circle-exclamation":[576,512,[],"e4fe","M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9l2.6-2.4C267.2 438.6 256 404.6 256 368c0-97.2 78.8-176 176-176c28.3 0 55 6.7 78.7 18.5c.9-6.5 1.3-13 1.3-19.6v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z"],house:[576,512,[127968,63498,63500,"home","home-alt","home-lg-alt"],"f015","M575.8 255.5c0 18-15 32.1-32 32.1h-32l.7 160.2c0 2.7-.2 5.4-.5 8.1V472c0 22.1-17.9 40-40 40H456c-1.1 0-2.2 0-3.3-.1c-1.4 .1-2.8 .1-4.2 .1H416 392c-22.1 0-40-17.9-40-40V448 384c0-17.7-14.3-32-32-32H256c-17.7 0-32 14.3-32 32v64 24c0 22.1-17.9 40-40 40H160 128.1c-1.5 0-3-.1-4.5-.2c-1.2 .1-2.4 .2-3.6 .2H104c-22.1 0-40-17.9-40-40V360c0-.9 0-1.9 .1-2.8V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L564.8 231.5c8 7 12 15 11 24z"],"calendar-week":[448,512,[],"f784","M128 0c17.7 0 32 14.3 32 32V64H288V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64h48c26.5 0 48 21.5 48 48v48H0V112C0 85.5 21.5 64 48 64H96V32c0-17.7 14.3-32 32-32zM0 192H448V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V192zm80 64c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16H368c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H80z"],"laptop-medical":[640,512,[],"f812","M64 96c0-35.3 28.7-64 64-64H512c35.3 0 64 28.7 64 64V352H512V96H128V352H64V96zM0 403.2C0 392.6 8.6 384 19.2 384H620.8c10.6 0 19.2 8.6 19.2 19.2c0 42.4-34.4 76.8-76.8 76.8H76.8C34.4 480 0 445.6 0 403.2zM288 160c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H352v48c0 8.8-7.2 16-16 16H304c-8.8 0-16-7.2-16-16V272H240c-8.8 0-16-7.2-16-16V224c0-8.8 7.2-16 16-16h48V160z"],b:[320,512,[98],"42","M64 32C28.7 32 0 60.7 0 96V256 416c0 35.3 28.7 64 64 64H192c70.7 0 128-57.3 128-128c0-46.5-24.8-87.3-62-109.7c18.7-22.3 30-51 30-82.3c0-70.7-57.3-128-128-128H64zm96 192H64V96h96c35.3 0 64 28.7 64 64s-28.7 64-64 64zM64 288h96 32c35.3 0 64 28.7 64 64s-28.7 64-64 64H64V288z"],"file-medical":[384,512,[],"f477","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM160 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H224v48c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V352H112c-8.8 0-16-7.2-16-16V304c0-8.8 7.2-16 16-16h48V240z"],"dice-one":[448,512,[9856],"f525","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM224 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"kiwi-bird":[576,512,[],"f535","M291.2 388.4c31.2-18.8 64.7-36.4 101.1-36.4H448c4.6 0 9.1-.2 13.6-.7l85.3 121.9c4 5.7 11.3 8.2 17.9 6.1s11.2-8.3 11.2-15.3V224c0-70.7-57.3-128-128-128H392.3c-36.4 0-69.9-17.6-101.1-36.4C262.3 42.1 228.3 32 192 32C86 32 0 118 0 224c0 71.1 38.6 133.1 96 166.3V456c0 13.3 10.7 24 24 24s24-10.7 24-24V410c15.3 3.9 31.4 6 48 6c5.4 0 10.7-.2 16-.7V456c0 13.3 10.7 24 24 24s24-10.7 24-24V405.1c12.4-4.4 24.2-10 35.2-16.7zM448 200a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],"arrow-right-arrow-left":[448,512,[8644,"exchange"],"f0ec","M438.6 150.6c12.5-12.5 12.5-32.8 0-45.3l-96-96c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L338.7 96 32 96C14.3 96 0 110.3 0 128s14.3 32 32 32l306.7 0-41.4 41.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l96-96zm-333.3 352c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L109.3 416 416 416c17.7 0 32-14.3 32-32s-14.3-32-32-32l-306.7 0 41.4-41.4c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-96 96c-12.5 12.5-12.5 32.8 0 45.3l96 96z"],"rotate-right":[512,512,["redo-alt","rotate-forward"],"f2f9","M463.5 224H472c13.3 0 24-10.7 24-24V72c0-9.7-5.8-18.5-14.8-22.2s-19.3-1.7-26.2 5.2L413.4 96.6c-87.6-86.5-228.7-86.2-315.8 1c-87.5 87.5-87.5 229.3 0 316.8s229.3 87.5 316.8 0c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0c-62.5 62.5-163.8 62.5-226.3 0s-62.5-163.8 0-226.3c62.2-62.2 162.7-62.5 225.3-1L327 183c-6.9 6.9-8.9 17.2-5.2 26.2s12.5 14.8 22.2 14.8H463.5z"],utensils:[448,512,[127860,61685,"cutlery"],"f2e7","M416 0C400 0 288 32 288 176V288c0 35.3 28.7 64 64 64h32V480c0 17.7 14.3 32 32 32s32-14.3 32-32V352 240 32c0-17.7-14.3-32-32-32zM64 16C64 7.8 57.9 1 49.7 .1S34.2 4.6 32.4 12.5L2.1 148.8C.7 155.1 0 161.5 0 167.9c0 45.9 35.1 83.6 80 87.7V480c0 17.7 14.3 32 32 32s32-14.3 32-32V255.6c44.9-4.1 80-41.8 80-87.7c0-6.4-.7-12.8-2.1-19.1L191.6 12.5c-1.8-8-9.3-13.3-17.4-12.4S160 7.8 160 16V150.2c0 5.4-4.4 9.8-9.8 9.8c-5.1 0-9.3-3.9-9.8-9L127.9 14.6C127.2 6.3 120.3 0 112 0s-15.2 6.3-15.9 14.6L83.7 151c-.5 5.1-4.7 9-9.8 9c-5.4 0-9.8-4.4-9.8-9.8V16zm48.3 152l-.3 0-.3 0 .3-.7 .3 .7z"],"arrow-up-wide-short":[576,512,["sort-amount-up"],"f161","M151.6 42.4C145.5 35.8 137 32 128 32s-17.5 3.8-23.6 10.4l-88 96c-11.9 13-11.1 33.3 2 45.2s33.3 11.1 45.2-2L96 146.3V448c0 17.7 14.3 32 32 32s32-14.3 32-32V146.3l32.4 35.4c11.9 13 32.2 13.9 45.2 2s13.9-32.2 2-45.2l-88-96zM320 480h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H320c-17.7 0-32 14.3-32 32s14.3 32 32 32zm0-128h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H320c-17.7 0-32 14.3-32 32s14.3 32 32 32zm0-128H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H320c-17.7 0-32 14.3-32 32s14.3 32 32 32zm0-128H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H320c-17.7 0-32 14.3-32 32s14.3 32 32 32z"],"mill-sign":[384,512,[],"e1ed","M302.1 42.8c5.9-16.6-2.7-35-19.4-40.9s-35 2.7-40.9 19.4L208 116.1c-5.7 4-11.1 8.5-16 13.5C171.7 108.9 143.3 96 112 96c-19.5 0-37.8 5-53.7 13.7C52.5 101.4 42.9 96 32 96C14.3 96 0 110.3 0 128v80V416c0 17.7 14.3 32 32 32s32-14.3 32-32V208c0-26.5 21.5-48 48-48s48 21.5 48 48v42.5L81.9 469.2c-5.9 16.6 2.7 35 19.4 40.9s35-2.7 40.9-19.4l21.4-60C168.9 441 179.6 448 192 448c17.7 0 32-14.3 32-32V261.5l35.7-100c3.9-1 8.1-1.6 12.3-1.6c26.5 0 48 21.5 48 48V416c0 17.7 14.3 32 32 32s32-14.3 32-32V208c0-58.2-44.3-106-101.1-111.5l19.2-53.8z"],"bowl-rice":[512,512,[],"e2eb","M176 56c0-13.3 10.7-24 24-24h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H200c-13.3 0-24-10.7-24-24zm24 48h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H200c-13.3 0-24-10.7-24-24s10.7-24 24-24zM56 176H72c13.3 0 24 10.7 24 24s-10.7 24-24 24H56c-13.3 0-24-10.7-24-24s10.7-24 24-24zM0 283.4C0 268.3 12.3 256 27.4 256H484.6c15.1 0 27.4 12.3 27.4 27.4c0 70.5-44.4 130.7-106.7 154.1L403.5 452c-2 16-15.6 28-31.8 28H140.2c-16.1 0-29.8-12-31.8-28l-1.8-14.4C44.4 414.1 0 353.9 0 283.4zM224 200c0-13.3 10.7-24 24-24h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H248c-13.3 0-24-10.7-24-24zm-96 0c0-13.3 10.7-24 24-24h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H152c-13.3 0-24-10.7-24-24zm-24-96h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H104c-13.3 0-24-10.7-24-24s10.7-24 24-24zm216 96c0-13.3 10.7-24 24-24h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H344c-13.3 0-24-10.7-24-24zm-24-96h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H296c-13.3 0-24-10.7-24-24s10.7-24 24-24zm120 96c0-13.3 10.7-24 24-24h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H440c-13.3 0-24-10.7-24-24zm-24-96h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H392c-13.3 0-24-10.7-24-24s10.7-24 24-24zM296 32h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H296c-13.3 0-24-10.7-24-24s10.7-24 24-24z"],skull:[512,512,[128128],"f54c","M416 398.9c58.5-41.1 96-104.1 96-174.9C512 100.3 397.4 0 256 0S0 100.3 0 224c0 70.7 37.5 133.8 96 174.9c0 .4 0 .7 0 1.1v64c0 26.5 21.5 48 48 48h48V464c0-8.8 7.2-16 16-16s16 7.2 16 16v48h64V464c0-8.8 7.2-16 16-16s16 7.2 16 16v48h48c26.5 0 48-21.5 48-48V400c0-.4 0-.7 0-1.1zM96 256a64 64 0 1 1 128 0A64 64 0 1 1 96 256zm256-64a64 64 0 1 1 0 128 64 64 0 1 1 0-128z"],"tower-broadcast":[576,512,["broadcast-tower"],"f519","M80.3 44C69.8 69.9 64 98.2 64 128s5.8 58.1 16.3 84c6.6 16.4-1.3 35-17.7 41.7s-35-1.3-41.7-17.7C7.4 202.6 0 166.1 0 128S7.4 53.4 20.9 20C27.6 3.6 46.2-4.3 62.6 2.3S86.9 27.6 80.3 44zM555.1 20C568.6 53.4 576 89.9 576 128s-7.4 74.6-20.9 108c-6.6 16.4-25.3 24.3-41.7 17.7S489.1 228.4 495.7 212c10.5-25.9 16.3-54.2 16.3-84s-5.8-58.1-16.3-84C489.1 27.6 497 9 513.4 2.3s35 1.3 41.7 17.7zM352 128c0 23.7-12.9 44.4-32 55.4V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V183.4c-19.1-11.1-32-31.7-32-55.4c0-35.3 28.7-64 64-64s64 28.7 64 64zM170.6 76.8C163.8 92.4 160 109.7 160 128s3.8 35.6 10.6 51.2c7.1 16.2-.3 35.1-16.5 42.1s-35.1-.3-42.1-16.5c-10.3-23.6-16-49.6-16-76.8s5.7-53.2 16-76.8c7.1-16.2 25.9-23.6 42.1-16.5s23.6 25.9 16.5 42.1zM464 51.2c10.3 23.6 16 49.6 16 76.8s-5.7 53.2-16 76.8c-7.1 16.2-25.9 23.6-42.1 16.5s-23.6-25.9-16.5-42.1c6.8-15.6 10.6-32.9 10.6-51.2s-3.8-35.6-10.6-51.2c-7.1-16.2 .3-35.1 16.5-42.1s35.1 .3 42.1 16.5z"],"truck-pickup":[640,512,[128763],"f63c","M368.6 96l76.8 96H288V96h80.6zM224 80V192H64c-17.7 0-32 14.3-32 32v64c-17.7 0-32 14.3-32 32s14.3 32 32 32H65.1c-.7 5.2-1.1 10.6-1.1 16c0 61.9 50.1 112 112 112s112-50.1 112-112c0-5.4-.4-10.8-1.1-16h66.3c-.7 5.2-1.1 10.6-1.1 16c0 61.9 50.1 112 112 112s112-50.1 112-112c0-5.4-.4-10.8-1.1-16H608c17.7 0 32-14.3 32-32s-14.3-32-32-32V224c0-17.7-14.3-32-32-32H527.4L418.6 56c-12.1-15.2-30.5-24-50-24H272c-26.5 0-48 21.5-48 48zm0 288a48 48 0 1 1 -96 0 48 48 0 1 1 96 0zm288 0a48 48 0 1 1 -96 0 48 48 0 1 1 96 0z"],"up-long":[320,512,["long-arrow-alt-up"],"f30c","M318 177.5c3.8-8.8 2-19-4.6-26l-136-144C172.9 2.7 166.6 0 160 0s-12.9 2.7-17.4 7.5l-136 144c-6.6 7-8.4 17.2-4.6 26S14.4 192 24 192H96l0 288c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32l0-288h72c9.6 0 18.2-5.7 22-14.5z"],stop:[384,512,[9209],"f04d","M0 128C0 92.7 28.7 64 64 64H320c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128z"],"code-merge":[448,512,[],"f387","M80 56a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm32.4 97.2c28-12.4 47.6-40.5 47.6-73.2c0-44.2-35.8-80-80-80S0 35.8 0 80c0 32.8 19.7 61 48 73.3V358.7C19.7 371 0 399.2 0 432c0 44.2 35.8 80 80 80s80-35.8 80-80c0-32.8-19.7-61-48-73.3V272c26.7 20.1 60 32 96 32h86.7c12.3 28.3 40.5 48 73.3 48c44.2 0 80-35.8 80-80s-35.8-80-80-80c-32.8 0-61 19.7-73.3 48H208c-49.9 0-91-38.1-95.6-86.8zM80 408a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM344 272a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z"],upload:[512,512,[],"f093","M288 109.3V352c0 17.7-14.3 32-32 32s-32-14.3-32-32V109.3l-73.4 73.4c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l128-128c12.5-12.5 32.8-12.5 45.3 0l128 128c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L288 109.3zM64 352H192c0 35.3 28.7 64 64 64s64-28.7 64-64H448c35.3 0 64 28.7 64 64v32c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V416c0-35.3 28.7-64 64-64zM432 456a24 24 0 1 0 0-48 24 24 0 1 0 0 48z"],hurricane:[384,512,[],"f751","M0 208C0 104.4 75.7 18.5 174.9 2.6C184 1.2 192 8.6 192 17.9V81.2c0 8.4 6.5 15.3 14.7 16.5C307 112.5 384 199 384 303.4c0 103.6-75.7 189.5-174.9 205.4c-9.2 1.5-17.1-5.9-17.1-15.2V430.2c0-8.4-6.5-15.3-14.7-16.5C77 398.9 0 312.4 0 208zm288 48A96 96 0 1 0 96 256a96 96 0 1 0 192 0zm-96-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],mound:[576,512,[],"e52d","M144.1 179.2C173.8 127.7 228.6 96 288 96s114.2 31.7 143.9 83.2L540.4 368c12.3 21.3-3.1 48-27.7 48H63.3c-24.6 0-40-26.6-27.7-48L144.1 179.2z"],"toilet-portable":[320,512,[],"e583","M0 32V64H320V32c0-17.7-14.3-32-32-32H32C14.3 0 0 14.3 0 32zM24 96H0v24V488c0 13.3 10.7 24 24 24s24-10.7 24-24v-8H272v8c0 13.3 10.7 24 24 24s24-10.7 24-24V120 96H296 24zM256 240v64c0 8.8-7.2 16-16 16s-16-7.2-16-16V240c0-8.8 7.2-16 16-16s16 7.2 16 16z"],"compact-disc":[512,512,[128191,128192,128440],"f51f","M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm256 32a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm-96-32a96 96 0 1 0 192 0 96 96 0 1 0 -192 0zM96 240c0-35 17.5-71.1 45.2-98.8S205 96 240 96c8.8 0 16-7.2 16-16s-7.2-16-16-16c-45.4 0-89.2 22.3-121.5 54.5S64 194.6 64 240c0 8.8 7.2 16 16 16s16-7.2 16-16z"],"file-arrow-down":[384,512,["file-download"],"f56d","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM216 232V334.1l31-31c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-72 72c-9.4 9.4-24.6 9.4-33.9 0l-72-72c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l31 31V232c0-13.3 10.7-24 24-24s24 10.7 24 24z"],caravan:[640,512,[],"f8ff","M0 112C0 67.8 35.8 32 80 32H416c88.4 0 160 71.6 160 160V352h32c17.7 0 32 14.3 32 32s-14.3 32-32 32l-32 0H288c0 53-43 96-96 96s-96-43-96-96H80c-44.2 0-80-35.8-80-80V112zM320 352H448V256H416c-8.8 0-16-7.2-16-16s7.2-16 16-16h32V160c0-17.7-14.3-32-32-32H352c-17.7 0-32 14.3-32 32V352zM96 128c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H224c17.7 0 32-14.3 32-32V160c0-17.7-14.3-32-32-32H96zm96 336a48 48 0 1 0 0-96 48 48 0 1 0 0 96z"],"shield-cat":[512,512,[],"e572","M269.4 2.9C265.2 1 260.7 0 256 0s-9.2 1-13.4 2.9L54.3 82.8c-22 9.3-38.4 31-38.3 57.2c.5 99.2 41.3 280.7 213.6 363.2c16.7 8 36.1 8 52.8 0C454.7 420.7 495.5 239.2 496 140c.1-26.2-16.3-47.9-38.3-57.2L269.4 2.9zM160 154.4c0-5.8 4.7-10.4 10.4-10.4h.2c3.4 0 6.5 1.6 8.5 4.3l40 53.3c3 4 7.8 6.4 12.8 6.4h48c5 0 9.8-2.4 12.8-6.4l40-53.3c2-2.7 5.2-4.3 8.5-4.3h.2c5.8 0 10.4 4.7 10.4 10.4V272c0 53-43 96-96 96s-96-43-96-96V154.4zM216 288a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm96-16a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z"],bolt:[448,512,[9889,"zap"],"f0e7","M349.4 44.6c5.9-13.7 1.5-29.7-10.6-38.5s-28.6-8-39.9 1.8l-256 224c-10 8.8-13.6 22.9-8.9 35.3S50.7 288 64 288H175.5L98.6 467.4c-5.9 13.7-1.5 29.7 10.6 38.5s28.6 8 39.9-1.8l256-224c10-8.8 13.6-22.9 8.9-35.3s-16.6-20.7-30-20.7H272.5L349.4 44.6z"],"glass-water":[384,512,[],"e4f4","M32 0C23.1 0 14.6 3.7 8.6 10.2S-.6 25.4 .1 34.3L28.9 437.7c3 41.9 37.8 74.3 79.8 74.3H275.3c42 0 76.8-32.4 79.8-74.3L383.9 34.3c.6-8.9-2.4-17.6-8.5-24.1S360.9 0 352 0H32zM73 156.5L66.4 64H317.6L311 156.5l-24.2 12.1c-19.4 9.7-42.2 9.7-61.6 0c-20.9-10.4-45.5-10.4-66.4 0c-19.4 9.7-42.2 9.7-61.6 0L73 156.5z"],"oil-well":[576,512,[],"e532","M528.3 61.3c-11.4-42.7-55.3-68-98-56.6L414.9 8.8C397.8 13.4 387.7 31 392.3 48l24.5 91.4L308.5 167.5l-6.3-18.1C297.7 136.6 285.6 128 272 128s-25.7 8.6-30.2 21.4l-13.6 39L96 222.6V184c0-13.3-10.7-24-24-24s-24 10.7-24 24V448H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H406.7L340 257.5l-62.2 16.1L305.3 352H238.7L265 277l-74.6 19.3L137.3 448H96V288.8l337.4-87.5 25.2 94c4.6 17.1 22.1 27.2 39.2 22.6l15.5-4.1c42.7-11.4 68-55.3 56.6-98L528.3 61.3zM205.1 448l11.2-32H327.7l11.2 32H205.1z"],vault:[576,512,[],"e2c5","M64 0C28.7 0 0 28.7 0 64V416c0 35.3 28.7 64 64 64H80l16 32h64l16-32H400l16 32h64l16-32h16c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zM224 320a80 80 0 1 0 0-160 80 80 0 1 0 0 160zm0-240a160 160 0 1 1 0 320 160 160 0 1 1 0-320zM480 221.3V336c0 8.8-7.2 16-16 16s-16-7.2-16-16V221.3c-18.6-6.6-32-24.4-32-45.3c0-26.5 21.5-48 48-48s48 21.5 48 48c0 20.9-13.4 38.7-32 45.3z"],mars:[448,512,[9794],"f222","M289.8 46.8c3.7-9 12.5-14.8 22.2-14.8H424c13.3 0 24 10.7 24 24V168c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-33.4-33.4L321 204.2c19.5 28.4 31 62.7 31 99.8c0 97.2-78.8 176-176 176S0 401.2 0 304s78.8-176 176-176c37 0 71.4 11.4 99.8 31l52.6-52.6L295 73c-6.9-6.9-8.9-17.2-5.2-26.2zM400 80l0 0h0v0zM176 416a112 112 0 1 0 0-224 112 112 0 1 0 0 224z"],toilet:[448,512,[128701],"f7d8","M24 0C10.7 0 0 10.7 0 24S10.7 48 24 48h8V196.9c-1.9 1.4-3.8 2.9-5.6 4.4C10.9 214.5 0 232.9 0 256c0 46.9 14.3 84.1 37 112.5c14.2 17.7 31.1 31.3 48.5 41.8L65.6 469.9c-3.3 9.8-1.6 20.5 4.4 28.8s15.7 13.3 26 13.3H352c10.3 0 19.9-4.9 26-13.3s7.7-19.1 4.4-28.8l-19.8-59.5c17.4-10.5 34.3-24.1 48.5-41.8c22.7-28.4 37-65.5 37-112.5c0-23.1-10.9-41.5-26.4-54.6c-1.8-1.5-3.7-3-5.6-4.4V48h8c13.3 0 24-10.7 24-24s-10.7-24-24-24H24zM384 256.3c0 1-.3 2.6-3.8 5.6c-4.8 4.1-14 9-29.3 13.4C320.5 284 276.1 288 224 288s-96.5-4-126.9-12.8c-15.3-4.4-24.5-9.3-29.3-13.4c-3.5-3-3.8-4.6-3.8-5.6l0-.3 0-.1c0-1 0-2.5 3.8-5.8c4.8-4.1 14-9 29.3-13.4C127.5 228 171.9 224 224 224s96.5 4 126.9 12.8c15.3 4.4 24.5 9.3 29.3 13.4c3.8 3.2 3.8 4.8 3.8 5.8l0 .1 0 .3zM328.2 384l-.2 .5 0-.5h.2zM112 64h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],"plane-circle-xmark":[640,512,[],"e557","M256 0c-35 0-64 59.5-64 93.7v84.6L8.1 283.4c-5 2.8-8.1 8.2-8.1 13.9v65.5c0 10.6 10.2 18.3 20.4 15.4l171.6-49 0 70.9-57.6 43.2c-4 3-6.4 7.8-6.4 12.8v42c0 7.8 6.3 14 14 14c1.3 0 2.6-.2 3.9-.5L256 480l110.1 31.5c1.3 .4 2.6 .5 3.9 .5c6 0 11.1-3.7 13.1-9C344.5 470.7 320 422.2 320 368c0-60.6 30.6-114 77.1-145.6L320 178.3V93.7C320 59.5 292 0 256 0zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L518.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L496 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L473.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L496 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z"],"yen-sign":[320,512,[165,"cny","jpy","rmb","yen"],"f157","M58.6 46.2C48.8 31.5 29 27.6 14.3 37.4S-4.4 67 5.4 81.7L100.2 224H48c-17.7 0-32 14.3-32 32s14.3 32 32 32h80v32H48c-17.7 0-32 14.3-32 32s14.3 32 32 32h80v64c0 17.7 14.3 32 32 32s32-14.3 32-32V384h80c17.7 0 32-14.3 32-32s-14.3-32-32-32H192V288h80c17.7 0 32-14.3 32-32s-14.3-32-32-32H219.8L314.6 81.7c9.8-14.7 5.8-34.6-8.9-44.4s-34.6-5.8-44.4 8.9L160 198.3 58.6 46.2z"],"ruble-sign":[384,512,[8381,"rouble","rub","ruble"],"f158","M96 32C78.3 32 64 46.3 64 64V256H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64v32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64v32c0 17.7 14.3 32 32 32s32-14.3 32-32V416H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H128V320H240c79.5 0 144-64.5 144-144s-64.5-144-144-144H96zM240 256H128V96H240c44.2 0 80 35.8 80 80s-35.8 80-80 80z"],sun:[512,512,[9728],"f185","M361.5 1.2c5 2.1 8.6 6.6 9.6 11.9L391 121l107.9 19.8c5.3 1 9.8 4.6 11.9 9.6s1.5 10.7-1.6 15.2L446.9 256l62.3 90.3c3.1 4.5 3.7 10.2 1.6 15.2s-6.6 8.6-11.9 9.6L391 391 371.1 498.9c-1 5.3-4.6 9.8-9.6 11.9s-10.7 1.5-15.2-1.6L256 446.9l-90.3 62.3c-4.5 3.1-10.2 3.7-15.2 1.6s-8.6-6.6-9.6-11.9L121 391 13.1 371.1c-5.3-1-9.8-4.6-11.9-9.6s-1.5-10.7 1.6-15.2L65.1 256 2.8 165.7c-3.1-4.5-3.7-10.2-1.6-15.2s6.6-8.6 11.9-9.6L121 121 140.9 13.1c1-5.3 4.6-9.8 9.6-11.9s10.7-1.5 15.2 1.6L256 65.1 346.3 2.8c4.5-3.1 10.2-3.7 15.2-1.6zM160 256a96 96 0 1 1 192 0 96 96 0 1 1 -192 0zm224 0a128 128 0 1 0 -256 0 128 128 0 1 0 256 0z"],guitar:[512,512,[],"f7a6","M465 7c-9.4-9.4-24.6-9.4-33.9 0L383 55c-2.4 2.4-4.3 5.3-5.5 8.5l-15.4 41-77.5 77.6c-45.1-29.4-99.3-30.2-131 1.6c-11 11-18 24.6-21.4 39.6c-3.7 16.6-19.1 30.7-36.1 31.6c-25.6 1.3-49.3 10.7-67.3 28.6C-16 328.4-7.6 409.4 47.5 464.5s136.1 63.5 180.9 18.7c17.9-17.9 27.4-41.7 28.6-67.3c.9-17 15-32.3 31.6-36.1c15-3.4 28.6-10.5 39.6-21.4c31.8-31.8 31-85.9 1.6-131l77.6-77.6 41-15.4c3.2-1.2 6.1-3.1 8.5-5.5l48-48c9.4-9.4 9.4-24.6 0-33.9L465 7zM208 256a48 48 0 1 1 0 96 48 48 0 1 1 0-96z"],"face-laugh-wink":[512,512,["laugh-wink"],"f59c","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM96.8 314.1c-3.8-13.7 7.4-26.1 21.6-26.1H393.6c14.2 0 25.5 12.4 21.6 26.1C396.2 382 332.1 432 256 432s-140.2-50-159.2-117.9zM144.4 192a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm156.4 25.6c-5.3 7.1-15.3 8.5-22.4 3.2s-8.5-15.3-3.2-22.4c30.4-40.5 91.2-40.5 121.6 0c5.3 7.1 3.9 17.1-3.2 22.4s-17.1 3.9-22.4-3.2c-17.6-23.5-52.8-23.5-70.4 0z"],"horse-head":[640,512,[],"f7ab","M64 464V316.9c0-108.4 68.3-205.1 170.5-241.3L404.2 15.5C425.6 7.9 448 23.8 448 46.4c0 11-5.5 21.2-14.6 27.3L400 96c48.1 0 91.2 29.8 108.1 74.9l48.6 129.5c11.8 31.4 4.1 66.8-19.6 90.5c-16 16-37.8 25.1-60.5 25.1h-3.4c-26.1 0-50.9-11.6-67.6-31.7l-32.3-38.7c-11.7 4.1-24.2 6.4-37.3 6.4l-.1 0 0 0c-6.3 0-12.5-.5-18.6-1.5c-3.6-.6-7.2-1.4-10.7-2.3l0 0c-28.9-7.8-53.1-26.8-67.8-52.2c-4.4-7.6-14.2-10.3-21.9-5.8s-10.3 14.2-5.8 21.9c24 41.5 68.3 70 119.3 71.9l47.2 70.8c4 6.1 6.2 13.2 6.2 20.4c0 20.3-16.5 36.8-36.8 36.8H112c-26.5 0-48-21.5-48-48zM392 224a24 24 0 1 0 0-48 24 24 0 1 0 0 48z"],"bore-hole":[512,512,[],"e4c3","M256 0c-17.7 0-32 14.3-32 32V296.6c-19.1 11.1-32 31.7-32 55.4c0 35.3 28.7 64 64 64s64-28.7 64-64c0-23.7-12.9-44.4-32-55.4V32c0-17.7-14.3-32-32-32zM48 128c-26.5 0-48 21.5-48 48V464c0 26.5 21.5 48 48 48H464c26.5 0 48-21.5 48-48V176c0-26.5-21.5-48-48-48H384c-17.7 0-32 14.3-32 32V352c0 53-43 96-96 96s-96-43-96-96V160c0-17.7-14.3-32-32-32H48z"],industry:[576,512,[],"f275","M64 32C46.3 32 32 46.3 32 64V304v48 80c0 26.5 21.5 48 48 48H496c26.5 0 48-21.5 48-48V304 152.2c0-18.2-19.4-29.7-35.4-21.1L352 215.4V152.2c0-18.2-19.4-29.7-35.4-21.1L160 215.4V64c0-17.7-14.3-32-32-32H64z"],"circle-down":[512,512,[61466,"arrow-alt-circle-down"],"f358","M256 0a256 256 0 1 0 0 512A256 256 0 1 0 256 0zM376.9 294.6L269.8 394.5c-3.8 3.5-8.7 5.5-13.8 5.5s-10.1-2-13.8-5.5L135.1 294.6c-4.5-4.2-7.1-10.1-7.1-16.3c0-12.3 10-22.3 22.3-22.3l57.7 0 0-96c0-17.7 14.3-32 32-32l32 0c17.7 0 32 14.3 32 32l0 96 57.7 0c12.3 0 22.3 10 22.3 22.3c0 6.2-2.6 12.1-7.1 16.3z"],"arrows-turn-to-dots":[512,512,[],"e4c1","M249.4 25.4c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3L269.3 96 416 96c53 0 96 43 96 96v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V192c0-17.7-14.3-32-32-32l-146.7 0 25.4 25.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0l-80-80c-12.5-12.5-12.5-32.8 0-45.3l80-80zm13.3 256l80 80c12.5 12.5 12.5 32.8 0 45.3l-80 80c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L242.7 416 96 416c-17.7 0-32 14.3-32 32v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448c0-53 43-96 96-96l146.7 0-25.4-25.4c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0zM384 384a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM64 192A64 64 0 1 1 64 64a64 64 0 1 1 0 128z"],"florin-sign":[384,512,[],"e184","M314.7 32c-38.8 0-73.7 23.3-88.6 59.1L170.7 224H64c-17.7 0-32 14.3-32 32s14.3 32 32 32h80L98.9 396.3c-5 11.9-16.6 19.7-29.5 19.7H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H69.3c38.8 0 73.7-23.3 88.6-59.1L213.3 288H320c17.7 0 32-14.3 32-32s-14.3-32-32-32H240l45.1-108.3c5-11.9 16.6-19.7 29.5-19.7H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H314.7z"],"arrow-down-short-wide":[576,512,["sort-amount-desc","sort-amount-down-alt"],"f884","M151.6 469.6C145.5 476.2 137 480 128 480s-17.5-3.8-23.6-10.4l-88-96c-11.9-13-11.1-33.3 2-45.2s33.3-11.1 45.2 2L96 365.7V64c0-17.7 14.3-32 32-32s32 14.3 32 32V365.7l32.4-35.4c11.9-13 32.2-13.9 45.2-2s13.9 32.2 2 45.2l-88 96zM320 32h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H320c-17.7 0-32-14.3-32-32s14.3-32 32-32zm0 128h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H320c-17.7 0-32-14.3-32-32s14.3-32 32-32zm0 128H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H320c-17.7 0-32-14.3-32-32s14.3-32 32-32zm0 128H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H320c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],"less-than":[384,512,[62774],"3c","M380.6 81.7c7.9 15.8 1.5 35-14.3 42.9L103.6 256 366.3 387.4c15.8 7.9 22.2 27.1 14.3 42.9s-27.1 22.2-42.9 14.3l-320-160C6.8 279.2 0 268.1 0 256s6.8-23.2 17.7-28.6l320-160c15.8-7.9 35-1.5 42.9 14.3z"],"angle-down":[448,512,[8964],"f107","M201.4 374.6c12.5 12.5 32.8 12.5 45.3 0l160-160c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L224 306.7 86.6 169.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l160 160z"],"car-tunnel":[512,512,[],"e4de","M256 0C114.6 0 0 114.6 0 256V448c0 35.3 28.7 64 64 64h42.8c-6.6-5.9-10.8-14.4-10.8-24V376c0-20.8 11.3-38.9 28.1-48.6l21-64.7c7.5-23.1 29-38.7 53.3-38.7H313.6c24.3 0 45.8 15.6 53.3 38.7l21 64.7c16.8 9.7 28.2 27.8 28.2 48.6V488c0 9.6-4.2 18.1-10.8 24H448c35.3 0 64-28.7 64-64V256C512 114.6 397.4 0 256 0zM362.8 512c-6.6-5.9-10.8-14.4-10.8-24V448H160v40c0 9.6-4.2 18.1-10.8 24H362.8zM190.8 277.5L177 320H335l-13.8-42.5c-1.1-3.3-4.1-5.5-7.6-5.5H198.4c-3.5 0-6.5 2.2-7.6 5.5zM168 408a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm200-24a24 24 0 1 0 -48 0 24 24 0 1 0 48 0z"],"head-side-cough":[640,512,[],"e061","M0 224.2C0 100.6 100.2 0 224 0h24c95.2 0 181.2 69.3 197.3 160.2c2.3 13 6.8 25.7 15.1 36l42 52.6c6.2 7.8 9.6 17.4 9.6 27.4c0 24.2-19.6 43.8-43.8 43.8H448v0 32L339.2 365.6c-11 1.4-19.2 10.7-19.2 21.8c0 11.6 9 21.2 20.6 21.9L448 416v16c0 26.5-21.5 48-48 48H320v8c0 13.3-10.7 24-24 24H256v0H96c-17.7 0-32-14.3-32-32V407.3c0-16.7-6.9-32.5-17.1-45.8C16.6 322.4 0 274.1 0 224.2zm352-.2a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM464 384a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm152-24a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM592 480a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zM552 312a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm40-24a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zM552 408a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],"grip-lines":[448,512,[],"f7a4","M32 288c-17.7 0-32 14.3-32 32s14.3 32 32 32l384 0c17.7 0 32-14.3 32-32s-14.3-32-32-32L32 288zm0-128c-17.7 0-32 14.3-32 32s14.3 32 32 32l384 0c17.7 0 32-14.3 32-32s-14.3-32-32-32L32 160z"],"thumbs-down":[512,512,[128078,61576],"f165","M313.4 479.1c26-5.2 42.9-30.5 37.7-56.5l-2.3-11.4c-5.3-26.7-15.1-52.1-28.8-75.2H464c26.5 0 48-21.5 48-48c0-18.5-10.5-34.6-25.9-42.6C497 236.6 504 223.1 504 208c0-23.4-16.8-42.9-38.9-47.1c4.4-7.3 6.9-15.8 6.9-24.9c0-21.3-13.9-39.4-33.1-45.6c.7-3.3 1.1-6.8 1.1-10.4c0-26.5-21.5-48-48-48H294.5c-19 0-37.5 5.6-53.3 16.1L202.7 73.8C176 91.6 160 121.6 160 153.7V192v48 24.9c0 29.2 13.3 56.7 36 75l7.4 5.9c26.5 21.2 44.6 51 51.2 84.2l2.3 11.4c5.2 26 30.5 42.9 56.5 37.7zM32 384H96c17.7 0 32-14.3 32-32V128c0-17.7-14.3-32-32-32H32C14.3 96 0 110.3 0 128V352c0 17.7 14.3 32 32 32z"],"user-lock":[640,512,[],"f502","M224 256A128 128 0 1 0 224 0a128 128 0 1 0 0 256zm-45.7 48C79.8 304 0 383.8 0 482.3C0 498.7 13.3 512 29.7 512H392.6c-5.4-9.4-8.6-20.3-8.6-32V352c0-2.1 .1-4.2 .3-6.3c-31-26-71-41.7-114.6-41.7H178.3zM528 240c17.7 0 32 14.3 32 32v48H496V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z"],"arrow-right-long":[512,512,["long-arrow-right"],"f178","M502.6 278.6c12.5-12.5 12.5-32.8 0-45.3l-128-128c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L402.7 224 32 224c-17.7 0-32 14.3-32 32s14.3 32 32 32l370.7 0-73.4 73.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l128-128z"],"anchor-circle-xmark":[640,512,[],"e4ac","M320 96a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm21.1 80C367 158.8 384 129.4 384 96c0-53-43-96-96-96s-96 43-96 96c0 33.4 17 62.8 42.9 80H224c-17.7 0-32 14.3-32 32s14.3 32 32 32h32V448H208c-53 0-96-43-96-96v-6.1l7 7c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L97 263c-9.4-9.4-24.6-9.4-33.9 0L7 319c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l7-7V352c0 88.4 71.6 160 160 160h80 80c8.2 0 16.3-.6 24.2-1.8c-22.2-16.2-40.4-37.5-53-62.2H320V368 240h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H341.1zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L518.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L496 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L473.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L496 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z"],ellipsis:[448,512,["ellipsis-h"],"f141","M8 256a56 56 0 1 1 112 0A56 56 0 1 1 8 256zm160 0a56 56 0 1 1 112 0 56 56 0 1 1 -112 0zm216-56a56 56 0 1 1 0 112 56 56 0 1 1 0-112z"],"chess-pawn":[320,512,[9823],"f443","M215.5 224c29.2-18.4 48.5-50.9 48.5-88c0-57.4-46.6-104-104-104S56 78.6 56 136c0 37.1 19.4 69.6 48.5 88H96c-17.7 0-32 14.3-32 32c0 16.5 12.5 30 28.5 31.8L80 400H240L227.5 287.8c16-1.8 28.5-15.3 28.5-31.8c0-17.7-14.3-32-32-32h-8.5zM22.6 473.4c-4.2 4.2-6.6 10-6.6 16C16 501.9 26.1 512 38.6 512H281.4c12.5 0 22.6-10.1 22.6-22.6c0-6-2.4-11.8-6.6-16L256 432H64L22.6 473.4z"],"kit-medical":[576,512,["first-aid"],"f479","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H96V32H64zm64 0V480H448V32H128zM512 480c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H480V480h32zM256 176c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H320v48c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V288H208c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16h48V176z"],"person-through-window":[640,512,[],"e5a9","M64 64l224 0 0 9.8c0 39-23.7 74-59.9 88.4C167.6 186.5 128 245 128 310.2l0 73.8s0 0 0 0H64V64zm288 0l224 0V384H508.3l-3.7-4.5-75.2-90.2c-9.1-10.9-22.6-17.3-36.9-17.3l-71.1 0-41-63.1c-.3-.5-.6-1-1-1.4c44.7-29 72.5-79 72.5-133.6l0-9.8zm73 320H379.2l42.7 64H592c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H48C21.5 0 0 21.5 0 48V400c0 26.5 21.5 48 48 48H308.2l33.2 49.8c9.8 14.7 29.7 18.7 44.4 8.9s18.7-29.7 8.9-44.4L310.5 336l74.6 0 40 48zm-159.5 0H192s0 0 0 0l0-73.8c0-10.2 1.6-20.1 4.7-29.5L265.5 384zM192 128a48 48 0 1 0 -96 0 48 48 0 1 0 96 0z"],toolbox:[512,512,[129520],"f552","M176 88l0 40 160 0 0-40c0-4.4-3.6-8-8-8L184 80c-4.4 0-8 3.6-8 8zm-48 40l0-40c0-30.9 25.1-56 56-56l144 0c30.9 0 56 25.1 56 56l0 40 28.1 0c12.7 0 24.9 5.1 33.9 14.1l51.9 51.9c9 9 14.1 21.2 14.1 33.9l0 92.1-128 0 0-32c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 32-128 0 0-32c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 32L0 320l0-92.1c0-12.7 5.1-24.9 14.1-33.9l51.9-51.9c9-9 21.2-14.1 33.9-14.1l28.1 0zM0 416l0-64 128 0c0 17.7 14.3 32 32 32s32-14.3 32-32l128 0c0 17.7 14.3 32 32 32s32-14.3 32-32l128 0 0 64c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64z"],"hands-holding-circle":[640,512,[],"e4fb","M320 0a128 128 0 1 1 0 256A128 128 0 1 1 320 0zM40 64c22.1 0 40 17.9 40 40v40 80 40.2c0 17 6.7 33.3 18.7 45.3l51.1 51.1c8.3 8.3 21.3 9.6 31 3.1c12.9-8.6 14.7-26.9 3.7-37.8l-15.2-15.2-32-32c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0l32 32 15.2 15.2 0 0 25.3 25.3c21 21 32.8 49.5 32.8 79.2V464c0 26.5-21.5 48-48 48H173.3c-17 0-33.3-6.7-45.3-18.7L28.1 393.4C10.1 375.4 0 351 0 325.5V224 160 104C0 81.9 17.9 64 40 64zm560 0c22.1 0 40 17.9 40 40v56 64V325.5c0 25.5-10.1 49.9-28.1 67.9L512 493.3c-12 12-28.3 18.7-45.3 18.7H400c-26.5 0-48-21.5-48-48V385.1c0-29.7 11.8-58.2 32.8-79.2l25.3-25.3 0 0 15.2-15.2 32-32c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3l-32 32-15.2 15.2c-11 11-9.2 29.2 3.7 37.8c9.7 6.5 22.7 5.2 31-3.1l51.1-51.1c12-12 18.7-28.3 18.7-45.3V224 144 104c0-22.1 17.9-40 40-40z"],bug:[512,512,[],"f188","M256 0c53 0 96 43 96 96v3.6c0 15.7-12.7 28.4-28.4 28.4H188.4c-15.7 0-28.4-12.7-28.4-28.4V96c0-53 43-96 96-96zM41.4 105.4c12.5-12.5 32.8-12.5 45.3 0l64 64c.7 .7 1.3 1.4 1.9 2.1c14.2-7.3 30.4-11.4 47.5-11.4H312c17.1 0 33.2 4.1 47.5 11.4c.6-.7 1.2-1.4 1.9-2.1l64-64c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3l-64 64c-.7 .7-1.4 1.3-2.1 1.9c6.2 12 10.1 25.3 11.1 39.5H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H416c0 24.6-5.5 47.8-15.4 68.6c2.2 1.3 4.2 2.9 6 4.8l64 64c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0l-63.1-63.1c-24.5 21.8-55.8 36.2-90.3 39.6V240c0-8.8-7.2-16-16-16s-16 7.2-16 16V479.2c-34.5-3.4-65.8-17.8-90.3-39.6L86.6 502.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l64-64c1.9-1.9 3.9-3.4 6-4.8C101.5 367.8 96 344.6 96 320H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H96.3c1.1-14.1 5-27.5 11.1-39.5c-.7-.6-1.4-1.2-2.1-1.9l-64-64c-12.5-12.5-12.5-32.8 0-45.3z"],"credit-card":[576,512,[128179,62083,"credit-card-alt"],"f09d","M64 32C28.7 32 0 60.7 0 96v32H576V96c0-35.3-28.7-64-64-64H64zM576 224H0V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V224zM112 352h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zm112 16c0-8.8 7.2-16 16-16H368c8.8 0 16 7.2 16 16s-7.2 16-16 16H240c-8.8 0-16-7.2-16-16z"],car:[512,512,[128664,"automobile"],"f1b9","M135.2 117.4L109.1 192H402.9l-26.1-74.6C372.3 104.6 360.2 96 346.6 96H165.4c-13.6 0-25.7 8.6-30.2 21.4zM39.6 196.8L74.8 96.3C88.3 57.8 124.6 32 165.4 32H346.6c40.8 0 77.1 25.8 90.6 64.3l35.2 100.5c23.2 9.6 39.6 32.5 39.6 59.2V400v48c0 17.7-14.3 32-32 32H448c-17.7 0-32-14.3-32-32V400H96v48c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32V400 256c0-26.7 16.4-49.6 39.6-59.2zM128 288a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm288 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"hand-holding-hand":[576,512,[],"e4f7","M7.8 207.7c-13.1-17.8-9.3-42.8 8.5-55.9L142.9 58.5C166.2 41.3 194.5 32 223.5 32H384 544c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H507.2l-44.9 36c-22.7 18.2-50.9 28-80 28H304 288 224c-17.7 0-32-14.3-32-32s14.3-32 32-32h64 16c8.8 0 16-7.2 16-16s-7.2-16-16-16H183.4L63.7 216.2c-17.8 13.1-42.8 9.3-55.9-8.5zM382.4 160l0 0 .9 0c-.3 0-.6 0-.9 0zM568.2 304.3c13.1 17.8 9.3 42.8-8.5 55.9L433.1 453.5c-23.4 17.2-51.6 26.5-80.7 26.5H192 32c-17.7 0-32-14.3-32-32V384c0-17.7 14.3-32 32-32H68.8l44.9-36c22.7-18.2 50.9-28 80-28H272h16 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H288 272c-8.8 0-16 7.2-16 16s7.2 16 16 16H392.6l119.7-88.2c17.8-13.1 42.8-9.3 55.9 8.5zM193.6 352l0 0-.9 0c.3 0 .6 0 .9 0z"],"book-open-reader":[512,512,["book-reader"],"f5da","M160 96a96 96 0 1 1 192 0A96 96 0 1 1 160 96zm80 152V512l-48.4-24.2c-20.9-10.4-43.5-17-66.8-19.3l-96-9.6C12.5 457.2 0 443.5 0 427V224c0-17.7 14.3-32 32-32H62.3c63.6 0 125.6 19.6 177.7 56zm32 264V248c52.1-36.4 114.1-56 177.7-56H480c17.7 0 32 14.3 32 32V427c0 16.4-12.5 30.2-28.8 31.8l-96 9.6c-23.2 2.3-45.9 8.9-66.8 19.3L272 512z"],"mountain-sun":[640,512,[],"e52f","M560 160A80 80 0 1 0 560 0a80 80 0 1 0 0 160zM55.9 512H381.1h75H578.9c33.8 0 61.1-27.4 61.1-61.1c0-11.2-3.1-22.2-8.9-31.8l-132-216.3C495 196.1 487.8 192 480 192s-15 4.1-19.1 10.7l-48.2 79L286.8 81c-6.6-10.6-18.3-17-30.8-17s-24.1 6.4-30.8 17L8.6 426.4C3 435.3 0 445.6 0 456.1C0 487 25 512 55.9 512z"],"arrows-left-right-to-line":[640,512,[],"e4ba","M32 64c17.7 0 32 14.3 32 32l0 320c0 17.7-14.3 32-32 32s-32-14.3-32-32V96C0 78.3 14.3 64 32 64zm214.6 73.4c12.5 12.5 12.5 32.8 0 45.3L205.3 224l229.5 0-41.4-41.4c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0l96 96c12.5 12.5 12.5 32.8 0 45.3l-96 96c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L434.7 288l-229.5 0 41.4 41.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0l-96-96c-12.5-12.5-12.5-32.8 0-45.3l96-96c12.5-12.5 32.8-12.5 45.3 0zM640 96V416c0 17.7-14.3 32-32 32s-32-14.3-32-32V96c0-17.7 14.3-32 32-32s32 14.3 32 32z"],"dice-d20":[512,512,[],"f6cf","M48.7 125.8l53.2 31.9c7.8 4.7 17.8 2 22.2-5.9L201.6 12.1c3-5.4-.9-12.1-7.1-12.1c-1.6 0-3.2 .5-4.6 1.4L47.9 98.8c-9.6 6.6-9.2 20.9 .8 26.9zM16 171.7V295.3c0 8 10.4 11 14.7 4.4l60-92c5-7.6 2.6-17.8-5.2-22.5L40.2 158C29.6 151.6 16 159.3 16 171.7zM310.4 12.1l77.6 139.6c4.4 7.9 14.5 10.6 22.2 5.9l53.2-31.9c10-6 10.4-20.3 .8-26.9L322.1 1.4c-1.4-.9-3-1.4-4.6-1.4c-6.2 0-10.1 6.7-7.1 12.1zM496 171.7c0-12.4-13.6-20.1-24.2-13.7l-45.3 27.2c-7.8 4.7-10.1 14.9-5.2 22.5l60 92c4.3 6.7 14.7 3.6 14.7-4.4V171.7zm-49.3 246L286.1 436.6c-8.1 .9-14.1 7.8-14.1 15.9v52.8c0 3.7 3 6.8 6.8 6.8c.8 0 1.6-.1 2.4-.4l172.7-64c6.1-2.2 10.1-8 10.1-14.5c0-9.3-8.1-16.5-17.3-15.4zM233.2 512c3.7 0 6.8-3 6.8-6.8V452.6c0-8.1-6.1-14.9-14.1-15.9l-160.6-19c-9.2-1.1-17.3 6.1-17.3 15.4c0 6.5 4 12.3 10.1 14.5l172.7 64c.8 .3 1.6 .4 2.4 .4zM41.7 382.9l170.9 20.2c7.8 .9 13.4-7.5 9.5-14.3l-85.7-150c-5.9-10.4-20.7-10.8-27.3-.8L30.2 358.2c-6.5 9.9-.3 23.3 11.5 24.7zm439.6-24.8L402.9 238.1c-6.5-10-21.4-9.6-27.3 .8L290.2 388.5c-3.9 6.8 1.6 15.2 9.5 14.3l170.1-20c11.8-1.4 18-14.7 11.5-24.6zm-216.9 11l78.4-137.2c6.1-10.7-1.6-23.9-13.9-23.9H183.1c-12.3 0-20 13.3-13.9 23.9l78.4 137.2c3.7 6.4 13 6.4 16.7 0zM174.4 176H337.6c12.2 0 19.9-13.1 14-23.8l-80-144c-2.8-5.1-8.2-8.2-14-8.2h-3.2c-5.8 0-11.2 3.2-14 8.2l-80 144c-5.9 10.7 1.8 23.8 14 23.8z"],"truck-droplet":[640,512,[],"e58c","M0 48C0 21.5 21.5 0 48 0H368c26.5 0 48 21.5 48 48V96h50.7c17 0 33.3 6.7 45.3 18.7L589.3 192c12 12 18.7 28.3 18.7 45.3V256v32 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H576c0 53-43 96-96 96s-96-43-96-96H256c0 53-43 96-96 96s-96-43-96-96H48c-26.5 0-48-21.5-48-48V48zM416 256H544V237.3L466.7 160H416v96zM160 464a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm368-48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM208 272c39.8 0 72-29.6 72-66c0-27-39.4-82.9-59.9-110.3c-6.1-8.2-18.1-8.2-24.2 0C175.4 123 136 179 136 206c0 36.5 32.2 66 72 66z"],"file-circle-xmark":[576,512,[],"e5a1","M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v38.6C310.1 219.5 256 287.4 256 368c0 59.1 29.1 111.3 73.7 143.3c-3.2 .5-6.4 .7-9.7 .7H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zm48 96a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm59.3 107.3c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0L432 345.4l-36.7-36.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6L409.4 368l-36.7 36.7c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L432 390.6l36.7 36.7c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6L454.6 368l36.7-36.7z"],"temperature-arrow-up":[576,512,["temperature-up"],"e040","M128 112c0-26.5 21.5-48 48-48s48 21.5 48 48V276.5c0 17.3 7.1 31.9 15.3 42.5C249.8 332.6 256 349.5 256 368c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-18.5 6.2-35.4 16.7-48.9c8.2-10.6 15.3-25.2 15.3-42.5V112zM176 0C114.1 0 64 50.1 64 112V276.4c0 .1-.1 .3-.2 .6c-.2 .6-.8 1.6-1.7 2.8C43.2 304.2 32 334.8 32 368c0 79.5 64.5 144 144 144s144-64.5 144-144c0-33.2-11.2-63.8-30.1-88.1c-.9-1.2-1.5-2.2-1.7-2.8c-.1-.3-.2-.5-.2-.6V112C288 50.1 237.9 0 176 0zm0 416c26.5 0 48-21.5 48-48c0-20.9-13.4-38.7-32-45.3V112c0-8.8-7.2-16-16-16s-16 7.2-16 16V322.7c-18.6 6.6-32 24.4-32 45.3c0 26.5 21.5 48 48 48zM480 160h32c12.9 0 24.6-7.8 29.6-19.8s2.2-25.7-6.9-34.9l-64-64c-12.5-12.5-32.8-12.5-45.3 0l-64 64c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8h32V448c0 17.7 14.3 32 32 32s32-14.3 32-32V160z"],medal:[512,512,[127941],"f5a2","M4.1 38.2C1.4 34.2 0 29.4 0 24.6C0 11 11 0 24.6 0H133.9c11.2 0 21.7 5.9 27.4 15.5l68.5 114.1c-48.2 6.1-91.3 28.6-123.4 61.9L4.1 38.2zm503.7 0L405.6 191.5c-32.1-33.3-75.2-55.8-123.4-61.9L350.7 15.5C356.5 5.9 366.9 0 378.1 0H487.4C501 0 512 11 512 24.6c0 4.8-1.4 9.6-4.1 13.6zM80 336a176 176 0 1 1 352 0A176 176 0 1 1 80 336zm184.4-94.9c-3.4-7-13.3-7-16.8 0l-22.4 45.4c-1.4 2.8-4 4.7-7 5.1L168 298.9c-7.7 1.1-10.7 10.5-5.2 16l36.3 35.4c2.2 2.2 3.2 5.2 2.7 8.3l-8.6 49.9c-1.3 7.6 6.7 13.5 13.6 9.9l44.8-23.6c2.7-1.4 6-1.4 8.7 0l44.8 23.6c6.9 3.6 14.9-2.2 13.6-9.9l-8.6-49.9c-.5-3 .5-6.1 2.7-8.3l36.3-35.4c5.6-5.4 2.5-14.8-5.2-16l-50.1-7.3c-3-.4-5.7-2.4-7-5.1l-22.4-45.4z"],bed:[640,512,[128716],"f236","M32 32c17.7 0 32 14.3 32 32V320H288V160c0-17.7 14.3-32 32-32H544c53 0 96 43 96 96V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V416H352 320 64v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V64C0 46.3 14.3 32 32 32zm144 96a80 80 0 1 1 0 160 80 80 0 1 1 0-160z"],"square-h":[448,512,["h-square"],"f0fd","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM336 152V256 360c0 13.3-10.7 24-24 24s-24-10.7-24-24V280H160l0 80c0 13.3-10.7 24-24 24s-24-10.7-24-24l0-208c0-13.3 10.7-24 24-24s24 10.7 24 24v80H288V152c0-13.3 10.7-24 24-24s24 10.7 24 24z"],podcast:[448,512,[],"f2ce","M319.4 372c48.5-31.3 80.6-85.9 80.6-148c0-97.2-78.8-176-176-176S48 126.8 48 224c0 62.1 32.1 116.6 80.6 148c1.2 17.3 4 38 7.2 57.1l.2 1C56 395.8 0 316.5 0 224C0 100.3 100.3 0 224 0S448 100.3 448 224c0 92.5-56 171.9-136 206.1l.2-1.1c3.1-19.2 6-39.8 7.2-57zm-2.3-38.1c-1.6-5.7-3.9-11.1-7-16.2c-5.8-9.7-13.5-17-21.9-22.4c19.5-17.6 31.8-43 31.8-71.3c0-53-43-96-96-96s-96 43-96 96c0 28.3 12.3 53.8 31.8 71.3c-8.4 5.4-16.1 12.7-21.9 22.4c-3.1 5.1-5.4 10.5-7 16.2C99.8 307.5 80 268 80 224c0-79.5 64.5-144 144-144s144 64.5 144 144c0 44-19.8 83.5-50.9 109.9zM224 312c32.9 0 64 8.6 64 43.8c0 33-12.9 104.1-20.6 132.9c-5.1 19-24.5 23.4-43.4 23.4s-38.2-4.4-43.4-23.4c-7.8-28.5-20.6-99.7-20.6-132.8c0-35.1 31.1-43.8 64-43.8zm0-144a56 56 0 1 1 0 112 56 56 0 1 1 0-112z"],"temperature-full":[320,512,["temperature-4","thermometer-4","thermometer-full"],"f2c7","M160 64c-26.5 0-48 21.5-48 48V276.5c0 17.3-7.1 31.9-15.3 42.5C86.2 332.6 80 349.5 80 368c0 44.2 35.8 80 80 80s80-35.8 80-80c0-18.5-6.2-35.4-16.7-48.9c-8.2-10.6-15.3-25.2-15.3-42.5V112c0-26.5-21.5-48-48-48zM48 112C48 50.2 98.1 0 160 0s112 50.1 112 112V276.5c0 .1 .1 .3 .2 .6c.2 .6 .8 1.6 1.7 2.8c18.9 24.4 30.1 55 30.1 88.1c0 79.5-64.5 144-144 144S16 447.5 16 368c0-33.2 11.2-63.8 30.1-88.1c.9-1.2 1.5-2.2 1.7-2.8c.1-.3 .2-.5 .2-.6V112zM208 368c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-20.9 13.4-38.7 32-45.3V112c0-8.8 7.2-16 16-16s16 7.2 16 16V322.7c18.6 6.6 32 24.4 32 45.3z"],bell:[448,512,[128276,61602],"f0f3","M224 0c-17.7 0-32 14.3-32 32V51.2C119 66 64 130.6 64 208v18.8c0 47-17.3 92.4-48.5 127.6l-7.4 8.3c-8.4 9.4-10.4 22.9-5.3 34.4S19.4 416 32 416H416c12.6 0 24-7.4 29.2-18.9s3.1-25-5.3-34.4l-7.4-8.3C401.3 319.2 384 273.9 384 226.8V208c0-77.4-55-142-128-156.8V32c0-17.7-14.3-32-32-32zm45.3 493.3c12-12 18.7-28.3 18.7-45.3H224 160c0 17 6.7 33.3 18.7 45.3s28.3 18.7 45.3 18.7s33.3-6.7 45.3-18.7z"],superscript:[512,512,[],"f12b","M480 32c0-11.1-5.7-21.4-15.2-27.2s-21.2-6.4-31.1-1.4l-32 16c-15.8 7.9-22.2 27.1-14.3 42.9C393 73.5 404.3 80 416 80v80c-17.7 0-32 14.3-32 32s14.3 32 32 32h32 32c17.7 0 32-14.3 32-32s-14.3-32-32-32V32zM32 64C14.3 64 0 78.3 0 96s14.3 32 32 32H47.3l89.6 128L47.3 384H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64c10.4 0 20.2-5.1 26.2-13.6L176 311.8l85.8 122.6c6 8.6 15.8 13.6 26.2 13.6h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H304.7L215.1 256l89.6-128H320c17.7 0 32-14.3 32-32s-14.3-32-32-32H288c-10.4 0-20.2 5.1-26.2 13.6L176 200.2 90.2 77.6C84.2 69.1 74.4 64 64 64H32z"],"plug-circle-xmark":[576,512,[],"e560","M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8c12.3-2.5 24.1-6.4 35.1-11.5c-2.1-10.8-3.1-21.9-3.1-33.3c0-80.3 53.8-148 127.3-169.2c.5-2.2 .7-4.5 .7-6.8c0-17.7-14.3-32-32-32H32zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L454.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L432 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L409.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L432 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z"],"star-of-life":[512,512,[],"f621","M208 32c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V172.9l122-70.4c15.3-8.8 34.9-3.6 43.7 11.7l16 27.7c8.8 15.3 3.6 34.9-11.7 43.7L352 256l122 70.4c15.3 8.8 20.5 28.4 11.7 43.7l-16 27.7c-8.8 15.3-28.4 20.6-43.7 11.7L304 339.1V480c0 17.7-14.3 32-32 32H240c-17.7 0-32-14.3-32-32V339.1L86 409.6c-15.3 8.8-34.9 3.6-43.7-11.7l-16-27.7c-8.8-15.3-3.6-34.9 11.7-43.7L160 256 38 185.6c-15.3-8.8-20.5-28.4-11.7-43.7l16-27.7C51.1 98.8 70.7 93.6 86 102.4l122 70.4V32z"],"phone-slash":[640,512,[],"f3dd","M601.2 5.1c10.4-8.2 25.5-6.3 33.7 4.1s6.3 25.5-4.1 33.7l-592 464c-10.4 8.2-25.5 6.3-33.7-4.1s-6.3-25.5 4.1-33.7l155.6-122C101.8 270 64 171.4 64 64c0-18 12.1-33.8 29.5-38.6l88-24c19.4-5.3 39.7 4.6 47.4 23.2l40 96c6.8 16.3 2.1 35.2-11.6 46.3L208 207.3c12 25.5 27.2 49.2 45 70.7L601.2 5.1zM234.3 415.6l91.3-72c13.7 9.1 28 17.3 43 24.3L409 318.7c11.2-13.7 30-18.4 46.3-11.6l96 40c18.6 7.7 28.5 28 23.2 47.4l-24 88C545.8 499.9 530 512 512 512c-104.9 0-201.3-36-277.7-96.4z"],"paint-roller":[512,512,[],"f5aa","M0 64C0 28.7 28.7 0 64 0H352c35.3 0 64 28.7 64 64v64c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM160 352c0-17.7 14.3-32 32-32V304c0-44.2 35.8-80 80-80H416c17.7 0 32-14.3 32-32V160 69.5c37.3 13.2 64 48.7 64 90.5v32c0 53-43 96-96 96H272c-8.8 0-16 7.2-16 16v16c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32V352z"],"handshake-angle":[640,512,["hands-helping"],"f4c4","M544 248v3.3l69.7-69.7c21.9-21.9 21.9-57.3 0-79.2L535.6 24.4c-21.9-21.9-57.3-21.9-79.2 0L416.3 64.5c-2.7-.3-5.5-.5-8.3-.5H296c-37.1 0-67.6 28-71.6 64H224V248c0 22.1 17.9 40 40 40s40-17.9 40-40V176c0 0 0-.1 0-.1V160l16 0 136 0c0 0 0 0 .1 0H464c44.2 0 80 35.8 80 80v8zM336 192v56c0 39.8-32.2 72-72 72s-72-32.2-72-72V129.4c-35.9 6.2-65.8 32.3-76 68.2L99.5 255.2 26.3 328.4c-21.9 21.9-21.9 57.3 0 79.2l78.1 78.1c21.9 21.9 57.3 21.9 79.2 0l37.7-37.7c.9 0 1.8 .1 2.7 .1H384c26.5 0 48-21.5 48-48c0-5.6-1-11-2.7-16H432c26.5 0 48-21.5 48-48c0-12.8-5-24.4-13.2-33c25.7-5 45.1-27.6 45.2-54.8v-.4c-.1-30.8-25.1-55.8-56-55.8c0 0 0 0 0 0l-120 0z"],"location-dot":[384,512,["map-marker-alt"],"f3c5","M215.7 499.2C267 435 384 279.4 384 192C384 86 298 0 192 0S0 86 0 192c0 87.4 117 243 168.3 307.2c12.3 15.3 35.1 15.3 47.4 0zM192 128a64 64 0 1 1 0 128 64 64 0 1 1 0-128z"],file:[384,512,[128196,128459,61462],"f15b","M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128z"],"greater-than":[384,512,[62769],"3e","M3.4 81.7c-7.9 15.8-1.5 35 14.3 42.9L280.5 256 17.7 387.4C1.9 395.3-4.5 414.5 3.4 430.3s27.1 22.2 42.9 14.3l320-160c10.8-5.4 17.7-16.5 17.7-28.6s-6.8-23.2-17.7-28.6l-320-160c-15.8-7.9-35-1.5-42.9 14.3z"],"person-swimming":[576,512,[127946,"swimmer"],"f5c4","M309.5 178.4L447.9 297.1c-1.6 .9-3.2 2-4.8 3c-18 12.4-40.1 20.3-59.2 20.3c-19.6 0-40.8-7.7-59.2-20.3c-22.1-15.5-51.6-15.5-73.7 0c-17.1 11.8-38 20.3-59.2 20.3c-10.1 0-21.1-2.2-31.9-6.2C163.1 193.2 262.2 96 384 96h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384c-26.9 0-52.3 6.6-74.5 18.4zM160 160A64 64 0 1 1 32 160a64 64 0 1 1 128 0zM306.5 325.9C329 341.4 356.5 352 384 352c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 405.7 417 416 384 416c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 341.2 165.1 352 192 352c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z"],"arrow-down":[384,512,[8595],"f063","M169.4 470.6c12.5 12.5 32.8 12.5 45.3 0l160-160c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L224 370.8 224 64c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 306.7L54.6 265.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l160 160z"],droplet:[384,512,[128167,"tint"],"f043","M192 512C86 512 0 426 0 320C0 228.8 130.2 57.7 166.6 11.7C172.6 4.2 181.5 0 191.1 0h1.8c9.6 0 18.5 4.2 24.5 11.7C253.8 57.7 384 228.8 384 320c0 106-86 192-192 192zM96 336c0-8.8-7.2-16-16-16s-16 7.2-16 16c0 61.9 50.1 112 112 112c8.8 0 16-7.2 16-16s-7.2-16-16-16c-44.2 0-80-35.8-80-80z"],eraser:[576,512,[],"f12d","M290.7 57.4L57.4 290.7c-25 25-25 65.5 0 90.5l80 80c12 12 28.3 18.7 45.3 18.7H288h9.4H512c17.7 0 32-14.3 32-32s-14.3-32-32-32H387.9L518.6 285.3c25-25 25-65.5 0-90.5L381.3 57.4c-25-25-65.5-25-90.5 0zM297.4 416H288l-105.4 0-80-80L227.3 211.3 364.7 348.7 297.4 416z"],"earth-americas":[512,512,[127758,"earth","earth-america","globe-americas"],"f57d","M57.7 193l9.4 16.4c8.3 14.5 21.9 25.2 38 29.8L163 255.7c17.2 4.9 29 20.6 29 38.5v39.9c0 11 6.2 21 16 25.9s16 14.9 16 25.9v39c0 15.6 14.9 26.9 29.9 22.6c16.1-4.6 28.6-17.5 32.7-33.8l2.8-11.2c4.2-16.9 15.2-31.4 30.3-40l8.1-4.6c15-8.5 24.2-24.5 24.2-41.7v-8.3c0-12.7-5.1-24.9-14.1-33.9l-3.9-3.9c-9-9-21.2-14.1-33.9-14.1H257c-11.1 0-22.1-2.9-31.8-8.4l-34.5-19.7c-4.3-2.5-7.6-6.5-9.2-11.2c-3.2-9.6 1.1-20 10.2-24.5l5.9-3c6.6-3.3 14.3-3.9 21.3-1.5l23.2 7.7c8.2 2.7 17.2-.4 21.9-7.5c4.7-7 4.2-16.3-1.2-22.8l-13.6-16.3c-10-12-9.9-29.5 .3-41.3l15.7-18.3c8.8-10.3 10.2-25 3.5-36.7l-2.4-4.2c-3.5-.2-6.9-.3-10.4-.3C163.1 48 84.4 108.9 57.7 193zM464 256c0-36.8-9.6-71.4-26.4-101.5L412 164.8c-15.7 6.3-23.8 23.8-18.5 39.8l16.9 50.7c3.5 10.4 12 18.3 22.6 20.9l29.1 7.3c1.2-9 1.8-18.2 1.8-27.5zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256z"],"person-burst":[640,512,[],"e53b","M480 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-8 384V352h16V480c0 17.7 14.3 32 32 32s32-14.3 32-32V256.9l28.6 47.5c9.1 15.1 28.8 20 43.9 10.9s20-28.8 10.9-43.9l-58.3-97c-17.4-28.9-48.6-46.6-82.3-46.6H465.1c-33.7 0-64.9 17.7-82.3 46.6l-58.3 97c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9L408 256.9V480c0 17.7 14.3 32 32 32s32-14.3 32-32zM190.9 18.1C188.4 12 182.6 8 176 8s-12.4 4-14.9 10.1l-29.4 74L55.6 68.9c-6.3-1.9-13.1 .2-17.2 5.3s-4.6 12.2-1.4 17.9l39.5 69.1L10.9 206.4c-5.4 3.7-8 10.3-6.5 16.7s6.7 11.2 13.1 12.2l78.7 12.2L90.6 327c-.5 6.5 3.1 12.7 9 15.5s12.9 1.8 17.8-2.6L176 286.1l58.6 53.9c4.8 4.4 11.9 5.5 17.8 2.6s9.5-9 9-15.5l-5.6-79.4 50.5-7.8 24.3-40.5-55.2-38L315 92.2c3.3-5.7 2.7-12.8-1.4-17.9s-10.9-7.2-17.2-5.3L220.3 92.1l-29.4-74z"],dove:[512,512,[128330],"f4ba","M160.8 96.5c14 17 31 30.9 49.5 42.2c25.9 15.8 53.7 25.9 77.7 31.6V138.8C265.8 108.5 250 71.5 248.6 28c-.4-11.3-7.5-21.5-18.4-24.4c-7.6-2-15.8-.2-21 5.8c-13.3 15.4-32.7 44.6-48.4 87.2zM320 144v30.6l0 0v1.3l0 0 0 32.1c-60.8-5.1-185-43.8-219.3-157.2C97.4 40 87.9 32 76.6 32c-7.9 0-15.3 3.9-18.8 11C46.8 65.9 32 112.1 32 176c0 116.9 80.1 180.5 118.4 202.8L11.8 416.6C6.7 418 2.6 421.8 .9 426.8s-.8 10.6 2.3 14.8C21.7 466.2 77.3 512 160 512c3.6 0 7.2-1.2 10-3.5L245.6 448H320c88.4 0 160-71.6 160-160V128l29.9-44.9c1.3-2 2.1-4.4 2.1-6.8c0-6.8-5.5-12.3-12.3-12.3H400c-44.2 0-80 35.8-80 80zm80-16a16 16 0 1 1 0 32 16 16 0 1 1 0-32z"],"battery-empty":[576,512,["battery-0"],"f244","M80 160c-8.8 0-16 7.2-16 16V336c0 8.8 7.2 16 16 16H464c8.8 0 16-7.2 16-16V176c0-8.8-7.2-16-16-16H80zM0 176c0-44.2 35.8-80 80-80H464c44.2 0 80 35.8 80 80v16c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32v16c0 44.2-35.8 80-80 80H80c-44.2 0-80-35.8-80-80V176z"],socks:[512,512,[129510],"f696","M175.2 476.6c-9.7-18-15.2-38.7-15.2-60.6c0-40.3 19-78.2 51.2-102.4l64-48c8.1-6 12.8-15.5 12.8-25.6V96H128V240c0 20.1-9.5 39.1-25.6 51.2l-64 48C14.2 357.3 0 385.8 0 416c0 53 43 96 96 96c20.8 0 41-6.7 57.6-19.2l21.6-16.2zM128 64H288V48c0-14.5 3.9-28.2 10.7-39.9C291 3 281.9 0 272 0H176c-26.5 0-48 21.5-48 48V64zM320 96V240c0 20.1-9.5 39.1-25.6 51.2l-64 48C206.2 357.3 192 385.8 192 416c0 53 43 96 96 96c20.8 0 41-6.7 57.6-19.2l115.2-86.4C493 382.2 512 344.3 512 304V96H320zM512 64V48c0-26.5-21.5-48-48-48H368c-26.5 0-48 21.5-48 48V64H512z"],inbox:[512,512,[],"f01c","M121 32C91.6 32 66 52 58.9 80.5L1.9 308.4C.6 313.5 0 318.7 0 323.9V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V323.9c0-5.2-.6-10.4-1.9-15.5l-57-227.9C446 52 420.4 32 391 32H121zm0 64H391l48 192H387.8c-12.1 0-23.2 6.8-28.6 17.7l-14.3 28.6c-5.4 10.8-16.5 17.7-28.6 17.7H195.8c-12.1 0-23.2-6.8-28.6-17.7l-14.3-28.6c-5.4-10.8-16.5-17.7-28.6-17.7H73L121 96z"],section:[256,512,[],"e447","M64.9 96C67.1 84.4 73.7 76.2 86 70.6c13.8-6.2 34.8-8.9 61.2-4.5c8.8 1.4 36.1 7.1 44.1 9.3c17 4.8 34.7-5.1 39.5-22.2s-5.1-34.7-22.2-39.5c-11.1-3.1-41-9.2-50.9-10.8C123-2.7 88.3-.6 59.7 12.3C29.9 25.8 7.5 50.9 1.6 86.5c-.1 .5-.2 1.1-.2 1.6c-2.2 19.7 .3 37.9 8.1 54.1c7.7 16.1 19.4 28 32 36.9c.6 .5 1.3 .9 2 1.4C22.3 194.2 6.5 215.1 1.7 243c-.1 .6-.2 1.1-.2 1.7c-2.3 19.3 .4 37.1 8.4 53c7.9 15.6 19.8 27 32.3 35.5c22.4 15.2 51.9 24 75.4 31l0 0 3.7 1.1c27.2 8.2 46.9 14.6 59.4 23.8c5.5 4 8.2 7.6 9.5 10.9c1.3 3.2 2.6 8.6 .9 18.1c-1.7 10.1-7.7 18-20.7 23.5c-14 6-35.4 8.5-62 4.4c-12.8-2.1-35.1-9.7-54.1-16.2l0 0c-4.3-1.5-8.5-2.9-12.3-4.2C25.3 420 7.2 429.1 1.6 445.8s3.5 34.9 20.3 40.5c2.6 .8 5.7 1.9 9.2 3.1c18.6 6.3 48.5 16.6 67.3 19.6l0 0 .2 0c34.5 5.4 68.8 3.4 97.2-8.7c29.4-12.6 52.5-36.5 58.5-71.5c3.3-19.3 1.9-37.4-5-53.9c-6.3-15-16.4-26.4-27.6-35.2c16.5-13.9 28.5-33.2 32.6-58.2c3.2-19.8 1.9-38.3-4.8-55.1c-6.7-16.8-17.8-29.4-30.2-39c-22.8-17.6-53.6-27.4-77.7-35l-1.4-.5c-27.4-8.7-47.8-15.3-61.5-25c-6.1-4.4-9.5-8.5-11.4-12.4c-1.8-3.7-3.2-9.3-2.3-18.5zm76.7 208.5c-.2-.1-.4-.1-.6-.2l-1.4-.4c-27.4-8.2-47.9-14.5-61.7-23.8c-6.2-4.2-9.3-7.9-11-11.3c-1.5-3-2.9-7.7-2.1-15.7c1.9-9.7 7.9-17.3 20.5-22.7c14-6 35.4-8.5 62.1-4.3l16.4 2.6c6.3 2.9 11.7 6 16.2 9.5c5.5 4.2 8.4 8.2 10 12.2c1.6 4 2.8 10.4 1.1 20.9c-2.4 14.7-12.8 26.4-37.1 31l-12.4 2.3z"],"gauge-high":[512,512,[62461,"tachometer-alt","tachometer-alt-fast"],"f625","M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM288 96a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM256 416c35.3 0 64-28.7 64-64c0-17.4-6.9-33.1-18.1-44.6L366 161.7c5.3-12.1-.2-26.3-12.3-31.6s-26.3 .2-31.6 12.3L257.9 288c-.6 0-1.3 0-1.9 0c-35.3 0-64 28.7-64 64s28.7 64 64 64zM176 144a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM96 288a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm352-32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z"],"envelope-open-text":[512,512,[],"f658","M215.4 96H144 107.8 96v8.8V144v40.4 89L.2 202.5c1.6-18.1 10.9-34.9 25.7-45.8L48 140.3V96c0-26.5 21.5-48 48-48h76.6l49.9-36.9C232.2 3.9 243.9 0 256 0s23.8 3.9 33.5 11L339.4 48H416c26.5 0 48 21.5 48 48v44.3l22.1 16.4c14.8 10.9 24.1 27.7 25.7 45.8L416 273.4v-89V144 104.8 96H404.2 368 296.6 215.4zM0 448V242.1L217.6 403.3c11.1 8.2 24.6 12.7 38.4 12.7s27.3-4.4 38.4-12.7L512 242.1V448v0c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64v0zM176 160H336c8.8 0 16 7.2 16 16s-7.2 16-16 16H176c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H336c8.8 0 16 7.2 16 16s-7.2 16-16 16H176c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],hospital:[640,512,[127973,62589,"hospital-alt","hospital-wide"],"f0f8","M192 48c0-26.5 21.5-48 48-48H400c26.5 0 48 21.5 48 48V512H368V432c0-26.5-21.5-48-48-48s-48 21.5-48 48v80H192V48zM48 96H160V512H48c-26.5 0-48-21.5-48-48V320H80c8.8 0 16-7.2 16-16s-7.2-16-16-16H0V224H80c8.8 0 16-7.2 16-16s-7.2-16-16-16H0V144c0-26.5 21.5-48 48-48zm544 0c26.5 0 48 21.5 48 48v48H560c-8.8 0-16 7.2-16 16s7.2 16 16 16h80v64H560c-8.8 0-16 7.2-16 16s7.2 16 16 16h80V464c0 26.5-21.5 48-48 48H480V96H592zM312 64c-8.8 0-16 7.2-16 16v24H272c-8.8 0-16 7.2-16 16v16c0 8.8 7.2 16 16 16h24v24c0 8.8 7.2 16 16 16h16c8.8 0 16-7.2 16-16V152h24c8.8 0 16-7.2 16-16V120c0-8.8-7.2-16-16-16H344V80c0-8.8-7.2-16-16-16H312z"],"wine-bottle":[512,512,[],"f72f","M393.4 9.4c12.5-12.5 32.8-12.5 45.3 0l64 64c12.5 12.5 12.5 32.8 0 45.3c-11.8 11.8-30.7 12.5-43.2 1.9l-9.5 9.5-48.8 48.8c-9.2 9.2-11.5 22.9-8.6 35.6c9.4 40.9-1.9 85.6-33.8 117.5L197.3 493.3c-25 25-65.5 25-90.5 0l-88-88c-25-25-25-65.5 0-90.5L180.2 153.3c31.9-31.9 76.6-43.1 117.5-33.8c12.6 2.9 26.4 .5 35.5-8.6l48.8-48.8 9.5-9.5c-10.6-12.6-10-31.4 1.9-43.2zM99.3 347.3l65.4 65.4c6.2 6.2 16.4 6.2 22.6 0l97.4-97.4c6.2-6.2 6.2-16.4 0-22.6l-65.4-65.4c-6.2-6.2-16.4-6.2-22.6 0L99.3 324.7c-6.2 6.2-6.2 16.4 0 22.6z"],"chess-rook":[448,512,[9820],"f447","M32 192V48c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16V88c0 4.4 3.6 8 8 8h32c4.4 0 8-3.6 8-8V48c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16V88c0 4.4 3.6 8 8 8h32c4.4 0 8-3.6 8-8V48c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16V192c0 10.1-4.7 19.6-12.8 25.6L352 256l16 144H80L96 256 44.8 217.6C36.7 211.6 32 202.1 32 192zm176 96h32c8.8 0 16-7.2 16-16V224c0-17.7-14.3-32-32-32s-32 14.3-32 32v48c0 8.8 7.2 16 16 16zM22.6 473.4L64 432H384l41.4 41.4c4.2 4.2 6.6 10 6.6 16c0 12.5-10.1 22.6-22.6 22.6H38.6C26.1 512 16 501.9 16 489.4c0-6 2.4-11.8 6.6-16z"],"bars-staggered":[512,512,["reorder","stream"],"f550","M0 96C0 78.3 14.3 64 32 64H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 128 0 113.7 0 96zM64 256c0-17.7 14.3-32 32-32H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H96c-17.7 0-32-14.3-32-32zM448 416c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H416c17.7 0 32 14.3 32 32z"],dharmachakra:[512,512,[9784],"f655","M337.8 205.7l48.6-42.5c13.8 19.3 23.4 41.9 27.4 66.2l-64.4 4.3c-2.4-10.1-6.4-19.5-11.6-28zm140.1 19.5c-5.3-38.8-20.6-74.5-43.2-104.3l.8-.7C449 108.4 449.7 87.6 437 75s-33.4-12-45.2 1.5l-.7 .8c-29.8-22.6-65.5-37.9-104.3-43.2l.1-1.1c1.2-17.9-13-33-30.9-33s-32.1 15.2-30.9 33l.1 1.1c-38.8 5.3-74.5 20.6-104.3 43.2l-.7-.8C108.4 63 87.6 62.3 75 75s-12 33.4 1.5 45.2l.8 .7c-22.6 29.8-37.9 65.5-43.2 104.3l-1.1-.1c-17.9-1.2-33 13-33 30.9s15.2 32.1 33 30.9l1.1-.1c5.3 38.8 20.6 74.5 43.2 104.3l-.8 .7C63 403.6 62.3 424.4 75 437s33.4 12 45.2-1.5l.7-.8c29.8 22.6 65.5 37.9 104.3 43.2l-.1 1.1c-1.2 17.9 13 33 30.9 33s32.1-15.2 30.9-33l-.1-1.1c38.8-5.3 74.5-20.6 104.3-43.2l.7 .8c11.8 13.5 32.5 14.2 45.2 1.5s12-33.4-1.5-45.2l-.8-.7c22.6-29.8 37.9-65.5 43.2-104.3l1.1 .1c17.9 1.2 33-13 33-30.9s-15.2-32.1-33-30.9l-1.1 .1zM163.2 125.6c19.3-13.8 41.9-23.4 66.2-27.5l4.3 64.4c-10 2.4-19.5 6.4-28 11.6l-42.5-48.6zm-65 103.8c4.1-24.4 13.7-46.9 27.5-66.2l48.6 42.5c-5.3 8.5-9.2 18-11.6 28l-64.4-4.3zm27.5 119.4c-13.8-19.3-23.4-41.9-27.5-66.2l64.4-4.3c2.4 10 6.4 19.5 11.6 28l-48.6 42.5zm103.8 65c-24.4-4.1-46.9-13.7-66.2-27.4l42.5-48.6c8.5 5.3 18 9.2 28 11.6l-4.3 64.4zm119.4-27.4c-19.3 13.8-41.9 23.4-66.2 27.4l-4.3-64.4c10-2.4 19.5-6.4 28-11.6l42.5 48.6zm65-103.8c-4.1 24.4-13.7 46.9-27.4 66.2l-48.6-42.5c5.3-8.5 9.2-18 11.6-28l64.4 4.3zm-65-156.9l-42.5 48.6c-8.5-5.3-18-9.2-28-11.6l4.3-64.4c24.4 4.1 46.9 13.7 66.2 27.5zM256 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],hotdog:[512,512,[127789],"f80f","M488.6 23.4c31.2 31.2 31.2 81.9 0 113.1l-352 352c-31.2 31.2-81.9 31.2-113.1 0s-31.2-81.9 0-113.1l352-352c31.2-31.2 81.9-31.2 113.1 0zM443.3 92.7c-6.2-6.2-16.4-6.2-22.6 0c-12.5 12.5-23.8 15.1-37.5 17.6l-2.5 .4c-13.8 2.5-31.6 5.6-48 22c-16.7 16.7-20.9 36-24.1 50.9l0 0v0l-.2 1c-3.4 15.6-6 26.4-15.7 36.1s-20.5 12.3-36.1 15.7l-1 .2c-14.9 3.2-34.2 7.4-50.9 24.1s-20.9 36-24.1 50.9l-.2 1c-3.4 15.6-6 26.4-15.7 36.1c-9.2 9.2-18 10.8-32.7 13.4l0 0-.9 .2c-15.6 2.8-34.9 6.9-54.4 26.4c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0c12.5-12.5 23.8-15.1 37.5-17.6l2.5-.4c13.8-2.5 31.6-5.6 48-22c16.7-16.7 20.9-36 24.1-50.9l.2-1c3.4-15.6 6-26.4 15.7-36.1s20.5-12.3 36.1-15.7l1-.2c14.9-3.2 34.2-7.4 50.9-24.1s20.9-36 24.1-50.9l.2-1c3.4-15.6 6-26.4 15.7-36.1c9.2-9.2 18-10.8 32.7-13.4l.9-.2c15.6-2.8 34.9-6.9 54.4-26.4c6.2-6.2 6.2-16.4 0-22.6zM191.2 479.2l288-288L495 207c10.9 10.9 17 25.6 17 41s-6.1 30.1-17 41L289 495c-10.9 10.9-25.6 17-41 17s-30.1-6.1-41-17l-15.8-15.8zM17 305C6.1 294.1 0 279.4 0 264s6.1-30.1 17-41L223 17C233.9 6.1 248.6 0 264 0s30.1 6.1 41 17l15.8 15.8-288 288L17 305z"],"person-walking-with-cane":[512,512,["blind"],"f29d","M176 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-8.4 32c-36.4 0-69.6 20.5-85.9 53.1L35.4 273.7c-7.9 15.8-1.5 35 14.3 42.9s35 1.5 42.9-14.3L128 231.6v43.2c0 17 6.7 33.3 18.7 45.3L224 397.3V480c0 17.7 14.3 32 32 32s32-14.3 32-32V390.6c0-12.7-5.1-24.9-14.1-33.9L224 306.7V213.3l70.4 93.9c10.6 14.1 30.7 17 44.8 6.4s17-30.7 6.4-44.8L268.8 166.4C250.7 142.2 222.2 128 192 128H167.6zM128.3 346.8L97 472.2c-4.3 17.1 6.1 34.5 23.3 38.8s34.5-6.1 38.8-23.3l22-88.2-52.8-52.8zM450.8 505.1c5 7.3 15 9.1 22.3 4s9.1-15 4-22.3L358.9 316.1c-2.8 3.8-6.1 7.3-10.1 10.3c-5 3.8-10.5 6.4-16.2 7.9L450.8 505.1z"],drum:[512,512,[129345],"f569","M501.2 76.1c11.1-7.3 14.2-22.1 6.9-33.2s-22.1-14.2-33.2-6.9L370.2 104.5C335.8 98.7 297 96 256 96C114.6 96 0 128 0 208V368c0 31.3 27.4 58.8 72 78.7V344c0-13.3 10.7-24 24-24s24 10.7 24 24V463.4c33 8.9 71.1 14.5 112 16.1V376c0-13.3 10.7-24 24-24s24 10.7 24 24V479.5c40.9-1.6 79-7.2 112-16.1V344c0-13.3 10.7-24 24-24s24 10.7 24 24V446.7c44.6-19.9 72-47.4 72-78.7V208c0-41.1-30.2-69.5-78.8-87.4l67.9-44.5zM307.4 145.6l-64.6 42.3c-11.1 7.3-14.2 22.1-6.9 33.2s22.1 14.2 33.2 6.9l111.1-72.8c14.7 3.2 27.9 7 39.4 11.5C458.4 181.8 464 197.4 464 208c0 .8-2.7 17.2-46 35.9C379.1 260.7 322 272 256 272s-123.1-11.3-162-28.1C50.7 225.2 48 208.8 48 208c0-10.6 5.6-26.2 44.4-41.3C130.6 151.9 187.8 144 256 144c18 0 35.1 .5 51.4 1.6z"],"ice-cream":[448,512,[127848],"f810","M367.1 160c.6-5.3 .9-10.6 .9-16C368 64.5 303.5 0 224 0S80 64.5 80 144c0 5.4 .3 10.7 .9 16H80c-26.5 0-48 21.5-48 48s21.5 48 48 48h53.5 181H368c26.5 0 48-21.5 48-48s-21.5-48-48-48h-.9zM96 288L200.8 497.7c4.4 8.8 13.3 14.3 23.2 14.3s18.8-5.5 23.2-14.3L352 288H96z"],"heart-circle-bolt":[576,512,[],"e4fc","M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9l2.6-2.4C267.2 438.6 256 404.6 256 368c0-97.2 78.8-176 176-176c28.3 0 55 6.7 78.7 18.5c.9-6.5 1.3-13 1.3-19.6v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm47.9-225c4.3 3.7 5.4 9.9 2.6 14.9L452.4 356H488c5.2 0 9.8 3.3 11.4 8.2s-.1 10.3-4.2 13.4l-96 72c-4.5 3.4-10.8 3.2-15.1-.6s-5.4-9.9-2.6-14.9L411.6 380H376c-5.2 0-9.8-3.3-11.4-8.2s.1-10.3 4.2-13.4l96-72c4.5-3.4 10.8-3.2 15.1 .6z"],fax:[512,512,[128224,128439],"f1ac","M128 64v96h64V64H386.7L416 93.3V160h64V93.3c0-17-6.7-33.3-18.7-45.3L432 18.7C420 6.7 403.7 0 386.7 0H192c-35.3 0-64 28.7-64 64zM0 160V480c0 17.7 14.3 32 32 32H64c17.7 0 32-14.3 32-32V160c0-17.7-14.3-32-32-32H32c-17.7 0-32 14.3-32 32zm480 32H128V480c0 17.7 14.3 32 32 32H480c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32zM256 256a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm96 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm32 96a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM224 416a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],paragraph:[448,512,[182],"f1dd","M192 32h64H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H384l0 352c0 17.7-14.3 32-32 32s-32-14.3-32-32l0-352H288V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H192c-88.4 0-160-71.6-160-160s71.6-160 160-160z"],"check-to-slot":[576,512,["vote-yea"],"f772","M96 80c0-26.5 21.5-48 48-48H432c26.5 0 48 21.5 48 48V384H96V80zm313 47c-9.4-9.4-24.6-9.4-33.9 0l-111 111-47-47c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l64 64c9.4 9.4 24.6 9.4 33.9 0L409 161c9.4-9.4 9.4-24.6 0-33.9zM0 336c0-26.5 21.5-48 48-48H64V416H512V288h16c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V336z"],"star-half":[576,512,[61731],"f089","M288 0c-12.2 .1-23.3 7-28.6 18L195 150.3 51.4 171.5c-12 1.8-22 10.2-25.7 21.7s-.7 24.2 7.9 32.7L137.8 329 113.2 474.7c-2 12 3 24.2 12.9 31.3s23 8 33.8 2.3L288 439.8V0zM429.9 512c1.1 .1 2.1 .1 3.2 0h-3.2z"],"boxes-stacked":[576,512,[62625,"boxes","boxes-alt"],"f468","M248 0H208c-26.5 0-48 21.5-48 48V160c0 35.3 28.7 64 64 64H352c35.3 0 64-28.7 64-64V48c0-26.5-21.5-48-48-48H328V80c0 8.8-7.2 16-16 16H264c-8.8 0-16-7.2-16-16V0zM64 256c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64H224c35.3 0 64-28.7 64-64V320c0-35.3-28.7-64-64-64H184v80c0 8.8-7.2 16-16 16H120c-8.8 0-16-7.2-16-16V256H64zM352 512H512c35.3 0 64-28.7 64-64V320c0-35.3-28.7-64-64-64H472v80c0 8.8-7.2 16-16 16H408c-8.8 0-16-7.2-16-16V256H352c-15 0-28.8 5.1-39.7 13.8c4.9 10.4 7.7 22 7.7 34.2V464c0 12.2-2.8 23.8-7.7 34.2C323.2 506.9 337 512 352 512z"],link:[640,512,[128279,"chain"],"f0c1","M579.8 267.7c56.5-56.5 56.5-148 0-204.5c-50-50-128.8-56.5-186.3-15.4l-1.6 1.1c-14.4 10.3-17.7 30.3-7.4 44.6s30.3 17.7 44.6 7.4l1.6-1.1c32.1-22.9 76-19.3 103.8 8.6c31.5 31.5 31.5 82.5 0 114L422.3 334.8c-31.5 31.5-82.5 31.5-114 0c-27.9-27.9-31.5-71.8-8.6-103.8l1.1-1.6c10.3-14.4 6.9-34.4-7.4-44.6s-34.4-6.9-44.6 7.4l-1.1 1.6C206.5 251.2 213 330 263 380c56.5 56.5 148 56.5 204.5 0L579.8 267.7zM60.2 244.3c-56.5 56.5-56.5 148 0 204.5c50 50 128.8 56.5 186.3 15.4l1.6-1.1c14.4-10.3 17.7-30.3 7.4-44.6s-30.3-17.7-44.6-7.4l-1.6 1.1c-32.1 22.9-76 19.3-103.8-8.6C74 372 74 321 105.5 289.5L217.7 177.2c31.5-31.5 82.5-31.5 114 0c27.9 27.9 31.5 71.8 8.6 103.9l-1.1 1.6c-10.3 14.4-6.9 34.4 7.4 44.6s34.4 6.9 44.6-7.4l1.1-1.6C433.5 260.8 427 182 377 132c-56.5-56.5-148-56.5-204.5 0L60.2 244.3z"],"ear-listen":[512,512,["assistive-listening-systems"],"f2a2","M398.3 3.4c-15.8-7.9-35-1.5-42.9 14.3c-7.9 15.8-1.5 34.9 14.2 42.9l.4 .2c.4 .2 1.1 .6 2.1 1.2c2 1.2 5 3 8.7 5.6c7.5 5.2 17.6 13.2 27.7 24.2C428.5 113.4 448 146 448 192c0 17.7 14.3 32 32 32s32-14.3 32-32c0-66-28.5-113.4-56.5-143.7C441.6 33.2 427.7 22.2 417.3 15c-5.3-3.7-9.7-6.4-13-8.3c-1.6-1-3-1.7-4-2.2c-.5-.3-.9-.5-1.2-.7l-.4-.2-.2-.1-.1 0 0 0c0 0 0 0-14.3 28.6L398.3 3.4zM128.7 227.5c6.2-56 53.7-99.5 111.3-99.5c61.9 0 112 50.1 112 112c0 29.3-11.2 55.9-29.6 75.9c-17 18.4-34.4 45.1-34.4 78V400c0 26.5-21.5 48-48 48c-17.7 0-32 14.3-32 32s14.3 32 32 32c61.9 0 112-50.1 112-112v-6.1c0-9.8 5.4-21.7 17.4-34.7C398.3 327.9 416 286 416 240c0-97.2-78.8-176-176-176C149.4 64 74.8 132.5 65.1 220.5c-1.9 17.6 10.7 33.4 28.3 35.3s33.4-10.7 35.3-28.3zM32 512a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM192 352a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM41.4 361.4c-12.5 12.5-12.5 32.8 0 45.3l64 64c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-64-64c-12.5-12.5-32.8-12.5-45.3 0zM208 240c0-17.7 14.3-32 32-32s32 14.3 32 32c0 13.3 10.7 24 24 24s24-10.7 24-24c0-44.2-35.8-80-80-80s-80 35.8-80 80c0 13.3 10.7 24 24 24s24-10.7 24-24z"],"tree-city":[640,512,[],"e587","M288 48c0-26.5 21.5-48 48-48h96c26.5 0 48 21.5 48 48V192h40V120c0-13.3 10.7-24 24-24s24 10.7 24 24v72h24c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H432 336c-26.5 0-48-21.5-48-48V48zm64 32v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V80c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16zm16 80c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V176c0-8.8-7.2-16-16-16H368zM352 272v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16zm176-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H528zM512 368v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H528c-8.8 0-16 7.2-16 16zM224 160c0 6-1 11-2 16c20 14 34 38 34 64c0 45-36 80-80 80H160V480c0 18-15 32-32 32c-18 0-32-14-32-32V320H80c-45 0-80-35-80-80c0-26 13-50 33-64c-1-5-1-10-1-16c0-53 42-96 96-96c53 0 96 43 96 96z"],play:[384,512,[9654],"f04b","M73 39c-14.8-9.1-33.4-9.4-48.5-.9S0 62.6 0 80V432c0 17.4 9.4 33.4 24.5 41.9s33.7 8.1 48.5-.9L361 297c14.3-8.7 23-24.2 23-41s-8.7-32.2-23-41L73 39z"],font:[448,512,[],"f031","M254 52.8C249.3 40.3 237.3 32 224 32s-25.3 8.3-30 20.8L57.8 416H32c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32h-1.8l18-48H303.8l18 48H320c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H390.2L254 52.8zM279.8 304H168.2L224 155.1 279.8 304z"],"table-cells-row-lock":[640,512,[],"e67a","M0 96C0 60.7 28.7 32 64 32l384 0c35.3 0 64 28.7 64 64l0 65.1c-37.8 5.4-69.4 29.6-85.2 62.9L360 224l0 64 56 0 0 8.6c-19.1 11.1-32 31.7-32 55.4l-24 0 0 64 24 0 0 64L64 480c-35.3 0-64-28.7-64-64L0 96zM64 224l0 64 88 0 0-64-88 0zm232 0l-88 0 0 64 88 0 0-64zM152 352l-88 0 0 64 88 0 0-64zm56 0l0 64 88 0 0-64-88 0zM528 240c-17.7 0-32 14.3-32 32l0 48 64 0 0-48c0-17.7-14.3-32-32-32zm-80 32c0-44.2 35.8-80 80-80s80 35.8 80 80l0 48c17.7 0 32 14.3 32 32l0 128c0 17.7-14.3 32-32 32l-160 0c-17.7 0-32-14.3-32-32l0-128c0-17.7 14.3-32 32-32l0-48z"],"rupiah-sign":[512,512,[],"e23d","M0 64C0 46.3 14.3 32 32 32h80c79.5 0 144 64.5 144 144c0 58.8-35.2 109.3-85.7 131.7l51.4 128.4c6.6 16.4-1.4 35-17.8 41.6s-35-1.4-41.6-17.8L106.3 320H64V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V288 64zM64 256h48c44.2 0 80-35.8 80-80s-35.8-80-80-80H64V256zm256-96h80c61.9 0 112 50.1 112 112s-50.1 112-112 112H352v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352 192c0-17.7 14.3-32 32-32zm80 160c26.5 0 48-21.5 48-48s-21.5-48-48-48H352v96h48z"],"magnifying-glass":[512,512,[128269,"search"],"f002","M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zM208 352a144 144 0 1 0 0-288 144 144 0 1 0 0 288z"],"table-tennis-paddle-ball":[640,512,[127955,"ping-pong-paddle-ball","table-tennis"],"f45d","M480 288c-50.1 0-93.6 28.8-114.6 70.8L132.9 126.3l.6-.6 60.1-60.1c87.5-87.5 229.3-87.5 316.8 0c67.1 67.1 82.7 166.3 46.8 248.3C535.8 297.6 509 288 480 288zM113.3 151.9L354.1 392.7c-1.4 7.5-2.1 15.3-2.1 23.3c0 23.2 6.2 44.9 16.9 63.7c-3 .2-6.1 .3-9.2 .3H357c-33.9 0-66.5-13.5-90.5-37.5l-9.8-9.8c-13.1-13.1-34.6-12.4-46.8 1.7L152.2 501c-5.8 6.7-14.2 10.7-23 11s-17.5-3.1-23.8-9.4l-32-32c-6.3-6.3-9.7-14.9-9.4-23.8s4.3-17.2 11-23l66.6-57.7c14-12.2 14.8-33.7 1.7-46.8l-9.8-9.8c-24-24-37.5-56.6-37.5-90.5v-2.7c0-22.8 6.1-44.9 17.3-64.3zM480 320a96 96 0 1 1 0 192 96 96 0 1 1 0-192z"],"person-dots-from-line":[576,512,["diagnoses"],"f470","M288 176A88 88 0 1 0 288 0a88 88 0 1 0 0 176zM78.7 372.9c15-12.5 50-34.4 97.3-50.1V432H400V322.7c47.3 15.8 82.3 37.7 97.3 50.1c20.4 17 50.6 14.2 67.6-6.1s14.2-50.6-6.1-67.6c-12-10-30.1-22.5-53.2-35C497.2 278.4 481.7 288 464 288c-26.5 0-48-21.5-48-48c0-4.3 .6-8.4 1.6-12.4C379.1 215.9 335.3 208 288 208c-60.2 0-114.9 12.9-160 29.9c0 .7 0 1.4 0 2.1c0 26.5-21.5 48-48 48c-11.8 0-22.7-4.3-31-11.4c-13.1 8.1-23.7 15.9-31.7 22.5c-20.4 17-23.1 47.2-6.1 67.6s47.2 23.1 67.6 6.1zM24 464c-13.3 0-24 10.7-24 24s10.7 24 24 24H552c13.3 0 24-10.7 24-24s-10.7-24-24-24H24zM224 280a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm104 56a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM96 240a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm368 16a16 16 0 1 0 0-32 16 16 0 1 0 0 32z"],"trash-can-arrow-up":[448,512,["trash-restore-alt"],"f82a","M163.8 0H284.2c12.1 0 23.2 6.8 28.6 17.7L320 32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 96 0 81.7 0 64S14.3 32 32 32h96l7.2-14.3C140.6 6.8 151.7 0 163.8 0zM32 128H416V448c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V128zm192 64c-6.4 0-12.5 2.5-17 7l-80 80c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l39-39V408c0 13.3 10.7 24 24 24s24-10.7 24-24V273.9l39 39c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-80-80c-4.5-4.5-10.6-7-17-7z"],"naira-sign":[448,512,[],"e1f6","M122.6 46.3c-7.8-11.7-22.4-17-35.9-12.9S64 49.9 64 64V256H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64V448c0 17.7 14.3 32 32 32s32-14.3 32-32V320H228.2l97.2 145.8c7.8 11.7 22.4 17 35.9 12.9s22.7-16.5 22.7-30.6V320h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H384V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V256H262.5L122.6 46.3zM305.1 320H320v22.3L305.1 320zM185.5 256H128V169.7L185.5 256z"],"cart-arrow-down":[576,512,[],"f218","M24 0C10.7 0 0 10.7 0 24S10.7 48 24 48H69.5c3.8 0 7.1 2.7 7.9 6.5l51.6 271c6.5 34 36.2 58.5 70.7 58.5H488c13.3 0 24-10.7 24-24s-10.7-24-24-24H199.7c-11.5 0-21.4-8.2-23.6-19.5L170.7 288H459.2c32.6 0 61.1-21.8 69.5-53.3l41-152.3C576.6 57 557.4 32 531.1 32H360V134.1l23-23c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-64 64c-9.4 9.4-24.6 9.4-33.9 0l-64-64c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l23 23V32H120.1C111 12.8 91.6 0 69.5 0H24zM176 512a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm336-48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0z"],"walkie-talkie":[384,512,[],"f8ef","M112 24c0-13.3-10.7-24-24-24S64 10.7 64 24V96H48C21.5 96 0 117.5 0 144V300.1c0 12.7 5.1 24.9 14.1 33.9l3.9 3.9c9 9 14.1 21.2 14.1 33.9V464c0 26.5 21.5 48 48 48H304c26.5 0 48-21.5 48-48V371.9c0-12.7 5.1-24.9 14.1-33.9l3.9-3.9c9-9 14.1-21.2 14.1-33.9V144c0-26.5-21.5-48-48-48H320c0-17.7-14.3-32-32-32s-32 14.3-32 32H224c0-17.7-14.3-32-32-32s-32 14.3-32 32H112V24zm0 136H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],"file-pen":[576,512,[128221,"file-edit"],"f31c","M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384V299.6l-94.7 94.7c-8.2 8.2-14 18.5-16.8 29.7l-15 60.1c-2.3 9.4-1.8 19 1.4 27.8H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zM549.8 235.7l14.4 14.4c15.6 15.6 15.6 40.9 0 56.6l-29.4 29.4-71-71 29.4-29.4c15.6-15.6 40.9-15.6 56.6 0zM311.9 417L441.1 287.8l71 71L382.9 487.9c-4.1 4.1-9.2 7-14.9 8.4l-60.1 15c-5.5 1.4-11.2-.2-15.2-4.2s-5.6-9.7-4.2-15.2l15-60.1c1.4-5.6 4.3-10.8 8.4-14.9z"],receipt:[384,512,[129534],"f543","M14 2.2C22.5-1.7 32.5-.3 39.6 5.8L80 40.4 120.4 5.8c9-7.7 22.3-7.7 31.2 0L192 40.4 232.4 5.8c9-7.7 22.3-7.7 31.2 0L304 40.4 344.4 5.8c7.1-6.1 17.1-7.5 25.6-3.6s14 12.4 14 21.8V488c0 9.4-5.5 17.9-14 21.8s-18.5 2.5-25.6-3.6L304 471.6l-40.4 34.6c-9 7.7-22.3 7.7-31.2 0L192 471.6l-40.4 34.6c-9 7.7-22.3 7.7-31.2 0L80 471.6 39.6 506.2c-7.1 6.1-17.1 7.5-25.6 3.6S0 497.4 0 488V24C0 14.6 5.5 6.1 14 2.2zM96 144c-8.8 0-16 7.2-16 16s7.2 16 16 16H288c8.8 0 16-7.2 16-16s-7.2-16-16-16H96zM80 352c0 8.8 7.2 16 16 16H288c8.8 0 16-7.2 16-16s-7.2-16-16-16H96c-8.8 0-16 7.2-16 16zM96 240c-8.8 0-16 7.2-16 16s7.2 16 16 16H288c8.8 0 16-7.2 16-16s-7.2-16-16-16H96z"],"square-pen":[448,512,["pen-square","pencil-square"],"f14b","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM325.8 139.7l14.4 14.4c15.6 15.6 15.6 40.9 0 56.6l-21.4 21.4-71-71 21.4-21.4c15.6-15.6 40.9-15.6 56.6 0zM119.9 289L225.1 183.8l71 71L190.9 359.9c-4.1 4.1-9.2 7-14.9 8.4l-60.1 15c-5.5 1.4-11.2-.2-15.2-4.2s-5.6-9.7-4.2-15.2l15-60.1c1.4-5.6 4.3-10.8 8.4-14.9z"],"suitcase-rolling":[384,512,[],"f5c1","M144 56c0-4.4 3.6-8 8-8h80c4.4 0 8 3.6 8 8v72H144V56zm176 72H288V56c0-30.9-25.1-56-56-56H152C121.1 0 96 25.1 96 56v72H64c-35.3 0-64 28.7-64 64V416c0 35.3 28.7 64 64 64c0 17.7 14.3 32 32 32s32-14.3 32-32H256c0 17.7 14.3 32 32 32s32-14.3 32-32c35.3 0 64-28.7 64-64V192c0-35.3-28.7-64-64-64zM112 224H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 128H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],"person-circle-exclamation":[576,512,[],"e53f","M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l44.9 74.7c-16.1 17.6-28.6 38.5-36.6 61.5c-1.9-1.8-3.5-3.9-4.9-6.3L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152zM432 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm0 240a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm0-192c-8.8 0-16 7.2-16 16v80c0 8.8 7.2 16 16 16s16-7.2 16-16V288c0-8.8-7.2-16-16-16z"],"chevron-down":[512,512,[],"f078","M233.4 406.6c12.5 12.5 32.8 12.5 45.3 0l192-192c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L256 338.7 86.6 169.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l192 192z"],"battery-full":[576,512,[128267,"battery","battery-5"],"f240","M464 160c8.8 0 16 7.2 16 16V336c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16H464zM80 96C35.8 96 0 131.8 0 176V336c0 44.2 35.8 80 80 80H464c44.2 0 80-35.8 80-80V320c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32V176c0-44.2-35.8-80-80-80H80zm368 96H96V320H448V192z"],"skull-crossbones":[448,512,[128369,9760],"f714","M368 128c0 44.4-25.4 83.5-64 106.4V256c0 17.7-14.3 32-32 32H176c-17.7 0-32-14.3-32-32V234.4c-38.6-23-64-62.1-64-106.4C80 57.3 144.5 0 224 0s144 57.3 144 128zM168 176a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm144-32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM3.4 273.7c7.9-15.8 27.1-22.2 42.9-14.3L224 348.2l177.7-88.8c15.8-7.9 35-1.5 42.9 14.3s1.5 35-14.3 42.9L295.6 384l134.8 67.4c15.8 7.9 22.2 27.1 14.3 42.9s-27.1 22.2-42.9 14.3L224 419.8 46.3 508.6c-15.8 7.9-35 1.5-42.9-14.3s-1.5-35 14.3-42.9L152.4 384 17.7 316.6C1.9 308.7-4.5 289.5 3.4 273.7z"],"code-compare":[512,512,[],"e13a","M320 488c0 9.5-5.6 18.1-14.2 21.9s-18.8 2.3-25.8-4.1l-80-72c-5.1-4.6-7.9-11-7.9-17.8s2.9-13.3 7.9-17.8l80-72c7-6.3 17.2-7.9 25.8-4.1s14.2 12.4 14.2 21.9v40h16c35.3 0 64-28.7 64-64V153.3C371.7 141 352 112.8 352 80c0-44.2 35.8-80 80-80s80 35.8 80 80c0 32.8-19.7 61-48 73.3V320c0 70.7-57.3 128-128 128H320v40zM456 80a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zM192 24c0-9.5 5.6-18.1 14.2-21.9s18.8-2.3 25.8 4.1l80 72c5.1 4.6 7.9 11 7.9 17.8s-2.9 13.3-7.9 17.8l-80 72c-7 6.3-17.2 7.9-25.8 4.1s-14.2-12.4-14.2-21.9V128H176c-35.3 0-64 28.7-64 64V358.7c28.3 12.3 48 40.5 48 73.3c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-32.8 19.7-61 48-73.3V192c0-70.7 57.3-128 128-128h16V24zM56 432a24 24 0 1 0 48 0 24 24 0 1 0 -48 0z"],"list-ul":[512,512,["list-dots"],"f0ca","M64 144a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM192 64c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H192zm0 160c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H192zm0 160c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H192zM64 464a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm48-208a48 48 0 1 0 -96 0 48 48 0 1 0 96 0z"],"school-lock":[640,512,[],"e56f","M302.2 5.4c10.7-7.2 24.8-7.2 35.5 0L473.7 96H592c26.5 0 48 21.5 48 48V272c0-61.9-50.1-112-112-112s-112 50.1-112 112v24.6c-19.1 11.1-32 31.7-32 55.4H320.3l-.3 0c-35.3 0-64 28.7-64 64v96h64v0H48c-26.5 0-48-21.5-48-48V144c0-26.5 21.5-48 48-48H166.3L302.2 5.4zM80 208v64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V208c0-8.8-7.2-16-16-16H96c-8.8 0-16 7.2-16 16zm0 128v64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V336c0-8.8-7.2-16-16-16H96c-8.8 0-16 7.2-16 16zm240-72a88 88 0 1 0 0-176 88 88 0 1 0 0 176zm16-120v16h16c8.8 0 16 7.2 16 16s-7.2 16-16 16H320c-8.8 0-16-7.2-16-16V144c0-8.8 7.2-16 16-16s16 7.2 16 16zm192 96c-17.7 0-32 14.3-32 32v48h64V272c0-17.7-14.3-32-32-32zm-80 32c0-44.2 35.8-80 80-80s80 35.8 80 80v48c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H448c-17.7 0-32-14.3-32-32V352c0-17.7 14.3-32 32-32V272z"],"tower-cell":[576,512,[],"e585","M62.6 2.3C46.2-4.3 27.6 3.6 20.9 20C7.4 53.4 0 89.9 0 128s7.4 74.6 20.9 108c6.6 16.4 25.3 24.3 41.7 17.7S86.9 228.4 80.3 212C69.8 186.1 64 157.8 64 128s5.8-58.1 16.3-84C86.9 27.6 79 9 62.6 2.3zm450.8 0C497 9 489.1 27.6 495.7 44C506.2 69.9 512 98.2 512 128s-5.8 58.1-16.3 84c-6.6 16.4 1.3 35 17.7 41.7s35-1.3 41.7-17.7c13.5-33.4 20.9-69.9 20.9-108s-7.4-74.6-20.9-108C548.4 3.6 529.8-4.3 513.4 2.3zM340.1 165.2c7.5-10.5 11.9-23.3 11.9-37.2c0-35.3-28.7-64-64-64s-64 28.7-64 64c0 13.9 4.4 26.7 11.9 37.2L98.9 466.8c-7.3 16.1-.2 35.1 15.9 42.4s35.1 .2 42.4-15.9L177.7 448H398.3l20.6 45.2c7.3 16.1 26.3 23.2 42.4 15.9s23.2-26.3 15.9-42.4L340.1 165.2zM369.2 384H206.8l14.5-32H354.7l14.5 32zM288 205.3L325.6 288H250.4L288 205.3zM163.3 73.6c5.3-12.1-.2-26.3-12.4-31.6s-26.3 .2-31.6 12.4C109.5 77 104 101.9 104 128s5.5 51 15.3 73.6c5.3 12.1 19.5 17.7 31.6 12.4s17.7-19.5 12.4-31.6C156 165.8 152 147.4 152 128s4-37.8 11.3-54.4zM456.7 54.4c-5.3-12.1-19.5-17.7-31.6-12.4s-17.7 19.5-12.4 31.6C420 90.2 424 108.6 424 128s-4 37.8-11.3 54.4c-5.3 12.1 .2 26.3 12.4 31.6s26.3-.2 31.6-12.4C466.5 179 472 154.1 472 128s-5.5-51-15.3-73.6z"],"down-long":[320,512,["long-arrow-alt-down"],"f309","M2 334.5c-3.8 8.8-2 19 4.6 26l136 144c4.5 4.8 10.8 7.5 17.4 7.5s12.9-2.7 17.4-7.5l136-144c6.6-7 8.4-17.2 4.6-26s-12.5-14.5-22-14.5l-72 0 0-288c0-17.7-14.3-32-32-32L128 0C110.3 0 96 14.3 96 32l0 288-72 0c-9.6 0-18.2 5.7-22 14.5z"],"ranking-star":[640,512,[],"e561","M353.8 54.1L330.2 6.3c-3.9-8.3-16.1-8.6-20.4 0L286.2 54.1l-52.3 7.5c-9.3 1.4-13.3 12.9-6.4 19.8l38 37-9 52.1c-1.4 9.3 8.2 16.5 16.8 12.2l46.9-24.8 46.6 24.4c8.6 4.3 18.3-2.9 16.8-12.2l-9-52.1 38-36.6c6.8-6.8 2.9-18.3-6.4-19.8l-52.3-7.5zM256 256c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H384c17.7 0 32-14.3 32-32V288c0-17.7-14.3-32-32-32H256zM32 320c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H160c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32H32zm416 96v64c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32V416c0-17.7-14.3-32-32-32H480c-17.7 0-32 14.3-32 32z"],"chess-king":[448,512,[9818],"f43f","M224 0c17.7 0 32 14.3 32 32V48h16c17.7 0 32 14.3 32 32s-14.3 32-32 32H256v48H408c22.1 0 40 17.9 40 40c0 5.3-1 10.5-3.1 15.4L368 400H80L3.1 215.4C1 210.5 0 205.3 0 200c0-22.1 17.9-40 40-40H192V112H176c-17.7 0-32-14.3-32-32s14.3-32 32-32h16V32c0-17.7 14.3-32 32-32zM38.6 473.4L80 432H368l41.4 41.4c4.2 4.2 6.6 10 6.6 16c0 12.5-10.1 22.6-22.6 22.6H54.6C42.1 512 32 501.9 32 489.4c0-6 2.4-11.8 6.6-16z"],"person-harassing":[576,512,[],"e549","M192 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM59.4 304.5L88 256.9V480c0 17.7 14.3 32 32 32s32-14.3 32-32V352h16V480c0 17.7 14.3 32 32 32s32-14.3 32-32V235.3l47.4 57.1c11.3 13.6 31.5 15.5 45.1 4.2s15.5-31.5 4.2-45.1l-73.7-88.9c-18.2-22-45.3-34.7-73.9-34.7H145.1c-33.7 0-64.9 17.7-82.3 46.6l-58.3 97c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9zM480 240a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM464 344v58.7l-41.4-41.4c-7.3-7.3-17.6-10.6-27.8-9s-18.9 8.1-23.5 17.3l-48 96c-7.9 15.8-1.5 35 14.3 42.9s35 1.5 42.9-14.3L408.8 438l54.7 54.7c12.4 12.4 29.1 19.3 46.6 19.3c36.4 0 65.9-29.5 65.9-65.9V344c0-30.9-25.1-56-56-56s-56 25.1-56 56zM288 48c0 8.8 7.2 16 16 16h56c8.8 0 16-7.2 16-16s-7.2-16-16-16H304c-8.8 0-16 7.2-16 16zm-.8 49.7c-7.9-4-17.5-.7-21.5 7.2s-.7 17.5 7.2 21.5l48 24c7.9 4 17.5 .7 21.5-7.2s.7-17.5-7.2-21.5l-48-24z"],"brazilian-real-sign":[512,512,[],"e46c","M400 0c17.7 0 32 14.3 32 32V50.2c12.5 2.3 24.7 6.4 36.2 12.1l10.1 5.1c15.8 7.9 22.2 27.1 14.3 42.9s-27.1 22.2-42.9 14.3l-10.2-5.1c-9.9-5-20.9-7.5-32-7.5h-1.7c-29.8 0-53.9 24.1-53.9 53.9c0 22 13.4 41.8 33.9 50l52 20.8c44.7 17.9 74.1 61.2 74.1 109.4v3.4c0 51.2-33.6 94.6-80 109.2V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V460.6c-15-3.5-29.4-9.7-42.3-18.3l-23.4-15.6c-14.7-9.8-18.7-29.7-8.9-44.4s29.7-18.7 44.4-8.9L361.2 389c10.8 7.2 23.4 11 36.3 11c27.9 0 50.5-22.6 50.5-50.5v-3.4c0-22-13.4-41.8-33.9-50l-52-20.8C317.3 257.4 288 214.1 288 165.9C288 114 321.5 70 368 54.2V32c0-17.7 14.3-32 32-32zM0 64C0 46.3 14.3 32 32 32h80c79.5 0 144 64.5 144 144c0 58.8-35.2 109.3-85.7 131.7l51.4 128.4c6.6 16.4-1.4 35-17.8 41.6s-35-1.4-41.6-17.8L106.3 320H64V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V288 64zM64 256h48c44.2 0 80-35.8 80-80s-35.8-80-80-80H64V256z"],"landmark-dome":[512,512,["landmark-alt"],"f752","M248 0h16c13.3 0 24 10.7 24 24V34.7C368.4 48.1 431.9 111.6 445.3 192H448c17.7 0 32 14.3 32 32s-14.3 32-32 32H64c-17.7 0-32-14.3-32-32s14.3-32 32-32h2.7C80.1 111.6 143.6 48.1 224 34.7V24c0-13.3 10.7-24 24-24zM64 288h64V416h40V288h64V416h48V288h64V416h40V288h64V420.3c.6 .3 1.2 .7 1.7 1.1l48 32c11.7 7.8 17 22.4 12.9 35.9S494.1 512 480 512H32c-14.1 0-26.5-9.2-30.6-22.7s1.1-28.1 12.9-35.9l48-32c.6-.4 1.2-.7 1.8-1.1V288z"],"arrow-up":[384,512,[8593],"f062","M214.6 41.4c-12.5-12.5-32.8-12.5-45.3 0l-160 160c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L160 141.2V448c0 17.7 14.3 32 32 32s32-14.3 32-32V141.2L329.4 246.6c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-160-160z"],tv:[640,512,[63717,"television","tv-alt"],"f26c","M64 64V352H576V64H64zM0 64C0 28.7 28.7 0 64 0H576c35.3 0 64 28.7 64 64V352c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM128 448H512c17.7 0 32 14.3 32 32s-14.3 32-32 32H128c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],shrimp:[512,512,[129424],"e448","M64 32C28.7 32 0 60.7 0 96s28.7 64 64 64h1c3.7 88.9 77 160 167 160h56V128H264 88.8 64c-17.7 0-32-14.3-32-32s14.3-32 32-32H464c8.8 0 16-7.2 16-16s-7.2-16-16-16H64zM224 456c0 13.3 10.7 24 24 24h72V407.8l-64.1-22.4c-12.5-4.4-26.2 2.2-30.6 14.7s2.2 26.2 14.7 30.6l4.5 1.6C233 433.9 224 443.9 224 456zm128 23.3c36.4-3.3 69.5-17.6 96.1-39.6l-86.5-34.6c-3 1.8-6.2 3.2-9.6 4.3v69.9zM472.6 415c24.6-30.3 39.4-68.9 39.4-111c0-12.3-1.3-24.3-3.7-35.9L382.8 355.1c.8 3.4 1.2 7 1.2 10.6c0 4.6-.7 9-1.9 13.1L472.6 415zM336 128H320V320h18.3c9.9 0 19.1 3.2 26.6 8.5l133.5-92.4C471.8 172.6 409.1 128 336 128zM168 192a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z"],"list-check":[512,512,["tasks"],"f0ae","M152.1 38.2c9.9 8.9 10.7 24 1.8 33.9l-72 80c-4.4 4.9-10.6 7.8-17.2 7.9s-12.9-2.4-17.6-7L7 113C-2.3 103.6-2.3 88.4 7 79s24.6-9.4 33.9 0l22.1 22.1 55.1-61.2c8.9-9.9 24-10.7 33.9-1.8zm0 160c9.9 8.9 10.7 24 1.8 33.9l-72 80c-4.4 4.9-10.6 7.8-17.2 7.9s-12.9-2.4-17.6-7L7 273c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l22.1 22.1 55.1-61.2c8.9-9.9 24-10.7 33.9-1.8zM224 96c0-17.7 14.3-32 32-32H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H256c-17.7 0-32-14.3-32-32zm0 160c0-17.7 14.3-32 32-32H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H256c-17.7 0-32-14.3-32-32zM160 416c0-17.7 14.3-32 32-32H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H192c-17.7 0-32-14.3-32-32zM48 368a48 48 0 1 1 0 96 48 48 0 1 1 0-96z"],"jug-detergent":[384,512,[],"e519","M96 24c0-13.3 10.7-24 24-24h80c13.3 0 24 10.7 24 24V48h8c13.3 0 24 10.7 24 24s-10.7 24-24 24H88C74.7 96 64 85.3 64 72s10.7-24 24-24h8V24zM0 256c0-70.7 57.3-128 128-128H256c70.7 0 128 57.3 128 128V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V256zm256 0v96c0 17.7 14.3 32 32 32s32-14.3 32-32V256c0-17.7-14.3-32-32-32s-32 14.3-32 32z"],"circle-user":[512,512,[62142,"user-circle"],"f2bd","M399 384.2C376.9 345.8 335.4 320 288 320H224c-47.4 0-88.9 25.8-111 64.2c35.2 39.2 86.2 63.8 143 63.8s107.8-24.7 143-63.8zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm256 16a72 72 0 1 0 0-144 72 72 0 1 0 0 144z"],"user-shield":[640,512,[],"f505","M224 256A128 128 0 1 0 224 0a128 128 0 1 0 0 256zm-45.7 48C79.8 304 0 383.8 0 482.3C0 498.7 13.3 512 29.7 512H418.3c1.8 0 3.5-.2 5.3-.5c-76.3-55.1-99.8-141-103.1-200.2c-16.1-4.8-33.1-7.3-50.7-7.3H178.3zm308.8-78.3l-120 48C358 277.4 352 286.2 352 296c0 63.3 25.9 168.8 134.8 214.2c5.9 2.5 12.6 2.5 18.5 0C614.1 464.8 640 359.3 640 296c0-9.8-6-18.6-15.1-22.3l-120-48c-5.7-2.3-12.1-2.3-17.8 0zM591.4 312c-3.9 50.7-27.2 116.7-95.4 149.7V273.8L591.4 312z"],wind:[512,512,[],"f72e","M288 32c0 17.7 14.3 32 32 32h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H352c53 0 96-43 96-96s-43-96-96-96H320c-17.7 0-32 14.3-32 32zm64 352c0 17.7 14.3 32 32 32h32c53 0 96-43 96-96s-43-96-96-96H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H384c-17.7 0-32 14.3-32 32zM128 512h32c53 0 96-43 96-96s-43-96-96-96H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H160c17.7 0 32 14.3 32 32s-14.3 32-32 32H128c-17.7 0-32 14.3-32 32s14.3 32 32 32z"],"car-burst":[640,512,["car-crash"],"f5e1","M176 8c-6.6 0-12.4 4-14.9 10.1l-29.4 74L55.6 68.9c-6.3-1.9-13.1 .2-17.2 5.3s-4.6 12.2-1.4 17.9l39.5 69.1L10.9 206.4c-5.4 3.7-8 10.3-6.5 16.7s6.7 11.2 13.1 12.2l78.7 12.2L90.6 327c-.5 6.5 3.1 12.7 9 15.5s12.9 1.8 17.8-2.6l35.3-32.5 9.5-35.4 10.4-38.6c8-29.9 30.5-52.1 57.9-60.9l41-59.2c11.3-16.3 26.4-28.9 43.5-37.2c-.4-.6-.8-1.2-1.3-1.8c-4.1-5.1-10.9-7.2-17.2-5.3L220.3 92.1l-29.4-74C188.4 12 182.6 8 176 8zM367.7 161.5l135.6 36.3c6.5 1.8 11.3 7.4 11.8 14.2l4.6 56.5-201.5-54 32.2-46.6c3.8-5.6 10.8-8.1 17.3-6.4zm-69.9-30l-47.9 69.3c-21.6 3-40.3 18.6-46.3 41l-10.4 38.6-16.6 61.8-8.3 30.9c-4.6 17.1 5.6 34.6 22.6 39.2l15.5 4.1c17.1 4.6 34.6-5.6 39.2-22.6l8.3-30.9 247.3 66.3-8.3 30.9c-4.6 17.1 5.6 34.6 22.6 39.2l15.5 4.1c17.1 4.6 34.6-5.6 39.2-22.6l8.3-30.9L595 388l10.4-38.6c6-22.4-2.5-45.2-19.6-58.7l-6.8-84c-2.7-33.7-26.4-62-59-70.8L384.2 99.7c-32.7-8.8-67.3 4-86.5 31.8zm-17 131a24 24 0 1 1 -12.4 46.4 24 24 0 1 1 12.4-46.4zm217.9 83.2A24 24 0 1 1 545 358.1a24 24 0 1 1 -46.4-12.4z"],y:[384,512,[121],"59","M58 45.4C47.8 31 27.8 27.7 13.4 38S-4.3 68.2 6 82.6L160 298.3V448c0 17.7 14.3 32 32 32s32-14.3 32-32V298.3L378 82.6c10.3-14.4 6.9-34.4-7.4-44.6S336.2 31 326 45.4L192 232.9 58 45.4z"],"person-snowboarding":[512,512,[127938,"snowboarding"],"f7ce","M209.7 3.4c15.8-7.9 35-1.5 42.9 14.3l25 50 42.4 8.5c19.5 3.9 37.8 12.3 53.5 24.5l126.1 98.1c14 10.9 16.5 31 5.6 44.9s-31 16.5-44.9 5.6l-72.1-56.1-71.5 31.8 33.1 27.6c23.2 19.3 33.5 50 26.7 79.4l-17.4 75.2c-2.2 9.4-8.2 16.8-16.1 21l86.5 33.1c4.6 1.8 9.4 2.6 14.3 2.6H472c13.3 0 24 10.7 24 24s-10.7 24-24 24H443.8c-10.8 0-21.4-2-31.5-5.8L60.1 371.3c-11.5-4.4-22-11.2-30.8-20L7 329c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l22.4 22.4c4 4 8.7 7.1 14 9.1l22.4 8.6c-.8-1.6-1.5-3.2-2.1-4.9c-5.6-16.8 3.5-34.9 20.2-40.5L192 264.9l0-53.2c0-24.2 13.7-46.4 35.4-57.2l45.2-22.6-7.5-1.5c-19.4-3.9-35.9-16.5-44.7-34.1l-25-50c-7.9-15.8-1.5-35 14.3-42.9zM139 350.1l159 60.9c-2.1-5.6-2.6-11.9-1.1-18.2l17.4-75.2c1.4-5.9-.7-12-5.3-15.9l-52.8-44 0 18.8c0 20.7-13.2 39-32.8 45.5L139 350.1zM432 0a48 48 0 1 1 0 96 48 48 0 1 1 0-96z"],"truck-fast":[640,512,["shipping-fast"],"f48b","M112 0C85.5 0 64 21.5 64 48V96H16c-8.8 0-16 7.2-16 16s7.2 16 16 16H64 272c8.8 0 16 7.2 16 16s-7.2 16-16 16H64 48c-8.8 0-16 7.2-16 16s7.2 16 16 16H64 240c8.8 0 16 7.2 16 16s-7.2 16-16 16H64 16c-8.8 0-16 7.2-16 16s7.2 16 16 16H64 208c8.8 0 16 7.2 16 16s-7.2 16-16 16H64V416c0 53 43 96 96 96s96-43 96-96H384c0 53 43 96 96 96s96-43 96-96h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V288 256 237.3c0-17-6.7-33.3-18.7-45.3L512 114.7c-12-12-28.3-18.7-45.3-18.7H416V48c0-26.5-21.5-48-48-48H112zM544 237.3V256H416V160h50.7L544 237.3zM160 368a48 48 0 1 1 0 96 48 48 0 1 1 0-96zm272 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0z"],fish:[576,512,[128031],"f578","M180.5 141.5C219.7 108.5 272.6 80 336 80s116.3 28.5 155.5 61.5c39.1 33 66.9 72.4 81 99.8c4.7 9.2 4.7 20.1 0 29.3c-14.1 27.4-41.9 66.8-81 99.8C452.3 403.5 399.4 432 336 432s-116.3-28.5-155.5-61.5c-16.2-13.7-30.5-28.5-42.7-43.1L48.1 379.6c-12.5 7.3-28.4 5.3-38.7-4.9S-3 348.7 4.2 336.1L50 256 4.2 175.9c-7.2-12.6-5-28.4 5.3-38.6s26.1-12.2 38.7-4.9l89.7 52.3c12.2-14.6 26.5-29.4 42.7-43.1zM448 256a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z"],"user-graduate":[448,512,[],"f501","M219.3 .5c3.1-.6 6.3-.6 9.4 0l200 40C439.9 42.7 448 52.6 448 64s-8.1 21.3-19.3 23.5L352 102.9V160c0 70.7-57.3 128-128 128s-128-57.3-128-128V102.9L48 93.3v65.1l15.7 78.4c.9 4.7-.3 9.6-3.3 13.3s-7.6 5.9-12.4 5.9H16c-4.8 0-9.3-2.1-12.4-5.9s-4.3-8.6-3.3-13.3L16 158.4V86.6C6.5 83.3 0 74.3 0 64C0 52.6 8.1 42.7 19.3 40.5l200-40zM111.9 327.7c10.5-3.4 21.8 .4 29.4 8.5l71 75.5c6.3 6.7 17 6.7 23.3 0l71-75.5c7.6-8.1 18.9-11.9 29.4-8.5C401 348.6 448 409.4 448 481.3c0 17-13.8 30.7-30.7 30.7H30.7C13.8 512 0 498.2 0 481.3c0-71.9 47-132.7 111.9-153.6z"],"circle-half-stroke":[512,512,[9680,"adjust"],"f042","M448 256c0-106-86-192-192-192V448c106 0 192-86 192-192zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256z"],clapperboard:[512,512,[],"e131","M448 32H361.9l-1 1-127 127h92.1l1-1L453.8 32.3c-1.9-.2-3.8-.3-5.8-.3zm64 128V96c0-15.1-5.3-29.1-14-40l-104 104H512zM294.1 32H201.9l-1 1L73.9 160h92.1l1-1 127-127zM64 32C28.7 32 0 60.7 0 96v64H6.1l1-1 127-127H64zM512 192H0V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V192z"],"circle-radiation":[512,512,[9762,"radiation-alt"],"f7ba","M256 64a192 192 0 1 1 0 384 192 192 0 1 1 0-384zm0 448A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM200 256c0-20.7 11.3-38.8 28-48.5l-36-62.3c-8.8-15.3-28.7-20.8-42-9c-25.6 22.6-43.9 53.3-50.9 88.1C95.7 241.5 110.3 256 128 256l72 0zm28 48.5l-36 62.4c-8.8 15.3-3.6 35.2 13.1 40.8c16 5.4 33.1 8.3 50.9 8.3s34.9-2.9 50.9-8.3c16.7-5.6 21.9-25.5 13.1-40.8l-36-62.4c-8.2 4.8-17.8 7.5-28 7.5s-19.8-2.7-28-7.5zM312 256l72 0c17.7 0 32.3-14.5 28.8-31.8c-7-34.8-25.3-65.5-50.9-88.1c-13.2-11.7-33.1-6.3-42 9l-36 62.3c16.7 9.7 28 27.8 28 48.5zm-56 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],baseball:[512,512,[129358,9918,"baseball-ball"],"f433","M62.7 223.4c-4.8 .4-9.7 .6-14.7 .6c-15.6 0-30.8-2-45.2-5.9C19.2 107.1 107.1 19.2 218.1 2.8C222 17.2 224 32.4 224 48c0 4.9-.2 9.8-.6 14.7c-.7 8.8 5.8 16.5 14.6 17.3s16.5-5.8 17.3-14.6c.5-5.7 .7-11.5 .7-17.3c0-16.5-1.9-32.6-5.6-47.9c1.8 0 3.7-.1 5.6-.1C397.4 0 512 114.6 512 256c0 1.9 0 3.7-.1 5.6c-15.4-3.6-31.4-5.6-47.9-5.6c-5.8 0-11.6 .2-17.3 .7c-8.8 .7-15.4 8.5-14.6 17.3s8.5 15.4 17.3 14.6c4.8-.4 9.7-.6 14.7-.6c15.6 0 30.8 2 45.2 5.9C492.8 404.9 404.9 492.8 293.9 509.2C290 494.8 288 479.6 288 464c0-4.9 .2-9.8 .6-14.7c.7-8.8-5.8-16.5-14.6-17.3s-16.5 5.8-17.3 14.6c-.5 5.7-.7 11.5-.7 17.3c0 16.5 1.9 32.6 5.6 47.9c-1.8 0-3.7 .1-5.6 .1C114.6 512 0 397.4 0 256c0-1.9 0-3.7 .1-5.6C15.4 254.1 31.5 256 48 256c5.8 0 11.6-.2 17.3-.7c8.8-.7 15.4-8.5 14.6-17.3s-8.5-15.4-17.3-14.6zM121.3 208c-8 3.7-11.6 13.2-7.9 21.2s13.2 11.6 21.2 7.9c45.2-20.8 81.7-57.2 102.5-102.5c3.7-8 .2-17.5-7.9-21.2s-17.5-.2-21.2 7.9c-17.6 38.3-48.5 69.2-86.7 86.7zm277.2 74.7c-3.7-8-13.2-11.6-21.2-7.9c-45.2 20.8-81.7 57.2-102.5 102.5c-3.7 8-.2 17.5 7.9 21.2s17.5 .2 21.2-7.9c17.6-38.3 48.5-69.2 86.7-86.7c8-3.7 11.6-13.2 7.9-21.2z"],"jet-fighter-up":[512,512,[],"e518","M270.7 9.7C268.2 3.8 262.4 0 256 0s-12.2 3.8-14.7 9.7L197.2 112.6c-3.4 8-5.2 16.5-5.2 25.2v77l-144 84V280c0-13.3-10.7-24-24-24s-24 10.7-24 24v56 32 24c0 13.3 10.7 24 24 24s24-10.7 24-24v-8H192v32.7L133.5 468c-3.5 3-5.5 7.4-5.5 12v16c0 8.8 7.2 16 16 16h96V448c0-8.8 7.2-16 16-16s16 7.2 16 16v64h96c8.8 0 16-7.2 16-16V480c0-4.6-2-9-5.5-12L320 416.7V384H464v8c0 13.3 10.7 24 24 24s24-10.7 24-24V368 336 280c0-13.3-10.7-24-24-24s-24 10.7-24 24v18.8l-144-84v-77c0-8.7-1.8-17.2-5.2-25.2L270.7 9.7z"],"diagram-project":[576,512,["project-diagram"],"f542","M0 80C0 53.5 21.5 32 48 32h96c26.5 0 48 21.5 48 48V96H384V80c0-26.5 21.5-48 48-48h96c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H432c-26.5 0-48-21.5-48-48V160H192v16c0 1.7-.1 3.4-.3 5L272 288h96c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H272c-26.5 0-48-21.5-48-48V336c0-1.7 .1-3.4 .3-5L144 224H48c-26.5 0-48-21.5-48-48V80z"],copy:[448,512,[],"f0c5","M208 0H332.1c12.7 0 24.9 5.1 33.9 14.1l67.9 67.9c9 9 14.1 21.2 14.1 33.9V336c0 26.5-21.5 48-48 48H208c-26.5 0-48-21.5-48-48V48c0-26.5 21.5-48 48-48zM48 128h80v64H64V448H256V416h64v48c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V176c0-26.5 21.5-48 48-48z"],"volume-xmark":[576,512,["volume-mute","volume-times"],"f6a9","M301.1 34.8C312.6 40 320 51.4 320 64V448c0 12.6-7.4 24-18.9 29.2s-25 3.1-34.4-5.3L131.8 352H64c-35.3 0-64-28.7-64-64V224c0-35.3 28.7-64 64-64h67.8L266.7 40.1c9.4-8.4 22.9-10.4 34.4-5.3zM425 167l55 55 55-55c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-55 55 55 55c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-55-55-55 55c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l55-55-55-55c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0z"],"hand-sparkles":[640,512,[],"e05d","M320 0c17.7 0 32 14.3 32 32V240c0 8.8 7.2 16 16 16s16-7.2 16-16V64c0-17.7 14.3-32 32-32s32 14.3 32 32V240c0 8.8 7.2 16 16 16s16-7.2 16-16V128c0-17.7 14.3-32 32-32s32 14.3 32 32V323.1c-11.9 4.8-21.3 14.9-25 27.8l-8.9 31.2L478.9 391C460.6 396.3 448 413 448 432c0 18.9 12.5 35.6 30.6 40.9C448.4 497.4 409.9 512 368 512H348.8c-59.6 0-116.9-22.9-160-64L76.4 341c-16-15.2-16.6-40.6-1.4-56.6s40.6-16.6 56.6-1.4l60.5 57.6c0-1.5-.1-3.1-.1-4.6V64c0-17.7 14.3-32 32-32s32 14.3 32 32V240c0 8.8 7.2 16 16 16s16-7.2 16-16V32c0-17.7 14.3-32 32-32zm-7.3 326.6c-1.1-3.9-4.7-6.6-8.7-6.6s-7.6 2.7-8.7 6.6L288 352l-25.4 7.3c-3.9 1.1-6.6 4.7-6.6 8.7s2.7 7.6 6.6 8.7L288 384l7.3 25.4c1.1 3.9 4.7 6.6 8.7 6.6s7.6-2.7 8.7-6.6L320 384l25.4-7.3c3.9-1.1 6.6-4.7 6.6-8.7s-2.7-7.6-6.6-8.7L320 352l-7.3-25.4zM104 120l48.3 13.8c4.6 1.3 7.7 5.5 7.7 10.2s-3.1 8.9-7.7 10.2L104 168 90.2 216.3c-1.3 4.6-5.5 7.7-10.2 7.7s-8.9-3.1-10.2-7.7L56 168 7.7 154.2C3.1 152.9 0 148.7 0 144s3.1-8.9 7.7-10.2L56 120 69.8 71.7C71.1 67.1 75.3 64 80 64s8.9 3.1 10.2 7.7L104 120zM584 408l48.3 13.8c4.6 1.3 7.7 5.5 7.7 10.2s-3.1 8.9-7.7 10.2L584 456l-13.8 48.3c-1.3 4.6-5.5 7.7-10.2 7.7s-8.9-3.1-10.2-7.7L536 456l-48.3-13.8c-4.6-1.3-7.7-5.5-7.7-10.2s3.1-8.9 7.7-10.2L536 408l13.8-48.3c1.3-4.6 5.5-7.7 10.2-7.7s8.9 3.1 10.2 7.7L584 408z"],grip:[448,512,["grip-horizontal"],"f58d","M128 136c0-22.1-17.9-40-40-40L40 96C17.9 96 0 113.9 0 136l0 48c0 22.1 17.9 40 40 40H88c22.1 0 40-17.9 40-40l0-48zm0 192c0-22.1-17.9-40-40-40H40c-22.1 0-40 17.9-40 40l0 48c0 22.1 17.9 40 40 40H88c22.1 0 40-17.9 40-40V328zm32-192v48c0 22.1 17.9 40 40 40h48c22.1 0 40-17.9 40-40V136c0-22.1-17.9-40-40-40l-48 0c-22.1 0-40 17.9-40 40zM288 328c0-22.1-17.9-40-40-40H200c-22.1 0-40 17.9-40 40l0 48c0 22.1 17.9 40 40 40h48c22.1 0 40-17.9 40-40V328zm32-192v48c0 22.1 17.9 40 40 40h48c22.1 0 40-17.9 40-40V136c0-22.1-17.9-40-40-40l-48 0c-22.1 0-40 17.9-40 40zM448 328c0-22.1-17.9-40-40-40H360c-22.1 0-40 17.9-40 40v48c0 22.1 17.9 40 40 40h48c22.1 0 40-17.9 40-40V328z"],"share-from-square":[576,512,[61509,"share-square"],"f14d","M352 224H305.5c-45 0-81.5 36.5-81.5 81.5c0 22.3 10.3 34.3 19.2 40.5c6.8 4.7 12.8 12 12.8 20.3c0 9.8-8 17.8-17.8 17.8h-2.5c-2.4 0-4.8-.4-7.1-1.4C210.8 374.8 128 333.4 128 240c0-79.5 64.5-144 144-144h80V34.7C352 15.5 367.5 0 386.7 0c8.6 0 16.8 3.2 23.2 8.9L548.1 133.3c7.6 6.8 11.9 16.5 11.9 26.7s-4.3 19.9-11.9 26.7l-139 125.1c-5.9 5.3-13.5 8.2-21.4 8.2H384c-17.7 0-32-14.3-32-32V224zM80 96c-8.8 0-16 7.2-16 16V432c0 8.8 7.2 16 16 16H400c8.8 0 16-7.2 16-16V384c0-17.7 14.3-32 32-32s32 14.3 32 32v48c0 44.2-35.8 80-80 80H80c-44.2 0-80-35.8-80-80V112C0 67.8 35.8 32 80 32h48c17.7 0 32 14.3 32 32s-14.3 32-32 32H80z"],"child-combatant":[576,512,["child-rifle"],"e4e0","M176 128A64 64 0 1 0 176 0a64 64 0 1 0 0 128zm-8 352V352h16V480c0 17.7 14.3 32 32 32s32-14.3 32-32V300.5L260.9 321c9.4 15 29.2 19.4 44.1 10s19.4-29.2 10-44.1l-51.7-82.1c-17.6-27.9-48.3-44.9-81.2-44.9H169.8c-33 0-63.7 16.9-81.2 44.9L36.9 287c-9.4 15-4.9 34.7 10 44.1s34.7 4.9 44.1-10L104 300.5V480c0 17.7 14.3 32 32 32s32-14.3 32-32zM448 0H432 416c-8.8 0-16 7.2-16 16s7.2 16 16 16V132.3c-9.6 5.5-16 15.9-16 27.7v32c-17.7 0-32 14.3-32 32V368c0 17.7 14.3 32 32 32h16v96c0 8.8 7.2 16 16 16h59.5c10.4 0 18-9.8 15.5-19.9L484 400h44c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H480V325.3l53.1-17.7c6.5-2.2 10.9-8.3 10.9-15.2V208c0-8.8-7.2-16-16-16H512c-8.8 0-16 7.2-16 16v56l-16 5.3V160c0-11.8-6.4-22.2-16-27.7V16c0-8.8-7.2-16-16-16z"],gun:[576,512,[],"e19b","M528 56c0-13.3-10.7-24-24-24s-24 10.7-24 24v8H32C14.3 64 0 78.3 0 96V208c0 17.7 14.3 32 32 32H42c20.8 0 36.1 19.6 31 39.8L33 440.2c-2.4 9.6-.2 19.7 5.8 27.5S54.1 480 64 480h96c14.7 0 27.5-10 31-24.2L217 352H321.4c23.7 0 44.8-14.9 52.7-37.2L400.9 240H432c8.5 0 16.6-3.4 22.6-9.4L477.3 208H544c17.7 0 32-14.3 32-32V96c0-17.7-14.3-32-32-32H528V56zM321.4 304H229l16-64h105l-21 58.7c-1.1 3.2-4.2 5.3-7.5 5.3zM80 128H464c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],"square-phone":[448,512,["phone-square"],"f098","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm90.7 96.7c9.7-2.6 19.9 2.3 23.7 11.6l20 48c3.4 8.2 1 17.6-5.8 23.2L168 231.7c16.6 35.2 45.1 63.7 80.3 80.3l20.2-24.7c5.6-6.8 15-9.2 23.2-5.8l48 20c9.3 3.9 14.2 14 11.6 23.7l-12 44C336.9 378 329 384 320 384C196.3 384 96 283.7 96 160c0-9 6-16.9 14.7-19.3l44-12z"],plus:[448,512,[10133,61543,"add"],"2b","M256 80c0-17.7-14.3-32-32-32s-32 14.3-32 32V224H48c-17.7 0-32 14.3-32 32s14.3 32 32 32H192V432c0 17.7 14.3 32 32 32s32-14.3 32-32V288H400c17.7 0 32-14.3 32-32s-14.3-32-32-32H256V80z"],expand:[448,512,[],"f065","M32 32C14.3 32 0 46.3 0 64v96c0 17.7 14.3 32 32 32s32-14.3 32-32V96h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H32zM64 352c0-17.7-14.3-32-32-32s-32 14.3-32 32v96c0 17.7 14.3 32 32 32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H64V352zM320 32c-17.7 0-32 14.3-32 32s14.3 32 32 32h64v64c0 17.7 14.3 32 32 32s32-14.3 32-32V64c0-17.7-14.3-32-32-32H320zM448 352c0-17.7-14.3-32-32-32s-32 14.3-32 32v64H320c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c17.7 0 32-14.3 32-32V352z"],computer:[640,512,[],"e4e5","M384 96V320H64L64 96H384zM64 32C28.7 32 0 60.7 0 96V320c0 35.3 28.7 64 64 64H181.3l-10.7 32H96c-17.7 0-32 14.3-32 32s14.3 32 32 32H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H277.3l-10.7-32H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm464 0c-26.5 0-48 21.5-48 48V432c0 26.5 21.5 48 48 48h64c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48H528zm16 64h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H544c-8.8 0-16-7.2-16-16s7.2-16 16-16zm-16 80c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H544c-8.8 0-16-7.2-16-16zm32 160a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],xmark:[384,512,[128473,10005,10006,10060,215,"close","multiply","remove","times"],"f00d","M342.6 150.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L192 210.7 86.6 105.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L146.7 256 41.4 361.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L192 301.3 297.4 406.6c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L237.3 256 342.6 150.6z"],"arrows-up-down-left-right":[512,512,["arrows"],"f047","M278.6 9.4c-12.5-12.5-32.8-12.5-45.3 0l-64 64c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l9.4-9.4V224H109.3l9.4-9.4c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-64 64c-12.5 12.5-12.5 32.8 0 45.3l64 64c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-9.4-9.4H224V402.7l-9.4-9.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l64 64c12.5 12.5 32.8 12.5 45.3 0l64-64c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-9.4 9.4V288H402.7l-9.4 9.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l64-64c12.5-12.5 12.5-32.8 0-45.3l-64-64c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l9.4 9.4H288V109.3l9.4 9.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-64-64z"],"chalkboard-user":[640,512,["chalkboard-teacher"],"f51c","M160 64c0-35.3 28.7-64 64-64H576c35.3 0 64 28.7 64 64V352c0 35.3-28.7 64-64 64H336.8c-11.8-25.5-29.9-47.5-52.4-64H384V320c0-17.7 14.3-32 32-32h64c17.7 0 32 14.3 32 32v32h64V64L224 64v49.1C205.2 102.2 183.3 96 160 96V64zm0 64a96 96 0 1 1 0 192 96 96 0 1 1 0-192zM133.3 352h53.3C260.3 352 320 411.7 320 485.3c0 14.7-11.9 26.7-26.7 26.7H26.7C11.9 512 0 500.1 0 485.3C0 411.7 59.7 352 133.3 352z"],"peso-sign":[384,512,[],"e222","M64 32C46.3 32 32 46.3 32 64v64c-17.7 0-32 14.3-32 32s14.3 32 32 32l0 32c-17.7 0-32 14.3-32 32s14.3 32 32 32l0 64v96c0 17.7 14.3 32 32 32s32-14.3 32-32V384h80c68.4 0 127.7-39 156.8-96H352c17.7 0 32-14.3 32-32s-14.3-32-32-32h-.7c.5-5.3 .7-10.6 .7-16s-.2-10.7-.7-16h.7c17.7 0 32-14.3 32-32s-14.3-32-32-32H332.8C303.7 71 244.4 32 176 32H64zm190.4 96H96V96h80c30.5 0 58.2 12.2 78.4 32zM96 192H286.9c.7 5.2 1.1 10.6 1.1 16s-.4 10.8-1.1 16H96V192zm158.4 96c-20.2 19.8-47.9 32-78.4 32H96V288H254.4z"],"building-shield":[576,512,[],"e4d8","M0 48C0 21.5 21.5 0 48 0H336c26.5 0 48 21.5 48 48V207l-42.4 17H304 272c-8.8 0-16 7.2-16 16v32 24.2V304c0 .9 .1 1.7 .2 2.6c2.3 58.1 24.1 144.8 98.7 201.5c-5.8 2.5-12.2 3.9-18.9 3.9H240V432c0-26.5-21.5-48-48-48s-48 21.5-48 48v80H48c-26.5 0-48-21.5-48-48V48zM80 224c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H80zm80 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H176c-8.8 0-16 7.2-16 16zM64 112v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H80c-8.8 0-16 7.2-16 16zM176 96c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H176zm80 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16zM423.1 225.7c5.7-2.3 12.1-2.3 17.8 0l120 48C570 277.4 576 286.2 576 296c0 63.3-25.9 168.8-134.8 214.2c-5.9 2.5-12.6 2.5-18.5 0C313.9 464.8 288 359.3 288 296c0-9.8 6-18.6 15.1-22.3l120-48zM527.4 312L432 273.8V461.7c68.2-33 91.5-99 95.4-149.7z"],baby:[448,512,[],"f77c","M152 88a72 72 0 1 1 144 0A72 72 0 1 1 152 88zM39.7 144.5c13-17.9 38-21.8 55.9-8.8L131.8 162c26.8 19.5 59.1 30 92.2 30s65.4-10.5 92.2-30l36.2-26.4c17.9-13 42.9-9 55.9 8.8s9 42.9-8.8 55.9l-36.2 26.4c-13.6 9.9-28.1 18.2-43.3 25V288H128V251.7c-15.2-6.7-29.7-15.1-43.3-25L48.5 200.3c-17.9-13-21.8-38-8.8-55.9zm89.8 184.8l60.6 53-26 37.2 24.3 24.3c15.6 15.6 15.6 40.9 0 56.6s-40.9 15.6-56.6 0l-48-48C70 438.6 68.1 417 79.2 401.1l50.2-71.8zm128.5 53l60.6-53 50.2 71.8c11.1 15.9 9.2 37.5-4.5 51.2l-48 48c-15.6 15.6-40.9 15.6-56.6 0s-15.6-40.9 0-56.6L284 419.4l-26-37.2z"],"users-line":[640,512,[],"e592","M211.2 96a64 64 0 1 0 -128 0 64 64 0 1 0 128 0zM32 256c0 17.7 14.3 32 32 32h85.6c10.1-39.4 38.6-71.5 75.8-86.6c-9.7-6-21.2-9.4-33.4-9.4H96c-35.3 0-64 28.7-64 64zm461.6 32H576c17.7 0 32-14.3 32-32c0-35.3-28.7-64-64-64H448c-11.7 0-22.7 3.1-32.1 8.6c38.1 14.8 67.4 47.3 77.7 87.4zM391.2 226.4c-6.9-1.6-14.2-2.4-21.6-2.4h-96c-8.5 0-16.7 1.1-24.5 3.1c-30.8 8.1-55.6 31.1-66.1 60.9c-3.5 10-5.5 20.8-5.5 32c0 17.7 14.3 32 32 32h224c17.7 0 32-14.3 32-32c0-11.2-1.9-22-5.5-32c-10.8-30.7-36.8-54.2-68.9-61.6zM563.2 96a64 64 0 1 0 -128 0 64 64 0 1 0 128 0zM321.6 192a80 80 0 1 0 0-160 80 80 0 1 0 0 160zM32 416c-17.7 0-32 14.3-32 32s14.3 32 32 32H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H32z"],"quote-left":[448,512,[8220,"quote-left-alt"],"f10d","M0 216C0 149.7 53.7 96 120 96h8c17.7 0 32 14.3 32 32s-14.3 32-32 32h-8c-30.9 0-56 25.1-56 56v8h64c35.3 0 64 28.7 64 64v64c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V320 288 216zm256 0c0-66.3 53.7-120 120-120h8c17.7 0 32 14.3 32 32s-14.3 32-32 32h-8c-30.9 0-56 25.1-56 56v8h64c35.3 0 64 28.7 64 64v64c0 35.3-28.7 64-64 64H320c-35.3 0-64-28.7-64-64V320 288 216z"],tractor:[640,512,[128668],"f722","M96 64c0-35.3 28.7-64 64-64H266.3c26.2 0 49.7 15.9 59.4 40.2L373.7 160H480V126.2c0-24.8 5.8-49.3 16.9-71.6l2.5-5c7.9-15.8 27.1-22.2 42.9-14.3s22.2 27.1 14.3 42.9l-2.5 5c-6.7 13.3-10.1 28-10.1 42.9V160h56c22.1 0 40 17.9 40 40v45.4c0 16.5-8.5 31.9-22.6 40.7l-43.3 27.1c-14.2-5.9-29.8-9.2-46.1-9.2c-39.3 0-74.1 18.9-96 48H352c0 17.7-14.3 32-32 32h-8.2c-1.7 4.8-3.7 9.5-5.8 14.1l5.8 5.8c12.5 12.5 12.5 32.8 0 45.3l-22.6 22.6c-12.5 12.5-32.8 12.5-45.3 0l-5.8-5.8c-4.6 2.2-9.3 4.1-14.1 5.8V480c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32v-8.2c-4.8-1.7-9.5-3.7-14.1-5.8l-5.8 5.8c-12.5 12.5-32.8 12.5-45.3 0L40.2 449.1c-12.5-12.5-12.5-32.8 0-45.3l5.8-5.8c-2.2-4.6-4.1-9.3-5.8-14.1H32c-17.7 0-32-14.3-32-32V320c0-17.7 14.3-32 32-32h8.2c1.7-4.8 3.7-9.5 5.8-14.1l-5.8-5.8c-12.5-12.5-12.5-32.8 0-45.3l22.6-22.6c9-9 21.9-11.5 33.1-7.6V192 160 64zm170.3 0H160v96h32H304.7L266.3 64zM176 256a80 80 0 1 0 0 160 80 80 0 1 0 0-160zM528 448a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm0 64c-48.6 0-88-39.4-88-88c0-29.8 14.8-56.1 37.4-72c14.3-10.1 31.8-16 50.6-16c2.7 0 5.3 .1 7.9 .3c44.9 4 80.1 41.7 80.1 87.7c0 48.6-39.4 88-88 88z"],"trash-arrow-up":[448,512,["trash-restore"],"f829","M163.8 0H284.2c12.1 0 23.2 6.8 28.6 17.7L320 32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 96 0 81.7 0 64S14.3 32 32 32h96l7.2-14.3C140.6 6.8 151.7 0 163.8 0zM32 128H416L394.8 467c-1.6 25.3-22.6 45-47.9 45H101.1c-25.3 0-46.3-19.7-47.9-45L32 128zm192 64c-6.4 0-12.5 2.5-17 7l-80 80c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l39-39V408c0 13.3 10.7 24 24 24s24-10.7 24-24V273.9l39 39c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-80-80c-4.5-4.5-10.6-7-17-7z"],"arrow-down-up-lock":[640,512,[],"e4b0","M150.6 502.6l96-96c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L160 402.7V288H416V272c0-17.2 3.9-33.5 10.8-48H352V109.3l41.4 41.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-96-96c-6-6-14.1-9.4-22.6-9.4s-16.6 3.4-22.6 9.4l-96 96c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L288 109.3V224l-128 0H96l-64 0c-17.7 0-32 14.3-32 32s14.3 32 32 32H96V402.7L54.6 361.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l96 96c12.5 12.5 32.8 12.5 45.3 0zM160 192V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V192h64zM288 320V448c0 17.7 14.3 32 32 32s32-14.3 32-32V320H288zm240-80c17.7 0 32 14.3 32 32v48H496V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z"],"lines-leaning":[384,512,[],"e51e","M190.4 74.1c5.6-16.8-3.5-34.9-20.2-40.5s-34.9 3.5-40.5 20.2l-128 384c-5.6 16.8 3.5 34.9 20.2 40.5s34.9-3.5 40.5-20.2l128-384zm70.9-41.7c-17.4-2.9-33.9 8.9-36.8 26.3l-64 384c-2.9 17.4 8.9 33.9 26.3 36.8s33.9-8.9 36.8-26.3l64-384c2.9-17.4-8.9-33.9-26.3-36.8zM352 32c-17.7 0-32 14.3-32 32V448c0 17.7 14.3 32 32 32s32-14.3 32-32V64c0-17.7-14.3-32-32-32z"],"ruler-combined":[512,512,[],"f546","M.2 468.9C2.7 493.1 23.1 512 48 512l96 0 320 0c26.5 0 48-21.5 48-48l0-96c0-26.5-21.5-48-48-48l-48 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-80 0c-8.8 0-16-7.2-16-16s7.2-16 16-16l80 0 0-64-80 0c-8.8 0-16-7.2-16-16s7.2-16 16-16l80 0 0-64-80 0c-8.8 0-16-7.2-16-16s7.2-16 16-16l80 0 0-48c0-26.5-21.5-48-48-48L48 0C21.5 0 0 21.5 0 48L0 368l0 96c0 1.7 .1 3.3 .2 4.9z"],copyright:[512,512,[169],"f1f9","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM199.4 312.6c31.2 31.2 81.9 31.2 113.1 0c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9c-50 50-131 50-181 0s-50-131 0-181s131-50 181 0c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0c-31.2-31.2-81.9-31.2-113.1 0s-31.2 81.9 0 113.1z"],equals:[448,512,[62764],"3d","M48 128c-17.7 0-32 14.3-32 32s14.3 32 32 32H400c17.7 0 32-14.3 32-32s-14.3-32-32-32H48zm0 192c-17.7 0-32 14.3-32 32s14.3 32 32 32H400c17.7 0 32-14.3 32-32s-14.3-32-32-32H48z"],blender:[512,512,[],"f517","M0 64C0 28.7 28.7 0 64 0h64 32H470.1c21.1 0 36.4 20.1 30.9 40.4L494.5 64H336c-8.8 0-16 7.2-16 16s7.2 16 16 16H485.8l-17.5 64H336c-8.8 0-16 7.2-16 16s7.2 16 16 16H459.6l-17.5 64H336c-8.8 0-16 7.2-16 16s7.2 16 16 16h97.5L416 352H160l-8.7-96H64c-35.3 0-64-28.7-64-64V64zM145.5 192L133.8 64H64V192h81.5zM144 384H432c26.5 0 48 21.5 48 48v32c0 26.5-21.5 48-48 48H144c-26.5 0-48-21.5-48-48V432c0-26.5 21.5-48 48-48zm144 96a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],teeth:[576,512,[],"f62e","M0 128C0 75 43 32 96 32H480c53 0 96 43 96 96V384c0 53-43 96-96 96H96c-53 0-96-43-96-96V128zm176 48v56c0 13.3 10.7 24 24 24h48c13.3 0 24-10.7 24-24V176c0-26.5-21.5-48-48-48s-48 21.5-48 48zm176-48c-26.5 0-48 21.5-48 48v56c0 13.3 10.7 24 24 24h48c13.3 0 24-10.7 24-24V176c0-26.5-21.5-48-48-48zM48 208v24c0 13.3 10.7 24 24 24h48c13.3 0 24-10.7 24-24V208c0-26.5-21.5-48-48-48s-48 21.5-48 48zM96 384c26.5 0 48-21.5 48-48V312c0-13.3-10.7-24-24-24H72c-13.3 0-24 10.7-24 24v24c0 26.5 21.5 48 48 48zm80-48c0 26.5 21.5 48 48 48s48-21.5 48-48V312c0-13.3-10.7-24-24-24H200c-13.3 0-24 10.7-24 24v24zm176 48c26.5 0 48-21.5 48-48V312c0-13.3-10.7-24-24-24H328c-13.3 0-24 10.7-24 24v24c0 26.5 21.5 48 48 48zm80-176v24c0 13.3 10.7 24 24 24h48c13.3 0 24-10.7 24-24V208c0-26.5-21.5-48-48-48s-48 21.5-48 48zm48 176c26.5 0 48-21.5 48-48V312c0-13.3-10.7-24-24-24H456c-13.3 0-24 10.7-24 24v24c0 26.5 21.5 48 48 48z"],"shekel-sign":[448,512,[8362,"ils","shekel","sheqel","sheqel-sign"],"f20b","M32 32C14.3 32 0 46.3 0 64V448c0 17.7 14.3 32 32 32s32-14.3 32-32V96H192c35.3 0 64 28.7 64 64V320c0 17.7 14.3 32 32 32s32-14.3 32-32V160c0-70.7-57.3-128-128-128H32zM320 480c70.7 0 128-57.3 128-128V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V352c0 35.3-28.7 64-64 64H192V192c0-17.7-14.3-32-32-32s-32 14.3-32 32V448c0 17.7 14.3 32 32 32H320z"],map:[576,512,[128506,62072],"f279","M384 476.1L192 421.2V35.9L384 90.8V476.1zm32-1.2V88.4L543.1 37.5c15.8-6.3 32.9 5.3 32.9 22.3V394.6c0 9.8-6 18.6-15.1 22.3L416 474.8zM15.1 95.1L160 37.2V423.6L32.9 474.5C17.1 480.8 0 469.2 0 452.2V117.4c0-9.8 6-18.6 15.1-22.3z"],rocket:[512,512,[],"f135","M156.6 384.9L125.7 354c-8.5-8.5-11.5-20.8-7.7-32.2c3-8.9 7-20.5 11.8-33.8L24 288c-8.6 0-16.6-4.6-20.9-12.1s-4.2-16.7 .2-24.1l52.5-88.5c13-21.9 36.5-35.3 61.9-35.3l82.3 0c2.4-4 4.8-7.7 7.2-11.3C289.1-4.1 411.1-8.1 483.9 5.3c11.6 2.1 20.6 11.2 22.8 22.8c13.4 72.9 9.3 194.8-111.4 276.7c-3.5 2.4-7.3 4.8-11.3 7.2v82.3c0 25.4-13.4 49-35.3 61.9l-88.5 52.5c-7.4 4.4-16.6 4.5-24.1 .2s-12.1-12.2-12.1-20.9V380.8c-14.1 4.9-26.4 8.9-35.7 11.9c-11.2 3.6-23.4 .5-31.8-7.8zM384 168a40 40 0 1 0 0-80 40 40 0 1 0 0 80z"],"photo-film":[640,512,["photo-video"],"f87c","M256 0H576c35.3 0 64 28.7 64 64V288c0 35.3-28.7 64-64 64H256c-35.3 0-64-28.7-64-64V64c0-35.3 28.7-64 64-64zM476 106.7C471.5 100 464 96 456 96s-15.5 4-20 10.7l-56 84L362.7 169c-4.6-5.7-11.5-9-18.7-9s-14.2 3.3-18.7 9l-64 80c-5.8 7.2-6.9 17.1-2.9 25.4s12.4 13.6 21.6 13.6h80 48H552c8.9 0 17-4.9 21.2-12.7s3.7-17.3-1.2-24.6l-96-144zM336 96a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM64 128h96V384v32c0 17.7 14.3 32 32 32H320c17.7 0 32-14.3 32-32V384H512v64c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V192c0-35.3 28.7-64 64-64zm8 64c-8.8 0-16 7.2-16 16v16c0 8.8 7.2 16 16 16H88c8.8 0 16-7.2 16-16V208c0-8.8-7.2-16-16-16H72zm0 104c-8.8 0-16 7.2-16 16v16c0 8.8 7.2 16 16 16H88c8.8 0 16-7.2 16-16V312c0-8.8-7.2-16-16-16H72zm0 104c-8.8 0-16 7.2-16 16v16c0 8.8 7.2 16 16 16H88c8.8 0 16-7.2 16-16V416c0-8.8-7.2-16-16-16H72zm336 16v16c0 8.8 7.2 16 16 16h16c8.8 0 16-7.2 16-16V416c0-8.8-7.2-16-16-16H424c-8.8 0-16 7.2-16 16z"],"folder-minus":[512,512,[],"f65d","M448 480H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H192c20.1 0 39.1 9.5 51.2 25.6l19.2 25.6c6 8.1 15.5 12.8 25.6 12.8H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64zM184 272c-13.3 0-24 10.7-24 24s10.7 24 24 24H328c13.3 0 24-10.7 24-24s-10.7-24-24-24H184z"],store:[576,512,[],"f54e","M547.6 103.8L490.3 13.1C485.2 5 476.1 0 466.4 0H109.6C99.9 0 90.8 5 85.7 13.1L28.3 103.8c-29.6 46.8-3.4 111.9 51.9 119.4c4 .5 8.1 .8 12.1 .8c26.1 0 49.3-11.4 65.2-29c15.9 17.6 39.1 29 65.2 29c26.1 0 49.3-11.4 65.2-29c15.9 17.6 39.1 29 65.2 29c26.2 0 49.3-11.4 65.2-29c16 17.6 39.1 29 65.2 29c4.1 0 8.1-.3 12.1-.8c55.5-7.4 81.8-72.5 52.1-119.4zM499.7 254.9l-.1 0c-5.3 .7-10.7 1.1-16.2 1.1c-12.4 0-24.3-1.9-35.4-5.3V384H128V250.6c-11.2 3.5-23.2 5.4-35.6 5.4c-5.5 0-11-.4-16.3-1.1l-.1 0c-4.1-.6-8.1-1.3-12-2.3V384v64c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V384 252.6c-4 1-8 1.8-12.3 2.3z"],"arrow-trend-up":[576,512,[],"e098","M384 160c-17.7 0-32-14.3-32-32s14.3-32 32-32H544c17.7 0 32 14.3 32 32V288c0 17.7-14.3 32-32 32s-32-14.3-32-32V205.3L342.6 374.6c-12.5 12.5-32.8 12.5-45.3 0L192 269.3 54.6 406.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l160-160c12.5-12.5 32.8-12.5 45.3 0L320 306.7 466.7 160H384z"],"plug-circle-minus":[576,512,[],"e55e","M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8c12.3-2.5 24.1-6.4 35.1-11.5c-2.1-10.8-3.1-21.9-3.1-33.3c0-80.3 53.8-148 127.3-169.2c.5-2.2 .7-4.5 .7-6.8c0-17.7-14.3-32-32-32H32zM576 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-64 0c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16s7.2-16 16-16H496c8.8 0 16 7.2 16 16z"],"sign-hanging":[512,512,["sign"],"f4d9","M96 0c17.7 0 32 14.3 32 32V64l352 0c17.7 0 32 14.3 32 32s-14.3 32-32 32l-352 0V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V128H32C14.3 128 0 113.7 0 96S14.3 64 32 64H64V32C64 14.3 78.3 0 96 0zm96 160H448c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32V192c0-17.7 14.3-32 32-32z"],"bezier-curve":[640,512,[],"f55b","M296 136V88h48v48H296zM288 32c-26.5 0-48 21.5-48 48v4H121.6C111.2 62.7 89.3 48 64 48C28.7 48 0 76.7 0 112s28.7 64 64 64c25.3 0 47.2-14.7 57.6-36h66.9c-58.9 39.6-98.9 105-104 180H80c-26.5 0-48 21.5-48 48v64c0 26.5 21.5 48 48 48h64c26.5 0 48-21.5 48-48V368c0-26.5-21.5-48-48-48h-3.3c5.9-67 48.5-123.4 107.5-149.1c8.6 12.7 23.2 21.1 39.8 21.1h64c16.6 0 31.1-8.4 39.8-21.1c59 25.7 101.6 82.1 107.5 149.1H496c-26.5 0-48 21.5-48 48v64c0 26.5 21.5 48 48 48h64c26.5 0 48-21.5 48-48V368c0-26.5-21.5-48-48-48h-4.5c-5-75-45.1-140.4-104-180h66.9c10.4 21.3 32.3 36 57.6 36c35.3 0 64-28.7 64-64s-28.7-64-64-64c-25.3 0-47.2 14.7-57.6 36H400V80c0-26.5-21.5-48-48-48H288zM88 376h48v48H88V376zm416 48V376h48v48H504z"],"bell-slash":[640,512,[128277,61943],"f1f6","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-90.2-70.7c.2-.4 .4-.9 .6-1.3c5.2-11.5 3.1-25-5.3-34.4l-7.4-8.3C497.3 319.2 480 273.9 480 226.8V208c0-77.4-55-142-128-156.8V32c0-17.7-14.3-32-32-32s-32 14.3-32 32V51.2c-42.6 8.6-79 34.2-102 69.3L38.8 5.1zM406.2 416L160 222.1v4.8c0 47-17.3 92.4-48.5 127.6l-7.4 8.3c-8.4 9.4-10.4 22.9-5.3 34.4S115.4 416 128 416H406.2zm-40.9 77.3c12-12 18.7-28.3 18.7-45.3H320 256c0 17 6.7 33.3 18.7 45.3s28.3 18.7 45.3 18.7s33.3-6.7 45.3-18.7z"],tablet:[448,512,["tablet-android"],"f3fb","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zM176 432h96c8.8 0 16 7.2 16 16s-7.2 16-16 16H176c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],"school-flag":[576,512,[],"e56e","M288 0H400c8.8 0 16 7.2 16 16V80c0 8.8-7.2 16-16 16H320.7l89.6 64H512c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H336V400c0-26.5-21.5-48-48-48s-48 21.5-48 48V512H64c-35.3 0-64-28.7-64-64V224c0-35.3 28.7-64 64-64H165.7L256 95.5V32c0-17.7 14.3-32 32-32zm48 240a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM80 224c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H80zm368 16v64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H464c-8.8 0-16 7.2-16 16zM80 352c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H80zm384 0c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H464z"],fill:[512,512,[],"f575","M86.6 9.4C74.1-3.1 53.9-3.1 41.4 9.4s-12.5 32.8 0 45.3L122.7 136 30.6 228.1c-37.5 37.5-37.5 98.3 0 135.8L148.1 481.4c37.5 37.5 98.3 37.5 135.8 0L474.3 290.9c28.1-28.1 28.1-73.7 0-101.8L322.9 37.7c-28.1-28.1-73.7-28.1-101.8 0L168 90.7 86.6 9.4zM168 181.3l49.4 49.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L213.3 136l53.1-53.1c3.1-3.1 8.2-3.1 11.3 0L429.1 234.3c3.1 3.1 3.1 8.2 0 11.3L386.7 288H67.5c1.4-5.4 4.2-10.4 8.4-14.6L168 181.3z"],"angle-up":[448,512,[8963],"f106","M201.4 137.4c12.5-12.5 32.8-12.5 45.3 0l160 160c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L224 205.3 86.6 342.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l160-160z"],"drumstick-bite":[512,512,[],"f6d7","M160 265.2c0 8.5-3.4 16.6-9.4 22.6l-26.8 26.8c-12.3 12.3-32.5 11.4-49.4 7.2C69.8 320.6 65 320 60 320c-33.1 0-60 26.9-60 60s26.9 60 60 60c6.3 0 12 5.7 12 12c0 33.1 26.9 60 60 60s60-26.9 60-60c0-5-.6-9.8-1.8-14.5c-4.2-16.9-5.2-37.1 7.2-49.4l26.8-26.8c6-6 14.1-9.4 22.6-9.4H336c6.3 0 12.4-.3 18.5-1c11.9-1.2 16.4-15.5 10.8-26c-8.5-15.8-13.3-33.8-13.3-53c0-61.9 50.1-112 112-112c8 0 15.7 .8 23.2 2.4c11.7 2.5 24.1-5.9 22-17.6C494.5 62.5 422.5 0 336 0C238.8 0 160 78.8 160 176v89.2z"],"holly-berry":[512,512,[],"f7aa","M256 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-80 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM276.8 383.8c1 .1 2.1 .2 3.2 .2c39.8 0 72 32.2 72 72v22.7c0 16.4 16 27.9 31.6 22.8l12.8-4.3c18-6 37.3-6.5 55.6-1.5l19.4 5.3c17.9 4.9 34.4-11.6 29.5-29.5L495.6 452c-5-18.3-4.4-37.6 1.5-55.6l4.3-12.8c5.2-15.5-6.4-31.6-22.8-31.6c-34.6 0-62.7-28.1-62.7-62.7v-32c0-16.4-16-27.9-31.6-22.8l-12.8 4.3c-18 6-37.3 6.5-55.6 1.5l-29.6-8.1c-2.9-.8-5.9-1-8.7-.7c4.2 9.7 5.8 20.8 3.7 32.3L275 298.7c-1.5 8.4-1.4 17 .5 25.3l5.3 23.9c2.8 12.7 1.1 25.2-4 35.9zM127.6 234.5c-15.5-5.2-31.6 6.4-31.6 22.8v32C96 323.9 67.9 352 33.3 352c-16.4 0-27.9 16-22.8 31.6l4.3 12.8c6 18 6.5 37.3 1.5 55.6l-5.3 19.4C6.2 489.4 22.6 505.8 40.5 501L60 495.6c18.3-5 37.6-4.5 55.6 1.5l12.8 4.3c15.5 5.2 31.6-6.4 31.6-22.8v-32c0-34.6 28.1-62.7 62.7-62.7c16.4 0 27.9-16 22.8-31.6l-4.3-12.8c-6-18-6.5-37.3-1.5-55.6l5.3-19.4c4.9-17.9-11.6-34.4-29.5-29.5L196 240.4c-18.3 5-37.6 4.4-55.6-1.5l-12.8-4.3zM384 144a48 48 0 1 0 -96 0 48 48 0 1 0 96 0z"],"chevron-left":[320,512,[9001],"f053","M9.4 233.4c-12.5 12.5-12.5 32.8 0 45.3l192 192c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L77.3 256 246.6 86.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-192 192z"],bacteria:[640,512,[],"e059","M304.9 .7c-9.6-2.7-19.5 2.8-22.3 12.4l-4.3 15.2c-8.3-.6-16.8 0-25.2 1.9c-7.3 1.7-14.3 3.5-21.1 5.5l-5.5-12.7c-3.9-9.1-14.5-13.4-23.6-9.5s-13.4 14.5-9.5 23.6l4.4 10.4c-16.6 6.7-31.7 14.4-45.4 22.8L147 62c-5.5-8.3-16.7-10.5-25-5s-10.5 16.7-5 25l6 9c-13.7 11-25.5 22.8-35.8 34.9l-10-8c-7.8-6.2-19.1-5-25.3 2.8s-5 19.1 2.8 25.3L65.9 155c-1.8 2.8-3.5 5.7-5.1 8.5c-6.6 11.4-11.8 22.6-16 33l-8-3.2c-9.2-3.7-19.7 .8-23.4 10s.8 19.7 10 23.4l10.4 4.2c-.2 .8-.4 1.5-.5 2.3c-2.2 9.3-3.4 17.3-4.1 23.4c-.4 3.1-.6 5.7-.8 7.8c-.1 1.1-.1 2-.2 2.8l-.1 1.1 0 .5 0 .2 0 .1c0 0 0 .1 29.1 1l-.1 0L28 269.3c-.1 3.1 0 6.1 .2 9.1l-15.2 4.3C3.5 285.4-2 295.4 .7 304.9s12.7 15.1 22.3 12.4l15.6-4.5c7.6 13.6 18.9 25 32.6 32.6L66.7 361c-2.7 9.6 2.8 19.5 12.4 22.3s19.5-2.8 22.3-12.4l4.3-15.2c1.2 .1 2.4 .2 3.6 .2c15.6 .5 30.3-3.3 43-10.2l9 9c7 7 18.4 7 25.5 0s7-18.4 0-25.5l-7.2-7.2c9.3-12.6 15.2-27.8 16.3-44.5l7.1 3c9.1 3.9 19.7-.3 23.6-9.5s-.3-19.7-9.5-23.6l-8.6-3.7c6.4-9.9 17.3-22.4 36.9-33.3l1.3 4.4c2.7 9.6 12.7 15.1 22.3 12.4s15.1-12.7 12.4-22.3l-2.3-8.1c3.8-1.1 7.7-2.1 11.9-3.1c11.6-2.7 22.1-7.7 31.1-14.4l7.2 7.2c7 7 18.4 7 25.5 0s7-18.4 0-25.5l-9-9c7.6-13.9 11.3-30.1 10.1-46.6l15.2-4.3c9.6-2.7 15.1-12.7 12.4-22.3S370.6 64 361 66.7l-15.6 4.5c-7.7-13.9-19.1-25.1-32.6-32.6l4.5-15.6c2.7-9.6-2.8-19.5-12.4-22.3zM112 272l-48-1.5 0 0c11.7 .4 27.3 .9 48 1.6zm16-80a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm64-48a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zM322.7 489c-2.7 9.6 2.8 19.5 12.4 22.3s19.5-2.8 22.2-12.4l4.3-15.2c8.3 .6 16.8 0 25.2-1.9c7.3-1.7 14.3-3.5 21.1-5.5l5.5 12.7c3.9 9.1 14.5 13.4 23.6 9.5s13.4-14.5 9.5-23.6l-4.4-10.4c16.6-6.7 31.7-14.4 45.4-22.8L493 450c5.5 8.3 16.7 10.5 25 5s10.5-16.7 5-25l-6-9c13.7-11 25.5-22.8 35.8-34.9l10 8c7.8 6.2 19.1 5 25.3-2.8s5-19.1-2.8-25.3L574.1 357c1.8-2.8 3.5-5.7 5.1-8.5c6.6-11.4 11.8-22.6 16-33l8 3.2c9.2 3.7 19.7-.8 23.4-10s-.8-19.7-10-23.4l-10.4-4.2c.2-.8 .4-1.5 .5-2.3c2.2-9.3 3.4-17.3 4.1-23.4c.4-3.1 .6-5.7 .8-7.8c.1-1.1 .1-2 .2-2.8l.1-1.1 0-.5 0-.2 0-.1c0 0 0-.1-29.1-1l.1 0 29.1 .9c.1-3.1 0-6.1-.2-9.1l15.2-4.3c9.6-2.7 15.1-12.7 12.4-22.3s-12.7-15.1-22.3-12.4l-15.6 4.5c-7.6-13.6-18.9-25-32.6-32.6l4.5-15.6c2.7-9.6-2.8-19.5-12.4-22.3s-19.5 2.8-22.3 12.4l-4.3 15.2c-1.2-.1-2.4-.2-3.6-.2c-15.6-.5-30.3 3.3-43 10.2l-9-9c-7-7-18.4-7-25.5 0s-7 18.4 0 25.5l7.2 7.2c-9.3 12.6-15.2 27.8-16.3 44.5l-7.1-3c-9.1-3.9-19.7 .3-23.6 9.5s.3 19.7 9.5 23.6l8.6 3.7c-6.4 9.9-17.3 22.4-36.9 33.3l-1.3-4.4c-2.7-9.6-12.7-15.1-22.3-12.4s-15.1 12.7-12.4 22.3l2.3 8.1c-3.8 1.1-7.7 2.1-11.9 3.1c-11.6 2.7-22.1 7.7-31.1 14.4l-7.2-7.2c-7-7-18.4-7-25.5 0s-7 18.4 0 25.5l9 9c-7.6 13.9-11.3 30.1-10.1 46.6l-15.2 4.3c-9.6 2.7-15.1 12.7-12.4 22.2s12.7 15.1 22.3 12.4l15.6-4.5c7.7 13.9 19.1 25.1 32.6 32.6L322.7 489zM576 241.5l0 0c-11.7-.4-27.3-.9-48-1.6l48 1.5zM448 384a32 32 0 1 1 -64 0 32 32 0 1 1 64 0z"],"hand-lizard":[512,512,[],"f258","M0 112C0 85.5 21.5 64 48 64H160h80 46.5c36.8 0 71.2 18 92.1 48.2l113.5 164c13 18.7 19.9 41 19.9 63.8v12 16 48c0 17.7-14.3 32-32 32H384c-17.7 0-32-14.3-32-32V402.2L273.9 352H240 160 112c-26.5 0-48-21.5-48-48s21.5-48 48-48h48 80c26.5 0 48-21.5 48-48s-21.5-48-48-48H160 48c-26.5 0-48-21.5-48-48z"],notdef:[384,512,[],"e1fe","M64 390.3L153.5 256 64 121.7V390.3zM102.5 448H281.5L192 313.7 102.5 448zm128-192L320 390.3V121.7L230.5 256zM281.5 64H102.5L192 198.3 281.5 64zM0 48C0 21.5 21.5 0 48 0H336c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V48z"],disease:[512,512,[],"f7fa","M236.4 61.4L227 75.5c-21.3 32-59.4 48.5-97.3 42.1l-59.6-9.9C33.4 101.6 0 129.9 .1 167.1c0 15.9 6.4 31.2 17.6 42.5l29.2 29.2c11 11 17.2 25.9 17.2 41.5c0 15.8-6.4 30.9-17.7 42L33.3 335.1C22.2 345.9 16 360.7 16 376.2c0 36.8 34.1 64.2 70.1 56.2l62.3-13.8c7.7-1.7 15.7-2.6 23.6-2.6h10c27.2 0 53.7 9.3 75 26.3L287.8 467c10.5 8.4 23.6 13 37 13c32.7 0 59.3-26.5 59.3-59.3l0-25.2c0-34.9 21.4-66.2 53.9-78.8l36.9-14.3c22.4-8.7 37.2-30.3 37.2-54.3c0-28.1-20.1-52.3-47.8-57.3l-28-5.1c-36.5-6.7-65.4-34.5-73.6-70.7l-7.1-31.5C348.9 53.4 322.1 32 291.3 32c-22 0-42.6 11-54.9 29.4zM160 192a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zm0 80a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"briefcase-medical":[512,512,[],"f469","M184 48H328c4.4 0 8 3.6 8 8V96H176V56c0-4.4 3.6-8 8-8zm-56 8V96H64C28.7 96 0 124.7 0 160V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H384V56c0-30.9-25.1-56-56-56H184c-30.9 0-56 25.1-56 56zm96 152c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H288v48c0 8.8-7.2 16-16 16H240c-8.8 0-16-7.2-16-16V320H176c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h48V208z"],genderless:[384,512,[],"f22d","M192 144a112 112 0 1 1 0 224 112 112 0 1 1 0-224zm0 288a176 176 0 1 0 0-352 176 176 0 1 0 0 352z"],"chevron-right":[320,512,[9002],"f054","M310.6 233.4c12.5 12.5 12.5 32.8 0 45.3l-192 192c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L242.7 256 73.4 86.6c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0l192 192z"],retweet:[576,512,[],"f079","M272 416c17.7 0 32-14.3 32-32s-14.3-32-32-32H160c-17.7 0-32-14.3-32-32V192h32c12.9 0 24.6-7.8 29.6-19.8s2.2-25.7-6.9-34.9l-64-64c-12.5-12.5-32.8-12.5-45.3 0l-64 64c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8l32 0 0 128c0 53 43 96 96 96H272zM304 96c-17.7 0-32 14.3-32 32s14.3 32 32 32l112 0c17.7 0 32 14.3 32 32l0 128H416c-12.9 0-24.6 7.8-29.6 19.8s-2.2 25.7 6.9 34.9l64 64c12.5 12.5 32.8 12.5 45.3 0l64-64c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8l-32 0V192c0-53-43-96-96-96L304 96z"],"car-rear":[512,512,["car-alt"],"f5de","M165.4 96H346.6c13.6 0 25.7 8.6 30.2 21.4L402.9 192H109.1l26.1-74.6c4.5-12.8 16.6-21.4 30.2-21.4zm-90.6 .3L39.6 196.8C16.4 206.4 0 229.3 0 256v80c0 23.7 12.9 44.4 32 55.4V448c0 17.7 14.3 32 32 32H96c17.7 0 32-14.3 32-32V400H384v48c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V391.4c19.1-11.1 32-31.7 32-55.4V256c0-26.7-16.4-49.6-39.6-59.2L437.2 96.3C423.7 57.8 387.4 32 346.6 32H165.4c-40.8 0-77.1 25.8-90.6 64.3zM208 272h96c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H208c-8.8 0-16-7.2-16-16V288c0-8.8 7.2-16 16-16zM48 280c0-13.3 10.7-24 24-24h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H72c-13.3 0-24-10.7-24-24zm360-24h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H408c-13.3 0-24-10.7-24-24s10.7-24 24-24z"],"pump-soap":[448,512,[],"e06b","M128 32v96H256V96h60.1c4.2 0 8.3 1.7 11.3 4.7l33.9 33.9c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L372.7 55.4c-15-15-35.4-23.4-56.6-23.4H256c0-17.7-14.3-32-32-32H160c-17.7 0-32 14.3-32 32zM117.4 160c-33.3 0-61 25.5-63.8 58.7L35 442.7C31.9 480 61.3 512 98.8 512H285.2c37.4 0 66.9-32 63.8-69.3l-18.7-224c-2.8-33.2-30.5-58.7-63.8-58.7H117.4zM256 360c0 35.3-28.7 56-64 56s-64-20.7-64-56c0-32.5 37-80.9 50.9-97.9c3.2-3.9 8.1-6.1 13.1-6.1s9.9 2.2 13.1 6.1C219 279.1 256 327.5 256 360z"],"video-slash":[640,512,[],"f4e2","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-86.4-67.7 13.8 9.2c9.8 6.5 22.4 7.2 32.9 1.6s16.9-16.4 16.9-28.2V128c0-11.8-6.5-22.6-16.9-28.2s-23-5-32.9 1.6l-96 64L448 174.9V192 320v5.8l-32-25.1V128c0-35.3-28.7-64-64-64H113.9L38.8 5.1zM407 416.7L32.3 121.5c-.2 2.1-.3 4.3-.3 6.5V384c0 35.3 28.7 64 64 64H352c23.4 0 43.9-12.6 55-31.3z"],"battery-quarter":[576,512,["battery-2"],"f243","M464 160c8.8 0 16 7.2 16 16V336c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16H464zM80 96C35.8 96 0 131.8 0 176V336c0 44.2 35.8 80 80 80H464c44.2 0 80-35.8 80-80V320c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32V176c0-44.2-35.8-80-80-80H80zm112 96H96V320h96V192z"],radio:[512,512,[128251],"f8d7","M494.8 47c12.7-3.7 20-17.1 16.3-29.8S494-2.8 481.2 1L51.7 126.9c-9.4 2.7-17.9 7.3-25.1 13.2C10.5 151.7 0 170.6 0 192v4V304 448c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V192c0-35.3-28.7-64-64-64H218.5L494.8 47zM368 240a80 80 0 1 1 0 160 80 80 0 1 1 0-160zM80 256c0-8.8 7.2-16 16-16h96c8.8 0 16 7.2 16 16s-7.2 16-16 16H96c-8.8 0-16-7.2-16-16zM64 320c0-8.8 7.2-16 16-16H208c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16zm16 64c0-8.8 7.2-16 16-16h96c8.8 0 16 7.2 16 16s-7.2 16-16 16H96c-8.8 0-16-7.2-16-16z"],"baby-carriage":[512,512,["carriage-baby"],"f77d","M256 192H.1C2.7 117.9 41.3 52.9 99 14.1c13.3-8.9 30.8-4.3 39.9 8.8L256 192zm128-32c0-35.3 28.7-64 64-64h32c17.7 0 32 14.3 32 32s-14.3 32-32 32l-32 0v64c0 25.2-5.8 50.2-17 73.5s-27.8 44.5-48.6 62.3s-45.5 32-72.7 41.6S253.4 416 224 416s-58.5-5-85.7-14.6s-51.9-23.8-72.7-41.6s-37.3-39-48.6-62.3S0 249.2 0 224l224 0 160 0V160zM80 416a48 48 0 1 1 0 96 48 48 0 1 1 0-96zm240 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0z"],"traffic-light":[320,512,[128678],"f637","M64 0C28.7 0 0 28.7 0 64V352c0 88.4 71.6 160 160 160s160-71.6 160-160V64c0-35.3-28.7-64-64-64H64zm96 416a48 48 0 1 1 0-96 48 48 0 1 1 0 96zm48-176a48 48 0 1 1 -96 0 48 48 0 1 1 96 0zm-48-80a48 48 0 1 1 0-96 48 48 0 1 1 0 96z"],thermometer:[512,512,[],"f491","M96 382.1V293.3c0-14.9 5.9-29.1 16.4-39.6l27.3-27.3 57 57c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-57-57 41.4-41.4 57 57c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-57-57 41.4-41.4 57 57c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-57-57 45.5-45.5C355.2 10.9 381.4 0 408.8 0C465.8 0 512 46.2 512 103.2c0 27.4-10.9 53.6-30.2 73L258.3 399.6c-10.5 10.5-24.7 16.4-39.6 16.4H129.9L41 505c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l89-89z"],"vr-cardboard":[640,512,[],"f729","M576 64H64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H184.4c24.2 0 46.4-13.7 57.2-35.4l32-64c8.8-17.5 26.7-28.6 46.3-28.6s37.5 11.1 46.3 28.6l32 64c10.8 21.7 33 35.4 57.2 35.4H576c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64zM96 240a64 64 0 1 1 128 0A64 64 0 1 1 96 240zm384-64a64 64 0 1 1 0 128 64 64 0 1 1 0-128z"],"hand-middle-finger":[448,512,[128405],"f806","M232 0c-22.1 0-40 17.9-40 40V204.2c-8.5-7.6-19.7-12.2-32-12.2c-26.5 0-48 21.5-48 48v7 73c0 8.8-7.2 16-16 16s-16-7.2-16-16V264.3c-2 1.4-3.9 3-5.8 4.5L55 284.8C40.4 297 32 315 32 334V372c0 38 16.9 74 46.1 98.3l5.4 4.5c28.8 24 65 37.1 102.4 37.1H304c70.7 0 128-57.3 128-128V320 288c0-26.5-21.5-48-48-48c-12.4 0-23.6 4.7-32.1 12.3C350 227.5 329.3 208 304 208c-12.3 0-23.5 4.6-32 12.2V40c0-22.1-17.9-40-40-40z"],percent:[384,512,[62101,62785,"percentage"],"25","M374.6 118.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-320 320c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l320-320zM128 128A64 64 0 1 0 0 128a64 64 0 1 0 128 0zM384 384a64 64 0 1 0 -128 0 64 64 0 1 0 128 0z"],"truck-moving":[640,512,[],"f4df","M64 32C28.7 32 0 60.7 0 96V304v80 16c0 44.2 35.8 80 80 80c26.2 0 49.4-12.6 64-32c14.6 19.4 37.8 32 64 32c44.2 0 80-35.8 80-80c0-5.5-.6-10.8-1.6-16H416h33.6c-1 5.2-1.6 10.5-1.6 16c0 44.2 35.8 80 80 80s80-35.8 80-80c0-5.5-.6-10.8-1.6-16H608c17.7 0 32-14.3 32-32V288 272 261.7c0-9.2-3.2-18.2-9-25.3l-58.8-71.8c-10.6-13-26.5-20.5-43.3-20.5H480V96c0-35.3-28.7-64-64-64H64zM585 256H480V192h48.8c2.4 0 4.7 1.1 6.2 2.9L585 256zM528 368a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM176 400a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM80 368a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"glass-water-droplet":[384,512,[],"e4f5","M32 0C23.1 0 14.6 3.7 8.6 10.2S-.6 25.4 .1 34.3L28.9 437.7c3 41.9 37.8 74.3 79.8 74.3H275.3c42 0 76.8-32.4 79.8-74.3L383.9 34.3c.6-8.9-2.4-17.6-8.5-24.1S360.9 0 352 0H32zM83 297.5L66.4 64H317.6L301 297.5 288 304c-20.1 10.1-43.9 10.1-64 0s-43.9-10.1-64 0s-43.9 10.1-64 0l-13-6.5zM256 196c0-24-33.7-70.1-52.2-93.5c-6.1-7.7-17.5-7.7-23.6 0C161.7 125.9 128 172 128 196c0 33.1 28.7 60 64 60s64-26.9 64-60z"],display:[576,512,[],"e163","M64 0C28.7 0 0 28.7 0 64V352c0 35.3 28.7 64 64 64H240l-10.7 32H160c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H346.7L336 416H512c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zM512 64V352H64V64H512z"],"face-smile":[512,512,[128578,"smile"],"f118","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM164.1 325.5C182 346.2 212.6 368 256 368s74-21.8 91.9-42.5c5.8-6.7 15.9-7.4 22.6-1.6s7.4 15.9 1.6 22.6C349.8 372.1 311.1 400 256 400s-93.8-27.9-116.1-53.5c-5.8-6.7-5.1-16.8 1.6-22.6s16.8-5.1 22.6 1.6zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],thumbtack:[384,512,[128204,128392,"thumb-tack"],"f08d","M32 32C32 14.3 46.3 0 64 0H320c17.7 0 32 14.3 32 32s-14.3 32-32 32H290.5l11.4 148.2c36.7 19.9 65.7 53.2 79.5 94.7l1 3c3.3 9.8 1.6 20.5-4.4 28.8s-15.7 13.3-26 13.3H32c-10.3 0-19.9-4.9-26-13.3s-7.7-19.1-4.4-28.8l1-3c13.8-41.5 42.8-74.8 79.5-94.7L93.5 64H64C46.3 64 32 49.7 32 32zM160 384h64v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384z"],trophy:[576,512,[127942],"f091","M400 0H176c-26.5 0-48.1 21.8-47.1 48.2c.2 5.3 .4 10.6 .7 15.8H24C10.7 64 0 74.7 0 88c0 92.6 33.5 157 78.5 200.7c44.3 43.1 98.3 64.8 138.1 75.8c23.4 6.5 39.4 26 39.4 45.6c0 20.9-17 37.9-37.9 37.9H192c-17.7 0-32 14.3-32 32s14.3 32 32 32H384c17.7 0 32-14.3 32-32s-14.3-32-32-32H357.9C337 448 320 431 320 410.1c0-19.6 15.9-39.2 39.4-45.6c39.9-11 93.9-32.7 138.2-75.8C542.5 245 576 180.6 576 88c0-13.3-10.7-24-24-24H446.4c.3-5.2 .5-10.4 .7-15.8C448.1 21.8 426.5 0 400 0zM48.9 112h84.4c9.1 90.1 29.2 150.3 51.9 190.6c-24.9-11-50.8-26.5-73.2-48.3c-32-31.1-58-76-63-142.3zM464.1 254.3c-22.4 21.8-48.3 37.3-73.2 48.3c22.7-40.3 42.8-100.5 51.9-190.6h84.4c-5.1 66.3-31.1 111.2-63 142.3z"],"person-praying":[448,512,[128720,"pray"],"f683","M352 64A64 64 0 1 0 224 64a64 64 0 1 0 128 0zM232.7 264l22.9 31.5c6.5 8.9 16.3 14.7 27.2 16.1s21.9-1.7 30.4-8.7l88-72c17.1-14 19.6-39.2 5.6-56.3s-39.2-19.6-56.3-5.6l-55.2 45.2-26.2-36C253.6 156.7 228.6 144 202 144c-30.9 0-59.2 17.1-73.6 44.4L79.8 280.9c-20.2 38.5-9.4 85.9 25.6 111.8L158.6 432H72c-22.1 0-40 17.9-40 40s17.9 40 40 40H280c17.3 0 32.6-11.1 38-27.5s-.3-34.4-14.2-44.7L187.7 354l45-90z"],hammer:[576,512,[128296],"f6e3","M413.5 237.5c-28.2 4.8-58.2-3.6-80-25.4l-38.1-38.1C280.4 159 272 138.8 272 117.6V105.5L192.3 62c-5.3-2.9-8.6-8.6-8.3-14.7s3.9-11.5 9.5-14l47.2-21C259.1 4.2 279 0 299.2 0h18.1c36.7 0 72 14 98.7 39.1l44.6 42c24.2 22.8 33.2 55.7 26.6 86L503 183l8-8c9.4-9.4 24.6-9.4 33.9 0l24 24c9.4 9.4 9.4 24.6 0 33.9l-88 88c-9.4 9.4-24.6 9.4-33.9 0l-24-24c-9.4-9.4-9.4-24.6 0-33.9l8-8-17.5-17.5zM27.4 377.1L260.9 182.6c3.5 4.9 7.5 9.6 11.8 14l38.1 38.1c6 6 12.4 11.2 19.2 15.7L134.9 484.6c-14.5 17.4-36 27.4-58.6 27.4C34.1 512 0 477.8 0 435.7c0-22.6 10.1-44.1 27.4-58.6z"],"hand-peace":[512,512,[9996],"f25b","M224 0c17.7 0 32 14.3 32 32V240H192V32c0-17.7 14.3-32 32-32zm96 160c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V192c0-17.7 14.3-32 32-32zm64 64c0-17.7 14.3-32 32-32s32 14.3 32 32v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V224zM93.3 51.2L175.9 240H106.1L34.7 76.8C27.6 60.6 35 41.8 51.2 34.7s35.1 .3 42.1 16.5zm27 221.3l-.2-.5h69.9H216c22.1 0 40 17.9 40 40s-17.9 40-40 40H160c-8.8 0-16 7.2-16 16s7.2 16 16 16h56c39.8 0 72-32.2 72-72l0-.6c9.4 5.4 20.3 8.6 32 8.6c13.2 0 25.4-4 35.6-10.8c8.7 24.9 32.5 42.8 60.4 42.8c11.7 0 22.6-3.1 32-8.6V352c0 88.4-71.6 160-160 160H226.3c-42.4 0-83.1-16.9-113.1-46.9l-11.6-11.6C77.5 429.5 64 396.9 64 363V336c0-32.7 24.6-59.7 56.3-63.5z"],rotate:[512,512,[128260,"sync-alt"],"f2f1","M142.9 142.9c62.2-62.2 162.7-62.5 225.3-1L327 183c-6.9 6.9-8.9 17.2-5.2 26.2s12.5 14.8 22.2 14.8H463.5c0 0 0 0 0 0H472c13.3 0 24-10.7 24-24V72c0-9.7-5.8-18.5-14.8-22.2s-19.3-1.7-26.2 5.2L413.4 96.6c-87.6-86.5-228.7-86.2-315.8 1C73.2 122 55.6 150.7 44.8 181.4c-5.9 16.7 2.9 34.9 19.5 40.8s34.9-2.9 40.8-19.5c7.7-21.8 20.2-42.3 37.8-59.8zM16 312v7.6 .7V440c0 9.7 5.8 18.5 14.8 22.2s19.3 1.7 26.2-5.2l41.6-41.6c87.6 86.5 228.7 86.2 315.8-1c24.4-24.4 42.1-53.1 52.9-83.7c5.9-16.7-2.9-34.9-19.5-40.8s-34.9 2.9-40.8 19.5c-7.7 21.8-20.2 42.3-37.8 59.8c-62.2 62.2-162.7 62.5-225.3 1L185 329c6.9-6.9 8.9-17.2 5.2-26.2s-12.5-14.8-22.2-14.8H48.4h-.7H40c-13.3 0-24 10.7-24 24z"],spinner:[512,512,[],"f110","M304 48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zm0 416a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM48 304a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm464-48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM142.9 437A48 48 0 1 0 75 369.1 48 48 0 1 0 142.9 437zm0-294.2A48 48 0 1 0 75 75a48 48 0 1 0 67.9 67.9zM369.1 437A48 48 0 1 0 437 369.1 48 48 0 1 0 369.1 437z"],robot:[640,512,[129302],"f544","M320 0c17.7 0 32 14.3 32 32V96H472c39.8 0 72 32.2 72 72V440c0 39.8-32.2 72-72 72H168c-39.8 0-72-32.2-72-72V168c0-39.8 32.2-72 72-72H288V32c0-17.7 14.3-32 32-32zM208 384c-8.8 0-16 7.2-16 16s7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H208zm96 0c-8.8 0-16 7.2-16 16s7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H304zm96 0c-8.8 0-16 7.2-16 16s7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H400zM264 256a40 40 0 1 0 -80 0 40 40 0 1 0 80 0zm152 40a40 40 0 1 0 0-80 40 40 0 1 0 0 80zM48 224H64V416H48c-26.5 0-48-21.5-48-48V272c0-26.5 21.5-48 48-48zm544 0c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H576V224h16z"],peace:[512,512,[9774],"f67c","M224 445.3V323.5l-94.3 77.1c26.1 22.8 58.5 38.7 94.3 44.7zM89.2 351.1L224 240.8V66.7C133.2 81.9 64 160.9 64 256c0 34.6 9.2 67.1 25.2 95.1zm293.1 49.5L288 323.5V445.3c35.7-6 68.1-21.9 94.3-44.7zm40.6-49.5c16-28 25.2-60.5 25.2-95.1c0-95.1-69.2-174.1-160-189.3V240.8L422.8 351.1zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256z"],gears:[640,512,["cogs"],"f085","M308.5 135.3c7.1-6.3 9.9-16.2 6.2-25c-2.3-5.3-4.8-10.5-7.6-15.5L304 89.4c-3-5-6.3-9.9-9.8-14.6c-5.7-7.6-15.7-10.1-24.7-7.1l-28.2 9.3c-10.7-8.8-23-16-36.2-20.9L199 27.1c-1.9-9.3-9.1-16.7-18.5-17.8C173.9 8.4 167.2 8 160.4 8h-.7c-6.8 0-13.5 .4-20.1 1.2c-9.4 1.1-16.6 8.6-18.5 17.8L115 56.1c-13.3 5-25.5 12.1-36.2 20.9L50.5 67.8c-9-3-19-.5-24.7 7.1c-3.5 4.7-6.8 9.6-9.9 14.6l-3 5.3c-2.8 5-5.3 10.2-7.6 15.6c-3.7 8.7-.9 18.6 6.2 25l22.2 19.8C32.6 161.9 32 168.9 32 176s.6 14.1 1.7 20.9L11.5 216.7c-7.1 6.3-9.9 16.2-6.2 25c2.3 5.3 4.8 10.5 7.6 15.6l3 5.2c3 5.1 6.3 9.9 9.9 14.6c5.7 7.6 15.7 10.1 24.7 7.1l28.2-9.3c10.7 8.8 23 16 36.2 20.9l6.1 29.1c1.9 9.3 9.1 16.7 18.5 17.8c6.7 .8 13.5 1.2 20.4 1.2s13.7-.4 20.4-1.2c9.4-1.1 16.6-8.6 18.5-17.8l6.1-29.1c13.3-5 25.5-12.1 36.2-20.9l28.2 9.3c9 3 19 .5 24.7-7.1c3.5-4.7 6.8-9.5 9.8-14.6l3.1-5.4c2.8-5 5.3-10.2 7.6-15.5c3.7-8.7 .9-18.6-6.2-25l-22.2-19.8c1.1-6.8 1.7-13.8 1.7-20.9s-.6-14.1-1.7-20.9l22.2-19.8zM112 176a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM504.7 500.5c6.3 7.1 16.2 9.9 25 6.2c5.3-2.3 10.5-4.8 15.5-7.6l5.4-3.1c5-3 9.9-6.3 14.6-9.8c7.6-5.7 10.1-15.7 7.1-24.7l-9.3-28.2c8.8-10.7 16-23 20.9-36.2l29.1-6.1c9.3-1.9 16.7-9.1 17.8-18.5c.8-6.7 1.2-13.5 1.2-20.4s-.4-13.7-1.2-20.4c-1.1-9.4-8.6-16.6-17.8-18.5L583.9 307c-5-13.3-12.1-25.5-20.9-36.2l9.3-28.2c3-9 .5-19-7.1-24.7c-4.7-3.5-9.6-6.8-14.6-9.9l-5.3-3c-5-2.8-10.2-5.3-15.6-7.6c-8.7-3.7-18.6-.9-25 6.2l-19.8 22.2c-6.8-1.1-13.8-1.7-20.9-1.7s-14.1 .6-20.9 1.7l-19.8-22.2c-6.3-7.1-16.2-9.9-25-6.2c-5.3 2.3-10.5 4.8-15.6 7.6l-5.2 3c-5.1 3-9.9 6.3-14.6 9.9c-7.6 5.7-10.1 15.7-7.1 24.7l9.3 28.2c-8.8 10.7-16 23-20.9 36.2L315.1 313c-9.3 1.9-16.7 9.1-17.8 18.5c-.8 6.7-1.2 13.5-1.2 20.4s.4 13.7 1.2 20.4c1.1 9.4 8.6 16.6 17.8 18.5l29.1 6.1c5 13.3 12.1 25.5 20.9 36.2l-9.3 28.2c-3 9-.5 19 7.1 24.7c4.7 3.5 9.5 6.8 14.6 9.8l5.4 3.1c5 2.8 10.2 5.3 15.5 7.6c8.7 3.7 18.6 .9 25-6.2l19.8-22.2c6.8 1.1 13.8 1.7 20.9 1.7s14.1-.6 20.9-1.7l19.8 22.2zM464 304a48 48 0 1 1 0 96 48 48 0 1 1 0-96z"],warehouse:[640,512,[],"f494","M0 488V171.3c0-26.2 15.9-49.7 40.2-59.4L308.1 4.8c7.6-3.1 16.1-3.1 23.8 0L599.8 111.9c24.3 9.7 40.2 33.3 40.2 59.4V488c0 13.3-10.7 24-24 24H568c-13.3 0-24-10.7-24-24V224c0-17.7-14.3-32-32-32H128c-17.7 0-32 14.3-32 32V488c0 13.3-10.7 24-24 24H24c-13.3 0-24-10.7-24-24zm488 24l-336 0c-13.3 0-24-10.7-24-24V432H512l0 56c0 13.3-10.7 24-24 24zM128 400V336H512v64H128zm0-96V224H512l0 80H128z"],"arrow-up-right-dots":[576,512,[],"e4b7","M160 0c-17.7 0-32 14.3-32 32s14.3 32 32 32h50.7L9.4 265.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L256 109.3V160c0 17.7 14.3 32 32 32s32-14.3 32-32V32c0-17.7-14.3-32-32-32H160zM576 80a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM448 208a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM400 384a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm48 80a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zm128 0a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM272 384a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm48 80a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM144 512a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM576 336a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zm-48-80a48 48 0 1 0 0-96 48 48 0 1 0 0 96z"],splotch:[512,512,[],"f5bc","M208.5 62.3l28.1-36.9C248.8 9.4 267.8 0 288 0c28.5 0 53.6 18.7 61.8 46l17.8 59.4c10.3 34.4 36.1 62 69.8 74.6l39.8 14.9c20.9 7.9 34.8 27.9 34.8 50.2c0 16.9-7.9 32.8-21.5 42.9l-67.3 50.5c-24.3 18.2-37.2 47.9-33.8 78.1l2.5 22.7c4.3 38.7-26 72.6-65 72.6c-14.8 0-29.3-5.1-40.8-14.3l-55.4-44.3c-4.5-3.6-9.3-6.7-14.5-9.2c-15.8-7.9-33.7-10.4-51-7.3L82.4 451.9C47.8 458.2 16 431.6 16 396.5c0-13.2 4.7-26 13.1-36.2l11.2-13.4c14.6-17.4 22.6-39.4 22.6-62.1c0-18.8-5.5-37.2-15.8-53L8.8 173.5C3.1 164.7 0 154.4 0 143.9c0-33.4 30.1-58.8 63-53.2l51.3 8.7c35.9 6.1 72.2-8.2 94.2-37.1z"],"face-grin-hearts":[512,512,[128525,"grin-hearts"],"f584","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM388.1 312.8c12.3-3.8 24.3 6.9 19.3 18.7C382.4 390.6 324.2 432 256.3 432s-126.2-41.4-151.1-100.5c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19zM199.3 129.1c17.8 4.8 28.4 23.1 23.6 40.8l-17.4 65c-2.3 8.5-11.1 13.6-19.6 11.3l-65.1-17.4c-17.8-4.8-28.4-23.1-23.6-40.8s23.1-28.4 40.8-23.6l16.1 4.3 4.3-16.1c4.8-17.8 23.1-28.4 40.8-23.6zm154.3 23.6l4.3 16.1 16.1-4.3c17.8-4.8 36.1 5.8 40.8 23.6s-5.8 36.1-23.6 40.8l-65.1 17.4c-8.5 2.3-17.3-2.8-19.6-11.3l-17.4-65c-4.8-17.8 5.8-36.1 23.6-40.8s36.1 5.8 40.9 23.6z"],"dice-four":[448,512,[9859],"f524","M0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm160 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM128 384a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM352 160a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM320 384a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"sim-card":[384,512,[],"f7c4","M64 0H242.7c17 0 33.3 6.7 45.3 18.7L365.3 96c12 12 18.7 28.3 18.7 45.3V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64C0 28.7 28.7 0 64 0zM96 192c-17.7 0-32 14.3-32 32v32h64V192H96zM64 352h80 96 80V288H240 144 64v64zM320 224c0-17.7-14.3-32-32-32H256v64h64V224zM160 192v64h64V192H160zM288 448c17.7 0 32-14.3 32-32V384H256v64h32zM160 384v64h64V384H160zM64 416c0 17.7 14.3 32 32 32h32V384H64v32z"],transgender:[512,512,[9895,"transgender-alt"],"f225","M112 0c6.5 0 12.3 3.9 14.8 9.9s1.1 12.9-3.5 17.4l-31 31L112 78.1l7-7c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-7 7 15.2 15.2C187.7 107.6 220.5 96 256 96s68.3 11.6 94.9 31.2l68.8-68.8-31-31c-4.6-4.6-5.9-11.5-3.5-17.4s8.3-9.9 14.8-9.9h96c8.8 0 16 7.2 16 16v96c0 6.5-3.9 12.3-9.9 14.8s-12.9 1.1-17.4-3.5l-31-31-68.8 68.8C404.4 187.7 416 220.5 416 256c0 80.2-59 146.6-136 158.2V432h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H280v8c0 13.3-10.7 24-24 24s-24-10.7-24-24v-8H216c-13.3 0-24-10.7-24-24s10.7-24 24-24h16V414.2C155 402.6 96 336.2 96 256c0-35.5 11.6-68.3 31.2-94.9L112 145.9l-7 7c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l7-7L58.3 92.3l-31 31c-4.6 4.6-11.5 5.9-17.4 3.5S0 118.5 0 112V16C0 7.2 7.2 0 16 0h96zM352 256a96 96 0 1 0 -192 0 96 96 0 1 0 192 0z"],mercury:[384,512,[9791],"f223","M72.1 7C85.8-4 106-1.8 117 12c17.6 22 44.7 36 75 36s57.3-14 75-36c11.1-13.8 31.2-16 45-5s16 31.2 5 45c-7.8 9.7-16.6 18.4-26.4 26.1C337.3 109.7 368 163.3 368 224c0 89.1-66.2 162.7-152 174.4V424h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H216v16c0 13.3-10.7 24-24 24s-24-10.7-24-24V472H136c-13.3 0-24-10.7-24-24s10.7-24 24-24h32V398.4C82.2 386.7 16 313.1 16 224c0-60.7 30.7-114.3 77.5-145.9C83.7 70.5 74.9 61.7 67.1 52c-11.1-13.8-8.8-33.9 5-45zM80 224a112 112 0 1 0 224 0A112 112 0 1 0 80 224z"],"arrow-turn-down":[384,512,["level-down"],"f149","M32 64C14.3 64 0 49.7 0 32S14.3 0 32 0l96 0c53 0 96 43 96 96l0 306.7 73.4-73.4c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3l-128 128c-12.5 12.5-32.8 12.5-45.3 0l-128-128c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0L160 402.7 160 96c0-17.7-14.3-32-32-32L32 64z"],"person-falling-burst":[640,512,[],"e547","M256 32c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 9.8c0 39-23.7 74-59.9 88.4C71.6 154.5 32 213 32 278.2V352c0 17.7 14.3 32 32 32s32-14.3 32-32l0-73.8c0-10 1.6-19.8 4.5-29L261.1 497.4c9.6 14.8 29.4 19.1 44.3 9.5s19.1-29.4 9.5-44.3L222.6 320H224l80 0 38.4 51.2c10.6 14.1 30.7 17 44.8 6.4s17-30.7 6.4-44.8l-43.2-57.6C341.3 263.1 327.1 256 312 256l-71.5 0-56.8-80.2-.2-.3c44.7-29 72.5-79 72.5-133.6l0-9.8zM96 80A48 48 0 1 0 0 80a48 48 0 1 0 96 0zM464 286.1l58.6 53.9c4.8 4.4 11.9 5.5 17.8 2.6s9.5-9 9-15.5l-5.6-79.4 78.7-12.2c6.5-1 11.7-5.9 13.1-12.2s-1.1-13-6.5-16.7l-65.6-45.1L603 92.2c3.3-5.7 2.7-12.8-1.4-17.9s-10.9-7.2-17.2-5.3L508.3 92.1l-29.4-74C476.4 12 470.6 8 464 8s-12.4 4-14.9 10.1l-29.4 74L343.6 68.9c-6.3-1.9-13.1 .2-17.2 5.3s-4.6 12.2-1.4 17.9l39.5 69.1-65.6 45.1c-5.4 3.7-8 10.3-6.5 16.7c.1 .3 .1 .6 .2 .8l19.4 0c20.1 0 39.2 7.5 53.8 20.8l18.4 2.9L383 265.3l36.2 48.3c2.1 2.8 3.9 5.7 5.5 8.6L464 286.1z"],award:[384,512,[],"f559","M173.8 5.5c11-7.3 25.4-7.3 36.4 0L228 17.2c6 3.9 13 5.8 20.1 5.4l21.3-1.3c13.2-.8 25.6 6.4 31.5 18.2l9.6 19.1c3.2 6.4 8.4 11.5 14.7 14.7L344.5 83c11.8 5.9 19 18.3 18.2 31.5l-1.3 21.3c-.4 7.1 1.5 14.2 5.4 20.1l11.8 17.8c7.3 11 7.3 25.4 0 36.4L366.8 228c-3.9 6-5.8 13-5.4 20.1l1.3 21.3c.8 13.2-6.4 25.6-18.2 31.5l-19.1 9.6c-6.4 3.2-11.5 8.4-14.7 14.7L301 344.5c-5.9 11.8-18.3 19-31.5 18.2l-21.3-1.3c-7.1-.4-14.2 1.5-20.1 5.4l-17.8 11.8c-11 7.3-25.4 7.3-36.4 0L156 366.8c-6-3.9-13-5.8-20.1-5.4l-21.3 1.3c-13.2 .8-25.6-6.4-31.5-18.2l-9.6-19.1c-3.2-6.4-8.4-11.5-14.7-14.7L39.5 301c-11.8-5.9-19-18.3-18.2-31.5l1.3-21.3c.4-7.1-1.5-14.2-5.4-20.1L5.5 210.2c-7.3-11-7.3-25.4 0-36.4L17.2 156c3.9-6 5.8-13 5.4-20.1l-1.3-21.3c-.8-13.2 6.4-25.6 18.2-31.5l19.1-9.6C65 70.2 70.2 65 73.4 58.6L83 39.5c5.9-11.8 18.3-19 31.5-18.2l21.3 1.3c7.1 .4 14.2-1.5 20.1-5.4L173.8 5.5zM272 192a80 80 0 1 0 -160 0 80 80 0 1 0 160 0zM1.3 441.8L44.4 339.3c.2 .1 .3 .2 .4 .4l9.6 19.1c11.7 23.2 36 37.3 62 35.8l21.3-1.3c.2 0 .5 0 .7 .2l17.8 11.8c5.1 3.3 10.5 5.9 16.1 7.7l-37.6 89.3c-2.3 5.5-7.4 9.2-13.3 9.7s-11.6-2.2-14.8-7.2L74.4 455.5l-56.1 8.3c-5.7 .8-11.4-1.5-15-6s-4.3-10.7-2.1-16zm248 60.4L211.7 413c5.6-1.8 11-4.3 16.1-7.7l17.8-11.8c.2-.1 .4-.2 .7-.2l21.3 1.3c26 1.5 50.3-12.6 62-35.8l9.6-19.1c.1-.2 .2-.3 .4-.4l43.2 102.5c2.2 5.3 1.4 11.4-2.1 16s-9.3 6.9-15 6l-56.1-8.3-32.2 49.2c-3.2 5-8.9 7.7-14.8 7.2s-11-4.3-13.3-9.7z"],"ticket-simple":[576,512,["ticket-alt"],"f3ff","M0 128C0 92.7 28.7 64 64 64H512c35.3 0 64 28.7 64 64v64c0 8.8-7.4 15.7-15.7 18.6C541.5 217.1 528 235 528 256s13.5 38.9 32.3 45.4c8.3 2.9 15.7 9.8 15.7 18.6v64c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V320c0-8.8 7.4-15.7 15.7-18.6C34.5 294.9 48 277 48 256s-13.5-38.9-32.3-45.4C7.4 207.7 0 200.8 0 192V128z"],building:[384,512,[127970,61687],"f1ad","M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16z"],"angles-left":[512,512,[171,"angle-double-left"],"f100","M41.4 233.4c-12.5 12.5-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L109.3 256 246.6 118.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-160 160zm352-160l-160 160c-12.5 12.5-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L301.3 256 438.6 118.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0z"],qrcode:[448,512,[],"f029","M0 80C0 53.5 21.5 32 48 32h96c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80zM64 96v64h64V96H64zM0 336c0-26.5 21.5-48 48-48h96c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V336zm64 16v64h64V352H64zM304 32h96c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H304c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48zm80 64H320v64h64V96zM256 304c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16s7.2 16 16 16h32c8.8 0 16-7.2 16-16s7.2-16 16-16s16 7.2 16 16v96c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16s-7.2-16-16-16s-16 7.2-16 16v64c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V304zM368 480a16 16 0 1 1 0-32 16 16 0 1 1 0 32zm64 0a16 16 0 1 1 0-32 16 16 0 1 1 0 32z"],"clock-rotate-left":[512,512,["history"],"f1da","M75 75L41 41C25.9 25.9 0 36.6 0 57.9V168c0 13.3 10.7 24 24 24H134.1c21.4 0 32.1-25.9 17-41l-30.8-30.8C155 85.5 203 64 256 64c106 0 192 86 192 192s-86 192-192 192c-40.8 0-78.6-12.7-109.7-34.4c-14.5-10.1-34.4-6.6-44.6 7.9s-6.6 34.4 7.9 44.6C151.2 495 201.7 512 256 512c141.4 0 256-114.6 256-256S397.4 0 256 0C185.3 0 121.3 28.7 75 75zm181 53c-13.3 0-24 10.7-24 24V256c0 6.4 2.5 12.5 7 17l72 72c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-65-65V152c0-13.3-10.7-24-24-24z"],"face-grin-beam-sweat":[512,512,[128517,"grin-beam-sweat"],"f583","M476.8 126.3c-4.1 1.1-8.4 1.7-12.8 1.7c-26.5 0-48-21-48-47c0-5 1.8-11.3 4.6-18.1c.3-.7 .6-1.4 .9-2.1c9-20.2 26.5-44.9 36-57.5c3.2-4.4 9.6-4.4 12.8 0C483.4 20.6 512 61 512 81c0 21.7-14.9 39.8-35.2 45.3zM256 0c51.4 0 99.3 15.2 139.4 41.2c-1.5 3.1-3 6.2-4.3 9.3c-3.4 8-7.1 19-7.1 30.5c0 44.3 36.6 79 80 79c9.6 0 18.8-1.7 27.4-4.8c13.3 30.9 20.6 65 20.6 100.8c0 141.4-114.6 256-256 256S0 397.4 0 256S114.6 0 256 0zM388.1 312.8c-39.7 12.2-84.5 19-131.8 19s-92.1-6.8-131.8-19c-12.3-3.8-24.3 6.9-19.3 18.7c25 59.1 83.2 100.5 151.1 100.5s126.2-41.4 151.1-100.5c5-11.8-7-22.5-19.3-18.7zm-170.5-84l0 0 0 0c2.1 2.8 5.7 3.9 8.9 2.8s5.5-4.1 5.5-7.6c0-17.9-6.7-35.6-16.6-48.8c-9.8-13-23.9-23.2-39.4-23.2s-29.6 10.2-39.4 23.2C126.7 188.4 120 206.1 120 224c0 3.4 2.2 6.5 5.5 7.6s6.9 0 8.9-2.8l0 0 0 0 0 0 .2-.2c.2-.2 .4-.5 .7-.9c.6-.8 1.6-2 2.8-3.4c2.5-2.8 6-6.6 10.2-10.3c8.8-7.8 18.8-14 27.7-14s18.9 6.2 27.7 14c4.2 3.7 7.7 7.5 10.2 10.3c1.2 1.4 2.2 2.6 2.8 3.4c.3 .4 .6 .7 .7 .9l.2 .2 0 0 0 0zm160 0l0 0 0 0 0 0c2.1 2.8 5.7 3.9 8.9 2.8s5.5-4.1 5.5-7.6c0-17.9-6.7-35.6-16.6-48.8c-9.8-13-23.9-23.2-39.4-23.2s-29.6 10.2-39.4 23.2C286.7 188.4 280 206.1 280 224c0 3.4 2.2 6.5 5.5 7.6s6.9 0 8.9-2.8l0 0 0 0 0 0 .2-.2c.2-.2 .4-.5 .7-.9c.6-.8 1.6-2 2.8-3.4c2.5-2.8 6-6.6 10.2-10.3c8.8-7.8 18.8-14 27.7-14s18.9 6.2 27.7 14c4.2 3.7 7.7 7.5 10.2 10.3c1.2 1.4 2.2 2.6 2.8 3.4c.3 .4 .6 .7 .7 .9l.2 .2 0 0z"],"file-export":[576,512,["arrow-right-from-file"],"f56e","M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384V288H216c-13.3 0-24 10.7-24 24s10.7 24 24 24H384V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM384 336V288H494.1l-39-39c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l80 80c9.4 9.4 9.4 24.6 0 33.9l-80 80c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l39-39H384zm0-208H256V0L384 128z"],shield:[512,512,[128737,"shield-blank"],"f132","M256 0c4.6 0 9.2 1 13.4 2.9L457.7 82.8c22 9.3 38.4 31 38.3 57.2c-.5 99.2-41.3 280.7-213.6 363.2c-16.7 8-36.1 8-52.8 0C57.3 420.7 16.5 239.2 16 140c-.1-26.2 16.3-47.9 38.3-57.2L242.7 2.9C246.8 1 251.4 0 256 0z"],"arrow-up-short-wide":[576,512,["sort-amount-up-alt"],"f885","M151.6 42.4C145.5 35.8 137 32 128 32s-17.5 3.8-23.6 10.4l-88 96c-11.9 13-11.1 33.3 2 45.2s33.3 11.1 45.2-2L96 146.3V448c0 17.7 14.3 32 32 32s32-14.3 32-32V146.3l32.4 35.4c11.9 13 32.2 13.9 45.2 2s13.9-32.2 2-45.2l-88-96zM320 32c-17.7 0-32 14.3-32 32s14.3 32 32 32h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H320zm0 128c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H320zm0 128c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H320zm0 128c-17.7 0-32 14.3-32 32s14.3 32 32 32H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H320z"],"house-medical":[576,512,[],"e3b2","M543.8 287.6c17 0 32-14 32-32.1c1-9-3-17-11-24L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32V448c0 35.3 28.7 64 64 64H448.5c35.5 0 64.2-28.8 64-64.3l-.7-160.2h32zM256 208c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H320v48c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V320H208c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h48V208z"],"golf-ball-tee":[384,512,["golf-ball"],"f450","M384 192c0 66.8-34.1 125.6-85.8 160H85.8C34.1 317.6 0 258.8 0 192C0 86 86 0 192 0S384 86 384 192zM242.1 256.6c0 18.5-15 33.5-33.5 33.5c-4.9 0-9.1 5.1-5.4 8.4c5.9 5.2 13.7 8.4 22.1 8.4c18.5 0 33.5-15 33.5-33.5c0-8.5-3.2-16.2-8.4-22.1c-3.3-3.7-8.4 .5-8.4 5.4zm-52.3-49.3c-4.9 0-9.1 5.1-5.4 8.4c5.9 5.2 13.7 8.4 22.1 8.4c18.5 0 33.5-15 33.5-33.5c0-8.5-3.2-16.2-8.4-22.1c-3.3-3.7-8.4 .5-8.4 5.4c0 18.5-15 33.5-33.5 33.5zm113.5-17.5c0 18.5-15 33.5-33.5 33.5c-4.9 0-9.1 5.1-5.4 8.4c5.9 5.2 13.7 8.4 22.1 8.4c18.5 0 33.5-15 33.5-33.5c0-8.5-3.2-16.2-8.4-22.1c-3.3-3.7-8.4 .5-8.4 5.4zM96 416c0-17.7 14.3-32 32-32h64 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H240c-8.8 0-16 7.2-16 16v16c0 17.7-14.3 32-32 32s-32-14.3-32-32V464c0-8.8-7.2-16-16-16H128c-17.7 0-32-14.3-32-32z"],"circle-chevron-left":[512,512,["chevron-circle-left"],"f137","M512 256A256 256 0 1 0 0 256a256 256 0 1 0 512 0zM271 135c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-87 87 87 87c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L167 273c-9.4-9.4-9.4-24.6 0-33.9L271 135z"],"house-chimney-window":[576,512,[],"e00d","M575.8 255.5c0 18-15 32.1-32 32.1h-32l.7 160.2c.2 35.5-28.5 64.3-64 64.3H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L416 100.7V64c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V185l52.8 46.4c8 7 12 15 11 24zM248 192c-13.3 0-24 10.7-24 24v80c0 13.3 10.7 24 24 24h80c13.3 0 24-10.7 24-24V216c0-13.3-10.7-24-24-24H248z"],"pen-nib":[512,512,[10001],"f5ad","M368.4 18.3L312.7 74.1 437.9 199.3l55.7-55.7c21.9-21.9 21.9-57.3 0-79.2L447.6 18.3c-21.9-21.9-57.3-21.9-79.2 0zM288 94.6l-9.2 2.8L134.7 140.6c-19.9 6-35.7 21.2-42.3 41L3.8 445.8c-3.8 11.3-1 23.9 7.3 32.4L164.7 324.7c-3-6.3-4.7-13.3-4.7-20.7c0-26.5 21.5-48 48-48s48 21.5 48 48s-21.5 48-48 48c-7.4 0-14.4-1.7-20.7-4.7L33.7 500.9c8.6 8.3 21.1 11.2 32.4 7.3l264.3-88.6c19.7-6.6 35-22.4 41-42.3l43.2-144.1 2.8-9.2L288 94.6z"],"tent-arrow-turn-left":[576,512,[],"e580","M120.1 41.8c9.9-8.9 10.7-24 1.8-33.9S97.8-2.7 87.9 6.2l-80 72C2.9 82.7 0 89.2 0 96s2.9 13.3 7.9 17.8l80 72c9.9 8.9 25 8.1 33.9-1.8s8.1-25-1.8-33.9L86.5 120 456 120c39.8 0 72 32.2 72 72v40c0 13.3 10.7 24 24 24s24-10.7 24-24V192c0-66.3-53.7-120-120-120L86.5 72l33.5-30.2zM307.4 166.5c-11.5-8.7-27.3-8.7-38.8 0l-168 128c-6.6 5-11 12.5-12.3 20.7l-24 160c-1.4 9.2 1.3 18.6 7.4 25.6S86.7 512 96 512H240h16c17.7 0 32-14.3 32-32V361.9c0-5.5 4.4-9.9 9.9-9.9c3.7 0 7.2 2.1 8.8 5.5l68.4 136.8c5.4 10.8 16.5 17.7 28.6 17.7H464h16c9.3 0 18.2-4.1 24.2-11.1s8.8-16.4 7.4-25.6l-24-160c-1.2-8.2-5.6-15.7-12.3-20.7l-168-128z"],tents:[640,512,[],"e582","M396.6 6.5L235.8 129.1c9.6 1.8 18.9 5.8 27 12l168 128c13.2 10.1 22 24.9 24.5 41.4l6.2 41.5H608c9.3 0 18.2-4.1 24.2-11.1s8.8-16.4 7.4-25.6l-24-160c-1.2-8.2-5.6-15.7-12.3-20.7l-168-128c-11.5-8.7-27.3-8.7-38.8 0zm-153.2 160c-11.5-8.7-27.3-8.7-38.8 0l-168 128c-6.6 5-11 12.5-12.3 20.7l-24 160c-1.4 9.2 1.3 18.6 7.4 25.6S22.7 512 32 512H176h16c17.7 0 32-14.3 32-32V361.9c0-5.5 4.4-9.9 9.9-9.9c3.7 0 7.2 2.1 8.8 5.5l68.4 136.8c5.4 10.8 16.5 17.7 28.6 17.7H400h16c9.3 0 18.2-4.1 24.2-11.1s8.8-16.4 7.4-25.6l-24-160c-1.2-8.2-5.6-15.7-12.3-20.7l-168-128z"],"wand-magic":[512,512,["magic"],"f0d0","M14.1 463.3c-18.7-18.7-18.7-49.1 0-67.9L395.4 14.1c18.7-18.7 49.1-18.7 67.9 0l34.6 34.6c18.7 18.7 18.7 49.1 0 67.9L116.5 497.9c-18.7 18.7-49.1 18.7-67.9 0L14.1 463.3zM347.6 187.6l105-105L429.4 59.3l-105 105 23.3 23.3z"],dog:[576,512,[128021],"f6d3","M309.6 158.5L332.7 19.8C334.6 8.4 344.5 0 356.1 0c7.5 0 14.5 3.5 19 9.5L392 32h52.1c12.7 0 24.9 5.1 33.9 14.1L496 64h56c13.3 0 24 10.7 24 24v24c0 44.2-35.8 80-80 80H464 448 426.7l-5.1 30.5-112-64zM416 256.1L416 480c0 17.7-14.3 32-32 32H352c-17.7 0-32-14.3-32-32V364.8c-24 12.3-51.2 19.2-80 19.2s-56-6.9-80-19.2V480c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V249.8c-28.8-10.9-51.4-35.3-59.2-66.5L1 167.8c-4.3-17.1 6.1-34.5 23.3-38.8s34.5 6.1 38.8 23.3l3.9 15.5C70.5 182 83.3 192 98 192h30 16H303.8L416 256.1zM464 80a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z"],carrot:[512,512,[129365],"f787","M346.7 6C337.6 17 320 42.3 320 72c0 40 15.3 55.3 40 80s40 40 80 40c29.7 0 55-17.6 66-26.7c4-3.3 6-8.2 6-13.3s-2-10-6-13.2c-11.4-9.1-38.3-26.8-74-26.8c-32 0-40 8-40 8s8-8 8-40c0-35.7-17.7-62.6-26.8-74C370 2 365.1 0 360 0s-10 2-13.3 6zM244.6 136c-40 0-77.1 18.1-101.7 48.2l60.5 60.5c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0l-55.3-55.3 0 .1L2.2 477.9C-2 487-.1 497.8 7 505s17.9 9 27.1 4.8l134.7-62.4-52.1-52.1c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L199.7 433l100.2-46.4c46.4-21.5 76.2-68 76.2-119.2C376 194.8 317.2 136 244.6 136z"],moon:[384,512,[127769,9214],"f186","M223.5 32C100 32 0 132.3 0 256S100 480 223.5 480c60.6 0 115.5-24.2 155.8-63.4c5-4.9 6.3-12.5 3.1-18.7s-10.1-9.7-17-8.5c-9.8 1.7-19.8 2.6-30.1 2.6c-96.9 0-175.5-78.8-175.5-176c0-65.8 36-123.1 89.3-153.3c6.1-3.5 9.2-10.5 7.7-17.3s-7.3-11.9-14.3-12.5c-6.3-.5-12.6-.8-19-.8z"],"wine-glass-empty":[320,512,["wine-glass-alt"],"f5ce","M64 0C47.4 0 33.5 12.8 32.1 29.3l-14 168.4c-6 72 42.5 135.2 109.9 150.6V448H80c-17.7 0-32 14.3-32 32s14.3 32 32 32h80 80c17.7 0 32-14.3 32-32s-14.3-32-32-32H192V348.4c67.4-15.4 115.9-78.6 109.9-150.6l-14-168.4C286.5 12.8 272.6 0 256 0H64zM81.9 203.1L93.4 64H226.6l11.6 139.1C242 248.8 205.9 288 160 288s-82-39.2-78.1-84.9z"],cheese:[512,512,[],"f7ef","M512 240.2V256H0c0-20 10-38.7 26.6-49.8L274.9 40.7c8.6-5.7 18.6-8.7 28.9-8.7C418.8 32 512 125.2 512 240.2zm0 47.8V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V288H512z"],"yin-yang":[512,512,[9775],"f6ad","M256 64c53 0 96 43 96 96s-43 96-96 96s-96 43-96 96s43 96 96 96C150 448 64 362 64 256S150 64 256 64zm0 448A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm32-352a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM224 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z"],music:[512,512,[127925],"f001","M499.1 6.3c8.1 6 12.9 15.6 12.9 25.7v72V368c0 44.2-43 80-96 80s-96-35.8-96-80s43-80 96-80c11.2 0 22 1.6 32 4.6V147L192 223.8V432c0 44.2-43 80-96 80s-96-35.8-96-80s43-80 96-80c11.2 0 22 1.6 32 4.6V200 128c0-14.1 9.3-26.6 22.8-30.7l320-96c9.7-2.9 20.2-1.1 28.3 5z"],"code-commit":[640,512,[],"f386","M320 336a80 80 0 1 0 0-160 80 80 0 1 0 0 160zm156.8-48C462 361 397.4 416 320 416s-142-55-156.8-128H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H163.2C178 151 242.6 96 320 96s142 55 156.8 128H608c17.7 0 32 14.3 32 32s-14.3 32-32 32H476.8z"],"temperature-low":[512,512,[],"f76b","M448 96a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM320 96a96 96 0 1 1 192 0A96 96 0 1 1 320 96zM144 64c-26.5 0-48 21.5-48 48V276.5c0 17.3-7.1 31.9-15.3 42.5C70.2 332.6 64 349.5 64 368c0 44.2 35.8 80 80 80s80-35.8 80-80c0-18.5-6.2-35.4-16.7-48.9c-8.2-10.6-15.3-25.2-15.3-42.5V112c0-26.5-21.5-48-48-48zM32 112C32 50.2 82.1 0 144 0s112 50.1 112 112V276.5c0 .1 .1 .3 .2 .6c.2 .6 .8 1.6 1.7 2.8c18.9 24.4 30.1 55 30.1 88.1c0 79.5-64.5 144-144 144S0 447.5 0 368c0-33.2 11.2-63.8 30.1-88.1c.9-1.2 1.5-2.2 1.7-2.8c.1-.3 .2-.5 .2-.6V112zM192 368c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-20.9 13.4-38.7 32-45.3V272c0-8.8 7.2-16 16-16s16 7.2 16 16v50.7c18.6 6.6 32 24.4 32 45.3z"],"person-biking":[640,512,[128692,"biking"],"f84a","M400 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm27.2 64l-61.8-48.8c-17.3-13.6-41.7-13.8-59.1-.3l-83.1 64.2c-30.7 23.8-28.5 70.8 4.3 91.6L288 305.1V416c0 17.7 14.3 32 32 32s32-14.3 32-32V288c0-10.7-5.3-20.7-14.2-26.6L295 232.9l60.3-48.5L396 217c5.7 4.5 12.7 7 20 7h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H427.2zM56 384a72 72 0 1 1 144 0A72 72 0 1 1 56 384zm200 0A128 128 0 1 0 0 384a128 128 0 1 0 256 0zm184 0a72 72 0 1 1 144 0 72 72 0 1 1 -144 0zm200 0a128 128 0 1 0 -256 0 128 128 0 1 0 256 0z"],broom:[576,512,[129529],"f51a","M566.6 54.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-192 192-34.7-34.7c-4.2-4.2-10-6.6-16-6.6c-12.5 0-22.6 10.1-22.6 22.6v29.1L364.3 320h29.1c12.5 0 22.6-10.1 22.6-22.6c0-6-2.4-11.8-6.6-16l-34.7-34.7 192-192zM341.1 353.4L222.6 234.9c-42.7-3.7-85.2 11.7-115.8 42.3l-8 8C76.5 307.5 64 337.7 64 369.2c0 6.8 7.1 11.2 13.2 8.2l51.1-25.5c5-2.5 9.5 4.1 5.4 7.9L7.3 473.4C2.7 477.6 0 483.6 0 489.9C0 502.1 9.9 512 22.1 512l173.3 0c38.8 0 75.9-15.4 103.4-42.8c30.6-30.6 45.9-73.1 42.3-115.8z"],"shield-heart":[512,512,[],"e574","M269.4 2.9C265.2 1 260.7 0 256 0s-9.2 1-13.4 2.9L54.3 82.8c-22 9.3-38.4 31-38.3 57.2c.5 99.2 41.3 280.7 213.6 363.2c16.7 8 36.1 8 52.8 0C454.7 420.7 495.5 239.2 496 140c.1-26.2-16.3-47.9-38.3-57.2L269.4 2.9zM144 221.3c0-33.8 27.4-61.3 61.3-61.3c16.2 0 31.8 6.5 43.3 17.9l7.4 7.4 7.4-7.4c11.5-11.5 27.1-17.9 43.3-17.9c33.8 0 61.3 27.4 61.3 61.3c0 16.2-6.5 31.8-17.9 43.3l-82.7 82.7c-6.2 6.2-16.4 6.2-22.6 0l-82.7-82.7c-11.5-11.5-17.9-27.1-17.9-43.3z"],gopuram:[512,512,[],"f664","M120 0c13.3 0 24 10.7 24 24v8h40V24c0-13.3 10.7-24 24-24s24 10.7 24 24v8h48V24c0-13.3 10.7-24 24-24s24 10.7 24 24v8h40V24c0-13.3 10.7-24 24-24s24 10.7 24 24v8V64v64c17.7 0 32 14.3 32 32v64c17.7 0 32 14.3 32 32v96c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H416V352H384V224H352V128H320v96h32V352h32V512H304V464c0-26.5-21.5-48-48-48s-48 21.5-48 48v48H128V352h32V224h32V128H160v96H128V352H96V512H32c-17.7 0-32-14.3-32-32V384c0-17.7 14.3-32 32-32V256c0-17.7 14.3-32 32-32V160c0-17.7 14.3-32 32-32V64 32 24c0-13.3 10.7-24 24-24zM256 272c-17.7 0-32 14.3-32 32v48h64V304c0-17.7-14.3-32-32-32zm-32-80v32h64V192c0-17.7-14.3-32-32-32s-32 14.3-32 32z"],"earth-oceania":[512,512,["globe-oceania"],"e47b","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM208.6 357.3l-39-13.5c-6.5-2.2-13.6-2.3-20.1-.3l-15.3 4.9c-18.5 5.9-38.5-2.4-47.5-19.5l-3.3-6.2c-10.6-20.1-2.3-45 18.2-54.7l35.3-16.8c2.3-1.1 4.4-2.8 5.9-4.8l5.3-7c7.2-9.6 18.6-15.3 30.6-15.3s23.4 5.7 30.6 15.3l4.6 6.1c2 2.6 4.9 4.5 8.1 5.1c7.8 1.6 15.7-1.5 20.4-7.9l10.4-14.2c2-2.8 5.3-4.4 8.7-4.4c4.4 0 8.4 2.7 10 6.8l10.1 25.9c2.8 7.2 6.7 14 11.5 20.2L311 299.8c5.8 7.4 9 16.6 9 26s-3.2 18.6-9 26L299 367.2c-8.3 10.6-21 16.8-34.4 16.8c-8.4 0-16.6-2.4-23.7-7l-25.4-16.4c-2.2-1.4-4.5-2.5-6.9-3.4zm65.2-214.8L296 164.7c10.1 10.1 2.9 27.3-11.3 27.3H254.8c-5.6 0-11.1-1.2-16.2-3.4l-42.8-19c-14.3-6.3-11.9-27.3 3.4-30.3l38.5-7.7c13.1-2.6 26.7 1.5 36.1 10.9zM248 432c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16s-7.2 16-16 16H264c-8.8 0-16-7.2-16-16zM431.2 298.9l8 24c2.8 8.4-1.7 17.4-10.1 20.2s-17.4-1.7-20.2-10.1l-8-24c-2.8-8.4 1.7-17.4 10.1-20.2s17.4 1.7 20.2 10.1zm-19.9 80.4l-32 32c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6l32-32c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z"],"square-xmark":[448,512,[10062,"times-square","xmark-square"],"f2d3","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm79 143c9.4-9.4 24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9z"],hashtag:[448,512,[62098],"23","M181.3 32.4c17.4 2.9 29.2 19.4 26.3 36.8L197.8 128h95.1l11.5-69.3c2.9-17.4 19.4-29.2 36.8-26.3s29.2 19.4 26.3 36.8L357.8 128H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H347.1L325.8 320H384c17.7 0 32 14.3 32 32s-14.3 32-32 32H315.1l-11.5 69.3c-2.9 17.4-19.4 29.2-36.8 26.3s-29.2-19.4-26.3-36.8l9.8-58.7H155.1l-11.5 69.3c-2.9 17.4-19.4 29.2-36.8 26.3s-29.2-19.4-26.3-36.8L90.2 384H32c-17.7 0-32-14.3-32-32s14.3-32 32-32h68.9l21.3-128H64c-17.7 0-32-14.3-32-32s14.3-32 32-32h68.9l11.5-69.3c2.9-17.4 19.4-29.2 36.8-26.3zM187.1 192L165.8 320h95.1l21.3-128H187.1z"],"up-right-and-down-left-from-center":[512,512,["expand-alt"],"f424","M344 0H488c13.3 0 24 10.7 24 24V168c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-39-39-87 87c-9.4 9.4-24.6 9.4-33.9 0l-32-32c-9.4-9.4-9.4-24.6 0-33.9l87-87L327 41c-6.9-6.9-8.9-17.2-5.2-26.2S334.3 0 344 0zM168 512H24c-13.3 0-24-10.7-24-24V344c0-9.7 5.8-18.5 14.8-22.2s19.3-1.7 26.2 5.2l39 39 87-87c9.4-9.4 24.6-9.4 33.9 0l32 32c9.4 9.4 9.4 24.6 0 33.9l-87 87 39 39c6.9 6.9 8.9 17.2 5.2 26.2s-12.5 14.8-22.2 14.8z"],"oil-can":[640,512,[],"f613","M320 128c17.7 0 32-14.3 32-32s-14.3-32-32-32H192c-17.7 0-32 14.3-32 32s14.3 32 32 32h32v32H144 96 48c-26.5 0-48 21.5-48 48v64.8c0 19 11.2 36.2 28.5 43.9l67.5 30V368c0 26.5 21.5 48 48 48H403.1c18.4 0 35.8-7.9 48-21.7L633.5 187.7c12.3-13.9-.3-35.4-18.4-31.5L448 192l-50.5-25.2c-8.9-4.4-18.7-6.8-28.6-6.8H288V128h32zM96 208v86.1L48 272.8V208H96z"],t:[384,512,[116],"54","M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96H160V448c0 17.7 14.3 32 32 32s32-14.3 32-32V96H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H192 32z"],hippo:[640,512,[129435],"f6ed","M407 47c9.4-9.4 24.6-9.4 33.9 0l17.2 17.2c1.9-.1 3.9-.2 5.8-.2h32c11.2 0 21.9 2.3 31.6 6.5L543 55c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9L564 101.9c7.6 12.2 12 26.7 12 42.1c0 10.2 7.4 18.8 16.7 23c27.9 12.5 47.3 40.5 47.3 73c0 26.2-12.6 49.4-32 64v32c0 8.8-7.2 16-16 16H560c-8.8 0-16-7.2-16-16V320H480v16c0 8.8-7.2 16-16 16H432c-8.8 0-16-7.2-16-16V318.4c-11.8-2.4-22.7-7.4-32-14.4c-1.5-1.1-2.9-2.3-4.3-3.5c-17-14.7-27.7-36.4-27.7-60.5c0-8.8-7.2-16-16-16s-16 7.2-16 16c0 44.7 26.2 83.2 64 101.2V352c0 17.7 14.3 32 32 32h32v64c0 17.7-14.3 32-32 32H352c-17.7 0-32-14.3-32-32V372c-19.8 7.7-41.4 12-64 12s-44.2-4.3-64-12v76c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V329.1L45.9 369.7c-5.4 12.1-19.6 17.6-31.7 12.2S-3.3 362.4 2.1 350.3L24 300.9c5.3-11.9 8-24.7 8-37.7C32 155.7 117.2 68 223.8 64.1l.2-.1h7.2H256h32c41.7 0 83.4 12.1 117.2 25.7c1.7-1.8 3.5-3.6 5.3-5.2L407 81c-9.4-9.4-9.4-24.6 0-33.9zm73 185a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zm88 24a24 24 0 1 0 0-48 24 24 0 1 0 0 48zM480 144a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm48 16a16 16 0 1 0 0-32 16 16 0 1 0 0 32z"],"chart-column":[512,512,[],"e0e3","M32 32c17.7 0 32 14.3 32 32V400c0 8.8 7.2 16 16 16H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H80c-44.2 0-80-35.8-80-80V64C0 46.3 14.3 32 32 32zM160 224c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V256c0-17.7 14.3-32 32-32zm128-64V320c0 17.7-14.3 32-32 32s-32-14.3-32-32V160c0-17.7 14.3-32 32-32s32 14.3 32 32zm64 32c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V224c0-17.7 14.3-32 32-32zM480 96V320c0 17.7-14.3 32-32 32s-32-14.3-32-32V96c0-17.7 14.3-32 32-32s32 14.3 32 32z"],infinity:[640,512,[8734,9854],"f534","M0 241.1C0 161 65 96 145.1 96c38.5 0 75.4 15.3 102.6 42.5L320 210.7l72.2-72.2C419.5 111.3 456.4 96 494.9 96C575 96 640 161 640 241.1v29.7C640 351 575 416 494.9 416c-38.5 0-75.4-15.3-102.6-42.5L320 301.3l-72.2 72.2C220.5 400.7 183.6 416 145.1 416C65 416 0 351 0 270.9V241.1zM274.7 256l-72.2-72.2c-15.2-15.2-35.9-23.8-57.4-23.8C100.3 160 64 196.3 64 241.1v29.7c0 44.8 36.3 81.1 81.1 81.1c21.5 0 42.2-8.5 57.4-23.8L274.7 256zm90.5 0l72.2 72.2c15.2 15.2 35.9 23.8 57.4 23.8c44.8 0 81.1-36.3 81.1-81.1V241.1c0-44.8-36.3-81.1-81.1-81.1c-21.5 0-42.2 8.5-57.4 23.8L365.3 256z"],"vial-circle-check":[512,512,[],"e596","M0 64C0 46.3 14.3 32 32 32H96h64 64c17.7 0 32 14.3 32 32s-14.3 32-32 32V266.8c-20.2 28.6-32 63.5-32 101.2c0 25.2 5.3 49.1 14.8 70.8C189.5 463.7 160.6 480 128 480c-53 0-96-43-96-96V96C14.3 96 0 81.7 0 64zM96 96v96h64V96H96zM224 368a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm211.3-43.3c-6.2-6.2-16.4-6.2-22.6 0L352 385.4l-28.7-28.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l40 40c6.2 6.2 16.4 6.2 22.6 0l72-72c6.2-6.2 6.2-16.4 0-22.6z"],"person-arrow-down-to-line":[640,512,[],"e538","M192 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-8 352V352h16v96H184zm-64 0H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H152h80H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H264V256.9l28.6 47.5c9.1 15.1 28.8 20 43.9 10.9s20-28.8 10.9-43.9l-58.3-97c-17.4-28.9-48.6-46.6-82.3-46.6H177.1c-33.7 0-64.9 17.7-82.3 46.6l-58.3 97c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9L120 256.9V448zM464 64V306.7l-25.4-25.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l80 80c12.5 12.5 32.8 12.5 45.3 0l80-80c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L528 306.7V64c0-17.7-14.3-32-32-32s-32 14.3-32 32z"],voicemail:[640,512,[],"f897","M144 320a80 80 0 1 1 0-160 80 80 0 1 1 0 160zm119.8 0c15.3-22.9 24.2-50.4 24.2-80c0-79.5-64.5-144-144-144S0 160.5 0 240s64.5 144 144 144H496c79.5 0 144-64.5 144-144s-64.5-144-144-144s-144 64.5-144 144c0 29.6 8.9 57.1 24.2 80H263.8zM496 160a80 80 0 1 1 0 160 80 80 0 1 1 0-160z"],fan:[512,512,[],"f863","M258.6 0c-1.7 0-3.4 .1-5.1 .5C168 17 115.6 102.3 130.5 189.3c2.9 17 8.4 32.9 15.9 47.4L32 224H29.4C13.2 224 0 237.2 0 253.4c0 1.7 .1 3.4 .5 5.1C17 344 102.3 396.4 189.3 381.5c17-2.9 32.9-8.4 47.4-15.9L224 480v2.6c0 16.2 13.2 29.4 29.4 29.4c1.7 0 3.4-.1 5.1-.5C344 495 396.4 409.7 381.5 322.7c-2.9-17-8.4-32.9-15.9-47.4L480 288h2.6c16.2 0 29.4-13.2 29.4-29.4c0-1.7-.1-3.4-.5-5.1C495 168 409.7 115.6 322.7 130.5c-17 2.9-32.9 8.4-47.4 15.9L288 32V29.4C288 13.2 274.8 0 258.6 0zM256 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"person-walking-luggage":[576,512,[],"e554","M432 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM347.7 200.5c1-.4 1.9-.8 2.9-1.2l-16.9 63.5c-5.6 21.1-.1 43.6 14.7 59.7l70.7 77.1 22 88.1c4.3 17.1 21.7 27.6 38.8 23.3s27.6-21.7 23.3-38.8l-23-92.1c-1.9-7.8-5.8-14.9-11.2-20.8l-49.5-54 19.3-65.5 9.6 23c4.4 10.6 12.5 19.3 22.8 24.5l26.7 13.3c15.8 7.9 35 1.5 42.9-14.3s1.5-35-14.3-42.9L505 232.7l-15.3-36.8C472.5 154.8 432.3 128 387.7 128c-22.8 0-45.3 4.8-66.1 14l-8 3.5c-32.9 14.6-58.1 42.4-69.4 76.5l-2.6 7.8c-5.6 16.8 3.5 34.9 20.2 40.5s34.9-3.5 40.5-20.2l2.6-7.8c5.7-17.1 18.3-30.9 34.7-38.2l8-3.5zm-30 135.1l-25 62.4-59.4 59.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L340.3 441c4.6-4.6 8.2-10.1 10.6-16.1l14.5-36.2-40.7-44.4c-2.5-2.7-4.8-5.6-7-8.6zM256 274.1c-7.7-4.4-17.4-1.8-21.9 5.9l-32 55.4L147.7 304c-15.3-8.8-34.9-3.6-43.7 11.7L40 426.6c-8.8 15.3-3.6 34.9 11.7 43.7l55.4 32c15.3 8.8 34.9 3.6 43.7-11.7l64-110.9c1.5-2.6 2.6-5.2 3.3-8L261.9 296c4.4-7.7 1.8-17.4-5.9-21.9z"],"up-down":[256,512,[8597,11021,"arrows-alt-v"],"f338","M145.6 7.7C141 2.8 134.7 0 128 0s-13 2.8-17.6 7.7l-104 112c-6.5 7-8.2 17.2-4.4 25.9S14.5 160 24 160H80V352H24c-9.5 0-18.2 5.7-22 14.4s-2.1 18.9 4.4 25.9l104 112c4.5 4.9 10.9 7.7 17.6 7.7s13-2.8 17.6-7.7l104-112c6.5-7 8.2-17.2 4.4-25.9s-12.5-14.4-22-14.4H176V160h56c9.5 0 18.2-5.7 22-14.4s2.1-18.9-4.4-25.9l-104-112z"],"cloud-moon-rain":[576,512,[],"f73c","M481.2 0C417 0 363.5 46.5 353.7 107.6c35.4 17.6 60.2 53.3 62.1 95.1c23.2 11 42 29.7 53.1 52.7c4 .4 8.1 .6 12.3 .6c34.9 0 66.7-13.8 89.9-36.1c5.1-4.9 6.4-12.5 3.2-18.7s-10.1-9.7-17-8.6c-4.9 .8-10 1.3-15.2 1.3c-49 0-88.4-39.3-88.4-87.4c0-32.6 18-61.1 44.9-76.1c6.1-3.4 9.3-10.5 7.8-17.4s-7.3-12-14.3-12.6c-3.6-.3-7.3-.5-10.9-.5zM367.9 383.9c44.2 0 80-35.8 80-80c0-39.3-28.4-72.1-65.8-78.7c1.2-5.6 1.9-11.3 1.9-17.2c0-44.2-35.8-80-80-80c-17 0-32.8 5.3-45.8 14.4C241.3 114.6 210.8 96 176 96c-53 0-96 43-96 96l0 1.3c-45.4 7.6-80 47.1-80 94.6c0 53 43 96 96 96H367.9zM85.4 420.1c-11-7.4-25.9-4.4-33.3 6.7l-32 48c-7.4 11-4.4 25.9 6.7 33.3s25.9 4.4 33.3-6.7l32-48c7.4-11 4.4-25.9-6.7-33.3zm96 0c-11-7.4-25.9-4.4-33.3 6.7l-32 48c-7.4 11-4.4 25.9 6.7 33.3s25.9 4.4 33.3-6.7l32-48c7.4-11 4.4-25.9-6.7-33.3zm96 0c-11-7.4-25.9-4.4-33.3 6.7l-32 48c-7.4 11-4.4 25.9 6.7 33.3s25.9 4.4 33.3-6.7l32-48c7.4-11 4.4-25.9-6.7-33.3zm96 0c-11-7.4-25.9-4.4-33.3 6.7l-32 48c-7.4 11-4.4 25.9 6.7 33.3s25.9 4.4 33.3-6.7l32-48c7.4-11 4.4-25.9-6.7-33.3z"],calendar:[448,512,[128197,128198],"f133","M96 32V64H48C21.5 64 0 85.5 0 112v48H448V112c0-26.5-21.5-48-48-48H352V32c0-17.7-14.3-32-32-32s-32 14.3-32 32V64H160V32c0-17.7-14.3-32-32-32S96 14.3 96 32zM448 192H0V464c0 26.5 21.5 48 48 48H400c26.5 0 48-21.5 48-48V192z"],trailer:[640,512,[],"e041","M48 32C21.5 32 0 53.5 0 80V336c0 26.5 21.5 48 48 48H65.1c7.8-54.3 54.4-96 110.9-96s103.1 41.7 110.9 96H488h8H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H544V80c0-26.5-21.5-48-48-48H48zM80 96c8.8 0 16 7.2 16 16l0 131.2c-11.4 5.9-22.2 12.9-32 21V112c0-8.8 7.2-16 16-16zm96 128c-5.4 0-10.7 .2-16 .7L160 112c0-8.8 7.2-16 16-16s16 7.2 16 16l0 112.7c-5.3-.5-10.6-.7-16-.7zm80 19.2L256 112c0-8.8 7.2-16 16-16s16 7.2 16 16l0 152.2c-9.8-8.1-20.6-15.2-32-21zM368 96c8.8 0 16 7.2 16 16l0 192c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-192c0-8.8 7.2-16 16-16zm112 16l0 192c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-192c0-8.8 7.2-16 16-16s16 7.2 16 16zM176 480a80 80 0 1 0 0-160 80 80 0 1 0 0 160zm0-112a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],bahai:[576,512,["haykal"],"f666","M288 0c14.5 0 27.2 9.7 30.9 23.8l23.9 89.6 75.9-53.3c11.9-8.3 27.8-7.6 39 1.7s14.6 24.9 8.4 38.1l-39.3 84 92.4 8c14.4 1.2 26.2 12 28.8 26.3s-4.9 28.5-18 34.6l-84.1 39.1 65.7 65.5c10.3 10.2 12.4 26.1 5.1 38.7s-22 18.7-36 14.9L391 386.8l8.2 92.4c1.3 14.4-7.3 27.9-20.9 32.9s-28.9 .1-37.2-11.7l-53.1-76-53.1 76c-8.3 11.9-23.6 16.7-37.2 11.7s-22.2-18.5-20.9-32.9l8.2-92.4L95.4 410.9c-14 3.8-28.8-2.3-36-14.9s-5.2-28.4 5.1-38.7l65.7-65.5L46 252.7c-13.1-6.1-20.5-20.3-18-34.6s14.3-25.1 28.8-26.3l92.4-8-39.3-84c-6.1-13.1-2.7-28.8 8.4-38.1s27.1-10 39-1.7l75.9 53.3 23.9-89.6C260.8 9.7 273.5 0 288 0zm0 156.2l-4.8 18c-2.7 10.1-10.2 18.2-20 21.8s-20.8 2.1-29.3-3.9l-15.2-10.7 7.9 16.8c4.4 9.5 4 20.5-1.3 29.6s-14.5 15-25 15.9l-18.5 1.6 16.8 7.8c9.5 4.4 16.2 13.2 18 23.5s-1.5 20.8-8.9 28.2l-13.2 13.1 17.9-4.8c10.1-2.7 20.9-.3 28.9 6.4s12.2 16.9 11.3 27.3l-1.6 18.5 10.6-15.2c6-8.6 15.8-13.7 26.2-13.7s20.2 5.1 26.2 13.7l10.6 15.2-1.6-18.5c-.9-10.4 3.3-20.6 11.3-27.3s18.8-9.1 28.9-6.4l17.9 4.8-13.2-13.1c-7.4-7.4-10.7-17.9-8.9-28.2s8.5-19.1 18-23.5l16.8-7.8-18.5-1.6c-10.4-.9-19.7-6.8-25-15.9s-5.7-20.1-1.3-29.6l7.9-16.8-15.2 10.7c-8.6 6-19.5 7.5-29.3 3.9s-17.3-11.7-20-21.8l-4.8-18z"],"sd-card":[384,512,[],"f7c2","M320 0H141.3C124.3 0 108 6.7 96 18.7L18.7 96C6.7 108 0 124.3 0 141.3V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64zM160 88v48c0 13.3-10.7 24-24 24s-24-10.7-24-24V88c0-13.3 10.7-24 24-24s24 10.7 24 24zm80 0v48c0 13.3-10.7 24-24 24s-24-10.7-24-24V88c0-13.3 10.7-24 24-24s24 10.7 24 24zm80 0v48c0 13.3-10.7 24-24 24s-24-10.7-24-24V88c0-13.3 10.7-24 24-24s24 10.7 24 24z"],dragon:[640,512,[128009],"f6d5","M352 124.5l-51.9-13c-6.5-1.6-11.3-7.1-12-13.8s2.8-13.1 8.7-16.1l40.8-20.4L294.4 28.8c-5.5-4.1-7.8-11.3-5.6-17.9S297.1 0 304 0H416h32 16c30.2 0 58.7 14.2 76.8 38.4l57.6 76.8c6.2 8.3 9.6 18.4 9.6 28.8c0 26.5-21.5 48-48 48H538.5c-17 0-33.3-6.7-45.3-18.7L480 160H448v21.5c0 24.8 12.8 47.9 33.8 61.1l106.6 66.6c32.1 20.1 51.6 55.2 51.6 93.1C640 462.9 590.9 512 530.2 512H496 432 32.3c-3.3 0-6.6-.4-9.6-1.4C13.5 507.8 6 501 2.4 492.1C1 488.7 .2 485.2 0 481.4c-.2-3.7 .3-7.3 1.3-10.7c2.8-9.2 9.6-16.7 18.6-20.4c3-1.2 6.2-2 9.5-2.2L433.3 412c8.3-.7 14.7-7.7 14.7-16.1c0-4.3-1.7-8.4-4.7-11.4l-44.4-44.4c-30-30-46.9-70.7-46.9-113.1V181.5v-57zM512 72.3c0-.1 0-.2 0-.3s0-.2 0-.3v.6zm-1.3 7.4L464.3 68.1c-.2 1.3-.3 2.6-.3 3.9c0 13.3 10.7 24 24 24c10.6 0 19.5-6.8 22.7-16.3zM130.9 116.5c16.3-14.5 40.4-16.2 58.5-4.1l130.6 87V227c0 32.8 8.4 64.8 24 93H112c-6.7 0-12.7-4.2-15-10.4s-.5-13.3 4.6-17.7L171 232.3 18.4 255.8c-7 1.1-13.9-2.6-16.9-9s-1.5-14.1 3.8-18.8L130.9 116.5z"],"shoe-prints":[640,512,[],"f54b","M416 0C352.3 0 256 32 256 32V160c48 0 76 16 104 32s56 32 104 32c56.4 0 176-16 176-96S512 0 416 0zM128 96c0 35.3 28.7 64 64 64h32V32H192c-35.3 0-64 28.7-64 64zM288 512c96 0 224-48 224-128s-119.6-96-176-96c-48 0-76 16-104 32s-56 32-104 32V480s96.3 32 160 32zM0 416c0 35.3 28.7 64 64 64H96V352H64c-35.3 0-64 28.7-64 64z"],"circle-plus":[512,512,["plus-circle"],"f055","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM232 344V280H168c-13.3 0-24-10.7-24-24s10.7-24 24-24h64V168c0-13.3 10.7-24 24-24s24 10.7 24 24v64h64c13.3 0 24 10.7 24 24s-10.7 24-24 24H280v64c0 13.3-10.7 24-24 24s-24-10.7-24-24z"],"face-grin-tongue-wink":[512,512,[128540,"grin-tongue-wink"],"f58b","M174.5 498.8C73.1 464.7 0 368.9 0 256C0 114.6 114.6 0 256 0S512 114.6 512 256c0 112.9-73.1 208.7-174.5 242.8C346.7 484 352 466.6 352 448V401.1c24.3-17.5 43.6-41.6 55.4-69.6c5-11.8-7-22.5-19.3-18.7c-39.7 12.2-84.5 19-131.8 19s-92.1-6.8-131.8-19c-12.3-3.8-24.3 6.9-19.3 18.7c11.7 27.8 30.8 51.7 54.8 69.2V448c0 18.6 5.3 36 14.5 50.8zm20.7-265.2c5.3 7.1 15.3 8.5 22.4 3.2s8.5-15.3 3.2-22.4c-30.4-40.5-91.2-40.5-121.6 0c-5.3 7.1-3.9 17.1 3.2 22.4s17.1 3.9 22.4-3.2c17.6-23.5 52.8-23.5 70.4 0zM336 272a64 64 0 1 0 0-128 64 64 0 1 0 0 128zM320 402.6V448c0 35.3-28.7 64-64 64s-64-28.7-64-64V402.6c0-14.7 11.9-26.6 26.6-26.6h2c11.3 0 21.1 7.9 23.6 18.9c2.8 12.6 20.8 12.6 23.6 0c2.5-11.1 12.3-18.9 23.6-18.9h2c14.7 0 26.6 11.9 26.6 26.6zM336 184a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],"hand-holding":[576,512,[],"f4bd","M559.7 392.2c17.8-13.1 21.6-38.1 8.5-55.9s-38.1-21.6-55.9-8.5L392.6 416H272c-8.8 0-16-7.2-16-16s7.2-16 16-16h16 64c17.7 0 32-14.3 32-32s-14.3-32-32-32H288 272 193.7c-29.1 0-57.3 9.9-80 28L68.8 384H32c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H192 352.5c29 0 57.3-9.3 80.7-26.5l126.6-93.3zm-366.1-8.3a.5 .5 0 1 1 -.9 .2 .5 .5 0 1 1 .9-.2z"],"plug-circle-exclamation":[576,512,[],"e55d","M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8c12.3-2.5 24.1-6.4 35.1-11.5c-2.1-10.8-3.1-21.9-3.1-33.3c0-80.3 53.8-148 127.3-169.2c.5-2.2 .7-4.5 .7-6.8c0-17.7-14.3-32-32-32H32zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z"],"link-slash":[640,512,["chain-broken","chain-slash","unlink"],"f127","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L489.3 358.2l90.5-90.5c56.5-56.5 56.5-148 0-204.5c-50-50-128.8-56.5-186.3-15.4l-1.6 1.1c-14.4 10.3-17.7 30.3-7.4 44.6s30.3 17.7 44.6 7.4l1.6-1.1c32.1-22.9 76-19.3 103.8 8.6c31.5 31.5 31.5 82.5 0 114l-96 96-31.9-25C430.9 239.6 420.1 175.1 377 132c-52.2-52.3-134.5-56.2-191.3-11.7L38.8 5.1zM239 162c30.1-14.9 67.7-9.9 92.8 15.3c20 20 27.5 48.3 21.7 74.5L239 162zM116.6 187.9L60.2 244.3c-56.5 56.5-56.5 148 0 204.5c50 50 128.8 56.5 186.3 15.4l1.6-1.1c14.4-10.3 17.7-30.3 7.4-44.6s-30.3-17.7-44.6-7.4l-1.6 1.1c-32.1 22.9-76 19.3-103.8-8.6C74 372 74 321 105.5 289.5l61.8-61.8-50.6-39.9zM220.9 270c-2.1 39.8 12.2 80.1 42.2 110c38.9 38.9 94.4 51 143.6 36.3L220.9 270z"],clone:[512,512,[],"f24d","M288 448H64V224h64V160H64c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64H288c35.3 0 64-28.7 64-64V384H288v64zm-64-96H448c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H224c-35.3 0-64 28.7-64 64V288c0 35.3 28.7 64 64 64z"],"person-walking-arrow-loop-left":[640,512,[],"e551","M208 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM123.7 200.5c1-.4 1.9-.8 2.9-1.2l-16.9 63.5c-5.6 21.1-.1 43.6 14.7 59.7l70.7 77.1 22 88.1c4.3 17.1 21.7 27.6 38.8 23.3s27.6-21.7 23.3-38.8l-23-92.1c-1.9-7.8-5.8-14.9-11.2-20.8l-49.5-54 19.3-65.5 9.6 23c4.4 10.6 12.5 19.3 22.8 24.5l26.7 13.3c15.8 7.9 35 1.5 42.9-14.3s1.5-35-14.3-42.9L281 232.7l-15.3-36.8C248.5 154.8 208.3 128 163.7 128c-22.8 0-45.3 4.8-66.1 14l-8 3.5c-32.9 14.6-58.1 42.4-69.4 76.5l-2.6 7.8c-5.6 16.8 3.5 34.9 20.2 40.5s34.9-3.5 40.5-20.2l2.6-7.8c5.7-17.1 18.3-30.9 34.7-38.2l8-3.5zm-30 135.1L68.7 398 9.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L116.3 441c4.6-4.6 8.2-10.1 10.6-16.1l14.5-36.2-40.7-44.4c-2.5-2.7-4.8-5.6-7-8.6zm347.7 119c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L461.3 384H480c88.4 0 160-71.6 160-160s-71.6-160-160-160L352 64c-17.7 0-32 14.3-32 32s14.3 32 32 32l128 0c53 0 96 43 96 96s-43 96-96 96H461.3l25.4-25.4c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-80 80c-12.5 12.5-12.5 32.8 0 45.3l80 80z"],"arrow-up-z-a":[576,512,["sort-alpha-up-alt"],"f882","M183.6 42.4C177.5 35.8 169 32 160 32s-17.5 3.8-23.6 10.4l-88 96c-11.9 13-11.1 33.3 2 45.2s33.3 11.1 45.2-2L128 146.3V448c0 17.7 14.3 32 32 32s32-14.3 32-32V146.3l32.4 35.4c11.9 13 32.2 13.9 45.2 2s13.9-32.2 2-45.2l-88-96zM320 64c0 17.7 14.3 32 32 32h50.7l-73.4 73.4c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H429.3l73.4-73.4c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8H352c-17.7 0-32 14.3-32 32zm96 192c-12.1 0-23.2 6.8-28.6 17.7l-64 128-16 32c-7.9 15.8-1.5 35 14.3 42.9s35 1.5 42.9-14.3l7.2-14.3h88.4l7.2 14.3c7.9 15.8 27.1 22.2 42.9 14.3s22.2-27.1 14.3-42.9l-16-32-64-128C439.2 262.8 428.1 256 416 256zM395.8 400L416 359.6 436.2 400H395.8z"],"fire-flame-curved":[384,512,["fire-alt"],"f7e4","M153.6 29.9l16-21.3C173.6 3.2 180 0 186.7 0C198.4 0 208 9.6 208 21.3V43.5c0 13.1 5.4 25.7 14.9 34.7L307.6 159C356.4 205.6 384 270.2 384 337.7C384 434 306 512 209.7 512H192C86 512 0 426 0 320v-3.8c0-48.8 19.4-95.6 53.9-130.1l3.5-3.5c4.2-4.2 10-6.6 16-6.6C85.9 176 96 186.1 96 198.6V288c0 35.3 28.7 64 64 64s64-28.7 64-64v-3.9c0-18-7.2-35.3-19.9-48l-38.6-38.6c-24-24-37.5-56.7-37.5-90.7c0-27.7 9-54.8 25.6-76.9z"],tornado:[448,512,[127786],"f76f","M0 32V45.6C0 62.7 1.7 79.6 5 96H357.8c3.2-6.9 7.5-13.3 13-18.8l38.6-38.6c4.2-4.2 6.6-10 6.6-16C416 10.1 405.9 0 393.4 0H32C14.3 0 0 14.3 0 32zm352.2 96H13.6c12.2 35.9 32.3 68.7 58.8 96H412l-47.2-62.9c-7.3-9.7-11.6-21.2-12.6-33.1zm-226 138.2l116.4 68.5c8.2 4.8 15.8 10.7 22.5 17.3H445c2-9.8 3-19.9 3-30.1c0-23-5.3-45.5-15.3-65.9H110.2c5.2 3.6 10.5 7 16 10.2zM288 384c10.3 21.4 13.8 45.5 9.9 69l-5.9 35.7c-2 12.2 7.4 23.4 19.8 23.4c5.3 0 10.4-2.1 14.2-5.9l78.2-78.2c12.8-12.8 23.1-27.7 30.4-43.9H288z"],"file-circle-plus":[576,512,[58606],"e494","M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v38.6C310.1 219.5 256 287.4 256 368c0 59.1 29.1 111.3 73.7 143.3c-3.2 .5-6.4 .7-9.7 .7H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zm48 96a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm16 80c0-8.8-7.2-16-16-16s-16 7.2-16 16v48H368c-8.8 0-16 7.2-16 16s7.2 16 16 16h48v48c0 8.8 7.2 16 16 16s16-7.2 16-16V384h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H448V304z"],"book-quran":[448,512,["quran"],"f687","M352 0c53 0 96 43 96 96V416c0 53-43 96-96 96H64 32c-17.7 0-32-14.3-32-32s14.3-32 32-32V384c-17.7 0-32-14.3-32-32V32C0 14.3 14.3 0 32 0H64 352zm0 384H96v64H352c17.7 0 32-14.3 32-32s-14.3-32-32-32zM274.1 150.2l-8.9 21.4-23.1 1.9c-5.7 .5-8 7.5-3.7 11.2L256 199.8l-5.4 22.6c-1.3 5.5 4.7 9.9 9.6 6.9L280 217.2l19.8 12.1c4.9 3 10.9-1.4 9.6-6.9L304 199.8l17.6-15.1c4.3-3.7 2-10.8-3.7-11.2l-23.1-1.9-8.9-21.4c-2.2-5.3-9.6-5.3-11.8 0zM96 192c0 70.7 57.3 128 128 128c25.6 0 49.5-7.5 69.5-20.5c3.2-2.1 4.5-6.2 3.1-9.7s-5.2-5.6-9-4.8c-6.1 1.2-12.5 1.9-19 1.9c-52.4 0-94.9-42.5-94.9-94.9s42.5-94.9 94.9-94.9c6.5 0 12.8 .7 19 1.9c3.8 .8 7.5-1.3 9-4.8s.2-7.6-3.1-9.7C273.5 71.5 249.6 64 224 64C153.3 64 96 121.3 96 192z"],anchor:[576,512,[9875],"f13d","M320 96a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm21.1 80C367 158.8 384 129.4 384 96c0-53-43-96-96-96s-96 43-96 96c0 33.4 17 62.8 42.9 80H224c-17.7 0-32 14.3-32 32s14.3 32 32 32h32V448H208c-53 0-96-43-96-96v-6.1l7 7c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L97 263c-9.4-9.4-24.6-9.4-33.9 0L7 319c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l7-7V352c0 88.4 71.6 160 160 160h80 80c88.4 0 160-71.6 160-160v-6.1l7 7c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-56-56c-9.4-9.4-24.6-9.4-33.9 0l-56 56c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l7-7V352c0 53-43 96-96 96H320V240h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H341.1z"],"border-all":[448,512,[],"f84c","M384 96V224H256V96H384zm0 192V416H256V288H384zM192 224H64V96H192V224zM64 288H192V416H64V288zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64z"],"face-angry":[512,512,[128544,"angry"],"f556","M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM338.7 395.9c6.6-5.9 7.1-16 1.2-22.6C323.8 355.4 295.7 336 256 336s-67.8 19.4-83.9 37.3c-5.9 6.6-5.4 16.7 1.2 22.6s16.7 5.4 22.6-1.2c11.7-13 31.6-26.7 60.1-26.7s48.4 13.7 60.1 26.7c5.9 6.6 16 7.1 22.6 1.2zM176.4 272c17.7 0 32-14.3 32-32c0-1.5-.1-3-.3-4.4l10.9 3.6c8.4 2.8 17.4-1.7 20.2-10.1s-1.7-17.4-10.1-20.2l-96-32c-8.4-2.8-17.4 1.7-20.2 10.1s1.7 17.4 10.1 20.2l30.7 10.2c-5.8 5.8-9.3 13.8-9.3 22.6c0 17.7 14.3 32 32 32zm192-32c0-8.9-3.6-17-9.5-22.8l30.2-10.1c8.4-2.8 12.9-11.9 10.1-20.2s-11.9-12.9-20.2-10.1l-96 32c-8.4 2.8-12.9 11.9-10.1 20.2s11.9 12.9 20.2 10.1l11.7-3.9c-.2 1.5-.3 3.1-.3 4.7c0 17.7 14.3 32 32 32s32-14.3 32-32z"],"cookie-bite":[512,512,[],"f564","M257.5 27.6c-.8-5.4-4.9-9.8-10.3-10.6v0c-22.1-3.1-44.6 .9-64.4 11.4l-74 39.5C89.1 78.4 73.2 94.9 63.4 115L26.7 190.6c-9.8 20.1-13 42.9-9.1 64.9l14.5 82.8c3.9 22.1 14.6 42.3 30.7 57.9l60.3 58.4c16.1 15.6 36.6 25.6 58.7 28.7l83 11.7c22.1 3.1 44.6-.9 64.4-11.4l74-39.5c19.7-10.5 35.6-27 45.4-47.2l36.7-75.5c9.8-20.1 13-42.9 9.1-64.9v0c-.9-5.3-5.3-9.3-10.6-10.1c-51.5-8.2-92.8-47.1-104.5-97.4c-1.8-7.6-8-13.4-15.7-14.6c-54.6-8.7-97.7-52-106.2-106.8zM208 144a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM144 336a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm224-64a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"arrow-trend-down":[576,512,[],"e097","M384 352c-17.7 0-32 14.3-32 32s14.3 32 32 32H544c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32s-32 14.3-32 32v82.7L342.6 137.4c-12.5-12.5-32.8-12.5-45.3 0L192 242.7 54.6 105.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0L320 205.3 466.7 352H384z"],rss:[448,512,["feed"],"f09e","M0 64C0 46.3 14.3 32 32 32c229.8 0 416 186.2 416 416c0 17.7-14.3 32-32 32s-32-14.3-32-32C384 253.6 226.4 96 32 96C14.3 96 0 81.7 0 64zM0 416a64 64 0 1 1 128 0A64 64 0 1 1 0 416zM32 160c159.1 0 288 128.9 288 288c0 17.7-14.3 32-32 32s-32-14.3-32-32c0-123.7-100.3-224-224-224c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],"draw-polygon":[448,512,[],"f5ee","M96 151.4V360.6c9.7 5.6 17.8 13.7 23.4 23.4H328.6c0-.1 .1-.2 .1-.3l-4.5-7.9-32-56 0 0c-1.4 .1-2.8 .1-4.2 .1c-35.3 0-64-28.7-64-64s28.7-64 64-64c1.4 0 2.8 0 4.2 .1l0 0 32-56 4.5-7.9-.1-.3H119.4c-5.6 9.7-13.7 17.8-23.4 23.4zM384.3 352c35.2 .2 63.7 28.7 63.7 64c0 35.3-28.7 64-64 64c-23.7 0-44.4-12.9-55.4-32H119.4c-11.1 19.1-31.7 32-55.4 32c-35.3 0-64-28.7-64-64c0-23.7 12.9-44.4 32-55.4V151.4C12.9 140.4 0 119.7 0 96C0 60.7 28.7 32 64 32c23.7 0 44.4 12.9 55.4 32H328.6c11.1-19.1 31.7-32 55.4-32c35.3 0 64 28.7 64 64c0 35.3-28.5 63.8-63.7 64l-4.5 7.9-32 56-2.3 4c4.2 8.5 6.5 18 6.5 28.1s-2.3 19.6-6.5 28.1l2.3 4 32 56 4.5 7.9z"],"scale-balanced":[640,512,[9878,"balance-scale"],"f24e","M384 32H512c17.7 0 32 14.3 32 32s-14.3 32-32 32H398.4c-5.2 25.8-22.9 47.1-46.4 57.3V448H512c17.7 0 32 14.3 32 32s-14.3 32-32 32H320 128c-17.7 0-32-14.3-32-32s14.3-32 32-32H288V153.3c-23.5-10.3-41.2-31.6-46.4-57.3H128c-17.7 0-32-14.3-32-32s14.3-32 32-32H256c14.6-19.4 37.8-32 64-32s49.4 12.6 64 32zm55.6 288H584.4L512 195.8 439.6 320zM512 416c-62.9 0-115.2-34-126-78.9c-2.6-11 1-22.3 6.7-32.1l95.2-163.2c5-8.6 14.2-13.8 24.1-13.8s19.1 5.3 24.1 13.8l95.2 163.2c5.7 9.8 9.3 21.1 6.7 32.1C627.2 382 574.9 416 512 416zM126.8 195.8L54.4 320H199.3L126.8 195.8zM.9 337.1c-2.6-11 1-22.3 6.7-32.1l95.2-163.2c5-8.6 14.2-13.8 24.1-13.8s19.1 5.3 24.1 13.8l95.2 163.2c5.7 9.8 9.3 21.1 6.7 32.1C242 382 189.7 416 126.8 416S11.7 382 .9 337.1z"],"gauge-simple-high":[512,512,[61668,"tachometer","tachometer-fast"],"f62a","M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm320 96c0-15.9-5.8-30.4-15.3-41.6l76.6-147.4c6.1-11.8 1.5-26.3-10.2-32.4s-26.2-1.5-32.4 10.2L262.1 288.3c-2-.2-4-.3-6.1-.3c-35.3 0-64 28.7-64 64s28.7 64 64 64s64-28.7 64-64z"],shower:[512,512,[128703],"f2cc","M64 131.9C64 112.1 80.1 96 99.9 96c9.5 0 18.6 3.8 25.4 10.5l16.2 16.2c-21 38.9-17.4 87.5 10.9 123L151 247c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0L345 121c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-1.3 1.3c-35.5-28.3-84.2-31.9-123-10.9L170.5 61.3C151.8 42.5 126.4 32 99.9 32C44.7 32 0 76.7 0 131.9V448c0 17.7 14.3 32 32 32s32-14.3 32-32V131.9zM256 352a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm64 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm0-128a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm64 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm0-128a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm64 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm32-32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],desktop:[576,512,[128421,61704,"desktop-alt"],"f390","M64 0C28.7 0 0 28.7 0 64V352c0 35.3 28.7 64 64 64H240l-10.7 32H160c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H346.7L336 416H512c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zM512 64V288H64V64H512z"],m:[448,512,[109],"4d","M22.7 33.4c13.5-4.1 28.1 1.1 35.9 12.9L224 294.3 389.4 46.2c7.8-11.7 22.4-17 35.9-12.9S448 49.9 448 64V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V169.7L250.6 369.8c-5.9 8.9-15.9 14.2-26.6 14.2s-20.7-5.3-26.6-14.2L64 169.7V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V64C0 49.9 9.2 37.5 22.7 33.4z"],"table-list":[512,512,["th-list"],"f00b","M0 96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm64 0v64h64V96H64zm384 0H192v64H448V96zM64 224v64h64V224H64zm384 0H192v64H448V224zM64 352v64h64V352H64zm384 0H192v64H448V352z"],"comment-sms":[512,512,["sms"],"f7cd","M256 448c141.4 0 256-93.1 256-208S397.4 32 256 32S0 125.1 0 240c0 45.1 17.7 86.8 47.7 120.9c-1.9 24.5-11.4 46.3-21.4 62.9c-5.5 9.2-11.1 16.6-15.2 21.6c-2.1 2.5-3.7 4.4-4.9 5.7c-.6 .6-1 1.1-1.3 1.4l-.3 .3 0 0 0 0 0 0 0 0c-4.6 4.6-5.9 11.4-3.4 17.4c2.5 6 8.3 9.9 14.8 9.9c28.7 0 57.6-8.9 81.6-19.3c22.9-10 42.4-21.9 54.3-30.6c31.8 11.5 67 17.9 104.1 17.9zM96 212.8c0-20.3 16.5-36.8 36.8-36.8H152c8.8 0 16 7.2 16 16s-7.2 16-16 16H132.8c-2.7 0-4.8 2.2-4.8 4.8c0 1.6 .8 3.1 2.2 4l29.4 19.6c10.3 6.8 16.4 18.3 16.4 30.7c0 20.3-16.5 36.8-36.8 36.8H112c-8.8 0-16-7.2-16-16s7.2-16 16-16h27.2c2.7 0 4.8-2.2 4.8-4.8c0-1.6-.8-3.1-2.2-4l-29.4-19.6C102.2 236.7 96 225.2 96 212.8zM372.8 176H392c8.8 0 16 7.2 16 16s-7.2 16-16 16H372.8c-2.7 0-4.8 2.2-4.8 4.8c0 1.6 .8 3.1 2.2 4l29.4 19.6c10.2 6.8 16.4 18.3 16.4 30.7c0 20.3-16.5 36.8-36.8 36.8H352c-8.8 0-16-7.2-16-16s7.2-16 16-16h27.2c2.7 0 4.8-2.2 4.8-4.8c0-1.6-.8-3.1-2.2-4l-29.4-19.6c-10.2-6.8-16.4-18.3-16.4-30.7c0-20.3 16.5-36.8 36.8-36.8zm-152 6.4L256 229.3l35.2-46.9c4.1-5.5 11.3-7.8 17.9-5.6s10.9 8.3 10.9 15.2v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V240l-19.2 25.6c-3 4-7.8 6.4-12.8 6.4s-9.8-2.4-12.8-6.4L224 240v48c0 8.8-7.2 16-16 16s-16-7.2-16-16V192c0-6.9 4.4-13 10.9-15.2s13.7 .1 17.9 5.6z"],book:[448,512,[128212],"f02d","M96 0C43 0 0 43 0 96V416c0 53 43 96 96 96H384h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V384c17.7 0 32-14.3 32-32V32c0-17.7-14.3-32-32-32H384 96zm0 384H352v64H96c-17.7 0-32-14.3-32-32s14.3-32 32-32zm32-240c0-8.8 7.2-16 16-16H336c8.8 0 16 7.2 16 16s-7.2 16-16 16H144c-8.8 0-16-7.2-16-16zm16 48H336c8.8 0 16 7.2 16 16s-7.2 16-16 16H144c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],"user-plus":[640,512,[],"f234","M96 128a128 128 0 1 1 256 0A128 128 0 1 1 96 128zM0 482.3C0 383.8 79.8 304 178.3 304h91.4C368.2 304 448 383.8 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3zM504 312V248H440c-13.3 0-24-10.7-24-24s10.7-24 24-24h64V136c0-13.3 10.7-24 24-24s24 10.7 24 24v64h64c13.3 0 24 10.7 24 24s-10.7 24-24 24H552v64c0 13.3-10.7 24-24 24s-24-10.7-24-24z"],check:[448,512,[10003,10004],"f00c","M438.6 105.4c12.5 12.5 12.5 32.8 0 45.3l-256 256c-12.5 12.5-32.8 12.5-45.3 0l-128-128c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0L160 338.7 393.4 105.4c12.5-12.5 32.8-12.5 45.3 0z"],"battery-three-quarters":[576,512,["battery-4"],"f241","M464 160c8.8 0 16 7.2 16 16V336c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16H464zM80 96C35.8 96 0 131.8 0 176V336c0 44.2 35.8 80 80 80H464c44.2 0 80-35.8 80-80V320c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32V176c0-44.2-35.8-80-80-80H80zm272 96H96V320H352V192z"],"house-circle-check":[640,512,[],"e509","M320.7 352c8.1-89.7 83.5-160 175.3-160c8.9 0 17.6 .7 26.1 1.9L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32v69.7c-.1 .9-.1 1.8-.1 2.8V472c0 22.1 17.9 40 40 40h16c1.2 0 2.4-.1 3.6-.2c1.5 .1 3 .2 4.5 .2H160h24c22.1 0 40-17.9 40-40V448 384c0-17.7 14.3-32 32-32h64l.7 0zM640 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z"],"angle-left":[320,512,[8249],"f104","M41.4 233.4c-12.5 12.5-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L109.3 256 246.6 118.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-160 160z"],"diagram-successor":[512,512,[],"e47a","M512 416l0-64c0-35.3-28.7-64-64-64L64 288c-35.3 0-64 28.7-64 64l0 64c0 35.3 28.7 64 64 64l384 0c35.3 0 64-28.7 64-64zM64 160l0-64 144 0 16 0 0 64L64 160zm224 0l0-64 80 0c8.8 0 16 7.2 16 16l0 16-38.1 0c-21.4 0-32.1 25.9-17 41L399 239c9.4 9.4 24.6 9.4 33.9 0L503 169c15.1-15.1 4.4-41-17-41L448 128l0-16c0-44.2-35.8-80-80-80L224 32l-16 0L64 32C28.7 32 0 60.7 0 96l0 64c0 35.3 28.7 64 64 64l160 0c35.3 0 64-28.7 64-64z"],"truck-arrow-right":[640,512,[],"e58b","M0 48C0 21.5 21.5 0 48 0H368c26.5 0 48 21.5 48 48V96h50.7c17 0 33.3 6.7 45.3 18.7L589.3 192c12 12 18.7 28.3 18.7 45.3V256v32 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H576c0 53-43 96-96 96s-96-43-96-96H256c0 53-43 96-96 96s-96-43-96-96H48c-26.5 0-48-21.5-48-48V48zM416 256H544V237.3L466.7 160H416v96zM160 464a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm368-48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM257 95c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l39 39H96c-13.3 0-24 10.7-24 24s10.7 24 24 24H262.1l-39 39c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l80-80c9.4-9.4 9.4-24.6 0-33.9L257 95z"],"arrows-split-up-and-left":[512,512,[],"e4bc","M246.6 150.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l96-96c12.5-12.5 32.8-12.5 45.3 0l96 96c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L352 109.3V384c0 35.3 28.7 64 64 64h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H416c-70.7 0-128-57.3-128-128c0-35.3-28.7-64-64-64H109.3l41.4 41.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0l-96-96c-12.5-12.5-12.5-32.8 0-45.3l96-96c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3L109.3 256H224c23.3 0 45.2 6.2 64 17.1V109.3l-41.4 41.4z"],"hand-fist":[448,512,[9994,"fist-raised"],"f6de","M192 0c17.7 0 32 14.3 32 32V144H160V32c0-17.7 14.3-32 32-32zM64 64c0-17.7 14.3-32 32-32s32 14.3 32 32v80H64V64zm192 0c0-17.7 14.3-32 32-32s32 14.3 32 32v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V64zm96 64c0-17.7 14.3-32 32-32s32 14.3 32 32v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V128zm-96 88l0-.6c9.4 5.4 20.3 8.6 32 8.6c13.2 0 25.4-4 35.6-10.8c8.7 24.9 32.5 42.8 60.4 42.8c11.7 0 22.6-3.1 32-8.6V256c0 52.3-25.1 98.8-64 128v96c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32V401.6c-17.3-7.9-33.2-18.8-46.9-32.5L69.5 357.5C45.5 333.5 32 300.9 32 267V240c0-35.3 28.7-64 64-64h88c22.1 0 40 17.9 40 40s-17.9 40-40 40H128c-8.8 0-16 7.2-16 16s7.2 16 16 16h56c39.8 0 72-32.2 72-72z"],"cloud-moon":[640,512,[],"f6c3","M495.8 0c5.5 0 10.9 .2 16.3 .7c7 .6 12.8 5.7 14.3 12.5s-1.6 13.9-7.7 17.3c-44.4 25.2-74.4 73-74.4 127.8c0 81 65.5 146.6 146.2 146.6c8.6 0 17-.7 25.1-2.1c6.9-1.2 13.8 2.2 17 8.5s1.9 13.8-3.1 18.7c-34.5 33.6-81.7 54.4-133.6 54.4c-9.3 0-18.4-.7-27.4-1.9c-11.2-22.6-29.8-40.9-52.6-51.7c-2.7-58.5-50.3-105.3-109.2-106.7c-1.7-10.4-2.6-21-2.6-31.8C304 86.1 389.8 0 495.8 0zM447.9 431.9c0 44.2-35.8 80-80 80H96c-53 0-96-43-96-96c0-47.6 34.6-87 80-94.6l0-1.3c0-53 43-96 96-96c34.9 0 65.4 18.6 82.2 46.4c13-9.1 28.8-14.4 45.8-14.4c44.2 0 80 35.8 80 80c0 5.9-.6 11.7-1.9 17.2c37.4 6.7 65.8 39.4 65.8 78.7z"],briefcase:[512,512,[128188],"f0b1","M184 48H328c4.4 0 8 3.6 8 8V96H176V56c0-4.4 3.6-8 8-8zm-56 8V96H64C28.7 96 0 124.7 0 160v96H192 320 512V160c0-35.3-28.7-64-64-64H384V56c0-30.9-25.1-56-56-56H184c-30.9 0-56 25.1-56 56zM512 288H320v32c0 17.7-14.3 32-32 32H224c-17.7 0-32-14.3-32-32V288H0V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V288z"],"person-falling":[512,512,[],"e546","M288 0c17.7 0 32 14.3 32 32l0 9.8c0 54.6-27.9 104.6-72.5 133.6l.2 .3L304.5 256l87.5 0c15.1 0 29.3 7.1 38.4 19.2l43.2 57.6c10.6 14.1 7.7 34.2-6.4 44.8s-34.2 7.7-44.8-6.4L384 320l-96 0h-1.4l92.3 142.6c9.6 14.8 5.4 34.6-9.5 44.3s-34.6 5.4-44.3-9.5L164.5 249.2c-2.9 9.2-4.5 19-4.5 29l0 73.8c0 17.7-14.3 32-32 32s-32-14.3-32-32V278.2c0-65.1 39.6-123.7 100.1-147.9C232.3 115.8 256 80.8 256 41.8l0-9.8c0-17.7 14.3-32 32-32zM112 32a48 48 0 1 1 0 96 48 48 0 1 1 0-96z"],"image-portrait":[384,512,["portrait"],"f3e0","M384 64c0-35.3-28.7-64-64-64H64C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64l0-384zM128 192a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM80 356.6c0-37.9 30.7-68.6 68.6-68.6h86.9c37.9 0 68.6 30.7 68.6 68.6c0 15.1-12.3 27.4-27.4 27.4H107.4C92.3 384 80 371.7 80 356.6z"],"user-tag":[640,512,[],"f507","M224 256A128 128 0 1 0 224 0a128 128 0 1 0 0 256zm-45.7 48C79.8 304 0 383.8 0 482.3C0 498.7 13.3 512 29.7 512H418.3c10 0 18.8-4.9 24.2-12.5l-99.2-99.2c-14.9-14.9-23.3-35.1-23.3-56.1v-33c-15.9-4.7-32.8-7.2-50.3-7.2H178.3zM384 224c-17.7 0-32 14.3-32 32v82.7c0 17 6.7 33.3 18.7 45.3L478.1 491.3c18.7 18.7 49.1 18.7 67.9 0l73.4-73.4c18.7-18.7 18.7-49.1 0-67.9L512 242.7c-12-12-28.3-18.7-45.3-18.7H384zm24 80a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z"],rug:[640,512,[],"e569","M24 64H56 80V88v88 80 80 88 24H56 24c-13.3 0-24-10.7-24-24s10.7-24 24-24h8V360H24c-13.3 0-24-10.7-24-24s10.7-24 24-24h8V280H24c-13.3 0-24-10.7-24-24s10.7-24 24-24h8V200H24c-13.3 0-24-10.7-24-24s10.7-24 24-24h8V112H24C10.7 112 0 101.3 0 88S10.7 64 24 64zm88 0H528V448H112V64zM640 88c0 13.3-10.7 24-24 24h-8v40h8c13.3 0 24 10.7 24 24s-10.7 24-24 24h-8v32h8c13.3 0 24 10.7 24 24s-10.7 24-24 24h-8v32h8c13.3 0 24 10.7 24 24s-10.7 24-24 24h-8v40h8c13.3 0 24 10.7 24 24s-10.7 24-24 24H584 560V424 336 256 176 88 64h24 32c13.3 0 24 10.7 24 24z"],"earth-europe":[512,512,["globe-europe"],"f7a2","M266.3 48.3L232.5 73.6c-5.4 4-8.5 10.4-8.5 17.1v9.1c0 6.8 5.5 12.3 12.3 12.3c2.4 0 4.8-.7 6.8-2.1l41.8-27.9c2-1.3 4.4-2.1 6.8-2.1h1c6.2 0 11.3 5.1 11.3 11.3c0 3-1.2 5.9-3.3 8l-19.9 19.9c-5.8 5.8-12.9 10.2-20.7 12.8l-26.5 8.8c-5.8 1.9-9.6 7.3-9.6 13.4c0 3.7-1.5 7.3-4.1 10l-17.9 17.9c-6.4 6.4-9.9 15-9.9 24v4.3c0 16.4 13.6 29.7 29.9 29.7c11 0 21.2-6.2 26.1-16l4-8.1c2.4-4.8 7.4-7.9 12.8-7.9c4.5 0 8.7 2.1 11.4 5.7l16.3 21.7c2.1 2.9 5.5 4.5 9.1 4.5c8.4 0 13.9-8.9 10.1-16.4l-1.1-2.3c-3.5-7 0-15.5 7.5-18l21.2-7.1c7.6-2.5 12.7-9.6 12.7-17.6c0-10.3 8.3-18.6 18.6-18.6H400c8.8 0 16 7.2 16 16s-7.2 16-16 16H379.3c-7.2 0-14.2 2.9-19.3 8l-4.7 4.7c-2.1 2.1-3.3 5-3.3 8c0 6.2 5.1 11.3 11.3 11.3h11.3c6 0 11.8 2.4 16 6.6l6.5 6.5c1.8 1.8 2.8 4.3 2.8 6.8s-1 5-2.8 6.8l-7.5 7.5C386 262 384 266.9 384 272s2 10 5.7 13.7L408 304c10.2 10.2 24.1 16 38.6 16H454c6.5-20.2 10-41.7 10-64c0-111.4-87.6-202.4-197.7-207.7zm172 307.9c-3.7-2.6-8.2-4.1-13-4.1c-6 0-11.8-2.4-16-6.6L396 332c-7.7-7.7-18-12-28.9-12c-9.7 0-19.2-3.5-26.6-9.8L314 287.4c-11.6-9.9-26.4-15.4-41.7-15.4H251.4c-12.6 0-25 3.7-35.5 10.7L188.5 301c-17.8 11.9-28.5 31.9-28.5 53.3v3.2c0 17 6.7 33.3 18.7 45.3l16 16c8.5 8.5 20 13.3 32 13.3H248c13.3 0 24 10.7 24 24c0 2.5 .4 5 1.1 7.3c71.3-5.8 132.5-47.6 165.2-107.2zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM187.3 100.7c-6.2-6.2-16.4-6.2-22.6 0l-32 32c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0l32-32c6.2-6.2 6.2-16.4 0-22.6z"],"cart-flatbed-suitcase":[640,512,["luggage-cart"],"f59d","M0 32C0 14.3 14.3 0 32 0H48c44.2 0 80 35.8 80 80V368c0 8.8 7.2 16 16 16H608c17.7 0 32 14.3 32 32s-14.3 32-32 32H541.3c1.8 5 2.7 10.4 2.7 16c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-5.6 1-11 2.7-16H253.3c1.8 5 2.7 10.4 2.7 16c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-5.6 1-11 2.7-16H144c-44.2 0-80-35.8-80-80V80c0-8.8-7.2-16-16-16H32C14.3 64 0 49.7 0 32zM432 96V56c0-4.4-3.6-8-8-8H344c-4.4 0-8 3.6-8 8V96h96zM288 96V56c0-30.9 25.1-56 56-56h80c30.9 0 56 25.1 56 56V96 320H288V96zM512 320V96h16c26.5 0 48 21.5 48 48V272c0 26.5-21.5 48-48 48H512zM240 96h16V320H240c-26.5 0-48-21.5-48-48V144c0-26.5 21.5-48 48-48z"],"rectangle-xmark":[512,512,[62164,"rectangle-times","times-rectangle","window-close"],"f410","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM175 175c9.4-9.4 24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9z"],"baht-sign":[320,512,[],"e0ac","M144 0c-17.7 0-32 14.3-32 32V64H37.6C16.8 64 0 80.8 0 101.6V224v41.7V288 406.3c0 23 18.7 41.7 41.7 41.7H112v32c0 17.7 14.3 32 32 32s32-14.3 32-32V448h32c61.9 0 112-50.1 112-112c0-40.1-21.1-75.3-52.7-95.1C280.3 222.6 288 200.2 288 176c0-61.9-50.1-112-112-112V32c0-17.7-14.3-32-32-32zM112 128v96H64V128h48zm64 96V128c26.5 0 48 21.5 48 48s-21.5 48-48 48zm-64 64v96H64V288h48zm64 96V288h32c26.5 0 48 21.5 48 48s-21.5 48-48 48H176z"],"book-open":[576,512,[128214,128366],"f518","M249.6 471.5c10.8 3.8 22.4-4.1 22.4-15.5V78.6c0-4.2-1.6-8.4-5-11C247.4 52 202.4 32 144 32C93.5 32 46.3 45.3 18.1 56.1C6.8 60.5 0 71.7 0 83.8V454.1c0 11.9 12.8 20.2 24.1 16.5C55.6 460.1 105.5 448 144 448c33.9 0 79 14 105.6 23.5zm76.8 0C353 462 398.1 448 432 448c38.5 0 88.4 12.1 119.9 22.6c11.3 3.8 24.1-4.6 24.1-16.5V83.8c0-12.1-6.8-23.3-18.1-27.6C529.7 45.3 482.5 32 432 32c-58.4 0-103.4 20-123 35.6c-3.3 2.6-5 6.8-5 11V456c0 11.4 11.7 19.3 22.4 15.5z"],"book-journal-whills":[448,512,["journal-whills"],"f66a","M0 96C0 43 43 0 96 0H384h32c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384 96c-53 0-96-43-96-96V96zM64 416c0 17.7 14.3 32 32 32H352V384H96c-17.7 0-32 14.3-32 32zm90.4-234.4l-21.2-21.2c-3 10.1-5.1 20.6-5.1 31.6c0 .2 0 .5 .1 .8s.1 .5 .1 .8L165.2 226c2.5 2.1 3.4 5.8 2.3 8.9c-1.3 3-4.1 5.1-7.5 5.1c-1.9-.1-3.8-.8-5.2-2l-23.6-20.6C142.8 267 186.9 304 240 304s97.3-37 108.9-86.6L325.3 238c-1.4 1.2-3.3 2-5.3 2c-2.2-.1-4.4-1.1-6-2.8c-1.2-1.5-1.9-3.4-2-5.2c.1-2.2 1.1-4.4 2.8-6l37.1-32.5c0-.3 0-.5 .1-.8s.1-.5 .1-.8c0-11-2.1-21.5-5.1-31.6l-21.2 21.2c-3.1 3.1-8.1 3.1-11.3 0s-3.1-8.1 0-11.2l26.4-26.5c-8.2-17-20.5-31.7-35.9-42.6c-2.7-1.9-6.2 1.4-5 4.5c8.5 22.4 3.6 48-13 65.6c-3.2 3.4-3.6 8.9-.9 12.7c9.8 14 12.7 31.9 7.5 48.5c-5.9 19.4-22 34.1-41.9 38.3l-1.4-34.3 12.6 8.6c.6 .4 1.5 .6 2.3 .6c1.5 0 2.7-.8 3.5-2s.6-2.8-.1-4L260 225.4l18-3.6c1.8-.4 3.1-2.1 3.1-4s-1.4-3.5-3.1-3.9l-18-3.7 8.5-14.3c.8-1.2 .9-2.9 .1-4.1s-2-2-3.5-2l-.1 0c-.7 .1-1.5 .3-2.1 .7l-14.1 9.6L244 87.9c-.1-2.2-1.9-3.9-4-3.9s-3.9 1.6-4 3.9l-4.6 110.8-12-8.1c-1.5-1.1-3.6-.9-5 .4s-1.6 3.4-.8 5l8.6 14.3-18 3.7c-1.8 .4-3.1 2-3.1 3.9s1.4 3.6 3.1 4l18 3.8-8.6 14.2c-.2 .6-.5 1.4-.5 2c0 1.1 .5 2.1 1.2 3c.8 .6 1.8 1 2.8 1c.7 0 1.6-.2 2.2-.6l10.4-7.1-1.4 32.8c-19.9-4.1-36-18.9-41.9-38.3c-5.1-16.6-2.2-34.4 7.6-48.5c2.7-3.9 2.3-9.3-.9-12.7c-16.6-17.5-21.6-43.1-13.1-65.5c1.2-3.1-2.3-6.4-5-4.5c-15.3 10.9-27.6 25.6-35.8 42.6l26.4 26.5c3.1 3.1 3.1 8.1 0 11.2s-8.1 3.1-11.2 0z"],handcuffs:[640,512,[],"e4f8","M240 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM192 48a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm-32 80c17.7 0 32 14.3 32 32h8c13.3 0 24 10.7 24 24v16c0 1.7-.2 3.4-.5 5.1C280.3 229.6 320 286.2 320 352c0 88.4-71.6 160-160 160S0 440.4 0 352c0-65.8 39.7-122.4 96.5-146.9c-.4-1.6-.5-3.3-.5-5.1V184c0-13.3 10.7-24 24-24h8c0-17.7 14.3-32 32-32zm0 320a96 96 0 1 0 0-192 96 96 0 1 0 0 192zm192-96c0-25.9-5.1-50.5-14.4-73.1c16.9-32.9 44.8-59.1 78.9-73.9c-.4-1.6-.5-3.3-.5-5.1V184c0-13.3 10.7-24 24-24h8c0-17.7 14.3-32 32-32s32 14.3 32 32h8c13.3 0 24 10.7 24 24v16c0 1.7-.2 3.4-.5 5.1C600.3 229.6 640 286.2 640 352c0 88.4-71.6 160-160 160c-62 0-115.8-35.3-142.4-86.9c9.3-22.5 14.4-47.2 14.4-73.1zm224 0a96 96 0 1 0 -192 0 96 96 0 1 0 192 0zM368 0a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm80 48a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"triangle-exclamation":[512,512,[9888,"exclamation-triangle","warning"],"f071","M256 32c14.2 0 27.3 7.5 34.5 19.8l216 368c7.3 12.4 7.3 27.7 .2 40.1S486.3 480 472 480H40c-14.3 0-27.6-7.7-34.7-20.1s-7-27.8 .2-40.1l216-368C228.7 39.5 241.8 32 256 32zm0 128c-13.3 0-24 10.7-24 24V296c0 13.3 10.7 24 24 24s24-10.7 24-24V184c0-13.3-10.7-24-24-24zm32 224a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z"],database:[448,512,[],"f1c0","M448 80v48c0 44.2-100.3 80-224 80S0 172.2 0 128V80C0 35.8 100.3 0 224 0S448 35.8 448 80zM393.2 214.7c20.8-7.4 39.9-16.9 54.8-28.6V288c0 44.2-100.3 80-224 80S0 332.2 0 288V186.1c14.9 11.8 34 21.2 54.8 28.6C99.7 230.7 159.5 240 224 240s124.3-9.3 169.2-25.3zM0 346.1c14.9 11.8 34 21.2 54.8 28.6C99.7 390.7 159.5 400 224 400s124.3-9.3 169.2-25.3c20.8-7.4 39.9-16.9 54.8-28.6V432c0 44.2-100.3 80-224 80S0 476.2 0 432V346.1z"],share:[512,512,["mail-forward"],"f064","M307 34.8c-11.5 5.1-19 16.6-19 29.2v64H176C78.8 128 0 206.8 0 304C0 417.3 81.5 467.9 100.2 478.1c2.5 1.4 5.3 1.9 8.1 1.9c10.9 0 19.7-8.9 19.7-19.7c0-7.5-4.3-14.4-9.8-19.5C108.8 431.9 96 414.4 96 384c0-53 43-96 96-96h96v64c0 12.6 7.4 24.1 19 29.2s25 3 34.4-5.4l160-144c6.7-6.1 10.6-14.7 10.6-23.8s-3.8-17.7-10.6-23.8l-160-144c-9.4-8.5-22.9-10.6-34.4-5.4z"],"bottle-droplet":[320,512,[],"e4c4","M96 0C82.7 0 72 10.7 72 24s10.7 24 24 24c4.4 0 8 3.6 8 8v64.9c0 12.2-7.2 23.1-17.2 30.1C53.7 174.1 32 212.5 32 256V448c0 35.3 28.7 64 64 64H224c35.3 0 64-28.7 64-64V256c0-43.5-21.7-81.9-54.8-105c-10-7-17.2-17.9-17.2-30.1V56c0-4.4 3.6-8 8-8c13.3 0 24-10.7 24-24s-10.7-24-24-24l-8 0 0 0 0 0H104l0 0 0 0L96 0zm64 382c-26.5 0-48-20.1-48-45c0-16.8 22.1-48.1 36.3-66.4c6-7.8 17.5-7.8 23.5 0C185.9 288.9 208 320.2 208 337c0 24.9-21.5 45-48 45z"],"mask-face":[640,512,[],"e1d7","M320 64c-27.2 0-53.8 8-76.4 23.1l-37.1 24.8c-15.8 10.5-34.3 16.1-53.3 16.1H144 128 56c-30.9 0-56 25.1-56 56v85c0 55.1 37.5 103.1 90.9 116.4l108 27C233.8 435 275.4 448 320 448s86.2-13 121.1-35.5l108-27C602.5 372.1 640 324.1 640 269V184c0-30.9-25.1-56-56-56H512 496h-9.2c-19 0-37.5-5.6-53.3-16.1L396.4 87.1C373.8 72 347.2 64 320 64zM132.3 346.3l-29.8-7.4C70.5 330.9 48 302.1 48 269V184c0-4.4 3.6-8 8-8H96v48c0 45.1 13.4 87.2 36.3 122.3zm405.1-7.4l-29.8 7.4c23-35.2 36.3-77.2 36.3-122.3V176h40c4.4 0 8 3.6 8 8v85c0 33-22.5 61.8-54.5 69.9zM192 208c0-8.8 7.2-16 16-16H432c8.8 0 16 7.2 16 16s-7.2 16-16 16H208c-8.8 0-16-7.2-16-16zm16 48H432c8.8 0 16 7.2 16 16s-7.2 16-16 16H208c-8.8 0-16-7.2-16-16s7.2-16 16-16zm16 80c0-8.8 7.2-16 16-16H400c8.8 0 16 7.2 16 16s-7.2 16-16 16H240c-8.8 0-16-7.2-16-16z"],"hill-rockslide":[576,512,[],"e508","M252.4 103.8l27 48c2.8 5 8.2 8.2 13.9 8.2l53.3 0c5.8 0 11.1-3.1 13.9-8.2l27-48c2.7-4.9 2.7-10.8 0-15.7l-27-48c-2.8-5-8.2-8.2-13.9-8.2H293.4c-5.8 0-11.1 3.1-13.9 8.2l-27 48c-2.7 4.9-2.7 10.8 0 15.7zM68.3 87C43.1 61.8 0 79.7 0 115.3V432c0 44.2 35.8 80 80 80H396.7c35.6 0 53.5-43.1 28.3-68.3L68.3 87zM504.2 403.6c4.9 2.7 10.8 2.7 15.7 0l48-27c5-2.8 8.2-8.2 8.2-13.9V309.4c0-5.8-3.1-11.1-8.2-13.9l-48-27c-4.9-2.7-10.8-2.7-15.7 0l-48 27c-5 2.8-8.2 8.2-8.2 13.9v53.3c0 5.8 3.1 11.1 8.2 13.9l48 27zM192 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM384 288a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"right-left":[512,512,["exchange-alt"],"f362","M32 96l320 0V32c0-12.9 7.8-24.6 19.8-29.6s25.7-2.2 34.9 6.9l96 96c6 6 9.4 14.1 9.4 22.6s-3.4 16.6-9.4 22.6l-96 96c-9.2 9.2-22.9 11.9-34.9 6.9s-19.8-16.6-19.8-29.6V160L32 160c-17.7 0-32-14.3-32-32s14.3-32 32-32zM480 352c17.7 0 32 14.3 32 32s-14.3 32-32 32H160v64c0 12.9-7.8 24.6-19.8 29.6s-25.7 2.2-34.9-6.9l-96-96c-6-6-9.4-14.1-9.4-22.6s3.4-16.6 9.4-22.6l96-96c9.2-9.2 22.9-11.9 34.9-6.9s19.8 16.6 19.8 29.6l0 64H480z"],"paper-plane":[512,512,[61913],"f1d8","M498.1 5.6c10.1 7 15.4 19.1 13.5 31.2l-64 416c-1.5 9.7-7.4 18.2-16 23s-18.9 5.4-28 1.6L284 427.7l-68.5 74.1c-8.9 9.7-22.9 12.9-35.2 8.1S160 493.2 160 480V396.4c0-4 1.5-7.8 4.2-10.7L331.8 202.8c5.8-6.3 5.6-16-.4-22s-15.7-6.4-22-.7L106 360.8 17.7 316.6C7.1 311.3 .3 300.7 0 288.9s5.9-22.8 16.1-28.7l448-256c10.7-6.1 23.9-5.5 34 1.4z"],"road-circle-exclamation":[640,512,[],"e565","M213.2 32H288V96c0 17.7 14.3 32 32 32s32-14.3 32-32V32h74.8c27.1 0 51.3 17.1 60.3 42.6l42.7 120.6c-10.9-2.1-22.2-3.2-33.8-3.2c-59.5 0-112.1 29.6-144 74.8V224c0-17.7-14.3-32-32-32s-32 14.3-32 32v64c0 17.7 14.3 32 32 32c2.3 0 4.6-.3 6.8-.7c-4.5 15.5-6.8 31.8-6.8 48.7c0 5.4 .2 10.7 .7 16l-.7 0c-17.7 0-32 14.3-32 32v64H86.6C56.5 480 32 455.5 32 425.4c0-6.2 1.1-12.4 3.1-18.2L152.9 74.6C162 49.1 186.1 32 213.2 32zM496 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm0 240a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm0-192c-8.8 0-16 7.2-16 16v80c0 8.8 7.2 16 16 16s16-7.2 16-16V288c0-8.8-7.2-16-16-16z"],dungeon:[512,512,[],"f6d9","M336.6 156.5c1.3 1.1 2.7 2.2 3.9 3.3c9.3 8.2 23 10.5 33.4 3.6l67.6-45.1c11.4-7.6 14.2-23.2 5.1-33.4C430 66.6 410.9 50.6 389.7 37.6c-11.9-7.3-26.9-1.4-32.1 11.6l-30.5 76.2c-4.5 11.1 .2 23.6 9.5 31.2zM328 36.8c5.1-12.8-1.6-27.4-15-30.5C294.7 2.2 275.6 0 256 0s-38.7 2.2-57 6.4C185.5 9.4 178.8 24 184 36.8l30.3 75.8c4.5 11.3 16.8 17.2 29 16c4.2-.4 8.4-.6 12.7-.6s8.6 .2 12.7 .6c12.1 1.2 24.4-4.7 29-16L328 36.8zM65.5 85c-9.1 10.2-6.3 25.8 5.1 33.4l67.6 45.1c10.3 6.9 24.1 4.6 33.4-3.6c1.3-1.1 2.6-2.3 4-3.3c9.3-7.5 13.9-20.1 9.5-31.2L154.4 49.2c-5.2-12.9-20.3-18.8-32.1-11.6C101.1 50.6 82 66.6 65.5 85zm314 137.1c.9 3.3 1.7 6.6 2.3 10c2.5 13 13 23.9 26.2 23.9h80c13.3 0 24.1-10.8 22.9-24c-2.5-27.2-9.3-53.2-19.7-77.3c-5.5-12.9-21.4-16.6-33.1-8.9l-68.6 45.7c-9.8 6.5-13.2 19.2-10 30.5zM53.9 145.8c-11.6-7.8-27.6-4-33.1 8.9C10.4 178.8 3.6 204.8 1.1 232c-1.2 13.2 9.6 24 22.9 24h80c13.3 0 23.8-10.8 26.2-23.9c.6-3.4 1.4-6.7 2.3-10c3.1-11.4-.2-24-10-30.5L53.9 145.8zM104 288H24c-13.3 0-24 10.7-24 24v48c0 13.3 10.7 24 24 24h80c13.3 0 24-10.7 24-24V312c0-13.3-10.7-24-24-24zm304 0c-13.3 0-24 10.7-24 24v48c0 13.3 10.7 24 24 24h80c13.3 0 24-10.7 24-24V312c0-13.3-10.7-24-24-24H408zM24 416c-13.3 0-24 10.7-24 24v48c0 13.3 10.7 24 24 24h80c13.3 0 24-10.7 24-24V440c0-13.3-10.7-24-24-24H24zm384 0c-13.3 0-24 10.7-24 24v48c0 13.3 10.7 24 24 24h80c13.3 0 24-10.7 24-24V440c0-13.3-10.7-24-24-24H408zM272 192c0-8.8-7.2-16-16-16s-16 7.2-16 16V464c0 8.8 7.2 16 16 16s16-7.2 16-16V192zm-64 32c0-8.8-7.2-16-16-16s-16 7.2-16 16V464c0 8.8 7.2 16 16 16s16-7.2 16-16V224zm128 0c0-8.8-7.2-16-16-16s-16 7.2-16 16V464c0 8.8 7.2 16 16 16s16-7.2 16-16V224z"],"align-right":[448,512,[],"f038","M448 64c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32s14.3-32 32-32H416c17.7 0 32 14.3 32 32zm0 256c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32s14.3-32 32-32H416c17.7 0 32 14.3 32 32zM0 192c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32zM448 448c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H416c17.7 0 32 14.3 32 32z"],"money-bill-1-wave":[576,512,["money-bill-wave-alt"],"f53b","M0 112.5V422.3c0 18 10.1 35 27 41.3c87 32.5 174 10.3 261-11.9c79.8-20.3 159.6-40.7 239.3-18.9c23 6.3 48.7-9.5 48.7-33.4V89.7c0-18-10.1-35-27-41.3C462 15.9 375 38.1 288 60.3C208.2 80.6 128.4 100.9 48.7 79.1C25.6 72.8 0 88.6 0 112.5zM128 416H64V352c35.3 0 64 28.7 64 64zM64 224V160h64c0 35.3-28.7 64-64 64zM448 352c0-35.3 28.7-64 64-64v64H448zm64-192c-35.3 0-64-28.7-64-64h64v64zM384 256c0 61.9-43 112-96 112s-96-50.1-96-112s43-112 96-112s96 50.1 96 112zM252 208c0 9.7 6.9 17.7 16 19.6V276h-4c-11 0-20 9-20 20s9 20 20 20h24 24c11 0 20-9 20-20s-9-20-20-20h-4V208c0-11-9-20-20-20H272c-11 0-20 9-20 20z"],"life-ring":[512,512,[],"f1cd","M367.2 412.5C335.9 434.9 297.5 448 256 448s-79.9-13.1-111.2-35.5l58-58c15.8 8.6 34 13.5 53.3 13.5s37.4-4.9 53.3-13.5l58 58zm90.7 .8c33.8-43.4 54-98 54-157.3s-20.2-113.9-54-157.3c9-12.5 7.9-30.1-3.4-41.3S425.8 45 413.3 54C369.9 20.2 315.3 0 256 0S142.1 20.2 98.7 54c-12.5-9-30.1-7.9-41.3 3.4S45 86.2 54 98.7C20.2 142.1 0 196.7 0 256s20.2 113.9 54 157.3c-9 12.5-7.9 30.1 3.4 41.3S86.2 467 98.7 458c43.4 33.8 98 54 157.3 54s113.9-20.2 157.3-54c12.5 9 30.1 7.9 41.3-3.4s12.4-28.8 3.4-41.3zm-45.5-46.1l-58-58c8.6-15.8 13.5-34 13.5-53.3s-4.9-37.4-13.5-53.3l58-58C434.9 176.1 448 214.5 448 256s-13.1 79.9-35.5 111.2zM367.2 99.5l-58 58c-15.8-8.6-34-13.5-53.3-13.5s-37.4 4.9-53.3 13.5l-58-58C176.1 77.1 214.5 64 256 64s79.9 13.1 111.2 35.5zM157.5 309.3l-58 58C77.1 335.9 64 297.5 64 256s13.1-79.9 35.5-111.2l58 58c-8.6 15.8-13.5 34-13.5 53.3s4.9 37.4 13.5 53.3zM208 256a48 48 0 1 1 96 0 48 48 0 1 1 -96 0z"],hands:[576,512,["sign-language","signing"],"f2a7","M544 160l-.1 72.6c-.1 52.2-24 101-64 133.1c.1-1.9 .1-3.8 .1-5.7v-8c0-71.8-37-138.6-97.9-176.7l-60.2-37.6c-8.6-5.4-17.9-8.4-27.3-9.4L248.7 48.8c-6.6-11.5-2.7-26.2 8.8-32.8s26.2-2.7 32.8 8.8l78 135.1c3.3 5.7 10.7 7.7 16.4 4.4s7.7-10.7 4.4-16.4l-62-107.4c-6.6-11.5-2.7-26.2 8.8-32.8S362 5 368.6 16.5l68 117.8 0 0 0 0 43.3 75L480 160c0-17.7 14.4-32 32-32s32 14.4 32 32zM243.9 88.5L268.5 131c-13.9 4.5-26.4 13.7-34.7 27c-.9 1.4-1.7 2.9-2.5 4.4l-28.9-50c-6.6-11.5-2.7-26.2 8.8-32.8s26.2-2.7 32.8 8.8zm-46.4 63.7l26.8 46.4c.6 6 2.1 11.8 4.3 17.4H224 210.7l0 0H179l-23-39.8c-6.6-11.5-2.7-26.2 8.8-32.8s26.2-2.7 32.8 8.8zM260.9 175c9.4-15 29.1-19.5 44.1-10.2l60.2 37.6C416.7 234.7 448 291.2 448 352v8c0 83.9-68.1 152-152 152H120c-13.3 0-24-10.7-24-24s10.7-24 24-24h92c6.6 0 12-5.4 12-12s-5.4-12-12-12H88c-13.3 0-24-10.7-24-24s10.7-24 24-24H212c6.6 0 12-5.4 12-12s-5.4-12-12-12H56c-13.3 0-24-10.7-24-24s10.7-24 24-24H212c6.6 0 12-5.4 12-12s-5.4-12-12-12H88c-13.3 0-24-10.7-24-24s10.7-24 24-24H224l0 0 0 0h93.2L271 219.1c-15-9.4-19.5-29.1-10.2-44.1z"],"calendar-day":[448,512,[],"f783","M128 0c17.7 0 32 14.3 32 32V64H288V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64h48c26.5 0 48 21.5 48 48v48H0V112C0 85.5 21.5 64 48 64H96V32c0-17.7 14.3-32 32-32zM0 192H448V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V192zm80 64c-8.8 0-16 7.2-16 16v96c0 8.8 7.2 16 16 16h96c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H80z"],"water-ladder":[576,512,["ladder-water","swimming-pool"],"f5c5","M128 127.7C128 74.9 170.9 32 223.7 32c48.3 0 89 36 95 83.9l1 8.2c2.2 17.5-10.2 33.5-27.8 35.7s-33.5-10.2-35.7-27.8l-1-8.2c-2-15.9-15.5-27.8-31.5-27.8c-17.5 0-31.7 14.2-31.7 31.7V224H384V127.7C384 74.9 426.9 32 479.7 32c48.3 0 89 36 95 83.9l1 8.2c2.2 17.5-10.2 33.5-27.8 35.7s-33.5-10.2-35.7-27.8l-1-8.2c-2-15.9-15.5-27.8-31.5-27.8c-17.5 0-31.7 14.2-31.7 31.7V361c-1.6 1-3.3 2-4.8 3.1c-18 12.4-40.1 20.3-59.2 20.3h0V288H192v96.5c-19 0-41.2-7.9-59.1-20.3c-1.6-1.1-3.2-2.2-4.9-3.1V127.7zM306.5 389.9C329 405.4 356.5 416 384 416c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 469.7 417 480 384 480c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 405.2 165.1 416 192 416c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z"],"arrows-up-down":[320,512,["arrows-v"],"f07d","M182.6 9.4c-12.5-12.5-32.8-12.5-45.3 0l-96 96c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L128 109.3V402.7L86.6 361.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l96 96c12.5 12.5 32.8 12.5 45.3 0l96-96c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L192 402.7V109.3l41.4 41.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-96-96z"],"face-grimace":[512,512,[128556,"grimace"],"f57f","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm96-112h-8V360l55.3 0c-3.8 22.7-23.6 40-47.3 40zm47.3-56L344 344V304h8c23.8 0 43.5 17.3 47.3 40zM328 344H264V304h64v40zm0 56H264V360h64v40zm-80-96v40l-64 0V304h64zm0 56v40H184V360l64 0zm-80-16H112.7c3.8-22.7 23.6-40 47.3-40h8v40zm0 56h-8c-23.8 0-43.5-17.3-47.3-40H168v40zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],"wheelchair-move":[448,512,["wheelchair-alt"],"e2ce","M320 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM204.5 121.3c-5.4-2.5-11.7-1.9-16.4 1.7l-40.9 30.7c-14.1 10.6-34.2 7.7-44.8-6.4s-7.7-34.2 6.4-44.8l40.9-30.7c23.7-17.8 55.3-21 82.1-8.4l90.4 42.5c29.1 13.7 36.8 51.6 15.2 75.5L299.1 224h97.4c30.3 0 53 27.7 47.1 57.4L415.4 422.3c-3.5 17.3-20.3 28.6-37.7 25.1s-28.6-20.3-25.1-37.7L377 288H306.7c8.6 19.6 13.3 41.2 13.3 64c0 88.4-71.6 160-160 160S0 440.4 0 352s71.6-160 160-160c11.1 0 22 1.1 32.4 3.3l54.2-54.2-42.1-19.8zM160 448a96 96 0 1 0 0-192 96 96 0 1 0 0 192z"],"turn-down":[384,512,[10549,"level-down-alt"],"f3be","M350 334.5c3.8 8.8 2 19-4.6 26l-136 144c-4.5 4.8-10.8 7.5-17.4 7.5s-12.9-2.7-17.4-7.5l-136-144c-6.6-7-8.4-17.2-4.6-26s12.5-14.5 22-14.5h88l0-192c0-17.7-14.3-32-32-32H32C14.3 96 0 81.7 0 64V32C0 14.3 14.3 0 32 0l80 0c70.7 0 128 57.3 128 128l0 192h88c9.6 0 18.2 5.7 22 14.5z"],"person-walking-arrow-right":[640,512,[],"e552","M208 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM123.7 200.5c1-.4 1.9-.8 2.9-1.2l-16.9 63.5c-5.6 21.1-.1 43.6 14.7 59.7l70.7 77.1 22 88.1c4.3 17.1 21.7 27.6 38.8 23.3s27.6-21.7 23.3-38.8l-23-92.1c-1.9-7.8-5.8-14.9-11.2-20.8l-49.5-54 19.3-65.5 9.6 23c4.4 10.6 12.5 19.3 22.8 24.5l26.7 13.3c15.8 7.9 35 1.5 42.9-14.3s1.5-35-14.3-42.9L281 232.7l-15.3-36.8C248.5 154.8 208.3 128 163.7 128c-22.8 0-45.3 4.8-66.1 14l-8 3.5c-32.9 14.6-58.1 42.4-69.4 76.5l-2.6 7.8c-5.6 16.8 3.5 34.9 20.2 40.5s34.9-3.5 40.5-20.2l2.6-7.8c5.7-17.1 18.3-30.9 34.7-38.2l8-3.5zm-30 135.1L68.7 398 9.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L116.3 441c4.6-4.6 8.2-10.1 10.6-16.1l14.5-36.2-40.7-44.4c-2.5-2.7-4.8-5.6-7-8.6zM550.6 153.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L530.7 224H384c-17.7 0-32 14.3-32 32s14.3 32 32 32H530.7l-25.4 25.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l80-80c12.5-12.5 12.5-32.8 0-45.3l-80-80z"],"square-envelope":[448,512,["envelope-square"],"f199","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM218 271.7L64.2 172.4C66 156.4 79.5 144 96 144H352c16.5 0 30 12.4 31.8 28.4L230 271.7c-1.8 1.2-3.9 1.8-6 1.8s-4.2-.6-6-1.8zm29.4 26.9L384 210.4V336c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V210.4l136.6 88.2c7 4.5 15.1 6.9 23.4 6.9s16.4-2.4 23.4-6.9z"],dice:[640,512,[127922],"f522","M274.9 34.3c-28.1-28.1-73.7-28.1-101.8 0L34.3 173.1c-28.1 28.1-28.1 73.7 0 101.8L173.1 413.7c28.1 28.1 73.7 28.1 101.8 0L413.7 274.9c28.1-28.1 28.1-73.7 0-101.8L274.9 34.3zM200 224a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zM96 200a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM224 376a24 24 0 1 1 0-48 24 24 0 1 1 0 48zM352 200a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM224 120a24 24 0 1 1 0-48 24 24 0 1 1 0 48zm96 328c0 35.3 28.7 64 64 64H576c35.3 0 64-28.7 64-64V256c0-35.3-28.7-64-64-64H461.7c11.6 36 3.1 77-25.4 105.5L320 413.8V448zM480 328a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],"bowling-ball":[512,512,[],"f436","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM240 80a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM208 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm-64-64a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],brain:[512,512,[129504],"f5dc","M184 0c30.9 0 56 25.1 56 56V456c0 30.9-25.1 56-56 56c-28.9 0-52.7-21.9-55.7-50.1c-5.2 1.4-10.7 2.1-16.3 2.1c-35.3 0-64-28.7-64-64c0-7.4 1.3-14.6 3.6-21.2C21.4 367.4 0 338.2 0 304c0-31.9 18.7-59.5 45.8-72.3C37.1 220.8 32 207 32 192c0-30.7 21.6-56.3 50.4-62.6C80.8 123.9 80 118 80 112c0-29.9 20.6-55.1 48.3-62.1C131.3 21.9 155.1 0 184 0zM328 0c28.9 0 52.6 21.9 55.7 49.9c27.8 7 48.3 32.1 48.3 62.1c0 6-.8 11.9-2.4 17.4c28.8 6.2 50.4 31.9 50.4 62.6c0 15-5.1 28.8-13.8 39.7C493.3 244.5 512 272.1 512 304c0 34.2-21.4 63.4-51.6 74.8c2.3 6.6 3.6 13.8 3.6 21.2c0 35.3-28.7 64-64 64c-5.6 0-11.1-.7-16.3-2.1c-3 28.2-26.8 50.1-55.7 50.1c-30.9 0-56-25.1-56-56V56c0-30.9 25.1-56 56-56z"],bandage:[640,512,[129657,"band-aid"],"f462","M480 416h96c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H480V416zM448 96H192V416H448V96zM64 96C28.7 96 0 124.7 0 160V352c0 35.3 28.7 64 64 64h96V96H64zM248 208a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm120-24a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM248 304a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm120-24a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],"calendar-minus":[448,512,[],"f272","M128 0c17.7 0 32 14.3 32 32V64H288V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64h48c26.5 0 48 21.5 48 48v48H0V112C0 85.5 21.5 64 48 64H96V32c0-17.7 14.3-32 32-32zM0 192H448V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V192zM312 376c13.3 0 24-10.7 24-24s-10.7-24-24-24H136c-13.3 0-24 10.7-24 24s10.7 24 24 24H312z"],"circle-xmark":[512,512,[61532,"times-circle","xmark-circle"],"f057","M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM175 175c9.4-9.4 24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9z"],gifts:[640,512,[],"f79c","M200.6 32C205 19.5 198.5 5.8 186 1.4S159.8 3.5 155.4 16L144.7 46.2l-9.9-29.8C130.6 3.8 117-3 104.4 1.2S85 19 89.2 31.6l8.3 25-27.4-20c-10.7-7.8-25.7-5.4-33.5 5.3s-5.4 25.7 5.3 33.5L70.2 96H48C21.5 96 0 117.5 0 144V464c0 26.5 21.5 48 48 48H200.6c-5.4-9.4-8.6-20.3-8.6-32V256c0-29.9 20.5-55 48.2-62c1.8-31 17.1-58.2 40.1-76.1C271.7 104.7 256.9 96 240 96H217.8l28.3-20.6c10.7-7.8 13.1-22.8 5.3-33.5s-22.8-13.1-33.5-5.3L192.5 55.1 200.6 32zM363.5 185.5L393.1 224H344c-13.3 0-24-10.7-24-24c0-13.1 10.8-24 24.2-24c7.6 0 14.7 3.5 19.3 9.5zM272 200c0 8.4 1.4 16.5 4.1 24H272c-26.5 0-48 21.5-48 48v80H416V256h32v96H640V272c0-26.5-21.5-48-48-48h-4.1c2.7-7.5 4.1-15.6 4.1-24c0-39.9-32.5-72-72.2-72c-22.4 0-43.6 10.4-57.3 28.2L432 195.8l-30.5-39.6c-13.7-17.8-35-28.2-57.3-28.2c-39.7 0-72.2 32.1-72.2 72zM224 464c0 26.5 21.5 48 48 48H416V384H224v80zm224 48H592c26.5 0 48-21.5 48-48V384H448V512zm96-312c0 13.3-10.7 24-24 24H470.9l29.6-38.5c4.6-5.9 11.7-9.5 19.3-9.5c13.4 0 24.2 10.9 24.2 24z"],hotel:[512,512,[127976],"f594","M0 32C0 14.3 14.3 0 32 0H480c17.7 0 32 14.3 32 32s-14.3 32-32 32V448c17.7 0 32 14.3 32 32s-14.3 32-32 32H304V464c0-26.5-21.5-48-48-48s-48 21.5-48 48v48H32c-17.7 0-32-14.3-32-32s14.3-32 32-32V64C14.3 64 0 49.7 0 32zm96 80v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H112c-8.8 0-16 7.2-16 16zM240 96c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H240zm112 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16zM112 192c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V208c0-8.8-7.2-16-16-16H112zm112 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V208c0-8.8-7.2-16-16-16H240c-8.8 0-16 7.2-16 16zm144-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V208c0-8.8-7.2-16-16-16H368zM328 384c13.3 0 24.3-10.9 21-23.8c-10.6-41.5-48.2-72.2-93-72.2s-82.5 30.7-93 72.2c-3.3 12.8 7.8 23.8 21 23.8H328z"],"earth-asia":[512,512,[127759,"globe-asia"],"f57e","M51.7 295.1l31.7 6.3c7.9 1.6 16-.9 21.7-6.6l15.4-15.4c11.6-11.6 31.1-8.4 38.4 6.2l9.3 18.5c4.8 9.6 14.6 15.7 25.4 15.7c15.2 0 26.1-14.6 21.7-29.2l-6-19.9c-4.6-15.4 6.9-30.9 23-30.9h2.3c13.4 0 25.9-6.7 33.3-17.8l10.7-16.1c5.6-8.5 5.3-19.6-.8-27.7l-16.1-21.5c-10.3-13.7-3.3-33.5 13.4-37.7l17-4.3c7.5-1.9 13.6-7.2 16.5-14.4l16.4-40.9C303.4 52.1 280.2 48 256 48C141.1 48 48 141.1 48 256c0 13.4 1.3 26.5 3.7 39.1zm407.7 4.6c-3-.3-6-.1-9 .8l-15.8 4.4c-6.7 1.9-13.8-.9-17.5-6.7l-2-3.1c-6-9.4-16.4-15.1-27.6-15.1s-21.6 5.7-27.6 15.1l-6.1 9.5c-1.4 2.2-3.4 4.1-5.7 5.3L312 330.1c-18.1 10.1-25.5 32.4-17 51.3l5.5 12.4c8.6 19.2 30.7 28.5 50.5 21.1l2.6-1c10-3.7 21.3-2.2 29.9 4.1l1.5 1.1c37.2-29.5 64.1-71.4 74.4-119.5zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm144.5 92.1c-2.1 8.6 3.1 17.3 11.6 19.4l32 8c8.6 2.1 17.3-3.1 19.4-11.6s-3.1-17.3-11.6-19.4l-32-8c-8.6-2.1-17.3 3.1-19.4 11.6zm92-20c-2.1 8.6 3.1 17.3 11.6 19.4s17.3-3.1 19.4-11.6l8-32c2.1-8.6-3.1-17.3-11.6-19.4s-17.3 3.1-19.4 11.6l-8 32zM343.2 113.7c-7.9-4-17.5-.7-21.5 7.2l-16 32c-4 7.9-.7 17.5 7.2 21.5s17.5 .7 21.5-7.2l16-32c4-7.9 .7-17.5-7.2-21.5z"],"id-card-clip":[576,512,["id-card-alt"],"f47f","M256 0h64c17.7 0 32 14.3 32 32V96c0 17.7-14.3 32-32 32H256c-17.7 0-32-14.3-32-32V32c0-17.7 14.3-32 32-32zM64 64H192v48c0 26.5 21.5 48 48 48h96c26.5 0 48-21.5 48-48V64H512c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128C0 92.7 28.7 64 64 64zM176 437.3c0 5.9 4.8 10.7 10.7 10.7H389.3c5.9 0 10.7-4.8 10.7-10.7c0-29.5-23.9-53.3-53.3-53.3H229.3c-29.5 0-53.3 23.9-53.3 53.3zM288 352a64 64 0 1 0 0-128 64 64 0 1 0 0 128z"],"magnifying-glass-plus":[512,512,["search-plus"],"f00e","M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zM184 296c0 13.3 10.7 24 24 24s24-10.7 24-24V232h64c13.3 0 24-10.7 24-24s-10.7-24-24-24H232V120c0-13.3-10.7-24-24-24s-24 10.7-24 24v64H120c-13.3 0-24 10.7-24 24s10.7 24 24 24h64v64z"],"thumbs-up":[512,512,[128077,61575],"f164","M313.4 32.9c26 5.2 42.9 30.5 37.7 56.5l-2.3 11.4c-5.3 26.7-15.1 52.1-28.8 75.2H464c26.5 0 48 21.5 48 48c0 18.5-10.5 34.6-25.9 42.6C497 275.4 504 288.9 504 304c0 23.4-16.8 42.9-38.9 47.1c4.4 7.3 6.9 15.8 6.9 24.9c0 21.3-13.9 39.4-33.1 45.6c.7 3.3 1.1 6.8 1.1 10.4c0 26.5-21.5 48-48 48H294.5c-19 0-37.5-5.6-53.3-16.1l-38.5-25.7C176 420.4 160 390.4 160 358.3V320 272 247.1c0-29.2 13.3-56.7 36-75l7.4-5.9c26.5-21.2 44.6-51 51.2-84.2l2.3-11.4c5.2-26 30.5-42.9 56.5-37.7zM32 192H96c17.7 0 32 14.3 32 32V448c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32V224c0-17.7 14.3-32 32-32z"],"user-clock":[640,512,[],"f4fd","M224 0a128 128 0 1 1 0 256A128 128 0 1 1 224 0zM178.3 304h91.4c20.6 0 40.4 3.5 58.8 9.9C323 331 320 349.1 320 368c0 59.5 29.5 112.1 74.8 144H29.7C13.3 512 0 498.7 0 482.3C0 383.8 79.8 304 178.3 304zM352 368a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm144-80c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H512V304c0-8.8-7.2-16-16-16z"],"hand-dots":[512,512,["allergies"],"f461","M288 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V336c0 1.5 0 3.1 .1 4.6L67.6 283c-16-15.2-41.3-14.6-56.6 1.4s-14.6 41.3 1.4 56.6L124.8 448c43.1 41.1 100.4 64 160 64H304c97.2 0 176-78.8 176-176V128c0-17.7-14.3-32-32-32s-32 14.3-32 32V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V32zM240 336a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zm80 16a16 16 0 1 1 0 32 16 16 0 1 1 0-32zm48-16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zm-16 80a16 16 0 1 1 0 32 16 16 0 1 1 0-32zM240 432a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zm-48-48a16 16 0 1 1 0 32 16 16 0 1 1 0-32z"],"file-invoice":[384,512,[],"f570","M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM80 64h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16s7.2-16 16-16zm16 96H288c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V256c0-17.7 14.3-32 32-32zm0 32v64H288V256H96zM240 416h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H240c-8.8 0-16-7.2-16-16s7.2-16 16-16z"],"window-minimize":[512,512,[128469],"f2d1","M32 416c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H32z"],"mug-saucer":[640,512,["coffee"],"f0f4","M96 64c0-17.7 14.3-32 32-32H448h64c70.7 0 128 57.3 128 128s-57.3 128-128 128H480c0 53-43 96-96 96H192c-53 0-96-43-96-96V64zM480 224h32c35.3 0 64-28.7 64-64s-28.7-64-64-64H480V224zM32 416H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32z"],brush:[384,512,[],"f55d","M162.4 6c-1.5-3.6-5-6-8.9-6h-19c-3.9 0-7.5 2.4-8.9 6L104.9 57.7c-3.2 8-14.6 8-17.8 0L66.4 6c-1.5-3.6-5-6-8.9-6H48C21.5 0 0 21.5 0 48V224v22.4V256H9.6 374.4 384v-9.6V224 48c0-26.5-21.5-48-48-48H230.5c-3.9 0-7.5 2.4-8.9 6L200.9 57.7c-3.2 8-14.6 8-17.8 0L162.4 6zM0 288v32c0 35.3 28.7 64 64 64h64v64c0 35.3 28.7 64 64 64s64-28.7 64-64V384h64c35.3 0 64-28.7 64-64V288H0zM192 432a16 16 0 1 1 0 32 16 16 0 1 1 0-32z"],mask:[576,512,[],"f6fa","M288 64C64 64 0 160 0 272S80 448 176 448h8.4c24.2 0 46.4-13.7 57.2-35.4l23.2-46.3c4.4-8.8 13.3-14.3 23.2-14.3s18.8 5.5 23.2 14.3l23.2 46.3c10.8 21.7 33 35.4 57.2 35.4H400c96 0 176-64 176-176s-64-208-288-208zM96 256a64 64 0 1 1 128 0A64 64 0 1 1 96 256zm320-64a64 64 0 1 1 0 128 64 64 0 1 1 0-128z"],"magnifying-glass-minus":[512,512,["search-minus"],"f010","M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zM136 184c-13.3 0-24 10.7-24 24s10.7 24 24 24H280c13.3 0 24-10.7 24-24s-10.7-24-24-24H136z"],"ruler-vertical":[256,512,[],"f548","M0 48C0 21.5 21.5 0 48 0H208c26.5 0 48 21.5 48 48V96H176c-8.8 0-16 7.2-16 16s7.2 16 16 16h80v64H176c-8.8 0-16 7.2-16 16s7.2 16 16 16h80v64H176c-8.8 0-16 7.2-16 16s7.2 16 16 16h80v64H176c-8.8 0-16 7.2-16 16s7.2 16 16 16h80v48c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V48z"],"user-large":[512,512,["user-alt"],"f406","M256 288A144 144 0 1 0 256 0a144 144 0 1 0 0 288zm-94.7 32C72.2 320 0 392.2 0 481.3c0 17 13.8 30.7 30.7 30.7H481.3c17 0 30.7-13.8 30.7-30.7C512 392.2 439.8 320 350.7 320H161.3z"],"train-tram":[448,512,[128650],"e5b4","M86.8 48c-12.2 0-23.6 5.5-31.2 15L42.7 79C34.5 89.3 19.4 91 9 82.7S-3 59.4 5.3 49L18 33C34.7 12.2 60 0 86.8 0H361.2c26.7 0 52 12.2 68.7 33l12.8 16c8.3 10.4 6.6 25.5-3.8 33.7s-25.5 6.6-33.7-3.7L392.5 63c-7.6-9.5-19.1-15-31.2-15H248V96h40c53 0 96 43 96 96V352c0 30.6-14.3 57.8-36.6 75.4l65.5 65.5c7.1 7.1 2.1 19.1-7.9 19.1H365.3c-8.5 0-16.6-3.4-22.6-9.4L288 448H160l-54.6 54.6c-6 6-14.1 9.4-22.6 9.4H43c-10 0-15-12.1-7.9-19.1l65.5-65.5C78.3 409.8 64 382.6 64 352V192c0-53 43-96 96-96h40V48H86.8zM160 160c-17.7 0-32 14.3-32 32v32c0 17.7 14.3 32 32 32H288c17.7 0 32-14.3 32-32V192c0-17.7-14.3-32-32-32H160zm32 192a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm96 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z"],"user-nurse":[448,512,[],"f82f","M96 128V70.2c0-13.3 8.3-25.3 20.8-30l96-36c7.2-2.7 15.2-2.7 22.5 0l96 36c12.5 4.7 20.8 16.6 20.8 30V128h-.3c.2 2.6 .3 5.3 .3 8v40c0 70.7-57.3 128-128 128s-128-57.3-128-128V136c0-2.7 .1-5.4 .3-8H96zm48 48c0 44.2 35.8 80 80 80s80-35.8 80-80V160H144v16zM111.9 327.7c10.5-3.4 21.8 .4 29.4 8.5l71 75.5c6.3 6.7 17 6.7 23.3 0l71-75.5c7.6-8.1 18.9-11.9 29.4-8.5C401 348.6 448 409.4 448 481.3c0 17-13.8 30.7-30.7 30.7H30.7C13.8 512 0 498.2 0 481.3c0-71.9 47-132.7 111.9-153.6zM208 48V64H192c-4.4 0-8 3.6-8 8V88c0 4.4 3.6 8 8 8h16v16c0 4.4 3.6 8 8 8h16c4.4 0 8-3.6 8-8V96h16c4.4 0 8-3.6 8-8V72c0-4.4-3.6-8-8-8H240V48c0-4.4-3.6-8-8-8H216c-4.4 0-8 3.6-8 8z"],syringe:[512,512,[128137],"f48e","M441 7l32 32 32 32c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-15-15L417.9 128l55 55c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-72-72L295 73c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l55 55L422.1 56 407 41c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0zM210.3 155.7l61.1-61.1c.3 .3 .6 .7 1 1l16 16 56 56 56 56 16 16c.3 .3 .6 .6 1 1l-191 191c-10.5 10.5-24.7 16.4-39.6 16.4H97.9L41 505c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l57-57V325.3c0-14.9 5.9-29.1 16.4-39.6l43.3-43.3 57 57c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-57-57 41.4-41.4 57 57c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-57-57z"],"cloud-sun":[640,512,[9925],"f6c4","M294.2 1.2c5.1 2.1 8.7 6.7 9.6 12.1l14.1 84.7 84.7 14.1c5.4 .9 10 4.5 12.1 9.6s1.5 10.9-1.6 15.4l-38.5 55c-2.2-.1-4.4-.2-6.7-.2c-23.3 0-45.1 6.2-64 17.1l0-1.1c0-53-43-96-96-96s-96 43-96 96s43 96 96 96c8.1 0 15.9-1 23.4-2.9c-36.6 18.1-63.3 53.1-69.8 94.9l-24.4 17c-4.5 3.1-10.3 3.8-15.4 1.6s-8.7-6.7-9.6-12.1L98.1 317.9 13.4 303.8c-5.4-.9-10-4.5-12.1-9.6s-1.5-10.9 1.6-15.4L52.5 208 2.9 137.2c-3.2-4.5-3.8-10.3-1.6-15.4s6.7-8.7 12.1-9.6L98.1 98.1l14.1-84.7c.9-5.4 4.5-10 9.6-12.1s10.9-1.5 15.4 1.6L208 52.5 278.8 2.9c4.5-3.2 10.3-3.8 15.4-1.6zM144 208a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM639.9 431.9c0 44.2-35.8 80-80 80H288c-53 0-96-43-96-96c0-47.6 34.6-87 80-94.6l0-1.3c0-53 43-96 96-96c34.9 0 65.4 18.6 82.2 46.4c13-9.1 28.8-14.4 45.8-14.4c44.2 0 80 35.8 80 80c0 5.9-.6 11.7-1.9 17.2c37.4 6.7 65.8 39.4 65.8 78.7z"],"stopwatch-20":[448,512,[],"e06f","M176 0c-17.7 0-32 14.3-32 32s14.3 32 32 32h16V98.4C92.3 113.8 16 200 16 304c0 114.9 93.1 208 208 208s208-93.1 208-208c0-41.8-12.3-80.7-33.5-113.2l24.1-24.1c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L355.7 143c-28.1-23-62.2-38.8-99.7-44.6V64h16c17.7 0 32-14.3 32-32s-14.3-32-32-32H176zM288 204c28.7 0 52 23.3 52 52v96c0 28.7-23.3 52-52 52s-52-23.3-52-52V256c0-28.7 23.3-52 52-52zm-12 52v96c0 6.6 5.4 12 12 12s12-5.4 12-12V256c0-6.6-5.4-12-12-12s-12 5.4-12 12zM159.5 244c-5.4 0-10.2 3.5-11.9 8.6l-.6 1.7c-3.5 10.5-14.8 16.1-25.3 12.6s-16.1-14.8-12.6-25.3l.6-1.7c7.2-21.5 27.2-35.9 49.8-35.9c29 0 52.5 23.5 52.5 52.5v2.2c0 13.4-4.9 26.4-13.8 36.4l-39 43.9c-6.2 7-10 15.7-10.9 24.9H192c11 0 20 9 20 20s-9 20-20 20H128c-11 0-20-9-20-20V368.3c0-20.6 7.5-40.4 21.2-55.8l39-43.9c2.4-2.7 3.7-6.2 3.7-9.8v-2.2c0-6.9-5.6-12.5-12.5-12.5z"],"square-full":[512,512,[128997,128998,128999,129e3,129001,129002,129003,11035,11036],"f45c","M0 0H512V512H0V0z"],magnet:[448,512,[129522],"f076","M0 160v96C0 379.7 100.3 480 224 480s224-100.3 224-224V160H320v96c0 53-43 96-96 96s-96-43-96-96V160H0zm0-32H128V64c0-17.7-14.3-32-32-32H32C14.3 32 0 46.3 0 64v64zm320 0H448V64c0-17.7-14.3-32-32-32H352c-17.7 0-32 14.3-32 32v64z"],jar:[320,512,[],"e516","M32 32C32 14.3 46.3 0 64 0H256c17.7 0 32 14.3 32 32s-14.3 32-32 32H64C46.3 64 32 49.7 32 32zM0 160c0-35.3 28.7-64 64-64H256c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V160zm96 64c-17.7 0-32 14.3-32 32v96c0 17.7 14.3 32 32 32H224c17.7 0 32-14.3 32-32V256c0-17.7-14.3-32-32-32H96z"],"note-sticky":[448,512,[62026,"sticky-note"],"f249","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H288V368c0-26.5 21.5-48 48-48H448V96c0-35.3-28.7-64-64-64H64zM448 352H402.7 336c-8.8 0-16 7.2-16 16v66.7V480l32-32 64-64 32-32z"],"bug-slash":[640,512,[],"e490","M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L477.4 348.9c1.7-9.4 2.6-19 2.6-28.9h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H479.7c-1.1-14.1-5-27.5-11.1-39.5c.7-.6 1.4-1.2 2.1-1.9l64-64c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-64 64c-.7 .7-1.3 1.4-1.9 2.1C409.2 164.1 393.1 160 376 160H264c-8.3 0-16.3 1-24 2.8L38.8 5.1zm392 430.3L336 360.7V479.2c36.6-3.6 69.7-19.6 94.8-43.8zM166.7 227.3c-3.4 9-5.6 18.7-6.4 28.7H96c-17.7 0-32 14.3-32 32s14.3 32 32 32h64c0 24.6 5.5 47.8 15.4 68.6c-2.2 1.3-4.2 2.9-6 4.8l-64 64c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l63.1-63.1c24.5 21.8 55.8 36.2 90.3 39.6V335.5L166.7 227.3zM320 0c-53 0-96 43-96 96v3.6c0 15.7 12.7 28.4 28.4 28.4H387.6c15.7 0 28.4-12.7 28.4-28.4V96c0-53-43-96-96-96z"],"arrow-up-from-water-pump":[576,512,[],"e4b6","M112 0C85.5 0 64 21.5 64 48V256H48c-26.5 0-48 21.5-48 48v96c0 8 2 15.6 5.4 22.2c3.8-1.7 7.8-3.1 12-4.1c13.1-3.1 26.7-9.8 37.3-18.6c22.2-18.7 54.3-20.1 78.1-3.4c18 12.4 40.1 20.3 59.2 20.3c21.1 0 42-8.5 59.2-20.3c22.1-15.5 51.6-15.5 73.7 0c18.4 12.7 39.6 20.3 59.2 20.3c19 0 41.2-7.9 59.2-20.3c23.8-16.7 55.8-15.3 78.1 3.4c10.6 8.8 24.2 15.6 37.3 18.6c4.2 1 8.2 2.4 12 4.1C574 415.6 576 408 576 400V304c0-26.5-21.5-48-48-48H480l0-146.7 25.4 25.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-80-80c-12.5-12.5-32.8-12.5-45.3 0l-80 80c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L416 109.3 416 256H288V48c0-26.5-21.5-48-48-48H112zM306.5 421.9c-11.1-7.9-25.9-7.9-37 0C247 437.4 219.5 448 192 448c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C126.5 501.7 159 512 192 512c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.4-27.3-10.1-39.2-1.7l0 0C439.4 437.2 410.9 448 384 448c-27.5 0-55-10.6-77.5-26.1z"],bone:[576,512,[129460],"f5d7","M153.7 144.8c6.9 16.3 20.6 31.2 38.3 31.2H384c17.7 0 31.4-14.9 38.3-31.2C434.4 116.1 462.9 96 496 96c44.2 0 80 35.8 80 80c0 30.4-17 56.9-42 70.4c-3.6 1.9-6 5.5-6 9.6s2.4 7.7 6 9.6c25 13.5 42 40 42 70.4c0 44.2-35.8 80-80 80c-33.1 0-61.6-20.1-73.7-48.8C415.4 350.9 401.7 336 384 336H192c-17.7 0-31.4 14.9-38.3 31.2C141.6 395.9 113.1 416 80 416c-44.2 0-80-35.8-80-80c0-30.4 17-56.9 42-70.4c3.6-1.9 6-5.5 6-9.6s-2.4-7.7-6-9.6C17 232.9 0 206.4 0 176c0-44.2 35.8-80 80-80c33.1 0 61.6 20.1 73.7 48.8z"],"user-injured":[448,512,[],"f728","M240 80H342.7c-7.9-19.5-20.4-36.5-36.2-49.9L240 80zm37.7-68.2C261.3 4.2 243.2 0 224 0c-53.7 0-99.7 33.1-118.7 80h81.4l91-68.2zM224 256c70.7 0 128-57.3 128-128c0-5.4-.3-10.8-1-16H97c-.7 5.2-1 10.6-1 16c0 70.7 57.3 128 128 128zM124 312.4c-9.7 3.1-19.1 7-28 11.7V512H243.7L181.5 408.2 124 312.4zm33-7.2L204.3 384H272c44.2 0 80 35.8 80 80c0 18-6 34.6-16 48h82.3c16.4 0 29.7-13.3 29.7-29.7C448 383.8 368.2 304 269.7 304H178.3c-7.2 0-14.3 .4-21.3 1.3zM0 482.3C0 498.7 13.3 512 29.7 512H64V345.4C24.9 378.1 0 427.3 0 482.3zM320 464c0-26.5-21.5-48-48-48H223.5l57.1 95.2C303 507.2 320 487.6 320 464z"],"face-sad-tear":[512,512,[128546,"sad-tear"],"f5b4","M0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zm240 80c0-8.8 7.2-16 16-16c45 0 85.6 20.5 115.7 53.1c6 6.5 5.6 16.6-.9 22.6s-16.6 5.6-22.6-.9c-25-27.1-57.4-42.9-92.3-42.9c-8.8 0-16-7.2-16-16zm-80 80c-26.5 0-48-21-48-47c0-20 28.6-60.4 41.6-77.7c3.2-4.4 9.6-4.4 12.8 0C179.6 308.6 208 349 208 369c0 26-21.5 47-48 47zM367.6 208a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm-192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],plane:[576,512,[],"f072","M482.3 192c34.2 0 93.7 29 93.7 64c0 36-59.5 64-93.7 64l-116.6 0L265.2 495.9c-5.7 10-16.3 16.1-27.8 16.1l-56.2 0c-10.6 0-18.3-10.2-15.4-20.4l49-171.6L112 320 68.8 377.6c-3 4-7.8 6.4-12.8 6.4l-42 0c-7.8 0-14-6.3-14-14c0-1.3 .2-2.6 .5-3.9L32 256 .5 145.9c-.4-1.3-.5-2.6-.5-3.9c0-7.8 6.3-14 14-14l42 0c5 0 9.8 2.4 12.8 6.4L112 192l102.9 0-49-171.6C162.9 10.2 170.6 0 181.2 0l56.2 0c11.5 0 22.1 6.2 27.8 16.1L365.7 192l116.6 0z"],"tent-arrows-down":[576,512,[],"e581","M209.8 111.9c-8.9-9.9-24-10.7-33.9-1.8l-39.9 36L136 24c0-13.3-10.7-24-24-24S88 10.7 88 24l0 122.1-39.9-36c-9.9-8.9-25-8.1-33.9 1.8s-8.1 25 1.8 33.9l80 72c9.1 8.2 23 8.2 32.1 0l80-72c9.9-8.9 10.7-24 1.8-33.9zm352 0c-8.9-9.9-24-10.7-33.9-1.8l-39.9 36V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V146.1l-39.9-36c-9.9-8.9-25-8.1-33.9 1.8s-8.1 25 1.8 33.9l80 72c9.1 8.2 23 8.2 32.1 0l80-72c9.9-8.9 10.7-24 1.8-33.9zM307.4 166.5c-11.5-8.7-27.3-8.7-38.8 0l-168 128c-6.6 5-11 12.5-12.3 20.7l-24 160c-1.4 9.2 1.3 18.6 7.4 25.6S86.7 512 96 512H240h16c17.7 0 32-14.3 32-32V361.9c0-5.5 4.4-9.9 9.9-9.9c3.7 0 7.2 2.1 8.8 5.5l68.4 136.8c5.4 10.8 16.5 17.7 28.6 17.7H464h16c9.3 0 18.2-4.1 24.2-11.1s8.8-16.4 7.4-25.6l-24-160c-1.2-8.2-5.6-15.7-12.3-20.7l-168-128z"],exclamation:[64,512,[10069,10071,61738],"21","M64 64c0-17.7-14.3-32-32-32S0 46.3 0 64V320c0 17.7 14.3 32 32 32s32-14.3 32-32V64zM32 480a40 40 0 1 0 0-80 40 40 0 1 0 0 80z"],"arrows-spin":[512,512,[],"e4bb","M256 96c38.4 0 73.7 13.5 101.3 36.1l-32.6 32.6c-4.6 4.6-5.9 11.5-3.5 17.4s8.3 9.9 14.8 9.9H448c8.8 0 16-7.2 16-16V64c0-6.5-3.9-12.3-9.9-14.8s-12.9-1.1-17.4 3.5l-34 34C363.4 52.6 312.1 32 256 32c-10.9 0-21.5 .8-32 2.3V99.2c10.3-2.1 21-3.2 32-3.2zM132.1 154.7l32.6 32.6c4.6 4.6 11.5 5.9 17.4 3.5s9.9-8.3 9.9-14.8V64c0-8.8-7.2-16-16-16H64c-6.5 0-12.3 3.9-14.8 9.9s-1.1 12.9 3.5 17.4l34 34C52.6 148.6 32 199.9 32 256c0 10.9 .8 21.5 2.3 32H99.2c-2.1-10.3-3.2-21-3.2-32c0-38.4 13.5-73.7 36.1-101.3zM477.7 224H412.8c2.1 10.3 3.2 21 3.2 32c0 38.4-13.5 73.7-36.1 101.3l-32.6-32.6c-4.6-4.6-11.5-5.9-17.4-3.5s-9.9 8.3-9.9 14.8V448c0 8.8 7.2 16 16 16H448c6.5 0 12.3-3.9 14.8-9.9s1.1-12.9-3.5-17.4l-34-34C459.4 363.4 480 312.1 480 256c0-10.9-.8-21.5-2.3-32zM256 416c-38.4 0-73.7-13.5-101.3-36.1l32.6-32.6c4.6-4.6 5.9-11.5 3.5-17.4s-8.3-9.9-14.8-9.9H64c-8.8 0-16 7.2-16 16l0 112c0 6.5 3.9 12.3 9.9 14.8s12.9 1.1 17.4-3.5l34-34C148.6 459.4 199.9 480 256 480c10.9 0 21.5-.8 32-2.3V412.8c-10.3 2.1-21 3.2-32 3.2z"],print:[512,512,[128424,128438,9113],"f02f","M128 0C92.7 0 64 28.7 64 64v96h64V64H354.7L384 93.3V160h64V93.3c0-17-6.7-33.3-18.7-45.3L400 18.7C388 6.7 371.7 0 354.7 0H128zM384 352v32 64H128V384 368 352H384zm64 32h32c17.7 0 32-14.3 32-32V256c0-35.3-28.7-64-64-64H64c-35.3 0-64 28.7-64 64v96c0 17.7 14.3 32 32 32H64v64c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V384zM432 248a24 24 0 1 1 0 48 24 24 0 1 1 0-48z"],"turkish-lira-sign":[384,512,["try","turkish-lira"],"e2bb","M96 32c17.7 0 32 14.3 32 32V99.3L247.2 65.2c17-4.9 34.7 5 39.6 22s-5 34.7-22 39.6L128 165.9v29.4l119.2-34.1c17-4.9 34.7 5 39.6 22s-5 34.7-22 39.6L128 261.9V416h63.8c68.2 0 124.4-53.5 127.8-121.6l.4-8c.9-17.7 15.9-31.2 33.6-30.4s31.2 15.9 30.4 33.6l-.4 8C378.5 399.8 294.1 480 191.8 480H96c-17.7 0-32-14.3-32-32V280.1l-23.2 6.6c-17 4.9-34.7-5-39.6-22s5-34.7 22-39.6L64 213.6V184.1l-23.2 6.6c-17 4.9-34.7-5-39.6-22s5-34.7 22-39.6L64 117.6V64c0-17.7 14.3-32 32-32z"],"dollar-sign":[320,512,[128178,61781,"dollar","usd"],"24","M160 0c17.7 0 32 14.3 32 32V67.7c1.6 .2 3.1 .4 4.7 .7c.4 .1 .7 .1 1.1 .2l48 8.8c17.4 3.2 28.9 19.9 25.7 37.2s-19.9 28.9-37.2 25.7l-47.5-8.7c-31.3-4.6-58.9-1.5-78.3 6.2s-27.2 18.3-29 28.1c-2 10.7-.5 16.7 1.2 20.4c1.8 3.9 5.5 8.3 12.8 13.2c16.3 10.7 41.3 17.7 73.7 26.3l2.9 .8c28.6 7.6 63.6 16.8 89.6 33.8c14.2 9.3 27.6 21.9 35.9 39.5c8.5 17.9 10.3 37.9 6.4 59.2c-6.9 38-33.1 63.4-65.6 76.7c-13.7 5.6-28.6 9.2-44.4 11V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V445.1c-.4-.1-.9-.1-1.3-.2l-.2 0 0 0c-24.4-3.8-64.5-14.3-91.5-26.3c-16.1-7.2-23.4-26.1-16.2-42.2s26.1-23.4 42.2-16.2c20.9 9.3 55.3 18.5 75.2 21.6c31.9 4.7 58.2 2 76-5.3c16.9-6.9 24.6-16.9 26.8-28.9c1.9-10.6 .4-16.7-1.3-20.4c-1.9-4-5.6-8.4-13-13.3c-16.4-10.7-41.5-17.7-74-26.3l-2.8-.7 0 0C119.4 279.3 84.4 270 58.4 253c-14.2-9.3-27.5-22-35.8-39.6c-8.4-17.9-10.1-37.9-6.1-59.2C23.7 116 52.3 91.2 84.8 78.3c13.3-5.3 27.9-8.9 43.2-11V32c0-17.7 14.3-32 32-32z"],x:[384,512,[120],"58","M376.6 84.5c11.3-13.6 9.5-33.8-4.1-45.1s-33.8-9.5-45.1 4.1L192 206 56.6 43.5C45.3 29.9 25.1 28.1 11.5 39.4S-3.9 70.9 7.4 84.5L150.3 256 7.4 427.5c-11.3 13.6-9.5 33.8 4.1 45.1s33.8 9.5 45.1-4.1L192 306 327.4 468.5c11.3 13.6 31.5 15.4 45.1 4.1s15.4-31.5 4.1-45.1L233.7 256 376.6 84.5z"],"magnifying-glass-dollar":[512,512,["search-dollar"],"f688","M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zM228 104c0-11-9-20-20-20s-20 9-20 20v14c-7.6 1.7-15.2 4.4-22.2 8.5c-13.9 8.3-25.9 22.8-25.8 43.9c.1 20.3 12 33.1 24.7 40.7c11 6.6 24.7 10.8 35.6 14l1.7 .5c12.6 3.8 21.8 6.8 28 10.7c5.1 3.2 5.8 5.4 5.9 8.2c.1 5-1.8 8-5.9 10.5c-5 3.1-12.9 5-21.4 4.7c-11.1-.4-21.5-3.9-35.1-8.5c-2.3-.8-4.7-1.6-7.2-2.4c-10.5-3.5-21.8 2.2-25.3 12.6s2.2 21.8 12.6 25.3c1.9 .6 4 1.3 6.1 2.1l0 0 0 0c8.3 2.9 17.9 6.2 28.2 8.4V312c0 11 9 20 20 20s20-9 20-20V298.2c8-1.7 16-4.5 23.2-9c14.3-8.9 25.1-24.1 24.8-45c-.3-20.3-11.7-33.4-24.6-41.6c-11.5-7.2-25.9-11.6-37.1-15l-.7-.2c-12.8-3.9-21.9-6.7-28.3-10.5c-5.2-3.1-5.3-4.9-5.3-6.7c0-3.7 1.4-6.5 6.2-9.3c5.4-3.2 13.6-5.1 21.5-5c9.6 .1 20.2 2.2 31.2 5.2c10.7 2.8 21.6-3.5 24.5-14.2s-3.5-21.6-14.2-24.5c-6.5-1.7-13.7-3.4-21.1-4.7V104z"],"users-gear":[640,512,["users-cog"],"f509","M144 160A80 80 0 1 0 144 0a80 80 0 1 0 0 160zm368 0A80 80 0 1 0 512 0a80 80 0 1 0 0 160zM0 298.7C0 310.4 9.6 320 21.3 320H234.7c.2 0 .4 0 .7 0c-26.6-23.5-43.3-57.8-43.3-96c0-7.6 .7-15 1.9-22.3c-13.6-6.3-28.7-9.7-44.6-9.7H106.7C47.8 192 0 239.8 0 298.7zM320 320c24 0 45.9-8.8 62.7-23.3c2.5-3.7 5.2-7.3 8-10.7c2.7-3.3 5.7-6.1 9-8.3C410 262.3 416 243.9 416 224c0-53-43-96-96-96s-96 43-96 96s43 96 96 96zm65.4 60.2c-10.3-5.9-18.1-16.2-20.8-28.2H261.3C187.7 352 128 411.7 128 485.3c0 14.7 11.9 26.7 26.7 26.7H455.2c-2.1-5.2-3.2-10.9-3.2-16.4v-3c-1.3-.7-2.7-1.5-4-2.3l-2.6 1.5c-16.8 9.7-40.5 8-54.7-9.7c-4.5-5.6-8.6-11.5-12.4-17.6l-.1-.2-.1-.2-2.4-4.1-.1-.2-.1-.2c-3.4-6.2-6.4-12.6-9-19.3c-8.2-21.2 2.2-42.6 19-52.3l2.7-1.5c0-.8 0-1.5 0-2.3s0-1.5 0-2.3l-2.7-1.5zM533.3 192H490.7c-15.9 0-31 3.5-44.6 9.7c1.3 7.2 1.9 14.7 1.9 22.3c0 17.4-3.5 33.9-9.7 49c2.5 .9 4.9 2 7.1 3.3l2.6 1.5c1.3-.8 2.6-1.6 4-2.3v-3c0-19.4 13.3-39.1 35.8-42.6c7.9-1.2 16-1.9 24.2-1.9s16.3 .6 24.2 1.9c22.5 3.5 35.8 23.2 35.8 42.6v3c1.3 .7 2.7 1.5 4 2.3l2.6-1.5c16.8-9.7 40.5-8 54.7 9.7c2.3 2.8 4.5 5.8 6.6 8.7c-2.1-57.1-49-102.7-106.6-102.7zm91.3 163.9c6.3-3.6 9.5-11.1 6.8-18c-2.1-5.5-4.6-10.8-7.4-15.9l-2.3-4c-3.1-5.1-6.5-9.9-10.2-14.5c-4.6-5.7-12.7-6.7-19-3l-2.9 1.7c-9.2 5.3-20.4 4-29.6-1.3s-16.1-14.5-16.1-25.1v-3.4c0-7.3-4.9-13.8-12.1-14.9c-6.5-1-13.1-1.5-19.9-1.5s-13.4 .5-19.9 1.5c-7.2 1.1-12.1 7.6-12.1 14.9v3.4c0 10.6-6.9 19.8-16.1 25.1s-20.4 6.6-29.6 1.3l-2.9-1.7c-6.3-3.6-14.4-2.6-19 3c-3.7 4.6-7.1 9.5-10.2 14.6l-2.3 3.9c-2.8 5.1-5.3 10.4-7.4 15.9c-2.6 6.8 .5 14.3 6.8 17.9l2.9 1.7c9.2 5.3 13.7 15.8 13.7 26.4s-4.5 21.1-13.7 26.4l-3 1.7c-6.3 3.6-9.5 11.1-6.8 17.9c2.1 5.5 4.6 10.7 7.4 15.8l2.4 4.1c3 5.1 6.4 9.9 10.1 14.5c4.6 5.7 12.7 6.7 19 3l2.9-1.7c9.2-5.3 20.4-4 29.6 1.3s16.1 14.5 16.1 25.1v3.4c0 7.3 4.9 13.8 12.1 14.9c6.5 1 13.1 1.5 19.9 1.5s13.4-.5 19.9-1.5c7.2-1.1 12.1-7.6 12.1-14.9v-3.4c0-10.6 6.9-19.8 16.1-25.1s20.4-6.6 29.6-1.3l2.9 1.7c6.3 3.6 14.4 2.6 19-3c3.7-4.6 7.1-9.4 10.1-14.5l2.4-4.2c2.8-5.1 5.3-10.3 7.4-15.8c2.6-6.8-.5-14.3-6.8-17.9l-3-1.7c-9.2-5.3-13.7-15.8-13.7-26.4s4.5-21.1 13.7-26.4l3-1.7zM472 384a40 40 0 1 1 80 0 40 40 0 1 1 -80 0z"],"person-military-pointing":[576,512,[],"e54a","M246.9 14.1C234 15.2 224 26 224 39c0 13.8 11.2 25 25 25H400c8.8 0 16-7.2 16-16V17.4C416 8 408 .7 398.7 1.4L246.9 14.1zM240 112c0 44.2 35.8 80 80 80s80-35.8 80-80c0-5.5-.6-10.8-1.6-16H241.6c-1 5.2-1.6 10.5-1.6 16zM72 224c-22.1 0-40 17.9-40 40s17.9 40 40 40H224v89.4L386.8 230.5c-13.3-4.3-27.3-6.5-41.6-6.5H240 72zm345.7 20.9L246.6 416H416V369.7l53.6 90.6c11.2 19 35.8 25.3 54.8 14.1s25.3-35.8 14.1-54.8L462.3 290.8c-11.2-18.9-26.6-34.5-44.6-45.9zM224 448v32c0 17.7 14.3 32 32 32H384c17.7 0 32-14.3 32-32V448H224z"],"building-columns":[512,512,["bank","institution","museum","university"],"f19c","M243.4 2.6l-224 96c-14 6-21.8 21-18.7 35.8S16.8 160 32 160v8c0 13.3 10.7 24 24 24H456c13.3 0 24-10.7 24-24v-8c15.2 0 28.3-10.7 31.3-25.6s-4.8-29.9-18.7-35.8l-224-96c-8-3.4-17.2-3.4-25.2 0zM128 224H64V420.3c-.6 .3-1.2 .7-1.8 1.1l-48 32c-11.7 7.8-17 22.4-12.9 35.9S17.9 512 32 512H480c14.1 0 26.5-9.2 30.6-22.7s-1.1-28.1-12.9-35.9l-48-32c-.6-.4-1.2-.7-1.8-1.1V224H384V416H344V224H280V416H232V224H168V416H128V224zM256 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64z"],umbrella:[576,512,[],"f0e9","M288 0c17.7 0 32 14.3 32 32V49.7C451.8 63.4 557.7 161 573.9 285.9c2 15.6-17.3 24.4-27.8 12.7C532.1 283 504.8 272 480 272c-38.7 0-71 27.5-78.4 64.1c-1.7 8.7-8.7 15.9-17.6 15.9s-15.8-7.2-17.6-15.9C359 299.5 326.7 272 288 272s-71 27.5-78.4 64.1c-1.7 8.7-8.7 15.9-17.6 15.9s-15.8-7.2-17.6-15.9C167 299.5 134.7 272 96 272c-24.8 0-52.1 11-66.1 26.7C19.4 310.4 .1 301.5 2.1 285.9C18.3 161 124.2 63.4 256 49.7V32c0-17.7 14.3-32 32-32zm0 304c12.3 0 23.5 4.6 32 12.2V430.6c0 45-36.5 81.4-81.4 81.4c-30.8 0-59-17.4-72.8-45l-2.3-4.7c-7.9-15.8-1.5-35 14.3-42.9s35-1.5 42.9 14.3l2.3 4.7c3 5.9 9 9.6 15.6 9.6c9.6 0 17.4-7.8 17.4-17.4V316.2c8.5-7.6 19.7-12.2 32-12.2z"],trowel:[512,512,[],"e589","M343.9 213.4L245.3 312l65.4 65.4c7.9 7.9 11.1 19.4 8.4 30.3s-10.8 19.6-21.5 22.9l-256 80c-11.4 3.5-23.8 .5-32.2-7.9S-2.1 481.8 1.5 470.5l80-256c3.3-10.7 12-18.9 22.9-21.5s22.4 .5 30.3 8.4L200 266.7l98.6-98.6c-14.3-14.6-14.2-38 .3-52.5l95.4-95.4c26.9-26.9 70.5-26.9 97.5 0s26.9 70.5 0 97.5l-95.4 95.4c-14.5 14.5-37.9 14.6-52.5 .3z"],d:[384,512,[100],"44","M0 96C0 60.7 28.7 32 64 32h96c123.7 0 224 100.3 224 224s-100.3 224-224 224H64c-35.3 0-64-28.7-64-64V96zm160 0H64V416h96c88.4 0 160-71.6 160-160s-71.6-160-160-160z"],stapler:[640,512,[],"e5af","M640 299.3V304 432c0 26.5-21.5 48-48 48H512 448 64c-17.7 0-32-14.3-32-32s14.3-32 32-32H448V368H96c-17.7 0-32-14.3-32-32V219.4L33.8 214C14.2 210.5 0 193.5 0 173.7c0-8.9 2.9-17.5 8.2-24.6l35.6-47.5C76.7 57.8 128.2 32 182.9 32c27 0 53.6 6.3 77.8 18.4L586.9 213.5C619.5 229.7 640 263 640 299.3zM448 304V288L128 230.9V304H448z"],"masks-theater":[640,512,[127917,"theater-masks"],"f630","M74.6 373.2c41.7 36.1 108 82.5 166.1 73.7c6.1-.9 12.1-2.5 18-4.5c-9.2-12.3-17.3-24.4-24.2-35.4c-21.9-35-28.8-75.2-25.9-113.6c-20.6 4.1-39.2 13-54.7 25.4c-6.5 5.2-16.3 1.3-14.8-7c6.4-33.5 33-60.9 68.2-66.3c2.6-.4 5.3-.7 7.9-.8l19.4-131.3c2-13.8 8-32.7 25-45.9C278.2 53.2 310.5 37 363.2 32.2c-.8-.7-1.6-1.4-2.4-2.1C340.6 14.5 288.4-11.5 175.7 5.6S20.5 63 5.7 83.9C0 91.9-.8 102 .6 111.8L24.8 276.1c5.5 37.3 21.5 72.6 49.8 97.2zm87.7-219.6c4.4-3.1 10.8-2 11.8 3.3c.1 .5 .2 1.1 .3 1.6c3.2 21.8-11.6 42-33.1 45.3s-41.5-11.8-44.7-33.5c-.1-.5-.1-1.1-.2-1.6c-.6-5.4 5.2-8.4 10.3-6.7c9 3 18.8 3.9 28.7 2.4s19.1-5.3 26.8-10.8zM261.6 390c29.4 46.9 79.5 110.9 137.6 119.7s124.5-37.5 166.1-73.7c28.3-24.5 44.3-59.8 49.8-97.2l24.2-164.3c1.4-9.8 .6-19.9-5.1-27.9c-14.8-20.9-57.3-61.2-170-78.3S299.4 77.2 279.2 92.8c-7.8 6-11.5 15.4-12.9 25.2L242.1 282.3c-5.5 37.3-.4 75.8 19.6 107.7zM404.5 235.3c-7.7-5.5-16.8-9.3-26.8-10.8s-19.8-.6-28.7 2.4c-5.1 1.7-10.9-1.3-10.3-6.7c.1-.5 .1-1.1 .2-1.6c3.2-21.8 23.2-36.8 44.7-33.5s36.3 23.5 33.1 45.3c-.1 .5-.2 1.1-.3 1.6c-1 5.3-7.4 6.4-11.8 3.3zm136.2 15.5c-1 5.3-7.4 6.4-11.8 3.3c-7.7-5.5-16.8-9.3-26.8-10.8s-19.8-.6-28.7 2.4c-5.1 1.7-10.9-1.3-10.3-6.7c.1-.5 .1-1.1 .2-1.6c3.2-21.8 23.2-36.8 44.7-33.5s36.3 23.5 33.1 45.3c-.1 .5-.2 1.1-.3 1.6zM530 350.2c-19.6 44.7-66.8 72.5-116.8 64.9s-87.1-48.2-93-96.7c-1-8.3 8.9-12.1 15.2-6.7c23.9 20.8 53.6 35.3 87 40.3s66.1 .1 94.9-12.8c7.6-3.4 16 3.2 12.6 10.9z"],"kip-sign":[384,512,[],"e1c4","M340.8 88.3c13.4-11.5 15-31.7 3.5-45.1s-31.7-15-45.1-3.5L128 186.4V64c0-17.7-14.3-32-32-32S64 46.3 64 64V224H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64V448c0 17.7 14.3 32 32 32s32-14.3 32-32V325.6L299.2 472.3c13.4 11.5 33.6 9.9 45.1-3.5s9.9-33.6-3.5-45.1L182.5 288H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H182.5L340.8 88.3z"],"hand-point-left":[512,512,[],"f0a5","M32 96C14.3 96 0 110.3 0 128s14.3 32 32 32l208 0 0-64L32 96zM192 288c-17.7 0-32 14.3-32 32s14.3 32 32 32l64 0c17.7 0 32-14.3 32-32s-14.3-32-32-32l-64 0zm-64-64c0 17.7 14.3 32 32 32l48 0c17.7 0 32-14.3 32-32s-14.3-32-32-32l-48 0c-17.7 0-32 14.3-32 32zm96 160c-17.7 0-32 14.3-32 32s14.3 32 32 32l64 0c17.7 0 32-14.3 32-32s-14.3-32-32-32l-64 0zm88-96l-.6 0c5.4 9.4 8.6 20.3 8.6 32c0 13.2-4 25.4-10.8 35.6c24.9 8.7 42.8 32.5 42.8 60.4c0 11.7-3.1 22.6-8.6 32l8.6 0c88.4 0 160-71.6 160-160l0-61.7c0-42.4-16.9-83.1-46.9-113.1l-11.6-11.6C429.5 77.5 396.9 64 363 64l-27 0c-35.3 0-64 28.7-64 64l0 88c0 22.1 17.9 40 40 40s40-17.9 40-40l0-56c0-8.8 7.2-16 16-16s16 7.2 16 16l0 56c0 39.8-32.2 72-72 72z"],"handshake-simple":[640,512,[129309,"handshake-alt"],"f4c6","M323.4 85.2l-96.8 78.4c-16.1 13-19.2 36.4-7 53.1c12.9 17.8 38 21.3 55.3 7.8l99.3-77.2c7-5.4 17-4.2 22.5 2.8s4.2 17-2.8 22.5l-20.9 16.2L550.2 352H592c26.5 0 48-21.5 48-48V176c0-26.5-21.5-48-48-48H516h-4-.7l-3.9-2.5L434.8 79c-15.3-9.8-33.2-15-51.4-15c-21.8 0-43 7.5-60 21.2zm22.8 124.4l-51.7 40.2C263 274.4 217.3 268 193.7 235.6c-22.2-30.5-16.6-73.1 12.7-96.8l83.2-67.3c-11.6-4.9-24.1-7.4-36.8-7.4C234 64 215.7 69.6 200 80l-72 48H48c-26.5 0-48 21.5-48 48V304c0 26.5 21.5 48 48 48H156.2l91.4 83.4c19.6 17.9 49.9 16.5 67.8-3.1c5.5-6.1 9.2-13.2 11.1-20.6l17 15.6c19.5 17.9 49.9 16.6 67.8-2.9c4.5-4.9 7.8-10.6 9.9-16.5c19.4 13 45.8 10.3 62.1-7.5c17.9-19.5 16.6-49.9-2.9-67.8l-134.2-123z"],"jet-fighter":[640,512,["fighter-jet"],"f0fb","M160 24c0-13.3 10.7-24 24-24H296c13.3 0 24 10.7 24 24s-10.7 24-24 24H280L384 192H500.4c7.7 0 15.3 1.4 22.5 4.1L625 234.4c9 3.4 15 12 15 21.6s-6 18.2-15 21.6L522.9 315.9c-7.2 2.7-14.8 4.1-22.5 4.1H384L280 464h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H184c-13.3 0-24-10.7-24-24s10.7-24 24-24h8V320H160l-54.6 54.6c-6 6-14.1 9.4-22.6 9.4H64c-17.7 0-32-14.3-32-32V288c-17.7 0-32-14.3-32-32s14.3-32 32-32V160c0-17.7 14.3-32 32-32H82.7c8.5 0 16.6 3.4 22.6 9.4L160 192h32V48h-8c-13.3 0-24-10.7-24-24zM80 240c-8.8 0-16 7.2-16 16s7.2 16 16 16h64c8.8 0 16-7.2 16-16s-7.2-16-16-16H80z"],"square-share-nodes":[448,512,["share-alt-square"],"f1e1","M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM384 160c0 35.3-28.7 64-64 64c-15.4 0-29.5-5.4-40.6-14.5L194.1 256l85.3 46.5c11-9.1 25.2-14.5 40.6-14.5c35.3 0 64 28.7 64 64s-28.7 64-64 64s-64-28.7-64-64c0-2.5 .1-4.9 .4-7.3L174.5 300c-11.7 12.3-28.2 20-46.5 20c-35.3 0-64-28.7-64-64s28.7-64 64-64c18.3 0 34.8 7.7 46.5 20l81.9-44.7c-.3-2.4-.4-4.9-.4-7.3c0-35.3 28.7-64 64-64s64 28.7 64 64z"],barcode:[512,512,[],"f02a","M24 32C10.7 32 0 42.7 0 56V456c0 13.3 10.7 24 24 24H40c13.3 0 24-10.7 24-24V56c0-13.3-10.7-24-24-24H24zm88 0c-8.8 0-16 7.2-16 16V464c0 8.8 7.2 16 16 16s16-7.2 16-16V48c0-8.8-7.2-16-16-16zm72 0c-13.3 0-24 10.7-24 24V456c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V56c0-13.3-10.7-24-24-24H184zm96 0c-13.3 0-24 10.7-24 24V456c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V56c0-13.3-10.7-24-24-24H280zM448 56V456c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V56c0-13.3-10.7-24-24-24H472c-13.3 0-24 10.7-24 24zm-64-8V464c0 8.8 7.2 16 16 16s16-7.2 16-16V48c0-8.8-7.2-16-16-16s-16 7.2-16 16z"],"plus-minus":[384,512,[],"e43c","M224 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V144H48c-17.7 0-32 14.3-32 32s14.3 32 32 32H160V320c0 17.7 14.3 32 32 32s32-14.3 32-32V208H336c17.7 0 32-14.3 32-32s-14.3-32-32-32H224V32zM0 480c0 17.7 14.3 32 32 32H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H32c-17.7 0-32 14.3-32 32z"],video:[576,512,["video-camera"],"f03d","M0 128C0 92.7 28.7 64 64 64H320c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128zM559.1 99.8c10.4 5.6 16.9 16.4 16.9 28.2V384c0 11.8-6.5 22.6-16.9 28.2s-23 5-32.9-1.6l-96-64L416 337.1V320 192 174.9l14.2-9.5 96-64c9.8-6.5 22.4-7.2 32.9-1.6z"],"graduation-cap":[640,512,[127891,"mortar-board"],"f19d","M320 32c-8.1 0-16.1 1.4-23.7 4.1L15.8 137.4C6.3 140.9 0 149.9 0 160s6.3 19.1 15.8 22.6l57.9 20.9C57.3 229.3 48 259.8 48 291.9v28.1c0 28.4-10.8 57.7-22.3 80.8c-6.5 13-13.9 25.8-22.5 37.6C0 442.7-.9 448.3 .9 453.4s6 8.9 11.2 10.2l64 16c4.2 1.1 8.7 .3 12.4-2s6.3-6.1 7.1-10.4c8.6-42.8 4.3-81.2-2.1-108.7C90.3 344.3 86 329.8 80 316.5V291.9c0-30.2 10.2-58.7 27.9-81.5c12.9-15.5 29.6-28 49.2-35.7l157-61.7c8.2-3.2 17.5 .8 20.7 9s-.8 17.5-9 20.7l-157 61.7c-12.4 4.9-23.3 12.4-32.2 21.6l159.6 57.6c7.6 2.7 15.6 4.1 23.7 4.1s16.1-1.4 23.7-4.1L624.2 182.6c9.5-3.4 15.8-12.5 15.8-22.6s-6.3-19.1-15.8-22.6L343.7 36.1C336.1 33.4 328.1 32 320 32zM128 408c0 35.3 86 72 192 72s192-36.7 192-72L496.7 262.6 354.5 314c-11.1 4-22.8 6-34.5 6s-23.5-2-34.5-6L143.3 262.6 128 408z"],"hand-holding-medical":[576,512,[],"e05c","M224 24V80H168c-13.3 0-24 10.7-24 24v48c0 13.3 10.7 24 24 24h56v56c0 13.3 10.7 24 24 24h48c13.3 0 24-10.7 24-24V176h56c13.3 0 24-10.7 24-24V104c0-13.3-10.7-24-24-24H320V24c0-13.3-10.7-24-24-24H248c-13.3 0-24 10.7-24 24zM559.7 392.2c17.8-13.1 21.6-38.1 8.5-55.9s-38.1-21.6-55.9-8.5L392.6 416H272c-8.8 0-16-7.2-16-16s7.2-16 16-16h16 64c17.7 0 32-14.3 32-32s-14.3-32-32-32H288 272 193.7c-29.1 0-57.3 9.9-80 28L68.8 384H32c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H192 352.5c29 0 57.3-9.3 80.7-26.5l126.6-93.3zm-367-8.2l.9 0 0 0c-.3 0-.6 0-.9 0z"],"person-circle-check":[576,512,[],"e53e","M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l44.9 74.7c-16.1 17.6-28.6 38.5-36.6 61.5c-1.9-1.8-3.5-3.9-4.9-6.3L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152zm136 16a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm211.3-43.3c-6.2-6.2-16.4-6.2-22.6 0L416 385.4l-28.7-28.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l40 40c6.2 6.2 16.4 6.2 22.6 0l72-72c6.2-6.2 6.2-16.4 0-22.6z"],"turn-up":[384,512,[10548,"level-up-alt"],"f3bf","M350 177.5c3.8-8.8 2-19-4.6-26l-136-144C204.9 2.7 198.6 0 192 0s-12.9 2.7-17.4 7.5l-136 144c-6.6 7-8.4 17.2-4.6 26s12.5 14.5 22 14.5h88l0 192c0 17.7-14.3 32-32 32H32c-17.7 0-32 14.3-32 32v32c0 17.7 14.3 32 32 32l80 0c70.7 0 128-57.3 128-128l0-192h88c9.6 0 18.2-5.7 22-14.5z"]};!function(c){try{for(var l=arguments.length,s=new Array(1<l?l-1:0),a=1;a<l;a++)s[a-1]=arguments[a];c.apply(void 0,s)}catch(c){if(!M)throw c}}((function(){d("fas",p),d("fa-solid",p)}))}(),function(){"use strict";function c(c,l){var s,a=Object.keys(c);return Object.getOwnPropertySymbols&&(s=Object.getOwnPropertySymbols(c),l&&(s=s.filter((function(l){return Object.getOwnPropertyDescriptor(c,l).enumerable}))),a.push.apply(a,s)),a}function l(l){for(var s=1;s<arguments.length;s++){var z=null!=arguments[s]?arguments[s]:{};s%2?c(Object(z),!0).forEach((function(c){a(l,c,z[c])})):Object.getOwnPropertyDescriptors?Object.defineProperties(l,Object.getOwnPropertyDescriptors(z)):c(Object(z)).forEach((function(c){Object.defineProperty(l,c,Object.getOwnPropertyDescriptor(z,c))}))}return l}function s(c){return(s="function"==typeof Symbol&&"symbol"==typeof Symbol.iterator?function(c){return typeof c}:function(c){return c&&"function"==typeof Symbol&&c.constructor===Symbol&&c!==Symbol.prototype?"symbol":typeof c})(c)}function a(c,l,s){return l in c?Object.defineProperty(c,l,{value:s,enumerable:!0,configurable:!0,writable:!0}):c[l]=s,c}function z(c,l){return function(c){if(Array.isArray(c))return c}(c)||function(c,l){var s=null==c?null:"undefined"!=typeof Symbol&&c[Symbol.iterator]||c["@@iterator"];if(null!=s){var a,z,e=[],H=!0,t=!1;try{for(s=s.call(c);!(H=(a=s.next()).done)&&(e.push(a.value),!l||e.length!==l);H=!0);}catch(c){t=!0,z=c}finally{try{H||null==s.return||s.return()}finally{if(t)throw z}}return e}}(c,l)||H(c,l)||function(){throw new TypeError("Invalid attempt to destructure non-iterable instance.\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method.")}()}function e(c){return function(c){if(Array.isArray(c))return t(c)}(c)||function(c){if("undefined"!=typeof Symbol&&null!=c[Symbol.iterator]||null!=c["@@iterator"])return Array.from(c)}(c)||H(c)||function(){throw new TypeError("Invalid attempt to spread non-iterable instance.\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method.")}()}function H(c,l){if(c){if("string"==typeof c)return t(c,l);var s=Object.prototype.toString.call(c).slice(8,-1);return"Map"===(s="Object"===s&&c.constructor?c.constructor.name:s)||"Set"===s?Array.from(c):"Arguments"===s||/^(?:Ui|I)nt(?:8|16|32)(?:Clamped)?Array$/.test(s)?t(c,l):void 0}}function t(c,l){(null==l||l>c.length)&&(l=c.length);for(var s=0,a=new Array(l);s<l;s++)a[s]=c[s];return a}function V(){}var r={},M={},h=null,n={mark:V,measure:V};try{"undefined"!=typeof window&&(r=window),"undefined"!=typeof document&&(M=document),"undefined"!=typeof MutationObserver&&(h=MutationObserver),"undefined"!=typeof performance&&(n=performance)}catch(V){}var i=void 0===(C=(r.navigator||{}).userAgent)?"":C,m=r,o=M,f=h,C=n,v=!!m.document,L=!!o.documentElement&&!!o.head&&"function"==typeof o.addEventListener&&"function"==typeof o.createElement,u=~i.indexOf("MSIE")||~i.indexOf("Trident/"),d=(r="___FONT_AWESOME___","svg-inline--fa"),p="data-fa-i2svg",b="data-fa-pseudo-element",g="data-fa-pseudo-element-pending",w="data-prefix",k="data-icon",y="fontawesome-i2svg",S="async",A=["HTML","HEAD","STYLE","SCRIPT"],x=function(){try{return!0}catch(c){return!1}}(),q="classic",Z="sharp",O=[q,Z];function j(c){return new Proxy(c,{get:function(c,l){return l in c?c[l]:c[q]}})}var P=j((a(M={},q,{fa:"solid",fas:"solid","fa-solid":"solid",far:"regular","fa-regular":"regular",fal:"light","fa-light":"light",fat:"thin","fa-thin":"thin",fad:"duotone","fa-duotone":"duotone",fab:"brands","fa-brands":"brands",fak:"kit",fakd:"kit","fa-kit":"kit","fa-kit-duotone":"kit"}),a(M,Z,{fa:"solid",fass:"solid","fa-solid":"solid",fasr:"regular","fa-regular":"regular",fasl:"light","fa-light":"light",fast:"thin","fa-thin":"thin"}),M)),N=j((a(h={},q,{solid:"fas",regular:"far",light:"fal",thin:"fat",duotone:"fad",brands:"fab",kit:"fak"}),a(h,Z,{solid:"fass",regular:"fasr",light:"fasl",thin:"fast"}),h)),E=j((a(n={},q,{fab:"fa-brands",fad:"fa-duotone",fak:"fa-kit",fal:"fa-light",far:"fa-regular",fas:"fa-solid",fat:"fa-thin"}),a(n,Z,{fass:"fa-solid",fasr:"fa-regular",fasl:"fa-light",fast:"fa-thin"}),n)),I=j((a(i={},q,{"fa-brands":"fab","fa-duotone":"fad","fa-kit":"fak","fa-light":"fal","fa-regular":"far","fa-solid":"fas","fa-thin":"fat"}),a(i,Z,{"fa-solid":"fass","fa-regular":"fasr","fa-light":"fasl","fa-thin":"fast"}),i)),T=/fa(s|r|l|t|d|b|k|ss|sr|sl|st)?[\-\ ]/,D="fa-layers-text",Y=/Font ?Awesome ?([56 ]*)(Solid|Regular|Light|Thin|Duotone|Brands|Free|Pro|Sharp|Kit)?.*/i,R=j((a(M={},q,{900:"fas",400:"far",normal:"far",300:"fal",100:"fat"}),a(M,Z,{900:"fass",400:"fasr",300:"fasl",100:"fast"}),M)),_=(n=(h=[1,2,3,4,5,6,7,8,9,10]).concat([11,12,13,14,15,16,17,18,19,20]),["class","data-prefix","data-icon","data-fa-transform","data-fa-mask"]),F={GROUP:"duotone-group",SWAP_OPACITY:"swap-opacity",PRIMARY:"primary",SECONDARY:"secondary"};i=new Set,Object.keys(N[q]).map(i.add.bind(i)),Object.keys(N[Z]).map(i.add.bind(i));var W=[].concat(O,e(i),["2xs","xs","sm","lg","xl","2xl","beat","border","fade","beat-fade","bounce","flip-both","flip-horizontal","flip-vertical","flip","fw","inverse","layers-counter","layers-text","layers","li","pull-left","pull-right","pulse","rotate-180","rotate-270","rotate-90","rotate-by","shake","spin-pulse","spin-reverse","spin","stack-1x","stack-2x","stack","ul",F.GROUP,F.SWAP_OPACITY,F.PRIMARY,F.SECONDARY]).concat(h.map((function(c){return"".concat(c,"x")}))).concat(n.map((function(c){return"w-".concat(c)}))),U=m.FontAwesomeConfig||{};o&&"function"==typeof o.querySelector&&[["data-family-prefix","familyPrefix"],["data-css-prefix","cssPrefix"],["data-family-default","familyDefault"],["data-style-default","styleDefault"],["data-replacement-class","replacementClass"],["data-auto-replace-svg","autoReplaceSvg"],["data-auto-add-css","autoAddCss"],["data-auto-a11y","autoA11y"],["data-search-pseudo-elements","searchPseudoElements"],["data-observe-mutations","observeMutations"],["data-mutate-approach","mutateApproach"],["data-keep-original-source","keepOriginalSource"],["data-measure-performance","measurePerformance"],["data-show-missing-icons","showMissingIcons"]].forEach((function(c){c=(l=z(c,2))[0];var l=l[1];c=""===(c=function(c){var l=o.querySelector("script["+c+"]");if(l)return l.getAttribute(c)}(c))||"false"!==c&&("true"===c||c),null!=c&&(U[l]=c)})),M={styleDefault:"solid",familyDefault:"classic",cssPrefix:"fa",replacementClass:d,autoReplaceSvg:!0,autoAddCss:!0,autoA11y:!0,searchPseudoElements:!1,observeMutations:!0,mutateApproach:"async",keepOriginalSource:!0,measurePerformance:!1,showMissingIcons:!0},U.familyPrefix&&(U.cssPrefix=U.familyPrefix);var X=l(l({},M),U);X.autoReplaceSvg||(X.observeMutations=!1);var B={};Object.keys(M).forEach((function(c){Object.defineProperty(B,c,{enumerable:!0,set:function(l){X[c]=l,G.forEach((function(c){return c(B)}))},get:function(){return X[c]}})})),Object.defineProperty(B,"familyPrefix",{enumerable:!0,set:function(c){X.cssPrefix=c,G.forEach((function(c){return c(B)}))},get:function(){return X.cssPrefix}}),m.FontAwesomeConfig=B;var G=[],Q=16,K={size:16,x:0,y:0,rotate:0,flipX:!1,flipY:!1},$="0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ";function J(){for(var c=12,l="";0<c--;)l+=$[62*Math.random()|0];return l}function cc(c){for(var l=[],s=(c||[]).length>>>0;s--;)l[s]=c[s];return l}function lc(c){return c.classList?cc(c.classList):(c.getAttribute("class")||"").split(" ").filter((function(c){return c}))}function sc(c){return"".concat(c).replace(/&/g,"&amp;").replace(/"/g,"&quot;").replace(/'/g,"&#39;").replace(/</g,"&lt;").replace(/>/g,"&gt;")}function ac(c){return Object.keys(c||{}).reduce((function(l,s){return l+"".concat(s,": ").concat(c[s].trim(),";")}),"")}function zc(c){return c.size!==K.size||c.x!==K.x||c.y!==K.y||c.rotate!==K.rotate||c.flipX||c.flipY}function ec(){var c,l,s=d,a=B.cssPrefix,z=B.replacementClass,e=':host,:root{--fa-font-solid:normal 900 1em/1 "Font Awesome 6 Solid";--fa-font-regular:normal 400 1em/1 "Font Awesome 6 Regular";--fa-font-light:normal 300 1em/1 "Font Awesome 6 Light";--fa-font-thin:normal 100 1em/1 "Font Awesome 6 Thin";--fa-font-duotone:normal 900 1em/1 "Font Awesome 6 Duotone";--fa-font-sharp-solid:normal 900 1em/1 "Font Awesome 6 Sharp";--fa-font-sharp-regular:normal 400 1em/1 "Font Awesome 6 Sharp";--fa-font-sharp-light:normal 300 1em/1 "Font Awesome 6 Sharp";--fa-font-sharp-thin:normal 100 1em/1 "Font Awesome 6 Sharp";--fa-font-brands:normal 400 1em/1 "Font Awesome 6 Brands"}svg:not(:host).svg-inline--fa,svg:not(:root).svg-inline--fa{overflow:visible;box-sizing:content-box}.svg-inline--fa{display:var(--fa-display,inline-block);height:1em;overflow:visible;vertical-align:-.125em}.svg-inline--fa.fa-2xs{vertical-align:.1em}.svg-inline--fa.fa-xs{vertical-align:0}.svg-inline--fa.fa-sm{vertical-align:-.0714285705em}.svg-inline--fa.fa-lg{vertical-align:-.2em}.svg-inline--fa.fa-xl{vertical-align:-.25em}.svg-inline--fa.fa-2xl{vertical-align:-.3125em}.svg-inline--fa.fa-pull-left{margin-right:var(--fa-pull-margin,.3em);width:auto}.svg-inline--fa.fa-pull-right{margin-left:var(--fa-pull-margin,.3em);width:auto}.svg-inline--fa.fa-li{width:var(--fa-li-width,2em);top:.25em}.svg-inline--fa.fa-fw{width:var(--fa-fw-width,1.25em)}.fa-layers svg.svg-inline--fa{bottom:0;left:0;margin:auto;position:absolute;right:0;top:0}.fa-layers-counter,.fa-layers-text{display:inline-block;position:absolute;text-align:center}.fa-layers{display:inline-block;height:1em;position:relative;text-align:center;vertical-align:-.125em;width:1em}.fa-layers svg.svg-inline--fa{-webkit-transform-origin:center center;transform-origin:center center}.fa-layers-text{left:50%;top:50%;-webkit-transform:translate(-50%,-50%);transform:translate(-50%,-50%);-webkit-transform-origin:center center;transform-origin:center center}.fa-layers-counter{background-color:var(--fa-counter-background-color,#ff253a);border-radius:var(--fa-counter-border-radius,1em);box-sizing:border-box;color:var(--fa-inverse,#fff);line-height:var(--fa-counter-line-height,1);max-width:var(--fa-counter-max-width,5em);min-width:var(--fa-counter-min-width,1.5em);overflow:hidden;padding:var(--fa-counter-padding,.25em .5em);right:var(--fa-right,0);text-overflow:ellipsis;top:var(--fa-top,0);-webkit-transform:scale(var(--fa-counter-scale,.25));transform:scale(var(--fa-counter-scale,.25));-webkit-transform-origin:top right;transform-origin:top right}.fa-layers-bottom-right{bottom:var(--fa-bottom,0);right:var(--fa-right,0);top:auto;-webkit-transform:scale(var(--fa-layers-scale,.25));transform:scale(var(--fa-layers-scale,.25));-webkit-transform-origin:bottom right;transform-origin:bottom right}.fa-layers-bottom-left{bottom:var(--fa-bottom,0);left:var(--fa-left,0);right:auto;top:auto;-webkit-transform:scale(var(--fa-layers-scale,.25));transform:scale(var(--fa-layers-scale,.25));-webkit-transform-origin:bottom left;transform-origin:bottom left}.fa-layers-top-right{top:var(--fa-top,0);right:var(--fa-right,0);-webkit-transform:scale(var(--fa-layers-scale,.25));transform:scale(var(--fa-layers-scale,.25));-webkit-transform-origin:top right;transform-origin:top right}.fa-layers-top-left{left:var(--fa-left,0);right:auto;top:var(--fa-top,0);-webkit-transform:scale(var(--fa-layers-scale,.25));transform:scale(var(--fa-layers-scale,.25));-webkit-transform-origin:top left;transform-origin:top left}.fa-1x{font-size:1em}.fa-2x{font-size:2em}.fa-3x{font-size:3em}.fa-4x{font-size:4em}.fa-5x{font-size:5em}.fa-6x{font-size:6em}.fa-7x{font-size:7em}.fa-8x{font-size:8em}.fa-9x{font-size:9em}.fa-10x{font-size:10em}.fa-2xs{font-size:.625em;line-height:.1em;vertical-align:.225em}.fa-xs{font-size:.75em;line-height:.0833333337em;vertical-align:.125em}.fa-sm{font-size:.875em;line-height:.0714285718em;vertical-align:.0535714295em}.fa-lg{font-size:1.25em;line-height:.05em;vertical-align:-.075em}.fa-xl{font-size:1.5em;line-height:.0416666682em;vertical-align:-.125em}.fa-2xl{font-size:2em;line-height:.03125em;vertical-align:-.1875em}.fa-fw{text-align:center;width:1.25em}.fa-ul{list-style-type:none;margin-left:var(--fa-li-margin,2.5em);padding-left:0}.fa-ul>li{position:relative}.fa-li{left:calc(var(--fa-li-width,2em) * -1);position:absolute;text-align:center;width:var(--fa-li-width,2em);line-height:inherit}.fa-border{border-color:var(--fa-border-color,#eee);border-radius:var(--fa-border-radius,.1em);border-style:var(--fa-border-style,solid);border-width:var(--fa-border-width,.08em);padding:var(--fa-border-padding,.2em .25em .15em)}.fa-pull-left{float:left;margin-right:var(--fa-pull-margin,.3em)}.fa-pull-right{float:right;margin-left:var(--fa-pull-margin,.3em)}.fa-beat{-webkit-animation-name:fa-beat;animation-name:fa-beat;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,ease-in-out);animation-timing-function:var(--fa-animation-timing,ease-in-out)}.fa-bounce{-webkit-animation-name:fa-bounce;animation-name:fa-bounce;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,cubic-bezier(.28,.84,.42,1));animation-timing-function:var(--fa-animation-timing,cubic-bezier(.28,.84,.42,1))}.fa-fade{-webkit-animation-name:fa-fade;animation-name:fa-fade;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,cubic-bezier(.4,0,.6,1));animation-timing-function:var(--fa-animation-timing,cubic-bezier(.4,0,.6,1))}.fa-beat-fade{-webkit-animation-name:fa-beat-fade;animation-name:fa-beat-fade;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,cubic-bezier(.4,0,.6,1));animation-timing-function:var(--fa-animation-timing,cubic-bezier(.4,0,.6,1))}.fa-flip{-webkit-animation-name:fa-flip;animation-name:fa-flip;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,ease-in-out);animation-timing-function:var(--fa-animation-timing,ease-in-out)}.fa-shake{-webkit-animation-name:fa-shake;animation-name:fa-shake;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,linear);animation-timing-function:var(--fa-animation-timing,linear)}.fa-spin{-webkit-animation-name:fa-spin;animation-name:fa-spin;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,2s);animation-duration:var(--fa-animation-duration,2s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,linear);animation-timing-function:var(--fa-animation-timing,linear)}.fa-spin-reverse{--fa-animation-direction:reverse}.fa-pulse,.fa-spin-pulse{-webkit-animation-name:fa-spin;animation-name:fa-spin;-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,steps(8));animation-timing-function:var(--fa-animation-timing,steps(8))}@media (prefers-reduced-motion:reduce){.fa-beat,.fa-beat-fade,.fa-bounce,.fa-fade,.fa-flip,.fa-pulse,.fa-shake,.fa-spin,.fa-spin-pulse{-webkit-animation-delay:-1ms;animation-delay:-1ms;-webkit-animation-duration:1ms;animation-duration:1ms;-webkit-animation-iteration-count:1;animation-iteration-count:1;-webkit-transition-delay:0s;transition-delay:0s;-webkit-transition-duration:0s;transition-duration:0s}}@-webkit-keyframes fa-beat{0%,90%{-webkit-transform:scale(1);transform:scale(1)}45%{-webkit-transform:scale(var(--fa-beat-scale,1.25));transform:scale(var(--fa-beat-scale,1.25))}}@keyframes fa-beat{0%,90%{-webkit-transform:scale(1);transform:scale(1)}45%{-webkit-transform:scale(var(--fa-beat-scale,1.25));transform:scale(var(--fa-beat-scale,1.25))}}@-webkit-keyframes fa-bounce{0%{-webkit-transform:scale(1,1) translateY(0);transform:scale(1,1) translateY(0)}10%{-webkit-transform:scale(var(--fa-bounce-start-scale-x,1.1),var(--fa-bounce-start-scale-y,.9)) translateY(0);transform:scale(var(--fa-bounce-start-scale-x,1.1),var(--fa-bounce-start-scale-y,.9)) translateY(0)}30%{-webkit-transform:scale(var(--fa-bounce-jump-scale-x,.9),var(--fa-bounce-jump-scale-y,1.1)) translateY(var(--fa-bounce-height,-.5em));transform:scale(var(--fa-bounce-jump-scale-x,.9),var(--fa-bounce-jump-scale-y,1.1)) translateY(var(--fa-bounce-height,-.5em))}50%{-webkit-transform:scale(var(--fa-bounce-land-scale-x,1.05),var(--fa-bounce-land-scale-y,.95)) translateY(0);transform:scale(var(--fa-bounce-land-scale-x,1.05),var(--fa-bounce-land-scale-y,.95)) translateY(0)}57%{-webkit-transform:scale(1,1) translateY(var(--fa-bounce-rebound,-.125em));transform:scale(1,1) translateY(var(--fa-bounce-rebound,-.125em))}64%{-webkit-transform:scale(1,1) translateY(0);transform:scale(1,1) translateY(0)}100%{-webkit-transform:scale(1,1) translateY(0);transform:scale(1,1) translateY(0)}}@keyframes fa-bounce{0%{-webkit-transform:scale(1,1) translateY(0);transform:scale(1,1) translateY(0)}10%{-webkit-transform:scale(var(--fa-bounce-start-scale-x,1.1),var(--fa-bounce-start-scale-y,.9)) translateY(0);transform:scale(var(--fa-bounce-start-scale-x,1.1),var(--fa-bounce-start-scale-y,.9)) translateY(0)}30%{-webkit-transform:scale(var(--fa-bounce-jump-scale-x,.9),var(--fa-bounce-jump-scale-y,1.1)) translateY(var(--fa-bounce-height,-.5em));transform:scale(var(--fa-bounce-jump-scale-x,.9),var(--fa-bounce-jump-scale-y,1.1)) translateY(var(--fa-bounce-height,-.5em))}50%{-webkit-transform:scale(var(--fa-bounce-land-scale-x,1.05),var(--fa-bounce-land-scale-y,.95)) translateY(0);transform:scale(var(--fa-bounce-land-scale-x,1.05),var(--fa-bounce-land-scale-y,.95)) translateY(0)}57%{-webkit-transform:scale(1,1) translateY(var(--fa-bounce-rebound,-.125em));transform:scale(1,1) translateY(var(--fa-bounce-rebound,-.125em))}64%{-webkit-transform:scale(1,1) translateY(0);transform:scale(1,1) translateY(0)}100%{-webkit-transform:scale(1,1) translateY(0);transform:scale(1,1) translateY(0)}}@-webkit-keyframes fa-fade{50%{opacity:var(--fa-fade-opacity,.4)}}@keyframes fa-fade{50%{opacity:var(--fa-fade-opacity,.4)}}@-webkit-keyframes fa-beat-fade{0%,100%{opacity:var(--fa-beat-fade-opacity,.4);-webkit-transform:scale(1);transform:scale(1)}50%{opacity:1;-webkit-transform:scale(var(--fa-beat-fade-scale,1.125));transform:scale(var(--fa-beat-fade-scale,1.125))}}@keyframes fa-beat-fade{0%,100%{opacity:var(--fa-beat-fade-opacity,.4);-webkit-transform:scale(1);transform:scale(1)}50%{opacity:1;-webkit-transform:scale(var(--fa-beat-fade-scale,1.125));transform:scale(var(--fa-beat-fade-scale,1.125))}}@-webkit-keyframes fa-flip{50%{-webkit-transform:rotate3d(var(--fa-flip-x,0),var(--fa-flip-y,1),var(--fa-flip-z,0),var(--fa-flip-angle,-180deg));transform:rotate3d(var(--fa-flip-x,0),var(--fa-flip-y,1),var(--fa-flip-z,0),var(--fa-flip-angle,-180deg))}}@keyframes fa-flip{50%{-webkit-transform:rotate3d(var(--fa-flip-x,0),var(--fa-flip-y,1),var(--fa-flip-z,0),var(--fa-flip-angle,-180deg));transform:rotate3d(var(--fa-flip-x,0),var(--fa-flip-y,1),var(--fa-flip-z,0),var(--fa-flip-angle,-180deg))}}@-webkit-keyframes fa-shake{0%{-webkit-transform:rotate(-15deg);transform:rotate(-15deg)}4%{-webkit-transform:rotate(15deg);transform:rotate(15deg)}24%,8%{-webkit-transform:rotate(-18deg);transform:rotate(-18deg)}12%,28%{-webkit-transform:rotate(18deg);transform:rotate(18deg)}16%{-webkit-transform:rotate(-22deg);transform:rotate(-22deg)}20%{-webkit-transform:rotate(22deg);transform:rotate(22deg)}32%{-webkit-transform:rotate(-12deg);transform:rotate(-12deg)}36%{-webkit-transform:rotate(12deg);transform:rotate(12deg)}100%,40%{-webkit-transform:rotate(0);transform:rotate(0)}}@keyframes fa-shake{0%{-webkit-transform:rotate(-15deg);transform:rotate(-15deg)}4%{-webkit-transform:rotate(15deg);transform:rotate(15deg)}24%,8%{-webkit-transform:rotate(-18deg);transform:rotate(-18deg)}12%,28%{-webkit-transform:rotate(18deg);transform:rotate(18deg)}16%{-webkit-transform:rotate(-22deg);transform:rotate(-22deg)}20%{-webkit-transform:rotate(22deg);transform:rotate(22deg)}32%{-webkit-transform:rotate(-12deg);transform:rotate(-12deg)}36%{-webkit-transform:rotate(12deg);transform:rotate(12deg)}100%,40%{-webkit-transform:rotate(0);transform:rotate(0)}}@-webkit-keyframes fa-spin{0%{-webkit-transform:rotate(0);transform:rotate(0)}100%{-webkit-transform:rotate(360deg);transform:rotate(360deg)}}@keyframes fa-spin{0%{-webkit-transform:rotate(0);transform:rotate(0)}100%{-webkit-transform:rotate(360deg);transform:rotate(360deg)}}.fa-rotate-90{-webkit-transform:rotate(90deg);transform:rotate(90deg)}.fa-rotate-180{-webkit-transform:rotate(180deg);transform:rotate(180deg)}.fa-rotate-270{-webkit-transform:rotate(270deg);transform:rotate(270deg)}.fa-flip-horizontal{-webkit-transform:scale(-1,1);transform:scale(-1,1)}.fa-flip-vertical{-webkit-transform:scale(1,-1);transform:scale(1,-1)}.fa-flip-both,.fa-flip-horizontal.fa-flip-vertical{-webkit-transform:scale(-1,-1);transform:scale(-1,-1)}.fa-rotate-by{-webkit-transform:rotate(var(--fa-rotate-angle,0));transform:rotate(var(--fa-rotate-angle,0))}.fa-stack{display:inline-block;vertical-align:middle;height:2em;position:relative;width:2.5em}.fa-stack-1x,.fa-stack-2x{bottom:0;left:0;margin:auto;position:absolute;right:0;top:0;z-index:var(--fa-stack-z-index,auto)}.svg-inline--fa.fa-stack-1x{height:1em;width:1.25em}.svg-inline--fa.fa-stack-2x{height:2em;width:2.5em}.fa-inverse{color:var(--fa-inverse,#fff)}.fa-sr-only,.sr-only{position:absolute;width:1px;height:1px;padding:0;margin:-1px;overflow:hidden;clip:rect(0,0,0,0);white-space:nowrap;border-width:0}.fa-sr-only-focusable:not(:focus),.sr-only-focusable:not(:focus){position:absolute;width:1px;height:1px;padding:0;margin:-1px;overflow:hidden;clip:rect(0,0,0,0);white-space:nowrap;border-width:0}.svg-inline--fa .fa-primary{fill:var(--fa-primary-color,currentColor);opacity:var(--fa-primary-opacity,1)}.svg-inline--fa .fa-secondary{fill:var(--fa-secondary-color,currentColor);opacity:var(--fa-secondary-opacity,.4)}.svg-inline--fa.fa-swap-opacity .fa-primary{opacity:var(--fa-secondary-opacity,.4)}.svg-inline--fa.fa-swap-opacity .fa-secondary{opacity:var(--fa-primary-opacity,1)}.svg-inline--fa mask .fa-primary,.svg-inline--fa mask .fa-secondary{fill:#000}.fa-duotone.fa-inverse,.fad.fa-inverse{color:var(--fa-inverse,#fff)}';return"fa"===a&&z===s||(c=new RegExp("\\.".concat("fa","\\-"),"g"),l=new RegExp("\\--".concat("fa","\\-"),"g"),s=new RegExp("\\.".concat(s),"g"),e=e.replace(c,".".concat(a,"-")).replace(l,"--".concat(a,"-")).replace(s,".".concat(z))),e}var Hc=!1;function tc(){B.autoAddCss&&!Hc&&(function(c){if(c&&L){var l=o.createElement("style");l.setAttribute("type","text/css"),l.innerHTML=c;for(var s=o.head.childNodes,a=null,z=s.length-1;-1<z;z--){var e=s[z],H=(e.tagName||"").toUpperCase();-1<["STYLE","LINK"].indexOf(H)&&(a=e)}o.head.insertBefore(l,a)}}(ec()),Hc=!0)}i={mixout:function(){return{dom:{css:ec,insertCss:tc}}},hooks:function(){return{beforeDOMElementCreation:function(){tc()},beforeI2svg:function(){tc()}}}},(h=m||{})[r]||(h[r]={}),h[r].styles||(h[r].styles={}),h[r].hooks||(h[r].hooks={}),h[r].shims||(h[r].shims=[]);var Vc=h[r],rc=[],Mc=!1;function hc(c){L&&(Mc?setTimeout(c,0):rc.push(c))}function nc(c){var l,s=c.tag,a=void 0===(z=c.attributes)?{}:z,z=void 0===(z=c.children)?[]:z;return"string"==typeof c?sc(c):"<".concat(s," ").concat((l=a,Object.keys(l||{}).reduce((function(c,s){return c+"".concat(s,'="').concat(sc(l[s]),'" ')}),"").trim()),">").concat(z.map(nc).join(""),"</").concat(s,">")}function ic(c,l,s){if(c&&c[l]&&c[l][s])return{prefix:l,iconName:s,icon:c[l][s]}}function mc(c,l,s,a){for(var z,e,H=Object.keys(c),t=H.length,V=void 0!==a?oc(l,a):l,r=void 0===s?(z=1,c[H[0]]):(z=0,s);z<t;z++)r=V(r,c[e=H[z]],e,c);return r}L&&((Mc=(o.documentElement.doScroll?/^loaded|^c/:/^loaded|^i|^c/).test(o.readyState))||o.addEventListener("DOMContentLoaded",(function c(){o.removeEventListener("DOMContentLoaded",c),Mc=1,rc.map((function(c){return c()}))})));var oc=function(c,l){return function(s,a,z,e){return c.call(l,s,a,z,e)}};function fc(c){return c=function(c){for(var l=[],s=0,a=c.length;s<a;){var z,e=c.charCodeAt(s++);55296<=e&&e<=56319&&s<a?56320==(64512&(z=c.charCodeAt(s++)))?l.push(((1023&e)<<10)+(1023&z)+65536):(l.push(e),s--):l.push(e)}return l}(c),1===c.length?c[0].toString(16):null}function Cc(c){return Object.keys(c).reduce((function(l,s){var a=c[s];return a.icon?l[a.iconName]=a.icon:l[s]=a,l}),{})}function vc(c,s,a){a=void 0!==(z=(2<arguments.length&&void 0!==a?a:{}).skipHooks)&&z;var z=Cc(s);"function"!=typeof Vc.hooks.addPack||a?Vc.styles[c]=l(l({},Vc.styles[c]||{}),z):Vc.hooks.addPack(c,Cc(s)),"fas"===c&&vc("fa",s)}var Lc=Vc.styles,uc=Vc.shims,dc=(a(n={},q,Object.values(E[q])),a(n,Z,Object.values(E[Z])),n),pc=null,bc={},gc={},wc={},kc={},yc={},Sc=(a(M={},q,Object.keys(P[q])),a(M,Z,Object.keys(P[Z])),M);function Ac(){function c(c){return mc(Lc,(function(l,s,a){return l[a]=mc(s,c,{}),l}),{})}bc=c((function(c,l,s){return l[3]&&(c[l[3]]=s),l[2]&&l[2].filter((function(c){return"number"==typeof c})).forEach((function(l){c[l.toString(16)]=s})),c})),gc=c((function(c,l,s){return c[s]=s,l[2]&&l[2].filter((function(c){return"string"==typeof c})).forEach((function(l){c[l]=s})),c})),yc=c((function(c,l,s){return l=l[2],c[s]=s,l.forEach((function(l){c[l]=s})),c}));var l="far"in Lc||B.autoFetchSvg,s=mc(uc,(function(c,s){var a=s[0],z=s[1];return s=s[2],"far"!==z||l||(z="fas"),"string"==typeof a&&(c.names[a]={prefix:z,iconName:s}),"number"==typeof a&&(c.unicodes[a.toString(16)]={prefix:z,iconName:s}),c}),{names:{},unicodes:{}});wc=s.names,kc=s.unicodes,pc=Oc(B.styleDefault,{family:B.familyDefault})}function xc(c,l){return(bc[c]||{})[l]}function qc(c,l){return(yc[c]||{})[l]}function Zc(c){return wc[c]||{prefix:null,iconName:null}}function Oc(c,l){var s=(1<arguments.length&&void 0!==l?l:{}).family;return s=P[l=void 0===s?q:s][c],s=N[l][c]||N[l][s],c=c in Vc.styles?c:null,s||c||null}h=function(c){pc=Oc(c.styleDefault,{family:B.familyDefault})},G.push(h),Ac();var jc=(a(r={},q,Object.keys(E[q])),a(r,Z,Object.keys(E[Z])),r);function Pc(c,l){var s=void 0!==(l=(1<arguments.length&&void 0!==l?l:{}).skipLookups)&&l,z=(a(l={},q,"".concat(B.cssPrefix,"-").concat(q)),a(l,Z,"".concat(B.cssPrefix,"-").concat(Z)),l),e=null,H=q;return(c.includes(z[q])||c.some((function(c){return jc[q].includes(c)})))&&(H=q),(c.includes(z[Z])||c.some((function(c){return jc[Z].includes(c)})))&&(H=Z),l=c.reduce((function(c,l){var a=function(c,l){l=(s=l.split("-"))[0];var s=s.slice(1).join("-");return l!==c||""===s||~W.indexOf(s)?null:s}(B.cssPrefix,l);return Lc[l]?(l=dc[H].includes(l)?I[H][l]:l,e=l,c.prefix=l):-1<Sc[H].indexOf(l)?(e=l,c.prefix=Oc(l,{family:H})):a?c.iconName=a:l!==B.replacementClass&&l!==z[q]&&l!==z[Z]&&c.rest.push(l),!s&&c.prefix&&c.iconName&&(a="fa"===e?Zc(c.iconName):{},l=qc(c.prefix,c.iconName),a.prefix&&(e=null),c.iconName=a.iconName||l||c.iconName,c.prefix=a.prefix||c.prefix,"far"!==c.prefix||Lc.far||!Lc.fas||B.autoFetchSvg||(c.prefix="fas")),c}),{prefix:null,iconName:null,rest:[]}),(c.includes("fa-brands")||c.includes("fab"))&&(l.prefix="fab"),(c.includes("fa-duotone")||c.includes("fad"))&&(l.prefix="fad"),l.prefix||H!==Z||!Lc.fass&&!B.autoFetchSvg||(l.prefix="fass",l.iconName=qc(l.prefix,l.iconName)||l.iconName),"fa"!==l.prefix&&"fa"!==e||(l.prefix=pc||"fas"),l}n=function(){function c(){!function(c,l){if(!(c instanceof l))throw new TypeError("Cannot call a class as a function")}(this,c),this.definitions={}}var s,a;return s=c,(a=[{key:"add",value:function(){for(var c=this,s=arguments.length,a=new Array(s),z=0;z<s;z++)a[z]=arguments[z];var e=a.reduce(this._pullDefinitions,{});Object.keys(e).forEach((function(s){c.definitions[s]=l(l({},c.definitions[s]||{}),e[s]),vc(s,e[s]);var a=E[q][s];a&&vc(a,e[s]),Ac()}))}},{key:"reset",value:function(){this.definitions={}}},{key:"_pullDefinitions",value:function(c,l){var s=l.prefix&&l.iconName&&l.icon?{0:l}:l;return Object.keys(s).map((function(l){var a=(e=s[l]).prefix,z=(l=e.iconName,e.icon),e=z[2];c[a]||(c[a]={}),0<e.length&&e.forEach((function(l){"string"==typeof l&&(c[a][l]=z)})),c[a][l]=z})),c}}])&&function(c,l){for(var s=0;s<l.length;s++){var a=l[s];a.enumerable=a.enumerable||!1,a.configurable=!0,"value"in a&&(a.writable=!0),Object.defineProperty(c,a.key,a)}}(s.prototype,a),Object.defineProperty(s,"prototype",{writable:!1}),c}(),M=[];var Nc={},Ec={},Ic=Object.keys(Ec);function Tc(c,l){for(var s=arguments.length,a=new Array(2<s?s-2:0),z=2;z<s;z++)a[z-2]=arguments[z];return(Nc[c]||[]).forEach((function(c){l=c.apply(null,[l].concat(a))})),l}function Dc(c){for(var l=arguments.length,s=new Array(1<l?l-1:0),a=1;a<l;a++)s[a-1]=arguments[a];(Nc[c]||[]).forEach((function(c){c.apply(null,s)}))}function Yc(c){var l=c;return c=Array.prototype.slice.call(arguments,1),Ec[l]?Ec[l].apply(null,c):void 0}function Rc(c){"fa"===c.prefix&&(c.prefix="fas");var l=c.iconName;if(c=c.prefix||pc,l)return l=qc(c,l)||l,ic(_c.definitions,c,l)||ic(Vc.styles,c,l)}var _c=new n,Fc={noAuto:function(){B.autoReplaceSvg=!1,B.observeMutations=!1,Dc("noAuto")},config:B,dom:{i2svg:function(){var c=0<arguments.length&&void 0!==arguments[0]?arguments[0]:{};return L?(Dc("beforeI2svg",c),Yc("pseudoElements2svg",c),Yc("i2svg",c)):Promise.reject("Operation requires a DOM of some kind.")},watch:function(){var c=0<arguments.length&&void 0!==arguments[0]?arguments[0]:{},l=c.autoReplaceSvgRoot;!1===B.autoReplaceSvg&&(B.autoReplaceSvg=!0),B.observeMutations=!0,hc((function(){Wc({autoReplaceSvgRoot:l}),Dc("watch",c)}))}},parse:{icon:function(c){if(null===c)return null;if("object"===s(c)&&c.prefix&&c.iconName)return{prefix:c.prefix,iconName:qc(c.prefix,c.iconName)||c.iconName};if(Array.isArray(c)&&2===c.length){var l=0===c[1].indexOf("fa-")?c[1].slice(3):c[1],a=Oc(c[0]);return{prefix:a,iconName:qc(a,l)||l}}return"string"==typeof c&&(-1<c.indexOf("".concat(B.cssPrefix,"-"))||c.match(T))?{prefix:(l=Pc(c.split(" "),{skipLookups:!0})).prefix||pc,iconName:qc(l.prefix,l.iconName)||l.iconName}:"string"==typeof c?{prefix:pc,iconName:qc(pc,c)||c}:void 0}},library:_c,findIconDefinition:Rc,toHtml:nc},Wc=function(){var c=void 0===(c=(0<arguments.length&&void 0!==arguments[0]?arguments[0]:{}).autoReplaceSvgRoot)?o:c;(0<Object.keys(Vc.styles).length||B.autoFetchSvg)&&L&&B.autoReplaceSvg&&Fc.dom.i2svg({node:c})};function Uc(c,l){return Object.defineProperty(c,"abstract",{get:l}),Object.defineProperty(c,"html",{get:function(){return c.abstract.map(nc)}}),Object.defineProperty(c,"node",{get:function(){if(L){var l=o.createElement("div");return l.innerHTML=c.html,l.children}}}),c}function Xc(c){var s,a,z,e,H,t=(v=c.icons).main,V=v.mask,r=c.prefix,M=c.iconName,h=c.transform,n=c.symbol,i=c.title,m=c.maskId,o=c.titleId,f=c.extra,C=void 0!==(L=c.watchable)&&L,v=(u=V.found?V:t).width,L=(c=u.height,"fak"===r),u=[B.replacementClass,M?"".concat(B.cssPrefix,"-").concat(M):""].filter((function(c){return-1===f.classes.indexOf(c)})).filter((function(c){return""!==c||!!c})).concat(f.classes).join(" ");return u={children:[],attributes:l(l({},f.attributes),{},{"data-prefix":r,"data-icon":M,class:u,role:f.attributes.role||"img",xmlns:"http://www.w3.org/2000/svg",viewBox:"0 0 ".concat(v," ").concat(c)})},c=L&&!~f.classes.indexOf("fa-fw")?{width:"".concat(v/c*16*.0625,"em")}:{},C&&(u.attributes[p]=""),i&&(u.children.push({tag:"title",attributes:{id:u.attributes["aria-labelledby"]||"title-".concat(o||J())},children:[i]}),delete u.attributes.title),c=l(l({},u),{},{prefix:r,iconName:M,main:t,mask:V,maskId:m,transform:h,symbol:n,styles:l(l({},c),f.styles)}),t=(V=V.found&&t.found?Yc("generateAbstractMask",c)||{children:[],attributes:{}}:Yc("generateAbstractIcon",c)||{children:[],attributes:{}}).children,V=V.attributes,c.children=t,c.attributes=V,n?(a=(s=c).prefix,z=s.iconName,e=s.children,H=s.attributes,s=!0===(s=s.symbol)?"".concat(a,"-").concat(B.cssPrefix,"-").concat(z):s,[{tag:"svg",attributes:{style:"display: none;"},children:[{tag:"symbol",attributes:l(l({},H),{},{id:s}),children:e}]}]):(z=(a=c).children,H=a.main,s=a.mask,e=a.attributes,c=a.styles,zc(a=a.transform)&&H.found&&!s.found&&(s=H.width/H.height/2,H=.5,e.style=ac(l(l({},c),{},{"transform-origin":"".concat(s+a.x/16,"em ").concat(H+a.y/16,"em")}))),[{tag:"svg",attributes:e,children:z}])}function Bc(c){var s=c.content,a=c.width,z=c.height,e=c.transform,H=c.title,t=c.extra,V=c.watchable,r=void 0!==V&&V;return c=l(l(l({},t.attributes),H?{title:H}:{}),{},{class:t.classes.join(" ")}),r&&(c[p]=""),V=l({},t.styles),zc(e)&&(V.transform=(t=(r={transform:e,startCentered:!0,width:a,height:z}).transform,e=r.width,z=void 0===(a=r.height)?16:a,r=void 0!==(a=r.startCentered)&&a,a="",a+=r&&u?"translate(".concat(t.x/Q-(void 0===e?16:e)/2,"em, ").concat(t.y/Q-z/2,"em) "):r?"translate(calc(-50% + ".concat(t.x/Q,"em), calc(-50% + ").concat(t.y/Q,"em)) "):"translate(".concat(t.x/Q,"em, ").concat(t.y/Q,"em) "),a+="scale(".concat(t.size/Q*(t.flipX?-1:1),", ").concat(t.size/Q*(t.flipY?-1:1),") "),a+="rotate(".concat(t.rotate,"deg) ")),V["-webkit-transform"]=V.transform),0<(V=ac(V)).length&&(c.style=V),(V=[]).push({tag:"span",attributes:c,children:[s]}),H&&V.push({tag:"span",attributes:{class:"sr-only"},children:[H]}),V}var Gc=Vc.styles;function Qc(c){var l=c[0],s=c[1];return c=z(c.slice(4),1)[0],{found:!0,width:l,height:s,icon:Array.isArray(c)?{tag:"g",attributes:{class:"".concat(B.cssPrefix,"-").concat(F.GROUP)},children:[{tag:"path",attributes:{class:"".concat(B.cssPrefix,"-").concat(F.SECONDARY),fill:"currentColor",d:c[0]}},{tag:"path",attributes:{class:"".concat(B.cssPrefix,"-").concat(F.PRIMARY),fill:"currentColor",d:c[1]}}]}:{tag:"path",attributes:{fill:"currentColor",d:c}}}}var Kc={found:!1,width:512,height:512};function $c(c,s){var a=s;return"fa"===s&&null!==B.styleDefault&&(s=pc),new Promise((function(z,e){var H,t;if(Yc("missingIconAbstract"),"fa"===a&&(t=Zc(c)||{},c=t.iconName||c,s=t.prefix||s),c&&s&&Gc[s]&&Gc[s][c])return z(Qc(Gc[s][c]));H=c,t=s,x||B.showMissingIcons||!H||console.error('Icon with name "'.concat(H,'" and prefix "').concat(t,'" is missing.')),z(l(l({},Kc),{},{icon:B.showMissingIcons&&c&&Yc("missingIconAbstract")||{}}))}))}function Jc(){}function cl(c){ll.mark("".concat(sl," ").concat(c," ends")),ll.measure("".concat(sl," ").concat(c),"".concat(sl," ").concat(c," begins"),"".concat(sl," ").concat(c," ends"))}var ll=B.measurePerformance&&C&&C.mark&&C.measure?C:{mark:Jc,measure:Jc},sl='FA "6.5.2"',al={begin:function(c){return ll.mark("".concat(sl," ").concat(c," begins")),function(){return cl(c)}},end:cl},zl=function(){};function el(c){return"string"==typeof(c.getAttribute?c.getAttribute(p):null)}function Hl(c){return o.createElementNS("http://www.w3.org/2000/svg",c)}function tl(c){return o.createElement(c)}var Vl={replace:function(c){var l=c[0];l.parentNode&&(c[1].forEach((function(c){l.parentNode.insertBefore(function c(l,s){var a=void 0===(s=(1<arguments.length&&void 0!==s?s:{}).ceFn)?"svg"===l.tag?Hl:tl:s;if("string"==typeof l)return o.createTextNode(l);var z=a(l.tag);return Object.keys(l.attributes||[]).forEach((function(c){z.setAttribute(c,l.attributes[c])})),(l.children||[]).forEach((function(l){z.appendChild(c(l,{ceFn:a}))})),z}(c),l)})),null===l.getAttribute(p)&&B.keepOriginalSource?(c=o.createComment((c=" ".concat((c=l).outerHTML," "),c="".concat(c,"Font Awesome fontawesome.com "))),l.parentNode.replaceChild(c,l)):l.remove())},nest:function(c){var l=c[0],s=c[1];if(~lc(l).indexOf(B.replacementClass))return Vl.replace(c);var a=new RegExp("".concat(B.cssPrefix,"-.*"));delete s[0].attributes.id,s[0].attributes.class&&(c=s[0].attributes.class.split(" ").reduce((function(c,l){return(l===B.replacementClass||l.match(a)?c.toSvg:c.toNode).push(l),c}),{toNode:[],toSvg:[]}),s[0].attributes.class=c.toSvg.join(" "),0===c.toNode.length?l.removeAttribute("class"):l.setAttribute("class",c.toNode.join(" "))),s=s.map(nc).join("\n"),l.setAttribute(p,""),l.innerHTML=s}};function rl(c){c()}function Ml(c,l){var s="function"==typeof l?l:zl;0===c.length?s():(B.mutateApproach===S&&m.requestAnimationFrame||rl)((function(){var l=!0!==B.autoReplaceSvg&&Vl[B.autoReplaceSvg]||Vl.replace,a=al.begin("mutate");c.map(l),a(),s()}))}var hl=!1;function nl(){hl=!0}function il(){hl=!1}var ml=null;function ol(c){var l,s,a,z;f&&B.observeMutations&&(a=c.treeCallback,l=void 0===a?zl:a,a=c.nodeCallback,s=void 0===a?zl:a,a=c.pseudoElementsCallback,z=void 0===a?zl:a,c=void 0===(c=c.observeMutationsRoot)?o:c,ml=new f((function(c){var a;hl||(a=pc,cc(c).forEach((function(c){var e,H,t;"childList"===c.type&&0<c.addedNodes.length&&!el(c.addedNodes[0])&&(B.searchPseudoElements&&z(c.target),l(c.target)),"attributes"===c.type&&c.target.parentNode&&B.searchPseudoElements&&z(c.target.parentNode),"attributes"===c.type&&el(c.target)&&~_.indexOf(c.attributeName)&&("class"===c.attributeName&&(t=(H=c.target).getAttribute?H.getAttribute(w):null,H=H.getAttribute?H.getAttribute(k):null,t&&H)?(H=(e=Pc(lc(c.target))).prefix,e=e.iconName,c.target.setAttribute(w,H||a),e&&c.target.setAttribute(k,e)):(e=c.target)&&e.classList&&e.classList.contains&&e.classList.contains(B.replacementClass)&&s(c.target))})))})),L&&ml.observe(c,{childList:!0,attributes:!0,characterData:!0,subtree:!0}))}function fl(c,s){var a=1<arguments.length&&void 0!==s?s:{styleParser:!0},z=function(c){var l=c.getAttribute("data-prefix"),s=c.getAttribute("data-icon"),a=void 0!==c.innerText?c.innerText.trim():"",z=Pc(lc(c));return z.prefix||(z.prefix=pc),l&&s&&(z.prefix=l,z.iconName=s),z.iconName&&z.prefix||(z.prefix&&0<a.length&&(z.iconName=(s=z.prefix,a=c.innerText,(gc[s]||{})[a]||xc(z.prefix,fc(c.innerText)))),!z.iconName&&B.autoFetchSvg&&c.firstChild&&c.firstChild.nodeType===Node.TEXT_NODE&&(z.iconName=c.firstChild.data)),z}(c),e=z.iconName,H=z.prefix,t=z.rest,V=(s=cc((V=c).attributes).reduce((function(c,l){return"class"!==c.name&&"style"!==c.name&&(c[l.name]=l.value),c}),{}),z=V.getAttribute("title"),V=V.getAttribute("data-fa-title-id"),B.autoA11y&&(z?s["aria-labelledby"]="".concat(B.replacementClass,"-title-").concat(V||J()):(s["aria-hidden"]="true",s.focusable="false")),s),r=(s=Tc("parseNodeAttributes",{},c),a.styleParser?(a=(r=c).getAttribute("style"),r=[],r=a?a.split(";").reduce((function(c,l){l=(s=l.split(":"))[0];var s=s.slice(1);return l&&0<s.length&&(c[l]=s.join(":").trim()),c}),{}):r):[]);return l({iconName:e,title:c.getAttribute("title"),titleId:c.getAttribute("data-fa-title-id"),prefix:H,transform:K,mask:{iconName:null,prefix:null,rest:[]},maskId:null,symbol:!1,extra:{classes:t,styles:r,attributes:V}},s)}var Cl=Vc.styles;function vl(c){var l="nest"===B.autoReplaceSvg?fl(c,{styleParser:!1}):fl(c);return~l.extra.classes.indexOf(D)?Yc("generateLayersText",c,l):Yc("generateSvgReplacementMutation",c,l)}var Ll=new Set;function ul(c){var l=1<arguments.length&&void 0!==arguments[1]?arguments[1]:null;if(!L)return Promise.resolve();function s(c){return z.add("".concat(y,"-").concat(c))}function a(c){return z.remove("".concat(y,"-").concat(c))}var z=o.documentElement.classList,e=B.autoFetchSvg?Ll:O.map((function(c){return"fa-".concat(c)})).concat(Object.keys(Cl));e.includes("fa")||e.push("fa");var H=[".".concat(D,":not([").concat(p,"])")].concat(e.map((function(c){return".".concat(c,":not([").concat(p,"])")}))).join(", ");if(0===H.length)return Promise.resolve();e=[];try{e=cc(c.querySelectorAll(H))}catch(c){}if(!(0<e.length))return Promise.resolve();s("pending"),a("complete");var t=al.begin("onTree"),V=e.reduce((function(c,l){try{var s=vl(l);s&&c.push(s)}catch(c){x||"MissingIcon"===c.name&&console.error(c)}return c}),[]);return new Promise((function(c,z){Promise.all(V).then((function(z){Ml(z,(function(){s("active"),s("complete"),a("pending"),"function"==typeof l&&l(),t(),c()}))})).catch((function(c){t(),z(c)}))}))}function dl(c){var l=1<arguments.length&&void 0!==arguments[1]?arguments[1]:null;vl(c).then((function(c){c&&Ml([c],l)}))}function pl(c){var s=1<arguments.length&&void 0!==arguments[1]?arguments[1]:{},a=s.transform,z=void 0===a?K:a,e=void 0!==(a=s.symbol)&&a,H=void 0===(a=s.mask)?null:a,t=void 0===(a=s.maskId)?null:a,V=void 0===(a=s.title)?null:a,r=void 0===(a=s.titleId)?null:a,M=void 0===(a=s.classes)?[]:a,h=void 0===(a=s.attributes)?{}:a,n=void 0===(a=s.styles)?{}:a;if(c){var i=c.prefix,m=c.iconName,o=c.icon;return Uc(l({type:"icon"},c),(function(){return Dc("beforeDOMElementCreation",{iconDefinition:c,params:s}),B.autoA11y&&(V?h["aria-labelledby"]="".concat(B.replacementClass,"-title-").concat(r||J()):(h["aria-hidden"]="true",h.focusable="false")),Xc({icons:{main:Qc(o),mask:H?Qc(H.icon):{found:!1,width:null,height:null,icon:{}}},prefix:i,iconName:m,transform:l(l({},K),z),symbol:e,title:V,maskId:t,titleId:r,extra:{attributes:h,styles:n,classes:M}})}))}}O.map((function(c){Ll.add("fa-".concat(c))})),Object.keys(P[q]).map(Ll.add.bind(Ll)),Object.keys(P[Z]).map(Ll.add.bind(Ll)),Ll=e(Ll),h={mixout:function(){return{icon:(c=pl,function(s){var a=1<arguments.length&&void 0!==arguments[1]?arguments[1]:{},z=(s||{}).icon?s:Rc(s||{});return s=(s=a.mask)&&((s||{}).icon?s:Rc(s||{})),c(z,l(l({},a),{},{mask:s}))})};var c},hooks:function(){return{mutationObserverCallbacks:function(c){return c.treeCallback=ul,c.nodeCallback=dl,c}}},provides:function(c){c.i2svg=function(c){var l=c.node;return c=c.callback,ul(void 0===l?o:l,void 0===c?function(){}:c)},c.generateSvgReplacementMutation=function(c,l){var s=l.iconName,a=l.title,e=l.titleId,H=l.prefix,t=l.transform,V=l.symbol,r=l.mask,M=l.maskId,h=l.extra;return new Promise((function(l,n){Promise.all([$c(s,H),r.iconName?$c(r.iconName,r.prefix):Promise.resolve({found:!1,width:512,height:512,icon:{}})]).then((function(r){r=(n=z(r,2))[0];var n=n[1];l([c,Xc({icons:{main:r,mask:n},prefix:H,iconName:s,transform:t,symbol:V,maskId:M,title:a,titleId:e,extra:h,watchable:!0})])})).catch(n)}))},c.generateAbstractIcon=function(c){var l,s=c.children,a=c.attributes,z=c.main,e=c.transform;return 0<(c=ac(c.styles)).length&&(a.style=c),zc(e)&&(l=Yc("generateAbstractTransformGrouping",{main:z,transform:e,containerWidth:z.width,iconWidth:z.width})),s.push(l||z.icon),{children:s,attributes:a}}}},r={mixout:function(){return{layer:function(c){var l=1<arguments.length&&void 0!==arguments[1]?arguments[1]:{},s=l.classes,a=void 0===s?[]:s;return Uc({type:"layer"},(function(){Dc("beforeDOMElementCreation",{assembler:c,params:l});var s=[];return c((function(c){Array.isArray(c)?c.map((function(c){s=s.concat(c.abstract)})):s=s.concat(c.abstract)})),[{tag:"span",attributes:{class:["".concat(B.cssPrefix,"-layers")].concat(e(a)).join(" ")},children:s}]}))}}}},n={mixout:function(){return{counter:function(c){var s,a=1<arguments.length&&void 0!==arguments[1]?arguments[1]:{},z=void 0===(s=a.title)?null:s,H=void 0===(s=a.classes)?[]:s,t=void 0===(s=a.attributes)?{}:s,V=void 0===(s=a.styles)?{}:s;return Uc({type:"counter",content:c},(function(){return Dc("beforeDOMElementCreation",{content:c,params:a}),r=(s={content:c.toString(),title:z,extra:{attributes:t,styles:V,classes:["".concat(B.cssPrefix,"-layers-counter")].concat(e(H))}}).content,M=s.title,s=l(l(l({},(h=s.extra).attributes),M?{title:M}:{}),{},{class:h.classes.join(" ")}),0<(h=ac(h.styles)).length&&(s.style=h),(h=[]).push({tag:"span",attributes:s,children:[r]}),M&&h.push({tag:"span",attributes:{class:"sr-only"},children:[M]}),h;var s,r,M,h}))}}}},C={mixout:function(){return{text:function(c){var s,a=1<arguments.length&&void 0!==arguments[1]?arguments[1]:{},z=void 0===(s=a.transform)?K:s,H=void 0===(s=a.title)?null:s,t=void 0===(s=a.classes)?[]:s,V=void 0===(s=a.attributes)?{}:s,r=void 0===(s=a.styles)?{}:s;return Uc({type:"text",content:c},(function(){return Dc("beforeDOMElementCreation",{content:c,params:a}),Bc({content:c,transform:l(l({},K),z),title:H,extra:{attributes:V,styles:r,classes:["".concat(B.cssPrefix,"-layers-text")].concat(e(t))}})}))}}},provides:function(c){c.generateLayersText=function(c,l){var s,a=l.title,z=l.transform,e=l.extra,H=null,t=null;return u&&(s=parseInt(getComputedStyle(c).fontSize,10),H=(l=c.getBoundingClientRect()).width/s,t=l.height/s),B.autoA11y&&!a&&(e.attributes["aria-hidden"]="true"),Promise.resolve([c,Bc({content:c.innerHTML,width:H,height:t,transform:z,title:a,extra:e,watchable:!0})])}}};var bl=new RegExp('"',"ug"),gl=[1105920,1112319];function wl(c,s){var a="".concat(g).concat(s.replace(":","-"));return new Promise((function(z,e){if(null!==c.getAttribute(a))return z();var H,t,V,r,M,h,n,i=cc(c.children).filter((function(c){return c.getAttribute(b)===s}))[0],f=m.getComputedStyle(c,s),C=f.getPropertyValue("font-family").match(Y),v=f.getPropertyValue("font-weight"),L=f.getPropertyValue("content");if(i&&!C)return c.removeChild(i),z();C&&"none"!==L&&""!==L?(h=f.getPropertyValue("content"),M=~["Sharp"].indexOf(C[2])?Z:q,H=~["Solid","Regular","Light","Thin","Duotone","Brands","Kit"].indexOf(C[2])?N[M][C[2].toLowerCase()]:R[M][v],L=(f=(L=h).replace(bl,""),M=0,h=(v=f).length,n=55296<=(L=v.charCodeAt(M))&&L<=56319&&M+1<h&&56320<=(n=v.charCodeAt(M+1))&&n<=57343?1024*(L-55296)+n-56320+65536:L,L=gl[0]<=n&&n<=gl[1],f={value:fc((n=2===f.length&&f[0]===f[1])?f[0]:f),isSecondary:L||n}).value,n=f.isSecondary,f=C[0].startsWith("FontAwesome"),C=xc(H,L),t=C,f&&(L=kc[f=L],f=xc("fas",f),(f=L||(f?{prefix:"fas",iconName:f}:null)||{prefix:null,iconName:null}).iconName&&f.prefix&&(C=f.iconName,H=f.prefix)),!C||n||i&&i.getAttribute(w)===H&&i.getAttribute(k)===t?z():(c.setAttribute(a,t),i&&c.removeChild(i),(r=(V={iconName:null,title:null,titleId:null,prefix:null,transform:K,symbol:!1,mask:{iconName:null,prefix:null,rest:[]},maskId:null,extra:{classes:[],styles:{},attributes:{}}}).extra).attributes[b]=s,$c(C,H).then((function(e){var M=Xc(l(l({},V),{},{icons:{main:e,mask:{prefix:null,iconName:null,rest:[]}},prefix:H,iconName:t,extra:r,watchable:!0}));e=o.createElementNS("http://www.w3.org/2000/svg","svg"),"::before"===s?c.insertBefore(e,c.firstChild):c.appendChild(e),e.outerHTML=M.map(nc).join("\n"),c.removeAttribute(a),z()})).catch(e))):z()}))}function kl(c){return Promise.all([wl(c,"::before"),wl(c,"::after")])}function yl(c){return!(c.parentNode===document.head||~A.indexOf(c.tagName.toUpperCase())||c.getAttribute(b)||c.parentNode&&"svg"===c.parentNode.tagName)}function Sl(c){if(L)return new Promise((function(l,s){var a=cc(c.querySelectorAll("*")).filter(yl).map(kl),z=al.begin("searchPseudoElements");nl(),Promise.all(a).then((function(){z(),il(),l()})).catch((function(){z(),il(),s()}))}))}function Al(c){return c.toLowerCase().split(" ").reduce((function(c,l){var s=l.toLowerCase().split("-"),a=(l=s[0],s.slice(1).join("-"));if(l&&"h"===a)return c.flipX=!0,c;if(l&&"v"===a)return c.flipY=!0,c;if(a=parseFloat(a),isNaN(a))return c;switch(l){case"grow":c.size=c.size+a;break;case"shrink":c.size=c.size-a;break;case"left":c.x=c.x-a;break;case"right":c.x=c.x+a;break;case"up":c.y=c.y-a;break;case"down":c.y=c.y+a;break;case"rotate":c.rotate=c.rotate+a}return c}),{size:16,x:0,y:0,flipX:!1,flipY:!1,rotate:0})}var xl,ql=!1,Zl={x:0,y:0,width:"100%",height:"100%"};function Ol(c){return c.attributes&&(c.attributes.fill||!(1<arguments.length&&void 0!==arguments[1])||arguments[1])&&(c.attributes.fill="black"),c}xl=Fc,M=[i,h,r,n,C,{hooks:function(){return{mutationObserverCallbacks:function(c){return c.pseudoElementsCallback=Sl,c}}},provides:function(c){c.pseudoElements2svg=function(c){c=c.node,B.searchPseudoElements&&Sl(void 0===c?o:c)}}},{mixout:function(){return{dom:{unwatch:function(){nl(),ql=!0}}}},hooks:function(){return{bootstrap:function(){ol(Tc("mutationObserverCallbacks",{}))},noAuto:function(){ml&&ml.disconnect()},watch:function(c){c=c.observeMutationsRoot,ql?il():ol(Tc("mutationObserverCallbacks",{observeMutationsRoot:c}))}}}},{mixout:function(){return{parse:{transform:Al}}},hooks:function(){return{parseNodeAttributes:function(c,l){return(l=l.getAttribute("data-fa-transform"))&&(c.transform=Al(l)),c}}},provides:function(c){c.generateAbstractTransformGrouping=function(c){var s=c.main,a=c.transform,z=c.containerWidth,e=c.iconWidth,H={transform:"translate(".concat(z/2," 256)")};return c="translate(".concat(32*a.x,", ").concat(32*a.y,") "),z="scale(".concat(a.size/16*(a.flipX?-1:1),", ").concat(a.size/16*(a.flipY?-1:1),") "),a="rotate(".concat(a.rotate," 0 0)"),{tag:"g",attributes:l({},(e={outer:H,inner:{transform:"".concat(c," ").concat(z," ").concat(a)},path:{transform:"translate(".concat(e/2*-1," -256)")}}).outer),children:[{tag:"g",attributes:l({},e.inner),children:[{tag:s.icon.tag,children:s.icon.children,attributes:l(l({},s.icon.attributes),e.path)}]}]}}}},{hooks:function(){return{parseNodeAttributes:function(c,l){var s=(s=l.getAttribute("data-fa-mask"))?Pc(s.split(" ").map((function(c){return c.trim()}))):{prefix:null,iconName:null,rest:[]};return s.prefix||(s.prefix=pc),c.mask=s,c.maskId=l.getAttribute("data-fa-mask-id"),c}}},provides:function(c){c.generateAbstractMask=function(c){var s=c.children,a=c.attributes,z=c.main,e=c.mask,H=c.maskId,t=c.transform,V=z.width,r=z.icon,M=e.width;return c=e.icon,e=(z={transform:t,containerWidth:M,iconWidth:V}).transform,t=z.containerWidth,M=z.iconWidth,V={transform:"translate(".concat(t/2," 256)")},z="translate(".concat(32*e.x,", ").concat(32*e.y,") "),t="scale(".concat(e.size/16*(e.flipX?-1:1),", ").concat(e.size/16*(e.flipY?-1:1),") "),e="rotate(".concat(e.rotate," 0 0)"),t={outer:V,inner:{transform:"".concat(z," ").concat(t," ").concat(e)},path:{transform:"translate(".concat(M/2*-1," -256)")}},e={tag:"rect",attributes:l(l({},Zl),{},{fill:"white"})},M=r.children?{children:r.children.map(Ol)}:{},M={tag:"g",attributes:l({},t.inner),children:[Ol(l({tag:r.tag,attributes:l(l({},r.attributes),t.path)},M))]},t={tag:"g",attributes:l({},t.outer),children:[M]},M="mask-".concat(H||J()),H="clip-".concat(H||J()),t={tag:"mask",attributes:l(l({},Zl),{},{id:M,maskUnits:"userSpaceOnUse",maskContentUnits:"userSpaceOnUse"}),children:[e,t]},t={tag:"defs",children:[{tag:"clipPath",attributes:{id:H},children:"g"===c.tag?c.children:[c]},t]},s.push(t,{tag:"rect",attributes:l({fill:"currentColor","clip-path":"url(#".concat(H,")"),mask:"url(#".concat(M,")")},Zl)}),{children:s,attributes:a}}}},{provides:function(c){var s=!1;m.matchMedia&&(s=m.matchMedia("(prefers-reduced-motion: reduce)").matches),c.missingIconAbstract=function(){var c=[],a={fill:"currentColor"},z={attributeType:"XML",repeatCount:"indefinite",dur:"2s"};c.push({tag:"path",attributes:l(l({},a),{},{d:"M156.5,447.7l-12.6,29.5c-18.7-9.5-35.9-21.2-51.5-34.9l22.7-22.7C127.6,430.5,141.5,440,156.5,447.7z M40.6,272H8.5 c1.4,21.2,5.4,41.7,11.7,61.1L50,321.2C45.1,305.5,41.8,289,40.6,272z M40.6,240c1.4-18.8,5.2-37,11.1-54.1l-29.5-12.6 C14.7,194.3,10,216.7,8.5,240H40.6z M64.3,156.5c7.8-14.9,17.2-28.8,28.1-41.5L69.7,92.3c-13.7,15.6-25.5,32.8-34.9,51.5 L64.3,156.5z M397,419.6c-13.9,12-29.4,22.3-46.1,30.4l11.9,29.8c20.7-9.9,39.8-22.6,56.9-37.6L397,419.6z M115,92.4 c13.9-12,29.4-22.3,46.1-30.4l-11.9-29.8c-20.7,9.9-39.8,22.6-56.8,37.6L115,92.4z M447.7,355.5c-7.8,14.9-17.2,28.8-28.1,41.5 l22.7,22.7c13.7-15.6,25.5-32.9,34.9-51.5L447.7,355.5z M471.4,272c-1.4,18.8-5.2,37-11.1,54.1l29.5,12.6 c7.5-21.1,12.2-43.5,13.6-66.8H471.4z M321.2,462c-15.7,5-32.2,8.2-49.2,9.4v32.1c21.2-1.4,41.7-5.4,61.1-11.7L321.2,462z M240,471.4c-18.8-1.4-37-5.2-54.1-11.1l-12.6,29.5c21.1,7.5,43.5,12.2,66.8,13.6V471.4z M462,190.8c5,15.7,8.2,32.2,9.4,49.2h32.1 c-1.4-21.2-5.4-41.7-11.7-61.1L462,190.8z M92.4,397c-12-13.9-22.3-29.4-30.4-46.1l-29.8,11.9c9.9,20.7,22.6,39.8,37.6,56.9 L92.4,397z M272,40.6c18.8,1.4,36.9,5.2,54.1,11.1l12.6-29.5C317.7,14.7,295.3,10,272,8.5V40.6z M190.8,50 c15.7-5,32.2-8.2,49.2-9.4V8.5c-21.2,1.4-41.7,5.4-61.1,11.7L190.8,50z M442.3,92.3L419.6,115c12,13.9,22.3,29.4,30.5,46.1 l29.8-11.9C470,128.5,457.3,109.4,442.3,92.3z M397,92.4l22.7-22.7c-15.6-13.7-32.8-25.5-51.5-34.9l-12.6,29.5 C370.4,72.1,384.4,81.5,397,92.4z"})});var e=l(l({},z),{},{attributeName:"opacity"}),H={tag:"circle",attributes:l(l({},a),{},{cx:"256",cy:"364",r:"28"}),children:[]};return s||H.children.push({tag:"animate",attributes:l(l({},z),{},{attributeName:"r",values:"28;14;28;28;14;28;"})},{tag:"animate",attributes:l(l({},e),{},{values:"1;0;1;1;0;1;"})}),c.push(H),c.push({tag:"path",attributes:l(l({},a),{},{opacity:"1",d:"M263.7,312h-16c-6.6,0-12-5.4-12-12c0-71,77.4-63.9,77.4-107.8c0-20-17.8-40.2-57.4-40.2c-29.1,0-44.3,9.6-59.2,28.7 c-3.9,5-11.1,6-16.2,2.4l-13.1-9.2c-5.6-3.9-6.9-11.8-2.6-17.2c21.2-27.2,46.4-44.7,91.2-44.7c52.3,0,97.4,29.8,97.4,80.2 c0,67.6-77.4,63.5-77.4,107.8C275.7,306.6,270.3,312,263.7,312z"}),children:s?[]:[{tag:"animate",attributes:l(l({},e),{},{values:"1;0;0;0;0;1;"})}]}),s||c.push({tag:"path",attributes:l(l({},a),{},{opacity:"0",d:"M232.5,134.5l7,168c0.3,6.4,5.6,11.5,12,11.5h9c6.4,0,11.7-5.1,12-11.5l7-168c0.3-6.8-5.2-12.5-12-12.5h-23 C237.7,122,232.2,127.7,232.5,134.5z"}),children:[{tag:"animate",attributes:l(l({},e),{},{values:"0;0;1;1;0;0;"})}]}),{tag:"g",attributes:{class:"missing"},children:c}}}},{hooks:function(){return{parseNodeAttributes:function(c,l){return l=l.getAttribute("data-fa-symbol"),c.symbol=null!==l&&(""===l||l),c}}}}],Nc={},Object.keys(Ec).forEach((function(c){-1===Ic.indexOf(c)&&delete Ec[c]})),M.forEach((function(c){var l,a=c.mixout?c.mixout():{};Object.keys(a).forEach((function(c){"function"==typeof a[c]&&(xl[c]=a[c]),"object"===s(a[c])&&Object.keys(a[c]).forEach((function(l){xl[c]||(xl[c]={}),xl[c][l]=a[c][l]}))})),c.hooks&&(l=c.hooks(),Object.keys(l).forEach((function(c){Nc[c]||(Nc[c]=[]),Nc[c].push(l[c])}))),c.provides&&c.provides(Ec)})),function(c){try{for(var l=arguments.length,s=new Array(1<l?l-1:0),a=1;a<l;a++)s[a-1]=arguments[a];c.apply(void 0,s)}catch(c){if(!x)throw c}}((function(c){v&&(m.FontAwesome||(m.FontAwesome=Fc),hc((function(){Wc(),Dc("bootstrap")}))),Vc.hooks=l(l({},Vc.hooks),{},{addPack:function(c,s){Vc.styles[c]=l(l({},Vc.styles[c]||{}),s),Ac(),Wc()},addPacks:function(c){c.forEach((function(c){c=(s=z(c,2))[0];var s=s[1];Vc.styles[c]=l(l({},Vc.styles[c]||{}),s)})),Ac(),Wc()},addShims:function(c){var l;(l=Vc.shims).push.apply(l,e(c)),Ac(),Wc()}})}))}()}},l={};function s(a){var z=l[a];if(void 0!==z)return z.exports;var e=l[a]={exports:{}};return c[a](e,e.exports,s),e.exports}s.n=c=>{var l=c&&c.__esModule?()=>c.default:()=>c;return s.d(l,{a:l}),l},s.d=(c,l)=>{for(var a in l)s.o(l,a)&&!s.o(c,a)&&Object.defineProperty(c,a,{enumerable:!0,get:l[a]})},s.o=(c,l)=>Object.prototype.hasOwnProperty.call(c,l),(()=>{"use strict";s(654)})()})();
+//# sourceMappingURL=fontawesome.js.map
\ No newline at end of file
diff --git a/_static/scripts/fontawesome.js.LICENSE.txt b/_static/scripts/fontawesome.js.LICENSE.txt
new file mode 100644
index 000000000..a91750aff
--- /dev/null
+++ b/_static/scripts/fontawesome.js.LICENSE.txt
@@ -0,0 +1,5 @@
+/*!
+ * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com
+ * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)
+ * Copyright 2024 Fonticons, Inc.
+ */
diff --git a/_static/scripts/fontawesome.js.map b/_static/scripts/fontawesome.js.map
new file mode 100644
index 000000000..8fa96a6db
--- /dev/null
+++ b/_static/scripts/fontawesome.js.map
@@ -0,0 +1 @@
+{"version":3,"file":"scripts/fontawesome.js","mappings":";uBAKC,WAAW,aAAa,IAAIA,EAAE,CAAC,EAAEC,EAAE,CAAC,EAAE,IAAI,oBAAoBC,SAASF,EAAEE,QAAQ,oBAAoBC,WAAWF,EAAEE,SAAS,CAAC,MAAMH,GAAG,CAAC,IAAkCI,OAAE,KAAhCC,GAAGL,EAAEM,WAAW,CAAC,GAAGC,WAAuB,GAAGF,EAAEG,EAAER,EAAES,EAAER,EAAyI,SAASS,EAAET,EAAED,GAAG,IAAIK,EAAED,EAAEO,OAAOC,KAAKX,GAAG,OAAOU,OAAOE,wBAAwBR,EAAEM,OAAOE,sBAAsBZ,GAAGD,IAAIK,EAAEA,EAAES,QAAO,SAASd,GAAG,OAAOW,OAAOI,yBAAyBd,EAAED,GAAGgB,UAAU,KAAIZ,EAAEa,KAAKC,MAAMd,EAAEC,IAAID,CAAC,CAAC,SAASe,EAAElB,GAAG,IAAI,IAAID,EAAE,EAAEA,EAAEoB,UAAUC,OAAOrB,IAAI,CAAC,IAAIK,EAAE,MAAMe,UAAUpB,GAAGoB,UAAUpB,GAAG,CAAC,EAAEA,EAAE,EAAEU,EAAEC,OAAON,IAAG,GAAIiB,SAAQ,SAAStB,GAAGuB,EAAEtB,EAAED,EAAEK,EAAEL,GAAG,IAAGW,OAAOa,0BAA0Bb,OAAOc,iBAAiBxB,EAAEU,OAAOa,0BAA0BnB,IAAIK,EAAEC,OAAON,IAAIiB,SAAQ,SAAStB,GAAGW,OAAOe,eAAezB,EAAED,EAAEW,OAAOI,yBAAyBV,EAAEL,GAAG,GAAE,CAAC,OAAOC,CAAC,CAAC,SAASsB,EAAEvB,EAAEC,EAAEI,GAAG,OAAOJ,KAAKD,EAAEW,OAAOe,eAAe1B,EAAEC,EAAE,CAAC0B,MAAMtB,EAAEW,YAAW,EAAGY,cAAa,EAAGC,UAAS,IAAK7B,EAAEC,GAAGI,EAAEL,CAAC,CAAC,SAAS8B,EAAE9B,EAAEC,IAAI,MAAMA,GAAGA,EAAED,EAAEqB,UAAUpB,EAAED,EAAEqB,QAAQ,IAAI,IAAIhB,EAAE,EAAED,EAAE,IAAI2B,MAAM9B,GAAGI,EAAEJ,EAAEI,IAAID,EAAEC,GAAGL,EAAEK,GAAG,OAAOD,CAAC,CAA95BI,EAAEL,SAASM,EAAEuB,iBAAiBvB,EAAEwB,MAAM,mBAAmBxB,EAAEyB,kBAAkBzB,EAAE0B,eAAe/B,EAAEgC,QAAQ,SAAShC,EAAEgC,QAAQ,YAAoyB,IAAIC,EAAE,qBAAqBC,EAAE,WAAW,IAAI,OAAM,CAAE,CAAC,MAAMtC,GAAG,OAAM,CAAE,CAAC,CAA1C,GAA8CuC,EAAE,UAAUC,EAAE,QAAQC,EAAE,CAACF,EAAEC,GAAG,SAASE,EAAE1C,GAAG,OAAO,IAAI2C,MAAM3C,EAAE,CAAC4C,IAAI,SAAS5C,EAAEC,GAAG,OAAOA,KAAKD,EAAEA,EAAEC,GAAGD,EAAEuC,EAAE,GAAG,CAACG,GAAGnB,EAAEsB,EAAE,CAAC,EAAEN,EAAE,CAACO,GAAG,QAAQC,IAAI,QAAQ,WAAW,QAAQC,IAAI,UAAU,aAAa,UAAUC,IAAI,QAAQ,WAAW,QAAQC,IAAI,OAAO,UAAU,OAAOC,IAAI,UAAU,aAAa,UAAUC,IAAI,SAAS,YAAY,SAASC,IAAI,MAAMC,KAAK,MAAM,SAAS,MAAM,iBAAiB,QAAQ/B,EAAEsB,EAAEL,EAAE,CAACM,GAAG,QAAQS,KAAK,QAAQ,WAAW,QAAQC,KAAK,UAAU,aAAa,UAAUC,KAAK,QAAQ,WAAW,QAAQC,KAAK,OAAO,UAAU,SAASb,IAAI,IAAIc,EAAEjB,GAAGnB,EAAEqC,EAAE,CAAC,EAAErB,EAAE,CAACsB,MAAM,MAAMC,QAAQ,MAAMC,MAAM,MAAMC,KAAK,MAAMC,QAAQ,MAAMC,OAAO,MAAMC,IAAI,QAAQ5C,EAAEqC,EAAEpB,EAAE,CAACqB,MAAM,OAAOC,QAAQ,OAAOC,MAAM,OAAOC,KAAK,SAASJ,IAA8mBf,GAA5CzC,GAA3jBsC,GAAGnB,EAAElB,EAAE,CAAC,EAAEkC,EAAE,CAACa,IAAI,YAAYD,IAAI,aAAaE,IAAI,SAASJ,IAAI,WAAWD,IAAI,aAAaD,IAAI,WAAWG,IAAI,YAAY3B,EAAElB,EAAEmC,EAAE,CAACe,KAAK,WAAWC,KAAK,aAAaC,KAAK,WAAWC,KAAK,YAAYrD,IAAIqC,GAAGnB,EAAEvB,EAAE,CAAC,EAAEuC,EAAE,CAAC,YAAY,MAAM,aAAa,MAAM,SAAS,MAAM,WAAW,MAAM,aAAa,MAAM,WAAW,MAAM,UAAU,QAAQhB,EAAEvB,EAAEwC,EAAE,CAAC,WAAW,OAAO,aAAa,OAAO,WAAW,OAAO,UAAU,SAASxC,IAAI0C,GAAGnB,EAAEtB,EAAE,CAAC,EAAEsC,EAAE,CAAC,IAAI,MAAM,IAAI,MAAM6B,OAAO,MAAM,IAAI,MAAM,IAAI,QAAQ7C,EAAEtB,EAAEuC,EAAE,CAAC,IAAI,OAAO,IAAI,OAAO,IAAI,OAAO,IAAI,SAASvC,IAAliBQ,EAAsiB,CAAC,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,KAAS4D,OAAO,CAAC,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,KAAO,iBAAgBT,EAAE,eAAevD,EAAE,UAAUL,EAAE,YAAYC,EAAE,IAAIqE,IAAI3D,OAAOC,KAAK+C,EAAEpB,IAAIgC,IAAItE,EAAEuE,IAAIC,KAAKxE,IAAIU,OAAOC,KAAK+C,EAAEnB,IAAI+B,IAAItE,EAAEuE,IAAIC,KAAKxE,IAAI,GAAGoE,OAAO5B,EAAE,SAASzC,GAAG,GAAG+B,MAAM2C,QAAQ1E,GAAG,OAAO8B,EAAE9B,EAAE,CAA3C,CAA+CC,IAAI,SAASD,GAAG,GAAG,oBAAoB2E,QAAQ,MAAM3E,EAAE2E,OAAOC,WAAW,MAAM5E,EAAE,cAAc,OAAO+B,MAAM8C,KAAK7E,EAAE,CAA/G,CAAiHC,IAAI,SAASD,EAAEC,GAAG,GAAGD,EAAE,CAAC,GAAG,iBAAiBA,EAAE,OAAO8B,EAAE9B,EAAEC,GAAG,IAAII,EAAEM,OAAOmE,UAAUC,SAASC,KAAKhF,GAAGiF,MAAM,GAAG,GAAG,MAAM,SAAS5E,EAAE,WAAWA,GAAGL,EAAEkF,YAAYlF,EAAEkF,YAAYC,KAAK9E,IAAI,QAAQA,EAAE0B,MAAM8C,KAAK7E,GAAG,cAAcK,GAAG,2CAA2C+E,KAAK/E,GAAGyB,EAAE9B,EAAEC,QAAG,CAAM,CAAC,CAA1R,CAA4RA,IAAI,WAAW,MAAM,IAAIoF,UAAU,uIAAuI,CAAtK,GAA0K,CAAC,MAAM,KAAK,KAAK,KAAK,KAAK,MAAM,OAAO,SAAS,OAAO,YAAY,SAAS,YAAY,kBAAkB,gBAAgB,OAAO,KAAK,UAAU,iBAAiB,cAAc,SAAS,KAAK,YAAY,aAAa,QAAQ,aAAa,aAAa,YAAY,YAAY,QAAQ,aAAa,eAAe,OAAO,WAAW,WAAW,QAAQ,KAAKxC,EAAEe,EAAEvD,EAAEL,IAAIqE,OAAO5D,EAAE8D,KAAI,SAASvE,GAAG,MAAM,GAAGqE,OAAOrE,EAAE,IAAI,KAAIqE,OAAOjE,EAAEmE,KAAI,SAASvE,GAAG,MAAM,KAAKqE,OAAOrE,EAAE,MAAIQ,EAAEA,GAAG,CAAC,GAAI6B,KAAK7B,EAAE6B,GAAG,CAAC,GAAG7B,EAAE6B,GAAGiD,SAAS9E,EAAE6B,GAAGiD,OAAO,CAAC,GAAG9E,EAAE6B,GAAGkD,QAAQ/E,EAAE6B,GAAGkD,MAAM,CAAC,GAAG/E,EAAE6B,GAAGmD,QAAQhF,EAAE6B,GAAGmD,MAAM,IAAI,IAAIC,EAAEjF,EAAE6B,GAAG,SAASqD,EAAEtF,GAAG,OAAOO,OAAOC,KAAKR,GAAGuF,QAAO,SAAS3F,EAAEC,GAAG,IAAII,EAAED,EAAEH,GAAG,OAAQI,EAAEuF,KAAK5F,EAAEK,EAAEwF,UAAUxF,EAAEuF,KAAK5F,EAAEC,GAAGI,EAAEL,CAAC,GAAE,CAAC,EAAE,CAAC,SAAS8F,EAAE9F,EAAEC,EAAEI,GAAyDA,OAAE,KAApDD,GAAG,EAAEgB,UAAUC,aAAQ,IAAShB,EAAEA,EAAE,CAAC,GAAG0F,YAAwB3F,EAApE,IAAsEA,EAAEsF,EAAEzF,GAAG,mBAAmBwF,EAAEF,MAAMS,SAAS3F,EAAEoF,EAAEH,OAAOtF,GAAGmB,EAAEA,EAAE,CAAC,EAAEsE,EAAEH,OAAOtF,IAAI,CAAC,GAAGI,GAAGqF,EAAEF,MAAMS,QAAQhG,EAAE0F,EAAEzF,IAAI,QAAQD,GAAG8F,EAAE,KAAK7F,EAAE,CAAC,IAAIgG,EAAE,CAACC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,0NAA0NC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,k4CAAk4CC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,yvBAAyvB,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,01BAA01BC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,6jBAA6jBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,mgBAAmgB,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,0pBAA0pBC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,0rEAA0rEC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,ujBAAujBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,4bAA4bC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,0gCAA0gCC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,+pBAA+pBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,0lBAA0lBC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,ybAAybC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,wLAAwLC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,2qBAA2qBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,ssBAAssBC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,6NAA6N,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,8oCAA8oCC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,4KAA4KC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,uYAAuYC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,uwBAAuwB,YAAY,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,6mBAA6mBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,+GAA+GC,GAAG,CAAC,IAAI,IAAI,GAAG,OAAO,2jBAA2jBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,+WAA+WC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,klCAAklCC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,+lBAA+lB,0BAA0B,CAAC,IAAI,IAAI,GAAG,OAAO,8hBAA8hBC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,2XAA2X,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,0qBAA0qBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,iyCAAiyCC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,8VAA8V,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,09BAA09B,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,ufAAufC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,guBAAguBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,8pCAA8pCC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,uwCAAuwCC,GAAG,CAAC,IAAI,IAAI,GAAG,OAAO,szBAAszBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,shCAAshCC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,q/FAAq/F,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,iPAAiP,gBAAgB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,85BAA85B,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,iYAAiYC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,4dAA4d,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,4WAA4WC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,mmBAAmmBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,4jFAA4jFC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,yRAAyR,mBAAmB,CAAC,IAAI,IAAI,CAAC,oBAAoB,OAAO,yyCAAyyC,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,i8EAAi8E,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,68LAA68L,qBAAqB,CAAC,IAAI,IAAI,CAAC,sBAAsB,OAAO,0UAA0UC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,0nCAA0nCC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,okCAAokCC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,w4BAAw4B,kBAAkB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,ojGAAojG,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,ieAAieC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,qMAAqMC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,2RAA2RC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,kQAAkQ,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,8YAA8YC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,0ZAA0ZC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,miBAAmiB,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,+NAA+N,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,oyBAAoyB,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,i/CAAi/CC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,weAAweC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,8pBAA8pB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,6ZAA6Z,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,8fAA8fC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,mlDAAmlDC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,2PAA2P,6BAA6B,CAAC,IAAI,IAAI,CAAC,oBAAoB,OAAO,ieAAieC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,kSAAkS,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,gTAAgT,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,4SAA4SC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,8dAA8dC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,sZAAsZC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,q6BAAq6B,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,wLAAwL,4BAA4B,CAAC,IAAI,IAAI,GAAG,OAAO,uoCAAuoCC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,8sCAA8sC,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,s3BAAs3B,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,+zBAA+zBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,ojBAAojBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,w2BAAw2BC,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,w+CAAw+CC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,6pBAA6pB,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,8cAA8cC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,ulBAAulBC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,+WAA+WC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,2IAA2I,uBAAuB,CAAC,IAAI,IAAI,GAAG,OAAO,+3PAA+3P,gBAAgB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,i1BAAi1BC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,wgEAAwgEC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,q0BAAq0BC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,+LAA+LC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,uIAAuIC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,irCAAirC,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,4jDAA4jD,kBAAkB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,2zBAA2zBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,6cAA6cC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,i+FAAi+FC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,oJAAoJC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,guBAAguB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,yrCAAyrCC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,qpBAAqpBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,uuJAAuuJC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,kQAAkQC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,oKAAoKC,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,s3BAAs3BC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,kvBAAkvB,iBAAiB,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,2mBAA2mB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,qiBAAqiBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,sdAAsdC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,wgBAAwgBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,4HAA4HC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,wwBAAwwBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,mnBAAmnB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,0hBAA0hB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,8wBAA8wB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,g4FAAg4F,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,0kBAA0kBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,y+DAAy+DC,YAAY,CAAC,IAAI,IAAI,CAAC,sBAAsB,OAAO,mXAAmXC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,ghDAAghDC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,45BAA45BC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,g4CAAg4CC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,+LAA+L,iBAAiB,CAAC,IAAI,IAAI,CAAC,MAAM,kBAAkB,OAAO,gcAAgc,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,6rBAA6rBC,QAAQ,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,gnCAAgnCC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,0xCAA0xC,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,u/CAAu/C,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,s6JAAs6JC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,qsBAAqsBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,iyBAAiyBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,u7BAAu7B,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,mTAAmTC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,2vBAA2vBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,imDAAimDC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,uSAAuS,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,uSAAuS,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,8sBAA8sB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,uhDAAuhD,wBAAwB,CAAC,IAAI,IAAI,GAAG,OAAO,4kBAA4kBC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,y1CAAy1CC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,gmBAAgmBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,u7BAAu7BC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,kxCAAkxCC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,iiBAAiiB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,0oBAA0oB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,yoCAAyoCC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,sfAAsfC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,ogBAAogBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,8vEAA8vEC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,6iBAA6iBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,+LAA+L,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,2rBAA2rBC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,6/BAA6/BC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,ooBAAooBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,yFAAyFC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,wzCAAwzC,eAAe,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,2oBAA2oB,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,ipCAAipC,eAAe,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,qhBAAqhBC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,grBAAgrB,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,MAAM,oBAAoB,0BAA0B,OAAO,gQAAgQC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,iQAAiQC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,ubAAubC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,4uBAA4uBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,8vCAA8vCC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,i9CAAi9C,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,0lCAA0lCC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,upBAAupB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,8zGAA8zG,gBAAgB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,k9CAAk9CC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,saAAsaC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,uuBAAuuB,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,g3EAAg3EC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,ytCAAytCC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,2WAA2W,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,ooCAAooCC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,+vBAA+vBC,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,guBAAguBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,q8CAAq8CC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,mKAAmKC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,+iBAA+iBC,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,m1DAAm1D,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,qIAAqIC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,2uDAA2uDC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,kSAAkS,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,maAAma,gBAAgB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,soBAAsoBC,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,2eAA2e,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,2hBAA2hBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,ogCAAogC,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,4PAA4PC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,sjEAAsjEC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,oiBAAoiBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,+tBAA+tB,uBAAuB,CAAC,IAAI,IAAI,CAAC,wBAAwB,OAAO,smBAAsmBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,myDAAmyDC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,8uBAA8uBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,moGAAmoGC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,wtBAAwtBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,uWAAuWC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,gxFAAgxF,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,spDAAspDC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,qlCAAqlC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,4rCAA4rC,yBAAyB,CAAC,IAAI,IAAI,GAAG,OAAO,qiBAAqiBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,ikCAAikCC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,m6CAAm6CC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,kaAAka,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,qqCAAqqCC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,m4BAAm4B,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,gjFAAgjFC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,uwBAAuwBC,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,+rBAA+rBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,mmCAAmmCC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,kEAAkEC,GAAG,CAAC,IAAI,IAAI,GAAG,OAAO,yQAAyQC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,2vBAA2vB,mBAAmB,CAAC,IAAI,IAAI,CAAC,oBAAoB,OAAO,8rBAA8rBC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,+XAA+XC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,wlBAAwlB,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,skBAAskBC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,slBAAslBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,8UAA8UC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,ouDAAouD,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,wzBAAwzBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,8VAA8VC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,yaAAya,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,+uDAA+uDC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,6iBAA6iBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,gwBAAgwBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,sUAAsU,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,+yBAA+yBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,8MAA8MC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,0dAA0dC,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,wUAAwU,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,0pBAA0pBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,+sBAA+sBC,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,2kBAA2kB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,0jCAA0jCC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,+NAA+NC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,mGAAmG,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,4WAA4W,yBAAyB,CAAC,IAAI,IAAI,GAAG,OAAO,4uBAA4uBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,8NAA8NC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,+eAA+eC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,gWAAgWC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,4WAA4WC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,yZAAyZ,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,wiGAAwiG,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,yKAAyK,qBAAqB,CAAC,IAAI,IAAI,CAAC,sBAAsB,OAAO,6aAA6a,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,qYAAqYC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,wxLAAwxL,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,86CAA86CC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,+2BAA+2B,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,0gBAA0gB,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,g1HAAg1HC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,ohBAAohBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,uQAAuQC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,2iBAA2iBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,6hBAA6hBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,yWAAyWC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,oxBAAoxB,cAAc,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,wfAAwfC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,wGAAwGC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,kgBAAkgBC,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,swCAAswCC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,6eAA6eC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,iWAAiWC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,2MAA2MC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,yxBAAyxB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,mkJAAmkJC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,wpBAAwpBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,qjEAAqjEC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,42BAA42BC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,8PAA8PC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,mmCAAmmCC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,+SAA+SC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,q2CAAq2CC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,gkEAAgkEC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,oTAAoT,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,oaAAoaC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,gmBAAgmBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,mlCAAmlCC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,gOAAgOC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,0kBAA0kB,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,8dAA8dC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,kfAAkf,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,uIAAuIC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,mjDAAmjDC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,oWAAoW,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,gKAAgK,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,8rCAA8rCC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,oPAAoP,WAAW,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,0pBAA0pB,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,gOAAgOC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,wNAAwN,oBAAoB,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,+RAA+R,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,sTAAsTC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,8vBAA8vBC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,oUAAoUC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,6dAA6dC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,+3BAA+3B,4BAA4B,CAAC,IAAI,IAAI,GAAG,OAAO,ymBAAymBC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,yqBAAyqBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,+NAA+N,kBAAkB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,+QAA+QC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,gVAAgVC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,o6GAAo6GC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,2wBAA2wBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,4aAA4aC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,ovBAAovBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,gwBAAgwBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,4MAA4MC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,miDAAmiDC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,8aAA8a,gBAAgB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,yvBAAyvBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,mfAAmfC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,opBAAopBC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,48BAA48BC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,q6CAAq6C,yBAAyB,CAAC,IAAI,IAAI,GAAG,OAAO,4jBAA4jB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,wxGAAwxGC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,0OAA0OC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,85BAA85BC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,4tDAA4tDC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,6gBAA6gBC,SAAS,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,gRAAgRC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,smBAAsmB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,opBAAopBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,+JAA+J,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,+hHAA+hHC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,w7BAAw7BC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,g4DAAg4DC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,2zDAA2zD,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,sxCAAsxCC,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,2MAA2MC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,03BAA03B,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,+qBAA+qBC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,kxBAAkxB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,0PAA0PC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,quBAAquBC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,smCAAsmC,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,6IAA6IC,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,8KAA8KC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,ycAAycC,UAAU,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,sOAAsOC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,m/DAAm/DC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,uRAAuR,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,+pHAA+pHC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,+NAA+NC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,2kBAA2kBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,wnEAAwnEC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,sSAAsSC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,sHAAsH,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,21IAA21IC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,wiBAAwiB,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,4gBAA4gB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,8UAA8U,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,giBAAgiB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,uQAAuQC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,8MAA8MC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,29BAA29BC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,2MAA2MC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,8mCAA8mC,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,qfAAqfC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,m8FAAm8FC,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,iPAAiPC,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,8YAA8YC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,0mIAA0mI,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,6vBAA6vBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,ozBAAozBC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,4gEAA4gE,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,mfAAmfC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,qnDAAqnD,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,+6BAA+6BC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,i9BAAi9BC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,+iBAA+iB,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,+rBAA+rBC,SAAS,CAAC,IAAI,IAAI,CAAC,MAAM,kBAAkB,OAAO,0jGAA0jG,uBAAuB,CAAC,IAAI,IAAI,GAAG,OAAO,mgCAAmgCC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,wvHAAwvHC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,goCAAgoC,iBAAiB,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,otBAAotBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,omDAAomDC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,qKAAqKC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,2iBAA2iBC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,mIAAmIC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,+sBAA+sBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,kPAAkP,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,2LAA2LC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,gLAAgLC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,20CAA20C,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,uoBAAuoB,aAAa,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,6oCAA6oC,gBAAgB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,keAAkeC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,wZAAwZ,yBAAyB,CAAC,IAAI,IAAI,GAAG,OAAO,0fAA0f,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,4VAA4VC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,8pBAA8pBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,syKAAsyKC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,o6BAAo6BC,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,s+FAAs+FC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,+jCAA+jC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,qGAAqGC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,wqBAAwqBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,gwBAAgwBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,iTAAiTC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,o8DAAo8D,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,spCAAspCC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,0/BAA0/BC,GAAG,CAAC,IAAI,IAAI,GAAG,OAAO,uhBAAuhBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,6wBAA6wBC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,+/BAA+/BC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,irBAAirBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,8UAA8U,uBAAuB,CAAC,IAAI,IAAI,GAAG,OAAO,gFAAgF,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,6lBAA6lBC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,gYAAgYC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,4xCAA4xCC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,wwBAAwwBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,6pBAA6pBC,QAAQ,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,0dAA0dC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,upBAAupB,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,6mDAA6mDC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,qsCAAqsCC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,stDAAstDC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,yqIAAyqIC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,2xBAA2xBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,2QAA2QC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,8pBAA8pBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,ynBAAynBC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,kLAAkLC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,w3CAAw3CC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,0qBAA0qBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,+hCAA+hC,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,+fAA+f,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,61BAA61BC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,uUAAuUC,SAAS,CAAC,IAAI,IAAI,CAAC,MAAM,kBAAkB,OAAO,ksBAAksB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,u4VAAu4VC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,q9DAAq9D,kBAAkB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,wjCAAwjC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,gkCAAgkC,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,8ZAA8ZC,MAAM,CAAC,IAAI,IAAI,CAAC,MAAM,cAAc,OAAO,i+BAAi+BC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,wvBAAwvBC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,6qBAA6qBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,iXAAiXC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,irBAAirBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,gfAAgf,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,0QAA0QC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,oyKAAoyKC,OAAO,CAAC,IAAI,IAAI,CAAC,MAAM,YAAY,OAAO,qaAAqaC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,ihBAAihBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,mbAAmbC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,gFAAgFC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,gKAAgK,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,i5NAAi5NC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,gRAAgRC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,2QAA2QC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,6iCAA6iCC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,2mBAA2mBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,qwBAAqwB,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,41EAA41E,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,4wBAA4wBC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,69EAA69E,iCAAiC,CAAC,IAAI,IAAI,GAAG,OAAO,ukCAAukCC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,4GAA4GC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,i0DAAi0D,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,sxBAAsxBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,0xDAA0xDC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,mdAAmd,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,u3BAAu3BC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,84CAA84CC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,2lCAA2lC,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,63CAA63CC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,0kBAA0kBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,47CAA47CC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,8PAA8PC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,43BAA43B,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,4mBAA4mB,SAASnb,GAAG,IAAI,IAAI,IAAIC,EAAEmB,UAAUC,OAAOhB,EAAE,IAAI0B,MAAM,EAAE9B,EAAEA,EAAE,EAAE,GAAGG,EAAE,EAAEA,EAAEH,EAAEG,IAAIC,EAAED,EAAE,GAAGgB,UAAUhB,GAAGJ,EAAEkB,WAAM,EAAOb,EAAE,CAAC,MAAML,GAAG,IAAIsC,EAAE,MAAMtC,CAAC,CAAC,CAA3I,EAA6I,WAAW8F,EAAE,MAAMG,GAAGH,EAAE,YAAYG,EAAE,GAAE,CAArrne,GAAyrne,WAAW,aAAa,IAAIjG,EAAE,CAAC,EAAEC,EAAE,CAAC,EAAE,IAAI,oBAAoBC,SAASF,EAAEE,QAAQ,oBAAoBC,WAAWF,EAAEE,SAAS,CAAC,MAAMH,GAAG,CAAC,IAAkCI,OAAE,KAAhCC,GAAGL,EAAEM,WAAW,CAAC,GAAGC,WAAuB,GAAGF,EAAEG,EAAER,EAAES,EAAER,EAAyI,SAASS,EAAET,EAAED,GAAG,IAAIK,EAAED,EAAEO,OAAOC,KAAKX,GAAG,OAAOU,OAAOE,wBAAwBR,EAAEM,OAAOE,sBAAsBZ,GAAGD,IAAIK,EAAEA,EAAES,QAAO,SAASd,GAAG,OAAOW,OAAOI,yBAAyBd,EAAED,GAAGgB,UAAU,KAAIZ,EAAEa,KAAKC,MAAMd,EAAEC,IAAID,CAAC,CAAC,SAASe,EAAElB,GAAG,IAAI,IAAID,EAAE,EAAEA,EAAEoB,UAAUC,OAAOrB,IAAI,CAAC,IAAIK,EAAE,MAAMe,UAAUpB,GAAGoB,UAAUpB,GAAG,CAAC,EAAEA,EAAE,EAAEU,EAAEC,OAAON,IAAG,GAAIiB,SAAQ,SAAStB,GAAGuB,EAAEtB,EAAED,EAAEK,EAAEL,GAAG,IAAGW,OAAOa,0BAA0Bb,OAAOc,iBAAiBxB,EAAEU,OAAOa,0BAA0BnB,IAAIK,EAAEC,OAAON,IAAIiB,SAAQ,SAAStB,GAAGW,OAAOe,eAAezB,EAAED,EAAEW,OAAOI,yBAAyBV,EAAEL,GAAG,GAAE,CAAC,OAAOC,CAAC,CAAC,SAASsB,EAAEvB,EAAEC,EAAEI,GAAG,OAAOJ,KAAKD,EAAEW,OAAOe,eAAe1B,EAAEC,EAAE,CAAC0B,MAAMtB,EAAEW,YAAW,EAAGY,cAAa,EAAGC,UAAS,IAAK7B,EAAEC,GAAGI,EAAEL,CAAC,CAAC,SAAS8B,EAAE9B,EAAEC,IAAI,MAAMA,GAAGA,EAAED,EAAEqB,UAAUpB,EAAED,EAAEqB,QAAQ,IAAI,IAAIhB,EAAE,EAAED,EAAE,IAAI2B,MAAM9B,GAAGI,EAAEJ,EAAEI,IAAID,EAAEC,GAAGL,EAAEK,GAAG,OAAOD,CAAC,CAA95BI,EAAEL,SAASM,EAAEuB,iBAAiBvB,EAAEwB,MAAM,mBAAmBxB,EAAEyB,kBAAkBzB,EAAE0B,eAAe/B,EAAEgC,QAAQ,SAAShC,EAAEgC,QAAQ,YAAoyB,IAAIC,EAAE,qBAAqBC,EAAE,WAAW,IAAI,OAAM,CAAE,CAAC,MAAMtC,GAAG,OAAM,CAAE,CAAC,CAA1C,GAA8CuC,EAAE,UAAUC,EAAE,QAAQC,EAAE,CAACF,EAAEC,GAAG,SAASE,EAAE1C,GAAG,OAAO,IAAI2C,MAAM3C,EAAE,CAAC4C,IAAI,SAAS5C,EAAEC,GAAG,OAAOA,KAAKD,EAAEA,EAAEC,GAAGD,EAAEuC,EAAE,GAAG,CAACG,GAAGnB,EAAEsB,EAAE,CAAC,EAAEN,EAAE,CAACO,GAAG,QAAQC,IAAI,QAAQ,WAAW,QAAQC,IAAI,UAAU,aAAa,UAAUC,IAAI,QAAQ,WAAW,QAAQC,IAAI,OAAO,UAAU,OAAOC,IAAI,UAAU,aAAa,UAAUC,IAAI,SAAS,YAAY,SAASC,IAAI,MAAMC,KAAK,MAAM,SAAS,MAAM,iBAAiB,QAAQ/B,EAAEsB,EAAEL,EAAE,CAACM,GAAG,QAAQS,KAAK,QAAQ,WAAW,QAAQC,KAAK,UAAU,aAAa,UAAUC,KAAK,QAAQ,WAAW,QAAQC,KAAK,OAAO,UAAU,SAASb,IAAI,IAAIc,EAAEjB,GAAGnB,EAAEqC,EAAE,CAAC,EAAErB,EAAE,CAACsB,MAAM,MAAMC,QAAQ,MAAMC,MAAM,MAAMC,KAAK,MAAMC,QAAQ,MAAMC,OAAO,MAAMC,IAAI,QAAQ5C,EAAEqC,EAAEpB,EAAE,CAACqB,MAAM,OAAOC,QAAQ,OAAOC,MAAM,OAAOC,KAAK,SAASJ,IAA8mBf,GAA5CzC,GAA3jBsC,GAAGnB,EAAElB,EAAE,CAAC,EAAEkC,EAAE,CAACa,IAAI,YAAYD,IAAI,aAAaE,IAAI,SAASJ,IAAI,WAAWD,IAAI,aAAaD,IAAI,WAAWG,IAAI,YAAY3B,EAAElB,EAAEmC,EAAE,CAACe,KAAK,WAAWC,KAAK,aAAaC,KAAK,WAAWC,KAAK,YAAYrD,IAAIqC,GAAGnB,EAAEvB,EAAE,CAAC,EAAEuC,EAAE,CAAC,YAAY,MAAM,aAAa,MAAM,SAAS,MAAM,WAAW,MAAM,aAAa,MAAM,WAAW,MAAM,UAAU,QAAQhB,EAAEvB,EAAEwC,EAAE,CAAC,WAAW,OAAO,aAAa,OAAO,WAAW,OAAO,UAAU,SAASxC,IAAI0C,GAAGnB,EAAEtB,EAAE,CAAC,EAAEsC,EAAE,CAAC,IAAI,MAAM,IAAI,MAAM6B,OAAO,MAAM,IAAI,MAAM,IAAI,QAAQ7C,EAAEtB,EAAEuC,EAAE,CAAC,IAAI,OAAO,IAAI,OAAO,IAAI,OAAO,IAAI,SAASvC,IAAliBQ,EAAsiB,CAAC,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,KAAS4D,OAAO,CAAC,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,KAAO,iBAAgBT,EAAE,eAAevD,EAAE,UAAUL,EAAE,YAAYC,EAAE,IAAIqE,IAAI3D,OAAOC,KAAK+C,EAAEpB,IAAIgC,IAAItE,EAAEuE,IAAIC,KAAKxE,IAAIU,OAAOC,KAAK+C,EAAEnB,IAAI+B,IAAItE,EAAEuE,IAAIC,KAAKxE,IAAI,GAAGoE,OAAO5B,EAAE,SAASzC,GAAG,GAAG+B,MAAM2C,QAAQ1E,GAAG,OAAO8B,EAAE9B,EAAE,CAA3C,CAA+CC,IAAI,SAASD,GAAG,GAAG,oBAAoB2E,QAAQ,MAAM3E,EAAE2E,OAAOC,WAAW,MAAM5E,EAAE,cAAc,OAAO+B,MAAM8C,KAAK7E,EAAE,CAA/G,CAAiHC,IAAI,SAASD,EAAEC,GAAG,GAAGD,EAAE,CAAC,GAAG,iBAAiBA,EAAE,OAAO8B,EAAE9B,EAAEC,GAAG,IAAII,EAAEM,OAAOmE,UAAUC,SAASC,KAAKhF,GAAGiF,MAAM,GAAG,GAAG,MAAM,SAAS5E,EAAE,WAAWA,GAAGL,EAAEkF,YAAYlF,EAAEkF,YAAYC,KAAK9E,IAAI,QAAQA,EAAE0B,MAAM8C,KAAK7E,GAAG,cAAcK,GAAG,2CAA2C+E,KAAK/E,GAAGyB,EAAE9B,EAAEC,QAAG,CAAM,CAAC,CAA1R,CAA4RA,IAAI,WAAW,MAAM,IAAIoF,UAAU,uIAAuI,CAAtK,GAA0K,CAAC,MAAM,KAAK,KAAK,KAAK,KAAK,MAAM,OAAO,SAAS,OAAO,YAAY,SAAS,YAAY,kBAAkB,gBAAgB,OAAO,KAAK,UAAU,iBAAiB,cAAc,SAAS,KAAK,YAAY,aAAa,QAAQ,aAAa,aAAa,YAAY,YAAY,QAAQ,aAAa,eAAe,OAAO,WAAW,WAAW,QAAQ,KAAKxC,EAAEe,EAAEvD,EAAEL,IAAIqE,OAAO5D,EAAE8D,KAAI,SAASvE,GAAG,MAAM,GAAGqE,OAAOrE,EAAE,IAAI,KAAIqE,OAAOjE,EAAEmE,KAAI,SAASvE,GAAG,MAAM,KAAKqE,OAAOrE,EAAE,MAAIQ,EAAEA,GAAG,CAAC,GAAI6B,KAAK7B,EAAE6B,GAAG,CAAC,GAAG7B,EAAE6B,GAAGiD,SAAS9E,EAAE6B,GAAGiD,OAAO,CAAC,GAAG9E,EAAE6B,GAAGkD,QAAQ/E,EAAE6B,GAAGkD,MAAM,CAAC,GAAG/E,EAAE6B,GAAGmD,QAAQhF,EAAE6B,GAAGmD,MAAM,IAAI,IAAIC,EAAEjF,EAAE6B,GAAG,SAASqD,EAAEtF,GAAG,OAAOO,OAAOC,KAAKR,GAAGuF,QAAO,SAAS3F,EAAEC,GAAG,IAAII,EAAED,EAAEH,GAAG,OAAQI,EAAEuF,KAAK5F,EAAEK,EAAEwF,UAAUxF,EAAEuF,KAAK5F,EAAEC,GAAGI,EAAEL,CAAC,GAAE,CAAC,EAAE,CAAC,SAAS8F,EAAE9F,EAAEC,EAAEI,GAAyDA,OAAE,KAApDD,GAAG,EAAEgB,UAAUC,aAAQ,IAAShB,EAAEA,EAAE,CAAC,GAAG0F,YAAwB3F,EAApE,IAAsEA,EAAEsF,EAAEzF,GAAG,mBAAmBwF,EAAEF,MAAMS,SAAS3F,EAAEoF,EAAEH,OAAOtF,GAAGmB,EAAEA,EAAE,CAAC,EAAEsE,EAAEH,OAAOtF,IAAI,CAAC,GAAGI,GAAGqF,EAAEF,MAAMS,QAAQhG,EAAE0F,EAAEzF,IAAI,QAAQD,GAAG8F,EAAE,KAAK7F,EAAE,CAAC,IAAIgG,EAAE,CAAC,YAAY,CAAC,IAAI,IAAI,CAAC,MAAM,aAAa,OAAO,unBAAunBmV,QAAQ,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,8XAA8X,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,MAAM,WAAW,aAAa,OAAO,8aAA8a,gBAAgB,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,+bAA+b,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,o7BAAo7B,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,8lCAA8lC,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,21BAA21B,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,gBAAgB,OAAO,mlBAAmlBC,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,6mCAA6mCC,MAAM,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,6hBAA6hB,0BAA0B,CAAC,IAAI,IAAI,CAAC,OAAO,sBAAsB,OAAO,i5BAAi5B,eAAe,CAAC,IAAI,IAAI,CAAC,OAAO,WAAW,OAAO,mcAAmc,qBAAqB,CAAC,IAAI,IAAI,CAAC,sBAAsB,OAAO,2YAA2Y,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,gBAAgB,OAAO,wSAAwSC,QAAQ,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,qTAAqT,oBAAoB,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,6YAA6Y,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,aAAa,OAAO,20DAA20DC,UAAU,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,8vBAA8vBC,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,2dAA2d,eAAe,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,MAAM,gBAAgB,OAAO,+VAA+V,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,iJAAiJ,aAAa,CAAC,IAAI,IAAI,CAAC,SAAS,OAAO,6lBAA6lBC,OAAO,CAAC,IAAI,IAAI,CAAC,KAAK,cAAc,eAAe,OAAO,i4BAAi4B,gBAAgB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2pBAA2pB,iBAAiB,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,0eAA0e,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,0hCAA0hCC,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,KAAK,cAAc,OAAO,26BAA26B,aAAa,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,y+BAAy+B,YAAY,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,OAAO,umCAAumC,mBAAmB,CAAC,IAAI,IAAI,CAAC,OAAO,eAAe,OAAO,qpBAAqpB,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,y1BAAy1B,iBAAiB,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,ggBAAggB,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,YAAY,OAAO,iaAAia,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,qWAAqW,gBAAgB,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,23BAA23BC,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,2NAA2N,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,06BAA06BC,OAAO,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,MAAM,gBAAgB,OAAO,wTAAwTC,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,+SAA+S,oBAAoB,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,yYAAyYC,KAAK,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,OAAO,klBAAklB,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,m2BAAm2B,oBAAoB,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,wjBAAwjB,aAAa,CAAC,IAAI,IAAI,CAAC,SAAS,OAAO,4TAA4T,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,OAAO,4YAA4YC,UAAU,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,4WAA4W,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,ozBAAozB,iBAAiB,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,uhCAAuhC,kBAAkB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,wYAAwY,YAAY,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,oaAAoa,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,wVAAwV,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,OAAO,mYAAmYC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,mWAAmW,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,gVAAgVC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,02BAA02BC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,80CAA80CC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,oVAAoV,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,OAAO,sRAAsR,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,gBAAgB,OAAO,oPAAoP,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,OAAO,kNAAkN,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,oZAAoZ,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,whCAAwhCC,WAAW,CAAC,IAAI,IAAI,CAAC,KAAK,OAAO,mWAAmW,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,SAAS,OAAO,0gBAA0gB,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,SAAS,OAAO,usBAAusB,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,MAAM,oBAAoB,0BAA0B,OAAO,qVAAqV,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,4hBAA4hB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,uiBAAuiB,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,waAAwa,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,ucAAuc,cAAc,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,kuBAAkuBC,OAAO,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,KAAK,OAAO,OAAO,+MAA+MC,UAAU,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,OAAO,0qCAA0qCC,UAAU,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,mvBAAmvB,uBAAuB,CAAC,IAAI,IAAI,CAAC,OAAO,mBAAmB,OAAO,opDAAopD,mBAAmB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,ifAAif,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,0hBAA0hB,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,aAAa,OAAO,skCAAskC,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,88CAA88C,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,0BAA0B,OAAO,uVAAuV,oBAAoB,CAAC,IAAI,IAAI,CAAC,OAAO,oBAAoB,OAAO,inBAAinB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,2qBAA2qBC,MAAM,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,KAAK,MAAM,OAAO,OAAO,4sBAA4sB,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,YAAY,OAAO,yOAAyO,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,gBAAgB,OAAO,kRAAkRC,OAAO,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,KAAK,KAAK,KAAK,MAAM,MAAM,OAAO,OAAO,qGAAqG,YAAY,CAAC,IAAI,IAAI,CAAC,MAAM,uBAAuB,OAAO,qVAAqV,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,2mBAA2mB,aAAa,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,6gBAA6gB,kBAAkB,CAAC,IAAI,IAAI,CAAC,MAAM,mBAAmB,OAAO,oeAAoe,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,aAAa,OAAO,6LAA6LC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,ktBAAktB,eAAe,CAAC,IAAI,IAAI,CAAC,OAAO,WAAW,OAAO,+vBAA+vB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,yhBAAyhB,kBAAkB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,mSAAmS,aAAa,CAAC,IAAI,IAAI,CAAC,KAAK,SAAS,OAAO,gZAAgZ,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,QAAQ,OAAO,maAAma,eAAe,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,cAAc,OAAO,0wBAA0wB,mBAAmB,CAAC,IAAI,IAAI,CAAC,OAAO,eAAe,OAAO,qmBAAqmB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,ikCAAikC,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,qiCAAqiC,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,22DAA22D,iBAAiB,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,wfAAwf,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,ueAAue,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,mwBAAmwBC,QAAQ,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,6qBAA6qBC,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,KAAK,OAAO,OAAO,mUAAmUC,UAAU,CAAC,IAAI,IAAI,CAAC,KAAK,MAAM,mBAAmB,OAAO,2eAA2e,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,waAAwa,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,gZAAgZ,yBAAyB,CAAC,IAAI,IAAI,CAAC,OAAO,qBAAqB,OAAO,4vCAA4vC,iBAAiB,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,smBAAsmB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,idAAid,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,yBAAyB,OAAO,sVAAsV,eAAe,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,yaAAyaC,MAAM,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,wOAAwOC,SAAS,CAAC,IAAI,IAAI,CAAC,KAAK,OAAO,6wCAA6wC,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,2sBAA2sBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,yhBAAyhB,YAAY,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,OAAO,wWAAwW,WAAW,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,sRAAsR,UAAU,CAAC,IAAI,IAAI,CAAC,MAAM,mBAAmB,OAAO,wbAAwbC,IAAI,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,8+BAA8+B,kBAAkB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,odAAod,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,yBAAyB,OAAO,uVAAuV,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,ujCAAujC,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,yfAAyf,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,mBAAmB,OAAO,mZAAmZC,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,+fAA+fC,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,OAAO,mQAAmQC,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,eAAe,iBAAiB,OAAO,kkHAAkkH,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,0tBAA0tB,YAAY,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,mXAAmX,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,qoBAAqoB,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,OAAO,obAAobC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,maAAma,oBAAoB,CAAC,IAAI,IAAI,CAAC,MAAM,gBAAgB,OAAO,i+BAAi+BC,UAAU,CAAC,IAAI,IAAI,CAAC,KAAK,OAAO,4TAA4ThZ,IAAI,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,6XAA6X,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,wpBAAwpB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,uiBAAuiB,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,SAAS,OAAO,oYAAoY,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,8iCAA8iC,mBAAmB,CAAC,IAAI,IAAI,CAAC,OAAO,eAAe,OAAO,0nBAA0nBiZ,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,ooBAAooB,uBAAuB,CAAC,IAAI,IAAI,CAAC,OAAO,mBAAmB,OAAO,ygDAAygDC,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,OAAO,+qBAA+qBC,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,OAAO,uRAAuR,wBAAwB,CAAC,IAAI,IAAI,CAAC,OAAO,oBAAoB,OAAO,m4BAAm4BC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,yXAAyX,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,SAAS,OAAO,ytBAAytB,kBAAkB,CAAC,IAAI,IAAI,CAAC,MAAM,kBAAkB,kBAAkB,gBAAgB,OAAO,qbAAqb,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,8TAA8T,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,m/BAAm/B,eAAe,CAAC,IAAI,IAAI,CAAC,OAAO,WAAW,OAAO,saAAsa,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,mXAAmX,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,gBAAgB,OAAO,0UAA0U,YAAY,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,sjCAAsjC,kBAAkB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2FAA2F,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,MAAM,OAAO,OAAO,OAAO,MAAM,OAAO,OAAO,sEAAsE,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,OAAO,4QAA4Q,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,YAAY,OAAO,whBAAwhB,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,s7BAAs7B,SAAS3d,GAAG,IAAI,IAAI,IAAIC,EAAEmB,UAAUC,OAAOhB,EAAE,IAAI0B,MAAM,EAAE9B,EAAEA,EAAE,EAAE,GAAGG,EAAE,EAAEA,EAAEH,EAAEG,IAAIC,EAAED,EAAE,GAAGgB,UAAUhB,GAAGJ,EAAEkB,WAAM,EAAOb,EAAE,CAAC,MAAML,GAAG,IAAIsC,EAAE,MAAMtC,CAAC,CAAC,CAA3I,EAA6I,WAAW8F,EAAE,MAAMG,GAAGH,EAAE,aAAaG,EAAE,GAAE,CAArr+G,GAAyr+G,WAAW,aAAa,IAAIjG,EAAE,CAAC,EAAEC,EAAE,CAAC,EAAE,IAAI,oBAAoBC,SAASF,EAAEE,QAAQ,oBAAoBC,WAAWF,EAAEE,SAAS,CAAC,MAAMH,GAAG,CAAC,IAAkCI,OAAE,KAAhCC,GAAGL,EAAEM,WAAW,CAAC,GAAGC,WAAuB,GAAGF,EAAEG,EAAER,EAAES,EAAER,EAAyI,SAASS,EAAET,EAAED,GAAG,IAAIK,EAAED,EAAEO,OAAOC,KAAKX,GAAG,OAAOU,OAAOE,wBAAwBR,EAAEM,OAAOE,sBAAsBZ,GAAGD,IAAIK,EAAEA,EAAES,QAAO,SAASd,GAAG,OAAOW,OAAOI,yBAAyBd,EAAED,GAAGgB,UAAU,KAAIZ,EAAEa,KAAKC,MAAMd,EAAEC,IAAID,CAAC,CAAC,SAASe,EAAElB,GAAG,IAAI,IAAID,EAAE,EAAEA,EAAEoB,UAAUC,OAAOrB,IAAI,CAAC,IAAIK,EAAE,MAAMe,UAAUpB,GAAGoB,UAAUpB,GAAG,CAAC,EAAEA,EAAE,EAAEU,EAAEC,OAAON,IAAG,GAAIiB,SAAQ,SAAStB,GAAGuB,EAAEtB,EAAED,EAAEK,EAAEL,GAAG,IAAGW,OAAOa,0BAA0Bb,OAAOc,iBAAiBxB,EAAEU,OAAOa,0BAA0BnB,IAAIK,EAAEC,OAAON,IAAIiB,SAAQ,SAAStB,GAAGW,OAAOe,eAAezB,EAAED,EAAEW,OAAOI,yBAAyBV,EAAEL,GAAG,GAAE,CAAC,OAAOC,CAAC,CAAC,SAASsB,EAAEvB,EAAEC,EAAEI,GAAG,OAAOJ,KAAKD,EAAEW,OAAOe,eAAe1B,EAAEC,EAAE,CAAC0B,MAAMtB,EAAEW,YAAW,EAAGY,cAAa,EAAGC,UAAS,IAAK7B,EAAEC,GAAGI,EAAEL,CAAC,CAAC,SAAS8B,EAAE9B,EAAEC,IAAI,MAAMA,GAAGA,EAAED,EAAEqB,UAAUpB,EAAED,EAAEqB,QAAQ,IAAI,IAAIhB,EAAE,EAAED,EAAE,IAAI2B,MAAM9B,GAAGI,EAAEJ,EAAEI,IAAID,EAAEC,GAAGL,EAAEK,GAAG,OAAOD,CAAC,CAA95BI,EAAEL,SAASM,EAAEuB,iBAAiBvB,EAAEwB,MAAM,mBAAmBxB,EAAEyB,kBAAkBzB,EAAE0B,eAAe/B,EAAEgC,QAAQ,SAAShC,EAAEgC,QAAQ,YAAoyB,IAAIC,EAAE,qBAAqBC,EAAE,WAAW,IAAI,OAAM,CAAE,CAAC,MAAMtC,GAAG,OAAM,CAAE,CAAC,CAA1C,GAA8CuC,EAAE,UAAUC,EAAE,QAAQC,EAAE,CAACF,EAAEC,GAAG,SAASE,EAAE1C,GAAG,OAAO,IAAI2C,MAAM3C,EAAE,CAAC4C,IAAI,SAAS5C,EAAEC,GAAG,OAAOA,KAAKD,EAAEA,EAAEC,GAAGD,EAAEuC,EAAE,GAAG,CAACG,GAAGnB,EAAEsB,EAAE,CAAC,EAAEN,EAAE,CAACO,GAAG,QAAQC,IAAI,QAAQ,WAAW,QAAQC,IAAI,UAAU,aAAa,UAAUC,IAAI,QAAQ,WAAW,QAAQC,IAAI,OAAO,UAAU,OAAOC,IAAI,UAAU,aAAa,UAAUC,IAAI,SAAS,YAAY,SAASC,IAAI,MAAMC,KAAK,MAAM,SAAS,MAAM,iBAAiB,QAAQ/B,EAAEsB,EAAEL,EAAE,CAACM,GAAG,QAAQS,KAAK,QAAQ,WAAW,QAAQC,KAAK,UAAU,aAAa,UAAUC,KAAK,QAAQ,WAAW,QAAQC,KAAK,OAAO,UAAU,SAASb,IAAI,IAAIc,EAAEjB,GAAGnB,EAAEqC,EAAE,CAAC,EAAErB,EAAE,CAACsB,MAAM,MAAMC,QAAQ,MAAMC,MAAM,MAAMC,KAAK,MAAMC,QAAQ,MAAMC,OAAO,MAAMC,IAAI,QAAQ5C,EAAEqC,EAAEpB,EAAE,CAACqB,MAAM,OAAOC,QAAQ,OAAOC,MAAM,OAAOC,KAAK,SAASJ,IAA8mBf,GAA5CzC,GAA3jBsC,GAAGnB,EAAElB,EAAE,CAAC,EAAEkC,EAAE,CAACa,IAAI,YAAYD,IAAI,aAAaE,IAAI,SAASJ,IAAI,WAAWD,IAAI,aAAaD,IAAI,WAAWG,IAAI,YAAY3B,EAAElB,EAAEmC,EAAE,CAACe,KAAK,WAAWC,KAAK,aAAaC,KAAK,WAAWC,KAAK,YAAYrD,IAAIqC,GAAGnB,EAAEvB,EAAE,CAAC,EAAEuC,EAAE,CAAC,YAAY,MAAM,aAAa,MAAM,SAAS,MAAM,WAAW,MAAM,aAAa,MAAM,WAAW,MAAM,UAAU,QAAQhB,EAAEvB,EAAEwC,EAAE,CAAC,WAAW,OAAO,aAAa,OAAO,WAAW,OAAO,UAAU,SAASxC,IAAI0C,GAAGnB,EAAEtB,EAAE,CAAC,EAAEsC,EAAE,CAAC,IAAI,MAAM,IAAI,MAAM6B,OAAO,MAAM,IAAI,MAAM,IAAI,QAAQ7C,EAAEtB,EAAEuC,EAAE,CAAC,IAAI,OAAO,IAAI,OAAO,IAAI,OAAO,IAAI,SAASvC,IAAliBQ,EAAsiB,CAAC,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,KAAS4D,OAAO,CAAC,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,KAAO,iBAAgBT,EAAE,eAAevD,EAAE,UAAUL,EAAE,YAAYC,EAAE,IAAIqE,IAAI3D,OAAOC,KAAK+C,EAAEpB,IAAIgC,IAAItE,EAAEuE,IAAIC,KAAKxE,IAAIU,OAAOC,KAAK+C,EAAEnB,IAAI+B,IAAItE,EAAEuE,IAAIC,KAAKxE,IAAI,GAAGoE,OAAO5B,EAAE,SAASzC,GAAG,GAAG+B,MAAM2C,QAAQ1E,GAAG,OAAO8B,EAAE9B,EAAE,CAA3C,CAA+CC,IAAI,SAASD,GAAG,GAAG,oBAAoB2E,QAAQ,MAAM3E,EAAE2E,OAAOC,WAAW,MAAM5E,EAAE,cAAc,OAAO+B,MAAM8C,KAAK7E,EAAE,CAA/G,CAAiHC,IAAI,SAASD,EAAEC,GAAG,GAAGD,EAAE,CAAC,GAAG,iBAAiBA,EAAE,OAAO8B,EAAE9B,EAAEC,GAAG,IAAII,EAAEM,OAAOmE,UAAUC,SAASC,KAAKhF,GAAGiF,MAAM,GAAG,GAAG,MAAM,SAAS5E,EAAE,WAAWA,GAAGL,EAAEkF,YAAYlF,EAAEkF,YAAYC,KAAK9E,IAAI,QAAQA,EAAE0B,MAAM8C,KAAK7E,GAAG,cAAcK,GAAG,2CAA2C+E,KAAK/E,GAAGyB,EAAE9B,EAAEC,QAAG,CAAM,CAAC,CAA1R,CAA4RA,IAAI,WAAW,MAAM,IAAIoF,UAAU,uIAAuI,CAAtK,GAA0K,CAAC,MAAM,KAAK,KAAK,KAAK,KAAK,MAAM,OAAO,SAAS,OAAO,YAAY,SAAS,YAAY,kBAAkB,gBAAgB,OAAO,KAAK,UAAU,iBAAiB,cAAc,SAAS,KAAK,YAAY,aAAa,QAAQ,aAAa,aAAa,YAAY,YAAY,QAAQ,aAAa,eAAe,OAAO,WAAW,WAAW,QAAQ,KAAKxC,EAAEe,EAAEvD,EAAEL,IAAIqE,OAAO5D,EAAE8D,KAAI,SAASvE,GAAG,MAAM,GAAGqE,OAAOrE,EAAE,IAAI,KAAIqE,OAAOjE,EAAEmE,KAAI,SAASvE,GAAG,MAAM,KAAKqE,OAAOrE,EAAE,MAAIQ,EAAEA,GAAG,CAAC,GAAI6B,KAAK7B,EAAE6B,GAAG,CAAC,GAAG7B,EAAE6B,GAAGiD,SAAS9E,EAAE6B,GAAGiD,OAAO,CAAC,GAAG9E,EAAE6B,GAAGkD,QAAQ/E,EAAE6B,GAAGkD,MAAM,CAAC,GAAG/E,EAAE6B,GAAGmD,QAAQhF,EAAE6B,GAAGmD,MAAM,IAAI,IAAIC,EAAEjF,EAAE6B,GAAG,SAASqD,EAAEtF,GAAG,OAAOO,OAAOC,KAAKR,GAAGuF,QAAO,SAAS3F,EAAEC,GAAG,IAAII,EAAED,EAAEH,GAAG,OAAQI,EAAEuF,KAAK5F,EAAEK,EAAEwF,UAAUxF,EAAEuF,KAAK5F,EAAEC,GAAGI,EAAEL,CAAC,GAAE,CAAC,EAAE,CAAC,SAAS8F,EAAE9F,EAAEC,EAAEI,GAAyDA,OAAE,KAApDD,GAAG,EAAEgB,UAAUC,aAAQ,IAAShB,EAAEA,EAAE,CAAC,GAAG0F,YAAwB3F,EAApE,IAAsEA,EAAEsF,EAAEzF,GAAG,mBAAmBwF,EAAEF,MAAMS,SAAS3F,EAAEoF,EAAEH,OAAOtF,GAAGmB,EAAEA,EAAE,CAAC,EAAEsE,EAAEH,OAAOtF,IAAI,CAAC,GAAGI,GAAGqF,EAAEF,MAAMS,QAAQhG,EAAE0F,EAAEzF,IAAI,QAAQD,GAAG8F,EAAE,KAAK7F,EAAE,CAAC,IAAIgG,EAAE,CAAC,EAAE,CAAC,IAAI,IAAI,GAAG,KAAK,iLAAiL,EAAE,CAAC,IAAI,IAAI,GAAG,KAAK,gNAAgN,EAAE,CAAC,IAAI,IAAI,GAAG,KAAK,oWAAoW,EAAE,CAAC,IAAI,IAAI,GAAG,KAAK,sXAAsX,EAAE,CAAC,IAAI,IAAI,GAAG,KAAK,8PAA8P,EAAE,CAAC,IAAI,IAAI,GAAG,KAAK,oWAAoW,EAAE,CAAC,IAAI,IAAI,GAAG,KAAK,0QAA0Q,EAAE,CAAC,IAAI,IAAI,GAAG,KAAK,wKAAwK,EAAE,CAAC,IAAI,IAAI,GAAG,KAAK,wcAAwc,EAAE,CAAC,IAAI,IAAI,GAAG,KAAK,qPAAqP,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,wkBAAwkB,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,4jCAA4jC,uBAAuB,CAAC,IAAI,IAAI,CAAC,wBAAwB,OAAO,mMAAmM2X,GAAG,CAAC,IAAI,IAAI,CAAC,OAAO,KAAK,wbAAwb,YAAY,CAAC,IAAI,IAAI,CAAC,MAAM,aAAa,OAAO,2eAA2e,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,giBAAgiB,aAAa,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,4ZAA4ZC,YAAY,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,ijBAAijBzC,QAAQ,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,oKAAoK0C,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,oNAAoN,mCAAmC,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,oaAAoaC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,wpBAAwpB,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,MAAM,WAAW,aAAa,OAAO,0XAA0X,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,sQAAsQC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,8iBAA8iB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,ixBAAixB,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,0OAA0O,gBAAgB,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,gzBAAgzB,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,krBAAkrB,8BAA8B,CAAC,IAAI,IAAI,GAAG,OAAO,uiCAAuiCC,WAAW,CAAC,IAAI,IAAI,CAAC,OAAO,mBAAmB,OAAO,kxBAAkxB,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,8gBAA8gB,YAAY,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,+JAA+J,eAAe,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,8IAA8I,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,6UAA6U,qBAAqB,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,qcAAqcC,KAAK,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,2mCAA2mCC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,0cAA0cC,MAAM,CAAC,IAAI,IAAI,CAAC,2BAA2B,OAAO,81BAA81B,yBAAyB,CAAC,IAAI,IAAI,CAAC,wBAAwB,OAAO,kqBAAkqB,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,uiBAAuiB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,2gBAA2gBC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,yzCAAyzC,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,qrBAAqrB,4BAA4B,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,mWAAmW,eAAe,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,0NAA0N,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,ijCAAijC,iBAAiB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,i4BAAi4BC,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,iBAAiB,OAAO,swBAAswB,4BAA4B,CAAC,IAAI,IAAI,GAAG,OAAO,s1BAAs1BC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,+UAA+U,cAAc,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,mVAAmV,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,qxBAAqxB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,s/BAAs/B,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,46BAA46B,iBAAiB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,mgBAAmgB,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,+ZAA+Z,gBAAgB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,uNAAuN,oBAAoB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,4dAA4dC,OAAO,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,OAAO,uVAAuVC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,+MAA+M,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,+8BAA+8BC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,+rDAA+rD,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,ouBAAouB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,sLAAsL,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,+yBAA+yBC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,6mBAA6mB,2BAA2B,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,g6EAAg6E,4BAA4B,CAAC,IAAI,IAAI,GAAG,OAAO,giBAAgiBC,MAAM,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,iBAAiB,OAAO,0MAA0M,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,owBAAowB,iBAAiB,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,2iBAA2iB,0BAA0B,CAAC,IAAI,IAAI,GAAG,OAAO,mXAAmX,WAAW,CAAC,IAAI,IAAI,CAAC,KAAK,mBAAmB,OAAO,kPAAkPC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,qYAAqY,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,gBAAgB,OAAO,gfAAgfC,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,g3BAAg3BC,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,iPAAiP,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,6XAA6XC,OAAO,CAAC,IAAI,IAAI,CAAC,KAAK,MAAM,cAAc,OAAO,2mBAA2mBC,SAAS,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,qTAAqT,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,gKAAgK5D,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,+sBAA+sBC,MAAM,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,sXAAsX,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,snBAAsnB,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,6cAA6c,iBAAiB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,ydAAyd,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,qUAAqU,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,y6BAAy6B,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,iWAAiW4D,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,wbAAwb,aAAa,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,4qBAA4qB,4BAA4B,CAAC,IAAI,IAAI,GAAG,OAAO,wkBAAwkB,eAAe,CAAC,IAAI,IAAI,CAAC,SAAS,OAAO,g2BAAg2B,aAAa,CAAC,IAAI,IAAI,CAAC,YAAY,wBAAwB,OAAO,giBAAgiB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,kPAAkP,cAAc,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,sQAAsQ,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,m9BAAm9B,iBAAiB,CAAC,IAAI,IAAI,CAAC,oBAAoB,yBAAyB,OAAO,6pBAA6pB,0BAA0B,CAAC,IAAI,IAAI,CAAC,OAAO,sBAAsB,OAAO,k2BAAk2B,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,gdAAgd,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,6vDAA6vDC,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,sUAAsU,eAAe,CAAC,IAAI,IAAI,CAAC,OAAO,gBAAgB,OAAO,61BAA61BC,QAAQ,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,qiBAAqiB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,+SAA+S,uBAAuB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,6KAA6KC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,sqBAAsqBC,MAAM,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,+PAA+P,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,klBAAklB,eAAe,CAAC,IAAI,IAAI,CAAC,OAAO,WAAW,OAAO,0TAA0T,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,omBAAomB,wBAAwB,CAAC,IAAI,IAAI,GAAG,OAAO,0qBAA0qBC,MAAM,CAAC,IAAI,IAAI,CAAC,SAAS,OAAO,8fAA8fC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,+cAA+c,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,+hBAA+hB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,wZAAwZC,WAAW,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,05BAA05B,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,kBAAkB,OAAO,+PAA+P,YAAY,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,2qBAA2qB,gBAAgB,CAAC,IAAI,IAAI,CAAC,4BAA4B,OAAO,i1BAAi1B,uBAAuB,CAAC,IAAI,IAAI,GAAG,OAAO,q0BAAq0B,qBAAqB,CAAC,IAAI,IAAI,CAAC,sBAAsB,OAAO,4RAA4RC,SAAS,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,KAAK,OAAO,OAAO,8fAA8f,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,wlCAAwlC,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,0gBAA0gB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,mPAAmPC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,qMAAqM,+BAA+B,CAAC,IAAI,IAAI,GAAG,OAAO,uXAAuX,qBAAqB,CAAC,IAAI,IAAI,CAAC,sBAAsB,OAAO,slBAAslB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,8tBAA8tBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,srCAAsrCC,MAAM,CAAC,IAAI,IAAI,CAAC,MAAM,cAAc,OAAO,mWAAmW,oBAAoB,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,8nBAA8nB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,mqBAAmqB,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,gBAAgB,OAAO,oMAAoMC,WAAW,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,qjBAAqjBvE,QAAQ,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,qQAAqQ,oBAAoB,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,+RAA+R,uBAAuB,CAAC,IAAI,IAAI,GAAG,OAAO,2kBAA2kB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,ueAAuewE,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,qXAAqX,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,6SAA6SC,KAAK,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,2RAA2R,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,+5BAA+5B,gBAAgB,CAAC,IAAI,IAAI,CAAC,KAAK,eAAe,OAAO,2aAA2a,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,gBAAgB,OAAO,4aAA4a,kBAAkB,CAAC,IAAI,IAAI,CAAC,KAAK,4BAA4B,OAAO,sXAAsX,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,aAAa,OAAO,8wDAA8wDC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,q5BAAq5B,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,wWAAwW,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,qqBAAqqBzE,UAAU,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,gaAAga,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,8JAA8J,qBAAqB,CAAC,IAAI,IAAI,CAAC,sBAAsB,OAAO,8LAA8L,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,y9BAAy9B,2BAA2B,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,+ZAA+Z,sBAAsB,CAAC,IAAI,IAAI,CAAC,uBAAuB,OAAO,kMAAkM,iBAAiB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,+VAA+V,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,4wBAA4wB,oBAAoB,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,uZAAuZ0E,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,khBAAkhB,wBAAwB,CAAC,IAAI,IAAI,CAAC,UAAU,OAAO,yqCAAyqCC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,2vBAA2vB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,yTAAyT,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,80BAA80B1E,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,gTAAgT2E,SAAS,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,q7DAAq7DC,QAAQ,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,8gBAA8gB,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,eAAe,OAAO,4YAA4Y,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,skBAAskB,iBAAiB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,+vCAA+vCC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,4cAA4c,YAAY,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,ojCAAojCC,KAAK,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,8YAA8Y,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,gmCAAgmCC,SAAS,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,KAAK,qXAAqX,eAAe,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,MAAM,gBAAgB,OAAO,4PAA4P,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,uTAAuTC,QAAQ,CAAC,IAAI,IAAI,CAAC,UAAU,OAAO,0YAA0YC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,kcAAkcC,KAAK,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,skBAAskB,oBAAoB,CAAC,IAAI,IAAI,CAAC,oBAAoB,OAAO,4XAA4X,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,uhBAAuhBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,qWAAqW,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,gGAAgG,aAAa,CAAC,IAAI,IAAI,CAAC,SAAS,OAAO,gnBAAgnBC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,oRAAoR,6BAA6B,CAAC,IAAI,IAAI,GAAG,OAAO,upBAAupBC,WAAW,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,yRAAyR,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,sWAAsWpF,OAAO,CAAC,IAAI,IAAI,CAAC,KAAK,cAAc,eAAe,OAAO,qtBAAqtBqF,WAAW,CAAC,IAAI,IAAI,CAAC,OAAO,eAAe,OAAO,sVAAsVC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,wNAAwN,WAAW,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,0eAA0e,iBAAiB,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,k/BAAk/B,eAAe,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,uhBAAuhB,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,q9CAAq9CC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,sdAAsd,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,sjBAAsjB,cAAc,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,s/BAAs/B,gBAAgB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,wfAAwf,WAAW,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,ofAAof,cAAc,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,gWAAgW,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,geAAge,iBAAiB,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,oeAAoeC,WAAW,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,+cAA+cC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,sbAAsb,eAAe,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,sSAAsS,iBAAiB,CAAC,IAAI,IAAI,CAAC,kBAAkB,uBAAuB,OAAO,8qBAA8qBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,uUAAuU,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,wvBAAwvBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,6lBAA6lB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,qwBAAqwB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,2gBAA2gB1F,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,KAAK,cAAc,OAAO,waAAwa2F,GAAG,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,mpEAAmpEC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,wVAAwV,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,umBAAumBC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,iTAAiT,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,4KAA4K,aAAa,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2nBAA2nBC,UAAU,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,6XAA6X,YAAY,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,OAAO,ijCAAijC,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,woBAAwoB,mBAAmB,CAAC,IAAI,IAAI,CAAC,OAAO,eAAe,OAAO,omBAAomB,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,ucAAuc,iBAAiB,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,waAAwa,WAAW,CAAC,IAAI,IAAI,CAAC,OAAO,WAAW,mBAAmB,OAAO,ivBAAivB,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,inBAAinB,YAAY,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,+TAA+T,aAAa,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,iaAAia,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,gjBAAgjB,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,gkBAAgkBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,yYAAyYC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,kQAAkQnf,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,wLAAwL,eAAe,CAAC,IAAI,IAAI,CAAC,KAAK,WAAW,OAAO,mqBAAmqBof,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,kdAAkd,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,YAAY,OAAO,8WAA8WC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,ugBAAugBje,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,sMAAsM,wBAAwB,CAAC,IAAI,IAAI,GAAG,OAAO,+YAA+Yke,QAAQ,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,60BAA60B,cAAc,CAAC,IAAI,IAAI,CAAC,gBAAgB,YAAY,qBAAqB,OAAO,mnBAAmnB,wBAAwB,CAAC,IAAI,IAAI,GAAG,OAAO,yyBAAyyB,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,aAAa,OAAO,+kBAA+kB,+BAA+B,CAAC,IAAI,IAAI,GAAG,OAAO,wsCAAwsCC,QAAQ,CAAC,IAAI,IAAI,CAAC,KAAK,MAAM,OAAO,ujCAAujC,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,6WAA6W,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,4YAA4YC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,4vBAA4vBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,qWAAqWC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,0lCAA0lC,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,6KAA6KC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,quBAAquBC,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,sYAAsYC,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,6TAA6T,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,qfAAqf,mBAAmB,CAAC,IAAI,IAAI,CAAC,OAAO,gBAAgB,gBAAgB,oBAAoB,OAAO,gkBAAgkB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,2aAA2aC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,idAAid,YAAY,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,qxBAAqxB,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,oTAAoT,gBAAgB,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,ipBAAipB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,4SAA4S1G,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,+KAA+K,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,sXAAsX,iBAAiB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,0mBAA0mB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,snBAAsnB2G,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,4OAA4OC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,upBAAupB,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,+oBAA+oB,wBAAwB,CAAC,IAAI,IAAI,GAAG,OAAO,2ZAA2Z3G,OAAO,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,MAAM,gBAAgB,OAAO,mLAAmL,gBAAgB,CAAC,IAAI,IAAI,CAAC,oBAAoB,OAAO,mYAAmY4G,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,8eAA8e,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,qUAAqU,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,geAAgeC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,0RAA0RC,MAAM,CAAC,IAAI,IAAI,CAAC,YAAY,YAAY,0BAA0B,OAAO,0WAA0W,sBAAsB,CAAC,IAAI,IAAI,CAAC,uBAAuB,OAAO,k4BAAk4BliB,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,oMAAoM,WAAW,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,0aAA0a,4BAA4B,CAAC,IAAI,IAAI,GAAG,OAAO,mgBAAmgBqb,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,uLAAuL,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,k3BAAk3B8G,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,4dAA4d,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,eAAe,OAAO,gZAAgZ,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,87BAA87B,oBAAoB,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,2RAA2RC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,4bAA4bC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,4SAA4SC,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,OAAO,+aAA+aC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,s9BAAs9BC,UAAU,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,wtBAAwtB,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,kZAAkZ,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,2jBAA2jB,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,kLAAkLjjB,EAAE,CAAC,IAAI,IAAI,CAAC,IAAI,KAAK,oPAAoP,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,qJAAqJ,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,q/BAAq/B,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,0cAA0c,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,kcAAkc,aAAa,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,8WAA8W,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,qWAAqW,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,weAAwekjB,IAAI,CAAC,IAAI,IAAI,CAAC,OAAO,UAAU,OAAO,8PAA8P,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,mzBAAmzB,qBAAqB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,wnCAAwnCnH,KAAK,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,OAAO,kVAAkVoH,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,6cAA6cC,MAAM,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,8NAA8NC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,0LAA0L,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,ytBAAytB,gBAAgB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,8PAA8PC,SAAS,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,mkBAAmkB,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,qrBAAqrBC,OAAO,CAAC,IAAI,IAAI,CAAC,0BAA0B,OAAO,wUAAwUC,QAAQ,CAAC,IAAI,IAAI,CAAC,OAAO,UAAU,OAAO,6uBAA6uB,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,WAAW,OAAO,oqBAAoqB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,giBAAgiB,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,2KAA2KC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,uzCAAuzC,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,qwBAAqwB,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,4lCAA4lC,2BAA2B,CAAC,IAAI,IAAI,GAAG,OAAO,ygBAAygB,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,yuBAAyuB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,4pBAA4pBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,oIAAoI,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,SAAS,OAAO,kKAAkK,yBAAyB,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,4ZAA4Z,aAAa,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,sfAAsfC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,8XAA8X,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,mkCAAmkC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,uiBAAuiB,kBAAkB,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,2aAA2a,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,+qBAA+qB,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,o3CAAo3C,uBAAuB,CAAC,IAAI,IAAI,GAAG,OAAO,knBAAknB,mBAAmB,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,qcAAqcC,MAAM,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,uSAAuSC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,qyBAAqyB,cAAc,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,irBAAirB,mBAAmB,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,s7BAAs7B,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,6pBAA6pBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,8vBAA8vBC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2aAA2a,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,2SAA2S,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,01BAA01B,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,khBAAkhB,eAAe,CAAC,IAAI,IAAI,CAAC,UAAU,OAAO,8WAA8W,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,gBAAgB,OAAO,6eAA6e,eAAe,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,orBAAorB,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,mhBAAmhB,oBAAoB,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,sgBAAsgBC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,8pBAA8pB,kBAAkB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,qPAAqP,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,sIAAsI,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,WAAW,OAAO,0rBAA0rB/jB,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,6HAA6HgkB,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,ygBAAygB,YAAY,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,imBAAimB,gBAAgB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,4fAA4f,aAAa,CAAC,IAAI,IAAI,CAAC,SAAS,OAAO,yQAAyQ,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,OAAO,6VAA6V,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,gjBAAgjB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,yfAAyfC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,q1CAAq1C,mBAAmB,CAAC,IAAI,IAAI,CAAC,OAAO,kBAAkB,OAAO,8eAA8e,aAAa,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,glBAAglBC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,yQAAyQ,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,8tBAA8tBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,8gBAA8gB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,otBAAotB,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,gtBAAgtB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,8QAA8Q,iBAAiB,CAAC,IAAI,IAAI,CAAC,KAAK,iBAAiB,OAAO,yaAAya,oBAAoB,CAAC,IAAI,IAAI,CAAC,OAAO,gBAAgB,OAAO,mnBAAmnBpI,UAAU,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,kSAAkS,2BAA2B,CAAC,IAAI,IAAI,GAAG,OAAO,keAAke,gBAAgB,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,mUAAmUqI,KAAK,CAAC,IAAI,IAAI,CAAC,SAAS,eAAe,OAAO,ucAAucC,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,WAAW,OAAO,ikBAAikBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,yWAAyW,WAAW,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,oeAAoeC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,qxBAAqxBC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,wkCAAwkCC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,iVAAiV,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,ssBAAssB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,s8BAAs8B,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,uSAAuS,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,qWAAqW,kBAAkB,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,sbAAsb,mBAAmB,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,6jBAA6jBC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,k5CAAk5C,uBAAuB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,6XAA6X,UAAU,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,ilBAAilB,cAAc,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,+cAA+cC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,4jBAA4jB,WAAW,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,oMAAoM,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,wkBAAwkBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,u5CAAu5C,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,4TAA4T,wBAAwB,CAAC,IAAI,IAAI,GAAG,OAAO,m/BAAm/B,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,waAAwa,oBAAoB,CAAC,IAAI,IAAI,CAAC,KAAK,iBAAiB,OAAO,iZAAiZ,iBAAiB,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,8ZAA8Z,kBAAkB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,+QAA+Q,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,quCAAquC,YAAY,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,gaAAga,gBAAgB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,8tCAA8tC,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,qRAAqRC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,wYAAwY,iBAAiB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,0sBAA0sBC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,mbAAmbC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,yZAAyZ,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,iXAAiX,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,OAAO,gSAAgS,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,8aAA8aC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,2mBAA2mBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,iTAAiTjJ,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,sYAAsYkJ,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,mXAAmXC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,o6BAAo6B,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,+JAA+JC,YAAY,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,uVAAuV,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,oMAAoM,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,irDAAirD,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,6iBAA6iB,6BAA6B,CAAC,IAAI,IAAI,GAAG,OAAO,2kCAA2kC,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,qBAAqB,OAAO,2PAA2P,cAAc,CAAC,IAAI,IAAI,CAAC,cAAc,kBAAkB,YAAY,OAAO,oWAAoW,gBAAgB,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,0JAA0JnJ,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,sfAAsf,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,slBAAslBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,u0BAAu0BC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2cAA2ckJ,MAAM,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,4dAA4dC,QAAQ,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,uuBAAuuBC,SAAS,CAAC,IAAI,IAAI,CAAC,uBAAuB,OAAO,k8BAAk8BC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,qSAAqSC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,ivBAAivB,eAAe,CAAC,IAAI,IAAI,CAAC,IAAI,sBAAsB,OAAO,oVAAoVC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,i5BAAi5B,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,OAAO,oOAAoOC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,ghBAAghBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,4YAA4Y,YAAY,CAAC,IAAI,IAAI,CAAC,KAAK,MAAM,QAAQ,OAAO,odAAodC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,iaAAia,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,gBAAgB,OAAO,qMAAqM,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,OAAO,mKAAmK,mBAAmB,CAAC,IAAI,IAAI,CAAC,oBAAoB,OAAO,qvBAAqvB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,6/BAA6/BC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,4YAA4Y,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,wiBAAwiBC,OAAO,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,0MAA0M,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,uUAAuUC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,2YAA2Y,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,u+BAAu+B,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,mfAAmf,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,okBAAokB,oBAAoB,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,sMAAsM,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,+iBAA+iBC,QAAQ,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,+sBAA+sB,yCAAyC,CAAC,IAAI,IAAI,GAAG,OAAO,qvCAAqvC,gBAAgB,CAAC,IAAI,IAAI,CAAC,IAAI,MAAM,cAAc,OAAO,ocAAocC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,+wCAA+wC,yBAAyB,CAAC,IAAI,IAAI,GAAG,OAAO,sYAAsY,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,gZAAgZ,kBAAkB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,+NAA+N,0BAA0B,CAAC,IAAI,IAAI,GAAG,OAAO,+fAA+fC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,muBAAmuB,YAAY,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,qgBAAqgBC,SAAS,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,0QAA0Q,iBAAiB,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,mvBAAmvB,uBAAuB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,olCAAolC,eAAe,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,0WAA0W,oBAAoB,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,qPAAqP,cAAc,CAAC,IAAI,IAAI,CAAC,uBAAuB,OAAO,uVAAuV,wBAAwB,CAAC,IAAI,IAAI,CAAC,wBAAwB,OAAO,8RAA8R,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,mdAAmdC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,8TAA8TC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,ieAAie,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,qcAAqcC,MAAM,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,wNAAwN,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,wnBAAwnB,oBAAoB,CAAC,IAAI,IAAI,CAAC,gBAAgB,gBAAgB,qBAAqB,OAAO,qeAAqeC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,kmBAAkmBpK,WAAW,CAAC,IAAI,IAAI,CAAC,KAAK,OAAO,4SAA4S,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,SAAS,OAAO,0eAA0e,wBAAwB,CAAC,IAAI,IAAI,CAAC,uBAAuB,OAAO,wwBAAwwBqK,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,yjBAAyjB,qBAAqB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,sYAAsYC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,2hBAA2hB,eAAe,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,scAAsc,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,SAAS,OAAO,8tBAA8tB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,2bAA2bC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,smBAAsmBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,2jBAA2jB,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,MAAM,oBAAoB,0BAA0B,OAAO,gQAAgQ,iBAAiB,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,oBAAoB,OAAO,+bAA+bC,QAAQ,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2aAA2a,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,yhBAAyhBC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,61BAA61BC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,ojCAAojC,wBAAwB,CAAC,IAAI,IAAI,GAAG,OAAO,itCAAitCC,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,UAAU,OAAO,sPAAsP,aAAa,CAAC,IAAI,IAAI,CAAC,KAAK,gBAAgB,OAAO,6UAA6U,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,+mBAA+mB,oBAAoB,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,0PAA0P,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,45DAA45D,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,ydAAyd,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,iuBAAiuB,wBAAwB,CAAC,IAAI,IAAI,CAAC,kBAAkB,oBAAoB,OAAO,+jBAA+jB,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,gBAAgB,OAAO,2jBAA2jB,aAAa,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,oeAAoe,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,0cAA0c,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,4cAA4c,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,+SAA+S,oBAAoB,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,+UAA+U,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,8lBAA8lB,mBAAmB,CAAC,IAAI,IAAI,CAAC,MAAM,iBAAiB,sBAAsB,OAAO,icAAicC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,uuBAAuuBC,WAAW,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,qOAAqO,mBAAmB,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,mTAAmT,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,kSAAkS,yBAAyB,CAAC,IAAI,IAAI,CAAC,uBAAuB,OAAO,8qBAA8qB,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,iPAAiP,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,wZAAwZ,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,UAAU,QAAQ,OAAO,gvBAAgvB,oBAAoB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,shBAAshB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,miBAAmiB,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,wRAAwR,gBAAgB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,6QAA6Q,aAAa,CAAC,IAAI,IAAI,CAAC,SAAS,OAAO,m2BAAm2BC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,kvBAAkvB,wBAAwB,CAAC,IAAI,IAAI,GAAG,OAAO,kiBAAkiB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,ynBAAynB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,wzBAAwzBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,mOAAmO,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,0iBAA0iB,sBAAsB,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,0ZAA0Z,cAAc,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,6oBAA6oB/K,OAAO,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,KAAK,OAAO,OAAO,6GAA6G,sBAAsB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,0PAA0PgL,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,kZAAkZ,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,gaAAgaC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,2MAA2M,0BAA0B,CAAC,IAAI,IAAI,GAAG,OAAO,whBAAwhBC,OAAO,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,4bAA4b,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,m8CAAm8CC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,k+CAAk+CjnB,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,8MAA8M,gBAAgB,CAAC,IAAI,IAAI,CAAC,KAAK,UAAU,OAAO,qyBAAqyB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,wlBAAwlBJ,EAAE,CAAC,IAAI,IAAI,CAAC,IAAI,KAAK,2OAA2O,yBAAyB,CAAC,IAAI,IAAI,CAAC,oBAAoB,OAAO,wxBAAwxB,kBAAkB,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,gmBAAgmB6F,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,6KAA6KsW,UAAU,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,OAAO,urCAAurCC,UAAU,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,8nBAA8nB,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,ijBAAijB,qBAAqB,CAAC,IAAI,IAAI,CAAC,sBAAsB,OAAO,uPAAuP,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,0iBAA0iBkL,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,0wBAA0wBC,KAAK,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,mTAAmT,UAAU,CAAC,IAAI,IAAI,CAAC,WAAW,gBAAgB,OAAO,ktBAAktB,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,y0BAAy0B,qBAAqB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,2gCAA2gC,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,ofAAof,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,mJAAmJC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,m2BAAm2B,uBAAuB,CAAC,IAAI,IAAI,CAAC,OAAO,mBAAmB,OAAO,mhDAAmhD9mB,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,yNAAyN+mB,SAAS,CAAC,IAAI,IAAI,CAAC,MAAM,MAAM,OAAO,KAAK,gXAAgX,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,u5BAAu5B,qBAAqB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,4lBAA4lB,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,yZAAyZ,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,khBAAkhB,eAAe,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,q5BAAq5B,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,oYAAoY,mBAAmB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,kdAAkdC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,kdAAkd,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,iBAAiB,OAAO,iNAAiN,8BAA8B,CAAC,IAAI,IAAI,GAAG,OAAO,+8BAA+8B,yBAAyB,CAAC,IAAI,IAAI,GAAG,OAAO,6bAA6b,6BAA6B,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,gaAAga,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,6rBAA6rB,WAAW,CAAC,IAAI,IAAI,CAAC,KAAK,MAAM,OAAO,OAAO,2hBAA2hB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,stCAAstC,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,wdAAwdnkB,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,uKAAuKokB,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,qbAAqbC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2XAA2XC,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,+gBAA+gB,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,wjBAAwjB,YAAY,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,2VAA2V,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,2RAA2R,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,ieAAie,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,0ZAA0Z,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,2sBAA2sB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,kkBAAkkB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,i1BAAi1B,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,aAAa,OAAO,ghCAAghC,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,+1BAA+1BC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,mcAAmc,eAAe,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,yHAAyHC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,43BAA43B,eAAe,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,qhBAAqhB,gBAAgB,CAAC,IAAI,IAAI,CAAC,WAAW,YAAY,OAAO,uaAAuaC,MAAM,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,yQAAyQ,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,0BAA0B,OAAO,+SAA+S,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,6iBAA6iB,oBAAoB,CAAC,IAAI,IAAI,CAAC,OAAO,oBAAoB,OAAO,6fAA6f,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,urBAAurB,aAAa,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,8VAA8V,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,sqBAAsqB,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,iOAAiO,aAAa,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,sZAAsZ,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,uwBAAuwB,yBAAyB,CAAC,IAAI,IAAI,GAAG,OAAO,wuBAAwuB9lB,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,wMAAwM+lB,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,iMAAiM,qBAAqB,CAAC,IAAI,IAAI,CAAC,SAAS,OAAO,uwBAAuwB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,+4BAA+4B,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,ioBAAioB5L,MAAM,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,KAAK,MAAM,OAAO,OAAO,oRAAoR,iBAAiB,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,0fAA0f,aAAa,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,gWAAgW,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,+kCAA+kC,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,uVAAuV,uBAAuB,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,0aAA0a,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,YAAY,OAAO,0LAA0L,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,quBAAquB,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,gBAAgB,OAAO,mOAAmO,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,klBAAklB,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,aAAa,OAAO,yWAAyW,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,iwBAAiwB3a,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,uQAAuQ,sBAAsB,CAAC,IAAI,IAAI,CAAC,gBAAgB,gBAAgB,uBAAuB,OAAO,+jBAA+jBwmB,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,+RAA+R,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,ofAAof,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,ylBAAylB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,6eAA6e,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,iPAAiPC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,6qBAA6qB,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,0zBAA0zB,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,koBAAkoB,uCAAuC,CAAC,IAAI,IAAI,GAAG,OAAO,gsDAAgsD,yBAAyB,CAAC,IAAI,IAAI,CAAC,UAAU,OAAO,uZAAuZ7L,OAAO,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,KAAK,KAAK,KAAK,MAAM,MAAM,OAAO,OAAO,qDAAqD,gBAAgB,CAAC,IAAI,IAAI,CAAC,KAAK,iBAAiB,OAAO,6UAA6U8L,QAAQ,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,MAAM,OAAO,mgCAAmgC,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,m4BAAm4B,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,wgBAAwgBC,UAAU,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,gaAAgaC,WAAW,CAAC,IAAI,IAAI,CAAC,OAAO,mBAAmB,OAAO,i6BAAi6B,iBAAiB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2iBAA2iB,YAAY,CAAC,IAAI,IAAI,CAAC,MAAM,uBAAuB,OAAO,sSAAsS,uBAAuB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,+KAA+K,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,aAAa,OAAO,gzBAAgzB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,qhCAAqhCC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,oOAAoO,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,+VAA+V,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,6rBAA6rBC,OAAO,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,meAAmeC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,mcAAmcC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,+0CAA+0C,aAAa,CAAC,IAAI,IAAI,CAAC,KAAK,SAAS,OAAO,2iCAA2iC,aAAa,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,yZAAyZ,kBAAkB,CAAC,IAAI,IAAI,CAAC,MAAM,mBAAmB,OAAO,qbAAqb,kBAAkB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,icAAic,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,q1BAAq1B,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,yRAAyR,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,mgBAAmgB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,4jBAA4jBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,oVAAoV,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,aAAa,OAAO,4RAA4R,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,6YAA6YC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,wwBAAwwB,4BAA4B,CAAC,IAAI,IAAI,GAAG,OAAO,8oBAA8oB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,idAAidC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,ylBAAylB,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,+kBAA+kBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,soBAAsoB,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,kjBAAkjB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,k+BAAk+B,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,8UAA8UC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,wZAAwZ,aAAa,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,oRAAoR,eAAe,CAAC,IAAI,IAAI,CAAC,KAAK,gBAAgB,OAAO,wUAAwU,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,aAAa,OAAO,0IAA0I,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,WAAW,OAAO,mRAAmR,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,ilBAAilB,gBAAgB,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,wQAAwQ,cAAc,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,8nBAA8nB,eAAe,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,0jBAA0jB,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,eAAe,OAAO,wkBAAwkBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,iQAAiQ,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,uJAAuJ,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,8cAA8c,kBAAkB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,ijBAAijBC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,6OAA6O,aAAa,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,8ZAA8Z1M,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,gkBAAgkB,eAAe,CAAC,IAAI,IAAI,CAAC,OAAO,WAAW,OAAO,8nBAA8nB,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,0eAA0e,4BAA4B,CAAC,IAAI,IAAI,GAAG,OAAO,q+BAAq+B,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,6bAA6b,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,gUAAgU2M,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,wQAAwQ,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,2mBAA2mB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,iaAAiaC,OAAO,CAAC,IAAI,IAAI,CAAC,OAAO,WAAW,kBAAkB,OAAO,2cAA2cC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,goBAAgoB,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,2oBAA2oB,gBAAgB,CAAC,IAAI,IAAI,CAAC,MAAM,WAAW,OAAO,+fAA+f,kBAAkB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,kMAAkM,aAAa,CAAC,IAAI,IAAI,CAAC,KAAK,SAAS,OAAO,0VAA0VC,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,0ZAA0ZC,KAAK,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,gVAAgV,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,QAAQ,OAAO,4SAA4SC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,2xCAA2xC,mBAAmB,CAAC,IAAI,IAAI,CAAC,sBAAsB,OAAO,wwBAAwwB,UAAU,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,8JAA8J,eAAe,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,cAAc,OAAO,meAAme,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,qhBAAqhBC,QAAQ,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,gKAAgK,mBAAmB,CAAC,IAAI,IAAI,CAAC,OAAO,eAAe,OAAO,kjBAAkjB,sBAAsB,CAAC,IAAI,IAAI,CAAC,oBAAoB,OAAO,ipCAAipCC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,68CAA68C,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,0lBAA0lB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,urBAAurB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,obAAob,gBAAgB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,g5BAAg5B,qBAAqB,CAAC,IAAI,IAAI,CAAC,KAAK,qBAAqB,uBAAuB,QAAQ,OAAO,wUAAwUC,UAAU,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,w1CAAw1C,sBAAsB,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,qaAAqa,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,8vBAA8vB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,wZAAwZ,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,kwBAAkwB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,yyDAAyyD,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,myDAAmyDC,MAAM,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,OAAO,yRAAyR,iBAAiB,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,icAAic,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,mcAAmc,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,i4BAAi4B,YAAY,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,y0CAAy0C,eAAe,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,8pBAA8pB,cAAc,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,oMAAoM,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,mzBAAmzB,WAAW,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,6WAA6W,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,qjDAAqjD,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,w5BAAw5B,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,mxBAAmxB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,giBAAgiB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,6vBAA6vB,YAAY,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,oYAAoY,iBAAiB,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,2lBAA2lBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,kQAAkQ,eAAe,CAAC,IAAI,IAAI,CAAC,mBAAmB,sBAAsB,OAAO,8LAA8L,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,4ZAA4ZC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,qrBAAqrB,cAAc,CAAC,IAAI,IAAI,CAAC,KAAK,mBAAmB,OAAO,8WAA8WC,MAAM,CAAC,IAAI,IAAI,CAAC,OAAO,UAAU,UAAU,OAAO,obAAobC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,iwBAAiwBC,OAAO,CAAC,IAAI,IAAI,CAAC,MAAM,KAAK,OAAO,iLAAiL,aAAa,CAAC,IAAI,IAAI,CAAC,KAAK,SAAS,OAAO,+OAA+OC,WAAW,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,uTAAuT,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,+tBAA+tB,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,w8BAAw8BC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,4tCAA4tC,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,gOAAgO,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,opBAAopBC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,saAAsa,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,sXAAsX,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,8eAA8eC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,6yBAA6yBC,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,0KAA0K,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,uhCAAuhC,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,wXAAwX,uBAAuB,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,gcAAgc,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,mSAAmS,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,wrBAAwrB,aAAa,CAAC,IAAI,IAAI,CAAC,SAAS,OAAO,kZAAkZ/nB,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,qGAAqG,mBAAmB,CAAC,IAAI,IAAI,CAAC,UAAU,OAAO,sdAAsd,cAAc,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2zBAA2zBgoB,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,izCAAizC,eAAe,CAAC,IAAI,IAAI,CAAC,UAAU,OAAO,waAAwa,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,qlCAAqlC,gBAAgB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,6gBAA6gB,oBAAoB,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,wMAAwM,cAAc,CAAC,IAAI,IAAI,CAAC,UAAU,OAAO,0qBAA0qB,eAAe,CAAC,IAAI,IAAI,CAAC,wBAAwB,OAAO,ukBAAukB,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,iaAAia,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,4VAA4VC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,gUAAgU,YAAY,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,2hBAA2hBC,UAAU,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,kpBAAkpB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,ymBAAymBC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2MAA2MjO,QAAQ,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,uWAAuW,eAAe,CAAC,IAAI,IAAI,CAAC,OAAO,gBAAgB,QAAQ,OAAO,wzDAAwzDC,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,KAAK,OAAO,OAAO,uQAAuQ,YAAY,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,mVAAmViO,UAAU,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,+bAA+b,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,43BAA43BC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,shBAAshBC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,yzBAAyzB,eAAe,CAAC,IAAI,IAAI,CAAC,uBAAuB,OAAO,2pBAA2pB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,oQAAoQ,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,k3BAAk3B,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,6aAA6a,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,klBAAklBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,yeAAye,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,4yBAA4yBnO,UAAU,CAAC,IAAI,IAAI,CAAC,KAAK,MAAM,mBAAmB,OAAO,0gBAA0gBoO,SAAS,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,6RAA6R,cAAc,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,qkBAAqkB,cAAc,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,gQAAgQ,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,+mCAA+mCC,OAAO,CAAC,IAAI,IAAI,CAAC,MAAM,cAAc,OAAO,wPAAwP,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,kuBAAkuBC,OAAO,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,iZAAiZ,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,ivBAAivBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,0bAA0b,iBAAiB,CAAC,IAAI,IAAI,CAAC,mBAAmB,qBAAqB,OAAO,6pBAA6pB,uBAAuB,CAAC,IAAI,IAAI,CAAC,sBAAsB,OAAO,2fAA2fC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,urDAAurD,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,iXAAiXC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,g2BAAg2B,8BAA8B,CAAC,IAAI,IAAI,CAAC,2BAA2B,OAAO,sXAAsXC,SAAS,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,wZAAwZC,MAAM,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,OAAO,kYAAkYC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,orBAAorB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,qfAAqfC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,yUAAyU,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,8gCAA8gC,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,4UAA4U,iBAAiB,CAAC,IAAI,IAAI,CAAC,iBAAiB,mBAAmB,OAAO,8qBAA8qB,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,ykCAAykCC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,6gBAA6gB,oBAAoB,CAAC,IAAI,IAAI,CAAC,KAAK,oBAAoB,oBAAoB,wBAAwB,QAAQ,OAAO,oUAAoU,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,uTAAuT,yBAAyB,CAAC,IAAI,IAAI,CAAC,OAAO,qBAAqB,OAAO,mzCAAmzCC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,6bAA6b,iBAAiB,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,kfAAkf,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,qXAAqX,6BAA6B,CAAC,IAAI,IAAI,GAAG,OAAO,qjBAAqjB,uBAAuB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,m1BAAm1B,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,0ZAA0Z,gBAAgB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,ihBAAihB,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,yBAAyB,OAAO,gTAAgT,eAAe,CAAC,IAAI,IAAI,CAAC,UAAU,OAAO,8hBAA8hB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,6ZAA6Z,oBAAoB,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,OAAO,yaAAya,cAAc,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,8QAA8Q,eAAe,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,gcAAgc,YAAY,CAAC,IAAI,IAAI,CAAC,uBAAuB,OAAO,oOAAoOC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,myBAAmyB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,2zBAA2zBC,MAAM,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,MAAM,YAAY,OAAO,wGAAwGC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,y5BAAy5B,kBAAkB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,8NAA8N,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,+jBAA+jB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,w5BAAw5B,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,qQAAqQ,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,iLAAiL,yBAAyB,CAAC,IAAI,IAAI,CAAC,sCAAsC,mBAAmB,6CAA6C,OAAO,ugCAAugCC,KAAK,CAAC,IAAI,IAAI,CAAC,KAAK,OAAO,OAAO,+gCAA+gC,gBAAgB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,mfAAmfC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,gzBAAgzBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,46BAA46B,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,mpBAAmpB,wBAAwB,CAAC,IAAI,IAAI,GAAG,OAAO,qwBAAqwB,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,iBAAiB,OAAO,+XAA+XC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,iUAAiU,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,qkBAAqkB,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,iaAAia,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,grBAAgrB,gBAAgB,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,4QAA4Q1mB,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,gLAAgL,uBAAuB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,ohBAAohBqX,MAAM,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,sLAAsL,gBAAgB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,2NAA2NsP,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,wPAAwPC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,mdAAmd,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,+cAA+cjsB,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,o6BAAo6BksB,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,mgBAAmgBvP,SAAS,CAAC,IAAI,IAAI,CAAC,KAAK,OAAO,s3CAAs3C,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,iKAAiK,wBAAwB,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,0dAA0d,6BAA6B,CAAC,IAAI,IAAI,CAAC,gBAAgB,gBAAgB,8BAA8B,OAAO,gkBAAgkB,gBAAgB,CAAC,IAAI,IAAI,CAAC,sBAAsB,OAAO,2NAA2N,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,gZAAgZ,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,6rBAA6rB,eAAe,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,6RAA6R,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,k6BAAk6BwP,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,61CAA61CC,OAAO,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,gkEAAgkEC,QAAQ,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,2tBAA2tB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,6bAA6b,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,gfAAgf,UAAU,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,mQAAmQC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,0oBAA0oB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,iZAAiZC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,qYAAqY,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,iaAAia,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,ulCAAulC,oBAAoB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,sIAAsIC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2gBAA2gB,YAAY,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,gZAAgZ,aAAa,CAAC,IAAI,IAAI,CAAC,wBAAwB,OAAO,yOAAyO,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,ohCAAohC,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,0RAA0RC,IAAI,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,ykCAAykC,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,6dAA6d,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,mlBAAmlB,mCAAmC,CAAC,IAAI,IAAI,GAAG,OAAO,qhBAAqhB,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,ymBAAymB,gBAAgB,CAAC,IAAI,IAAI,CAAC,UAAU,OAAO,0qBAA0qB,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,grBAAgrB7P,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,qgBAAqgB8P,WAAW,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,wlBAAwlB,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,giCAAgiCxqB,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,uNAAuN,YAAY,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,QAAQ,OAAO,snBAAsnB,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,OAAO,usBAAusB,wBAAwB,CAAC,IAAI,IAAI,GAAG,OAAO,olCAAolCyqB,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,4+BAA4+B,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,shBAAshBC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,0ZAA0Z,YAAY,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,OAAO,oTAAoT,cAAc,CAAC,IAAI,IAAI,CAAC,KAAK,aAAa,OAAO,4ZAA4Z,cAAc,CAAC,IAAI,IAAI,CAAC,oBAAoB,cAAc,OAAO,ycAAyc,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,wZAAwZC,QAAQ,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,qTAAqTC,OAAO,CAAC,IAAI,IAAI,CAAC,OAAO,iBAAiB,gBAAgB,OAAO,6LAA6L,WAAW,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,gOAAgO,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,2XAA2X,aAAa,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,2rBAA2rB,UAAU,CAAC,IAAI,IAAI,CAAC,MAAM,mBAAmB,OAAO,yiBAAyiBC,QAAQ,CAAC,IAAI,IAAI,CAAC,UAAU,OAAO,qeAAqe,2BAA2B,CAAC,IAAI,IAAI,GAAG,OAAO,sgBAAsgBC,MAAM,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,MAAM,OAAO,WAAW,eAAe,OAAO,ycAAyc,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,oUAAoU,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,0ZAA0ZC,EAAE,CAAC,IAAI,IAAI,CAAC,IAAI,KAAK,mRAAmR,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,oVAAoV,WAAW,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,uJAAuJ,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,6cAA6c,yBAAyB,CAAC,IAAI,IAAI,CAAC,KAAK,YAAY,OAAO,gbAAgb,eAAe,CAAC,IAAI,IAAI,CAAC,WAAW,kBAAkB,OAAO,0WAA0WC,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,WAAW,OAAO,gkBAAgkB,sBAAsB,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,4jBAA4jB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,ohBAAohB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,uqCAAuqCC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,6WAA6W,kBAAkB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,65BAA65B,eAAe,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2fAA2f,UAAU,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,6NAA6NC,KAAK,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,+GAA+G,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,4cAA4cC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,qZAAqZC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,kUAAkUC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,+IAA+I,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,6PAA6P,eAAe,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,QAAQ,OAAO,uSAAuS,kBAAkB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,iUAAiUC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,ibAAib,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,4fAA4fC,KAAK,CAAC,IAAI,IAAI,CAAC,KAAK,OAAO,OAAO,qPAAqP,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,oTAAoT,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,ugBAAugBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,sXAAsXC,KAAK,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,6VAA6VC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,w0BAAw0B,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,8oBAA8oB,WAAW,CAAC,IAAI,IAAI,CAAC,IAAI,MAAM,MAAM,MAAM,OAAO,OAAO,gXAAgX,aAAa,CAAC,IAAI,IAAI,CAAC,KAAK,SAAS,MAAM,SAAS,OAAO,+TAA+T/Q,IAAI,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,umBAAumBgR,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,keAAke,kBAAkB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,8XAA8X,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,wkBAAwkB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,uVAAuVC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,8MAA8M,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,yBAAyB,OAAO,+SAA+S,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,kmBAAkmB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,sVAAsV,wBAAwB,CAAC,IAAI,IAAI,CAAC,mBAAmB,wBAAwB,OAAO,8jBAA8jB,YAAY,CAAC,IAAI,IAAI,CAAC,OAAO,KAAK,wMAAwM,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,4KAA4K,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,gjBAAgjB,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,mqBAAmqB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,6LAA6L,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,8jBAA8jB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,2ZAA2Z,mBAAmB,CAAC,IAAI,IAAI,CAAC,oBAAoB,OAAO,+NAA+N,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,mxBAAmxBC,SAAS,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,uIAAuI,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,kXAAkX,cAAc,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,oWAAoW,wBAAwB,CAAC,IAAI,IAAI,GAAG,OAAO,oiBAAoiBC,QAAQ,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,miBAAmiB,uBAAuB,CAAC,IAAI,IAAI,GAAG,OAAO,uxBAAuxBC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,m2BAAm2B,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,mBAAmB,OAAO,4SAA4SC,IAAI,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,6cAA6c,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,grBAAgrB,mBAAmB,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,4UAA4U,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,6SAA6S,4BAA4B,CAAC,IAAI,IAAI,GAAG,OAAO,4gBAA4gB,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,2xCAA2xC,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,kgBAAkgB,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,ufAAuf,uBAAuB,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,4wBAA4wBC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,+pBAA+pBC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,wPAAwP,WAAW,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,ySAAySC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,qvBAAqvB,mBAAmB,CAAC,IAAI,IAAI,CAAC,gBAAgB,gBAAgB,oBAAoB,OAAO,gkBAAgkBvR,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,uYAAuYwR,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,ojBAAojB,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,gqBAAgqB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,4cAA4c,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,udAAud,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,wVAAwV,kBAAkB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,2nBAA2nB,eAAe,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,4KAA4KvR,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,OAAO,yIAAyI,eAAe,CAAC,IAAI,IAAI,CAAC,OAAO,KAAK,6MAA6M,kBAAkB,CAAC,IAAI,IAAI,CAAC,OAAO,WAAW,OAAO,84BAA84B,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,+NAA+NwR,QAAQ,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,OAAO,+RAA+RC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,yQAAyQ,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,gBAAgB,kBAAkB,OAAO,6zBAA6zB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,kvBAAkvBC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,mnBAAmnB,gBAAgB,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,kQAAkQC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,4hBAA4hBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,0WAA0WC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,k0CAAk0C,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,iBAAiB,uBAAuB,OAAO,+YAA+Y,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,slBAAslB5R,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,eAAe,iBAAiB,OAAO,0mBAA0mB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,+fAA+f,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,2hBAA2hB,iBAAiB,CAAC,IAAI,IAAI,CAAC,UAAU,UAAU,OAAO,4RAA4R6R,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,46CAA46CC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,kjCAAkjC,2BAA2B,CAAC,IAAI,IAAI,CAAC,SAAS,OAAO,2kBAA2kBC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,+sBAA+sB,YAAY,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,wQAAwQ,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,4mBAA4mBC,IAAI,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,OAAO,8eAA8eC,UAAU,CAAC,IAAI,IAAI,CAAC,KAAK,OAAO,yMAAyM,gBAAgB,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,mVAAmV,YAAY,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,+MAA+M,gBAAgB,CAAC,IAAI,IAAI,CAAC,MAAM,QAAQ,aAAa,OAAO,qiBAAqiBC,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,SAAS,OAAO,8wBAA8wB,aAAa,CAAC,IAAI,IAAI,CAAC,+BAA+B,OAAO,klCAAklC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,02BAA02BC,KAAK,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,sJAAsJC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,4SAA4S,uBAAuB,CAAC,IAAI,IAAI,GAAG,OAAO,yhBAAyhB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,mbAAmb,mBAAmB,CAAC,IAAI,IAAI,CAAC,OAAO,UAAU,OAAO,4OAA4O,2BAA2B,CAAC,IAAI,IAAI,CAAC,OAAO,wBAAwB,gBAAgB,OAAO,okBAAokB,wBAAwB,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,4tBAA4tB,qBAAqB,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,kbAAkb,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,2XAA2X,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,ygBAAygB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,8lBAA8lB,WAAW,CAAC,IAAI,IAAI,CAAC,OAAO,aAAa,OAAO,8bAA8bC,QAAQ,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,yoBAAyoB,aAAa,CAAC,IAAI,IAAI,CAAC,aAAa,iBAAiB,OAAO,4WAA4W,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,wcAAwc,4BAA4B,CAAC,IAAI,IAAI,GAAG,OAAO,ghBAAghB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,4KAA4K,eAAe,CAAC,IAAI,IAAI,CAAC,OAAO,UAAU,aAAa,OAAO,6RAA6R,mBAAmB,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,OAAO,2hBAA2hB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,gsBAAgsB,UAAU,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,yYAAyY,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,8yBAA8yB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,gmCAAgmC,YAAY,CAAC,IAAI,IAAI,CAAC,uBAAuB,OAAO,uOAAuO,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,sjBAAsjB,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,iZAAiZ,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,szBAAszB,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,+0BAA+0B,gBAAgB,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,saAAsa,WAAW,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,0NAA0NC,GAAG,CAAC,IAAI,IAAI,CAAC,MAAM,aAAa,UAAU,OAAO,yNAAyNC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,8nBAA8nB,aAAa,CAAC,IAAI,IAAI,CAAC,SAAS,OAAO,8qBAA8qB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,mVAAmV,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,OAAO,wNAAwN,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,qcAAqcC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,mfAAmf,YAAY,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,89BAA89BC,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,wLAAwL,sBAAsB,CAAC,IAAI,IAAI,CAAC,OAAO,gBAAgB,OAAO,6yBAA6yB,aAAa,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,gmBAAgmBC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,kbAAkb,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,giBAAgiB,qBAAqB,CAAC,IAAI,IAAI,CAAC,KAAK,UAAU,OAAO,0GAA0GC,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,wRAAwR,mBAAmB,CAAC,IAAI,IAAI,CAAC,KAAK,iBAAiB,OAAO,4jBAA4jBC,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,KAAK,iBAAiB,OAAO,olCAAolC,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,+eAA+e,kBAAkB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,yUAAyU3S,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,kQAAkQ,eAAe,CAAC,IAAI,IAAI,CAAC,cAAc,gBAAgB,OAAO,4ZAA4Z,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,swCAAswC4S,KAAK,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,opBAAopB,oBAAoB,CAAC,IAAI,IAAI,CAAC,MAAM,gBAAgB,OAAO,umBAAumB,kBAAkB,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,utBAAutBC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,weAAwe,eAAe,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,gXAAgXC,KAAK,CAAC,IAAI,IAAI,CAAC,MAAM,MAAM,OAAO,KAAK,yLAAyLC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,ycAAycC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,6iBAA6iBC,MAAM,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,MAAM,MAAM,IAAI,QAAQ,WAAW,SAAS,SAAS,OAAO,2RAA2R,4BAA4B,CAAC,IAAI,IAAI,CAAC,UAAU,OAAO,qoBAAqoB,kBAAkB,CAAC,IAAI,IAAI,CAAC,sBAAsB,OAAO,2YAA2Y,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,+eAA+e,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,+9BAA+9BC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,qjBAAqjB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,2qBAA2qB,aAAa,CAAC,IAAI,IAAI,CAAC,KAAK,kBAAkB,OAAO,8WAA8WC,QAAQ,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,qlCAAqlC,iBAAiB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,ucAAuc,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,4vBAA4vB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,iWAAiW,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,+cAA+clT,UAAU,CAAC,IAAI,IAAI,CAAC,KAAK,OAAO,8QAA8QmT,OAAO,CAAC,IAAI,IAAI,CAAC,OAAO,KAAK,iLAAiLC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,scAAscC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,63BAA63B,cAAc,CAAC,IAAI,IAAI,CAAC,KAAK,MAAM,SAAS,SAAS,eAAe,OAAO,kVAAkVrsB,IAAI,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,qOAAqOssB,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,2fAA2f,aAAa,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,84BAA84B,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,2QAA2QC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,klBAAklB,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,8RAA8R,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,qfAAqf,eAAe,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,6SAA6S,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,6rBAA6rB,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,yhBAAyhBC,OAAO,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,4LAA4L,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,qrBAAqrBC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,ybAAyb,WAAW,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,4KAA4K,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,ieAAie,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,69BAA69B,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,yKAAyKC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,ivFAAivF,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,wSAAwSC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,yPAAyPC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,urBAAurB,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,0aAA0aC,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,oGAAoG,gBAAgB,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,4KAA4KC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,gfAAgf,WAAW,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,yrBAAyrB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,meAAme,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,+YAA+Y,kBAAkB,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,4RAA4RC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,wgBAAwgB,gBAAgB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,2ZAA2Z,gBAAgB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,oOAAoOC,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,4aAA4a,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,kUAAkU,qBAAqB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,wZAAwZC,QAAQ,CAAC,IAAI,IAAI,CAAC,MAAM,MAAM,cAAc,KAAK,uNAAuN,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,wiBAAwiB,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,oZAAoZC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,2OAA2O,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,SAAS,OAAO,qVAAqVC,UAAU,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,cAAc,OAAO,gVAAgVC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,uqBAAuqB,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,OAAO,iZAAiZC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,6iBAA6iB,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,6qBAA6qBC,OAAO,CAAC,IAAI,IAAI,CAAC,OAAO,YAAY,OAAO,4pBAA4pBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,4UAA4UC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,0mBAA0mBC,MAAM,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,uUAAuUC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,y7DAAy7DC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,qYAAqY,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,omBAAomBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,2jBAA2jB,mBAAmB,CAAC,IAAI,IAAI,CAAC,OAAO,eAAe,OAAO,skBAAskB,YAAY,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,0RAA0R,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,iZAAiZC,YAAY,CAAC,IAAI,IAAI,CAAC,KAAK,mBAAmB,OAAO,mvBAAmvBC,QAAQ,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,0eAA0e,kBAAkB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,sQAAsQ,uBAAuB,CAAC,IAAI,IAAI,GAAG,OAAO,k5BAAk5BC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,w3CAAw3C,gBAAgB,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,wTAAwT9U,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,muBAAmuB,cAAc,CAAC,IAAI,IAAI,CAAC,IAAI,qBAAqB,OAAO,iVAAiV+U,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,urBAAurB,oBAAoB,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,odAAod,uBAAuB,CAAC,IAAI,IAAI,CAAC,OAAO,mBAAmB,OAAO,u2CAAu2C,cAAc,CAAC,IAAI,IAAI,CAAC,yBAAyB,OAAO,oVAAoVC,OAAO,CAAC,IAAI,IAAI,CAAC,OAAO,gBAAgB,OAAO,oNAAoN,sBAAsB,CAAC,IAAI,IAAI,CAAC,sBAAsB,OAAO,2jBAA2jB,gBAAgB,CAAC,IAAI,IAAI,GAAG,OAAO,yZAAyZ,gBAAgB,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,0vBAA0vB,sBAAsB,CAAC,IAAI,IAAI,CAAC,uBAAuB,OAAO,qMAAqM,uBAAuB,CAAC,IAAI,IAAI,GAAG,OAAO,yXAAyX,UAAU,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,saAAsa,uBAAuB,CAAC,IAAI,IAAI,GAAG,OAAO,kmBAAkmBC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,qiBAAqiB,aAAa,CAAC,IAAI,IAAI,CAAC,SAAS,OAAO,6NAA6NC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,6iBAA6iBC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,ohBAAohBlV,KAAK,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,OAAO,kRAAkR,mBAAmB,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,sUAAsUmV,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,gLAAgL,WAAW,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,6OAA6OC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,yQAAyQ,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,2OAA2O,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,qpBAAqpB,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,UAAU,OAAO,4eAA4eC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2eAA2e,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,ocAAocC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,ooBAAooB,gBAAgB,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,4/BAA4/B,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,gBAAgB,OAAO,kVAAkVC,QAAQ,CAAC,IAAI,IAAI,CAAC,OAAO,KAAK,4iBAA4iB,qCAAqC,CAAC,IAAI,IAAI,CAAC,cAAc,OAAO,+ZAA+Z,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,0VAA0V7xB,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,0IAA0I8xB,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,4iCAA4iC,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,ufAAufC,SAAS,CAAC,IAAI,IAAI,CAAC,KAAK,MAAM,OAAO,2jBAA2jB,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,obAAob,4BAA4B,CAAC,IAAI,IAAI,GAAG,OAAO,+jBAA+jBC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,wRAAwRC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,ygBAAygB,yBAAyB,CAAC,IAAI,IAAI,GAAG,OAAO,g6BAAg6B,UAAU,CAAC,IAAI,IAAI,CAAC,KAAK,MAAM,gBAAgB,OAAO,+TAA+T,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,kgCAAkgC1V,SAAS,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,OAAO,+NAA+N2V,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,ssBAAssBC,MAAM,CAAC,IAAI,IAAI,CAAC,UAAU,OAAO,2pCAA2pC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,8YAA8YC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,k8BAAk8B,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,gUAAgU,cAAc,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,yOAAyO,wBAAwB,CAAC,IAAI,IAAI,CAAC,OAAO,oBAAoB,OAAO,wwBAAwwB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,wVAAwV,0BAA0B,CAAC,IAAI,IAAI,GAAG,OAAO,2hBAA2hB,aAAa,CAAC,IAAI,IAAI,CAAC,eAAe,cAAc,UAAU,OAAO,mwBAAmwB5V,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,6NAA6N,iCAAiC,CAAC,IAAI,IAAI,GAAG,OAAO,i+BAAi+B,eAAe,CAAC,IAAI,IAAI,CAAC,qBAAqB,OAAO,srBAAsrB,oBAAoB,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,6ZAA6Z6V,QAAQ,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,sfAAsf,mBAAmB,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,2ZAA2Z,aAAa,CAAC,IAAI,IAAI,CAAC,SAAS,OAAO,osBAAosBC,OAAO,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,0lBAA0lB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,wMAAwM,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,SAAS,OAAO,6qBAA6qB,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,kkBAAkkB,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,6RAA6RC,IAAI,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,yTAAyT,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,onBAAonB,iBAAiB,CAAC,IAAI,IAAI,CAAC,KAAK,iBAAiB,OAAO,0tBAA0tB,oBAAoB,CAAC,IAAI,IAAI,CAAC,MAAM,aAAa,mBAAmB,OAAO,gOAAgOC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,8nBAA8nBC,QAAQ,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,eAAe,OAAO,2OAA2OnxB,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,oSAAoS,aAAa,CAAC,IAAI,IAAI,CAAC,WAAW,OAAO,0OAA0O,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,otCAAotCoxB,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,+WAA+W,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,yWAAyWC,MAAM,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,OAAO,wLAAwL,yBAAyB,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,6RAA6R,qBAAqB,CAAC,IAAI,IAAI,GAAG,OAAO,sgBAAsgB,aAAa,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,4KAA4K,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,kaAAka,oBAAoB,CAAC,IAAI,IAAI,GAAG,OAAO,4jBAA4jB,2BAA2B,CAAC,IAAI,IAAI,GAAG,OAAO,wdAAwd,YAAY,CAAC,IAAI,IAAI,CAAC,KAAK,eAAe,OAAO,kqBAAkqB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,slBAAslBC,UAAU,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,sTAAsT,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,8cAA8c,iBAAiB,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,8RAA8R,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,2bAA2bC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,4hBAA4hB,eAAe,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,64CAA64C,wBAAwB,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,8lBAA8lB,kBAAkB,CAAC,IAAI,IAAI,CAAC,MAAM,kBAAkB,kBAAkB,gBAAgB,OAAO,mVAAmV,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,8aAA8a,YAAY,CAAC,IAAI,IAAI,CAAC,OAAO,QAAQ,OAAO,wcAAwc,sBAAsB,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,q8CAAq8CC,UAAU,CAAC,IAAI,IAAI,GAAG,OAAO,qyBAAqyB,uBAAuB,CAAC,IAAI,IAAI,CAAC,KAAK,uBAAuB,WAAW,OAAO,mTAAmTC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,saAAsaC,MAAM,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,qWAAqW,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,wbAAwb,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,4xBAA4xB,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,qkBAAqkB,aAAa,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,qaAAqa,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,8UAA8U,0BAA0B,CAAC,IAAI,IAAI,GAAG,OAAO,8kBAA8kBC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,+lDAA+lD,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,0XAA0X,oBAAoB,CAAC,IAAI,IAAI,CAAC,uBAAuB,OAAO,ylBAAylB,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,o4BAAo4BC,MAAM,CAAC,IAAI,IAAI,CAAC,gBAAgB,WAAW,OAAO,khCAAkhC,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,mUAAmU,eAAe,CAAC,IAAI,IAAI,CAAC,eAAe,iBAAiB,OAAO,0iCAA0iC,iBAAiB,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,yUAAyU,eAAe,CAAC,IAAI,IAAI,CAAC,OAAO,WAAW,OAAO,4ZAA4Z,kBAAkB,CAAC,IAAI,IAAI,CAAC,kBAAkB,OAAO,8fAA8f,YAAY,CAAC,IAAI,IAAI,CAAC,MAAM,kBAAkB,OAAO,oRAAoR,6BAA6B,CAAC,IAAI,IAAI,GAAG,OAAO,g5BAAg5B,kBAAkB,CAAC,IAAI,IAAI,CAAC,mBAAmB,OAAO,yWAAyWC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,+iBAA+iB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,qLAAqLC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,oqBAAoqBC,QAAQ,CAAC,IAAI,IAAI,CAAC,OAAO,YAAY,OAAO,wUAAwU,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,2TAA2T,eAAe,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,gBAAgB,OAAO,4RAA4RC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,q9BAAq9BC,MAAM,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,65BAA65B,aAAa,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,OAAO,qlCAAqlC,eAAe,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,4bAA4b,wBAAwB,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,+WAA+W,YAAY,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,gkBAAgkB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,6WAA6W,YAAY,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,uqBAAuqB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,ofAAof,kBAAkB,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,2FAA2F,aAAa,CAAC,IAAI,IAAI,CAAC,UAAU,OAAO,8QAA8QC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,6ZAA6ZC,KAAK,CAAC,IAAI,IAAI,GAAG,OAAO,4SAA4S,yBAAyB,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,oRAAoR,iBAAiB,CAAC,IAAI,IAAI,GAAG,OAAO,sRAAsR,aAAa,CAAC,IAAI,IAAI,CAAC,YAAY,OAAO,oLAAoL,aAAa,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,krBAAkrB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,woBAAwoBC,QAAQ,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,ukBAAukB,YAAY,CAAC,IAAI,IAAI,CAAC,MAAM,OAAO,qzBAAqzB,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,+zBAA+zB,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,OAAO,MAAM,OAAO,OAAO,OAAO,MAAM,OAAO,OAAO,qBAAqBC,OAAO,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,qOAAqOC,IAAI,CAAC,IAAI,IAAI,GAAG,OAAO,qTAAqT,cAAc,CAAC,IAAI,IAAI,CAAC,MAAM,eAAe,OAAO,sLAAsL,YAAY,CAAC,IAAI,IAAI,GAAG,OAAO,uxBAAuxB,2BAA2B,CAAC,IAAI,IAAI,GAAG,OAAO,wrCAAwrCC,KAAK,CAAC,IAAI,IAAI,CAAC,QAAQ,OAAO,ifAAif,eAAe,CAAC,IAAI,IAAI,GAAG,OAAO,klBAAklB,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,YAAY,OAAO,sZAAsZC,MAAM,CAAC,IAAI,IAAI,GAAG,OAAO,0aAA0a,mBAAmB,CAAC,IAAI,IAAI,GAAG,OAAO,yuBAAyuBC,YAAY,CAAC,GAAG,IAAI,CAAC,MAAM,MAAM,OAAO,KAAK,gIAAgI,cAAc,CAAC,IAAI,IAAI,GAAG,OAAO,u+BAAu+BC,MAAM,CAAC,IAAI,IAAI,CAAC,OAAO,OAAO,MAAM,OAAO,gXAAgX,oBAAoB,CAAC,IAAI,IAAI,CAAC,MAAM,gBAAgB,OAAO,idAAid,cAAc,CAAC,IAAI,IAAI,CAAC,OAAO,MAAM,SAAS,OAAO,KAAK,i5BAAi5BC,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,8RAA8R,0BAA0B,CAAC,IAAI,IAAI,CAAC,iBAAiB,OAAO,66BAA66B,aAAa,CAAC,IAAI,IAAI,CAAC,aAAa,OAAO,imEAAimE,2BAA2B,CAAC,IAAI,IAAI,GAAG,OAAO,mgBAAmgB,mBAAmB,CAAC,IAAI,IAAI,CAAC,OAAO,cAAc,SAAS,cAAc,OAAO,wcAAwcC,SAAS,CAAC,IAAI,IAAI,GAAG,OAAO,+oBAA+oBC,OAAO,CAAC,IAAI,IAAI,GAAG,OAAO,6UAA6UxvB,EAAE,CAAC,IAAI,IAAI,CAAC,KAAK,KAAK,sKAAsKyvB,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,qUAAqU,gBAAgB,CAAC,IAAI,IAAI,CAAC,OAAO,iBAAiB,OAAO,q5CAAq5C,WAAW,CAAC,IAAI,IAAI,GAAG,OAAO,qUAAqU,kBAAkB,CAAC,IAAI,IAAI,GAAG,OAAO,orBAAorB,mBAAmB,CAAC,IAAI,IAAI,CAAC,OAAO,iBAAiB,OAAO,4qBAA4qB,cAAc,CAAC,IAAI,IAAI,CAAC,eAAe,OAAO,6jBAA6jB,qBAAqB,CAAC,IAAI,IAAI,CAAC,oBAAoB,OAAO,+bAA+bC,QAAQ,CAAC,IAAI,IAAI,GAAG,OAAO,2kBAA2kB,aAAa,CAAC,IAAI,IAAI,GAAG,OAAO,sRAAsRC,MAAM,CAAC,IAAI,IAAI,CAAC,gBAAgB,OAAO,6QAA6Q,iBAAiB,CAAC,IAAI,IAAI,CAAC,OAAO,gBAAgB,OAAO,wvBAAwvB,uBAAuB,CAAC,IAAI,IAAI,GAAG,OAAO,qiBAAqiB,sBAAsB,CAAC,IAAI,IAAI,GAAG,OAAO,kjBAAkjB,UAAU,CAAC,IAAI,IAAI,CAAC,MAAM,gBAAgB,OAAO,uRAAuR,SAASz1B,GAAG,IAAI,IAAI,IAAIC,EAAEmB,UAAUC,OAAOhB,EAAE,IAAI0B,MAAM,EAAE9B,EAAEA,EAAE,EAAE,GAAGG,EAAE,EAAEA,EAAEH,EAAEG,IAAIC,EAAED,EAAE,GAAGgB,UAAUhB,GAAGJ,EAAEkB,WAAM,EAAOb,EAAE,CAAC,MAAML,GAAG,IAAIsC,EAAE,MAAMtC,CAAC,CAAC,CAA3I,EAA6I,WAAW8F,EAAE,MAAMG,GAAGH,EAAE,WAAWG,EAAE,GAAE,CAApwnyB,GAAwwnyB,WAAW,aAAa,SAAS7F,EAAEH,EAAED,GAAG,IAAIK,EAAED,EAAEO,OAAOC,KAAKX,GAAG,OAAOU,OAAOE,wBAAwBR,EAAEM,OAAOE,sBAAsBZ,GAAGD,IAAIK,EAAEA,EAAES,QAAO,SAASd,GAAG,OAAOW,OAAOI,yBAAyBd,EAAED,GAAGgB,UAAU,KAAIZ,EAAEa,KAAKC,MAAMd,EAAEC,IAAID,CAAC,CAAC,SAASsF,EAAEzF,GAAG,IAAI,IAAID,EAAE,EAAEA,EAAEoB,UAAUC,OAAOrB,IAAI,CAAC,IAAIK,EAAE,MAAMe,UAAUpB,GAAGoB,UAAUpB,GAAG,CAAC,EAAEA,EAAE,EAAEI,EAAEO,OAAON,IAAG,GAAIiB,SAAQ,SAAStB,GAAGmB,EAAElB,EAAED,EAAEK,EAAEL,GAAG,IAAGW,OAAOa,0BAA0Bb,OAAOc,iBAAiBxB,EAAEU,OAAOa,0BAA0BnB,IAAID,EAAEO,OAAON,IAAIiB,SAAQ,SAAStB,GAAGW,OAAOe,eAAezB,EAAED,EAAEW,OAAOI,yBAAyBV,EAAEL,GAAG,GAAE,CAAC,OAAOC,CAAC,CAAC,SAASO,EAAER,GAAG,OAAOQ,EAAE,mBAAmBmE,QAAQ,iBAAiBA,OAAOC,SAAS,SAAS5E,GAAG,cAAcA,CAAC,EAAE,SAASA,GAAG,OAAOA,GAAG,mBAAmB2E,QAAQ3E,EAAEkF,cAAcP,QAAQ3E,IAAI2E,OAAOG,UAAU,gBAAgB9E,CAAC,GAAGA,EAAE,CAAuK,SAASmB,EAAEnB,EAAEC,EAAEI,GAAG,OAAOJ,KAAKD,EAAEW,OAAOe,eAAe1B,EAAEC,EAAE,CAAC0B,MAAMtB,EAAEW,YAAW,EAAGY,cAAa,EAAGC,UAAS,IAAK7B,EAAEC,GAAGI,EAAEL,CAAC,CAAC,SAASuC,EAAEvC,EAAEC,GAAG,OAAO,SAASD,GAAG,GAAG+B,MAAM2C,QAAQ1E,GAAG,OAAOA,CAAC,CAAxC,CAA0CA,IAAI,SAASA,EAAEC,GAAG,IAAII,EAAE,MAAML,EAAE,KAAK,oBAAoB2E,QAAQ3E,EAAE2E,OAAOC,WAAW5E,EAAE,cAAc,GAAG,MAAMK,EAAE,CAAC,IAAID,EAAEI,EAAEC,EAAE,GAAGC,GAAE,EAAGS,GAAE,EAAG,IAAI,IAAId,EAAEA,EAAE2E,KAAKhF,KAAKU,GAAGN,EAAEC,EAAEq1B,QAAQC,QAAQl1B,EAAEQ,KAAKb,EAAEuB,QAAQ1B,GAAGQ,EAAEY,SAASpB,GAAGS,GAAE,GAAI,CAAC,MAAMV,GAAGmB,GAAE,EAAGX,EAAER,CAAC,CAAC,QAAQ,IAAIU,GAAG,MAAML,EAAEu1B,QAAQv1B,EAAEu1B,QAAQ,CAAC,QAAQ,GAAGz0B,EAAE,MAAMX,CAAC,CAAC,CAAC,OAAOC,CAAC,CAAC,CAArT,CAAuTT,EAAEC,IAAII,EAAEL,EAAEC,IAAI,WAAW,MAAM,IAAIoF,UAAU,4IAA4I,CAA3K,EAA8K,CAAC,SAAShD,EAAErC,GAAG,OAAO,SAASA,GAAG,GAAG+B,MAAM2C,QAAQ1E,GAAG,OAAOU,EAAEV,EAAE,CAA3C,CAA6CA,IAAI,SAASA,GAAG,GAAG,oBAAoB2E,QAAQ,MAAM3E,EAAE2E,OAAOC,WAAW,MAAM5E,EAAE,cAAc,OAAO+B,MAAM8C,KAAK7E,EAAE,CAA/G,CAAiHA,IAAIK,EAAEL,IAAI,WAAW,MAAM,IAAIqF,UAAU,uIAAuI,CAAtK,EAAyK,CAAC,SAAShF,EAAEL,EAAEC,GAAG,GAAGD,EAAE,CAAC,GAAG,iBAAiBA,EAAE,OAAOU,EAAEV,EAAEC,GAAG,IAAII,EAAEM,OAAOmE,UAAUC,SAASC,KAAKhF,GAAGiF,MAAM,GAAG,GAAG,MAAM,SAAS5E,EAAE,WAAWA,GAAGL,EAAEkF,YAAYlF,EAAEkF,YAAYC,KAAK9E,IAAI,QAAQA,EAAE0B,MAAM8C,KAAK7E,GAAG,cAAcK,GAAG,2CAA2C+E,KAAK/E,GAAGK,EAAEV,EAAEC,QAAG,CAAM,CAAC,CAAC,SAASS,EAAEV,EAAEC,IAAI,MAAMA,GAAGA,EAAED,EAAEqB,UAAUpB,EAAED,EAAEqB,QAAQ,IAAI,IAAIhB,EAAE,EAAED,EAAE,IAAI2B,MAAM9B,GAAGI,EAAEJ,EAAEI,IAAID,EAAEC,GAAGL,EAAEK,GAAG,OAAOD,CAAC,CAAC,SAASJ,IAAI,CAAC,IAAIC,EAAE,CAAC,EAAEsB,EAAE,CAAC,EAAEO,EAAE,KAAKQ,EAAE,CAACuzB,KAAK71B,EAAE81B,QAAQ91B,GAAG,IAAI,oBAAoBE,SAASD,EAAEC,QAAQ,oBAAoBC,WAAWoB,EAAEpB,UAAU,oBAAoB41B,mBAAmBj0B,EAAEi0B,kBAAkB,oBAAoBC,cAAc1zB,EAAE0zB,YAAY,CAAC,MAAMh2B,GAAG,CAAC,IAAkCyC,OAAE,KAAhCD,GAAGvC,EAAEK,WAAW,CAAC,GAAGC,WAAuB,GAAGiC,EAAEK,EAAE5C,EAAE2D,EAAErC,EAAEmB,EAAEZ,EAAEU,EAAEF,EAAEqB,IAAId,EAAE1C,SAASsF,IAAI7B,EAAE5B,mBAAmB4B,EAAE3B,MAAM,mBAAmB2B,EAAE1B,kBAAkB,mBAAmB0B,EAAEzB,cAAc2D,GAAGrD,EAAEL,QAAQ,UAAUK,EAAEL,QAAQ,YAAwCkrB,GAA5BrtB,EAAE,qBAA4B,kBAAiBoiB,EAAE,gBAAgBlD,EAAE,yBAAyBJ,EAAE,iCAAiC+Q,EAAE,cAAcmG,EAAE,YAAYC,EAAE,oBAAoBd,EAAE,QAAQhT,EAAE,CAAC,OAAO,OAAO,QAAQ,UAAU+T,EAAE,WAAW,IAAI,OAAM,CAAE,CAAC,MAAMn2B,GAAG,OAAM,CAAE,CAAC,CAA1C,GAA8Co2B,EAAE,UAAU3L,EAAE,QAAQ4L,EAAE,CAACD,EAAE3L,GAAG,SAAS6L,EAAEt2B,GAAG,OAAO,IAAI2C,MAAM3C,EAAE,CAAC4C,IAAI,SAAS5C,EAAEC,GAAG,OAAOA,KAAKD,EAAEA,EAAEC,GAAGD,EAAEo2B,EAAE,GAAG,CAAC,IAAIG,EAAED,GAAGn1B,EAAEI,EAAE,CAAC,EAAE60B,EAAE,CAACtzB,GAAG,QAAQC,IAAI,QAAQ,WAAW,QAAQC,IAAI,UAAU,aAAa,UAAUC,IAAI,QAAQ,WAAW,QAAQC,IAAI,OAAO,UAAU,OAAOC,IAAI,UAAU,aAAa,UAAUC,IAAI,SAAS,YAAY,SAASC,IAAI,MAAMC,KAAK,MAAM,SAAS,MAAM,iBAAiB,QAAQnC,EAAEI,EAAEkpB,EAAE,CAAC3nB,GAAG,QAAQS,KAAK,QAAQ,WAAW,QAAQC,KAAK,UAAU,aAAa,UAAUC,KAAK,QAAQ,WAAW,QAAQC,KAAK,OAAO,UAAU,SAASnC,IAAIi1B,EAAEF,GAAGn1B,EAAEW,EAAE,CAAC,EAAEs0B,EAAE,CAACvyB,MAAM,MAAMC,QAAQ,MAAMC,MAAM,MAAMC,KAAK,MAAMC,QAAQ,MAAMC,OAAO,MAAMC,IAAI,QAAQhD,EAAEW,EAAE2oB,EAAE,CAAC5mB,MAAM,OAAOC,QAAQ,OAAOC,MAAM,OAAOC,KAAK,SAASlC,IAAI20B,EAAEH,GAAGn1B,EAAEmB,EAAE,CAAC,EAAE8zB,EAAE,CAAChzB,IAAI,YAAYD,IAAI,aAAaE,IAAI,SAASJ,IAAI,WAAWD,IAAI,aAAaD,IAAI,WAAWG,IAAI,YAAY/B,EAAEmB,EAAEmoB,EAAE,CAAClnB,KAAK,WAAWC,KAAK,aAAaC,KAAK,WAAWC,KAAK,YAAYpB,IAAIo0B,EAAEJ,GAAGn1B,EAAEsB,EAAE,CAAC,EAAE2zB,EAAE,CAAC,YAAY,MAAM,aAAa,MAAM,SAAS,MAAM,WAAW,MAAM,aAAa,MAAM,WAAW,MAAM,UAAU,QAAQj1B,EAAEsB,EAAEgoB,EAAE,CAAC,WAAW,OAAO,aAAa,OAAO,WAAW,OAAO,UAAU,SAAShoB,IAAIk0B,EAAE,uCAAuCC,EAAE,iBAAiBC,EAAE,0FAA0FC,EAAER,GAAGn1B,EAAEI,EAAE,CAAC,EAAE60B,EAAE,CAAC,IAAI,MAAM,IAAI,MAAMhyB,OAAO,MAAM,IAAI,MAAM,IAAI,QAAQjD,EAAEI,EAAEkpB,EAAE,CAAC,IAAI,OAAO,IAAI,OAAO,IAAI,OAAO,IAAI,SAASlpB,IAAyEw1B,GAA5Cz0B,GAAzBR,EAAE,CAAC,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,EAAE,KAAQuC,OAAO,CAAC,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,GAAG,KAAO,CAAC,QAAQ,cAAc,YAAY,oBAAoB,iBAAgB2yB,EAAE,CAACC,MAAM,gBAAgBC,aAAa,eAAeC,QAAQ,UAAUC,UAAU,aAAa30B,EAAE,IAAI6B,IAAI3D,OAAOC,KAAK41B,EAAEJ,IAAI7xB,IAAI9B,EAAE+B,IAAIC,KAAKhC,IAAI9B,OAAOC,KAAK41B,EAAE/L,IAAIlmB,IAAI9B,EAAE+B,IAAIC,KAAKhC,IAAI,IAAI40B,EAAE,GAAGhzB,OAAOgyB,EAAEh0B,EAAEI,GAAG,CAAC,MAAM,KAAK,KAAK,KAAK,KAAK,MAAM,OAAO,SAAS,OAAO,YAAY,SAAS,YAAY,kBAAkB,gBAAgB,OAAO,KAAK,UAAU,iBAAiB,cAAc,SAAS,KAAK,YAAY,aAAa,QAAQ,aAAa,aAAa,YAAY,YAAY,QAAQ,aAAa,eAAe,OAAO,WAAW,WAAW,QAAQ,KAAKu0B,EAAEC,MAAMD,EAAEE,aAAaF,EAAEG,QAAQH,EAAEI,YAAY/yB,OAAOvC,EAAEyC,KAAI,SAASvE,GAAG,MAAM,GAAGqE,OAAOrE,EAAE,IAAI,KAAIqE,OAAO/B,EAAEiC,KAAI,SAASvE,GAAG,MAAM,KAAKqE,OAAOrE,EAAE,KAAIs3B,EAAEz0B,EAAE00B,mBAAmB,CAAC,EAAE3zB,GAAG,mBAAmBA,EAAE4zB,eAAe,CAAC,CAAC,qBAAqB,gBAAgB,CAAC,kBAAkB,aAAa,CAAC,sBAAsB,iBAAiB,CAAC,qBAAqB,gBAAgB,CAAC,yBAAyB,oBAAoB,CAAC,wBAAwB,kBAAkB,CAAC,oBAAoB,cAAc,CAAC,iBAAiB,YAAY,CAAC,8BAA8B,wBAAwB,CAAC,yBAAyB,oBAAoB,CAAC,uBAAuB,kBAAkB,CAAC,4BAA4B,sBAAsB,CAAC,2BAA2B,sBAAsB,CAAC,0BAA0B,qBAAqBl2B,SAAQ,SAAStB,GAAgBA,GAATC,EAAEsC,EAAEvC,EAAE,IAAO,GAAjB,IAAoBC,EAAEA,EAAE,GAAGD,EAAE,MAAMA,EAAE,SAASA,GAAG,IAAIC,EAAE2D,EAAE4zB,cAAc,UAAUx3B,EAAE,KAAK,GAAGC,EAAE,OAAOA,EAAEw3B,aAAaz3B,EAAE,CAAhF,CAAkFA,KAAK,UAAUA,IAAI,SAASA,GAAGA,GAAG,MAAMA,IAAIs3B,EAAEr3B,GAAGD,EAAE,IAAGuB,EAAE,CAACm2B,aAAa,QAAQC,cAAc,UAAUC,UAAU,KAAKC,iBAAiBvK,EAAEwK,gBAAe,EAAGC,YAAW,EAAGC,UAAS,EAAGC,sBAAqB,EAAGC,kBAAiB,EAAGC,eAAe,QAAQC,oBAAmB,EAAGC,oBAAmB,EAAGC,kBAAiB,GAAIhB,EAAEiB,eAAejB,EAAEM,UAAUN,EAAEiB,cAAc,IAAIC,EAAE9yB,EAAEA,EAAE,CAAC,EAAEnE,GAAG+1B,GAAGkB,EAAEV,iBAAiBU,EAAEN,kBAAiB,GAAI,IAAIO,EAAE,CAAC,EAAE93B,OAAOC,KAAKW,GAAGD,SAAQ,SAASrB,GAAGU,OAAOe,eAAe+2B,EAAEx4B,EAAE,CAACe,YAAW,EAAG03B,IAAI,SAAS14B,GAAGw4B,EAAEv4B,GAAGD,EAAE24B,EAAEr3B,SAAQ,SAAStB,GAAG,OAAOA,EAAEy4B,EAAE,GAAE,EAAE71B,IAAI,WAAW,OAAO41B,EAAEv4B,EAAE,GAAG,IAAGU,OAAOe,eAAe+2B,EAAE,eAAe,CAACz3B,YAAW,EAAG03B,IAAI,SAAS14B,GAAGw4B,EAAEZ,UAAU53B,EAAE24B,EAAEr3B,SAAQ,SAAStB,GAAG,OAAOA,EAAEy4B,EAAE,GAAE,EAAE71B,IAAI,WAAW,OAAO41B,EAAEZ,SAAS,IAAI/0B,EAAE00B,kBAAkBkB,EAAE,IAAIE,EAAE,GAAOC,EAAtgI,GAA0gIC,EAAE,CAACC,KAAK,GAAG1D,EAAE,EAAEtF,EAAE,EAAE+B,OAAO,EAAEkH,OAAM,EAAGC,OAAM,GAAQC,EAAG,iEAAiE,SAASC,IAAK,IAAI,IAAIl5B,EAAE,GAAGC,EAAE,GAAG,EAAED,KAAKC,GAAGg5B,EAAG,GAAGE,KAAKC,SAAS,GAAG,OAAOn5B,CAAC,CAAC,SAASo5B,GAAGr5B,GAAG,IAAI,IAAIC,EAAE,GAAGI,GAAGL,GAAG,IAAIqB,SAAS,EAAEhB,KAAKJ,EAAEI,GAAGL,EAAEK,GAAG,OAAOJ,CAAC,CAAC,SAASq5B,GAAGt5B,GAAG,OAAOA,EAAEu5B,UAAUF,GAAGr5B,EAAEu5B,YAAYv5B,EAAEy3B,aAAa,UAAU,IAAI+B,MAAM,KAAK14B,QAAO,SAASd,GAAG,OAAOA,CAAC,GAAE,CAAC,SAASy5B,GAAGz5B,GAAG,MAAM,GAAGqE,OAAOrE,GAAG05B,QAAQ,KAAK,SAASA,QAAQ,KAAK,UAAUA,QAAQ,KAAK,SAASA,QAAQ,KAAK,QAAQA,QAAQ,KAAK,OAAO,CAAC,SAASC,GAAGt5B,GAAG,OAAOM,OAAOC,KAAKP,GAAG,CAAC,GAAGsF,QAAO,SAAS3F,EAAEC,GAAG,OAAOD,EAAE,GAAGqE,OAAOpE,EAAE,MAAMoE,OAAOhE,EAAEJ,GAAG25B,OAAO,IAAI,GAAE,GAAG,CAAC,SAASC,GAAG75B,GAAG,OAAOA,EAAE84B,OAAOD,EAAEC,MAAM94B,EAAEo1B,IAAIyD,EAAEzD,GAAGp1B,EAAE8vB,IAAI+I,EAAE/I,GAAG9vB,EAAE6xB,SAASgH,EAAEhH,QAAQ7xB,EAAE+4B,OAAO/4B,EAAEg5B,KAAK,CAAC,SAASc,KAAK,IAAI95B,EAAEC,EAAEI,EAAEitB,EAAEltB,EAAEq4B,EAAEb,UAAUp3B,EAAEi4B,EAAEZ,iBAAiBp3B,EAAE,koiBAAkoiB,MAAM,OAAOL,GAAGI,IAAIH,IAAIL,EAAE,IAAI+5B,OAAO,MAAM11B,OAAO,KAAK,OAAO,KAAKpE,EAAE,IAAI85B,OAAO,OAAO11B,OAAO,KAAK,OAAO,KAAKhE,EAAE,IAAI05B,OAAO,MAAM11B,OAAOhE,GAAG,KAAKI,EAAEA,EAAEi5B,QAAQ15B,EAAE,IAAIqE,OAAOjE,EAAE,MAAMs5B,QAAQz5B,EAAE,KAAKoE,OAAOjE,EAAE,MAAMs5B,QAAQr5B,EAAE,IAAIgE,OAAO7D,KAAKC,CAAC,CAAC,IAAIu5B,IAAG,EAAG,SAASC,KAAKxB,EAAEV,aAAaiC,KAAK,SAASh6B,GAAG,GAAGA,GAAGyF,EAAE,CAAC,IAAIxF,EAAE2D,EAAEzB,cAAc,SAASlC,EAAEi6B,aAAa,OAAO,YAAYj6B,EAAEk6B,UAAUn6B,EAAE,IAAI,IAAIK,EAAEuD,EAAE3B,KAAKm4B,WAAWh6B,EAAE,KAAKI,EAAEH,EAAEgB,OAAO,GAAG,EAAEb,EAAEA,IAAI,CAAC,IAAIC,EAAEJ,EAAEG,GAAGE,GAAGD,EAAE45B,SAAS,IAAIC,eAAe,EAAE,CAAC,QAAQ,QAAQl4B,QAAQ1B,KAAKN,EAAEK,EAAE,CAACmD,EAAE3B,KAAKs4B,aAAat6B,EAAEG,EAAE,CAAC,CAAxQ,CAA0Q05B,MAAME,IAAG,EAAG,CAACv3B,EAAE,CAAC+3B,OAAO,WAAW,MAAM,CAACC,IAAI,CAACC,IAAIZ,GAAGa,UAAUV,IAAI,EAAE10B,MAAM,WAAW,MAAM,CAACq1B,yBAAyB,WAAWX,IAAI,EAAEY,YAAY,WAAWZ,IAAI,EAAE,IAAGn4B,EAAEe,GAAG,CAAC,GAAI5C,KAAK6B,EAAE7B,GAAG,CAAC,GAAG6B,EAAE7B,GAAGqF,SAASxD,EAAE7B,GAAGqF,OAAO,CAAC,GAAGxD,EAAE7B,GAAGsF,QAAQzD,EAAE7B,GAAGsF,MAAM,CAAC,GAAGzD,EAAE7B,GAAGuF,QAAQ1D,EAAE7B,GAAGuF,MAAM,IAAoG,IAAIs1B,GAAGh5B,EAAE7B,GAAG86B,GAAG,GAAGC,IAAG,EAAG,SAASC,GAAGj7B,GAAGyF,IAAIu1B,GAAGE,WAAWl7B,EAAE,GAAG+6B,GAAG95B,KAAKjB,GAAG,CAAC,SAASm7B,GAAGn7B,GAAG,IAAIK,EAAEJ,EAAED,EAAE6qB,IAAmBrqB,OAAE,KAAjBJ,EAAEJ,EAAEo7B,YAAwB,CAAC,EAAEh7B,EAAeA,OAAE,KAAfA,EAAEJ,EAAEknB,UAAsB,GAAG9mB,EAAE,MAAM,iBAAiBJ,EAAEy5B,GAAGz5B,GAAG,IAAIqE,OAAOpE,EAAE,KAAKoE,QAAQhE,EAAEG,EAAEG,OAAOC,KAAKP,GAAG,CAAC,GAAGsF,QAAO,SAAS3F,EAAEC,GAAG,OAAOD,EAAE,GAAGqE,OAAOpE,EAAE,MAAMoE,OAAOo1B,GAAGp5B,EAAEJ,IAAI,KAAK,GAAE,IAAI25B,QAAQ,KAAKv1B,OAAOjE,EAAEmE,IAAI42B,IAAIE,KAAK,IAAI,MAAMh3B,OAAOpE,EAAE,IAAI,CAAC,SAASq7B,GAAGt7B,EAAEC,EAAEI,GAAG,GAAGL,GAAGA,EAAEC,IAAID,EAAEC,GAAGI,GAAG,MAAM,CAACk7B,OAAOt7B,EAAE4F,SAASxF,EAAEuF,KAAK5F,EAAEC,GAAGI,GAAG,CAAmI,SAASm7B,GAAGx7B,EAAEC,EAAEI,EAAED,GAAG,IAAI,IAAII,EAAEC,EAAEC,EAAEC,OAAOC,KAAKZ,GAAGmB,EAAET,EAAEW,OAAOE,OAAE,IAASnB,EAAEq7B,GAAGx7B,EAAEG,GAAGH,EAAE6B,OAAE,IAASzB,GAAGG,EAAE,EAAER,EAAEU,EAAE,MAAMF,EAAE,EAAEH,GAAGG,EAAEW,EAAEX,IAAIsB,EAAEP,EAAEO,EAAE9B,EAAES,EAAEC,EAAEF,IAAIC,EAAET,GAAG,OAAO8B,CAAC,CAA9R2D,KAAKu1B,IAAIp3B,EAAE5B,gBAAgB05B,SAAS,aAAa,iBAAiBt2B,KAAKxB,EAAE+3B,cAAc/3B,EAAE1B,iBAAiB,oBAAvpB,SAAS05B,IAAKh4B,EAAEi4B,oBAAoB,mBAAmBD,GAAIZ,GAAG,EAAED,GAAGx2B,KAAI,SAASvE,GAAG,OAAOA,GAAG,GAAE,KAA6uB,IAAIy7B,GAAG,SAASj7B,EAAEC,GAAG,OAAO,SAAST,EAAEC,EAAEI,EAAED,GAAG,OAAOI,EAAEwE,KAAKvE,EAAET,EAAEC,EAAEI,EAAED,EAAE,CAAC,EAAE,SAAS07B,GAAG97B,GAAuN,OAApNA,EAAE,SAASA,GAAG,IAAI,IAAIC,EAAE,GAAGI,EAAE,EAAED,EAAEJ,EAAEqB,OAAOhB,EAAED,GAAG,CAAC,IAAII,EAAEC,EAAET,EAAE+7B,WAAW17B,KAAK,OAAOI,GAAGA,GAAG,OAAOJ,EAAED,EAAE,QAAQ,OAAOI,EAAER,EAAE+7B,WAAW17B,OAAOJ,EAAEgB,OAAO,KAAKR,IAAI,KAAK,KAAKD,GAAG,QAAQP,EAAEgB,KAAKR,GAAGJ,KAAKJ,EAAEgB,KAAKR,EAAE,CAAC,OAAOR,CAAC,CAA7M,CAA+MD,GAAU,IAAIA,EAAEqB,OAAOrB,EAAE,GAAG+E,SAAS,IAAI,IAAI,CAAC,SAASi3B,GAAG57B,GAAG,OAAOO,OAAOC,KAAKR,GAAGuF,QAAO,SAAS3F,EAAEC,GAAG,IAAII,EAAED,EAAEH,GAAG,OAAQI,EAAEuF,KAAK5F,EAAEK,EAAEwF,UAAUxF,EAAEuF,KAAK5F,EAAEC,GAAGI,EAAEL,CAAC,GAAE,CAAC,EAAE,CAAC,SAASi8B,GAAGj8B,EAAEC,EAAEI,GAAyDA,OAAE,KAApDD,GAAG,EAAEgB,UAAUC,aAAQ,IAAShB,EAAEA,EAAE,CAAC,GAAG0F,YAAwB3F,EAApE,IAAsEA,EAAE47B,GAAG/7B,GAAG,mBAAmB66B,GAAGv1B,MAAMS,SAAS3F,EAAEy6B,GAAGx1B,OAAOtF,GAAG0F,EAAEA,EAAE,CAAC,EAAEo1B,GAAGx1B,OAAOtF,IAAI,CAAC,GAAGI,GAAG06B,GAAGv1B,MAAMS,QAAQhG,EAAEg8B,GAAG/7B,IAAI,QAAQD,GAAGi8B,GAAG,KAAKh8B,EAAE,CAAC,IAAIi8B,GAAGpB,GAAGx1B,OAAO62B,GAAGrB,GAAGt1B,MAAM42B,IAAIj7B,EAAEmB,EAAE,CAAC,EAAE8zB,EAAEz1B,OAAO07B,OAAO5F,EAAEL,KAAKj1B,EAAEmB,EAAEmoB,EAAE9pB,OAAO07B,OAAO5F,EAAEhM,KAAKnoB,GAAGg6B,GAAG,KAAKC,GAAG,CAAC,EAAEC,GAAG,CAAC,EAAEC,GAAG,CAAC,EAAEC,GAAG,CAAC,EAAEC,GAAG,CAAC,EAAEC,IAAIz7B,EAAEI,EAAE,CAAC,EAAE60B,EAAEz1B,OAAOC,KAAK21B,EAAEH,KAAKj1B,EAAEI,EAAEkpB,EAAE9pB,OAAOC,KAAK21B,EAAE9L,KAAKlpB,GAAgH,SAASs7B,KAAK,SAAS78B,EAAEI,GAAG,OAAOo7B,GAAGU,IAAG,SAASl8B,EAAEC,EAAEI,GAAG,OAAOL,EAAEK,GAAGm7B,GAAGv7B,EAAEG,EAAE,CAAC,GAAGJ,CAAC,GAAE,CAAC,EAAE,CAACu8B,GAAGv8B,GAAE,SAASC,EAAED,EAAEK,GAAG,OAAOL,EAAE,KAAKC,EAAED,EAAE,IAAIK,GAAGL,EAAE,IAAIA,EAAE,GAAGc,QAAO,SAASd,GAAG,MAAM,iBAAiBA,CAAC,IAAGsB,SAAQ,SAAStB,GAAGC,EAAED,EAAE+E,SAAS,KAAK1E,CAAC,IAAGJ,CAAC,IAAGu8B,GAAGx8B,GAAE,SAASC,EAAED,EAAEK,GAAG,OAAOJ,EAAEI,GAAGA,EAAEL,EAAE,IAAIA,EAAE,GAAGc,QAAO,SAASd,GAAG,MAAM,iBAAiBA,CAAC,IAAGsB,SAAQ,SAAStB,GAAGC,EAAED,GAAGK,CAAC,IAAGJ,CAAC,IAAG08B,GAAG38B,GAAE,SAASC,EAAED,EAAEK,GAAU,OAAPL,EAAEA,EAAE,GAAUC,EAAEI,GAAGA,EAAEL,EAAEsB,SAAQ,SAAStB,GAAGC,EAAED,GAAGK,CAAC,IAAGJ,CAAC,IAAG,IAAIO,EAAE,QAAQ07B,IAAIzD,EAAEqE,aAAa78B,EAAEu7B,GAAGW,IAAG,SAASn8B,EAAEC,GAAG,IAAII,EAAEJ,EAAE,GAAGG,EAAEH,EAAE,GAAU,OAAPA,EAAEA,EAAE,GAAS,QAAQG,GAAGI,IAAIJ,EAAE,OAAO,iBAAiBC,IAAIL,EAAE+8B,MAAM18B,GAAG,CAACk7B,OAAOn7B,EAAEyF,SAAS5F,IAAI,iBAAiBI,IAAIL,EAAEg9B,SAAS38B,EAAE0E,SAAS,KAAK,CAACw2B,OAAOn7B,EAAEyF,SAAS5F,IAAID,CAAC,GAAE,CAAC+8B,MAAM,CAAC,EAAEC,SAAS,CAAC,IAAIP,GAAGx8B,EAAE88B,MAAML,GAAGz8B,EAAE+8B,SAASV,GAAGW,GAAGxE,EAAEf,aAAa,CAACwF,OAAOzE,EAAEd,eAAe,CAAC,SAASwF,GAAGn9B,EAAEC,GAAG,OAAOs8B,GAAGv8B,IAAI,CAAC,GAAGC,EAAE,CAAC,SAASm9B,GAAGp9B,EAAEC,GAAG,OAAO08B,GAAG38B,IAAI,CAAC,GAAGC,EAAE,CAAC,SAASo9B,GAAGr9B,GAAG,OAAOy8B,GAAGz8B,IAAI,CAACu7B,OAAO,KAAK11B,SAAS,KAAK,CAAsI,SAASo3B,GAAGj9B,EAAEC,GAAG,IAAII,GAAG,EAAEe,UAAUC,aAAQ,IAASpB,EAAEA,EAAE,CAAC,GAAGi9B,OAA6E,OAArD78B,EAAEk2B,EAAnBt2B,OAAE,IAASI,EAAE+1B,EAAE/1B,GAASL,GAAGK,EAAEm2B,EAAEv2B,GAAGD,IAAIw2B,EAAEv2B,GAAGI,GAAGL,EAAEA,KAAK86B,GAAGx1B,OAAOtF,EAAE,KAAYK,GAAGL,GAAG,IAAI,CAAhS8B,EAAE,SAAS9B,GAAGs8B,GAAGW,GAAGj9B,EAAE03B,aAAa,CAACwF,OAAOzE,EAAEd,eAAe,EAAEgB,EAAE13B,KAAKa,GAAG+6B,KAAyN,IAAIS,IAAIn8B,EAAElB,EAAE,CAAC,EAAEm2B,EAAEz1B,OAAOC,KAAK61B,EAAEL,KAAKj1B,EAAElB,EAAEwqB,EAAE9pB,OAAOC,KAAK61B,EAAEhM,KAAKxqB,GAAG,SAASs9B,GAAGv9B,EAAEC,GAAG,IAAwDG,OAAE,KAAtDH,GAAG,EAAEmB,UAAUC,aAAQ,IAASpB,EAAEA,EAAE,CAAC,GAAGu9B,cAA0Bv9B,EAAEO,GAAGW,EAAElB,EAAE,CAAC,EAAEm2B,EAAE,GAAG/xB,OAAOo0B,EAAEb,UAAU,KAAKvzB,OAAO+xB,IAAIj1B,EAAElB,EAAEwqB,EAAE,GAAGpmB,OAAOo0B,EAAEb,UAAU,KAAKvzB,OAAOomB,IAAIxqB,GAAGQ,EAAE,KAAKC,EAAE01B,EAAmnB,OAAhnBp2B,EAAEy9B,SAASj9B,EAAE41B,KAAKp2B,EAAE09B,MAAK,SAAS19B,GAAG,OAAOs9B,GAAGlH,GAAGqH,SAASz9B,EAAE,OAAMU,EAAE01B,IAAIp2B,EAAEy9B,SAASj9B,EAAEiqB,KAAKzqB,EAAE09B,MAAK,SAAS19B,GAAG,OAAOs9B,GAAG7S,GAAGgT,SAASz9B,EAAE,OAAMU,EAAE+pB,GAAGxqB,EAAED,EAAE2F,QAAO,SAAS3F,EAAEC,GAAG,IAAII,EAA5sD,SAAYL,EAAEC,GAAsBA,GAAfI,EAAEJ,EAAEu5B,MAAM,MAAS,GAAvB,IAA0Bn5B,EAAEA,EAAE4E,MAAM,GAAGo2B,KAAK,KAAK,OAAOp7B,IAAID,GAAG,KAAKK,IAAIg3B,EAAEj1B,QAAQ/B,GAAG,KAAKA,CAAC,CAAkmDs9B,CAAGlF,EAAEb,UAAU33B,GAAG,OAAOi8B,GAAGj8B,IAAIA,EAAEm8B,GAAG17B,GAAG+8B,SAASx9B,GAAGy2B,EAAEh2B,GAAGT,GAAGA,EAAEQ,EAAER,EAAED,EAAEu7B,OAAOt7B,IAAI,EAAE28B,GAAGl8B,GAAG0B,QAAQnC,IAAIQ,EAAER,EAAED,EAAEu7B,OAAO0B,GAAGh9B,EAAE,CAACi9B,OAAOx8B,KAAKL,EAAEL,EAAE6F,SAASxF,EAAEJ,IAAIw4B,EAAEZ,kBAAkB53B,IAAIO,EAAE41B,IAAIn2B,IAAIO,EAAEiqB,IAAIzqB,EAAE49B,KAAK38B,KAAKhB,IAAIG,GAAGJ,EAAEu7B,QAAQv7B,EAAE6F,WAAWxF,EAAE,OAAOI,EAAE48B,GAAGr9B,EAAE6F,UAAU,CAAC,EAAE5F,EAAEm9B,GAAGp9B,EAAEu7B,OAAOv7B,EAAE6F,UAAUxF,EAAEk7B,SAAS96B,EAAE,MAAMT,EAAE6F,SAASxF,EAAEwF,UAAU5F,GAAGD,EAAE6F,SAAS7F,EAAEu7B,OAAOl7B,EAAEk7B,QAAQv7B,EAAEu7B,OAAO,QAAQv7B,EAAEu7B,QAAQW,GAAGl5B,MAAMk5B,GAAGn5B,KAAK01B,EAAEqE,eAAe98B,EAAEu7B,OAAO,QAAQv7B,CAAC,GAA9iC,CAACu7B,OAAO,KAAK11B,SAAS,KAAK+3B,KAAK,MAA6hC59B,EAAEy9B,SAAS,cAAcz9B,EAAEy9B,SAAS,UAAUx9B,EAAEs7B,OAAO,QAAQv7B,EAAEy9B,SAAS,eAAez9B,EAAEy9B,SAAS,UAAUx9B,EAAEs7B,OAAO,OAAOt7B,EAAEs7B,QAAQ76B,IAAI+pB,IAAIyR,GAAG34B,OAAOk1B,EAAEqE,eAAe78B,EAAEs7B,OAAO,OAAOt7B,EAAE4F,SAASu3B,GAAGn9B,EAAEs7B,OAAOt7B,EAAE4F,WAAW5F,EAAE4F,UAAU,OAAO5F,EAAEs7B,QAAQ,OAAO96B,IAAIR,EAAEs7B,OAAOe,IAAI,OAAOr8B,CAAC,CAAKqC,EAAE,WAAW,SAAStC,KAAK,SAASA,EAAEC,GAAG,KAAKD,aAAaC,GAAG,MAAM,IAAIoF,UAAU,oCAAoC,CAA3F,CAA6Fw4B,KAAK79B,GAAG69B,KAAKC,YAAY,CAAC,CAAC,CAAC,IAAI79B,EAAEI,EAAI,OAAOJ,EAAED,GAAGK,EAAE,CAAC,CAACyiB,IAAI,MAAMnhB,MAAM,WAAW,IAAI,IAAItB,EAAEw9B,KAAK79B,EAAEoB,UAAUC,OAAOpB,EAAE,IAAI8B,MAAM/B,GAAGI,EAAE,EAAEA,EAAEJ,EAAEI,IAAIH,EAAEG,GAAGgB,UAAUhB,GAAG,IAAII,EAAEP,EAAE0F,OAAOk4B,KAAKE,iBAAiB,CAAC,GAAGp9B,OAAOC,KAAKJ,GAAGc,SAAQ,SAAStB,GAAGK,EAAEy9B,YAAY99B,GAAG0F,EAAEA,EAAE,CAAC,EAAErF,EAAEy9B,YAAY99B,IAAI,CAAC,GAAGQ,EAAER,IAAIi8B,GAAGj8B,EAAEQ,EAAER,IAAI,IAAIC,EAAEw2B,EAAEL,GAAGp2B,GAAGC,GAAGg8B,GAAGh8B,EAAEO,EAAER,IAAI68B,IAAI,GAAE,GAAG,CAAC/Z,IAAI,QAAQnhB,MAAM,WAAWk8B,KAAKC,YAAY,CAAC,CAAC,GAAG,CAAChb,IAAI,mBAAmBnhB,MAAM,SAASnB,EAAER,GAAG,IAAIS,EAAET,EAAEu7B,QAAQv7B,EAAE6F,UAAU7F,EAAE4F,KAAK,CAAC,EAAE5F,GAAGA,EAAE,OAAOW,OAAOC,KAAKH,GAAG8D,KAAI,SAASvE,GAAG,IAAWK,GAAPJ,EAAEQ,EAAET,IAAOu7B,OAAoBn7B,GAAbJ,EAAEC,EAAE4F,SAAW5F,EAAE2F,MAAK3F,EAAEG,EAAE,GAAGI,EAAEH,KAAKG,EAAEH,GAAG,CAAC,GAAG,EAAEJ,EAAEoB,QAAQpB,EAAEqB,SAAQ,SAAStB,GAAG,iBAAiBA,IAAIQ,EAAEH,GAAGL,GAAGI,EAAE,IAAGI,EAAEH,GAAGL,GAAGI,CAAC,IAAGI,CAAC,MAA3r7B,SAAWR,EAAEC,GAAG,IAAI,IAAII,EAAE,EAAEA,EAAEJ,EAAEoB,OAAOhB,IAAI,CAAC,IAAID,EAAEH,EAAEI,GAAGD,EAAEY,WAAWZ,EAAEY,aAAY,EAAGZ,EAAEwB,cAAa,EAAG,UAAUxB,IAAIA,EAAEyB,UAAS,GAAIlB,OAAOe,eAAe1B,EAAEI,EAAE0iB,IAAI1iB,EAAE,CAAC,CAA4h7BK,CAAER,EAAE6E,UAAUzE,GAAaM,OAAOe,eAAezB,EAAE,YAAY,CAAC4B,UAAS,IAAK7B,CAAC,CAAh2B,GAAo2BuB,EAAE,GAA52B,IAA+2By8B,GAAG,CAAC,EAAEC,GAAG,CAAC,EAAEC,GAAGv9B,OAAOC,KAAKq9B,IAAI,SAASE,GAAGn+B,EAAEC,GAAG,IAAI,IAAII,EAAEe,UAAUC,OAAOjB,EAAE,IAAI2B,MAAM,EAAE1B,EAAEA,EAAE,EAAE,GAAGG,EAAE,EAAEA,EAAEH,EAAEG,IAAIJ,EAAEI,EAAE,GAAGY,UAAUZ,GAAG,OAAOw9B,GAAGh+B,IAAI,IAAIsB,SAAQ,SAAStB,GAAGC,EAAED,EAAEkB,MAAM,KAAK,CAACjB,GAAGoE,OAAOjE,GAAG,IAAGH,CAAC,CAAC,SAASm+B,GAAGp+B,GAAG,IAAI,IAAIC,EAAEmB,UAAUC,OAAOhB,EAAE,IAAI0B,MAAM,EAAE9B,EAAEA,EAAE,EAAE,GAAGG,EAAE,EAAEA,EAAEH,EAAEG,IAAIC,EAAED,EAAE,GAAGgB,UAAUhB,IAAI49B,GAAGh+B,IAAI,IAAIsB,SAAQ,SAAStB,GAAGA,EAAEkB,MAAM,KAAKb,EAAE,GAAE,CAAC,SAASg+B,GAAGr+B,GAAG,IAAIC,EAAED,EAA4C,OAA1CA,EAAE+B,MAAM+C,UAAUG,MAAMD,KAAK5D,UAAU,GAAU68B,GAAGh+B,GAAGg+B,GAAGh+B,GAAGiB,MAAM,KAAKlB,QAAG,CAAM,CAAC,SAASs+B,GAAGt+B,GAAG,OAAOA,EAAEu7B,SAASv7B,EAAEu7B,OAAO,OAAO,IAAIt7B,EAAED,EAAE6F,SAAwB,GAAf7F,EAAEA,EAAEu7B,QAAQe,GAAMr8B,EAAE,OAAOA,EAAEm9B,GAAGp9B,EAAEC,IAAIA,EAAEq7B,GAAGiD,GAAGT,YAAY99B,EAAEC,IAAIq7B,GAAGR,GAAGx1B,OAAOtF,EAAEC,EAAE,CAAC,IAAIs+B,GAAG,IAAIj8B,EAAEk8B,GAAG,CAACC,OAAO,WAAWhG,EAAEX,gBAAe,EAAGW,EAAEP,kBAAiB,EAAGkG,GAAG,SAAS,EAAEM,OAAOjG,EAAEgC,IAAI,CAACkE,MAAM,WAAW,IAAI3+B,EAAE,EAAEoB,UAAUC,aAAQ,IAASD,UAAU,GAAGA,UAAU,GAAG,CAAC,EAAE,OAAOqE,GAAG24B,GAAG,cAAcp+B,GAAGq+B,GAAG,qBAAqBr+B,GAAGq+B,GAAG,QAAQr+B,IAAI4+B,QAAQC,OAAO,yCAAyC,EAAEC,MAAM,WAAW,IAAI9+B,EAAE,EAAEoB,UAAUC,aAAQ,IAASD,UAAU,GAAGA,UAAU,GAAG,CAAC,EAAEnB,EAAED,EAAE++B,oBAAmB,IAAKtG,EAAEX,iBAAiBW,EAAEX,gBAAe,GAAIW,EAAEP,kBAAiB,EAAG+C,IAAG,WAAW+D,GAAG,CAACD,mBAAmB9+B,IAAIm+B,GAAG,QAAQp+B,EAAE,GAAE,GAAGi/B,MAAM,CAACr5B,KAAK,SAAS5F,GAAG,GAAG,OAAOA,EAAE,OAAO,KAAK,GAAG,WAAWQ,EAAER,IAAIA,EAAEu7B,QAAQv7B,EAAE6F,SAAS,MAAM,CAAC01B,OAAOv7B,EAAEu7B,OAAO11B,SAASu3B,GAAGp9B,EAAEu7B,OAAOv7B,EAAE6F,WAAW7F,EAAE6F,UAAU,GAAG9D,MAAM2C,QAAQ1E,IAAI,IAAIA,EAAEqB,OAAO,CAAC,IAAIpB,EAAE,IAAID,EAAE,GAAGoC,QAAQ,OAAOpC,EAAE,GAAGiF,MAAM,GAAGjF,EAAE,GAAGK,EAAE48B,GAAGj9B,EAAE,IAAI,MAAM,CAACu7B,OAAOl7B,EAAEwF,SAASu3B,GAAG/8B,EAAEJ,IAAIA,EAAE,CAAC,MAAG,iBAAiBD,KAAK,EAAEA,EAAEoC,QAAQ,GAAGiC,OAAOo0B,EAAEb,UAAU,OAAO53B,EAAEk/B,MAAMvI,IAA+C,CAAC4E,QAA3Ct7B,EAAEs9B,GAAGv9B,EAAEw5B,MAAM,KAAK,CAACgE,aAAY,KAAqBjC,QAAQe,GAAGz2B,SAASu3B,GAAGn9B,EAAEs7B,OAAOt7B,EAAE4F,WAAW5F,EAAE4F,UAAgB,iBAAiB7F,EAAE,CAACu7B,OAAOe,GAAGz2B,SAASu3B,GAAGd,GAAGt8B,IAAIA,QAAG,CAAM,GAAGm/B,QAAQZ,GAAGa,mBAAmBd,GAAGe,OAAOlE,IAAI6D,GAAG,WAAW,IAAqFh/B,OAAE,KAAnFA,GAAG,EAAEoB,UAAUC,aAAQ,IAASD,UAAU,GAAGA,UAAU,GAAG,CAAC,GAAG29B,oBAAgCn7B,EAAE5D,GAAG,EAAEW,OAAOC,KAAKk6B,GAAGx1B,QAAQjE,QAAQo3B,EAAEqE,eAAer3B,GAAGgzB,EAAEX,gBAAgB0G,GAAG/D,IAAIkE,MAAM,CAACt0B,KAAKrK,GAAG,EAAE,SAASs/B,GAAGr/B,EAAED,GAAG,OAAOW,OAAOe,eAAezB,EAAE,WAAW,CAAC2C,IAAI5C,IAAIW,OAAOe,eAAezB,EAAE,OAAO,CAAC2C,IAAI,WAAW,OAAO3C,EAAEs/B,SAASh7B,IAAI42B,GAAG,IAAIx6B,OAAOe,eAAezB,EAAE,OAAO,CAAC2C,IAAI,WAAW,GAAG6C,EAAE,CAAC,IAAIzF,EAAE4D,EAAEzB,cAAc,OAAO,OAAOnC,EAAEm6B,UAAUl6B,EAAEu/B,KAAKx/B,EAAEknB,QAAQ,CAAC,IAAIjnB,CAAC,CAAC,SAASw/B,GAAGz/B,GAAG,IAAi0B0C,EAAEiB,EAAEd,EAAEe,EAAE6B,EAA3zBpF,GAAVJ,EAAED,EAAEoe,OAAUshB,KAAKt/B,EAAEH,EAAE20B,KAAKp0B,EAAER,EAAEu7B,OAAO96B,EAAET,EAAE6F,SAASnF,EAAEV,EAAE2/B,UAAUx+B,EAAEnB,EAAE4/B,OAAOr+B,EAAEvB,EAAE6/B,MAAM/9B,EAAE9B,EAAE8/B,OAAOz9B,EAAErC,EAAE+/B,QAAQz9B,EAAEtC,EAAEggC,MAAoBx9B,OAAE,KAAhBD,EAAEvC,EAAEigC,YAAwB19B,EAAgBtC,GAAdwC,EAAErC,EAAE8/B,MAAM9/B,EAAEC,GAAM8/B,MAAiB59B,GAAXvC,EAAEyC,EAAE29B,OAAS,QAAQ5/B,GAAEiC,EAAE,CAACg2B,EAAEZ,iBAAiBp3B,EAAE,GAAG4D,OAAOo0B,EAAEb,UAAU,KAAKvzB,OAAO5D,GAAG,IAAIK,QAAO,SAASd,GAAG,OAAO,IAAIsC,EAAE+9B,QAAQj+B,QAAQpC,EAAE,IAAGc,QAAO,SAASd,GAAG,MAAM,KAAKA,KAAKA,CAAC,IAAGqE,OAAO/B,EAAE+9B,SAAShF,KAAK,KAA8tB,OAAztB54B,EAAE,CAACykB,SAAS,GAAGkU,WAAW11B,EAAEA,EAAE,CAAC,EAAEpD,EAAE84B,YAAY,CAAC,EAAE,CAAC,cAAc56B,EAAE,YAAYC,EAAE6/B,MAAM79B,EAAE89B,KAAKj+B,EAAE84B,WAAWmF,MAAM,MAAMC,MAAM,6BAA6BC,QAAQ,OAAOp8B,OAAOpE,EAAE,KAAKoE,OAAOrE,MAAMA,EAAEuC,KAAKD,EAAE+9B,QAAQj+B,QAAQ,SAAS,CAAC+9B,MAAM,GAAG97B,OAAOpE,EAAED,EAAE,GAAG,MAAM,OAAO,CAAC,EAAEwC,IAAIC,EAAE24B,WAAW/Y,GAAG,IAAI9gB,IAAIkB,EAAEykB,SAASjmB,KAAK,CAAC4pB,IAAI,QAAQuQ,WAAW,CAACsF,GAAGj+B,EAAE24B,WAAW,oBAAoB,SAAS/2B,OAAOhC,GAAG62B,MAAOhS,SAAS,CAAC3lB,YAAYkB,EAAE24B,WAAWyE,OAAqB7/B,EAAE0F,EAAEA,EAAE,CAAC,EAAEjD,GAAG,CAAC,EAAE,CAAC84B,OAAO/6B,EAAEqF,SAASpF,EAAEi/B,KAAKr/B,EAAEu0B,KAAKx0B,EAAE0/B,OAAOh+B,EAAE69B,UAAUj/B,EAAEk/B,OAAOz+B,EAAEmE,OAAOI,EAAEA,EAAE,CAAC,EAAE1F,GAAGsC,EAAEgD,UAAiJjF,GAAvID,EAAEA,EAAE8/B,OAAO7/B,EAAE6/B,MAAM7B,GAAG,uBAAuBr+B,IAAI,CAACknB,SAAS,GAAGkU,WAAW,CAAC,GAAGiD,GAAG,uBAAuBr+B,IAAI,CAACknB,SAAS,GAAGkU,WAAW,CAAC,IAAOlU,SAAS9mB,EAAEA,EAAEg7B,WAAkBp7B,EAAEknB,SAAS7mB,EAAEL,EAAEo7B,WAAWh7B,EAAEe,GAAGwC,GAAGjB,EAAE1C,GAAGu7B,OAAO14B,EAAEH,EAAEmD,SAASjC,EAAElB,EAAEwkB,SAASzhB,EAAE/C,EAAE04B,WAAW14B,GAAE,KAAMA,EAAEA,EAAEk9B,QAAQ,GAAGv7B,OAAOV,EAAE,KAAKU,OAAOo0B,EAAEb,UAAU,KAAKvzB,OAAOxB,GAAGH,EAAE,CAAC,CAACmoB,IAAI,MAAMuQ,WAAW,CAACuF,MAAM,kBAAkBzZ,SAAS,CAAC,CAAC2D,IAAI,SAASuQ,WAAW11B,EAAEA,EAAE,CAAC,EAAED,GAAG,CAAC,EAAE,CAACi7B,GAAGh+B,IAAIwkB,SAAStjB,QAAQf,GAAGc,EAAE3D,GAAGknB,SAASzhB,EAAE9B,EAAE+7B,KAAKh9B,EAAEiB,EAAEixB,KAAKhxB,EAAED,EAAEy3B,WAAWp7B,EAAE2D,EAAE2B,OAAOu0B,GAAGl2B,EAAEA,EAAEg8B,YAAYl6B,EAAEy6B,QAAQx9B,EAAEw9B,QAAQx9B,EAAE+C,EAAE06B,MAAM16B,EAAE26B,OAAO,EAAE36B,EAAE,GAAG7B,EAAE+8B,MAAMhH,GAAGj0B,EAAEA,EAAE,CAAC,EAAE1F,GAAG,CAAC,EAAE,CAAC,mBAAmB,GAAGqE,OAAO3B,EAAEiB,EAAEyxB,EAAE,GAAG,OAAO/wB,OAAOoB,EAAE9B,EAAEmsB,EAAE,GAAG,UAAU,CAAC,CAACjF,IAAI,MAAMuQ,WAAWx3B,EAAEsjB,SAASrkB,IAAI,CAAC,SAAS+9B,GAAG5gC,GAAG,IAAIC,EAAED,EAAE6gC,QAAQxgC,EAAEL,EAAEmgC,MAAM//B,EAAEJ,EAAEogC,OAAO5/B,EAAER,EAAE2/B,UAAUl/B,EAAET,EAAE6/B,MAAMn/B,EAAEV,EAAEggC,MAAM7+B,EAAEnB,EAAEigC,UAAU1+B,OAAE,IAASJ,GAAGA,EAAirB,OAA/qBnB,EAAE0F,EAAEA,EAAEA,EAAE,CAAC,EAAEhF,EAAE06B,YAAY36B,EAAE,CAACo/B,MAAMp/B,GAAG,CAAC,GAAG,CAAC,EAAE,CAAC6/B,MAAM5/B,EAAE2/B,QAAQhF,KAAK,OAAO95B,IAAIvB,EAAEqiB,GAAG,IAAIlhB,EAAEuE,EAAE,CAAC,EAAEhF,EAAE4E,QAAQu0B,GAAGr5B,KAAKW,EAAEw+B,WAAWj/B,GAAGa,EAAE,CAACo+B,UAAUn/B,EAAEsgC,eAAc,EAAGX,MAAM9/B,EAAE+/B,OAAOhgC,IAAIu/B,UAAUn/B,EAAEe,EAAE4+B,MAAM//B,OAAE,KAAUC,EAAEkB,EAAE6+B,QAA30/B,GAAq1/B//B,EAAEkB,OAAE,KAAUlB,EAAEkB,EAAEu/B,gBAAgBzgC,EAAEA,EAAE,GAAGA,GAAGkB,GAAGuE,EAAE,aAAazB,OAAO3D,EAAE00B,EAAEwD,QAAG,IAASp4B,EAA16/B,GAA86/BA,GAAG,EAAE,QAAQ6D,OAAO3D,EAAEovB,EAAE8I,EAAEx4B,EAAE,EAAE,QAAQmB,EAAE,yBAAyB8C,OAAO3D,EAAE00B,EAAEwD,EAAE,qBAAqBv0B,OAAO3D,EAAEovB,EAAE8I,EAAE,SAAS,aAAav0B,OAAO3D,EAAE00B,EAAEwD,EAAE,QAAQv0B,OAAO3D,EAAEovB,EAAE8I,EAAE,QAAQv4B,GAAG,SAASgE,OAAO3D,EAAEo4B,KAAKF,GAAGl4B,EAAEq4B,OAAO,EAAE,GAAG,MAAM10B,OAAO3D,EAAEo4B,KAAKF,GAAGl4B,EAAEs4B,OAAO,EAAE,GAAG,MAAM34B,GAAG,UAAUgE,OAAO3D,EAAEmxB,OAAO,UAAU1wB,EAAE,qBAAqBA,EAAEw+B,WAAmB,GAARx+B,EAAEw4B,GAAGx4B,IAAOE,SAASrB,EAAE2gC,MAAMx/B,IAAGA,EAAE,IAAYF,KAAK,CAAC4pB,IAAI,OAAOuQ,WAAWp7B,EAAEknB,SAAS,CAACjnB,KAAKQ,GAAGU,EAAEF,KAAK,CAAC4pB,IAAI,OAAOuQ,WAAW,CAACkF,MAAM,WAAWpZ,SAAS,CAACzmB,KAAKU,CAAC,CAAC,IAAI4/B,GAAGjG,GAAGx1B,OAAO,SAAS07B,GAAGhhC,GAAG,IAAIC,EAAED,EAAE,GAAGK,EAAEL,EAAE,GAAwB,OAArBA,EAAEuC,EAAEvC,EAAEiF,MAAM,GAAG,GAAG,GAAS,CAACi7B,OAAM,EAAGC,MAAMlgC,EAAEmgC,OAAO//B,EAAEuF,KAAK7D,MAAM2C,QAAQ1E,GAAG,CAAC6qB,IAAI,IAAIuQ,WAAW,CAACkF,MAAM,GAAGj8B,OAAOo0B,EAAEb,UAAU,KAAKvzB,OAAO2yB,EAAEC,QAAQ/P,SAAS,CAAC,CAAC2D,IAAI,OAAOuQ,WAAW,CAACkF,MAAM,GAAGj8B,OAAOo0B,EAAEb,UAAU,KAAKvzB,OAAO2yB,EAAEI,WAAWpG,KAAK,eAAelrB,EAAE9F,EAAE,KAAK,CAAC6qB,IAAI,OAAOuQ,WAAW,CAACkF,MAAM,GAAGj8B,OAAOo0B,EAAEb,UAAU,KAAKvzB,OAAO2yB,EAAEG,SAASnG,KAAK,eAAelrB,EAAE9F,EAAE,OAAO,CAAC6qB,IAAI,OAAOuQ,WAAW,CAACpK,KAAK,eAAelrB,EAAE9F,IAAI,CAAC,IAAIihC,GAAG,CAACf,OAAM,EAAGC,MAAM,IAAIC,OAAO,KAAK,SAASc,GAAG1gC,EAAEC,GAAG,IAAIC,EAAED,EAAE,MAAM,OAAOA,GAAG,OAAOg4B,EAAEf,eAAej3B,EAAE67B,IAAI,IAAIsC,SAAQ,SAAS5+B,EAAEC,GAAG,IAAII,EAAED,EAA4B,GAA1Bi+B,GAAG,uBAA0B,OAAO39B,IAAIN,EAAEi9B,GAAG78B,IAAI,CAAC,EAAEA,EAAEJ,EAAEyF,UAAUrF,EAAEC,EAAEL,EAAEm7B,QAAQ96B,GAAGD,GAAGC,GAAGsgC,GAAGtgC,IAAIsgC,GAAGtgC,GAAGD,GAAG,OAAOR,EAAEghC,GAAGD,GAAGtgC,GAAGD,KAAKH,EAAEG,EAAEJ,EAAEK,EAAE01B,GAAGsC,EAAEH,mBAAmBj4B,GAAG8gC,QAAQC,MAAM,mBAAmB/8B,OAAOhE,EAAE,kBAAkBgE,OAAOjE,EAAE,kBAAkBJ,EAAE0F,EAAEA,EAAE,CAAC,EAAEu7B,IAAI,CAAC,EAAE,CAACr7B,KAAK6yB,EAAEH,kBAAkB93B,GAAG69B,GAAG,wBAAwB,CAAC,IAAI,GAAE,CAAC,SAASgD,KAAK,CAAC,SAASC,GAAGthC,GAAGuhC,GAAG1L,KAAK,GAAGxxB,OAAOm9B,GAAG,KAAKn9B,OAAOrE,EAAE,UAAUuhC,GAAGzL,QAAQ,GAAGzxB,OAAOm9B,GAAG,KAAKn9B,OAAOrE,GAAG,GAAGqE,OAAOm9B,GAAG,KAAKn9B,OAAOrE,EAAE,WAAW,GAAGqE,OAAOm9B,GAAG,KAAKn9B,OAAOrE,EAAE,SAAS,CAAC,IAAIuhC,GAAG9I,EAAEJ,oBAAoB71B,GAAGA,EAAEqzB,MAAMrzB,EAAEszB,QAAQtzB,EAAE,CAACqzB,KAAKwL,GAAGvL,QAAQuL,IAAIG,GAAG,aAAaC,GAAG,CAACC,MAAM,SAAS1hC,GAAG,OAAOuhC,GAAG1L,KAAK,GAAGxxB,OAAOm9B,GAAG,KAAKn9B,OAAOrE,EAAE,YAAY,WAAW,OAAOshC,GAAGthC,EAAE,CAAC,EAAE2hC,IAAIL,IAAIM,GAAG,WAAW,EAAE,SAASC,GAAG7hC,GAAG,MAAM,iBAAiBA,EAAEy3B,aAAaz3B,EAAEy3B,aAAapV,GAAG,KAAK,CAAC,SAASyf,GAAG9hC,GAAG,OAAO4D,EAAEm+B,gBAAgB,6BAA6B/hC,EAAE,CAAC,SAASgiC,GAAGhiC,GAAG,OAAO4D,EAAEzB,cAAcnC,EAAE,CAAC,IAAIiiC,GAAG,CAACvI,QAAQ,SAAS15B,GAAG,IAAIC,EAAED,EAAE,GAAGC,EAAEiiC,aAAaliC,EAAE,GAAGsB,SAAQ,SAAStB,GAAGC,EAAEiiC,WAAW3H,aAAa,SAASt6B,EAAEI,EAAEL,GAAG,IAAiDI,OAAE,KAA/CJ,GAAG,EAAEoB,UAAUC,aAAQ,IAASrB,EAAEA,EAAE,CAAC,GAAGmiC,MAAkB,QAAQ9hC,EAAEwqB,IAAIiX,GAAGE,GAAGhiC,EAAE,GAAG,iBAAiBK,EAAE,OAAOuD,EAAEw+B,eAAe/hC,GAAG,IAAIG,EAAEJ,EAAEC,EAAEwqB,KAAK,OAAOlqB,OAAOC,KAAKP,EAAE+6B,YAAY,IAAI95B,SAAQ,SAAStB,GAAGQ,EAAE05B,aAAal6B,EAAEK,EAAE+6B,WAAWp7B,GAAG,KAAIK,EAAE6mB,UAAU,IAAI5lB,SAAQ,SAAStB,GAAGQ,EAAE6hC,YAAYpiC,EAAED,EAAE,CAACmiC,KAAK/hC,IAAI,IAAGI,CAAC,CAAtU,CAAwUR,GAAGC,EAAE,IAAG,OAAOA,EAAEw3B,aAAapV,IAAIoW,EAAEL,oBAAoBp4B,EAAE4D,EAAE0+B,eAAetiC,EAAE,IAAIqE,QAAQrE,EAAEC,GAAGsiC,UAAU,KAAKviC,EAAE,GAAGqE,OAAOrE,EAAE,mCAAmCC,EAAEiiC,WAAWM,aAAaxiC,EAAEC,IAAIA,EAAEwiC,SAAS,EAAEC,KAAK,SAAS1iC,GAAG,IAAIC,EAAED,EAAE,GAAGK,EAAEL,EAAE,GAAG,IAAIs5B,GAAGr5B,GAAGmC,QAAQq2B,EAAEZ,kBAAkB,OAAOoK,GAAGvI,QAAQ15B,GAAG,IAAII,EAAE,IAAI25B,OAAO,GAAG11B,OAAOo0B,EAAEb,UAAU,eAAev3B,EAAE,GAAG+6B,WAAWsF,GAAGrgC,EAAE,GAAG+6B,WAAWkF,QAAQtgC,EAAEK,EAAE,GAAG+6B,WAAWkF,MAAM9G,MAAM,KAAK7zB,QAAO,SAAS3F,EAAEC,GAAG,OAAOA,IAAIw4B,EAAEZ,kBAAkB53B,EAAEi/B,MAAM9+B,GAAGJ,EAAE2iC,MAAM3iC,EAAE4iC,QAAQ3hC,KAAKhB,GAAGD,CAAC,GAAE,CAAC4iC,OAAO,GAAGD,MAAM,KAAKtiC,EAAE,GAAG+6B,WAAWkF,MAAMtgC,EAAE2iC,MAAMtH,KAAK,KAAK,IAAIr7B,EAAE4iC,OAAOvhC,OAAOpB,EAAE4iC,gBAAgB,SAAS5iC,EAAEi6B,aAAa,QAAQl6B,EAAE4iC,OAAOvH,KAAK,OAAOh7B,EAAEA,EAAEkE,IAAI42B,IAAIE,KAAK,MAAMp7B,EAAEi6B,aAAa7X,EAAE,IAAIpiB,EAAEk6B,UAAU95B,CAAC,GAAG,SAASyiC,GAAG9iC,GAAGA,GAAG,CAAC,SAAS+iC,GAAG1iC,EAAEL,GAAG,IAAII,EAAE,mBAAmBJ,EAAEA,EAAE4hC,GAAG,IAAIvhC,EAAEgB,OAAOjB,KAAKq4B,EAAEN,iBAAiB/C,GAAEvyB,EAAEmgC,uBAA0BF,KAAI,WAAW,IAAI9iC,GAAE,IAAKy4B,EAAEX,gBAAgBmK,GAAGxJ,EAAEX,iBAAiBmK,GAAGvI,QAAQz5B,EAAEwhC,GAAGC,MAAM,UAAUrhC,EAAEkE,IAAIvE,GAAGC,IAAIG,GAAG,GAAE,CAAC,IAAI6iC,IAAG,EAAG,SAASC,KAAKD,IAAG,CAAE,CAAC,SAASE,KAAKF,IAAG,CAAE,CAAC,IAAIG,GAAG,KAAK,SAASC,GAAGrjC,GAAG,IAAIS,EAAEC,EAAET,EAAEkB,EAAEuB,GAAG+1B,EAAEP,mBAAmBj4B,EAAED,EAAEsjC,aAAa7iC,OAAE,IAASR,EAAE2hC,GAAG3hC,EAAEA,EAAED,EAAEujC,aAAa7iC,OAAE,IAAST,EAAE2hC,GAAG3hC,EAAEA,EAAED,EAAEwjC,uBAAuBriC,OAAE,IAASlB,EAAE2hC,GAAG3hC,EAAED,OAAE,KAAUA,EAAEA,EAAEyjC,sBAAsB7/B,EAAE5D,EAAEojC,GAAG,IAAI1gC,GAAE,SAAS1C,GAAG,IAAIQ,EAAEyiC,KAAKziC,EAAE87B,GAAGjD,GAAGr5B,GAAGsB,SAAQ,SAAStB,GAAG,IAAIC,EAAEI,EAAED,EAAE,cAAcJ,EAAE0jC,MAAM,EAAE1jC,EAAE2jC,WAAWtiC,SAASwgC,GAAG7hC,EAAE2jC,WAAW,MAAMlL,EAAER,sBAAsB92B,EAAEnB,EAAE4jC,QAAQnjC,EAAET,EAAE4jC,SAAS,eAAe5jC,EAAE0jC,MAAM1jC,EAAE4jC,OAAO1B,YAAYzJ,EAAER,sBAAsB92B,EAAEnB,EAAE4jC,OAAO1B,YAAY,eAAeliC,EAAE0jC,MAAM7B,GAAG7hC,EAAE4jC,UAAU7M,EAAE30B,QAAQpC,EAAE6jC,iBAAiB,UAAU7jC,EAAE6jC,gBAA2BzjC,GAAXC,EAAEL,EAAE4jC,QAAWnM,aAAap3B,EAAEo3B,aAAa3H,GAAG,KAAKzvB,EAAEA,EAAEo3B,aAAap3B,EAAEo3B,aAAaxB,GAAG,KAAK71B,GAAGC,IAAIA,GAAGJ,EAAEs9B,GAAGjE,GAAGt5B,EAAE4jC,UAAUrI,OAAOt7B,EAAEA,EAAE4F,SAAS7F,EAAE4jC,OAAO1J,aAAapK,EAAEzvB,GAAGG,GAAGP,GAAGD,EAAE4jC,OAAO1J,aAAajE,EAAEh2B,KAAKA,EAAED,EAAE4jC,SAAS3jC,EAAEs5B,WAAWt5B,EAAEs5B,UAAUuK,UAAU7jC,EAAEs5B,UAAUuK,SAASrL,EAAEZ,mBAAmBn3B,EAAEV,EAAE4jC,QAAQ,IAAG,IAAGn+B,GAAG29B,GAAGW,QAAQ/jC,EAAE,CAACgkC,WAAU,EAAG5I,YAAW,EAAG6I,eAAc,EAAGC,SAAQ,IAAK,CAAic,SAASC,GAAGnkC,EAAEC,GAAG,IAAII,EAAE,EAAEe,UAAUC,aAAQ,IAASpB,EAAEA,EAAE,CAACmkC,aAAY,GAAIhkC,EAAzgB,SAAYJ,GAAG,IAAIC,EAAED,EAAEy3B,aAAa,eAAep3B,EAAEL,EAAEy3B,aAAa,aAAar3B,OAAE,IAASJ,EAAEqkC,UAAUrkC,EAAEqkC,UAAUzK,OAAO,GAAGp5B,EAAE+8B,GAAGjE,GAAGt5B,IAAI,OAAOQ,EAAE+6B,SAAS/6B,EAAE+6B,OAAOe,IAAIr8B,GAAGI,IAAIG,EAAE+6B,OAAOt7B,EAAEO,EAAEqF,SAASxF,GAAGG,EAAEqF,UAAUrF,EAAE+6B,SAAS/6B,EAAE+6B,QAAQ,EAAEn7B,EAAEiB,SAASb,EAAEqF,UAAUxF,EAAEG,EAAE+6B,OAAOn7B,EAAEJ,EAAEqkC,WAAW7H,GAAGn8B,IAAI,CAAC,GAAGD,IAAI+8B,GAAG38B,EAAE+6B,OAAOO,GAAG97B,EAAEqkC,eAAe7jC,EAAEqF,UAAU4yB,EAAEqE,cAAc98B,EAAEskC,YAAYtkC,EAAEskC,WAAWC,WAAWC,KAAKC,YAAYjkC,EAAEqF,SAAS7F,EAAEskC,WAAWI,OAAOlkC,CAAC,CAA4EmkC,CAAG3kC,GAAGQ,EAAEJ,EAAEyF,SAASpF,EAAEL,EAAEm7B,OAAO76B,EAAEN,EAAEw9B,KAAKz8B,GAAGlB,EAAEo5B,IAAIl4B,EAAEnB,GAAGo7B,YAAYz1B,QAAO,SAAS3F,EAAEC,GAAG,MAAM,UAAUD,EAAEmF,MAAM,UAAUnF,EAAEmF,OAAOnF,EAAEC,EAAEkF,MAAMlF,EAAE0B,OAAO3B,CAAC,GAAE,CAAC,GAAGI,EAAEe,EAAEs2B,aAAa,SAASt2B,EAAEA,EAAEs2B,aAAa,oBAAoBgB,EAAET,WAAW53B,EAAEH,EAAE,mBAAmB,GAAGoE,OAAOo0B,EAAEZ,iBAAiB,WAAWxzB,OAAOlD,GAAG+3B,MAAOj5B,EAAE,eAAe,OAAOA,EAAE2kC,UAAU,UAAU3kC,GAAoCsB,GAAjCtB,EAAEk+B,GAAG,sBAAsB,CAAC,EAAEn+B,GAAKK,EAAE+jC,aAAa/jC,GAAGkB,EAAEvB,GAAGy3B,aAAa,SAASl2B,EAAE,GAAGA,EAAElB,EAAEA,EAAEm5B,MAAM,KAAK7zB,QAAO,SAAS3F,EAAEC,GAAsBA,GAAfI,EAAEJ,EAAEu5B,MAAM,MAAS,GAAvB,IAA0Bn5B,EAAEA,EAAE4E,MAAM,GAAG,OAAOhF,GAAG,EAAEI,EAAEgB,SAASrB,EAAEC,GAAGI,EAAEg7B,KAAK,KAAKzB,QAAQ55B,CAAC,GAAE,CAAC,GAAGuB,GAAG,IAAG,OAAOmE,EAAE,CAACG,SAASrF,EAAEq/B,MAAM7/B,EAAEy3B,aAAa,SAASsI,QAAQ//B,EAAEy3B,aAAa,oBAAoB8D,OAAO96B,EAAEk/B,UAAU9G,EAAEjE,KAAK,CAAC/uB,SAAS,KAAK01B,OAAO,KAAKqC,KAAK,IAAIkC,OAAO,KAAKF,QAAO,EAAGI,MAAM,CAACK,QAAQ3/B,EAAE4E,OAAO/D,EAAE65B,WAAWj6B,IAAIlB,EAAE,CAAC,IAAI4kC,GAAG/J,GAAGx1B,OAAO,SAASw/B,GAAG9kC,GAAG,IAAIC,EAAE,SAASw4B,EAAEX,eAAeqM,GAAGnkC,EAAE,CAACokC,aAAY,IAAKD,GAAGnkC,GAAG,OAAOC,EAAE+/B,MAAMK,QAAQj+B,QAAQw0B,GAAGyH,GAAG,qBAAqBr+B,EAAEC,GAAGo+B,GAAG,iCAAiCr+B,EAAEC,EAAE,CAAC,IAAI8kC,GAAG,IAAIzgC,IAAI,SAAS0gC,GAAGhlC,GAAG,IAAII,EAAE,EAAEgB,UAAUC,aAAQ,IAASD,UAAU,GAAGA,UAAU,GAAG,KAAK,IAAIqE,EAAE,OAAOm5B,QAAQqG,UAAU,SAASzkC,EAAER,GAAG,OAAOC,EAAEuE,IAAI,GAAGH,OAAO6xB,EAAE,KAAK7xB,OAAOrE,GAAG,CAAC,SAASS,EAAET,GAAG,OAAOC,EAAEwiC,OAAO,GAAGp+B,OAAO6xB,EAAE,KAAK7xB,OAAOrE,GAAG,CAAC,IAAIC,EAAE2D,EAAE5B,gBAAgBu3B,UAAUl5B,EAAEo4B,EAAEqE,aAAaiI,GAAG1O,EAAE9xB,KAAI,SAASvE,GAAG,MAAM,MAAMqE,OAAOrE,EAAE,IAAGqE,OAAO1D,OAAOC,KAAKikC,KAAKxkC,EAAEo9B,SAAS,OAAOp9B,EAAEY,KAAK,MAAM,IAAIP,EAAE,CAAC,IAAI2D,OAAOuyB,EAAE,UAAUvyB,OAAOge,EAAE,OAAOhe,OAAOhE,EAAEkE,KAAI,SAASvE,GAAG,MAAM,IAAIqE,OAAOrE,EAAE,UAAUqE,OAAOge,EAAE,KAAK,KAAIgZ,KAAK,MAAM,GAAG,IAAI36B,EAAEW,OAAO,OAAOu9B,QAAQqG,UAAU5kC,EAAE,GAAG,IAAIA,EAAEg5B,GAAGr5B,EAAEklC,iBAAiBxkC,GAAG,CAAC,MAAMV,GAAG,CAAC,KAAK,EAAEK,EAAEgB,QAAQ,OAAOu9B,QAAQqG,UAAUzkC,EAAE,WAAWC,EAAE,YAAY,IAAIU,EAAEsgC,GAAGC,MAAM,UAAUngC,EAAElB,EAAEsF,QAAO,SAAS3F,EAAEC,GAAG,IAAI,IAAII,EAAEykC,GAAG7kC,GAAGI,GAAGL,EAAEiB,KAAKZ,EAAE,CAAC,MAAML,GAAGm2B,GAAG,gBAAgBn2B,EAAEmF,MAAMg8B,QAAQC,MAAMphC,EAAE,CAAC,OAAOA,CAAC,GAAE,IAAI,OAAO,IAAI4+B,SAAQ,SAAS3+B,EAAEI,GAAGu+B,QAAQuG,IAAI5jC,GAAG6jC,MAAK,SAASplC,GAAG+iC,GAAG/iC,GAAE,WAAWQ,EAAE,UAAUA,EAAE,YAAYC,EAAE,WAAW,mBAAmBL,GAAGA,IAAIe,IAAIlB,GAAG,GAAE,IAAGolC,OAAM,SAASrlC,GAAGmB,IAAId,EAAEL,EAAE,GAAE,GAAE,CAAC,SAASslC,GAAGtlC,GAAG,IAAIC,EAAE,EAAEmB,UAAUC,aAAQ,IAASD,UAAU,GAAGA,UAAU,GAAG,KAAK0jC,GAAG9kC,GAAGolC,MAAK,SAASplC,GAAGA,GAAG+iC,GAAG,CAAC/iC,GAAGC,EAAE,GAAE,CAA2H,SAASslC,GAAGvlC,GAAG,IAAIC,EAAE,EAAEmB,UAAUC,aAAQ,IAASD,UAAU,GAAGA,UAAU,GAAG,CAAC,EAAEf,EAAEJ,EAAE0/B,UAAUv/B,OAAE,IAASC,EAAEw4B,EAAEx4B,EAAEG,OAAE,KAAUH,EAAEJ,EAAE2/B,SAASv/B,EAAEI,OAAE,KAAUJ,EAAEJ,EAAE20B,MAAM,KAAKv0B,EAAEK,OAAE,KAAUL,EAAEJ,EAAE6/B,QAAQ,KAAKz/B,EAAEc,OAAE,KAAUd,EAAEJ,EAAE4/B,OAAO,KAAKx/B,EAAEkB,OAAE,KAAUlB,EAAEJ,EAAE8/B,SAAS,KAAK1/B,EAAEyB,OAAE,KAAUzB,EAAEJ,EAAEogC,SAAS,GAAGhgC,EAAEgC,OAAE,KAAUhC,EAAEJ,EAAEm7B,YAAY,CAAC,EAAE/6B,EAAEiC,OAAE,KAAUjC,EAAEJ,EAAEqF,QAAQ,CAAC,EAAEjF,EAAE,GAAGL,EAAE,CAAC,IAAIuC,EAAEvC,EAAEu7B,OAAO/4B,EAAExC,EAAE6F,SAASpD,EAAEzC,EAAE4F,KAAK,OAAO05B,GAAG55B,EAAE,CAACg+B,KAAK,QAAQ1jC,IAAG,WAAW,OAAOo+B,GAAG,2BAA2B,CAACoH,eAAexlC,EAAEylC,OAAOxlC,IAAIw4B,EAAET,WAAW72B,EAAEkB,EAAE,mBAAmB,GAAGgC,OAAOo0B,EAAEZ,iBAAiB,WAAWxzB,OAAO9C,GAAG23B,MAAO72B,EAAE,eAAe,OAAOA,EAAEuiC,UAAU,UAAUnF,GAAG,CAACrhB,MAAM,CAACshB,KAAKsB,GAAGv+B,GAAGmyB,KAAKn0B,EAAEugC,GAAGvgC,EAAEmF,MAAM,CAACs6B,OAAM,EAAGC,MAAM,KAAKC,OAAO,KAAKx6B,KAAK,CAAC,IAAI21B,OAAOh5B,EAAEsD,SAASrD,EAAEm9B,UAAUj6B,EAAEA,EAAE,CAAC,EAAEmzB,GAAGz4B,GAAGw/B,OAAOp/B,EAAEq/B,MAAM1+B,EAAE2+B,OAAOp/B,EAAEq/B,QAAQx+B,EAAEy+B,MAAM,CAAC5E,WAAW/4B,EAAEiD,OAAOhD,EAAE+9B,QAAQv+B,IAAI,GAAE,CAAC,CAAv8Bu0B,EAAE9xB,KAAI,SAASvE,GAAG+kC,GAAGvgC,IAAI,MAAMH,OAAOrE,GAAG,IAAGW,OAAOC,KAAK21B,EAAEH,IAAI7xB,IAAIwgC,GAAGvgC,IAAIC,KAAKsgC,KAAKpkC,OAAOC,KAAK21B,EAAE9L,IAAIlmB,IAAIwgC,GAAGvgC,IAAIC,KAAKsgC,KAAu1BA,GAAG1iC,EAAE0iC,IAAIjjC,EAAE,CAAC04B,OAAO,WAAW,MAAM,CAAC50B,MAAMxF,EAAEmlC,GAAG,SAASvlC,GAAG,IAAIC,EAAE,EAAEmB,UAAUC,aAAQ,IAASD,UAAU,GAAGA,UAAU,GAAG,CAAC,EAAEf,GAAGL,GAAG,CAAC,GAAG4F,KAAK5F,EAAEs+B,GAAGt+B,GAAG,CAAC,GAA4C,OAAzCA,GAAGA,EAAEC,EAAE20B,SAAS50B,GAAG,CAAC,GAAG4F,KAAK5F,EAAEs+B,GAAGt+B,GAAG,CAAC,IAAWI,EAAEC,EAAEqF,EAAEA,EAAE,CAAC,EAAEzF,GAAG,CAAC,EAAE,CAAC20B,KAAK50B,IAAI,IAAI,IAAII,CAAC,EAAEmF,MAAM,WAAW,MAAM,CAACmgC,0BAA0B,SAAS1lC,GAAG,OAAOA,EAAEsjC,aAAa0B,GAAGhlC,EAAEujC,aAAa+B,GAAGtlC,CAAC,EAAE,EAAE2lC,SAAS,SAAS3lC,GAAGA,EAAE2+B,MAAM,SAAS3+B,GAAG,IAAIC,EAAED,EAAEqK,KAAkB,OAAbrK,EAAEA,EAAE4lC,SAAgBZ,QAAG,IAAS/kC,EAAE2D,EAAE3D,OAAE,IAASD,EAAE,WAAW,EAAEA,EAAE,EAAEA,EAAE6lC,+BAA+B,SAASzlC,EAAEJ,GAAG,IAAIQ,EAAER,EAAE6F,SAASpF,EAAET,EAAE6/B,MAAMn/B,EAAEV,EAAE+/B,QAAQ5+B,EAAEnB,EAAEu7B,OAAOh6B,EAAEvB,EAAE2/B,UAAU79B,EAAE9B,EAAE4/B,OAAO3/B,EAAED,EAAE40B,KAAKvyB,EAAErC,EAAE8/B,OAAOx9B,EAAEtC,EAAEggC,MAAM,OAAO,IAAIpB,SAAQ,SAASv+B,EAAEL,GAAG4+B,QAAQuG,IAAI,CAACjE,GAAG1gC,EAAEW,GAAGlB,EAAE4F,SAASq7B,GAAGjhC,EAAE4F,SAAS5F,EAAEs7B,QAAQqD,QAAQqG,QAAQ,CAAC/E,OAAM,EAAGC,MAAM,IAAIC,OAAO,IAAIx6B,KAAK,CAAC,MAAMw/B,MAAK,SAASplC,GAAgBA,GAATC,EAAEsC,EAAEvC,EAAE,IAAO,GAAjB,IAAoBC,EAAEA,EAAE,GAAGI,EAAE,CAACD,EAAEq/B,GAAG,CAACrhB,MAAM,CAACshB,KAAK1/B,EAAE40B,KAAK30B,GAAGs7B,OAAOp6B,EAAE0E,SAASrF,EAAEm/B,UAAUp+B,EAAEq+B,OAAO99B,EAAEg+B,OAAOz9B,EAAEw9B,MAAMp/B,EAAEs/B,QAAQr/B,EAAEs/B,MAAM19B,EAAE29B,WAAU,KAAM,IAAGoF,MAAMrlC,EAAE,GAAE,EAAEA,EAAE8lC,qBAAqB,SAAS9lC,GAAG,IAAIC,EAAEI,EAAEL,EAAEknB,SAAS9mB,EAAEJ,EAAEo7B,WAAW56B,EAAER,EAAE0/B,KAAKj/B,EAAET,EAAE2/B,UAAyB,OAAO,GAAtB3/B,EAAE25B,GAAG35B,EAAEsF,SAAmBjE,SAASjB,EAAEugC,MAAM3gC,GAAG65B,GAAGp5B,KAAKR,EAAEo+B,GAAG,oCAAoC,CAACqB,KAAKl/B,EAAEm/B,UAAUl/B,EAAEslC,eAAevlC,EAAE2/B,MAAM6F,UAAUxlC,EAAE2/B,SAAS9/B,EAAEY,KAAKhB,GAAGO,EAAEoF,MAAM,CAACshB,SAAS7mB,EAAE+6B,WAAWh7B,EAAE,CAAC,GAAGH,EAAE,CAACu6B,OAAO,WAAW,MAAM,CAACyL,MAAM,SAASjmC,GAAG,IAAIK,EAAE,EAAEe,UAAUC,aAAQ,IAASD,UAAU,GAAGA,UAAU,GAAG,CAAC,EAAEnB,EAAEI,EAAEggC,QAAQjgC,OAAE,IAASH,EAAE,GAAGA,EAAE,OAAOq/B,GAAG,CAACoE,KAAK,UAAS,WAAWtF,GAAG,2BAA2B,CAAC8H,UAAUlmC,EAAEylC,OAAOplC,IAAI,IAAIJ,EAAE,GAAG,OAAOD,GAAE,SAASA,GAAG+B,MAAM2C,QAAQ1E,GAAGA,EAAEuE,KAAI,SAASvE,GAAGC,EAAEA,EAAEoE,OAAOrE,EAAEu/B,SAAS,IAAGt/B,EAAEA,EAAEoE,OAAOrE,EAAEu/B,SAAS,IAAG,CAAC,CAAC1U,IAAI,OAAOuQ,WAAW,CAACkF,MAAM,CAAC,GAAGj8B,OAAOo0B,EAAEb,UAAU,YAAYvzB,OAAOhC,EAAEjC,IAAIi7B,KAAK,MAAMnU,SAASjnB,GAAG,GAAE,EAAE,GAAGqC,EAAE,CAACk4B,OAAO,WAAW,MAAM,CAAC2L,QAAQ,SAAS3lC,GAAG,IAA6JR,EAAzJS,EAAE,EAAEW,UAAUC,aAAQ,IAASD,UAAU,GAAGA,UAAU,GAAG,CAAC,EAAYV,OAAE,KAAZV,EAAES,EAAEo/B,OAAmB,KAAK7/B,EAAcmB,OAAE,KAAdnB,EAAES,EAAE4/B,SAAqB,GAAGrgC,EAAiBuB,OAAE,KAAjBvB,EAAES,EAAE26B,YAAwB,CAAC,EAAEp7B,EAAa8B,OAAE,KAAb9B,EAAES,EAAE6E,QAAoB,CAAC,EAAEtF,EAAE,OAAOs/B,GAAG,CAACoE,KAAK,UAAU7C,QAAQrgC,IAAG,WAAW,OAAO49B,GAAG,2BAA2B,CAACyC,QAAQrgC,EAAEilC,OAAOhlC,IAAmIR,GAA/HD,EAAE,CAAC6gC,QAAQrgC,EAAEuE,WAAW86B,MAAMn/B,EAAEs/B,MAAM,CAAC5E,WAAW75B,EAAE+D,OAAOxD,EAAEu+B,QAAQ,CAAC,GAAGh8B,OAAOo0B,EAAEb,UAAU,oBAAoBvzB,OAAOhC,EAAElB,OAAU0/B,QAAQxgC,EAAEL,EAAE6/B,MAAgB7/B,EAAE0F,EAAEA,EAAEA,EAAE,CAAC,GAAnBtF,EAAEJ,EAAEggC,OAAmB5E,YAAY/6B,EAAE,CAACw/B,MAAMx/B,GAAG,CAAC,GAAG,CAAC,EAAE,CAACigC,MAAMlgC,EAAEigC,QAAQhF,KAAK,OAAO,GAAGj7B,EAAEu5B,GAAGv5B,EAAEkF,SAASjE,SAASrB,EAAE2gC,MAAMvgC,IAAIA,EAAE,IAAIa,KAAK,CAAC4pB,IAAI,OAAOuQ,WAAWp7B,EAAEknB,SAAS,CAACjnB,KAAKI,GAAGD,EAAEa,KAAK,CAAC4pB,IAAI,OAAOuQ,WAAW,CAACkF,MAAM,WAAWpZ,SAAS,CAAC7mB,KAAKD,EAAE,IAAIJ,EAAEC,EAAEI,EAAED,CAAC,GAAE,EAAE,GAAGoC,EAAE,CAACg4B,OAAO,WAAW,MAAM,CAAC4L,KAAK,SAASpmC,GAAG,IAA4LK,EAAxLJ,EAAE,EAAEmB,UAAUC,aAAQ,IAASD,UAAU,GAAGA,UAAU,GAAG,CAAC,EAAgBhB,OAAE,KAAhBC,EAAEJ,EAAE0/B,WAAuB9G,EAAEx4B,EAAYG,OAAE,KAAZH,EAAEJ,EAAE4/B,OAAmB,KAAKx/B,EAAcI,OAAE,KAAdJ,EAAEJ,EAAEogC,SAAqB,GAAGhgC,EAAiBK,OAAE,KAAjBL,EAAEJ,EAAEm7B,YAAwB,CAAC,EAAE/6B,EAAac,OAAE,KAAbd,EAAEJ,EAAEqF,QAAoB,CAAC,EAAEjF,EAAE,OAAOi/B,GAAG,CAACoE,KAAK,OAAO7C,QAAQ7gC,IAAG,WAAW,OAAOo+B,GAAG,2BAA2B,CAACyC,QAAQ7gC,EAAEylC,OAAOxlC,IAAI2gC,GAAG,CAACC,QAAQ7gC,EAAE2/B,UAAUj6B,EAAEA,EAAE,CAAC,EAAEmzB,GAAGz4B,GAAGy/B,MAAMr/B,EAAEw/B,MAAM,CAAC5E,WAAW16B,EAAE4E,OAAOnE,EAAEk/B,QAAQ,CAAC,GAAGh8B,OAAOo0B,EAAEb,UAAU,iBAAiBvzB,OAAOhC,EAAE5B,MAAM,GAAE,EAAE,EAAEklC,SAAS,SAAS3lC,GAAGA,EAAEqmC,mBAAmB,SAASrmC,EAAEC,GAAG,IAAII,EAAED,EAAEH,EAAE4/B,MAAMr/B,EAAEP,EAAE0/B,UAAUl/B,EAAER,EAAE+/B,MAAMt/B,EAAE,KAAKS,EAAE,KAAK,OAAO2E,IAAIzF,EAAEimC,SAASC,iBAAiBvmC,GAAGwmC,SAAS,IAAI9lC,GAAGT,EAAED,EAAEymC,yBAAyBtG,MAAM9/B,EAAEc,EAAElB,EAAEmgC,OAAO//B,GAAGo4B,EAAET,WAAW53B,IAAIK,EAAE26B,WAAW,eAAe,QAAQwD,QAAQqG,QAAQ,CAACjlC,EAAE4gC,GAAG,CAACC,QAAQ7gC,EAAEm6B,UAAUgG,MAAMz/B,EAAE0/B,OAAOj/B,EAAEw+B,UAAUn/B,EAAEq/B,MAAMz/B,EAAE4/B,MAAMv/B,EAAEw/B,WAAU,KAAM,CAAC,GAA7wG,IAAgxGyG,GAAG,IAAI3M,OAAO,IAAI,MAAM4M,GAAG,CAAC,QAAQ,SAAS,SAASC,GAAGnkC,EAAEC,GAAG,IAAIiB,EAAE,GAAGU,OAAO0a,GAAG1a,OAAO3B,EAAEg3B,QAAQ,IAAI,MAAM,OAAO,IAAIkF,SAAQ,SAASv+B,EAAEL,GAAG,GAAG,OAAOyC,EAAEg1B,aAAa9zB,GAAG,OAAOtD,IAAI,IAAID,EAAEI,EAAEC,EAAEC,EAAET,EAAEkB,EAAEI,EAAEO,EAAEu3B,GAAG52B,EAAEykB,UAAUpmB,QAAO,SAASd,GAAG,OAAOA,EAAEy3B,aAAatY,KAAKzc,CAAC,IAAG,GAAGL,EAAEQ,EAAE0jC,iBAAiB9jC,EAAEC,GAAGJ,EAAED,EAAEwkC,iBAAiB,eAAe3H,MAAMrI,GAAGt0B,EAAEF,EAAEwkC,iBAAiB,eAAerkC,EAAEH,EAAEwkC,iBAAiB,WAAW,GAAG/kC,IAAIQ,EAAE,OAAOG,EAAEqkC,YAAYhlC,GAAGzB,IAAIiC,GAAG,SAASE,GAAG,KAAKA,GAAGrB,EAAEkB,EAAEwkC,iBAAiB,WAAW5mC,GAAG,CAAC,SAASmC,QAAQE,EAAE,IAAImoB,EAAE2L,EAAEh2B,GAAG,CAAC,QAAQ,UAAU,QAAQ,OAAO,UAAU,SAAS,OAAOgC,QAAQE,EAAE,IAAIk0B,EAAEv2B,GAAGqC,EAAE,GAAGykC,eAAejQ,EAAE72B,GAAGsC,GAAGC,GAAMH,GAAGG,EAAErB,GAAGu4B,QAAQgN,GAAG,IAAIzmC,EAAE,EAAEkB,GAAGoB,EAAEF,GAAGhB,OAAOE,EAAE,QAAQiB,EAAED,EAAEw5B,WAAW97B,KAAKuC,GAAG,OAAOvC,EAAE,EAAEkB,GAAG,QAAQI,EAAEgB,EAAEw5B,WAAW97B,EAAE,KAAKsB,GAAG,MAAM,MAAMiB,EAAE,OAAOjB,EAAE,MAAM,MAAMiB,EAAEA,EAAEmkC,GAAG,IAAIplC,GAAGA,GAAGolC,GAAG,GAApLtkC,EAAuL,CAACV,MAAMm6B,IAAIv6B,EAAE,IAAIc,EAAEhB,QAAQgB,EAAE,KAAKA,EAAE,IAAIA,EAAE,GAAGA,GAAG2kC,YAAYxkC,GAAGjB,IAAKI,MAAMJ,EAAEc,EAAE2kC,YAAY3kC,EAAEC,EAAE,GAAG2kC,WAAW,eAAe3kC,EAAE66B,GAAG/8B,EAAEoC,GAAGhC,EAAE8B,EAAED,IAAIG,EAAEk6B,GAAGr6B,EAAEG,GAAGH,EAAE86B,GAAG,MAAM96B,IAAIA,EAAEG,IAAIH,EAAE,CAACk5B,OAAO,MAAM11B,SAASxD,GAAG,OAAO,CAACk5B,OAAO,KAAK11B,SAAS,OAAOA,UAAUxD,EAAEk5B,SAASj5B,EAAED,EAAEwD,SAASzF,EAAEiC,EAAEk5B,UAAUj5B,GAAGf,GAAGO,GAAGA,EAAE21B,aAAa3H,KAAK1vB,GAAG0B,EAAE21B,aAAaxB,KAAKz1B,EAAEH,KAAKoC,EAAEy3B,aAAav2B,EAAEnD,GAAGsB,GAAGW,EAAEqkC,YAAYhlC,IAAIpB,GAAGD,EAAE,CAACoF,SAAS,KAAKg6B,MAAM,KAAKE,QAAQ,KAAKxE,OAAO,KAAKoE,UAAU9G,EAAE+G,QAAO,EAAGhL,KAAK,CAAC/uB,SAAS,KAAK01B,OAAO,KAAKqC,KAAK,IAAIkC,OAAO,KAAKE,MAAM,CAACK,QAAQ,GAAG/6B,OAAO,CAAC,EAAE81B,WAAW,CAAC,KAAK4E,OAAO5E,WAAWjc,GAAGzc,EAAEw+B,GAAG5+B,EAAElC,GAAGglC,MAAK,SAASplC,GAAG,IAAIC,EAAEw/B,GAAG/5B,EAAEA,EAAE,CAAC,EAAEjF,GAAG,CAAC,EAAE,CAAC2d,MAAM,CAACshB,KAAK1/B,EAAE40B,KAA3imB,CAAC2G,OAAO,KAAK11B,SAAS,KAAK+3B,KAAK,KAAshmBrC,OAAOn7B,EAAEyF,SAASrF,EAAEw/B,MAAMt/B,EAAEu/B,WAAU,KAAMjgC,EAAE4D,EAAEm+B,gBAAgB,6BAA6B,OAAO,aAAar/B,EAAED,EAAE83B,aAAav6B,EAAEyC,EAAE6hC,YAAY7hC,EAAE4/B,YAAYriC,GAAGA,EAAEuiC,UAAUtiC,EAAEsE,IAAI42B,IAAIE,KAAK,MAAM54B,EAAEogC,gBAAgBl/B,GAAGtD,GAAG,IAAGglC,MAAMrlC,KAAKK,GAAG,GAAE,CAAC,SAAS6mC,GAAGlnC,GAAG,OAAO4+B,QAAQuG,IAAI,CAACyB,GAAG5mC,EAAE,YAAY4mC,GAAG5mC,EAAE,YAAY,CAAC,SAASmnC,GAAGnnC,GAAG,QAAQA,EAAEkiC,aAAa/hC,SAAS8B,OAAOmgB,EAAEhgB,QAAQpC,EAAEq6B,QAAQC,gBAAgBt6B,EAAEy3B,aAAatY,IAAInf,EAAEkiC,YAAY,QAAQliC,EAAEkiC,WAAW7H,QAAQ,CAAC,SAAS+M,GAAG5mC,GAAG,GAAGiF,EAAE,OAAO,IAAIm5B,SAAQ,SAAS5+B,EAAEC,GAAG,IAAII,EAAEg5B,GAAG74B,EAAE0kC,iBAAiB,MAAMpkC,OAAOqmC,IAAI5iC,IAAI2iC,IAAI9mC,EAAEqhC,GAAGC,MAAM,wBAAwBwB,KAAKtE,QAAQuG,IAAI9kC,GAAG+kC,MAAK,WAAWhlC,IAAI+iC,KAAKnjC,GAAG,IAAGqlC,OAAM,WAAWjlC,IAAI+iC,KAAKljC,GAAG,GAAE,GAAE,CAAC,SAASonC,GAAGrnC,GAAG,OAAOA,EAAE+mC,cAAcvN,MAAM,KAAK7zB,QAAO,SAAS3F,EAAEC,GAAG,IAAII,EAAEJ,EAAE8mC,cAAcvN,MAAM,KAAYp5B,GAAPH,EAAEI,EAAE,GAAKA,EAAE4E,MAAM,GAAGo2B,KAAK,MAAK,GAAGp7B,GAAG,MAAMG,EAAE,OAAOJ,EAAE+4B,OAAM,EAAG/4B,EAAE,GAAGC,GAAG,MAAMG,EAAE,OAAOJ,EAAEg5B,OAAM,EAAGh5B,EAAE,GAAGI,EAAEknC,WAAWlnC,GAAGmnC,MAAMnnC,GAAG,OAAOJ,EAAE,OAAOC,GAAG,IAAI,OAAOD,EAAE84B,KAAK94B,EAAE84B,KAAK14B,EAAE,MAAM,IAAI,SAASJ,EAAE84B,KAAK94B,EAAE84B,KAAK14B,EAAE,MAAM,IAAI,OAAOJ,EAAEo1B,EAAEp1B,EAAEo1B,EAAEh1B,EAAE,MAAM,IAAI,QAAQJ,EAAEo1B,EAAEp1B,EAAEo1B,EAAEh1B,EAAE,MAAM,IAAI,KAAKJ,EAAE8vB,EAAE9vB,EAAE8vB,EAAE1vB,EAAE,MAAM,IAAI,OAAOJ,EAAE8vB,EAAE9vB,EAAE8vB,EAAE1vB,EAAE,MAAM,IAAI,SAASJ,EAAE6xB,OAAO7xB,EAAE6xB,OAAOzxB,EAAE,OAAOJ,CAAC,GAAE,CAAC84B,KAAK,GAAG1D,EAAE,EAAEtF,EAAE,EAAEiJ,OAAM,EAAGC,OAAM,EAAGnH,OAAO,GAAG,CAAC,IAA2M2V,GAAvMC,IAAG,EAAGC,GAAG,CAACtS,EAAE,EAAEtF,EAAE,EAAEqQ,MAAM,OAAOC,OAAO,QAAQ,SAASuH,GAAG3nC,GAAG,OAAOA,EAAEo7B,aAAap7B,EAAEo7B,WAAWpK,QAAS,EAAE5vB,UAAUC,aAAQ,IAASD,UAAU,KAAKA,UAAU,MAAOpB,EAAEo7B,WAAWpK,KAAK,SAAShxB,CAAC,CAAQwnC,GAAchJ,GAAcj9B,EAAE,CAACkB,EAAEX,EAAE7B,EAAEqC,EAAEE,EAAE,CAAC+C,MAAM,WAAW,MAAM,CAACmgC,0BAA0B,SAAS1lC,GAAG,OAAOA,EAAEwjC,uBAAuB4D,GAAGpnC,CAAC,EAAE,EAAE2lC,SAAS,SAAS3lC,GAAGA,EAAE4nC,mBAAmB,SAAS5nC,GAAGA,EAAEA,EAAEqK,KAAKouB,EAAER,sBAAsBmP,QAAG,IAASpnC,EAAE4D,EAAE5D,EAAE,CAAC,GAAG,CAACw6B,OAAO,WAAW,MAAM,CAACC,IAAI,CAACoN,QAAQ,WAAW3E,KAAKuE,IAAG,CAAE,GAAG,EAAEliC,MAAM,WAAW,MAAM,CAAC8K,UAAU,WAAWgzB,GAAGlF,GAAG,4BAA4B,CAAC,GAAG,EAAEM,OAAO,WAAW2E,IAAIA,GAAG0E,YAAY,EAAEhJ,MAAM,SAAS9+B,GAAGA,EAAEA,EAAEyjC,qBAAqBgE,GAAGtE,KAAKE,GAAGlF,GAAG,4BAA4B,CAACsF,qBAAqBzjC,IAAI,EAAE,GAAG,CAACw6B,OAAO,WAAW,MAAM,CAACyE,MAAM,CAACU,UAAU0H,IAAI,EAAE9hC,MAAM,WAAW,MAAM,CAACwiC,oBAAoB,SAAS/nC,EAAEC,GAAyC,OAAtCA,EAAEA,EAAEw3B,aAAa,wBAAgCz3B,EAAE2/B,UAAU0H,GAAGpnC,IAAID,CAAC,EAAE,EAAE2lC,SAAS,SAAS3lC,GAAGA,EAAEgoC,kCAAkC,SAAShoC,GAAG,IAAIC,EAAED,EAAE0/B,KAAKr/B,EAAEL,EAAE2/B,UAAUv/B,EAAEJ,EAAE+lC,eAAevlC,EAAER,EAAEgmC,UAAUvlC,EAAE,CAACk/B,UAAU,aAAat7B,OAAOjE,EAAE,EAAE,UAA0T,OAAhTJ,EAAE,aAAaqE,OAAO,GAAGhE,EAAE+0B,EAAE,MAAM/wB,OAAO,GAAGhE,EAAEyvB,EAAE,MAAM1vB,EAAE,SAASiE,OAAOhE,EAAEy4B,KAAK,IAAIz4B,EAAE04B,OAAO,EAAE,GAAG,MAAM10B,OAAOhE,EAAEy4B,KAAK,IAAIz4B,EAAE24B,OAAO,EAAE,GAAG,MAAM34B,EAAE,UAAUgE,OAAOhE,EAAEwxB,OAAO,SAA4I,CAAChH,IAAI,IAAIuQ,WAAW11B,EAAE,CAAC,GAA1JlF,EAAE,CAACynC,MAAMxnC,EAAEynC,MAAM,CAACvI,UAAU,GAAGt7B,OAAOrE,EAAE,KAAKqE,OAAOjE,EAAE,KAAKiE,OAAOhE,IAAI8nC,KAAK,CAACxI,UAAU,aAAat7B,OAAO7D,EAAE,GAAG,EAAE,aAA6CynC,OAAO/gB,SAAS,CAAC,CAAC2D,IAAI,IAAIuQ,WAAW11B,EAAE,CAAC,EAAElF,EAAE0nC,OAAOhhB,SAAS,CAAC,CAAC2D,IAAI5qB,EAAE2F,KAAKilB,IAAI3D,SAASjnB,EAAE2F,KAAKshB,SAASkU,WAAW11B,EAAEA,EAAE,CAAC,EAAEzF,EAAE2F,KAAKw1B,YAAY56B,EAAE2nC,UAAU,CAAC,GAAG,CAAC5iC,MAAM,WAAW,MAAM,CAACwiC,oBAAoB,SAAS/nC,EAAEC,GAAG,IAAqCI,GAAjCA,EAAEJ,EAAEw3B,aAAa,iBAAoB8F,GAAGl9B,EAAEm5B,MAAM,KAAKj1B,KAAI,SAASvE,GAAG,OAAOA,EAAE45B,MAAM,KAAt8rB,CAAC2B,OAAO,KAAK11B,SAAS,KAAK+3B,KAAK,IAA+6rB,OAAOv9B,EAAEk7B,SAASl7B,EAAEk7B,OAAOe,IAAIt8B,EAAE40B,KAAKv0B,EAAEL,EAAE8/B,OAAO7/B,EAAEw3B,aAAa,mBAAmBz3B,CAAC,EAAE,EAAE2lC,SAAS,SAAS3lC,GAAGA,EAAEooC,qBAAqB,SAASpoC,GAAG,IAAIC,EAAED,EAAEknB,SAAS7mB,EAAEL,EAAEo7B,WAAWh7B,EAAEJ,EAAE0/B,KAAKl/B,EAAER,EAAE40B,KAAKn0B,EAAET,EAAE8/B,OAAOp/B,EAAEV,EAAE2/B,UAAUx+B,EAAEf,EAAE+/B,MAAM5+B,EAAEnB,EAAEwF,KAAK9D,EAAEtB,EAAE2/B,MAAw+B,OAAl+BngC,EAAEQ,EAAEoF,KAAQpF,GAAGJ,EAAE,CAACu/B,UAAUj/B,EAAEqlC,eAAejkC,EAAEkkC,UAAU7kC,IAAIw+B,UAAUj/B,EAAEN,EAAE2lC,eAAejkC,EAAE1B,EAAE4lC,UAAU7kC,EAAE,CAACw+B,UAAU,aAAat7B,OAAO3D,EAAE,EAAE,UAAUN,EAAE,aAAaiE,OAAO,GAAG7D,EAAE40B,EAAE,MAAM/wB,OAAO,GAAG7D,EAAEsvB,EAAE,MAAMpvB,EAAE,SAAS2D,OAAO7D,EAAEs4B,KAAK,IAAIt4B,EAAEu4B,OAAO,EAAE,GAAG,MAAM10B,OAAO7D,EAAEs4B,KAAK,IAAIt4B,EAAEw4B,OAAO,EAAE,GAAG,MAAMx4B,EAAE,UAAU6D,OAAO7D,EAAEqxB,OAAO,SAAxTnxB,EAAiU,CAACunC,MAAM9mC,EAAE+mC,MAAM,CAACvI,UAAU,GAAGt7B,OAAOjE,EAAE,KAAKiE,OAAO3D,EAAE,KAAK2D,OAAO7D,IAAI2nC,KAAK,CAACxI,UAAU,aAAat7B,OAAOvC,EAAE,GAAG,EAAE,YAAatB,EAAE,CAACqqB,IAAI,OAAOuQ,WAAW11B,EAAEA,EAAE,CAAC,EAAEgiC,IAAI,CAAC,EAAE,CAAC1W,KAAK,WAAWlvB,EAAEP,EAAE2lB,SAAS,CAACA,SAAS3lB,EAAE2lB,SAAS3iB,IAAIojC,KAAK,CAAC,EAAE7lC,EAAE,CAAC+oB,IAAI,IAAIuQ,WAAW11B,EAAE,CAAC,EAAEhF,EAAEwnC,OAAOhhB,SAAS,CAACygB,GAAGjiC,EAAE,CAACmlB,IAAItpB,EAAEspB,IAAIuQ,WAAW11B,EAAEA,EAAE,CAAC,EAAEnE,EAAE65B,YAAY16B,EAAEynC,OAAOrmC,MAAMpB,EAAE,CAACmqB,IAAI,IAAIuQ,WAAW11B,EAAE,CAAC,EAAEhF,EAAEunC,OAAO/gB,SAAS,CAACplB,IAAIA,EAAE,QAAQuC,OAAO5D,GAAGy4B,KAAMz4B,EAAE,QAAQ4D,OAAO5D,GAAGy4B,KAAMx4B,EAAE,CAACmqB,IAAI,OAAOuQ,WAAW11B,EAAEA,EAAE,CAAC,EAAEgiC,IAAI,CAAC,EAAE,CAAChH,GAAG5+B,EAAEumC,UAAU,iBAAiBC,iBAAiB,mBAAmBphB,SAAS,CAAC1mB,EAAEE,IAAIA,EAAE,CAACmqB,IAAI,OAAO3D,SAAS,CAAC,CAAC2D,IAAI,WAAWuQ,WAAW,CAACsF,GAAGjgC,GAAGymB,SAAS,MAASlnB,EAAG6qB,IAAI7qB,EAAEknB,SAAS,CAAClnB,IAAIU,IAAWT,EAAEgB,KAAKP,EAAE,CAACmqB,IAAI,OAAOuQ,WAAW11B,EAAE,CAACsrB,KAAK,eAAe,YAAY,QAAQ3sB,OAAO5D,EAAE,KAAKm0B,KAAK,QAAQvwB,OAAOvC,EAAE,MAAM4lC,MAAM,CAACxgB,SAASjnB,EAAEm7B,WAAW/6B,EAAE,CAAC,GAAG,CAACslC,SAAS,SAAS3lC,GAAG,IAAIS,GAAE,EAAGoC,EAAE0lC,aAAa9nC,EAAEoC,EAAE0lC,WAAW,oCAAoCC,SAASxoC,EAAEyoC,oBAAoB,WAAW,IAAIzoC,EAAE,GAAGC,EAAE,CAAC+wB,KAAK,gBAAgB3wB,EAAE,CAACqoC,cAAc,MAAMC,YAAY,aAAaC,IAAI,MAAM5oC,EAAEiB,KAAK,CAAC4pB,IAAI,OAAOuQ,WAAW11B,EAAEA,EAAE,CAAC,EAAEzF,GAAG,CAAC,EAAE,CAAC6F,EAAE,u4CAAu4C,IAAI1F,EAAEsF,EAAEA,EAAE,CAAC,EAAErF,GAAG,CAAC,EAAE,CAACwjC,cAAc,YAAYrjC,EAAE,CAACqqB,IAAI,SAASuQ,WAAW11B,EAAEA,EAAE,CAAC,EAAEzF,GAAG,CAAC,EAAE,CAAC4oC,GAAG,MAAMC,GAAG,MAAMhnC,EAAE,OAAOolB,SAAS,IAAI,OAAOzmB,GAAGD,EAAE0mB,SAASjmB,KAAK,CAAC4pB,IAAI,UAAUuQ,WAAW11B,EAAEA,EAAE,CAAC,EAAErF,GAAG,CAAC,EAAE,CAACwjC,cAAc,IAAIxH,OAAO,wBAAwB,CAACxR,IAAI,UAAUuQ,WAAW11B,EAAEA,EAAE,CAAC,EAAEtF,GAAG,CAAC,EAAE,CAACi8B,OAAO,mBAAmBr8B,EAAEiB,KAAKT,GAAGR,EAAEiB,KAAK,CAAC4pB,IAAI,OAAOuQ,WAAW11B,EAAEA,EAAE,CAAC,EAAEzF,GAAG,CAAC,EAAE,CAAC8oC,QAAQ,IAAIjjC,EAAE,ySAAySohB,SAASzmB,EAAE,GAAG,CAAC,CAACoqB,IAAI,UAAUuQ,WAAW11B,EAAEA,EAAE,CAAC,EAAEtF,GAAG,CAAC,EAAE,CAACi8B,OAAO,qBAAqB57B,GAAGT,EAAEiB,KAAK,CAAC4pB,IAAI,OAAOuQ,WAAW11B,EAAEA,EAAE,CAAC,EAAEzF,GAAG,CAAC,EAAE,CAAC8oC,QAAQ,IAAIjjC,EAAE,gJAAgJohB,SAAS,CAAC,CAAC2D,IAAI,UAAUuQ,WAAW11B,EAAEA,EAAE,CAAC,EAAEtF,GAAG,CAAC,EAAE,CAACi8B,OAAO,qBAAqB,CAACxR,IAAI,IAAIuQ,WAAW,CAACkF,MAAM,WAAWpZ,SAASlnB,EAAE,CAAC,GAAG,CAACuF,MAAM,WAAW,MAAM,CAACwiC,oBAAoB,SAAS/nC,EAAEC,GAAsC,OAAnCA,EAAEA,EAAEw3B,aAAa,kBAAyBz3B,EAAE4/B,OAAO,OAAO3/B,IAAI,KAAKA,GAAGA,GAAGD,CAAC,EAAE,IAAIg+B,GAAG,CAAC,EAAEr9B,OAAOC,KAAKq9B,IAAI38B,SAAQ,SAAStB,IAAI,IAAIk+B,GAAG97B,QAAQpC,WAAWi+B,GAAGj+B,EAAE,IAAGuB,EAAED,SAAQ,SAAStB,GAAG,IAAIC,EAAEI,EAAEL,EAAEw6B,OAAOx6B,EAAEw6B,SAAS,CAAC,EAAE75B,OAAOC,KAAKP,GAAGiB,SAAQ,SAASrB,GAAG,mBAAmBI,EAAEJ,KAAKunC,GAAGvnC,GAAGI,EAAEJ,IAAI,WAAWO,EAAEH,EAAEJ,KAAKU,OAAOC,KAAKP,EAAEJ,IAAIqB,SAAQ,SAAStB,GAAGwnC,GAAGvnC,KAAKunC,GAAGvnC,GAAG,CAAC,GAAGunC,GAAGvnC,GAAGD,GAAGK,EAAEJ,GAAGD,EAAE,GAAE,IAAGA,EAAEuF,QAAQtF,EAAED,EAAEuF,QAAQ5E,OAAOC,KAAKX,GAAGqB,SAAQ,SAAStB,GAAGg+B,GAAGh+B,KAAKg+B,GAAGh+B,GAAG,IAAIg+B,GAAGh+B,GAAGiB,KAAKhB,EAAED,GAAG,KAAIA,EAAE2lC,UAAU3lC,EAAE2lC,SAAS1H,GAAG,IAAG,SAASj+B,GAAG,IAAI,IAAI,IAAIC,EAAEmB,UAAUC,OAAOhB,EAAE,IAAI0B,MAAM,EAAE9B,EAAEA,EAAE,EAAE,GAAGG,EAAE,EAAEA,EAAEH,EAAEG,IAAIC,EAAED,EAAE,GAAGgB,UAAUhB,GAAGJ,EAAEkB,WAAM,EAAOb,EAAE,CAAC,MAAML,GAAG,IAAIm2B,EAAE,MAAMn2B,CAAC,CAAC,CAA3I,EAA6I,SAASA,GAAG2D,IAAId,EAAEmmC,cAAcnmC,EAAEmmC,YAAYxK,IAAIvD,IAAG,WAAW+D,KAAKZ,GAAG,YAAY,KAAItD,GAAGv1B,MAAMG,EAAEA,EAAE,CAAC,EAAEo1B,GAAGv1B,OAAO,CAAC,EAAE,CAACS,QAAQ,SAAShG,EAAEC,GAAG66B,GAAGx1B,OAAOtF,GAAG0F,EAAEA,EAAE,CAAC,EAAEo1B,GAAGx1B,OAAOtF,IAAI,CAAC,GAAGC,GAAG48B,KAAKmC,IAAI,EAAEiK,SAAS,SAASjpC,GAAGA,EAAEsB,SAAQ,SAAStB,GAAgBA,GAATC,EAAEsC,EAAEvC,EAAE,IAAO,GAAjB,IAAoBC,EAAEA,EAAE,GAAG66B,GAAGx1B,OAAOtF,GAAG0F,EAAEA,EAAE,CAAC,EAAEo1B,GAAGx1B,OAAOtF,IAAI,CAAC,GAAGC,EAAE,IAAG48B,KAAKmC,IAAI,EAAEkK,SAAS,SAASlpC,GAAG,IAAIC,GAAGA,EAAE66B,GAAGt1B,OAAOvE,KAAKC,MAAMjB,EAAEoC,EAAErC,IAAI68B,KAAKmC,IAAI,GAAG,GAAE,CAA/+uD,KCJvnt3CmK,EAA2B,CAAC,EAGhC,SAASC,EAAoBC,GAE5B,IAAIC,EAAeH,EAAyBE,GAC5C,QAAqBE,IAAjBD,EACH,OAAOA,EAAaE,QAGrB,IAAIC,EAASN,EAAyBE,GAAY,CAGjDG,QAAS,CAAC,GAOX,OAHAE,EAAoBL,GAAUI,EAAQA,EAAOD,QAASJ,GAG/CK,EAAOD,OACf,CCrBAJ,EAAoB7mC,EAAKknC,IACxB,IAAIE,EAASF,GAAUA,EAAOG,WAC7B,IAAOH,EAAiB,QACxB,IAAM,EAEP,OADAL,EAAoBtjC,EAAE6jC,EAAQ,CAAEvpC,EAAGupC,IAC5BA,CAAM,ECLdP,EAAoBtjC,EAAI,CAAC0jC,EAASK,KACjC,IAAI,IAAI/mB,KAAO+mB,EACXT,EAAoB1mC,EAAEmnC,EAAY/mB,KAASsmB,EAAoB1mC,EAAE8mC,EAAS1mB,IAC5EniB,OAAOe,eAAe8nC,EAAS1mB,EAAK,CAAE9hB,YAAY,EAAM4B,IAAKinC,EAAW/mB,IAE1E,ECNDsmB,EAAoB1mC,EAAI,CAAConC,EAAKC,IAAUppC,OAAOmE,UAAUklC,eAAehlC,KAAK8kC,EAAKC","sources":["webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/js/all.min.js","webpack://pydata_sphinx_theme/webpack/bootstrap","webpack://pydata_sphinx_theme/webpack/runtime/compat get default export","webpack://pydata_sphinx_theme/webpack/runtime/define property getters","webpack://pydata_sphinx_theme/webpack/runtime/hasOwnProperty shorthand"],"sourcesContent":["/*!\n * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com\n * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)\n * Copyright 2024 Fonticons, Inc.\n */\n!function(){\"use strict\";var c={},l={};try{\"undefined\"!=typeof window&&(c=window),\"undefined\"!=typeof document&&(l=document)}catch(c){}var s=(c.navigator||{}).userAgent,a=void 0===s?\"\":s,z=c,e=l;z.document,e.documentElement&&e.head&&\"function\"==typeof e.addEventListener&&e.createElement,~a.indexOf(\"MSIE\")||a.indexOf(\"Trident/\");function H(l,c){var s,a=Object.keys(l);return Object.getOwnPropertySymbols&&(s=Object.getOwnPropertySymbols(l),c&&(s=s.filter(function(c){return Object.getOwnPropertyDescriptor(l,c).enumerable})),a.push.apply(a,s)),a}function t(l){for(var c=1;c<arguments.length;c++){var s=null!=arguments[c]?arguments[c]:{};c%2?H(Object(s),!0).forEach(function(c){V(l,c,s[c])}):Object.getOwnPropertyDescriptors?Object.defineProperties(l,Object.getOwnPropertyDescriptors(s)):H(Object(s)).forEach(function(c){Object.defineProperty(l,c,Object.getOwnPropertyDescriptor(s,c))})}return l}function V(c,l,s){return l in c?Object.defineProperty(c,l,{value:s,enumerable:!0,configurable:!0,writable:!0}):c[l]=s,c}function r(c,l){(null==l||l>c.length)&&(l=c.length);for(var s=0,a=new Array(l);s<l;s++)a[s]=c[s];return a}var M=\"___FONT_AWESOME___\",h=function(){try{return!0}catch(c){return!1}}(),n=\"classic\",i=\"sharp\",m=[n,i];function o(c){return new Proxy(c,{get:function(c,l){return l in c?c[l]:c[n]}})}o((V(C={},n,{fa:\"solid\",fas:\"solid\",\"fa-solid\":\"solid\",far:\"regular\",\"fa-regular\":\"regular\",fal:\"light\",\"fa-light\":\"light\",fat:\"thin\",\"fa-thin\":\"thin\",fad:\"duotone\",\"fa-duotone\":\"duotone\",fab:\"brands\",\"fa-brands\":\"brands\",fak:\"kit\",fakd:\"kit\",\"fa-kit\":\"kit\",\"fa-kit-duotone\":\"kit\"}),V(C,i,{fa:\"solid\",fass:\"solid\",\"fa-solid\":\"solid\",fasr:\"regular\",\"fa-regular\":\"regular\",fasl:\"light\",\"fa-light\":\"light\",fast:\"thin\",\"fa-thin\":\"thin\"}),C));var f=o((V(v={},n,{solid:\"fas\",regular:\"far\",light:\"fal\",thin:\"fat\",duotone:\"fad\",brands:\"fab\",kit:\"fak\"}),V(v,i,{solid:\"fass\",regular:\"fasr\",light:\"fasl\",thin:\"fast\"}),v)),e=(o((V(s={},n,{fab:\"fa-brands\",fad:\"fa-duotone\",fak:\"fa-kit\",fal:\"fa-light\",far:\"fa-regular\",fas:\"fa-solid\",fat:\"fa-thin\"}),V(s,i,{fass:\"fa-solid\",fasr:\"fa-regular\",fasl:\"fa-light\",fast:\"fa-thin\"}),s)),o((V(c={},n,{\"fa-brands\":\"fab\",\"fa-duotone\":\"fad\",\"fa-kit\":\"fak\",\"fa-light\":\"fal\",\"fa-regular\":\"far\",\"fa-solid\":\"fas\",\"fa-thin\":\"fat\"}),V(c,i,{\"fa-solid\":\"fass\",\"fa-regular\":\"fasr\",\"fa-light\":\"fasl\",\"fa-thin\":\"fast\"}),c)),o((V(l={},n,{900:\"fas\",400:\"far\",normal:\"far\",300:\"fal\",100:\"fat\"}),V(l,i,{900:\"fass\",400:\"fasr\",300:\"fasl\",100:\"fast\"}),l)),[1,2,3,4,5,6,7,8,9,10]),a=e.concat([11,12,13,14,15,16,17,18,19,20]),C=\"duotone-group\",v=\"swap-opacity\",s=\"primary\",c=\"secondary\",l=new Set;Object.keys(f[n]).map(l.add.bind(l)),Object.keys(f[i]).map(l.add.bind(l));[].concat(m,function(c){if(Array.isArray(c))return r(c)}(l=l)||function(c){if(\"undefined\"!=typeof Symbol&&null!=c[Symbol.iterator]||null!=c[\"@@iterator\"])return Array.from(c)}(l)||function(c,l){if(c){if(\"string\"==typeof c)return r(c,l);var s=Object.prototype.toString.call(c).slice(8,-1);return\"Map\"===(s=\"Object\"===s&&c.constructor?c.constructor.name:s)||\"Set\"===s?Array.from(c):\"Arguments\"===s||/^(?:Ui|I)nt(?:8|16|32)(?:Clamped)?Array$/.test(s)?r(c,l):void 0}}(l)||function(){throw new TypeError(\"Invalid attempt to spread non-iterable instance.\\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method.\")}(),[\"2xs\",\"xs\",\"sm\",\"lg\",\"xl\",\"2xl\",\"beat\",\"border\",\"fade\",\"beat-fade\",\"bounce\",\"flip-both\",\"flip-horizontal\",\"flip-vertical\",\"flip\",\"fw\",\"inverse\",\"layers-counter\",\"layers-text\",\"layers\",\"li\",\"pull-left\",\"pull-right\",\"pulse\",\"rotate-180\",\"rotate-270\",\"rotate-90\",\"rotate-by\",\"shake\",\"spin-pulse\",\"spin-reverse\",\"spin\",\"stack-1x\",\"stack-2x\",\"stack\",\"ul\",C,v,s,c]).concat(e.map(function(c){return\"\".concat(c,\"x\")})).concat(a.map(function(c){return\"w-\".concat(c)}));z=z||{};z[M]||(z[M]={}),z[M].styles||(z[M].styles={}),z[M].hooks||(z[M].hooks={}),z[M].shims||(z[M].shims=[]);var L=z[M];function u(a){return Object.keys(a).reduce(function(c,l){var s=a[l];return!!s.icon?c[s.iconName]=s.icon:c[l]=s,c},{})}function d(c,l,s){var a=(2<arguments.length&&void 0!==s?s:{}).skipHooks,s=void 0!==a&&a,a=u(l);\"function\"!=typeof L.hooks.addPack||s?L.styles[c]=t(t({},L.styles[c]||{}),a):L.hooks.addPack(c,u(l)),\"fas\"===c&&d(\"fa\",l)}var p={monero:[496,512,[],\"f3d0\",\"M352 384h108.4C417 455.9 338.1 504 248 504S79 455.9 35.6 384H144V256.2L248 361l104-105v128zM88 336V128l159.4 159.4L408 128v208h74.8c8.5-25.1 13.2-52 13.2-80C496 119 385 8 248 8S0 119 0 256c0 28 4.6 54.9 13.2 80H88z\"],hooli:[640,512,[],\"f427\",\"M144.5 352l38.3.8c-13.2-4.6-26-10.2-38.3-16.8zm57.7-5.3v5.3l-19.4.8c36.5 12.5 69.9 14.2 94.7 7.2-19.9.2-45.8-2.6-75.3-13.3zm408.9-115.2c15.9 0 28.9-12.9 28.9-28.9s-12.9-24.5-28.9-24.5c-15.9 0-28.9 8.6-28.9 24.5s12.9 28.9 28.9 28.9zm-29 120.5H640V241.5h-57.9zm-73.7 0h57.9V156.7L508.4 184zm-31-119.4c-18.2-18.2-50.4-17.1-50.4-17.1s-32.3-1.1-50.4 17.1c-18.2 18.2-16.8 33.9-16.8 52.6s-1.4 34.3 16.8 52.5 50.4 17.1 50.4 17.1 32.3 1.1 50.4-17.1c18.2-18.2 16.8-33.8 16.8-52.5-.1-18.8 1.3-34.5-16.8-52.6zm-39.8 71.9c0 3.6-1.8 12.5-10.7 12.5s-10.7-8.9-10.7-12.5v-40.4c0-8.7 7.3-10.9 10.7-10.9s10.7 2.1 10.7 10.9zm-106.2-71.9c-18.2-18.2-50.4-17.1-50.4-17.1s-32.2-1.1-50.4 17.1c-1.9 1.9-3.7 3.9-5.3 6-38.2-29.6-72.5-46.5-102.1-61.1v-20.7l-22.5 10.6c-54.4-22.1-89-18.2-97.3.1 0 0-24.9 32.8 61.8 110.8V352h57.9v-28.6c-6.5-4.2-13-8.7-19.4-13.6-14.8-11.2-27.4-21.6-38.4-31.4v-31c13.1 14.7 30.5 31.4 53.4 50.3l4.5 3.6v-29.8c0-6.9 1.7-18.2 10.8-18.2s10.6 6.9 10.6 15V317c18 12.2 37.3 22.1 57.7 29.6v-93.9c0-18.7-13.4-37.4-40.6-37.4-15.8-.1-30.5 8.2-38.5 21.9v-54.3c41.9 20.9 83.9 46.5 99.9 58.3-10.2 14.6-9.3 28.1-9.3 43.7 0 18.7-1.4 34.3 16.8 52.5s50.4 17.1 50.4 17.1 32.3 1.1 50.4-17.1c18.2-18.2 16.7-33.8 16.7-52.5 0-18.5 1.5-34.2-16.7-52.3zM65.2 184v63.3c-48.7-54.5-38.9-76-35.2-79.1 13.5-11.4 37.5-8 64.4 2.1zm226.5 120.5c0 3.6-1.8 12.5-10.7 12.5s-10.7-8.9-10.7-12.5v-40.4c0-8.7 7.3-10.9 10.7-10.9s10.7 2.1 10.7 10.9z\"],yelp:[384,512,[],\"f1e9\",\"M42.9 240.32l99.62 48.61c19.2 9.4 16.2 37.51-4.5 42.71L30.5 358.45a22.79 22.79 0 0 1-28.21-19.6 197.16 197.16 0 0 1 9-85.32 22.8 22.8 0 0 1 31.61-13.21zm44 239.25a199.45 199.45 0 0 0 79.42 32.11A22.78 22.78 0 0 0 192.94 490l3.9-110.82c.7-21.3-25.5-31.91-39.81-16.1l-74.21 82.4a22.82 22.82 0 0 0 4.09 34.09zm145.34-109.92l58.81 94a22.93 22.93 0 0 0 34 5.5 198.36 198.36 0 0 0 52.71-67.61A23 23 0 0 0 364.17 370l-105.42-34.26c-20.31-6.5-37.81 15.8-26.51 33.91zm148.33-132.23a197.44 197.44 0 0 0-50.41-69.31 22.85 22.85 0 0 0-34 4.4l-62 91.92c-11.9 17.7 4.7 40.61 25.2 34.71L366 268.63a23 23 0 0 0 14.61-31.21zM62.11 30.18a22.86 22.86 0 0 0-9.9 32l104.12 180.44c11.7 20.2 42.61 11.9 42.61-11.4V22.88a22.67 22.67 0 0 0-24.5-22.8 320.37 320.37 0 0 0-112.33 30.1z\"],\"cc-visa\":[576,512,[],\"f1f0\",\"M470.1 231.3s7.6 37.2 9.3 45H446c3.3-8.9 16-43.5 16-43.5-.2.3 3.3-9.1 5.3-14.9l2.8 13.4zM576 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h480c26.5 0 48 21.5 48 48zM152.5 331.2L215.7 176h-42.5l-39.3 106-4.3-21.5-14-71.4c-2.3-9.9-9.4-12.7-18.2-13.1H32.7l-.7 3.1c15.8 4 29.9 9.8 42.2 17.1l35.8 135h42.5zm94.4.2L272.1 176h-40.2l-25.1 155.4h40.1zm139.9-50.8c.2-17.7-10.6-31.2-33.7-42.3-14.1-7.1-22.7-11.9-22.7-19.2.2-6.6 7.3-13.4 23.1-13.4 13.1-.3 22.7 2.8 29.9 5.9l3.6 1.7 5.5-33.6c-7.9-3.1-20.5-6.6-36-6.6-39.7 0-67.6 21.2-67.8 51.4-.3 22.3 20 34.7 35.2 42.2 15.5 7.6 20.8 12.6 20.8 19.3-.2 10.4-12.6 15.2-24.1 15.2-16 0-24.6-2.5-37.7-8.3l-5.3-2.5-5.6 34.9c9.4 4.3 26.8 8.1 44.8 8.3 42.2.1 69.7-20.8 70-53zM528 331.4L495.6 176h-31.1c-9.6 0-16.9 2.8-21 12.9l-59.7 142.5H426s6.9-19.2 8.4-23.3H486c1.2 5.5 4.8 23.3 4.8 23.3H528z\"],lastfm:[512,512,[],\"f202\",\"M225.8 367.1l-18.8-51s-30.5 34-76.2 34c-40.5 0-69.2-35.2-69.2-91.5 0-72.1 36.4-97.9 72.1-97.9 66.5 0 74.8 53.3 100.9 134.9 18.8 56.9 54 102.6 155.4 102.6 72.7 0 122-22.3 122-80.9 0-72.9-62.7-80.6-115-92.1-25.8-5.9-33.4-16.4-33.4-34 0-19.9 15.8-31.7 41.6-31.7 28.2 0 43.4 10.6 45.7 35.8l58.6-7c-4.7-52.8-41.1-74.5-100.9-74.5-52.8 0-104.4 19.9-104.4 83.9 0 39.9 19.4 65.1 68 76.8 44.9 10.6 79.8 13.8 79.8 45.7 0 21.7-21.1 30.5-61 30.5-59.2 0-83.9-31.1-97.9-73.9-32-96.8-43.6-163-161.3-163C45.7 113.8 0 168.3 0 261c0 89.1 45.7 137.2 127.9 137.2 66.2 0 97.9-31.1 97.9-31.1z\"],shopware:[512,512,[],\"f5b5\",\"M403.5 455.41A246.17 246.17 0 0 1 256 504C118.81 504 8 393 8 256 8 118.81 119 8 256 8a247.39 247.39 0 0 1 165.7 63.5 3.57 3.57 0 0 1-2.86 6.18A418.62 418.62 0 0 0 362.13 74c-129.36 0-222.4 53.47-222.4 155.35 0 109 92.13 145.88 176.83 178.73 33.64 13 65.4 25.36 87 41.59a3.58 3.58 0 0 1 0 5.72zM503 233.09a3.64 3.64 0 0 0-1.27-2.44c-51.76-43-93.62-60.48-144.48-60.48-84.13 0-80.25 52.17-80.25 53.63 0 42.6 52.06 62 112.34 84.49 31.07 11.59 63.19 23.57 92.68 39.93a3.57 3.57 0 0 0 5-1.82A249 249 0 0 0 503 233.09z\"],\"creative-commons-nc\":[496,512,[],\"f4e8\",\"M247.6 8C387.4 8 496 115.9 496 256c0 147.2-118.5 248-248.4 248C113.1 504 0 393.2 0 256 0 123.1 104.7 8 247.6 8zM55.8 189.1c-7.4 20.4-11.1 42.7-11.1 66.9 0 110.9 92.1 202.4 203.7 202.4 122.4 0 177.2-101.8 178.5-104.1l-93.4-41.6c-7.7 37.1-41.2 53-68.2 55.4v38.1h-28.8V368c-27.5-.3-52.6-10.2-75.3-29.7l34.1-34.5c31.7 29.4 86.4 31.8 86.4-2.2 0-6.2-2.2-11.2-6.6-15.1-14.2-6-1.8-.1-219.3-97.4zM248.4 52.3c-38.4 0-112.4 8.7-170.5 93l94.8 42.5c10-31.3 40.4-42.9 63.8-44.3v-38.1h28.8v38.1c22.7 1.2 43.4 8.9 62 23L295 199.7c-42.7-29.9-83.5-8-70 11.1 53.4 24.1 43.8 19.8 93 41.6l127.1 56.7c4.1-17.4 6.2-35.1 6.2-53.1 0-57-19.8-105-59.3-143.9-39.3-39.9-87.2-59.8-143.6-59.8z\"],aws:[640,512,[],\"f375\",\"M180.41 203.01c-.72 22.65 10.6 32.68 10.88 39.05a8.164 8.164 0 0 1-4.1 6.27l-12.8 8.96a10.66 10.66 0 0 1-5.63 1.92c-.43-.02-8.19 1.83-20.48-25.61a78.608 78.608 0 0 1-62.61 29.45c-16.28.89-60.4-9.24-58.13-56.21-1.59-38.28 34.06-62.06 70.93-60.05 7.1.02 21.6.37 46.99 6.27v-15.62c2.69-26.46-14.7-46.99-44.81-43.91-2.4.01-19.4-.5-45.84 10.11-7.36 3.38-8.3 2.82-10.75 2.82-7.41 0-4.36-21.48-2.94-24.2 5.21-6.4 35.86-18.35 65.94-18.18a76.857 76.857 0 0 1 55.69 17.28 70.285 70.285 0 0 1 17.67 52.36l-.01 69.29zM93.99 235.4c32.43-.47 46.16-19.97 49.29-30.47 2.46-10.05 2.05-16.41 2.05-27.4-9.67-2.32-23.59-4.85-39.56-4.87-15.15-1.14-42.82 5.63-41.74 32.26-1.24 16.79 11.12 31.4 29.96 30.48zm170.92 23.05c-7.86.72-11.52-4.86-12.68-10.37l-49.8-164.65c-.97-2.78-1.61-5.65-1.92-8.58a4.61 4.61 0 0 1 3.86-5.25c.24-.04-2.13 0 22.25 0 8.78-.88 11.64 6.03 12.55 10.37l35.72 140.83 33.16-140.83c.53-3.22 2.94-11.07 12.8-10.24h17.16c2.17-.18 11.11-.5 12.68 10.37l33.42 142.63L420.98 80.1c.48-2.18 2.72-11.37 12.68-10.37h19.72c.85-.13 6.15-.81 5.25 8.58-.43 1.85 3.41-10.66-52.75 169.9-1.15 5.51-4.82 11.09-12.68 10.37h-18.69c-10.94 1.15-12.51-9.66-12.68-10.75L328.67 110.7l-32.78 136.99c-.16 1.09-1.73 11.9-12.68 10.75h-18.3zm273.48 5.63c-5.88.01-33.92-.3-57.36-12.29a12.802 12.802 0 0 1-7.81-11.91v-10.75c0-8.45 6.2-6.9 8.83-5.89 10.04 4.06 16.48 7.14 28.81 9.6 36.65 7.53 52.77-2.3 56.72-4.48 13.15-7.81 14.19-25.68 5.25-34.95-10.48-8.79-15.48-9.12-53.13-21-4.64-1.29-43.7-13.61-43.79-52.36-.61-28.24 25.05-56.18 69.52-55.95 12.67-.01 46.43 4.13 55.57 15.62 1.35 2.09 2.02 4.55 1.92 7.04v10.11c0 4.44-1.62 6.66-4.87 6.66-7.71-.86-21.39-11.17-49.16-10.75-6.89-.36-39.89.91-38.41 24.97-.43 18.96 26.61 26.07 29.7 26.89 36.46 10.97 48.65 12.79 63.12 29.58 17.14 22.25 7.9 48.3 4.35 55.44-19.08 37.49-68.42 34.44-69.26 34.42zm40.2 104.86c-70.03 51.72-171.69 79.25-258.49 79.25A469.127 469.127 0 0 1 2.83 327.46c-6.53-5.89-.77-13.96 7.17-9.47a637.37 637.37 0 0 0 316.88 84.12 630.22 630.22 0 0 0 241.59-49.55c11.78-5 21.77 7.8 10.12 16.38zm29.19-33.29c-8.96-11.52-59.28-5.38-81.81-2.69-6.79.77-7.94-5.12-1.79-9.47 40.07-28.17 105.88-20.1 113.44-10.63 7.55 9.47-2.05 75.41-39.56 106.91-5.76 4.87-11.27 2.3-8.71-4.1 8.44-21.25 27.39-68.49 18.43-80.02z\"],redhat:[512,512,[],\"f7bc\",\"M341.52 285.56c33.65 0 82.34-6.94 82.34-47 .22-6.74.86-1.82-20.88-96.24-4.62-19.15-8.68-27.84-42.31-44.65-26.09-13.34-82.92-35.37-99.73-35.37-15.66 0-20.2 20.17-38.87 20.17-18 0-31.31-15.06-48.12-15.06-16.14 0-26.66 11-34.78 33.62-27.5 77.55-26.28 74.27-26.12 78.27 0 24.8 97.64 106.11 228.47 106.11M429 254.84c4.65 22 4.65 24.35 4.65 27.25 0 37.66-42.33 58.56-98 58.56-125.74.08-235.91-73.65-235.91-122.33a49.55 49.55 0 0 1 4.06-19.72C58.56 200.86 0 208.93 0 260.63c0 84.67 200.63 189 359.49 189 121.79 0 152.51-55.08 152.51-98.58 0-34.21-29.59-73.05-82.93-96.24\"],yoast:[448,512,[],\"f2b1\",\"M91.3 76h186l-7 18.9h-179c-39.7 0-71.9 31.6-71.9 70.3v205.4c0 35.4 24.9 70.3 84 70.3V460H91.3C41.2 460 0 419.8 0 370.5V165.2C0 115.9 40.7 76 91.3 76zm229.1-56h66.5C243.1 398.1 241.2 418.9 202.2 459.3c-20.8 21.6-49.3 31.7-78.3 32.7v-51.1c49.2-7.7 64.6-49.9 64.6-75.3 0-20.1.6-12.6-82.1-223.2h61.4L218.2 299 320.4 20zM448 161.5V460H234c6.6-9.6 10.7-16.3 12.1-19.4h182.5V161.5c0-32.5-17.1-51.9-48.2-62.9l6.7-17.6c41.7 13.6 60.9 43.1 60.9 80.5z\"],cloudflare:[640,512,[],\"e07d\",\"M407.906,319.913l-230.8-2.928a4.58,4.58,0,0,1-3.632-1.926,4.648,4.648,0,0,1-.494-4.147,6.143,6.143,0,0,1,5.361-4.076L411.281,303.9c27.631-1.26,57.546-23.574,68.022-50.784l13.286-34.542a7.944,7.944,0,0,0,.524-2.936,7.735,7.735,0,0,0-.164-1.631A151.91,151.91,0,0,0,201.257,198.4,68.12,68.12,0,0,0,94.2,269.59C41.924,271.106,0,313.728,0,366.12a96.054,96.054,0,0,0,1.029,13.958,4.508,4.508,0,0,0,4.445,3.871l426.1.051c.043,0,.08-.019.122-.02a5.606,5.606,0,0,0,5.271-4l3.273-11.265c3.9-13.4,2.448-25.8-4.1-34.9C430.124,325.423,420.09,320.487,407.906,319.913ZM513.856,221.1c-2.141,0-4.271.062-6.391.164a3.771,3.771,0,0,0-3.324,2.653l-9.077,31.193c-3.9,13.4-2.449,25.786,4.1,34.89,6.02,8.4,16.054,13.323,28.238,13.9l49.2,2.939a4.491,4.491,0,0,1,3.51,1.894,4.64,4.64,0,0,1,.514,4.169,6.153,6.153,0,0,1-5.351,4.075l-51.125,2.939c-27.754,1.27-57.669,23.574-68.145,50.784l-3.695,9.606a2.716,2.716,0,0,0,2.427,3.68c.046,0,.088.017.136.017h175.91a4.69,4.69,0,0,0,4.539-3.37,124.807,124.807,0,0,0,4.682-34C640,277.3,583.524,221.1,513.856,221.1Z\"],ups:[384,512,[],\"f7e0\",\"M103.2 303c-5.2 3.6-32.6 13.1-32.6-19V180H37.9v102.6c0 74.9 80.2 51.1 97.9 39V180h-32.6zM4 74.82v220.9c0 103.7 74.9 135.2 187.7 184.1 112.4-48.9 187.7-80.2 187.7-184.1V74.82c-116.3-61.6-281.8-49.6-375.4 0zm358.1 220.9c0 86.6-53.2 113.6-170.4 165.3-117.5-51.8-170.5-78.7-170.5-165.3v-126.4c102.3-93.8 231.6-100 340.9-89.8zm-209.6-107.4v212.8h32.7v-68.7c24.4 7.3 71.7-2.6 71.7-78.5 0-97.4-80.7-80.92-104.4-65.6zm32.7 117.3v-100.3c8.4-4.2 38.4-12.7 38.4 49.3 0 67.9-36.4 51.8-38.4 51zm79.1-86.4c.1 47.3 51.6 42.5 52.2 70.4.6 23.5-30.4 23-50.8 4.9v30.1c36.2 21.5 81.9 8.1 83.2-33.5 1.7-51.5-54.1-46.6-53.4-73.2.6-20.3 30.6-20.5 48.5-2.2v-28.4c-28.5-22-79.9-9.2-79.7 31.9z\"],pixiv:[448,512,[],\"e640\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm171.5 84c41 0 76.3 12.9 101.4 35.2l0 0c25.2 22.2 39.8 54.1 39.8 88.8c.1 35.3-16.6 66.3-42.4 87c-25.9 20.8-60.6 32.4-98.8 32.4c-43.5 0-83.8-16.1-83.8-16.1v51.8c7.4 2.2 19.7 7 11.9 14.8H104.8c-7.7-7.8 3.6-12.4 12.1-14.8V175.5C97.1 190.9 87 204.3 81.8 214.2c6 19.4-5.3 18.5-5.3 18.5L56 199.7s72.7-83.7 179.5-83.7zm-3.6 222.9c30 0 56-11.3 73.9-29.2c17.9-18.1 27.9-41.6 28-70.2c-.1-29.3-9.5-54.6-26.7-73.6c-17.2-18.9-42.7-31.3-75.2-31.4c-26.7-.1-59.8 9-80.2 23.7V323.1c18.6 9.3 46.8 15.9 80.2 15.8z\"],wpexplorer:[512,512,[],\"f2de\",\"M512 256c0 141.2-114.7 256-256 256C114.8 512 0 397.3 0 256S114.7 0 256 0s256 114.7 256 256zm-32 0c0-123.2-100.3-224-224-224C132.5 32 32 132.5 32 256s100.5 224 224 224 224-100.5 224-224zM160.9 124.6l86.9 37.1-37.1 86.9-86.9-37.1 37.1-86.9zm110 169.1l46.6 94h-14.6l-50-100-48.9 100h-14l51.1-106.9-22.3-9.4 6-14 68.6 29.1-6 14.3-16.5-7.1zm-11.8-116.3l68.6 29.4-29.4 68.3L230 246l29.1-68.6zm80.3 42.9l54.6 23.1-23.4 54.3-54.3-23.1 23.1-54.3z\"],dyalog:[416,512,[],\"f399\",\"M0 32v119.2h64V96h107.2C284.6 96 352 176.2 352 255.9 352 332 293.4 416 171.2 416H0v64h171.2C331.9 480 416 367.3 416 255.9c0-58.7-22.1-113.4-62.3-154.3C308.9 56 245.7 32 171.2 32H0z\"],bity:[496,512,[],\"f37a\",\"M78.4 67.2C173.8-22 324.5-24 421.5 71c14.3 14.1-6.4 37.1-22.4 21.5-84.8-82.4-215.8-80.3-298.9-3.2-16.3 15.1-36.5-8.3-21.8-22.1zm98.9 418.6c19.3 5.7 29.3-23.6 7.9-30C73 421.9 9.4 306.1 37.7 194.8c5-19.6-24.9-28.1-30.2-7.1-32.1 127.4 41.1 259.8 169.8 298.1zm148.1-2c121.9-40.2 192.9-166.9 164.4-291-4.5-19.7-34.9-13.8-30 7.9 24.2 107.7-37.1 217.9-143.2 253.4-21.2 7-10.4 36 8.8 29.7zm-62.9-79l.2-71.8c0-8.2-6.6-14.8-14.8-14.8-8.2 0-14.8 6.7-14.8 14.8l-.2 71.8c0 8.2 6.6 14.8 14.8 14.8s14.8-6.6 14.8-14.8zm71-269c2.1 90.9 4.7 131.9-85.5 132.5-92.5-.7-86.9-44.3-85.5-132.5 0-21.8-32.5-19.6-32.5 0v71.6c0 69.3 60.7 90.9 118 90.1 57.3.8 118-20.8 118-90.1v-71.6c0-19.6-32.5-21.8-32.5 0z\"],stackpath:[448,512,[],\"f842\",\"M244.6 232.4c0 8.5-4.26 20.49-21.34 20.49h-19.61v-41.47h19.61c17.13 0 21.34 12.36 21.34 20.98zM448 32v448H0V32zM151.3 287.84c0-21.24-12.12-34.54-46.72-44.85-20.57-7.41-26-10.91-26-18.63s7-14.61 20.41-14.61c14.09 0 20.79 8.45 20.79 18.35h30.7l.19-.57c.5-19.57-15.06-41.65-51.12-41.65-23.37 0-52.55 10.75-52.55 38.29 0 19.4 9.25 31.29 50.74 44.37 17.26 6.15 21.91 10.4 21.91 19.48 0 15.2-19.13 14.23-19.47 14.23-20.4 0-25.65-9.1-25.65-21.9h-30.8l-.18.56c-.68 31.32 28.38 45.22 56.63 45.22 29.98 0 51.12-13.55 51.12-38.29zm125.38-55.63c0-25.3-18.43-45.46-53.42-45.46h-51.78v138.18h32.17v-47.36h19.61c30.25 0 53.42-15.95 53.42-45.36zM297.94 325L347 186.78h-31.09L268 325zm106.52-138.22h-31.09L325.46 325h29.94z\"],buysellads:[448,512,[],\"f20d\",\"M224 150.7l42.9 160.7h-85.8L224 150.7zM448 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h352c26.5 0 48 21.5 48 48zm-65.3 325.3l-94.5-298.7H159.8L65.3 405.3H156l111.7-91.6 24.2 91.6h90.8z\"],\"first-order\":[448,512,[],\"f2b0\",\"M12.9 229.2c.1-.1.2-.3.3-.4 0 .1 0 .3-.1.4h-.2zM224 96.6c-7.1 0-14.6.6-21.4 1.7l3.7 67.4-22-64c-14.3 3.7-27.7 9.4-40 16.6l29.4 61.4-45.1-50.9c-11.4 8.9-21.7 19.1-30.6 30.9l50.6 45.4-61.1-29.7c-7.1 12.3-12.9 25.7-16.6 40l64.3 22.6-68-4c-.9 7.1-1.4 14.6-1.4 22s.6 14.6 1.4 21.7l67.7-4-64 22.6c3.7 14.3 9.4 27.7 16.6 40.3l61.1-29.7L97.7 352c8.9 11.7 19.1 22.3 30.9 30.9l44.9-50.9-29.5 61.4c12.3 7.4 25.7 13.1 40 16.9l22.3-64.6-4 68c7.1 1.1 14.6 1.7 21.7 1.7 7.4 0 14.6-.6 21.7-1.7l-4-68.6 22.6 65.1c14.3-4 27.7-9.4 40-16.9L274.9 332l44.9 50.9c11.7-8.9 22-19.1 30.6-30.9l-50.6-45.1 61.1 29.4c7.1-12.3 12.9-25.7 16.6-40.3l-64-22.3 67.4 4c1.1-7.1 1.4-14.3 1.4-21.7s-.3-14.9-1.4-22l-67.7 4 64-22.3c-3.7-14.3-9.1-28-16.6-40.3l-60.9 29.7 50.6-45.4c-8.9-11.7-19.1-22-30.6-30.9l-45.1 50.9 29.4-61.1c-12.3-7.4-25.7-13.1-40-16.9L241.7 166l4-67.7c-7.1-1.2-14.3-1.7-21.7-1.7zM443.4 128v256L224 512 4.6 384V128L224 0l219.4 128zm-17.1 10.3L224 20.9 21.7 138.3v235.1L224 491.1l202.3-117.7V138.3zM224 37.1l187.7 109.4v218.9L224 474.9 36.3 365.4V146.6L224 37.1zm0 50.9c-92.3 0-166.9 75.1-166.9 168 0 92.6 74.6 167.7 166.9 167.7 92 0 166.9-75.1 166.9-167.7 0-92.9-74.9-168-166.9-168z\"],modx:[448,512,[],\"f285\",\"M356 241.8l36.7 23.7V480l-133-83.8L356 241.8zM440 75H226.3l-23 37.8 153.5 96.5L440 75zm-89 142.8L55.2 32v214.5l46 29L351 217.8zM97 294.2L8 437h213.7l125-200.5L97 294.2z\"],guilded:[448,512,[],\"e07e\",\"M443.427,64H4.571c0,103.26,22.192,180.06,43.418,222.358C112.046,414.135,224,448,225.256,448a312.824,312.824,0,0,0,140.55-103.477c25.907-33.923,53.1-87.19,65.916-145.761H171.833c4.14,36.429,22.177,67.946,45.1,86.944h88.589c-17.012,28.213-48.186,54.4-80.456,69.482-31.232-13.259-69.09-46.544-96.548-98.362-26.726-53.833-27.092-105.883-27.092-105.883H437.573A625.91,625.91,0,0,0,443.427,64Z\"],vnv:[640,512,[],\"f40b\",\"M104.9 352c-34.1 0-46.4-30.4-46.4-30.4L2.6 210.1S-7.8 192 13 192h32.8c10.4 0 13.2 8.7 18.8 18.1l36.7 74.5s5.2 13.1 21.1 13.1 21.1-13.1 21.1-13.1l36.7-74.5c5.6-9.5 8.4-18.1 18.8-18.1h32.8c20.8 0 10.4 18.1 10.4 18.1l-55.8 111.5S174.2 352 140 352h-35.1zm395 0c-34.1 0-46.4-30.4-46.4-30.4l-55.9-111.5S387.2 192 408 192h32.8c10.4 0 13.2 8.7 18.8 18.1l36.7 74.5s5.2 13.1 21.1 13.1 21.1-13.1 21.1-13.1l36.8-74.5c5.6-9.5 8.4-18.1 18.8-18.1H627c20.8 0 10.4 18.1 10.4 18.1l-55.9 111.5S569.3 352 535.1 352h-35.2zM337.6 192c34.1 0 46.4 30.4 46.4 30.4l55.9 111.5s10.4 18.1-10.4 18.1h-32.8c-10.4 0-13.2-8.7-18.8-18.1l-36.7-74.5s-5.2-13.1-21.1-13.1c-15.9 0-21.1 13.1-21.1 13.1l-36.7 74.5c-5.6 9.4-8.4 18.1-18.8 18.1h-32.9c-20.8 0-10.4-18.1-10.4-18.1l55.9-111.5s12.2-30.4 46.4-30.4h35.1z\"],\"square-js\":[448,512,[\"js-square\"],\"f3b9\",\"M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM180.9 444.9c-33.7 0-53.2-17.4-63.2-38.5L152 385.7c6.6 11.7 12.6 21.6 27.1 21.6c13.8 0 22.6-5.4 22.6-26.5V237.7h42.1V381.4c0 43.6-25.6 63.5-62.9 63.5zm85.8-43L301 382.1c9 14.7 20.8 25.6 41.5 25.6c17.4 0 28.6-8.7 28.6-20.8c0-14.4-11.4-19.5-30.7-28l-10.5-4.5c-30.4-12.9-50.5-29.2-50.5-63.5c0-31.6 24.1-55.6 61.6-55.6c26.8 0 46 9.3 59.8 33.7L368 290c-7.2-12.9-15-18-27.1-18c-12.3 0-20.1 7.8-20.1 18c0 12.6 7.8 17.7 25.9 25.6l10.5 4.5c35.8 15.3 55.9 31 55.9 66.2c0 37.8-29.8 58.6-69.7 58.6c-39.1 0-64.4-18.6-76.7-43z\"],microsoft:[448,512,[],\"f3ca\",\"M0 32h214.6v214.6H0V32zm233.4 0H448v214.6H233.4V32zM0 265.4h214.6V480H0V265.4zm233.4 0H448V480H233.4V265.4z\"],qq:[448,512,[],\"f1d6\",\"M433.754 420.445c-11.526 1.393-44.86-52.741-44.86-52.741 0 31.345-16.136 72.247-51.051 101.786 16.842 5.192 54.843 19.167 45.803 34.421-7.316 12.343-125.51 7.881-159.632 4.037-34.122 3.844-152.316 8.306-159.632-4.037-9.045-15.25 28.918-29.214 45.783-34.415-34.92-29.539-51.059-70.445-51.059-101.792 0 0-33.334 54.134-44.859 52.741-5.37-.65-12.424-29.644 9.347-99.704 10.261-33.024 21.995-60.478 40.144-105.779C60.683 98.063 108.982.006 224 0c113.737.006 163.156 96.133 160.264 214.963 18.118 45.223 29.912 72.85 40.144 105.778 21.768 70.06 14.716 99.053 9.346 99.704z\"],orcid:[512,512,[],\"f8d2\",\"M294.75 188.19h-45.92V342h47.47c67.62 0 83.12-51.34 83.12-76.91 0-41.64-26.54-76.9-84.67-76.9zM256 8C119 8 8 119 8 256s111 248 248 248 248-111 248-248S393 8 256 8zm-80.79 360.76h-29.84v-207.5h29.84zm-14.92-231.14a19.57 19.57 0 1 1 19.57-19.57 19.64 19.64 0 0 1-19.57 19.57zM300 369h-81V161.26h80.6c76.73 0 110.44 54.83 110.44 103.85C410 318.39 368.38 369 300 369z\"],java:[384,512,[],\"f4e4\",\"M277.74 312.9c9.8-6.7 23.4-12.5 23.4-12.5s-38.7 7-77.2 10.2c-47.1 3.9-97.7 4.7-123.1 1.3-60.1-8 33-30.1 33-30.1s-36.1-2.4-80.6 19c-52.5 25.4 130 37 224.5 12.1zm-85.4-32.1c-19-42.7-83.1-80.2 0-145.8C296 53.2 242.84 0 242.84 0c21.5 84.5-75.6 110.1-110.7 162.6-23.9 35.9 11.7 74.4 60.2 118.2zm114.6-176.2c.1 0-175.2 43.8-91.5 140.2 24.7 28.4-6.5 54-6.5 54s62.7-32.4 33.9-72.9c-26.9-37.8-47.5-56.6 64.1-121.3zm-6.1 270.5a12.19 12.19 0 0 1-2 2.6c128.3-33.7 81.1-118.9 19.8-97.3a17.33 17.33 0 0 0-8.2 6.3 70.45 70.45 0 0 1 11-3c31-6.5 75.5 41.5-20.6 91.4zM348 437.4s14.5 11.9-15.9 21.2c-57.9 17.5-240.8 22.8-291.6.7-18.3-7.9 16-19 26.8-21.3 11.2-2.4 17.7-2 17.7-2-20.3-14.3-131.3 28.1-56.4 40.2C232.84 509.4 401 461.3 348 437.4zM124.44 396c-78.7 22 47.9 67.4 148.1 24.5a185.89 185.89 0 0 1-28.2-13.8c-44.7 8.5-65.4 9.1-106 4.5-33.5-3.8-13.9-15.2-13.9-15.2zm179.8 97.2c-78.7 14.8-175.8 13.1-233.3 3.6 0-.1 11.8 9.7 72.4 13.6 92.2 5.9 233.8-3.3 237.1-46.9 0 0-6.4 16.5-76.2 29.7zM260.64 353c-59.2 11.4-93.5 11.1-136.8 6.6-33.5-3.5-11.6-19.7-11.6-19.7-86.8 28.8 48.2 61.4 169.5 25.9a60.37 60.37 0 0 1-21.1-12.8z\"],invision:[448,512,[],\"f7b0\",\"M407.4 32H40.6C18.2 32 0 50.2 0 72.6v366.8C0 461.8 18.2 480 40.6 480h366.8c22.4 0 40.6-18.2 40.6-40.6V72.6c0-22.4-18.2-40.6-40.6-40.6zM176.1 145.6c.4 23.4-22.4 27.3-26.6 27.4-14.9 0-27.1-12-27.1-27 .1-35.2 53.1-35.5 53.7-.4zM332.8 377c-65.6 0-34.1-74-25-106.6 14.1-46.4-45.2-59-59.9.7l-25.8 103.3H177l8.1-32.5c-31.5 51.8-94.6 44.4-94.6-4.3.1-14.3.9-14 23-104.1H81.7l9.7-35.6h76.4c-33.6 133.7-32.6 126.9-32.9 138.2 0 20.9 40.9 13.5 57.4-23.2l19.8-79.4h-32.3l9.7-35.6h68.8l-8.9 40.5c40.5-75.5 127.9-47.8 101.8 38-14.2 51.1-14.6 50.7-14.9 58.8 0 15.5 17.5 22.6 31.8-16.9L386 325c-10.5 36.7-29.4 52-53.2 52z\"],\"creative-commons-pd-alt\":[496,512,[],\"f4ed\",\"M247.6 8C104.7 8 0 123.1 0 256c0 138.5 113.6 248 247.6 248C377.5 504 496 403.1 496 256 496 118.1 389.4 8 247.6 8zm.8 450.8c-112.5 0-203.7-93-203.7-202.8 0-105.4 85.5-203.3 203.7-203.3 112.6 0 202.9 89.5 202.8 203.3 0 121.7-99.6 202.8-202.8 202.8zM316.7 186h-53.2v137.2h53.2c21.4 0 70-5.1 70-68.6 0-63.4-48.6-68.6-70-68.6zm.8 108.5h-19.9v-79.7l19.4-.1c3.8 0 35-2.1 35 39.9 0 24.6-10.5 39.9-34.5 39.9zM203.7 186h-68.2v137.3h34.6V279h27c54.1 0 57.1-37.5 57.1-46.5 0-31-16.8-46.5-50.5-46.5zm-4.9 67.3h-29.2v-41.6h28.3c30.9 0 28.8 41.6.9 41.6z\"],centercode:[512,512,[],\"f380\",\"M329.2 268.6c-3.8 35.2-35.4 60.6-70.6 56.8-35.2-3.8-60.6-35.4-56.8-70.6 3.8-35.2 35.4-60.6 70.6-56.8 35.1 3.8 60.6 35.4 56.8 70.6zm-85.8 235.1C96.7 496-8.2 365.5 10.1 224.3c11.2-86.6 65.8-156.9 139.1-192 161-77.1 349.7 37.4 354.7 216.6 4.1 147-118.4 262.2-260.5 254.8zm179.9-180c27.9-118-160.5-205.9-237.2-234.2-57.5 56.3-69.1 188.6-33.8 344.4 68.8 15.8 169.1-26.4 271-110.2z\"],\"glide-g\":[448,512,[],\"f2a6\",\"M407.1 211.2c-3.5-1.4-11.6-3.8-15.4-3.8-37.1 0-62.2 16.8-93.5 34.5l-.9-.9c7-47.3 23.5-91.9 23.5-140.4C320.8 29.1 282.6 0 212.4 0 97.3 0 39 113.7 39 198.4 39 286.3 90.3 335 177.6 335c12 0 11-1 11 3.8-16.9 128.9-90.8 133.1-90.8 94.6 0-39.2 45-58.6 45.5-61-.3-12.2-47-27.6-58.9-27.6-33.9.1-52.4 51.2-52.4 79.3C32 476 64.8 512 117.5 512c77.4 0 134-77.8 151.4-145.4 15.1-60.5 11.2-63.3 19.7-67.6 32.2-16.2 57.5-27 93.8-27 17.8 0 30.5 3.7 58.9 8.4 2.9 0 6.7-2.9 6.7-5.8 0-8-33.4-60.5-40.9-63.4zm-175.3-84.4c-9.3 44.7-18.6 89.6-27.8 134.3-2.3 10.2-13.3 7.8-22 7.8-38.3 0-49-41.8-49-73.1 0-47 18-109.3 61.8-133.4 7-4.1 14.8-6.7 22.6-6.7 18.6 0 20 13.3 20 28.7-.1 14.3-2.7 28.5-5.6 42.4z\"],drupal:[448,512,[],\"f1a9\",\"M303.973,108.136C268.2,72.459,234.187,38.35,224.047,0c-9.957,38.35-44.25,72.459-80.019,108.136C90.467,161.7,29.716,222.356,29.716,313.436c-2.337,107.3,82.752,196.18,190.053,198.517S415.948,429.2,418.285,321.9q.091-4.231,0-8.464C418.285,222.356,357.534,161.7,303.973,108.136Zm-174.326,223a130.282,130.282,0,0,0-15.211,24.153,4.978,4.978,0,0,1-3.319,2.766h-1.659c-4.333,0-9.219-8.481-9.219-8.481h0c-1.29-2.028-2.489-4.149-3.687-6.361l-.83-1.752c-11.247-25.72-1.475-62.318-1.475-62.318h0a160.585,160.585,0,0,1,23.231-49.873A290.8,290.8,0,0,1,138.5,201.613l9.219,9.219,43.512,44.434a4.979,4.979,0,0,1,0,6.638L145.78,312.33h0Zm96.612,127.311a67.2,67.2,0,0,1-49.781-111.915c14.2-16.871,31.528-33.464,50.334-55.313,22.309,23.785,36.875,40.1,51.164,57.986a28.413,28.413,0,0,1,2.95,4.425,65.905,65.905,0,0,1,11.984,37.981,66.651,66.651,0,0,1-66.466,66.836ZM352.371,351.6h0a7.743,7.743,0,0,1-6.176,5.347H344.9a11.249,11.249,0,0,1-6.269-5.07h0a348.21,348.21,0,0,0-39.456-48.952L281.387,284.49,222.3,223.185a497.888,497.888,0,0,1-35.4-36.322,12.033,12.033,0,0,0-.922-1.382,35.4,35.4,0,0,1-4.7-9.219V174.51a31.346,31.346,0,0,1,9.218-27.656c11.432-11.431,22.955-22.954,33.833-34.939,11.984,13.275,24.8,26,37.428,38.627h0a530.991,530.991,0,0,1,69.6,79.1,147.494,147.494,0,0,1,27.011,83.8A134.109,134.109,0,0,1,352.371,351.6Z\"],jxl:[448,512,[],\"e67b\",\"M412.2 32H35.8C16 32 0 48 0 67.8V444.2C0 464 16 480 35.8 480H412.2c19.8 0 35.8-16 35.8-35.8V67.8C448 48 432 32 412.2 32zM378.6 333.7c0 40.2-32.6 72.8-72.8 72.8H70.2c0-40.2 32.6-72.8 72.8-72.8H378.6zm0-113.9c0 40.2-32.6 72.8-72.8 72.8H70.2c0-40.2 32.6-72.8 72.8-72.8H378.6zm0-113.9c0 40.2-32.6 72.8-72.8 72.8H70.2c0-40.2 32.6-72.8 72.8-72.8H378.6z\"],\"hire-a-helper\":[512,512,[],\"f3b0\",\"M443.1 0H71.9C67.9 37.3 37.4 67.8 0 71.7v371.5c37.4 4.9 66 32.4 71.9 68.8h372.2c3-36.4 32.5-65.8 67.9-69.8V71.7c-36.4-5.9-65-35.3-68.9-71.7zm-37 404.9c-36.3 0-18.8-2-55.1-2-35.8 0-21 2-56.1 2-5.9 0-4.9-8.2 0-9.8 22.8-7.6 22.9-10.2 24.6-12.8 10.4-15.6 5.9-83 5.9-113 0-5.3-6.4-12.8-13.8-12.8H200.4c-7.4 0-13.8 7.5-13.8 12.8 0 30-4.5 97.4 5.9 113 1.7 2.5 1.8 5.2 24.6 12.8 4.9 1.6 6 9.8 0 9.8-35.1 0-20.3-2-56.1-2-36.3 0-18.8 2-55.1 2-7.9 0-5.8-10.8 0-10.8 10.2-3.4 13.5-3.5 21.7-13.8 7.7-12.9 7.9-44.4 7.9-127.8V151.3c0-22.2-12.2-28.3-28.6-32.4-8.8-2.2-4-11.8 1-11.8 36.5 0 20.6 2 57.1 2 32.7 0 16.5-2 49.2-2 3.3 0 8.5 8.3 1 10.8-4.9 1.6-27.6 3.7-27.6 39.3 0 45.6-.2 55.8 1 68.8 0 1.3 2.3 12.8 12.8 12.8h109.2c10.5 0 12.8-11.5 12.8-12.8 1.2-13 1-23.2 1-68.8 0-35.6-22.7-37.7-27.6-39.3-7.5-2.5-2.3-10.8 1-10.8 32.7 0 16.5 2 49.2 2 36.5 0 20.6-2 57.1-2 4.9 0 9.9 9.6 1 11.8-16.4 4.1-28.6 10.3-28.6 32.4v101.2c0 83.4.1 114.9 7.9 127.8 8.2 10.2 11.4 10.4 21.7 13.8 5.8 0 7.8 10.8 0 10.8z\"],\"creative-commons-by\":[496,512,[],\"f4e7\",\"M314.9 194.4v101.4h-28.3v120.5h-77.1V295.9h-28.3V194.4c0-4.4 1.6-8.2 4.6-11.3 3.1-3.1 6.9-4.7 11.3-4.7H299c4.1 0 7.8 1.6 11.1 4.7 3.1 3.2 4.8 6.9 4.8 11.3zm-101.5-63.7c0-23.3 11.5-35 34.5-35s34.5 11.7 34.5 35c0 23-11.5 34.5-34.5 34.5s-34.5-11.5-34.5-34.5zM247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3z\"],unity:[448,512,[],\"e049\",\"M243.583 91.6027L323.695 138.384C326.575 140.026 326.68 144.583 323.695 146.225L228.503 201.854C225.623 203.55 222.22 203.444 219.549 201.854L124.357 146.225C121.425 144.636 121.373 139.973 124.357 138.384L204.417 91.6027V0L0 119.417V358.252L78.3843 312.477V218.914C78.3319 215.576 82.2066 213.192 85.0865 214.993L180.279 270.622C183.159 272.318 184.782 275.338 184.782 278.464V389.669C184.834 393.007 180.959 395.391 178.079 393.589L97.9673 346.808L19.583 392.583L224 512L428.417 392.583L350.033 346.808L269.921 393.589C267.093 395.338 263.114 393.06 263.218 389.669V278.464C263.218 275.126 265.051 272.159 267.721 270.622L362.914 214.993C365.741 213.245 369.72 215.47 369.616 218.914V312.477L448 358.252V119.417L243.583 0V91.6027Z\"],whmcs:[448,512,[],\"f40d\",\"M448 161v-21.3l-28.5-8.8-2.2-10.4 20.1-20.7L427 80.4l-29 7.5-7.2-7.5 7.5-28.2-19.1-11.6-21.3 21-10.7-3.2-7-26.4h-22.6l-6.2 26.4-12.1 3.2-19.7-21-19.4 11 8.1 27.7-8.1 8.4-28.5-7.5-11 19.1 20.7 21-2.9 10.4-28.5 7.8-.3 21.7 28.8 7.5 2.4 12.1-20.1 19.9 10.4 18.5 29.6-7.5 7.2 8.6-8.1 26.9 19.9 11.6 19.4-20.4 11.6 2.9 6.7 28.5 22.6.3 6.7-28.8 11.6-3.5 20.7 21.6 20.4-12.1-8.8-28 7.8-8.1 28.8 8.8 10.3-20.1-20.9-18.8 2.2-12.1 29.1-7zm-119.2 45.2c-31.3 0-56.8-25.4-56.8-56.8s25.4-56.8 56.8-56.8 56.8 25.4 56.8 56.8c0 31.5-25.4 56.8-56.8 56.8zm72.3 16.4l46.9 14.5V277l-55.1 13.4-4.1 22.7 38.9 35.3-19.2 37.9-54-16.7-14.6 15.2 16.7 52.5-38.3 22.7-38.9-40.5-21.7 6.6-12.6 54-42.4-.5-12.6-53.6-21.7-5.6-36.4 38.4-37.4-21.7 15.2-50.5-13.7-16.1-55.5 14.1-19.7-34.8 37.9-37.4-4.8-22.8-54-14.1.5-40.9L54 219.9l5.7-19.7-38.9-39.4L41.5 125l53.6 14.1 15.2-15.7-15.2-52 36.4-20.7 36.8 39.4L191 84l11.6-52H245l11.6 45.9L234 72l-6.3-1.7-3.3 5.7-11 19.1-3.3 5.6 4.6 4.6 17.2 17.4-.3 1-23.8 6.5-6.2 1.7-.1 6.4-.2 12.9C153.8 161.6 118 204 118 254.7c0 58.3 47.3 105.7 105.7 105.7 50.5 0 92.7-35.4 103.2-82.8l13.2.2 6.9.1 1.6-6.7 5.6-24 1.9-.6 17.1 17.8 4.7 4.9 5.8-3.4 20.4-12.1 5.8-3.5-2-6.5-6.8-21.2z\"],rocketchat:[576,512,[],\"f3e8\",\"M284.046,224.8a34.114,34.114,0,1,0,34.317,34.113A34.217,34.217,0,0,0,284.046,224.8Zm-110.45,0a34.114,34.114,0,1,0,34.317,34.113A34.217,34.217,0,0,0,173.6,224.8Zm220.923,0a34.114,34.114,0,1,0,34.317,34.113A34.215,34.215,0,0,0,394.519,224.8Zm153.807-55.319c-15.535-24.172-37.31-45.57-64.681-63.618-52.886-34.817-122.374-54-195.666-54a405.975,405.975,0,0,0-72.032,6.357,238.524,238.524,0,0,0-49.51-36.588C99.684-11.7,40.859.711,11.135,11.421A14.291,14.291,0,0,0,5.58,34.782C26.542,56.458,61.222,99.3,52.7,138.252c-33.142,33.9-51.112,74.776-51.112,117.337,0,43.372,17.97,84.248,51.112,118.148,8.526,38.956-26.154,81.816-47.116,103.491a14.284,14.284,0,0,0,5.555,23.34c29.724,10.709,88.549,23.147,155.324-10.2a238.679,238.679,0,0,0,49.51-36.589A405.972,405.972,0,0,0,288,460.14c73.313,0,142.8-19.159,195.667-53.975,27.371-18.049,49.145-39.426,64.679-63.619,17.309-26.923,26.07-55.916,26.07-86.125C574.394,225.4,565.634,196.43,548.326,169.485ZM284.987,409.9a345.65,345.65,0,0,1-89.446-11.5l-20.129,19.393a184.366,184.366,0,0,1-37.138,27.585,145.767,145.767,0,0,1-52.522,14.87c.983-1.771,1.881-3.563,2.842-5.356q30.258-55.68,16.325-100.078c-32.992-25.962-52.778-59.2-52.778-95.4,0-83.1,104.254-150.469,232.846-150.469s232.867,67.373,232.867,150.469C517.854,342.525,413.6,409.9,284.987,409.9Z\"],vk:[448,512,[],\"f189\",\"M31.4907 63.4907C0 94.9813 0 145.671 0 247.04V264.96C0 366.329 0 417.019 31.4907 448.509C62.9813 480 113.671 480 215.04 480H232.96C334.329 480 385.019 480 416.509 448.509C448 417.019 448 366.329 448 264.96V247.04C448 145.671 448 94.9813 416.509 63.4907C385.019 32 334.329 32 232.96 32H215.04C113.671 32 62.9813 32 31.4907 63.4907ZM75.6 168.267H126.747C128.427 253.76 166.133 289.973 196 297.44V168.267H244.16V242C273.653 238.827 304.64 205.227 315.093 168.267H363.253C359.313 187.435 351.46 205.583 340.186 221.579C328.913 237.574 314.461 251.071 297.733 261.227C316.41 270.499 332.907 283.63 346.132 299.751C359.357 315.873 369.01 334.618 374.453 354.747H321.44C316.555 337.262 306.614 321.61 292.865 309.754C279.117 297.899 262.173 290.368 244.16 288.107V354.747H238.373C136.267 354.747 78.0267 284.747 75.6 168.267Z\"],untappd:[640,512,[],\"f405\",\"M401.3 49.9c-79.8 160.1-84.6 152.5-87.9 173.2l-5.2 32.8c-1.9 12-6.6 23.5-13.7 33.4L145.6 497.1c-7.6 10.6-20.4 16.2-33.4 14.6-40.3-5-77.8-32.2-95.3-68.5-5.7-11.8-4.5-25.8 3.1-36.4l148.9-207.9c7.1-9.9 16.4-18 27.2-23.7l29.3-15.5c18.5-9.8 9.7-11.9 135.6-138.9 1-4.8 1-7.3 3.6-8 3-.7 6.6-1 6.3-4.6l-.4-4.6c-.2-1.9 1.3-3.6 3.2-3.6 4.5-.1 13.2 1.2 25.6 10 12.3 8.9 16.4 16.8 17.7 21.1.6 1.8-.6 3.7-2.4 4.2l-4.5 1.1c-3.4.9-2.5 4.4-2.3 7.4.1 2.8-2.3 3.6-6.5 6.1zM230.1 36.4c3.4.9 2.5 4.4 2.3 7.4-.2 2.7 2.1 3.5 6.4 6 7.9 15.9 15.3 30.5 22.2 44 .7 1.3 2.3 1.5 3.3.5 11.2-12 24.6-26.2 40.5-42.6 1.3-1.4 1.4-3.5.1-4.9-8-8.2-16.5-16.9-25.6-26.1-1-4.7-1-7.3-3.6-8-3-.8-6.6-1-6.3-4.6.3-3.3 1.4-8.1-2.8-8.2-4.5-.1-13.2 1.1-25.6 10-12.3 8.9-16.4 16.8-17.7 21.1-1.4 4.2 3.6 4.6 6.8 5.4zM620 406.7L471.2 198.8c-13.2-18.5-26.6-23.4-56.4-39.1-11.2-5.9-14.2-10.9-30.5-28.9-1-1.1-2.9-.9-3.6.5-46.3 88.8-47.1 82.8-49 94.8-1.7 10.7-1.3 20 .3 29.8 1.9 12 6.6 23.5 13.7 33.4l148.9 207.9c7.6 10.6 20.2 16.2 33.1 14.7 40.3-4.9 78-32 95.7-68.6 5.4-11.9 4.3-25.9-3.4-36.6z\"],mailchimp:[448,512,[],\"f59e\",\"M330.61 243.52a36.15 36.15 0 0 1 9.3 0c1.66-3.83 1.95-10.43.45-17.61-2.23-10.67-5.25-17.14-11.48-16.13s-6.47 8.74-4.24 19.42c1.26 6 3.49 11.14 6 14.32zM277.05 252c4.47 2 7.2 3.26 8.28 2.13 1.89-1.94-3.48-9.39-12.12-13.09a31.44 31.44 0 0 0-30.61 3.68c-3 2.18-5.81 5.22-5.41 7.06.85 3.74 10-2.71 22.6-3.48 7-.44 12.8 1.75 17.26 3.71zm-9 5.13c-9.07 1.42-15 6.53-13.47 10.1.9.34 1.17.81 5.21-.81a37 37 0 0 1 18.72-1.95c2.92.34 4.31.52 4.94-.49 1.46-2.22-5.71-8-15.39-6.85zm54.17 17.1c3.38-6.87-10.9-13.93-14.3-7s10.92 13.88 14.32 6.97zm15.66-20.47c-7.66-.13-7.95 15.8-.26 15.93s7.98-15.81.28-15.96zm-218.79 78.9c-1.32.31-6 1.45-8.47-2.35-5.2-8 11.11-20.38 3-35.77-9.1-17.47-27.82-13.54-35.05-5.54-8.71 9.6-8.72 23.54-5 24.08 4.27.57 4.08-6.47 7.38-11.63a12.83 12.83 0 0 1 17.85-3.72c11.59 7.59 1.37 17.76 2.28 28.62 1.39 16.68 18.42 16.37 21.58 9a2.08 2.08 0 0 0-.2-2.33c.03.89.68-1.3-3.35-.39zm299.72-17.07c-3.35-11.73-2.57-9.22-6.78-20.52 2.45-3.67 15.29-24-3.07-43.25-10.4-10.92-33.9-16.54-41.1-18.54-1.5-11.39 4.65-58.7-21.52-83 20.79-21.55 33.76-45.29 33.73-65.65-.06-39.16-48.15-51-107.42-26.47l-12.55 5.33c-.06-.05-22.71-22.27-23.05-22.57C169.5-18-41.77 216.81 25.78 273.85l14.76 12.51a72.49 72.49 0 0 0-4.1 33.5c3.36 33.4 36 60.42 67.53 60.38 57.73 133.06 267.9 133.28 322.29 3 1.74-4.47 9.11-24.61 9.11-42.38s-10.09-25.27-16.53-25.27zm-316 48.16c-22.82-.61-47.46-21.15-49.91-45.51-6.17-61.31 74.26-75.27 84-12.33 4.54 29.64-4.67 58.49-34.12 57.81zM84.3 249.55C69.14 252.5 55.78 261.09 47.6 273c-4.88-4.07-14-12-15.59-15-13.01-24.85 14.24-73 33.3-100.21C112.42 90.56 186.19 39.68 220.36 48.91c5.55 1.57 23.94 22.89 23.94 22.89s-34.15 18.94-65.8 45.35c-42.66 32.85-74.89 80.59-94.2 132.4zM323.18 350.7s-35.74 5.3-69.51-7.07c6.21-20.16 27 6.1 96.4-13.81 15.29-4.38 35.37-13 51-25.35a102.85 102.85 0 0 1 7.12 24.28c3.66-.66 14.25-.52 11.44 18.1-3.29 19.87-11.73 36-25.93 50.84A106.86 106.86 0 0 1 362.55 421a132.45 132.45 0 0 1-20.34 8.58c-53.51 17.48-108.3-1.74-126-43a66.33 66.33 0 0 1-3.55-9.74c-7.53-27.2-1.14-59.83 18.84-80.37 1.23-1.31 2.48-2.85 2.48-4.79a8.45 8.45 0 0 0-1.92-4.54c-7-10.13-31.19-27.4-26.33-60.83 3.5-24 24.49-40.91 44.07-39.91l5 .29c8.48.5 15.89 1.59 22.88 1.88 11.69.5 22.2-1.19 34.64-11.56 4.2-3.5 7.57-6.54 13.26-7.51a17.45 17.45 0 0 1 13.6 2.24c10 6.64 11.4 22.73 11.92 34.49.29 6.72 1.1 23 1.38 27.63.63 10.67 3.43 12.17 9.11 14 3.19 1.05 6.15 1.83 10.51 3.06 13.21 3.71 21 7.48 26 12.31a16.38 16.38 0 0 1 4.74 9.29c1.56 11.37-8.82 25.4-36.31 38.16-46.71 21.68-93.68 14.45-100.48 13.68-20.15-2.71-31.63 23.32-19.55 41.15 22.64 33.41 122.4 20 151.37-21.35.69-1 .12-1.59-.73-1-41.77 28.58-97.06 38.21-128.46 26-4.77-1.85-14.73-6.44-15.94-16.67 43.6 13.49 71 .74 71 .74s2.03-2.79-.56-2.53zm-68.47-5.7zm-83.4-187.5c16.74-19.35 37.36-36.18 55.83-45.63a.73.73 0 0 1 1 1c-1.46 2.66-4.29 8.34-5.19 12.65a.75.75 0 0 0 1.16.79c11.49-7.83 31.48-16.22 49-17.3a.77.77 0 0 1 .52 1.38 41.86 41.86 0 0 0-7.71 7.74.75.75 0 0 0 .59 1.19c12.31.09 29.66 4.4 41 10.74.76.43.22 1.91-.64 1.72-69.55-15.94-123.08 18.53-134.5 26.83a.76.76 0 0 1-1-1.12z\"],\"css3-alt\":[384,512,[],\"f38b\",\"M0 32l34.9 395.8L192 480l157.1-52.2L384 32H0zm313.1 80l-4.8 47.3L193 208.6l-.3.1h111.5l-12.8 146.6-98.2 28.7-98.8-29.2-6.4-73.9h48.9l3.2 38.3 52.6 13.3 54.7-15.4 3.7-61.6-166.3-.5v-.1l-.2.1-3.6-46.3L193.1 162l6.5-2.7H76.7L70.9 112h242.2z\"],\"square-reddit\":[448,512,[\"reddit-square\"],\"f1a2\",\"M64 32l320 0c35.3 0 64 28.7 64 64l0 320c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64L0 96C0 60.7 28.7 32 64 32zM305.9 166.4c20.6 0 37.3-16.7 37.3-37.3s-16.7-37.3-37.3-37.3c-18 0-33.1 12.8-36.6 29.8c-30.2 3.2-53.8 28.8-53.8 59.9l0 .2c-32.8 1.4-62.8 10.7-86.6 25.5c-8.8-6.8-19.9-10.9-32-10.9c-28.9 0-52.3 23.4-52.3 52.3c0 21 12.3 39 30.1 47.4c1.7 60.7 67.9 109.6 149.3 109.6s147.6-48.9 149.3-109.7c17.7-8.4 29.9-26.4 29.9-47.3c0-28.9-23.4-52.3-52.3-52.3c-12 0-23 4-31.9 10.8c-24-14.9-54.3-24.2-87.5-25.4l0-.1c0-22.2 16.5-40.7 37.9-43.7l0 0c3.9 16.5 18.7 28.7 36.3 28.7zM155 248.1c14.6 0 25.8 15.4 25 34.4s-11.8 25.9-26.5 25.9s-27.5-7.7-26.6-26.7s13.5-33.5 28.1-33.5zm166.4 33.5c.9 19-12 26.7-26.6 26.7s-25.6-6.9-26.5-25.9c-.9-19 10.3-34.4 25-34.4s27.3 14.6 28.1 33.5zm-42.1 49.6c-9 21.5-30.3 36.7-55.1 36.7s-46.1-15.1-55.1-36.7c-1.1-2.6 .7-5.4 3.4-5.7c16.1-1.6 33.5-2.5 51.7-2.5s35.6 .9 51.7 2.5c2.7 .3 4.5 3.1 3.4 5.7z\"],\"vimeo-v\":[448,512,[],\"f27d\",\"M447.8 153.6c-2 43.6-32.4 103.3-91.4 179.1-60.9 79.2-112.4 118.8-154.6 118.8-26.1 0-48.2-24.1-66.3-72.3C100.3 250 85.3 174.3 56.2 174.3c-3.4 0-15.1 7.1-35.2 21.1L0 168.2c51.6-45.3 100.9-95.7 131.8-98.5 34.9-3.4 56.3 20.5 64.4 71.5 28.7 181.5 41.4 208.9 93.6 126.7 18.7-29.6 28.8-52.1 30.2-67.6 4.8-45.9-35.8-42.8-63.3-31 22-72.1 64.1-107.1 126.2-105.1 45.8 1.2 67.5 31.1 64.9 89.4z\"],contao:[512,512,[],\"f26d\",\"M45.4 305c14.4 67.1 26.4 129 68.2 175H34c-18.7 0-34-15.2-34-34V66c0-18.7 15.2-34 34-34h57.7C77.9 44.6 65.6 59.2 54.8 75.6c-45.4 70-27 146.8-9.4 229.4zM478 32h-90.2c21.4 21.4 39.2 49.5 52.7 84.1l-137.1 29.3c-14.9-29-37.8-53.3-82.6-43.9-24.6 5.3-41 19.3-48.3 34.6-8.8 18.7-13.2 39.8 8.2 140.3 21.1 100.2 33.7 117.7 49.5 131.2 12.9 11.1 33.4 17 58.3 11.7 44.5-9.4 55.7-40.7 57.4-73.2l137.4-29.6c3.2 71.5-18.7 125.2-57.4 163.6H478c18.7 0 34-15.2 34-34V66c0-18.8-15.2-34-34-34z\"],\"square-font-awesome\":[448,512,[],\"e5ad\",\"M64 32C28.7 32 0 60.7 0 96L0 416c0 35.3 28.7 64 64 64l320 0c35.3 0 64-28.7 64-64l0-320c0-35.3-28.7-64-64-64L64 32zm90 122c0 9.3-4.8 17.4-12.1 22l188.9 0c7.3 0 13.2 5.9 13.2 13.2c0 1.8-.4 3.7-1.1 5.4L312 264l30.9 69.4c.7 1.7 1.1 3.5 1.1 5.4c0 7.3-5.9 13.2-13.2 13.2L144 352l0 32-32 0 0-32 0-176 0-1.5c-6.1-4.8-10-12.2-10-20.5c0-14.4 11.6-26 26-26s26 11.6 26 26z\"],deskpro:[480,512,[],\"f38f\",\"M205.9 512l31.1-38.4c12.3-.2 25.6-1.4 36.5-6.6 38.9-18.6 38.4-61.9 38.3-63.8-.1-5-.8-4.4-28.9-37.4H362c-.2 50.1-7.3 68.5-10.2 75.7-9.4 23.7-43.9 62.8-95.2 69.4-8.7 1.1-32.8 1.2-50.7 1.1zm200.4-167.7c38.6 0 58.5-13.6 73.7-30.9l-175.5-.3-17.4 31.3 119.2-.1zm-43.6-223.9v168.3h-73.5l-32.7 55.5H250c-52.3 0-58.1-56.5-58.3-58.9-1.2-13.2-21.3-11.6-20.1 1.8 1.4 15.8 8.8 40 26.4 57.1h-91c-25.5 0-110.8-26.8-107-114V16.9C0 .9 9.7.3 15 .1h82c.2 0 .3.1.5.1 4.3-.4 50.1-2.1 50.1 43.7 0 13.3 20.2 13.4 20.2 0 0-18.2-5.5-32.8-15.8-43.7h84.2c108.7-.4 126.5 79.4 126.5 120.2zm-132.5 56l64 29.3c13.3-45.5-42.2-71.7-64-29.3z\"],brave:[448,512,[],\"e63c\",\"M145.5 0H224h78.5l44.7 50.8s39.3-10.9 57.8 7.6s33.8 34.9 33.8 34.9l-12 29.5 15.3 43.7s-44.9 170.2-50.1 191c-10.4 40.9-17.4 56.8-46.9 77.5s-82.9 56.8-91.6 62.2c-1.9 1.2-3.9 2.5-5.9 3.9c-7.5 5.1-15.8 10.8-23.5 10.8l0 0 0 0c-7.7 0-16.1-5.7-23.5-10.8c-2-1.4-4-2.8-5.9-3.9c-8.7-5.5-62.1-41.5-91.6-62.2s-36.5-36.6-46.9-77.5c-5.3-20.8-50.1-191-50.1-191l15.3-43.7L9.2 93.3s15.3-16.4 33.8-34.9s57.8-7.6 57.8-7.6L145.5 0zM224 407.6l0 0c3.7 0 8.9-4.7 13-8.4c.6-.5 1.2-1.1 1.7-1.5c4.2-3.7 47.8-37.5 51-39.8s5.4-6.5 1.9-8.7c-2.8-1.7-10-5.5-20.3-10.8c-3-1.6-6.3-3.2-9.7-5c-15.4-8-34.5-14.7-37.5-14.7l0 0 0 0c-3 0-22.1 6.8-37.5 14.7c-3.5 1.8-6.7 3.5-9.7 5c-10.3 5.3-17.6 9.1-20.3 10.8c-3.6 2.2-1.4 6.4 1.9 8.7s46.8 36.1 51 39.8c.5 .5 1.1 1 1.7 1.5c4.1 3.7 9.3 8.4 13 8.4l0 0zm0-165.7l0 0c4.7 0 17.6-3 26.4-5l0 0 2-.5c7.8-1.8 7.3-6.3 6.4-13c-.1-.8-.2-1.6-.3-2.4c-.6-6.1-5.8-33.1-9.1-50.3c-1.1-5.8-2-10.5-2.4-12.9c-1.5-8.1-.6-9.4 .7-11.3c.2-.3 .5-.7 .7-1.1c1.4-2.3 16-6.2 27.9-9.5l0 0c2.5-.7 4.8-1.3 6.9-1.9c10.6-3 32.4-.6 44.2 .6c1.8 .2 3.4 .4 4.7 .5c9.6 .9 10.4 2.3 7.2 3.8c-2.3 1.1-16.2 6.3-28.7 10.9l0 0 0 0c-4.7 1.8-9.2 3.5-12.8 4.8c-1.5 .5-3 1.1-4.5 1.7c-12.5 4.6-27.2 10-28.9 19.4c-1.5 8.3 5.2 19.9 11.3 30.3l0 0c1.6 2.8 3.2 5.5 4.6 8.1c6.3 11.9 6.5 13.3 6.1 18.1c-.4 3.9-14.5 12.7-22.4 17.6l0 0c-1.8 1.1-3.3 2.1-4.2 2.7c-.8 .5-2.1 1.4-3.8 2.4c-8.6 5.2-26.3 16-26.3 22.5c0 7.8 24.6 28.1 32.4 33.2s28.9 16.1 37.9 17.8s23-8.5 31.2-23.8c7.7-14.4 1.7-28.5-3.2-40l-.9-2.2c-4.5-10.6 1.9-17 6.2-21.3l0 0c.5-.5 1-1 1.4-1.4L377.7 194c1.3-1.3 2.5-2.6 3.7-3.8l0 0c5.8-5.7 10.8-10.5 10.8-22.8c0-14.9-57.5-84.5-57.5-84.5s-48.5 9.3-55.1 9.3c-5.2 0-15.3-3.5-25.8-7.1l0 0c-2.7-.9-5.4-1.9-8-2.7C232.8 78.1 224 78 224 78l0 0 0 0s-8.7 0-21.8 4.4c-2.7 .9-5.4 1.8-8 2.7l0 0c-10.5 3.6-20.6 7.1-25.8 7.1c-6.5 0-55.1-9.3-55.1-9.3s-57.5 69.6-57.5 84.5c0 12.3 4.9 17.1 10.8 22.8l0 0c1.2 1.2 2.5 2.4 3.7 3.8l43.1 45.8c.4 .5 .9 .9 1.4 1.4l0 0c4.3 4.3 10.6 10.7 6.2 21.3l-.9 2.2c-4.9 11.5-11 25.6-3.2 40c8.2 15.3 22.2 25.5 31.2 23.8s30.1-12.7 37.9-17.8s32.4-25.4 32.4-33.2c0-6.5-17.7-17.3-26.3-22.5c-1.7-1-3.1-1.9-3.8-2.4c-.9-.6-2.4-1.5-4.2-2.7c-7.9-4.9-22-13.7-22.4-17.6c-.4-4.8-.3-6.2 6.1-18.1c1.3-2.5 2.9-5.3 4.6-8.1c6-10.4 12.8-22 11.3-30.3c-1.7-9.4-16.4-14.8-28.9-19.4c-1.6-.6-3.1-1.1-4.5-1.7c-3.6-1.4-8.1-3.1-12.8-4.8l-.1 0c-12.5-4.7-26.4-9.9-28.7-10.9c-3.2-1.5-2.3-2.8 7.2-3.8c1.3-.1 2.9-.3 4.7-.5c11.8-1.3 33.6-3.6 44.2-.6c2.1 .6 4.4 1.2 6.9 1.9c11.9 3.2 26.5 7.2 27.9 9.5c.2 .4 .5 .7 .7 1.1c1.3 1.9 2.2 3.2 .7 11.3c-.4 2.4-1.3 7.1-2.4 12.9c-3.3 17.2-8.5 44.2-9.1 50.3c-.1 .8-.2 1.7-.3 2.4c-.8 6.7-1.4 11.2 6.4 13l2 .5 0 0c8.8 2 21.8 5 26.4 5l0 0z\"],sistrix:[448,512,[],\"f3ee\",\"M448 449L301.2 300.2c20-27.9 31.9-62.2 31.9-99.2 0-93.1-74.7-168.9-166.5-168.9C74.7 32 0 107.8 0 200.9s74.7 168.9 166.5 168.9c39.8 0 76.3-14.2 105-37.9l146 148.1 30.5-31zM166.5 330.8c-70.6 0-128.1-58.3-128.1-129.9S95.9 71 166.5 71s128.1 58.3 128.1 129.9-57.4 129.9-128.1 129.9z\"],\"square-instagram\":[448,512,[\"instagram-square\"],\"e055\",\"M194.4 211.7a53.3 53.3 0 1 0 59.3 88.7 53.3 53.3 0 1 0 -59.3-88.7zm142.3-68.4c-5.2-5.2-11.5-9.3-18.4-12c-18.1-7.1-57.6-6.8-83.1-6.5c-4.1 0-7.9 .1-11.2 .1c-3.3 0-7.2 0-11.4-.1c-25.5-.3-64.8-.7-82.9 6.5c-6.9 2.7-13.1 6.8-18.4 12s-9.3 11.5-12 18.4c-7.1 18.1-6.7 57.7-6.5 83.2c0 4.1 .1 7.9 .1 11.1s0 7-.1 11.1c-.2 25.5-.6 65.1 6.5 83.2c2.7 6.9 6.8 13.1 12 18.4s11.5 9.3 18.4 12c18.1 7.1 57.6 6.8 83.1 6.5c4.1 0 7.9-.1 11.2-.1c3.3 0 7.2 0 11.4 .1c25.5 .3 64.8 .7 82.9-6.5c6.9-2.7 13.1-6.8 18.4-12s9.3-11.5 12-18.4c7.2-18 6.8-57.4 6.5-83c0-4.2-.1-8.1-.1-11.4s0-7.1 .1-11.4c.3-25.5 .7-64.9-6.5-83l0 0c-2.7-6.9-6.8-13.1-12-18.4zm-67.1 44.5A82 82 0 1 1 178.4 324.2a82 82 0 1 1 91.1-136.4zm29.2-1.3c-3.1-2.1-5.6-5.1-7.1-8.6s-1.8-7.3-1.1-11.1s2.6-7.1 5.2-9.8s6.1-4.5 9.8-5.2s7.6-.4 11.1 1.1s6.5 3.9 8.6 7s3.2 6.8 3.2 10.6c0 2.5-.5 5-1.4 7.3s-2.4 4.4-4.1 6.2s-3.9 3.2-6.2 4.2s-4.8 1.5-7.3 1.5l0 0c-3.8 0-7.5-1.1-10.6-3.2zM448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM357 389c-18.7 18.7-41.4 24.6-67 25.9c-26.4 1.5-105.6 1.5-132 0c-25.6-1.3-48.3-7.2-67-25.9s-24.6-41.4-25.8-67c-1.5-26.4-1.5-105.6 0-132c1.3-25.6 7.1-48.3 25.8-67s41.5-24.6 67-25.8c26.4-1.5 105.6-1.5 132 0c25.6 1.3 48.3 7.1 67 25.8s24.6 41.4 25.8 67c1.5 26.3 1.5 105.4 0 131.9c-1.3 25.6-7.1 48.3-25.8 67z\"],\"battle-net\":[512,512,[],\"f835\",\"M448.61 225.62c26.87.18 35.57-7.43 38.92-12.37 12.47-16.32-7.06-47.6-52.85-71.33 17.76-33.58 30.11-63.68 36.34-85.3 3.38-11.83 1.09-19 .45-20.25-1.72 10.52-15.85 48.46-48.2 100.05-25-11.22-56.52-20.1-93.77-23.8-8.94-16.94-34.88-63.86-60.48-88.93C252.18 7.14 238.7 1.07 228.18.22h-.05c-13.83-1.55-22.67 5.85-27.4 11-17.2 18.53-24.33 48.87-25 84.07-7.24-12.35-17.17-24.63-28.5-25.93h-.18c-20.66-3.48-38.39 29.22-36 81.29-38.36 1.38-71 5.75-93 11.23-9.9 2.45-16.22 7.27-17.76 9.72 1-.38 22.4-9.22 111.56-9.22 5.22 53 29.75 101.82 26 93.19-9.73 15.4-38.24 62.36-47.31 97.7-5.87 22.88-4.37 37.61.15 47.14 5.57 12.75 16.41 16.72 23.2 18.26 25 5.71 55.38-3.63 86.7-21.14-7.53 12.84-13.9 28.51-9.06 39.34 7.31 19.65 44.49 18.66 88.44-9.45 20.18 32.18 40.07 57.94 55.7 74.12a39.79 39.79 0 0 0 8.75 7.09c5.14 3.21 8.58 3.37 8.58 3.37-8.24-6.75-34-38-62.54-91.78 22.22-16 45.65-38.87 67.47-69.27 122.82 4.6 143.29-24.76 148-31.64 14.67-19.88 3.43-57.44-57.32-93.69zm-77.85 106.22c23.81-37.71 30.34-67.77 29.45-92.33 27.86 17.57 47.18 37.58 49.06 58.83 1.14 12.93-8.1 29.12-78.51 33.5zM216.9 387.69c9.76-6.23 19.53-13.12 29.2-20.49 6.68 13.33 13.6 26.1 20.6 38.19-40.6 21.86-68.84 12.76-49.8-17.7zm215-171.35c-10.29-5.34-21.16-10.34-32.38-15.05a722.459 722.459 0 0 0 22.74-36.9c39.06 24.1 45.9 53.18 9.64 51.95zM279.18 398c-5.51-11.35-11-23.5-16.5-36.44 43.25 1.27 62.42-18.73 63.28-20.41 0 .07-25 15.64-62.53 12.25a718.78 718.78 0 0 0 85.06-84q13.06-15.31 24.93-31.11c-.36-.29-1.54-3-16.51-12-51.7 60.27-102.34 98-132.75 115.92-20.59-11.18-40.84-31.78-55.71-61.49-20-39.92-30-82.39-31.57-116.07 12.3.91 25.27 2.17 38.85 3.88-22.29 36.8-14.39 63-13.47 64.23 0-.07-.95-29.17 20.14-59.57a695.23 695.23 0 0 0 44.67 152.84c.93-.38 1.84.88 18.67-8.25-26.33-74.47-33.76-138.17-34-173.43 20-12.42 48.18-19.8 81.63-17.81 44.57 2.67 86.36 15.25 116.32 30.71q-10.69 15.66-23.33 32.47C365.63 152 339.1 145.84 337.5 146c.11 0 25.9 14.07 41.52 47.22a717.63 717.63 0 0 0-115.34-31.71 646.608 646.608 0 0 0-39.39-6.05c-.07.45-1.81 1.85-2.16 20.33C300 190.28 358.78 215.68 389.36 233c.74 23.55-6.95 51.61-25.41 79.57-24.6 37.31-56.39 67.23-84.77 85.43zm27.4-287c-44.56-1.66-73.58 7.43-94.69 20.67 2-52.3 21.31-76.38 38.21-75.28C267 52.15 305 108.55 306.58 111zm-130.65 3.1c.48 12.11 1.59 24.62 3.21 37.28-14.55-.85-28.74-1.25-42.4-1.26-.08 3.24-.12-51 24.67-49.59h.09c5.76 1.09 10.63 6.88 14.43 13.57zm-28.06 162c20.76 39.7 43.3 60.57 65.25 72.31-46.79 24.76-77.53 20-84.92 4.51-.2-.21-11.13-15.3 19.67-76.81zm210.06 74.8\"],\"the-red-yeti\":[512,512,[],\"f69d\",\"M488.23 241.7l20.7 7.1c-9.6-23.9-23.9-37-31.7-44.8l7.1-18.2c.2 0 12.3-27.8-2.5-30.7-.6-11.3-6.6-27-18.4-27-7.6-10.6-17.7-12.3-30.7-5.9a122.2 122.2 0 0 0-25.3 16.5c-5.3-6.4-3 .4-3-29.8-37.1-24.3-45.4-11.7-74.8 3l.5.5a239.36 239.36 0 0 0-68.4-13.3c-5.5-8.7-18.6-19.1-25.1-25.1l24.8 7.1c-5.5-5.5-26.8-12.9-34.2-15.2 18.2-4.1 29.8-20.8 42.5-33-34.9-10.1-67.9-5.9-97.9 11.8l12-44.2L182 0c-31.6 24.2-33 41.9-33.7 45.5-.9-2.4-6.3-19.6-15.2-27a35.12 35.12 0 0 0-.5 25.3c3 8.4 5.9 14.8 8.4 18.9-16-3.3-28.3-4.9-49.2 0h-3.7l33 14.3a194.26 194.26 0 0 0-46.7 67.4l-1.7 8.4 1.7 1.7 7.6-4.7c-3.3 11.6-5.3 19.4-6.6 25.8a200.18 200.18 0 0 0-27.8 40.3c-15 1-31.8 10.8-40.3 14.3l3 3.4 28.8 1c-.5 1-.7 2.2-1.2 3.2-7.3 6.4-39.8 37.7-33 80.7l20.2-22.4c.5 1.7.7 3.4 1.2 5.2 0 25.5.4 89.6 64.9 150.5 43.6 40 96 60.2 157.5 60.2 121.7 0 223-87.3 223-211.5 6.8-9.7-1.2 3 16.7-25.1l13 14.3 2.5-.5A181.84 181.84 0 0 0 495 255a44.74 44.74 0 0 0-6.8-13.3zM398 111.2l-.5 21.9c5.5 18.1 16.9 17.2 22.4 17.2l-3.4-4.7 22.4-5.4a242.44 242.44 0 0 1-27 0c12.8-2.1 33.3-29 43-11.3 3.4 7.6 6.4 17.2 9.3 27.8l1.7-5.9a56.38 56.38 0 0 1-1.7-15.2c5.4.5 8.8 3.4 9.3 10.1.5 6.4 1.7 14.8 3.4 25.3l4.7-11.3c4.6 0 4.5-3.6-2.5 20.7-20.9-8.7-35.1-8.4-46.5-8.4l18.2-16c-25.3 8.2-33 10.8-54.8 20.9-1.1-5.4-5-13.5-16-19.9-3.2 3.8-2.8.9-.7 14.8h-2.5a62.32 62.32 0 0 0-8.4-23.1l4.2-3.4c8.4-7.1 11.8-14.3 10.6-21.9-.5-6.4-5.4-13.5-13.5-20.7 5.6-3.4 15.2-.4 28.3 8.5zm-39.6-10.1c2.7 1.9 11.4 5.4 18.9 17.2 4.2 8.4 4 9.8 3.4 11.1-.5 2.4-.5 4.3-3 7.1-1.7 2.5-5.4 4.7-11.8 7.6-7.6-13-16.5-23.6-27.8-31.2zM91 143.1l1.2-1.7c1.2-2.9 4.2-7.6 9.3-15.2l2.5-3.4-13 12.3 5.4-4.7-10.1 9.3-4.2 1.2c12.3-24.1 23.1-41.3 32.5-50.2 9.3-9.3 16-16 20.2-19.4l-6.4 1.2c-11.3-4.2-19.4-7.1-24.8-8.4 2.5-.5 3.7-.5 3.2-.5 10.3 0 17.5.5 20.9 1.2a52.35 52.35 0 0 0 16 2.5l.5-1.7-8.4-35.8 13.5 29a42.89 42.89 0 0 0 5.9-14.3c1.7-6.4 5.4-13 10.1-19.4s7.6-10.6 9.3-11.3a234.68 234.68 0 0 0-6.4 25.3l-1.7 7.1-.5 4.7 2.5 2.5C190.4 39.9 214 34 239.8 34.5l21.1.5c-11.8 13.5-27.8 21.9-48.5 24.8a201.26 201.26 0 0 1-23.4 2.9l-.2-.5-2.5-1.2a20.75 20.75 0 0 0-14 2c-2.5-.2-4.9-.5-7.1-.7l-2.5 1.7.5 1.2c2 .2 3.9.5 6.2.7l-2 3.4 3.4-.5-10.6 11.3c-4.2 3-5.4 6.4-4.2 9.3l5.4-3.4h1.2a39.4 39.4 0 0 1 25.3-15.2v-3c6.4.5 13 1 19.4 1.2 6.4 0 8.4.5 5.4 1.2a189.6 189.6 0 0 1 20.7 13.5c13.5 10.1 23.6 21.9 30 35.4 8.8 18.2 13.5 37.1 13.5 56.6a141.13 141.13 0 0 1-3 28.3 209.91 209.91 0 0 1-16 46l2.5.5c18.2-19.7 41.9-16 49.2-16l-6.4 5.9 22.4 17.7-1.7 30.7c-5.4-12.3-16.5-21.1-33-27.8 16.5 14.8 23.6 21.1 21.9 20.2-4.8-2.8-3.5-1.9-10.8-3.7 4.1 4.1 17.5 18.8 18.2 20.7l.2.2-.2.2c0 1.8 1.6-1.2-14 22.9-75.2-15.3-106.27-42.7-141.2-63.2l11.8 1.2c-11.8-18.5-15.6-17.7-38.4-26.1L149 225c-8.8-3-18.2-3-28.3.5l7.6-10.6-1.2-1.7c-14.9 4.3-19.8 9.2-22.6 11.3-1.1-5.5-2.8-12.4-12.3-28.8l-1.2 27-13.2-5c1.5-25.2 5.4-50.5 13.2-74.6zm276.5 330c-49.9 25-56.1 22.4-59 23.9-29.8-11.8-50.9-31.7-63.5-58.8l30 16.5c-9.8-9.3-18.3-16.5-38.4-44.3l11.8 23.1-17.7-7.6c14.2 21.1 23.5 51.7 66.6 73.5-120.8 24.2-199-72.1-200.9-74.3a262.57 262.57 0 0 0 35.4 24.8c3.4 1.7 7.1 2.5 10.1 1.2l-16-20.7c9.2 4.2 9.5 4.5 69.1 29-42.5-20.7-73.8-40.8-93.2-60.2-.5 6.4-1.2 10.1-1.2 10.1a80.25 80.25 0 0 1 20.7 26.6c-39-18.9-57.6-47.6-71.3-82.6 49.9 55.1 118.9 37.5 120.5 37.1 34.8 16.4 69.9 23.6 113.9 10.6 3.3 0 20.3 17 25.3 39.1l4.2-3-2.5-23.6c9 9 24.9 22.6 34.4 13-15.6-5.3-23.5-9.5-29.5-31.7 4.6 4.2 7.6 9 27.8 15l1.2-1.2-10.5-14.2c11.7-4.8-3.5 1 32-10.8 4.3 34.3 9 49.2.7 89.5zm115.3-214.4l-2.5.5 3 9.3c-3.5 5.9-23.7 44.3-71.6 79.7-39.5 29.8-76.6 39.1-80.9 40.3l-7.6-7.1-1.2 3 14.3 16-7.1-4.7 3.4 4.2h-1.2l-21.9-13.5 9.3 26.6-19-27.9-1.2 2.5 7.6 29c-6.1-8.2-21-32.6-56.8-39.6l32.5 21.2a214.82 214.82 0 0 1-93.2-6.4c-4.2-1.2-8.9-2.5-13.5-4.2l1.2-3-44.8-22.4 26.1 22.4c-57.7 9.1-113-25.4-126.4-83.4l-2.5-16.4-22.27 22.3c19.5-57.5 25.6-57.9 51.4-70.1-9.1-5.3-1.6-3.3-38.4-9.3 15.8-5.8 33-15.4 73 5.2a18.5 18.5 0 0 1 3.7-1.7c.6-3.2.4-.8 1-11.8 3.9 10 3.6 8.7 3 9.3l1.7.5c12.7-6.5 8.9-4.5 17-8.9l-5.4 13.5 22.3-5.8-8.4 8.4 2.5 2.5c4.5-1.8 30.3 3.4 40.8 16l-23.6-2.5c39.4 23 51.5 54 55.8 69.6l1.7-1.2c-2.8-22.3-12.4-33.9-16-40.1 4.2 5 39.2 34.6 110.4 46-11.3-.5-23.1 5.4-34.9 18.9l46.7-20.2-9.3 21.9c7.6-10.1 14.8-23.6 21.2-39.6v-.5l1.2-3-1.2 16c13.5-41.8 25.3-78.5 35.4-109.7l13.5-27.8v-2l-5.4-4.2h10.1l5.9 4.2 2.5-1.2-3.4-16 12.3 18.9 41.8-20.2-14.8 13 .5 2.9 17.7-.5a184 184 0 0 1 33 4.2l-23.6 2.5-1.2 3 26.6 23.1a254.21 254.21 0 0 1 27 32c-11.2-3.3-10.3-3.4-21.2-3.4l12.3 32.5zm-6.1-71.3l-3.9 13-14.3-11.8zm-254.8 7.1c1.7 10.6 4.7 17.7 8.8 21.9-9.3 6.6-27.5 13.9-46.5 16l.5 1.2a50.22 50.22 0 0 0 24.8-2.5l-7.1 13c4.2-1.7 10.1-7.1 17.7-14.8 11.9-5.5 12.7-5.1 20.2-16-12.7-6.4-15.7-13.7-18.4-18.8zm3.7-102.3c-6.4-3.4-10.6 3-12.3 18.9s2.5 29.5 11.8 39.6 18.2 10.6 26.1 3 3.4-23.6-11.3-47.7a39.57 39.57 0 0 0-14.27-13.8zm-4.7 46.3c5.4 2.2 10.5 1.9 12.3-10.6v-4.7l-1.2.5c-4.3-3.1-2.5-4.5-1.7-6.2l.5-.5c-.9-1.2-5-8.1-12.5 4.7-.5-13.5.5-21.9 3-24.8 1.2-2.5 4.7-1.2 11.3 4.2 6.4 5.4 11.3 16 15.2 32.5 6.5 28-19.8 26.2-26.9 4.9zm-45-5.5c1.6.3 9.3-1.1 9.3-14.8h-.5c-5.4-1.1-2.2-5.5-.7-5.9-1.7-3-3.4-4.2-5.4-4.7-8.1 0-11.6 12.7-8.1 21.2a7.51 7.51 0 0 0 5.43 4.2zM216 82.9l-2.5.5.5 3a48.94 48.94 0 0 1 26.1 5.9c-2.5-5.5-10-14.3-28.3-14.3l.5 2.5zm-71.8 49.4c21.7 16.8 16.5 21.4 46.5 23.6l-2.9-4.7a42.67 42.67 0 0 0 14.8-28.3c1.7-16-1.2-29.5-8.8-41.3l13-7.6a2.26 2.26 0 0 0-.5-1.7 14.21 14.21 0 0 0-13.5 1.7c-12.7 6.7-28 20.9-29 22.4-1.7 1.7-3.4 5.9-5.4 13.5a99.61 99.61 0 0 0-2.9 23.6c-4.7-8-10.5-6.4-19.9-5.9l7.1 7.6c-16.5 0-23.3 15.4-23.6 16 6.8 0 4.6-7.6 30-12.3-4.3-6.3-3.3-5-4.9-6.6zm18.7-18.7c1.2-7.6 3.4-13 6.4-17.2 5.4-6.4 10.6-10.1 16-11.8 4.2-1.7 7.1 1.2 10.1 9.3a72.14 72.14 0 0 1 3 25.3c-.5 9.3-3.4 17.2-8.4 23.1-2.9 3.4-5.4 5.9-6.4 7.6a39.21 39.21 0 0 1-11.3-.5l-7.1-3.4-5.4-6.4c.8-10 1.3-18.8 3.1-26zm42 56.1c-34.8 14.4-34.7 14-36.1 14.3-20.8 4.7-19-24.4-18.9-24.8l5.9-1.2-.5-2.5c-20.2-2.6-31 4.2-32.5 4.9.5.5 3 3.4 5.9 9.3 4.2-6.4 8.8-10.1 15.2-10.6a83.47 83.47 0 0 0 1.7 33.7c.1.5 2.6 17.4 27.5 24.1 11.3 3 27 1.2 48.9-5.4l-9.2.5c-4.2-14.8-6.4-24.8-5.9-29.5 11.3-8.8 21.9-11.3 30.7-7.6h2.5l-11.8-7.6-7.1.5c-5.9 1.2-12.3 4.2-19.4 8.4z\"],\"square-hacker-news\":[448,512,[\"hacker-news-square\"],\"f3af\",\"M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM21 229.2c0 0 .1-.1 .1-.2s.1-.2 .2-.2c0 .1 0 .3-.1 .4H21zM239.2 384H207.8V281.3L128 128h37.3c41.5 77.7 48.1 95.8 54.1 112c1.6 4.3 3.1 8.5 5.2 13.6c3.2-7 5.1-11.9 7.1-17.3c5.9-15.3 12.8-33.2 53.5-108.3H320L239.2 283.1V384z\"],edge:[512,512,[],\"f282\",\"M120.1 37.44C161.1 12.23 207.7-.7753 255 .0016C423 .0016 512 123.8 512 219.5C511.9 252.2 499 283.4 476.1 306.7C453.2 329.9 422.1 343.2 389.4 343.7C314.2 343.7 297.9 320.6 297.9 311.7C297.9 307.9 299.1 305.5 302.7 302.3L303.7 301.1L304.1 299.5C314.6 288 320 273.3 320 257.9C320 179.2 237.8 115.2 136 115.2C98.46 114.9 61.46 124.1 28.48 142.1C55.48 84.58 111.2 44.5 119.8 38.28C120.6 37.73 120.1 37.44 120.1 37.44V37.44zM135.7 355.5C134.3 385.5 140.3 415.5 152.1 442.7C165.7 469.1 184.8 493.7 208.6 512C149.1 500.5 97.11 468.1 59.2 422.7C21.12 376.3 0 318.4 0 257.9C0 206.7 62.4 163.5 136 163.5C172.6 162.9 208.4 174.4 237.8 196.2L234.2 197.4C182.7 215 135.7 288.1 135.7 355.5V355.5zM469.8 400L469.1 400.1C457.3 418.9 443.2 435.2 426.9 449.6C396.1 477.6 358.8 495.1 318.1 499.5C299.5 499.8 281.3 496.3 264.3 488.1C238.7 477.8 217.2 458.1 202.7 435.1C188.3 411.2 181.6 383.4 183.7 355.5C183.1 335.4 189.1 315.2 198.7 297.3C212.6 330.4 236.2 358.6 266.3 378.1C296.4 397.6 331.8 407.6 367.7 406.7C398.7 407 429.8 400 457.9 386.2L459.8 385.3C463.7 383 467.5 381.4 471.4 385.3C475.9 390.2 473.2 394.5 470.2 399.3C470 399.5 469.9 399.8 469.8 400V400z\"],threads:[448,512,[],\"e618\",\"M331.5 235.7c2.2 .9 4.2 1.9 6.3 2.8c29.2 14.1 50.6 35.2 61.8 61.4c15.7 36.5 17.2 95.8-30.3 143.2c-36.2 36.2-80.3 52.5-142.6 53h-.3c-70.2-.5-124.1-24.1-160.4-70.2c-32.3-41-48.9-98.1-49.5-169.6V256v-.2C17 184.3 33.6 127.2 65.9 86.2C102.2 40.1 156.2 16.5 226.4 16h.3c70.3 .5 124.9 24 162.3 69.9c18.4 22.7 32 50 40.6 81.7l-40.4 10.8c-7.1-25.8-17.8-47.8-32.2-65.4c-29.2-35.8-73-54.2-130.5-54.6c-57 .5-100.1 18.8-128.2 54.4C72.1 146.1 58.5 194.3 58 256c.5 61.7 14.1 109.9 40.3 143.3c28 35.6 71.2 53.9 128.2 54.4c51.4-.4 85.4-12.6 113.7-40.9c32.3-32.2 31.7-71.8 21.4-95.9c-6.1-14.2-17.1-26-31.9-34.9c-3.7 26.9-11.8 48.3-24.7 64.8c-17.1 21.8-41.4 33.6-72.7 35.3c-23.6 1.3-46.3-4.4-63.9-16c-20.8-13.8-33-34.8-34.3-59.3c-2.5-48.3 35.7-83 95.2-86.4c21.1-1.2 40.9-.3 59.2 2.8c-2.4-14.8-7.3-26.6-14.6-35.2c-10-11.7-25.6-17.7-46.2-17.8H227c-16.6 0-39 4.6-53.3 26.3l-34.4-23.6c19.2-29.1 50.3-45.1 87.8-45.1h.8c62.6 .4 99.9 39.5 103.7 107.7l-.2 .2zm-156 68.8c1.3 25.1 28.4 36.8 54.6 35.3c25.6-1.4 54.6-11.4 59.5-73.2c-13.2-2.9-27.8-4.4-43.4-4.4c-4.8 0-9.6 .1-14.4 .4c-42.9 2.4-57.2 23.2-56.2 41.8l-.1 .1z\"],napster:[496,512,[],\"f3d2\",\"M298.3 373.6c-14.2 13.6-31.3 24.1-50.4 30.5-19-6.4-36.2-16.9-50.3-30.5h100.7zm44-199.6c20-16.9 43.6-29.2 69.6-36.2V299c0 219.4-328 217.6-328 .3V137.7c25.9 6.9 49.6 19.6 69.5 36.4 56.8-40 132.5-39.9 188.9-.1zm-208.8-58.5c64.4-60 164.3-60.1 228.9-.2-7.1 3.5-13.9 7.3-20.6 11.5-58.7-30.5-129.2-30.4-187.9.1-6.3-4-13.9-8.2-20.4-11.4zM43.8 93.2v69.3c-58.4 36.5-58.4 121.1.1 158.3 26.4 245.1 381.7 240.3 407.6 1.5l.3-1.7c58.7-36.3 58.9-121.7.2-158.2V93.2c-17.3.5-34 3-50.1 7.4-82-91.5-225.5-91.5-307.5.1-16.3-4.4-33.1-7-50.6-7.5zM259.2 352s36-.3 61.3-1.5c10.2-.5 21.1-4 25.5-6.5 26.3-15.1 25.4-39.2 26.2-47.4-79.5-.6-99.9-3.9-113 55.4zm-135.5-55.3c.8 8.2-.1 32.3 26.2 47.4 4.4 2.5 15.2 6 25.5 6.5 25.3 1.1 61.3 1.5 61.3 1.5-13.2-59.4-33.7-56.1-113-55.4zm169.1 123.4c-3.2-5.3-6.9-7.3-6.9-7.3-24.8 7.3-52.2 6.9-75.9 0 0 0-2.9 1.5-6.4 6.6-2.8 4.1-3.7 9.6-3.7 9.6 29.1 17.6 67.1 17.6 96.2 0-.1-.1-.3-4-3.3-8.9z\"],\"square-snapchat\":[448,512,[\"snapchat-square\"],\"f2ad\",\"M384,32H64A64,64,0,0,0,0,96V416a64,64,0,0,0,64,64H384a64,64,0,0,0,64-64V96A64,64,0,0,0,384,32Zm-3.907,319.309-.083.1a32.364,32.364,0,0,1-8.717,6.823,90.26,90.26,0,0,1-20.586,8.2,12.694,12.694,0,0,0-3.852,1.76c-2.158,1.909-2.1,4.64-4.4,8.55a23.137,23.137,0,0,1-6.84,7.471c-6.707,4.632-14.244,4.923-22.23,5.23-7.214.274-15.39.581-24.729,3.669-3.761,1.245-7.753,3.694-12.377,6.533-11.265,6.9-26.68,16.353-52.3,16.353s-40.925-9.4-52.106-16.279c-4.657-2.888-8.675-5.362-12.543-6.64-9.339-3.08-17.516-3.4-24.729-3.67-7.986-.307-15.523-.6-22.231-5.229a23.085,23.085,0,0,1-6.01-6.11c-3.2-4.632-2.855-7.8-5.254-9.895a13.428,13.428,0,0,0-4.1-1.834,89.986,89.986,0,0,1-20.313-8.127,32.905,32.905,0,0,1-8.3-6.284c-6.583-6.757-8.276-14.776-5.686-21.824,3.436-9.338,11.571-12.111,19.4-16.262,14.776-8.027,26.348-18.055,34.433-29.884a68.236,68.236,0,0,0,5.985-10.567c.789-2.158.772-3.329.241-4.416a7.386,7.386,0,0,0-2.208-2.217c-2.532-1.676-5.113-3.353-6.882-4.5-3.27-2.141-5.868-3.818-7.529-4.98-6.267-4.383-10.65-9.04-13.4-14.245a28.4,28.4,0,0,1-1.369-23.584c4.134-10.924,14.469-17.706,26.978-17.706a37.141,37.141,0,0,1,7.845.83c.689.15,1.37.307,2.042.482-.108-7.43.058-15.357.722-23.119,2.358-27.261,11.912-41.589,21.874-52.994a86.836,86.836,0,0,1,22.28-17.931C188.254,100.383,205.312,96,224,96s35.828,4.383,50.944,13.016a87.169,87.169,0,0,1,22.239,17.9c9.961,11.406,19.516,25.709,21.874,52.995a231.194,231.194,0,0,1,.713,23.118c.673-.174,1.362-.332,2.051-.481a37.131,37.131,0,0,1,7.844-.83c12.5,0,22.82,6.782,26.971,17.706a28.37,28.37,0,0,1-1.4,23.559c-2.74,5.2-7.123,9.861-13.39,14.244-1.668,1.187-4.258,2.864-7.529,4.981-1.835,1.187-4.541,2.947-7.164,4.682a6.856,6.856,0,0,0-1.951,2.034c-.506,1.046-.539,2.191.166,4.208a69.015,69.015,0,0,0,6.085,10.792c8.268,12.1,20.188,22.313,35.454,30.407,1.486.772,2.98,1.5,4.441,2.258.722.332,1.569.763,2.491,1.3,4.9,2.723,9.2,6.01,11.455,12.153C387.821,336.915,386.269,344.7,380.093,351.309Zm-16.719-18.461c-50.313-24.314-58.332-61.918-58.689-64.749-.431-3.379-.921-6.035,2.806-9.472,3.594-3.328,19.541-13.19,23.965-16.278,7.33-5.114,10.534-10.219,8.16-16.495-1.66-4.316-5.686-5.976-9.961-5.976a18.5,18.5,0,0,0-3.993.448c-8.035,1.743-15.838,5.769-20.354,6.857a7.1,7.1,0,0,1-1.66.224c-2.408,0-3.279-1.071-3.088-3.968.564-8.783,1.759-25.925.373-41.937-1.884-22.032-8.99-32.948-17.432-42.6-4.051-4.624-23.135-24.654-59.536-24.654S168.53,134.359,164.479,139c-8.434,9.654-15.531,20.57-17.432,42.6-1.386,16.013-.141,33.147.373,41.937.166,2.756-.68,3.968-3.088,3.968a7.1,7.1,0,0,1-1.66-.224c-4.507-1.087-12.31-5.113-20.346-6.856a18.494,18.494,0,0,0-3.993-.449c-4.25,0-8.3,1.636-9.961,5.977-2.374,6.276.847,11.381,8.168,16.494,4.425,3.088,20.371,12.958,23.966,16.279,3.719,3.437,3.237,6.093,2.805,9.471-.356,2.79-8.384,40.394-58.689,64.749-2.946,1.428-7.96,4.45.88,9.331,13.88,7.628,23.111,6.807,30.3,11.43,6.093,3.927,2.5,12.394,6.923,15.449,5.454,3.76,21.583-.266,42.335,6.6,17.433,5.744,28.116,22.015,58.963,22.015s41.788-16.3,58.938-21.973c20.795-6.865,36.89-2.839,42.336-6.6,4.433-3.055.822-11.522,6.923-15.448,7.181-4.624,16.411-3.8,30.3-11.472C371.36,337.355,366.346,334.333,363.374,332.848Z\"],\"google-plus-g\":[640,512,[],\"f0d5\",\"M386.061 228.496c1.834 9.692 3.143 19.384 3.143 31.956C389.204 370.205 315.599 448 204.8 448c-106.084 0-192-85.915-192-192s85.916-192 192-192c51.864 0 95.083 18.859 128.611 50.292l-52.126 50.03c-14.145-13.621-39.028-29.599-76.485-29.599-65.484 0-118.92 54.221-118.92 121.277 0 67.056 53.436 121.277 118.92 121.277 75.961 0 104.513-54.745 108.965-82.773H204.8v-66.009h181.261zm185.406 6.437V179.2h-56.001v55.733h-55.733v56.001h55.733v55.733h56.001v-55.733H627.2v-56.001h-55.733z\"],artstation:[512,512,[],\"f77a\",\"M2 377.4l43 74.3A51.35 51.35 0 0 0 90.9 480h285.4l-59.2-102.6zM501.8 350L335.6 59.3A51.38 51.38 0 0 0 290.2 32h-88.4l257.3 447.6 40.7-70.5c1.9-3.2 21-29.7 2-59.1zM275 304.5l-115.5-200L44 304.5z\"],markdown:[640,512,[],\"f60f\",\"M593.8 59.1H46.2C20.7 59.1 0 79.8 0 105.2v301.5c0 25.5 20.7 46.2 46.2 46.2h547.7c25.5 0 46.2-20.7 46.1-46.1V105.2c0-25.4-20.7-46.1-46.2-46.1zM338.5 360.6H277v-120l-61.5 76.9-61.5-76.9v120H92.3V151.4h61.5l61.5 76.9 61.5-76.9h61.5v209.2zm135.3 3.1L381.5 256H443V151.4h61.5V256H566z\"],sourcetree:[448,512,[],\"f7d3\",\"M427.2 203c0-112.1-90.9-203-203-203C112.1-.2 21.2 90.6 21 202.6A202.86 202.86 0 0 0 161.5 396v101.7a14.3 14.3 0 0 0 14.3 14.3h96.4a14.3 14.3 0 0 0 14.3-14.3V396.1A203.18 203.18 0 0 0 427.2 203zm-271.6 0c0-90.8 137.3-90.8 137.3 0-.1 89.9-137.3 91-137.3 0z\"],\"google-plus\":[512,512,[],\"f2b3\",\"M256,8C119.1,8,8,119.1,8,256S119.1,504,256,504,504,392.9,504,256,392.9,8,256,8ZM185.3,380a124,124,0,0,1,0-248c31.3,0,60.1,11,83,32.3l-33.6,32.6c-13.2-12.9-31.3-19.1-49.4-19.1-42.9,0-77.2,35.5-77.2,78.1S142.3,334,185.3,334c32.6,0,64.9-19.1,70.1-53.3H185.3V238.1H302.2a109.2,109.2,0,0,1,1.9,20.7c0,70.8-47.5,121.2-118.8,121.2ZM415.5,273.8v35.5H380V273.8H344.5V238.3H380V202.8h35.5v35.5h35.2v35.5Z\"],diaspora:[512,512,[],\"f791\",\"M251.64 354.55c-1.4 0-88 119.9-88.7 119.9S76.34 414 76 413.25s86.6-125.7 86.6-127.4c0-2.2-129.6-44-137.6-47.1-1.3-.5 31.4-101.8 31.7-102.1.6-.7 144.4 47 145.5 47 .4 0 .9-.6 1-1.3.4-2 1-148.6 1.7-149.6.8-1.2 104.5-.7 105.1-.3 1.5 1 3.5 156.1 6.1 156.1 1.4 0 138.7-47 139.3-46.3.8.9 31.9 102.2 31.5 102.6-.9.9-140.2 47.1-140.6 48.8-.3 1.4 82.8 122.1 82.5 122.9s-85.5 63.5-86.3 63.5c-1-.2-89-125.5-90.9-125.5z\"],foursquare:[368,512,[],\"f180\",\"M323.1 3H49.9C12.4 3 0 31.3 0 49.1v433.8c0 20.3 12.1 27.7 18.2 30.1 6.2 2.5 22.8 4.6 32.9-7.1C180 356.5 182.2 354 182.2 354c3.1-3.4 3.4-3.1 6.8-3.1h83.4c35.1 0 40.6-25.2 44.3-39.7l48.6-243C373.8 25.8 363.1 3 323.1 3zm-16.3 73.8l-11.4 59.7c-1.2 6.5-9.5 13.2-16.9 13.2H172.1c-12 0-20.6 8.3-20.6 20.3v13c0 12 8.6 20.6 20.6 20.6h90.4c8.3 0 16.6 9.2 14.8 18.2-1.8 8.9-10.5 53.8-11.4 58.8-.9 4.9-6.8 13.5-16.9 13.5h-73.5c-13.5 0-17.2 1.8-26.5 12.6 0 0-8.9 11.4-89.5 108.3-.9.9-1.8.6-1.8-.3V75.9c0-7.7 6.8-16.6 16.6-16.6h219c8.2 0 15.6 7.7 13.5 17.5z\"],\"stack-overflow\":[384,512,[],\"f16c\",\"M290.7 311L95 269.7 86.8 309l195.7 41zm51-87L188.2 95.7l-25.5 30.8 153.5 128.3zm-31.2 39.7L129.2 179l-16.7 36.5L293.7 300zM262 32l-32 24 119.3 160.3 32-24zm20.5 328h-200v39.7h200zm39.7 80H42.7V320h-40v160h359.5V320h-40z\"],\"github-alt\":[480,512,[],\"f113\",\"M186.1 328.7c0 20.9-10.9 55.1-36.7 55.1s-36.7-34.2-36.7-55.1 10.9-55.1 36.7-55.1 36.7 34.2 36.7 55.1zM480 278.2c0 31.9-3.2 65.7-17.5 95-37.9 76.6-142.1 74.8-216.7 74.8-75.8 0-186.2 2.7-225.6-74.8-14.6-29-20.2-63.1-20.2-95 0-41.9 13.9-81.5 41.5-113.6-5.2-15.8-7.7-32.4-7.7-48.8 0-21.5 4.9-32.3 14.6-51.8 45.3 0 74.3 9 108.8 36 29-6.9 58.8-10 88.7-10 27 0 54.2 2.9 80.4 9.2 34-26.7 63-35.2 107.8-35.2 9.8 19.5 14.6 30.3 14.6 51.8 0 16.4-2.6 32.7-7.7 48.2 27.5 32.4 39 72.3 39 114.2zm-64.3 50.5c0-43.9-26.7-82.6-73.5-82.6-18.9 0-37 3.4-56 6-14.9 2.3-29.8 3.2-45.1 3.2-15.2 0-30.1-.9-45.1-3.2-18.7-2.6-37-6-56-6-46.8 0-73.5 38.7-73.5 82.6 0 87.8 80.4 101.3 150.4 101.3h48.2c70.3 0 150.6-13.4 150.6-101.3zm-82.6-55.1c-25.8 0-36.7 34.2-36.7 55.1s10.9 55.1 36.7 55.1 36.7-34.2 36.7-55.1-10.9-55.1-36.7-55.1z\"],\"phoenix-squadron\":[512,512,[],\"f511\",\"M96 63.38C142.49 27.25 201.55 7.31 260.51 8.81c29.58-.38 59.11 5.37 86.91 15.33-24.13-4.63-49-6.34-73.38-2.45C231.17 27 191 48.84 162.21 80.87c5.67-1 10.78-3.67 16-5.86 18.14-7.87 37.49-13.26 57.23-14.83 19.74-2.13 39.64-.43 59.28 1.92-14.42 2.79-29.12 4.57-43 9.59-34.43 11.07-65.27 33.16-86.3 62.63-13.8 19.71-23.63 42.86-24.67 67.13-.35 16.49 5.22 34.81 19.83 44a53.27 53.27 0 0 0 37.52 6.74c15.45-2.46 30.07-8.64 43.6-16.33 11.52-6.82 22.67-14.55 32-24.25 3.79-3.22 2.53-8.45 2.62-12.79-2.12-.34-4.38-1.11-6.3.3a203 203 0 0 1-35.82 15.37c-20 6.17-42.16 8.46-62.1.78 12.79 1.73 26.06.31 37.74-5.44 20.23-9.72 36.81-25.2 54.44-38.77a526.57 526.57 0 0 1 88.9-55.31c25.71-12 52.94-22.78 81.57-24.12-15.63 13.72-32.15 26.52-46.78 41.38-14.51 14-27.46 29.5-40.11 45.18-3.52 4.6-8.95 6.94-13.58 10.16a150.7 150.7 0 0 0-51.89 60.1c-9.33 19.68-14.5 41.85-11.77 63.65 1.94 13.69 8.71 27.59 20.9 34.91 12.9 8 29.05 8.07 43.48 5.1 32.8-7.45 61.43-28.89 81-55.84 20.44-27.52 30.52-62.2 29.16-96.35-.52-7.5-1.57-15-1.66-22.49 8 19.48 14.82 39.71 16.65 60.83 2 14.28.75 28.76-1.62 42.9-1.91 11-5.67 21.51-7.78 32.43a165 165 0 0 0 39.34-81.07 183.64 183.64 0 0 0-14.21-104.64c20.78 32 32.34 69.58 35.71 107.48.49 12.73.49 25.51 0 38.23A243.21 243.21 0 0 1 482 371.34c-26.12 47.34-68 85.63-117.19 108-78.29 36.23-174.68 31.32-248-14.68A248.34 248.34 0 0 1 25.36 366 238.34 238.34 0 0 1 0 273.08v-31.34C3.93 172 40.87 105.82 96 63.38m222 80.33a79.13 79.13 0 0 0 16-4.48c5-1.77 9.24-5.94 10.32-11.22-8.96 4.99-17.98 9.92-26.32 15.7z\"],pagelines:[384,512,[],\"f18c\",\"M384 312.7c-55.1 136.7-187.1 54-187.1 54-40.5 81.8-107.4 134.4-184.6 134.7-16.1 0-16.6-24.4 0-24.4 64.4-.3 120.5-42.7 157.2-110.1-41.1 15.9-118.6 27.9-161.6-82.2 109-44.9 159.1 11.2 178.3 45.5 9.9-24.4 17-50.9 21.6-79.7 0 0-139.7 21.9-149.5-98.1 119.1-47.9 152.6 76.7 152.6 76.7 1.6-16.7 3.3-52.6 3.3-53.4 0 0-106.3-73.7-38.1-165.2 124.6 43 61.4 162.4 61.4 162.4.5 1.6.5 23.8 0 33.4 0 0 45.2-89 136.4-57.5-4.2 134-141.9 106.4-141.9 106.4-4.4 27.4-11.2 53.4-20 77.5 0 0 83-91.8 172-20z\"],algolia:[512,512,[],\"f36c\",\"M256 0C116.1 0 2 112.7 0 252.1C-2 393.6 112.9 510.8 254.5 511.6c43.7 .3 85.9-10.4 123.3-30.7c3.6-2 4.2-7 1.1-9.7l-24-21.2c-4.9-4.3-11.8-5.5-17.8-3c-26.1 11.1-54.5 16.8-83.7 16.4C139 461.9 46.5 366.8 48.3 252.4C50.1 139.5 142.6 48.2 256 48.2H463.7V417.2L345.9 312.5c-3.8-3.4-9.7-2.7-12.7 1.3c-18.9 25-49.7 40.6-83.9 38.2c-47.5-3.3-85.9-41.5-89.5-88.9c-4.2-56.6 40.6-103.9 96.3-103.9c50.4 0 91.9 38.8 96.2 88c.4 4.4 2.4 8.5 5.7 11.4l30.7 27.2c3.5 3.1 9 1.2 9.9-3.4c2.2-11.8 3-24.2 2.1-36.8c-4.9-72-63.3-130-135.4-134.4c-82.7-5.1-151.8 59.5-154 140.6c-2.1 78.9 62.6 147 141.6 148.7c33 .7 63.6-9.6 88.3-27.6L495 509.4c6.6 5.8 17 1.2 17-7.7V9.7c0-5.4-4.4-9.7-9.7-9.7H256z\"],\"red-river\":[448,512,[],\"f3e3\",\"M353.2 32H94.8C42.4 32 0 74.4 0 126.8v258.4C0 437.6 42.4 480 94.8 480h258.4c52.4 0 94.8-42.4 94.8-94.8V126.8c0-52.4-42.4-94.8-94.8-94.8zM144.9 200.9v56.3c0 27-21.9 48.9-48.9 48.9V151.9c0-13.2 10.7-23.9 23.9-23.9h154.2c0 27-21.9 48.9-48.9 48.9h-56.3c-12.3-.6-24.6 11.6-24 24zm176.3 72h-56.3c-12.3-.6-24.6 11.6-24 24v56.3c0 27-21.9 48.9-48.9 48.9V247.9c0-13.2 10.7-23.9 23.9-23.9h154.2c0 27-21.9 48.9-48.9 48.9z\"],\"creative-commons-sa\":[496,512,[],\"f4ef\",\"M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zM137.7 221c13-83.9 80.5-95.7 108.9-95.7 99.8 0 127.5 82.5 127.5 134.2 0 63.6-41 132.9-128.9 132.9-38.9 0-99.1-20-109.4-97h62.5c1.5 30.1 19.6 45.2 54.5 45.2 23.3 0 58-18.2 58-82.8 0-82.5-49.1-80.6-56.7-80.6-33.1 0-51.7 14.6-55.8 43.8h18.2l-49.2 49.2-49-49.2h19.4z\"],safari:[512,512,[],\"f267\",\"M274.69,274.69l-37.38-37.38L166,346ZM256,8C119,8,8,119,8,256S119,504,256,504,504,393,504,256,393,8,256,8ZM411.85,182.79l14.78-6.13A8,8,0,0,1,437.08,181h0a8,8,0,0,1-4.33,10.46L418,197.57a8,8,0,0,1-10.45-4.33h0A8,8,0,0,1,411.85,182.79ZM314.43,94l6.12-14.78A8,8,0,0,1,331,74.92h0a8,8,0,0,1,4.33,10.45l-6.13,14.78a8,8,0,0,1-10.45,4.33h0A8,8,0,0,1,314.43,94ZM256,60h0a8,8,0,0,1,8,8V84a8,8,0,0,1-8,8h0a8,8,0,0,1-8-8V68A8,8,0,0,1,256,60ZM181,74.92a8,8,0,0,1,10.46,4.33L197.57,94a8,8,0,1,1-14.78,6.12l-6.13-14.78A8,8,0,0,1,181,74.92Zm-63.58,42.49h0a8,8,0,0,1,11.31,0L140,128.72A8,8,0,0,1,140,140h0a8,8,0,0,1-11.31,0l-11.31-11.31A8,8,0,0,1,117.41,117.41ZM60,256h0a8,8,0,0,1,8-8H84a8,8,0,0,1,8,8h0a8,8,0,0,1-8,8H68A8,8,0,0,1,60,256Zm40.15,73.21-14.78,6.13A8,8,0,0,1,74.92,331h0a8,8,0,0,1,4.33-10.46L94,314.43a8,8,0,0,1,10.45,4.33h0A8,8,0,0,1,100.15,329.21Zm4.33-136h0A8,8,0,0,1,94,197.57l-14.78-6.12A8,8,0,0,1,74.92,181h0a8,8,0,0,1,10.45-4.33l14.78,6.13A8,8,0,0,1,104.48,193.24ZM197.57,418l-6.12,14.78a8,8,0,0,1-14.79-6.12l6.13-14.78A8,8,0,1,1,197.57,418ZM264,444a8,8,0,0,1-8,8h0a8,8,0,0,1-8-8V428a8,8,0,0,1,8-8h0a8,8,0,0,1,8,8Zm67-6.92h0a8,8,0,0,1-10.46-4.33L314.43,418a8,8,0,0,1,4.33-10.45h0a8,8,0,0,1,10.45,4.33l6.13,14.78A8,8,0,0,1,331,437.08Zm63.58-42.49h0a8,8,0,0,1-11.31,0L372,383.28A8,8,0,0,1,372,372h0a8,8,0,0,1,11.31,0l11.31,11.31A8,8,0,0,1,394.59,394.59ZM286.25,286.25,110.34,401.66,225.75,225.75,401.66,110.34ZM437.08,331h0a8,8,0,0,1-10.45,4.33l-14.78-6.13a8,8,0,0,1-4.33-10.45h0A8,8,0,0,1,418,314.43l14.78,6.12A8,8,0,0,1,437.08,331ZM444,264H428a8,8,0,0,1-8-8h0a8,8,0,0,1,8-8h16a8,8,0,0,1,8,8h0A8,8,0,0,1,444,264Z\"],google:[488,512,[],\"f1a0\",\"M488 261.8C488 403.3 391.1 504 248 504 110.8 504 0 393.2 0 256S110.8 8 248 8c66.8 0 123 24.5 166.3 64.9l-67.5 64.9C258.5 52.6 94.3 116.6 94.3 256c0 86.5 69.1 156.6 153.7 156.6 98.2 0 135-70.4 140.8-106.9H248v-85.3h236.1c2.3 12.7 3.9 24.9 3.9 41.4z\"],\"square-font-awesome-stroke\":[448,512,[\"font-awesome-alt\"],\"f35c\",\"M64 64C46.3 64 32 78.3 32 96l0 320c0 17.7 14.3 32 32 32l320 0c17.7 0 32-14.3 32-32l0-320c0-17.7-14.3-32-32-32L64 64zM0 96C0 60.7 28.7 32 64 32l320 0c35.3 0 64 28.7 64 64l0 320c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64L0 96zm154 58c0 9.3-4.8 17.4-12.1 22l188.9 0c7.3 0 13.2 5.9 13.2 13.2c0 1.8-.4 3.7-1.1 5.4L312 264l30.9 69.4c.8 1.7 1.1 3.5 1.1 5.4c0 7.3-5.9 13.2-13.2 13.2L144 352l0 32-32 0 0-32 0-176 0-1.5c-6.1-4.8-10-12.2-10-20.5c0-14.4 11.6-26 26-26s26 11.6 26 26z\"],atlassian:[512,512,[],\"f77b\",\"M152.2 236.4c-7.7-8.2-19.7-7.7-24.8 2.8L1.6 490.2c-5 10 2.4 21.7 13.4 21.7h175c5.8.1 11-3.2 13.4-8.4 37.9-77.8 15.1-196.3-51.2-267.1zM244.4 8.1c-122.3 193.4-8.5 348.6 65 495.5 2.5 5.1 7.7 8.4 13.4 8.4H497c11.2 0 18.4-11.8 13.4-21.7 0 0-234.5-470.6-240.4-482.3-5.3-10.6-18.8-10.8-25.6.1z\"],\"linkedin-in\":[448,512,[],\"f0e1\",\"M100.28 448H7.4V148.9h92.88zM53.79 108.1C24.09 108.1 0 83.5 0 53.8a53.79 53.79 0 0 1 107.58 0c0 29.7-24.1 54.3-53.79 54.3zM447.9 448h-92.68V302.4c0-34.7-.7-79.2-48.29-79.2-48.29 0-55.69 37.7-55.69 76.7V448h-92.78V148.9h89.08v40.8h1.3c12.4-23.5 42.69-48.3 87.88-48.3 94 0 111.28 61.9 111.28 142.3V448z\"],\"digital-ocean\":[512,512,[],\"f391\",\"M87 481.8h73.7v-73.6H87zM25.4 346.6v61.6H87v-61.6zm466.2-169.7c-23-74.2-82.4-133.3-156.6-156.6C164.9-32.8 8 93.7 8 255.9h95.8c0-101.8 101-180.5 208.1-141.7 39.7 14.3 71.5 46.1 85.8 85.7 39.1 107-39.7 207.8-141.4 208v.3h-.3V504c162.6 0 288.8-156.8 235.6-327.1zm-235.3 231v-95.3h-95.6v95.6H256v-.3z\"],nimblr:[384,512,[],\"f5a8\",\"M246.6 299.29c15.57 0 27.15 11.46 27.15 27s-11.62 27-27.15 27c-15.7 0-27.15-11.57-27.15-27s11.55-27 27.15-27zM113 326.25c0-15.61 11.68-27 27.15-27s27.15 11.46 27.15 27-11.47 27-27.15 27c-15.44 0-27.15-11.31-27.15-27M191.76 159C157 159 89.45 178.77 59.25 227L14 0v335.48C14 433.13 93.61 512 191.76 512s177.76-78.95 177.76-176.52S290.13 159 191.76 159zm0 308.12c-73.27 0-132.51-58.9-132.51-131.59s59.24-131.59 132.51-131.59 132.51 58.86 132.51 131.54S265 467.07 191.76 467.07z\"],chromecast:[512,512,[],\"f838\",\"M447.8,64H64c-23.6,0-42.7,19.1-42.7,42.7v63.9H64v-63.9h383.8v298.6H298.6V448H448c23.6,0,42.7-19.1,42.7-42.7V106.7 C490.7,83.1,471.4,64,447.8,64z M21.3,383.6L21.3,383.6l0,63.9h63.9C85.2,412.2,56.6,383.6,21.3,383.6L21.3,383.6z M21.3,298.6V341 c58.9,0,106.6,48.1,106.6,107h42.7C170.7,365.6,103.7,298.7,21.3,298.6z M213.4,448h42.7c-0.5-129.5-105.3-234.3-234.8-234.6l0,42.4 C127.3,255.6,213.3,342,213.4,448z\"],evernote:[384,512,[],\"f839\",\"M120.82 132.21c1.6 22.31-17.55 21.59-21.61 21.59-68.93 0-73.64-1-83.58 3.34-.56.22-.74 0-.37-.37L123.79 46.45c.38-.37.6-.22.38.37-4.35 9.99-3.35 15.09-3.35 85.39zm79 308c-14.68-37.08 13-76.93 52.52-76.62 17.49 0 22.6 23.21 7.95 31.42-6.19 3.3-24.95 1.74-25.14 19.2-.05 17.09 19.67 25 31.2 24.89A45.64 45.64 0 0 0 312 393.45v-.08c0-11.63-7.79-47.22-47.54-55.34-7.72-1.54-65-6.35-68.35-50.52-3.74 16.93-17.4 63.49-43.11 69.09-8.74 1.94-69.68 7.64-112.92-36.77 0 0-18.57-15.23-28.23-57.95-3.38-15.75-9.28-39.7-11.14-62 0-18 11.14-30.45 25.07-32.2 81 0 90 2.32 101-7.8 9.82-9.24 7.8-15.5 7.8-102.78 1-8.3 7.79-30.81 53.41-24.14 6 .86 31.91 4.18 37.48 30.64l64.26 11.15c20.43 3.71 70.94 7 80.6 57.94 22.66 121.09 8.91 238.46 7.8 238.46C362.15 485.53 267.06 480 267.06 480c-18.95-.23-54.25-9.4-67.27-39.83zm80.94-204.84c-1 1.92-2.2 6 .85 7 14.09 4.93 39.75 6.84 45.88 5.53 3.11-.25 3.05-4.43 2.48-6.65-3.53-21.85-40.83-26.5-49.24-5.92z\"],\"hacker-news\":[448,512,[],\"f1d4\",\"M0 32v448h448V32H0zm21.2 197.2H21c.1-.1.2-.3.3-.4 0 .1 0 .3-.1.4zm218 53.9V384h-31.4V281.3L128 128h37.3c52.5 98.3 49.2 101.2 59.3 125.6 12.3-27 5.8-24.4 60.6-125.6H320l-80.8 155.1z\"],\"creative-commons-sampling\":[496,512,[],\"f4f0\",\"M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zm3.6 53.2c2.8-.3 11.5 1 11.5 11.5l6.6 107.2 4.9-59.3c0-6 4.7-10.6 10.6-10.6 5.9 0 10.6 4.7 10.6 10.6 0 2.5-.5-5.7 5.7 81.5l5.8-64.2c.3-2.9 2.9-9.3 10.2-9.3 3.8 0 9.9 2.3 10.6 8.9l11.5 96.5 5.3-12.8c1.8-4.4 5.2-6.6 10.2-6.6h58v21.3h-50.9l-18.2 44.3c-3.9 9.9-19.5 9.1-20.8-3.1l-4-31.9-7.5 92.6c-.3 3-3 9.3-10.2 9.3-3 0-9.8-2.1-10.6-9.3 0-1.9.6 5.8-6.2-77.9l-5.3 72.2c-1.1 4.8-4.8 9.3-10.6 9.3-2.9 0-9.8-2-10.6-9.3 0-1.9.5 6.7-5.8-87.7l-5.8 94.8c0 6.3-3.6 12.4-10.6 12.4-5.2 0-10.6-4.1-10.6-12l-5.8-87.7c-5.8 92.5-5.3 84-5.3 85.9-1.1 4.8-4.8 9.3-10.6 9.3-3 0-9.8-2.1-10.6-9.3 0-.7-.4-1.1-.4-2.6l-6.2-88.6L182 348c-.7 6.5-6.7 9.3-10.6 9.3-5.8 0-9.6-4.1-10.6-8.9L149.7 272c-2 4-3.5 8.4-11.1 8.4H87.2v-21.3H132l13.7-27.9c4.4-9.9 18.2-7.2 19.9 2.7l3.1 20.4 8.4-97.9c0-6 4.8-10.6 10.6-10.6.5 0 10.6-.2 10.6 12.4l4.9 69.1 6.6-92.6c0-10.1 9.5-10.6 10.2-10.6.6 0 10.6.7 10.6 10.6l5.3 80.6 6.2-97.9c.1-1.1-.6-10.3 9.9-11.5z\"],adversal:[512,512,[],\"f36a\",\"M482.1 32H28.7C5.8 32 0 37.9 0 60.9v390.2C0 474.4 5.8 480 28.7 480h453.4c24.4 0 29.9-5.2 29.9-29.7V62.2c0-24.6-5.4-30.2-29.9-30.2zM178.4 220.3c-27.5-20.2-72.1-8.7-84.2 23.4-4.3 11.1-9.3 9.5-17.5 8.3-9.7-1.5-17.2-3.2-22.5-5.5-28.8-11.4 8.6-55.3 24.9-64.3 41.1-21.4 83.4-22.2 125.3-4.8 40.9 16.8 34.5 59.2 34.5 128.5 2.7 25.8-4.3 58.3 9.3 88.8 1.9 4.4.4 7.9-2.7 10.7-8.4 6.7-39.3 2.2-46.6-7.4-1.9-2.2-1.8-3.6-3.9-6.2-3.6-3.9-7.3-2.2-11.9 1-57.4 36.4-140.3 21.4-147-43.3-3.1-29.3 12.4-57.1 39.6-71 38.2-19.5 112.2-11.8 114-30.9 1.1-10.2-1.9-20.1-11.3-27.3zm286.7 222c0 15.1-11.1 9.9-17.8 9.9H52.4c-7.4 0-18.2 4.8-17.8-10.7.4-13.9 10.5-9.1 17.1-9.1 132.3-.4 264.5-.4 396.8 0 6.8 0 16.6-4.4 16.6 9.9zm3.8-340.5v291c0 5.7-.7 13.9-8.1 13.9-12.4-.4-27.5 7.1-36.1-5.6-5.8-8.7-7.8-4-12.4-1.2-53.4 29.7-128.1 7.1-144.4-85.2-6.1-33.4-.7-67.1 15.7-100 11.8-23.9 56.9-76.1 136.1-30.5v-71c0-26.2-.1-26.2 26-26.2 3.1 0 6.6.4 9.7 0 10.1-.8 13.6 4.4 13.6 14.3-.1.2-.1.3-.1.5zm-51.5 232.3c-19.5 47.6-72.9 43.3-90 5.2-15.1-33.3-15.5-68.2.4-101.5 16.3-34.1 59.7-35.7 81.5-4.8 20.6 28.8 14.9 84.6 8.1 101.1zm-294.8 35.3c-7.5-1.3-33-3.3-33.7-27.8-.4-13.9 7.8-23 19.8-25.8 24.4-5.9 49.3-9.9 73.7-14.7 8.9-2 7.4 4.4 7.8 9.5 1.4 33-26.1 59.2-67.6 58.8z\"],\"creative-commons\":[496,512,[],\"f25e\",\"M245.83 214.87l-33.22 17.28c-9.43-19.58-25.24-19.93-27.46-19.93-22.13 0-33.22 14.61-33.22 43.84 0 23.57 9.21 43.84 33.22 43.84 14.47 0 24.65-7.09 30.57-21.26l30.55 15.5c-6.17 11.51-25.69 38.98-65.1 38.98-22.6 0-73.96-10.32-73.96-77.05 0-58.69 43-77.06 72.63-77.06 30.72-.01 52.7 11.95 65.99 35.86zm143.05 0l-32.78 17.28c-9.5-19.77-25.72-19.93-27.9-19.93-22.14 0-33.22 14.61-33.22 43.84 0 23.55 9.23 43.84 33.22 43.84 14.45 0 24.65-7.09 30.54-21.26l31 15.5c-2.1 3.75-21.39 38.98-65.09 38.98-22.69 0-73.96-9.87-73.96-77.05 0-58.67 42.97-77.06 72.63-77.06 30.71-.01 52.58 11.95 65.56 35.86zM247.56 8.05C104.74 8.05 0 123.11 0 256.05c0 138.49 113.6 248 247.56 248 129.93 0 248.44-100.87 248.44-248 0-137.87-106.62-248-248.44-248zm.87 450.81c-112.54 0-203.7-93.04-203.7-202.81 0-105.42 85.43-203.27 203.72-203.27 112.53 0 202.82 89.46 202.82 203.26-.01 121.69-99.68 202.82-202.84 202.82z\"],\"watchman-monitoring\":[512,512,[],\"e087\",\"M256,16C123.452,16,16,123.452,16,256S123.452,496,256,496,496,388.548,496,256,388.548,16,256,16ZM121.69,429.122C70.056,388.972,36.741,326.322,36.741,256a218.519,218.519,0,0,1,9.587-64.122l102.9-17.895-.121,10.967-13.943,2.013s-.144,12.5-.144,19.549a12.778,12.778,0,0,0,4.887,10.349l9.468,7.4Zm105.692-283.27,8.48-7.618s6.934-5.38-.143-9.344c-7.188-4.024-39.53-34.5-39.53-34.5-5.348-5.477-8.257-7.347-15.46,0,0,0-32.342,30.474-39.529,34.5-7.078,3.964-.144,9.344-.144,9.344l8.481,7.618-.048,4.369L75.982,131.045c39.644-56.938,105.532-94.3,180.018-94.3A218.754,218.754,0,0,1,420.934,111.77l-193.512,37.7Zm34.063,329.269-33.9-250.857,9.467-7.4a12.778,12.778,0,0,0,4.888-10.349c0-7.044-.144-19.549-.144-19.549l-13.943-2.013-.116-10.474,241.711,31.391A218.872,218.872,0,0,1,475.259,256C475.259,375.074,379.831,472.212,261.445,475.121Z\"],fonticons:[448,512,[],\"f280\",\"M0 32v448h448V32zm187 140.9c-18.4 0-19 9.9-19 27.4v23.3c0 2.4-3.5 4.4-.6 4.4h67.4l-11.1 37.3H168v112.9c0 5.8-2 6.7 3.2 7.3l43.5 4.1v25.1H84V389l21.3-2c5.2-.6 6.7-2.3 6.7-7.9V267.7c0-2.3-2.9-2.3-5.8-2.3H84V228h28v-21c0-49.6 26.5-70 77.3-70 34.1 0 64.7 8.2 64.7 52.8l-50.7 6.1c.3-18.7-4.4-23-16.3-23zm74.3 241.8v-25.1l20.4-2.6c5.2-.6 7.6-1.7 7.6-7.3V271.8c0-4.1-2.9-6.7-6.7-7.9l-24.2-6.4 6.7-29.5h80.2v151.7c0 5.8-2.6 6.4 2.9 7.3l15.7 2.6v25.1zm80.8-255.5l9 33.2-7.3 7.3-31.2-16.6-31.2 16.6-7.3-7.3 9-33.2-21.8-24.2 3.5-9.6h27.7l15.5-28h9.3l15.5 28h27.7l3.5 9.6z\"],weixin:[576,512,[],\"f1d7\",\"M385.2 167.6c6.4 0 12.6.3 18.8 1.1C387.4 90.3 303.3 32 207.7 32 100.5 32 13 104.8 13 197.4c0 53.4 29.3 97.5 77.9 131.6l-19.3 58.6 68-34.1c24.4 4.8 43.8 9.7 68.2 9.7 6.2 0 12.1-.3 18.3-.8-4-12.9-6.2-26.6-6.2-40.8-.1-84.9 72.9-154 165.3-154zm-104.5-52.9c14.5 0 24.2 9.7 24.2 24.4 0 14.5-9.7 24.2-24.2 24.2-14.8 0-29.3-9.7-29.3-24.2.1-14.7 14.6-24.4 29.3-24.4zm-136.4 48.6c-14.5 0-29.3-9.7-29.3-24.2 0-14.8 14.8-24.4 29.3-24.4 14.8 0 24.4 9.7 24.4 24.4 0 14.6-9.6 24.2-24.4 24.2zM563 319.4c0-77.9-77.9-141.3-165.4-141.3-92.7 0-165.4 63.4-165.4 141.3S305 460.7 397.6 460.7c19.3 0 38.9-5.1 58.6-9.9l53.4 29.3-14.8-48.6C534 402.1 563 363.2 563 319.4zm-219.1-24.5c-9.7 0-19.3-9.7-19.3-19.6 0-9.7 9.7-19.3 19.3-19.3 14.8 0 24.4 9.7 24.4 19.3 0 10-9.7 19.6-24.4 19.6zm107.1 0c-9.7 0-19.3-9.7-19.3-19.6 0-9.7 9.7-19.3 19.3-19.3 14.5 0 24.4 9.7 24.4 19.3.1 10-9.9 19.6-24.4 19.6z\"],shirtsinbulk:[448,512,[],\"f214\",\"M100 410.3l30.6 13.4 4.4-9.9-30.6-13.4zm39.4 17.5l30.6 13.4 4.4-9.9-30.6-13.4zm172.1-14l4.4 9.9 30.6-13.4-4.4-9.9zM179.1 445l30.3 13.7 4.4-9.9-30.3-13.4zM60.4 392.8L91 406.2l4.4-9.6-30.6-13.7zm211.4 38.5l4.4 9.9 30.6-13.4-4.4-9.9zm-39.3 17.5l4.4 9.9 30.6-13.7-4.4-9.6zm118.4-52.2l4.4 9.6 30.6-13.4-4.4-9.9zM170 46.6h-33.5v10.5H170zm-47.2 0H89.2v10.5h33.5zm-47.3 0H42.3v10.5h33.3zm141.5 0h-33.2v10.5H217zm94.5 0H278v10.5h33.5zm47.3 0h-33.5v10.5h33.5zm-94.6 0H231v10.5h33.2zm141.5 0h-33.3v10.5h33.3zM52.8 351.1H42v33.5h10.8zm70-215.9H89.2v10.5h33.5zm-70 10.6h22.8v-10.5H42v33.5h10.8zm168.9 228.6c50.5 0 91.3-40.8 91.3-91.3 0-50.2-40.8-91.3-91.3-91.3-50.2 0-91.3 41.1-91.3 91.3 0 50.5 41.1 91.3 91.3 91.3zm-48.2-111.1c0-25.4 29.5-31.8 49.6-31.8 16.9 0 29.2 5.8 44.3 12l-8.8 16.9h-.9c-6.4-9.9-24.8-13.1-35.6-13.1-9 0-29.8 1.8-29.8 14.9 0 21.6 78.5-10.2 78.5 37.9 0 25.4-31.5 31.2-51 31.2-18.1 0-32.4-2.9-47.2-12.2l9-18.4h.9c6.1 12.2 23.6 14.9 35.9 14.9 8.7 0 32.7-1.2 32.7-14.3 0-26.1-77.6 6.3-77.6-38zM52.8 178.4H42V212h10.8zm342.4 206.2H406v-33.5h-10.8zM52.8 307.9H42v33.5h10.8zM0 3.7v406l221.7 98.6L448 409.7V3.7zm418.8 387.1L222 476.5 29.2 390.8V120.7h389.7v270.1zm0-299.3H29.2V32.9h389.7v58.6zm-366 130.1H42v33.5h10.8zm0 43.2H42v33.5h10.8zM170 135.2h-33.5v10.5H170zm225.2 163.1H406v-33.5h-10.8zm0-43.2H406v-33.5h-10.8zM217 135.2h-33.2v10.5H217zM395.2 212H406v-33.5h-10.8zm0 129.5H406V308h-10.8zm-131-206.3H231v10.5h33.2zm47.3 0H278v10.5h33.5zm83.7 33.6H406v-33.5h-33.5v10.5h22.8zm-36.4-33.6h-33.5v10.5h33.5z\"],codepen:[512,512,[],\"f1cb\",\"M502.285 159.704l-234-156c-7.987-4.915-16.511-4.96-24.571 0l-234 156C3.714 163.703 0 170.847 0 177.989v155.999c0 7.143 3.714 14.286 9.715 18.286l234 156.022c7.987 4.915 16.511 4.96 24.571 0l234-156.022c6-3.999 9.715-11.143 9.715-18.286V177.989c-.001-7.142-3.715-14.286-9.716-18.285zM278 63.131l172.286 114.858-76.857 51.429L278 165.703V63.131zm-44 0v102.572l-95.429 63.715-76.857-51.429L234 63.131zM44 219.132l55.143 36.857L44 292.846v-73.714zm190 229.715L61.714 333.989l76.857-51.429L234 346.275v102.572zm22-140.858l-77.715-52 77.715-52 77.715 52-77.715 52zm22 140.858V346.275l95.429-63.715 76.857 51.429L278 448.847zm190-156.001l-55.143-36.857L468 219.132v73.714z\"],\"git-alt\":[448,512,[],\"f841\",\"M439.55 236.05L244 40.45a28.87 28.87 0 0 0-40.81 0l-40.66 40.63 51.52 51.52c27.06-9.14 52.68 16.77 43.39 43.68l49.66 49.66c34.23-11.8 61.18 31 35.47 56.69-26.49 26.49-70.21-2.87-56-37.34L240.22 199v121.85c25.3 12.54 22.26 41.85 9.08 55a34.34 34.34 0 0 1-48.55 0c-17.57-17.6-11.07-46.91 11.25-56v-123c-20.8-8.51-24.6-30.74-18.64-45L142.57 101 8.45 235.14a28.86 28.86 0 0 0 0 40.81l195.61 195.6a28.86 28.86 0 0 0 40.8 0l194.69-194.69a28.86 28.86 0 0 0 0-40.81z\"],lyft:[512,512,[],\"f3c3\",\"M0 81.1h77.8v208.7c0 33.1 15 52.8 27.2 61-12.7 11.1-51.2 20.9-80.2-2.8C7.8 334 0 310.7 0 289V81.1zm485.9 173.5v-22h23.8v-76.8h-26.1c-10.1-46.3-51.2-80.7-100.3-80.7-56.6 0-102.7 46-102.7 102.7V357c16 2.3 35.4-.3 51.7-14 17.1-14 24.8-37.2 24.8-59v-6.7h38.8v-76.8h-38.8v-23.3c0-34.6 52.2-34.6 52.2 0v77.1c0 56.6 46 102.7 102.7 102.7v-76.5c-14.5 0-26.1-11.7-26.1-25.9zm-294.3-99v113c0 15.4-23.8 15.4-23.8 0v-113H91v132.7c0 23.8 8 54 45 63.9 37 9.8 58.2-10.6 58.2-10.6-2.1 13.4-14.5 23.3-34.9 25.3-15.5 1.6-35.2-3.6-45-7.8v70.3c25.1 7.5 51.5 9.8 77.6 4.7 47.1-9.1 76.8-48.4 76.8-100.8V155.1h-77.1v.5z\"],rev:[448,512,[],\"f5b2\",\"M289.67 274.89a65.57 65.57 0 1 1-65.56-65.56 65.64 65.64 0 0 1 65.56 65.56zm139.55-5.05h-.13a204.69 204.69 0 0 0-74.32-153l-45.38 26.2a157.07 157.07 0 0 1 71.81 131.84C381.2 361.5 310.73 432 224.11 432S67 361.5 67 274.88c0-81.88 63-149.27 143-156.43v39.12l108.77-62.79L210 32v38.32c-106.7 7.25-191 96-191 204.57 0 111.59 89.12 202.29 200.06 205v.11h210.16V269.84z\"],windows:[448,512,[],\"f17a\",\"M0 93.7l183.6-25.3v177.4H0V93.7zm0 324.6l183.6 25.3V268.4H0v149.9zm203.8 28L448 480V268.4H203.8v177.9zm0-380.6v180.1H448V32L203.8 65.7z\"],\"wizards-of-the-coast\":[640,512,[],\"f730\",\"M219.19 345.69c-1.9 1.38-11.07 8.44-.26 23.57 4.64 6.42 14.11 12.79 21.73 6.55 6.5-4.88 7.35-12.92.26-23.04-5.47-7.76-14.28-12.88-21.73-7.08zm336.75 75.94c-.34 1.7-.55 1.67.79 0 2.09-4.19 4.19-10.21 4.98-19.9 3.14-38.49-40.33-71.49-101.34-78.03-54.73-6.02-124.38 9.17-188.8 60.49l-.26 1.57c2.62 4.98 4.98 10.74 3.4 21.21l.79.26c63.89-58.4 131.19-77.25 184.35-73.85 58.4 3.67 100.03 34.04 100.03 68.08-.01 9.96-2.63 15.72-3.94 20.17zM392.28 240.42c.79 7.07 4.19 10.21 9.17 10.47 5.5.26 9.43-2.62 10.47-6.55.79-3.4 2.09-29.85 2.09-29.85s-11.26 6.55-14.93 10.47c-3.66 3.68-7.33 8.39-6.8 15.46zm-50.02-151.1C137.75 89.32 13.1 226.8.79 241.2c-1.05.52-1.31.79.79 1.31 60.49 16.5 155.81 81.18 196.13 202.16l1.05.26c55.25-69.92 140.88-128.05 236.99-128.05 80.92 0 130.15 42.16 130.15 80.39 0 18.33-6.55 33.52-22.26 46.35 0 .96-.2.79.79.79 14.66-10.74 27.5-28.8 27.5-48.18 0-22.78-12.05-38.23-12.05-38.23 7.07 7.07 10.74 16.24 10.74 16.24 5.76-40.85 26.97-62.32 26.97-62.32-2.36-9.69-6.81-17.81-6.81-17.81 7.59 8.12 14.4 27.5 14.4 41.37 0 10.47-3.4 22.78-12.57 31.95l.26.52c8.12-4.98 16.5-16.76 16.5-37.97 0-15.71-4.71-25.92-4.71-25.92 5.76-5.24 11.26-9.17 15.97-11.78.79 3.4 2.09 9.69 2.36 14.93 0 1.05.79 1.83 1.05 0 .79-5.76-.26-16.24-.26-16.5 6.02-3.14 9.69-4.45 9.69-4.45C617.74 176 489.43 89.32 342.26 89.32zm-99.24 289.62c-11.06 8.99-24.2 4.08-30.64-4.19-7.45-9.58-6.76-24.09 4.19-32.47 14.85-11.35 27.08-.49 31.16 5.5.28.39 12.13 16.57-4.71 31.16zm2.09-136.43l9.43-17.81 11.78 70.96-12.57 6.02-24.62-28.8 14.14-26.71 3.67 4.45-1.83-8.11zm18.59 117.58l-.26-.26c2.05-4.1-2.5-6.61-17.54-31.69-1.31-2.36-3.14-2.88-4.45-2.62l-.26-.52c7.86-5.76 15.45-10.21 25.4-15.71l.52.26c1.31 1.83 2.09 2.88 3.4 4.71l-.26.52c-1.05-.26-2.36-.79-5.24.26-2.09.79-7.86 3.67-12.31 7.59v1.31c1.57 2.36 3.93 6.55 5.76 9.69h.26c10.05-6.28 7.56-4.55 11.52-7.86h.26c.52 1.83.52 1.83 1.83 5.5l-.26.26c-3.06.61-4.65.34-11.52 5.5v.26c9.46 17.02 11.01 16.75 12.57 15.97l.26.26c-2.34 1.59-6.27 4.21-9.68 6.57zm55.26-32.47c-3.14 1.57-6.02 2.88-9.95 4.98l-.26-.26c1.29-2.59 1.16-2.71-11.78-32.47l-.26-.26c-.15 0-8.9 3.65-9.95 7.33h-.52l-1.05-5.76.26-.52c7.29-4.56 25.53-11.64 27.76-12.57l.52.26 3.14 4.98-.26.52c-3.53-1.76-7.35.76-12.31 2.62v.26c12.31 32.01 12.67 30.64 14.66 30.64v.25zm44.77-16.5c-4.19 1.05-5.24 1.31-9.69 2.88l-.26-.26.52-4.45c-1.05-3.4-3.14-11.52-3.67-13.62l-.26-.26c-3.4.79-8.9 2.62-12.83 3.93l-.26.26c.79 2.62 3.14 9.95 4.19 13.88.79 2.36 1.83 2.88 2.88 3.14v.52c-3.67 1.05-7.07 2.62-10.21 3.93l-.26-.26c1.05-1.31 1.05-2.88.26-4.98-1.05-3.14-8.12-23.83-9.17-27.23-.52-1.83-1.57-3.14-2.62-3.14v-.52c3.14-1.05 6.02-2.09 10.74-3.4l.26.26-.26 4.71c1.31 3.93 2.36 7.59 3.14 9.69h.26c3.93-1.31 9.43-2.88 12.83-3.93l.26-.26-2.62-9.43c-.52-1.83-1.05-3.4-2.62-3.93v-.26c4.45-1.05 7.33-1.83 10.74-2.36l.26.26c-1.05 1.31-1.05 2.88-.52 4.45 1.57 6.28 4.71 20.43 6.28 26.45.54 2.62 1.85 3.41 2.63 3.93zm32.21-6.81l-.26.26c-4.71.52-14.14 2.36-22.52 4.19l-.26-.26.79-4.19c-1.57-7.86-3.4-18.59-4.98-26.19-.26-1.83-.79-2.88-2.62-3.67l.79-.52c9.17-1.57 20.16-2.36 24.88-2.62l.26.26c.52 2.36.79 3.14 1.57 5.5l-.26.26c-1.14-1.14-3.34-3.2-16.24-.79l-.26.26c.26 1.57 1.05 6.55 1.57 9.95l.26.26c9.52-1.68 4.76-.06 10.74-2.36h.26c0 1.57-.26 1.83-.26 5.24h-.26c-4.81-1.03-2.15-.9-10.21 0l-.26.26c.26 2.09 1.57 9.43 2.09 12.57l.26.26c1.15.38 14.21-.65 16.24-4.71h.26c-.53 2.38-1.05 4.21-1.58 6.04zm10.74-44.51c-4.45 2.36-8.12 2.88-11 2.88-.25.02-11.41 1.09-17.54-9.95-6.74-10.79-.98-25.2 5.5-31.69 8.8-8.12 23.35-10.1 28.54-17.02 8.03-10.33-13.04-22.31-29.59-5.76l-2.62-2.88 5.24-16.24c25.59-1.57 45.2-3.04 50.02 16.24.79 3.14 0 9.43-.26 12.05 0 2.62-1.83 18.85-2.09 23.04-.52 4.19-.79 18.33-.79 20.69.26 2.36.52 4.19 1.57 5.5 1.57 1.83 5.76 1.83 5.76 1.83l-.79 4.71c-11.82-1.07-10.28-.59-20.43-1.05-3.22-5.15-2.23-3.28-4.19-7.86 0 .01-4.19 3.94-7.33 5.51zm37.18 21.21c-6.35-10.58-19.82-7.16-21.73 5.5-2.63 17.08 14.3 19.79 20.69 10.21l.26.26c-.52 1.83-1.83 6.02-1.83 6.28l-.52.52c-10.3 6.87-28.5-2.5-25.66-18.59 1.94-10.87 14.44-18.93 28.8-9.95l.26.52c0 1.06-.27 3.41-.27 5.25zm5.77-87.73v-6.55c.69 0 19.65 3.28 27.76 7.33l-1.57 17.54s10.21-9.43 15.45-10.74c5.24-1.57 14.93 7.33 14.93 7.33l-11.26 11.26c-12.07-6.35-19.59-.08-20.69.79-5.29 38.72-8.6 42.17 4.45 46.09l-.52 4.71c-17.55-4.29-18.53-4.5-36.92-7.33l.79-4.71c7.25 0 7.48-5.32 7.59-6.81 0 0 4.98-53.16 4.98-55.25-.02-2.87-4.99-3.66-4.99-3.66zm10.99 114.44c-8.12-2.09-14.14-11-10.74-20.69 3.14-9.43 12.31-12.31 18.85-10.21 9.17 2.62 12.83 11.78 10.74 19.38-2.61 8.9-9.42 13.87-18.85 11.52zm42.16 9.69c-2.36-.52-7.07-2.36-8.64-2.88v-.26l1.57-1.83c.59-8.24.59-7.27.26-7.59-4.82-1.81-6.66-2.36-7.07-2.36-1.31 1.83-2.88 4.45-3.67 5.5l-.79 3.4v.26c-1.31-.26-3.93-1.31-6.02-1.57v-.26l2.62-1.83c3.4-4.71 9.95-14.14 13.88-20.16v-2.09l.52-.26c2.09.79 5.5 2.09 7.59 2.88.48.48.18-1.87-1.05 25.14-.24 1.81.02 2.6.8 3.91zm-4.71-89.82c11.25-18.27 30.76-16.19 34.04-3.4L539.7 198c2.34-6.25-2.82-9.9-4.45-11.26l1.83-3.67c12.22 10.37 16.38 13.97 22.52 20.43-25.91 73.07-30.76 80.81-24.62 84.32l-1.83 4.45c-6.37-3.35-8.9-4.42-17.81-8.64l2.09-6.81c-.26-.26-3.93 3.93-9.69 3.67-19.06-1.3-22.89-31.75-9.67-52.9zm29.33 79.34c0-5.71-6.34-7.89-7.86-5.24-1.31 2.09 1.05 4.98 2.88 8.38 1.57 2.62 2.62 6.28 1.05 9.43-2.64 6.34-12.4 5.31-15.45-.79 0-.7-.27.09 1.83-4.71l.79-.26c-.57 5.66 6.06 9.61 8.38 4.98 1.05-2.09-.52-5.5-2.09-8.38-1.57-2.62-3.67-6.28-1.83-9.69 2.72-5.06 11.25-4.47 14.66 2.36v.52l-2.36 3.4zm21.21 13.36c-1.96-3.27-.91-2.14-4.45-4.71h-.26c-2.36 4.19-5.76 10.47-8.64 16.24-1.31 2.36-1.05 3.4-.79 3.93l-.26.26-5.76-4.45.26-.26 2.09-1.31c3.14-5.76 6.55-12.05 9.17-17.02v-.26c-2.64-1.98-1.22-1.51-6.02-1.83v-.26l3.14-3.4h.26c3.67 2.36 9.95 6.81 12.31 8.9l.26.26-1.31 3.91zm27.23-44.26l-2.88-2.88c.79-2.36 1.83-4.98 2.09-7.59.75-9.74-11.52-11.84-11.52-4.98 0 4.98 7.86 19.38 7.86 27.76 0 10.21-5.76 15.71-13.88 16.5-8.38.79-20.16-10.47-20.16-10.47l4.98-14.4 2.88 2.09c-2.97 17.8 17.68 20.37 13.35 5.24-1.06-4.02-18.75-34.2 2.09-38.23 13.62-2.36 23.04 16.5 23.04 16.5l-7.85 10.46zm35.62-10.21c-11-30.38-60.49-127.53-191.95-129.62-53.42-1.05-94.27 15.45-132.76 37.97l85.63-9.17-91.39 20.69 25.14 19.64-3.93-16.5c7.5-1.71 39.15-8.45 66.77-8.9l-22.26 80.39c13.61-.7 18.97-8.98 19.64-22.78l4.98-1.05.26 26.71c-22.46 3.21-37.3 6.69-49.49 9.95l13.09-43.21-61.54-36.66 2.36 8.12 10.21 4.98c6.28 18.59 19.38 56.56 20.43 58.66 1.95 4.28 3.16 5.78 12.05 4.45l1.05 4.98c-16.08 4.86-23.66 7.61-39.02 14.4l-2.36-4.71c4.4-2.94 8.73-3.94 5.5-12.83-23.7-62.5-21.48-58.14-22.78-59.44l2.36-4.45 33.52 67.3c-3.84-11.87 1.68 1.69-32.99-78.82l-41.9 88.51 4.71-13.88-35.88-42.16 27.76 93.48-11.78 8.38C95 228.58 101.05 231.87 93.23 231.52c-5.5-.26-13.62 5.5-13.62 5.5L74.63 231c30.56-23.53 31.62-24.33 58.4-42.68l4.19 7.07s-5.76 4.19-7.86 7.07c-5.9 9.28 1.67 13.28 61.8 75.68l-18.85-58.92 39.8-10.21 25.66 30.64 4.45-12.31-4.98-24.62 13.09-3.4.52 3.14 3.67-10.47-94.27 29.33 11.26-4.98-13.62-42.42 17.28-9.17 30.11 36.14 28.54-13.09c-1.41-7.47-2.47-14.5-4.71-19.64l17.28 13.88 4.71-2.09-59.18-42.68 23.08 11.5c18.98-6.07 25.23-7.47 32.21-9.69l2.62 11c-12.55 12.55 1.43 16.82 6.55 19.38l-13.62-61.01 12.05 28.28c4.19-1.31 7.33-2.09 7.33-2.09l2.62 8.64s-3.14 1.05-6.28 2.09l8.9 20.95 33.78-65.73-20.69 61.01c42.42-24.09 81.44-36.66 131.98-35.88 67.04 1.05 167.33 40.85 199.8 139.83.78 2.1-.01 2.63-.79.27zM203.48 152.43s1.83-.52 4.19-1.31l9.43 7.59c-.4 0-3.44-.25-11.26 2.36l-2.36-8.64zm143.76 38.5c-1.57-.6-26.46-4.81-33.26 20.69l21.73 17.02 11.53-37.71zM318.43 67.07c-58.4 0-106.05 12.05-114.96 14.4v.79c8.38 2.09 14.4 4.19 21.21 11.78l1.57.26c6.55-1.83 48.97-13.88 110.24-13.88 180.16 0 301.67 116.79 301.67 223.37v9.95c0 1.31.79 2.62 1.05.52.52-2.09.79-8.64.79-19.64.26-83.79-96.63-227.55-321.57-227.55zm211.06 169.68c1.31-5.76 0-12.31-7.33-13.09-9.62-1.13-16.14 23.79-17.02 33.52-.79 5.5-1.31 14.93 6.02 14.93 4.68-.01 9.72-.91 18.33-35.36zm-61.53 42.95c-2.62-.79-9.43-.79-12.57 10.47-1.83 6.81.52 13.35 6.02 14.66 3.67 1.05 8.9.52 11.78-10.74 2.62-9.94-1.83-13.61-5.23-14.39zM491 300.65c1.83.52 3.14 1.05 5.76 1.83 0-1.83.52-8.38.79-12.05-1.05 1.31-5.5 8.12-6.55 9.95v.27z\"],\"square-viadeo\":[448,512,[\"viadeo-square\"],\"f2aa\",\"M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM118.3 381.2c-68-73.6-19.8-196.1 81.2-196.1c13.3 0 26.6 2.1 39.1 6.7c-4.3 8.4-7.3 17.6-8.4 27.1c-9.7-4.1-20.2-6-30.7-6c-48.8 0-84.6 41.7-84.6 88.9c0 43 28.5 78.7 69.5 85.9c61.5-24 72.9-117.6 72.9-175v0c0-7.3 0-14.8-.6-22.1c-11.2-32.9-26.6-64.6-44.2-94.5c27.1 18.3 41.9 62.5 44.2 94.1v.4c7.7 22.5 11.8 46.2 11.8 70c0 54.1-21.9 99-68.3 128.2l-2.4 .2c50 1 86.2-38.6 86.2-87.2c0-12.2-2.1-24.3-6.9-35.7c9.5-1.9 18.5-5.6 26.4-10.5c15.3 36.6 12.6 87.3-22.8 125.6c-42.4 46.2-120 46.6-162.4 0zM274.6 217.6c21.9-12 49.6-30.7 62.3-53c1.5-3 4.1-8.6 4.5-12c-12.5 27.9-44.2 49.8-73.9 56.7c-4.7-7.3-7.5-15.5-7.5-24.3c0-10.3 5.2-24.1 12.9-31.6c8.3-7.9 18-10.9 27.9-14.1c16-5.1 32.5-10.3 44.5-35.9c32.5 46.2 13.1 130.3-36.3 130.3c-13.3 0-25.1-7.1-34.4-16.1z\"],meetup:[512,512,[],\"f2e0\",\"M99 414.3c1.1 5.7-2.3 11.1-8 12.3-5.4 1.1-10.9-2.3-12-8-1.1-5.4 2.3-11.1 7.7-12.3 5.4-1.2 11.1 2.3 12.3 8zm143.1 71.4c-6.3 4.6-8 13.4-3.7 20 4.6 6.6 13.4 8.3 20 3.7 6.3-4.6 8-13.4 3.4-20-4.2-6.5-13.1-8.3-19.7-3.7zm-86-462.3c6.3-1.4 10.3-7.7 8.9-14-1.1-6.6-7.4-10.6-13.7-9.1-6.3 1.4-10.3 7.7-9.1 14 1.4 6.6 7.6 10.6 13.9 9.1zM34.4 226.3c-10-6.9-23.7-4.3-30.6 6-6.9 10-4.3 24 5.7 30.9 10 7.1 23.7 4.6 30.6-5.7 6.9-10.4 4.3-24.1-5.7-31.2zm272-170.9c10.6-6.3 13.7-20 7.7-30.3-6.3-10.6-19.7-14-30-7.7s-13.7 20-7.4 30.6c6 10.3 19.4 13.7 29.7 7.4zm-191.1 58c7.7-5.4 9.4-16 4.3-23.7s-15.7-9.4-23.1-4.3c-7.7 5.4-9.4 16-4.3 23.7 5.1 7.8 15.6 9.5 23.1 4.3zm372.3 156c-7.4 1.7-12.3 9.1-10.6 16.9 1.4 7.4 8.9 12.3 16.3 10.6 7.4-1.4 12.3-8.9 10.6-16.6-1.5-7.4-8.9-12.3-16.3-10.9zm39.7-56.8c-1.1-5.7-6.6-9.1-12-8-5.7 1.1-9.1 6.9-8 12.6 1.1 5.4 6.6 9.1 12.3 8 5.4-1.5 9.1-6.9 7.7-12.6zM447 138.9c-8.6 6-10.6 17.7-4.9 26.3 5.7 8.6 17.4 10.6 26 4.9 8.3-6 10.3-17.7 4.6-26.3-5.7-8.7-17.4-10.9-25.7-4.9zm-6.3 139.4c26.3 43.1 15.1 100-26.3 129.1-17.4 12.3-37.1 17.7-56.9 17.1-12 47.1-69.4 64.6-105.1 32.6-1.1.9-2.6 1.7-3.7 2.9-39.1 27.1-92.3 17.4-119.4-22.3-9.7-14.3-14.6-30.6-15.1-46.9-65.4-10.9-90-94-41.1-139.7-28.3-46.9.6-107.4 53.4-114.9C151.6 70 234.1 38.6 290.1 82c67.4-22.3 136.3 29.4 130.9 101.1 41.1 12.6 52.8 66.9 19.7 95.2zm-70 74.3c-3.1-20.6-40.9-4.6-43.1-27.1-3.1-32 43.7-101.1 40-128-3.4-24-19.4-29.1-33.4-29.4-13.4-.3-16.9 2-21.4 4.6-2.9 1.7-6.6 4.9-11.7-.3-6.3-6-11.1-11.7-19.4-12.9-12.3-2-17.7 2-26.6 9.7-3.4 2.9-12 12.9-20 9.1-3.4-1.7-15.4-7.7-24-11.4-16.3-7.1-40 4.6-48.6 20-12.9 22.9-38 113.1-41.7 125.1-8.6 26.6 10.9 48.6 36.9 47.1 11.1-.6 18.3-4.6 25.4-17.4 4-7.4 41.7-107.7 44.6-112.6 2-3.4 8.9-8 14.6-5.1 5.7 3.1 6.9 9.4 6 15.1-1.1 9.7-28 70.9-28.9 77.7-3.4 22.9 26.9 26.6 38.6 4 3.7-7.1 45.7-92.6 49.4-98.3 4.3-6.3 7.4-8.3 11.7-8 3.1 0 8.3.9 7.1 10.9-1.4 9.4-35.1 72.3-38.9 87.7-4.6 20.6 6.6 41.4 24.9 50.6 11.4 5.7 62.5 15.7 58.5-11.1zm5.7 92.3c-10.3 7.4-12.9 22-5.7 32.6 7.1 10.6 21.4 13.1 32 6 10.6-7.4 13.1-22 6-32.6-7.4-10.6-21.7-13.5-32.3-6z\"],centos:[448,512,[],\"f789\",\"M289.6 97.5l31.6 31.7-76.3 76.5V97.5zm-162.4 31.7l76.3 76.5V97.5h-44.7zm41.5-41.6h44.7v127.9l10.8 10.8 10.8-10.8V87.6h44.7L224.2 32zm26.2 168.1l-10.8-10.8H55.5v-44.8L0 255.7l55.5 55.6v-44.8h128.6l10.8-10.8zm79.3-20.7h107.9v-44.8l-31.6-31.7zm173.3 20.7L392 200.1v44.8H264.3l-10.8 10.8 10.8 10.8H392v44.8l55.5-55.6zM65.4 176.2l32.5-31.7 90.3 90.5h15.3v-15.3l-90.3-90.5 31.6-31.7H65.4zm316.7-78.7h-78.5l31.6 31.7-90.3 90.5V235h15.3l90.3-90.5 31.6 31.7zM203.5 413.9V305.8l-76.3 76.5 31.6 31.7h44.7zM65.4 235h108.8l-76.3-76.5-32.5 31.7zm316.7 100.2l-31.6 31.7-90.3-90.5h-15.3v15.3l90.3 90.5-31.6 31.7h78.5zm0-58.8H274.2l76.3 76.5 31.6-31.7zm-60.9 105.8l-76.3-76.5v108.1h44.7zM97.9 352.9l76.3-76.5H65.4v44.8zm181.8 70.9H235V295.9l-10.8-10.8-10.8 10.8v127.9h-44.7l55.5 55.6zm-166.5-41.6l90.3-90.5v-15.3h-15.3l-90.3 90.5-32.5-31.7v78.7h79.4z\"],adn:[496,512,[],\"f170\",\"M248 167.5l64.9 98.8H183.1l64.9-98.8zM496 256c0 136.9-111.1 248-248 248S0 392.9 0 256 111.1 8 248 8s248 111.1 248 248zm-99.8 82.7L248 115.5 99.8 338.7h30.4l33.6-51.7h168.6l33.6 51.7h30.2z\"],cloudsmith:[512,512,[],\"f384\",\"M512 227.6v56.9L284.4 512H227.6L0 284.4V227.6L227.6 0h56.9L512 227.6zm-256 162a133.6 133.6 0 1 0 0-267.1 133.6 133.6 0 1 0 0 267.1z\"],opensuse:[640,512,[],\"e62b\",\"M471.1 102.7s-.3 18.3-.3 20.3c-9.1-3-74.4-24.1-135.7-26.3c-51.9-1.8-122.8-4.3-223 57.3c-19.4 12.4-73.9 46.1-99.6 109.7C7 277-.1 307 7 335.1c3.3 12.8 8.9 24.9 16.5 35.7c17.4 25 46.6 41.6 78.1 44.4c44.4 3.9 78.1-16 90-53.3c8.2-25.8 0-63.6-31.5-82.9c-25.6-15.7-53.3-12.1-69.2-1.6c-13.9 9.2-21.8 23.5-21.6 39.2c.3 27.8 24.3 42.6 41.5 42.6c5.4 0 10.7-.9 15.8-2.7c6.5-1.8 13.3-6.5 13.3-14.9c0-12.1-11.6-14.8-16.8-13.9c-2.9 .5-4.5 2-11.8 2.4c-2-.2-12-3.1-12-14V316c.2-12.3 13.2-18 25.5-16.9c32.3 2.8 47.7 40.7 28.5 65.7C135 388.5 76.7 388 53.6 344.4c-26-49.2 12.7-111.2 87-98.4c33.2 5.7 83.6 35.5 102.4 104.3h45.9c-5.7-17.6-8.9-68.3 42.7-68.3c56.7 0 63.9 39.9 79.8 68.3H460c-12.8-18.3-21.7-38.7-18.9-55.8c5.6-33.8 39.7-18.4 82.4-17.4c66.5 .4 102.1-27 103.1-28c3.7-3.1 6.5-15.8 7-17.7c1.3-5.1-3.2-2.4-3.2-2.4c-8.7 5.2-30.5 15.2-50.9 15.6c-25.3 .5-76.2-25.4-81.6-28.2c-.3-.4 .1 1.2-11-25.5c88.4 58.3 118.3 40.5 145.2 21.7c.8-.6 4.3-2.9 3.6-5.7c-13.8-48.1-22.4-62.7-34.5-69.6c-37-21.6-125-34.7-129.2-35.3c.1-.1-.9-.3-.9 .7l0 0zm135.6 75.4a37.6 37.6 0 1 1 -75.2-2.6 37.6 37.6 0 1 1 75.2 2.6zm-36.6-27.9a26.3 26.3 0 1 0 -1.7 52.5 26.3 26.3 0 1 0 1.7-52.5zm4.3 28.8c-15.4 0-15.4-15.6 0-15.6s15.4 15.6 0 15.6v0z\"],\"pied-piper-alt\":[576,512,[],\"f1a8\",\"M244 246c-3.2-2-6.3-2.9-10.1-2.9-6.6 0-12.6 3.2-19.3 3.7l1.7 4.9zm135.9 197.9c-19 0-64.1 9.5-79.9 19.8l6.9 45.1c35.7 6.1 70.1 3.6 106-9.8-4.8-10-23.5-55.1-33-55.1zM340.8 177c6.6 2.8 11.5 9.2 22.7 22.1 2-1.4 7.5-5.2 7.5-8.6 0-4.9-11.8-13.2-13.2-23 11.2-5.7 25.2-6 37.6-8.9 68.1-16.4 116.3-52.9 146.8-116.7C548.3 29.3 554 16.1 554.6 2l-2 2.6c-28.4 50-33 63.2-81.3 100-31.9 24.4-69.2 40.2-106.6 54.6l-6.3-.3v-21.8c-19.6 1.6-19.7-14.6-31.6-23-18.7 20.6-31.6 40.8-58.9 51.1-12.7 4.8-19.6 10-25.9 21.8 34.9-16.4 91.2-13.5 98.8-10zM555.5 0l-.6 1.1-.3.9.6-.6zm-59.2 382.1c-33.9-56.9-75.3-118.4-150-115.5l-.3-6c-1.1-13.5 32.8 3.2 35.1-31l-14.4 7.2c-19.8-45.7-8.6-54.3-65.5-54.3-14.7 0-26.7 1.7-41.4 4.6 2.9 18.6 2.2 36.7-10.9 50.3l19.5 5.5c-1.7 3.2-2.9 6.3-2.9 9.8 0 21 42.8 2.9 42.8 33.6 0 18.4-36.8 60.1-54.9 60.1-8 0-53.7-50-53.4-60.1l.3-4.6 52.3-11.5c13-2.6 12.3-22.7-2.9-22.7-3.7 0-43.1 9.2-49.4 10.6-2-5.2-7.5-14.1-13.8-14.1-3.2 0-6.3 3.2-9.5 4-9.2 2.6-31 2.9-21.5 20.1L15.9 298.5c-5.5 1.1-8.9 6.3-8.9 11.8 0 6 5.5 10.9 11.5 10.9 8 0 131.3-28.4 147.4-32.2 2.6 3.2 4.6 6.3 7.8 8.6 20.1 14.4 59.8 85.9 76.4 85.9 24.1 0 58-22.4 71.3-41.9 3.2-4.3 6.9-7.5 12.4-6.9.6 13.8-31.6 34.2-33 43.7-1.4 10.2-1 35.2-.3 41.1 26.7 8.1 52-3.6 77.9-2.9 4.3-21 10.6-41.9 9.8-63.5l-.3-9.5c-1.4-34.2-10.9-38.5-34.8-58.6-1.1-1.1-2.6-2.6-3.7-4 2.2-1.4 1.1-1 4.6-1.7 88.5 0 56.3 183.6 111.5 229.9 33.1-15 72.5-27.9 103.5-47.2-29-25.6-52.6-45.7-72.7-79.9zm-196.2 46.1v27.2l11.8-3.4-2.9-23.8zm-68.7-150.4l24.1 61.2 21-13.8-31.3-50.9zm84.4 154.9l2 12.4c9-1.5 58.4-6.6 58.4-14.1 0-1.4-.6-3.2-.9-4.6-26.8 0-36.9 3.8-59.5 6.3z\"],\"square-dribbble\":[448,512,[\"dribbble-square\"],\"f397\",\"M165.9 132.5c-38.3 18-66.8 53.3-75.7 95.7c6.1 .1 62.4 .3 126.4-16.7c-22.7-40.2-47.1-74.1-50.7-79zm26.1-9.1c3.8 5.1 28.6 38.9 51 80c48.6-18.3 69.1-45.9 71.6-49.4C281 124.2 235.3 112.9 192 123.4zM277.4 382c-2-12-10-53.8-29.2-103.6c-55.1 18.8-93.8 56.4-108.1 85.6c40.5 31.6 93.3 36.7 137.3 18zM227.8 232.6C159.6 253 93.4 252.2 87.4 252c0 .7 0 1.4 0 2.1s0 1.4 0 2.1c0 35.1 13.3 67.1 35.1 91.4c22.2-37.9 67.1-77.9 116.5-91.8c-3.4-7.8-7.2-15.5-11.1-23.2zm72.5 136.9c30.7-20.7 52.5-53.6 58.6-91.6c-4.6-1.5-42.3-12.7-85.1-5.8c17.9 49.1 25.1 89.1 26.5 97.4zm-34.8-119c45.5-5.7 90.7 3.4 95.2 4.4c-.3-32.3-11.8-61.9-30.9-85.1c-2.9 3.9-25.8 33.2-76.3 53.9c4.8 9.8 8.3 17.8 12 26.8zM384 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64zM224 416a160 160 0 1 1 0-320 160 160 0 1 1 0 320z\"],codiepie:[472,512,[],\"f284\",\"M422.5 202.9c30.7 0 33.5 53.1-.3 53.1h-10.8v44.3h-26.6v-97.4h37.7zM472 352.6C429.9 444.5 350.4 504 248 504 111 504 0 393 0 256S111 8 248 8c97.4 0 172.8 53.7 218.2 138.4l-186 108.8L472 352.6zm-38.5 12.5l-60.3-30.7c-27.1 44.3-70.4 71.4-122.4 71.4-82.5 0-149.2-66.7-149.2-148.9 0-82.5 66.7-149.2 149.2-149.2 48.4 0 88.9 23.5 116.9 63.4l59.5-34.6c-40.7-62.6-104.7-100-179.2-100-121.2 0-219.5 98.3-219.5 219.5S126.8 475.5 248 475.5c78.6 0 146.5-42.1 185.5-110.4z\"],node:[640,512,[],\"f419\",\"M316.3 452c-2.1 0-4.2-.6-6.1-1.6L291 439c-2.9-1.6-1.5-2.2-.5-2.5 3.8-1.3 4.6-1.6 8.7-4 .4-.2 1-.1 1.4.1l14.8 8.8c.5.3 1.3.3 1.8 0L375 408c.5-.3.9-.9.9-1.6v-66.7c0-.7-.3-1.3-.9-1.6l-57.8-33.3c-.5-.3-1.2-.3-1.8 0l-57.8 33.3c-.6.3-.9 1-.9 1.6v66.7c0 .6.4 1.2.9 1.5l15.8 9.1c8.6 4.3 13.9-.8 13.9-5.8v-65.9c0-.9.7-1.7 1.7-1.7h7.3c.9 0 1.7.7 1.7 1.7v65.9c0 11.5-6.2 18-17.1 18-3.3 0-6 0-13.3-3.6l-15.2-8.7c-3.7-2.2-6.1-6.2-6.1-10.5v-66.7c0-4.3 2.3-8.4 6.1-10.5l57.8-33.4c3.7-2.1 8.5-2.1 12.1 0l57.8 33.4c3.7 2.2 6.1 6.2 6.1 10.5v66.7c0 4.3-2.3 8.4-6.1 10.5l-57.8 33.4c-1.7 1.1-3.8 1.7-6 1.7zm46.7-65.8c0-12.5-8.4-15.8-26.2-18.2-18-2.4-19.8-3.6-19.8-7.8 0-3.5 1.5-8.1 14.8-8.1 11.9 0 16.3 2.6 18.1 10.6.2.8.8 1.3 1.6 1.3h7.5c.5 0 .9-.2 1.2-.5.3-.4.5-.8.4-1.3-1.2-13.8-10.3-20.2-28.8-20.2-16.5 0-26.3 7-26.3 18.6 0 12.7 9.8 16.1 25.6 17.7 18.9 1.9 20.4 4.6 20.4 8.3 0 6.5-5.2 9.2-17.4 9.2-15.3 0-18.7-3.8-19.8-11.4-.1-.8-.8-1.4-1.7-1.4h-7.5c-.9 0-1.7.7-1.7 1.7 0 9.7 5.3 21.3 30.6 21.3 18.5 0 29-7.2 29-19.8zm54.5-50.1c0 6.1-5 11.1-11.1 11.1s-11.1-5-11.1-11.1c0-6.3 5.2-11.1 11.1-11.1 6-.1 11.1 4.8 11.1 11.1zm-1.8 0c0-5.2-4.2-9.3-9.4-9.3-5.1 0-9.3 4.1-9.3 9.3 0 5.2 4.2 9.4 9.3 9.4 5.2-.1 9.4-4.3 9.4-9.4zm-4.5 6.2h-2.6c-.1-.6-.5-3.8-.5-3.9-.2-.7-.4-1.1-1.3-1.1h-2.2v5h-2.4v-12.5h4.3c1.5 0 4.4 0 4.4 3.3 0 2.3-1.5 2.8-2.4 3.1 1.7.1 1.8 1.2 2.1 2.8.1 1 .3 2.7.6 3.3zm-2.8-8.8c0-1.7-1.2-1.7-1.8-1.7h-2v3.5h1.9c1.6 0 1.9-1.1 1.9-1.8zM137.3 191c0-2.7-1.4-5.1-3.7-6.4l-61.3-35.3c-1-.6-2.2-.9-3.4-1h-.6c-1.2 0-2.3.4-3.4 1L3.7 184.6C1.4 185.9 0 188.4 0 191l.1 95c0 1.3.7 2.5 1.8 3.2 1.1.7 2.5.7 3.7 0L42 268.3c2.3-1.4 3.7-3.8 3.7-6.4v-44.4c0-2.6 1.4-5.1 3.7-6.4l15.5-8.9c1.2-.7 2.4-1 3.7-1 1.3 0 2.6.3 3.7 1l15.5 8.9c2.3 1.3 3.7 3.8 3.7 6.4v44.4c0 2.6 1.4 5.1 3.7 6.4l36.4 20.9c1.1.7 2.6.7 3.7 0 1.1-.6 1.8-1.9 1.8-3.2l.2-95zM472.5 87.3v176.4c0 2.6-1.4 5.1-3.7 6.4l-61.3 35.4c-2.3 1.3-5.1 1.3-7.4 0l-61.3-35.4c-2.3-1.3-3.7-3.8-3.7-6.4v-70.8c0-2.6 1.4-5.1 3.7-6.4l61.3-35.4c2.3-1.3 5.1-1.3 7.4 0l15.3 8.8c1.7 1 3.9-.3 3.9-2.2v-94c0-2.8 3-4.6 5.5-3.2l36.5 20.4c2.3 1.2 3.8 3.7 3.8 6.4zm-46 128.9c0-.7-.4-1.3-.9-1.6l-21-12.2c-.6-.3-1.3-.3-1.9 0l-21 12.2c-.6.3-.9.9-.9 1.6v24.3c0 .7.4 1.3.9 1.6l21 12.1c.6.3 1.3.3 1.8 0l21-12.1c.6-.3.9-.9.9-1.6v-24.3zm209.8-.7c2.3-1.3 3.7-3.8 3.7-6.4V192c0-2.6-1.4-5.1-3.7-6.4l-60.9-35.4c-2.3-1.3-5.1-1.3-7.4 0l-61.3 35.4c-2.3 1.3-3.7 3.8-3.7 6.4v70.8c0 2.7 1.4 5.1 3.7 6.4l60.9 34.7c2.2 1.3 5 1.3 7.3 0l36.8-20.5c2.5-1.4 2.5-5 0-6.4L550 241.6c-1.2-.7-1.9-1.9-1.9-3.2v-22.2c0-1.3.7-2.5 1.9-3.2l19.2-11.1c1.1-.7 2.6-.7 3.7 0l19.2 11.1c1.1.7 1.9 1.9 1.9 3.2v17.4c0 2.8 3.1 4.6 5.6 3.2l36.7-21.3zM559 219c-.4.3-.7.7-.7 1.2v13.6c0 .5.3 1 .7 1.2l11.8 6.8c.4.3 1 .3 1.4 0L584 235c.4-.3.7-.7.7-1.2v-13.6c0-.5-.3-1-.7-1.2l-11.8-6.8c-.4-.3-1-.3-1.4 0L559 219zm-254.2 43.5v-70.4c0-2.6-1.6-5.1-3.9-6.4l-61.1-35.2c-2.1-1.2-5-1.4-7.4 0l-61.1 35.2c-2.3 1.3-3.9 3.7-3.9 6.4v70.4c0 2.8 1.9 5.2 4 6.4l61.2 35.2c2.4 1.4 5.2 1.3 7.4 0l61-35.2c1.8-1 3.1-2.7 3.6-4.7.1-.5.2-1.1.2-1.7zm-74.3-124.9l-.8.5h1.1l-.3-.5zm76.2 130.2l-.4-.7v.9l.4-.2z\"],mix:[448,512,[],\"f3cb\",\"M0 64v348.9c0 56.2 88 58.1 88 0V174.3c7.9-52.9 88-50.4 88 6.5v175.3c0 57.9 96 58 96 0V240c5.3-54.7 88-52.5 88 4.3v23.8c0 59.9 88 56.6 88 0V64H0z\"],steam:[496,512,[],\"f1b6\",\"M496 256c0 137-111.2 248-248.4 248-113.8 0-209.6-76.3-239-180.4l95.2 39.3c6.4 32.1 34.9 56.4 68.9 56.4 39.2 0 71.9-32.4 70.2-73.5l84.5-60.2c52.1 1.3 95.8-40.9 95.8-93.5 0-51.6-42-93.5-93.7-93.5s-93.7 42-93.7 93.5v1.2L176.6 279c-15.5-.9-30.7 3.4-43.5 12.1L0 236.1C10.2 108.4 117.1 8 247.6 8 384.8 8 496 119 496 256zM155.7 384.3l-30.5-12.6a52.79 52.79 0 0 0 27.2 25.8c26.9 11.2 57.8-1.6 69-28.4 5.4-13 5.5-27.3.1-40.3-5.4-13-15.5-23.2-28.5-28.6-12.9-5.4-26.7-5.2-38.9-.6l31.5 13c19.8 8.2 29.2 30.9 20.9 50.7-8.3 19.9-31 29.2-50.8 21zm173.8-129.9c-34.4 0-62.4-28-62.4-62.3s28-62.3 62.4-62.3 62.4 28 62.4 62.3-27.9 62.3-62.4 62.3zm.1-15.6c25.9 0 46.9-21 46.9-46.8 0-25.9-21-46.8-46.9-46.8s-46.9 21-46.9 46.8c.1 25.8 21.1 46.8 46.9 46.8z\"],\"cc-apple-pay\":[576,512,[],\"f416\",\"M302.2 218.4c0 17.2-10.5 27.1-29 27.1h-24.3v-54.2h24.4c18.4 0 28.9 9.8 28.9 27.1zm47.5 62.6c0 8.3 7.2 13.7 18.5 13.7 14.4 0 25.2-9.1 25.2-21.9v-7.7l-23.5 1.5c-13.3.9-20.2 5.8-20.2 14.4zM576 79v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V79c0-26.5 21.5-48 48-48h480c26.5 0 48 21.5 48 48zM127.8 197.2c8.4.7 16.8-4.2 22.1-10.4 5.2-6.4 8.6-15 7.7-23.7-7.4.3-16.6 4.9-21.9 11.3-4.8 5.5-8.9 14.4-7.9 22.8zm60.6 74.5c-.2-.2-19.6-7.6-19.8-30-.2-18.7 15.3-27.7 16-28.2-8.8-13-22.4-14.4-27.1-14.7-12.2-.7-22.6 6.9-28.4 6.9-5.9 0-14.7-6.6-24.3-6.4-12.5.2-24.2 7.3-30.5 18.6-13.1 22.6-3.4 56 9.3 74.4 6.2 9.1 13.7 19.1 23.5 18.7 9.3-.4 13-6 24.2-6 11.3 0 14.5 6 24.3 5.9 10.2-.2 16.5-9.1 22.8-18.2 6.9-10.4 9.8-20.4 10-21zm135.4-53.4c0-26.6-18.5-44.8-44.9-44.8h-51.2v136.4h21.2v-46.6h29.3c26.8 0 45.6-18.4 45.6-45zm90 23.7c0-19.7-15.8-32.4-40-32.4-22.5 0-39.1 12.9-39.7 30.5h19.1c1.6-8.4 9.4-13.9 20-13.9 13 0 20.2 6 20.2 17.2v7.5l-26.4 1.6c-24.6 1.5-37.9 11.6-37.9 29.1 0 17.7 13.7 29.4 33.4 29.4 13.3 0 25.6-6.7 31.2-17.4h.4V310h19.6v-68zM516 210.9h-21.5l-24.9 80.6h-.4l-24.9-80.6H422l35.9 99.3-1.9 6c-3.2 10.2-8.5 14.2-17.9 14.2-1.7 0-4.9-.2-6.2-.3v16.4c1.2.4 6.5.5 8.1.5 20.7 0 30.4-7.9 38.9-31.8L516 210.9z\"],scribd:[384,512,[],\"f28a\",\"M42.3 252.7c-16.1-19-24.7-45.9-24.8-79.9 0-100.4 75.2-153.1 167.2-153.1 98.6-1.6 156.8 49 184.3 70.6l-50.5 72.1-37.3-24.6 26.9-38.6c-36.5-24-79.4-36.5-123-35.8-50.7-.8-111.7 27.2-111.7 76.2 0 18.7 11.2 20.7 28.6 15.6 23.3-5.3 41.9.6 55.8 14 26.4 24.3 23.2 67.6-.7 91.9-29.2 29.5-85.2 27.3-114.8-8.4zm317.7 5.9c-15.5-18.8-38.9-29.4-63.2-28.6-38.1-2-71.1 28-70.5 67.2-.7 16.8 6 33 18.4 44.3 14.1 13.9 33 19.7 56.3 14.4 17.4-5.1 28.6-3.1 28.6 15.6 0 4.3-.5 8.5-1.4 12.7-16.7 40.9-59.5 64.4-121.4 64.4-51.9.2-102.4-16.4-144.1-47.3l33.7-39.4-35.6-27.4L0 406.3l15.4 13.8c52.5 46.8 120.4 72.5 190.7 72.2 51.4 0 94.4-10.5 133.6-44.1 57.1-51.4 54.2-149.2 20.3-189.6z\"],debian:[448,512,[],\"e60b\",\"M380.2 245.6c3-7.6 5.5-14 5.2-24.4l-4.3 9c4.4-13.2 4-27.1 3.6-40.4c-.2-6-.3-11.8 0-17.4l-1.8-.5c-1.5-45.2-40.6-93.1-75.3-109.4c-30-13.8-76.1-16.2-97.3-5.8c1.3-1.1 4.2-2 6.8-2.7l.3-.1c3.3-1 6-1.7 4-2.9c-19.2 1.9-24.9 5.5-31.1 9.4l-.1 0c-4.6 2.9-9.5 6-20.3 8.7c-3.5 3.4 1.7 2 5.8 .9l0 0c4.1-1.1 7.2-1.9-.1 2.4c-3.5 1-6.6 1.3-9.6 1.6l-.1 0c-8.3 .8-15.8 1.6-30.7 17c.8 1.3 3.4-.2 5.3-1.3l.1-.1c2.3-1.4 3.4-2-1.7 4.4c-19.1-2.4-60.3 43.7-69.1 59l4.6 .8c-3.2 8-6.8 14.8-10 20.8c-4.3 8.1-7.9 14.9-8.7 21.3c-.3 5.1-1 11-1.7 17.3l0 0c-.1 1-.2 2-.3 3l-.1 .6c-3 27.3-6.7 60.8 3.9 73l-1.3 13c.6 1.2 1.1 2.3 1.6 3.5c.2 .4 .4 .8 .5 1.1l0 0 0 0 0 0 0 0 0 0 0 0 0 0c1 2.1 2 4.2 3.3 6.2l-3 .2c7 22.1 10.8 22.5 15.1 22.9l0 0c4.4 .4 9.3 .9 18.7 24.2c-2.7-.9-5.5-1.9-9.4-7.2c-.5 4.1 5.8 16.3 13.1 25.8l-3.1 3.6c2.1 3.7 4.8 6.2 7.6 8.8l0 0 0 0c1 .9 2.1 1.9 3.1 2.9c-11.9-6.5 3.2 13.7 11.9 25.2c.8 1.1 1.5 2 2.2 2.9l0 0 0 0 0 0 0 0 0 0c1.4 1.9 2.5 3.4 2.9 4.1l2.4-4.2c-.3 6.1 4.3 13.9 13.1 24.7l7.3-.3c3 6 14 16.7 20.7 17.2l-4.4 5.8c8.1 2.6 10.3 4.3 12.7 6.2c2.6 2.1 5.4 4.3 16.1 8.1l-4.2-7.4c3.5 3 6.2 5.9 8.8 8.7l.1 .1c5.2 5.6 9.9 10.6 19.7 15.3c10.7 3.7 16.6 4.7 22.7 5.8c.3 0 .6 .1 .9 .1c5.4 .8 11.2 1.8 20.8 4.5c-1.1-.1-2.2-.1-3.3-.1h0c-2.3-.1-4.7-.1-7-.1l0 0 0 0 0 0 0 0 0 0 0 0 0 0c-14.4-.2-29.2-.4-42.7-5.2C107.8 480.5 19.5 367.2 26 250.6c-.6-9.9-.3-20.9 0-30.7c.4-13.5 .7-24.8-1.6-28.3l1-3.1c5.3-17.4 11.7-38.2 23.8-62.8l-.1-.2v-.1c.4 .4 3.4 3.4 8.8-5.8c.8-1.8 1.6-3.7 2.4-5.6c.5-1.1 .9-2.2 1.4-3.2c2.5-6.1 5.1-12.3 8.4-17.9l2.6-.6c1.7-10.1 17-23.8 29.8-35.2l1.1-1c5.7-5.1 10.7-9.7 13.6-13.1l.7 4.4c17-15.9 44.6-27.5 65.6-36.4l.5-.2c4.8-2 9.3-3.9 13.3-5.7c-3.4 3.8 2.2 2.7 10 1c4.8-1 10.4-2.1 15.3-2.4l-3.9 2.1c-2.7 1.4-5.4 2.8-8 4.6c8.1-2 11.7-1.4 15.7-.8l.3 0c3.5 .6 7.3 1.2 14.6 .2c-5.6 .8-12.3 3-11.2 3.8c7.9 .9 12.8-.1 17.2-1l.2 0c5.5-1.1 10.3-2 19.3 .9l-1-4.8c7.3 2.6 12.7 4.3 17.5 5.8l.5 .1c10 3 17.6 5.3 34.2 14.1c3.2 .2 5.3-.5 7.4-1.2l.1 0c3.6-1.1 7-2.1 15.2 1.2c.3 .5 .5 1 .7 1.4c.1 .2 .2 .5 .3 .7l0 .1c1 2.6 1.8 4.6 14.6 12.1c1.7-.7-2.7-4.7-6.4-8.2c0 0 0 0-.1-.1c-.2-.1-.3-.3-.5-.4c32.2 17.3 67.3 54.1 78 93.5c-6-11.1-5.2-5.5-4.3 .5c.6 4 1.2 8.1-.2 7.5c4.5 12.1 8.1 24.5 10.4 37.4l-.8-2.9-.1-.3c-3.3-11.9-9.6-34.3-19.9-49.3c-.4 4.3-2.8 3.9-5.2 3.5l-.1 0 0 0c-3.3-.6-6.2-1.1-1.9 12.6c2.6 3.8 3.1 2.4 3.5 1.1l0 0c.5-1.5 .9-2.7 4.7 5.2c.1 4.1 1 8.2 2.1 12.7l0 0 0 0 .1 .6c.1 .3 .1 .5 .2 .8l.1 .6c.6 2.6 1.3 5.4 1.8 8.4c-1.1-.2-2.3-2.2-3.4-4.2c-1.4-2.4-2.8-4.7-3.7-3.2c2.4 11.5 6.5 17.4 8 18.3c-.3 .6-.6 .7-1.1 .7c-.8 0-1.8 .1-1.9 5.3c.7 13.7 3.3 12.5 5.3 11.6l0 0c.6-.3 1.2-.6 1.7-.4c-.6 2.5-1.6 5.1-2.7 7.9c-2.8 7.1-6 15.4-3.4 26.1c-.8-3-2-6-3.1-8.9l-.1-.4c-.2-.5-.4-1-.6-1.5l0 0c-.3-.8-.6-1.6-.9-2.3c-.6 4.4-.3 7.7-.1 10.6c0 .2 0 .5 0 .7c.4 5.3 .7 10-3 19.9c4.3-14.2 3.8-26.9-.2-20.8c1 10.9-3.7 20.4-8 28.9l-.1 .2c-3.6 7.1-6.8 13.5-5.9 19.3l-5.2-7.1c-7.5 10.9-7 13.3-6.5 15.5l0 .1c.5 1.9 1 3.8-3.4 10.8c1.7-2.9 1.3-3.6 1-4.2l0 0c-.4-.8-.7-1.5 1.7-5.1c-1.6 .1-5.5 3.9-10.1 8.5c-3.9 3.9-8.5 8.4-12.8 11.8c-37.5 30.1-82.3 34-125.6 17.8c.2-1-.2-2.1-3.1-4.1c-36.8-28.2-58.5-52.1-50.9-107.5c2.1-1.6 3.6-5.8 5.3-10.8l0 0 0 0 .2-.4 .1-.3 0-.1c2.9-8.4 6.5-18.8 14.3-23.8c7.8-17.3 31.3-33.3 56.4-33.7c25.6-1.4 47.2 13.7 58.1 27.9c-19.8-18.4-52.1-24-79.7-10.4c-28.2 12.7-45 43.8-42.5 74.7c.3-.4 .6-.6 .9-.8l0 0s0 0 0 0c0 0 .1-.1 .1-.1l.1-.1c.6-.5 1.1-.9 1.4-3.3c-.9 60.2 64.8 104.3 112.1 82l.6 1.3c12.7-3.5 15.9-6.5 20.3-10.7l.1-.1 0 0c2.2-2.1 4.7-4.5 8.9-7.3c-.3 .7-1.3 1.7-2.4 2.7c-2.2 2.1-4.6 4.5-1.6 4.6c5-1.3 18.5-13.4 28.5-22.3l0 0 0 0c.6-.5 1.2-1 1.7-1.5c1.5-1.3 2.8-2.5 4-3.6l0 0 .3-.3c1.9-4.2 1.6-5.6 1.3-7l0-.1c-.4-1.6-.8-3.3 2.4-9.6l7.3-3.7c.8-2.1 1.5-4.1 2.2-6c.2-.6 .5-1.2 .7-1.8l-.4-.2zM349.3 34.3l-.2-.1 .2 .1 0 0zM247.8 334.1c-6-3-13.7-8.9-14.8-11.4l-.4 .3c-.3 .6-.5 1.3-.2 2.2c-12.2-5.7-23.4-14.3-32.6-24.9c4.9 7.1 10.1 14.1 17 19.5c-6.9-2.3-15.1-11.8-21.6-19.3l-.1-.1c-4.3-5-7.9-9.1-9.7-9.5c19.8 35.5 80.5 62.3 112.3 49c-14.7 .5-33.4 .3-49.9-5.8zm79.3-119.7l-.1-.2c-.5-1.5-1.1-3.1-1.7-3.4c1.4-5.8 5.4-10.7 4.4 4.6c-1 3.8-1.8 1.5-2.6-1zm-4.2 22.2c-1.3 7.9-5 15.5-10.1 22.5c.2-2-1.2-2.4-2.6-2.8l0 0c-2.9-.8-5.9-1.6 5.6-16.1c-.5 1.9-2.1 4.6-3.7 7.3l0 0 0 0-.3 .4c-3.6 5.9-6.7 11 4 4.3l1-1.8c2.6-4.5 5-8.8 6-13.8h.1zm-55.6 33.9c7.1 .6 14.1 .6 21-1.1c-2.5 2.4-5.2 4.8-8.3 7.2c-11.1-1.7-21.2-6-12.7-6.1zm-92.6 11.6c3.6 7.1 6.4 11.5 9 15.7l.1 .2c2.3 3.7 4.4 7.1 6.8 11.7c-5.1-4.2-8.7-9.5-12.5-15l-.3-.5c-1.4-2.1-2.8-4.2-4.4-6.2l1.2-5.9h.1zm7.5-9.6c1.6 3.3 3.2 6.4 5.7 9.1l2.6 7.7-1.3-2.1c-3.2-5.3-6.3-10.6-8-16.7l.8 1.6 .2 .4zm238.9-41.6c-2.3 17.4-7.7 34.6-16 50.3c7.6-14.9 12.5-30.9 14.8-47.2l1.2-3.1zM35.6 110.6c.4 .8 1.4 .5 2.3 .3c1.9-.5 3.6-.9-.1 7.6c-.5 .3-1 .7-1.5 1l0 0 0 0c-1.4 .9-2.8 1.9-3.9 3c1.9-3.8 3.5-7.4 3.2-11.9zM25.3 152.3c-.7 3.7-1.5 7.9-3.4 13.9c.2-1.9 0-3.5-.2-4.9l0-.1c-.4-3.4-.7-6.3 4.3-12.8c-.3 1.2-.5 2.5-.7 3.8v.1z\"],openid:[448,512,[],\"f19b\",\"M271.5 432l-68 32C88.5 453.7 0 392.5 0 318.2c0-71.5 82.5-131 191.7-144.3v43c-71.5 12.5-124 53-124 101.3 0 51 58.5 93.3 135.7 103v-340l68-33.2v384zM448 291l-131.3-28.5 36.8-20.7c-19.5-11.5-43.5-20-70-24.8v-43c46.2 5.5 87.7 19.5 120.3 39.3l35-19.8L448 291z\"],instalod:[512,512,[],\"e081\",\"M153.384,480H387.113L502.554,275.765,204.229,333.211ZM504.726,240.078,387.113,32H155.669L360.23,267.9ZM124.386,48.809,7.274,256,123.236,461.154,225.627,165.561Z\"],expeditedssl:[496,512,[],\"f23e\",\"M248 43.4C130.6 43.4 35.4 138.6 35.4 256S130.6 468.6 248 468.6 460.6 373.4 460.6 256 365.4 43.4 248 43.4zm-97.4 132.9c0-53.7 43.7-97.4 97.4-97.4s97.4 43.7 97.4 97.4v26.6c0 5-3.9 8.9-8.9 8.9h-17.7c-5 0-8.9-3.9-8.9-8.9v-26.6c0-82.1-124-82.1-124 0v26.6c0 5-3.9 8.9-8.9 8.9h-17.7c-5 0-8.9-3.9-8.9-8.9v-26.6zM389.7 380c0 9.7-8 17.7-17.7 17.7H124c-9.7 0-17.7-8-17.7-17.7V238.3c0-9.7 8-17.7 17.7-17.7h248c9.7 0 17.7 8 17.7 17.7V380zm-248-137.3v132.9c0 2.5-1.9 4.4-4.4 4.4h-8.9c-2.5 0-4.4-1.9-4.4-4.4V242.7c0-2.5 1.9-4.4 4.4-4.4h8.9c2.5 0 4.4 1.9 4.4 4.4zm141.7 48.7c0 13-7.2 24.4-17.7 30.4v31.6c0 5-3.9 8.9-8.9 8.9h-17.7c-5 0-8.9-3.9-8.9-8.9v-31.6c-10.5-6.1-17.7-17.4-17.7-30.4 0-19.7 15.8-35.4 35.4-35.4s35.5 15.8 35.5 35.4zM248 8C111 8 0 119 0 256s111 248 248 248 248-111 248-248S385 8 248 8zm0 478.3C121 486.3 17.7 383 17.7 256S121 25.7 248 25.7 478.3 129 478.3 256 375 486.3 248 486.3z\"],sellcast:[448,512,[],\"f2da\",\"M353.4 32H94.7C42.6 32 0 74.6 0 126.6v258.7C0 437.4 42.6 480 94.7 480h258.7c52.1 0 94.7-42.6 94.7-94.6V126.6c0-52-42.6-94.6-94.7-94.6zm-50 316.4c-27.9 48.2-89.9 64.9-138.2 37.2-22.9 39.8-54.9 8.6-42.3-13.2l15.7-27.2c5.9-10.3 19.2-13.9 29.5-7.9 18.6 10.8-.1-.1 18.5 10.7 27.6 15.9 63.4 6.3 79.4-21.3 15.9-27.6 6.3-63.4-21.3-79.4-17.8-10.2-.6-.4-18.6-10.6-24.6-14.2-3.4-51.9 21.6-37.5 18.6 10.8-.1-.1 18.5 10.7 48.4 28 65.1 90.3 37.2 138.5zm21.8-208.8c-17 29.5-16.3 28.8-19 31.5-6.5 6.5-16.3 8.7-26.5 3.6-18.6-10.8.1.1-18.5-10.7-27.6-15.9-63.4-6.3-79.4 21.3s-6.3 63.4 21.3 79.4c0 0 18.5 10.6 18.6 10.6 24.6 14.2 3.4 51.9-21.6 37.5-18.6-10.8.1.1-18.5-10.7-48.2-27.8-64.9-90.1-37.1-138.4 27.9-48.2 89.9-64.9 138.2-37.2l4.8-8.4c14.3-24.9 52-3.3 37.7 21.5z\"],\"square-twitter\":[448,512,[\"twitter-square\"],\"f081\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM351.3 199.3v0c0 86.7-66 186.6-186.6 186.6c-37.2 0-71.7-10.8-100.7-29.4c5.3 .6 10.4 .8 15.8 .8c30.7 0 58.9-10.4 81.4-28c-28.8-.6-53-19.5-61.3-45.5c10.1 1.5 19.2 1.5 29.6-1.2c-30-6.1-52.5-32.5-52.5-64.4v-.8c8.7 4.9 18.9 7.9 29.6 8.3c-9-6-16.4-14.1-21.5-23.6s-7.8-20.2-7.7-31c0-12.2 3.2-23.4 8.9-33.1c32.3 39.8 80.8 65.8 135.2 68.6c-9.3-44.5 24-80.6 64-80.6c18.9 0 35.9 7.9 47.9 20.7c14.8-2.8 29-8.3 41.6-15.8c-4.9 15.2-15.2 28-28.8 36.1c13.2-1.4 26-5.1 37.8-10.2c-8.9 13.1-20.1 24.7-32.9 34c.2 2.8 .2 5.7 .2 8.5z\"],\"r-project\":[581,512,[],\"f4f7\",\"M581 226.6C581 119.1 450.9 32 290.5 32S0 119.1 0 226.6C0 322.4 103.3 402 239.4 418.1V480h99.1v-61.5c24.3-2.7 47.6-7.4 69.4-13.9L448 480h112l-67.4-113.7c54.5-35.4 88.4-84.9 88.4-139.7zm-466.8 14.5c0-73.5 98.9-133 220.8-133s211.9 40.7 211.9 133c0 50.1-26.5 85-70.3 106.4-2.4-1.6-4.7-2.9-6.4-3.7-10.2-5.2-27.8-10.5-27.8-10.5s86.6-6.4 86.6-92.7-90.6-87.9-90.6-87.9h-199V361c-74.1-21.5-125.2-67.1-125.2-119.9zm225.1 38.3v-55.6c57.8 0 87.8-6.8 87.8 27.3 0 36.5-38.2 28.3-87.8 28.3zm-.9 72.5H365c10.8 0 18.9 11.7 24 19.2-16.1 1.9-33 2.8-50.6 2.9v-22.1z\"],delicious:[448,512,[],\"f1a5\",\"M446.5 68c-.4-1.5-.9-3-1.4-4.5-.9-2.5-2-4.8-3.3-7.1-1.4-2.4-3-4.8-4.7-6.9-2.1-2.5-4.4-4.8-6.9-6.8-1.1-.9-2.2-1.7-3.3-2.5-1.3-.9-2.6-1.7-4-2.4-1.8-1-3.6-1.8-5.5-2.5-1.7-.7-3.5-1.3-5.4-1.7-3.8-1-7.9-1.5-12-1.5H48C21.5 32 0 53.5 0 80v352c0 4.1.5 8.2 1.5 12 2 7.7 5.8 14.6 11 20.3 1 1.1 2.1 2.2 3.3 3.3 5.7 5.2 12.6 9 20.3 11 3.8 1 7.9 1.5 12 1.5h352c26.5 0 48-21.5 48-48V80c-.1-4.1-.6-8.2-1.6-12zM416 432c0 8.8-7.2 16-16 16H224V256H32V80c0-8.8 7.2-16 16-16h176v192h192z\"],freebsd:[448,512,[],\"f3a4\",\"M303.7 96.2c11.1-11.1 115.5-77 139.2-53.2 23.7 23.7-42.1 128.1-53.2 139.2-11.1 11.1-39.4.9-63.1-22.9-23.8-23.7-34.1-52-22.9-63.1zM109.9 68.1C73.6 47.5 22 24.6 5.6 41.1c-16.6 16.6 7.1 69.4 27.9 105.7 18.5-32.2 44.8-59.3 76.4-78.7zM406.7 174c3.3 11.3 2.7 20.7-2.7 26.1-20.3 20.3-87.5-27-109.3-70.1-18-32.3-11.1-53.4 14.9-48.7 5.7-3.6 12.3-7.6 19.6-11.6-29.8-15.5-63.6-24.3-99.5-24.3-119.1 0-215.6 96.5-215.6 215.6 0 119 96.5 215.6 215.6 215.6S445.3 380.1 445.3 261c0-38.4-10.1-74.5-27.7-105.8-3.9 7-7.6 13.3-10.9 18.8z\"],vuejs:[448,512,[],\"f41f\",\"M356.9 64.3H280l-56 88.6-48-88.6H0L224 448 448 64.3h-91.1zm-301.2 32h53.8L224 294.5 338.4 96.3h53.8L224 384.5 55.7 96.3z\"],accusoft:[640,512,[],\"f369\",\"M322.1 252v-1l-51.2-65.8s-12 1.6-25 15.1c-9 9.3-242.1 239.1-243.4 240.9-7 10 1.6 6.8 15.7 1.7.8 0 114.5-36.6 114.5-36.6.5-.6-.1-.1.6-.6-.4-5.1-.8-26.2-1-27.7-.6-5.2 2.2-6.9 7-8.9l92.6-33.8c.6-.8 88.5-81.7 90.2-83.3zm160.1 120.1c13.3 16.1 20.7 13.3 30.8 9.3 3.2-1.2 115.4-47.6 117.8-48.9 8-4.3-1.7-16.7-7.2-23.4-2.1-2.5-205.1-245.6-207.2-248.3-9.7-12.2-14.3-12.9-38.4-12.8-10.2 0-106.8.5-116.5.6-19.2.1-32.9-.3-19.2 16.9C250 75 476.5 365.2 482.2 372.1zm152.7 1.6c-2.3-.3-24.6-4.7-38-7.2 0 0-115 50.4-117.5 51.6-16 7.3-26.9-3.2-36.7-14.6l-57.1-74c-5.4-.9-60.4-9.6-65.3-9.3-3.1.2-9.6.8-14.4 2.9-4.9 2.1-145.2 52.8-150.2 54.7-5.1 2-11.4 3.6-11.1 7.6.2 2.5 2 2.6 4.6 3.5 2.7.8 300.9 67.6 308 69.1 15.6 3.3 38.5 10.5 53.6 1.7 2.1-1.2 123.8-76.4 125.8-77.8 5.4-4 4.3-6.8-1.7-8.2z\"],ioxhost:[640,512,[],\"f208\",\"M616 160h-67.3C511.2 70.7 422.9 8 320 8 183 8 72 119 72 256c0 16.4 1.6 32.5 4.7 48H24c-13.3 0-24 10.8-24 24 0 13.3 10.7 24 24 24h67.3c37.5 89.3 125.8 152 228.7 152 137 0 248-111 248-248 0-16.4-1.6-32.5-4.7-48H616c13.3 0 24-10.8 24-24 0-13.3-10.7-24-24-24zm-96 96c0 110.5-89.5 200-200 200-75.7 0-141.6-42-175.5-104H424c13.3 0 24-10.8 24-24 0-13.3-10.7-24-24-24H125.8c-3.8-15.4-5.8-31.4-5.8-48 0-110.5 89.5-200 200-200 75.7 0 141.6 42 175.5 104H216c-13.3 0-24 10.8-24 24 0 13.3 10.7 24 24 24h298.2c3.8 15.4 5.8 31.4 5.8 48zm-304-24h208c13.3 0 24 10.7 24 24 0 13.2-10.7 24-24 24H216c-13.3 0-24-10.7-24-24 0-13.2 10.7-24 24-24z\"],\"fonticons-fi\":[384,512,[],\"f3a2\",\"M114.4 224h92.4l-15.2 51.2h-76.4V433c0 8-2.8 9.2 4.4 10l59.6 5.6V483H0v-35.2l29.2-2.8c7.2-.8 9.2-3.2 9.2-10.8V278.4c0-3.2-4-3.2-8-3.2H0V224h38.4v-28.8c0-68 36.4-96 106-96 46.8 0 88.8 11.2 88.8 72.4l-69.6 8.4c.4-25.6-6-31.6-22.4-31.6-25.2 0-26 13.6-26 37.6v32c0 3.2-4.8 6-.8 6zM384 483H243.2v-34.4l28-3.6c7.2-.8 10.4-2.4 10.4-10V287c0-5.6-4-9.2-9.2-10.8l-33.2-8.8 9.2-40.4h110v208c0 8-3.6 8.8 4 10l21.6 3.6V483zm-30-347.2l12.4 45.6-10 10-42.8-22.8-42.8 22.8-10-10 12.4-45.6-30-36.4 4.8-10h38L307.2 51H320l21.2 38.4h38l4.8 13.2-30 33.2z\"],\"app-store\":[512,512,[],\"f36f\",\"M255.9 120.9l9.1-15.7c5.6-9.8 18.1-13.1 27.9-7.5 9.8 5.6 13.1 18.1 7.5 27.9l-87.5 151.5h63.3c20.5 0 32 24.1 23.1 40.8H113.8c-11.3 0-20.4-9.1-20.4-20.4 0-11.3 9.1-20.4 20.4-20.4h52l66.6-115.4-20.8-36.1c-5.6-9.8-2.3-22.2 7.5-27.9 9.8-5.6 22.2-2.3 27.9 7.5l8.9 15.7zm-78.7 218l-19.6 34c-5.6 9.8-18.1 13.1-27.9 7.5-9.8-5.6-13.1-18.1-7.5-27.9l14.6-25.2c16.4-5.1 29.8-1.2 40.4 11.6zm168.9-61.7h53.1c11.3 0 20.4 9.1 20.4 20.4 0 11.3-9.1 20.4-20.4 20.4h-29.5l19.9 34.5c5.6 9.8 2.3 22.2-7.5 27.9-9.8 5.6-22.2 2.3-27.9-7.5-33.5-58.1-58.7-101.6-75.4-130.6-17.1-29.5-4.9-59.1 7.2-69.1 13.4 23 33.4 57.7 60.1 104zM256 8C119 8 8 119 8 256s111 248 248 248 248-111 248-248S393 8 256 8zm216 248c0 118.7-96.1 216-216 216-118.7 0-216-96.1-216-216 0-118.7 96.1-216 216-216 118.7 0 216 96.1 216 216z\"],\"cc-mastercard\":[576,512,[],\"f1f1\",\"M482.9 410.3c0 6.8-4.6 11.7-11.2 11.7-6.8 0-11.2-5.2-11.2-11.7 0-6.5 4.4-11.7 11.2-11.7 6.6 0 11.2 5.2 11.2 11.7zm-310.8-11.7c-7.1 0-11.2 5.2-11.2 11.7 0 6.5 4.1 11.7 11.2 11.7 6.5 0 10.9-4.9 10.9-11.7-.1-6.5-4.4-11.7-10.9-11.7zm117.5-.3c-5.4 0-8.7 3.5-9.5 8.7h19.1c-.9-5.7-4.4-8.7-9.6-8.7zm107.8.3c-6.8 0-10.9 5.2-10.9 11.7 0 6.5 4.1 11.7 10.9 11.7 6.8 0 11.2-4.9 11.2-11.7 0-6.5-4.4-11.7-11.2-11.7zm105.9 26.1c0 .3.3.5.3 1.1 0 .3-.3.5-.3 1.1-.3.3-.3.5-.5.8-.3.3-.5.5-1.1.5-.3.3-.5.3-1.1.3-.3 0-.5 0-1.1-.3-.3 0-.5-.3-.8-.5-.3-.3-.5-.5-.5-.8-.3-.5-.3-.8-.3-1.1 0-.5 0-.8.3-1.1 0-.5.3-.8.5-1.1.3-.3.5-.3.8-.5.5-.3.8-.3 1.1-.3.5 0 .8 0 1.1.3.5.3.8.3 1.1.5s.2.6.5 1.1zm-2.2 1.4c.5 0 .5-.3.8-.3.3-.3.3-.5.3-.8 0-.3 0-.5-.3-.8-.3 0-.5-.3-1.1-.3h-1.6v3.5h.8V426h.3l1.1 1.4h.8l-1.1-1.3zM576 81v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V81c0-26.5 21.5-48 48-48h480c26.5 0 48 21.5 48 48zM64 220.6c0 76.5 62.1 138.5 138.5 138.5 27.2 0 53.9-8.2 76.5-23.1-72.9-59.3-72.4-171.2 0-230.5-22.6-15-49.3-23.1-76.5-23.1-76.4-.1-138.5 62-138.5 138.2zm224 108.8c70.5-55 70.2-162.2 0-217.5-70.2 55.3-70.5 162.6 0 217.5zm-142.3 76.3c0-8.7-5.7-14.4-14.7-14.7-4.6 0-9.5 1.4-12.8 6.5-2.4-4.1-6.5-6.5-12.2-6.5-3.8 0-7.6 1.4-10.6 5.4V392h-8.2v36.7h8.2c0-18.9-2.5-30.2 9-30.2 10.2 0 8.2 10.2 8.2 30.2h7.9c0-18.3-2.5-30.2 9-30.2 10.2 0 8.2 10 8.2 30.2h8.2v-23zm44.9-13.7h-7.9v4.4c-2.7-3.3-6.5-5.4-11.7-5.4-10.3 0-18.2 8.2-18.2 19.3 0 11.2 7.9 19.3 18.2 19.3 5.2 0 9-1.9 11.7-5.4v4.6h7.9V392zm40.5 25.6c0-15-22.9-8.2-22.9-15.2 0-5.7 11.9-4.8 18.5-1.1l3.3-6.5c-9.4-6.1-30.2-6-30.2 8.2 0 14.3 22.9 8.3 22.9 15 0 6.3-13.5 5.8-20.7.8l-3.5 6.3c11.2 7.6 32.6 6 32.6-7.5zm35.4 9.3l-2.2-6.8c-3.8 2.1-12.2 4.4-12.2-4.1v-16.6h13.1V392h-13.1v-11.2h-8.2V392h-7.6v7.3h7.6V416c0 17.6 17.3 14.4 22.6 10.9zm13.3-13.4h27.5c0-16.2-7.4-22.6-17.4-22.6-10.6 0-18.2 7.9-18.2 19.3 0 20.5 22.6 23.9 33.8 14.2l-3.8-6c-7.8 6.4-19.6 5.8-21.9-4.9zm59.1-21.5c-4.6-2-11.6-1.8-15.2 4.4V392h-8.2v36.7h8.2V408c0-11.6 9.5-10.1 12.8-8.4l2.4-7.6zm10.6 18.3c0-11.4 11.6-15.1 20.7-8.4l3.8-6.5c-11.6-9.1-32.7-4.1-32.7 15 0 19.8 22.4 23.8 32.7 15l-3.8-6.5c-9.2 6.5-20.7 2.6-20.7-8.6zm66.7-18.3H408v4.4c-8.3-11-29.9-4.8-29.9 13.9 0 19.2 22.4 24.7 29.9 13.9v4.6h8.2V392zm33.7 0c-2.4-1.2-11-2.9-15.2 4.4V392h-7.9v36.7h7.9V408c0-11 9-10.3 12.8-8.4l2.4-7.6zm40.3-14.9h-7.9v19.3c-8.2-10.9-29.9-5.1-29.9 13.9 0 19.4 22.5 24.6 29.9 13.9v4.6h7.9v-51.7zm7.6-75.1v4.6h.8V302h1.9v-.8h-4.6v.8h1.9zm6.6 123.8c0-.5 0-1.1-.3-1.6-.3-.3-.5-.8-.8-1.1-.3-.3-.8-.5-1.1-.8-.5 0-1.1-.3-1.6-.3-.3 0-.8.3-1.4.3-.5.3-.8.5-1.1.8-.5.3-.8.8-.8 1.1-.3.5-.3 1.1-.3 1.6 0 .3 0 .8.3 1.4 0 .3.3.8.8 1.1.3.3.5.5 1.1.8.5.3 1.1.3 1.4.3.5 0 1.1 0 1.6-.3.3-.3.8-.5 1.1-.8.3-.3.5-.8.8-1.1.3-.6.3-1.1.3-1.4zm3.2-124.7h-1.4l-1.6 3.5-1.6-3.5h-1.4v5.4h.8v-4.1l1.6 3.5h1.1l1.4-3.5v4.1h1.1v-5.4zm4.4-80.5c0-76.2-62.1-138.3-138.5-138.3-27.2 0-53.9 8.2-76.5 23.1 72.1 59.3 73.2 171.5 0 230.5 22.6 15 49.5 23.1 76.5 23.1 76.4.1 138.5-61.9 138.5-138.4z\"],\"itunes-note\":[384,512,[],\"f3b5\",\"M381.9 388.2c-6.4 27.4-27.2 42.8-55.1 48-24.5 4.5-44.9 5.6-64.5-10.2-23.9-20.1-24.2-53.4-2.7-74.4 17-16.2 40.9-19.5 76.8-25.8 6-1.1 11.2-2.5 15.6-7.4 6.4-7.2 4.4-4.1 4.4-163.2 0-11.2-5.5-14.3-17-12.3-8.2 1.4-185.7 34.6-185.7 34.6-10.2 2.2-13.4 5.2-13.4 16.7 0 234.7 1.1 223.9-2.5 239.5-4.2 18.2-15.4 31.9-30.2 39.5-16.8 9.3-47.2 13.4-63.4 10.4-43.2-8.1-58.4-58-29.1-86.6 17-16.2 40.9-19.5 76.8-25.8 6-1.1 11.2-2.5 15.6-7.4 10.1-11.5 1.8-256.6 5.2-270.2.8-5.2 3-9.6 7.1-12.9 4.2-3.5 11.8-5.5 13.4-5.5 204-38.2 228.9-43.1 232.4-43.1 11.5-.8 18.1 6 18.1 17.6.2 344.5 1.1 326-1.8 338.5z\"],golang:[640,512,[],\"e40f\",\"M400.1 194.8C389.2 197.6 380.2 199.1 371 202.4C363.7 204.3 356.3 206.3 347.8 208.5L347.2 208.6C343 209.8 342.6 209.9 338.7 205.4C334 200.1 330.6 196.7 324.1 193.5C304.4 183.9 285.4 186.7 267.7 198.2C246.5 211.9 235.6 232.2 235.9 257.4C236.2 282.4 253.3 302.9 277.1 306.3C299.1 309.1 316.9 301.7 330.9 285.8C333 283.2 334.9 280.5 337 277.5V277.5L337 277.5C337.8 276.5 338.5 275.4 339.3 274.2H279.2C272.7 274.2 271.1 270.2 273.3 264.9C277.3 255.2 284.8 239 289.2 230.9C290.1 229.1 292.3 225.1 296.1 225.1H397.2C401.7 211.7 409 198.2 418.8 185.4C441.5 155.5 468.1 139.9 506 133.4C537.8 127.8 567.7 130.9 594.9 149.3C619.5 166.1 634.7 188.9 638.8 218.8C644.1 260.9 631.9 295.1 602.1 324.4C582.4 345.3 557.2 358.4 528.2 364.3C522.6 365.3 517.1 365.8 511.7 366.3C508.8 366.5 506 366.8 503.2 367.1C474.9 366.5 449 358.4 427.2 339.7C411.9 326.4 401.3 310.1 396.1 291.2C392.4 298.5 388.1 305.6 382.1 312.3C360.5 341.9 331.2 360.3 294.2 365.2C263.6 369.3 235.3 363.4 210.3 344.7C187.3 327.2 174.2 304.2 170.8 275.5C166.7 241.5 176.7 210.1 197.2 184.2C219.4 155.2 248.7 136.8 284.5 130.3C313.8 124.1 341.8 128.4 367.1 145.6C383.6 156.5 395.4 171.4 403.2 189.5C405.1 192.3 403.8 193.9 400.1 194.8zM48.3 200.4C47.05 200.4 46.74 199.8 47.36 198.8L53.91 190.4C54.53 189.5 56.09 188.9 57.34 188.9H168.6C169.8 188.9 170.1 189.8 169.5 190.7L164.2 198.8C163.6 199.8 162 200.7 161.1 200.7L48.3 200.4zM1.246 229.1C0 229.1-.3116 228.4 .3116 227.5L6.855 219.1C7.479 218.2 9.037 217.5 10.28 217.5H152.4C153.6 217.5 154.2 218.5 153.9 219.4L151.4 226.9C151.1 228.1 149.9 228.8 148.6 228.8L1.246 229.1zM75.72 255.9C75.1 256.8 75.41 257.7 76.65 257.7L144.6 258C145.5 258 146.8 257.1 146.8 255.9L147.4 248.4C147.4 247.1 146.8 246.2 145.5 246.2H83.2C81.95 246.2 80.71 247.1 80.08 248.1L75.72 255.9zM577.2 237.9C577 235.3 576.9 233.1 576.5 230.9C570.9 200.1 542.5 182.6 512.9 189.5C483.9 196 465.2 214.4 458.4 243.7C452.8 268 464.6 292.6 487 302.6C504.2 310.1 521.3 309.2 537.8 300.7C562.4 287.1 575.8 268 577.4 241.2C577.3 240 577.3 238.9 577.2 237.9z\"],kickstarter:[448,512,[\"square-kickstarter\"],\"f3bb\",\"M64 32C28.7 32 0 60.7 0 96L0 416c0 35.3 28.7 64 64 64l320 0c35.3 0 64-28.7 64-64l0-320c0-35.3-28.7-64-64-64L64 32zM320.8 233l-23.3 23.1L320.8 279c24.1 23.9 24.1 63 0 86.9s-63.4 23.9-87.6 0l-8.5-8.4c-11.3 16-29.7 26.5-50.9 26.5c-34.1 0-61.9-27.5-61.9-61.4l0-133.2c0-33.8 27.7-61.4 61.9-61.4c21.1 0 39.6 10.5 50.9 26.5l8.5-8.4c24.1-23.9 63.4-23.9 87.6 0s24.1 63 0 86.9z\"],grav:[512,512,[],\"f2d6\",\"M301.1 212c4.4 4.4 4.4 11.9 0 16.3l-9.7 9.7c-4.4 4.7-11.9 4.7-16.6 0l-10.5-10.5c-4.4-4.7-4.4-11.9 0-16.6l9.7-9.7c4.4-4.4 11.9-4.4 16.6 0l10.5 10.8zm-30.2-19.7c3-3 3-7.8 0-10.5-2.8-3-7.5-3-10.5 0-2.8 2.8-2.8 7.5 0 10.5 3.1 2.8 7.8 2.8 10.5 0zm-26 5.3c-3 2.8-3 7.5 0 10.2 2.8 3 7.5 3 10.5 0 2.8-2.8 2.8-7.5 0-10.2-3-3-7.7-3-10.5 0zm72.5-13.3c-19.9-14.4-33.8-43.2-11.9-68.1 21.6-24.9 40.7-17.2 59.8.8 11.9 11.3 29.3 24.9 17.2 48.2-12.5 23.5-45.1 33.2-65.1 19.1zm47.7-44.5c-8.9-10-23.3 6.9-15.5 16.1 7.4 9 32.1 2.4 15.5-16.1zM504 256c0 137-111 248-248 248S8 393 8 256 119 8 256 8s248 111 248 248zm-66.2 42.6c2.5-16.1-20.2-16.6-25.2-25.7-13.6-24.1-27.7-36.8-54.5-30.4 11.6-8 23.5-6.1 23.5-6.1.3-6.4 0-13-9.4-24.9 3.9-12.5.3-22.4.3-22.4 15.5-8.6 26.8-24.4 29.1-43.2 3.6-31-18.8-59.2-49.8-62.8-22.1-2.5-43.7 7.7-54.3 25.7-23.2 40.1 1.4 70.9 22.4 81.4-14.4-1.4-34.3-11.9-40.1-34.3-6.6-25.7 2.8-49.8 8.9-61.4 0 0-4.4-5.8-8-8.9 0 0-13.8 0-24.6 5.3 11.9-15.2 25.2-14.4 25.2-14.4 0-6.4-.6-14.9-3.6-21.6-5.4-11-23.8-12.9-31.7 2.8.1-.2.3-.4.4-.5-5 11.9-1.1 55.9 16.9 87.2-2.5 1.4-9.1 6.1-13 10-21.6 9.7-56.2 60.3-56.2 60.3-28.2 10.8-77.2 50.9-70.6 79.7.3 3 1.4 5.5 3 7.5-2.8 2.2-5.5 5-8.3 8.3-11.9 13.8-5.3 35.2 17.7 24.4 15.8-7.2 29.6-20.2 36.3-30.4 0 0-5.5-5-16.3-4.4 27.7-6.6 34.3-9.4 46.2-9.1 8 3.9 8-34.3 8-34.3 0-14.7-2.2-31-11.1-41.5 12.5 12.2 29.1 32.7 28 60.6-.8 18.3-15.2 23-15.2 23-9.1 16.6-43.2 65.9-30.4 106 0 0-9.7-14.9-10.2-22.1-17.4 19.4-46.5 52.3-24.6 64.5 26.6 14.7 108.8-88.6 126.2-142.3 34.6-20.8 55.4-47.3 63.9-65 22 43.5 95.3 94.5 101.1 59z\"],weibo:[512,512,[],\"f18a\",\"M407 177.6c7.6-24-13.4-46.8-37.4-41.7-22 4.8-28.8-28.1-7.1-32.8 50.1-10.9 92.3 37.1 76.5 84.8-6.8 21.2-38.8 10.8-32-10.3zM214.8 446.7C108.5 446.7 0 395.3 0 310.4c0-44.3 28-95.4 76.3-143.7C176 67 279.5 65.8 249.9 161c-4 13.1 12.3 5.7 12.3 6 79.5-33.6 140.5-16.8 114 51.4-3.7 9.4 1.1 10.9 8.3 13.1 135.7 42.3 34.8 215.2-169.7 215.2zm143.7-146.3c-5.4-55.7-78.5-94-163.4-85.7-84.8 8.6-148.8 60.3-143.4 116s78.5 94 163.4 85.7c84.8-8.6 148.8-60.3 143.4-116zM347.9 35.1c-25.9 5.6-16.8 43.7 8.3 38.3 72.3-15.2 134.8 52.8 111.7 124-7.4 24.2 29.1 37 37.4 12 31.9-99.8-55.1-195.9-157.4-174.3zm-78.5 311c-17.1 38.8-66.8 60-109.1 46.3-40.8-13.1-58-53.4-40.3-89.7 17.7-35.4 63.1-55.4 103.4-45.1 42 10.8 63.1 50.2 46 88.5zm-86.3-30c-12.9-5.4-30 .3-38 12.9-8.3 12.9-4.3 28 8.6 34 13.1 6 30.8.3 39.1-12.9 8-13.1 3.7-28.3-9.7-34zm32.6-13.4c-5.1-1.7-11.4.6-14.3 5.4-2.9 5.1-1.4 10.6 3.7 12.9 5.1 2 11.7-.3 14.6-5.4 2.8-5.2 1.1-10.9-4-12.9z\"],uncharted:[448,512,[],\"e084\",\"M171.73,232.813A5.381,5.381,0,0,0,176.7,229.5,48.081,48.081,0,0,1,191.6,204.244c1.243-.828,1.657-2.484,1.657-4.141a4.22,4.22,0,0,0-2.071-3.312L74.429,128.473,148.958,85a9.941,9.941,0,0,0,4.968-8.281,9.108,9.108,0,0,0-4.968-8.281L126.6,55.6a9.748,9.748,0,0,0-9.523,0l-100.2,57.966a9.943,9.943,0,0,0-4.969,8.281V236.954a9.109,9.109,0,0,0,4.969,8.281L39.235,258.07a8.829,8.829,0,0,0,4.968,1.242,9.4,9.4,0,0,0,6.625-2.484,10.8,10.8,0,0,0,2.9-7.039V164.5L169.66,232.4A4.5,4.5,0,0,0,171.73,232.813ZM323.272,377.73a12.478,12.478,0,0,0-4.969,1.242l-74.528,43.062V287.882c0-2.9-2.9-5.8-6.211-4.555a53.036,53.036,0,0,1-28.984.414,4.86,4.86,0,0,0-6.21,4.555V421.619l-74.529-43.061a8.83,8.83,0,0,0-4.969-1.242,9.631,9.631,0,0,0-9.523,9.523v26.085a9.107,9.107,0,0,0,4.969,8.281l100.2,57.553A8.829,8.829,0,0,0,223.486,480a11.027,11.027,0,0,0,4.969-1.242l100.2-57.553a9.941,9.941,0,0,0,4.968-8.281V386.839C332.8,382.285,328.24,377.73,323.272,377.73ZM286.007,78a23,23,0,1,0-23-23A23,23,0,0,0,286.007,78Zm63.627-10.086a23,23,0,1,0,23,23A23,23,0,0,0,349.634,67.914ZM412.816,151.6a23,23,0,1,0-23-23A23,23,0,0,0,412.816,151.6Zm-63.182-9.2a23,23,0,1,0,23,23A23,23,0,0,0,349.634,142.4Zm-63.627,83.244a23,23,0,1,0-23-23A23,23,0,0,0,286.007,225.648Zm-62.074,36.358a23,23,0,1,0-23-23A23,23,0,0,0,223.933,262.006Zm188.883-82.358a23,23,0,1,0,23,23A23,23,0,0,0,412.816,179.648Zm0,72.272a23,23,0,1,0,23,23A23,23,0,0,0,412.816,251.92Z\"],firstdraft:[384,512,[],\"f3a1\",\"M384 192h-64v128H192v128H0v-25.6h166.4v-128h128v-128H384V192zm-25.6 38.4v128h-128v128H64V512h192V384h128V230.4h-25.6zm25.6 192h-89.6V512H320v-64h64v-25.6zM0 0v384h128V256h128V128h128V0H0z\"],\"square-youtube\":[448,512,[61798,\"youtube-square\"],\"f431\",\"M282 256.2l-95.2-54.1V310.3L282 256.2zM384 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64zm14.4 136.1c7.6 28.6 7.6 88.2 7.6 88.2s0 59.6-7.6 88.1c-4.2 15.8-16.5 27.7-32.2 31.9C337.9 384 224 384 224 384s-113.9 0-142.2-7.6c-15.7-4.2-28-16.1-32.2-31.9C42 315.9 42 256.3 42 256.3s0-59.7 7.6-88.2c4.2-15.8 16.5-28.2 32.2-32.4C110.1 128 224 128 224 128s113.9 0 142.2 7.7c15.7 4.2 28 16.6 32.2 32.4z\"],\"wikipedia-w\":[640,512,[],\"f266\",\"M640 51.2l-.3 12.2c-28.1.8-45 15.8-55.8 40.3-25 57.8-103.3 240-155.3 358.6H415l-81.9-193.1c-32.5 63.6-68.3 130-99.2 193.1-.3.3-15 0-15-.3C172 352.3 122.8 243.4 75.8 133.4 64.4 106.7 26.4 63.4.2 63.7c0-3.1-.3-10-.3-14.2h161.9v13.9c-19.2 1.1-52.8 13.3-43.3 34.2 21.9 49.7 103.6 240.3 125.6 288.6 15-29.7 57.8-109.2 75.3-142.8-13.9-28.3-58.6-133.9-72.8-160-9.7-17.8-36.1-19.4-55.8-19.7V49.8l142.5.3v13.1c-19.4.6-38.1 7.8-29.4 26.1 18.9 40 30.6 68.1 48.1 104.7 5.6-10.8 34.7-69.4 48.1-100.8 8.9-20.6-3.9-28.6-38.6-29.4.3-3.6 0-10.3.3-13.6 44.4-.3 111.1-.3 123.1-.6v13.6c-22.5.8-45.8 12.8-58.1 31.7l-59.2 122.8c6.4 16.1 63.3 142.8 69.2 156.7L559.2 91.8c-8.6-23.1-36.4-28.1-47.2-28.3V49.6l127.8 1.1.2.5z\"],wpressr:[496,512,[\"rendact\"],\"f3e4\",\"M248 8C111.03 8 0 119.03 0 256s111.03 248 248 248 248-111.03 248-248S384.97 8 248 8zm171.33 158.6c-15.18 34.51-30.37 69.02-45.63 103.5-2.44 5.51-6.89 8.24-12.97 8.24-23.02-.01-46.03.06-69.05-.05-5.12-.03-8.25 1.89-10.34 6.72-10.19 23.56-20.63 47-30.95 70.5-1.54 3.51-4.06 5.29-7.92 5.29-45.94-.01-91.87-.02-137.81 0-3.13 0-5.63-1.15-7.72-3.45-11.21-12.33-22.46-24.63-33.68-36.94-2.69-2.95-2.79-6.18-1.21-9.73 8.66-19.54 17.27-39.1 25.89-58.66 12.93-29.35 25.89-58.69 38.75-88.08 1.7-3.88 4.28-5.68 8.54-5.65 14.24.1 28.48.02 42.72.05 6.24.01 9.2 4.84 6.66 10.59-13.6 30.77-27.17 61.55-40.74 92.33-5.72 12.99-11.42 25.99-17.09 39-3.91 8.95 7.08 11.97 10.95 5.6.23-.37-1.42 4.18 30.01-67.69 1.36-3.1 3.41-4.4 6.77-4.39 15.21.08 30.43.02 45.64.04 5.56.01 7.91 3.64 5.66 8.75-8.33 18.96-16.71 37.9-24.98 56.89-4.98 11.43 8.08 12.49 11.28 5.33.04-.08 27.89-63.33 32.19-73.16 2.02-4.61 5.44-6.51 10.35-6.5 26.43.05 52.86 0 79.29.05 12.44.02 13.93-13.65 3.9-13.64-25.26.03-50.52.02-75.78.02-6.27 0-7.84-2.47-5.27-8.27 5.78-13.06 11.59-26.11 17.3-39.21 1.73-3.96 4.52-5.79 8.84-5.78 23.09.06 25.98.02 130.78.03 6.08-.01 8.03 2.79 5.62 8.27z\"],angellist:[448,512,[],\"f209\",\"M347.1 215.4c11.7-32.6 45.4-126.9 45.4-157.1 0-26.6-15.7-48.9-43.7-48.9-44.6 0-84.6 131.7-97.1 163.1C242 144 196.6 0 156.6 0c-31.1 0-45.7 22.9-45.7 51.7 0 35.3 34.2 126.8 46.6 162-6.3-2.3-13.1-4.3-20-4.3-23.4 0-48.3 29.1-48.3 52.6 0 8.9 4.9 21.4 8 29.7-36.9 10-51.1 34.6-51.1 71.7C46 435.6 114.4 512 210.6 512c118 0 191.4-88.6 191.4-202.9 0-43.1-6.9-82-54.9-93.7zM311.7 108c4-12.3 21.1-64.3 37.1-64.3 8.6 0 10.9 8.9 10.9 16 0 19.1-38.6 124.6-47.1 148l-34-6 33.1-93.7zM142.3 48.3c0-11.9 14.5-45.7 46.3 47.1l34.6 100.3c-15.6-1.3-27.7-3-35.4 1.4-10.9-28.8-45.5-119.7-45.5-148.8zM140 244c29.3 0 67.1 94.6 67.1 107.4 0 5.1-4.9 11.4-10.6 11.4-20.9 0-76.9-76.9-76.9-97.7.1-7.7 12.7-21.1 20.4-21.1zm184.3 186.3c-29.1 32-66.3 48.6-109.7 48.6-59.4 0-106.3-32.6-128.9-88.3-17.1-43.4 3.8-68.3 20.6-68.3 11.4 0 54.3 60.3 54.3 73.1 0 4.9-7.7 8.3-11.7 8.3-16.1 0-22.4-15.5-51.1-51.4-29.7 29.7 20.5 86.9 58.3 86.9 26.1 0 43.1-24.2 38-42 3.7 0 8.3.3 11.7-.6 1.1 27.1 9.1 59.4 41.7 61.7 0-.9 2-7.1 2-7.4 0-17.4-10.6-32.6-10.6-50.3 0-28.3 21.7-55.7 43.7-71.7 8-6 17.7-9.7 27.1-13.1 9.7-3.7 20-8 27.4-15.4-1.1-11.2-5.7-21.1-16.9-21.1-27.7 0-120.6 4-120.6-39.7 0-6.7.1-13.1 17.4-13.1 32.3 0 114.3 8 138.3 29.1 18.1 16.1 24.3 113.2-31 174.7zm-98.6-126c9.7 3.1 19.7 4 29.7 6-7.4 5.4-14 12-20.3 19.1-2.8-8.5-6.2-16.8-9.4-25.1z\"],\"galactic-republic\":[496,512,[],\"f50c\",\"M248 504C111.25 504 0 392.75 0 256S111.25 8 248 8s248 111.25 248 248-111.25 248-248 248zm0-479.47C120.37 24.53 16.53 128.37 16.53 256S120.37 487.47 248 487.47 479.47 383.63 479.47 256 375.63 24.53 248 24.53zm27.62 21.81v24.62a185.933 185.933 0 0 1 83.57 34.54l17.39-17.36c-28.75-22.06-63.3-36.89-100.96-41.8zm-55.37.07c-37.64 4.94-72.16 19.8-100.88 41.85l17.28 17.36h.08c24.07-17.84 52.55-30.06 83.52-34.67V46.41zm12.25 50.17v82.87c-10.04 2.03-19.42 5.94-27.67 11.42l-58.62-58.59-21.93 21.93 58.67 58.67c-5.47 8.23-9.45 17.59-11.47 27.62h-82.9v31h82.9c2.02 10.02 6.01 19.31 11.47 27.54l-58.67 58.69 21.93 21.93 58.62-58.62a77.873 77.873 0 0 0 27.67 11.47v82.9h31v-82.9c10.05-2.03 19.37-6.06 27.62-11.55l58.67 58.69 21.93-21.93-58.67-58.69c5.46-8.23 9.47-17.52 11.5-27.54h82.87v-31h-82.87c-2.02-10.02-6.03-19.38-11.5-27.62l58.67-58.67-21.93-21.93-58.67 58.67c-8.25-5.49-17.57-9.47-27.62-11.5V96.58h-31zm183.24 30.72l-17.36 17.36a186.337 186.337 0 0 1 34.67 83.67h24.62c-4.95-37.69-19.83-72.29-41.93-101.03zm-335.55.13c-22.06 28.72-36.91 63.26-41.85 100.91h24.65c4.6-30.96 16.76-59.45 34.59-83.52l-17.39-17.39zM38.34 283.67c4.92 37.64 19.75 72.18 41.8 100.9l17.36-17.39c-17.81-24.07-29.92-52.57-34.51-83.52H38.34zm394.7 0c-4.61 30.99-16.8 59.5-34.67 83.6l17.36 17.36c22.08-28.74 36.98-63.29 41.93-100.96h-24.62zM136.66 406.38l-17.36 17.36c28.73 22.09 63.3 36.98 100.96 41.93v-24.64c-30.99-4.63-59.53-16.79-83.6-34.65zm222.53.05c-24.09 17.84-52.58 30.08-83.57 34.67v24.57c37.67-4.92 72.21-19.79 100.96-41.85l-17.31-17.39h-.08z\"],\"nfc-directional\":[512,512,[],\"e530\",\"M211.8 488.6C213.4 491.1 213.9 494.2 213.2 497.1C212.6 500 210.8 502.6 208.3 504.2C205.7 505.8 202.7 506.3 199.7 505.7C138.3 491.8 84.1 455.8 47.53 404.5C10.97 353.2-5.395 290.3 1.57 227.7C8.536 165 38.34 107.2 85.29 65.21C132.2 23.2 193-.0131 256 0C257.5 0 258.1 .2931 260.3 .8627C261.7 1.432 262.1 2.267 264 3.319C265.1 4.371 265.9 5.619 266.5 6.993C267 8.367 267.3 9.839 267.3 11.32V112.3L291.8 86.39C292.8 85.31 294 84.44 295.4 83.84C296.7 83.23 298.2 82.9 299.7 82.86C301.2 82.81 302.6 83.06 304 83.59C305.4 84.12 306.7 84.92 307.8 85.94C308.8 86.96 309.7 88.18 310.3 89.54C310.9 90.89 311.3 92.35 311.3 93.84C311.3 95.32 311.1 96.8 310.6 98.18C310 99.57 309.2 100.8 308.2 101.9L264.2 148.5C263.1 149.6 261.9 150.5 260.5 151.1C259 151.7 257.5 152 255.1 152C254.5 152 252.9 151.7 251.5 151.1C250.1 150.5 248.8 149.6 247.8 148.5L203.7 101.9C201.7 99.74 200.6 96.83 200.7 93.84C200.7 90.84 202 87.1 204.2 85.94C206.4 83.88 209.3 82.77 212.3 82.86C215.3 82.94 218.1 84.21 220.2 86.39L244.7 112.4V22.89C188.3 25.64 134.9 48.73 94.23 87.87C53.58 127 28.49 179.6 23.61 235.8C18.73 292 34.38 348.1 67.68 393.7C100.1 439.2 149.7 471.2 204.7 483.6C207.6 484.3 210.2 486.1 211.8 488.6L211.8 488.6zM171.4 126.1C170.6 127.4 169.5 128.5 168.3 129.3C147.8 143.2 131.1 161.9 119.5 183.8C107.9 205.7 101.8 230.1 101.8 254.9C101.8 279.7 107.9 304.1 119.5 325.1C131.1 347.9 147.8 366.6 168.3 380.5C170.8 382.2 172.5 384.8 173 387.8C173.6 390.7 172.1 393.8 171.3 396.2C169.6 398.7 166.1 400.4 164 400.1C161.1 401.5 158 400.9 155.6 399.2C132 383.2 112.8 361.7 99.46 336.5C86.15 311.4 79.19 283.4 79.19 254.9C79.19 226.5 86.15 198.4 99.46 173.3C112.8 148.1 132 126.6 155.6 110.6C156.8 109.8 158.2 109.2 159.6 108.8C161.1 108.5 162.6 108.5 164.1 108.8C165.5 109 166.9 109.6 168.2 110.4C169.5 111.2 170.5 112.3 171.4 113.5C172.2 114.7 172.8 116.1 173.1 117.6C173.4 119.1 173.4 120.6 173.1 122C172.8 123.5 172.3 124.9 171.4 126.1H171.4zM340.9 383.5C341.7 382.3 342.8 381.2 343.1 380.4V380.3C364.4 366.3 381.1 347.6 392.7 325.7C404.2 303.9 410.2 279.5 410.2 254.8C410.2 230.1 404.2 205.7 392.7 183.8C381.1 161.1 364.4 143.3 343.1 129.3C342.8 128.5 341.7 127.4 340.9 126.2C340.1 124.9 339.5 123.5 339.3 122.1C338.1 120.6 339 119.1 339.3 117.7C339.6 116.2 340.2 114.8 341 113.6C341.9 112.4 342.1 111.3 344.2 110.5C345.4 109.7 346.8 109.2 348.3 108.9C349.8 108.6 351.2 108.6 352.7 108.9C354.2 109.2 355.5 109.8 356.8 110.7C380.2 126.7 399.5 148.2 412.7 173.3C426 198.4 432.1 226.4 432.1 254.8C432.1 283.3 426 311.3 412.7 336.4C399.5 361.5 380.2 383 356.8 399C355.5 399.9 354.2 400.5 352.7 400.8C351.2 401.1 349.8 401.1 348.3 400.8C346.8 400.5 345.4 399.1 344.2 399.2C342.1 398.4 341.9 397.3 341 396.1C340.2 394.9 339.6 393.5 339.3 392C339 390.6 338.1 389.1 339.3 387.6C339.5 386.2 340.1 384.8 340.9 383.5V383.5zM312.3 6.307C368.5 19.04 418.7 50.28 455 95.01C485.4 132.6 504.6 178 510.3 226C515.9 274 507.9 322.7 487.1 366.3C466.2 409.9 433.5 446.8 392.6 472.6C351.7 498.3 304.4 512 256 512C254.5 512 253.1 511.7 251.7 511.1C250.3 510.6 249.1 509.7 248 508.7C246.1 507.6 246.1 506.4 245.6 505C245 503.6 244.7 502.2 244.7 500.7V401.5L220.2 427.5C218.1 429.7 215.3 430.1 212.3 431.1C209.3 431.2 206.4 430 204.2 427.1C202 425.9 200.7 423.1 200.7 420.1C200.6 417.1 201.7 414.2 203.7 412L247.8 365.4C249.1 363.2 252.9 362 255.1 362C259.1 362 262 363.2 264.2 365.4L308.2 412C310.3 414.2 311.4 417.1 311.3 420.1C311.2 423.1 309.9 425.9 307.8 427.1C305.6 430 302.7 431.2 299.7 431.1C296.7 430.1 293.8 429.7 291.8 427.5L267.3 401.6V489.1C323.7 486.3 377.1 463.3 417.8 424.1C458.5 384.1 483.6 332.4 488.5 276.2C493.3 219.1 477.7 163.9 444.4 118.3C411.1 72.75 362.4 40.79 307.4 28.36C305.9 28.03 304.6 27.42 303.3 26.57C302.1 25.71 301.1 24.63 300.3 23.37C299.5 22.12 298.1 20.72 298.7 19.26C298.5 17.8 298.5 16.3 298.8 14.85C299.2 13.41 299.8 12.04 300.6 10.82C301.5 9.61 302.6 8.577 303.8 7.784C305.1 6.99 306.5 6.451 307.9 6.198C309.4 5.945 310.9 5.982 312.3 6.307L312.3 6.307zM353.1 256.1C353.1 287.5 335.6 317.2 303.8 339.6C301.7 341.1 299 341.9 296.4 341.6C293.7 341.4 291.2 340.3 289.4 338.4L219.3 268.6C217.1 266.5 215.1 263.6 215.9 260.6C215.9 257.6 217.1 254.7 219.2 252.6C221.4 250.5 224.2 249.3 227.2 249.3C230.2 249.3 233.1 250.5 235.2 252.6L298.3 315.4C319.1 298.3 330.5 277.5 330.5 256.1C330.5 232.2 316.4 209.1 290.8 191C288.3 189.3 286.7 186.7 286.2 183.7C285.7 180.8 286.3 177.7 288.1 175.3C289.8 172.8 292.4 171.2 295.4 170.7C298.3 170.2 301.4 170.8 303.8 172.6C335.6 195 353.1 224.7 353.1 256.1V256.1zM216.7 341.5C213.7 342 210.7 341.3 208.2 339.6C176.5 317.2 158.1 287.5 158.1 256.1C158.1 224.7 176.5 195 208.2 172.6C210.4 171 213.1 170.3 215.7 170.5C218.4 170.8 220.8 171.9 222.7 173.8L292.8 243.6C294.9 245.7 296.1 248.6 296.1 251.6C296.1 254.6 294.1 257.4 292.8 259.6C290.7 261.7 287.8 262.9 284.9 262.9C281.9 262.9 278.1 261.7 276.9 259.6L213.8 196.7C192.9 214 181.6 234.7 181.6 256.1C181.6 279.1 195.7 303.1 221.3 321.1C223.7 322.9 225.4 325.5 225.9 328.5C226.4 331.4 225.7 334.4 224 336.9C222.3 339.3 219.6 341 216.7 341.5L216.7 341.5z\"],skype:[448,512,[],\"f17e\",\"M424.7 299.8c2.9-14 4.7-28.9 4.7-43.8 0-113.5-91.9-205.3-205.3-205.3-14.9 0-29.7 1.7-43.8 4.7C161.3 40.7 137.7 32 112 32 50.2 32 0 82.2 0 144c0 25.7 8.7 49.3 23.3 68.2-2.9 14-4.7 28.9-4.7 43.8 0 113.5 91.9 205.3 205.3 205.3 14.9 0 29.7-1.7 43.8-4.7 19 14.6 42.6 23.3 68.2 23.3 61.8 0 112-50.2 112-112 .1-25.6-8.6-49.2-23.2-68.1zm-194.6 91.5c-65.6 0-120.5-29.2-120.5-65 0-16 9-30.6 29.5-30.6 31.2 0 34.1 44.9 88.1 44.9 25.7 0 42.3-11.4 42.3-26.3 0-18.7-16-21.6-42-28-62.5-15.4-117.8-22-117.8-87.2 0-59.2 58.6-81.1 109.1-81.1 55.1 0 110.8 21.9 110.8 55.4 0 16.9-11.4 31.8-30.3 31.8-28.3 0-29.2-33.5-75-33.5-25.7 0-42 7-42 22.5 0 19.8 20.8 21.8 69.1 33 41.4 9.3 90.7 26.8 90.7 77.6 0 59.1-57.1 86.5-112 86.5z\"],joget:[496,512,[],\"f3b7\",\"M378.1 45C337.6 19.9 292.6 8 248.2 8 165 8 83.8 49.9 36.9 125.9c-71.9 116.6-35.6 269.3 81 341.2s269.3 35.6 341.2-80.9c71.9-116.6 35.6-269.4-81-341.2zm51.8 323.2c-40.4 65.5-110.4 101.5-182 101.5-6.8 0-13.6-.4-20.4-1-9-13.6-19.9-33.3-23.7-42.4-5.7-13.7-27.2-45.6 31.2-67.1 51.7-19.1 176.7-16.5 208.8-17.6-4 9-8.6 17.9-13.9 26.6zm-200.8-86.3c-55.5-1.4-81.7-20.8-58.5-48.2s51.1-40.7 68.9-51.2c17.9-10.5 27.3-33.7-23.6-29.7C87.3 161.5 48.6 252.1 37.6 293c-8.8-49.7-.1-102.7 28.5-149.1C128 43.4 259.6 12.2 360.1 74.1c74.8 46.1 111.2 130.9 99.3 212.7-24.9-.5-179.3-3.6-230.3-4.9zm183.8-54.8c-22.7-6-57 11.3-86.7 27.2-29.7 15.8-31.1 8.2-31.1 8.2s40.2-28.1 50.7-34.5 31.9-14 13.4-24.6c-3.2-1.8-6.7-2.7-10.4-2.7-17.8 0-41.5 18.7-67.5 35.6-31.5 20.5-65.3 31.3-65.3 31.3l169.5-1.6 46.5-23.4s3.6-9.5-19.1-15.5z\"],fedora:[448,512,[],\"f798\",\"M.0413 255.8C.1219 132.2 100.3 32 224 32C347.7 32 448 132.3 448 256C448 379.7 347.8 479.9 224.1 480H50.93C22.84 480 .0832 457.3 .0416 429.2H0V255.8H.0413zM342.6 192.7C342.6 153 307 124.2 269.4 124.2C234.5 124.2 203.6 150.5 199.3 184.1C199.1 187.9 198.9 189.1 198.9 192.6C198.8 213.7 198.9 235.4 198.1 257C199 283.1 199.1 309.1 198.1 333.6C198.1 360.7 178.7 379.1 153.4 379.1C128.1 379.1 107.6 358.9 107.6 333.6C108.1 305.9 130.2 288.3 156.1 287.5H156.3L182.6 287.3V250L156.3 250.2C109.2 249.8 71.72 286.7 70.36 333.6C70.36 379.2 107.9 416.5 153.4 416.5C196.4 416.5 232.1 382.9 236 340.9L236.2 287.4L268.8 287.1C294.1 287.3 293.8 249.3 268.6 249.8L236.2 250.1C236.2 243.7 236.3 237.3 236.3 230.9C236.4 218.2 236.4 205.5 236.2 192.7C236.3 176.2 252 161.5 269.4 161.5C286.9 161.5 305.3 170.2 305.3 192.7C305.3 195.9 305.2 197.8 305 199C303.1 209.5 310.2 219.4 320.7 220.9C331.3 222.4 340.9 214.8 341.9 204.3C342.5 200.1 342.6 196.4 342.6 192.7H342.6z\"],\"stripe-s\":[384,512,[],\"f42a\",\"M155.3 154.6c0-22.3 18.6-30.9 48.4-30.9 43.4 0 98.5 13.3 141.9 36.7V26.1C298.3 7.2 251.1 0 203.8 0 88.1 0 11 60.4 11 161.4c0 157.9 216.8 132.3 216.8 200.4 0 26.4-22.9 34.9-54.7 34.9-47.2 0-108.2-19.5-156.1-45.5v128.5a396.09 396.09 0 0 0 156 32.4c118.6 0 200.3-51 200.3-153.6 0-170.2-218-139.7-218-203.9z\"],meta:[640,512,[],\"e49b\",\"M640 317.9C640 409.2 600.6 466.4 529.7 466.4C467.1 466.4 433.9 431.8 372.8 329.8L341.4 277.2C333.1 264.7 326.9 253 320.2 242.2C300.1 276 273.1 325.2 273.1 325.2C206.1 441.8 168.5 466.4 116.2 466.4C43.42 466.4 0 409.1 0 320.5C0 177.5 79.78 42.4 183.9 42.4C234.1 42.4 277.7 67.08 328.7 131.9C365.8 81.8 406.8 42.4 459.3 42.4C558.4 42.4 640 168.1 640 317.9H640zM287.4 192.2C244.5 130.1 216.5 111.7 183 111.7C121.1 111.7 69.22 217.8 69.22 321.7C69.22 370.2 87.7 397.4 118.8 397.4C149 397.4 167.8 378.4 222 293.6C222 293.6 246.7 254.5 287.4 192.2V192.2zM531.2 397.4C563.4 397.4 578.1 369.9 578.1 322.5C578.1 198.3 523.8 97.08 454.9 97.08C421.7 97.08 393.8 123 360 175.1C369.4 188.9 379.1 204.1 389.3 220.5L426.8 282.9C485.5 377 500.3 397.4 531.2 397.4L531.2 397.4z\"],laravel:[512,512,[],\"f3bd\",\"M504.4,115.83a5.72,5.72,0,0,0-.28-.68,8.52,8.52,0,0,0-.53-1.25,6,6,0,0,0-.54-.71,9.36,9.36,0,0,0-.72-.94c-.23-.22-.52-.4-.77-.6a8.84,8.84,0,0,0-.9-.68L404.4,55.55a8,8,0,0,0-8,0L300.12,111h0a8.07,8.07,0,0,0-.88.69,7.68,7.68,0,0,0-.78.6,8.23,8.23,0,0,0-.72.93c-.17.24-.39.45-.54.71a9.7,9.7,0,0,0-.52,1.25c-.08.23-.21.44-.28.68a8.08,8.08,0,0,0-.28,2.08V223.18l-80.22,46.19V63.44a7.8,7.8,0,0,0-.28-2.09c-.06-.24-.2-.45-.28-.68a8.35,8.35,0,0,0-.52-1.24c-.14-.26-.37-.47-.54-.72a9.36,9.36,0,0,0-.72-.94,9.46,9.46,0,0,0-.78-.6,9.8,9.8,0,0,0-.88-.68h0L115.61,1.07a8,8,0,0,0-8,0L11.34,56.49h0a6.52,6.52,0,0,0-.88.69,7.81,7.81,0,0,0-.79.6,8.15,8.15,0,0,0-.71.93c-.18.25-.4.46-.55.72a7.88,7.88,0,0,0-.51,1.24,6.46,6.46,0,0,0-.29.67,8.18,8.18,0,0,0-.28,2.1v329.7a8,8,0,0,0,4,6.95l192.5,110.84a8.83,8.83,0,0,0,1.33.54c.21.08.41.2.63.26a7.92,7.92,0,0,0,4.1,0c.2-.05.37-.16.55-.22a8.6,8.6,0,0,0,1.4-.58L404.4,400.09a8,8,0,0,0,4-6.95V287.88l92.24-53.11a8,8,0,0,0,4-7V117.92A8.63,8.63,0,0,0,504.4,115.83ZM111.6,17.28h0l80.19,46.15-80.2,46.18L31.41,63.44Zm88.25,60V278.6l-46.53,26.79-33.69,19.4V123.5l46.53-26.79Zm0,412.78L23.37,388.5V77.32L57.06,96.7l46.52,26.8V338.68a6.94,6.94,0,0,0,.12.9,8,8,0,0,0,.16,1.18h0a5.92,5.92,0,0,0,.38.9,6.38,6.38,0,0,0,.42,1v0a8.54,8.54,0,0,0,.6.78,7.62,7.62,0,0,0,.66.84l0,0c.23.22.52.38.77.58a8.93,8.93,0,0,0,.86.66l0,0,0,0,92.19,52.18Zm8-106.17-80.06-45.32,84.09-48.41,92.26-53.11,80.13,46.13-58.8,33.56Zm184.52,4.57L215.88,490.11V397.8L346.6,323.2l45.77-26.15Zm0-119.13L358.68,250l-46.53-26.79V131.79l33.69,19.4L392.37,178Zm8-105.28-80.2-46.17,80.2-46.16,80.18,46.15Zm8,105.28V178L455,151.19l33.68-19.4v91.39h0Z\"],hotjar:[512,512,[],\"f3b1\",\"M361.5 0c0 131.6-80.7 176.8-140.2 209.4c-.6 .3-1.1 .6-1.6 .9c-53.8 30.2-88.7 49.8-89.6 122H32C32 200.8 112.7 155.6 172.2 123C227 93.2 262.5 73 262.5 0h98.9zM301 302.6c54.8-29.8 90.3-50 90.3-123h98c0 131.6-80.7 176.7-140.2 209.4c-54.8 29.8-90.3 50-90.3 123h-98c0-131.6 80.7-176.8 140.2-209.4z\"],\"bluetooth-b\":[320,512,[],\"f294\",\"M196.48 260.023l92.626-103.333L143.125 0v206.33l-86.111-86.111-31.406 31.405 108.061 108.399L25.608 368.422l31.406 31.405 86.111-86.111L145.84 512l148.552-148.644-97.912-103.333zm40.86-102.996l-49.977 49.978-.338-100.295 50.315 50.317zM187.363 313.04l49.977 49.978-50.315 50.316.338-100.294z\"],\"square-letterboxd\":[448,512,[],\"e62e\",\"M384 32c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H384zM105.1 187C66.4 187 35 218.3 35 257s31.4 70 70.1 70c24.8 0 46.5-12.8 59-32.2l.5-.7-.4-.6c-6.5-10.6-10.2-23.1-10.2-36.5c0-13.6 3.9-26.3 10.6-37.1c-12.4-19.8-34.4-32.9-59.5-32.9zM224 187c-24.8 0-46.5 12.8-59 32.2l-.5 .7 .4 .6c6.5 10.6 10.2 23.1 10.2 36.5c0 13.6-3.9 26.3-10.6 37.1C176.9 313.8 198.9 327 224 327c24.8 0 46.5-12.8 59-32.2l.5-.7-.4-.6c-6.5-10.6-10.2-23.1-10.2-36.5c0-13.6 3.9-26.3 10.6-37.1C271.1 200.2 249.1 187 224 187zm118.9 0c-24.8 0-46.5 12.8-59 32.2l-.5 .7 .4 .6c6.5 10.6 10.2 23.1 10.2 36.5c0 13.6-3.9 26.3-10.6 37.1c12.4 19.8 34.4 32.9 59.5 32.9c38.7 0 70.1-31.3 70.1-70s-31.4-70-70.1-70z\"],\"sticker-mule\":[576,512,[],\"f3f7\",\"M561.7 199.6c-1.3.3.3 0 0 0zm-6.2-77.4c-7.7-22.3-5.1-7.2-13.4-36.9-1.6-6.5-3.6-14.5-6.2-20-4.4-8.7-4.6-7.5-4.6-9.5 0-5.3 30.7-45.3 19-46.9-5.7-.6-12.2 11.6-20.6 17-8.6 4.2-8 5-10.3 5-2.6 0-5.7-3-6.2-5-2-5.7 1.9-25.9-3.6-25.9-3.6 0-12.3 24.8-17 25.8-5.2 1.3-27.9-11.4-75.1 18-25.3 13.2-86.9 65.2-87 65.3-6.7 4.7-20 4.7-35.5 16-44.4 30.1-109.6 9.4-110.7 9-110.6-26.8-128-15.2-159 11.5-20.8 17.9-23.7 36.5-24.2 38.9-4.2 20.4 5.2 48.3 6.7 64.3 1.8 19.3-2.7 17.7 7.7 98.3.5 1 4.1 0 5.1 1.5 0 8.4-3.8 12.1-4.1 13-1.5 4.5-1.5 10.5 0 16 2.3 8.2 8.2 37.2 8.2 46.9 0 41.8.4 44 2.6 49.4 3.9 10 12.5 9.1 17 12 3.1 3.5-.5 8.5 1 12.5.5 2 3.6 4 6.2 5 9.2 3.6 27 .3 29.9-2.5 1.6-1.5.5-4.5 3.1-5 5.1 0 10.8-.5 14.4-2.5 5.1-2.5 4.1-6 1.5-10.5-.4-.8-7-13.3-9.8-16-2.1-2-5.1-3-7.2-4.5-5.8-4.9-10.3-19.4-10.3-19.5-4.6-19.4-10.3-46.3-4.1-66.8 4.6-17.2 39.5-87.7 39.6-87.8 4.1-6.5 17-11.5 27.3-7 6 1.9 19.3 22 65.4 30.9 47.9 8.7 97.4-2 112.2-2 2.8 2-1.9 13-.5 38.9 0 26.4-.4 13.7-4.1 29.9-2.2 9.7 3.4 23.2-1.5 46.9-1.4 9.8-9.9 32.7-8.2 43.4.5 1 1 2 1.5 3.5.5 4.5 1.5 8.5 4.6 10 7.3 3.6 12-3.5 9.8 11.5-.7 3.1-2.6 12 1.5 15 4.4 3.7 30.6 3.4 36.5.5 2.6-1.5 1.6-4.5 6.4-7.4 1.9-.9 11.3-.4 11.3-6.5.3-1.8-9.2-19.9-9.3-20-2.6-3.5-9.2-4.5-11.3-8-6.9-10.1-1.7-52.6.5-59.4 3-11 5.6-22.4 8.7-32.4 11-42.5 10.3-50.6 16.5-68.3.8-1.8 6.4-23.1 10.3-29.9 9.3-17 21.7-32.4 33.5-47.4 18-22.9 34-46.9 52-69.8 6.1-7 8.2-13.7 18-8 10.8 5.7 21.6 7 31.9 17 14.6 12.8 10.2 18.2 11.8 22.9 1.5 5 7.7 10.5 14.9 9.5 10.4-2 13-2.5 13.4-2.5 2.6-.5 5.7-5 7.2-8 3.1-5.5 7.2-9 7.2-16.5 0-7.7-.4-2.8-20.6-52.9z\"],\"creative-commons-zero\":[496,512,[],\"f4f3\",\"M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zm-.4 60.5c-81.9 0-102.5 77.3-102.5 142.8 0 65.5 20.6 142.8 102.5 142.8S350.5 321.5 350.5 256c0-65.5-20.6-142.8-102.5-142.8zm0 53.9c3.3 0 6.4.5 9.2 1.2 5.9 5.1 8.8 12.1 3.1 21.9l-54.5 100.2c-1.7-12.7-1.9-25.1-1.9-34.4 0-28.8 2-88.9 44.1-88.9zm40.8 46.2c2.9 15.4 3.3 31.4 3.3 42.7 0 28.9-2 88.9-44.1 88.9-13.5 0-32.6-7.7-20.1-26.4l60.9-105.2z\"],hips:[640,512,[],\"f452\",\"M251.6 157.6c0-1.9-.9-2.8-2.8-2.8h-40.9c-1.6 0-2.7 1.4-2.7 2.8v201.8c0 1.4 1.1 2.8 2.7 2.8h40.9c1.9 0 2.8-.9 2.8-2.8zM156.5 168c-16.1-11.8-36.3-17.9-60.3-18-18.1-.1-34.6 3.7-49.8 11.4V80.2c0-1.8-.9-2.7-2.8-2.7H2.7c-1.8 0-2.7.9-2.7 2.7v279.2c0 1.9.9 2.8 2.7 2.8h41c1.9 0 2.8-.9 2.8-2.8V223.3c0-.8-2.8-27 45.8-27 48.5 0 45.8 26.1 45.8 27v122.6c0 9 7.3 16.3 16.4 16.3h27.3c1.8 0 2.7-.9 2.7-2.8V223.3c0-23.4-9.3-41.8-28-55.3zm478.4 110.1c-6.8-15.7-18.4-27-34.9-34.1l-57.6-25.3c-8.6-3.6-9.2-11.2-2.6-16.1 7.4-5.5 44.3-13.9 84 6.8 1.7 1 4-.3 4-2.4v-44.7c0-1.3-.6-2.1-1.9-2.6-17.7-6.6-36.1-9.9-55.1-9.9-26.5 0-45.3 5.8-58.5 15.4-.5.4-28.4 20-22.7 53.7 3.4 19.6 15.8 34.2 37.2 43.6l53.6 23.5c11.6 5.1 15.2 13.3 12.2 21.2-3.7 9.1-13.2 13.6-36.5 13.6-24.3 0-44.7-8.9-58.4-19.1-2.1-1.4-4.4.2-4.4 2.3v34.4c0 10.4 4.9 17.3 14.6 20.7 15.6 5.5 31.6 8.2 48.2 8.2 12.7 0 25.8-1.2 36.3-4.3.7-.3 36-8.9 45.6-45.8 3.5-13.5 2.4-26.5-3.1-39.1zM376.2 149.8c-31.7 0-104.2 20.1-104.2 103.5v183.5c0 .8.6 2.7 2.7 2.7h40.9c1.9 0 2.8-.9 2.8-2.7V348c16.5 12.7 35.8 19.1 57.7 19.1 60.5 0 108.7-48.5 108.7-108.7.1-60.3-48.2-108.6-108.6-108.6zm0 170.9c-17.2 0-31.9-6.1-44-18.2-12.2-12.2-18.2-26.8-18.2-44 0-34.5 27.6-62.2 62.2-62.2 34.5 0 62.2 27.6 62.2 62.2.1 34.3-27.3 62.2-62.2 62.2zM228.3 72.5c-15.9 0-28.8 12.9-28.9 28.9 0 15.6 12.7 28.9 28.9 28.9s28.9-13.1 28.9-28.9c0-16.2-13-28.9-28.9-28.9z\"],behance:[576,512,[],\"f1b4\",\"M232 237.2c31.8-15.2 48.4-38.2 48.4-74 0-70.6-52.6-87.8-113.3-87.8H0v354.4h171.8c64.4 0 124.9-30.9 124.9-102.9 0-44.5-21.1-77.4-64.7-89.7zM77.9 135.9H151c28.1 0 53.4 7.9 53.4 40.5 0 30.1-19.7 42.2-47.5 42.2h-79v-82.7zm83.3 233.7H77.9V272h84.9c34.3 0 56 14.3 56 50.6 0 35.8-25.9 47-57.6 47zm358.5-240.7H376V94h143.7v34.9zM576 305.2c0-75.9-44.4-139.2-124.9-139.2-78.2 0-131.3 58.8-131.3 135.8 0 79.9 50.3 134.7 131.3 134.7 61.3 0 101-27.6 120.1-86.3H509c-6.7 21.9-34.3 33.5-55.7 33.5-41.3 0-63-24.2-63-65.3h185.1c.3-4.2.6-8.7.6-13.2zM390.4 274c2.3-33.7 24.7-54.8 58.5-54.8 35.4 0 53.2 20.8 56.2 54.8H390.4z\"],reddit:[512,512,[],\"f1a1\",\"M0 256C0 114.6 114.6 0 256 0S512 114.6 512 256s-114.6 256-256 256L37.1 512c-13.7 0-20.5-16.5-10.9-26.2L75 437C28.7 390.7 0 326.7 0 256zM349.6 153.6c23.6 0 42.7-19.1 42.7-42.7s-19.1-42.7-42.7-42.7c-20.6 0-37.8 14.6-41.8 34c-34.5 3.7-61.4 33-61.4 68.4l0 .2c-37.5 1.6-71.8 12.3-99 29.1c-10.1-7.8-22.8-12.5-36.5-12.5c-33 0-59.8 26.8-59.8 59.8c0 24 14.1 44.6 34.4 54.1c2 69.4 77.6 125.2 170.6 125.2s168.7-55.9 170.6-125.3c20.2-9.6 34.1-30.2 34.1-54c0-33-26.8-59.8-59.8-59.8c-13.7 0-26.3 4.6-36.4 12.4c-27.4-17-62.1-27.7-100-29.1l0-.2c0-25.4 18.9-46.5 43.4-49.9l0 0c4.4 18.8 21.3 32.8 41.5 32.8zM177.1 246.9c16.7 0 29.5 17.6 28.5 39.3s-13.5 29.6-30.3 29.6s-31.4-8.8-30.4-30.5s15.4-38.3 32.1-38.3zm190.1 38.3c1 21.7-13.7 30.5-30.4 30.5s-29.3-7.9-30.3-29.6c-1-21.7 11.8-39.3 28.5-39.3s31.2 16.6 32.1 38.3zm-48.1 56.7c-10.3 24.6-34.6 41.9-63 41.9s-52.7-17.3-63-41.9c-1.2-2.9 .8-6.2 3.9-6.5c18.4-1.9 38.3-2.9 59.1-2.9s40.7 1 59.1 2.9c3.1 .3 5.1 3.6 3.9 6.5z\"],discord:[640,512,[],\"f392\",\"M524.531,69.836a1.5,1.5,0,0,0-.764-.7A485.065,485.065,0,0,0,404.081,32.03a1.816,1.816,0,0,0-1.923.91,337.461,337.461,0,0,0-14.9,30.6,447.848,447.848,0,0,0-134.426,0,309.541,309.541,0,0,0-15.135-30.6,1.89,1.89,0,0,0-1.924-.91A483.689,483.689,0,0,0,116.085,69.137a1.712,1.712,0,0,0-.788.676C39.068,183.651,18.186,294.69,28.43,404.354a2.016,2.016,0,0,0,.765,1.375A487.666,487.666,0,0,0,176.02,479.918a1.9,1.9,0,0,0,2.063-.676A348.2,348.2,0,0,0,208.12,430.4a1.86,1.86,0,0,0-1.019-2.588,321.173,321.173,0,0,1-45.868-21.853,1.885,1.885,0,0,1-.185-3.126c3.082-2.309,6.166-4.711,9.109-7.137a1.819,1.819,0,0,1,1.9-.256c96.229,43.917,200.41,43.917,295.5,0a1.812,1.812,0,0,1,1.924.233c2.944,2.426,6.027,4.851,9.132,7.16a1.884,1.884,0,0,1-.162,3.126,301.407,301.407,0,0,1-45.89,21.83,1.875,1.875,0,0,0-1,2.611,391.055,391.055,0,0,0,30.014,48.815,1.864,1.864,0,0,0,2.063.7A486.048,486.048,0,0,0,610.7,405.729a1.882,1.882,0,0,0,.765-1.352C623.729,277.594,590.933,167.465,524.531,69.836ZM222.491,337.58c-28.972,0-52.844-26.587-52.844-59.239S193.056,219.1,222.491,219.1c29.665,0,53.306,26.82,52.843,59.239C275.334,310.993,251.924,337.58,222.491,337.58Zm195.38,0c-28.971,0-52.843-26.587-52.843-59.239S388.437,219.1,417.871,219.1c29.667,0,53.307,26.82,52.844,59.239C470.715,310.993,447.538,337.58,417.871,337.58Z\"],chrome:[512,512,[],\"f268\",\"M0 256C0 209.4 12.47 165.6 34.27 127.1L144.1 318.3C166 357.5 207.9 384 256 384C270.3 384 283.1 381.7 296.8 377.4L220.5 509.6C95.9 492.3 0 385.3 0 256zM365.1 321.6C377.4 302.4 384 279.1 384 256C384 217.8 367.2 183.5 340.7 160H493.4C505.4 189.6 512 222.1 512 256C512 397.4 397.4 511.1 256 512L365.1 321.6zM477.8 128H256C193.1 128 142.3 172.1 130.5 230.7L54.19 98.47C101 38.53 174 0 256 0C350.8 0 433.5 51.48 477.8 128V128zM168 256C168 207.4 207.4 168 256 168C304.6 168 344 207.4 344 256C344 304.6 304.6 344 256 344C207.4 344 168 304.6 168 256z\"],\"app-store-ios\":[448,512,[],\"f370\",\"M400 32H48C21.5 32 0 53.5 0 80v352c0 26.5 21.5 48 48 48h352c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48zM127 384.5c-5.5 9.6-17.8 12.8-27.3 7.3-9.6-5.5-12.8-17.8-7.3-27.3l14.3-24.7c16.1-4.9 29.3-1.1 39.6 11.4L127 384.5zm138.9-53.9H84c-11 0-20-9-20-20s9-20 20-20h51l65.4-113.2-20.5-35.4c-5.5-9.6-2.2-21.8 7.3-27.3 9.6-5.5 21.8-2.2 27.3 7.3l8.9 15.4 8.9-15.4c5.5-9.6 17.8-12.8 27.3-7.3 9.6 5.5 12.8 17.8 7.3 27.3l-85.8 148.6h62.1c20.2 0 31.5 23.7 22.7 40zm98.1 0h-29l19.6 33.9c5.5 9.6 2.2 21.8-7.3 27.3-9.6 5.5-21.8 2.2-27.3-7.3-32.9-56.9-57.5-99.7-74-128.1-16.7-29-4.8-58 7.1-67.8 13.1 22.7 32.7 56.7 58.9 102h52c11 0 20 9 20 20 0 11.1-9 20-20 20z\"],\"cc-discover\":[576,512,[],\"f1f2\",\"M520.4 196.1c0-7.9-5.5-12.1-15.6-12.1h-4.9v24.9h4.7c10.3 0 15.8-4.4 15.8-12.8zM528 32H48C21.5 32 0 53.5 0 80v352c0 26.5 21.5 48 48 48h480c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48zm-44.1 138.9c22.6 0 52.9-4.1 52.9 24.4 0 12.6-6.6 20.7-18.7 23.2l25.8 34.4h-19.6l-22.2-32.8h-2.2v32.8h-16zm-55.9.1h45.3v14H444v18.2h28.3V217H444v22.2h29.3V253H428zm-68.7 0l21.9 55.2 22.2-55.2h17.5l-35.5 84.2h-8.6l-35-84.2zm-55.9-3c24.7 0 44.6 20 44.6 44.6 0 24.7-20 44.6-44.6 44.6-24.7 0-44.6-20-44.6-44.6 0-24.7 20-44.6 44.6-44.6zm-49.3 6.1v19c-20.1-20.1-46.8-4.7-46.8 19 0 25 27.5 38.5 46.8 19.2v19c-29.7 14.3-63.3-5.7-63.3-38.2 0-31.2 33.1-53 63.3-38zm-97.2 66.3c11.4 0 22.4-15.3-3.3-24.4-15-5.5-20.2-11.4-20.2-22.7 0-23.2 30.6-31.4 49.7-14.3l-8.4 10.8c-10.4-11.6-24.9-6.2-24.9 2.5 0 4.4 2.7 6.9 12.3 10.3 18.2 6.6 23.6 12.5 23.6 25.6 0 29.5-38.8 37.4-56.6 11.3l10.3-9.9c3.7 7.1 9.9 10.8 17.5 10.8zM55.4 253H32v-82h23.4c26.1 0 44.1 17 44.1 41.1 0 18.5-13.2 40.9-44.1 40.9zm67.5 0h-16v-82h16zM544 433c0 8.2-6.8 15-15 15H128c189.6-35.6 382.7-139.2 416-160zM74.1 191.6c-5.2-4.9-11.6-6.6-21.9-6.6H48v54.2h4.2c10.3 0 17-2 21.9-6.4 5.7-5.2 8.9-12.8 8.9-20.7s-3.2-15.5-8.9-20.5z\"],wpbeginner:[512,512,[],\"f297\",\"M462.799 322.374C519.01 386.682 466.961 480 370.944 480c-39.602 0-78.824-17.687-100.142-50.04-6.887.356-22.702.356-29.59 0C219.848 462.381 180.588 480 141.069 480c-95.49 0-148.348-92.996-91.855-157.626C-29.925 190.523 80.479 32 256.006 32c175.632 0 285.87 158.626 206.793 290.374zm-339.647-82.972h41.529v-58.075h-41.529v58.075zm217.18 86.072v-23.839c-60.506 20.915-132.355 9.198-187.589-33.971l.246 24.897c51.101 46.367 131.746 57.875 187.343 32.913zm-150.753-86.072h166.058v-58.075H189.579v58.075z\"],confluence:[512,512,[],\"f78d\",\"M2.3 412.2c-4.5 7.6-2.1 17.5 5.5 22.2l105.9 65.2c7.7 4.7 17.7 2.4 22.4-5.3 0-.1.1-.2.1-.2 67.1-112.2 80.5-95.9 280.9-.7 8.1 3.9 17.8.4 21.7-7.7.1-.1.1-.3.2-.4l50.4-114.1c3.6-8.1-.1-17.6-8.1-21.3-22.2-10.4-66.2-31.2-105.9-50.3C127.5 179 44.6 345.3 2.3 412.2zm507.4-312.1c4.5-7.6 2.1-17.5-5.5-22.2L398.4 12.8c-7.5-5-17.6-3.1-22.6 4.4-.2.3-.4.6-.6 1-67.3 112.6-81.1 95.6-280.6.9-8.1-3.9-17.8-.4-21.7 7.7-.1.1-.1.3-.2.4L22.2 141.3c-3.6 8.1.1 17.6 8.1 21.3 22.2 10.4 66.3 31.2 106 50.4 248 120 330.8-45.4 373.4-112.9z\"],shoelace:[512,512,[],\"e60c\",\"M404.9 331c2.2-1.1 4.4-2.3 6.5-3.7l8.3-4.8c1.5-1.1 4.4-3.4 8.7-6.7l.5-.5c3.4-3.4 7.2-5 11.3-4.9c1.8 0 3.9 .5 6.4 1.5l31-27.5c.9-.7 1.7-1.3 2.6-1.8h.2c3.3-1.9 6-1.8 8.2 .4c3.9 2.1 4.2 5.6 .9 10.6L456.9 322c.2 .5 .4 1 .4 1.5c.5 2.2 .3 4.4-.5 6.6c-.7 1.5-1.8 2.9-3.1 4.2c-1.4 1.4-2.7 2.8-4.2 4.2l-18.8 13.7c-1.7 1.2-3.4 2.3-5.1 3.3c-2.1 1.3-4.3 2.5-6.6 3.6c-1 .4-1.9 .9-2.9 1.3c-5.9 2.5-11.9 4.2-18.2 5c-2.9 24.5-11.3 47.1-25.1 67.8c-17.5 25.7-41.4 45.4-71.8 58.8c-30.2 13.5-63 20.2-98.2 20.2c-48.6-.5-88-11.4-118.2-32.8C49.5 454.4 32 421.5 32 380.3v-5.6c1.2-28.1 9.5-54.6 24.8-79.8c15.1-24.9 37.1-41.7 66.1-50.5c14.9-4.4 29.9-6.6 45-6.6c15.5 0 31.6 2.9 48.1 8.6s35.2 15.5 55.9 29.5L326 312.2c15.1 9.8 28.8 16.5 41.2 20c-2.6-25.1-11.7-46.6-27.3-64.5c-15.7-18.1-35.6-31.3-59.9-39.7l-23.3-8c-21.4-7.5-37.3-14.9-47.7-22.2c-28.2-19.1-43.8-45.2-47-78.5l-.5-9.8c0-32.1 13-58.9 39-80.5C223.5 9.7 251.1 0 283 0c24 0 45.6 6.9 64.7 20.8c19.2 14 30.1 33.8 32.6 59.4l.5 10c0 18.6-4.8 34.5-14.4 47.7c-9.8 13.2-18.5 19.9-26 19.9c-1.6-.1-3.1-.3-4.5-.6l-34 32c-5.5 3-9.2 2.5-11.1-1.6c-1.9-2.2-1.8-4.9 .5-8.2l.2-.2c.5-.7 1.2-1.5 2-2.4l31.6-30c-.4-1.5-.6-3.1-.6-4.8c0-4.1 1.6-7.6 4.9-10.4c13.8-12.4 20.8-26.7 20.8-42.8c0-16-6.1-29.5-18.2-40.4s-28.7-16.5-49.7-16.8c-26.2 0-47.8 7.9-64.7 23.7S192.3 89.9 192.3 112c0 17.8 6.9 33.9 20.6 48.3c13.6 14.2 34.6 25.4 63 33.5c39.8 11.5 70.2 31 91.3 58.3c18.7 24.2 29.1 51.3 31.3 81.4c2.2-.7 4.3-1.5 6.5-2.6zM294.1 178.7c0 1.1 .6 1.6 1.8 1.6c.1 0 9.7-8.9 28.8-26.6c0-2.4-5.1 .9-15.3 10c-10.2 9.2-15.3 14.2-15.3 14.9zm8 6.4c0-1-.5-1.5-1.5-1.5c-1.1 0-2.1 .5-2.9 1.6c-1.9-.1-3.3 .1-4.2 .7c-.4 .2-.5 .5-.5 .7c0 .7 .5 1.3 1.5 1.6h3.3c2.9-1.1 4.4-2.2 4.4-3.3zm22.6-19.9c0-2.8-1.6-2.8-4.9 0c-1.6 1.5-3.6 3.5-6 6.2c-.8 .6-2.6 2.2-5.3 4.9c-2.8 2.9-4.2 4.7-4.2 5.3l.2 1.3c.7 .2 1.2 .4 1.5 .4c.1 0 3.3-2.9 9.5-8.7s9.3-8.9 9.3-9.3zm159.7 120l-30.6 27c1.8 1 3.2 2.4 4 4.2l30.2-27c.2-1.2 .1-2.2-.5-2.9c-.6-.5-1.6-.9-3.1-1.3zm-1.6-.9l-.7-.7-27 21.9 1.6 2 26-23.1zM366.6 363.9c-8-2.1-15.4-4.6-22.2-7.5c-15.3-6.2-34.3-17-57-32.4L250 298.7c-15.8-10.1-30.2-17.6-43.2-22.6c-13.1-4.9-26-7.3-38.6-7.3h-5.5c-32.2 1.7-57.2 13.8-75 36.2c-16.6 20.8-25 45.3-25 73.6c0 31.8 12.8 56.7 38.2 74.7c25.4 18.1 60.2 27.1 104.4 27.1c34.7 0 64-6.2 87.8-18.6c23.7-12.4 42.1-28.8 55.2-49.2c9.8-15.5 15.9-31.8 18.2-48.8z\"],mdb:[576,512,[],\"f8ca\",\"M17.37 160.41L7 352h43.91l5.59-79.83L84.43 352h44.71l25.54-77.43 4.79 77.43H205l-12.79-191.59H146.7L106 277.74 63.67 160.41zm281 0h-47.9V352h47.9s95 .8 94.2-95.79c-.78-94.21-94.18-95.78-94.18-95.78zm-1.2 146.46V204.78s46 4.27 46.8 50.57-46.78 51.54-46.78 51.54zm238.29-74.24a56.16 56.16 0 0 0 8-38.31c-5.34-35.76-55.08-34.32-55.08-34.32h-51.9v191.58H482s87 4.79 87-63.85c0-43.14-33.52-55.08-33.52-55.08zm-51.9-31.94s13.57-1.59 16 9.59c1.43 6.66-4 12-4 12h-12v-21.57zm-.1 109.46l.1-24.92V267h.08s41.58-4.73 41.19 22.43c-.33 25.65-41.35 20.74-41.35 20.74z\"],dochub:[416,512,[],\"f394\",\"M397.9 160H256V19.6L397.9 160zM304 192v130c0 66.8-36.5 100.1-113.3 100.1H96V84.8h94.7c12 0 23.1.8 33.1 2.5v-84C212.9 1.1 201.4 0 189.2 0H0v512h189.2C329.7 512 400 447.4 400 318.1V192h-96z\"],\"accessible-icon\":[448,512,[62107],\"f368\",\"M423.9 255.8L411 413.1c-3.3 40.7-63.9 35.1-60.6-4.9l10-122.5-41.1 2.3c10.1 20.7 15.8 43.9 15.8 68.5 0 41.2-16.1 78.7-42.3 106.5l-39.3-39.3c57.9-63.7 13.1-167.2-74-167.2-25.9 0-49.5 9.9-67.2 26L73 243.2c22-20.7 50.1-35.1 81.4-40.2l75.3-85.7-42.6-24.8-51.6 46c-30 26.8-70.6-18.5-40.5-45.4l68-60.7c9.8-8.8 24.1-10.2 35.5-3.6 0 0 139.3 80.9 139.5 81.1 16.2 10.1 20.7 36 6.1 52.6L285.7 229l106.1-5.9c18.5-1.1 33.6 14.4 32.1 32.7zm-64.9-154c28.1 0 50.9-22.8 50.9-50.9C409.9 22.8 387.1 0 359 0c-28.1 0-50.9 22.8-50.9 50.9 0 28.1 22.8 50.9 50.9 50.9zM179.6 456.5c-80.6 0-127.4-90.6-82.7-156.1l-39.7-39.7C36.4 287 24 320.3 24 356.4c0 130.7 150.7 201.4 251.4 122.5l-39.7-39.7c-16 10.9-35.3 17.3-56.1 17.3z\"],ebay:[640,512,[],\"f4f4\",\"M606 189.5l-54.8 109.9-54.9-109.9h-37.5l10.9 20.6c-11.5-19-35.9-26-63.3-26-31.8 0-67.9 8.7-71.5 43.1h33.7c1.4-13.8 15.7-21.8 35-21.8 26 0 41 9.6 41 33v3.4c-12.7 0-28 .1-41.7.4-42.4.9-69.6 10-76.7 34.4 1-5.2 1.5-10.6 1.5-16.2 0-52.1-39.7-76.2-75.4-76.2-21.3 0-43 5.5-58.7 24.2v-80.6h-32.1v169.5c0 10.3-.6 22.9-1.1 33.1h31.5c.7-6.3 1.1-12.9 1.1-19.5 13.6 16.6 35.4 24.9 58.7 24.9 36.9 0 64.9-21.9 73.3-54.2-.5 2.8-.7 5.8-.7 9 0 24.1 21.1 45 60.6 45 26.6 0 45.8-5.7 61.9-25.5 0 6.6.3 13.3 1.1 20.2h29.8c-.7-8.2-1-17.5-1-26.8v-65.6c0-9.3-1.7-17.2-4.8-23.8l61.5 116.1-28.5 54.1h35.9L640 189.5zM243.7 313.8c-29.6 0-50.2-21.5-50.2-53.8 0-32.4 20.6-53.8 50.2-53.8 29.8 0 50.2 21.4 50.2 53.8 0 32.3-20.4 53.8-50.2 53.8zm200.9-47.3c0 30-17.9 48.4-51.6 48.4-25.1 0-35-13.4-35-25.8 0-19.1 18.1-24.4 47.2-25.3 13.1-.5 27.6-.6 39.4-.6zm-411.9 1.6h128.8v-8.5c0-51.7-33.1-75.4-78.4-75.4-56.8 0-83 30.8-83 77.6 0 42.5 25.3 74 82.5 74 31.4 0 68-11.7 74.4-46.1h-33.1c-12 35.8-87.7 36.7-91.2-21.6zm95-21.4H33.3c6.9-56.6 92.1-54.7 94.4 0z\"],amazon:[448,512,[],\"f270\",\"M257.2 162.7c-48.7 1.8-169.5 15.5-169.5 117.5 0 109.5 138.3 114 183.5 43.2 6.5 10.2 35.4 37.5 45.3 46.8l56.8-56S341 288.9 341 261.4V114.3C341 89 316.5 32 228.7 32 140.7 32 94 87 94 136.3l73.5 6.8c16.3-49.5 54.2-49.5 54.2-49.5 40.7-.1 35.5 29.8 35.5 69.1zm0 86.8c0 80-84.2 68-84.2 17.2 0-47.2 50.5-56.7 84.2-57.8v40.6zm136 163.5c-7.7 10-70 67-174.5 67S34.2 408.5 9.7 379c-6.8-7.7 1-11.3 5.5-8.3C88.5 415.2 203 488.5 387.7 401c7.5-3.7 13.3 2 5.5 12zm39.8 2.2c-6.5 15.8-16 26.8-21.2 31-5.5 4.5-9.5 2.7-6.5-3.8s19.3-46.5 12.7-55c-6.5-8.3-37-4.3-48-3.2-10.8 1-13 2-14-.3-2.3-5.7 21.7-15.5 37.5-17.5 15.7-1.8 41-.8 46 5.7 3.7 5.1 0 27.1-6.5 43.1z\"],unsplash:[448,512,[],\"e07c\",\"M448,230.17V480H0V230.17H141.13V355.09H306.87V230.17ZM306.87,32H141.13V156.91H306.87Z\"],yarn:[496,512,[],\"f7e3\",\"M393.9 345.2c-39 9.3-48.4 32.1-104 47.4 0 0-2.7 4-10.4 5.8-13.4 3.3-63.9 6-68.5 6.1-12.4.1-19.9-3.2-22-8.2-6.4-15.3 9.2-22 9.2-22-8.1-5-9-9.9-9.8-8.1-2.4 5.8-3.6 20.1-10.1 26.5-8.8 8.9-25.5 5.9-35.3.8-10.8-5.7.8-19.2.8-19.2s-5.8 3.4-10.5-3.6c-6-9.3-17.1-37.3 11.5-62-1.3-10.1-4.6-53.7 40.6-85.6 0 0-20.6-22.8-12.9-43.3 5-13.4 7-13.3 8.6-13.9 5.7-2.2 11.3-4.6 15.4-9.1 20.6-22.2 46.8-18 46.8-18s12.4-37.8 23.9-30.4c3.5 2.3 16.3 30.6 16.3 30.6s13.6-7.9 15.1-5c8.2 16 9.2 46.5 5.6 65.1-6.1 30.6-21.4 47.1-27.6 57.5-1.4 2.4 16.5 10 27.8 41.3 10.4 28.6 1.1 52.7 2.8 55.3.8 1.4 13.7.8 36.4-13.2 12.8-7.9 28.1-16.9 45.4-17 16.7-.5 17.6 19.2 4.9 22.2zM496 256c0 136.9-111.1 248-248 248S0 392.9 0 256 111.1 8 248 8s248 111.1 248 248zm-79.3 75.2c-1.7-13.6-13.2-23-28-22.8-22 .3-40.5 11.7-52.8 19.2-4.8 3-8.9 5.2-12.4 6.8 3.1-44.5-22.5-73.1-28.7-79.4 7.8-11.3 18.4-27.8 23.4-53.2 4.3-21.7 3-55.5-6.9-74.5-1.6-3.1-7.4-11.2-21-7.4-9.7-20-13-22.1-15.6-23.8-1.1-.7-23.6-16.4-41.4 28-12.2.9-31.3 5.3-47.5 22.8-2 2.2-5.9 3.8-10.1 5.4h.1c-8.4 3-12.3 9.9-16.9 22.3-6.5 17.4.2 34.6 6.8 45.7-17.8 15.9-37 39.8-35.7 82.5-34 36-11.8 73-5.6 79.6-1.6 11.1 3.7 19.4 12 23.8 12.6 6.7 30.3 9.6 43.9 2.8 4.9 5.2 13.8 10.1 30 10.1 6.8 0 58-2.9 72.6-6.5 6.8-1.6 11.5-4.5 14.6-7.1 9.8-3.1 36.8-12.3 62.2-28.7 18-11.7 24.2-14.2 37.6-17.4 12.9-3.2 21-15.1 19.4-28.2z\"],\"square-steam\":[448,512,[\"steam-square\"],\"f1b7\",\"M165.6 309.1c18.6 7.7 27.3 28.9 19.6 47.4s-29 27.2-47.6 19.4l-28.5-11.8c5 10.6 13.8 19.4 25.4 24.2c25.2 10.5 54.1-1.4 64.6-26.5c5.1-12.1 5.1-25.5 .1-37.7c-5.1-12.1-14.5-21.6-26.7-26.7c-12.1-5-25-4.8-36.4-.5l29.5 12.2zM448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V240.7l116.6 48.1c12-8.2 26.2-12.1 40.7-11.3l55.4-80.2v-1.1c0-48.2 39.3-87.5 87.6-87.5s87.6 39.3 87.6 87.5c0 49.2-40.9 88.7-89.6 87.5l-79 56.3c1.6 38.5-29.1 68.8-65.7 68.8c-31.8 0-58.5-22.7-64.5-52.7L0 319.2V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM241.9 196.2a58.4 58.4 0 1 0 116.8 0 58.4 58.4 0 1 0 -116.8 0zm14.6-.1a43.9 43.9 0 1 1 87.8 0 43.9 43.9 0 1 1 -87.8 0z\"],\"500px\":[448,512,[],\"f26e\",\"M103.3 344.3c-6.5-14.2-6.9-18.3 7.4-23.1 25.6-8 8 9.2 43.2 49.2h.3v-93.9c1.2-50.2 44-92.2 97.7-92.2 53.9 0 97.7 43.5 97.7 96.8 0 63.4-60.8 113.2-128.5 93.3-10.5-4.2-2.1-31.7 8.5-28.6 53 0 89.4-10.1 89.4-64.4 0-61-77.1-89.6-116.9-44.6-23.5 26.4-17.6 42.1-17.6 157.6 50.7 31 118.3 22 160.4-20.1 24.8-24.8 38.5-58 38.5-93 0-35.2-13.8-68.2-38.8-93.3-24.8-24.8-57.8-38.5-93.3-38.5s-68.8 13.8-93.5 38.5c-.3.3-16 16.5-21.2 23.9l-.5.6c-3.3 4.7-6.3 9.1-20.1 6.1-6.9-1.7-14.3-5.8-14.3-11.8V20c0-5 3.9-10.5 10.5-10.5h241.3c8.3 0 8.3 11.6 8.3 15.1 0 3.9 0 15.1-8.3 15.1H130.3v132.9h.3c104.2-109.8 282.8-36 282.8 108.9 0 178.1-244.8 220.3-310.1 62.8zm63.3-260.8c-.5 4.2 4.6 24.5 14.6 20.6C306 56.6 384 144.5 390.6 144.5c4.8 0 22.8-15.3 14.3-22.8-93.2-89-234.5-57-238.3-38.2zM393 414.7C283 524.6 94 475.5 61 310.5c0-12.2-30.4-7.4-28.9 3.3 24 173.4 246 256.9 381.6 121.3 6.9-7.8-12.6-28.4-20.7-20.4zM213.6 306.6c0 4 4.3 7.3 5.5 8.5 3 3 6.1 4.4 8.5 4.4 3.8 0 2.6.2 22.3-19.5 19.6 19.3 19.1 19.5 22.3 19.5 5.4 0 18.5-10.4 10.7-18.2L265.6 284l18.2-18.2c6.3-6.8-10.1-21.8-16.2-15.7L249.7 268c-18.6-18.8-18.4-19.5-21.5-19.5-5 0-18 11.7-12.4 17.3L234 284c-18.1 17.9-20.4 19.2-20.4 22.6z\"],\"square-vimeo\":[448,512,[\"vimeo-square\"],\"f194\",\"M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM318.5 309.5C275 366 238.2 394.3 208.1 394.3c-18.7 0-34.4-17.2-47.3-51.6c-3.8-13.9-7.3-26.9-10.5-39c-18.2-68.3-28.6-107.4-46.2-107.4c-2.4 0-10.8 5-25.1 15.1L64 192c7-6.1 13.9-12.4 20.6-18.5c29.1-26.3 55.6-50.3 73.5-51.9c24.9-2.4 40.2 14.6 46 51.1c20.5 129.6 29.6 149.2 66.8 90.5c13.4-21.2 20.6-37.2 21.5-48.3c3.4-32.8-25.6-30.6-45.2-22.2c15.7-51.5 45.8-76.5 90.1-75.1c32.9 1 48.4 22.4 46.5 64c-1.4 31.1-23.2 73.8-65.3 127.9z\"],asymmetrik:[576,512,[],\"f372\",\"M517.5 309.2c38.8-40 58.1-80 58.5-116.1.8-65.5-59.4-118.2-169.4-135C277.9 38.4 118.1 73.6 0 140.5 52 114 110.6 92.3 170.7 82.3c74.5-20.5 153-25.4 221.3-14.8C544.5 91.3 588.8 195 490.8 299.2c-10.2 10.8-22 21.1-35 30.6L304.9 103.4 114.7 388.9c-65.6-29.4-76.5-90.2-19.1-151.2 20.8-22.2 48.3-41.9 79.5-58.1 20-12.2 39.7-22.6 62-30.7-65.1 20.3-122.7 52.9-161.6 92.9-27.7 28.6-41.4 57.1-41.7 82.9-.5 35.1 23.4 65.1 68.4 83l-34.5 51.7h101.6l22-34.4c22.2 1 45.3 0 68.6-2.7l-22.8 37.1h135.5L340 406.3c18.6-5.3 36.9-11.5 54.5-18.7l45.9 71.8H542L468.6 349c18.5-12.1 35-25.5 48.9-39.8zm-187.6 80.5l-25-40.6-32.7 53.3c-23.4 3.5-46.7 5.1-69.2 4.4l101.9-159.3 78.7 123c-17.2 7.4-35.3 13.9-53.7 19.2z\"],\"font-awesome\":[512,512,[62501,62694,\"font-awesome-flag\",\"font-awesome-logo-full\"],\"f2b4\",\"M91.7 96C106.3 86.8 116 70.5 116 52C116 23.3 92.7 0 64 0S12 23.3 12 52c0 16.7 7.8 31.5 20 41l0 3 0 352 0 64 64 0 0-64 373.6 0c14.6 0 26.4-11.8 26.4-26.4c0-3.7-.8-7.3-2.3-10.7L432 272l61.7-138.9c1.5-3.4 2.3-7 2.3-10.7c0-14.6-11.8-26.4-26.4-26.4L91.7 96z\"],gratipay:[496,512,[],\"f184\",\"M248 8C111.1 8 0 119.1 0 256s111.1 248 248 248 248-111.1 248-248S384.9 8 248 8zm114.6 226.4l-113 152.7-112.7-152.7c-8.7-11.9-19.1-50.4 13.6-72 28.1-18.1 54.6-4.2 68.5 11.9 15.9 17.9 46.6 16.9 61.7 0 13.9-16.1 40.4-30 68.1-11.9 32.9 21.6 22.6 60 13.8 72z\"],apple:[384,512,[],\"f179\",\"M318.7 268.7c-.2-36.7 16.4-64.4 50-84.8-18.8-26.9-47.2-41.7-84.7-44.6-35.5-2.8-74.3 20.7-88.5 20.7-15 0-49.4-19.7-76.4-19.7C63.3 141.2 4 184.8 4 273.5q0 39.3 14.4 81.2c12.8 36.7 59 126.7 107.2 125.2 25.2-.6 43-17.9 75.8-17.9 31.8 0 48.3 17.9 76.4 17.9 48.6-.7 90.4-82.5 102.6-119.3-65.2-30.7-61.7-90-61.7-91.9zm-56.6-164.2c27.3-32.4 24.8-61.9 24-72.5-24.1 1.4-52 16.4-67.9 34.9-17.5 19.8-27.8 44.3-25.6 71.9 26.1 2 49.9-11.4 69.5-34.3z\"],hive:[512,512,[],\"e07f\",\"M260.353,254.878,131.538,33.1a2.208,2.208,0,0,0-3.829.009L.3,254.887A2.234,2.234,0,0,0,.3,257.122L129.116,478.9a2.208,2.208,0,0,0,3.83-.009L260.358,257.113A2.239,2.239,0,0,0,260.353,254.878Zm39.078-25.713a2.19,2.19,0,0,0,1.9,1.111h66.509a2.226,2.226,0,0,0,1.9-3.341L259.115,33.111a2.187,2.187,0,0,0-1.9-1.111H190.707a2.226,2.226,0,0,0-1.9,3.341ZM511.7,254.886,384.9,33.112A2.2,2.2,0,0,0,382.99,32h-66.6a2.226,2.226,0,0,0-1.906,3.34L440.652,256,314.481,476.66a2.226,2.226,0,0,0,1.906,3.34h66.6a2.2,2.2,0,0,0,1.906-1.112L511.7,257.114A2.243,2.243,0,0,0,511.7,254.886ZM366.016,284.917H299.508a2.187,2.187,0,0,0-1.9,1.111l-108.8,190.631a2.226,2.226,0,0,0,1.9,3.341h66.509a2.187,2.187,0,0,0,1.9-1.111l108.8-190.631A2.226,2.226,0,0,0,366.016,284.917Z\"],gitkraken:[592,512,[],\"f3a6\",\"M565.7 118.1c-2.3-6.1-9.3-9.2-15.3-6.6-5.7 2.4-8.5 8.9-6.3 14.6 10.9 29 16.9 60.5 16.9 93.3 0 134.6-100.3 245.7-230.2 262.7V358.4c7.9-1.5 15.5-3.6 23-6.2v104c106.7-25.9 185.9-122.1 185.9-236.8 0-91.8-50.8-171.8-125.8-213.3-5.7-3.2-13-.9-15.9 5-2.7 5.5-.6 12.2 4.7 15.1 67.9 37.6 113.9 110 113.9 193.2 0 93.3-57.9 173.1-139.8 205.4v-92.2c14.2-4.5 24.9-17.7 24.9-33.5 0-13.1-6.8-24.4-17.3-30.5 8.3-79.5 44.5-58.6 44.5-83.9V170c0-38-87.9-161.8-129-164.7-2.5-.2-5-.2-7.6 0C251.1 8.3 163.2 132 163.2 170v14.8c0 25.3 36.3 4.3 44.5 83.9-10.6 6.1-17.3 17.4-17.3 30.5 0 15.8 10.6 29 24.8 33.5v92.2c-81.9-32.2-139.8-112-139.8-205.4 0-83.1 46-155.5 113.9-193.2 5.4-3 7.4-9.6 4.7-15.1-2.9-5.9-10.1-8.2-15.9-5-75 41.5-125.8 121.5-125.8 213.3 0 114.7 79.2 210.8 185.9 236.8v-104c7.6 2.5 15.1 4.6 23 6.2v123.7C131.4 465.2 31 354.1 31 219.5c0-32.8 6-64.3 16.9-93.3 2.2-5.8-.6-12.2-6.3-14.6-6-2.6-13 .4-15.3 6.6C14.5 149.7 8 183.8 8 219.5c0 155.1 122.6 281.6 276.3 287.8V361.4c6.8.4 15 .5 23.4 0v145.8C461.4 501.1 584 374.6 584 219.5c0-35.7-6.5-69.8-18.3-101.4zM365.9 275.5c13 0 23.7 10.5 23.7 23.7 0 13.1-10.6 23.7-23.7 23.7-13 0-23.7-10.5-23.7-23.7 0-13.1 10.6-23.7 23.7-23.7zm-139.8 47.3c-13.2 0-23.7-10.7-23.7-23.7s10.5-23.7 23.7-23.7c13.1 0 23.7 10.6 23.7 23.7 0 13-10.5 23.7-23.7 23.7z\"],keybase:[448,512,[],\"f4f5\",\"M286.17 419a18 18 0 1 0 18 18 18 18 0 0 0-18-18zm111.92-147.6c-9.5-14.62-39.37-52.45-87.26-73.71q-9.1-4.06-18.38-7.27a78.43 78.43 0 0 0-47.88-104.13c-12.41-4.1-23.33-6-32.41-5.77-.6-2-1.89-11 9.4-35L198.66 32l-5.48 7.56c-8.69 12.06-16.92 23.55-24.34 34.89a51 51 0 0 0-8.29-1.25c-41.53-2.45-39-2.33-41.06-2.33-50.61 0-50.75 52.12-50.75 45.88l-2.36 36.68c-1.61 27 19.75 50.21 47.63 51.85l8.93.54a214 214 0 0 0-46.29 35.54C14 304.66 14 374 14 429.77v33.64l23.32-29.8a148.6 148.6 0 0 0 14.56 37.56c5.78 10.13 14.87 9.45 19.64 7.33 4.21-1.87 10-6.92 3.75-20.11a178.29 178.29 0 0 1-15.76-53.13l46.82-59.83-24.66 74.11c58.23-42.4 157.38-61.76 236.25-38.59 34.2 10.05 67.45.69 84.74-23.84.72-1 1.2-2.16 1.85-3.22a156.09 156.09 0 0 1 2.8 28.43c0 23.3-3.69 52.93-14.88 81.64-2.52 6.46 1.76 14.5 8.6 15.74 7.42 1.57 15.33-3.1 18.37-11.15C429 443 434 414 434 382.32c0-38.58-13-77.46-35.91-110.92zM142.37 128.58l-15.7-.93-1.39 21.79 13.13.78a93 93 0 0 0 .32 19.57l-22.38-1.34a12.28 12.28 0 0 1-11.76-12.79L107 119c1-12.17 13.87-11.27 13.26-11.32l29.11 1.73a144.35 144.35 0 0 0-7 19.17zm148.42 172.18a10.51 10.51 0 0 1-14.35-1.39l-9.68-11.49-34.42 27a8.09 8.09 0 0 1-11.13-1.08l-15.78-18.64a7.38 7.38 0 0 1 1.34-10.34l34.57-27.18-14.14-16.74-17.09 13.45a7.75 7.75 0 0 1-10.59-1s-3.72-4.42-3.8-4.53a7.38 7.38 0 0 1 1.37-10.34L214 225.19s-18.51-22-18.6-22.14a9.56 9.56 0 0 1 1.74-13.42 10.38 10.38 0 0 1 14.3 1.37l81.09 96.32a9.58 9.58 0 0 1-1.74 13.44zM187.44 419a18 18 0 1 0 18 18 18 18 0 0 0-18-18z\"],\"apple-pay\":[640,512,[],\"f415\",\"M116.9 158.5c-7.5 8.9-19.5 15.9-31.5 14.9-1.5-12 4.4-24.8 11.3-32.6 7.5-9.1 20.6-15.6 31.3-16.1 1.2 12.4-3.7 24.7-11.1 33.8m10.9 17.2c-17.4-1-32.3 9.9-40.5 9.9-8.4 0-21-9.4-34.8-9.1-17.9.3-34.5 10.4-43.6 26.5-18.8 32.3-4.9 80 13.3 106.3 8.9 13 19.5 27.3 33.5 26.8 13.3-.5 18.5-8.6 34.5-8.6 16.1 0 20.8 8.6 34.8 8.4 14.5-.3 23.6-13 32.5-26 10.1-14.8 14.3-29.1 14.5-29.9-.3-.3-28-10.9-28.3-42.9-.3-26.8 21.9-39.5 22.9-40.3-12.5-18.6-32-20.6-38.8-21.1m100.4-36.2v194.9h30.3v-66.6h41.9c38.3 0 65.1-26.3 65.1-64.3s-26.4-64-64.1-64h-73.2zm30.3 25.5h34.9c26.3 0 41.3 14 41.3 38.6s-15 38.8-41.4 38.8h-34.8V165zm162.2 170.9c19 0 36.6-9.6 44.6-24.9h.6v23.4h28v-97c0-28.1-22.5-46.3-57.1-46.3-32.1 0-55.9 18.4-56.8 43.6h27.3c2.3-12 13.4-19.9 28.6-19.9 18.5 0 28.9 8.6 28.9 24.5v10.8l-37.8 2.3c-35.1 2.1-54.1 16.5-54.1 41.5.1 25.2 19.7 42 47.8 42zm8.2-23.1c-16.1 0-26.4-7.8-26.4-19.6 0-12.3 9.9-19.4 28.8-20.5l33.6-2.1v11c0 18.2-15.5 31.2-36 31.2zm102.5 74.6c29.5 0 43.4-11.3 55.5-45.4L640 193h-30.8l-35.6 115.1h-.6L537.4 193h-31.6L557 334.9l-2.8 8.6c-4.6 14.6-12.1 20.3-25.5 20.3-2.4 0-7-.3-8.9-.5v23.4c1.8.4 9.3.7 11.6.7z\"],padlet:[640,512,[],\"e4a0\",\"M297.9 0L298 .001C305.6 .1078 312.4 4.72 315.5 11.78L447.5 320.3L447.8 320.2L448 320.6L445.2 330.6L402.3 488.6C398.6 504.8 382.6 514.9 366.5 511.2L298.1 495.6L229.6 511.2C213.5 514.9 197.5 504.8 193.8 488.6L150.9 330.6L148.2 320.6L148.3 320.2L280.4 11.78C283.4 4.797 290.3 .1837 297.9 .0006L297.9 0zM160.1 322.1L291.1 361.2L298 483.7L305.9 362.2L436.5 322.9L436.7 322.8L305.7 347.9L297.1 27.72L291.9 347.9L160.1 322.1zM426 222.6L520.4 181.6H594.2L437.2 429.2L468.8 320.2L426 222.6zM597.5 181.4L638.9 257.6C642.9 265.1 635 273.5 627.3 269.8L579.7 247.1L597.5 181.4zM127.3 318.5L158.7 430L1.61 154.5C-4.292 144.1 7.128 132.5 17.55 138.3L169.4 222.5L127.3 318.5z\"],\"amazon-pay\":[640,512,[],\"f42c\",\"M14 325.3c2.3-4.2 5.2-4.9 9.7-2.5 10.4 5.6 20.6 11.4 31.2 16.7a595.88 595.88 0 0 0 127.4 46.3 616.61 616.61 0 0 0 63.2 11.8 603.33 603.33 0 0 0 95 5.2c17.4-.4 34.8-1.8 52.1-3.8a603.66 603.66 0 0 0 163.3-42.8c2.9-1.2 5.9-2 9.1-1.2 6.7 1.8 9 9 4.1 13.9a70 70 0 0 1-9.6 7.4c-30.7 21.1-64.2 36.4-99.6 47.9a473.31 473.31 0 0 1-75.1 17.6 431 431 0 0 1-53.2 4.8 21.3 21.3 0 0 0-2.5.3H308a21.3 21.3 0 0 0-2.5-.3c-3.6-.2-7.2-.3-10.7-.4a426.3 426.3 0 0 1-50.4-5.3A448.4 448.4 0 0 1 164 420a443.33 443.33 0 0 1-145.6-87c-1.8-1.6-3-3.8-4.4-5.7zM172 65.1l-4.3.6a80.92 80.92 0 0 0-38 15.1c-2.4 1.7-4.6 3.5-7.1 5.4a4.29 4.29 0 0 1-.4-1.4c-.4-2.7-.8-5.5-1.3-8.2-.7-4.6-3-6.6-7.6-6.6h-11.5c-6.9 0-8.2 1.3-8.2 8.2v209.3c0 1 0 2 .1 3 .2 3 2 4.9 4.9 5 7 .1 14.1.1 21.1 0 2.9 0 4.7-2 5-5 .1-1 .1-2 .1-3v-72.4c1.1.9 1.7 1.4 2.2 1.9 17.9 14.9 38.5 19.8 61 15.4 20.4-4 34.6-16.5 43.8-34.9 7-13.9 9.9-28.7 10.3-44.1.5-17.1-1.2-33.9-8.1-49.8-8.5-19.6-22.6-32.5-43.9-36.9-3.2-.7-6.5-1-9.8-1.5-2.8-.1-5.5-.1-8.3-.1zM124.6 107a3.48 3.48 0 0 1 1.7-3.3c13.7-9.5 28.8-14.5 45.6-13.2 14.9 1.1 27.1 8.4 33.5 25.9 3.9 10.7 4.9 21.8 4.9 33 0 10.4-.8 20.6-4 30.6-6.8 21.3-22.4 29.4-42.6 28.5-14-.6-26.2-6-37.4-13.9a3.57 3.57 0 0 1-1.7-3.3c.1-14.1 0-28.1 0-42.2s.1-28 0-42.1zm205.7-41.9c-1 .1-2 .3-2.9.4a148 148 0 0 0-28.9 4.1c-6.1 1.6-12 3.8-17.9 5.8-3.6 1.2-5.4 3.8-5.3 7.7.1 3.3-.1 6.6 0 9.9.1 4.8 2.1 6.1 6.8 4.9 7.8-2 15.6-4.2 23.5-5.7 12.3-2.3 24.7-3.3 37.2-1.4 6.5 1 12.6 2.9 16.8 8.4 3.7 4.8 5.1 10.5 5.3 16.4.3 8.3.2 16.6.3 24.9a7.84 7.84 0 0 1-.2 1.4c-.5-.1-.9 0-1.3-.1a180.56 180.56 0 0 0-32-4.9c-11.3-.6-22.5.1-33.3 3.9-12.9 4.5-23.3 12.3-29.4 24.9-4.7 9.8-5.4 20.2-3.9 30.7 2 14 9 24.8 21.4 31.7 11.9 6.6 24.8 7.4 37.9 5.4 15.1-2.3 28.5-8.7 40.3-18.4a7.36 7.36 0 0 1 1.6-1.1c.6 3.8 1.1 7.4 1.8 11 .6 3.1 2.5 5.1 5.4 5.2 5.4.1 10.9.1 16.3 0a4.84 4.84 0 0 0 4.8-4.7 26.2 26.2 0 0 0 .1-2.8v-106a80 80 0 0 0-.9-12.9c-1.9-12.9-7.4-23.5-19-30.4-6.7-4-14.1-6-21.8-7.1-3.6-.5-7.2-.8-10.8-1.3-3.9.1-7.9.1-11.9.1zm35 127.7a3.33 3.33 0 0 1-1.5 3c-11.2 8.1-23.5 13.5-37.4 14.9-5.7.6-11.4.4-16.8-1.8a20.08 20.08 0 0 1-12.4-13.3 32.9 32.9 0 0 1-.1-19.4c2.5-8.3 8.4-13 16.4-15.6a61.33 61.33 0 0 1 24.8-2.2c8.4.7 16.6 2.3 25 3.4 1.6.2 2.1 1 2.1 2.6-.1 4.8 0 9.5 0 14.3s-.2 9.4-.1 14.1zm259.9 129.4c-1-5-4.8-6.9-9.1-8.3a88.42 88.42 0 0 0-21-3.9 147.32 147.32 0 0 0-39.2 1.9c-14.3 2.7-27.9 7.3-40 15.6a13.75 13.75 0 0 0-3.7 3.5 5.11 5.11 0 0 0-.5 4c.4 1.5 2.1 1.9 3.6 1.8a16.2 16.2 0 0 0 2.2-.1c7.8-.8 15.5-1.7 23.3-2.5 11.4-1.1 22.9-1.8 34.3-.9a71.64 71.64 0 0 1 14.4 2.7c5.1 1.4 7.4 5.2 7.6 10.4.4 8-1.4 15.7-3.5 23.3-4.1 15.4-10 30.3-15.8 45.1a17.6 17.6 0 0 0-1 3c-.5 2.9 1.2 4.8 4.1 4.1a10.56 10.56 0 0 0 4.8-2.5 145.91 145.91 0 0 0 12.7-13.4c12.8-16.4 20.3-35.3 24.7-55.6.8-3.6 1.4-7.3 2.1-10.9v-17.3zM493.1 199q-19.35-53.55-38.7-107.2c-2-5.7-4.2-11.3-6.3-16.9-1.1-2.9-3.2-4.8-6.4-4.8-7.6-.1-15.2-.2-22.9-.1-2.5 0-3.7 2-3.2 4.5a43.1 43.1 0 0 0 1.9 6.1q29.4 72.75 59.1 145.5c1.7 4.1 2.1 7.6.2 11.8-3.3 7.3-5.9 15-9.3 22.3-3 6.5-8 11.4-15.2 13.3a42.13 42.13 0 0 1-15.4 1.1c-2.5-.2-5-.8-7.5-1-3.4-.2-5.1 1.3-5.2 4.8q-.15 5 0 9.9c.1 5.5 2 8 7.4 8.9a108.18 108.18 0 0 0 16.9 2c17.1.4 30.7-6.5 39.5-21.4a131.63 131.63 0 0 0 9.2-18.4q35.55-89.7 70.6-179.6a26.62 26.62 0 0 0 1.6-5.5c.4-2.8-.9-4.4-3.7-4.4-6.6-.1-13.3 0-19.9 0a7.54 7.54 0 0 0-7.7 5.2c-.5 1.4-1.1 2.7-1.6 4.1l-34.8 100c-2.5 7.2-5.1 14.5-7.7 22.2-.4-1.1-.6-1.7-.9-2.4z\"],\"square-github\":[448,512,[\"github-square\"],\"f092\",\"M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM265.8 407.7c0-1.8 0-6 .1-11.6c.1-11.4 .1-28.8 .1-43.7c0-15.6-5.2-25.5-11.3-30.7c37-4.1 76-9.2 76-73.1c0-18.2-6.5-27.3-17.1-39c1.7-4.3 7.4-22-1.7-45c-13.9-4.3-45.7 17.9-45.7 17.9c-13.2-3.7-27.5-5.6-41.6-5.6s-28.4 1.9-41.6 5.6c0 0-31.8-22.2-45.7-17.9c-9.1 22.9-3.5 40.6-1.7 45c-10.6 11.7-15.6 20.8-15.6 39c0 63.6 37.3 69 74.3 73.1c-4.8 4.3-9.1 11.7-10.6 22.3c-9.5 4.3-33.8 11.7-48.3-13.9c-9.1-15.8-25.5-17.1-25.5-17.1c-16.2-.2-1.1 10.2-1.1 10.2c10.8 5 18.4 24.2 18.4 24.2c9.7 29.7 56.1 19.7 56.1 19.7c0 9 .1 21.7 .1 30.6c0 4.8 .1 8.6 .1 10c0 4.3-3 9.5-11.5 8C106 393.6 59.8 330.8 59.8 257.4c0-91.8 70.2-161.5 162-161.5s166.2 69.7 166.2 161.5c.1 73.4-44.7 136.3-110.7 158.3c-8.4 1.5-11.5-3.7-11.5-8zm-90.5-54.8c-.2-1.5 1.1-2.8 3-3.2c1.9-.2 3.7 .6 3.9 1.9c.3 1.3-1 2.6-3 3c-1.9 .4-3.7-.4-3.9-1.7zm-9.1 3.2c-2.2 .2-3.7-.9-3.7-2.4c0-1.3 1.5-2.4 3.5-2.4c1.9-.2 3.7 .9 3.7 2.4c0 1.3-1.5 2.4-3.5 2.4zm-14.3-2.2c-1.9-.4-3.2-1.9-2.8-3.2s2.4-1.9 4.1-1.5c2 .6 3.3 2.1 2.8 3.4c-.4 1.3-2.4 1.9-4.1 1.3zm-12.5-7.3c-1.5-1.3-1.9-3.2-.9-4.1c.9-1.1 2.8-.9 4.3 .6c1.3 1.3 1.8 3.3 .9 4.1c-.9 1.1-2.8 .9-4.3-.6zm-8.5-10c-1.1-1.5-1.1-3.2 0-3.9c1.1-.9 2.8-.2 3.7 1.3c1.1 1.5 1.1 3.3 0 4.1c-.9 .6-2.6 0-3.7-1.5zm-6.3-8.8c-1.1-1.3-1.3-2.8-.4-3.5c.9-.9 2.4-.4 3.5 .6c1.1 1.3 1.3 2.8 .4 3.5c-.9 .9-2.4 .4-3.5-.6zm-6-6.4c-1.3-.6-1.9-1.7-1.5-2.6c.4-.6 1.5-.9 2.8-.4c1.3 .7 1.9 1.8 1.5 2.6c-.4 .9-1.7 1.1-2.8 .4z\"],stumbleupon:[512,512,[],\"f1a4\",\"M502.9 266v69.7c0 62.1-50.3 112.4-112.4 112.4-61.8 0-112.4-49.8-112.4-111.3v-70.2l34.3 16 51.1-15.2V338c0 14.7 12 26.5 26.7 26.5S417 352.7 417 338v-72h85.9zm-224.7-58.2l34.3 16 51.1-15.2V173c0-60.5-51.1-109-112.1-109-60.8 0-112.1 48.2-112.1 108.2v162.4c0 14.9-12 26.7-26.7 26.7S86 349.5 86 334.6V266H0v69.7C0 397.7 50.3 448 112.4 448c61.6 0 112.4-49.5 112.4-110.8V176.9c0-14.7 12-26.7 26.7-26.7s26.7 12 26.7 26.7v30.9z\"],fedex:[640,512,[],\"f797\",\"M586 284.5l53.3-59.9h-62.4l-21.7 24.8-22.5-24.8H414v-16h56.1v-48.1H318.9V236h-.5c-9.6-11-21.5-14.8-35.4-14.8-28.4 0-49.8 19.4-57.3 44.9-18-59.4-97.4-57.6-121.9-14v-24.2H49v-26.2h60v-41.1H0V345h49v-77.5h48.9c-1.5 5.7-2.3 11.8-2.3 18.2 0 73.1 102.6 91.4 130.2 23.7h-42c-14.7 20.9-45.8 8.9-45.8-14.6h85.5c3.7 30.5 27.4 56.9 60.1 56.9 14.1 0 27-6.9 34.9-18.6h.5V345h212.2l22.1-25 22.3 25H640l-54-60.5zm-446.7-16.6c6.1-26.3 41.7-25.6 46.5 0h-46.5zm153.4 48.9c-34.6 0-34-62.8 0-62.8 32.6 0 34.5 62.8 0 62.8zm167.8 19.1h-94.4V169.4h95v30.2H405v33.9h55.5v28.1h-56.1v44.7h56.1v29.6zm-45.9-39.8v-24.4h56.1v-44l50.7 57-50.7 57v-45.6h-56.1zm138.6 10.3l-26.1 29.5H489l45.6-51.2-45.6-51.2h39.7l26.6 29.3 25.6-29.3h38.5l-45.4 51 46 51.4h-40.5l-26.3-29.5z\"],\"phoenix-framework\":[640,512,[],\"f3dc\",\"M212.9 344.3c3.8-.1 22.8-1.4 25.6-2.2-2.4-2.6-43.6-1-68-49.6-4.3-8.6-7.5-17.6-6.4-27.6 2.9-25.5 32.9-30 52-18.5 36 21.6 63.3 91.3 113.7 97.5 37 4.5 84.6-17 108.2-45.4-.6-.1-.8-.2-1-.1-.4.1-.8.2-1.1.3-33.3 12.1-94.3 9.7-134.7-14.8-37.6-22.8-53.1-58.7-51.8-74.6 1.8-21.3 22.9-23.2 35.9-19.6 14.4 3.9 24.4 17.6 38.9 27.4 15.6 10.4 32.9 13.7 51.3 10.3 14.9-2.7 34.4-12.3 36.5-14.5-1.1-.1-1.8-.1-2.5-.2-6.2-.6-12.4-.8-18.5-1.7C279.8 194.5 262.1 47.4 138.5 37.9 94.2 34.5 39.1 46 2.2 72.9c-.8.6-1.5 1.2-2.2 1.8.1.2.1.3.2.5.8 0 1.6-.1 2.4-.2 6.3-1 12.5-.8 18.7.3 23.8 4.3 47.7 23.1 55.9 76.5 5.3 34.3-.7 50.8 8 86.1 19 77.1 91 107.6 127.7 106.4zM75.3 64.9c-.9-1-.9-1.2-1.3-2 12.1-2.6 24.2-4.1 36.6-4.8-1.1 14.7-22.2 21.3-35.3 6.8zm196.9 350.5c-42.8 1.2-92-26.7-123.5-61.4-4.6-5-16.8-20.2-18.6-23.4l.4-.4c6.6 4.1 25.7 18.6 54.8 27 24.2 7 48.1 6.3 71.6-3.3 22.7-9.3 41-.5 43.1 2.9-18.5 3.8-20.1 4.4-24 7.9-5.1 4.4-4.6 11.7 7 17.2 26.2 12.4 63-2.8 97.2 25.4 2.4 2 8.1 7.8 10.1 10.7-.1.2-.3.3-.4.5-4.8-1.5-16.4-7.5-40.2-9.3-24.7-2-46.3 5.3-77.5 6.2zm174.8-252c16.4-5.2 41.3-13.4 66.5-3.3 16.1 6.5 26.2 18.7 32.1 34.6 3.5 9.4 5.1 19.7 5.1 28.7-.2 0-.4 0-.6.1-.2-.4-.4-.9-.5-1.3-5-22-29.9-43.8-67.6-29.9-50.2 18.6-130.4 9.7-176.9-48-.7-.9-2.4-1.7-1.3-3.2.1-.2 2.1.6 3 1.3 18.1 13.4 38.3 21.9 60.3 26.2 30.5 6.1 54.6 2.9 79.9-5.2zm102.7 117.5c-32.4.2-33.8 50.1-103.6 64.4-18.2 3.7-38.7 4.6-44.9 4.2v-.4c2.8-1.5 14.7-2.6 29.7-16.6 7.9-7.3 15.3-15.1 22.8-22.9 19.5-20.2 41.4-42.2 81.9-39 23.1 1.8 29.3 8.2 36.1 12.7.3.2.4.5.7.9-.5 0-.7.1-.9 0-7-2.7-14.3-3.3-21.8-3.3zm-12.3-24.1c-.1.2-.1.4-.2.6-28.9-4.4-48-7.9-68.5 4-17 9.9-31.4 20.5-62 24.4-27.1 3.4-45.1 2.4-66.1-8-.3-.2-.6-.4-1-.6 0-.2.1-.3.1-.5 24.9 3.8 36.4 5.1 55.5-5.8 22.3-12.9 40.1-26.6 71.3-31 29.6-4.1 51.3 2.5 70.9 16.9zM268.6 97.3c-.6-.6-1.1-1.2-2.1-2.3 7.6 0 29.7-1.2 53.4 8.4 19.7 8 32.2 21 50.2 32.9 11.1 7.3 23.4 9.3 36.4 8.1 4.3-.4 8.5-1.2 12.8-1.7.4-.1.9 0 1.5.3-.6.4-1.2.9-1.8 1.2-8.1 4-16.7 6.3-25.6 7.1-26.1 2.6-50.3-3.7-73.4-15.4-19.3-9.9-36.4-22.9-51.4-38.6zM640 335.7c-3.5 3.1-22.7 11.6-42.7 5.3-12.3-3.9-19.5-14.9-31.6-24.1-10-7.6-20.9-7.9-28.1-8.4.6-.8.9-1.2 1.2-1.4 14.8-9.2 30.5-12.2 47.3-6.5 12.5 4.2 19.2 13.5 30.4 24.2 10.8 10.4 21 9.9 23.1 10.5.1-.1.2 0 .4.4zm-212.5 137c2.2 1.2 1.6 1.5 1.5 2-18.5-1.4-33.9-7.6-46.8-22.2-21.8-24.7-41.7-27.9-48.6-29.7.5-.2.8-.4 1.1-.4 13.1.1 26.1.7 38.9 3.9 25.3 6.4 35 25.4 41.6 35.3 3.2 4.8 7.3 8.3 12.3 11.1z\"],shopify:[448,512,[],\"e057\",\"M388.32,104.1a4.66,4.66,0,0,0-4.4-4c-2,0-37.23-.8-37.23-.8s-21.61-20.82-29.62-28.83V503.2L442.76,472S388.72,106.5,388.32,104.1ZM288.65,70.47a116.67,116.67,0,0,0-7.21-17.61C271,32.85,255.42,22,237,22a15,15,0,0,0-4,.4c-.4-.8-1.2-1.2-1.6-2C223.4,11.63,213,7.63,200.58,8c-24,.8-48,18-67.25,48.83-13.61,21.62-24,48.84-26.82,70.06-27.62,8.4-46.83,14.41-47.23,14.81-14,4.4-14.41,4.8-16,18-1.2,10-38,291.82-38,291.82L307.86,504V65.67a41.66,41.66,0,0,0-4.4.4S297.86,67.67,288.65,70.47ZM233.41,87.69c-16,4.8-33.63,10.4-50.84,15.61,4.8-18.82,14.41-37.63,25.62-50,4.4-4.4,10.41-9.61,17.21-12.81C232.21,54.86,233.81,74.48,233.41,87.69ZM200.58,24.44A27.49,27.49,0,0,1,215,28c-6.4,3.2-12.81,8.41-18.81,14.41-15.21,16.42-26.82,42-31.62,66.45-14.42,4.41-28.83,8.81-42,12.81C131.33,83.28,163.75,25.24,200.58,24.44ZM154.15,244.61c1.6,25.61,69.25,31.22,73.25,91.66,2.8,47.64-25.22,80.06-65.65,82.47-48.83,3.2-75.65-25.62-75.65-25.62l10.4-44s26.82,20.42,48.44,18.82c14-.8,19.22-12.41,18.81-20.42-2-33.62-57.24-31.62-60.84-86.86-3.2-46.44,27.22-93.27,94.47-97.68,26-1.6,39.23,4.81,39.23,4.81L221.4,225.39s-17.21-8-37.63-6.4C154.15,221,153.75,239.8,154.15,244.61ZM249.42,82.88c0-12-1.6-29.22-7.21-43.63,18.42,3.6,27.22,24,31.23,36.43Q262.63,78.68,249.42,82.88Z\"],neos:[512,512,[],\"f612\",\"M415.44 512h-95.11L212.12 357.46v91.1L125.69 512H28V29.82L68.47 0h108.05l123.74 176.13V63.45L386.69 0h97.69v461.5zM38.77 35.27V496l72-52.88V194l215.5 307.64h84.79l52.35-38.17h-78.27L69 13zm82.54 466.61l80-58.78v-101l-79.76-114.4v220.94L49 501.89h72.34zM80.63 10.77l310.6 442.57h82.37V10.77h-79.75v317.56L170.91 10.77zM311 191.65l72 102.81V15.93l-72 53v122.72z\"],\"square-threads\":[448,512,[],\"e619\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM294.2 244.3c19.5 9.3 33.7 23.5 41.2 40.9c10.4 24.3 11.4 63.9-20.2 95.4c-24.2 24.1-53.5 35-95.1 35.3h-.2c-46.8-.3-82.8-16.1-106.9-46.8C91.5 341.8 80.4 303.7 80 256v-.1-.1c.4-47.7 11.5-85.7 33-113.1c24.2-30.7 60.2-46.5 106.9-46.8h.2c46.9 .3 83.3 16 108.2 46.6c12.3 15.1 21.3 33.3 27 54.4l-26.9 7.2c-4.7-17.2-11.9-31.9-21.4-43.6c-19.4-23.9-48.7-36.1-87-36.4c-38 .3-66.8 12.5-85.5 36.2c-17.5 22.3-26.6 54.4-26.9 95.5c.3 41.1 9.4 73.3 26.9 95.5c18.7 23.8 47.4 36 85.5 36.2c34.3-.3 56.9-8.4 75.8-27.3c21.5-21.5 21.1-47.9 14.2-64c-4-9.4-11.4-17.3-21.3-23.3c-2.4 18-7.9 32.2-16.5 43.2c-11.4 14.5-27.7 22.4-48.4 23.5c-15.7 .9-30.8-2.9-42.6-10.7c-13.9-9.2-22-23.2-22.9-39.5c-1.7-32.2 23.8-55.3 63.5-57.6c14.1-.8 27.3-.2 39.5 1.9c-1.6-9.9-4.9-17.7-9.8-23.4c-6.7-7.8-17.1-11.8-30.8-11.9h-.4c-11 0-26 3.1-35.6 17.6l-23-15.8c12.8-19.4 33.6-30.1 58.5-30.1h.6c41.8 .3 66.6 26.3 69.1 71.8c1.4 .6 2.8 1.2 4.2 1.9l.1 .5zm-71.8 67.5c17-.9 36.4-7.6 39.7-48.8c-8.8-1.9-18.6-2.9-29-2.9c-3.2 0-6.4 .1-9.6 .3c-28.6 1.6-38.1 15.5-37.4 27.9c.9 16.7 19 24.5 36.4 23.6l-.1-.1z\"],hackerrank:[512,512,[],\"f5f7\",\"M477.5 128C463 103.05 285.13 0 256.16 0S49.25 102.79 34.84 128s-14.49 230.8 0 256 192.38 128 221.32 128S463 409.08 477.49 384s14.51-231 .01-256zM316.13 414.22c-4 0-40.91-35.77-38-38.69.87-.87 6.26-1.48 17.55-1.83 0-26.23.59-68.59.94-86.32 0-2-.44-3.43-.44-5.85h-79.93c0 7.1-.46 36.2 1.37 72.88.23 4.54-1.58 6-5.74 5.94-10.13 0-20.27-.11-30.41-.08-4.1 0-5.87-1.53-5.74-6.11.92-33.44 3-84-.15-212.67v-3.17c-9.67-.35-16.38-1-17.26-1.84-2.92-2.92 34.54-38.69 38.49-38.69s41.17 35.78 38.27 38.69c-.87.87-7.9 1.49-16.77 1.84v3.16c-2.42 25.75-2 79.59-2.63 105.39h80.26c0-4.55.39-34.74-1.2-83.64-.1-3.39.95-5.17 4.21-5.2 11.07-.08 22.15-.13 33.23-.06 3.46 0 4.57 1.72 4.5 5.38C333 354.64 336 341.29 336 373.69c8.87.35 16.82 1 17.69 1.84 2.88 2.91-33.62 38.69-37.58 38.69z\"],researchgate:[448,512,[],\"f4f8\",\"M0 32v448h448V32H0zm262.2 334.4c-6.6 3-33.2 6-50-14.2-9.2-10.6-25.3-33.3-42.2-63.6-8.9 0-14.7 0-21.4-.6v46.4c0 23.5 6 21.2 25.8 23.9v8.1c-6.9-.3-23.1-.8-35.6-.8-13.1 0-26.1.6-33.6.8v-8.1c15.5-2.9 22-1.3 22-23.9V225c0-22.6-6.4-21-22-23.9V193c25.8 1 53.1-.6 70.9-.6 31.7 0 55.9 14.4 55.9 45.6 0 21.1-16.7 42.2-39.2 47.5 13.6 24.2 30 45.6 42.2 58.9 7.2 7.8 17.2 14.7 27.2 14.7v7.3zm22.9-135c-23.3 0-32.2-15.7-32.2-32.2V167c0-12.2 8.8-30.4 34-30.4s30.4 17.9 30.4 17.9l-10.7 7.2s-5.5-12.5-19.7-12.5c-7.9 0-19.7 7.3-19.7 19.7v26.8c0 13.4 6.6 23.3 17.9 23.3 14.1 0 21.5-10.9 21.5-26.8h-17.9v-10.7h30.4c0 20.5 4.7 49.9-34 49.9zm-116.5 44.7c-9.4 0-13.6-.3-20-.8v-69.7c6.4-.6 15-.6 22.5-.6 23.3 0 37.2 12.2 37.2 34.5 0 21.9-15 36.6-39.7 36.6z\"],swift:[448,512,[],\"f8e1\",\"M448 156.09c0-4.51-.08-9-.2-13.52a196.31 196.31 0 0 0-2.58-29.42 99.62 99.62 0 0 0-9.22-28A94.08 94.08 0 0 0 394.84 44a99.17 99.17 0 0 0-28-9.22 195 195 0 0 0-29.43-2.59c-4.51-.12-9-.17-13.52-.2H124.14c-4.51 0-9 .08-13.52.2-2.45.07-4.91.15-7.37.27a171.68 171.68 0 0 0-22.06 2.32 103.06 103.06 0 0 0-21.21 6.1q-3.46 1.45-6.81 3.12a94.66 94.66 0 0 0-18.39 12.32c-1.88 1.61-3.69 3.28-5.43 5A93.86 93.86 0 0 0 12 85.17a99.45 99.45 0 0 0-9.22 28 196.31 196.31 0 0 0-2.54 29.4c-.13 4.51-.18 9-.21 13.52v199.83c0 4.51.08 9 .21 13.51a196.08 196.08 0 0 0 2.58 29.42 99.3 99.3 0 0 0 9.22 28A94.31 94.31 0 0 0 53.17 468a99.47 99.47 0 0 0 28 9.21 195 195 0 0 0 29.43 2.59c4.5.12 9 .17 13.52.2H323.91c4.51 0 9-.08 13.52-.2a196.59 196.59 0 0 0 29.44-2.59 99.57 99.57 0 0 0 28-9.21A94.22 94.22 0 0 0 436 426.84a99.3 99.3 0 0 0 9.22-28 194.79 194.79 0 0 0 2.59-29.42c.12-4.5.17-9 .2-13.51V172.14c-.01-5.35-.01-10.7-.01-16.05zm-69.88 241c-20-38.93-57.23-29.27-76.31-19.47-1.72 1-3.48 2-5.25 3l-.42.25c-39.5 21-92.53 22.54-145.85-.38A234.64 234.64 0 0 1 45 290.12a230.63 230.63 0 0 0 39.17 23.37c56.36 26.4 113 24.49 153 0-57-43.85-104.6-101-141.09-147.22a197.09 197.09 0 0 1-18.78-25.9c43.7 40 112.7 90.22 137.48 104.12-52.57-55.49-98.89-123.94-96.72-121.74 82.79 83.42 159.18 130.59 159.18 130.59 2.88 1.58 5 2.85 6.73 4a127.44 127.44 0 0 0 4.16-12.47c13.22-48.33-1.66-103.58-35.31-149.2C329.61 141.75 375 229.34 356.4 303.42c-.44 1.73-.95 3.4-1.44 5.09 38.52 47.4 28.04 98.17 23.13 88.59z\"],angular:[448,512,[],\"f420\",\"M185.7 268.1h76.2l-38.1-91.6-38.1 91.6zM223.8 32L16 106.4l31.8 275.7 176 97.9 176-97.9 31.8-275.7zM354 373.8h-48.6l-26.2-65.4H168.6l-26.2 65.4H93.7L223.8 81.5z\"],speakap:[448,512,[],\"f3f3\",\"M64 391.78C-15.41 303.59-8 167.42 80.64 87.64s224.8-73 304.21 15.24 72 224.36-16.64 304.14c-18.74 16.87 64 43.09 42 52.26-82.06 34.21-253.91 35-346.23-67.5zm213.31-211.6l38.5-40.86c-9.61-8.89-32-26.83-76.17-27.6-52.33-.91-95.86 28.3-96.77 80-.2 11.33.29 36.72 29.42 54.83 34.46 21.42 86.52 21.51 86 52.26-.37 21.28-26.42 25.81-38.59 25.6-3-.05-30.23-.46-47.61-24.62l-40 42.61c28.16 27 59 32.62 83.49 33.05 10.23.18 96.42.33 97.84-81 .28-15.81-2.07-39.72-28.86-56.59-34.36-21.64-85-19.45-84.43-49.75.41-23.25 31-25.37 37.53-25.26.43 0 26.62.26 39.62 17.37z\"],angrycreative:[640,512,[],\"f36e\",\"M640 238.2l-3.2 28.2-34.5 2.3-2 18.1 34.5-2.3-3.2 28.2-34.4 2.2-2.3 20.1 34.4-2.2-3 26.1-64.7 4.1 12.7-113.2L527 365.2l-31.9 2-23.8-117.8 30.3-2 13.6 79.4 31.7-82.4 93.1-6.2zM426.8 371.5l28.3-1.8L468 249.6l-28.4 1.9-12.8 120zM162 388.1l-19.4-36-3.5 37.4-28.2 1.7 2.7-29.1c-11 18-32 34.3-56.9 35.8C23.9 399.9-3 377 .3 339.7c2.6-29.3 26.7-62.8 67.5-65.4 37.7-2.4 47.6 23.2 51.3 28.8l2.8-30.8 38.9-2.5c20.1-1.3 38.7 3.7 42.5 23.7l2.6-26.6 64.8-4.2-2.7 27.9-36.4 2.4-1.7 17.9 36.4-2.3-2.7 27.9-36.4 2.3-1.9 19.9 36.3-2.3-2.1 20.8 55-117.2 23.8-1.6L370.4 369l8.9-85.6-22.3 1.4 2.9-27.9 75-4.9-3 28-24.3 1.6-9.7 91.9-58 3.7-4.3-15.6-39.4 2.5-8 16.3-126.2 7.7zm-44.3-70.2l-26.4 1.7C84.6 307.2 76.9 303 65 303.8c-19 1.2-33.3 17.5-34.6 33.3-1.4 16 7.3 32.5 28.7 31.2 12.8-.8 21.3-8.6 28.9-18.9l27-1.7 2.7-29.8zm56.1-7.7c1.2-12.9-7.6-13.6-26.1-12.4l-2.7 28.5c14.2-.9 27.5-2.1 28.8-16.1zm21.1 70.8l5.8-60c-5 13.5-14.7 21.1-27.9 26.6l22.1 33.4zm135.4-45l-7.9-37.8-15.8 39.3 23.7-1.5zm-170.1-74.6l-4.3-17.5-39.6 2.6-8.1 18.2-31.9 2.1 57-121.9 23.9-1.6 30.7 102 9.9-104.7 27-1.8 37.8 63.6 6.5-66.6 28.5-1.9-4 41.2c7.4-13.5 22.9-44.7 63.6-47.5 40.5-2.8 52.4 29.3 53.4 30.3l3.3-32 39.3-2.7c12.7-.9 27.8.3 36.3 9.7l-4.4-11.9 32.2-2.2 12.9 43.2 23-45.7 31-2.2-43.6 78.4-4.8 44.3-28.4 1.9 4.8-44.3-15.8-43c1 22.3-9.2 40.1-32 49.6l25.2 38.8-36.4 2.4-19.2-36.8-4 38.3-28.4 1.9 3.3-31.5c-6.7 9.3-19.7 35.4-59.6 38-26.2 1.7-45.6-10.3-55.4-39.2l-4 40.3-25 1.6-37.6-63.3-6.3 66.2-56.8 3.7zm276.6-82.1c10.2-.7 17.5-2.1 21.6-4.3 4.5-2.4 7-6.4 7.6-12.1.6-5.3-.6-8.8-3.4-10.4-3.6-2.1-10.6-2.8-22.9-2l-2.9 28.8zM327.7 214c5.6 5.9 12.7 8.5 21.3 7.9 4.7-.3 9.1-1.8 13.3-4.1 5.5-3 10.6-8 15.1-14.3l-34.2 2.3 2.4-23.9 63.1-4.3 1.2-12-31.2 2.1c-4.1-3.7-7.8-6.6-11.1-8.1-4-1.7-8.1-2.8-12.2-2.5-8 .5-15.3 3.6-22 9.2-7.7 6.4-12 14.5-12.9 24.4-1.1 9.6 1.4 17.3 7.2 23.3zm-201.3 8.2l23.8-1.6-8.3-37.6-15.5 39.2z\"],\"y-combinator\":[448,512,[],\"f23b\",\"M448 32v448H0V32h448zM236 287.5L313.5 142h-32.7L235 233c-4.7 9.3-9 18.3-12.8 26.8L210 233l-45.2-91h-35l76.7 143.8v94.5H236v-92.8z\"],empire:[496,512,[],\"f1d1\",\"M287.6 54.2c-10.8-2.2-22.1-3.3-33.5-3.6V32.4c78.1 2.2 146.1 44 184.6 106.6l-15.8 9.1c-6.1-9.7-12.7-18.8-20.2-27.1l-18 15.5c-26-29.6-61.4-50.7-101.9-58.4l4.8-23.9zM53.4 322.4l23-7.7c-6.4-18.3-10-38.2-10-58.7s3.3-40.4 9.7-58.7l-22.7-7.7c3.6-10.8 8.3-21.3 13.6-31l-15.8-9.1C34 181 24.1 217.5 24.1 256s10 75 27.1 106.6l15.8-9.1c-5.3-10-9.7-20.3-13.6-31.1zM213.1 434c-40.4-8-75.8-29.1-101.9-58.7l-18 15.8c-7.5-8.6-14.4-17.7-20.2-27.4l-16 9.4c38.5 62.3 106.8 104.3 184.9 106.6v-18.3c-11.3-.3-22.7-1.7-33.5-3.6l4.7-23.8zM93.3 120.9l18 15.5c26-29.6 61.4-50.7 101.9-58.4l-4.7-23.8c10.8-2.2 22.1-3.3 33.5-3.6V32.4C163.9 34.6 95.9 76.4 57.4 139l15.8 9.1c6-9.7 12.6-18.9 20.1-27.2zm309.4 270.2l-18-15.8c-26 29.6-61.4 50.7-101.9 58.7l4.7 23.8c-10.8 1.9-22.1 3.3-33.5 3.6v18.3c78.1-2.2 146.4-44.3 184.9-106.6l-16.1-9.4c-5.7 9.7-12.6 18.8-20.1 27.4zM496 256c0 137-111 248-248 248S0 393 0 256 111 8 248 8s248 111 248 248zm-12.2 0c0-130.1-105.7-235.8-235.8-235.8S12.2 125.9 12.2 256 117.9 491.8 248 491.8 483.8 386.1 483.8 256zm-39-106.6l-15.8 9.1c5.3 9.7 10 20.2 13.6 31l-22.7 7.7c6.4 18.3 9.7 38.2 9.7 58.7s-3.6 40.4-10 58.7l23 7.7c-3.9 10.8-8.3 21-13.6 31l15.8 9.1C462 331 471.9 294.5 471.9 256s-9.9-75-27.1-106.6zm-183 177.7c16.3-3.3 30.4-11.6 40.7-23.5l51.2 44.8c11.9-13.6 21.3-29.3 27.1-46.8l-64.2-22.1c2.5-7.5 3.9-15.2 3.9-23.5s-1.4-16.1-3.9-23.5l64.5-22.1c-6.1-17.4-15.5-33.2-27.4-46.8l-51.2 44.8c-10.2-11.9-24.4-20.5-40.7-23.8l13.3-66.4c-8.6-1.9-17.7-2.8-27.1-2.8-9.4 0-18.5.8-27.1 2.8l13.3 66.4c-16.3 3.3-30.4 11.9-40.7 23.8l-51.2-44.8c-11.9 13.6-21.3 29.3-27.4 46.8l64.5 22.1c-2.5 7.5-3.9 15.2-3.9 23.5s1.4 16.1 3.9 23.5l-64.2 22.1c5.8 17.4 15.2 33.2 27.1 46.8l51.2-44.8c10.2 11.9 24.4 20.2 40.7 23.5l-13.3 66.7c8.6 1.7 17.7 2.8 27.1 2.8 9.4 0 18.5-1.1 27.1-2.8l-13.3-66.7z\"],envira:[448,512,[],\"f299\",\"M0 32c477.6 0 366.6 317.3 367.1 366.3L448 480h-26l-70.4-71.2c-39 4.2-124.4 34.5-214.4-37C47 300.3 52 214.7 0 32zm79.7 46c-49.7-23.5-5.2 9.2-5.2 9.2 45.2 31.2 66 73.7 90.2 119.9 31.5 60.2 79 139.7 144.2 167.7 65 28 34.2 12.5 6-8.5-28.2-21.2-68.2-87-91-130.2-31.7-60-61-118.6-144.2-158.1z\"],\"google-scholar\":[512,512,[],\"e63b\",\"M390.9 298.5c0 0 0 .1 .1 .1c9.2 19.4 14.4 41.1 14.4 64C405.3 445.1 338.5 512 256 512s-149.3-66.9-149.3-149.3c0-22.9 5.2-44.6 14.4-64h0c1.7-3.6 3.6-7.2 5.6-10.7c4.4-7.6 9.4-14.7 15-21.3c27.4-32.6 68.5-53.3 114.4-53.3c33.6 0 64.6 11.1 89.6 29.9c9.1 6.9 17.4 14.7 24.8 23.5c5.6 6.6 10.6 13.8 15 21.3c2 3.4 3.8 7 5.5 10.5zm26.4-18.8c-30.1-58.4-91-98.4-161.3-98.4s-131.2 40-161.3 98.4L0 202.7 256 0 512 202.7l-94.7 77.1z\"],\"square-gitlab\":[448,512,[\"gitlab-square\"],\"e5ae\",\"M0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96zm337.5 12.5l44.6 116.4 .4 1.2c5.6 16.8 7.2 35.2 2.3 52.5c-5 17.2-15.4 32.4-29.8 43.3l-.2 .1-68.4 51.2-54.1 40.9c-.5 .2-1.1 .5-1.7 .8c-2 1-4.4 2-6.7 2c-3 0-6.8-1.8-8.3-2.8l-54.2-40.9L93.5 322.3l-.4-.3-.2-.1c-14.3-10.8-24.8-26-29.7-43.3s-4.2-35.7 2.2-52.5l.5-1.2 44.7-116.4c.9-2.3 2.5-4.3 4.5-5.6c1.6-1 3.4-1.6 5.2-1.8c1.3-.7 2.1-.4 3.4 .1c.6 .2 1.2 .5 2 .7c1 .4 1.6 .9 2.4 1.5c.6 .4 1.2 1 2.1 1.5c1.2 1.4 2.2 3 2.7 4.8l29.2 92.2H285l30.2-92.2c.5-1.8 1.4-3.4 2.6-4.8s2.8-2.4 4.5-3.1c1.7-.6 3.6-.9 5.4-.7s3.6 .8 5.2 1.8c2 1.3 3.7 3.3 4.6 5.6z\"],studiovinari:[512,512,[],\"f3f8\",\"M480.3 187.7l4.2 28v28l-25.1 44.1-39.8 78.4-56.1 67.5-79.1 37.8-17.7 24.5-7.7 12-9.6 4s17.3-63.6 19.4-63.6c2.1 0 20.3.7 20.3.7l66.7-38.6-92.5 26.1-55.9 36.8-22.8 28-6.6 1.4 20.8-73.6 6.9-5.5 20.7 12.9 88.3-45.2 56.8-51.5 14.8-68.4-125.4 23.3 15.2-18.2-173.4-53.3 81.9-10.5-166-122.9L133.5 108 32.2 0l252.9 126.6-31.5-38L378 163 234.7 64l18.7 38.4-49.6-18.1L158.3 0l194.6 122L310 66.2l108 96.4 12-8.9-21-16.4 4.2-37.8L451 89.1l29.2 24.7 11.5 4.2-7 6.2 8.5 12-13.1 7.4-10.3 20.2 10.5 23.9z\"],\"pied-piper\":[480,512,[],\"f2ae\",\"M455.93,23.2C429.23,30,387.79,51.69,341.35,90.66A206,206,0,0,0,240,64C125.13,64,32,157.12,32,272s93.13,208,208,208,208-93.13,208-208a207.25,207.25,0,0,0-58.75-144.81,155.35,155.35,0,0,0-17,27.4A176.16,176.16,0,0,1,417.1,272c0,97.66-79.44,177.11-177.09,177.11a175.81,175.81,0,0,1-87.63-23.4c82.94-107.33,150.79-37.77,184.31-226.65,5.79-32.62,28-94.26,126.23-160.18C471,33.45,465.35,20.8,455.93,23.2ZM125,406.4A176.66,176.66,0,0,1,62.9,272C62.9,174.34,142.35,94.9,240,94.9a174,174,0,0,1,76.63,17.75C250.64,174.76,189.77,265.52,125,406.4Z\"],wordpress:[512,512,[],\"f19a\",\"M61.7 169.4l101.5 278C92.2 413 43.3 340.2 43.3 256c0-30.9 6.6-60.1 18.4-86.6zm337.9 75.9c0-26.3-9.4-44.5-17.5-58.7-10.8-17.5-20.9-32.4-20.9-49.9 0-19.6 14.8-37.8 35.7-37.8.9 0 1.8.1 2.8.2-37.9-34.7-88.3-55.9-143.7-55.9-74.3 0-139.7 38.1-177.8 95.9 5 .2 9.7.3 13.7.3 22.2 0 56.7-2.7 56.7-2.7 11.5-.7 12.8 16.2 1.4 17.5 0 0-11.5 1.3-24.3 2l77.5 230.4L249.8 247l-33.1-90.8c-11.5-.7-22.3-2-22.3-2-11.5-.7-10.1-18.2 1.3-17.5 0 0 35.1 2.7 56 2.7 22.2 0 56.7-2.7 56.7-2.7 11.5-.7 12.8 16.2 1.4 17.5 0 0-11.5 1.3-24.3 2l76.9 228.7 21.2-70.9c9-29.4 16-50.5 16-68.7zm-139.9 29.3l-63.8 185.5c19.1 5.6 39.2 8.7 60.1 8.7 24.8 0 48.5-4.3 70.6-12.1-.6-.9-1.1-1.9-1.5-2.9l-65.4-179.2zm183-120.7c.9 6.8 1.4 14 1.4 21.9 0 21.6-4 45.8-16.2 76.2l-65 187.9C426.2 403 468.7 334.5 468.7 256c0-37-9.4-71.8-26-102.1zM504 256c0 136.8-111.3 248-248 248C119.2 504 8 392.7 8 256 8 119.2 119.2 8 256 8c136.7 0 248 111.2 248 248zm-11.4 0c0-130.5-106.2-236.6-236.6-236.6C125.5 19.4 19.4 125.5 19.4 256S125.6 492.6 256 492.6c130.5 0 236.6-106.1 236.6-236.6z\"],\"product-hunt\":[512,512,[],\"f288\",\"M326.3 218.8c0 20.5-16.7 37.2-37.2 37.2h-70.3v-74.4h70.3c20.5 0 37.2 16.7 37.2 37.2zM504 256c0 137-111 248-248 248S8 393 8 256 119 8 256 8s248 111 248 248zm-128.1-37.2c0-47.9-38.9-86.8-86.8-86.8H169.2v248h49.6v-74.4h70.3c47.9 0 86.8-38.9 86.8-86.8z\"],firefox:[512,512,[],\"f269\",\"M503.52,241.48c-.12-1.56-.24-3.12-.24-4.68v-.12l-.36-4.68v-.12a245.86,245.86,0,0,0-7.32-41.15c0-.12,0-.12-.12-.24l-1.08-4c-.12-.24-.12-.48-.24-.6-.36-1.2-.72-2.52-1.08-3.72-.12-.24-.12-.6-.24-.84-.36-1.2-.72-2.4-1.08-3.48-.12-.36-.24-.6-.36-1-.36-1.2-.72-2.28-1.2-3.48l-.36-1.08c-.36-1.08-.84-2.28-1.2-3.36a8.27,8.27,0,0,0-.36-1c-.48-1.08-.84-2.28-1.32-3.36-.12-.24-.24-.6-.36-.84-.48-1.2-1-2.28-1.44-3.48,0-.12-.12-.24-.12-.36-1.56-3.84-3.24-7.68-5-11.4l-.36-.72c-.48-1-.84-1.8-1.32-2.64-.24-.48-.48-1.08-.72-1.56-.36-.84-.84-1.56-1.2-2.4-.36-.6-.6-1.2-1-1.8s-.84-1.44-1.2-2.28c-.36-.6-.72-1.32-1.08-1.92s-.84-1.44-1.2-2.16a18.07,18.07,0,0,0-1.2-2c-.36-.72-.84-1.32-1.2-2s-.84-1.32-1.2-2-.84-1.32-1.2-1.92-.84-1.44-1.32-2.16a15.63,15.63,0,0,0-1.2-1.8L463.2,119a15.63,15.63,0,0,0-1.2-1.8c-.48-.72-1.08-1.56-1.56-2.28-.36-.48-.72-1.08-1.08-1.56l-1.8-2.52c-.36-.48-.6-.84-1-1.32-1-1.32-1.8-2.52-2.76-3.72a248.76,248.76,0,0,0-23.51-26.64A186.82,186.82,0,0,0,412,62.46c-4-3.48-8.16-6.72-12.48-9.84a162.49,162.49,0,0,0-24.6-15.12c-2.4-1.32-4.8-2.52-7.2-3.72a254,254,0,0,0-55.43-19.56c-1.92-.36-3.84-.84-5.64-1.2h-.12c-1-.12-1.8-.36-2.76-.48a236.35,236.35,0,0,0-38-4H255.14a234.62,234.62,0,0,0-45.48,5c-33.59,7.08-63.23,21.24-82.91,39-1.08,1-1.92,1.68-2.4,2.16l-.48.48H124l-.12.12.12-.12a.12.12,0,0,0,.12-.12l-.12.12a.42.42,0,0,1,.24-.12c14.64-8.76,34.92-16,49.44-19.56l5.88-1.44c.36-.12.84-.12,1.2-.24,1.68-.36,3.36-.72,5.16-1.08.24,0,.6-.12.84-.12C250.94,20.94,319.34,40.14,367,85.61a171.49,171.49,0,0,1,26.88,32.76c30.36,49.2,27.48,111.11,3.84,147.59-34.44,53-111.35,71.27-159,24.84a84.19,84.19,0,0,1-25.56-59,74.05,74.05,0,0,1,6.24-31c1.68-3.84,13.08-25.67,18.24-24.59-13.08-2.76-37.55,2.64-54.71,28.19-15.36,22.92-14.52,58.2-5,83.28a132.85,132.85,0,0,1-12.12-39.24c-12.24-82.55,43.31-153,94.31-170.51-27.48-24-96.47-22.31-147.71,15.36-29.88,22-51.23,53.16-62.51,90.36,1.68-20.88,9.6-52.08,25.8-83.88-17.16,8.88-39,37-49.8,62.88-15.6,37.43-21,82.19-16.08,124.79.36,3.24.72,6.36,1.08,9.6,19.92,117.11,122,206.38,244.78,206.38C392.77,503.42,504,392.19,504,255,503.88,250.48,503.76,245.92,503.52,241.48Z\"],linode:[448,512,[],\"f2b8\",\"M366.036,186.867l-59.5,36.871-.838,36.871-29.329-19.273-39.384,24.3c2.238,55.211,2.483,59.271,2.51,59.5l-97.2,65.359L127.214,285.748l108.1-62.01L195.09,197.761l-75.417,38.547L98.723,93.015,227.771,43.574,136.432,0,10.737,39.385,38.39,174.3l41.9,32.681L48.445,222.062,69.394,323.457,98.723,351.11,77.774,363.679l16.76,78.769L160.733,512c-10.8-74.842-11.658-78.641-11.725-78.773l77.925-55.3c16.759-12.57,15.083-10.894,15.083-10.894l.838,24.3,33.519,28.491-.838-77.093,46.927-33.519,26.815-18.435-2.514,36.033,25.139,17.6,6.7-74.579,58.657-43.575Z\"],goodreads:[448,512,[],\"f3a8\",\"M299.9 191.2c5.1 37.3-4.7 79-35.9 100.7-22.3 15.5-52.8 14.1-70.8 5.7-37.1-17.3-49.5-58.6-46.8-97.2 4.3-60.9 40.9-87.9 75.3-87.5 46.9-.2 71.8 31.8 78.2 78.3zM448 88v336c0 30.9-25.1 56-56 56H56c-30.9 0-56-25.1-56-56V88c0-30.9 25.1-56 56-56h336c30.9 0 56 25.1 56 56zM330 313.2s-.1-34-.1-217.3h-29v40.3c-.8.3-1.2-.5-1.6-1.2-9.6-20.7-35.9-46.3-76-46-51.9.4-87.2 31.2-100.6 77.8-4.3 14.9-5.8 30.1-5.5 45.6 1.7 77.9 45.1 117.8 112.4 115.2 28.9-1.1 54.5-17 69-45.2.5-1 1.1-1.9 1.7-2.9.2.1.4.1.6.2.3 3.8.2 30.7.1 34.5-.2 14.8-2 29.5-7.2 43.5-7.8 21-22.3 34.7-44.5 39.5-17.8 3.9-35.6 3.8-53.2-1.2-21.5-6.1-36.5-19-41.1-41.8-.3-1.6-1.3-1.3-2.3-1.3h-26.8c.8 10.6 3.2 20.3 8.5 29.2 24.2 40.5 82.7 48.5 128.2 37.4 49.9-12.3 67.3-54.9 67.4-106.3z\"],\"square-odnoklassniki\":[448,512,[\"odnoklassniki-square\"],\"f264\",\"M224 137.1a39.9 39.9 0 1 0 0 79.7 39.9 39.9 0 1 0 0-79.7zM384 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64zM224 95.9A81 81 0 1 1 224 258a81 81 0 1 1 0-162.1zm59.3 168.3c16.8-13.2 29.5-5.5 34.1 3.6c7.8 16-1.1 23.7-21.5 37c-17.1 10.9-40.7 15.2-56.2 16.8l13 12.9 47.7 47.7c17.4 17.9-11 45.8-28.6 28.6c-12-12.2-29.5-29.7-47.7-47.9l0 0-47.7 47.9c-17.7 17.2-46-11-28.4-28.6c3.7-3.7 7.9-7.9 12.5-12.5c10.4-10.4 22.6-22.7 35.2-35.2l12.9-12.9c-15.4-1.6-39.3-5.7-56.6-16.8c-20.3-13.3-29.3-20.9-21.4-37c4.6-9.1 17.3-16.8 34.1-3.6c0 0 22.7 18 59.3 18s59.3-18 59.3-18z\"],jsfiddle:[576,512,[],\"f1cc\",\"M510.634 237.462c-4.727-2.621-5.664-5.748-6.381-10.776-2.352-16.488-3.539-33.619-9.097-49.095-35.895-99.957-153.99-143.386-246.849-91.646-27.37 15.25-48.971 36.369-65.493 63.903-3.184-1.508-5.458-2.71-7.824-3.686-30.102-12.421-59.049-10.121-85.331 9.167-25.531 18.737-36.422 44.548-32.676 76.408.355 3.025-1.967 7.621-4.514 9.545-39.712 29.992-56.031 78.065-41.902 124.615 13.831 45.569 57.514 79.796 105.608 81.433 30.291 1.031 60.637.546 90.959.539 84.041-.021 168.09.531 252.12-.48 52.664-.634 96.108-36.873 108.212-87.293 11.54-48.074-11.144-97.3-56.832-122.634zm21.107 156.88c-18.23 22.432-42.343 35.253-71.28 35.65-56.874.781-113.767.23-170.652.23 0 .7-163.028.159-163.728.154-43.861-.332-76.739-19.766-95.175-59.995-18.902-41.245-4.004-90.848 34.186-116.106 9.182-6.073 12.505-11.566 10.096-23.136-5.49-26.361 4.453-47.956 26.42-62.981 22.987-15.723 47.422-16.146 72.034-3.083 10.269 5.45 14.607 11.564 22.198-2.527 14.222-26.399 34.557-46.727 60.671-61.294 97.46-54.366 228.37 7.568 230.24 132.697.122 8.15 2.412 12.428 9.848 15.894 57.56 26.829 74.456 96.122 35.142 144.497zm-87.789-80.499c-5.848 31.157-34.622 55.096-66.666 55.095-16.953-.001-32.058-6.545-44.079-17.705-27.697-25.713-71.141-74.98-95.937-93.387-20.056-14.888-41.99-12.333-60.272 3.782-49.996 44.071 15.859 121.775 67.063 77.188 4.548-3.96 7.84-9.543 12.744-12.844 8.184-5.509 20.766-.884 13.168 10.622-17.358 26.284-49.33 38.197-78.863 29.301-28.897-8.704-48.84-35.968-48.626-70.179 1.225-22.485 12.364-43.06 35.414-55.965 22.575-12.638 46.369-13.146 66.991 2.474C295.68 280.7 320.467 323.97 352.185 343.47c24.558 15.099 54.254 7.363 68.823-17.506 28.83-49.209-34.592-105.016-78.868-63.46-3.989 3.744-6.917 8.932-11.41 11.72-10.975 6.811-17.333-4.113-12.809-10.353 20.703-28.554 50.464-40.44 83.271-28.214 31.429 11.714 49.108 44.366 42.76 78.186z\"],sith:[448,512,[],\"f512\",\"M0 32l69.71 118.75-58.86-11.52 69.84 91.03a146.741 146.741 0 0 0 0 51.45l-69.84 91.03 58.86-11.52L0 480l118.75-69.71-11.52 58.86 91.03-69.84c17.02 3.04 34.47 3.04 51.48 0l91.03 69.84-11.52-58.86L448 480l-69.71-118.78 58.86 11.52-69.84-91.03c3.03-17.01 3.04-34.44 0-51.45l69.84-91.03-58.86 11.52L448 32l-118.75 69.71 11.52-58.9-91.06 69.87c-8.5-1.52-17.1-2.29-25.71-2.29s-17.21.78-25.71 2.29l-91.06-69.87 11.52 58.9L0 32zm224 99.78c31.8 0 63.6 12.12 87.85 36.37 48.5 48.5 48.49 127.21 0 175.7s-127.2 48.46-175.7-.03c-48.5-48.5-48.49-127.21 0-175.7 24.24-24.25 56.05-36.34 87.85-36.34zm0 36.66c-22.42 0-44.83 8.52-61.92 25.61-34.18 34.18-34.19 89.68 0 123.87s89.65 34.18 123.84 0c34.18-34.18 34.19-89.68 0-123.87-17.09-17.09-39.5-25.61-61.92-25.61z\"],themeisle:[512,512,[],\"f2b2\",\"M208 88.286c0-10 6.286-21.714 17.715-21.714 11.142 0 17.714 11.714 17.714 21.714 0 10.285-6.572 21.714-17.714 21.714C214.286 110 208 98.571 208 88.286zm304 160c0 36.001-11.429 102.286-36.286 129.714-22.858 24.858-87.428 61.143-120.857 70.572l-1.143.286v32.571c0 16.286-12.572 30.571-29.143 30.571-10 0-19.429-5.714-24.572-14.286-5.427 8.572-14.856 14.286-24.856 14.286-10 0-19.429-5.714-24.858-14.286-5.142 8.572-14.571 14.286-24.57 14.286-10.286 0-19.429-5.714-24.858-14.286-5.143 8.572-14.571 14.286-24.571 14.286-18.857 0-29.429-15.714-29.429-32.857-16.286 12.285-35.715 19.428-56.571 19.428-22 0-43.429-8.285-60.286-22.857 10.285-.286 20.571-2.286 30.285-5.714-20.857-5.714-39.428-18.857-52-36.286 21.37 4.645 46.209 1.673 67.143-11.143-22-22-56.571-58.857-68.572-87.428C1.143 321.714 0 303.714 0 289.429c0-49.714 20.286-160 86.286-160 10.571 0 18.857 4.858 23.143 14.857a158.792 158.792 0 0 1 12-15.428c2-2.572 5.714-5.429 7.143-8.286 7.999-12.571 11.714-21.142 21.714-34C182.571 45.428 232 17.143 285.143 17.143c6 0 12 .285 17.714 1.143C313.714 6.571 328.857 0 344.572 0c14.571 0 29.714 6 40 16.286.857.858 1.428 2.286 1.428 3.428 0 3.714-10.285 13.429-12.857 16.286 4.286 1.429 15.714 6.858 15.714 12 0 2.857-2.857 5.143-4.571 7.143 31.429 27.714 49.429 67.143 56.286 108 4.286-5.143 10.285-8.572 17.143-8.572 10.571 0 20.857 7.144 28.571 14.001C507.143 187.143 512 221.714 512 248.286zM188 89.428c0 18.286 12.571 37.143 32.286 37.143 19.714 0 32.285-18.857 32.285-37.143 0-18-12.571-36.857-32.285-36.857-19.715 0-32.286 18.858-32.286 36.857zM237.714 194c0-19.714 3.714-39.143 8.571-58.286-52.039 79.534-13.531 184.571 68.858 184.571 21.428 0 42.571-7.714 60-20 2-7.429 3.714-14.857 3.714-22.572 0-14.286-6.286-21.428-20.572-21.428-4.571 0-9.143.857-13.429 1.714-63.343 12.668-107.142 3.669-107.142-63.999zm-41.142 254.858c0-11.143-8.858-20.857-20.286-20.857-11.429 0-20 9.715-20 20.857v32.571c0 11.143 8.571 21.142 20 21.142 11.428 0 20.286-9.715 20.286-21.142v-32.571zm49.143 0c0-11.143-8.572-20.857-20-20.857-11.429 0-20.286 9.715-20.286 20.857v32.571c0 11.143 8.857 21.142 20.286 21.142 11.428 0 20-10 20-21.142v-32.571zm49.713 0c0-11.143-8.857-20.857-20.285-20.857-11.429 0-20.286 9.715-20.286 20.857v32.571c0 11.143 8.857 21.142 20.286 21.142 11.428 0 20.285-9.715 20.285-21.142v-32.571zm49.715 0c0-11.143-8.857-20.857-20.286-20.857-11.428 0-20.286 9.715-20.286 20.857v32.571c0 11.143 8.858 21.142 20.286 21.142 11.429 0 20.286-10 20.286-21.142v-32.571zM421.714 286c-30.857 59.142-90.285 102.572-158.571 102.572-96.571 0-160.571-84.572-160.571-176.572 0-16.857 2-33.429 6-49.714-20 33.715-29.714 72.572-29.714 111.429 0 60.286 24.857 121.715 71.429 160.857 5.143-9.714 14.857-16.286 26-16.286 10 0 19.428 5.714 24.571 14.286 5.429-8.571 14.571-14.286 24.858-14.286 10 0 19.428 5.714 24.571 14.286 5.429-8.571 14.857-14.286 24.858-14.286 10 0 19.428 5.714 24.857 14.286 5.143-8.571 14.571-14.286 24.572-14.286 10.857 0 20.857 6.572 25.714 16 43.427-36.286 68.569-92 71.426-148.286zm10.572-99.714c0-53.714-34.571-105.714-92.572-105.714-30.285 0-58.571 15.143-78.857 36.857C240.862 183.812 233.41 254 302.286 254c28.805 0 97.357-28.538 84.286 36.857 28.857-26 45.714-65.714 45.714-104.571z\"],page4:[496,512,[],\"f3d7\",\"M248 504C111 504 0 393 0 256S111 8 248 8c20.9 0 41.3 2.6 60.7 7.5L42.3 392H248v112zm0-143.6V146.8L98.6 360.4H248zm96 31.6v92.7c45.7-19.2 84.5-51.7 111.4-92.7H344zm57.4-138.2l-21.2 8.4 21.2 8.3v-16.7zm-20.3 54.5c-6.7 0-8 6.3-8 12.9v7.7h16.2v-10c0-5.9-2.3-10.6-8.2-10.6zM496 256c0 37.3-8.2 72.7-23 104.4H344V27.3C433.3 64.8 496 153.1 496 256zM360.4 143.6h68.2V96h-13.9v32.6h-13.9V99h-13.9v29.6h-12.7V96h-13.9v47.6zm68.1 185.3H402v-11c0-15.4-5.6-25.2-20.9-25.2-15.4 0-20.7 10.6-20.7 25.9v25.3h68.2v-15zm0-103l-68.2 29.7V268l68.2 29.5v-16.6l-14.4-5.7v-26.5l14.4-5.9v-16.9zm-4.8-68.5h-35.6V184H402v-12.2h11c8.6 15.8 1.3 35.3-18.6 35.3-22.5 0-28.3-25.3-15.5-37.7l-11.6-10.6c-16.2 17.5-12.2 63.9 27.1 63.9 34 0 44.7-35.9 29.3-65.3z\"],hashnode:[512,512,[],\"e499\",\"M35.19 171.1C-11.72 217.1-11.72 294 35.19 340.9L171.1 476.8C217.1 523.7 294 523.7 340.9 476.8L476.8 340.9C523.7 294 523.7 217.1 476.8 171.1L340.9 35.19C294-11.72 217.1-11.72 171.1 35.19L35.19 171.1zM315.5 315.5C282.6 348.3 229.4 348.3 196.6 315.5C163.7 282.6 163.7 229.4 196.6 196.6C229.4 163.7 282.6 163.7 315.5 196.6C348.3 229.4 348.3 282.6 315.5 315.5z\"],react:[512,512,[],\"f41b\",\"M418.2 177.2c-5.4-1.8-10.8-3.5-16.2-5.1.9-3.7 1.7-7.4 2.5-11.1 12.3-59.6 4.2-107.5-23.1-123.3-26.3-15.1-69.2.6-112.6 38.4-4.3 3.7-8.5 7.6-12.5 11.5-2.7-2.6-5.5-5.2-8.3-7.7-45.5-40.4-91.1-57.4-118.4-41.5-26.2 15.2-34 60.3-23 116.7 1.1 5.6 2.3 11.1 3.7 16.7-6.4 1.8-12.7 3.8-18.6 5.9C38.3 196.2 0 225.4 0 255.6c0 31.2 40.8 62.5 96.3 81.5 4.5 1.5 9 3 13.6 4.3-1.5 6-2.8 11.9-4 18-10.5 55.5-2.3 99.5 23.9 114.6 27 15.6 72.4-.4 116.6-39.1 3.5-3.1 7-6.3 10.5-9.7 4.4 4.3 9 8.4 13.6 12.4 42.8 36.8 85.1 51.7 111.2 36.6 27-15.6 35.8-62.9 24.4-120.5-.9-4.4-1.9-8.9-3-13.5 3.2-.9 6.3-1.9 9.4-2.9 57.7-19.1 99.5-50 99.5-81.7 0-30.3-39.4-59.7-93.8-78.4zM282.9 92.3c37.2-32.4 71.9-45.1 87.7-36 16.9 9.7 23.4 48.9 12.8 100.4-.7 3.4-1.4 6.7-2.3 10-22.2-5-44.7-8.6-67.3-10.6-13-18.6-27.2-36.4-42.6-53.1 3.9-3.7 7.7-7.2 11.7-10.7zM167.2 307.5c5.1 8.7 10.3 17.4 15.8 25.9-15.6-1.7-31.1-4.2-46.4-7.5 4.4-14.4 9.9-29.3 16.3-44.5 4.6 8.8 9.3 17.5 14.3 26.1zm-30.3-120.3c14.4-3.2 29.7-5.8 45.6-7.8-5.3 8.3-10.5 16.8-15.4 25.4-4.9 8.5-9.7 17.2-14.2 26-6.3-14.9-11.6-29.5-16-43.6zm27.4 68.9c6.6-13.8 13.8-27.3 21.4-40.6s15.8-26.2 24.4-38.9c15-1.1 30.3-1.7 45.9-1.7s31 .6 45.9 1.7c8.5 12.6 16.6 25.5 24.3 38.7s14.9 26.7 21.7 40.4c-6.7 13.8-13.9 27.4-21.6 40.8-7.6 13.3-15.7 26.2-24.2 39-14.9 1.1-30.4 1.6-46.1 1.6s-30.9-.5-45.6-1.4c-8.7-12.7-16.9-25.7-24.6-39s-14.8-26.8-21.5-40.6zm180.6 51.2c5.1-8.8 9.9-17.7 14.6-26.7 6.4 14.5 12 29.2 16.9 44.3-15.5 3.5-31.2 6.2-47 8 5.4-8.4 10.5-17 15.5-25.6zm14.4-76.5c-4.7-8.8-9.5-17.6-14.5-26.2-4.9-8.5-10-16.9-15.3-25.2 16.1 2 31.5 4.7 45.9 8-4.6 14.8-10 29.2-16.1 43.4zM256.2 118.3c10.5 11.4 20.4 23.4 29.6 35.8-19.8-.9-39.7-.9-59.5 0 9.8-12.9 19.9-24.9 29.9-35.8zM140.2 57c16.8-9.8 54.1 4.2 93.4 39 2.5 2.2 5 4.6 7.6 7-15.5 16.7-29.8 34.5-42.9 53.1-22.6 2-45 5.5-67.2 10.4-1.3-5.1-2.4-10.3-3.5-15.5-9.4-48.4-3.2-84.9 12.6-94zm-24.5 263.6c-4.2-1.2-8.3-2.5-12.4-3.9-21.3-6.7-45.5-17.3-63-31.2-10.1-7-16.9-17.8-18.8-29.9 0-18.3 31.6-41.7 77.2-57.6 5.7-2 11.5-3.8 17.3-5.5 6.8 21.7 15 43 24.5 63.6-9.6 20.9-17.9 42.5-24.8 64.5zm116.6 98c-16.5 15.1-35.6 27.1-56.4 35.3-11.1 5.3-23.9 5.8-35.3 1.3-15.9-9.2-22.5-44.5-13.5-92 1.1-5.6 2.3-11.2 3.7-16.7 22.4 4.8 45 8.1 67.9 9.8 13.2 18.7 27.7 36.6 43.2 53.4-3.2 3.1-6.4 6.1-9.6 8.9zm24.5-24.3c-10.2-11-20.4-23.2-30.3-36.3 9.6.4 19.5.6 29.5.6 10.3 0 20.4-.2 30.4-.7-9.2 12.7-19.1 24.8-29.6 36.4zm130.7 30c-.9 12.2-6.9 23.6-16.5 31.3-15.9 9.2-49.8-2.8-86.4-34.2-4.2-3.6-8.4-7.5-12.7-11.5 15.3-16.9 29.4-34.8 42.2-53.6 22.9-1.9 45.7-5.4 68.2-10.5 1 4.1 1.9 8.2 2.7 12.2 4.9 21.6 5.7 44.1 2.5 66.3zm18.2-107.5c-2.8.9-5.6 1.8-8.5 2.6-7-21.8-15.6-43.1-25.5-63.8 9.6-20.4 17.7-41.4 24.5-62.9 5.2 1.5 10.2 3.1 15 4.7 46.6 16 79.3 39.8 79.3 58 0 19.6-34.9 44.9-84.8 61.4zm-149.7-15c25.3 0 45.8-20.5 45.8-45.8s-20.5-45.8-45.8-45.8c-25.3 0-45.8 20.5-45.8 45.8s20.5 45.8 45.8 45.8z\"],\"cc-paypal\":[576,512,[],\"f1f4\",\"M186.3 258.2c0 12.2-9.7 21.5-22 21.5-9.2 0-16-5.2-16-15 0-12.2 9.5-22 21.7-22 9.3 0 16.3 5.7 16.3 15.5zM80.5 209.7h-4.7c-1.5 0-3 1-3.2 2.7l-4.3 26.7 8.2-.3c11 0 19.5-1.5 21.5-14.2 2.3-13.4-6.2-14.9-17.5-14.9zm284 0H360c-1.8 0-3 1-3.2 2.7l-4.2 26.7 8-.3c13 0 22-3 22-18-.1-10.6-9.6-11.1-18.1-11.1zM576 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h480c26.5 0 48 21.5 48 48zM128.3 215.4c0-21-16.2-28-34.7-28h-40c-2.5 0-5 2-5.2 4.7L32 294.2c-.3 2 1.2 4 3.2 4h19c2.7 0 5.2-2.9 5.5-5.7l4.5-26.6c1-7.2 13.2-4.7 18-4.7 28.6 0 46.1-17 46.1-45.8zm84.2 8.8h-19c-3.8 0-4 5.5-4.2 8.2-5.8-8.5-14.2-10-23.7-10-24.5 0-43.2 21.5-43.2 45.2 0 19.5 12.2 32.2 31.7 32.2 9 0 20.2-4.9 26.5-11.9-.5 1.5-1 4.7-1 6.2 0 2.3 1 4 3.2 4H200c2.7 0 5-2.9 5.5-5.7l10.2-64.3c.3-1.9-1.2-3.9-3.2-3.9zm40.5 97.9l63.7-92.6c.5-.5.5-1 .5-1.7 0-1.7-1.5-3.5-3.2-3.5h-19.2c-1.7 0-3.5 1-4.5 2.5l-26.5 39-11-37.5c-.8-2.2-3-4-5.5-4h-18.7c-1.7 0-3.2 1.8-3.2 3.5 0 1.2 19.5 56.8 21.2 62.1-2.7 3.8-20.5 28.6-20.5 31.6 0 1.8 1.5 3.2 3.2 3.2h19.2c1.8-.1 3.5-1.1 4.5-2.6zm159.3-106.7c0-21-16.2-28-34.7-28h-39.7c-2.7 0-5.2 2-5.5 4.7l-16.2 102c-.2 2 1.3 4 3.2 4h20.5c2 0 3.5-1.5 4-3.2l4.5-29c1-7.2 13.2-4.7 18-4.7 28.4 0 45.9-17 45.9-45.8zm84.2 8.8h-19c-3.8 0-4 5.5-4.3 8.2-5.5-8.5-14-10-23.7-10-24.5 0-43.2 21.5-43.2 45.2 0 19.5 12.2 32.2 31.7 32.2 9.3 0 20.5-4.9 26.5-11.9-.3 1.5-1 4.7-1 6.2 0 2.3 1 4 3.2 4H484c2.7 0 5-2.9 5.5-5.7l10.2-64.3c.3-1.9-1.2-3.9-3.2-3.9zm47.5-33.3c0-2-1.5-3.5-3.2-3.5h-18.5c-1.5 0-3 1.2-3.2 2.7l-16.2 104-.3.5c0 1.8 1.5 3.5 3.5 3.5h16.5c2.5 0 5-2.9 5.2-5.7L544 191.2v-.3zm-90 51.8c-12.2 0-21.7 9.7-21.7 22 0 9.7 7 15 16.2 15 12 0 21.7-9.2 21.7-21.5.1-9.8-6.9-15.5-16.2-15.5z\"],squarespace:[512,512,[],\"f5be\",\"M186.12 343.34c-9.65 9.65-9.65 25.29 0 34.94 9.65 9.65 25.29 9.65 34.94 0L378.24 221.1c19.29-19.29 50.57-19.29 69.86 0s19.29 50.57 0 69.86L293.95 445.1c19.27 19.29 50.53 19.31 69.82.04l.04-.04 119.25-119.24c38.59-38.59 38.59-101.14 0-139.72-38.59-38.59-101.15-38.59-139.72 0l-157.22 157.2zm244.53-104.8c-9.65-9.65-25.29-9.65-34.93 0l-157.2 157.18c-19.27 19.29-50.53 19.31-69.82.05l-.05-.05c-9.64-9.64-25.27-9.65-34.92-.01l-.01.01c-9.65 9.64-9.66 25.28-.02 34.93l.02.02c38.58 38.57 101.14 38.57 139.72 0l157.2-157.2c9.65-9.65 9.65-25.29.01-34.93zm-261.99 87.33l157.18-157.18c9.64-9.65 9.64-25.29 0-34.94-9.64-9.64-25.27-9.64-34.91 0L133.72 290.93c-19.28 19.29-50.56 19.3-69.85.01l-.01-.01c-19.29-19.28-19.31-50.54-.03-69.84l.03-.03L218.03 66.89c-19.28-19.29-50.55-19.3-69.85-.02l-.02.02L28.93 186.14c-38.58 38.59-38.58 101.14 0 139.72 38.6 38.59 101.13 38.59 139.73.01zm-87.33-52.4c9.64 9.64 25.27 9.64 34.91 0l157.21-157.19c19.28-19.29 50.55-19.3 69.84-.02l.02.02c9.65 9.65 25.29 9.65 34.93 0 9.65-9.65 9.65-25.29 0-34.93-38.59-38.59-101.13-38.59-139.72 0L81.33 238.54c-9.65 9.64-9.65 25.28-.01 34.93h.01z\"],\"cc-stripe\":[576,512,[],\"f1f5\",\"M492.4 220.8c-8.9 0-18.7 6.7-18.7 22.7h36.7c0-16-9.3-22.7-18-22.7zM375 223.4c-8.2 0-13.3 2.9-17 7l.2 52.8c3.5 3.7 8.5 6.7 16.8 6.7 13.1 0 21.9-14.3 21.9-33.4 0-18.6-9-33.2-21.9-33.1zM528 32H48C21.5 32 0 53.5 0 80v352c0 26.5 21.5 48 48 48h480c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48zM122.2 281.1c0 25.6-20.3 40.1-49.9 40.3-12.2 0-25.6-2.4-38.8-8.1v-33.9c12 6.4 27.1 11.3 38.9 11.3 7.9 0 13.6-2.1 13.6-8.7 0-17-54-10.6-54-49.9 0-25.2 19.2-40.2 48-40.2 11.8 0 23.5 1.8 35.3 6.5v33.4c-10.8-5.8-24.5-9.1-35.3-9.1-7.5 0-12.1 2.2-12.1 7.7 0 16 54.3 8.4 54.3 50.7zm68.8-56.6h-27V275c0 20.9 22.5 14.4 27 12.6v28.9c-4.7 2.6-13.3 4.7-24.9 4.7-21.1 0-36.9-15.5-36.9-36.5l.2-113.9 34.7-7.4v30.8H191zm74 2.4c-4.5-1.5-18.7-3.6-27.1 7.4v84.4h-35.5V194.2h30.7l2.2 10.5c8.3-15.3 24.9-12.2 29.6-10.5h.1zm44.1 91.8h-35.7V194.2h35.7zm0-142.9l-35.7 7.6v-28.9l35.7-7.6zm74.1 145.5c-12.4 0-20-5.3-25.1-9l-.1 40.2-35.5 7.5V194.2h31.3l1.8 8.8c4.9-4.5 13.9-11.1 27.8-11.1 24.9 0 48.4 22.5 48.4 63.8 0 45.1-23.2 65.5-48.6 65.6zm160.4-51.5h-69.5c1.6 16.6 13.8 21.5 27.6 21.5 14.1 0 25.2-3 34.9-7.9V312c-9.7 5.3-22.4 9.2-39.4 9.2-34.6 0-58.8-21.7-58.8-64.5 0-36.2 20.5-64.9 54.3-64.9 33.7 0 51.3 28.7 51.3 65.1 0 3.5-.3 10.9-.4 12.9z\"],\"creative-commons-share\":[496,512,[],\"f4f2\",\"M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zm101 132.4c7.8 0 13.7 6.1 13.7 13.7v182.5c0 7.7-6.1 13.7-13.7 13.7H214.3c-7.7 0-13.7-6-13.7-13.7v-54h-54c-7.8 0-13.7-6-13.7-13.7V131.1c0-8.2 6.6-12.7 12.4-13.7h136.4c7.7 0 13.7 6 13.7 13.7v54h54zM159.9 300.3h40.7V198.9c0-7.4 5.8-12.6 12-13.7h55.8v-40.3H159.9v155.4zm176.2-88.1H227.6v155.4h108.5V212.2z\"],bitcoin:[512,512,[],\"f379\",\"M504 256c0 136.967-111.033 248-248 248S8 392.967 8 256 119.033 8 256 8s248 111.033 248 248zm-141.651-35.33c4.937-32.999-20.191-50.739-54.55-62.573l11.146-44.702-27.213-6.781-10.851 43.524c-7.154-1.783-14.502-3.464-21.803-5.13l10.929-43.81-27.198-6.781-11.153 44.686c-5.922-1.349-11.735-2.682-17.377-4.084l.031-.14-37.53-9.37-7.239 29.062s20.191 4.627 19.765 4.913c11.022 2.751 13.014 10.044 12.68 15.825l-12.696 50.925c.76.194 1.744.473 2.829.907-.907-.225-1.876-.473-2.876-.713l-17.796 71.338c-1.349 3.348-4.767 8.37-12.471 6.464.271.395-19.78-4.937-19.78-4.937l-13.51 31.147 35.414 8.827c6.588 1.651 13.045 3.379 19.4 5.006l-11.262 45.213 27.182 6.781 11.153-44.733a1038.209 1038.209 0 0 0 21.687 5.627l-11.115 44.523 27.213 6.781 11.262-45.128c46.404 8.781 81.299 5.239 95.986-36.727 11.836-33.79-.589-53.281-25.004-65.991 17.78-4.098 31.174-15.792 34.747-39.949zm-62.177 87.179c-8.41 33.79-65.308 15.523-83.755 10.943l14.944-59.899c18.446 4.603 77.6 13.717 68.811 48.956zm8.417-87.667c-7.673 30.736-55.031 15.12-70.393 11.292l13.548-54.327c15.363 3.828 64.836 10.973 56.845 43.035z\"],keycdn:[512,512,[],\"f3ba\",\"M63.8 409.3l60.5-59c32.1 42.8 71.1 66 126.6 67.4 30.5.7 60.3-7 86.4-22.4 5.1 5.3 18.5 19.5 20.9 22-32.2 20.7-69.6 31.1-108.1 30.2-43.3-1.1-84.6-16.7-117.7-44.4.3-.6-38.2 37.5-38.6 37.9 9.5 29.8-13.1 62.4-46.3 62.4C20.7 503.3 0 481.7 0 454.9c0-34.3 33.1-56.6 63.8-45.6zm354.9-252.4c19.1 31.3 29.6 67.4 28.7 104-1.1 44.8-19 87.5-48.6 121 .3.3 23.8 25.2 24.1 25.5 9.6-1.3 19.2 2 25.9 9.1 11.3 12 10.9 30.9-1.1 42.4-12 11.3-30.9 10.9-42.4-1.1-6.7-7-9.4-16.8-7.6-26.3-24.9-26.6-44.4-47.2-44.4-47.2 42.7-34.1 63.3-79.6 64.4-124.2.7-28.9-7.2-57.2-21.1-82.2l22.1-21zM104 53.1c6.7 7 9.4 16.8 7.6 26.3l45.9 48.1c-4.7 3.8-13.3 10.4-22.8 21.3-25.4 28.5-39.6 64.8-40.7 102.9-.7 28.9 6.1 57.2 20 82.4l-22 21.5C72.7 324 63.1 287.9 64.2 250.9c1-44.6 18.3-87.6 47.5-121.1l-25.3-26.4c-9.6 1.3-19.2-2-25.9-9.1-11.3-12-10.9-30.9 1.1-42.4C73.5 40.7 92.2 41 104 53.1zM464.9 8c26 0 47.1 22.4 47.1 48.3S490.9 104 464.9 104c-6.3.1-14-1.1-15.9-1.8l-62.9 59.7c-32.7-43.6-76.7-65.9-126.9-67.2-30.5-.7-60.3 6.8-86.2 22.4l-21.1-22C184.1 74.3 221.5 64 260 64.9c43.3 1.1 84.6 16.7 117.7 44.6l41.1-38.6c-1.5-4.7-2.2-9.6-2.2-14.5C416.5 29.7 438.9 8 464.9 8zM256.7 113.4c5.5 0 10.9.4 16.4 1.1 78.1 9.8 133.4 81.1 123.8 159.1-9.8 78.1-81.1 133.4-159.1 123.8-78.1-9.8-133.4-81.1-123.8-159.2 9.3-72.4 70.1-124.6 142.7-124.8zm-59 119.4c.6 22.7 12.2 41.8 32.4 52.2l-11 51.7h73.7l-11-51.7c20.1-10.9 32.1-29 32.4-52.2-.4-32.8-25.8-57.5-58.3-58.3-32.1.8-57.3 24.8-58.2 58.3zM256 160\"],opera:[496,512,[],\"f26a\",\"M313.9 32.7c-170.2 0-252.6 223.8-147.5 355.1 36.5 45.4 88.6 75.6 147.5 75.6 36.3 0 70.3-11.1 99.4-30.4-43.8 39.2-101.9 63-165.3 63-3.9 0-8 0-11.9-.3C104.6 489.6 0 381.1 0 248 0 111 111 0 248 0h.8c63.1.3 120.7 24.1 164.4 63.1-29-19.4-63.1-30.4-99.3-30.4zm101.8 397.7c-40.9 24.7-90.7 23.6-132-5.8 56.2-20.5 97.7-91.6 97.7-176.6 0-84.7-41.2-155.8-97.4-176.6 41.8-29.2 91.2-30.3 132.9-5 105.9 98.7 105.5 265.7-1.2 364z\"],\"itch-io\":[512,512,[],\"f83a\",\"M71.92 34.77C50.2 47.67 7.4 96.84 7 109.73v21.34c0 27.06 25.29 50.84 48.25 50.84 27.57 0 50.54-22.85 50.54-50 0 27.12 22.18 50 49.76 50s49-22.85 49-50c0 27.12 23.59 50 51.16 50h.5c27.57 0 51.16-22.85 51.16-50 0 27.12 21.47 50 49 50s49.76-22.85 49.76-50c0 27.12 23 50 50.54 50 23 0 48.25-23.78 48.25-50.84v-21.34c-.4-12.9-43.2-62.07-64.92-75C372.56 32.4 325.76 32 256 32S91.14 33.1 71.92 34.77zm132.32 134.39c-22 38.4-77.9 38.71-99.85.25-13.17 23.14-43.17 32.07-56 27.66-3.87 40.15-13.67 237.13 17.73 269.15 80 18.67 302.08 18.12 379.76 0 31.65-32.27 21.32-232 17.75-269.15-12.92 4.44-42.88-4.6-56-27.66-22 38.52-77.85 38.1-99.85-.24-7.1 12.49-23.05 28.94-51.76 28.94a57.54 57.54 0 0 1-51.75-28.94zm-41.58 53.77c16.47 0 31.09 0 49.22 19.78a436.91 436.91 0 0 1 88.18 0C318.22 223 332.85 223 349.31 223c52.33 0 65.22 77.53 83.87 144.45 17.26 62.15-5.52 63.67-33.95 63.73-42.15-1.57-65.49-32.18-65.49-62.79-39.25 6.43-101.93 8.79-155.55 0 0 30.61-23.34 61.22-65.49 62.79-28.42-.06-51.2-1.58-33.94-63.73 18.67-67 31.56-144.45 83.88-144.45zM256 270.79s-44.38 40.77-52.35 55.21l29-1.17v25.32c0 1.55 21.34.16 23.33.16 11.65.54 23.31 1 23.31-.16v-25.28l29 1.17c-8-14.48-52.35-55.24-52.35-55.24z\"],umbraco:[510,512,[],\"f8e8\",\"M255.35 8C118.36 7.83 7.14 118.72 7 255.68c-.07 137 111 248.2 248 248.27 136.85 0 247.82-110.7 248-247.67S392.34 8.17 255.35 8zm145 266q-1.14 40.68-14 65t-43.51 35q-30.61 10.7-85.45 10.47h-4.6q-54.78.22-85.44-10.47t-43.52-35q-12.85-24.36-14-65a224.81 224.81 0 0 1 0-30.71 418.37 418.37 0 0 1 3.6-43.88c1.88-13.39 3.57-22.58 5.4-32 1-4.88 1.28-6.42 1.82-8.45a5.09 5.09 0 0 1 4.9-3.89h.69l32 5a5.07 5.07 0 0 1 4.16 5 5 5 0 0 1 0 .77l-1.7 8.78q-2.41 13.25-4.84 33.68a380.62 380.62 0 0 0-2.64 42.15q-.28 40.43 8.13 59.83a43.87 43.87 0 0 0 31.31 25.18A243 243 0 0 0 250 340.6h10.25a242.64 242.64 0 0 0 57.27-5.16 43.86 43.86 0 0 0 31.15-25.23q8.53-19.42 8.13-59.78a388 388 0 0 0-2.6-42.15q-2.48-20.38-4.89-33.68l-1.69-8.78a5 5 0 0 1 0-.77 5 5 0 0 1 4.2-5l32-5h.82a5 5 0 0 1 4.9 3.89c.55 2.05.81 3.57 1.83 8.45 1.82 9.62 3.52 18.78 5.39 32a415.71 415.71 0 0 1 3.61 43.88 228.06 228.06 0 0 1-.04 30.73z\"],\"galactic-senate\":[512,512,[],\"f50d\",\"M249.86 33.48v26.07C236.28 80.17 226 168.14 225.39 274.9c11.74-15.62 19.13-33.33 19.13-48.24v-16.88c-.03-5.32.75-10.53 2.19-15.65.65-2.14 1.39-4.08 2.62-5.82 1.23-1.75 3.43-3.79 6.68-3.79 3.24 0 5.45 2.05 6.68 3.79 1.23 1.75 1.97 3.68 2.62 5.82 1.44 5.12 2.22 10.33 2.19 15.65v16.88c0 14.91 7.39 32.62 19.13 48.24-.63-106.76-10.91-194.73-24.49-215.35V33.48h-12.28zm-26.34 147.77c-9.52 2.15-18.7 5.19-27.46 9.08 8.9 16.12 9.76 32.64 1.71 37.29-8 4.62-21.85-4.23-31.36-19.82-11.58 8.79-21.88 19.32-30.56 31.09 14.73 9.62 22.89 22.92 18.32 30.66-4.54 7.7-20.03 7.14-35.47-.96-5.78 13.25-9.75 27.51-11.65 42.42 9.68.18 18.67 2.38 26.18 6.04 17.78-.3 32.77-1.96 40.49-4.22 5.55-26.35 23.02-48.23 46.32-59.51.73-25.55 1.88-49.67 3.48-72.07zm64.96 0c1.59 22.4 2.75 46.52 3.47 72.07 23.29 11.28 40.77 33.16 46.32 59.51 7.72 2.26 22.71 3.92 40.49 4.22 7.51-3.66 16.5-5.85 26.18-6.04-1.9-14.91-5.86-29.17-11.65-42.42-15.44 8.1-30.93 8.66-35.47.96-4.57-7.74 3.6-21.05 18.32-30.66-8.68-11.77-18.98-22.3-30.56-31.09-9.51 15.59-23.36 24.44-31.36 19.82-8.05-4.65-7.19-21.16 1.71-37.29a147.49 147.49 0 0 0-27.45-9.08zm-32.48 8.6c-3.23 0-5.86 8.81-6.09 19.93h-.05v16.88c0 41.42-49.01 95.04-93.49 95.04-52 0-122.75-1.45-156.37 29.17v2.51c9.42 17.12 20.58 33.17 33.18 47.97C45.7 380.26 84.77 360.4 141.2 360c45.68 1.02 79.03 20.33 90.76 40.87.01.01-.01.04 0 .05 7.67 2.14 15.85 3.23 24.04 3.21 8.19.02 16.37-1.07 24.04-3.21.01-.01-.01-.04 0-.05 11.74-20.54 45.08-39.85 90.76-40.87 56.43.39 95.49 20.26 108.02 41.35 12.6-14.8 23.76-30.86 33.18-47.97v-2.51c-33.61-30.62-104.37-29.17-156.37-29.17-44.48 0-93.49-53.62-93.49-95.04v-16.88h-.05c-.23-11.12-2.86-19.93-6.09-19.93zm0 96.59c22.42 0 40.6 18.18 40.6 40.6s-18.18 40.65-40.6 40.65-40.6-18.23-40.6-40.65c0-22.42 18.18-40.6 40.6-40.6zm0 7.64c-18.19 0-32.96 14.77-32.96 32.96S237.81 360 256 360s32.96-14.77 32.96-32.96-14.77-32.96-32.96-32.96zm0 6.14c14.81 0 26.82 12.01 26.82 26.82s-12.01 26.82-26.82 26.82-26.82-12.01-26.82-26.82 12.01-26.82 26.82-26.82zm-114.8 66.67c-10.19.07-21.6.36-30.5 1.66.43 4.42 1.51 18.63 7.11 29.76 9.11-2.56 18.36-3.9 27.62-3.9 41.28.94 71.48 34.35 78.26 74.47l.11 4.7c10.4 1.91 21.19 2.94 32.21 2.94 11.03 0 21.81-1.02 32.21-2.94l.11-4.7c6.78-40.12 36.98-73.53 78.26-74.47 9.26 0 18.51 1.34 27.62 3.9 5.6-11.13 6.68-25.34 7.11-29.76-8.9-1.3-20.32-1.58-30.5-1.66-18.76.42-35.19 4.17-48.61 9.67-12.54 16.03-29.16 30.03-49.58 33.07-.09.02-.17.04-.27.05-.05.01-.11.04-.16.05-5.24 1.07-10.63 1.6-16.19 1.6-5.55 0-10.95-.53-16.19-1.6-.05-.01-.11-.04-.16-.05-.1-.02-.17-.04-.27-.05-20.42-3.03-37.03-17.04-49.58-33.07-13.42-5.49-29.86-9.25-48.61-9.67z\"],ubuntu:[576,512,[],\"f7df\",\"M469.2 75A75.6 75.6 0 1 0 317.9 75a75.6 75.6 0 1 0 151.2 0zM154.2 240.7A75.6 75.6 0 1 0 3 240.7a75.6 75.6 0 1 0 151.2 0zM57 346C75.6 392.9 108 433 150 461.1s91.5 42.6 142 41.7c-14.7-18.6-22.9-41.5-23.2-65.2c-6.8-.9-13.3-2.1-19.5-3.4c-26.8-5.7-51.9-17.3-73.6-34s-39.3-38.1-51.7-62.5c-20.9 9.9-44.5 12.8-67.1 8.2zm395.1 89.8a75.6 75.6 0 1 0 -151.2 0 75.6 75.6 0 1 0 151.2 0zM444 351.6c18.5 14.8 31.6 35.2 37.2 58.2c33.3-41.3 52.6-92.2 54.8-145.2s-12.5-105.4-42.2-149.4c-8.6 21.5-24 39.6-43.8 51.6c15.4 28.6 22.9 60.8 21.9 93.2s-10.7 64-28 91.6zM101.1 135.4c12.4 2.7 24.3 7.5 35.1 14.3c16.6-24.2 38.9-44.1 64.8-58S255.8 70.4 285.2 70c.2-5.9 .9-11.9 2-17.7c3.6-16.7 11.1-32.3 21.8-45.5c-47.7-3.8-95.4 6-137.6 28.5S94.3 91.7 70.8 133.4c2.7-.2 5.3-.3 8-.3c7.5 0 15 .8 22.4 2.3z\"],draft2digital:[480,512,[],\"f396\",\"M480 398.1l-144-82.2v64.7h-91.3c30.8-35 81.8-95.9 111.8-149.3 35.2-62.6 16.1-123.4-12.8-153.3-4.4-4.6-62.2-62.9-166-41.2-59.1 12.4-89.4 43.4-104.3 67.3-13.1 20.9-17 39.8-18.2 47.7-5.5 33 19.4 67.1 56.7 67.1 31.7 0 57.3-25.7 57.3-57.4 0-27.1-19.7-52.1-48-56.8 1.8-7.3 17.7-21.1 26.3-24.7 41.1-17.3 78 5.2 83.3 33.5 8.3 44.3-37.1 90.4-69.7 127.6C84.5 328.1 18.3 396.8 0 415.9l336-.1V480zM369.9 371l47.1 27.2-47.1 27.2zM134.2 161.4c0 12.4-10 22.4-22.4 22.4s-22.4-10-22.4-22.4 10-22.4 22.4-22.4 22.4 10.1 22.4 22.4zM82.5 380.5c25.6-27.4 97.7-104.7 150.8-169.9 35.1-43.1 40.3-82.4 28.4-112.7-7.4-18.8-17.5-30.2-24.3-35.7 45.3 2.1 68 23.4 82.2 38.3 0 0 42.4 48.2 5.8 113.3-37 65.9-110.9 147.5-128.5 166.7z\"],stripe:[640,512,[],\"f429\",\"M165 144.7l-43.3 9.2-.2 142.4c0 26.3 19.8 43.3 46.1 43.3 14.6 0 25.3-2.7 31.2-5.9v-33.8c-5.7 2.3-33.7 10.5-33.7-15.7V221h33.7v-37.8h-33.7zm89.1 51.6l-2.7-13.1H213v153.2h44.3V233.3c10.5-13.8 28.2-11.1 33.9-9.3v-40.8c-6-2.1-26.7-6-37.1 13.1zm92.3-72.3l-44.6 9.5v36.2l44.6-9.5zM44.9 228.3c0-6.9 5.8-9.6 15.1-9.7 13.5 0 30.7 4.1 44.2 11.4v-41.8c-14.7-5.8-29.4-8.1-44.1-8.1-36 0-60 18.8-60 50.2 0 49.2 67.5 41.2 67.5 62.4 0 8.2-7.1 10.9-17 10.9-14.7 0-33.7-6.1-48.6-14.2v40c16.5 7.1 33.2 10.1 48.5 10.1 36.9 0 62.3-15.8 62.3-47.8 0-52.9-67.9-43.4-67.9-63.4zM640 261.6c0-45.5-22-81.4-64.2-81.4s-67.9 35.9-67.9 81.1c0 53.5 30.3 78.2 73.5 78.2 21.2 0 37.1-4.8 49.2-11.5v-33.4c-12.1 6.1-26 9.8-43.6 9.8-17.3 0-32.5-6.1-34.5-26.9h86.9c.2-2.3.6-11.6.6-15.9zm-87.9-16.8c0-20 12.3-28.4 23.4-28.4 10.9 0 22.5 8.4 22.5 28.4zm-112.9-64.6c-17.4 0-28.6 8.2-34.8 13.9l-2.3-11H363v204.8l44.4-9.4.1-50.2c6.4 4.7 15.9 11.2 31.4 11.2 31.8 0 60.8-23.2 60.8-79.6.1-51.6-29.3-79.7-60.5-79.7zm-10.6 122.5c-10.4 0-16.6-3.8-20.9-8.4l-.3-66c4.6-5.1 11-8.8 21.2-8.8 16.2 0 27.4 18.2 27.4 41.4.1 23.9-10.9 41.8-27.4 41.8zm-126.7 33.7h44.6V183.2h-44.6z\"],houzz:[448,512,[],\"f27c\",\"M275.9 330.7H171.3V480H17V32h109.5v104.5l305.1 85.6V480H275.9z\"],gg:[512,512,[],\"f260\",\"M179.2 230.4l102.4 102.4-102.4 102.4L0 256 179.2 76.8l44.8 44.8-25.6 25.6-19.2-19.2-128 128 128 128 51.5-51.5-77.1-76.5 25.6-25.6zM332.8 76.8L230.4 179.2l102.4 102.4 25.6-25.6-77.1-76.5 51.5-51.5 128 128-128 128-19.2-19.2-25.6 25.6 44.8 44.8L512 256 332.8 76.8z\"],dhl:[640,512,[],\"f790\",\"M238 301.2h58.7L319 271h-58.7L238 301.2zM0 282.9v6.4h81.8l4.7-6.4H0zM172.9 271c-8.7 0-6-3.6-4.6-5.5 2.8-3.8 7.6-10.4 10.4-14.1 2.8-3.7 2.8-5.9-2.8-5.9h-51l-41.1 55.8h100.1c33.1 0 51.5-22.5 57.2-30.3h-68.2zm317.5-6.9l39.3-53.4h-62.2l-39.3 53.4h62.2zM95.3 271H0v6.4h90.6l4.7-6.4zm111-26.6c-2.8 3.8-7.5 10.4-10.3 14.2-1.4 2-4.1 5.5 4.6 5.5h45.6s7.3-10 13.5-18.4c8.4-11.4.7-35-29.2-35H112.6l-20.4 27.8h111.4c5.6 0 5.5 2.2 2.7 5.9zM0 301.2h73.1l4.7-6.4H0v6.4zm323 0h58.7L404 271h-58.7c-.1 0-22.3 30.2-22.3 30.2zm222 .1h95v-6.4h-90.3l-4.7 6.4zm22.3-30.3l-4.7 6.4H640V271h-72.7zm-13.5 18.3H640v-6.4h-81.5l-4.7 6.4zm-164.2-78.6l-22.5 30.6h-26.2l22.5-30.6h-58.7l-39.3 53.4H409l39.3-53.4h-58.7zm33.5 60.3s-4.3 5.9-6.4 8.7c-7.4 10-.9 21.6 23.2 21.6h94.3l22.3-30.3H423.1z\"],\"square-pinterest\":[448,512,[\"pinterest-square\"],\"f0d3\",\"M384 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64h72.6l-2.2-.8c-5.4-48.1-3.1-57.5 15.7-134.7c3.9-16 8.5-35 13.9-57.9c0 0-7.3-14.8-7.3-36.5c0-70.7 75.5-78 75.5-25c0 13.5-5.4 31.1-11.2 49.8c-3.3 10.6-6.6 21.5-9.1 32c-5.7 24.5 12.3 44.4 36.4 44.4c43.7 0 77.2-46 77.2-112.4c0-58.8-42.3-99.9-102.6-99.9C153 139 112 191.4 112 245.6c0 21.1 8.2 43.7 18.3 56c2 2.4 2.3 4.5 1.7 7c-1.1 4.7-3.1 12.9-4.7 19.2c-1 4-1.8 7.3-2.1 8.6c-1.1 4.5-3.5 5.5-8.2 3.3c-30.6-14.3-49.8-59.1-49.8-95.1C67.2 167.1 123.4 96 229.4 96c85.2 0 151.4 60.7 151.4 141.8c0 84.6-53.3 152.7-127.4 152.7c-24.9 0-48.3-12.9-56.3-28.2c0 0-12.3 46.9-15.3 58.4c-5 19.3-17.6 42.9-27.4 59.3H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64z\"],xing:[384,512,[],\"f168\",\"M162.7 210c-1.8 3.3-25.2 44.4-70.1 123.5-4.9 8.3-10.8 12.5-17.7 12.5H9.8c-7.7 0-12.1-7.5-8.5-14.4l69-121.3c.2 0 .2-.1 0-.3l-43.9-75.6c-4.3-7.8.3-14.1 8.5-14.1H100c7.3 0 13.3 4.1 18 12.2l44.7 77.5zM382.6 46.1l-144 253v.3L330.2 466c3.9 7.1.2 14.1-8.5 14.1h-65.2c-7.6 0-13.6-4-18-12.2l-92.4-168.5c3.3-5.8 51.5-90.8 144.8-255.2 4.6-8.1 10.4-12.2 17.5-12.2h65.7c8 0 12.3 6.7 8.5 14.1z\"],blackberry:[512,512,[],\"f37b\",\"M166 116.9c0 23.4-16.4 49.1-72.5 49.1H23.4l21-88.8h67.8c42.1 0 53.8 23.3 53.8 39.7zm126.2-39.7h-67.8L205.7 166h70.1c53.8 0 70.1-25.7 70.1-49.1.1-16.4-11.6-39.7-53.7-39.7zM88.8 208.1H21L0 296.9h70.1c56.1 0 72.5-23.4 72.5-49.1 0-16.3-11.7-39.7-53.8-39.7zm180.1 0h-67.8l-18.7 88.8h70.1c53.8 0 70.1-23.4 70.1-49.1 0-16.3-11.7-39.7-53.7-39.7zm189.3-53.8h-67.8l-18.7 88.8h70.1c53.8 0 70.1-23.4 70.1-49.1.1-16.3-11.6-39.7-53.7-39.7zm-28 137.9h-67.8L343.7 381h70.1c56.1 0 70.1-23.4 70.1-49.1 0-16.3-11.6-39.7-53.7-39.7zM240.8 346H173l-18.7 88.8h70.1c56.1 0 70.1-25.7 70.1-49.1.1-16.3-11.6-39.7-53.7-39.7z\"],\"creative-commons-pd\":[496,512,[],\"f4ec\",\"M248 8C111 8 0 119.1 0 256c0 137 111 248 248 248s248-111 248-248C496 119.1 385 8 248 8zm0 449.5c-139.2 0-235.8-138-190.2-267.9l78.8 35.1c-2.1 10.5-3.3 21.5-3.3 32.9 0 99 73.9 126.9 120.4 126.9 22.9 0 53.5-6.7 79.4-29.5L297 311.1c-5.5 6.3-17.6 16.7-36.3 16.7-37.8 0-53.7-39.9-53.9-71.9 230.4 102.6 216.5 96.5 217.9 96.8-34.3 62.4-100.6 104.8-176.7 104.8zm194.2-150l-224-100c18.8-34 54.9-30.7 74.7-11l40.4-41.6c-27.1-23.3-58-27.5-78.1-27.5-47.4 0-80.9 20.5-100.7 51.6l-74.9-33.4c36.1-54.9 98.1-91.2 168.5-91.2 111.1 0 201.5 90.4 201.5 201.5 0 18-2.4 35.4-6.8 52-.3-.1-.4-.2-.6-.4z\"],playstation:[576,512,[],\"f3df\",\"M570.9 372.3c-11.3 14.2-38.8 24.3-38.8 24.3L327 470.2v-54.3l150.9-53.8c17.1-6.1 19.8-14.8 5.8-19.4-13.9-4.6-39.1-3.3-56.2 2.9L327 381.1v-56.4c23.2-7.8 47.1-13.6 75.7-16.8 40.9-4.5 90.9.6 130.2 15.5 44.2 14 49.2 34.7 38 48.9zm-224.4-92.5v-139c0-16.3-3-31.3-18.3-35.6-11.7-3.8-19 7.1-19 23.4v347.9l-93.8-29.8V32c39.9 7.4 98 24.9 129.2 35.4C424.1 94.7 451 128.7 451 205.2c0 74.5-46 102.8-104.5 74.6zM43.2 410.2c-45.4-12.8-53-39.5-32.3-54.8 19.1-14.2 51.7-24.9 51.7-24.9l134.5-47.8v54.5l-96.8 34.6c-17.1 6.1-19.7 14.8-5.8 19.4 13.9 4.6 39.1 3.3 56.2-2.9l46.4-16.9v48.8c-51.6 9.3-101.4 7.3-153.9-10z\"],quinscape:[512,512,[],\"f459\",\"M313.6 474.6h-1a158.1 158.1 0 0 1 0-316.2c94.9 0 168.2 83.1 157 176.6 4 5.1 8.2 9.6 11.2 15.3 13.4-30.3 20.3-62.4 20.3-97.7C501.1 117.5 391.6 8 256.5 8S12 117.5 12 252.6s109.5 244.6 244.5 244.6a237.36 237.36 0 0 0 70.4-10.1c-5.2-3.5-8.9-8.1-13.3-12.5zm-.1-.1l.4.1zm78.4-168.9a99.2 99.2 0 1 0 99.2 99.2 99.18 99.18 0 0 0-99.2-99.2z\"],less:[640,512,[],\"f41d\",\"M612.7 219c0-20.5 3.2-32.6 3.2-54.6 0-34.2-12.6-45.2-40.5-45.2h-20.5v24.2h6.3c14.2 0 17.3 4.7 17.3 22.1 0 16.3-1.6 32.6-1.6 51.5 0 24.2 7.9 33.6 23.6 37.3v1.6c-15.8 3.7-23.6 13.1-23.6 37.3 0 18.9 1.6 34.2 1.6 51.5 0 17.9-3.7 22.6-17.3 22.6v.5h-6.3V393h20.5c27.8 0 40.5-11 40.5-45.2 0-22.6-3.2-34.2-3.2-54.6 0-11 6.8-22.6 27.3-23.6v-27.3c-20.5-.7-27.3-12.3-27.3-23.3zm-105.6 32c-15.8-6.3-30.5-10-30.5-20.5 0-7.9 6.3-12.6 17.9-12.6s22.1 4.7 33.6 13.1l21-27.8c-13.1-10-31-20.5-55.2-20.5-35.7 0-59.9 20.5-59.9 49.4 0 25.7 22.6 38.9 41.5 46.2 16.3 6.3 32.1 11.6 32.1 22.1 0 7.9-6.3 13.1-20.5 13.1-13.1 0-26.3-5.3-40.5-16.3l-21 30.5c15.8 13.1 39.9 22.1 59.9 22.1 42 0 64.6-22.1 64.6-51s-22.5-41-43-47.8zm-358.9 59.4c-3.7 0-8.4-3.2-8.4-13.1V119.1H65.2c-28.4 0-41 11-41 45.2 0 22.6 3.2 35.2 3.2 54.6 0 11-6.8 22.6-27.3 23.6v27.3c20.5.5 27.3 12.1 27.3 23.1 0 19.4-3.2 31-3.2 53.6 0 34.2 12.6 45.2 40.5 45.2h20.5v-24.2h-6.3c-13.1 0-17.3-5.3-17.3-22.6s1.6-32.1 1.6-51.5c0-24.2-7.9-33.6-23.6-37.3v-1.6c15.8-3.7 23.6-13.1 23.6-37.3 0-18.9-1.6-34.2-1.6-51.5s3.7-22.1 17.3-22.1H93v150.8c0 32.1 11 53.1 43.1 53.1 10 0 17.9-1.6 23.6-3.7l-5.3-34.2c-3.1.8-4.6.8-6.2.8zM379.9 251c-16.3-6.3-31-10-31-20.5 0-7.9 6.3-12.6 17.9-12.6 11.6 0 22.1 4.7 33.6 13.1l21-27.8c-13.1-10-31-20.5-55.2-20.5-35.7 0-59.9 20.5-59.9 49.4 0 25.7 22.6 38.9 41.5 46.2 16.3 6.3 32.1 11.6 32.1 22.1 0 7.9-6.3 13.1-20.5 13.1-13.1 0-26.3-5.3-40.5-16.3l-20.5 30.5c15.8 13.1 39.9 22.1 59.9 22.1 42 0 64.6-22.1 64.6-51 .1-28.9-22.5-41-43-47.8zm-155-68.8c-38.4 0-75.1 32.1-74.1 82.5 0 52 34.2 82.5 79.3 82.5 18.9 0 39.9-6.8 56.2-17.9l-15.8-27.8c-11.6 6.8-22.6 10-34.2 10-21 0-37.3-10-41.5-34.2H290c.5-3.7 1.6-11 1.6-19.4.6-42.6-22.6-75.7-66.7-75.7zm-30 66.2c3.2-21 15.8-31 30.5-31 18.9 0 26.3 13.1 26.3 31h-56.8z\"],\"blogger-b\":[448,512,[],\"f37d\",\"M446.6 222.7c-1.8-8-6.8-15.4-12.5-18.5-1.8-1-13-2.2-25-2.7-20.1-.9-22.3-1.3-28.7-5-10.1-5.9-12.8-12.3-12.9-29.5-.1-33-13.8-63.7-40.9-91.3-19.3-19.7-40.9-33-65.5-40.5-5.9-1.8-19.1-2.4-63.3-2.9-69.4-.8-84.8.6-108.4 10C45.9 59.5 14.7 96.1 3.3 142.9 1.2 151.7.7 165.8.2 246.8c-.6 101.5.1 116.4 6.4 136.5 15.6 49.6 59.9 86.3 104.4 94.3 14.8 2.7 197.3 3.3 216 .8 32.5-4.4 58-17.5 81.9-41.9 17.3-17.7 28.1-36.8 35.2-62.1 4.9-17.6 4.5-142.8 2.5-151.7zm-322.1-63.6c7.8-7.9 10-8.2 58.8-8.2 43.9 0 45.4.1 51.8 3.4 9.3 4.7 13.4 11.3 13.4 21.9 0 9.5-3.8 16.2-12.3 21.6-4.6 2.9-7.3 3.1-50.3 3.3-26.5.2-47.7-.4-50.8-1.2-16.6-4.7-22.8-28.5-10.6-40.8zm191.8 199.8l-14.9 2.4-77.5.9c-68.1.8-87.3-.4-90.9-2-7.1-3.1-13.8-11.7-14.9-19.4-1.1-7.3 2.6-17.3 8.2-22.4 7.1-6.4 10.2-6.6 97.3-6.7 89.6-.1 89.1-.1 97.6 7.8 12.1 11.3 9.5 31.2-4.9 39.4z\"],opencart:[640,512,[],\"f23d\",\"M423.3 440.7c0 25.3-20.3 45.6-45.6 45.6s-45.8-20.3-45.8-45.6 20.6-45.8 45.8-45.8c25.4 0 45.6 20.5 45.6 45.8zm-253.9-45.8c-25.3 0-45.6 20.6-45.6 45.8s20.3 45.6 45.6 45.6 45.8-20.3 45.8-45.6-20.5-45.8-45.8-45.8zm291.7-270C158.9 124.9 81.9 112.1 0 25.7c34.4 51.7 53.3 148.9 373.1 144.2 333.3-5 130 86.1 70.8 188.9 186.7-166.7 319.4-233.9 17.2-233.9z\"],vine:[384,512,[],\"f1ca\",\"M384 254.7v52.1c-18.4 4.2-36.9 6.1-52.1 6.1-36.9 77.4-103 143.8-125.1 156.2-14 7.9-27.1 8.4-42.7-.8C137 452 34.2 367.7 0 102.7h74.5C93.2 261.8 139 343.4 189.3 404.5c27.9-27.9 54.8-65.1 75.6-106.9-49.8-25.3-80.1-80.9-80.1-145.6 0-65.6 37.7-115.1 102.2-115.1 114.9 0 106.2 127.9 81.6 181.5 0 0-46.4 9.2-63.5-20.5 3.4-11.3 8.2-30.8 8.2-48.5 0-31.3-11.3-46.6-28.4-46.6-18.2 0-30.8 17.1-30.8 50 .1 79.2 59.4 118.7 129.9 101.9z\"],\"signal-messenger\":[512,512,[],\"e663\",\"M194.6 7.5l5.8 23.3C177.7 36.3 156 45.3 136 57.4L123.7 36.8c22-13.3 45.9-23.2 70.9-29.3zm122.9 0l-5.8 23.3C334.3 36.3 356 45.3 376 57.4l12.4-20.6c-22-13.3-46-23.2-71-29.3zM36.8 123.7c-13.3 22-23.2 45.9-29.3 70.9l23.3 5.8C36.3 177.7 45.3 156 57.4 136L36.8 123.7zM24 256c0-11.6 .9-23.3 2.6-34.8L2.9 217.6c-3.8 25.4-3.8 51.3 0 76.7l23.7-3.6C24.9 279.3 24 267.6 24 256zM388.3 475.2L376 454.6c-20 12.1-41.6 21-64.2 26.6l5.8 23.3c24.9-6.2 48.8-16 70.8-29.3zM488 256c0 11.6-.9 23.3-2.6 34.8l23.7 3.6c3.8-25.4 3.8-51.3 0-76.7l-23.7 3.6c1.7 11.5 2.6 23.1 2.6 34.8zm16.5 61.4l-23.3-5.8c-5.6 22.7-14.5 44.3-26.6 64.3l20.6 12.4c13.3-22 23.2-46 29.3-71zm-213.8 168c-23 3.5-46.5 3.5-69.5 0l-3.6 23.7c25.4 3.8 51.3 3.8 76.7 0l-3.6-23.7zm152-91.8c-13.8 18.7-30.4 35.3-49.2 49.1l14.2 19.3c20.7-15.2 39-33.4 54.2-54.1l-19.3-14.4zM393.6 69.2c18.8 13.8 35.3 30.4 49.2 49.2L462.1 104C446.9 83.4 428.6 65.1 408 49.9L393.6 69.2zM69.2 118.4c13.8-18.8 30.4-35.3 49.2-49.2L104 49.9C83.4 65.1 65.1 83.4 49.9 104l19.3 14.4zm406 5.3L454.6 136c12.1 20 21 41.6 26.6 64.2l23.3-5.8c-6.2-24.9-16-48.8-29.3-70.8zm-254-97.1c23-3.5 46.5-3.5 69.5 0l3.6-23.7C268.9-1 243.1-1 217.6 2.9l3.6 23.7zM81.6 468.4L32 480l11.6-49.6L20.2 425 8.6 474.5c-.9 4-.8 8.1 .3 12.1s3.2 7.5 6.1 10.4s6.5 5 10.4 6.1s8.1 1.2 12.1 .3L87 492l-5.4-23.6zM25.2 403.6L48.6 409l8-34.4c-11.7-19.6-20.4-40.8-25.8-63L7.5 317.4c5.2 21.2 13.2 41.7 23.6 60.8l-5.9 25.3zm112 52l-34.4 8 5.4 23.4 25.3-5.9c19.2 10.4 39.6 18.4 60.8 23.6l5.8-23.3c-22.1-5.5-43.3-14.3-62.8-26l-.2 .2zM256 48c-37.2 0-73.6 10-105.6 28.9s-58.4 46-76.3 78.6s-26.9 69.3-25.8 106.4s12 73.3 31.8 104.8L60 452l85.3-20c27.3 17.2 58.2 27.8 90.3 31s64.5-1.1 94.6-12.6s57.2-29.8 79-53.6s37.8-52.2 46.8-83.2s10.5-63.6 4.7-95.3s-19-61.6-38.4-87.4s-44.5-46.7-73.4-61S288.3 48 256 48z\"],paypal:[384,512,[],\"f1ed\",\"M111.4 295.9c-3.5 19.2-17.4 108.7-21.5 134-.3 1.8-1 2.5-3 2.5H12.3c-7.6 0-13.1-6.6-12.1-13.9L58.8 46.6c1.5-9.6 10.1-16.9 20-16.9 152.3 0 165.1-3.7 204 11.4 60.1 23.3 65.6 79.5 44 140.3-21.5 62.6-72.5 89.5-140.1 90.3-43.4.7-69.5-7-75.3 24.2zM357.1 152c-1.8-1.3-2.5-1.8-3 1.3-2 11.4-5.1 22.5-8.8 33.6-39.9 113.8-150.5 103.9-204.5 103.9-6.1 0-10.1 3.3-10.9 9.4-22.6 140.4-27.1 169.7-27.1 169.7-1 7.1 3.5 12.9 10.6 12.9h63.5c8.6 0 15.7-6.3 17.4-14.9.7-5.4-1.1 6.1 14.4-91.3 4.6-22 14.3-19.7 29.3-19.7 71 0 126.4-28.8 142.9-112.3 6.5-34.8 4.6-71.4-23.8-92.6z\"],gitlab:[512,512,[],\"f296\",\"M503.5 204.6L502.8 202.8L433.1 21.02C431.7 17.45 429.2 14.43 425.9 12.38C423.5 10.83 420.8 9.865 417.9 9.57C415 9.275 412.2 9.653 409.5 10.68C406.8 11.7 404.4 13.34 402.4 15.46C400.5 17.58 399.1 20.13 398.3 22.9L351.3 166.9H160.8L113.7 22.9C112.9 20.13 111.5 17.59 109.6 15.47C107.6 13.35 105.2 11.72 102.5 10.7C99.86 9.675 96.98 9.295 94.12 9.587C91.26 9.878 88.51 10.83 86.08 12.38C82.84 14.43 80.33 17.45 78.92 21.02L9.267 202.8L8.543 204.6C-1.484 230.8-2.72 259.6 5.023 286.6C12.77 313.5 29.07 337.3 51.47 354.2L51.74 354.4L52.33 354.8L158.3 434.3L210.9 474L242.9 498.2C246.6 500.1 251.2 502.5 255.9 502.5C260.6 502.5 265.2 500.1 268.9 498.2L300.9 474L353.5 434.3L460.2 354.4L460.5 354.1C482.9 337.2 499.2 313.5 506.1 286.6C514.7 259.6 513.5 230.8 503.5 204.6z\"],typo3:[448,512,[],\"f42b\",\"M178.7 78.4c0-24.7 5.4-32.4 13.9-39.4-69.5 8.5-149.3 34-176.3 66.4-5.4 7.7-9.3 20.8-9.3 37.1C7 246 113.8 480 191.1 480c36.3 0 97.3-59.5 146.7-139-7 2.3-11.6 2.3-18.5 2.3-57.2 0-140.6-198.5-140.6-264.9zM301.5 32c-30.1 0-41.7 5.4-41.7 36.3 0 66.4 53.8 198.5 101.7 198.5 26.3 0 78.8-99.7 78.8-182.3 0-40.9-67-52.5-138.8-52.5z\"],\"reddit-alien\":[512,512,[],\"f281\",\"M373 138.6c-25.2 0-46.3-17.5-51.9-41l0 0c-30.6 4.3-54.2 30.7-54.2 62.4l0 .2c47.4 1.8 90.6 15.1 124.9 36.3c12.6-9.7 28.4-15.5 45.5-15.5c41.3 0 74.7 33.4 74.7 74.7c0 29.8-17.4 55.5-42.7 67.5c-2.4 86.8-97 156.6-213.2 156.6S45.5 410.1 43 323.4C17.6 311.5 0 285.7 0 255.7c0-41.3 33.4-74.7 74.7-74.7c17.2 0 33 5.8 45.7 15.6c34-21.1 76.8-34.4 123.7-36.4l0-.3c0-44.3 33.7-80.9 76.8-85.5C325.8 50.2 347.2 32 373 32c29.4 0 53.3 23.9 53.3 53.3s-23.9 53.3-53.3 53.3zM157.5 255.3c-20.9 0-38.9 20.8-40.2 47.9s17.1 38.1 38 38.1s36.6-9.8 37.8-36.9s-14.7-49.1-35.7-49.1zM395 303.1c-1.2-27.1-19.2-47.9-40.2-47.9s-36.9 22-35.7 49.1c1.2 27.1 16.9 36.9 37.8 36.9s39.3-11 38-38.1zm-60.1 70.8c1.5-3.6-1-7.7-4.9-8.1c-23-2.3-47.9-3.6-73.8-3.6s-50.8 1.3-73.8 3.6c-3.9 .4-6.4 4.5-4.9 8.1c12.9 30.8 43.3 52.4 78.7 52.4s65.8-21.6 78.7-52.4z\"],yahoo:[512,512,[],\"f19e\",\"M223.69,141.06,167,284.23,111,141.06H14.93L120.76,390.19,82.19,480h94.17L317.27,141.06Zm105.4,135.79a58.22,58.22,0,1,0,58.22,58.22A58.22,58.22,0,0,0,329.09,276.85ZM394.65,32l-93,223.47H406.44L499.07,32Z\"],dailymotion:[448,512,[],\"e052\",\"M298.93,267a48.4,48.4,0,0,0-24.36-6.21q-19.83,0-33.44,13.27t-13.61,33.42q0,21.16,13.28,34.6t33.43,13.44q20.5,0,34.11-13.78T322,307.47A47.13,47.13,0,0,0,315.9,284,44.13,44.13,0,0,0,298.93,267ZM0,32V480H448V32ZM374.71,405.26h-53.1V381.37h-.67q-15.79,26.2-55.78,26.2-27.56,0-48.89-13.1a88.29,88.29,0,0,1-32.94-35.77q-11.6-22.68-11.59-50.89,0-27.56,11.76-50.22a89.9,89.9,0,0,1,32.93-35.78q21.18-13.09,47.72-13.1a80.87,80.87,0,0,1,29.74,5.21q13.28,5.21,25,17V153l55.79-12.09Z\"],affiliatetheme:[512,512,[],\"f36b\",\"M159.7 237.4C108.4 308.3 43.1 348.2 14 326.6-15.2 304.9 2.8 230 54.2 159.1c51.3-70.9 116.6-110.8 145.7-89.2 29.1 21.6 11.1 96.6-40.2 167.5zm351.2-57.3C437.1 303.5 319 367.8 246.4 323.7c-25-15.2-41.3-41.2-49-73.8-33.6 64.8-92.8 113.8-164.1 133.2 49.8 59.3 124.1 96.9 207 96.9 150 0 271.6-123.1 271.6-274.9.1-8.5-.3-16.8-1-25z\"],\"pied-piper-pp\":[448,512,[],\"f1a7\",\"M205.3 174.6c0 21.1-14.2 38.1-31.7 38.1-7.1 0-12.8-1.2-17.2-3.7v-68c4.4-2.7 10.1-4.2 17.2-4.2 17.5 0 31.7 16.9 31.7 37.8zm52.6 67c-7.1 0-12.8 1.5-17.2 4.2v68c4.4 2.5 10.1 3.7 17.2 3.7 17.4 0 31.7-16.9 31.7-37.8 0-21.1-14.3-38.1-31.7-38.1zM448 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h352c26.5 0 48 21.5 48 48zM185 255.1c41 0 74.2-35.6 74.2-79.6 0-44-33.2-79.6-74.2-79.6-12 0-24.1 3.2-34.6 8.8h-45.7V311l51.8-10.1v-50.6c8.6 3.1 18.1 4.8 28.5 4.8zm158.4 25.3c0-44-33.2-79.6-73.9-79.6-3.2 0-6.4.2-9.6.7-3.7 12.5-10.1 23.8-19.2 33.4-13.8 15-32.2 23.8-51.8 24.8V416l51.8-10.1v-50.6c8.6 3.2 18.2 4.7 28.7 4.7 40.8 0 74-35.6 74-79.6z\"],bootstrap:[576,512,[],\"f836\",\"M333.5,201.4c0-22.1-15.6-34.3-43-34.3h-50.4v71.2h42.5C315.4,238.2,333.5,225,333.5,201.4z M517,188.6 c-9.5-30.9-10.9-68.8-9.8-98.1c1.1-30.5-22.7-58.5-54.7-58.5H123.7c-32.1,0-55.8,28.1-54.7,58.5c1,29.3-0.3,67.2-9.8,98.1 c-9.6,31-25.7,50.6-52.2,53.1v28.5c26.4,2.5,42.6,22.1,52.2,53.1c9.5,30.9,10.9,68.8,9.8,98.1c-1.1,30.5,22.7,58.5,54.7,58.5h328.7 c32.1,0,55.8-28.1,54.7-58.5c-1-29.3,0.3-67.2,9.8-98.1c9.6-31,25.7-50.6,52.1-53.1v-28.5C542.7,239.2,526.5,219.6,517,188.6z M300.2,375.1h-97.9V136.8h97.4c43.3,0,71.7,23.4,71.7,59.4c0,25.3-19.1,47.9-43.5,51.8v1.3c33.2,3.6,55.5,26.6,55.5,58.3 C383.4,349.7,352.1,375.1,300.2,375.1z M290.2,266.4h-50.1v78.4h52.3c34.2,0,52.3-13.7,52.3-39.5 C344.7,279.6,326.1,266.4,290.2,266.4z\"],odnoklassniki:[320,512,[],\"f263\",\"M275.1 334c-27.4 17.4-65.1 24.3-90 26.9l20.9 20.6 76.3 76.3c27.9 28.6-17.5 73.3-45.7 45.7-19.1-19.4-47.1-47.4-76.3-76.6L84 503.4c-28.2 27.5-73.6-17.6-45.4-45.7 19.4-19.4 47.1-47.4 76.3-76.3l20.6-20.6c-24.6-2.6-62.9-9.1-90.6-26.9-32.6-21-46.9-33.3-34.3-59 7.4-14.6 27.7-26.9 54.6-5.7 0 0 36.3 28.9 94.9 28.9s94.9-28.9 94.9-28.9c26.9-21.1 47.1-8.9 54.6 5.7 12.4 25.7-1.9 38-34.5 59.1zM30.3 129.7C30.3 58 88.6 0 160 0s129.7 58 129.7 129.7c0 71.4-58.3 129.4-129.7 129.4s-129.7-58-129.7-129.4zm66 0c0 35.1 28.6 63.7 63.7 63.7s63.7-28.6 63.7-63.7c0-35.4-28.6-64-63.7-64s-63.7 28.6-63.7 64z\"],\"nfc-symbol\":[576,512,[],\"e531\",\"M392.9 32.43C400.6 31.1 408.6 32.89 414.1 37.41C498.2 96.14 544 173.7 544 255.1C544 338.2 498.2 415.9 414.1 474.6C409.3 478.6 402.4 480.5 395.5 479.9C388.5 479.3 382 476.3 377.1 471.4L193.7 288.7C188.1 283.2 185 275.7 184.1 267.8C184.1 260 188.1 252.5 193.6 246.9C199.2 241.4 206.7 238.2 214.5 238.2C222.4 238.2 229.9 241.3 235.4 246.8L400.5 411.2C455.1 366.5 484.8 312 484.8 255.1C484.8 193.5 447.9 132.9 380.9 85.76C374.5 81.24 370.1 74.35 368.8 66.62C367.4 58.89 369.2 50.94 373.8 44.53C378.3 38.12 385.2 33.77 392.9 32.43V32.43zM186.9 479.6C179.2 480.9 171.3 479.1 164.8 474.6C81.67 415.9 35.84 338.2 35.84 255.1C35.84 173.7 81.67 96.14 164.8 37.41C170.5 33.4 177.4 31.53 184.4 32.12C191.3 32.71 197.8 35.72 202.7 40.63L386.1 223.3C391.7 228.8 394.8 236.3 394.8 244.2C394.9 251.1 391.8 259.5 386.2 265.1C380.7 270.6 373.2 273.8 365.3 273.8C357.5 273.8 349.1 270.7 344.4 265.2L179.3 100.7C124.7 145.9 95.03 199.9 95.03 255.1C95.03 318.5 131.9 379.1 198.1 426.2C205.4 430.8 209.7 437.6 211.1 445.4C212.4 453.1 210.6 461.1 206.1 467.5C201.6 473.9 194.7 478.2 186.9 479.6V479.6z\"],mintbit:[512,512,[],\"e62f\",\"M73.2 512V438.9H365.7V365.7h73.2V219.4H512V0H292.6V73.1H146.3v73.2H73.2V438.9H0V512H73.2zm73.1-219.4h73.2v73.1H146.3V292.6zm73.2-73.1h73.1v73.1H219.4V219.4zm73.1 0V146.3h73.2v73.1H292.6zM365.7 73.1h73.2v73.2H365.7V73.1z\"],ethereum:[320,512,[],\"f42e\",\"M311.9 260.8L160 353.6 8 260.8 160 0l151.9 260.8zM160 383.4L8 290.6 160 512l152-221.4-152 92.8z\"],\"speaker-deck\":[512,512,[],\"f83c\",\"M213.86 296H100a100 100 0 0 1 0-200h132.84a40 40 0 0 1 0 80H98c-26.47 0-26.45 40 0 40h113.82a100 100 0 0 1 0 200H40a40 40 0 0 1 0-80h173.86c26.48 0 26.46-40 0-40zM298 416a120.21 120.21 0 0 0 51.11-80h64.55a19.83 19.83 0 0 0 19.66-20V196a19.83 19.83 0 0 0-19.66-20H296.42a60.77 60.77 0 0 0 0-80h136.93c43.44 0 78.65 35.82 78.65 80v160c0 44.18-35.21 80-78.65 80z\"],\"creative-commons-nc-eu\":[496,512,[],\"f4e9\",\"M247.7 8C103.6 8 0 124.8 0 256c0 136.3 111.7 248 247.7 248C377.9 504 496 403.1 496 256 496 117 388.4 8 247.7 8zm.6 450.7c-112 0-203.6-92.5-203.6-202.7 0-23.2 3.7-45.2 10.9-66l65.7 29.1h-4.7v29.5h23.3c0 6.2-.4 3.2-.4 19.5h-22.8v29.5h27c11.4 67 67.2 101.3 124.6 101.3 26.6 0 50.6-7.9 64.8-15.8l-10-46.1c-8.7 4.6-28.2 10.8-47.3 10.8-28.2 0-58.1-10.9-67.3-50.2h90.3l128.3 56.8c-1.5 2.1-56.2 104.3-178.8 104.3zm-16.7-190.6l-.5-.4.9.4h-.4zm77.2-19.5h3.7v-29.5h-70.3l-28.6-12.6c2.5-5.5 5.4-10.5 8.8-14.3 12.9-15.8 31.1-22.4 51.1-22.4 18.3 0 35.3 5.4 46.1 10l11.6-47.3c-15-6.6-37-12.4-62.3-12.4-39 0-72.2 15.8-95.9 42.3-5.3 6.1-9.8 12.9-13.9 20.1l-81.6-36.1c64.6-96.8 157.7-93.6 170.7-93.6 113 0 203 90.2 203 203.4 0 18.7-2.1 36.3-6.3 52.9l-136.1-60.5z\"],patreon:[512,512,[],\"f3d9\",\"M489.7 153.8c-.1-65.4-51-119-110.7-138.3C304.8-8.5 207-5 136.1 28.4C50.3 68.9 23.3 157.7 22.3 246.2C21.5 319 28.7 510.6 136.9 512c80.3 1 92.3-102.5 129.5-152.3c26.4-35.5 60.5-45.5 102.4-55.9c72-17.8 121.1-74.7 121-150z\"],avianex:[512,512,[],\"f374\",\"M453.1 32h-312c-38.9 0-76.2 31.2-83.3 69.7L1.2 410.3C-5.9 448.8 19.9 480 58.9 480h312c38.9 0 76.2-31.2 83.3-69.7l56.7-308.5c7-38.6-18.8-69.8-57.8-69.8zm-58.2 347.3l-32 13.5-115.4-110c-14.7 10-29.2 19.5-41.7 27.1l22.1 64.2-17.9 12.7-40.6-61-52.4-48.1 15.7-15.4 58 31.1c9.3-10.5 20.8-22.6 32.8-34.9L203 228.9l-68.8-99.8 18.8-28.9 8.9-4.8L265 207.8l4.9 4.5c19.4-18.8 33.8-32.4 33.8-32.4 7.7-6.5 21.5-2.9 30.7 7.9 9 10.5 10.6 24.7 2.7 31.3-1.8 1.3-15.5 11.4-35.3 25.6l4.5 7.3 94.9 119.4-6.3 7.9z\"],ello:[496,512,[],\"f5f1\",\"M248 8C111.03 8 0 119.03 0 256s111.03 248 248 248 248-111.03 248-248S384.97 8 248 8zm143.84 285.2C375.31 358.51 315.79 404.8 248 404.8s-127.31-46.29-143.84-111.6c-1.65-7.44 2.48-15.71 9.92-17.36 7.44-1.65 15.71 2.48 17.36 9.92 14.05 52.91 62 90.11 116.56 90.11s102.51-37.2 116.56-90.11c1.65-7.44 9.92-12.4 17.36-9.92 7.44 1.65 12.4 9.92 9.92 17.36z\"],gofore:[400,512,[],\"f3a7\",\"M324 319.8h-13.2v34.7c-24.5 23.1-56.3 35.8-89.9 35.8-73.2 0-132.4-60.2-132.4-134.4 0-74.1 59.2-134.4 132.4-134.4 35.3 0 68.6 14 93.6 39.4l62.3-63.3C335 55.3 279.7 32 220.7 32 98 32 0 132.6 0 256c0 122.5 97 224 220.7 224 63.2 0 124.5-26.2 171-82.5-2-27.6-13.4-77.7-67.7-77.7zm-12.1-112.5H205.6v89H324c33.5 0 60.5 15.1 76 41.8v-30.6c0-65.2-40.4-100.2-88.1-100.2z\"],bimobject:[448,512,[],\"f378\",\"M416 32H32C14.4 32 0 46.4 0 64v384c0 17.6 14.4 32 32 32h384c17.6 0 32-14.4 32-32V64c0-17.6-14.4-32-32-32zm-64 257.4c0 49.4-11.4 82.6-103.8 82.6h-16.9c-44.1 0-62.4-14.9-70.4-38.8h-.9V368H96V136h64v74.7h1.1c4.6-30.5 39.7-38.8 69.7-38.8h17.3c92.4 0 103.8 33.1 103.8 82.5v35zm-64-28.9v22.9c0 21.7-3.4 33.8-38.4 33.8h-45.3c-28.9 0-44.1-6.5-44.1-35.7v-19c0-29.3 15.2-35.7 44.1-35.7h45.3c35-.2 38.4 12 38.4 33.7z\"],\"brave-reverse\":[448,512,[],\"e63d\",\"M298 0c3 0 5.8 1.3 7.8 3.6l38.1 44c.5-.1 1-.2 1.5-.3c9.2-1.6 18.6-2.2 27.7-1.2c11.6 1.4 21.5 5.4 28.9 12.9c7.7 7.8 15.4 15.8 22.6 23.6c2.5 2.7 4.9 5.2 6.9 7.4c.7 .8 1.4 1.5 1.9 2c3.4 3.7 4.2 8.1 2.7 11.9l-9.8 24.6 13.1 38.1c.7 2 .8 4.1 .2 6.2c-.1 .4-.1 .4-.5 2.1c-.6 2.3-.6 2.3-1.5 5.8c-1.6 6.3-3.5 13.3-5.4 20.9c-5.6 21.6-11.2 43.2-16.4 63.4c-12.9 49.9-21.4 82.7-23.4 90.9c-11.1 44.5-19.9 60-48.3 80.3c-24.9 17.8-76.8 53.6-86.8 60c-1 .6-2 1.3-3.4 2.3c-.5 .4-3.2 2.2-3.9 2.7c-4.9 3.3-8.3 5.5-12.1 7.3c-4.7 2.2-9.3 3.5-13.9 3.5s-9.1-1.2-13.9-3.5c-3.7-1.8-7.2-3.9-12.1-7.3c-.8-.5-3.4-2.4-3.9-2.7c-1.4-1-2.5-1.7-3.4-2.3c-10-6.4-61.9-42.1-86.8-60c-28.4-20.4-37.2-35.8-48.3-80.3c-2-8.2-10.5-41-23.3-90.5c-5.3-20.6-10.9-42.2-16.5-63.8c-2-7.6-3.8-14.6-5.4-20.9c-.9-3.5-.9-3.5-1.5-5.8c-.4-1.7-.4-1.7-.5-2.1c-.5-2-.4-4.2 .2-6.2l13.1-38.1L11.8 104c-1.5-3.8-.7-8.2 2-11.2c1.2-1.3 1.8-2 2.6-2.8c2-2.2 4.4-4.7 6.9-7.4C30.6 74.9 38.3 66.9 46 59.1c7.4-7.5 17.3-11.6 28.9-12.9c9.1-1.1 18.5-.5 27.7 1.2c.5 .1 1 .2 1.5 .3l38.1-44C144.2 1.3 147 0 150 0H298zm-4.7 21.1H154.7L115.6 66.2c-2.6 3-6.7 4.3-10.6 3.2c-.2-.1-.7-.2-1.5-.4c-1.3-.3-2.9-.6-4.5-.9c-7.4-1.3-14.9-1.8-21.7-1C70 68 64.3 70.3 60.7 74c-7.6 7.7-15.2 15.6-22.3 23.3c-1.7 1.8-3.3 3.5-4.8 5.1l8.8 22c1 2.4 1 5 .2 7.5L29.2 170.6c.4 1.4 .5 1.9 1.2 4.8c1.6 6.3 3.5 13.3 5.4 20.9c5.6 21.6 11.2 43.2 16.4 63.4c12.9 50 21.4 82.8 23.4 91C85.7 390.8 92 402 115.8 419c24.6 17.6 76.3 53.2 85.9 59.3c1.2 .8 2.5 1.6 4 2.7c.6 .4 3.2 2.2 3.9 2.7c4 2.8 6.7 4.4 9.2 5.6c2.2 1 3.9 1.5 5.1 1.5s2.9-.5 5.1-1.5c2.5-1.2 5.2-2.8 9.2-5.6c.7-.5 3.3-2.3 3.9-2.7c1.6-1.1 2.8-1.9 4-2.7c9.6-6.1 61.3-41.7 85.9-59.3c23.8-17.1 30.2-28.2 40.1-68.3c2.1-8.3 10.5-41.1 23.3-90.7c5.3-20.6 10.9-42.2 16.5-63.8c2-7.6 3.8-14.6 5.4-20.9c.7-2.9 .9-3.4 1.2-4.8l-13.3-38.8c-.8-2.4-.8-5.1 .2-7.5l8.8-22c-1.5-1.6-3.1-3.3-4.8-5.1c-7.2-7.6-14.7-15.5-22.3-23.3c-3.7-3.7-9.3-6-16.6-6.9c-6.8-.8-14.4-.3-21.7 1c-1.7 .3-3.2 .6-4.5 .9c-.8 .2-1.3 .3-1.5 .4c-3.8 1.1-7.9-.2-10.6-3.2L293.3 21.1zM224 316c2.8 0 20.9 6.5 35.4 14.1s25 13 28.3 15.2s1.3 6.2-1.7 8.4s-44.1 34.6-48.1 38.2s-9.8 9.5-13.8 9.5s-9.8-5.9-13.8-9.5s-45.1-36-48.1-38.2s-5.1-6.2-1.7-8.4s13.9-7.5 28.3-15.2s32.5-14.1 35.4-14.1zm.1-230.7c.7 0 8.8 .2 20.5 4.2c12.3 4.2 25.7 9.4 31.9 9.4s51.9-8.9 51.9-8.9s54.2 66.7 54.2 81s-6.8 18-13.7 25.4s-36.8 39.8-40.7 43.9s-11.9 10.5-7.1 21.8s11.7 25.8 3.9 40.4s-21 24.4-29.4 22.8s-28.4-12.2-35.7-17.1s-30.5-24.3-30.5-31.8s24-20.8 28.4-23.9s24.7-14.8 25.1-19.4s.3-6-5.7-17.4s-16.7-26.7-14.9-36.8s19.1-15.4 31.5-20.2s36.2-13.7 39.2-15.1s2.2-2.7-6.8-3.6s-34.6-4.3-46.1-1.1s-31.2 8.2-32.8 10.9s-3 2.7-1.4 11.8s10.1 52.8 10.9 60.6s2.4 12.9-5.8 14.8s-22.1 5.2-26.8 5.2s-18.6-3.3-26.8-5.2s-6.6-7-5.8-14.8s9.3-51.5 10.9-60.6s.2-9.2-1.4-11.8s-21.3-7.6-32.8-10.9s-37.1 .2-46.1 1.1s-9.8 2.2-6.8 3.6s26.8 10.4 39.2 15.1s29.7 10 31.5 20.2s-9 25.4-14.9 36.8s-6.1 12.8-5.7 17.4s20.6 16.4 25.1 19.4s28.4 16.4 28.4 23.9s-23.2 27-30.5 31.8s-27.2 15.4-35.7 17.1s-21.7-8.2-29.4-22.8s-.8-29.1 3.9-40.4s-3.3-17.7-7.1-21.8s-33.8-36.5-40.7-43.9s-13.7-11.2-13.7-25.4s54.2-81 54.2-81s45.8 8.9 51.9 8.9s19.5-5.2 31.9-9.4s20.6-4.2 20.6-4.2l.1 0z\"],\"facebook-f\":[320,512,[],\"f39e\",\"M80 299.3V512H196V299.3h86.5l18-97.8H196V166.9c0-51.7 20.3-71.5 72.7-71.5c16.3 0 29.4 .4 37 1.2V7.9C291.4 4 256.4 0 236.2 0C129.3 0 80 50.5 80 159.4v42.1H14v97.8H80z\"],\"square-google-plus\":[448,512,[\"google-plus-square\"],\"f0d4\",\"M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM64 256c0-55.3 44.7-100 100-100c27 0 49.5 9.8 67 26.2l-27.1 26.1c-7.4-7.1-20.3-15.4-39.8-15.4c-34.1 0-61.9 28.2-61.9 63.2c0 34.9 27.8 63.2 61.9 63.2c39.6 0 54.4-28.5 56.8-43.1H164V241.8h94.4c1 5 1.6 10.1 1.6 16.6c0 57.1-38.3 97.6-96 97.6c-55.3 0-100-44.7-100-100zm291 18.2v29H325.8v-29h-29V245h29V216H355v29h29v29.2H355z\"],\"web-awesome\":[640,512,[],\"e682\",\"M372.2 52c0 20.9-12.4 39-30.2 47.2L448 192l104.4-20.9c-5.3-7.7-8.4-17.1-8.4-27.1c0-26.5 21.5-48 48-48s48 21.5 48 48c0 26-20.6 47.1-46.4 48L481 442.3c-10.3 23-33.2 37.7-58.4 37.7l-205.2 0c-25.2 0-48-14.8-58.4-37.7L46.4 192C20.6 191.1 0 170 0 144c0-26.5 21.5-48 48-48s48 21.5 48 48c0 10.1-3.1 19.4-8.4 27.1L192 192 298.1 99.1c-17.7-8.3-30-26.3-30-47.1c0-28.7 23.3-52 52-52s52 23.3 52 52z\"],mandalorian:[448,512,[],\"f50f\",\"M232.27 511.89c-1-3.26-1.69-15.83-1.39-24.58.55-15.89 1-24.72 1.4-28.76.64-6.2 2.87-20.72 3.28-21.38.6-1 .4-27.87-.24-33.13-.31-2.58-.63-11.9-.69-20.73-.13-16.47-.53-20.12-2.73-24.76-1.1-2.32-1.23-3.84-1-11.43a92.38 92.38 0 0 0-.34-12.71c-2-13-3.46-27.7-3.25-33.9s.43-7.15 2.06-9.67c3.05-4.71 6.51-14 8.62-23.27 2.26-9.86 3.88-17.18 4.59-20.74a109.54 109.54 0 0 1 4.42-15.05c2.27-6.25 2.49-15.39.37-15.39-.3 0-1.38 1.22-2.41 2.71s-4.76 4.8-8.29 7.36c-8.37 6.08-11.7 9.39-12.66 12.58s-1 7.23-.16 7.76c.34.21 1.29 2.4 2.11 4.88a28.83 28.83 0 0 1 .72 15.36c-.39 1.77-1 5.47-1.46 8.23s-1 6.46-1.25 8.22a9.85 9.85 0 0 1-1.55 4.26c-1 1-1.14.91-2.05-.53a14.87 14.87 0 0 1-1.44-4.75c-.25-1.74-1.63-7.11-3.08-11.93-3.28-10.9-3.52-16.15-1-21a14.24 14.24 0 0 0 1.67-4.61c0-2.39-2.2-5.32-7.41-9.89-7-6.18-8.63-7.92-10.23-11.3-1.71-3.6-3.06-4.06-4.54-1.54-1.78 3-2.6 9.11-3 22l-.34 12.19 2 2.25c3.21 3.7 12.07 16.45 13.78 19.83 3.41 6.74 4.34 11.69 4.41 23.56s.95 22.75 2 24.71c.36.66.51 1.35.34 1.52s.41 2.09 1.29 4.27a38.14 38.14 0 0 1 2.06 9 91 91 0 0 0 1.71 10.37c2.23 9.56 2.77 14.08 2.39 20.14-.2 3.27-.53 11.07-.73 17.32-1.31 41.76-1.85 58-2 61.21-.12 2-.39 11.51-.6 21.07-.36 16.3-1.3 27.37-2.42 28.65-.64.73-8.07-4.91-12.52-9.49-3.75-3.87-4-4.79-2.83-9.95.7-3 2.26-18.29 3.33-32.62.36-4.78.81-10.5 1-12.71.83-9.37 1.66-20.35 2.61-34.78.56-8.46 1.33-16.44 1.72-17.73s.89-9.89 1.13-19.11l.43-16.77-2.26-4.3c-1.72-3.28-4.87-6.94-13.22-15.34-6-6.07-11.84-12.3-12.91-13.85l-1.95-2.81.75-10.9c1.09-15.71 1.1-48.57 0-59.06l-.89-8.7-3.28-4.52c-5.86-8.08-5.8-7.75-6.22-33.27-.1-6.07-.38-11.5-.63-12.06-.83-1.87-3.05-2.66-8.54-3.05-8.86-.62-11-1.9-23.85-14.55-6.15-6-12.34-12-13.75-13.19-2.81-2.42-2.79-2-.56-9.63l1.35-4.65-1.69-3a32.22 32.22 0 0 0-2.59-4.07c-1.33-1.51-5.5-10.89-6-13.49a4.24 4.24 0 0 1 .87-3.9c2.23-2.86 3.4-5.68 4.45-10.73 2.33-11.19 7.74-26.09 10.6-29.22 3.18-3.47 7.7-1 9.41 5 1.34 4.79 1.37 9.79.1 18.55a101.2 101.2 0 0 0-1 11.11c0 4 .19 4.69 2.25 7.39 3.33 4.37 7.73 7.41 15.2 10.52a18.67 18.67 0 0 1 4.72 2.85c11.17 10.72 18.62 16.18 22.95 16.85 5.18.8 8 4.54 10 13.39 1.31 5.65 4 11.14 5.46 11.14a9.38 9.38 0 0 0 3.33-1.39c2-1.22 2.25-1.73 2.25-4.18a132.88 132.88 0 0 0-2-17.84c-.37-1.66-.78-4.06-.93-5.35s-.61-3.85-1-5.69c-2.55-11.16-3.65-15.46-4.1-16-1.55-2-4.08-10.2-4.93-15.92-1.64-11.11-4-14.23-12.91-17.39A43.15 43.15 0 0 1 165.24 78c-1.15-1-4-3.22-6.35-5.06s-4.41-3.53-4.6-3.76a22.7 22.7 0 0 0-2.69-2c-6.24-4.22-8.84-7-11.26-12l-2.44-5-.22-13-.22-13 6.91-6.55c3.95-3.75 8.48-7.35 10.59-8.43 3.31-1.69 4.45-1.89 11.37-2 8.53-.19 10.12 0 11.66 1.56s1.36 6.4-.29 8.5a6.66 6.66 0 0 0-1.34 2.32c0 .58-2.61 4.91-5.42 9a30.39 30.39 0 0 0-2.37 6.82c20.44 13.39 21.55 3.77 14.07 29L194 66.92c3.11-8.66 6.47-17.26 8.61-26.22.29-7.63-12-4.19-15.4-8.68-2.33-5.93 3.13-14.18 6.06-19.2 1.6-2.34 6.62-4.7 8.82-4.15.88.22 4.16-.35 7.37-1.28a45.3 45.3 0 0 1 7.55-1.68 29.57 29.57 0 0 0 6-1.29c3.65-1.11 4.5-1.17 6.35-.4a29.54 29.54 0 0 0 5.82 1.36 18.18 18.18 0 0 1 6 1.91 22.67 22.67 0 0 0 5 2.17c2.51.68 3 .57 7.05-1.67l4.35-2.4L268.32 5c10.44-.4 10.81-.47 15.26-2.68L288.16 0l2.46 1.43c1.76 1 3.14 2.73 4.85 6 2.36 4.51 2.38 4.58 1.37 7.37-.88 2.44-.89 3.3-.1 6.39a35.76 35.76 0 0 0 2.1 5.91 13.55 13.55 0 0 1 1.31 4c.31 4.33 0 5.3-2.41 6.92-2.17 1.47-7 7.91-7 9.34a14.77 14.77 0 0 1-1.07 3c-5 11.51-6.76 13.56-14.26 17-9.2 4.2-12.3 5.19-16.21 5.19-3.1 0-4 .25-4.54 1.26a18.33 18.33 0 0 1-4.09 3.71 13.62 13.62 0 0 0-4.38 4.78 5.89 5.89 0 0 1-2.49 2.91 6.88 6.88 0 0 0-2.45 1.71 67.62 67.62 0 0 1-7 5.38c-3.33 2.34-6.87 5-7.87 6A7.27 7.27 0 0 1 224 100a5.76 5.76 0 0 0-2.13 1.65c-1.31 1.39-1.49 2.11-1.14 4.6a36.45 36.45 0 0 0 1.42 5.88c1.32 3.8 1.31 7.86 0 10.57s-.89 6.65 1.35 9.59c2 2.63 2.16 4.56.71 8.84a33.45 33.45 0 0 0-1.06 8.91c0 4.88.22 6.28 1.46 8.38s1.82 2.48 3.24 2.32c2-.23 2.3-1.05 4.71-12.12 2.18-10 3.71-11.92 13.76-17.08 2.94-1.51 7.46-4 10-5.44s6.79-3.69 9.37-4.91a40.09 40.09 0 0 0 15.22-11.67c7.11-8.79 10-16.22 12.85-33.3a18.37 18.37 0 0 1 2.86-7.73 20.39 20.39 0 0 0 2.89-7.31c1-5.3 2.85-9.08 5.58-11.51 4.7-4.18 6-1.09 4.59 10.87-.46 3.86-1.1 10.33-1.44 14.38l-.61 7.36 4.45 4.09 4.45 4.09.11 8.42c.06 4.63.47 9.53.92 10.89l.82 2.47-6.43 6.28c-8.54 8.33-12.88 13.93-16.76 21.61-1.77 3.49-3.74 7.11-4.38 8-2.18 3.11-6.46 13-8.76 20.26l-2.29 7.22-7 6.49c-3.83 3.57-8 7.25-9.17 8.17-3.05 2.32-4.26 5.15-4.26 10a14.62 14.62 0 0 0 1.59 7.26 42 42 0 0 1 2.09 4.83 9.28 9.28 0 0 0 1.57 2.89c1.4 1.59 1.92 16.12.83 23.22-.68 4.48-3.63 12-4.7 12-1.79 0-4.06 9.27-5.07 20.74-.18 2-.62 5.94-1 8.7s-1 10-1.35 16.05c-.77 12.22-.19 18.77 2 23.15 3.41 6.69.52 12.69-11 22.84l-4 3.49.07 5.19a40.81 40.81 0 0 0 1.14 8.87c4.61 16 4.73 16.92 4.38 37.13-.46 26.4-.26 40.27.63 44.15a61.31 61.31 0 0 1 1.08 7c.17 2 .66 5.33 1.08 7.36.47 2.26.78 11 .79 22.74v19.06l-1.81 2.63c-2.71 3.91-15.11 13.54-15.49 12.29zm29.53-45.11c-.18-.3-.33-6.87-.33-14.59 0-14.06-.89-27.54-2.26-34.45-.4-2-.81-9.7-.9-17.06-.15-11.93-1.4-24.37-2.64-26.38-.66-1.07-3-17.66-3-21.3 0-4.23 1-6 5.28-9.13s4.86-3.14 5.48-.72c.28 1.1 1.45 5.62 2.6 10 3.93 15.12 4.14 16.27 4.05 21.74-.1 5.78-.13 6.13-1.74 17.73-1 7.07-1.17 12.39-1 28.43.17 19.4-.64 35.73-2 41.27-.71 2.78-2.8 5.48-3.43 4.43zm-71-37.58a101 101 0 0 1-1.73-10.79 100.5 100.5 0 0 0-1.73-10.79 37.53 37.53 0 0 1-1-6.49c-.31-3.19-.91-7.46-1.33-9.48-1-4.79-3.35-19.35-3.42-21.07 0-.74-.34-4.05-.7-7.36-.67-6.21-.84-27.67-.22-28.29 1-1 6.63 2.76 11.33 7.43l5.28 5.25-.45 6.47c-.25 3.56-.6 10.23-.78 14.83s-.49 9.87-.67 11.71-.61 9.36-.94 16.72c-.79 17.41-1.94 31.29-2.65 32a.62.62 0 0 1-1-.14zm-87.18-266.59c21.07 12.79 17.84 14.15 28.49 17.66 13 4.29 18.87 7.13 23.15 16.87C111.6 233.28 86.25 255 78.55 268c-31 52-6 101.59 62.75 87.21-14.18 29.23-78 28.63-98.68-4.9-24.68-39.95-22.09-118.3 61-187.66zm210.79 179c56.66 6.88 82.32-37.74 46.54-89.23 0 0-26.87-29.34-64.28-68 3-15.45 9.49-32.12 30.57-53.82 89.2 63.51 92 141.61 92.46 149.36 4.3 70.64-78.7 91.18-105.29 61.71z\"],\"first-order-alt\":[496,512,[],\"f50a\",\"M248 8C111.03 8 0 119.03 0 256s111.03 248 248 248 248-111.03 248-248S384.97 8 248 8zm0 488.21C115.34 496.21 7.79 388.66 7.79 256S115.34 15.79 248 15.79 488.21 123.34 488.21 256 380.66 496.21 248 496.21zm0-459.92C126.66 36.29 28.29 134.66 28.29 256S126.66 475.71 248 475.71 467.71 377.34 467.71 256 369.34 36.29 248 36.29zm0 431.22c-116.81 0-211.51-94.69-211.51-211.51S131.19 44.49 248 44.49 459.51 139.19 459.51 256 364.81 467.51 248 467.51zm186.23-162.98a191.613 191.613 0 0 1-20.13 48.69l-74.13-35.88 61.48 54.82a193.515 193.515 0 0 1-37.2 37.29l-54.8-61.57 35.88 74.27a190.944 190.944 0 0 1-48.63 20.23l-27.29-78.47 4.79 82.93c-8.61 1.18-17.4 1.8-26.33 1.8s-17.72-.62-26.33-1.8l4.76-82.46-27.15 78.03a191.365 191.365 0 0 1-48.65-20.2l35.93-74.34-54.87 61.64a193.85 193.85 0 0 1-37.22-37.28l61.59-54.9-74.26 35.93a191.638 191.638 0 0 1-20.14-48.69l77.84-27.11-82.23 4.76c-1.16-8.57-1.78-17.32-1.78-26.21 0-9 .63-17.84 1.82-26.51l82.38 4.77-77.94-27.16a191.726 191.726 0 0 1 20.23-48.67l74.22 35.92-61.52-54.86a193.85 193.85 0 0 1 37.28-37.22l54.76 61.53-35.83-74.17a191.49 191.49 0 0 1 48.65-20.13l26.87 77.25-4.71-81.61c8.61-1.18 17.39-1.8 26.32-1.8s17.71.62 26.32 1.8l-4.74 82.16 27.05-77.76c17.27 4.5 33.6 11.35 48.63 20.17l-35.82 74.12 54.72-61.47a193.13 193.13 0 0 1 37.24 37.23l-61.45 54.77 74.12-35.86a191.515 191.515 0 0 1 20.2 48.65l-77.81 27.1 82.24-4.75c1.19 8.66 1.82 17.5 1.82 26.49 0 8.88-.61 17.63-1.78 26.19l-82.12-4.75 77.72 27.09z\"],osi:[512,512,[],\"f41a\",\"M8 266.44C10.3 130.64 105.4 34 221.8 18.34c138.8-18.6 255.6 75.8 278 201.1 21.3 118.8-44 230-151.6 274-9.3 3.8-14.4 1.7-18-7.7q-26.7-69.45-53.4-139c-3.1-8.1-1-13.2 7-16.8 24.2-11 39.3-29.4 43.3-55.8a71.47 71.47 0 0 0-64.5-82.2c-39-3.4-71.8 23.7-77.5 59.7-5.2 33 11.1 63.7 41.9 77.7 9.6 4.4 11.5 8.6 7.8 18.4q-26.85 69.9-53.7 139.9c-2.6 6.9-8.3 9.3-15.5 6.5-52.6-20.3-101.4-61-130.8-119-24.9-49.2-25.2-87.7-26.8-108.7zm20.9-1.9c.4 6.6.6 14.3 1.3 22.1 6.3 71.9 49.6 143.5 131 183.1 3.2 1.5 4.4.8 5.6-2.3q22.35-58.65 45-117.3c1.3-3.3.6-4.8-2.4-6.7-31.6-19.9-47.3-48.5-45.6-86 1-21.6 9.3-40.5 23.8-56.3 30-32.7 77-39.8 115.5-17.6a91.64 91.64 0 0 1 45.2 90.4c-3.6 30.6-19.3 53.9-45.7 69.8-2.7 1.6-3.5 2.9-2.3 6q22.8 58.8 45.2 117.7c1.2 3.1 2.4 3.8 5.6 2.3 35.5-16.6 65.2-40.3 88.1-72 34.8-48.2 49.1-101.9 42.3-161-13.7-117.5-119.4-214.8-255.5-198-106.1 13-195.3 102.5-197.1 225.8z\"],\"google-wallet\":[448,512,[],\"f1ee\",\"M156.8 126.8c37.6 60.6 64.2 113.1 84.3 162.5-8.3 33.8-18.8 66.5-31.3 98.3-13.2-52.3-26.5-101.3-56-148.5 6.5-36.4 2.3-73.6 3-112.3zM109.3 200H16.1c-6.5 0-10.5 7.5-6.5 12.7C51.8 267 81.3 330.5 101.3 400h103.5c-16.2-69.7-38.7-133.7-82.5-193.5-3-4-8-6.5-13-6.5zm47.8-88c68.5 108 130 234.5 138.2 368H409c-12-138-68.4-265-143.2-368H157.1zm251.8-68.5c-1.8-6.8-8.2-11.5-15.2-11.5h-88.3c-5.3 0-9 5-7.8 10.3 13.2 46.5 22.3 95.5 26.5 146 48.2 86.2 79.7 178.3 90.6 270.8 15.8-60.5 25.3-133.5 25.3-203 0-73.6-12.1-145.1-31.1-212.6z\"],\"d-and-d-beyond\":[640,512,[],\"f6ca\",\"M313.8 241.5c13.8 0 21-10.1 24.8-17.9-1-1.1-5-4.2-7.4-6.6-2.4 4.3-8.2 10.7-13.9 10.7-10.2 0-15.4-14.7-3.2-26.6-.5-.2-4.3-1.8-8 2.4 0-3 1-5.1 2.1-6.6-3.5 1.3-9.8 5.6-11.4 7.9.2-5.8 1.6-7.5.6-9l-.2-.2s-8.5 5.6-9.3 14.7c0 0 1.1-1.6 2.1-1.9.6-.3 1.3 0 .6 1.9-.2.6-5.8 15.7 5.1 26-.6-1.6-1.9-7.6 2.4-1.9-.3.1 5.8 7.1 15.7 7.1zm52.4-21.1c0-4-4.9-4.4-5.6-4.5 2 3.9.9 7.5.2 9 2.5-.4 5.4-1.6 5.4-4.5zm10.3 5.2c0-6.4-6.2-11.4-13.5-10.7 8 1.3 5.6 13.8-5 11.4 3.7-2.6 3.2-9.9-1.3-12.5 1.4 4.2-3 8.2-7.4 4.6-2.4-1.9-8-6.6-10.6-8.6-2.4-2.1-5.5-1-6.6-1.8-1.3-1.1-.5-3.8-2.2-5-1.6-.8-3-.3-4.8-1-1.6-.6-2.7-1.9-2.6-3.5-2.5 4.4 3.4 6.3 4.5 8.5 1 1.9-.8 4.8 4 8.5 14.8 11.6 9.1 8 10.4 18.1.6 4.3 4.2 6.7 6.4 7.4-2.1-1.9-2.9-6.4 0-9.3 0 13.9 19.2 13.3 23.1 6.4-2.4 1.1-7-.2-9-1.9 7.7 1 14.2-4.1 14.6-10.6zm-39.4-18.4c2 .8 1.6.7 6.4 4.5 10.2-24.5 21.7-15.7 22-15.5 2.2-1.9 9.8-3.8 13.8-2.7-2.4-2.7-7.5-6.2-13.3-6.2-4.7 0-7.4 2.2-8 1.3-.8-1.4 3.2-3.4 3.2-3.4-5.4.2-9.6 6.7-11.2 5.9-1.1-.5 1.4-3.7 1.4-3.7-5.1 2.9-9.3 9.1-10.2 13 4.6-5.8 13.8-9.8 19.7-9-10.5.5-19.5 9.7-23.8 15.8zm242.5 51.9c-20.7 0-40 1.3-50.3 2.1l7.4 8.2v77.2l-7.4 8.2c10.4.8 30.9 2.1 51.6 2.1 42.1 0 59.1-20.7 59.1-48.9 0-29.3-23.2-48.9-60.4-48.9zm-15.1 75.6v-53.3c30.1-3.3 46.8 3.8 46.8 26.3 0 25.6-21.4 30.2-46.8 27zM301.6 181c-1-3.4-.2-6.9 1.1-9.4 1 3 2.6 6.4 7.5 9-.5-2.4-.2-5.6.5-8-1.4-5.4 2.1-9.9 6.4-9.9 6.9 0 8.5 8.8 4.7 14.4 2.1 3.2 5.5 5.6 7.7 7.8 3.2-3.7 5.5-9.5 5.5-13.8 0-8.2-5.5-15.9-16.7-16.5-20-.9-20.2 16.6-20 18.9.5 5.2 3.4 7.8 3.3 7.5zm-.4 6c-.5 1.8-7 3.7-10.2 6.9 4.8-1 7-.2 7.8 1.8.5 1.4-.2 3.4-.5 5.6 1.6-1.8 7-5.5 11-6.2-1-.3-3.4-.8-4.3-.8 2.9-3.4 9.3-4.5 12.8-3.7-2.2-.2-6.7 1.1-8.5 2.6 1.6.3 3 .6 4.3 1.1-2.1.8-4.8 3.4-5.8 6.1 7-5 13.1 5.2 7 8.2.8.2 2.7 0 3.5-.5-.3 1.1-1.9 3-3 3.4 2.9 0 7-1.9 8.2-4.6 0 0-1.8.6-2.6-.2s.3-4.3.3-4.3c-2.3 2.9-3.4-1.3-1.3-4.2-1-.3-3.5-.6-4.6-.5 3.2-1.1 10.4-1.8 11.2-.3.6 1.1-1 3.4-1 3.4 4-.5 8.3 1.1 6.7 5.1 2.9-1.4 5.5-5.9 4.8-10.4-.3 1-1.6 2.4-2.9 2.7.2-1.4-1-2.2-1.9-2.6 1.7-9.6-14.6-14.2-14.1-23.9-1 1.3-1.8 5-.8 7.1 2.7 3.2 8.7 6.7 10.1 12.2-2.6-6.4-15.1-11.4-14.6-20.2-1.6 1.6-2.6 7.8-1.3 11 2.4 1.4 4.5 3.8 4.8 6.1-2.2-5.1-11.4-6.1-13.9-12.2-.6 2.2-.3 5 1 6.7 0 0-2.2-.8-7-.6 1.7.6 5.1 3.5 4.8 5.2zm25.9 7.4c-2.7 0-3.5-2.1-4.2-4.3 3.3 1.3 4.2 4.3 4.2 4.3zm38.9 3.7l-1-.6c-1.1-1-2.9-1.4-4.7-1.4-2.9 0-5.8 1.3-7.5 3.4-.8.8-1.4 1.8-2.1 2.6v15.7c3.5 2.6 7.1-2.9 3-7.2 1.5.3 4.6 2.7 5.1 3.2 0 0 2.6-.5 5-.5 2.1 0 3.9.3 5.6 1.1V196c-1.1.5-2.2 1-2.7 1.4zM79.9 305.9c17.2-4.6 16.2-18 16.2-19.9 0-20.6-24.1-25-37-25H3l8.3 8.6v29.5H0l11.4 14.6V346L3 354.6c61.7 0 73.8 1.5 86.4-5.9 6.7-4 9.9-9.8 9.9-17.6 0-5.1 2.6-18.8-19.4-25.2zm-41.3-27.5c20 0 29.6-.8 29.6 9.1v3c0 12.1-19 8.8-29.6 8.8zm0 59.2V315c12.2 0 32.7-2.3 32.7 8.8v4.5h.2c0 11.2-12.5 9.3-32.9 9.3zm101.2-19.3l23.1.2v-.2l14.1-21.2h-37.2v-14.9h52.4l-14.1-21v-.2l-73.5.2 7.4 8.2v77.1l-7.4 8.2h81.2l14.1-21.2-60.1.2zm214.7-60.1c-73.9 0-77.5 99.3-.3 99.3 77.9 0 74.1-99.3.3-99.3zm-.3 77.5c-37.4 0-36.9-55.3.2-55.3 36.8.1 38.8 55.3-.2 55.3zm-91.3-8.3l44.1-66.2h-41.7l6.1 7.2-20.5 37.2h-.3l-21-37.2 6.4-7.2h-44.9l44.1 65.8.2 19.4-7.7 8.2h42.6l-7.2-8.2zm-28.4-151.3c1.6 1.3 2.9 2.4 2.9 6.6v38.8c0 4.2-.8 5.3-2.7 6.4-.1.1-7.5 4.5-7.9 4.6h35.1c10 0 17.4-1.5 26-8.6-.6-5 .2-9.5.8-12 0-.2-1.8 1.4-2.7 3.5 0-5.7 1.6-15.4 9.6-20.5-.1 0-3.7-.8-9 1.1 2-3.1 10-7.9 10.4-7.9-8.2-26-38-22.9-32.2-22.9-30.9 0-32.6.3-39.9-4 .1.8.5 8.2 9.6 14.9zm21.5 5.5c4.6 0 23.1-3.3 23.1 17.3 0 20.7-18.4 17.3-23.1 17.3zm228.9 79.6l7 8.3V312h-.3c-5.4-14.4-42.3-41.5-45.2-50.9h-31.6l7.4 8.5v76.9l-7.2 8.3h39l-7.4-8.2v-47.4h.3c3.7 10.6 44.5 42.9 48.5 55.6h21.3v-85.2l7.4-8.3zm-106.7-96.1c-32.2 0-32.8.2-39.9-4 .1.7.5 8.3 9.6 14.9 3.1 2 2.9 4.3 2.9 9.5 1.8-1.1 3.8-2.2 6.1-3-1.1 1.1-2.7 2.7-3.5 4.5 1-1.1 7.5-5.1 14.6-3.5-1.6.3-4 1.1-6.1 2.9.1 0 2.1-1.1 7.5-.3v-4.3c4.7 0 23.1-3.4 23.1 17.3 0 20.5-18.5 17.3-19.7 17.3 5.7 4.4 5.8 12 2.2 16.3h.3c33.4 0 36.7-27.3 36.7-34 0-3.8-1.1-32-33.8-33.6z\"],periscope:[448,512,[],\"f3da\",\"M370 63.6C331.4 22.6 280.5 0 226.6 0 111.9 0 18.5 96.2 18.5 214.4c0 75.1 57.8 159.8 82.7 192.7C137.8 455.5 192.6 512 226.6 512c41.6 0 112.9-94.2 120.9-105 24.6-33.1 82-118.3 82-192.6 0-56.5-21.1-110.1-59.5-150.8zM226.6 493.9c-42.5 0-190-167.3-190-279.4 0-107.4 83.9-196.3 190-196.3 100.8 0 184.7 89 184.7 196.3.1 112.1-147.4 279.4-184.7 279.4zM338 206.8c0 59.1-51.1 109.7-110.8 109.7-100.6 0-150.7-108.2-92.9-181.8v.4c0 24.5 20.1 44.4 44.8 44.4 24.7 0 44.8-19.9 44.8-44.4 0-18.2-11.1-33.8-26.9-40.7 76.6-19.2 141 39.3 141 112.4z\"],fulcrum:[320,512,[],\"f50b\",\"M95.75 164.14l-35.38 43.55L25 164.14l35.38-43.55zM144.23 0l-20.54 198.18L72.72 256l51 57.82L144.23 512V300.89L103.15 256l41.08-44.89zm79.67 164.14l35.38 43.55 35.38-43.55-35.38-43.55zm-48.48 47L216.5 256l-41.08 44.89V512L196 313.82 247 256l-51-57.82L175.42 0z\"],cloudscale:[448,512,[],\"f383\",\"M318.1 154l-9.4 7.6c-22.5-19.3-51.5-33.6-83.3-33.6C153.8 128 96 188.8 96 260.3c0 6.6.4 13.1 1.4 19.4-2-56 41.8-97.4 92.6-97.4 24.2 0 46.2 9.4 62.6 24.7l-25.2 20.4c-8.3-.9-16.8 1.8-23.1 8.1-11.1 11-11.1 28.9 0 40 11.1 11 28.9 11 40 0 6.3-6.3 9-14.9 8.1-23.1l75.2-88.8c6.3-6.5-3.3-15.9-9.5-9.6zm-83.8 111.5c-5.6 5.5-14.6 5.5-20.2 0-5.6-5.6-5.6-14.6 0-20.2s14.6-5.6 20.2 0 5.6 14.7 0 20.2zM224 32C100.5 32 0 132.5 0 256s100.5 224 224 224 224-100.5 224-224S347.5 32 224 32zm0 384c-88.2 0-160-71.8-160-160S135.8 96 224 96s160 71.8 160 160-71.8 160-160 160z\"],forumbee:[448,512,[],\"f211\",\"M5.8 309.7C2 292.7 0 275.5 0 258.3 0 135 99.8 35 223.1 35c16.6 0 33.3 2 49.3 5.5C149 87.5 51.9 186 5.8 309.7zm392.9-189.2C385 103 369 87.8 350.9 75.2c-149.6 44.3-266.3 162.1-309.7 312 12.5 18.1 28 35.6 45.2 49 43.1-151.3 161.2-271.7 312.3-315.7zm15.8 252.7c15.2-25.1 25.4-53.7 29.5-82.8-79.4 42.9-145 110.6-187.6 190.3 30-4.4 58.9-15.3 84.6-31.3 35 13.1 70.9 24.3 107 33.6-9.3-36.5-20.4-74.5-33.5-109.8zm29.7-145.5c-2.6-19.5-7.9-38.7-15.8-56.8C290.5 216.7 182 327.5 137.1 466c18.1 7.6 37 12.5 56.6 15.2C240 367.1 330.5 274.4 444.2 227.7z\"],mizuni:[496,512,[],\"f3cc\",\"M248 8C111 8 0 119.1 0 256c0 137 111 248 248 248s248-111 248-248C496 119.1 385 8 248 8zm-80 351.9c-31.4 10.6-58.8 27.3-80 48.2V136c0-22.1 17.9-40 40-40s40 17.9 40 40v223.9zm120-9.9c-12.9-2-26.2-3.1-39.8-3.1-13.8 0-27.2 1.1-40.2 3.1V136c0-22.1 17.9-40 40-40s40 17.9 40 40v214zm120 57.7c-21.2-20.8-48.6-37.4-80-48V136c0-22.1 17.9-40 40-40s40 17.9 40 40v271.7z\"],schlix:[448,512,[],\"f3ea\",\"M350.5 157.7l-54.2-46.1 73.4-39 78.3 44.2-97.5 40.9zM192 122.1l45.7-28.2 34.7 34.6-55.4 29-25-35.4zm-65.1 6.6l31.9-22.1L176 135l-36.7 22.5-12.4-28.8zm-23.3 88.2l-8.8-34.8 29.6-18.3 13.1 35.3-33.9 17.8zm-21.2-83.7l23.9-18.1 8.9 24-26.7 18.3-6.1-24.2zM59 206.5l-3.6-28.4 22.3-15.5 6.1 28.7L59 206.5zm-30.6 16.6l20.8-12.8 3.3 33.4-22.9 12-1.2-32.6zM1.4 268l19.2-10.2.4 38.2-21 8.8L1.4 268zm59.1 59.3l-28.3 8.3-1.6-46.8 25.1-10.7 4.8 49.2zM99 263.2l-31.1 13-5.2-40.8L90.1 221l8.9 42.2zM123.2 377l-41.6 5.9-8.1-63.5 35.2-10.8 14.5 68.4zm28.5-139.9l21.2 57.1-46.2 13.6-13.7-54.1 38.7-16.6zm85.7 230.5l-70.9-3.3-24.3-95.8 55.2-8.6 40 107.7zm-84.9-279.7l42.2-22.4 28 45.9-50.8 21.3-19.4-44.8zm41 94.9l61.3-18.7 52.8 86.6-79.8 11.3-34.3-79.2zm51.4-85.6l67.3-28.8 65.5 65.4-88.6 26.2-44.2-62.8z\"],\"square-xing\":[448,512,[\"xing-square\"],\"f169\",\"M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM93.8 320.2c-5.5 0-8.7-5.3-6-10.3l49.3-86.7c.1 0 .1-.1 0-.2l-31.4-54c-3-5.6 .2-10.1 6-10.1h46.6c5.2 0 9.5 2.9 12.9 8.7l31.9 55.3c-1.3 2.3-18 31.7-50.1 88.2c-3.5 6.2-7.7 9.1-12.6 9.1H93.8zm163.5-33.4v.2l65.5 119c2.8 5.1 .1 10.1-6 10.1H270.2c-5.5 0-9.7-2.9-12.9-8.7l-66-120.3c1.8-3.2 22.9-40.4 63.3-111.6c11.7-20.7 25.1-44.3 40.1-70.8c3.3-5.8 7.4-8.7 12.5-8.7h46.9c5.7-.1 8.8 4.7 6 10L257.3 286.8z\"],bandcamp:[512,512,[],\"f2d5\",\"M256,8C119,8,8,119,8,256S119,504,256,504,504,393,504,256,393,8,256,8Zm48.2,326.1h-181L207.9,178h181Z\"],wpforms:[448,512,[],\"f298\",\"M448 75.2v361.7c0 24.3-19 43.2-43.2 43.2H43.2C19.3 480 0 461.4 0 436.8V75.2C0 51.1 18.8 32 43.2 32h361.7c24 0 43.1 18.8 43.1 43.2zm-37.3 361.6V75.2c0-3-2.6-5.8-5.8-5.8h-9.3L285.3 144 224 94.1 162.8 144 52.5 69.3h-9.3c-3.2 0-5.8 2.8-5.8 5.8v361.7c0 3 2.6 5.8 5.8 5.8h361.7c3.2.1 5.8-2.7 5.8-5.8zM150.2 186v37H76.7v-37h73.5zm0 74.4v37.3H76.7v-37.3h73.5zm11.1-147.3l54-43.7H96.8l64.5 43.7zm210 72.9v37h-196v-37h196zm0 74.4v37.3h-196v-37.3h196zm-84.6-147.3l64.5-43.7H232.8l53.9 43.7zM371.3 335v37.3h-99.4V335h99.4z\"],cloudversify:[616,512,[],\"f385\",\"M148.6 304c8.2 68.5 67.4 115.5 146 111.3 51.2 43.3 136.8 45.8 186.4-5.6 69.2 1.1 118.5-44.6 131.5-99.5 14.8-62.5-18.2-132.5-92.1-155.1-33-88.1-131.4-101.5-186.5-85-57.3 17.3-84.3 53.2-99.3 109.7-7.8 2.7-26.5 8.9-45 24.1 11.7 0 15.2 8.9 15.2 19.5v20.4c0 10.7-8.7 19.5-19.5 19.5h-20.2c-10.7 0-19.5-6-19.5-16.7V240H98.8C95 240 88 244.3 88 251.9v40.4c0 6.4 5.3 11.8 11.7 11.8h48.9zm227.4 8c-10.7 46.3 21.7 72.4 55.3 86.8C324.1 432.6 259.7 348 296 288c-33.2 21.6-33.7 71.2-29.2 92.9-17.9-12.4-53.8-32.4-57.4-79.8-3-39.9 21.5-75.7 57-93.9C297 191.4 369.9 198.7 400 248c-14.1-48-53.8-70.1-101.8-74.8 30.9-30.7 64.4-50.3 114.2-43.7 69.8 9.3 133.2 82.8 67.7 150.5 35-16.3 48.7-54.4 47.5-76.9l10.5 19.6c11.8 22 15.2 47.6 9.4 72-9.2 39-40.6 68.8-79.7 76.5-32.1 6.3-83.1-5.1-91.8-59.2zM128 208H88.2c-8.9 0-16.2-7.3-16.2-16.2v-39.6c0-8.9 7.3-16.2 16.2-16.2H128c8.9 0 16.2 7.3 16.2 16.2v39.6c0 8.9-7.3 16.2-16.2 16.2zM10.1 168C4.5 168 0 163.5 0 157.9v-27.8c0-5.6 4.5-10.1 10.1-10.1h27.7c5.5 0 10.1 4.5 10.1 10.1v27.8c0 5.6-4.5 10.1-10.1 10.1H10.1zM168 142.7v-21.4c0-5.1 4.2-9.3 9.3-9.3h21.4c5.1 0 9.3 4.2 9.3 9.3v21.4c0 5.1-4.2 9.3-9.3 9.3h-21.4c-5.1 0-9.3-4.2-9.3-9.3zM56 235.5v25c0 6.3-5.1 11.5-11.4 11.5H19.4C13.1 272 8 266.8 8 260.5v-25c0-6.3 5.1-11.5 11.4-11.5h25.1c6.4 0 11.5 5.2 11.5 11.5z\"],usps:[576,512,[],\"f7e1\",\"M460.3 241.7c25.8-41.3 15.2-48.8-11.7-48.8h-27c-.1 0-1.5-1.4-10.9 8-11.2 5.6-37.9 6.3-37.9 8.7 0 4.5 70.3-3.1 88.1 0 9.5 1.5-1.5 20.4-4.4 32-.5 4.5 2.4 2.3 3.8.1zm-112.1 22.6c64-21.3 97.3-23.9 102-26.2 4.4-2.9-4.4-6.6-26.2-5.8-51.7 2.2-137.6 37.1-172.6 53.9l-30.7-93.3h196.6c-2.7-28.2-152.9-22.6-337.9-22.6L27 415.8c196.4-97.3 258.9-130.3 321.2-151.5zM94.7 96c253.3 53.7 330 65.7 332.1 85.2 36.4 0 45.9 0 52.4 6.6 21.1 19.7-14.6 67.7-14.6 67.7-4.4 2.9-406.4 160.2-406.4 160.2h423.1L549 96z\"],megaport:[496,512,[],\"f5a3\",\"M214.5 209.6v66.2l33.5 33.5 33.3-33.3v-66.4l-33.4-33.4zM248 8C111 8 0 119 0 256s111 248 248 248 248-111 248-248S385 8 248 8zm145.1 414.4L367 441.6l-26-19.2v-65.5l-33.4-33.4-33.4 33.4v65.5L248 441.6l-26.1-19.2v-65.5l-33.4-33.4-33.5 33.4v65.5l-26.1 19.2-26.1-19.2v-87l59.5-59.5V188l59.5-59.5V52.9l26.1-19.2L274 52.9v75.6l59.5 59.5v87.6l59.7 59.7v87.1z\"],magento:[448,512,[],\"f3c4\",\"M445.7 127.9V384l-63.4 36.5V164.7L223.8 73.1 65.2 164.7l.4 255.9L2.3 384V128.1L224.2 0l221.5 127.9zM255.6 420.5L224 438.9l-31.8-18.2v-256l-63.3 36.6.1 255.9 94.9 54.9 95.1-54.9v-256l-63.4-36.6v255.9z\"],spotify:[496,512,[],\"f1bc\",\"M248 8C111.1 8 0 119.1 0 256s111.1 248 248 248 248-111.1 248-248S384.9 8 248 8zm100.7 364.9c-4.2 0-6.8-1.3-10.7-3.6-62.4-37.6-135-39.2-206.7-24.5-3.9 1-9 2.6-11.9 2.6-9.7 0-15.8-7.7-15.8-15.8 0-10.3 6.1-15.2 13.6-16.8 81.9-18.1 165.6-16.5 237 26.2 6.1 3.9 9.7 7.4 9.7 16.5s-7.1 15.4-15.2 15.4zm26.9-65.6c-5.2 0-8.7-2.3-12.3-4.2-62.5-37-155.7-51.9-238.6-29.4-4.8 1.3-7.4 2.6-11.9 2.6-10.7 0-19.4-8.7-19.4-19.4s5.2-17.8 15.5-20.7c27.8-7.8 56.2-13.6 97.8-13.6 64.9 0 127.6 16.1 177 45.5 8.1 4.8 11.3 11 11.3 19.7-.1 10.8-8.5 19.5-19.4 19.5zm31-76.2c-5.2 0-8.4-1.3-12.9-3.9-71.2-42.5-198.5-52.7-280.9-29.7-3.6 1-8.1 2.6-12.9 2.6-13.2 0-23.3-10.3-23.3-23.6 0-13.6 8.4-21.3 17.4-23.9 35.2-10.3 74.6-15.2 117.5-15.2 73 0 149.5 15.2 205.4 47.8 7.8 4.5 12.9 10.7 12.9 22.6 0 13.6-11 23.3-23.2 23.3z\"],\"optin-monster\":[576,512,[],\"f23c\",\"M572.6 421.4c5.6-9.5 4.7-15.2-5.4-11.6-3-4.9-7-9.5-11.1-13.8 2.9-9.7-.7-14.2-10.8-9.2-4.6-3.2-10.3-6.5-15.9-9.2 0-15.1-11.6-11.6-17.6-5.7-10.4-1.5-18.7-.3-26.8 5.7.3-6.5.3-13 .3-19.7 12.6 0 40.2-11 45.9-36.2 1.4-6.8 1.6-13.8-.3-21.9-3-13.5-14.3-21.3-25.1-25.7-.8-5.9-7.6-14.3-14.9-15.9s-12.4 4.9-14.1 10.3c-8.5 0-19.2 2.8-21.1 8.4-5.4-.5-11.1-1.4-16.8-1.9 2.7-1.9 5.4-3.5 8.4-4.6 5.4-9.2 14.6-11.4 25.7-11.6V256c19.5-.5 43-5.9 53.8-18.1 12.7-13.8 14.6-37.3 12.4-55.1-2.4-17.3-9.7-37.6-24.6-48.1-8.4-5.9-21.6-.8-22.7 9.5-2.2 19.6 1.2 30-38.6 25.1-10.3-23.8-24.6-44.6-42.7-60C341 49.6 242.9 55.5 166.4 71.7c19.7 4.6 41.1 8.6 59.7 16.5-26.2 2.4-52.7 11.3-76.2 23.2-32.8 17-44 29.9-56.7 42.4 14.9-2.2 28.9-5.1 43.8-3.8-9.7 5.4-18.4 12.2-26.5 20-25.8.9-23.8-5.3-26.2-25.9-1.1-10.5-14.3-15.4-22.7-9.7-28.1 19.9-33.5 79.9-12.2 103.5 10.8 12.2 35.1 17.3 54.9 17.8-.3 1.1-.3 1.9-.3 2.7 10.8.5 19.5 2.7 24.6 11.6 3 1.1 5.7 2.7 8.1 4.6-5.4.5-11.1 1.4-16.5 1.9-3.3-6.6-13.7-8.1-21.1-8.1-1.6-5.7-6.5-12.2-14.1-10.3-6.8 1.9-14.1 10-14.9 15.9-22.5 9.5-30.1 26.8-25.1 47.6 5.3 24.8 33 36.2 45.9 36.2v19.7c-6.6-5-14.3-7.5-26.8-5.7-5.5-5.5-17.3-10.1-17.3 5.7-5.9 2.7-11.4 5.9-15.9 9.2-9.8-4.9-13.6-1.7-11.1 9.2-4.1 4.3-7.8 8.6-11.1 13.8-10.2-3.7-11 2.2-5.4 11.6-1.1 3.5-1.6 7-1.9 10.8-.5 31.6 44.6 64 73.5 65.1 17.3.5 34.6-8.4 43-23.5 113.2 4.9 226.7 4.1 340.2 0 8.1 15.1 25.4 24.3 42.7 23.5 29.2-1.1 74.3-33.5 73.5-65.1.2-3.7-.7-7.2-1.7-10.7zm-73.8-254c1.1-3 2.4-8.4 2.4-14.6 0-5.9 6.8-8.1 14.1-.8 11.1 11.6 14.9 40.5 13.8 51.1-4.1-13.6-13-29-30.3-35.7zm-4.6 6.7c19.5 6.2 28.6 27.6 29.7 48.9-1.1 2.7-3 5.4-4.9 7.6-5.7 5.9-15.4 10-26.2 12.2 4.3-21.3.3-47.3-12.7-63 4.9-.8 10.9-2.4 14.1-5.7zm-24.1 6.8c13.8 11.9 20 39.2 14.1 63.5-4.1.5-8.1.8-11.6.8-1.9-21.9-6.8-44-14.3-64.6 3.7.3 8.1.3 11.8.3zM47.5 203c-1.1-10.5 2.4-39.5 13.8-51.1 7-7.3 14.1-5.1 14.1.8 0 6.2 1.4 11.6 2.4 14.6-17.3 6.8-26.2 22.2-30.3 35.7zm9.7 27.6c-1.9-2.2-3.5-4.9-4.9-7.6 1.4-21.3 10.3-42.7 29.7-48.9 3.2 3.2 9.2 4.9 14.1 5.7-13 15.7-17 41.6-12.7 63-10.8-2.2-20.5-6-26.2-12.2zm47.9 14.6c-4.1 0-8.1-.3-12.7-.8-4.6-18.6-1.9-38.9 5.4-53v.3l12.2-5.1c4.9-1.9 9.7-3.8 14.9-4.9-10.7 19.7-17.4 41.3-19.8 63.5zm184-162.7c41.9 0 76.2 34 76.2 75.9 0 42.2-34.3 76.2-76.2 76.2s-76.2-34-76.2-76.2c0-41.8 34.3-75.9 76.2-75.9zm115.6 174.3c-.3 17.8-7 48.9-23 57-13.2 6.6-6.5-7.5-16.5-58.1 13.3.3 26.6.3 39.5 1.1zm-54-1.6c.8 4.9 3.8 40.3-1.6 41.9-11.6 3.5-40 4.3-51.1-1.1-4.1-3-4.6-35.9-4.3-41.1v.3c18.9-.3 38.1-.3 57 0zM278.3 309c-13 3.5-41.6 4.1-54.6-1.6-6.5-2.7-3.8-42.4-1.9-51.6 19.2-.5 38.4-.5 57.8-.8v.3c1.1 8.3 3.3 51.2-1.3 53.7zm-106.5-51.1c12.2-.8 24.6-1.4 36.8-1.6-2.4 15.4-3 43.5-4.9 52.2-1.1 6.8-4.3 6.8-9.7 4.3-21.9-9.8-27.6-35.2-22.2-54.9zm-35.4 31.3c7.8-1.1 15.7-1.9 23.5-2.7 1.6 6.2 3.8 11.9 7 17.6 10 17 44 35.7 45.1 7 6.2 14.9 40.8 12.2 54.9 10.8 15.7-1.4 23.8-1.4 26.8-14.3 12.4 4.3 30.8 4.1 44 3 11.3-.8 20.8-.5 24.6-8.9 1.1 5.1 1.9 11.6 4.6 16.8 10.8 21.3 37.3 1.4 46.8-31.6 8.6.8 17.6 1.9 26.5 2.7-.4 1.3-3.8 7.3 7.3 11.6-47.6 47-95.7 87.8-163.2 107-63.2-20.8-112.1-59.5-155.9-106.5 9.6-3.4 10.4-8.8 8-12.5zm-21.6 172.5c-3.8 17.8-21.9 29.7-39.7 28.9-19.2-.8-46.5-17-59.2-36.5-2.7-31.1 43.8-61.3 66.2-54.6 14.9 4.3 27.8 30.8 33.5 54 0 3-.3 5.7-.8 8.2zm-8.7-66c-.5-13.5-.5-27-.3-40.5h.3c2.7-1.6 5.7-3.8 7.8-6.5 6.5-1.6 13-5.1 15.1-9.2 3.3-7.1-7-7.5-5.4-12.4 2.7-1.1 5.7-2.2 7.8-3.5 29.2 29.2 58.6 56.5 97.3 77-36.8 11.3-72.4 27.6-105.9 47-1.2-18.6-7.7-35.9-16.7-51.9zm337.6 64.6c-103 3.5-206.2 4.1-309.4 0 0 .3 0 .3-.3.3v-.3h.3c35.1-21.6 72.2-39.2 112.4-50.8 11.6 5.1 23 9.5 34.9 13.2 2.2.8 2.2.8 4.3 0 14.3-4.1 28.4-9.2 42.2-15.4 41.5 11.7 78.8 31.7 115.6 53zm10.5-12.4c-35.9-19.5-73-35.9-111.9-47.6 38.1-20 71.9-47.3 103.5-76.7 2.2 1.4 4.6 2.4 7.6 3.2 0 .8.3 1.9.5 2.4-4.6 2.7-7.8 6.2-5.9 10.3 2.2 3.8 8.6 7.6 15.1 8.9 2.4 2.7 5.1 5.1 8.1 6.8 0 13.8-.3 27.6-.8 41.3l.3-.3c-9.3 15.9-15.5 37-16.5 51.7zm105.9 6.2c-12.7 19.5-40 35.7-59.2 36.5-19.3.9-40.5-13.2-40.5-37 5.7-23.2 18.9-49.7 33.5-54 22.7-6.9 69.2 23.4 66.2 54.5zM372.9 75.2c-3.8-72.1-100.8-79.7-126-23.5 44.6-24.3 90.3-15.7 126 23.5zM74.8 407.1c-15.7 1.6-49.5 25.4-49.5 43.2 0 11.6 15.7 19.5 32.2 14.9 12.2-3.2 31.1-17.6 35.9-27.3 6-11.6-3.7-32.7-18.6-30.8zm215.9-176.2c28.6 0 51.9-21.6 51.9-48.4 0-36.1-40.5-58.1-72.2-44.3 9.5 3 16.5 11.6 16.5 21.6 0 23.3-33.3 32-46.5 11.3-7.3 34.1 19.4 59.8 50.3 59.8zM68 474.1c.5 6.5 12.2 12.7 21.6 9.5 6.8-2.7 14.6-10.5 17.3-16.2 3-7-1.1-20-9.7-18.4-8.9 1.6-29.7 16.7-29.2 25.1zm433.2-67c-14.9-1.9-24.6 19.2-18.9 30.8 4.9 9.7 24.1 24.1 36.2 27.3 16.5 4.6 32.2-3.2 32.2-14.9 0-17.8-33.8-41.6-49.5-43.2zM478.8 449c-8.4-1.6-12.4 11.3-9.5 18.4 2.4 5.7 10.3 13.5 17.3 16.2 9.2 3.2 21.1-3 21.3-9.5.9-8.4-20.2-23.5-29.1-25.1z\"],fly:[384,512,[],\"f417\",\"M197.8 427.8c12.9 11.7 33.7 33.3 33.2 50.7 0 .8-.1 1.6-.1 2.5-1.8 19.8-18.8 31.1-39.1 31-25-.1-39.9-16.8-38.7-35.8 1-16.2 20.5-36.7 32.4-47.6 2.3-2.1 2.7-2.7 5.6-3.6 3.4 0 3.9.3 6.7 2.8zM331.9 67.3c-16.3-25.7-38.6-40.6-63.3-52.1C243.1 4.5 214-.2 192 0c-44.1 0-71.2 13.2-81.1 17.3C57.3 45.2 26.5 87.2 28 158.6c7.1 82.2 97 176 155.8 233.8 1.7 1.6 4.5 4.5 6.2 5.1l3.3.1c2.1-.7 1.8-.5 3.5-2.1 52.3-49.2 140.7-145.8 155.9-215.7 7-39.2 3.1-72.5-20.8-112.5zM186.8 351.9c-28-51.1-65.2-130.7-69.3-189-3.4-47.5 11.4-131.2 69.3-136.7v325.7zM328.7 180c-16.4 56.8-77.3 128-118.9 170.3C237.6 298.4 275 217 277 158.4c1.6-45.9-9.8-105.8-48-131.4 88.8 18.3 115.5 98.1 99.7 153z\"],aviato:[640,512,[],\"f421\",\"M107.2 283.5l-19-41.8H36.1l-19 41.8H0l62.2-131.4 62.2 131.4h-17.2zm-45-98.1l-19.6 42.5h39.2l-19.6-42.5zm112.7 102.4l-62.2-131.4h17.1l45.1 96 45.1-96h17l-62.1 131.4zm80.6-4.3V156.4H271v127.1h-15.5zm209.1-115.6v115.6h-17.3V167.9h-41.2v-11.5h99.6v11.5h-41.1zM640 218.8c0 9.2-1.7 17.8-5.1 25.8-3.4 8-8.2 15.1-14.2 21.1-6 6-13.1 10.8-21.1 14.2-8 3.4-16.6 5.1-25.8 5.1s-17.8-1.7-25.8-5.1c-8-3.4-15.1-8.2-21.1-14.2-6-6-10.8-13-14.2-21.1-3.4-8-5.1-16.6-5.1-25.8s1.7-17.8 5.1-25.8c3.4-8 8.2-15.1 14.2-21.1 6-6 13-8.4 21.1-11.9 8-3.4 16.6-5.1 25.8-5.1s17.8 1.7 25.8 5.1c8 3.4 15.1 5.8 21.1 11.9 6 6 10.7 13.1 14.2 21.1 3.4 8 5.1 16.6 5.1 25.8zm-15.5 0c0-7.3-1.3-14-3.9-20.3-2.6-6.3-6.2-11.7-10.8-16.3-4.6-4.6-10-8.2-16.2-10.9-6.2-2.7-12.8-4-19.8-4s-13.6 1.3-19.8 4c-6.2 2.7-11.6 6.3-16.2 10.9-4.6 4.6-8.2 10-10.8 16.3-2.6 6.3-3.9 13.1-3.9 20.3 0 7.3 1.3 14 3.9 20.3 2.6 6.3 6.2 11.7 10.8 16.3 4.6 4.6 10 8.2 16.2 10.9 6.2 2.7 12.8 4 19.8 4s13.6-1.3 19.8-4c6.2-2.7 11.6-6.3 16.2-10.9 4.6-4.6 8.2-10 10.8-16.3 2.6-6.3 3.9-13.1 3.9-20.3zm-94.8 96.7v-6.3l88.9-10-242.9 13.4c.6-2.2 1.1-4.6 1.4-7.2.3-2 .5-4.2.6-6.5l64.8-8.1-64.9 1.9c0-.4-.1-.7-.1-1.1-2.8-17.2-25.5-23.7-25.5-23.7l-1.1-26.3h23.8l19 41.8h17.1L348.6 152l-62.2 131.4h17.1l19-41.8h23.6L345 268s-22.7 6.5-25.5 23.7c-.1.3-.1.7-.1 1.1l-64.9-1.9 64.8 8.1c.1 2.3.3 4.4.6 6.5.3 2.6.8 5 1.4 7.2L78.4 299.2l88.9 10v6.3c-5.9.9-10.5 6-10.5 12.2 0 6.8 5.6 12.4 12.4 12.4 6.8 0 12.4-5.6 12.4-12.4 0-6.2-4.6-11.3-10.5-12.2v-5.8l80.3 9v5.4c-5.7 1.1-9.9 6.2-9.9 12.1 0 6.8 5.6 10.2 12.4 10.2 6.8 0 12.4-3.4 12.4-10.2 0-6-4.3-11-9.9-12.1v-4.9l28.4 3.2v23.7h-5.9V360h5.9v-6.6h5v6.6h5.9v-13.8h-5.9V323l38.3 4.3c8.1 11.4 19 13.6 19 13.6l-.1 6.7-5.1.2-.1 12.1h4.1l.1-5h5.2l.1 5h4.1l-.1-12.1-5.1-.2-.1-6.7s10.9-2.1 19-13.6l38.3-4.3v23.2h-5.9V360h5.9v-6.6h5v6.6h5.9v-13.8h-5.9v-23.7l28.4-3.2v4.9c-5.7 1.1-9.9 6.2-9.9 12.1 0 6.8 5.6 10.2 12.4 10.2 6.8 0 12.4-3.4 12.4-10.2 0-6-4.3-11-9.9-12.1v-5.4l80.3-9v5.8c-5.9.9-10.5 6-10.5 12.2 0 6.8 5.6 12.4 12.4 12.4 6.8 0 12.4-5.6 12.4-12.4-.2-6.3-4.7-11.4-10.7-12.3zm-200.8-87.6l19.6-42.5 19.6 42.5h-17.9l-1.7-40.3-1.7 40.3h-17.9z\"],itunes:[448,512,[],\"f3b4\",\"M223.6 80.3C129 80.3 52.5 157 52.5 251.5S129 422.8 223.6 422.8s171.2-76.7 171.2-171.2c0-94.6-76.7-171.3-171.2-171.3zm79.4 240c-3.2 13.6-13.5 21.2-27.3 23.8-12.1 2.2-22.2 2.8-31.9-5-11.8-10-12-26.4-1.4-36.8 8.4-8 20.3-9.6 38-12.8 3-.5 5.6-1.2 7.7-3.7 3.2-3.6 2.2-2 2.2-80.8 0-5.6-2.7-7.1-8.4-6.1-4 .7-91.9 17.1-91.9 17.1-5 1.1-6.7 2.6-6.7 8.3 0 116.1.5 110.8-1.2 118.5-2.1 9-7.6 15.8-14.9 19.6-8.3 4.6-23.4 6.6-31.4 5.2-21.4-4-28.9-28.7-14.4-42.9 8.4-8 20.3-9.6 38-12.8 3-.5 5.6-1.2 7.7-3.7 5-5.7.9-127 2.6-133.7.4-2.6 1.5-4.8 3.5-6.4 2.1-1.7 5.8-2.7 6.7-2.7 101-19 113.3-21.4 115.1-21.4 5.7-.4 9 3 9 8.7-.1 170.6.4 161.4-1 167.6zM345.2 32H102.8C45.9 32 0 77.9 0 134.8v242.4C0 434.1 45.9 480 102.8 480h242.4c57 0 102.8-45.9 102.8-102.8V134.8C448 77.9 402.1 32 345.2 32zM223.6 444c-106.3 0-192.5-86.2-192.5-192.5S117.3 59 223.6 59s192.5 86.2 192.5 192.5S329.9 444 223.6 444z\"],cuttlefish:[440,512,[],\"f38c\",\"M344 305.5c-17.5 31.6-57.4 54.5-96 54.5-56.6 0-104-47.4-104-104s47.4-104 104-104c38.6 0 78.5 22.9 96 54.5 13.7-50.9 41.7-93.3 87-117.8C385.7 39.1 320.5 8 248 8 111 8 0 119 0 256s111 248 248 248c72.5 0 137.7-31.1 183-80.7-45.3-24.5-73.3-66.9-87-117.8z\"],blogger:[448,512,[],\"f37c\",\"M162.4 196c4.8-4.9 6.2-5.1 36.4-5.1 27.2 0 28.1.1 32.1 2.1 5.8 2.9 8.3 7 8.3 13.6 0 5.9-2.4 10-7.6 13.4-2.8 1.8-4.5 1.9-31.1 2.1-16.4.1-29.5-.2-31.5-.8-10.3-2.9-14.1-17.7-6.6-25.3zm61.4 94.5c-53.9 0-55.8.2-60.2 4.1-3.5 3.1-5.7 9.4-5.1 13.9.7 4.7 4.8 10.1 9.2 12 2.2 1 14.1 1.7 56.3 1.2l47.9-.6 9.2-1.5c9-5.1 10.5-17.4 3.1-24.4-5.3-4.7-5-4.7-60.4-4.7zm223.4 130.1c-3.5 28.4-23 50.4-51.1 57.5-7.2 1.8-9.7 1.9-172.9 1.8-157.8 0-165.9-.1-172-1.8-8.4-2.2-15.6-5.5-22.3-10-5.6-3.8-13.9-11.8-17-16.4-3.8-5.6-8.2-15.3-10-22C.1 423 0 420.3 0 256.3 0 93.2 0 89.7 1.8 82.6 8.1 57.9 27.7 39 53 33.4c7.3-1.6 332.1-1.9 340-.3 21.2 4.3 37.9 17.1 47.6 36.4 7.7 15.3 7-1.5 7.3 180.6.2 115.8 0 164.5-.7 170.5zm-85.4-185.2c-1.1-5-4.2-9.6-7.7-11.5-1.1-.6-8-1.3-15.5-1.7-12.4-.6-13.8-.8-17.8-3.1-6.2-3.6-7.9-7.6-8-18.3 0-20.4-8.5-39.4-25.3-56.5-12-12.2-25.3-20.5-40.6-25.1-3.6-1.1-11.8-1.5-39.2-1.8-42.9-.5-52.5.4-67.1 6.2-27 10.7-46.3 33.4-53.4 62.4-1.3 5.4-1.6 14.2-1.9 64.3-.4 62.8 0 72.1 4 84.5 9.7 30.7 37.1 53.4 64.6 58.4 9.2 1.7 122.2 2.1 133.7.5 20.1-2.7 35.9-10.8 50.7-25.9 10.7-10.9 17.4-22.8 21.8-38.5 3.2-10.9 2.9-88.4 1.7-93.9z\"],flickr:[448,512,[],\"f16e\",\"M400 32H48C21.5 32 0 53.5 0 80v352c0 26.5 21.5 48 48 48h352c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48zM144.5 319c-35.1 0-63.5-28.4-63.5-63.5s28.4-63.5 63.5-63.5 63.5 28.4 63.5 63.5-28.4 63.5-63.5 63.5zm159 0c-35.1 0-63.5-28.4-63.5-63.5s28.4-63.5 63.5-63.5 63.5 28.4 63.5 63.5-28.4 63.5-63.5 63.5z\"],viber:[512,512,[],\"f409\",\"M444 49.9C431.3 38.2 379.9.9 265.3.4c0 0-135.1-8.1-200.9 52.3C27.8 89.3 14.9 143 13.5 209.5c-1.4 66.5-3.1 191.1 117 224.9h.1l-.1 51.6s-.8 20.9 13 25.1c16.6 5.2 26.4-10.7 42.3-27.8 8.7-9.4 20.7-23.2 29.8-33.7 82.2 6.9 145.3-8.9 152.5-11.2 16.6-5.4 110.5-17.4 125.7-142 15.8-128.6-7.6-209.8-49.8-246.5zM457.9 287c-12.9 104-89 110.6-103 115.1-6 1.9-61.5 15.7-131.2 11.2 0 0-52 62.7-68.2 79-5.3 5.3-11.1 4.8-11-5.7 0-6.9.4-85.7.4-85.7-.1 0-.1 0 0 0-101.8-28.2-95.8-134.3-94.7-189.8 1.1-55.5 11.6-101 42.6-131.6 55.7-50.5 170.4-43 170.4-43 96.9.4 143.3 29.6 154.1 39.4 35.7 30.6 53.9 103.8 40.6 211.1zm-139-80.8c.4 8.6-12.5 9.2-12.9.6-1.1-22-11.4-32.7-32.6-33.9-8.6-.5-7.8-13.4.7-12.9 27.9 1.5 43.4 17.5 44.8 46.2zm20.3 11.3c1-42.4-25.5-75.6-75.8-79.3-8.5-.6-7.6-13.5.9-12.9 58 4.2 88.9 44.1 87.8 92.5-.1 8.6-13.1 8.2-12.9-.3zm47 13.4c.1 8.6-12.9 8.7-12.9.1-.6-81.5-54.9-125.9-120.8-126.4-8.5-.1-8.5-12.9 0-12.9 73.7.5 133 51.4 133.7 139.2zM374.9 329v.2c-10.8 19-31 40-51.8 33.3l-.2-.3c-21.1-5.9-70.8-31.5-102.2-56.5-16.2-12.8-31-27.9-42.4-42.4-10.3-12.9-20.7-28.2-30.8-46.6-21.3-38.5-26-55.7-26-55.7-6.7-20.8 14.2-41 33.3-51.8h.2c9.2-4.8 18-3.2 23.9 3.9 0 0 12.4 14.8 17.7 22.1 5 6.8 11.7 17.7 15.2 23.8 6.1 10.9 2.3 22-3.7 26.6l-12 9.6c-6.1 4.9-5.3 14-5.3 14s17.8 67.3 84.3 84.3c0 0 9.1.8 14-5.3l9.6-12c4.6-6 15.7-9.8 26.6-3.7 14.7 8.3 33.4 21.2 45.8 32.9 7 5.7 8.6 14.4 3.8 23.6z\"],soundcloud:[640,512,[],\"f1be\",\"M111.4 256.3l5.8 65-5.8 68.3c-.3 2.5-2.2 4.4-4.4 4.4s-4.2-1.9-4.2-4.4l-5.6-68.3 5.6-65c0-2.2 1.9-4.2 4.2-4.2 2.2 0 4.1 2 4.4 4.2zm21.4-45.6c-2.8 0-4.7 2.2-5 5l-5 105.6 5 68.3c.3 2.8 2.2 5 5 5 2.5 0 4.7-2.2 4.7-5l5.8-68.3-5.8-105.6c0-2.8-2.2-5-4.7-5zm25.5-24.1c-3.1 0-5.3 2.2-5.6 5.3l-4.4 130 4.4 67.8c.3 3.1 2.5 5.3 5.6 5.3 2.8 0 5.3-2.2 5.3-5.3l5.3-67.8-5.3-130c0-3.1-2.5-5.3-5.3-5.3zM7.2 283.2c-1.4 0-2.2 1.1-2.5 2.5L0 321.3l4.7 35c.3 1.4 1.1 2.5 2.5 2.5s2.2-1.1 2.5-2.5l5.6-35-5.6-35.6c-.3-1.4-1.1-2.5-2.5-2.5zm23.6-21.9c-1.4 0-2.5 1.1-2.5 2.5l-6.4 57.5 6.4 56.1c0 1.7 1.1 2.8 2.5 2.8s2.5-1.1 2.8-2.5l7.2-56.4-7.2-57.5c-.3-1.4-1.4-2.5-2.8-2.5zm25.3-11.4c-1.7 0-3.1 1.4-3.3 3.3L47 321.3l5.8 65.8c.3 1.7 1.7 3.1 3.3 3.1 1.7 0 3.1-1.4 3.1-3.1l6.9-65.8-6.9-68.1c0-1.9-1.4-3.3-3.1-3.3zm25.3-2.2c-1.9 0-3.6 1.4-3.6 3.6l-5.8 70 5.8 67.8c0 2.2 1.7 3.6 3.6 3.6s3.6-1.4 3.9-3.6l6.4-67.8-6.4-70c-.3-2.2-2-3.6-3.9-3.6zm241.4-110.9c-1.1-.8-2.8-1.4-4.2-1.4-2.2 0-4.2.8-5.6 1.9-1.9 1.7-3.1 4.2-3.3 6.7v.8l-3.3 176.7 1.7 32.5 1.7 31.7c.3 4.7 4.2 8.6 8.9 8.6s8.6-3.9 8.6-8.6l3.9-64.2-3.9-177.5c-.4-3-2-5.8-4.5-7.2zm-26.7 15.3c-1.4-.8-2.8-1.4-4.4-1.4s-3.1.6-4.4 1.4c-2.2 1.4-3.6 3.9-3.6 6.7l-.3 1.7-2.8 160.8s0 .3 3.1 65.6v.3c0 1.7.6 3.3 1.7 4.7 1.7 1.9 3.9 3.1 6.4 3.1 2.2 0 4.2-1.1 5.6-2.5 1.7-1.4 2.5-3.3 2.5-5.6l.3-6.7 3.1-58.6-3.3-162.8c-.3-2.8-1.7-5.3-3.9-6.7zm-111.4 22.5c-3.1 0-5.8 2.8-5.8 6.1l-4.4 140.6 4.4 67.2c.3 3.3 2.8 5.8 5.8 5.8 3.3 0 5.8-2.5 6.1-5.8l5-67.2-5-140.6c-.2-3.3-2.7-6.1-6.1-6.1zm376.7 62.8c-10.8 0-21.1 2.2-30.6 6.1-6.4-70.8-65.8-126.4-138.3-126.4-17.8 0-35 3.3-50.3 9.4-6.1 2.2-7.8 4.4-7.8 9.2v249.7c0 5 3.9 8.6 8.6 9.2h218.3c43.3 0 78.6-35 78.6-78.3.1-43.6-35.2-78.9-78.5-78.9zm-296.7-60.3c-4.2 0-7.5 3.3-7.8 7.8l-3.3 136.7 3.3 65.6c.3 4.2 3.6 7.5 7.8 7.5 4.2 0 7.5-3.3 7.5-7.5l3.9-65.6-3.9-136.7c-.3-4.5-3.3-7.8-7.5-7.8zm-53.6-7.8c-3.3 0-6.4 3.1-6.4 6.7l-3.9 145.3 3.9 66.9c.3 3.6 3.1 6.4 6.4 6.4 3.6 0 6.4-2.8 6.7-6.4l4.4-66.9-4.4-145.3c-.3-3.6-3.1-6.7-6.7-6.7zm26.7 3.4c-3.9 0-6.9 3.1-6.9 6.9L227 321.3l3.9 66.4c.3 3.9 3.1 6.9 6.9 6.9s6.9-3.1 6.9-6.9l4.2-66.4-4.2-141.7c0-3.9-3-6.9-6.9-6.9z\"],digg:[512,512,[],\"f1a6\",\"M81.7 172.3H0v174.4h132.7V96h-51v76.3zm0 133.4H50.9v-92.3h30.8v92.3zm297.2-133.4v174.4h81.8v28.5h-81.8V416H512V172.3H378.9zm81.8 133.4h-30.8v-92.3h30.8v92.3zm-235.6 41h82.1v28.5h-82.1V416h133.3V172.3H225.1v174.4zm51.2-133.3h30.8v92.3h-30.8v-92.3zM153.3 96h51.3v51h-51.3V96zm0 76.3h51.3v174.4h-51.3V172.3z\"],\"tencent-weibo\":[384,512,[],\"f1d5\",\"M72.3 495.8c1.4 19.9-27.6 22.2-29.7 2.9C31 368.8 73.7 259.2 144 185.5c-15.6-34 9.2-77.1 50.6-77.1 30.3 0 55.1 24.6 55.1 55.1 0 44-49.5 70.8-86.9 45.1-65.7 71.3-101.4 169.8-90.5 287.2zM192 .1C66.1.1-12.3 134.3 43.7 242.4 52.4 259.8 79 246.9 70 229 23.7 136.4 91 29.8 192 29.8c75.4 0 136.9 61.4 136.9 136.9 0 90.8-86.9 153.9-167.7 133.1-19.1-4.1-25.6 24.4-6.6 29.1 110.7 23.2 204-60 204-162.3C358.6 74.7 284 .1 192 .1z\"],letterboxd:[640,512,[],\"e62d\",\"M521.3 128C586.9 128 640 181.1 640 246.6s-53.1 118.6-118.7 118.6c-42.5 0-79.7-22.3-100.7-55.8c11.4-18.2 18-39.7 18-62.8s-6.6-44.6-18-62.8l0 0 .8-1.2c20.8-32.3 56.8-53.9 97.9-54.6l2 0zM320 128c42.5 0 79.7 22.3 100.7 55.8c-11.4 18.2-18 39.7-18 62.8s6.6 44.6 18 62.8l0 0-.8 1.2c-20.8 32.3-56.8 53.9-97.9 54.6l-2 0c-42.5 0-79.7-22.3-100.7-55.8c11.4-18.2 18-39.7 18-62.8s-6.6-44.6-18-62.8l0 0 .8-1.2c20.8-32.3 56.8-53.9 97.9-54.6l2 0zm-201.3 0c42.5 0 79.7 22.3 100.7 55.8c-11.4 18.2-18 39.7-18 62.8s6.6 44.6 18 62.8l0 0-.8 1.2c-20.8 32.3-56.8 53.9-97.9 54.6l-2 0C53.1 365.1 0 312.1 0 246.6S53.1 128 118.7 128z\"],symfony:[512,512,[],\"f83d\",\"M256 8C119 8 8 119 8 256s111 248 248 248 248-111 248-248S393 8 256 8zm133.74 143.54c-11.47.41-19.4-6.45-19.77-16.87-.27-9.18 6.68-13.44 6.53-18.85-.23-6.55-10.16-6.82-12.87-6.67-39.78 1.29-48.59 57-58.89 113.85 21.43 3.15 36.65-.72 45.14-6.22 12-7.75-3.34-15.72-1.42-24.56 4-18.16 32.55-19 32 5.3-.36 17.86-25.92 41.81-77.6 35.7-10.76 59.52-18.35 115-58.2 161.72-29 34.46-58.4 39.82-71.58 40.26-24.65.85-41-12.31-41.58-29.84-.56-17 14.45-26.26 24.31-26.59 21.89-.75 30.12 25.67 14.88 34-12.09 9.71.11 12.61 2.05 12.55 10.42-.36 17.34-5.51 22.18-9 24-20 33.24-54.86 45.35-118.35 8.19-49.66 17-78 18.23-82-16.93-12.75-27.08-28.55-49.85-34.72-15.61-4.23-25.12-.63-31.81 7.83-7.92 10-5.29 23 2.37 30.7l12.63 14c15.51 17.93 24 31.87 20.8 50.62-5.06 29.93-40.72 52.9-82.88 39.94-36-11.11-42.7-36.56-38.38-50.62 7.51-24.15 42.36-11.72 34.62 13.6-2.79 8.6-4.92 8.68-6.28 13.07-4.56 14.77 41.85 28.4 51-1.39 4.47-14.52-5.3-21.71-22.25-39.85-28.47-31.75-16-65.49 2.95-79.67C204.23 140.13 251.94 197 262 205.29c37.17-109 100.53-105.46 102.43-105.53 25.16-.81 44.19 10.59 44.83 28.65.25 7.69-4.17 22.59-19.52 23.13z\"],maxcdn:[512,512,[],\"f136\",\"M461.1 442.7h-97.4L415.6 200c2.3-10.2.9-19.5-4.4-25.7-5-6.1-13.7-9.6-24.2-9.6h-49.3l-59.5 278h-97.4l59.5-278h-83.4l-59.5 278H0l59.5-278-44.6-95.4H387c39.4 0 75.3 16.3 98.3 44.9 23.3 28.6 31.8 67.4 23.6 105.9l-47.8 222.6z\"],etsy:[384,512,[],\"f2d7\",\"M384 348c-1.75 10.75-13.75 110-15.5 132-117.879-4.299-219.895-4.743-368.5 0v-25.5c45.457-8.948 60.627-8.019 61-35.25 1.793-72.322 3.524-244.143 0-322-1.029-28.46-12.13-26.765-61-36v-25.5c73.886 2.358 255.933 8.551 362.999-3.75-3.5 38.25-7.75 126.5-7.75 126.5H332C320.947 115.665 313.241 68 277.25 68h-137c-10.25 0-10.75 3.5-10.75 9.75V241.5c58 .5 88.5-2.5 88.5-2.5 29.77-.951 27.56-8.502 40.75-65.251h25.75c-4.407 101.351-3.91 61.829-1.75 160.25H257c-9.155-40.086-9.065-61.045-39.501-61.5 0 0-21.5-2-88-2v139c0 26 14.25 38.25 44.25 38.25H263c63.636 0 66.564-24.996 98.751-99.75H384z\"],\"facebook-messenger\":[512,512,[],\"f39f\",\"M256.55 8C116.52 8 8 110.34 8 248.57c0 72.3 29.71 134.78 78.07 177.94 8.35 7.51 6.63 11.86 8.05 58.23A19.92 19.92 0 0 0 122 502.31c52.91-23.3 53.59-25.14 62.56-22.7C337.85 521.8 504 423.7 504 248.57 504 110.34 396.59 8 256.55 8zm149.24 185.13l-73 115.57a37.37 37.37 0 0 1-53.91 9.93l-58.08-43.47a15 15 0 0 0-18 0l-78.37 59.44c-10.46 7.93-24.16-4.6-17.11-15.67l73-115.57a37.36 37.36 0 0 1 53.91-9.93l58.06 43.46a15 15 0 0 0 18 0l78.41-59.38c10.44-7.98 24.14 4.54 17.09 15.62z\"],audible:[640,512,[],\"f373\",\"M640 199.9v54l-320 200L0 254v-54l320 200 320-200.1zm-194.5 72l47.1-29.4c-37.2-55.8-100.7-92.6-172.7-92.6-72 0-135.5 36.7-172.6 92.4h.3c2.5-2.3 5.1-4.5 7.7-6.7 89.7-74.4 219.4-58.1 290.2 36.3zm-220.1 18.8c16.9-11.9 36.5-18.7 57.4-18.7 34.4 0 65.2 18.4 86.4 47.6l45.4-28.4c-20.9-29.9-55.6-49.5-94.8-49.5-38.9 0-73.4 19.4-94.4 49zM103.6 161.1c131.8-104.3 318.2-76.4 417.5 62.1l.7 1 48.8-30.4C517.1 112.1 424.8 58.1 319.9 58.1c-103.5 0-196.6 53.5-250.5 135.6 9.9-10.5 22.7-23.5 34.2-32.6zm467 32.7z\"],\"think-peaks\":[576,512,[],\"f731\",\"M465.4 409.4l87.1-150.2-32-.3-55.1 95L259.2 0 23 407.4l32 .3L259.2 55.6zm-355.3-44.1h32.1l117.4-202.5L463 511.9l32.5.1-235.8-404.6z\"],bilibili:[512,512,[],\"e3d9\",\"M488.6 104.1C505.3 122.2 513 143.8 511.9 169.8V372.2C511.5 398.6 502.7 420.3 485.4 437.3C468.2 454.3 446.3 463.2 419.9 464H92.02C65.57 463.2 43.81 454.2 26.74 436.8C9.682 419.4 .7667 396.5 0 368.2V169.8C.7667 143.8 9.682 122.2 26.74 104.1C43.81 87.75 65.57 78.77 92.02 78H121.4L96.05 52.19C90.3 46.46 87.42 39.19 87.42 30.4C87.42 21.6 90.3 14.34 96.05 8.603C101.8 2.868 109.1 0 117.9 0C126.7 0 134 2.868 139.8 8.603L213.1 78H301.1L375.6 8.603C381.7 2.868 389.2 0 398 0C406.8 0 414.1 2.868 419.9 8.603C425.6 14.34 428.5 21.6 428.5 30.4C428.5 39.19 425.6 46.46 419.9 52.19L394.6 78L423.9 78C450.3 78.77 471.9 87.75 488.6 104.1H488.6zM449.8 173.8C449.4 164.2 446.1 156.4 439.1 150.3C433.9 144.2 425.1 140.9 416.4 140.5H96.05C86.46 140.9 78.6 144.2 72.47 150.3C66.33 156.4 63.07 164.2 62.69 173.8V368.2C62.69 377.4 65.95 385.2 72.47 391.7C78.99 398.2 86.85 401.5 96.05 401.5H416.4C425.6 401.5 433.4 398.2 439.7 391.7C446 385.2 449.4 377.4 449.8 368.2L449.8 173.8zM185.5 216.5C191.8 222.8 195.2 230.6 195.6 239.7V273C195.2 282.2 191.9 289.9 185.8 296.2C179.6 302.5 171.8 305.7 162.2 305.7C152.6 305.7 144.7 302.5 138.6 296.2C132.5 289.9 129.2 282.2 128.8 273V239.7C129.2 230.6 132.6 222.8 138.9 216.5C145.2 210.2 152.1 206.9 162.2 206.5C171.4 206.9 179.2 210.2 185.5 216.5H185.5zM377 216.5C383.3 222.8 386.7 230.6 387.1 239.7V273C386.7 282.2 383.4 289.9 377.3 296.2C371.2 302.5 363.3 305.7 353.7 305.7C344.1 305.7 336.3 302.5 330.1 296.2C323.1 289.9 320.7 282.2 320.4 273V239.7C320.7 230.6 324.1 222.8 330.4 216.5C336.7 210.2 344.5 206.9 353.7 206.5C362.9 206.9 370.7 210.2 377 216.5H377z\"],erlang:[640,512,[],\"f39d\",\"M87.2 53.5H0v405h100.4c-49.7-52.6-78.8-125.3-78.7-212.1-.1-76.7 24-142.7 65.5-192.9zm238.2 9.7c-45.9.1-85.1 33.5-89.2 83.2h169.9c-1.1-49.7-34.5-83.1-80.7-83.2zm230.7-9.6h.3l-.1-.1zm.3 0c31.4 42.7 48.7 97.5 46.2 162.7.5 6 .5 11.7 0 24.1H230.2c-.2 109.7 38.9 194.9 138.6 195.3 68.5-.3 118-51 151.9-106.1l96.4 48.2c-17.4 30.9-36.5 57.8-57.9 80.8H640v-405z\"],\"x-twitter\":[512,512,[],\"e61b\",\"M389.2 48h70.6L305.6 224.2 487 464H345L233.7 318.6 106.5 464H35.8L200.7 275.5 26.8 48H172.4L272.9 180.9 389.2 48zM364.4 421.8h39.1L151.1 88h-42L364.4 421.8z\"],\"cotton-bureau\":[512,512,[],\"f89e\",\"M474.31 330.41c-23.66 91.85-94.23 144.59-201.9 148.35V429.6c0-48 26.41-74.39 74.39-74.39 62 0 99.2-37.2 99.2-99.21 0-61.37-36.53-98.28-97.38-99.06-33-69.32-146.5-64.65-177.24 0C110.52 157.72 74 194.63 74 256c0 62.13 37.27 99.41 99.4 99.41 48 0 74.55 26.23 74.55 74.39V479c-134.43-5-211.1-85.07-211.1-223 0-141.82 81.35-223.2 223.2-223.2 114.77 0 189.84 53.2 214.69 148.81H500C473.88 71.51 388.22 8 259.82 8 105 8 12 101.19 12 255.82 12 411.14 105.19 504.34 259.82 504c128.27 0 213.87-63.81 239.67-173.59zM357 182.33c41.37 3.45 64.2 29 64.2 73.67 0 48-26.43 74.41-74.4 74.41-28.61 0-49.33-9.59-61.59-27.33 83.06-16.55 75.59-99.67 71.79-120.75zm-81.68 97.36c-2.46-10.34-16.33-87 56.23-97 2.27 10.09 16.52 87.11-56.26 97zM260 132c28.61 0 49 9.67 61.44 27.61-28.36 5.48-49.36 20.59-61.59 43.45-12.23-22.86-33.23-38-61.6-43.45 12.41-17.69 33.27-27.35 61.57-27.35zm-71.52 50.72c73.17 10.57 58.91 86.81 56.49 97-72.41-9.84-59-86.95-56.25-97zM173.2 330.41c-48 0-74.4-26.4-74.4-74.41 0-44.36 22.86-70 64.22-73.67-6.75 37.2-1.38 106.53 71.65 120.75-12.14 17.63-32.84 27.3-61.14 27.3zm53.21 12.39A80.8 80.8 0 0 0 260 309.25c7.77 14.49 19.33 25.54 33.82 33.55a80.28 80.28 0 0 0-33.58 33.83c-8-14.5-19.07-26.23-33.56-33.83z\"],dashcube:[448,512,[],\"f210\",\"M326.6 104H110.4c-51.1 0-91.2 43.3-91.2 93.5V427c0 50.5 40.1 85 91.2 85h227.2c51.1 0 91.2-34.5 91.2-85V0L326.6 104zM153.9 416.5c-17.7 0-32.4-15.1-32.4-32.8V240.8c0-17.7 14.7-32.5 32.4-32.5h140.7c17.7 0 32 14.8 32 32.5v123.5l51.1 52.3H153.9z\"],\"42-group\":[640,512,[\"innosoft\"],\"e080\",\"M320 96V416C341.011 416 361.818 411.861 381.23 403.821C400.641 395.78 418.28 383.995 433.138 369.138C447.995 354.28 459.78 336.641 467.821 317.23C475.861 297.818 480 277.011 480 256C480 234.989 475.861 214.182 467.821 194.771C459.78 175.359 447.995 157.72 433.138 142.863C418.28 128.005 400.641 116.22 381.23 108.179C361.818 100.139 341.011 96 320 96ZM0 256L160.002 416L320.003 256L160.002 96L0 256ZM480 256C480 277.011 484.138 297.818 492.179 317.23C500.219 336.643 512.005 354.28 526.862 369.138C541.72 383.995 559.357 395.781 578.77 403.821C598.182 411.862 618.989 416 640 416V96C597.565 96 556.869 112.858 526.862 142.863C496.857 172.869 480 213.565 480 256Z\"],\"stack-exchange\":[448,512,[],\"f18d\",\"M17.7 332.3h412.7v22c0 37.7-29.3 68-65.3 68h-19L259.3 512v-89.7H83c-36 0-65.3-30.3-65.3-68v-22zm0-23.6h412.7v-85H17.7v85zm0-109.4h412.7v-85H17.7v85zM365 0H83C47 0 17.7 30.3 17.7 67.7V90h412.7V67.7C430.3 30.3 401 0 365 0z\"],elementor:[512,512,[],\"f430\",\"M.361 256C.361 397 114 511 255 511C397 511 511 397 511 256C511 116 397 2.05 255 2.05C114 2.05 .361 116 .361 256zM192 150V363H149V150H192zM234 150H362V193H234V150zM362 235V278H234V235H362zM234 320H362V363H234V320z\"],\"square-pied-piper\":[448,512,[\"pied-piper-square\"],\"e01e\",\"M32 419L0 479.2l.8-328C.8 85.3 54 32 120 32h327.2c-93 28.9-189.9 94.2-253.9 168.6C122.7 282 82.6 338 32 419M448 32S305.2 98.8 261.6 199.1c-23.2 53.6-28.9 118.1-71 158.6-28.9 27.8-69.8 38.2-105.3 56.3-23.2 12-66.4 40.5-84.9 66h328.4c66 0 119.3-53.3 119.3-119.2-.1 0-.1-328.8-.1-328.8z\"],\"creative-commons-nd\":[496,512,[],\"f4eb\",\"M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zm94 144.3v42.5H162.1V197h180.3zm0 79.8v42.5H162.1v-42.5h180.3z\"],palfed:[576,512,[],\"f3d8\",\"M384.9 193.9c0-47.4-55.2-44.2-95.4-29.8-1.3 39.4-2.5 80.7-3 119.8.7 2.8 2.6 6.2 15.1 6.2 36.8 0 83.4-42.8 83.3-96.2zm-194.5 72.2c.2 0 6.5-2.7 11.2-2.7 26.6 0 20.7 44.1-14.4 44.1-21.5 0-37.1-18.1-37.1-43 0-42 42.9-95.6 100.7-126.5 1-12.4 3-22 10.5-28.2 11.2-9 26.6-3.5 29.5 11.1 72.2-22.2 135.2 1 135.2 72 0 77.9-79.3 152.6-140.1 138.2-.1 39.4.9 74.4 2.7 100v.2c.2 3.4.6 12.5-5.3 19.1-9.6 10.6-33.4 10-36.4-22.3-4.1-44.4.2-206.1 1.4-242.5-21.5 15-58.5 50.3-58.5 75.9.2 2.5.4 4 .6 4.6zM8 181.1s-.1 37.4 38.4 37.4h30l22.4 217.2s0 44.3 44.7 44.3h288.9s44.7-.4 44.7-44.3l22.4-217.2h30s38.4 1.2 38.4-37.4c0 0 .1-37.4-38.4-37.4h-30.1c-7.3-25.6-30.2-74.3-119.4-74.3h-28V50.3s-2.7-18.4-21.1-18.4h-85.8s-21.1 0-21.1 18.4v19.1h-28.1s-105 4.2-120.5 74.3h-29S8 142.5 8 181.1z\"],superpowers:[448,512,[],\"f2dd\",\"M448 32c-83.3 11-166.8 22-250 33-92 12.5-163.3 86.7-169 180-3.3 55.5 18 109.5 57.8 148.2L0 480c83.3-11 166.5-22 249.8-33 91.8-12.5 163.3-86.8 168.7-179.8 3.5-55.5-18-109.5-57.7-148.2L448 32zm-79.7 232.3c-4.2 79.5-74 139.2-152.8 134.5-79.5-4.7-140.7-71-136.3-151 4.5-79.2 74.3-139.3 153-134.5 79.3 4.7 140.5 71 136.1 151z\"],resolving:[496,512,[],\"f3e7\",\"M281.2 278.2c46-13.3 49.6-23.5 44-43.4L314 195.5c-6.1-20.9-18.4-28.1-71.1-12.8L54.7 236.8l28.6 98.6 197.9-57.2zM248.5 8C131.4 8 33.2 88.7 7.2 197.5l221.9-63.9c34.8-10.2 54.2-11.7 79.3-8.2 36.3 6.1 52.7 25 61.4 55.2l10.7 37.8c8.2 28.1 1 50.6-23.5 73.6-19.4 17.4-31.2 24.5-61.4 33.2L203 351.8l220.4 27.1 9.7 34.2-48.1 13.3-286.8-37.3 23 80.2c36.8 22 80.3 34.7 126.3 34.7 137 0 248.5-111.4 248.5-248.3C497 119.4 385.5 8 248.5 8zM38.3 388.6L0 256.8c0 48.5 14.3 93.4 38.3 131.8z\"],xbox:[512,512,[],\"f412\",\"M369.9 318.2c44.3 54.3 64.7 98.8 54.4 118.7-7.9 15.1-56.7 44.6-92.6 55.9-29.6 9.3-68.4 13.3-100.4 10.2-38.2-3.7-76.9-17.4-110.1-39C93.3 445.8 87 438.3 87 423.4c0-29.9 32.9-82.3 89.2-142.1 32-33.9 76.5-73.7 81.4-72.6 9.4 2.1 84.3 75.1 112.3 109.5zM188.6 143.8c-29.7-26.9-58.1-53.9-86.4-63.4-15.2-5.1-16.3-4.8-28.7 8.1-29.2 30.4-53.5 79.7-60.3 122.4-5.4 34.2-6.1 43.8-4.2 60.5 5.6 50.5 17.3 85.4 40.5 120.9 9.5 14.6 12.1 17.3 9.3 9.9-4.2-11-.3-37.5 9.5-64 14.3-39 53.9-112.9 120.3-194.4zm311.6 63.5C483.3 127.3 432.7 77 425.6 77c-7.3 0-24.2 6.5-36 13.9-23.3 14.5-41 31.4-64.3 52.8C367.7 197 427.5 283.1 448.2 346c6.8 20.7 9.7 41.1 7.4 52.3-1.7 8.5-1.7 8.5 1.4 4.6 6.1-7.7 19.9-31.3 25.4-43.5 7.4-16.2 15-40.2 18.6-58.7 4.3-22.5 3.9-70.8-.8-93.4zM141.3 43C189 40.5 251 77.5 255.6 78.4c.7.1 10.4-4.2 21.6-9.7 63.9-31.1 94-25.8 107.4-25.2-63.9-39.3-152.7-50-233.9-11.7-23.4 11.1-24 11.9-9.4 11.2z\"],\"square-web-awesome-stroke\":[448,512,[],\"e684\",\"M64 64C46.3 64 32 78.3 32 96l0 320c0 17.7 14.3 32 32 32l320 0c17.7 0 32-14.3 32-32l0-320c0-17.7-14.3-32-32-32L64 64zM0 96C0 60.7 28.7 32 64 32l320 0c35.3 0 64 28.7 64 64l0 320c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64L0 96zm235 81.6L288 224l52.2-10.4c-2.6-3.9-4.2-8.5-4.2-13.6c0-13.3 10.7-24 24-24s24 10.7 24 24c0 13-10.3 23.6-23.2 24L304.5 349.1c-5.2 11.5-16.6 18.9-29.2 18.9l-102.6 0c-12.6 0-24-7.4-29.2-18.9L87.2 224C74.3 223.6 64 213 64 200c0-13.3 10.7-24 24-24s24 10.7 24 24c0 5-1.5 9.7-4.2 13.6L160 224l53.1-46.4c-8.9-4.1-15-13.1-15-23.6c0-14.4 11.6-26 26-26s26 11.6 26 26c0 10.5-6.2 19.5-15.1 23.6z\"],searchengin:[460,512,[],\"f3eb\",\"M220.6 130.3l-67.2 28.2V43.2L98.7 233.5l54.7-24.2v130.3l67.2-209.3zm-83.2-96.7l-1.3 4.7-15.2 52.9C80.6 106.7 52 145.8 52 191.5c0 52.3 34.3 95.9 83.4 105.5v53.6C57.5 340.1 0 272.4 0 191.6c0-80.5 59.8-147.2 137.4-158zm311.4 447.2c-11.2 11.2-23.1 12.3-28.6 10.5-5.4-1.8-27.1-19.9-60.4-44.4-33.3-24.6-33.6-35.7-43-56.7-9.4-20.9-30.4-42.6-57.5-52.4l-9.7-14.7c-24.7 16.9-53 26.9-81.3 28.7l2.1-6.6 15.9-49.5c46.5-11.9 80.9-54 80.9-104.2 0-54.5-38.4-102.1-96-107.1V32.3C254.4 37.4 320 106.8 320 191.6c0 33.6-11.2 64.7-29 90.4l14.6 9.6c9.8 27.1 31.5 48 52.4 57.4s32.2 9.7 56.8 43c24.6 33.2 42.7 54.9 44.5 60.3s.7 17.3-10.5 28.5zm-9.9-17.9c0-4.4-3.6-8-8-8s-8 3.6-8 8 3.6 8 8 8 8-3.6 8-8z\"],tiktok:[448,512,[],\"e07b\",\"M448,209.91a210.06,210.06,0,0,1-122.77-39.25V349.38A162.55,162.55,0,1,1,185,188.31V278.2a74.62,74.62,0,1,0,52.23,71.18V0l88,0a121.18,121.18,0,0,0,1.86,22.17h0A122.18,122.18,0,0,0,381,102.39a121.43,121.43,0,0,0,67,20.14Z\"],\"square-facebook\":[448,512,[\"facebook-square\"],\"f082\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64h98.2V334.2H109.4V256h52.8V222.3c0-87.1 39.4-127.5 125-127.5c16.2 0 44.2 3.2 55.7 6.4V172c-6-.6-16.5-1-29.6-1c-42 0-58.2 15.9-58.2 57.2V256h83.6l-14.4 78.2H255V480H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64z\"],renren:[512,512,[],\"f18b\",\"M214 169.1c0 110.4-61 205.4-147.6 247.4C30 373.2 8 317.7 8 256.6 8 133.9 97.1 32.2 214 12.5v156.6zM255 504c-42.9 0-83.3-11-118.5-30.4C193.7 437.5 239.9 382.9 255 319c15.5 63.9 61.7 118.5 118.8 154.7C338.7 493 298.3 504 255 504zm190.6-87.5C359 374.5 298 279.6 298 169.1V12.5c116.9 19.7 206 121.4 206 244.1 0 61.1-22 116.6-58.4 159.9z\"],linux:[448,512,[],\"f17c\",\"M220.8 123.3c1 .5 1.8 1.7 3 1.7 1.1 0 2.8-.4 2.9-1.5.2-1.4-1.9-2.3-3.2-2.9-1.7-.7-3.9-1-5.5-.1-.4.2-.8.7-.6 1.1.3 1.3 2.3 1.1 3.4 1.7zm-21.9 1.7c1.2 0 2-1.2 3-1.7 1.1-.6 3.1-.4 3.5-1.6.2-.4-.2-.9-.6-1.1-1.6-.9-3.8-.6-5.5.1-1.3.6-3.4 1.5-3.2 2.9.1 1 1.8 1.5 2.8 1.4zM420 403.8c-3.6-4-5.3-11.6-7.2-19.7-1.8-8.1-3.9-16.8-10.5-22.4-1.3-1.1-2.6-2.1-4-2.9-1.3-.8-2.7-1.5-4.1-2 9.2-27.3 5.6-54.5-3.7-79.1-11.4-30.1-31.3-56.4-46.5-74.4-17.1-21.5-33.7-41.9-33.4-72C311.1 85.4 315.7.1 234.8 0 132.4-.2 158 103.4 156.9 135.2c-1.7 23.4-6.4 41.8-22.5 64.7-18.9 22.5-45.5 58.8-58.1 96.7-6 17.9-8.8 36.1-6.2 53.3-6.5 5.8-11.4 14.7-16.6 20.2-4.2 4.3-10.3 5.9-17 8.3s-14 6-18.5 14.5c-2.1 3.9-2.8 8.1-2.8 12.4 0 3.9.6 7.9 1.2 11.8 1.2 8.1 2.5 15.7.8 20.8-5.2 14.4-5.9 24.4-2.2 31.7 3.8 7.3 11.4 10.5 20.1 12.3 17.3 3.6 40.8 2.7 59.3 12.5 19.8 10.4 39.9 14.1 55.9 10.4 11.6-2.6 21.1-9.6 25.9-20.2 12.5-.1 26.3-5.4 48.3-6.6 14.9-1.2 33.6 5.3 55.1 4.1.6 2.3 1.4 4.6 2.5 6.7v.1c8.3 16.7 23.8 24.3 40.3 23 16.6-1.3 34.1-11 48.3-27.9 13.6-16.4 36-23.2 50.9-32.2 7.4-4.5 13.4-10.1 13.9-18.3.4-8.2-4.4-17.3-15.5-29.7zM223.7 87.3c9.8-22.2 34.2-21.8 44-.4 6.5 14.2 3.6 30.9-4.3 40.4-1.6-.8-5.9-2.6-12.6-4.9 1.1-1.2 3.1-2.7 3.9-4.6 4.8-11.8-.2-27-9.1-27.3-7.3-.5-13.9 10.8-11.8 23-4.1-2-9.4-3.5-13-4.4-1-6.9-.3-14.6 2.9-21.8zM183 75.8c10.1 0 20.8 14.2 19.1 33.5-3.5 1-7.1 2.5-10.2 4.6 1.2-8.9-3.3-20.1-9.6-19.6-8.4.7-9.8 21.2-1.8 28.1 1 .8 1.9-.2-5.9 5.5-15.6-14.6-10.5-52.1 8.4-52.1zm-13.6 60.7c6.2-4.6 13.6-10 14.1-10.5 4.7-4.4 13.5-14.2 27.9-14.2 7.1 0 15.6 2.3 25.9 8.9 6.3 4.1 11.3 4.4 22.6 9.3 8.4 3.5 13.7 9.7 10.5 18.2-2.6 7.1-11 14.4-22.7 18.1-11.1 3.6-19.8 16-38.2 14.9-3.9-.2-7-1-9.6-2.1-8-3.5-12.2-10.4-20-15-8.6-4.8-13.2-10.4-14.7-15.3-1.4-4.9 0-9 4.2-12.3zm3.3 334c-2.7 35.1-43.9 34.4-75.3 18-29.9-15.8-68.6-6.5-76.5-21.9-2.4-4.7-2.4-12.7 2.6-26.4v-.2c2.4-7.6.6-16-.6-23.9-1.2-7.8-1.8-15 .9-20 3.5-6.7 8.5-9.1 14.8-11.3 10.3-3.7 11.8-3.4 19.6-9.9 5.5-5.7 9.5-12.9 14.3-18 5.1-5.5 10-8.1 17.7-6.9 8.1 1.2 15.1 6.8 21.9 16l19.6 35.6c9.5 19.9 43.1 48.4 41 68.9zm-1.4-25.9c-4.1-6.6-9.6-13.6-14.4-19.6 7.1 0 14.2-2.2 16.7-8.9 2.3-6.2 0-14.9-7.4-24.9-13.5-18.2-38.3-32.5-38.3-32.5-13.5-8.4-21.1-18.7-24.6-29.9s-3-23.3-.3-35.2c5.2-22.9 18.6-45.2 27.2-59.2 2.3-1.7.8 3.2-8.7 20.8-8.5 16.1-24.4 53.3-2.6 82.4.6-20.7 5.5-41.8 13.8-61.5 12-27.4 37.3-74.9 39.3-112.7 1.1.8 4.6 3.2 6.2 4.1 4.6 2.7 8.1 6.7 12.6 10.3 12.4 10 28.5 9.2 42.4 1.2 6.2-3.5 11.2-7.5 15.9-9 9.9-3.1 17.8-8.6 22.3-15 7.7 30.4 25.7 74.3 37.2 95.7 6.1 11.4 18.3 35.5 23.6 64.6 3.3-.1 7 .4 10.9 1.4 13.8-35.7-11.7-74.2-23.3-84.9-4.7-4.6-4.9-6.6-2.6-6.5 12.6 11.2 29.2 33.7 35.2 59 2.8 11.6 3.3 23.7.4 35.7 16.4 6.8 35.9 17.9 30.7 34.8-2.2-.1-3.2 0-4.2 0 3.2-10.1-3.9-17.6-22.8-26.1-19.6-8.6-36-8.6-38.3 12.5-12.1 4.2-18.3 14.7-21.4 27.3-2.8 11.2-3.6 24.7-4.4 39.9-.5 7.7-3.6 18-6.8 29-32.1 22.9-76.7 32.9-114.3 7.2zm257.4-11.5c-.9 16.8-41.2 19.9-63.2 46.5-13.2 15.7-29.4 24.4-43.6 25.5s-26.5-4.8-33.7-19.3c-4.7-11.1-2.4-23.1 1.1-36.3 3.7-14.2 9.2-28.8 9.9-40.6.8-15.2 1.7-28.5 4.2-38.7 2.6-10.3 6.6-17.2 13.7-21.1.3-.2.7-.3 1-.5.8 13.2 7.3 26.6 18.8 29.5 12.6 3.3 30.7-7.5 38.4-16.3 9-.3 15.7-.9 22.6 5.1 9.9 8.5 7.1 30.3 17.1 41.6 10.6 11.6 14 19.5 13.7 24.6zM173.3 148.7c2 1.9 4.7 4.5 8 7.1 6.6 5.2 15.8 10.6 27.3 10.6 11.6 0 22.5-5.9 31.8-10.8 4.9-2.6 10.9-7 14.8-10.4s5.9-6.3 3.1-6.6-2.6 2.6-6 5.1c-4.4 3.2-9.7 7.4-13.9 9.8-7.4 4.2-19.5 10.2-29.9 10.2s-18.7-4.8-24.9-9.7c-3.1-2.5-5.7-5-7.7-6.9-1.5-1.4-1.9-4.6-4.3-4.9-1.4-.1-1.8 3.7 1.7 6.5z\"],glide:[448,512,[],\"f2a5\",\"M252.8 148.6c0 8.8-1.6 17.7-3.4 26.4-5.8 27.8-11.6 55.8-17.3 83.6-1.4 6.3-8.3 4.9-13.7 4.9-23.8 0-30.5-26-30.5-45.5 0-29.3 11.2-68.1 38.5-83.1 4.3-2.5 9.2-4.2 14.1-4.2 11.4 0 12.3 8.3 12.3 17.9zM448 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h352c26.5 0 48 21.5 48 48zm-64 187c0-5.1-20.8-37.7-25.5-39.5-2.2-.9-7.2-2.3-9.6-2.3-23.1 0-38.7 10.5-58.2 21.5l-.5-.5c4.3-29.4 14.6-57.2 14.6-87.4 0-44.6-23.8-62.7-67.5-62.7-71.7 0-108 70.8-108 123.5 0 54.7 32 85 86.3 85 7.5 0 6.9-.6 6.9 2.3-10.5 80.3-56.5 82.9-56.5 58.9 0-24.4 28-36.5 28.3-38-.2-7.6-29.3-17.2-36.7-17.2-21.1 0-32.7 33-32.7 50.6 0 32.3 20.4 54.7 53.3 54.7 48.2 0 83.4-49.7 94.3-91.7 9.4-37.7 7-39.4 12.3-42.1 20-10.1 35.8-16.8 58.4-16.8 11.1 0 19 2.3 36.7 5.2 1.8.1 4.1-1.7 4.1-3.5z\"],linkedin:[448,512,[],\"f08c\",\"M416 32H31.9C14.3 32 0 46.5 0 64.3v383.4C0 465.5 14.3 480 31.9 480H416c17.6 0 32-14.5 32-32.3V64.3c0-17.8-14.4-32.3-32-32.3zM135.4 416H69V202.2h66.5V416zm-33.2-243c-21.3 0-38.5-17.3-38.5-38.5S80.9 96 102.2 96c21.2 0 38.5 17.3 38.5 38.5 0 21.3-17.2 38.5-38.5 38.5zm282.1 243h-66.4V312c0-24.8-.5-56.7-34.5-56.7-34.6 0-39.9 27-39.9 54.9V416h-66.4V202.2h63.7v29.2h.9c8.9-16.8 30.6-34.5 62.9-34.5 67.2 0 79.7 44.3 79.7 101.9V416z\"],hubspot:[512,512,[],\"f3b2\",\"M267.4 211.6c-25.1 23.7-40.8 57.3-40.8 94.6 0 29.3 9.7 56.3 26 78L203.1 434c-4.4-1.6-9.1-2.5-14-2.5-10.8 0-20.9 4.2-28.5 11.8-7.6 7.6-11.8 17.8-11.8 28.6s4.2 20.9 11.8 28.5c7.6 7.6 17.8 11.6 28.5 11.6 10.8 0 20.9-3.9 28.6-11.6 7.6-7.6 11.8-17.8 11.8-28.5 0-4.2-.6-8.2-1.9-12.1l50-50.2c22 16.9 49.4 26.9 79.3 26.9 71.9 0 130-58.3 130-130.2 0-65.2-47.7-119.2-110.2-128.7V116c17.5-7.4 28.2-23.8 28.2-42.9 0-26.1-20.9-47.9-47-47.9S311.2 47 311.2 73.1c0 19.1 10.7 35.5 28.2 42.9v61.2c-15.2 2.1-29.6 6.7-42.7 13.6-27.6-20.9-117.5-85.7-168.9-124.8 1.2-4.4 2-9 2-13.8C129.8 23.4 106.3 0 77.4 0 48.6 0 25.2 23.4 25.2 52.2c0 28.9 23.4 52.3 52.2 52.3 9.8 0 18.9-2.9 26.8-7.6l163.2 114.7zm89.5 163.6c-38.1 0-69-30.9-69-69s30.9-69 69-69 69 30.9 69 69-30.9 69-69 69z\"],deploydog:[512,512,[],\"f38e\",\"M382.2 136h51.7v239.6h-51.7v-20.7c-19.8 24.8-52.8 24.1-73.8 14.7-26.2-11.7-44.3-38.1-44.3-71.8 0-29.8 14.8-57.9 43.3-70.8 20.2-9.1 52.7-10.6 74.8 12.9V136zm-64.7 161.8c0 18.2 13.6 33.5 33.2 33.5 19.8 0 33.2-16.4 33.2-32.9 0-17.1-13.7-33.2-33.2-33.2-19.6 0-33.2 16.4-33.2 32.6zM188.5 136h51.7v239.6h-51.7v-20.7c-19.8 24.8-52.8 24.1-73.8 14.7-26.2-11.7-44.3-38.1-44.3-71.8 0-29.8 14.8-57.9 43.3-70.8 20.2-9.1 52.7-10.6 74.8 12.9V136zm-64.7 161.8c0 18.2 13.6 33.5 33.2 33.5 19.8 0 33.2-16.4 33.2-32.9 0-17.1-13.7-33.2-33.2-33.2-19.7 0-33.2 16.4-33.2 32.6zM448 96c17.5 0 32 14.4 32 32v256c0 17.5-14.4 32-32 32H64c-17.5 0-32-14.4-32-32V128c0-17.5 14.4-32 32-32h384m0-32H64C28.8 64 0 92.8 0 128v256c0 35.2 28.8 64 64 64h384c35.2 0 64-28.8 64-64V128c0-35.2-28.8-64-64-64z\"],twitch:[512,512,[],\"f1e8\",\"M391.17,103.47H352.54v109.7h38.63ZM285,103H246.37V212.75H285ZM120.83,0,24.31,91.42V420.58H140.14V512l96.53-91.42h77.25L487.69,256V0ZM449.07,237.75l-77.22,73.12H294.61l-67.6,64v-64H140.14V36.58H449.07Z\"],ravelry:[512,512,[],\"f2d9\",\"M498.252,234.223c-1.208-10.34-1.7-20.826-3.746-31a310.306,310.306,0,0,0-9.622-36.6,184.068,184.068,0,0,0-30.874-57.5,251.154,251.154,0,0,0-18.818-21.689,237.362,237.362,0,0,0-47.113-36.116A240.8,240.8,0,0,0,331.356,26.65c-11.018-3.1-22.272-5.431-33.515-7.615-6.78-1.314-13.749-1.667-20.627-2.482-.316-.036-.6-.358-.9-.553q-16.143.009-32.288.006c-2.41.389-4.808.925-7.236,1.15a179.331,179.331,0,0,0-34.256,7.1,221.5,221.5,0,0,0-39.768,16.355,281.385,281.385,0,0,0-38.08,24.158c-6.167,4.61-12.268,9.36-17.974,14.518C96.539,88.494,86.34,97.72,76.785,107.555a243.878,243.878,0,0,0-33.648,43.95,206.488,206.488,0,0,0-20.494,44.6,198.2,198.2,0,0,0-7.691,34.759A201.13,201.13,0,0,0,13.4,266.385a299.716,299.716,0,0,0,4.425,40.24,226.865,226.865,0,0,0,16.73,53.3,210.543,210.543,0,0,0,24,39.528,213.589,213.589,0,0,0,26.358,28.416A251.313,251.313,0,0,0,126.7,458.455a287.831,287.831,0,0,0,55.9,25.277,269.5,269.5,0,0,0,40.641,9.835c6.071,1.01,12.275,1.253,18.412,1.873a4.149,4.149,0,0,1,1.19.56h32.289c2.507-.389,5-.937,7.527-1.143,16.336-1.332,32.107-5.335,47.489-10.717A219.992,219.992,0,0,0,379.1,460.322c9.749-6.447,19.395-13.077,28.737-20.1,5.785-4.348,10.988-9.5,16.3-14.457,3.964-3.7,7.764-7.578,11.51-11.5a232.162,232.162,0,0,0,31.427-41.639c9.542-16.045,17.355-32.905,22.3-50.926,2.859-10.413,4.947-21.045,7.017-31.652,1.032-5.279,1.251-10.723,1.87-16.087.036-.317.358-.6.552-.9V236.005A9.757,9.757,0,0,1,498.252,234.223Zm-161.117-1.15s-16.572-2.98-28.47-2.98c-27.2,0-33.57,14.9-33.57,37.04V360.8H201.582V170.062H275.1v31.931c8.924-26.822,26.771-36.189,62.04-36.189Z\"],mixer:[512,512,[],\"e056\",\"M114.57,76.07a45.71,45.71,0,0,0-67.51-6.41c-17.58,16.18-19,43.52-4.75,62.77l91.78,123L41.76,379.58c-14.23,19.25-13.11,46.59,4.74,62.77A45.71,45.71,0,0,0,114,435.94L242.89,262.7a12.14,12.14,0,0,0,0-14.23ZM470.24,379.58,377.91,255.45l91.78-123c14.22-19.25,12.83-46.59-4.75-62.77a45.71,45.71,0,0,0-67.51,6.41l-128,172.12a12.14,12.14,0,0,0,0,14.23L398,435.94a45.71,45.71,0,0,0,67.51,6.41C483.35,426.17,484.47,398.83,470.24,379.58Z\"],\"square-lastfm\":[448,512,[\"lastfm-square\"],\"f203\",\"M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM210.7 280.8c-1.8-5.5-3.4-10.8-5-15.9c-12.9-41.9-21-68.4-58-68.4c-22.4 0-45.1 16.1-45.1 61.2c0 35.2 18 57.2 43.3 57.2c28.6 0 47.6-21.3 47.6-21.3l11.7 31.9s-19.8 19.4-61.2 19.4c-51.3 0-79.9-30.1-79.9-85.8c0-57.9 28.6-92 82.5-92c67.9 0 79.3 35.3 96.4 88.4c1.4 4.4 2.9 8.9 4.4 13.5c8.8 26.8 24.2 46.2 61.2 46.2c24.9 0 38.1-5.5 38.1-19.1c0-17.5-16.9-21.2-40-26.4c-3.2-.7-6.5-1.4-9.9-2.2c-30.4-7.3-42.5-23.1-42.5-48c0-40 32.3-52.4 65.2-52.4c37.4 0 60.1 13.6 63 46.6l-36.7 4.4c-1.5-15.8-11-22.4-28.6-22.4c-16.1 0-26 7.3-26 19.8c0 11 4.8 17.6 20.9 21.3c2.2 .5 4.5 1 6.7 1.4c31.1 6.5 65.1 13.7 65.1 56.1c.1 36.7-30.7 50.6-76.1 50.6c-63.4 0-85.4-28.6-97.1-64.1z\"],vimeo:[448,512,[],\"f40a\",\"M403.2 32H44.8C20.1 32 0 52.1 0 76.8v358.4C0 459.9 20.1 480 44.8 480h358.4c24.7 0 44.8-20.1 44.8-44.8V76.8c0-24.7-20.1-44.8-44.8-44.8zM377 180.8c-1.4 31.5-23.4 74.7-66 129.4-44 57.2-81.3 85.8-111.7 85.8-18.9 0-34.8-17.4-47.9-52.3-25.5-93.3-36.4-148-57.4-148-2.4 0-10.9 5.1-25.4 15.2l-15.2-19.6c37.3-32.8 72.9-69.2 95.2-71.2 25.2-2.4 40.7 14.8 46.5 51.7 20.7 131.2 29.9 151 67.6 91.6 13.5-21.4 20.8-37.7 21.8-48.9 3.5-33.2-25.9-30.9-45.8-22.4 15.9-52.1 46.3-77.4 91.2-76 33.3.9 49 22.5 47.1 64.7z\"],mendeley:[640,512,[],\"f7b3\",\"M624.6 325.2c-12.3-12.4-29.7-19.2-48.4-17.2-43.3-1-49.7-34.9-37.5-98.8 22.8-57.5-14.9-131.5-87.4-130.8-77.4.7-81.7 82-130.9 82-48.1 0-54-81.3-130.9-82-72.9-.8-110.1 73.3-87.4 130.8 12.2 63.9 5.8 97.8-37.5 98.8-21.2-2.3-37 6.5-53 22.5-19.9 19.7-19.3 94.8 42.6 102.6 47.1 5.9 81.6-42.9 61.2-87.8-47.3-103.7 185.9-106.1 146.5-8.2-.1.1-.2.2-.3.4-26.8 42.8 6.8 97.4 58.8 95.2 52.1 2.1 85.4-52.6 58.8-95.2-.1-.2-.2-.3-.3-.4-39.4-97.9 193.8-95.5 146.5 8.2-4.6 10-6.7 21.3-5.7 33 4.9 53.4 68.7 74.1 104.9 35.2 17.8-14.8 23.1-65.6 0-88.3zm-303.9-19.1h-.6c-43.4 0-62.8-37.5-62.8-62.8 0-34.7 28.2-62.8 62.8-62.8h.6c34.7 0 62.8 28.1 62.8 62.8 0 25-19.2 62.8-62.8 62.8z\"],uniregistry:[384,512,[],\"f404\",\"M192 480c39.5 0 76.2-11.8 106.8-32.2H85.3C115.8 468.2 152.5 480 192 480zm-89.1-193.1v-12.4H0v12.4c0 2.5 0 5 .1 7.4h103.1c-.2-2.4-.3-4.9-.3-7.4zm20.5 57H8.5c2.6 8.5 5.8 16.8 9.6 24.8h138.3c-12.9-5.7-24.1-14.2-33-24.8zm-17.7-34.7H1.3c.9 7.6 2.2 15 3.9 22.3h109.7c-4-6.9-7.2-14.4-9.2-22.3zm-2.8-69.3H0v17.3h102.9zm0-173.2H0v4.9h102.9zm0-34.7H0v2.5h102.9zm0 69.3H0v7.4h102.9zm0 104H0v14.8h102.9zm0-69.3H0v9.9h102.9zm0 34.6H0V183h102.9zm166.2 160.9h109.7c1.8-7.3 3.1-14.7 3.9-22.3H278.3c-2.1 7.9-5.2 15.4-9.2 22.3zm12-185.7H384V136H281.1zm0 37.2H384v-12.4H281.1zm0-74.3H384v-7.4H281.1zm0-76.7v2.5H384V32zm-203 410.9h227.7c11.8-8.7 22.7-18.6 32.2-29.7H44.9c9.6 11 21.4 21 33.2 29.7zm203-371.3H384v-4.9H281.1zm0 148.5H384v-14.8H281.1zM38.8 405.7h305.3c6.7-8.5 12.6-17.6 17.8-27.2H23c5.2 9.6 9.2 18.7 15.8 27.2zm188.8-37.1H367c3.7-8 5.8-16.2 8.5-24.8h-115c-8.8 10.7-20.1 19.2-32.9 24.8zm53.5-81.7c0 2.5-.1 5-.4 7.4h103.1c.1-2.5.2-4.9.2-7.4v-12.4H281.1zm0-29.7H384v-17.3H281.1z\"],figma:[384,512,[],\"f799\",\"M14 95.7924C14 42.8877 56.8878 0 109.793 0H274.161C327.066 0 369.954 42.8877 369.954 95.7924C369.954 129.292 352.758 158.776 326.711 175.897C352.758 193.019 369.954 222.502 369.954 256.002C369.954 308.907 327.066 351.795 274.161 351.795H272.081C247.279 351.795 224.678 342.369 207.666 326.904V415.167C207.666 468.777 163.657 512 110.309 512C57.5361 512 14 469.243 14 416.207C14 382.709 31.1945 353.227 57.2392 336.105C31.1945 318.983 14 289.5 14 256.002C14 222.502 31.196 193.019 57.2425 175.897C31.196 158.776 14 129.292 14 95.7924ZM176.288 191.587H109.793C74.2172 191.587 45.3778 220.427 45.3778 256.002C45.3778 291.44 73.9948 320.194 109.381 320.416C109.518 320.415 109.655 320.415 109.793 320.415H176.288V191.587ZM207.666 256.002C207.666 291.577 236.505 320.417 272.081 320.417H274.161C309.737 320.417 338.576 291.577 338.576 256.002C338.576 220.427 309.737 191.587 274.161 191.587H272.081C236.505 191.587 207.666 220.427 207.666 256.002ZM109.793 351.795C109.655 351.795 109.518 351.794 109.381 351.794C73.9948 352.015 45.3778 380.769 45.3778 416.207C45.3778 451.652 74.6025 480.622 110.309 480.622C146.591 480.622 176.288 451.186 176.288 415.167V351.795H109.793ZM109.793 31.3778C74.2172 31.3778 45.3778 60.2173 45.3778 95.7924C45.3778 131.368 74.2172 160.207 109.793 160.207H176.288V31.3778H109.793ZM207.666 160.207H274.161C309.737 160.207 338.576 131.368 338.576 95.7924C338.576 60.2173 309.737 31.3778 274.161 31.3778H207.666V160.207Z\"],\"creative-commons-remix\":[496,512,[],\"f4ee\",\"M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zm161.7 207.7l4.9 2.2v70c-7.2 3.6-63.4 27.5-67.3 28.8-6.5-1.8-113.7-46.8-137.3-56.2l-64.2 26.6-63.3-27.5v-63.8l59.3-24.8c-.7-.7-.4 5-.4-70.4l67.3-29.7L361 178.5v61.6l49.1 20.3zm-70.4 81.5v-43.8h-.4v-1.8l-113.8-46.5V295l113.8 46.9v-.4l.4.4zm7.5-57.6l39.9-16.4-36.8-15.5-39 16.4 35.9 15.5zm52.3 38.1v-43L355.2 298v43.4l44.3-19z\"],\"cc-amazon-pay\":[576,512,[],\"f42d\",\"M124.7 201.8c.1-11.8 0-23.5 0-35.3v-35.3c0-1.3.4-2 1.4-2.7 11.5-8 24.1-12.1 38.2-11.1 12.5.9 22.7 7 28.1 21.7 3.3 8.9 4.1 18.2 4.1 27.7 0 8.7-.7 17.3-3.4 25.6-5.7 17.8-18.7 24.7-35.7 23.9-11.7-.5-21.9-5-31.4-11.7-.9-.8-1.4-1.6-1.3-2.8zm154.9 14.6c4.6 1.8 9.3 2 14.1 1.5 11.6-1.2 21.9-5.7 31.3-12.5.9-.6 1.3-1.3 1.3-2.5-.1-3.9 0-7.9 0-11.8 0-4-.1-8 0-12 0-1.4-.4-2-1.8-2.2-7-.9-13.9-2.2-20.9-2.9-7-.6-14-.3-20.8 1.9-6.7 2.2-11.7 6.2-13.7 13.1-1.6 5.4-1.6 10.8.1 16.2 1.6 5.5 5.2 9.2 10.4 11.2zM576 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h480c26.5 0 48 21.5 48 48zm-207.5 23.9c.4 1.7.9 3.4 1.6 5.1 16.5 40.6 32.9 81.3 49.5 121.9 1.4 3.5 1.7 6.4.2 9.9-2.8 6.2-4.9 12.6-7.8 18.7-2.6 5.5-6.7 9.5-12.7 11.2-4.2 1.1-8.5 1.3-12.9.9-2.1-.2-4.2-.7-6.3-.8-2.8-.2-4.2 1.1-4.3 4-.1 2.8-.1 5.6 0 8.3.1 4.6 1.6 6.7 6.2 7.5 4.7.8 9.4 1.6 14.2 1.7 14.3.3 25.7-5.4 33.1-17.9 2.9-4.9 5.6-10.1 7.7-15.4 19.8-50.1 39.5-100.3 59.2-150.5.6-1.5 1.1-3 1.3-4.6.4-2.4-.7-3.6-3.1-3.7-5.6-.1-11.1 0-16.7 0-3.1 0-5.3 1.4-6.4 4.3-.4 1.1-.9 2.3-1.3 3.4l-29.1 83.7c-2.1 6.1-4.2 12.1-6.5 18.6-.4-.9-.6-1.4-.8-1.9-10.8-29.9-21.6-59.9-32.4-89.8-1.7-4.7-3.5-9.5-5.3-14.2-.9-2.5-2.7-4-5.4-4-6.4-.1-12.8-.2-19.2-.1-2.2 0-3.3 1.6-2.8 3.7zM242.4 206c1.7 11.7 7.6 20.8 18 26.6 9.9 5.5 20.7 6.2 31.7 4.6 12.7-1.9 23.9-7.3 33.8-15.5.4-.3.8-.6 1.4-1 .5 3.2.9 6.2 1.5 9.2.5 2.6 2.1 4.3 4.5 4.4 4.6.1 9.1.1 13.7 0 2.3-.1 3.8-1.6 4-3.9.1-.8.1-1.6.1-2.3v-88.8c0-3.6-.2-7.2-.7-10.8-1.6-10.8-6.2-19.7-15.9-25.4-5.6-3.3-11.8-5-18.2-5.9-3-.4-6-.7-9.1-1.1h-10c-.8.1-1.6.3-2.5.3-8.2.4-16.3 1.4-24.2 3.5-5.1 1.3-10 3.2-15 4.9-3 1-4.5 3.2-4.4 6.5.1 2.8-.1 5.6 0 8.3.1 4.1 1.8 5.2 5.7 4.1 6.5-1.7 13.1-3.5 19.7-4.8 10.3-1.9 20.7-2.7 31.1-1.2 5.4.8 10.5 2.4 14.1 7 3.1 4 4.2 8.8 4.4 13.7.3 6.9.2 13.9.3 20.8 0 .4-.1.7-.2 1.2-.4 0-.8 0-1.1-.1-8.8-2.1-17.7-3.6-26.8-4.1-9.5-.5-18.9.1-27.9 3.2-10.8 3.8-19.5 10.3-24.6 20.8-4.1 8.3-4.6 17-3.4 25.8zM98.7 106.9v175.3c0 .8 0 1.7.1 2.5.2 2.5 1.7 4.1 4.1 4.2 5.9.1 11.8.1 17.7 0 2.5 0 4-1.7 4.1-4.1.1-.8.1-1.7.1-2.5v-60.7c.9.7 1.4 1.2 1.9 1.6 15 12.5 32.2 16.6 51.1 12.9 17.1-3.4 28.9-13.9 36.7-29.2 5.8-11.6 8.3-24.1 8.7-37 .5-14.3-1-28.4-6.8-41.7-7.1-16.4-18.9-27.3-36.7-30.9-2.7-.6-5.5-.8-8.2-1.2h-7c-1.2.2-2.4.3-3.6.5-11.7 1.4-22.3 5.8-31.8 12.7-2 1.4-3.9 3-5.9 4.5-.1-.5-.3-.8-.4-1.2-.4-2.3-.7-4.6-1.1-6.9-.6-3.9-2.5-5.5-6.4-5.6h-9.7c-5.9-.1-6.9 1-6.9 6.8zM493.6 339c-2.7-.7-5.1 0-7.6 1-43.9 18.4-89.5 30.2-136.8 35.8-14.5 1.7-29.1 2.8-43.7 3.2-26.6.7-53.2-.8-79.6-4.3-17.8-2.4-35.5-5.7-53-9.9-37-8.9-72.7-21.7-106.7-38.8-8.8-4.4-17.4-9.3-26.1-14-3.8-2.1-6.2-1.5-8.2 2.1v1.7c1.2 1.6 2.2 3.4 3.7 4.8 36 32.2 76.6 56.5 122 72.9 21.9 7.9 44.4 13.7 67.3 17.5 14 2.3 28 3.8 42.2 4.5 3 .1 6 .2 9 .4.7 0 1.4.2 2.1.3h17.7c.7-.1 1.4-.3 2.1-.3 14.9-.4 29.8-1.8 44.6-4 21.4-3.2 42.4-8.1 62.9-14.7 29.6-9.6 57.7-22.4 83.4-40.1 2.8-1.9 5.7-3.8 8-6.2 4.3-4.4 2.3-10.4-3.3-11.9zm50.4-27.7c-.8-4.2-4-5.8-7.6-7-5.7-1.9-11.6-2.8-17.6-3.3-11-.9-22-.4-32.8 1.6-12 2.2-23.4 6.1-33.5 13.1-1.2.8-2.4 1.8-3.1 3-.6.9-.7 2.3-.5 3.4.3 1.3 1.7 1.6 3 1.5.6 0 1.2 0 1.8-.1l19.5-2.1c9.6-.9 19.2-1.5 28.8-.8 4.1.3 8.1 1.2 12 2.2 4.3 1.1 6.2 4.4 6.4 8.7.3 6.7-1.2 13.1-2.9 19.5-3.5 12.9-8.3 25.4-13.3 37.8-.3.8-.7 1.7-.8 2.5-.4 2.5 1 4 3.4 3.5 1.4-.3 3-1.1 4-2.1 3.7-3.6 7.5-7.2 10.6-11.2 10.7-13.8 17-29.6 20.7-46.6.7-3 1.2-6.1 1.7-9.1.2-4.7.2-9.6.2-14.5z\"],dropbox:[528,512,[],\"f16b\",\"M264.4 116.3l-132 84.3 132 84.3-132 84.3L0 284.1l132.3-84.3L0 116.3 132.3 32l132.1 84.3zM131.6 395.7l132-84.3 132 84.3-132 84.3-132-84.3zm132.8-111.6l132-84.3-132-83.6L395.7 32 528 116.3l-132.3 84.3L528 284.8l-132.3 84.3-131.3-85z\"],instagram:[448,512,[],\"f16d\",\"M224.1 141c-63.6 0-114.9 51.3-114.9 114.9s51.3 114.9 114.9 114.9S339 319.5 339 255.9 287.7 141 224.1 141zm0 189.6c-41.1 0-74.7-33.5-74.7-74.7s33.5-74.7 74.7-74.7 74.7 33.5 74.7 74.7-33.6 74.7-74.7 74.7zm146.4-194.3c0 14.9-12 26.8-26.8 26.8-14.9 0-26.8-12-26.8-26.8s12-26.8 26.8-26.8 26.8 12 26.8 26.8zm76.1 27.2c-1.7-35.9-9.9-67.7-36.2-93.9-26.2-26.2-58-34.4-93.9-36.2-37-2.1-147.9-2.1-184.9 0-35.8 1.7-67.6 9.9-93.9 36.1s-34.4 58-36.2 93.9c-2.1 37-2.1 147.9 0 184.9 1.7 35.9 9.9 67.7 36.2 93.9s58 34.4 93.9 36.2c37 2.1 147.9 2.1 184.9 0 35.9-1.7 67.7-9.9 93.9-36.2 26.2-26.2 34.4-58 36.2-93.9 2.1-37 2.1-147.8 0-184.8zM398.8 388c-7.8 19.6-22.9 34.7-42.6 42.6-29.5 11.7-99.5 9-132.1 9s-102.7 2.6-132.1-9c-19.6-7.8-34.7-22.9-42.6-42.6-11.7-29.5-9-99.5-9-132.1s-2.6-102.7 9-132.1c7.8-19.6 22.9-34.7 42.6-42.6 29.5-11.7 99.5-9 132.1-9s102.7-2.6 132.1 9c19.6 7.8 34.7 22.9 42.6 42.6 11.7 29.5 9 99.5 9 132.1s2.7 102.7-9 132.1z\"],cmplid:[640,512,[],\"e360\",\"M226.119,388.165a3.816,3.816,0,0,0-2.294-3.5,3.946,3.946,0,0,0-1.629-.385L72.6,384.3a19.243,19.243,0,0,1-17.924-26.025L81.585,255.692a35.72,35.72,0,0,1,32.373-26H262.525a7.07,7.07,0,0,0,6.392-5.194l10.769-41.131a3.849,3.849,0,0,0-2.237-4.937,3.755,3.755,0,0,0-1.377-.261c-.063,0-.126,0-.189.005H127.38a106.8,106.8,0,0,0-96.99,77.1L3.483,358.824A57.469,57.469,0,0,0,57.314,436q1.43,0,2.86-.072H208.742a7.131,7.131,0,0,0,6.391-5.193L225.839,389.6A3.82,3.82,0,0,0,226.119,388.165ZM306.658,81.2a3.861,3.861,0,0,0,.251-1.367A3.813,3.813,0,0,0,303.079,76c-.064,0-.128,0-.192,0h-41A7.034,7.034,0,0,0,255.5,81.2l-21.347,80.915h51.131ZM180.364,368.249H231.5L263.452,245.69H212.321ZM511.853,79.723a3.809,3.809,0,0,0-3.8-3.661c-.058,0-.137,0-.23.007h-41a7.1,7.1,0,0,0-6.584,5.129L368.91,430.634a3.54,3.54,0,0,0-.262,1.335,3.873,3.873,0,0,0,3.864,3.863c.056,0,.112,0,.169,0h41a7.068,7.068,0,0,0,6.392-5.193L511.533,81.2A3.624,3.624,0,0,0,511.853,79.723ZM324.649,384.47h-41a7.2,7.2,0,0,0-6.392,5.194L266.52,430.8a3.662,3.662,0,0,0-.268,1.374A3.783,3.783,0,0,0,270.023,436c.06,0,.166,0,.3-.012h40.905a7.036,7.036,0,0,0,6.391-5.193l10.769-41.131a3.75,3.75,0,0,0-3.445-5.208c-.108,0-.217,0-.326.014Zm311.324-308.4h-41a7.066,7.066,0,0,0-6.392,5.129l-91.46,349.436a4.073,4.073,0,0,0-.229,1.347,3.872,3.872,0,0,0,3.863,3.851c.056,0,.112,0,.169,0h40.968a7.1,7.1,0,0,0,6.392-5.193L639.68,81.2a3.624,3.624,0,0,0,.32-1.475,3.841,3.841,0,0,0-3.821-3.564c-.068,0-.137,0-.206.006ZM371.562,225.236l10.8-41.1a4.369,4.369,0,0,0,.227-1.388,3.869,3.869,0,0,0-3.861-3.842c-.057,0-.113,0-.169,0h-41.1a7.292,7.292,0,0,0-6.391,5.226l-10.834,41.1a4.417,4.417,0,0,0-.26,1.493c0,.069,0,.138,0,.206a3.776,3.776,0,0,0,3.757,3.507c.076,0,.18,0,.3-.012h41.129A7.034,7.034,0,0,0,371.562,225.236Z\"],upwork:[641,512,[],\"e641\",\"M494.7 295.6c-50.3 0-83.5-38.9-92.8-53.9c11.9-95.3 46.8-125.4 92.8-125.4c45.5 0 80.9 36.4 80.9 89.7s-35.4 89.7-80.9 89.7zm0-237.8c-81.9 0-127.8 53.4-141 108.4c-14.9-28-25.9-65.5-34.5-100.3H206v141c0 51.1-23.3 89-68.8 89s-71.6-37.8-71.6-89l.5-141H.8v141c0 41.1 13.3 78.4 37.6 105.1c25 27.5 59.2 41.8 98.8 41.8c78.8 0 133.8-60.4 133.8-146.9V112.1c8.2 31.2 27.8 91.1 65.3 143.6l-35 199.4h66.4l23.1-141.3c7.6 6.3 15.7 12 24.2 17c22.2 14 47.7 21.9 73.9 22.8c0 0 4 .2 6.1 .2c81.2 0 145.9-62.9 145.9-147.8s-64.8-148.1-146-148.1z\"],facebook:[512,512,[62e3],\"f09a\",\"M512 256C512 114.6 397.4 0 256 0S0 114.6 0 256C0 376 82.7 476.8 194.2 504.5V334.2H141.4V256h52.8V222.3c0-87.1 39.4-127.5 125-127.5c16.2 0 44.2 3.2 55.7 6.4V172c-6-.6-16.5-1-29.6-1c-42 0-58.2 15.9-58.2 57.2V256h83.6l-14.4 78.2H287V510.1C413.8 494.8 512 386.9 512 256h0z\"],gripfire:[384,512,[],\"f3ac\",\"M112.5 301.4c0-73.8 105.1-122.5 105.1-203 0-47.1-34-88-39.1-90.4.4 3.3.6 6.7.6 10C179.1 110.1 32 171.9 32 286.6c0 49.8 32.2 79.2 66.5 108.3 65.1 46.7 78.1 71.4 78.1 86.6 0 10.1-4.8 17-4.8 22.3 13.1-16.7 17.4-31.9 17.5-46.4 0-29.6-21.7-56.3-44.2-86.5-16-22.3-32.6-42.6-32.6-69.5zm205.3-39c-12.1-66.8-78-124.4-94.7-130.9l4 7.2c2.4 5.1 3.4 10.9 3.4 17.1 0 44.7-54.2 111.2-56.6 116.7-2.2 5.1-3.2 10.5-3.2 15.8 0 20.1 15.2 42.1 17.9 42.1 2.4 0 56.6-55.4 58.1-87.7 6.4 11.7 9.1 22.6 9.1 33.4 0 41.2-41.8 96.9-41.8 96.9 0 11.6 31.9 53.2 35.5 53.2 1 0 2.2-1.4 3.2-2.4 37.9-39.3 67.3-85 67.3-136.8 0-8-.7-16.2-2.2-24.6z\"],\"jedi-order\":[448,512,[],\"f50e\",\"M398.5 373.6c95.9-122.1 17.2-233.1 17.2-233.1 45.4 85.8-41.4 170.5-41.4 170.5 105-171.5-60.5-271.5-60.5-271.5 96.9 72.7-10.1 190.7-10.1 190.7 85.8 158.4-68.6 230.1-68.6 230.1s-.4-16.9-2.2-85.7c4.3 4.5 34.5 36.2 34.5 36.2l-24.2-47.4 62.6-9.1-62.6-9.1 20.2-55.5-31.4 45.9c-2.2-87.7-7.8-305.1-7.9-306.9v-2.4 1-1 2.4c0 1-5.6 219-7.9 306.9l-31.4-45.9 20.2 55.5-62.6 9.1 62.6 9.1-24.2 47.4 34.5-36.2c-1.8 68.8-2.2 85.7-2.2 85.7s-154.4-71.7-68.6-230.1c0 0-107-118.1-10.1-190.7 0 0-165.5 99.9-60.5 271.5 0 0-86.8-84.8-41.4-170.5 0 0-78.7 111 17.2 233.1 0 0-26.2-16.1-49.4-77.7 0 0 16.9 183.3 222 185.7h4.1c205-2.4 222-185.7 222-185.7-23.6 61.5-49.9 77.7-49.9 77.7z\"],uikit:[448,512,[],\"f403\",\"M443.9 128v256L218 512 0 384V169.7l87.6 45.1v117l133.5 75.5 135.8-75.5v-151l-101.1-57.6 87.6-53.1L443.9 128zM308.6 49.1L223.8 0l-88.6 54.8 86 47.3 87.4-53z\"],\"fort-awesome-alt\":[512,512,[],\"f3a3\",\"M208 237.4h-22.2c-2.1 0-3.7 1.6-3.7 3.7v51.7c0 2.1 1.6 3.7 3.7 3.7H208c2.1 0 3.7-1.6 3.7-3.7v-51.7c0-2.1-1.6-3.7-3.7-3.7zm118.2 0H304c-2.1 0-3.7 1.6-3.7 3.7v51.7c0 2.1 1.6 3.7 3.7 3.7h22.2c2.1 0 3.7-1.6 3.7-3.7v-51.7c-.1-2.1-1.7-3.7-3.7-3.7zm132-125.1c-2.3-3.2-4.6-6.4-7.1-9.5-9.8-12.5-20.8-24-32.8-34.4-4.5-3.9-9.1-7.6-13.9-11.2-1.6-1.2-3.2-2.3-4.8-3.5C372 34.1 340.3 20 306 13c-16.2-3.3-32.9-5-50-5s-33.9 1.7-50 5c-34.3 7.1-66 21.2-93.3 40.8-1.6 1.1-3.2 2.3-4.8 3.5-4.8 3.6-9.4 7.3-13.9 11.2-3 2.6-5.9 5.3-8.8 8s-5.7 5.5-8.4 8.4c-5.5 5.7-10.7 11.8-15.6 18-2.4 3.1-4.8 6.3-7.1 9.5C25.2 153 8.3 202.5 8.3 256c0 2 .1 4 .1 6 .1.7.1 1.3.1 2 .1 1.3.1 2.7.2 4 0 .8.1 1.5.1 2.3 0 1.3.1 2.5.2 3.7.1.8.1 1.6.2 2.4.1 1.1.2 2.3.3 3.5 0 .8.1 1.6.2 2.4.1 1.2.3 2.4.4 3.6.1.8.2 1.5.3 2.3.1 1.3.3 2.6.5 3.9.1.6.2 1.3.3 1.9l.9 5.7c.1.6.2 1.1.3 1.7.3 1.3.5 2.7.8 4 .2.8.3 1.6.5 2.4.2 1 .5 2.1.7 3.2.2.9.4 1.7.6 2.6.2 1 .4 2 .7 3 .2.9.5 1.8.7 2.7.3 1 .5 1.9.8 2.9.3.9.5 1.8.8 2.7.2.9.5 1.9.8 2.8s.5 1.8.8 2.7c.3 1 .6 1.9.9 2.8.6 1.6 1.1 3.3 1.7 4.9.4 1 .7 1.9 1 2.8.3 1 .7 2 1.1 3 .3.8.6 1.5.9 2.3l1.2 3c.3.7.6 1.5.9 2.2.4 1 .9 2 1.3 3l.9 2.1c.5 1 .9 2 1.4 3 .3.7.6 1.3.9 2 .5 1 1 2.1 1.5 3.1.2.6.5 1.1.8 1.7.6 1.1 1.1 2.2 1.7 3.3.1.2.2.3.3.5 2.2 4.1 4.4 8.2 6.8 12.2.2.4.5.8.7 1.2.7 1.1 1.3 2.2 2 3.3.3.5.6.9.9 1.4.6 1.1 1.3 2.1 2 3.2.3.5.6.9.9 1.4.7 1.1 1.4 2.1 2.1 3.2.2.4.5.8.8 1.2.7 1.1 1.5 2.2 2.3 3.3.2.2.3.5.5.7 37.5 51.7 94.4 88.5 160 99.4.9.1 1.7.3 2.6.4 1 .2 2.1.4 3.1.5s1.9.3 2.8.4c1 .2 2 .3 3 .4.9.1 1.9.2 2.9.3s1.9.2 2.9.3 2.1.2 3.1.3c.9.1 1.8.1 2.7.2 1.1.1 2.3.1 3.4.2.8 0 1.7.1 2.5.1 1.3 0 2.6.1 3.9.1.7.1 1.4.1 2.1.1 2 .1 4 .1 6 .1s4-.1 6-.1c.7 0 1.4-.1 2.1-.1 1.3 0 2.6 0 3.9-.1.8 0 1.7-.1 2.5-.1 1.1-.1 2.3-.1 3.4-.2.9 0 1.8-.1 2.7-.2 1-.1 2.1-.2 3.1-.3s1.9-.2 2.9-.3c.9-.1 1.9-.2 2.9-.3s2-.3 3-.4 1.9-.3 2.8-.4c1-.2 2.1-.3 3.1-.5.9-.1 1.7-.3 2.6-.4 65.6-11 122.5-47.7 160.1-102.4.2-.2.3-.5.5-.7.8-1.1 1.5-2.2 2.3-3.3.2-.4.5-.8.8-1.2.7-1.1 1.4-2.1 2.1-3.2.3-.5.6-.9.9-1.4.6-1.1 1.3-2.1 2-3.2.3-.5.6-.9.9-1.4.7-1.1 1.3-2.2 2-3.3.2-.4.5-.8.7-1.2 2.4-4 4.6-8.1 6.8-12.2.1-.2.2-.3.3-.5.6-1.1 1.1-2.2 1.7-3.3.2-.6.5-1.1.8-1.7.5-1 1-2.1 1.5-3.1.3-.7.6-1.3.9-2 .5-1 1-2 1.4-3l.9-2.1c.5-1 .9-2 1.3-3 .3-.7.6-1.5.9-2.2l1.2-3c.3-.8.6-1.5.9-2.3.4-1 .7-2 1.1-3s.7-1.9 1-2.8c.6-1.6 1.2-3.3 1.7-4.9.3-1 .6-1.9.9-2.8s.5-1.8.8-2.7c.2-.9.5-1.9.8-2.8s.6-1.8.8-2.7c.3-1 .5-1.9.8-2.9.2-.9.5-1.8.7-2.7.2-1 .5-2 .7-3 .2-.9.4-1.7.6-2.6.2-1 .5-2.1.7-3.2.2-.8.3-1.6.5-2.4.3-1.3.6-2.7.8-4 .1-.6.2-1.1.3-1.7l.9-5.7c.1-.6.2-1.3.3-1.9.1-1.3.3-2.6.5-3.9.1-.8.2-1.5.3-2.3.1-1.2.3-2.4.4-3.6 0-.8.1-1.6.2-2.4.1-1.1.2-2.3.3-3.5.1-.8.1-1.6.2-2.4.1 1.7.1.5.2-.7 0-.8.1-1.5.1-2.3.1-1.3.2-2.7.2-4 .1-.7.1-1.3.1-2 .1-2 .1-4 .1-6 0-53.5-16.9-103-45.8-143.7zM448 371.5c-9.4 15.5-20.6 29.9-33.6 42.9-20.6 20.6-44.5 36.7-71.2 48-13.9 5.8-28.2 10.3-42.9 13.2v-75.8c0-58.6-88.6-58.6-88.6 0v75.8c-14.7-2.9-29-7.3-42.9-13.2-26.7-11.3-50.6-27.4-71.2-48-13-13-24.2-27.4-33.6-42.9v-71.3c0-2.1 1.6-3.7 3.7-3.7h22.1c2.1 0 3.7 1.6 3.7 3.7V326h29.6V182c0-2.1 1.6-3.7 3.7-3.7h22.1c2.1 0 3.7 1.6 3.7 3.7v25.9h29.5V182c0-2.1 1.6-3.7 3.7-3.7H208c2.1 0 3.7 1.6 3.7 3.7v25.9h29.5V182c0-4.8 6.5-3.7 9.5-3.7V88.1c-4.4-2-7.4-6.7-7.4-11.5 0-16.8 25.4-16.8 25.4 0 0 4.8-3 9.4-7.4 11.5V92c6.3-1.4 12.7-2.3 19.2-2.3 9.4 0 18.4 3.5 26.3 3.5 7.2 0 15.2-3.5 19.4-3.5 2.1 0 3.7 1.6 3.7 3.7v48.4c0 5.6-18.7 6.5-22.4 6.5-8.6 0-16.6-3.5-25.4-3.5-7 0-14.1 1.2-20.8 2.8v30.7c3 0 9.5-1.1 9.5 3.7v25.9h29.5V182c0-2.1 1.6-3.7 3.7-3.7h22.2c2.1 0 3.7 1.6 3.7 3.7v25.9h29.5V182c0-2.1 1.6-3.7 3.7-3.7h22.1c2.1 0 3.7 1.6 3.7 3.7v144h29.5v-25.8c0-2.1 1.6-3.7 3.7-3.7h22.2c2.1 0 3.7 1.6 3.7 3.7z\"],phabricator:[496,512,[],\"f3db\",\"M323 262.1l-.1-13s21.7-19.8 21.1-21.2l-9.5-20c-.6-1.4-29.5-.5-29.5-.5l-9.4-9.3s.2-28.5-1.2-29.1l-20.1-9.2c-1.4-.6-20.7 21-20.7 21l-13.1-.2s-20.5-21.4-21.9-20.8l-20 8.3c-1.4.5.2 28.9.2 28.9l-9.1 9.1s-29.2-.9-29.7.4l-8.1 19.8c-.6 1.4 21 21 21 21l.1 12.9s-21.7 19.8-21.1 21.2l9.5 20c.6 1.4 29.5.5 29.5.5l9.4 9.3s-.2 31.8 1.2 32.3l20.1 8.3c1.4.6 20.7-23.5 20.7-23.5l13.1.2s20.5 23.8 21.8 23.3l20-7.5c1.4-.6-.2-32.1-.2-32.1l9.1-9.1s29.2.9 29.7-.5l8.1-19.8c.7-1.1-20.9-20.7-20.9-20.7zm-44.9-8.7c.7 17.1-12.8 31.6-30.1 32.4-17.3.8-32.1-12.5-32.8-29.6-.7-17.1 12.8-31.6 30.1-32.3 17.3-.8 32.1 12.5 32.8 29.5zm201.2-37.9l-97-97-.1.1c-75.1-73.3-195.4-72.8-269.8 1.6-50.9 51-27.8 27.9-95.7 95.3-22.3 22.3-22.3 58.7 0 81 69.9 69.4 46.4 46 97.4 97l.1-.1c75.1 73.3 195.4 72.9 269.8-1.6 51-50.9 27.9-27.9 95.3-95.3 22.3-22.3 22.3-58.7 0-81zM140.4 363.8c-59.6-59.5-59.6-156 0-215.5 59.5-59.6 156-59.5 215.6 0 59.5 59.5 59.6 156 0 215.6-59.6 59.5-156 59.4-215.6-.1z\"],ussunnah:[482,512,[],\"f407\",\"M481.9 268.1A240.9 240.9 0 1 1 .1 268a240.9 240.9 0 1 1 481.9 0zM24.5 268a216.5 216.5 0 1 0 432.9 0A216.5 216.5 0 1 0 24.5 268zm385.9 63.3c-12.7 0-21.6-1.9-26.7-5.9c-5.5-4.3-8.2-12.3-8.2-23.8V205.1c0-6.5-5.2-20.2-15.7-41.2c7 0 17-9.1 30-27.2V284.5c0 11 2.4 19.4 7 25.3c3.7 4.7 10.1 8.9 19 12.6c1.2 .4 2.6 .9 4.1 1.4c2.9 .9 6.3 2.1 10.3 3.5c-1.8 2.7-8.3 4-19.9 4zm-219 0c-1.3 2.4-3.6 5.5-6.8 9.4l-18.5 22.5c-1-6.1-4-13-9.3-20.6s-9.7-11.4-13.4-11.4h-8.3H53.6c3.3-5.3 4.9-8.8 4.9-10.8c0-2-.8-5.3-2.4-9.7c-1.5-4.4-2.4-8.5-2.4-12.4c0-7.4 2.1-13.9 6.3-19.3L80 253.4l-7.1-17.7L89 215.9l6.7 16.8 8-10.3c-1.8 6.4-2.6 12.3-2.6 17.7c0 4.2 2.8 13.3 8.3 27.3l16.2 40.7H135h8 .3c2.8 .4 7.7 5 14.6 13.9c1.8 2.4 4.3 5.8 7.7 10.2c1.4 1.9 2.9 3.9 4.6 6.1c1.3-2.3 2-4.6 2-7.1c0-2-1.3-6.6-4-13.4L163 304.1c-4-10.6-6.1-17.7-6.1-21.3c0-6.3 1.9-12.3 5.8-17.9c.5-.6 1-1.3 1.5-1.9c4.4-5.6 8.8-11.1 13.3-16.5c-1.1 4.6-1.7 8.7-1.7 12c0 3.7 1.7 9.9 5.1 18.8l7.9 20.4c1.9 4.7 3 8.2 3.7 10.3h17.6 8.3l-.9-2.6c-1.4-3.9-4-7-7.7-9.3l15.6-20.1 12.3 32h13.4L245 292.2c-1.5-3.9-4-7-7.7-9.3L253 262.8 270.3 308h13.4l-11.4-29.4c-1.5-3.9-4-7-7.7-9.3l15.6-20L302.6 308h10.3 8.3 7.6c1.5 0 3-1.1 4.5-3.1s2.2-4.1 2.2-6.3V205.1c0-6.5-4.5-20.3-13.7-41.2c5.4 0 14.1-9.1 26.2-27.2V300.2c0 7.2 .6 12 1.7 14.6c1.6 3.4 5.3 6.2 11.1 8.2c-3.9 5.6-8.7 8.5-14.5 8.5H321.1h-8.3H210.5h-19zM93.4 287.3c-2.7-6.7-4-11.7-4-15c-.6 1.2-2.4 3.7-5.4 7.6c-1.4 1.9-2.2 3.7-2.2 5.3c0 2.6 .8 5.7 2.2 9.3l5.6 13.9h0c5 0 9 0 11.9-.1l-8.2-20.9zm13.5-72.4c-3-5.2-7-9.3-11.9-11.9c-3.5-1.9-5.3-4.3-5.3-7.4c0-2.4 4.6-8.6 14-18.3c.2 3.8 1.9 7.6 4.9 11.2c3.1 3.6 4.6 7 4.6 10.1c0 2.6-2.1 8-6.2 16.3zm-27.6 0c-3-5.2-7-9.3-11.9-11.9c-3.5-1.9-5.3-4.3-5.3-7.4c0-2.4 4.6-8.6 14-18.3c.2 3.8 1.9 7.6 4.9 11.2c3.1 3.6 4.6 7 4.6 10.1c0 2.6-2.1 8-6.2 16.3zm87 27.5c-3-5.2-7-9.3-11.9-11.9c-3.5-1.9-5.3-4.3-5.3-7.4c0-2.4 4.6-8.6 14-18.3c.2 3.8 1.9 7.6 4.9 11.2c3.1 3.6 4.6 7 4.6 10.1c0 2.6-2.1 8-6.2 16.3z\"],earlybirds:[480,512,[],\"f39a\",\"M313.2 47.5c1.2-13 21.3-14 36.6-8.7.9.3 26.2 9.7 19 15.2-27.9-7.4-56.4 18.2-55.6-6.5zm-201 6.9c30.7-8.1 62 20 61.1-7.1-1.3-14.2-23.4-15.3-40.2-9.6-1 .3-28.7 10.5-20.9 16.7zM319.4 160c-8.8 0-16 7.2-16 16s7.2 16 16 16 16-7.2 16-16-7.2-16-16-16zm-159.7 0c-8.8 0-16 7.2-16 16s7.2 16 16 16 16-7.2 16-16-7.2-16-16-16zm318.5 163.2c-9.9 24-40.7 11-63.9-1.2-13.5 69.1-58.1 111.4-126.3 124.2.3.9-2-.1 24 1 33.6 1.4 63.8-3.1 97.4-8-19.8-13.8-11.4-37.1-9.8-38.1 1.4-.9 14.7 1.7 21.6 11.5 8.6-12.5 28.4-14.8 30.2-13.6 1.6 1.1 6.6 20.9-6.9 34.6 4.7-.9 8.2-1.6 9.8-2.1 2.6-.8 17.7 11.3 3.1 13.3-14.3 2.3-22.6 5.1-47.1 10.8-45.9 10.7-85.9 11.8-117.7 12.8l1 11.6c3.8 18.1-23.4 24.3-27.6 6.2.8 17.9-27.1 21.8-28.4-1l-.5 5.3c-.7 18.4-28.4 17.9-28.3-.6-7.5 13.5-28.1 6.8-26.4-8.5l1.2-12.4c-36.7.9-59.7 3.1-61.8 3.1-20.9 0-20.9-31.6 0-31.6 2.4 0 27.7 1.3 63.2 2.8-61.1-15.5-103.7-55-114.9-118.2-25 12.8-57.5 26.8-68.2.8-10.5-25.4 21.5-42.6 66.8-73.4.7-6.6 1.6-13.3 2.7-19.8-14.4-19.6-11.6-36.3-16.1-60.4-16.8 2.4-23.2-9.1-23.6-23.1.3-7.3 2.1-14.9 2.4-15.4 1.1-1.8 10.1-2 12.7-2.6 6-31.7 50.6-33.2 90.9-34.5 19.7-21.8 45.2-41.5 80.9-48.3C203.3 29 215.2 8.5 216.2 8c1.7-.8 21.2 4.3 26.3 23.2 5.2-8.8 18.3-11.4 19.6-10.7 1.1.6 6.4 15-4.9 25.9 40.3 3.5 72.2 24.7 96 50.7 36.1 1.5 71.8 5.9 77.1 34 2.7.6 11.6.8 12.7 2.6.3.5 2.1 8.1 2.4 15.4-.5 13.9-6.8 25.4-23.6 23.1-3.2 17.3-2.7 32.9-8.7 47.7 2.4 11.7 4 23.8 4.8 36.4 37 25.4 70.3 42.5 60.3 66.9zM207.4 159.9c.9-44-37.9-42.2-78.6-40.3-21.7 1-38.9 1.9-45.5 13.9-11.4 20.9 5.9 92.9 23.2 101.2 9.8 4.7 73.4 7.9 86.3-7.1 8.2-9.4 15-49.4 14.6-67.7zm52 58.3c-4.3-12.4-6-30.1-15.3-32.7-2-.5-9-.5-11 0-10 2.8-10.8 22.1-17 37.2 15.4 0 19.3 9.7 23.7 9.7 4.3 0 6.3-11.3 19.6-14.2zm135.7-84.7c-6.6-12.1-24.8-12.9-46.5-13.9-40.2-1.9-78.2-3.8-77.3 40.3-.5 18.3 5 58.3 13.2 67.8 13 14.9 76.6 11.8 86.3 7.1 15.8-7.6 36.5-78.9 24.3-101.3z\"],\"trade-federation\":[496,512,[],\"f513\",\"M248 8.8c-137 0-248 111-248 248s111 248 248 248 248-111 248-248-111-248-248-248zm0 482.8c-129.7 0-234.8-105.1-234.8-234.8S118.3 22 248 22s234.8 105.1 234.8 234.8S377.7 491.6 248 491.6zm155.1-328.5v-46.8H209.3V198H54.2l36.7 46h117.7v196.8h48.8V245h83.3v-47h-83.3v-34.8h145.7zm-73.3 45.1v23.9h-82.9v197.4h-26.8V232.1H96.3l-20.1-23.9h143.9v-80.6h171.8V152h-145v56.2zm-161.3-69l-12.4-20.7 2.1 23.8-23.5 5.4 23.3 5.4-2.1 24 12.3-20.5 22.2 9.5-15.7-18.1 15.8-18.1zm-29.6-19.7l9.3-11.5-12.7 5.9-8-12.4 1.7 13.9-14.3 3.8 13.7 2.7-.8 14.7 6.8-12.2 13.8 5.3zm165.4 145.2l-13.1 5.6-7.3-12.2 1.3 14.2-13.9 3.2 13.9 3.2-1.2 14.2 7.3-12.2 13.1 5.5-9.4-10.7zm106.9-77.2l-20.9 9.1-12-19.6 2.2 22.7-22.3 5.4 22.2 4.9-1.8 22.9 11.5-19.6 21.2 8.8-15.1-17zM248 29.9c-125.3 0-226.9 101.6-226.9 226.9S122.7 483.7 248 483.7s226.9-101.6 226.9-226.9S373.3 29.9 248 29.9zM342.6 196v51h-83.3v195.7h-52.7V245.9H89.9l-40-49.9h157.4v-81.6h197.8v50.7H259.4V196zM248 43.2c60.3 0 114.8 25 153.6 65.2H202.5V190H45.1C73.1 104.8 153.4 43.2 248 43.2zm0 427.1c-117.9 0-213.6-95.6-213.6-213.5 0-21.2 3.1-41.8 8.9-61.1L87.1 252h114.7v196.8h64.6V253h83.3v-62.7h-83.2v-19.2h145.6v-50.8c30.8 37 49.3 84.6 49.3 136.5.1 117.9-95.5 213.5-213.4 213.5zM178.8 275l-11-21.4 1.7 24.5-23.7 3.9 23.8 5.9-3.7 23.8 13-20.9 21.5 10.8-15.8-18.8 16.9-17.1z\"],autoprefixer:[640,512,[],\"f41c\",\"M318.4 16l-161 480h77.5l25.4-81.4h119.5L405 496h77.5L318.4 16zm-40.3 341.9l41.2-130.4h1.5l40.9 130.4h-83.6zM640 405l-10-31.4L462.1 358l19.4 56.5L640 405zm-462.1-47L10 373.7 0 405l158.5 9.4 19.4-56.4z\"],whatsapp:[448,512,[],\"f232\",\"M380.9 97.1C339 55.1 283.2 32 223.9 32c-122.4 0-222 99.6-222 222 0 39.1 10.2 77.3 29.6 111L0 480l117.7-30.9c32.4 17.7 68.9 27 106.1 27h.1c122.3 0 224.1-99.6 224.1-222 0-59.3-25.2-115-67.1-157zm-157 341.6c-33.2 0-65.7-8.9-94-25.7l-6.7-4-69.8 18.3L72 359.2l-4.4-7c-18.5-29.4-28.2-63.3-28.2-98.2 0-101.7 82.8-184.5 184.6-184.5 49.3 0 95.6 19.2 130.4 54.1 34.8 34.9 56.2 81.2 56.1 130.5 0 101.8-84.9 184.6-186.6 184.6zm101.2-138.2c-5.5-2.8-32.8-16.2-37.9-18-5.1-1.9-8.8-2.8-12.5 2.8-3.7 5.6-14.3 18-17.6 21.8-3.2 3.7-6.5 4.2-12 1.4-32.6-16.3-54-29.1-75.5-66-5.7-9.8 5.7-9.1 16.3-30.3 1.8-3.7.9-6.9-.5-9.7-1.4-2.8-12.5-30.1-17.1-41.2-4.5-10.8-9.1-9.3-12.5-9.5-3.2-.2-6.9-.2-10.6-.2-3.7 0-9.7 1.4-14.8 6.9-5.1 5.6-19.4 19-19.4 46.3 0 27.3 19.9 53.7 22.6 57.4 2.8 3.7 39.1 59.7 94.8 83.8 35.2 15.2 49 16.5 66.6 13.9 10.7-1.6 32.8-13.4 37.4-26.4 4.6-13 4.6-24.1 3.2-26.4-1.3-2.5-5-3.9-10.5-6.6z\"],\"square-upwork\":[448,512,[],\"e67c\",\"M56 32l336 0c30.9 0 56 25.1 56 56l0 336c0 30.9-25.1 56-56 56L56 480c-30.9 0-56-25.1-56-56L0 88C0 57.1 25.1 32 56 32zM270.9 274.2c6.6-52.9 25.9-69.5 51.4-69.5c25.3 0 44.9 20.2 44.9 49.7s-19.7 49.7-44.9 49.7c-27.9 0-46.3-21.5-51.4-29.9zm-26.7-41.8c-8.2-15.5-14.3-36.3-19.2-55.6l-29.7 0-33.2 0 0 78.1c0 28.4-12.9 49.4-38.2 49.4s-39.8-20.9-39.8-49.3l.3-78.1-36.2 0 0 78.1c0 22.8 7.4 43.5 20.9 58.2c13.9 15.2 32.8 23.2 54.8 23.2c43.7 0 74.2-33.5 74.2-81.5l0-52.5c4.6 17.3 15.4 50.5 36.2 79.7L215 392.6l36.8 0 12.8-78.4c4.2 3.5 8.7 6.6 13.4 9.4c12.3 7.8 26.4 12.2 40.9 12.6l.1 0c.5 0 1.1 0 1.6 0c.6 0 1.1 0 1.7 0c45.1 0 80.9-34.9 80.9-81.9s-35.9-82.2-80.9-82.2c-45.4 0-70.9 29.7-78.1 60.1z\"],slideshare:[512,512,[],\"f1e7\",\"M187.7 153.7c-34 0-61.7 25.7-61.7 57.7 0 31.7 27.7 57.7 61.7 57.7s61.7-26 61.7-57.7c0-32-27.7-57.7-61.7-57.7zm143.4 0c-34 0-61.7 25.7-61.7 57.7 0 31.7 27.7 57.7 61.7 57.7 34.3 0 61.7-26 61.7-57.7.1-32-27.4-57.7-61.7-57.7zm156.6 90l-6 4.3V49.7c0-27.4-20.6-49.7-46-49.7H76.6c-25.4 0-46 22.3-46 49.7V248c-2-1.4-4.3-2.9-6.3-4.3-15.1-10.6-25.1 4-16 17.7 18.3 22.6 53.1 50.3 106.3 72C58.3 525.1 252 555.7 248.9 457.5c0-.7.3-56.6.3-96.6 5.1 1.1 9.4 2.3 13.7 3.1 0 39.7.3 92.8.3 93.5-3.1 98.3 190.6 67.7 134.3-124 53.1-21.7 88-49.4 106.3-72 9.1-13.8-.9-28.3-16.1-17.8zm-30.5 19.2c-68.9 37.4-128.3 31.1-160.6 29.7-23.7-.9-32.6 9.1-33.7 24.9-10.3-7.7-18.6-15.5-20.3-17.1-5.1-5.4-13.7-8-27.1-7.7-31.7 1.1-89.7 7.4-157.4-28V72.3c0-34.9 8.9-45.7 40.6-45.7h317.7c30.3 0 40.9 12.9 40.9 45.7v190.6z\"],\"google-play\":[512,512,[],\"f3ab\",\"M325.3 234.3L104.6 13l280.8 161.2-60.1 60.1zM47 0C34 6.8 25.3 19.2 25.3 35.3v441.3c0 16.1 8.7 28.5 21.7 35.3l256.6-256L47 0zm425.2 225.6l-58.9-34.1-65.7 64.5 65.7 64.5 60.1-34.1c18-14.3 18-46.5-1.2-60.8zM104.6 499l280.8-161.2-60.1-60.1L104.6 499z\"],viadeo:[448,512,[],\"f2a9\",\"M276.2 150.5v.7C258.3 98.6 233.6 47.8 205.4 0c43.3 29.2 67 100 70.8 150.5zm32.7 121.7c7.6 18.2 11 37.5 11 57 0 77.7-57.8 141-137.8 139.4l3.8-.3c74.2-46.7 109.3-118.6 109.3-205.1 0-38.1-6.5-75.9-18.9-112 1 11.7 1 23.7 1 35.4 0 91.8-18.1 241.6-116.6 280C95 455.2 49.4 398 49.4 329.2c0-75.6 57.4-142.3 135.4-142.3 16.8 0 33.7 3.1 49.1 9.6 1.7-15.1 6.5-29.9 13.4-43.3-19.9-7.2-41.2-10.7-62.5-10.7-161.5 0-238.7 195.9-129.9 313.7 67.9 74.6 192 73.9 259.8 0 56.6-61.3 60.9-142.4 36.4-201-12.7 8-27.1 13.9-42.2 17zM418.1 11.7c-31 66.5-81.3 47.2-115.8 80.1-12.4 12-20.6 34-20.6 50.5 0 14.1 4.5 27.1 12 38.8 47.4-11 98.3-46 118.2-90.7-.7 5.5-4.8 14.4-7.2 19.2-20.3 35.7-64.6 65.6-99.7 84.9 14.8 14.4 33.7 25.8 55 25.8 79 0 110.1-134.6 58.1-208.6z\"],line:[512,512,[],\"f3c0\",\"M311 196.8v81.3c0 2.1-1.6 3.7-3.7 3.7h-13c-1.3 0-2.4-.7-3-1.5l-37.3-50.3v48.2c0 2.1-1.6 3.7-3.7 3.7h-13c-2.1 0-3.7-1.6-3.7-3.7V196.9c0-2.1 1.6-3.7 3.7-3.7h12.9c1.1 0 2.4 .6 3 1.6l37.3 50.3V196.9c0-2.1 1.6-3.7 3.7-3.7h13c2.1-.1 3.8 1.6 3.8 3.5zm-93.7-3.7h-13c-2.1 0-3.7 1.6-3.7 3.7v81.3c0 2.1 1.6 3.7 3.7 3.7h13c2.1 0 3.7-1.6 3.7-3.7V196.8c0-1.9-1.6-3.7-3.7-3.7zm-31.4 68.1H150.3V196.8c0-2.1-1.6-3.7-3.7-3.7h-13c-2.1 0-3.7 1.6-3.7 3.7v81.3c0 1 .3 1.8 1 2.5c.7 .6 1.5 1 2.5 1h52.2c2.1 0 3.7-1.6 3.7-3.7v-13c0-1.9-1.6-3.7-3.5-3.7zm193.7-68.1H327.3c-1.9 0-3.7 1.6-3.7 3.7v81.3c0 1.9 1.6 3.7 3.7 3.7h52.2c2.1 0 3.7-1.6 3.7-3.7V265c0-2.1-1.6-3.7-3.7-3.7H344V247.7h35.5c2.1 0 3.7-1.6 3.7-3.7V230.9c0-2.1-1.6-3.7-3.7-3.7H344V213.5h35.5c2.1 0 3.7-1.6 3.7-3.7v-13c-.1-1.9-1.7-3.7-3.7-3.7zM512 93.4V419.4c-.1 51.2-42.1 92.7-93.4 92.6H92.6C41.4 511.9-.1 469.8 0 418.6V92.6C.1 41.4 42.2-.1 93.4 0H419.4c51.2 .1 92.7 42.1 92.6 93.4zM441.6 233.5c0-83.4-83.7-151.3-186.4-151.3s-186.4 67.9-186.4 151.3c0 74.7 66.3 137.4 155.9 149.3c21.8 4.7 19.3 12.7 14.4 42.1c-.8 4.7-3.8 18.4 16.1 10.1s107.3-63.2 146.5-108.2c27-29.7 39.9-59.8 39.9-93.1z\"],\"google-drive\":[512,512,[],\"f3aa\",\"M339 314.9L175.4 32h161.2l163.6 282.9H339zm-137.5 23.6L120.9 480h310.5L512 338.5H201.5zM154.1 67.4L0 338.5 80.6 480 237 208.8 154.1 67.4z\"],servicestack:[496,512,[],\"f3ec\",\"M88 216c81.7 10.2 273.7 102.3 304 232H0c99.5-8.1 184.5-137 88-232zm32-152c32.3 35.6 47.7 83.9 46.4 133.6C249.3 231.3 373.7 321.3 400 448h96C455.3 231.9 222.8 79.5 120 64z\"],simplybuilt:[512,512,[],\"f215\",\"M481.2 64h-106c-14.5 0-26.6 11.8-26.6 26.3v39.6H163.3V90.3c0-14.5-12-26.3-26.6-26.3h-106C16.1 64 4.3 75.8 4.3 90.3v331.4c0 14.5 11.8 26.3 26.6 26.3h450.4c14.8 0 26.6-11.8 26.6-26.3V90.3c-.2-14.5-12-26.3-26.7-26.3zM149.8 355.8c-36.6 0-66.4-29.7-66.4-66.4 0-36.9 29.7-66.6 66.4-66.6 36.9 0 66.6 29.7 66.6 66.6 0 36.7-29.7 66.4-66.6 66.4zm212.4 0c-36.9 0-66.6-29.7-66.6-66.6 0-36.6 29.7-66.4 66.6-66.4 36.6 0 66.4 29.7 66.4 66.4 0 36.9-29.8 66.6-66.4 66.6z\"],bitbucket:[512,512,[61810],\"f171\",\"M22.2 32A16 16 0 0 0 6 47.8a26.35 26.35 0 0 0 .2 2.8l67.9 412.1a21.77 21.77 0 0 0 21.3 18.2h325.7a16 16 0 0 0 16-13.4L505 50.7a16 16 0 0 0-13.2-18.3 24.58 24.58 0 0 0-2.8-.2L22.2 32zm285.9 297.8h-104l-28.1-147h157.3l-25.2 147z\"],imdb:[448,512,[],\"f2d8\",\"M89.5 323.6H53.93V186.2H89.5V323.6zM156.1 250.5L165.2 186.2H211.5V323.6H180.5V230.9L167.1 323.6H145.8L132.8 232.9L132.7 323.6H101.5V186.2H147.6C148.1 194.5 150.4 204.3 151.9 215.6L156.1 250.5zM223.7 323.6V186.2H250.3C267.3 186.2 277.3 187.1 283.3 188.6C289.4 190.3 294 192.8 297.2 196.5C300.3 199.8 302.3 203.1 303 208.5C303.9 212.9 304.4 221.6 304.4 234.7V282.9C304.4 295.2 303.7 303.4 302.5 307.6C301.4 311.7 299.4 315 296.5 317.3C293.7 319.7 290.1 321.4 285.8 322.3C281.6 323.1 275.2 323.6 266.7 323.6H223.7zM259.2 209.7V299.1C264.3 299.1 267.5 298.1 268.6 296.8C269.7 294.8 270.4 289.2 270.4 280.1V226.8C270.4 220.6 270.3 216.6 269.7 214.8C269.4 213 268.5 211.8 267.1 210.1C265.7 210.1 263 209.7 259.2 209.7V209.7zM316.5 323.6V186.2H350.6V230.1C353.5 227.7 356.7 225.2 360.1 223.5C363.7 222 368.9 221.1 372.9 221.1C377.7 221.1 381.8 221.9 385.2 223.3C388.6 224.8 391.2 226.8 393.2 229.5C394.9 232.1 395.9 234.8 396.3 237.3C396.7 239.9 396.1 245.3 396.1 253.5V292.1C396.1 300.3 396.3 306.4 395.3 310.5C394.2 314.5 391.5 318.1 387.5 320.1C383.4 324 378.6 325.4 372.9 325.4C368.9 325.4 363.7 324.5 360.2 322.9C356.7 321.1 353.5 318.4 350.6 314.9L348.5 323.6L316.5 323.6zM361.6 302.9C362.3 301.1 362.6 296.9 362.6 290.4V255C362.6 249.4 362.3 245.5 361.5 243.8C360.8 241.9 357.8 241.1 355.7 241.1C353.7 241.1 352.3 241.9 351.6 243.4C351 244.9 350.6 248.8 350.6 255V291.4C350.6 297.5 351 301.4 351.8 303C352.4 304.7 353.9 305.5 355.9 305.5C358.1 305.5 360.1 304.7 361.6 302.9L361.6 302.9zM418.4 32.04C434.1 33.27 447.1 47.28 447.1 63.92V448.1C447.1 464.5 435.2 478.5 418.9 479.1C418.6 479.1 418.4 480 418.1 480H29.88C29.6 480 29.32 479.1 29.04 479.9C13.31 478.5 1.093 466.1 0 449.7L.0186 61.78C1.081 45.88 13.82 33.09 30.26 31.1H417.7C417.9 31.1 418.2 32.01 418.4 32.04L418.4 32.04zM30.27 41.26C19 42.01 10.02 51.01 9.257 62.4V449.7C9.63 455.1 11.91 460.2 15.7 464C19.48 467.9 24.51 470.3 29.89 470.7H418.1C429.6 469.7 438.7 459.1 438.7 448.1V63.91C438.7 58.17 436.6 52.65 432.7 48.45C428.8 44.24 423.4 41.67 417.7 41.26L30.27 41.26z\"],deezer:[576,512,[],\"e077\",\"M451.46,244.71H576V172H451.46Zm0-173.89v72.67H576V70.82Zm0,275.06H576V273.2H451.46ZM0,447.09H124.54V374.42H0Zm150.47,0H275V374.42H150.47Zm150.52,0H425.53V374.42H301Zm150.47,0H576V374.42H451.46ZM301,345.88H425.53V273.2H301Zm-150.52,0H275V273.2H150.47Zm0-101.17H275V172H150.47Z\"],\"raspberry-pi\":[407,512,[],\"f7bb\",\"M372 232.5l-3.7-6.5c.1-46.4-21.4-65.3-46.5-79.7 7.6-2 15.4-3.6 17.6-13.2 13.1-3.3 15.8-9.4 17.1-15.8 3.4-2.3 14.8-8.7 13.6-19.7 6.4-4.4 10-10.1 8.1-18.1 6.9-7.5 8.7-13.7 5.8-19.4 8.3-10.3 4.6-15.6 1.1-20.9 6.2-11.2.7-23.2-16.6-21.2-6.9-10.1-21.9-7.8-24.2-7.8-2.6-3.2-6-6-16.5-4.7-6.8-6.1-14.4-5-22.3-2.1-9.3-7.3-15.5-1.4-22.6.8C271.6.6 269 5.5 263.5 7.6c-12.3-2.6-16.1 3-22 8.9l-6.9-.1c-18.6 10.8-27.8 32.8-31.1 44.1-3.3-11.3-12.5-33.3-31.1-44.1l-6.9.1c-5.9-5.9-9.7-11.5-22-8.9-5.6-2-8.1-7-19.4-3.4-4.6-1.4-8.9-4.4-13.9-4.3-2.6.1-5.5 1-8.7 3.5-7.9-3-15.5-4-22.3 2.1-10.5-1.3-14 1.4-16.5 4.7-2.3 0-17.3-2.3-24.2 7.8C21.2 16 15.8 28 22 39.2c-3.5 5.4-7.2 10.7 1.1 20.9-2.9 5.7-1.1 11.9 5.8 19.4-1.8 8 1.8 13.7 8.1 18.1-1.2 11 10.2 17.4 13.6 19.7 1.3 6.4 4 12.4 17.1 15.8 2.2 9.5 10 11.2 17.6 13.2-25.1 14.4-46.6 33.3-46.5 79.7l-3.7 6.5c-28.8 17.2-54.7 72.7-14.2 117.7 2.6 14.1 7.1 24.2 11 35.4 5.9 45.2 44.5 66.3 54.6 68.8 14.9 11.2 30.8 21.8 52.2 29.2C159 504.2 181 512 203 512h1c22.1 0 44-7.8 64.2-28.4 21.5-7.4 37.3-18 52.2-29.2 10.2-2.5 48.7-23.6 54.6-68.8 3.9-11.2 8.4-21.3 11-35.4 40.6-45.1 14.7-100.5-14-117.7zm-22.2-8c-1.5 18.7-98.9-65.1-82.1-67.9 45.7-7.5 83.6 19.2 82.1 67.9zm-43 93.1c-24.5 15.8-59.8 5.6-78.8-22.8s-14.6-64.2 9.9-80c24.5-15.8 59.8-5.6 78.8 22.8s14.6 64.2-9.9 80zM238.9 29.3c.8 4.2 1.8 6.8 2.9 7.6 5.4-5.8 9.8-11.7 16.8-17.3 0 3.3-1.7 6.8 2.5 9.4 3.7-5 8.8-9.5 15.5-13.3-3.2 5.6-.6 7.3 1.2 9.6 5.1-4.4 10-8.8 19.4-12.3-2.6 3.1-6.2 6.2-2.4 9.8 5.3-3.3 10.6-6.6 23.1-8.9-2.8 3.1-8.7 6.3-5.1 9.4 6.6-2.5 14-4.4 22.1-5.4-3.9 3.2-7.1 6.3-3.9 8.8 7.1-2.2 16.9-5.1 26.4-2.6l-6 6.1c-.7.8 14.1.6 23.9.8-3.6 5-7.2 9.7-9.3 18.2 1 1 5.8.4 10.4 0-4.7 9.9-12.8 12.3-14.7 16.6 2.9 2.2 6.8 1.6 11.2.1-3.4 6.9-10.4 11.7-16 17.3 1.4 1 3.9 1.6 9.7.9-5.2 5.5-11.4 10.5-18.8 15 1.3 1.5 5.8 1.5 10 1.6-6.7 6.5-15.3 9.9-23.4 14.2 4 2.7 6.9 2.1 10 2.1-5.7 4.7-15.4 7.1-24.4 10 1.7 2.7 3.4 3.4 7.1 4.1-9.5 5.3-23.2 2.9-27 5.6.9 2.7 3.6 4.4 6.7 5.8-15.4.9-57.3-.6-65.4-32.3 15.7-17.3 44.4-37.5 93.7-62.6-38.4 12.8-73 30-102 53.5-34.3-15.9-10.8-55.9 5.8-71.8zm-34.4 114.6c24.2-.3 54.1 17.8 54 34.7-.1 15-21 27.1-53.8 26.9-32.1-.4-53.7-15.2-53.6-29.8 0-11.9 26.2-32.5 53.4-31.8zm-123-12.8c3.7-.7 5.4-1.5 7.1-4.1-9-2.8-18.7-5.3-24.4-10 3.1 0 6 .7 10-2.1-8.1-4.3-16.7-7.7-23.4-14.2 4.2-.1 8.7 0 10-1.6-7.4-4.5-13.6-9.5-18.8-15 5.8.7 8.3.1 9.7-.9-5.6-5.6-12.7-10.4-16-17.3 4.3 1.5 8.3 2 11.2-.1-1.9-4.2-10-6.7-14.7-16.6 4.6.4 9.4 1 10.4 0-2.1-8.5-5.8-13.3-9.3-18.2 9.8-.1 24.6 0 23.9-.8l-6-6.1c9.5-2.5 19.3.4 26.4 2.6 3.2-2.5-.1-5.6-3.9-8.8 8.1 1.1 15.4 2.9 22.1 5.4 3.5-3.1-2.3-6.3-5.1-9.4 12.5 2.3 17.8 5.6 23.1 8.9 3.8-3.6.2-6.7-2.4-9.8 9.4 3.4 14.3 7.9 19.4 12.3 1.7-2.3 4.4-4 1.2-9.6 6.7 3.8 11.8 8.3 15.5 13.3 4.1-2.6 2.5-6.2 2.5-9.4 7 5.6 11.4 11.5 16.8 17.3 1.1-.8 2-3.4 2.9-7.6 16.6 15.9 40.1 55.9 6 71.8-29-23.5-63.6-40.7-102-53.5 49.3 25 78 45.3 93.7 62.6-8 31.8-50 33.2-65.4 32.3 3.1-1.4 5.8-3.2 6.7-5.8-4-2.8-17.6-.4-27.2-5.6zm60.1 24.1c16.8 2.8-80.6 86.5-82.1 67.9-1.5-48.7 36.5-75.5 82.1-67.9zM38.2 342c-23.7-18.8-31.3-73.7 12.6-98.3 26.5-7 9 107.8-12.6 98.3zm91 98.2c-13.3 7.9-45.8 4.7-68.8-27.9-15.5-27.4-13.5-55.2-2.6-63.4 16.3-9.8 41.5 3.4 60.9 25.6 16.9 20 24.6 55.3 10.5 65.7zm-26.4-119.7c-24.5-15.8-28.9-51.6-9.9-80s54.3-38.6 78.8-22.8 28.9 51.6 9.9 80c-19.1 28.4-54.4 38.6-78.8 22.8zM205 496c-29.4 1.2-58.2-23.7-57.8-32.3-.4-12.7 35.8-22.6 59.3-22 23.7-1 55.6 7.5 55.7 18.9.5 11-28.8 35.9-57.2 35.4zm58.9-124.9c.2 29.7-26.2 53.8-58.8 54-32.6.2-59.2-23.8-59.4-53.4v-.6c-.2-29.7 26.2-53.8 58.8-54 32.6-.2 59.2 23.8 59.4 53.4v.6zm82.2 42.7c-25.3 34.6-59.6 35.9-72.3 26.3-13.3-12.4-3.2-50.9 15.1-72 20.9-23.3 43.3-38.5 58.9-26.6 10.5 10.3 16.7 49.1-1.7 72.3zm22.9-73.2c-21.5 9.4-39-105.3-12.6-98.3 43.9 24.7 36.3 79.6 12.6 98.3z\"],jira:[496,512,[],\"f7b1\",\"M490 241.7C417.1 169 320.6 71.8 248.5 0 83 164.9 6 241.7 6 241.7c-7.9 7.9-7.9 20.7 0 28.7C138.8 402.7 67.8 331.9 248.5 512c379.4-378 15.7-16.7 241.5-241.7 8-7.9 8-20.7 0-28.6zm-241.5 90l-76-75.7 76-75.7 76 75.7-76 75.7z\"],docker:[640,512,[],\"f395\",\"M349.9 236.3h-66.1v-59.4h66.1v59.4zm0-204.3h-66.1v60.7h66.1V32zm78.2 144.8H362v59.4h66.1v-59.4zm-156.3-72.1h-66.1v60.1h66.1v-60.1zm78.1 0h-66.1v60.1h66.1v-60.1zm276.8 100c-14.4-9.7-47.6-13.2-73.1-8.4-3.3-24-16.7-44.9-41.1-63.7l-14-9.3-9.3 14c-18.4 27.8-23.4 73.6-3.7 103.8-8.7 4.7-25.8 11.1-48.4 10.7H2.4c-8.7 50.8 5.8 116.8 44 162.1 37.1 43.9 92.7 66.2 165.4 66.2 157.4 0 273.9-72.5 328.4-204.2 21.4.4 67.6.1 91.3-45.2 1.5-2.5 6.6-13.2 8.5-17.1l-13.3-8.9zm-511.1-27.9h-66v59.4h66.1v-59.4zm78.1 0h-66.1v59.4h66.1v-59.4zm78.1 0h-66.1v59.4h66.1v-59.4zm-78.1-72.1h-66.1v60.1h66.1v-60.1z\"],screenpal:[512,512,[],\"e570\",\"M233.5 22.49C233.5 10.07 243.6 0 256 0C268.4 0 278.5 10.07 278.5 22.49C278.5 34.91 268.4 44.98 256 44.98C243.6 44.98 233.5 34.91 233.5 22.49zM313.4 259C313.4 290.7 287.7 316.4 256 316.4C224.3 316.4 198.6 290.7 198.6 259C198.6 227.3 224.3 201.6 256 201.6C287.7 201.6 313.4 227.3 313.4 259zM337.2 350C359.5 330.1 373.7 302.7 377.1 273H496.6C493.1 334.4 466.2 392.2 421.4 434.4C376.7 476.6 317.5 500.2 256 500.2C194.5 500.2 135.3 476.6 90.56 434.4C45.83 392.2 18.94 334.4 15.39 273H135.1C138.5 302.7 152.7 330.1 175 350C197.3 369.9 226.2 380.9 256.1 380.9C285.1 380.9 314.8 369.9 337.2 350zM73.14 140.3C73.54 152.7 63.81 163.1 51.39 163.5C38.97 163.9 28.59 154.2 28.18 141.8C27.78 129.3 37.52 118.9 49.94 118.5C62.35 118.1 72.74 127.9 73.14 140.3zM438.9 141C438.9 128.6 448.9 118.5 461.4 118.5C473.8 118.5 483.8 128.6 483.8 141C483.8 153.5 473.8 163.5 461.4 163.5C448.9 163.5 438.9 153.5 438.9 141zM317.9 95.27C300.6 109.1 278.7 118.1 256 118.1C233.3 118.1 211.4 109.1 194.1 95.27C176.8 80.55 165.3 60.18 161.7 37.78C176.8 31.37 192.5 26.52 208.6 23.31C208.6 35.88 213.6 47.93 222.5 56.82C231.4 65.7 243.4 70.7 256 70.7C268.6 70.7 280.6 65.7 289.5 56.82C298.4 47.93 303.4 35.88 303.4 23.31C319.5 26.52 335.2 31.37 350.3 37.78C346.7 60.18 335.2 80.55 317.9 95.27H317.9zM82.78 231C61.42 238.6 38.06 238.4 16.86 230.4C18.82 214.1 22.46 198.1 27.71 182.5C33.1 185.6 39.05 187.6 45.22 188.5C51.39 189.3 57.67 188.9 63.68 187.3C69.69 185.6 75.33 182.9 80.27 179.1C85.21 175.3 89.36 170.6 92.47 165.2C95.58 159.8 97.61 153.8 98.42 147.7C99.23 141.5 98.83 135.2 97.22 129.2C95.61 123.2 92.83 117.6 89.04 112.6C85.25 107.7 80.53 103.5 75.14 100.4C85.96 88.11 98.01 76.94 111.1 67.07C128.7 81.42 140.6 101.6 144.7 123.9C148.8 146.2 144.8 169.3 133.5 188.9C122.1 208.5 104.1 223.4 82.78 231V231zM429.2 231.1C407.9 223.5 389.9 208.5 378.5 188.9C367.2 169.3 363.3 146.2 367.4 123.9C371.5 101.7 383.4 81.54 400.9 67.19C414 77.04 426.1 88.21 436.9 100.5C426.2 106.9 418.5 117.2 415.4 129.3C412.2 141.3 413.1 154.1 420.2 164.9C426.4 175.7 436.6 183.6 448.6 186.9C460.6 190.2 473.5 188.6 484.3 182.6C489.6 198.1 493.2 214.2 495.2 230.4C473.1 238.5 450.6 238.7 429.2 231.1L429.2 231.1z\"],bluetooth:[448,512,[],\"f293\",\"M292.6 171.1L249.7 214l-.3-86 43.2 43.1m-43.2 219.8l43.1-43.1-42.9-42.9-.2 86zM416 259.4C416 465 344.1 512 230.9 512S32 465 32 259.4 115.4 0 228.6 0 416 53.9 416 259.4zm-158.5 0l79.4-88.6L211.8 36.5v176.9L138 139.6l-27 26.9 92.7 93-92.7 93 26.9 26.9 73.8-73.8 2.3 170 127.4-127.5-83.9-88.7z\"],gitter:[384,512,[],\"f426\",\"M66.4 322.5H16V0h50.4v322.5zM166.9 76.1h-50.4V512h50.4V76.1zm100.6 0h-50.4V512h50.4V76.1zM368 76h-50.4v247H368V76z\"],\"d-and-d\":[576,512,[],\"f38d\",\"M82.5 98.9c-.6-17.2 2-33.8 12.7-48.2.3 7.4 1.2 14.5 4.2 21.6 5.9-27.5 19.7-49.3 42.3-65.5-1.9 5.9-3.5 11.8-3 17.7 8.7-7.4 18.8-17.8 44.4-22.7 14.7-2.8 29.7-2 42.1 1 38.5 9.3 61 34.3 69.7 72.3 5.3 23.1.7 45-8.3 66.4-5.2 12.4-12 24.4-20.7 35.1-2-1.9-3.9-3.8-5.8-5.6-42.8-40.8-26.8-25.2-37.4-37.4-1.1-1.2-1-2.2-.1-3.6 8.3-13.5 11.8-28.2 10-44-1.1-9.8-4.3-18.9-11.3-26.2-14.5-15.3-39.2-15-53.5.6-11.4 12.5-14.1 27.4-10.9 43.6.2 1.3.4 2.7 0 3.9-3.4 13.7-4.6 27.6-2.5 41.6.1.5.1 1.1.1 1.6 0 .3-.1.5-.2 1.1-21.8-11-36-28.3-43.2-52.2-8.3 17.8-11.1 35.5-6.6 54.1-15.6-15.2-21.3-34.3-22-55.2zm469.6 123.2c-11.6-11.6-25-20.4-40.1-26.6-12.8-5.2-26-7.9-39.9-7.1-10 .6-19.6 3.1-29 6.4-2.5.9-5.1 1.6-7.7 2.2-4.9 1.2-7.3-3.1-4.7-6.8 3.2-4.6 3.4-4.2 15-12 .6-.4 1.2-.8 2.2-1.5h-2.5c-.6 0-1.2.2-1.9.3-19.3 3.3-30.7 15.5-48.9 29.6-10.4 8.1-13.8 3.8-12-.5 1.4-3.5 3.3-6.7 5.1-10 1-1.8 2.3-3.4 3.5-5.1-.2-.2-.5-.3-.7-.5-27 18.3-46.7 42.4-57.7 73.3.3.3.7.6 1 .9.3-.6.5-1.2.9-1.7 10.4-12.1 22.8-21.8 36.6-29.8 18.2-10.6 37.5-18.3 58.7-20.2 4.3-.4 8.7-.1 13.1-.1-1.8.7-3.5.9-5.3 1.1-18.5 2.4-35.5 9-51.5 18.5-30.2 17.9-54.5 42.2-75.1 70.4-.3.4-.4.9-.7 1.3 14.5 5.3 24 17.3 36.1 25.6.2-.1.3-.2.4-.4l1.2-2.7c12.2-26.9 27-52.3 46.7-74.5 16.7-18.8 38-25.3 62.5-20 5.9 1.3 11.4 4.4 17.2 6.8 2.3-1.4 5.1-3.2 8-4.7 8.4-4.3 17.4-7 26.7-9 14.7-3.1 29.5-4.9 44.5-1.3v-.5c-.5-.4-1.2-.8-1.7-1.4zM316.7 397.6c-39.4-33-22.8-19.5-42.7-35.6-.8.9 0-.2-1.9 3-11.2 19.1-25.5 35.3-44 47.6-10.3 6.8-21.5 11.8-34.1 11.8-21.6 0-38.2-9.5-49.4-27.8-12-19.5-13.3-40.7-8.2-62.6 7.8-33.8 30.1-55.2 38.6-64.3-18.7-6.2-33 1.7-46.4 13.9.8-13.9 4.3-26.2 11.8-37.3-24.3 10.6-45.9 25-64.8 43.9-.3-5.8 5.4-43.7 5.6-44.7.3-2.7-.6-5.3-3-7.4-24.2 24.7-44.5 51.8-56.1 84.6 7.4-5.9 14.9-11.4 23.6-16.2-8.3 22.3-19.6 52.8-7.8 101.1 4.6 19 11.9 36.8 24.1 52.3 2.9 3.7 6.3 6.9 9.5 10.3.2-.2.4-.3.6-.5-1.4-7-2.2-14.1-1.5-21.9 2.2 3.2 3.9 6 5.9 8.6 12.6 16 28.7 27.4 47.2 35.6 25 11.3 51.1 13.3 77.9 8.6 54.9-9.7 90.7-48.6 116-98.8 1-1.8.6-2.9-.9-4.2zm172-46.4c-9.5-3.1-22.2-4.2-28.7-2.9 9.9 4 14.1 6.6 18.8 12 12.6 14.4 10.4 34.7-5.4 45.6-11.7 8.1-24.9 10.5-38.9 9.1-1.2-.1-2.3-.4-3-.6 2.8-3.7 6-7 8.1-10.8 9.4-16.8 5.4-42.1-8.7-56.1-2.1-2.1-4.6-3.9-7-5.9-.3 1.3-.1 2.1.1 2.8 4.2 16.6-8.1 32.4-24.8 31.8-7.6-.3-13.9-3.8-19.6-8.5-19.5-16.1-39.1-32.1-58.5-48.3-5.9-4.9-12.5-8.1-20.1-8.7-4.6-.4-9.3-.6-13.9-.9-5.9-.4-8.8-2.8-10.4-8.4-.9-3.4-1.5-6.8-2.2-10.2-1.5-8.1-6.2-13-14.3-14.2-4.4-.7-8.9-1-13.3-1.5-13-1.4-19.8-7.4-22.6-20.3-5 11-1.6 22.4 7.3 29.9 4.5 3.8 9.3 7.3 13.8 11.2 4.6 3.8 7.4 8.7 7.9 14.8.4 4.7.8 9.5 1.8 14.1 2.2 10.6 8.9 18.4 17 25.1 16.5 13.7 33 27.3 49.5 41.1 17.9 15 13.9 32.8 13 56-.9 22.9 12.2 42.9 33.5 51.2 1 .4 2 .6 3.6 1.1-15.7-18.2-10.1-44.1.7-52.3.3 2.2.4 4.3.9 6.4 9.4 44.1 45.4 64.2 85 56.9 16-2.9 30.6-8.9 42.9-19.8 2-1.8 3.7-4.1 5.9-6.5-19.3 4.6-35.8.1-50.9-10.6.7-.3 1.3-.3 1.9-.3 21.3 1.8 40.6-3.4 57-17.4 19.5-16.6 26.6-42.9 17.4-66-8.3-20.1-23.6-32.3-43.8-38.9zM99.4 179.3c-5.3-9.2-13.2-15.6-22.1-21.3 13.7-.5 26.6.2 39.6 3.7-7-12.2-8.5-24.7-5-38.7 5.3 11.9 13.7 20.1 23.6 26.8 19.7 13.2 35.7 19.6 46.7 30.2 3.4 3.3 6.3 7.1 9.6 10.9-.8-2.1-1.4-4.1-2.2-6-5-10.6-13-18.6-22.6-25-1.8-1.2-2.8-2.5-3.4-4.5-3.3-12.5-3-25.1-.7-37.6 1-5.5 2.8-10.9 4.5-16.3.8-2.4 2.3-4.6 4-6.6.6 6.9 0 25.5 19.6 46 10.8 11.3 22.4 21.9 33.9 32.7 9 8.5 18.3 16.7 25.5 26.8 1.1 1.6 2.2 3.3 3.8 4.7-5-13-14.2-24.1-24.2-33.8-9.6-9.3-19.4-18.4-29.2-27.4-3.3-3-4.6-6.7-5.1-10.9-1.2-10.4 0-20.6 4.3-30.2.5-1 1.1-2 1.9-3.3.5 4.2.6 7.9 1.4 11.6 4.8 23.1 20.4 36.3 49.3 63.5 10 9.4 19.3 19.2 25.6 31.6 4.8 9.3 7.3 19 5.7 29.6-.1.6.5 1.7 1.1 2 6.2 2.6 10 6.9 9.7 14.3 7.7-2.6 12.5-8 16.4-14.5 4.2 20.2-9.1 50.3-27.2 58.7.4-4.5 5-23.4-16.5-27.7-6.8-1.3-12.8-1.3-22.9-2.1 4.7-9 10.4-20.6.5-22.4-24.9-4.6-52.8 1.9-57.8 4.6 8.2.4 16.3 1 23.5 3.3-2 6.5-4 12.7-5.8 18.9-1.9 6.5 2.1 14.6 9.3 9.6 1.2-.9 2.3-1.9 3.3-2.7-3.1 17.9-2.9 15.9-2.8 18.3.3 10.2 9.5 7.8 15.7 7.3-2.5 11.8-29.5 27.3-45.4 25.8 7-4.7 12.7-10.3 15.9-17.9-6.5.8-12.9 1.6-19.2 2.4l-.3-.9c4.7-3.4 8-7.8 10.2-13.1 8.7-21.1-3.6-38-25-39.9-9.1-.8-17.8.8-25.9 5.5 6.2-15.6 17.2-26.6 32.6-34.5-15.2-4.3-8.9-2.7-24.6-6.3 14.6-9.3 30.2-13.2 46.5-14.6-5.2-3.2-48.1-3.6-70.2 20.9 7.9 1.4 15.5 2.8 23.2 4.2-23.8 7-44 19.7-62.4 35.6 1.1-4.8 2.7-9.5 3.3-14.3.6-4.5.8-9.2.1-13.6-1.5-9.4-8.9-15.1-19.7-16.3-7.9-.9-15.6.1-23.3 1.3-.9.1-1.7.3-2.9 0 15.8-14.8 36-21.7 53.1-33.5 6-4.5 6.8-8.2 3-14.9zm128.4 26.8c3.3 16 12.6 25.5 23.8 24.3-4.6-11.3-12.1-19.5-23.8-24.3z\"],microblog:[448,512,[],\"e01a\",\"M399.36,362.23c29.49-34.69,47.1-78.34,47.1-125.79C446.46,123.49,346.86,32,224,32S1.54,123.49,1.54,236.44,101.14,440.87,224,440.87a239.28,239.28,0,0,0,79.44-13.44,7.18,7.18,0,0,1,8.12,2.56c18.58,25.09,47.61,42.74,79.89,49.92a4.42,4.42,0,0,0,5.22-3.43,4.37,4.37,0,0,0-.85-3.62,87,87,0,0,1,3.69-110.69ZM329.52,212.4l-57.3,43.49L293,324.75a6.5,6.5,0,0,1-9.94,7.22L224,290.92,164.94,332a6.51,6.51,0,0,1-9.95-7.22l20.79-68.86-57.3-43.49a6.5,6.5,0,0,1,3.8-11.68l71.88-1.51,23.66-67.92a6.5,6.5,0,0,1,12.28,0l23.66,67.92,71.88,1.51a6.5,6.5,0,0,1,3.88,11.68Z\"],\"cc-diners-club\":[576,512,[],\"f24c\",\"M239.7 79.9c-96.9 0-175.8 78.6-175.8 175.8 0 96.9 78.9 175.8 175.8 175.8 97.2 0 175.8-78.9 175.8-175.8 0-97.2-78.6-175.8-175.8-175.8zm-39.9 279.6c-41.7-15.9-71.4-56.4-71.4-103.8s29.7-87.9 71.4-104.1v207.9zm79.8.3V151.6c41.7 16.2 71.4 56.7 71.4 104.1s-29.7 87.9-71.4 104.1zM528 32H48C21.5 32 0 53.5 0 80v352c0 26.5 21.5 48 48 48h480c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48zM329.7 448h-90.3c-106.2 0-193.8-85.5-193.8-190.2C45.6 143.2 133.2 64 239.4 64h90.3c105 0 200.7 79.2 200.7 193.8 0 104.7-95.7 190.2-200.7 190.2z\"],\"gg-circle\":[512,512,[],\"f261\",\"M257 8C120 8 9 119 9 256s111 248 248 248 248-111 248-248S394 8 257 8zm-49.5 374.8L81.8 257.1l125.7-125.7 35.2 35.4-24.2 24.2-11.1-11.1-77.2 77.2 77.2 77.2 26.6-26.6-53.1-52.9 24.4-24.4 77.2 77.2-75 75.2zm99-2.2l-35.2-35.2 24.1-24.4 11.1 11.1 77.2-77.2-77.2-77.2-26.5 26.5 53.1 52.9-24.4 24.4-77.2-77.2 75-75L432.2 255 306.5 380.6z\"],\"pied-piper-hat\":[640,512,[],\"f4e5\",\"M640 24.9c-80.8 53.6-89.4 92.5-96.4 104.4-6.7 12.2-11.7 60.3-23.3 83.6-11.7 23.6-54.2 42.2-66.1 50-11.7 7.8-28.3 38.1-41.9 64.2-108.1-4.4-167.4 38.8-259.2 93.6 29.4-9.7 43.3-16.7 43.3-16.7 94.2-36 139.3-68.3 281.1-49.2 1.1 0 1.9.6 2.8.8 3.9 2.2 5.3 6.9 3.1 10.8l-53.9 95.8c-2.5 4.7-7.8 7.2-13.1 6.1-126.8-23.8-226.9 17.3-318.9 18.6C24.1 488 0 453.4 0 451.8c0-1.1.6-1.7 1.7-1.7 0 0 38.3 0 103.1-15.3C178.4 294.5 244 245.4 315.4 245.4c0 0 71.7 0 90.6 61.9 22.8-39.7 28.3-49.2 28.3-49.2 5.3-9.4 35-77.2 86.4-141.4 51.5-64 90.4-79.9 119.3-91.8z\"],\"kickstarter-k\":[448,512,[],\"f3bc\",\"M356.6 256.2l40.8-40.5c42.2-41.9 42.2-110.3 0-152.1s-111-41.9-153.2 0L229.3 78.4C209.6 50.3 177.1 32 140.2 32C80.5 32 32 80.2 32 139.5V372.5C32 431.9 80.5 480 140.2 480c37.1 0 69.3-18.3 89-46.4l14.9 14.7c42.2 41.9 111 41.9 153.2 0s42.2-110.3 0-152.1l-40.8-40z\"],yandex:[256,512,[],\"f413\",\"M153.1 315.8L65.7 512H2l96-209.8c-45.1-22.9-75.2-64.4-75.2-141.1C22.7 53.7 90.8 0 171.7 0H254v512h-55.1V315.8h-45.8zm45.8-269.3h-29.4c-44.4 0-87.4 29.4-87.4 114.6 0 82.3 39.4 108.8 87.4 108.8h29.4V46.5z\"],readme:[576,512,[],\"f4d5\",\"M528.3 46.5H388.5c-48.1 0-89.9 33.3-100.4 80.3-10.6-47-52.3-80.3-100.4-80.3H48c-26.5 0-48 21.5-48 48v245.8c0 26.5 21.5 48 48 48h89.7c102.2 0 132.7 24.4 147.3 75 .7 2.8 5.2 2.8 6 0 14.7-50.6 45.2-75 147.3-75H528c26.5 0 48-21.5 48-48V94.6c0-26.4-21.3-47.9-47.7-48.1zM242 311.9c0 1.9-1.5 3.5-3.5 3.5H78.2c-1.9 0-3.5-1.5-3.5-3.5V289c0-1.9 1.5-3.5 3.5-3.5h160.4c1.9 0 3.5 1.5 3.5 3.5v22.9zm0-60.9c0 1.9-1.5 3.5-3.5 3.5H78.2c-1.9 0-3.5-1.5-3.5-3.5v-22.9c0-1.9 1.5-3.5 3.5-3.5h160.4c1.9 0 3.5 1.5 3.5 3.5V251zm0-60.9c0 1.9-1.5 3.5-3.5 3.5H78.2c-1.9 0-3.5-1.5-3.5-3.5v-22.9c0-1.9 1.5-3.5 3.5-3.5h160.4c1.9 0 3.5 1.5 3.5 3.5v22.9zm259.3 121.7c0 1.9-1.5 3.5-3.5 3.5H337.5c-1.9 0-3.5-1.5-3.5-3.5v-22.9c0-1.9 1.5-3.5 3.5-3.5h160.4c1.9 0 3.5 1.5 3.5 3.5v22.9zm0-60.9c0 1.9-1.5 3.5-3.5 3.5H337.5c-1.9 0-3.5-1.5-3.5-3.5V228c0-1.9 1.5-3.5 3.5-3.5h160.4c1.9 0 3.5 1.5 3.5 3.5v22.9zm0-60.9c0 1.9-1.5 3.5-3.5 3.5H337.5c-1.9 0-3.5-1.5-3.5-3.5v-22.8c0-1.9 1.5-3.5 3.5-3.5h160.4c1.9 0 3.5 1.5 3.5 3.5V190z\"],html5:[384,512,[],\"f13b\",\"M0 32l34.9 395.8L191.5 480l157.6-52.2L384 32H0zm308.2 127.9H124.4l4.1 49.4h175.6l-13.6 148.4-97.9 27v.3h-1.1l-98.7-27.3-6-75.8h47.7L138 320l53.5 14.5 53.7-14.5 6-62.2H84.3L71.5 112.2h241.1l-4.4 47.7z\"],sellsy:[640,512,[],\"f213\",\"M539.71 237.308c3.064-12.257 4.29-24.821 4.29-37.384C544 107.382 468.618 32 376.076 32c-77.22 0-144.634 53.012-163.02 127.781-15.322-13.176-34.934-20.53-55.157-20.53-46.271 0-83.962 37.69-83.962 83.961 0 7.354.92 15.015 3.065 22.369-42.9 20.225-70.785 63.738-70.785 111.234C6.216 424.843 61.68 480 129.401 480h381.198c67.72 0 123.184-55.157 123.184-123.184.001-56.384-38.916-106.025-94.073-119.508zM199.88 401.554c0 8.274-7.048 15.321-15.321 15.321H153.61c-8.274 0-15.321-7.048-15.321-15.321V290.626c0-8.273 7.048-15.321 15.321-15.321h30.949c8.274 0 15.321 7.048 15.321 15.321v110.928zm89.477 0c0 8.274-7.048 15.321-15.322 15.321h-30.949c-8.274 0-15.321-7.048-15.321-15.321V270.096c0-8.274 7.048-15.321 15.321-15.321h30.949c8.274 0 15.322 7.048 15.322 15.321v131.458zm89.477 0c0 8.274-7.047 15.321-15.321 15.321h-30.949c-8.274 0-15.322-7.048-15.322-15.321V238.84c0-8.274 7.048-15.321 15.322-15.321h30.949c8.274 0 15.321 7.048 15.321 15.321v162.714zm87.027 0c0 8.274-7.048 15.321-15.322 15.321h-28.497c-8.274 0-15.321-7.048-15.321-15.321V176.941c0-8.579 7.047-15.628 15.321-15.628h28.497c8.274 0 15.322 7.048 15.322 15.628v224.613z\"],\"square-web-awesome\":[448,512,[],\"e683\",\"M64 32C28.7 32 0 60.7 0 96L0 416c0 35.3 28.7 64 64 64l320 0c35.3 0 64-28.7 64-64l0-320c0-35.3-28.7-64-64-64L64 32zM235 177.6L288 224l52.2-10.4c-2.6-3.9-4.2-8.5-4.2-13.6c0-13.3 10.7-24 24-24s24 10.7 24 24c0 13-10.3 23.6-23.2 24L304.5 349.1c-5.2 11.5-16.6 18.9-29.2 18.9l-102.6 0c-12.6 0-24-7.4-29.2-18.9L87.2 224C74.3 223.6 64 213 64 200c0-13.3 10.7-24 24-24s24 10.7 24 24c0 5-1.5 9.7-4.2 13.6L160 224l53.1-46.4c-8.9-4.1-15-13.1-15-23.6c0-14.4 11.6-26 26-26s26 11.6 26 26c0 10.5-6.2 19.5-15.1 23.6z\"],sass:[640,512,[],\"f41e\",\"M301.84 378.92c-.3.6-.6 1.08 0 0zm249.13-87a131.16 131.16 0 0 0-58 13.5c-5.9-11.9-12-22.3-13-30.1-1.2-9.1-2.5-14.5-1.1-25.3s7.7-26.1 7.6-27.2-1.4-6.6-14.3-6.7-24 2.5-25.29 5.9a122.83 122.83 0 0 0-5.3 19.1c-2.3 11.7-25.79 53.5-39.09 75.3-4.4-8.5-8.1-16-8.9-22-1.2-9.1-2.5-14.5-1.1-25.3s7.7-26.1 7.6-27.2-1.4-6.6-14.29-6.7-24 2.5-25.3 5.9-2.7 11.4-5.3 19.1-33.89 77.3-42.08 95.4c-4.2 9.2-7.8 16.6-10.4 21.6-.4.8-.7 1.3-.9 1.7.3-.5.5-1 .5-.8-2.2 4.3-3.5 6.7-3.5 6.7v.1c-1.7 3.2-3.6 6.1-4.5 6.1-.6 0-1.9-8.4.3-19.9 4.7-24.2 15.8-61.8 15.7-63.1-.1-.7 2.1-7.2-7.3-10.7-9.1-3.3-12.4 2.2-13.2 2.2s-1.4 2-1.4 2 10.1-42.4-19.39-42.4c-18.4 0-44 20.2-56.58 38.5-7.9 4.3-25 13.6-43 23.5-6.9 3.8-14 7.7-20.7 11.4-.5-.5-.9-1-1.4-1.5-35.79-38.2-101.87-65.2-99.07-116.5 1-18.7 7.5-67.8 127.07-127.4 98-48.8 176.35-35.4 189.84-5.6 19.4 42.5-41.89 121.6-143.66 133-38.79 4.3-59.18-10.7-64.28-16.3-5.3-5.9-6.1-6.2-8.1-5.1-3.3 1.8-1.2 7 0 10.1 3 7.9 15.5 21.9 36.79 28.9 18.7 6.1 64.18 9.5 119.17-11.8 61.78-23.8 109.87-90.1 95.77-145.6C386.52 18.32 293-.18 204.57 31.22c-52.69 18.7-109.67 48.1-150.66 86.4-48.69 45.6-56.48 85.3-53.28 101.9 11.39 58.9 92.57 97.3 125.06 125.7-1.6.9-3.1 1.7-4.5 2.5-16.29 8.1-78.18 40.5-93.67 74.7-17.5 38.8 2.9 66.6 16.29 70.4 41.79 11.6 84.58-9.3 107.57-43.6s20.2-79.1 9.6-99.5c-.1-.3-.3-.5-.4-.8 4.2-2.5 8.5-5 12.8-7.5 8.29-4.9 16.39-9.4 23.49-13.3-4 10.8-6.9 23.8-8.4 42.6-1.8 22 7.3 50.5 19.1 61.7 5.2 4.9 11.49 5 15.39 5 13.8 0 20-11.4 26.89-25 8.5-16.6 16-35.9 16-35.9s-9.4 52.2 16.3 52.2c9.39 0 18.79-12.1 23-18.3v.1s.2-.4.7-1.2c1-1.5 1.5-2.4 1.5-2.4v-.3c3.8-6.5 12.1-21.4 24.59-46 16.2-31.8 31.69-71.5 31.69-71.5a201.24 201.24 0 0 0 6.2 25.8c2.8 9.5 8.7 19.9 13.4 30-3.8 5.2-6.1 8.2-6.1 8.2a.31.31 0 0 0 .1.2c-3 4-6.4 8.3-9.9 12.5-12.79 15.2-28 32.6-30 37.6-2.4 5.9-1.8 10.3 2.8 13.7 3.4 2.6 9.4 3 15.69 2.5 11.5-.8 19.6-3.6 23.5-5.4a82.2 82.2 0 0 0 20.19-10.6c12.5-9.2 20.1-22.4 19.4-39.8-.4-9.6-3.5-19.2-7.3-28.2 1.1-1.6 2.3-3.3 3.4-5C434.8 301.72 450.1 270 450.1 270a201.24 201.24 0 0 0 6.2 25.8c2.4 8.1 7.09 17 11.39 25.7-18.59 15.1-30.09 32.6-34.09 44.1-7.4 21.3-1.6 30.9 9.3 33.1 4.9 1 11.9-1.3 17.1-3.5a79.46 79.46 0 0 0 21.59-11.1c12.5-9.2 24.59-22.1 23.79-39.6-.3-7.9-2.5-15.8-5.4-23.4 15.7-6.6 36.09-10.2 62.09-7.2 55.68 6.5 66.58 41.3 64.48 55.8s-13.8 22.6-17.7 25-5.1 3.3-4.8 5.1c.5 2.6 2.3 2.5 5.6 1.9 4.6-.8 29.19-11.8 30.29-38.7 1.6-34-31.09-71.4-89-71.1zm-429.18 144.7c-18.39 20.1-44.19 27.7-55.28 21.3C54.61 451 59.31 421.42 82 400c13.8-13 31.59-25 43.39-32.4 2.7-1.6 6.6-4 11.4-6.9.8-.5 1.2-.7 1.2-.7.9-.6 1.9-1.1 2.9-1.7 8.29 30.4.3 57.2-19.1 78.3zm134.36-91.4c-6.4 15.7-19.89 55.7-28.09 53.6-7-1.8-11.3-32.3-1.4-62.3 5-15.1 15.6-33.1 21.9-40.1 10.09-11.3 21.19-14.9 23.79-10.4 3.5 5.9-12.2 49.4-16.2 59.2zm111 53c-2.7 1.4-5.2 2.3-6.4 1.6-.9-.5 1.1-2.4 1.1-2.4s13.9-14.9 19.4-21.7c3.2-4 6.9-8.7 10.89-13.9 0 .5.1 1 .1 1.6-.13 17.9-17.32 30-25.12 34.8zm85.58-19.5c-2-1.4-1.7-6.1 5-20.7 2.6-5.7 8.59-15.3 19-24.5a36.18 36.18 0 0 1 1.9 10.8c-.1 22.5-16.2 30.9-25.89 34.4z\"],wirsindhandwerk:[512,512,[\"wsh\"],\"e2d0\",\"M50.77161,479.81213h83.36071V367.84741l-83.36071,47.009Zm329.04675,0h82.35022V414.85645l-82.35022-47.009Zm.00568-448V251.568L256.1759,179.1861,134.50378,251.568V31.81213H50.77161V392.60565L256.1759,270.31909,462.16858,392.60565V31.81213Z\"],buromobelexperte:[448,512,[],\"f37f\",\"M0 32v128h128V32H0zm120 120H8V40h112v112zm40-120v128h128V32H160zm120 120H168V40h112v112zm40-120v128h128V32H320zm120 120H328V40h112v112zM0 192v128h128V192H0zm120 120H8V200h112v112zm40-120v128h128V192H160zm120 120H168V200h112v112zm40-120v128h128V192H320zm120 120H328V200h112v112zM0 352v128h128V352H0zm120 120H8V360h112v112zm40-120v128h128V352H160zm120 120H168V360h112v112zm40-120v128h128V352H320z\"],salesforce:[640,512,[],\"f83b\",\"M248.89 245.64h-26.35c.69-5.16 3.32-14.12 13.64-14.12 6.75 0 11.97 3.82 12.71 14.12zm136.66-13.88c-.47 0-14.11-1.77-14.11 20s13.63 20 14.11 20c13 0 14.11-13.54 14.11-20 0-21.76-13.66-20-14.11-20zm-243.22 23.76a8.63 8.63 0 0 0-3.29 7.29c0 4.78 2.08 6.05 3.29 7.05 4.7 3.7 15.07 2.12 20.93.95v-16.94c-5.32-1.07-16.73-1.96-20.93 1.65zM640 232c0 87.58-80 154.39-165.36 136.43-18.37 33-70.73 70.75-132.2 41.63-41.16 96.05-177.89 92.18-213.81-5.17C8.91 428.78-50.19 266.52 53.36 205.61 18.61 126.18 76 32 167.67 32a124.24 124.24 0 0 1 98.56 48.7c20.7-21.4 49.4-34.81 81.15-34.81 42.34 0 79 23.52 98.8 58.57C539 63.78 640 132.69 640 232zm-519.55 31.8c0-11.76-11.69-15.17-17.87-17.17-5.27-2.11-13.41-3.51-13.41-8.94 0-9.46 17-6.66 25.17-2.12 0 0 1.17.71 1.64-.47.24-.7 2.36-6.58 2.59-7.29a1.13 1.13 0 0 0-.7-1.41c-12.33-7.63-40.7-8.51-40.7 12.7 0 12.46 11.49 15.44 17.88 17.17 4.72 1.58 13.17 3 13.17 8.7 0 4-3.53 7.06-9.17 7.06a31.76 31.76 0 0 1-19-6.35c-.47-.23-1.42-.71-1.65.71l-2.4 7.47c-.47.94.23 1.18.23 1.41 1.75 1.4 10.3 6.59 22.82 6.59 13.17 0 21.4-7.06 21.4-18.11zm32-42.58c-10.13 0-18.66 3.17-21.4 5.18a1 1 0 0 0-.24 1.41l2.59 7.06a1 1 0 0 0 1.18.7c.65 0 6.8-4 16.93-4 4 0 7.06.71 9.18 2.36 3.6 2.8 3.06 8.29 3.06 10.58-4.79-.3-19.11-3.44-29.41 3.76a16.92 16.92 0 0 0-7.34 14.54c0 5.9 1.51 10.4 6.59 14.35 12.24 8.16 36.28 2 38.1 1.41 1.58-.32 3.53-.66 3.53-1.88v-33.88c.04-4.61.32-21.64-22.78-21.64zM199 200.24a1.11 1.11 0 0 0-1.18-1.18H188a1.11 1.11 0 0 0-1.17 1.18v79a1.11 1.11 0 0 0 1.17 1.18h9.88a1.11 1.11 0 0 0 1.18-1.18zm55.75 28.93c-2.1-2.31-6.79-7.53-17.65-7.53-3.51 0-14.16.23-20.7 8.94-6.35 7.63-6.58 18.11-6.58 21.41 0 3.12.15 14.26 7.06 21.17 2.64 2.91 9.06 8.23 22.81 8.23 10.82 0 16.47-2.35 18.58-3.76.47-.24.71-.71.24-1.88l-2.35-6.83a1.26 1.26 0 0 0-1.41-.7c-2.59.94-6.35 2.82-15.29 2.82-17.42 0-16.85-14.74-16.94-16.7h37.17a1.23 1.23 0 0 0 1.17-.94c-.29 0 2.07-14.7-6.09-24.23zm36.69 52.69c13.17 0 21.41-7.06 21.41-18.11 0-11.76-11.7-15.17-17.88-17.17-4.14-1.66-13.41-3.38-13.41-8.94 0-3.76 3.29-6.35 8.47-6.35a38.11 38.11 0 0 1 16.7 4.23s1.18.71 1.65-.47c.23-.7 2.35-6.58 2.58-7.29a1.13 1.13 0 0 0-.7-1.41c-7.91-4.9-16.74-4.94-20.23-4.94-12 0-20.46 7.29-20.46 17.64 0 12.46 11.48 15.44 17.87 17.17 6.11 2 13.17 3.26 13.17 8.7 0 4-3.52 7.06-9.17 7.06a31.8 31.8 0 0 1-19-6.35 1 1 0 0 0-1.65.71l-2.35 7.52c-.47.94.23 1.18.23 1.41 1.72 1.4 10.33 6.59 22.79 6.59zM357.09 224c0-.71-.24-1.18-1.18-1.18h-11.76c0-.14.94-8.94 4.47-12.47 4.16-4.15 11.76-1.64 12-1.64 1.17.47 1.41 0 1.64-.47l2.83-7.77c.7-.94 0-1.17-.24-1.41-5.09-2-17.35-2.87-24.46 4.24-5.48 5.48-7 13.92-8 19.52h-8.47a1.28 1.28 0 0 0-1.17 1.18l-1.42 7.76c0 .7.24 1.17 1.18 1.17h8.23c-8.51 47.9-8.75 50.21-10.35 55.52-1.08 3.62-3.29 6.9-5.88 7.76-.09 0-3.88 1.68-9.64-.24 0 0-.94-.47-1.41.71-.24.71-2.59 6.82-2.83 7.53s0 1.41.47 1.41c5.11 2 13 1.77 17.88 0 6.28-2.28 9.72-7.89 11.53-12.94 2.75-7.71 2.81-9.79 11.76-59.74h12.23a1.29 1.29 0 0 0 1.18-1.18zm53.39 16c-.56-1.68-5.1-18.11-25.17-18.11-15.25 0-23 10-25.16 18.11-1 3-3.18 14 0 23.52.09.3 4.41 18.12 25.16 18.12 14.95 0 22.9-9.61 25.17-18.12 3.21-9.61 1.01-20.52 0-23.52zm45.4-16.7c-5-1.65-16.62-1.9-22.11 5.41v-4.47a1.11 1.11 0 0 0-1.18-1.17h-9.4a1.11 1.11 0 0 0-1.18 1.17v55.28a1.12 1.12 0 0 0 1.18 1.18h9.64a1.12 1.12 0 0 0 1.18-1.18v-27.77c0-2.91.05-11.37 4.46-15.05 4.9-4.9 12-3.36 13.41-3.06a1.57 1.57 0 0 0 1.41-.94 74 74 0 0 0 3.06-8 1.16 1.16 0 0 0-.47-1.41zm46.81 54.1l-2.12-7.29c-.47-1.18-1.41-.71-1.41-.71-4.23 1.82-10.15 1.89-11.29 1.89-4.64 0-17.17-1.13-17.17-19.76 0-6.23 1.85-19.76 16.47-19.76a34.85 34.85 0 0 1 11.52 1.65s.94.47 1.18-.71c.94-2.59 1.64-4.47 2.59-7.53.23-.94-.47-1.17-.71-1.17-11.59-3.87-22.34-2.53-27.76 0-1.59.74-16.23 6.49-16.23 27.52 0 2.9-.58 30.11 28.94 30.11a44.45 44.45 0 0 0 15.52-2.83 1.3 1.3 0 0 0 .47-1.42zm53.87-39.52c-.8-3-5.37-16.23-22.35-16.23-16 0-23.52 10.11-25.64 18.59a38.58 38.58 0 0 0-1.65 11.76c0 25.87 18.84 29.4 29.88 29.4 10.82 0 16.46-2.35 18.58-3.76.47-.24.71-.71.24-1.88l-2.36-6.83a1.26 1.26 0 0 0-1.41-.7c-2.59.94-6.35 2.82-15.29 2.82-17.42 0-16.85-14.74-16.93-16.7h37.16a1.25 1.25 0 0 0 1.18-.94c-.24-.01.94-7.07-1.41-15.54zm-23.29-6.35c-10.33 0-13 9-13.64 14.12H546c-.88-11.92-7.62-14.13-12.73-14.13z\"],\"octopus-deploy\":[512,512,[],\"e082\",\"M455.6,349.2c-45.891-39.09-36.67-77.877-16.095-128.11C475.16,134.04,415.967,34.14,329.93,8.3,237.04-19.6,134.252,24.341,99.677,117.147a180.862,180.862,0,0,0-10.988,73.544c1.733,29.543,14.717,52.97,24.09,80.3,17.2,50.161-28.1,92.743-66.662,117.582-46.806,30.2-36.319,39.857-8.428,41.858,23.378,1.68,44.478-4.548,65.265-15.045,9.2-4.647,40.687-18.931,45.13-28.588C135.9,413.388,111.122,459.5,126.621,488.9c19.1,36.229,67.112-31.77,76.709-45.812,8.591-12.572,42.963-81.279,63.627-46.926,18.865,31.361,8.6,76.391,35.738,104.622,32.854,34.2,51.155-18.312,51.412-44.221.163-16.411-6.1-95.852,29.9-59.944C405.428,418,436.912,467.8,472.568,463.642c38.736-4.516-22.123-67.967-28.262-78.695,5.393,4.279,53.665,34.128,53.818,9.52C498.234,375.678,468.039,359.8,455.6,349.2Z\"],medapps:[320,512,[],\"f3c6\",\"M118.3 238.4c3.5-12.5 6.9-33.6 13.2-33.6 8.3 1.8 9.6 23.4 18.6 36.6 4.6-23.5 5.3-85.1 14.1-86.7 9-.7 19.7 66.5 22 77.5 9.9 4.1 48.9 6.6 48.9 6.6 1.9 7.3-24 7.6-40 7.8-4.6 14.8-5.4 27.7-11.4 28-4.7.2-8.2-28.8-17.5-49.6l-9.4 65.5c-4.4 13-15.5-22.5-21.9-39.3-3.3-.1-62.4-1.6-47.6-7.8l31-5zM228 448c21.2 0 21.2-32 0-32H92c-21.2 0-21.2 32 0 32h136zm-24 64c21.2 0 21.2-32 0-32h-88c-21.2 0-21.2 32 0 32h88zm34.2-141.5c3.2-18.9 5.2-36.4 11.9-48.8 7.9-14.7 16.1-28.1 24-41 24.6-40.4 45.9-75.2 45.9-125.5C320 69.6 248.2 0 160 0S0 69.6 0 155.2c0 50.2 21.3 85.1 45.9 125.5 7.9 12.9 16 26.3 24 41 6.7 12.5 8.7 29.8 11.9 48.9 3.5 21 36.1 15.7 32.6-5.1-3.6-21.7-5.6-40.7-15.3-58.6C66.5 246.5 33 211.3 33 155.2 33 87.3 90 32 160 32s127 55.3 127 123.2c0 56.1-33.5 91.3-66.1 151.6-9.7 18-11.7 37.4-15.3 58.6-3.4 20.6 29 26.4 32.6 5.1z\"],ns8:[640,512,[],\"f3d5\",\"M104.324,269.172h26.067V242.994H104.324Zm52.466-26.178-.055-26.178v-.941a39.325,39.325,0,0,0-78.644.941v.166h26.4v-.166a12.98,12.98,0,0,1,25.956,0v26.178Zm52.356,25.846a91.1,91.1,0,0,1-91.1,91.1h-.609a91.1,91.1,0,0,1-91.1-91.1H0v.166A117.33,117.33,0,0,0,117.44,386.28h.775A117.331,117.331,0,0,0,235.49,268.84V242.828H209.146Zm-157.233,0a65.362,65.362,0,0,0,130.723,0H156.292a39.023,39.023,0,0,1-78.035,0V242.883H51.968v-26.62A65.42,65.42,0,0,1,182.8,217.48v25.293h26.344V217.48a91.761,91.761,0,0,0-183.522,0v25.4H51.913Zm418.4-71.173c13.67,0,24.573,6.642,30.052,18.264l.719,1.549,23.245-11.511-.609-1.439c-8.025-19.26-28.5-31.27-53.407-31.27-23.134,0-43.611,11.4-50.972,28.447-.123,26.876-.158,23.9,0,24.85,4.7,11.013,14.555,19.37,28.668,24.241a102.033,102.033,0,0,0,19.813,3.984c5.479.72,10.626,1.384,15.829,3.1,6.364,2.1,10.46,5.257,12.84,9.851v9.851c-3.708,7.527-13.781,12.342-25.791,12.342-14.334,0-25.956-6.918-31.933-19.039l-.72-1.494L415.026,280.9l.553,1.439c7.915,19.426,29.609,32.044,55.289,32.044,23.632,0,44.608-11.4,52.3-28.447l.166-25.9-.166-.664c-4.87-11.014-15.219-19.647-28.944-24.241-7.693-2.712-14.335-3.6-20.7-4.427a83.777,83.777,0,0,1-14.832-2.878c-6.31-1.937-10.4-5.092-12.619-9.63v-8.412C449.45,202.427,458.969,197.667,470.315,197.667ZM287.568,311.344h26.067v-68.4H287.568Zm352.266-53.3c-2.933-6.254-8.3-12.01-15.441-16.714A37.99,37.99,0,0,0,637.4,226l.166-25.347-.166-.664C630.038,184,610.667,173.26,589.25,173.26S548.461,184,541.1,199.992l-.166,25.347.166.664a39.643,39.643,0,0,0,13.006,15.331c-7.2,4.7-12.508,10.46-15.441,16.714l-.166,28.889.166.72c7.582,15.994,27.893,26.731,50.585,26.731s43.057-10.737,50.584-26.731l.166-28.89Zm-73.22-50.806c3.6-6.31,12.563-10.516,22.58-10.516s19.038,4.206,22.636,10.516v13.725c-3.542,6.2-12.563,10.349-22.636,10.349s-19.094-4.15-22.58-10.349Zm47.319,72.169c-3.764,6.641-13.338,10.9-24.683,10.9-11.125,0-20.976-4.372-24.684-10.9V263.25c3.708-6.309,13.5-10.515,24.684-10.515,11.345,0,20.919,4.15,24.683,10.515ZM376.4,265.962l-59.827-89.713h-29v40.623h26.51v.387l62.539,94.085H402.3V176.249H376.4Z\"],\"pinterest-p\":[384,512,[],\"f231\",\"M204 6.5C101.4 6.5 0 74.9 0 185.6 0 256 39.6 296 63.6 296c9.9 0 15.6-27.6 15.6-35.4 0-9.3-23.7-29.1-23.7-67.8 0-80.4 61.2-137.4 140.4-137.4 68.1 0 118.5 38.7 118.5 109.8 0 53.1-21.3 152.7-90.3 152.7-24.9 0-46.2-18-46.2-43.8 0-37.8 26.4-74.4 26.4-113.4 0-66.2-93.9-54.2-93.9 25.8 0 16.8 2.1 35.4 9.6 50.7-13.8 59.4-42 147.9-42 209.1 0 18.9 2.7 37.5 4.5 56.4 3.4 3.8 1.7 3.4 6.9 1.5 50.4-69 48.6-82.5 71.4-172.8 12.3 23.4 44.1 36 69.3 36 106.2 0 153.9-103.5 153.9-196.8C384 71.3 298.2 6.5 204 6.5z\"],apper:[640,512,[],\"f371\",\"M42.1 239.1c22.2 0 29 2.8 33.5 14.6h.8v-22.9c0-11.3-4.8-15.4-17.9-15.4-11.3 0-14.4 2.5-15.1 12.8H4.8c.3-13.9 1.5-19.1 5.8-24.4C17.9 195 29.5 192 56.7 192c33 0 47.1 5 53.9 18.9 2 4.3 4 15.6 4 23.7v76.3H76.3l1.3-19.1h-1c-5.3 15.6-13.6 20.4-35.5 20.4-30.3 0-41.1-10.1-41.1-37.3 0-25.2 12.3-35.8 42.1-35.8zm17.1 48.1c13.1 0 16.9-3 16.9-13.4 0-9.1-4.3-11.6-19.6-11.6-13.1 0-17.9 3-17.9 12.1-.1 10.4 3.7 12.9 20.6 12.9zm77.8-94.9h38.3l-1.5 20.6h.8c9.1-17.1 15.9-20.9 37.5-20.9 14.4 0 24.7 3 31.5 9.1 9.8 8.6 12.8 20.4 12.8 48.1 0 30-3 43.1-12.1 52.9-6.8 7.3-16.4 10.1-33.2 10.1-20.4 0-29.2-5.5-33.8-21.2h-.8v70.3H137v-169zm80.9 60.7c0-27.5-3.3-32.5-20.7-32.5-16.9 0-20.7 5-20.7 28.7 0 28 3.5 33.5 21.2 33.5 16.4 0 20.2-5.6 20.2-29.7zm57.9-60.7h38.3l-1.5 20.6h.8c9.1-17.1 15.9-20.9 37.5-20.9 14.4 0 24.7 3 31.5 9.1 9.8 8.6 12.8 20.4 12.8 48.1 0 30-3 43.1-12.1 52.9-6.8 7.3-16.4 10.1-33.3 10.1-20.4 0-29.2-5.5-33.8-21.2h-.8v70.3h-39.5v-169zm80.9 60.7c0-27.5-3.3-32.5-20.7-32.5-16.9 0-20.7 5-20.7 28.7 0 28 3.5 33.5 21.2 33.5 16.4 0 20.2-5.6 20.2-29.7zm53.8-3.8c0-25.4 3.3-37.8 12.3-45.8 8.8-8.1 22.2-11.3 45.1-11.3 42.8 0 55.7 12.8 55.7 55.7v11.1h-75.3c-.3 2-.3 4-.3 4.8 0 16.9 4.5 21.9 20.1 21.9 13.9 0 17.9-3 17.9-13.9h37.5v2.3c0 9.8-2.5 18.9-6.8 24.7-7.3 9.8-19.6 13.6-44.3 13.6-27.5 0-41.6-3.3-50.6-12.3-8.5-8.5-11.3-21.3-11.3-50.8zm76.4-11.6c-.3-1.8-.3-3.3-.3-3.8 0-12.3-3.3-14.6-19.6-14.6-14.4 0-17.1 3-18.1 15.1l-.3 3.3h38.3zm55.6-45.3h38.3l-1.8 19.9h.7c6.8-14.9 14.4-20.2 29.7-20.2 10.8 0 19.1 3.3 23.4 9.3 5.3 7.3 6.8 14.4 6.8 34 0 1.5 0 5 .2 9.3h-35c.3-1.8.3-3.3.3-4 0-15.4-2-19.4-10.3-19.4-6.3 0-10.8 3.3-13.1 9.3-1 3-1 4.3-1 12.3v68h-38.3V192.3z\"],\"fort-awesome\":[512,512,[],\"f286\",\"M489.2 287.9h-27.4c-2.6 0-4.6 2-4.6 4.6v32h-36.6V146.2c0-2.6-2-4.6-4.6-4.6h-27.4c-2.6 0-4.6 2-4.6 4.6v32h-36.6v-32c0-2.6-2-4.6-4.6-4.6h-27.4c-2.6 0-4.6 2-4.6 4.6v32h-36.6v-32c0-6-8-4.6-11.7-4.6v-38c8.3-2 17.1-3.4 25.7-3.4 10.9 0 20.9 4.3 31.4 4.3 4.6 0 27.7-1.1 27.7-8v-60c0-2.6-2-4.6-4.6-4.6-5.1 0-15.1 4.3-24 4.3-9.7 0-20.9-4.3-32.6-4.3-8 0-16 1.1-23.7 2.9v-4.9c5.4-2.6 9.1-8.3 9.1-14.3 0-20.7-31.4-20.8-31.4 0 0 6 3.7 11.7 9.1 14.3v111.7c-3.7 0-11.7-1.4-11.7 4.6v32h-36.6v-32c0-2.6-2-4.6-4.6-4.6h-27.4c-2.6 0-4.6 2-4.6 4.6v32H128v-32c0-2.6-2-4.6-4.6-4.6H96c-2.6 0-4.6 2-4.6 4.6v178.3H54.8v-32c0-2.6-2-4.6-4.6-4.6H22.8c-2.6 0-4.6 2-4.6 4.6V512h182.9v-96c0-72.6 109.7-72.6 109.7 0v96h182.9V292.5c.1-2.6-1.9-4.6-4.5-4.6zm-288.1-4.5c0 2.6-2 4.6-4.6 4.6h-27.4c-2.6 0-4.6-2-4.6-4.6v-64c0-2.6 2-4.6 4.6-4.6h27.4c2.6 0 4.6 2 4.6 4.6v64zm146.4 0c0 2.6-2 4.6-4.6 4.6h-27.4c-2.6 0-4.6-2-4.6-4.6v-64c0-2.6 2-4.6 4.6-4.6h27.4c2.6 0 4.6 2 4.6 4.6v64z\"],waze:[512,512,[],\"f83f\",\"M502.17 201.67C516.69 287.53 471.23 369.59 389 409.8c13 34.1-12.4 70.2-48.32 70.2a51.68 51.68 0 0 1-51.57-49c-6.44.19-64.2 0-76.33-.64A51.69 51.69 0 0 1 159 479.92c-33.86-1.36-57.95-34.84-47-67.92-37.21-13.11-72.54-34.87-99.62-70.8-13-17.28-.48-41.8 20.84-41.8 46.31 0 32.22-54.17 43.15-110.26C94.8 95.2 193.12 32 288.09 32c102.48 0 197.15 70.67 214.08 169.67zM373.51 388.28c42-19.18 81.33-56.71 96.29-102.14 40.48-123.09-64.15-228-181.71-228-83.45 0-170.32 55.42-186.07 136-9.53 48.91 5 131.35-68.75 131.35C58.21 358.6 91.6 378.11 127 389.54c24.66-21.8 63.87-15.47 79.83 14.34 14.22 1 79.19 1.18 87.9.82a51.69 51.69 0 0 1 78.78-16.42zM205.12 187.13c0-34.74 50.84-34.75 50.84 0s-50.84 34.74-50.84 0zm116.57 0c0-34.74 50.86-34.75 50.86 0s-50.86 34.75-50.86 0zm-122.61 70.69c-3.44-16.94 22.18-22.18 25.62-5.21l.06.28c4.14 21.42 29.85 44 64.12 43.07 35.68-.94 59.25-22.21 64.11-42.77 4.46-16.05 28.6-10.36 25.47 6-5.23 22.18-31.21 62-91.46 62.9-42.55 0-80.88-27.84-87.9-64.25z\"],bluesky:[576,512,[],\"e671\",\"M407.8 294.7c-3.3-.4-6.7-.8-10-1.3c3.4 .4 6.7 .9 10 1.3zM288 227.1C261.9 176.4 190.9 81.9 124.9 35.3C61.6-9.4 37.5-1.7 21.6 5.5C3.3 13.8 0 41.9 0 58.4S9.1 194 15 213.9c19.5 65.7 89.1 87.9 153.2 80.7c3.3-.5 6.6-.9 10-1.4c-3.3 .5-6.6 1-10 1.4C74.3 308.6-9.1 342.8 100.3 464.5C220.6 589.1 265.1 437.8 288 361.1c22.9 76.7 49.2 222.5 185.6 103.4c102.4-103.4 28.1-156-65.8-169.9c-3.3-.4-6.7-.8-10-1.3c3.4 .4 6.7 .9 10 1.3c64.1 7.1 133.6-15.1 153.2-80.7C566.9 194 576 75 576 58.4s-3.3-44.7-21.6-52.9c-15.8-7.1-40-14.9-103.2 29.8C385.1 81.9 314.1 176.4 288 227.1z\"],\"cc-jcb\":[576,512,[],\"f24b\",\"M431.5 244.3V212c41.2 0 38.5.2 38.5.2 7.3 1.3 13.3 7.3 13.3 16 0 8.8-6 14.5-13.3 15.8-1.2.4-3.3.3-38.5.3zm42.8 20.2c-2.8-.7-3.3-.5-42.8-.5v35c39.6 0 40 .2 42.8-.5 7.5-1.5 13.5-8 13.5-17 0-8.7-6-15.5-13.5-17zM576 80v352c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48h480c26.5 0 48 21.5 48 48zM182 192.3h-57c0 67.1 10.7 109.7-35.8 109.7-19.5 0-38.8-5.7-57.2-14.8v28c30 8.3 68 8.3 68 8.3 97.9 0 82-47.7 82-131.2zm178.5 4.5c-63.4-16-165-14.9-165 59.3 0 77.1 108.2 73.6 165 59.2V287C312.9 311.7 253 309 253 256s59.8-55.6 107.5-31.2v-28zM544 286.5c0-18.5-16.5-30.5-38-32v-.8c19.5-2.7 30.3-15.5 30.3-30.2 0-19-15.7-30-37-31 0 0 6.3-.3-120.3-.3v127.5h122.7c24.3.1 42.3-12.9 42.3-33.2z\"],snapchat:[512,512,[62124,\"snapchat-ghost\"],\"f2ab\",\"M496.926,366.6c-3.373-9.176-9.8-14.086-17.112-18.153-1.376-.806-2.641-1.451-3.72-1.947-2.182-1.128-4.414-2.22-6.634-3.373-22.8-12.09-40.609-27.341-52.959-45.42a102.889,102.889,0,0,1-9.089-16.12c-1.054-3.013-1-4.724-.248-6.287a10.221,10.221,0,0,1,2.914-3.038c3.918-2.591,7.96-5.22,10.7-6.993,4.885-3.162,8.754-5.667,11.246-7.44,9.362-6.547,15.909-13.5,20-21.278a42.371,42.371,0,0,0,2.1-35.191c-6.2-16.318-21.613-26.449-40.287-26.449a55.543,55.543,0,0,0-11.718,1.24c-1.029.224-2.059.459-3.063.72.174-11.16-.074-22.94-1.066-34.534-3.522-40.758-17.794-62.123-32.674-79.16A130.167,130.167,0,0,0,332.1,36.443C309.515,23.547,283.91,17,256,17S202.6,23.547,180,36.443a129.735,129.735,0,0,0-33.281,26.783c-14.88,17.038-29.152,38.44-32.673,79.161-.992,11.594-1.24,23.435-1.079,34.533-1-.26-2.021-.5-3.051-.719a55.461,55.461,0,0,0-11.717-1.24c-18.687,0-34.125,10.131-40.3,26.449a42.423,42.423,0,0,0,2.046,35.228c4.105,7.774,10.652,14.731,20.014,21.278,2.48,1.736,6.361,4.24,11.246,7.44,2.641,1.711,6.5,4.216,10.28,6.72a11.054,11.054,0,0,1,3.3,3.311c.794,1.624.818,3.373-.36,6.6a102.02,102.02,0,0,1-8.94,15.785c-12.077,17.669-29.363,32.648-51.434,44.639C32.355,348.608,20.2,352.75,15.069,366.7c-3.868,10.528-1.339,22.506,8.494,32.6a49.137,49.137,0,0,0,12.4,9.387,134.337,134.337,0,0,0,30.342,12.139,20.024,20.024,0,0,1,6.126,2.741c3.583,3.137,3.075,7.861,7.849,14.78a34.468,34.468,0,0,0,8.977,9.127c10.019,6.919,21.278,7.353,33.207,7.811,10.776.41,22.989.881,36.939,5.481,5.778,1.91,11.78,5.605,18.736,9.92C194.842,480.951,217.707,495,255.973,495s61.292-14.123,78.118-24.428c6.907-4.24,12.872-7.9,18.489-9.758,13.949-4.613,26.163-5.072,36.939-5.481,11.928-.459,23.187-.893,33.206-7.812a34.584,34.584,0,0,0,10.218-11.16c3.434-5.84,3.348-9.919,6.572-12.771a18.971,18.971,0,0,1,5.753-2.629A134.893,134.893,0,0,0,476.02,408.71a48.344,48.344,0,0,0,13.019-10.193l.124-.149C498.389,388.5,500.708,376.867,496.926,366.6Zm-34.013,18.277c-20.745,11.458-34.533,10.23-45.259,17.137-9.114,5.865-3.72,18.513-10.342,23.076-8.134,5.617-32.177-.4-63.239,9.858-25.618,8.469-41.961,32.822-88.038,32.822s-62.036-24.3-88.076-32.884c-31-10.255-55.092-4.241-63.239-9.858-6.609-4.563-1.24-17.211-10.341-23.076-10.739-6.907-24.527-5.679-45.26-17.075-13.206-7.291-5.716-11.8-1.314-13.937,75.143-36.381,87.133-92.552,87.666-96.719.645-5.046,1.364-9.014-4.191-14.148-5.369-4.96-29.189-19.7-35.8-24.316-10.937-7.638-15.748-15.264-12.2-24.638,2.48-6.485,8.531-8.928,14.879-8.928a27.643,27.643,0,0,1,5.965.67c12,2.6,23.659,8.617,30.392,10.242a10.749,10.749,0,0,0,2.48.335c3.6,0,4.86-1.811,4.612-5.927-.768-13.132-2.628-38.725-.558-62.644,2.84-32.909,13.442-49.215,26.04-63.636,6.051-6.932,34.484-36.976,88.857-36.976s82.88,29.92,88.931,36.827c12.611,14.421,23.225,30.727,26.04,63.636,2.071,23.919.285,49.525-.558,62.644-.285,4.327,1.017,5.927,4.613,5.927a10.648,10.648,0,0,0,2.48-.335c6.745-1.624,18.4-7.638,30.4-10.242a27.641,27.641,0,0,1,5.964-.67c6.386,0,12.4,2.48,14.88,8.928,3.546,9.374-1.24,17-12.189,24.639-6.609,4.612-30.429,19.343-35.8,24.315-5.568,5.134-4.836,9.1-4.191,14.149.533,4.228,12.511,60.4,87.666,96.718C468.629,373.011,476.119,377.524,462.913,384.877Z\"],\"fantasy-flight-games\":[512,512,[],\"f6dc\",\"M256 32.86L32.86 256 256 479.14 479.14 256 256 32.86zM88.34 255.83c1.96-2 11.92-12.3 96.49-97.48 41.45-41.75 86.19-43.77 119.77-18.69 24.63 18.4 62.06 58.9 62.15 59 .68.74 1.07 2.86.58 3.38-11.27 11.84-22.68 23.54-33.5 34.69-34.21-32.31-40.52-38.24-48.51-43.95-17.77-12.69-41.4-10.13-56.98 5.1-2.17 2.13-1.79 3.43.12 5.35 2.94 2.95 28.1 28.33 35.09 35.78-11.95 11.6-23.66 22.97-35.69 34.66-12.02-12.54-24.48-25.53-36.54-38.11-21.39 21.09-41.69 41.11-61.85 60.99zm234.82 101.6c-35.49 35.43-78.09 38.14-106.99 20.47-22.08-13.5-39.38-32.08-72.93-66.84 12.05-12.37 23.79-24.42 35.37-36.31 33.02 31.91 37.06 36.01 44.68 42.09 18.48 14.74 42.52 13.67 59.32-1.8 3.68-3.39 3.69-3.64.14-7.24-10.59-10.73-21.19-21.44-31.77-32.18-1.32-1.34-3.03-2.48-.8-4.69 10.79-10.71 21.48-21.52 32.21-32.29.26-.26.65-.38 1.91-1.07 12.37 12.87 24.92 25.92 37.25 38.75 21.01-20.73 41.24-40.68 61.25-60.42 13.68 13.4 27.13 26.58 40.86 40.03-20.17 20.86-81.68 82.71-100.5 101.5zM256 0L0 256l256 256 256-256L256 0zM16 256L256 16l240 240-240 240L16 256z\"],rust:[512,512,[],\"e07a\",\"M508.52,249.75,486.7,236.24c-.17-2-.34-3.93-.55-5.88l18.72-17.5a7.35,7.35,0,0,0-2.44-12.25l-24-9c-.54-1.88-1.08-3.78-1.67-5.64l15-20.83a7.35,7.35,0,0,0-4.79-11.54l-25.42-4.15c-.9-1.73-1.79-3.45-2.73-5.15l10.68-23.42a7.35,7.35,0,0,0-6.95-10.39l-25.82.91q-1.79-2.22-3.61-4.4L439,81.84A7.36,7.36,0,0,0,430.16,73L405,78.93q-2.17-1.83-4.4-3.61l.91-25.82a7.35,7.35,0,0,0-10.39-7L367.7,53.23c-1.7-.94-3.43-1.84-5.15-2.73L358.4,25.08a7.35,7.35,0,0,0-11.54-4.79L326,35.26c-1.86-.59-3.75-1.13-5.64-1.67l-9-24a7.35,7.35,0,0,0-12.25-2.44l-17.5,18.72c-1.95-.21-3.91-.38-5.88-.55L262.25,3.48a7.35,7.35,0,0,0-12.5,0L236.24,25.3c-2,.17-3.93.34-5.88.55L212.86,7.13a7.35,7.35,0,0,0-12.25,2.44l-9,24c-1.89.55-3.79,1.08-5.66,1.68l-20.82-15a7.35,7.35,0,0,0-11.54,4.79l-4.15,25.41c-1.73.9-3.45,1.79-5.16,2.73L120.88,42.55a7.35,7.35,0,0,0-10.39,7l.92,25.81c-1.49,1.19-3,2.39-4.42,3.61L81.84,73A7.36,7.36,0,0,0,73,81.84L78.93,107c-1.23,1.45-2.43,2.93-3.62,4.41l-25.81-.91a7.42,7.42,0,0,0-6.37,3.26,7.35,7.35,0,0,0-.57,7.13l10.66,23.41c-.94,1.7-1.83,3.43-2.73,5.16L25.08,153.6a7.35,7.35,0,0,0-4.79,11.54l15,20.82c-.59,1.87-1.13,3.77-1.68,5.66l-24,9a7.35,7.35,0,0,0-2.44,12.25l18.72,17.5c-.21,1.95-.38,3.91-.55,5.88L3.48,249.75a7.35,7.35,0,0,0,0,12.5L25.3,275.76c.17,2,.34,3.92.55,5.87L7.13,299.13a7.35,7.35,0,0,0,2.44,12.25l24,9c.55,1.89,1.08,3.78,1.68,5.65l-15,20.83a7.35,7.35,0,0,0,4.79,11.54l25.42,4.15c.9,1.72,1.79,3.45,2.73,5.14L42.56,391.12a7.35,7.35,0,0,0,.57,7.13,7.13,7.13,0,0,0,6.37,3.26l25.83-.91q1.77,2.22,3.6,4.4L73,430.16A7.36,7.36,0,0,0,81.84,439L107,433.07q2.18,1.83,4.41,3.61l-.92,25.82a7.35,7.35,0,0,0,10.39,6.95l23.43-10.68c1.69.94,3.42,1.83,5.14,2.73l4.15,25.42a7.34,7.34,0,0,0,11.54,4.78l20.83-15c1.86.6,3.76,1.13,5.65,1.68l9,24a7.36,7.36,0,0,0,12.25,2.44l17.5-18.72c1.95.21,3.92.38,5.88.55l13.51,21.82a7.35,7.35,0,0,0,12.5,0l13.51-21.82c2-.17,3.93-.34,5.88-.56l17.5,18.73a7.36,7.36,0,0,0,12.25-2.44l9-24c1.89-.55,3.78-1.08,5.65-1.68l20.82,15a7.34,7.34,0,0,0,11.54-4.78l4.15-25.42c1.72-.9,3.45-1.79,5.15-2.73l23.42,10.68a7.35,7.35,0,0,0,10.39-6.95l-.91-25.82q2.22-1.79,4.4-3.61L430.16,439a7.36,7.36,0,0,0,8.84-8.84L433.07,405q1.83-2.17,3.61-4.4l25.82.91a7.23,7.23,0,0,0,6.37-3.26,7.35,7.35,0,0,0,.58-7.13L458.77,367.7c.94-1.7,1.83-3.43,2.73-5.15l25.42-4.15a7.35,7.35,0,0,0,4.79-11.54l-15-20.83c.59-1.87,1.13-3.76,1.67-5.65l24-9a7.35,7.35,0,0,0,2.44-12.25l-18.72-17.5c.21-1.95.38-3.91.55-5.87l21.82-13.51a7.35,7.35,0,0,0,0-12.5Zm-151,129.08A13.91,13.91,0,0,0,341,389.51l-7.64,35.67A187.51,187.51,0,0,1,177,424.44l-7.64-35.66a13.87,13.87,0,0,0-16.46-10.68l-31.51,6.76a187.38,187.38,0,0,1-16.26-19.21H258.3c1.72,0,2.89-.29,2.89-1.91V309.55c0-1.57-1.17-1.91-2.89-1.91H213.47l.05-34.35H262c4.41,0,23.66,1.28,29.79,25.87,1.91,7.55,6.17,32.14,9.06,40,2.89,8.82,14.6,26.46,27.1,26.46H407a187.3,187.3,0,0,1-17.34,20.09Zm25.77,34.49A15.24,15.24,0,1,1,368,398.08h.44A15.23,15.23,0,0,1,383.24,413.32Zm-225.62-.68a15.24,15.24,0,1,1-15.25-15.25h.45A15.25,15.25,0,0,1,157.62,412.64ZM69.57,234.15l32.83-14.6a13.88,13.88,0,0,0,7.06-18.33L102.69,186h26.56V305.73H75.65A187.65,187.65,0,0,1,69.57,234.15ZM58.31,198.09a15.24,15.24,0,0,1,15.23-15.25H74a15.24,15.24,0,1,1-15.67,15.24Zm155.16,24.49.05-35.32h63.26c3.28,0,23.07,3.77,23.07,18.62,0,12.29-15.19,16.7-27.68,16.7ZM399,306.71c-9.8,1.13-20.63-4.12-22-10.09-5.78-32.49-15.39-39.4-30.57-51.4,18.86-11.95,38.46-29.64,38.46-53.26,0-25.52-17.49-41.59-29.4-49.48-16.76-11-35.28-13.23-40.27-13.23H116.32A187.49,187.49,0,0,1,221.21,70.06l23.47,24.6a13.82,13.82,0,0,0,19.6.44l26.26-25a187.51,187.51,0,0,1,128.37,91.43l-18,40.57A14,14,0,0,0,408,220.43l34.59,15.33a187.12,187.12,0,0,1,.4,32.54H423.71c-1.91,0-2.69,1.27-2.69,3.13v8.82C421,301,409.31,305.58,399,306.71ZM240,60.21A15.24,15.24,0,0,1,255.21,45h.45A15.24,15.24,0,1,1,240,60.21ZM436.84,214a15.24,15.24,0,1,1,0-30.48h.44a15.24,15.24,0,0,1-.44,30.48Z\"],wix:[640,512,[],\"f5cf\",\"M393.38 131.69c0 13.03 2.08 32.69-28.68 43.83-9.52 3.45-15.95 9.66-15.95 9.66 0-31 4.72-42.22 17.4-48.86 9.75-5.11 27.23-4.63 27.23-4.63zm-115.8 35.54l-34.24 132.66-28.48-108.57c-7.69-31.99-20.81-48.53-48.43-48.53-27.37 0-40.66 16.18-48.43 48.53L89.52 299.89 55.28 167.23C49.73 140.51 23.86 128.96 0 131.96l65.57 247.93s21.63 1.56 32.46-3.96c14.22-7.25 20.98-12.84 29.59-46.57 7.67-30.07 29.11-118.41 31.12-124.7 4.76-14.94 11.09-13.81 15.4 0 1.97 6.3 23.45 94.63 31.12 124.7 8.6 33.73 15.37 39.32 29.59 46.57 10.82 5.52 32.46 3.96 32.46 3.96l65.57-247.93c-24.42-3.07-49.82 8.93-55.3 35.27zm115.78 5.21s-4.1 6.34-13.46 11.57c-6.01 3.36-11.78 5.64-17.97 8.61-15.14 7.26-13.18 13.95-13.18 35.2v152.07s16.55 2.09 27.37-3.43c13.93-7.1 17.13-13.95 17.26-44.78V181.41l-.02.01v-8.98zm163.44 84.08L640 132.78s-35.11-5.98-52.5 9.85c-13.3 12.1-24.41 29.55-54.18 72.47-.47.73-6.25 10.54-13.07 0-29.29-42.23-40.8-60.29-54.18-72.47-17.39-15.83-52.5-9.85-52.5-9.85l83.2 123.74-82.97 123.36s36.57 4.62 53.95-11.21c11.49-10.46 17.58-20.37 52.51-70.72 6.81-10.52 12.57-.77 13.07 0 29.4 42.38 39.23 58.06 53.14 70.72 17.39 15.83 53.32 11.21 53.32 11.21L556.8 256.52z\"],\"square-behance\":[448,512,[\"behance-square\"],\"f1b5\",\"M155.3 318.4c17.2 0 31.2-6.1 31.2-25.4c0-19.7-11.7-27.4-30.3-27.5h-46v52.9h45.1zm-5.4-129.6H110.3v44.8H153c15.1 0 25.8-6.6 25.8-22.9c0-17.7-13.7-21.9-28.9-21.9zm129.5 74.8h62.2c-1.7-18.5-11.3-29.7-30.5-29.7c-18.3 0-30.5 11.4-31.7 29.7zM384 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64zM349.5 185H271.7V166.1h77.8V185zM193.7 243.7c23.6 6.7 35 27.5 35 51.6c0 39-32.7 55.7-67.6 55.9H68v-192h90.5c32.9 0 61.4 9.3 61.4 47.5c0 19.3-9 28.8-26.2 37zm118.7-38.6c43.5 0 67.6 34.3 67.6 75.4c0 1.6-.1 3.3-.2 5c0 .8-.1 1.5-.1 2.2H279.5c0 22.2 11.7 35.3 34.1 35.3c11.6 0 26.5-6.2 30.2-18.1h33.7c-10.4 31.9-31.9 46.8-65.1 46.8c-43.8 0-71.1-29.7-71.1-73c0-41.8 28.7-73.6 71.1-73.6z\"],supple:[640,512,[],\"f3f9\",\"M640 262.5c0 64.1-109 116.1-243.5 116.1-24.8 0-48.6-1.8-71.1-5 7.7.4 15.5.6 23.4.6 134.5 0 243.5-56.9 243.5-127.1 0-29.4-19.1-56.4-51.2-78 60 21.1 98.9 55.1 98.9 93.4zM47.7 227.9c-.1-70.2 108.8-127.3 243.3-127.6 7.9 0 15.6.2 23.3.5-22.5-3.2-46.3-4.9-71-4.9C108.8 96.3-.1 148.5 0 212.6c.1 38.3 39.1 72.3 99.3 93.3-32.3-21.5-51.5-48.6-51.6-78zm60.2 39.9s10.5 13.2 29.3 13.2c17.9 0 28.4-11.5 28.4-25.1 0-28-40.2-25.1-40.2-39.7 0-5.4 5.3-9.1 12.5-9.1 5.7 0 11.3 2.6 11.3 6.6v3.9h14.2v-7.9c0-12.1-15.4-16.8-25.4-16.8-16.5 0-28.5 10.2-28.5 24.1 0 26.6 40.2 25.4 40.2 39.9 0 6.6-5.8 10.1-12.3 10.1-11.9 0-20.7-10.1-20.7-10.1l-8.8 10.9zm120.8-73.6v54.4c0 11.3-7.1 17.8-17.8 17.8-10.7 0-17.8-6.5-17.8-17.7v-54.5h-15.8v55c0 18.9 13.4 31.9 33.7 31.9 20.1 0 33.4-13 33.4-31.9v-55h-15.7zm34.4 85.4h15.8v-29.5h15.5c16 0 27.2-11.5 27.2-28.1s-11.2-27.8-27.2-27.8h-39.1v13.4h7.8v72zm15.8-43v-29.1h12.9c8.7 0 13.7 5.7 13.7 14.4 0 8.9-5.1 14.7-14 14.7h-12.6zm57 43h15.8v-29.5h15.5c16 0 27.2-11.5 27.2-28.1s-11.2-27.8-27.2-27.8h-39.1v13.4h7.8v72zm15.7-43v-29.1h12.9c8.7 0 13.7 5.7 13.7 14.4 0 8.9-5 14.7-14 14.7h-12.6zm57.1 34.8c0 5.8 2.4 8.2 8.2 8.2h37.6c5.8 0 8.2-2.4 8.2-8.2v-13h-14.3v5.2c0 1.7-1 2.6-2.6 2.6h-18.6c-1.7 0-2.6-1-2.6-2.6v-61.2c0-5.7-2.4-8.2-8.2-8.2H401v13.4h5.2c1.7 0 2.6 1 2.6 2.6v61.2zm63.4 0c0 5.8 2.4 8.2 8.2 8.2H519c5.7 0 8.2-2.4 8.2-8.2v-13h-14.3v5.2c0 1.7-1 2.6-2.6 2.6h-19.7c-1.7 0-2.6-1-2.6-2.6v-20.3h27.7v-13.4H488v-22.4h19.2c1.7 0 2.6 1 2.6 2.6v5.2H524v-13c0-5.7-2.5-8.2-8.2-8.2h-51.6v13.4h7.8v63.9zm58.9-76v5.9h1.6v-5.9h2.7v-1.2h-7v1.2h2.7zm5.7-1.2v7.1h1.5v-5.7l2.3 5.7h1.3l2.3-5.7v5.7h1.5v-7.1h-2.3l-2.1 5.1-2.1-5.1h-2.4z\"],webflow:[640,512,[],\"e65c\",\"M640 64L435.8 463.2H244l85.5-165.5h-3.8C255.1 389.3 149.9 449.5 0 463.2V300.1s95.9-5.7 152.3-64.9H0V64H171.1V204.8l3.8 0L244.9 64H374.3V203.9l3.8 0L450.7 64H640z\"],rebel:[512,512,[],\"f1d0\",\"M256.5 504C117.2 504 9 387.8 13.2 249.9 16 170.7 56.4 97.7 129.7 49.5c.3 0 1.9-.6 1.1.8-5.8 5.5-111.3 129.8-14.1 226.4 49.8 49.5 90 2.5 90 2.5 38.5-50.1-.6-125.9-.6-125.9-10-24.9-45.7-40.1-45.7-40.1l28.8-31.8c24.4 10.5 43.2 38.7 43.2 38.7.8-29.6-21.9-61.4-21.9-61.4L255.1 8l44.3 50.1c-20.5 28.8-21.9 62.6-21.9 62.6 13.8-23 43.5-39.3 43.5-39.3l28.5 31.8c-27.4 8.9-45.4 39.9-45.4 39.9-15.8 28.5-27.1 89.4.6 127.3 32.4 44.6 87.7-2.8 87.7-2.8 102.7-91.9-10.5-225-10.5-225-6.1-5.5.8-2.8.8-2.8 50.1 36.5 114.6 84.4 116.2 204.8C500.9 400.2 399 504 256.5 504z\"],css3:[512,512,[],\"f13c\",\"M480 32l-64 368-223.3 80L0 400l19.6-94.8h82l-8 40.6L210 390.2l134.1-44.4 18.8-97.1H29.5l16-82h333.7l10.5-52.7H56.3l16.3-82H480z\"],staylinked:[440,512,[],\"f3f5\",\"M382.7 292.5l2.7 2.7-170-167.3c-3.5-3.5-9.7-3.7-13.8-.5L144.3 171c-4.2 3.2-4.6 8.7-1.1 12.2l68.1 64.3c3.6 3.5 9.9 3.7 14 .5l.1-.1c4.1-3.2 10.4-3 14 .5l84 81.3c3.6 3.5 3.2 9-.9 12.2l-93.2 74c-4.2 3.3-10.5 3.1-14.2-.4L63.2 268c-3.5-3.5-9.7-3.7-13.9-.5L3.5 302.4c-4.2 3.2-4.7 8.7-1.2 12.2L211 510.7s7.4 6.8 17.3-.8l198-163.9c4-3.2 4.4-8.7.7-12.2zm54.5-83.4L226.7 2.5c-1.5-1.2-8-5.5-16.3 1.1L3.6 165.7c-4.2 3.2-4.8 8.7-1.2 12.2l42.3 41.7 171.7 165.1c3.7 3.5 10.1 3.7 14.3.4l50.2-38.8-.3-.3 7.7-6c4.2-3.2 4.6-8.7.9-12.2l-57.1-54.4c-3.6-3.5-10-3.7-14.2-.5l-.1.1c-4.2 3.2-10.5 3.1-14.2-.4L109 180.8c-3.6-3.5-3.1-8.9 1.1-12.2l92.2-71.5c4.1-3.2 10.3-3 13.9.5l160.4 159c3.7 3.5 10 3.7 14.1.5l45.8-35.8c4.1-3.2 4.4-8.7.7-12.2z\"],kaggle:[320,512,[],\"f5fa\",\"M304.2 501.5L158.4 320.3 298.2 185c2.6-2.7 1.7-10.5-5.3-10.5h-69.2c-3.5 0-7 1.8-10.5 5.3L80.9 313.5V7.5q0-7.5-7.5-7.5H21.5Q14 0 14 7.5v497q0 7.5 7.5 7.5h51.9q7.5 0 7.5-7.5v-109l30.8-29.3 110.5 140.6c3 3.5 6.5 5.3 10.5 5.3h66.9q5.25 0 6-3z\"],\"space-awesome\":[512,512,[],\"e5ac\",\"M96 256H128V512H0V352H32V320H64V288H96V256zM512 352V512H384V256H416V288H448V320H480V352H512zM320 64H352V448H320V416H192V448H160V64H192V32H224V0H288V32H320V64zM288 128H224V192H288V128z\"],deviantart:[320,512,[],\"f1bd\",\"M320 93.2l-98.2 179.1 7.4 9.5H320v127.7H159.1l-13.5 9.2-43.7 84c-.3 0-8.6 8.6-9.2 9.2H0v-93.2l93.2-179.4-7.4-9.2H0V102.5h156l13.5-9.2 43.7-84c.3 0 8.6-8.6 9.2-9.2H320v93.1z\"],cpanel:[640,512,[],\"f388\",\"M210.3 220.2c-5.6-24.8-26.9-41.2-51-41.2h-37c-7.1 0-12.5 4.5-14.3 10.9L73.1 320l24.7-.1c6.8 0 12.3-4.5 14.2-10.7l25.8-95.7h19.8c8.4 0 16.2 5.6 18.3 14.8 2.5 10.9-5.9 22.6-18.3 22.6h-10.3c-7 0-12.5 4.6-14.3 10.8l-6.4 23.8h32c37.2 0 58.3-36.2 51.7-65.3zm-156.5 28h18.6c6.9 0 12.4-4.4 14.3-10.9l6.2-23.6h-40C30 213.7 9 227.8 1.7 254.8-7 288.6 18.5 320 52 320h12.4l7.1-26.1c1.2-4.4-2.2-8.3-6.4-8.3H53.8c-24.7 0-24.9-37.4 0-37.4zm247.5-34.8h-77.9l-3.5 13.4c-2.4 9.6 4.5 18.5 14.2 18.5h57.5c4 0 2.4 4.3 2.1 5.3l-8.6 31.8c-.4 1.4-.9 5.3-5.5 5.3h-34.9c-5.3 0-5.3-7.9 0-7.9h21.6c6.8 0 12.3-4.6 14.2-10.8l3.5-13.2h-48.4c-39.2 0-43.6 63.8-.7 63.8l57.5.2c11.2 0 20.6-7.2 23.4-17.8l14-51.8c4.8-19.2-9.7-36.8-28.5-36.8zM633.1 179h-18.9c-4.9 0-9.2 3.2-10.4 7.9L568.2 320c20.7 0 39.8-13.8 44.9-34.5l26.5-98.2c1.2-4.3-2-8.3-6.5-8.3zm-236.3 34.7v.1h-48.3l-26.2 98c-1.2 4.4 2.2 8.3 6.4 8.3h18.9c4.8 0 9.2-3 10.4-7.8l17.2-64H395c12.5 0 21.4 11.8 18.1 23.4l-10.6 40c-1.2 4.3 1.9 8.3 6.4 8.3H428c4.6 0 9.1-2.9 10.3-7.8l8.8-33.1c9-33.1-15.9-65.4-50.3-65.4zm98.3 74.6c-3.6 0-6-3.4-5.1-6.7l8-30c.9-3.9 3.7-6 7.8-6h32.9c2.6 0 4.6 2.4 3.9 5.1l-.7 2.6c-.6 2-1.9 3-3.9 3h-21.6c-7 0-12.6 4.6-14.2 10.8l-3.5 13h53.4c10.5 0 20.3-6.6 23.2-17.6l3.2-12c4.9-19.1-9.3-36.8-28.3-36.8h-47.3c-17.9 0-33.8 12-38.6 29.6l-10.8 40c-5 17.7 8.3 36.7 28.3 36.7h66.7c6.8 0 12.3-4.5 14.2-10.7l5.7-21z\"],\"goodreads-g\":[384,512,[],\"f3a9\",\"M42.6 403.3h2.8c12.7 0 25.5 0 38.2.1 1.6 0 3.1-.4 3.6 2.1 7.1 34.9 30 54.6 62.9 63.9 26.9 7.6 54.1 7.8 81.3 1.8 33.8-7.4 56-28.3 68-60.4 8-21.5 10.7-43.8 11-66.5.1-5.8.3-47-.2-52.8l-.9-.3c-.8 1.5-1.7 2.9-2.5 4.4-22.1 43.1-61.3 67.4-105.4 69.1-103 4-169.4-57-172-176.2-.5-23.7 1.8-46.9 8.3-69.7C58.3 47.7 112.3.6 191.6 0c61.3-.4 101.5 38.7 116.2 70.3.5 1.1 1.3 2.3 2.4 1.9V10.6h44.3c0 280.3.1 332.2.1 332.2-.1 78.5-26.7 143.7-103 162.2-69.5 16.9-159 4.8-196-57.2-8-13.5-11.8-28.3-13-44.5zM188.9 36.5c-52.5-.5-108.5 40.7-115 133.8-4.1 59 14.8 122.2 71.5 148.6 27.6 12.9 74.3 15 108.3-8.7 47.6-33.2 62.7-97 54.8-154-9.7-71.1-47.8-120-119.6-119.7z\"],\"square-git\":[448,512,[\"git-square\"],\"f1d2\",\"M120.8 335.5c-5.9-.4-12.6-.8-20.2-1.3c-3.3 4.1-6.6 8.4-6.6 13.5c0 18.5 65.5 18.5 65.5-1.5c0-8.3-7.4-8.7-38.8-10.7zm7.8-117.9c-32.3 0-33.7 44.5-.7 44.5c32.5 0 31.7-44.5 .7-44.5zM384 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64zM243.9 172.2c-14.5 0-22.9-8.4-22.9-22.9c0-14.5 8.4-22.3 22.9-22.3c14.7 0 23.1 7.8 23.1 22.3s-8.4 22.9-23.1 22.9zM149.6 195h49.5l0 21.6-23.4 1.8c4.6 5.8 9.4 14 9.4 25.7c0 48.7-57.2 47.2-74.2 42.4l-8.4 13.4c5 .3 9.8 .6 14.3 .8c56.3 3.2 80.5 4.6 80.5 38.5c0 29.2-25.7 45.7-69.9 45.7c-46 0-63.5-11.6-63.5-31.7c0-11.4 5.1-17.5 14-25.9c-8.4-3.5-11.2-9.9-11.2-16.8c0-9.6 7.4-16.3 23-30.6l.2-.2c-12.4-6.1-21.8-19.3-21.8-38.1c0-51.6 56.6-53.3 81.6-46.8zM270.5 303.1l13 1.8 0 20.1H211.1V304.9c2.7-.4 5-.7 6.9-.9c9.9-1.2 10.1-1.3 10.1-6V223.3c0-4.4-.9-4.7-10.1-7.8c-1.9-.7-4.2-1.4-6.9-2.4l2.8-20.6h52.6V298c0 4.1 .2 4.6 4.1 5.1zm106.6-10.4L384 315c-10.9 5.4-26.9 10.2-41.4 10.2c-30.2 0-41.7-12.2-41.7-40.9V217.7c0-.8 0-1.4-.2-1.8c-.8-1.2-4.2-.7-19.6-.7V192.6c22.3-2.5 31.2-13.7 34-41.4h24.2c0 33.3-.6 38 .7 38.6c.3 .1 .7 0 1.3 0h35.8v25.4H339.3v60.7c0 .2 0 .5 0 .9c-.2 6.3-.9 30.4 37.9 15.9z\"],\"square-tumblr\":[448,512,[\"tumblr-square\"],\"f174\",\"M448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM256.8 416c-75.5 0-91.9-55.5-91.9-87.9v-90H135.2c-3.4 0-6.2-2.8-6.2-6.2V189.4c0-4.5 2.8-8.5 7.1-10c38.8-13.7 50.9-47.5 52.7-73.2c.5-6.9 4.1-10.2 10-10.2h44.3c3.4 0 6.2 2.8 6.2 6.2v72h51.9c3.4 0 6.2 2.8 6.2 6.2v51.1c0 3.4-2.8 6.2-6.2 6.2H249.1V321c0 21.4 14.8 33.5 42.5 22.4c3-1.2 5.6-2 8-1.4c2.2 .5 3.6 2.1 4.6 4.9L318 387.1c1 3.2 2 6.7-.3 9.1c-8.5 9.1-31.2 19.8-60.9 19.8z\"],trello:[448,512,[],\"f181\",\"M392.3 32H56.1C25.1 32 0 57.1 0 88c-.1 0 0-4 0 336 0 30.9 25.1 56 56 56h336.2c30.8-.2 55.7-25.2 55.7-56V88c.1-30.8-24.8-55.8-55.6-56zM197 371.3c-.2 14.7-12.1 26.6-26.9 26.6H87.4c-14.8.1-26.9-11.8-27-26.6V117.1c0-14.8 12-26.9 26.9-26.9h82.9c14.8 0 26.9 12 26.9 26.9v254.2zm193.1-112c0 14.8-12 26.9-26.9 26.9h-81c-14.8 0-26.9-12-26.9-26.9V117.2c0-14.8 12-26.9 26.8-26.9h81.1c14.8 0 26.9 12 26.9 26.9v142.1z\"],\"creative-commons-nc-jp\":[496,512,[],\"f4ea\",\"M247.7 8C103.6 8 0 124.8 0 256c0 136.4 111.8 248 247.7 248C377.9 504 496 403.2 496 256 496 117.2 388.5 8 247.7 8zm.6 450.7c-112 0-203.6-92.5-203.6-202.7 0-21.1 3-41.2 9-60.3l127 56.5h-27.9v38.6h58.1l5.7 11.8v18.7h-63.8V360h63.8v56h61.7v-56h64.2v-35.7l81 36.1c-1.5 2.2-57.1 98.3-175.2 98.3zm87.6-137.3h-57.6v-18.7l2.9-5.6 54.7 24.3zm6.5-51.4v-17.8h-38.6l63-116H301l-43.4 96-23-10.2-39.6-85.7h-65.8l27.3 51-81.9-36.5c27.8-44.1 82.6-98.1 173.7-98.1 112.8 0 203 90 203 203.4 0 21-2.7 40.6-7.9 59l-101-45.1z\"],\"get-pocket\":[448,512,[],\"f265\",\"M407.6 64h-367C18.5 64 0 82.5 0 104.6v135.2C0 364.5 99.7 464 224.2 464c124 0 223.8-99.5 223.8-224.2V104.6c0-22.4-17.7-40.6-40.4-40.6zm-162 268.5c-12.4 11.8-31.4 11.1-42.4 0C89.5 223.6 88.3 227.4 88.3 209.3c0-16.9 13.8-30.7 30.7-30.7 17 0 16.1 3.8 105.2 89.3 90.6-86.9 88.6-89.3 105.5-89.3 16.9 0 30.7 13.8 30.7 30.7 0 17.8-2.9 15.7-114.8 123.2z\"],perbyte:[448,512,[],\"e083\",\"M305.314,284.578H246.6V383.3h58.711q24.423,0,38.193-13.77t13.77-36.11q0-21.826-14.032-35.335T305.314,284.578ZM149.435,128.7H90.724v98.723h58.711q24.42,0,38.19-13.773t13.77-36.107q0-21.826-14.029-35.338T149.435,128.7ZM366.647,32H81.353A81.445,81.445,0,0,0,0,113.352V398.647A81.445,81.445,0,0,0,81.353,480H366.647A81.445,81.445,0,0,0,448,398.647V113.352A81.445,81.445,0,0,0,366.647,32Zm63.635,366.647a63.706,63.706,0,0,1-63.635,63.635H81.353a63.706,63.706,0,0,1-63.635-63.635V113.352A63.706,63.706,0,0,1,81.353,49.718H366.647a63.706,63.706,0,0,1,63.635,63.634ZM305.314,128.7H246.6v98.723h58.711q24.423,0,38.193-13.773t13.77-36.107q0-21.826-14.032-35.338T305.314,128.7Z\"],grunt:[384,512,[],\"f3ad\",\"M61.3 189.3c-1.1 10 5.2 19.1 5.2 19.1.7-7.5 2.2-12.8 4-16.6.4 10.3 3.2 23.5 12.8 34.1 6.9 7.6 35.6 23.3 54.9 6.1 1 2.4 2.1 5.3 3 8.5 2.9 10.3-2.7 25.3-2.7 25.3s15.1-17.1 13.9-32.5c10.8-.5 21.4-8.4 21.1-19.5 0 0-18.9 10.4-35.5-8.8-9.7-11.2-40.9-42-83.1-31.8 4.3 1 8.9 2.4 13.5 4.1h-.1c-4.2 2-6.5 7.1-7 12zm28.3-1.8c19.5 11 37.4 25.7 44.9 37-5.7 3.3-21.7 10.4-38-1.7-10.3-7.6-9.8-26.2-6.9-35.3zm142.1 45.8c-1.2 15.5 13.9 32.5 13.9 32.5s-5.6-15-2.7-25.3c.9-3.2 2-6 3-8.5 19.3 17.3 48 1.5 54.8-6.1 9.6-10.6 12.3-23.8 12.8-34.1 1.8 3.8 3.4 9.1 4 16.6 0 0 6.4-9.1 5.2-19.1-.6-5-2.9-10-7-11.8h-.1c4.6-1.8 9.2-3.2 13.5-4.1-42.3-10.2-73.4 20.6-83.1 31.8-16.7 19.2-35.5 8.8-35.5 8.8-.2 10.9 10.4 18.9 21.2 19.3zm62.7-45.8c3 9.1 3.4 27.7-7 35.4-16.3 12.1-32.2 5-37.9 1.6 7.5-11.4 25.4-26 44.9-37zM160 418.5h-29.4c-5.5 0-8.2 1.6-9.5 2.9-1.9 2-2.2 4.7-.9 8.1 3.5 9.1 11.4 16.5 13.7 18.6 3.1 2.7 7.5 4.3 11.8 4.3 4.4 0 8.3-1.7 11-4.6 7.5-8.2 11.9-17.1 13-19.8.6-1.5 1.3-4.5-.9-6.8-1.8-1.8-4.7-2.7-8.8-2.7zm189.2-101.2c-2.4 17.9-13 33.8-24.6 43.7-3.1-22.7-3.7-55.5-3.7-62.4 0-14.7 9.5-24.5 12.2-26.1 2.5-1.5 5.4-3 8.3-4.6 18-9.6 40.4-21.6 40.4-43.7 0-16.2-9.3-23.2-15.4-27.8-.8-.6-1.5-1.1-2.2-1.7-2.1-1.7-3.7-3-4.3-4.4-4.4-9.8-3.6-34.2-1.7-37.6.6-.6 16.7-20.9 11.8-39.2-2-7.4-6.9-13.3-14.1-17-5.3-2.7-11.9-4.2-19.5-4.5-.1-2-.5-3.9-.9-5.9-.6-2.6-1.1-5.3-.9-8.1.4-4.7.8-9 2.2-11.3 8.4-13.3 28.8-17.6 29-17.6l12.3-2.4-8.1-9.5c-.1-.2-17.3-17.5-46.3-17.5-7.9 0-16 1.3-24.1 3.9-24.2 7.8-42.9 30.5-49.4 39.3-3.1-1-6.3-1.9-9.6-2.7-4.2-15.8 9-38.5 9-38.5s-13.6-3-33.7 15.2c-2.6-6.5-8.1-20.5-1.8-37.2C184.6 10.1 177.2 26 175 40.4c-7.6-5.4-6.7-23.1-7.2-27.6-7.5.9-29.2 21.9-28.2 48.3-2 .5-3.9 1.1-5.9 1.7-6.5-8.8-25.1-31.5-49.4-39.3-7.9-2.2-16-3.5-23.9-3.5-29 0-46.1 17.3-46.3 17.5L6 46.9l12.3 2.4c.2 0 20.6 4.3 29 17.6 1.4 2.2 1.8 6.6 2.2 11.3.2 2.8-.4 5.5-.9 8.1-.4 1.9-.8 3.9-.9 5.9-7.7.3-14.2 1.8-19.5 4.5-7.2 3.7-12.1 9.6-14.1 17-5 18.2 11.2 38.5 11.8 39.2 1.9 3.4 2.7 27.8-1.7 37.6-.6 1.4-2.2 2.7-4.3 4.4-.7.5-1.4 1.1-2.2 1.7-6.1 4.6-15.4 11.7-15.4 27.8 0 22.1 22.4 34.1 40.4 43.7 3 1.6 5.8 3.1 8.3 4.6 2.7 1.6 12.2 11.4 12.2 26.1 0 6.9-.6 39.7-3.7 62.4-11.6-9.9-22.2-25.9-24.6-43.8 0 0-29.2 22.6-20.6 70.8 5.2 29.5 23.2 46.1 47 54.7 8.8 19.1 29.4 45.7 67.3 49.6C143 504.3 163 512 192.2 512h.2c29.1 0 49.1-7.7 63.6-19.5 37.9-3.9 58.5-30.5 67.3-49.6 23.8-8.7 41.7-25.2 47-54.7 8.2-48.4-21.1-70.9-21.1-70.9zM305.7 37.7c5.6-1.8 11.6-2.7 17.7-2.7 11 0 19.9 3 24.7 5-3.1 1.4-6.4 3.2-9.7 5.3-2.4-.4-5.6-.8-9.2-.8-10.5 0-20.5 3.1-28.7 8.9-12.3 8.7-18 16.9-20.7 22.4-2.2-1.3-4.5-2.5-7.1-3.7-1.6-.8-3.1-1.5-4.7-2.2 6.1-9.1 19.9-26.5 37.7-32.2zm21 18.2c-.8 1-1.6 2.1-2.3 3.2-3.3 5.2-3.9 11.6-4.4 17.8-.5 6.4-1.1 12.5-4.4 17-4.2.8-8.1 1.7-11.5 2.7-2.3-3.1-5.6-7-10.5-11.2 1.4-4.8 5.5-16.1 13.5-22.5 5.6-4.3 12.2-6.7 19.6-7zM45.6 45.3c-3.3-2.2-6.6-4-9.7-5.3 4.8-2 13.7-5 24.7-5 6.1 0 12 .9 17.7 2.7 17.8 5.8 31.6 23.2 37.7 32.1-1.6.7-3.2 1.4-4.8 2.2-2.5 1.2-4.9 2.5-7.1 3.7-2.6-5.4-8.3-13.7-20.7-22.4-8.3-5.8-18.2-8.9-28.8-8.9-3.4.1-6.6.5-9 .9zm44.7 40.1c-4.9 4.2-8.3 8-10.5 11.2-3.4-.9-7.3-1.9-11.5-2.7C65 89.5 64.5 83.4 64 77c-.5-6.2-1.1-12.6-4.4-17.8-.7-1.1-1.5-2.2-2.3-3.2 7.4.3 14 2.6 19.5 7 8 6.3 12.1 17.6 13.5 22.4zM58.1 259.9c-2.7-1.6-5.6-3.1-8.4-4.6-14.9-8-30.2-16.3-30.2-30.5 0-11.1 4.3-14.6 8.9-18.2l.5-.4c.7-.6 1.4-1.2 2.2-1.8-.9 7.2-1.9 13.3-2.7 14.9 0 0 12.1-15 15.7-44.3 1.4-11.5-1.1-34.3-5.1-43 .2 4.9 0 9.8-.3 14.4-.4-.8-.8-1.6-1.3-2.2-3.2-4-11.8-17.5-9.4-26.6.9-3.5 3.1-6 6.7-7.8 3.8-1.9 8.8-2.9 15.1-2.9 12.3 0 25.9 3.7 32.9 6 25.1 8 55.4 30.9 64.1 37.7.2.2.4.3.4.3l5.6 3.9-3.5-5.8c-.2-.3-19.1-31.4-53.2-46.5 2-2.9 7.4-8.1 21.6-15.1 21.4-10.5 46.5-15.8 74.3-15.8 27.9 0 52.9 5.3 74.3 15.8 14.2 6.9 19.6 12.2 21.6 15.1-34 15.1-52.9 46.2-53.1 46.5l-3.5 5.8 5.6-3.9s.2-.1.4-.3c8.7-6.8 39-29.8 64.1-37.7 7-2.2 20.6-6 32.9-6 6.3 0 11.3 1 15.1 2.9 3.5 1.8 5.7 4.4 6.7 7.8 2.5 9.1-6.1 22.6-9.4 26.6-.5.6-.9 1.3-1.3 2.2-.3-4.6-.5-9.5-.3-14.4-4 8.8-6.5 31.5-5.1 43 3.6 29.3 15.7 44.3 15.7 44.3-.8-1.6-1.8-7.7-2.7-14.9.7.6 1.5 1.2 2.2 1.8l.5.4c4.6 3.7 8.9 7.1 8.9 18.2 0 14.2-15.4 22.5-30.2 30.5-2.9 1.5-5.7 3.1-8.4 4.6-8.7 5-18 16.7-19.1 34.2-.9 14.6.9 49.9 3.4 75.9-12.4 4.8-26.7 6.4-39.7 6.8-2-4.1-3.9-8.5-5.5-13.1-.7-2-19.6-51.1-26.4-62.2 5.5 39 17.5 73.7 23.5 89.6-3.5-.5-7.3-.7-11.7-.7h-117c-4.4 0-8.3.3-11.7.7 6-15.9 18.1-50.6 23.5-89.6-6.8 11.2-25.7 60.3-26.4 62.2-1.6 4.6-3.5 9-5.5 13.1-13-.4-27.2-2-39.7-6.8 2.5-26 4.3-61.2 3.4-75.9-.9-17.4-10.3-29.2-19-34.2zM34.8 404.6c-12.1-20-8.7-54.1-3.7-59.1 10.9 34.4 47.2 44.3 74.4 45.4-2.7 4.2-5.2 7.6-7 10l-1.4 1.4c-7.2 7.8-8.6 18.5-4.1 31.8-22.7-.1-46.3-9.8-58.2-29.5zm45.7 43.5c6 1.1 12.2 1.9 18.6 2.4 3.5 8 7.4 15.9 12.3 23.1-14.4-5.9-24.4-16-30.9-25.5zM192 498.2c-60.6-.1-78.3-45.8-84.9-64.7-3.7-10.5-3.4-18.2.9-23.1 2.9-3.3 9.5-7.2 24.6-7.2h118.8c15.1 0 21.8 3.9 24.6 7.2 4.2 4.8 4.5 12.6.9 23.1-6.6 18.8-24.3 64.6-84.9 64.7zm80.6-24.6c4.9-7.2 8.8-15.1 12.3-23.1 6.4-.5 12.6-1.3 18.6-2.4-6.5 9.5-16.5 19.6-30.9 25.5zm76.6-69c-12 19.7-35.6 29.3-58.1 29.7 4.5-13.3 3.1-24.1-4.1-31.8-.4-.5-.9-1-1.4-1.5-1.8-2.4-4.3-5.8-7-10 27.2-1.2 63.5-11 74.4-45.4 5 5 8.4 39.1-3.8 59zM191.9 187.7h.2c12.7-.1 27.2-17.8 27.2-17.8-9.9 6-18.8 8.1-27.3 8.3-8.5-.2-17.4-2.3-27.3-8.3 0 0 14.5 17.6 27.2 17.8zm61.7 230.7h-29.4c-4.2 0-7.2.9-8.9 2.7-2.2 2.3-1.5 5.2-.9 6.7 1 2.6 5.5 11.3 13 19.3 2.7 2.9 6.6 4.5 11 4.5s8.7-1.6 11.8-4.2c2.3-2 10.2-9.2 13.7-18.1 1.3-3.3 1-6-.9-7.9-1.3-1.3-4-2.9-9.4-3z\"],weebly:[512,512,[],\"f5cc\",\"M425.09 65.83c-39.88 0-73.28 25.73-83.66 64.33-18.16-58.06-65.5-64.33-84.95-64.33-19.78 0-66.8 6.28-85.28 64.33-10.38-38.6-43.45-64.33-83.66-64.33C38.59 65.83 0 99.72 0 143.03c0 28.96 4.18 33.27 77.17 233.48 22.37 60.57 67.77 69.35 92.74 69.35 39.23 0 70.04-19.46 85.93-53.98 15.89 34.83 46.69 54.29 85.93 54.29 24.97 0 70.36-9.1 92.74-69.67 76.55-208.65 77.5-205.58 77.5-227.2.63-48.32-36.01-83.47-86.92-83.47zm26.34 114.81l-65.57 176.44c-7.92 21.49-21.22 37.22-46.24 37.22-23.44 0-37.38-12.41-44.03-33.9l-39.28-117.42h-.95L216.08 360.4c-6.96 21.5-20.9 33.6-44.02 33.6-25.02 0-38.33-15.74-46.24-37.22L60.88 181.55c-5.38-14.83-7.92-23.91-7.92-34.5 0-16.34 15.84-29.36 38.33-29.36 18.69 0 31.99 11.8 36.11 29.05l44.03 139.82h.95l44.66-136.79c6.02-19.67 16.47-32.08 38.96-32.08s32.94 12.11 38.96 32.08l44.66 136.79h.95l44.03-139.82c4.12-17.25 17.42-29.05 36.11-29.05 22.17 0 38.33 13.32 38.33 35.71-.32 7.87-4.12 16.04-7.61 27.24z\"],connectdevelop:[576,512,[],\"f20e\",\"M550.5 241l-50.089-86.786c1.071-2.142 1.875-4.553 1.875-7.232 0-8.036-6.696-14.733-14.732-15.001l-55.447-95.893c.536-1.607 1.071-3.214 1.071-4.821 0-8.571-6.964-15.268-15.268-15.268-4.821 0-8.839 2.143-11.786 5.625H299.518C296.839 18.143 292.821 16 288 16s-8.839 2.143-11.518 5.625H170.411C167.464 18.143 163.447 16 158.625 16c-8.303 0-15.268 6.696-15.268 15.268 0 1.607.536 3.482 1.072 4.821l-55.983 97.233c-5.356 2.41-9.107 7.5-9.107 13.661 0 .535.268 1.071.268 1.607l-53.304 92.143c-7.232 1.339-12.59 7.5-12.59 15 0 7.232 5.089 13.393 12.054 15l55.179 95.358c-.536 1.607-.804 2.946-.804 4.821 0 7.232 5.089 13.393 12.054 14.732l51.697 89.732c-.536 1.607-1.071 3.482-1.071 5.357 0 8.571 6.964 15.268 15.268 15.268 4.821 0 8.839-2.143 11.518-5.357h106.875C279.161 493.857 283.447 496 288 496s8.839-2.143 11.518-5.357h107.143c2.678 2.946 6.696 4.821 10.982 4.821 8.571 0 15.268-6.964 15.268-15.268 0-1.607-.267-2.946-.803-4.285l51.697-90.268c6.964-1.339 12.054-7.5 12.054-14.732 0-1.607-.268-3.214-.804-4.821l54.911-95.358c6.964-1.339 12.322-7.5 12.322-15-.002-7.232-5.092-13.393-11.788-14.732zM153.535 450.732l-43.66-75.803h43.66v75.803zm0-83.839h-43.66c-.268-1.071-.804-2.142-1.339-3.214l44.999-47.41v50.624zm0-62.411l-50.357 53.304c-1.339-.536-2.679-1.34-4.018-1.607L43.447 259.75c.535-1.339.535-2.679.535-4.018s0-2.41-.268-3.482l51.965-90c2.679-.268 5.357-1.072 7.768-2.679l50.089 51.965v92.946zm0-102.322l-45.803-47.41c1.339-2.143 2.143-4.821 2.143-7.767 0-.268-.268-.804-.268-1.072l43.928-15.804v72.053zm0-80.625l-43.66 15.804 43.66-75.536v59.732zm326.519 39.108l.804 1.339L445.5 329.125l-63.75-67.232 98.036-101.518.268.268zM291.75 355.107l11.518 11.786H280.5l11.25-11.786zm-.268-11.25l-83.303-85.446 79.553-84.375 83.036 87.589-79.286 82.232zm5.357 5.893l79.286-82.232 67.5 71.25-5.892 28.125H313.714l-16.875-17.143zM410.411 44.393c1.071.536 2.142 1.072 3.482 1.34l57.857 100.714v.536c0 2.946.803 5.624 2.143 7.767L376.393 256l-83.035-87.589L410.411 44.393zm-9.107-2.143L287.732 162.518l-57.054-60.268 166.339-60h4.287zm-123.483 0c2.678 2.678 6.16 4.285 10.179 4.285s7.5-1.607 10.179-4.285h75L224.786 95.821 173.893 42.25h103.928zm-116.249 5.625l1.071-2.142a33.834 33.834 0 0 0 2.679-.804l51.161 53.84-54.911 19.821V47.875zm0 79.286l60.803-21.964 59.732 63.214-79.553 84.107-40.982-42.053v-83.304zm0 92.678L198 257.607l-36.428 38.304v-76.072zm0 87.858l42.053-44.464 82.768 85.982-17.143 17.678H161.572v-59.196zm6.964 162.053c-1.607-1.607-3.482-2.678-5.893-3.482l-1.071-1.607v-89.732h99.91l-91.607 94.821h-1.339zm129.911 0c-2.679-2.41-6.428-4.285-10.447-4.285s-7.767 1.875-10.447 4.285h-96.429l91.607-94.821h38.304l91.607 94.821H298.447zm120-11.786l-4.286 7.5c-1.339.268-2.41.803-3.482 1.339l-89.196-91.875h114.376l-17.412 83.036zm12.856-22.232l12.858-60.803h21.964l-34.822 60.803zm34.822-68.839h-20.357l4.553-21.16 17.143 18.214c-.535.803-1.071 1.874-1.339 2.946zm66.161-107.411l-55.447 96.697c-1.339.535-2.679 1.071-4.018 1.874l-20.625-21.964 34.554-163.928 45.803 79.286c-.267 1.339-.803 2.678-.803 4.285 0 1.339.268 2.411.536 3.75z\"],leanpub:[576,512,[],\"f212\",\"M386.539 111.485l15.096 248.955-10.979-.275c-36.232-.824-71.64 8.783-102.657 27.997-31.016-19.214-66.424-27.997-102.657-27.997-45.564 0-82.07 10.705-123.516 27.723L93.117 129.6c28.546-11.803 61.484-18.115 92.226-18.115 41.173 0 73.836 13.175 102.657 42.544 27.723-28.271 59.013-41.721 98.539-42.544zM569.07 448c-25.526 0-47.485-5.215-70.542-15.645-34.31-15.645-69.993-24.978-107.871-24.978-38.977 0-74.934 12.901-102.657 40.623-27.723-27.723-63.68-40.623-102.657-40.623-37.878 0-73.561 9.333-107.871 24.978C55.239 442.236 32.731 448 8.303 448H6.93L49.475 98.859C88.726 76.626 136.486 64 181.775 64 218.83 64 256.984 71.685 288 93.095 319.016 71.685 357.17 64 394.225 64c45.289 0 93.049 12.626 132.3 34.859L569.07 448zm-43.368-44.741l-34.036-280.246c-30.742-13.999-67.248-21.41-101.009-21.41-38.428 0-74.385 12.077-102.657 38.702-28.272-26.625-64.228-38.702-102.657-38.702-33.761 0-70.267 7.411-101.009 21.41L50.298 403.259c47.211-19.487 82.894-33.486 135.045-33.486 37.604 0 70.817 9.606 102.657 29.644 31.84-20.038 65.052-29.644 102.657-29.644 52.151 0 87.834 13.999 135.045 33.486z\"],\"black-tie\":[448,512,[],\"f27e\",\"M0 32v448h448V32H0zm316.5 325.2L224 445.9l-92.5-88.7 64.5-184-64.5-86.6h184.9L252 173.2l64.5 184z\"],themeco:[448,512,[],\"f5c6\",\"M202.9 8.43c9.9-5.73 26-5.82 35.95-.21L430 115.85c10 5.6 18 19.44 18 30.86V364c0 11.44-8.06 25.29-18 31L238.81 503.74c-9.93 5.66-26 5.57-35.85-.21L17.86 395.12C8 389.34 0 375.38 0 364V146.71c0-11.44 8-25.36 17.91-31.08zm-77.4 199.83c-15.94 0-31.89.14-47.83.14v101.45H96.8V280h28.7c49.71 0 49.56-71.74 0-71.74zm140.14 100.29l-30.73-34.64c37-7.51 34.8-65.23-10.87-65.51-16.09 0-32.17-.14-48.26-.14v101.59h19.13v-33.91h18.41l29.56 33.91h22.76zm-41.59-82.32c23.34 0 23.26 32.46 0 32.46h-29.13v-32.46zm-95.56-1.6c21.18 0 21.11 38.85 0 38.85H96.18v-38.84zm192.65-18.25c-68.46 0-71 105.8 0 105.8 69.48-.01 69.41-105.8 0-105.8zm0 17.39c44.12 0 44.8 70.86 0 70.86s-44.43-70.86 0-70.86z\"],python:[448,512,[],\"f3e2\",\"M439.8 200.5c-7.7-30.9-22.3-54.2-53.4-54.2h-40.1v47.4c0 36.8-31.2 67.8-66.8 67.8H172.7c-29.2 0-53.4 25-53.4 54.3v101.8c0 29 25.2 46 53.4 54.3 33.8 9.9 66.3 11.7 106.8 0 26.9-7.8 53.4-23.5 53.4-54.3v-40.7H226.2v-13.6h160.2c31.1 0 42.6-21.7 53.4-54.2 11.2-33.5 10.7-65.7 0-108.6zM286.2 404c11.1 0 20.1 9.1 20.1 20.3 0 11.3-9 20.4-20.1 20.4-11 0-20.1-9.2-20.1-20.4.1-11.3 9.1-20.3 20.1-20.3zM167.8 248.1h106.8c29.7 0 53.4-24.5 53.4-54.3V91.9c0-29-24.4-50.7-53.4-55.6-35.8-5.9-74.7-5.6-106.8.1-45.2 8-53.4 24.7-53.4 55.6v40.7h106.9v13.6h-147c-31.1 0-58.3 18.7-66.8 54.2-9.8 40.7-10.2 66.1 0 108.6 7.6 31.6 25.7 54.2 56.8 54.2H101v-48.8c0-35.3 30.5-66.4 66.8-66.4zm-6.7-142.6c-11.1 0-20.1-9.1-20.1-20.3.1-11.3 9-20.4 20.1-20.4 11 0 20.1 9.2 20.1 20.4s-9 20.3-20.1 20.3z\"],android:[576,512,[],\"f17b\",\"M420.55,301.93a24,24,0,1,1,24-24,24,24,0,0,1-24,24m-265.1,0a24,24,0,1,1,24-24,24,24,0,0,1-24,24m273.7-144.48,47.94-83a10,10,0,1,0-17.27-10h0l-48.54,84.07a301.25,301.25,0,0,0-246.56,0L116.18,64.45a10,10,0,1,0-17.27,10h0l47.94,83C64.53,202.22,8.24,285.55,0,384H576c-8.24-98.45-64.54-181.78-146.85-226.55\"],bots:[640,512,[],\"e340\",\"M86.344,197.834a51.767,51.767,0,0,0-41.57,20.058V156.018a8.19,8.19,0,0,0-8.19-8.19H8.19A8.19,8.19,0,0,0,0,156.018V333.551a8.189,8.189,0,0,0,8.19,8.189H36.584a8.189,8.189,0,0,0,8.19-8.189v-8.088c11.628,13.373,25.874,19.769,41.573,19.769,34.6,0,61.922-26.164,61.922-73.843C148.266,225.452,121.229,197.834,86.344,197.834ZM71.516,305.691c-9.593,0-21.221-4.942-26.745-12.5V250.164c5.528-7.558,17.152-12.791,26.745-12.791,17.734,0,31.107,13.082,31.107,34.013C102.623,292.609,89.25,305.691,71.516,305.691Zm156.372-59.032a17.4,17.4,0,1,0,17.4,17.4A17.4,17.4,0,0,0,227.888,246.659ZM273.956,156.7V112.039a13.308,13.308,0,1,0-10.237,0V156.7a107.49,107.49,0,1,0,10.237,0Zm85.993,107.367c0,30.531-40.792,55.281-91.112,55.281s-91.111-24.75-91.111-55.281,40.792-55.281,91.111-55.281S359.949,233.532,359.949,264.062Zm-50.163,17.4a17.4,17.4,0,1,0-17.4-17.4h0A17.4,17.4,0,0,0,309.786,281.466ZM580.7,250.455c-14.828-2.617-22.387-3.78-22.387-9.885,0-5.523,7.268-9.884,17.735-9.884a65.56,65.56,0,0,1,34.484,10.1,8.171,8.171,0,0,0,11.288-2.468c.07-.11.138-.221.2-.333l8.611-14.886a8.2,8.2,0,0,0-2.867-11.123,99.863,99.863,0,0,0-52.014-14.138c-38.956,0-60.179,21.514-60.179,46.225,0,36.342,33.725,41.864,57.563,45.642,13.373,2.326,24.13,4.361,24.13,11.048,0,6.4-5.523,10.757-18.9,10.757-13.552,0-30.994-6.222-42.623-13.579a8.206,8.206,0,0,0-11.335,2.491c-.035.054-.069.108-.1.164l-10.2,16.891a8.222,8.222,0,0,0,2.491,11.066c15.224,10.3,37.663,16.692,59.441,16.692,40.409,0,63.957-19.769,63.957-46.515C640,260.63,604.537,254.816,580.7,250.455Zm-95.928,60.787a8.211,8.211,0,0,0-9.521-5.938,23.168,23.168,0,0,1-4.155.387c-7.849,0-12.5-6.106-12.5-14.245V240.28h20.349a8.143,8.143,0,0,0,8.141-8.143V209.466a8.143,8.143,0,0,0-8.141-8.143H458.594V171.091a8.143,8.143,0,0,0-8.143-8.143H422.257a8.143,8.143,0,0,0-8.143,8.143h0v30.232H399a8.143,8.143,0,0,0-8.143,8.143h0v22.671A8.143,8.143,0,0,0,399,240.28h15.115v63.667c0,27.037,15.408,41.282,43.9,41.282,12.183,0,21.383-2.2,27.6-5.446a8.161,8.161,0,0,0,4.145-9.278Z\"],\"free-code-camp\":[576,512,[],\"f2c5\",\"M97.22,96.21c10.36-10.65,16-17.12,16-21.9,0-2.76-1.92-5.51-3.83-7.42A14.81,14.81,0,0,0,101,64.05c-8.48,0-20.92,8.79-35.84,25.69C23.68,137,2.51,182.81,3.37,250.34s17.47,117,54.06,161.87C76.22,435.86,90.62,448,100.9,448a13.55,13.55,0,0,0,8.37-3.84c1.91-2.76,3.81-5.63,3.81-8.38,0-5.63-3.86-12.2-13.2-20.55-44.45-42.33-67.32-97-67.48-165C32.25,188.8,54,137.83,97.22,96.21ZM239.47,420.07c.58.37.91.55.91.55Zm93.79.55.17-.13C333.24,420.62,333.17,420.67,333.26,420.62Zm3.13-158.18c-16.24-4.15,50.41-82.89-68.05-177.17,0,0,15.54,49.38-62.83,159.57-74.27,104.35,23.46,168.73,34,175.23-6.73-4.35-47.4-35.7,9.55-128.64,11-18.3,25.53-34.87,43.5-72.16,0,0,15.91,22.45,7.6,71.13C287.7,364,354,342.91,355,343.94c22.75,26.78-17.72,73.51-21.58,76.55,5.49-3.65,117.71-78,33-188.1C360.43,238.4,352.62,266.59,336.39,262.44ZM510.88,89.69C496,72.79,483.52,64,475,64a14.81,14.81,0,0,0-8.39,2.84c-1.91,1.91-3.83,4.66-3.83,7.42,0,4.78,5.6,11.26,16,21.9,43.23,41.61,65,92.59,64.82,154.06-.16,68-23,122.63-67.48,165-9.34,8.35-13.18,14.92-13.2,20.55,0,2.75,1.9,5.62,3.81,8.38A13.61,13.61,0,0,0,475.1,448c10.28,0,24.68-12.13,43.47-35.79,36.59-44.85,53.14-94.38,54.06-161.87S552.32,137,510.88,89.69Z\"],hornbill:[512,512,[],\"f592\",\"M76.38 370.3a37.8 37.8 0 1 1-32.07-32.42c-78.28-111.35 52-190.53 52-190.53-5.86 43-8.24 91.16-8.24 91.16-67.31 41.49.93 64.06 39.81 72.87a140.38 140.38 0 0 0 131.66 91.94c1.92 0 3.77-.21 5.67-.28l.11 18.86c-99.22 1.39-158.7-29.14-188.94-51.6zm108-327.7A37.57 37.57 0 0 0 181 21.45a37.95 37.95 0 1 0-31.17 54.22c-22.55 29.91-53.83 89.57-52.42 190l21.84-.15c0-.9-.14-1.77-.14-2.68A140.42 140.42 0 0 1 207 132.71c8-37.71 30.7-114.3 73.8-44.29 0 0 48.14 2.38 91.18 8.24 0 0-77.84-128-187.59-54.06zm304.19 134.17a37.94 37.94 0 1 0-53.84-28.7C403 126.13 344.89 99 251.28 100.33l.14 22.5c2.7-.15 5.39-.41 8.14-.41a140.37 140.37 0 0 1 130.49 88.76c39.1 9 105.06 31.58 38.46 72.54 0 0-2.34 48.13-8.21 91.16 0 0 133.45-81.16 49-194.61a37.45 37.45 0 0 0 19.31-3.5zM374.06 436.24c21.43-32.46 46.42-89.69 45.14-179.66l-19.52.14c.08 2.06.3 4.07.3 6.15a140.34 140.34 0 0 1-91.39 131.45c-8.85 38.95-31.44 106.66-72.77 39.49 0 0-48.12-2.34-91.19-8.22 0 0 79.92 131.34 191.9 51a37.5 37.5 0 0 0 3.64 14 37.93 37.93 0 1 0 33.89-54.29z\"],js:[448,512,[],\"f3b8\",\"M0 32v448h448V32H0zm243.8 349.4c0 43.6-25.6 63.5-62.9 63.5-33.7 0-53.2-17.4-63.2-38.5l34.3-20.7c6.6 11.7 12.6 21.6 27.1 21.6 13.8 0 22.6-5.4 22.6-26.5V237.7h42.1v143.7zm99.6 63.5c-39.1 0-64.4-18.6-76.7-43l34.3-19.8c9 14.7 20.8 25.6 41.5 25.6 17.4 0 28.6-8.7 28.6-20.8 0-14.4-11.4-19.5-30.7-28l-10.5-4.5c-30.4-12.9-50.5-29.2-50.5-63.5 0-31.6 24.1-55.6 61.6-55.6 26.8 0 46 9.3 59.8 33.7L368 290c-7.2-12.9-15-18-27.1-18-12.3 0-20.1 7.8-20.1 18 0 12.6 7.8 17.7 25.9 25.6l10.5 4.5c35.8 15.3 55.9 31 55.9 66.2 0 37.8-29.8 58.6-69.7 58.6z\"],ideal:[576,512,[],\"e013\",\"M125.61,165.48a49.07,49.07,0,1,0,49.06,49.06A49.08,49.08,0,0,0,125.61,165.48ZM86.15,425.84h78.94V285.32H86.15Zm151.46-211.6c0-20-10-22.53-18.74-22.53H204.82V237.5h14.05C228.62,237.5,237.61,234.69,237.61,214.24Zm201.69,46V168.93h22.75V237.5h33.69C486.5,113.08,388.61,86.19,299.67,86.19H204.84V169h14c25.6,0,41.5,17.35,41.5,45.26,0,28.81-15.52,46-41.5,46h-14V425.88h94.83c144.61,0,194.94-67.16,196.72-165.64Zm-109.75,0H273.3V169h54.43v22.73H296v10.58h30V225H296V237.5h33.51Zm74.66,0-5.16-17.67H369.31l-5.18,17.67H340.47L368,168.92h32.35l27.53,91.34ZM299.65,32H32V480H299.65c161.85,0,251-79.73,251-224.52C550.62,172,518,32,299.65,32Zm0,426.92H53.07V53.07H299.65c142.1,0,229.9,64.61,229.9,202.41C529.55,389.57,448.55,458.92,299.65,458.92Zm83.86-264.85L376,219.88H392.4l-7.52-25.81Z\"],git:[512,512,[],\"f1d3\",\"M216.29 158.39H137C97 147.9 6.51 150.63 6.51 233.18c0 30.09 15 51.23 35 61-25.1 23-37 33.85-37 49.21 0 11 4.47 21.14 17.89 26.81C8.13 383.61 0 393.35 0 411.65c0 32.11 28.05 50.82 101.63 50.82 70.75 0 111.79-26.42 111.79-73.18 0-58.66-45.16-56.5-151.63-63l13.43-21.55c27.27 7.58 118.7 10 118.7-67.89 0-18.7-7.73-31.71-15-41.07l37.41-2.84zm-63.42 241.9c0 32.06-104.89 32.1-104.89 2.43 0-8.14 5.27-15 10.57-21.54 77.71 5.3 94.32 3.37 94.32 19.11zm-50.81-134.58c-52.8 0-50.46-71.16 1.2-71.16 49.54 0 50.82 71.16-1.2 71.16zm133.3 100.51v-32.1c26.75-3.66 27.24-2 27.24-11V203.61c0-8.5-2.05-7.38-27.24-16.26l4.47-32.92H324v168.71c0 6.51.4 7.32 6.51 8.14l20.73 2.84v32.1zm52.45-244.31c-23.17 0-36.59-13.43-36.59-36.61s13.42-35.77 36.59-35.77c23.58 0 37 12.62 37 35.77s-13.42 36.61-37 36.61zM512 350.46c-17.49 8.53-43.1 16.26-66.28 16.26-48.38 0-66.67-19.5-66.67-65.46V194.75c0-5.42 1.05-4.06-31.71-4.06V154.5c35.78-4.07 50-22 54.47-66.27h38.63c0 65.83-1.34 61.81 3.26 61.81H501v40.65h-60.56v97.15c0 6.92-4.92 51.41 60.57 26.84z\"],dev:[448,512,[],\"f6cc\",\"M120.12 208.29c-3.88-2.9-7.77-4.35-11.65-4.35H91.03v104.47h17.45c3.88 0 7.77-1.45 11.65-4.35 3.88-2.9 5.82-7.25 5.82-13.06v-69.65c-.01-5.8-1.96-10.16-5.83-13.06zM404.1 32H43.9C19.7 32 .06 51.59 0 75.8v360.4C.06 460.41 19.7 480 43.9 480h360.2c24.21 0 43.84-19.59 43.9-43.8V75.8c-.06-24.21-19.7-43.8-43.9-43.8zM154.2 291.19c0 18.81-11.61 47.31-48.36 47.25h-46.4V172.98h47.38c35.44 0 47.36 28.46 47.37 47.28l.01 70.93zm100.68-88.66H201.6v38.42h32.57v29.57H201.6v38.41h53.29v29.57h-62.18c-11.16.29-20.44-8.53-20.72-19.69V193.7c-.27-11.15 8.56-20.41 19.71-20.69h63.19l-.01 29.52zm103.64 115.29c-13.2 30.75-36.85 24.63-47.44 0l-38.53-144.8h32.57l29.71 113.72 29.57-113.72h32.58l-38.46 144.8z\"],sketch:[512,512,[],\"f7c6\",\"M27.5 162.2L9 187.1h90.5l6.9-130.7-78.9 105.8zM396.3 45.7L267.7 32l135.7 147.2-7.1-133.5zM112.2 218.3l-11.2-22H9.9L234.8 458zm2-31.2h284l-81.5-88.5L256.3 33zm297.3 9.1L277.6 458l224.8-261.7h-90.9zM415.4 69L406 56.4l.9 17.3 6.1 113.4h90.3zM113.5 93.5l-4.6 85.6L244.7 32 116.1 45.7zm287.7 102.7h-290l42.4 82.9L256.3 480l144.9-283.8z\"],\"yandex-international\":[320,512,[],\"f414\",\"M129.5 512V345.9L18.5 48h55.8l81.8 229.7L250.2 0h51.3L180.8 347.8V512h-51.3z\"],\"cc-amex\":[576,512,[],\"f1f3\",\"M0 432c0 26.5 21.5 48 48 48H528c26.5 0 48-21.5 48-48v-1.1H514.3l-31.9-35.1-31.9 35.1H246.8V267.1H181L262.7 82.4h78.6l28.1 63.2V82.4h97.2L483.5 130l17-47.6H576V80c0-26.5-21.5-48-48-48H48C21.5 32 0 53.5 0 80V432zm440.4-21.7L482.6 364l42 46.3H576l-68-72.1 68-72.1H525.4l-42 46.7-41.5-46.7H390.5L458 338.6l-67.4 71.6V377.1h-83V354.9h80.9V322.6H307.6V300.2h83V267.1h-122V410.3H440.4zm96.3-72L576 380.2V296.9l-39.3 41.4zm-36.3-92l36.9-100.6V246.3H576V103H515.8l-32.2 89.3L451.7 103H390.5V246.1L327.3 103H276.1L213.7 246.3h43l11.9-28.7h65.9l12 28.7h82.7V146L466 246.3h34.4zM282 185.4l19.5-46.9 19.4 46.9H282z\"],uber:[448,512,[],\"f402\",\"M414.1 32H33.9C15.2 32 0 47.2 0 65.9V446c0 18.8 15.2 34 33.9 34H414c18.7 0 33.9-15.2 33.9-33.9V65.9C448 47.2 432.8 32 414.1 32zM237.6 391.1C163 398.6 96.4 344.2 88.9 269.6h94.4V290c0 3.7 3 6.8 6.8 6.8H258c3.7 0 6.8-3 6.8-6.8v-67.9c0-3.7-3-6.8-6.8-6.8h-67.9c-3.7 0-6.8 3-6.8 6.8v20.4H88.9c7-69.4 65.4-122.2 135.1-122.2 69.7 0 128.1 52.8 135.1 122.2 7.5 74.5-46.9 141.1-121.5 148.6z\"],github:[496,512,[],\"f09b\",\"M165.9 397.4c0 2-2.3 3.6-5.2 3.6-3.3.3-5.6-1.3-5.6-3.6 0-2 2.3-3.6 5.2-3.6 3-.3 5.6 1.3 5.6 3.6zm-31.1-4.5c-.7 2 1.3 4.3 4.3 4.9 2.6 1 5.6 0 6.2-2s-1.3-4.3-4.3-5.2c-2.6-.7-5.5.3-6.2 2.3zm44.2-1.7c-2.9.7-4.9 2.6-4.6 4.9.3 2 2.9 3.3 5.9 2.6 2.9-.7 4.9-2.6 4.6-4.6-.3-1.9-3-3.2-5.9-2.9zM244.8 8C106.1 8 0 113.3 0 252c0 110.9 69.8 205.8 169.5 239.2 12.8 2.3 17.3-5.6 17.3-12.1 0-6.2-.3-40.4-.3-61.4 0 0-70 15-84.7-29.8 0 0-11.4-29.1-27.8-36.6 0 0-22.9-15.7 1.6-15.4 0 0 24.9 2 38.6 25.8 21.9 38.6 58.6 27.5 72.9 20.9 2.3-16 8.8-27.1 16-33.7-55.9-6.2-112.3-14.3-112.3-110.5 0-27.5 7.6-41.3 23.6-58.9-2.6-6.5-11.1-33.3 2.6-67.9 20.9-6.5 69 27 69 27 20-5.6 41.5-8.5 62.8-8.5s42.8 2.9 62.8 8.5c0 0 48.1-33.6 69-27 13.7 34.7 5.2 61.4 2.6 67.9 16 17.7 25.8 31.5 25.8 58.9 0 96.5-58.9 104.2-114.8 110.5 9.2 7.9 17 22.9 17 46.4 0 33.7-.3 75.4-.3 83.6 0 6.5 4.6 14.4 17.3 12.1C428.2 457.8 496 362.9 496 252 496 113.3 383.5 8 244.8 8zM97.2 352.9c-1.3 1-1 3.3.7 5.2 1.6 1.6 3.9 2.3 5.2 1 1.3-1 1-3.3-.7-5.2-1.6-1.6-3.9-2.3-5.2-1zm-10.8-8.1c-.7 1.3.3 2.9 2.3 3.9 1.6 1 3.6.7 4.3-.7.7-1.3-.3-2.9-2.3-3.9-2-.6-3.6-.3-4.3.7zm32.4 35.6c-1.6 1.3-1 4.3 1.3 6.2 2.3 2.3 5.2 2.6 6.5 1 1.3-1.3.7-4.3-1.3-6.2-2.2-2.3-5.2-2.6-6.5-1zm-11.4-14.7c-1.6 1-1.6 3.6 0 5.9 1.6 2.3 4.3 3.3 5.6 2.3 1.6-1.3 1.6-3.9 0-6.2-1.4-2.3-4-3.3-5.6-2z\"],php:[640,512,[],\"f457\",\"M320 104.5c171.4 0 303.2 72.2 303.2 151.5S491.3 407.5 320 407.5c-171.4 0-303.2-72.2-303.2-151.5S148.7 104.5 320 104.5m0-16.8C143.3 87.7 0 163 0 256s143.3 168.3 320 168.3S640 349 640 256 496.7 87.7 320 87.7zM218.2 242.5c-7.9 40.5-35.8 36.3-70.1 36.3l13.7-70.6c38 0 63.8-4.1 56.4 34.3zM97.4 350.3h36.7l8.7-44.8c41.1 0 66.6 3 90.2-19.1 26.1-24 32.9-66.7 14.3-88.1-9.7-11.2-25.3-16.7-46.5-16.7h-70.7L97.4 350.3zm185.7-213.6h36.5l-8.7 44.8c31.5 0 60.7-2.3 74.8 10.7 14.8 13.6 7.7 31-8.3 113.1h-37c15.4-79.4 18.3-86 12.7-92-5.4-5.8-17.7-4.6-47.4-4.6l-18.8 96.6h-36.5l32.7-168.6zM505 242.5c-8 41.1-36.7 36.3-70.1 36.3l13.7-70.6c38.2 0 63.8-4.1 56.4 34.3zM384.2 350.3H421l8.7-44.8c43.2 0 67.1 2.5 90.2-19.1 26.1-24 32.9-66.7 14.3-88.1-9.7-11.2-25.3-16.7-46.5-16.7H417l-32.8 168.7z\"],alipay:[448,512,[],\"f642\",\"M377.74 32H70.26C31.41 32 0 63.41 0 102.26v307.48C0 448.59 31.41 480 70.26 480h307.48c38.52 0 69.76-31.08 70.26-69.6-45.96-25.62-110.59-60.34-171.6-88.44-32.07 43.97-84.14 81-148.62 81-70.59 0-93.73-45.3-97.04-76.37-3.97-39.01 14.88-81.5 99.52-81.5 35.38 0 79.35 10.25 127.13 24.96 16.53-30.09 26.45-60.34 26.45-60.34h-178.2v-16.7h92.08v-31.24H88.28v-19.01h109.44V92.34h50.92v50.42h109.44v19.01H248.63v31.24h88.77s-15.21 46.62-38.35 90.92c48.93 16.7 100.01 36.04 148.62 52.74V102.26C447.83 63.57 416.43 32 377.74 32zM47.28 322.95c.99 20.17 10.25 53.73 69.93 53.73 52.07 0 92.58-39.68 117.87-72.9-44.63-18.68-84.48-31.41-109.44-31.41-67.45 0-79.35 33.06-78.36 50.58z\"],youtube:[576,512,[61802],\"f167\",\"M549.655 124.083c-6.281-23.65-24.787-42.276-48.284-48.597C458.781 64 288 64 288 64S117.22 64 74.629 75.486c-23.497 6.322-42.003 24.947-48.284 48.597-11.412 42.867-11.412 132.305-11.412 132.305s0 89.438 11.412 132.305c6.281 23.65 24.787 41.5 48.284 47.821C117.22 448 288 448 288 448s170.78 0 213.371-11.486c23.497-6.321 42.003-24.171 48.284-47.821 11.412-42.867 11.412-132.305 11.412-132.305s0-89.438-11.412-132.305zm-317.51 213.508V175.185l142.739 81.205-142.739 81.201z\"],skyatlas:[640,512,[],\"f216\",\"M640 329.3c0 65.9-52.5 114.4-117.5 114.4-165.9 0-196.6-249.7-359.7-249.7-146.9 0-147.1 212.2 5.6 212.2 42.5 0 90.9-17.8 125.3-42.5 5.6-4.1 16.9-16.3 22.8-16.3s10.9 5 10.9 10.9c0 7.8-13.1 19.1-18.7 24.1-40.9 35.6-100.3 61.2-154.7 61.2-83.4.1-154-59-154-144.9s67.5-149.1 152.8-149.1c185.3 0 222.5 245.9 361.9 245.9 99.9 0 94.8-139.7 3.4-139.7-17.5 0-35 11.6-46.9 11.6-8.4 0-15.9-7.2-15.9-15.6 0-11.6 5.3-23.7 5.3-36.3 0-66.6-50.9-114.7-116.9-114.7-53.1 0-80 36.9-88.8 36.9-6.2 0-11.2-5-11.2-11.2 0-5.6 4.1-10.3 7.8-14.4 25.3-28.8 64.7-43.7 102.8-43.7 79.4 0 139.1 58.4 139.1 137.8 0 6.9-.3 13.7-1.2 20.6 11.9-3.1 24.1-4.7 35.9-4.7 60.7 0 111.9 45.3 111.9 107.2z\"],\"firefox-browser\":[512,512,[],\"e007\",\"M130.22 127.548C130.38 127.558 130.3 127.558 130.22 127.548V127.548ZM481.64 172.898C471.03 147.398 449.56 119.898 432.7 111.168C446.42 138.058 454.37 165.048 457.4 185.168C457.405 185.306 457.422 185.443 457.45 185.578C429.87 116.828 383.098 89.1089 344.9 28.7479C329.908 5.05792 333.976 3.51792 331.82 4.08792L331.7 4.15792C284.99 30.1109 256.365 82.5289 249.12 126.898C232.503 127.771 216.219 131.895 201.19 139.035C199.838 139.649 198.736 140.706 198.066 142.031C197.396 143.356 197.199 144.87 197.506 146.323C197.7 147.162 198.068 147.951 198.586 148.639C199.103 149.327 199.76 149.899 200.512 150.318C201.264 150.737 202.096 150.993 202.954 151.071C203.811 151.148 204.676 151.045 205.491 150.768L206.011 150.558C221.511 143.255 238.408 139.393 255.541 139.238C318.369 138.669 352.698 183.262 363.161 201.528C350.161 192.378 326.811 183.338 304.341 187.248C392.081 231.108 368.541 381.784 246.951 376.448C187.487 373.838 149.881 325.467 146.421 285.648C146.421 285.648 157.671 243.698 227.041 243.698C234.541 243.698 255.971 222.778 256.371 216.698C256.281 214.698 213.836 197.822 197.281 181.518C188.434 172.805 184.229 168.611 180.511 165.458C178.499 163.75 176.392 162.158 174.201 160.688C168.638 141.231 168.399 120.638 173.51 101.058C148.45 112.468 128.96 130.508 114.8 146.428H114.68C105.01 134.178 105.68 93.7779 106.25 85.3479C106.13 84.8179 99.022 89.0159 98.1 89.6579C89.5342 95.7103 81.5528 102.55 74.26 110.088C57.969 126.688 30.128 160.242 18.76 211.318C14.224 231.701 12 255.739 12 263.618C12 398.318 121.21 507.508 255.92 507.508C376.56 507.508 478.939 420.281 496.35 304.888C507.922 228.192 481.64 173.82 481.64 172.898Z\"],replyd:[448,512,[],\"f3e6\",\"M320 480H128C57.6 480 0 422.4 0 352V160C0 89.6 57.6 32 128 32h192c70.4 0 128 57.6 128 128v192c0 70.4-57.6 128-128 128zM193.4 273.2c-6.1-2-11.6-3.1-16.4-3.1-7.2 0-13.5 1.9-18.9 5.6-5.4 3.7-9.6 9-12.8 15.8h-1.1l-4.2-18.3h-28v138.9h36.1v-89.7c1.5-5.4 4.4-9.8 8.7-13.2 4.3-3.4 9.8-5.1 16.2-5.1 4.6 0 9.8 1 15.6 3.1l4.8-34zm115.2 103.4c-3.2 2.4-7.7 4.8-13.7 7.1-6 2.3-12.8 3.5-20.4 3.5-12.2 0-21.1-3-26.5-8.9-5.5-5.9-8.5-14.7-9-26.4h83.3c.9-4.8 1.6-9.4 2.1-13.9.5-4.4.7-8.6.7-12.5 0-10.7-1.6-19.7-4.7-26.9-3.2-7.2-7.3-13-12.5-17.2-5.2-4.3-11.1-7.3-17.8-9.2-6.7-1.8-13.5-2.8-20.6-2.8-21.1 0-37.5 6.1-49.2 18.3s-17.5 30.5-17.5 55c0 22.8 5.2 40.7 15.6 53.7 10.4 13.1 26.8 19.6 49.2 19.6 10.7 0 20.9-1.5 30.4-4.6 9.5-3.1 17.1-6.8 22.6-11.2l-12-23.6zm-21.8-70.3c3.8 5.4 5.3 13.1 4.6 23.1h-51.7c.9-9.4 3.7-17 8.2-22.6 4.5-5.6 11.5-8.5 21-8.5 8.2-.1 14.1 2.6 17.9 8zm79.9 2.5c4.1 3.9 9.4 5.8 16.1 5.8 7 0 12.6-1.9 16.7-5.8s6.1-9.1 6.1-15.6-2-11.6-6.1-15.4c-4.1-3.8-9.6-5.7-16.7-5.7-6.7 0-12 1.9-16.1 5.7-4.1 3.8-6.1 8.9-6.1 15.4s2 11.7 6.1 15.6zm0 100.5c4.1 3.9 9.4 5.8 16.1 5.8 7 0 12.6-1.9 16.7-5.8s6.1-9.1 6.1-15.6-2-11.6-6.1-15.4c-4.1-3.8-9.6-5.7-16.7-5.7-6.7 0-12 1.9-16.1 5.7-4.1 3.8-6.1 8.9-6.1 15.4 0 6.6 2 11.7 6.1 15.6z\"],suse:[640,512,[],\"f7d6\",\"M593.1 192.6A10.4 10.4 0 1 1 604.5 210a10.4 10.4 0 1 1 -11.4-17.4zm-47.1 12.2a38.5 38.5 0 1 1 75-17.6 38.5 38.5 0 1 1 -75 17.6zM433.7 336.7c3.2 4.6 5.8 9 7.3 13.4c1 3.1 2.4 7.3 5.5 8.9c.2 .1 .3 .2 .5 .2c5.7 2.1 20.3 1.7 20.3 1.7h26.8c2.3 0 22.4 0 21.9-2.3c-2.4-10.8-14.9-12.7-24.4-18.3c-8.7-5.2-17-11.1-20.8-21.3c-2-5.2-.8-17.4 2.6-21.8c2.5-3.2 6.1-5.3 10-6.2c4.3-.9 8.8-.1 13.1 .3c5.3 .5 10.6 1.5 15.9 2.2c10.3 1.3 20.6 1.9 31 1.6c17.1-.5 34.2-3.2 50.4-8.7c11.3-3.8 22.4-8.9 32-16.1c10.9-8.1 8.1-7.4-3-6.2c-13.3 1.4-26.6 1.6-39.9 .8c-12.4-.7-24.7-2.2-35.9-7.9c-8.8-4.6-16.4-9.1-23.4-16.2c-1-1.1-1.7-4.2 .2-6.2c1.9-1.9 5.8-.8 7 .2c12.2 10.2 30.5 18.6 49.3 19.5c10.2 .5 20.1 .7 30.4 .3c5.1-.2 12.8-.2 17.9-.3c2.6 0 9.8 .7 11.2-2.1c.4-.8 .4-1.8 .3-2.7c-1.5-40.9-4.5-86.9-47.3-106.5c-31.9-14.6-79.7-37.2-99.9-46.6c-4.7-2.2-10.2 1.3-10.2 6.5c0 13.6 .7 33.3 .7 51.1c-9.7-9.9-26-16.1-38.4-21.8c-14.1-6.5-28.7-12-43.5-16.6c-29.8-9.2-60.7-14.9-91.7-18c-35.2-3.5-71-1.8-105.7 5.3C147 115.1 90.8 142.6 48.2 182.7C22.1 207.3 1.6 242.4 .2 277.9c-2 50.3 12.1 77.3 38 105.2c41.3 44.4 130.2 50.6 166.2-2c16.2-23.7 19.7-55.8 8-82c-11.8-26.2-38.8-45.1-67.4-46c-22.2-.7-45.9 10.6-54.5 31.1c-6.5 15.7-2.8 35.1 9 47.3c4.6 4.8 10.9 8.7 17.7 7.1c4-.9 7.4-3.9 8-8c.9-6-4.4-9.9-7.6-14.5c-5.8-8.3-4.7-20.9 2.7-27.9c6.2-6 15.3-7.8 23.9-7.7c8 0 16.2 1.4 23.1 5.5c9.7 5.7 16.2 16.2 18.4 27.2c6.7 33-20.2 59.9-56.6 62c-18.6 1.1-37.6-3.8-52.1-15.5C40.1 329.9 31.1 269.4 73.2 237c40-30.7 90.4-22.8 120.2-6.8c23.8 12.8 41.5 33.6 55 56.7c6.7 11.6 12.5 23.7 17.8 36.1c5.1 11.8 9.9 23.8 20.2 32.5c6.8 5.8 15.2 5.6 24.1 5.6h50.8c6.9 0 5.2-4.6 2.2-7.7c-6.7-6.9-16.4-8.4-25.4-10.9c-20.5-5.6-18.4-32.8-12.7-32.8c18.3 0 18.9 .6 34.9 .3c23.2-.3 30.2-1.7 48.3 5c9.7 3.6 19 13.1 25.1 21.7z\"],jenkins:[512,512,[],\"f3b6\",\"M487.1 425c-1.4-11.2-19-23.1-28.2-31.9-5.1-5-29-23.1-30.4-29.9-1.4-6.6 9.7-21.5 13.3-28.9 5.1-10.7 8.8-23.7 11.3-32.6 18.8-66.1 20.7-156.9-6.2-211.2-10.2-20.6-38.6-49-56.4-62.5-42-31.7-119.6-35.3-170.1-16.6-14.1 5.2-27.8 9.8-40.1 17.1-33.1 19.4-68.3 32.5-78.1 71.6-24.2 10.8-31.5 41.8-30.3 77.8.2 7 4.1 15.8 2.7 22.4-.7 3.3-5.2 7.6-6.1 9.8-11.6 27.7-2.3 64 11.1 83.7 8.1 11.9 21.5 22.4 39.2 25.2.7 10.6 3.3 19.7 8.2 30.4 3.1 6.8 14.7 19 10.4 27.7-2.2 4.4-21 13.8-27.3 17.6C89 407.2 73.7 415 54.2 429c-12.6 9-32.3 10.2-29.2 31.1 2.1 14.1 10.1 31.6 14.7 45.8.7 2 1.4 4.1 2.1 6h422c4.9-15.3 9.7-30.9 14.6-47.2 3.4-11.4 10.2-27.8 8.7-39.7zM205.9 33.7c1.8-.5 3.4.7 4.9 2.4-.2 5.2-5.4 5.1-8.9 6.8-5.4 6.7-13.4 9.8-20 17.2-6.8 7.5-14.4 27.7-23.4 30-4.5 1.1-9.7-.8-13.6-.5-10.4.7-17.7 6-28.3 7.5 13.6-29.9 56.1-54 89.3-63.4zm-104.8 93.6c13.5-14.9 32.1-24.1 54.8-25.9 11.7 29.7-8.4 65-.9 97.6 2.3 9.9 10.2 25.4-2.4 25.7.3-28.3-34.8-46.3-61.3-29.6-1.8-21.5-4.9-51.7 9.8-67.8zm36.7 200.2c-1-4.1-2.7-12.9-2.3-15.1 1.6-8.7 17.1-12.5 11-24.7-11.3-.1-13.8 10.2-24.1 11.3-26.7 2.6-45.6-35.4-44.4-58.4 1-19.5 17.6-38.2 40.1-35.8 16 1.8 21.4 19.2 24.5 34.7 9.2.5 22.5-.4 26.9-7.6-.6-17.5-8.8-31.6-8.2-47.7 1-30.3 17.5-57.6 4.8-87.4 13.6-30.9 53.5-55.3 83.1-70 36.6-18.3 94.9-3.7 129.3 15.8 19.7 11.1 34.4 32.7 48.3 50.7-19.5-5.8-36.1 4.2-33.1 20.3 16.3-14.9 44.2-.2 52.5 16.4 7.9 15.8 7.8 39.3 9 62.8 2.9 57-10.4 115.9-39.1 157.1-7.7 11-14.1 23-24.9 30.6-26 18.2-65.4 34.7-99.2 23.4-44.7-15-65-44.8-89.5-78.8.7 18.7 13.8 34.1 26.8 48.4 11.3 12.5 25 26.6 39.7 32.4-12.3-2.9-31.1-3.8-36.2 7.2-28.6-1.9-55.1-4.8-68.7-24.2-10.6-15.4-21.4-41.4-26.3-61.4zm222 124.1c4.1-3 11.1-2.9 17.4-3.6-5.4-2.7-13-3.7-19.3-2.2-.1-4.2-2-6.8-3.2-10.2 10.6-3.8 35.5-28.5 49.6-20.3 6.7 3.9 9.5 26.2 10.1 37 .4 9-.8 18-4.5 22.8-18.8-.6-35.8-2.8-50.7-7 .9-6.1-1-12.1.6-16.5zm-17.2-20c-16.8.8-26-1.2-38.3-10.8.2-.8 1.4-.5 1.5-1.4 18 8 40.8-3.3 59-4.9-7.9 5.1-14.6 11.6-22.2 17.1zm-12.1 33.2c-1.6-9.4-3.5-12-2.8-20.2 25-16.6 29.7 28.6 2.8 20.2zM226 438.6c-11.6-.7-48.1-14-38.5-23.7 9.4 6.5 27.5 4.9 41.3 7.3.8 4.4-2.8 10.2-2.8 16.4zM57.7 497.1c-4.3-12.7-9.2-25.1-14.8-36.9 30.8-23.8 65.3-48.9 102.2-63.5 2.8-1.1 23.2 25.4 26.2 27.6 16.5 11.7 37 21 56.2 30.2 1.2 8.8 3.9 20.2 8.7 35.5.7 2.3 1.4 4.7 2.2 7.2H57.7zm240.6 5.7h-.8c.3-.2.5-.4.8-.5v.5zm7.5-5.7c2.1-1.4 4.3-2.8 6.4-4.3 1.1 1.4 2.2 2.8 3.2 4.3h-9.6zm15.1-24.7c-10.8 7.3-20.6 18.3-33.3 25.2-6 3.3-27 11.7-33.4 10.2-3.6-.8-3.9-5.3-5.4-9.5-3.1-9-10.1-23.4-10.8-37-.8-17.2-2.5-46 16-42.4 14.9 2.9 32.3 9.7 43.9 16.1 7.1 3.9 11.1 8.6 21.9 9.5-.1 1.4-.1 2.8-.2 4.3-5.9 3.9-15.3 3.8-21.8 7.1 9.5.4 17 2.7 23.5 5.9-.1 3.4-.3 7-.4 10.6zm53.4 24.7h-14c-.1-3.2-2.8-5.8-6.1-5.8s-5.9 2.6-6.1 5.8h-17.4c-2.8-4.4-5.7-8.6-8.9-12.5 2.1-2.2 4-4.7 6-6.9 9 3.7 14.8-4.9 21.7-4.2 7.9.8 14.2 11.7 25.4 11l-.6 12.6zm8.7 0c.2-4 .4-7.8.6-11.5 15.6-7.3 29 1.3 35.7 11.5H383zm83.4-37c-2.3 11.2-5.8 24-9.9 37.1-.2-.1-.4-.1-.6-.1H428c.6-1.1 1.2-2.2 1.9-3.3-2.6-6.1-9-8.7-10.9-15.5 12.1-22.7 6.5-93.4-24.2-78.5 4.3-6.3 15.6-11.5 20.8-19.3 13 10.4 20.8 20.3 33.2 31.4 6.8 6 20 13.3 21.4 23.1.8 5.5-2.6 18.9-3.8 25.1zM222.2 130.5c5.4-14.9 27.2-34.7 45-32 7.7 1.2 18 8.2 12.2 17.7-30.2-7-45.2 12.6-54.4 33.1-8.1-2-4.9-13.1-2.8-18.8zm184.1 63.1c8.2-3.6 22.4-.7 29.6-5.3-4.2-11.5-10.3-21.4-9.3-37.7.5 0 1 0 1.4.1 6.8 14.2 12.7 29.2 21.4 41.7-5.7 13.5-43.6 25.4-43.1 1.2zm20.4-43zm-117.2 45.7c-6.8-10.9-19-32.5-14.5-45.3 6.5 11.9 8.6 24.4 17.8 33.3 4.1 4 12.2 9 8.2 20.2-.9 2.7-7.8 8.6-11.7 9.7-14.4 4.3-47.9.9-36.6-17.1 11.9.7 27.9 7.8 36.8-.8zm27.3 70c3.8 6.6 1.4 18.7 12.1 20.6 20.2 3.4 43.6-12.3 58.1-17.8 9-15.2-.8-20.7-8.9-30.5-16.6-20-38.8-44.8-38-74.7 6.7-4.9 7.3 7.4 8.2 9.7 8.7 20.3 30.4 46.2 46.3 63.5 3.9 4.3 10.3 8.4 11 11.2 2.1 8.2-5.4 18-4.5 23.5-21.7 13.9-45.8 29.1-81.4 25.6-7.4-6.7-10.3-21.4-2.9-31.1zm-201.3-9.2c-6.8-3.9-8.4-21-16.4-21.4-11.4-.7-9.3 22.2-9.3 35.5-7.8-7.1-9.2-29.1-3.5-40.3-6.6-3.2-9.5 3.6-13.1 5.9 4.7-34.1 49.8-15.8 42.3 20.3zm299.6 28.8c-10.1 19.2-24.4 40.4-54 41-.6-6.2-1.1-15.6 0-19.4 22.7-2.2 36.6-13.7 54-21.6zm-141.9 12.4c18.9 9.9 53.6 11 79.3 10.2 1.4 5.6 1.3 12.6 1.4 19.4-33 1.8-72-6.4-80.7-29.6zm92.2 46.7c-1.7 4.3-5.3 9.3-9.8 11.1-12.1 4.9-45.6 8.7-62.4-.3-10.7-5.7-17.5-18.5-23.4-26-2.8-3.6-16.9-12.9-.2-12.9 13.1 32.7 58 29 95.8 28.1z\"],twitter:[512,512,[],\"f099\",\"M459.37 151.716c.325 4.548.325 9.097.325 13.645 0 138.72-105.583 298.558-298.558 298.558-59.452 0-114.68-17.219-161.137-47.106 8.447.974 16.568 1.299 25.34 1.299 49.055 0 94.213-16.568 130.274-44.832-46.132-.975-84.792-31.188-98.112-72.772 6.498.974 12.995 1.624 19.818 1.624 9.421 0 18.843-1.3 27.614-3.573-48.081-9.747-84.143-51.98-84.143-102.985v-1.299c13.969 7.797 30.214 12.67 47.431 13.319-28.264-18.843-46.781-51.005-46.781-87.391 0-19.492 5.197-37.36 14.294-52.954 51.655 63.675 129.3 105.258 216.365 109.807-1.624-7.797-2.599-15.918-2.599-24.04 0-57.828 46.782-104.934 104.934-104.934 30.213 0 57.502 12.67 76.67 33.137 23.715-4.548 46.456-13.32 66.599-25.34-7.798 24.366-24.366 44.833-46.132 57.827 21.117-2.273 41.584-8.122 60.426-16.243-14.292 20.791-32.161 39.308-52.628 54.253z\"],rockrms:[496,512,[],\"f3e9\",\"M248 8C111 8 0 119 0 256s111 248 248 248 248-111 248-248S385 8 248 8zm157.4 419.5h-90l-112-131.3c-17.9-20.4-3.9-56.1 26.6-56.1h75.3l-84.6-99.3-84.3 98.9h-90L193.5 67.2c14.4-18.4 41.3-17.3 54.5 0l157.7 185.1c19 22.8 2 57.2-27.6 56.1-.6 0-74.2.2-74.2.2l101.5 118.9z\"],pinterest:[496,512,[],\"f0d2\",\"M496 256c0 137-111 248-248 248-25.6 0-50.2-3.9-73.4-11.1 10.1-16.5 25.2-43.5 30.8-65 3-11.6 15.4-59 15.4-59 8.1 15.4 31.7 28.5 56.8 28.5 74.8 0 128.7-68.8 128.7-154.3 0-81.9-66.9-143.2-152.9-143.2-107 0-163.9 71.8-163.9 150.1 0 36.4 19.4 81.7 50.3 96.1 4.7 2.2 7.2 1.2 8.3-3.3.8-3.4 5-20.3 6.9-28.1.6-2.5.3-4.7-1.7-7.1-10.1-12.5-18.3-35.3-18.3-56.6 0-54.7 41.4-107.6 112-107.6 60.9 0 103.6 41.5 103.6 100.9 0 67.1-33.9 113.6-78 113.6-24.3 0-42.6-20.1-36.7-44.8 7-29.5 20.5-61.3 20.5-82.6 0-19-10.2-34.9-31.4-34.9-24.9 0-44.9 25.7-44.9 60.2 0 22 7.4 36.8 7.4 36.8s-24.5 103.8-29 123.2c-5 21.4-3 51.6-.9 71.2C65.4 450.9 0 361.1 0 256 0 119 111 8 248 8s248 111 248 248z\"],buffer:[448,512,[],\"f837\",\"M427.84 380.67l-196.5 97.82a18.6 18.6 0 0 1-14.67 0L20.16 380.67c-4-2-4-5.28 0-7.29L67.22 350a18.65 18.65 0 0 1 14.69 0l134.76 67a18.51 18.51 0 0 0 14.67 0l134.76-67a18.62 18.62 0 0 1 14.68 0l47.06 23.43c4.05 1.96 4.05 5.24 0 7.24zm0-136.53l-47.06-23.43a18.62 18.62 0 0 0-14.68 0l-134.76 67.08a18.68 18.68 0 0 1-14.67 0L81.91 220.71a18.65 18.65 0 0 0-14.69 0l-47.06 23.43c-4 2-4 5.29 0 7.31l196.51 97.8a18.6 18.6 0 0 0 14.67 0l196.5-97.8c4.05-2.02 4.05-5.3 0-7.31zM20.16 130.42l196.5 90.29a20.08 20.08 0 0 0 14.67 0l196.51-90.29c4-1.86 4-4.89 0-6.74L231.33 33.4a19.88 19.88 0 0 0-14.67 0l-196.5 90.28c-4.05 1.85-4.05 4.88 0 6.74z\"],npm:[576,512,[],\"f3d4\",\"M288 288h-32v-64h32v64zm288-128v192H288v32H160v-32H0V160h576zm-416 32H32v128h64v-96h32v96h32V192zm160 0H192v160h64v-32h64V192zm224 0H352v128h64v-96h32v96h32v-96h32v96h32V192z\"],yammer:[512,512,[],\"f840\",\"M500.676,159.486a12.779,12.779,0,0,0-6.4-8.282,13.954,13.954,0,0,0-10.078-1.125L457.8,156.7l-.043-.2-22.3,5.785-1.243.333-.608-2.17A369.037,369.037,0,0,0,347.538,4.289a14.1,14.1,0,0,0-19.784-.463l-102.9,102.747H24.947A24.9,24.9,0,0,0,0,131.417V380.38a24.963,24.963,0,0,0,24.918,24.9H224.986L328.072,508a13.667,13.667,0,0,0,19.327,0c.126-.126.249-.255.37-.385a368.025,368.025,0,0,0,69.577-107.374,403.45,403.45,0,0,0,17.3-50.8v-.028l20.406,5.336.029-.073L483.345,362a20.253,20.253,0,0,0,2.619.5,13.359,13.359,0,0,0,4.139-.072,13.5,13.5,0,0,0,10.515-9.924,415.855,415.855,0,0,0,.058-193.013ZM337.125,24.65l.013.014h-.013Zm-110.2,165.161L174.311,281.1a11.338,11.338,0,0,0-1.489,5.655v46.189a22.04,22.04,0,0,1-22.041,22h-3.4A22.068,22.068,0,0,1,125.3,332.962V287.294a11.532,11.532,0,0,0-1.388-5.51l-51.6-92.2a21.988,21.988,0,0,1,19.264-32.726h3.268a22.059,22.059,0,0,1,19.611,11.916l36.357,70.281,37.515-70.512a22.066,22.066,0,0,1,38.556-.695,21.7,21.7,0,0,1,0,21.967ZM337.145,24.673a348.147,348.147,0,0,1,75.8,141.335l.564,1.952-114.134,29.6V131.417a25.006,25.006,0,0,0-24.947-24.9H255.067Zm60.5,367.305v-.043l-.014.014a347.19,347.19,0,0,1-60.177,95.227l-82.2-81.893h19.177a24.978,24.978,0,0,0,24.947-24.9v-66.2l114.6,29.862A385.191,385.191,0,0,1,397.648,391.978Zm84-52.45.015.014-50.618-13.131L299.379,292.1V219.572l119.746-30.99,4.468-1.157,39.54-10.253,18.511-4.816A393,393,0,0,1,481.644,339.528Z\"],btc:[384,512,[],\"f15a\",\"M310.204 242.638c27.73-14.18 45.377-39.39 41.28-81.3-5.358-57.351-52.458-76.573-114.85-81.929V0h-48.528v77.203c-12.605 0-25.525.315-38.444.63V0h-48.528v79.409c-17.842.539-38.622.276-97.37 0v51.678c38.314-.678 58.417-3.14 63.023 21.427v217.429c-2.925 19.492-18.524 16.685-53.255 16.071L3.765 443.68c88.481 0 97.37.315 97.37.315V512h48.528v-67.06c13.234.315 26.154.315 38.444.315V512h48.528v-68.005c81.299-4.412 135.647-24.894 142.895-101.467 5.671-61.446-23.32-88.862-69.326-99.89zM150.608 134.553c27.415 0 113.126-8.507 113.126 48.528 0 54.515-85.71 48.212-113.126 48.212v-96.74zm0 251.776V279.821c32.772 0 133.127-9.138 133.127 53.255-.001 60.186-100.355 53.253-133.127 53.253z\"],dribbble:[512,512,[],\"f17d\",\"M256 8C119.252 8 8 119.252 8 256s111.252 248 248 248 248-111.252 248-248S392.748 8 256 8zm163.97 114.366c29.503 36.046 47.369 81.957 47.835 131.955-6.984-1.477-77.018-15.682-147.502-6.818-5.752-14.041-11.181-26.393-18.617-41.614 78.321-31.977 113.818-77.482 118.284-83.523zM396.421 97.87c-3.81 5.427-35.697 48.286-111.021 76.519-34.712-63.776-73.185-116.168-79.04-124.008 67.176-16.193 137.966 1.27 190.061 47.489zm-230.48-33.25c5.585 7.659 43.438 60.116 78.537 122.509-99.087 26.313-186.36 25.934-195.834 25.809C62.38 147.205 106.678 92.573 165.941 64.62zM44.17 256.323c0-2.166.043-4.322.108-6.473 9.268.19 111.92 1.513 217.706-30.146 6.064 11.868 11.857 23.915 17.174 35.949-76.599 21.575-146.194 83.527-180.531 142.306C64.794 360.405 44.17 310.73 44.17 256.323zm81.807 167.113c22.127-45.233 82.178-103.622 167.579-132.756 29.74 77.283 42.039 142.053 45.189 160.638-68.112 29.013-150.015 21.053-212.768-27.882zm248.38 8.489c-2.171-12.886-13.446-74.897-41.152-151.033 66.38-10.626 124.7 6.768 131.947 9.055-9.442 58.941-43.273 109.844-90.795 141.978z\"],\"stumbleupon-circle\":[496,512,[],\"f1a3\",\"M256 8C119 8 8 119 8 256s111 248 248 248 248-111 248-248S393 8 256 8zm0 177.5c-9.8 0-17.8 8-17.8 17.8v106.9c0 40.9-33.9 73.9-74.9 73.9-41.4 0-74.9-33.5-74.9-74.9v-46.5h57.3v45.8c0 10 8 17.8 17.8 17.8s17.8-7.9 17.8-17.8V200.1c0-40 34.2-72.1 74.7-72.1 40.7 0 74.7 32.3 74.7 72.6v23.7l-34.1 10.1-22.9-10.7v-20.6c.1-9.6-7.9-17.6-17.7-17.6zm167.6 123.6c0 41.4-33.5 74.9-74.9 74.9-41.2 0-74.9-33.2-74.9-74.2V263l22.9 10.7 34.1-10.1v47.1c0 9.8 8 17.6 17.8 17.6s17.8-7.9 17.8-17.6v-48h57.3c-.1 45.9-.1 46.4-.1 46.4z\"],\"internet-explorer\":[512,512,[],\"f26b\",\"M483.049 159.706c10.855-24.575 21.424-60.438 21.424-87.871 0-72.722-79.641-98.371-209.673-38.577-107.632-7.181-211.221 73.67-237.098 186.457 30.852-34.862 78.271-82.298 121.977-101.158C125.404 166.85 79.128 228.002 43.992 291.725 23.246 329.651 0 390.94 0 436.747c0 98.575 92.854 86.5 180.251 42.006 31.423 15.43 66.559 15.573 101.695 15.573 97.124 0 184.249-54.294 216.814-146.022H377.927c-52.509 88.593-196.819 52.996-196.819-47.436H509.9c6.407-43.581-1.655-95.715-26.851-141.162zM64.559 346.877c17.711 51.15 53.703 95.871 100.266 123.304-88.741 48.94-173.267 29.096-100.266-123.304zm115.977-108.873c2-55.151 50.276-94.871 103.98-94.871 53.418 0 101.981 39.72 103.981 94.871H180.536zm184.536-187.6c21.425-10.287 48.563-22.003 72.558-22.003 31.422 0 54.274 21.717 54.274 53.722 0 20.003-7.427 49.007-14.569 67.867-26.28-42.292-65.986-81.584-112.263-99.586z\"],stubber:[448,512,[],\"e5c7\",\"M136.5 294.2l58.8 22.9c9.1-36.8 25.4-61.1 55-61.1c49.4 0 71.4 63.6 142.4 63.6c15.6 0 35.9-2.8 55.3-13.3V368c0 61.8-50.4 112-112.3 112H0l41.8-56L0 368l41.7-56L0 256.1l41.8-56L0 144.1 41.8 88 0 32H335.7C397.6 32 448 82.3 448 144.1v51.3c-9.2 36.3-25.9 60.6-55 60.6c-49.6 0-71.6-63.5-142.4-63.5c-35.9 0-95.2 14.6-114.1 101.6h0z\"],telegram:[496,512,[62462,\"telegram-plane\"],\"f2c6\",\"M248,8C111.033,8,0,119.033,0,256S111.033,504,248,504,496,392.967,496,256,384.967,8,248,8ZM362.952,176.66c-3.732,39.215-19.881,134.378-28.1,178.3-3.476,18.584-10.322,24.816-16.948,25.425-14.4,1.326-25.338-9.517-39.287-18.661-21.827-14.308-34.158-23.215-55.346-37.177-24.485-16.135-8.612-25,5.342-39.5,3.652-3.793,67.107-61.51,68.335-66.746.153-.655.3-3.1-1.154-4.384s-3.59-.849-5.135-.5q-3.283.746-104.608,69.142-14.845,10.194-26.894,9.934c-8.855-.191-25.888-5.006-38.551-9.123-15.531-5.048-27.875-7.717-26.8-16.291q.84-6.7,18.45-13.7,108.446-47.248,144.628-62.3c68.872-28.647,83.183-33.623,92.511-33.789,2.052-.034,6.639.474,9.61,2.885a10.452,10.452,0,0,1,3.53,6.716A43.765,43.765,0,0,1,362.952,176.66Z\"],\"old-republic\":[496,512,[],\"f510\",\"M235.76 10.23c7.5-.31 15-.28 22.5-.09 3.61.14 7.2.4 10.79.73 4.92.27 9.79 1.03 14.67 1.62 2.93.43 5.83.98 8.75 1.46 7.9 1.33 15.67 3.28 23.39 5.4 12.24 3.47 24.19 7.92 35.76 13.21 26.56 12.24 50.94 29.21 71.63 49.88 20.03 20.09 36.72 43.55 48.89 69.19 1.13 2.59 2.44 5.1 3.47 7.74 2.81 6.43 5.39 12.97 7.58 19.63 4.14 12.33 7.34 24.99 9.42 37.83.57 3.14 1.04 6.3 1.4 9.47.55 3.83.94 7.69 1.18 11.56.83 8.34.84 16.73.77 25.1-.07 4.97-.26 9.94-.75 14.89-.24 3.38-.51 6.76-.98 10.12-.39 2.72-.63 5.46-1.11 8.17-.9 5.15-1.7 10.31-2.87 15.41-4.1 18.5-10.3 36.55-18.51 53.63-15.77 32.83-38.83 62.17-67.12 85.12a246.503 246.503 0 0 1-56.91 34.86c-6.21 2.68-12.46 5.25-18.87 7.41-3.51 1.16-7.01 2.38-10.57 3.39-6.62 1.88-13.29 3.64-20.04 5-4.66.91-9.34 1.73-14.03 2.48-5.25.66-10.5 1.44-15.79 1.74-6.69.66-13.41.84-20.12.81-6.82.03-13.65-.12-20.45-.79-3.29-.23-6.57-.5-9.83-.95-2.72-.39-5.46-.63-8.17-1.11-4.12-.72-8.25-1.37-12.35-2.22-4.25-.94-8.49-1.89-12.69-3.02-8.63-2.17-17.08-5.01-25.41-8.13-10.49-4.12-20.79-8.75-30.64-14.25-2.14-1.15-4.28-2.29-6.35-3.57-11.22-6.58-21.86-14.1-31.92-22.34-34.68-28.41-61.41-66.43-76.35-108.7-3.09-8.74-5.71-17.65-7.8-26.68-1.48-6.16-2.52-12.42-3.58-18.66-.4-2.35-.61-4.73-.95-7.09-.6-3.96-.75-7.96-1.17-11.94-.8-9.47-.71-18.99-.51-28.49.14-3.51.34-7.01.7-10.51.31-3.17.46-6.37.92-9.52.41-2.81.65-5.65 1.16-8.44.7-3.94 1.3-7.9 2.12-11.82 3.43-16.52 8.47-32.73 15.26-48.18 1.15-2.92 2.59-5.72 3.86-8.59 8.05-16.71 17.9-32.56 29.49-47.06 20-25.38 45.1-46.68 73.27-62.47 7.5-4.15 15.16-8.05 23.07-11.37 15.82-6.88 32.41-11.95 49.31-15.38 3.51-.67 7.04-1.24 10.56-1.85 2.62-.47 5.28-.7 7.91-1.08 3.53-.53 7.1-.68 10.65-1.04 2.46-.24 4.91-.36 7.36-.51m8.64 24.41c-9.23.1-18.43.99-27.57 2.23-7.3 1.08-14.53 2.6-21.71 4.3-13.91 3.5-27.48 8.34-40.46 14.42-10.46 4.99-20.59 10.7-30.18 17.22-4.18 2.92-8.4 5.8-12.34 9.03-5.08 3.97-9.98 8.17-14.68 12.59-2.51 2.24-4.81 4.7-7.22 7.06-28.22 28.79-48.44 65.39-57.5 104.69-2.04 8.44-3.54 17.02-4.44 25.65-1.1 8.89-1.44 17.85-1.41 26.8.11 7.14.38 14.28 1.22 21.37.62 7.12 1.87 14.16 3.2 21.18 1.07 4.65 2.03 9.32 3.33 13.91 6.29 23.38 16.5 45.7 30.07 65.75 8.64 12.98 18.78 24.93 29.98 35.77 16.28 15.82 35.05 29.04 55.34 39.22 7.28 3.52 14.66 6.87 22.27 9.63 5.04 1.76 10.06 3.57 15.22 4.98 11.26 3.23 22.77 5.6 34.39 7.06 2.91.29 5.81.61 8.72.9 13.82 1.08 27.74 1 41.54-.43 4.45-.6 8.92-.99 13.35-1.78 3.63-.67 7.28-1.25 10.87-2.1 4.13-.98 8.28-1.91 12.36-3.07 26.5-7.34 51.58-19.71 73.58-36.2 15.78-11.82 29.96-25.76 42.12-41.28 3.26-4.02 6.17-8.31 9.13-12.55 3.39-5.06 6.58-10.25 9.6-15.54 2.4-4.44 4.74-8.91 6.95-13.45 5.69-12.05 10.28-24.62 13.75-37.49 2.59-10.01 4.75-20.16 5.9-30.45 1.77-13.47 1.94-27.1 1.29-40.65-.29-3.89-.67-7.77-1-11.66-2.23-19.08-6.79-37.91-13.82-55.8-5.95-15.13-13.53-29.63-22.61-43.13-12.69-18.8-28.24-35.68-45.97-49.83-25.05-20-54.47-34.55-85.65-42.08-7.78-1.93-15.69-3.34-23.63-4.45-3.91-.59-7.85-.82-11.77-1.24-7.39-.57-14.81-.72-22.22-.58zM139.26 83.53c13.3-8.89 28.08-15.38 43.3-20.18-3.17 1.77-6.44 3.38-9.53 5.29-11.21 6.68-21.52 14.9-30.38 24.49-6.8 7.43-12.76 15.73-17.01 24.89-3.29 6.86-5.64 14.19-6.86 21.71-.93 4.85-1.3 9.81-1.17 14.75.13 13.66 4.44 27.08 11.29 38.82 5.92 10.22 13.63 19.33 22.36 27.26 4.85 4.36 10.24 8.09 14.95 12.6 2.26 2.19 4.49 4.42 6.43 6.91 2.62 3.31 4.89 6.99 5.99 11.1.9 3.02.66 6.2.69 9.31.02 4.1-.04 8.2.03 12.3.14 3.54-.02 7.09.11 10.63.08 2.38.02 4.76.05 7.14.16 5.77.06 11.53.15 17.3.11 2.91.02 5.82.13 8.74.03 1.63.13 3.28-.03 4.91-.91.12-1.82.18-2.73.16-10.99 0-21.88-2.63-31.95-6.93-6-2.7-11.81-5.89-17.09-9.83-5.75-4.19-11.09-8.96-15.79-14.31-6.53-7.24-11.98-15.39-16.62-23.95-1.07-2.03-2.24-4.02-3.18-6.12-1.16-2.64-2.62-5.14-3.67-7.82-4.05-9.68-6.57-19.94-8.08-30.31-.49-4.44-1.09-8.88-1.2-13.35-.7-15.73.84-31.55 4.67-46.82 2.12-8.15 4.77-16.18 8.31-23.83 6.32-14.2 15.34-27.18 26.3-38.19 6.28-6.2 13.13-11.84 20.53-16.67zm175.37-20.12c2.74.74 5.41 1.74 8.09 2.68 6.36 2.33 12.68 4.84 18.71 7.96 13.11 6.44 25.31 14.81 35.82 24.97 10.2 9.95 18.74 21.6 25.14 34.34 1.28 2.75 2.64 5.46 3.81 8.26 6.31 15.1 10 31.26 11.23 47.57.41 4.54.44 9.09.45 13.64.07 11.64-1.49 23.25-4.3 34.53-1.97 7.27-4.35 14.49-7.86 21.18-3.18 6.64-6.68 13.16-10.84 19.24-6.94 10.47-15.6 19.87-25.82 27.22-10.48 7.64-22.64 13.02-35.4 15.38-3.51.69-7.08 1.08-10.66 1.21-1.85.06-3.72.16-5.56-.1-.28-2.15 0-4.31-.01-6.46-.03-3.73.14-7.45.1-11.17.19-7.02.02-14.05.21-21.07.03-2.38-.03-4.76.03-7.14.17-5.07-.04-10.14.14-15.21.1-2.99-.24-6.04.51-8.96.66-2.5 1.78-4.86 3.09-7.08 4.46-7.31 11.06-12.96 17.68-18.26 5.38-4.18 10.47-8.77 15.02-13.84 7.68-8.37 14.17-17.88 18.78-28.27 2.5-5.93 4.52-12.1 5.55-18.46.86-4.37 1.06-8.83 1.01-13.27-.02-7.85-1.4-15.65-3.64-23.17-1.75-5.73-4.27-11.18-7.09-16.45-3.87-6.93-8.65-13.31-13.96-19.2-9.94-10.85-21.75-19.94-34.6-27.1-1.85-1.02-3.84-1.82-5.63-2.97zm-100.8 58.45c.98-1.18 1.99-2.33 3.12-3.38-.61.93-1.27 1.81-1.95 2.68-3.1 3.88-5.54 8.31-7.03 13.06-.87 3.27-1.68 6.6-1.73 10-.07 2.52-.08 5.07.32 7.57 1.13 7.63 4.33 14.85 8.77 21.12 2 2.7 4.25 5.27 6.92 7.33 1.62 1.27 3.53 2.09 5.34 3.05 3.11 1.68 6.32 3.23 9.07 5.48 2.67 2.09 4.55 5.33 4.4 8.79-.01 73.67 0 147.34-.01 221.02 0 1.35-.08 2.7.04 4.04.13 1.48.82 2.83 1.47 4.15.86 1.66 1.78 3.34 3.18 4.62.85.77 1.97 1.4 3.15 1.24 1.5-.2 2.66-1.35 3.45-2.57.96-1.51 1.68-3.16 2.28-4.85.76-2.13.44-4.42.54-6.63.14-4.03-.02-8.06.14-12.09.03-5.89.03-11.77.06-17.66.14-3.62.03-7.24.11-10.86.15-4.03-.02-8.06.14-12.09.03-5.99.03-11.98.07-17.97.14-3.62.02-7.24.11-10.86.14-3.93-.02-7.86.14-11.78.03-5.99.03-11.98.06-17.97.16-3.94-.01-7.88.19-11.82.29 1.44.13 2.92.22 4.38.19 3.61.42 7.23.76 10.84.32 3.44.44 6.89.86 10.32.37 3.1.51 6.22.95 9.31.57 4.09.87 8.21 1.54 12.29 1.46 9.04 2.83 18.11 5.09 26.99 1.13 4.82 2.4 9.61 4 14.3 2.54 7.9 5.72 15.67 10.31 22.62 1.73 2.64 3.87 4.98 6.1 7.21.27.25.55.51.88.71.6.25 1.31-.07 1.7-.57.71-.88 1.17-1.94 1.7-2.93 4.05-7.8 8.18-15.56 12.34-23.31.7-1.31 1.44-2.62 2.56-3.61 1.75-1.57 3.84-2.69 5.98-3.63 2.88-1.22 5.9-2.19 9.03-2.42 6.58-.62 13.11.75 19.56 1.85 3.69.58 7.4 1.17 11.13 1.41 3.74.1 7.48.05 11.21-.28 8.55-.92 16.99-2.96 24.94-6.25 5.3-2.24 10.46-4.83 15.31-7.93 11.46-7.21 21.46-16.57 30.04-27.01 1.17-1.42 2.25-2.9 3.46-4.28-1.2 3.24-2.67 6.37-4.16 9.48-1.25 2.9-2.84 5.61-4.27 8.42-5.16 9.63-11.02 18.91-17.75 27.52-4.03 5.21-8.53 10.05-13.33 14.57-6.64 6.05-14.07 11.37-22.43 14.76-8.21 3.37-17.31 4.63-26.09 3.29-3.56-.58-7.01-1.69-10.41-2.88-2.79-.97-5.39-2.38-8.03-3.69-3.43-1.71-6.64-3.81-9.71-6.08 2.71 3.06 5.69 5.86 8.7 8.61 4.27 3.76 8.74 7.31 13.63 10.23 3.98 2.45 8.29 4.4 12.84 5.51 1.46.37 2.96.46 4.45.6-1.25 1.1-2.63 2.04-3.99 2.98-9.61 6.54-20.01 11.86-30.69 16.43-20.86 8.7-43.17 13.97-65.74 15.34-4.66.24-9.32.36-13.98.36-4.98-.11-9.97-.13-14.92-.65-11.2-.76-22.29-2.73-33.17-5.43-10.35-2.71-20.55-6.12-30.3-10.55-8.71-3.86-17.12-8.42-24.99-13.79-1.83-1.31-3.74-2.53-5.37-4.08 6.6-1.19 13.03-3.39 18.99-6.48 5.74-2.86 10.99-6.66 15.63-11.07 2.24-2.19 4.29-4.59 6.19-7.09-3.43 2.13-6.93 4.15-10.62 5.78-4.41 2.16-9.07 3.77-13.81 5.02-5.73 1.52-11.74 1.73-17.61 1.14-8.13-.95-15.86-4.27-22.51-8.98-4.32-2.94-8.22-6.43-11.96-10.06-9.93-10.16-18.2-21.81-25.66-33.86-3.94-6.27-7.53-12.75-11.12-19.22-1.05-2.04-2.15-4.05-3.18-6.1 2.85 2.92 5.57 5.97 8.43 8.88 8.99 8.97 18.56 17.44 29.16 24.48 7.55 4.9 15.67 9.23 24.56 11.03 3.11.73 6.32.47 9.47.81 2.77.28 5.56.2 8.34.3 5.05.06 10.11.04 15.16-.16 3.65-.16 7.27-.66 10.89-1.09 2.07-.25 4.11-.71 6.14-1.2 3.88-.95 8.11-.96 11.83.61 4.76 1.85 8.44 5.64 11.38 9.71 2.16 3.02 4.06 6.22 5.66 9.58 1.16 2.43 2.46 4.79 3.55 7.26 1 2.24 2.15 4.42 3.42 6.52.67 1.02 1.4 2.15 2.62 2.55 1.06-.75 1.71-1.91 2.28-3.03 2.1-4.16 3.42-8.65 4.89-13.05 2.02-6.59 3.78-13.27 5.19-20.02 2.21-9.25 3.25-18.72 4.54-28.13.56-3.98.83-7.99 1.31-11.97.87-10.64 1.9-21.27 2.24-31.94.08-1.86.24-3.71.25-5.57.01-4.35.25-8.69.22-13.03-.01-2.38-.01-4.76 0-7.13.05-5.07-.2-10.14-.22-15.21-.2-6.61-.71-13.2-1.29-19.78-.73-5.88-1.55-11.78-3.12-17.51-2.05-7.75-5.59-15.03-9.8-21.82-3.16-5.07-6.79-9.88-11.09-14.03-3.88-3.86-8.58-7.08-13.94-8.45-1.5-.41-3.06-.45-4.59-.64.07-2.99.7-5.93 1.26-8.85 1.59-7.71 3.8-15.3 6.76-22.6 1.52-4.03 3.41-7.9 5.39-11.72 3.45-6.56 7.62-12.79 12.46-18.46zm31.27 1.7c.35-.06.71-.12 1.07-.19.19 1.79.09 3.58.1 5.37v38.13c-.01 1.74.13 3.49-.15 5.22-.36-.03-.71-.05-1.06-.05-.95-3.75-1.72-7.55-2.62-11.31-.38-1.53-.58-3.09-1.07-4.59-1.7-.24-3.43-.17-5.15-.2-5.06-.01-10.13 0-15.19-.01-1.66-.01-3.32.09-4.98-.03-.03-.39-.26-.91.16-1.18 1.28-.65 2.72-.88 4.06-1.35 3.43-1.14 6.88-2.16 10.31-3.31 1.39-.48 2.9-.72 4.16-1.54.04-.56.02-1.13-.05-1.68-1.23-.55-2.53-.87-3.81-1.28-3.13-1.03-6.29-1.96-9.41-3.02-1.79-.62-3.67-1-5.41-1.79-.03-.37-.07-.73-.11-1.09 5.09-.19 10.2.06 15.3-.12 3.36-.13 6.73.08 10.09-.07.12-.39.26-.77.37-1.16 1.08-4.94 2.33-9.83 3.39-14.75zm5.97-.2c.36.05.72.12 1.08.2.98 3.85 1.73 7.76 2.71 11.61.36 1.42.56 2.88 1.03 4.27 2.53.18 5.07-.01 7.61.05 5.16.12 10.33.12 15.49.07.76-.01 1.52.03 2.28.08-.04.36-.07.72-.1 1.08-1.82.83-3.78 1.25-5.67 1.89-3.73 1.23-7.48 2.39-11.22 3.57-.57.17-1.12.42-1.67.64-.15.55-.18 1.12-.12 1.69.87.48 1.82.81 2.77 1.09 4.88 1.52 9.73 3.14 14.63 4.6.38.13.78.27 1.13.49.4.27.23.79.15 1.18-1.66.13-3.31.03-4.97.04-5.17.01-10.33-.01-15.5.01-1.61.03-3.22-.02-4.82.21-.52 1.67-.72 3.42-1.17 5.11-.94 3.57-1.52 7.24-2.54 10.78-.36.01-.71.02-1.06.06-.29-1.73-.15-3.48-.15-5.22v-38.13c.02-1.78-.08-3.58.11-5.37zM65.05 168.33c1.12-2.15 2.08-4.4 3.37-6.46-1.82 7.56-2.91 15.27-3.62 23-.8 7.71-.85 15.49-.54 23.23 1.05 19.94 5.54 39.83 14.23 57.88 2.99 5.99 6.35 11.83 10.5 17.11 6.12 7.47 12.53 14.76 19.84 21.09 4.8 4.1 9.99 7.78 15.54 10.8 3.27 1.65 6.51 3.39 9.94 4.68 5.01 2.03 10.19 3.61 15.42 4.94 3.83.96 7.78 1.41 11.52 2.71 5 1.57 9.47 4.61 13.03 8.43 4.93 5.23 8.09 11.87 10.2 18.67.99 2.9 1.59 5.91 2.17 8.92.15.75.22 1.52.16 2.29-6.5 2.78-13.26 5.06-20.26 6.18-4.11.78-8.29.99-12.46 1.08-10.25.24-20.47-1.76-30.12-5.12-3.74-1.42-7.49-2.85-11.03-4.72-8.06-3.84-15.64-8.7-22.46-14.46-2.92-2.55-5.83-5.13-8.4-8.03-9.16-9.83-16.3-21.41-21.79-33.65-2.39-5.55-4.61-11.18-6.37-16.96-1.17-3.94-2.36-7.89-3.26-11.91-.75-2.94-1.22-5.95-1.87-8.92-.46-2.14-.69-4.32-1.03-6.48-.85-5.43-1.28-10.93-1.33-16.43.11-6.18.25-12.37 1.07-18.5.4-2.86.67-5.74 1.15-8.6.98-5.7 2.14-11.37 3.71-16.93 3.09-11.65 7.48-22.95 12.69-33.84zm363.73-6.44c1.1 1.66 1.91 3.48 2.78 5.26 2.1 4.45 4.24 8.9 6.02 13.49 7.61 18.76 12.3 38.79 13.04 59.05.02 1.76.07 3.52.11 5.29.13 9.57-1.27 19.09-3.18 28.45-.73 3.59-1.54 7.17-2.58 10.69-4.04 14.72-10 29-18.41 41.78-8.21 12.57-19.01 23.55-31.84 31.41-5.73 3.59-11.79 6.64-18.05 9.19-5.78 2.19-11.71 4.03-17.8 5.11-6.4 1.05-12.91 1.52-19.4 1.23-7.92-.48-15.78-2.07-23.21-4.85-1.94-.8-3.94-1.46-5.84-2.33-.21-1.51.25-2.99.53-4.46 1.16-5.74 3.03-11.36 5.7-16.58 2.37-4.51 5.52-8.65 9.46-11.9 2.43-2.05 5.24-3.61 8.16-4.83 3.58-1.5 7.47-1.97 11.24-2.83 7.23-1.71 14.37-3.93 21.15-7 10.35-4.65 19.71-11.38 27.65-19.46 1.59-1.61 3.23-3.18 4.74-4.87 3.37-3.76 6.71-7.57 9.85-11.53 7.48-10.07 12.82-21.59 16.71-33.48 1.58-5.3 3.21-10.6 4.21-16.05.63-2.87 1.04-5.78 1.52-8.68.87-6.09 1.59-12.22 1.68-18.38.12-6.65.14-13.32-.53-19.94-.73-7.99-1.87-15.96-3.71-23.78z\"],odysee:[512,512,[],\"e5c6\",\"M406.7 463c-42.3 30.8-94.4 49-150.7 49C144.9 512 50.3 441.2 14.9 342.2c2.4 1.7 5.9 3.6 7.9 4.4c16.3 7.4 40.1-5.4 62.9-28.7c6.9-6.9 14.4-12.4 22.8-17.3c18.3-11.9 37.6-20.8 58.4-27.2c0 0 22.3 34.2 43.1 74.8s-22.3 54-27.2 54c-.3 0-.8 0-1.5-.1c-11-.5-70-3-56 51.1c14.9 57.4 97.5 36.6 139.6 8.9s31.7-118.3 31.7-118.3c41.1-6.4 54 37.1 57.9 59.4c.8 4.6 1.1 9.9 1.4 15.5c1.1 21.2 2.3 45.6 35.3 46.4c5.3 0 10.6-.8 15.5-2zm-95.3-23.7c-2-.5-3.5-2.5-3-5c1-2.5 3-3.5 5-3s3.5 3 3 5s-2.5 3.5-5 3zm-207-95.6c1.5-.5 3.5 1 4 3c0 2-1 4-3 4c-1.5 .5-3.5-1-4-3c-.5-1.5 1-3.5 3-4zM451.8 421C489.3 376.4 512 318.8 512 256c0-67.5-26.1-128.9-68.8-174.7c-.1 23.5-6.1 48.2-16.8 69.2c-11.9 20.3-49 58.9-69.8 78.7c-.7 .3-1.1 .9-1.5 1.4c-.2 .2-.3 .4-.5 .6c-5 6.9-4 16.8 3 21.8c21.3 15.8 56.4 45.6 59.4 72.8c3.5 34.9 27.9 75.6 34.2 86.2l0 0c.8 1.3 1.3 2.1 1.4 2.4c0 2.2-.4 4.3-.8 6.5zM390.7 251c-.5 3 1 5.9 4 6.4s5.9-1 6.4-4s-1-5.9-4-6.4c-3-1-5.9 1-6.4 4zm61.4-60.9l-11.4 5.4-3 12.9-5.4-11.4-12.9-3 11.4-5.4 3-12.9 5.4 11.4 12.9 3zM395.5 41.3c-16.2 8.2-22.1 32.8-29 61.4l0 0c-.3 1.4-.7 2.8-1 4.2c-9.5 38.5-30.6 37.6-41.7 37.2c-1.1 0-2-.1-2.9-.1c-5.1 0-6-4-8.9-17.1c-2.6-12.1-6.9-32-17.9-63.6C271.4-2.5 211.4 13.9 165.9 41.1C110.6 74.2 131.5 143 146.1 190.5c.7 2.2 1.4 4.4 2 6.6c-4 4-13.8 7.5-26 11.9c-12.1 4.3-26.6 9.5-40.3 16.9C47.9 243.9 11.5 274.9 2 288.5C.7 277.8 0 267 0 256C0 114.6 114.6 0 256 0c51.4 0 99.4 15.2 139.5 41.3zM58.9 189.6c-1.5-2-4.5-3-6.4-1.5s-3 4.5-1.5 6.4s4.5 3 6.4 1.5c2.5-1.5 3-4.5 1.5-6.4zM327.3 64.9c2-1.5 5-.5 6.4 1.5c1.5 2.5 1 5.4-1.5 6.4c-2 1.5-5 .5-6.4-1.5s-.5-5 1.5-6.4zM95.1 105c-.5 1.5 .5 3 2 3c1.5 .5 3-.5 3-2c.5-1.5-.5-3-2-3s-3 .5-3 2zm84.7-.5c-3.5-43.1 37.1-54 37.1-54c44.1-15.4 56 5.9 66.4 37.6s3 42.6-38.6 58.9s-61.9-4.5-64.9-42.6zm89.6 14.9h1c2.5 0 5-2 5-5c2-6.9 1-14.4-2-20.8c-1.5-2-4-3.5-6.4-2.5c-3 1-4.5 4-3.5 6.9c2 4.5 3 9.9 1.5 14.9c-.5 3 1.5 5.9 4.5 6.4zm-9.9-41.6c-2 0-4-1-5-3s-2-3.5-3-5c-2-2-2-5.4 0-7.4s5.4-2 7.4 0c2 2.5 3.5 5 5 7.4s.5 5.9-2.5 7.4c-.6 0-1 .2-1.3 .3c-.2 .1-.4 .2-.6 .2z\"],\"square-whatsapp\":[448,512,[\"whatsapp-square\"],\"f40c\",\"M92.1 254.6c0 24.9 7 49.2 20.2 70.1l3.1 5-13.3 48.6L152 365.2l4.8 2.9c20.2 12 43.4 18.4 67.1 18.4h.1c72.6 0 133.3-59.1 133.3-131.8c0-35.2-15.2-68.3-40.1-93.2c-25-25-58-38.7-93.2-38.7c-72.7 0-131.8 59.1-131.9 131.8zM274.8 330c-12.6 1.9-22.4 .9-47.5-9.9c-36.8-15.9-61.8-51.5-66.9-58.7c-.4-.6-.7-.9-.8-1.1c-2-2.6-16.2-21.5-16.2-41c0-18.4 9-27.9 13.2-32.3c.3-.3 .5-.5 .7-.8c3.6-4 7.9-5 10.6-5c2.6 0 5.3 0 7.6 .1c.3 0 .5 0 .8 0c2.3 0 5.2 0 8.1 6.8c1.2 2.9 3 7.3 4.9 11.8c3.3 8 6.7 16.3 7.3 17.6c1 2 1.7 4.3 .3 6.9c-3.4 6.8-6.9 10.4-9.3 13c-3.1 3.2-4.5 4.7-2.3 8.6c15.3 26.3 30.6 35.4 53.9 47.1c4 2 6.3 1.7 8.6-1c2.3-2.6 9.9-11.6 12.5-15.5c2.6-4 5.3-3.3 8.9-2s23.1 10.9 27.1 12.9c.8 .4 1.5 .7 2.1 1c2.8 1.4 4.7 2.3 5.5 3.6c.9 1.9 .9 9.9-2.4 19.1c-3.3 9.3-19.1 17.7-26.7 18.8zM448 96c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96zM148.1 393.9L64 416l22.5-82.2c-13.9-24-21.2-51.3-21.2-79.3C65.4 167.1 136.5 96 223.9 96c42.4 0 82.2 16.5 112.2 46.5c29.9 30 47.9 69.8 47.9 112.2c0 87.4-72.7 158.5-160.1 158.5c-26.6 0-52.7-6.7-75.8-19.3z\"],\"node-js\":[448,512,[],\"f3d3\",\"M224 508c-6.7 0-13.5-1.8-19.4-5.2l-61.7-36.5c-9.2-5.2-4.7-7-1.7-8 12.3-4.3 14.8-5.2 27.9-12.7 1.4-.8 3.2-.5 4.6.4l47.4 28.1c1.7 1 4.1 1 5.7 0l184.7-106.6c1.7-1 2.8-3 2.8-5V149.3c0-2.1-1.1-4-2.9-5.1L226.8 37.7c-1.7-1-4-1-5.7 0L36.6 144.3c-1.8 1-2.9 3-2.9 5.1v213.1c0 2 1.1 4 2.9 4.9l50.6 29.2c27.5 13.7 44.3-2.4 44.3-18.7V167.5c0-3 2.4-5.3 5.4-5.3h23.4c2.9 0 5.4 2.3 5.4 5.3V378c0 36.6-20 57.6-54.7 57.6-10.7 0-19.1 0-42.5-11.6l-48.4-27.9C8.1 389.2.7 376.3.7 362.4V149.3c0-13.8 7.4-26.8 19.4-33.7L204.6 9c11.7-6.6 27.2-6.6 38.8 0l184.7 106.7c12 6.9 19.4 19.8 19.4 33.7v213.1c0 13.8-7.4 26.7-19.4 33.7L243.4 502.8c-5.9 3.4-12.6 5.2-19.4 5.2zm149.1-210.1c0-39.9-27-50.5-83.7-58-57.4-7.6-63.2-11.5-63.2-24.9 0-11.1 4.9-25.9 47.4-25.9 37.9 0 51.9 8.2 57.7 33.8.5 2.4 2.7 4.2 5.2 4.2h24c1.5 0 2.9-.6 3.9-1.7s1.5-2.6 1.4-4.1c-3.7-44.1-33-64.6-92.2-64.6-52.7 0-84.1 22.2-84.1 59.5 0 40.4 31.3 51.6 81.8 56.6 60.5 5.9 65.2 14.8 65.2 26.7 0 20.6-16.6 29.4-55.5 29.4-48.9 0-59.6-12.3-63.2-36.6-.4-2.6-2.6-4.5-5.3-4.5h-23.9c-3 0-5.3 2.4-5.3 5.3 0 31.1 16.9 68.2 97.8 68.2 58.4-.1 92-23.2 92-63.4z\"],\"edge-legacy\":[512,512,[],\"e078\",\"M25.71,228.16l.35-.48c0,.16,0,.32-.07.48Zm460.58,15.51c0-44-7.76-84.46-28.81-122.4C416.5,47.88,343.91,8,258.89,8,119,7.72,40.62,113.21,26.06,227.68c42.42-61.31,117.07-121.38,220.37-125,0,0,109.67,0,99.42,105H170c6.37-37.39,18.55-59,34.34-78.93-75.05,34.9-121.85,96.1-120.75,188.32.83,71.45,50.13,144.84,120.75,172,83.35,31.84,192.77,7.2,240.13-21.33V363.31C363.6,419.8,173.6,424.23,172.21,295.74H486.29V243.67Z\"],slack:[448,512,[62447,\"slack-hash\"],\"f198\",\"M94.12 315.1c0 25.9-21.16 47.06-47.06 47.06S0 341 0 315.1c0-25.9 21.16-47.06 47.06-47.06h47.06v47.06zm23.72 0c0-25.9 21.16-47.06 47.06-47.06s47.06 21.16 47.06 47.06v117.84c0 25.9-21.16 47.06-47.06 47.06s-47.06-21.16-47.06-47.06V315.1zm47.06-188.98c-25.9 0-47.06-21.16-47.06-47.06S139 32 164.9 32s47.06 21.16 47.06 47.06v47.06H164.9zm0 23.72c25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06H47.06C21.16 243.96 0 222.8 0 196.9s21.16-47.06 47.06-47.06H164.9zm188.98 47.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06h-47.06V196.9zm-23.72 0c0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06V79.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06V196.9zM283.1 385.88c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06v-47.06h47.06zm0-23.72c-25.9 0-47.06-21.16-47.06-47.06 0-25.9 21.16-47.06 47.06-47.06h117.84c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06H283.1z\"],medrt:[544,512,[],\"f3c8\",\"M113.7 256c0 121.8 83.9 222.8 193.5 241.1-18.7 4.5-38.2 6.9-58.2 6.9C111.4 504 0 393 0 256S111.4 8 248.9 8c20.1 0 39.6 2.4 58.2 6.9C197.5 33.2 113.7 134.2 113.7 256m297.4 100.3c-77.7 55.4-179.6 47.5-240.4-14.6 5.5 14.1 12.7 27.7 21.7 40.5 61.6 88.2 182.4 109.3 269.7 47 87.3-62.3 108.1-184.3 46.5-272.6-9-12.9-19.3-24.3-30.5-34.2 37.4 78.8 10.7 178.5-67 233.9m-218.8-244c-1.4 1-2.7 2.1-4 3.1 64.3-17.8 135.9 4 178.9 60.5 35.7 47 42.9 106.6 24.4 158 56.7-56.2 67.6-142.1 22.3-201.8-50-65.5-149.1-74.4-221.6-19.8M296 224c-4.4 0-8-3.6-8-8v-40c0-4.4-3.6-8-8-8h-48c-4.4 0-8 3.6-8 8v40c0 4.4-3.6 8-8 8h-40c-4.4 0-8 3.6-8 8v48c0 4.4 3.6 8 8 8h40c4.4 0 8 3.6 8 8v40c0 4.4 3.6 8 8 8h48c4.4 0 8-3.6 8-8v-40c0-4.4 3.6-8 8-8h40c4.4 0 8-3.6 8-8v-48c0-4.4-3.6-8-8-8h-40z\"],usb:[640,512,[],\"f287\",\"M641.5 256c0 3.1-1.7 6.1-4.5 7.5L547.9 317c-1.4.8-2.8 1.4-4.5 1.4-1.4 0-3.1-.3-4.5-1.1-2.8-1.7-4.5-4.5-4.5-7.8v-35.6H295.7c25.3 39.6 40.5 106.9 69.6 106.9H392V354c0-5 3.9-8.9 8.9-8.9H490c5 0 8.9 3.9 8.9 8.9v89.1c0 5-3.9 8.9-8.9 8.9h-89.1c-5 0-8.9-3.9-8.9-8.9v-26.7h-26.7c-75.4 0-81.1-142.5-124.7-142.5H140.3c-8.1 30.6-35.9 53.5-69 53.5C32 327.3 0 295.3 0 256s32-71.3 71.3-71.3c33.1 0 61 22.8 69 53.5 39.1 0 43.9 9.5 74.6-60.4C255 88.7 273 95.7 323.8 95.7c7.5-20.9 27-35.6 50.4-35.6 29.5 0 53.5 23.9 53.5 53.5s-23.9 53.5-53.5 53.5c-23.4 0-42.9-14.8-50.4-35.6H294c-29.1 0-44.3 67.4-69.6 106.9h310.1v-35.6c0-3.3 1.7-6.1 4.5-7.8 2.8-1.7 6.4-1.4 8.9.3l89.1 53.5c2.8 1.1 4.5 4.1 4.5 7.2z\"],tumblr:[320,512,[],\"f173\",\"M309.8 480.3c-13.6 14.5-50 31.7-97.4 31.7-120.8 0-147-88.8-147-140.6v-144H17.9c-5.5 0-10-4.5-10-10v-68c0-7.2 4.5-13.6 11.3-16 62-21.8 81.5-76 84.3-117.1.8-11 6.5-16.3 16.1-16.3h70.9c5.5 0 10 4.5 10 10v115.2h83c5.5 0 10 4.4 10 9.9v81.7c0 5.5-4.5 10-10 10h-83.4V360c0 34.2 23.7 53.6 68 35.8 4.8-1.9 9-3.2 12.7-2.2 3.5.9 5.8 3.4 7.4 7.9l22 64.3c1.8 5 3.3 10.6-.4 14.5z\"],vaadin:[448,512,[],\"f408\",\"M224.5 140.7c1.5-17.6 4.9-52.7 49.8-52.7h98.6c20.7 0 32.1-7.8 32.1-21.6V54.1c0-12.2 9.3-22.1 21.5-22.1S448 41.9 448 54.1v36.5c0 42.9-21.5 62-66.8 62H280.7c-30.1 0-33 14.7-33 27.1 0 1.3-.1 2.5-.2 3.7-.7 12.3-10.9 22.2-23.4 22.2s-22.7-9.8-23.4-22.2c-.1-1.2-.2-2.4-.2-3.7 0-12.3-3-27.1-33-27.1H66.8c-45.3 0-66.8-19.1-66.8-62V54.1C0 41.9 9.4 32 21.6 32s21.5 9.9 21.5 22.1v12.3C43.1 80.2 54.5 88 75.2 88h98.6c44.8 0 48.3 35.1 49.8 52.7h.9zM224 456c11.5 0 21.4-7 25.7-16.3 1.1-1.8 97.1-169.6 98.2-171.4 11.9-19.6-3.2-44.3-27.2-44.3-13.9 0-23.3 6.4-29.8 20.3L224 362l-66.9-117.7c-6.4-13.9-15.9-20.3-29.8-20.3-24 0-39.1 24.6-27.2 44.3 1.1 1.9 97.1 169.6 98.2 171.4 4.3 9.3 14.2 16.3 25.7 16.3z\"],quora:[448,512,[],\"f2c4\",\"M440.5 386.7h-29.3c-1.5 13.5-10.5 30.8-33 30.8-20.5 0-35.3-14.2-49.5-35.8 44.2-34.2 74.7-87.5 74.7-153C403.5 111.2 306.8 32 205 32 105.3 32 7.3 111.7 7.3 228.7c0 134.1 131.3 221.6 249 189C276 451.3 302 480 351.5 480c81.8 0 90.8-75.3 89-93.3zM297 329.2C277.5 300 253.3 277 205.5 277c-30.5 0-54.3 10-69 22.8l12.2 24.3c6.2-3 13-4 19.8-4 35.5 0 53.7 30.8 69.2 61.3-10 3-20.7 4.2-32.7 4.2-75 0-107.5-53-107.5-156.7C97.5 124.5 130 71 205 71c76.2 0 108.7 53.5 108.7 157.7.1 41.8-5.4 75.6-16.7 100.5z\"],\"square-x-twitter\":[448,512,[],\"e61a\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm297.1 84L257.3 234.6 379.4 396H283.8L209 298.1 123.3 396H75.8l111-126.9L69.7 116h98l67.7 89.5L313.6 116h47.5zM323.3 367.6L153.4 142.9H125.1L296.9 367.6h26.3z\"],reacteurope:[576,512,[],\"f75d\",\"M250.6 211.74l5.8-4.1 5.8 4.1-2.1-6.8 5.7-4.3-7.1-.1-2.3-6.8-2.3 6.8-7.2.1 5.7 4.3zm63.7 0l5.8-4.1 5.8 4.1-2.1-6.8 5.7-4.3-7.2-.1-2.3-6.8-2.3 6.8-7.2.1 5.7 4.3zm-91.3 50.5h-3.4c-4.8 0-3.8 4-3.8 12.1 0 4.7-2.3 6.1-5.8 6.1s-5.8-1.4-5.8-6.1v-36.6c0-4.7 2.3-6.1 5.8-6.1s5.8 1.4 5.8 6.1c0 7.2-.7 10.5 3.8 10.5h3.4c4.7-.1 3.8-3.9 3.8-12.3 0-9.9-6.7-14.1-16.8-14.1h-.2c-10.1 0-16.8 4.2-16.8 14.1V276c0 10.4 6.7 14.1 16.8 14.1h.2c10.1 0 16.8-3.8 16.8-14.1 0-9.86 1.1-13.76-3.8-13.76zm-80.7 17.4h-14.7v-19.3H139c2.5 0 3.8-1.3 3.8-3.8v-2.1c0-2.5-1.3-3.8-3.8-3.8h-11.4v-18.3H142c2.5 0 3.8-1.3 3.8-3.8v-2.1c0-2.5-1.3-3.8-3.8-3.8h-21.7c-2.4-.1-3.7 1.3-3.7 3.8v59.1c0 2.5 1.3 3.8 3.8 3.8h21.9c2.5 0 3.8-1.3 3.8-3.8v-2.1c0-2.5-1.3-3.8-3.8-3.8zm-42-18.5c4.6-2 7.3-6 7.3-12.4v-11.9c0-10.1-6.7-14.1-16.8-14.1H77.4c-2.5 0-3.8 1.3-3.8 3.8v59.1c0 2.5 1.3 3.8 3.8 3.8h3.4c2.5 0 3.8-1.3 3.8-3.8v-22.9h5.6l7.4 23.5a4.1 4.1 0 0 0 4.3 3.2h3.3c2.8 0 4-1.8 3.2-4.4zm-3.8-14c0 4.8-2.5 6.1-6.1 6.1h-5.8v-20.9h5.8c3.6 0 6.1 1.3 6.1 6.1zM176 226a3.82 3.82 0 0 0-4.2-3.4h-6.9a3.68 3.68 0 0 0-4 3.4l-11 59.2c-.5 2.7.9 4.1 3.4 4.1h3a3.74 3.74 0 0 0 4.1-3.5l1.8-11.3h12.2l1.8 11.3a3.74 3.74 0 0 0 4.1 3.5h3.5c2.6 0 3.9-1.4 3.4-4.1zm-12.3 39.3l4.7-29.7 4.7 29.7zm89.3 20.2v-53.2h7.5c2.5 0 3.8-1.3 3.8-3.8v-2.1c0-2.5-1.3-3.8-3.8-3.8h-25.8c-2.5 0-3.8 1.3-3.8 3.8v2.1c0 2.5 1.3 3.8 3.8 3.8h7.3v53.2c0 2.5 1.3 3.8 3.8 3.8h3.4c2.5.04 3.8-1.3 3.8-3.76zm248-.8h-19.4V258h16.1a1.89 1.89 0 0 0 2-2v-.8a1.89 1.89 0 0 0-2-2h-16.1v-25.8h19.1a1.89 1.89 0 0 0 2-2v-.8a1.77 1.77 0 0 0-2-1.9h-22.2a1.62 1.62 0 0 0-2 1.8v63a1.81 1.81 0 0 0 2 1.9H501a1.81 1.81 0 0 0 2-1.9v-.8a1.84 1.84 0 0 0-2-1.96zm-93.1-62.9h-.8c-10.1 0-15.3 4.7-15.3 14.1V276c0 9.3 5.2 14.1 15.3 14.1h.8c10.1 0 15.3-4.8 15.3-14.1v-40.1c0-9.36-5.2-14.06-15.3-14.06zm10.2 52.4c-.1 8-3 11.1-10.5 11.1s-10.5-3.1-10.5-11.1v-36.6c0-7.9 3-11.1 10.5-11.1s10.5 3.2 10.5 11.1zm-46.5-14.5c6.1-1.6 9.2-6.1 9.2-13.3v-9.7c0-9.4-5.2-14.1-15.3-14.1h-13.7a1.81 1.81 0 0 0-2 1.9v63a1.81 1.81 0 0 0 2 1.9h1.2a1.74 1.74 0 0 0 1.9-1.9v-26.9h11.6l10.4 27.2a2.32 2.32 0 0 0 2.3 1.5h1.5c1.4 0 2-1 1.5-2.3zm-6.4-3.9H355v-28.5h10.2c7.5 0 10.5 3.1 10.5 11.1v6.4c0 7.84-3 11.04-10.5 11.04zm85.9-33.1h-13.7a1.62 1.62 0 0 0-2 1.8v63a1.81 1.81 0 0 0 2 1.9h1.2a1.74 1.74 0 0 0 1.9-1.9v-26.1h10.6c10.1 0 15.3-4.8 15.3-14.1v-10.5c0-9.4-5.2-14.1-15.3-14.1zm10.2 22.8c0 7.9-3 11.1-10.5 11.1h-10.2v-29.2h10.2c7.5-.1 10.5 3.1 10.5 11zM259.5 308l-2.3-6.8-2.3 6.8-7.1.1 5.7 4.3-2.1 6.8 5.8-4.1 5.8 4.1-2.1-6.8 5.7-4.3zm227.6-136.1a364.42 364.42 0 0 0-35.6-11.3c19.6-78 11.6-134.7-22.3-153.9C394.7-12.66 343.3 11 291 61.94q5.1 4.95 10.2 10.2c82.5-80 119.6-53.5 120.9-52.8 22.4 12.7 36 55.8 15.5 137.8a587.83 587.83 0 0 0-84.6-13C281.1 43.64 212.4 2 170.8 2 140 2 127 23 123.2 29.74c-18.1 32-13.3 84.2.1 133.8-70.5 20.3-120.7 54.1-120.3 95 .5 59.6 103.2 87.8 122.1 92.8-20.5 81.9-10.1 135.6 22.3 153.9 28 15.8 75.1 6 138.2-55.2q-5.1-4.95-10.2-10.2c-82.5 80-119.7 53.5-120.9 52.8-22.3-12.6-36-55.6-15.5-137.9 12.4 2.9 41.8 9.5 84.6 13 71.9 100.4 140.6 142 182.1 142 30.8 0 43.8-21 47.6-27.7 18-31.9 13.3-84.1-.1-133.8 152.3-43.8 156.2-130.2 33.9-176.3zM135.9 36.84c2.9-5.1 11.9-20.3 34.9-20.3 36.8 0 98.8 39.6 163.3 126.2a714 714 0 0 0-93.9.9 547.76 547.76 0 0 1 42.2-52.4Q277.3 86 272.2 81a598.25 598.25 0 0 0-50.7 64.2 569.69 569.69 0 0 0-84.4 14.6c-.2-1.4-24.3-82.2-1.2-123zm304.8 438.3c-2.9 5.1-11.8 20.3-34.9 20.3-36.7 0-98.7-39.4-163.3-126.2a695.38 695.38 0 0 0 93.9-.9 547.76 547.76 0 0 1-42.2 52.4q5.1 5.25 10.2 10.2a588.47 588.47 0 0 0 50.7-64.2c47.3-4.7 80.3-13.5 84.4-14.6 22.7 84.4 4.5 117 1.2 123zm9.1-138.6c-3.6-11.9-7.7-24.1-12.4-36.4a12.67 12.67 0 0 1-10.7-5.7l-.1.1a19.61 19.61 0 0 1-5.4 3.6c5.7 14.3 10.6 28.4 14.7 42.2a535.3 535.3 0 0 1-72 13c3.5-5.3 17.2-26.2 32.2-54.2a24.6 24.6 0 0 1-6-3.2c-1.1 1.2-3.6 4.2-10.9 4.2-6.2 11.2-17.4 30.9-33.9 55.2a711.91 711.91 0 0 1-112.4 1c-7.9-11.2-21.5-31.1-36.8-57.8a21 21 0 0 1-3-1.5c-1.9 1.6-3.9 3.2-12.6 3.2 6.3 11.2 17.5 30.7 33.8 54.6a548.81 548.81 0 0 1-72.2-11.7q5.85-21 14.1-42.9c-3.2 0-5.4.2-8.4-1a17.58 17.58 0 0 1-6.9 1c-4.9 13.4-9.1 26.5-12.7 39.4C-31.7 297-12.1 216 126.7 175.64c3.6 11.9 7.7 24.1 12.4 36.4 10.4 0 12.9 3.4 14.4 5.3a12 12 0 0 1 2.3-2.2c-5.8-14.7-10.9-29.2-15.2-43.3 7-1.8 32.4-8.4 72-13-15.9 24.3-26.7 43.9-32.8 55.3a14.22 14.22 0 0 1 6.4 8 23.42 23.42 0 0 1 10.2-8.4c6.5-11.7 17.9-31.9 34.8-56.9a711.72 711.72 0 0 1 112.4-1c31.5 44.6 28.9 48.1 42.5 64.5a21.42 21.42 0 0 1 10.4-7.4c-6.4-11.4-17.6-31-34.3-55.5 40.4 4.1 65 10 72.2 11.7-4 14.4-8.9 29.2-14.6 44.2a20.74 20.74 0 0 1 6.8 4.3l.1.1a12.72 12.72 0 0 1 8.9-5.6c4.9-13.4 9.2-26.6 12.8-39.5a359.71 359.71 0 0 1 34.5 11c106.1 39.9 74 87.9 72.6 90.4-19.8 35.1-80.1 55.2-105.7 62.5zm-114.4-114h-1.2a1.74 1.74 0 0 0-1.9 1.9v49.8c0 7.9-2.6 11.1-10.1 11.1s-10.1-3.1-10.1-11.1v-49.8a1.69 1.69 0 0 0-1.9-1.9H309a1.81 1.81 0 0 0-2 1.9v51.5c0 9.6 5 14.1 15.1 14.1h.4c10.1 0 15.1-4.6 15.1-14.1v-51.5a2 2 0 0 0-2.2-1.9zM321.7 308l-2.3-6.8-2.3 6.8-7.1.1 5.7 4.3-2.1 6.8 5.8-4.1 5.8 4.1-2.1-6.8 5.7-4.3zm-31.1 7.4l-2.3-6.8-2.3 6.8-7.1.1 5.7 4.3-2.1 6.8 5.8-4.1 5.8 4.1-2.1-6.8 5.7-4.3zm5.1-30.8h-19.4v-26.7h16.1a1.89 1.89 0 0 0 2-2v-.8a1.89 1.89 0 0 0-2-2h-16.1v-25.8h19.1a1.89 1.89 0 0 0 2-2v-.8a1.77 1.77 0 0 0-2-1.9h-22.2a1.81 1.81 0 0 0-2 1.9v63a1.81 1.81 0 0 0 2 1.9h22.5a1.77 1.77 0 0 0 2-1.9v-.8a1.83 1.83 0 0 0-2-2.06zm-7.4-99.4L286 192l-7.1.1 5.7 4.3-2.1 6.8 5.8-4.1 5.8 4.1-2.1-6.8 5.7-4.3-7.1-.1z\"],medium:[640,512,[62407,\"medium-m\"],\"f23a\",\"M180.5,74.262C80.813,74.262,0,155.633,0,256S80.819,437.738,180.5,437.738,361,356.373,361,256,280.191,74.262,180.5,74.262Zm288.25,10.646c-49.845,0-90.245,76.619-90.245,171.095s40.406,171.1,90.251,171.1,90.251-76.619,90.251-171.1H559C559,161.5,518.6,84.908,468.752,84.908Zm139.506,17.821c-17.526,0-31.735,68.628-31.735,153.274s14.2,153.274,31.735,153.274S640,340.631,640,256C640,171.351,625.785,102.729,608.258,102.729Z\"],amilia:[448,512,[],\"f36d\",\"M240.1 32c-61.9 0-131.5 16.9-184.2 55.4-5.1 3.1-9.1 9.2-7.2 19.4 1.1 5.1 5.1 27.4 10.2 39.6 4.1 10.2 14.2 10.2 20.3 6.1 32.5-22.3 96.5-47.7 152.3-47.7 57.9 0 58.9 28.4 58.9 73.1v38.5C203 227.7 78.2 251 46.7 264.2 11.2 280.5 16.3 357.7 16.3 376s15.2 104 124.9 104c47.8 0 113.7-20.7 153.3-42.1v25.4c0 3 2.1 8.2 6.1 9.1 3.1 1 50.7 2 59.9 2s62.5.3 66.5-.7c4.1-1 5.1-6.1 5.1-9.1V168c-.1-80.3-57.9-136-192-136zm50.2 348c-21.4 13.2-48.7 24.4-79.1 24.4-52.8 0-58.9-33.5-59-44.7 0-12.2-3-42.7 18.3-52.9 24.3-13.2 75.1-29.4 119.8-33.5z\"],mixcloud:[640,512,[],\"f289\",\"M212.98 346.566H179.789V195.114L185.973 173.47H175.262L137.127 346.566H76.1069L37.7323 173.47H27.276L33.1913 195.114V346.566H0V165H65.6506L102.248 338.096H110.747L147.329 165H212.98L212.98 346.566ZM544.459 283.589L458.434 345.655V307.534L531.329 255.776L458.434 204.017V165.896L544.459 228.231H553.721L640 165.896V204.017L566.866 255.776L640 307.549V345.655L553.721 283.589H544.459ZM430.157 272.311H248.113V239.255H430.157V272.311Z\"],flipboard:[448,512,[],\"f44d\",\"M0 32v448h448V32H0zm358.4 179.2h-89.6v89.6h-89.6v89.6H89.6V121.6h268.8v89.6z\"],viacoin:[384,512,[],\"f237\",\"M384 32h-64l-80.7 192h-94.5L64 32H0l48 112H0v48h68.5l13.8 32H0v48h102.8L192 480l89.2-208H384v-48h-82.3l13.8-32H384v-48h-48l48-112zM192 336l-27-64h54l-27 64z\"],\"critical-role\":[448,512,[],\"f6c9\",\"M225.82 0c.26.15 216.57 124.51 217.12 124.72 3 1.18 3.7 3.46 3.7 6.56q-.11 125.17 0 250.36a5.88 5.88 0 0 1-3.38 5.78c-21.37 12-207.86 118.29-218.93 124.58h-3C142 466.34 3.08 386.56 2.93 386.48a3.29 3.29 0 0 1-1.88-3.24c0-.87 0-225.94-.05-253.1a5 5 0 0 1 2.93-4.93C27.19 112.11 213.2 6 224.07 0zM215.4 20.42l-.22-.16Q118.06 75.55 21 130.87c0 .12.08.23.13.35l30.86 11.64c-7.71 6-8.32 6-10.65 5.13-.1 0-24.17-9.28-26.8-10v230.43c.88-1.41 64.07-110.91 64.13-111 1.62-2.82 3-1.92 9.12-1.52 1.4.09 1.48.22.78 1.42-41.19 71.33-36.4 63-67.48 116.94-.81 1.4-.61 1.13 1.25 1.13h186.5c1.44 0 1.69-.23 1.7-1.64v-8.88c0-1.34 2.36-.81-18.37-1-7.46-.07-14.14-3.22-21.38-12.7-7.38-9.66-14.62-19.43-21.85-29.21-2.28-3.08-3.45-2.38-16.76-2.38-1.75 0-1.78 0-1.76 1.82.29 26.21.15 25.27 1 32.66.52 4.37 2.16 4.2 9.69 4.81 3.14.26 3.88 4.08.52 4.92-1.57.39-31.6.51-33.67-.1a2.42 2.42 0 0 1 .3-4.73c3.29-.76 6.16.81 6.66-4.44 1.3-13.66 1.17-9 1.1-79.42 0-10.82-.35-12.58-5.36-13.55-1.22-.24-3.54-.16-4.69-.55-2.88-1-2-4.84 1.77-4.85 33.67 0 46.08-1.07 56.06 4.86 7.74 4.61 12 11.48 12.51 20.4.88 14.59-6.51 22.35-15 32.59a1.46 1.46 0 0 0 0 2.22c2.6 3.25 5 6.63 7.71 9.83 27.56 33.23 24.11 30.54 41.28 33.06.89.13 1-.42 1-1.15v-11c0-1 .32-1.43 1.41-1.26a72.37 72.37 0 0 0 23.58-.3c1.08-.15 1.5.2 1.48 1.33 0 .11.88 26.69.87 26.8-.05 1.52.67 1.62 1.89 1.62h186.71Q386.51 304.6 346 234.33c2.26-.66-.4 0 6.69-1.39 2-.39 2.05-.41 3.11 1.44 7.31 12.64 77.31 134 77.37 134.06V138c-1.72.5-103.3 38.72-105.76 39.68-1.08.42-1.55.2-1.91-.88-.63-1.9-1.34-3.76-2.09-5.62-.32-.79-.09-1.13.65-1.39.1 0 95.53-35.85 103-38.77-65.42-37.57-130.56-75-196-112.6l86.82 150.39-.28.33c-9.57-.9-10.46-1.6-11.8-3.94-1-1.69-73.5-127.71-82-142.16-9.1 14.67-83.56 146.21-85.37 146.32-2.93.17-5.88.08-9.25.08q43.25-74.74 86.18-149zm51.93 129.92a37.68 37.68 0 0 0 5.54-.85c1.69-.3 2.53.2 2.6 1.92 0 .11.07 19.06-.86 20.45s-1.88 1.22-2.6-.19c-5-9.69 6.22-9.66-39.12-12-.7 0-1 .23-1 .93 0 .13 3.72 122 3.73 122.11 0 .89.52 1.2 1.21 1.51a83.92 83.92 0 0 1 8.7 4.05c7.31 4.33 11.38 10.84 12.41 19.31 1.44 11.8-2.77 35.77-32.21 37.14-2.75.13-28.26 1.08-34.14-23.25-4.66-19.26 8.26-32.7 19.89-36.4a2.45 2.45 0 0 0 2-2.66c.1-5.63 3-107.1 3.71-121.35.05-1.08-.62-1.16-1.35-1.15-32.35.52-36.75-.34-40.22 8.52-2.42 6.18-4.14 1.32-3.95.23q1.59-9 3.31-18c.4-2.11 1.43-2.61 3.43-1.86 5.59 2.11 6.72 1.7 37.25 1.92 1.73 0 1.78-.08 1.82-1.85.68-27.49.58-22.59 1-29.55a2.69 2.69 0 0 0-1.63-2.8c-5.6-2.91-8.75-7.55-8.9-13.87-.35-14.81 17.72-21.67 27.38-11.51 6.84 7.19 5.8 18.91-2.45 24.15a4.35 4.35 0 0 0-2.22 4.34c0 .59-.11-4.31 1 30.05 0 .9.43 1.12 1.24 1.11.1 0 23-.09 34.47-.37zM68.27 141.7c19.84-4.51 32.68-.56 52.49 1.69 2.76.31 3.74 1.22 3.62 4-.21 5-1.16 22.33-1.24 23.15a2.65 2.65 0 0 1-1.63 2.34c-4.06 1.7-3.61-4.45-4-7.29-3.13-22.43-73.87-32.7-74.63 25.4-.31 23.92 17 53.63 54.08 50.88 27.24-2 19-20.19 24.84-20.47a2.72 2.72 0 0 1 3 3.36c-1.83 10.85-3.42 18.95-3.45 19.15-1.54 9.17-86.7 22.09-93.35-42.06-2.71-25.85 10.44-53.37 40.27-60.15zm80 87.67h-19.49a2.57 2.57 0 0 1-2.66-1.79c2.38-3.75 5.89.92 5.86-6.14-.08-25.75.21-38 .23-40.1 0-3.42-.53-4.65-3.32-4.94-7-.72-3.11-3.37-1.11-3.38 11.84-.1 22.62-.18 30.05.72 8.77 1.07 16.71 12.63 7.93 22.62-2 2.25-4 4.42-6.14 6.73.95 1.15 6.9 8.82 17.28 19.68 2.66 2.78 6.15 3.51 9.88 3.13a2.21 2.21 0 0 0 2.23-2.12c.3-3.42.26 4.73.45-40.58 0-5.65-.34-6.58-3.23-6.83-3.95-.35-4-2.26-.69-3.37l19.09-.09c.32 0 4.49.53 1 3.38 0 .05-.16 0-.24 0-3.61.26-3.94 1-4 4.62-.27 43.93.07 40.23.41 42.82.11.84.27 2.23 5.1 2.14 2.49 0 3.86 3.37 0 3.4-10.37.08-20.74 0-31.11.07-10.67 0-13.47-6.2-24.21-20.82-1.6-2.18-8.31-2.36-8.2-.37.88 16.47 0 17.78 4 17.67 4.75-.1 4.73 3.57.83 3.55zm275-10.15c-1.21 7.13.17 10.38-5.3 10.34-61.55-.42-47.82-.22-50.72-.31a18.4 18.4 0 0 1-3.63-.73c-2.53-.6 1.48-1.23-.38-5.6-1.43-3.37-2.78-6.78-4.11-10.19a1.94 1.94 0 0 0-2-1.44 138 138 0 0 0-14.58.07 2.23 2.23 0 0 0-1.62 1.06c-1.58 3.62-3.07 7.29-4.51 11-1.27 3.23 7.86 1.32 12.19 2.16 3 .57 4.53 3.72.66 3.73H322.9c-2.92 0-3.09-3.15-.74-3.21a6.3 6.3 0 0 0 5.92-3.47c1.5-3 2.8-6 4.11-9.09 18.18-42.14 17.06-40.17 18.42-41.61a1.83 1.83 0 0 1 3 0c2.93 3.34 18.4 44.71 23.62 51.92 2 2.7 5.74 2 6.36 2 3.61.13 4-1.11 4.13-4.29.09-1.87.08 1.17.07-41.24 0-4.46-2.36-3.74-5.55-4.27-.26 0-2.56-.63-.08-3.06.21-.2-.89-.24 21.7-.15 2.32 0 5.32 2.75-1.21 3.45a2.56 2.56 0 0 0-2.66 2.83c-.07 1.63-.19 38.89.29 41.21a3.06 3.06 0 0 0 3.23 2.43c13.25.43 14.92.44 16-3.41 1.67-5.78 4.13-2.52 3.73-.19zm-104.72 64.37c-4.24 0-4.42-3.39-.61-3.41 35.91-.16 28.11.38 37.19-.65 1.68-.19 2.38.24 2.25 1.89-.26 3.39-.64 6.78-1 10.16-.25 2.16-3.2 2.61-3.4-.15-.38-5.31-2.15-4.45-15.63-5.08-1.58-.07-1.64 0-1.64 1.52V304c0 1.65 0 1.6 1.62 1.47 3.12-.25 10.31.34 15.69-1.52.47-.16 3.3-1.79 3.07 1.76 0 .21-.76 10.35-1.18 11.39-.53 1.29-1.88 1.51-2.58.32-1.17-2 0-5.08-3.71-5.3-15.42-.9-12.91-2.55-12.91 6 0 12.25-.76 16.11 3.89 16.24 16.64.48 14.4 0 16.43-5.71.84-2.37 3.5-1.77 3.18.58-.44 3.21-.85 6.43-1.23 9.64 0 .36-.16 2.4-4.66 2.39-37.16-.08-34.54-.19-35.21-.31-2.72-.51-2.2-3 .22-3.45 1.1-.19 4 .54 4.16-2.56 2.44-56.22-.07-51.34-3.91-51.33zm-.41-109.52c2.46.61 3.13 1.76 2.95 4.65-.33 5.3-.34 9-.55 9.69-.66 2.23-3.15 2.12-3.34-.27-.38-4.81-3.05-7.82-7.57-9.15-26.28-7.73-32.81 15.46-27.17 30.22 5.88 15.41 22 15.92 28.86 13.78 5.92-1.85 5.88-6.5 6.91-7.58 1.23-1.3 2.25-1.84 3.12 1.1 0 .1.57 11.89-6 12.75-1.6.21-19.38 3.69-32.68-3.39-21-11.19-16.74-35.47-6.88-45.33 14-14.06 39.91-7.06 42.32-6.47zM289.8 280.14c3.28 0 3.66 3 .16 3.43-2.61.32-5-.42-5 5.46 0 2-.19 29.05.4 41.45.11 2.29 1.15 3.52 3.44 3.65 22 1.21 14.95-1.65 18.79-6.34 1.83-2.24 2.76.84 2.76 1.08.35 13.62-4 12.39-5.19 12.4l-38.16-.19c-1.93-.23-2.06-3-.42-3.38 2-.48 4.94.4 5.13-2.8 1-15.87.57-44.65.34-47.81-.27-3.77-2.8-3.27-5.68-3.71-2.47-.38-2-3.22.34-3.22 1.45-.02 17.97-.03 23.09-.02zm-31.63-57.79c.07 4.08 2.86 3.46 6 3.58 2.61.1 2.53 3.41-.07 3.43-6.48 0-13.7 0-21.61-.06-3.84 0-3.38-3.35 0-3.37 4.49 0 3.24 1.61 3.41-45.54 0-5.08-3.27-3.54-4.72-4.23-2.58-1.23-1.36-3.09.41-3.15 1.29 0 20.19-.41 21.17.21s1.87 1.65-.42 2.86c-1 .52-3.86-.28-4.15 2.47 0 .21-.82 1.63-.07 43.8zm-36.91 274.27a2.93 2.93 0 0 0 3.26 0c17-9.79 182-103.57 197.42-112.51-.14-.43 11.26-.18-181.52-.27-1.22 0-1.57.37-1.53 1.56 0 .1 1.25 44.51 1.22 50.38a28.33 28.33 0 0 1-1.36 7.71c-.55 1.83.38-.5-13.5 32.23-.73 1.72-1 2.21-2-.08-4.19-10.34-8.28-20.72-12.57-31a23.6 23.6 0 0 1-2-10.79c.16-2.46.8-16.12 1.51-48 0-1.95 0-2-2-2h-183c2.58 1.63 178.32 102.57 196 112.76zm-90.9-188.75c0 2.4.36 2.79 2.76 3 11.54 1.17 21 3.74 25.64-7.32 6-14.46 2.66-34.41-12.48-38.84-2-.59-16-2.76-15.94 1.51.05 8.04.01 11.61.02 41.65zm105.75-15.05c0 2.13 1.07 38.68 1.09 39.13.34 9.94-25.58 5.77-25.23-2.59.08-2 1.37-37.42 1.1-39.43-14.1 7.44-14.42 40.21 6.44 48.8a17.9 17.9 0 0 0 22.39-7.07c4.91-7.76 6.84-29.47-5.43-39a2.53 2.53 0 0 1-.36.12zm-12.28-198c-9.83 0-9.73 14.75-.07 14.87s10.1-14.88.07-14.91zm-80.15 103.83c0 1.8.41 2.4 2.17 2.58 13.62 1.39 12.51-11 12.16-13.36-1.69-11.22-14.38-10.2-14.35-7.81.05 4.5-.03 13.68.02 18.59zm212.32 6.4l-6.1-15.84c-2.16 5.48-4.16 10.57-6.23 15.84z\"],sitrox:[448,512,[],\"e44a\",\"M212.439 0.00846128V0H448V128H64C64 57.6008 141.755 0.475338 212.439 0.00846128ZM237.256 192V192.007C307.135 192.475 384 249.6 384 320H210.809V319.995C140.915 319.563 64 262.424 64 192H237.256ZM235.565 511.993C306.251 511.521 384 454.399 384 384H0V512H235.565V511.993Z\"],discourse:[448,512,[],\"f393\",\"M225.9 32C103.3 32 0 130.5 0 252.1 0 256 .1 480 .1 480l225.8-.2c122.7 0 222.1-102.3 222.1-223.9C448 134.3 348.6 32 225.9 32zM224 384c-19.4 0-37.9-4.3-54.4-12.1L88.5 392l22.9-75c-9.8-18.1-15.4-38.9-15.4-61 0-70.7 57.3-128 128-128s128 57.3 128 128-57.3 128-128 128z\"],joomla:[448,512,[],\"f1aa\",\"M.6 92.1C.6 58.8 27.4 32 60.4 32c30 0 54.5 21.9 59.2 50.2 32.6-7.6 67.1.6 96.5 30l-44.3 44.3c-20.5-20.5-42.6-16.3-55.4-3.5-14.3 14.3-14.3 37.9 0 52.2l99.5 99.5-44 44.3c-87.7-87.2-49.7-49.7-99.8-99.7-26.8-26.5-35-64.8-24.8-98.9C20.4 144.6.6 120.7.6 92.1zm129.5 116.4l44.3 44.3c10-10 89.7-89.7 99.7-99.8 14.3-14.3 37.6-14.3 51.9 0 12.8 12.8 17 35-3.5 55.4l44 44.3c31.2-31.2 38.5-67.6 28.9-101.2 29.2-4.1 51.9-29.2 51.9-59.5 0-33.2-26.8-60.1-59.8-60.1-30.3 0-55.4 22.5-59.5 51.6-33.8-9.9-71.7-1.5-98.3 25.1-18.3 19.1-71.1 71.5-99.6 99.9zm266.3 152.2c8.2-32.7-.9-68.5-26.3-93.9-11.8-12.2 5 4.7-99.5-99.7l-44.3 44.3 99.7 99.7c14.3 14.3 14.3 37.6 0 51.9-12.8 12.8-35 17-55.4-3.5l-44 44.3c27.6 30.2 68 38.8 102.7 28 5.5 27.4 29.7 48.1 58.9 48.1 33 0 59.8-26.8 59.8-60.1 0-30.2-22.5-55-51.6-59.1zm-84.3-53.1l-44-44.3c-87 86.4-50.4 50.4-99.7 99.8-14.3 14.3-37.6 14.3-51.9 0-13.1-13.4-16.9-35.3 3.2-55.4l-44-44.3c-30.2 30.2-38 65.2-29.5 98.3-26.7 6-46.2 29.9-46.2 58.2C0 453.2 26.8 480 59.8 480c28.6 0 52.5-19.8 58.6-46.7 32.7 8.2 68.5-.6 94.2-26 32.1-32 12.2-12.4 99.5-99.7z\"],mastodon:[448,512,[],\"f4f6\",\"M433 179.11c0-97.2-63.71-125.7-63.71-125.7-62.52-28.7-228.56-28.4-290.48 0 0 0-63.72 28.5-63.72 125.7 0 115.7-6.6 259.4 105.63 289.1 40.51 10.7 75.32 13 103.33 11.4 50.81-2.8 79.32-18.1 79.32-18.1l-1.7-36.9s-36.31 11.4-77.12 10.1c-40.41-1.4-83-4.4-89.63-54a102.54 102.54 0 0 1-.9-13.9c85.63 20.9 158.65 9.1 178.75 6.7 56.12-6.7 105-41.3 111.23-72.9 9.8-49.8 9-121.5 9-121.5zm-75.12 125.2h-46.63v-114.2c0-49.7-64-51.6-64 6.9v62.5h-46.33V197c0-58.5-64-56.6-64-6.9v114.2H90.19c0-122.1-5.2-147.9 18.41-175 25.9-28.9 79.82-30.8 103.83 6.1l11.6 19.5 11.6-19.5c24.11-37.1 78.12-34.8 103.83-6.1 23.71 27.3 18.4 53 18.4 175z\"],airbnb:[448,512,[],\"f834\",\"M224 373.12c-25.24-31.67-40.08-59.43-45-83.18-22.55-88 112.61-88 90.06 0-5.45 24.25-20.29 52-45 83.18zm138.15 73.23c-42.06 18.31-83.67-10.88-119.3-50.47 103.9-130.07 46.11-200-18.85-200-54.92 0-85.16 46.51-73.28 100.5 6.93 29.19 25.23 62.39 54.43 99.5-32.53 36.05-60.55 52.69-85.15 54.92-50 7.43-89.11-41.06-71.3-91.09 15.1-39.16 111.72-231.18 115.87-241.56 15.75-30.07 25.56-57.4 59.38-57.4 32.34 0 43.4 25.94 60.37 59.87 36 70.62 89.35 177.48 114.84 239.09 13.17 33.07-1.37 71.29-37.01 86.64zm47-136.12C280.27 35.93 273.13 32 224 32c-45.52 0-64.87 31.67-84.66 72.79C33.18 317.1 22.89 347.19 22 349.81-3.22 419.14 48.74 480 111.63 480c21.71 0 60.61-6.06 112.37-62.4 58.68 63.78 101.26 62.4 112.37 62.4 62.89.05 114.85-60.86 89.61-130.19.02-3.89-16.82-38.9-16.82-39.58z\"],\"wolf-pack-battalion\":[512,512,[],\"f514\",\"M267.73 471.53l10.56 15.84 5.28-12.32 5.28 7V512c21.06-7.92 21.11-66.86 25.51-97.21 4.62-31.89-.88-92.81 81.37-149.11-8.88-23.61-12-49.43-2.64-80.05C421 189 447 196.21 456.43 239.73l-30.35 8.36c11.15 23 17 46.76 13.2 72.14L412 313.18l-6.16 33.43-18.47-7-8.8 33.39-19.35-7 26.39 21.11 8.8-28.15L419 364.2l7-35.63 26.39 14.52c.25-20 7-58.06-8.8-84.45l26.39 5.28c4-22.07-2.38-39.21-7.92-56.74l22.43 9.68c-.44-25.07-29.94-56.79-61.58-58.5-20.22-1.09-56.74-25.17-54.1-51.9 2-19.87 17.45-42.62 43.11-49.7-44 36.51-9.68 67.3 5.28 73.46 4.4-11.44 17.54-69.08 0-130.2-40.39 22.87-89.65 65.1-93.2 147.79l-58 38.71-3.52 93.25L369.78 220l7 7-17.59 3.52-44 38.71-15.84-5.28-28.1 49.25-3.52 119.64 21.11 15.84-32.55 15.84-32.55-15.84 21.11-15.84-3.52-119.64-28.15-49.26-15.84 5.28-44-38.71-17.58-3.51 7-7 107.33 59.82-3.52-93.25-58.06-38.71C185 65.1 135.77 22.87 95.3 0c-17.54 61.12-4.4 118.76 0 130.2 15-6.16 49.26-36.95 5.28-73.46 25.66 7.08 41.15 29.83 43.11 49.7 2.63 26.74-33.88 50.81-54.1 51.9-31.65 1.72-61.15 33.44-61.59 58.51l22.43-9.68c-5.54 17.53-11.91 34.67-7.92 56.74l26.39-5.28c-15.76 26.39-9.05 64.43-8.8 84.45l26.39-14.52 7 35.63 24.63-5.28 8.8 28.15L153.35 366 134 373l-8.8-33.43-18.47 7-6.16-33.43-27.27 7c-3.82-25.38 2-49.1 13.2-72.14l-30.35-8.36c9.4-43.52 35.47-50.77 63.34-54.1 9.36 30.62 6.24 56.45-2.64 80.05 82.25 56.3 76.75 117.23 81.37 149.11 4.4 30.35 4.45 89.29 25.51 97.21v-29.83l5.28-7 5.28 12.32 10.56-15.84 11.44 21.11 11.43-21.1zm79.17-95L331.06 366c7.47-4.36 13.76-8.42 19.35-12.32-.6 7.22-.27 13.84-3.51 22.84zm28.15-49.26c-.4 10.94-.9 21.66-1.76 31.67-7.85-1.86-15.57-3.8-21.11-7 8.24-7.94 15.55-16.32 22.87-24.68zm24.63 5.28c0-13.43-2.05-24.21-5.28-33.43a235 235 0 0 1-18.47 27.27zm3.52-80.94c19.44 12.81 27.8 33.66 29.91 56.3-12.32-4.53-24.63-9.31-36.95-10.56 5.06-12 6.65-28.14 7-45.74zm-1.76-45.74c.81 14.3 1.84 28.82 1.76 42.23 19.22-8.11 29.78-9.72 44-14.08-10.61-18.96-27.2-25.53-45.76-28.16zM165.68 376.52L181.52 366c-7.47-4.36-13.76-8.42-19.35-12.32.6 7.26.27 13.88 3.51 22.88zm-28.15-49.26c.4 10.94.9 21.66 1.76 31.67 7.85-1.86 15.57-3.8 21.11-7-8.24-7.93-15.55-16.31-22.87-24.67zm-24.64 5.28c0-13.43 2-24.21 5.28-33.43a235 235 0 0 0 18.47 27.27zm-3.52-80.94c-19.44 12.81-27.8 33.66-29.91 56.3 12.32-4.53 24.63-9.31 37-10.56-5-12-6.65-28.14-7-45.74zm1.76-45.74c-.81 14.3-1.84 28.82-1.76 42.23-19.22-8.11-29.78-9.72-44-14.08 10.63-18.95 27.23-25.52 45.76-28.15z\"],\"buy-n-large\":[576,512,[],\"f8a6\",\"M288 32C133.27 32 7.79 132.32 7.79 256S133.27 480 288 480s280.21-100.32 280.21-224S442.73 32 288 32zm-85.39 357.19L64.1 390.55l77.25-290.74h133.44c63.15 0 84.93 28.65 78 72.84a60.24 60.24 0 0 1-1.5 6.85 77.39 77.39 0 0 0-17.21-1.93c-42.35 0-76.69 33.88-76.69 75.65 0 37.14 27.14 68 62.93 74.45-18.24 37.16-56.16 60.92-117.71 61.52zM358 207.11h32l-22.16 90.31h-35.41l-11.19-35.63-7.83 35.63h-37.83l26.63-90.31h31.34l15 36.75zm145.86 182.08H306.79L322.63 328a78.8 78.8 0 0 0 11.47.83c42.34 0 76.69-33.87 76.69-75.65 0-32.65-21-60.46-50.38-71.06l21.33-82.35h92.5l-53.05 205.36h103.87zM211.7 269.39H187l-13.8 56.47h24.7c16.14 0 32.11-3.18 37.94-26.65 5.56-22.31-7.99-29.82-24.14-29.82zM233 170h-21.34L200 217.71h21.37c18 0 35.38-14.64 39.21-30.14C265.23 168.71 251.07 170 233 170z\"],gulp:[256,512,[],\"f3ae\",\"M209.8 391.1l-14.1 24.6-4.6 80.2c0 8.9-28.3 16.1-63.1 16.1s-63.1-7.2-63.1-16.1l-5.8-79.4-14.9-25.4c41.2 17.3 126 16.7 165.6 0zm-196-253.3l13.6 125.5c5.9-20 20.8-47 40-55.2 6.3-2.7 12.7-2.7 18.7.9 5.2 3 9.6 9.3 10.1 11.8 1.2 6.5-2 9.1-4.5 9.1-3 0-5.3-4.6-6.8-7.3-4.1-7.3-10.3-7.6-16.9-2.8-6.9 5-12.9 13.4-17.1 20.7-5.1 8.8-9.4 18.5-12 28.2-1.5 5.6-2.9 14.6-.6 19.9 1 2.2 2.5 3.6 4.9 3.6 5 0 12.3-6.6 15.8-10.1 4.5-4.5 10.3-11.5 12.5-16l5.2-15.5c2.6-6.8 9.9-5.6 9.9 0 0 10.2-3.7 13.6-10 34.7-5.8 19.5-7.6 25.8-7.6 25.8-.7 2.8-3.4 7.5-6.3 7.5-1.2 0-2.1-.4-2.6-1.2-1-1.4-.9-5.3-.8-6.3.2-3.2 6.3-22.2 7.3-25.2-2 2.2-4.1 4.4-6.4 6.6-5.4 5.1-14.1 11.8-21.5 11.8-3.4 0-5.6-.9-7.7-2.4l7.6 79.6c2 5 39.2 17.1 88.2 17.1 49.1 0 86.3-12.2 88.2-17.1l10.9-94.6c-5.7 5.2-12.3 11.6-19.6 14.8-5.4 2.3-17.4 3.8-17.4-5.7 0-5.2 9.1-14.8 14.4-21.5 1.4-1.7 4.7-5.9 4.7-8.1 0-2.9-6-2.2-11.7 2.5-3.2 2.7-6.2 6.3-8.7 9.7-4.3 6-6.6 11.2-8.5 15.5-6.2 14.2-4.1 8.6-9.1 22-5 13.3-4.2 11.8-5.2 14-.9 1.9-2.2 3.5-4 4.5-1.9 1-4.5.9-6.1-.3-.9-.6-1.3-1.9-1.3-3.7 0-.9.1-1.8.3-2.7 1.5-6.1 7.8-18.1 15-34.3 1.6-3.7 1-2.6.8-2.3-6.2 6-10.9 8.9-14.4 10.5-5.8 2.6-13 2.6-14.5-4.1-.1-.4-.1-.8-.2-1.2-11.8 9.2-24.3 11.7-20-8.1-4.6 8.2-12.6 14.9-22.4 14.9-4.1 0-7.1-1.4-8.6-5.1-2.3-5.5 1.3-14.9 4.6-23.8 1.7-4.5 4-9.9 7.1-16.2 1.6-3.4 4.2-5.4 7.6-4.5.6.2 1.1.4 1.6.7 2.6 1.8 1.6 4.5.3 7.2-3.8 7.5-7.1 13-9.3 20.8-.9 3.3-2 9 1.5 9 2.4 0 4.7-.8 6.9-2.4 4.6-3.4 8.3-8.5 11.1-13.5 2-3.6 4.4-8.3 5.6-12.3.5-1.7 1.1-3.3 1.8-4.8 1.1-2.5 2.6-5.1 5.2-5.1 1.3 0 2.4.5 3.2 1.5 1.7 2.2 1.3 4.5.4 6.9-2 5.6-4.7 10.6-6.9 16.7-1.3 3.5-2.7 8-2.7 11.7 0 3.4 3.7 2.6 6.8 1.2 2.4-1.1 4.8-2.8 6.8-4.5 1.2-4.9.9-3.8 26.4-68.2 1.3-3.3 3.7-4.7 6.1-4.7 1.2 0 2.2.4 3.2 1.1 1.7 1.3 1.7 4.1 1 6.2-.7 1.9-.6 1.3-4.5 10.5-5.2 12.1-8.6 20.8-13.2 31.9-1.9 4.6-7.7 18.9-8.7 22.3-.6 2.2-1.3 5.8 1 5.8 5.4 0 19.3-13.1 23.1-17 .2-.3.5-.4.9-.6.6-1.9 1.2-3.7 1.7-5.5 1.4-3.8 2.7-8.2 5.3-11.3.8-1 1.7-1.6 2.7-1.6 2.8 0 4.2 1.2 4.2 4 0 1.1-.7 5.1-1.1 6.2 1.4-1.5 2.9-3 4.5-4.5 15-13.9 25.7-6.8 25.7.2 0 7.4-8.9 17.7-13.8 23.4-1.6 1.9-4.9 5.4-5 6.4 0 1.3.9 1.8 2.2 1.8 2 0 6.4-3.5 8-4.7 5-3.9 11.8-9.9 16.6-14.1l14.8-136.8c-30.5 17.1-197.6 17.2-228.3.2zm229.7-8.5c0 21-231.2 21-231.2 0 0-8.8 51.8-15.9 115.6-15.9 9 0 17.8.1 26.3.4l12.6-48.7L228.1.6c1.4-1.4 5.8-.2 9.9 3.5s6.6 7.9 5.3 9.3l-.1.1L185.9 74l-10 40.7c39.9 2.6 67.6 8.1 67.6 14.6zm-69.4 4.6c0-.8-.9-1.5-2.5-2.1l-.2.8c0 1.3-5 2.4-11.1 2.4s-11.1-1.1-11.1-2.4c0-.1 0-.2.1-.3l.2-.7c-1.8.6-3 1.4-3 2.3 0 2.1 6.2 3.7 13.7 3.7 7.7.1 13.9-1.6 13.9-3.7z\"],\"creative-commons-sampling-plus\":[496,512,[],\"f4f1\",\"M247.6 8C389.4 8 496 118.1 496 256c0 147.1-118.5 248-248.4 248C113.6 504 0 394.5 0 256 0 123.1 104.7 8 247.6 8zm.8 44.7C130.2 52.7 44.7 150.6 44.7 256c0 109.8 91.2 202.8 203.7 202.8 103.2 0 202.8-81.1 202.8-202.8.1-113.8-90.2-203.3-202.8-203.3zm107 205.6c-4.7 0-9 2.8-10.7 7.2l-4 9.5-11-92.8c-1.7-13.9-22-13.4-23.1.4l-4.3 51.4-5.2-68.8c-1.1-14.3-22.1-14.2-23.2 0l-3.5 44.9-5.9-94.3c-.9-14.5-22.3-14.4-23.2 0l-5.1 83.7-4.3-66.3c-.9-14.4-22.2-14.4-23.2 0l-5.3 80.2-4.1-57c-1.1-14.3-22-14.3-23.2-.2l-7.7 89.8-1.8-12.2c-1.7-11.4-17.1-13.6-22-3.3l-13.2 27.7H87.5v23.2h51.3c4.4 0 8.4-2.5 10.4-6.4l10.7 73.1c2 13.5 21.9 13 23.1-.7l3.8-43.6 5.7 78.3c1.1 14.4 22.3 14.2 23.2-.1l4.6-70.4 4.8 73.3c.9 14.4 22.3 14.4 23.2-.1l4.9-80.5 4.5 71.8c.9 14.3 22.1 14.5 23.2.2l4.6-58.6 4.9 64.4c1.1 14.3 22 14.2 23.1.1l6.8-83 2.7 22.3c1.4 11.8 17.7 14.1 22.3 3.1l18-43.4h50.5V258l-58.4.3zm-78 5.2h-21.9v21.9c0 4.1-3.3 7.5-7.5 7.5-4.1 0-7.5-3.3-7.5-7.5v-21.9h-21.9c-4.1 0-7.5-3.3-7.5-7.5 0-4.1 3.4-7.5 7.5-7.5h21.9v-21.9c0-4.1 3.4-7.5 7.5-7.5s7.5 3.3 7.5 7.5v21.9h21.9c4.1 0 7.5 3.3 7.5 7.5 0 4.1-3.4 7.5-7.5 7.5z\"],strava:[384,512,[],\"f428\",\"M158.4 0L7 292h89.2l62.2-116.1L220.1 292h88.5zm150.2 292l-43.9 88.2-44.6-88.2h-67.6l112.2 220 111.5-220z\"],ember:[640,512,[],\"f423\",\"M639.9 254.6c-1.1-10.7-10.7-6.8-10.7-6.8s-15.6 12.1-29.3 10.7c-13.7-1.3-9.4-32-9.4-32s3-28.1-5.1-30.4c-8.1-2.4-18 7.3-18 7.3s-12.4 13.7-18.3 31.2l-1.6.5s1.9-30.6-.3-37.6c-1.6-3.5-16.4-3.2-18.8 3s-14.2 49.2-15 67.2c0 0-23.1 19.6-43.3 22.8s-25-9.4-25-9.4 54.8-15.3 52.9-59.1-44.2-27.6-49-24c-4.6 3.5-29.4 18.4-36.6 59.7-.2 1.4-.7 7.5-.7 7.5s-21.2 14.2-33 18c0 0 33-55.6-7.3-80.9-11.4-6.8-21.3-.5-27.2 5.3 13.6-17.3 46.4-64.2 36.9-105.2-5.8-24.4-18-27.1-29.2-23.1-17 6.7-23.5 16.7-23.5 16.7s-22 32-27.1 79.5-12.6 105.1-12.6 105.1-10.5 10.2-20.2 10.7-5.4-28.7-5.4-28.7 7.5-44.6 7-52.1-1.1-11.6-9.9-14.2c-8.9-2.7-18.5 8.6-18.5 8.6s-25.5 38.7-27.7 44.6l-1.3 2.4-1.3-1.6s18-52.7.8-53.5-28.5 18.8-28.5 18.8-19.6 32.8-20.4 36.5l-1.3-1.6s8.1-38.2 6.4-47.6c-1.6-9.4-10.5-7.5-10.5-7.5s-11.3-1.3-14.2 5.9-13.7 55.3-15 70.7c0 0-28.2 20.2-46.8 20.4-18.5.3-16.7-11.8-16.7-11.8s68-23.3 49.4-69.2c-8.3-11.8-18-15.5-31.7-15.3-13.7.3-30.3 8.6-41.3 33.3-5.3 11.8-6.8 23-7.8 31.5 0 0-12.3 2.4-18.8-2.9s-10 0-10 0-11.2 14-.1 18.3 28.1 6.1 28.1 6.1c1.6 7.5 6.2 19.5 19.6 29.7 20.2 15.3 58.8-1.3 58.8-1.3l15.9-8.8s.5 14.6 12.1 16.7 16.4 1 36.5-47.9c11.8-25 12.6-23.6 12.6-23.6l1.3-.3s-9.1 46.8-5.6 59.7C187.7 319.4 203 318 203 318s8.3 2.4 15-21.2 19.6-49.9 19.6-49.9h1.6s-5.6 48.1 3 63.7 30.9 5.3 30.9 5.3 15.6-7.8 18-10.2c0 0 18.5 15.8 44.6 12.9 58.3-11.5 79.1-25.9 79.1-25.9s10 24.4 41.1 26.7c35.5 2.7 54.8-18.6 54.8-18.6s-.3 13.5 12.1 18.6 20.7-22.8 20.7-22.8l20.7-57.2h1.9s1.1 37.3 21.5 43.2 47-13.7 47-13.7 6.4-3.5 5.3-14.3zm-578 5.3c.8-32 21.8-45.9 29-39 7.3 7 4.6 22-9.1 31.4-13.7 9.5-19.9 7.6-19.9 7.6zm272.8-123.8s19.1-49.7 23.6-25.5-40 96.2-40 96.2c.5-16.2 16.4-70.7 16.4-70.7zm22.8 138.4c-12.6 33-43.3 19.6-43.3 19.6s-3.5-11.8 6.4-44.9 33.3-20.2 33.3-20.2 16.2 12.4 3.6 45.5zm84.6-14.6s-3-10.5 8.1-30.6c11-20.2 19.6-9.1 19.6-9.1s9.4 10.2-1.3 25.5-26.4 14.2-26.4 14.2z\"],\"canadian-maple-leaf\":[512,512,[],\"f785\",\"M383.8 351.7c2.5-2.5 105.2-92.4 105.2-92.4l-17.5-7.5c-10-4.9-7.4-11.5-5-17.4 2.4-7.6 20.1-67.3 20.1-67.3s-47.7 10-57.7 12.5c-7.5 2.4-10-2.5-12.5-7.5s-15-32.4-15-32.4-52.6 59.9-55.1 62.3c-10 7.5-20.1 0-17.6-10 0-10 27.6-129.6 27.6-129.6s-30.1 17.4-40.1 22.4c-7.5 5-12.6 5-17.6-5C293.5 72.3 255.9 0 255.9 0s-37.5 72.3-42.5 79.8c-5 10-10 10-17.6 5-10-5-40.1-22.4-40.1-22.4S183.3 182 183.3 192c2.5 10-7.5 17.5-17.6 10-2.5-2.5-55.1-62.3-55.1-62.3S98.1 167 95.6 172s-5 9.9-12.5 7.5C73 177 25.4 167 25.4 167s17.6 59.7 20.1 67.3c2.4 6 5 12.5-5 17.4L23 259.3s102.6 89.9 105.2 92.4c5.1 5 10 7.5 5.1 22.5-5.1 15-10.1 35.1-10.1 35.1s95.2-20.1 105.3-22.6c8.7-.9 18.3 2.5 18.3 12.5S241 512 241 512h30s-5.8-102.7-5.8-112.8 9.5-13.4 18.4-12.5c10 2.5 105.2 22.6 105.2 22.6s-5-20.1-10-35.1 0-17.5 5-22.5z\"],teamspeak:[576,512,[],\"f4f9\",\"M152.8 37.2c-32.2 38.1-56.1 82.6-69.9 130.5c0 .2-.1 .3-.1 .5C43.5 184.4 16 223 16 268c0 59.6 48.4 108 108 108s108-48.4 108-108c0-53.5-38.9-97.9-90-106.5c15.7-41.8 40.4-79.6 72.3-110.7c1.8-1.6 4-2.6 6.3-3.1c37.2-11.5 76.7-13.3 114.8-5.2C454.7 67.6 534 180.7 517.1 301.3c-8.4 62.6-38.6 112.7-87.7 151.4c-50.1 39.7-107.5 54.3-170.2 52.2l-24-1c12.4 2.8 25 4.9 37.6 6.3c40.7 4.2 81.4 2.1 120.1-12.5c94-35.5 149.3-102.3 162.9-202.5c4.8-52.6-5.8-105.4-30.8-152C454.6 11.3 290.8-38.4 159 32c-2.4 1.4-4.5 3.1-6.3 5.2zM309.4 433.9c-2.1 11.5-4.2 21.9-14.6 31.3c53.2-1 123.2-29.2 161.8-97.1c39.7-69.9 37.6-139.9-6.3-207.8C413.8 105 360.5 77.9 293.7 73.7c1.5 2.3 3.2 4.4 5.2 6.3l5.2 6.3c25.1 31.3 37.6 67.9 42.8 107.5c2.1 15.7-1 30.3-13.6 41.8c-4.2 3.1-5.2 6.3-4.2 10.4l7.3 17.7L365.7 318c5.2 11.5 4.2 19.8-6.3 28.2c-3.2 2.5-6.7 4.6-10.4 6.3l-18.8 8.4 3.1 13.6c3.1 6.3 1 12.5-3.1 17.7c-2.5 2.4-3.8 5.9-3.1 9.4c2.1 11.5-2.1 19.8-12.5 25.1c-2.1 1-4.2 5.2-5.2 7.3zm-133.6-3.1c16.7 11.5 34.5 20.9 53.2 26.1c24 5.2 41.8-6.3 44.9-30.3c1-8.4 5.2-14.6 12.5-17.7c7.3-4.2 8.4-7.3 2.1-13.6l-9.4-8.4 13.6-4.2c6.3-2.1 7.3-5.2 5.2-11.5c-1.4-3-2.4-6.2-3.1-9.4c-3.1-14.6-2.1-15.7 11.5-18.8c8.4-3.1 15.7-6.3 21.9-12.5c3.1-2.1 3.1-4.2 1-8.4l-16.7-30.3c-1-1.9-2.1-3.8-3.1-5.7c-6.4-11.7-13-23.6-15.7-37.1c-2.1-9.4-1-17.7 8.4-24c5.2-4.2 8.4-9.4 8.4-16.7c-.4-10.1-1.5-20.3-3.1-30.3c-6.3-37.6-23-68.9-51.2-95c-5.2-4.2-9.4-6.3-16.7-4.2L203.9 91.5c2 1.2 4 2.4 6 3.6l0 0c6.3 3.7 12.2 7.3 17 12.1c30.3 26.1 41.8 61.6 45.9 100.2c1 8.4 0 16.7-7.3 21.9c-8.4 5.2-10.4 12.5-7.3 20.9c4.9 13.2 10.4 26 16.7 38.6L291.6 318c-6.3 8.4-13.6 11.5-21.9 14.6c-12.5 3.1-14.6 7.3-10.4 20.9c.6 1.5 1.4 2.8 2.1 4.2c2.1 5.2 1 8.4-4.2 10.4l-12.5 3.1 5.2 4.2 4.2 4.2c4.2 5.2 4.2 8.4-2.1 10.4c-7.3 4.2-11.5 9.4-11.5 17.7c0 12.5-7.3 19.8-18.8 24c-3.8 1-7.6 1.5-11.5 1l-34.5-2.1z\"],pushed:[432,512,[],\"f3e1\",\"M407 111.9l-98.5-9 14-33.4c10.4-23.5-10.8-40.4-28.7-37L22.5 76.9c-15.1 2.7-26 18.3-21.4 36.6l105.1 348.3c6.5 21.3 36.7 24.2 47.7 7l35.3-80.8 235.2-231.3c16.4-16.8 4.3-42.9-17.4-44.8zM297.6 53.6c5.1-.7 7.5 2.5 5.2 7.4L286 100.9 108.6 84.6l189-31zM22.7 107.9c-3.1-5.1 1-10 6.1-9.1l248.7 22.7-96.9 230.7L22.7 107.9zM136 456.4c-2.6 4-7.9 3.1-9.4-1.2L43.5 179.7l127.7 197.6c-7 15-35.2 79.1-35.2 79.1zm272.8-314.5L210.1 337.3l89.7-213.7 106.4 9.7c4 1.1 5.7 5.3 2.6 8.6z\"],\"wordpress-simple\":[512,512,[],\"f411\",\"M256 8C119.3 8 8 119.2 8 256c0 136.7 111.3 248 248 248s248-111.3 248-248C504 119.2 392.7 8 256 8zM33 256c0-32.3 6.9-63 19.3-90.7l106.4 291.4C84.3 420.5 33 344.2 33 256zm223 223c-21.9 0-43-3.2-63-9.1l66.9-194.4 68.5 187.8c.5 1.1 1 2.1 1.6 3.1-23.1 8.1-48 12.6-74 12.6zm30.7-327.5c13.4-.7 25.5-2.1 25.5-2.1 12-1.4 10.6-19.1-1.4-18.4 0 0-36.1 2.8-59.4 2.8-21.9 0-58.7-2.8-58.7-2.8-12-.7-13.4 17.7-1.4 18.4 0 0 11.4 1.4 23.4 2.1l34.7 95.2L200.6 393l-81.2-241.5c13.4-.7 25.5-2.1 25.5-2.1 12-1.4 10.6-19.1-1.4-18.4 0 0-36.1 2.8-59.4 2.8-4.2 0-9.1-.1-14.4-.3C109.6 73 178.1 33 256 33c58 0 110.9 22.2 150.6 58.5-1-.1-1.9-.2-2.9-.2-21.9 0-37.4 19.1-37.4 39.6 0 18.4 10.6 33.9 21.9 52.3 8.5 14.8 18.4 33.9 18.4 61.5 0 19.1-7.3 41.2-17 72.1l-22.2 74.3-80.7-239.6zm81.4 297.2l68.1-196.9c12.7-31.8 17-57.2 17-79.9 0-8.2-.5-15.8-1.5-22.9 17.4 31.8 27.3 68.2 27.3 107 0 82.3-44.6 154.1-110.9 192.7z\"],nutritionix:[400,512,[],\"f3d6\",\"M88 8.1S221.4-.1 209 112.5c0 0 19.1-74.9 103-40.6 0 0-17.7 74-88 56 0 0 14.6-54.6 66.1-56.6 0 0-39.9-10.3-82.1 48.8 0 0-19.8-94.5-93.6-99.7 0 0 75.2 19.4 77.6 107.5 0 .1-106.4 7-104-119.8zm312 315.6c0 48.5-9.7 95.3-32 132.3-42.2 30.9-105 48-168 48-62.9 0-125.8-17.1-168-48C9.7 419 0 372.2 0 323.7 0 275.3 17.7 229 40 192c42.2-30.9 97.1-48.6 160-48.6 63 0 117.8 17.6 160 48.6 22.3 37 40 83.3 40 131.7zM120 428c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zM192 428c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zM264 428c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zM336 428c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm0-66.2c0-15.5-12.5-28-28-28s-28 12.5-28 28 12.5 28 28 28 28-12.5 28-28zm24-39.6c-4.8-22.3-7.4-36.9-16-56-38.8-19.9-90.5-32-144-32S94.8 180.1 56 200c-8.8 19.5-11.2 33.9-16 56 42.2-7.9 98.7-14.8 160-14.8s117.8 6.9 160 14.8z\"],wodu:[640,512,[],\"e088\",\"M178.414 339.706H141.1L112.166 223.475h-.478L83.228 339.706H45.2L0 168.946H37.548L64.574 285.177h.478L94.707 168.946h35.157l29.178 117.667h.479L187.5 168.946h36.831zM271.4 212.713c38.984 0 64.1 25.828 64.1 65.291 0 39.222-25.111 65.05-64.1 65.05-38.743 0-63.855-25.828-63.855-65.05C207.547 238.541 232.659 212.713 271.4 212.713zm0 104.753c23.2 0 30.133-19.852 30.133-39.462 0-19.852-6.934-39.7-30.133-39.7-27.7 0-29.894 19.85-29.894 39.7C241.508 297.614 248.443 317.466 271.4 317.466zM435.084 323.922h-.478c-7.893 13.392-21.765 19.132-37.548 19.132-37.31 0-55.485-32.045-55.485-66.246 0-33.243 18.415-64.095 54.767-64.095 14.589 0 28.938 6.218 36.831 18.416h.24V168.946h33.96v170.76H435.084zM405.428 238.3c-22.24 0-29.894 19.134-29.894 39.463 0 19.371 8.848 39.7 29.894 39.7 22.482 0 29.178-19.613 29.178-39.94C434.606 257.436 427.432 238.3 405.428 238.3zM592.96 339.706H560.673V322.487h-.718c-8.609 13.87-23.436 20.567-37.786 20.567-36.113 0-45.2-20.328-45.2-50.941V216.061h33.959V285.9c0 20.329 5.979 30.372 21.765 30.372 18.415 0 26.306-10.283 26.306-35.393V216.061H592.96zM602.453 302.876H640v36.83H602.453z\"],\"google-pay\":[640,512,[],\"e079\",\"M105.72,215v41.25h57.1a49.66,49.66,0,0,1-21.14,32.6c-9.54,6.55-21.72,10.28-36,10.28-27.6,0-50.93-18.91-59.3-44.22a65.61,65.61,0,0,1,0-41l0,0c8.37-25.46,31.7-44.37,59.3-44.37a56.43,56.43,0,0,1,40.51,16.08L176.47,155a101.24,101.24,0,0,0-70.75-27.84,105.55,105.55,0,0,0-94.38,59.11,107.64,107.64,0,0,0,0,96.18v.15a105.41,105.41,0,0,0,94.38,59c28.47,0,52.55-9.53,70-25.91,20-18.61,31.41-46.15,31.41-78.91A133.76,133.76,0,0,0,205.38,215Zm389.41-4c-10.13-9.38-23.93-14.14-41.39-14.14-22.46,0-39.34,8.34-50.5,24.86l20.85,13.26q11.45-17,31.26-17a34.05,34.05,0,0,1,22.75,8.79A28.14,28.14,0,0,1,487.79,248v5.51c-9.1-5.07-20.55-7.75-34.64-7.75-16.44,0-29.65,3.88-39.49,11.77s-14.82,18.31-14.82,31.56a39.74,39.74,0,0,0,13.94,31.27c9.25,8.34,21,12.51,34.79,12.51,16.29,0,29.21-7.3,39-21.89h1v17.72h22.61V250C510.25,233.45,505.26,220.34,495.13,211ZM475.9,300.3a37.32,37.32,0,0,1-26.57,11.16A28.61,28.61,0,0,1,431,305.21a19.41,19.41,0,0,1-7.77-15.63c0-7,3.22-12.81,9.54-17.42s14.53-7,24.07-7C470,265,480.3,268,487.64,273.94,487.64,284.07,483.68,292.85,475.9,300.3Zm-93.65-142A55.71,55.71,0,0,0,341.74,142H279.07V328.74H302.7V253.1h39c16,0,29.5-5.36,40.51-15.93.88-.89,1.76-1.79,2.65-2.68A54.45,54.45,0,0,0,382.25,158.26Zm-16.58,62.23a30.65,30.65,0,0,1-23.34,9.68H302.7V165h39.63a32,32,0,0,1,22.6,9.23A33.18,33.18,0,0,1,365.67,220.49ZM614.31,201,577.77,292.7h-.45L539.9,201H514.21L566,320.55l-29.35,64.32H561L640,201Z\"],intercom:[448,512,[],\"f7af\",\"M392 32H56C25.1 32 0 57.1 0 88v336c0 30.9 25.1 56 56 56h336c30.9 0 56-25.1 56-56V88c0-30.9-25.1-56-56-56zm-108.3 82.1c0-19.8 29.9-19.8 29.9 0v199.5c0 19.8-29.9 19.8-29.9 0V114.1zm-74.6-7.5c0-19.8 29.9-19.8 29.9 0v216.5c0 19.8-29.9 19.8-29.9 0V106.6zm-74.7 7.5c0-19.8 29.9-19.8 29.9 0v199.5c0 19.8-29.9 19.8-29.9 0V114.1zM59.7 144c0-19.8 29.9-19.8 29.9 0v134.3c0 19.8-29.9 19.8-29.9 0V144zm323.4 227.8c-72.8 63-241.7 65.4-318.1 0-15-12.8 4.4-35.5 19.4-22.7 65.9 55.3 216.1 53.9 279.3 0 14.9-12.9 34.3 9.8 19.4 22.7zm5.2-93.5c0 19.8-29.9 19.8-29.9 0V144c0-19.8 29.9-19.8 29.9 0v134.3z\"],zhihu:[640,512,[],\"f63f\",\"M170.54 148.13v217.54l23.43.01 7.71 26.37 42.01-26.37h49.53V148.13H170.54zm97.75 193.93h-27.94l-27.9 17.51-5.08-17.47-11.9-.04V171.75h72.82v170.31zm-118.46-94.39H97.5c1.74-27.1 2.2-51.59 2.2-73.46h51.16s1.97-22.56-8.58-22.31h-88.5c3.49-13.12 7.87-26.66 13.12-40.67 0 0-24.07 0-32.27 21.57-3.39 8.9-13.21 43.14-30.7 78.12 5.89-.64 25.37-1.18 36.84-22.21 2.11-5.89 2.51-6.66 5.14-14.53h28.87c0 10.5-1.2 66.88-1.68 73.44H20.83c-11.74 0-15.56 23.62-15.56 23.62h65.58C66.45 321.1 42.83 363.12 0 396.34c20.49 5.85 40.91-.93 51-9.9 0 0 22.98-20.9 35.59-69.25l53.96 64.94s7.91-26.89-1.24-39.99c-7.58-8.92-28.06-33.06-36.79-41.81L87.9 311.95c4.36-13.98 6.99-27.55 7.87-40.67h61.65s-.09-23.62-7.59-23.62v.01zm412.02-1.6c20.83-25.64 44.98-58.57 44.98-58.57s-18.65-14.8-27.38-4.06c-6 8.15-36.83 48.2-36.83 48.2l19.23 14.43zm-150.09-59.09c-9.01-8.25-25.91 2.13-25.91 2.13s39.52 55.04 41.12 57.45l19.46-13.73s-25.67-37.61-34.66-45.86h-.01zM640 258.35c-19.78 0-130.91.93-131.06.93v-101c4.81 0 12.42-.4 22.85-1.2 40.88-2.41 70.13-4 87.77-4.81 0 0 12.22-27.19-.59-33.44-3.07-1.18-23.17 4.58-23.17 4.58s-165.22 16.49-232.36 18.05c1.6 8.82 7.62 17.08 15.78 19.55 13.31 3.48 22.69 1.7 49.15.89 24.83-1.6 43.68-2.43 56.51-2.43v99.81H351.41s2.82 22.31 25.51 22.85h107.94v70.92c0 13.97-11.19 21.99-24.48 21.12-14.08.11-26.08-1.15-41.69-1.81 1.99 3.97 6.33 14.39 19.31 21.84 9.88 4.81 16.17 6.57 26.02 6.57 29.56 0 45.67-17.28 44.89-45.31v-73.32h122.36c9.68 0 8.7-23.78 8.7-23.78l.03-.01z\"],korvue:[446,512,[],\"f42f\",\"M386.5 34h-327C26.8 34 0 60.8 0 93.5v327.1C0 453.2 26.8 480 59.5 480h327.1c33 0 59.5-26.8 59.5-59.5v-327C446 60.8 419.2 34 386.5 34zM87.1 120.8h96v116l61.8-116h110.9l-81.2 132H87.1v-132zm161.8 272.1l-65.7-113.6v113.6h-96V262.1h191.5l88.6 130.8H248.9z\"],pix:[512,512,[],\"e43a\",\"M242.4 292.5C247.8 287.1 257.1 287.1 262.5 292.5L339.5 369.5C353.7 383.7 372.6 391.5 392.6 391.5H407.7L310.6 488.6C280.3 518.1 231.1 518.1 200.8 488.6L103.3 391.2H112.6C132.6 391.2 151.5 383.4 165.7 369.2L242.4 292.5zM262.5 218.9C256.1 224.4 247.9 224.5 242.4 218.9L165.7 142.2C151.5 127.1 132.6 120.2 112.6 120.2H103.3L200.7 22.76C231.1-7.586 280.3-7.586 310.6 22.76L407.8 119.9H392.6C372.6 119.9 353.7 127.7 339.5 141.9L262.5 218.9zM112.6 142.7C126.4 142.7 139.1 148.3 149.7 158.1L226.4 234.8C233.6 241.1 243 245.6 252.5 245.6C261.9 245.6 271.3 241.1 278.5 234.8L355.5 157.8C365.3 148.1 378.8 142.5 392.6 142.5H430.3L488.6 200.8C518.9 231.1 518.9 280.3 488.6 310.6L430.3 368.9H392.6C378.8 368.9 365.3 363.3 355.5 353.5L278.5 276.5C264.6 262.6 240.3 262.6 226.4 276.6L149.7 353.2C139.1 363 126.4 368.6 112.6 368.6H80.78L22.76 310.6C-7.586 280.3-7.586 231.1 22.76 200.8L80.78 142.7H112.6z\"],\"steam-symbol\":[448,512,[],\"f3f6\",\"M395.5 177.5c0 33.8-27.5 61-61 61-33.8 0-61-27.3-61-61s27.3-61 61-61c33.5 0 61 27.2 61 61zm52.5.2c0 63-51 113.8-113.7 113.8L225 371.3c-4 43-40.5 76.8-84.5 76.8-40.5 0-74.7-28.8-83-67L0 358V250.7L97.2 290c15.1-9.2 32.2-13.3 52-11.5l71-101.7c.5-62.3 51.5-112.8 114-112.8C397 64 448 115 448 177.7zM203 363c0-34.7-27.8-62.5-62.5-62.5-4.5 0-9 .5-13.5 1.5l26 10.5c25.5 10.2 38 39 27.7 64.5-10.2 25.5-39.2 38-64.7 27.5-10.2-4-20.5-8.3-30.7-12.2 10.5 19.7 31.2 33.2 55.2 33.2 34.7 0 62.5-27.8 62.5-62.5zm207.5-185.3c0-42-34.3-76.2-76.2-76.2-42.3 0-76.5 34.2-76.5 76.2 0 42.2 34.3 76.2 76.5 76.2 41.9.1 76.2-33.9 76.2-76.2z\"]};!function(c){try{for(var l=arguments.length,s=new Array(1<l?l-1:0),a=1;a<l;a++)s[a-1]=arguments[a];c.apply(void 0,s)}catch(c){if(!h)throw c}}(function(){d(\"fab\",p),d(\"fa-brands\",p)})}(),function(){\"use strict\";var c={},l={};try{\"undefined\"!=typeof window&&(c=window),\"undefined\"!=typeof document&&(l=document)}catch(c){}var s=(c.navigator||{}).userAgent,a=void 0===s?\"\":s,z=c,e=l;z.document,e.documentElement&&e.head&&\"function\"==typeof e.addEventListener&&e.createElement,~a.indexOf(\"MSIE\")||a.indexOf(\"Trident/\");function H(l,c){var s,a=Object.keys(l);return Object.getOwnPropertySymbols&&(s=Object.getOwnPropertySymbols(l),c&&(s=s.filter(function(c){return Object.getOwnPropertyDescriptor(l,c).enumerable})),a.push.apply(a,s)),a}function t(l){for(var c=1;c<arguments.length;c++){var s=null!=arguments[c]?arguments[c]:{};c%2?H(Object(s),!0).forEach(function(c){V(l,c,s[c])}):Object.getOwnPropertyDescriptors?Object.defineProperties(l,Object.getOwnPropertyDescriptors(s)):H(Object(s)).forEach(function(c){Object.defineProperty(l,c,Object.getOwnPropertyDescriptor(s,c))})}return l}function V(c,l,s){return l in c?Object.defineProperty(c,l,{value:s,enumerable:!0,configurable:!0,writable:!0}):c[l]=s,c}function r(c,l){(null==l||l>c.length)&&(l=c.length);for(var s=0,a=new Array(l);s<l;s++)a[s]=c[s];return a}var M=\"___FONT_AWESOME___\",h=function(){try{return!0}catch(c){return!1}}(),n=\"classic\",i=\"sharp\",m=[n,i];function o(c){return new Proxy(c,{get:function(c,l){return l in c?c[l]:c[n]}})}o((V(C={},n,{fa:\"solid\",fas:\"solid\",\"fa-solid\":\"solid\",far:\"regular\",\"fa-regular\":\"regular\",fal:\"light\",\"fa-light\":\"light\",fat:\"thin\",\"fa-thin\":\"thin\",fad:\"duotone\",\"fa-duotone\":\"duotone\",fab:\"brands\",\"fa-brands\":\"brands\",fak:\"kit\",fakd:\"kit\",\"fa-kit\":\"kit\",\"fa-kit-duotone\":\"kit\"}),V(C,i,{fa:\"solid\",fass:\"solid\",\"fa-solid\":\"solid\",fasr:\"regular\",\"fa-regular\":\"regular\",fasl:\"light\",\"fa-light\":\"light\",fast:\"thin\",\"fa-thin\":\"thin\"}),C));var f=o((V(v={},n,{solid:\"fas\",regular:\"far\",light:\"fal\",thin:\"fat\",duotone:\"fad\",brands:\"fab\",kit:\"fak\"}),V(v,i,{solid:\"fass\",regular:\"fasr\",light:\"fasl\",thin:\"fast\"}),v)),e=(o((V(s={},n,{fab:\"fa-brands\",fad:\"fa-duotone\",fak:\"fa-kit\",fal:\"fa-light\",far:\"fa-regular\",fas:\"fa-solid\",fat:\"fa-thin\"}),V(s,i,{fass:\"fa-solid\",fasr:\"fa-regular\",fasl:\"fa-light\",fast:\"fa-thin\"}),s)),o((V(c={},n,{\"fa-brands\":\"fab\",\"fa-duotone\":\"fad\",\"fa-kit\":\"fak\",\"fa-light\":\"fal\",\"fa-regular\":\"far\",\"fa-solid\":\"fas\",\"fa-thin\":\"fat\"}),V(c,i,{\"fa-solid\":\"fass\",\"fa-regular\":\"fasr\",\"fa-light\":\"fasl\",\"fa-thin\":\"fast\"}),c)),o((V(l={},n,{900:\"fas\",400:\"far\",normal:\"far\",300:\"fal\",100:\"fat\"}),V(l,i,{900:\"fass\",400:\"fasr\",300:\"fasl\",100:\"fast\"}),l)),[1,2,3,4,5,6,7,8,9,10]),a=e.concat([11,12,13,14,15,16,17,18,19,20]),C=\"duotone-group\",v=\"swap-opacity\",s=\"primary\",c=\"secondary\",l=new Set;Object.keys(f[n]).map(l.add.bind(l)),Object.keys(f[i]).map(l.add.bind(l));[].concat(m,function(c){if(Array.isArray(c))return r(c)}(l=l)||function(c){if(\"undefined\"!=typeof Symbol&&null!=c[Symbol.iterator]||null!=c[\"@@iterator\"])return Array.from(c)}(l)||function(c,l){if(c){if(\"string\"==typeof c)return r(c,l);var s=Object.prototype.toString.call(c).slice(8,-1);return\"Map\"===(s=\"Object\"===s&&c.constructor?c.constructor.name:s)||\"Set\"===s?Array.from(c):\"Arguments\"===s||/^(?:Ui|I)nt(?:8|16|32)(?:Clamped)?Array$/.test(s)?r(c,l):void 0}}(l)||function(){throw new TypeError(\"Invalid attempt to spread non-iterable instance.\\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method.\")}(),[\"2xs\",\"xs\",\"sm\",\"lg\",\"xl\",\"2xl\",\"beat\",\"border\",\"fade\",\"beat-fade\",\"bounce\",\"flip-both\",\"flip-horizontal\",\"flip-vertical\",\"flip\",\"fw\",\"inverse\",\"layers-counter\",\"layers-text\",\"layers\",\"li\",\"pull-left\",\"pull-right\",\"pulse\",\"rotate-180\",\"rotate-270\",\"rotate-90\",\"rotate-by\",\"shake\",\"spin-pulse\",\"spin-reverse\",\"spin\",\"stack-1x\",\"stack-2x\",\"stack\",\"ul\",C,v,s,c]).concat(e.map(function(c){return\"\".concat(c,\"x\")})).concat(a.map(function(c){return\"w-\".concat(c)}));z=z||{};z[M]||(z[M]={}),z[M].styles||(z[M].styles={}),z[M].hooks||(z[M].hooks={}),z[M].shims||(z[M].shims=[]);var L=z[M];function u(a){return Object.keys(a).reduce(function(c,l){var s=a[l];return!!s.icon?c[s.iconName]=s.icon:c[l]=s,c},{})}function d(c,l,s){var a=(2<arguments.length&&void 0!==s?s:{}).skipHooks,s=void 0!==a&&a,a=u(l);\"function\"!=typeof L.hooks.addPack||s?L.styles[c]=t(t({},L.styles[c]||{}),a):L.hooks.addPack(c,u(l)),\"fas\"===c&&d(\"fa\",l)}var p={\"trash-can\":[448,512,[61460,\"trash-alt\"],\"f2ed\",\"M170.5 51.6L151.5 80h145l-19-28.4c-1.5-2.2-4-3.6-6.7-3.6H177.1c-2.7 0-5.2 1.3-6.7 3.6zm147-26.6L354.2 80H368h48 8c13.3 0 24 10.7 24 24s-10.7 24-24 24h-8V432c0 44.2-35.8 80-80 80H112c-44.2 0-80-35.8-80-80V128H24c-13.3 0-24-10.7-24-24S10.7 80 24 80h8H80 93.8l36.7-55.1C140.9 9.4 158.4 0 177.1 0h93.7c18.7 0 36.2 9.4 46.6 24.9zM80 128V432c0 17.7 14.3 32 32 32H336c17.7 0 32-14.3 32-32V128H80zm80 64V400c0 8.8-7.2 16-16 16s-16-7.2-16-16V192c0-8.8 7.2-16 16-16s16 7.2 16 16zm80 0V400c0 8.8-7.2 16-16 16s-16-7.2-16-16V192c0-8.8 7.2-16 16-16s16 7.2 16 16zm80 0V400c0 8.8-7.2 16-16 16s-16-7.2-16-16V192c0-8.8 7.2-16 16-16s16 7.2 16 16z\"],message:[512,512,[\"comment-alt\"],\"f27a\",\"M160 368c26.5 0 48 21.5 48 48v16l72.5-54.4c8.3-6.2 18.4-9.6 28.8-9.6H448c8.8 0 16-7.2 16-16V64c0-8.8-7.2-16-16-16H64c-8.8 0-16 7.2-16 16V352c0 8.8 7.2 16 16 16h96zm48 124l-.2 .2-5.1 3.8-17.1 12.8c-4.8 3.6-11.3 4.2-16.8 1.5s-8.8-8.2-8.8-14.3V474.7v-6.4V468v-4V416H112 64c-35.3 0-64-28.7-64-64V64C0 28.7 28.7 0 64 0H448c35.3 0 64 28.7 64 64V352c0 35.3-28.7 64-64 64H309.3L208 492z\"],\"file-lines\":[384,512,[128441,128462,61686,\"file-alt\",\"file-text\"],\"f15c\",\"M64 464c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16H224v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm56 256c-13.3 0-24 10.7-24 24s10.7 24 24 24H264c13.3 0 24-10.7 24-24s-10.7-24-24-24H120zm0 96c-13.3 0-24 10.7-24 24s10.7 24 24 24H264c13.3 0 24-10.7 24-24s-10.7-24-24-24H120z\"],\"calendar-days\":[448,512,[\"calendar-alt\"],\"f073\",\"M152 24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H64C28.7 64 0 92.7 0 128v16 48V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V192 144 128c0-35.3-28.7-64-64-64H344V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H152V24zM48 192h80v56H48V192zm0 104h80v64H48V296zm128 0h96v64H176V296zm144 0h80v64H320V296zm80-48H320V192h80v56zm0 160v40c0 8.8-7.2 16-16 16H320V408h80zm-128 0v56H176V408h96zm-144 0v56H64c-8.8 0-16-7.2-16-16V408h80zM272 248H176V192h96v56z\"],\"hand-point-right\":[512,512,[],\"f0a4\",\"M448 128l-177.6 0c1 5.2 1.6 10.5 1.6 16l0 16 32 0 144 0c8.8 0 16-7.2 16-16s-7.2-16-16-16zM224 144c0-17.7-14.3-32-32-32c0 0 0 0 0 0l-24 0c-66.3 0-120 53.7-120 120l0 48c0 52.5 33.7 97.1 80.7 113.4c-.5-3.1-.7-6.2-.7-9.4c0-20 9.2-37.9 23.6-49.7c-4.9-9-7.6-19.4-7.6-30.3c0-15.1 5.3-29 14-40c-8.8-11-14-24.9-14-40l0-40c0-13.3 10.7-24 24-24s24 10.7 24 24l0 40c0 8.8 7.2 16 16 16s16-7.2 16-16l0-40 0-40zM192 64s0 0 0 0c18 0 34.6 6 48 16l208 0c35.3 0 64 28.7 64 64s-28.7 64-64 64l-82 0c1.3 5.1 2 10.5 2 16c0 25.3-14.7 47.2-36 57.6c2.6 7 4 14.5 4 22.4c0 20-9.2 37.9-23.6 49.7c4.9 9 7.6 19.4 7.6 30.3c0 35.3-28.7 64-64 64l-64 0-24 0C75.2 448 0 372.8 0 280l0-48C0 139.2 75.2 64 168 64l24 0zm64 336c8.8 0 16-7.2 16-16s-7.2-16-16-16l-48 0-16 0c-8.8 0-16 7.2-16 16s7.2 16 16 16l64 0zm16-176c0 5.5-.7 10.9-2 16l2 0 32 0c8.8 0 16-7.2 16-16s-7.2-16-16-16l-32 0 0 16zm-24 64l-40 0c-8.8 0-16 7.2-16 16s7.2 16 16 16l48 0 16 0c8.8 0 16-7.2 16-16s-7.2-16-16-16l-24 0z\"],\"face-smile-beam\":[512,512,[128522,\"smile-beam\"],\"f5b8\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm177.6 62.1C192.8 334.5 218.8 352 256 352s63.2-17.5 78.4-33.9c9-9.7 24.2-10.4 33.9-1.4s10.4 24.2 1.4 33.9c-22 23.8-60 49.4-113.6 49.4s-91.7-25.5-113.6-49.4c-9-9.7-8.4-24.9 1.4-33.9s24.9-8.4 33.9 1.4zm40-89.3l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z\"],\"face-grin-stars\":[512,512,[129321,\"grin-stars\"],\"f587\",\"M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM183.2 132.6c-1.3-2.8-4.1-4.6-7.2-4.6s-5.9 1.8-7.2 4.6l-16.6 34.7-38.1 5c-3.1 .4-5.6 2.5-6.6 5.5s-.1 6.2 2.1 8.3l27.9 26.5-7 37.8c-.6 3 .7 6.1 3.2 7.9s5.8 2 8.5 .6L176 240.5l33.8 18.3c2.7 1.5 6 1.3 8.5-.6s3.7-4.9 3.2-7.9l-7-37.8L242.4 186c2.2-2.1 3.1-5.3 2.1-8.3s-3.5-5.1-6.6-5.5l-38.1-5-16.6-34.7zm160 0c-1.3-2.8-4.1-4.6-7.2-4.6s-5.9 1.8-7.2 4.6l-16.6 34.7-38.1 5c-3.1 .4-5.6 2.5-6.6 5.5s-.1 6.2 2.1 8.3l27.9 26.5-7 37.8c-.6 3 .7 6.1 3.2 7.9s5.8 2 8.5 .6L336 240.5l33.8 18.3c2.7 1.5 6 1.3 8.5-.6s3.7-4.9 3.2-7.9l-7-37.8L402.4 186c2.2-2.1 3.1-5.3 2.1-8.3s-3.5-5.1-6.6-5.5l-38.1-5-16.6-34.7zm6.3 175.8c-28.9 6.8-60.5 10.5-93.6 10.5s-64.7-3.7-93.6-10.5c-18.7-4.4-35.9 12-25.5 28.1c24.6 38.1 68.7 63.5 119.1 63.5s94.5-25.4 119.1-63.5c10.4-16.1-6.8-32.5-25.5-28.1z\"],\"address-book\":[512,512,[62138,\"contact-book\"],\"f2b9\",\"M384 48c8.8 0 16 7.2 16 16V448c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16H384zM96 0C60.7 0 32 28.7 32 64V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H96zM240 256a64 64 0 1 0 0-128 64 64 0 1 0 0 128zm-32 32c-44.2 0-80 35.8-80 80c0 8.8 7.2 16 16 16H336c8.8 0 16-7.2 16-16c0-44.2-35.8-80-80-80H208zM512 80c0-8.8-7.2-16-16-16s-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V80zM496 192c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V208c0-8.8-7.2-16-16-16zm16 144c0-8.8-7.2-16-16-16s-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V336z\"],comments:[640,512,[128490,61670],\"f086\",\"M88.2 309.1c9.8-18.3 6.8-40.8-7.5-55.8C59.4 230.9 48 204 48 176c0-63.5 63.8-128 160-128s160 64.5 160 128s-63.8 128-160 128c-13.1 0-25.8-1.3-37.8-3.6c-10.4-2-21.2-.6-30.7 4.2c-4.1 2.1-8.3 4.1-12.6 6c-16 7.2-32.9 13.5-49.9 18c2.8-4.6 5.4-9.1 7.9-13.6c1.1-1.9 2.2-3.9 3.2-5.9zM0 176c0 41.8 17.2 80.1 45.9 110.3c-.9 1.7-1.9 3.5-2.8 5.1c-10.3 18.4-22.3 36.5-36.6 52.1c-6.6 7-8.3 17.2-4.6 25.9C5.8 378.3 14.4 384 24 384c43 0 86.5-13.3 122.7-29.7c4.8-2.2 9.6-4.5 14.2-6.8c15.1 3 30.9 4.5 47.1 4.5c114.9 0 208-78.8 208-176S322.9 0 208 0S0 78.8 0 176zM432 480c16.2 0 31.9-1.6 47.1-4.5c4.6 2.3 9.4 4.6 14.2 6.8C529.5 498.7 573 512 616 512c9.6 0 18.2-5.7 22-14.5c3.8-8.8 2-19-4.6-25.9c-14.2-15.6-26.2-33.7-36.6-52.1c-.9-1.7-1.9-3.4-2.8-5.1C622.8 384.1 640 345.8 640 304c0-94.4-87.9-171.5-198.2-175.8c4.1 15.2 6.2 31.2 6.2 47.8l0 .6c87.2 6.7 144 67.5 144 127.4c0 28-11.4 54.9-32.7 77.2c-14.3 15-17.3 37.6-7.5 55.8c1.1 2 2.2 4 3.2 5.9c2.5 4.5 5.2 9 7.9 13.6c-17-4.5-33.9-10.7-49.9-18c-4.3-1.9-8.5-3.9-12.6-6c-9.5-4.8-20.3-6.2-30.7-4.2c-12.1 2.4-24.7 3.6-37.8 3.6c-61.7 0-110-26.5-136.8-62.3c-16 5.4-32.8 9.4-50 11.8C279 439.8 350 480 432 480z\"],paste:[512,512,[\"file-clipboard\"],\"f0ea\",\"M104.6 48H64C28.7 48 0 76.7 0 112V384c0 35.3 28.7 64 64 64h96V400H64c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16H80c0 17.7 14.3 32 32 32h72.4C202 108.4 227.6 96 256 96h62c-7.1-27.6-32.2-48-62-48H215.4C211.6 20.9 188.2 0 160 0s-51.6 20.9-55.4 48zM144 56a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zM448 464H256c-8.8 0-16-7.2-16-16V192c0-8.8 7.2-16 16-16l140.1 0L464 243.9V448c0 8.8-7.2 16-16 16zM256 512H448c35.3 0 64-28.7 64-64V243.9c0-12.7-5.1-24.9-14.1-33.9l-67.9-67.9c-9-9-21.2-14.1-33.9-14.1H256c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64z\"],\"face-grin-tongue-squint\":[512,512,[128541,\"grin-tongue-squint\"],\"f58a\",\"M464 256c0-114.9-93.1-208-208-208S48 141.1 48 256c0 81.7 47.1 152.4 115.7 186.4c-2.4-8.4-3.7-17.3-3.7-26.4V392.7c-24-17.5-43.1-41.4-54.8-69.2c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19c12.3-3.8 24.3 6.9 19.3 18.7c-11.8 28-31.1 52-55.4 69.6V416c0 9.2-1.3 18-3.7 26.4C416.9 408.4 464 337.7 464 256zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm116-98.9c0-9 9.6-14.7 17.5-10.5l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6zm262.5-10.5c7.9-4.2 17.5 1.5 17.5 10.5c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9zM320 416V378.6c0-14.7-11.9-26.6-26.6-26.6h-2c-11.3 0-21.1 7.9-23.6 18.9c-2.8 12.6-20.8 12.6-23.6 0c-2.5-11.1-12.3-18.9-23.6-18.9h-2c-14.7 0-26.6 11.9-26.6 26.6V416c0 35.3 28.7 64 64 64s64-28.7 64-64z\"],\"face-flushed\":[512,512,[128563,\"flushed\"],\"f579\",\"M464 256A208 208 0 1 1 48 256a208 208 0 1 1 416 0zM256 0a256 256 0 1 0 0 512A256 256 0 1 0 256 0zM160.4 248a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm216-24a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zM192 336c-13.3 0-24 10.7-24 24s10.7 24 24 24H320c13.3 0 24-10.7 24-24s-10.7-24-24-24H192zM160 176a48 48 0 1 1 0 96 48 48 0 1 1 0-96zm0 128a80 80 0 1 0 0-160 80 80 0 1 0 0 160zm144-80a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm128 0a80 80 0 1 0 -160 0 80 80 0 1 0 160 0z\"],\"square-caret-right\":[448,512,[\"caret-square-right\"],\"f152\",\"M400 96c0-8.8-7.2-16-16-16L64 80c-8.8 0-16 7.2-16 16l0 320c0 8.8 7.2 16 16 16l320 0c8.8 0 16-7.2 16-16l0-320zM384 32c35.3 0 64 28.7 64 64l0 320c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64L0 96C0 60.7 28.7 32 64 32l320 0zM320 256c0 6.7-2.8 13-7.7 17.6l-112 104c-7 6.5-17.2 8.2-25.9 4.4s-14.4-12.5-14.4-22l0-208c0-9.5 5.7-18.2 14.4-22s18.9-2.1 25.9 4.4l112 104c4.9 4.5 7.7 10.9 7.7 17.6z\"],\"square-minus\":[448,512,[61767,\"minus-square\"],\"f146\",\"M64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V96c0-8.8-7.2-16-16-16H64zM0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM152 232H296c13.3 0 24 10.7 24 24s-10.7 24-24 24H152c-13.3 0-24-10.7-24-24s10.7-24 24-24z\"],compass:[512,512,[129517],\"f14e\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm306.7 69.1L162.4 380.6c-19.4 7.5-38.5-11.6-31-31l55.5-144.3c3.3-8.5 9.9-15.1 18.4-18.4l144.3-55.5c19.4-7.5 38.5 11.6 31 31L325.1 306.7c-3.2 8.5-9.9 15.1-18.4 18.4zM288 256a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z\"],\"square-caret-down\":[448,512,[\"caret-square-down\"],\"f150\",\"M384 432c8.8 0 16-7.2 16-16l0-320c0-8.8-7.2-16-16-16L64 80c-8.8 0-16 7.2-16 16l0 320c0 8.8 7.2 16 16 16l320 0zm64-16c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64L0 96C0 60.7 28.7 32 64 32l320 0c35.3 0 64 28.7 64 64l0 320zM224 352c-6.7 0-13-2.8-17.6-7.7l-104-112c-6.5-7-8.2-17.2-4.4-25.9s12.5-14.4 22-14.4l208 0c9.5 0 18.2 5.7 22 14.4s2.1 18.9-4.4 25.9l-104 112c-4.5 4.9-10.9 7.7-17.6 7.7z\"],\"face-kiss-beam\":[512,512,[128537,\"kiss-beam\"],\"f597\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm304.7 41.7c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4c-2.7 1.5-5.7 3-8.7 4.3c3.1 1.3 6 2.7 8.7 4.3c6.6 3.7 12.5 8.2 16.7 13.4c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4C274.7 427.1 257.4 432 240 432c-3.6 0-6.8-2.5-7.7-6s.6-7.2 3.8-9l0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1c-.4-.2-.7-.4-.9-.5l-.2-.1 0 0 0 0 0 0c-2.5-1.4-4.1-4.1-4.1-7s1.6-5.6 4.1-7l0 0 0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1c-.4-.2-.7-.4-.9-.5l-.2-.1 0 0 0 0 0 0c-3.2-1.8-4.7-5.5-3.8-9s4.1-6 7.7-6c17.4 0 34.7 4.9 47.9 12.3c6.6 3.7 12.5 8.2 16.7 13.4zm-87.1-68.9l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z\"],lightbulb:[384,512,[128161],\"f0eb\",\"M297.2 248.9C311.6 228.3 320 203.2 320 176c0-70.7-57.3-128-128-128S64 105.3 64 176c0 27.2 8.4 52.3 22.8 72.9c3.7 5.3 8.1 11.3 12.8 17.7l0 0c12.9 17.7 28.3 38.9 39.8 59.8c10.4 19 15.7 38.8 18.3 57.5H109c-2.2-12-5.9-23.7-11.8-34.5c-9.9-18-22.2-34.9-34.5-51.8l0 0 0 0c-5.2-7.1-10.4-14.2-15.4-21.4C27.6 247.9 16 213.3 16 176C16 78.8 94.8 0 192 0s176 78.8 176 176c0 37.3-11.6 71.9-31.4 100.3c-5 7.2-10.2 14.3-15.4 21.4l0 0 0 0c-12.3 16.8-24.6 33.7-34.5 51.8c-5.9 10.8-9.6 22.5-11.8 34.5H226.4c2.6-18.7 7.9-38.6 18.3-57.5c11.5-20.9 26.9-42.1 39.8-59.8l0 0 0 0 0 0c4.7-6.4 9-12.4 12.7-17.7zM192 128c-26.5 0-48 21.5-48 48c0 8.8-7.2 16-16 16s-16-7.2-16-16c0-44.2 35.8-80 80-80c8.8 0 16 7.2 16 16s-7.2 16-16 16zm0 384c-44.2 0-80-35.8-80-80V416H272v16c0 44.2-35.8 80-80 80z\"],flag:[448,512,[127988,61725],\"f024\",\"M48 24C48 10.7 37.3 0 24 0S0 10.7 0 24V64 350.5 400v88c0 13.3 10.7 24 24 24s24-10.7 24-24V388l80.3-20.1c41.1-10.3 84.6-5.5 122.5 13.4c44.2 22.1 95.5 24.8 141.7 7.4l34.7-13c12.5-4.7 20.8-16.6 20.8-30V66.1c0-23-24.2-38-44.8-27.7l-9.6 4.8c-46.3 23.2-100.8 23.2-147.1 0c-35.1-17.6-75.4-22-113.5-12.5L48 52V24zm0 77.5l96.6-24.2c27-6.7 55.5-3.6 80.4 8.8c54.9 27.4 118.7 29.7 175 6.8V334.7l-24.4 9.1c-33.7 12.6-71.2 10.7-103.4-5.4c-48.2-24.1-103.3-30.1-155.6-17.1L48 338.5v-237z\"],\"square-check\":[448,512,[9745,9989,61510,\"check-square\"],\"f14a\",\"M64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V96c0-8.8-7.2-16-16-16H64zM0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM337 209L209 337c-9.4 9.4-24.6 9.4-33.9 0l-64-64c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47L303 175c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z\"],\"circle-dot\":[512,512,[128280,\"dot-circle\"],\"f192\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm256-96a96 96 0 1 1 0 192 96 96 0 1 1 0-192z\"],\"face-dizzy\":[512,512,[\"dizzy\"],\"f567\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm256 32a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM103 135c9.4-9.4 24.6-9.4 33.9 0l23 23 23-23c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-23 23 23 23c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-23-23-23 23c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l23-23-23-23c-9.4-9.4-9.4-24.6 0-33.9zm192 0c9.4-9.4 24.6-9.4 33.9 0l23 23 23-23c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-23 23 23 23c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-23-23-23 23c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l23-23-23-23c-9.4-9.4-9.4-24.6 0-33.9z\"],futbol:[512,512,[9917,\"futbol-ball\",\"soccer-ball\"],\"f1e3\",\"M435.4 361.3l-89.7-6c-5.2-.3-10.3 1.1-14.5 4.2s-7.2 7.4-8.4 12.5l-22 87.2c-14.4 3.2-29.4 4.8-44.8 4.8s-30.3-1.7-44.8-4.8l-22-87.2c-1.3-5-4.3-9.4-8.4-12.5s-9.3-4.5-14.5-4.2l-89.7 6C61.7 335.9 51.9 307 49 276.2L125 228.3c4.4-2.8 7.6-7 9.2-11.9s1.4-10.2-.5-15L100.4 118c19.9-22.4 44.6-40.5 72.4-52.7l69.1 57.6c4 3.3 9 5.1 14.1 5.1s10.2-1.8 14.1-5.1l69.1-57.6c27.8 12.2 52.5 30.3 72.4 52.7l-33.4 83.4c-1.9 4.8-2.1 10.1-.5 15s4.9 9.1 9.2 11.9L463 276.2c-3 30.8-12.7 59.7-27.6 85.1zM256 48l.9 0h-1.8l.9 0zM56.7 196.2c.9-3 1.9-6.1 2.9-9.1l-2.9 9.1zM132 423l3.8 2.7c-1.3-.9-2.5-1.8-3.8-2.7zm248.1-.1c-1.3 1-2.7 2-4 2.9l4-2.9zm75.2-226.6l-3-9.2c1.1 3 2.1 6.1 3 9.2zM256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm14.1-325.7c-8.4-6.1-19.8-6.1-28.2 0L194 221c-8.4 6.1-11.9 16.9-8.7 26.8l18.3 56.3c3.2 9.9 12.4 16.6 22.8 16.6h59.2c10.4 0 19.6-6.7 22.8-16.6l18.3-56.3c3.2-9.9-.3-20.7-8.7-26.8l-47.9-34.8z\"],\"pen-to-square\":[512,512,[\"edit\"],\"f044\",\"M441 58.9L453.1 71c9.4 9.4 9.4 24.6 0 33.9L424 134.1 377.9 88 407 58.9c9.4-9.4 24.6-9.4 33.9 0zM209.8 256.2L344 121.9 390.1 168 255.8 302.2c-2.9 2.9-6.5 5-10.4 6.1l-58.5 16.7 16.7-58.5c1.1-3.9 3.2-7.5 6.1-10.4zM373.1 25L175.8 222.2c-8.7 8.7-15 19.4-18.3 31.1l-28.6 100c-2.4 8.4-.1 17.4 6.1 23.6s15.2 8.5 23.6 6.1l100-28.6c11.8-3.4 22.5-9.7 31.1-18.3L487 138.9c28.1-28.1 28.1-73.7 0-101.8L474.9 25C446.8-3.1 401.2-3.1 373.1 25zM88 64C39.4 64 0 103.4 0 152V424c0 48.6 39.4 88 88 88H360c48.6 0 88-39.4 88-88V312c0-13.3-10.7-24-24-24s-24 10.7-24 24V424c0 22.1-17.9 40-40 40H88c-22.1 0-40-17.9-40-40V152c0-22.1 17.9-40 40-40H200c13.3 0 24-10.7 24-24s-10.7-24-24-24H88z\"],\"hourglass-half\":[384,512,[\"hourglass-2\"],\"f252\",\"M0 24C0 10.7 10.7 0 24 0H360c13.3 0 24 10.7 24 24s-10.7 24-24 24h-8V67c0 40.3-16 79-44.5 107.5L225.9 256l81.5 81.5C336 366 352 404.7 352 445v19h8c13.3 0 24 10.7 24 24s-10.7 24-24 24H24c-13.3 0-24-10.7-24-24s10.7-24 24-24h8V445c0-40.3 16-79 44.5-107.5L158.1 256 76.5 174.5C48 146 32 107.3 32 67V48H24C10.7 48 0 37.3 0 24zM110.5 371.5c-3.9 3.9-7.5 8.1-10.7 12.5H284.2c-3.2-4.4-6.8-8.6-10.7-12.5L192 289.9l-81.5 81.5zM284.2 128C297 110.4 304 89 304 67V48H80V67c0 22.1 7 43.4 19.8 61H284.2z\"],\"eye-slash\":[640,512,[],\"f070\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L525.6 386.7c39.6-40.6 66.4-86.1 79.9-118.4c3.3-7.9 3.3-16.7 0-24.6c-14.9-35.7-46.2-87.7-93-131.1C465.5 68.8 400.8 32 320 32c-68.2 0-125 26.3-169.3 60.8L38.8 5.1zm151 118.3C226 97.7 269.5 80 320 80c65.2 0 118.8 29.6 159.9 67.7C518.4 183.5 545 226 558.6 256c-12.6 28-36.6 66.8-70.9 100.9l-53.8-42.2c9.1-17.6 14.2-37.5 14.2-58.7c0-70.7-57.3-128-128-128c-32.2 0-61.7 11.9-84.2 31.5l-46.1-36.1zM394.9 284.2l-81.5-63.9c4.2-8.5 6.6-18.2 6.6-28.3c0-5.5-.7-10.9-2-16c.7 0 1.3 0 2 0c44.2 0 80 35.8 80 80c0 9.9-1.8 19.4-5.1 28.2zm51.3 163.3l-41.9-33C378.8 425.4 350.7 432 320 432c-65.2 0-118.8-29.6-159.9-67.7C121.6 328.5 95 286 81.4 256c8.3-18.4 21.5-41.5 39.4-64.8L83.1 161.5C60.3 191.2 44 220.8 34.5 243.7c-3.3 7.9-3.3 16.7 0 24.6c14.9 35.7 46.2 87.7 93 131.1C174.5 443.2 239.2 480 320 480c47.8 0 89.9-12.9 126.2-32.5zm-88-69.3L302 334c-23.5-5.4-43.1-21.2-53.7-42.3l-56.1-44.2c-.2 2.8-.3 5.6-.3 8.5c0 70.7 57.3 128 128 128c13.3 0 26.1-2 38.2-5.8z\"],hand:[512,512,[129306,9995,\"hand-paper\"],\"f256\",\"M256 0c-25.3 0-47.2 14.7-57.6 36c-7-2.6-14.5-4-22.4-4c-35.3 0-64 28.7-64 64V261.5l-2.7-2.7c-25-25-65.5-25-90.5 0s-25 65.5 0 90.5L106.5 437c48 48 113.1 75 181 75H296h8c1.5 0 3-.1 4.5-.4c91.7-6.2 165-79.4 171.1-171.1c.3-1.5 .4-3 .4-4.5V160c0-35.3-28.7-64-64-64c-5.5 0-10.9 .7-16 2V96c0-35.3-28.7-64-64-64c-7.9 0-15.4 1.4-22.4 4C303.2 14.7 281.3 0 256 0zM240 96.1c0 0 0-.1 0-.1V64c0-8.8 7.2-16 16-16s16 7.2 16 16V95.9c0 0 0 .1 0 .1V232c0 13.3 10.7 24 24 24s24-10.7 24-24V96c0 0 0 0 0-.1c0-8.8 7.2-16 16-16s16 7.2 16 16v55.9c0 0 0 .1 0 .1v80c0 13.3 10.7 24 24 24s24-10.7 24-24V160.1c0 0 0-.1 0-.1c0-8.8 7.2-16 16-16s16 7.2 16 16V332.9c-.1 .6-.1 1.3-.2 1.9c-3.4 69.7-59.3 125.6-129 129c-.6 0-1.3 .1-1.9 .2H296h-8.5c-55.2 0-108.1-21.9-147.1-60.9L52.7 315.3c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L119 336.4c6.9 6.9 17.2 8.9 26.2 5.2s14.8-12.5 14.8-22.2V96c0-8.8 7.2-16 16-16c8.8 0 16 7.1 16 15.9V232c0 13.3 10.7 24 24 24s24-10.7 24-24V96.1z\"],\"hand-spock\":[576,512,[128406],\"f259\",\"M170.2 80.8C161 47 180.8 12 214.6 2.4c34-9.6 69.4 10.2 79 44.2l30.3 107.1L337.1 84c6.6-34.7 40.1-57.5 74.8-50.9c31.4 6 53 33.9 52 64.9c10-2.6 20.8-2.8 31.5-.1c34.3 8.6 55.1 43.3 46.6 77.6L486.7 397.2C469.8 464.7 409.2 512 339.6 512c-11.2 0-22.5 0-33.7 0c-56.9 0-112.2-19-157.2-53.9l-92-71.6c-27.9-21.7-32.9-61.9-11.2-89.8s61.9-32.9 89.8-11.2l17 13.2L100.5 167.5c-13-32.9 3.2-70.1 36-83c11.1-4.4 22.7-5.4 33.7-3.7zm77.1-21.2c-2.4-8.5-11.2-13.4-19.7-11s-13.4 11.2-11 19.7l54.8 182.4c3.5 12.3-3.3 25.2-15.4 29.3s-25.3-2-30-13.9L174.9 138.1c-3.2-8.2-12.5-12.3-20.8-9s-12.3 12.5-9 20.8l73.3 185.6c12 30.3-23.7 57-49.4 37l-63.1-49.1c-7-5.4-17-4.2-22.5 2.8s-4.2 17 2.8 22.5l92 71.6c36.5 28.4 81.4 43.8 127.7 43.8c11.2 0 22.5 0 33.7 0c47.5 0 89-32.4 100.5-78.5l55.4-221.6c2.1-8.6-3.1-17.3-11.6-19.4s-17.3 3.1-19.4 11.6l-26 104C435.6 271.8 425 280 413 280c-16.5 0-28.9-15-25.8-31.2L415.7 99c1.7-8.7-4-17.1-12.7-18.7s-17.1 4-18.7 12.7L352.5 260c-2.2 11.6-12.4 20-24.2 20c-11 0-20.7-7.3-23.7-17.9L247.4 59.6z\"],\"face-kiss\":[512,512,[128535,\"kiss\"],\"f596\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm304.7 25.7c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4c-2.7 1.5-5.7 3-8.7 4.3c3.1 1.3 6 2.7 8.7 4.3c6.6 3.7 12.5 8.2 16.7 13.4c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4C274.7 411.1 257.4 416 240 416c-3.6 0-6.8-2.5-7.7-6s.6-7.2 3.8-9l0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1l-.8-.5-.1-.1-.2-.1 0 0 0 0 0 0c-2.5-1.4-4.1-4.1-4.1-7s1.6-5.6 4.1-7l0 0 0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1c-.4-.2-.7-.4-.9-.5l-.2-.1 0 0 0 0 0 0c-3.2-1.8-4.7-5.5-3.8-9s4.1-6 7.7-6c17.4 0 34.7 4.9 47.9 12.3c6.6 3.7 12.5 8.2 16.7 13.4zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"face-grin-tongue\":[512,512,[128539,\"grin-tongue\"],\"f589\",\"M464 256c0-114.9-93.1-208-208-208S48 141.1 48 256c0 81.7 47.1 152.4 115.7 186.4c-2.4-8.4-3.7-17.3-3.7-26.4V363.6c-8.9-8-16.7-17.1-23.1-27.1c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5c18.7-4.4 35.9 12 25.5 28.1c-6.4 9.9-14.2 19-23 27V416c0 9.2-1.3 18-3.7 26.4C416.9 408.4 464 337.7 464 256zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm176.4-80a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM320 416V378.6c0-14.7-11.9-26.6-26.6-26.6h-2c-11.3 0-21.1 7.9-23.6 18.9c-2.8 12.6-20.8 12.6-23.6 0c-2.5-11.1-12.3-18.9-23.6-18.9h-2c-14.7 0-26.6 11.9-26.6 26.6V416c0 35.3 28.7 64 64 64s64-28.7 64-64z\"],\"chess-bishop\":[320,512,[9821],\"f43a\",\"M104 0C90.7 0 80 10.7 80 24c0 11.2 7.6 20.6 18 23.2c-7.8 8-16.1 17-24.4 27C38.2 116.7 0 178.8 0 250.9c0 44.8 24.6 72.2 48 87.8V352H96V325c0-9-5-17.2-13-21.3c-18-9.3-35-24.7-35-52.7c0-55.5 29.8-106.8 62.4-145.9c16-19.2 32.1-34.8 44.2-45.5c1.9-1.7 3.7-3.2 5.3-4.6c1.7 1.4 3.4 3 5.3 4.6c12.1 10.7 28.2 26.3 44.2 45.5c5.3 6.3 10.5 13 15.5 20L159 191c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l57.8-57.8c12.8 25.9 21.2 54.3 21.2 83.8c0 28-17 43.4-35 52.7c-8 4.1-13 12.3-13 21.3v27h48V338.7c23.4-15.6 48-42.9 48-87.8c0-72.1-38.2-134.2-73.6-176.7c-8.3-9.9-16.6-19-24.4-27c10.3-2.7 18-12.1 18-23.2c0-13.3-10.7-24-24-24H160 104zM52.7 464l16.6-32H250.8l16.6 32H52.7zm207.9-80H59.5c-12 0-22.9 6.7-28.4 17.3L4.6 452.5c-3 5.8-4.6 12.2-4.6 18.7C0 493.8 18.2 512 40.8 512H279.2c22.5 0 40.8-18.2 40.8-40.8c0-6.5-1.6-12.9-4.6-18.7l-26.5-51.2c-5.5-10.6-16.5-17.3-28.4-17.3z\"],\"face-grin-wink\":[512,512,[\"grin-wink\"],\"f58c\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm349.5 52.4c18.7-4.4 35.9 12 25.5 28.1C350.4 374.6 306.3 400 255.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm165.8 21.7c-7.6 8.1-20.2 8.5-28.3 .9s-8.5-20.2-.9-28.3c14.5-15.5 35.2-22.3 54.6-22.3s40.1 6.8 54.6 22.3c7.6 8.1 7.1 20.7-.9 28.3s-20.7 7.1-28.3-.9c-5.5-5.8-14.8-9.7-25.4-9.7s-19.9 3.8-25.4 9.7z\"],\"face-grin-wide\":[512,512,[128515,\"grin-alt\"],\"f581\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm349.5 52.4c18.7-4.4 35.9 12 25.5 28.1C350.4 374.6 306.3 400 255.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5zM224 192c0 35.3-14.3 64-32 64s-32-28.7-32-64s14.3-64 32-64s32 28.7 32 64zm96 64c-17.7 0-32-28.7-32-64s14.3-64 32-64s32 28.7 32 64s-14.3 64-32 64z\"],\"face-frown-open\":[512,512,[128550,\"frown-open\"],\"f57a\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM182.4 382.5c-12.4 5.2-26.5-4.1-21.1-16.4c16-36.6 52.4-62.1 94.8-62.1s78.8 25.6 94.8 62.1c5.4 12.3-8.7 21.6-21.1 16.4c-22.4-9.5-47.4-14.8-73.7-14.8s-51.3 5.3-73.7 14.8zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"hand-point-up\":[384,512,[9757],\"f0a6\",\"M64 64V241.6c5.2-1 10.5-1.6 16-1.6H96V208 64c0-8.8-7.2-16-16-16s-16 7.2-16 16zM80 288c-17.7 0-32 14.3-32 32c0 0 0 0 0 0v24c0 66.3 53.7 120 120 120h48c52.5 0 97.1-33.7 113.4-80.7c-3.1 .5-6.2 .7-9.4 .7c-20 0-37.9-9.2-49.7-23.6c-9 4.9-19.4 7.6-30.3 7.6c-15.1 0-29-5.3-40-14c-11 8.8-24.9 14-40 14H120c-13.3 0-24-10.7-24-24s10.7-24 24-24h40c8.8 0 16-7.2 16-16s-7.2-16-16-16H120 80zM0 320s0 0 0 0c0-18 6-34.6 16-48V64C16 28.7 44.7 0 80 0s64 28.7 64 64v82c5.1-1.3 10.5-2 16-2c25.3 0 47.2 14.7 57.6 36c7-2.6 14.5-4 22.4-4c20 0 37.9 9.2 49.7 23.6c9-4.9 19.4-7.6 30.3-7.6c35.3 0 64 28.7 64 64v64 24c0 92.8-75.2 168-168 168H168C75.2 512 0 436.8 0 344V320zm336-64c0-8.8-7.2-16-16-16s-16 7.2-16 16v48 16c0 8.8 7.2 16 16 16s16-7.2 16-16V256zM160 240c5.5 0 10.9 .7 16 2v-2V208c0-8.8-7.2-16-16-16s-16 7.2-16 16v32h16zm64 24v40c0 8.8 7.2 16 16 16s16-7.2 16-16V256 240c0-8.8-7.2-16-16-16s-16 7.2-16 16v24z\"],bookmark:[384,512,[128278,61591],\"f02e\",\"M0 48C0 21.5 21.5 0 48 0l0 48V441.4l130.1-92.9c8.3-6 19.6-6 27.9 0L336 441.4V48H48V0H336c26.5 0 48 21.5 48 48V488c0 9-5 17.2-13 21.3s-17.6 3.4-24.9-1.8L192 397.5 37.9 507.5c-7.3 5.2-16.9 5.9-24.9 1.8S0 497 0 488V48z\"],\"hand-point-down\":[384,512,[],\"f0a7\",\"M64 448l0-177.6c5.2 1 10.5 1.6 16 1.6l16 0 0 32 0 144c0 8.8-7.2 16-16 16s-16-7.2-16-16zM80 224c-17.7 0-32-14.3-32-32c0 0 0 0 0 0l0-24c0-66.3 53.7-120 120-120l48 0c52.5 0 97.1 33.7 113.4 80.7c-3.1-.5-6.2-.7-9.4-.7c-20 0-37.9 9.2-49.7 23.6c-9-4.9-19.4-7.6-30.3-7.6c-15.1 0-29 5.3-40 14c-11-8.8-24.9-14-40-14l-40 0c-13.3 0-24 10.7-24 24s10.7 24 24 24l40 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-40 0-40 0zM0 192s0 0 0 0c0 18 6 34.6 16 48l0 208c0 35.3 28.7 64 64 64s64-28.7 64-64l0-82c5.1 1.3 10.5 2 16 2c25.3 0 47.2-14.7 57.6-36c7 2.6 14.5 4 22.4 4c20 0 37.9-9.2 49.7-23.6c9 4.9 19.4 7.6 30.3 7.6c35.3 0 64-28.7 64-64l0-64 0-24C384 75.2 308.8 0 216 0L168 0C75.2 0 0 75.2 0 168l0 24zm336 64c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-48 0-16c0-8.8 7.2-16 16-16s16 7.2 16 16l0 64zM160 272c5.5 0 10.9-.7 16-2l0 2 0 32c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-32 16 0zm64-24l0-40c0-8.8 7.2-16 16-16s16 7.2 16 16l0 48 0 16c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-24z\"],folder:[512,512,[128193,128447,61716,\"folder-blank\"],\"f07b\",\"M0 96C0 60.7 28.7 32 64 32H196.1c19.1 0 37.4 7.6 50.9 21.1L289.9 96H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H448c8.8 0 16-7.2 16-16V160c0-8.8-7.2-16-16-16H286.6c-10.6 0-20.8-4.2-28.3-11.7L213.1 87c-4.5-4.5-10.6-7-17-7H64z\"],user:[448,512,[128100,62144],\"f007\",\"M304 128a80 80 0 1 0 -160 0 80 80 0 1 0 160 0zM96 128a128 128 0 1 1 256 0A128 128 0 1 1 96 128zM49.3 464H398.7c-8.9-63.3-63.3-112-129-112H178.3c-65.7 0-120.1 48.7-129 112zM0 482.3C0 383.8 79.8 304 178.3 304h91.4C368.2 304 448 383.8 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3z\"],\"square-caret-left\":[448,512,[\"caret-square-left\"],\"f191\",\"M48 416c0 8.8 7.2 16 16 16l320 0c8.8 0 16-7.2 16-16l0-320c0-8.8-7.2-16-16-16L64 80c-8.8 0-16 7.2-16 16l0 320zm16 64c-35.3 0-64-28.7-64-64L0 96C0 60.7 28.7 32 64 32l320 0c35.3 0 64 28.7 64 64l0 320c0 35.3-28.7 64-64 64L64 480zm64-224c0-6.7 2.8-13 7.7-17.6l112-104c7-6.5 17.2-8.2 25.9-4.4s14.4 12.5 14.4 22l0 208c0 9.5-5.7 18.2-14.4 22s-18.9 2.1-25.9-4.4l-112-104c-4.9-4.5-7.7-10.9-7.7-17.6z\"],star:[576,512,[11088,61446],\"f005\",\"M287.9 0c9.2 0 17.6 5.2 21.6 13.5l68.6 141.3 153.2 22.6c9 1.3 16.5 7.6 19.3 16.3s.5 18.1-5.9 24.5L433.6 328.4l26.2 155.6c1.5 9-2.2 18.1-9.7 23.5s-17.3 6-25.3 1.7l-137-73.2L151 509.1c-8.1 4.3-17.9 3.7-25.3-1.7s-11.2-14.5-9.7-23.5l26.2-155.6L31.1 218.2c-6.5-6.4-8.7-15.9-5.9-24.5s10.3-14.9 19.3-16.3l153.2-22.6L266.3 13.5C270.4 5.2 278.7 0 287.9 0zm0 79L235.4 187.2c-3.5 7.1-10.2 12.1-18.1 13.3L99 217.9 184.9 303c5.5 5.5 8.1 13.3 6.8 21L171.4 443.7l105.2-56.2c7.1-3.8 15.6-3.8 22.6 0l105.2 56.2L384.2 324.1c-1.3-7.7 1.2-15.5 6.8-21l85.9-85.1L358.6 200.5c-7.8-1.2-14.6-6.1-18.1-13.3L287.9 79z\"],\"chess-knight\":[448,512,[9822],\"f441\",\"M226.6 48H117.3l17.1 12.8c6 4.5 9.6 11.6 9.6 19.2s-3.6 14.7-9.6 19.2l-6.5 4.9c-10 7.5-16 19.3-16 31.9l-.3 91c0 10.2 4.9 19.9 13.2 25.8l1.9 1.3c9.9 7.1 23.3 7 33.2-.1l49.9-36.3c10.7-7.8 25.7-5.4 33.5 5.3s5.4 25.7-5.3 33.5l-49.9 36.3-53.8 39.1c-7.3 5.3-13 12.2-16.9 20.1H66.8c5.3-22.1 17.8-41.9 35.9-56.3c-1.3-.8-2.6-1.7-3.8-2.6L97 291.8c-21-15-33.4-39.2-33.3-65l.3-91c.1-19.8 6.7-38.7 18.6-53.9l-.4-.3C70.7 73 64 59.6 64 45.3C64 20.3 84.3 0 109.3 0H226.6C331.2 0 416 84.8 416 189.4c0 11.1-1 22.2-2.9 33.2L390.1 352H341.3l24.5-137.8c1.5-8.2 2.2-16.5 2.2-24.8C368 111.3 304.7 48 226.6 48zM85.2 432L68.7 464H379.3l-16.6-32H85.2zm315.7-30.7l26.5 51.2c3 5.8 4.6 12.2 4.6 18.7c0 22.5-18.2 40.8-40.8 40.8H56.8C34.2 512 16 493.8 16 471.2c0-6.5 1.6-12.9 4.6-18.7l26.5-51.2C52.5 390.7 63.5 384 75.5 384h297c12 0 22.9 6.7 28.4 17.3zM172 128a20 20 0 1 1 0 40 20 20 0 1 1 0-40z\"],\"face-laugh-squint\":[512,512,[\"laugh-squint\"],\"f59b\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm130.7 57.9c-4.2-13.6 7.1-25.9 21.3-25.9H364.5c14.2 0 25.5 12.4 21.3 25.9C369 368.4 318.2 408 258.2 408s-110.8-39.6-127.5-94.1zm2.8-183.3l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6c0-9 9.6-14.7 17.5-10.5zM396 141.1c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9c7.9-4.2 17.5 1.5 17.5 10.5z\"],\"face-laugh\":[512,512,[\"laugh\"],\"f599\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm130.7 57.9c-4.2-13.6 7.1-25.9 21.3-25.9H364.5c14.2 0 25.5 12.4 21.3 25.9C369 368.4 318.2 408 258.2 408s-110.8-39.6-127.5-94.1zM144.4 192a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"folder-open\":[576,512,[128194,128449,61717],\"f07c\",\"M384 480h48c11.4 0 21.9-6 27.6-15.9l112-192c5.8-9.9 5.8-22.1 .1-32.1S555.5 224 544 224H144c-11.4 0-21.9 6-27.6 15.9L48 357.1V96c0-8.8 7.2-16 16-16H181.5c4.2 0 8.3 1.7 11.3 4.7l26.5 26.5c21 21 49.5 32.8 79.2 32.8H416c8.8 0 16 7.2 16 16v32h48V160c0-35.3-28.7-64-64-64H298.5c-17 0-33.3-6.7-45.3-18.7L226.7 50.7c-12-12-28.3-18.7-45.3-18.7H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H87.7 384z\"],clipboard:[384,512,[128203],\"f328\",\"M280 64h40c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128C0 92.7 28.7 64 64 64h40 9.6C121 27.5 153.3 0 192 0s71 27.5 78.4 64H280zM64 112c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16H320c8.8 0 16-7.2 16-16V128c0-8.8-7.2-16-16-16H304v24c0 13.3-10.7 24-24 24H192 104c-13.3 0-24-10.7-24-24V112H64zm128-8a24 24 0 1 0 0-48 24 24 0 1 0 0 48z\"],\"chess-queen\":[512,512,[9819],\"f445\",\"M256 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-95.2-8c-18.1 0-31.3 12.8-35.6 26.9c-8 26.2-32.4 45.2-61.2 45.2c-10 0-19.4-2.3-27.7-6.3c-7.6-3.7-16.7-3.3-24 1.2C.7 162.1-3.1 177.1 3.7 188.9L97.6 352H153l-83-144.1c40.5-2.2 75.3-25.9 93.1-59.8c22 26.8 55.4 43.9 92.8 43.9s70.8-17.1 92.8-43.9c17.8 34 52.6 57.7 93.1 59.8L359 352h55.4l93.9-163.1c6.8-11.7 3-26.7-8.6-33.8c-7.3-4.5-16.4-4.9-24-1.2c-8.4 4-17.7 6.3-27.7 6.3c-28.8 0-53.2-19-61.2-45.2C382.5 100.8 369.3 88 351.2 88c-14.5 0-26.3 8.5-32.4 19.3c-12.4 22-35.9 36.7-62.8 36.7s-50.4-14.8-62.8-36.7C187.1 96.5 175.4 88 160.8 88zM133.2 432H378.8l16.6 32H116.7l16.6-32zm283.7-30.7c-5.5-10.6-16.5-17.3-28.4-17.3h-265c-12 0-22.9 6.7-28.4 17.3L68.6 452.5c-3 5.8-4.6 12.2-4.6 18.7c0 22.5 18.2 40.8 40.8 40.8H407.2c22.5 0 40.8-18.2 40.8-40.8c0-6.5-1.6-12.9-4.6-18.7l-26.5-51.2z\"],\"hand-back-fist\":[448,512,[\"hand-rock\"],\"f255\",\"M144 64c0-8.8 7.2-16 16-16s16 7.2 16 16c0 9.1 5.1 17.4 13.3 21.5s17.9 3.2 25.1-2.3c2.7-2 6-3.2 9.6-3.2c8.8 0 16 7.2 16 16c0 9.1 5.1 17.4 13.3 21.5s17.9 3.2 25.1-2.3c2.7-2 6-3.2 9.6-3.2c8.8 0 16 7.2 16 16c0 9.1 5.1 17.4 13.3 21.5s17.9 3.2 25.1-2.3c2.7-2 6-3.2 9.6-3.2c8.8 0 16 7.2 16 16V264c0 31.3-20 58-48 67.9c-9.6 3.4-16 12.5-16 22.6V488c0 13.3 10.7 24 24 24s24-10.7 24-24V370.2c38-20.1 64-60.1 64-106.2V160c0-35.3-28.7-64-64-64c-2.8 0-5.6 .2-8.3 .5C332.8 77.1 311.9 64 288 64c-2.8 0-5.6 .2-8.3 .5C268.8 45.1 247.9 32 224 32c-2.8 0-5.6 .2-8.3 .5C204.8 13.1 183.9 0 160 0C124.7 0 96 28.7 96 64v64.3c-11.7 7.4-22.5 16.4-32 26.9l17.8 16.1L64 155.2l-9.4 10.5C40 181.8 32 202.8 32 224.6v12.8c0 49.6 24.2 96.1 64.8 124.5l13.8-19.7L96.8 361.9l8.9 6.2c6.9 4.8 14.4 8.6 22.3 11.3V488c0 13.3 10.7 24 24 24s24-10.7 24-24V359.9c0-12.6-9.8-23.1-22.4-23.9c-7.3-.5-14.3-2.9-20.3-7.1l-13.1 18.7 13.1-18.7-8.9-6.2C96.6 303.1 80 271.3 80 237.4V224.6c0-9.9 3.7-19.4 10.3-26.8l9.4-10.5c3.8-4.2 7.9-8.1 12.3-11.6V208c0 8.8 7.2 16 16 16s16-7.2 16-16V142.3 128 64z\"],\"square-caret-up\":[448,512,[\"caret-square-up\"],\"f151\",\"M64 80c-8.8 0-16 7.2-16 16l0 320c0 8.8 7.2 16 16 16l320 0c8.8 0 16-7.2 16-16l0-320c0-8.8-7.2-16-16-16L64 80zM0 96C0 60.7 28.7 32 64 32l320 0c35.3 0 64 28.7 64 64l0 320c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64L0 96zm224 64c6.7 0 13 2.8 17.6 7.7l104 112c6.5 7 8.2 17.2 4.4 25.9s-12.5 14.4-22 14.4l-208 0c-9.5 0-18.2-5.7-22-14.4s-2.1-18.9 4.4-25.9l104-112c4.5-4.9 10.9-7.7 17.6-7.7z\"],\"chart-bar\":[512,512,[\"bar-chart\"],\"f080\",\"M24 32c13.3 0 24 10.7 24 24V408c0 13.3 10.7 24 24 24H488c13.3 0 24 10.7 24 24s-10.7 24-24 24H72c-39.8 0-72-32.2-72-72V56C0 42.7 10.7 32 24 32zM128 136c0-13.3 10.7-24 24-24l208 0c13.3 0 24 10.7 24 24s-10.7 24-24 24l-208 0c-13.3 0-24-10.7-24-24zm24 72H296c13.3 0 24 10.7 24 24s-10.7 24-24 24H152c-13.3 0-24-10.7-24-24s10.7-24 24-24zm0 96H424c13.3 0 24 10.7 24 24s-10.7 24-24 24H152c-13.3 0-24-10.7-24-24s10.7-24 24-24z\"],\"window-restore\":[512,512,[],\"f2d2\",\"M432 48H208c-17.7 0-32 14.3-32 32V96H128V80c0-44.2 35.8-80 80-80H432c44.2 0 80 35.8 80 80V304c0 44.2-35.8 80-80 80H416V336h16c17.7 0 32-14.3 32-32V80c0-17.7-14.3-32-32-32zM48 448c0 8.8 7.2 16 16 16H320c8.8 0 16-7.2 16-16V256H48V448zM64 128H320c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V192c0-35.3 28.7-64 64-64z\"],\"square-plus\":[448,512,[61846,\"plus-square\"],\"f0fe\",\"M64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V96c0-8.8-7.2-16-16-16H64zM0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM200 344V280H136c-13.3 0-24-10.7-24-24s10.7-24 24-24h64V168c0-13.3 10.7-24 24-24s24 10.7 24 24v64h64c13.3 0 24 10.7 24 24s-10.7 24-24 24H248v64c0 13.3-10.7 24-24 24s-24-10.7-24-24z\"],image:[512,512,[],\"f03e\",\"M448 80c8.8 0 16 7.2 16 16V415.8l-5-6.5-136-176c-4.5-5.9-11.6-9.3-19-9.3s-14.4 3.4-19 9.3L202 340.7l-30.5-42.7C167 291.7 159.8 288 152 288s-15 3.7-19.5 10.1l-80 112L48 416.3l0-.3V96c0-8.8 7.2-16 16-16H448zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm80 192a48 48 0 1 0 0-96 48 48 0 1 0 0 96z\"],\"folder-closed\":[512,512,[],\"e185\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H289.9L247 53.1C233.5 39.6 215.2 32 196.1 32H64zM48 96c0-8.8 7.2-16 16-16H196.1c6.4 0 12.5 2.5 17 7l45.3 45.3c7.5 7.5 17.7 11.7 28.3 11.7H448c8.8 0 16 7.2 16 16v32H48V96zm0 144H464V416c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V240z\"],lemon:[448,512,[127819],\"f094\",\"M368 80c-3.2 0-6.2 .4-8.9 1.3C340 86.8 313 91.9 284.8 84.6C227.4 69.7 160.2 92 110.1 142.1S37.7 259.4 52.6 316.8c7.3 28.2 2.2 55.2-3.3 74.3c-.8 2.8-1.3 5.8-1.3 8.9c0 17.7 14.3 32 32 32c3.2 0 6.2-.4 8.9-1.3c19.1-5.5 46.1-10.7 74.3-3.3c57.4 14.9 124.6-7.4 174.7-57.5s72.4-117.3 57.5-174.7c-7.3-28.2-2.2-55.2 3.3-74.3c.8-2.8 1.3-5.8 1.3-8.9c0-17.7-14.3-32-32-32zm0-48c44.2 0 80 35.8 80 80c0 7.7-1.1 15.2-3.1 22.3c-4.6 15.8-7.1 32.9-3 48.9c20.1 77.6-10.9 161.5-70 220.7s-143.1 90.2-220.7 70c-16-4.1-33-1.6-48.9 3c-7.1 2-14.6 3.1-22.3 3.1c-44.2 0-80-35.8-80-80c0-7.7 1.1-15.2 3.1-22.3c4.6-15.8 7.1-32.9 3-48.9C-14 251.3 17 167.3 76.2 108.2S219.3 18 296.8 38.1c16 4.1 33 1.6 48.9-3c7.1-2 14.6-3.1 22.3-3.1zM246.7 167c-52 15.2-96.5 59.7-111.7 111.7c-3.7 12.7-17.1 20-29.8 16.3s-20-17.1-16.3-29.8c19.8-67.7 76.6-124.5 144.3-144.3c12.7-3.7 26.1 3.6 29.8 16.3s-3.6 26.1-16.3 29.8z\"],handshake:[640,512,[],\"f2b5\",\"M272.2 64.6l-51.1 51.1c-15.3 4.2-29.5 11.9-41.5 22.5L153 161.9C142.8 171 129.5 176 115.8 176H96V304c20.4 .6 39.8 8.9 54.3 23.4l35.6 35.6 7 7 0 0L219.9 397c6.2 6.2 16.4 6.2 22.6 0c1.7-1.7 3-3.7 3.7-5.8c2.8-7.7 9.3-13.5 17.3-15.3s16.4 .6 22.2 6.5L296.5 393c11.6 11.6 30.4 11.6 41.9 0c5.4-5.4 8.3-12.3 8.6-19.4c.4-8.8 5.6-16.6 13.6-20.4s17.3-3 24.4 2.1c9.4 6.7 22.5 5.8 30.9-2.6c9.4-9.4 9.4-24.6 0-33.9L340.1 243l-35.8 33c-27.3 25.2-69.2 25.6-97 .9c-31.7-28.2-32.4-77.4-1.6-106.5l70.1-66.2C303.2 78.4 339.4 64 377.1 64c36.1 0 71 13.3 97.9 37.2L505.1 128H544h40 40c8.8 0 16 7.2 16 16V352c0 17.7-14.3 32-32 32H576c-11.8 0-22.2-6.4-27.7-16H463.4c-3.4 6.7-7.9 13.1-13.5 18.7c-17.1 17.1-40.8 23.8-63 20.1c-3.6 7.3-8.5 14.1-14.6 20.2c-27.3 27.3-70 30-100.4 8.1c-25.1 20.8-62.5 19.5-86-4.1L159 404l-7-7-35.6-35.6c-5.5-5.5-12.7-8.7-20.4-9.3C96 369.7 81.6 384 64 384H32c-17.7 0-32-14.3-32-32V144c0-8.8 7.2-16 16-16H56 96h19.8c2 0 3.9-.7 5.3-2l26.5-23.6C175.5 77.7 211.4 64 248.7 64H259c4.4 0 8.9 .2 13.2 .6zM544 320V176H496c-5.9 0-11.6-2.2-15.9-6.1l-36.9-32.8c-18.2-16.2-41.7-25.1-66.1-25.1c-25.4 0-49.8 9.7-68.3 27.1l-70.1 66.2c-10.3 9.8-10.1 26.3 .5 35.7c9.3 8.3 23.4 8.1 32.5-.3l71.9-66.4c9.7-9 24.9-8.4 33.9 1.4s8.4 24.9-1.4 33.9l-.8 .8 74.4 74.4c10 10 16.5 22.3 19.4 35.1H544zM64 336a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm528 16a16 16 0 1 0 0-32 16 16 0 1 0 0 32z\"],gem:[512,512,[128142],\"f3a5\",\"M168.5 72L256 165l87.5-93h-175zM383.9 99.1L311.5 176h129L383.9 99.1zm50 124.9H256 78.1L256 420.3 433.9 224zM71.5 176h129L128.1 99.1 71.5 176zm434.3 40.1l-232 256c-4.5 5-11 7.9-17.8 7.9s-13.2-2.9-17.8-7.9l-232-256c-7.7-8.5-8.3-21.2-1.5-30.4l112-152c4.5-6.1 11.7-9.8 19.3-9.8H376c7.6 0 14.8 3.6 19.3 9.8l112 152c6.8 9.2 6.1 21.9-1.5 30.4z\"],\"circle-play\":[512,512,[61469,\"play-circle\"],\"f144\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM188.3 147.1c7.6-4.2 16.8-4.1 24.3 .5l144 88c7.1 4.4 11.5 12.1 11.5 20.5s-4.4 16.1-11.5 20.5l-144 88c-7.4 4.5-16.7 4.7-24.3 .5s-12.3-12.2-12.3-20.9V168c0-8.7 4.7-16.7 12.3-20.9z\"],\"circle-check\":[512,512,[61533,\"check-circle\"],\"f058\",\"M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM369 209c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-111 111-47-47c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l64 64c9.4 9.4 24.6 9.4 33.9 0L369 209z\"],\"circle-stop\":[512,512,[62094,\"stop-circle\"],\"f28d\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm192-96H320c17.7 0 32 14.3 32 32V320c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32V192c0-17.7 14.3-32 32-32z\"],\"id-badge\":[384,512,[],\"f2c1\",\"M256 48V64c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32V48H64c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16H320c8.8 0 16-7.2 16-16V64c0-8.8-7.2-16-16-16H256zM0 64C0 28.7 28.7 0 64 0H320c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM160 320h64c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80zm-32-96a64 64 0 1 1 128 0 64 64 0 1 1 -128 0z\"],\"face-laugh-beam\":[512,512,[128513,\"laugh-beam\"],\"f59a\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm130.7 57.9c-4.2-13.6 7.1-25.9 21.3-25.9H364.5c14.2 0 25.5 12.4 21.3 25.9C369 368.4 318.2 408 258.2 408s-110.8-39.6-127.5-94.1zm86.9-85.1l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z\"],registered:[512,512,[174],\"f25d\",\"M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM160 152V264v96c0 13.3 10.7 24 24 24s24-10.7 24-24V288h60.9l37.2 81.9c5.5 12.1 19.7 17.4 31.8 11.9s17.4-19.7 11.9-31.8L315.7 275c21.8-14.3 36.3-39 36.3-67c0-44.2-35.8-80-80-80H184c-13.3 0-24 10.7-24 24zm48 88V176h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H208z\"],\"address-card\":[576,512,[62140,\"contact-card\",\"vcard\"],\"f2bb\",\"M512 80c8.8 0 16 7.2 16 16V416c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V96c0-8.8 7.2-16 16-16H512zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM208 256a64 64 0 1 0 0-128 64 64 0 1 0 0 128zm-32 32c-44.2 0-80 35.8-80 80c0 8.8 7.2 16 16 16H304c8.8 0 16-7.2 16-16c0-44.2-35.8-80-80-80H176zM376 144c-13.3 0-24 10.7-24 24s10.7 24 24 24h80c13.3 0 24-10.7 24-24s-10.7-24-24-24H376zm0 96c-13.3 0-24 10.7-24 24s10.7 24 24 24h80c13.3 0 24-10.7 24-24s-10.7-24-24-24H376z\"],\"face-tired\":[512,512,[128555,\"tired\"],\"f5c8\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm176.5 64.3C196.1 302.1 223.8 288 256 288s59.9 14.1 79.5 32.3C354.5 338.1 368 362 368 384c0 5.4-2.7 10.4-7.2 13.4s-10.2 3.4-15.2 1.3l-17.2-7.5c-22.8-10-47.5-15.1-72.4-15.1s-49.6 5.2-72.4 15.1l-17.2 7.5c-4.9 2.2-10.7 1.7-15.2-1.3s-7.2-8-7.2-13.4c0-22 13.5-45.9 32.5-63.7zm-43-173.6l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6c0-9 9.6-14.7 17.5-10.5zM396 157.1c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9c7.9-4.2 17.5 1.5 17.5 10.5z\"],\"font-awesome\":[512,512,[62501,62694,\"font-awesome-flag\",\"font-awesome-logo-full\"],\"f2b4\",\"M91.7 96C106.3 86.8 116 70.5 116 52C116 23.3 92.7 0 64 0S12 23.3 12 52c0 16.7 7.8 31.5 20 41l0 3 0 48 0 256 0 48 0 64 48 0 0-64 389.6 0c14.6 0 26.4-11.8 26.4-26.4c0-3.7-.8-7.3-2.3-10.7L432 272l61.7-138.9c1.5-3.4 2.3-7 2.3-10.7c0-14.6-11.8-26.4-26.4-26.4L91.7 96zM80 400l0-256 356.4 0L388.1 252.5c-5.5 12.4-5.5 26.6 0 39L436.4 400 80 400z\"],\"face-smile-wink\":[512,512,[128521,\"smile-wink\"],\"f4da\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm177.6 62.1C192.8 334.5 218.8 352 256 352s63.2-17.5 78.4-33.9c9-9.7 24.2-10.4 33.9-1.4s10.4 24.2 1.4 33.9c-22 23.8-60 49.4-113.6 49.4s-91.7-25.5-113.6-49.4c-9-9.7-8.4-24.9 1.4-33.9s24.9-8.4 33.9 1.4zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm165.8 21.7c-7.6 8.1-20.2 8.5-28.3 .9s-8.5-20.2-.9-28.3c14.5-15.5 35.2-22.3 54.6-22.3s40.1 6.8 54.6 22.3c7.6 8.1 7.1 20.7-.9 28.3s-20.7 7.1-28.3-.9c-5.5-5.8-14.8-9.7-25.4-9.7s-19.9 3.8-25.4 9.7z\"],\"file-word\":[384,512,[],\"f1c2\",\"M48 448V64c0-8.8 7.2-16 16-16H224v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm55 241.1c-3.8-12.7-17.2-19.9-29.9-16.1s-19.9 17.2-16.1 29.9l48 160c3 10.2 12.4 17.1 23 17.1s19.9-7 23-17.1l25-83.4 25 83.4c3 10.2 12.4 17.1 23 17.1s19.9-7 23-17.1l48-160c3.8-12.7-3.4-26.1-16.1-29.9s-26.1 3.4-29.9 16.1l-25 83.4-25-83.4c-3-10.2-12.4-17.1-23-17.1s-19.9 7-23 17.1l-25 83.4-25-83.4z\"],\"file-powerpoint\":[384,512,[],\"f1c4\",\"M64 464c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16H224v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm72 208c-13.3 0-24 10.7-24 24V336v56c0 13.3 10.7 24 24 24s24-10.7 24-24V360h44c42 0 76-34 76-76s-34-76-76-76H136zm68 104H160V256h44c15.5 0 28 12.5 28 28s-12.5 28-28 28z\"],\"envelope-open\":[512,512,[62135],\"f2b6\",\"M255.4 48.2c.2-.1 .4-.2 .6-.2s.4 .1 .6 .2L460.6 194c2.1 1.5 3.4 3.9 3.4 6.5v13.6L291.5 355.7c-20.7 17-50.4 17-71.1 0L48 214.1V200.5c0-2.6 1.2-5 3.4-6.5L255.4 48.2zM48 276.2L190 392.8c38.4 31.5 93.7 31.5 132 0L464 276.2V456c0 4.4-3.6 8-8 8H56c-4.4 0-8-3.6-8-8V276.2zM256 0c-10.2 0-20.2 3.2-28.5 9.1L23.5 154.9C8.7 165.4 0 182.4 0 200.5V456c0 30.9 25.1 56 56 56H456c30.9 0 56-25.1 56-56V200.5c0-18.1-8.7-35.1-23.4-45.6L284.5 9.1C276.2 3.2 266.2 0 256 0z\"],\"file-zipper\":[384,512,[\"file-archive\"],\"f1c6\",\"M64 464c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16h48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16h48v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm48 112c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H128c-8.8 0-16 7.2-16 16zm0 64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H128c-8.8 0-16 7.2-16 16zm-6.3 71.8L82.1 335.9c-1.4 5.4-2.1 10.9-2.1 16.4c0 35.2 28.8 63.7 64 63.7s64-28.5 64-63.7c0-5.5-.7-11.1-2.1-16.4l-23.5-88.2c-3.7-14-16.4-23.8-30.9-23.8H136.6c-14.5 0-27.2 9.7-30.9 23.8zM128 336h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H128c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],square:[448,512,[9632,9723,9724,61590],\"f0c8\",\"M384 80c8.8 0 16 7.2 16 16V416c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V96c0-8.8 7.2-16 16-16H384zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64z\"],snowflake:[448,512,[10052,10054],\"f2dc\",\"M224 0c13.3 0 24 10.7 24 24V70.1l23-23c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-57 57v76.5l66.2-38.2 20.9-77.8c3.4-12.8 16.6-20.4 29.4-17s20.4 16.6 17 29.4L373 142.2l37.1-21.4c11.5-6.6 26.2-2.7 32.8 8.8s2.7 26.2-8.8 32.8L397 183.8l31.5 8.4c12.8 3.4 20.4 16.6 17 29.4s-16.6 20.4-29.4 17l-77.8-20.9L272 256l66.2 38.2 77.8-20.9c12.8-3.4 26 4.2 29.4 17s-4.2 26-17 29.4L397 328.2l37.1 21.4c11.5 6.6 15.4 21.3 8.8 32.8s-21.3 15.4-32.8 8.8L373 369.8l8.4 31.5c3.4 12.8-4.2 26-17 29.4s-26-4.2-29.4-17l-20.9-77.8L248 297.6v76.5l57 57c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-23-23V488c0 13.3-10.7 24-24 24s-24-10.7-24-24V441.9l-23 23c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l57-57V297.6l-66.2 38.2-20.9 77.8c-3.4 12.8-16.6 20.4-29.4 17s-20.4-16.6-17-29.4L75 369.8 37.9 391.2c-11.5 6.6-26.2 2.7-32.8-8.8s-2.7-26.2 8.8-32.8L51 328.2l-31.5-8.4c-12.8-3.4-20.4-16.6-17-29.4s16.6-20.4 29.4-17l77.8 20.9L176 256l-66.2-38.2L31.9 238.6c-12.8 3.4-26-4.2-29.4-17s4.2-26 17-29.4L51 183.8 13.9 162.4c-11.5-6.6-15.4-21.3-8.8-32.8s21.3-15.4 32.8-8.8L75 142.2l-8.4-31.5c-3.4-12.8 4.2-26 17-29.4s26 4.2 29.4 17l20.9 77.8L200 214.4V137.9L143 81c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l23 23V24c0-13.3 10.7-24 24-24z\"],newspaper:[512,512,[128240],\"f1ea\",\"M168 80c-13.3 0-24 10.7-24 24V408c0 8.4-1.4 16.5-4.1 24H440c13.3 0 24-10.7 24-24V104c0-13.3-10.7-24-24-24H168zM72 480c-39.8 0-72-32.2-72-72V112C0 98.7 10.7 88 24 88s24 10.7 24 24V408c0 13.3 10.7 24 24 24s24-10.7 24-24V104c0-39.8 32.2-72 72-72H440c39.8 0 72 32.2 72 72V408c0 39.8-32.2 72-72 72H72zM176 136c0-13.3 10.7-24 24-24h96c13.3 0 24 10.7 24 24v80c0 13.3-10.7 24-24 24H200c-13.3 0-24-10.7-24-24V136zm200-24h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H376c-13.3 0-24-10.7-24-24s10.7-24 24-24zm0 80h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H376c-13.3 0-24-10.7-24-24s10.7-24 24-24zM200 272H408c13.3 0 24 10.7 24 24s-10.7 24-24 24H200c-13.3 0-24-10.7-24-24s10.7-24 24-24zm0 80H408c13.3 0 24 10.7 24 24s-10.7 24-24 24H200c-13.3 0-24-10.7-24-24s10.7-24 24-24z\"],\"face-kiss-wink-heart\":[512,512,[128536,\"kiss-wink-heart\"],\"f598\",\"M338.9 446.8c-25.4 11-53.4 17.2-82.9 17.2C141.1 464 48 370.9 48 256S141.1 48 256 48s208 93.1 208 208c0 22.4-3.5 43.9-10.1 64.1c3.1 4.5 5.7 9.4 7.8 14.6c12.7-1.6 25.1 .4 36.2 5c9.1-26.2 14-54.4 14-83.7C512 114.6 397.4 0 256 0S0 114.6 0 256S114.6 512 256 512c35.4 0 69.1-7.2 99.7-20.2c-4.8-5.5-8.5-12.2-10.4-19.7l-6.5-25.3zM296 316c0-6.9-3.1-13.2-7.3-18.3c-4.3-5.2-10.1-9.7-16.7-13.4C258.7 276.9 241.4 272 224 272c-3.6 0-6.8 2.5-7.7 6s.6 7.2 3.8 9l0 0 0 0 0 0 .2 .1c.2 .1 .5 .3 .9 .5c.8 .5 2 1.2 3.4 2.1c2.8 1.9 6.5 4.5 10.2 7.6c3.7 3.1 7.2 6.6 9.6 10.1c2.5 3.5 3.5 6.4 3.5 8.6s-1 5-3.5 8.6c-2.5 3.5-5.9 6.9-9.6 10.1c-3.7 3.1-7.4 5.7-10.2 7.6c-1.4 .9-2.6 1.6-3.4 2.1c-.4 .2-.7 .4-.9 .5l-.2 .1 0 0 0 0 0 0 0 0 0 0c-2.5 1.4-4.1 4.1-4.1 7s1.6 5.6 4.1 7l0 0 0 0 0 0 .2 .1c.2 .1 .5 .3 .9 .5c.8 .5 2 1.2 3.4 2.1c2.8 1.9 6.5 4.5 10.2 7.6c3.7 3.1 7.2 6.6 9.6 10.1c2.5 3.5 3.5 6.4 3.5 8.6s-1 5-3.5 8.6c-2.5 3.5-5.9 6.9-9.6 10.1c-3.7 3.1-7.4 5.7-10.2 7.6c-1.4 .9-2.6 1.6-3.4 2.1c-.4 .2-.7 .4-.9 .5l-.2 .1 0 0 0 0 0 0 0 0c-3.2 1.8-4.7 5.5-3.8 9s4.1 6 7.7 6c17.4 0 34.7-4.9 47.9-12.3c6.6-3.7 12.5-8.2 16.7-13.4c4.3-5.1 7.3-11.4 7.3-18.3s-3.1-13.2-7.3-18.3c-4.3-5.2-10.1-9.7-16.7-13.4c-2.7-1.5-5.7-3-8.7-4.3c3.1-1.3 6-2.7 8.7-4.3c6.6-3.7 12.5-8.2 16.7-13.4c4.3-5.1 7.3-11.4 7.3-18.3zM176.4 240a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm159.3-20c10.6 0 19.9 3.8 25.4 9.7c7.6 8.1 20.2 8.5 28.3 .9s8.5-20.2 .9-28.3C375.7 186.8 355 180 335.6 180s-40.1 6.8-54.6 22.3c-7.6 8.1-7.1 20.7 .9 28.3s20.7 7.1 28.3-.9c5.5-5.8 14.8-9.7 25.4-9.7zM434 352.3c-6-23.2-28.8-37-51.1-30.8s-35.4 30.1-29.5 53.4l22.9 89.3c2.2 8.7 11.2 13.9 19.8 11.4l84.9-23.8c22.2-6.2 35.4-30.1 29.5-53.4s-28.8-37-51.1-30.8l-20.2 5.6-5.4-21z\"],\"star-half-stroke\":[576,512,[\"star-half-alt\"],\"f5c0\",\"M309.5 13.5C305.5 5.2 297.1 0 287.9 0s-17.6 5.2-21.6 13.5L197.7 154.8 44.5 177.5c-9 1.3-16.5 7.6-19.3 16.3s-.5 18.1 5.9 24.5L142.2 328.4 116 483.9c-1.5 9 2.2 18.1 9.7 23.5s17.3 6 25.3 1.7l137-73.2 137 73.2c8.1 4.3 17.9 3.7 25.3-1.7s11.2-14.5 9.7-23.5L433.6 328.4 544.8 218.2c6.5-6.4 8.7-15.9 5.9-24.5s-10.3-14.9-19.3-16.3L378.1 154.8 309.5 13.5zM288 384.7V79.1l52.5 108.1c3.5 7.1 10.2 12.1 18.1 13.3l118.3 17.5L391 303c-5.5 5.5-8.1 13.3-6.8 21l20.2 119.6L299.2 387.5c-3.5-1.9-7.4-2.8-11.2-2.8z\"],\"file-excel\":[384,512,[],\"f1c3\",\"M48 448V64c0-8.8 7.2-16 16-16H224v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm90.9 233.3c-8.1-10.5-23.2-12.3-33.7-4.2s-12.3 23.2-4.2 33.7L161.6 320l-44.5 57.3c-8.1 10.5-6.3 25.5 4.2 33.7s25.5 6.3 33.7-4.2L192 359.1l37.1 47.6c8.1 10.5 23.2 12.3 33.7 4.2s12.3-23.2 4.2-33.7L222.4 320l44.5-57.3c8.1-10.5 6.3-25.5-4.2-33.7s-25.5-6.3-33.7 4.2L192 280.9l-37.1-47.6z\"],\"face-grin-beam\":[512,512,[128516,\"grin-beam\"],\"f582\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm349.5 52.4c18.7-4.4 35.9 12 25.5 28.1C350.4 374.6 306.3 400 255.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5zM217.6 228.8l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z\"],\"object-ungroup\":[640,512,[],\"f248\",\"M48.2 66.8c-.1-.8-.2-1.7-.2-2.5c0-.1 0-.1 0-.2c0-8.8 7.2-16 16-16c.9 0 1.9 .1 2.8 .2C74.3 49.5 80 56.1 80 64c0 8.8-7.2 16-16 16c-7.9 0-14.5-5.7-15.8-13.2zM0 64c0 26.9 16.5 49.9 40 59.3V228.7C16.5 238.1 0 261.1 0 288c0 35.3 28.7 64 64 64c26.9 0 49.9-16.5 59.3-40H324.7c9.5 23.5 32.5 40 59.3 40c35.3 0 64-28.7 64-64c0-26.9-16.5-49.9-40-59.3V123.3c23.5-9.5 40-32.5 40-59.3c0-35.3-28.7-64-64-64c-26.9 0-49.9 16.5-59.3 40H123.3C113.9 16.5 90.9 0 64 0C28.7 0 0 28.7 0 64zm368 0a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zM324.7 88c6.5 16 19.3 28.9 35.3 35.3V228.7c-16 6.5-28.9 19.3-35.3 35.3H123.3c-6.5-16-19.3-28.9-35.3-35.3V123.3c16-6.5 28.9-19.3 35.3-35.3H324.7zM384 272a16 16 0 1 1 0 32 16 16 0 1 1 0-32zM80 288c0 7.9-5.7 14.5-13.2 15.8c-.8 .1-1.7 .2-2.5 .2l-.2 0c-8.8 0-16-7.2-16-16c0-.9 .1-1.9 .2-2.8C49.5 277.7 56.1 272 64 272c8.8 0 16 7.2 16 16zm391.3-40h45.4c6.5 16 19.3 28.9 35.3 35.3V388.7c-16 6.5-28.9 19.3-35.3 35.3H315.3c-6.5-16-19.3-28.9-35.3-35.3V352H232v36.7c-23.5 9.5-40 32.5-40 59.3c0 35.3 28.7 64 64 64c26.9 0 49.9-16.5 59.3-40H516.7c9.5 23.5 32.5 40 59.3 40c35.3 0 64-28.7 64-64c0-26.9-16.5-49.9-40-59.3V283.3c23.5-9.5 40-32.5 40-59.3c0-35.3-28.7-64-64-64c-26.9 0-49.9 16.5-59.3 40H448v16.4c9.8 8.8 17.8 19.5 23.3 31.6zm88.9-26.7a16 16 0 1 1 31.5 5.5 16 16 0 1 1 -31.5-5.5zM271.8 450.7a16 16 0 1 1 -31.5-5.5 16 16 0 1 1 31.5 5.5zm301.5 13c-7.5-1.3-13.2-7.9-13.2-15.8c0-8.8 7.2-16 16-16c7.9 0 14.5 5.7 15.8 13.2l0 .1c.1 .9 .2 1.8 .2 2.7c0 8.8-7.2 16-16 16c-.9 0-1.9-.1-2.8-.2z\"],\"circle-right\":[512,512,[61838,\"arrow-alt-circle-right\"],\"f35a\",\"M464 256A208 208 0 1 1 48 256a208 208 0 1 1 416 0zM0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zM294.6 135.1c-4.2-4.5-10.1-7.1-16.3-7.1C266 128 256 138 256 150.3V208H160c-17.7 0-32 14.3-32 32v32c0 17.7 14.3 32 32 32h96v57.7c0 12.3 10 22.3 22.3 22.3c6.2 0 12.1-2.6 16.3-7.1l99.9-107.1c3.5-3.8 5.5-8.7 5.5-13.8s-2-10.1-5.5-13.8L294.6 135.1z\"],\"face-rolling-eyes\":[512,512,[128580,\"meh-rolling-eyes\"],\"f5a5\",\"M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM168 376c0 13.3 10.7 24 24 24H320c13.3 0 24-10.7 24-24s-10.7-24-24-24H192c-13.3 0-24 10.7-24 24zm-8-104c-26.5 0-48-21.5-48-48c0-14.3 6.3-27.2 16.2-36c-.2 1.3-.2 2.6-.2 4c0 17.7 14.3 32 32 32s32-14.3 32-32c0-1.4-.1-2.7-.2-4c10 8.8 16.2 21.7 16.2 36c0 26.5-21.5 48-48 48zm0 32a80 80 0 1 0 0-160 80 80 0 1 0 0 160zm192-32c-26.5 0-48-21.5-48-48c0-14.3 6.3-27.2 16.2-36c-.2 1.3-.2 2.6-.2 4c0 17.7 14.3 32 32 32s32-14.3 32-32c0-1.4-.1-2.7-.2-4c10 8.8 16.2 21.7 16.2 36c0 26.5-21.5 48-48 48zm0 32a80 80 0 1 0 0-160 80 80 0 1 0 0 160z\"],\"object-group\":[576,512,[],\"f247\",\"M48 115.8C38.2 107 32 94.2 32 80c0-26.5 21.5-48 48-48c14.2 0 27 6.2 35.8 16H460.2c8.8-9.8 21.6-16 35.8-16c26.5 0 48 21.5 48 48c0 14.2-6.2 27-16 35.8V396.2c9.8 8.8 16 21.6 16 35.8c0 26.5-21.5 48-48 48c-14.2 0-27-6.2-35.8-16H115.8c-8.8 9.8-21.6 16-35.8 16c-26.5 0-48-21.5-48-48c0-14.2 6.2-27 16-35.8V115.8zM125.3 96c-4.8 13.6-15.6 24.4-29.3 29.3V386.7c13.6 4.8 24.4 15.6 29.3 29.3H450.7c4.8-13.6 15.6-24.4 29.3-29.3V125.3c-13.6-4.8-24.4-15.6-29.3-29.3H125.3zm2.7 64c0-17.7 14.3-32 32-32H288c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32V160zM256 320h32c35.3 0 64-28.7 64-64V224h64c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H288c-17.7 0-32-14.3-32-32V320z\"],heart:[512,512,[128153,128154,128155,128156,128420,129293,129294,129505,9829,10084,61578],\"f004\",\"M225.8 468.2l-2.5-2.3L48.1 303.2C17.4 274.7 0 234.7 0 192.8v-3.3c0-70.4 50-130.8 119.2-144C158.6 37.9 198.9 47 231 69.6c9 6.4 17.4 13.8 25 22.3c4.2-4.8 8.7-9.2 13.5-13.3c3.7-3.2 7.5-6.2 11.5-9c0 0 0 0 0 0C313.1 47 353.4 37.9 392.8 45.4C462 58.6 512 119.1 512 189.5v3.3c0 41.9-17.4 81.9-48.1 110.4L288.7 465.9l-2.5 2.3c-8.2 7.6-19 11.9-30.2 11.9s-22-4.2-30.2-11.9zM239.1 145c-.4-.3-.7-.7-1-1.1l-17.8-20c0 0-.1-.1-.1-.1c0 0 0 0 0 0c-23.1-25.9-58-37.7-92-31.2C81.6 101.5 48 142.1 48 189.5v3.3c0 28.5 11.9 55.8 32.8 75.2L256 430.7 431.2 268c20.9-19.4 32.8-46.7 32.8-75.2v-3.3c0-47.3-33.6-88-80.1-96.9c-34-6.5-69 5.4-92 31.2c0 0 0 0-.1 .1s0 0-.1 .1l-17.8 20c-.3 .4-.7 .7-1 1.1c-4.5 4.5-10.6 7-16.9 7s-12.4-2.5-16.9-7z\"],\"face-surprise\":[512,512,[128558,\"surprise\"],\"f5c2\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm176.4-80a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM256 288a64 64 0 1 1 0 128 64 64 0 1 1 0-128z\"],\"circle-pause\":[512,512,[62092,\"pause-circle\"],\"f28b\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm224-72V328c0 13.3-10.7 24-24 24s-24-10.7-24-24V184c0-13.3 10.7-24 24-24s24 10.7 24 24zm112 0V328c0 13.3-10.7 24-24 24s-24-10.7-24-24V184c0-13.3 10.7-24 24-24s24 10.7 24 24z\"],circle:[512,512,[128308,128309,128992,128993,128994,128995,128996,9679,9898,9899,11044,61708,61915],\"f111\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256z\"],\"circle-up\":[512,512,[61467,\"arrow-alt-circle-up\"],\"f35b\",\"M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM135.1 217.4c-4.5 4.2-7.1 10.1-7.1 16.3c0 12.3 10 22.3 22.3 22.3H208v96c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V256h57.7c12.3 0 22.3-10 22.3-22.3c0-6.2-2.6-12.1-7.1-16.3L269.8 117.5c-3.8-3.5-8.7-5.5-13.8-5.5s-10.1 2-13.8 5.5L135.1 217.4z\"],\"file-audio\":[384,512,[],\"f1c7\",\"M64 464H320c8.8 0 16-7.2 16-16V160H256c-17.7 0-32-14.3-32-32V48H64c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16zM0 64C0 28.7 28.7 0 64 0H229.5c17 0 33.3 6.7 45.3 18.7l90.5 90.5c12 12 18.7 28.3 18.7 45.3V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM192 272V400c0 6.5-3.9 12.3-9.9 14.8s-12.9 1.1-17.4-3.5L129.4 376H112c-8.8 0-16-7.2-16-16V312c0-8.8 7.2-16 16-16h17.4l35.3-35.3c4.6-4.6 11.5-5.9 17.4-3.5s9.9 8.3 9.9 14.8zm85.8-4c11.6 20 18.2 43.3 18.2 68s-6.6 48-18.2 68c-6.6 11.5-21.3 15.4-32.8 8.8s-15.4-21.3-8.8-32.8c7.5-12.9 11.8-27.9 11.8-44s-4.3-31.1-11.8-44c-6.6-11.5-2.7-26.2 8.8-32.8s26.2-2.7 32.8 8.8z\"],\"file-image\":[384,512,[128443],\"f1c5\",\"M64 464c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16H224v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm96 256a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm69.2 46.9c-3-4.3-7.9-6.9-13.2-6.9s-10.2 2.6-13.2 6.9l-41.3 59.7-11.9-19.1c-2.9-4.7-8.1-7.5-13.6-7.5s-10.6 2.8-13.6 7.5l-40 64c-3.1 4.9-3.2 11.1-.4 16.2s8.2 8.2 14 8.2h48 32 40 72c6 0 11.4-3.3 14.2-8.6s2.4-11.6-1-16.5l-72-104z\"],\"circle-question\":[512,512,[62108,\"question-circle\"],\"f059\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm169.8-90.7c7.9-22.3 29.1-37.3 52.8-37.3h58.3c34.9 0 63.1 28.3 63.1 63.1c0 22.6-12.1 43.5-31.7 54.8L280 264.4c-.2 13-10.9 23.6-24 23.6c-13.3 0-24-10.7-24-24V250.5c0-8.6 4.6-16.5 12.1-20.8l44.3-25.4c4.7-2.7 7.6-7.7 7.6-13.1c0-8.4-6.8-15.1-15.1-15.1H222.6c-3.4 0-6.4 2.1-7.5 5.3l-.4 1.2c-4.4 12.5-18.2 19-30.6 14.6s-19-18.2-14.6-30.6l.4-1.2zM224 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],\"face-meh-blank\":[512,512,[128566,\"meh-blank\"],\"f5a4\",\"M256 48a208 208 0 1 0 0 416 208 208 0 1 0 0-416zM512 256A256 256 0 1 1 0 256a256 256 0 1 1 512 0zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],eye:[576,512,[128065],\"f06e\",\"M288 80c-65.2 0-118.8 29.6-159.9 67.7C89.6 183.5 63 226 49.4 256c13.6 30 40.2 72.5 78.6 108.3C169.2 402.4 222.8 432 288 432s118.8-29.6 159.9-67.7C486.4 328.5 513 286 526.6 256c-13.6-30-40.2-72.5-78.6-108.3C406.8 109.6 353.2 80 288 80zM95.4 112.6C142.5 68.8 207.2 32 288 32s145.5 36.8 192.6 80.6c46.8 43.5 78.1 95.4 93 131.1c3.3 7.9 3.3 16.7 0 24.6c-14.9 35.7-46.2 87.7-93 131.1C433.5 443.2 368.8 480 288 480s-145.5-36.8-192.6-80.6C48.6 356 17.3 304 2.5 268.3c-3.3-7.9-3.3-16.7 0-24.6C17.3 208 48.6 156 95.4 112.6zM288 336c44.2 0 80-35.8 80-80s-35.8-80-80-80c-.7 0-1.3 0-2 0c1.3 5.1 2 10.5 2 16c0 35.3-28.7 64-64 64c-5.5 0-10.9-.7-16-2c0 .7 0 1.3 0 2c0 44.2 35.8 80 80 80zm0-208a128 128 0 1 1 0 256 128 128 0 1 1 0-256z\"],\"face-sad-cry\":[512,512,[128557,\"sad-cry\"],\"f5b3\",\"M400 406.1V288c0-13.3-10.7-24-24-24s-24 10.7-24 24V440.6c-28.7 15-61.4 23.4-96 23.4s-67.3-8.5-96-23.4V288c0-13.3-10.7-24-24-24s-24 10.7-24 24V406.1C72.6 368.2 48 315 48 256C48 141.1 141.1 48 256 48s208 93.1 208 208c0 59-24.6 112.2-64 150.1zM256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM159.6 220c10.6 0 19.9 3.8 25.4 9.7c7.6 8.1 20.2 8.5 28.3 .9s8.5-20.2 .9-28.3C199.7 186.8 179 180 159.6 180s-40.1 6.8-54.6 22.3c-7.6 8.1-7.1 20.7 .9 28.3s20.7 7.1 28.3-.9c5.5-5.8 14.8-9.7 25.4-9.7zm166.6 9.7c5.5-5.8 14.8-9.7 25.4-9.7s19.9 3.8 25.4 9.7c7.6 8.1 20.2 8.5 28.3 .9s8.5-20.2 .9-28.3C391.7 186.8 371 180 351.6 180s-40.1 6.8-54.6 22.3c-7.6 8.1-7.1 20.7 .9 28.3s20.7 7.1 28.3-.9zM208 320v32c0 26.5 21.5 48 48 48s48-21.5 48-48V320c0-26.5-21.5-48-48-48s-48 21.5-48 48z\"],\"file-code\":[384,512,[],\"f1c9\",\"M64 464c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16H224v80c0 17.7 14.3 32 32 32h80V448c0 8.8-7.2 16-16 16H64zM64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V154.5c0-17-6.7-33.3-18.7-45.3L274.7 18.7C262.7 6.7 246.5 0 229.5 0H64zm97 289c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L79 303c-9.4 9.4-9.4 24.6 0 33.9l48 48c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-31-31 31-31zM257 255c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l31 31-31 31c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l48-48c9.4-9.4 9.4-24.6 0-33.9l-48-48z\"],\"window-maximize\":[512,512,[128470],\"f2d0\",\"M.3 89.5C.1 91.6 0 93.8 0 96V224 416c0 35.3 28.7 64 64 64l384 0c35.3 0 64-28.7 64-64V224 96c0-35.3-28.7-64-64-64H64c-2.2 0-4.4 .1-6.5 .3c-9.2 .9-17.8 3.8-25.5 8.2C21.8 46.5 13.4 55.1 7.7 65.5c-3.9 7.3-6.5 15.4-7.4 24zM48 224H464l0 192c0 8.8-7.2 16-16 16L64 432c-8.8 0-16-7.2-16-16l0-192z\"],\"face-frown\":[512,512,[9785,\"frown\"],\"f119\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM174.6 384.1c-4.5 12.5-18.2 18.9-30.7 14.4s-18.9-18.2-14.4-30.7C146.9 319.4 198.9 288 256 288s109.1 31.4 126.6 79.9c4.5 12.5-2 26.2-14.4 30.7s-26.2-2-30.7-14.4C328.2 358.5 297.2 336 256 336s-72.2 22.5-81.4 48.1zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"floppy-disk\":[448,512,[128190,128426,\"save\"],\"f0c7\",\"M48 96V416c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V170.5c0-4.2-1.7-8.3-4.7-11.3l33.9-33.9c12 12 18.7 28.3 18.7 45.3V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H309.5c17 0 33.3 6.7 45.3 18.7l74.5 74.5-33.9 33.9L320.8 84.7c-.3-.3-.5-.5-.8-.8V184c0 13.3-10.7 24-24 24H104c-13.3 0-24-10.7-24-24V80H64c-8.8 0-16 7.2-16 16zm80-16v80H272V80H128zm32 240a64 64 0 1 1 128 0 64 64 0 1 1 -128 0z\"],\"comment-dots\":[512,512,[128172,62075,\"commenting\"],\"f4ad\",\"M168.2 384.9c-15-5.4-31.7-3.1-44.6 6.4c-8.2 6-22.3 14.8-39.4 22.7c5.6-14.7 9.9-31.3 11.3-49.4c1-12.9-3.3-25.7-11.8-35.5C60.4 302.8 48 272 48 240c0-79.5 83.3-160 208-160s208 80.5 208 160s-83.3 160-208 160c-31.6 0-61.3-5.5-87.8-15.1zM26.3 423.8c-1.6 2.7-3.3 5.4-5.1 8.1l-.3 .5c-1.6 2.3-3.2 4.6-4.8 6.9c-3.5 4.7-7.3 9.3-11.3 13.5c-4.6 4.6-5.9 11.4-3.4 17.4c2.5 6 8.3 9.9 14.8 9.9c5.1 0 10.2-.3 15.3-.8l.7-.1c4.4-.5 8.8-1.1 13.2-1.9c.8-.1 1.6-.3 2.4-.5c17.8-3.5 34.9-9.5 50.1-16.1c22.9-10 42.4-21.9 54.3-30.6c31.8 11.5 67 17.9 104.1 17.9c141.4 0 256-93.1 256-208S397.4 32 256 32S0 125.1 0 240c0 45.1 17.7 86.8 47.7 120.9c-1.9 24.5-11.4 46.3-21.4 62.9zM144 272a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm144-32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm80 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"face-grin-squint\":[512,512,[128518,\"grin-squint\"],\"f585\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm349.5 52.4c18.7-4.4 35.9 12 25.5 28.1C350.4 374.6 306.3 400 255.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5zm-216-161.7l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6c0-9 9.6-14.7 17.5-10.5zM396 157.1c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9c7.9-4.2 17.5 1.5 17.5 10.5z\"],\"hand-pointer\":[448,512,[],\"f25a\",\"M160 64c0-8.8 7.2-16 16-16s16 7.2 16 16V200c0 10.3 6.6 19.5 16.4 22.8s20.6-.1 26.8-8.3c3-3.9 7.6-6.4 12.8-6.4c8.8 0 16 7.2 16 16c0 10.3 6.6 19.5 16.4 22.8s20.6-.1 26.8-8.3c3-3.9 7.6-6.4 12.8-6.4c7.8 0 14.3 5.6 15.7 13c1.6 8.2 7.3 15.1 15.1 18s16.7 1.6 23.3-3.6c2.7-2.1 6.1-3.4 9.9-3.4c8.8 0 16 7.2 16 16l0 16V392c0 39.8-32.2 72-72 72H272 212.3h-.9c-37.4 0-72.4-18.7-93.2-49.9L50.7 312.9c-4.9-7.4-2.9-17.3 4.4-22.2s17.3-2.9 22.2 4.4L116 353.2c5.9 8.8 16.8 12.7 26.9 9.7s17-12.4 17-23V320 64zM176 0c-35.3 0-64 28.7-64 64V261.7C91.2 238 55.5 232.8 28.5 250.7C-.9 270.4-8.9 310.1 10.8 339.5L78.3 440.8c29.7 44.5 79.6 71.2 133.1 71.2h.9H272h56c66.3 0 120-53.7 120-120V288l0-16c0-35.3-28.7-64-64-64c-4.5 0-8.8 .5-13 1.3c-11.7-15.4-30.2-25.3-51-25.3c-6.9 0-13.5 1.1-19.7 3.1C288.7 170.7 269.6 160 248 160c-2.7 0-5.4 .2-8 .5V64c0-35.3-28.7-64-64-64zm48 304c0-8.8-7.2-16-16-16s-16 7.2-16 16v96c0 8.8 7.2 16 16 16s16-7.2 16-16V304zm48-16c-8.8 0-16 7.2-16 16v96c0 8.8 7.2 16 16 16s16-7.2 16-16V304c0-8.8-7.2-16-16-16zm80 16c0-8.8-7.2-16-16-16s-16 7.2-16 16v96c0 8.8 7.2 16 16 16s16-7.2 16-16V304z\"],\"hand-scissors\":[512,512,[],\"f257\",\"M.2 276.3c-1.2-35.3 26.4-65 61.7-66.2l3.3-.1L57 208.1C22.5 200.5 .7 166.3 8.3 131.8S50.2 75.5 84.7 83.2l173 38.3c2.3-2.9 4.7-5.7 7.1-8.5l18.4-20.3C299.9 74.5 323.5 64 348.3 64l10.2 0c54.1 0 104.1 28.7 131.3 75.4l1.5 2.6c13.6 23.2 20.7 49.7 20.7 76.6L512 344c0 66.3-53.7 120-120 120l-8 0-96 0c-35.3 0-64-28.7-64-64c0-2.8 .2-5.6 .5-8.3c-19.4-11-32.5-31.8-32.5-55.7c0-.8 0-1.6 0-2.4L66.4 338c-35.3 1.2-65-26.4-66.2-61.7zm63.4-18.2c-8.8 .3-15.7 7.7-15.4 16.5s7.7 15.7 16.5 15.4l161.5-5.6c9.8-.3 18.7 5.3 22.7 14.2s2.2 19.3-4.5 26.4c-2.8 2.9-4.4 6.7-4.4 11c0 8.8 7.2 16 16 16c9.1 0 17.4 5.1 21.5 13.3s3.2 17.9-2.3 25.1c-2 2.7-3.2 6-3.2 9.6c0 8.8 7.2 16 16 16l96 0 8 0c39.8 0 72-32.2 72-72l0-125.4c0-18.4-4.9-36.5-14.2-52.4l-1.5-2.6c-18.6-32-52.8-51.6-89.8-51.6l-10.2 0c-11.3 0-22 4.8-29.6 13.1l-17.5-15.9 17.5 15.9-18.4 20.3c-.6 .6-1.1 1.3-1.7 1.9l57 13.2c8.6 2 14 10.6 12 19.2s-10.6 14-19.2 12l-85.6-19.7L74.3 130c-8.6-1.9-17.2 3.5-19.1 12.2s3.5 17.2 12.2 19.1l187.5 41.6c10.2 2.3 17.8 10.9 18.7 21.4l.1 1c.6 6.6-1.5 13.1-5.8 18.1s-10.6 7.9-17.2 8.2L63.6 258.1z\"],\"face-grin-tears\":[640,512,[128514,\"grin-tears\"],\"f588\",\"M516.1 325.5c1 3 2.1 6 3.3 8.9c3.3 8.1 8.4 18.5 16.5 26.6c3.9 3.9 8.2 7.4 12.7 10.3C506.4 454.8 419.9 512 320 512s-186.4-57.2-228.6-140.6c4.5-2.9 8.7-6.3 12.7-10.3c8.1-8.1 13.2-18.6 16.5-26.6c1.2-2.9 2.3-5.9 3.3-8.9C152.5 406.2 229.5 464 320 464s167.5-57.8 196.1-138.5zM320 48c-101.4 0-185.8 72.5-204.3 168.5c-6.7-3.1-14.3-4.3-22.3-3.1c-6.8 .9-16.2 2.4-26.6 4.4C85.3 94.5 191.6 0 320 0S554.7 94.5 573.2 217.7c-10.3-2-19.8-3.5-26.6-4.4c-8-1.2-15.7 .1-22.3 3.1C505.8 120.5 421.4 48 320 48zM78.5 341.1C60 356.7 32 355.5 14.3 337.7c-18.7-18.7-19.1-48.8-.7-67.2c8.6-8.6 30.1-15.1 50.5-19.6c13-2.8 25.5-4.8 33.9-6c5.4-.8 9.9 3.7 9 9c-3.1 21.5-11.4 70.2-25.5 84.4c-.9 1-1.9 1.8-2.9 2.7zm483 0c-.8-.6-1.5-1.3-2.3-2c-.2-.2-.5-.4-.7-.7c-14.1-14.1-22.5-62.9-25.5-84.4c-.8-5.4 3.7-9.9 9-9c1 .1 2.2 .3 3.3 .5c8.2 1.2 19.2 3 30.6 5.5c20.4 4.4 41.9 10.9 50.5 19.6c18.4 18.4 18 48.5-.7 67.2c-17.7 17.7-45.7 19-64.2 3.4zM439 336.5C414.4 374.6 370.3 400 319.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5c18.7-4.4 35.9 12 25.5 28.1zM281.6 228.8l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0zm160 0l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0z\"],\"calendar-xmark\":[448,512,[\"calendar-times\"],\"f273\",\"M128 0c13.3 0 24 10.7 24 24V64H296V24c0-13.3 10.7-24 24-24s24 10.7 24 24V64h40c35.3 0 64 28.7 64 64v16 48V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V192 144 128C0 92.7 28.7 64 64 64h40V24c0-13.3 10.7-24 24-24zM400 192H48V448c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V192zm-95 89l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z\"],\"file-video\":[384,512,[],\"f1c8\",\"M320 464c8.8 0 16-7.2 16-16V160H256c-17.7 0-32-14.3-32-32V48H64c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16H320zM0 64C0 28.7 28.7 0 64 0H229.5c17 0 33.3 6.7 45.3 18.7l90.5 90.5c12 12 18.7 28.3 18.7 45.3V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM80 288c0-17.7 14.3-32 32-32h96c17.7 0 32 14.3 32 32v16l44.9-29.9c2-1.3 4.4-2.1 6.8-2.1c6.8 0 12.3 5.5 12.3 12.3V387.7c0 6.8-5.5 12.3-12.3 12.3c-2.4 0-4.8-.7-6.8-2.1L240 368v16c0 17.7-14.3 32-32 32H112c-17.7 0-32-14.3-32-32V288z\"],\"file-pdf\":[512,512,[],\"f1c1\",\"M64 464l48 0 0 48-48 0c-35.3 0-64-28.7-64-64L0 64C0 28.7 28.7 0 64 0L229.5 0c17 0 33.3 6.7 45.3 18.7l90.5 90.5c12 12 18.7 28.3 18.7 45.3L384 304l-48 0 0-144-80 0c-17.7 0-32-14.3-32-32l0-80L64 48c-8.8 0-16 7.2-16 16l0 384c0 8.8 7.2 16 16 16zM176 352l32 0c30.9 0 56 25.1 56 56s-25.1 56-56 56l-16 0 0 32c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-48 0-80c0-8.8 7.2-16 16-16zm32 80c13.3 0 24-10.7 24-24s-10.7-24-24-24l-16 0 0 48 16 0zm96-80l32 0c26.5 0 48 21.5 48 48l0 64c0 26.5-21.5 48-48 48l-32 0c-8.8 0-16-7.2-16-16l0-128c0-8.8 7.2-16 16-16zm32 128c8.8 0 16-7.2 16-16l0-64c0-8.8-7.2-16-16-16l-16 0 0 96 16 0zm80-112c0-8.8 7.2-16 16-16l48 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-32 0 0 32 32 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-32 0 0 48c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-64 0-64z\"],comment:[512,512,[128489,61669],\"f075\",\"M123.6 391.3c12.9-9.4 29.6-11.8 44.6-6.4c26.5 9.6 56.2 15.1 87.8 15.1c124.7 0 208-80.5 208-160s-83.3-160-208-160S48 160.5 48 240c0 32 12.4 62.8 35.7 89.2c8.6 9.7 12.8 22.5 11.8 35.5c-1.4 18.1-5.7 34.7-11.3 49.4c17-7.9 31.1-16.7 39.4-22.7zM21.2 431.9c1.8-2.7 3.5-5.4 5.1-8.1c10-16.6 19.5-38.4 21.4-62.9C17.7 326.8 0 285.1 0 240C0 125.1 114.6 32 256 32s256 93.1 256 208s-114.6 208-256 208c-37.1 0-72.3-6.4-104.1-17.9c-11.9 8.7-31.3 20.6-54.3 30.6c-15.1 6.6-32.3 12.6-50.1 16.1c-.8 .2-1.6 .3-2.4 .5c-4.4 .8-8.7 1.5-13.2 1.9c-.2 0-.5 .1-.7 .1c-5.1 .5-10.2 .8-15.3 .8c-6.5 0-12.3-3.9-14.8-9.9c-2.5-6-1.1-12.8 3.4-17.4c4.1-4.2 7.8-8.7 11.3-13.5c1.7-2.3 3.3-4.6 4.8-6.9c.1-.2 .2-.3 .3-.5z\"],envelope:[512,512,[128386,9993,61443],\"f0e0\",\"M64 112c-8.8 0-16 7.2-16 16v22.1L220.5 291.7c20.7 17 50.4 17 71.1 0L464 150.1V128c0-8.8-7.2-16-16-16H64zM48 212.2V384c0 8.8 7.2 16 16 16H448c8.8 0 16-7.2 16-16V212.2L322 328.8c-38.4 31.5-93.7 31.5-132 0L48 212.2zM0 128C0 92.7 28.7 64 64 64H448c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128z\"],hourglass:[384,512,[9203,62032,\"hourglass-empty\"],\"f254\",\"M24 0C10.7 0 0 10.7 0 24S10.7 48 24 48h8V67c0 40.3 16 79 44.5 107.5L158.1 256 76.5 337.5C48 366 32 404.7 32 445v19H24c-13.3 0-24 10.7-24 24s10.7 24 24 24H360c13.3 0 24-10.7 24-24s-10.7-24-24-24h-8V445c0-40.3-16-79-44.5-107.5L225.9 256l81.5-81.5C336 146 352 107.3 352 67V48h8c13.3 0 24-10.7 24-24s-10.7-24-24-24H24zM192 289.9l81.5 81.5C293 391 304 417.4 304 445v19H80V445c0-27.6 11-54 30.5-73.5L192 289.9zm0-67.9l-81.5-81.5C91 121 80 94.6 80 67V48H304V67c0 27.6-11 54-30.5 73.5L192 222.1z\"],\"calendar-check\":[448,512,[],\"f274\",\"M128 0c13.3 0 24 10.7 24 24V64H296V24c0-13.3 10.7-24 24-24s24 10.7 24 24V64h40c35.3 0 64 28.7 64 64v16 48V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V192 144 128C0 92.7 28.7 64 64 64h40V24c0-13.3 10.7-24 24-24zM400 192H48V448c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V192zM329 297L217 409c-9.4 9.4-24.6 9.4-33.9 0l-64-64c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47 95-95c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z\"],\"hard-drive\":[512,512,[128436,\"hdd\"],\"f0a0\",\"M64 80c-8.8 0-16 7.2-16 16V258c5.1-1.3 10.5-2 16-2H448c5.5 0 10.9 .7 16 2V96c0-8.8-7.2-16-16-16H64zM48 320v96c0 8.8 7.2 16 16 16H448c8.8 0 16-7.2 16-16V320c0-8.8-7.2-16-16-16H64c-8.8 0-16 7.2-16 16zM0 320V96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V320v96c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V320zm280 48a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm120-24a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],\"face-grin-squint-tears\":[512,512,[129315,\"grin-squint-tears\"],\"f586\",\"M426.8 14.2C446-5 477.5-4.6 497.1 14.9s20 51 .7 70.3c-14.8 14.8-65.7 23.6-88.3 26.7c-5.6 .9-10.3-3.9-9.5-9.5C403.3 79.9 412 29 426.8 14.2zM75 75C158.2-8.3 284.5-22.2 382.2 33.2c-1.5 4.8-2.9 9.6-4.1 14.3c-3.1 12.2-5.5 24.6-7.3 35c-80.8-53.6-190.7-44.8-261.9 26.4C37.7 180.1 28.9 290 82.5 370.8c-10.5 1.8-22.9 4.2-35 7.3c-4.7 1.2-9.5 2.5-14.3 4.1C-22.2 284.5-8.2 158.2 75 75zm389.6 58.9c4.7-1.2 9.5-2.5 14.3-4.1C534.2 227.5 520.2 353.8 437 437c-83.2 83.2-209.5 97.2-307.2 41.8c1.5-4.8 2.8-9.6 4-14.3c3.1-12.2 5.5-24.6 7.3-35c80.8 53.6 190.7 44.8 261.9-26.4c71.2-71.2 80-181.1 26.4-261.9c10.5-1.8 22.9-4.2 35-7.3zm-105.4 93c10.1-16.3 33.9-16.9 37.9 1.9c9.5 44.4-3.7 93.5-39.3 129.1s-84.8 48.8-129.1 39.3c-18.7-4-18.2-27.8-1.9-37.9c25.2-15.7 50.2-35.4 73.6-58.8s43.1-48.4 58.8-73.6zM92 265.3l97.4-29.7c11.6-3.5 22.5 7.3 19 19l-29.7 97.4c-2.6 8.6-13.4 11.3-19.8 4.9c-2-2-3.2-4.6-3.4-7.3l-5.1-56.1-56.1-5.1c-2.8-.3-5.4-1.5-7.3-3.4c-6.3-6.3-3.6-17.2 4.9-19.8zm193-178.2c2 2 3.2 4.6 3.4 7.3l5.1 56.1 56.1 5.1c2.8 .3 5.4 1.5 7.3 3.4c6.3 6.3 3.6 17.2-4.9 19.8l-97.4 29.7c-11.6 3.5-22.5-7.3-19-19L265.3 92c2.6-8.6 13.4-11.3 19.8-4.9zM14.9 497.1c-19.6-19.6-20-51-.7-70.3C29 412 79.8 403.2 102.4 400.1c5.6-.9 10.3 3.9 9.5 9.5c-3.2 22.5-11.9 73.5-26.7 88.3C66 517 34.5 516.6 14.9 497.1z\"],\"rectangle-list\":[576,512,[\"list-alt\"],\"f022\",\"M64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H512c8.8 0 16-7.2 16-16V96c0-8.8-7.2-16-16-16H64zM0 96C0 60.7 28.7 32 64 32H512c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm96 64a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm104 0c0-13.3 10.7-24 24-24H448c13.3 0 24 10.7 24 24s-10.7 24-24 24H224c-13.3 0-24-10.7-24-24zm0 96c0-13.3 10.7-24 24-24H448c13.3 0 24 10.7 24 24s-10.7 24-24 24H224c-13.3 0-24-10.7-24-24zm0 96c0-13.3 10.7-24 24-24H448c13.3 0 24 10.7 24 24s-10.7 24-24 24H224c-13.3 0-24-10.7-24-24zm-72-64a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM96 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],\"calendar-plus\":[448,512,[],\"f271\",\"M152 24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H64C28.7 64 0 92.7 0 128v16 48V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V192 144 128c0-35.3-28.7-64-64-64H344V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H152V24zM48 192H400V448c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V192zm176 40c-13.3 0-24 10.7-24 24v48H152c-13.3 0-24 10.7-24 24s10.7 24 24 24h48v48c0 13.3 10.7 24 24 24s24-10.7 24-24V352h48c13.3 0 24-10.7 24-24s-10.7-24-24-24H248V256c0-13.3-10.7-24-24-24z\"],\"circle-left\":[512,512,[61840,\"arrow-alt-circle-left\"],\"f359\",\"M48 256a208 208 0 1 1 416 0A208 208 0 1 1 48 256zm464 0A256 256 0 1 0 0 256a256 256 0 1 0 512 0zM217.4 376.9c4.2 4.5 10.1 7.1 16.3 7.1c12.3 0 22.3-10 22.3-22.3V304h96c17.7 0 32-14.3 32-32V240c0-17.7-14.3-32-32-32H256V150.3c0-12.3-10-22.3-22.3-22.3c-6.2 0-12.1 2.6-16.3 7.1L117.5 242.2c-3.5 3.8-5.5 8.7-5.5 13.8s2 10.1 5.5 13.8l99.9 107.1z\"],\"money-bill-1\":[576,512,[\"money-bill-alt\"],\"f3d1\",\"M112 112c0 35.3-28.7 64-64 64V336c35.3 0 64 28.7 64 64H464c0-35.3 28.7-64 64-64V176c-35.3 0-64-28.7-64-64H112zM0 128C0 92.7 28.7 64 64 64H512c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128zM176 256a112 112 0 1 1 224 0 112 112 0 1 1 -224 0zm80-48c0 8.8 7.2 16 16 16v64h-8c-8.8 0-16 7.2-16 16s7.2 16 16 16h24 24c8.8 0 16-7.2 16-16s-7.2-16-16-16h-8V208c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16z\"],clock:[512,512,[128339,\"clock-four\"],\"f017\",\"M464 256A208 208 0 1 1 48 256a208 208 0 1 1 416 0zM0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zM232 120V256c0 8 4 15.5 10.7 20l96 64c11 7.4 25.9 4.4 33.3-6.7s4.4-25.9-6.7-33.3L280 243.2V120c0-13.3-10.7-24-24-24s-24 10.7-24 24z\"],keyboard:[576,512,[9e3],\"f11c\",\"M64 112c-8.8 0-16 7.2-16 16V384c0 8.8 7.2 16 16 16H512c8.8 0 16-7.2 16-16V128c0-8.8-7.2-16-16-16H64zM0 128C0 92.7 28.7 64 64 64H512c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128zM176 320H400c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zm-72-72c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H120c-8.8 0-16-7.2-16-16V248zm16-96h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H120c-8.8 0-16-7.2-16-16V168c0-8.8 7.2-16 16-16zm64 96c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H200c-8.8 0-16-7.2-16-16V248zm16-96h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H200c-8.8 0-16-7.2-16-16V168c0-8.8 7.2-16 16-16zm64 96c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H280c-8.8 0-16-7.2-16-16V248zm16-96h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H280c-8.8 0-16-7.2-16-16V168c0-8.8 7.2-16 16-16zm64 96c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H360c-8.8 0-16-7.2-16-16V248zm16-96h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H360c-8.8 0-16-7.2-16-16V168c0-8.8 7.2-16 16-16zm64 96c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H440c-8.8 0-16-7.2-16-16V248zm16-96h16c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H440c-8.8 0-16-7.2-16-16V168c0-8.8 7.2-16 16-16z\"],\"closed-captioning\":[576,512,[],\"f20a\",\"M512 80c8.8 0 16 7.2 16 16V416c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V96c0-8.8 7.2-16 16-16H512zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM200 208c14.2 0 27 6.1 35.8 16c8.8 9.9 24 10.7 33.9 1.9s10.7-24 1.9-33.9c-17.5-19.6-43.1-32-71.5-32c-53 0-96 43-96 96s43 96 96 96c28.4 0 54-12.4 71.5-32c8.8-9.9 8-25-1.9-33.9s-25-8-33.9 1.9c-8.8 9.9-21.6 16-35.8 16c-26.5 0-48-21.5-48-48s21.5-48 48-48zm144 48c0-26.5 21.5-48 48-48c14.2 0 27 6.1 35.8 16c8.8 9.9 24 10.7 33.9 1.9s10.7-24 1.9-33.9c-17.5-19.6-43.1-32-71.5-32c-53 0-96 43-96 96s43 96 96 96c28.4 0 54-12.4 71.5-32c8.8-9.9 8-25-1.9-33.9s-25-8-33.9 1.9c-8.8 9.9-21.6 16-35.8 16c-26.5 0-48-21.5-48-48z\"],images:[576,512,[],\"f302\",\"M160 80H512c8.8 0 16 7.2 16 16V320c0 8.8-7.2 16-16 16H490.8L388.1 178.9c-4.4-6.8-12-10.9-20.1-10.9s-15.7 4.1-20.1 10.9l-52.2 79.8-12.4-16.9c-4.5-6.2-11.7-9.8-19.4-9.8s-14.8 3.6-19.4 9.8L175.6 336H160c-8.8 0-16-7.2-16-16V96c0-8.8 7.2-16 16-16zM96 96V320c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H160c-35.3 0-64 28.7-64 64zM48 120c0-13.3-10.7-24-24-24S0 106.7 0 120V344c0 75.1 60.9 136 136 136H456c13.3 0 24-10.7 24-24s-10.7-24-24-24H136c-48.6 0-88-39.4-88-88V120zm208 24a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z\"],\"face-grin\":[512,512,[128512,\"grin\"],\"f580\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm349.5 52.4c18.7-4.4 35.9 12 25.5 28.1C350.4 374.6 306.3 400 255.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"face-meh\":[512,512,[128528,\"meh\"],\"f11a\",\"M464 256A208 208 0 1 1 48 256a208 208 0 1 1 416 0zM256 0a256 256 0 1 0 0 512A256 256 0 1 0 256 0zM176.4 240a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm192-32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM184 328c-13.3 0-24 10.7-24 24s10.7 24 24 24H328c13.3 0 24-10.7 24-24s-10.7-24-24-24H184z\"],\"id-card\":[576,512,[62147,\"drivers-license\"],\"f2c2\",\"M528 160V416c0 8.8-7.2 16-16 16H320c0-44.2-35.8-80-80-80H176c-44.2 0-80 35.8-80 80H64c-8.8 0-16-7.2-16-16V160H528zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM272 256a64 64 0 1 0 -128 0 64 64 0 1 0 128 0zm104-48c-13.3 0-24 10.7-24 24s10.7 24 24 24h80c13.3 0 24-10.7 24-24s-10.7-24-24-24H376zm0 96c-13.3 0-24 10.7-24 24s10.7 24 24 24h80c13.3 0 24-10.7 24-24s-10.7-24-24-24H376z\"],sun:[512,512,[9728],\"f185\",\"M375.7 19.7c-1.5-8-6.9-14.7-14.4-17.8s-16.1-2.2-22.8 2.4L256 61.1 173.5 4.2c-6.7-4.6-15.3-5.5-22.8-2.4s-12.9 9.8-14.4 17.8l-18.1 98.5L19.7 136.3c-8 1.5-14.7 6.9-17.8 14.4s-2.2 16.1 2.4 22.8L61.1 256 4.2 338.5c-4.6 6.7-5.5 15.3-2.4 22.8s9.8 13 17.8 14.4l98.5 18.1 18.1 98.5c1.5 8 6.9 14.7 14.4 17.8s16.1 2.2 22.8-2.4L256 450.9l82.5 56.9c6.7 4.6 15.3 5.5 22.8 2.4s12.9-9.8 14.4-17.8l18.1-98.5 98.5-18.1c8-1.5 14.7-6.9 17.8-14.4s2.2-16.1-2.4-22.8L450.9 256l56.9-82.5c4.6-6.7 5.5-15.3 2.4-22.8s-9.8-12.9-17.8-14.4l-98.5-18.1L375.7 19.7zM269.6 110l65.6-45.2 14.4 78.3c1.8 9.8 9.5 17.5 19.3 19.3l78.3 14.4L402 242.4c-5.7 8.2-5.7 19 0 27.2l45.2 65.6-78.3 14.4c-9.8 1.8-17.5 9.5-19.3 19.3l-14.4 78.3L269.6 402c-8.2-5.7-19-5.7-27.2 0l-65.6 45.2-14.4-78.3c-1.8-9.8-9.5-17.5-19.3-19.3L64.8 335.2 110 269.6c5.7-8.2 5.7-19 0-27.2L64.8 176.8l78.3-14.4c9.8-1.8 17.5-9.5 19.3-19.3l14.4-78.3L242.4 110c8.2 5.7 19 5.7 27.2 0zM256 368a112 112 0 1 0 0-224 112 112 0 1 0 0 224zM192 256a64 64 0 1 1 128 0 64 64 0 1 1 -128 0z\"],\"face-laugh-wink\":[512,512,[\"laugh-wink\"],\"f59c\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm130.7 57.9c-4.2-13.6 7.1-25.9 21.3-25.9H364.5c14.2 0 25.5 12.4 21.3 25.9C369 368.4 318.2 408 258.2 408s-110.8-39.6-127.5-94.1zM144.4 192a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm165.8 21.7c-7.6 8.1-20.2 8.5-28.3 .9s-8.5-20.2-.9-28.3c14.5-15.5 35.2-22.3 54.6-22.3s40.1 6.8 54.6 22.3c7.6 8.1 7.1 20.7-.9 28.3s-20.7 7.1-28.3-.9c-5.5-5.8-14.8-9.7-25.4-9.7s-19.9 3.8-25.4 9.7z\"],\"circle-down\":[512,512,[61466,\"arrow-alt-circle-down\"],\"f358\",\"M256 464a208 208 0 1 1 0-416 208 208 0 1 1 0 416zM256 0a256 256 0 1 0 0 512A256 256 0 1 0 256 0zM376.9 294.6c4.5-4.2 7.1-10.1 7.1-16.3c0-12.3-10-22.3-22.3-22.3H304V160c0-17.7-14.3-32-32-32l-32 0c-17.7 0-32 14.3-32 32v96H150.3C138 256 128 266 128 278.3c0 6.2 2.6 12.1 7.1 16.3l107.1 99.9c3.8 3.5 8.7 5.5 13.8 5.5s10.1-2 13.8-5.5l107.1-99.9z\"],\"thumbs-down\":[512,512,[128078,61576],\"f165\",\"M323.8 477.2c-38.2 10.9-78.1-11.2-89-49.4l-5.7-20c-3.7-13-10.4-25-19.5-35l-51.3-56.4c-8.9-9.8-8.2-25 1.6-33.9s25-8.2 33.9 1.6l51.3 56.4c14.1 15.5 24.4 34 30.1 54.1l5.7 20c3.6 12.7 16.9 20.1 29.7 16.5s20.1-16.9 16.5-29.7l-5.7-20c-5.7-19.9-14.7-38.7-26.6-55.5c-5.2-7.3-5.8-16.9-1.7-24.9s12.3-13 21.3-13L448 288c8.8 0 16-7.2 16-16c0-6.8-4.3-12.7-10.4-15c-7.4-2.8-13-9-14.9-16.7s.1-15.8 5.3-21.7c2.5-2.8 4-6.5 4-10.6c0-7.8-5.6-14.3-13-15.7c-8.2-1.6-15.1-7.3-18-15.2s-1.6-16.7 3.6-23.3c2.1-2.7 3.4-6.1 3.4-9.9c0-6.7-4.2-12.6-10.2-14.9c-11.5-4.5-17.7-16.9-14.4-28.8c.4-1.3 .6-2.8 .6-4.3c0-8.8-7.2-16-16-16H286.5c-12.6 0-25 3.7-35.5 10.7l-61.7 41.1c-11 7.4-25.9 4.4-33.3-6.7s-4.4-25.9 6.7-33.3l61.7-41.1c18.4-12.3 40-18.8 62.1-18.8H384c34.7 0 62.9 27.6 64 62c14.6 11.7 24 29.7 24 50c0 4.5-.5 8.8-1.3 13c15.4 11.7 25.3 30.2 25.3 51c0 6.5-1 12.8-2.8 18.7C504.8 238.3 512 254.3 512 272c0 35.3-28.6 64-64 64l-92.3 0c4.7 10.4 8.7 21.2 11.8 32.2l5.7 20c10.9 38.2-11.2 78.1-49.4 89zM32 384c-17.7 0-32-14.3-32-32V128c0-17.7 14.3-32 32-32H96c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32H32z\"],\"chess-pawn\":[320,512,[9823],\"f443\",\"M232 152A72 72 0 1 0 88 152a72 72 0 1 0 144 0zm24 120H243.4l10.7 80H205.7L195 272H160 125l-10.7 80H65.9l10.7-80H64c-13.3 0-24-10.7-24-24s10.7-24 24-24c-15.1-20.1-24-45-24-72C40 85.7 93.7 32 160 32s120 53.7 120 120c0 27-8.9 51.9-24 72c13.3 0 24 10.7 24 24s-10.7 24-24 24zM52.7 464H267.3l-16.6-32H69.2L52.7 464zm207.9-80c12 0 22.9 6.7 28.4 17.3l26.5 51.2c3 5.8 4.6 12.2 4.6 18.7c0 22.5-18.2 40.8-40.8 40.8H40.8C18.2 512 0 493.8 0 471.2c0-6.5 1.6-12.9 4.6-18.7l26.5-51.2C36.5 390.7 47.5 384 59.5 384h201z\"],\"credit-card\":[576,512,[128179,62083,\"credit-card-alt\"],\"f09d\",\"M512 80c8.8 0 16 7.2 16 16v32H48V96c0-8.8 7.2-16 16-16H512zm16 144V416c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V224H528zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm56 304c-13.3 0-24 10.7-24 24s10.7 24 24 24h48c13.3 0 24-10.7 24-24s-10.7-24-24-24H120zm128 0c-13.3 0-24 10.7-24 24s10.7 24 24 24H360c13.3 0 24-10.7 24-24s-10.7-24-24-24H248z\"],bell:[448,512,[128276,61602],\"f0f3\",\"M224 0c-17.7 0-32 14.3-32 32V51.2C119 66 64 130.6 64 208v25.4c0 45.4-15.5 89.5-43.8 124.9L5.3 377c-5.8 7.2-6.9 17.1-2.9 25.4S14.8 416 24 416H424c9.2 0 17.6-5.3 21.6-13.6s2.9-18.2-2.9-25.4l-14.9-18.6C399.5 322.9 384 278.8 384 233.4V208c0-77.4-55-142-128-156.8V32c0-17.7-14.3-32-32-32zm0 96c61.9 0 112 50.1 112 112v25.4c0 47.9 13.9 94.6 39.7 134.6H72.3C98.1 328 112 281.3 112 233.4V208c0-61.9 50.1-112 112-112zm64 352H224 160c0 17 6.7 33.3 18.7 45.3s28.3 18.7 45.3 18.7s33.3-6.7 45.3-18.7s18.7-28.3 18.7-45.3z\"],file:[384,512,[128196,128459,61462],\"f15b\",\"M320 464c8.8 0 16-7.2 16-16V160H256c-17.7 0-32-14.3-32-32V48H64c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16H320zM0 64C0 28.7 28.7 0 64 0H229.5c17 0 33.3 6.7 45.3 18.7l90.5 90.5c12 12 18.7 28.3 18.7 45.3V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64z\"],hospital:[640,512,[127973,62589,\"hospital-alt\",\"hospital-wide\"],\"f0f8\",\"M232 0c-39.8 0-72 32.2-72 72v8H72C32.2 80 0 112.2 0 152V440c0 39.8 32.2 72 72 72h.2 .2 .2 .2 .2H73h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2H75h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2H77h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2H79h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2H82h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2H85h.2 .2 .2 .2H86h.2 .2 .2 .2H87h.2 .2 .2 .2H88h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2H98h.2 .2 .2 .2H99h.2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2 .2v0H456h8v0H568c39.8 0 72-32.2 72-72V152c0-39.8-32.2-72-72-72H480V72c0-39.8-32.2-72-72-72H232zM480 128h88c13.3 0 24 10.7 24 24v40H536c-13.3 0-24 10.7-24 24s10.7 24 24 24h56v48H536c-13.3 0-24 10.7-24 24s10.7 24 24 24h56V440c0 13.3-10.7 24-24 24H480V336 128zM72 128h88V464h-.1-.2-.2-.2H159h-.2-.2-.2H158h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H154h-.2-.2-.2H153h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H150h-.2-.2-.2H149h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H146h-.2-.2-.2H145h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H142h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H139h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H136h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H133h-.2-.2-.2-.2-.2-.2-.2-.2H131h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H128h-.2-.2-.2-.2-.2-.2-.2-.2H126h-.2-.2-.2-.2-.2-.2-.2-.2H124h-.2-.2-.2-.2-.2-.2-.2-.2H122h-.2-.2-.2-.2-.2-.2-.2-.2H120h-.2-.2-.2-.2-.2-.2-.2-.2H118h-.2-.2-.2-.2-.2-.2-.2-.2H116h-.2-.2-.2-.2-.2-.2-.2-.2H114h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H111h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H108h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H105h-.2-.2-.2-.2H104h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H100h-.2-.2-.2-.2H99h-.2-.2-.2-.2H98h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H88h-.2-.2-.2-.2H87h-.2-.2-.2-.2H86h-.2-.2-.2-.2H85h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H82h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H79h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H77h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H75h-.2-.2-.2-.2-.2-.2-.2-.2-.2-.2H73h-.2-.2-.2-.2-.2H72c-13.2 0-24-10.7-24-24V336h56c13.3 0 24-10.7 24-24s-10.7-24-24-24H48V240h56c13.3 0 24-10.7 24-24s-10.7-24-24-24H48V152c0-13.3 10.7-24 24-24zM208 72c0-13.3 10.7-24 24-24H408c13.3 0 24 10.7 24 24V336 464H368V400c0-26.5-21.5-48-48-48s-48 21.5-48 48v64H208V72zm88 24v24H272c-8.8 0-16 7.2-16 16v16c0 8.8 7.2 16 16 16h24v24c0 8.8 7.2 16 16 16h16c8.8 0 16-7.2 16-16V168h24c8.8 0 16-7.2 16-16V136c0-8.8-7.2-16-16-16H344V96c0-8.8-7.2-16-16-16H312c-8.8 0-16 7.2-16 16z\"],\"chess-rook\":[448,512,[9820],\"f447\",\"M80 80V192c0 2.5 1.2 4.9 3.2 6.4l51.2 38.4c6.8 5.1 10.4 13.4 9.5 21.9L133.5 352H85.2l9.4-85L54.4 236.8C40.3 226.2 32 209.6 32 192V72c0-22.1 17.9-40 40-40H376c22.1 0 40 17.9 40 40V192c0 17.6-8.3 34.2-22.4 44.8L353.4 267l9.4 85H314.5l-10.4-93.3c-.9-8.4 2.7-16.8 9.5-21.9l51.2-38.4c2-1.5 3.2-3.9 3.2-6.4V80H304v24c0 13.3-10.7 24-24 24s-24-10.7-24-24V80H192v24c0 13.3-10.7 24-24 24s-24-10.7-24-24V80H80zm4.7 384H363.3l-16.6-32H101.2L84.7 464zm271.9-80c12 0 22.9 6.7 28.4 17.3l26.5 51.2c3 5.8 4.6 12.2 4.6 18.7c0 22.5-18.2 40.8-40.8 40.8H72.8C50.2 512 32 493.8 32 471.2c0-6.5 1.6-12.9 4.6-18.7l26.5-51.2C68.5 390.7 79.5 384 91.5 384h265zM208 288c-8.8 0-16-7.2-16-16V224c0-17.7 14.3-32 32-32s32 14.3 32 32v48c0 8.8-7.2 16-16 16H208z\"],\"star-half\":[576,512,[61731],\"f089\",\"M293.3 .6c10.9 2.5 18.6 12.2 18.6 23.4V408.7c0 8.9-4.9 17-12.7 21.2L151 509.1c-8.1 4.3-17.9 3.7-25.3-1.7s-11.2-14.5-9.7-23.5l26.2-155.6L31.1 218.3c-6.5-6.4-8.7-15.9-5.9-24.5s10.3-14.9 19.3-16.3l153.2-22.6L266.3 13.5c4.9-10.1 16.1-15.4 27-12.9zM263.9 128.4l-28.6 58.8c-3.5 7.1-10.2 12.1-18.1 13.3L99 217.9 184.9 303c5.5 5.5 8.1 13.3 6.8 21L171.4 443.7l92.5-49.4V128.4z\"],\"chess-king\":[448,512,[9818],\"f43f\",\"M248 24c0-13.3-10.7-24-24-24s-24 10.7-24 24V56H168c-13.3 0-24 10.7-24 24s10.7 24 24 24h32v40H59.6C26.7 144 0 170.7 0 203.6c0 8.2 1.7 16.3 4.9 23.8L59.1 352h52.3L49 208.2c-.6-1.5-1-3-1-4.6c0-6.4 5.2-11.6 11.6-11.6H224 388.4c6.4 0 11.6 5.2 11.6 11.6c0 1.6-.3 3.2-1 4.6L336.5 352h52.3l54.2-124.6c3.3-7.5 4.9-15.6 4.9-23.8c0-32.9-26.7-59.6-59.6-59.6H248V104h32c13.3 0 24-10.7 24-24s-10.7-24-24-24H248V24zM101.2 432H346.8l16.6 32H84.7l16.6-32zm283.7-30.7c-5.5-10.6-16.5-17.3-28.4-17.3H91.5c-12 0-22.9 6.7-28.4 17.3L36.6 452.5c-3 5.8-4.6 12.2-4.6 18.7C32 493.8 50.2 512 72.8 512H375.2c22.5 0 40.8-18.2 40.8-40.8c0-6.5-1.6-12.9-4.6-18.7l-26.5-51.2z\"],\"circle-user\":[512,512,[62142,\"user-circle\"],\"f2bd\",\"M406.5 399.6C387.4 352.9 341.5 320 288 320H224c-53.5 0-99.4 32.9-118.5 79.6C69.9 362.2 48 311.7 48 256C48 141.1 141.1 48 256 48s208 93.1 208 208c0 55.7-21.9 106.2-57.5 143.6zm-40.1 32.7C334.4 452.4 296.6 464 256 464s-78.4-11.6-110.5-31.7c7.3-36.7 39.7-64.3 78.5-64.3h64c38.8 0 71.2 27.6 78.5 64.3zM256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm0-272a40 40 0 1 1 0-80 40 40 0 1 1 0 80zm-88-40a88 88 0 1 0 176 0 88 88 0 1 0 -176 0z\"],copy:[448,512,[],\"f0c5\",\"M384 336H192c-8.8 0-16-7.2-16-16V64c0-8.8 7.2-16 16-16l140.1 0L400 115.9V320c0 8.8-7.2 16-16 16zM192 384H384c35.3 0 64-28.7 64-64V115.9c0-12.7-5.1-24.9-14.1-33.9L366.1 14.1c-9-9-21.2-14.1-33.9-14.1H192c-35.3 0-64 28.7-64 64V320c0 35.3 28.7 64 64 64zM64 128c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64H256c35.3 0 64-28.7 64-64V416H272v32c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V192c0-8.8 7.2-16 16-16H96V128H64z\"],\"share-from-square\":[576,512,[61509,\"share-square\"],\"f14d\",\"M400 255.4V240 208c0-8.8-7.2-16-16-16H352 336 289.5c-50.9 0-93.9 33.5-108.3 79.6c-3.3-9.4-5.2-19.8-5.2-31.6c0-61.9 50.1-112 112-112h48 16 32c8.8 0 16-7.2 16-16V80 64.6L506 160 400 255.4zM336 240h16v48c0 17.7 14.3 32 32 32h3.7c7.9 0 15.5-2.9 21.4-8.2l139-125.1c7.6-6.8 11.9-16.5 11.9-26.7s-4.3-19.9-11.9-26.7L409.9 8.9C403.5 3.2 395.3 0 386.7 0C367.5 0 352 15.5 352 34.7V80H336 304 288c-88.4 0-160 71.6-160 160c0 60.4 34.6 99.1 63.9 120.9c5.9 4.4 11.5 8.1 16.7 11.2c4.4 2.7 8.5 4.9 11.9 6.6c3.4 1.7 6.2 3 8.2 3.9c2.2 1 4.6 1.4 7.1 1.4h2.5c9.8 0 17.8-8 17.8-17.8c0-7.8-5.3-14.7-11.6-19.5l0 0c-.4-.3-.7-.5-1.1-.8c-1.7-1.1-3.4-2.5-5-4.1c-.8-.8-1.7-1.6-2.5-2.6s-1.6-1.9-2.4-2.9c-1.8-2.5-3.5-5.3-5-8.5c-2.6-6-4.3-13.3-4.3-22.4c0-36.1 29.3-65.5 65.5-65.5H304h32zM72 32C32.2 32 0 64.2 0 104V440c0 39.8 32.2 72 72 72H408c39.8 0 72-32.2 72-72V376c0-13.3-10.7-24-24-24s-24 10.7-24 24v64c0 13.3-10.7 24-24 24H72c-13.3 0-24-10.7-24-24V104c0-13.3 10.7-24 24-24h64c13.3 0 24-10.7 24-24s-10.7-24-24-24H72z\"],copyright:[512,512,[169],\"f1f9\",\"M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM199.4 312.6c-31.2-31.2-31.2-81.9 0-113.1s81.9-31.2 113.1 0c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9c-50-50-131-50-181 0s-50 131 0 181s131 50 181 0c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0c-31.2 31.2-81.9 31.2-113.1 0z\"],map:[576,512,[128506,62072],\"f279\",\"M565.6 36.2C572.1 40.7 576 48.1 576 56V392c0 10-6.2 18.9-15.5 22.4l-168 64c-5.2 2-10.9 2.1-16.1 .3L192.5 417.5l-160 61c-7.4 2.8-15.7 1.8-22.2-2.7S0 463.9 0 456V120c0-10 6.1-18.9 15.5-22.4l168-64c5.2-2 10.9-2.1 16.1-.3L383.5 94.5l160-61c7.4-2.8 15.7-1.8 22.2 2.7zM48 136.5V421.2l120-45.7V90.8L48 136.5zM360 422.7V137.3l-144-48V374.7l144 48zm48-1.5l120-45.7V90.8L408 136.5V421.2z\"],\"bell-slash\":[640,512,[128277,61943],\"f1f6\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L542.6 400c2.7-7.8 1.3-16.5-3.9-23l-14.9-18.6C495.5 322.9 480 278.8 480 233.4V208c0-77.4-55-142-128-156.8V32c0-17.7-14.3-32-32-32s-32 14.3-32 32V51.2c-42.6 8.6-79 34.2-102 69.3L38.8 5.1zM224 150.3C243.6 117.7 279.3 96 320 96c61.9 0 112 50.1 112 112v25.4c0 32.7 6.4 64.8 18.7 94.5L224 150.3zM406.2 416l-60.9-48H168.3c21.2-32.8 34.4-70.3 38.4-109.1L160 222.1v11.4c0 45.4-15.5 89.5-43.8 124.9L101.3 377c-5.8 7.2-6.9 17.1-2.9 25.4s12.4 13.6 21.6 13.6H406.2zM384 448H320 256c0 17 6.7 33.3 18.7 45.3s28.3 18.7 45.3 18.7s33.3-6.7 45.3-18.7s18.7-28.3 18.7-45.3z\"],\"hand-lizard\":[512,512,[],\"f258\",\"M72 112c-13.3 0-24 10.7-24 24s10.7 24 24 24H240c35.3 0 64 28.7 64 64s-28.7 64-64 64H136c-13.3 0-24 10.7-24 24s10.7 24 24 24H288c4.5 0 8.9 1.3 12.7 3.6l64 40c7 4.4 11.3 12.1 11.3 20.4v24c0 13.3-10.7 24-24 24s-24-10.7-24-24V413.3L281.1 384H136c-39.8 0-72-32.2-72-72s32.2-72 72-72H240c8.8 0 16-7.2 16-16s-7.2-16-16-16H72c-39.8 0-72-32.2-72-72S32.2 64 72 64H281.6c46.7 0 90.9 21.5 119.7 58.3l78.4 100.1c20.9 26.7 32.3 59.7 32.3 93.7V424c0 13.3-10.7 24-24 24s-24-10.7-24-24V316.1c0-23.2-7.8-45.8-22.1-64.1L363.5 151.9c-19.7-25.2-49.9-39.9-81.9-39.9H72z\"],\"face-smile\":[512,512,[128578,\"smile\"],\"f118\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm177.6 62.1C192.8 334.5 218.8 352 256 352s63.2-17.5 78.4-33.9c9-9.7 24.2-10.4 33.9-1.4s10.4 24.2 1.4 33.9c-22 23.8-60 49.4-113.6 49.4s-91.7-25.5-113.6-49.4c-9-9.7-8.4-24.9 1.4-33.9s24.9-8.4 33.9 1.4zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"hand-peace\":[512,512,[9996],\"f25b\",\"M250.8 1.4c-35.2-3.7-66.6 21.8-70.3 57L174 119 156.7 69.6C145 36.3 108.4 18.8 75.1 30.5S24.2 78.8 35.9 112.1L88.7 262.2C73.5 276.7 64 297.3 64 320v0 24c0 92.8 75.2 168 168 168h48c92.8 0 168-75.2 168-168V272 256 224c0-35.3-28.7-64-64-64c-7.9 0-15.4 1.4-22.4 4c-10.4-21.3-32.3-36-57.6-36c-.7 0-1.5 0-2.2 0l5.9-56.3c3.7-35.2-21.8-66.6-57-70.3zm-.2 155.4C243.9 166.9 240 179 240 192v48c0 .7 0 1.4 0 2c-5.1-1.3-10.5-2-16-2h-7.4l-5.4-15.3 17-161.3c.9-8.8 8.8-15.2 17.6-14.2s15.2 8.8 14.2 17.6l-9.5 90.1zM111.4 85.6L165.7 240H144c-4 0-8 .3-11.9 .9L81.2 96.2c-2.9-8.3 1.5-17.5 9.8-20.4s17.5 1.5 20.4 9.8zM288 192c0-8.8 7.2-16 16-16s16 7.2 16 16v32 16c0 8.8-7.2 16-16 16s-16-7.2-16-16V192zm38.4 108c10.4 21.3 32.3 36 57.6 36c5.5 0 10.9-.7 16-2v10c0 66.3-53.7 120-120 120H232c-66.3 0-120-53.7-120-120l0-24 0 0c0-17.7 14.3-32 32-32h80c8.8 0 16 7.2 16 16s-7.2 16-16 16H184c-13.3 0-24 10.7-24 24s10.7 24 24 24h40c35.3 0 64-28.7 64-64c0-.7 0-1.4 0-2c5.1 1.3 10.5 2 16 2c7.9 0 15.4-1.4 22.4-4zM400 272c0 8.8-7.2 16-16 16s-16-7.2-16-16V240 224c0-8.8 7.2-16 16-16s16 7.2 16 16v32 16z\"],\"face-grin-hearts\":[512,512,[128525,\"grin-hearts\"],\"f584\",\"M464 256A208 208 0 1 0 48 256a208 208 0 1 0 416 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm349.5 52.4c18.7-4.4 35.9 12 25.5 28.1C350.4 374.6 306.3 400 255.9 400s-94.5-25.4-119.1-63.5c-10.4-16.1 6.8-32.5 25.5-28.1c28.9 6.8 60.5 10.5 93.6 10.5s64.7-3.7 93.6-10.5zM215.3 137.1c17.8 4.8 28.4 23.1 23.6 40.8l-17.4 65c-2.3 8.5-11.1 13.6-19.6 11.3l-65.1-17.4c-17.8-4.8-28.4-23.1-23.6-40.8s23.1-28.4 40.8-23.6l16.1 4.3 4.3-16.1c4.8-17.8 23.1-28.4 40.8-23.6zm122.3 23.6l4.3 16.1 16.1-4.3c17.8-4.8 36.1 5.8 40.8 23.6s-5.8 36.1-23.6 40.8l-65.1 17.4c-8.5 2.3-17.3-2.8-19.6-11.3l-17.4-65c-4.8-17.8 5.8-36.1 23.6-40.8s36.1 5.8 40.9 23.6z\"],building:[384,512,[127970,61687],\"f1ad\",\"M64 48c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16h80V400c0-26.5 21.5-48 48-48s48 21.5 48 48v64h80c8.8 0 16-7.2 16-16V64c0-8.8-7.2-16-16-16H64zM0 64C0 28.7 28.7 0 64 0H320c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zm88 40c0-8.8 7.2-16 16-16h48c8.8 0 16 7.2 16 16v48c0 8.8-7.2 16-16 16H104c-8.8 0-16-7.2-16-16V104zM232 88h48c8.8 0 16 7.2 16 16v48c0 8.8-7.2 16-16 16H232c-8.8 0-16-7.2-16-16V104c0-8.8 7.2-16 16-16zM88 232c0-8.8 7.2-16 16-16h48c8.8 0 16 7.2 16 16v48c0 8.8-7.2 16-16 16H104c-8.8 0-16-7.2-16-16V232zm144-16h48c8.8 0 16 7.2 16 16v48c0 8.8-7.2 16-16 16H232c-8.8 0-16-7.2-16-16V232c0-8.8 7.2-16 16-16z\"],\"face-grin-beam-sweat\":[512,512,[128517,\"grin-beam-sweat\"],\"f583\",\"M476.8 126.3C497.1 120.8 512 102.7 512 81c0-20-28.6-60.4-41.6-77.7c-3.2-4.4-9.6-4.4-12.8 0c-9.5 12.6-27.1 37.2-36 57.5c-.3 .7-.6 1.4-.9 2.1C417.8 69.7 416 76 416 81c0 26 21.5 47 48 47c4.4 0 8.7-.6 12.8-1.7zM395.4 41.2C355.3 15.2 307.4 0 256 0C114.6 0 0 114.6 0 256S114.6 512 256 512s256-114.6 256-256c0-35.8-7.3-69.9-20.6-100.8c-8.6 3.1-17.8 4.8-27.4 4.8c-8.9 0-17.6-1.5-25.7-4.2C454.7 185.5 464 219.7 464 256c0 114.9-93.1 208-208 208S48 370.9 48 256S141.1 48 256 48c48.7 0 93.4 16.7 128.9 44.7c-.6-3.8-.9-7.7-.9-11.7c0-11.4 3.8-22.4 7.1-30.5c1.3-3.1 2.7-6.2 4.3-9.3zM375 336.5c10.4-16.1-6.8-32.5-25.5-28.1c-28.9 6.8-60.5 10.5-93.6 10.5s-64.7-3.7-93.6-10.5c-18.7-4.4-35.9 12-25.5 28.1c24.6 38.1 68.7 63.5 119.1 63.5s94.5-25.4 119.1-63.5zM217.6 228.8l0 0 0 0 0 0c2.1 2.8 5.7 3.9 8.9 2.8s5.5-4.1 5.5-7.6c0-17.9-6.7-35.6-16.6-48.8c-9.8-13-23.9-23.2-39.4-23.2s-29.6 10.2-39.4 23.2C126.7 188.4 120 206.1 120 224c0 3.4 2.2 6.5 5.5 7.6s6.9 0 8.9-2.8l0 0 0 0 0 0 .2-.2c.2-.2 .4-.5 .7-.9c.6-.8 1.6-2 2.8-3.4c2.5-2.8 6-6.6 10.2-10.3c8.8-7.8 18.8-14 27.7-14s18.9 6.2 27.7 14c4.2 3.7 7.7 7.5 10.2 10.3c1.2 1.4 2.2 2.6 2.8 3.4c.3 .4 .6 .7 .7 .9l.2 .2 0 0zm160 0l0 0 0 0c2.1 2.8 5.7 3.9 8.9 2.8s5.5-4.1 5.5-7.6c0-17.9-6.7-35.6-16.6-48.8c-9.8-13-23.9-23.2-39.4-23.2s-29.6 10.2-39.4 23.2C286.7 188.4 280 206.1 280 224c0 3.4 2.2 6.5 5.5 7.6s6.9 0 8.9-2.8l0 0 0 0 0 0 .2-.2c.2-.2 .4-.5 .7-.9c.6-.8 1.6-2 2.8-3.4c2.5-2.8 6-6.6 10.2-10.3c8.8-7.8 18.8-14 27.7-14s18.9 6.2 27.7 14c4.2 3.7 7.7 7.5 10.2 10.3c1.2 1.4 2.2 2.6 2.8 3.4c.3 .4 .6 .7 .7 .9l.2 .2 0 0 0 0z\"],moon:[384,512,[127769,9214],\"f186\",\"M144.7 98.7c-21 34.1-33.1 74.3-33.1 117.3c0 98 62.8 181.4 150.4 211.7c-12.4 2.8-25.3 4.3-38.6 4.3C126.6 432 48 353.3 48 256c0-68.9 39.4-128.4 96.8-157.3zm62.1-66C91.1 41.2 0 137.9 0 256C0 379.7 100 480 223.5 480c47.8 0 92-15 128.4-40.6c1.9-1.3 3.7-2.7 5.5-4c4.8-3.6 9.4-7.4 13.9-11.4c2.7-2.4 5.3-4.8 7.9-7.3c5-4.9 6.3-12.5 3.1-18.7s-10.1-9.7-17-8.5c-3.7 .6-7.4 1.2-11.1 1.6c-5 .5-10.1 .9-15.3 1c-1.2 0-2.5 0-3.7 0c-.1 0-.2 0-.3 0c-96.8-.2-175.2-78.9-175.2-176c0-54.8 24.9-103.7 64.1-136c1-.9 2.1-1.7 3.2-2.6c4-3.2 8.2-6.2 12.5-9c3.1-2 6.3-4 9.6-5.8c6.1-3.5 9.2-10.5 7.7-17.3s-7.3-11.9-14.3-12.5c-3.6-.3-7.1-.5-10.7-.6c-2.7-.1-5.5-.1-8.2-.1c-3.3 0-6.5 .1-9.8 .2c-2.3 .1-4.6 .2-6.9 .4z\"],calendar:[448,512,[128197,128198],\"f133\",\"M152 24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H64C28.7 64 0 92.7 0 128v16 48V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V192 144 128c0-35.3-28.7-64-64-64H344V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H152V24zM48 192H400V448c0 8.8-7.2 16-16 16H64c-8.8 0-16-7.2-16-16V192z\"],\"face-grin-tongue-wink\":[512,512,[128540,\"grin-tongue-wink\"],\"f58b\",\"M348.3 442.4c2.4-8.4 3.7-17.3 3.7-26.4V363.5c8.8-8 16.6-17.1 23-27c10.4-16.1-6.8-32.5-25.5-28.1c-28.9 6.8-60.5 10.5-93.6 10.5s-64.7-3.7-93.6-10.5c-18.7-4.4-35.9 12-25.5 28.1c6.5 10 14.3 19.1 23.1 27.1V416c0 9.2 1.3 18 3.7 26.4C95.1 408.4 48 337.7 48 256C48 141.1 141.1 48 256 48s208 93.1 208 208c0 81.7-47.1 152.4-115.7 186.4zM256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM159.6 220c10.6 0 19.9 3.8 25.4 9.7c7.6 8.1 20.2 8.5 28.3 .9s8.5-20.2 .9-28.3C199.7 186.8 179 180 159.6 180s-40.1 6.8-54.6 22.3c-7.6 8.1-7.1 20.7 .9 28.3s20.7 7.1 28.3-.9c5.5-5.8 14.8-9.7 25.4-9.7zm176.7 12a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm-.4-72a48 48 0 1 1 0 96 48 48 0 1 1 0-96zm0 128a80 80 0 1 0 0-160 80 80 0 1 0 0 160zM320 416c0 35.3-28.7 64-64 64s-64-28.7-64-64V378.6c0-14.7 11.9-26.6 26.6-26.6h2c11.3 0 21.1 7.9 23.6 18.9c2.8 12.6 20.8 12.6 23.6 0c2.5-11.1 12.3-18.9 23.6-18.9h2c14.7 0 26.6 11.9 26.6 26.6V416z\"],clone:[512,512,[],\"f24d\",\"M64 464H288c8.8 0 16-7.2 16-16V384h48v64c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V224c0-35.3 28.7-64 64-64h64v48H64c-8.8 0-16 7.2-16 16V448c0 8.8 7.2 16 16 16zM224 304H448c8.8 0 16-7.2 16-16V64c0-8.8-7.2-16-16-16H224c-8.8 0-16 7.2-16 16V288c0 8.8 7.2 16 16 16zm-64-16V64c0-35.3 28.7-64 64-64H448c35.3 0 64 28.7 64 64V288c0 35.3-28.7 64-64 64H224c-35.3 0-64-28.7-64-64z\"],\"face-angry\":[512,512,[128544,\"angry\"],\"f556\",\"M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm72.4-118.5c9.7-9 10.2-24.2 1.2-33.9C315.3 344.3 290.6 328 256 328s-59.3 16.3-73.5 31.6c-9 9.7-8.5 24.9 1.2 33.9s24.9 8.5 33.9-1.2c7.4-7.9 20-16.4 38.5-16.4s31.1 8.5 38.5 16.4c9 9.7 24.2 10.2 33.9 1.2zM176.4 272c17.7 0 32-14.3 32-32c0-1.5-.1-3-.3-4.4l10.9 3.6c8.4 2.8 17.4-1.7 20.2-10.1s-1.7-17.4-10.1-20.2l-96-32c-8.4-2.8-17.4 1.7-20.2 10.1s1.7 17.4 10.1 20.2l30.7 10.2c-5.8 5.8-9.3 13.8-9.3 22.6c0 17.7 14.3 32 32 32zm192-32c0-8.9-3.6-17-9.5-22.8l30.2-10.1c8.4-2.8 12.9-11.9 10.1-20.2s-11.9-12.9-20.2-10.1l-96 32c-8.4 2.8-12.9 11.9-10.1 20.2s11.9 12.9 20.2 10.1l11.7-3.9c-.2 1.5-.3 3.1-.3 4.7c0 17.7 14.3 32 32 32s32-14.3 32-32z\"],\"rectangle-xmark\":[512,512,[62164,\"rectangle-times\",\"times-rectangle\",\"window-close\"],\"f410\",\"M64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H448c8.8 0 16-7.2 16-16V96c0-8.8-7.2-16-16-16H64zM0 96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm175 79c9.4-9.4 24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9z\"],\"paper-plane\":[512,512,[61913],\"f1d8\",\"M16.1 260.2c-22.6 12.9-20.5 47.3 3.6 57.3L160 376V479.3c0 18.1 14.6 32.7 32.7 32.7c9.7 0 18.9-4.3 25.1-11.8l62-74.3 123.9 51.6c18.9 7.9 40.8-4.5 43.9-24.7l64-416c1.9-12.1-3.4-24.3-13.5-31.2s-23.3-7.5-34-1.4l-448 256zm52.1 25.5L409.7 90.6 190.1 336l1.2 1L68.2 285.7zM403.3 425.4L236.7 355.9 450.8 116.6 403.3 425.4z\"],\"life-ring\":[512,512,[],\"f1cd\",\"M385.1 419.1C349.7 447.2 304.8 464 256 464s-93.7-16.8-129.1-44.9l80.4-80.4c14.3 8.4 31 13.3 48.8 13.3s34.5-4.8 48.8-13.3l80.4 80.4zm68.1 .2C489.9 374.9 512 318.1 512 256s-22.1-118.9-58.8-163.3L465 81c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L419.3 58.8C374.9 22.1 318.1 0 256 0S137.1 22.1 92.7 58.8L81 47c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9L58.8 92.7C22.1 137.1 0 193.9 0 256s22.1 118.9 58.8 163.3L47 431c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l11.8-11.8C137.1 489.9 193.9 512 256 512s118.9-22.1 163.3-58.8L431 465c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-11.8-11.8zm-34.1-34.1l-80.4-80.4c8.4-14.3 13.3-31 13.3-48.8s-4.8-34.5-13.3-48.8l80.4-80.4C447.2 162.3 464 207.2 464 256s-16.8 93.7-44.9 129.1zM385.1 92.9l-80.4 80.4c-14.3-8.4-31-13.3-48.8-13.3s-34.5 4.8-48.8 13.3L126.9 92.9C162.3 64.8 207.2 48 256 48s93.7 16.8 129.1 44.9zM173.3 304.8L92.9 385.1C64.8 349.7 48 304.8 48 256s16.8-93.7 44.9-129.1l80.4 80.4c-8.4 14.3-13.3 31-13.3 48.8s4.8 34.5 13.3 48.8zM208 256a48 48 0 1 1 96 0 48 48 0 1 1 -96 0z\"],\"face-grimace\":[512,512,[128556,\"grimace\"],\"f57f\",\"M256 48a208 208 0 1 0 0 416 208 208 0 1 0 0-416zM512 256A256 256 0 1 1 0 256a256 256 0 1 1 512 0zM168 320c-13.3 0-24 10.7-24 24s10.7 24 24 24h8V320h-8zm40 48h32V320H208v48zm96 0V320H272v48h32zm32 0h8c13.3 0 24-10.7 24-24s-10.7-24-24-24h-8v48zM168 288H344c30.9 0 56 25.1 56 56s-25.1 56-56 56H168c-30.9 0-56-25.1-56-56s25.1-56 56-56zm-23.6-80a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"calendar-minus\":[448,512,[],\"f272\",\"M128 0c13.3 0 24 10.7 24 24V64H296V24c0-13.3 10.7-24 24-24s24 10.7 24 24V64h40c35.3 0 64 28.7 64 64v16 48V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V192 144 128C0 92.7 28.7 64 64 64h40V24c0-13.3 10.7-24 24-24zM400 192H48V448c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16V192zM296 352H152c-13.3 0-24-10.7-24-24s10.7-24 24-24H296c13.3 0 24 10.7 24 24s-10.7 24-24 24z\"],\"circle-xmark\":[512,512,[61532,\"times-circle\",\"xmark-circle\"],\"f057\",\"M256 48a208 208 0 1 1 0 416 208 208 0 1 1 0-416zm0 464A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM175 175c-9.4 9.4-9.4 24.6 0 33.9l47 47-47 47c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l47-47 47 47c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-47-47 47-47c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-47 47-47-47c-9.4-9.4-24.6-9.4-33.9 0z\"],\"thumbs-up\":[512,512,[128077,61575],\"f164\",\"M323.8 34.8c-38.2-10.9-78.1 11.2-89 49.4l-5.7 20c-3.7 13-10.4 25-19.5 35l-51.3 56.4c-8.9 9.8-8.2 25 1.6 33.9s25 8.2 33.9-1.6l51.3-56.4c14.1-15.5 24.4-34 30.1-54.1l5.7-20c3.6-12.7 16.9-20.1 29.7-16.5s20.1 16.9 16.5 29.7l-5.7 20c-5.7 19.9-14.7 38.7-26.6 55.5c-5.2 7.3-5.8 16.9-1.7 24.9s12.3 13 21.3 13L448 224c8.8 0 16 7.2 16 16c0 6.8-4.3 12.7-10.4 15c-7.4 2.8-13 9-14.9 16.7s.1 15.8 5.3 21.7c2.5 2.8 4 6.5 4 10.6c0 7.8-5.6 14.3-13 15.7c-8.2 1.6-15.1 7.3-18 15.2s-1.6 16.7 3.6 23.3c2.1 2.7 3.4 6.1 3.4 9.9c0 6.7-4.2 12.6-10.2 14.9c-11.5 4.5-17.7 16.9-14.4 28.8c.4 1.3 .6 2.8 .6 4.3c0 8.8-7.2 16-16 16H286.5c-12.6 0-25-3.7-35.5-10.7l-61.7-41.1c-11-7.4-25.9-4.4-33.3 6.7s-4.4 25.9 6.7 33.3l61.7 41.1c18.4 12.3 40 18.8 62.1 18.8H384c34.7 0 62.9-27.6 64-62c14.6-11.7 24-29.7 24-50c0-4.5-.5-8.8-1.3-13c15.4-11.7 25.3-30.2 25.3-51c0-6.5-1-12.8-2.8-18.7C504.8 273.7 512 257.7 512 240c0-35.3-28.6-64-64-64l-92.3 0c4.7-10.4 8.7-21.2 11.8-32.2l5.7-20c10.9-38.2-11.2-78.1-49.4-89zM32 192c-17.7 0-32 14.3-32 32V448c0 17.7 14.3 32 32 32H96c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32H32z\"],\"window-minimize\":[512,512,[128469],\"f2d1\",\"M24 432c-13.3 0-24 10.7-24 24s10.7 24 24 24H488c13.3 0 24-10.7 24-24s-10.7-24-24-24H24z\"],\"square-full\":[512,512,[128997,128998,128999,129e3,129001,129002,129003,11035,11036],\"f45c\",\"M464 48V464H48V48H464zM48 0H0V48 464v48H48 464h48V464 48 0H464 48z\"],\"note-sticky\":[448,512,[62026,\"sticky-note\"],\"f249\",\"M64 80c-8.8 0-16 7.2-16 16V416c0 8.8 7.2 16 16 16H288V352c0-17.7 14.3-32 32-32h80V96c0-8.8-7.2-16-16-16H64zM288 480H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V320v5.5c0 17-6.7 33.3-18.7 45.3l-90.5 90.5c-12 12-28.3 18.7-45.3 18.7H288z\"],\"face-sad-tear\":[512,512,[128546,\"sad-tear\"],\"f5b4\",\"M175.9 448c-35-.1-65.5-22.6-76-54.6C67.6 356.8 48 308.7 48 256C48 141.1 141.1 48 256 48s208 93.1 208 208s-93.1 208-208 208c-28.4 0-55.5-5.7-80.1-16zM0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zM128 369c0 26 21.5 47 48 47s48-21 48-47c0-20-28.4-60.4-41.6-77.7c-3.2-4.4-9.6-4.4-12.8 0C156.6 308.6 128 349 128 369zm128-65c-13.3 0-24 10.7-24 24s10.7 24 24 24c30.7 0 58.7 11.5 80 30.6c9.9 8.8 25 8 33.9-1.9s8-25-1.9-33.9C338.3 320.2 299 304 256 304zm47.6-96a32 32 0 1 0 64 0 32 32 0 1 0 -64 0zm-128 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"hand-point-left\":[512,512,[],\"f0a5\",\"M64 128l177.6 0c-1 5.2-1.6 10.5-1.6 16l0 16-32 0L64 160c-8.8 0-16-7.2-16-16s7.2-16 16-16zm224 16c0-17.7 14.3-32 32-32c0 0 0 0 0 0l24 0c66.3 0 120 53.7 120 120l0 48c0 52.5-33.7 97.1-80.7 113.4c.5-3.1 .7-6.2 .7-9.4c0-20-9.2-37.9-23.6-49.7c4.9-9 7.6-19.4 7.6-30.3c0-15.1-5.3-29-14-40c8.8-11 14-24.9 14-40l0-40c0-13.3-10.7-24-24-24s-24 10.7-24 24l0 40c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-40 0-40zm32-80s0 0 0 0c-18 0-34.6 6-48 16L64 80C28.7 80 0 108.7 0 144s28.7 64 64 64l82 0c-1.3 5.1-2 10.5-2 16c0 25.3 14.7 47.2 36 57.6c-2.6 7-4 14.5-4 22.4c0 20 9.2 37.9 23.6 49.7c-4.9 9-7.6 19.4-7.6 30.3c0 35.3 28.7 64 64 64l64 0 24 0c92.8 0 168-75.2 168-168l0-48c0-92.8-75.2-168-168-168l-24 0zM256 400c-8.8 0-16-7.2-16-16s7.2-16 16-16l48 0 16 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-64 0zM240 224c0 5.5 .7 10.9 2 16l-2 0-32 0c-8.8 0-16-7.2-16-16s7.2-16 16-16l32 0 0 16zm24 64l40 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-48 0-16 0c-8.8 0-16-7.2-16-16s7.2-16 16-16l24 0z\"]};!function(c){try{for(var l=arguments.length,s=new Array(1<l?l-1:0),a=1;a<l;a++)s[a-1]=arguments[a];c.apply(void 0,s)}catch(c){if(!h)throw c}}(function(){d(\"far\",p),d(\"fa-regular\",p)})}(),function(){\"use strict\";var c={},l={};try{\"undefined\"!=typeof window&&(c=window),\"undefined\"!=typeof document&&(l=document)}catch(c){}var s=(c.navigator||{}).userAgent,a=void 0===s?\"\":s,z=c,e=l;z.document,e.documentElement&&e.head&&\"function\"==typeof e.addEventListener&&e.createElement,~a.indexOf(\"MSIE\")||a.indexOf(\"Trident/\");function H(l,c){var s,a=Object.keys(l);return Object.getOwnPropertySymbols&&(s=Object.getOwnPropertySymbols(l),c&&(s=s.filter(function(c){return Object.getOwnPropertyDescriptor(l,c).enumerable})),a.push.apply(a,s)),a}function t(l){for(var c=1;c<arguments.length;c++){var s=null!=arguments[c]?arguments[c]:{};c%2?H(Object(s),!0).forEach(function(c){V(l,c,s[c])}):Object.getOwnPropertyDescriptors?Object.defineProperties(l,Object.getOwnPropertyDescriptors(s)):H(Object(s)).forEach(function(c){Object.defineProperty(l,c,Object.getOwnPropertyDescriptor(s,c))})}return l}function V(c,l,s){return l in c?Object.defineProperty(c,l,{value:s,enumerable:!0,configurable:!0,writable:!0}):c[l]=s,c}function r(c,l){(null==l||l>c.length)&&(l=c.length);for(var s=0,a=new Array(l);s<l;s++)a[s]=c[s];return a}var M=\"___FONT_AWESOME___\",h=function(){try{return!0}catch(c){return!1}}(),n=\"classic\",i=\"sharp\",m=[n,i];function o(c){return new Proxy(c,{get:function(c,l){return l in c?c[l]:c[n]}})}o((V(C={},n,{fa:\"solid\",fas:\"solid\",\"fa-solid\":\"solid\",far:\"regular\",\"fa-regular\":\"regular\",fal:\"light\",\"fa-light\":\"light\",fat:\"thin\",\"fa-thin\":\"thin\",fad:\"duotone\",\"fa-duotone\":\"duotone\",fab:\"brands\",\"fa-brands\":\"brands\",fak:\"kit\",fakd:\"kit\",\"fa-kit\":\"kit\",\"fa-kit-duotone\":\"kit\"}),V(C,i,{fa:\"solid\",fass:\"solid\",\"fa-solid\":\"solid\",fasr:\"regular\",\"fa-regular\":\"regular\",fasl:\"light\",\"fa-light\":\"light\",fast:\"thin\",\"fa-thin\":\"thin\"}),C));var f=o((V(v={},n,{solid:\"fas\",regular:\"far\",light:\"fal\",thin:\"fat\",duotone:\"fad\",brands:\"fab\",kit:\"fak\"}),V(v,i,{solid:\"fass\",regular:\"fasr\",light:\"fasl\",thin:\"fast\"}),v)),e=(o((V(s={},n,{fab:\"fa-brands\",fad:\"fa-duotone\",fak:\"fa-kit\",fal:\"fa-light\",far:\"fa-regular\",fas:\"fa-solid\",fat:\"fa-thin\"}),V(s,i,{fass:\"fa-solid\",fasr:\"fa-regular\",fasl:\"fa-light\",fast:\"fa-thin\"}),s)),o((V(c={},n,{\"fa-brands\":\"fab\",\"fa-duotone\":\"fad\",\"fa-kit\":\"fak\",\"fa-light\":\"fal\",\"fa-regular\":\"far\",\"fa-solid\":\"fas\",\"fa-thin\":\"fat\"}),V(c,i,{\"fa-solid\":\"fass\",\"fa-regular\":\"fasr\",\"fa-light\":\"fasl\",\"fa-thin\":\"fast\"}),c)),o((V(l={},n,{900:\"fas\",400:\"far\",normal:\"far\",300:\"fal\",100:\"fat\"}),V(l,i,{900:\"fass\",400:\"fasr\",300:\"fasl\",100:\"fast\"}),l)),[1,2,3,4,5,6,7,8,9,10]),a=e.concat([11,12,13,14,15,16,17,18,19,20]),C=\"duotone-group\",v=\"swap-opacity\",s=\"primary\",c=\"secondary\",l=new Set;Object.keys(f[n]).map(l.add.bind(l)),Object.keys(f[i]).map(l.add.bind(l));[].concat(m,function(c){if(Array.isArray(c))return r(c)}(l=l)||function(c){if(\"undefined\"!=typeof Symbol&&null!=c[Symbol.iterator]||null!=c[\"@@iterator\"])return Array.from(c)}(l)||function(c,l){if(c){if(\"string\"==typeof c)return r(c,l);var s=Object.prototype.toString.call(c).slice(8,-1);return\"Map\"===(s=\"Object\"===s&&c.constructor?c.constructor.name:s)||\"Set\"===s?Array.from(c):\"Arguments\"===s||/^(?:Ui|I)nt(?:8|16|32)(?:Clamped)?Array$/.test(s)?r(c,l):void 0}}(l)||function(){throw new TypeError(\"Invalid attempt to spread non-iterable instance.\\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method.\")}(),[\"2xs\",\"xs\",\"sm\",\"lg\",\"xl\",\"2xl\",\"beat\",\"border\",\"fade\",\"beat-fade\",\"bounce\",\"flip-both\",\"flip-horizontal\",\"flip-vertical\",\"flip\",\"fw\",\"inverse\",\"layers-counter\",\"layers-text\",\"layers\",\"li\",\"pull-left\",\"pull-right\",\"pulse\",\"rotate-180\",\"rotate-270\",\"rotate-90\",\"rotate-by\",\"shake\",\"spin-pulse\",\"spin-reverse\",\"spin\",\"stack-1x\",\"stack-2x\",\"stack\",\"ul\",C,v,s,c]).concat(e.map(function(c){return\"\".concat(c,\"x\")})).concat(a.map(function(c){return\"w-\".concat(c)}));z=z||{};z[M]||(z[M]={}),z[M].styles||(z[M].styles={}),z[M].hooks||(z[M].hooks={}),z[M].shims||(z[M].shims=[]);var L=z[M];function u(a){return Object.keys(a).reduce(function(c,l){var s=a[l];return!!s.icon?c[s.iconName]=s.icon:c[l]=s,c},{})}function d(c,l,s){var a=(2<arguments.length&&void 0!==s?s:{}).skipHooks,s=void 0!==a&&a,a=u(l);\"function\"!=typeof L.hooks.addPack||s?L.styles[c]=t(t({},L.styles[c]||{}),a):L.hooks.addPack(c,u(l)),\"fas\"===c&&d(\"fa\",l)}var p={0:[320,512,[],\"30\",\"M0 192C0 103.6 71.6 32 160 32s160 71.6 160 160V320c0 88.4-71.6 160-160 160S0 408.4 0 320V192zM160 96c-53 0-96 43-96 96V320c0 53 43 96 96 96s96-43 96-96V192c0-53-43-96-96-96z\"],1:[256,512,[],\"31\",\"M160 64c0-11.8-6.5-22.6-16.9-28.2s-23-5-32.8 1.6l-96 64C-.5 111.2-4.4 131 5.4 145.8s29.7 18.7 44.4 8.9L96 123.8V416H32c-17.7 0-32 14.3-32 32s14.3 32 32 32h96 96c17.7 0 32-14.3 32-32s-14.3-32-32-32H160V64z\"],2:[320,512,[],\"32\",\"M142.9 96c-21.5 0-42.2 8.5-57.4 23.8L54.6 150.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L40.2 74.5C67.5 47.3 104.4 32 142.9 32C223 32 288 97 288 177.1c0 38.5-15.3 75.4-42.5 102.6L109.3 416H288c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-12.9 0-24.6-7.8-29.6-19.8s-2.2-25.7 6.9-34.9L200.2 234.5c15.2-15.2 23.8-35.9 23.8-57.4c0-44.8-36.3-81.1-81.1-81.1z\"],3:[320,512,[],\"33\",\"M0 64C0 46.3 14.3 32 32 32H272c13.2 0 25 8.1 29.8 20.4s1.5 26.3-8.2 35.2L162.3 208H184c75.1 0 136 60.9 136 136s-60.9 136-136 136H105.4C63 480 24.2 456 5.3 418.1l-1.9-3.8c-7.9-15.8-1.5-35 14.3-42.9s35-1.5 42.9 14.3l1.9 3.8c8.1 16.3 24.8 26.5 42.9 26.5H184c39.8 0 72-32.2 72-72s-32.2-72-72-72H80c-13.2 0-25-8.1-29.8-20.4s-1.5-26.3 8.2-35.2L189.7 96H32C14.3 96 0 81.7 0 64z\"],4:[384,512,[],\"34\",\"M189 77.6c7.5-16 .7-35.1-15.3-42.6s-35.1-.7-42.6 15.3L3 322.4c-4.7 9.9-3.9 21.5 1.9 30.8S21 368 32 368H256v80c0 17.7 14.3 32 32 32s32-14.3 32-32V368h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H320V160c0-17.7-14.3-32-32-32s-32 14.3-32 32V304H82.4L189 77.6z\"],5:[320,512,[],\"35\",\"M32.5 58.3C35.3 43.1 48.5 32 64 32H256c17.7 0 32 14.3 32 32s-14.3 32-32 32H90.7L70.3 208H184c75.1 0 136 60.9 136 136s-60.9 136-136 136H100.5c-39.4 0-75.4-22.3-93-57.5l-4.1-8.2c-7.9-15.8-1.5-35 14.3-42.9s35-1.5 42.9 14.3l4.1 8.2c6.8 13.6 20.6 22.1 35.8 22.1H184c39.8 0 72-32.2 72-72s-32.2-72-72-72H32c-9.5 0-18.5-4.2-24.6-11.5s-8.6-16.9-6.9-26.2l32-176z\"],6:[320,512,[],\"36\",\"M232.4 84.7c11.4-13.5 9.7-33.7-3.8-45.1s-33.7-9.7-45.1 3.8L38.6 214.7C14.7 242.9 1.1 278.4 .1 315.2c0 1.4-.1 2.9-.1 4.3c0 .2 0 .3 0 .5c0 88.4 71.6 160 160 160s160-71.6 160-160c0-85.5-67.1-155.4-151.5-159.8l63.9-75.6zM256 320A96 96 0 1 1 64 320a96 96 0 1 1 192 0z\"],7:[320,512,[],\"37\",\"M0 64C0 46.3 14.3 32 32 32H288c11.5 0 22 6.1 27.7 16.1s5.7 22.2-.1 32.1l-224 384c-8.9 15.3-28.5 20.4-43.8 11.5s-20.4-28.5-11.5-43.8L232.3 96H32C14.3 96 0 81.7 0 64z\"],8:[320,512,[],\"38\",\"M304 160c0-70.7-57.3-128-128-128H144C73.3 32 16 89.3 16 160c0 34.6 13.7 66 36 89C20.5 272.3 0 309.8 0 352c0 70.7 57.3 128 128 128h64c70.7 0 128-57.3 128-128c0-42.2-20.5-79.7-52-103c22.3-23 36-54.4 36-89zM176.1 288H192c35.3 0 64 28.7 64 64s-28.7 64-64 64H128c-35.3 0-64-28.7-64-64s28.7-64 64-64h15.9c0 0 .1 0 .1 0h32c0 0 .1 0 .1 0zm0-64c0 0 0 0 0 0H144c0 0 0 0 0 0c-35.3 0-64-28.7-64-64c0-35.3 28.7-64 64-64h32c35.3 0 64 28.7 64 64c0 35.3-28.6 64-64 64z\"],9:[320,512,[],\"39\",\"M64 192a96 96 0 1 0 192 0A96 96 0 1 0 64 192zm87.5 159.8C67.1 347.4 0 277.5 0 192C0 103.6 71.6 32 160 32s160 71.6 160 160c0 2.6-.1 5.3-.2 7.9c-1.7 35.7-15.2 70-38.4 97.4l-145 171.4c-11.4 13.5-31.6 15.2-45.1 3.8s-15.2-31.6-3.8-45.1l63.9-75.6z\"],\"fill-drip\":[576,512,[],\"f576\",\"M41.4 9.4C53.9-3.1 74.1-3.1 86.6 9.4L168 90.7l53.1-53.1c28.1-28.1 73.7-28.1 101.8 0L474.3 189.1c28.1 28.1 28.1 73.7 0 101.8L283.9 481.4c-37.5 37.5-98.3 37.5-135.8 0L30.6 363.9c-37.5-37.5-37.5-98.3 0-135.8L122.7 136 41.4 54.6c-12.5-12.5-12.5-32.8 0-45.3zm176 221.3L168 181.3 75.9 273.4c-4.2 4.2-7 9.3-8.4 14.6H386.7l42.3-42.3c3.1-3.1 3.1-8.2 0-11.3L277.7 82.9c-3.1-3.1-8.2-3.1-11.3 0L213.3 136l49.4 49.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0zM512 512c-35.3 0-64-28.7-64-64c0-25.2 32.6-79.6 51.2-108.7c6-9.4 19.5-9.4 25.5 0C543.4 368.4 576 422.8 576 448c0 35.3-28.7 64-64 64z\"],\"arrows-to-circle\":[640,512,[],\"e4bd\",\"M9.4 9.4C21.9-3.1 42.1-3.1 54.6 9.4L160 114.7V96c0-17.7 14.3-32 32-32s32 14.3 32 32v96c0 4.3-.9 8.5-2.4 12.2c-1.6 3.7-3.8 7.3-6.9 10.3l-.1 .1c-3.1 3-6.6 5.3-10.3 6.9c-3.8 1.6-7.9 2.4-12.2 2.4H96c-17.7 0-32-14.3-32-32s14.3-32 32-32h18.7L9.4 54.6C-3.1 42.1-3.1 21.9 9.4 9.4zM256 256a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM114.7 352H96c-17.7 0-32-14.3-32-32s14.3-32 32-32h96 0l.1 0c8.8 0 16.7 3.6 22.5 9.3l.1 .1c3 3.1 5.3 6.6 6.9 10.3c1.6 3.8 2.4 7.9 2.4 12.2v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V397.3L54.6 502.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L114.7 352zM416 96c0-17.7 14.3-32 32-32s32 14.3 32 32v18.7L585.4 9.4c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3L525.3 160H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H448c-8.8 0-16.8-3.6-22.6-9.3l-.1-.1c-3-3.1-5.3-6.6-6.9-10.3s-2.4-7.8-2.4-12.2l0-.1v0V96zM525.3 352L630.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L480 397.3V416c0 17.7-14.3 32-32 32s-32-14.3-32-32V320v0c0 0 0-.1 0-.1c0-4.3 .9-8.4 2.4-12.2c1.6-3.8 3.9-7.3 6.9-10.4c5.8-5.8 13.7-9.3 22.5-9.4c0 0 .1 0 .1 0h0 96c17.7 0 32 14.3 32 32s-14.3 32-32 32H525.3z\"],\"circle-chevron-right\":[512,512,[\"chevron-circle-right\"],\"f138\",\"M0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zM241 377c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l87-87-87-87c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0L345 239c9.4 9.4 9.4 24.6 0 33.9L241 377z\"],at:[512,512,[61946],\"40\",\"M256 64C150 64 64 150 64 256s86 192 192 192c17.7 0 32 14.3 32 32s-14.3 32-32 32C114.6 512 0 397.4 0 256S114.6 0 256 0S512 114.6 512 256v32c0 53-43 96-96 96c-29.3 0-55.6-13.2-73.2-33.9C320 371.1 289.5 384 256 384c-70.7 0-128-57.3-128-128s57.3-128 128-128c27.9 0 53.7 8.9 74.7 24.1c5.7-5 13.1-8.1 21.3-8.1c17.7 0 32 14.3 32 32v80 32c0 17.7 14.3 32 32 32s32-14.3 32-32V256c0-106-86-192-192-192zm64 192a64 64 0 1 0 -128 0 64 64 0 1 0 128 0z\"],\"trash-can\":[448,512,[61460,\"trash-alt\"],\"f2ed\",\"M135.2 17.7C140.6 6.8 151.7 0 163.8 0H284.2c12.1 0 23.2 6.8 28.6 17.7L320 32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 96 0 81.7 0 64S14.3 32 32 32h96l7.2-14.3zM32 128H416V448c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V128zm96 64c-8.8 0-16 7.2-16 16V432c0 8.8 7.2 16 16 16s16-7.2 16-16V208c0-8.8-7.2-16-16-16zm96 0c-8.8 0-16 7.2-16 16V432c0 8.8 7.2 16 16 16s16-7.2 16-16V208c0-8.8-7.2-16-16-16zm96 0c-8.8 0-16 7.2-16 16V432c0 8.8 7.2 16 16 16s16-7.2 16-16V208c0-8.8-7.2-16-16-16z\"],\"text-height\":[576,512,[],\"f034\",\"M64 128V96h64l0 320H96c-17.7 0-32 14.3-32 32s14.3 32 32 32H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H192l0-320h64v32c0 17.7 14.3 32 32 32s32-14.3 32-32V80c0-26.5-21.5-48-48-48H160 48C21.5 32 0 53.5 0 80v48c0 17.7 14.3 32 32 32s32-14.3 32-32zM502.6 41.4c-12.5-12.5-32.8-12.5-45.3 0l-64 64c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8h32V352H416c-12.9 0-24.6 7.8-29.6 19.8s-2.2 25.7 6.9 34.9l64 64c12.5 12.5 32.8 12.5 45.3 0l64-64c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8H512V160h32c12.9 0 24.6-7.8 29.6-19.8s2.2-25.7-6.9-34.9l-64-64z\"],\"user-xmark\":[640,512,[\"user-times\"],\"f235\",\"M96 128a128 128 0 1 1 256 0A128 128 0 1 1 96 128zM0 482.3C0 383.8 79.8 304 178.3 304h91.4C368.2 304 448 383.8 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3zM471 143c9.4-9.4 24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9z\"],stethoscope:[576,512,[129658],\"f0f1\",\"M142.4 21.9c5.6 16.8-3.5 34.9-20.2 40.5L96 71.1V192c0 53 43 96 96 96s96-43 96-96V71.1l-26.1-8.7c-16.8-5.6-25.8-23.7-20.2-40.5s23.7-25.8 40.5-20.2l26.1 8.7C334.4 19.1 352 43.5 352 71.1V192c0 77.2-54.6 141.6-127.3 156.7C231 404.6 278.4 448 336 448c61.9 0 112-50.1 112-112V265.3c-28.3-12.3-48-40.5-48-73.3c0-44.2 35.8-80 80-80s80 35.8 80 80c0 32.8-19.7 61-48 73.3V336c0 97.2-78.8 176-176 176c-92.9 0-168.9-71.9-175.5-163.1C87.2 334.2 32 269.6 32 192V71.1c0-27.5 17.6-52 43.8-60.7l26.1-8.7c16.8-5.6 34.9 3.5 40.5 20.2zM480 224a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],message:[512,512,[\"comment-alt\"],\"f27a\",\"M64 0C28.7 0 0 28.7 0 64V352c0 35.3 28.7 64 64 64h96v80c0 6.1 3.4 11.6 8.8 14.3s11.9 2.1 16.8-1.5L309.3 416H448c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64z\"],info:[192,512,[],\"f129\",\"M48 80a48 48 0 1 1 96 0A48 48 0 1 1 48 80zM0 224c0-17.7 14.3-32 32-32H96c17.7 0 32 14.3 32 32V448h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H64V256H32c-17.7 0-32-14.3-32-32z\"],\"down-left-and-up-right-to-center\":[512,512,[\"compress-alt\"],\"f422\",\"M439 7c9.4-9.4 24.6-9.4 33.9 0l32 32c9.4 9.4 9.4 24.6 0 33.9l-87 87 39 39c6.9 6.9 8.9 17.2 5.2 26.2s-12.5 14.8-22.2 14.8H296c-13.3 0-24-10.7-24-24V72c0-9.7 5.8-18.5 14.8-22.2s19.3-1.7 26.2 5.2l39 39L439 7zM72 272H216c13.3 0 24 10.7 24 24V440c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-39-39L73 505c-9.4 9.4-24.6 9.4-33.9 0L7 473c-9.4-9.4-9.4-24.6 0-33.9l87-87L55 313c-6.9-6.9-8.9-17.2-5.2-26.2s12.5-14.8 22.2-14.8z\"],explosion:[576,512,[],\"e4e9\",\"M499.6 11.3c6.7-10.7 20.5-14.5 31.7-8.5s15.8 19.5 10.6 31L404.8 338.6c2.2 2.3 4.3 4.7 6.3 7.1l97.2-54.7c10.5-5.9 23.6-3.1 30.9 6.4s6.3 23-2.2 31.5l-87 87H378.5c-13.2-37.3-48.7-64-90.5-64s-77.4 26.7-90.5 64H117.8L42.3 363.7c-9.7-6.7-13.1-19.6-7.9-30.3s17.4-15.9 28.7-12.4l97.2 30.4c3-3.9 6.1-7.7 9.4-11.3L107.4 236.3c-6.1-10.1-3.9-23.1 5.1-30.7s22.2-7.5 31.1 .1L246 293.6c1.5-.4 3-.8 4.5-1.1l13.6-142.7c1.2-12.3 11.5-21.7 23.9-21.7s22.7 9.4 23.9 21.7l13.5 141.9L499.6 11.3zM64 448v0H512v0h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H64zM288 0c13.3 0 24 10.7 24 24V72c0 13.3-10.7 24-24 24s-24-10.7-24-24V24c0-13.3 10.7-24 24-24z\"],\"file-lines\":[384,512,[128441,128462,61686,\"file-alt\",\"file-text\"],\"f15c\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM112 256H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],\"wave-square\":[640,512,[],\"f83e\",\"M128 64c0-17.7 14.3-32 32-32H320c17.7 0 32 14.3 32 32V416h96V256c0-17.7 14.3-32 32-32H608c17.7 0 32 14.3 32 32s-14.3 32-32 32H512V448c0 17.7-14.3 32-32 32H320c-17.7 0-32-14.3-32-32V96H192V256c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32h96V64z\"],ring:[512,512,[],\"f70b\",\"M64 208c0 7.8 4.4 18.7 17.1 30.3C126.5 214.1 188.9 200 256 200s129.5 14.1 174.9 38.3C443.6 226.7 448 215.8 448 208c0-12.3-10.8-32-47.9-50.6C364.9 139.8 314 128 256 128s-108.9 11.8-144.1 29.4C74.8 176 64 195.7 64 208zm192 40c-47 0-89.3 7.6-122.9 19.7C166.3 280.2 208.8 288 256 288s89.7-7.8 122.9-20.3C345.3 255.6 303 248 256 248zM0 208c0-49.6 39.4-85.8 83.3-107.8C129.1 77.3 190.3 64 256 64s126.9 13.3 172.7 36.2c43.9 22 83.3 58.2 83.3 107.8v96c0 49.6-39.4 85.8-83.3 107.8C382.9 434.7 321.7 448 256 448s-126.9-13.3-172.7-36.2C39.4 389.8 0 353.6 0 304V208z\"],\"building-un\":[384,512,[],\"e4d9\",\"M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM237.3 71.1l34.7 52V80c0-8.8 7.2-16 16-16s16 7.2 16 16v96c0 7.1-4.6 13.3-11.4 15.3s-14-.6-17.9-6.4l-34.7-52V176c0 8.8-7.2 16-16 16s-16-7.2-16-16V80c0-7.1 4.6-13.3 11.4-15.3s14 .6 17.9 6.4zM112 80v64c0 8.8 7.2 16 16 16s16-7.2 16-16V80c0-8.8 7.2-16 16-16s16 7.2 16 16v64c0 26.5-21.5 48-48 48s-48-21.5-48-48V80c0-8.8 7.2-16 16-16s16 7.2 16 16z\"],\"dice-three\":[448,512,[9858],\"f527\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm64 96a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm64 128a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm128 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"calendar-days\":[448,512,[\"calendar-alt\"],\"f073\",\"M128 0c17.7 0 32 14.3 32 32V64H288V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64h48c26.5 0 48 21.5 48 48v48H0V112C0 85.5 21.5 64 48 64H96V32c0-17.7 14.3-32 32-32zM0 192H448V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V192zm64 80v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H80c-8.8 0-16 7.2-16 16zm128 0v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H208c-8.8 0-16 7.2-16 16zm144-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H336zM64 400v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V400c0-8.8-7.2-16-16-16H80c-8.8 0-16 7.2-16 16zm144-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V400c0-8.8-7.2-16-16-16H208zm112 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V400c0-8.8-7.2-16-16-16H336c-8.8 0-16 7.2-16 16z\"],\"anchor-circle-check\":[640,512,[],\"e4aa\",\"M320 96a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm21.1 80C367 158.8 384 129.4 384 96c0-53-43-96-96-96s-96 43-96 96c0 33.4 17 62.8 42.9 80H224c-17.7 0-32 14.3-32 32s14.3 32 32 32h32V448H208c-53 0-96-43-96-96v-6.1l7 7c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L97 263c-9.4-9.4-24.6-9.4-33.9 0L7 319c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l7-7V352c0 88.4 71.6 160 160 160h80 80c8.2 0 16.3-.6 24.2-1.8c-22.2-16.2-40.4-37.5-53-62.2H320V368 240h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H341.1zM640 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z\"],\"building-circle-arrow-right\":[640,512,[],\"e4d1\",\"M0 48C0 21.5 21.5 0 48 0H336c26.5 0 48 21.5 48 48V232.2c-39.1 32.3-64 81.1-64 135.8c0 49.5 20.4 94.2 53.3 126.2C364.5 505.1 351.1 512 336 512H240V432c0-26.5-21.5-48-48-48s-48 21.5-48 48v80H48c-26.5 0-48-21.5-48-48V48zM80 224c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H80zm80 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H176c-8.8 0-16 7.2-16 16zm112-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H272zM64 112v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H80c-8.8 0-16 7.2-16 16zM176 96c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H176zm80 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16zm96 256a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm140.7-67.3c-6.2 6.2-6.2 16.4 0 22.6L521.4 352H432c-8.8 0-16 7.2-16 16s7.2 16 16 16h89.4l-28.7 28.7c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0l56-56c6.2-6.2 6.2-16.4 0-22.6l-56-56c-6.2-6.2-16.4-6.2-22.6 0z\"],volleyball:[512,512,[127952,\"volleyball-ball\"],\"f45f\",\"M511.8 267.4c-26.1 8.7-53.4 13.8-81 15.1c9.2-105.3-31.5-204.2-103.2-272.4C434.1 41.1 512 139.5 512 256c0 3.8-.1 7.6-.2 11.4zm-3.9 34.7c-5.8 32-17.6 62-34.2 88.7c-97.5 48.5-217.7 42.6-311.9-24.5c23.7-36.2 55.4-67.7 94.5-91.8c79.9 43.2 170.1 50.8 251.6 27.6zm-236-55.5c-2.5-90.9-41.1-172.7-101.9-231.7C196.8 5.2 225.8 0 256 0c2.7 0 5.3 0 7.9 .1c90.8 60.2 145.7 167.2 134.7 282.3c-43.1-2.4-86.4-14.1-126.8-35.9zM138 28.8c20.6 18.3 38.7 39.4 53.7 62.6C95.9 136.1 30.6 220.8 7.3 316.9C2.5 297.4 0 277 0 256C0 157.2 56 71.5 138 28.8zm69.6 90.5c19.5 38.6 31 81.9 32.3 127.7C162.5 294.6 110.9 368.9 90.2 451C66 430.4 45.6 405.4 30.4 377.2c6.7-108.7 71.9-209.9 177.1-257.9zM256 512c-50.7 0-98-14.7-137.8-40.2c5.6-27 14.8-53.1 27.4-77.7C232.2 454.6 338.1 468.8 433 441c-46 44-108.3 71-177 71z\"],\"arrows-up-to-line\":[576,512,[],\"e4c2\",\"M32 96l512 0c17.7 0 32-14.3 32-32s-14.3-32-32-32L32 32C14.3 32 0 46.3 0 64S14.3 96 32 96zM9.4 233.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L96 237.3 96 448c0 17.7 14.3 32 32 32s32-14.3 32-32l0-210.7 41.4 41.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-96-96c-12.5-12.5-32.8-12.5-45.3 0l-96 96zm320 45.3c12.5 12.5 32.8 12.5 45.3 0L416 237.3 416 448c0 17.7 14.3 32 32 32s32-14.3 32-32l0-210.7 41.4 41.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-96-96c-12.5-12.5-32.8-12.5-45.3 0l-96 96c-12.5 12.5-12.5 32.8 0 45.3z\"],\"sort-down\":[320,512,[\"sort-desc\"],\"f0dd\",\"M182.6 470.6c-12.5 12.5-32.8 12.5-45.3 0l-128-128c-9.2-9.2-11.9-22.9-6.9-34.9s16.6-19.8 29.6-19.8H288c12.9 0 24.6 7.8 29.6 19.8s2.2 25.7-6.9 34.9l-128 128z\"],\"circle-minus\":[512,512,[\"minus-circle\"],\"f056\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM184 232H328c13.3 0 24 10.7 24 24s-10.7 24-24 24H184c-13.3 0-24-10.7-24-24s10.7-24 24-24z\"],\"door-open\":[576,512,[],\"f52b\",\"M320 32c0-9.9-4.5-19.2-12.3-25.2S289.8-1.4 280.2 1l-179.9 45C79 51.3 64 70.5 64 92.5V448H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H96 288h32V480 32zM256 256c0 17.7-10.7 32-24 32s-24-14.3-24-32s10.7-32 24-32s24 14.3 24 32zm96-128h96V480c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H512V128c0-35.3-28.7-64-64-64H352v64z\"],\"right-from-bracket\":[512,512,[\"sign-out-alt\"],\"f2f5\",\"M377.9 105.9L500.7 228.7c7.2 7.2 11.3 17.1 11.3 27.3s-4.1 20.1-11.3 27.3L377.9 406.1c-6.4 6.4-15 9.9-24 9.9c-18.7 0-33.9-15.2-33.9-33.9l0-62.1-128 0c-17.7 0-32-14.3-32-32l0-64c0-17.7 14.3-32 32-32l128 0 0-62.1c0-18.7 15.2-33.9 33.9-33.9c9 0 17.6 3.6 24 9.9zM160 96L96 96c-17.7 0-32 14.3-32 32l0 256c0 17.7 14.3 32 32 32l64 0c17.7 0 32 14.3 32 32s-14.3 32-32 32l-64 0c-53 0-96-43-96-96L0 128C0 75 43 32 96 32l64 0c17.7 0 32 14.3 32 32s-14.3 32-32 32z\"],atom:[512,512,[9883],\"f5d2\",\"M256 398.8c-11.8 5.1-23.4 9.7-34.9 13.5c16.7 33.8 31 35.7 34.9 35.7s18.1-1.9 34.9-35.7c-11.4-3.9-23.1-8.4-34.9-13.5zM446 256c33 45.2 44.3 90.9 23.6 128c-20.2 36.3-62.5 49.3-115.2 43.2c-22 52.1-55.6 84.8-98.4 84.8s-76.4-32.7-98.4-84.8c-52.7 6.1-95-6.8-115.2-43.2C21.7 346.9 33 301.2 66 256c-33-45.2-44.3-90.9-23.6-128c20.2-36.3 62.5-49.3 115.2-43.2C179.6 32.7 213.2 0 256 0s76.4 32.7 98.4 84.8c52.7-6.1 95 6.8 115.2 43.2c20.7 37.1 9.4 82.8-23.6 128zm-65.8 67.4c-1.7 14.2-3.9 28-6.7 41.2c31.8 1.4 38.6-8.7 40.2-11.7c2.3-4.2 7-17.9-11.9-48.1c-6.8 6.3-14 12.5-21.6 18.6zm-6.7-175.9c2.8 13.1 5 26.9 6.7 41.2c7.6 6.1 14.8 12.3 21.6 18.6c18.9-30.2 14.2-44 11.9-48.1c-1.6-2.9-8.4-13-40.2-11.7zM290.9 99.7C274.1 65.9 259.9 64 256 64s-18.1 1.9-34.9 35.7c11.4 3.9 23.1 8.4 34.9 13.5c11.8-5.1 23.4-9.7 34.9-13.5zm-159 88.9c1.7-14.3 3.9-28 6.7-41.2c-31.8-1.4-38.6 8.7-40.2 11.7c-2.3 4.2-7 17.9 11.9 48.1c6.8-6.3 14-12.5 21.6-18.6zM110.2 304.8C91.4 335 96 348.7 98.3 352.9c1.6 2.9 8.4 13 40.2 11.7c-2.8-13.1-5-26.9-6.7-41.2c-7.6-6.1-14.8-12.3-21.6-18.6zM336 256a80 80 0 1 0 -160 0 80 80 0 1 0 160 0zm-80-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],soap:[512,512,[129532],\"e06e\",\"M208 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM320 256a64 64 0 1 0 0-128 64 64 0 1 0 0 128zM416 32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm0 160c0 27.6-11.7 52.5-30.4 70.1C422.1 275.7 448 310.8 448 352c0 53-43 96-96 96H160c-53 0-96-43-96-96s43-96 96-96h88.4c-15.2-17-24.4-39.4-24.4-64H96c-53 0-96 43-96 96V416c0 53 43 96 96 96H416c53 0 96-43 96-96V288c0-53-43-96-96-96zM160 288c-35.3 0-64 28.7-64 64s28.7 64 64 64H352c35.3 0 64-28.7 64-64s-28.7-64-64-64H320 160z\"],icons:[512,512,[\"heart-music-camera-bolt\"],\"f86d\",\"M500.3 7.3C507.7 13.3 512 22.4 512 32V176c0 26.5-28.7 48-64 48s-64-21.5-64-48s28.7-48 64-48V71L352 90.2V208c0 26.5-28.7 48-64 48s-64-21.5-64-48s28.7-48 64-48V64c0-15.3 10.8-28.4 25.7-31.4l160-32c9.4-1.9 19.1 .6 26.6 6.6zM74.7 304l11.8-17.8c5.9-8.9 15.9-14.2 26.6-14.2h61.7c10.7 0 20.7 5.3 26.6 14.2L213.3 304H240c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V352c0-26.5 21.5-48 48-48H74.7zM192 408a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM478.7 278.3L440.3 368H496c6.7 0 12.6 4.1 15 10.4s.6 13.3-4.4 17.7l-128 112c-5.6 4.9-13.9 5.3-19.9 .9s-8.2-12.4-5.3-19.2L391.7 400H336c-6.7 0-12.6-4.1-15-10.4s-.6-13.3 4.4-17.7l128-112c5.6-4.9 13.9-5.3 19.9-.9s8.2 12.4 5.3 19.2zm-339-59.2c-6.5 6.5-17 6.5-23 0L19.9 119.2c-28-29-26.5-76.9 5-103.9c27-23.5 68.4-19 93.4 6.5l10 10.5 9.5-10.5c25-25.5 65.9-30 93.9-6.5c31 27 32.5 74.9 4.5 103.9l-96.4 99.9z\"],\"microphone-lines-slash\":[640,512,[\"microphone-alt-slash\"],\"f539\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L472.1 344.7c15.2-26 23.9-56.3 23.9-88.7V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v24 16c0 21.2-5.1 41.1-14.2 58.7L416 300.8V256H358.9l-34.5-27c2.9-3.1 7-5 11.6-5h80V192H336c-8.8 0-16-7.2-16-16s7.2-16 16-16h80V128H336c-8.8 0-16-7.2-16-16s7.2-16 16-16h80c0-53-43-96-96-96s-96 43-96 96v54.3L38.8 5.1zM358.2 378.2C346.1 382 333.3 384 320 384c-70.7 0-128-57.3-128-128v-8.7L144.7 210c-.5 1.9-.7 3.9-.7 6v40c0 89.1 66.2 162.7 152 174.4V464H248c-13.3 0-24 10.7-24 24s10.7 24 24 24h72 72c13.3 0 24-10.7 24-24s-10.7-24-24-24H344V430.4c20.4-2.8 39.7-9.1 57.3-18.2l-43.1-33.9z\"],\"bridge-circle-check\":[640,512,[],\"e4c9\",\"M64 32C46.3 32 32 46.3 32 64s14.3 32 32 32h40v64H32V288c53 0 96 43 96 96v64c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V384c0-53 43-96 96-96c6.3 0 12.4 .6 18.3 1.7C367.1 231.8 426.9 192 496 192c42.5 0 81.6 15.1 112 40.2V160H536V96h40c17.7 0 32-14.3 32-32s-14.3-32-32-32H64zM488 96v64H408V96h80zM360 96v64H280V96h80zM232 96v64H152V96h80zM640 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z\"],\"pump-medical\":[448,512,[],\"e06a\",\"M128 32v96H256V96h60.1c4.2 0 8.3 1.7 11.3 4.7l33.9 33.9c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L372.7 55.4c-15-15-35.4-23.4-56.6-23.4H256c0-17.7-14.3-32-32-32H160c-17.7 0-32 14.3-32 32zM117.4 160c-33.3 0-61 25.5-63.8 58.7L35 442.7C31.9 480 61.3 512 98.8 512H285.2c37.4 0 66.9-32 63.8-69.3l-18.7-224c-2.8-33.2-30.5-58.7-63.8-58.7H117.4zM216 280v32h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H216v32c0 13.3-10.7 24-24 24s-24-10.7-24-24V360H136c-13.3 0-24-10.7-24-24s10.7-24 24-24h32V280c0-13.3 10.7-24 24-24s24 10.7 24 24z\"],fingerprint:[512,512,[],\"f577\",\"M48 256C48 141.1 141.1 48 256 48c63.1 0 119.6 28.1 157.8 72.5c8.6 10.1 23.8 11.2 33.8 2.6s11.2-23.8 2.6-33.8C403.3 34.6 333.7 0 256 0C114.6 0 0 114.6 0 256v40c0 13.3 10.7 24 24 24s24-10.7 24-24V256zm458.5-52.9c-2.7-13-15.5-21.3-28.4-18.5s-21.3 15.5-18.5 28.4c2.9 13.9 4.5 28.3 4.5 43.1v40c0 13.3 10.7 24 24 24s24-10.7 24-24V256c0-18.1-1.9-35.8-5.5-52.9zM256 80c-19 0-37.4 3-54.5 8.6c-15.2 5-18.7 23.7-8.3 35.9c7.1 8.3 18.8 10.8 29.4 7.9c10.6-2.9 21.8-4.4 33.4-4.4c70.7 0 128 57.3 128 128v24.9c0 25.2-1.5 50.3-4.4 75.3c-1.7 14.6 9.4 27.8 24.2 27.8c11.8 0 21.9-8.6 23.3-20.3c3.3-27.4 5-55 5-82.7V256c0-97.2-78.8-176-176-176zM150.7 148.7c-9.1-10.6-25.3-11.4-33.9-.4C93.7 178 80 215.4 80 256v24.9c0 24.2-2.6 48.4-7.8 71.9C68.8 368.4 80.1 384 96.1 384c10.5 0 19.9-7 22.2-17.3c6.4-28.1 9.7-56.8 9.7-85.8V256c0-27.2 8.5-52.4 22.9-73.1c7.2-10.4 8-24.6-.2-34.2zM256 160c-53 0-96 43-96 96v24.9c0 35.9-4.6 71.5-13.8 106.1c-3.8 14.3 6.7 29 21.5 29c9.5 0 17.9-6.2 20.4-15.4c10.5-39 15.9-79.2 15.9-119.7V256c0-28.7 23.3-52 52-52s52 23.3 52 52v24.9c0 36.3-3.5 72.4-10.4 107.9c-2.7 13.9 7.7 27.2 21.8 27.2c10.2 0 19-7 21-17c7.7-38.8 11.6-78.3 11.6-118.1V256c0-53-43-96-96-96zm24 96c0-13.3-10.7-24-24-24s-24 10.7-24 24v24.9c0 59.9-11 119.3-32.5 175.2l-5.9 15.3c-4.8 12.4 1.4 26.3 13.8 31s26.3-1.4 31-13.8l5.9-15.3C267.9 411.9 280 346.7 280 280.9V256z\"],\"hand-point-right\":[512,512,[],\"f0a4\",\"M480 96c17.7 0 32 14.3 32 32s-14.3 32-32 32l-208 0 0-64 208 0zM320 288c17.7 0 32 14.3 32 32s-14.3 32-32 32l-64 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l64 0zm64-64c0 17.7-14.3 32-32 32l-48 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l48 0c17.7 0 32 14.3 32 32zM288 384c17.7 0 32 14.3 32 32s-14.3 32-32 32l-64 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l64 0zm-88-96l.6 0c-5.4 9.4-8.6 20.3-8.6 32c0 13.2 4 25.4 10.8 35.6C177.9 364.3 160 388.1 160 416c0 11.7 3.1 22.6 8.6 32l-8.6 0C71.6 448 0 376.4 0 288l0-61.7c0-42.4 16.9-83.1 46.9-113.1l11.6-11.6C82.5 77.5 115.1 64 149 64l27 0c35.3 0 64 28.7 64 64l0 88c0 22.1-17.9 40-40 40s-40-17.9-40-40l0-56c0-8.8-7.2-16-16-16s-16 7.2-16 16l0 56c0 39.8 32.2 72 72 72z\"],\"magnifying-glass-location\":[512,512,[\"search-location\"],\"f689\",\"M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zM288 176c0-44.2-35.8-80-80-80s-80 35.8-80 80c0 48.8 46.5 111.6 68.6 138.6c6 7.3 16.8 7.3 22.7 0c22.1-27 68.6-89.8 68.6-138.6zm-112 0a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],\"forward-step\":[320,512,[\"step-forward\"],\"f051\",\"M52.5 440.6c-9.5 7.9-22.8 9.7-34.1 4.4S0 428.4 0 416V96C0 83.6 7.2 72.3 18.4 67s24.5-3.6 34.1 4.4l192 160L256 241V96c0-17.7 14.3-32 32-32s32 14.3 32 32V416c0 17.7-14.3 32-32 32s-32-14.3-32-32V271l-11.5 9.6-192 160z\"],\"face-smile-beam\":[512,512,[128522,\"smile-beam\"],\"f5b8\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM164.1 325.5C182 346.2 212.6 368 256 368s74-21.8 91.9-42.5c5.8-6.7 15.9-7.4 22.6-1.6s7.4 15.9 1.6 22.6C349.8 372.1 311.1 400 256 400s-93.8-27.9-116.1-53.5c-5.8-6.7-5.1-16.8 1.6-22.6s16.8-5.1 22.6 1.6zm53.5-96.7l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z\"],\"flag-checkered\":[448,512,[127937],\"f11e\",\"M32 0C49.7 0 64 14.3 64 32V48l69-17.2c38.1-9.5 78.3-5.1 113.5 12.5c46.3 23.2 100.8 23.2 147.1 0l9.6-4.8C423.8 28.1 448 43.1 448 66.1V345.8c0 13.3-8.3 25.3-20.8 30l-34.7 13c-46.2 17.3-97.6 14.6-141.7-7.4c-37.9-19-81.3-23.7-122.5-13.4L64 384v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V400 334 64 32C0 14.3 14.3 0 32 0zM64 187.1l64-13.9v65.5L64 252.6V318l48.8-12.2c5.1-1.3 10.1-2.4 15.2-3.3V238.7l38.9-8.4c8.3-1.8 16.7-2.5 25.1-2.1l0-64c13.6 .4 27.2 2.6 40.4 6.4l23.6 6.9v66.7l-41.7-12.3c-7.3-2.1-14.8-3.4-22.3-3.8v71.4c21.8 1.9 43.3 6.7 64 14.4V244.2l22.7 6.7c13.5 4 27.3 6.4 41.3 7.4V194c-7.8-.8-15.6-2.3-23.2-4.5l-40.8-12v-62c-13-3.8-25.8-8.8-38.2-15c-8.2-4.1-16.9-7-25.8-8.8v72.4c-13-.4-26 .8-38.7 3.6L128 173.2V98L64 114v73.1zM320 335.7c16.8 1.5 33.9-.7 50-6.8l14-5.2V251.9l-7.9 1.8c-18.4 4.3-37.3 5.7-56.1 4.5v77.4zm64-149.4V115.4c-20.9 6.1-42.4 9.1-64 9.1V194c13.9 1.4 28 .5 41.7-2.6l22.3-5.2z\"],football:[512,512,[127944,\"football-ball\"],\"f44e\",\"M247.5 25.4c-13.5 3.3-26.4 7.2-38.6 11.7C142.9 61.6 96.7 103.6 66 153.6c-18.3 29.8-30.9 62.3-39.2 95.4L264.5 486.6c13.5-3.3 26.4-7.2 38.6-11.7c66-24.5 112.2-66.5 142.9-116.5c18.3-29.8 30.9-62.3 39.1-95.3L247.5 25.4zM495.2 205.3c6.1-56.8 1.4-112.2-7.7-156.4c-2.7-12.9-13-22.9-26.1-25.1c-58.2-9.7-109.9-12-155.6-7.9L495.2 205.3zM206.1 496L16.8 306.7c-6.1 56.8-1.4 112.2 7.7 156.4c2.7 12.9 13 22.9 26.1 25.1c58.2 9.7 109.9 12 155.6 7.9zm54.6-331.3c6.2-6.2 16.4-6.2 22.6 0l64 64c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0l-64-64c-6.2-6.2-6.2-16.4 0-22.6zm-48 48c6.2-6.2 16.4-6.2 22.6 0l64 64c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0l-64-64c-6.2-6.2-6.2-16.4 0-22.6zm-48 48c6.2-6.2 16.4-6.2 22.6 0l64 64c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0l-64-64c-6.2-6.2-6.2-16.4 0-22.6z\"],\"school-circle-exclamation\":[640,512,[],\"e56c\",\"M337.8 5.4C327-1.8 313-1.8 302.2 5.4L166.3 96H48C21.5 96 0 117.5 0 144V464c0 26.5 21.5 48 48 48H320v0H256V416c0-35.3 28.7-64 64-64l.3 0h.5c3.4-37.7 18.7-72.1 42.2-99.1C350.2 260 335.6 264 320 264c-48.6 0-88-39.4-88-88s39.4-88 88-88s88 39.4 88 88c0 18.3-5.6 35.3-15.1 49.4c29-21 64.6-33.4 103.1-33.4c59.5 0 112.1 29.6 144 74.8V144c0-26.5-21.5-48-48-48H473.7L337.8 5.4zM96 192h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V208c0-8.8 7.2-16 16-16zm0 128h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zM320 128c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H336V144c0-8.8-7.2-16-16-16zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z\"],crop:[512,512,[],\"f125\",\"M448 109.3l54.6-54.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L402.7 64 160 64v64l178.7 0L128 338.7V32c0-17.7-14.3-32-32-32S64 14.3 64 32V64H32C14.3 64 0 78.3 0 96s14.3 32 32 32H64V384c0 35.3 28.7 64 64 64H352V384H173.3L384 173.3 384 480c0 17.7 14.3 32 32 32s32-14.3 32-32V448h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H448l0-274.7z\"],\"angles-down\":[448,512,[\"angle-double-down\"],\"f103\",\"M246.6 470.6c-12.5 12.5-32.8 12.5-45.3 0l-160-160c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0L224 402.7 361.4 265.4c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3l-160 160zm160-352l-160 160c-12.5 12.5-32.8 12.5-45.3 0l-160-160c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0L224 210.7 361.4 73.4c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3z\"],\"users-rectangle\":[640,512,[],\"e594\",\"M96 0C43 0 0 43 0 96V416c0 53 43 96 96 96H544c53 0 96-43 96-96V96c0-53-43-96-96-96H96zM64 96c0-17.7 14.3-32 32-32H544c17.7 0 32 14.3 32 32V416c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V96zm159.8 80a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM96 309.3c0 14.7 11.9 26.7 26.7 26.7h56.1c8-34.1 32.8-61.7 65.2-73.6c-7.5-4.1-16.2-6.4-25.3-6.4H149.3C119.9 256 96 279.9 96 309.3zM461.2 336h56.1c14.7 0 26.7-11.9 26.7-26.7c0-29.5-23.9-53.3-53.3-53.3H421.3c-9.2 0-17.8 2.3-25.3 6.4c32.4 11.9 57.2 39.5 65.2 73.6zM372 289c-3.9-.7-7.9-1-12-1H280c-4.1 0-8.1 .3-12 1c-26 4.4-47.3 22.7-55.9 47c-2.7 7.5-4.1 15.6-4.1 24c0 13.3 10.7 24 24 24H408c13.3 0 24-10.7 24-24c0-8.4-1.4-16.5-4.1-24c-8.6-24.3-29.9-42.6-55.9-47zM512 176a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM320 256a64 64 0 1 0 0-128 64 64 0 1 0 0 128z\"],\"people-roof\":[640,512,[],\"e537\",\"M335.5 4l288 160c15.4 8.6 21 28.1 12.4 43.5s-28.1 21-43.5 12.4L320 68.6 47.5 220c-15.4 8.6-34.9 3-43.5-12.4s-3-34.9 12.4-43.5L304.5 4c9.7-5.4 21.4-5.4 31.1 0zM320 160a40 40 0 1 1 0 80 40 40 0 1 1 0-80zM144 256a40 40 0 1 1 0 80 40 40 0 1 1 0-80zm312 40a40 40 0 1 1 80 0 40 40 0 1 1 -80 0zM226.9 491.4L200 441.5V480c0 17.7-14.3 32-32 32H120c-17.7 0-32-14.3-32-32V441.5L61.1 491.4c-6.3 11.7-20.8 16-32.5 9.8s-16-20.8-9.8-32.5l37.9-70.3c15.3-28.5 45.1-46.3 77.5-46.3h19.5c16.3 0 31.9 4.5 45.4 12.6l33.6-62.3c15.3-28.5 45.1-46.3 77.5-46.3h19.5c32.4 0 62.1 17.8 77.5 46.3l33.6 62.3c13.5-8.1 29.1-12.6 45.4-12.6h19.5c32.4 0 62.1 17.8 77.5 46.3l37.9 70.3c6.3 11.7 1.9 26.2-9.8 32.5s-26.2 1.9-32.5-9.8L552 441.5V480c0 17.7-14.3 32-32 32H472c-17.7 0-32-14.3-32-32V441.5l-26.9 49.9c-6.3 11.7-20.8 16-32.5 9.8s-16-20.8-9.8-32.5l36.3-67.5c-1.7-1.7-3.2-3.6-4.3-5.8L376 345.5V400c0 17.7-14.3 32-32 32H296c-17.7 0-32-14.3-32-32V345.5l-26.9 49.9c-1.2 2.2-2.6 4.1-4.3 5.8l36.3 67.5c6.3 11.7 1.9 26.2-9.8 32.5s-26.2 1.9-32.5-9.8z\"],\"people-line\":[640,512,[],\"e534\",\"M360 72a40 40 0 1 0 -80 0 40 40 0 1 0 80 0zM144 208a40 40 0 1 0 0-80 40 40 0 1 0 0 80zM32 416c-17.7 0-32 14.3-32 32s14.3 32 32 32H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H32zM496 208a40 40 0 1 0 0-80 40 40 0 1 0 0 80zM200 313.5l26.9 49.9c6.3 11.7 20.8 16 32.5 9.8s16-20.8 9.8-32.5l-36.3-67.5c1.7-1.7 3.2-3.6 4.3-5.8L264 217.5V272c0 17.7 14.3 32 32 32h48c17.7 0 32-14.3 32-32V217.5l26.9 49.9c1.2 2.2 2.6 4.1 4.3 5.8l-36.3 67.5c-6.3 11.7-1.9 26.2 9.8 32.5s26.2 1.9 32.5-9.8L440 313.5V352c0 17.7 14.3 32 32 32h48c17.7 0 32-14.3 32-32V313.5l26.9 49.9c6.3 11.7 20.8 16 32.5 9.8s16-20.8 9.8-32.5l-37.9-70.3c-15.3-28.5-45.1-46.3-77.5-46.3H486.2c-16.3 0-31.9 4.5-45.4 12.6l-33.6-62.3c-15.3-28.5-45.1-46.3-77.5-46.3H310.2c-32.4 0-62.1 17.8-77.5 46.3l-33.6 62.3c-13.5-8.1-29.1-12.6-45.4-12.6H134.2c-32.4 0-62.1 17.8-77.5 46.3L18.9 340.6c-6.3 11.7-1.9 26.2 9.8 32.5s26.2 1.9 32.5-9.8L88 313.5V352c0 17.7 14.3 32 32 32h48c17.7 0 32-14.3 32-32V313.5z\"],\"beer-mug-empty\":[512,512,[\"beer\"],\"f0fc\",\"M32 64c0-17.7 14.3-32 32-32H352c17.7 0 32 14.3 32 32V96h51.2c42.4 0 76.8 34.4 76.8 76.8V274.9c0 30.4-17.9 57.9-45.6 70.2L384 381.7V416c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V64zM384 311.6l56.4-25.1c4.6-2.1 7.6-6.6 7.6-11.7V172.8c0-7.1-5.7-12.8-12.8-12.8H384V311.6zM160 144c0-8.8-7.2-16-16-16s-16 7.2-16 16V368c0 8.8 7.2 16 16 16s16-7.2 16-16V144zm64 0c0-8.8-7.2-16-16-16s-16 7.2-16 16V368c0 8.8 7.2 16 16 16s16-7.2 16-16V144zm64 0c0-8.8-7.2-16-16-16s-16 7.2-16 16V368c0 8.8 7.2 16 16 16s16-7.2 16-16V144z\"],\"diagram-predecessor\":[512,512,[],\"e477\",\"M448 416l0-64L64 352l0 64 384 0zm0 64L64 480c-35.3 0-64-28.7-64-64l0-64c0-35.3 28.7-64 64-64l384 0c35.3 0 64 28.7 64 64l0 64c0 35.3-28.7 64-64 64zM288 160c0 35.3-28.7 64-64 64L64 224c-35.3 0-64-28.7-64-64L0 96C0 60.7 28.7 32 64 32l144 0 16 0 144 0c44.2 0 80 35.8 80 80l0 16 38.1 0c21.4 0 32.1 25.9 17 41L433 239c-9.4 9.4-24.6 9.4-33.9 0L329 169c-15.1-15.1-4.4-41 17-41l38.1 0 0-16c0-8.8-7.2-16-16-16l-80 0 0 64z\"],\"arrow-up-long\":[384,512,[\"long-arrow-up\"],\"f176\",\"M214.6 9.4c-12.5-12.5-32.8-12.5-45.3 0l-128 128c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L160 109.3V480c0 17.7 14.3 32 32 32s32-14.3 32-32V109.3l73.4 73.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-128-128z\"],\"fire-flame-simple\":[384,512,[\"burn\"],\"f46a\",\"M372.5 256.5l-.7-1.9C337.8 160.8 282 76.5 209.1 8.5l-3.3-3C202.1 2 197.1 0 192 0s-10.1 2-13.8 5.5l-3.3 3C102 76.5 46.2 160.8 12.2 254.6l-.7 1.9C3.9 277.3 0 299.4 0 321.6C0 426.7 86.8 512 192 512s192-85.3 192-190.4c0-22.2-3.9-44.2-11.5-65.1zm-90.8 49.5c4.1 9.3 6.2 19.4 6.2 29.5c0 53-43 96.5-96 96.5s-96-43.5-96-96.5c0-10.1 2.1-20.3 6.2-29.5l1.9-4.3c15.8-35.4 37.9-67.7 65.3-95.1l8.9-8.9c3.6-3.6 8.5-5.6 13.6-5.6s10 2 13.6 5.6l8.9 8.9c27.4 27.4 49.6 59.7 65.3 95.1l1.9 4.3z\"],person:[320,512,[129485,\"male\"],\"f183\",\"M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l58.3 97c9.1 15.1 4.2 34.8-10.9 43.9s-34.8 4.2-43.9-10.9L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152z\"],laptop:[640,512,[128187],\"f109\",\"M128 32C92.7 32 64 60.7 64 96V352h64V96H512V352h64V96c0-35.3-28.7-64-64-64H128zM19.2 384C8.6 384 0 392.6 0 403.2C0 445.6 34.4 480 76.8 480H563.2c42.4 0 76.8-34.4 76.8-76.8c0-10.6-8.6-19.2-19.2-19.2H19.2z\"],\"file-csv\":[512,512,[],\"f6dd\",\"M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384V304H176c-35.3 0-64 28.7-64 64V512H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zM200 352h16c22.1 0 40 17.9 40 40v8c0 8.8-7.2 16-16 16s-16-7.2-16-16v-8c0-4.4-3.6-8-8-8H200c-4.4 0-8 3.6-8 8v80c0 4.4 3.6 8 8 8h16c4.4 0 8-3.6 8-8v-8c0-8.8 7.2-16 16-16s16 7.2 16 16v8c0 22.1-17.9 40-40 40H200c-22.1 0-40-17.9-40-40V392c0-22.1 17.9-40 40-40zm133.1 0H368c8.8 0 16 7.2 16 16s-7.2 16-16 16H333.1c-7.2 0-13.1 5.9-13.1 13.1c0 5.2 3 9.9 7.8 12l37.4 16.6c16.3 7.2 26.8 23.4 26.8 41.2c0 24.9-20.2 45.1-45.1 45.1H304c-8.8 0-16-7.2-16-16s7.2-16 16-16h42.9c7.2 0 13.1-5.9 13.1-13.1c0-5.2-3-9.9-7.8-12l-37.4-16.6c-16.3-7.2-26.8-23.4-26.8-41.2c0-24.9 20.2-45.1 45.1-45.1zm98.9 0c8.8 0 16 7.2 16 16v31.6c0 23 5.5 45.6 16 66c10.5-20.3 16-42.9 16-66V368c0-8.8 7.2-16 16-16s16 7.2 16 16v31.6c0 34.7-10.3 68.7-29.6 97.6l-5.1 7.7c-3 4.5-8 7.1-13.3 7.1s-10.3-2.7-13.3-7.1l-5.1-7.7c-19.3-28.9-29.6-62.9-29.6-97.6V368c0-8.8 7.2-16 16-16z\"],menorah:[640,512,[],\"f676\",\"M20.8 7.4C22.8 2.9 27.1 0 32 0s9.2 2.9 11.2 7.4L61.3 49.7c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32S0 81.7 0 64V62.8c0-4.5 .9-8.9 2.7-13.1L20.8 7.4zm96 0C118.8 2.9 123.1 0 128 0s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32s-32-14.3-32-32V62.8c0-4.5 .9-8.9 2.7-13.1L116.8 7.4zm77.8 42.4L212.8 7.4C214.8 2.9 219.1 0 224 0s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32s-32-14.3-32-32V62.8c0-4.5 .9-8.9 2.7-13.1zM308.8 7.4C310.8 2.9 315.1 0 320 0s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32s-32-14.3-32-32V62.8c0-4.5 .9-8.9 2.7-13.1L308.8 7.4zm77.8 42.4L404.8 7.4C406.8 2.9 411.1 0 416 0s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32s-32-14.3-32-32V62.8c0-4.5 .9-8.9 2.7-13.1zM500.8 7.4C502.8 2.9 507.1 0 512 0s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32s-32-14.3-32-32V62.8c0-4.5 .9-8.9 2.7-13.1L500.8 7.4zm77.8 42.4L596.8 7.4C598.8 2.9 603.1 0 608 0s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V64c0 17.7-14.3 32-32 32s-32-14.3-32-32V62.8c0-4.5 .9-8.9 2.7-13.1zM32 128c17.7 0 32 14.3 32 32V288c0 17.7 14.3 32 32 32H288V160c0-17.7 14.3-32 32-32s32 14.3 32 32V320H544c17.7 0 32-14.3 32-32V160c0-17.7 14.3-32 32-32s32 14.3 32 32V288c0 53-43 96-96 96H352v64H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H320 160c-17.7 0-32-14.3-32-32s14.3-32 32-32H288V384H96c-53 0-96-43-96-96V160c0-17.7 14.3-32 32-32zm96 0c17.7 0 32 14.3 32 32v96 32H96V256 160c0-17.7 14.3-32 32-32zm96 0c17.7 0 32 14.3 32 32v96 32H192V256 160c0-17.7 14.3-32 32-32zm192 0c17.7 0 32 14.3 32 32v96 32H384V256 160c0-17.7 14.3-32 32-32zm96 0c17.7 0 32 14.3 32 32v96 32H480V256 160c0-17.7 14.3-32 32-32z\"],\"truck-plane\":[640,512,[],\"e58f\",\"M200 0c-30.6 0-56 54.7-56 86.1V192.5L7.8 274.3C2.9 277.2 0 282.4 0 288v64c0 5.1 2.4 9.8 6.4 12.8s9.3 3.9 14.1 2.5l123.4-37v81.2l-50 40c-3.8 3-6 7.6-6 12.5v32c0 5.1 2.5 10 6.6 13s9.5 3.8 14.4 2.2L200 480.9 290.4 511c-1.6-4.7-2.4-9.8-2.4-15V463.4c-18.2-10.5-30.7-29.7-31.9-51.8l-.1-.1V408 325.5 184l0-1.1 0 0V86.1C256 54.7 231.5 0 200 0zm88 176V400c0 20.9 13.4 38.7 32 45.3V488c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V448H544v40c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V445.3c18.6-6.6 32-24.4 32-45.3V176c0-26.5-21.5-48-48-48H336c-26.5 0-48 21.5-48 48zm79.8 78.7c3.3-8.7 11.2-14.7 20.5-14.7H539.7c9.2 0 17.2 6 20.5 14.7L576 304H352l15.8-49.3zM568 352a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM336 376a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z\"],\"record-vinyl\":[512,512,[],\"f8d9\",\"M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm256-96a96 96 0 1 1 0 192 96 96 0 1 1 0-192zm0 224a128 128 0 1 0 0-256 128 128 0 1 0 0 256zm0-96a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"face-grin-stars\":[512,512,[129321,\"grin-stars\"],\"f587\",\"M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm407.4 75.5c5-11.8-7-22.5-19.3-18.7c-39.7 12.2-84.5 19-131.8 19s-92.1-6.8-131.8-19c-12.3-3.8-24.3 6.9-19.3 18.7c25 59.1 83.2 100.5 151.1 100.5s126.2-41.4 151.1-100.5zM160 120c-3.1 0-5.9 1.8-7.2 4.6l-16.6 34.7-38.1 5c-3.1 .4-5.6 2.5-6.6 5.5s-.1 6.2 2.1 8.3l27.9 26.5-7 37.8c-.6 3 .7 6.1 3.2 7.9s5.8 2 8.5 .6L160 232.5l33.8 18.3c2.7 1.5 6 1.3 8.5-.6s3.7-4.9 3.2-7.9l-7-37.8L226.4 178c2.2-2.1 3.1-5.3 2.1-8.3s-3.5-5.1-6.6-5.5l-38.1-5-16.6-34.7c-1.3-2.8-4.1-4.6-7.2-4.6zm192 0c-3.1 0-5.9 1.8-7.2 4.6l-16.6 34.7-38.1 5c-3.1 .4-5.6 2.5-6.6 5.5s-.1 6.2 2.1 8.3l27.9 26.5-7 37.8c-.6 3 .7 6.1 3.2 7.9s5.8 2 8.5 .6L352 232.5l33.8 18.3c2.7 1.5 6 1.3 8.5-.6s3.7-4.9 3.2-7.9l-7-37.8L418.4 178c2.2-2.1 3.1-5.3 2.1-8.3s-3.5-5.1-6.6-5.5l-38.1-5-16.6-34.7c-1.3-2.8-4.1-4.6-7.2-4.6z\"],bong:[448,512,[],\"f55c\",\"M160 208.5c0 29.1-15.6 53.9-37.2 67.8c-17.2 11.1-31.5 26.1-41.7 43.7H302.9c-10.2-17.6-24.5-32.6-41.7-43.7c-21.6-13.9-37.2-38.7-37.2-67.8V64H160V208.5zM288 64V208.5c0 5.7 3.1 10.9 7.9 14c11.2 7.2 21.5 15.5 30.9 24.8L366.1 208l-7-7c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l24 24 24 24c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-7-7-43.3 43.3C374 314.1 384 347.9 384 384c0 39.4-11.9 76.1-32.2 106.5c-9.6 14.4-26.5 21.5-43.8 21.5H76.1c-17.3 0-34.2-7.1-43.8-21.5C11.9 460.1 0 423.4 0 384c0-67.8 35.1-127.3 88.1-161.5c4.8-3.1 7.9-8.3 7.9-14V64C78.3 64 64 49.7 64 32S78.3 0 96 0h16H272h16c17.7 0 32 14.3 32 32s-14.3 32-32 32z\"],\"spaghetti-monster-flying\":[640,512,[\"pastafarianism\"],\"f67b\",\"M208 64a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm48 0c0 16.2-6 31.1-16 42.3l15.6 31.2c18.7-6 39.9-9.5 64.4-9.5s45.8 3.5 64.4 9.5L400 106.3C390 95.1 384 80.2 384 64c0-35.3 28.7-64 64-64s64 28.7 64 64s-28.7 64-64 64c-1.7 0-3.4-.1-5.1-.2L427.8 158c21.1 13.6 37.7 30.2 51.4 46.4c7.1 8.3 13.5 16.6 19.3 24l1.4 1.8c6.3 8.1 11.6 14.8 16.7 20.4C527.3 262.3 532.7 264 536 264c2.5 0 4.3-.6 7.1-3.3c3.7-3.5 7.1-8.8 12.5-17.4l.6-.9c4.6-7.4 11-17.6 19.4-25.7c9.7-9.3 22.9-16.7 40.4-16.7c13.3 0 24 10.7 24 24s-10.7 24-24 24c-2.5 0-4.3 .6-7.1 3.3c-3.7 3.5-7.1 8.8-12.5 17.4l-.6 .9c-4.6 7.4-11 17.6-19.4 25.7c-9.7 9.3-22.9 16.7-40.4 16.7c-18.5 0-32.9-8.5-44.3-18.6c-3.1 4-6.6 8.3-10.5 12.7c1.4 4.3 2.8 8.5 4 12.5c.9 3 1.8 5.8 2.6 8.6c3 9.8 5.5 18.2 8.6 25.9c3.9 9.8 7.4 15.4 10.8 18.5c2.6 2.4 5.9 4.3 12.8 4.3c8.7 0 16.9-4.2 33.7-13.2c15-8 35.7-18.8 62.3-18.8c13.3 0 24 10.7 24 24s-10.7 24-24 24c-13.4 0-24.7 5.2-39.7 13.2c-1 .6-2.1 1.1-3.2 1.7C559.9 414 541.4 424 520 424c-18.4 0-33.6-6.1-45.5-17.2c-11.1-10.3-17.9-23.7-22.7-36c-3.6-9-6.7-19.1-9.5-28.5c-16.4 12.3-36.1 23.6-58.9 31.3c3.6 10.8 8.4 23.5 14.4 36.2c7.5 15.9 16.2 30.4 25.8 40.5C433 460.5 441.2 464 448 464c13.3 0 24 10.7 24 24s-10.7 24-24 24c-25.2 0-45-13.5-59.5-28.8c-14.5-15.4-25.7-34.9-34.2-53c-8-17-14.1-33.8-18.3-46.9c-5.2 .4-10.6 .6-16 .6s-10.8-.2-16-.6c-4.2 13-10.3 29.9-18.3 46.9c-8.5 18.1-19.8 37.6-34.2 53C237 498.5 217.2 512 192 512c-13.3 0-24-10.7-24-24s10.7-24 24-24c6.8 0 15-3.5 24.5-13.7c9.5-10.1 18.3-24.6 25.8-40.5c5.9-12.6 10.7-25.4 14.4-36.2c-22.8-7.7-42.5-19-58.9-31.3c-2.9 9.4-6 19.5-9.5 28.5c-4.8 12.2-11.6 25.6-22.7 36C153.6 417.9 138.4 424 120 424c-21.4 0-39.9-10-53.1-17.1l0 0c-1.1-.6-2.2-1.2-3.2-1.7c-15-8-26.3-13.2-39.7-13.2c-13.3 0-24-10.7-24-24s10.7-24 24-24c26.6 0 47.3 10.8 62.3 18.8c16.8 9 25 13.2 33.7 13.2c6.8 0 10.2-1.9 12.8-4.3c3.4-3.2 7-8.8 10.8-18.5c3-7.7 5.6-16.1 8.6-25.9c.8-2.7 1.7-5.6 2.6-8.6c1.2-4 2.6-8.2 4-12.5c-3.9-4.5-7.4-8.8-10.5-12.7C136.9 303.5 122.5 312 104 312c-17.5 0-30.7-7.4-40.4-16.7c-8.4-8.1-14.8-18.3-19.4-25.7l-.6-.9c-5.4-8.6-8.8-13.9-12.5-17.4c-2.8-2.7-4.6-3.3-7.1-3.3c-13.3 0-24-10.7-24-24s10.7-24 24-24c17.5 0 30.7 7.4 40.4 16.7c8.4 8.1 14.8 18.3 19.4 25.7l.6 .9c5.4 8.6 8.8 13.9 12.5 17.4c2.8 2.7 4.6 3.3 7.1 3.3c3.3 0 8.7-1.7 19.4-13.4c5.1-5.6 10.4-12.3 16.7-20.4l1.4-1.8c5.8-7.4 12.2-15.7 19.3-24c13.8-16.2 30.3-32.8 51.4-46.4l-15.1-30.2c-1.7 .1-3.4 .2-5.1 .2c-35.3 0-64-28.7-64-64s28.7-64 64-64s64 28.7 64 64zm208 0a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z\"],\"arrow-down-up-across-line\":[576,512,[],\"e4af\",\"M137.4 502.6c12.5 12.5 32.8 12.5 45.3 0l96-96c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L192 402.7V288H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H448V109.3l41.4 41.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-96-96c-12.5-12.5-32.8-12.5-45.3 0l-96 96c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L384 109.3V224H192 128 32c-17.7 0-32 14.3-32 32s14.3 32 32 32h96V402.7L86.6 361.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l96 96zM128 192h64V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V192zM448 320H384V448c0 17.7 14.3 32 32 32s32-14.3 32-32V320z\"],spoon:[512,512,[129348,61873,\"utensil-spoon\"],\"f2e5\",\"M245.8 220.9c-14.5-17.6-21.8-39.2-21.8-60.8C224 80 320 0 416 0c53 0 96 43 96 96c0 96-80 192-160.2 192c-21.6 0-43.2-7.3-60.8-21.8L54.6 502.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L245.8 220.9z\"],\"jar-wheat\":[320,512,[],\"e517\",\"M32 32C32 14.3 46.3 0 64 0H256c17.7 0 32 14.3 32 32s-14.3 32-32 32H64C46.3 64 32 49.7 32 32zM0 160c0-35.3 28.7-64 64-64H256c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V160zm112 0H69.8c-3.2 0-5.8 2.6-5.8 5.8C64 198 90 224 122.2 224H144h32 21.8c32.1 0 58.2-26 58.2-58.2c0-3.2-2.6-5.8-5.8-5.8H208c-19.1 0-36.3 8.4-48 21.7c-11.7-13.3-28.9-21.7-48-21.7zm48 117.7c-11.7-13.3-28.9-21.7-48-21.7H69.8c-3.2 0-5.8 2.6-5.8 5.8C64 294 90 320 122.2 320H144h32 21.8c32.1 0 58.2-26 58.2-58.2c0-3.2-2.6-5.8-5.8-5.8H208c-19.1 0-36.3 8.4-48 21.7zM112 352H69.8c-3.2 0-5.8 2.6-5.8 5.8C64 390 90 416 122.2 416H144v32c0 8.8 7.2 16 16 16s16-7.2 16-16V416h21.8c32.1 0 58.2-26 58.2-58.2c0-3.2-2.6-5.8-5.8-5.8H208c-19.1 0-36.3 8.4-48 21.7c-11.7-13.3-28.9-21.7-48-21.7z\"],\"envelopes-bulk\":[640,512,[\"mail-bulk\"],\"f674\",\"M128 0C110.3 0 96 14.3 96 32V224h96V192c0-35.3 28.7-64 64-64H480V32c0-17.7-14.3-32-32-32H128zM256 160c-17.7 0-32 14.3-32 32v32h96c35.3 0 64 28.7 64 64V416H576c17.7 0 32-14.3 32-32V192c0-17.7-14.3-32-32-32H256zm240 64h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H496c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zM64 256c-17.7 0-32 14.3-32 32v13L187.1 415.9c1.4 1 3.1 1.6 4.9 1.6s3.5-.6 4.9-1.6L352 301V288c0-17.7-14.3-32-32-32H64zm288 84.8L216 441.6c-6.9 5.1-15.3 7.9-24 7.9s-17-2.8-24-7.9L32 340.8V480c0 17.7 14.3 32 32 32H320c17.7 0 32-14.3 32-32V340.8z\"],\"file-circle-exclamation\":[576,512,[],\"e4eb\",\"M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v38.6C310.1 219.5 256 287.4 256 368c0 59.1 29.1 111.3 73.7 143.3c-3.2 .5-6.4 .7-9.7 .7H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zm48 96a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm0 240a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm0-192c-8.8 0-16 7.2-16 16v80c0 8.8 7.2 16 16 16s16-7.2 16-16V288c0-8.8-7.2-16-16-16z\"],\"circle-h\":[512,512,[9405,\"hospital-symbol\"],\"f47e\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM368 152V256 360c0 13.3-10.7 24-24 24s-24-10.7-24-24V280H192l0 80c0 13.3-10.7 24-24 24s-24-10.7-24-24l0-208c0-13.3 10.7-24 24-24s24 10.7 24 24v80H320V152c0-13.3 10.7-24 24-24s24 10.7 24 24z\"],pager:[512,512,[128223],\"f815\",\"M0 128C0 92.7 28.7 64 64 64H448c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128zm64 32v64c0 17.7 14.3 32 32 32H416c17.7 0 32-14.3 32-32V160c0-17.7-14.3-32-32-32H96c-17.7 0-32 14.3-32 32zM80 320c-13.3 0-24 10.7-24 24s10.7 24 24 24h56c13.3 0 24-10.7 24-24s-10.7-24-24-24H80zm136 0c-13.3 0-24 10.7-24 24s10.7 24 24 24h48c13.3 0 24-10.7 24-24s-10.7-24-24-24H216z\"],\"address-book\":[512,512,[62138,\"contact-book\"],\"f2b9\",\"M96 0C60.7 0 32 28.7 32 64V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H96zM208 288h64c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16H144c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80zm-32-96a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM512 80c0-8.8-7.2-16-16-16s-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V80zM496 192c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V208c0-8.8-7.2-16-16-16zm16 144c0-8.8-7.2-16-16-16s-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V336z\"],strikethrough:[512,512,[],\"f0cc\",\"M161.3 144c3.2-17.2 14-30.1 33.7-38.6c21.1-9 51.8-12.3 88.6-6.5c11.9 1.9 48.8 9.1 60.1 12c17.1 4.5 34.6-5.6 39.2-22.7s-5.6-34.6-22.7-39.2c-14.3-3.8-53.6-11.4-66.6-13.4c-44.7-7-88.3-4.2-123.7 10.9c-36.5 15.6-64.4 44.8-71.8 87.3c-.1 .6-.2 1.1-.2 1.7c-2.8 23.9 .5 45.6 10.1 64.6c4.5 9 10.2 16.9 16.7 23.9H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H270.1c-.1 0-.3-.1-.4-.1l-1.1-.3c-36-10.8-65.2-19.6-85.2-33.1c-9.3-6.3-15-12.6-18.2-19.1c-3.1-6.1-5.2-14.6-3.8-27.4zM348.9 337.2c2.7 6.5 4.4 15.8 1.9 30.1c-3 17.6-13.8 30.8-33.9 39.4c-21.1 9-51.7 12.3-88.5 6.5c-18-2.9-49.1-13.5-74.4-22.1c-5.6-1.9-11-3.7-15.9-5.4c-16.8-5.6-34.9 3.5-40.5 20.3s3.5 34.9 20.3 40.5c3.6 1.2 7.9 2.7 12.7 4.3l0 0 0 0c24.9 8.5 63.6 21.7 87.6 25.6l0 0 .2 0c44.7 7 88.3 4.2 123.7-10.9c36.5-15.6 64.4-44.8 71.8-87.3c3.6-21 2.7-40.4-3.1-58.1H335.1c7 5.6 11.4 11.2 13.9 17.2z\"],k:[320,512,[107],\"4b\",\"M311 86.3c12.3-12.7 12-32.9-.7-45.2s-32.9-12-45.2 .7l-155.2 160L64 249V64c0-17.7-14.3-32-32-32S0 46.3 0 64V328 448c0 17.7 14.3 32 32 32s32-14.3 32-32V341l64.7-66.7 133 192c10.1 14.5 30 18.1 44.5 8.1s18.1-30 8.1-44.5L174.1 227.4 311 86.3z\"],\"landmark-flag\":[512,512,[],\"e51c\",\"M272 0h80c8.8 0 16 7.2 16 16V80c0 8.8-7.2 16-16 16H272v32H464c17.7 0 32 14.3 32 32s-14.3 32-32 32H48c-17.7 0-32-14.3-32-32s14.3-32 32-32H240V16c0-8.8 7.2-16 16-16h16zM64 224h64V416h40V224h64V416h48V224h64V416h40V224h64V420.3c.6 .3 1.2 .7 1.8 1.1l48 32c11.7 7.8 17 22.4 12.9 35.9S494.1 512 480 512H32c-14.1 0-26.5-9.2-30.6-22.7s1.1-28.1 12.9-35.9l48-32c.6-.4 1.2-.7 1.8-1.1V224z\"],pencil:[512,512,[9999,61504,\"pencil-alt\"],\"f303\",\"M410.3 231l11.3-11.3-33.9-33.9-62.1-62.1L291.7 89.8l-11.3 11.3-22.6 22.6L58.6 322.9c-10.4 10.4-18 23.3-22.2 37.4L1 480.7c-2.5 8.4-.2 17.5 6.1 23.7s15.3 8.5 23.7 6.1l120.3-35.4c14.1-4.2 27-11.8 37.4-22.2L387.7 253.7 410.3 231zM160 399.4l-9.1 22.7c-4 3.1-8.5 5.4-13.3 6.9L59.4 452l23-78.1c1.4-4.9 3.8-9.4 6.9-13.3l22.7-9.1v32c0 8.8 7.2 16 16 16h32zM362.7 18.7L348.3 33.2 325.7 55.8 314.3 67.1l33.9 33.9 62.1 62.1 33.9 33.9 11.3-11.3 22.6-22.6 14.5-14.5c25-25 25-65.5 0-90.5L453.3 18.7c-25-25-65.5-25-90.5 0zm-47.4 168l-144 144c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6l144-144c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z\"],backward:[512,512,[9194],\"f04a\",\"M459.5 440.6c9.5 7.9 22.8 9.7 34.1 4.4s18.4-16.6 18.4-29V96c0-12.4-7.2-23.7-18.4-29s-24.5-3.6-34.1 4.4L288 214.3V256v41.7L459.5 440.6zM256 352V256 128 96c0-12.4-7.2-23.7-18.4-29s-24.5-3.6-34.1 4.4l-192 160C4.2 237.5 0 246.5 0 256s4.2 18.5 11.5 24.6l192 160c9.5 7.9 22.8 9.7 34.1 4.4s18.4-16.6 18.4-29V352z\"],\"caret-right\":[256,512,[],\"f0da\",\"M246.6 278.6c12.5-12.5 12.5-32.8 0-45.3l-128-128c-9.2-9.2-22.9-11.9-34.9-6.9s-19.8 16.6-19.8 29.6l0 256c0 12.9 7.8 24.6 19.8 29.6s25.7 2.2 34.9-6.9l128-128z\"],comments:[640,512,[128490,61670],\"f086\",\"M208 352c114.9 0 208-78.8 208-176S322.9 0 208 0S0 78.8 0 176c0 38.6 14.7 74.3 39.6 103.4c-3.5 9.4-8.7 17.7-14.2 24.7c-4.8 6.2-9.7 11-13.3 14.3c-1.8 1.6-3.3 2.9-4.3 3.7c-.5 .4-.9 .7-1.1 .8l-.2 .2 0 0 0 0C1 327.2-1.4 334.4 .8 340.9S9.1 352 16 352c21.8 0 43.8-5.6 62.1-12.5c9.2-3.5 17.8-7.4 25.3-11.4C134.1 343.3 169.8 352 208 352zM448 176c0 112.3-99.1 196.9-216.5 207C255.8 457.4 336.4 512 432 512c38.2 0 73.9-8.7 104.7-23.9c7.5 4 16 7.9 25.2 11.4c18.3 6.9 40.3 12.5 62.1 12.5c6.9 0 13.1-4.5 15.2-11.1c2.1-6.6-.2-13.8-5.8-17.9l0 0 0 0-.2-.2c-.2-.2-.6-.4-1.1-.8c-1-.8-2.5-2-4.3-3.7c-3.6-3.3-8.5-8.1-13.3-14.3c-5.5-7-10.7-15.4-14.2-24.7c24.9-29 39.6-64.7 39.6-103.4c0-92.8-84.9-168.9-192.6-175.5c.4 5.1 .6 10.3 .6 15.5z\"],paste:[512,512,[\"file-clipboard\"],\"f0ea\",\"M160 0c-23.7 0-44.4 12.9-55.4 32H48C21.5 32 0 53.5 0 80V400c0 26.5 21.5 48 48 48H192V176c0-44.2 35.8-80 80-80h48V80c0-26.5-21.5-48-48-48H215.4C204.4 12.9 183.7 0 160 0zM272 128c-26.5 0-48 21.5-48 48V448v16c0 26.5 21.5 48 48 48H464c26.5 0 48-21.5 48-48V243.9c0-12.7-5.1-24.9-14.1-33.9l-67.9-67.9c-9-9-21.2-14.1-33.9-14.1H320 272zM160 40a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],\"code-pull-request\":[512,512,[],\"e13c\",\"M305.8 2.1C314.4 5.9 320 14.5 320 24V64h16c70.7 0 128 57.3 128 128V358.7c28.3 12.3 48 40.5 48 73.3c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-32.8 19.7-61 48-73.3V192c0-35.3-28.7-64-64-64H320v40c0 9.5-5.6 18.1-14.2 21.9s-18.8 2.3-25.8-4.1l-80-72c-5.1-4.6-7.9-11-7.9-17.8s2.9-13.3 7.9-17.8l80-72c7-6.3 17.2-7.9 25.8-4.1zM104 80A24 24 0 1 0 56 80a24 24 0 1 0 48 0zm8 73.3V358.7c28.3 12.3 48 40.5 48 73.3c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-32.8 19.7-61 48-73.3V153.3C19.7 141 0 112.8 0 80C0 35.8 35.8 0 80 0s80 35.8 80 80c0 32.8-19.7 61-48 73.3zM104 432a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zm328 24a24 24 0 1 0 0-48 24 24 0 1 0 0 48z\"],\"clipboard-list\":[384,512,[],\"f46d\",\"M192 0c-41.8 0-77.4 26.7-90.5 64H64C28.7 64 0 92.7 0 128V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H282.5C269.4 26.7 233.8 0 192 0zm0 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM72 272a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm104-16H304c8.8 0 16 7.2 16 16s-7.2 16-16 16H176c-8.8 0-16-7.2-16-16s7.2-16 16-16zM72 368a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm88 0c0-8.8 7.2-16 16-16H304c8.8 0 16 7.2 16 16s-7.2 16-16 16H176c-8.8 0-16-7.2-16-16z\"],\"truck-ramp-box\":[640,512,[\"truck-loading\"],\"f4de\",\"M640 0V400c0 61.9-50.1 112-112 112c-61 0-110.5-48.7-112-109.3L48.4 502.9c-17.1 4.6-34.6-5.4-39.3-22.5s5.4-34.6 22.5-39.3L352 353.8V64c0-35.3 28.7-64 64-64H640zM576 400a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM23.1 207.7c-4.6-17.1 5.6-34.6 22.6-39.2l46.4-12.4 20.7 77.3c2.3 8.5 11.1 13.6 19.6 11.3l30.9-8.3c8.5-2.3 13.6-11.1 11.3-19.6l-20.7-77.3 46.4-12.4c17.1-4.6 34.6 5.6 39.2 22.6l41.4 154.5c4.6 17.1-5.6 34.6-22.6 39.2L103.7 384.9c-17.1 4.6-34.6-5.6-39.2-22.6L23.1 207.7z\"],\"user-check\":[640,512,[],\"f4fc\",\"M96 128a128 128 0 1 1 256 0A128 128 0 1 1 96 128zM0 482.3C0 383.8 79.8 304 178.3 304h91.4C368.2 304 448 383.8 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3zM625 177L497 305c-9.4 9.4-24.6 9.4-33.9 0l-64-64c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47L591 143c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z\"],\"vial-virus\":[512,512,[],\"e597\",\"M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96V384c0 53 43 96 96 96c28.6 0 54.2-12.5 71.8-32.3c.1-14.2 5.6-28.3 16.4-39.1c.2-.2 .1-.6-.2-.6c-30.9 0-56-25.1-56-56s25.1-56 56-56c.3 0 .4-.4 .2-.6c-21.9-21.9-21.9-57.3 0-79.2c2.4-2.4 5-4.6 7.8-6.5V96c17.7 0 32-14.3 32-32s-14.3-32-32-32H160 96 32zM96 192V96h64v96H96zM216 376c28.8 0 43.2 34.8 22.9 55.2c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0c20.4-20.4 55.2-5.9 55.2 22.9c0 13.3 10.7 24 24 24s24-10.7 24-24c0-28.8 34.8-43.2 55.2-22.9c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9C444.8 410.8 459.2 376 488 376c13.3 0 24-10.7 24-24s-10.7-24-24-24c-28.8 0-43.2-34.8-22.9-55.2c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0C410.8 259.2 376 244.8 376 216c0-13.3-10.7-24-24-24s-24 10.7-24 24c0 28.8-34.8 43.2-55.2 22.9c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9c20.4 20.4 5.9 55.2-22.9 55.2c-13.3 0-24 10.7-24 24s10.7 24 24 24zm104-88a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm40 96a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z\"],\"sheet-plastic\":[384,512,[],\"e571\",\"M0 448c0 35.3 28.7 64 64 64H224V384c0-17.7 14.3-32 32-32H384V64c0-35.3-28.7-64-64-64H64C28.7 0 0 28.7 0 64V448zM171.3 75.3l-96 96c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6l96-96c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6zm96 32l-160 160c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6l160-160c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6zM384 384H256V512L384 384z\"],blog:[512,512,[],\"f781\",\"M192 32c0 17.7 14.3 32 32 32c123.7 0 224 100.3 224 224c0 17.7 14.3 32 32 32s32-14.3 32-32C512 128.9 383.1 0 224 0c-17.7 0-32 14.3-32 32zm0 96c0 17.7 14.3 32 32 32c70.7 0 128 57.3 128 128c0 17.7 14.3 32 32 32s32-14.3 32-32c0-106-86-192-192-192c-17.7 0-32 14.3-32 32zM96 144c0-26.5-21.5-48-48-48S0 117.5 0 144V368c0 79.5 64.5 144 144 144s144-64.5 144-144s-64.5-144-144-144H128v96h16c26.5 0 48 21.5 48 48s-21.5 48-48 48s-48-21.5-48-48V144z\"],\"user-ninja\":[448,512,[129399],\"f504\",\"M224 256c-57.2 0-105.6-37.5-122-89.3c-1.1 1.3-2.2 2.6-3.5 3.8c-15.8 15.8-38.8 20.7-53.6 22.1c-8.1 .8-14.6-5.7-13.8-13.8c1.4-14.7 6.3-37.8 22.1-53.6c5.8-5.8 12.6-10.1 19.6-13.4c-7-3.2-13.8-7.6-19.6-13.4C37.4 82.7 32.6 59.7 31.1 44.9c-.8-8.1 5.7-14.6 13.8-13.8c14.7 1.4 37.8 6.3 53.6 22.1c4.8 4.8 8.7 10.4 11.7 16.1C131.4 28.2 174.4 0 224 0c70.7 0 128 57.3 128 128s-57.3 128-128 128zM0 482.3C0 399.5 56.4 330 132.8 309.9c6-1.6 12.2 .9 15.9 5.8l62.5 83.3c6.4 8.5 19.2 8.5 25.6 0l62.5-83.3c3.7-4.9 9.9-7.4 15.9-5.8C391.6 330 448 399.5 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3zM160 96c-8.8 0-16 7.2-16 16s7.2 16 16 16H288c8.8 0 16-7.2 16-16s-7.2-16-16-16H160z\"],\"person-arrow-up-from-line\":[640,512,[],\"e539\",\"M192 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-8 352V352h16v96H184zm-64 0H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H152h80H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H264V256.9l28.6 47.5c9.1 15.1 28.8 20 43.9 10.9s20-28.8 10.9-43.9l-58.3-97c-17.4-28.9-48.6-46.6-82.3-46.6H177.1c-33.7 0-64.9 17.7-82.3 46.6l-58.3 97c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9L120 256.9V448zM598.6 121.4l-80-80c-12.5-12.5-32.8-12.5-45.3 0l-80 80c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L464 141.3 464 384c0 17.7 14.3 32 32 32s32-14.3 32-32V141.3l25.4 25.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3z\"],\"scroll-torah\":[640,512,[\"torah\"],\"f6a0\",\"M96 480V32C96 14.3 74.5 0 48 0S0 14.3 0 32V480c0 17.7 21.5 32 48 32s48-14.3 48-32zM512 32H128V480H512V32zM592 0c-26.5 0-48 14.3-48 32V480c0 17.7 21.5 32 48 32s48-14.3 48-32V32c0-17.7-21.5-32-48-32zM196 313.7c0-3.2 .9-6.4 2.5-9.2L226.7 256l-28.3-48.5c-1.6-2.8-2.5-6-2.5-9.2c0-10.1 8.2-18.3 18.3-18.3H271l31.4-53.9c3.6-6.3 10.3-10.1 17.6-10.1s13.9 3.8 17.6 10.1L369 180h56.7c10.1 0 18.3 8.2 18.3 18.3c0 3.2-.9 6.4-2.5 9.2L413.3 256l28.3 48.5c1.6 2.8 2.5 6 2.5 9.2c0 10.1-8.2 18.3-18.3 18.3H369l-31.4 53.9c-3.6 6.3-10.3 10.1-17.6 10.1s-13.9-3.8-17.6-10.1L271 332H214.3c-10.1 0-18.3-8.2-18.3-18.3zm124 54.7L341.2 332H298.8L320 368.4zM254.5 256l30.3 52h70.4l30.3-52-30.3-52H284.8l-30.3 52zm144.9 23.8L383 308h32.8l-16.4-28.2zM415.8 204H383l16.4 28.2L415.8 204zM320 143.6L298.8 180h42.4L320 143.6zM224.2 204l16.4 28.2L257 204H224.2zM257 308l-16.4-28.2L224.2 308H257z\"],\"broom-ball\":[576,512,[\"quidditch\",\"quidditch-broom-ball\"],\"f458\",\"M566.6 9.4c12.5 12.5 12.5 32.8 0 45.3l-192 192 34.7 34.7c4.2 4.2 6.6 10 6.6 16c0 12.5-10.1 22.6-22.6 22.6H364.3L256 211.7V182.6c0-12.5 10.1-22.6 22.6-22.6c6 0 11.8 2.4 16 6.6l34.7 34.7 192-192c12.5-12.5 32.8-12.5 45.3 0zm-344 225.5L341.1 353.4c3.7 42.7-11.7 85.2-42.3 115.8C271.4 496.6 234.2 512 195.5 512L22.1 512C9.9 512 0 502.1 0 489.9c0-6.3 2.7-12.3 7.3-16.5L133.7 359.7c4.2-3.7-.4-10.4-5.4-7.9L77.2 377.4c-6.1 3-13.2-1.4-13.2-8.2c0-31.5 12.5-61.7 34.8-84l8-8c30.6-30.6 73.1-45.9 115.8-42.3zM464 352a80 80 0 1 1 0 160 80 80 0 1 1 0-160z\"],\"toggle-off\":[576,512,[],\"f204\",\"M384 128c70.7 0 128 57.3 128 128s-57.3 128-128 128H192c-70.7 0-128-57.3-128-128s57.3-128 128-128H384zM576 256c0-106-86-192-192-192H192C86 64 0 150 0 256S86 448 192 448H384c106 0 192-86 192-192zM192 352a96 96 0 1 0 0-192 96 96 0 1 0 0 192z\"],\"box-archive\":[512,512,[\"archive\"],\"f187\",\"M32 32H480c17.7 0 32 14.3 32 32V96c0 17.7-14.3 32-32 32H32C14.3 128 0 113.7 0 96V64C0 46.3 14.3 32 32 32zm0 128H480V416c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V160zm128 80c0 8.8 7.2 16 16 16H336c8.8 0 16-7.2 16-16s-7.2-16-16-16H176c-8.8 0-16 7.2-16 16z\"],\"person-drowning\":[576,512,[],\"e545\",\"M192 64c0-17.7-14.3-32-32-32s-32 14.3-32 32V96.2c0 54.1 23.5 104 62.2 138.3l-21 146.7c7.8 2.1 15.5 3.3 22.8 3.3c21.1 0 42-8.5 59.2-20.3c22.1-15.5 51.6-15.5 73.7 0c12.4 8.5 26.1 14.8 39.7 18l17.7-97.6c10.7-1.2 21.3-3.1 31.9-5.5l105-23.9c17.2-3.9 28-21.1 24.1-38.3s-21.1-28-38.3-24.1L400 216.6c-41 9.3-83.7 7.5-123.7-5.2c-50.2-16-84.3-62.6-84.3-115.3V64zM320 192a64 64 0 1 0 0-128 64 64 0 1 0 0 128zM306.5 389.9c-11.1-7.9-25.9-7.9-37 0C247 405.4 219.5 416 192 416c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C126.5 469.7 159 480 192 480c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.4-27.3-10.1-39.2-1.7l0 0C439.4 405.2 410.9 416 384 416c-27.5 0-55-10.6-77.5-26.1z\"],\"arrow-down-9-1\":[576,512,[\"sort-numeric-desc\",\"sort-numeric-down-alt\"],\"f886\",\"M160 480c9 0 17.5-3.8 23.6-10.4l88-96c11.9-13 11.1-33.3-2-45.2s-33.3-11.1-45.2 2L192 365.7V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V365.7L95.6 330.4c-11.9-13-32.2-13.9-45.2-2s-13.9 32.2-2 45.2l88 96C142.5 476.2 151 480 160 480zM450.7 294c-8.3-6-19.1-7.7-28.8-4.4l-48 16c-16.8 5.6-25.8 23.7-20.2 40.5s23.7 25.8 40.5 20.2l5.9-2V416H384c-17.7 0-32 14.3-32 32s14.3 32 32 32h48 48c17.7 0 32-14.3 32-32s-14.3-32-32-32H464V320c0-10.3-4.9-19.9-13.3-26zM418.3 91a32 32 0 1 1 27.4 57.9A32 32 0 1 1 418.3 91zM405.1 203.8l-6.8 9.2c-10.5 14.2-7.5 34.2 6.7 44.8s34.2 7.5 44.8-6.7l48.8-65.8c14-18.9 21.5-41.7 21.5-65.2c0-48.6-39.4-88-88-88s-88 39.4-88 88c0 39.2 25.6 72.4 61.1 83.8z\"],\"face-grin-tongue-squint\":[512,512,[128541,\"grin-tongue-squint\"],\"f58a\",\"M0 256C0 368.9 73.1 464.7 174.5 498.8C165.3 484 160 466.6 160 448V400.7c-24-17.5-43.1-41.4-54.8-69.2c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19c12.3-3.8 24.3 6.9 19.3 18.7c-11.8 28-31.1 52-55.4 69.6V448c0 18.6-5.3 36-14.5 50.8C438.9 464.7 512 368.9 512 256C512 114.6 397.4 0 256 0S0 114.6 0 256zM116 141.1c0-9 9.6-14.7 17.5-10.5l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6zm262.5-10.5c7.9-4.2 17.5 1.5 17.5 10.5c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9zM320 448V402.6c0-14.7-11.9-26.6-26.6-26.6h-2c-11.3 0-21.1 7.9-23.6 18.9c-2.8 12.6-20.8 12.6-23.6 0c-2.5-11.1-12.3-18.9-23.6-18.9h-2c-14.7 0-26.6 11.9-26.6 26.6V448c0 35.3 28.7 64 64 64s64-28.7 64-64z\"],\"spray-can\":[512,512,[],\"f5bd\",\"M128 0h64c17.7 0 32 14.3 32 32v96H96V32c0-17.7 14.3-32 32-32zM0 256c0-53 43-96 96-96H224c53 0 96 43 96 96V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V256zm240 80A80 80 0 1 0 80 336a80 80 0 1 0 160 0zM256 64a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM384 32a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm64 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm32 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM448 256a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM384 128a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"truck-monster\":[640,512,[],\"f63b\",\"M288 64v64H416L368 64H288zM419.2 25.6L496 128h80c17.7 0 32 14.3 32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32c-29.2-38.9-75.7-64-128-64s-98.8 25.1-128 64H288c-29.2-38.9-75.7-64-128-64s-98.8 25.1-128 64c-17.7 0-32-14.3-32-32s14.3-32 32-32V160c0-17.7 14.3-32 32-32H224V48c0-26.5 21.5-48 48-48h96c20.1 0 39.1 9.5 51.2 25.6zM152 256h16c12.1 0 22.1 8.9 23.8 20.6c7.6 2.2 14.9 5.3 21.7 9c9.4-7 22.8-6.3 31.3 2.3l11.3 11.3c8.6 8.6 9.3 21.9 2.3 31.3c3.7 6.8 6.8 14.1 9 21.7c11.6 1.7 20.6 11.7 20.6 23.8v16c0 12.1-8.9 22.1-20.6 23.8c-2.2 7.6-5.3 14.9-9 21.7c7 9.4 6.3 22.8-2.3 31.3l-11.3 11.3c-8.6 8.6-21.9 9.3-31.3 2.2c-6.8 3.7-14.1 6.8-21.7 9C190.1 503.1 180.1 512 168 512H152c-12.1 0-22.1-8.9-23.8-20.6c-7.6-2.2-14.9-5.3-21.7-9c-9.4 7.1-22.8 6.3-31.3-2.2L63.8 468.9c-8.6-8.6-9.3-21.9-2.3-31.3c-3.7-6.9-6.8-14.1-9-21.8C40.9 414.1 32 404.1 32 392V376c0-12.1 8.9-22.1 20.6-23.8c2.2-7.6 5.3-14.9 9-21.8c-7-9.4-6.3-22.8 2.3-31.3l11.3-11.3c8.6-8.6 21.9-9.3 31.3-2.3c6.8-3.7 14.1-6.8 21.7-9c1.7-11.6 11.7-20.6 23.8-20.6zm8 176a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM448.2 276.6c1.7-11.6 11.7-20.6 23.8-20.6h16c12.1 0 22.1 8.9 23.8 20.6c7.6 2.2 14.9 5.3 21.8 9c9.4-7 22.8-6.3 31.3 2.3l11.3 11.3c8.6 8.6 9.3 21.9 2.2 31.3c3.7 6.8 6.8 14.1 9 21.7c11.6 1.7 20.6 11.7 20.6 23.8v16c0 12.1-8.9 22.1-20.6 23.8c-2.2 7.6-5.3 14.9-9 21.7c7 9.4 6.3 22.8-2.2 31.3l-11.3 11.3c-8.6 8.6-21.9 9.3-31.3 2.2c-6.9 3.7-14.1 6.8-21.8 9C510.1 503.1 500.1 512 488 512H472c-12.1 0-22.1-8.9-23.8-20.6c-7.6-2.2-14.9-5.3-21.7-9c-9.4 7.1-22.8 6.3-31.3-2.2l-11.3-11.3c-8.6-8.6-9.3-21.9-2.2-31.3c-3.7-6.9-6.8-14.1-9-21.8C360.9 414.1 352 404.1 352 392V376c0-12.1 8.9-22.1 20.6-23.8c2.2-7.6 5.3-14.9 9-21.8c-7-9.4-6.3-22.8 2.2-31.3l11.3-11.3c8.6-8.6 21.9-9.3 31.3-2.3c6.8-3.7 14.1-6.8 21.7-9zM528 384a48 48 0 1 0 -96 0 48 48 0 1 0 96 0z\"],w:[576,512,[119],\"57\",\"M20.8 34c16.5-6.2 35 2.2 41.2 18.7l110.2 294L257.3 55c4-13.7 16.5-23 30.7-23s26.7 9.4 30.7 23l85.1 291.7L514 52.8c6.2-16.5 24.6-24.9 41.2-18.7s24.9 24.7 18.7 41.2l-144 384c-4.8 12.9-17.4 21.3-31.2 20.7s-25.7-9.8-29.5-23L288 178.3 206.7 457c-3.9 13.2-15.8 22.5-29.5 23s-26.3-7.8-31.2-20.7L2 75.2C-4.2 58.7 4.2 40.2 20.8 34z\"],\"earth-africa\":[512,512,[127757,\"globe-africa\"],\"f57c\",\"M177.8 63.2l10 17.4c2.8 4.8 4.2 10.3 4.2 15.9v41.4c0 3.9 1.6 7.7 4.3 10.4c6.2 6.2 16.5 5.7 22-1.2l13.6-17c4.7-5.9 12.9-7.7 19.6-4.3l15.2 7.6c3.4 1.7 7.2 2.6 11 2.6c6.5 0 12.8-2.6 17.4-7.2l3.9-3.9c2.9-2.9 7.3-3.6 11-1.8l29.2 14.6c7.8 3.9 12.6 11.8 12.6 20.5c0 10.5-7.1 19.6-17.3 22.2l-35.4 8.8c-7.4 1.8-15.1 1.5-22.4-.9l-32-10.7c-3.3-1.1-6.7-1.7-10.2-1.7c-7 0-13.8 2.3-19.4 6.5L176 212c-10.1 7.6-16 19.4-16 32v28c0 26.5 21.5 48 48 48h32c8.8 0 16 7.2 16 16v48c0 17.7 14.3 32 32 32c10.1 0 19.6-4.7 25.6-12.8l25.6-34.1c8.3-11.1 12.8-24.6 12.8-38.4V318.6c0-3.9 2.6-7.3 6.4-8.2l5.3-1.3c11.9-3 20.3-13.7 20.3-26c0-7.1-2.8-13.9-7.8-18.9l-33.5-33.5c-3.7-3.7-3.7-9.7 0-13.4c5.7-5.7 14.1-7.7 21.8-5.1l14.1 4.7c12.3 4.1 25.7-1.5 31.5-13c3.5-7 11.2-10.8 18.9-9.2l27.4 5.5C432 112.4 351.5 48 256 48c-27.7 0-54 5.4-78.2 15.2zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256z\"],rainbow:[640,512,[127752],\"f75b\",\"M320 96C178.6 96 64 210.6 64 352v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352C0 175.3 143.3 32 320 32s320 143.3 320 320v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352C576 210.6 461.4 96 320 96zm0 192c-35.3 0-64 28.7-64 64v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352c0-70.7 57.3-128 128-128s128 57.3 128 128v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352c0-35.3-28.7-64-64-64zM160 352v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352c0-123.7 100.3-224 224-224s224 100.3 224 224v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352c0-88.4-71.6-160-160-160s-160 71.6-160 160z\"],\"circle-notch\":[512,512,[],\"f1ce\",\"M222.7 32.1c5 16.9-4.6 34.8-21.5 39.8C121.8 95.6 64 169.1 64 256c0 106 86 192 192 192s192-86 192-192c0-86.9-57.8-160.4-137.1-184.1c-16.9-5-26.6-22.9-21.5-39.8s22.9-26.6 39.8-21.5C434.9 42.1 512 140 512 256c0 141.4-114.6 256-256 256S0 397.4 0 256C0 140 77.1 42.1 182.9 10.6c16.9-5 34.8 4.6 39.8 21.5z\"],\"tablet-screen-button\":[448,512,[\"tablet-alt\"],\"f3fa\",\"M0 64C0 28.7 28.7 0 64 0H384c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM256 448a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM384 64H64V384H384V64z\"],paw:[512,512,[],\"f1b0\",\"M226.5 92.9c14.3 42.9-.3 86.2-32.6 96.8s-70.1-15.6-84.4-58.5s.3-86.2 32.6-96.8s70.1 15.6 84.4 58.5zM100.4 198.6c18.9 32.4 14.3 70.1-10.2 84.1s-59.7-.9-78.5-33.3S-2.7 179.3 21.8 165.3s59.7 .9 78.5 33.3zM69.2 401.2C121.6 259.9 214.7 224 256 224s134.4 35.9 186.8 177.2c3.6 9.7 5.2 20.1 5.2 30.5v1.6c0 25.8-20.9 46.7-46.7 46.7c-11.5 0-22.9-1.4-34-4.2l-88-22c-15.3-3.8-31.3-3.8-46.6 0l-88 22c-11.1 2.8-22.5 4.2-34 4.2C84.9 480 64 459.1 64 433.3v-1.6c0-10.4 1.6-20.8 5.2-30.5zM421.8 282.7c-24.5-14-29.1-51.7-10.2-84.1s54-47.3 78.5-33.3s29.1 51.7 10.2 84.1s-54 47.3-78.5 33.3zM310.1 189.7c-32.3-10.6-46.9-53.9-32.6-96.8s52.1-69.1 84.4-58.5s46.9 53.9 32.6 96.8s-52.1 69.1-84.4 58.5z\"],cloud:[640,512,[9729],\"f0c2\",\"M0 336c0 79.5 64.5 144 144 144H512c70.7 0 128-57.3 128-128c0-61.9-44-113.6-102.4-125.4c4.1-10.7 6.4-22.4 6.4-34.6c0-53-43-96-96-96c-19.7 0-38.1 6-53.3 16.2C367 64.2 315.3 32 256 32C167.6 32 96 103.6 96 192c0 2.7 .1 5.4 .2 8.1C40.2 219.8 0 273.2 0 336z\"],\"trowel-bricks\":[512,512,[],\"e58a\",\"M240.8 4.8C250.3 10.6 256 20.9 256 32v72h89c3.6-13.8 16.1-24 31-24h88c26.5 0 48 21.5 48 48s-21.5 48-48 48H376c-14.9 0-27.4-10.2-31-24H256v72c0 11.1-5.7 21.4-15.2 27.2s-21.2 6.4-31.1 1.4l-192-96C6.8 151.2 0 140.1 0 128s6.8-23.2 17.7-28.6l192-96c9.9-5 21.7-4.4 31.1 1.4zM288 256c0-17.7 14.3-32 32-32H480c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H320c-17.7 0-32-14.3-32-32V256zM32 384h96c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32V416c0-17.7 14.3-32 32-32zm192 0H480c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H224c-17.7 0-32-14.3-32-32V416c0-17.7 14.3-32 32-32z\"],\"face-flushed\":[512,512,[128563,\"flushed\"],\"f579\",\"M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM176 384c0 8.8 7.2 16 16 16H320c8.8 0 16-7.2 16-16s-7.2-16-16-16H192c-8.8 0-16 7.2-16 16zm-16-88a72 72 0 1 0 0-144 72 72 0 1 0 0 144zm264-72a72 72 0 1 0 -144 0 72 72 0 1 0 144 0zm-288 0a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm192 0a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z\"],\"hospital-user\":[576,512,[],\"f80d\",\"M48 0C21.5 0 0 21.5 0 48V256H144c8.8 0 16 7.2 16 16s-7.2 16-16 16H0v64H144c8.8 0 16 7.2 16 16s-7.2 16-16 16H0v80c0 26.5 21.5 48 48 48H265.9c-6.3-10.2-9.9-22.2-9.9-35.1c0-46.9 25.8-87.8 64-109.2V271.8 48c0-26.5-21.5-48-48-48H48zM152 64h16c8.8 0 16 7.2 16 16v24h24c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H184v24c0 8.8-7.2 16-16 16H152c-8.8 0-16-7.2-16-16V152H112c-8.8 0-16-7.2-16-16V120c0-8.8 7.2-16 16-16h24V80c0-8.8 7.2-16 16-16zM512 272a80 80 0 1 0 -160 0 80 80 0 1 0 160 0zM288 477.1c0 19.3 15.6 34.9 34.9 34.9H541.1c19.3 0 34.9-15.6 34.9-34.9c0-51.4-41.7-93.1-93.1-93.1H381.1c-51.4 0-93.1 41.7-93.1 93.1z\"],\"tent-arrow-left-right\":[576,512,[],\"e57f\",\"M488.1 6.2c-9.9-8.9-25-8.1-33.9 1.8s-8.1 25 1.8 33.9L489.5 72 86.5 72l33.5-30.2c9.9-8.9 10.7-24 1.8-33.9S97.8-2.7 87.9 6.2l-80 72C2.9 82.7 0 89.2 0 96s2.9 13.3 7.9 17.8l80 72c9.9 8.9 25 8.1 33.9-1.8s8.1-25-1.8-33.9L86.5 120l402.9 0-33.5 30.2c-9.9 8.9-10.7 24-1.8 33.9s24 10.7 33.9 1.8l80-72c5.1-4.6 7.9-11 7.9-17.8s-2.9-13.3-7.9-17.8l-80-72zM307.4 166.5c-11.5-8.7-27.3-8.7-38.8 0l-168 128c-6.6 5-11 12.5-12.3 20.7l-24 160c-1.4 9.2 1.3 18.6 7.4 25.6S86.7 512 96 512H240h16c17.7 0 32-14.3 32-32V361.9c0-5.5 4.4-9.9 9.9-9.9c3.7 0 7.2 2.1 8.8 5.5l68.4 136.8c5.4 10.8 16.5 17.7 28.6 17.7H464h16c9.3 0 18.2-4.1 24.2-11.1s8.8-16.4 7.4-25.6l-24-160c-1.2-8.2-5.6-15.7-12.3-20.7l-168-128z\"],gavel:[512,512,[\"legal\"],\"f0e3\",\"M318.6 9.4c-12.5-12.5-32.8-12.5-45.3 0l-120 120c-12.5 12.5-12.5 32.8 0 45.3l16 16c12.5 12.5 32.8 12.5 45.3 0l4-4L325.4 293.4l-4 4c-12.5 12.5-12.5 32.8 0 45.3l16 16c12.5 12.5 32.8 12.5 45.3 0l120-120c12.5-12.5 12.5-32.8 0-45.3l-16-16c-12.5-12.5-32.8-12.5-45.3 0l-4 4L330.6 74.6l4-4c12.5-12.5 12.5-32.8 0-45.3l-16-16zm-152 288c-12.5-12.5-32.8-12.5-45.3 0l-112 112c-12.5 12.5-12.5 32.8 0 45.3l48 48c12.5 12.5 32.8 12.5 45.3 0l112-112c12.5-12.5 12.5-32.8 0-45.3l-1.4-1.4L272 285.3 226.7 240 168 298.7l-1.4-1.4z\"],binoculars:[512,512,[],\"f1e5\",\"M128 32h32c17.7 0 32 14.3 32 32V96H96V64c0-17.7 14.3-32 32-32zm64 96V448c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32V388.9c0-34.6 9.4-68.6 27.2-98.3C40.9 267.8 49.7 242.4 53 216L60.5 156c2-16 15.6-28 31.8-28H192zm227.8 0c16.1 0 29.8 12 31.8 28L459 216c3.3 26.4 12.1 51.8 25.8 74.6c17.8 29.7 27.2 63.7 27.2 98.3V448c0 17.7-14.3 32-32 32H352c-17.7 0-32-14.3-32-32V128h99.8zM320 64c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V96H320V64zm-32 64V288H224V128h64z\"],\"microphone-slash\":[640,512,[],\"f131\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L472.1 344.7c15.2-26 23.9-56.3 23.9-88.7V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v40c0 21.2-5.1 41.1-14.2 58.7L416 300.8V96c0-53-43-96-96-96s-96 43-96 96v54.3L38.8 5.1zm362.5 407l-43.1-33.9C346.1 382 333.3 384 320 384c-70.7 0-128-57.3-128-128v-8.7L144.7 210c-.5 1.9-.7 3.9-.7 6v40c0 89.1 66.2 162.7 152 174.4V464H248c-13.3 0-24 10.7-24 24s10.7 24 24 24h72 72c13.3 0 24-10.7 24-24s-10.7-24-24-24H344V430.4c20.4-2.8 39.7-9.1 57.3-18.2z\"],\"box-tissue\":[512,512,[],\"e05b\",\"M92.5 0H208c40 0 52 24 64 48s24 48 64 48h85.2C436 96 448 108 448 122.8c0 3.4-.7 6.8-1.9 10L409.6 224 384 288H128l-16-64L64.9 35.4c-.6-2.3-.9-4.6-.9-6.9C64 12.8 76.8 0 92.5 0zM79 224l16 64H80c-8.8 0-16 7.2-16 16s7.2 16 16 16h48H384h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H418.5l25.6-64H464c26.5 0 48 21.5 48 48V384H0V272c0-26.5 21.5-48 48-48H79zM0 416H512v48c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V416z\"],motorcycle:[640,512,[127949],\"f21c\",\"M280 32c-13.3 0-24 10.7-24 24s10.7 24 24 24h57.7l16.4 30.3L256 192l-45.3-45.3c-12-12-28.3-18.7-45.3-18.7H64c-17.7 0-32 14.3-32 32v32h96c88.4 0 160 71.6 160 160c0 11-1.1 21.7-3.2 32h70.4c-2.1-10.3-3.2-21-3.2-32c0-52.2 25-98.6 63.7-127.8l15.4 28.6C402.4 276.3 384 312 384 352c0 70.7 57.3 128 128 128s128-57.3 128-128s-57.3-128-128-128c-13.5 0-26.5 2.1-38.7 6L418.2 128H480c17.7 0 32-14.3 32-32V64c0-17.7-14.3-32-32-32H459.6c-7.5 0-14.7 2.6-20.5 7.4L391.7 78.9l-14-26c-7-12.9-20.5-21-35.2-21H280zM462.7 311.2l28.2 52.2c6.3 11.7 20.9 16 32.5 9.7s16-20.9 9.7-32.5l-28.2-52.2c2.3-.3 4.7-.4 7.1-.4c35.3 0 64 28.7 64 64s-28.7 64-64 64s-64-28.7-64-64c0-15.5 5.5-29.7 14.7-40.8zM187.3 376c-9.5 23.5-32.5 40-59.3 40c-35.3 0-64-28.7-64-64s28.7-64 64-64c26.9 0 49.9 16.5 59.3 40h66.4C242.5 268.8 190.5 224 128 224C57.3 224 0 281.3 0 352s57.3 128 128 128c62.5 0 114.5-44.8 125.8-104H187.3zM128 384a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"bell-concierge\":[512,512,[128718,\"concierge-bell\"],\"f562\",\"M216 64c-13.3 0-24 10.7-24 24s10.7 24 24 24h16v33.3C119.6 157.2 32 252.4 32 368H480c0-115.6-87.6-210.8-200-222.7V112h16c13.3 0 24-10.7 24-24s-10.7-24-24-24H256 216zM24 400c-13.3 0-24 10.7-24 24s10.7 24 24 24H488c13.3 0 24-10.7 24-24s-10.7-24-24-24H24z\"],\"pen-ruler\":[512,512,[\"pencil-ruler\"],\"f5ae\",\"M469.3 19.3l23.4 23.4c25 25 25 65.5 0 90.5l-56.4 56.4L322.3 75.7l56.4-56.4c25-25 65.5-25 90.5 0zM44.9 353.2L299.7 98.3 413.7 212.3 158.8 467.1c-6.7 6.7-15.1 11.6-24.2 14.2l-104 29.7c-8.4 2.4-17.4 .1-23.6-6.1s-8.5-15.2-6.1-23.6l29.7-104c2.6-9.2 7.5-17.5 14.2-24.2zM249.4 103.4L103.4 249.4 16 161.9c-18.7-18.7-18.7-49.1 0-67.9L94.1 16c18.7-18.7 49.1-18.7 67.9 0l19.8 19.8c-.3 .3-.7 .6-1 .9l-64 64c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0l64-64c.3-.3 .6-.7 .9-1l45.1 45.1zM408.6 262.6l45.1 45.1c-.3 .3-.7 .6-1 .9l-64 64c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0l64-64c.3-.3 .6-.7 .9-1L496 350.1c18.7 18.7 18.7 49.1 0 67.9L417.9 496c-18.7 18.7-49.1 18.7-67.9 0l-87.4-87.4L408.6 262.6z\"],\"people-arrows\":[640,512,[\"people-arrows-left-right\"],\"e068\",\"M64 64a64 64 0 1 1 128 0A64 64 0 1 1 64 64zM25.9 233.4C29.3 191.9 64 160 105.6 160h44.8c27 0 51 13.4 65.5 34.1c-2.7 1.9-5.2 4-7.5 6.3l-64 64c-21.9 21.9-21.9 57.3 0 79.2L192 391.2V464c0 26.5-21.5 48-48 48H112c-26.5 0-48-21.5-48-48V348.3c-26.5-9.5-44.7-35.8-42.2-65.6l4.1-49.3zM448 64a64 64 0 1 1 128 0A64 64 0 1 1 448 64zM431.6 200.4c-2.3-2.3-4.9-4.4-7.5-6.3c14.5-20.7 38.6-34.1 65.5-34.1h44.8c41.6 0 76.3 31.9 79.7 73.4l4.1 49.3c2.5 29.8-15.7 56.1-42.2 65.6V464c0 26.5-21.5 48-48 48H496c-26.5 0-48-21.5-48-48V391.2l47.6-47.6c21.9-21.9 21.9-57.3 0-79.2l-64-64zM272 240v32h96V240c0-9.7 5.8-18.5 14.8-22.2s19.3-1.7 26.2 5.2l64 64c9.4 9.4 9.4 24.6 0 33.9l-64 64c-6.9 6.9-17.2 8.9-26.2 5.2s-14.8-12.5-14.8-22.2V336H272v32c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-64-64c-9.4-9.4-9.4-24.6 0-33.9l64-64c6.9-6.9 17.2-8.9 26.2-5.2s14.8 12.5 14.8 22.2z\"],\"mars-and-venus-burst\":[640,512,[],\"e523\",\"M504 0c-9.7 0-18.5 5.8-22.2 14.8s-1.7 19.3 5.2 26.2l39 39-22.2 22.2C475.9 78.4 439.6 64 400 64c-88.4 0-160 71.6-160 160c0 80.2 59 146.6 136 158.2V408H352c-13.3 0-24 10.7-24 24s10.7 24 24 24h24v32c0 13.3 10.7 24 24 24s24-10.7 24-24V456h24c13.3 0 24-10.7 24-24s-10.7-24-24-24H424V382.2c77-11.6 136-78 136-158.2c0-31.4-9-60.7-24.7-85.4L560 113.9l39 39c6.9 6.9 17.2 8.9 26.2 5.2s14.8-12.5 14.8-22.2V24c0-13.3-10.7-24-24-24H504zM400 128a96 96 0 1 1 0 192 96 96 0 1 1 0-192zM190.9 18.1C188.4 12 182.6 8 176 8s-12.4 4-14.9 10.1l-29.4 74L55.6 68.9c-6.3-1.9-13.1 .2-17.2 5.3s-4.6 12.2-1.4 17.9l39.5 69.1L10.9 206.4c-5.4 3.7-8 10.3-6.5 16.7s6.7 11.2 13.1 12.2l78.7 12.2L90.6 327c-.5 6.5 3.1 12.7 9 15.5s12.9 1.8 17.8-2.6L176 286.1l58.6 53.9c4.1 3.8 9.9 5.1 15.2 3.6C223.6 310.8 208 269.2 208 224c0-60.8 28.3-115 72.4-150.2L220.3 92.1l-29.4-74z\"],\"square-caret-right\":[448,512,[\"caret-square-right\"],\"f152\",\"M448 96c0-35.3-28.7-64-64-64L64 32C28.7 32 0 60.7 0 96L0 416c0 35.3 28.7 64 64 64l320 0c35.3 0 64-28.7 64-64l0-320zM320 256c0 6.7-2.8 13-7.7 17.6l-112 104c-7 6.5-17.2 8.2-25.9 4.4s-14.4-12.5-14.4-22l0-208c0-9.5 5.7-18.2 14.4-22s18.9-2.1 25.9 4.4l112 104c4.9 4.5 7.7 10.9 7.7 17.6z\"],scissors:[512,512,[9984,9986,9988,\"cut\"],\"f0c4\",\"M256 192l-39.5-39.5c4.9-12.6 7.5-26.2 7.5-40.5C224 50.1 173.9 0 112 0S0 50.1 0 112s50.1 112 112 112c14.3 0 27.9-2.7 40.5-7.5L192 256l-39.5 39.5c-12.6-4.9-26.2-7.5-40.5-7.5C50.1 288 0 338.1 0 400s50.1 112 112 112s112-50.1 112-112c0-14.3-2.7-27.9-7.5-40.5L499.2 76.8c7.1-7.1 7.1-18.5 0-25.6c-28.3-28.3-74.1-28.3-102.4 0L256 192zm22.6 150.6L396.8 460.8c28.3 28.3 74.1 28.3 102.4 0c7.1-7.1 7.1-18.5 0-25.6L342.6 278.6l-64 64zM64 112a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm48 240a48 48 0 1 1 0 96 48 48 0 1 1 0-96z\"],\"sun-plant-wilt\":[640,512,[],\"e57a\",\"M160 0c-6.3 0-12 3.7-14.6 9.5L120.6 64.9 63.9 43.2c-5.9-2.3-12.6-.8-17 3.6s-5.9 11.1-3.6 17l21.7 56.7L9.5 145.4C3.7 148 0 153.7 0 160s3.7 12 9.5 14.6l55.4 24.8L43.2 256.1c-2.3 5.9-.8 12.6 3.6 17s11.1 5.9 17 3.6l56.7-21.7 24.8 55.4c2.6 5.8 8.3 9.5 14.6 9.5s12-3.7 14.6-9.5l24.8-55.4 56.7 21.7c5.9 2.3 12.6 .8 17-3.6s5.9-11.1 3.6-17l-21.7-56.7 55.4-24.8c5.8-2.6 9.5-8.3 9.5-14.6s-3.7-12-9.5-14.6l-55.4-24.8 21.7-56.7c2.3-5.9 .8-12.6-3.6-17s-11.1-5.9-17-3.6L199.4 64.9 174.6 9.5C172 3.7 166.3 0 160 0zm0 96a64 64 0 1 1 0 128 64 64 0 1 1 0-128zm32 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm312 16c0-17.7 14.3-32 32-32s32 14.3 32 32v53.4c-14.8 7.7-24 23.1-24 44.6c0 16.8 16 44 37.4 67.2c5.8 6.2 15.5 6.2 21.2 0C624 318 640 290.7 640 274c0-21.5-9.2-37-24-44.6V176c0-44.2-35.8-80-80-80s-80 35.8-80 80v22.7c-9.8-4.3-20.6-6.7-32-6.7c-44.2 0-80 35.8-80 80v21.4c-14.8 7.7-24 23.1-24 44.6c0 16.8 16 44 37.4 67.2c5.8 6.2 15.5 6.2 21.2 0C400 382 416 354.7 416 338c0-21.5-9.2-37-24-44.6V272c0-17.7 14.3-32 32-32s32 14.3 32 32v8V448H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H504V280v-8V176z\"],\"toilets-portable\":[576,512,[],\"e584\",\"M32 0H224c17.7 0 32 14.3 32 32V64H0V32C0 14.3 14.3 0 32 0zM0 96H24 232h24v24V488c0 13.3-10.7 24-24 24s-24-10.7-24-24v-8H48v8c0 13.3-10.7 24-24 24s-24-10.7-24-24V120 96zM192 224c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V240c0-8.8-7.2-16-16-16zM352 0H544c17.7 0 32 14.3 32 32V64H320V32c0-17.7 14.3-32 32-32zM320 96h24H552h24v24V488c0 13.3-10.7 24-24 24s-24-10.7-24-24v-8H368v8c0 13.3-10.7 24-24 24s-24-10.7-24-24V120 96zM512 224c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16s16-7.2 16-16V240c0-8.8-7.2-16-16-16z\"],\"hockey-puck\":[512,512,[],\"f453\",\"M256 256C114.6 256 0 213 0 160s114.6-96 256-96s256 43 256 96s-114.6 96-256 96zm192.3 1.8c24.7-9.3 46.9-21 63.7-35.6V352c0 53-114.6 96-256 96S0 405 0 352V222.3c16.8 14.6 39 26.3 63.7 35.6C114.5 276.9 182.5 288 256 288s141.5-11.1 192.3-30.2z\"],table:[512,512,[],\"f0ce\",\"M64 256V160H224v96H64zm0 64H224v96H64V320zm224 96V320H448v96H288zM448 256H288V160H448v96zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64z\"],\"magnifying-glass-arrow-right\":[512,512,[],\"e521\",\"M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zM241 119c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l31 31H120c-13.3 0-24 10.7-24 24s10.7 24 24 24H238.1l-31 31c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l72-72c9.4-9.4 9.4-24.6 0-33.9l-72-72z\"],\"tachograph-digital\":[640,512,[\"digital-tachograph\"],\"f566\",\"M64 64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H576c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H64zm32 64H320c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V160c0-17.7 14.3-32 32-32zM64 368c0-8.8 7.2-16 16-16H336c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16zm320 0c0-8.8 7.2-16 16-16H560c8.8 0 16 7.2 16 16s-7.2 16-16 16H400c-8.8 0-16-7.2-16-16zM80 288a16 16 0 1 1 0 32 16 16 0 1 1 0-32zm48 16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zm80-16a16 16 0 1 1 0 32 16 16 0 1 1 0-32zm48 16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zm80-16a16 16 0 1 1 0 32 16 16 0 1 1 0-32z\"],\"users-slash\":[640,512,[],\"e073\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L440.6 320H618.7c11.8 0 21.3-9.6 21.3-21.3C640 239.8 592.2 192 533.3 192H490.7c-15.9 0-31 3.5-44.6 9.7c1.3 7.2 1.9 14.7 1.9 22.3c0 30.2-10.5 58-28 79.9l-25.2-19.7C408.1 267.7 416 246.8 416 224c0-53-43-96-96-96c-31.1 0-58.7 14.8-76.3 37.7l-40.6-31.8c13-14.2 20.9-33.1 20.9-53.9c0-44.2-35.8-80-80-80C116.3 0 91.9 14.1 77.5 35.5L38.8 5.1zM106.7 192C47.8 192 0 239.8 0 298.7C0 310.4 9.6 320 21.3 320H234.7c.2 0 .4 0 .7 0c-20.6-18.2-35.2-42.8-40.8-70.8L121.8 192H106.7zM261.3 352C187.7 352 128 411.7 128 485.3c0 14.7 11.9 26.7 26.7 26.7H485.3c10.5 0 19.5-6 23.9-14.8L324.9 352H261.3zM512 160A80 80 0 1 0 512 0a80 80 0 1 0 0 160z\"],clover:[448,512,[],\"e139\",\"M216.6 49.9C205.1 38.5 189.5 32 173.3 32C139.4 32 112 59.4 112 93.3v4.9c0 12 3.3 23.7 9.4 34l18.8 31.3c1.1 1.8 1.2 3.1 1 4.2c-.2 1.2-.8 2.5-2 3.6s-2.4 1.8-3.6 2c-1 .2-2.4 .1-4.2-1l-31.3-18.8c-10.3-6.2-22-9.4-34-9.4H61.3C27.4 144 0 171.4 0 205.3c0 16.2 6.5 31.8 17.9 43.3l1.2 1.2c3.4 3.4 3.4 9 0 12.4l-1.2 1.2C6.5 274.9 0 290.5 0 306.7C0 340.6 27.4 368 61.3 368h4.9c12 0 23.7-3.3 34-9.4l31.3-18.8c1.8-1.1 3.1-1.2 4.2-1c1.2 .2 2.5 .8 3.6 2s1.8 2.4 2 3.6c.2 1 .1 2.4-1 4.2l-18.8 31.3c-6.2 10.3-9.4 22-9.4 34v4.9c0 33.8 27.4 61.3 61.3 61.3c16.2 0 31.8-6.5 43.3-17.9l1.2-1.2c3.4-3.4 9-3.4 12.4 0l1.2 1.2c11.5 11.5 27.1 17.9 43.3 17.9c33.8 0 61.3-27.4 61.3-61.3v-4.9c0-12-3.3-23.7-9.4-34l-18.8-31.3c-1.1-1.8-1.2-3.1-1-4.2c.2-1.2 .8-2.5 2-3.6s2.4-1.8 3.6-2c1-.2 2.4-.1 4.2 1l31.3 18.8c10.3 6.2 22 9.4 34 9.4h4.9c33.8 0 61.3-27.4 61.3-61.3c0-16.2-6.5-31.8-17.9-43.3l-1.2-1.2c-3.4-3.4-3.4-9 0-12.4l1.2-1.2c11.5-11.5 17.9-27.1 17.9-43.3c0-33.8-27.4-61.3-61.3-61.3h-4.9c-12 0-23.7 3.3-34 9.4l-31.3 18.8c-1.8 1.1-3.1 1.2-4.2 1c-1.2-.2-2.5-.8-3.6-2s-1.8-2.4-2-3.6c-.2-1-.1-2.4 1-4.2l18.8-31.3c6.2-10.3 9.4-22 9.4-34V93.3C336 59.4 308.6 32 274.7 32c-16.2 0-31.8 6.5-43.3 17.9l-1.2 1.2c-3.4 3.4-9 3.4-12.4 0l-1.2-1.2z\"],reply:[512,512,[61714,\"mail-reply\"],\"f3e5\",\"M205 34.8c11.5 5.1 19 16.6 19 29.2v64H336c97.2 0 176 78.8 176 176c0 113.3-81.5 163.9-100.2 174.1c-2.5 1.4-5.3 1.9-8.1 1.9c-10.9 0-19.7-8.9-19.7-19.7c0-7.5 4.3-14.4 9.8-19.5c9.4-8.8 22.2-26.4 22.2-56.7c0-53-43-96-96-96H224v64c0 12.6-7.4 24.1-19 29.2s-25 3-34.4-5.4l-160-144C3.9 225.7 0 217.1 0 208s3.9-17.7 10.6-23.8l160-144c9.4-8.5 22.9-10.6 34.4-5.4z\"],\"star-and-crescent\":[512,512,[9770],\"f699\",\"M0 256C0 114.6 114.6 0 256 0c33 0 64.6 6.3 93.6 17.7c7.4 2.9 11.5 10.7 9.8 18.4s-8.8 13-16.7 12.4c-4.8-.3-9.7-.5-14.6-.5c-114.9 0-208 93.1-208 208s93.1 208 208 208c4.9 0 9.8-.2 14.6-.5c7.9-.5 15 4.7 16.7 12.4s-2.4 15.5-9.8 18.4C320.6 505.7 289 512 256 512C114.6 512 0 397.4 0 256zM375.4 137.4c3.5-7.1 13.7-7.1 17.2 0l31.5 63.8c1.4 2.8 4.1 4.8 7.2 5.3l70.4 10.2c7.9 1.1 11 10.8 5.3 16.4l-50.9 49.6c-2.3 2.2-3.3 5.4-2.8 8.5l12 70.1c1.3 7.8-6.9 13.8-13.9 10.1l-63-33.1c-2.8-1.5-6.1-1.5-8.9 0l-63 33.1c-7 3.7-15.3-2.3-13.9-10.1l12-70.1c.5-3.1-.5-6.3-2.8-8.5L261 233.1c-5.7-5.6-2.6-15.2 5.3-16.4l70.4-10.2c3.1-.5 5.8-2.4 7.2-5.3l31.5-63.8z\"],\"house-fire\":[640,512,[],\"e50c\",\"M288 350.1l0 1.9H256c-17.7 0-32 14.3-32 32v64 24c0 22.1-17.9 40-40 40H160 128.1c-1.5 0-3-.1-4.5-.2c-1.2 .1-2.4 .2-3.6 .2H104c-22.1 0-40-17.9-40-40V360c0-.9 0-1.9 .1-2.8V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L447.3 128.1c-12.3-1-25 3-34.8 11.7c-35.4 31.6-65.6 67.7-87.3 102.8C304.3 276.5 288 314.9 288 350.1zM480 512c-88.4 0-160-71.6-160-160c0-76.7 62.5-144.7 107.2-179.4c5-3.9 10.9-5.8 16.8-5.8c7.9-.1 16 3.1 22 9.2l46 46 11.3-11.3c11.7-11.7 30.6-12.7 42.3-1C624.5 268 640 320.2 640 352c0 88.4-71.6 160-160 160zm64-111.8c0-36.5-37-73-54.8-88.4c-5.4-4.7-13.1-4.7-18.5 0C453 327.1 416 363.6 416 400.2c0 35.3 28.7 64 64 64s64-28.7 64-64z\"],\"square-minus\":[448,512,[61767,\"minus-square\"],\"f146\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm88 200H296c13.3 0 24 10.7 24 24s-10.7 24-24 24H152c-13.3 0-24-10.7-24-24s10.7-24 24-24z\"],helicopter:[640,512,[128641],\"f533\",\"M128 32c0-17.7 14.3-32 32-32H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H384v64h32c88.4 0 160 71.6 160 160v64c0 17.7-14.3 32-32 32H384 320c-20.1 0-39.1-9.5-51.2-25.6l-71.4-95.2c-3.5-4.7-8.3-8.3-13.7-10.5L47.2 198.1c-9.5-3.8-16.7-12-19.2-22L5 83.9C2.4 73.8 10.1 64 20.5 64H48c10.1 0 19.6 4.7 25.6 12.8L112 128H320V64H160c-17.7 0-32-14.3-32-32zM384 320H512V288c0-53-43-96-96-96H384V320zM630.6 425.4c12.5 12.5 12.5 32.8 0 45.3l-3.9 3.9c-24 24-56.6 37.5-90.5 37.5H256c-17.7 0-32-14.3-32-32s14.3-32 32-32H536.2c17 0 33.3-6.7 45.3-18.7l3.9-3.9c12.5-12.5 32.8-12.5 45.3 0z\"],compass:[512,512,[129517],\"f14e\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm50.7-186.9L162.4 380.6c-19.4 7.5-38.5-11.6-31-31l55.5-144.3c3.3-8.5 9.9-15.1 18.4-18.4l144.3-55.5c19.4-7.5 38.5 11.6 31 31L325.1 306.7c-3.2 8.5-9.9 15.1-18.4 18.4zM288 256a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z\"],\"square-caret-down\":[448,512,[\"caret-square-down\"],\"f150\",\"M384 480c35.3 0 64-28.7 64-64l0-320c0-35.3-28.7-64-64-64L64 32C28.7 32 0 60.7 0 96L0 416c0 35.3 28.7 64 64 64l320 0zM224 352c-6.7 0-13-2.8-17.6-7.7l-104-112c-6.5-7-8.2-17.2-4.4-25.9s12.5-14.4 22-14.4l208 0c9.5 0 18.2 5.7 22 14.4s2.1 18.9-4.4 25.9l-104 112c-4.5 4.9-10.9 7.7-17.6 7.7z\"],\"file-circle-question\":[576,512,[],\"e4ef\",\"M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v38.6C310.1 219.5 256 287.4 256 368c0 59.1 29.1 111.3 73.7 143.3c-3.2 .5-6.4 .7-9.7 .7H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zm48 96a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm0 240a24 24 0 1 0 0-48 24 24 0 1 0 0 48zM368 321.6V328c0 8.8 7.2 16 16 16s16-7.2 16-16v-6.4c0-5.3 4.3-9.6 9.6-9.6h40.5c7.7 0 13.9 6.2 13.9 13.9c0 5.2-2.9 9.9-7.4 12.3l-32 16.8c-5.3 2.8-8.6 8.2-8.6 14.2V384c0 8.8 7.2 16 16 16s16-7.2 16-16v-5.1l23.5-12.3c15.1-7.9 24.5-23.6 24.5-40.6c0-25.4-20.6-45.9-45.9-45.9H409.6c-23 0-41.6 18.6-41.6 41.6z\"],\"laptop-code\":[640,512,[],\"f5fc\",\"M64 96c0-35.3 28.7-64 64-64H512c35.3 0 64 28.7 64 64V352H512V96H128V352H64V96zM0 403.2C0 392.6 8.6 384 19.2 384H620.8c10.6 0 19.2 8.6 19.2 19.2c0 42.4-34.4 76.8-76.8 76.8H76.8C34.4 480 0 445.6 0 403.2zM281 209l-31 31 31 31c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-48-48c-9.4-9.4-9.4-24.6 0-33.9l48-48c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9zM393 175l48 48c9.4 9.4 9.4 24.6 0 33.9l-48 48c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l31-31-31-31c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0z\"],swatchbook:[512,512,[],\"f5c3\",\"M0 32C0 14.3 14.3 0 32 0H160c17.7 0 32 14.3 32 32V416c0 53-43 96-96 96s-96-43-96-96V32zM223.6 425.9c.3-3.3 .4-6.6 .4-9.9V154l75.4-75.4c12.5-12.5 32.8-12.5 45.3 0l90.5 90.5c12.5 12.5 12.5 32.8 0 45.3L223.6 425.9zM182.8 512l192-192H480c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H182.8zM128 64H64v64h64V64zM64 192v64h64V192H64zM96 440a24 24 0 1 0 0-48 24 24 0 1 0 0 48z\"],\"prescription-bottle\":[384,512,[],\"f485\",\"M0 32C0 14.3 14.3 0 32 0H352c17.7 0 32 14.3 32 32V64c0 17.7-14.3 32-32 32H32C14.3 96 0 81.7 0 64V32zm32 96H352V448c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V416H144c8.8 0 16-7.2 16-16s-7.2-16-16-16H32V320H144c8.8 0 16-7.2 16-16s-7.2-16-16-16H32V224H144c8.8 0 16-7.2 16-16s-7.2-16-16-16H32V128z\"],bars:[448,512,[\"navicon\"],\"f0c9\",\"M0 96C0 78.3 14.3 64 32 64H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 128 0 113.7 0 96zM0 256c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32zM448 416c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H416c17.7 0 32 14.3 32 32z\"],\"people-group\":[640,512,[],\"e533\",\"M72 88a56 56 0 1 1 112 0A56 56 0 1 1 72 88zM64 245.7C54 256.9 48 271.8 48 288s6 31.1 16 42.3V245.7zm144.4-49.3C178.7 222.7 160 261.2 160 304c0 34.3 12 65.8 32 90.5V416c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V389.2C26.2 371.2 0 332.7 0 288c0-61.9 50.1-112 112-112h32c24 0 46.2 7.5 64.4 20.3zM448 416V394.5c20-24.7 32-56.2 32-90.5c0-42.8-18.7-81.3-48.4-107.7C449.8 183.5 472 176 496 176h32c61.9 0 112 50.1 112 112c0 44.7-26.2 83.2-64 101.2V416c0 17.7-14.3 32-32 32H480c-17.7 0-32-14.3-32-32zm8-328a56 56 0 1 1 112 0A56 56 0 1 1 456 88zM576 245.7v84.7c10-11.3 16-26.1 16-42.3s-6-31.1-16-42.3zM320 32a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM240 304c0 16.2 6 31 16 42.3V261.7c-10 11.3-16 26.1-16 42.3zm144-42.3v84.7c10-11.3 16-26.1 16-42.3s-6-31.1-16-42.3zM448 304c0 44.7-26.2 83.2-64 101.2V448c0 17.7-14.3 32-32 32H288c-17.7 0-32-14.3-32-32V405.2c-37.8-18-64-56.5-64-101.2c0-61.9 50.1-112 112-112h32c61.9 0 112 50.1 112 112z\"],\"hourglass-end\":[384,512,[8987,\"hourglass-3\"],\"f253\",\"M32 0C14.3 0 0 14.3 0 32S14.3 64 32 64V75c0 42.4 16.9 83.1 46.9 113.1L146.7 256 78.9 323.9C48.9 353.9 32 394.6 32 437v11c-17.7 0-32 14.3-32 32s14.3 32 32 32H64 320h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V437c0-42.4-16.9-83.1-46.9-113.1L237.3 256l67.9-67.9c30-30 46.9-70.7 46.9-113.1V64c17.7 0 32-14.3 32-32s-14.3-32-32-32H320 64 32zM96 75V64H288V75c0 25.5-10.1 49.9-28.1 67.9L192 210.7l-67.9-67.9C106.1 124.9 96 100.4 96 75z\"],\"heart-crack\":[512,512,[128148,\"heart-broken\"],\"f7a9\",\"M119.4 44.1c23.3-3.9 46.8-1.9 68.6 5.3l49.8 77.5-75.4 75.4c-1.5 1.5-2.4 3.6-2.3 5.8s1 4.2 2.6 5.7l112 104c2.9 2.7 7.4 2.9 10.5 .3s3.8-7 1.7-10.4l-60.4-98.1 90.7-75.6c2.6-2.1 3.5-5.7 2.4-8.8L296.8 61.8c28.5-16.7 62.4-23.2 95.7-17.6C461.5 55.6 512 115.2 512 185.1v5.8c0 41.5-17.2 81.2-47.6 109.5L283.7 469.1c-7.5 7-17.4 10.9-27.7 10.9s-20.2-3.9-27.7-10.9L47.6 300.4C17.2 272.1 0 232.4 0 190.9v-5.8c0-69.9 50.5-129.5 119.4-141z\"],\"square-up-right\":[448,512,[8599,\"external-link-square-alt\"],\"f360\",\"M384 32c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H384zM320 313.4V176c0-8.8-7.2-16-16-16H166.6c-12.5 0-22.6 10.1-22.6 22.6c0 6 2.4 11.8 6.6 16L184 232l-66.3 66.3C114 302 112 306.9 112 312s2 10 5.7 13.7l36.7 36.7c3.6 3.6 8.5 5.7 13.7 5.7s10-2 13.7-5.7L248 296l33.4 33.4c4.2 4.2 10 6.6 16 6.6c12.5 0 22.6-10.1 22.6-22.6z\"],\"face-kiss-beam\":[512,512,[128537,\"kiss-beam\"],\"f597\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm48.7-198.3c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4c-2.7 1.5-5.7 3-8.7 4.3c3.1 1.3 6 2.7 8.7 4.3c6.6 3.7 12.5 8.2 16.7 13.4c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4C274.7 443.1 257.4 448 240 448c-3.6 0-6.8-2.5-7.7-6s.6-7.2 3.8-9l0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1c-.4-.2-.7-.4-.9-.5l-.2-.1 0 0 0 0 0 0c-2.5-1.4-4.1-4.1-4.1-7s1.6-5.6 4.1-7l0 0 0 0 0 0 0 0 0 0 .2-.1 .3-.2 .6-.4c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1l-.4-.3-.5-.3-.2-.1 0 0 0 0 0 0c-3.2-1.8-4.7-5.5-3.8-9s4.1-6 7.7-6c17.4 0 34.7 4.9 47.9 12.3c6.6 3.7 12.5 8.2 16.7 13.4zm-87.1-84.9l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z\"],film:[512,512,[127902],\"f008\",\"M0 96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM48 368v32c0 8.8 7.2 16 16 16H96c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H64c-8.8 0-16 7.2-16 16zm368-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H416zM48 240v32c0 8.8 7.2 16 16 16H96c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H64c-8.8 0-16 7.2-16 16zm368-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H416zM48 112v32c0 8.8 7.2 16 16 16H96c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H64c-8.8 0-16 7.2-16 16zM416 96c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H416zM160 128v64c0 17.7 14.3 32 32 32H320c17.7 0 32-14.3 32-32V128c0-17.7-14.3-32-32-32H192c-17.7 0-32 14.3-32 32zm32 160c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H320c17.7 0 32-14.3 32-32V320c0-17.7-14.3-32-32-32H192z\"],\"ruler-horizontal\":[640,512,[],\"f547\",\"M0 336c0 26.5 21.5 48 48 48l544 0c26.5 0 48-21.5 48-48l0-160c0-26.5-21.5-48-48-48l-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0c-26.5 0-48 21.5-48 48L0 336z\"],\"people-robbery\":[576,512,[],\"e536\",\"M488.2 59.1C478.1 99.6 441.7 128 400 128s-78.1-28.4-88.2-68.9L303 24.2C298.8 7.1 281.4-3.3 264.2 1S236.7 22.6 241 39.8l8.7 34.9c11 44 40.2 79.6 78.3 99.6V480c0 17.7 14.3 32 32 32s32-14.3 32-32V352h16V480c0 17.7 14.3 32 32 32s32-14.3 32-32V174.3c38.1-20 67.3-55.6 78.3-99.6L559 39.8c4.3-17.1-6.1-34.5-23.3-38.8S501.2 7.1 497 24.2l-8.7 34.9zM400 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM80 96A48 48 0 1 0 80 0a48 48 0 1 0 0 96zm-8 32c-35.3 0-64 28.7-64 64v96l0 .6V480c0 17.7 14.3 32 32 32s32-14.3 32-32V352H88V480c0 17.7 14.3 32 32 32s32-14.3 32-32V252.7l13 20.5c5.9 9.2 16.1 14.9 27 14.9h48c17.7 0 32-14.3 32-32s-14.3-32-32-32H209.6l-37.4-58.9C157.6 142 132.1 128 104.7 128H72z\"],lightbulb:[384,512,[128161],\"f0eb\",\"M272 384c9.6-31.9 29.5-59.1 49.2-86.2l0 0c5.2-7.1 10.4-14.2 15.4-21.4c19.8-28.5 31.4-63 31.4-100.3C368 78.8 289.2 0 192 0S16 78.8 16 176c0 37.3 11.6 71.9 31.4 100.3c5 7.2 10.2 14.3 15.4 21.4l0 0c19.8 27.1 39.7 54.4 49.2 86.2H272zM192 512c44.2 0 80-35.8 80-80V416H112v16c0 44.2 35.8 80 80 80zM112 176c0 8.8-7.2 16-16 16s-16-7.2-16-16c0-61.9 50.1-112 112-112c8.8 0 16 7.2 16 16s-7.2 16-16 16c-44.2 0-80 35.8-80 80z\"],\"caret-left\":[256,512,[],\"f0d9\",\"M9.4 278.6c-12.5-12.5-12.5-32.8 0-45.3l128-128c9.2-9.2 22.9-11.9 34.9-6.9s19.8 16.6 19.8 29.6l0 256c0 12.9-7.8 24.6-19.8 29.6s-25.7 2.2-34.9-6.9l-128-128z\"],\"circle-exclamation\":[512,512,[\"exclamation-circle\"],\"f06a\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm0-384c13.3 0 24 10.7 24 24V264c0 13.3-10.7 24-24 24s-24-10.7-24-24V152c0-13.3 10.7-24 24-24zM224 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],\"school-circle-xmark\":[640,512,[],\"e56d\",\"M337.8 5.4C327-1.8 313-1.8 302.2 5.4L166.3 96H48C21.5 96 0 117.5 0 144V464c0 26.5 21.5 48 48 48H320v0H256V416c0-35.3 28.7-64 64-64l.3 0h.5c3.4-37.7 18.7-72.1 42.2-99.1C350.2 260 335.6 264 320 264c-48.6 0-88-39.4-88-88s39.4-88 88-88s88 39.4 88 88c0 18.3-5.6 35.3-15.1 49.4c29-21 64.6-33.4 103.1-33.4c59.5 0 112.1 29.6 144 74.8V144c0-26.5-21.5-48-48-48H473.7L337.8 5.4zM96 192h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V208c0-8.8 7.2-16 16-16zm0 128h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zM320 128c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H336V144c0-8.8-7.2-16-16-16zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm22.6-144l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L496 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L473.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L496 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6L518.6 368z\"],\"arrow-right-from-bracket\":[512,512,[\"sign-out\"],\"f08b\",\"M502.6 278.6c12.5-12.5 12.5-32.8 0-45.3l-128-128c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L402.7 224 192 224c-17.7 0-32 14.3-32 32s14.3 32 32 32l210.7 0-73.4 73.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l128-128zM160 96c17.7 0 32-14.3 32-32s-14.3-32-32-32L96 32C43 32 0 75 0 128L0 384c0 53 43 96 96 96l64 0c17.7 0 32-14.3 32-32s-14.3-32-32-32l-64 0c-17.7 0-32-14.3-32-32l0-256c0-17.7 14.3-32 32-32l64 0z\"],\"circle-chevron-down\":[512,512,[\"chevron-circle-down\"],\"f13a\",\"M256 0a256 256 0 1 0 0 512A256 256 0 1 0 256 0zM135 241c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l87 87 87-87c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9L273 345c-9.4 9.4-24.6 9.4-33.9 0L135 241z\"],\"unlock-keyhole\":[448,512,[\"unlock-alt\"],\"f13e\",\"M224 64c-44.2 0-80 35.8-80 80v48H384c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V256c0-35.3 28.7-64 64-64H80V144C80 64.5 144.5 0 224 0c57.5 0 107 33.7 130.1 82.3c7.6 16 .8 35.1-15.2 42.6s-35.1 .8-42.6-15.2C283.4 82.6 255.9 64 224 64zm32 320c17.7 0 32-14.3 32-32s-14.3-32-32-32H192c-17.7 0-32 14.3-32 32s14.3 32 32 32h64z\"],\"cloud-showers-heavy\":[512,512,[],\"f740\",\"M96 320c-53 0-96-43-96-96c0-42.5 27.6-78.6 65.9-91.2C64.7 126.1 64 119.1 64 112C64 50.1 114.1 0 176 0c43.1 0 80.5 24.3 99.2 60c14.7-17.1 36.5-28 60.8-28c44.2 0 80 35.8 80 80c0 5.5-.6 10.8-1.6 16c.5 0 1.1 0 1.6 0c53 0 96 43 96 96s-43 96-96 96H96zM81.5 353.9c12.2 5.2 17.8 19.3 12.6 31.5l-48 112c-5.2 12.2-19.3 17.8-31.5 12.6S-3.3 490.7 1.9 478.5l48-112c5.2-12.2 19.3-17.8 31.5-12.6zm120 0c12.2 5.2 17.8 19.3 12.6 31.5l-48 112c-5.2 12.2-19.3 17.8-31.5 12.6s-17.8-19.3-12.6-31.5l48-112c5.2-12.2 19.3-17.8 31.5-12.6zm244.6 31.5l-48 112c-5.2 12.2-19.3 17.8-31.5 12.6s-17.8-19.3-12.6-31.5l48-112c5.2-12.2 19.3-17.8 31.5-12.6s17.8 19.3 12.6 31.5zM313.5 353.9c12.2 5.2 17.8 19.3 12.6 31.5l-48 112c-5.2 12.2-19.3 17.8-31.5 12.6s-17.8-19.3-12.6-31.5l48-112c5.2-12.2 19.3-17.8 31.5-12.6z\"],\"headphones-simple\":[512,512,[\"headphones-alt\"],\"f58f\",\"M256 80C141.1 80 48 173.1 48 288V392c0 13.3-10.7 24-24 24s-24-10.7-24-24V288C0 146.6 114.6 32 256 32s256 114.6 256 256V392c0 13.3-10.7 24-24 24s-24-10.7-24-24V288c0-114.9-93.1-208-208-208zM80 352c0-35.3 28.7-64 64-64h16c17.7 0 32 14.3 32 32V448c0 17.7-14.3 32-32 32H144c-35.3 0-64-28.7-64-64V352zm288-64c35.3 0 64 28.7 64 64v64c0 35.3-28.7 64-64 64H352c-17.7 0-32-14.3-32-32V320c0-17.7 14.3-32 32-32h16z\"],sitemap:[576,512,[],\"f0e8\",\"M208 80c0-26.5 21.5-48 48-48h64c26.5 0 48 21.5 48 48v64c0 26.5-21.5 48-48 48h-8v40H464c30.9 0 56 25.1 56 56v32h8c26.5 0 48 21.5 48 48v64c0 26.5-21.5 48-48 48H464c-26.5 0-48-21.5-48-48V368c0-26.5 21.5-48 48-48h8V288c0-4.4-3.6-8-8-8H312v40h8c26.5 0 48 21.5 48 48v64c0 26.5-21.5 48-48 48H256c-26.5 0-48-21.5-48-48V368c0-26.5 21.5-48 48-48h8V280H112c-4.4 0-8 3.6-8 8v32h8c26.5 0 48 21.5 48 48v64c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V368c0-26.5 21.5-48 48-48h8V288c0-30.9 25.1-56 56-56H264V192h-8c-26.5 0-48-21.5-48-48V80z\"],\"circle-dollar-to-slot\":[512,512,[\"donate\"],\"f4b9\",\"M326.7 403.7c-22.1 8-45.9 12.3-70.7 12.3s-48.7-4.4-70.7-12.3c-.3-.1-.5-.2-.8-.3c-30-11-56.8-28.7-78.6-51.4C70 314.6 48 263.9 48 208C48 93.1 141.1 0 256 0S464 93.1 464 208c0 55.9-22 106.6-57.9 144c-1 1-2 2.1-3 3.1c-21.4 21.4-47.4 38.1-76.3 48.6zM256 84c-11 0-20 9-20 20v14c-7.6 1.7-15.2 4.4-22.2 8.5c-13.9 8.3-25.9 22.8-25.8 43.9c.1 20.3 12 33.1 24.7 40.7c11 6.6 24.7 10.8 35.6 14l1.7 .5c12.6 3.8 21.8 6.8 28 10.7c5.1 3.2 5.8 5.4 5.9 8.2c.1 5-1.8 8-5.9 10.5c-5 3.1-12.9 5-21.4 4.7c-11.1-.4-21.5-3.9-35.1-8.5c-2.3-.8-4.7-1.6-7.2-2.4c-10.5-3.5-21.8 2.2-25.3 12.6s2.2 21.8 12.6 25.3c1.9 .6 4 1.3 6.1 2.1l0 0 0 0c8.3 2.9 17.9 6.2 28.2 8.4V312c0 11 9 20 20 20s20-9 20-20V298.2c8-1.7 16-4.5 23.2-9c14.3-8.9 25.1-24.1 24.8-45c-.3-20.3-11.7-33.4-24.6-41.6c-11.5-7.2-25.9-11.6-37.1-15l-.7-.2c-12.8-3.9-21.9-6.7-28.3-10.5c-5.2-3.1-5.3-4.9-5.3-6.7c0-3.7 1.4-6.5 6.2-9.3c5.4-3.2 13.6-5.1 21.5-5c9.6 .1 20.2 2.2 31.2 5.2c10.7 2.8 21.6-3.5 24.5-14.2s-3.5-21.6-14.2-24.5c-6.5-1.7-13.7-3.4-21.1-4.7V104c0-11-9-20-20-20zM48 352H64c19.5 25.9 44 47.7 72.2 64H64v32H256 448V416H375.8c28.2-16.3 52.8-38.1 72.2-64h16c26.5 0 48 21.5 48 48v64c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V400c0-26.5 21.5-48 48-48z\"],memory:[576,512,[],\"f538\",\"M64 64C28.7 64 0 92.7 0 128v7.4c0 6.8 4.4 12.6 10.1 16.3C23.3 160.3 32 175.1 32 192s-8.7 31.7-21.9 40.3C4.4 236 0 241.8 0 248.6V320H576V248.6c0-6.8-4.4-12.6-10.1-16.3C552.7 223.7 544 208.9 544 192s8.7-31.7 21.9-40.3c5.7-3.7 10.1-9.5 10.1-16.3V128c0-35.3-28.7-64-64-64H64zM576 352H0v64c0 17.7 14.3 32 32 32H80V416c0-8.8 7.2-16 16-16s16 7.2 16 16v32h96V416c0-8.8 7.2-16 16-16s16 7.2 16 16v32h96V416c0-8.8 7.2-16 16-16s16 7.2 16 16v32h96V416c0-8.8 7.2-16 16-16s16 7.2 16 16v32h48c17.7 0 32-14.3 32-32V352zM192 160v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V160c0-17.7 14.3-32 32-32s32 14.3 32 32zm128 0v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V160c0-17.7 14.3-32 32-32s32 14.3 32 32zm128 0v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V160c0-17.7 14.3-32 32-32s32 14.3 32 32z\"],\"road-spikes\":[640,512,[],\"e568\",\"M64 116.8c0-15.8 20.5-22 29.3-8.9L192 256V116.8c0-15.8 20.5-22 29.3-8.9L320 256V116.8c0-15.8 20.5-22 29.3-8.9L448 256V116.8c0-15.8 20.5-22 29.3-8.9L606.8 302.2c14.2 21.3-1.1 49.7-26.6 49.7H512 448 384 320 256 192 64V116.8zM32 384H608c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],\"fire-burner\":[640,512,[],\"e4f1\",\"M345.7 48.3L358 34.5c5.4-6.1 13.3-8.8 20.9-8.9c7.2 0 14.3 2.6 19.9 7.8c19.7 18.3 39.8 43.2 55 70.6C469 131.2 480 162.2 480 192.2C480 280.8 408.7 352 320 352c-89.6 0-160-71.3-160-159.8c0-37.3 16-73.4 36.8-104.5c20.9-31.3 47.5-59 70.9-80.2C273.4 2.3 280.7-.2 288 0c14.1 .3 23.8 11.4 32.7 21.6c0 0 0 0 0 0c2 2.3 4 4.6 6 6.7l19 19.9zM384 240.2c0-36.5-37-73-54.8-88.4c-5.4-4.7-13.1-4.7-18.5 0C293 167.1 256 203.6 256 240.2c0 35.3 28.7 64 64 64s64-28.7 64-64zM32 288c0-17.7 14.3-32 32-32l32 0c17.7 0 32 14.3 32 32s-14.3 32-32 32l0 64 448 0 0-64c-17.7 0-32-14.3-32-32s14.3-32 32-32l32 0c17.7 0 32 14.3 32 32l0 96c17.7 0 32 14.3 32 32l0 64c0 17.7-14.3 32-32 32L32 512c-17.7 0-32-14.3-32-32l0-64c0-17.7 14.3-32 32-32l0-96zM320 480a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm160-32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM192 480a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],flag:[448,512,[127988,61725],\"f024\",\"M64 32C64 14.3 49.7 0 32 0S0 14.3 0 32V64 368 480c0 17.7 14.3 32 32 32s32-14.3 32-32V352l64.3-16.1c41.1-10.3 84.6-5.5 122.5 13.4c44.2 22.1 95.5 24.8 141.7 7.4l34.7-13c12.5-4.7 20.8-16.6 20.8-30V66.1c0-23-24.2-38-44.8-27.7l-9.6 4.8c-46.3 23.2-100.8 23.2-147.1 0c-35.1-17.6-75.4-22-113.5-12.5L64 48V32z\"],hanukiah:[640,512,[128334],\"f6e6\",\"M314.2 3.3C309.1 12.1 296 36.6 296 56c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7C324.6 1.2 322.4 0 320 0s-4.6 1.2-5.8 3.3zm-288 48C21.1 60.1 8 84.6 8 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7C36.6 49.2 34.4 48 32 48s-4.6 1.2-5.8 3.3zM88 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3C101.1 60.1 88 84.6 88 104zm82.2-52.7C165.1 60.1 152 84.6 152 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3zM216 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3C229.1 60.1 216 84.6 216 104zM394.2 51.3C389.1 60.1 376 84.6 376 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3zM440 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3C453.1 60.1 440 84.6 440 104zm82.2-52.7C517.1 60.1 504 84.6 504 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3zM584 104c0 13.3 10.7 24 24 24s24-10.7 24-24c0-19.4-13.1-43.9-18.2-52.7c-1.2-2.1-3.4-3.3-5.8-3.3s-4.6 1.2-5.8 3.3C597.1 60.1 584 84.6 584 104zM112 160c-8.8 0-16 7.2-16 16v96 16h32V272 176c0-8.8-7.2-16-16-16zm64 0c-8.8 0-16 7.2-16 16v96 16h32V272 176c0-8.8-7.2-16-16-16zm64 0c-8.8 0-16 7.2-16 16v96 16h32V272 176c0-8.8-7.2-16-16-16zm160 0c-8.8 0-16 7.2-16 16v96 16h32V272 176c0-8.8-7.2-16-16-16zm64 0c-8.8 0-16 7.2-16 16v96 16h32V272 176c0-8.8-7.2-16-16-16zm64 0c-8.8 0-16 7.2-16 16v96 16h32V272 176c0-8.8-7.2-16-16-16zM352 144c0-17.7-14.3-32-32-32s-32 14.3-32 32V320H96c-17.7 0-32-14.3-32-32V192c0-17.7-14.3-32-32-32s-32 14.3-32 32v96c0 53 43 96 96 96H288v64H160c-17.7 0-32 14.3-32 32s14.3 32 32 32H320 480c17.7 0 32-14.3 32-32s-14.3-32-32-32H352V384H544c53 0 96-43 96-96V192c0-17.7-14.3-32-32-32s-32 14.3-32 32v96c0 17.7-14.3 32-32 32H352V144z\"],feather:[512,512,[129718],\"f52d\",\"M278.5 215.6L23 471c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l57-57h68c49.7 0 97.9-14.4 139-41c11.1-7.2 5.5-23-7.8-23c-5.1 0-9.2-4.1-9.2-9.2c0-4.1 2.7-7.6 6.5-8.8l81-24.3c2.5-.8 4.8-2.1 6.7-4l22.4-22.4c10.1-10.1 2.9-27.3-11.3-27.3l-32.2 0c-5.1 0-9.2-4.1-9.2-9.2c0-4.1 2.7-7.6 6.5-8.8l112-33.6c4-1.2 7.4-3.9 9.3-7.7C506.4 207.6 512 184.1 512 160c0-41-16.3-80.3-45.3-109.3l-5.5-5.5C432.3 16.3 393 0 352 0s-80.3 16.3-109.3 45.3L139 149C91 197 64 262.1 64 330v55.3L253.6 195.8c6.2-6.2 16.4-6.2 22.6 0c5.4 5.4 6.1 13.6 2.2 19.8z\"],\"volume-low\":[448,512,[128264,\"volume-down\"],\"f027\",\"M301.1 34.8C312.6 40 320 51.4 320 64V448c0 12.6-7.4 24-18.9 29.2s-25 3.1-34.4-5.3L131.8 352H64c-35.3 0-64-28.7-64-64V224c0-35.3 28.7-64 64-64h67.8L266.7 40.1c9.4-8.4 22.9-10.4 34.4-5.3zM412.6 181.5C434.1 199.1 448 225.9 448 256s-13.9 56.9-35.4 74.5c-10.3 8.4-25.4 6.8-33.8-3.5s-6.8-25.4 3.5-33.8C393.1 284.4 400 271 400 256s-6.9-28.4-17.7-37.3c-10.3-8.4-11.8-23.5-3.5-33.8s23.5-11.8 33.8-3.5z\"],\"comment-slash\":[640,512,[],\"f4b3\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L512.9 376.7C552.2 340.2 576 292.3 576 240C576 125.1 461.4 32 320 32c-67.7 0-129.3 21.4-175.1 56.3L38.8 5.1zm385.2 425L82.9 161.3C70.7 185.6 64 212.2 64 240c0 45.1 17.7 86.8 47.7 120.9c-1.9 24.5-11.4 46.3-21.4 62.9c-5.5 9.2-11.1 16.6-15.2 21.6c-2.1 2.5-3.7 4.4-4.9 5.7c-.6 .6-1 1.1-1.3 1.4l-.3 .3 0 0 0 0 0 0 0 0c-4.6 4.6-5.9 11.4-3.4 17.4c2.5 6 8.3 9.9 14.8 9.9c28.7 0 57.6-8.9 81.6-19.3c22.9-10 42.4-21.9 54.3-30.6c31.8 11.5 67 17.9 104.1 17.9c37 0 72.3-6.4 104.1-17.9z\"],\"cloud-sun-rain\":[640,512,[127782],\"f743\",\"M294.2 1.2c5.1 2.1 8.7 6.7 9.6 12.1l10.4 62.4c-23.3 10.8-42.9 28.4-56 50.3c-14.6-9-31.8-14.1-50.2-14.1c-53 0-96 43-96 96c0 35.5 19.3 66.6 48 83.2c.8 31.8 13.2 60.7 33.1 82.7l-56 39.2c-4.5 3.2-10.3 3.8-15.4 1.6s-8.7-6.7-9.6-12.1L98.1 317.9 13.4 303.8c-5.4-.9-10-4.5-12.1-9.6s-1.5-10.9 1.6-15.4L52.5 208 2.9 137.2c-3.2-4.5-3.8-10.3-1.6-15.4s6.7-8.7 12.1-9.6L98.1 98.1l14.1-84.7c.9-5.4 4.5-10 9.6-12.1s10.9-1.5 15.4 1.6L208 52.5 278.8 2.9c4.5-3.2 10.3-3.8 15.4-1.6zM208 144c13.8 0 26.7 4.4 37.1 11.9c-1.2 4.1-2.2 8.3-3 12.6c-37.9 14.6-67.2 46.6-77.8 86.4C151.8 243.1 144 226.5 144 208c0-35.3 28.7-64 64-64zm69.4 276c11 7.4 14 22.3 6.7 33.3l-32 48c-7.4 11-22.3 14-33.3 6.7s-14-22.3-6.7-33.3l32-48c7.4-11 22.3-14 33.3-6.7zm96 0c11 7.4 14 22.3 6.7 33.3l-32 48c-7.4 11-22.3 14-33.3 6.7s-14-22.3-6.7-33.3l32-48c7.4-11 22.3-14 33.3-6.7zm96 0c11 7.4 14 22.3 6.7 33.3l-32 48c-7.4 11-22.3 14-33.3 6.7s-14-22.3-6.7-33.3l32-48c7.4-11 22.3-14 33.3-6.7zm96 0c11 7.4 14 22.3 6.7 33.3l-32 48c-7.4 11-22.3 14-33.3 6.7s-14-22.3-6.7-33.3l32-48c7.4-11 22.3-14 33.3-6.7zm74.5-116.1c0 44.2-35.8 80-80 80H288c-53 0-96-43-96-96c0-47.6 34.6-87 80-94.6l0-1.3c0-53 43-96 96-96c34.9 0 65.4 18.6 82.2 46.4c13-9.1 28.8-14.4 45.8-14.4c44.2 0 80 35.8 80 80c0 5.9-.6 11.7-1.9 17.2c37.4 6.7 65.8 39.4 65.8 78.7z\"],compress:[448,512,[],\"f066\",\"M160 64c0-17.7-14.3-32-32-32s-32 14.3-32 32v64H32c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c17.7 0 32-14.3 32-32V64zM32 320c-17.7 0-32 14.3-32 32s14.3 32 32 32H96v64c0 17.7 14.3 32 32 32s32-14.3 32-32V352c0-17.7-14.3-32-32-32H32zM352 64c0-17.7-14.3-32-32-32s-32 14.3-32 32v96c0 17.7 14.3 32 32 32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H352V64zM320 320c-17.7 0-32 14.3-32 32v96c0 17.7 14.3 32 32 32s32-14.3 32-32V384h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H320z\"],\"wheat-awn\":[512,512,[\"wheat-alt\"],\"e2cd\",\"M505 41c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L383 95c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l88-88zM305.5 27.3c-6.2-6.2-16.4-6.2-22.6 0L271.5 38.6c-37.5 37.5-37.5 98.3 0 135.8l10.4 10.4-30.5 30.5c-3.4-27.3-15.5-53.8-36.5-74.8l-11.3-11.3c-6.2-6.2-16.4-6.2-22.6 0l-11.3 11.3c-37.5 37.5-37.5 98.3 0 135.8l10.4 10.4-30.5 30.5c-3.4-27.3-15.5-53.8-36.5-74.8L101.8 231c-6.2-6.2-16.4-6.2-22.6 0L67.9 242.3c-37.5 37.5-37.5 98.3 0 135.8l10.4 10.4L9.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l68.9-68.9 12.2 12.2c37.5 37.5 98.3 37.5 135.8 0l11.3-11.3c6.2-6.2 6.2-16.4 0-22.6l-11.3-11.3c-21.8-21.8-49.6-34.1-78.1-36.9l31.9-31.9 12.2 12.2c37.5 37.5 98.3 37.5 135.8 0l11.3-11.3c6.2-6.2 6.2-16.4 0-22.6l-11.3-11.3c-21.8-21.8-49.6-34.1-78.1-36.9l31.9-31.9 12.2 12.2c37.5 37.5 98.3 37.5 135.8 0L486.5 231c6.2-6.2 6.2-16.4 0-22.6L475.2 197c-5.2-5.2-10.6-9.8-16.4-13.9L505 137c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-59.4 59.4c-20.6-4.4-42-3.7-62.3 2.1c6.1-21.3 6.6-43.8 1.4-65.3L409 41c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L329.1 52.9c-3.7-5-7.8-9.8-12.4-14.3L305.5 27.3z\"],ankh:[320,512,[9765],\"f644\",\"M96 128c0-35.3 28.7-64 64-64s64 28.7 64 64c0 41.6-20.7 76.6-46.6 104.1c-5.9 6.2-11.8 11.8-17.4 16.7c-5.6-4.9-11.5-10.5-17.4-16.7C116.7 204.6 96 169.6 96 128zM160 0C89.3 0 32 57.3 32 128c0 52.4 21.5 95.5 46.8 128H32c-17.7 0-32 14.3-32 32s14.3 32 32 32h96V480c0 17.7 14.3 32 32 32s32-14.3 32-32V320h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H241.2c25.4-32.5 46.8-75.6 46.8-128C288 57.3 230.7 0 160 0z\"],\"hands-holding-child\":[640,512,[],\"e4fa\",\"M320 0a40 40 0 1 1 0 80 40 40 0 1 1 0-80zm44.7 164.3L375.8 253c1.6 13.2-7.7 25.1-20.8 26.8s-25.1-7.7-26.8-20.8l-4.4-35h-7.6l-4.4 35c-1.6 13.2-13.6 22.5-26.8 20.8s-22.5-13.6-20.8-26.8l11.1-88.8L255.5 181c-10.1 8.6-25.3 7.3-33.8-2.8s-7.3-25.3 2.8-33.8l27.9-23.6C271.3 104.8 295.3 96 320 96s48.7 8.8 67.6 24.7l27.9 23.6c10.1 8.6 11.4 23.7 2.8 33.8s-23.7 11.4-33.8 2.8l-19.8-16.7zM40 64c22.1 0 40 17.9 40 40v40 80 40.2c0 17 6.7 33.3 18.7 45.3l51.1 51.1c8.3 8.3 21.3 9.6 31 3.1c12.9-8.6 14.7-26.9 3.7-37.8l-15.2-15.2-32-32c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0l32 32 15.2 15.2 0 0 25.3 25.3c21 21 32.8 49.5 32.8 79.2V464c0 26.5-21.5 48-48 48H173.3c-17 0-33.3-6.7-45.3-18.7L28.1 393.4C10.1 375.4 0 351 0 325.5V224 160 104C0 81.9 17.9 64 40 64zm560 0c22.1 0 40 17.9 40 40v56 64V325.5c0 25.5-10.1 49.9-28.1 67.9L512 493.3c-12 12-28.3 18.7-45.3 18.7H400c-26.5 0-48-21.5-48-48V385.1c0-29.7 11.8-58.2 32.8-79.2l25.3-25.3 0 0 15.2-15.2 32-32c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3l-32 32-15.2 15.2c-11 11-9.2 29.2 3.7 37.8c9.7 6.5 22.7 5.2 31-3.1l51.1-51.1c12-12 18.7-28.3 18.7-45.3V224 144 104c0-22.1 17.9-40 40-40z\"],asterisk:[384,512,[10033,61545],\"2a\",\"M192 32c17.7 0 32 14.3 32 32V199.5l111.5-66.9c15.2-9.1 34.8-4.2 43.9 11s4.2 34.8-11 43.9L254.2 256l114.3 68.6c15.2 9.1 20.1 28.7 11 43.9s-28.7 20.1-43.9 11L224 312.5V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V312.5L48.5 379.4c-15.2 9.1-34.8 4.2-43.9-11s-4.2-34.8 11-43.9L129.8 256 15.5 187.4c-15.2-9.1-20.1-28.7-11-43.9s28.7-20.1 43.9-11L160 199.5V64c0-17.7 14.3-32 32-32z\"],\"square-check\":[448,512,[9745,9989,61510,\"check-square\"],\"f14a\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM337 209L209 337c-9.4 9.4-24.6 9.4-33.9 0l-64-64c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47L303 175c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z\"],\"peseta-sign\":[384,512,[],\"e221\",\"M64 32C46.3 32 32 46.3 32 64v96c-17.7 0-32 14.3-32 32s14.3 32 32 32l0 96V448c0 17.7 14.3 32 32 32s32-14.3 32-32V352h96c77.4 0 142-55 156.8-128H352c17.7 0 32-14.3 32-32s-14.3-32-32-32h-3.2C334 87 269.4 32 192 32H64zM282.5 160H96V96h96c41.8 0 77.4 26.7 90.5 64zM96 224H282.5c-13.2 37.3-48.7 64-90.5 64H96V224z\"],heading:[448,512,[\"header\"],\"f1dc\",\"M0 64C0 46.3 14.3 32 32 32H80h48c17.7 0 32 14.3 32 32s-14.3 32-32 32H112V208H336V96H320c-17.7 0-32-14.3-32-32s14.3-32 32-32h48 48c17.7 0 32 14.3 32 32s-14.3 32-32 32H400V240 416h16c17.7 0 32 14.3 32 32s-14.3 32-32 32H368 320c-17.7 0-32-14.3-32-32s14.3-32 32-32h16V272H112V416h16c17.7 0 32 14.3 32 32s-14.3 32-32 32H80 32c-17.7 0-32-14.3-32-32s14.3-32 32-32H48V240 96H32C14.3 96 0 81.7 0 64z\"],ghost:[384,512,[128123],\"f6e2\",\"M40.1 467.1l-11.2 9c-3.2 2.5-7.1 3.9-11.1 3.9C8 480 0 472 0 462.2V192C0 86 86 0 192 0S384 86 384 192V462.2c0 9.8-8 17.8-17.8 17.8c-4 0-7.9-1.4-11.1-3.9l-11.2-9c-13.4-10.7-32.8-9-44.1 3.9L269.3 506c-3.3 3.8-8.2 6-13.3 6s-9.9-2.2-13.3-6l-26.6-30.5c-12.7-14.6-35.4-14.6-48.2 0L141.3 506c-3.3 3.8-8.2 6-13.3 6s-9.9-2.2-13.3-6L84.2 471c-11.3-12.9-30.7-14.6-44.1-3.9zM160 192a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm96 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],list:[512,512,[\"list-squares\"],\"f03a\",\"M40 48C26.7 48 16 58.7 16 72v48c0 13.3 10.7 24 24 24H88c13.3 0 24-10.7 24-24V72c0-13.3-10.7-24-24-24H40zM192 64c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H192zm0 160c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H192zm0 160c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H192zM16 232v48c0 13.3 10.7 24 24 24H88c13.3 0 24-10.7 24-24V232c0-13.3-10.7-24-24-24H40c-13.3 0-24 10.7-24 24zM40 368c-13.3 0-24 10.7-24 24v48c0 13.3 10.7 24 24 24H88c13.3 0 24-10.7 24-24V392c0-13.3-10.7-24-24-24H40z\"],\"square-phone-flip\":[448,512,[\"phone-square-alt\"],\"f87b\",\"M384 32c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H384zm-90.7 96.7c-9.7-2.6-19.9 2.3-23.7 11.6l-20 48c-3.4 8.2-1 17.6 5.8 23.2L280 231.7c-16.6 35.2-45.1 63.7-80.3 80.3l-20.2-24.7c-5.6-6.8-15-9.2-23.2-5.8l-48 20c-9.3 3.9-14.2 14-11.6 23.7l12 44C111.1 378 119 384 128 384c123.7 0 224-100.3 224-224c0-9-6-16.9-14.7-19.3l-44-12z\"],\"cart-plus\":[576,512,[],\"f217\",\"M0 24C0 10.7 10.7 0 24 0H69.5c22 0 41.5 12.8 50.6 32h411c26.3 0 45.5 25 38.6 50.4l-41 152.3c-8.5 31.4-37 53.3-69.5 53.3H170.7l5.4 28.5c2.2 11.3 12.1 19.5 23.6 19.5H488c13.3 0 24 10.7 24 24s-10.7 24-24 24H199.7c-34.6 0-64.3-24.6-70.7-58.5L77.4 54.5c-.7-3.8-4-6.5-7.9-6.5H24C10.7 48 0 37.3 0 24zM128 464a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm336-48a48 48 0 1 1 0 96 48 48 0 1 1 0-96zM252 160c0 11 9 20 20 20h44v44c0 11 9 20 20 20s20-9 20-20V180h44c11 0 20-9 20-20s-9-20-20-20H356V96c0-11-9-20-20-20s-20 9-20 20v44H272c-11 0-20 9-20 20z\"],gamepad:[640,512,[],\"f11b\",\"M192 64C86 64 0 150 0 256S86 448 192 448H448c106 0 192-86 192-192s-86-192-192-192H192zM496 168a40 40 0 1 1 0 80 40 40 0 1 1 0-80zM392 304a40 40 0 1 1 80 0 40 40 0 1 1 -80 0zM168 200c0-13.3 10.7-24 24-24s24 10.7 24 24v32h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H216v32c0 13.3-10.7 24-24 24s-24-10.7-24-24V280H136c-13.3 0-24-10.7-24-24s10.7-24 24-24h32V200z\"],\"circle-dot\":[512,512,[128280,\"dot-circle\"],\"f192\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm0-352a96 96 0 1 1 0 192 96 96 0 1 1 0-192z\"],\"face-dizzy\":[512,512,[\"dizzy\"],\"f567\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm0-224a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM100.7 132.7c6.2-6.2 16.4-6.2 22.6 0L160 169.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6L182.6 192l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L160 214.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L137.4 192l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6zm192 0c6.2-6.2 16.4-6.2 22.6 0L352 169.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6L374.6 192l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L352 214.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L329.4 192l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6z\"],egg:[384,512,[129370],\"f7fb\",\"M192 496C86 496 0 394 0 288C0 176 64 16 192 16s192 160 192 272c0 106-86 208-192 208zM154.8 134c6.5-6 7-16.1 1-22.6s-16.1-7-22.6-1c-23.9 21.8-41.1 52.7-52.3 84.2C69.7 226.1 64 259.7 64 288c0 8.8 7.2 16 16 16s16-7.2 16-16c0-24.5 5-54.4 15.1-82.8c10.1-28.5 25-54.1 43.7-71.2z\"],\"house-medical-circle-xmark\":[640,512,[],\"e513\",\"M320 368c0 59.5 29.5 112.1 74.8 144H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L522.1 193.9c-8.5-1.3-17.3-1.9-26.1-1.9c-54.7 0-103.5 24.9-135.8 64H320V208c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16v48H208c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16zM496 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm22.6 144l36.7-36.7c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0L496 345.4l-36.7-36.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6L473.4 368l-36.7 36.7c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L496 390.6l36.7 36.7c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6L518.6 368z\"],campground:[576,512,[9978],\"f6bb\",\"M377 52c11-13.8 8.8-33.9-5-45s-33.9-8.8-45 5L288 60.8 249 12c-11-13.8-31.2-16-45-5s-16 31.2-5 45l48 60L12.3 405.4C4.3 415.4 0 427.7 0 440.4V464c0 26.5 21.5 48 48 48H288 528c26.5 0 48-21.5 48-48V440.4c0-12.7-4.3-25.1-12.3-35L329 112l48-60zM288 448H168.5L288 291.7 407.5 448H288z\"],\"folder-plus\":[512,512,[],\"f65e\",\"M512 416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H192c20.1 0 39.1 9.5 51.2 25.6l19.2 25.6c6 8.1 15.5 12.8 25.6 12.8H448c35.3 0 64 28.7 64 64V416zM232 376c0 13.3 10.7 24 24 24s24-10.7 24-24V312h64c13.3 0 24-10.7 24-24s-10.7-24-24-24H280V200c0-13.3-10.7-24-24-24s-24 10.7-24 24v64H168c-13.3 0-24 10.7-24 24s10.7 24 24 24h64v64z\"],futbol:[512,512,[9917,\"futbol-ball\",\"soccer-ball\"],\"f1e3\",\"M417.3 360.1l-71.6-4.8c-5.2-.3-10.3 1.1-14.5 4.2s-7.2 7.4-8.4 12.5l-17.6 69.6C289.5 445.8 273 448 256 448s-33.5-2.2-49.2-6.4L189.2 372c-1.3-5-4.3-9.4-8.4-12.5s-9.3-4.5-14.5-4.2l-71.6 4.8c-17.6-27.2-28.5-59.2-30.4-93.6L125 228.3c4.4-2.8 7.6-7 9.2-11.9s1.4-10.2-.5-15l-26.7-66.6C128 109.2 155.3 89 186.7 76.9l55.2 46c4 3.3 9 5.1 14.1 5.1s10.2-1.8 14.1-5.1l55.2-46c31.3 12.1 58.7 32.3 79.6 57.9l-26.7 66.6c-1.9 4.8-2.1 10.1-.5 15s4.9 9.1 9.2 11.9l60.7 38.2c-1.9 34.4-12.8 66.4-30.4 93.6zM256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm14.1-325.7c-8.4-6.1-19.8-6.1-28.2 0L194 221c-8.4 6.1-11.9 16.9-8.7 26.8l18.3 56.3c3.2 9.9 12.4 16.6 22.8 16.6h59.2c10.4 0 19.6-6.7 22.8-16.6l18.3-56.3c3.2-9.9-.3-20.7-8.7-26.8l-47.9-34.8z\"],paintbrush:[576,512,[128396,\"paint-brush\"],\"f1fc\",\"M339.3 367.1c27.3-3.9 51.9-19.4 67.2-42.9L568.2 74.1c12.6-19.5 9.4-45.3-7.6-61.2S517.7-4.4 499.1 9.6L262.4 187.2c-24 18-38.2 46.1-38.4 76.1L339.3 367.1zm-19.6 25.4l-116-104.4C143.9 290.3 96 339.6 96 400c0 3.9 .2 7.8 .6 11.6C98.4 429.1 86.4 448 68.8 448H64c-17.7 0-32 14.3-32 32s14.3 32 32 32H208c61.9 0 112-50.1 112-112c0-2.5-.1-5-.2-7.5z\"],lock:[448,512,[128274],\"f023\",\"M144 144v48H304V144c0-44.2-35.8-80-80-80s-80 35.8-80 80zM80 192V144C80 64.5 144.5 0 224 0s144 64.5 144 144v48h16c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V256c0-35.3 28.7-64 64-64H80z\"],\"gas-pump\":[512,512,[9981],\"f52f\",\"M32 64C32 28.7 60.7 0 96 0H256c35.3 0 64 28.7 64 64V256h8c48.6 0 88 39.4 88 88v32c0 13.3 10.7 24 24 24s24-10.7 24-24V222c-27.6-7.1-48-32.2-48-62V96L384 64c-8.8-8.8-8.8-23.2 0-32s23.2-8.8 32 0l77.3 77.3c12 12 18.7 28.3 18.7 45.3V168v24 32V376c0 39.8-32.2 72-72 72s-72-32.2-72-72V344c0-22.1-17.9-40-40-40h-8V448c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32V64zM96 80v96c0 8.8 7.2 16 16 16H240c8.8 0 16-7.2 16-16V80c0-8.8-7.2-16-16-16H112c-8.8 0-16 7.2-16 16z\"],\"hot-tub-person\":[512,512,[\"hot-tub\"],\"f593\",\"M272 24c0-13.3-10.7-24-24-24s-24 10.7-24 24v5.2c0 34 14.4 66.4 39.7 89.2l16.4 14.8c15.2 13.7 23.8 33.1 23.8 53.5V200c0 13.3 10.7 24 24 24s24-10.7 24-24V186.8c0-34-14.4-66.4-39.7-89.2L295.8 82.8C280.7 69.1 272 49.7 272 29.2V24zM0 320v16V448c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V320c0-35.3-28.7-64-64-64H277.3c-13.8 0-27.3-4.5-38.4-12.8l-85.3-64C137 166.7 116.8 160 96 160c-53 0-96 43-96 96v64zm128 16v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V336c0-8.8 7.2-16 16-16s16 7.2 16 16zm80-16c8.8 0 16 7.2 16 16v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V336c0-8.8 7.2-16 16-16zm112 16v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V336c0-8.8 7.2-16 16-16s16 7.2 16 16zm80-16c8.8 0 16 7.2 16 16v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V336c0-8.8 7.2-16 16-16zM360 0c-13.3 0-24 10.7-24 24v5.2c0 34 14.4 66.4 39.7 89.2l16.4 14.8c15.2 13.7 23.8 33.1 23.8 53.5V200c0 13.3 10.7 24 24 24s24-10.7 24-24V186.8c0-34-14.4-66.4-39.7-89.2L407.8 82.8C392.7 69.1 384 49.7 384 29.2V24c0-13.3-10.7-24-24-24zM64 128A64 64 0 1 0 64 0a64 64 0 1 0 0 128z\"],\"map-location\":[576,512,[\"map-marked\"],\"f59f\",\"M302.8 312C334.9 271.9 408 174.6 408 120C408 53.7 354.3 0 288 0S168 53.7 168 120c0 54.6 73.1 151.9 105.2 192c7.7 9.6 22 9.6 29.6 0zM416 503l144.9-58c9.1-3.6 15.1-12.5 15.1-22.3V152c0-17-17.1-28.6-32.9-22.3l-116 46.4c-.5 1.2-1 2.5-1.5 3.7c-2.9 6.8-6.1 13.7-9.6 20.6V503zM15.1 187.3C6 191 0 199.8 0 209.6V480.4c0 17 17.1 28.6 32.9 22.3L160 451.8V200.4c-3.5-6.9-6.7-13.8-9.6-20.6c-5.6-13.2-10.4-27.4-12.8-41.5l-122.6 49zM384 255c-20.5 31.3-42.3 59.6-56.2 77c-20.5 25.6-59.1 25.6-79.6 0c-13.9-17.4-35.7-45.7-56.2-77V449.4l192 54.9V255z\"],\"house-flood-water\":[576,512,[],\"e50e\",\"M306.8 6.1C295.6-2 280.4-2 269.2 6.1l-176 128c-11.2 8.2-15.9 22.6-11.6 35.8S98.1 192 112 192h16v73c1.7 1 3.3 2 4.9 3.1c18 12.4 40.1 20.3 59.2 20.3c21.1 0 42-8.5 59.2-20.3c22.1-15.5 51.6-15.5 73.7 0c18.4 12.7 39.6 20.3 59.2 20.3c19 0 41.2-7.9 59.2-20.3c1.5-1 3-2 4.5-2.9l-.3-73.2H464c13.9 0 26.1-8.9 30.4-22.1s-.4-27.6-11.6-35.8l-176-128zM269.5 309.9C247 325.4 219.5 336 192 336c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C126.5 389.7 159 400 192 400c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.5-27.3-10.1-39.2-1.7l0 0C439.4 325.2 410.9 336 384 336c-27.5 0-55-10.6-77.5-26.1c-11.1-7.9-25.9-7.9-37 0zM384 448c-27.5 0-55-10.6-77.5-26.1c-11.1-7.9-25.9-7.9-37 0C247 437.4 219.5 448 192 448c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C126.5 501.7 159 512 192 512c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.4-27.3-10.1-39.2-1.7l0 0C439.4 437.2 410.9 448 384 448z\"],tree:[448,512,[127794],\"f1bb\",\"M210.6 5.9L62 169.4c-3.9 4.2-6 9.8-6 15.5C56 197.7 66.3 208 79.1 208H104L30.6 281.4c-4.2 4.2-6.6 10-6.6 16C24 309.9 34.1 320 46.6 320H80L5.4 409.5C1.9 413.7 0 419 0 424.5c0 13 10.5 23.5 23.5 23.5H192v32c0 17.7 14.3 32 32 32s32-14.3 32-32V448H424.5c13 0 23.5-10.5 23.5-23.5c0-5.5-1.9-10.8-5.4-15L368 320h33.4c12.5 0 22.6-10.1 22.6-22.6c0-6-2.4-11.8-6.6-16L344 208h24.9c12.7 0 23.1-10.3 23.1-23.1c0-5.7-2.1-11.3-6-15.5L237.4 5.9C234 2.1 229.1 0 224 0s-10 2.1-13.4 5.9z\"],\"bridge-lock\":[640,512,[],\"e4cc\",\"M32 64c0-17.7 14.3-32 32-32H576c17.7 0 32 14.3 32 32s-14.3 32-32 32H536v64h-8c-61.9 0-112 50.1-112 112v24.6c-9.9 5.8-18.2 14.1-23.8 24.1c-17.6-20-43.4-32.7-72.2-32.7c-53 0-96 43-96 96v64c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32V384c0-53-43-96-96-96V160h72V96H64C46.3 96 32 81.7 32 64zM408 96v64h80V96H408zm-48 64V96H280v64h80zM152 96v64h80V96H152zM528 240c-17.7 0-32 14.3-32 32v48h64V272c0-17.7-14.3-32-32-32zm-80 32c0-44.2 35.8-80 80-80s80 35.8 80 80v48c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H448c-17.7 0-32-14.3-32-32V352c0-17.7 14.3-32 32-32V272z\"],\"sack-dollar\":[512,512,[128176],\"f81d\",\"M320 96H192L144.6 24.9C137.5 14.2 145.1 0 157.9 0H354.1c12.8 0 20.4 14.2 13.3 24.9L320 96zM192 128H320c3.8 2.5 8.1 5.3 13 8.4C389.7 172.7 512 250.9 512 416c0 53-43 96-96 96H96c-53 0-96-43-96-96C0 250.9 122.3 172.7 179 136.4l0 0 0 0c4.8-3.1 9.2-5.9 13-8.4zm84 88c0-11-9-20-20-20s-20 9-20 20v14c-7.6 1.7-15.2 4.4-22.2 8.5c-13.9 8.3-25.9 22.8-25.8 43.9c.1 20.3 12 33.1 24.7 40.7c11 6.6 24.7 10.8 35.6 14l1.7 .5c12.6 3.8 21.8 6.8 28 10.7c5.1 3.2 5.8 5.4 5.9 8.2c.1 5-1.8 8-5.9 10.5c-5 3.1-12.9 5-21.4 4.7c-11.1-.4-21.5-3.9-35.1-8.5c-2.3-.8-4.7-1.6-7.2-2.4c-10.5-3.5-21.8 2.2-25.3 12.6s2.2 21.8 12.6 25.3c1.9 .6 4 1.3 6.1 2.1l0 0 0 0c8.3 2.9 17.9 6.2 28.2 8.4V424c0 11 9 20 20 20s20-9 20-20V410.2c8-1.7 16-4.5 23.2-9c14.3-8.9 25.1-24.1 24.8-45c-.3-20.3-11.7-33.4-24.6-41.6c-11.5-7.2-25.9-11.6-37.1-15l0 0-.7-.2c-12.8-3.9-21.9-6.7-28.3-10.5c-5.2-3.1-5.3-4.9-5.3-6.7c0-3.7 1.4-6.5 6.2-9.3c5.4-3.2 13.6-5.1 21.5-5c9.6 .1 20.2 2.2 31.2 5.2c10.7 2.8 21.6-3.5 24.5-14.2s-3.5-21.6-14.2-24.5c-6.5-1.7-13.7-3.4-21.1-4.7V216z\"],\"pen-to-square\":[512,512,[\"edit\"],\"f044\",\"M471.6 21.7c-21.9-21.9-57.3-21.9-79.2 0L362.3 51.7l97.9 97.9 30.1-30.1c21.9-21.9 21.9-57.3 0-79.2L471.6 21.7zm-299.2 220c-6.1 6.1-10.8 13.6-13.5 21.9l-29.6 88.8c-2.9 8.6-.6 18.1 5.8 24.6s15.9 8.7 24.6 5.8l88.8-29.6c8.2-2.7 15.7-7.4 21.9-13.5L437.7 172.3 339.7 74.3 172.4 241.7zM96 64C43 64 0 107 0 160V416c0 53 43 96 96 96H352c53 0 96-43 96-96V320c0-17.7-14.3-32-32-32s-32 14.3-32 32v96c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V160c0-17.7 14.3-32 32-32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H96z\"],\"car-side\":[640,512,[128663],\"f5e4\",\"M171.3 96H224v96H111.3l30.4-75.9C146.5 104 158.2 96 171.3 96zM272 192V96h81.2c9.7 0 18.9 4.4 25 12l67.2 84H272zm256.2 1L428.2 68c-18.2-22.8-45.8-36-75-36H171.3c-39.3 0-74.6 23.9-89.1 60.3L40.6 196.4C16.8 205.8 0 228.9 0 256V368c0 17.7 14.3 32 32 32H65.3c7.6 45.4 47.1 80 94.7 80s87.1-34.6 94.7-80H385.3c7.6 45.4 47.1 80 94.7 80s87.1-34.6 94.7-80H608c17.7 0 32-14.3 32-32V320c0-65.2-48.8-119-111.8-127zM434.7 368a48 48 0 1 1 90.5 32 48 48 0 1 1 -90.5-32zM160 336a48 48 0 1 1 0 96 48 48 0 1 1 0-96z\"],\"share-nodes\":[448,512,[\"share-alt\"],\"f1e0\",\"M352 224c53 0 96-43 96-96s-43-96-96-96s-96 43-96 96c0 4 .2 8 .7 11.9l-94.1 47C145.4 170.2 121.9 160 96 160c-53 0-96 43-96 96s43 96 96 96c25.9 0 49.4-10.2 66.6-26.9l94.1 47c-.5 3.9-.7 7.8-.7 11.9c0 53 43 96 96 96s96-43 96-96s-43-96-96-96c-25.9 0-49.4 10.2-66.6 26.9l-94.1-47c.5-3.9 .7-7.8 .7-11.9s-.2-8-.7-11.9l94.1-47C302.6 213.8 326.1 224 352 224z\"],\"heart-circle-minus\":[576,512,[],\"e4ff\",\"M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9l2.6-2.4C267.2 438.6 256 404.6 256 368c0-97.2 78.8-176 176-176c28.3 0 55 6.7 78.7 18.5c.9-6.5 1.3-13 1.3-19.6v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5zM576 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-64 0c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16s7.2-16 16-16H496c8.8 0 16 7.2 16 16z\"],\"hourglass-half\":[384,512,[\"hourglass-2\"],\"f252\",\"M32 0C14.3 0 0 14.3 0 32S14.3 64 32 64V75c0 42.4 16.9 83.1 46.9 113.1L146.7 256 78.9 323.9C48.9 353.9 32 394.6 32 437v11c-17.7 0-32 14.3-32 32s14.3 32 32 32H64 320h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V437c0-42.4-16.9-83.1-46.9-113.1L237.3 256l67.9-67.9c30-30 46.9-70.7 46.9-113.1V64c17.7 0 32-14.3 32-32s-14.3-32-32-32H320 64 32zM96 75V64H288V75c0 19-5.6 37.4-16 53H112c-10.3-15.6-16-34-16-53zm16 309c3.5-5.3 7.6-10.3 12.1-14.9L192 301.3l67.9 67.9c4.6 4.6 8.6 9.6 12.1 14.9H112z\"],microscope:[512,512,[128300],\"f610\",\"M160 32c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32c17.7 0 32 14.3 32 32V288c0 17.7-14.3 32-32 32c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32c-17.7 0-32-14.3-32-32V64c0-17.7 14.3-32 32-32zM32 448H320c70.7 0 128-57.3 128-128s-57.3-128-128-128V128c106 0 192 86 192 192c0 49.2-18.5 94-48.9 128H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H320 32c-17.7 0-32-14.3-32-32s14.3-32 32-32zm80-64H304c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],sink:[512,512,[],\"e06d\",\"M288 96c0-17.7 14.3-32 32-32s32 14.3 32 32s14.3 32 32 32s32-14.3 32-32c0-53-43-96-96-96s-96 43-96 96V288H160V264c0-30.9-25.1-56-56-56H56c-13.3 0-24 10.7-24 24s10.7 24 24 24h48c4.4 0 8 3.6 8 8v24H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H256 480c17.7 0 32-14.3 32-32s-14.3-32-32-32H400V264c0-4.4 3.6-8 8-8h56c13.3 0 24-10.7 24-24s-10.7-24-24-24H408c-30.9 0-56 25.1-56 56v24H288V96zM480 416V384H32v32c0 53 43 96 96 96H384c53 0 96-43 96-96z\"],\"bag-shopping\":[448,512,[\"shopping-bag\"],\"f290\",\"M160 112c0-35.3 28.7-64 64-64s64 28.7 64 64v48H160V112zm-48 48H48c-26.5 0-48 21.5-48 48V416c0 53 43 96 96 96H352c53 0 96-43 96-96V208c0-26.5-21.5-48-48-48H336V112C336 50.1 285.9 0 224 0S112 50.1 112 112v48zm24 48a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm152 24a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z\"],\"arrow-down-z-a\":[576,512,[\"sort-alpha-desc\",\"sort-alpha-down-alt\"],\"f881\",\"M183.6 469.6C177.5 476.2 169 480 160 480s-17.5-3.8-23.6-10.4l-88-96c-11.9-13-11.1-33.3 2-45.2s33.3-11.1 45.2 2L128 365.7V64c0-17.7 14.3-32 32-32s32 14.3 32 32V365.7l32.4-35.4c11.9-13 32.2-13.9 45.2-2s13.9 32.2 2 45.2l-88 96zM320 64c0-17.7 14.3-32 32-32H480c12.9 0 24.6 7.8 29.6 19.8s2.2 25.7-6.9 34.9L429.3 160H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H352c-12.9 0-24.6-7.8-29.6-19.8s-2.2-25.7 6.9-34.9L402.7 96H352c-17.7 0-32-14.3-32-32zm96 192c12.1 0 23.2 6.8 28.6 17.7l64 128 16 32c7.9 15.8 1.5 35-14.3 42.9s-35 1.5-42.9-14.3L460.2 448H371.8l-7.2 14.3c-7.9 15.8-27.1 22.2-42.9 14.3s-22.2-27.1-14.3-42.9l16-32 64-128c5.4-10.8 16.5-17.7 28.6-17.7zM395.8 400h40.4L416 359.6 395.8 400z\"],mitten:[448,512,[],\"f7b5\",\"M352 384H64L5.4 178.9C1.8 166.4 0 153.4 0 140.3C0 62.8 62.8 0 140.3 0h3.4c66 0 123.5 44.9 139.5 108.9l31.4 125.8 17.6-20.1C344.8 200.2 362.9 192 382 192h2.8c34.9 0 63.3 28.3 63.3 63.3c0 15.9-6 31.2-16.8 42.9L352 384zM32 448c0-17.7 14.3-32 32-32H352c17.7 0 32 14.3 32 32v32c0 17.7-14.3 32-32 32H64c-17.7 0-32-14.3-32-32V448z\"],\"person-rays\":[512,512,[],\"e54d\",\"M208 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9l-28.6 47.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l58.3 97c9.1 15.1 4.2 34.8-10.9 43.9s-34.8 4.2-43.9-10.9L328 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H248zM7 7C16.4-2.3 31.6-2.3 41 7l80 80c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L7 41C-2.3 31.6-2.3 16.4 7 7zM471 7c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-80 80c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9L471 7zM7 505c-9.4-9.4-9.4-24.6 0-33.9l80-80c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9L41 505c-9.4 9.4-24.6 9.4-33.9 0zm464 0l-80-80c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l80 80c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0z\"],users:[640,512,[],\"f0c0\",\"M144 0a80 80 0 1 1 0 160A80 80 0 1 1 144 0zM512 0a80 80 0 1 1 0 160A80 80 0 1 1 512 0zM0 298.7C0 239.8 47.8 192 106.7 192h42.7c15.9 0 31 3.5 44.6 9.7c-1.3 7.2-1.9 14.7-1.9 22.3c0 38.2 16.8 72.5 43.3 96c-.2 0-.4 0-.7 0H21.3C9.6 320 0 310.4 0 298.7zM405.3 320c-.2 0-.4 0-.7 0c26.6-23.5 43.3-57.8 43.3-96c0-7.6-.7-15-1.9-22.3c13.6-6.3 28.7-9.7 44.6-9.7h42.7C592.2 192 640 239.8 640 298.7c0 11.8-9.6 21.3-21.3 21.3H405.3zM224 224a96 96 0 1 1 192 0 96 96 0 1 1 -192 0zM128 485.3C128 411.7 187.7 352 261.3 352H378.7C452.3 352 512 411.7 512 485.3c0 14.7-11.9 26.7-26.7 26.7H154.7c-14.7 0-26.7-11.9-26.7-26.7z\"],\"eye-slash\":[640,512,[],\"f070\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L525.6 386.7c39.6-40.6 66.4-86.1 79.9-118.4c3.3-7.9 3.3-16.7 0-24.6c-14.9-35.7-46.2-87.7-93-131.1C465.5 68.8 400.8 32 320 32c-68.2 0-125 26.3-169.3 60.8L38.8 5.1zM223.1 149.5C248.6 126.2 282.7 112 320 112c79.5 0 144 64.5 144 144c0 24.9-6.3 48.3-17.4 68.7L408 294.5c8.4-19.3 10.6-41.4 4.8-63.3c-11.1-41.5-47.8-69.4-88.6-71.1c-5.8-.2-9.2 6.1-7.4 11.7c2.1 6.4 3.3 13.2 3.3 20.3c0 10.2-2.4 19.8-6.6 28.3l-90.3-70.8zM373 389.9c-16.4 6.5-34.3 10.1-53 10.1c-79.5 0-144-64.5-144-144c0-6.9 .5-13.6 1.4-20.2L83.1 161.5C60.3 191.2 44 220.8 34.5 243.7c-3.3 7.9-3.3 16.7 0 24.6c14.9 35.7 46.2 87.7 93 131.1C174.5 443.2 239.2 480 320 480c47.8 0 89.9-12.9 126.2-32.5L373 389.9z\"],\"flask-vial\":[640,512,[],\"e4f3\",\"M175 389.4c-9.8 16-15 34.3-15 53.1c-10 3.5-20.8 5.5-32 5.5c-53 0-96-43-96-96V64C14.3 64 0 49.7 0 32S14.3 0 32 0H96h64 64c17.7 0 32 14.3 32 32s-14.3 32-32 32V309.9l-49 79.6zM96 64v96h64V64H96zM352 0H480h32c17.7 0 32 14.3 32 32s-14.3 32-32 32V214.9L629.7 406.2c6.7 10.9 10.3 23.5 10.3 36.4c0 38.3-31.1 69.4-69.4 69.4H261.4c-38.3 0-69.4-31.1-69.4-69.4c0-12.8 3.6-25.4 10.3-36.4L320 214.9V64c-17.7 0-32-14.3-32-32s14.3-32 32-32h32zm32 64V224c0 5.9-1.6 11.7-4.7 16.8L330.5 320h171l-48.8-79.2c-3.1-5-4.7-10.8-4.7-16.8V64H384z\"],hand:[512,512,[129306,9995,\"hand-paper\"],\"f256\",\"M288 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V336c0 1.5 0 3.1 .1 4.6L67.6 283c-16-15.2-41.3-14.6-56.6 1.4s-14.6 41.3 1.4 56.6L124.8 448c43.1 41.1 100.4 64 160 64H304c97.2 0 176-78.8 176-176V128c0-17.7-14.3-32-32-32s-32 14.3-32 32V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V32z\"],om:[512,512,[128329],\"f679\",\"M379.3 4.7c-6.2-6.2-16.4-6.2-22.6 0l-16 16c-6.2 6.2-6.2 16.4 0 22.6l16 16c6.2 6.2 16.4 6.2 22.6 0l16-16c6.2-6.2 6.2-16.4 0-22.6l-16-16zM281 66.7c-2.2-1.5-4.9-2.5-7.7-2.7c-.6 0-1.3-.1-1.9 0c-3.9 .2-7.4 1.7-10.1 4.2c-.9 .8-1.6 1.7-2.3 2.6c-1.7 2.4-2.7 5.3-2.9 8.5c0 .7 0 1.4 0 2.1c.2 2.2 .9 4.3 1.9 6.2l.3 .6c.3 .6 .8 1.4 1.4 2.4c1.2 2 2.9 4.8 5.1 8.2c4.4 6.7 11.1 15.5 20 24.4C302.4 141.1 330.3 160 368 160c31.2 0 56.6-10.4 73.9-20.2c8.7-5 15.6-9.9 20.4-13.8c2.4-1.9 4.3-3.6 5.7-4.9c.7-.6 1.3-1.2 1.7-1.6l.6-.5 .1-.1 .1-.1 0 0 0 0c5.9-5.8 9.5-13.9 9.5-22.8c0-17.7-14.3-32-32-32c-8.7 0-16.7 3.5-22.4 9.2c-.1 .1-.2 .2-.5 .4c-.5 .5-1.5 1.3-2.8 2.4c-2.7 2.2-6.8 5.2-12.1 8.2C399.4 90.4 384.8 96 368 96c-20.8 0-42.4-7-59.5-14.6c-8.4-3.7-15.4-7.5-20.3-10.3c-2.4-1.4-4.3-2.5-5.6-3.3c-.6-.4-1.1-.7-1.4-.9l-.3-.2zM115.2 169.6c8-6 17.9-9.6 28.8-9.6c26.5 0 48 21.5 48 48s-21.5 48-48 48H109.8c-7.6 0-13.8 6.2-13.8 13.8c0 1.5 .2 2.9 .7 4.4l8 24c4.4 13.1 16.6 21.9 30.4 21.9H144h16c35.3 0 64 28.7 64 64s-28.7 64-64 64c-50.8 0-82.7-21.5-102.2-42.8c-9.9-10.8-16.6-21.6-20.9-29.7c-2.1-4-3.6-7.3-4.5-9.6c-.5-1.1-.8-2-1-2.5l-.2-.5c-.3-.9-.7-1.8-1.1-2.6c-1.2-2.2-2.8-4-4.7-5.4c-1.9-1.4-4.1-2.3-6.5-2.8c-1.4-.3-2.9-.3-4.4-.2c-2.5 .2-4.8 1-6.8 2.3c-1.1 .7-2.2 1.5-3.1 2.5c-2.4 2.5-4.1 5.8-4.5 9.5c-.1 .6-.1 1.1-.1 1.7c0 0 0 0 0 0c0 .8 .1 1.7 .2 2.5l0 .1c0 .3 .1 .8 .2 1.3c.2 1.1 .4 2.7 .8 4.6c.8 3.9 2 9.4 3.9 15.9c3.8 13 10.3 30.4 21.3 48C48.7 476.2 89.4 512 160 512c70.7 0 128-57.3 128-128c0-23.3-6.2-45.2-17.1-64h22.6c25.5 0 49.9-10.1 67.9-28.1l26.5-26.5c6-6 14.1-9.4 22.6-9.4H416c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32c-25.7 0-41.4-12.5-51.2-25.6c-5-6.7-8.4-13.4-10.5-18.6c-1.1-2.5-1.8-4.6-2.2-6c-.2-.7-.4-1.2-.5-1.5l-.1-.2c-.3-1.3-.8-2.6-1.5-3.8c-1.1-2-2.6-3.8-4.4-5.1c-2.7-2-6-3.2-9.6-3.2l-.2 0c-8 .1-14.6 6.1-15.6 13.9l0 0c0 .3-.1 .6-.2 1.1c-.1 .9-.3 2.1-.4 3.6c-.3 3-.6 7.3-.6 12.4c0 10.1 1.1 23.9 5.8 38.1c4.8 14.3 13.4 29.3 28.6 40.7C368.7 473.3 389.3 480 416 480c53 0 96-43 96-96V288c0-53-43-96-96-96h-5.5c-25.5 0-49.9 10.1-67.9 28.1l-26.5 26.5c-6 6-14.1 9.4-22.6 9.4H245.2c6.9-14.5 10.8-30.8 10.8-48c0-61.9-50.1-112-112-112c-25.2 0-48.5 8.3-67.2 22.4c-14.1 10.6-17 30.7-6.4 44.8s30.7 17 44.8 6.4z\"],worm:[512,512,[],\"e599\",\"M256 96c0-53 43-96 96-96h38.4C439.9 0 480 40.1 480 89.6V176v16V376c0 75.1-60.9 136-136 136s-136-60.9-136-136V296c0-22.1-17.9-40-40-40s-40 17.9-40 40V464c0 26.5-21.5 48-48 48s-48-21.5-48-48V296c0-75.1 60.9-136 136-136s136 60.9 136 136v80c0 22.1 17.9 40 40 40s40-17.9 40-40V192H352c-53 0-96-43-96-96zm144-8a24 24 0 1 0 -48 0 24 24 0 1 0 48 0z\"],\"house-circle-xmark\":[640,512,[],\"e50b\",\"M320.7 352c8.1-89.7 83.5-160 175.3-160c8.9 0 17.6 .7 26.1 1.9L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32v69.7c-.1 .9-.1 1.8-.1 2.8V472c0 22.1 17.9 40 40 40h16c1.2 0 2.4-.1 3.6-.2c1.5 .1 3 .2 4.5 .2H160h24c22.1 0 40-17.9 40-40V448 384c0-17.7 14.3-32 32-32h64l.7 0zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L518.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L496 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L473.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L496 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z\"],plug:[384,512,[128268],\"f1e6\",\"M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8C297 398 352 333.4 352 256V224c17.7 0 32-14.3 32-32s-14.3-32-32-32H32z\"],\"chevron-up\":[512,512,[],\"f077\",\"M233.4 105.4c12.5-12.5 32.8-12.5 45.3 0l192 192c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L256 173.3 86.6 342.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l192-192z\"],\"hand-spock\":[576,512,[128406],\"f259\",\"M246.9 23.7C242.3 6.6 224.8-3.5 207.7 1.1s-27.2 22.1-22.6 39.2L238 237.8c2.5 9.2-4.5 18.2-14 18.2c-6.4 0-12-4.2-13.9-10.3L166.6 102.7c-5.1-16.9-23-26.4-39.9-21.3s-26.4 23-21.3 39.9l62.8 206.4c2.4 7.9-7.2 13.8-13.2 8.1L99.6 283c-16-15.2-41.3-14.6-56.6 1.4s-14.6 41.3 1.4 56.6L156.8 448c43.1 41.1 100.4 64 160 64h10.9 8.2c.1 0 .1-.1 .1-.1s.1-.1 .1-.1c58.3-3.5 108.6-43.2 125.3-99.7l81.2-275c5-16.9-4.7-34.7-21.6-39.8s-34.7 4.7-39.8 21.6L443.5 247.1c-1.6 5.3-6.4 8.9-12 8.9c-7.9 0-13.8-7.3-12.2-15.1l36-170.3c3.7-17.3-7.4-34.3-24.7-37.9s-34.3 7.4-37.9 24.7L355.1 235.1c-2.6 12.2-13.3 20.9-25.8 20.9c-11.9 0-22.4-8-25.4-19.5l-57-212.8z\"],stopwatch:[448,512,[9201],\"f2f2\",\"M176 0c-17.7 0-32 14.3-32 32s14.3 32 32 32h16V98.4C92.3 113.8 16 200 16 304c0 114.9 93.1 208 208 208s208-93.1 208-208c0-41.8-12.3-80.7-33.5-113.2l24.1-24.1c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L355.7 143c-28.1-23-62.2-38.8-99.7-44.6V64h16c17.7 0 32-14.3 32-32s-14.3-32-32-32H224 176zm72 192V320c0 13.3-10.7 24-24 24s-24-10.7-24-24V192c0-13.3 10.7-24 24-24s24 10.7 24 24z\"],\"face-kiss\":[512,512,[128535,\"kiss\"],\"f596\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm48.7-198.3c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4c-2.7 1.5-5.7 3-8.7 4.3c3.1 1.3 6 2.7 8.7 4.3c6.6 3.7 12.5 8.2 16.7 13.4c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4C274.7 443.1 257.4 448 240 448c-3.6 0-6.8-2.5-7.7-6s.6-7.2 3.8-9l0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1c-.4-.2-.7-.4-.9-.5l-.2-.1 0 0 0 0 0 0c-2.5-1.4-4.1-4.1-4.1-7s1.6-5.6 4.1-7l0 0 0 0 0 0 0 0 0 0 .2-.1 .3-.2 .6-.4c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1l-.4-.3-.5-.3-.2-.1 0 0 0 0 0 0c-3.2-1.8-4.7-5.5-3.8-9s4.1-6 7.7-6c17.4 0 34.7 4.9 47.9 12.3c6.6 3.7 12.5 8.2 16.7 13.4zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"bridge-circle-xmark\":[640,512,[],\"e4cb\",\"M64 32C46.3 32 32 46.3 32 64s14.3 32 32 32h40v64H32V288c53 0 96 43 96 96v64c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V384c0-53 43-96 96-96c6.3 0 12.4 .6 18.3 1.7C367.1 231.8 426.9 192 496 192c42.5 0 81.6 15.1 112 40.2V160H536V96h40c17.7 0 32-14.3 32-32s-14.3-32-32-32H64zM488 96v64H408V96h80zM360 96v64H280V96h80zM232 96v64H152V96h80zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L518.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L496 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L473.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L496 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z\"],\"face-grin-tongue\":[512,512,[128539,\"grin-tongue\"],\"f589\",\"M0 256C0 368.9 73.1 464.7 174.5 498.8C165.3 484 160 466.6 160 448V400.7c-24-17.5-43.1-41.4-54.8-69.2c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19c12.3-3.8 24.3 6.9 19.3 18.7c-11.8 28-31.1 52-55.4 69.6V448c0 18.6-5.3 36-14.5 50.8C438.9 464.7 512 368.9 512 256C512 114.6 397.4 0 256 0S0 114.6 0 256zm176.4-80a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM320 448V402.6c0-14.7-11.9-26.6-26.6-26.6h-2c-11.3 0-21.1 7.9-23.6 18.9c-2.8 12.6-20.8 12.6-23.6 0c-2.5-11.1-12.3-18.9-23.6-18.9h-2c-14.7 0-26.6 11.9-26.6 26.6V448c0 35.3 28.7 64 64 64s64-28.7 64-64z\"],\"chess-bishop\":[320,512,[9821],\"f43a\",\"M128 0C110.3 0 96 14.3 96 32c0 16.1 11.9 29.4 27.4 31.7C78.4 106.8 8 190 8 288c0 47.4 30.8 72.3 56 84.7V400H256V372.7c25.2-12.5 56-37.4 56-84.7c0-37.3-10.2-72.4-25.3-104.1l-99.4 99.4c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L270.8 154.6c-23.2-38.1-51.8-69.5-74.2-90.9C212.1 61.4 224 48.1 224 32c0-17.7-14.3-32-32-32H128zM48 432L6.6 473.4c-4.2 4.2-6.6 10-6.6 16C0 501.9 10.1 512 22.6 512H297.4c12.5 0 22.6-10.1 22.6-22.6c0-6-2.4-11.8-6.6-16L272 432H48z\"],\"face-grin-wink\":[512,512,[\"grin-wink\"],\"f58c\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM388.1 312.8c12.3-3.8 24.3 6.9 19.3 18.7C382.4 390.6 324.2 432 256.3 432s-126.2-41.4-151.1-100.5c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19zm-16.9-79.2c-17.6-23.5-52.8-23.5-70.4 0c-5.3 7.1-15.3 8.5-22.4 3.2s-8.5-15.3-3.2-22.4c30.4-40.5 91.2-40.5 121.6 0c5.3 7.1 3.9 17.1-3.2 22.4s-17.1 3.9-22.4-3.2zM176.4 176a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"ear-deaf\":[512,512,[\"deaf\",\"deafness\",\"hard-of-hearing\"],\"f2a4\",\"M502.6 54.6l-40 40c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l40-40c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3zm-320 320l-128 128c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l128-128c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3zM240 128c-57.6 0-105.1 43.6-111.3 99.5c-1.9 17.6-17.8 30.2-35.3 28.3s-30.2-17.8-28.3-35.3C74.8 132.5 149.4 64 240 64c97.2 0 176 78.8 176 176c0 46-17.7 87.9-46.6 119.3c-12 13-17.4 24.8-17.4 34.7V400c0 61.9-50.1 112-112 112c-17.7 0-32-14.3-32-32s14.3-32 32-32c26.5 0 48-21.5 48-48v-6.1c0-32.9 17.4-59.6 34.4-78c18.4-20 29.6-46.6 29.6-75.9c0-61.9-50.1-112-112-112zm0 80c-17.7 0-32 14.3-32 32c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-44.2 35.8-80 80-80s80 35.8 80 80c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-17.7-14.3-32-32-32z\"],\"road-circle-check\":[640,512,[],\"e564\",\"M213.2 32H288V96c0 17.7 14.3 32 32 32s32-14.3 32-32V32h74.8c27.1 0 51.3 17.1 60.3 42.6l42.7 120.6c-10.9-2.1-22.2-3.2-33.8-3.2c-59.5 0-112.1 29.6-144 74.8V224c0-17.7-14.3-32-32-32s-32 14.3-32 32v64c0 17.7 14.3 32 32 32c2.3 0 4.6-.3 6.8-.7c-4.5 15.5-6.8 31.8-6.8 48.7c0 5.4 .2 10.7 .7 16l-.7 0c-17.7 0-32 14.3-32 32v64H86.6C56.5 480 32 455.5 32 425.4c0-6.2 1.1-12.4 3.1-18.2L152.9 74.6C162 49.1 186.1 32 213.2 32zM352 368a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm211.3-43.3c-6.2-6.2-16.4-6.2-22.6 0L480 385.4l-28.7-28.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l40 40c6.2 6.2 16.4 6.2 22.6 0l72-72c6.2-6.2 6.2-16.4 0-22.6z\"],\"dice-five\":[448,512,[9860],\"f523\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm64 96a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM96 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM224 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm64-64a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm32 160a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"square-rss\":[448,512,[\"rss-square\"],\"f143\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM96 136c0-13.3 10.7-24 24-24c137 0 248 111 248 248c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-110.5-89.5-200-200-200c-13.3 0-24-10.7-24-24zm0 96c0-13.3 10.7-24 24-24c83.9 0 152 68.1 152 152c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-57.4-46.6-104-104-104c-13.3 0-24-10.7-24-24zm0 120a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],\"land-mine-on\":[640,512,[],\"e51b\",\"M344 24V168c0 13.3-10.7 24-24 24s-24-10.7-24-24V24c0-13.3 10.7-24 24-24s24 10.7 24 24zM192 320c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32v32H192V320zm-77.3 90.5c8.1-16.3 24.8-26.5 42.9-26.5H482.3c18.2 0 34.8 10.3 42.9 26.5l27.6 55.2C563.5 487 548 512 524.2 512H115.8c-23.8 0-39.3-25-28.6-46.3l27.6-55.2zM36.3 138.3c7.5-10.9 22.5-13.6 33.4-6.1l104 72c10.9 7.5 13.6 22.5 6.1 33.4s-22.5 13.6-33.4 6.1l-104-72c-10.9-7.5-13.6-22.5-6.1-33.4zm534.1-6.1c10.9-7.5 25.8-4.8 33.4 6.1s4.8 25.8-6.1 33.4l-104 72c-10.9 7.5-25.8 4.8-33.4-6.1s-4.8-25.8 6.1-33.4l104-72z\"],\"i-cursor\":[256,512,[],\"f246\",\"M.1 29.3C-1.4 47 11.7 62.4 29.3 63.9l8 .7C70.5 67.3 96 95 96 128.3V224H64c-17.7 0-32 14.3-32 32s14.3 32 32 32H96v95.7c0 33.3-25.5 61-58.7 63.8l-8 .7C11.7 449.6-1.4 465 .1 482.7s16.9 30.7 34.5 29.2l8-.7c34.1-2.8 64.2-18.9 85.4-42.9c21.2 24 51.2 40.1 85.4 42.9l8 .7c17.6 1.5 33.1-11.6 34.5-29.2s-11.6-33.1-29.2-34.5l-8-.7C185.5 444.7 160 417 160 383.7V288h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H160V128.3c0-33.3 25.5-61 58.7-63.8l8-.7c17.6-1.5 30.7-16.9 29.2-34.5S239-1.4 221.3 .1l-8 .7C179.2 3.6 149.2 19.7 128 43.7c-21.2-24-51.2-40-85.4-42.9l-8-.7C17-1.4 1.6 11.7 .1 29.3z\"],stamp:[512,512,[],\"f5bf\",\"M312 201.8c0-17.4 9.2-33.2 19.9-47C344.5 138.5 352 118.1 352 96c0-53-43-96-96-96s-96 43-96 96c0 22.1 7.5 42.5 20.1 58.8c10.7 13.8 19.9 29.6 19.9 47c0 29.9-24.3 54.2-54.2 54.2H112C50.1 256 0 306.1 0 368c0 20.9 13.4 38.7 32 45.3V464c0 26.5 21.5 48 48 48H432c26.5 0 48-21.5 48-48V413.3c18.6-6.6 32-24.4 32-45.3c0-61.9-50.1-112-112-112H366.2c-29.9 0-54.2-24.3-54.2-54.2zM416 416v32H96V416H416z\"],stairs:[576,512,[],\"e289\",\"M384 64c0-17.7 14.3-32 32-32H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H448v96c0 17.7-14.3 32-32 32H320v96c0 17.7-14.3 32-32 32H192v96c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32h96V320c0-17.7 14.3-32 32-32h96V192c0-17.7 14.3-32 32-32h96V64z\"],i:[320,512,[105],\"49\",\"M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96h96V416H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H192V96h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H160 32z\"],\"hryvnia-sign\":[384,512,[8372,\"hryvnia\"],\"f6f2\",\"M121.9 116.2C138.3 103.1 158.7 96 179.6 96H223c27.1 0 49 21.9 49 49c0 11.5-4 22.4-11.1 31H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H155.5l-50.6 28.9c-1.7 1-3.4 2-5.1 3.1H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H52.3c-2.8 9.9-4.3 20.4-4.3 31c0 62.4 50.6 113 113 113h43.4c35.5 0 70-12.1 97.7-34.3L308 441c13.8-11 16-31.2 5-45s-31.2-16-45-5l-5.9 4.7c-16.4 13.1-36.7 20.2-57.7 20.2H161c-27.1 0-49-21.9-49-49c0-11.5 4-22.4 11.1-31H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H228.5l50.6-28.9c1.7-1 3.4-2 5.1-3.1H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H331.7c2.8-10 4.3-20.4 4.3-31c0-62.4-50.6-113-113-113H179.6c-35.5 0-70 12.1-97.7 34.3L76 71c-13.8 11-16 31.2-5 45s31.2 16 45 5l5.9-4.7z\"],pills:[576,512,[],\"f484\",\"M112 96c-26.5 0-48 21.5-48 48V256h96V144c0-26.5-21.5-48-48-48zM0 144C0 82.1 50.1 32 112 32s112 50.1 112 112V368c0 61.9-50.1 112-112 112S0 429.9 0 368V144zM554.9 399.4c-7.1 12.3-23.7 13.1-33.8 3.1L333.5 214.9c-10-10-9.3-26.7 3.1-33.8C360 167.7 387.1 160 416 160c88.4 0 160 71.6 160 160c0 28.9-7.7 56-21.1 79.4zm-59.5 59.5C472 472.3 444.9 480 416 480c-88.4 0-160-71.6-160-160c0-28.9 7.7-56 21.1-79.4c7.1-12.3 23.7-13.1 33.8-3.1L498.5 425.1c10 10 9.3 26.7-3.1 33.8z\"],\"face-grin-wide\":[512,512,[128515,\"grin-alt\"],\"f581\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM388.1 312.8c12.3-3.8 24.3 6.9 19.3 18.7C382.4 390.6 324.2 432 256.3 432s-126.2-41.4-151.1-100.5c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19zM208 192c0 35.3-14.3 64-32 64s-32-28.7-32-64s14.3-64 32-64s32 28.7 32 64zm128 64c-17.7 0-32-28.7-32-64s14.3-64 32-64s32 28.7 32 64s-14.3 64-32 64z\"],tooth:[448,512,[129463],\"f5c9\",\"M186.1 52.1C169.3 39.1 148.7 32 127.5 32C74.7 32 32 74.7 32 127.5v6.2c0 15.8 3.7 31.3 10.7 45.5l23.5 47.1c4.5 8.9 7.6 18.4 9.4 28.2l36.7 205.8c2 11.2 11.6 19.4 22.9 19.8s21.4-7.4 24-18.4l28.9-121.3C192.2 323.7 207 312 224 312s31.8 11.7 35.8 28.3l28.9 121.3c2.6 11.1 12.7 18.8 24 18.4s20.9-8.6 22.9-19.8l36.7-205.8c1.8-9.8 4.9-19.3 9.4-28.2l23.5-47.1c7.1-14.1 10.7-29.7 10.7-45.5v-2.1c0-55-44.6-99.6-99.6-99.6c-24.1 0-47.4 8.8-65.6 24.6l-3.2 2.8 19.5 15.2c7 5.4 8.2 15.5 2.8 22.5s-15.5 8.2-22.5 2.8l-24.4-19-37-28.8z\"],v:[384,512,[118],\"56\",\"M19.7 34.5c16.3-6.8 35 .9 41.8 17.2L192 364.8 322.5 51.7c6.8-16.3 25.5-24 41.8-17.2s24 25.5 17.2 41.8l-160 384c-5 11.9-16.6 19.7-29.5 19.7s-24.6-7.8-29.5-19.7L2.5 76.3c-6.8-16.3 .9-35 17.2-41.8z\"],\"bangladeshi-taka-sign\":[384,512,[],\"e2e6\",\"M36 32.2C18.4 30.1 2.4 42.5 .2 60S10.5 93.6 28 95.8l7.9 1c16 2 28 15.6 28 31.8V160H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64V384c0 53 43 96 96 96h32c106 0 192-86 192-192V256c0-53-43-96-96-96H272c-17.7 0-32 14.3-32 32s14.3 32 32 32h16c17.7 0 32 14.3 32 32v32c0 70.7-57.3 128-128 128H160c-17.7 0-32-14.3-32-32V224h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H128V128.5c0-48.4-36.1-89.3-84.1-95.3l-7.9-1z\"],bicycle:[640,512,[128690],\"f206\",\"M312 32c-13.3 0-24 10.7-24 24s10.7 24 24 24h25.7l34.6 64H222.9l-27.4-38C191 99.7 183.7 96 176 96H120c-13.3 0-24 10.7-24 24s10.7 24 24 24h43.7l22.1 30.7-26.6 53.1c-10-2.5-20.5-3.8-31.2-3.8C57.3 224 0 281.3 0 352s57.3 128 128 128c65.3 0 119.1-48.9 127-112h49c8.5 0 16.3-4.5 20.7-11.8l84.8-143.5 21.7 40.1C402.4 276.3 384 312 384 352c0 70.7 57.3 128 128 128s128-57.3 128-128s-57.3-128-128-128c-13.5 0-26.5 2.1-38.7 6L375.4 48.8C369.8 38.4 359 32 347.2 32H312zM458.6 303.7l32.3 59.7c6.3 11.7 20.9 16 32.5 9.7s16-20.9 9.7-32.5l-32.3-59.7c3.6-.6 7.4-.9 11.2-.9c39.8 0 72 32.2 72 72s-32.2 72-72 72s-72-32.2-72-72c0-18.6 7-35.5 18.6-48.3zM133.2 368h65c-7.3 32.1-36 56-70.2 56c-39.8 0-72-32.2-72-72s32.2-72 72-72c1.7 0 3.4 .1 5.1 .2l-24.2 48.5c-9 18.1 4.1 39.4 24.3 39.4zm33.7-48l50.7-101.3 72.9 101.2-.1 .1H166.8zm90.6-128H365.9L317 274.8 257.4 192z\"],\"staff-snake\":[384,512,[\"rod-asclepius\",\"rod-snake\",\"staff-aesculapius\"],\"e579\",\"M222.6 43.2l-.1 4.8H288c53 0 96 43 96 96s-43 96-96 96H248V160h40c8.8 0 16-7.2 16-16s-7.2-16-16-16H248 220l-4.5 144H256c53 0 96 43 96 96s-43 96-96 96H240V384h16c8.8 0 16-7.2 16-16s-7.2-16-16-16H213l-3.1 99.5L208.5 495l0 1c-.3 8.9-7.6 16-16.5 16s-16.2-7.1-16.5-16l0-1-1-31H136c-22.1 0-40-17.9-40-40s17.9-40 40-40h36l-1-32H152c-53 0-96-43-96-96c0-47.6 34.6-87.1 80-94.7V256c0 8.8 7.2 16 16 16h16.5L164 128H136 122.6c-9 18.9-28.3 32-50.6 32H56c-30.9 0-56-25.1-56-56S25.1 48 56 48h8 8 89.5l-.1-4.8L161 32c0-.7 0-1.3 0-1.9c.5-16.6 14.1-30 31-30s30.5 13.4 31 30c0 .6 0 1.3 0 1.9l-.4 11.2zM64 112a16 16 0 1 0 0-32 16 16 0 1 0 0 32z\"],\"head-side-cough-slash\":[640,512,[],\"e062\",\"M448 325.8l44 34.5c8.1 1.4 14.8 6.8 18 14.1L552.9 408c10.6 .4 19.5 7.6 22.2 17.4l39.1 30.6c.6 0 1.2-.1 1.8-.1c11.1 0 20.4 7.5 23.2 17.8h-3.9c6.2 8.5 6.4 20.4-.4 29c-8.2 10.4-23.3 12.3-33.7 4.1L9.2 42.9C-1.2 34.7-3.1 19.6 5.1 9.2S28.4-3.1 38.8 5.1L89.6 44.9C127 16.7 173.5 0 224 0h24c95.2 0 181.2 69.3 197.3 160.2c2.3 13 6.8 25.7 15.1 36l42 52.6c6.2 7.8 9.6 17.4 9.6 27.4c0 24.2-19.6 43.8-43.8 43.8H448v0 5.8zM0 224.2c0-38.7 9.8-75.1 27.1-106.9L341.8 365.3l-2.5 .3c-11 1.4-19.2 10.7-19.2 21.8c0 11.6 9 21.2 20.6 21.9l62 3.9 43 33.9C439.3 466.2 421.2 480 400 480H320v8c0 13.3-10.7 24-24 24H256v0H96c-17.7 0-32-14.3-32-32V407.3c0-16.7-6.9-32.5-17.1-45.8C16.6 322.4 0 274.1 0 224.2zM616 360a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm-64-48a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm40-24a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z\"],\"truck-medical\":[640,512,[128657,\"ambulance\"],\"f0f9\",\"M0 48C0 21.5 21.5 0 48 0H368c26.5 0 48 21.5 48 48V96h50.7c17 0 33.3 6.7 45.3 18.7L589.3 192c12 12 18.7 28.3 18.7 45.3V256v32 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H576c0 53-43 96-96 96s-96-43-96-96H256c0 53-43 96-96 96s-96-43-96-96H48c-26.5 0-48-21.5-48-48V48zM416 256H544V237.3L466.7 160H416v96zM160 464a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm368-48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM176 80v48l-48 0c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V192h48c8.8 0 16-7.2 16-16V144c0-8.8-7.2-16-16-16H240V80c0-8.8-7.2-16-16-16H192c-8.8 0-16 7.2-16 16z\"],\"wheat-awn-circle-exclamation\":[640,512,[],\"e598\",\"M505 41c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L383 95c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l88-88zM305.5 27.3c-6.2-6.2-16.4-6.2-22.6 0L271.5 38.6c-37.5 37.5-37.5 98.3 0 135.8l10.4 10.4-30.5 30.5c-3.4-27.3-15.5-53.8-36.5-74.8l-11.3-11.3c-6.2-6.2-16.4-6.2-22.6 0l-11.3 11.3c-37.5 37.5-37.5 98.3 0 135.8l10.4 10.4-30.5 30.5c-3.4-27.3-15.5-53.8-36.5-74.8L101.8 231c-6.2-6.2-16.4-6.2-22.6 0L67.9 242.3c-37.5 37.5-37.5 98.3 0 135.8l10.4 10.4L9.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l68.9-68.9 12.2 12.2c37.5 37.5 98.3 37.5 135.8 0l11.3-11.3c6.2-6.2 6.2-16.4 0-22.6l-11.3-11.3c-21.8-21.8-49.6-34.1-78.1-36.9l31.9-31.9 12.2 12.2c22.5 22.5 53.3 31.5 82.4 27c0-1 0-2.1 0-3.1c0-33.1 9.1-64.1 25-90.6c-15.5-8.7-32.5-13.8-49.8-15.5l31.9-31.9 12.2 12.2c6 6 12.6 11.1 19.7 15.2c27.5-34 67.3-57.5 112.6-63.8c-4.1-3.8-8.4-7.3-12.9-10.5L505 137c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-59.4 59.4c-20.6-4.4-42-3.7-62.3 2.1c6.1-21.3 6.6-43.8 1.4-65.3L409 41c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L329.1 52.9c-3.7-5-7.8-9.8-12.4-14.3L305.5 27.3zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z\"],snowman:[512,512,[9731,9924],\"f7d0\",\"M341.1 140.6c-2 3.9-1.6 8.6 1.2 12c7 8.5 12.9 18.1 17.2 28.4L408 160.2V120c0-13.3 10.7-24 24-24s24 10.7 24 24v19.6l22.5-9.7c12.2-5.2 26.3 .4 31.5 12.6s-.4 26.3-12.6 31.5l-56 24-73.6 31.5c-.5 9.5-2.1 18.6-4.8 27.3c-1.2 3.8-.1 8 2.8 10.8C396.7 296.9 416 338.2 416 384c0 44.7-18.3 85-47.8 114.1c-9.9 9.7-23.7 13.9-37.5 13.9H181.3c-13.9 0-27.7-4.2-37.5-13.9C114.3 469 96 428.7 96 384c0-45.8 19.3-87.1 50.1-116.3c2.9-2.8 4-6.9 2.8-10.8c-2.7-8.7-4.3-17.9-4.8-27.3L70.5 198.1l-56-24C2.4 168.8-3.3 154.7 1.9 142.5s19.3-17.8 31.5-12.6L56 139.6V120c0-13.3 10.7-24 24-24s24 10.7 24 24v40.2L152.6 181c4.3-10.3 10.1-19.9 17.2-28.4c2.8-3.4 3.3-8.1 1.2-12C164 127.2 160 112.1 160 96c0-53 43-96 96-96s96 43 96 96c0 16.1-4 31.2-10.9 44.6zM224 96a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm48 128a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm-16 80a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm16 48a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zM288 96a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm-48 24v3.2c0 3.2 .8 6.3 2.3 9l9 16.9c.9 1.7 2.7 2.8 4.7 2.8s3.8-1.1 4.7-2.8l9-16.9c1.5-2.8 2.3-5.9 2.3-9V120c0-8.8-7.2-16-16-16s-16 7.2-16 16z\"],\"mortar-pestle\":[512,512,[],\"f5a7\",\"M504.3 11.1C493.3-1.6 474.5-3.7 461 6.2L252.3 160H397.3L502.6 54.6c11.8-11.8 12.6-30.8 1.6-43.5zM32 192c-17.7 0-32 14.3-32 32s14.3 32 32 32c0 82.5 43.4 147.7 123.9 176.2c-11.1 13.9-19.4 30.3-23.9 48.1C127.6 497.4 142.3 512 160 512H352c17.7 0 32.4-14.6 28.1-31.7c-4.5-17.8-12.8-34.1-23.9-48.1C436.6 403.7 480 338.5 480 256c17.7 0 32-14.3 32-32s-14.3-32-32-32H32z\"],\"road-barrier\":[640,512,[],\"e562\",\"M32 32C14.3 32 0 46.3 0 64V448c0 17.7 14.3 32 32 32s32-14.3 32-32V266.3L149.2 96H64V64c0-17.7-14.3-32-32-32zM405.2 96H330.8l-5.4 10.7L234.8 288h74.3l5.4-10.7L405.2 96zM362.8 288h74.3l5.4-10.7L533.2 96H458.8l-5.4 10.7L362.8 288zM202.8 96l-5.4 10.7L106.8 288h74.3l5.4-10.7L277.2 96H202.8zm288 192H576V448c0 17.7 14.3 32 32 32s32-14.3 32-32V64c0-17.7-14.3-32-32-32s-32 14.3-32 32v53.7L490.8 288z\"],school:[640,512,[127979],\"f549\",\"M337.8 5.4C327-1.8 313-1.8 302.2 5.4L166.3 96H48C21.5 96 0 117.5 0 144V464c0 26.5 21.5 48 48 48H256V416c0-35.3 28.7-64 64-64s64 28.7 64 64v96H592c26.5 0 48-21.5 48-48V144c0-26.5-21.5-48-48-48H473.7L337.8 5.4zM96 192h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V208c0-8.8 7.2-16 16-16zm400 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H512c-8.8 0-16-7.2-16-16V208zM96 320h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zm400 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H512c-8.8 0-16-7.2-16-16V336zM232 176a88 88 0 1 1 176 0 88 88 0 1 1 -176 0zm88-48c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H336V144c0-8.8-7.2-16-16-16z\"],igloo:[576,512,[],\"f7ae\",\"M320 33.8V160H48.5C100.2 82.8 188.1 32 288 32c10.8 0 21.5 .6 32 1.8zM352 160V39.1C424.9 55.7 487.2 99.8 527.5 160H352zM29.9 192H96V320H0c0-46 10.8-89.4 29.9-128zM192 320H128V192H448V320H384v32H576v80c0 26.5-21.5 48-48 48H352V352c0-35.3-28.7-64-64-64s-64 28.7-64 64V480H48c-26.5 0-48-21.5-48-48V352H192V320zm288 0V192h66.1c19.2 38.6 29.9 82 29.9 128H480z\"],joint:[640,512,[],\"f595\",\"M448 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V43c0 55.2 21.9 108.1 60.9 147.1l21 21c9 9 14.1 21.2 14.1 33.9v11c0 17.7 14.3 32 32 32s32-14.3 32-32V245c0-29.7-11.8-58.2-32.8-79.2l-21-21C463.2 117.8 448 81.2 448 43V32zM576 256c0 17.7 14.3 32 32 32s32-14.3 32-32V245c0-55.2-21.9-108.1-60.9-147.1l-21-21c-9-9-14.1-21.2-14.1-33.9V32c0-17.7-14.3-32-32-32s-32 14.3-32 32V43c0 29.7 11.8 58.2 32.8 79.2l21 21c27 27 42.2 63.6 42.2 101.8v11zM229.8 360c-4.7-2.3-10-2.7-15.2-2c-37.8 5.6-75.2 14.3-106.9 22.8C81.3 388 58.3 395.1 42 400.4c-8.2 2.7-14.7 4.9-19.2 6.5c-2.3 .8-4 1.4-5.2 1.8l-1.3 .5C6.8 412.5 0 421.4 0 432s6.8 19.5 16.3 22.7l1.3 .5c1.2 .4 3 1.1 5.2 1.8c4.5 1.6 11 3.8 19.2 6.5c16.3 5.4 39.2 12.5 65.7 19.6C160.3 497.3 228.8 512 288 512h67.3c4.1 0 6.3-5.1 3.6-8.3L256.5 380.8c-7.4-8.9-16.5-15.9-26.7-20.8zM445 512h19 51.3c4.1 0 6.3-5.1 3.6-8.3L416.5 380.8C401.3 362.5 378.8 352 355 352H336 288c-1.1 0-2.3 0-3.4 0c-4.1 0-6.2 5.1-3.5 8.3L383.5 483.2C398.7 501.5 421.2 512 445 512zm-3.9-151.7L543.5 483.2c14.6 17.5 35.9 27.9 58.6 28.7c21.1-1.1 37.9-18.6 37.9-39.9V392c0-22.1-17.9-40-40-40H444.7c-4.1 0-6.3 5.1-3.6 8.3z\"],\"angle-right\":[320,512,[8250],\"f105\",\"M278.6 233.4c12.5 12.5 12.5 32.8 0 45.3l-160 160c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L210.7 256 73.4 118.6c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0l160 160z\"],horse:[576,512,[128014],\"f6f0\",\"M448 238.1V160h16l9.8 19.6c12.5 25.1 42.2 36.4 68.3 26c20.5-8.2 33.9-28 33.9-50.1V80c0-19.1-8.4-36.3-21.7-48H560c8.8 0 16-7.2 16-16s-7.2-16-16-16H480 448C377.3 0 320 57.3 320 128H224 203.2 148.8c-30.7 0-57.6 16.3-72.5 40.8C33.2 174.5 0 211.4 0 256v56c0 13.3 10.7 24 24 24s24-10.7 24-24V256c0-13.4 6.6-25.2 16.7-32.5c1.6 13 6.3 25.4 13.6 36.4l28.2 42.4c8.3 12.4 6.4 28.7-1.2 41.6c-16.5 28-20.6 62.2-10 93.9l17.5 52.4c4.4 13.1 16.6 21.9 30.4 21.9h33.7c21.8 0 37.3-21.4 30.4-42.1l-20.8-62.5c-2.1-6.4-.5-13.4 4.3-18.2l12.7-12.7c13.2-13.2 20.6-31.1 20.6-49.7c0-2.3-.1-4.6-.3-6.9l84 24c4.1 1.2 8.2 2.1 12.3 2.8V480c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V315.7c19.2-19.2 31.5-45.7 32-75.7h0v-1.9zM496 64a16 16 0 1 1 0 32 16 16 0 1 1 0-32z\"],q:[448,512,[113],\"51\",\"M64 256c0 88.4 71.6 160 160 160c28.9 0 56-7.7 79.4-21.1l-72-86.4c-11.3-13.6-9.5-33.8 4.1-45.1s33.8-9.5 45.1 4.1l70.9 85.1C371.9 325.8 384 292.3 384 256c0-88.4-71.6-160-160-160S64 167.6 64 256zM344.9 444.6C310 467 268.5 480 224 480C100.3 480 0 379.7 0 256S100.3 32 224 32s224 100.3 224 224c0 56.1-20.6 107.4-54.7 146.7l47.3 56.8c11.3 13.6 9.5 33.8-4.1 45.1s-33.8 9.5-45.1-4.1l-46.6-55.9z\"],g:[448,512,[103],\"47\",\"M224 96C135.6 96 64 167.6 64 256s71.6 160 160 160c77.4 0 142-55 156.8-128H256c-17.7 0-32-14.3-32-32s14.3-32 32-32H400c25.8 0 49.6 21.4 47.2 50.6C437.8 389.6 341.4 480 224 480C100.3 480 0 379.7 0 256S100.3 32 224 32c57.4 0 109.7 21.6 149.3 57c13.2 11.8 14.3 32 2.5 45.2s-32 14.3-45.2 2.5C302.3 111.4 265 96 224 96z\"],\"notes-medical\":[512,512,[],\"f481\",\"M96 352V96c0-35.3 28.7-64 64-64H416c35.3 0 64 28.7 64 64V293.5c0 17-6.7 33.3-18.7 45.3l-58.5 58.5c-12 12-28.3 18.7-45.3 18.7H160c-35.3 0-64-28.7-64-64zM272 128c-8.8 0-16 7.2-16 16v48H208c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V256h48c8.8 0 16-7.2 16-16V208c0-8.8-7.2-16-16-16H320V144c0-8.8-7.2-16-16-16H272zm24 336c13.3 0 24 10.7 24 24s-10.7 24-24 24H136C60.9 512 0 451.1 0 376V152c0-13.3 10.7-24 24-24s24 10.7 24 24l0 224c0 48.6 39.4 88 88 88H296z\"],\"temperature-half\":[320,512,[127777,\"temperature-2\",\"thermometer-2\",\"thermometer-half\"],\"f2c9\",\"M160 64c-26.5 0-48 21.5-48 48V276.5c0 17.3-7.1 31.9-15.3 42.5C86.2 332.6 80 349.5 80 368c0 44.2 35.8 80 80 80s80-35.8 80-80c0-18.5-6.2-35.4-16.7-48.9c-8.2-10.6-15.3-25.2-15.3-42.5V112c0-26.5-21.5-48-48-48zM48 112C48 50.2 98.1 0 160 0s112 50.1 112 112V276.5c0 .1 .1 .3 .2 .6c.2 .6 .8 1.6 1.7 2.8c18.9 24.4 30.1 55 30.1 88.1c0 79.5-64.5 144-144 144S16 447.5 16 368c0-33.2 11.2-63.8 30.1-88.1c.9-1.2 1.5-2.2 1.7-2.8c.1-.3 .2-.5 .2-.6V112zM208 368c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-20.9 13.4-38.7 32-45.3V208c0-8.8 7.2-16 16-16s16 7.2 16 16V322.7c18.6 6.6 32 24.4 32 45.3z\"],\"dong-sign\":[384,512,[],\"e169\",\"M288 32c-17.7 0-32 14.3-32 32l-32 0c-17.7 0-32 14.3-32 32s14.3 32 32 32h32v49.1c-18.8-10.9-40.7-17.1-64-17.1c-70.7 0-128 57.3-128 128s57.3 128 128 128c24.5 0 47.4-6.9 66.8-18.8c5 11.1 16.2 18.8 29.2 18.8c17.7 0 32-14.3 32-32V288 128c17.7 0 32-14.3 32-32s-14.3-32-32-32c0-17.7-14.3-32-32-32zM128 288a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM32 448c-17.7 0-32 14.3-32 32s14.3 32 32 32H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H32z\"],capsules:[576,512,[],\"f46b\",\"M64 144c0-26.5 21.5-48 48-48s48 21.5 48 48V256H64V144zM0 144V368c0 61.9 50.1 112 112 112s112-50.1 112-112V189.6c1.8 19.1 8.2 38 19.8 54.8L372.3 431.7c35.5 51.7 105.3 64.3 156 28.1s63-107.5 27.5-159.2L427.3 113.3C391.8 61.5 321.9 49 271.3 85.2c-28 20-44.3 50.8-47.3 83V144c0-61.9-50.1-112-112-112S0 82.1 0 144zm296.6 64.2c-16-23.3-10-55.3 11.9-71c21.2-15.1 50.5-10.3 66 12.2l67 97.6L361.6 303l-65-94.8zM491 407.7c-.8 .6-1.6 1.1-2.4 1.6l4-2.8c-.5 .4-1 .8-1.6 1.2z\"],\"poo-storm\":[448,512,[\"poo-bolt\"],\"f75a\",\"M236.9 .2c-5.5-.7-11 1.4-14.5 5.7s-4.6 10.1-2.8 15.3c2.8 8.2 4.3 16.9 4.3 26.1c0 21.7-8.5 37.2-21.9 47.6c-13.8 10.8-34 17-57.8 17H128c-35.3 0-64 28.7-64 64c0 12.2 3.4 23.5 9.3 33.2C31.7 216.2 0 252.4 0 296c0 40.9 28 75.4 65.8 85.2c-5.3-18.5 1-38.5 16.2-50.7l160-128c17.6-14.1 42.6-14 60.2 .2s22.8 38.6 12.8 58.8L285.7 320H304c20.4 0 38.5 12.9 45.3 32.1c3.7 10.6 3.5 21.8 0 31.9H360c48.6 0 88-39.4 88-88c0-43.6-31.7-79.8-73.3-86.8c5.9-9.7 9.3-21.1 9.3-33.2c0-35.3-28.7-64-64-64h-1.4c.9-5.4 1.4-10.9 1.4-16.6c0-48.7-36.1-88.9-83.1-95.2zm45.1 227.4c-5.8-4.7-14.2-4.7-20.1-.1l-160 128c-5.3 4.2-7.4 11.4-5.1 17.8s8.3 10.7 15.1 10.7h70.1L129.7 488.8c-3.4 6.7-1.6 14.9 4.3 19.6s14.2 4.7 20.1 .1l160-128c5.3-4.2 7.4-11.4 5.1-17.8s-8.3-10.7-15.1-10.7H233.9l52.4-104.8c3.4-6.7 1.6-14.9-4.3-19.6z\"],\"face-frown-open\":[512,512,[128550,\"frown-open\"],\"f57a\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM176.4 176a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm-122 174.5c-12.4 5.2-26.5-4.1-21.1-16.4c16-36.6 52.4-62.1 94.8-62.1s78.8 25.6 94.8 62.1c5.4 12.3-8.7 21.6-21.1 16.4c-22.4-9.5-47.4-14.8-73.7-14.8s-51.3 5.3-73.7 14.8z\"],\"hand-point-up\":[384,512,[9757],\"f0a6\",\"M32 32C32 14.3 46.3 0 64 0S96 14.3 96 32V240H32V32zM224 192c0-17.7 14.3-32 32-32s32 14.3 32 32v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V192zm-64-64c17.7 0 32 14.3 32 32v48c0 17.7-14.3 32-32 32s-32-14.3-32-32V160c0-17.7 14.3-32 32-32zm160 96c0-17.7 14.3-32 32-32s32 14.3 32 32v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V224zm-96 88l0-.6c9.4 5.4 20.3 8.6 32 8.6c13.2 0 25.4-4 35.6-10.8c8.7 24.9 32.5 42.8 60.4 42.8c11.7 0 22.6-3.1 32-8.6V352c0 88.4-71.6 160-160 160H162.3c-42.4 0-83.1-16.9-113.1-46.9L37.5 453.5C13.5 429.5 0 396.9 0 363V336c0-35.3 28.7-64 64-64h88c22.1 0 40 17.9 40 40s-17.9 40-40 40H96c-8.8 0-16 7.2-16 16s7.2 16 16 16h56c39.8 0 72-32.2 72-72z\"],\"money-bill\":[576,512,[],\"f0d6\",\"M64 64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H64zm64 320H64V320c35.3 0 64 28.7 64 64zM64 192V128h64c0 35.3-28.7 64-64 64zM448 384c0-35.3 28.7-64 64-64v64H448zm64-192c-35.3 0-64-28.7-64-64h64v64zM288 160a96 96 0 1 1 0 192 96 96 0 1 1 0-192z\"],bookmark:[384,512,[128278,61591],\"f02e\",\"M0 48V487.7C0 501.1 10.9 512 24.3 512c5 0 9.9-1.5 14-4.4L192 400 345.7 507.6c4.1 2.9 9 4.4 14 4.4c13.4 0 24.3-10.9 24.3-24.3V48c0-26.5-21.5-48-48-48H48C21.5 0 0 21.5 0 48z\"],\"align-justify\":[448,512,[],\"f039\",\"M448 64c0-17.7-14.3-32-32-32H32C14.3 32 0 46.3 0 64S14.3 96 32 96H416c17.7 0 32-14.3 32-32zm0 256c0-17.7-14.3-32-32-32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32zM0 192c0 17.7 14.3 32 32 32H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H32c-17.7 0-32 14.3-32 32zM448 448c0-17.7-14.3-32-32-32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32z\"],\"umbrella-beach\":[576,512,[127958],\"f5ca\",\"M346.3 271.8l-60.1-21.9L214 448H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H282.1l64.1-176.2zm121.1-.2l-3.3 9.1 67.7 24.6c18.1 6.6 38-4.2 39.6-23.4c6.5-78.5-23.9-155.5-80.8-208.5c2 8 3.2 16.3 3.4 24.8l.2 6c1.8 57-7.3 113.8-26.8 167.4zM462 99.1c-1.1-34.4-22.5-64.8-54.4-77.4c-.9-.4-1.9-.7-2.8-1.1c-33-11.7-69.8-2.4-93.1 23.8l-4 4.5C272.4 88.3 245 134.2 226.8 184l-3.3 9.1L434 269.7l3.3-9.1c18.1-49.8 26.6-102.5 24.9-155.5l-.2-6zM107.2 112.9c-11.1 15.7-2.8 36.8 15.3 43.4l71 25.8 3.3-9.1c19.5-53.6 49.1-103 87.1-145.5l4-4.5c6.2-6.9 13.1-13 20.5-18.2c-79.6 2.5-154.7 42.2-201.2 108z\"],\"helmet-un\":[512,512,[],\"e503\",\"M479.5 224C471.2 98.9 367.2 0 240 0C107.5 0 0 107.5 0 240v56.3C0 344.8 39.2 384 87.7 384H200h14.9L343.5 505.4c4.5 4.2 10.4 6.6 16.5 6.6h96c13.3 0 24-10.7 24-24s-10.7-24-24-24H369.5l-1.5-1.5V288h80 32c17.7 0 32-14.3 32-32s-14.3-32-32-32h-.5zM320 417.2l-78-73.7L274.4 288H320V417.2zM285.3 103.1l34.7 52V112c0-8.8 7.2-16 16-16s16 7.2 16 16v96c0 7.1-4.6 13.3-11.4 15.3s-14-.6-17.9-6.4l-34.7-52V208c0 8.8-7.2 16-16 16s-16-7.2-16-16V112c0-7.1 4.6-13.3 11.4-15.3s14 .6 17.9 6.4zM160 112v64c0 8.8 7.2 16 16 16s16-7.2 16-16V112c0-8.8 7.2-16 16-16s16 7.2 16 16v64c0 26.5-21.5 48-48 48s-48-21.5-48-48V112c0-8.8 7.2-16 16-16s16 7.2 16 16z\"],bullseye:[512,512,[],\"f140\",\"M448 256A192 192 0 1 0 64 256a192 192 0 1 0 384 0zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm256 80a80 80 0 1 0 0-160 80 80 0 1 0 0 160zm0-224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zM224 256a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],bacon:[576,512,[129363],\"f7e5\",\"M439.2 1.2c11.2-3.2 23.2-.1 31.4 8.1L518 56.7l-26.5 7.9c-58 16.6-98.1 39.6-129.6 67.4c-31.2 27.5-53.2 59.1-75.1 90.9l-2.3 3.3C241.6 288.7 195 356.6 72.8 417.7L37.9 435.2 9.4 406.6c-7.3-7.3-10.6-17.6-9-27.8s8.1-18.9 17.3-23.5C136.1 296.2 180.9 231 223.3 169.3l2.3-3.4c21.8-31.8 44.9-64.9 77.7-93.9c33.4-29.5 75.8-53.6 135.9-70.8zM61.8 459l25.4-12.7c129.5-64.7 179.9-138.1 223.8-202l2.2-3.3c22.1-32.1 42.1-60.5 69.9-85.1c27.5-24.3 63.4-45.2 117.3-60.6l0 0 .2-.1 43.1-12.9 23 23c8 8 11.2 19.7 8.3 30.7s-11.3 19.6-22.2 22.7c-51.9 14.8-85.6 34.7-111.1 57.2c-26.1 23-45.1 49.9-67.3 82.1l-2.2 3.2C327.8 365.9 275.5 442 142.3 508.6c-12.3 6.2-27.2 3.7-36.9-6L61.8 459z\"],\"hand-point-down\":[384,512,[],\"f0a7\",\"M32 480c0 17.7 14.3 32 32 32s32-14.3 32-32V272H32V480zM224 320c0 17.7 14.3 32 32 32s32-14.3 32-32V256c0-17.7-14.3-32-32-32s-32 14.3-32 32v64zm-64 64c17.7 0 32-14.3 32-32V304c0-17.7-14.3-32-32-32s-32 14.3-32 32v48c0 17.7 14.3 32 32 32zm160-96c0 17.7 14.3 32 32 32s32-14.3 32-32V224c0-17.7-14.3-32-32-32s-32 14.3-32 32v64zm-96-88l0 .6c9.4-5.4 20.3-8.6 32-8.6c13.2 0 25.4 4 35.6 10.8c8.7-24.9 32.5-42.8 60.4-42.8c11.7 0 22.6 3.1 32 8.6V160C384 71.6 312.4 0 224 0H162.3C119.8 0 79.1 16.9 49.1 46.9L37.5 58.5C13.5 82.5 0 115.1 0 149v27c0 35.3 28.7 64 64 64h88c22.1 0 40-17.9 40-40s-17.9-40-40-40H96c-8.8 0-16-7.2-16-16s7.2-16 16-16h56c39.8 0 72 32.2 72 72z\"],\"arrow-up-from-bracket\":[448,512,[],\"e09a\",\"M246.6 9.4c-12.5-12.5-32.8-12.5-45.3 0l-128 128c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L192 109.3 192 320c0 17.7 14.3 32 32 32s32-14.3 32-32l0-210.7 73.4 73.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-128-128zM64 352c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 64c0 53 43 96 96 96l256 0c53 0 96-43 96-96l0-64c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 64c0 17.7-14.3 32-32 32L96 448c-17.7 0-32-14.3-32-32l0-64z\"],folder:[512,512,[128193,128447,61716,\"folder-blank\"],\"f07b\",\"M64 480H448c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H288c-10.1 0-19.6-4.7-25.6-12.8L243.2 57.6C231.1 41.5 212.1 32 192 32H64C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64z\"],\"file-waveform\":[448,512,[\"file-medical-alt\"],\"f478\",\"M96 0C60.7 0 32 28.7 32 64V288H144c6.1 0 11.6 3.4 14.3 8.8L176 332.2l49.7-99.4c2.7-5.4 8.3-8.8 14.3-8.8s11.6 3.4 14.3 8.8L281.9 288H352c8.8 0 16 7.2 16 16s-7.2 16-16 16H272c-6.1 0-11.6-3.4-14.3-8.8L240 275.8l-49.7 99.4c-2.7 5.4-8.3 8.8-14.3 8.8s-11.6-3.4-14.3-8.8L134.1 320H32V448c0 35.3 28.7 64 64 64H352c35.3 0 64-28.7 64-64V160H288c-17.7 0-32-14.3-32-32V0H96zM288 0V128H416L288 0z\"],radiation:[512,512,[],\"f7b9\",\"M216 186.7c-23.9 13.8-40 39.7-40 69.3L32 256C14.3 256-.2 241.6 2 224.1C10.7 154 47.8 92.7 101.3 52c14.1-10.7 33.8-5.3 42.7 10l72 124.7zM256 336c14.6 0 28.2-3.9 40-10.7l72 124.8c8.8 15.3 3.7 35.1-12.6 41.9c-30.6 12.9-64.2 20-99.4 20s-68.9-7.1-99.4-20c-16.3-6.9-21.4-26.6-12.6-41.9l72-124.8c11.8 6.8 25.4 10.7 40 10.7zm224-80l-144 0c0-29.6-16.1-55.5-40-69.3L368 62c8.8-15.3 28.6-20.7 42.7-10c53.6 40.7 90.6 102 99.4 172.1c2.2 17.5-12.4 31.9-30 31.9zM256 208a48 48 0 1 1 0 96 48 48 0 1 1 0-96z\"],\"chart-simple\":[448,512,[],\"e473\",\"M160 80c0-26.5 21.5-48 48-48h32c26.5 0 48 21.5 48 48V432c0 26.5-21.5 48-48 48H208c-26.5 0-48-21.5-48-48V80zM0 272c0-26.5 21.5-48 48-48H80c26.5 0 48 21.5 48 48V432c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V272zM368 96h32c26.5 0 48 21.5 48 48V432c0 26.5-21.5 48-48 48H368c-26.5 0-48-21.5-48-48V144c0-26.5 21.5-48 48-48z\"],\"mars-stroke\":[512,512,[9894],\"f229\",\"M376 0c-9.7 0-18.5 5.8-22.2 14.8s-1.7 19.3 5.2 26.2l33.4 33.4L370.3 96.4 345 71c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l25.4 25.4L307.8 159c-28.4-19.5-62.7-31-99.8-31c-97.2 0-176 78.8-176 176s78.8 176 176 176s176-78.8 176-176c0-37-11.4-71.4-31-99.8l28.6-28.6L407 201c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-25.4-25.4 22.1-22.1L471 153c6.9 6.9 17.2 8.9 26.2 5.2s14.8-12.5 14.8-22.2V24c0-13.3-10.7-24-24-24H376zm88 48h0v0l0 0zM96 304a112 112 0 1 1 224 0A112 112 0 1 1 96 304z\"],vial:[512,512,[129514],\"f492\",\"M342.6 9.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l9.4 9.4L28.1 342.6C10.1 360.6 0 385 0 410.5V416c0 53 43 96 96 96h5.5c25.5 0 49.9-10.1 67.9-28.1L448 205.3l9.4 9.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-32-32-96-96-32-32zM205.3 256L352 109.3 402.7 160l-96 96H205.3z\"],gauge:[512,512,[\"dashboard\",\"gauge-med\",\"tachometer-alt-average\"],\"f624\",\"M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm320 96c0-26.9-16.5-49.9-40-59.3V88c0-13.3-10.7-24-24-24s-24 10.7-24 24V292.7c-23.5 9.5-40 32.5-40 59.3c0 35.3 28.7 64 64 64s64-28.7 64-64zM144 176a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm-16 80a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm288 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM400 144a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z\"],\"wand-magic-sparkles\":[576,512,[\"magic-wand-sparkles\"],\"e2ca\",\"M234.7 42.7L197 56.8c-3 1.1-5 4-5 7.2s2 6.1 5 7.2l37.7 14.1L248.8 123c1.1 3 4 5 7.2 5s6.1-2 7.2-5l14.1-37.7L315 71.2c3-1.1 5-4 5-7.2s-2-6.1-5-7.2L277.3 42.7 263.2 5c-1.1-3-4-5-7.2-5s-6.1 2-7.2 5L234.7 42.7zM46.1 395.4c-18.7 18.7-18.7 49.1 0 67.9l34.6 34.6c18.7 18.7 49.1 18.7 67.9 0L529.9 116.5c18.7-18.7 18.7-49.1 0-67.9L495.3 14.1c-18.7-18.7-49.1-18.7-67.9 0L46.1 395.4zM484.6 82.6l-105 105-23.3-23.3 105-105 23.3 23.3zM7.5 117.2C3 118.9 0 123.2 0 128s3 9.1 7.5 10.8L64 160l21.2 56.5c1.7 4.5 6 7.5 10.8 7.5s9.1-3 10.8-7.5L128 160l56.5-21.2c4.5-1.7 7.5-6 7.5-10.8s-3-9.1-7.5-10.8L128 96 106.8 39.5C105.1 35 100.8 32 96 32s-9.1 3-10.8 7.5L64 96 7.5 117.2zm352 256c-4.5 1.7-7.5 6-7.5 10.8s3 9.1 7.5 10.8L416 416l21.2 56.5c1.7 4.5 6 7.5 10.8 7.5s9.1-3 10.8-7.5L480 416l56.5-21.2c4.5-1.7 7.5-6 7.5-10.8s-3-9.1-7.5-10.8L480 352l-21.2-56.5c-1.7-4.5-6-7.5-10.8-7.5s-9.1 3-10.8 7.5L416 352l-56.5 21.2z\"],e:[320,512,[101],\"45\",\"M64 32C28.7 32 0 60.7 0 96V256 416c0 35.3 28.7 64 64 64H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H64V288H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H64V96H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H64z\"],\"pen-clip\":[512,512,[\"pen-alt\"],\"f305\",\"M453.3 19.3l39.4 39.4c25 25 25 65.5 0 90.5l-52.1 52.1 0 0-1-1 0 0-16-16-96-96-17-17 52.1-52.1c25-25 65.5-25 90.5 0zM241 114.9c-9.4-9.4-24.6-9.4-33.9 0L105 217c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9L173.1 81c28.1-28.1 73.7-28.1 101.8 0L288 94.1l17 17 96 96 16 16 1 1-17 17L229.5 412.5c-48 48-109.2 80.8-175.8 94.1l-25 5c-7.9 1.6-16-.9-21.7-6.6s-8.1-13.8-6.6-21.7l5-25c13.3-66.6 46.1-127.8 94.1-175.8L254.1 128 241 114.9z\"],\"bridge-circle-exclamation\":[640,512,[],\"e4ca\",\"M64 32C46.3 32 32 46.3 32 64s14.3 32 32 32h40v64H32V288c53 0 96 43 96 96v64c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V384c0-53 43-96 96-96c6.3 0 12.4 .6 18.3 1.7C367.1 231.8 426.9 192 496 192c42.5 0 81.6 15.1 112 40.2V160H536V96h40c17.7 0 32-14.3 32-32s-14.3-32-32-32H64zM488 96v64H408V96h80zM360 96v64H280V96h80zM232 96v64H152V96h80zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z\"],user:[448,512,[128100,62144],\"f007\",\"M224 256A128 128 0 1 0 224 0a128 128 0 1 0 0 256zm-45.7 48C79.8 304 0 383.8 0 482.3C0 498.7 13.3 512 29.7 512H418.3c16.4 0 29.7-13.3 29.7-29.7C448 383.8 368.2 304 269.7 304H178.3z\"],\"school-circle-check\":[640,512,[],\"e56b\",\"M337.8 5.4C327-1.8 313-1.8 302.2 5.4L166.3 96H48C21.5 96 0 117.5 0 144V464c0 26.5 21.5 48 48 48H320v0H256V416c0-35.3 28.7-64 64-64l.3 0h.5c3.4-37.7 18.7-72.1 42.2-99.1C350.2 260 335.6 264 320 264c-48.6 0-88-39.4-88-88s39.4-88 88-88s88 39.4 88 88c0 18.3-5.6 35.3-15.1 49.4c29-21 64.6-33.4 103.1-33.4c59.5 0 112.1 29.6 144 74.8V144c0-26.5-21.5-48-48-48H473.7L337.8 5.4zM96 192h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V208c0-8.8 7.2-16 16-16zm0 128h32c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zM320 128c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H336V144c0-8.8-7.2-16-16-16zM640 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-99.3-43.3c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 385.4l60.7-60.7z\"],dumpster:[576,512,[],\"f793\",\"M49.7 32c-10.5 0-19.8 6.9-22.9 16.9L.9 133c-.6 2-.9 4.1-.9 6.1C0 150.7 9.3 160 20.9 160h94L140.5 32H49.7zM272 160V32H173.1L147.5 160H272zm32 0H428.5L402.9 32H304V160zm157.1 0h94c11.5 0 20.9-9.3 20.9-20.9c0-2.1-.3-4.1-.9-6.1L549.2 48.9C546.1 38.9 536.8 32 526.3 32H435.5l25.6 128zM32 192l4 32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H44L64 448c0 17.7 14.3 32 32 32s32-14.3 32-32H448c0 17.7 14.3 32 32 32s32-14.3 32-32l20-160h12c17.7 0 32-14.3 32-32s-14.3-32-32-32h-4l4-32H32z\"],\"van-shuttle\":[640,512,[128656,\"shuttle-van\"],\"f5b6\",\"M64 104v88h96V96H72c-4.4 0-8 3.6-8 8zm482 88L465.1 96H384v96H546zm-226 0V96H224v96h96zM592 384H576c0 53-43 96-96 96s-96-43-96-96H256c0 53-43 96-96 96s-96-43-96-96H48c-26.5 0-48-21.5-48-48V104C0 64.2 32.2 32 72 32H192 352 465.1c18.9 0 36.8 8.3 49 22.8L625 186.5c9.7 11.5 15 26.1 15 41.2V336c0 26.5-21.5 48-48 48zm-64 0a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM160 432a48 48 0 1 0 0-96 48 48 0 1 0 0 96z\"],\"building-user\":[640,512,[],\"e4da\",\"M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h89.9c-6.3-10.2-9.9-22.2-9.9-35.1c0-46.9 25.8-87.8 64-109.2V271.8 48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zM576 272a80 80 0 1 0 -160 0 80 80 0 1 0 160 0zM352 477.1c0 19.3 15.6 34.9 34.9 34.9H605.1c19.3 0 34.9-15.6 34.9-34.9c0-51.4-41.7-93.1-93.1-93.1H445.1c-51.4 0-93.1 41.7-93.1 93.1z\"],\"square-caret-left\":[448,512,[\"caret-square-left\"],\"f191\",\"M0 416c0 35.3 28.7 64 64 64l320 0c35.3 0 64-28.7 64-64l0-320c0-35.3-28.7-64-64-64L64 32C28.7 32 0 60.7 0 96L0 416zM128 256c0-6.7 2.8-13 7.7-17.6l112-104c7-6.5 17.2-8.2 25.9-4.4s14.4 12.5 14.4 22l0 208c0 9.5-5.7 18.2-14.4 22s-18.9 2.1-25.9-4.4l-112-104c-4.9-4.5-7.7-10.9-7.7-17.6z\"],highlighter:[576,512,[],\"f591\",\"M315 315l158.4-215L444.1 70.6 229 229 315 315zm-187 5l0 0V248.3c0-15.3 7.2-29.6 19.5-38.6L420.6 8.4C428 2.9 437 0 446.2 0c11.4 0 22.4 4.5 30.5 12.6l54.8 54.8c8.1 8.1 12.6 19 12.6 30.5c0 9.2-2.9 18.2-8.4 25.6L334.4 396.5c-9 12.3-23.4 19.5-38.6 19.5H224l-25.4 25.4c-12.5 12.5-32.8 12.5-45.3 0l-50.7-50.7c-12.5-12.5-12.5-32.8 0-45.3L128 320zM7 466.3l63-63 70.6 70.6-31 31c-4.5 4.5-10.6 7-17 7H24c-13.3 0-24-10.7-24-24v-4.7c0-6.4 2.5-12.5 7-17z\"],key:[512,512,[128273],\"f084\",\"M336 352c97.2 0 176-78.8 176-176S433.2 0 336 0S160 78.8 160 176c0 18.7 2.9 36.8 8.3 53.7L7 391c-4.5 4.5-7 10.6-7 17v80c0 13.3 10.7 24 24 24h80c13.3 0 24-10.7 24-24V448h40c13.3 0 24-10.7 24-24V384h40c6.4 0 12.5-2.5 17-7l33.3-33.3c16.9 5.4 35 8.3 53.7 8.3zM376 96a40 40 0 1 1 0 80 40 40 0 1 1 0-80z\"],bullhorn:[512,512,[128226,128363],\"f0a1\",\"M480 32c0-12.9-7.8-24.6-19.8-29.6s-25.7-2.2-34.9 6.9L381.7 53c-48 48-113.1 75-181 75H192 160 64c-35.3 0-64 28.7-64 64v96c0 35.3 28.7 64 64 64l0 128c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V352l8.7 0c67.9 0 133 27 181 75l43.6 43.6c9.2 9.2 22.9 11.9 34.9 6.9s19.8-16.6 19.8-29.6V300.4c18.6-8.8 32-32.5 32-60.4s-13.4-51.6-32-60.4V32zm-64 76.7V240 371.3C357.2 317.8 280.5 288 200.7 288H192V192h8.7c79.8 0 156.5-29.8 215.3-83.3z\"],globe:[512,512,[127760],\"f0ac\",\"M352 256c0 22.2-1.2 43.6-3.3 64H163.3c-2.2-20.4-3.3-41.8-3.3-64s1.2-43.6 3.3-64H348.7c2.2 20.4 3.3 41.8 3.3 64zm28.8-64H503.9c5.3 20.5 8.1 41.9 8.1 64s-2.8 43.5-8.1 64H380.8c2.1-20.6 3.2-42 3.2-64s-1.1-43.4-3.2-64zm112.6-32H376.7c-10-63.9-29.8-117.4-55.3-151.6c78.3 20.7 142 77.5 171.9 151.6zm-149.1 0H167.7c6.1-36.4 15.5-68.6 27-94.7c10.5-23.6 22.2-40.7 33.5-51.5C239.4 3.2 248.7 0 256 0s16.6 3.2 27.8 13.8c11.3 10.8 23 27.9 33.5 51.5c11.6 26 20.9 58.2 27 94.7zm-209 0H18.6C48.6 85.9 112.2 29.1 190.6 8.4C165.1 42.6 145.3 96.1 135.3 160zM8.1 192H131.2c-2.1 20.6-3.2 42-3.2 64s1.1 43.4 3.2 64H8.1C2.8 299.5 0 278.1 0 256s2.8-43.5 8.1-64zM194.7 446.6c-11.6-26-20.9-58.2-27-94.6H344.3c-6.1 36.4-15.5 68.6-27 94.6c-10.5 23.6-22.2 40.7-33.5 51.5C272.6 508.8 263.3 512 256 512s-16.6-3.2-27.8-13.8c-11.3-10.8-23-27.9-33.5-51.5zM135.3 352c10 63.9 29.8 117.4 55.3 151.6C112.2 482.9 48.6 426.1 18.6 352H135.3zm358.1 0c-30 74.1-93.6 130.9-171.9 151.6c25.5-34.2 45.2-87.7 55.3-151.6H493.4z\"],synagogue:[640,512,[128333],\"f69b\",\"M309.8 3.7c5.9-4.9 14.6-4.9 20.5 0l121 100.8C469.5 119.7 480 142.2 480 166V280.1 512H464 352V416c0-17.7-14.3-32-32-32s-32 14.3-32 32v96H176 160V280.1 166c0-23.7 10.5-46.3 28.8-61.5L309.8 3.7zM512 512V244.5l28.1-31.2c3-3.4 7.4-5.3 11.9-5.3s8.9 1.9 11.9 5.3l63.8 70.9c7.9 8.8 12.3 20.3 12.3 32.1V448c0 35.3-28.7 64-64 64H512zM128 244.5V512H64c-35.3 0-64-28.7-64-64V316.3c0-11.9 4.4-23.3 12.3-32.1l63.8-70.9c3-3.4 7.4-5.3 11.9-5.3s8.9 1.9 11.9 5.3L128 244.5zM327 124.3c-3.1-5.4-10.9-5.4-13.9 0l-15.9 28.1-32.3-.3c-6.2-.1-10.1 6.7-7 12.1L274.3 192l-16.4 27.8c-3.2 5.4 .7 12.1 7 12.1l32.3-.3L313 259.7c3.1 5.4 10.9 5.4 13.9 0l15.9-28.1 32.3 .3c6.2 .1 10.1-6.7 7-12.1L365.7 192l16.4-27.8c3.2-5.4-.7-12.1-7-12.1l-32.3 .3L327 124.3z\"],\"person-half-dress\":[320,512,[],\"e548\",\"M160 0a48 48 0 1 1 0 96 48 48 0 1 1 0-96zm8 352V128h6.9c33.7 0 64.9 17.7 82.3 46.6l58.3 97c9.1 15.1 4.2 34.8-10.9 43.9s-34.8 4.2-43.9-10.9L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352h0zM58.2 182.3c19.9-33.1 55.3-53.5 93.8-54.3V384h0v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384H70.2c-10.9 0-18.6-10.7-15.2-21.1L93.3 248.1 59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l53.6-89.2z\"],\"road-bridge\":[640,512,[],\"e563\",\"M352 0H608c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H352c-17.7 0-32-14.3-32-32V32c0-17.7 14.3-32 32-32zM480 200c-13.3 0-24 10.7-24 24v64c0 13.3 10.7 24 24 24s24-10.7 24-24V224c0-13.3-10.7-24-24-24zm24 184c0-13.3-10.7-24-24-24s-24 10.7-24 24v64c0 13.3 10.7 24 24 24s24-10.7 24-24V384zM480 40c-13.3 0-24 10.7-24 24v64c0 13.3 10.7 24 24 24s24-10.7 24-24V64c0-13.3-10.7-24-24-24zM32 96H288v64H248v64h40v96c-53 0-96 43-96 96v64c0 17.7-14.3 32-32 32H128c-17.7 0-32-14.3-32-32V416c0-53-43-96-96-96V224H72V160H32c-17.7 0-32-14.3-32-32s14.3-32 32-32zm168 64H120v64h80V160z\"],\"location-arrow\":[448,512,[],\"f124\",\"M429.6 92.1c4.9-11.9 2.1-25.6-7-34.7s-22.8-11.9-34.7-7l-352 144c-14.2 5.8-22.2 20.8-19.3 35.8s16.1 25.8 31.4 25.8H224V432c0 15.3 10.8 28.4 25.8 31.4s30-5.1 35.8-19.3l144-352z\"],c:[384,512,[99],\"43\",\"M329.1 142.9c-62.5-62.5-155.8-62.5-218.3 0s-62.5 163.8 0 226.3s155.8 62.5 218.3 0c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3c-87.5 87.5-221.3 87.5-308.8 0s-87.5-229.3 0-316.8s221.3-87.5 308.8 0c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0z\"],\"tablet-button\":[448,512,[],\"f10a\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zM224 400a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"building-lock\":[576,512,[],\"e4d6\",\"M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h88.6c-5.4-9.4-8.6-20.3-8.6-32V352c0-23.7 12.9-44.4 32-55.4V272c0-30.5 12.2-58.2 32-78.4V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zM464 240c17.7 0 32 14.3 32 32v48H432V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H544c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z\"],\"pizza-slice\":[512,512,[],\"f818\",\"M169.7 .9c-22.8-1.6-41.9 14-47.5 34.7L110.4 80c.5 0 1.1 0 1.6 0c176.7 0 320 143.3 320 320c0 .5 0 1.1 0 1.6l44.4-11.8c20.8-5.5 36.3-24.7 34.7-47.5C498.5 159.5 352.5 13.5 169.7 .9zM399.8 410.2c.1-3.4 .2-6.8 .2-10.2c0-159.1-128.9-288-288-288c-3.4 0-6.8 .1-10.2 .2L.5 491.9c-1.5 5.5 .1 11.4 4.1 15.4s9.9 5.6 15.4 4.1L399.8 410.2zM176 208a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm64 128a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM96 384a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],\"money-bill-wave\":[576,512,[],\"f53a\",\"M0 112.5V422.3c0 18 10.1 35 27 41.3c87 32.5 174 10.3 261-11.9c79.8-20.3 159.6-40.7 239.3-18.9c23 6.3 48.7-9.5 48.7-33.4V89.7c0-18-10.1-35-27-41.3C462 15.9 375 38.1 288 60.3C208.2 80.6 128.4 100.9 48.7 79.1C25.6 72.8 0 88.6 0 112.5zM288 352c-44.2 0-80-43-80-96s35.8-96 80-96s80 43 80 96s-35.8 96-80 96zM64 352c35.3 0 64 28.7 64 64H64V352zm64-208c0 35.3-28.7 64-64 64V144h64zM512 304v64H448c0-35.3 28.7-64 64-64zM448 96h64v64c-35.3 0-64-28.7-64-64z\"],\"chart-area\":[512,512,[\"area-chart\"],\"f1fe\",\"M64 64c0-17.7-14.3-32-32-32S0 46.3 0 64V400c0 44.2 35.8 80 80 80H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H80c-8.8 0-16-7.2-16-16V64zm96 288H448c17.7 0 32-14.3 32-32V251.8c0-7.6-2.7-15-7.7-20.8l-65.8-76.8c-12.1-14.2-33.7-15-46.9-1.8l-21 21c-10 10-26.4 9.2-35.4-1.6l-39.2-47c-12.6-15.1-35.7-15.4-48.7-.6L135.9 215c-5.1 5.8-7.9 13.3-7.9 21.1v84c0 17.7 14.3 32 32 32z\"],\"house-flag\":[640,512,[],\"e50d\",\"M480 0c-17.7 0-32 14.3-32 32V192 512h64V192H624c8.8 0 16-7.2 16-16V48c0-8.8-7.2-16-16-16H512c0-17.7-14.3-32-32-32zM416 159L276.8 39.7c-12-10.3-29.7-10.3-41.7 0l-224 192C1 240.4-2.7 254.5 2 267.1S18.6 288 32 288H64V480c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V384c0-17.7 14.3-32 32-32h64c17.7 0 32 14.3 32 32v96c0 17.7 14.3 32 32 32h64.7l.2 0h-1V159z\"],\"person-circle-minus\":[576,512,[],\"e540\",\"M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l44.9 74.7c-16.1 17.6-28.6 38.5-36.6 61.5c-1.9-1.8-3.5-3.9-4.9-6.3L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152zm136 16a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm224 0c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16s7.2 16 16 16H496c8.8 0 16-7.2 16-16z\"],ban:[512,512,[128683,\"cancel\"],\"f05e\",\"M367.2 412.5L99.5 144.8C77.1 176.1 64 214.5 64 256c0 106 86 192 192 192c41.5 0 79.9-13.1 111.2-35.5zm45.3-45.3C434.9 335.9 448 297.5 448 256c0-106-86-192-192-192c-41.5 0-79.9 13.1-111.2 35.5L412.5 367.2zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256z\"],\"camera-rotate\":[640,512,[],\"e0d8\",\"M213.1 64.8L202.7 96H128c-35.3 0-64 28.7-64 64V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H437.3L426.9 64.8C420.4 45.2 402.1 32 381.4 32H258.6c-20.7 0-39 13.2-45.5 32.8zM448 256c0 8.8-7.2 16-16 16H355.3c-6.2 0-11.3-5.1-11.3-11.3c0-3 1.2-5.9 3.3-8L371 229c-13.6-13.4-31.9-21-51-21c-19.2 0-37.7 7.6-51.3 21.3L249 249c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l19.7-19.7C257.4 172.7 288 160 320 160c31.8 0 62.4 12.6 85 35l23.7-23.7c2.1-2.1 5-3.3 8-3.3c6.2 0 11.3 5.1 11.3 11.3V256zM192 320c0-8.8 7.2-16 16-16h76.7c6.2 0 11.3 5.1 11.3 11.3c0 3-1.2 5.9-3.3 8L269 347c13.6 13.4 31.9 21 51 21c19.2 0 37.7-7.6 51.3-21.3L391 327c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-19.7 19.7C382.6 403.3 352 416 320 416c-31.8 0-62.4-12.6-85-35l-23.7 23.7c-2.1 2.1-5 3.3-8 3.3c-6.2 0-11.3-5.1-11.3-11.3V320z\"],\"spray-can-sparkles\":[512,512,[\"air-freshener\"],\"f5d0\",\"M96 32v96H224V32c0-17.7-14.3-32-32-32H128C110.3 0 96 14.3 96 32zm0 128c-53 0-96 43-96 96V464c0 26.5 21.5 48 48 48H272c26.5 0 48-21.5 48-48V256c0-53-43-96-96-96H96zm64 96a80 80 0 1 1 0 160 80 80 0 1 1 0-160zM384 48c0-1.4-1-3-2.2-3.6L352 32 339.6 2.2C339 1 337.4 0 336 0s-3 1-3.6 2.2L320 32 290.2 44.4C289 45 288 46.6 288 48c0 1.4 1 3 2.2 3.6L320 64l12.4 29.8C333 95 334.6 96 336 96s3-1 3.6-2.2L352 64l29.8-12.4C383 51 384 49.4 384 48zm76.4 45.8C461 95 462.6 96 464 96s3-1 3.6-2.2L480 64l29.8-12.4C511 51 512 49.4 512 48c0-1.4-1-3-2.2-3.6L480 32 467.6 2.2C467 1 465.4 0 464 0s-3 1-3.6 2.2L448 32 418.2 44.4C417 45 416 46.6 416 48c0 1.4 1 3 2.2 3.6L448 64l12.4 29.8zm7.2 100.4c-.6-1.2-2.2-2.2-3.6-2.2s-3 1-3.6 2.2L448 224l-29.8 12.4c-1.2 .6-2.2 2.2-2.2 3.6c0 1.4 1 3 2.2 3.6L448 256l12.4 29.8c.6 1.2 2.2 2.2 3.6 2.2s3-1 3.6-2.2L480 256l29.8-12.4c1.2-.6 2.2-2.2 2.2-3.6c0-1.4-1-3-2.2-3.6L480 224l-12.4-29.8zM448 144c0-1.4-1-3-2.2-3.6L416 128 403.6 98.2C403 97 401.4 96 400 96s-3 1-3.6 2.2L384 128l-29.8 12.4c-1.2 .6-2.2 2.2-2.2 3.6c0 1.4 1 3 2.2 3.6L384 160l12.4 29.8c.6 1.2 2.2 2.2 3.6 2.2s3-1 3.6-2.2L416 160l29.8-12.4c1.2-.6 2.2-2.2 2.2-3.6z\"],star:[576,512,[11088,61446],\"f005\",\"M316.9 18C311.6 7 300.4 0 288.1 0s-23.4 7-28.8 18L195 150.3 51.4 171.5c-12 1.8-22 10.2-25.7 21.7s-.7 24.2 7.9 32.7L137.8 329 113.2 474.7c-2 12 3 24.2 12.9 31.3s23 8 33.8 2.3l128.3-68.5 128.3 68.5c10.8 5.7 23.9 4.9 33.8-2.3s14.9-19.3 12.9-31.3L438.5 329 542.7 225.9c8.6-8.5 11.7-21.2 7.9-32.7s-13.7-19.9-25.7-21.7L381.2 150.3 316.9 18z\"],repeat:[512,512,[128257],\"f363\",\"M0 224c0 17.7 14.3 32 32 32s32-14.3 32-32c0-53 43-96 96-96H320v32c0 12.9 7.8 24.6 19.8 29.6s25.7 2.2 34.9-6.9l64-64c12.5-12.5 12.5-32.8 0-45.3l-64-64c-9.2-9.2-22.9-11.9-34.9-6.9S320 19.1 320 32V64H160C71.6 64 0 135.6 0 224zm512 64c0-17.7-14.3-32-32-32s-32 14.3-32 32c0 53-43 96-96 96H192V352c0-12.9-7.8-24.6-19.8-29.6s-25.7-2.2-34.9 6.9l-64 64c-12.5 12.5-12.5 32.8 0 45.3l64 64c9.2 9.2 22.9 11.9 34.9 6.9s19.8-16.6 19.8-29.6V448H352c88.4 0 160-71.6 160-160z\"],cross:[384,512,[128327,10013],\"f654\",\"M176 0c-26.5 0-48 21.5-48 48v80H48c-26.5 0-48 21.5-48 48v32c0 26.5 21.5 48 48 48h80V464c0 26.5 21.5 48 48 48h32c26.5 0 48-21.5 48-48V256h80c26.5 0 48-21.5 48-48V176c0-26.5-21.5-48-48-48H256V48c0-26.5-21.5-48-48-48H176z\"],box:[448,512,[128230],\"f466\",\"M50.7 58.5L0 160H208V32H93.7C75.5 32 58.9 42.3 50.7 58.5zM240 160H448L397.3 58.5C389.1 42.3 372.5 32 354.3 32H240V160zm208 32H0V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V192z\"],\"venus-mars\":[640,512,[9892],\"f228\",\"M176 288a112 112 0 1 0 0-224 112 112 0 1 0 0 224zM352 176c0 86.3-62.1 158.1-144 173.1V384h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H208v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H112c-17.7 0-32-14.3-32-32s14.3-32 32-32h32V349.1C62.1 334.1 0 262.3 0 176C0 78.8 78.8 0 176 0s176 78.8 176 176zM271.9 360.6c19.3-10.1 36.9-23.1 52.1-38.4c20 18.5 46.7 29.8 76.1 29.8c61.9 0 112-50.1 112-112s-50.1-112-112-112c-7.2 0-14.3 .7-21.1 2c-4.9-21.5-13-41.7-24-60.2C369.3 66 384.4 64 400 64c37 0 71.4 11.4 99.8 31l20.6-20.6L487 41c-6.9-6.9-8.9-17.2-5.2-26.2S494.3 0 504 0H616c13.3 0 24 10.7 24 24V136c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-33.4-33.4L545 140.2c19.5 28.4 31 62.7 31 99.8c0 97.2-78.8 176-176 176c-50.5 0-96-21.3-128.1-55.4z\"],\"arrow-pointer\":[320,512,[\"mouse-pointer\"],\"f245\",\"M0 55.2V426c0 12.2 9.9 22 22 22c6.3 0 12.4-2.7 16.6-7.5L121.2 346l58.1 116.3c7.9 15.8 27.1 22.2 42.9 14.3s22.2-27.1 14.3-42.9L179.8 320H297.9c12.2 0 22.1-9.9 22.1-22.1c0-6.3-2.7-12.3-7.4-16.5L38.6 37.9C34.3 34.1 28.9 32 23.2 32C10.4 32 0 42.4 0 55.2z\"],maximize:[512,512,[\"expand-arrows-alt\"],\"f31e\",\"M200 32H56C42.7 32 32 42.7 32 56V200c0 9.7 5.8 18.5 14.8 22.2s19.3 1.7 26.2-5.2l40-40 79 79-79 79L73 295c-6.9-6.9-17.2-8.9-26.2-5.2S32 302.3 32 312V456c0 13.3 10.7 24 24 24H200c9.7 0 18.5-5.8 22.2-14.8s1.7-19.3-5.2-26.2l-40-40 79-79 79 79-40 40c-6.9 6.9-8.9 17.2-5.2 26.2s12.5 14.8 22.2 14.8H456c13.3 0 24-10.7 24-24V312c0-9.7-5.8-18.5-14.8-22.2s-19.3-1.7-26.2 5.2l-40 40-79-79 79-79 40 40c6.9 6.9 17.2 8.9 26.2 5.2s14.8-12.5 14.8-22.2V56c0-13.3-10.7-24-24-24H312c-9.7 0-18.5 5.8-22.2 14.8s-1.7 19.3 5.2 26.2l40 40-79 79-79-79 40-40c6.9-6.9 8.9-17.2 5.2-26.2S209.7 32 200 32z\"],\"charging-station\":[576,512,[],\"f5e7\",\"M96 0C60.7 0 32 28.7 32 64V448c-17.7 0-32 14.3-32 32s14.3 32 32 32H320c17.7 0 32-14.3 32-32s-14.3-32-32-32V304h16c22.1 0 40 17.9 40 40v32c0 39.8 32.2 72 72 72s72-32.2 72-72V252.3c32.5-10.2 56-40.5 56-76.3V144c0-8.8-7.2-16-16-16H544V80c0-8.8-7.2-16-16-16s-16 7.2-16 16v48H480V80c0-8.8-7.2-16-16-16s-16 7.2-16 16v48H432c-8.8 0-16 7.2-16 16v32c0 35.8 23.5 66.1 56 76.3V376c0 13.3-10.7 24-24 24s-24-10.7-24-24V344c0-48.6-39.4-88-88-88H320V64c0-35.3-28.7-64-64-64H96zM216.9 82.7c6 4 8.5 11.5 6.3 18.3l-25 74.9H256c6.7 0 12.7 4.2 15 10.4s.5 13.3-4.6 17.7l-112 96c-5.5 4.7-13.4 5.1-19.3 1.1s-8.5-11.5-6.3-18.3l25-74.9H96c-6.7 0-12.7-4.2-15-10.4s-.5-13.3 4.6-17.7l112-96c5.5-4.7 13.4-5.1 19.3-1.1z\"],shapes:[512,512,[\"triangle-circle-square\"],\"f61f\",\"M315.4 15.5C309.7 5.9 299.2 0 288 0s-21.7 5.9-27.4 15.5l-96 160c-5.9 9.9-6.1 22.2-.4 32.2s16.3 16.2 27.8 16.2H384c11.5 0 22.2-6.2 27.8-16.2s5.5-22.3-.4-32.2l-96-160zM288 312V456c0 22.1 17.9 40 40 40H472c22.1 0 40-17.9 40-40V312c0-22.1-17.9-40-40-40H328c-22.1 0-40 17.9-40 40zM128 512a128 128 0 1 0 0-256 128 128 0 1 0 0 256z\"],shuffle:[512,512,[128256,\"random\"],\"f074\",\"M403.8 34.4c12-5 25.7-2.2 34.9 6.9l64 64c6 6 9.4 14.1 9.4 22.6s-3.4 16.6-9.4 22.6l-64 64c-9.2 9.2-22.9 11.9-34.9 6.9s-19.8-16.6-19.8-29.6V160H352c-10.1 0-19.6 4.7-25.6 12.8L284 229.3 244 176l31.2-41.6C293.3 110.2 321.8 96 352 96h32V64c0-12.9 7.8-24.6 19.8-29.6zM164 282.7L204 336l-31.2 41.6C154.7 401.8 126.2 416 96 416H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H96c10.1 0 19.6-4.7 25.6-12.8L164 282.7zm274.6 188c-9.2 9.2-22.9 11.9-34.9 6.9s-19.8-16.6-19.8-29.6V416H352c-30.2 0-58.7-14.2-76.8-38.4L121.6 172.8c-6-8.1-15.5-12.8-25.6-12.8H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H96c30.2 0 58.7 14.2 76.8 38.4L326.4 339.2c6 8.1 15.5 12.8 25.6 12.8h32V320c0-12.9 7.8-24.6 19.8-29.6s25.7-2.2 34.9 6.9l64 64c6 6 9.4 14.1 9.4 22.6s-3.4 16.6-9.4 22.6l-64 64z\"],\"person-running\":[448,512,[127939,\"running\"],\"f70c\",\"M320 48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM125.7 175.5c9.9-9.9 23.4-15.5 37.5-15.5c1.9 0 3.8 .1 5.6 .3L137.6 254c-9.3 28 1.7 58.8 26.8 74.5l86.2 53.9-25.4 88.8c-4.9 17 5 34.7 22 39.6s34.7-5 39.6-22l28.7-100.4c5.9-20.6-2.6-42.6-20.7-53.9L238 299l30.9-82.4 5.1 12.3C289 264.7 323.9 288 362.7 288H384c17.7 0 32-14.3 32-32s-14.3-32-32-32H362.7c-12.9 0-24.6-7.8-29.5-19.7l-6.3-15c-14.6-35.1-44.1-61.9-80.5-73.1l-48.7-15c-11.1-3.4-22.7-5.2-34.4-5.2c-31 0-60.8 12.3-82.7 34.3L57.4 153.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l23.1-23.1zM91.2 352H32c-17.7 0-32 14.3-32 32s14.3 32 32 32h69.6c19 0 36.2-11.2 43.9-28.5L157 361.6l-9.5-6c-17.5-10.9-30.5-26.8-37.9-44.9L91.2 352z\"],\"mobile-retro\":[320,512,[],\"e527\",\"M0 64C0 28.7 28.7 0 64 0H256c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zm64 96v64c0 17.7 14.3 32 32 32H224c17.7 0 32-14.3 32-32V160c0-17.7-14.3-32-32-32H96c-17.7 0-32 14.3-32 32zM80 352a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm24 56a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zm56-56a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm24 56a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zm56-56a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm24 56a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zM128 48c-8.8 0-16 7.2-16 16s7.2 16 16 16h64c8.8 0 16-7.2 16-16s-7.2-16-16-16H128z\"],\"grip-lines-vertical\":[192,512,[],\"f7a5\",\"M64 64c0-17.7-14.3-32-32-32S0 46.3 0 64V448c0 17.7 14.3 32 32 32s32-14.3 32-32V64zm128 0c0-17.7-14.3-32-32-32s-32 14.3-32 32V448c0 17.7 14.3 32 32 32s32-14.3 32-32V64z\"],spider:[512,512,[128375],\"f717\",\"M158.4 32.6c4.8-12.4-1.4-26.3-13.8-31s-26.3 1.4-31 13.8L81.1 100c-7.9 20.7-3 44.1 12.7 59.7l57.4 57.4L70.8 190.3c-2.4-.8-4.3-2.7-5.1-5.1L46.8 128.4C42.6 115.8 29 109 16.4 113.2S-3 131 1.2 143.6l18.9 56.8c5.6 16.7 18.7 29.8 35.4 35.4L116.1 256 55.6 276.2c-16.7 5.6-29.8 18.7-35.4 35.4L1.2 368.4C-3 381 3.8 394.6 16.4 398.8s26.2-2.6 30.4-15.2l18.9-56.8c.8-2.4 2.7-4.3 5.1-5.1l80.4-26.8L93.7 352.3C78.1 368 73.1 391.4 81.1 412l32.5 84.6c4.8 12.4 18.6 18.5 31 13.8s18.5-18.6 13.8-31l-32.5-84.6c-1.1-3-.4-6.3 1.8-8.5L160 353.9c1 52.1 43.6 94.1 96 94.1s95-41.9 96-94.1l32.3 32.3c2.2 2.2 2.9 5.6 1.8 8.5l-32.5 84.6c-4.8 12.4 1.4 26.3 13.8 31s26.3-1.4 31-13.8L430.9 412c7.9-20.7 3-44.1-12.7-59.7l-57.4-57.4 80.4 26.8c2.4 .8 4.3 2.7 5.1 5.1l18.9 56.8c4.2 12.6 17.8 19.4 30.4 15.2s19.4-17.8 15.2-30.4l-18.9-56.8c-5.6-16.7-18.7-29.8-35.4-35.4L395.9 256l60.5-20.2c16.7-5.6 29.8-18.7 35.4-35.4l18.9-56.8c4.2-12.6-2.6-26.2-15.2-30.4s-26.2 2.6-30.4 15.2l-18.9 56.8c-.8 2.4-2.7 4.3-5.1 5.1l-80.4 26.8 57.4-57.4c15.6-15.6 20.6-39 12.7-59.7L398.4 15.4C393.6 3 379.8-3.2 367.4 1.6s-18.5 18.6-13.8 31l32.5 84.6c1.1 3 .4 6.3-1.8 8.5L336 174.1V160c0-31.8-18.6-59.3-45.5-72.2c-9.1-4.4-18.5 3.3-18.5 13.4V112c0 8.8-7.2 16-16 16s-16-7.2-16-16V101.2c0-10.1-9.4-17.7-18.5-13.4C194.6 100.7 176 128.2 176 160v14.1l-48.3-48.3c-2.2-2.2-2.9-5.6-1.8-8.5l32.5-84.6z\"],\"hands-bound\":[640,512,[],\"e4f9\",\"M96 32C96 14.3 81.7 0 64 0S32 14.3 32 32V96v59.1 .7V192v21.9c0 14.2 5.1 27.9 14.3 38.7L131.6 352H128c-13.3 0-24 10.7-24 24s10.7 24 24 24h32H288h64H480h32c13.3 0 24-10.7 24-24s-10.7-24-24-24h-3.6l85.3-99.5c9.2-10.8 14.3-24.5 14.3-38.7V192 155.8v-.7V96 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V96v48.8l-69.3 92.4c-5.7 7.6-16.1 9.6-24.2 4.8c-9.7-5.7-12.1-18.7-5.1-27.5L473 180c10.8-13.5 8.9-33.3-4.4-44.5s-33-9.8-44.5 3.2l-46.7 52.5C361 209.7 352 233.4 352 258.1V320v32H288V320 258.1c0-24.6-9-48.4-25.4-66.8l-46.7-52.5c-11.5-13-31.3-14.4-44.5-3.2s-15.2 30.9-4.4 44.5l27.6 34.5c7 8.8 4.7 21.8-5.1 27.5c-8.1 4.8-18.6 2.7-24.2-4.8L96 144.8V96 32zm64 448v32H288V480h64v32H480V480h32c13.3 0 24-10.7 24-24s-10.7-24-24-24H480 352 288 160 128c-13.3 0-24 10.7-24 24s10.7 24 24 24h32z\"],\"file-invoice-dollar\":[384,512,[],\"f571\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM64 80c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16zm0 64c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16zm128 72c8.8 0 16 7.2 16 16v17.3c8.5 1.2 16.7 3.1 24.1 5.1c8.5 2.3 13.6 11 11.3 19.6s-11 13.6-19.6 11.3c-11.1-3-22-5.2-32.1-5.3c-8.4-.1-17.4 1.8-23.6 5.5c-5.7 3.4-8.1 7.3-8.1 12.8c0 3.7 1.3 6.5 7.3 10.1c6.9 4.1 16.6 7.1 29.2 10.9l.5 .1 0 0 0 0c11.3 3.4 25.3 7.6 36.3 14.6c12.1 7.6 22.4 19.7 22.7 38.2c.3 19.3-9.6 33.3-22.9 41.6c-7.7 4.8-16.4 7.6-25.1 9.1V440c0 8.8-7.2 16-16 16s-16-7.2-16-16V422.2c-11.2-2.1-21.7-5.7-30.9-8.9l0 0 0 0c-2.1-.7-4.2-1.4-6.2-2.1c-8.4-2.8-12.9-11.9-10.1-20.2s11.9-12.9 20.2-10.1c2.5 .8 4.8 1.6 7.1 2.4l0 0 0 0 0 0c13.6 4.6 24.6 8.4 36.3 8.7c9.1 .3 17.9-1.7 23.7-5.3c5.1-3.2 7.9-7.3 7.8-14c-.1-4.6-1.8-7.8-7.7-11.6c-6.8-4.3-16.5-7.4-29-11.2l-1.6-.5 0 0c-11-3.3-24.3-7.3-34.8-13.7c-12-7.2-22.6-18.9-22.7-37.3c-.1-19.4 10.8-32.8 23.8-40.5c7.5-4.4 15.8-7.2 24.1-8.7V232c0-8.8 7.2-16 16-16z\"],\"plane-circle-exclamation\":[640,512,[],\"e556\",\"M256 0c-35 0-64 59.5-64 93.7v84.6L8.1 283.4c-5 2.8-8.1 8.2-8.1 13.9v65.5c0 10.6 10.2 18.3 20.4 15.4l171.6-49 0 70.9-57.6 43.2c-4 3-6.4 7.8-6.4 12.8v42c0 7.8 6.3 14 14 14c1.3 0 2.6-.2 3.9-.5L256 480l110.1 31.5c1.3 .4 2.6 .5 3.9 .5c6 0 11.1-3.7 13.1-9C344.5 470.7 320 422.2 320 368c0-60.6 30.6-114 77.1-145.6L320 178.3V93.7C320 59.5 292 0 256 0zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z\"],\"x-ray\":[512,512,[],\"f497\",\"M0 64C0 46.3 14.3 32 32 32H480c17.7 0 32 14.3 32 32s-14.3 32-32 32V416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32V96C14.3 96 0 81.7 0 64zM256 96c-8.8 0-16 7.2-16 16v32H160c-8.8 0-16 7.2-16 16s7.2 16 16 16h80v48H128c-8.8 0-16 7.2-16 16s7.2 16 16 16H240v70.6L189.1 307c-5.2-2-10.6-3-16.2-3h-2.1c-23.6 0-42.8 19.2-42.8 42.8c0 9.6 3.2 18.9 9.1 26.4l18.2 23.2c9.7 12.4 24.6 19.6 40.3 19.6H316.4c15.7 0 30.6-7.2 40.3-19.6l18.2-23.2c5.9-7.5 9.1-16.8 9.1-26.4c0-23.6-19.2-42.8-42.8-42.8H339c-5.5 0-11 1-16.2 3L272 326.6V256H384c8.8 0 16-7.2 16-16s-7.2-16-16-16H272V176h80c8.8 0 16-7.2 16-16s-7.2-16-16-16H272V112c0-8.8-7.2-16-16-16zM208 352a16 16 0 1 1 0 32 16 16 0 1 1 0-32zm80 16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0z\"],\"spell-check\":[576,512,[],\"f891\",\"M112 0C99.1 0 87.4 7.8 82.5 19.7l-66.7 160-13.3 32c-6.8 16.3 .9 35 17.2 41.8s35-.9 41.8-17.2L66.7 224h90.7l5.1 12.3c6.8 16.3 25.5 24 41.8 17.2s24-25.5 17.2-41.8l-13.3-32-66.7-160C136.6 7.8 124.9 0 112 0zm18.7 160H93.3L112 115.2 130.7 160zM256 32v96 96c0 17.7 14.3 32 32 32h80c44.2 0 80-35.8 80-80c0-23.1-9.8-43.8-25.4-58.4c6-11.2 9.4-24 9.4-37.6c0-44.2-35.8-80-80-80H288c-17.7 0-32 14.3-32 32zm96 64H320V64h32c8.8 0 16 7.2 16 16s-7.2 16-16 16zm-32 64h32 16c8.8 0 16 7.2 16 16s-7.2 16-16 16H320V160zM566.6 310.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L352 434.7l-73.4-73.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l96 96c12.5 12.5 32.8 12.5 45.3 0l192-192z\"],slash:[640,512,[],\"f715\",\"M5.1 9.2C13.3-1.2 28.4-3.1 38.8 5.1l592 464c10.4 8.2 12.3 23.3 4.1 33.7s-23.3 12.3-33.7 4.1L9.2 42.9C-1.2 34.7-3.1 19.6 5.1 9.2z\"],\"computer-mouse\":[384,512,[128433,\"mouse\"],\"f8cc\",\"M0 192H176V0H160C71.6 0 0 71.6 0 160v32zm0 32V352c0 88.4 71.6 160 160 160h64c88.4 0 160-71.6 160-160V224H192 0zm384-32V160C384 71.6 312.4 0 224 0H208V192H384z\"],\"arrow-right-to-bracket\":[512,512,[\"sign-in\"],\"f090\",\"M352 96l64 0c17.7 0 32 14.3 32 32l0 256c0 17.7-14.3 32-32 32l-64 0c-17.7 0-32 14.3-32 32s14.3 32 32 32l64 0c53 0 96-43 96-96l0-256c0-53-43-96-96-96l-64 0c-17.7 0-32 14.3-32 32s14.3 32 32 32zm-9.4 182.6c12.5-12.5 12.5-32.8 0-45.3l-128-128c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L242.7 224 32 224c-17.7 0-32 14.3-32 32s14.3 32 32 32l210.7 0-73.4 73.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l128-128z\"],\"shop-slash\":[640,512,[\"store-alt-slash\"],\"e070\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-54.8-43V224H512V376L384 275.7V224H320v1.5L277.2 192H603.2c20.3 0 36.8-16.5 36.8-36.8c0-7.3-2.2-14.4-6.2-20.4L558.2 21.4C549.3 8 534.4 0 518.3 0H121.7c-16 0-31 8-39.9 21.4L74.1 32.8 38.8 5.1zM36.8 192h85L21 112.5 6.2 134.7c-4 6.1-6.2 13.2-6.2 20.4C0 175.5 16.5 192 36.8 192zM320 384H128V224H64V384v80c0 26.5 21.5 48 48 48H336c26.5 0 48-21.5 48-48V398.5l-64-50.4V384zM544 512l-.3 0h.6l-.3 0z\"],server:[512,512,[],\"f233\",\"M64 32C28.7 32 0 60.7 0 96v64c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm280 72a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm48 24a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zM64 288c-35.3 0-64 28.7-64 64v64c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V352c0-35.3-28.7-64-64-64H64zm280 72a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm56 24a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z\"],\"virus-covid-slash\":[640,512,[],\"e4a9\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L472.1 344.7c11.4-19.5 19.1-41.4 22.3-64.7H528v16c0 13.3 10.7 24 24 24s24-10.7 24-24V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v16H494.4c-4.2-30.7-16.3-58.8-34.1-82.3L484 125.9l11.3 11.3c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L472.7 46.7c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9L450.1 92l-23.8 23.8C402.8 97.9 374.7 85.8 344 81.6V48h16c13.3 0 24-10.7 24-24s-10.7-24-24-24H280c-13.3 0-24 10.7-24 24s10.7 24 24 24h16V81.6c-30.7 4.2-58.8 16.3-82.3 34.1L189.9 92l11.3-11.3c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L134.1 79.8 38.8 5.1zm362.5 407L149.2 213.5c-1.5 6-2.7 12.2-3.5 18.5H112V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v80c0 13.3 10.7 24 24 24s24-10.7 24-24V280h33.6c4.2 30.7 16.3 58.8 34.1 82.3L156 386.1l-11.3-11.3c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l56.6 56.6c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L189.9 420l23.8-23.8c23.5 17.9 51.7 29.9 82.3 34.1V464H280c-13.3 0-24 10.7-24 24s10.7 24 24 24h80c13.3 0 24-10.7 24-24s-10.7-24-24-24H344V430.4c20.4-2.8 39.7-9.1 57.3-18.2z\"],\"shop-lock\":[640,512,[],\"e4a5\",\"M36.8 192H449.6c20.2-19.8 47.9-32 78.4-32c30.5 0 58.1 12.2 78.3 31.9c18.9-1.6 33.7-17.4 33.7-36.7c0-7.3-2.2-14.4-6.2-20.4L558.2 21.4C549.3 8 534.4 0 518.3 0H121.7c-16 0-31 8-39.9 21.4L6.2 134.7c-4 6.1-6.2 13.2-6.2 20.4C0 175.5 16.5 192 36.8 192zM384 224H320V384H128V224H64V384v80c0 26.5 21.5 48 48 48H336c26.5 0 48-21.5 48-48V384 352 224zm144 16c17.7 0 32 14.3 32 32v48H496V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z\"],\"hourglass-start\":[384,512,[\"hourglass-1\"],\"f251\",\"M32 0C14.3 0 0 14.3 0 32S14.3 64 32 64V75c0 42.4 16.9 83.1 46.9 113.1L146.7 256 78.9 323.9C48.9 353.9 32 394.6 32 437v11c-17.7 0-32 14.3-32 32s14.3 32 32 32H64 320h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V437c0-42.4-16.9-83.1-46.9-113.1L237.3 256l67.9-67.9c30-30 46.9-70.7 46.9-113.1V64c17.7 0 32-14.3 32-32s-14.3-32-32-32H320 64 32zM288 437v11H96V437c0-25.5 10.1-49.9 28.1-67.9L192 301.3l67.9 67.9c18 18 28.1 42.4 28.1 67.9z\"],\"blender-phone\":[576,512,[],\"f6b6\",\"M224 352L196.8 52.3C194.2 24.2 216.3 0 244.6 0H534.1c21.1 0 36.4 20.1 30.9 40.4L558.5 64H400c-8.8 0-16 7.2-16 16s7.2 16 16 16H549.8l-17.5 64H400c-8.8 0-16 7.2-16 16s7.2 16 16 16H523.6l-17.5 64H400c-8.8 0-16 7.2-16 16s7.2 16 16 16h97.5L480 352H224zm-16 32H496c26.5 0 48 21.5 48 48v32c0 26.5-21.5 48-48 48H208c-26.5 0-48-21.5-48-48V432c0-26.5 21.5-48 48-48zm144 96a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM147.5 30.7c10.8 6.7 15.3 21 10.6 33.4l-22 57.8c-4.2 10.9-14.5 17.6-25.3 16.4l-33.3-3.6c-13.6 42.2-13.6 88.4 0 130.7l33.3-3.6c10.9-1.2 21.2 5.5 25.3 16.4l22 57.8c4.7 12.4 .2 26.7-10.6 33.4l-44 27.2c-9.7 6-21.9 4.2-29.8-4.3C-24.6 286-24.6 114 73.7 7.8C81.6-.7 93.8-2.5 103.5 3.5l44 27.2z\"],\"building-wheat\":[640,512,[],\"e4db\",\"M0 48C0 21.5 21.5 0 48 0H336c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H240V432c0-26.5-21.5-48-48-48s-48 21.5-48 48v80H48c-26.5 0-48-21.5-48-48V48zM80 224c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H80zm80 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H176c-8.8 0-16 7.2-16 16zm112-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H272zM64 112v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H80c-8.8 0-16 7.2-16 16zM176 96c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H176zm80 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16zm384 80v16c0 44.2-35.8 80-80 80H544V272c0-44.2 35.8-80 80-80h16zm0 128c0 44.2-35.8 80-80 80H544V384c0-44.2 35.8-80 80-80h16v16zm0 112c0 44.2-35.8 80-80 80H544V496c0-44.2 35.8-80 80-80h16v16zM512 496v16H496c-44.2 0-80-35.8-80-80V416h16c44.2 0 80 35.8 80 80zm0-96H496c-44.2 0-80-35.8-80-80V304h16c44.2 0 80 35.8 80 80v16zm0-128v16H496c-44.2 0-80-35.8-80-80V192h16c44.2 0 80 35.8 80 80zM528 32c13.3 0 24 10.7 24 24V160c0 13.3-10.7 24-24 24s-24-10.7-24-24V56c0-13.3 10.7-24 24-24zm96 64v32c0 13.3-10.7 24-24 24s-24-10.7-24-24V96c0-13.3 10.7-24 24-24s24 10.7 24 24zM456 72c13.3 0 24 10.7 24 24v32c0 13.3-10.7 24-24 24s-24-10.7-24-24V96c0-13.3 10.7-24 24-24z\"],\"person-breastfeeding\":[448,512,[],\"e53a\",\"M224 0a80 80 0 1 1 0 160A80 80 0 1 1 224 0zM436.8 382.8L373.5 462c-16.6 20.7-46.8 24.1-67.5 7.5c-17.6-14.1-22.7-38.1-13.5-57.7l-.8-.1c-38.9-5.6-74.3-25.1-99.7-54.8V320c0-17.7-14.3-32-32-32s-32 14.3-32 32v48c0 .8 0 1.6 .1 2.4l101.4 50.7c23.7 11.9 33.3 40.7 21.5 64.4s-40.7 33.3-64.4 21.5L27.2 427.3c-1.1-.5-2.2-1.1-3.3-1.7c-4.9-2.8-9.2-6.4-12.6-10.6c-4.6-5.4-7.8-11.7-9.6-18.4c-3.3-12-1.9-25.2 4.8-36.6c.6-1.1 1.3-2.2 2-3.2L75.6 256.1c26.7-40.1 71.7-64.1 119.8-64.1h75.2c46.5 0 90.1 22.5 117.2 60.3l50.7 70.9c2.2 3 4 6.1 5.5 9.4c2.9 6.7 4.3 13.8 4 20.8c-.3 10.6-4.2 21-11.2 29.4zM320 332a44 44 0 1 0 -88 0 44 44 0 1 0 88 0z\"],\"right-to-bracket\":[512,512,[\"sign-in-alt\"],\"f2f6\",\"M217.9 105.9L340.7 228.7c7.2 7.2 11.3 17.1 11.3 27.3s-4.1 20.1-11.3 27.3L217.9 406.1c-6.4 6.4-15 9.9-24 9.9c-18.7 0-33.9-15.2-33.9-33.9l0-62.1L32 320c-17.7 0-32-14.3-32-32l0-64c0-17.7 14.3-32 32-32l128 0 0-62.1c0-18.7 15.2-33.9 33.9-33.9c9 0 17.6 3.6 24 9.9zM352 416l64 0c17.7 0 32-14.3 32-32l0-256c0-17.7-14.3-32-32-32l-64 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l64 0c53 0 96 43 96 96l0 256c0 53-43 96-96 96l-64 0c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],venus:[384,512,[9792],\"f221\",\"M80 176a112 112 0 1 1 224 0A112 112 0 1 1 80 176zM224 349.1c81.9-15 144-86.8 144-173.1C368 78.8 289.2 0 192 0S16 78.8 16 176c0 86.3 62.1 158.1 144 173.1V384H128c-17.7 0-32 14.3-32 32s14.3 32 32 32h32v32c0 17.7 14.3 32 32 32s32-14.3 32-32V448h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H224V349.1z\"],passport:[448,512,[],\"f5ab\",\"M0 64C0 28.7 28.7 0 64 0H384c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM183 278.8c-27.9-13.2-48.4-39.4-53.7-70.8h39.1c1.6 30.4 7.7 53.8 14.6 70.8zm41.3 9.2l-.3 0-.3 0c-2.4-3.5-5.7-8.9-9.1-16.5c-6-13.6-12.4-34.3-14.2-63.5h47.1c-1.8 29.2-8.1 49.9-14.2 63.5c-3.4 7.6-6.7 13-9.1 16.5zm40.7-9.2c6.8-17.1 12.9-40.4 14.6-70.8h39.1c-5.3 31.4-25.8 57.6-53.7 70.8zM279.6 176c-1.6-30.4-7.7-53.8-14.6-70.8c27.9 13.2 48.4 39.4 53.7 70.8H279.6zM223.7 96l.3 0 .3 0c2.4 3.5 5.7 8.9 9.1 16.5c6 13.6 12.4 34.3 14.2 63.5H200.5c1.8-29.2 8.1-49.9 14.2-63.5c3.4-7.6 6.7-13 9.1-16.5zM183 105.2c-6.8 17.1-12.9 40.4-14.6 70.8H129.3c5.3-31.4 25.8-57.6 53.7-70.8zM352 192A128 128 0 1 0 96 192a128 128 0 1 0 256 0zM112 384c-8.8 0-16 7.2-16 16s7.2 16 16 16H336c8.8 0 16-7.2 16-16s-7.2-16-16-16H112z\"],\"heart-pulse\":[512,512,[\"heartbeat\"],\"f21e\",\"M228.3 469.1L47.6 300.4c-4.2-3.9-8.2-8.1-11.9-12.4h87c22.6 0 43-13.6 51.7-34.5l10.5-25.2 49.3 109.5c3.8 8.5 12.1 14 21.4 14.1s17.8-5 22-13.3L320 253.7l1.7 3.4c9.5 19 28.9 31 50.1 31H476.3c-3.7 4.3-7.7 8.5-11.9 12.4L283.7 469.1c-7.5 7-17.4 10.9-27.7 10.9s-20.2-3.9-27.7-10.9zM503.7 240h-132c-3 0-5.8-1.7-7.2-4.4l-23.2-46.3c-4.1-8.1-12.4-13.3-21.5-13.3s-17.4 5.1-21.5 13.3l-41.4 82.8L205.9 158.2c-3.9-8.7-12.7-14.3-22.2-14.1s-18.1 5.9-21.8 14.8l-31.8 76.3c-1.2 3-4.2 4.9-7.4 4.9H16c-2.6 0-5 .4-7.3 1.1C3 225.2 0 208.2 0 190.9v-5.8c0-69.9 50.5-129.5 119.4-141C165 36.5 211.4 51.4 244 84l12 12 12-12c32.6-32.6 79-47.5 124.6-39.9C461.5 55.6 512 115.2 512 185.1v5.8c0 16.9-2.8 33.5-8.3 49.1z\"],\"people-carry-box\":[640,512,[\"people-carry\"],\"f4ce\",\"M80 48a48 48 0 1 1 96 0A48 48 0 1 1 80 48zm64 193.7v65.1l51 51c7.1 7.1 11.8 16.2 13.4 26.1l15.2 90.9c2.9 17.4-8.9 33.9-26.3 36.8s-33.9-8.9-36.8-26.3l-14.3-85.9L66.8 320C54.8 308 48 291.7 48 274.7V186.6c0-32.4 26.2-58.6 58.6-58.6c24.1 0 46.5 12 59.9 32l47.4 71.1 10.1 5V160c0-17.7 14.3-32 32-32H384c17.7 0 32 14.3 32 32v76.2l10.1-5L473.5 160c13.3-20 35.8-32 59.9-32c32.4 0 58.6 26.2 58.6 58.6v88.1c0 17-6.7 33.3-18.7 45.3l-79.4 79.4-14.3 85.9c-2.9 17.4-19.4 29.2-36.8 26.3s-29.2-19.4-26.3-36.8l15.2-90.9c1.6-9.9 6.3-19 13.4-26.1l51-51V241.7l-19 28.5c-4.6 7-11 12.6-18.5 16.3l-59.6 29.8c-2.4 1.3-4.9 2.2-7.6 2.8c-2.6 .6-5.3 .9-7.9 .8H256.7c-2.5 .1-5-.2-7.5-.7c-2.9-.6-5.6-1.6-8.1-3l-59.5-29.8c-7.5-3.7-13.8-9.4-18.5-16.3l-19-28.5zM2.3 468.1L50.1 348.6l49.2 49.2-37.6 94c-6.6 16.4-25.2 24.4-41.6 17.8S-4.3 484.5 2.3 468.1zM512 0a48 48 0 1 1 0 96 48 48 0 1 1 0-96zm77.9 348.6l47.8 119.5c6.6 16.4-1.4 35-17.8 41.6s-35-1.4-41.6-17.8l-37.6-94 49.2-49.2z\"],\"temperature-high\":[512,512,[],\"f769\",\"M416 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm0 128A96 96 0 1 0 416 0a96 96 0 1 0 0 192zM96 112c0-26.5 21.5-48 48-48s48 21.5 48 48V276.5c0 17.3 7.1 31.9 15.3 42.5C217.8 332.6 224 349.5 224 368c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-18.5 6.2-35.4 16.7-48.9C88.9 308.4 96 293.8 96 276.5V112zM144 0C82.1 0 32 50.2 32 112V276.5c0 .1-.1 .3-.2 .6c-.2 .6-.8 1.6-1.7 2.8C11.2 304.2 0 334.8 0 368c0 79.5 64.5 144 144 144s144-64.5 144-144c0-33.2-11.2-63.8-30.1-88.1c-.9-1.2-1.5-2.2-1.7-2.8c-.1-.3-.2-.5-.2-.6V112C256 50.2 205.9 0 144 0zm0 416c26.5 0 48-21.5 48-48c0-20.9-13.4-38.7-32-45.3V112c0-8.8-7.2-16-16-16s-16 7.2-16 16V322.7c-18.6 6.6-32 24.4-32 45.3c0 26.5 21.5 48 48 48z\"],microchip:[512,512,[],\"f2db\",\"M176 24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64c-35.3 0-64 28.7-64 64H24c-13.3 0-24 10.7-24 24s10.7 24 24 24H64v56H24c-13.3 0-24 10.7-24 24s10.7 24 24 24H64v56H24c-13.3 0-24 10.7-24 24s10.7 24 24 24H64c0 35.3 28.7 64 64 64v40c0 13.3 10.7 24 24 24s24-10.7 24-24V448h56v40c0 13.3 10.7 24 24 24s24-10.7 24-24V448h56v40c0 13.3 10.7 24 24 24s24-10.7 24-24V448c35.3 0 64-28.7 64-64h40c13.3 0 24-10.7 24-24s-10.7-24-24-24H448V280h40c13.3 0 24-10.7 24-24s-10.7-24-24-24H448V176h40c13.3 0 24-10.7 24-24s-10.7-24-24-24H448c0-35.3-28.7-64-64-64V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H280V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V64H176V24zM160 128H352c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32V160c0-17.7 14.3-32 32-32zm192 32H160V352H352V160z\"],crown:[576,512,[128081],\"f521\",\"M309 106c11.4-7 19-19.7 19-34c0-22.1-17.9-40-40-40s-40 17.9-40 40c0 14.4 7.6 27 19 34L209.7 220.6c-9.1 18.2-32.7 23.4-48.6 10.7L72 160c5-6.7 8-15 8-24c0-22.1-17.9-40-40-40S0 113.9 0 136s17.9 40 40 40c.2 0 .5 0 .7 0L86.4 427.4c5.5 30.4 32 52.6 63 52.6H426.6c30.9 0 57.4-22.1 63-52.6L535.3 176c.2 0 .5 0 .7 0c22.1 0 40-17.9 40-40s-17.9-40-40-40s-40 17.9-40 40c0 9 3 17.3 8 24l-89.1 71.3c-15.9 12.7-39.5 7.5-48.6-10.7L309 106z\"],\"weight-hanging\":[512,512,[],\"f5cd\",\"M224 96a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm122.5 32c3.5-10 5.5-20.8 5.5-32c0-53-43-96-96-96s-96 43-96 96c0 11.2 1.9 22 5.5 32H120c-22 0-41.2 15-46.6 36.4l-72 288c-3.6 14.3-.4 29.5 8.7 41.2S33.2 512 48 512H464c14.8 0 28.7-6.8 37.8-18.5s12.3-26.8 8.7-41.2l-72-288C433.2 143 414 128 392 128H346.5z\"],\"xmarks-lines\":[640,512,[],\"e59a\",\"M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H32zm0 384c-17.7 0-32 14.3-32 32s14.3 32 32 32H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H32zM7 167c-9.4 9.4-9.4 24.6 0 33.9l55 55L7 311c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l55-55 55 55c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-55-55 55-55c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-55 55L41 167c-9.4-9.4-24.6-9.4-33.9 0zM265 167c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l55 55-55 55c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l55-55 55 55c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-55-55 55-55c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-55 55-55-55zM455 167c-9.4 9.4-9.4 24.6 0 33.9l55 55-55 55c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l55-55 55 55c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-55-55 55-55c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-55 55-55-55c-9.4-9.4-24.6-9.4-33.9 0z\"],\"file-prescription\":[384,512,[],\"f572\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM104 196h72c33.1 0 60 26.9 60 60c0 25.5-15.9 47.2-38.3 55.9l43 40.3 33.8-31c8.1-7.5 20.8-6.9 28.3 1.2s6.9 20.8-1.2 28.3L270 379.7l31.7 29.7c8.1 7.6 8.5 20.2 .9 28.3s-20.2 8.5-28.3 .9l-33.9-31.8-34.9 32c-8.1 7.5-20.8 6.9-28.3-1.2s-6.9-20.8 1.2-28.3l32.6-29.9-64.8-60.8c-.9-.8-1.6-1.7-2.3-2.6H124v44c0 11-9 20-20 20s-20-9-20-20V296 216c0-11 9-20 20-20zm72 80c11 0 20-9 20-20s-9-20-20-20H124v40h52z\"],\"weight-scale\":[512,512,[\"weight\"],\"f496\",\"M128 176a128 128 0 1 1 256 0 128 128 0 1 1 -256 0zM391.8 64C359.5 24.9 310.7 0 256 0S152.5 24.9 120.2 64H64C28.7 64 0 92.7 0 128V448c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H391.8zM296 224c0-10.6-4.1-20.2-10.9-27.4l33.6-78.3c3.5-8.1-.3-17.5-8.4-21s-17.5 .3-21 8.4L255.7 184c-22 .1-39.7 18-39.7 40c0 22.1 17.9 40 40 40s40-17.9 40-40z\"],\"user-group\":[640,512,[128101,\"user-friends\"],\"f500\",\"M96 128a128 128 0 1 1 256 0A128 128 0 1 1 96 128zM0 482.3C0 383.8 79.8 304 178.3 304h91.4C368.2 304 448 383.8 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3zM609.3 512H471.4c5.4-9.4 8.6-20.3 8.6-32v-8c0-60.7-27.1-115.2-69.8-151.8c2.4-.1 4.7-.2 7.1-.2h61.4C567.8 320 640 392.2 640 481.3c0 17-13.8 30.7-30.7 30.7zM432 256c-31 0-59-12.6-79.3-32.9C372.4 196.5 384 163.6 384 128c0-26.8-6.6-52.1-18.3-74.3C384.3 40.1 407.2 32 432 32c61.9 0 112 50.1 112 112s-50.1 112-112 112z\"],\"arrow-up-a-z\":[576,512,[\"sort-alpha-up\"],\"f15e\",\"M183.6 42.4C177.5 35.8 169 32 160 32s-17.5 3.8-23.6 10.4l-88 96c-11.9 13-11.1 33.3 2 45.2s33.3 11.1 45.2-2L128 146.3V448c0 17.7 14.3 32 32 32s32-14.3 32-32V146.3l32.4 35.4c11.9 13 32.2 13.9 45.2 2s13.9-32.2 2-45.2l-88-96zM320 320c0 17.7 14.3 32 32 32h50.7l-73.4 73.4c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H429.3l73.4-73.4c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8H352c-17.7 0-32 14.3-32 32zM416 32c-12.1 0-23.2 6.8-28.6 17.7l-64 128-16 32c-7.9 15.8-1.5 35 14.3 42.9s35 1.5 42.9-14.3l7.2-14.3h88.4l7.2 14.3c7.9 15.8 27.1 22.2 42.9 14.3s22.2-27.1 14.3-42.9l-16-32-64-128C439.2 38.8 428.1 32 416 32zM395.8 176L416 135.6 436.2 176H395.8z\"],\"chess-knight\":[448,512,[9822],\"f441\",\"M96 48L82.7 61.3C70.7 73.3 64 89.5 64 106.5V238.9c0 10.7 5.3 20.7 14.2 26.6l10.6 7c14.3 9.6 32.7 10.7 48.1 3l3.2-1.6c2.6-1.3 5-2.8 7.3-4.5l49.4-37c6.6-5 15.7-5 22.3 0c10.2 7.7 9.9 23.1-.7 30.3L90.4 350C73.9 361.3 64 380 64 400H384l28.9-159c2.1-11.3 3.1-22.8 3.1-34.3V192C416 86 330 0 224 0H83.8C72.9 0 64 8.9 64 19.8c0 7.5 4.2 14.3 10.9 17.7L96 48zm24 68a20 20 0 1 1 40 0 20 20 0 1 1 -40 0zM22.6 473.4c-4.2 4.2-6.6 10-6.6 16C16 501.9 26.1 512 38.6 512H409.4c12.5 0 22.6-10.1 22.6-22.6c0-6-2.4-11.8-6.6-16L384 432H64L22.6 473.4z\"],\"face-laugh-squint\":[512,512,[\"laugh-squint\"],\"f59b\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM96.8 314.1c-3.8-13.7 7.4-26.1 21.6-26.1H393.6c14.2 0 25.5 12.4 21.6 26.1C396.2 382 332.1 432 256 432s-140.2-50-159.2-117.9zm36.7-199.4l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6c0-9 9.6-14.7 17.5-10.5zM396 125.1c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9c7.9-4.2 17.5 1.5 17.5 10.5z\"],wheelchair:[512,512,[],\"f193\",\"M192 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM120.5 247.2c12.4-4.7 18.7-18.5 14-30.9s-18.5-18.7-30.9-14C43.1 225.1 0 283.5 0 352c0 88.4 71.6 160 160 160c61.2 0 114.3-34.3 141.2-84.7c6.2-11.7 1.8-26.2-9.9-32.5s-26.2-1.8-32.5 9.9C240 440 202.8 464 160 464C98.1 464 48 413.9 48 352c0-47.9 30.1-88.8 72.5-104.8zM259.8 176l-1.9-9.7c-4.5-22.3-24-38.3-46.8-38.3c-30.1 0-52.7 27.5-46.8 57l23.1 115.5c6 29.9 32.2 51.4 62.8 51.4h5.1c.4 0 .8 0 1.3 0h94.1c6.7 0 12.6 4.1 15 10.4L402 459.2c6 16.1 23.8 24.6 40.1 19.1l48-16c16.8-5.6 25.8-23.7 20.2-40.5s-23.7-25.8-40.5-20.2l-18.7 6.2-25.5-68c-11.7-31.2-41.6-51.9-74.9-51.9H282.2l-9.6-48H336c17.7 0 32-14.3 32-32s-14.3-32-32-32H259.8z\"],\"circle-arrow-up\":[512,512,[\"arrow-circle-up\"],\"f0aa\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM385 231c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-71-71V376c0 13.3-10.7 24-24 24s-24-10.7-24-24V193.9l-71 71c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9L239 119c9.4-9.4 24.6-9.4 33.9 0L385 231z\"],\"toggle-on\":[576,512,[],\"f205\",\"M192 64C86 64 0 150 0 256S86 448 192 448H384c106 0 192-86 192-192s-86-192-192-192H192zm192 96a96 96 0 1 1 0 192 96 96 0 1 1 0-192z\"],\"person-walking\":[320,512,[128694,\"walking\"],\"f554\",\"M160 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM126.5 199.3c-1 .4-1.9 .8-2.9 1.2l-8 3.5c-16.4 7.3-29 21.2-34.7 38.2l-2.6 7.8c-5.6 16.8-23.7 25.8-40.5 20.2s-25.8-23.7-20.2-40.5l2.6-7.8c11.4-34.1 36.6-61.9 69.4-76.5l8-3.5c20.8-9.2 43.3-14 66.1-14c44.6 0 84.8 26.8 101.9 67.9L281 232.7l21.4 10.7c15.8 7.9 22.2 27.1 14.3 42.9s-27.1 22.2-42.9 14.3L247 287.3c-10.3-5.2-18.4-13.8-22.8-24.5l-9.6-23-19.3 65.5 49.5 54c5.4 5.9 9.2 13 11.2 20.8l23 92.1c4.3 17.1-6.1 34.5-23.3 38.8s-34.5-6.1-38.8-23.3l-22-88.1-70.7-77.1c-14.8-16.1-20.3-38.6-14.7-59.7l16.9-63.5zM68.7 398l25-62.4c2.1 3 4.5 5.8 7 8.6l40.7 44.4-14.5 36.2c-2.4 6-6 11.5-10.6 16.1L54.6 502.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L68.7 398z\"],l:[320,512,[108],\"4c\",\"M64 32c17.7 0 32 14.3 32 32V416H288c17.7 0 32 14.3 32 32s-14.3 32-32 32H64c-17.7 0-32-14.3-32-32V64c0-17.7 14.3-32 32-32z\"],fire:[448,512,[128293],\"f06d\",\"M159.3 5.4c7.8-7.3 19.9-7.2 27.7 .1c27.6 25.9 53.5 53.8 77.7 84c11-14.4 23.5-30.1 37-42.9c7.9-7.4 20.1-7.4 28 .1c34.6 33 63.9 76.6 84.5 118c20.3 40.8 33.8 82.5 33.8 111.9C448 404.2 348.2 512 224 512C98.4 512 0 404.1 0 276.5c0-38.4 17.8-85.3 45.4-131.7C73.3 97.7 112.7 48.6 159.3 5.4zM225.7 416c25.3 0 47.7-7 68.8-21c42.1-29.4 53.4-88.2 28.1-134.4c-4.5-9-16-9.6-22.5-2l-25.2 29.3c-6.6 7.6-18.5 7.4-24.7-.5c-16.5-21-46-58.5-62.8-79.8c-6.3-8-18.3-8.1-24.7-.1c-33.8 42.5-50.8 69.3-50.8 99.4C112 375.4 162.6 416 225.7 416z\"],\"bed-pulse\":[640,512,[\"procedures\"],\"f487\",\"M483.2 9.6L524 64h92c13.3 0 24 10.7 24 24s-10.7 24-24 24H512c-7.6 0-14.7-3.6-19.2-9.6L468.7 70.3l-47 99.9c-3.7 7.8-11.3 13.1-19.9 13.7s-16.9-3.4-21.7-10.6L339.2 112H216c-13.3 0-24-10.7-24-24s10.7-24 24-24H352c8 0 15.5 4 20 10.7l24.4 36.6 45.9-97.5C445.9 6.2 453.2 1 461.6 .1s16.6 2.7 21.6 9.5zM320 160h12.7l20.7 31.1c11.2 16.8 30.6 26.3 50.7 24.8s37.9-13.7 46.5-32L461.9 160H544c53 0 96 43 96 96V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H352 320 64v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V96C0 78.3 14.3 64 32 64s32 14.3 32 32V352H288V192c0-17.7 14.3-32 32-32zm-144 0a80 80 0 1 1 0 160 80 80 0 1 1 0-160z\"],\"shuttle-space\":[640,512,[\"space-shuttle\"],\"f197\",\"M130 480c40.6 0 80.4-11 115.2-31.9L352 384l-224 0 0 96h2zM352 128L245.2 63.9C210.4 43 170.6 32 130 32h-2v96l224 0zM96 128l0-96H80C53.5 32 32 53.5 32 80v48h8c-22.1 0-40 17.9-40 40v16V328v16c0 22.1 17.9 40 40 40H32v48c0 26.5 21.5 48 48 48H96l0-96h8c26.2 0 49.4-12.6 64-32H456c69.3 0 135-22.7 179.2-81.6c6.4-8.5 6.4-20.3 0-28.8C591 182.7 525.3 160 456 160H168c-14.6-19.4-37.8-32-64-32l-8 0zM512 243.6v24.9c0 19.6-15.9 35.6-35.6 35.6c-2.5 0-4.4-2-4.4-4.4V212.4c0-2.5 2-4.4 4.4-4.4c19.6 0 35.6 15.9 35.6 35.6z\"],\"face-laugh\":[512,512,[\"laugh\"],\"f599\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM96.8 314.1c-3.8-13.7 7.4-26.1 21.6-26.1H393.6c14.2 0 25.5 12.4 21.6 26.1C396.2 382 332.1 432 256 432s-140.2-50-159.2-117.9zM144.4 192a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"folder-open\":[576,512,[128194,128449,61717],\"f07c\",\"M88.7 223.8L0 375.8V96C0 60.7 28.7 32 64 32H181.5c17 0 33.3 6.7 45.3 18.7l26.5 26.5c12 12 28.3 18.7 45.3 18.7H416c35.3 0 64 28.7 64 64v32H144c-22.8 0-43.8 12.1-55.3 31.8zm27.6 16.1C122.1 230 132.6 224 144 224H544c11.5 0 22 6.1 27.7 16.1s5.7 22.2-.1 32.1l-112 192C453.9 474 443.4 480 432 480H32c-11.5 0-22-6.1-27.7-16.1s-5.7-22.2 .1-32.1l112-192z\"],\"heart-circle-plus\":[576,512,[],\"e500\",\"M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9l2.6-2.4C267.2 438.6 256 404.6 256 368c0-97.2 78.8-176 176-176c28.3 0 55 6.7 78.7 18.5c.9-6.5 1.3-13 1.3-19.6v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm16-208v48h48c8.8 0 16 7.2 16 16s-7.2 16-16 16H448v48c0 8.8-7.2 16-16 16s-16-7.2-16-16V384H368c-8.8 0-16-7.2-16-16s7.2-16 16-16h48V304c0-8.8 7.2-16 16-16s16 7.2 16 16z\"],\"code-fork\":[448,512,[],\"e13b\",\"M80 104a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm80-24c0 32.8-19.7 61-48 73.3V192c0 17.7 14.3 32 32 32H304c17.7 0 32-14.3 32-32V153.3C307.7 141 288 112.8 288 80c0-44.2 35.8-80 80-80s80 35.8 80 80c0 32.8-19.7 61-48 73.3V192c0 53-43 96-96 96H256v70.7c28.3 12.3 48 40.5 48 73.3c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-32.8 19.7-61 48-73.3V288H144c-53 0-96-43-96-96V153.3C19.7 141 0 112.8 0 80C0 35.8 35.8 0 80 0s80 35.8 80 80zm208 24a24 24 0 1 0 0-48 24 24 0 1 0 0 48zM248 432a24 24 0 1 0 -48 0 24 24 0 1 0 48 0z\"],city:[640,512,[127961],\"f64f\",\"M480 48c0-26.5-21.5-48-48-48H336c-26.5 0-48 21.5-48 48V96H224V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V96H112V24c0-13.3-10.7-24-24-24S64 10.7 64 24V96H48C21.5 96 0 117.5 0 144v96V464c0 26.5 21.5 48 48 48H304h32 96H592c26.5 0 48-21.5 48-48V240c0-26.5-21.5-48-48-48H480V48zm96 320v32c0 8.8-7.2 16-16 16H528c-8.8 0-16-7.2-16-16V368c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16zM240 416H208c-8.8 0-16-7.2-16-16V368c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16zM128 400c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V368c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32zM560 256c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H528c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h32zM256 176v32c0 8.8-7.2 16-16 16H208c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16zM112 160c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16h32zM256 304c0 8.8-7.2 16-16 16H208c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32zM112 320H80c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16zm304-48v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16zM400 64c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V80c0-8.8 7.2-16 16-16h32zm16 112v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16z\"],\"microphone-lines\":[384,512,[127897,\"microphone-alt\"],\"f3c9\",\"M96 96V256c0 53 43 96 96 96s96-43 96-96H208c-8.8 0-16-7.2-16-16s7.2-16 16-16h80V192H208c-8.8 0-16-7.2-16-16s7.2-16 16-16h80V128H208c-8.8 0-16-7.2-16-16s7.2-16 16-16h80c0-53-43-96-96-96S96 43 96 96zM320 240v16c0 70.7-57.3 128-128 128s-128-57.3-128-128V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v40c0 89.1 66.2 162.7 152 174.4V464H120c-13.3 0-24 10.7-24 24s10.7 24 24 24h72 72c13.3 0 24-10.7 24-24s-10.7-24-24-24H216V430.4c85.8-11.7 152-85.3 152-174.4V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v24z\"],\"pepper-hot\":[512,512,[127798],\"f816\",\"M428.3 3c11.6-6.4 26.2-2.3 32.6 9.3l4.8 8.7c19.3 34.7 19.8 75.7 3.4 110C495.8 159.6 512 197.9 512 240c0 18.5-3.1 36.3-8.9 52.8c-6.1 17.3-28.5 16.3-36.8-.1l-11.7-23.4c-4.1-8.1-12.4-13.3-21.5-13.3H360c-13.3 0-24-10.7-24-24V152c0-13.3-10.7-24-24-24l-17.1 0c-21.3 0-30-23.9-10.8-32.9C304.7 85.4 327.7 80 352 80c28.3 0 54.8 7.3 77.8 20.2c5.5-18.2 3.7-38.4-6-55.8L419 35.7c-6.4-11.6-2.3-26.2 9.3-32.6zM171.2 345.5L264 160l40 0v80c0 26.5 21.5 48 48 48h76.2l23.9 47.8C372.3 443.9 244.3 512 103.2 512H44.4C19.9 512 0 492.1 0 467.6c0-20.8 14.5-38.8 34.8-43.3l49.8-11.1c37.6-8.4 69.5-33.2 86.7-67.7z\"],unlock:[448,512,[128275],\"f09c\",\"M144 144c0-44.2 35.8-80 80-80c31.9 0 59.4 18.6 72.3 45.7c7.6 16 26.7 22.8 42.6 15.2s22.8-26.7 15.2-42.6C331 33.7 281.5 0 224 0C144.5 0 80 64.5 80 144v48H64c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V256c0-35.3-28.7-64-64-64H144V144z\"],\"colon-sign\":[384,512,[],\"e140\",\"M255 39.8c4.3-17.1-6.1-34.5-23.3-38.8S197.2 7.1 193 24.2L181.9 68.6C96.1 87.8 32 164.4 32 256c0 58.1 25.8 110.2 66.7 145.4L81 472.2c-4.3 17.1 6.1 34.5 23.3 38.8s34.5-6.1 38.8-23.3l13-52.1c9 3.4 18.4 6.2 28 8.2L177 472.2c-4.3 17.1 6.1 34.5 23.3 38.8s34.5-6.1 38.8-23.3l10.4-41.4c33.4-4.4 64.1-17.4 89.8-36.7c14.1-10.6 17-30.7 6.4-44.8s-30.7-17-44.8-6.4c-10.2 7.7-21.7 13.9-34 18.3L321 160c9.4-.3 18.5-4.7 24.6-12.8c10.6-14.1 7.8-34.2-6.4-44.8c-1.1-.8-2.2-1.6-3.3-2.4L351 39.8c4.3-17.1-6.1-34.5-23.3-38.8S293.2 7.1 289 24.2L277.2 71.5c-9.3-2.7-18.8-4.6-28.6-5.9L255 39.8zM163.2 143.3L117.3 326.8C103.9 306.5 96 282.2 96 256c0-48.7 27.2-91 67.2-112.7zm8.6 229.5l61.1-244.6c9.9 .7 19.5 2.5 28.7 5.3l-62 248.1c-9.7-1.9-19-4.8-27.8-8.8z\"],headset:[512,512,[],\"f590\",\"M256 48C141.1 48 48 141.1 48 256v40c0 13.3-10.7 24-24 24s-24-10.7-24-24V256C0 114.6 114.6 0 256 0S512 114.6 512 256V400.1c0 48.6-39.4 88-88.1 88L313.6 488c-8.3 14.3-23.8 24-41.6 24H240c-26.5 0-48-21.5-48-48s21.5-48 48-48h32c17.8 0 33.3 9.7 41.6 24l110.4 .1c22.1 0 40-17.9 40-40V256c0-114.9-93.1-208-208-208zM144 208h16c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32H144c-35.3 0-64-28.7-64-64V272c0-35.3 28.7-64 64-64zm224 0c35.3 0 64 28.7 64 64v48c0 35.3-28.7 64-64 64H352c-17.7 0-32-14.3-32-32V240c0-17.7 14.3-32 32-32h16z\"],\"store-slash\":[640,512,[],\"e071\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-86.8-68V384 252.6c-4 1-8 1.8-12.3 2.3l-.1 0c-5.3 .7-10.7 1.1-16.2 1.1c-12.4 0-24.3-1.9-35.4-5.3V350.9L301.2 210.7c7-4.4 13.3-9.7 18.8-15.7c15.9 17.6 39.1 29 65.2 29c26.2 0 49.3-11.4 65.2-29c16 17.6 39.1 29 65.2 29c4.1 0 8.1-.3 12.1-.8c55.5-7.4 81.8-72.5 52.1-119.4L522.3 13.1C517.2 5 508.1 0 498.4 0H141.6c-9.7 0-18.8 5-23.9 13.1l-22.7 36L38.8 5.1zM514.9 501.7L365.5 384H160V250.6c-11.2 3.5-23.2 5.4-35.6 5.4c-5.5 0-11-.4-16.3-1.1l-.1 0c-4.1-.6-8.1-1.3-12-2.3V384v64c0 35.3 28.7 64 64 64H480c12.9 0 24.8-3.8 34.9-10.3zM155.3 218.4L48.9 134.5c-6.1 40.6 19.5 82.8 63.3 88.7c4 .5 8.1 .8 12.1 .8c11 0 21.4-2 31-5.6z\"],\"road-circle-xmark\":[640,512,[],\"e566\",\"M213.2 32H288V96c0 17.7 14.3 32 32 32s32-14.3 32-32V32h74.8c27.1 0 51.3 17.1 60.3 42.6l42.7 120.6c-10.9-2.1-22.2-3.2-33.8-3.2c-59.5 0-112.1 29.6-144 74.8V224c0-17.7-14.3-32-32-32s-32 14.3-32 32v64c0 17.7 14.3 32 32 32c2.3 0 4.6-.3 6.8-.7c-4.5 15.5-6.8 31.8-6.8 48.7c0 5.4 .2 10.7 .7 16l-.7 0c-17.7 0-32 14.3-32 32v64H86.6C56.5 480 32 455.5 32 425.4c0-6.2 1.1-12.4 3.1-18.2L152.9 74.6C162 49.1 186.1 32 213.2 32zM496 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm22.6 144l36.7-36.7c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0L496 345.4l-36.7-36.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6L473.4 368l-36.7 36.7c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L496 390.6l36.7 36.7c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6L518.6 368z\"],\"user-minus\":[640,512,[],\"f503\",\"M96 128a128 128 0 1 1 256 0A128 128 0 1 1 96 128zM0 482.3C0 383.8 79.8 304 178.3 304h91.4C368.2 304 448 383.8 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3zM472 200H616c13.3 0 24 10.7 24 24s-10.7 24-24 24H472c-13.3 0-24-10.7-24-24s10.7-24 24-24z\"],\"mars-stroke-up\":[320,512,[9896,\"mars-stroke-v\"],\"f22a\",\"M148.7 4.7c6.2-6.2 16.4-6.2 22.6 0l64 64c4.6 4.6 5.9 11.5 3.5 17.4s-8.3 9.9-14.8 9.9H184v24h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H184v24c0 .6 0 1.2-.1 1.8c77 11.6 136.1 78 136.1 158.2c0 88.4-71.6 160-160 160S0 440.4 0 352c0-80.2 59.1-146.7 136.1-158.2c0-.6-.1-1.2-.1-1.8V168H104c-13.3 0-24-10.7-24-24s10.7-24 24-24h32V96H96c-6.5 0-12.3-3.9-14.8-9.9s-1.1-12.9 3.5-17.4l64-64zM256 352A96 96 0 1 0 64 352a96 96 0 1 0 192 0z\"],\"champagne-glasses\":[640,512,[129346,\"glass-cheers\"],\"f79f\",\"M155.6 17.3C163 3 179.9-3.6 195 1.9L320 47.5l125-45.6c15.1-5.5 32 1.1 39.4 15.4l78.8 152.9c28.8 55.8 10.3 122.3-38.5 156.6L556.1 413l41-15c16.6-6 35 2.5 41 19.1s-2.5 35-19.1 41l-71.1 25.9L476.8 510c-16.6 6.1-35-2.5-41-19.1s2.5-35 19.1-41l41-15-31.3-86.2c-59.4 5.2-116.2-34-130-95.2L320 188.8l-14.6 64.7c-13.8 61.3-70.6 100.4-130 95.2l-31.3 86.2 41 15c16.6 6 25.2 24.4 19.1 41s-24.4 25.2-41 19.1L92.2 484.1 21.1 458.2c-16.6-6.1-25.2-24.4-19.1-41s24.4-25.2 41-19.1l41 15 31.3-86.2C66.5 292.5 48.1 226 76.9 170.2L155.6 17.3zm44 54.4l-27.2 52.8L261.6 157l13.1-57.9L199.6 71.7zm240.9 0L365.4 99.1 378.5 157l89.2-32.5L440.5 71.7z\"],clipboard:[384,512,[128203],\"f328\",\"M192 0c-41.8 0-77.4 26.7-90.5 64H64C28.7 64 0 92.7 0 128V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H282.5C269.4 26.7 233.8 0 192 0zm0 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM112 192H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],\"house-circle-exclamation\":[640,512,[],\"e50a\",\"M320.7 352c8.1-89.7 83.5-160 175.3-160c8.9 0 17.6 .7 26.1 1.9L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32v69.7c-.1 .9-.1 1.8-.1 2.8V472c0 22.1 17.9 40 40 40h16c1.2 0 2.4-.1 3.6-.2c1.5 .1 3 .2 4.5 .2H160h24c22.1 0 40-17.9 40-40V448 384c0-17.7 14.3-32 32-32h64l.7 0zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z\"],\"file-arrow-up\":[384,512,[\"file-upload\"],\"f574\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM216 408c0 13.3-10.7 24-24 24s-24-10.7-24-24V305.9l-31 31c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l72-72c9.4-9.4 24.6-9.4 33.9 0l72 72c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-31-31V408z\"],wifi:[640,512,[\"wifi-3\",\"wifi-strong\"],\"f1eb\",\"M54.2 202.9C123.2 136.7 216.8 96 320 96s196.8 40.7 265.8 106.9c12.8 12.2 33 11.8 45.2-.9s11.8-33-.9-45.2C549.7 79.5 440.4 32 320 32S90.3 79.5 9.8 156.7C-2.9 169-3.3 189.2 8.9 202s32.5 13.2 45.2 .9zM320 256c56.8 0 108.6 21.1 148.2 56c13.3 11.7 33.5 10.4 45.2-2.8s10.4-33.5-2.8-45.2C459.8 219.2 393 192 320 192s-139.8 27.2-190.5 72c-13.3 11.7-14.5 31.9-2.8 45.2s31.9 14.5 45.2 2.8c39.5-34.9 91.3-56 148.2-56zm64 160a64 64 0 1 0 -128 0 64 64 0 1 0 128 0z\"],bath:[512,512,[128705,\"bathtub\"],\"f2cd\",\"M96 77.3c0-7.3 5.9-13.3 13.3-13.3c3.5 0 6.9 1.4 9.4 3.9l14.9 14.9C130 91.8 128 101.7 128 112c0 19.9 7.2 38 19.2 52c-5.3 9.2-4 21.1 3.8 29c9.4 9.4 24.6 9.4 33.9 0L289 89c9.4-9.4 9.4-24.6 0-33.9c-7.9-7.9-19.8-9.1-29-3.8C246 39.2 227.9 32 208 32c-10.3 0-20.2 2-29.2 5.5L163.9 22.6C149.4 8.1 129.7 0 109.3 0C66.6 0 32 34.6 32 77.3V256c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H96V77.3zM32 352v16c0 28.4 12.4 54 32 71.6V480c0 17.7 14.3 32 32 32s32-14.3 32-32V464H384v16c0 17.7 14.3 32 32 32s32-14.3 32-32V439.6c19.6-17.6 32-43.1 32-71.6V352H32z\"],underline:[448,512,[],\"f0cd\",\"M16 64c0-17.7 14.3-32 32-32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H128V224c0 53 43 96 96 96s96-43 96-96V96H304c-17.7 0-32-14.3-32-32s14.3-32 32-32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H384V224c0 88.4-71.6 160-160 160s-160-71.6-160-160V96H48C30.3 96 16 81.7 16 64zM0 448c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32z\"],\"user-pen\":[640,512,[\"user-edit\"],\"f4ff\",\"M224 256A128 128 0 1 0 224 0a128 128 0 1 0 0 256zm-45.7 48C79.8 304 0 383.8 0 482.3C0 498.7 13.3 512 29.7 512H322.8c-3.1-8.8-3.7-18.4-1.4-27.8l15-60.1c2.8-11.3 8.6-21.5 16.8-29.7l40.3-40.3c-32.1-31-75.7-50.1-123.9-50.1H178.3zm435.5-68.3c-15.6-15.6-40.9-15.6-56.6 0l-29.4 29.4 71 71 29.4-29.4c15.6-15.6 15.6-40.9 0-56.6l-14.4-14.4zM375.9 417c-4.1 4.1-7 9.2-8.4 14.9l-15 60.1c-1.4 5.5 .2 11.2 4.2 15.2s9.7 5.6 15.2 4.2l60.1-15c5.6-1.4 10.8-4.3 14.9-8.4L576.1 358.7l-71-71L375.9 417z\"],signature:[640,512,[],\"f5b7\",\"M192 128c0-17.7 14.3-32 32-32s32 14.3 32 32v7.8c0 27.7-2.4 55.3-7.1 82.5l-84.4 25.3c-40.6 12.2-68.4 49.6-68.4 92v71.9c0 40 32.5 72.5 72.5 72.5c26 0 50-13.9 62.9-36.5l13.9-24.3c26.8-47 46.5-97.7 58.4-150.5l94.4-28.3-12.5 37.5c-3.3 9.8-1.6 20.5 4.4 28.8s15.7 13.3 26 13.3H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H460.4l18-53.9c3.8-11.3 .9-23.8-7.4-32.4s-20.7-11.8-32.2-8.4L316.4 198.1c2.4-20.7 3.6-41.4 3.6-62.3V128c0-53-43-96-96-96s-96 43-96 96v32c0 17.7 14.3 32 32 32s32-14.3 32-32V128zm-9.2 177l49-14.7c-10.4 33.8-24.5 66.4-42.1 97.2l-13.9 24.3c-1.5 2.6-4.3 4.3-7.4 4.3c-4.7 0-8.5-3.8-8.5-8.5V335.6c0-14.1 9.3-26.6 22.8-30.7zM24 368c-13.3 0-24 10.7-24 24s10.7 24 24 24H64.3c-.2-2.8-.3-5.6-.3-8.5V368H24zm592 48c13.3 0 24-10.7 24-24s-10.7-24-24-24H305.9c-6.7 16.3-14.2 32.3-22.3 48H616z\"],stroopwafel:[512,512,[],\"f551\",\"M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM312.6 63.7c-6.2-6.2-16.4-6.2-22.6 0L256 97.6 222.1 63.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l33.9 33.9-45.3 45.3-56.6-56.6c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l56.6 56.6-45.3 45.3L86.3 199.4c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6L97.6 256 63.7 289.9c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0l33.9-33.9 45.3 45.3-56.6 56.6c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0l56.6-56.6 45.3 45.3-33.9 33.9c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L256 414.4l33.9 33.9c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-33.9-33.9 45.3-45.3 56.6 56.6c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-56.6-56.6 45.3-45.3 33.9 33.9c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6L414.4 256l33.9-33.9c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0l-33.9 33.9-45.3-45.3 56.6-56.6c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0l-56.6 56.6-45.3-45.3 33.9-33.9c6.2-6.2 6.2-16.4 0-22.6zM142.9 256l45.3-45.3L233.4 256l-45.3 45.3L142.9 256zm67.9 67.9L256 278.6l45.3 45.3L256 369.1l-45.3-45.3zM278.6 256l45.3-45.3L369.1 256l-45.3 45.3L278.6 256zm22.6-67.9L256 233.4l-45.3-45.3L256 142.9l45.3 45.3z\"],bold:[384,512,[],\"f032\",\"M0 64C0 46.3 14.3 32 32 32H80 96 224c70.7 0 128 57.3 128 128c0 31.3-11.3 60.1-30 82.3c37.1 22.4 62 63.1 62 109.7c0 70.7-57.3 128-128 128H96 80 32c-17.7 0-32-14.3-32-32s14.3-32 32-32H48V256 96H32C14.3 96 0 81.7 0 64zM224 224c35.3 0 64-28.7 64-64s-28.7-64-64-64H112V224H224zM112 288V416H256c35.3 0 64-28.7 64-64s-28.7-64-64-64H224 112z\"],\"anchor-lock\":[640,512,[],\"e4ad\",\"M320 96a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm21.1 80C367 158.8 384 129.4 384 96c0-53-43-96-96-96s-96 43-96 96c0 33.4 17 62.8 42.9 80H224c-17.7 0-32 14.3-32 32s14.3 32 32 32h32V448H208c-53 0-96-43-96-96v-6.1l7 7c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L97 263c-9.4-9.4-24.6-9.4-33.9 0L7 319c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l7-7V352c0 88.4 71.6 160 160 160h80 80c8 0 15.9-.6 23.6-1.7c-4.8-9-7.6-19.3-7.6-30.3V446.7c-5.2 .9-10.5 1.3-16 1.3H320V240h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H341.1zM528 240c17.7 0 32 14.3 32 32v48H496V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z\"],\"building-ngo\":[384,512,[],\"e4d7\",\"M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM168 64h48c8.8 0 16 7.2 16 16s-7.2 16-16 16H184v64h16V144c0-8.8 7.2-16 16-16s16 7.2 16 16v24c0 13.3-10.7 24-24 24H176c-13.3 0-24-10.7-24-24V80c0-8.8 7.2-16 16-16zM304 96c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16s16-7.2 16-16V112c0-8.8-7.2-16-16-16zm-48 16c0-26.5 21.5-48 48-48s48 21.5 48 48v32c0 26.5-21.5 48-48 48s-48-21.5-48-48V112zM61.3 71.1l34.7 52V80c0-8.8 7.2-16 16-16s16 7.2 16 16v96c0 7.1-4.6 13.3-11.4 15.3s-14-.6-17.9-6.4L64 132.8V176c0 8.8-7.2 16-16 16s-16-7.2-16-16V80c0-7.1 4.6-13.3 11.4-15.3s14 .6 17.9 6.4z\"],\"manat-sign\":[384,512,[],\"e1d5\",\"M192 32c-17.7 0-32 14.3-32 32V98.7C69.2 113.9 0 192.9 0 288V448c0 17.7 14.3 32 32 32s32-14.3 32-32V288c0-59.6 40.8-109.8 96-124V448c0 17.7 14.3 32 32 32s32-14.3 32-32V164c55.2 14.2 96 64.3 96 124V448c0 17.7 14.3 32 32 32s32-14.3 32-32V288c0-95.1-69.2-174.1-160-189.3V64c0-17.7-14.3-32-32-32z\"],\"not-equal\":[448,512,[],\"f53e\",\"M369.8 37.4c14.7 9.8 18.7 29.7 8.9 44.4L337.1 144H400c17.7 0 32 14.3 32 32s-14.3 32-32 32H294.5l-64 96H400c17.7 0 32 14.3 32 32s-14.3 32-32 32H187.8l-65.2 97.7c-9.8 14.7-29.7 18.7-44.4 8.9s-18.7-29.7-8.9-44.4L110.9 368H48c-17.7 0-32-14.3-32-32s14.3-32 32-32H153.5l64-96H48c-17.7 0-32-14.3-32-32s14.3-32 32-32H260.2l65.2-97.7c9.8-14.7 29.7-18.7 44.4-8.9z\"],\"border-top-left\":[448,512,[\"border-style\"],\"f853\",\"M0 448c0 17.7 14.3 32 32 32s32-14.3 32-32l0-336c0-8.8 7.2-16 16-16l336 0c17.7 0 32-14.3 32-32s-14.3-32-32-32L80 32C35.8 32 0 67.8 0 112L0 448zm160 0a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm192 0a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm-96 0a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm192 0a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM416 288a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm0 32a32 32 0 1 0 0 64 32 32 0 1 0 0-64zm0-128a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"map-location-dot\":[576,512,[\"map-marked-alt\"],\"f5a0\",\"M408 120c0 54.6-73.1 151.9-105.2 192c-7.7 9.6-22 9.6-29.6 0C241.1 271.9 168 174.6 168 120C168 53.7 221.7 0 288 0s120 53.7 120 120zm8 80.4c3.5-6.9 6.7-13.8 9.6-20.6c.5-1.2 1-2.5 1.5-3.7l116-46.4C558.9 123.4 576 135 576 152V422.8c0 9.8-6 18.6-15.1 22.3L416 503V200.4zM137.6 138.3c2.4 14.1 7.2 28.3 12.8 41.5c2.9 6.8 6.1 13.7 9.6 20.6V451.8L32.9 502.7C17.1 509 0 497.4 0 480.4V209.6c0-9.8 6-18.6 15.1-22.3l122.6-49zM327.8 332c13.9-17.4 35.7-45.7 56.2-77V504.3L192 449.4V255c20.5 31.3 42.3 59.6 56.2 77c20.5 25.6 59.1 25.6 79.6 0zM288 152a40 40 0 1 0 0-80 40 40 0 1 0 0 80z\"],jedi:[576,512,[],\"f669\",\"M246 315.7l-21.2-31.9c-2.1-3.2-1.7-7.4 1-10.1s6.9-3.1 10.1-1l29.5 19.7c2.1 1.4 4.9 0 5-2.6L279.7 8c.1-4.5 3.8-8 8.3-8s8.1 3.5 8.3 8l9.4 281.9c.1 2.5 2.9 3.9 5 2.6l29.5-19.7c3.2-2.1 7.4-1.7 10.1 1s3.1 6.9 1 10.1L330 315.7c-1.3 1.9-.2 4.5 2 4.9l37.6 7.5c3.7 .7 6.4 4 6.4 7.8s-2.7 7.1-6.4 7.8L332 351.4c-2.2 .4-3.3 3-2 4.9l21.2 31.9c2.1 3.2 1.7 7.4-1 10.1s-6.9 3.1-10.1 1l-26.3-17.6c-2.2-1.4-5.1 .2-5 2.8l2.1 61.5C370.6 435.2 416 382.9 416 320c0-37-15.7-70.4-40.8-93.7c-7-6.5-6.5-18.6 1-24.4C410.1 175.5 432 134.3 432 88c0-16.8-2.9-33-8.2-48c-4.6-13 10.2-30 21.4-22c53.5 38 92.7 94.8 107.8 160.7c.5 2.1-.2 4.3-1.7 5.9l-28.4 28.4c-4 4-1.2 10.9 4.5 10.9h26c3.4 0 6.2 2.6 6.3 6c.1 3.3 .2 6.6 .2 10c0 17.5-1.7 34.7-4.8 51.3c-.2 1.2-.9 2.4-1.7 3.3l-46.5 46.5c-4 4-1.2 10.9 4.5 10.9H526c4.6 0 7.7 4.8 5.7 9C487.2 450.5 394.8 512 288 512S88.8 450.5 44.3 361c-2.1-4.2 1-9 5.7-9H64.5c5.7 0 8.6-6.9 4.5-10.9L22.6 294.6c-.9-.9-1.5-2-1.7-3.3C17.7 274.7 16 257.5 16 240c0-3.3 .1-6.7 .2-10c.1-3.4 2.9-6 6.3-6h26c5.7 0 8.6-6.9 4.5-10.9L24.6 184.6c-1.5-1.5-2.2-3.8-1.7-5.9C38.1 112.8 77.3 56 130.8 18c11.3-8 26 8.9 21.4 22c-5.3 15-8.2 31.2-8.2 48c0 46.3 21.9 87.5 55.8 113.9c7.5 5.8 8 17.9 1 24.4C175.7 249.6 160 283 160 320c0 62.9 45.4 115.2 105.1 126l2.1-61.5c.1-2.6-2.8-4.2-5-2.8l-26.3 17.6c-3.2 2.1-7.4 1.7-10.1-1s-3.1-6.9-1-10.1L246 356.3c1.3-1.9 .2-4.5-2-4.9l-37.6-7.5c-3.7-.7-6.4-4-6.4-7.8s2.7-7.1 6.4-7.8l37.6-7.5c2.2-.4 3.3-3 2-4.9z\"],\"square-poll-vertical\":[448,512,[\"poll\"],\"f681\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm64 192c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V256c0-17.7 14.3-32 32-32zm64-64c0-17.7 14.3-32 32-32s32 14.3 32 32V352c0 17.7-14.3 32-32 32s-32-14.3-32-32V160zM320 288c17.7 0 32 14.3 32 32v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V320c0-17.7 14.3-32 32-32z\"],\"mug-hot\":[512,512,[9749],\"f7b6\",\"M88 0C74.7 0 64 10.7 64 24c0 38.9 23.4 59.4 39.1 73.1l1.1 1C120.5 112.3 128 119.9 128 136c0 13.3 10.7 24 24 24s24-10.7 24-24c0-38.9-23.4-59.4-39.1-73.1l-1.1-1C119.5 47.7 112 40.1 112 24c0-13.3-10.7-24-24-24zM32 192c-17.7 0-32 14.3-32 32V416c0 53 43 96 96 96H288c53 0 96-43 96-96h16c61.9 0 112-50.1 112-112s-50.1-112-112-112H352 32zm352 64h16c26.5 0 48 21.5 48 48s-21.5 48-48 48H384V256zM224 24c0-13.3-10.7-24-24-24s-24 10.7-24 24c0 38.9 23.4 59.4 39.1 73.1l1.1 1C232.5 112.3 240 119.9 240 136c0 13.3 10.7 24 24 24s24-10.7 24-24c0-38.9-23.4-59.4-39.1-73.1l-1.1-1C231.5 47.7 224 40.1 224 24z\"],\"car-battery\":[512,512,[\"battery-car\"],\"f5df\",\"M80 96c0-17.7 14.3-32 32-32h64c17.7 0 32 14.3 32 32l96 0c0-17.7 14.3-32 32-32h64c17.7 0 32 14.3 32 32h16c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V160c0-35.3 28.7-64 64-64l16 0zm304 96c0-8.8-7.2-16-16-16s-16 7.2-16 16v32H320c-8.8 0-16 7.2-16 16s7.2 16 16 16h32v32c0 8.8 7.2 16 16 16s16-7.2 16-16V256h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H384V192zM80 240c0 8.8 7.2 16 16 16h96c8.8 0 16-7.2 16-16s-7.2-16-16-16H96c-8.8 0-16 7.2-16 16z\"],gift:[512,512,[127873],\"f06b\",\"M190.5 68.8L225.3 128H224 152c-22.1 0-40-17.9-40-40s17.9-40 40-40h2.2c14.9 0 28.8 7.9 36.3 20.8zM64 88c0 14.4 3.5 28 9.6 40H32c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H480c17.7 0 32-14.3 32-32V160c0-17.7-14.3-32-32-32H438.4c6.1-12 9.6-25.6 9.6-40c0-48.6-39.4-88-88-88h-2.2c-31.9 0-61.5 16.9-77.7 44.4L256 85.5l-24.1-41C215.7 16.9 186.1 0 154.2 0H152C103.4 0 64 39.4 64 88zm336 0c0 22.1-17.9 40-40 40H288h-1.3l34.8-59.2C329.1 55.9 342.9 48 357.8 48H360c22.1 0 40 17.9 40 40zM32 288V464c0 26.5 21.5 48 48 48H224V288H32zM288 512H432c26.5 0 48-21.5 48-48V288H288V512z\"],\"dice-two\":[448,512,[9857],\"f528\",\"M0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM352 352a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM128 192a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"chess-queen\":[512,512,[9819],\"f445\",\"M256 0a56 56 0 1 1 0 112A56 56 0 1 1 256 0zM134.1 143.8c3.3-13 15-23.8 30.2-23.8c12.3 0 22.6 7.2 27.7 17c12 23.2 36.2 39 64 39s52-15.8 64-39c5.1-9.8 15.4-17 27.7-17c15.3 0 27 10.8 30.2 23.8c7 27.8 32.2 48.3 62.1 48.3c10.8 0 21-2.7 29.8-7.4c8.4-4.4 18.9-4.5 27.6 .9c13 8 17.1 25 9.2 38L399.7 400H384 343.6 168.4 128 112.3L5.4 223.6c-7.9-13-3.8-30 9.2-38c8.7-5.3 19.2-5.3 27.6-.9c8.9 4.7 19 7.4 29.8 7.4c29.9 0 55.1-20.5 62.1-48.3zM256 224l0 0 0 0h0zM112 432H400l41.4 41.4c4.2 4.2 6.6 10 6.6 16c0 12.5-10.1 22.6-22.6 22.6H86.6C74.1 512 64 501.9 64 489.4c0-6 2.4-11.8 6.6-16L112 432z\"],glasses:[576,512,[],\"f530\",\"M118.6 80c-11.5 0-21.4 7.9-24 19.1L57 260.3c20.5-6.2 48.3-12.3 78.7-12.3c32.3 0 61.8 6.9 82.8 13.5c10.6 3.3 19.3 6.7 25.4 9.2c3.1 1.3 5.5 2.4 7.3 3.2c.9 .4 1.6 .7 2.1 1l.6 .3 .2 .1 .1 0 0 0 0 0s0 0-6.3 12.7h0l6.3-12.7c5.8 2.9 10.4 7.3 13.5 12.7h40.6c3.1-5.3 7.7-9.8 13.5-12.7l6.3 12.7h0c-6.3-12.7-6.3-12.7-6.3-12.7l0 0 0 0 .1 0 .2-.1 .6-.3c.5-.2 1.2-.6 2.1-1c1.8-.8 4.2-1.9 7.3-3.2c6.1-2.6 14.8-5.9 25.4-9.2c21-6.6 50.4-13.5 82.8-13.5c30.4 0 58.2 6.1 78.7 12.3L481.4 99.1c-2.6-11.2-12.6-19.1-24-19.1c-3.1 0-6.2 .6-9.2 1.8L416.9 94.3c-12.3 4.9-26.3-1.1-31.2-13.4s1.1-26.3 13.4-31.2l31.3-12.5c8.6-3.4 17.7-5.2 27-5.2c33.8 0 63.1 23.3 70.8 56.2l43.9 188c1.7 7.3 2.9 14.7 3.5 22.1c.3 1.9 .5 3.8 .5 5.7v6.7V352v16c0 61.9-50.1 112-112 112H419.7c-59.4 0-108.5-46.4-111.8-105.8L306.6 352H269.4l-1.2 22.2C264.9 433.6 215.8 480 156.3 480H112C50.1 480 0 429.9 0 368V352 310.7 304c0-1.9 .2-3.8 .5-5.7c.6-7.4 1.8-14.8 3.5-22.1l43.9-188C55.5 55.3 84.8 32 118.6 32c9.2 0 18.4 1.8 27 5.2l31.3 12.5c12.3 4.9 18.3 18.9 13.4 31.2s-18.9 18.3-31.2 13.4L127.8 81.8c-2.9-1.2-6-1.8-9.2-1.8zM64 325.4V368c0 26.5 21.5 48 48 48h44.3c25.5 0 46.5-19.9 47.9-45.3l2.5-45.6c-2.3-.8-4.9-1.7-7.5-2.5c-17.2-5.4-39.9-10.5-63.6-10.5c-23.7 0-46.2 5.1-63.2 10.5c-3.1 1-5.9 1.9-8.5 2.9zM512 368V325.4c-2.6-.9-5.5-1.9-8.5-2.9c-17-5.4-39.5-10.5-63.2-10.5c-23.7 0-46.4 5.1-63.6 10.5c-2.7 .8-5.2 1.7-7.5 2.5l2.5 45.6c1.4 25.4 22.5 45.3 47.9 45.3H464c26.5 0 48-21.5 48-48z\"],\"chess-board\":[448,512,[],\"f43c\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm64 64v64h64V96h64v64h64V96h64v64H320v64h64v64H320v64h64v64H320V352H256v64H192V352H128v64H64V352h64V288H64V224h64V160H64V96h64zm64 128h64V160H192v64zm0 64V224H128v64h64zm64 0H192v64h64V288zm0 0h64V224H256v64z\"],\"building-circle-check\":[640,512,[],\"e4d2\",\"M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c15.1 0 28.5-6.9 37.3-17.8C340.4 462.2 320 417.5 320 368c0-54.7 24.9-103.5 64-135.8V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zM640 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z\"],\"person-chalkboard\":[640,512,[],\"e53d\",\"M192 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-8 384V352h16V480c0 17.7 14.3 32 32 32s32-14.3 32-32V192h56 64 16c17.7 0 32-14.3 32-32s-14.3-32-32-32H384V64H576V256H384V224H320v48c0 26.5 21.5 48 48 48H592c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H368c-26.5 0-48 21.5-48 48v80H243.1 177.1c-33.7 0-64.9 17.7-82.3 46.6l-58.3 97c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9L120 256.9V480c0 17.7 14.3 32 32 32s32-14.3 32-32z\"],\"mars-stroke-right\":[640,512,[9897,\"mars-stroke-h\"],\"f22b\",\"M208 368a112 112 0 1 0 0-224 112 112 0 1 0 0 224zm174.4-88C370.7 365.8 297.1 432 208 432c-97.2 0-176-78.8-176-176s78.8-176 176-176c89.1 0 162.7 66.2 174.4 152H416V176c0-13.3 10.7-24 24-24s24 10.7 24 24v56h32V176c0-9.7 5.8-18.5 14.8-22.2s19.3-1.7 26.2 5.2l80 80c9.4 9.4 9.4 24.6 0 33.9l-80 80c-6.9 6.9-17.2 8.9-26.2 5.2s-14.8-12.5-14.8-22.2V280H464v56c0 13.3-10.7 24-24 24s-24-10.7-24-24V280H382.4z\"],\"hand-back-fist\":[448,512,[\"hand-rock\"],\"f255\",\"M144 0C117.5 0 96 21.5 96 48V96v28.5V176c0 8.8-7.2 16-16 16s-16-7.2-16-16V149.3l-9 7.5C40.4 169 32 187 32 206V244c0 38 16.9 74 46.1 98.3L128 384v96c0 17.7 14.3 32 32 32H320c17.7 0 32-14.3 32-32V374.7c46.9-19 80-65 80-118.7V176 160 144c0-26.5-21.5-48-48-48c-12.4 0-23.6 4.7-32.1 12.3C350 83.5 329.3 64 304 64c-12.4 0-23.6 4.7-32.1 12.3C270 51.5 249.3 32 224 32c-12.4 0-23.6 4.7-32.1 12.3C190 19.5 169.3 0 144 0z\"],\"square-caret-up\":[448,512,[\"caret-square-up\"],\"f151\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM224 160c6.7 0 13 2.8 17.6 7.7l104 112c6.5 7 8.2 17.2 4.4 25.9s-12.5 14.4-22 14.4H120c-9.5 0-18.2-5.7-22-14.4s-2.1-18.9 4.4-25.9l104-112c4.5-4.9 10.9-7.7 17.6-7.7z\"],\"cloud-showers-water\":[576,512,[],\"e4e4\",\"M224 0c38.6 0 71.9 22.8 87.2 55.7C325.7 41.1 345.8 32 368 32c38.7 0 71 27.5 78.4 64H448c35.3 0 64 28.7 64 64s-28.7 64-64 64H128c-35.3 0-64-28.7-64-64s28.7-64 64-64c0-53 43-96 96-96zM140.6 292.3l-48 80c-6.8 11.4-21.6 15-32.9 8.2s-15.1-21.6-8.2-32.9l48-80c6.8-11.4 21.6-15.1 32.9-8.2s15.1 21.6 8.2 32.9zm327.8-32.9c11.4 6.8 15 21.6 8.2 32.9l-48 80c-6.8 11.4-21.6 15-32.9 8.2s-15-21.6-8.2-32.9l48-80c6.8-11.4 21.6-15.1 32.9-8.2zM252.6 292.3l-48 80c-6.8 11.4-21.6 15-32.9 8.2s-15.1-21.6-8.2-32.9l48-80c6.8-11.4 21.6-15.1 32.9-8.2s15.1 21.6 8.2 32.9zm103.8-32.9c11.4 6.8 15 21.6 8.2 32.9l-48 80c-6.8 11.4-21.6 15-32.9 8.2s-15.1-21.6-8.2-32.9l48-80c6.8-11.4 21.6-15.1 32.9-8.2zM306.5 421.9C329 437.4 356.5 448 384 448c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 501.7 417 512 384 512c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 437.2 165.1 448 192 448c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z\"],\"chart-bar\":[512,512,[\"bar-chart\"],\"f080\",\"M32 32c17.7 0 32 14.3 32 32V400c0 8.8 7.2 16 16 16H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H80c-44.2 0-80-35.8-80-80V64C0 46.3 14.3 32 32 32zm96 96c0-17.7 14.3-32 32-32l192 0c17.7 0 32 14.3 32 32s-14.3 32-32 32l-192 0c-17.7 0-32-14.3-32-32zm32 64H288c17.7 0 32 14.3 32 32s-14.3 32-32 32H160c-17.7 0-32-14.3-32-32s14.3-32 32-32zm0 96H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H160c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],\"hands-bubbles\":[576,512,[\"hands-wash\"],\"e05e\",\"M416 64a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm96 128a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM160 464a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM32 160l.1 72.6c.1 52.2 24 101 64 133.1c-.1-1.9-.1-3.8-.1-5.7v-8c0-71.8 37-138.6 97.9-176.7l60.2-37.6c8.6-5.4 17.9-8.4 27.3-9.4l45.9-79.5c6.6-11.5 2.7-26.2-8.8-32.8s-26.2-2.7-32.8 8.8l-78 135.1c-3.3 5.7-10.7 7.7-16.4 4.4s-7.7-10.7-4.4-16.4l62-107.4c6.6-11.5 2.7-26.2-8.8-32.8S214 5 207.4 16.5l-68 117.8 0 0 0 0-43.3 75L96 160c0-17.7-14.4-32-32-32s-32 14.4-32 32zM332.1 88.5L307.5 131c13.9 4.5 26.4 13.7 34.7 27c.9 1.5 1.8 2.9 2.5 4.4l28.9-50c6.6-11.5 2.7-26.2-8.8-32.8s-26.2-2.7-32.8 8.8zm46.4 63.7l-26.8 46.4c-.6 6-2.1 11.8-4.3 17.4H352h13.3l0 0H397l23-39.8c6.6-11.5 2.7-26.2-8.8-32.8s-26.2-2.7-32.8 8.8zM315.1 175c-9.4-15-29.1-19.5-44.1-10.2l-60.2 37.6C159.3 234.7 128 291.2 128 352v8c0 8.9 .8 17.6 2.2 26.1c35.4 8.2 61.8 40 61.8 77.9c0 6.3-.7 12.5-2.1 18.4C215.1 501 246.3 512 280 512H456c13.3 0 24-10.7 24-24s-10.7-24-24-24H364c-6.6 0-12-5.4-12-12s5.4-12 12-12H488c13.3 0 24-10.7 24-24s-10.7-24-24-24H364c-6.6 0-12-5.4-12-12s5.4-12 12-12H520c13.3 0 24-10.7 24-24s-10.7-24-24-24H364c-6.6 0-12-5.4-12-12s5.4-12 12-12H488c13.3 0 24-10.7 24-24s-10.7-24-24-24H352l0 0 0 0H258.8L305 219.1c15-9.4 19.5-29.1 10.2-44.1z\"],\"less-than-equal\":[448,512,[],\"f537\",\"M395.9 93.7c16.4-6.6 24.4-25.2 17.8-41.6s-25.2-24.4-41.6-17.8l-320 128C40 167.1 32 178.9 32 192s8 24.9 20.1 29.7l320 128c16.4 6.6 35-1.4 41.6-17.8s-1.4-35-17.8-41.6L150.2 192 395.9 93.7zM32 416c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H32z\"],train:[448,512,[128646],\"f238\",\"M96 0C43 0 0 43 0 96V352c0 48 35.2 87.7 81.1 94.9l-46 46C28.1 499.9 33.1 512 43 512H82.7c8.5 0 16.6-3.4 22.6-9.4L160 448H288l54.6 54.6c6 6 14.1 9.4 22.6 9.4H405c10 0 15-12.1 7.9-19.1l-46-46c46-7.1 81.1-46.9 81.1-94.9V96c0-53-43-96-96-96H96zM64 96c0-17.7 14.3-32 32-32H352c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V96zM224 288a48 48 0 1 1 0 96 48 48 0 1 1 0-96z\"],\"eye-low-vision\":[640,512,[\"low-vision\"],\"f2a8\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L525.6 386.7c39.6-40.6 66.4-86.1 79.9-118.4c3.3-7.9 3.3-16.7 0-24.6c-14.9-35.7-46.2-87.7-93-131.1C465.5 68.8 400.8 32 320 32c-68.2 0-125 26.3-169.3 60.8L38.8 5.1zM223 149.5c48.6-44.3 123-50.8 179.3-11.7c60.8 42.4 78.9 123.2 44.2 186.9L408 294.5c8.4-19.3 10.6-41.4 4.8-63.3c-11.1-41.5-47.8-69.4-88.6-71.1c-5.8-.2-9.2 6.1-7.4 11.7c2.1 6.4 3.3 13.2 3.3 20.3c0 10.2-2.4 19.8-6.6 28.3L223 149.5zm-139.9 12c-11 14.4-20.5 28.7-28.4 42.2l339 265.7c18.7-5.5 36.2-13 52.6-21.8L83.1 161.5zm-50 86.3c-1.8 6.8-1.3 14 1.4 20.5c14.9 35.7 46.2 87.7 93 131.1C174.5 443.2 239.2 480 320 480c3.1 0 6.1-.1 9.2-.2L33.1 247.8z\"],crow:[640,512,[],\"f520\",\"M456 0c-48.6 0-88 39.4-88 88v29.2L12.5 390.6c-14 10.8-16.6 30.9-5.9 44.9s30.9 16.6 44.9 5.9L126.1 384H259.2l46.6 113.1c5 12.3 19.1 18.1 31.3 13.1s18.1-19.1 13.1-31.3L311.1 384H352c1.1 0 2.1 0 3.2 0l46.6 113.2c5 12.3 19.1 18.1 31.3 13.1s18.1-19.1 13.1-31.3l-42-102C484.9 354.1 544 280 544 192V128v-8l80.5-20.1c8.6-2.1 13.8-10.8 11.6-19.4C629 52 603.4 32 574 32H523.9C507.7 12.5 483.3 0 456 0zm0 64a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],sailboat:[576,512,[],\"e445\",\"M256 16c0-7 4.5-13.2 11.2-15.3s13.9 .4 17.9 6.1l224 320c3.4 4.9 3.8 11.3 1.1 16.6s-8.2 8.6-14.2 8.6H272c-8.8 0-16-7.2-16-16V16zM212.1 96.5c7 1.9 11.9 8.2 11.9 15.5V336c0 8.8-7.2 16-16 16H80c-5.7 0-11-3-13.8-8s-2.9-11-.1-16l128-224c3.6-6.3 11-9.4 18-7.5zM5.7 404.3C2.8 394.1 10.5 384 21.1 384H554.9c10.6 0 18.3 10.1 15.4 20.3l-4 14.3C550.7 473.9 500.4 512 443 512H133C75.6 512 25.3 473.9 9.7 418.7l-4-14.3z\"],\"window-restore\":[512,512,[],\"f2d2\",\"M432 64H208c-8.8 0-16 7.2-16 16V96H128V80c0-44.2 35.8-80 80-80H432c44.2 0 80 35.8 80 80V304c0 44.2-35.8 80-80 80H416V320h16c8.8 0 16-7.2 16-16V80c0-8.8-7.2-16-16-16zM0 192c0-35.3 28.7-64 64-64H320c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V192zm64 32c0 17.7 14.3 32 32 32H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H96c-17.7 0-32 14.3-32 32z\"],\"square-plus\":[448,512,[61846,\"plus-square\"],\"f0fe\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM200 344V280H136c-13.3 0-24-10.7-24-24s10.7-24 24-24h64V168c0-13.3 10.7-24 24-24s24 10.7 24 24v64h64c13.3 0 24 10.7 24 24s-10.7 24-24 24H248v64c0 13.3-10.7 24-24 24s-24-10.7-24-24z\"],\"torii-gate\":[512,512,[9961],\"f6a1\",\"M0 80c0 26.5 21.5 48 48 48H64v64h64V128h96v64h64V128h96v64h64V128h16c26.5 0 48-21.5 48-48V13.4C512 6 506 0 498.6 0c-1.7 0-3.4 .3-5 1l-49 19.6C425.7 28.1 405.5 32 385.2 32H126.8c-20.4 0-40.5-3.9-59.4-11.4L18.4 1c-1.6-.6-3.3-1-5-1C6 0 0 6 0 13.4V80zM64 288V480c0 17.7 14.3 32 32 32s32-14.3 32-32V288H384V480c0 17.7 14.3 32 32 32s32-14.3 32-32V288h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64z\"],frog:[576,512,[],\"f52e\",\"M368 32c41.7 0 75.9 31.8 79.7 72.5l85.6 26.3c25.4 7.8 42.8 31.3 42.8 57.9c0 21.8-11.7 41.9-30.7 52.7L400.8 323.5 493.3 416H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H480c-8.5 0-16.6-3.4-22.6-9.4L346.9 360.2c11.7-36 3.2-77.1-25.4-105.7c-40.6-40.6-106.3-40.6-146.9-.1L101 324.4c-6.4 6.1-6.7 16.2-.6 22.6s16.2 6.6 22.6 .6l73.8-70.2 .1-.1 .1-.1c3.5-3.5 7.3-6.6 11.3-9.2c27.9-18.5 65.9-15.4 90.5 9.2c24.7 24.7 27.7 62.9 9 90.9c-2.6 3.8-5.6 7.5-9 10.9L261.8 416H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H64c-35.3 0-64-28.7-64-64C0 249.6 127 112.9 289.3 97.5C296.2 60.2 328.8 32 368 32zm0 104a24 24 0 1 0 0-48 24 24 0 1 0 0 48z\"],bucket:[448,512,[],\"e4cf\",\"M96 152v8H48v-8C48 68.1 116.1 0 200 0h48c83.9 0 152 68.1 152 152v8H352v-8c0-57.4-46.6-104-104-104H200C142.6 48 96 94.6 96 152zM0 224c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32h-5.1L388.5 469c-2.6 24.4-23.2 43-47.7 43H107.2c-24.6 0-45.2-18.5-47.7-43L37.1 256H32c-17.7 0-32-14.3-32-32z\"],image:[512,512,[],\"f03e\",\"M0 96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM323.8 202.5c-4.5-6.6-11.9-10.5-19.8-10.5s-15.4 3.9-19.8 10.5l-87 127.6L170.7 297c-4.6-5.7-11.5-9-18.7-9s-14.2 3.3-18.7 9l-64 80c-5.8 7.2-6.9 17.1-2.9 25.4s12.4 13.6 21.6 13.6h96 32H424c8.9 0 17.1-4.9 21.2-12.8s3.6-17.4-1.4-24.7l-120-176zM112 192a48 48 0 1 0 0-96 48 48 0 1 0 0 96z\"],microphone:[384,512,[],\"f130\",\"M192 0C139 0 96 43 96 96V256c0 53 43 96 96 96s96-43 96-96V96c0-53-43-96-96-96zM64 216c0-13.3-10.7-24-24-24s-24 10.7-24 24v40c0 89.1 66.2 162.7 152 174.4V464H120c-13.3 0-24 10.7-24 24s10.7 24 24 24h72 72c13.3 0 24-10.7 24-24s-10.7-24-24-24H216V430.4c85.8-11.7 152-85.3 152-174.4V216c0-13.3-10.7-24-24-24s-24 10.7-24 24v40c0 70.7-57.3 128-128 128s-128-57.3-128-128V216z\"],cow:[640,512,[128004],\"f6c8\",\"M96 224v32V416c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V327.8c9.9 6.6 20.6 12 32 16.1V368c0 8.8 7.2 16 16 16s16-7.2 16-16V351.1c5.3 .6 10.6 .9 16 .9s10.7-.3 16-.9V368c0 8.8 7.2 16 16 16s16-7.2 16-16V343.8c11.4-4 22.1-9.4 32-16.1V416c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V256l32 32v49.5c0 9.5 2.8 18.7 8.1 26.6L530 427c8.8 13.1 23.5 21 39.3 21c22.5 0 41.9-15.9 46.3-38l20.3-101.6c2.6-13-.3-26.5-8-37.3l-3.9-5.5V184c0-13.3-10.7-24-24-24s-24 10.7-24 24v14.4l-52.9-74.1C496 86.5 452.4 64 405.9 64H272 256 192 144C77.7 64 24 117.7 24 184v54C9.4 249.8 0 267.8 0 288v17.6c0 8 6.4 14.4 14.4 14.4C46.2 320 72 294.2 72 262.4V256 224 184c0-24.3 12.1-45.8 30.5-58.9C98.3 135.9 96 147.7 96 160v64zM560 336a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zM166.6 166.6c-4.2-4.2-6.6-10-6.6-16c0-12.5 10.1-22.6 22.6-22.6H361.4c12.5 0 22.6 10.1 22.6 22.6c0 6-2.4 11.8-6.6 16l-23.4 23.4C332.2 211.8 302.7 224 272 224s-60.2-12.2-81.9-33.9l-23.4-23.4z\"],\"caret-up\":[320,512,[],\"f0d8\",\"M182.6 137.4c-12.5-12.5-32.8-12.5-45.3 0l-128 128c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8H288c12.9 0 24.6-7.8 29.6-19.8s2.2-25.7-6.9-34.9l-128-128z\"],screwdriver:[512,512,[129691],\"f54a\",\"M465 7c-8.5-8.5-22-9.4-31.6-2.1l-104 80c-5.9 4.5-9.4 11.6-9.4 19v54.1l-85.6 85.6c6.7 4.2 13 9.3 18.8 15.1s10.9 12.2 15.1 18.8L353.9 192H408c7.5 0 14.5-3.5 19-9.4l80-104c7.4-9.6 6.5-23.1-2.1-31.6L465 7zM121.4 281.4l-112 112c-12.5 12.5-12.5 32.8 0 45.3l64 64c12.5 12.5 32.8 12.5 45.3 0l112-112c30.2-30.2 30.2-79.1 0-109.3s-79.1-30.2-109.3 0z\"],\"folder-closed\":[512,512,[],\"e185\",\"M448 480H64c-35.3 0-64-28.7-64-64V192H512V416c0 35.3-28.7 64-64 64zm64-320H0V96C0 60.7 28.7 32 64 32H192c20.1 0 39.1 9.5 51.2 25.6l19.2 25.6c6 8.1 15.5 12.8 25.6 12.8H448c35.3 0 64 28.7 64 64z\"],\"house-tsunami\":[576,512,[],\"e515\",\"M80.8 136.5C104.9 93.8 152.6 64 209 64c16.9 0 33.1 2.7 48.2 7.7c16.8 5.5 34.9-3.6 40.4-20.4s-3.6-34.9-20.4-40.4C255.8 3.8 232.8 0 209 0C95.2 0 0 88 0 200c0 91.6 53.5 172.1 142.2 194.1c13.4 3.8 27.5 5.9 42.2 5.9c.7 0 1.4 0 2.1-.1c1.8 0 3.7 .1 5.5 .1l0 0c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.5-27.3-10.1-39.2-1.7l0 0C439.4 325.2 410.9 336 384 336c-27.5 0-55-10.6-77.5-26.1c-11.1-7.9-25.9-7.9-37 0c-22.4 15.5-49.9 26.1-77.4 26.1c0 0-.1 0-.1 0c-12.4 0-24-1.5-34.9-4.3C121.6 320.2 96 287 96 248c0-48.5 39.5-88 88.4-88c13.5 0 26.1 3 37.5 8.3c16 7.5 35.1 .6 42.5-15.5s.6-35.1-15.5-42.5C229.3 101.1 207.4 96 184.4 96c-40 0-76.4 15.4-103.6 40.5zm252-18.1c-8.1 6-12.8 15.5-12.8 25.6V265c1.6 1 3.3 2 4.8 3.1c18.4 12.7 39.6 20.3 59.2 20.3c19 0 41.2-7.9 59.2-20.3c23.8-16.7 55.8-15.3 78.1 3.4c10.6 8.8 24.2 15.6 37.3 18.6c5.8 1.4 11.2 3.4 16.2 6.2c.7-2.7 1.1-5.5 1.1-8.4l-.4-144c0-10-4.7-19.4-12.7-25.5l-95.5-72c-11.4-8.6-27.1-8.6-38.5 0l-96 72zM384 448c-27.5 0-55-10.6-77.5-26.1c-11.1-7.9-25.9-7.9-37 0C247 437.4 219.5 448 192 448c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C126.5 501.7 159 512 192 512c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.4-27.3-10.1-39.2-1.7l0 0C439.4 437.2 410.9 448 384 448z\"],\"square-nfi\":[448,512,[],\"e576\",\"M0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm75.7 64.6C68.8 162.5 64 168.8 64 176V336c0 8.8 7.2 16 16 16s16-7.2 16-16V233.8l66.3 110.5c3.7 6.2 11.1 9.1 18 7.2s11.7-8.2 11.7-15.4V176c0-8.8-7.2-16-16-16s-16 7.2-16 16V278.2L93.7 167.8c-3.7-6.2-11.1-9.1-18-7.2zM224 176v64 96c0 8.8 7.2 16 16 16s16-7.2 16-16V256h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H256V192h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H240c-8.8 0-16 7.2-16 16zm160 0c0-8.8-7.2-16-16-16s-16 7.2-16 16V336c0 8.8 7.2 16 16 16s16-7.2 16-16V176z\"],\"arrow-up-from-ground-water\":[576,512,[],\"e4b5\",\"M288 352c17.7 0 32-14.3 32-32V109.3l25.4 25.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-80-80c-12.5-12.5-32.8-12.5-45.3 0l-80 80c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L256 109.3V320c0 17.7 14.3 32 32 32zm-18.5 69.9C247 437.4 219.5 448 192 448c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C126.5 501.7 159 512 192 512c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.4-27.3-10.1-39.2-1.7l0 0C439.4 437.2 410.9 448 384 448c-27.5 0-55-10.6-77.5-26.1c-11.1-7.9-25.9-7.9-37 0zM192 192H48c-26.5 0-48 21.5-48 48V425c5.3-3.1 11.2-5.4 17.5-6.9c13.1-3.1 26.7-9.8 37.3-18.6c22.2-18.7 54.3-20.1 78.1-3.4c18 12.4 40.1 20.3 59.1 20.3V192zm384 48c0-26.5-21.5-48-48-48H384V416.5h0c19 0 41.2-7.9 59.2-20.3c23.8-16.7 55.8-15.3 78.1 3.4c10.6 8.8 24.2 15.6 37.3 18.6c6.3 1.5 12.1 3.8 17.5 6.9V240z\"],\"martini-glass\":[512,512,[127864,\"glass-martini-alt\"],\"f57b\",\"M32 0C19.1 0 7.4 7.8 2.4 19.8s-2.2 25.7 6.9 34.9L224 269.3V448H160c-17.7 0-32 14.3-32 32s14.3 32 32 32h96 96c17.7 0 32-14.3 32-32s-14.3-32-32-32H288V269.3L502.6 54.6c9.2-9.2 11.9-22.9 6.9-34.9S492.9 0 480 0H32zM173.3 128l-64-64H402.7l-64 64H173.3z\"],\"rotate-left\":[512,512,[\"rotate-back\",\"rotate-backward\",\"undo-alt\"],\"f2ea\",\"M48.5 224H40c-13.3 0-24-10.7-24-24V72c0-9.7 5.8-18.5 14.8-22.2s19.3-1.7 26.2 5.2L98.6 96.6c87.6-86.5 228.7-86.2 315.8 1c87.5 87.5 87.5 229.3 0 316.8s-229.3 87.5-316.8 0c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0c62.5 62.5 163.8 62.5 226.3 0s62.5-163.8 0-226.3c-62.2-62.2-162.7-62.5-225.3-1L185 183c6.9 6.9 8.9 17.2 5.2 26.2s-12.5 14.8-22.2 14.8H48.5z\"],\"table-columns\":[512,512,[\"columns\"],\"f0db\",\"M0 96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm64 64V416H224V160H64zm384 0H288V416H448V160z\"],lemon:[448,512,[127819],\"f094\",\"M448 96c0-35.3-28.7-64-64-64c-6.6 0-13 1-19 2.9c-22.5 7-48.1 14.9-71 9c-75.2-19.1-156.4 11-213.7 68.3S-7.2 250.8 11.9 326c5.8 22.9-2 48.4-9 71C1 403 0 409.4 0 416c0 35.3 28.7 64 64 64c6.6 0 13-1 19.1-2.9c22.5-7 48.1-14.9 71-9c75.2 19.1 156.4-11 213.7-68.3s87.5-138.5 68.3-213.7c-5.8-22.9 2-48.4 9-71c1.9-6 2.9-12.4 2.9-19.1zM212.5 127.4c-54.6 16-101.1 62.5-117.1 117.1C92.9 253 84 257.8 75.5 255.4S62.2 244 64.6 235.5c19.1-65.1 73.7-119.8 138.9-138.9c8.5-2.5 17.4 2.4 19.9 10.9s-2.4 17.4-10.9 19.9z\"],\"head-side-mask\":[576,512,[],\"e063\",\"M32 224.2c0-22.2 3.2-43.6 9.2-63.9L262.2 321c-4 9.5-6.2 20-6.2 31V512H128c-17.7 0-32-14.3-32-32V407.3c0-16.7-6.9-32.5-17.1-45.8C48.6 322.4 32 274.1 32 224.2zm248.3 70.4L53 129.3C88.7 53 166.2 0 256 0h24c95.2 0 181.2 69.3 197.3 160.2c2.3 13 6.8 25.7 15.1 36l42 52.6c5.4 6.7 8.6 14.8 9.4 23.2H336c-21.7 0-41.3 8.6-55.7 22.6zM336 304H534l0 0h10l-19.7 64H368c-8.8 0-16 7.2-16 16s7.2 16 16 16H514.5l-9.8 32H368c-8.8 0-16 7.2-16 16s7.2 16 16 16H494.8l-.9 2.8c-8.3 26.9-33.1 45.2-61.2 45.2H288V352c0-14 6-26.7 15.6-35.4c0 0 0 0 0 0c8.5-7.8 19.9-12.6 32.4-12.6zm48-80a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],handshake:[640,512,[],\"f2b5\",\"M323.4 85.2l-96.8 78.4c-16.1 13-19.2 36.4-7 53.1c12.9 17.8 38 21.3 55.3 7.8l99.3-77.2c7-5.4 17-4.2 22.5 2.8s4.2 17-2.8 22.5l-20.9 16.2L512 316.8V128h-.7l-3.9-2.5L434.8 79c-15.3-9.8-33.2-15-51.4-15c-21.8 0-43 7.5-60 21.2zm22.8 124.4l-51.7 40.2C263 274.4 217.3 268 193.7 235.6c-22.2-30.5-16.6-73.1 12.7-96.8l83.2-67.3c-11.6-4.9-24.1-7.4-36.8-7.4C234 64 215.7 69.6 200 80l-72 48V352h28.2l91.4 83.4c19.6 17.9 49.9 16.5 67.8-3.1c5.5-6.1 9.2-13.2 11.1-20.6l17 15.6c19.5 17.9 49.9 16.6 67.8-2.9c4.5-4.9 7.8-10.6 9.9-16.5c19.4 13 45.8 10.3 62.1-7.5c17.9-19.5 16.6-49.9-2.9-67.8l-134.2-123zM16 128c-8.8 0-16 7.2-16 16V352c0 17.7 14.3 32 32 32H64c17.7 0 32-14.3 32-32V128H16zM48 320a16 16 0 1 1 0 32 16 16 0 1 1 0-32zM544 128V352c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V144c0-8.8-7.2-16-16-16H544zm32 208a16 16 0 1 1 32 0 16 16 0 1 1 -32 0z\"],gem:[512,512,[128142],\"f3a5\",\"M116.7 33.8c4.5-6.1 11.7-9.8 19.3-9.8H376c7.6 0 14.8 3.6 19.3 9.8l112 152c6.8 9.2 6.1 21.9-1.5 30.4l-232 256c-4.5 5-11 7.9-17.8 7.9s-13.2-2.9-17.8-7.9l-232-256c-7.7-8.5-8.3-21.2-1.5-30.4l112-152zm38.5 39.8c-3.3 2.5-4.2 7-2.1 10.5l57.4 95.6L63.3 192c-4.1 .3-7.3 3.8-7.3 8s3.2 7.6 7.3 8l192 16c.4 0 .9 0 1.3 0l192-16c4.1-.3 7.3-3.8 7.3-8s-3.2-7.6-7.3-8L301.5 179.8l57.4-95.6c2.1-3.5 1.2-8.1-2.1-10.5s-7.9-2-10.7 1L256 172.2 165.9 74.6c-2.8-3-7.4-3.4-10.7-1z\"],dolly:[576,512,[\"dolly-box\"],\"f472\",\"M0 32C0 14.3 14.3 0 32 0h72.9c27.5 0 52 17.6 60.7 43.8L257.7 320c30.1 .5 56.8 14.9 74 37l202.1-67.4c16.8-5.6 34.9 3.5 40.5 20.2s-3.5 34.9-20.2 40.5L352 417.7c-.9 52.2-43.5 94.3-96 94.3c-53 0-96-43-96-96c0-30.8 14.5-58.2 37-75.8L104.9 64H32C14.3 64 0 49.7 0 32zM244.8 134.5c-5.5-16.8 3.7-34.9 20.5-40.3L311 79.4l19.8 60.9 60.9-19.8L371.8 59.6l45.7-14.8c16.8-5.5 34.9 3.7 40.3 20.5l49.4 152.2c5.5 16.8-3.7 34.9-20.5 40.3L334.5 307.2c-16.8 5.5-34.9-3.7-40.3-20.5L244.8 134.5z\"],smoking:[640,512,[128684],\"f48d\",\"M448 32V43c0 38.2 15.2 74.8 42.2 101.8l21 21c21 21 32.8 49.5 32.8 79.2v11c0 17.7-14.3 32-32 32s-32-14.3-32-32V245c0-12.7-5.1-24.9-14.1-33.9l-21-21C405.9 151.1 384 98.1 384 43V32c0-17.7 14.3-32 32-32s32 14.3 32 32zM576 256V245c0-38.2-15.2-74.8-42.2-101.8l-21-21c-21-21-32.8-49.5-32.8-79.2V32c0-17.7 14.3-32 32-32s32 14.3 32 32V43c0 12.7 5.1 24.9 14.1 33.9l21 21c39 39 60.9 91.9 60.9 147.1v11c0 17.7-14.3 32-32 32s-32-14.3-32-32zM0 416c0-35.3 28.7-64 64-64H416c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H64c-35.3 0-64-28.7-64-64V416zm224 0v32H384V416H224zm288-64c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384c0-17.7 14.3-32 32-32zm96 0c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384c0-17.7 14.3-32 32-32z\"],minimize:[512,512,[\"compress-arrows-alt\"],\"f78c\",\"M456 224H312c-13.3 0-24-10.7-24-24V56c0-9.7 5.8-18.5 14.8-22.2s19.3-1.7 26.2 5.2l40 40L442.3 5.7C446 2 450.9 0 456 0s10 2 13.7 5.7l36.7 36.7C510 46 512 50.9 512 56s-2 10-5.7 13.7L433 143l40 40c6.9 6.9 8.9 17.2 5.2 26.2s-12.5 14.8-22.2 14.8zm0 64c9.7 0 18.5 5.8 22.2 14.8s1.7 19.3-5.2 26.2l-40 40 73.4 73.4c3.6 3.6 5.7 8.5 5.7 13.7s-2 10-5.7 13.7l-36.7 36.7C466 510 461.1 512 456 512s-10-2-13.7-5.7L369 433l-40 40c-6.9 6.9-17.2 8.9-26.2 5.2s-14.8-12.5-14.8-22.2V312c0-13.3 10.7-24 24-24H456zm-256 0c13.3 0 24 10.7 24 24V456c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-40-40L69.7 506.3C66 510 61.1 512 56 512s-10-2-13.7-5.7L5.7 469.7C2 466 0 461.1 0 456s2-10 5.7-13.7L79 369 39 329c-6.9-6.9-8.9-17.2-5.2-26.2s12.5-14.8 22.2-14.8H200zM56 224c-9.7 0-18.5-5.8-22.2-14.8s-1.7-19.3 5.2-26.2l40-40L5.7 69.7C2 66 0 61.1 0 56s2-10 5.7-13.7L42.3 5.7C46 2 50.9 0 56 0s10 2 13.7 5.7L143 79l40-40c6.9-6.9 17.2-8.9 26.2-5.2s14.8 12.5 14.8 22.2V200c0 13.3-10.7 24-24 24H56z\"],monument:[384,512,[],\"f5a6\",\"M180.7 4.7c6.2-6.2 16.4-6.2 22.6 0l80 80c2.5 2.5 4.1 5.8 4.6 9.3l40.2 322H55.9L96.1 94c.4-3.5 2-6.8 4.6-9.3l80-80zM152 272c-13.3 0-24 10.7-24 24s10.7 24 24 24h80c13.3 0 24-10.7 24-24s-10.7-24-24-24H152zM32 448H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],snowplow:[640,512,[],\"f7d2\",\"M298.9 64l68.6 160H256l-64-64V64H298.9zM445.1 242.7l-87.4-204C347.6 15.3 324.5 0 298.9 0H176c-26.5 0-48 21.5-48 48V160H96c-17.7 0-32 14.3-32 32V298.8C26.2 316.8 0 355.3 0 400c0 61.9 50.1 112 112 112H368c61.9 0 112-50.1 112-112c0-17.2-3.9-33.5-10.8-48H512v50.7c0 17 6.7 33.3 18.7 45.3l54.6 54.6c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L576 402.7V320 235.2L633 164c11-13.8 8.8-33.9-5-45s-33.9-8.8-45 5l-57 71.2c-9.1 11.3-14 25.4-14 40V288H448V256.7c.1-2.4-.2-4.8-.6-7.1s-1.2-4.7-2.2-6.8zM368 352c26.5 0 48 21.5 48 48s-21.5 48-48 48H112c-26.5 0-48-21.5-48-48s21.5-48 48-48H368zM144 400a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zm216 24a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm-56-24a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zM200 424a24 24 0 1 0 0-48 24 24 0 1 0 0 48z\"],\"angles-right\":[512,512,[187,\"angle-double-right\"],\"f101\",\"M470.6 278.6c12.5-12.5 12.5-32.8 0-45.3l-160-160c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L402.7 256 265.4 393.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l160-160zm-352 160l160-160c12.5-12.5 12.5-32.8 0-45.3l-160-160c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L210.7 256 73.4 393.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0z\"],cannabis:[512,512,[],\"f55f\",\"M256 0c5.3 0 10.3 2.7 13.3 7.1c15.8 23.5 36.7 63.7 49.2 109c7.2 26.4 11.8 55.2 10.4 84c11.5-8.8 23.7-16.7 35.8-23.6c41-23.3 84.4-36.9 112.2-42.5c5.2-1 10.7 .6 14.4 4.4s5.4 9.2 4.4 14.5c-5.6 27.7-19.3 70.9-42.7 111.7c-9.1 15.9-19.9 31.7-32.4 46.3c27.8 6.6 52.4 17.3 67.2 25.5c5.1 2.8 8.2 8.2 8.2 14s-3.2 11.2-8.2 14c-15.2 8.4-40.9 19.5-69.8 26.1c-20.2 4.6-42.9 7.2-65.2 4.6l8.3 33.1c1.5 6.1-.6 12.4-5.5 16.4s-11.6 4.6-17.2 1.9L280 417.2V488c0 13.3-10.7 24-24 24s-24-10.7-24-24V417.2l-58.5 29.1c-5.6 2.8-12.3 2.1-17.2-1.9s-7-10.3-5.5-16.4l8.3-33.1c-22.2 2.6-45 0-65.2-4.6c-28.9-6.6-54.6-17.6-69.8-26.1c-5.1-2.8-8.2-8.2-8.2-14s3.2-11.2 8.2-14c14.8-8.2 39.4-18.8 67.2-25.5C78.9 296.3 68.1 280.5 59 264.6c-23.4-40.8-37.1-84-42.7-111.7c-1.1-5.2 .6-10.7 4.4-14.5s9.2-5.4 14.4-4.4c27.9 5.5 71.2 19.2 112.2 42.5c12.1 6.9 24.3 14.7 35.8 23.6c-1.4-28.7 3.1-57.6 10.4-84c12.5-45.3 33.4-85.5 49.2-109c3-4.4 8-7.1 13.3-7.1z\"],\"circle-play\":[512,512,[61469,\"play-circle\"],\"f144\",\"M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM188.3 147.1c-7.6 4.2-12.3 12.3-12.3 20.9V344c0 8.7 4.7 16.7 12.3 20.9s16.8 4.1 24.3-.5l144-88c7.1-4.4 11.5-12.1 11.5-20.5s-4.4-16.1-11.5-20.5l-144-88c-7.4-4.5-16.7-4.7-24.3-.5z\"],tablets:[640,512,[],\"f490\",\"M614.3 247c16.3-25 25.7-54.9 25.7-87C640 71.6 568.4 0 480 0c-32.1 0-61.9 9.4-87 25.7c-7.9 5.2-8.5 16.2-1.8 22.9L591.4 248.8c6.7 6.7 17.8 6.2 22.9-1.8zM567 294.3c7.9-5.2 8.5-16.2 1.8-22.9L368.6 71.2c-6.7-6.7-17.8-6.2-22.9 1.8c-16.3 25-25.7 54.9-25.7 87c0 88.4 71.6 160 160 160c32.1 0 61.9-9.4 87-25.7zM301.5 368H18.5c-9.5 0-16.9 8.2-15 17.5C18.9 457.8 83.1 512 160 512s141.1-54.2 156.5-126.5c2-9.3-5.5-17.5-15-17.5zm0-32c9.5 0 16.9-8.2 15-17.5C301.1 246.2 236.9 192 160 192S18.9 246.2 3.5 318.5c-2 9.3 5.5 17.5 15 17.5H301.5z\"],ethernet:[512,512,[],\"f796\",\"M0 224V416c0 17.7 14.3 32 32 32H96V336c0-8.8 7.2-16 16-16s16 7.2 16 16V448h64V336c0-8.8 7.2-16 16-16s16 7.2 16 16V448h64V336c0-8.8 7.2-16 16-16s16 7.2 16 16V448h64V336c0-8.8 7.2-16 16-16s16 7.2 16 16V448h64c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32H448V160c0-17.7-14.3-32-32-32H384V96c0-17.7-14.3-32-32-32H160c-17.7 0-32 14.3-32 32v32H96c-17.7 0-32 14.3-32 32v32H32c-17.7 0-32 14.3-32 32z\"],\"euro-sign\":[320,512,[8364,\"eur\",\"euro\"],\"f153\",\"M48.1 240c-.1 2.7-.1 5.3-.1 8v16c0 2.7 0 5.3 .1 8H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H60.3C89.9 419.9 170 480 264 480h24c17.7 0 32-14.3 32-32s-14.3-32-32-32H264c-57.9 0-108.2-32.4-133.9-80H256c17.7 0 32-14.3 32-32s-14.3-32-32-32H112.2c-.1-2.6-.2-5.3-.2-8V248c0-2.7 .1-5.4 .2-8H256c17.7 0 32-14.3 32-32s-14.3-32-32-32H130.1c25.7-47.6 76-80 133.9-80h24c17.7 0 32-14.3 32-32s-14.3-32-32-32H264C170 32 89.9 92.1 60.3 176H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H48.1z\"],chair:[448,512,[129681],\"f6c0\",\"M248 48V256h48V58.7c23.9 13.8 40 39.7 40 69.3V256h48V128C384 57.3 326.7 0 256 0H192C121.3 0 64 57.3 64 128V256h48V128c0-29.6 16.1-55.5 40-69.3V256h48V48h48zM48 288c-12.1 0-23.2 6.8-28.6 17.7l-16 32c-5 9.9-4.4 21.7 1.4 31.1S20.9 384 32 384l0 96c0 17.7 14.3 32 32 32s32-14.3 32-32V384H352v96c0 17.7 14.3 32 32 32s32-14.3 32-32V384c11.1 0 21.4-5.7 27.2-15.2s6.4-21.2 1.4-31.1l-16-32C423.2 294.8 412.1 288 400 288H48z\"],\"circle-check\":[512,512,[61533,\"check-circle\"],\"f058\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM369 209L241 337c-9.4 9.4-24.6 9.4-33.9 0l-64-64c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47L335 175c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z\"],\"circle-stop\":[512,512,[62094,\"stop-circle\"],\"f28d\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM192 160H320c17.7 0 32 14.3 32 32V320c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32V192c0-17.7 14.3-32 32-32z\"],\"compass-drafting\":[512,512,[\"drafting-compass\"],\"f568\",\"M352 96c0 14.3-3.1 27.9-8.8 40.2L396 227.4c-23.7 25.3-54.2 44.1-88.5 53.6L256 192h0 0l-68 117.5c21.5 6.8 44.3 10.5 68.1 10.5c70.7 0 133.8-32.7 174.9-84c11.1-13.8 31.2-16 45-5s16 31.2 5 45C428.1 341.8 347 384 256 384c-35.4 0-69.4-6.4-100.7-18.1L98.7 463.7C94 471.8 87 478.4 78.6 482.6L23.2 510.3c-5 2.5-10.9 2.2-15.6-.7S0 501.5 0 496V440.6c0-8.4 2.2-16.7 6.5-24.1l60-103.7C53.7 301.6 41.8 289.3 31.2 276c-11.1-13.8-8.8-33.9 5-45s33.9-8.8 45 5c5.7 7.1 11.8 13.8 18.2 20.1l69.4-119.9c-5.6-12.2-8.8-25.8-8.8-40.2c0-53 43-96 96-96s96 43 96 96zm21 297.9c32.6-12.8 62.5-30.8 88.9-52.9l43.7 75.5c4.2 7.3 6.5 15.6 6.5 24.1V496c0 5.5-2.9 10.7-7.6 13.6s-10.6 3.2-15.6 .7l-55.4-27.7c-8.4-4.2-15.4-10.8-20.1-18.9L373 393.9zM256 128a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"plate-wheat\":[512,512,[],\"e55a\",\"M176 32c44.2 0 80 35.8 80 80v16c0 8.8-7.2 16-16 16c-44.2 0-80-35.8-80-80V48c0-8.8 7.2-16 16-16zM56 64h48c13.3 0 24 10.7 24 24s-10.7 24-24 24H56c-13.3 0-24-10.7-24-24s10.7-24 24-24zM24 136H136c13.3 0 24 10.7 24 24s-10.7 24-24 24H24c-13.3 0-24-10.7-24-24s10.7-24 24-24zm8 96c0-13.3 10.7-24 24-24h48c13.3 0 24 10.7 24 24s-10.7 24-24 24H56c-13.3 0-24-10.7-24-24zM272 48c0-8.8 7.2-16 16-16c44.2 0 80 35.8 80 80v16c0 8.8-7.2 16-16 16c-44.2 0-80-35.8-80-80V48zM400 32c44.2 0 80 35.8 80 80v16c0 8.8-7.2 16-16 16c-44.2 0-80-35.8-80-80V48c0-8.8 7.2-16 16-16zm80 160v16c0 44.2-35.8 80-80 80c-8.8 0-16-7.2-16-16V256c0-44.2 35.8-80 80-80c8.8 0 16 7.2 16 16zM352 176c8.8 0 16 7.2 16 16v16c0 44.2-35.8 80-80 80c-8.8 0-16-7.2-16-16V256c0-44.2 35.8-80 80-80zm-96 16v16c0 44.2-35.8 80-80 80c-8.8 0-16-7.2-16-16V256c0-44.2 35.8-80 80-80c8.8 0 16 7.2 16 16zM3.5 347.6C1.6 332.9 13 320 27.8 320H484.2c14.8 0 26.2 12.9 24.4 27.6C502.3 397.8 464.2 437 416 446v2c0 17.7-14.3 32-32 32H128c-17.7 0-32-14.3-32-32v-2c-48.2-9-86.3-48.2-92.5-98.4z\"],icicles:[512,512,[],\"f7ad\",\"M75.8 304.8L1 35.7c-.7-2.5-1-5-1-7.5C0 12.6 12.6 0 28.2 0H482.4C498.8 0 512 13.2 512 29.6c0 1.6-.1 3.3-.4 4.9L434.6 496.1c-1.5 9.2-9.5 15.9-18.8 15.9c-9.2 0-17.1-6.6-18.7-15.6L336 160 307.2 303.9c-1.9 9.3-10.1 16.1-19.6 16.1c-9.2 0-17.2-6.2-19.4-15.1L240 192 210.6 368.2c-1.5 9.1-9.4 15.8-18.6 15.8s-17.1-6.7-18.6-15.8L144 192 115.9 304.3c-2.3 9.2-10.6 15.7-20.1 15.7c-9.3 0-17.5-6.2-20-15.2z\"],\"person-shelter\":[512,512,[],\"e54f\",\"M271.9 4.2c-9.8-5.6-21.9-5.6-31.8 0l-224 128C6.2 137.9 0 148.5 0 160V480c0 17.7 14.3 32 32 32s32-14.3 32-32V178.6L256 68.9 448 178.6V480c0 17.7 14.3 32 32 32s32-14.3 32-32V160c0-11.5-6.2-22.1-16.1-27.8l-224-128zM256 208a40 40 0 1 0 0-80 40 40 0 1 0 0 80zm-8 280V400h16v88c0 13.3 10.7 24 24 24s24-10.7 24-24V313.5l26.9 49.9c6.3 11.7 20.8 16 32.5 9.8s16-20.8 9.8-32.5l-37.9-70.3c-15.3-28.5-45.1-46.3-77.5-46.3H246.2c-32.4 0-62.1 17.8-77.5 46.3l-37.9 70.3c-6.3 11.7-1.9 26.2 9.8 32.5s26.2 1.9 32.5-9.8L200 313.5V488c0 13.3 10.7 24 24 24s24-10.7 24-24z\"],neuter:[384,512,[9906],\"f22c\",\"M80 176a112 112 0 1 1 224 0A112 112 0 1 1 80 176zM224 349.1c81.9-15 144-86.8 144-173.1C368 78.8 289.2 0 192 0S16 78.8 16 176c0 86.3 62.1 158.1 144 173.1V480c0 17.7 14.3 32 32 32s32-14.3 32-32V349.1z\"],\"id-badge\":[384,512,[],\"f2c1\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zm96 320h64c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80zm-32-96a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM144 64h96c8.8 0 16 7.2 16 16s-7.2 16-16 16H144c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],marker:[512,512,[],\"f5a1\",\"M481 31C445.1-4.8 386.9-4.8 351 31l-15 15L322.9 33C294.8 4.9 249.2 4.9 221.1 33L135 119c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0L255 66.9c9.4-9.4 24.6-9.4 33.9 0L302.1 80 186.3 195.7 316.3 325.7 481 161c35.9-35.9 35.9-94.1 0-129.9zM293.7 348.3L163.7 218.3 99.5 282.5c-48 48-80.8 109.2-94.1 175.8l-5 25c-1.6 7.9 .9 16 6.6 21.7s13.8 8.1 21.7 6.6l25-5c66.6-13.3 127.8-46.1 175.8-94.1l64.2-64.2z\"],\"face-laugh-beam\":[512,512,[128513,\"laugh-beam\"],\"f59a\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM96.8 314.1c-3.8-13.7 7.4-26.1 21.6-26.1H393.6c14.2 0 25.5 12.4 21.6 26.1C396.2 382 332.1 432 256 432s-140.2-50-159.2-117.9zM217.6 212.8l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z\"],\"helicopter-symbol\":[512,512,[],\"e502\",\"M445.3 224H510C495.6 108.2 403.8 16.4 288 2V66.7C368.4 80.1 431.9 143.6 445.3 224zM510 288H445.3C431.9 368.4 368.4 431.9 288 445.4V510c115.8-14.4 207.6-106.2 222-222zM2 288C16.4 403.8 108.2 495.6 224 510V445.4C143.6 431.9 80.1 368.4 66.7 288H2zm0-64H66.7C80.1 143.6 143.6 80.1 224 66.7V2C108.2 16.4 16.4 108.2 2 224zm206-64c0-17.7-14.3-32-32-32s-32 14.3-32 32V352c0 17.7 14.3 32 32 32s32-14.3 32-32V288h96v64c0 17.7 14.3 32 32 32s32-14.3 32-32V160c0-17.7-14.3-32-32-32s-32 14.3-32 32v64H208V160z\"],\"universal-access\":[512,512,[],\"f29a\",\"M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm161.5-86.1c-12.2-5.2-26.3 .4-31.5 12.6s.4 26.3 12.6 31.5l11.9 5.1c17.3 7.4 35.2 12.9 53.6 16.3v50.1c0 4.3-.7 8.6-2.1 12.6l-28.7 86.1c-4.2 12.6 2.6 26.2 15.2 30.4s26.2-2.6 30.4-15.2l24.4-73.2c1.3-3.8 4.8-6.4 8.8-6.4s7.6 2.6 8.8 6.4l24.4 73.2c4.2 12.6 17.8 19.4 30.4 15.2s19.4-17.8 15.2-30.4l-28.7-86.1c-1.4-4.1-2.1-8.3-2.1-12.6V235.5c18.4-3.5 36.3-8.9 53.6-16.3l11.9-5.1c12.2-5.2 17.8-19.3 12.6-31.5s-19.3-17.8-31.5-12.6L338.7 175c-26.1 11.2-54.2 17-82.7 17s-56.5-5.8-82.7-17l-11.9-5.1zM256 160a40 40 0 1 0 0-80 40 40 0 1 0 0 80z\"],\"circle-chevron-up\":[512,512,[\"chevron-circle-up\"],\"f139\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM377 271c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-87-87-87 87c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9L239 167c9.4-9.4 24.6-9.4 33.9 0L377 271z\"],\"lari-sign\":[384,512,[],\"e1c8\",\"M144 32c17.7 0 32 14.3 32 32V96.7c5.3-.4 10.6-.7 16-.7s10.7 .2 16 .7V64c0-17.7 14.3-32 32-32s32 14.3 32 32v49.4c54.9 25.2 95.8 75.5 108.2 136.2c3.5 17.3-7.7 34.2-25 37.7s-34.2-7.7-37.7-25c-6.1-29.9-22.5-55.9-45.4-74.3V256c0 17.7-14.3 32-32 32s-32-14.3-32-32V161c-5.2-.7-10.6-1-16-1s-10.8 .3-16 1v95c0 17.7-14.3 32-32 32s-32-14.3-32-32V188.1C82.7 211.5 64 247.6 64 288c0 70.7 57.3 128 128 128H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H192 32c-17.7 0-32-14.3-32-32s14.3-32 32-32H48.9C18.5 382 0 337.2 0 288c0-77.5 45.9-144.3 112-174.6V64c0-17.7 14.3-32 32-32z\"],volcano:[512,512,[127755],\"f770\",\"M160 144c-35.3 0-64-28.7-64-64s28.7-64 64-64c15.7 0 30 5.6 41.2 15C212.4 12.4 232.7 0 256 0s43.6 12.4 54.8 31C322 21.6 336.3 16 352 16c35.3 0 64 28.7 64 64s-28.7 64-64 64c-14.7 0-28.3-5-39.1-13.3l-32 48C275.3 187 266 192 256 192s-19.3-5-24.9-13.3l-32-48C188.3 139 174.7 144 160 144zM144 352l48.4-24.2c10.2-5.1 21.6-7.8 33-7.8c19.6 0 38.4 7.8 52.2 21.6l32.5 32.5c6.3 6.3 14.9 9.9 23.8 9.9c11.3 0 21.8-5.6 28-15l9.7-14.6-58.9-66.3c-9.1-10.2-22.2-16.1-35.9-16.1H235.1c-13.7 0-26.8 5.9-35.9 16.1l-59.9 67.4L144 352zm19.4-95.8c18.2-20.5 44.3-32.2 71.8-32.2h41.8c27.4 0 53.5 11.7 71.8 32.2l150.2 169c8.5 9.5 13.2 21.9 13.2 34.7c0 28.8-23.4 52.2-52.2 52.2H52.2C23.4 512 0 488.6 0 459.8c0-12.8 4.7-25.1 13.2-34.7l150.2-169z\"],\"person-walking-dashed-line-arrow-right\":[640,512,[],\"e553\",\"M208 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM123.7 200.5c1-.4 1.9-.8 2.9-1.2l-16.9 63.5c-5.6 21.1-.1 43.6 14.7 59.7l70.7 77.1 22 88.1c4.3 17.1 21.7 27.6 38.8 23.3s27.6-21.7 23.3-38.8l-23-92.1c-1.9-7.8-5.8-14.9-11.2-20.8l-49.5-54 19.3-65.5 9.6 23c4.4 10.6 12.5 19.3 22.8 24.5l26.7 13.3c15.8 7.9 35 1.5 42.9-14.3s1.5-35-14.3-42.9L281 232.7l-15.3-36.8C248.5 154.8 208.3 128 163.7 128c-22.8 0-45.3 4.8-66.1 14l-8 3.5c-32.9 14.6-58.1 42.4-69.4 76.5l-2.6 7.8c-5.6 16.8 3.5 34.9 20.2 40.5s34.9-3.5 40.5-20.2l2.6-7.8c5.7-17.1 18.3-30.9 34.7-38.2l8-3.5zm-30 135.1L68.7 398 9.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L116.3 441c4.6-4.6 8.2-10.1 10.6-16.1l14.5-36.2-40.7-44.4c-2.5-2.7-4.8-5.6-7-8.6zM550.6 153.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L530.7 224H384c-17.7 0-32 14.3-32 32s14.3 32 32 32H530.7l-25.4 25.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l80-80c12.5-12.5 12.5-32.8 0-45.3l-80-80zM392 0c-13.3 0-24 10.7-24 24V72c0 13.3 10.7 24 24 24s24-10.7 24-24V24c0-13.3-10.7-24-24-24zm24 152c0-13.3-10.7-24-24-24s-24 10.7-24 24v16c0 13.3 10.7 24 24 24s24-10.7 24-24V152zM392 320c-13.3 0-24 10.7-24 24v16c0 13.3 10.7 24 24 24s24-10.7 24-24V344c0-13.3-10.7-24-24-24zm24 120c0-13.3-10.7-24-24-24s-24 10.7-24 24v48c0 13.3 10.7 24 24 24s24-10.7 24-24V440z\"],\"sterling-sign\":[320,512,[163,\"gbp\",\"pound-sign\"],\"f154\",\"M112 160.4c0-35.5 28.8-64.4 64.4-64.4c6.9 0 13.8 1.1 20.4 3.3l81.2 27.1c16.8 5.6 34.9-3.5 40.5-20.2s-3.5-34.9-20.2-40.5L217 38.6c-13.1-4.4-26.8-6.6-40.6-6.6C105.5 32 48 89.5 48 160.4V224H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H48v44.5c0 17.4-4.7 34.5-13.7 49.4L4.6 431.5c-5.9 9.9-6.1 22.2-.4 32.2S20.5 480 32 480H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H88.5l.7-1.1C104.1 390 112 361.5 112 332.5V288H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H112V160.4z\"],viruses:[640,512,[],\"e076\",\"M192 0c13.3 0 24 10.7 24 24V37.5c0 35.6 43.1 53.5 68.3 28.3l9.5-9.5c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-9.5 9.5C293 124.9 310.9 168 346.5 168H360c13.3 0 24 10.7 24 24s-10.7 24-24 24H346.5c-35.6 0-53.5 43.1-28.3 68.3l9.5 9.5c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-9.5-9.5C259.1 293 216 310.9 216 346.5V360c0 13.3-10.7 24-24 24s-24-10.7-24-24V346.5c0-35.6-43.1-53.5-68.3-28.3l-9.5 9.5c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l9.5-9.5C91 259.1 73.1 216 37.5 216H24c-13.3 0-24-10.7-24-24s10.7-24 24-24H37.5c35.6 0 53.5-43.1 28.3-68.3l-9.5-9.5c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l9.5 9.5C124.9 91 168 73.1 168 37.5V24c0-13.3 10.7-24 24-24zm48 224a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm-48-64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm320 80c0 33 39.9 49.5 63.2 26.2c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6C574.5 312.1 591 352 624 352c8.8 0 16 7.2 16 16s-7.2 16-16 16c-33 0-49.5 39.9-26.2 63.2c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0C551.9 446.5 512 463 512 496c0 8.8-7.2 16-16 16s-16-7.2-16-16c0-33-39.9-49.5-63.2-26.2c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6C417.5 423.9 401 384 368 384c-8.8 0-16-7.2-16-16s7.2-16 16-16c33 0 49.5-39.9 26.2-63.2c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0C440.1 289.5 480 273 480 240c0-8.8 7.2-16 16-16s16 7.2 16 16zm0 112a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z\"],\"square-person-confined\":[448,512,[],\"e577\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm96 112a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm80 104c0-30.9 25.1-56 56-56s56 25.1 56 56V350.1c0 36.4-29.5 65.9-65.9 65.9c-17.5 0-34.3-6.9-46.6-19.3L184.8 342l-28.1 56.3c-7.9 15.8-27.1 22.2-42.9 14.3s-22.2-27.1-14.3-42.9l48-96c4.6-9.2 13.3-15.6 23.5-17.3s20.5 1.7 27.8 9L240 306.7V248z\"],\"user-tie\":[448,512,[],\"f508\",\"M96 128a128 128 0 1 0 256 0A128 128 0 1 0 96 128zm94.5 200.2l18.6 31L175.8 483.1l-36-146.9c-2-8.1-9.8-13.4-17.9-11.3C51.9 342.4 0 405.8 0 481.3c0 17 13.8 30.7 30.7 30.7H162.5c0 0 0 0 .1 0H168 280h5.5c0 0 0 0 .1 0H417.3c17 0 30.7-13.8 30.7-30.7c0-75.5-51.9-138.9-121.9-156.4c-8.1-2-15.9 3.3-17.9 11.3l-36 146.9L238.9 359.2l18.6-31c6.4-10.7-1.3-24.2-13.7-24.2H224 204.3c-12.4 0-20.1 13.6-13.7 24.2z\"],\"arrow-down-long\":[384,512,[\"long-arrow-down\"],\"f175\",\"M169.4 502.6c12.5 12.5 32.8 12.5 45.3 0l128-128c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L224 402.7 224 32c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 370.7L86.6 329.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l128 128z\"],\"tent-arrow-down-to-line\":[640,512,[],\"e57e\",\"M241.8 111.9c8.9 9.9 8.1 25-1.8 33.9l-80 72c-9.1 8.2-23 8.2-32.1 0l-80-72c-9.9-8.9-10.7-24-1.8-33.9s24-10.7 33.9-1.8l39.9 36L120 24c0-13.3 10.7-24 24-24s24 10.7 24 24l0 122.1 39.9-36c9.9-8.9 25-8.1 33.9 1.8zm122.8 22.6c11.5-8.7 27.3-8.7 38.8 0l168 128c6.6 5 11 12.5 12.3 20.7l24 160 .7 4.7c17.5 .2 31.6 14.4 31.6 32c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H159.6l.7-4.7 24-160c1.2-8.2 5.6-15.7 12.3-20.7l168-128zM384 448h80L402.7 325.5c-1.7-3.4-5.1-5.5-8.8-5.5c-5.5 0-9.9 4.4-9.9 9.9V448z\"],certificate:[512,512,[],\"f0a3\",\"M211 7.3C205 1 196-1.4 187.6 .8s-14.9 8.9-17.1 17.3L154.7 80.6l-62-17.5c-8.4-2.4-17.4 0-23.5 6.1s-8.5 15.1-6.1 23.5l17.5 62L18.1 170.6c-8.4 2.1-15 8.7-17.3 17.1S1 205 7.3 211l46.2 45L7.3 301C1 307-1.4 316 .8 324.4s8.9 14.9 17.3 17.1l62.5 15.8-17.5 62c-2.4 8.4 0 17.4 6.1 23.5s15.1 8.5 23.5 6.1l62-17.5 15.8 62.5c2.1 8.4 8.7 15 17.1 17.3s17.3-.2 23.4-6.4l45-46.2 45 46.2c6.1 6.2 15 8.7 23.4 6.4s14.9-8.9 17.1-17.3l15.8-62.5 62 17.5c8.4 2.4 17.4 0 23.5-6.1s8.5-15.1 6.1-23.5l-17.5-62 62.5-15.8c8.4-2.1 15-8.7 17.3-17.1s-.2-17.4-6.4-23.4l-46.2-45 46.2-45c6.2-6.1 8.7-15 6.4-23.4s-8.9-14.9-17.3-17.1l-62.5-15.8 17.5-62c2.4-8.4 0-17.4-6.1-23.5s-15.1-8.5-23.5-6.1l-62 17.5L341.4 18.1c-2.1-8.4-8.7-15-17.1-17.3S307 1 301 7.3L256 53.5 211 7.3z\"],\"reply-all\":[576,512,[\"mail-reply-all\"],\"f122\",\"M209.4 39.5c-9.1-9.6-24.3-10-33.9-.9L33.8 173.2c-19.9 18.9-19.9 50.7 0 69.6L175.5 377.4c9.6 9.1 24.8 8.7 33.9-.9s8.7-24.8-.9-33.9L66.8 208 208.5 73.4c9.6-9.1 10-24.3 .9-33.9zM352 64c0-12.6-7.4-24.1-19-29.2s-25-3-34.4 5.4l-160 144c-6.7 6.1-10.6 14.7-10.6 23.8s3.9 17.7 10.6 23.8l160 144c9.4 8.5 22.9 10.6 34.4 5.4s19-16.6 19-29.2V288h32c53 0 96 43 96 96c0 30.4-12.8 47.9-22.2 56.7c-5.5 5.1-9.8 12-9.8 19.5c0 10.9 8.8 19.7 19.7 19.7c2.8 0 5.6-.6 8.1-1.9C494.5 467.9 576 417.3 576 304c0-97.2-78.8-176-176-176H352V64z\"],suitcase:[512,512,[129523],\"f0f2\",\"M176 56V96H336V56c0-4.4-3.6-8-8-8H184c-4.4 0-8 3.6-8 8zM128 96V56c0-30.9 25.1-56 56-56H328c30.9 0 56 25.1 56 56V96v32V480H128V128 96zM64 96H96V480H64c-35.3 0-64-28.7-64-64V160c0-35.3 28.7-64 64-64zM448 480H416V96h32c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64z\"],\"person-skating\":[448,512,[\"skating\"],\"f7c5\",\"M352 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM128 128c0-17.7 14.3-32 32-32H319.4c43.6 0 64.6 53.4 32.8 83.1l-74.4 69.4 60.2 60.2c9 9 14.1 21.2 14.1 33.9V416c0 17.7-14.3 32-32 32s-32-14.3-32-32V349.3l-77.9-77.8c-26.6-26.6-24.6-70.3 4.3-94.4l20.4-17H160c-17.7 0-32-14.3-32-32zM81.4 353.4l86.9-86.9c4.6 10 11 19.3 19.3 27.5l21.8 21.8-82.7 82.7c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3zm322.5 95.1c8.6 2.1 13.8 10.8 11.6 19.4l-.4 1.7c-6.2 24.9-28.6 42.4-54.3 42.4H272c-8.8 0-16-7.2-16-16s7.2-16 16-16h88.8c11 0 20.6-7.5 23.3-18.2l.4-1.7c2.1-8.6 10.8-13.8 19.4-11.6zM135.2 478.3l-6.2 3.1c-21.6 10.8-47.6 6.6-64.6-10.5L4.7 411.3c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0l59.6 59.6c7.3 7.3 18.5 9.1 27.7 4.5l6.2-3.1c7.9-4 17.5-.7 21.5 7.2s.7 17.5-7.2 21.5z\"],\"filter-circle-dollar\":[576,512,[\"funnel-dollar\"],\"f662\",\"M3.9 22.9C10.5 8.9 24.5 0 40 0H472c15.5 0 29.5 8.9 36.1 22.9s4.6 30.5-5.2 42.5L396.4 195.6C316.2 212.1 256 283 256 368c0 27.4 6.3 53.4 17.5 76.5c-1.6-.8-3.2-1.8-4.7-2.9l-64-48c-8.1-6-12.8-15.5-12.8-25.6V288.9L9 65.3C-.7 53.4-2.8 36.8 3.9 22.9zM288 368a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm120.8-32.6c.6-.9 1.8-2.1 4.2-3.4c5.1-2.7 12.5-4.1 18.7-4c8.2 .1 17.1 1.8 26.4 4.1c8.6 2.1 17.3-3.1 19.4-11.7s-3.1-17.3-11.7-19.4c-5.6-1.4-11.6-2.7-17.9-3.7V288c0-8.8-7.2-16-16-16s-16 7.2-16 16v9.5c-6.1 1.2-12.3 3.2-18 6.3c-11.8 6.3-23 18.4-21.8 37.2c1 16 11.7 25.3 21.6 30.7c8.8 4.7 19.7 7.8 28.6 10.3l1.8 .5c10.3 2.9 17.9 5.2 23.2 8.3c4.5 2.7 4.7 4.2 4.7 5.6c.1 2.4-.5 3.7-1 4.5c-.6 1-1.8 2.2-4 3.3c-4.7 2.5-11.8 3.8-18.5 3.6c-9.5-.3-18.5-3.1-29.9-6.8c-1.9-.6-3.8-1.2-5.8-1.8c-8.4-2.6-17.4 2.1-20 10.5s2.1 17.4 10.5 20c1.6 .5 3.3 1 5 1.6l0 0 0 0c7 2.3 15.1 4.8 23.7 6.6v11.4c0 8.8 7.2 16 16 16s16-7.2 16-16V438.7c6.2-1.1 12.5-3.1 18.3-6.2c12.1-6.5 22.3-18.7 21.7-36.9c-.5-16.2-10.3-26.3-20.5-32.3c-9.4-5.6-21.2-8.9-30.5-11.5l-.2 0c-10.4-2.9-18.3-5.2-23.9-8.2c-4.8-2.6-4.8-4-4.8-4.5l0-.1c-.1-1.9 .3-2.9 .8-3.6z\"],\"camera-retro\":[512,512,[128247],\"f083\",\"M220.6 121.2L271.1 96 448 96v96H333.2c-21.9-15.1-48.5-24-77.2-24s-55.2 8.9-77.2 24H64V128H192c9.9 0 19.7-2.3 28.6-6.8zM0 128V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H271.1c-9.9 0-19.7 2.3-28.6 6.8L192 64H160V48c0-8.8-7.2-16-16-16H80c-8.8 0-16 7.2-16 16l0 16C28.7 64 0 92.7 0 128zM168 304a88 88 0 1 1 176 0 88 88 0 1 1 -176 0z\"],\"circle-arrow-down\":[512,512,[\"arrow-circle-down\"],\"f0ab\",\"M256 0a256 256 0 1 0 0 512A256 256 0 1 0 256 0zM127 281c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l71 71L232 136c0-13.3 10.7-24 24-24s24 10.7 24 24l0 182.1 71-71c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9L273 393c-9.4 9.4-24.6 9.4-33.9 0L127 281z\"],\"file-import\":[512,512,[\"arrow-right-to-file\"],\"f56f\",\"M128 64c0-35.3 28.7-64 64-64H352V128c0 17.7 14.3 32 32 32H512V448c0 35.3-28.7 64-64 64H192c-35.3 0-64-28.7-64-64V336H302.1l-39 39c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l80-80c9.4-9.4 9.4-24.6 0-33.9l-80-80c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l39 39H128V64zm0 224v48H24c-13.3 0-24-10.7-24-24s10.7-24 24-24H128zM512 128H384V0L512 128z\"],\"square-arrow-up-right\":[448,512,[\"external-link-square\"],\"f14c\",\"M384 32c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H384zM160 144c-13.3 0-24 10.7-24 24s10.7 24 24 24h94.1L119 327c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l135-135V328c0 13.3 10.7 24 24 24s24-10.7 24-24V168c0-13.3-10.7-24-24-24H160z\"],\"box-open\":[640,512,[],\"f49e\",\"M58.9 42.1c3-6.1 9.6-9.6 16.3-8.7L320 64 564.8 33.4c6.7-.8 13.3 2.7 16.3 8.7l41.7 83.4c9 17.9-.6 39.6-19.8 45.1L439.6 217.3c-13.9 4-28.8-1.9-36.2-14.3L320 64 236.6 203c-7.4 12.4-22.3 18.3-36.2 14.3L37.1 170.6c-19.3-5.5-28.8-27.2-19.8-45.1L58.9 42.1zM321.1 128l54.9 91.4c14.9 24.8 44.6 36.6 72.5 28.6L576 211.6v167c0 22-15 41.2-36.4 46.6l-204.1 51c-10.2 2.6-20.9 2.6-31 0l-204.1-51C79 419.7 64 400.5 64 378.5v-167L191.6 248c27.8 8 57.6-3.8 72.5-28.6L318.9 128h2.2z\"],scroll:[576,512,[128220],\"f70e\",\"M0 80v48c0 17.7 14.3 32 32 32H48 96V80c0-26.5-21.5-48-48-48S0 53.5 0 80zM112 32c10 13.4 16 30 16 48V384c0 35.3 28.7 64 64 64s64-28.7 64-64v-5.3c0-32.4 26.3-58.7 58.7-58.7H480V128c0-53-43-96-96-96H112zM464 480c61.9 0 112-50.1 112-112c0-8.8-7.2-16-16-16H314.7c-14.7 0-26.7 11.9-26.7 26.7V384c0 53-43 96-96 96H368h96z\"],spa:[576,512,[],\"f5bb\",\"M183.1 235.3c33.7 20.7 62.9 48.1 85.8 80.5c7 9.9 13.4 20.3 19.1 31c5.7-10.8 12.1-21.1 19.1-31c22.9-32.4 52.1-59.8 85.8-80.5C437.6 207.8 490.1 192 546 192h9.9c11.1 0 20.1 9 20.1 20.1C576 360.1 456.1 480 308.1 480H288 267.9C119.9 480 0 360.1 0 212.1C0 201 9 192 20.1 192H30c55.9 0 108.4 15.8 153.1 43.3zM301.5 37.6c15.7 16.9 61.1 71.8 84.4 164.6c-38 21.6-71.4 50.8-97.9 85.6c-26.5-34.8-59.9-63.9-97.9-85.6c23.2-92.8 68.6-147.7 84.4-164.6C278 33.9 282.9 32 288 32s10 1.9 13.5 5.6z\"],\"location-pin-lock\":[512,512,[],\"e51f\",\"M215.7 499.2c11-13.8 25.1-31.7 40.3-52.3V352c0-23.7 12.9-44.4 32-55.4V272c0-55.6 40.5-101.7 93.6-110.5C367 70 287.7 0 192 0C86 0 0 86 0 192c0 87.4 117 243 168.3 307.2c12.3 15.3 35.1 15.3 47.4 0zM192 128a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM400 240c17.7 0 32 14.3 32 32v48H368V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H480c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z\"],pause:[320,512,[9208],\"f04c\",\"M48 64C21.5 64 0 85.5 0 112V400c0 26.5 21.5 48 48 48H80c26.5 0 48-21.5 48-48V112c0-26.5-21.5-48-48-48H48zm192 0c-26.5 0-48 21.5-48 48V400c0 26.5 21.5 48 48 48h32c26.5 0 48-21.5 48-48V112c0-26.5-21.5-48-48-48H240z\"],\"hill-avalanche\":[576,512,[],\"e507\",\"M439.7 401.9c34.2 23.1 81.1 19.5 111.4-10.8c34.4-34.4 34.4-90.1 0-124.4c-27.8-27.8-69.5-33.1-102.6-16c-11.8 6.1-16.4 20.6-10.3 32.3s20.6 16.4 32.3 10.3c15.1-7.8 34-5.3 46.6 7.3c15.6 15.6 15.6 40.9 0 56.6s-40.9 15.6-56.6 0l-81.7-81.7C401.2 261.3 416 236.4 416 208c0-33.9-21.1-62.9-50.9-74.5c1.9-6.8 2.9-14 2.9-21.5c0-44.2-35.8-80-80-80c-27.3 0-51.5 13.7-65.9 34.6C216.3 46.6 197.9 32 176 32c-26.5 0-48 21.5-48 48c0 4 .5 7.9 1.4 11.6L439.7 401.9zM480 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm0 128a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM68.3 87C43.1 61.8 0 79.7 0 115.3V432c0 44.2 35.8 80 80 80H396.7c35.6 0 53.5-43.1 28.3-68.3L68.3 87z\"],\"temperature-empty\":[320,512,[\"temperature-0\",\"thermometer-0\",\"thermometer-empty\"],\"f2cb\",\"M112 112c0-26.5 21.5-48 48-48s48 21.5 48 48V276.5c0 17.3 7.1 31.9 15.3 42.5C233.8 332.6 240 349.5 240 368c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-18.5 6.2-35.4 16.7-48.9c8.2-10.6 15.3-25.2 15.3-42.5V112zM160 0C98.1 0 48 50.2 48 112V276.5c0 .1-.1 .3-.2 .6c-.2 .6-.8 1.6-1.7 2.8C27.2 304.2 16 334.8 16 368c0 79.5 64.5 144 144 144s144-64.5 144-144c0-33.2-11.2-63.8-30.1-88.1c-.9-1.2-1.5-2.2-1.7-2.8c-.1-.3-.2-.5-.2-.6V112C272 50.2 221.9 0 160 0zm0 416a48 48 0 1 0 0-96 48 48 0 1 0 0 96z\"],bomb:[512,512,[128163],\"f1e2\",\"M459.1 52.4L442.6 6.5C440.7 2.6 436.5 0 432.1 0s-8.5 2.6-10.4 6.5L405.2 52.4l-46 16.8c-4.3 1.6-7.3 5.9-7.2 10.4c0 4.5 3 8.7 7.2 10.2l45.7 16.8 16.8 45.8c1.5 4.4 5.8 7.5 10.4 7.5s8.9-3.1 10.4-7.5l16.5-45.8 45.7-16.8c4.2-1.5 7.2-5.7 7.2-10.2c0-4.6-3-8.9-7.2-10.4L459.1 52.4zm-132.4 53c-12.5-12.5-32.8-12.5-45.3 0l-2.9 2.9C256.5 100.3 232.7 96 208 96C93.1 96 0 189.1 0 304S93.1 512 208 512s208-93.1 208-208c0-24.7-4.3-48.5-12.2-70.5l2.9-2.9c12.5-12.5 12.5-32.8 0-45.3l-80-80zM200 192c-57.4 0-104 46.6-104 104v8c0 8.8-7.2 16-16 16s-16-7.2-16-16v-8c0-75.1 60.9-136 136-136h8c8.8 0 16 7.2 16 16s-7.2 16-16 16h-8z\"],registered:[512,512,[174],\"f25d\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM160 152c0-13.3 10.7-24 24-24h88c44.2 0 80 35.8 80 80c0 28-14.4 52.7-36.3 67l34.1 75.1c5.5 12.1 .1 26.3-11.9 31.8s-26.3 .1-31.8-11.9L268.9 288H208v72c0 13.3-10.7 24-24 24s-24-10.7-24-24V264 152zm48 88h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H208v64z\"],\"address-card\":[576,512,[62140,\"contact-card\",\"vcard\"],\"f2bb\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm80 256h64c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80zm-32-96a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zm256-32H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H368c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H368c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H368c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],\"scale-unbalanced-flip\":[640,512,[\"balance-scale-right\"],\"f516\",\"M117.9 62.4c-16.8-5.6-25.8-23.7-20.2-40.5s23.7-25.8 40.5-20.2l113 37.7C265 15.8 290.7 0 320 0c44.2 0 80 35.8 80 80c0 3-.2 5.9-.5 8.8l122.6 40.9c16.8 5.6 25.8 23.7 20.2 40.5s-23.7 25.8-40.5 20.2L366.4 145.2c-4.5 3.2-9.3 5.9-14.4 8.2V480c0 17.7-14.3 32-32 32H128c-17.7 0-32-14.3-32-32s14.3-32 32-32H288V153.3c-21-9.2-37.2-27-44.2-49l-125.9-42zM200.4 288L128 163.8 55.6 288H200.4zM128 384C65.1 384 12.8 350 2 305.1c-2.6-11 1-22.3 6.7-32.1l95.2-163.2c5-8.6 14.2-13.8 24.1-13.8s19.1 5.3 24.1 13.8l95.2 163.2c5.7 9.8 9.3 21.1 6.7 32.1C243.2 350 190.9 384 128 384zm382.8-92.2L438.4 416H583.3L510.8 291.8zm126 141.3C626 478 573.7 512 510.8 512s-115.2-34-126-78.9c-2.6-11 1-22.3 6.7-32.1l95.2-163.2c5-8.6 14.2-13.8 24.1-13.8s19.1 5.3 24.1 13.8l95.2 163.2c5.7 9.8 9.3 21.1 6.7 32.1z\"],subscript:[512,512,[],\"f12c\",\"M32 64C14.3 64 0 78.3 0 96s14.3 32 32 32H47.3l89.6 128L47.3 384H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64c10.4 0 20.2-5.1 26.2-13.6L176 311.8l85.8 122.6c6 8.6 15.8 13.6 26.2 13.6h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H304.7L215.1 256l89.6-128H320c17.7 0 32-14.3 32-32s-14.3-32-32-32H288c-10.4 0-20.2 5.1-26.2 13.6L176 200.2 90.2 77.6C84.2 69.1 74.4 64 64 64H32zM480 320c0-11.1-5.7-21.4-15.2-27.2s-21.2-6.4-31.1-1.4l-32 16c-15.8 7.9-22.2 27.1-14.3 42.9C393 361.5 404.3 368 416 368v80c-17.7 0-32 14.3-32 32s14.3 32 32 32h32 32c17.7 0 32-14.3 32-32s-14.3-32-32-32V320z\"],\"diamond-turn-right\":[512,512,[\"directions\"],\"f5eb\",\"M227.7 11.7c15.6-15.6 40.9-15.6 56.6 0l216 216c15.6 15.6 15.6 40.9 0 56.6l-216 216c-15.6 15.6-40.9 15.6-56.6 0l-216-216c-15.6-15.6-15.6-40.9 0-56.6l216-216zm87.6 137c-4.6-4.6-11.5-5.9-17.4-3.5s-9.9 8.3-9.9 14.8v56H224c-35.3 0-64 28.7-64 64v48c0 13.3 10.7 24 24 24s24-10.7 24-24V280c0-8.8 7.2-16 16-16h64v56c0 6.5 3.9 12.3 9.9 14.8s12.9 1.1 17.4-3.5l80-80c6.2-6.2 6.2-16.4 0-22.6l-80-80z\"],burst:[512,512,[],\"e4dc\",\"M37.6 4.2C28-2.3 15.2-1.1 7 7s-9.4 21-2.8 30.5l112 163.3L16.6 233.2C6.7 236.4 0 245.6 0 256s6.7 19.6 16.6 22.8l103.1 33.4L66.8 412.8c-4.9 9.3-3.2 20.7 4.3 28.1s18.8 9.2 28.1 4.3l100.6-52.9 33.4 103.1c3.2 9.9 12.4 16.6 22.8 16.6s19.6-6.7 22.8-16.6l33.4-103.1 100.6 52.9c9.3 4.9 20.7 3.2 28.1-4.3s9.2-18.8 4.3-28.1L392.3 312.2l103.1-33.4c9.9-3.2 16.6-12.4 16.6-22.8s-6.7-19.6-16.6-22.8L388.9 198.7l25.7-70.4c3.2-8.8 1-18.6-5.6-25.2s-16.4-8.8-25.2-5.6l-70.4 25.7L278.8 16.6C275.6 6.7 266.4 0 256 0s-19.6 6.7-22.8 16.6l-32.3 99.6L37.6 4.2z\"],\"house-laptop\":[640,512,[\"laptop-house\"],\"e066\",\"M218.3 8.5c12.3-11.3 31.2-11.3 43.4 0l208 192c6.7 6.2 10.3 14.8 10.3 23.5H336c-19.1 0-36.3 8.4-48 21.7V208c0-8.8-7.2-16-16-16H208c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h64V416H112c-26.5 0-48-21.5-48-48V256H32c-13.2 0-25-8.1-29.8-20.3s-1.6-26.2 8.1-35.2l208-192zM352 304V448H544V304H352zm-48-16c0-17.7 14.3-32 32-32H560c17.7 0 32 14.3 32 32V448h32c8.8 0 16 7.2 16 16c0 26.5-21.5 48-48 48H544 352 304c-26.5 0-48-21.5-48-48c0-8.8 7.2-16 16-16h32V288z\"],\"face-tired\":[512,512,[128555,\"tired\"],\"f5c8\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM164.7 328.7c22-22 53.9-40.7 91.3-40.7s69.3 18.7 91.3 40.7c11.1 11.1 20.1 23.4 26.4 35.4c6.2 11.7 10.3 24.4 10.3 35.9c0 5.2-2.6 10.2-6.9 13.2s-9.8 3.7-14.7 1.8l-20.5-7.7c-26.9-10.1-55.5-15.3-84.3-15.3h-3.2c-28.8 0-57.3 5.2-84.3 15.3L149.6 415c-4.9 1.8-10.4 1.2-14.7-1.8s-6.9-7.9-6.9-13.2c0-11.6 4.2-24.2 10.3-35.9c6.3-12 15.3-24.3 26.4-35.4zm-31.2-182l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6c0-9 9.6-14.7 17.5-10.5zM396 157.1c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9c7.9-4.2 17.5 1.5 17.5 10.5z\"],\"money-bills\":[640,512,[],\"e1f3\",\"M96 96V320c0 35.3 28.7 64 64 64H576c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H160c-35.3 0-64 28.7-64 64zm64 160c35.3 0 64 28.7 64 64H160V256zM224 96c0 35.3-28.7 64-64 64V96h64zM576 256v64H512c0-35.3 28.7-64 64-64zM512 96h64v64c-35.3 0-64-28.7-64-64zM288 208a80 80 0 1 1 160 0 80 80 0 1 1 -160 0zM48 120c0-13.3-10.7-24-24-24S0 106.7 0 120V360c0 66.3 53.7 120 120 120H520c13.3 0 24-10.7 24-24s-10.7-24-24-24H120c-39.8 0-72-32.2-72-72V120z\"],smog:[640,512,[],\"f75f\",\"M32 144c0 79.5 64.5 144 144 144H299.3c22.6 19.9 52.2 32 84.7 32s62.1-12.1 84.7-32H496c61.9 0 112-50.1 112-112s-50.1-112-112-112c-10.7 0-21 1.5-30.8 4.3C443.8 27.7 401.1 0 352 0c-32.6 0-62.4 12.2-85.1 32.3C242.1 12.1 210.5 0 176 0C96.5 0 32 64.5 32 144zM616 368H280c-13.3 0-24 10.7-24 24s10.7 24 24 24H616c13.3 0 24-10.7 24-24s-10.7-24-24-24zm-64 96H440c-13.3 0-24 10.7-24 24s10.7 24 24 24H552c13.3 0 24-10.7 24-24s-10.7-24-24-24zm-192 0H24c-13.3 0-24 10.7-24 24s10.7 24 24 24H360c13.3 0 24-10.7 24-24s-10.7-24-24-24zM224 392c0-13.3-10.7-24-24-24H96c-13.3 0-24 10.7-24 24s10.7 24 24 24H200c13.3 0 24-10.7 24-24z\"],crutch:[512,512,[],\"f7f7\",\"M297.4 9.4c-12.5 12.5-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-160-160c-12.5-12.5-32.8-12.5-45.3 0zm-96 144l-34.8 34.8c-12.9 12.9-21.9 29.2-25.8 47.1L116.8 342.9c-1.3 5.9-4.3 11.4-8.6 15.7L9.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l98.8-98.8c4.3-4.3 9.7-7.3 15.7-8.6l107.6-23.9c17.8-4 34.1-12.9 47.1-25.8l34.7-34.7c0 0 .1-.1 .1-.1s.1-.1 .1-.1l74.6-74.6-45.3-45.3L336 242.7 269.3 176l52.1-52.1L276.1 78.6l-74.7 74.7zM224 221.3L290.7 288l-12.2 12.2c-4.3 4.3-9.7 7.3-15.7 8.6l-76.7 17 17-76.7c1.3-5.9 4.3-11.4 8.6-15.7L224 221.3z\"],\"font-awesome\":[512,512,[62501,62694,\"font-awesome-flag\",\"font-awesome-logo-full\"],\"f2b4\",\"M91.7 96C106.3 86.8 116 70.5 116 52C116 23.3 92.7 0 64 0S12 23.3 12 52c0 16.7 7.8 31.5 20 41l0 3 0 352 0 64 64 0 0-64 373.6 0c14.6 0 26.4-11.8 26.4-26.4c0-3.7-.8-7.3-2.3-10.7L432 272l61.7-138.9c1.5-3.4 2.3-7 2.3-10.7c0-14.6-11.8-26.4-26.4-26.4L91.7 96z\"],\"cloud-arrow-up\":[640,512,[62338,\"cloud-upload\",\"cloud-upload-alt\"],\"f0ee\",\"M144 480C64.5 480 0 415.5 0 336c0-62.8 40.2-116.2 96.2-135.9c-.1-2.7-.2-5.4-.2-8.1c0-88.4 71.6-160 160-160c59.3 0 111 32.2 138.7 80.2C409.9 102 428.3 96 448 96c53 0 96 43 96 96c0 12.2-2.3 23.8-6.4 34.6C596 238.4 640 290.1 640 352c0 70.7-57.3 128-128 128H144zm79-217c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l39-39V392c0 13.3 10.7 24 24 24s24-10.7 24-24V257.9l39 39c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-80-80c-9.4-9.4-24.6-9.4-33.9 0l-80 80z\"],palette:[512,512,[127912],\"f53f\",\"M512 256c0 .9 0 1.8 0 2.7c-.4 36.5-33.6 61.3-70.1 61.3H344c-26.5 0-48 21.5-48 48c0 3.4 .4 6.7 1 9.9c2.1 10.2 6.5 20 10.8 29.9c6.1 13.8 12.1 27.5 12.1 42c0 31.8-21.6 60.7-53.4 62c-3.5 .1-7 .2-10.6 .2C114.6 512 0 397.4 0 256S114.6 0 256 0S512 114.6 512 256zM128 288a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm0-96a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM288 96a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm96 96a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"arrows-turn-right\":[448,512,[],\"e4c0\",\"M297.4 9.4c12.5-12.5 32.8-12.5 45.3 0l96 96c12.5 12.5 12.5 32.8 0 45.3l-96 96c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L338.7 160H128c-35.3 0-64 28.7-64 64v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V224C0 153.3 57.3 96 128 96H338.7L297.4 54.6c-12.5-12.5-12.5-32.8 0-45.3zm-96 256c12.5-12.5 32.8-12.5 45.3 0l96 96c12.5 12.5 12.5 32.8 0 45.3l-96 96c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L242.7 416H96c-17.7 0-32 14.3-32 32v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448c0-53 43-96 96-96H242.7l-41.4-41.4c-12.5-12.5-12.5-32.8 0-45.3z\"],vest:[448,512,[],\"e085\",\"M207.1 237.4L151.2 69.7C168.6 79.7 192.6 88 224 88s55.4-8.3 72.8-18.3L226.5 280.6c-1.6 4.9-2.5 10-2.5 15.2V464c0 26.5 21.5 48 48 48H400c26.5 0 48-21.5 48-48V270.5c0-9.5-2.8-18.7-8.1-26.6l-47.9-71.8c-5.3-7.9-8.1-17.1-8.1-26.6V128 54.3 48c0-26.5-21.5-48-48-48h-4.5c-.2 0-.4 0-.6 0c-.4 0-.8 0-1.2 0C311 0 295.7 9.7 285.7 18.8C276.4 27.2 257.2 40 224 40s-52.4-12.8-61.7-21.2C152.3 9.7 137 0 118.3 0c-.4 0-.8 0-1.2 0c-.2 0-.4 0-.6 0H112C85.5 0 64 21.5 64 48v6.3V128v17.5c0 9.5-2.8 18.7-8.1 26.6L8.1 243.9C2.8 251.8 0 261.1 0 270.5V464c0 26.5 21.5 48 48 48H176c9.9 0 19-3 26.7-8.1C195.9 492.2 192 478.5 192 464V295.8c0-8.6 1.4-17.1 4.1-25.3l11-33.1zM347.3 356.7l48 48c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0l-48-48c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0zm-294.6 48l48-48c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6l-48 48c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6z\"],ferry:[576,512,[],\"e4ea\",\"M224 0H352c17.7 0 32 14.3 32 32h75.1c20.6 0 31.6 24.3 18.1 39.8L456 96H120L98.8 71.8C85.3 56.3 96.3 32 116.9 32H192c0-17.7 14.3-32 32-32zM96 128H480c17.7 0 32 14.3 32 32V283.5c0 13.3-4.2 26.3-11.9 37.2l-51.4 71.9c-1.9 1.1-3.7 2.2-5.5 3.5c-15.5 10.7-34 18-51 19.9H375.6c-17.1-1.8-35-9-50.8-19.9c-22.1-15.5-51.6-15.5-73.7 0c-14.8 10.2-32.5 18-50.6 19.9H183.9c-17-1.8-35.6-9.2-51-19.9c-1.8-1.3-3.7-2.4-5.6-3.5L75.9 320.7C68.2 309.8 64 296.8 64 283.5V160c0-17.7 14.3-32 32-32zm32 64v96H448V192H128zM306.5 421.9C329 437.4 356.5 448 384 448c26.9 0 55.3-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 501.7 417 512 384 512c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 437.2 165.1 448 192 448c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z\"],\"arrows-down-to-people\":[640,512,[],\"e4b9\",\"M144 0c-13.3 0-24 10.7-24 24V142.1L97 119c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l64 64c9.4 9.4 24.6 9.4 33.9 0l64-64c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-23 23V24c0-13.3-10.7-24-24-24zM360 200a40 40 0 1 0 -80 0 40 40 0 1 0 80 0zM184 296a40 40 0 1 0 -80 0 40 40 0 1 0 80 0zm312 40a40 40 0 1 0 0-80 40 40 0 1 0 0 80zM200 441.5l26.9 49.9c6.3 11.7 20.8 16 32.5 9.8s16-20.8 9.8-32.5l-36.3-67.5c1.7-1.7 3.2-3.6 4.3-5.8L264 345.5V400c0 17.7 14.3 32 32 32h48c17.7 0 32-14.3 32-32V345.5l26.9 49.9c1.2 2.2 2.6 4.1 4.3 5.8l-36.3 67.5c-6.3 11.7-1.9 26.2 9.8 32.5s26.2 1.9 32.5-9.8L440 441.5V480c0 17.7 14.3 32 32 32h48c17.7 0 32-14.3 32-32V441.5l26.9 49.9c6.3 11.7 20.8 16 32.5 9.8s16-20.8 9.8-32.5l-37.9-70.3c-15.3-28.5-45.1-46.3-77.5-46.3H486.2c-16.3 0-31.9 4.5-45.4 12.6l-33.6-62.3c-15.3-28.5-45.1-46.3-77.5-46.3H310.2c-32.4 0-62.1 17.8-77.5 46.3l-33.6 62.3c-13.5-8.1-29.1-12.6-45.4-12.6H134.2c-32.4 0-62.1 17.8-77.5 46.3L18.9 468.6c-6.3 11.7-1.9 26.2 9.8 32.5s26.2 1.9 32.5-9.8L88 441.5V480c0 17.7 14.3 32 32 32h48c17.7 0 32-14.3 32-32V441.5zM415 153l64 64c9.4 9.4 24.6 9.4 33.9 0l64-64c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-23 23V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V142.1l-23-23c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9z\"],seedling:[512,512,[127793,\"sprout\"],\"f4d8\",\"M512 32c0 113.6-84.6 207.5-194.2 222c-7.1-53.4-30.6-101.6-65.3-139.3C290.8 46.3 364 0 448 0h32c17.7 0 32 14.3 32 32zM0 96C0 78.3 14.3 64 32 64H64c123.7 0 224 100.3 224 224v32V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V320C100.3 320 0 219.7 0 96z\"],\"left-right\":[512,512,[8596,\"arrows-alt-h\"],\"f337\",\"M504.3 273.6c4.9-4.5 7.7-10.9 7.7-17.6s-2.8-13-7.7-17.6l-112-104c-7-6.5-17.2-8.2-25.9-4.4s-14.4 12.5-14.4 22l0 56-192 0 0-56c0-9.5-5.7-18.2-14.4-22s-18.9-2.1-25.9 4.4l-112 104C2.8 243 0 249.3 0 256s2.8 13 7.7 17.6l112 104c7 6.5 17.2 8.2 25.9 4.4s14.4-12.5 14.4-22l0-56 192 0 0 56c0 9.5 5.7 18.2 14.4 22s18.9 2.1 25.9-4.4l112-104z\"],\"boxes-packing\":[640,512,[],\"e4c7\",\"M256 48c0-26.5 21.5-48 48-48H592c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H381.3c1.8-5 2.7-10.4 2.7-16V253.3c18.6-6.6 32-24.4 32-45.3V176c0-26.5-21.5-48-48-48H256V48zM571.3 347.3c6.2-6.2 6.2-16.4 0-22.6l-64-64c-6.2-6.2-16.4-6.2-22.6 0l-64 64c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L480 310.6V432c0 8.8 7.2 16 16 16s16-7.2 16-16V310.6l36.7 36.7c6.2 6.2 16.4 6.2 22.6 0zM0 176c0-8.8 7.2-16 16-16H368c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H16c-8.8 0-16-7.2-16-16V176zm352 80V480c0 17.7-14.3 32-32 32H64c-17.7 0-32-14.3-32-32V256H352zM144 320c-8.8 0-16 7.2-16 16s7.2 16 16 16h96c8.8 0 16-7.2 16-16s-7.2-16-16-16H144z\"],\"circle-arrow-left\":[512,512,[\"arrow-circle-left\"],\"f0a8\",\"M512 256A256 256 0 1 0 0 256a256 256 0 1 0 512 0zM231 127c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-71 71L376 232c13.3 0 24 10.7 24 24s-10.7 24-24 24l-182.1 0 71 71c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L119 273c-9.4-9.4-9.4-24.6 0-33.9L231 127z\"],\"group-arrows-rotate\":[512,512,[],\"e4f6\",\"M201.1 71.9c16.9-5 26.6-22.9 21.5-39.8s-22.9-26.6-39.8-21.5c-21.5 6.4-41.8 15.5-60.6 27C114.3 34 105.4 32 96 32C60.7 32 32 60.7 32 96c0 9.4 2 18.3 5.6 26.3c-11.5 18.7-20.6 39-27 60.6c-5 16.9 4.6 34.8 21.5 39.8s34.8-4.6 39.8-21.5c4.3-14.6 10.4-28.5 17.9-41.4c2 .2 4.1 .3 6.1 .3c35.3 0 64-28.7 64-64c0-2.1-.1-4.1-.3-6.1c12.9-7.5 26.8-13.6 41.4-17.9zm128-61.3c-16.9-5-34.8 4.6-39.8 21.5s4.6 34.8 21.5 39.8c14.6 4.3 28.5 10.4 41.4 17.9c-.2 2-.3 4.1-.3 6.1c0 35.3 28.7 64 64 64c2.1 0 4.1-.1 6.2-.3c7.5 12.9 13.6 26.8 17.9 41.4c5 16.9 22.9 26.6 39.8 21.5s26.6-22.9 21.5-39.8c-6.4-21.5-15.5-41.8-27-60.6c3.6-8 5.6-16.9 5.6-26.3c0-35.3-28.7-64-64-64c-9.4 0-18.3 2-26.3 5.6c-18.7-11.5-39-20.6-60.6-27zM71.9 310.9c-5-16.9-22.9-26.6-39.8-21.5s-26.6 22.9-21.5 39.8c6.4 21.5 15.5 41.8 27 60.6C34 397.7 32 406.6 32 416c0 35.3 28.7 64 64 64c9.4 0 18.3-2 26.3-5.6c18.7 11.5 39 20.6 60.6 27c16.9 5 34.8-4.6 39.8-21.5s-4.6-34.8-21.5-39.8c-14.6-4.3-28.5-10.4-41.4-17.9c.2-2 .3-4.1 .3-6.2c0-35.3-28.7-64-64-64c-2.1 0-4.1 .1-6.2 .3c-7.5-12.9-13.6-26.8-17.9-41.4zm429.4 18.3c5-16.9-4.6-34.8-21.5-39.8s-34.8 4.6-39.8 21.5c-4.3 14.6-10.4 28.5-17.9 41.4c-2-.2-4.1-.3-6.2-.3c-35.3 0-64 28.7-64 64c0 2.1 .1 4.1 .3 6.2c-12.9 7.5-26.8 13.6-41.4 17.9c-16.9 5-26.6 22.9-21.5 39.8s22.9 26.6 39.8 21.5c21.5-6.4 41.8-15.5 60.6-27c8 3.6 16.9 5.6 26.3 5.6c35.3 0 64-28.7 64-64c0-9.4-2-18.3-5.6-26.3c11.5-18.7 20.6-39 27-60.6zM192.8 256.8c0-15.6 5.6-29.9 14.9-41.1L223 231c6.6 6.6 17.8 1.9 17.8-7.4V163.2c0-5.7-4.7-10.4-10.4-10.4H169.9c-9.3 0-13.9 11.2-7.4 17.8l11.2 11.2c-17.9 19.8-28.9 46.2-28.9 75.1c0 43.6 24.9 81.3 61.1 99.8c11.8 6 26.3 1.4 32.3-10.4s1.4-26.3-10.4-32.3c-20.8-10.6-34.9-32.2-34.9-57zm93.1-58.6c20.8 10.6 34.9 32.2 34.9 57c0 15.6-5.6 29.9-14.9 41.1L290.6 281c-6.6-6.6-17.8-1.9-17.8 7.4v60.5c0 5.7 4.7 10.4 10.4 10.4h60.5c9.3 0 13.9-11.2 7.4-17.8l-11.2-11.2c17.9-19.8 28.9-46.2 28.9-75.1c0-43.6-24.9-81.3-61.1-99.8c-11.8-6-26.3-1.4-32.3 10.4s-1.4 26.3 10.4 32.3z\"],\"bowl-food\":[512,512,[],\"e4c6\",\"M0 192c0-35.3 28.7-64 64-64c.5 0 1.1 0 1.6 0C73 91.5 105.3 64 144 64c15 0 29 4.1 40.9 11.2C198.2 49.6 225.1 32 256 32s57.8 17.6 71.1 43.2C339 68.1 353 64 368 64c38.7 0 71 27.5 78.4 64c.5 0 1.1 0 1.6 0c35.3 0 64 28.7 64 64c0 11.7-3.1 22.6-8.6 32H8.6C3.1 214.6 0 203.7 0 192zm0 91.4C0 268.3 12.3 256 27.4 256H484.6c15.1 0 27.4 12.3 27.4 27.4c0 70.5-44.4 130.7-106.7 154.1L403.5 452c-2 16-15.6 28-31.8 28H140.2c-16.1 0-29.8-12-31.8-28l-1.8-14.4C44.4 414.1 0 353.9 0 283.4z\"],\"candy-cane\":[512,512,[],\"f786\",\"M348.8 131.5c3.7-2.3 7.9-3.5 12.2-3.5c12.7 0 23 10.3 23 23v5.6c0 9.9-5.1 19.1-13.5 24.3L30.1 393.7C.1 412.5-9 451.9 9.7 481.9s58.2 39.1 88.2 20.4L438.4 289.5c45.8-28.6 73.6-78.8 73.6-132.8V151C512 67.6 444.4 0 361 0c-28.3 0-56 8-80.1 23L254.1 39.7c-30 18.7-39.1 58.2-20.4 88.2s58.2 39.1 88.2 20.4l26.8-16.8zM298.4 49.8c9.2-5.7 19.1-10.1 29.4-13.1L348 97.5c-5.7 1.4-11.2 3.7-16.3 6.8l-12.6 7.9L298.4 49.8zm88.5 52.7l46.2-46.2c8.5 6.5 16.1 14.1 22.6 22.6l-46.2 46.2c-5.1-9.6-13-17.5-22.6-22.6zm28.9 59.3l61.6 20.5c-2.2 10.5-5.8 20.7-10.5 30.2l-62-20.7c6.2-8.8 10.1-19.1 11-30.1zm-86.1 82.5l60.4 37.7-30.2 18.9-60.4-37.7 30.2-18.9zm-107.2 67l60.4 37.7-30.2 18.9-60.4-37.7 30.2-18.9zM119.3 375.7l60.4 37.7-30.2 18.9L89.1 394.6l30.2-18.9z\"],\"arrow-down-wide-short\":[576,512,[\"sort-amount-asc\",\"sort-amount-down\"],\"f160\",\"M151.6 469.6C145.5 476.2 137 480 128 480s-17.5-3.8-23.6-10.4l-88-96c-11.9-13-11.1-33.3 2-45.2s33.3-11.1 45.2 2L96 365.7V64c0-17.7 14.3-32 32-32s32 14.3 32 32V365.7l32.4-35.4c11.9-13 32.2-13.9 45.2-2s13.9 32.2 2 45.2l-88 96zM320 480c-17.7 0-32-14.3-32-32s14.3-32 32-32h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H320zm0-128c-17.7 0-32-14.3-32-32s14.3-32 32-32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H320zm0-128c-17.7 0-32-14.3-32-32s14.3-32 32-32H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H320zm0-128c-17.7 0-32-14.3-32-32s14.3-32 32-32H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H320z\"],\"cloud-bolt\":[512,512,[127785,\"thunderstorm\"],\"f76c\",\"M0 224c0 53 43 96 96 96h47.2L290 202.5c17.6-14.1 42.6-14 60.2 .2s22.8 38.6 12.8 58.8L333.7 320H352h64c53 0 96-43 96-96s-43-96-96-96c-.5 0-1.1 0-1.6 0c1.1-5.2 1.6-10.5 1.6-16c0-44.2-35.8-80-80-80c-24.3 0-46.1 10.9-60.8 28C256.5 24.3 219.1 0 176 0C114.1 0 64 50.1 64 112c0 7.1 .7 14.1 1.9 20.8C27.6 145.4 0 181.5 0 224zm330.1 3.6c-5.8-4.7-14.2-4.7-20.1-.1l-160 128c-5.3 4.2-7.4 11.4-5.1 17.8s8.3 10.7 15.1 10.7h70.1L177.7 488.8c-3.4 6.7-1.6 14.9 4.3 19.6s14.2 4.7 20.1 .1l160-128c5.3-4.2 7.4-11.4 5.1-17.8s-8.3-10.7-15.1-10.7H281.9l52.4-104.8c3.4-6.7 1.6-14.9-4.2-19.6z\"],\"text-slash\":[640,512,[\"remove-format\"],\"f87d\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L355.7 253.5 400.2 96H503L497 120.2c-4.3 17.1 6.1 34.5 23.3 38.8s34.5-6.1 38.8-23.3l11-44.1C577.6 61.3 554.7 32 523.5 32H376.1h-.3H204.5c-22 0-41.2 15-46.6 36.4l-6.3 25.2L38.8 5.1zm168 131.7c.1-.3 .2-.7 .3-1L217 96H333.7L301.3 210.8l-94.5-74.1zM327.3 353.9L272.9 311 243.3 416H192c-17.7 0-32 14.3-32 32s14.3 32 32 32H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H309.8l17.6-62.1z\"],\"face-smile-wink\":[512,512,[128521,\"smile-wink\"],\"f4da\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM164.1 325.5C182 346.2 212.6 368 256 368s74-21.8 91.9-42.5c5.8-6.7 15.9-7.4 22.6-1.6s7.4 15.9 1.6 22.6C349.8 372.1 311.1 400 256 400s-93.8-27.9-116.1-53.5c-5.8-6.7-5.1-16.8 1.6-22.6s16.8-5.1 22.6 1.6zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm156.4 25.6c-5.3 7.1-15.3 8.5-22.4 3.2s-8.5-15.3-3.2-22.4c30.4-40.5 91.2-40.5 121.6 0c5.3 7.1 3.9 17.1-3.2 22.4s-17.1 3.9-22.4-3.2c-17.6-23.5-52.8-23.5-70.4 0z\"],\"file-word\":[384,512,[],\"f1c2\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM111 257.1l26.8 89.2 31.6-90.3c3.4-9.6 12.5-16.1 22.7-16.1s19.3 6.4 22.7 16.1l31.6 90.3L273 257.1c3.8-12.7 17.2-19.9 29.9-16.1s19.9 17.2 16.1 29.9l-48 160c-3 10-12 16.9-22.4 17.1s-19.8-6.2-23.2-16.1L192 336.6l-33.3 95.3c-3.4 9.8-12.8 16.3-23.2 16.1s-19.5-7.1-22.4-17.1l-48-160c-3.8-12.7 3.4-26.1 16.1-29.9s26.1 3.4 29.9 16.1z\"],\"file-powerpoint\":[384,512,[],\"f1c4\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM136 240h68c42 0 76 34 76 76s-34 76-76 76H160v32c0 13.3-10.7 24-24 24s-24-10.7-24-24V368 264c0-13.3 10.7-24 24-24zm68 104c15.5 0 28-12.5 28-28s-12.5-28-28-28H160v56h44z\"],\"arrows-left-right\":[512,512,[\"arrows-h\"],\"f07e\",\"M406.6 374.6l96-96c12.5-12.5 12.5-32.8 0-45.3l-96-96c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L402.7 224l-293.5 0 41.4-41.4c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-96 96c-12.5 12.5-12.5 32.8 0 45.3l96 96c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L109.3 288l293.5 0-41.4 41.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0z\"],\"house-lock\":[640,512,[],\"e510\",\"M384 480c0 11.7 3.1 22.6 8.6 32H392c-22.1 0-40-17.9-40-40V448 384c0-17.7-14.3-32-32-32H256c-17.7 0-32 14.3-32 32v64 24c0 22.1-17.9 40-40 40H160 128.1c-1.5 0-3-.1-4.5-.2c-1.2 .1-2.4 .2-3.6 .2H104c-22.1 0-40-17.9-40-40V360c0-.9 0-1.9 .1-2.8V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L490.7 166.3C447.2 181.7 416 223.2 416 272v24.6c-19.1 11.1-32 31.7-32 55.4V480zM528 240c-17.7 0-32 14.3-32 32v48h64V272c0-17.7-14.3-32-32-32zm-80 32c0-44.2 35.8-80 80-80s80 35.8 80 80v48c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H448c-17.7 0-32-14.3-32-32V352c0-17.7 14.3-32 32-32V272z\"],\"cloud-arrow-down\":[640,512,[62337,\"cloud-download\",\"cloud-download-alt\"],\"f0ed\",\"M144 480C64.5 480 0 415.5 0 336c0-62.8 40.2-116.2 96.2-135.9c-.1-2.7-.2-5.4-.2-8.1c0-88.4 71.6-160 160-160c59.3 0 111 32.2 138.7 80.2C409.9 102 428.3 96 448 96c53 0 96 43 96 96c0 12.2-2.3 23.8-6.4 34.6C596 238.4 640 290.1 640 352c0 70.7-57.3 128-128 128H144zm79-167l80 80c9.4 9.4 24.6 9.4 33.9 0l80-80c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-39 39V184c0-13.3-10.7-24-24-24s-24 10.7-24 24V318.1l-39-39c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9z\"],children:[640,512,[],\"e4e1\",\"M160 0a64 64 0 1 1 0 128A64 64 0 1 1 160 0zM88 480V400H70.2c-10.9 0-18.6-10.7-15.2-21.1l31.1-93.4L57.5 323.3c-10.7 14.1-30.8 16.8-44.8 6.2s-16.8-30.7-6.2-44.8L65.4 207c22.4-29.6 57.5-47 94.6-47s72.2 17.4 94.6 47l58.9 77.7c10.7 14.1 7.9 34.2-6.2 44.8s-34.2 7.9-44.8-6.2l-28.6-37.8L265 378.9c3.5 10.4-4.3 21.1-15.2 21.1H232v80c0 17.7-14.3 32-32 32s-32-14.3-32-32V400H152v80c0 17.7-14.3 32-32 32s-32-14.3-32-32zM480 0a64 64 0 1 1 0 128A64 64 0 1 1 480 0zm-8 384v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V300.5L395.1 321c-9.4 15-29.2 19.4-44.1 10s-19.4-29.2-10-44.1l51.7-82.1c17.6-27.9 48.3-44.9 81.2-44.9h12.3c33 0 63.7 16.9 81.2 44.9L619.1 287c9.4 15 4.9 34.7-10 44.1s-34.7 4.9-44.1-10L552 300.5V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V384H472z\"],chalkboard:[576,512,[\"blackboard\"],\"f51b\",\"M96 32C60.7 32 32 60.7 32 96V384H96V96l384 0V384h64V96c0-35.3-28.7-64-64-64H96zM224 384v32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H416V384c0-17.7-14.3-32-32-32H256c-17.7 0-32 14.3-32 32z\"],\"user-large-slash\":[640,512,[\"user-alt-slash\"],\"f4fa\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L381.9 274c48.5-23.2 82.1-72.7 82.1-130C464 64.5 399.5 0 320 0C250.4 0 192.4 49.3 178.9 114.9L38.8 5.1zM545.5 512H528L284.3 320h-59C136.2 320 64 392.2 64 481.3c0 17 13.8 30.7 30.7 30.7H545.3l.3 0z\"],\"envelope-open\":[512,512,[62135],\"f2b6\",\"M64 208.1L256 65.9 448 208.1v47.4L289.5 373c-9.7 7.2-21.4 11-33.5 11s-23.8-3.9-33.5-11L64 255.5V208.1zM256 0c-12.1 0-23.8 3.9-33.5 11L25.9 156.7C9.6 168.8 0 187.8 0 208.1V448c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V208.1c0-20.3-9.6-39.4-25.9-51.4L289.5 11C279.8 3.9 268.1 0 256 0z\"],\"handshake-simple-slash\":[640,512,[\"handshake-alt-slash\"],\"e05f\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-135-105.8c-1.1-11.3-6.3-22.3-15.3-30.7l-134.2-123-23.4 18.2-26-20.3 77.2-60.1c7-5.4 17-4.2 22.5 2.8s4.2 17-2.8 22.5l-20.9 16.2L550.2 352H592c26.5 0 48-21.5 48-48V176c0-26.5-21.5-48-48-48H516h-4-.7l-3.9-2.5L434.8 79c-15.3-9.8-33.2-15-51.4-15c-21.8 0-43 7.5-60 21.2l-89.7 72.6-25.8-20.3 81.8-66.2c-11.6-4.9-24.1-7.4-36.8-7.4C234 64 215.7 69.6 200 80l-35.5 23.7L38.8 5.1zM413.6 421.9L41.2 128.5C17.9 131.8 0 151.8 0 176V304c0 26.5 21.5 48 48 48H156.2l91.4 83.4c19.6 17.9 49.9 16.5 67.8-3.1c5.5-6.1 9.2-13.2 11.1-20.6l17 15.6c19.5 17.9 49.9 16.6 67.8-2.9c.8-.8 1.5-1.7 2.2-2.6z\"],\"mattress-pillow\":[640,512,[],\"e525\",\"M256 64H64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H256V64zm32 384H576c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H288V448zM64 160c0-17.7 14.3-32 32-32h64c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V160z\"],\"guarani-sign\":[384,512,[],\"e19a\",\"M192 0c-17.7 0-32 14.3-32 32V66.7C69.2 81.9 0 160.9 0 256s69.2 174.1 160 189.3V480c0 17.7 14.3 32 32 32s32-14.3 32-32V445.3c90.8-15.2 160-94.2 160-189.3c0-17.7-14.3-32-32-32H224V132c22.1 5.7 41.8 17.1 57.6 32.6c12.6 12.4 32.9 12.2 45.3-.4s12.2-32.9-.5-45.3C299 92 263.5 73.3 224 66.7V32c0-17.7-14.3-32-32-32zM160 132V380c-55.2-14.2-96-64.3-96-124s40.8-109.8 96-124zM224 380V288h92c-11.6 45-47 80.4-92 92z\"],\"arrows-rotate\":[512,512,[128472,\"refresh\",\"sync\"],\"f021\",\"M105.1 202.6c7.7-21.8 20.2-42.3 37.8-59.8c62.5-62.5 163.8-62.5 226.3 0L386.3 160H352c-17.7 0-32 14.3-32 32s14.3 32 32 32H463.5c0 0 0 0 0 0h.4c17.7 0 32-14.3 32-32V80c0-17.7-14.3-32-32-32s-32 14.3-32 32v35.2L414.4 97.6c-87.5-87.5-229.3-87.5-316.8 0C73.2 122 55.6 150.7 44.8 181.4c-5.9 16.7 2.9 34.9 19.5 40.8s34.9-2.9 40.8-19.5zM39 289.3c-5 1.5-9.8 4.2-13.7 8.2c-4 4-6.7 8.8-8.1 14c-.3 1.2-.6 2.5-.8 3.8c-.3 1.7-.4 3.4-.4 5.1V432c0 17.7 14.3 32 32 32s32-14.3 32-32V396.9l17.6 17.5 0 0c87.5 87.4 229.3 87.4 316.7 0c24.4-24.4 42.1-53.1 52.9-83.7c5.9-16.7-2.9-34.9-19.5-40.8s-34.9 2.9-40.8 19.5c-7.7 21.8-20.2 42.3-37.8 59.8c-62.5 62.5-163.8 62.5-226.3 0l-.1-.1L125.6 352H160c17.7 0 32-14.3 32-32s-14.3-32-32-32H48.4c-1.6 0-3.2 .1-4.8 .3s-3.1 .5-4.6 1z\"],\"fire-extinguisher\":[512,512,[129519],\"f134\",\"M500.3 7.3C507.7 13.3 512 22.4 512 32v96c0 9.6-4.3 18.7-11.7 24.7s-17.2 8.5-26.6 6.6l-160-32C301.5 124.9 292 115.7 289 104H224v34.8c37.8 18 64 56.5 64 101.2V384H64V240c0-44.7 26.2-83.2 64-101.2V110c-36.2 11.1-66 36.9-82.3 70.5c-5.8 11.9-20.2 16.9-32.1 11.1S-3.3 171.4 2.5 159.5C26.7 109.8 72.7 72.6 128 60.4V32c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V56h65c3-11.7 12.5-20.9 24.7-23.4l160-32c9.4-1.9 19.1 .6 26.6 6.6zM288 416v32c0 35.3-28.7 64-64 64H128c-35.3 0-64-28.7-64-64V416H288zM176 96a16 16 0 1 0 0-32 16 16 0 1 0 0 32z\"],\"cruzeiro-sign\":[448,512,[],\"e152\",\"M96 256c0-88.4 71.6-160 160-160c41 0 78.3 15.4 106.7 40.7c13.2 11.8 33.4 10.7 45.2-2.5s10.7-33.4-2.5-45.2c-39.6-35.5-92-57-149.3-57C132.3 32 32 132.3 32 256s100.3 224 224 224c57.4 0 109.7-21.6 149.3-57c13.2-11.8 14.3-32 2.5-45.2s-32-14.3-45.2-2.5C334.3 400.6 297 416 256 416V320v-8.7c0-12.8 10.4-23.3 23.3-23.3c4.6 0 9.1 1.4 12.9 3.9l10.1 6.7c14.7 9.8 34.6 5.8 44.4-8.9s5.8-34.6-8.9-44.4l-10.1-6.7c-14.3-9.6-31.2-14.7-48.4-14.7c-12.4 0-24.2 2.6-34.9 7.3c-5.5-4.5-12.6-7.3-20.3-7.3c-17.7 0-32 14.3-32 32v55.3V320v82.7C135.5 378 96 321.6 96 256z\"],\"greater-than-equal\":[448,512,[],\"f532\",\"M52.1 93.7C35.7 87.1 27.7 68.5 34.3 52.1s25.2-24.4 41.6-17.8l320 128C408 167.1 416 178.9 416 192s-8 24.9-20.1 29.7l-320 128c-16.4 6.6-35-1.4-41.6-17.8s1.4-35 17.8-41.6L297.8 192 52.1 93.7zM416 416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H416z\"],\"shield-halved\":[512,512,[\"shield-alt\"],\"f3ed\",\"M256 0c4.6 0 9.2 1 13.4 2.9L457.7 82.8c22 9.3 38.4 31 38.3 57.2c-.5 99.2-41.3 280.7-213.6 363.2c-16.7 8-36.1 8-52.8 0C57.3 420.7 16.5 239.2 16 140c-.1-26.2 16.3-47.9 38.3-57.2L242.7 2.9C246.8 1 251.4 0 256 0zm0 66.8V444.8C394 378 431.1 230.1 432 141.4L256 66.8l0 0z\"],\"book-atlas\":[448,512,[\"atlas\"],\"f558\",\"M0 96C0 43 43 0 96 0H384h32c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384 96c-53 0-96-43-96-96V96zM64 416c0 17.7 14.3 32 32 32H352V384H96c-17.7 0-32 14.3-32 32zM247.4 283.8c-3.7 3.7-6.2 4.2-7.4 4.2s-3.7-.5-7.4-4.2c-3.8-3.7-8-10-11.8-18.9c-6.2-14.5-10.8-34.3-12.2-56.9h63c-1.5 22.6-6 42.4-12.2 56.9c-3.8 8.9-8 15.2-11.8 18.9zm42.7-9.9c7.3-18.3 12-41.1 13.4-65.9h31.1c-4.7 27.9-21.4 51.7-44.5 65.9zm0-163.8c23.2 14.2 39.9 38 44.5 65.9H303.5c-1.4-24.7-6.1-47.5-13.4-65.9zM368 192a128 128 0 1 0 -256 0 128 128 0 1 0 256 0zM145.3 208h31.1c1.4 24.7 6.1 47.5 13.4 65.9c-23.2-14.2-39.9-38-44.5-65.9zm31.1-32H145.3c4.7-27.9 21.4-51.7 44.5-65.9c-7.3 18.3-12 41.1-13.4 65.9zm56.1-75.8c3.7-3.7 6.2-4.2 7.4-4.2s3.7 .5 7.4 4.2c3.8 3.7 8 10 11.8 18.9c6.2 14.5 10.8 34.3 12.2 56.9h-63c1.5-22.6 6-42.4 12.2-56.9c3.8-8.9 8-15.2 11.8-18.9z\"],virus:[512,512,[],\"e074\",\"M288 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V43.5c0 49.9-60.3 74.9-95.6 39.6L120.2 75C107.7 62.5 87.5 62.5 75 75s-12.5 32.8 0 45.3l8.2 8.2C118.4 163.7 93.4 224 43.5 224H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H43.5c49.9 0 74.9 60.3 39.6 95.6L75 391.8c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l8.2-8.2c35.3-35.3 95.6-10.3 95.6 39.6V480c0 17.7 14.3 32 32 32s32-14.3 32-32V468.5c0-49.9 60.3-74.9 95.6-39.6l8.2 8.2c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-8.2-8.2c-35.3-35.3-10.3-95.6 39.6-95.6H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H468.5c-49.9 0-74.9-60.3-39.6-95.6l8.2-8.2c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-8.2 8.2C348.3 118.4 288 93.4 288 43.5V32zM176 224a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm128 56a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],\"envelope-circle-check\":[640,512,[],\"e4e8\",\"M48 64C21.5 64 0 85.5 0 112c0 15.1 7.1 29.3 19.2 38.4L236.8 313.6c11.4 8.5 27 8.5 38.4 0l57.4-43c23.9-59.8 79.7-103.3 146.3-109.8l13.9-10.4c12.1-9.1 19.2-23.3 19.2-38.4c0-26.5-21.5-48-48-48H48zM294.4 339.2c-22.8 17.1-54 17.1-76.8 0L0 176V384c0 35.3 28.7 64 64 64H360.2C335.1 417.6 320 378.5 320 336c0-5.6 .3-11.1 .8-16.6l-26.4 19.8zM640 336a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 353.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z\"],\"layer-group\":[576,512,[],\"f5fd\",\"M264.5 5.2c14.9-6.9 32.1-6.9 47 0l218.6 101c8.5 3.9 13.9 12.4 13.9 21.8s-5.4 17.9-13.9 21.8l-218.6 101c-14.9 6.9-32.1 6.9-47 0L45.9 149.8C37.4 145.8 32 137.3 32 128s5.4-17.9 13.9-21.8L264.5 5.2zM476.9 209.6l53.2 24.6c8.5 3.9 13.9 12.4 13.9 21.8s-5.4 17.9-13.9 21.8l-218.6 101c-14.9 6.9-32.1 6.9-47 0L45.9 277.8C37.4 273.8 32 265.3 32 256s5.4-17.9 13.9-21.8l53.2-24.6 152 70.2c23.4 10.8 50.4 10.8 73.8 0l152-70.2zm-152 198.2l152-70.2 53.2 24.6c8.5 3.9 13.9 12.4 13.9 21.8s-5.4 17.9-13.9 21.8l-218.6 101c-14.9 6.9-32.1 6.9-47 0L45.9 405.8C37.4 401.8 32 393.3 32 384s5.4-17.9 13.9-21.8l53.2-24.6 152 70.2c23.4 10.8 50.4 10.8 73.8 0z\"],\"arrows-to-dot\":[512,512,[],\"e4be\",\"M256 0c17.7 0 32 14.3 32 32V64h32c12.9 0 24.6 7.8 29.6 19.8s2.2 25.7-6.9 34.9l-64 64c-12.5 12.5-32.8 12.5-45.3 0l-64-64c-9.2-9.2-11.9-22.9-6.9-34.9s16.6-19.8 29.6-19.8h32V32c0-17.7 14.3-32 32-32zM169.4 393.4l64-64c12.5-12.5 32.8-12.5 45.3 0l64 64c9.2 9.2 11.9 22.9 6.9 34.9s-16.6 19.8-29.6 19.8H288v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H192c-12.9 0-24.6-7.8-29.6-19.8s-2.2-25.7 6.9-34.9zM32 224H64V192c0-12.9 7.8-24.6 19.8-29.6s25.7-2.2 34.9 6.9l64 64c12.5 12.5 12.5 32.8 0 45.3l-64 64c-9.2 9.2-22.9 11.9-34.9 6.9s-19.8-16.6-19.8-29.6V288H32c-17.7 0-32-14.3-32-32s14.3-32 32-32zm297.4 54.6c-12.5-12.5-12.5-32.8 0-45.3l64-64c9.2-9.2 22.9-11.9 34.9-6.9s19.8 16.6 19.8 29.6v32h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H448v32c0 12.9-7.8 24.6-19.8 29.6s-25.7 2.2-34.9-6.9l-64-64zM256 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],archway:[512,512,[],\"f557\",\"M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H32zm0 384c-17.7 0-32 14.3-32 32s14.3 32 32 32H96h64V352c0-53 43-96 96-96s96 43 96 96V480h64 64c17.7 0 32-14.3 32-32s-14.3-32-32-32V128H32V416z\"],\"heart-circle-check\":[576,512,[],\"e4fd\",\"M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9l2.6-2.4C267.2 438.6 256 404.6 256 368c0-97.2 78.8-176 176-176c28.3 0 55 6.7 78.7 18.5c.9-6.5 1.3-13 1.3-19.6v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5zM576 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L416 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z\"],\"house-chimney-crack\":[576,512,[\"house-damage\"],\"f6f1\",\"M575.8 255.5c0 18-15 32.1-32 32.1h-32l.7 160.2c.2 35.5-28.5 64.3-64 64.3H326.4L288 448l80.8-67.3c7.8-6.5 7.6-18.6-.4-24.9L250.6 263.2c-14.6-11.5-33.8 7-22.8 22L288 368l-85.5 71.2c-6.1 5-7.5 13.8-3.5 20.5L230.4 512H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L416 100.7V64c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V185l52.8 46.4c8 7 12 15 11 24z\"],\"file-zipper\":[384,512,[\"file-archive\"],\"f1c6\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM96 48c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16zm0 64c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16zm0 64c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16zm-6.3 71.8c3.7-14 16.4-23.8 30.9-23.8h14.8c14.5 0 27.2 9.7 30.9 23.8l23.5 88.2c1.4 5.4 2.1 10.9 2.1 16.4c0 35.2-28.8 63.7-64 63.7s-64-28.5-64-63.7c0-5.5 .7-11.1 2.1-16.4l23.5-88.2zM112 336c-8.8 0-16 7.2-16 16s7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H112z\"],square:[448,512,[9632,9723,9724,61590],\"f0c8\",\"M0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96z\"],\"martini-glass-empty\":[512,512,[\"glass-martini\"],\"f000\",\"M32 0C19.1 0 7.4 7.8 2.4 19.8s-2.2 25.7 6.9 34.9L224 269.3V448H160c-17.7 0-32 14.3-32 32s14.3 32 32 32h96 96c17.7 0 32-14.3 32-32s-14.3-32-32-32H288V269.3L502.6 54.6c9.2-9.2 11.9-22.9 6.9-34.9S492.9 0 480 0H32zM256 210.7L109.3 64H402.7L256 210.7z\"],couch:[640,512,[],\"f4b8\",\"M64 160C64 89.3 121.3 32 192 32H448c70.7 0 128 57.3 128 128v33.6c-36.5 7.4-64 39.7-64 78.4v48H128V272c0-38.7-27.5-71-64-78.4V160zM544 272c0-20.9 13.4-38.7 32-45.3c5-1.8 10.4-2.7 16-2.7c26.5 0 48 21.5 48 48V448c0 17.7-14.3 32-32 32H576c-17.7 0-32-14.3-32-32H96c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32V272c0-26.5 21.5-48 48-48c5.6 0 11 1 16 2.7c18.6 6.6 32 24.4 32 45.3v48 32h32H512h32V320 272z\"],\"cedi-sign\":[384,512,[],\"e0df\",\"M256 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V66.7C101.2 81.9 32 160.9 32 256s69.2 174.1 160 189.3V480c0 17.7 14.3 32 32 32s32-14.3 32-32V445.3c30.9-5.2 59.2-17.7 83.2-35.8c14.1-10.6 17-30.7 6.4-44.8s-30.7-17-44.8-6.4c-13.2 9.9-28.3 17.3-44.8 21.6V132c16.4 4.2 31.6 11.6 44.8 21.6c14.1 10.6 34.2 7.8 44.8-6.4s7.8-34.2-6.4-44.8c-24-18-52.4-30.6-83.2-35.8V32zM192 132V380c-55.2-14.2-96-64.3-96-124s40.8-109.8 96-124z\"],italic:[384,512,[],\"f033\",\"M128 64c0-17.7 14.3-32 32-32H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H293.3L160 416h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H90.7L224 96H160c-17.7 0-32-14.3-32-32z\"],\"table-cells-column-lock\":[640,512,[],\"e678\",\"M0 96C0 60.7 28.7 32 64 32l384 0c35.3 0 64 28.7 64 64l0 65.1c-37.8 5.4-69.4 29.6-85.2 62.9L360 224l0 64 56 0 0 8.6c-19.1 11.1-32 31.7-32 55.4l-24 0 0 64 24 0 0 64L64 480c-35.3 0-64-28.7-64-64L0 96zm208 0l0 64 88 0 0-64-88 0zm240 0l-88 0 0 64 88 0 0-64zM208 224l0 64 88 0 0-64-88 0zm0 128l0 64 88 0 0-64-88 0zM528 240c-17.7 0-32 14.3-32 32l0 48 64 0 0-48c0-17.7-14.3-32-32-32zm-80 32c0-44.2 35.8-80 80-80s80 35.8 80 80l0 48c17.7 0 32 14.3 32 32l0 128c0 17.7-14.3 32-32 32l-160 0c-17.7 0-32-14.3-32-32l0-128c0-17.7 14.3-32 32-32l0-48z\"],church:[640,512,[9962],\"f51d\",\"M344 24c0-13.3-10.7-24-24-24s-24 10.7-24 24V48H264c-13.3 0-24 10.7-24 24s10.7 24 24 24h32v46.4L183.3 210c-14.5 8.7-23.3 24.3-23.3 41.2V512h96V416c0-35.3 28.7-64 64-64s64 28.7 64 64v96h96V251.2c0-16.9-8.8-32.5-23.3-41.2L344 142.4V96h32c13.3 0 24-10.7 24-24s-10.7-24-24-24H344V24zM24.9 330.3C9.5 338.8 0 354.9 0 372.4V464c0 26.5 21.5 48 48 48h80V273.6L24.9 330.3zM592 512c26.5 0 48-21.5 48-48V372.4c0-17.5-9.5-33.6-24.9-42.1L512 273.6V512h80z\"],\"comments-dollar\":[640,512,[],\"f653\",\"M416 176c0 97.2-93.1 176-208 176c-38.2 0-73.9-8.7-104.7-23.9c-7.5 4-16 7.9-25.2 11.4C59.8 346.4 37.8 352 16 352c-6.9 0-13.1-4.5-15.2-11.1s.2-13.8 5.8-17.9l0 0 0 0 .2-.2c.2-.2 .6-.4 1.1-.8c1-.8 2.5-2 4.3-3.7c3.6-3.3 8.5-8.1 13.3-14.3c5.5-7 10.7-15.4 14.2-24.7C14.7 250.3 0 214.6 0 176C0 78.8 93.1 0 208 0S416 78.8 416 176zM231.5 383C348.9 372.9 448 288.3 448 176c0-5.2-.2-10.4-.6-15.5C555.1 167.1 640 243.2 640 336c0 38.6-14.7 74.3-39.6 103.4c3.5 9.4 8.7 17.7 14.2 24.7c4.8 6.2 9.7 11 13.3 14.3c1.8 1.6 3.3 2.9 4.3 3.7c.5 .4 .9 .7 1.1 .8l.2 .2 0 0 0 0c5.6 4.1 7.9 11.3 5.8 17.9c-2.1 6.6-8.3 11.1-15.2 11.1c-21.8 0-43.8-5.6-62.1-12.5c-9.2-3.5-17.8-7.4-25.2-11.4C505.9 503.3 470.2 512 432 512c-95.6 0-176.2-54.6-200.5-129zM228 72c0-11-9-20-20-20s-20 9-20 20V86c-7.6 1.7-15.2 4.4-22.2 8.5c-13.9 8.3-25.9 22.8-25.8 43.9c.1 20.3 12 33.1 24.7 40.7c11 6.6 24.7 10.8 35.6 14l1.7 .5c12.6 3.8 21.8 6.8 28 10.7c5.1 3.2 5.8 5.4 5.9 8.2c.1 5-1.8 8-5.9 10.5c-5 3.1-12.9 5-21.4 4.7c-11.1-.4-21.5-3.9-35.1-8.5c-2.3-.8-4.7-1.6-7.2-2.4c-10.5-3.5-21.8 2.2-25.3 12.6s2.2 21.8 12.6 25.3c1.9 .6 4 1.3 6.1 2.1l0 0 0 0c8.3 2.9 17.9 6.2 28.2 8.4V280c0 11 9 20 20 20s20-9 20-20V266.2c8-1.7 16-4.5 23.2-9c14.3-8.9 25.1-24.1 24.8-45c-.3-20.3-11.7-33.4-24.6-41.6c-11.5-7.2-25.9-11.6-37.1-15l-.7-.2c-12.8-3.9-21.9-6.7-28.3-10.5c-5.2-3.1-5.3-4.9-5.3-6.7c0-3.7 1.4-6.5 6.2-9.3c5.4-3.2 13.6-5.1 21.5-5c9.6 .1 20.2 2.2 31.2 5.2c10.7 2.8 21.6-3.5 24.5-14.2s-3.5-21.6-14.2-24.5c-6.5-1.7-13.7-3.4-21.1-4.7V72z\"],democrat:[640,512,[],\"f747\",\"M64 32c0-8.9 3.8-20.9 6.2-27.3C71.2 1.8 74 0 77 0c1.9 0 3.8 .7 5.2 2.1L128 45.7 173.8 2.1C175.2 .7 177.1 0 179 0c3 0 5.8 1.8 6.8 4.7c2.4 6.5 6.2 18.4 6.2 27.3c0 26.5-21.9 42-29.5 46.6l76.2 72.6c6 5.7 13.9 8.8 22.1 8.8H480l32 0c40.3 0 78.2 19 102.4 51.2l19.2 25.6c10.6 14.1 7.7 34.2-6.4 44.8s-34.2 7.7-44.8-6.4l-19.2-25.6c-5.3-7-11.8-12.8-19.2-17V320H192l-40.4-94.3c-3.9-9.2-15.3-12.6-23.6-7l-42.1 28c-9.1 6.1-19.7 9.3-30.7 9.3h-2C23.9 256 0 232.1 0 202.7c0-12.1 4.1-23.8 11.7-33.3L87.6 74.6C78.1 67.4 64 53.2 64 32zM448 352h96v64 64c0 17.7-14.3 32-32 32H480c-17.7 0-32-14.3-32-32V416H288v64c0 17.7-14.3 32-32 32H224c-17.7 0-32-14.3-32-32V416 352h96H448zM260.9 210.9c-.9-1.8-2.8-2.9-4.8-2.9s-3.9 1.1-4.8 2.9l-10.5 20.5-23.5 3.3c-2 .3-3.7 1.6-4.3 3.5s-.1 3.9 1.3 5.3l17 16-4 22.6c-.3 1.9 .5 3.9 2.1 5s3.8 1.3 5.6 .4l21-10.7 21 10.7c1.8 .9 4 .8 5.6-.4s2.5-3.1 2.1-5l-4-22.6 17-16c1.5-1.4 2-3.4 1.3-5.3s-2.3-3.2-4.3-3.5l-23.5-3.3-10.5-20.5zM368.1 208c-2 0-3.9 1.1-4.8 2.9l-10.5 20.5-23.5 3.3c-2 .3-3.7 1.6-4.3 3.5s-.1 3.9 1.3 5.3l17 16-4 22.6c-.3 1.9 .5 3.9 2.1 5s3.8 1.3 5.6 .4l21-10.7 21 10.7c1.8 .9 4 .8 5.6-.4s2.5-3.1 2.1-5l-4-22.6 17-16c1.5-1.4 2-3.4 1.4-5.3s-2.3-3.2-4.3-3.5l-23.5-3.3-10.5-20.5c-.9-1.8-2.8-2.9-4.8-2.9zm116.8 2.9c-.9-1.8-2.8-2.9-4.8-2.9s-3.9 1.1-4.8 2.9l-10.5 20.5-23.5 3.3c-2 .3-3.7 1.6-4.3 3.5s-.1 3.9 1.3 5.3l17 16-4 22.6c-.3 1.9 .5 3.9 2.1 5s3.8 1.3 5.6 .4l21-10.7 21 10.7c1.8 .9 4 .8 5.6-.4s2.5-3.1 2.1-5l-4-22.6 17-16c1.5-1.4 2-3.4 1.4-5.3s-2.3-3.2-4.3-3.5l-23.5-3.3-10.5-20.5z\"],z:[384,512,[122],\"5a\",\"M0 64C0 46.3 14.3 32 32 32H352c12.4 0 23.7 7.2 29 18.4s3.6 24.5-4.4 34.1L100.3 416H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-12.4 0-23.7-7.2-29-18.4s-3.6-24.5 4.4-34.1L283.7 96H32C14.3 96 0 81.7 0 64z\"],\"person-skiing\":[512,512,[9975,\"skiing\"],\"f7c9\",\"M380.7 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM2.7 268.9c6.1-11.8 20.6-16.3 32.4-10.2L232.7 361.3l46.2-69.2-75.1-75.1c-14.6-14.6-20.4-33.9-18.4-52.1l108.8 52 39.3 39.3c16.2 16.2 18.7 41.5 6 60.6L289.8 391l128.7 66.8c13.6 7.1 29.8 7.2 43.6 .3l15.2-7.6c11.9-5.9 26.3-1.1 32.2 10.7s1.1 26.3-10.7 32.2l-15.2 7.6c-27.5 13.7-59.9 13.5-87.2-.7L12.9 301.3C1.2 295.2-3.4 280.7 2.7 268.9zM118.9 65.6L137 74.2l8.7-17.4c4-7.9 13.6-11.1 21.5-7.2s11.1 13.6 7.2 21.5l-8.5 16.9 54.7 26.2c1.5-.7 3.1-1.4 4.7-2.1l83.4-33.4c34.2-13.7 72.8 4.2 84.5 39.2l17.1 51.2 52.1 26.1c15.8 7.9 22.2 27.1 14.3 42.9s-27.1 22.2-42.9 14.3l-58.1-29c-11.4-5.7-20-15.7-24.1-27.8l-5.8-17.3-27.3 12.1-6.8 3-6.7-3.2L151.5 116.7l-9.2 18.4c-4 7.9-13.6 11.1-21.5 7.2s-11.1-13.6-7.2-21.5l9-18-17.6-8.4c-8-3.8-11.3-13.4-7.5-21.3s13.4-11.3 21.3-7.5z\"],\"road-lock\":[640,512,[],\"e567\",\"M288 32H213.2c-27.1 0-51.3 17.1-60.3 42.6L35.1 407.2c-2.1 5.9-3.1 12-3.1 18.2C32 455.5 56.5 480 86.6 480H288V416c0-17.7 14.3-32 32-32s32 14.3 32 32v64h32V352c0-23.7 12.9-44.4 32-55.4V272c0-58.3 44.6-106.2 101.5-111.5L487.1 74.6C478 49.1 453.9 32 426.8 32H352V96c0 17.7-14.3 32-32 32s-32-14.3-32-32V32zm64 192v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V224c0-17.7 14.3-32 32-32s32 14.3 32 32zm176 16c17.7 0 32 14.3 32 32v48H496V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z\"],a:[384,512,[97],\"41\",\"M221.5 51.7C216.6 39.8 204.9 32 192 32s-24.6 7.8-29.5 19.7l-120 288-40 96c-6.8 16.3 .9 35 17.2 41.8s35-.9 41.8-17.2L93.3 384H290.7l31.8 76.3c6.8 16.3 25.5 24 41.8 17.2s24-25.5 17.2-41.8l-40-96-120-288zM264 320H120l72-172.8L264 320z\"],\"temperature-arrow-down\":[576,512,[\"temperature-down\"],\"e03f\",\"M128 112c0-26.5 21.5-48 48-48s48 21.5 48 48V276.5c0 17.3 7.1 31.9 15.3 42.5C249.8 332.6 256 349.5 256 368c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-18.5 6.2-35.4 16.7-48.9c8.2-10.6 15.3-25.2 15.3-42.5V112zM176 0C114.1 0 64 50.1 64 112V276.4c0 .1-.1 .3-.2 .6c-.2 .6-.8 1.6-1.7 2.8C43.2 304.2 32 334.8 32 368c0 79.5 64.5 144 144 144s144-64.5 144-144c0-33.2-11.2-63.8-30.1-88.1c-.9-1.2-1.5-2.2-1.7-2.8c-.1-.3-.2-.5-.2-.6V112C288 50.1 237.9 0 176 0zm0 416c26.5 0 48-21.5 48-48c0-20.9-13.4-38.7-32-45.3V272c0-8.8-7.2-16-16-16s-16 7.2-16 16v50.7c-18.6 6.6-32 24.4-32 45.3c0 26.5 21.5 48 48 48zm336-64H480V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V352H384c-12.9 0-24.6 7.8-29.6 19.8s-2.2 25.7 6.9 34.9l64 64c6 6 14.1 9.4 22.6 9.4s16.6-3.4 22.6-9.4l64-64c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8z\"],\"feather-pointed\":[512,512,[\"feather-alt\"],\"f56b\",\"M278.5 215.6L23 471c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l74.8-74.8c7.4 4.6 15.3 8.2 23.8 10.5C200.3 452.8 270 454.5 338 409.4c12.2-8.1 5.8-25.4-8.8-25.4l-16.1 0c-5.1 0-9.2-4.1-9.2-9.2c0-4.1 2.7-7.6 6.5-8.8l97.7-29.3c3.4-1 6.4-3.1 8.4-6.1c4.4-6.4 8.6-12.9 12.6-19.6c6.2-10.3-1.5-23-13.5-23l-38.6 0c-5.1 0-9.2-4.1-9.2-9.2c0-4.1 2.7-7.6 6.5-8.8l80.9-24.3c4.6-1.4 8.4-4.8 10.2-9.3C494.5 163 507.8 86.1 511.9 36.8c.8-9.9-3-19.6-10-26.6s-16.7-10.8-26.6-10C391.5 7 228.5 40.5 137.4 131.6C57.3 211.7 56.7 302.3 71.3 356.4c2.1 7.9 12 9.6 17.8 3.8L253.6 195.8c6.2-6.2 16.4-6.2 22.6 0c5.4 5.4 6.1 13.6 2.2 19.8z\"],p:[320,512,[112],\"50\",\"M0 96C0 60.7 28.7 32 64 32h96c88.4 0 160 71.6 160 160s-71.6 160-160 160H64v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V320 96zM64 288h96c53 0 96-43 96-96s-43-96-96-96H64V288z\"],snowflake:[448,512,[10052,10054],\"f2dc\",\"M224 0c17.7 0 32 14.3 32 32V62.1l15-15c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-49 49v70.3l61.4-35.8 17.7-66.1c3.4-12.8 16.6-20.4 29.4-17s20.4 16.6 17 29.4l-5.2 19.3 23.6-13.8c15.3-8.9 34.9-3.7 43.8 11.5s3.8 34.9-11.5 43.8l-25.3 14.8 21.7 5.8c12.8 3.4 20.4 16.6 17 29.4s-16.6 20.4-29.4 17l-67.7-18.1L287.5 256l60.9 35.5 67.7-18.1c12.8-3.4 26 4.2 29.4 17s-4.2 26-17 29.4l-21.7 5.8 25.3 14.8c15.3 8.9 20.4 28.5 11.5 43.8s-28.5 20.4-43.8 11.5l-23.6-13.8 5.2 19.3c3.4 12.8-4.2 26-17 29.4s-26-4.2-29.4-17l-17.7-66.1L256 311.7v70.3l49 49c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-15-15V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V449.9l-15 15c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l49-49V311.7l-61.4 35.8-17.7 66.1c-3.4 12.8-16.6 20.4-29.4 17s-20.4-16.6-17-29.4l5.2-19.3L48.1 395.6c-15.3 8.9-34.9 3.7-43.8-11.5s-3.7-34.9 11.5-43.8l25.3-14.8-21.7-5.8c-12.8-3.4-20.4-16.6-17-29.4s16.6-20.4 29.4-17l67.7 18.1L160.5 256 99.6 220.5 31.9 238.6c-12.8 3.4-26-4.2-29.4-17s4.2-26 17-29.4l21.7-5.8L15.9 171.6C.6 162.7-4.5 143.1 4.4 127.9s28.5-20.4 43.8-11.5l23.6 13.8-5.2-19.3c-3.4-12.8 4.2-26 17-29.4s26 4.2 29.4 17l17.7 66.1L192 200.3V129.9L143 81c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l15 15V32c0-17.7 14.3-32 32-32z\"],newspaper:[512,512,[128240],\"f1ea\",\"M96 96c0-35.3 28.7-64 64-64H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H80c-44.2 0-80-35.8-80-80V128c0-17.7 14.3-32 32-32s32 14.3 32 32V400c0 8.8 7.2 16 16 16s16-7.2 16-16V96zm64 24v80c0 13.3 10.7 24 24 24H296c13.3 0 24-10.7 24-24V120c0-13.3-10.7-24-24-24H184c-13.3 0-24 10.7-24 24zm208-8c0 8.8 7.2 16 16 16h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H384c-8.8 0-16 7.2-16 16zm0 96c0 8.8 7.2 16 16 16h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H384c-8.8 0-16 7.2-16 16zM160 304c0 8.8 7.2 16 16 16H432c8.8 0 16-7.2 16-16s-7.2-16-16-16H176c-8.8 0-16 7.2-16 16zm0 96c0 8.8 7.2 16 16 16H432c8.8 0 16-7.2 16-16s-7.2-16-16-16H176c-8.8 0-16 7.2-16 16z\"],\"rectangle-ad\":[576,512,[\"ad\"],\"f641\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM229.5 173.3l72 144c5.9 11.9 1.1 26.3-10.7 32.2s-26.3 1.1-32.2-10.7L253.2 328H162.8l-5.4 10.7c-5.9 11.9-20.3 16.7-32.2 10.7s-16.7-20.3-10.7-32.2l72-144c4.1-8.1 12.4-13.3 21.5-13.3s17.4 5.1 21.5 13.3zM208 237.7L186.8 280h42.3L208 237.7zM392 256a24 24 0 1 0 0 48 24 24 0 1 0 0-48zm24-43.9V184c0-13.3 10.7-24 24-24s24 10.7 24 24v96 48c0 13.3-10.7 24-24 24c-6.6 0-12.6-2.7-17-7c-9.4 4.5-19.9 7-31 7c-39.8 0-72-32.2-72-72s32.2-72 72-72c8.4 0 16.5 1.4 24 4.1z\"],\"circle-arrow-right\":[512,512,[\"arrow-circle-right\"],\"f0a9\",\"M0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zM281 385c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l71-71L136 280c-13.3 0-24-10.7-24-24s10.7-24 24-24l182.1 0-71-71c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0L393 239c9.4 9.4 9.4 24.6 0 33.9L281 385z\"],\"filter-circle-xmark\":[576,512,[],\"e17b\",\"M3.9 22.9C10.5 8.9 24.5 0 40 0H472c15.5 0 29.5 8.9 36.1 22.9s4.6 30.5-5.2 42.5L396.4 195.6C316.2 212.1 256 283 256 368c0 27.4 6.3 53.4 17.5 76.5c-1.6-.8-3.2-1.8-4.7-2.9l-64-48c-8.1-6-12.8-15.5-12.8-25.6V288.9L9 65.3C-.7 53.4-2.8 36.8 3.9 22.9zM432 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm59.3 107.3c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0L432 345.4l-36.7-36.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6L409.4 368l-36.7 36.7c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L432 390.6l36.7 36.7c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6L454.6 368l36.7-36.7z\"],locust:[576,512,[],\"e520\",\"M312 32c-13.3 0-24 10.7-24 24s10.7 24 24 24h16c98.7 0 180.6 71.4 197 165.4c-9-3.5-18.8-5.4-29-5.4H431.8l-41.8-97.5c-3.4-7.9-10.8-13.4-19.3-14.4s-17 2.7-22.1 9.6l-40.9 55.5-21.7-50.7c-3.3-7.8-10.5-13.2-18.9-14.3s-16.7 2.3-22 8.9l-240 304c-8.2 10.4-6.4 25.5 4 33.7s25.5 6.4 33.7-4l79.4-100.5 43 16.4-40.5 55c-7.9 10.7-5.6 25.7 5.1 33.6s25.7 5.6 33.6-5.1L215.1 400h74.5l-29.3 42.3c-7.5 10.9-4.8 25.8 6.1 33.4s25.8 4.8 33.4-6.1L348 400h80.4l38.8 67.9c6.6 11.5 21.2 15.5 32.7 8.9s15.5-21.2 8.9-32.7L483.6 400H496c44.1 0 79.8-35.7 80-79.7c0-.1 0-.2 0-.3V280C576 143 465 32 328 32H312zm50.5 168l17.1 40H333l29.5-40zm-87.7 38.1l-1.4 1.9H225.1l32.7-41.5 16.9 39.5zM88.8 240C57.4 240 32 265.4 32 296.8c0 15.5 6.3 30 16.9 40.4L126.7 240H88.8zM496 288a16 16 0 1 1 0 32 16 16 0 1 1 0-32z\"],sort:[320,512,[\"unsorted\"],\"f0dc\",\"M137.4 41.4c12.5-12.5 32.8-12.5 45.3 0l128 128c9.2 9.2 11.9 22.9 6.9 34.9s-16.6 19.8-29.6 19.8H32c-12.9 0-24.6-7.8-29.6-19.8s-2.2-25.7 6.9-34.9l128-128zm0 429.3l-128-128c-9.2-9.2-11.9-22.9-6.9-34.9s16.6-19.8 29.6-19.8H288c12.9 0 24.6 7.8 29.6 19.8s2.2 25.7-6.9 34.9l-128 128c-12.5 12.5-32.8 12.5-45.3 0z\"],\"list-ol\":[512,512,[\"list-1-2\",\"list-numeric\"],\"f0cb\",\"M24 56c0-13.3 10.7-24 24-24H80c13.3 0 24 10.7 24 24V176h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H40c-13.3 0-24-10.7-24-24s10.7-24 24-24H56V80H48C34.7 80 24 69.3 24 56zM86.7 341.2c-6.5-7.4-18.3-6.9-24 1.2L51.5 357.9c-7.7 10.8-22.7 13.3-33.5 5.6s-13.3-22.7-5.6-33.5l11.1-15.6c23.7-33.2 72.3-35.6 99.2-4.9c21.3 24.4 20.8 60.9-1.1 84.7L86.8 432H120c13.3 0 24 10.7 24 24s-10.7 24-24 24H32c-9.5 0-18.2-5.6-22-14.4s-2.1-18.9 4.3-25.9l72-78c5.3-5.8 5.4-14.6 .3-20.5zM224 64H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32s14.3-32 32-32zm0 160H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32s14.3-32 32-32zm0 160H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],\"person-dress-burst\":[640,512,[],\"e544\",\"M528 48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM390.2 384H408v96c0 17.7 14.3 32 32 32s32-14.3 32-32V384h16v96c0 17.7 14.3 32 32 32s32-14.3 32-32V384h17.8c10.9 0 18.6-10.7 15.2-21.1L546.7 248.1l33.9 56.3c9.1 15.1 28.8 20 43.9 10.9s20-28.8 10.9-43.9l-53.6-89.2c-20.2-33.7-56.7-54.3-96-54.3H474.2c-39.3 0-75.7 20.6-96 54.3l-53.6 89.2c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9l33.9-56.3L375 362.9c-3.5 10.4 4.3 21.1 15.2 21.1zM190.9 18.1C188.4 12 182.6 8 176 8s-12.4 4-14.9 10.1l-29.4 74L55.6 68.9c-6.3-1.9-13.1 .2-17.2 5.3s-4.6 12.2-1.4 17.9l39.5 69.1L10.9 206.4c-5.4 3.7-8 10.3-6.5 16.7s6.7 11.2 13.1 12.2l78.7 12.2L90.6 327c-.5 6.5 3.1 12.7 9 15.5s12.9 1.8 17.8-2.6L176 286.1l58.6 53.9c4.8 4.4 11.9 5.5 17.8 2.6s9.5-9 9-15.5l-5.6-79.4 50.5-7.8 24.4-40.5-55.2-38L315 92.2c3.3-5.7 2.7-12.8-1.4-17.9s-10.9-7.2-17.2-5.3L220.3 92.1l-29.4-74z\"],\"money-check-dollar\":[576,512,[\"money-check-alt\"],\"f53d\",\"M64 64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H64zM272 192H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H272c-8.8 0-16-7.2-16-16s7.2-16 16-16zM256 304c0-8.8 7.2-16 16-16H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H272c-8.8 0-16-7.2-16-16zM164 152v13.9c7.5 1.2 14.6 2.9 21.1 4.7c10.7 2.8 17 13.8 14.2 24.5s-13.8 17-24.5 14.2c-11-2.9-21.6-5-31.2-5.2c-7.9-.1-16 1.8-21.5 5c-4.8 2.8-6.2 5.6-6.2 9.3c0 1.8 .1 3.5 5.3 6.7c6.3 3.8 15.5 6.7 28.3 10.5l.7 .2c11.2 3.4 25.6 7.7 37.1 15c12.9 8.1 24.3 21.3 24.6 41.6c.3 20.9-10.5 36.1-24.8 45c-7.2 4.5-15.2 7.3-23.2 9V360c0 11-9 20-20 20s-20-9-20-20V345.4c-10.3-2.2-20-5.5-28.2-8.4l0 0 0 0c-2.1-.7-4.1-1.4-6.1-2.1c-10.5-3.5-16.1-14.8-12.6-25.3s14.8-16.1 25.3-12.6c2.5 .8 4.9 1.7 7.2 2.4c13.6 4.6 24 8.1 35.1 8.5c8.6 .3 16.5-1.6 21.4-4.7c4.1-2.5 6-5.5 5.9-10.5c0-2.9-.8-5-5.9-8.2c-6.3-4-15.4-6.9-28-10.7l-1.7-.5c-10.9-3.3-24.6-7.4-35.6-14c-12.7-7.7-24.6-20.5-24.7-40.7c-.1-21.1 11.8-35.7 25.8-43.9c6.9-4.1 14.5-6.8 22.2-8.5V152c0-11 9-20 20-20s20 9 20 20z\"],\"vector-square\":[448,512,[],\"f5cb\",\"M368 80h32v32H368V80zM352 32c-17.7 0-32 14.3-32 32H128c0-17.7-14.3-32-32-32H32C14.3 32 0 46.3 0 64v64c0 17.7 14.3 32 32 32V352c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H96c17.7 0 32-14.3 32-32H320c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V384c0-17.7-14.3-32-32-32V160c17.7 0 32-14.3 32-32V64c0-17.7-14.3-32-32-32H352zM96 160c17.7 0 32-14.3 32-32H320c0 17.7 14.3 32 32 32V352c-17.7 0-32 14.3-32 32H128c0-17.7-14.3-32-32-32V160zM48 400H80v32H48V400zm320 32V400h32v32H368zM48 112V80H80v32H48z\"],\"bread-slice\":[512,512,[],\"f7ec\",\"M256 32C192 32 0 64 0 192c0 35.3 28.7 64 64 64V432c0 26.5 21.5 48 48 48H400c26.5 0 48-21.5 48-48V256c35.3 0 64-28.7 64-64C512 64 320 32 256 32z\"],language:[640,512,[],\"f1ab\",\"M0 128C0 92.7 28.7 64 64 64H256h48 16H576c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H320 304 256 64c-35.3 0-64-28.7-64-64V128zm320 0V384H576V128H320zM178.3 175.9c-3.2-7.2-10.4-11.9-18.3-11.9s-15.1 4.7-18.3 11.9l-64 144c-4.5 10.1 .1 21.9 10.2 26.4s21.9-.1 26.4-10.2l8.9-20.1h73.6l8.9 20.1c4.5 10.1 16.3 14.6 26.4 10.2s14.6-16.3 10.2-26.4l-64-144zM160 233.2L179 276H141l19-42.8zM448 164c11 0 20 9 20 20v4h44 16c11 0 20 9 20 20s-9 20-20 20h-2l-1.6 4.5c-8.9 24.4-22.4 46.6-39.6 65.4c.9 .6 1.8 1.1 2.7 1.6l18.9 11.3c9.5 5.7 12.5 18 6.9 27.4s-18 12.5-27.4 6.9l-18.9-11.3c-4.5-2.7-8.8-5.5-13.1-8.5c-10.6 7.5-21.9 14-34 19.4l-3.6 1.6c-10.1 4.5-21.9-.1-26.4-10.2s.1-21.9 10.2-26.4l3.6-1.6c6.4-2.9 12.6-6.1 18.5-9.8l-12.2-12.2c-7.8-7.8-7.8-20.5 0-28.3s20.5-7.8 28.3 0l14.6 14.6 .5 .5c12.4-13.1 22.5-28.3 29.8-45H448 376c-11 0-20-9-20-20s9-20 20-20h52v-4c0-11 9-20 20-20z\"],\"face-kiss-wink-heart\":[512,512,[128536,\"kiss-wink-heart\"],\"f598\",\"M498 339.7c9.1-26.2 14-54.4 14-83.7C512 114.6 397.4 0 256 0S0 114.6 0 256S114.6 512 256 512c35.4 0 69.1-7.2 99.7-20.2c-4.8-5.5-8.5-12.2-10.4-19.7l-22.9-89.3c-10-39 11.8-80.9 51.8-92.1c37.2-10.4 73.8 10.1 87.5 44c12.7-1.6 25.1 .4 36.2 5zM296 332c0 6.9-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4c-2.7 1.5-5.7 3-8.7 4.3c3.1 1.3 6 2.7 8.7 4.3c6.6 3.7 12.5 8.2 16.7 13.4c4.3 5.1 7.3 11.4 7.3 18.3s-3.1 13.2-7.3 18.3c-4.3 5.2-10.1 9.7-16.7 13.4C258.7 443.1 241.4 448 224 448c-3.6 0-6.8-2.5-7.7-6s.6-7.2 3.8-9l0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1l-.6-.4-.3-.2-.2-.1 0 0 0 0 0 0c-2.5-1.4-4.1-4.1-4.1-7s1.6-5.6 4.1-7l0 0 0 0 0 0 0 0 0 0 .2-.1c.2-.1 .5-.3 .9-.5c.8-.5 2-1.2 3.4-2.1c2.8-1.9 6.5-4.5 10.2-7.6c3.7-3.1 7.2-6.6 9.6-10.1c2.5-3.5 3.5-6.4 3.5-8.6s-1-5-3.5-8.6c-2.5-3.5-5.9-6.9-9.6-10.1c-3.7-3.1-7.4-5.7-10.2-7.6c-1.4-.9-2.6-1.6-3.4-2.1c-.4-.2-.7-.4-.9-.5l-.2-.1 0 0 0 0 0 0c-3.2-1.8-4.7-5.5-3.8-9s4.1-6 7.7-6c17.4 0 34.7 4.9 47.9 12.3c6.6 3.7 12.5 8.2 16.7 13.4c4.3 5.1 7.3 11.4 7.3 18.3zM176.4 176a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm194.8 57.6c-17.6-23.5-52.8-23.5-70.4 0c-5.3 7.1-15.3 8.5-22.4 3.2s-8.5-15.3-3.2-22.4c30.4-40.5 91.2-40.5 121.6 0c5.3 7.1 3.9 17.1-3.2 22.4s-17.1 3.9-22.4-3.2zM434 352.3c-6-23.2-28.8-37-51.1-30.8s-35.4 30.1-29.5 53.4l22.9 89.3c2.2 8.7 11.2 13.9 19.8 11.4l84.9-23.8c22.2-6.2 35.4-30.1 29.5-53.4s-28.8-37-51.1-30.8l-20.2 5.6-5.4-21z\"],filter:[512,512,[],\"f0b0\",\"M3.9 54.9C10.5 40.9 24.5 32 40 32H472c15.5 0 29.5 8.9 36.1 22.9s4.6 30.5-5.2 42.5L320 320.9V448c0 12.1-6.8 23.2-17.7 28.6s-23.8 4.3-33.5-3l-64-48c-8.1-6-12.8-15.5-12.8-25.6V320.9L9 97.3C-.7 85.4-2.8 68.8 3.9 54.9z\"],question:[320,512,[10067,10068,61736],\"3f\",\"M80 160c0-35.3 28.7-64 64-64h32c35.3 0 64 28.7 64 64v3.6c0 21.8-11.1 42.1-29.4 53.8l-42.2 27.1c-25.2 16.2-40.4 44.1-40.4 74V320c0 17.7 14.3 32 32 32s32-14.3 32-32v-1.4c0-8.2 4.2-15.8 11-20.2l42.2-27.1c36.6-23.6 58.8-64.1 58.8-107.7V160c0-70.7-57.3-128-128-128H144C73.3 32 16 89.3 16 160c0 17.7 14.3 32 32 32s32-14.3 32-32zm80 320a40 40 0 1 0 0-80 40 40 0 1 0 0 80z\"],\"file-signature\":[576,512,[],\"f573\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V428.7c-2.7 1.1-5.4 2-8.2 2.7l-60.1 15c-3 .7-6 1.2-9 1.4c-.9 .1-1.8 .2-2.7 .2H240c-6.1 0-11.6-3.4-14.3-8.8l-8.8-17.7c-1.7-3.4-5.1-5.5-8.8-5.5s-7.2 2.1-8.8 5.5l-8.8 17.7c-2.9 5.9-9.2 9.4-15.7 8.8s-12.1-5.1-13.9-11.3L144 381l-9.8 32.8c-6.1 20.3-24.8 34.2-46 34.2H80c-8.8 0-16-7.2-16-16s7.2-16 16-16h8.2c7.1 0 13.3-4.6 15.3-11.4l14.9-49.5c3.4-11.3 13.8-19.1 25.6-19.1s22.2 7.8 25.6 19.1l11.6 38.6c7.4-6.2 16.8-9.7 26.8-9.7c15.9 0 30.4 9 37.5 23.2l4.4 8.8h8.9c-3.1-8.8-3.7-18.4-1.4-27.8l15-60.1c2.8-11.3 8.6-21.5 16.8-29.7L384 203.6V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM549.8 139.7c-15.6-15.6-40.9-15.6-56.6 0l-29.4 29.4 71 71 29.4-29.4c15.6-15.6 15.6-40.9 0-56.6l-14.4-14.4zM311.9 321c-4.1 4.1-7 9.2-8.4 14.9l-15 60.1c-1.4 5.5 .2 11.2 4.2 15.2s9.7 5.6 15.2 4.2l60.1-15c5.6-1.4 10.8-4.3 14.9-8.4L512.1 262.7l-71-71L311.9 321z\"],\"up-down-left-right\":[512,512,[\"arrows-alt\"],\"f0b2\",\"M278.6 9.4c-12.5-12.5-32.8-12.5-45.3 0l-64 64c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8h32v96H128V192c0-12.9-7.8-24.6-19.8-29.6s-25.7-2.2-34.9 6.9l-64 64c-12.5 12.5-12.5 32.8 0 45.3l64 64c9.2 9.2 22.9 11.9 34.9 6.9s19.8-16.6 19.8-29.6V288h96v96H192c-12.9 0-24.6 7.8-29.6 19.8s-2.2 25.7 6.9 34.9l64 64c12.5 12.5 32.8 12.5 45.3 0l64-64c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8H288V288h96v32c0 12.9 7.8 24.6 19.8 29.6s25.7 2.2 34.9-6.9l64-64c12.5-12.5 12.5-32.8 0-45.3l-64-64c-9.2-9.2-22.9-11.9-34.9-6.9s-19.8 16.6-19.8 29.6v32H288V128h32c12.9 0 24.6-7.8 29.6-19.8s2.2-25.7-6.9-34.9l-64-64z\"],\"house-chimney-user\":[576,512,[],\"e065\",\"M543.8 287.6c17 0 32-14 32-32.1c1-9-3-17-11-24L512 185V64c0-17.7-14.3-32-32-32H448c-17.7 0-32 14.3-32 32v36.7L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32V448c0 35.3 28.7 64 64 64H448.5c35.5 0 64.2-28.8 64-64.3l-.7-160.2h32zM288 160a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM176 400c0-44.2 35.8-80 80-80h64c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16H192c-8.8 0-16-7.2-16-16z\"],\"hand-holding-heart\":[576,512,[],\"f4be\",\"M163.9 136.9c-29.4-29.8-29.4-78.2 0-108s77-29.8 106.4 0l17.7 18 17.7-18c29.4-29.8 77-29.8 106.4 0s29.4 78.2 0 108L310.5 240.1c-6.2 6.3-14.3 9.4-22.5 9.4s-16.3-3.1-22.5-9.4L163.9 136.9zM568.2 336.3c13.1 17.8 9.3 42.8-8.5 55.9L433.1 485.5c-23.4 17.2-51.6 26.5-80.7 26.5H192 32c-17.7 0-32-14.3-32-32V416c0-17.7 14.3-32 32-32H68.8l44.9-36c22.7-18.2 50.9-28 80-28H272h16 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H288 272c-8.8 0-16 7.2-16 16s7.2 16 16 16H392.6l119.7-88.2c17.8-13.1 42.8-9.3 55.9 8.5zM193.6 384l0 0-.9 0c.3 0 .6 0 .9 0z\"],\"puzzle-piece\":[512,512,[129513],\"f12e\",\"M192 104.8c0-9.2-5.8-17.3-13.2-22.8C167.2 73.3 160 61.3 160 48c0-26.5 28.7-48 64-48s64 21.5 64 48c0 13.3-7.2 25.3-18.8 34c-7.4 5.5-13.2 13.6-13.2 22.8c0 12.8 10.4 23.2 23.2 23.2H336c26.5 0 48 21.5 48 48v56.8c0 12.8 10.4 23.2 23.2 23.2c9.2 0 17.3-5.8 22.8-13.2c8.7-11.6 20.7-18.8 34-18.8c26.5 0 48 28.7 48 64s-21.5 64-48 64c-13.3 0-25.3-7.2-34-18.8c-5.5-7.4-13.6-13.2-22.8-13.2c-12.8 0-23.2 10.4-23.2 23.2V464c0 26.5-21.5 48-48 48H279.2c-12.8 0-23.2-10.4-23.2-23.2c0-9.2 5.8-17.3 13.2-22.8c11.6-8.7 18.8-20.7 18.8-34c0-26.5-28.7-48-64-48s-64 21.5-64 48c0 13.3 7.2 25.3 18.8 34c7.4 5.5 13.2 13.6 13.2 22.8c0 12.8-10.4 23.2-23.2 23.2H48c-26.5 0-48-21.5-48-48V343.2C0 330.4 10.4 320 23.2 320c9.2 0 17.3 5.8 22.8 13.2C54.7 344.8 66.7 352 80 352c26.5 0 48-28.7 48-64s-21.5-64-48-64c-13.3 0-25.3 7.2-34 18.8C40.5 250.2 32.4 256 23.2 256C10.4 256 0 245.6 0 232.8V176c0-26.5 21.5-48 48-48H168.8c12.8 0 23.2-10.4 23.2-23.2z\"],\"money-check\":[576,512,[],\"f53c\",\"M64 64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H64zm48 160H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zM96 336c0-8.8 7.2-16 16-16H464c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16zM376 160h80c13.3 0 24 10.7 24 24v48c0 13.3-10.7 24-24 24H376c-13.3 0-24-10.7-24-24V184c0-13.3 10.7-24 24-24z\"],\"star-half-stroke\":[576,512,[\"star-half-alt\"],\"f5c0\",\"M288 376.4l.1-.1 26.4 14.1 85.2 45.5-16.5-97.6-4.8-28.7 20.7-20.5 70.1-69.3-96.1-14.2-29.3-4.3-12.9-26.6L288.1 86.9l-.1 .3V376.4zm175.1 98.3c2 12-3 24.2-12.9 31.3s-23 8-33.8 2.3L288.1 439.8 159.8 508.3C149 514 135.9 513.1 126 506s-14.9-19.3-12.9-31.3L137.8 329 33.6 225.9c-8.6-8.5-11.7-21.2-7.9-32.7s13.7-19.9 25.7-21.7L195 150.3 259.4 18c5.4-11 16.5-18 28.8-18s23.4 7 28.8 18l64.3 132.3 143.6 21.2c12 1.8 22 10.2 25.7 21.7s.7 24.2-7.9 32.7L438.5 329l24.6 145.7z\"],code:[640,512,[],\"f121\",\"M392.8 1.2c-17-4.9-34.7 5-39.6 22l-128 448c-4.9 17 5 34.7 22 39.6s34.7-5 39.6-22l128-448c4.9-17-5-34.7-22-39.6zm80.6 120.1c-12.5 12.5-12.5 32.8 0 45.3L562.7 256l-89.4 89.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l112-112c12.5-12.5 12.5-32.8 0-45.3l-112-112c-12.5-12.5-32.8-12.5-45.3 0zm-306.7 0c-12.5-12.5-32.8-12.5-45.3 0l-112 112c-12.5 12.5-12.5 32.8 0 45.3l112 112c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L77.3 256l89.4-89.4c12.5-12.5 12.5-32.8 0-45.3z\"],\"whiskey-glass\":[512,512,[129347,\"glass-whiskey\"],\"f7a0\",\"M32 32c-9.3 0-18.1 4-24.2 11.1S-1 59.4 .3 68.6l50 342.9c5.7 39.3 39.4 68.5 79.2 68.5h253c39.7 0 73.4-29.1 79.2-68.5l50-342.9c1.3-9.2-1.4-18.5-7.5-25.5S489.3 32 480 32H32zM87.7 224L69 96H443L424.3 224H87.7z\"],\"building-circle-exclamation\":[640,512,[],\"e4d3\",\"M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c15.1 0 28.5-6.9 37.3-17.8C340.4 462.2 320 417.5 320 368c0-54.7 24.9-103.5 64-135.8V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z\"],\"magnifying-glass-chart\":[512,512,[],\"e522\",\"M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zm-312 8v64c0 13.3 10.7 24 24 24s24-10.7 24-24l0-64c0-13.3-10.7-24-24-24s-24 10.7-24 24zm80-96V280c0 13.3 10.7 24 24 24s24-10.7 24-24V120c0-13.3-10.7-24-24-24s-24 10.7-24 24zm80 64v96c0 13.3 10.7 24 24 24s24-10.7 24-24V184c0-13.3-10.7-24-24-24s-24 10.7-24 24z\"],\"arrow-up-right-from-square\":[512,512,[\"external-link\"],\"f08e\",\"M320 0c-17.7 0-32 14.3-32 32s14.3 32 32 32h82.7L201.4 265.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L448 109.3V192c0 17.7 14.3 32 32 32s32-14.3 32-32V32c0-17.7-14.3-32-32-32H320zM80 32C35.8 32 0 67.8 0 112V432c0 44.2 35.8 80 80 80H400c44.2 0 80-35.8 80-80V320c0-17.7-14.3-32-32-32s-32 14.3-32 32V432c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16H192c17.7 0 32-14.3 32-32s-14.3-32-32-32H80z\"],\"cubes-stacked\":[448,512,[],\"e4e6\",\"M192 64v64c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V64c0-17.7-14.3-32-32-32H224c-17.7 0-32 14.3-32 32zM82.7 207c-15.3 8.8-20.5 28.4-11.7 43.7l32 55.4c8.8 15.3 28.4 20.5 43.7 11.7l55.4-32c15.3-8.8 20.5-28.4 11.7-43.7l-32-55.4c-8.8-15.3-28.4-20.5-43.7-11.7L82.7 207zM288 192c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32H288zm64 160c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V384c0-17.7-14.3-32-32-32H352zM160 384v64c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32V384c0-17.7-14.3-32-32-32H192c-17.7 0-32 14.3-32 32zM32 352c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H96c17.7 0 32-14.3 32-32V384c0-17.7-14.3-32-32-32H32z\"],\"won-sign\":[512,512,[8361,\"krw\",\"won\"],\"f159\",\"M62.4 53.9C56.8 37.1 38.6 28.1 21.9 33.6S-3.9 57.4 1.6 74.1L51.6 224H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H72.9l56.7 170.1c4.5 13.5 17.4 22.4 31.6 21.9s26.4-10.4 29.8-24.2L233 288h46L321 455.8c3.4 13.8 15.6 23.7 29.8 24.2s27.1-8.4 31.6-21.9L439.1 288H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H460.4l50-149.9c5.6-16.8-3.5-34.9-20.2-40.5s-34.9 3.5-40.5 20.2L392.9 224H329L287 56.2C283.5 42 270.7 32 256 32s-27.5 10-31 24.2L183 224h-64L62.4 53.9zm78 234.1H167l-11.4 45.6L140.4 288zM249 224l7-28.1 7 28.1H249zm96 64h26.6l-15.2 45.6L345 288z\"],\"virus-covid\":[512,512,[],\"e4a8\",\"M192 24c0-13.3 10.7-24 24-24h80c13.3 0 24 10.7 24 24s-10.7 24-24 24H280V81.6c30.7 4.2 58.8 16.3 82.3 34.1L386.1 92 374.8 80.6c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l56.6 56.6c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L420 125.9l-23.8 23.8c17.9 23.5 29.9 51.7 34.1 82.3H464V216c0-13.3 10.7-24 24-24s24 10.7 24 24v80c0 13.3-10.7 24-24 24s-24-10.7-24-24V280H430.4c-4.2 30.7-16.3 58.8-34.1 82.3L420 386.1l11.3-11.3c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-56.6 56.6c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9L386.1 420l-23.8-23.8c-23.5 17.9-51.7 29.9-82.3 34.1V464h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H216c-13.3 0-24-10.7-24-24s10.7-24 24-24h16V430.4c-30.7-4.2-58.8-16.3-82.3-34.1L125.9 420l11.3 11.3c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L46.7 408.7c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0L92 386.1l23.8-23.8C97.9 338.8 85.8 310.7 81.6 280H48v16c0 13.3-10.7 24-24 24s-24-10.7-24-24V216c0-13.3 10.7-24 24-24s24 10.7 24 24v16H81.6c4.2-30.7 16.3-58.8 34.1-82.3L92 125.9 80.6 137.2c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l56.6-56.6c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9L125.9 92l23.8 23.8c23.5-17.9 51.7-29.9 82.3-34.1V48H216c-13.3 0-24-10.7-24-24zm48 200a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zm64 104a24 24 0 1 0 0-48 24 24 0 1 0 0 48z\"],\"austral-sign\":[448,512,[],\"e0a9\",\"M253.5 51.7C248.6 39.8 236.9 32 224 32s-24.6 7.8-29.5 19.7L122.7 224H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H96L82.7 320H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H56L34.5 435.7c-6.8 16.3 .9 35 17.2 41.8s35-.9 41.8-17.2L125.3 384H322.7l31.8 76.3c6.8 16.3 25.5 24 41.8 17.2s24-25.5 17.2-41.8L392 384h24c17.7 0 32-14.3 32-32s-14.3-32-32-32H365.3L352 288h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H325.3L253.5 51.7zM256 224H192l32-76.8L256 224zm-90.7 64H282.7L296 320H152l13.3-32z\"],f:[320,512,[102],\"46\",\"M64 32C28.7 32 0 60.7 0 96V256 448c0 17.7 14.3 32 32 32s32-14.3 32-32V288H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H64V96H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H64z\"],leaf:[512,512,[],\"f06c\",\"M272 96c-78.6 0-145.1 51.5-167.7 122.5c33.6-17 71.5-26.5 111.7-26.5h88c8.8 0 16 7.2 16 16s-7.2 16-16 16H288 216s0 0 0 0c-16.6 0-32.7 1.9-48.3 5.4c-25.9 5.9-49.9 16.4-71.4 30.7c0 0 0 0 0 0C38.3 298.8 0 364.9 0 440v16c0 13.3 10.7 24 24 24s24-10.7 24-24V440c0-48.7 20.7-92.5 53.8-123.2C121.6 392.3 190.3 448 272 448l1 0c132.1-.7 239-130.9 239-291.4c0-42.6-7.5-83.1-21.1-119.6c-2.6-6.9-12.7-6.6-16.2-.1C455.9 72.1 418.7 96 376 96L272 96z\"],road:[576,512,[128739],\"f018\",\"M256 32H181.2c-27.1 0-51.3 17.1-60.3 42.6L3.1 407.2C1.1 413 0 419.2 0 425.4C0 455.5 24.5 480 54.6 480H256V416c0-17.7 14.3-32 32-32s32 14.3 32 32v64H521.4c30.2 0 54.6-24.5 54.6-54.6c0-6.2-1.1-12.4-3.1-18.2L455.1 74.6C446 49.1 421.9 32 394.8 32H320V96c0 17.7-14.3 32-32 32s-32-14.3-32-32V32zm64 192v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V224c0-17.7 14.3-32 32-32s32 14.3 32 32z\"],taxi:[512,512,[128662,\"cab\"],\"f1ba\",\"M192 0c-17.7 0-32 14.3-32 32V64c0 .1 0 .1 0 .2c-38.6 2.2-72.3 27.3-85.2 64.1L39.6 228.8C16.4 238.4 0 261.3 0 288V432v48c0 17.7 14.3 32 32 32H64c17.7 0 32-14.3 32-32V432H416v48c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V432 288c0-26.7-16.4-49.6-39.6-59.2L437.2 128.3c-12.9-36.8-46.6-62-85.2-64.1c0-.1 0-.1 0-.2V32c0-17.7-14.3-32-32-32H192zM165.4 128H346.6c13.6 0 25.7 8.6 30.2 21.4L402.9 224H109.1l26.1-74.6c4.5-12.8 16.6-21.4 30.2-21.4zM96 288a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm288 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],\"person-circle-plus\":[576,512,[],\"e541\",\"M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l44.9 74.7c-16.1 17.6-28.6 38.5-36.6 61.5c-1.9-1.8-3.5-3.9-4.9-6.3L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152zM432 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm16 80c0-8.8-7.2-16-16-16s-16 7.2-16 16v48H368c-8.8 0-16 7.2-16 16s7.2 16 16 16h48v48c0 8.8 7.2 16 16 16s16-7.2 16-16V384h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H448V304z\"],\"chart-pie\":[576,512,[\"pie-chart\"],\"f200\",\"M304 240V16.6c0-9 7-16.6 16-16.6C443.7 0 544 100.3 544 224c0 9-7.6 16-16.6 16H304zM32 272C32 150.7 122.1 50.3 239 34.3c9.2-1.3 17 6.1 17 15.4V288L412.5 444.5c6.7 6.7 6.2 17.7-1.5 23.1C371.8 495.6 323.8 512 272 512C139.5 512 32 404.6 32 272zm526.4 16c9.3 0 16.6 7.8 15.4 17c-7.7 55.9-34.6 105.6-73.9 142.3c-6 5.6-15.4 5.2-21.2-.7L320 288H558.4z\"],\"bolt-lightning\":[384,512,[],\"e0b7\",\"M0 256L28.5 28c2-16 15.6-28 31.8-28H228.9c15 0 27.1 12.1 27.1 27.1c0 3.2-.6 6.5-1.7 9.5L208 160H347.3c20.2 0 36.7 16.4 36.7 36.7c0 7.4-2.2 14.6-6.4 20.7l-192.2 281c-5.9 8.6-15.6 13.7-25.9 13.7h-2.9c-15.7 0-28.5-12.8-28.5-28.5c0-2.3 .3-4.6 .9-6.9L176 288H32c-17.7 0-32-14.3-32-32z\"],\"sack-xmark\":[512,512,[],\"e56a\",\"M192 96H320l47.4-71.1C374.5 14.2 366.9 0 354.1 0H157.9c-12.8 0-20.4 14.2-13.3 24.9L192 96zm128 32H192c-3.8 2.5-8.1 5.3-13 8.4l0 0 0 0C122.3 172.7 0 250.9 0 416c0 53 43 96 96 96H416c53 0 96-43 96-96c0-165.1-122.3-243.3-179-279.6c-4.8-3.1-9.2-5.9-13-8.4zM289.9 336l47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47z\"],\"file-excel\":[384,512,[],\"f1c3\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM155.7 250.2L192 302.1l36.3-51.9c7.6-10.9 22.6-13.5 33.4-5.9s13.5 22.6 5.9 33.4L221.3 344l46.4 66.2c7.6 10.9 5 25.8-5.9 33.4s-25.8 5-33.4-5.9L192 385.8l-36.3 51.9c-7.6 10.9-22.6 13.5-33.4 5.9s-13.5-22.6-5.9-33.4L162.7 344l-46.4-66.2c-7.6-10.9-5-25.8 5.9-33.4s25.8-5 33.4 5.9z\"],\"file-contract\":[384,512,[],\"f56c\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM80 64h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16s7.2-16 16-16zm54.2 253.8c-6.1 20.3-24.8 34.2-46 34.2H80c-8.8 0-16-7.2-16-16s7.2-16 16-16h8.2c7.1 0 13.3-4.6 15.3-11.4l14.9-49.5c3.4-11.3 13.8-19.1 25.6-19.1s22.2 7.7 25.6 19.1l11.6 38.6c7.4-6.2 16.8-9.7 26.8-9.7c15.9 0 30.4 9 37.5 23.2l4.4 8.8H304c8.8 0 16 7.2 16 16s-7.2 16-16 16H240c-6.1 0-11.6-3.4-14.3-8.8l-8.8-17.7c-1.7-3.4-5.1-5.5-8.8-5.5s-7.2 2.1-8.8 5.5l-8.8 17.7c-2.9 5.9-9.2 9.4-15.7 8.8s-12.1-5.1-13.9-11.3L144 349l-9.8 32.8z\"],\"fish-fins\":[576,512,[],\"e4f2\",\"M275.2 38.4c-10.6-8-25-8.5-36.3-1.5S222 57.3 224.6 70.3l9.7 48.6c-19.4 9-36.9 19.9-52.4 31.5c-15.3 11.5-29 23.9-40.7 36.3L48.1 132.4c-12.5-7.3-28.4-5.3-38.6 4.9S-3 163.3 4.2 175.9L50 256 4.2 336.1c-7.2 12.6-5 28.4 5.3 38.6s26.1 12.2 38.6 4.9l93.1-54.3c11.8 12.3 25.4 24.8 40.7 36.3c15.5 11.6 33 22.5 52.4 31.5l-9.7 48.6c-2.6 13 3.1 26.3 14.3 33.3s25.6 6.5 36.3-1.5l77.6-58.2c54.9-4 101.5-27 137.2-53.8c39.2-29.4 67.2-64.7 81.6-89.5c5.8-9.9 5.8-22.2 0-32.1c-14.4-24.8-42.5-60.1-81.6-89.5c-35.8-26.8-82.3-49.8-137.2-53.8L275.2 38.4zM384 256a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],\"building-flag\":[640,512,[],\"e4d5\",\"M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zM448 0c-17.7 0-32 14.3-32 32V512h64V192H624c8.8 0 16-7.2 16-16V48c0-8.8-7.2-16-16-16H480c0-17.7-14.3-32-32-32z\"],\"face-grin-beam\":[512,512,[128516,\"grin-beam\"],\"f582\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM388.1 312.8c12.3-3.8 24.3 6.9 19.3 18.7C382.4 390.6 324.2 432 256.3 432s-126.2-41.4-151.1-100.5c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19zm-170.5-84l0 0 0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0zm160 0l0 0-.2-.2c-.2-.2-.4-.5-.7-.9c-.6-.8-1.6-2-2.8-3.4c-2.5-2.8-6-6.6-10.2-10.3c-8.8-7.8-18.8-14-27.7-14s-18.9 6.2-27.7 14c-4.2 3.7-7.7 7.5-10.2 10.3c-1.2 1.4-2.2 2.6-2.8 3.4c-.3 .4-.6 .7-.7 .9l-.2 .2 0 0 0 0 0 0c-2.1 2.8-5.7 3.9-8.9 2.8s-5.5-4.1-5.5-7.6c0-17.9 6.7-35.6 16.6-48.8c9.8-13 23.9-23.2 39.4-23.2s29.6 10.2 39.4 23.2c9.9 13.2 16.6 30.9 16.6 48.8c0 3.4-2.2 6.5-5.5 7.6s-6.9 0-8.9-2.8l0 0 0 0 0 0z\"],\"object-ungroup\":[640,512,[],\"f248\",\"M32 119.4C12.9 108.4 0 87.7 0 64C0 28.7 28.7 0 64 0c23.7 0 44.4 12.9 55.4 32H328.6C339.6 12.9 360.3 0 384 0c35.3 0 64 28.7 64 64c0 23.7-12.9 44.4-32 55.4V232.6c19.1 11.1 32 31.7 32 55.4c0 35.3-28.7 64-64 64c-23.7 0-44.4-12.9-55.4-32H119.4c-11.1 19.1-31.7 32-55.4 32c-35.3 0-64-28.7-64-64c0-23.7 12.9-44.4 32-55.4V119.4zM119.4 96c-5.6 9.7-13.7 17.8-23.4 23.4V232.6c9.7 5.6 17.8 13.7 23.4 23.4H328.6c5.6-9.7 13.7-17.8 23.4-23.4V119.4c-9.7-5.6-17.8-13.7-23.4-23.4H119.4zm192 384c-11.1 19.1-31.7 32-55.4 32c-35.3 0-64-28.7-64-64c0-23.7 12.9-44.4 32-55.4V352h64v40.6c9.7 5.6 17.8 13.7 23.4 23.4H520.6c5.6-9.7 13.7-17.8 23.4-23.4V279.4c-9.7-5.6-17.8-13.7-23.4-23.4h-46c-5.4-15.4-14.6-28.9-26.5-39.6V192h72.6c11.1-19.1 31.7-32 55.4-32c35.3 0 64 28.7 64 64c0 23.7-12.9 44.4-32 55.4V392.6c19.1 11.1 32 31.7 32 55.4c0 35.3-28.7 64-64 64c-23.7 0-44.4-12.9-55.4-32H311.4z\"],poop:[512,512,[],\"f619\",\"M254.4 6.6c3.5-4.3 9-6.5 14.5-5.7C315.8 7.2 352 47.4 352 96c0 11.2-1.9 22-5.5 32H352c35.3 0 64 28.7 64 64c0 19.1-8.4 36.3-21.7 48H408c39.8 0 72 32.2 72 72c0 23.2-11 43.8-28 57c34.1 5.7 60 35.3 60 71c0 39.8-32.2 72-72 72H72c-39.8 0-72-32.2-72-72c0-35.7 25.9-65.3 60-71c-17-13.2-28-33.8-28-57c0-39.8 32.2-72 72-72h13.7C104.4 228.3 96 211.1 96 192c0-35.3 28.7-64 64-64h16.2c44.1-.1 79.8-35.9 79.8-80c0-9.2-1.5-17.9-4.3-26.1c-1.8-5.2-.8-11.1 2.8-15.4z\"],\"location-pin\":[384,512,[\"map-marker\"],\"f041\",\"M384 192c0 87.4-117 243-168.3 307.2c-12.3 15.3-35.1 15.3-47.4 0C117 435 0 279.4 0 192C0 86 86 0 192 0S384 86 384 192z\"],kaaba:[576,512,[128331],\"f66b\",\"M60 120l228 71.2L516 120 288 48.8 60 120zM278.5 1.5c6.2-1.9 12.9-1.9 19.1 0l256 80C566.9 85.6 576 98 576 112v16 0 21.2L292.8 237.7c-3.1 1-6.4 1-9.5 0L0 149.2V128 112C0 98 9.1 85.6 22.5 81.5l256-80zm23.9 266.8L576 182.8v46.5l-52.8 16.5c-8.4 2.6-13.1 11.6-10.5 20s11.6 13.1 20 10.5L576 262.8V400c0 14-9.1 26.4-22.5 30.5l-256 80c-6.2 1.9-12.9 1.9-19.1 0l-256-80C9.1 426.4 0 414 0 400V262.8l43.2 13.5c8.4 2.6 17.4-2.1 20-10.5s-2.1-17.4-10.5-20L0 229.2V182.8l273.7 85.5c9.3 2.9 19.3 2.9 28.6 0zm-185.5-2.6c-8.4-2.6-17.4 2.1-20 10.5s2.1 17.4 10.5 20l64 20c8.4 2.6 17.4-2.1 20-10.5s-2.1-17.4-10.5-20l-64-20zm352 30.5c8.4-2.6 13.1-11.6 10.5-20s-11.6-13.1-20-10.5l-64 20c-8.4 2.6-13.1 11.6-10.5 20s11.6 13.1 20 10.5l64-20zm-224 9.5c-8.4-2.6-17.4 2.1-20 10.5s2.1 17.4 10.5 20l38.5 12c9.3 2.9 19.3 2.9 28.6 0l38.5-12c8.4-2.6 13.1-11.6 10.5-20s-11.6-13.1-20-10.5l-38.5 12c-3.1 1-6.4 1-9.5 0l-38.5-12z\"],\"toilet-paper\":[640,512,[129531],\"f71e\",\"M444.2 0C397.2 49.6 384 126.5 384 192c0 158.8-27.3 247-42.7 283.9c-10 24-33.2 36.1-55.4 36.1H48c-11.5 0-22.2-6.2-27.8-16.2s-5.6-22.3 .4-32.2c9.8-17.7 15.4-38.2 20.5-57.7C52.3 362.8 64 293.5 64 192C64 86 107 0 160 0H444.2zM512 384c-53 0-96-86-96-192S459 0 512 0s96 86 96 192s-43 192-96 192zm0-128c17.7 0 32-28.7 32-64s-14.3-64-32-64s-32 28.7-32 64s14.3 64 32 64zM144 208a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm64 0a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm48 16a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm80-16a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z\"],\"helmet-safety\":[576,512,[\"hard-hat\",\"hat-hard\"],\"f807\",\"M256 32c-17.7 0-32 14.3-32 32v2.3 99.6c0 5.6-4.5 10.1-10.1 10.1c-3.6 0-7-1.9-8.8-5.1L157.1 87C83 123.5 32 199.8 32 288v64H544l0-66.4c-.9-87.2-51.7-162.4-125.1-198.6l-48 83.9c-1.8 3.2-5.2 5.1-8.8 5.1c-5.6 0-10.1-4.5-10.1-10.1V66.3 64c0-17.7-14.3-32-32-32H256zM16.6 384C7.4 384 0 391.4 0 400.6c0 4.7 2 9.2 5.8 11.9C27.5 428.4 111.8 480 288 480s260.5-51.6 282.2-67.5c3.8-2.8 5.8-7.2 5.8-11.9c0-9.2-7.4-16.6-16.6-16.6H16.6z\"],eject:[448,512,[9167],\"f052\",\"M224 32c13.5 0 26.3 5.6 35.4 15.6l176 192c12.9 14 16.2 34.3 8.6 51.8S419 320 400 320H48c-19 0-36.3-11.2-43.9-28.7s-4.3-37.7 8.6-51.8l176-192C197.7 37.6 210.5 32 224 32zM0 432c0-26.5 21.5-48 48-48H400c26.5 0 48 21.5 48 48s-21.5 48-48 48H48c-26.5 0-48-21.5-48-48z\"],\"circle-right\":[512,512,[61838,\"arrow-alt-circle-right\"],\"f35a\",\"M0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zM294.6 135.1l99.9 107.1c3.5 3.8 5.5 8.7 5.5 13.8s-2 10.1-5.5 13.8L294.6 376.9c-4.2 4.5-10.1 7.1-16.3 7.1C266 384 256 374 256 361.7l0-57.7-96 0c-17.7 0-32-14.3-32-32l0-32c0-17.7 14.3-32 32-32l96 0 0-57.7c0-12.3 10-22.3 22.3-22.3c6.2 0 12.1 2.6 16.3 7.1z\"],\"plane-circle-check\":[640,512,[],\"e555\",\"M256 0c-35 0-64 59.5-64 93.7v84.6L8.1 283.4c-5 2.8-8.1 8.2-8.1 13.9v65.5c0 10.6 10.2 18.3 20.4 15.4l171.6-49 0 70.9-57.6 43.2c-4 3-6.4 7.8-6.4 12.8v42c0 7.8 6.3 14 14 14c1.3 0 2.6-.2 3.9-.5L256 480l110.1 31.5c1.3 .4 2.6 .5 3.9 .5c6 0 11.1-3.7 13.1-9C344.5 470.7 320 422.2 320 368c0-60.6 30.6-114 77.1-145.6L320 178.3V93.7C320 59.5 292 0 256 0zM640 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z\"],\"face-rolling-eyes\":[512,512,[128580,\"meh-rolling-eyes\"],\"f5a5\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM192 368H320c8.8 0 16 7.2 16 16s-7.2 16-16 16H192c-8.8 0-16-7.2-16-16s7.2-16 16-16zm32-144c0 35.3-28.7 64-64 64s-64-28.7-64-64c0-26 15.5-48.4 37.8-58.4c-3.7 5.2-5.8 11.6-5.8 18.4c0 17.7 14.3 32 32 32s32-14.3 32-32c0-6.9-2.2-13.2-5.8-18.4C208.5 175.6 224 198 224 224zm128 64c-35.3 0-64-28.7-64-64c0-26 15.5-48.4 37.8-58.4c-3.7 5.2-5.8 11.6-5.8 18.4c0 17.7 14.3 32 32 32s32-14.3 32-32c0-6.9-2.2-13.2-5.8-18.4C400.5 175.6 416 198 416 224c0 35.3-28.7 64-64 64z\"],\"object-group\":[576,512,[],\"f247\",\"M32 119.4C12.9 108.4 0 87.7 0 64C0 28.7 28.7 0 64 0c23.7 0 44.4 12.9 55.4 32H456.6C467.6 12.9 488.3 0 512 0c35.3 0 64 28.7 64 64c0 23.7-12.9 44.4-32 55.4V392.6c19.1 11.1 32 31.7 32 55.4c0 35.3-28.7 64-64 64c-23.7 0-44.4-12.9-55.4-32H119.4c-11.1 19.1-31.7 32-55.4 32c-35.3 0-64-28.7-64-64c0-23.7 12.9-44.4 32-55.4V119.4zM456.6 96H119.4c-5.6 9.7-13.7 17.8-23.4 23.4V392.6c9.7 5.6 17.8 13.7 23.4 23.4H456.6c5.6-9.7 13.7-17.8 23.4-23.4V119.4c-9.7-5.6-17.8-13.7-23.4-23.4zM128 160c0-17.7 14.3-32 32-32H288c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32V160zM256 320h32c35.3 0 64-28.7 64-64V224h64c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H288c-17.7 0-32-14.3-32-32V320z\"],\"chart-line\":[512,512,[\"line-chart\"],\"f201\",\"M64 64c0-17.7-14.3-32-32-32S0 46.3 0 64V400c0 44.2 35.8 80 80 80H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H80c-8.8 0-16-7.2-16-16V64zm406.6 86.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L320 210.7l-57.4-57.4c-12.5-12.5-32.8-12.5-45.3 0l-112 112c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L240 221.3l57.4 57.4c12.5 12.5 32.8 12.5 45.3 0l128-128z\"],\"mask-ventilator\":[640,512,[],\"e524\",\"M159.1 176C139.4 219.2 128 264.7 128 300.8c0 15.9 2.2 31.4 6.3 46l-31.8-7.9C70.5 330.9 48 302.1 48 269V184c0-4.4 3.6-8 8-8H159.1zm26-48H56c-30.9 0-56 25.1-56 56v85c0 55.1 37.5 103.1 90.9 116.4l71.3 17.8c22.7 30.5 55.4 54.1 93.8 66.6V393.3c-19.7-16.4-32-40.3-32-66.9c0-49.5 43-134.4 96-134.4c52.5 0 96 84.9 96 134.4c0 26.7-12.4 50.4-32 66.8v76.6c38-12.6 70.6-36 93.5-66.4l71.6-17.9C602.5 372.1 640 324.1 640 269V184c0-30.9-25.1-56-56-56H454.5C419.7 73.8 372.1 32 320 32c-52.6 0-100.2 41.8-134.9 96zm295.6 48H584c4.4 0 8 3.6 8 8v85c0 33-22.5 61.8-54.5 69.9l-31.8 8c4.2-14.7 6.4-30.1 6.4-46.1c0-36.1-11.6-81.6-31.3-124.8zM288 320V512h64V320c0-17.7-14.3-32-32-32s-32 14.3-32 32z\"],\"arrow-right\":[448,512,[8594],\"f061\",\"M438.6 278.6c12.5-12.5 12.5-32.8 0-45.3l-160-160c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L338.8 224 32 224c-17.7 0-32 14.3-32 32s14.3 32 32 32l306.7 0L233.4 393.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l160-160z\"],\"signs-post\":[512,512,[\"map-signs\"],\"f277\",\"M224 32H64C46.3 32 32 46.3 32 64v64c0 17.7 14.3 32 32 32H441.4c4.2 0 8.3-1.7 11.3-4.7l48-48c6.2-6.2 6.2-16.4 0-22.6l-48-48c-3-3-7.1-4.7-11.3-4.7H288c0-17.7-14.3-32-32-32s-32 14.3-32 32zM480 256c0-17.7-14.3-32-32-32H288V192H224v32H70.6c-4.2 0-8.3 1.7-11.3 4.7l-48 48c-6.2 6.2-6.2 16.4 0 22.6l48 48c3 3 7.1 4.7 11.3 4.7H448c17.7 0 32-14.3 32-32V256zM288 480V384H224v96c0 17.7 14.3 32 32 32s32-14.3 32-32z\"],\"cash-register\":[512,512,[],\"f788\",\"M64 0C46.3 0 32 14.3 32 32V96c0 17.7 14.3 32 32 32h80v32H87c-31.6 0-58.5 23.1-63.3 54.4L1.1 364.1C.4 368.8 0 373.6 0 378.4V448c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V378.4c0-4.8-.4-9.6-1.1-14.4L488.2 214.4C483.5 183.1 456.6 160 425 160H208V128h80c17.7 0 32-14.3 32-32V32c0-17.7-14.3-32-32-32H64zM96 48H256c8.8 0 16 7.2 16 16s-7.2 16-16 16H96c-8.8 0-16-7.2-16-16s7.2-16 16-16zM64 432c0-8.8 7.2-16 16-16H432c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16zm48-168a24 24 0 1 1 0-48 24 24 0 1 1 0 48zm120-24a24 24 0 1 1 -48 0 24 24 0 1 1 48 0zM160 344a24 24 0 1 1 0-48 24 24 0 1 1 0 48zM328 240a24 24 0 1 1 -48 0 24 24 0 1 1 48 0zM256 344a24 24 0 1 1 0-48 24 24 0 1 1 0 48zM424 240a24 24 0 1 1 -48 0 24 24 0 1 1 48 0zM352 344a24 24 0 1 1 0-48 24 24 0 1 1 0 48z\"],\"person-circle-question\":[576,512,[],\"e542\",\"M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l44.9 74.7c-16.1 17.6-28.6 38.5-36.6 61.5c-1.9-1.8-3.5-3.9-4.9-6.3L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152zM432 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm0 240a24 24 0 1 0 0-48 24 24 0 1 0 0 48zM368 321.6V328c0 8.8 7.2 16 16 16s16-7.2 16-16v-6.4c0-5.3 4.3-9.6 9.6-9.6h40.5c7.7 0 13.9 6.2 13.9 13.9c0 5.2-2.9 9.9-7.4 12.3l-32 16.8c-5.3 2.8-8.6 8.2-8.6 14.2V384c0 8.8 7.2 16 16 16s16-7.2 16-16v-5.1l23.5-12.3c15.1-7.9 24.5-23.6 24.5-40.6c0-25.4-20.6-45.9-45.9-45.9H409.6c-23 0-41.6 18.6-41.6 41.6z\"],h:[384,512,[104],\"48\",\"M320 256l0 192c0 17.7 14.3 32 32 32s32-14.3 32-32l0-224V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V192L64 192 64 64c0-17.7-14.3-32-32-32S0 46.3 0 64V448c0 17.7 14.3 32 32 32s32-14.3 32-32l0-192 256 0z\"],tarp:[576,512,[],\"e57b\",\"M576 128c0-35.3-28.7-64-64-64H64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64l352 0 0-128c0-17.7 14.3-32 32-32H576V128zM448 448L576 320H448l0 128zM96 128a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"screwdriver-wrench\":[512,512,[\"tools\"],\"f7d9\",\"M78.6 5C69.1-2.4 55.6-1.5 47 7L7 47c-8.5 8.5-9.4 22-2.1 31.6l80 104c4.5 5.9 11.6 9.4 19 9.4h54.1l109 109c-14.7 29-10 65.4 14.3 89.6l112 112c12.5 12.5 32.8 12.5 45.3 0l64-64c12.5-12.5 12.5-32.8 0-45.3l-112-112c-24.2-24.2-60.6-29-89.6-14.3l-109-109V104c0-7.5-3.5-14.5-9.4-19L78.6 5zM19.9 396.1C7.2 408.8 0 426.1 0 444.1C0 481.6 30.4 512 67.9 512c18 0 35.3-7.2 48-19.9L233.7 374.3c-7.8-20.9-9-43.6-3.6-65.1l-61.7-61.7L19.9 396.1zM512 144c0-10.5-1.1-20.7-3.2-30.5c-2.4-11.2-16.1-14.1-24.2-6l-63.9 63.9c-3 3-7.1 4.7-11.3 4.7H352c-8.8 0-16-7.2-16-16V102.6c0-4.2 1.7-8.3 4.7-11.3l63.9-63.9c8.1-8.1 5.2-21.8-6-24.2C388.7 1.1 378.5 0 368 0C288.5 0 224 64.5 224 144l0 .8 85.3 85.3c36-9.1 75.8 .5 104 28.7L429 274.5c49-23 83-72.8 83-130.5zM56 432a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z\"],\"arrows-to-eye\":[640,512,[],\"e4bf\",\"M15 15C24.4 5.7 39.6 5.7 49 15l63 63V40c0-13.3 10.7-24 24-24s24 10.7 24 24v96c0 13.3-10.7 24-24 24H40c-13.3 0-24-10.7-24-24s10.7-24 24-24H78.1L15 49C5.7 39.6 5.7 24.4 15 15zM133.5 243.9C158.6 193.6 222.7 112 320 112s161.4 81.6 186.5 131.9c3.8 7.6 3.8 16.5 0 24.2C481.4 318.4 417.3 400 320 400s-161.4-81.6-186.5-131.9c-3.8-7.6-3.8-16.5 0-24.2zM320 320a64 64 0 1 0 0-128 64 64 0 1 0 0 128zM591 15c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-63 63H600c13.3 0 24 10.7 24 24s-10.7 24-24 24H504c-13.3 0-24-10.7-24-24V40c0-13.3 10.7-24 24-24s24 10.7 24 24V78.1l63-63zM15 497c-9.4-9.4-9.4-24.6 0-33.9l63-63H40c-13.3 0-24-10.7-24-24s10.7-24 24-24h96c13.3 0 24 10.7 24 24v96c0 13.3-10.7 24-24 24s-24-10.7-24-24V433.9L49 497c-9.4 9.4-24.6 9.4-33.9 0zm576 0l-63-63V472c0 13.3-10.7 24-24 24s-24-10.7-24-24V376c0-13.3 10.7-24 24-24h96c13.3 0 24 10.7 24 24s-10.7 24-24 24H561.9l63 63c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0z\"],\"plug-circle-bolt\":[576,512,[],\"e55b\",\"M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8c12.3-2.5 24.1-6.4 35.1-11.5c-2.1-10.8-3.1-21.9-3.1-33.3c0-80.3 53.8-148 127.3-169.2c.5-2.2 .7-4.5 .7-6.8c0-17.7-14.3-32-32-32H32zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm47.9-225c4.3 3.7 5.4 9.9 2.6 14.9L452.4 356H488c5.2 0 9.8 3.3 11.4 8.2s-.1 10.3-4.2 13.4l-96 72c-4.5 3.4-10.8 3.2-15.1-.6s-5.4-9.9-2.6-14.9L411.6 380H376c-5.2 0-9.8-3.3-11.4-8.2s.1-10.3 4.2-13.4l96-72c4.5-3.4 10.8-3.2 15.1 .6z\"],heart:[512,512,[128153,128154,128155,128156,128420,129293,129294,129505,9829,10084,61578],\"f004\",\"M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9L464.4 300.4c30.4-28.3 47.6-68 47.6-109.5v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5z\"],\"mars-and-venus\":[512,512,[9893],\"f224\",\"M337.8 14.8C341.5 5.8 350.3 0 360 0H472c13.3 0 24 10.7 24 24V136c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-39-39-24.7 24.7C407 163.3 416 192.6 416 224c0 80.2-59 146.6-136 158.2V408h24c13.3 0 24 10.7 24 24s-10.7 24-24 24H280v32c0 13.3-10.7 24-24 24s-24-10.7-24-24V456H208c-13.3 0-24-10.7-24-24s10.7-24 24-24h24V382.2C155 370.6 96 304.2 96 224c0-88.4 71.6-160 160-160c39.6 0 75.9 14.4 103.8 38.2L382.1 80 343 41c-6.9-6.9-8.9-17.2-5.2-26.2zM448 48l0 0h0v0zM352 224a96 96 0 1 0 -192 0 96 96 0 1 0 192 0z\"],\"house-user\":[576,512,[\"home-user\"],\"e1b0\",\"M575.8 255.5c0 18-15 32.1-32 32.1h-32l.7 160.2c.2 35.5-28.5 64.3-64 64.3H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L564.8 231.5c8 7 12 15 11 24zM352 224a64 64 0 1 0 -128 0 64 64 0 1 0 128 0zm-96 96c-44.2 0-80 35.8-80 80c0 8.8 7.2 16 16 16H384c8.8 0 16-7.2 16-16c0-44.2-35.8-80-80-80H256z\"],\"dumpster-fire\":[640,512,[],\"f794\",\"M49.7 32l90.8 0L114.9 160l-94 0C9.3 160 0 150.7 0 139.1c0-2.1 .3-4.1 .9-6.1L26.8 48.9C29.9 38.9 39.2 32 49.7 32zM272 160l-124.5 0L173.1 32 272 32l0 128zm32 0l0-128 98.9 0 14.4 72.1c-1.7 1.1-3.3 2.4-4.8 3.8c-18.4 16.4-35.4 34-50.5 52.1l-58 0zm209.9-23.7c-1.7 1.6-3.4 3.2-5 4.8C498 129.6 486.7 118.6 475 108c-7.6-6.9-17-10.8-26.6-11.8L435.5 32l90.8 0c10.5 0 19.8 6.9 22.9 16.9L575.1 133c.2 .7 .4 1.4 .5 2.1c-17.8-15-44.3-14.6-61.7 1.2zM325.2 210.7C304.3 244.5 288 282.9 288 318.1c0 49.3 18.6 95.2 49.6 129.9L128 448c0 17.7-14.3 32-32 32s-32-14.3-32-32L44 288l-12 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l4 0-4-32 305.5 0c-4.4 6.2-8.5 12.5-12.3 18.7zm180.6-34.5L518 162.5c5.4-6.1 13.3-8.8 20.9-8.9c7.2 0 14.3 2.6 19.9 7.8c19.7 18.3 39.8 43.2 55 70.6C629 259.2 640 290.2 640 320.2C640 408.8 568.7 480 480 480c-89.6 0-160-71.3-160-159.8c0-37.3 16-73.4 36.8-104.5c20.9-31.3 47.5-59 70.9-80.2c5.7-5.2 13.1-7.7 20.3-7.5c14.1 .3 23.8 11.4 32.7 21.6c0 0 0 0 0 0c2 2.3 4 4.6 6 6.7l19 19.9zM544 368.2c0-36.5-37-73-54.8-88.4c-5.4-4.7-13.1-4.7-18.5 0C453 295.1 416 331.6 416 368.2c0 35.3 28.7 64 64 64s64-28.7 64-64z\"],\"house-crack\":[576,512,[],\"e3b1\",\"M543.8 287.6c17 0 32-14 32-32.1c1-9-3-17-11-24L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32V448c0 35.3 28.7 64 64 64H230.4l-31.3-52.2c-4.1-6.8-2.6-15.5 3.5-20.5L288 368l-60.2-82.8c-10.9-15 8.2-33.5 22.8-22l117.9 92.6c8 6.3 8.2 18.4 .4 24.9L288 448l38.4 64H448.5c35.5 0 64.2-28.8 64-64.3l-.7-160.2h32z\"],\"martini-glass-citrus\":[576,512,[\"cocktail\"],\"f561\",\"M432 240c53 0 96-43 96-96s-43-96-96-96c-35.5 0-66.6 19.3-83.2 48H296.2C316 40.1 369.3 0 432 0c79.5 0 144 64.5 144 144s-64.5 144-144 144c-27.7 0-53.5-7.8-75.5-21.3l35.4-35.4c12.2 5.6 25.8 8.7 40.1 8.7zM1.8 142.8C5.5 133.8 14.3 128 24 128H392c9.7 0 18.5 5.8 22.2 14.8s1.7 19.3-5.2 26.2l-177 177V464h64c13.3 0 24 10.7 24 24s-10.7 24-24 24H208 120c-13.3 0-24-10.7-24-24s10.7-24 24-24h64V345.9L7 169c-6.9-6.9-8.9-17.2-5.2-26.2z\"],\"face-surprise\":[512,512,[128558,\"surprise\"],\"f5c2\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM176.4 176a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM256 288a64 64 0 1 1 0 128 64 64 0 1 1 0-128z\"],\"bottle-water\":[320,512,[],\"e4c5\",\"M120 0l80 0c13.3 0 24 10.7 24 24l0 40L96 64l0-40c0-13.3 10.7-24 24-24zM32 167.5c0-19.5 10-37.6 26.6-47.9l15.8-9.9C88.7 100.7 105.2 96 122.1 96l75.8 0c16.9 0 33.4 4.7 47.7 13.7l15.8 9.9C278 129.9 288 148 288 167.5c0 17-7.5 32.3-19.4 42.6C280.6 221.7 288 238 288 256c0 19.1-8.4 36.3-21.7 48c13.3 11.7 21.7 28.9 21.7 48s-8.4 36.3-21.7 48c13.3 11.7 21.7 28.9 21.7 48c0 35.3-28.7 64-64 64L96 512c-35.3 0-64-28.7-64-64c0-19.1 8.4-36.3 21.7-48C40.4 388.3 32 371.1 32 352s8.4-36.3 21.7-48C40.4 292.3 32 275.1 32 256c0-18 7.4-34.3 19.4-45.9C39.5 199.7 32 184.5 32 167.5zM96 240c0 8.8 7.2 16 16 16l96 0c8.8 0 16-7.2 16-16s-7.2-16-16-16l-96 0c-8.8 0-16 7.2-16 16zm16 112c-8.8 0-16 7.2-16 16s7.2 16 16 16l96 0c8.8 0 16-7.2 16-16s-7.2-16-16-16l-96 0z\"],\"circle-pause\":[512,512,[62092,\"pause-circle\"],\"f28b\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM224 192V320c0 17.7-14.3 32-32 32s-32-14.3-32-32V192c0-17.7 14.3-32 32-32s32 14.3 32 32zm128 0V320c0 17.7-14.3 32-32 32s-32-14.3-32-32V192c0-17.7 14.3-32 32-32s32 14.3 32 32z\"],\"toilet-paper-slash\":[640,512,[],\"e072\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-109.7-86C569.9 374 608 291.9 608 192C608 86 565 0 512 0s-96 86-96 192c0 49.1 9.2 93.9 24.4 127.9l-59-46.2c1.6-24.8 2.6-52 2.6-81.6c0-65.5 13.2-142.4 60.2-192H160c-24.8 0-47.4 18.8-64.4 49.6L38.8 5.1zM367.3 385.4L66.5 148.4C64.9 162.4 64 177 64 192c0 101.5-11.7 170.8-23 213.9c-5.1 19.4-10.7 39.9-20.5 57.7c-5.9 9.9-6.1 22.1-.4 32.2S36.5 512 48 512H285.9c22.3 0 45.4-12.1 55.4-36.1c7.4-17.7 17.5-47.2 26-90.6zM544 192c0 35.3-14.3 64-32 64s-32-28.7-32-64s14.3-64 32-64s32 28.7 32 64z\"],\"apple-whole\":[448,512,[127822,127823,\"apple-alt\"],\"f5d1\",\"M224 112c-8.8 0-16-7.2-16-16V80c0-44.2 35.8-80 80-80h16c8.8 0 16 7.2 16 16V32c0 44.2-35.8 80-80 80H224zM0 288c0-76.3 35.7-160 112-160c27.3 0 59.7 10.3 82.7 19.3c18.8 7.3 39.9 7.3 58.7 0c22.9-8.9 55.4-19.3 82.7-19.3c76.3 0 112 83.7 112 160c0 128-80 224-160 224c-16.5 0-38.1-6.6-51.5-11.3c-8.1-2.8-16.9-2.8-25 0c-13.4 4.7-35 11.3-51.5 11.3C80 512 0 416 0 288z\"],\"kitchen-set\":[576,512,[],\"e51a\",\"M240 144A96 96 0 1 0 48 144a96 96 0 1 0 192 0zm44.4 32C269.9 240.1 212.5 288 144 288C64.5 288 0 223.5 0 144S64.5 0 144 0c68.5 0 125.9 47.9 140.4 112h71.8c8.8-9.8 21.6-16 35.8-16H496c26.5 0 48 21.5 48 48s-21.5 48-48 48H392c-14.2 0-27-6.2-35.8-16H284.4zM144 80a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM400 240c13.3 0 24 10.7 24 24v8h96c13.3 0 24 10.7 24 24s-10.7 24-24 24H280c-13.3 0-24-10.7-24-24s10.7-24 24-24h96v-8c0-13.3 10.7-24 24-24zM288 464V352H512V464c0 26.5-21.5 48-48 48H336c-26.5 0-48-21.5-48-48zM48 320h80 16 32c26.5 0 48 21.5 48 48s-21.5 48-48 48H160c0 17.7-14.3 32-32 32H64c-17.7 0-32-14.3-32-32V336c0-8.8 7.2-16 16-16zm128 64c8.8 0 16-7.2 16-16s-7.2-16-16-16H160v32h16zM24 464H200c13.3 0 24 10.7 24 24s-10.7 24-24 24H24c-13.3 0-24-10.7-24-24s10.7-24 24-24z\"],r:[320,512,[114],\"52\",\"M64 32C28.7 32 0 60.7 0 96V288 448c0 17.7 14.3 32 32 32s32-14.3 32-32V320h95.3L261.8 466.4c10.1 14.5 30.1 18 44.6 7.9s18-30.1 7.9-44.6L230.1 309.5C282.8 288.1 320 236.4 320 176c0-79.5-64.5-144-144-144H64zM176 256H64V96H176c44.2 0 80 35.8 80 80s-35.8 80-80 80z\"],\"temperature-quarter\":[320,512,[\"temperature-1\",\"thermometer-1\",\"thermometer-quarter\"],\"f2ca\",\"M160 64c-26.5 0-48 21.5-48 48V276.5c0 17.3-7.1 31.9-15.3 42.5C86.2 332.6 80 349.5 80 368c0 44.2 35.8 80 80 80s80-35.8 80-80c0-18.5-6.2-35.4-16.7-48.9c-8.2-10.6-15.3-25.2-15.3-42.5V112c0-26.5-21.5-48-48-48zM48 112C48 50.2 98.1 0 160 0s112 50.1 112 112V276.5c0 .1 .1 .3 .2 .6c.2 .6 .8 1.6 1.7 2.8c18.9 24.4 30.1 55 30.1 88.1c0 79.5-64.5 144-144 144S16 447.5 16 368c0-33.2 11.2-63.8 30.1-88.1c.9-1.2 1.5-2.2 1.7-2.8c.1-.3 .2-.5 .2-.6V112zM208 368c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-20.9 13.4-38.7 32-45.3V272c0-8.8 7.2-16 16-16s16 7.2 16 16v50.7c18.6 6.6 32 24.4 32 45.3z\"],cube:[512,512,[],\"f1b2\",\"M234.5 5.7c13.9-5 29.1-5 43.1 0l192 68.6C495 83.4 512 107.5 512 134.6V377.4c0 27-17 51.2-42.5 60.3l-192 68.6c-13.9 5-29.1 5-43.1 0l-192-68.6C17 428.6 0 404.5 0 377.4V134.6c0-27 17-51.2 42.5-60.3l192-68.6zM256 66L82.3 128 256 190l173.7-62L256 66zm32 368.6l160-57.1v-188L288 246.6v188z\"],\"bitcoin-sign\":[320,512,[],\"e0b4\",\"M48 32C48 14.3 62.3 0 80 0s32 14.3 32 32V64h32V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64c0 1.5-.1 3.1-.3 4.5C254.1 82.2 288 125.1 288 176c0 24.2-7.7 46.6-20.7 64.9c31.7 19.8 52.7 55 52.7 95.1c0 61.9-50.1 112-112 112v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H112v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H41.7C18.7 448 0 429.3 0 406.3V288 265.7 224 101.6C0 80.8 16.8 64 37.6 64H48V32zM64 224H176c26.5 0 48-21.5 48-48s-21.5-48-48-48H64v96zm112 64H64v96H208c26.5 0 48-21.5 48-48s-21.5-48-48-48H176z\"],\"shield-dog\":[512,512,[],\"e573\",\"M269.4 2.9C265.2 1 260.7 0 256 0s-9.2 1-13.4 2.9L54.3 82.8c-22 9.3-38.4 31-38.3 57.2c.5 99.2 41.3 280.7 213.6 363.2c16.7 8 36.1 8 52.8 0C454.7 420.7 495.5 239.2 496 140c.1-26.2-16.3-47.9-38.3-57.2L269.4 2.9zM160.9 286.2c4.8 1.2 9.9 1.8 15.1 1.8c35.3 0 64-28.7 64-64V160h44.2c12.1 0 23.2 6.8 28.6 17.7L320 192h64c8.8 0 16 7.2 16 16v32c0 44.2-35.8 80-80 80H272v50.7c0 7.3-5.9 13.3-13.3 13.3c-1.8 0-3.6-.4-5.2-1.1l-98.7-42.3c-6.6-2.8-10.8-9.3-10.8-16.4c0-2.8 .6-5.5 1.9-8l15-30zM160 160h40 8v32 32c0 17.7-14.3 32-32 32s-32-14.3-32-32V176c0-8.8 7.2-16 16-16zm128 48a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z\"],\"solar-panel\":[640,512,[],\"f5ba\",\"M122.2 0C91.7 0 65.5 21.5 59.5 51.4L8.3 307.4C.4 347 30.6 384 71 384H288v64H224c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H352V384H569c40.4 0 70.7-36.9 62.8-76.6l-51.2-256C574.5 21.5 548.3 0 517.8 0H122.2zM260.9 64H379.1l10.4 104h-139L260.9 64zM202.3 168H101.4L122.2 64h90.4L202.3 168zM91.8 216H197.5L187.1 320H71L91.8 216zm153.9 0H394.3l10.4 104-169.4 0 10.4-104zm196.8 0H548.2L569 320h-116L442.5 216zm96-48H437.7L427.3 64h90.4l31.4-6.3L517.8 64l20.8 104z\"],\"lock-open\":[576,512,[],\"f3c1\",\"M352 144c0-44.2 35.8-80 80-80s80 35.8 80 80v48c0 17.7 14.3 32 32 32s32-14.3 32-32V144C576 64.5 511.5 0 432 0S288 64.5 288 144v48H64c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V256c0-35.3-28.7-64-64-64H352V144z\"],elevator:[512,512,[],\"e16d\",\"M132.7 4.7l-64 64c-4.6 4.6-5.9 11.5-3.5 17.4s8.3 9.9 14.8 9.9H208c6.5 0 12.3-3.9 14.8-9.9s1.1-12.9-3.5-17.4l-64-64c-6.2-6.2-16.4-6.2-22.6 0zM64 128c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V192c0-35.3-28.7-64-64-64H64zm96 96a48 48 0 1 1 0 96 48 48 0 1 1 0-96zM80 400c0-26.5 21.5-48 48-48h64c26.5 0 48 21.5 48 48v16c0 17.7-14.3 32-32 32H112c-17.7 0-32-14.3-32-32V400zm192 0c0-26.5 21.5-48 48-48h64c26.5 0 48 21.5 48 48v16c0 17.7-14.3 32-32 32H304c-17.7 0-32-14.3-32-32V400zm32-128a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM356.7 91.3c6.2 6.2 16.4 6.2 22.6 0l64-64c4.6-4.6 5.9-11.5 3.5-17.4S438.5 0 432 0H304c-6.5 0-12.3 3.9-14.8 9.9s-1.1 12.9 3.5 17.4l64 64z\"],\"money-bill-transfer\":[640,512,[],\"e528\",\"M535 41c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l64 64c4.5 4.5 7 10.6 7 17s-2.5 12.5-7 17l-64 64c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l23-23L384 112c-13.3 0-24-10.7-24-24s10.7-24 24-24l174.1 0L535 41zM105 377l-23 23L256 400c13.3 0 24 10.7 24 24s-10.7 24-24 24L81.9 448l23 23c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L7 441c-4.5-4.5-7-10.6-7-17s2.5-12.5 7-17l64-64c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9zM96 64H337.9c-3.7 7.2-5.9 15.3-5.9 24c0 28.7 23.3 52 52 52l117.4 0c-4 17 .6 35.5 13.8 48.8c20.3 20.3 53.2 20.3 73.5 0L608 169.5V384c0 35.3-28.7 64-64 64H302.1c3.7-7.2 5.9-15.3 5.9-24c0-28.7-23.3-52-52-52l-117.4 0c4-17-.6-35.5-13.8-48.8c-20.3-20.3-53.2-20.3-73.5 0L32 342.5V128c0-35.3 28.7-64 64-64zm64 64H96v64c35.3 0 64-28.7 64-64zM544 320c-35.3 0-64 28.7-64 64h64V320zM320 352a96 96 0 1 0 0-192 96 96 0 1 0 0 192z\"],\"money-bill-trend-up\":[512,512,[],\"e529\",\"M470.7 9.4c3 3.1 5.3 6.6 6.9 10.3s2.4 7.8 2.4 12.2l0 .1v0 96c0 17.7-14.3 32-32 32s-32-14.3-32-32V109.3L310.6 214.6c-11.8 11.8-30.8 12.6-43.5 1.7L176 138.1 84.8 216.3c-13.4 11.5-33.6 9.9-45.1-3.5s-9.9-33.6 3.5-45.1l112-96c12-10.3 29.7-10.3 41.7 0l89.5 76.7L370.7 64H352c-17.7 0-32-14.3-32-32s14.3-32 32-32h96 0c8.8 0 16.8 3.6 22.6 9.3l.1 .1zM0 304c0-26.5 21.5-48 48-48H464c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V304zM48 416v48H96c0-26.5-21.5-48-48-48zM96 304H48v48c26.5 0 48-21.5 48-48zM464 416c-26.5 0-48 21.5-48 48h48V416zM416 304c0 26.5 21.5 48 48 48V304H416zm-96 80a64 64 0 1 0 -128 0 64 64 0 1 0 128 0z\"],\"house-flood-water-circle-arrow-right\":[640,512,[],\"e50f\",\"M288 144A144 144 0 1 0 0 144a144 144 0 1 0 288 0zM140.7 76.7c6.2-6.2 16.4-6.2 22.6 0l56 56c6.2 6.2 6.2 16.4 0 22.6l-56 56c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L169.4 160H80c-8.8 0-16-7.2-16-16s7.2-16 16-16h89.4L140.7 99.3c-6.2-6.2-6.2-16.4 0-22.6zM320 144c0 57.3-27.4 108.2-69.8 140.3c11.8-3.6 23-9.4 33-16.2c22.1-15.5 51.6-15.5 73.7 0c18.4 12.7 39.6 20.3 59.2 20.3c19 0 41.2-7.9 59.2-20.3c23.8-16.7 55.8-15.4 78.1 3.4c2.1 1.7 4.2 3.3 6.5 4.9l-.3-84.4H576c13.9 0 26.1-8.9 30.4-22.1s-.4-27.6-11.6-35.8l-176-128C407.6-2 392.4-2 381.2 6.1L301 64.4c12.1 23.9 19 50.9 19 79.6zm18.5 165.9c-11.1-7.9-25.9-7.9-37 0C279 325.4 251.5 336 224 336c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C158.5 389.7 191 400 224 400c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.6 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.5-27.3-10.1-39.2-1.7l0 0C471.4 325.2 442.9 336 416 336c-27.5 0-55-10.6-77.5-26.1zm0 112c-11.1-7.9-25.9-7.9-37 0C279 437.4 251.5 448 224 448c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C158.5 501.7 191 512 224 512c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.6 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.4-27.3-10.1-39.2-1.7l0 0C471.4 437.2 442.9 448 416 448c-27.5 0-55-10.6-77.5-26.1z\"],\"square-poll-horizontal\":[448,512,[\"poll-h\"],\"f682\",\"M448 96c0-35.3-28.7-64-64-64L64 32C28.7 32 0 60.7 0 96L0 416c0 35.3 28.7 64 64 64l320 0c35.3 0 64-28.7 64-64l0-320zM256 160c0 17.7-14.3 32-32 32l-96 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l96 0c17.7 0 32 14.3 32 32zm64 64c17.7 0 32 14.3 32 32s-14.3 32-32 32l-192 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l192 0zM192 352c0 17.7-14.3 32-32 32l-32 0c-17.7 0-32-14.3-32-32s14.3-32 32-32l32 0c17.7 0 32 14.3 32 32z\"],circle:[512,512,[128308,128309,128992,128993,128994,128995,128996,9679,9898,9899,11044,61708,61915],\"f111\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512z\"],\"backward-fast\":[512,512,[9198,\"fast-backward\"],\"f049\",\"M493.6 445c-11.2 5.3-24.5 3.6-34.1-4.4L288 297.7V416c0 12.4-7.2 23.7-18.4 29s-24.5 3.6-34.1-4.4L64 297.7V416c0 17.7-14.3 32-32 32s-32-14.3-32-32V96C0 78.3 14.3 64 32 64s32 14.3 32 32V214.3L235.5 71.4c9.5-7.9 22.8-9.7 34.1-4.4S288 83.6 288 96V214.3L459.5 71.4c9.5-7.9 22.8-9.7 34.1-4.4S512 83.6 512 96V416c0 12.4-7.2 23.7-18.4 29z\"],recycle:[512,512,[9842,9850,9851],\"f1b8\",\"M174.7 45.1C192.2 17 223 0 256 0s63.8 17 81.3 45.1l38.6 61.7 27-15.6c8.4-4.9 18.9-4.2 26.6 1.7s11.1 15.9 8.6 25.3l-23.4 87.4c-3.4 12.8-16.6 20.4-29.4 17l-87.4-23.4c-9.4-2.5-16.3-10.4-17.6-20s3.4-19.1 11.8-23.9l28.4-16.4L283 79c-5.8-9.3-16-15-27-15s-21.2 5.7-27 15l-17.5 28c-9.2 14.8-28.6 19.5-43.6 10.5c-15.3-9.2-20.2-29.2-10.7-44.4l17.5-28zM429.5 251.9c15-9 34.4-4.3 43.6 10.5l24.4 39.1c9.4 15.1 14.4 32.4 14.6 50.2c.3 53.1-42.7 96.4-95.8 96.4L320 448v32c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-64-64c-9.4-9.4-9.4-24.6 0-33.9l64-64c6.9-6.9 17.2-8.9 26.2-5.2s14.8 12.5 14.8 22.2v32l96.2 0c17.6 0 31.9-14.4 31.8-32c0-5.9-1.7-11.7-4.8-16.7l-24.4-39.1c-9.5-15.2-4.7-35.2 10.7-44.4zm-364.6-31L36 204.2c-8.4-4.9-13.1-14.3-11.8-23.9s8.2-17.5 17.6-20l87.4-23.4c12.8-3.4 26 4.2 29.4 17L182 241.2c2.5 9.4-.9 19.3-8.6 25.3s-18.2 6.6-26.6 1.7l-26.5-15.3L68.8 335.3c-3.1 5-4.8 10.8-4.8 16.7c-.1 17.6 14.2 32 31.8 32l32.2 0c17.7 0 32 14.3 32 32s-14.3 32-32 32l-32.2 0C42.7 448-.3 404.8 0 351.6c.1-17.8 5.1-35.1 14.6-50.2l50.3-80.5z\"],\"user-astronaut\":[448,512,[],\"f4fb\",\"M370.7 96.1C346.1 39.5 289.7 0 224 0S101.9 39.5 77.3 96.1C60.9 97.5 48 111.2 48 128v64c0 16.8 12.9 30.5 29.3 31.9C101.9 280.5 158.3 320 224 320s122.1-39.5 146.7-96.1c16.4-1.4 29.3-15.1 29.3-31.9V128c0-16.8-12.9-30.5-29.3-31.9zM336 144v16c0 53-43 96-96 96H208c-53 0-96-43-96-96V144c0-26.5 21.5-48 48-48H288c26.5 0 48 21.5 48 48zM189.3 162.7l-6-21.2c-.9-3.3-3.9-5.5-7.3-5.5s-6.4 2.2-7.3 5.5l-6 21.2-21.2 6c-3.3 .9-5.5 3.9-5.5 7.3s2.2 6.4 5.5 7.3l21.2 6 6 21.2c.9 3.3 3.9 5.5 7.3 5.5s6.4-2.2 7.3-5.5l6-21.2 21.2-6c3.3-.9 5.5-3.9 5.5-7.3s-2.2-6.4-5.5-7.3l-21.2-6zM112.7 316.5C46.7 342.6 0 407 0 482.3C0 498.7 13.3 512 29.7 512H128V448c0-17.7 14.3-32 32-32H288c17.7 0 32 14.3 32 32v64l98.3 0c16.4 0 29.7-13.3 29.7-29.7c0-75.3-46.7-139.7-112.7-165.8C303.9 338.8 265.5 352 224 352s-79.9-13.2-111.3-35.5zM176 448c-8.8 0-16 7.2-16 16v48h32V464c0-8.8-7.2-16-16-16zm96 32a16 16 0 1 0 0-32 16 16 0 1 0 0 32z\"],\"plane-slash\":[640,512,[],\"e069\",\"M440.6 320h73.8c34.2 0 93.7-28 93.7-64c0-35-59.5-64-93.7-64l-116.6 0L297.2 16.1C291.5 6.2 280.9 0 269.4 0H213.2c-10.6 0-18.3 10.2-15.4 20.4l40.3 140.9L38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L440.6 320zm-78.1 61.6L41.5 128.7C36 130.6 32 135.9 32 142c0 1.3 .2 2.6 .5 3.9L64 256 32.5 366.1c-.4 1.3-.5 2.6-.5 3.9c0 7.8 6.3 14 14 14H88c5 0 9.8-2.4 12.8-6.4L144 320H246.9l-49 171.6c-2.9 10.2 4.8 20.4 15.4 20.4l56.2 0c11.5 0 22.1-6.2 27.8-16.1l65.3-114.3z\"],trademark:[640,512,[8482],\"f25c\",\"M345.6 108.8c-8.3-11-22.7-15.5-35.7-11.2S288 114.2 288 128V384c0 17.7 14.3 32 32 32s32-14.3 32-32V224l86.4 115.2c6 8.1 15.5 12.8 25.6 12.8s19.6-4.7 25.6-12.8L576 224V384c0 17.7 14.3 32 32 32s32-14.3 32-32V128c0-13.8-8.8-26-21.9-30.4s-27.5 .1-35.7 11.2L464 266.7 345.6 108.8zM0 128c0 17.7 14.3 32 32 32H96V384c0 17.7 14.3 32 32 32s32-14.3 32-32V160h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H32C14.3 96 0 110.3 0 128z\"],basketball:[512,512,[127936,\"basketball-ball\"],\"f434\",\"M86.6 64l85.2 85.2C194.5 121.7 208 86.4 208 48c0-14.7-2-28.9-5.7-42.4C158.6 15 119 35.5 86.6 64zM64 86.6C35.5 119 15 158.6 5.6 202.3C19.1 206 33.3 208 48 208c38.4 0 73.7-13.5 101.3-36.1L64 86.6zM256 0c-7.3 0-14.6 .3-21.8 .9C238 16 240 31.8 240 48c0 47.3-17.1 90.5-45.4 124L256 233.4 425.4 64C380.2 24.2 320.9 0 256 0zM48 240c-16.2 0-32-2-47.1-5.8C.3 241.4 0 248.7 0 256c0 64.9 24.2 124.2 64 169.4L233.4 256 172 194.6C138.5 222.9 95.3 240 48 240zm463.1 37.8c.6-7.2 .9-14.5 .9-21.8c0-64.9-24.2-124.2-64-169.4L278.6 256 340 317.4c33.4-28.3 76.7-45.4 124-45.4c16.2 0 32 2 47.1 5.8zm-4.7 31.9C492.9 306 478.7 304 464 304c-38.4 0-73.7 13.5-101.3 36.1L448 425.4c28.5-32.3 49.1-71.9 58.4-115.7zM340.1 362.7C317.5 390.3 304 425.6 304 464c0 14.7 2 28.9 5.7 42.4C353.4 497 393 476.5 425.4 448l-85.2-85.2zM317.4 340L256 278.6 86.6 448c45.1 39.8 104.4 64 169.4 64c7.3 0 14.6-.3 21.8-.9C274 496 272 480.2 272 464c0-47.3 17.1-90.5 45.4-124z\"],\"satellite-dish\":[512,512,[128225],\"f7c0\",\"M192 32c0-17.7 14.3-32 32-32C383.1 0 512 128.9 512 288c0 17.7-14.3 32-32 32s-32-14.3-32-32C448 164.3 347.7 64 224 64c-17.7 0-32-14.3-32-32zM60.6 220.6L164.7 324.7l28.4-28.4c-.7-2.6-1.1-5.4-1.1-8.3c0-17.7 14.3-32 32-32s32 14.3 32 32s-14.3 32-32 32c-2.9 0-5.6-.4-8.3-1.1l-28.4 28.4L291.4 451.4c14.5 14.5 11.8 38.8-7.3 46.3C260.5 506.9 234.9 512 208 512C93.1 512 0 418.9 0 304c0-26.9 5.1-52.5 14.4-76.1c7.5-19 31.8-21.8 46.3-7.3zM224 96c106 0 192 86 192 192c0 17.7-14.3 32-32 32s-32-14.3-32-32c0-70.7-57.3-128-128-128c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],\"circle-up\":[512,512,[61467,\"arrow-alt-circle-up\"],\"f35b\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM135.1 217.4l107.1-99.9c3.8-3.5 8.7-5.5 13.8-5.5s10.1 2 13.8 5.5l107.1 99.9c4.5 4.2 7.1 10.1 7.1 16.3c0 12.3-10 22.3-22.3 22.3H304v96c0 17.7-14.3 32-32 32H240c-17.7 0-32-14.3-32-32V256H150.3C138 256 128 246 128 233.7c0-6.2 2.6-12.1 7.1-16.3z\"],\"mobile-screen-button\":[384,512,[\"mobile-alt\"],\"f3cd\",\"M16 64C16 28.7 44.7 0 80 0H304c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H80c-35.3 0-64-28.7-64-64V64zM224 448a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM304 64H80V384H304V64z\"],\"volume-high\":[640,512,[128266,\"volume-up\"],\"f028\",\"M533.6 32.5C598.5 85.2 640 165.8 640 256s-41.5 170.7-106.4 223.5c-10.3 8.4-25.4 6.8-33.8-3.5s-6.8-25.4 3.5-33.8C557.5 398.2 592 331.2 592 256s-34.5-142.2-88.7-186.3c-10.3-8.4-11.8-23.5-3.5-33.8s23.5-11.8 33.8-3.5zM473.1 107c43.2 35.2 70.9 88.9 70.9 149s-27.7 113.8-70.9 149c-10.3 8.4-25.4 6.8-33.8-3.5s-6.8-25.4 3.5-33.8C475.3 341.3 496 301.1 496 256s-20.7-85.3-53.2-111.8c-10.3-8.4-11.8-23.5-3.5-33.8s23.5-11.8 33.8-3.5zm-60.5 74.5C434.1 199.1 448 225.9 448 256s-13.9 56.9-35.4 74.5c-10.3 8.4-25.4 6.8-33.8-3.5s-6.8-25.4 3.5-33.8C393.1 284.4 400 271 400 256s-6.9-28.4-17.7-37.3c-10.3-8.4-11.8-23.5-3.5-33.8s23.5-11.8 33.8-3.5zM301.1 34.8C312.6 40 320 51.4 320 64V448c0 12.6-7.4 24-18.9 29.2s-25 3.1-34.4-5.3L131.8 352H64c-35.3 0-64-28.7-64-64V224c0-35.3 28.7-64 64-64h67.8L266.7 40.1c9.4-8.4 22.9-10.4 34.4-5.3z\"],\"users-rays\":[640,512,[],\"e593\",\"M41 7C31.6-2.3 16.4-2.3 7 7S-2.3 31.6 7 41l72 72c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L41 7zM599 7L527 79c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l72-72c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0zM7 505c9.4 9.4 24.6 9.4 33.9 0l72-72c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0L7 471c-9.4 9.4-9.4 24.6 0 33.9zm592 0c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-72-72c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l72 72zM320 256a64 64 0 1 0 0-128 64 64 0 1 0 0 128zM212.1 336c-2.7 7.5-4.1 15.6-4.1 24c0 13.3 10.7 24 24 24H408c13.3 0 24-10.7 24-24c0-8.4-1.4-16.5-4.1-24c-.5-1.4-1-2.7-1.6-4c-9.4-22.3-29.8-38.9-54.3-43c-3.9-.7-7.9-1-12-1H280c-4.1 0-8.1 .3-12 1c-.8 .1-1.7 .3-2.5 .5c-24.9 5.1-45.1 23-53.4 46.5zM175.8 224a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-26.5 32C119.9 256 96 279.9 96 309.3c0 14.7 11.9 26.7 26.7 26.7h56.1c8-34.1 32.8-61.7 65.2-73.6c-7.5-4.1-16.2-6.4-25.3-6.4H149.3zm368 80c14.7 0 26.7-11.9 26.7-26.7c0-29.5-23.9-53.3-53.3-53.3H421.3c-9.2 0-17.8 2.3-25.3 6.4c32.4 11.9 57.2 39.5 65.2 73.6h56.1zM464 224a48 48 0 1 0 0-96 48 48 0 1 0 0 96z\"],wallet:[512,512,[],\"f555\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V192c0-35.3-28.7-64-64-64H80c-8.8 0-16-7.2-16-16s7.2-16 16-16H448c17.7 0 32-14.3 32-32s-14.3-32-32-32H64zM416 272a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"clipboard-check\":[384,512,[],\"f46c\",\"M192 0c-41.8 0-77.4 26.7-90.5 64H64C28.7 64 0 92.7 0 128V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H282.5C269.4 26.7 233.8 0 192 0zm0 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM305 273L177 401c-9.4 9.4-24.6 9.4-33.9 0L79 337c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l47 47L271 239c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9z\"],\"file-audio\":[384,512,[],\"f1c7\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zm2 226.3c37.1 22.4 62 63.1 62 109.7s-24.9 87.3-62 109.7c-7.6 4.6-17.4 2.1-22-5.4s-2.1-17.4 5.4-22C269.4 401.5 288 370.9 288 336s-18.6-65.5-46.5-82.3c-7.6-4.6-10-14.4-5.4-22s14.4-10 22-5.4zm-91.9 30.9c6 2.5 9.9 8.3 9.9 14.8V400c0 6.5-3.9 12.3-9.9 14.8s-12.9 1.1-17.4-3.5L113.4 376H80c-8.8 0-16-7.2-16-16V312c0-8.8 7.2-16 16-16h33.4l35.3-35.3c4.6-4.6 11.5-5.9 17.4-3.5zm51 34.9c6.6-5.9 16.7-5.3 22.6 1.3C249.8 304.6 256 319.6 256 336s-6.2 31.4-16.3 42.7c-5.9 6.6-16 7.1-22.6 1.3s-7.1-16-1.3-22.6c5.1-5.7 8.1-13.1 8.1-21.3s-3.1-15.7-8.1-21.3c-5.9-6.6-5.3-16.7 1.3-22.6z\"],burger:[512,512,[\"hamburger\"],\"f805\",\"M61.1 224C45 224 32 211 32 194.9c0-1.9 .2-3.7 .6-5.6C37.9 168.3 78.8 32 256 32s218.1 136.3 223.4 157.3c.5 1.9 .6 3.7 .6 5.6c0 16.1-13 29.1-29.1 29.1H61.1zM144 128a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm240 16a16 16 0 1 0 0-32 16 16 0 1 0 0 32zM272 96a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zM16 304c0-26.5 21.5-48 48-48H448c26.5 0 48 21.5 48 48s-21.5 48-48 48H64c-26.5 0-48-21.5-48-48zm16 96c0-8.8 7.2-16 16-16H464c8.8 0 16 7.2 16 16v16c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V400z\"],wrench:[512,512,[128295],\"f0ad\",\"M352 320c88.4 0 160-71.6 160-160c0-15.3-2.2-30.1-6.2-44.2c-3.1-10.8-16.4-13.2-24.3-5.3l-76.8 76.8c-3 3-7.1 4.7-11.3 4.7H336c-8.8 0-16-7.2-16-16V118.6c0-4.2 1.7-8.3 4.7-11.3l76.8-76.8c7.9-7.9 5.4-21.2-5.3-24.3C382.1 2.2 367.3 0 352 0C263.6 0 192 71.6 192 160c0 19.1 3.4 37.5 9.5 54.5L19.9 396.1C7.2 408.8 0 426.1 0 444.1C0 481.6 30.4 512 67.9 512c18 0 35.3-7.2 48-19.9L297.5 310.5c17 6.2 35.4 9.5 54.5 9.5zM80 408a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],bugs:[576,512,[],\"e4d0\",\"M164.5 107.4l33.4-73.5c5.5-12.1 .1-26.3-11.9-31.8s-26.3-.1-31.8 11.9L128 71.7 101.9 14.1C96.4 2 82.1-3.3 70.1 2.1S52.7 21.9 58.1 33.9l33.4 73.5c-10.2 7.1-18.2 17-22.9 28.6h-17l-4.1-20.7c-2.6-13-15.2-21.4-28.2-18.8S-2.1 111.7 .5 124.7l8 40C10.7 175.9 20.6 184 32 184H64v23.3l-37.8 9.5c-9.5 2.4-16.6 10.2-17.9 19.9l-8 56c-1.9 13.1 7.2 25.3 20.4 27.2s25.3-7.2 27.2-20.4l5.7-40 18.4-4.6C82.7 274.6 103.8 288 128 288s45.3-13.4 56.1-33.2l18.4 4.6 5.7 40c1.9 13.1 14 22.2 27.2 20.4s22.2-14 20.4-27.2l-8-56c-1.4-9.7-8.5-17.5-17.9-19.9L192 207.3V184h32c11.4 0 21.3-8.1 23.5-19.3l8-40c2.6-13-5.8-25.6-18.8-28.2s-25.6 5.8-28.2 18.8L204.3 136h-17c-4.7-11.6-12.7-21.5-22.9-28.6zM496 286.5l65.6-47c10.8-7.7 13.3-22.7 5.6-33.5s-22.7-13.3-33.5-5.6l-51.4 36.8 6.1-62.9c1.3-13.2-8.4-24.9-21.6-26.2s-24.9 8.4-26.2 21.6L432.8 250c-12.3 1-24.2 5.6-34.1 13.3L384 254.8l6.8-20c4.2-12.6-2.5-26.2-15-30.4s-26.2 2.5-30.4 15l-13.1 38.6c-3.7 10.8 .8 22.8 10.7 28.5l27.7 16L359 322.7 321.5 312c-9.4-2.7-19.5 .6-25.5 8.3l-34.9 44.5c-8.2 10.4-6.4 25.5 4.1 33.7s25.5 6.4 33.7-4.1l25-31.8 18.2 5.2c-.5 22.6 11 44.7 32 56.8s45.9 11 65.2-.7l13.6 13.2-15.1 37.5c-4.9 12.3 1 26.3 13.3 31.2s26.3-1 31.2-13.3L503.5 440c3.6-9.1 1.4-19.4-5.6-26.2l-28-27.1 11.6-20.1 27.7 16c9.9 5.7 22.5 3.7 30-4.9L566.2 347c8.7-10 7.8-25.1-2.2-33.9s-25.1-7.8-33.9 2.2l-13.9 15.9-14.7-8.5c1.7-12.4-.2-25-5.5-36.2z\"],\"rupee-sign\":[448,512,[8360,\"rupee\"],\"f156\",\"M0 64C0 46.3 14.3 32 32 32h80c79.5 0 144 64.5 144 144c0 58.8-35.2 109.3-85.7 131.7l51.4 128.4c6.6 16.4-1.4 35-17.8 41.6s-35-1.4-41.6-17.8L106.3 320H64V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V288 64zM64 256h48c44.2 0 80-35.8 80-80s-35.8-80-80-80H64V256zm256.5 16.4c-.9 6 0 8.7 .4 9.8c.4 1.1 1.4 2.6 4.2 4.9c7.2 5.7 18.7 10 37.9 16.8l1.3 .5c16 5.6 38.7 13.6 55.7 28.1c9.5 8.1 17.9 18.6 23.1 32.3c5.1 13.7 6.1 28.5 3.8 44c-4.2 28.1-20.5 49.3-43.8 60.9c-22.1 11-48.1 12.5-73.2 8l-.2 0 0 0c-9.3-1.8-20.5-5.7-29.3-9c-6-2.3-12.6-4.9-17.7-6.9l0 0c-2.5-1-4.6-1.8-6.3-2.5c-16.5-6.4-24.6-25-18.2-41.4s24.9-24.6 41.4-18.2c2.6 1 5.2 2 7.9 3.1l0 0c4.8 1.9 9.8 3.9 15.4 6c8.8 3.3 15.3 5.4 18.7 6c15.7 2.8 26.7 .8 32.9-2.3c5-2.5 8-6 9.1-13c1-6.9 .2-10.5-.5-12.3c-.6-1.7-1.8-3.6-4.5-5.9c-6.9-5.8-18.2-10.4-36.9-17l-3-1.1c-15.5-5.4-37-13-53.3-25.9c-9.5-7.5-18.3-17.6-23.7-31c-5.5-13.4-6.6-28-4.4-43.2c8.4-57.1 67-78 116.9-68.9c6.9 1.3 27.3 5.8 35.4 8.4c16.9 5.2 26.3 23.2 21.1 40.1s-23.2 26.3-40.1 21.1c-4.7-1.4-22.3-5.5-27.9-6.5c-14.6-2.7-25.8-.4-32.6 3.2c-6.3 3.3-8.9 7.6-9.5 12z\"],\"file-image\":[384,512,[128443],\"f1c5\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM64 256a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm152 32c5.3 0 10.2 2.6 13.2 6.9l88 128c3.4 4.9 3.7 11.3 1 16.5s-8.2 8.6-14.2 8.6H216 176 128 80c-5.8 0-11.1-3.1-13.9-8.1s-2.8-11.2 .2-16.1l48-80c2.9-4.8 8.1-7.8 13.7-7.8s10.8 2.9 13.7 7.8l12.8 21.4 48.3-70.2c3-4.3 7.9-6.9 13.2-6.9z\"],\"circle-question\":[512,512,[62108,\"question-circle\"],\"f059\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM169.8 165.3c7.9-22.3 29.1-37.3 52.8-37.3h58.3c34.9 0 63.1 28.3 63.1 63.1c0 22.6-12.1 43.5-31.7 54.8L280 264.4c-.2 13-10.9 23.6-24 23.6c-13.3 0-24-10.7-24-24V250.5c0-8.6 4.6-16.5 12.1-20.8l44.3-25.4c4.7-2.7 7.6-7.7 7.6-13.1c0-8.4-6.8-15.1-15.1-15.1H222.6c-3.4 0-6.4 2.1-7.5 5.3l-.4 1.2c-4.4 12.5-18.2 19-30.6 14.6s-19-18.2-14.6-30.6l.4-1.2zM224 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],\"plane-departure\":[640,512,[128747],\"f5b0\",\"M381 114.9L186.1 41.8c-16.7-6.2-35.2-5.3-51.1 2.7L89.1 67.4C78 73 77.2 88.5 87.6 95.2l146.9 94.5L136 240 77.8 214.1c-8.7-3.9-18.8-3.7-27.3 .6L18.3 230.8c-9.3 4.7-11.8 16.8-5 24.7l73.1 85.3c6.1 7.1 15 11.2 24.3 11.2H248.4c5 0 9.9-1.2 14.3-3.4L535.6 212.2c46.5-23.3 82.5-63.3 100.8-112C645.9 75 627.2 48 600.2 48H542.8c-20.2 0-40.2 4.8-58.2 14L381 114.9zM0 480c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H32c-17.7 0-32 14.3-32 32z\"],\"handshake-slash\":[640,512,[],\"e060\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-135-105.8c-1.1-11.3-6.3-22.3-15.3-30.7l-134.2-123-23.4 18.2-26-20.3 77.2-60.1c7-5.4 17-4.2 22.5 2.8s4.2 17-2.8 22.5l-20.9 16.2L512 316.8V128h-.7l-3.9-2.5L434.8 79c-15.3-9.8-33.2-15-51.4-15c-21.8 0-43 7.5-60 21.2l-89.7 72.6-25.8-20.3 81.8-66.2c-11.6-4.9-24.1-7.4-36.8-7.4C234 64 215.7 69.6 200 80l-35.5 23.7L38.8 5.1zM413.6 421.9L128 196.8V352h28.2l91.4 83.4c19.6 17.9 49.9 16.5 67.8-3.1c5.5-6.1 9.2-13.2 11.1-20.6l17 15.6c19.5 17.9 49.9 16.6 67.8-2.9c.8-.8 1.5-1.7 2.2-2.6zM96 171.6L40.6 128H16c-8.8 0-16 7.2-16 16V352c0 17.7 14.3 32 32 32H64c17.7 0 32-14.3 32-32V171.6zM48 320a16 16 0 1 1 0 32 16 16 0 1 1 0-32zM544 128V352c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V144c0-8.8-7.2-16-16-16H544zm32 208a16 16 0 1 1 32 0 16 16 0 1 1 -32 0z\"],\"book-bookmark\":[448,512,[],\"e0bb\",\"M0 96C0 43 43 0 96 0h96V190.7c0 13.4 15.5 20.9 26 12.5L272 160l54 43.2c10.5 8.4 26 .9 26-12.5V0h32 32c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384 96c-53 0-96-43-96-96V96zM64 416c0 17.7 14.3 32 32 32H352V384H96c-17.7 0-32 14.3-32 32z\"],\"code-branch\":[448,512,[],\"f126\",\"M80 104a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm80-24c0 32.8-19.7 61-48 73.3v87.8c18.8-10.9 40.7-17.1 64-17.1h96c35.3 0 64-28.7 64-64v-6.7C307.7 141 288 112.8 288 80c0-44.2 35.8-80 80-80s80 35.8 80 80c0 32.8-19.7 61-48 73.3V160c0 70.7-57.3 128-128 128H176c-35.3 0-64 28.7-64 64v6.7c28.3 12.3 48 40.5 48 73.3c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-32.8 19.7-61 48-73.3V352 153.3C19.7 141 0 112.8 0 80C0 35.8 35.8 0 80 0s80 35.8 80 80zm232 0a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zM80 456a24 24 0 1 0 0-48 24 24 0 1 0 0 48z\"],\"hat-cowboy\":[640,512,[],\"f8c0\",\"M320 64c14.4 0 22.3-7 30.8-14.4C360.4 41.1 370.7 32 392 32c49.3 0 84.4 152.2 97.9 221.9C447.8 272.1 390.9 288 320 288s-127.8-15.9-169.9-34.1C163.6 184.2 198.7 32 248 32c21.3 0 31.6 9.1 41.2 17.6C297.7 57 305.6 64 320 64zM111.1 270.7c47.2 24.5 117.5 49.3 209 49.3s161.8-24.8 208.9-49.3c24.8-12.9 49.8-28.3 70.1-47.7c7.9-7.9 20.2-9.2 29.6-3.3c9.5 5.9 13.5 17.9 9.9 28.5c-13.5 37.7-38.4 72.3-66.1 100.6C523.7 398.9 443.6 448 320 448s-203.6-49.1-252.5-99.2C39.8 320.4 14.9 285.8 1.4 248.1c-3.6-10.6 .4-22.6 9.9-28.5c9.5-5.9 21.7-4.5 29.6 3.3c20.4 19.4 45.3 34.8 70.1 47.7z\"],bridge:[576,512,[],\"e4c8\",\"M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96H72v64H0V288c53 0 96 43 96 96v64c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V384c0-53 43-96 96-96s96 43 96 96v64c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V384c0-53 43-96 96-96V160H504V96h40c17.7 0 32-14.3 32-32s-14.3-32-32-32H32zM456 96v64H376V96h80zM328 96v64H248V96h80zM200 96v64H120V96h80z\"],\"phone-flip\":[512,512,[128381,\"phone-alt\"],\"f879\",\"M347.1 24.6c7.7-18.6 28-28.5 47.4-23.2l88 24C499.9 30.2 512 46 512 64c0 247.4-200.6 448-448 448c-18 0-33.8-12.1-38.6-29.5l-24-88c-5.3-19.4 4.6-39.7 23.2-47.4l96-40c16.3-6.8 35.2-2.1 46.3 11.6L207.3 368c70.4-33.3 127.4-90.3 160.7-160.7L318.7 167c-13.7-11.2-18.4-30-11.6-46.3l40-96z\"],\"truck-front\":[512,512,[],\"e2b7\",\"M0 80C0 35.8 35.8 0 80 0H432c44.2 0 80 35.8 80 80V368c0 26.2-12.6 49.4-32 64v48c0 17.7-14.3 32-32 32H416c-17.7 0-32-14.3-32-32V448H128v32c0 17.7-14.3 32-32 32H64c-17.7 0-32-14.3-32-32V432C12.6 417.4 0 394.2 0 368V80zm129.9 72.2L112 224H400l-17.9-71.8C378.5 138 365.7 128 351 128H161c-14.7 0-27.5 10-31 24.2zM128 320a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm288 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],cat:[576,512,[128008],\"f6be\",\"M320 192h17.1c22.1 38.3 63.5 64 110.9 64c11 0 21.8-1.4 32-4v4 32V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V339.2L280 448h56c17.7 0 32 14.3 32 32s-14.3 32-32 32H192c-53 0-96-43-96-96V192.5c0-16.1-12-29.8-28-31.8l-7.9-1c-17.5-2.2-30-18.2-27.8-35.7s18.2-30 35.7-27.8l7.9 1c48 6 84.1 46.8 84.1 95.3v85.3c34.4-51.7 93.2-85.8 160-85.8zm160 26.5v0c-10 3.5-20.8 5.5-32 5.5c-28.4 0-54-12.4-71.6-32h0c-3.7-4.1-7-8.5-9.9-13.2C357.3 164 352 146.6 352 128v0V32 12 10.7C352 4.8 356.7 .1 362.6 0h.2c3.3 0 6.4 1.6 8.4 4.2l0 .1L384 21.3l27.2 36.3L416 64h64l4.8-6.4L512 21.3 524.8 4.3l0-.1c2-2.6 5.1-4.2 8.4-4.2h.2C539.3 .1 544 4.8 544 10.7V12 32v96c0 17.3-4.6 33.6-12.6 47.6c-11.3 19.8-29.6 35.2-51.4 42.9zM432 128a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm48 16a16 16 0 1 0 0-32 16 16 0 1 0 0 32z\"],\"anchor-circle-exclamation\":[640,512,[],\"e4ab\",\"M320 96a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm21.1 80C367 158.8 384 129.4 384 96c0-53-43-96-96-96s-96 43-96 96c0 33.4 17 62.8 42.9 80H224c-17.7 0-32 14.3-32 32s14.3 32 32 32h32V448H208c-53 0-96-43-96-96v-6.1l7 7c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L97 263c-9.4-9.4-24.6-9.4-33.9 0L7 319c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l7-7V352c0 88.4 71.6 160 160 160h80 80c8.2 0 16.3-.6 24.2-1.8c-22.2-16.2-40.4-37.5-53-62.2H320V368 240h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H341.1zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z\"],\"truck-field\":[640,512,[],\"e58d\",\"M32 96c0-35.3 28.7-64 64-64H320c23.7 0 44.4 12.9 55.4 32h51.8c25.3 0 48.2 14.9 58.5 38l52.8 118.8c.5 1.1 .9 2.1 1.3 3.2H544c35.3 0 64 28.7 64 64v32c17.7 0 32 14.3 32 32s-14.3 32-32 32H576c0 53-43 96-96 96s-96-43-96-96H256c0 53-43 96-96 96s-96-43-96-96H32c-17.7 0-32-14.3-32-32s14.3-32 32-32V288c-17.7 0-32-14.3-32-32V160c0-17.7 14.3-32 32-32V96zM384 224h85.9l-42.7-96H384v96zM160 432a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm368-48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0z\"],route:[512,512,[],\"f4d7\",\"M512 96c0 50.2-59.1 125.1-84.6 155c-3.8 4.4-9.4 6.1-14.5 5H320c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c53 0 96 43 96 96s-43 96-96 96H139.6c8.7-9.9 19.3-22.6 30-36.8c6.3-8.4 12.8-17.6 19-27.2H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H320c-53 0-96-43-96-96s43-96 96-96h39.8c-21-31.5-39.8-67.7-39.8-96c0-53 43-96 96-96s96 43 96 96zM117.1 489.1c-3.8 4.3-7.2 8.1-10.1 11.3l-1.8 2-.2-.2c-6 4.6-14.6 4-20-1.8C59.8 473 0 402.5 0 352c0-53 43-96 96-96s96 43 96 96c0 30-21.1 67-43.5 97.9c-10.7 14.7-21.7 28-30.8 38.5l-.6 .7zM128 352a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM416 128a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"clipboard-question\":[384,512,[],\"e4e3\",\"M192 0c-41.8 0-77.4 26.7-90.5 64H64C28.7 64 0 92.7 0 128V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H282.5C269.4 26.7 233.8 0 192 0zm0 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM105.8 229.3c7.9-22.3 29.1-37.3 52.8-37.3h58.3c34.9 0 63.1 28.3 63.1 63.1c0 22.6-12.1 43.5-31.7 54.8L216 328.4c-.2 13-10.9 23.6-24 23.6c-13.3 0-24-10.7-24-24V314.5c0-8.6 4.6-16.5 12.1-20.8l44.3-25.4c4.7-2.7 7.6-7.7 7.6-13.1c0-8.4-6.8-15.1-15.1-15.1H158.6c-3.4 0-6.4 2.1-7.5 5.3l-.4 1.2c-4.4 12.5-18.2 19-30.6 14.6s-19-18.2-14.6-30.6l.4-1.2zM160 416a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],panorama:[640,512,[],\"e209\",\"M45.6 32C20.4 32 0 52.4 0 77.6V434.4C0 459.6 20.4 480 45.6 480c5.1 0 10-.8 14.7-2.4C74.6 472.8 177.6 440 320 440s245.4 32.8 259.6 37.6c4.7 1.6 9.7 2.4 14.7 2.4c25.2 0 45.6-20.4 45.6-45.6V77.6C640 52.4 619.6 32 594.4 32c-5 0-10 .8-14.7 2.4C565.4 39.2 462.4 72 320 72S74.6 39.2 60.4 34.4C55.6 32.8 50.7 32 45.6 32zM96 160a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm272 0c7.9 0 15.4 3.9 19.8 10.5L512.3 353c5.4 8 5.6 18.4 .4 26.5s-14.7 12.3-24.2 10.7C442.7 382.4 385.2 376 320 376c-65.6 0-123.4 6.5-169.3 14.4c-9.8 1.7-19.7-2.9-24.7-11.5s-4.3-19.4 1.9-27.2L197.3 265c4.6-5.7 11.4-9 18.7-9s14.2 3.3 18.7 9l26.4 33.1 87-127.6c4.5-6.6 11.9-10.5 19.8-10.5z\"],\"comment-medical\":[512,512,[],\"f7f5\",\"M256 448c141.4 0 256-93.1 256-208S397.4 32 256 32S0 125.1 0 240c0 45.1 17.7 86.8 47.7 120.9c-1.9 24.5-11.4 46.3-21.4 62.9c-5.5 9.2-11.1 16.6-15.2 21.6c-2.1 2.5-3.7 4.4-4.9 5.7c-.6 .6-1 1.1-1.3 1.4l-.3 .3 0 0 0 0 0 0 0 0c-4.6 4.6-5.9 11.4-3.4 17.4c2.5 6 8.3 9.9 14.8 9.9c28.7 0 57.6-8.9 81.6-19.3c22.9-10 42.4-21.9 54.3-30.6c31.8 11.5 67 17.9 104.1 17.9zM224 160c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H288v48c0 8.8-7.2 16-16 16H240c-8.8 0-16-7.2-16-16V272H176c-8.8 0-16-7.2-16-16V224c0-8.8 7.2-16 16-16h48V160z\"],\"teeth-open\":[576,512,[],\"f62f\",\"M96 32C43 32 0 75 0 128v64c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128c0-53-43-96-96-96H96zM224 96c26.5 0 48 21.5 48 48v56c0 13.3-10.7 24-24 24H200c-13.3 0-24-10.7-24-24V144c0-26.5 21.5-48 48-48zm80 48c0-26.5 21.5-48 48-48s48 21.5 48 48v56c0 13.3-10.7 24-24 24H328c-13.3 0-24-10.7-24-24V144zM96 128c26.5 0 48 21.5 48 48v24c0 13.3-10.7 24-24 24H72c-13.3 0-24-10.7-24-24V176c0-26.5 21.5-48 48-48zm336 48c0-26.5 21.5-48 48-48s48 21.5 48 48v24c0 13.3-10.7 24-24 24H456c-13.3 0-24-10.7-24-24V176zM96 480H480c53 0 96-43 96-96V352c0-35.3-28.7-64-64-64H64c-35.3 0-64 28.7-64 64v32c0 53 43 96 96 96zm0-64c-26.5 0-48-21.5-48-48V344c0-13.3 10.7-24 24-24h48c13.3 0 24 10.7 24 24v24c0 26.5-21.5 48-48 48zm80-48V344c0-13.3 10.7-24 24-24h48c13.3 0 24 10.7 24 24v24c0 26.5-21.5 48-48 48s-48-21.5-48-48zm176 48c-26.5 0-48-21.5-48-48V344c0-13.3 10.7-24 24-24h48c13.3 0 24 10.7 24 24v24c0 26.5-21.5 48-48 48zm80-48V344c0-13.3 10.7-24 24-24h48c13.3 0 24 10.7 24 24v24c0 26.5-21.5 48-48 48s-48-21.5-48-48z\"],\"file-circle-minus\":[576,512,[],\"e4ed\",\"M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v38.6C310.1 219.5 256 287.4 256 368c0 59.1 29.1 111.3 73.7 143.3c-3.2 .5-6.4 .7-9.7 .7H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zM288 368a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm224 0c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16s7.2 16 16 16H496c8.8 0 16-7.2 16-16z\"],tags:[512,512,[],\"f02c\",\"M345 39.1L472.8 168.4c52.4 53 52.4 138.2 0 191.2L360.8 472.9c-9.3 9.4-24.5 9.5-33.9 .2s-9.5-24.5-.2-33.9L438.6 325.9c33.9-34.3 33.9-89.4 0-123.7L310.9 72.9c-9.3-9.4-9.2-24.6 .2-33.9s24.6-9.2 33.9 .2zM0 229.5V80C0 53.5 21.5 32 48 32H197.5c17 0 33.3 6.7 45.3 18.7l168 168c25 25 25 65.5 0 90.5L277.3 442.7c-25 25-65.5 25-90.5 0l-168-168C6.7 262.7 0 246.5 0 229.5zM144 144a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z\"],\"wine-glass\":[320,512,[127863],\"f4e3\",\"M32.1 29.3C33.5 12.8 47.4 0 64 0H256c16.6 0 30.5 12.8 31.9 29.3l14 168.4c6 72-42.5 135.2-109.9 150.6V448h48c17.7 0 32 14.3 32 32s-14.3 32-32 32H160 80c-17.7 0-32-14.3-32-32s14.3-32 32-32h48V348.4C60.6 333 12.1 269.8 18.1 197.8l14-168.4zm56 98.7H231.9l-5.3-64H93.4l-5.3 64z\"],\"forward-fast\":[512,512,[9197,\"fast-forward\"],\"f050\",\"M18.4 445c11.2 5.3 24.5 3.6 34.1-4.4L224 297.7V416c0 12.4 7.2 23.7 18.4 29s24.5 3.6 34.1-4.4L448 297.7V416c0 17.7 14.3 32 32 32s32-14.3 32-32V96c0-17.7-14.3-32-32-32s-32 14.3-32 32V214.3L276.5 71.4c-9.5-7.9-22.8-9.7-34.1-4.4S224 83.6 224 96V214.3L52.5 71.4c-9.5-7.9-22.8-9.7-34.1-4.4S0 83.6 0 96V416c0 12.4 7.2 23.7 18.4 29z\"],\"face-meh-blank\":[512,512,[128566,\"meh-blank\"],\"f5a4\",\"M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm208.4-48a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm128 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"square-parking\":[448,512,[127359,\"parking\"],\"f540\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM192 256h48c17.7 0 32-14.3 32-32s-14.3-32-32-32H192v64zm48 64H192v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V288 168c0-22.1 17.9-40 40-40h72c53 0 96 43 96 96s-43 96-96 96z\"],\"house-signal\":[576,512,[],\"e012\",\"M357.7 8.5c-12.3-11.3-31.2-11.3-43.4 0l-208 192c-9.4 8.6-12.7 22-8.5 34c87.1 25.3 155.6 94.2 180.3 181.6H464c26.5 0 48-21.5 48-48V256h32c13.2 0 25-8.1 29.8-20.3s1.6-26.2-8.1-35.2l-208-192zM288 208c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16v64c0 8.8-7.2 16-16 16H304c-8.8 0-16-7.2-16-16V208zM24 256c-13.3 0-24 10.7-24 24s10.7 24 24 24c101.6 0 184 82.4 184 184c0 13.3 10.7 24 24 24s24-10.7 24-24c0-128.1-103.9-232-232-232zm8 256a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM0 376c0 13.3 10.7 24 24 24c48.6 0 88 39.4 88 88c0 13.3 10.7 24 24 24s24-10.7 24-24c0-75.1-60.9-136-136-136c-13.3 0-24 10.7-24 24z\"],\"bars-progress\":[512,512,[\"tasks-alt\"],\"f828\",\"M448 160H320V128H448v32zM48 64C21.5 64 0 85.5 0 112v64c0 26.5 21.5 48 48 48H464c26.5 0 48-21.5 48-48V112c0-26.5-21.5-48-48-48H48zM448 352v32H192V352H448zM48 288c-26.5 0-48 21.5-48 48v64c0 26.5 21.5 48 48 48H464c26.5 0 48-21.5 48-48V336c0-26.5-21.5-48-48-48H48z\"],\"faucet-drip\":[512,512,[128688],\"e006\",\"M224 0c17.7 0 32 14.3 32 32V44l96-12c17.7 0 32 14.3 32 32s-14.3 32-32 32L256 84l-31-3.9-1-.1-1 .1L192 84 96 96C78.3 96 64 81.7 64 64s14.3-32 32-32l96 12V32c0-17.7 14.3-32 32-32zM0 224c0-17.7 14.3-32 32-32h96l22.6-22.6c6-6 14.1-9.4 22.6-9.4H192V116.2l32-4 32 4V160h18.7c8.5 0 16.6 3.4 22.6 9.4L320 192h32c88.4 0 160 71.6 160 160c0 17.7-14.3 32-32 32H416c-17.7 0-32-14.3-32-32s-14.3-32-32-32H315.9c-20.2 29-53.9 48-91.9 48s-71.7-19-91.9-48H32c-17.7 0-32-14.3-32-32V224zM436.8 423.4c1.9-4.5 6.3-7.4 11.2-7.4s9.2 2.9 11.2 7.4l18.2 42.4c1.8 4.1 2.7 8.6 2.7 13.1V480c0 17.7-14.3 32-32 32s-32-14.3-32-32v-1.2c0-4.5 .9-8.9 2.7-13.1l18.2-42.4z\"],\"cart-flatbed\":[640,512,[\"dolly-flatbed\"],\"f474\",\"M32 0C14.3 0 0 14.3 0 32S14.3 64 32 64H48c8.8 0 16 7.2 16 16V368c0 44.2 35.8 80 80 80h18.7c-1.8 5-2.7 10.4-2.7 16c0 26.5 21.5 48 48 48s48-21.5 48-48c0-5.6-1-11-2.7-16H450.7c-1.8 5-2.7 10.4-2.7 16c0 26.5 21.5 48 48 48s48-21.5 48-48c0-5.6-1-11-2.7-16H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H144c-8.8 0-16-7.2-16-16V80C128 35.8 92.2 0 48 0H32zM192 80V272c0 26.5 21.5 48 48 48H560c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48H464V176c0 5.9-3.2 11.3-8.5 14.1s-11.5 2.5-16.4-.8L400 163.2l-39.1 26.1c-4.9 3.3-11.2 3.6-16.4 .8s-8.5-8.2-8.5-14.1V32H240c-26.5 0-48 21.5-48 48z\"],\"ban-smoking\":[512,512,[128685,\"smoking-ban\"],\"f54d\",\"M99.5 144.8L178.7 224l96 96 92.5 92.5C335.9 434.9 297.5 448 256 448C150 448 64 362 64 256c0-41.5 13.1-79.9 35.5-111.2zM333.3 288l-32-32H384v32H333.3zm32 32H400c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H269.3L144.8 99.5C176.1 77.1 214.5 64 256 64c106 0 192 86 192 192c0 41.5-13.1 79.9-35.5 111.2L365.3 320zM256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM272 96c-8.8 0-16 7.2-16 16c0 26.5 21.5 48 48 48h32c8.8 0 16 7.2 16 16s7.2 16 16 16s16-7.2 16-16c0-26.5-21.5-48-48-48H304c-8.8 0-16-7.2-16-16s-7.2-16-16-16zM229.5 320l-96-96H112c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16H229.5z\"],terminal:[576,512,[],\"f120\",\"M9.4 86.6C-3.1 74.1-3.1 53.9 9.4 41.4s32.8-12.5 45.3 0l192 192c12.5 12.5 12.5 32.8 0 45.3l-192 192c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L178.7 256 9.4 86.6zM256 416H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H256c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],\"mobile-button\":[384,512,[],\"f10b\",\"M80 0C44.7 0 16 28.7 16 64V448c0 35.3 28.7 64 64 64H304c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H80zM192 400a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"house-medical-flag\":[640,512,[],\"e514\",\"M480 0c17.7 0 32 14.3 32 32H624c8.8 0 16 7.2 16 16V176c0 8.8-7.2 16-16 16H512V512H448V192 32c0-17.7 14.3-32 32-32zM276.8 39.7L416 159V512h1l-.2 0H96c-17.7 0-32-14.3-32-32V288H32c-13.4 0-25.4-8.3-30-20.9s-1-26.7 9.2-35.4l224-192c12-10.3 29.7-10.3 41.7 0zM224 208v48H176c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V320h48c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H288V208c0-8.8-7.2-16-16-16H240c-8.8 0-16 7.2-16 16z\"],\"basket-shopping\":[576,512,[\"shopping-basket\"],\"f291\",\"M253.3 35.1c6.1-11.8 1.5-26.3-10.2-32.4s-26.3-1.5-32.4 10.2L117.6 192H32c-17.7 0-32 14.3-32 32s14.3 32 32 32L83.9 463.5C91 492 116.6 512 146 512H430c29.4 0 55-20 62.1-48.5L544 256c17.7 0 32-14.3 32-32s-14.3-32-32-32H458.4L365.3 12.9C359.2 1.2 344.7-3.4 332.9 2.7s-16.3 20.6-10.2 32.4L404.3 192H171.7L253.3 35.1zM192 304v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V304c0-8.8 7.2-16 16-16s16 7.2 16 16zm96-16c8.8 0 16 7.2 16 16v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V304c0-8.8 7.2-16 16-16zm128 16v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V304c0-8.8 7.2-16 16-16s16 7.2 16 16z\"],tape:[576,512,[],\"f4db\",\"M380.8 416c41.5-40.7 67.2-97.3 67.2-160C448 132.3 347.7 32 224 32S0 132.3 0 256S100.3 480 224 480H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H380.8zM224 160a96 96 0 1 1 0 192 96 96 0 1 1 0-192zm64 96a64 64 0 1 0 -128 0 64 64 0 1 0 128 0z\"],\"bus-simple\":[448,512,[\"bus-alt\"],\"f55e\",\"M224 0C348.8 0 448 35.2 448 80V96 416c0 17.7-14.3 32-32 32v32c0 17.7-14.3 32-32 32H352c-17.7 0-32-14.3-32-32V448H128v32c0 17.7-14.3 32-32 32H64c-17.7 0-32-14.3-32-32l0-32c-17.7 0-32-14.3-32-32V96 80C0 35.2 99.2 0 224 0zM64 128V256c0 17.7 14.3 32 32 32H352c17.7 0 32-14.3 32-32V128c0-17.7-14.3-32-32-32H96c-17.7 0-32 14.3-32 32zM80 400a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm288 0a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],eye:[576,512,[128065],\"f06e\",\"M288 32c-80.8 0-145.5 36.8-192.6 80.6C48.6 156 17.3 208 2.5 243.7c-3.3 7.9-3.3 16.7 0 24.6C17.3 304 48.6 356 95.4 399.4C142.5 443.2 207.2 480 288 480s145.5-36.8 192.6-80.6c46.8-43.5 78.1-95.4 93-131.1c3.3-7.9 3.3-16.7 0-24.6c-14.9-35.7-46.2-87.7-93-131.1C433.5 68.8 368.8 32 288 32zM144 256a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm144-64c0 35.3-28.7 64-64 64c-7.1 0-13.9-1.2-20.3-3.3c-5.5-1.8-11.9 1.6-11.7 7.4c.3 6.9 1.3 13.8 3.2 20.7c13.7 51.2 66.4 81.6 117.6 67.9s81.6-66.4 67.9-117.6c-11.1-41.5-47.8-69.4-88.6-71.1c-5.8-.2-9.2 6.1-7.4 11.7c2.1 6.4 3.3 13.2 3.3 20.3z\"],\"face-sad-cry\":[512,512,[128557,\"sad-cry\"],\"f5b3\",\"M352 493.4c-29.6 12-62.1 18.6-96 18.6s-66.4-6.6-96-18.6V288c0-8.8-7.2-16-16-16s-16 7.2-16 16V477.8C51.5 433.5 0 350.8 0 256C0 114.6 114.6 0 256 0S512 114.6 512 256c0 94.8-51.5 177.5-128 221.8V288c0-8.8-7.2-16-16-16s-16 7.2-16 16V493.4zM195.2 233.6c5.3 7.1 15.3 8.5 22.4 3.2s8.5-15.3 3.2-22.4c-30.4-40.5-91.2-40.5-121.6 0c-5.3 7.1-3.9 17.1 3.2 22.4s17.1 3.9 22.4-3.2c17.6-23.5 52.8-23.5 70.4 0zm121.6 0c17.6-23.5 52.8-23.5 70.4 0c5.3 7.1 15.3 8.5 22.4 3.2s8.5-15.3 3.2-22.4c-30.4-40.5-91.2-40.5-121.6 0c-5.3 7.1-3.9 17.1 3.2 22.4s17.1 3.9 22.4-3.2zM208 336v32c0 26.5 21.5 48 48 48s48-21.5 48-48V336c0-26.5-21.5-48-48-48s-48 21.5-48 48z\"],\"audio-description\":[576,512,[],\"f29e\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM213.5 173.3l72 144c5.9 11.9 1.1 26.3-10.7 32.2s-26.3 1.1-32.2-10.7l-9.4-18.9H150.9l-9.4 18.9c-5.9 11.9-20.3 16.7-32.2 10.7s-16.7-20.3-10.7-32.2l72-144c4.1-8.1 12.4-13.3 21.5-13.3s17.4 5.1 21.5 13.3zm-.4 106.6L192 237.7l-21.1 42.2h42.2zM304 184c0-13.3 10.7-24 24-24h56c53 0 96 43 96 96s-43 96-96 96H328c-13.3 0-24-10.7-24-24V184zm48 24v96h32c26.5 0 48-21.5 48-48s-21.5-48-48-48H352z\"],\"person-military-to-person\":[512,512,[],\"e54c\",\"M71 12.5c-8.6 1-15 8.2-15 16.8c0 9.3 7.5 16.8 16.7 16.9H184.1c8.8-.1 15.9-7.2 15.9-16V16c0-9.5-8.3-17-17.8-15.9L71 12.5zM189.5 78.1H66.5C64.9 83.8 64 89.8 64 96c0 35.3 28.7 64 64 64s64-28.7 64-64c0-6.2-.9-12.2-2.5-17.9zM32 256v32c0 17.7 14.3 32 32 32H192c1.8 0 3.5-.1 5.2-.4L53 208.6C40.1 220.3 32 237.2 32 256zm190.2 42.5c1.1-3.3 1.8-6.8 1.8-10.5V256c0-35.3-28.7-64-64-64H96c-3.7 0-7.4 .3-10.9 .9L222.2 298.5zM384 160a64 64 0 1 0 0-128 64 64 0 1 0 0 128zm-32 32c-35.3 0-64 28.7-64 64v32c0 17.7 14.3 32 32 32H448c17.7 0 32-14.3 32-32V256c0-35.3-28.7-64-64-64H352zM215.8 450.1c5.2-4.6 8.2-11.1 8.2-18.1s-3-13.5-8.2-18.1l-64-56c-7.1-6.2-17.1-7.7-25.7-3.8S112 366.6 112 376v32l-88 0c-13.3 0-24 10.7-24 24s10.7 24 24 24l88 0v32c0 9.4 5.5 18 14.1 21.9s18.6 2.4 25.7-3.8l64-56zM288 431.9c0 6.9 2.9 13.5 8.1 18.1l64 56.4c7.1 6.2 17.1 7.8 25.7 3.9s14.1-12.4 14.1-21.9l0-32.4 88 0c13.3 0 24-10.7 24-24s-10.7-24-24-24l-88 0 0-32c0-9.4-5.5-18-14.1-21.9s-18.6-2.4-25.7 3.8l-64 56c-5.2 4.5-8.2 11.1-8.2 18z\"],\"file-shield\":[576,512,[],\"e4f0\",\"M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v47l-92.8 37.1c-21.3 8.5-35.2 29.1-35.2 52c0 56.6 18.9 148 94.2 208.3c-9 4.8-19.3 7.6-30.2 7.6H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zm39.1 97.7c5.7-2.3 12.1-2.3 17.8 0l120 48C570 277.4 576 286.2 576 296c0 63.3-25.9 168.8-134.8 214.2c-5.9 2.5-12.6 2.5-18.5 0C313.9 464.8 288 359.3 288 296c0-9.8 6-18.6 15.1-22.3l120-48zM527.4 312L432 273.8V461.7c68.2-33 91.5-99 95.4-149.7z\"],\"user-slash\":[640,512,[],\"f506\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L353.3 251.6C407.9 237 448 187.2 448 128C448 57.3 390.7 0 320 0C250.2 0 193.5 55.8 192 125.2L38.8 5.1zM264.3 304.3C170.5 309.4 96 387.2 96 482.3c0 16.4 13.3 29.7 29.7 29.7H514.3c3.9 0 7.6-.7 11-2.1l-261-205.6z\"],pen:[512,512,[128394],\"f304\",\"M362.7 19.3L314.3 67.7 444.3 197.7l48.4-48.4c25-25 25-65.5 0-90.5L453.3 19.3c-25-25-65.5-25-90.5 0zm-71 71L58.6 323.5c-10.4 10.4-18 23.3-22.2 37.4L1 481.2C-1.5 489.7 .8 498.8 7 505s15.3 8.5 23.7 6.1l120.3-35.4c14.1-4.2 27-11.8 37.4-22.2L421.7 220.3 291.7 90.3z\"],\"tower-observation\":[512,512,[],\"e586\",\"M241.7 3.4c9-4.5 19.6-4.5 28.6 0l160 80c15.8 7.9 22.2 27.1 14.3 42.9C439 137.5 427.7 144 416 144v80c0 17.7-14.3 32-32 32h-4.9l32 192H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H384.5c-.4 0-.8 0-1.1 0H128.6c-.4 0-.8 0-1.1 0H32c-17.7 0-32-14.3-32-32s14.3-32 32-32h68.9l32-192H128c-17.7 0-32-14.3-32-32V144c-11.7 0-23-6.5-28.6-17.7c-7.9-15.8-1.5-35 14.3-42.9l160-80zM314.5 448L256 399.2 197.5 448h117zM197.8 256l-4.7 28.3L256 336.8l62.9-52.5L314.2 256H197.8zm-13.9 83.2l-11.2 67L218.5 368l-34.6-28.8zM293.5 368l45.8 38.1-11.2-67L293.5 368zM176 128c-8.8 0-16 7.2-16 16s7.2 16 16 16H336c8.8 0 16-7.2 16-16s-7.2-16-16-16H176z\"],\"file-code\":[384,512,[],\"f1c9\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM153 289l-31 31 31 31c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L71 337c-9.4-9.4-9.4-24.6 0-33.9l48-48c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9zM265 255l48 48c9.4 9.4 9.4 24.6 0 33.9l-48 48c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l31-31-31-31c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0z\"],signal:[640,512,[128246,\"signal-5\",\"signal-perfect\"],\"f012\",\"M576 0c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V32c0-17.7 14.3-32 32-32zM448 96c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V128c0-17.7 14.3-32 32-32zM352 224V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V224c0-17.7 14.3-32 32-32s32 14.3 32 32zM192 288c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V320c0-17.7 14.3-32 32-32zM96 416v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V416c0-17.7 14.3-32 32-32s32 14.3 32 32z\"],bus:[576,512,[128653],\"f207\",\"M288 0C422.4 0 512 35.2 512 80V96l0 32c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32l0 160c0 17.7-14.3 32-32 32v32c0 17.7-14.3 32-32 32H416c-17.7 0-32-14.3-32-32V448H192v32c0 17.7-14.3 32-32 32H128c-17.7 0-32-14.3-32-32l0-32c-17.7 0-32-14.3-32-32l0-160c-17.7 0-32-14.3-32-32V160c0-17.7 14.3-32 32-32h0V96h0V80C64 35.2 153.6 0 288 0zM128 160v96c0 17.7 14.3 32 32 32H272V128H160c-17.7 0-32 14.3-32 32zM304 288H416c17.7 0 32-14.3 32-32V160c0-17.7-14.3-32-32-32H304V288zM144 400a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm288 0a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM384 80c0-8.8-7.2-16-16-16H208c-8.8 0-16 7.2-16 16s7.2 16 16 16H368c8.8 0 16-7.2 16-16z\"],\"heart-circle-xmark\":[576,512,[],\"e501\",\"M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9l2.6-2.4C267.2 438.6 256 404.6 256 368c0-97.2 78.8-176 176-176c28.3 0 55 6.7 78.7 18.5c.9-6.5 1.3-13 1.3-19.6v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L454.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L432 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L409.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L432 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z\"],\"house-chimney\":[576,512,[63499,\"home-lg\"],\"e3af\",\"M543.8 287.6c17 0 32-14 32-32.1c1-9-3-17-11-24L512 185V64c0-17.7-14.3-32-32-32H448c-17.7 0-32 14.3-32 32v36.7L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32v69.7c-.1 .9-.1 1.8-.1 2.8V472c0 22.1 17.9 40 40 40h16c1.2 0 2.4-.1 3.6-.2c1.5 .1 3 .2 4.5 .2H160h24c22.1 0 40-17.9 40-40V448 384c0-17.7 14.3-32 32-32h64c17.7 0 32 14.3 32 32v64 24c0 22.1 17.9 40 40 40h24 32.5c1.4 0 2.8 0 4.2-.1c1.1 .1 2.2 .1 3.3 .1h16c22.1 0 40-17.9 40-40V455.8c.3-2.6 .5-5.3 .5-8.1l-.7-160.2h32z\"],\"window-maximize\":[512,512,[128470],\"f2d0\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM96 96H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H96c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],\"face-frown\":[512,512,[9785,\"frown\"],\"f119\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM159.3 388.7c-2.6 8.4-11.6 13.2-20 10.5s-13.2-11.6-10.5-20C145.2 326.1 196.3 288 256 288s110.8 38.1 127.3 91.3c2.6 8.4-2.1 17.4-10.5 20s-17.4-2.1-20-10.5C340.5 349.4 302.1 320 256 320s-84.5 29.4-96.7 68.7zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],prescription:[448,512,[],\"f5b1\",\"M32 0C14.3 0 0 14.3 0 32V192v96c0 17.7 14.3 32 32 32s32-14.3 32-32V224h50.7l128 128L137.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L288 397.3 393.4 502.6c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L333.3 352 438.6 246.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L288 306.7l-85.8-85.8C251.4 209.1 288 164.8 288 112C288 50.1 237.9 0 176 0H32zM176 160H64V64H176c26.5 0 48 21.5 48 48s-21.5 48-48 48z\"],shop:[640,512,[\"store-alt\"],\"f54f\",\"M36.8 192H603.2c20.3 0 36.8-16.5 36.8-36.8c0-7.3-2.2-14.4-6.2-20.4L558.2 21.4C549.3 8 534.4 0 518.3 0H121.7c-16 0-31 8-39.9 21.4L6.2 134.7c-4 6.1-6.2 13.2-6.2 20.4C0 175.5 16.5 192 36.8 192zM64 224V384v80c0 26.5 21.5 48 48 48H336c26.5 0 48-21.5 48-48V384 224H320V384H128V224H64zm448 0V480c0 17.7 14.3 32 32 32s32-14.3 32-32V224H512z\"],\"floppy-disk\":[448,512,[128190,128426,\"save\"],\"f0c7\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V173.3c0-17-6.7-33.3-18.7-45.3L352 50.7C340 38.7 323.7 32 306.7 32H64zm0 96c0-17.7 14.3-32 32-32H288c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V128zM224 288a64 64 0 1 1 0 128 64 64 0 1 1 0-128z\"],vihara:[640,512,[],\"f6a7\",\"M281 22L305.8 4.7c1.3-.9 2.7-1.8 4.1-2.4C313.1 .7 316.6 0 320 0s6.9 .7 10.1 2.2c1.4 .7 2.8 1.5 4.1 2.4L359 22C393 45.8 430.8 63.5 470.8 74.4l23 6.3c1.8 .5 3.6 1.1 5.2 2c3.2 1.7 5.9 4 8.1 6.8c3.8 4.9 5.6 11.3 4.7 17.8c-.4 2.8-1.2 5.4-2.5 7.8c-1.7 3.2-4 5.9-6.8 8.1c-4.3 3.2-9.6 5.1-15.1 4.9H480v56.1l6.4 5.1 5.2 4.1c21.1 16.7 45 29.6 70.5 38.1l28.9 9.6c1.6 .5 3.2 1.2 4.6 2c3.1 1.7 5.8 4.1 7.8 6.9s3.5 6.1 4.1 9.6c.5 2.7 .6 5.5 .1 8.3s-1.4 5.4-2.7 7.8c-1.7 3.1-4.1 5.8-6.9 7.8s-6.1 3.5-9.6 4.1c-1.6 .3-3.3 .4-5 .4H544v65.9c20.5 22.8 47.4 39.2 77.4 46.7C632 403 640 412.6 640 424c0 13.3-10.7 24-24 24H576v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H352v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H128v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H24c-13.3 0-24-10.7-24-24c0-11.4 8-21 18.6-23.4c30-7.6 56.9-23.9 77.4-46.7V288H56.6c-1.7 0-3.4-.1-5-.4c-3.5-.7-6.8-2.1-9.6-4.1s-5.2-4.7-7-7.8c-1.3-2.4-2.3-5-2.7-7.8s-.4-5.6 .1-8.3c.7-3.5 2.1-6.8 4.1-9.6s4.7-5.2 7.8-6.9c1.4-.8 3-1.5 4.6-2l28.9-9.6c25.5-8.5 49.4-21.4 70.5-38.1l5.2-4.1 6.4-5.1V176 128h-7.5c-5.5 .1-10.8-1.7-15.1-4.9c-2.8-2.1-5.1-4.8-6.8-8.1c-1.2-2.4-2.1-5-2.5-7.8c-.9-6.5 .9-12.8 4.7-17.8c2.1-2.8 4.8-5.1 8.1-6.8c1.6-.8 3.4-1.5 5.2-2l23-6.3C209.2 63.5 247 45.8 281 22zM416 128H320 224v64h72 48 72V128zM160 288v64H296h24 24H480V288H344 320h0H296 160z\"],\"scale-unbalanced\":[640,512,[\"balance-scale-left\"],\"f515\",\"M522.1 62.4c16.8-5.6 25.8-23.7 20.2-40.5S518.6-3.9 501.9 1.6l-113 37.7C375 15.8 349.3 0 320 0c-44.2 0-80 35.8-80 80c0 3 .2 5.9 .5 8.8L117.9 129.6c-16.8 5.6-25.8 23.7-20.2 40.5s23.7 25.8 40.5 20.2l135.5-45.2c4.5 3.2 9.3 5.9 14.4 8.2V480c0 17.7 14.3 32 32 32H512c17.7 0 32-14.3 32-32s-14.3-32-32-32H352V153.3c21-9.2 37.2-27 44.2-49l125.9-42zM439.6 288L512 163.8 584.4 288H439.6zM512 384c62.9 0 115.2-34 126-78.9c2.6-11-1-22.3-6.7-32.1L536.1 109.8c-5-8.6-14.2-13.8-24.1-13.8s-19.1 5.3-24.1 13.8L392.7 273.1c-5.7 9.8-9.3 21.1-6.7 32.1C396.8 350 449.1 384 512 384zM129.2 291.8L201.6 416H56.7l72.4-124.2zM3.2 433.1C14 478 66.3 512 129.2 512s115.2-34 126-78.9c2.6-11-1-22.3-6.7-32.1L153.2 237.8c-5-8.6-14.2-13.8-24.1-13.8s-19.1 5.3-24.1 13.8L9.9 401.1c-5.7 9.8-9.3 21.1-6.7 32.1z\"],\"sort-up\":[320,512,[\"sort-asc\"],\"f0de\",\"M182.6 41.4c-12.5-12.5-32.8-12.5-45.3 0l-128 128c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8H288c12.9 0 24.6-7.8 29.6-19.8s2.2-25.7-6.9-34.9l-128-128z\"],\"comment-dots\":[512,512,[128172,62075,\"commenting\"],\"f4ad\",\"M256 448c141.4 0 256-93.1 256-208S397.4 32 256 32S0 125.1 0 240c0 45.1 17.7 86.8 47.7 120.9c-1.9 24.5-11.4 46.3-21.4 62.9c-5.5 9.2-11.1 16.6-15.2 21.6c-2.1 2.5-3.7 4.4-4.9 5.7c-.6 .6-1 1.1-1.3 1.4l-.3 .3 0 0 0 0 0 0 0 0c-4.6 4.6-5.9 11.4-3.4 17.4c2.5 6 8.3 9.9 14.8 9.9c28.7 0 57.6-8.9 81.6-19.3c22.9-10 42.4-21.9 54.3-30.6c31.8 11.5 67 17.9 104.1 17.9zM128 208a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 0a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm96 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],\"plant-wilt\":[512,512,[],\"e5aa\",\"M288 120c0-30.9 25.1-56 56-56s56 25.1 56 56v13c-29.3 10-48 34.5-48 70.1c0 27.9 25.3 74.8 66 111.6c3.8 3.5 8.9 5.3 14 5.3s10.2-1.8 14-5.3c40.7-36.8 66-83.7 66-111.6c0-35.6-18.7-60.2-48-70.1V120C464 53.7 410.3 0 344 0S224 53.7 224 120v21.8C207.3 133 188.2 128 168 128c-66.3 0-120 53.7-120 120v13c-29.3 10-48 34.5-48 70.1C0 359 25.3 405.9 66 442.7c3.8 3.5 8.9 5.3 14 5.3s10.2-1.8 14-5.3c40.7-36.8 66-83.7 66-111.6c0-35.6-18.7-60.2-48-70.1V248c0-30.9 25.1-56 56-56s56 25.1 56 56v32V480c0 17.7 14.3 32 32 32s32-14.3 32-32V280 248 120z\"],diamond:[512,512,[9830],\"f219\",\"M284.3 11.7c-15.6-15.6-40.9-15.6-56.6 0l-216 216c-15.6 15.6-15.6 40.9 0 56.6l216 216c15.6 15.6 40.9 15.6 56.6 0l216-216c15.6-15.6 15.6-40.9 0-56.6l-216-216z\"],\"face-grin-squint\":[512,512,[128518,\"grin-squint\"],\"f585\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM388.1 312.8c12.3-3.8 24.3 6.9 19.3 18.7C382.4 390.6 324.2 432 256.3 432s-126.2-41.4-151.1-100.5c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19zM133.5 146.7l89.9 47.9c10.7 5.7 10.7 21.1 0 26.8l-89.9 47.9c-7.9 4.2-17.5-1.5-17.5-10.5c0-2.8 1-5.5 2.8-7.6l36-43.2-36-43.2c-1.8-2.1-2.8-4.8-2.8-7.6c0-9 9.6-14.7 17.5-10.5zM396 157.1c0 2.8-1 5.5-2.8 7.6l-36 43.2 36 43.2c1.8 2.1 2.8 4.8 2.8 7.6c0 9-9.6 14.7-17.5 10.5l-89.9-47.9c-10.7-5.7-10.7-21.1 0-26.8l89.9-47.9c7.9-4.2 17.5 1.5 17.5 10.5z\"],\"hand-holding-dollar\":[576,512,[\"hand-holding-usd\"],\"f4c0\",\"M312 24V34.5c6.4 1.2 12.6 2.7 18.2 4.2c12.8 3.4 20.4 16.6 17 29.4s-16.6 20.4-29.4 17c-10.9-2.9-21.1-4.9-30.2-5c-7.3-.1-14.7 1.7-19.4 4.4c-2.1 1.3-3.1 2.4-3.5 3c-.3 .5-.7 1.2-.7 2.8c0 .3 0 .5 0 .6c.2 .2 .9 1.2 3.3 2.6c5.8 3.5 14.4 6.2 27.4 10.1l.9 .3c11.1 3.3 25.9 7.8 37.9 15.3c13.7 8.6 26.1 22.9 26.4 44.9c.3 22.5-11.4 38.9-26.7 48.5c-6.7 4.1-13.9 7-21.3 8.8V232c0 13.3-10.7 24-24 24s-24-10.7-24-24V220.6c-9.5-2.3-18.2-5.3-25.6-7.8c-2.1-.7-4.1-1.4-6-2c-12.6-4.2-19.4-17.8-15.2-30.4s17.8-19.4 30.4-15.2c2.6 .9 5 1.7 7.3 2.5c13.6 4.6 23.4 7.9 33.9 8.3c8 .3 15.1-1.6 19.2-4.1c1.9-1.2 2.8-2.2 3.2-2.9c.4-.6 .9-1.8 .8-4.1l0-.2c0-1 0-2.1-4-4.6c-5.7-3.6-14.3-6.4-27.1-10.3l-1.9-.6c-10.8-3.2-25-7.5-36.4-14.4c-13.5-8.1-26.5-22-26.6-44.1c-.1-22.9 12.9-38.6 27.7-47.4c6.4-3.8 13.3-6.4 20.2-8.2V24c0-13.3 10.7-24 24-24s24 10.7 24 24zM568.2 336.3c13.1 17.8 9.3 42.8-8.5 55.9L433.1 485.5c-23.4 17.2-51.6 26.5-80.7 26.5H192 32c-17.7 0-32-14.3-32-32V416c0-17.7 14.3-32 32-32H68.8l44.9-36c22.7-18.2 50.9-28 80-28H272h16 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H288 272c-8.8 0-16 7.2-16 16s7.2 16 16 16H392.6l119.7-88.2c17.8-13.1 42.8-9.3 55.9 8.5zM193.6 384l0 0-.9 0c.3 0 .6 0 .9 0z\"],bacterium:[512,512,[],\"e05a\",\"M423.1 30.6c3.6-12.7-3.7-26-16.5-29.7s-26 3.7-29.7 16.5l-4.2 14.7c-9.8-.4-19.9 .5-29.9 2.8c-12.1 2.8-23.7 5.9-34.9 9.4l-5.9-13.7c-5.2-12.2-19.3-17.8-31.5-12.6s-17.8 19.3-12.6 31.5l4.9 11.3c-22 9.4-42 20.1-60.2 31.8L196 82.7c-7.4-11-22.3-14-33.3-6.7s-14 22.3-6.7 33.3l7.8 11.6c-18 15-33.7 30.8-47.3 47.1L103 157.3c-10.4-8.3-25.5-6.6-33.7 3.7s-6.6 25.5 3.7 33.7l15 12c-2.1 3.2-4.1 6.5-6 9.7c-9.4 15.7-17 31-23.2 45.3l-9.9-3.9c-12.3-4.9-26.3 1.1-31.2 13.4s1.1 26.3 13.4 31.2l11.6 4.6c-.3 1.1-.6 2.1-.9 3.1c-3.5 12.5-5.7 23.2-7.1 31.3c-.7 4.1-1.2 7.5-1.6 10.3c-.2 1.4-.3 2.6-.4 3.6l-.1 1.4-.1 .6 0 .3 0 .1c0 0 0 .1 39.2 3.7l0 0-39.2-3.6c-.5 5-.6 10-.4 14.9l-14.7 4.2C4.7 380.6-2.7 393.8 .9 406.6s16.9 20.1 29.7 16.5l13.8-3.9c10.6 20.7 27.6 37.8 48.5 48.5l-3.9 13.7c-3.6 12.7 3.7 26 16.5 29.7s26-3.7 29.7-16.5l4.2-14.7c23.8 1 46.3-5.5 65.1-17.6L215 473c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-10.6-10.6c9.1-14.1 15.1-30.5 17-48.3l.1-.8c.3-1.7 1-5.1 2.3-9.8l.2-.8 12.6 5.4c12.2 5.2 26.3-.4 31.5-12.6s-.4-26.3-12.6-31.5l-11.3-4.8c9.9-14.9 24.9-31.6 48.6-46l2.1 7.5c3.6 12.7 16.9 20.1 29.7 16.5s20.1-16.9 16.5-29.7L371 259.2c6.9-2.2 14.3-4.3 22.2-6.1c12.9-3 24.7-8 35.2-14.8L439 249c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-10.6-10.6c12.2-19 18.6-41.6 17.6-65.1l14.7-4.2c12.7-3.6 20.1-16.9 16.5-29.7s-16.9-20.1-29.7-16.5l-13.7 3.9c-10.8-21.2-28-38-48.5-48.5l3.9-13.8zM92.1 363.3l0 0L144 368l-51.9-4.7zM112 320a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM240 184a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],\"hand-pointer\":[448,512,[],\"f25a\",\"M128 40c0-22.1 17.9-40 40-40s40 17.9 40 40V188.2c8.5-7.6 19.7-12.2 32-12.2c20.6 0 38.2 13 45 31.2c8.8-9.3 21.2-15.2 35-15.2c25.3 0 46 19.5 47.9 44.3c8.5-7.7 19.8-12.3 32.1-12.3c26.5 0 48 21.5 48 48v48 16 48c0 70.7-57.3 128-128 128l-16 0H240l-.1 0h-5.2c-5 0-9.9-.3-14.7-1c-55.3-5.6-106.2-34-140-79L8 336c-13.3-17.7-9.7-42.7 8-56s42.7-9.7 56 8l56 74.7V40zM240 304c0-8.8-7.2-16-16-16s-16 7.2-16 16v96c0 8.8 7.2 16 16 16s16-7.2 16-16V304zm48-16c-8.8 0-16 7.2-16 16v96c0 8.8 7.2 16 16 16s16-7.2 16-16V304c0-8.8-7.2-16-16-16zm80 16c0-8.8-7.2-16-16-16s-16 7.2-16 16v96c0 8.8 7.2 16 16 16s16-7.2 16-16V304z\"],\"drum-steelpan\":[576,512,[],\"f56a\",\"M288 32c159.1 0 288 48 288 128V352c0 80-128.9 128-288 128S0 432 0 352V160C0 80 128.9 32 288 32zM528 160c0-9.9-8-29.9-55-49.8c-18.6-7.9-40.9-14.4-66-19.4l-27.8 43.6c-7.3 11.5-11.2 24.8-11.2 38.4c0 17.5 6.4 34.4 18.1 47.5l9.8 11c29.8-5.2 55.9-12.5 77.2-21.5c47.1-19.9 55-39.9 55-49.8zM349.2 237.3c-8-26.2-32.4-45.3-61.2-45.3s-53.3 19.1-61.2 45.3c19.4 1.7 39.9 2.7 61.2 2.7s41.8-.9 61.2-2.7zM169 90.8c-25.2 5-47.4 11.6-66 19.4C56 130.1 48 150.1 48 160s8 29.9 55 49.8c21.3 9 47.4 16.3 77.2 21.5l9.8-11c11.6-13.1 18.1-30 18.1-47.5c0-13.6-3.9-26.9-11.2-38.4L169 90.8zm56.3-8C224.5 87 224 91.5 224 96c0 35.3 28.7 64 64 64s64-28.7 64-64c0-4.5-.5-9-1.4-13.2C330.8 81 309.8 80 288 80s-42.8 1-62.6 2.8z\"],\"hand-scissors\":[512,512,[],\"f257\",\"M40 208c-22.1 0-40 17.9-40 40s17.9 40 40 40l180.2 0c-7.6 8.5-12.2 19.7-12.2 32c0 25.3 19.5 46 44.3 47.9c-7.7 8.5-12.3 19.8-12.3 32.1c0 26.5 21.5 48 48 48l32 0 64 0c70.7 0 128-57.3 128-128l0-113.1c0-40.2-16-78.8-44.4-107.3C444.8 76.8 413.9 64 381.7 64L336 64c-21.3 0-39.3 13.9-45.6 33.1l74.5 23.7c8.4 2.7 13.1 11.7 10.4 20.1s-11.7 13.1-20.1 10.4L288 129.9l0 .1L84 65.8C62.9 59.2 40.5 70.9 33.8 92s5.1 43.5 26.2 50.2L269.5 208 40 208z\"],\"hands-praying\":[640,512,[\"praying-hands\"],\"f684\",\"M351.2 4.8c3.2-2 6.6-3.3 10-4.1c4.7-1 9.6-.9 14.1 .1c7.7 1.8 14.8 6.5 19.4 13.6L514.6 194.2c8.8 13.1 13.4 28.6 13.4 44.4v73.5c0 6.9 4.4 13 10.9 15.2l79.2 26.4C631.2 358 640 370.2 640 384v96c0 9.9-4.6 19.3-12.5 25.4s-18.1 8.1-27.7 5.5L431 465.9c-56-14.9-95-65.7-95-123.7V224c0-17.7 14.3-32 32-32s32 14.3 32 32v80c0 8.8 7.2 16 16 16s16-7.2 16-16V219.1c0-7-1.8-13.8-5.3-19.8L340.3 48.1c-1.7-3-2.9-6.1-3.6-9.3c-1-4.7-1-9.6 .1-14.1c1.9-8 6.8-15.2 14.3-19.9zm-62.4 0c7.5 4.6 12.4 11.9 14.3 19.9c1.1 4.6 1.2 9.4 .1 14.1c-.7 3.2-1.9 6.3-3.6 9.3L213.3 199.3c-3.5 6-5.3 12.9-5.3 19.8V304c0 8.8 7.2 16 16 16s16-7.2 16-16V224c0-17.7 14.3-32 32-32s32 14.3 32 32V342.3c0 58-39 108.7-95 123.7l-168.7 45c-9.6 2.6-19.9 .5-27.7-5.5S0 490 0 480V384c0-13.8 8.8-26 21.9-30.4l79.2-26.4c6.5-2.2 10.9-8.3 10.9-15.2V238.5c0-15.8 4.7-31.2 13.4-44.4L245.2 14.5c4.6-7.1 11.7-11.8 19.4-13.6c4.6-1.1 9.4-1.2 14.1-.1c3.5 .8 6.9 2.1 10 4.1z\"],\"arrow-rotate-right\":[512,512,[8635,\"arrow-right-rotate\",\"arrow-rotate-forward\",\"redo\"],\"f01e\",\"M386.3 160H336c-17.7 0-32 14.3-32 32s14.3 32 32 32H464c17.7 0 32-14.3 32-32V64c0-17.7-14.3-32-32-32s-32 14.3-32 32v51.2L414.4 97.6c-87.5-87.5-229.3-87.5-316.8 0s-87.5 229.3 0 316.8s229.3 87.5 316.8 0c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0c-62.5 62.5-163.8 62.5-226.3 0s-62.5-163.8 0-226.3s163.8-62.5 226.3 0L386.3 160z\"],biohazard:[576,512,[9763],\"f780\",\"M173.2 0c-1.8 0-3.5 .7-4.8 2C138.5 32.3 120 74 120 120c0 26.2 6 50.9 16.6 73c-22 2.4-43.8 9.1-64.2 20.5C37.9 232.8 13.3 262.4 .4 296c-.7 1.7-.5 3.7 .5 5.2c2.2 3.7 7.4 4.3 10.6 1.3C64.2 254.3 158 245.1 205 324s-8.1 153.1-77.6 173.2c-4.2 1.2-6.3 5.9-4.1 9.6c1 1.6 2.6 2.7 4.5 3c36.5 5.9 75.2 .1 109.7-19.2c20.4-11.4 37.4-26.5 50.5-43.8c13.1 17.3 30.1 32.4 50.5 43.8c34.5 19.3 73.3 25.2 109.7 19.2c1.9-.3 3.5-1.4 4.5-3c2.2-3.7 .1-8.4-4.1-9.6C379.1 477.1 324 403 371 324s140.7-69.8 193.5-21.4c3.2 2.9 8.4 2.3 10.6-1.3c1-1.6 1.1-3.5 .5-5.2c-12.9-33.6-37.5-63.2-72.1-82.5c-20.4-11.4-42.2-18.1-64.2-20.5C450 170.9 456 146.2 456 120c0-46-18.5-87.7-48.4-118c-1.3-1.3-3-2-4.8-2c-5 0-8.4 5.2-6.7 9.9C421.7 80.5 385.6 176 288 176S154.3 80.5 179.9 9.9c1.7-4.7-1.6-9.9-6.7-9.9zM240 272a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM181.7 417.6c6.3-11.8 9.8-25.1 8.6-39.8c-19.5-18-34-41.4-41.2-67.8c-12.5-8.1-26.2-11.8-40-12.4c-9-.4-18.1 .6-27.1 2.7c7.8 57.1 38.7 106.8 82.9 139.4c6.8-6.7 12.6-14.1 16.8-22.1zM288 64c-28.8 0-56.3 5.9-81.2 16.5c2 8.3 5 16.2 9 23.5c6.8 12.4 16.7 23.1 30.1 30.3c13.3-4.1 27.5-6.3 42.2-6.3s28.8 2.2 42.2 6.3c13.4-7.2 23.3-17.9 30.1-30.3c4-7.3 7-15.2 9-23.5C344.3 69.9 316.8 64 288 64zM426.9 310c-7.2 26.4-21.7 49.7-41.2 67.8c-1.2 14.7 2.2 28.1 8.6 39.8c4.3 8 10 15.4 16.8 22.1c44.3-32.6 75.2-82.3 82.9-139.4c-9-2.2-18.1-3.1-27.1-2.7c-13.8 .6-27.5 4.4-40 12.4z\"],\"location-crosshairs\":[512,512,[\"location\"],\"f601\",\"M256 0c17.7 0 32 14.3 32 32V66.7C368.4 80.1 431.9 143.6 445.3 224H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H445.3C431.9 368.4 368.4 431.9 288 445.3V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V445.3C143.6 431.9 80.1 368.4 66.7 288H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H66.7C80.1 143.6 143.6 80.1 224 66.7V32c0-17.7 14.3-32 32-32zM128 256a128 128 0 1 0 256 0 128 128 0 1 0 -256 0zm128-80a80 80 0 1 1 0 160 80 80 0 1 1 0-160z\"],\"mars-double\":[640,512,[9891],\"f227\",\"M312 32c-9.7 0-18.5 5.8-22.2 14.8s-1.7 19.3 5.2 26.2l33.4 33.4L275.8 159c-28.4-19.5-62.7-31-99.8-31C78.8 128 0 206.8 0 304s78.8 176 176 176s176-78.8 176-176c0-37-11.4-71.4-31-99.8l52.6-52.6L407 185c6.9 6.9 17.2 8.9 26.2 5.2s14.8-12.5 14.8-22.2V56c0-13.3-10.7-24-24-24H312zm88 48h0v0l0 0zM64 304a112 112 0 1 1 224 0A112 112 0 1 1 64 304zM368 480c97.2 0 176-78.8 176-176c0-37-11.4-71.4-31-99.8l52.6-52.6L599 185c6.9 6.9 17.2 8.9 26.2 5.2s14.8-12.5 14.8-22.2V56c0-13.3-10.7-24-24-24H504c-9.7 0-18.5 5.8-22.2 14.8c-1.2 2.9-1.8 6-1.8 9l0 .2v.2c0 6.2 2.5 12.2 7 16.8l33.4 33.4L480 146.7V168c0 22.6-13.6 43.1-34.6 51.7c-.8 .3-1.7 .7-2.5 1C465.7 241.2 480 270.9 480 304c0 61.9-50.1 112-112 112c-5.4 0-10.8-.4-16-1.1c-12.9 20.4-29.1 38.3-48.1 53.1c19.8 7.8 41.4 12 64 12z\"],\"child-dress\":[320,512,[],\"e59c\",\"M224 64A64 64 0 1 0 96 64a64 64 0 1 0 128 0zM88 400v80c0 17.7 14.3 32 32 32s32-14.3 32-32V400h16v80c0 17.7 14.3 32 32 32s32-14.3 32-32V400h17.8c10.9 0 18.6-10.7 15.2-21.1l-31.1-93.4 28.6 37.8c10.7 14.1 30.8 16.8 44.8 6.2s16.8-30.7 6.2-44.8L254.6 207c-22.4-29.6-57.5-47-94.6-47s-72.2 17.4-94.6 47L6.5 284.7c-10.7 14.1-7.9 34.2 6.2 44.8s34.2 7.9 44.8-6.2l28.7-37.8L55 378.9C51.6 389.3 59.3 400 70.2 400H88z\"],\"users-between-lines\":[640,512,[],\"e591\",\"M0 24C0 10.7 10.7 0 24 0H616c13.3 0 24 10.7 24 24s-10.7 24-24 24H24C10.7 48 0 37.3 0 24zM0 488c0-13.3 10.7-24 24-24H616c13.3 0 24 10.7 24 24s-10.7 24-24 24H24c-13.3 0-24-10.7-24-24zM83.2 160a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM32 320c0-35.3 28.7-64 64-64h96c12.2 0 23.7 3.4 33.4 9.4c-37.2 15.1-65.6 47.2-75.8 86.6H64c-17.7 0-32-14.3-32-32zm461.6 32c-10.3-40.1-39.6-72.6-77.7-87.4c9.4-5.5 20.4-8.6 32.1-8.6h96c35.3 0 64 28.7 64 64c0 17.7-14.3 32-32 32H493.6zM391.2 290.4c32.1 7.4 58.1 30.9 68.9 61.6c3.5 10 5.5 20.8 5.5 32c0 17.7-14.3 32-32 32h-224c-17.7 0-32-14.3-32-32c0-11.2 1.9-22 5.5-32c10.5-29.7 35.3-52.8 66.1-60.9c7.8-2.1 16-3.1 24.5-3.1h96c7.4 0 14.7 .8 21.6 2.4zm44-130.4a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM321.6 96a80 80 0 1 1 0 160 80 80 0 1 1 0-160z\"],\"lungs-virus\":[640,512,[],\"e067\",\"M320 0c17.7 0 32 14.3 32 32V156.2c-8.5-7.6-19.7-12.2-32-12.2s-23.5 4.6-32 12.2V32c0-17.7 14.3-32 32-32zM444.5 195.5c-16.4-16.4-41.8-18.5-60.5-6.1V165.3C384 127 415 96 453.3 96c21.7 0 42.8 10.2 55.8 28.8c15.4 22.1 44.3 65.4 71 116.9c26.5 50.9 52.4 112.5 59.6 170.3c.2 1.3 .2 2.6 .2 4v7c0 49.1-39.8 89-89 89c-7.3 0-14.5-.9-21.6-2.7l-72.7-18.2c-20.9-5.2-38.7-17.1-51.5-32.9c14 1.5 28.5-3 39.2-13.8l-22.6-22.6 22.6 22.6c18.7-18.7 18.7-49.1 0-67.9c-1.1-1.1-1.4-2-1.5-2.5c-.1-.8-.1-1.8 .4-2.9s1.2-1.9 1.8-2.3c.5-.3 1.3-.8 2.9-.8c26.5 0 48-21.5 48-48s-21.5-48-48-48c-1.6 0-2.4-.4-2.9-.8c-.6-.4-1.3-1.2-1.8-2.3s-.5-2.2-.4-2.9c.1-.6 .4-1.4 1.5-2.5c18.7-18.7 18.7-49.1 0-67.9zM421.8 421.8c-6.2 6.2-16.4 6.2-22.6 0C375.9 398.5 336 415 336 448c0 8.8-7.2 16-16 16s-16-7.2-16-16c0-33-39.9-49.5-63.2-26.2c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6C241.5 375.9 225 336 192 336c-8.8 0-16-7.2-16-16s7.2-16 16-16c33 0 49.5-39.9 26.2-63.2c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0C264.1 241.5 304 225 304 192c0-8.8 7.2-16 16-16s16 7.2 16 16c0 33 39.9 49.5 63.2 26.2c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6C398.5 264.1 415 304 448 304c8.8 0 16 7.2 16 16s-7.2 16-16 16c-33 0-49.5 39.9-26.2 63.2c6.2 6.2 6.2 16.4 0 22.6zM183.3 491.2l-72.7 18.2c-7.1 1.8-14.3 2.7-21.6 2.7c-49.1 0-89-39.8-89-89v-7c0-1.3 .1-2.7 .2-4c7.2-57.9 33.1-119.4 59.6-170.3c26.8-51.5 55.6-94.8 71-116.9c13-18.6 34-28.8 55.8-28.8C225 96 256 127 256 165.3v24.1c-18.6-12.4-44-10.3-60.5 6.1c-18.7 18.7-18.7 49.1 0 67.9c1.1 1.1 1.4 2 1.5 2.5c.1 .8 .1 1.8-.4 2.9s-1.2 1.9-1.8 2.3c-.5 .3-1.3 .8-2.9 .8c-26.5 0-48 21.5-48 48s21.5 48 48 48c1.6 0 2.4 .4 2.9 .8c.6 .4 1.3 1.2 1.8 2.3s.5 2.2 .4 2.9c-.1 .6-.4 1.4-1.5 2.5c-18.7 18.7-18.7 49.1 0 67.9c10.7 10.7 25.3 15.3 39.2 13.8c-12.8 15.9-30.6 27.7-51.5 32.9zM296 320a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm72 32a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z\"],\"face-grin-tears\":[640,512,[128514,\"grin-tears\"],\"f588\",\"M548.6 371.4C506.4 454.8 419.9 512 320 512s-186.4-57.2-228.6-140.6c4.5-2.9 8.7-6.3 12.7-10.3c8.1-8.1 13.2-18.6 16.5-26.6c3.6-8.8 6.5-18.4 8.8-27.5c4.6-18.2 7.7-37 9.3-48.2c3.9-26.5-18.8-49.2-45.2-45.4c-6.8 .9-16.2 2.4-26.6 4.4C85.3 94.5 191.6 0 320 0S554.7 94.5 573.2 217.7c-10.3-2-19.8-3.5-26.6-4.4c-26.5-3.9-49.2 18.8-45.2 45.4c1.6 11.3 4.6 30 9.3 48.2c2.3 9.1 5.2 18.8 8.8 27.5c3.3 8.1 8.4 18.5 16.5 26.6c3.9 3.9 8.2 7.4 12.7 10.3zM107 254.1c-3.1 21.5-11.4 70.2-25.5 84.4c-.9 1-1.9 1.8-2.9 2.7C60 356.7 32 355.5 14.3 337.7c-18.7-18.7-19.1-48.8-.7-67.2c8.6-8.6 30.1-15.1 50.5-19.6c13-2.8 25.5-4.8 33.9-6c5.4-.8 9.9 3.7 9 9zm454.5 87.1c-.8-.6-1.5-1.3-2.3-2c-.2-.2-.5-.4-.7-.7c-14.1-14.1-22.5-62.9-25.5-84.4c-.8-5.4 3.7-9.9 9-9c1 .1 2.2 .3 3.3 .5c8.2 1.2 19.2 3 30.6 5.5c20.4 4.4 41.9 10.9 50.5 19.6c18.4 18.4 18 48.5-.7 67.2c-17.7 17.7-45.7 19-64.2 3.4zm-90.1-9.7c5-11.8-7-22.5-19.3-18.7c-39.7 12.2-84.5 19-131.8 19s-92.1-6.8-131.8-19c-12.3-3.8-24.3 6.9-19.3 18.7c25 59.1 83.2 100.5 151.1 100.5s126.2-41.4 151.1-100.5zM281.6 228.8l0 0 0 0 0 0c2.1 2.8 5.7 3.9 8.9 2.8s5.5-4.1 5.5-7.6c0-17.9-6.7-35.6-16.6-48.8c-9.8-13-23.9-23.2-39.4-23.2s-29.6 10.2-39.4 23.2C190.7 188.4 184 206.1 184 224c0 3.4 2.2 6.5 5.5 7.6s6.9 0 8.9-2.8l0 0 0 0 0 0 .2-.2c.2-.2 .4-.5 .7-.9c.6-.8 1.6-2 2.8-3.4c2.5-2.8 6-6.6 10.2-10.3c8.8-7.8 18.8-14 27.7-14s18.9 6.2 27.7 14c4.2 3.7 7.7 7.5 10.2 10.3c1.2 1.4 2.2 2.6 2.8 3.4c.3 .4 .6 .7 .7 .9l.2 .2 0 0zm160 0l0 0 0 0c2.1 2.8 5.7 3.9 8.9 2.8s5.5-4.1 5.5-7.6c0-17.9-6.7-35.6-16.6-48.8c-9.8-13-23.9-23.2-39.4-23.2s-29.6 10.2-39.4 23.2C350.7 188.4 344 206.1 344 224c0 3.4 2.2 6.5 5.5 7.6s6.9 0 8.9-2.8l0 0 0 0 0 0 .2-.2c.2-.2 .4-.5 .7-.9c.6-.8 1.6-2 2.8-3.4c2.5-2.8 6-6.6 10.2-10.3c8.8-7.8 18.8-14 27.7-14s18.9 6.2 27.7 14c4.2 3.7 7.7 7.5 10.2 10.3c1.2 1.4 2.2 2.6 2.8 3.4c.3 .4 .6 .7 .7 .9l.2 .2 0 0 0 0z\"],phone:[512,512,[128222,128379],\"f095\",\"M164.9 24.6c-7.7-18.6-28-28.5-47.4-23.2l-88 24C12.1 30.2 0 46 0 64C0 311.4 200.6 512 448 512c18 0 33.8-12.1 38.6-29.5l24-88c5.3-19.4-4.6-39.7-23.2-47.4l-96-40c-16.3-6.8-35.2-2.1-46.3 11.6L304.7 368C234.3 334.7 177.3 277.7 144 207.3L193.3 167c13.7-11.2 18.4-30 11.6-46.3l-40-96z\"],\"calendar-xmark\":[448,512,[\"calendar-times\"],\"f273\",\"M128 0c17.7 0 32 14.3 32 32V64H288V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64h48c26.5 0 48 21.5 48 48v48H0V112C0 85.5 21.5 64 48 64H96V32c0-17.7 14.3-32 32-32zM0 192H448V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V192zM305 305c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-47 47-47-47c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l47 47-47 47c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l47-47 47 47c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-47-47 47-47z\"],\"child-reaching\":[384,512,[],\"e59d\",\"M256 64A64 64 0 1 0 128 64a64 64 0 1 0 128 0zM152.9 169.3c-23.7-8.4-44.5-24.3-58.8-45.8L74.6 94.2C64.8 79.5 45 75.6 30.2 85.4s-18.7 29.7-8.9 44.4L40.9 159c18.1 27.1 42.8 48.4 71.1 62.4V480c0 17.7 14.3 32 32 32s32-14.3 32-32V384h32v96c0 17.7 14.3 32 32 32s32-14.3 32-32V221.6c29.1-14.2 54.4-36.2 72.7-64.2l18.2-27.9c9.6-14.8 5.4-34.6-9.4-44.3s-34.6-5.5-44.3 9.4L291 122.4c-21.8 33.4-58.9 53.6-98.8 53.6c-12.6 0-24.9-2-36.6-5.8c-.9-.3-1.8-.7-2.7-.9z\"],\"head-side-virus\":[512,512,[],\"e064\",\"M0 224.2C0 100.6 100.2 0 224 0h24c95.2 0 181.2 69.3 197.3 160.2c2.3 13 6.8 25.7 15.1 36l42 52.6c6.2 7.8 9.6 17.4 9.6 27.4c0 24.2-19.6 43.8-43.8 43.8H448v64c0 35.3-28.7 64-64 64H320v32c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V407.3c0-16.7-6.9-32.5-17.1-45.8C16.6 322.4 0 274.1 0 224.2zM224 64c-8.8 0-16 7.2-16 16c0 33-39.9 49.5-63.2 26.2c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6C145.5 152.1 129 192 96 192c-8.8 0-16 7.2-16 16s7.2 16 16 16c33 0 49.5 39.9 26.2 63.2c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0C168.1 286.5 208 303 208 336c0 8.8 7.2 16 16 16s16-7.2 16-16c0-33 39.9-49.5 63.2-26.2c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6C302.5 263.9 319 224 352 224c8.8 0 16-7.2 16-16s-7.2-16-16-16c-33 0-49.5-39.9-26.2-63.2c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0C279.9 129.5 240 113 240 80c0-8.8-7.2-16-16-16zm-24 96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm40 80a16 16 0 1 1 32 0 16 16 0 1 1 -32 0z\"],\"user-gear\":[640,512,[\"user-cog\"],\"f4fe\",\"M224 0a128 128 0 1 1 0 256A128 128 0 1 1 224 0zM178.3 304h91.4c11.8 0 23.4 1.2 34.5 3.3c-2.1 18.5 7.4 35.6 21.8 44.8c-16.6 10.6-26.7 31.6-20 53.3c4 12.9 9.4 25.5 16.4 37.6s15.2 23.1 24.4 33c15.7 16.9 39.6 18.4 57.2 8.7v.9c0 9.2 2.7 18.5 7.9 26.3H29.7C13.3 512 0 498.7 0 482.3C0 383.8 79.8 304 178.3 304zM436 218.2c0-7 4.5-13.3 11.3-14.8c10.5-2.4 21.5-3.7 32.7-3.7s22.2 1.3 32.7 3.7c6.8 1.5 11.3 7.8 11.3 14.8v17.7c0 7.8 4.8 14.8 11.6 18.7c6.8 3.9 15.1 4.5 21.8 .6l13.8-7.9c6.1-3.5 13.7-2.7 18.5 2.4c7.6 8.1 14.3 17.2 20.1 27.2s10.3 20.4 13.5 31c2.1 6.7-1.1 13.7-7.2 17.2l-14.4 8.3c-6.5 3.7-10 10.9-10 18.4s3.5 14.7 10 18.4l14.4 8.3c6.1 3.5 9.2 10.5 7.2 17.2c-3.3 10.6-7.8 21-13.5 31s-12.5 19.1-20.1 27.2c-4.8 5.1-12.5 5.9-18.5 2.4l-13.8-7.9c-6.7-3.9-15.1-3.3-21.8 .6c-6.8 3.9-11.6 10.9-11.6 18.7v17.7c0 7-4.5 13.3-11.3 14.8c-10.5 2.4-21.5 3.7-32.7 3.7s-22.2-1.3-32.7-3.7c-6.8-1.5-11.3-7.8-11.3-14.8V467.8c0-7.9-4.9-14.9-11.7-18.9c-6.8-3.9-15.2-4.5-22-.6l-13.5 7.8c-6.1 3.5-13.7 2.7-18.5-2.4c-7.6-8.1-14.3-17.2-20.1-27.2s-10.3-20.4-13.5-31c-2.1-6.7 1.1-13.7 7.2-17.2l14-8.1c6.5-3.8 10.1-11.1 10.1-18.6s-3.5-14.8-10.1-18.6l-14-8.1c-6.1-3.5-9.2-10.5-7.2-17.2c3.3-10.6 7.7-21 13.5-31s12.5-19.1 20.1-27.2c4.8-5.1 12.4-5.9 18.5-2.4l13.6 7.8c6.8 3.9 15.2 3.3 22-.6c6.9-3.9 11.7-11 11.7-18.9V218.2zm92.1 133.5a48.1 48.1 0 1 0 -96.1 0 48.1 48.1 0 1 0 96.1 0z\"],\"arrow-up-1-9\":[576,512,[\"sort-numeric-up\"],\"f163\",\"M450.7 38c8.3 6 13.3 15.7 13.3 26v96h16c17.7 0 32 14.3 32 32s-14.3 32-32 32H432 384c-17.7 0-32-14.3-32-32s14.3-32 32-32h16V108.4l-5.9 2c-16.8 5.6-34.9-3.5-40.5-20.2s3.5-34.9 20.2-40.5l48-16c9.8-3.3 20.5-1.6 28.8 4.4zM160 32c9 0 17.5 3.8 23.6 10.4l88 96c11.9 13 11.1 33.3-2 45.2s-33.3 11.1-45.2-2L192 146.3V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V146.3L95.6 181.6c-11.9 13-32.2 13.9-45.2 2s-13.9-32.2-2-45.2l88-96C142.5 35.8 151 32 160 32zM445.7 364.9A32 32 0 1 0 418.3 307a32 32 0 1 0 27.4 57.9zm-40.7 54.9C369.6 408.4 344 375.2 344 336c0-48.6 39.4-88 88-88s88 39.4 88 88c0 23.5-7.5 46.3-21.5 65.2L449.7 467c-10.5 14.2-30.6 17.2-44.8 6.7s-17.2-30.6-6.7-44.8l6.8-9.2z\"],\"door-closed\":[576,512,[128682],\"f52a\",\"M96 64c0-35.3 28.7-64 64-64H416c35.3 0 64 28.7 64 64V448h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H432 144 32c-17.7 0-32-14.3-32-32s14.3-32 32-32H96V64zM384 288a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"shield-virus\":[512,512,[],\"e06c\",\"M269.4 2.9C265.2 1 260.7 0 256 0s-9.2 1-13.4 2.9L54.3 82.8c-22 9.3-38.4 31-38.3 57.2c.5 99.2 41.3 280.7 213.6 363.2c16.7 8 36.1 8 52.8 0C454.7 420.7 495.5 239.2 496 140c.1-26.2-16.3-47.9-38.3-57.2L269.4 2.9zM256 112c8.8 0 16 7.2 16 16c0 33 39.9 49.5 63.2 26.2c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6C334.5 200.1 351 240 384 240c8.8 0 16 7.2 16 16s-7.2 16-16 16c-33 0-49.5 39.9-26.2 63.2c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0C311.9 334.5 272 351 272 384c0 8.8-7.2 16-16 16s-16-7.2-16-16c0-33-39.9-49.5-63.2-26.2c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6C177.5 311.9 161 272 128 272c-8.8 0-16-7.2-16-16s7.2-16 16-16c33 0 49.5-39.9 26.2-63.2c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0C200.1 177.5 240 161 240 128c0-8.8 7.2-16 16-16zM232 256a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm72 32a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z\"],\"dice-six\":[448,512,[9861],\"f526\",\"M0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm160 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM128 288a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm32 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM320 192a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm32 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM320 384a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"mosquito-net\":[640,512,[],\"e52c\",\"M168.8 462.3c-7.9-4-11.1-13.6-7.2-21.5L192 380.2l0-44.2c0-4.2 1.7-8.3 4.7-11.3L256 265.4V242.2L139.2 344C87.8 395.3 0 358.9 0 286.3c0-41.1 30.6-75.8 71.4-80.9l159.9-23.9-49.6-41.3c-5.1-4.2-7-11.1-4.9-17.4l13.9-41.7-29-58.1c-4-7.9-.7-17.5 7.2-21.5s17.5-.7 21.5 7.2l32 64c1.9 3.8 2.2 8.2 .9 12.2l-12.5 37.6L256 160.5V137.9c0-14.9 10.1-27.3 23.8-31V63.7c0-4.5 3.7-8.2 8.2-8.2s8.2 3.7 8.2 8.2V107c13.7 3.6 23.8 16.1 23.8 31v22.6l45.4-37.8L352.8 85.1c-1.3-4-1-8.4 .9-12.2l32-64c4-7.9 13.6-11.1 21.5-7.2s11.1 13.6 7.2 21.5l-29 58.1 13.9 41.7c2.1 6.2 .1 13.1-4.9 17.4l-49.6 41.3 159.9 23.9c22.5 2.8 41.8 14.6 54.7 31.4c-2.7 2.6-5.2 5.4-7.3 8.6c-8.6-12.9-23.3-21.5-40-21.5s-31.4 8.5-40 21.5c-8.6-12.9-23.3-21.5-40-21.5c-21.7 0-40 14.3-45.9 34.1c-10.7 3.2-19.8 10.1-25.9 19.2l-40.2-35v23.1l32.4 32.4c-.3 2-.4 4.1-.4 6.2c0 16.7 8.5 31.4 21.5 40c-4 2.6-7.5 5.9-10.6 9.5L320 310.6v50c0 17.7-14.3 32-32 32s-32-14.3-32-32v-50l-32 32 0 41.4c0 2.5-.6 4.9-1.7 7.2l-32 64c-4 7.9-13.6 11.1-21.5 7.2zM512 256c8.8 0 16 7.2 16 16v16h48V272c0-8.8 7.2-16 16-16s16 7.2 16 16v16h16c8.8 0 16 7.2 16 16s-7.2 16-16 16H608v48h16c8.8 0 16 7.2 16 16s-7.2 16-16 16H608v48h16c8.8 0 16 7.2 16 16s-7.2 16-16 16H608v16c0 8.8-7.2 16-16 16s-16-7.2-16-16V480H528v16c0 8.8-7.2 16-16 16s-16-7.2-16-16V480H448v16c0 8.8-7.2 16-16 16s-16-7.2-16-16V480H400c-8.8 0-16-7.2-16-16s7.2-16 16-16h16V400H400c-8.8 0-16-7.2-16-16s7.2-16 16-16h16V320H400c-8.8 0-16-7.2-16-16s7.2-16 16-16h16V272c0-8.8 7.2-16 16-16s16 7.2 16 16v16h48V272c0-8.8 7.2-16 16-16zm16 112h48V320H528v48zm0 80h48V400H528v48zM448 320v48h48V320H448zm0 80v48h48V400H448z\"],\"bridge-water\":[576,512,[],\"e4ce\",\"M0 96C0 78.3 14.3 64 32 64H544c17.7 0 32 14.3 32 32v35.6c0 15.7-12.7 28.4-28.4 28.4c-37.3 0-67.6 30.2-67.6 67.6V352.5c-12.9 0-25.8 3.9-36.8 11.7c-18 12.4-40.1 20.3-59.2 20.3h0l0-.5V256c0-53-43-96-96-96s-96 43-96 96V384l0 .5c-19 0-41.2-7.9-59.1-20.3c-11.1-7.8-24-11.7-36.9-11.7V227.6C96 190.2 65.8 160 28.4 160C12.7 160 0 147.3 0 131.6V96zM306.5 389.9C329 405.4 356.5 416 384 416c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 469.7 417 480 384 480c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 405.2 165.1 416 192 416c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z\"],\"person-booth\":[576,512,[],\"f756\",\"M256 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V192h64V32zm320 0c0-17.7-14.3-32-32-32s-32 14.3-32 32V480c0 17.7 14.3 32 32 32s32-14.3 32-32V32zM224 512c17.7 0 32-14.3 32-32V320H192V480c0 17.7 14.3 32 32 32zM320 0c-9.3 0-18.1 4-24.2 11s-8.8 16.3-7.5 25.5l31.2 218.6L288.6 409.7c-3.5 17.3 7.8 34.2 25.1 37.7s34.2-7.8 37.7-25.1l.7-3.6c1.3 16.4 15.1 29.4 31.9 29.4c17.7 0 32-14.3 32-32c0 17.7 14.3 32 32 32s32-14.3 32-32V32c0-17.7-14.3-32-32-32H320zM112 80A48 48 0 1 0 16 80a48 48 0 1 0 96 0zm0 261.3V269.3l4.7 4.7c9 9 21.2 14.1 33.9 14.1H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H157.3l-41.6-41.6c-14.3-14.3-33.8-22.4-54-22.4C27.6 160 0 187.6 0 221.6v55.7l0 .9V480c0 17.7 14.3 32 32 32s32-14.3 32-32V384l32 42.7V480c0 17.7 14.3 32 32 32s32-14.3 32-32V421.3c0-10.4-3.4-20.5-9.6-28.8L112 341.3z\"],\"text-width\":[448,512,[],\"f035\",\"M64 128V96H192l0 128H176c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H256l0-128H384v32c0 17.7 14.3 32 32 32s32-14.3 32-32V80c0-26.5-21.5-48-48-48H224 48C21.5 32 0 53.5 0 80v48c0 17.7 14.3 32 32 32s32-14.3 32-32zM9.4 361.4c-12.5 12.5-12.5 32.8 0 45.3l64 64c9.2 9.2 22.9 11.9 34.9 6.9s19.8-16.6 19.8-29.6V416H320v32c0 12.9 7.8 24.6 19.8 29.6s25.7 2.2 34.9-6.9l64-64c12.5-12.5 12.5-32.8 0-45.3l-64-64c-9.2-9.2-22.9-11.9-34.9-6.9s-19.8 16.6-19.8 29.6v32H128V320c0-12.9-7.8-24.6-19.8-29.6s-25.7-2.2-34.9 6.9l-64 64z\"],\"hat-wizard\":[512,512,[],\"f6e8\",\"M64 416L168.6 180.7c15.3-34.4 40.3-63.5 72-83.7l146.9-94c3-1.9 6.5-2.9 10-2.9C407.7 0 416 8.3 416 18.6v1.6c0 2.6-.5 5.1-1.4 7.5L354.8 176.9c-1.9 4.7-2.8 9.7-2.8 14.7c0 5.5 1.2 11 3.4 16.1L448 416H240.9l11.8-35.4 40.4-13.5c6.5-2.2 10.9-8.3 10.9-15.2s-4.4-13-10.9-15.2l-40.4-13.5-13.5-40.4C237 276.4 230.9 272 224 272s-13 4.4-15.2 10.9l-13.5 40.4-40.4 13.5C148.4 339 144 345.1 144 352s4.4 13 10.9 15.2l40.4 13.5L207.1 416H64zM279.6 141.5c-1.1-3.3-4.1-5.5-7.6-5.5s-6.5 2.2-7.6 5.5l-6.7 20.2-20.2 6.7c-3.3 1.1-5.5 4.1-5.5 7.6s2.2 6.5 5.5 7.6l20.2 6.7 6.7 20.2c1.1 3.3 4.1 5.5 7.6 5.5s6.5-2.2 7.6-5.5l6.7-20.2 20.2-6.7c3.3-1.1 5.5-4.1 5.5-7.6s-2.2-6.5-5.5-7.6l-20.2-6.7-6.7-20.2zM32 448H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],\"pen-fancy\":[512,512,[128395,10002],\"f5ac\",\"M373.5 27.1C388.5 9.9 410.2 0 433 0c43.6 0 79 35.4 79 79c0 22.8-9.9 44.6-27.1 59.6L277.7 319l-10.3-10.3-64-64L193 234.3 373.5 27.1zM170.3 256.9l10.4 10.4 64 64 10.4 10.4-19.2 83.4c-3.9 17.1-16.9 30.7-33.8 35.4L24.4 510.3l95.4-95.4c2.6 .7 5.4 1.1 8.3 1.1c17.7 0 32-14.3 32-32s-14.3-32-32-32s-32 14.3-32 32c0 2.9 .4 5.6 1.1 8.3L1.7 487.6 51.5 310c4.7-16.9 18.3-29.9 35.4-33.8l83.4-19.2z\"],\"person-digging\":[576,512,[\"digging\"],\"f85e\",\"M208 64a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM9.8 214.8c5.1-12.2 19.1-18 31.4-12.9L60.7 210l22.9-38.1C99.9 144.6 129.3 128 161 128c51.4 0 97 32.9 113.3 81.7l34.6 103.7 79.3 33.1 34.2-45.6c6.4-8.5 16.6-13.3 27.2-12.8s20.3 6.4 25.8 15.5l96 160c5.9 9.9 6.1 22.2 .4 32.2s-16.3 16.2-27.8 16.2H288c-11.1 0-21.4-5.7-27.2-15.2s-6.4-21.2-1.4-31.1l16-32c5.4-10.8 16.5-17.7 28.6-17.7h32l22.5-30L22.8 246.2c-12.2-5.1-18-19.1-12.9-31.4zm82.8 91.8l112 48c11.8 5 19.4 16.6 19.4 29.4v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V405.1l-60.6-26-37 111c-5.6 16.8-23.7 25.8-40.5 20.2S-3.9 486.6 1.6 469.9l48-144 11-33 32 13.7z\"],trash:[448,512,[],\"f1f8\",\"M135.2 17.7L128 32H32C14.3 32 0 46.3 0 64S14.3 96 32 96H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H320l-7.2-14.3C307.4 6.8 296.3 0 284.2 0H163.8c-12.1 0-23.2 6.8-28.6 17.7zM416 128H32L53.2 467c1.6 25.3 22.6 45 47.9 45H346.9c25.3 0 46.3-19.7 47.9-45L416 128z\"],\"gauge-simple\":[512,512,[\"gauge-simple-med\",\"tachometer-average\"],\"f629\",\"M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm320 96c0-26.9-16.5-49.9-40-59.3V88c0-13.3-10.7-24-24-24s-24 10.7-24 24V292.7c-23.5 9.5-40 32.5-40 59.3c0 35.3 28.7 64 64 64s64-28.7 64-64z\"],\"book-medical\":[448,512,[],\"f7e6\",\"M0 96C0 43 43 0 96 0H384h32c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384 96c-53 0-96-43-96-96V96zM64 416c0 17.7 14.3 32 32 32H352V384H96c-17.7 0-32 14.3-32 32zM208 112v48H160c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V224h48c8.8 0 16-7.2 16-16V176c0-8.8-7.2-16-16-16H272V112c0-8.8-7.2-16-16-16H224c-8.8 0-16 7.2-16 16z\"],poo:[512,512,[128169],\"f2fe\",\"M268.9 .9c-5.5-.7-11 1.4-14.5 5.7s-4.6 10.1-2.8 15.4c2.8 8.2 4.3 16.9 4.3 26.1c0 44.1-35.7 79.9-79.8 80H160c-35.3 0-64 28.7-64 64c0 19.1 8.4 36.3 21.7 48H104c-39.8 0-72 32.2-72 72c0 23.2 11 43.8 28 57c-34.1 5.7-60 35.3-60 71c0 39.8 32.2 72 72 72H440c39.8 0 72-32.2 72-72c0-35.7-25.9-65.3-60-71c17-13.2 28-33.8 28-57c0-39.8-32.2-72-72-72H394.3c13.3-11.7 21.7-28.9 21.7-48c0-35.3-28.7-64-64-64h-5.5c3.5-10 5.5-20.8 5.5-32c0-48.6-36.2-88.8-83.1-95.1zM192 256a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm96 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm64 108.3c0 2.4-.7 4.8-2.2 6.7c-8.2 10.5-39.5 45-93.8 45s-85.6-34.6-93.8-45c-1.5-1.9-2.2-4.3-2.2-6.7c0-6.8 5.5-12.3 12.3-12.3H339.7c6.8 0 12.3 5.5 12.3 12.3z\"],\"quote-right\":[448,512,[8221,\"quote-right-alt\"],\"f10e\",\"M448 296c0 66.3-53.7 120-120 120h-8c-17.7 0-32-14.3-32-32s14.3-32 32-32h8c30.9 0 56-25.1 56-56v-8H320c-35.3 0-64-28.7-64-64V160c0-35.3 28.7-64 64-64h64c35.3 0 64 28.7 64 64v32 32 72zm-256 0c0 66.3-53.7 120-120 120H64c-17.7 0-32-14.3-32-32s14.3-32 32-32h8c30.9 0 56-25.1 56-56v-8H64c-35.3 0-64-28.7-64-64V160c0-35.3 28.7-64 64-64h64c35.3 0 64 28.7 64 64v32 32 72z\"],shirt:[640,512,[128085,\"t-shirt\",\"tshirt\"],\"f553\",\"M211.8 0c7.8 0 14.3 5.7 16.7 13.2C240.8 51.9 277.1 80 320 80s79.2-28.1 91.5-66.8C413.9 5.7 420.4 0 428.2 0h12.6c22.5 0 44.2 7.9 61.5 22.3L628.5 127.4c6.6 5.5 10.7 13.5 11.4 22.1s-2.1 17.1-7.8 23.6l-56 64c-11.4 13.1-31.2 14.6-44.6 3.5L480 197.7V448c0 35.3-28.7 64-64 64H224c-35.3 0-64-28.7-64-64V197.7l-51.5 42.9c-13.3 11.1-33.1 9.6-44.6-3.5l-56-64c-5.7-6.5-8.5-15-7.8-23.6s4.8-16.6 11.4-22.1L137.7 22.3C155 7.9 176.7 0 199.2 0h12.6z\"],cubes:[576,512,[],\"f1b3\",\"M290.8 48.6l78.4 29.7L288 109.5 206.8 78.3l78.4-29.7c1.8-.7 3.8-.7 5.7 0zM136 92.5V204.7c-1.3 .4-2.6 .8-3.9 1.3l-96 36.4C14.4 250.6 0 271.5 0 294.7V413.9c0 22.2 13.1 42.3 33.5 51.3l96 42.2c14.4 6.3 30.7 6.3 45.1 0L288 457.5l113.5 49.9c14.4 6.3 30.7 6.3 45.1 0l96-42.2c20.3-8.9 33.5-29.1 33.5-51.3V294.7c0-23.3-14.4-44.1-36.1-52.4l-96-36.4c-1.3-.5-2.6-.9-3.9-1.3V92.5c0-23.3-14.4-44.1-36.1-52.4l-96-36.4c-12.8-4.8-26.9-4.8-39.7 0l-96 36.4C150.4 48.4 136 69.3 136 92.5zM392 210.6l-82.4 31.2V152.6L392 121v89.6zM154.8 250.9l78.4 29.7L152 311.7 70.8 280.6l78.4-29.7c1.8-.7 3.8-.7 5.7 0zm18.8 204.4V354.8L256 323.2v95.9l-82.4 36.2zM421.2 250.9c1.8-.7 3.8-.7 5.7 0l78.4 29.7L424 311.7l-81.2-31.1 78.4-29.7zM523.2 421.2l-77.6 34.1V354.8L528 323.2v90.7c0 3.2-1.9 6-4.8 7.3z\"],divide:[448,512,[10135,247],\"f529\",\"M272 96a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zm0 320a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM400 288c17.7 0 32-14.3 32-32s-14.3-32-32-32H48c-17.7 0-32 14.3-32 32s14.3 32 32 32H400z\"],\"tenge-sign\":[384,512,[8376,\"tenge\"],\"f7d7\",\"M0 64C0 46.3 14.3 32 32 32H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 96 0 81.7 0 64zM0 192c0-17.7 14.3-32 32-32H192 352c17.7 0 32 14.3 32 32s-14.3 32-32 32H224V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V224H32c-17.7 0-32-14.3-32-32z\"],headphones:[512,512,[127911],\"f025\",\"M256 80C149.9 80 62.4 159.4 49.6 262c9.4-3.8 19.6-6 30.4-6c26.5 0 48 21.5 48 48V432c0 26.5-21.5 48-48 48c-44.2 0-80-35.8-80-80V384 336 288C0 146.6 114.6 32 256 32s256 114.6 256 256v48 48 16c0 44.2-35.8 80-80 80c-26.5 0-48-21.5-48-48V304c0-26.5 21.5-48 48-48c10.8 0 21 2.1 30.4 6C449.6 159.4 362.1 80 256 80z\"],\"hands-holding\":[640,512,[],\"f4c2\",\"M80 104c0-22.1-17.9-40-40-40S0 81.9 0 104v56 64V325.5c0 25.5 10.1 49.9 28.1 67.9L128 493.3c12 12 28.3 18.7 45.3 18.7H240c26.5 0 48-21.5 48-48V385.1c0-29.7-11.8-58.2-32.8-79.2l-25.3-25.3 0 0-15.2-15.2-32-32c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l32 32 15.2 15.2c11 11 9.2 29.2-3.7 37.8c-9.7 6.5-22.7 5.2-31-3.1L98.7 309.5c-12-12-18.7-28.3-18.7-45.3V224 144 104zm480 0v40 80 40.2c0 17-6.7 33.3-18.7 45.3l-51.1 51.1c-8.3 8.3-21.3 9.6-31 3.1c-12.9-8.6-14.7-26.9-3.7-37.8l15.2-15.2 32-32c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-32 32-15.2 15.2 0 0-25.3 25.3c-21 21-32.8 49.5-32.8 79.2V464c0 26.5 21.5 48 48 48h66.7c17 0 33.3-6.7 45.3-18.7l99.9-99.9c18-18 28.1-42.4 28.1-67.9V224 160 104c0-22.1-17.9-40-40-40s-40 17.9-40 40z\"],\"hands-clapping\":[512,512,[],\"e1a8\",\"M336 16V80c0 8.8-7.2 16-16 16s-16-7.2-16-16V16c0-8.8 7.2-16 16-16s16 7.2 16 16zm-98.7 7.1l32 48c4.9 7.4 2.9 17.3-4.4 22.2s-17.3 2.9-22.2-4.4l-32-48c-4.9-7.4-2.9-17.3 4.4-22.2s17.3-2.9 22.2 4.4zM135 119c9.4-9.4 24.6-9.4 33.9 0L292.7 242.7c10.1 10.1 27.3 2.9 27.3-11.3V192c0-17.7 14.3-32 32-32s32 14.3 32 32V345.6c0 57.1-30 110-78.9 139.4c-64 38.4-145.8 28.3-198.5-24.4L7 361c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l53 53c6.1 6.1 16 6.1 22.1 0s6.1-16 0-22.1L23 265c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l93 93c6.1 6.1 16 6.1 22.1 0s6.1-16 0-22.1L55 185c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l117 117c6.1 6.1 16 6.1 22.1 0s6.1-16 0-22.1l-93-93c-9.4-9.4-9.4-24.6 0-33.9zM433.1 484.9c-24.2 14.5-50.9 22.1-77.7 23.1c48.1-39.6 76.6-99 76.6-162.4l0-98.1c8.2-.1 16-6.4 16-16V192c0-17.7 14.3-32 32-32s32 14.3 32 32V345.6c0 57.1-30 110-78.9 139.4zM424.9 18.7c7.4 4.9 9.3 14.8 4.4 22.2l-32 48c-4.9 7.4-14.8 9.3-22.2 4.4s-9.3-14.8-4.4-22.2l32-48c4.9-7.4 14.8-9.3 22.2-4.4z\"],republican:[640,512,[],\"f75e\",\"M0 192C0 103.6 71.6 32 160 32H384c88.4 0 160 71.6 160 160v64H0V192zm415.9-64c-2.4 0-4.7 1.3-5.7 3.4l-12.6 24.6-28.2 4c-2.4 .3-4.4 2-5.2 4.2s-.1 4.7 1.6 6.3l20.4 19.2-4.8 27.1c-.4 2.3 .6 4.7 2.5 6s4.6 1.6 6.7 .5l25.2-12.8 25.2 12.8c2.2 1.1 4.8 .9 6.7-.5s3-3.7 2.5-6l-4.8-27.1L466 170.5c1.7-1.6 2.4-4.1 1.6-6.3s-2.8-3.9-5.2-4.2l-28.2-4-12.6-24.6c-1.1-2.1-3.3-3.4-5.7-3.4zm-138.3 3.4c-1.1-2.1-3.3-3.4-5.7-3.4s-4.7 1.3-5.7 3.4l-12.6 24.6-28.2 4c-2.4 .3-4.4 2-5.2 4.2s-.1 4.7 1.6 6.3l20.4 19.2-4.8 27.1c-.4 2.3 .6 4.7 2.5 6s4.6 1.6 6.7 .5l25.2-12.8 25.2 12.8c2.2 1.1 4.8 .9 6.7-.5s3-3.7 2.5-6l-4.8-27.1L322 170.5c1.7-1.6 2.4-4.1 1.6-6.3s-2.8-3.9-5.2-4.2l-28.2-4-12.6-24.6zM127.9 128c-2.4 0-4.7 1.3-5.7 3.4l-12.6 24.6-28.2 4c-2.4 .3-4.4 2-5.2 4.2s-.1 4.7 1.6 6.3l20.4 19.2-4.8 27.1c-.4 2.3 .6 4.7 2.5 6s4.6 1.6 6.7 .5l25.2-12.8 25.2 12.8c2.2 1.1 4.8 .9 6.7-.5s3-3.7 2.5-6l-4.8-27.1L178 170.5c1.7-1.6 2.4-4.1 1.6-6.3s-2.8-3.9-5.2-4.2l-28.2-4-12.6-24.6c-1.1-2.1-3.3-3.4-5.7-3.4zm.1 160H320h96 32 64 32v32 80c0 8.8 7.2 16 16 16s16-7.2 16-16V352c0-17.7 14.3-32 32-32s32 14.3 32 32v48c0 44.2-35.8 80-80 80s-80-35.8-80-80V352H448v32 64c0 17.7-14.3 32-32 32H352c-17.7 0-32-14.3-32-32V384H128v64c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32V384 288H128z\"],\"arrow-left\":[448,512,[8592],\"f060\",\"M9.4 233.4c-12.5 12.5-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L109.2 288 416 288c17.7 0 32-14.3 32-32s-14.3-32-32-32l-306.7 0L214.6 118.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-160 160z\"],\"person-circle-xmark\":[576,512,[],\"e543\",\"M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l44.9 74.7c-16.1 17.6-28.6 38.5-36.6 61.5c-1.9-1.8-3.5-3.9-4.9-6.3L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152zM432 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm59.3 107.3c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0L432 345.4l-36.7-36.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6L409.4 368l-36.7 36.7c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L432 390.6l36.7 36.7c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6L454.6 368l36.7-36.7z\"],ruler:[512,512,[128207],\"f545\",\"M177.9 494.1c-18.7 18.7-49.1 18.7-67.9 0L17.9 401.9c-18.7-18.7-18.7-49.1 0-67.9l50.7-50.7 48 48c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-48-48 41.4-41.4 48 48c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-48-48 41.4-41.4 48 48c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-48-48 41.4-41.4 48 48c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-48-48 50.7-50.7c18.7-18.7 49.1-18.7 67.9 0l92.1 92.1c18.7 18.7 18.7 49.1 0 67.9L177.9 494.1z\"],\"align-left\":[448,512,[],\"f036\",\"M288 64c0 17.7-14.3 32-32 32H32C14.3 96 0 81.7 0 64S14.3 32 32 32H256c17.7 0 32 14.3 32 32zm0 256c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H256c17.7 0 32 14.3 32 32zM0 192c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32zM448 448c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H416c17.7 0 32 14.3 32 32z\"],\"dice-d6\":[448,512,[],\"f6d1\",\"M201 10.3c14.3-7.8 31.6-7.8 46 0L422.3 106c5.1 2.8 8.3 8.2 8.3 14s-3.2 11.2-8.3 14L231.7 238c-4.8 2.6-10.5 2.6-15.3 0L25.7 134c-5.1-2.8-8.3-8.2-8.3-14s3.2-11.2 8.3-14L201 10.3zM23.7 170l176 96c5.1 2.8 8.3 8.2 8.3 14V496c0 5.6-3 10.9-7.8 13.8s-10.9 3-15.8 .3L25 423.1C9.6 414.7 0 398.6 0 381V184c0-5.6 3-10.9 7.8-13.8s10.9-3 15.8-.3zm400.7 0c5-2.7 11-2.6 15.8 .3s7.8 8.1 7.8 13.8V381c0 17.6-9.6 33.7-25 42.1L263.7 510c-5 2.7-11 2.6-15.8-.3s-7.8-8.1-7.8-13.8V280c0-5.9 3.2-11.2 8.3-14l176-96z\"],restroom:[640,512,[],\"f7bd\",\"M80 48a48 48 0 1 1 96 0A48 48 0 1 1 80 48zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V325.2c-8.1 9.2-21.1 13.2-33.5 9.4c-16.9-5.3-26.3-23.2-21-40.1l30.9-99.1C44.9 155.3 82 128 124 128h8c42 0 79.1 27.3 91.6 67.4l30.9 99.1c5.3 16.9-4.1 34.8-21 40.1c-12.4 3.9-25.4-.2-33.5-9.4V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H120zM320 0c13.3 0 24 10.7 24 24V488c0 13.3-10.7 24-24 24s-24-10.7-24-24V24c0-13.3 10.7-24 24-24zM464 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM440 480V384H422.2c-10.9 0-18.6-10.7-15.2-21.1l9-26.9c-3.2 0-6.4-.5-9.5-1.5c-16.9-5.3-26.3-23.2-21-40.1l29.7-95.2C428.4 156.9 467.6 128 512 128s83.6 28.9 96.8 71.2l29.7 95.2c5.3 16.9-4.1 34.8-21 40.1c-3.2 1-6.4 1.5-9.5 1.5l9 26.9c3.5 10.4-4.3 21.1-15.2 21.1H584v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384H504v96c0 17.7-14.3 32-32 32s-32-14.3-32-32z\"],j:[320,512,[106],\"4a\",\"M288 32c17.7 0 32 14.3 32 32V320c0 88.4-71.6 160-160 160S0 408.4 0 320V288c0-17.7 14.3-32 32-32s32 14.3 32 32v32c0 53 43 96 96 96s96-43 96-96V64c0-17.7 14.3-32 32-32z\"],\"users-viewfinder\":[640,512,[],\"e595\",\"M48 48h88c13.3 0 24-10.7 24-24s-10.7-24-24-24H32C14.3 0 0 14.3 0 32V136c0 13.3 10.7 24 24 24s24-10.7 24-24V48zM175.8 224a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-26.5 32C119.9 256 96 279.9 96 309.3c0 14.7 11.9 26.7 26.7 26.7h56.1c8-34.1 32.8-61.7 65.2-73.6c-7.5-4.1-16.2-6.4-25.3-6.4H149.3zm368 80c14.7 0 26.7-11.9 26.7-26.7c0-29.5-23.9-53.3-53.3-53.3H421.3c-9.2 0-17.8 2.3-25.3 6.4c32.4 11.9 57.2 39.5 65.2 73.6h56.1zm-89.4 0c-8.6-24.3-29.9-42.6-55.9-47c-3.9-.7-7.9-1-12-1H280c-4.1 0-8.1 .3-12 1c-26 4.4-47.3 22.7-55.9 47c-2.7 7.5-4.1 15.6-4.1 24c0 13.3 10.7 24 24 24H408c13.3 0 24-10.7 24-24c0-8.4-1.4-16.5-4.1-24zM464 224a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-80-32a64 64 0 1 0 -128 0 64 64 0 1 0 128 0zM504 48h88v88c0 13.3 10.7 24 24 24s24-10.7 24-24V32c0-17.7-14.3-32-32-32H504c-13.3 0-24 10.7-24 24s10.7 24 24 24zM48 464V376c0-13.3-10.7-24-24-24s-24 10.7-24 24V480c0 17.7 14.3 32 32 32H136c13.3 0 24-10.7 24-24s-10.7-24-24-24H48zm456 0c-13.3 0-24 10.7-24 24s10.7 24 24 24H608c17.7 0 32-14.3 32-32V376c0-13.3-10.7-24-24-24s-24 10.7-24 24v88H504z\"],\"file-video\":[384,512,[],\"f1c8\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM64 288c0-17.7 14.3-32 32-32h96c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V288zM300.9 397.9L256 368V304l44.9-29.9c2-1.3 4.4-2.1 6.8-2.1c6.8 0 12.3 5.5 12.3 12.3V387.7c0 6.8-5.5 12.3-12.3 12.3c-2.4 0-4.8-.7-6.8-2.1z\"],\"up-right-from-square\":[512,512,[\"external-link-alt\"],\"f35d\",\"M352 0c-12.9 0-24.6 7.8-29.6 19.8s-2.2 25.7 6.9 34.9L370.7 96 201.4 265.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L416 141.3l41.4 41.4c9.2 9.2 22.9 11.9 34.9 6.9s19.8-16.6 19.8-29.6V32c0-17.7-14.3-32-32-32H352zM80 32C35.8 32 0 67.8 0 112V432c0 44.2 35.8 80 80 80H400c44.2 0 80-35.8 80-80V320c0-17.7-14.3-32-32-32s-32 14.3-32 32V432c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16H192c17.7 0 32-14.3 32-32s-14.3-32-32-32H80z\"],\"table-cells\":[512,512,[\"th\"],\"f00a\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm88 64v64H64V96h88zm56 0h88v64H208V96zm240 0v64H360V96h88zM64 224h88v64H64V224zm232 0v64H208V224h88zm64 0h88v64H360V224zM152 352v64H64V352h88zm56 0h88v64H208V352zm240 0v64H360V352h88z\"],\"file-pdf\":[512,512,[],\"f1c1\",\"M0 64C0 28.7 28.7 0 64 0L224 0l0 128c0 17.7 14.3 32 32 32l128 0 0 144-208 0c-35.3 0-64 28.7-64 64l0 144-48 0c-35.3 0-64-28.7-64-64L0 64zm384 64l-128 0L256 0 384 128zM176 352l32 0c30.9 0 56 25.1 56 56s-25.1 56-56 56l-16 0 0 32c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-48 0-80c0-8.8 7.2-16 16-16zm32 80c13.3 0 24-10.7 24-24s-10.7-24-24-24l-16 0 0 48 16 0zm96-80l32 0c26.5 0 48 21.5 48 48l0 64c0 26.5-21.5 48-48 48l-32 0c-8.8 0-16-7.2-16-16l0-128c0-8.8 7.2-16 16-16zm32 128c8.8 0 16-7.2 16-16l0-64c0-8.8-7.2-16-16-16l-16 0 0 96 16 0zm80-112c0-8.8 7.2-16 16-16l48 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-32 0 0 32 32 0c8.8 0 16 7.2 16 16s-7.2 16-16 16l-32 0 0 48c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-64 0-64z\"],\"book-bible\":[448,512,[\"bible\"],\"f647\",\"M96 0C43 0 0 43 0 96V416c0 53 43 96 96 96H384h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V384c17.7 0 32-14.3 32-32V32c0-17.7-14.3-32-32-32H384 96zm0 384H352v64H96c-17.7 0-32-14.3-32-32s14.3-32 32-32zM208 80c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272V304c0 8.8-7.2 16-16 16H224c-8.8 0-16-7.2-16-16V192H160c-8.8 0-16-7.2-16-16V144c0-8.8 7.2-16 16-16h48V80z\"],o:[448,512,[111],\"4f\",\"M224 96a160 160 0 1 0 0 320 160 160 0 1 0 0-320zM448 256A224 224 0 1 1 0 256a224 224 0 1 1 448 0z\"],\"suitcase-medical\":[512,512,[\"medkit\"],\"f0fa\",\"M184 48H328c4.4 0 8 3.6 8 8V96H176V56c0-4.4 3.6-8 8-8zm-56 8V96v32V480H384V128 96 56c0-30.9-25.1-56-56-56H184c-30.9 0-56 25.1-56 56zM96 96H64C28.7 96 0 124.7 0 160V416c0 35.3 28.7 64 64 64H96V96zM416 480h32c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H416V480zM224 208c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H288v48c0 8.8-7.2 16-16 16H240c-8.8 0-16-7.2-16-16V320H176c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h48V208z\"],\"user-secret\":[448,512,[128373],\"f21b\",\"M224 16c-6.7 0-10.8-2.8-15.5-6.1C201.9 5.4 194 0 176 0c-30.5 0-52 43.7-66 89.4C62.7 98.1 32 112.2 32 128c0 14.3 25 27.1 64.6 35.9c-.4 4-.6 8-.6 12.1c0 17 3.3 33.2 9.3 48H45.4C38 224 32 230 32 237.4c0 1.7 .3 3.4 1 5l38.8 96.9C28.2 371.8 0 423.8 0 482.3C0 498.7 13.3 512 29.7 512H418.3c16.4 0 29.7-13.3 29.7-29.7c0-58.5-28.2-110.4-71.7-143L415 242.4c.6-1.6 1-3.3 1-5c0-7.4-6-13.4-13.4-13.4H342.7c6-14.8 9.3-31 9.3-48c0-4.1-.2-8.1-.6-12.1C391 155.1 416 142.3 416 128c0-15.8-30.7-29.9-78-38.6C324 43.7 302.5 0 272 0c-18 0-25.9 5.4-32.5 9.9c-4.8 3.3-8.8 6.1-15.5 6.1zm56 208H267.6c-16.5 0-31.1-10.6-36.3-26.2c-2.3-7-12.2-7-14.5 0c-5.2 15.6-19.9 26.2-36.3 26.2H168c-22.1 0-40-17.9-40-40V169.6c28.2 4.1 61 6.4 96 6.4s67.8-2.3 96-6.4V184c0 22.1-17.9 40-40 40zm-88 96l16 32L176 480 128 288l64 32zm128-32L272 480 240 352l16-32 64-32z\"],otter:[640,512,[129446],\"f700\",\"M181.5 197.1l12.9 6.4c5.9 3 12.4 4.5 19.1 4.5c23.5 0 42.6-19.1 42.6-42.6V144c0-35.3-28.7-64-64-64H128c-35.3 0-64 28.7-64 64v21.4c0 23.5 19.1 42.6 42.6 42.6c6.6 0 13.1-1.5 19.1-4.5l12.9-6.4 8.4-4.2L135.1 185c-4.5-3-7.1-8-7.1-13.3V168c0-13.3 10.7-24 24-24h16c13.3 0 24 10.7 24 24v3.7c0 5.3-2.7 10.3-7.1 13.3l-11.8 7.9 8.4 4.2zm-8.6 49.4L160 240l-12.9 6.4c-12.6 6.3-26.5 9.6-40.5 9.6c-3.6 0-7.1-.2-10.6-.6v.6c0 35.3 28.7 64 64 64h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384V336 320c0-23.7 12.9-44.4 32-55.4c9.4-5.4 20.3-8.6 32-8.6V240c0-26.5 21.5-48 48-48c8.8 0 16 7.2 16 16v32 16 48c0 8.8 7.2 16 16 16s16-7.2 16-16V204.3c0-48.2-30.8-91-76.6-106.3l-8.5-2.8c-8-2.7-12.6-11.1-10.4-19.3s10.3-13.2 18.6-11.6l19.9 4C576 86.1 640 164.2 640 254.9l0 1.1h0c0 123.7-100.3 224-224 224h-1.1H256h-.6C132 480 32 380 32 256.6V256 216.8c-10.1-14.6-16-32.3-16-51.4V144l0-1.4C6.7 139.3 0 130.5 0 120c0-13.3 10.7-24 24-24h2.8C44.8 58.2 83.3 32 128 32h64c44.7 0 83.2 26.2 101.2 64H296c13.3 0 24 10.7 24 24c0 10.5-6.7 19.3-16 22.6l0 1.4v21.4c0 1.4 0 2.8-.1 4.3c12-6.2 25.7-9.6 40.1-9.6h8c17.7 0 32 14.3 32 32s-14.3 32-32 32h-8c-13.3 0-24 10.7-24 24v8h56.4c-15.2 17-24.4 39.4-24.4 64H320c-42.3 0-78.2-27.4-91-65.3c-5.1 .9-10.3 1.3-15.6 1.3c-14.1 0-27.9-3.3-40.5-9.6zM96 128a16 16 0 1 1 0 32 16 16 0 1 1 0-32zm112 16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0z\"],\"person-dress\":[320,512,[\"female\"],\"f182\",\"M160 0a48 48 0 1 1 0 96 48 48 0 1 1 0-96zM88 384H70.2c-10.9 0-18.6-10.7-15.2-21.1L93.3 248.1 59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l53.6-89.2c20.3-33.7 56.7-54.3 96-54.3h11.6c39.3 0 75.7 20.6 96 54.3l53.6 89.2c9.1 15.1 4.2 34.8-10.9 43.9s-34.8 4.2-43.9-10.9l-33.9-56.3L265 362.9c3.5 10.4-4.3 21.1-15.2 21.1H232v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384H152v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384z\"],\"comment-dollar\":[512,512,[],\"f651\",\"M256 448c141.4 0 256-93.1 256-208S397.4 32 256 32S0 125.1 0 240c0 45.1 17.7 86.8 47.7 120.9c-1.9 24.5-11.4 46.3-21.4 62.9c-5.5 9.2-11.1 16.6-15.2 21.6c-2.1 2.5-3.7 4.4-4.9 5.7c-.6 .6-1 1.1-1.3 1.4l-.3 .3 0 0 0 0 0 0 0 0c-4.6 4.6-5.9 11.4-3.4 17.4c2.5 6 8.3 9.9 14.8 9.9c28.7 0 57.6-8.9 81.6-19.3c22.9-10 42.4-21.9 54.3-30.6c31.8 11.5 67 17.9 104.1 17.9zm20-312v13.9c7.5 1.2 14.6 2.9 21.1 4.7c10.7 2.8 17 13.8 14.2 24.5s-13.8 17-24.5 14.2c-11-2.9-21.6-5-31.2-5.2c-7.9-.1-16 1.8-21.5 5c-4.8 2.8-6.2 5.6-6.2 9.3c0 1.8 .1 3.5 5.3 6.7c6.3 3.8 15.5 6.7 28.3 10.5l.7 .2c11.2 3.4 25.6 7.7 37.1 15c12.9 8.1 24.3 21.3 24.6 41.6c.3 20.9-10.5 36.1-24.8 45c-7.2 4.5-15.2 7.3-23.2 9V344c0 11-9 20-20 20s-20-9-20-20V329.4c-10.3-2.2-20-5.5-28.2-8.4l0 0 0 0c-2.1-.7-4.1-1.4-6.1-2.1c-10.5-3.5-16.1-14.8-12.6-25.3s14.8-16.1 25.3-12.6c2.5 .8 4.9 1.7 7.2 2.4c13.6 4.6 24 8.1 35.1 8.5c8.6 .3 16.5-1.6 21.4-4.7c4.1-2.5 6-5.5 5.9-10.5c0-2.9-.8-5-5.9-8.2c-6.3-4-15.4-6.9-28-10.7l-1.7-.5c-10.9-3.3-24.6-7.4-35.6-14c-12.7-7.7-24.6-20.5-24.7-40.7c-.1-21.1 11.8-35.7 25.8-43.9c6.9-4.1 14.5-6.8 22.2-8.5V136c0-11 9-20 20-20s20 9 20 20z\"],\"business-time\":[640,512,[\"briefcase-clock\"],\"f64a\",\"M184 48H328c4.4 0 8 3.6 8 8V96H176V56c0-4.4 3.6-8 8-8zm-56 8V96H64C28.7 96 0 124.7 0 160v96H192 352h8.2c32.3-39.1 81.1-64 135.8-64c5.4 0 10.7 .2 16 .7V160c0-35.3-28.7-64-64-64H384V56c0-30.9-25.1-56-56-56H184c-30.9 0-56 25.1-56 56zM320 352H224c-17.7 0-32-14.3-32-32V288H0V416c0 35.3 28.7 64 64 64H360.2C335.1 449.6 320 410.5 320 368c0-5.4 .2-10.7 .7-16l-.7 0zm320 16a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zM496 288c8.8 0 16 7.2 16 16v48h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H496c-8.8 0-16-7.2-16-16V304c0-8.8 7.2-16 16-16z\"],\"table-cells-large\":[512,512,[\"th-large\"],\"f009\",\"M448 96V224H288V96H448zm0 192V416H288V288H448zM224 224H64V96H224V224zM64 288H224V416H64V288zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64z\"],\"book-tanakh\":[448,512,[\"tanakh\"],\"f827\",\"M352 0c53 0 96 43 96 96V416c0 53-43 96-96 96H64 32c-17.7 0-32-14.3-32-32s14.3-32 32-32V384c-17.7 0-32-14.3-32-32V32C0 14.3 14.3 0 32 0H64 352zm0 384H96v64H352c17.7 0 32-14.3 32-32s-14.3-32-32-32zM138.7 208l13.9 24H124.9l13.9-24zm-13.9-24L97.1 232c-6.2 10.7 1.5 24 13.9 24h55.4l27.7 48c6.2 10.7 21.6 10.7 27.7 0l27.7-48H305c12.3 0 20-13.3 13.9-24l-27.7-48 27.7-48c6.2-10.7-1.5-24-13.9-24H249.6L221.9 64c-6.2-10.7-21.6-10.7-27.7 0l-27.7 48H111c-12.3 0-20 13.3-13.9 24l27.7 48zm27.7 0l27.7-48h55.4l27.7 48-27.7 48H180.3l-27.7-48zm0-48l-13.9 24-13.9-24h27.7zm41.6-24L208 88l13.9 24H194.1zm69.3 24h27.7l-13.9 24-13.9-24zm13.9 72l13.9 24H263.4l13.9-24zm-55.4 48L208 280l-13.9-24h27.7z\"],\"phone-volume\":[512,512,[\"volume-control-phone\"],\"f2a0\",\"M280 0C408.1 0 512 103.9 512 232c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-101.6-82.4-184-184-184c-13.3 0-24-10.7-24-24s10.7-24 24-24zm8 192a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm-32-72c0-13.3 10.7-24 24-24c75.1 0 136 60.9 136 136c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-48.6-39.4-88-88-88c-13.3 0-24-10.7-24-24zM117.5 1.4c19.4-5.3 39.7 4.6 47.4 23.2l40 96c6.8 16.3 2.1 35.2-11.6 46.3L144 207.3c33.3 70.4 90.3 127.4 160.7 160.7L345 318.7c11.2-13.7 30-18.4 46.3-11.6l96 40c18.6 7.7 28.5 28 23.2 47.4l-24 88C481.8 499.9 466 512 448 512C200.6 512 0 311.4 0 64C0 46 12.1 30.2 29.5 25.4l88-24z\"],\"hat-cowboy-side\":[640,512,[],\"f8c1\",\"M152.7 135.9l-10.4 57.2c6.8-.7 13.6-1.1 20.5-1.1h10.7c39.4 0 77.8 12.1 110.1 34.7L562.4 421.8l35.1 24.6c24.4-6 42.5-28.1 42.5-54.4c0-75.8-94.7-126.6-134.6-144.7L474 83.9C468.2 53.8 441.8 32 411.1 32h-2.7c-5.6 0-11.1 .7-16.5 2.2L199.2 85.5c-23.9 6.4-42 26-46.5 50.4zM0 384c0 35.3 28.7 64 64 64H544L265.3 252.9c-26.9-18.8-58.9-28.9-91.8-28.9H162.9c-60.6 0-116 34.2-143.1 88.4L13.5 325C4.6 342.7 0 362.3 0 382.2V384z\"],\"clipboard-user\":[384,512,[],\"f7f3\",\"M192 0c-41.8 0-77.4 26.7-90.5 64H64C28.7 64 0 92.7 0 128V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H282.5C269.4 26.7 233.8 0 192 0zm0 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM128 256a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM80 432c0-44.2 35.8-80 80-80h64c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16H96c-8.8 0-16-7.2-16-16z\"],child:[320,512,[],\"f1ae\",\"M96 64a64 64 0 1 1 128 0A64 64 0 1 1 96 64zm48 320v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V287.8L59.1 321c-9.4 15-29.2 19.4-44.1 10S-4.5 301.9 4.9 287l39.9-63.3C69.7 184 113.2 160 160 160s90.3 24 115.2 63.6L315.1 287c9.4 15 4.9 34.7-10 44.1s-34.7 4.9-44.1-10L240 287.8V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V384H144z\"],\"lira-sign\":[320,512,[8356],\"f195\",\"M112 160.4c0-35.5 28.8-64.4 64.4-64.4c6.9 0 13.8 1.1 20.4 3.3l81.2 27.1c16.8 5.6 34.9-3.5 40.5-20.2s-3.5-34.9-20.2-40.5L217 38.6c-13.1-4.4-26.8-6.6-40.6-6.6C105.5 32 48 89.5 48 160.4V192H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H48v32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H46c-2.2 10.5-6.1 20.6-11.7 29.9L4.6 431.5c-5.9 9.9-6.1 22.2-.4 32.2S20.5 480 32 480H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H88.5l.7-1.1c11.6-19.3 18.9-40.7 21.6-62.9H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H112V256H224c17.7 0 32-14.3 32-32s-14.3-32-32-32H112V160.4z\"],satellite:[512,512,[128752],\"f7bf\",\"M233 7c-9.4-9.4-24.6-9.4-33.9 0l-96 96c-9.4 9.4-9.4 24.6 0 33.9l89.4 89.4-15.5 15.5C152.3 230.4 124.9 224 96 224c-31.7 0-61.5 7.7-87.8 21.2c-9 4.7-10.3 16.7-3.1 23.8L112.7 376.7 96.3 393.1c-2.6-.7-5.4-1.1-8.3-1.1c-17.7 0-32 14.3-32 32s14.3 32 32 32s32-14.3 32-32c0-2.9-.4-5.6-1.1-8.3l16.4-16.4L242.9 506.9c7.2 7.2 19.2 5.9 23.8-3.1C280.3 477.5 288 447.7 288 416c0-28.9-6.4-56.3-17.8-80.9l15.5-15.5L375 409c9.4 9.4 24.6 9.4 33.9 0l96-96c9.4-9.4 9.4-24.6 0-33.9l-89.4-89.4 55-55c12.5-12.5 12.5-32.8 0-45.3l-48-48c-12.5-12.5-32.8-12.5-45.3 0l-55 55L233 7zm159 351l-72.4-72.4 62.1-62.1L454.1 296 392 358.1zM226.3 192.4L153.9 120 216 57.9l72.4 72.4-62.1 62.1z\"],\"plane-lock\":[640,512,[],\"e558\",\"M192 93.7C192 59.5 221 0 256 0c36 0 64 59.5 64 93.7v84.6l101.8 58.2C418 247.6 416 259.6 416 272v24.6c-17.9 10.4-30.3 29.1-31.8 50.9L320 329.1V400l57.6 43.2c4 3 6.4 7.8 6.4 12.8v24 18c0 7.8-6.3 14-14 14c-1.3 0-2.6-.2-3.9-.5L256 480 145.9 511.5c-1.3 .4-2.6 .5-3.9 .5c-7.8 0-14-6.3-14-14V456c0-5 2.4-9.8 6.4-12.8L192 400l0-70.9-171.6 49C10.2 381.1 0 373.4 0 362.8V297.3c0-5.7 3.1-11 8.1-13.9L192 178.3V93.7zM528 240c-17.7 0-32 14.3-32 32v48h64V272c0-17.7-14.3-32-32-32zm-80 32c0-44.2 35.8-80 80-80s80 35.8 80 80v48c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H448c-17.7 0-32-14.3-32-32V352c0-17.7 14.3-32 32-32V272z\"],tag:[448,512,[127991],\"f02b\",\"M0 80V229.5c0 17 6.7 33.3 18.7 45.3l176 176c25 25 65.5 25 90.5 0L418.7 317.3c25-25 25-65.5 0-90.5l-176-176c-12-12-28.3-18.7-45.3-18.7H48C21.5 32 0 53.5 0 80zm112 32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],comment:[512,512,[128489,61669],\"f075\",\"M512 240c0 114.9-114.6 208-256 208c-37.1 0-72.3-6.4-104.1-17.9c-11.9 8.7-31.3 20.6-54.3 30.6C73.6 471.1 44.7 480 16 480c-6.5 0-12.3-3.9-14.8-9.9c-2.5-6-1.1-12.8 3.4-17.4l0 0 0 0 0 0 0 0 .3-.3c.3-.3 .7-.7 1.3-1.4c1.1-1.2 2.8-3.1 4.9-5.7c4.1-5 9.6-12.4 15.2-21.6c10-16.6 19.5-38.4 21.4-62.9C17.7 326.8 0 285.1 0 240C0 125.1 114.6 32 256 32s256 93.1 256 208z\"],\"cake-candles\":[448,512,[127874,\"birthday-cake\",\"cake\"],\"f1fd\",\"M86.4 5.5L61.8 47.6C58 54.1 56 61.6 56 69.2V72c0 22.1 17.9 40 40 40s40-17.9 40-40V69.2c0-7.6-2-15-5.8-21.6L105.6 5.5C103.6 2.1 100 0 96 0s-7.6 2.1-9.6 5.5zm128 0L189.8 47.6c-3.8 6.5-5.8 14-5.8 21.6V72c0 22.1 17.9 40 40 40s40-17.9 40-40V69.2c0-7.6-2-15-5.8-21.6L233.6 5.5C231.6 2.1 228 0 224 0s-7.6 2.1-9.6 5.5zM317.8 47.6c-3.8 6.5-5.8 14-5.8 21.6V72c0 22.1 17.9 40 40 40s40-17.9 40-40V69.2c0-7.6-2-15-5.8-21.6L361.6 5.5C359.6 2.1 356 0 352 0s-7.6 2.1-9.6 5.5L317.8 47.6zM128 176c0-17.7-14.3-32-32-32s-32 14.3-32 32v48c-35.3 0-64 28.7-64 64v71c8.3 5.2 18.1 9 28.8 9c13.5 0 27.2-6.1 38.4-13.4c5.4-3.5 9.9-7.1 13-9.7c1.5-1.3 2.7-2.4 3.5-3.1c.4-.4 .7-.6 .8-.8l.1-.1 0 0 0 0s0 0 0 0s0 0 0 0c3.1-3.2 7.4-4.9 11.9-4.8s8.6 2.1 11.6 5.4l0 0 0 0 .1 .1c.1 .1 .4 .4 .7 .7c.7 .7 1.7 1.7 3.1 3c2.8 2.6 6.8 6.1 11.8 9.5c10.2 7.1 23 13.1 36.3 13.1s26.1-6 36.3-13.1c5-3.5 9-6.9 11.8-9.5c1.4-1.3 2.4-2.3 3.1-3c.3-.3 .6-.6 .7-.7l.1-.1c3-3.5 7.4-5.4 12-5.4s9 2 12 5.4l.1 .1c.1 .1 .4 .4 .7 .7c.7 .7 1.7 1.7 3.1 3c2.8 2.6 6.8 6.1 11.8 9.5c10.2 7.1 23 13.1 36.3 13.1s26.1-6 36.3-13.1c5-3.5 9-6.9 11.8-9.5c1.4-1.3 2.4-2.3 3.1-3c.3-.3 .6-.6 .7-.7l.1-.1c2.9-3.4 7.1-5.3 11.6-5.4s8.7 1.6 11.9 4.8l0 0 0 0 0 0 .1 .1c.2 .2 .4 .4 .8 .8c.8 .7 1.9 1.8 3.5 3.1c3.1 2.6 7.5 6.2 13 9.7c11.2 7.3 24.9 13.4 38.4 13.4c10.7 0 20.5-3.9 28.8-9V288c0-35.3-28.7-64-64-64V176c0-17.7-14.3-32-32-32s-32 14.3-32 32v48H256V176c0-17.7-14.3-32-32-32s-32 14.3-32 32v48H128V176zM448 394.6c-8.5 3.3-18.2 5.4-28.8 5.4c-22.5 0-42.4-9.9-55.8-18.6c-4.1-2.7-7.8-5.4-10.9-7.8c-2.8 2.4-6.1 5-9.8 7.5C329.8 390 310.6 400 288 400s-41.8-10-54.6-18.9c-3.5-2.4-6.7-4.9-9.4-7.2c-2.7 2.3-5.9 4.7-9.4 7.2C201.8 390 182.6 400 160 400s-41.8-10-54.6-18.9c-3.7-2.6-7-5.2-9.8-7.5c-3.1 2.4-6.8 5.1-10.9 7.8C71.2 390.1 51.3 400 28.8 400c-10.6 0-20.3-2.2-28.8-5.4V480c0 17.7 14.3 32 32 32H416c17.7 0 32-14.3 32-32V394.6z\"],envelope:[512,512,[128386,9993,61443],\"f0e0\",\"M48 64C21.5 64 0 85.5 0 112c0 15.1 7.1 29.3 19.2 38.4L236.8 313.6c11.4 8.5 27 8.5 38.4 0L492.8 150.4c12.1-9.1 19.2-23.3 19.2-38.4c0-26.5-21.5-48-48-48H48zM0 176V384c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V176L294.4 339.2c-22.8 17.1-54 17.1-76.8 0L0 176z\"],\"angles-up\":[448,512,[\"angle-double-up\"],\"f102\",\"M246.6 41.4c-12.5-12.5-32.8-12.5-45.3 0l-160 160c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L224 109.3 361.4 246.6c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-160-160zm160 352l-160-160c-12.5-12.5-32.8-12.5-45.3 0l-160 160c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L224 301.3 361.4 438.6c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3z\"],paperclip:[448,512,[128206],\"f0c6\",\"M364.2 83.8c-24.4-24.4-64-24.4-88.4 0l-184 184c-42.1 42.1-42.1 110.3 0 152.4s110.3 42.1 152.4 0l152-152c10.9-10.9 28.7-10.9 39.6 0s10.9 28.7 0 39.6l-152 152c-64 64-167.6 64-231.6 0s-64-167.6 0-231.6l184-184c46.3-46.3 121.3-46.3 167.6 0s46.3 121.3 0 167.6l-176 176c-28.6 28.6-75 28.6-103.6 0s-28.6-75 0-103.6l144-144c10.9-10.9 28.7-10.9 39.6 0s10.9 28.7 0 39.6l-144 144c-6.7 6.7-6.7 17.7 0 24.4s17.7 6.7 24.4 0l176-176c24.4-24.4 24.4-64 0-88.4z\"],\"arrow-right-to-city\":[640,512,[],\"e4b3\",\"M288 48c0-26.5 21.5-48 48-48h96c26.5 0 48 21.5 48 48V192h40V120c0-13.3 10.7-24 24-24s24 10.7 24 24v72h24c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H432 336c-26.5 0-48-21.5-48-48V48zm64 32v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V80c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16zm16 80c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V176c0-8.8-7.2-16-16-16H368zM352 272v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16zm176-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H528zM512 368v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H528c-8.8 0-16 7.2-16 16zM166.6 153.4l80 80c12.5 12.5 12.5 32.8 0 45.3l-80 80c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L146.7 288H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H146.7l-25.4-25.4c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0z\"],ribbon:[448,512,[127895],\"f4d6\",\"M333.2 322.8l0 0-133.9-146 0 0L146 118.6c7.8-5.1 37-22.6 78-22.6s70.2 17.4 78 22.6L245.7 180l85.6 93.4 27.4-29.8c16.3-17.7 25.3-40.9 25.3-65V149.1c0-19-5.6-37.5-16.1-53.3L327.8 35.6C312.9 13.4 287.9 0 261.2 0h-76c-25.8 0-50.1 12.5-65.1 33.5L81.9 87C70.3 103.2 64 122.8 64 142.8V164c0 23.2 8.4 45.6 23.6 63.1l56 64.2 0 0 83.3 95.6 0 0 91.8 105.3c10 11.5 26.8 14.3 40 6.8l54.5-31.1c17.8-10.2 21.6-34.3 7.7-49.4l-87.7-95.7zM205.2 410.6l-83.3-95.6L27.1 418.5c-13.9 15.1-10.1 39.2 7.7 49.4l55.1 31.5c13 7.4 29.3 4.9 39.4-6.1l75.9-82.6z\"],lungs:[640,512,[129729],\"f604\",\"M320 0c17.7 0 32 14.3 32 32V164.1c0 16.4 8.4 31.7 22.2 40.5l9.8 6.2V165.3C384 127 415 96 453.3 96c21.7 0 42.8 10.2 55.8 28.8c15.4 22.1 44.3 65.4 71 116.9c26.5 50.9 52.4 112.5 59.6 170.3c.2 1.3 .2 2.6 .2 4v7c0 49.1-39.8 89-89 89c-7.3 0-14.5-.9-21.6-2.7l-72.7-18.2C414 480.5 384 442.1 384 398V325l90.5 57.6c7.5 4.7 17.3 2.5 22.1-4.9s2.5-17.3-4.9-22.1L384 287.1v-.4l-44.1-28.1c-7.3-4.6-13.9-10.1-19.9-16.1c-5.9 6-12.6 11.5-19.9 16.1L256 286.7 161.2 347l-13.5 8.6c0 0 0 0-.1 0c-7.4 4.8-9.6 14.6-4.8 22.1c4.7 7.5 14.6 9.7 22.1 4.9l91.1-58V398c0 44.1-30 82.5-72.7 93.1l-72.7 18.2c-7.1 1.8-14.3 2.7-21.6 2.7c-49.1 0-89-39.8-89-89v-7c0-1.3 .1-2.7 .2-4c7.2-57.9 33.1-119.4 59.6-170.3c26.8-51.5 55.6-94.8 71-116.9c13-18.6 34-28.8 55.8-28.8C225 96 256 127 256 165.3v45.5l9.8-6.2c13.8-8.8 22.2-24.1 22.2-40.5V32c0-17.7 14.3-32 32-32z\"],\"arrow-up-9-1\":[576,512,[\"sort-numeric-up-alt\"],\"f887\",\"M160 32c9 0 17.5 3.8 23.6 10.4l88 96c11.9 13 11.1 33.3-2 45.2s-33.3 11.1-45.2-2L192 146.3V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V146.3L95.6 181.6c-11.9 13-32.2 13.9-45.2 2s-13.9-32.2-2-45.2l88-96C142.5 35.8 151 32 160 32zM450.7 294c8.3 6 13.3 15.7 13.3 26v96h16c17.7 0 32 14.3 32 32s-14.3 32-32 32H432 384c-17.7 0-32-14.3-32-32s14.3-32 32-32h16V364.4l-5.9 2c-16.8 5.6-34.9-3.5-40.5-20.2s3.5-34.9 20.2-40.5l48-16c9.8-3.3 20.5-1.6 28.8 4.4zm-5-145.1A32 32 0 1 0 418.3 91a32 32 0 1 0 27.4 57.9zm-40.7 54.9C369.6 192.4 344 159.2 344 120c0-48.6 39.4-88 88-88s88 39.4 88 88c0 23.5-7.5 46.3-21.5 65.2L449.7 251c-10.5 14.2-30.6 17.2-44.8 6.7s-17.2-30.6-6.7-44.8l6.8-9.2z\"],\"litecoin-sign\":[384,512,[],\"e1d3\",\"M128 64c0-17.7-14.3-32-32-32S64 46.3 64 64V213.6L23.2 225.2c-17 4.9-26.8 22.6-22 39.6s22.6 26.8 39.6 22L64 280.1V448c0 17.7 14.3 32 32 32H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H128V261.9l136.8-39.1c17-4.9 26.8-22.6 22-39.6s-22.6-26.8-39.6-22L128 195.3V64z\"],\"border-none\":[448,512,[],\"f850\",\"M32 480a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm96-64a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm0-384a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm0 256a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM320 416a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm0-320a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm0 128a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM224 480a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm0-448a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm0 256a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM416 416a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm0-384a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM32 96a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM416 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM32 288a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm192 32a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm192 64a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM32 320a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM416 192a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM32 128a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm192 64a32 32 0 1 1 0-64 32 32 0 1 1 0 64z\"],\"circle-nodes\":[512,512,[],\"e4e2\",\"M418.4 157.9c35.3-8.3 61.6-40 61.6-77.9c0-44.2-35.8-80-80-80c-43.4 0-78.7 34.5-80 77.5L136.2 151.1C121.7 136.8 101.9 128 80 128c-44.2 0-80 35.8-80 80s35.8 80 80 80c12.2 0 23.8-2.7 34.1-7.6L259.7 407.8c-2.4 7.6-3.7 15.8-3.7 24.2c0 44.2 35.8 80 80 80s80-35.8 80-80c0-27.7-14-52.1-35.4-66.4l37.8-207.7zM156.3 232.2c2.2-6.9 3.5-14.2 3.7-21.7l183.8-73.5c3.6 3.5 7.4 6.7 11.6 9.5L317.6 354.1c-5.5 1.3-10.8 3.1-15.8 5.5L156.3 232.2z\"],\"parachute-box\":[512,512,[],\"f4cd\",\"M383.5 192c.3-5.3 .5-10.6 .5-16c0-51-15.9-96-40.2-127.6C319.5 16.9 288.2 0 256 0s-63.5 16.9-87.8 48.4C143.9 80 128 125 128 176c0 5.4 .2 10.7 .5 16H240V320H208c-7 0-13.7 1.5-19.7 4.2L68.2 192H96.5c-.3-5.3-.5-10.6-.5-16c0-64 22.2-121.2 57.1-159.3C62 49.3 18.6 122.6 4.2 173.6C1.5 183.1 9 192 18.9 192h6L165.2 346.3c-3.3 6.5-5.2 13.9-5.2 21.7v96c0 26.5 21.5 48 48 48h96c26.5 0 48-21.5 48-48V368c0-7.8-1.9-15.2-5.2-21.7L487.1 192h6c9.9 0 17.4-8.9 14.7-18.4C493.4 122.6 450 49.3 358.9 16.7C393.8 54.8 416 112.1 416 176c0 5.4-.2 10.7-.5 16h28.3L323.7 324.2c-6-2.7-12.7-4.2-19.7-4.2H272V192H383.5z\"],indent:[448,512,[],\"f03c\",\"M0 64C0 46.3 14.3 32 32 32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 96 0 81.7 0 64zM192 192c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32zm32 96H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32s14.3-32 32-32zM0 448c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32zM127.8 268.6L25.8 347.9C15.3 356.1 0 348.6 0 335.3V176.7c0-13.3 15.3-20.8 25.8-12.6l101.9 79.3c8.2 6.4 8.2 18.9 0 25.3z\"],\"truck-field-un\":[640,512,[],\"e58e\",\"M96 32C60.7 32 32 60.7 32 96v32c-17.7 0-32 14.3-32 32v96c0 17.7 14.3 32 32 32v32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64c0 53 43 96 96 96s96-43 96-96H384c0 53 43 96 96 96s96-43 96-96h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V288c0-35.3-28.7-64-64-64h-4.2c-.4-1.1-.9-2.1-1.3-3.2L485.7 102c-10.3-23.1-33.2-38-58.5-38H375.4C364.4 44.9 343.7 32 320 32H96zm288 96h43.2l42.7 96H384V128zM112 384a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm368-48a48 48 0 1 1 0 96 48 48 0 1 1 0-96zM253.3 135.1l34.7 52V144c0-8.8 7.2-16 16-16s16 7.2 16 16v96c0 7.1-4.6 13.3-11.4 15.3s-14-.6-17.9-6.4l-34.7-52V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V144c0-7.1 4.6-13.3 11.4-15.3s14 .6 17.9 6.4zM128 144v64c0 8.8 7.2 16 16 16s16-7.2 16-16V144c0-8.8 7.2-16 16-16s16 7.2 16 16v64c0 26.5-21.5 48-48 48s-48-21.5-48-48V144c0-8.8 7.2-16 16-16s16 7.2 16 16z\"],hourglass:[384,512,[9203,62032,\"hourglass-empty\"],\"f254\",\"M0 32C0 14.3 14.3 0 32 0H64 320h32c17.7 0 32 14.3 32 32s-14.3 32-32 32V75c0 42.4-16.9 83.1-46.9 113.1L237.3 256l67.9 67.9c30 30 46.9 70.7 46.9 113.1v11c17.7 0 32 14.3 32 32s-14.3 32-32 32H320 64 32c-17.7 0-32-14.3-32-32s14.3-32 32-32V437c0-42.4 16.9-83.1 46.9-113.1L146.7 256 78.9 188.1C48.9 158.1 32 117.4 32 75V64C14.3 64 0 49.7 0 32zM96 64V75c0 25.5 10.1 49.9 28.1 67.9L192 210.7l67.9-67.9c18-18 28.1-42.4 28.1-67.9V64H96zm0 384H288V437c0-25.5-10.1-49.9-28.1-67.9L192 301.3l-67.9 67.9c-18 18-28.1 42.4-28.1 67.9v11z\"],mountain:[512,512,[127956],\"f6fc\",\"M256 32c12.5 0 24.1 6.4 30.8 17L503.4 394.4c5.6 8.9 8.6 19.2 8.6 29.7c0 30.9-25 55.9-55.9 55.9H55.9C25 480 0 455 0 424.1c0-10.5 3-20.8 8.6-29.7L225.2 49c6.6-10.6 18.3-17 30.8-17zm65 192L256 120.4 176.9 246.5l18.3 24.4c6.4 8.5 19.2 8.5 25.6 0l25.6-34.1c6-8.1 15.5-12.8 25.6-12.8h49z\"],\"user-doctor\":[448,512,[\"user-md\"],\"f0f0\",\"M224 256A128 128 0 1 0 224 0a128 128 0 1 0 0 256zm-96 55.2C54 332.9 0 401.3 0 482.3C0 498.7 13.3 512 29.7 512H418.3c16.4 0 29.7-13.3 29.7-29.7c0-81-54-149.4-128-171.1V362c27.6 7.1 48 32.2 48 62v40c0 8.8-7.2 16-16 16H336c-8.8 0-16-7.2-16-16s7.2-16 16-16V424c0-17.7-14.3-32-32-32s-32 14.3-32 32v24c8.8 0 16 7.2 16 16s-7.2 16-16 16H256c-8.8 0-16-7.2-16-16V424c0-29.8 20.4-54.9 48-62V304.9c-6-.6-12.1-.9-18.3-.9H178.3c-6.2 0-12.3 .3-18.3 .9v65.4c23.1 6.9 40 28.3 40 53.7c0 30.9-25.1 56-56 56s-56-25.1-56-56c0-25.4 16.9-46.8 40-53.7V311.2zM144 448a24 24 0 1 0 0-48 24 24 0 1 0 0 48z\"],\"circle-info\":[512,512,[\"info-circle\"],\"f05a\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM216 336h24V272H216c-13.3 0-24-10.7-24-24s10.7-24 24-24h48c13.3 0 24 10.7 24 24v88h8c13.3 0 24 10.7 24 24s-10.7 24-24 24H216c-13.3 0-24-10.7-24-24s10.7-24 24-24zm40-208a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"cloud-meatball\":[512,512,[],\"f73b\",\"M0 224c0 53 43 96 96 96h44.7c9.5-23.5 32.5-40 59.3-40c2 0 3.9 .1 5.8 .3C217.6 265.5 235.7 256 256 256s38.4 9.5 50.2 24.3c1.9-.2 3.9-.3 5.8-.3c26.9 0 49.9 16.5 59.3 40H416c53 0 96-43 96-96s-43-96-96-96c-.5 0-1.1 0-1.6 0c1.1-5.2 1.6-10.5 1.6-16c0-44.2-35.8-80-80-80c-24.3 0-46.1 10.9-60.8 28C256.5 24.3 219.1 0 176 0C114.1 0 64 50.1 64 112c0 7.1 .7 14.1 1.9 20.8C27.6 145.4 0 181.5 0 224zm288 96c0-17.7-14.3-32-32-32s-32 14.3-32 32c0 1 .1 2.1 .1 3.1c-.7-.8-1.4-1.6-2.1-2.3c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3c.7 .7 1.5 1.4 2.3 2.1c-1-.1-2.1-.1-3.1-.1c-17.7 0-32 14.3-32 32s14.3 32 32 32c1 0 2.1-.1 3.1-.1c-.8 .7-1.6 1.3-2.3 2.1c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0c.7-.7 1.4-1.5 2.1-2.3c-.1 1-.1 2.1-.1 3.1c0 17.7 14.3 32 32 32s32-14.3 32-32c0-1-.1-2.1-.1-3.1c.7 .8 1.3 1.6 2.1 2.3c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3c-.7-.7-1.5-1.4-2.3-2.1c1 .1 2.1 .1 3.1 .1c17.7 0 32-14.3 32-32s-14.3-32-32-32c-1 0-2.1 .1-3.1 .1c.8-.7 1.6-1.3 2.3-2.1c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0c-.7 .7-1.4 1.5-2.1 2.3c.1-1 .1-2.1 .1-3.1zM48 448a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm416 0a48 48 0 1 0 0-96 48 48 0 1 0 0 96z\"],camera:[512,512,[62258,\"camera-alt\"],\"f030\",\"M149.1 64.8L138.7 96H64C28.7 96 0 124.7 0 160V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H373.3L362.9 64.8C356.4 45.2 338.1 32 317.4 32H194.6c-20.7 0-39 13.2-45.5 32.8zM256 192a96 96 0 1 1 0 192 96 96 0 1 1 0-192z\"],\"square-virus\":[448,512,[],\"e578\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM223.8 93.7c13.3 0 24 10.7 24 24c0 29.3 35.4 43.9 56.1 23.2c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9c-20.7 20.7-6 56.1 23.2 56.1c13.3 0 24 10.7 24 24s-10.7 24-24 24c-29.3 0-43.9 35.4-23.2 56.1c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0c-20.7-20.7-56.1-6-56.1 23.2c0 13.3-10.7 24-24 24s-24-10.7-24-24c0-29.3-35.4-43.9-56.1-23.2c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9c20.7-20.7 6-56.1-23.2-56.1c-13.3 0-24-10.7-24-24s10.7-24 24-24c29.3 0 43.9-35.4 23.2-56.1c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0c20.7 20.7 56.1 6 56.1-23.2c0-13.3 10.7-24 24-24zM192 256a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm88 32a24 24 0 1 0 -48 0 24 24 0 1 0 48 0z\"],meteor:[512,512,[9732],\"f753\",\"M493.7 .9L299.4 75.6l2.3-29.3c1-12.8-12.8-21.5-24-15.1L101.3 133.4C38.6 169.7 0 236.6 0 309C0 421.1 90.9 512 203 512c72.4 0 139.4-38.6 175.7-101.3L480.8 234.3c6.5-11.1-2.2-25-15.1-24l-29.3 2.3L511.1 18.3c.6-1.5 .9-3.2 .9-4.8C512 6 506 0 498.5 0c-1.7 0-3.3 .3-4.8 .9zM192 192a128 128 0 1 1 0 256 128 128 0 1 1 0-256zm0 96a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm16 96a16 16 0 1 0 0-32 16 16 0 1 0 0 32z\"],\"car-on\":[512,512,[],\"e4dd\",\"M280 24c0-13.3-10.7-24-24-24s-24 10.7-24 24v80c0 13.3 10.7 24 24 24s24-10.7 24-24V24zM185.8 224H326.2c6.8 0 12.8 4.3 15.1 10.6L360.3 288H151.7l19.1-53.4c2.3-6.4 8.3-10.6 15.1-10.6zm-75.3-10.9L82.2 292.4C62.1 300.9 48 320.8 48 344v40 64 32c0 17.7 14.3 32 32 32H96c17.7 0 32-14.3 32-32V448H384v32c0 17.7 14.3 32 32 32h16c17.7 0 32-14.3 32-32V448 384 344c0-23.2-14.1-43.1-34.2-51.6l-28.3-79.3C390.1 181.3 360 160 326.2 160H185.8c-33.8 0-64 21.3-75.3 53.1zM128 344a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm232 24a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zM39 39c-9.4 9.4-9.4 24.6 0 33.9l48 48c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L73 39c-9.4-9.4-24.6-9.4-33.9 0zm400 0L391 87c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l48-48c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0z\"],sleigh:[640,512,[],\"f7cc\",\"M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96V256c0 53 43 96 96 96v32h64V352H384v32h64V352c53 0 96-43 96-96V160c17.7 0 32-14.3 32-32s-14.3-32-32-32H512 480c-17.7 0-32 14.3-32 32v41.3c0 30.2-24.5 54.7-54.7 54.7c-75.5 0-145.6-38.9-185.6-102.9l-4.3-6.9C174.2 67.6 125 37.6 70.7 32.7c-2.2-.5-4.4-.7-6.7-.7H55 32zM640 384c0-17.7-14.3-32-32-32s-32 14.3-32 32v8c0 13.3-10.7 24-24 24H64c-17.7 0-32 14.3-32 32s14.3 32 32 32H552c48.6 0 88-39.4 88-88v-8z\"],\"arrow-down-1-9\":[576,512,[\"sort-numeric-asc\",\"sort-numeric-down\"],\"f162\",\"M450.7 38c-8.3-6-19.1-7.7-28.8-4.4l-48 16c-16.8 5.6-25.8 23.7-20.2 40.5s23.7 25.8 40.5 20.2l5.9-2V160H384c-17.7 0-32 14.3-32 32s14.3 32 32 32h48 48c17.7 0 32-14.3 32-32s-14.3-32-32-32H464V64c0-10.3-4.9-19.9-13.3-26zM160 480c9 0 17.5-3.8 23.6-10.4l88-96c11.9-13 11.1-33.3-2-45.2s-33.3-11.1-45.2 2L192 365.7V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V365.7L95.6 330.4c-11.9-13-32.2-13.9-45.2-2s-13.9 32.2-2 45.2l88 96C142.5 476.2 151 480 160 480zM418.3 307a32 32 0 1 1 27.4 57.9A32 32 0 1 1 418.3 307zM405.1 419.8l-6.8 9.2c-10.5 14.2-7.5 34.2 6.7 44.8s34.2 7.5 44.8-6.7l48.8-65.8c14-18.9 21.5-41.7 21.5-65.2c0-48.6-39.4-88-88-88s-88 39.4-88 88c0 39.2 25.6 72.4 61.1 83.8z\"],\"hand-holding-droplet\":[576,512,[\"hand-holding-water\"],\"f4c1\",\"M275.5 6.6C278.3 2.5 283 0 288 0s9.7 2.5 12.5 6.6L366.8 103C378 119.3 384 138.6 384 158.3V160c0 53-43 96-96 96s-96-43-96-96v-1.7c0-19.8 6-39 17.2-55.3L275.5 6.6zM568.2 336.3c13.1 17.8 9.3 42.8-8.5 55.9L433.1 485.5c-23.4 17.2-51.6 26.5-80.7 26.5H192 32c-17.7 0-32-14.3-32-32V416c0-17.7 14.3-32 32-32H68.8l44.9-36c22.7-18.2 50.9-28 80-28H272h16 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H288 272c-8.8 0-16 7.2-16 16s7.2 16 16 16H392.6l119.7-88.2c17.8-13.1 42.8-9.3 55.9 8.5zM193.6 384l0 0-.9 0c.3 0 .6 0 .9 0z\"],water:[576,512,[],\"f773\",\"M269.5 69.9c11.1-7.9 25.9-7.9 37 0C329 85.4 356.5 96 384 96c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 149.7 417 160 384 160c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4C42.8 92.6 61 83.5 75.3 71.6c11.1-9.5 27.3-10.1 39.2-1.7l0 0C136.7 85.2 165.1 96 192 96c27.5 0 55-10.6 77.5-26.1zm37 288C329 373.4 356.5 384 384 384c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 437.7 417 448 384 448c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 373.2 165.1 384 192 384c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0zm0-144C329 229.4 356.5 240 384 240c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 293.7 417 304 384 304c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.5 27.3-10.1 39.2-1.7l0 0C136.7 229.2 165.1 240 192 240c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z\"],\"calendar-check\":[448,512,[],\"f274\",\"M128 0c17.7 0 32 14.3 32 32V64H288V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64h48c26.5 0 48 21.5 48 48v48H0V112C0 85.5 21.5 64 48 64H96V32c0-17.7 14.3-32 32-32zM0 192H448V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V192zM329 305c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-95 95-47-47c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l64 64c9.4 9.4 24.6 9.4 33.9 0L329 305z\"],braille:[640,512,[],\"f2a1\",\"M0 96a64 64 0 1 1 128 0A64 64 0 1 1 0 96zM224 272a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm0-80a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM80 416a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zM0 416a64 64 0 1 1 128 0A64 64 0 1 1 0 416zm240 0a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm-80 0a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM64 192a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM224 32a64 64 0 1 1 0 128 64 64 0 1 1 0-128zM352 96a64 64 0 1 1 128 0A64 64 0 1 1 352 96zm240 0a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm-80 0a64 64 0 1 1 128 0A64 64 0 1 1 512 96zm64 176a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm0-80a64 64 0 1 1 0 128 64 64 0 1 1 0-128zm16 224a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm-80 0a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM416 272a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm0-80a64 64 0 1 1 0 128 64 64 0 1 1 0-128zm16 224a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm-80 0a64 64 0 1 1 128 0 64 64 0 1 1 -128 0z\"],\"prescription-bottle-medical\":[384,512,[\"prescription-bottle-alt\"],\"f486\",\"M0 32C0 14.3 14.3 0 32 0H352c17.7 0 32 14.3 32 32V64c0 17.7-14.3 32-32 32H32C14.3 96 0 81.7 0 64V32zm32 96H352V448c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V128zM160 240v48H112c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V352h48c8.8 0 16-7.2 16-16V304c0-8.8-7.2-16-16-16H224V240c0-8.8-7.2-16-16-16H176c-8.8 0-16 7.2-16 16z\"],landmark:[512,512,[127963],\"f66f\",\"M240.1 4.2c9.8-5.6 21.9-5.6 31.8 0l171.8 98.1L448 104l0 .9 47.9 27.4c12.6 7.2 18.8 22 15.1 36s-16.4 23.8-30.9 23.8H32c-14.5 0-27.2-9.8-30.9-23.8s2.5-28.8 15.1-36L64 104.9V104l4.4-1.6L240.1 4.2zM64 224h64V416h40V224h64V416h48V224h64V416h40V224h64V420.3c.6 .3 1.2 .7 1.8 1.1l48 32c11.7 7.8 17 22.4 12.9 35.9S494.1 512 480 512H32c-14.1 0-26.5-9.2-30.6-22.7s1.1-28.1 12.9-35.9l48-32c.6-.4 1.2-.7 1.8-1.1V224z\"],truck:[640,512,[128666,9951],\"f0d1\",\"M48 0C21.5 0 0 21.5 0 48V368c0 26.5 21.5 48 48 48H64c0 53 43 96 96 96s96-43 96-96H384c0 53 43 96 96 96s96-43 96-96h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V288 256 237.3c0-17-6.7-33.3-18.7-45.3L512 114.7c-12-12-28.3-18.7-45.3-18.7H416V48c0-26.5-21.5-48-48-48H48zM416 160h50.7L544 237.3V256H416V160zM112 416a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm368-48a48 48 0 1 1 0 96 48 48 0 1 1 0-96z\"],crosshairs:[512,512,[],\"f05b\",\"M256 0c17.7 0 32 14.3 32 32V42.4c93.7 13.9 167.7 88 181.6 181.6H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H469.6c-13.9 93.7-88 167.7-181.6 181.6V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V469.6C130.3 455.7 56.3 381.7 42.4 288H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H42.4C56.3 130.3 130.3 56.3 224 42.4V32c0-17.7 14.3-32 32-32zM107.4 288c12.5 58.3 58.4 104.1 116.6 116.6V384c0-17.7 14.3-32 32-32s32 14.3 32 32v20.6c58.3-12.5 104.1-58.4 116.6-116.6H384c-17.7 0-32-14.3-32-32s14.3-32 32-32h20.6C392.1 165.7 346.3 119.9 288 107.4V128c0 17.7-14.3 32-32 32s-32-14.3-32-32V107.4C165.7 119.9 119.9 165.7 107.4 224H128c17.7 0 32 14.3 32 32s-14.3 32-32 32H107.4zM256 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"person-cane\":[448,512,[],\"e53c\",\"M272 48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zm-8 187.3l47.4 57.1c11.3 13.6 31.5 15.5 45.1 4.2s15.5-31.5 4.2-45.1l-73.7-88.9c-18.2-22-45.3-34.7-73.9-34.7H177.1c-33.7 0-64.9 17.7-82.3 46.6l-58.3 97c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9L120 256.9V480c0 17.7 14.3 32 32 32s32-14.3 32-32V352h16V480c0 17.7 14.3 32 32 32s32-14.3 32-32V235.3zM352 376c0-4.4 3.6-8 8-8s8 3.6 8 8V488c0 13.3 10.7 24 24 24s24-10.7 24-24V376c0-30.9-25.1-56-56-56s-56 25.1-56 56v8c0 13.3 10.7 24 24 24s24-10.7 24-24v-8z\"],tent:[576,512,[],\"e57d\",\"M269.4 6C280.5-2 295.5-2 306.6 6l224 160c7.4 5.3 12.2 13.5 13.2 22.5l32 288c1 9-1.9 18.1-8 24.9s-14.7 10.7-23.8 10.7H464 435.8c-12.1 0-23.2-6.8-28.6-17.7L306.7 293.5c-1.7-3.4-5.1-5.5-8.8-5.5c-5.5 0-9.9 4.4-9.9 9.9V480c0 17.7-14.3 32-32 32H240 32c-9.1 0-17.8-3.9-23.8-10.7s-9-15.8-8-24.9l32-288c1-9 5.8-17.2 13.2-22.5L269.4 6z\"],\"vest-patches\":[448,512,[],\"e086\",\"M151.2 69.7l55.9 167.7-11 33.1c-2.7 8.2-4.1 16.7-4.1 25.3V464c0 14.5 3.9 28.2 10.7 39.9C195 509 185.9 512 176 512H48c-26.5 0-48-21.5-48-48V270.5c0-9.5 2.8-18.7 8.1-26.6l47.9-71.8c5.3-7.9 8.1-17.1 8.1-26.6V128 54.3 48C64 21.5 85.5 0 112 0h4.5c.2 0 .4 0 .6 0c.4 0 .8 0 1.2 0c18.8 0 34.1 9.7 44.1 18.8C171.6 27.2 190.8 40 224 40s52.4-12.8 61.7-21.2C295.7 9.7 311 0 329.7 0c.4 0 .8 0 1.2 0c.2 0 .4 0 .6 0H336c26.5 0 48 21.5 48 48v6.3V128v17.5c0 9.5 2.8 18.7 8.1 26.6l47.9 71.8c5.3 7.9 8.1 17.1 8.1 26.6V464c0 26.5-21.5 48-48 48H272c-26.5 0-48-21.5-48-48V295.8c0-5.2 .8-10.3 2.5-15.2L296.8 69.7C279.4 79.7 255.4 88 224 88s-55.4-8.3-72.8-18.3zM96 456a40 40 0 1 0 0-80 40 40 0 1 0 0 80zM63.5 255.5c-4.7 4.7-4.7 12.3 0 17L79 288 63.5 303.5c-4.7 4.7-4.7 12.3 0 17s12.3 4.7 17 0L96 305l15.5 15.5c4.7 4.7 12.3 4.7 17 0s4.7-12.3 0-17L113 288l15.5-15.5c4.7-4.7 4.7-12.3 0-17s-12.3-4.7-17 0L96 271 80.5 255.5c-4.7-4.7-12.3-4.7-17 0zM304 280v8 32c0 8.8 7.2 16 16 16h32 8c13.3 0 24-10.7 24-24s-10.7-24-24-24h-8v-8c0-13.3-10.7-24-24-24s-24 10.7-24 24z\"],\"check-double\":[448,512,[],\"f560\",\"M342.6 86.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L160 178.7l-57.4-57.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l80 80c12.5 12.5 32.8 12.5 45.3 0l160-160zm96 128c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L160 402.7 54.6 297.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l128 128c12.5 12.5 32.8 12.5 45.3 0l256-256z\"],\"arrow-down-a-z\":[576,512,[\"sort-alpha-asc\",\"sort-alpha-down\"],\"f15d\",\"M183.6 469.6C177.5 476.2 169 480 160 480s-17.5-3.8-23.6-10.4l-88-96c-11.9-13-11.1-33.3 2-45.2s33.3-11.1 45.2 2L128 365.7V64c0-17.7 14.3-32 32-32s32 14.3 32 32V365.7l32.4-35.4c11.9-13 32.2-13.9 45.2-2s13.9 32.2 2 45.2l-88 96zM320 320c0-17.7 14.3-32 32-32H480c12.9 0 24.6 7.8 29.6 19.8s2.2 25.7-6.9 34.9L429.3 416H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H352c-12.9 0-24.6-7.8-29.6-19.8s-2.2-25.7 6.9-34.9L402.7 352H352c-17.7 0-32-14.3-32-32zM416 32c12.1 0 23.2 6.8 28.6 17.7l64 128 16 32c7.9 15.8 1.5 35-14.3 42.9s-35 1.5-42.9-14.3L460.2 224H371.8l-7.2 14.3c-7.9 15.8-27.1 22.2-42.9 14.3s-22.2-27.1-14.3-42.9l16-32 64-128C392.8 38.8 403.9 32 416 32zM395.8 176h40.4L416 135.6 395.8 176z\"],\"money-bill-wheat\":[512,512,[],\"e52a\",\"M176 0c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16c-44.2 0-80-35.8-80-80c0-8.8 7.2-16 16-16zM56 16h48c13.3 0 24 10.7 24 24s-10.7 24-24 24H56C42.7 64 32 53.3 32 40s10.7-24 24-24zM24 88H136c13.3 0 24 10.7 24 24s-10.7 24-24 24H24c-13.3 0-24-10.7-24-24S10.7 88 24 88zm8 96c0-13.3 10.7-24 24-24h48c13.3 0 24 10.7 24 24s-10.7 24-24 24H56c-13.3 0-24-10.7-24-24zM272 16c0-8.8 7.2-16 16-16c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16c-44.2 0-80-35.8-80-80zM400 0c44.2 0 80 35.8 80 80c0 8.8-7.2 16-16 16c-44.2 0-80-35.8-80-80c0-8.8 7.2-16 16-16zm80 144c0 44.2-35.8 80-80 80c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80c8.8 0 16 7.2 16 16zM352 128c8.8 0 16 7.2 16 16c0 44.2-35.8 80-80 80c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80zm-96 16c0 44.2-35.8 80-80 80c-8.8 0-16-7.2-16-16c0-44.2 35.8-80 80-80c8.8 0 16 7.2 16 16zM0 304c0-26.5 21.5-48 48-48H464c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V304zM48 416v48H96c0-26.5-21.5-48-48-48zM96 304H48v48c26.5 0 48-21.5 48-48zM464 416c-26.5 0-48 21.5-48 48h48V416zM416 304c0 26.5 21.5 48 48 48V304H416zm-96 80a64 64 0 1 0 -128 0 64 64 0 1 0 128 0z\"],cookie:[512,512,[127850],\"f563\",\"M247.2 17c-22.1-3.1-44.6 .9-64.4 11.4l-74 39.5C89.1 78.4 73.2 94.9 63.4 115L26.7 190.6c-9.8 20.1-13 42.9-9.1 64.9l14.5 82.8c3.9 22.1 14.6 42.3 30.7 57.9l60.3 58.4c16.1 15.6 36.6 25.6 58.7 28.7l83 11.7c22.1 3.1 44.6-.9 64.4-11.4l74-39.5c19.7-10.5 35.6-27 45.4-47.2l36.7-75.5c9.8-20.1 13-42.9 9.1-64.9l-14.6-82.8c-3.9-22.1-14.6-42.3-30.7-57.9L388.9 57.5c-16.1-15.6-36.6-25.6-58.7-28.7L247.2 17zM208 144a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM144 336a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm224-64a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"arrow-rotate-left\":[512,512,[8634,\"arrow-left-rotate\",\"arrow-rotate-back\",\"arrow-rotate-backward\",\"undo\"],\"f0e2\",\"M125.7 160H176c17.7 0 32 14.3 32 32s-14.3 32-32 32H48c-17.7 0-32-14.3-32-32V64c0-17.7 14.3-32 32-32s32 14.3 32 32v51.2L97.6 97.6c87.5-87.5 229.3-87.5 316.8 0s87.5 229.3 0 316.8s-229.3 87.5-316.8 0c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0c62.5 62.5 163.8 62.5 226.3 0s62.5-163.8 0-226.3s-163.8-62.5-226.3 0L125.7 160z\"],\"hard-drive\":[512,512,[128436,\"hdd\"],\"f0a0\",\"M0 96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V280.4c-17-15.2-39.4-24.4-64-24.4H64c-24.6 0-47 9.2-64 24.4V96zM64 288H448c35.3 0 64 28.7 64 64v64c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V352c0-35.3 28.7-64 64-64zM320 416a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm128-32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z\"],\"face-grin-squint-tears\":[512,512,[129315,\"grin-squint-tears\"],\"f586\",\"M426.8 14.2C446-5 477.5-4.6 497.1 14.9s20 51 .7 70.3c-6.8 6.8-21.4 12.4-37.4 16.7c-16.3 4.4-34.1 7.5-46.3 9.3c-1.6 .2-3.1 .5-4.6 .6c-5.6 .9-10.3-3.9-9.5-9.5c1.6-11.2 4.6-29.6 9-47c.3-1.3 .7-2.6 1-3.9c4.3-15.9 9.8-30.5 16.7-37.4zm-44.7 19c-1.5 4.8-2.9 9.6-4.1 14.3c-4.8 18.9-8 38.5-9.7 50.3c-4 26.8 18.9 49.7 45.7 45.8c11.9-1.6 31.5-4.8 50.4-9.7c4.7-1.2 9.5-2.5 14.3-4.1C534.2 227.5 520.2 353.8 437 437c-83.2 83.2-209.5 97.2-307.2 41.8c1.5-4.8 2.8-9.6 4-14.3c4.8-18.9 8-38.5 9.7-50.3c4-26.8-18.9-49.7-45.7-45.8c-11.9 1.6-31.5 4.8-50.4 9.7c-4.7 1.2-9.5 2.5-14.3 4.1C-22.2 284.5-8.2 158.2 75 75C158.2-8.3 284.5-22.2 382.2 33.2zM51.5 410.1c18.5-5 38.8-8.3 50.9-10c5.6-.9 10.3 3.9 9.5 9.5c-1.7 12.1-5 32.4-10 50.9C97.6 476.4 92 491 85.2 497.8C66 517 34.5 516.6 14.9 497.1s-20-51-.7-70.3c6.8-6.8 21.4-12.4 37.4-16.7zM416.4 202.3c-4.8-11.9-20.9-10.9-26.9 .4c-19.4 36.7-46.3 73.2-79.8 106.7s-70 60.3-106.7 79.8c-11.3 6-12.3 22.1-.4 26.9c59.4 24.1 129.9 12.2 177.9-35.8s59.9-118.5 35.8-177.9zM87.1 285.1c2 2 4.6 3.2 7.3 3.4l56.1 5.1 5.1 56.1c.3 2.8 1.5 5.4 3.4 7.3c6.3 6.3 17.2 3.6 19.8-4.9l29.7-97.4c3.5-11.6-7.3-22.5-19-19L92 265.3c-8.6 2.6-11.3 13.4-4.9 19.8zM265.3 92l-29.7 97.4c-3.5 11.6 7.3 22.5 19 19l97.4-29.7c8.6-2.6 11.3-13.4 4.9-19.8c-2-2-4.6-3.2-7.3-3.4l-56.1-5.1-5.1-56.1c-.3-2.8-1.5-5.4-3.4-7.3c-6.3-6.3-17.2-3.6-19.8 4.9z\"],dumbbell:[640,512,[],\"f44b\",\"M96 64c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V224v64V448c0 17.7-14.3 32-32 32H128c-17.7 0-32-14.3-32-32V384H64c-17.7 0-32-14.3-32-32V288c-17.7 0-32-14.3-32-32s14.3-32 32-32V160c0-17.7 14.3-32 32-32H96V64zm448 0v64h32c17.7 0 32 14.3 32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32v64c0 17.7-14.3 32-32 32H544v64c0 17.7-14.3 32-32 32H480c-17.7 0-32-14.3-32-32V288 224 64c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32zM416 224v64H224V224H416z\"],\"rectangle-list\":[576,512,[\"list-alt\"],\"f022\",\"M0 96C0 60.7 28.7 32 64 32H512c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM128 288a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm32-128a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM128 384a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm96-248c-13.3 0-24 10.7-24 24s10.7 24 24 24H448c13.3 0 24-10.7 24-24s-10.7-24-24-24H224zm0 96c-13.3 0-24 10.7-24 24s10.7 24 24 24H448c13.3 0 24-10.7 24-24s-10.7-24-24-24H224zm0 96c-13.3 0-24 10.7-24 24s10.7 24 24 24H448c13.3 0 24-10.7 24-24s-10.7-24-24-24H224z\"],\"tarp-droplet\":[576,512,[],\"e57c\",\"M288 160c-35.3 0-64-26.9-64-60c0-24 33.7-70.1 52.2-93.5c6.1-7.7 17.5-7.7 23.6 0C318.3 29.9 352 76 352 100c0 33.1-28.7 60-64 60zM64 128H197.5c13.2 37.3 48.7 64 90.5 64s77.4-26.7 90.5-64H512c35.3 0 64 28.7 64 64V352H448c-17.7 0-32 14.3-32 32l0 128L64 512c-35.3 0-64-28.7-64-64V192c0-35.3 28.7-64 64-64zM448 512l0-128H576L448 512zM96 256a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"house-medical-circle-check\":[640,512,[],\"e511\",\"M320 368c0 59.5 29.5 112.1 74.8 144H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L522.1 193.9c-8.5-1.3-17.3-1.9-26.1-1.9c-54.7 0-103.5 24.9-135.8 64H320V208c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16v48H208c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16zm32 0a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm211.3-43.3c-6.2-6.2-16.4-6.2-22.6 0L480 385.4l-28.7-28.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l40 40c6.2 6.2 16.4 6.2 22.6 0l72-72c6.2-6.2 6.2-16.4 0-22.6z\"],\"person-skiing-nordic\":[576,512,[\"skiing-nordic\"],\"f7ca\",\"M336 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM227.2 160c1.9 0 3.8 .1 5.6 .3L201.6 254c-9.3 28 1.7 58.8 26.8 74.5l86.2 53.9L291.3 464H202.8l41.1-88.1-32.4-20.3c-7.8-4.9-14.7-10.7-20.6-17.3L132.2 464H99.7l54.2-257.6c4.6-1.5 9-4.1 12.7-7.8l23.1-23.1c9.9-9.9 23.4-15.5 37.5-15.5zM121.4 198.6c.4 .4 .8 .8 1.3 1.2L67 464H24c-13.3 0-24 10.7-24 24s10.7 24 24 24H159.3c.4 0 .9 0 1.3 0H319.3c.5 0 1 0 1.4 0H504c39.8 0 72-32.2 72-72v-8c0-13.3-10.7-24-24-24s-24 10.7-24 24v8c0 13.3-10.7 24-24 24H434.6l27.6-179.3c10.5-5.2 17.8-16.1 17.8-28.7c0-17.7-14.3-32-32-32H426.7c-12.9 0-24.6-7.8-29.5-19.7l-6.3-15c-14.6-35.1-44.1-61.9-80.5-73.1l-48.7-15c-11.1-3.4-22.7-5.2-34.4-5.2c-31 0-60.8 12.3-82.7 34.3l-23.1 23.1c-12.5 12.5-12.5 32.8 0 45.3zm308 89.4L402.3 464H357.8l21.6-75.6c5.9-20.6-2.6-42.6-20.7-53.9L302 299l30.9-82.4 5.1 12.3C353 264.7 387.9 288 426.7 288h2.7z\"],\"calendar-plus\":[448,512,[],\"f271\",\"M96 32V64H48C21.5 64 0 85.5 0 112v48H448V112c0-26.5-21.5-48-48-48H352V32c0-17.7-14.3-32-32-32s-32 14.3-32 32V64H160V32c0-17.7-14.3-32-32-32S96 14.3 96 32zM448 192H0V464c0 26.5 21.5 48 48 48H400c26.5 0 48-21.5 48-48V192zM224 248c13.3 0 24 10.7 24 24v56h56c13.3 0 24 10.7 24 24s-10.7 24-24 24H248v56c0 13.3-10.7 24-24 24s-24-10.7-24-24V376H144c-13.3 0-24-10.7-24-24s10.7-24 24-24h56V272c0-13.3 10.7-24 24-24z\"],\"plane-arrival\":[640,512,[128748],\"f5af\",\"M.3 166.9L0 68C0 57.7 9.5 50.1 19.5 52.3l35.6 7.9c10.6 2.3 19.2 9.9 23 20L96 128l127.3 37.6L181.8 20.4C178.9 10.2 186.6 0 197.2 0h40.1c11.6 0 22.2 6.2 27.9 16.3l109 193.8 107.2 31.7c15.9 4.7 30.8 12.5 43.7 22.8l34.4 27.6c24 19.2 18.1 57.3-10.7 68.2c-41.2 15.6-86.2 18.1-128.8 7L121.7 289.8c-11.1-2.9-21.2-8.7-29.3-16.9L9.5 189.4c-5.9-6-9.3-14.1-9.3-22.5zM32 448H608c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32zm96-80a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm128-16a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"circle-left\":[512,512,[61840,\"arrow-alt-circle-left\"],\"f359\",\"M512 256A256 256 0 1 0 0 256a256 256 0 1 0 512 0zM217.4 376.9L117.5 269.8c-3.5-3.8-5.5-8.7-5.5-13.8s2-10.1 5.5-13.8l99.9-107.1c4.2-4.5 10.1-7.1 16.3-7.1c12.3 0 22.3 10 22.3 22.3l0 57.7 96 0c17.7 0 32 14.3 32 32l0 32c0 17.7-14.3 32-32 32l-96 0 0 57.7c0 12.3-10 22.3-22.3 22.3c-6.2 0-12.1-2.6-16.3-7.1z\"],\"train-subway\":[448,512,[\"subway\"],\"f239\",\"M96 0C43 0 0 43 0 96V352c0 48 35.2 87.7 81.1 94.9l-46 46C28.1 499.9 33.1 512 43 512H82.7c8.5 0 16.6-3.4 22.6-9.4L160 448H288l54.6 54.6c6 6 14.1 9.4 22.6 9.4H405c10 0 15-12.1 7.9-19.1l-46-46c46-7.1 81.1-46.9 81.1-94.9V96c0-53-43-96-96-96H96zM64 128c0-17.7 14.3-32 32-32h80c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V128zM272 96h80c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H272c-17.7 0-32-14.3-32-32V128c0-17.7 14.3-32 32-32zM64 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm288-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"chart-gantt\":[512,512,[],\"e0e4\",\"M32 32c17.7 0 32 14.3 32 32V400c0 8.8 7.2 16 16 16H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H80c-44.2 0-80-35.8-80-80V64C0 46.3 14.3 32 32 32zm96 96c0-17.7 14.3-32 32-32l96 0c17.7 0 32 14.3 32 32s-14.3 32-32 32H160c-17.7 0-32-14.3-32-32zm96 64H352c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32s14.3-32 32-32zm160 96h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],\"indian-rupee-sign\":[320,512,[\"indian-rupee\",\"inr\"],\"e1bc\",\"M0 64C0 46.3 14.3 32 32 32H96h16H288c17.7 0 32 14.3 32 32s-14.3 32-32 32H231.8c9.6 14.4 16.7 30.6 20.7 48H288c17.7 0 32 14.3 32 32s-14.3 32-32 32H252.4c-13.2 58.3-61.9 103.2-122.2 110.9L274.6 422c14.4 10.3 17.7 30.3 7.4 44.6s-30.3 17.7-44.6 7.4L13.4 314C2.1 306-2.7 291.5 1.5 278.2S18.1 256 32 256h80c32.8 0 61-19.7 73.3-48H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H185.3C173 115.7 144.8 96 112 96H96 32C14.3 96 0 81.7 0 64z\"],\"crop-simple\":[512,512,[\"crop-alt\"],\"f565\",\"M128 32c0-17.7-14.3-32-32-32S64 14.3 64 32V64H32C14.3 64 0 78.3 0 96s14.3 32 32 32H64V384c0 35.3 28.7 64 64 64H352V384H128V32zM384 480c0 17.7 14.3 32 32 32s32-14.3 32-32V448h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H448l0-256c0-35.3-28.7-64-64-64L160 64v64l224 0 0 352z\"],\"money-bill-1\":[576,512,[\"money-bill-alt\"],\"f3d1\",\"M64 64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H64zm64 320H64V320c35.3 0 64 28.7 64 64zM64 192V128h64c0 35.3-28.7 64-64 64zM448 384c0-35.3 28.7-64 64-64v64H448zm64-192c-35.3 0-64-28.7-64-64h64v64zM176 256a112 112 0 1 1 224 0 112 112 0 1 1 -224 0zm76-48c0 9.7 6.9 17.7 16 19.6V276h-4c-11 0-20 9-20 20s9 20 20 20h24 24c11 0 20-9 20-20s-9-20-20-20h-4V208c0-11-9-20-20-20H272c-11 0-20 9-20 20z\"],\"left-long\":[512,512,[\"long-arrow-alt-left\"],\"f30a\",\"M177.5 414c-8.8 3.8-19 2-26-4.6l-144-136C2.7 268.9 0 262.6 0 256s2.7-12.9 7.5-17.4l144-136c7-6.6 17.2-8.4 26-4.6s14.5 12.5 14.5 22l0 72 288 0c17.7 0 32 14.3 32 32l0 64c0 17.7-14.3 32-32 32l-288 0 0 72c0 9.6-5.7 18.2-14.5 22z\"],dna:[448,512,[129516],\"f471\",\"M416 0c17.7 0 32 14.3 32 32c0 59.8-30.3 107.5-69.4 146.6c-28 28-62.5 53.5-97.3 77.4l-2.5 1.7c-11.9 8.1-23.8 16.1-35.5 23.9l0 0 0 0 0 0-1.6 1c-6 4-11.9 7.9-17.8 11.9c-20.9 14-40.8 27.7-59.3 41.5H283.3c-9.8-7.4-20.1-14.7-30.7-22.1l7-4.7 3-2c15.1-10.1 30.9-20.6 46.7-31.6c25 18.1 48.9 37.3 69.4 57.7C417.7 372.5 448 420.2 448 480c0 17.7-14.3 32-32 32s-32-14.3-32-32H64c0 17.7-14.3 32-32 32s-32-14.3-32-32c0-59.8 30.3-107.5 69.4-146.6c28-28 62.5-53.5 97.3-77.4c-34.8-23.9-69.3-49.3-97.3-77.4C30.3 139.5 0 91.8 0 32C0 14.3 14.3 0 32 0S64 14.3 64 32H384c0-17.7 14.3-32 32-32zM338.6 384H109.4c-10.1 10.6-18.6 21.3-25.5 32H364.1c-6.8-10.7-15.3-21.4-25.5-32zM109.4 128H338.6c10.1-10.7 18.6-21.3 25.5-32H83.9c6.8 10.7 15.3 21.3 25.5 32zm55.4 48c18.4 13.8 38.4 27.5 59.3 41.5c20.9-14 40.8-27.7 59.3-41.5H164.7z\"],\"virus-slash\":[640,512,[],\"e075\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-154.3-121c-2-30.1 20.8-60.1 56-60.1H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H532.5c-49.9 0-74.9-60.3-39.6-95.6l8.2-8.2c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-8.2 8.2C412.3 118.4 352 93.4 352 43.5V32c0-17.7-14.3-32-32-32s-32 14.3-32 32V43.5c0 49.9-60.3 74.9-95.6 39.6L184.2 75c-12.5-12.5-32.8-12.5-45.3 0c-1.6 1.6-3.1 3.4-4.3 5.3L38.8 5.1zm225.8 177c6.9-3.9 14.9-6.1 23.4-6.1c26.5 0 48 21.5 48 48c0 4.4-.6 8.7-1.7 12.7l-69.7-54.6zM402 412.7L144.7 210c-9.5 8.5-22.2 14-37.2 14H96c-17.7 0-32 14.3-32 32s14.3 32 32 32h11.5c49.9 0 74.9 60.3 39.6 95.6l-8.2 8.2c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l8.2-8.2c35.3-35.3 95.6-10.3 95.6 39.6V480c0 17.7 14.3 32 32 32s32-14.3 32-32V468.5c0-31.2 23.6-52.7 50-55.7z\"],minus:[448,512,[8211,8722,10134,\"subtract\"],\"f068\",\"M432 256c0 17.7-14.3 32-32 32L48 288c-17.7 0-32-14.3-32-32s14.3-32 32-32l352 0c17.7 0 32 14.3 32 32z\"],chess:[512,512,[],\"f439\",\"M144 16c0-8.8-7.2-16-16-16s-16 7.2-16 16V32H96c-8.8 0-16 7.2-16 16s7.2 16 16 16h16V96H60.2C49.1 96 40 105.1 40 116.2c0 2.5 .5 4.9 1.3 7.3L73.8 208H72c-13.3 0-24 10.7-24 24s10.7 24 24 24h4L60 384H196L180 256h4c13.3 0 24-10.7 24-24s-10.7-24-24-24h-1.8l32.5-84.5c.9-2.3 1.3-4.8 1.3-7.3c0-11.2-9.1-20.2-20.2-20.2H144V64h16c8.8 0 16-7.2 16-16s-7.2-16-16-16H144V16zM48 416L4.8 473.6C1.7 477.8 0 482.8 0 488c0 13.3 10.7 24 24 24H232c13.3 0 24-10.7 24-24c0-5.2-1.7-10.2-4.8-14.4L208 416H48zm288 0l-43.2 57.6c-3.1 4.2-4.8 9.2-4.8 14.4c0 13.3 10.7 24 24 24H488c13.3 0 24-10.7 24-24c0-5.2-1.7-10.2-4.8-14.4L464 416H336zM304 208v51.9c0 7.8 2.8 15.3 8 21.1L339.2 312 337 384H462.5l-3.3-72 28.3-30.8c5.4-5.9 8.5-13.6 8.5-21.7V208c0-8.8-7.2-16-16-16H464c-8.8 0-16 7.2-16 16v16H424V208c0-8.8-7.2-16-16-16H392c-8.8 0-16 7.2-16 16v16H352V208c0-8.8-7.2-16-16-16H320c-8.8 0-16 7.2-16 16zm80 96c0-8.8 7.2-16 16-16s16 7.2 16 16v32H384V304z\"],\"arrow-left-long\":[512,512,[\"long-arrow-left\"],\"f177\",\"M9.4 233.4c-12.5 12.5-12.5 32.8 0 45.3l128 128c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L109.3 288 480 288c17.7 0 32-14.3 32-32s-14.3-32-32-32l-370.7 0 73.4-73.4c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-128 128z\"],\"plug-circle-check\":[576,512,[],\"e55c\",\"M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8c12.3-2.5 24.1-6.4 35.1-11.5c-2.1-10.8-3.1-21.9-3.1-33.3c0-80.3 53.8-148 127.3-169.2c.5-2.2 .7-4.5 .7-6.8c0-17.7-14.3-32-32-32H32zM576 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L416 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z\"],\"street-view\":[512,512,[],\"f21d\",\"M320 64A64 64 0 1 0 192 64a64 64 0 1 0 128 0zm-96 96c-35.3 0-64 28.7-64 64v48c0 17.7 14.3 32 32 32h1.8l11.1 99.5c1.8 16.2 15.5 28.5 31.8 28.5h38.7c16.3 0 30-12.3 31.8-28.5L318.2 304H320c17.7 0 32-14.3 32-32V224c0-35.3-28.7-64-64-64H224zM132.3 394.2c13-2.4 21.7-14.9 19.3-27.9s-14.9-21.7-27.9-19.3c-32.4 5.9-60.9 14.2-82 24.8c-10.5 5.3-20.3 11.7-27.8 19.6C6.4 399.5 0 410.5 0 424c0 21.4 15.5 36.1 29.1 45c14.7 9.6 34.3 17.3 56.4 23.4C130.2 504.7 190.4 512 256 512s125.8-7.3 170.4-19.6c22.1-6.1 41.8-13.8 56.4-23.4c13.7-8.9 29.1-23.6 29.1-45c0-13.5-6.4-24.5-14-32.6c-7.5-7.9-17.3-14.3-27.8-19.6c-21-10.6-49.5-18.9-82-24.8c-13-2.4-25.5 6.3-27.9 19.3s6.3 25.5 19.3 27.9c30.2 5.5 53.7 12.8 69 20.5c3.2 1.6 5.8 3.1 7.9 4.5c3.6 2.4 3.6 7.2 0 9.6c-8.8 5.7-23.1 11.8-43 17.3C374.3 457 318.5 464 256 464s-118.3-7-157.7-17.9c-19.9-5.5-34.2-11.6-43-17.3c-3.6-2.4-3.6-7.2 0-9.6c2.1-1.4 4.8-2.9 7.9-4.5c15.3-7.7 38.8-14.9 69-20.5z\"],\"franc-sign\":[320,512,[],\"e18f\",\"M80 32C62.3 32 48 46.3 48 64V224v96H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H48v64c0 17.7 14.3 32 32 32s32-14.3 32-32V384h80c17.7 0 32-14.3 32-32s-14.3-32-32-32H112V256H256c17.7 0 32-14.3 32-32s-14.3-32-32-32H112V96H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H80z\"],\"volume-off\":[320,512,[],\"f026\",\"M320 64c0-12.6-7.4-24-18.9-29.2s-25-3.1-34.4 5.3L131.8 160H64c-35.3 0-64 28.7-64 64v64c0 35.3 28.7 64 64 64h67.8L266.7 471.9c9.4 8.4 22.9 10.4 34.4 5.3S320 460.6 320 448V64z\"],\"hands-asl-interpreting\":[640,512,[\"american-sign-language-interpreting\",\"asl-interpreting\",\"hands-american-sign-language-interpreting\"],\"f2a3\",\"M156.6 46.3c7.9-15.8 1.5-35-14.3-42.9s-35-1.5-42.9 14.3L13.5 189.4C4.6 207.2 0 226.8 0 246.7V256c0 70.7 57.3 128 128 128h72 8v-.3c35.2-2.7 65.4-22.8 82.1-51.7c8.8-15.3 3.6-34.9-11.7-43.7s-34.9-3.6-43.7 11.7c-7 12-19.9 20-34.7 20c-22.1 0-40-17.9-40-40s17.9-40 40-40c14.8 0 27.7 8 34.7 20c8.8 15.3 28.4 20.5 43.7 11.7s20.5-28.4 11.7-43.7c-12.8-22.1-33.6-39.1-58.4-47.1l80.8-22c17-4.6 27.1-22.2 22.5-39.3s-22.2-27.1-39.3-22.5L194.9 124.6l81.6-68c13.6-11.3 15.4-31.5 4.1-45.1S249.1-3.9 235.5 7.4L133.6 92.3l23-46zM483.4 465.7c-7.9 15.8-1.5 35 14.3 42.9s35 1.5 42.9-14.3l85.9-171.7c8.9-17.8 13.5-37.4 13.5-57.2V256c0-70.7-57.3-128-128-128H440h-8v.3c-35.2 2.7-65.4 22.8-82.1 51.7c-8.9 15.3-3.6 34.9 11.7 43.7s34.9 3.6 43.7-11.7c7-12 19.9-20 34.7-20c22.1 0 40 17.9 40 40s-17.9 40-40 40c-14.8 0-27.7-8-34.7-20c-8.9-15.3-28.4-20.5-43.7-11.7s-20.5 28.4-11.7 43.7c12.8 22.1 33.6 39.1 58.4 47.1l-80.8 22c-17.1 4.7-27.1 22.2-22.5 39.3s22.2 27.1 39.3 22.5l100.7-27.5-81.6 68c-13.6 11.3-15.4 31.5-4.1 45.1s31.5 15.4 45.1 4.1l101.9-84.9-23 46z\"],gear:[512,512,[9881,\"cog\"],\"f013\",\"M495.9 166.6c3.2 8.7 .5 18.4-6.4 24.6l-43.3 39.4c1.1 8.3 1.7 16.8 1.7 25.4s-.6 17.1-1.7 25.4l43.3 39.4c6.9 6.2 9.6 15.9 6.4 24.6c-4.4 11.9-9.7 23.3-15.8 34.3l-4.7 8.1c-6.6 11-14 21.4-22.1 31.2c-5.9 7.2-15.7 9.6-24.5 6.8l-55.7-17.7c-13.4 10.3-28.2 18.9-44 25.4l-12.5 57.1c-2 9.1-9 16.3-18.2 17.8c-13.8 2.3-28 3.5-42.5 3.5s-28.7-1.2-42.5-3.5c-9.2-1.5-16.2-8.7-18.2-17.8l-12.5-57.1c-15.8-6.5-30.6-15.1-44-25.4L83.1 425.9c-8.8 2.8-18.6 .3-24.5-6.8c-8.1-9.8-15.5-20.2-22.1-31.2l-4.7-8.1c-6.1-11-11.4-22.4-15.8-34.3c-3.2-8.7-.5-18.4 6.4-24.6l43.3-39.4C64.6 273.1 64 264.6 64 256s.6-17.1 1.7-25.4L22.4 191.2c-6.9-6.2-9.6-15.9-6.4-24.6c4.4-11.9 9.7-23.3 15.8-34.3l4.7-8.1c6.6-11 14-21.4 22.1-31.2c5.9-7.2 15.7-9.6 24.5-6.8l55.7 17.7c13.4-10.3 28.2-18.9 44-25.4l12.5-57.1c2-9.1 9-16.3 18.2-17.8C227.3 1.2 241.5 0 256 0s28.7 1.2 42.5 3.5c9.2 1.5 16.2 8.7 18.2 17.8l12.5 57.1c15.8 6.5 30.6 15.1 44 25.4l55.7-17.7c8.8-2.8 18.6-.3 24.5 6.8c8.1 9.8 15.5 20.2 22.1 31.2l4.7 8.1c6.1 11 11.4 22.4 15.8 34.3zM256 336a80 80 0 1 0 0-160 80 80 0 1 0 0 160z\"],\"droplet-slash\":[640,512,[\"tint-slash\"],\"f5c7\",\"M320 512c53.2 0 101.4-21.6 136.1-56.6l-298.3-235C140 257.1 128 292.3 128 320c0 106 86 192 192 192zM505.2 370.7c4.4-16.1 6.8-33.1 6.8-50.7c0-91.2-130.2-262.3-166.6-308.3C339.4 4.2 330.5 0 320.9 0h-1.8c-9.6 0-18.5 4.2-24.5 11.7C277.8 33 240.7 81.3 205.8 136L38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L505.2 370.7zM224 336c0 44.2 35.8 80 80 80c8.8 0 16 7.2 16 16s-7.2 16-16 16c-61.9 0-112-50.1-112-112c0-8.8 7.2-16 16-16s16 7.2 16 16z\"],mosque:[640,512,[128332],\"f678\",\"M400 0c5 0 9.8 2.4 12.8 6.4c34.7 46.3 78.1 74.9 133.5 111.5l0 0 0 0c5.2 3.4 10.5 7 16 10.6c28.9 19.2 45.7 51.7 45.7 86.1c0 28.6-11.3 54.5-29.8 73.4H221.8c-18.4-19-29.8-44.9-29.8-73.4c0-34.4 16.7-66.9 45.7-86.1c5.4-3.6 10.8-7.1 16-10.6l0 0 0 0C309.1 81.3 352.5 52.7 387.2 6.4c3-4 7.8-6.4 12.8-6.4zM288 512V440c0-13.3-10.7-24-24-24s-24 10.7-24 24v72H192c-17.7 0-32-14.3-32-32V352c0-17.7 14.3-32 32-32H608c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H560V440c0-13.3-10.7-24-24-24s-24 10.7-24 24v72H448V454c0-19-8.4-37-23-49.2L400 384l-25 20.8C360.4 417 352 435 352 454v58H288zM70.4 5.2c5.7-4.3 13.5-4.3 19.2 0l16 12C139.8 42.9 160 83.2 160 126v2H0v-2C0 83.2 20.2 42.9 54.4 17.2l16-12zM0 160H160V296.6c-19.1 11.1-32 31.7-32 55.4V480c0 9.6 2.1 18.6 5.8 26.8c-6.6 3.4-14 5.2-21.8 5.2H48c-26.5 0-48-21.5-48-48V176 160z\"],mosquito:[640,512,[],\"e52b\",\"M463.7 505.9c9.8-8.9 10.7-24.3 2.1-34.3l-42.1-49 0-54.7c0-5.5-1.8-10.8-5.1-15.1L352 266.3l0-.3L485.4 387.8C542.4 447.6 640 405.2 640 320.6c0-47.9-34-88.3-79.4-94.2l-153-23.9 40.8-40.9c7.8-7.8 9.4-20.1 3.9-29.8L428.5 90.1l38.2-50.9c8-10.6 6.1-25.9-4.3-34.1s-25.2-6.3-33.2 4.4l-48 63.9c-5.9 7.9-6.6 18.6-1.7 27.2L402.2 140 352 190.3l0-38.2c0-14.9-10.2-27.4-24-31l0-57.2c0-4.4-3.6-8-8-8s-8 3.6-8 8l0 57.2c-13.8 3.6-24 16.1-24 31l0 38.1L237.8 140l22.6-39.5c4.9-8.6 4.2-19.3-1.7-27.2l-48-63.9c-8-10.6-22.8-12.6-33.2-4.4s-12.2 23.5-4.3 34.1l38.2 50.9-23.9 41.7c-5.5 9.7-3.9 22 3.9 29.8l40.8 40.9-153 23.9C34 232.3 0 272.7 0 320.6c0 84.6 97.6 127 154.6 67.1L288 266l0 .3-66.5 86.4c-3.3 4.3-5.1 9.6-5.1 15.1l0 54.7-42.1 49c-8.6 10.1-7.7 25.5 2.1 34.3s24.7 7.9 33.4-2.1l48-55.9c3.8-4.4 5.9-10.2 5.9-16.1l0-55.4L288 344.7l0 63.1c0 17.7 14.3 32 32 32s32-14.3 32-32l0-63.1 24.3 31.6 0 55.4c0 5.9 2.1 11.7 5.9 16.1l48 55.9c8.6 10.1 23.6 11 33.4 2.1z\"],\"star-of-david\":[512,512,[10017],\"f69a\",\"M404.2 309.5L383.1 344h42.3l-21.1-34.5zM371.4 256l-54-88H194.6l-54 88 54 88H317.4l54-88zm65.7 0l53.4 87c3.6 5.9 5.5 12.7 5.5 19.6c0 20.7-16.8 37.4-37.4 37.4H348.7l-56.2 91.5C284.8 504.3 270.9 512 256 512s-28.8-7.7-36.6-20.5L163.3 400H53.4C32.8 400 16 383.2 16 362.6c0-6.9 1.9-13.7 5.5-19.6l53.4-87L21.5 169c-3.6-5.9-5.5-12.7-5.5-19.6C16 128.8 32.8 112 53.4 112H163.3l56.2-91.5C227.2 7.7 241.1 0 256 0s28.8 7.7 36.6 20.5L348.7 112H458.6c20.7 0 37.4 16.8 37.4 37.4c0 6.9-1.9 13.7-5.5 19.6l-53.4 87zm-54-88l21.1 34.5L425.4 168H383.1zM283 112L256 68l-27 44h54zM128.9 168H86.6l21.1 34.5L128.9 168zM107.8 309.5L86.6 344h42.3l-21.1-34.5zM229 400l27 44 27-44H229z\"],\"person-military-rifle\":[512,512,[],\"e54b\",\"M160 39c0-13 10-23.8 22.9-24.9L334.7 1.4C344 .7 352 8 352 17.4V48c0 8.8-7.2 16-16 16H185c-13.8 0-25-11.2-25-25zm17.6 57H334.4c1 5.2 1.6 10.5 1.6 16c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-5.5 .6-10.8 1.6-16zm228 364.3L352 369.7V480c0 1.3-.1 2.5-.2 3.8L177.5 234.9c16.6-7.1 34.6-10.9 53.3-10.9h50.4c15.9 0 31.3 2.8 45.8 7.9L421.9 67.7c-7.7-4.4-10.3-14.2-5.9-21.9s14.2-10.3 21.9-5.9l13.9 8 13.9 8c7.7 4.4 10.3 14.2 5.9 21.9L416 173.9l1.6 .9c15.3 8.8 20.6 28.4 11.7 43.7L392.6 282c2 2.8 3.9 5.8 5.7 8.8l76.1 128.8c11.2 19 4.9 43.5-14.1 54.8s-43.5 4.9-54.8-14.1zM320 512H192c-17.7 0-32-14.3-32-32V369.7l-53.6 90.6c-11.2 19-35.8 25.3-54.8 14.1s-25.3-35.8-14.1-54.8l76.1-128.8c9.4-15.8 21.7-29.3 36-40L331.1 510c-3.5 1.3-7.2 2-11.1 2zM296 320a24 24 0 1 0 0-48 24 24 0 1 0 0 48z\"],\"cart-shopping\":[576,512,[128722,\"shopping-cart\"],\"f07a\",\"M0 24C0 10.7 10.7 0 24 0H69.5c22 0 41.5 12.8 50.6 32h411c26.3 0 45.5 25 38.6 50.4l-41 152.3c-8.5 31.4-37 53.3-69.5 53.3H170.7l5.4 28.5c2.2 11.3 12.1 19.5 23.6 19.5H488c13.3 0 24 10.7 24 24s-10.7 24-24 24H199.7c-34.6 0-64.3-24.6-70.7-58.5L77.4 54.5c-.7-3.8-4-6.5-7.9-6.5H24C10.7 48 0 37.3 0 24zM128 464a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm336-48a48 48 0 1 1 0 96 48 48 0 1 1 0-96z\"],vials:[512,512,[],\"f493\",\"M0 64C0 46.3 14.3 32 32 32H88h48 56c17.7 0 32 14.3 32 32s-14.3 32-32 32V400c0 44.2-35.8 80-80 80s-80-35.8-80-80V96C14.3 96 0 81.7 0 64zM136 96H88V256h48V96zM288 64c0-17.7 14.3-32 32-32h56 48 56c17.7 0 32 14.3 32 32s-14.3 32-32 32V400c0 44.2-35.8 80-80 80s-80-35.8-80-80V96c-17.7 0-32-14.3-32-32zM424 96H376V256h48V96z\"],\"plug-circle-plus\":[576,512,[],\"e55f\",\"M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8c12.3-2.5 24.1-6.4 35.1-11.5c-2.1-10.8-3.1-21.9-3.1-33.3c0-80.3 53.8-148 127.3-169.2c.5-2.2 .7-4.5 .7-6.8c0-17.7-14.3-32-32-32H32zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm16-208v48h48c8.8 0 16 7.2 16 16s-7.2 16-16 16H448v48c0 8.8-7.2 16-16 16s-16-7.2-16-16V384H368c-8.8 0-16-7.2-16-16s7.2-16 16-16h48V304c0-8.8 7.2-16 16-16s16 7.2 16 16z\"],\"place-of-worship\":[640,512,[],\"f67f\",\"M224 109.3V217.6L183.3 242c-14.5 8.7-23.3 24.3-23.3 41.2V512h96V416c0-35.3 28.7-64 64-64s64 28.7 64 64v96h96V283.2c0-16.9-8.8-32.5-23.3-41.2L416 217.6V109.3c0-8.5-3.4-16.6-9.4-22.6L331.3 11.3c-6.2-6.2-16.4-6.2-22.6 0L233.4 86.6c-6 6-9.4 14.1-9.4 22.6zM24.9 330.3C9.5 338.8 0 354.9 0 372.4V464c0 26.5 21.5 48 48 48h80V273.6L24.9 330.3zM592 512c26.5 0 48-21.5 48-48V372.4c0-17.5-9.5-33.6-24.9-42.1L512 273.6V512h80z\"],\"grip-vertical\":[320,512,[],\"f58e\",\"M40 352l48 0c22.1 0 40 17.9 40 40l0 48c0 22.1-17.9 40-40 40l-48 0c-22.1 0-40-17.9-40-40l0-48c0-22.1 17.9-40 40-40zm192 0l48 0c22.1 0 40 17.9 40 40l0 48c0 22.1-17.9 40-40 40l-48 0c-22.1 0-40-17.9-40-40l0-48c0-22.1 17.9-40 40-40zM40 320c-22.1 0-40-17.9-40-40l0-48c0-22.1 17.9-40 40-40l48 0c22.1 0 40 17.9 40 40l0 48c0 22.1-17.9 40-40 40l-48 0zM232 192l48 0c22.1 0 40 17.9 40 40l0 48c0 22.1-17.9 40-40 40l-48 0c-22.1 0-40-17.9-40-40l0-48c0-22.1 17.9-40 40-40zM40 160c-22.1 0-40-17.9-40-40L0 72C0 49.9 17.9 32 40 32l48 0c22.1 0 40 17.9 40 40l0 48c0 22.1-17.9 40-40 40l-48 0zM232 32l48 0c22.1 0 40 17.9 40 40l0 48c0 22.1-17.9 40-40 40l-48 0c-22.1 0-40-17.9-40-40l0-48c0-22.1 17.9-40 40-40z\"],\"arrow-turn-up\":[384,512,[\"level-up\"],\"f148\",\"M32 448c-17.7 0-32 14.3-32 32s14.3 32 32 32l96 0c53 0 96-43 96-96l0-306.7 73.4 73.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-128-128c-12.5-12.5-32.8-12.5-45.3 0l-128 128c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L160 109.3 160 416c0 17.7-14.3 32-32 32l-96 0z\"],u:[384,512,[117],\"55\",\"M32 32c17.7 0 32 14.3 32 32V288c0 70.7 57.3 128 128 128s128-57.3 128-128V64c0-17.7 14.3-32 32-32s32 14.3 32 32V288c0 106-86 192-192 192S0 394 0 288V64C0 46.3 14.3 32 32 32z\"],\"square-root-variable\":[576,512,[\"square-root-alt\"],\"f698\",\"M282.6 78.1c8-27.3 33-46.1 61.4-46.1H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H344L238.7 457c-3.6 12.3-14.1 21.2-26.8 22.8s-25.1-4.6-31.5-15.6L77.6 288H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H77.6c22.8 0 43.8 12.1 55.3 31.8l65.2 111.8L282.6 78.1zM393.4 233.4c12.5-12.5 32.8-12.5 45.3 0L480 274.7l41.4-41.4c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3L525.3 320l41.4 41.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L480 365.3l-41.4 41.4c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L434.7 320l-41.4-41.4c-12.5-12.5-12.5-32.8 0-45.3z\"],clock:[512,512,[128339,\"clock-four\"],\"f017\",\"M256 0a256 256 0 1 1 0 512A256 256 0 1 1 256 0zM232 120V256c0 8 4 15.5 10.7 20l96 64c11 7.4 25.9 4.4 33.3-6.7s4.4-25.9-6.7-33.3L280 243.2V120c0-13.3-10.7-24-24-24s-24 10.7-24 24z\"],\"backward-step\":[320,512,[\"step-backward\"],\"f048\",\"M267.5 440.6c9.5 7.9 22.8 9.7 34.1 4.4s18.4-16.6 18.4-29V96c0-12.4-7.2-23.7-18.4-29s-24.5-3.6-34.1 4.4l-192 160L64 241V96c0-17.7-14.3-32-32-32S0 78.3 0 96V416c0 17.7 14.3 32 32 32s32-14.3 32-32V271l11.5 9.6 192 160z\"],pallet:[640,512,[],\"f482\",\"M32 320c-17.7 0-32 14.3-32 32s14.3 32 32 32H64v64H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H96 320 544h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H576V384h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H544 320 96 32zm96 64H288v64H128V384zm224 0H512v64H352V384z\"],faucet:[512,512,[],\"e005\",\"M192 96v12L96 96c-17.7 0-32 14.3-32 32s14.3 32 32 32l96-12 31-3.9 1-.1 1 .1 31 3.9 96 12c17.7 0 32-14.3 32-32s-14.3-32-32-32l-96 12V96c0-17.7-14.3-32-32-32s-32 14.3-32 32zM32 256c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H132.1c20.2 29 53.9 48 91.9 48s71.7-19 91.9-48H352c17.7 0 32 14.3 32 32s14.3 32 32 32h64c17.7 0 32-14.3 32-32c0-88.4-71.6-160-160-160H320l-22.6-22.6c-6-6-14.1-9.4-22.6-9.4H256V180.2l-32-4-32 4V224H173.3c-8.5 0-16.6 3.4-22.6 9.4L128 256H32z\"],\"baseball-bat-ball\":[512,512,[],\"f432\",\"M424 0c-12.4 0-24.2 4.9-33 13.7L233.5 171.2c-10.5 10.5-19.8 22.1-27.7 34.6L132.7 321.6c-7.3 11.5-15.8 22.2-25.5 31.9L69.9 390.7l51.3 51.3 37.3-37.3c9.6-9.6 20.3-18.2 31.9-25.5l115.8-73.1c12.5-7.9 24.1-17.2 34.6-27.7L498.3 121c8.7-8.7 13.7-20.6 13.7-33s-4.9-24.2-13.7-33L457 13.7C448.2 4.9 436.4 0 424 0zm88 432a80 80 0 1 0 -160 0 80 80 0 1 0 160 0zM15 399c-9.4 9.4-9.4 24.6 0 33.9l64 64c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L49 399c-9.4-9.4-24.6-9.4-33.9 0z\"],s:[320,512,[115],\"53\",\"M99.1 105.4C79 114 68.2 127.2 65.2 144.8c-2.4 14.1-.7 23.2 2 29.4c2.8 6.3 7.9 12.4 16.7 18.6c19.2 13.4 48.3 22.1 84.9 32.5c1 .3 1.9 .6 2.9 .8c32.7 9.3 72 20.6 100.9 40.7c15.7 10.9 29.9 25.5 38.6 45.1c8.8 19.8 10.8 42 6.6 66.3c-7.3 42.5-35.3 71.7-71.8 87.3c-35.4 15.2-79.1 17.9-123.7 10.9l-.2 0 0 0c-24-3.9-62.7-17.1-87.6-25.6c-4.8-1.7-9.2-3.1-12.8-4.3C5.1 440.8-3.9 422.7 1.6 405.9s23.7-25.8 40.5-20.3c4.9 1.6 10.2 3.4 15.9 5.4c25.4 8.6 56.4 19.2 74.4 22.1c36.8 5.7 67.5 2.5 88.5-6.5c20.1-8.6 30.8-21.8 33.9-39.4c2.4-14.1 .7-23.2-2-29.4c-2.8-6.3-7.9-12.4-16.7-18.6c-19.2-13.4-48.3-22.1-84.9-32.5c-1-.3-1.9-.6-2.9-.8c-32.7-9.3-72-20.6-100.9-40.7c-15.7-10.9-29.9-25.5-38.6-45.1c-8.8-19.8-10.8-42-6.6-66.3l31.5 5.5L2.1 133.9C9.4 91.4 37.4 62.2 73.9 46.6c35.4-15.2 79.1-17.9 123.7-10.9c13 2 52.4 9.6 66.6 13.4c17.1 4.5 27.2 22.1 22.7 39.2s-22.1 27.2-39.2 22.7c-11.2-3-48.1-10.2-60.1-12l4.9-31.5-4.9 31.5c-36.9-5.8-67.5-2.5-88.6 6.5z\"],timeline:[640,512,[],\"e29c\",\"M128 72a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm32 97.3c28.3-12.3 48-40.5 48-73.3c0-44.2-35.8-80-80-80S48 51.8 48 96c0 32.8 19.7 61 48 73.3V224H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H288v54.7c-28.3 12.3-48 40.5-48 73.3c0 44.2 35.8 80 80 80s80-35.8 80-80c0-32.8-19.7-61-48-73.3V288H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H544V169.3c28.3-12.3 48-40.5 48-73.3c0-44.2-35.8-80-80-80s-80 35.8-80 80c0 32.8 19.7 61 48 73.3V224H160V169.3zM488 96a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zM320 392a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],keyboard:[576,512,[9e3],\"f11c\",\"M64 64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64H64zm16 64h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V144c0-8.8 7.2-16 16-16zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm16 80h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zm80-176c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V144zm16 80h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zM160 336c0-8.8 7.2-16 16-16H400c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V336zM272 128h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V144c0-8.8 7.2-16 16-16zM256 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM368 128h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V144c0-8.8 7.2-16 16-16zM352 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V240zM464 128h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H464c-8.8 0-16-7.2-16-16V144c0-8.8 7.2-16 16-16zM448 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H464c-8.8 0-16-7.2-16-16V240zm16 80h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H464c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16z\"],\"caret-down\":[320,512,[],\"f0d7\",\"M137.4 374.6c12.5 12.5 32.8 12.5 45.3 0l128-128c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8L32 192c-12.9 0-24.6 7.8-29.6 19.8s-2.2 25.7 6.9 34.9l128 128z\"],\"house-chimney-medical\":[576,512,[\"clinic-medical\"],\"f7f2\",\"M575.8 255.5c0 18-15 32.1-32 32.1h-32l.7 160.2c.2 35.5-28.5 64.3-64 64.3H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L416 100.7V64c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V185l52.8 46.4c8 7 12 15 11 24zM272 192c-8.8 0-16 7.2-16 16v48H208c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V320h48c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H320V208c0-8.8-7.2-16-16-16H272z\"],\"temperature-three-quarters\":[320,512,[\"temperature-3\",\"thermometer-3\",\"thermometer-three-quarters\"],\"f2c8\",\"M160 64c-26.5 0-48 21.5-48 48V276.5c0 17.3-7.1 31.9-15.3 42.5C86.2 332.6 80 349.5 80 368c0 44.2 35.8 80 80 80s80-35.8 80-80c0-18.5-6.2-35.4-16.7-48.9c-8.2-10.6-15.3-25.2-15.3-42.5V112c0-26.5-21.5-48-48-48zM48 112C48 50.2 98.1 0 160 0s112 50.1 112 112V276.5c0 .1 .1 .3 .2 .6c.2 .6 .8 1.6 1.7 2.8c18.9 24.4 30.1 55 30.1 88.1c0 79.5-64.5 144-144 144S16 447.5 16 368c0-33.2 11.2-63.8 30.1-88.1c.9-1.2 1.5-2.2 1.7-2.8c.1-.3 .2-.5 .2-.6V112zM208 368c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-20.9 13.4-38.7 32-45.3V144c0-8.8 7.2-16 16-16s16 7.2 16 16V322.7c18.6 6.6 32 24.4 32 45.3z\"],\"mobile-screen\":[384,512,[\"mobile-android-alt\"],\"f3cf\",\"M16 64C16 28.7 44.7 0 80 0H304c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H80c-35.3 0-64-28.7-64-64V64zM144 448c0 8.8 7.2 16 16 16h64c8.8 0 16-7.2 16-16s-7.2-16-16-16H160c-8.8 0-16 7.2-16 16zM304 64H80V384H304V64z\"],\"plane-up\":[512,512,[],\"e22d\",\"M192 93.7C192 59.5 221 0 256 0c36 0 64 59.5 64 93.7l0 66.3L497.8 278.5c8.9 5.9 14.2 15.9 14.2 26.6v56.7c0 10.9-10.7 18.6-21.1 15.2L320 320v80l57.6 43.2c4 3 6.4 7.8 6.4 12.8v42c0 7.8-6.3 14-14 14c-1.3 0-2.6-.2-3.9-.5L256 480 145.9 511.5c-1.3 .4-2.6 .5-3.9 .5c-7.8 0-14-6.3-14-14V456c0-5 2.4-9.8 6.4-12.8L192 400V320L21.1 377C10.7 380.4 0 372.7 0 361.8V305.1c0-10.7 5.3-20.7 14.2-26.6L192 160V93.7z\"],\"piggy-bank\":[576,512,[],\"f4d3\",\"M400 96l0 .7c-5.3-.4-10.6-.7-16-.7H256c-16.5 0-32.5 2.1-47.8 6c-.1-2-.2-4-.2-6c0-53 43-96 96-96s96 43 96 96zm-16 32c3.5 0 7 .1 10.4 .3c4.2 .3 8.4 .7 12.6 1.3C424.6 109.1 450.8 96 480 96h11.5c10.4 0 18 9.8 15.5 19.9l-13.8 55.2c15.8 14.8 28.7 32.8 37.5 52.9H544c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H512c-9.1 12.1-19.9 22.9-32 32v64c0 17.7-14.3 32-32 32H416c-17.7 0-32-14.3-32-32V448H256v32c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32V416c-34.9-26.2-58.7-66.3-63.2-112H68c-37.6 0-68-30.4-68-68s30.4-68 68-68h4c13.3 0 24 10.7 24 24s-10.7 24-24 24H68c-11 0-20 9-20 20s9 20 20 20H99.2c12.1-59.8 57.7-107.5 116.3-122.8c12.9-3.4 26.5-5.2 40.5-5.2H384zm64 136a24 24 0 1 0 -48 0 24 24 0 1 0 48 0z\"],\"battery-half\":[576,512,[\"battery-3\"],\"f242\",\"M464 160c8.8 0 16 7.2 16 16V336c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16H464zM80 96C35.8 96 0 131.8 0 176V336c0 44.2 35.8 80 80 80H464c44.2 0 80-35.8 80-80V320c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32V176c0-44.2-35.8-80-80-80H80zm208 96H96V320H288V192z\"],\"mountain-city\":[640,512,[],\"e52e\",\"M336 0c-26.5 0-48 21.5-48 48v92.1l71.4 118.4c2.5-1.6 5.4-2.5 8.6-2.5h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16h-3.5l73.8 122.4c12.4 20.6 12.9 46.3 1.2 67.3c-.4 .8-.9 1.6-1.4 2.3H592c26.5 0 48-21.5 48-48V240c0-26.5-21.5-48-48-48H568V120c0-13.3-10.7-24-24-24s-24 10.7-24 24v72H480V48c0-26.5-21.5-48-48-48H336zm32 64h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V80c0-8.8 7.2-16 16-16zM352 176c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16V176zm160 96c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H528c-8.8 0-16-7.2-16-16V272zm16 80h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H528c-8.8 0-16-7.2-16-16V368c0-8.8 7.2-16 16-16zM224 188.9L283.8 288H223l-48 64-24.6-41.2L224 188.9zm29.4-44.2C247.1 134.3 236 128 224 128s-23.1 6.3-29.4 16.7L5.1 458.9c-6.5 10.8-6.7 24.3-.7 35.3S22 512 34.5 512H413.5c12.5 0 24-6.8 30.1-17.8s5.8-24.5-.7-35.3L253.4 144.7z\"],coins:[512,512,[],\"f51e\",\"M512 80c0 18-14.3 34.6-38.4 48c-29.1 16.1-72.5 27.5-122.3 30.9c-3.7-1.8-7.4-3.5-11.3-5C300.6 137.4 248.2 128 192 128c-8.3 0-16.4 .2-24.5 .6l-1.1-.6C142.3 114.6 128 98 128 80c0-44.2 86-80 192-80S512 35.8 512 80zM160.7 161.1c10.2-.7 20.7-1.1 31.3-1.1c62.2 0 117.4 12.3 152.5 31.4C369.3 204.9 384 221.7 384 240c0 4-.7 7.9-2.1 11.7c-4.6 13.2-17 25.3-35 35.5c0 0 0 0 0 0c-.1 .1-.3 .1-.4 .2l0 0 0 0c-.3 .2-.6 .3-.9 .5c-35 19.4-90.8 32-153.6 32c-59.6 0-112.9-11.3-148.2-29.1c-1.9-.9-3.7-1.9-5.5-2.9C14.3 274.6 0 258 0 240c0-34.8 53.4-64.5 128-75.4c10.5-1.5 21.4-2.7 32.7-3.5zM416 240c0-21.9-10.6-39.9-24.1-53.4c28.3-4.4 54.2-11.4 76.2-20.5c16.3-6.8 31.5-15.2 43.9-25.5V176c0 19.3-16.5 37.1-43.8 50.9c-14.6 7.4-32.4 13.7-52.4 18.5c.1-1.8 .2-3.5 .2-5.3zm-32 96c0 18-14.3 34.6-38.4 48c-1.8 1-3.6 1.9-5.5 2.9C304.9 404.7 251.6 416 192 416c-62.8 0-118.6-12.6-153.6-32C14.3 370.6 0 354 0 336V300.6c12.5 10.3 27.6 18.7 43.9 25.5C83.4 342.6 135.8 352 192 352s108.6-9.4 148.1-25.9c7.8-3.2 15.3-6.9 22.4-10.9c6.1-3.4 11.8-7.2 17.2-11.2c1.5-1.1 2.9-2.3 4.3-3.4V304v5.7V336zm32 0V304 278.1c19-4.2 36.5-9.5 52.1-16c16.3-6.8 31.5-15.2 43.9-25.5V272c0 10.5-5 21-14.9 30.9c-16.3 16.3-45 29.7-81.3 38.4c.1-1.7 .2-3.5 .2-5.3zM192 448c56.2 0 108.6-9.4 148.1-25.9c16.3-6.8 31.5-15.2 43.9-25.5V432c0 44.2-86 80-192 80S0 476.2 0 432V396.6c12.5 10.3 27.6 18.7 43.9 25.5C83.4 438.6 135.8 448 192 448z\"],khanda:[512,512,[9772],\"f66d\",\"M245.8 3.7c5.9-4.9 14.6-4.9 20.5 0l48 40c5.9 4.9 7.5 13.2 3.8 19.9l0 0 0 0 0 0 0 0-.1 .1-.3 .6c-.3 .5-.7 1.3-1.2 2.3c-1 2-2.6 5-4.4 8.6c-.5 .9-.9 1.9-1.4 2.9C344.9 97.4 368 134 368 176s-23.1 78.6-57.3 97.8c.5 1 1 2 1.4 2.9c1.8 3.7 3.3 6.6 4.4 8.6c.5 1 .9 1.8 1.2 2.3l.3 .6 .1 .1 0 0 0 0c3.6 6.7 2 15-3.8 19.9L272 343.5v19.8l35.6-24.5 41.1-28.2c42.8-29.4 68.4-78 68.4-130c0-31.1-9.2-61.6-26.5-87.5l-2.8-4.2c-4-6-3.5-14 1.3-19.5s12.7-7 19.2-3.7L401.1 80c7.2-14.3 7.2-14.3 7.2-14.3l0 0 0 0 .1 0 .3 .2 1 .5c.8 .4 2 1.1 3.5 1.9c2.9 1.7 7 4.1 11.8 7.3c9.6 6.4 22.5 16.1 35.4 29c25.7 25.7 52.7 65.6 52.7 119.3c0 53.1-26.4 100.5-51.2 133.6c-12.6 16.7-25.1 30.3-34.5 39.7c-4.7 4.7-8.7 8.4-11.5 10.9c-1.4 1.3-2.5 2.2-3.3 2.9l-.9 .8-.3 .2-.1 .1 0 0 0 0s0 0-10.2-12.3l10.2 12.3c-5.1 4.3-12.4 4.9-18.2 1.6l-75.6-43-32.7 22.5 45.5 31.3c1.8-.4 3.7-.7 5.7-.7c13.3 0 24 10.7 24 24s-10.7 24-24 24c-12.2 0-22.3-9.1-23.8-21L272 423.4v28.9c9.6 5.5 16 15.9 16 27.7c0 17.7-14.3 32-32 32s-32-14.3-32-32c0-11.8 6.4-22.2 16-27.7V424.1l-40.3 27.7C197.8 463.3 187.9 472 176 472c-13.3 0-24-10.7-24-24s10.7-24 24-24c2.2 0 4.4 .3 6.5 .9l45.8-31.5-32.7-22.5-75.6 43c-5.8 3.3-13 2.7-18.2-1.6L112 400c-10.2 12.3-10.2 12.3-10.3 12.3l0 0 0 0-.1-.1-.3-.2-.9-.8c-.8-.7-1.9-1.7-3.3-2.9c-2.8-2.5-6.7-6.2-11.5-10.9c-9.4-9.4-21.9-23-34.5-39.7C26.4 324.5 0 277.1 0 224c0-53.7 26.9-93.6 52.7-119.3c12.9-12.9 25.8-22.6 35.4-29C93 72.5 97 70 99.9 68.4c1.5-.8 2.6-1.5 3.5-1.9l1-.5 .3-.2 .1 0 0 0 0 0s0 0 7.2 14.3l-7.2-14.3c6.5-3.2 14.3-1.7 19.2 3.7s5.3 13.4 1.3 19.5l-2.8 4.2C105.2 119 96 149.5 96 180.6c0 51.9 25.6 100.6 68.4 130l41.1 28.2L240 362.6V343.5l-42.2-35.2c-5.9-4.9-7.5-13.2-3.8-19.9l0 0 0 0 0 0 .1-.1 .3-.6c.3-.5 .7-1.3 1.2-2.3c1-2 2.6-5 4.4-8.6c.5-.9 .9-1.9 1.4-2.9C167.1 254.6 144 218 144 176s23.1-78.6 57.3-97.8c-.5-1-1-2-1.4-2.9c-1.8-3.7-3.3-6.6-4.4-8.6c-.5-1-.9-1.8-1.2-2.3l-.3-.6-.1-.1 0 0 0 0 0 0c-3.6-6.7-2-15 3.8-19.9l48-40zM220.2 122.9c-17 11.5-28.2 31-28.2 53.1s11.2 41.6 28.2 53.1C227 210.2 232 190.9 232 176s-5-34.2-11.8-53.1zm71.5 106.2c17-11.5 28.2-31 28.2-53.1s-11.2-41.6-28.2-53.1C285 141.8 280 161.1 280 176s5 34.2 11.8 53.1z\"],sliders:[512,512,[\"sliders-h\"],\"f1de\",\"M0 416c0 17.7 14.3 32 32 32l54.7 0c12.3 28.3 40.5 48 73.3 48s61-19.7 73.3-48L480 448c17.7 0 32-14.3 32-32s-14.3-32-32-32l-246.7 0c-12.3-28.3-40.5-48-73.3-48s-61 19.7-73.3 48L32 384c-17.7 0-32 14.3-32 32zm128 0a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM320 256a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm32-80c-32.8 0-61 19.7-73.3 48L32 224c-17.7 0-32 14.3-32 32s14.3 32 32 32l246.7 0c12.3 28.3 40.5 48 73.3 48s61-19.7 73.3-48l54.7 0c17.7 0 32-14.3 32-32s-14.3-32-32-32l-54.7 0c-12.3-28.3-40.5-48-73.3-48zM192 128a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm73.3-64C253 35.7 224.8 16 192 16s-61 19.7-73.3 48L32 64C14.3 64 0 78.3 0 96s14.3 32 32 32l86.7 0c12.3 28.3 40.5 48 73.3 48s61-19.7 73.3-48L480 128c17.7 0 32-14.3 32-32s-14.3-32-32-32L265.3 64z\"],\"folder-tree\":[576,512,[],\"f802\",\"M64 32C64 14.3 49.7 0 32 0S0 14.3 0 32v96V384c0 35.3 28.7 64 64 64H256V384H64V160H256V96H64V32zM288 192c0 17.7 14.3 32 32 32H544c17.7 0 32-14.3 32-32V64c0-17.7-14.3-32-32-32H445.3c-8.5 0-16.6-3.4-22.6-9.4L409.4 9.4c-6-6-14.1-9.4-22.6-9.4H320c-17.7 0-32 14.3-32 32V192zm0 288c0 17.7 14.3 32 32 32H544c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32H445.3c-8.5 0-16.6-3.4-22.6-9.4l-13.3-13.3c-6-6-14.1-9.4-22.6-9.4H320c-17.7 0-32 14.3-32 32V480z\"],\"network-wired\":[640,512,[],\"f6ff\",\"M256 64H384v64H256V64zM240 0c-26.5 0-48 21.5-48 48v96c0 26.5 21.5 48 48 48h48v32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32h96v32H80c-26.5 0-48 21.5-48 48v96c0 26.5 21.5 48 48 48H240c26.5 0 48-21.5 48-48V368c0-26.5-21.5-48-48-48H192V288H448v32H400c-26.5 0-48 21.5-48 48v96c0 26.5 21.5 48 48 48H560c26.5 0 48-21.5 48-48V368c0-26.5-21.5-48-48-48H512V288h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H352V192h48c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H240zM96 448V384H224v64H96zm320-64H544v64H416V384z\"],\"map-pin\":[320,512,[128205],\"f276\",\"M16 144a144 144 0 1 1 288 0A144 144 0 1 1 16 144zM160 80c8.8 0 16-7.2 16-16s-7.2-16-16-16c-53 0-96 43-96 96c0 8.8 7.2 16 16 16s16-7.2 16-16c0-35.3 28.7-64 64-64zM128 480V317.1c10.4 1.9 21.1 2.9 32 2.9s21.6-1 32-2.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32z\"],hamsa:[512,512,[],\"f665\",\"M34.6 288H80c8.8 0 16-7.2 16-16V72c0-22.1 17.9-40 40-40s40 17.9 40 40V204c0 11 9 20 20 20s20-9 20-20V40c0-22.1 17.9-40 40-40s40 17.9 40 40V204c0 11 9 20 20 20s20-9 20-20V72c0-22.1 17.9-40 40-40s40 17.9 40 40V272c0 8.8 7.2 16 16 16h45.4c19.1 0 34.6 15.5 34.6 34.6c0 8.6-3.2 16.9-9 23.3L416.6 441c-41.1 45.2-99.4 71-160.6 71s-119.4-25.8-160.6-71L9 345.9c-5.8-6.4-9-14.7-9-23.3C0 303.5 15.5 288 34.6 288zM256 288c-38.4 0-76.8 35.8-90.6 50.2c-3.6 3.7-5.4 8.7-5.4 13.8s1.8 10.1 5.4 13.8C179.2 380.2 217.6 416 256 416s76.8-35.8 90.6-50.2c3.6-3.7 5.4-8.7 5.4-13.8s-1.8-10.1-5.4-13.8C332.8 323.8 294.4 288 256 288zm0 32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"cent-sign\":[384,512,[],\"e3f5\",\"M224 0c17.7 0 32 14.3 32 32V66.7c30.9 5.2 59.2 17.7 83.2 35.8c14.1 10.6 17 30.7 6.4 44.8s-30.7 17-44.8 6.4C279.4 137.5 252.9 128 224 128c-70.7 0-128 57.3-128 128s57.3 128 128 128c28.9 0 55.4-9.5 76.8-25.6c14.1-10.6 34.2-7.8 44.8 6.4s7.8 34.2-6.4 44.8c-24 18-52.4 30.6-83.2 35.8V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V445.3C101.2 430.1 32 351.1 32 256s69.2-174.1 160-189.3V32c0-17.7 14.3-32 32-32z\"],flask:[448,512,[],\"f0c3\",\"M288 0H160 128C110.3 0 96 14.3 96 32s14.3 32 32 32V196.8c0 11.8-3.3 23.5-9.5 33.5L10.3 406.2C3.6 417.2 0 429.7 0 442.6C0 480.9 31.1 512 69.4 512H378.6c38.3 0 69.4-31.1 69.4-69.4c0-12.8-3.6-25.4-10.3-36.4L329.5 230.4c-6.2-10.1-9.5-21.7-9.5-33.5V64c17.7 0 32-14.3 32-32s-14.3-32-32-32H288zM192 196.8V64h64V196.8c0 23.7 6.6 46.9 19 67.1L309.5 320h-171L173 263.9c12.4-20.2 19-43.4 19-67.1z\"],\"person-pregnant\":[384,512,[],\"e31e\",\"M192 0a48 48 0 1 1 0 96 48 48 0 1 1 0-96zM120 383c-13.8-3.6-24-16.1-24-31V296.9l-4.6 7.6c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c15-24.9 40.3-41.5 68.7-45.6c4.1-.6 8.2-1 12.5-1h1.1 12.5H192c1.4 0 2.8 .1 4.1 .3c35.7 2.9 65.4 29.3 72.1 65l6.1 32.5c44.3 8.6 77.7 47.5 77.7 94.3v32c0 17.7-14.3 32-32 32H304 264v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384h-8-8v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V383z\"],\"wand-sparkles\":[512,512,[],\"f72b\",\"M464 6.1c9.5-8.5 24-8.1 33 .9l8 8c9 9 9.4 23.5 .9 33l-85.8 95.9c-2.6 2.9-4.1 6.7-4.1 10.7V176c0 8.8-7.2 16-16 16H384.2c-4.6 0-8.9 1.9-11.9 5.3L100.7 500.9C94.3 508 85.3 512 75.8 512c-8.8 0-17.3-3.5-23.5-9.8L9.7 459.7C3.5 453.4 0 445 0 436.2c0-9.5 4-18.5 11.1-24.8l111.6-99.8c3.4-3 5.3-7.4 5.3-11.9V272c0-8.8 7.2-16 16-16h34.6c3.9 0 7.7-1.5 10.7-4.1L464 6.1zM432 288c3.6 0 6.7 2.4 7.7 5.8l14.8 51.7 51.7 14.8c3.4 1 5.8 4.1 5.8 7.7s-2.4 6.7-5.8 7.7l-51.7 14.8-14.8 51.7c-1 3.4-4.1 5.8-7.7 5.8s-6.7-2.4-7.7-5.8l-14.8-51.7-51.7-14.8c-3.4-1-5.8-4.1-5.8-7.7s2.4-6.7 5.8-7.7l51.7-14.8 14.8-51.7c1-3.4 4.1-5.8 7.7-5.8zM87.7 69.8l14.8 51.7 51.7 14.8c3.4 1 5.8 4.1 5.8 7.7s-2.4 6.7-5.8 7.7l-51.7 14.8L87.7 218.2c-1 3.4-4.1 5.8-7.7 5.8s-6.7-2.4-7.7-5.8L57.5 166.5 5.8 151.7c-3.4-1-5.8-4.1-5.8-7.7s2.4-6.7 5.8-7.7l51.7-14.8L72.3 69.8c1-3.4 4.1-5.8 7.7-5.8s6.7 2.4 7.7 5.8zM208 0c3.7 0 6.9 2.5 7.8 6.1l6.8 27.3 27.3 6.8c3.6 .9 6.1 4.1 6.1 7.8s-2.5 6.9-6.1 7.8l-27.3 6.8-6.8 27.3c-.9 3.6-4.1 6.1-7.8 6.1s-6.9-2.5-7.8-6.1l-6.8-27.3-27.3-6.8c-3.6-.9-6.1-4.1-6.1-7.8s2.5-6.9 6.1-7.8l27.3-6.8 6.8-27.3c.9-3.6 4.1-6.1 7.8-6.1z\"],\"ellipsis-vertical\":[128,512,[\"ellipsis-v\"],\"f142\",\"M64 360a56 56 0 1 0 0 112 56 56 0 1 0 0-112zm0-160a56 56 0 1 0 0 112 56 56 0 1 0 0-112zM120 96A56 56 0 1 0 8 96a56 56 0 1 0 112 0z\"],ticket:[576,512,[127903],\"f145\",\"M64 64C28.7 64 0 92.7 0 128v64c0 8.8 7.4 15.7 15.7 18.6C34.5 217.1 48 235 48 256s-13.5 38.9-32.3 45.4C7.4 304.3 0 311.2 0 320v64c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V320c0-8.8-7.4-15.7-15.7-18.6C541.5 294.9 528 277 528 256s13.5-38.9 32.3-45.4c8.3-2.9 15.7-9.8 15.7-18.6V128c0-35.3-28.7-64-64-64H64zm64 112l0 160c0 8.8 7.2 16 16 16H432c8.8 0 16-7.2 16-16V176c0-8.8-7.2-16-16-16H144c-8.8 0-16 7.2-16 16zM96 160c0-17.7 14.3-32 32-32H448c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32H128c-17.7 0-32-14.3-32-32V160z\"],\"power-off\":[512,512,[9211],\"f011\",\"M288 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V256c0 17.7 14.3 32 32 32s32-14.3 32-32V32zM143.5 120.6c13.6-11.3 15.4-31.5 4.1-45.1s-31.5-15.4-45.1-4.1C49.7 115.4 16 181.8 16 256c0 132.5 107.5 240 240 240s240-107.5 240-240c0-74.2-33.8-140.6-86.6-184.6c-13.6-11.3-33.8-9.4-45.1 4.1s-9.4 33.8 4.1 45.1c38.9 32.3 63.5 81 63.5 135.4c0 97.2-78.8 176-176 176s-176-78.8-176-176c0-54.4 24.7-103.1 63.5-135.4z\"],\"right-long\":[512,512,[\"long-arrow-alt-right\"],\"f30b\",\"M334.5 414c8.8 3.8 19 2 26-4.6l144-136c4.8-4.5 7.5-10.8 7.5-17.4s-2.7-12.9-7.5-17.4l-144-136c-7-6.6-17.2-8.4-26-4.6s-14.5 12.5-14.5 22l0 72L32 192c-17.7 0-32 14.3-32 32l0 64c0 17.7 14.3 32 32 32l288 0 0 72c0 9.6 5.7 18.2 14.5 22z\"],\"flag-usa\":[448,512,[],\"f74d\",\"M32 0C49.7 0 64 14.3 64 32V48l69-17.2c38.1-9.5 78.3-5.1 113.5 12.5c46.3 23.2 100.8 23.2 147.1 0l9.6-4.8C423.8 28.1 448 43.1 448 66.1v36.1l-44.7 16.2c-42.8 15.6-90 13.9-131.6-4.6l-16.1-7.2c-20.3-9-41.8-14.7-63.6-16.9v32.2c17.4 2.1 34.4 6.7 50.6 13.9l16.1 7.2c49.2 21.9 105 23.8 155.6 5.4L448 136.3v62l-44.7 16.2c-42.8 15.6-90 13.9-131.6-4.6l-16.1-7.2c-40.2-17.9-85-22.5-128.1-13.3L64 203.1v32.7l70.2-15.1c36.4-7.8 74.3-3.9 108.4 11.3l16.1 7.2c49.2 21.9 105 23.8 155.6 5.4L448 232.3v62l-44.7 16.2c-42.8 15.6-90 13.9-131.6-4.6l-16.1-7.2c-40.2-17.9-85-22.5-128.1-13.3L64 299.1v32.7l70.2-15.1c36.4-7.8 74.3-3.9 108.4 11.3l16.1 7.2c49.2 21.9 105 23.8 155.6 5.4L448 328.3v33.5c0 13.3-8.3 25.3-20.8 30l-34.7 13c-46.2 17.3-97.6 14.6-141.7-7.4c-37.9-19-81.3-23.7-122.5-13.4L64 400v80c0 17.7-14.3 32-32 32s-32-14.3-32-32V416 345.5 312.8 249.5 216.8 153.5 120.8 64 32C0 14.3 14.3 0 32 0zm80 96A16 16 0 1 0 80 96a16 16 0 1 0 32 0zm32 0a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm-32 48a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm32 0a16 16 0 1 0 0-32 16 16 0 1 0 0 32z\"],\"laptop-file\":[640,512,[],\"e51d\",\"M128 0C92.7 0 64 28.7 64 64V288H19.2C8.6 288 0 296.6 0 307.2C0 349.6 34.4 384 76.8 384H320V288H128V64H448V96h64V64c0-35.3-28.7-64-64-64H128zM512 128H400c-26.5 0-48 21.5-48 48V464c0 26.5 21.5 48 48 48H592c26.5 0 48-21.5 48-48V256H544c-17.7 0-32-14.3-32-32V128zm32 0v96h96l-96-96z\"],tty:[512,512,[\"teletype\"],\"f1e4\",\"M38.3 241.3L15.1 200.6c-9.2-16.2-8.4-36.5 4.5-50C61.4 106.8 144.7 48 256 48s194.6 58.8 236.4 102.6c12.9 13.5 13.7 33.8 4.5 50l-23.1 40.7c-7.5 13.2-23.3 19.3-37.8 14.6l-81.1-26.6c-13.1-4.3-22-16.6-22-30.4V144c-49.6-18.1-104-18.1-153.6 0v54.8c0 13.8-8.9 26.1-22 30.4L76.1 255.8c-14.5 4.7-30.3-1.4-37.8-14.6zM32 336c0-8.8 7.2-16 16-16H80c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H48c-8.8 0-16-7.2-16-16V336zm0 96c0-8.8 7.2-16 16-16H80c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H48c-8.8 0-16-7.2-16-16V432zM144 320h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H144c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H240c-8.8 0-16-7.2-16-16V336zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H336c-8.8 0-16-7.2-16-16V336c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H432c-8.8 0-16-7.2-16-16V336zm16 80h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H432c-8.8 0-16-7.2-16-16V432c0-8.8 7.2-16 16-16zM128 432c0-8.8 7.2-16 16-16H368c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H144c-8.8 0-16-7.2-16-16V432z\"],\"diagram-next\":[512,512,[],\"e476\",\"M512 160c0 35.3-28.7 64-64 64H280v64h46.1c21.4 0 32.1 25.9 17 41L273 399c-9.4 9.4-24.6 9.4-33.9 0L169 329c-15.1-15.1-4.4-41 17-41H232V224H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64v64zM448 416V352H365.3l.4-.4c18.4-18.4 20.4-43.7 11-63.6l71.3 0c35.3 0 64 28.7 64 64v64c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64V352c0-35.3 28.7-64 64-64l71.3 0c-9.4 19.9-7.4 45.2 11 63.6l.4 .4H64v64H210.7l5.7 5.7c21.9 21.9 57.3 21.9 79.2 0l5.7-5.7H448z\"],\"person-rifle\":[576,512,[],\"e54e\",\"M265.2 192c25.4 0 49.8 7.1 70.8 19.9V512H144V337.7L90.4 428.3c-11.2 19-35.8 25.3-54.8 14.1s-25.3-35.8-14.1-54.8L97.7 258.8c24.5-41.4 69-66.8 117.1-66.8h50.4zM160 80a80 80 0 1 1 160 0A80 80 0 1 1 160 80zM448 0c8.8 0 16 7.2 16 16V132.3c9.6 5.5 16 15.9 16 27.7V269.3l16-5.3V208c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16v84.5c0 6.9-4.4 13-10.9 15.2L480 325.3V352h48c8.8 0 16 7.2 16 16v16c0 8.8-7.2 16-16 16H484l23 92.1c2.5 10.1-5.1 19.9-15.5 19.9H432c-8.8 0-16-7.2-16-16V400H400c-17.7 0-32-14.3-32-32V224c0-17.7 14.3-32 32-32V160c0-11.8 6.4-22.2 16-27.7V32c-8.8 0-16-7.2-16-16s7.2-16 16-16h16 16z\"],\"house-medical-circle-exclamation\":[640,512,[],\"e512\",\"M320 368c0 59.5 29.5 112.1 74.8 144H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L522.1 193.9c-8.5-1.3-17.3-1.9-26.1-1.9c-54.7 0-103.5 24.9-135.8 64H320V208c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16v48H208c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h48v48c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16zM496 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm0 240a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm0-192c-8.8 0-16 7.2-16 16v80c0 8.8 7.2 16 16 16s16-7.2 16-16V288c0-8.8-7.2-16-16-16z\"],\"closed-captioning\":[576,512,[],\"f20a\",\"M0 96C0 60.7 28.7 32 64 32H512c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zM200 208c14.2 0 27 6.1 35.8 16c8.8 9.9 24 10.7 33.9 1.9s10.7-24 1.9-33.9c-17.5-19.6-43.1-32-71.5-32c-53 0-96 43-96 96s43 96 96 96c28.4 0 54-12.4 71.5-32c8.8-9.9 8-25-1.9-33.9s-25-8-33.9 1.9c-8.8 9.9-21.6 16-35.8 16c-26.5 0-48-21.5-48-48s21.5-48 48-48zm144 48c0-26.5 21.5-48 48-48c14.2 0 27 6.1 35.8 16c8.8 9.9 24 10.7 33.9 1.9s10.7-24 1.9-33.9c-17.5-19.6-43.1-32-71.5-32c-53 0-96 43-96 96s43 96 96 96c28.4 0 54-12.4 71.5-32c8.8-9.9 8-25-1.9-33.9s-25-8-33.9 1.9c-8.8 9.9-21.6 16-35.8 16c-26.5 0-48-21.5-48-48z\"],\"person-hiking\":[384,512,[\"hiking\"],\"f6ec\",\"M192 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm51.3 182.7L224.2 307l49.7 49.7c9 9 14.1 21.2 14.1 33.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V397.3l-73.9-73.9c-15.8-15.8-22.2-38.6-16.9-60.3l20.4-84c8.3-34.1 42.7-54.9 76.7-46.4c19 4.8 35.6 16.4 46.4 32.7L305.1 208H336V184c0-13.3 10.7-24 24-24s24 10.7 24 24v55.8c0 .1 0 .2 0 .2s0 .2 0 .2V488c0 13.3-10.7 24-24 24s-24-10.7-24-24V272H296.6c-16 0-31-8-39.9-21.4l-13.3-20zM81.1 471.9L117.3 334c3 4.2 6.4 8.2 10.1 11.9l41.9 41.9L142.9 488.1c-4.5 17.1-22 27.3-39.1 22.8s-27.3-22-22.8-39.1zm55.5-346L101.4 266.5c-3 12.1-14.9 19.9-27.2 17.9l-47.9-8c-14-2.3-22.9-16.3-19.2-30L31.9 155c9.5-34.8 41.1-59 77.2-59h4.2c15.6 0 27.1 14.7 23.3 29.8z\"],\"venus-double\":[640,512,[9890],\"f226\",\"M192 288a112 112 0 1 0 0-224 112 112 0 1 0 0 224zM368 176c0 86.3-62.1 158.1-144 173.1V384h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H224v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H128c-17.7 0-32-14.3-32-32s14.3-32 32-32h32V349.1C78.1 334.1 16 262.3 16 176C16 78.8 94.8 0 192 0s176 78.8 176 176zM344 318c14.6-15.6 26.8-33.4 36-53c18.8 14.4 42.4 23 68 23c61.9 0 112-50.1 112-112s-50.1-112-112-112c-25.6 0-49.1 8.6-68 23c-9.3-19.5-21.5-37.4-36-53C373.1 12.6 409.1 0 448 0c97.2 0 176 78.8 176 176c0 86.3-62.1 158.1-144 173.1V384h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H480v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448H384c-17.7 0-32-14.3-32-32s14.3-32 32-32h32V349.1c-26.6-4.9-51.1-15.7-72-31.1z\"],images:[576,512,[],\"f302\",\"M160 32c-35.3 0-64 28.7-64 64V320c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H160zM396 138.7l96 144c4.9 7.4 5.4 16.8 1.2 24.6S480.9 320 472 320H328 280 200c-9.2 0-17.6-5.3-21.6-13.6s-2.9-18.2 2.9-25.4l64-80c4.6-5.7 11.4-9 18.7-9s14.2 3.3 18.7 9l17.3 21.6 56-84C360.5 132 368 128 376 128s15.5 4 20 10.7zM192 128a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM48 120c0-13.3-10.7-24-24-24S0 106.7 0 120V344c0 75.1 60.9 136 136 136H456c13.3 0 24-10.7 24-24s-10.7-24-24-24H136c-48.6 0-88-39.4-88-88V120z\"],calculator:[384,512,[128425],\"f1ec\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zM96 64H288c17.7 0 32 14.3 32 32v32c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V96c0-17.7 14.3-32 32-32zm32 160a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zM96 352a32 32 0 1 1 0-64 32 32 0 1 1 0 64zM64 416c0-17.7 14.3-32 32-32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H96c-17.7 0-32-14.3-32-32zM192 256a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm32 64a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm64-64a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm32 64a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zM288 448a32 32 0 1 1 0-64 32 32 0 1 1 0 64z\"],\"people-pulling\":[576,512,[],\"e535\",\"M80 96A48 48 0 1 0 80 0a48 48 0 1 0 0 96zM64 128c-35.3 0-64 28.7-64 64V320c0 17.7 14.3 32 32 32c9.8 0 18.5-4.4 24.4-11.2L80.4 485.3c2.9 17.4 19.4 29.2 36.8 26.3s29.2-19.4 26.3-36.8L123.1 352h15.7l30 134.9c3.8 17.3 20.9 28.1 38.2 24.3s28.1-20.9 24.3-38.2l-57.3-258 116.3 53.8c.5 .3 1.1 .5 1.6 .7c8.6 3.6 18 3.1 25.9-.7c3.4-1.6 6.6-3.9 9.3-6.7c3.1-3.2 5.5-7 7.1-11.4c.1-.3 .2-.7 .3-1l2.5-7.5c5.7-17.1 18.3-30.9 34.7-38.2l8-3.5c1-.4 1.9-.8 2.9-1.2l-16.9 63.5c-5.6 21.1-.1 43.6 14.7 59.7l70.7 77.1 22 88.1c4.3 17.1 21.7 27.6 38.8 23.3s27.6-21.7 23.3-38.8l-23-92.1c-1.9-7.8-5.8-14.9-11.2-20.8l-49.5-54 19.3-65.5 9.6 23c4.4 10.6 12.5 19.3 22.8 24.5l26.7 13.3c15.8 7.9 35 1.5 42.9-14.3s1.5-35-14.3-42.9L537 232.7l-15.3-36.8C504.5 154.8 464.3 128 419.7 128c-22.8 0-45.3 4.8-66.1 14l-8 3.5c-24.4 10.9-44.6 29-58.1 51.6L157.3 136.9C144.7 131 130.9 128 117 128H64zM464 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM349.7 335.6l-25 62.4-59.4 59.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L372.3 441c4.6-4.6 8.2-10.1 10.6-16.1l14.5-36.2-40.7-44.4c-2.5-2.7-4.8-5.6-7-8.6z\"],n:[384,512,[110],\"4e\",\"M21.1 33.9c12.7-4.6 26.9-.7 35.5 9.6L320 359.6V64c0-17.7 14.3-32 32-32s32 14.3 32 32V448c0 13.5-8.4 25.5-21.1 30.1s-26.9 .7-35.5-9.6L64 152.4V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V64C0 50.5 8.4 38.5 21.1 33.9z\"],\"cable-car\":[512,512,[128673,57551,\"tram\"],\"f7da\",\"M288 0a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM160 56a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM32 288c0-35.3 28.7-64 64-64H232V157.5l-203.1 42c-13 2.7-25.7-5.7-28.4-18.6s5.7-25.7 18.6-28.4l232-48 232-48c13-2.7 25.7 5.7 28.4 18.6s-5.7 25.7-18.6 28.4L280 147.5V224H416c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V288zm64 0c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h64c8.8 0 16-7.2 16-16V304c0-8.8-7.2-16-16-16H96zm112 16v64c0 8.8 7.2 16 16 16h64c8.8 0 16-7.2 16-16V304c0-8.8-7.2-16-16-16H224c-8.8 0-16 7.2-16 16zm144-16c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h64c8.8 0 16-7.2 16-16V304c0-8.8-7.2-16-16-16H352z\"],\"cloud-rain\":[512,512,[127783,9926],\"f73d\",\"M96 320c-53 0-96-43-96-96c0-42.5 27.6-78.6 65.9-91.2C64.7 126.1 64 119.1 64 112C64 50.1 114.1 0 176 0c43.1 0 80.5 24.3 99.2 60c14.7-17.1 36.5-28 60.8-28c44.2 0 80 35.8 80 80c0 5.5-.6 10.8-1.6 16c.5 0 1.1 0 1.6 0c53 0 96 43 96 96s-43 96-96 96H96zm-6.8 52c1.3-2.5 3.9-4 6.8-4s5.4 1.5 6.8 4l35.1 64.6c4.1 7.5 6.2 15.8 6.2 24.3v3c0 26.5-21.5 48-48 48s-48-21.5-48-48v-3c0-8.5 2.1-16.9 6.2-24.3L89.2 372zm160 0c1.3-2.5 3.9-4 6.8-4s5.4 1.5 6.8 4l35.1 64.6c4.1 7.5 6.2 15.8 6.2 24.3v3c0 26.5-21.5 48-48 48s-48-21.5-48-48v-3c0-8.5 2.1-16.9 6.2-24.3L249.2 372zm124.9 64.6L409.2 372c1.3-2.5 3.9-4 6.8-4s5.4 1.5 6.8 4l35.1 64.6c4.1 7.5 6.2 15.8 6.2 24.3v3c0 26.5-21.5 48-48 48s-48-21.5-48-48v-3c0-8.5 2.1-16.9 6.2-24.3z\"],\"building-circle-xmark\":[640,512,[],\"e4d4\",\"M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c15.1 0 28.5-6.9 37.3-17.8C340.4 462.2 320 417.5 320 368c0-54.7 24.9-103.5 64-135.8V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L518.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L496 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L473.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L496 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z\"],ship:[576,512,[128674],\"f21a\",\"M192 32c0-17.7 14.3-32 32-32H352c17.7 0 32 14.3 32 32V64h48c26.5 0 48 21.5 48 48V240l44.4 14.8c23.1 7.7 29.5 37.5 11.5 53.9l-101 92.6c-16.2 9.4-34.7 15.1-50.9 15.1c-19.6 0-40.8-7.7-59.2-20.3c-22.1-15.5-51.6-15.5-73.7 0c-17.1 11.8-38 20.3-59.2 20.3c-16.2 0-34.7-5.7-50.9-15.1l-101-92.6c-18-16.5-11.6-46.2 11.5-53.9L96 240V112c0-26.5 21.5-48 48-48h48V32zM160 218.7l107.8-35.9c13.1-4.4 27.3-4.4 40.5 0L416 218.7V128H160v90.7zM306.5 421.9C329 437.4 356.5 448 384 448c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 501.7 417 512 384 512c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 437.2 165.1 448 192 448c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z\"],\"arrows-down-to-line\":[576,512,[],\"e4b8\",\"M544 416L32 416c-17.7 0-32 14.3-32 32s14.3 32 32 32l512 0c17.7 0 32-14.3 32-32s-14.3-32-32-32zm22.6-137.4c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L480 274.7 480 64c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 210.7-41.4-41.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l96 96c12.5 12.5 32.8 12.5 45.3 0l96-96zm-320-45.3c-12.5-12.5-32.8-12.5-45.3 0L160 274.7 160 64c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 210.7L54.6 233.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l96 96c12.5 12.5 32.8 12.5 45.3 0l96-96c12.5-12.5 12.5-32.8 0-45.3z\"],download:[512,512,[],\"f019\",\"M288 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V274.7l-73.4-73.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l128 128c12.5 12.5 32.8 12.5 45.3 0l128-128c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L288 274.7V32zM64 352c-35.3 0-64 28.7-64 64v32c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V416c0-35.3-28.7-64-64-64H346.5l-45.3 45.3c-25 25-65.5 25-90.5 0L165.5 352H64zm368 56a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],\"face-grin\":[512,512,[128512,\"grin\"],\"f580\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM388.1 312.8c12.3-3.8 24.3 6.9 19.3 18.7C382.4 390.6 324.2 432 256.3 432s-126.2-41.4-151.1-100.5c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"delete-left\":[576,512,[9003,\"backspace\"],\"f55a\",\"M576 128c0-35.3-28.7-64-64-64H205.3c-17 0-33.3 6.7-45.3 18.7L9.4 233.4c-6 6-9.4 14.1-9.4 22.6s3.4 16.6 9.4 22.6L160 429.3c12 12 28.3 18.7 45.3 18.7H512c35.3 0 64-28.7 64-64V128zM271 175c9.4-9.4 24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9z\"],\"eye-dropper\":[512,512,[\"eye-dropper-empty\",\"eyedropper\"],\"f1fb\",\"M341.6 29.2L240.1 130.8l-9.4-9.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-9.4-9.4L482.8 170.4c39-39 39-102.2 0-141.1s-102.2-39-141.1 0zM55.4 323.3c-15 15-23.4 35.4-23.4 56.6v42.4L5.4 462.2c-8.5 12.7-6.8 29.6 4 40.4s27.7 12.5 40.4 4L89.7 480h42.4c21.2 0 41.6-8.4 56.6-23.4L309.4 335.9l-45.3-45.3L143.4 411.3c-3 3-7.1 4.7-11.3 4.7H96V379.9c0-4.2 1.7-8.3 4.7-11.3L221.4 247.9l-45.3-45.3L55.4 323.3z\"],\"file-circle-check\":[576,512,[],\"e5a0\",\"M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v38.6C310.1 219.5 256 287.4 256 368c0 59.1 29.1 111.3 73.7 143.3c-3.2 .5-6.4 .7-9.7 .7H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zM288 368a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm211.3-43.3c-6.2-6.2-16.4-6.2-22.6 0L416 385.4l-28.7-28.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l40 40c6.2 6.2 16.4 6.2 22.6 0l72-72c6.2-6.2 6.2-16.4 0-22.6z\"],forward:[512,512,[9193],\"f04e\",\"M52.5 440.6c-9.5 7.9-22.8 9.7-34.1 4.4S0 428.4 0 416V96C0 83.6 7.2 72.3 18.4 67s24.5-3.6 34.1 4.4L224 214.3V256v41.7L52.5 440.6zM256 352V256 128 96c0-12.4 7.2-23.7 18.4-29s24.5-3.6 34.1 4.4l192 160c7.3 6.1 11.5 15.1 11.5 24.6s-4.2 18.5-11.5 24.6l-192 160c-9.5 7.9-22.8 9.7-34.1 4.4s-18.4-16.6-18.4-29V352z\"],mobile:[384,512,[128241,\"mobile-android\",\"mobile-phone\"],\"f3ce\",\"M80 0C44.7 0 16 28.7 16 64V448c0 35.3 28.7 64 64 64H304c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H80zm80 432h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H160c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],\"face-meh\":[512,512,[128528,\"meh\"],\"f11a\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM176.4 176a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM160 336H352c8.8 0 16 7.2 16 16s-7.2 16-16 16H160c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],\"align-center\":[448,512,[],\"f037\",\"M352 64c0-17.7-14.3-32-32-32H128c-17.7 0-32 14.3-32 32s14.3 32 32 32H320c17.7 0 32-14.3 32-32zm96 128c0-17.7-14.3-32-32-32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32zM0 448c0 17.7 14.3 32 32 32H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H32c-17.7 0-32 14.3-32 32zM352 320c0-17.7-14.3-32-32-32H128c-17.7 0-32 14.3-32 32s14.3 32 32 32H320c17.7 0 32-14.3 32-32z\"],\"book-skull\":[448,512,[\"book-dead\"],\"f6b7\",\"M0 96C0 43 43 0 96 0H384h32c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384 96c-53 0-96-43-96-96V96zM64 416c0 17.7 14.3 32 32 32H352V384H96c-17.7 0-32 14.3-32 32zM320 112c0-35.3-35.8-64-80-64s-80 28.7-80 64c0 20.9 12.6 39.5 32 51.2V176c0 8.8 7.2 16 16 16h64c8.8 0 16-7.2 16-16V163.2c19.4-11.7 32-30.3 32-51.2zM208 96a16 16 0 1 1 0 32 16 16 0 1 1 0-32zm48 16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zM134.3 209.3c-8.1-3.5-17.5 .3-21 8.4s.3 17.5 8.4 21L199.4 272l-77.7 33.3c-8.1 3.5-11.9 12.9-8.4 21s12.9 11.9 21 8.4L240 289.4l105.7 45.3c8.1 3.5 17.5-.3 21-8.4s-.3-17.5-8.4-21L280.6 272l77.7-33.3c8.1-3.5 11.9-12.9 8.4-21s-12.9-11.9-21-8.4L240 254.6 134.3 209.3z\"],\"id-card\":[576,512,[62147,\"drivers-license\"],\"f2c2\",\"M0 96l576 0c0-35.3-28.7-64-64-64H64C28.7 32 0 60.7 0 96zm0 32V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V128H0zM64 405.3c0-29.5 23.9-53.3 53.3-53.3H234.7c29.5 0 53.3 23.9 53.3 53.3c0 5.9-4.8 10.7-10.7 10.7H74.7c-5.9 0-10.7-4.8-10.7-10.7zM176 192a64 64 0 1 1 0 128 64 64 0 1 1 0-128zm176 16c0-8.8 7.2-16 16-16H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H368c-8.8 0-16-7.2-16-16zm0 64c0-8.8 7.2-16 16-16H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H368c-8.8 0-16-7.2-16-16zm0 64c0-8.8 7.2-16 16-16H496c8.8 0 16 7.2 16 16s-7.2 16-16 16H368c-8.8 0-16-7.2-16-16z\"],outdent:[448,512,[\"dedent\"],\"f03b\",\"M0 64C0 46.3 14.3 32 32 32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 96 0 81.7 0 64zM192 192c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32zm32 96H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H224c-17.7 0-32-14.3-32-32s14.3-32 32-32zM0 448c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32zM.2 268.6c-8.2-6.4-8.2-18.9 0-25.3l101.9-79.3c10.5-8.2 25.8-.7 25.8 12.6V335.3c0 13.3-15.3 20.8-25.8 12.6L.2 268.6z\"],\"heart-circle-exclamation\":[576,512,[],\"e4fe\",\"M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9l2.6-2.4C267.2 438.6 256 404.6 256 368c0-97.2 78.8-176 176-176c28.3 0 55 6.7 78.7 18.5c.9-6.5 1.3-13 1.3-19.6v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z\"],house:[576,512,[127968,63498,63500,\"home\",\"home-alt\",\"home-lg-alt\"],\"f015\",\"M575.8 255.5c0 18-15 32.1-32 32.1h-32l.7 160.2c0 2.7-.2 5.4-.5 8.1V472c0 22.1-17.9 40-40 40H456c-1.1 0-2.2 0-3.3-.1c-1.4 .1-2.8 .1-4.2 .1H416 392c-22.1 0-40-17.9-40-40V448 384c0-17.7-14.3-32-32-32H256c-17.7 0-32 14.3-32 32v64 24c0 22.1-17.9 40-40 40H160 128.1c-1.5 0-3-.1-4.5-.2c-1.2 .1-2.4 .2-3.6 .2H104c-22.1 0-40-17.9-40-40V360c0-.9 0-1.9 .1-2.8V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L564.8 231.5c8 7 12 15 11 24z\"],\"calendar-week\":[448,512,[],\"f784\",\"M128 0c17.7 0 32 14.3 32 32V64H288V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64h48c26.5 0 48 21.5 48 48v48H0V112C0 85.5 21.5 64 48 64H96V32c0-17.7 14.3-32 32-32zM0 192H448V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V192zm80 64c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16H368c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H80z\"],\"laptop-medical\":[640,512,[],\"f812\",\"M64 96c0-35.3 28.7-64 64-64H512c35.3 0 64 28.7 64 64V352H512V96H128V352H64V96zM0 403.2C0 392.6 8.6 384 19.2 384H620.8c10.6 0 19.2 8.6 19.2 19.2c0 42.4-34.4 76.8-76.8 76.8H76.8C34.4 480 0 445.6 0 403.2zM288 160c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H352v48c0 8.8-7.2 16-16 16H304c-8.8 0-16-7.2-16-16V272H240c-8.8 0-16-7.2-16-16V224c0-8.8 7.2-16 16-16h48V160z\"],b:[320,512,[98],\"42\",\"M64 32C28.7 32 0 60.7 0 96V256 416c0 35.3 28.7 64 64 64H192c70.7 0 128-57.3 128-128c0-46.5-24.8-87.3-62-109.7c18.7-22.3 30-51 30-82.3c0-70.7-57.3-128-128-128H64zm96 192H64V96h96c35.3 0 64 28.7 64 64s-28.7 64-64 64zM64 288h96 32c35.3 0 64 28.7 64 64s-28.7 64-64 64H64V288z\"],\"file-medical\":[384,512,[],\"f477\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM160 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H224v48c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V352H112c-8.8 0-16-7.2-16-16V304c0-8.8 7.2-16 16-16h48V240z\"],\"dice-one\":[448,512,[9856],\"f525\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM224 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"kiwi-bird\":[576,512,[],\"f535\",\"M291.2 388.4c31.2-18.8 64.7-36.4 101.1-36.4H448c4.6 0 9.1-.2 13.6-.7l85.3 121.9c4 5.7 11.3 8.2 17.9 6.1s11.2-8.3 11.2-15.3V224c0-70.7-57.3-128-128-128H392.3c-36.4 0-69.9-17.6-101.1-36.4C262.3 42.1 228.3 32 192 32C86 32 0 118 0 224c0 71.1 38.6 133.1 96 166.3V456c0 13.3 10.7 24 24 24s24-10.7 24-24V410c15.3 3.9 31.4 6 48 6c5.4 0 10.7-.2 16-.7V456c0 13.3 10.7 24 24 24s24-10.7 24-24V405.1c12.4-4.4 24.2-10 35.2-16.7zM448 200a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],\"arrow-right-arrow-left\":[448,512,[8644,\"exchange\"],\"f0ec\",\"M438.6 150.6c12.5-12.5 12.5-32.8 0-45.3l-96-96c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L338.7 96 32 96C14.3 96 0 110.3 0 128s14.3 32 32 32l306.7 0-41.4 41.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l96-96zm-333.3 352c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L109.3 416 416 416c17.7 0 32-14.3 32-32s-14.3-32-32-32l-306.7 0 41.4-41.4c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-96 96c-12.5 12.5-12.5 32.8 0 45.3l96 96z\"],\"rotate-right\":[512,512,[\"redo-alt\",\"rotate-forward\"],\"f2f9\",\"M463.5 224H472c13.3 0 24-10.7 24-24V72c0-9.7-5.8-18.5-14.8-22.2s-19.3-1.7-26.2 5.2L413.4 96.6c-87.6-86.5-228.7-86.2-315.8 1c-87.5 87.5-87.5 229.3 0 316.8s229.3 87.5 316.8 0c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0c-62.5 62.5-163.8 62.5-226.3 0s-62.5-163.8 0-226.3c62.2-62.2 162.7-62.5 225.3-1L327 183c-6.9 6.9-8.9 17.2-5.2 26.2s12.5 14.8 22.2 14.8H463.5z\"],utensils:[448,512,[127860,61685,\"cutlery\"],\"f2e7\",\"M416 0C400 0 288 32 288 176V288c0 35.3 28.7 64 64 64h32V480c0 17.7 14.3 32 32 32s32-14.3 32-32V352 240 32c0-17.7-14.3-32-32-32zM64 16C64 7.8 57.9 1 49.7 .1S34.2 4.6 32.4 12.5L2.1 148.8C.7 155.1 0 161.5 0 167.9c0 45.9 35.1 83.6 80 87.7V480c0 17.7 14.3 32 32 32s32-14.3 32-32V255.6c44.9-4.1 80-41.8 80-87.7c0-6.4-.7-12.8-2.1-19.1L191.6 12.5c-1.8-8-9.3-13.3-17.4-12.4S160 7.8 160 16V150.2c0 5.4-4.4 9.8-9.8 9.8c-5.1 0-9.3-3.9-9.8-9L127.9 14.6C127.2 6.3 120.3 0 112 0s-15.2 6.3-15.9 14.6L83.7 151c-.5 5.1-4.7 9-9.8 9c-5.4 0-9.8-4.4-9.8-9.8V16zm48.3 152l-.3 0-.3 0 .3-.7 .3 .7z\"],\"arrow-up-wide-short\":[576,512,[\"sort-amount-up\"],\"f161\",\"M151.6 42.4C145.5 35.8 137 32 128 32s-17.5 3.8-23.6 10.4l-88 96c-11.9 13-11.1 33.3 2 45.2s33.3 11.1 45.2-2L96 146.3V448c0 17.7 14.3 32 32 32s32-14.3 32-32V146.3l32.4 35.4c11.9 13 32.2 13.9 45.2 2s13.9-32.2 2-45.2l-88-96zM320 480h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H320c-17.7 0-32 14.3-32 32s14.3 32 32 32zm0-128h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H320c-17.7 0-32 14.3-32 32s14.3 32 32 32zm0-128H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H320c-17.7 0-32 14.3-32 32s14.3 32 32 32zm0-128H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H320c-17.7 0-32 14.3-32 32s14.3 32 32 32z\"],\"mill-sign\":[384,512,[],\"e1ed\",\"M302.1 42.8c5.9-16.6-2.7-35-19.4-40.9s-35 2.7-40.9 19.4L208 116.1c-5.7 4-11.1 8.5-16 13.5C171.7 108.9 143.3 96 112 96c-19.5 0-37.8 5-53.7 13.7C52.5 101.4 42.9 96 32 96C14.3 96 0 110.3 0 128v80V416c0 17.7 14.3 32 32 32s32-14.3 32-32V208c0-26.5 21.5-48 48-48s48 21.5 48 48v42.5L81.9 469.2c-5.9 16.6 2.7 35 19.4 40.9s35-2.7 40.9-19.4l21.4-60C168.9 441 179.6 448 192 448c17.7 0 32-14.3 32-32V261.5l35.7-100c3.9-1 8.1-1.6 12.3-1.6c26.5 0 48 21.5 48 48V416c0 17.7 14.3 32 32 32s32-14.3 32-32V208c0-58.2-44.3-106-101.1-111.5l19.2-53.8z\"],\"bowl-rice\":[512,512,[],\"e2eb\",\"M176 56c0-13.3 10.7-24 24-24h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H200c-13.3 0-24-10.7-24-24zm24 48h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H200c-13.3 0-24-10.7-24-24s10.7-24 24-24zM56 176H72c13.3 0 24 10.7 24 24s-10.7 24-24 24H56c-13.3 0-24-10.7-24-24s10.7-24 24-24zM0 283.4C0 268.3 12.3 256 27.4 256H484.6c15.1 0 27.4 12.3 27.4 27.4c0 70.5-44.4 130.7-106.7 154.1L403.5 452c-2 16-15.6 28-31.8 28H140.2c-16.1 0-29.8-12-31.8-28l-1.8-14.4C44.4 414.1 0 353.9 0 283.4zM224 200c0-13.3 10.7-24 24-24h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H248c-13.3 0-24-10.7-24-24zm-96 0c0-13.3 10.7-24 24-24h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H152c-13.3 0-24-10.7-24-24zm-24-96h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H104c-13.3 0-24-10.7-24-24s10.7-24 24-24zm216 96c0-13.3 10.7-24 24-24h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H344c-13.3 0-24-10.7-24-24zm-24-96h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H296c-13.3 0-24-10.7-24-24s10.7-24 24-24zm120 96c0-13.3 10.7-24 24-24h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H440c-13.3 0-24-10.7-24-24zm-24-96h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H392c-13.3 0-24-10.7-24-24s10.7-24 24-24zM296 32h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H296c-13.3 0-24-10.7-24-24s10.7-24 24-24z\"],skull:[512,512,[128128],\"f54c\",\"M416 398.9c58.5-41.1 96-104.1 96-174.9C512 100.3 397.4 0 256 0S0 100.3 0 224c0 70.7 37.5 133.8 96 174.9c0 .4 0 .7 0 1.1v64c0 26.5 21.5 48 48 48h48V464c0-8.8 7.2-16 16-16s16 7.2 16 16v48h64V464c0-8.8 7.2-16 16-16s16 7.2 16 16v48h48c26.5 0 48-21.5 48-48V400c0-.4 0-.7 0-1.1zM96 256a64 64 0 1 1 128 0A64 64 0 1 1 96 256zm256-64a64 64 0 1 1 0 128 64 64 0 1 1 0-128z\"],\"tower-broadcast\":[576,512,[\"broadcast-tower\"],\"f519\",\"M80.3 44C69.8 69.9 64 98.2 64 128s5.8 58.1 16.3 84c6.6 16.4-1.3 35-17.7 41.7s-35-1.3-41.7-17.7C7.4 202.6 0 166.1 0 128S7.4 53.4 20.9 20C27.6 3.6 46.2-4.3 62.6 2.3S86.9 27.6 80.3 44zM555.1 20C568.6 53.4 576 89.9 576 128s-7.4 74.6-20.9 108c-6.6 16.4-25.3 24.3-41.7 17.7S489.1 228.4 495.7 212c10.5-25.9 16.3-54.2 16.3-84s-5.8-58.1-16.3-84C489.1 27.6 497 9 513.4 2.3s35 1.3 41.7 17.7zM352 128c0 23.7-12.9 44.4-32 55.4V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V183.4c-19.1-11.1-32-31.7-32-55.4c0-35.3 28.7-64 64-64s64 28.7 64 64zM170.6 76.8C163.8 92.4 160 109.7 160 128s3.8 35.6 10.6 51.2c7.1 16.2-.3 35.1-16.5 42.1s-35.1-.3-42.1-16.5c-10.3-23.6-16-49.6-16-76.8s5.7-53.2 16-76.8c7.1-16.2 25.9-23.6 42.1-16.5s23.6 25.9 16.5 42.1zM464 51.2c10.3 23.6 16 49.6 16 76.8s-5.7 53.2-16 76.8c-7.1 16.2-25.9 23.6-42.1 16.5s-23.6-25.9-16.5-42.1c6.8-15.6 10.6-32.9 10.6-51.2s-3.8-35.6-10.6-51.2c-7.1-16.2 .3-35.1 16.5-42.1s35.1 .3 42.1 16.5z\"],\"truck-pickup\":[640,512,[128763],\"f63c\",\"M368.6 96l76.8 96H288V96h80.6zM224 80V192H64c-17.7 0-32 14.3-32 32v64c-17.7 0-32 14.3-32 32s14.3 32 32 32H65.1c-.7 5.2-1.1 10.6-1.1 16c0 61.9 50.1 112 112 112s112-50.1 112-112c0-5.4-.4-10.8-1.1-16h66.3c-.7 5.2-1.1 10.6-1.1 16c0 61.9 50.1 112 112 112s112-50.1 112-112c0-5.4-.4-10.8-1.1-16H608c17.7 0 32-14.3 32-32s-14.3-32-32-32V224c0-17.7-14.3-32-32-32H527.4L418.6 56c-12.1-15.2-30.5-24-50-24H272c-26.5 0-48 21.5-48 48zm0 288a48 48 0 1 1 -96 0 48 48 0 1 1 96 0zm288 0a48 48 0 1 1 -96 0 48 48 0 1 1 96 0z\"],\"up-long\":[320,512,[\"long-arrow-alt-up\"],\"f30c\",\"M318 177.5c3.8-8.8 2-19-4.6-26l-136-144C172.9 2.7 166.6 0 160 0s-12.9 2.7-17.4 7.5l-136 144c-6.6 7-8.4 17.2-4.6 26S14.4 192 24 192H96l0 288c0 17.7 14.3 32 32 32h64c17.7 0 32-14.3 32-32l0-288h72c9.6 0 18.2-5.7 22-14.5z\"],stop:[384,512,[9209],\"f04d\",\"M0 128C0 92.7 28.7 64 64 64H320c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128z\"],\"code-merge\":[448,512,[],\"f387\",\"M80 56a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm32.4 97.2c28-12.4 47.6-40.5 47.6-73.2c0-44.2-35.8-80-80-80S0 35.8 0 80c0 32.8 19.7 61 48 73.3V358.7C19.7 371 0 399.2 0 432c0 44.2 35.8 80 80 80s80-35.8 80-80c0-32.8-19.7-61-48-73.3V272c26.7 20.1 60 32 96 32h86.7c12.3 28.3 40.5 48 73.3 48c44.2 0 80-35.8 80-80s-35.8-80-80-80c-32.8 0-61 19.7-73.3 48H208c-49.9 0-91-38.1-95.6-86.8zM80 408a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM344 272a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z\"],upload:[512,512,[],\"f093\",\"M288 109.3V352c0 17.7-14.3 32-32 32s-32-14.3-32-32V109.3l-73.4 73.4c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l128-128c12.5-12.5 32.8-12.5 45.3 0l128 128c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L288 109.3zM64 352H192c0 35.3 28.7 64 64 64s64-28.7 64-64H448c35.3 0 64 28.7 64 64v32c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V416c0-35.3 28.7-64 64-64zM432 456a24 24 0 1 0 0-48 24 24 0 1 0 0 48z\"],hurricane:[384,512,[],\"f751\",\"M0 208C0 104.4 75.7 18.5 174.9 2.6C184 1.2 192 8.6 192 17.9V81.2c0 8.4 6.5 15.3 14.7 16.5C307 112.5 384 199 384 303.4c0 103.6-75.7 189.5-174.9 205.4c-9.2 1.5-17.1-5.9-17.1-15.2V430.2c0-8.4-6.5-15.3-14.7-16.5C77 398.9 0 312.4 0 208zm288 48A96 96 0 1 0 96 256a96 96 0 1 0 192 0zm-96-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],mound:[576,512,[],\"e52d\",\"M144.1 179.2C173.8 127.7 228.6 96 288 96s114.2 31.7 143.9 83.2L540.4 368c12.3 21.3-3.1 48-27.7 48H63.3c-24.6 0-40-26.6-27.7-48L144.1 179.2z\"],\"toilet-portable\":[320,512,[],\"e583\",\"M0 32V64H320V32c0-17.7-14.3-32-32-32H32C14.3 0 0 14.3 0 32zM24 96H0v24V488c0 13.3 10.7 24 24 24s24-10.7 24-24v-8H272v8c0 13.3 10.7 24 24 24s24-10.7 24-24V120 96H296 24zM256 240v64c0 8.8-7.2 16-16 16s-16-7.2-16-16V240c0-8.8 7.2-16 16-16s16 7.2 16 16z\"],\"compact-disc\":[512,512,[128191,128192,128440],\"f51f\",\"M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm256 32a32 32 0 1 1 0-64 32 32 0 1 1 0 64zm-96-32a96 96 0 1 0 192 0 96 96 0 1 0 -192 0zM96 240c0-35 17.5-71.1 45.2-98.8S205 96 240 96c8.8 0 16-7.2 16-16s-7.2-16-16-16c-45.4 0-89.2 22.3-121.5 54.5S64 194.6 64 240c0 8.8 7.2 16 16 16s16-7.2 16-16z\"],\"file-arrow-down\":[384,512,[\"file-download\"],\"f56d\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM216 232V334.1l31-31c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-72 72c-9.4 9.4-24.6 9.4-33.9 0l-72-72c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l31 31V232c0-13.3 10.7-24 24-24s24 10.7 24 24z\"],caravan:[640,512,[],\"f8ff\",\"M0 112C0 67.8 35.8 32 80 32H416c88.4 0 160 71.6 160 160V352h32c17.7 0 32 14.3 32 32s-14.3 32-32 32l-32 0H288c0 53-43 96-96 96s-96-43-96-96H80c-44.2 0-80-35.8-80-80V112zM320 352H448V256H416c-8.8 0-16-7.2-16-16s7.2-16 16-16h32V160c0-17.7-14.3-32-32-32H352c-17.7 0-32 14.3-32 32V352zM96 128c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H224c17.7 0 32-14.3 32-32V160c0-17.7-14.3-32-32-32H96zm96 336a48 48 0 1 0 0-96 48 48 0 1 0 0 96z\"],\"shield-cat\":[512,512,[],\"e572\",\"M269.4 2.9C265.2 1 260.7 0 256 0s-9.2 1-13.4 2.9L54.3 82.8c-22 9.3-38.4 31-38.3 57.2c.5 99.2 41.3 280.7 213.6 363.2c16.7 8 36.1 8 52.8 0C454.7 420.7 495.5 239.2 496 140c.1-26.2-16.3-47.9-38.3-57.2L269.4 2.9zM160 154.4c0-5.8 4.7-10.4 10.4-10.4h.2c3.4 0 6.5 1.6 8.5 4.3l40 53.3c3 4 7.8 6.4 12.8 6.4h48c5 0 9.8-2.4 12.8-6.4l40-53.3c2-2.7 5.2-4.3 8.5-4.3h.2c5.8 0 10.4 4.7 10.4 10.4V272c0 53-43 96-96 96s-96-43-96-96V154.4zM216 288a16 16 0 1 0 0-32 16 16 0 1 0 0 32zm96-16a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z\"],bolt:[448,512,[9889,\"zap\"],\"f0e7\",\"M349.4 44.6c5.9-13.7 1.5-29.7-10.6-38.5s-28.6-8-39.9 1.8l-256 224c-10 8.8-13.6 22.9-8.9 35.3S50.7 288 64 288H175.5L98.6 467.4c-5.9 13.7-1.5 29.7 10.6 38.5s28.6 8 39.9-1.8l256-224c10-8.8 13.6-22.9 8.9-35.3s-16.6-20.7-30-20.7H272.5L349.4 44.6z\"],\"glass-water\":[384,512,[],\"e4f4\",\"M32 0C23.1 0 14.6 3.7 8.6 10.2S-.6 25.4 .1 34.3L28.9 437.7c3 41.9 37.8 74.3 79.8 74.3H275.3c42 0 76.8-32.4 79.8-74.3L383.9 34.3c.6-8.9-2.4-17.6-8.5-24.1S360.9 0 352 0H32zM73 156.5L66.4 64H317.6L311 156.5l-24.2 12.1c-19.4 9.7-42.2 9.7-61.6 0c-20.9-10.4-45.5-10.4-66.4 0c-19.4 9.7-42.2 9.7-61.6 0L73 156.5z\"],\"oil-well\":[576,512,[],\"e532\",\"M528.3 61.3c-11.4-42.7-55.3-68-98-56.6L414.9 8.8C397.8 13.4 387.7 31 392.3 48l24.5 91.4L308.5 167.5l-6.3-18.1C297.7 136.6 285.6 128 272 128s-25.7 8.6-30.2 21.4l-13.6 39L96 222.6V184c0-13.3-10.7-24-24-24s-24 10.7-24 24V448H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H406.7L340 257.5l-62.2 16.1L305.3 352H238.7L265 277l-74.6 19.3L137.3 448H96V288.8l337.4-87.5 25.2 94c4.6 17.1 22.1 27.2 39.2 22.6l15.5-4.1c42.7-11.4 68-55.3 56.6-98L528.3 61.3zM205.1 448l11.2-32H327.7l11.2 32H205.1z\"],vault:[576,512,[],\"e2c5\",\"M64 0C28.7 0 0 28.7 0 64V416c0 35.3 28.7 64 64 64H80l16 32h64l16-32H400l16 32h64l16-32h16c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zM224 320a80 80 0 1 0 0-160 80 80 0 1 0 0 160zm0-240a160 160 0 1 1 0 320 160 160 0 1 1 0-320zM480 221.3V336c0 8.8-7.2 16-16 16s-16-7.2-16-16V221.3c-18.6-6.6-32-24.4-32-45.3c0-26.5 21.5-48 48-48s48 21.5 48 48c0 20.9-13.4 38.7-32 45.3z\"],mars:[448,512,[9794],\"f222\",\"M289.8 46.8c3.7-9 12.5-14.8 22.2-14.8H424c13.3 0 24 10.7 24 24V168c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-33.4-33.4L321 204.2c19.5 28.4 31 62.7 31 99.8c0 97.2-78.8 176-176 176S0 401.2 0 304s78.8-176 176-176c37 0 71.4 11.4 99.8 31l52.6-52.6L295 73c-6.9-6.9-8.9-17.2-5.2-26.2zM400 80l0 0h0v0zM176 416a112 112 0 1 0 0-224 112 112 0 1 0 0 224z\"],toilet:[448,512,[128701],\"f7d8\",\"M24 0C10.7 0 0 10.7 0 24S10.7 48 24 48h8V196.9c-1.9 1.4-3.8 2.9-5.6 4.4C10.9 214.5 0 232.9 0 256c0 46.9 14.3 84.1 37 112.5c14.2 17.7 31.1 31.3 48.5 41.8L65.6 469.9c-3.3 9.8-1.6 20.5 4.4 28.8s15.7 13.3 26 13.3H352c10.3 0 19.9-4.9 26-13.3s7.7-19.1 4.4-28.8l-19.8-59.5c17.4-10.5 34.3-24.1 48.5-41.8c22.7-28.4 37-65.5 37-112.5c0-23.1-10.9-41.5-26.4-54.6c-1.8-1.5-3.7-3-5.6-4.4V48h8c13.3 0 24-10.7 24-24s-10.7-24-24-24H24zM384 256.3c0 1-.3 2.6-3.8 5.6c-4.8 4.1-14 9-29.3 13.4C320.5 284 276.1 288 224 288s-96.5-4-126.9-12.8c-15.3-4.4-24.5-9.3-29.3-13.4c-3.5-3-3.8-4.6-3.8-5.6l0-.3 0-.1c0-1 0-2.5 3.8-5.8c4.8-4.1 14-9 29.3-13.4C127.5 228 171.9 224 224 224s96.5 4 126.9 12.8c15.3 4.4 24.5 9.3 29.3 13.4c3.8 3.2 3.8 4.8 3.8 5.8l0 .1 0 .3zM328.2 384l-.2 .5 0-.5h.2zM112 64h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],\"plane-circle-xmark\":[640,512,[],\"e557\",\"M256 0c-35 0-64 59.5-64 93.7v84.6L8.1 283.4c-5 2.8-8.1 8.2-8.1 13.9v65.5c0 10.6 10.2 18.3 20.4 15.4l171.6-49 0 70.9-57.6 43.2c-4 3-6.4 7.8-6.4 12.8v42c0 7.8 6.3 14 14 14c1.3 0 2.6-.2 3.9-.5L256 480l110.1 31.5c1.3 .4 2.6 .5 3.9 .5c6 0 11.1-3.7 13.1-9C344.5 470.7 320 422.2 320 368c0-60.6 30.6-114 77.1-145.6L320 178.3V93.7C320 59.5 292 0 256 0zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L518.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L496 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L473.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L496 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z\"],\"yen-sign\":[320,512,[165,\"cny\",\"jpy\",\"rmb\",\"yen\"],\"f157\",\"M58.6 46.2C48.8 31.5 29 27.6 14.3 37.4S-4.4 67 5.4 81.7L100.2 224H48c-17.7 0-32 14.3-32 32s14.3 32 32 32h80v32H48c-17.7 0-32 14.3-32 32s14.3 32 32 32h80v64c0 17.7 14.3 32 32 32s32-14.3 32-32V384h80c17.7 0 32-14.3 32-32s-14.3-32-32-32H192V288h80c17.7 0 32-14.3 32-32s-14.3-32-32-32H219.8L314.6 81.7c9.8-14.7 5.8-34.6-8.9-44.4s-34.6-5.8-44.4 8.9L160 198.3 58.6 46.2z\"],\"ruble-sign\":[384,512,[8381,\"rouble\",\"rub\",\"ruble\"],\"f158\",\"M96 32C78.3 32 64 46.3 64 64V256H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64v32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64v32c0 17.7 14.3 32 32 32s32-14.3 32-32V416H288c17.7 0 32-14.3 32-32s-14.3-32-32-32H128V320H240c79.5 0 144-64.5 144-144s-64.5-144-144-144H96zM240 256H128V96H240c44.2 0 80 35.8 80 80s-35.8 80-80 80z\"],sun:[512,512,[9728],\"f185\",\"M361.5 1.2c5 2.1 8.6 6.6 9.6 11.9L391 121l107.9 19.8c5.3 1 9.8 4.6 11.9 9.6s1.5 10.7-1.6 15.2L446.9 256l62.3 90.3c3.1 4.5 3.7 10.2 1.6 15.2s-6.6 8.6-11.9 9.6L391 391 371.1 498.9c-1 5.3-4.6 9.8-9.6 11.9s-10.7 1.5-15.2-1.6L256 446.9l-90.3 62.3c-4.5 3.1-10.2 3.7-15.2 1.6s-8.6-6.6-9.6-11.9L121 391 13.1 371.1c-5.3-1-9.8-4.6-11.9-9.6s-1.5-10.7 1.6-15.2L65.1 256 2.8 165.7c-3.1-4.5-3.7-10.2-1.6-15.2s6.6-8.6 11.9-9.6L121 121 140.9 13.1c1-5.3 4.6-9.8 9.6-11.9s10.7-1.5 15.2 1.6L256 65.1 346.3 2.8c4.5-3.1 10.2-3.7 15.2-1.6zM160 256a96 96 0 1 1 192 0 96 96 0 1 1 -192 0zm224 0a128 128 0 1 0 -256 0 128 128 0 1 0 256 0z\"],guitar:[512,512,[],\"f7a6\",\"M465 7c-9.4-9.4-24.6-9.4-33.9 0L383 55c-2.4 2.4-4.3 5.3-5.5 8.5l-15.4 41-77.5 77.6c-45.1-29.4-99.3-30.2-131 1.6c-11 11-18 24.6-21.4 39.6c-3.7 16.6-19.1 30.7-36.1 31.6c-25.6 1.3-49.3 10.7-67.3 28.6C-16 328.4-7.6 409.4 47.5 464.5s136.1 63.5 180.9 18.7c17.9-17.9 27.4-41.7 28.6-67.3c.9-17 15-32.3 31.6-36.1c15-3.4 28.6-10.5 39.6-21.4c31.8-31.8 31-85.9 1.6-131l77.6-77.6 41-15.4c3.2-1.2 6.1-3.1 8.5-5.5l48-48c9.4-9.4 9.4-24.6 0-33.9L465 7zM208 256a48 48 0 1 1 0 96 48 48 0 1 1 0-96z\"],\"face-laugh-wink\":[512,512,[\"laugh-wink\"],\"f59c\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM96.8 314.1c-3.8-13.7 7.4-26.1 21.6-26.1H393.6c14.2 0 25.5 12.4 21.6 26.1C396.2 382 332.1 432 256 432s-140.2-50-159.2-117.9zM144.4 192a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm156.4 25.6c-5.3 7.1-15.3 8.5-22.4 3.2s-8.5-15.3-3.2-22.4c30.4-40.5 91.2-40.5 121.6 0c5.3 7.1 3.9 17.1-3.2 22.4s-17.1 3.9-22.4-3.2c-17.6-23.5-52.8-23.5-70.4 0z\"],\"horse-head\":[640,512,[],\"f7ab\",\"M64 464V316.9c0-108.4 68.3-205.1 170.5-241.3L404.2 15.5C425.6 7.9 448 23.8 448 46.4c0 11-5.5 21.2-14.6 27.3L400 96c48.1 0 91.2 29.8 108.1 74.9l48.6 129.5c11.8 31.4 4.1 66.8-19.6 90.5c-16 16-37.8 25.1-60.5 25.1h-3.4c-26.1 0-50.9-11.6-67.6-31.7l-32.3-38.7c-11.7 4.1-24.2 6.4-37.3 6.4l-.1 0 0 0c-6.3 0-12.5-.5-18.6-1.5c-3.6-.6-7.2-1.4-10.7-2.3l0 0c-28.9-7.8-53.1-26.8-67.8-52.2c-4.4-7.6-14.2-10.3-21.9-5.8s-10.3 14.2-5.8 21.9c24 41.5 68.3 70 119.3 71.9l47.2 70.8c4 6.1 6.2 13.2 6.2 20.4c0 20.3-16.5 36.8-36.8 36.8H112c-26.5 0-48-21.5-48-48zM392 224a24 24 0 1 0 0-48 24 24 0 1 0 0 48z\"],\"bore-hole\":[512,512,[],\"e4c3\",\"M256 0c-17.7 0-32 14.3-32 32V296.6c-19.1 11.1-32 31.7-32 55.4c0 35.3 28.7 64 64 64s64-28.7 64-64c0-23.7-12.9-44.4-32-55.4V32c0-17.7-14.3-32-32-32zM48 128c-26.5 0-48 21.5-48 48V464c0 26.5 21.5 48 48 48H464c26.5 0 48-21.5 48-48V176c0-26.5-21.5-48-48-48H384c-17.7 0-32 14.3-32 32V352c0 53-43 96-96 96s-96-43-96-96V160c0-17.7-14.3-32-32-32H48z\"],industry:[576,512,[],\"f275\",\"M64 32C46.3 32 32 46.3 32 64V304v48 80c0 26.5 21.5 48 48 48H496c26.5 0 48-21.5 48-48V304 152.2c0-18.2-19.4-29.7-35.4-21.1L352 215.4V152.2c0-18.2-19.4-29.7-35.4-21.1L160 215.4V64c0-17.7-14.3-32-32-32H64z\"],\"circle-down\":[512,512,[61466,\"arrow-alt-circle-down\"],\"f358\",\"M256 0a256 256 0 1 0 0 512A256 256 0 1 0 256 0zM376.9 294.6L269.8 394.5c-3.8 3.5-8.7 5.5-13.8 5.5s-10.1-2-13.8-5.5L135.1 294.6c-4.5-4.2-7.1-10.1-7.1-16.3c0-12.3 10-22.3 22.3-22.3l57.7 0 0-96c0-17.7 14.3-32 32-32l32 0c17.7 0 32 14.3 32 32l0 96 57.7 0c12.3 0 22.3 10 22.3 22.3c0 6.2-2.6 12.1-7.1 16.3z\"],\"arrows-turn-to-dots\":[512,512,[],\"e4c1\",\"M249.4 25.4c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3L269.3 96 416 96c53 0 96 43 96 96v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V192c0-17.7-14.3-32-32-32l-146.7 0 25.4 25.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0l-80-80c-12.5-12.5-12.5-32.8 0-45.3l80-80zm13.3 256l80 80c12.5 12.5 12.5 32.8 0 45.3l-80 80c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L242.7 416 96 416c-17.7 0-32 14.3-32 32v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V448c0-53 43-96 96-96l146.7 0-25.4-25.4c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0zM384 384a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM64 192A64 64 0 1 1 64 64a64 64 0 1 1 0 128z\"],\"florin-sign\":[384,512,[],\"e184\",\"M314.7 32c-38.8 0-73.7 23.3-88.6 59.1L170.7 224H64c-17.7 0-32 14.3-32 32s14.3 32 32 32h80L98.9 396.3c-5 11.9-16.6 19.7-29.5 19.7H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H69.3c38.8 0 73.7-23.3 88.6-59.1L213.3 288H320c17.7 0 32-14.3 32-32s-14.3-32-32-32H240l45.1-108.3c5-11.9 16.6-19.7 29.5-19.7H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H314.7z\"],\"arrow-down-short-wide\":[576,512,[\"sort-amount-desc\",\"sort-amount-down-alt\"],\"f884\",\"M151.6 469.6C145.5 476.2 137 480 128 480s-17.5-3.8-23.6-10.4l-88-96c-11.9-13-11.1-33.3 2-45.2s33.3-11.1 45.2 2L96 365.7V64c0-17.7 14.3-32 32-32s32 14.3 32 32V365.7l32.4-35.4c11.9-13 32.2-13.9 45.2-2s13.9 32.2 2 45.2l-88 96zM320 32h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H320c-17.7 0-32-14.3-32-32s14.3-32 32-32zm0 128h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H320c-17.7 0-32-14.3-32-32s14.3-32 32-32zm0 128H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H320c-17.7 0-32-14.3-32-32s14.3-32 32-32zm0 128H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H320c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],\"less-than\":[384,512,[62774],\"3c\",\"M380.6 81.7c7.9 15.8 1.5 35-14.3 42.9L103.6 256 366.3 387.4c15.8 7.9 22.2 27.1 14.3 42.9s-27.1 22.2-42.9 14.3l-320-160C6.8 279.2 0 268.1 0 256s6.8-23.2 17.7-28.6l320-160c15.8-7.9 35-1.5 42.9 14.3z\"],\"angle-down\":[448,512,[8964],\"f107\",\"M201.4 374.6c12.5 12.5 32.8 12.5 45.3 0l160-160c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L224 306.7 86.6 169.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l160 160z\"],\"car-tunnel\":[512,512,[],\"e4de\",\"M256 0C114.6 0 0 114.6 0 256V448c0 35.3 28.7 64 64 64h42.8c-6.6-5.9-10.8-14.4-10.8-24V376c0-20.8 11.3-38.9 28.1-48.6l21-64.7c7.5-23.1 29-38.7 53.3-38.7H313.6c24.3 0 45.8 15.6 53.3 38.7l21 64.7c16.8 9.7 28.2 27.8 28.2 48.6V488c0 9.6-4.2 18.1-10.8 24H448c35.3 0 64-28.7 64-64V256C512 114.6 397.4 0 256 0zM362.8 512c-6.6-5.9-10.8-14.4-10.8-24V448H160v40c0 9.6-4.2 18.1-10.8 24H362.8zM190.8 277.5L177 320H335l-13.8-42.5c-1.1-3.3-4.1-5.5-7.6-5.5H198.4c-3.5 0-6.5 2.2-7.6 5.5zM168 408a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm200-24a24 24 0 1 0 -48 0 24 24 0 1 0 48 0z\"],\"head-side-cough\":[640,512,[],\"e061\",\"M0 224.2C0 100.6 100.2 0 224 0h24c95.2 0 181.2 69.3 197.3 160.2c2.3 13 6.8 25.7 15.1 36l42 52.6c6.2 7.8 9.6 17.4 9.6 27.4c0 24.2-19.6 43.8-43.8 43.8H448v0 32L339.2 365.6c-11 1.4-19.2 10.7-19.2 21.8c0 11.6 9 21.2 20.6 21.9L448 416v16c0 26.5-21.5 48-48 48H320v8c0 13.3-10.7 24-24 24H256v0H96c-17.7 0-32-14.3-32-32V407.3c0-16.7-6.9-32.5-17.1-45.8C16.6 322.4 0 274.1 0 224.2zm352-.2a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM464 384a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm152-24a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM592 480a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zM552 312a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm40-24a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zM552 408a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],\"grip-lines\":[448,512,[],\"f7a4\",\"M32 288c-17.7 0-32 14.3-32 32s14.3 32 32 32l384 0c17.7 0 32-14.3 32-32s-14.3-32-32-32L32 288zm0-128c-17.7 0-32 14.3-32 32s14.3 32 32 32l384 0c17.7 0 32-14.3 32-32s-14.3-32-32-32L32 160z\"],\"thumbs-down\":[512,512,[128078,61576],\"f165\",\"M313.4 479.1c26-5.2 42.9-30.5 37.7-56.5l-2.3-11.4c-5.3-26.7-15.1-52.1-28.8-75.2H464c26.5 0 48-21.5 48-48c0-18.5-10.5-34.6-25.9-42.6C497 236.6 504 223.1 504 208c0-23.4-16.8-42.9-38.9-47.1c4.4-7.3 6.9-15.8 6.9-24.9c0-21.3-13.9-39.4-33.1-45.6c.7-3.3 1.1-6.8 1.1-10.4c0-26.5-21.5-48-48-48H294.5c-19 0-37.5 5.6-53.3 16.1L202.7 73.8C176 91.6 160 121.6 160 153.7V192v48 24.9c0 29.2 13.3 56.7 36 75l7.4 5.9c26.5 21.2 44.6 51 51.2 84.2l2.3 11.4c5.2 26 30.5 42.9 56.5 37.7zM32 384H96c17.7 0 32-14.3 32-32V128c0-17.7-14.3-32-32-32H32C14.3 96 0 110.3 0 128V352c0 17.7 14.3 32 32 32z\"],\"user-lock\":[640,512,[],\"f502\",\"M224 256A128 128 0 1 0 224 0a128 128 0 1 0 0 256zm-45.7 48C79.8 304 0 383.8 0 482.3C0 498.7 13.3 512 29.7 512H392.6c-5.4-9.4-8.6-20.3-8.6-32V352c0-2.1 .1-4.2 .3-6.3c-31-26-71-41.7-114.6-41.7H178.3zM528 240c17.7 0 32 14.3 32 32v48H496V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z\"],\"arrow-right-long\":[512,512,[\"long-arrow-right\"],\"f178\",\"M502.6 278.6c12.5-12.5 12.5-32.8 0-45.3l-128-128c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L402.7 224 32 224c-17.7 0-32 14.3-32 32s14.3 32 32 32l370.7 0-73.4 73.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l128-128z\"],\"anchor-circle-xmark\":[640,512,[],\"e4ac\",\"M320 96a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm21.1 80C367 158.8 384 129.4 384 96c0-53-43-96-96-96s-96 43-96 96c0 33.4 17 62.8 42.9 80H224c-17.7 0-32 14.3-32 32s14.3 32 32 32h32V448H208c-53 0-96-43-96-96v-6.1l7 7c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L97 263c-9.4-9.4-24.6-9.4-33.9 0L7 319c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l7-7V352c0 88.4 71.6 160 160 160h80 80c8.2 0 16.3-.6 24.2-1.8c-22.2-16.2-40.4-37.5-53-62.2H320V368 240h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H341.1zM496 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L518.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L496 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L473.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L496 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z\"],ellipsis:[448,512,[\"ellipsis-h\"],\"f141\",\"M8 256a56 56 0 1 1 112 0A56 56 0 1 1 8 256zm160 0a56 56 0 1 1 112 0 56 56 0 1 1 -112 0zm216-56a56 56 0 1 1 0 112 56 56 0 1 1 0-112z\"],\"chess-pawn\":[320,512,[9823],\"f443\",\"M215.5 224c29.2-18.4 48.5-50.9 48.5-88c0-57.4-46.6-104-104-104S56 78.6 56 136c0 37.1 19.4 69.6 48.5 88H96c-17.7 0-32 14.3-32 32c0 16.5 12.5 30 28.5 31.8L80 400H240L227.5 287.8c16-1.8 28.5-15.3 28.5-31.8c0-17.7-14.3-32-32-32h-8.5zM22.6 473.4c-4.2 4.2-6.6 10-6.6 16C16 501.9 26.1 512 38.6 512H281.4c12.5 0 22.6-10.1 22.6-22.6c0-6-2.4-11.8-6.6-16L256 432H64L22.6 473.4z\"],\"kit-medical\":[576,512,[\"first-aid\"],\"f479\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H96V32H64zm64 0V480H448V32H128zM512 480c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H480V480h32zM256 176c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H320v48c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V288H208c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16h48V176z\"],\"person-through-window\":[640,512,[],\"e5a9\",\"M64 64l224 0 0 9.8c0 39-23.7 74-59.9 88.4C167.6 186.5 128 245 128 310.2l0 73.8s0 0 0 0H64V64zm288 0l224 0V384H508.3l-3.7-4.5-75.2-90.2c-9.1-10.9-22.6-17.3-36.9-17.3l-71.1 0-41-63.1c-.3-.5-.6-1-1-1.4c44.7-29 72.5-79 72.5-133.6l0-9.8zm73 320H379.2l42.7 64H592c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H48C21.5 0 0 21.5 0 48V400c0 26.5 21.5 48 48 48H308.2l33.2 49.8c9.8 14.7 29.7 18.7 44.4 8.9s18.7-29.7 8.9-44.4L310.5 336l74.6 0 40 48zm-159.5 0H192s0 0 0 0l0-73.8c0-10.2 1.6-20.1 4.7-29.5L265.5 384zM192 128a48 48 0 1 0 -96 0 48 48 0 1 0 96 0z\"],toolbox:[512,512,[129520],\"f552\",\"M176 88l0 40 160 0 0-40c0-4.4-3.6-8-8-8L184 80c-4.4 0-8 3.6-8 8zm-48 40l0-40c0-30.9 25.1-56 56-56l144 0c30.9 0 56 25.1 56 56l0 40 28.1 0c12.7 0 24.9 5.1 33.9 14.1l51.9 51.9c9 9 14.1 21.2 14.1 33.9l0 92.1-128 0 0-32c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 32-128 0 0-32c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 32L0 320l0-92.1c0-12.7 5.1-24.9 14.1-33.9l51.9-51.9c9-9 21.2-14.1 33.9-14.1l28.1 0zM0 416l0-64 128 0c0 17.7 14.3 32 32 32s32-14.3 32-32l128 0c0 17.7 14.3 32 32 32s32-14.3 32-32l128 0 0 64c0 35.3-28.7 64-64 64L64 480c-35.3 0-64-28.7-64-64z\"],\"hands-holding-circle\":[640,512,[],\"e4fb\",\"M320 0a128 128 0 1 1 0 256A128 128 0 1 1 320 0zM40 64c22.1 0 40 17.9 40 40v40 80 40.2c0 17 6.7 33.3 18.7 45.3l51.1 51.1c8.3 8.3 21.3 9.6 31 3.1c12.9-8.6 14.7-26.9 3.7-37.8l-15.2-15.2-32-32c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0l32 32 15.2 15.2 0 0 25.3 25.3c21 21 32.8 49.5 32.8 79.2V464c0 26.5-21.5 48-48 48H173.3c-17 0-33.3-6.7-45.3-18.7L28.1 393.4C10.1 375.4 0 351 0 325.5V224 160 104C0 81.9 17.9 64 40 64zm560 0c22.1 0 40 17.9 40 40v56 64V325.5c0 25.5-10.1 49.9-28.1 67.9L512 493.3c-12 12-28.3 18.7-45.3 18.7H400c-26.5 0-48-21.5-48-48V385.1c0-29.7 11.8-58.2 32.8-79.2l25.3-25.3 0 0 15.2-15.2 32-32c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3l-32 32-15.2 15.2c-11 11-9.2 29.2 3.7 37.8c9.7 6.5 22.7 5.2 31-3.1l51.1-51.1c12-12 18.7-28.3 18.7-45.3V224 144 104c0-22.1 17.9-40 40-40z\"],bug:[512,512,[],\"f188\",\"M256 0c53 0 96 43 96 96v3.6c0 15.7-12.7 28.4-28.4 28.4H188.4c-15.7 0-28.4-12.7-28.4-28.4V96c0-53 43-96 96-96zM41.4 105.4c12.5-12.5 32.8-12.5 45.3 0l64 64c.7 .7 1.3 1.4 1.9 2.1c14.2-7.3 30.4-11.4 47.5-11.4H312c17.1 0 33.2 4.1 47.5 11.4c.6-.7 1.2-1.4 1.9-2.1l64-64c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3l-64 64c-.7 .7-1.4 1.3-2.1 1.9c6.2 12 10.1 25.3 11.1 39.5H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H416c0 24.6-5.5 47.8-15.4 68.6c2.2 1.3 4.2 2.9 6 4.8l64 64c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0l-63.1-63.1c-24.5 21.8-55.8 36.2-90.3 39.6V240c0-8.8-7.2-16-16-16s-16 7.2-16 16V479.2c-34.5-3.4-65.8-17.8-90.3-39.6L86.6 502.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l64-64c1.9-1.9 3.9-3.4 6-4.8C101.5 367.8 96 344.6 96 320H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H96.3c1.1-14.1 5-27.5 11.1-39.5c-.7-.6-1.4-1.2-2.1-1.9l-64-64c-12.5-12.5-12.5-32.8 0-45.3z\"],\"credit-card\":[576,512,[128179,62083,\"credit-card-alt\"],\"f09d\",\"M64 32C28.7 32 0 60.7 0 96v32H576V96c0-35.3-28.7-64-64-64H64zM576 224H0V416c0 35.3 28.7 64 64 64H512c35.3 0 64-28.7 64-64V224zM112 352h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zm112 16c0-8.8 7.2-16 16-16H368c8.8 0 16 7.2 16 16s-7.2 16-16 16H240c-8.8 0-16-7.2-16-16z\"],car:[512,512,[128664,\"automobile\"],\"f1b9\",\"M135.2 117.4L109.1 192H402.9l-26.1-74.6C372.3 104.6 360.2 96 346.6 96H165.4c-13.6 0-25.7 8.6-30.2 21.4zM39.6 196.8L74.8 96.3C88.3 57.8 124.6 32 165.4 32H346.6c40.8 0 77.1 25.8 90.6 64.3l35.2 100.5c23.2 9.6 39.6 32.5 39.6 59.2V400v48c0 17.7-14.3 32-32 32H448c-17.7 0-32-14.3-32-32V400H96v48c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32V400 256c0-26.7 16.4-49.6 39.6-59.2zM128 288a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm288 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"hand-holding-hand\":[576,512,[],\"e4f7\",\"M7.8 207.7c-13.1-17.8-9.3-42.8 8.5-55.9L142.9 58.5C166.2 41.3 194.5 32 223.5 32H384 544c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H507.2l-44.9 36c-22.7 18.2-50.9 28-80 28H304 288 224c-17.7 0-32-14.3-32-32s14.3-32 32-32h64 16c8.8 0 16-7.2 16-16s-7.2-16-16-16H183.4L63.7 216.2c-17.8 13.1-42.8 9.3-55.9-8.5zM382.4 160l0 0 .9 0c-.3 0-.6 0-.9 0zM568.2 304.3c13.1 17.8 9.3 42.8-8.5 55.9L433.1 453.5c-23.4 17.2-51.6 26.5-80.7 26.5H192 32c-17.7 0-32-14.3-32-32V384c0-17.7 14.3-32 32-32H68.8l44.9-36c22.7-18.2 50.9-28 80-28H272h16 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H288 272c-8.8 0-16 7.2-16 16s7.2 16 16 16H392.6l119.7-88.2c17.8-13.1 42.8-9.3 55.9 8.5zM193.6 352l0 0-.9 0c.3 0 .6 0 .9 0z\"],\"book-open-reader\":[512,512,[\"book-reader\"],\"f5da\",\"M160 96a96 96 0 1 1 192 0A96 96 0 1 1 160 96zm80 152V512l-48.4-24.2c-20.9-10.4-43.5-17-66.8-19.3l-96-9.6C12.5 457.2 0 443.5 0 427V224c0-17.7 14.3-32 32-32H62.3c63.6 0 125.6 19.6 177.7 56zm32 264V248c52.1-36.4 114.1-56 177.7-56H480c17.7 0 32 14.3 32 32V427c0 16.4-12.5 30.2-28.8 31.8l-96 9.6c-23.2 2.3-45.9 8.9-66.8 19.3L272 512z\"],\"mountain-sun\":[640,512,[],\"e52f\",\"M560 160A80 80 0 1 0 560 0a80 80 0 1 0 0 160zM55.9 512H381.1h75H578.9c33.8 0 61.1-27.4 61.1-61.1c0-11.2-3.1-22.2-8.9-31.8l-132-216.3C495 196.1 487.8 192 480 192s-15 4.1-19.1 10.7l-48.2 79L286.8 81c-6.6-10.6-18.3-17-30.8-17s-24.1 6.4-30.8 17L8.6 426.4C3 435.3 0 445.6 0 456.1C0 487 25 512 55.9 512z\"],\"arrows-left-right-to-line\":[640,512,[],\"e4ba\",\"M32 64c17.7 0 32 14.3 32 32l0 320c0 17.7-14.3 32-32 32s-32-14.3-32-32V96C0 78.3 14.3 64 32 64zm214.6 73.4c12.5 12.5 12.5 32.8 0 45.3L205.3 224l229.5 0-41.4-41.4c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0l96 96c12.5 12.5 12.5 32.8 0 45.3l-96 96c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L434.7 288l-229.5 0 41.4 41.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0l-96-96c-12.5-12.5-12.5-32.8 0-45.3l96-96c12.5-12.5 32.8-12.5 45.3 0zM640 96V416c0 17.7-14.3 32-32 32s-32-14.3-32-32V96c0-17.7 14.3-32 32-32s32 14.3 32 32z\"],\"dice-d20\":[512,512,[],\"f6cf\",\"M48.7 125.8l53.2 31.9c7.8 4.7 17.8 2 22.2-5.9L201.6 12.1c3-5.4-.9-12.1-7.1-12.1c-1.6 0-3.2 .5-4.6 1.4L47.9 98.8c-9.6 6.6-9.2 20.9 .8 26.9zM16 171.7V295.3c0 8 10.4 11 14.7 4.4l60-92c5-7.6 2.6-17.8-5.2-22.5L40.2 158C29.6 151.6 16 159.3 16 171.7zM310.4 12.1l77.6 139.6c4.4 7.9 14.5 10.6 22.2 5.9l53.2-31.9c10-6 10.4-20.3 .8-26.9L322.1 1.4c-1.4-.9-3-1.4-4.6-1.4c-6.2 0-10.1 6.7-7.1 12.1zM496 171.7c0-12.4-13.6-20.1-24.2-13.7l-45.3 27.2c-7.8 4.7-10.1 14.9-5.2 22.5l60 92c4.3 6.7 14.7 3.6 14.7-4.4V171.7zm-49.3 246L286.1 436.6c-8.1 .9-14.1 7.8-14.1 15.9v52.8c0 3.7 3 6.8 6.8 6.8c.8 0 1.6-.1 2.4-.4l172.7-64c6.1-2.2 10.1-8 10.1-14.5c0-9.3-8.1-16.5-17.3-15.4zM233.2 512c3.7 0 6.8-3 6.8-6.8V452.6c0-8.1-6.1-14.9-14.1-15.9l-160.6-19c-9.2-1.1-17.3 6.1-17.3 15.4c0 6.5 4 12.3 10.1 14.5l172.7 64c.8 .3 1.6 .4 2.4 .4zM41.7 382.9l170.9 20.2c7.8 .9 13.4-7.5 9.5-14.3l-85.7-150c-5.9-10.4-20.7-10.8-27.3-.8L30.2 358.2c-6.5 9.9-.3 23.3 11.5 24.7zm439.6-24.8L402.9 238.1c-6.5-10-21.4-9.6-27.3 .8L290.2 388.5c-3.9 6.8 1.6 15.2 9.5 14.3l170.1-20c11.8-1.4 18-14.7 11.5-24.6zm-216.9 11l78.4-137.2c6.1-10.7-1.6-23.9-13.9-23.9H183.1c-12.3 0-20 13.3-13.9 23.9l78.4 137.2c3.7 6.4 13 6.4 16.7 0zM174.4 176H337.6c12.2 0 19.9-13.1 14-23.8l-80-144c-2.8-5.1-8.2-8.2-14-8.2h-3.2c-5.8 0-11.2 3.2-14 8.2l-80 144c-5.9 10.7 1.8 23.8 14 23.8z\"],\"truck-droplet\":[640,512,[],\"e58c\",\"M0 48C0 21.5 21.5 0 48 0H368c26.5 0 48 21.5 48 48V96h50.7c17 0 33.3 6.7 45.3 18.7L589.3 192c12 12 18.7 28.3 18.7 45.3V256v32 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H576c0 53-43 96-96 96s-96-43-96-96H256c0 53-43 96-96 96s-96-43-96-96H48c-26.5 0-48-21.5-48-48V48zM416 256H544V237.3L466.7 160H416v96zM160 464a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm368-48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM208 272c39.8 0 72-29.6 72-66c0-27-39.4-82.9-59.9-110.3c-6.1-8.2-18.1-8.2-24.2 0C175.4 123 136 179 136 206c0 36.5 32.2 66 72 66z\"],\"file-circle-xmark\":[576,512,[],\"e5a1\",\"M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v38.6C310.1 219.5 256 287.4 256 368c0 59.1 29.1 111.3 73.7 143.3c-3.2 .5-6.4 .7-9.7 .7H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zm48 96a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm59.3 107.3c6.2-6.2 6.2-16.4 0-22.6s-16.4-6.2-22.6 0L432 345.4l-36.7-36.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6L409.4 368l-36.7 36.7c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0L432 390.6l36.7 36.7c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6L454.6 368l36.7-36.7z\"],\"temperature-arrow-up\":[576,512,[\"temperature-up\"],\"e040\",\"M128 112c0-26.5 21.5-48 48-48s48 21.5 48 48V276.5c0 17.3 7.1 31.9 15.3 42.5C249.8 332.6 256 349.5 256 368c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-18.5 6.2-35.4 16.7-48.9c8.2-10.6 15.3-25.2 15.3-42.5V112zM176 0C114.1 0 64 50.1 64 112V276.4c0 .1-.1 .3-.2 .6c-.2 .6-.8 1.6-1.7 2.8C43.2 304.2 32 334.8 32 368c0 79.5 64.5 144 144 144s144-64.5 144-144c0-33.2-11.2-63.8-30.1-88.1c-.9-1.2-1.5-2.2-1.7-2.8c-.1-.3-.2-.5-.2-.6V112C288 50.1 237.9 0 176 0zm0 416c26.5 0 48-21.5 48-48c0-20.9-13.4-38.7-32-45.3V112c0-8.8-7.2-16-16-16s-16 7.2-16 16V322.7c-18.6 6.6-32 24.4-32 45.3c0 26.5 21.5 48 48 48zM480 160h32c12.9 0 24.6-7.8 29.6-19.8s2.2-25.7-6.9-34.9l-64-64c-12.5-12.5-32.8-12.5-45.3 0l-64 64c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8h32V448c0 17.7 14.3 32 32 32s32-14.3 32-32V160z\"],medal:[512,512,[127941],\"f5a2\",\"M4.1 38.2C1.4 34.2 0 29.4 0 24.6C0 11 11 0 24.6 0H133.9c11.2 0 21.7 5.9 27.4 15.5l68.5 114.1c-48.2 6.1-91.3 28.6-123.4 61.9L4.1 38.2zm503.7 0L405.6 191.5c-32.1-33.3-75.2-55.8-123.4-61.9L350.7 15.5C356.5 5.9 366.9 0 378.1 0H487.4C501 0 512 11 512 24.6c0 4.8-1.4 9.6-4.1 13.6zM80 336a176 176 0 1 1 352 0A176 176 0 1 1 80 336zm184.4-94.9c-3.4-7-13.3-7-16.8 0l-22.4 45.4c-1.4 2.8-4 4.7-7 5.1L168 298.9c-7.7 1.1-10.7 10.5-5.2 16l36.3 35.4c2.2 2.2 3.2 5.2 2.7 8.3l-8.6 49.9c-1.3 7.6 6.7 13.5 13.6 9.9l44.8-23.6c2.7-1.4 6-1.4 8.7 0l44.8 23.6c6.9 3.6 14.9-2.2 13.6-9.9l-8.6-49.9c-.5-3 .5-6.1 2.7-8.3l36.3-35.4c5.6-5.4 2.5-14.8-5.2-16l-50.1-7.3c-3-.4-5.7-2.4-7-5.1l-22.4-45.4z\"],bed:[640,512,[128716],\"f236\",\"M32 32c17.7 0 32 14.3 32 32V320H288V160c0-17.7 14.3-32 32-32H544c53 0 96 43 96 96V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V416H352 320 64v32c0 17.7-14.3 32-32 32s-32-14.3-32-32V64C0 46.3 14.3 32 32 32zm144 96a80 80 0 1 1 0 160 80 80 0 1 1 0-160z\"],\"square-h\":[448,512,[\"h-square\"],\"f0fd\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM336 152V256 360c0 13.3-10.7 24-24 24s-24-10.7-24-24V280H160l0 80c0 13.3-10.7 24-24 24s-24-10.7-24-24l0-208c0-13.3 10.7-24 24-24s24 10.7 24 24v80H288V152c0-13.3 10.7-24 24-24s24 10.7 24 24z\"],podcast:[448,512,[],\"f2ce\",\"M319.4 372c48.5-31.3 80.6-85.9 80.6-148c0-97.2-78.8-176-176-176S48 126.8 48 224c0 62.1 32.1 116.6 80.6 148c1.2 17.3 4 38 7.2 57.1l.2 1C56 395.8 0 316.5 0 224C0 100.3 100.3 0 224 0S448 100.3 448 224c0 92.5-56 171.9-136 206.1l.2-1.1c3.1-19.2 6-39.8 7.2-57zm-2.3-38.1c-1.6-5.7-3.9-11.1-7-16.2c-5.8-9.7-13.5-17-21.9-22.4c19.5-17.6 31.8-43 31.8-71.3c0-53-43-96-96-96s-96 43-96 96c0 28.3 12.3 53.8 31.8 71.3c-8.4 5.4-16.1 12.7-21.9 22.4c-3.1 5.1-5.4 10.5-7 16.2C99.8 307.5 80 268 80 224c0-79.5 64.5-144 144-144s144 64.5 144 144c0 44-19.8 83.5-50.9 109.9zM224 312c32.9 0 64 8.6 64 43.8c0 33-12.9 104.1-20.6 132.9c-5.1 19-24.5 23.4-43.4 23.4s-38.2-4.4-43.4-23.4c-7.8-28.5-20.6-99.7-20.6-132.8c0-35.1 31.1-43.8 64-43.8zm0-144a56 56 0 1 1 0 112 56 56 0 1 1 0-112z\"],\"temperature-full\":[320,512,[\"temperature-4\",\"thermometer-4\",\"thermometer-full\"],\"f2c7\",\"M160 64c-26.5 0-48 21.5-48 48V276.5c0 17.3-7.1 31.9-15.3 42.5C86.2 332.6 80 349.5 80 368c0 44.2 35.8 80 80 80s80-35.8 80-80c0-18.5-6.2-35.4-16.7-48.9c-8.2-10.6-15.3-25.2-15.3-42.5V112c0-26.5-21.5-48-48-48zM48 112C48 50.2 98.1 0 160 0s112 50.1 112 112V276.5c0 .1 .1 .3 .2 .6c.2 .6 .8 1.6 1.7 2.8c18.9 24.4 30.1 55 30.1 88.1c0 79.5-64.5 144-144 144S16 447.5 16 368c0-33.2 11.2-63.8 30.1-88.1c.9-1.2 1.5-2.2 1.7-2.8c.1-.3 .2-.5 .2-.6V112zM208 368c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-20.9 13.4-38.7 32-45.3V112c0-8.8 7.2-16 16-16s16 7.2 16 16V322.7c18.6 6.6 32 24.4 32 45.3z\"],bell:[448,512,[128276,61602],\"f0f3\",\"M224 0c-17.7 0-32 14.3-32 32V51.2C119 66 64 130.6 64 208v18.8c0 47-17.3 92.4-48.5 127.6l-7.4 8.3c-8.4 9.4-10.4 22.9-5.3 34.4S19.4 416 32 416H416c12.6 0 24-7.4 29.2-18.9s3.1-25-5.3-34.4l-7.4-8.3C401.3 319.2 384 273.9 384 226.8V208c0-77.4-55-142-128-156.8V32c0-17.7-14.3-32-32-32zm45.3 493.3c12-12 18.7-28.3 18.7-45.3H224 160c0 17 6.7 33.3 18.7 45.3s28.3 18.7 45.3 18.7s33.3-6.7 45.3-18.7z\"],superscript:[512,512,[],\"f12b\",\"M480 32c0-11.1-5.7-21.4-15.2-27.2s-21.2-6.4-31.1-1.4l-32 16c-15.8 7.9-22.2 27.1-14.3 42.9C393 73.5 404.3 80 416 80v80c-17.7 0-32 14.3-32 32s14.3 32 32 32h32 32c17.7 0 32-14.3 32-32s-14.3-32-32-32V32zM32 64C14.3 64 0 78.3 0 96s14.3 32 32 32H47.3l89.6 128L47.3 384H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64c10.4 0 20.2-5.1 26.2-13.6L176 311.8l85.8 122.6c6 8.6 15.8 13.6 26.2 13.6h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H304.7L215.1 256l89.6-128H320c17.7 0 32-14.3 32-32s-14.3-32-32-32H288c-10.4 0-20.2 5.1-26.2 13.6L176 200.2 90.2 77.6C84.2 69.1 74.4 64 64 64H32z\"],\"plug-circle-xmark\":[576,512,[],\"e560\",\"M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8c12.3-2.5 24.1-6.4 35.1-11.5c-2.1-10.8-3.1-21.9-3.1-33.3c0-80.3 53.8-148 127.3-169.2c.5-2.2 .7-4.5 .7-6.8c0-17.7-14.3-32-32-32H32zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm59.3-180.7L454.6 368l36.7 36.7c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0L432 390.6l-36.7 36.7c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6L409.4 368l-36.7-36.7c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L432 345.4l36.7-36.7c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z\"],\"star-of-life\":[512,512,[],\"f621\",\"M208 32c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V172.9l122-70.4c15.3-8.8 34.9-3.6 43.7 11.7l16 27.7c8.8 15.3 3.6 34.9-11.7 43.7L352 256l122 70.4c15.3 8.8 20.5 28.4 11.7 43.7l-16 27.7c-8.8 15.3-28.4 20.6-43.7 11.7L304 339.1V480c0 17.7-14.3 32-32 32H240c-17.7 0-32-14.3-32-32V339.1L86 409.6c-15.3 8.8-34.9 3.6-43.7-11.7l-16-27.7c-8.8-15.3-3.6-34.9 11.7-43.7L160 256 38 185.6c-15.3-8.8-20.5-28.4-11.7-43.7l16-27.7C51.1 98.8 70.7 93.6 86 102.4l122 70.4V32z\"],\"phone-slash\":[640,512,[],\"f3dd\",\"M601.2 5.1c10.4-8.2 25.5-6.3 33.7 4.1s6.3 25.5-4.1 33.7l-592 464c-10.4 8.2-25.5 6.3-33.7-4.1s-6.3-25.5 4.1-33.7l155.6-122C101.8 270 64 171.4 64 64c0-18 12.1-33.8 29.5-38.6l88-24c19.4-5.3 39.7 4.6 47.4 23.2l40 96c6.8 16.3 2.1 35.2-11.6 46.3L208 207.3c12 25.5 27.2 49.2 45 70.7L601.2 5.1zM234.3 415.6l91.3-72c13.7 9.1 28 17.3 43 24.3L409 318.7c11.2-13.7 30-18.4 46.3-11.6l96 40c18.6 7.7 28.5 28 23.2 47.4l-24 88C545.8 499.9 530 512 512 512c-104.9 0-201.3-36-277.7-96.4z\"],\"paint-roller\":[512,512,[],\"f5aa\",\"M0 64C0 28.7 28.7 0 64 0H352c35.3 0 64 28.7 64 64v64c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM160 352c0-17.7 14.3-32 32-32V304c0-44.2 35.8-80 80-80H416c17.7 0 32-14.3 32-32V160 69.5c37.3 13.2 64 48.7 64 90.5v32c0 53-43 96-96 96H272c-8.8 0-16 7.2-16 16v16c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32V352z\"],\"handshake-angle\":[640,512,[\"hands-helping\"],\"f4c4\",\"M544 248v3.3l69.7-69.7c21.9-21.9 21.9-57.3 0-79.2L535.6 24.4c-21.9-21.9-57.3-21.9-79.2 0L416.3 64.5c-2.7-.3-5.5-.5-8.3-.5H296c-37.1 0-67.6 28-71.6 64H224V248c0 22.1 17.9 40 40 40s40-17.9 40-40V176c0 0 0-.1 0-.1V160l16 0 136 0c0 0 0 0 .1 0H464c44.2 0 80 35.8 80 80v8zM336 192v56c0 39.8-32.2 72-72 72s-72-32.2-72-72V129.4c-35.9 6.2-65.8 32.3-76 68.2L99.5 255.2 26.3 328.4c-21.9 21.9-21.9 57.3 0 79.2l78.1 78.1c21.9 21.9 57.3 21.9 79.2 0l37.7-37.7c.9 0 1.8 .1 2.7 .1H384c26.5 0 48-21.5 48-48c0-5.6-1-11-2.7-16H432c26.5 0 48-21.5 48-48c0-12.8-5-24.4-13.2-33c25.7-5 45.1-27.6 45.2-54.8v-.4c-.1-30.8-25.1-55.8-56-55.8c0 0 0 0 0 0l-120 0z\"],\"location-dot\":[384,512,[\"map-marker-alt\"],\"f3c5\",\"M215.7 499.2C267 435 384 279.4 384 192C384 86 298 0 192 0S0 86 0 192c0 87.4 117 243 168.3 307.2c12.3 15.3 35.1 15.3 47.4 0zM192 128a64 64 0 1 1 0 128 64 64 0 1 1 0-128z\"],file:[384,512,[128196,128459,61462],\"f15b\",\"M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128z\"],\"greater-than\":[384,512,[62769],\"3e\",\"M3.4 81.7c-7.9 15.8-1.5 35 14.3 42.9L280.5 256 17.7 387.4C1.9 395.3-4.5 414.5 3.4 430.3s27.1 22.2 42.9 14.3l320-160c10.8-5.4 17.7-16.5 17.7-28.6s-6.8-23.2-17.7-28.6l-320-160c-15.8-7.9-35-1.5-42.9 14.3z\"],\"person-swimming\":[576,512,[127946,\"swimmer\"],\"f5c4\",\"M309.5 178.4L447.9 297.1c-1.6 .9-3.2 2-4.8 3c-18 12.4-40.1 20.3-59.2 20.3c-19.6 0-40.8-7.7-59.2-20.3c-22.1-15.5-51.6-15.5-73.7 0c-17.1 11.8-38 20.3-59.2 20.3c-10.1 0-21.1-2.2-31.9-6.2C163.1 193.2 262.2 96 384 96h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384c-26.9 0-52.3 6.6-74.5 18.4zM160 160A64 64 0 1 1 32 160a64 64 0 1 1 128 0zM306.5 325.9C329 341.4 356.5 352 384 352c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 405.7 417 416 384 416c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 341.2 165.1 352 192 352c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z\"],\"arrow-down\":[384,512,[8595],\"f063\",\"M169.4 470.6c12.5 12.5 32.8 12.5 45.3 0l160-160c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L224 370.8 224 64c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 306.7L54.6 265.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l160 160z\"],droplet:[384,512,[128167,\"tint\"],\"f043\",\"M192 512C86 512 0 426 0 320C0 228.8 130.2 57.7 166.6 11.7C172.6 4.2 181.5 0 191.1 0h1.8c9.6 0 18.5 4.2 24.5 11.7C253.8 57.7 384 228.8 384 320c0 106-86 192-192 192zM96 336c0-8.8-7.2-16-16-16s-16 7.2-16 16c0 61.9 50.1 112 112 112c8.8 0 16-7.2 16-16s-7.2-16-16-16c-44.2 0-80-35.8-80-80z\"],eraser:[576,512,[],\"f12d\",\"M290.7 57.4L57.4 290.7c-25 25-25 65.5 0 90.5l80 80c12 12 28.3 18.7 45.3 18.7H288h9.4H512c17.7 0 32-14.3 32-32s-14.3-32-32-32H387.9L518.6 285.3c25-25 25-65.5 0-90.5L381.3 57.4c-25-25-65.5-25-90.5 0zM297.4 416H288l-105.4 0-80-80L227.3 211.3 364.7 348.7 297.4 416z\"],\"earth-americas\":[512,512,[127758,\"earth\",\"earth-america\",\"globe-americas\"],\"f57d\",\"M57.7 193l9.4 16.4c8.3 14.5 21.9 25.2 38 29.8L163 255.7c17.2 4.9 29 20.6 29 38.5v39.9c0 11 6.2 21 16 25.9s16 14.9 16 25.9v39c0 15.6 14.9 26.9 29.9 22.6c16.1-4.6 28.6-17.5 32.7-33.8l2.8-11.2c4.2-16.9 15.2-31.4 30.3-40l8.1-4.6c15-8.5 24.2-24.5 24.2-41.7v-8.3c0-12.7-5.1-24.9-14.1-33.9l-3.9-3.9c-9-9-21.2-14.1-33.9-14.1H257c-11.1 0-22.1-2.9-31.8-8.4l-34.5-19.7c-4.3-2.5-7.6-6.5-9.2-11.2c-3.2-9.6 1.1-20 10.2-24.5l5.9-3c6.6-3.3 14.3-3.9 21.3-1.5l23.2 7.7c8.2 2.7 17.2-.4 21.9-7.5c4.7-7 4.2-16.3-1.2-22.8l-13.6-16.3c-10-12-9.9-29.5 .3-41.3l15.7-18.3c8.8-10.3 10.2-25 3.5-36.7l-2.4-4.2c-3.5-.2-6.9-.3-10.4-.3C163.1 48 84.4 108.9 57.7 193zM464 256c0-36.8-9.6-71.4-26.4-101.5L412 164.8c-15.7 6.3-23.8 23.8-18.5 39.8l16.9 50.7c3.5 10.4 12 18.3 22.6 20.9l29.1 7.3c1.2-9 1.8-18.2 1.8-27.5zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256z\"],\"person-burst\":[640,512,[],\"e53b\",\"M480 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-8 384V352h16V480c0 17.7 14.3 32 32 32s32-14.3 32-32V256.9l28.6 47.5c9.1 15.1 28.8 20 43.9 10.9s20-28.8 10.9-43.9l-58.3-97c-17.4-28.9-48.6-46.6-82.3-46.6H465.1c-33.7 0-64.9 17.7-82.3 46.6l-58.3 97c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9L408 256.9V480c0 17.7 14.3 32 32 32s32-14.3 32-32zM190.9 18.1C188.4 12 182.6 8 176 8s-12.4 4-14.9 10.1l-29.4 74L55.6 68.9c-6.3-1.9-13.1 .2-17.2 5.3s-4.6 12.2-1.4 17.9l39.5 69.1L10.9 206.4c-5.4 3.7-8 10.3-6.5 16.7s6.7 11.2 13.1 12.2l78.7 12.2L90.6 327c-.5 6.5 3.1 12.7 9 15.5s12.9 1.8 17.8-2.6L176 286.1l58.6 53.9c4.8 4.4 11.9 5.5 17.8 2.6s9.5-9 9-15.5l-5.6-79.4 50.5-7.8 24.3-40.5-55.2-38L315 92.2c3.3-5.7 2.7-12.8-1.4-17.9s-10.9-7.2-17.2-5.3L220.3 92.1l-29.4-74z\"],dove:[512,512,[128330],\"f4ba\",\"M160.8 96.5c14 17 31 30.9 49.5 42.2c25.9 15.8 53.7 25.9 77.7 31.6V138.8C265.8 108.5 250 71.5 248.6 28c-.4-11.3-7.5-21.5-18.4-24.4c-7.6-2-15.8-.2-21 5.8c-13.3 15.4-32.7 44.6-48.4 87.2zM320 144v30.6l0 0v1.3l0 0 0 32.1c-60.8-5.1-185-43.8-219.3-157.2C97.4 40 87.9 32 76.6 32c-7.9 0-15.3 3.9-18.8 11C46.8 65.9 32 112.1 32 176c0 116.9 80.1 180.5 118.4 202.8L11.8 416.6C6.7 418 2.6 421.8 .9 426.8s-.8 10.6 2.3 14.8C21.7 466.2 77.3 512 160 512c3.6 0 7.2-1.2 10-3.5L245.6 448H320c88.4 0 160-71.6 160-160V128l29.9-44.9c1.3-2 2.1-4.4 2.1-6.8c0-6.8-5.5-12.3-12.3-12.3H400c-44.2 0-80 35.8-80 80zm80-16a16 16 0 1 1 0 32 16 16 0 1 1 0-32z\"],\"battery-empty\":[576,512,[\"battery-0\"],\"f244\",\"M80 160c-8.8 0-16 7.2-16 16V336c0 8.8 7.2 16 16 16H464c8.8 0 16-7.2 16-16V176c0-8.8-7.2-16-16-16H80zM0 176c0-44.2 35.8-80 80-80H464c44.2 0 80 35.8 80 80v16c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32v16c0 44.2-35.8 80-80 80H80c-44.2 0-80-35.8-80-80V176z\"],socks:[512,512,[129510],\"f696\",\"M175.2 476.6c-9.7-18-15.2-38.7-15.2-60.6c0-40.3 19-78.2 51.2-102.4l64-48c8.1-6 12.8-15.5 12.8-25.6V96H128V240c0 20.1-9.5 39.1-25.6 51.2l-64 48C14.2 357.3 0 385.8 0 416c0 53 43 96 96 96c20.8 0 41-6.7 57.6-19.2l21.6-16.2zM128 64H288V48c0-14.5 3.9-28.2 10.7-39.9C291 3 281.9 0 272 0H176c-26.5 0-48 21.5-48 48V64zM320 96V240c0 20.1-9.5 39.1-25.6 51.2l-64 48C206.2 357.3 192 385.8 192 416c0 53 43 96 96 96c20.8 0 41-6.7 57.6-19.2l115.2-86.4C493 382.2 512 344.3 512 304V96H320zM512 64V48c0-26.5-21.5-48-48-48H368c-26.5 0-48 21.5-48 48V64H512z\"],inbox:[512,512,[],\"f01c\",\"M121 32C91.6 32 66 52 58.9 80.5L1.9 308.4C.6 313.5 0 318.7 0 323.9V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V323.9c0-5.2-.6-10.4-1.9-15.5l-57-227.9C446 52 420.4 32 391 32H121zm0 64H391l48 192H387.8c-12.1 0-23.2 6.8-28.6 17.7l-14.3 28.6c-5.4 10.8-16.5 17.7-28.6 17.7H195.8c-12.1 0-23.2-6.8-28.6-17.7l-14.3-28.6c-5.4-10.8-16.5-17.7-28.6-17.7H73L121 96z\"],section:[256,512,[],\"e447\",\"M64.9 96C67.1 84.4 73.7 76.2 86 70.6c13.8-6.2 34.8-8.9 61.2-4.5c8.8 1.4 36.1 7.1 44.1 9.3c17 4.8 34.7-5.1 39.5-22.2s-5.1-34.7-22.2-39.5c-11.1-3.1-41-9.2-50.9-10.8C123-2.7 88.3-.6 59.7 12.3C29.9 25.8 7.5 50.9 1.6 86.5c-.1 .5-.2 1.1-.2 1.6c-2.2 19.7 .3 37.9 8.1 54.1c7.7 16.1 19.4 28 32 36.9c.6 .5 1.3 .9 2 1.4C22.3 194.2 6.5 215.1 1.7 243c-.1 .6-.2 1.1-.2 1.7c-2.3 19.3 .4 37.1 8.4 53c7.9 15.6 19.8 27 32.3 35.5c22.4 15.2 51.9 24 75.4 31l0 0 3.7 1.1c27.2 8.2 46.9 14.6 59.4 23.8c5.5 4 8.2 7.6 9.5 10.9c1.3 3.2 2.6 8.6 .9 18.1c-1.7 10.1-7.7 18-20.7 23.5c-14 6-35.4 8.5-62 4.4c-12.8-2.1-35.1-9.7-54.1-16.2l0 0c-4.3-1.5-8.5-2.9-12.3-4.2C25.3 420 7.2 429.1 1.6 445.8s3.5 34.9 20.3 40.5c2.6 .8 5.7 1.9 9.2 3.1c18.6 6.3 48.5 16.6 67.3 19.6l0 0 .2 0c34.5 5.4 68.8 3.4 97.2-8.7c29.4-12.6 52.5-36.5 58.5-71.5c3.3-19.3 1.9-37.4-5-53.9c-6.3-15-16.4-26.4-27.6-35.2c16.5-13.9 28.5-33.2 32.6-58.2c3.2-19.8 1.9-38.3-4.8-55.1c-6.7-16.8-17.8-29.4-30.2-39c-22.8-17.6-53.6-27.4-77.7-35l-1.4-.5c-27.4-8.7-47.8-15.3-61.5-25c-6.1-4.4-9.5-8.5-11.4-12.4c-1.8-3.7-3.2-9.3-2.3-18.5zm76.7 208.5c-.2-.1-.4-.1-.6-.2l-1.4-.4c-27.4-8.2-47.9-14.5-61.7-23.8c-6.2-4.2-9.3-7.9-11-11.3c-1.5-3-2.9-7.7-2.1-15.7c1.9-9.7 7.9-17.3 20.5-22.7c14-6 35.4-8.5 62.1-4.3l16.4 2.6c6.3 2.9 11.7 6 16.2 9.5c5.5 4.2 8.4 8.2 10 12.2c1.6 4 2.8 10.4 1.1 20.9c-2.4 14.7-12.8 26.4-37.1 31l-12.4 2.3z\"],\"gauge-high\":[512,512,[62461,\"tachometer-alt\",\"tachometer-alt-fast\"],\"f625\",\"M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM288 96a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM256 416c35.3 0 64-28.7 64-64c0-17.4-6.9-33.1-18.1-44.6L366 161.7c5.3-12.1-.2-26.3-12.3-31.6s-26.3 .2-31.6 12.3L257.9 288c-.6 0-1.3 0-1.9 0c-35.3 0-64 28.7-64 64s28.7 64 64 64zM176 144a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM96 288a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm352-32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z\"],\"envelope-open-text\":[512,512,[],\"f658\",\"M215.4 96H144 107.8 96v8.8V144v40.4 89L.2 202.5c1.6-18.1 10.9-34.9 25.7-45.8L48 140.3V96c0-26.5 21.5-48 48-48h76.6l49.9-36.9C232.2 3.9 243.9 0 256 0s23.8 3.9 33.5 11L339.4 48H416c26.5 0 48 21.5 48 48v44.3l22.1 16.4c14.8 10.9 24.1 27.7 25.7 45.8L416 273.4v-89V144 104.8 96H404.2 368 296.6 215.4zM0 448V242.1L217.6 403.3c11.1 8.2 24.6 12.7 38.4 12.7s27.3-4.4 38.4-12.7L512 242.1V448v0c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64v0zM176 160H336c8.8 0 16 7.2 16 16s-7.2 16-16 16H176c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H336c8.8 0 16 7.2 16 16s-7.2 16-16 16H176c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],hospital:[640,512,[127973,62589,\"hospital-alt\",\"hospital-wide\"],\"f0f8\",\"M192 48c0-26.5 21.5-48 48-48H400c26.5 0 48 21.5 48 48V512H368V432c0-26.5-21.5-48-48-48s-48 21.5-48 48v80H192V48zM48 96H160V512H48c-26.5 0-48-21.5-48-48V320H80c8.8 0 16-7.2 16-16s-7.2-16-16-16H0V224H80c8.8 0 16-7.2 16-16s-7.2-16-16-16H0V144c0-26.5 21.5-48 48-48zm544 0c26.5 0 48 21.5 48 48v48H560c-8.8 0-16 7.2-16 16s7.2 16 16 16h80v64H560c-8.8 0-16 7.2-16 16s7.2 16 16 16h80V464c0 26.5-21.5 48-48 48H480V96H592zM312 64c-8.8 0-16 7.2-16 16v24H272c-8.8 0-16 7.2-16 16v16c0 8.8 7.2 16 16 16h24v24c0 8.8 7.2 16 16 16h16c8.8 0 16-7.2 16-16V152h24c8.8 0 16-7.2 16-16V120c0-8.8-7.2-16-16-16H344V80c0-8.8-7.2-16-16-16H312z\"],\"wine-bottle\":[512,512,[],\"f72f\",\"M393.4 9.4c12.5-12.5 32.8-12.5 45.3 0l64 64c12.5 12.5 12.5 32.8 0 45.3c-11.8 11.8-30.7 12.5-43.2 1.9l-9.5 9.5-48.8 48.8c-9.2 9.2-11.5 22.9-8.6 35.6c9.4 40.9-1.9 85.6-33.8 117.5L197.3 493.3c-25 25-65.5 25-90.5 0l-88-88c-25-25-25-65.5 0-90.5L180.2 153.3c31.9-31.9 76.6-43.1 117.5-33.8c12.6 2.9 26.4 .5 35.5-8.6l48.8-48.8 9.5-9.5c-10.6-12.6-10-31.4 1.9-43.2zM99.3 347.3l65.4 65.4c6.2 6.2 16.4 6.2 22.6 0l97.4-97.4c6.2-6.2 6.2-16.4 0-22.6l-65.4-65.4c-6.2-6.2-16.4-6.2-22.6 0L99.3 324.7c-6.2 6.2-6.2 16.4 0 22.6z\"],\"chess-rook\":[448,512,[9820],\"f447\",\"M32 192V48c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16V88c0 4.4 3.6 8 8 8h32c4.4 0 8-3.6 8-8V48c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16V88c0 4.4 3.6 8 8 8h32c4.4 0 8-3.6 8-8V48c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16V192c0 10.1-4.7 19.6-12.8 25.6L352 256l16 144H80L96 256 44.8 217.6C36.7 211.6 32 202.1 32 192zm176 96h32c8.8 0 16-7.2 16-16V224c0-17.7-14.3-32-32-32s-32 14.3-32 32v48c0 8.8 7.2 16 16 16zM22.6 473.4L64 432H384l41.4 41.4c4.2 4.2 6.6 10 6.6 16c0 12.5-10.1 22.6-22.6 22.6H38.6C26.1 512 16 501.9 16 489.4c0-6 2.4-11.8 6.6-16z\"],\"bars-staggered\":[512,512,[\"reorder\",\"stream\"],\"f550\",\"M0 96C0 78.3 14.3 64 32 64H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 128 0 113.7 0 96zM64 256c0-17.7 14.3-32 32-32H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H96c-17.7 0-32-14.3-32-32zM448 416c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H416c17.7 0 32 14.3 32 32z\"],dharmachakra:[512,512,[9784],\"f655\",\"M337.8 205.7l48.6-42.5c13.8 19.3 23.4 41.9 27.4 66.2l-64.4 4.3c-2.4-10.1-6.4-19.5-11.6-28zm140.1 19.5c-5.3-38.8-20.6-74.5-43.2-104.3l.8-.7C449 108.4 449.7 87.6 437 75s-33.4-12-45.2 1.5l-.7 .8c-29.8-22.6-65.5-37.9-104.3-43.2l.1-1.1c1.2-17.9-13-33-30.9-33s-32.1 15.2-30.9 33l.1 1.1c-38.8 5.3-74.5 20.6-104.3 43.2l-.7-.8C108.4 63 87.6 62.3 75 75s-12 33.4 1.5 45.2l.8 .7c-22.6 29.8-37.9 65.5-43.2 104.3l-1.1-.1c-17.9-1.2-33 13-33 30.9s15.2 32.1 33 30.9l1.1-.1c5.3 38.8 20.6 74.5 43.2 104.3l-.8 .7C63 403.6 62.3 424.4 75 437s33.4 12 45.2-1.5l.7-.8c29.8 22.6 65.5 37.9 104.3 43.2l-.1 1.1c-1.2 17.9 13 33 30.9 33s32.1-15.2 30.9-33l-.1-1.1c38.8-5.3 74.5-20.6 104.3-43.2l.7 .8c11.8 13.5 32.5 14.2 45.2 1.5s12-33.4-1.5-45.2l-.8-.7c22.6-29.8 37.9-65.5 43.2-104.3l1.1 .1c17.9 1.2 33-13 33-30.9s-15.2-32.1-33-30.9l-1.1 .1zM163.2 125.6c19.3-13.8 41.9-23.4 66.2-27.5l4.3 64.4c-10 2.4-19.5 6.4-28 11.6l-42.5-48.6zm-65 103.8c4.1-24.4 13.7-46.9 27.5-66.2l48.6 42.5c-5.3 8.5-9.2 18-11.6 28l-64.4-4.3zm27.5 119.4c-13.8-19.3-23.4-41.9-27.5-66.2l64.4-4.3c2.4 10 6.4 19.5 11.6 28l-48.6 42.5zm103.8 65c-24.4-4.1-46.9-13.7-66.2-27.4l42.5-48.6c8.5 5.3 18 9.2 28 11.6l-4.3 64.4zm119.4-27.4c-19.3 13.8-41.9 23.4-66.2 27.4l-4.3-64.4c10-2.4 19.5-6.4 28-11.6l42.5 48.6zm65-103.8c-4.1 24.4-13.7 46.9-27.4 66.2l-48.6-42.5c5.3-8.5 9.2-18 11.6-28l64.4 4.3zm-65-156.9l-42.5 48.6c-8.5-5.3-18-9.2-28-11.6l4.3-64.4c24.4 4.1 46.9 13.7 66.2 27.5zM256 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],hotdog:[512,512,[127789],\"f80f\",\"M488.6 23.4c31.2 31.2 31.2 81.9 0 113.1l-352 352c-31.2 31.2-81.9 31.2-113.1 0s-31.2-81.9 0-113.1l352-352c31.2-31.2 81.9-31.2 113.1 0zM443.3 92.7c-6.2-6.2-16.4-6.2-22.6 0c-12.5 12.5-23.8 15.1-37.5 17.6l-2.5 .4c-13.8 2.5-31.6 5.6-48 22c-16.7 16.7-20.9 36-24.1 50.9l0 0v0l-.2 1c-3.4 15.6-6 26.4-15.7 36.1s-20.5 12.3-36.1 15.7l-1 .2c-14.9 3.2-34.2 7.4-50.9 24.1s-20.9 36-24.1 50.9l-.2 1c-3.4 15.6-6 26.4-15.7 36.1c-9.2 9.2-18 10.8-32.7 13.4l0 0-.9 .2c-15.6 2.8-34.9 6.9-54.4 26.4c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0c12.5-12.5 23.8-15.1 37.5-17.6l2.5-.4c13.8-2.5 31.6-5.6 48-22c16.7-16.7 20.9-36 24.1-50.9l.2-1c3.4-15.6 6-26.4 15.7-36.1s20.5-12.3 36.1-15.7l1-.2c14.9-3.2 34.2-7.4 50.9-24.1s20.9-36 24.1-50.9l.2-1c3.4-15.6 6-26.4 15.7-36.1c9.2-9.2 18-10.8 32.7-13.4l.9-.2c15.6-2.8 34.9-6.9 54.4-26.4c6.2-6.2 6.2-16.4 0-22.6zM191.2 479.2l288-288L495 207c10.9 10.9 17 25.6 17 41s-6.1 30.1-17 41L289 495c-10.9 10.9-25.6 17-41 17s-30.1-6.1-41-17l-15.8-15.8zM17 305C6.1 294.1 0 279.4 0 264s6.1-30.1 17-41L223 17C233.9 6.1 248.6 0 264 0s30.1 6.1 41 17l15.8 15.8-288 288L17 305z\"],\"person-walking-with-cane\":[512,512,[\"blind\"],\"f29d\",\"M176 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-8.4 32c-36.4 0-69.6 20.5-85.9 53.1L35.4 273.7c-7.9 15.8-1.5 35 14.3 42.9s35 1.5 42.9-14.3L128 231.6v43.2c0 17 6.7 33.3 18.7 45.3L224 397.3V480c0 17.7 14.3 32 32 32s32-14.3 32-32V390.6c0-12.7-5.1-24.9-14.1-33.9L224 306.7V213.3l70.4 93.9c10.6 14.1 30.7 17 44.8 6.4s17-30.7 6.4-44.8L268.8 166.4C250.7 142.2 222.2 128 192 128H167.6zM128.3 346.8L97 472.2c-4.3 17.1 6.1 34.5 23.3 38.8s34.5-6.1 38.8-23.3l22-88.2-52.8-52.8zM450.8 505.1c5 7.3 15 9.1 22.3 4s9.1-15 4-22.3L358.9 316.1c-2.8 3.8-6.1 7.3-10.1 10.3c-5 3.8-10.5 6.4-16.2 7.9L450.8 505.1z\"],drum:[512,512,[129345],\"f569\",\"M501.2 76.1c11.1-7.3 14.2-22.1 6.9-33.2s-22.1-14.2-33.2-6.9L370.2 104.5C335.8 98.7 297 96 256 96C114.6 96 0 128 0 208V368c0 31.3 27.4 58.8 72 78.7V344c0-13.3 10.7-24 24-24s24 10.7 24 24V463.4c33 8.9 71.1 14.5 112 16.1V376c0-13.3 10.7-24 24-24s24 10.7 24 24V479.5c40.9-1.6 79-7.2 112-16.1V344c0-13.3 10.7-24 24-24s24 10.7 24 24V446.7c44.6-19.9 72-47.4 72-78.7V208c0-41.1-30.2-69.5-78.8-87.4l67.9-44.5zM307.4 145.6l-64.6 42.3c-11.1 7.3-14.2 22.1-6.9 33.2s22.1 14.2 33.2 6.9l111.1-72.8c14.7 3.2 27.9 7 39.4 11.5C458.4 181.8 464 197.4 464 208c0 .8-2.7 17.2-46 35.9C379.1 260.7 322 272 256 272s-123.1-11.3-162-28.1C50.7 225.2 48 208.8 48 208c0-10.6 5.6-26.2 44.4-41.3C130.6 151.9 187.8 144 256 144c18 0 35.1 .5 51.4 1.6z\"],\"ice-cream\":[448,512,[127848],\"f810\",\"M367.1 160c.6-5.3 .9-10.6 .9-16C368 64.5 303.5 0 224 0S80 64.5 80 144c0 5.4 .3 10.7 .9 16H80c-26.5 0-48 21.5-48 48s21.5 48 48 48h53.5 181H368c26.5 0 48-21.5 48-48s-21.5-48-48-48h-.9zM96 288L200.8 497.7c4.4 8.8 13.3 14.3 23.2 14.3s18.8-5.5 23.2-14.3L352 288H96z\"],\"heart-circle-bolt\":[576,512,[],\"e4fc\",\"M47.6 300.4L228.3 469.1c7.5 7 17.4 10.9 27.7 10.9s20.2-3.9 27.7-10.9l2.6-2.4C267.2 438.6 256 404.6 256 368c0-97.2 78.8-176 176-176c28.3 0 55 6.7 78.7 18.5c.9-6.5 1.3-13 1.3-19.6v-5.8c0-69.9-50.5-129.5-119.4-141C347 36.5 300.6 51.4 268 84L256 96 244 84c-32.6-32.6-79-47.5-124.6-39.9C50.5 55.6 0 115.2 0 185.1v5.8c0 41.5 17.2 81.2 47.6 109.5zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm47.9-225c4.3 3.7 5.4 9.9 2.6 14.9L452.4 356H488c5.2 0 9.8 3.3 11.4 8.2s-.1 10.3-4.2 13.4l-96 72c-4.5 3.4-10.8 3.2-15.1-.6s-5.4-9.9-2.6-14.9L411.6 380H376c-5.2 0-9.8-3.3-11.4-8.2s.1-10.3 4.2-13.4l96-72c4.5-3.4 10.8-3.2 15.1 .6z\"],fax:[512,512,[128224,128439],\"f1ac\",\"M128 64v96h64V64H386.7L416 93.3V160h64V93.3c0-17-6.7-33.3-18.7-45.3L432 18.7C420 6.7 403.7 0 386.7 0H192c-35.3 0-64 28.7-64 64zM0 160V480c0 17.7 14.3 32 32 32H64c17.7 0 32-14.3 32-32V160c0-17.7-14.3-32-32-32H32c-17.7 0-32 14.3-32 32zm480 32H128V480c0 17.7 14.3 32 32 32H480c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32zM256 256a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm96 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm32 96a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM224 416a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],paragraph:[448,512,[182],\"f1dd\",\"M192 32h64H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H384l0 352c0 17.7-14.3 32-32 32s-32-14.3-32-32l0-352H288V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H192c-88.4 0-160-71.6-160-160s71.6-160 160-160z\"],\"check-to-slot\":[576,512,[\"vote-yea\"],\"f772\",\"M96 80c0-26.5 21.5-48 48-48H432c26.5 0 48 21.5 48 48V384H96V80zm313 47c-9.4-9.4-24.6-9.4-33.9 0l-111 111-47-47c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l64 64c9.4 9.4 24.6 9.4 33.9 0L409 161c9.4-9.4 9.4-24.6 0-33.9zM0 336c0-26.5 21.5-48 48-48H64V416H512V288h16c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V336z\"],\"star-half\":[576,512,[61731],\"f089\",\"M288 0c-12.2 .1-23.3 7-28.6 18L195 150.3 51.4 171.5c-12 1.8-22 10.2-25.7 21.7s-.7 24.2 7.9 32.7L137.8 329 113.2 474.7c-2 12 3 24.2 12.9 31.3s23 8 33.8 2.3L288 439.8V0zM429.9 512c1.1 .1 2.1 .1 3.2 0h-3.2z\"],\"boxes-stacked\":[576,512,[62625,\"boxes\",\"boxes-alt\"],\"f468\",\"M248 0H208c-26.5 0-48 21.5-48 48V160c0 35.3 28.7 64 64 64H352c35.3 0 64-28.7 64-64V48c0-26.5-21.5-48-48-48H328V80c0 8.8-7.2 16-16 16H264c-8.8 0-16-7.2-16-16V0zM64 256c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64H224c35.3 0 64-28.7 64-64V320c0-35.3-28.7-64-64-64H184v80c0 8.8-7.2 16-16 16H120c-8.8 0-16-7.2-16-16V256H64zM352 512H512c35.3 0 64-28.7 64-64V320c0-35.3-28.7-64-64-64H472v80c0 8.8-7.2 16-16 16H408c-8.8 0-16-7.2-16-16V256H352c-15 0-28.8 5.1-39.7 13.8c4.9 10.4 7.7 22 7.7 34.2V464c0 12.2-2.8 23.8-7.7 34.2C323.2 506.9 337 512 352 512z\"],link:[640,512,[128279,\"chain\"],\"f0c1\",\"M579.8 267.7c56.5-56.5 56.5-148 0-204.5c-50-50-128.8-56.5-186.3-15.4l-1.6 1.1c-14.4 10.3-17.7 30.3-7.4 44.6s30.3 17.7 44.6 7.4l1.6-1.1c32.1-22.9 76-19.3 103.8 8.6c31.5 31.5 31.5 82.5 0 114L422.3 334.8c-31.5 31.5-82.5 31.5-114 0c-27.9-27.9-31.5-71.8-8.6-103.8l1.1-1.6c10.3-14.4 6.9-34.4-7.4-44.6s-34.4-6.9-44.6 7.4l-1.1 1.6C206.5 251.2 213 330 263 380c56.5 56.5 148 56.5 204.5 0L579.8 267.7zM60.2 244.3c-56.5 56.5-56.5 148 0 204.5c50 50 128.8 56.5 186.3 15.4l1.6-1.1c14.4-10.3 17.7-30.3 7.4-44.6s-30.3-17.7-44.6-7.4l-1.6 1.1c-32.1 22.9-76 19.3-103.8-8.6C74 372 74 321 105.5 289.5L217.7 177.2c31.5-31.5 82.5-31.5 114 0c27.9 27.9 31.5 71.8 8.6 103.9l-1.1 1.6c-10.3 14.4-6.9 34.4 7.4 44.6s34.4 6.9 44.6-7.4l1.1-1.6C433.5 260.8 427 182 377 132c-56.5-56.5-148-56.5-204.5 0L60.2 244.3z\"],\"ear-listen\":[512,512,[\"assistive-listening-systems\"],\"f2a2\",\"M398.3 3.4c-15.8-7.9-35-1.5-42.9 14.3c-7.9 15.8-1.5 34.9 14.2 42.9l.4 .2c.4 .2 1.1 .6 2.1 1.2c2 1.2 5 3 8.7 5.6c7.5 5.2 17.6 13.2 27.7 24.2C428.5 113.4 448 146 448 192c0 17.7 14.3 32 32 32s32-14.3 32-32c0-66-28.5-113.4-56.5-143.7C441.6 33.2 427.7 22.2 417.3 15c-5.3-3.7-9.7-6.4-13-8.3c-1.6-1-3-1.7-4-2.2c-.5-.3-.9-.5-1.2-.7l-.4-.2-.2-.1-.1 0 0 0c0 0 0 0-14.3 28.6L398.3 3.4zM128.7 227.5c6.2-56 53.7-99.5 111.3-99.5c61.9 0 112 50.1 112 112c0 29.3-11.2 55.9-29.6 75.9c-17 18.4-34.4 45.1-34.4 78V400c0 26.5-21.5 48-48 48c-17.7 0-32 14.3-32 32s14.3 32 32 32c61.9 0 112-50.1 112-112v-6.1c0-9.8 5.4-21.7 17.4-34.7C398.3 327.9 416 286 416 240c0-97.2-78.8-176-176-176C149.4 64 74.8 132.5 65.1 220.5c-1.9 17.6 10.7 33.4 28.3 35.3s33.4-10.7 35.3-28.3zM32 512a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM192 352a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM41.4 361.4c-12.5 12.5-12.5 32.8 0 45.3l64 64c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-64-64c-12.5-12.5-32.8-12.5-45.3 0zM208 240c0-17.7 14.3-32 32-32s32 14.3 32 32c0 13.3 10.7 24 24 24s24-10.7 24-24c0-44.2-35.8-80-80-80s-80 35.8-80 80c0 13.3 10.7 24 24 24s24-10.7 24-24z\"],\"tree-city\":[640,512,[],\"e587\",\"M288 48c0-26.5 21.5-48 48-48h96c26.5 0 48 21.5 48 48V192h40V120c0-13.3 10.7-24 24-24s24 10.7 24 24v72h24c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H432 336c-26.5 0-48-21.5-48-48V48zm64 32v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V80c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16zm16 80c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V176c0-8.8-7.2-16-16-16H368zM352 272v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16zm176-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H528zM512 368v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H528c-8.8 0-16 7.2-16 16zM224 160c0 6-1 11-2 16c20 14 34 38 34 64c0 45-36 80-80 80H160V480c0 18-15 32-32 32c-18 0-32-14-32-32V320H80c-45 0-80-35-80-80c0-26 13-50 33-64c-1-5-1-10-1-16c0-53 42-96 96-96c53 0 96 43 96 96z\"],play:[384,512,[9654],\"f04b\",\"M73 39c-14.8-9.1-33.4-9.4-48.5-.9S0 62.6 0 80V432c0 17.4 9.4 33.4 24.5 41.9s33.7 8.1 48.5-.9L361 297c14.3-8.7 23-24.2 23-41s-8.7-32.2-23-41L73 39z\"],font:[448,512,[],\"f031\",\"M254 52.8C249.3 40.3 237.3 32 224 32s-25.3 8.3-30 20.8L57.8 416H32c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32h-1.8l18-48H303.8l18 48H320c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H390.2L254 52.8zM279.8 304H168.2L224 155.1 279.8 304z\"],\"table-cells-row-lock\":[640,512,[],\"e67a\",\"M0 96C0 60.7 28.7 32 64 32l384 0c35.3 0 64 28.7 64 64l0 65.1c-37.8 5.4-69.4 29.6-85.2 62.9L360 224l0 64 56 0 0 8.6c-19.1 11.1-32 31.7-32 55.4l-24 0 0 64 24 0 0 64L64 480c-35.3 0-64-28.7-64-64L0 96zM64 224l0 64 88 0 0-64-88 0zm232 0l-88 0 0 64 88 0 0-64zM152 352l-88 0 0 64 88 0 0-64zm56 0l0 64 88 0 0-64-88 0zM528 240c-17.7 0-32 14.3-32 32l0 48 64 0 0-48c0-17.7-14.3-32-32-32zm-80 32c0-44.2 35.8-80 80-80s80 35.8 80 80l0 48c17.7 0 32 14.3 32 32l0 128c0 17.7-14.3 32-32 32l-160 0c-17.7 0-32-14.3-32-32l0-128c0-17.7 14.3-32 32-32l0-48z\"],\"rupiah-sign\":[512,512,[],\"e23d\",\"M0 64C0 46.3 14.3 32 32 32h80c79.5 0 144 64.5 144 144c0 58.8-35.2 109.3-85.7 131.7l51.4 128.4c6.6 16.4-1.4 35-17.8 41.6s-35-1.4-41.6-17.8L106.3 320H64V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V288 64zM64 256h48c44.2 0 80-35.8 80-80s-35.8-80-80-80H64V256zm256-96h80c61.9 0 112 50.1 112 112s-50.1 112-112 112H352v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V352 192c0-17.7 14.3-32 32-32zm80 160c26.5 0 48-21.5 48-48s-21.5-48-48-48H352v96h48z\"],\"magnifying-glass\":[512,512,[128269,\"search\"],\"f002\",\"M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zM208 352a144 144 0 1 0 0-288 144 144 0 1 0 0 288z\"],\"table-tennis-paddle-ball\":[640,512,[127955,\"ping-pong-paddle-ball\",\"table-tennis\"],\"f45d\",\"M480 288c-50.1 0-93.6 28.8-114.6 70.8L132.9 126.3l.6-.6 60.1-60.1c87.5-87.5 229.3-87.5 316.8 0c67.1 67.1 82.7 166.3 46.8 248.3C535.8 297.6 509 288 480 288zM113.3 151.9L354.1 392.7c-1.4 7.5-2.1 15.3-2.1 23.3c0 23.2 6.2 44.9 16.9 63.7c-3 .2-6.1 .3-9.2 .3H357c-33.9 0-66.5-13.5-90.5-37.5l-9.8-9.8c-13.1-13.1-34.6-12.4-46.8 1.7L152.2 501c-5.8 6.7-14.2 10.7-23 11s-17.5-3.1-23.8-9.4l-32-32c-6.3-6.3-9.7-14.9-9.4-23.8s4.3-17.2 11-23l66.6-57.7c14-12.2 14.8-33.7 1.7-46.8l-9.8-9.8c-24-24-37.5-56.6-37.5-90.5v-2.7c0-22.8 6.1-44.9 17.3-64.3zM480 320a96 96 0 1 1 0 192 96 96 0 1 1 0-192z\"],\"person-dots-from-line\":[576,512,[\"diagnoses\"],\"f470\",\"M288 176A88 88 0 1 0 288 0a88 88 0 1 0 0 176zM78.7 372.9c15-12.5 50-34.4 97.3-50.1V432H400V322.7c47.3 15.8 82.3 37.7 97.3 50.1c20.4 17 50.6 14.2 67.6-6.1s14.2-50.6-6.1-67.6c-12-10-30.1-22.5-53.2-35C497.2 278.4 481.7 288 464 288c-26.5 0-48-21.5-48-48c0-4.3 .6-8.4 1.6-12.4C379.1 215.9 335.3 208 288 208c-60.2 0-114.9 12.9-160 29.9c0 .7 0 1.4 0 2.1c0 26.5-21.5 48-48 48c-11.8 0-22.7-4.3-31-11.4c-13.1 8.1-23.7 15.9-31.7 22.5c-20.4 17-23.1 47.2-6.1 67.6s47.2 23.1 67.6 6.1zM24 464c-13.3 0-24 10.7-24 24s10.7 24 24 24H552c13.3 0 24-10.7 24-24s-10.7-24-24-24H24zM224 280a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm104 56a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM96 240a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm368 16a16 16 0 1 0 0-32 16 16 0 1 0 0 32z\"],\"trash-can-arrow-up\":[448,512,[\"trash-restore-alt\"],\"f82a\",\"M163.8 0H284.2c12.1 0 23.2 6.8 28.6 17.7L320 32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 96 0 81.7 0 64S14.3 32 32 32h96l7.2-14.3C140.6 6.8 151.7 0 163.8 0zM32 128H416V448c0 35.3-28.7 64-64 64H96c-35.3 0-64-28.7-64-64V128zm192 64c-6.4 0-12.5 2.5-17 7l-80 80c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l39-39V408c0 13.3 10.7 24 24 24s24-10.7 24-24V273.9l39 39c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-80-80c-4.5-4.5-10.6-7-17-7z\"],\"naira-sign\":[448,512,[],\"e1f6\",\"M122.6 46.3c-7.8-11.7-22.4-17-35.9-12.9S64 49.9 64 64V256H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64V448c0 17.7 14.3 32 32 32s32-14.3 32-32V320H228.2l97.2 145.8c7.8 11.7 22.4 17 35.9 12.9s22.7-16.5 22.7-30.6V320h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H384V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V256H262.5L122.6 46.3zM305.1 320H320v22.3L305.1 320zM185.5 256H128V169.7L185.5 256z\"],\"cart-arrow-down\":[576,512,[],\"f218\",\"M24 0C10.7 0 0 10.7 0 24S10.7 48 24 48H69.5c3.8 0 7.1 2.7 7.9 6.5l51.6 271c6.5 34 36.2 58.5 70.7 58.5H488c13.3 0 24-10.7 24-24s-10.7-24-24-24H199.7c-11.5 0-21.4-8.2-23.6-19.5L170.7 288H459.2c32.6 0 61.1-21.8 69.5-53.3l41-152.3C576.6 57 557.4 32 531.1 32H360V134.1l23-23c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-64 64c-9.4 9.4-24.6 9.4-33.9 0l-64-64c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l23 23V32H120.1C111 12.8 91.6 0 69.5 0H24zM176 512a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm336-48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0z\"],\"walkie-talkie\":[384,512,[],\"f8ef\",\"M112 24c0-13.3-10.7-24-24-24S64 10.7 64 24V96H48C21.5 96 0 117.5 0 144V300.1c0 12.7 5.1 24.9 14.1 33.9l3.9 3.9c9 9 14.1 21.2 14.1 33.9V464c0 26.5 21.5 48 48 48H304c26.5 0 48-21.5 48-48V371.9c0-12.7 5.1-24.9 14.1-33.9l3.9-3.9c9-9 14.1-21.2 14.1-33.9V144c0-26.5-21.5-48-48-48H320c0-17.7-14.3-32-32-32s-32 14.3-32 32H224c0-17.7-14.3-32-32-32s-32 14.3-32 32H112V24zm0 136H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],\"file-pen\":[576,512,[128221,\"file-edit\"],\"f31c\",\"M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384V299.6l-94.7 94.7c-8.2 8.2-14 18.5-16.8 29.7l-15 60.1c-2.3 9.4-1.8 19 1.4 27.8H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zM549.8 235.7l14.4 14.4c15.6 15.6 15.6 40.9 0 56.6l-29.4 29.4-71-71 29.4-29.4c15.6-15.6 40.9-15.6 56.6 0zM311.9 417L441.1 287.8l71 71L382.9 487.9c-4.1 4.1-9.2 7-14.9 8.4l-60.1 15c-5.5 1.4-11.2-.2-15.2-4.2s-5.6-9.7-4.2-15.2l15-60.1c1.4-5.6 4.3-10.8 8.4-14.9z\"],receipt:[384,512,[129534],\"f543\",\"M14 2.2C22.5-1.7 32.5-.3 39.6 5.8L80 40.4 120.4 5.8c9-7.7 22.3-7.7 31.2 0L192 40.4 232.4 5.8c9-7.7 22.3-7.7 31.2 0L304 40.4 344.4 5.8c7.1-6.1 17.1-7.5 25.6-3.6s14 12.4 14 21.8V488c0 9.4-5.5 17.9-14 21.8s-18.5 2.5-25.6-3.6L304 471.6l-40.4 34.6c-9 7.7-22.3 7.7-31.2 0L192 471.6l-40.4 34.6c-9 7.7-22.3 7.7-31.2 0L80 471.6 39.6 506.2c-7.1 6.1-17.1 7.5-25.6 3.6S0 497.4 0 488V24C0 14.6 5.5 6.1 14 2.2zM96 144c-8.8 0-16 7.2-16 16s7.2 16 16 16H288c8.8 0 16-7.2 16-16s-7.2-16-16-16H96zM80 352c0 8.8 7.2 16 16 16H288c8.8 0 16-7.2 16-16s-7.2-16-16-16H96c-8.8 0-16 7.2-16 16zM96 240c-8.8 0-16 7.2-16 16s7.2 16 16 16H288c8.8 0 16-7.2 16-16s-7.2-16-16-16H96z\"],\"square-pen\":[448,512,[\"pen-square\",\"pencil-square\"],\"f14b\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM325.8 139.7l14.4 14.4c15.6 15.6 15.6 40.9 0 56.6l-21.4 21.4-71-71 21.4-21.4c15.6-15.6 40.9-15.6 56.6 0zM119.9 289L225.1 183.8l71 71L190.9 359.9c-4.1 4.1-9.2 7-14.9 8.4l-60.1 15c-5.5 1.4-11.2-.2-15.2-4.2s-5.6-9.7-4.2-15.2l15-60.1c1.4-5.6 4.3-10.8 8.4-14.9z\"],\"suitcase-rolling\":[384,512,[],\"f5c1\",\"M144 56c0-4.4 3.6-8 8-8h80c4.4 0 8 3.6 8 8v72H144V56zm176 72H288V56c0-30.9-25.1-56-56-56H152C121.1 0 96 25.1 96 56v72H64c-35.3 0-64 28.7-64 64V416c0 35.3 28.7 64 64 64c0 17.7 14.3 32 32 32s32-14.3 32-32H256c0 17.7 14.3 32 32 32s32-14.3 32-32c35.3 0 64-28.7 64-64V192c0-35.3-28.7-64-64-64zM112 224H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 128H272c8.8 0 16 7.2 16 16s-7.2 16-16 16H112c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],\"person-circle-exclamation\":[576,512,[],\"e53f\",\"M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l44.9 74.7c-16.1 17.6-28.6 38.5-36.6 61.5c-1.9-1.8-3.5-3.9-4.9-6.3L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152zM432 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm0 240a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm0-192c-8.8 0-16 7.2-16 16v80c0 8.8 7.2 16 16 16s16-7.2 16-16V288c0-8.8-7.2-16-16-16z\"],\"chevron-down\":[512,512,[],\"f078\",\"M233.4 406.6c12.5 12.5 32.8 12.5 45.3 0l192-192c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L256 338.7 86.6 169.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l192 192z\"],\"battery-full\":[576,512,[128267,\"battery\",\"battery-5\"],\"f240\",\"M464 160c8.8 0 16 7.2 16 16V336c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16H464zM80 96C35.8 96 0 131.8 0 176V336c0 44.2 35.8 80 80 80H464c44.2 0 80-35.8 80-80V320c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32V176c0-44.2-35.8-80-80-80H80zm368 96H96V320H448V192z\"],\"skull-crossbones\":[448,512,[128369,9760],\"f714\",\"M368 128c0 44.4-25.4 83.5-64 106.4V256c0 17.7-14.3 32-32 32H176c-17.7 0-32-14.3-32-32V234.4c-38.6-23-64-62.1-64-106.4C80 57.3 144.5 0 224 0s144 57.3 144 128zM168 176a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm144-32a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM3.4 273.7c7.9-15.8 27.1-22.2 42.9-14.3L224 348.2l177.7-88.8c15.8-7.9 35-1.5 42.9 14.3s1.5 35-14.3 42.9L295.6 384l134.8 67.4c15.8 7.9 22.2 27.1 14.3 42.9s-27.1 22.2-42.9 14.3L224 419.8 46.3 508.6c-15.8 7.9-35 1.5-42.9-14.3s-1.5-35 14.3-42.9L152.4 384 17.7 316.6C1.9 308.7-4.5 289.5 3.4 273.7z\"],\"code-compare\":[512,512,[],\"e13a\",\"M320 488c0 9.5-5.6 18.1-14.2 21.9s-18.8 2.3-25.8-4.1l-80-72c-5.1-4.6-7.9-11-7.9-17.8s2.9-13.3 7.9-17.8l80-72c7-6.3 17.2-7.9 25.8-4.1s14.2 12.4 14.2 21.9v40h16c35.3 0 64-28.7 64-64V153.3C371.7 141 352 112.8 352 80c0-44.2 35.8-80 80-80s80 35.8 80 80c0 32.8-19.7 61-48 73.3V320c0 70.7-57.3 128-128 128H320v40zM456 80a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zM192 24c0-9.5 5.6-18.1 14.2-21.9s18.8-2.3 25.8 4.1l80 72c5.1 4.6 7.9 11 7.9 17.8s-2.9 13.3-7.9 17.8l-80 72c-7 6.3-17.2 7.9-25.8 4.1s-14.2-12.4-14.2-21.9V128H176c-35.3 0-64 28.7-64 64V358.7c28.3 12.3 48 40.5 48 73.3c0 44.2-35.8 80-80 80s-80-35.8-80-80c0-32.8 19.7-61 48-73.3V192c0-70.7 57.3-128 128-128h16V24zM56 432a24 24 0 1 0 48 0 24 24 0 1 0 -48 0z\"],\"list-ul\":[512,512,[\"list-dots\"],\"f0ca\",\"M64 144a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM192 64c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H192zm0 160c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H192zm0 160c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H192zM64 464a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm48-208a48 48 0 1 0 -96 0 48 48 0 1 0 96 0z\"],\"school-lock\":[640,512,[],\"e56f\",\"M302.2 5.4c10.7-7.2 24.8-7.2 35.5 0L473.7 96H592c26.5 0 48 21.5 48 48V272c0-61.9-50.1-112-112-112s-112 50.1-112 112v24.6c-19.1 11.1-32 31.7-32 55.4H320.3l-.3 0c-35.3 0-64 28.7-64 64v96h64v0H48c-26.5 0-48-21.5-48-48V144c0-26.5 21.5-48 48-48H166.3L302.2 5.4zM80 208v64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V208c0-8.8-7.2-16-16-16H96c-8.8 0-16 7.2-16 16zm0 128v64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V336c0-8.8-7.2-16-16-16H96c-8.8 0-16 7.2-16 16zm240-72a88 88 0 1 0 0-176 88 88 0 1 0 0 176zm16-120v16h16c8.8 0 16 7.2 16 16s-7.2 16-16 16H320c-8.8 0-16-7.2-16-16V144c0-8.8 7.2-16 16-16s16 7.2 16 16zm192 96c-17.7 0-32 14.3-32 32v48h64V272c0-17.7-14.3-32-32-32zm-80 32c0-44.2 35.8-80 80-80s80 35.8 80 80v48c17.7 0 32 14.3 32 32V480c0 17.7-14.3 32-32 32H448c-17.7 0-32-14.3-32-32V352c0-17.7 14.3-32 32-32V272z\"],\"tower-cell\":[576,512,[],\"e585\",\"M62.6 2.3C46.2-4.3 27.6 3.6 20.9 20C7.4 53.4 0 89.9 0 128s7.4 74.6 20.9 108c6.6 16.4 25.3 24.3 41.7 17.7S86.9 228.4 80.3 212C69.8 186.1 64 157.8 64 128s5.8-58.1 16.3-84C86.9 27.6 79 9 62.6 2.3zm450.8 0C497 9 489.1 27.6 495.7 44C506.2 69.9 512 98.2 512 128s-5.8 58.1-16.3 84c-6.6 16.4 1.3 35 17.7 41.7s35-1.3 41.7-17.7c13.5-33.4 20.9-69.9 20.9-108s-7.4-74.6-20.9-108C548.4 3.6 529.8-4.3 513.4 2.3zM340.1 165.2c7.5-10.5 11.9-23.3 11.9-37.2c0-35.3-28.7-64-64-64s-64 28.7-64 64c0 13.9 4.4 26.7 11.9 37.2L98.9 466.8c-7.3 16.1-.2 35.1 15.9 42.4s35.1 .2 42.4-15.9L177.7 448H398.3l20.6 45.2c7.3 16.1 26.3 23.2 42.4 15.9s23.2-26.3 15.9-42.4L340.1 165.2zM369.2 384H206.8l14.5-32H354.7l14.5 32zM288 205.3L325.6 288H250.4L288 205.3zM163.3 73.6c5.3-12.1-.2-26.3-12.4-31.6s-26.3 .2-31.6 12.4C109.5 77 104 101.9 104 128s5.5 51 15.3 73.6c5.3 12.1 19.5 17.7 31.6 12.4s17.7-19.5 12.4-31.6C156 165.8 152 147.4 152 128s4-37.8 11.3-54.4zM456.7 54.4c-5.3-12.1-19.5-17.7-31.6-12.4s-17.7 19.5-12.4 31.6C420 90.2 424 108.6 424 128s-4 37.8-11.3 54.4c-5.3 12.1 .2 26.3 12.4 31.6s26.3-.2 31.6-12.4C466.5 179 472 154.1 472 128s-5.5-51-15.3-73.6z\"],\"down-long\":[320,512,[\"long-arrow-alt-down\"],\"f309\",\"M2 334.5c-3.8 8.8-2 19 4.6 26l136 144c4.5 4.8 10.8 7.5 17.4 7.5s12.9-2.7 17.4-7.5l136-144c6.6-7 8.4-17.2 4.6-26s-12.5-14.5-22-14.5l-72 0 0-288c0-17.7-14.3-32-32-32L128 0C110.3 0 96 14.3 96 32l0 288-72 0c-9.6 0-18.2 5.7-22 14.5z\"],\"ranking-star\":[640,512,[],\"e561\",\"M353.8 54.1L330.2 6.3c-3.9-8.3-16.1-8.6-20.4 0L286.2 54.1l-52.3 7.5c-9.3 1.4-13.3 12.9-6.4 19.8l38 37-9 52.1c-1.4 9.3 8.2 16.5 16.8 12.2l46.9-24.8 46.6 24.4c8.6 4.3 18.3-2.9 16.8-12.2l-9-52.1 38-36.6c6.8-6.8 2.9-18.3-6.4-19.8l-52.3-7.5zM256 256c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H384c17.7 0 32-14.3 32-32V288c0-17.7-14.3-32-32-32H256zM32 320c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H160c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32H32zm416 96v64c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32V416c0-17.7-14.3-32-32-32H480c-17.7 0-32 14.3-32 32z\"],\"chess-king\":[448,512,[9818],\"f43f\",\"M224 0c17.7 0 32 14.3 32 32V48h16c17.7 0 32 14.3 32 32s-14.3 32-32 32H256v48H408c22.1 0 40 17.9 40 40c0 5.3-1 10.5-3.1 15.4L368 400H80L3.1 215.4C1 210.5 0 205.3 0 200c0-22.1 17.9-40 40-40H192V112H176c-17.7 0-32-14.3-32-32s14.3-32 32-32h16V32c0-17.7 14.3-32 32-32zM38.6 473.4L80 432H368l41.4 41.4c4.2 4.2 6.6 10 6.6 16c0 12.5-10.1 22.6-22.6 22.6H54.6C42.1 512 32 501.9 32 489.4c0-6 2.4-11.8 6.6-16z\"],\"person-harassing\":[576,512,[],\"e549\",\"M192 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM59.4 304.5L88 256.9V480c0 17.7 14.3 32 32 32s32-14.3 32-32V352h16V480c0 17.7 14.3 32 32 32s32-14.3 32-32V235.3l47.4 57.1c11.3 13.6 31.5 15.5 45.1 4.2s15.5-31.5 4.2-45.1l-73.7-88.9c-18.2-22-45.3-34.7-73.9-34.7H145.1c-33.7 0-64.9 17.7-82.3 46.6l-58.3 97c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9zM480 240a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM464 344v58.7l-41.4-41.4c-7.3-7.3-17.6-10.6-27.8-9s-18.9 8.1-23.5 17.3l-48 96c-7.9 15.8-1.5 35 14.3 42.9s35 1.5 42.9-14.3L408.8 438l54.7 54.7c12.4 12.4 29.1 19.3 46.6 19.3c36.4 0 65.9-29.5 65.9-65.9V344c0-30.9-25.1-56-56-56s-56 25.1-56 56zM288 48c0 8.8 7.2 16 16 16h56c8.8 0 16-7.2 16-16s-7.2-16-16-16H304c-8.8 0-16 7.2-16 16zm-.8 49.7c-7.9-4-17.5-.7-21.5 7.2s-.7 17.5 7.2 21.5l48 24c7.9 4 17.5 .7 21.5-7.2s.7-17.5-7.2-21.5l-48-24z\"],\"brazilian-real-sign\":[512,512,[],\"e46c\",\"M400 0c17.7 0 32 14.3 32 32V50.2c12.5 2.3 24.7 6.4 36.2 12.1l10.1 5.1c15.8 7.9 22.2 27.1 14.3 42.9s-27.1 22.2-42.9 14.3l-10.2-5.1c-9.9-5-20.9-7.5-32-7.5h-1.7c-29.8 0-53.9 24.1-53.9 53.9c0 22 13.4 41.8 33.9 50l52 20.8c44.7 17.9 74.1 61.2 74.1 109.4v3.4c0 51.2-33.6 94.6-80 109.2V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V460.6c-15-3.5-29.4-9.7-42.3-18.3l-23.4-15.6c-14.7-9.8-18.7-29.7-8.9-44.4s29.7-18.7 44.4-8.9L361.2 389c10.8 7.2 23.4 11 36.3 11c27.9 0 50.5-22.6 50.5-50.5v-3.4c0-22-13.4-41.8-33.9-50l-52-20.8C317.3 257.4 288 214.1 288 165.9C288 114 321.5 70 368 54.2V32c0-17.7 14.3-32 32-32zM0 64C0 46.3 14.3 32 32 32h80c79.5 0 144 64.5 144 144c0 58.8-35.2 109.3-85.7 131.7l51.4 128.4c6.6 16.4-1.4 35-17.8 41.6s-35-1.4-41.6-17.8L106.3 320H64V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V288 64zM64 256h48c44.2 0 80-35.8 80-80s-35.8-80-80-80H64V256z\"],\"landmark-dome\":[512,512,[\"landmark-alt\"],\"f752\",\"M248 0h16c13.3 0 24 10.7 24 24V34.7C368.4 48.1 431.9 111.6 445.3 192H448c17.7 0 32 14.3 32 32s-14.3 32-32 32H64c-17.7 0-32-14.3-32-32s14.3-32 32-32h2.7C80.1 111.6 143.6 48.1 224 34.7V24c0-13.3 10.7-24 24-24zM64 288h64V416h40V288h64V416h48V288h64V416h40V288h64V420.3c.6 .3 1.2 .7 1.7 1.1l48 32c11.7 7.8 17 22.4 12.9 35.9S494.1 512 480 512H32c-14.1 0-26.5-9.2-30.6-22.7s1.1-28.1 12.9-35.9l48-32c.6-.4 1.2-.7 1.8-1.1V288z\"],\"arrow-up\":[384,512,[8593],\"f062\",\"M214.6 41.4c-12.5-12.5-32.8-12.5-45.3 0l-160 160c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L160 141.2V448c0 17.7 14.3 32 32 32s32-14.3 32-32V141.2L329.4 246.6c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-160-160z\"],tv:[640,512,[63717,\"television\",\"tv-alt\"],\"f26c\",\"M64 64V352H576V64H64zM0 64C0 28.7 28.7 0 64 0H576c35.3 0 64 28.7 64 64V352c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM128 448H512c17.7 0 32 14.3 32 32s-14.3 32-32 32H128c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],shrimp:[512,512,[129424],\"e448\",\"M64 32C28.7 32 0 60.7 0 96s28.7 64 64 64h1c3.7 88.9 77 160 167 160h56V128H264 88.8 64c-17.7 0-32-14.3-32-32s14.3-32 32-32H464c8.8 0 16-7.2 16-16s-7.2-16-16-16H64zM224 456c0 13.3 10.7 24 24 24h72V407.8l-64.1-22.4c-12.5-4.4-26.2 2.2-30.6 14.7s2.2 26.2 14.7 30.6l4.5 1.6C233 433.9 224 443.9 224 456zm128 23.3c36.4-3.3 69.5-17.6 96.1-39.6l-86.5-34.6c-3 1.8-6.2 3.2-9.6 4.3v69.9zM472.6 415c24.6-30.3 39.4-68.9 39.4-111c0-12.3-1.3-24.3-3.7-35.9L382.8 355.1c.8 3.4 1.2 7 1.2 10.6c0 4.6-.7 9-1.9 13.1L472.6 415zM336 128H320V320h18.3c9.9 0 19.1 3.2 26.6 8.5l133.5-92.4C471.8 172.6 409.1 128 336 128zM168 192a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z\"],\"list-check\":[512,512,[\"tasks\"],\"f0ae\",\"M152.1 38.2c9.9 8.9 10.7 24 1.8 33.9l-72 80c-4.4 4.9-10.6 7.8-17.2 7.9s-12.9-2.4-17.6-7L7 113C-2.3 103.6-2.3 88.4 7 79s24.6-9.4 33.9 0l22.1 22.1 55.1-61.2c8.9-9.9 24-10.7 33.9-1.8zm0 160c9.9 8.9 10.7 24 1.8 33.9l-72 80c-4.4 4.9-10.6 7.8-17.2 7.9s-12.9-2.4-17.6-7L7 273c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l22.1 22.1 55.1-61.2c8.9-9.9 24-10.7 33.9-1.8zM224 96c0-17.7 14.3-32 32-32H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H256c-17.7 0-32-14.3-32-32zm0 160c0-17.7 14.3-32 32-32H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H256c-17.7 0-32-14.3-32-32zM160 416c0-17.7 14.3-32 32-32H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H192c-17.7 0-32-14.3-32-32zM48 368a48 48 0 1 1 0 96 48 48 0 1 1 0-96z\"],\"jug-detergent\":[384,512,[],\"e519\",\"M96 24c0-13.3 10.7-24 24-24h80c13.3 0 24 10.7 24 24V48h8c13.3 0 24 10.7 24 24s-10.7 24-24 24H88C74.7 96 64 85.3 64 72s10.7-24 24-24h8V24zM0 256c0-70.7 57.3-128 128-128H256c70.7 0 128 57.3 128 128V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V256zm256 0v96c0 17.7 14.3 32 32 32s32-14.3 32-32V256c0-17.7-14.3-32-32-32s-32 14.3-32 32z\"],\"circle-user\":[512,512,[62142,\"user-circle\"],\"f2bd\",\"M399 384.2C376.9 345.8 335.4 320 288 320H224c-47.4 0-88.9 25.8-111 64.2c35.2 39.2 86.2 63.8 143 63.8s107.8-24.7 143-63.8zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm256 16a72 72 0 1 0 0-144 72 72 0 1 0 0 144z\"],\"user-shield\":[640,512,[],\"f505\",\"M224 256A128 128 0 1 0 224 0a128 128 0 1 0 0 256zm-45.7 48C79.8 304 0 383.8 0 482.3C0 498.7 13.3 512 29.7 512H418.3c1.8 0 3.5-.2 5.3-.5c-76.3-55.1-99.8-141-103.1-200.2c-16.1-4.8-33.1-7.3-50.7-7.3H178.3zm308.8-78.3l-120 48C358 277.4 352 286.2 352 296c0 63.3 25.9 168.8 134.8 214.2c5.9 2.5 12.6 2.5 18.5 0C614.1 464.8 640 359.3 640 296c0-9.8-6-18.6-15.1-22.3l-120-48c-5.7-2.3-12.1-2.3-17.8 0zM591.4 312c-3.9 50.7-27.2 116.7-95.4 149.7V273.8L591.4 312z\"],wind:[512,512,[],\"f72e\",\"M288 32c0 17.7 14.3 32 32 32h32c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H352c53 0 96-43 96-96s-43-96-96-96H320c-17.7 0-32 14.3-32 32zm64 352c0 17.7 14.3 32 32 32h32c53 0 96-43 96-96s-43-96-96-96H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H384c-17.7 0-32 14.3-32 32zM128 512h32c53 0 96-43 96-96s-43-96-96-96H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H160c17.7 0 32 14.3 32 32s-14.3 32-32 32H128c-17.7 0-32 14.3-32 32s14.3 32 32 32z\"],\"car-burst\":[640,512,[\"car-crash\"],\"f5e1\",\"M176 8c-6.6 0-12.4 4-14.9 10.1l-29.4 74L55.6 68.9c-6.3-1.9-13.1 .2-17.2 5.3s-4.6 12.2-1.4 17.9l39.5 69.1L10.9 206.4c-5.4 3.7-8 10.3-6.5 16.7s6.7 11.2 13.1 12.2l78.7 12.2L90.6 327c-.5 6.5 3.1 12.7 9 15.5s12.9 1.8 17.8-2.6l35.3-32.5 9.5-35.4 10.4-38.6c8-29.9 30.5-52.1 57.9-60.9l41-59.2c11.3-16.3 26.4-28.9 43.5-37.2c-.4-.6-.8-1.2-1.3-1.8c-4.1-5.1-10.9-7.2-17.2-5.3L220.3 92.1l-29.4-74C188.4 12 182.6 8 176 8zM367.7 161.5l135.6 36.3c6.5 1.8 11.3 7.4 11.8 14.2l4.6 56.5-201.5-54 32.2-46.6c3.8-5.6 10.8-8.1 17.3-6.4zm-69.9-30l-47.9 69.3c-21.6 3-40.3 18.6-46.3 41l-10.4 38.6-16.6 61.8-8.3 30.9c-4.6 17.1 5.6 34.6 22.6 39.2l15.5 4.1c17.1 4.6 34.6-5.6 39.2-22.6l8.3-30.9 247.3 66.3-8.3 30.9c-4.6 17.1 5.6 34.6 22.6 39.2l15.5 4.1c17.1 4.6 34.6-5.6 39.2-22.6l8.3-30.9L595 388l10.4-38.6c6-22.4-2.5-45.2-19.6-58.7l-6.8-84c-2.7-33.7-26.4-62-59-70.8L384.2 99.7c-32.7-8.8-67.3 4-86.5 31.8zm-17 131a24 24 0 1 1 -12.4 46.4 24 24 0 1 1 12.4-46.4zm217.9 83.2A24 24 0 1 1 545 358.1a24 24 0 1 1 -46.4-12.4z\"],y:[384,512,[121],\"59\",\"M58 45.4C47.8 31 27.8 27.7 13.4 38S-4.3 68.2 6 82.6L160 298.3V448c0 17.7 14.3 32 32 32s32-14.3 32-32V298.3L378 82.6c10.3-14.4 6.9-34.4-7.4-44.6S336.2 31 326 45.4L192 232.9 58 45.4z\"],\"person-snowboarding\":[512,512,[127938,\"snowboarding\"],\"f7ce\",\"M209.7 3.4c15.8-7.9 35-1.5 42.9 14.3l25 50 42.4 8.5c19.5 3.9 37.8 12.3 53.5 24.5l126.1 98.1c14 10.9 16.5 31 5.6 44.9s-31 16.5-44.9 5.6l-72.1-56.1-71.5 31.8 33.1 27.6c23.2 19.3 33.5 50 26.7 79.4l-17.4 75.2c-2.2 9.4-8.2 16.8-16.1 21l86.5 33.1c4.6 1.8 9.4 2.6 14.3 2.6H472c13.3 0 24 10.7 24 24s-10.7 24-24 24H443.8c-10.8 0-21.4-2-31.5-5.8L60.1 371.3c-11.5-4.4-22-11.2-30.8-20L7 329c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l22.4 22.4c4 4 8.7 7.1 14 9.1l22.4 8.6c-.8-1.6-1.5-3.2-2.1-4.9c-5.6-16.8 3.5-34.9 20.2-40.5L192 264.9l0-53.2c0-24.2 13.7-46.4 35.4-57.2l45.2-22.6-7.5-1.5c-19.4-3.9-35.9-16.5-44.7-34.1l-25-50c-7.9-15.8-1.5-35 14.3-42.9zM139 350.1l159 60.9c-2.1-5.6-2.6-11.9-1.1-18.2l17.4-75.2c1.4-5.9-.7-12-5.3-15.9l-52.8-44 0 18.8c0 20.7-13.2 39-32.8 45.5L139 350.1zM432 0a48 48 0 1 1 0 96 48 48 0 1 1 0-96z\"],\"truck-fast\":[640,512,[\"shipping-fast\"],\"f48b\",\"M112 0C85.5 0 64 21.5 64 48V96H16c-8.8 0-16 7.2-16 16s7.2 16 16 16H64 272c8.8 0 16 7.2 16 16s-7.2 16-16 16H64 48c-8.8 0-16 7.2-16 16s7.2 16 16 16H64 240c8.8 0 16 7.2 16 16s-7.2 16-16 16H64 16c-8.8 0-16 7.2-16 16s7.2 16 16 16H64 208c8.8 0 16 7.2 16 16s-7.2 16-16 16H64V416c0 53 43 96 96 96s96-43 96-96H384c0 53 43 96 96 96s96-43 96-96h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V288 256 237.3c0-17-6.7-33.3-18.7-45.3L512 114.7c-12-12-28.3-18.7-45.3-18.7H416V48c0-26.5-21.5-48-48-48H112zM544 237.3V256H416V160h50.7L544 237.3zM160 368a48 48 0 1 1 0 96 48 48 0 1 1 0-96zm272 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0z\"],fish:[576,512,[128031],\"f578\",\"M180.5 141.5C219.7 108.5 272.6 80 336 80s116.3 28.5 155.5 61.5c39.1 33 66.9 72.4 81 99.8c4.7 9.2 4.7 20.1 0 29.3c-14.1 27.4-41.9 66.8-81 99.8C452.3 403.5 399.4 432 336 432s-116.3-28.5-155.5-61.5c-16.2-13.7-30.5-28.5-42.7-43.1L48.1 379.6c-12.5 7.3-28.4 5.3-38.7-4.9S-3 348.7 4.2 336.1L50 256 4.2 175.9c-7.2-12.6-5-28.4 5.3-38.6s26.1-12.2 38.7-4.9l89.7 52.3c12.2-14.6 26.5-29.4 42.7-43.1zM448 256a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z\"],\"user-graduate\":[448,512,[],\"f501\",\"M219.3 .5c3.1-.6 6.3-.6 9.4 0l200 40C439.9 42.7 448 52.6 448 64s-8.1 21.3-19.3 23.5L352 102.9V160c0 70.7-57.3 128-128 128s-128-57.3-128-128V102.9L48 93.3v65.1l15.7 78.4c.9 4.7-.3 9.6-3.3 13.3s-7.6 5.9-12.4 5.9H16c-4.8 0-9.3-2.1-12.4-5.9s-4.3-8.6-3.3-13.3L16 158.4V86.6C6.5 83.3 0 74.3 0 64C0 52.6 8.1 42.7 19.3 40.5l200-40zM111.9 327.7c10.5-3.4 21.8 .4 29.4 8.5l71 75.5c6.3 6.7 17 6.7 23.3 0l71-75.5c7.6-8.1 18.9-11.9 29.4-8.5C401 348.6 448 409.4 448 481.3c0 17-13.8 30.7-30.7 30.7H30.7C13.8 512 0 498.2 0 481.3c0-71.9 47-132.7 111.9-153.6z\"],\"circle-half-stroke\":[512,512,[9680,\"adjust\"],\"f042\",\"M448 256c0-106-86-192-192-192V448c106 0 192-86 192-192zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256z\"],clapperboard:[512,512,[],\"e131\",\"M448 32H361.9l-1 1-127 127h92.1l1-1L453.8 32.3c-1.9-.2-3.8-.3-5.8-.3zm64 128V96c0-15.1-5.3-29.1-14-40l-104 104H512zM294.1 32H201.9l-1 1L73.9 160h92.1l1-1 127-127zM64 32C28.7 32 0 60.7 0 96v64H6.1l1-1 127-127H64zM512 192H0V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V192z\"],\"circle-radiation\":[512,512,[9762,\"radiation-alt\"],\"f7ba\",\"M256 64a192 192 0 1 1 0 384 192 192 0 1 1 0-384zm0 448A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM200 256c0-20.7 11.3-38.8 28-48.5l-36-62.3c-8.8-15.3-28.7-20.8-42-9c-25.6 22.6-43.9 53.3-50.9 88.1C95.7 241.5 110.3 256 128 256l72 0zm28 48.5l-36 62.4c-8.8 15.3-3.6 35.2 13.1 40.8c16 5.4 33.1 8.3 50.9 8.3s34.9-2.9 50.9-8.3c16.7-5.6 21.9-25.5 13.1-40.8l-36-62.4c-8.2 4.8-17.8 7.5-28 7.5s-19.8-2.7-28-7.5zM312 256l72 0c17.7 0 32.3-14.5 28.8-31.8c-7-34.8-25.3-65.5-50.9-88.1c-13.2-11.7-33.1-6.3-42 9l-36 62.3c16.7 9.7 28 27.8 28 48.5zm-56 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],baseball:[512,512,[129358,9918,\"baseball-ball\"],\"f433\",\"M62.7 223.4c-4.8 .4-9.7 .6-14.7 .6c-15.6 0-30.8-2-45.2-5.9C19.2 107.1 107.1 19.2 218.1 2.8C222 17.2 224 32.4 224 48c0 4.9-.2 9.8-.6 14.7c-.7 8.8 5.8 16.5 14.6 17.3s16.5-5.8 17.3-14.6c.5-5.7 .7-11.5 .7-17.3c0-16.5-1.9-32.6-5.6-47.9c1.8 0 3.7-.1 5.6-.1C397.4 0 512 114.6 512 256c0 1.9 0 3.7-.1 5.6c-15.4-3.6-31.4-5.6-47.9-5.6c-5.8 0-11.6 .2-17.3 .7c-8.8 .7-15.4 8.5-14.6 17.3s8.5 15.4 17.3 14.6c4.8-.4 9.7-.6 14.7-.6c15.6 0 30.8 2 45.2 5.9C492.8 404.9 404.9 492.8 293.9 509.2C290 494.8 288 479.6 288 464c0-4.9 .2-9.8 .6-14.7c.7-8.8-5.8-16.5-14.6-17.3s-16.5 5.8-17.3 14.6c-.5 5.7-.7 11.5-.7 17.3c0 16.5 1.9 32.6 5.6 47.9c-1.8 0-3.7 .1-5.6 .1C114.6 512 0 397.4 0 256c0-1.9 0-3.7 .1-5.6C15.4 254.1 31.5 256 48 256c5.8 0 11.6-.2 17.3-.7c8.8-.7 15.4-8.5 14.6-17.3s-8.5-15.4-17.3-14.6zM121.3 208c-8 3.7-11.6 13.2-7.9 21.2s13.2 11.6 21.2 7.9c45.2-20.8 81.7-57.2 102.5-102.5c3.7-8 .2-17.5-7.9-21.2s-17.5-.2-21.2 7.9c-17.6 38.3-48.5 69.2-86.7 86.7zm277.2 74.7c-3.7-8-13.2-11.6-21.2-7.9c-45.2 20.8-81.7 57.2-102.5 102.5c-3.7 8-.2 17.5 7.9 21.2s17.5 .2 21.2-7.9c17.6-38.3 48.5-69.2 86.7-86.7c8-3.7 11.6-13.2 7.9-21.2z\"],\"jet-fighter-up\":[512,512,[],\"e518\",\"M270.7 9.7C268.2 3.8 262.4 0 256 0s-12.2 3.8-14.7 9.7L197.2 112.6c-3.4 8-5.2 16.5-5.2 25.2v77l-144 84V280c0-13.3-10.7-24-24-24s-24 10.7-24 24v56 32 24c0 13.3 10.7 24 24 24s24-10.7 24-24v-8H192v32.7L133.5 468c-3.5 3-5.5 7.4-5.5 12v16c0 8.8 7.2 16 16 16h96V448c0-8.8 7.2-16 16-16s16 7.2 16 16v64h96c8.8 0 16-7.2 16-16V480c0-4.6-2-9-5.5-12L320 416.7V384H464v8c0 13.3 10.7 24 24 24s24-10.7 24-24V368 336 280c0-13.3-10.7-24-24-24s-24 10.7-24 24v18.8l-144-84v-77c0-8.7-1.8-17.2-5.2-25.2L270.7 9.7z\"],\"diagram-project\":[576,512,[\"project-diagram\"],\"f542\",\"M0 80C0 53.5 21.5 32 48 32h96c26.5 0 48 21.5 48 48V96H384V80c0-26.5 21.5-48 48-48h96c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H432c-26.5 0-48-21.5-48-48V160H192v16c0 1.7-.1 3.4-.3 5L272 288h96c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H272c-26.5 0-48-21.5-48-48V336c0-1.7 .1-3.4 .3-5L144 224H48c-26.5 0-48-21.5-48-48V80z\"],copy:[448,512,[],\"f0c5\",\"M208 0H332.1c12.7 0 24.9 5.1 33.9 14.1l67.9 67.9c9 9 14.1 21.2 14.1 33.9V336c0 26.5-21.5 48-48 48H208c-26.5 0-48-21.5-48-48V48c0-26.5 21.5-48 48-48zM48 128h80v64H64V448H256V416h64v48c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V176c0-26.5 21.5-48 48-48z\"],\"volume-xmark\":[576,512,[\"volume-mute\",\"volume-times\"],\"f6a9\",\"M301.1 34.8C312.6 40 320 51.4 320 64V448c0 12.6-7.4 24-18.9 29.2s-25 3.1-34.4-5.3L131.8 352H64c-35.3 0-64-28.7-64-64V224c0-35.3 28.7-64 64-64h67.8L266.7 40.1c9.4-8.4 22.9-10.4 34.4-5.3zM425 167l55 55 55-55c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-55 55 55 55c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-55-55-55 55c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l55-55-55-55c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0z\"],\"hand-sparkles\":[640,512,[],\"e05d\",\"M320 0c17.7 0 32 14.3 32 32V240c0 8.8 7.2 16 16 16s16-7.2 16-16V64c0-17.7 14.3-32 32-32s32 14.3 32 32V240c0 8.8 7.2 16 16 16s16-7.2 16-16V128c0-17.7 14.3-32 32-32s32 14.3 32 32V323.1c-11.9 4.8-21.3 14.9-25 27.8l-8.9 31.2L478.9 391C460.6 396.3 448 413 448 432c0 18.9 12.5 35.6 30.6 40.9C448.4 497.4 409.9 512 368 512H348.8c-59.6 0-116.9-22.9-160-64L76.4 341c-16-15.2-16.6-40.6-1.4-56.6s40.6-16.6 56.6-1.4l60.5 57.6c0-1.5-.1-3.1-.1-4.6V64c0-17.7 14.3-32 32-32s32 14.3 32 32V240c0 8.8 7.2 16 16 16s16-7.2 16-16V32c0-17.7 14.3-32 32-32zm-7.3 326.6c-1.1-3.9-4.7-6.6-8.7-6.6s-7.6 2.7-8.7 6.6L288 352l-25.4 7.3c-3.9 1.1-6.6 4.7-6.6 8.7s2.7 7.6 6.6 8.7L288 384l7.3 25.4c1.1 3.9 4.7 6.6 8.7 6.6s7.6-2.7 8.7-6.6L320 384l25.4-7.3c3.9-1.1 6.6-4.7 6.6-8.7s-2.7-7.6-6.6-8.7L320 352l-7.3-25.4zM104 120l48.3 13.8c4.6 1.3 7.7 5.5 7.7 10.2s-3.1 8.9-7.7 10.2L104 168 90.2 216.3c-1.3 4.6-5.5 7.7-10.2 7.7s-8.9-3.1-10.2-7.7L56 168 7.7 154.2C3.1 152.9 0 148.7 0 144s3.1-8.9 7.7-10.2L56 120 69.8 71.7C71.1 67.1 75.3 64 80 64s8.9 3.1 10.2 7.7L104 120zM584 408l48.3 13.8c4.6 1.3 7.7 5.5 7.7 10.2s-3.1 8.9-7.7 10.2L584 456l-13.8 48.3c-1.3 4.6-5.5 7.7-10.2 7.7s-8.9-3.1-10.2-7.7L536 456l-48.3-13.8c-4.6-1.3-7.7-5.5-7.7-10.2s3.1-8.9 7.7-10.2L536 408l13.8-48.3c1.3-4.6 5.5-7.7 10.2-7.7s8.9 3.1 10.2 7.7L584 408z\"],grip:[448,512,[\"grip-horizontal\"],\"f58d\",\"M128 136c0-22.1-17.9-40-40-40L40 96C17.9 96 0 113.9 0 136l0 48c0 22.1 17.9 40 40 40H88c22.1 0 40-17.9 40-40l0-48zm0 192c0-22.1-17.9-40-40-40H40c-22.1 0-40 17.9-40 40l0 48c0 22.1 17.9 40 40 40H88c22.1 0 40-17.9 40-40V328zm32-192v48c0 22.1 17.9 40 40 40h48c22.1 0 40-17.9 40-40V136c0-22.1-17.9-40-40-40l-48 0c-22.1 0-40 17.9-40 40zM288 328c0-22.1-17.9-40-40-40H200c-22.1 0-40 17.9-40 40l0 48c0 22.1 17.9 40 40 40h48c22.1 0 40-17.9 40-40V328zm32-192v48c0 22.1 17.9 40 40 40h48c22.1 0 40-17.9 40-40V136c0-22.1-17.9-40-40-40l-48 0c-22.1 0-40 17.9-40 40zM448 328c0-22.1-17.9-40-40-40H360c-22.1 0-40 17.9-40 40v48c0 22.1 17.9 40 40 40h48c22.1 0 40-17.9 40-40V328z\"],\"share-from-square\":[576,512,[61509,\"share-square\"],\"f14d\",\"M352 224H305.5c-45 0-81.5 36.5-81.5 81.5c0 22.3 10.3 34.3 19.2 40.5c6.8 4.7 12.8 12 12.8 20.3c0 9.8-8 17.8-17.8 17.8h-2.5c-2.4 0-4.8-.4-7.1-1.4C210.8 374.8 128 333.4 128 240c0-79.5 64.5-144 144-144h80V34.7C352 15.5 367.5 0 386.7 0c8.6 0 16.8 3.2 23.2 8.9L548.1 133.3c7.6 6.8 11.9 16.5 11.9 26.7s-4.3 19.9-11.9 26.7l-139 125.1c-5.9 5.3-13.5 8.2-21.4 8.2H384c-17.7 0-32-14.3-32-32V224zM80 96c-8.8 0-16 7.2-16 16V432c0 8.8 7.2 16 16 16H400c8.8 0 16-7.2 16-16V384c0-17.7 14.3-32 32-32s32 14.3 32 32v48c0 44.2-35.8 80-80 80H80c-44.2 0-80-35.8-80-80V112C0 67.8 35.8 32 80 32h48c17.7 0 32 14.3 32 32s-14.3 32-32 32H80z\"],\"child-combatant\":[576,512,[\"child-rifle\"],\"e4e0\",\"M176 128A64 64 0 1 0 176 0a64 64 0 1 0 0 128zm-8 352V352h16V480c0 17.7 14.3 32 32 32s32-14.3 32-32V300.5L260.9 321c9.4 15 29.2 19.4 44.1 10s19.4-29.2 10-44.1l-51.7-82.1c-17.6-27.9-48.3-44.9-81.2-44.9H169.8c-33 0-63.7 16.9-81.2 44.9L36.9 287c-9.4 15-4.9 34.7 10 44.1s34.7 4.9 44.1-10L104 300.5V480c0 17.7 14.3 32 32 32s32-14.3 32-32zM448 0H432 416c-8.8 0-16 7.2-16 16s7.2 16 16 16V132.3c-9.6 5.5-16 15.9-16 27.7v32c-17.7 0-32 14.3-32 32V368c0 17.7 14.3 32 32 32h16v96c0 8.8 7.2 16 16 16h59.5c10.4 0 18-9.8 15.5-19.9L484 400h44c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H480V325.3l53.1-17.7c6.5-2.2 10.9-8.3 10.9-15.2V208c0-8.8-7.2-16-16-16H512c-8.8 0-16 7.2-16 16v56l-16 5.3V160c0-11.8-6.4-22.2-16-27.7V16c0-8.8-7.2-16-16-16z\"],gun:[576,512,[],\"e19b\",\"M528 56c0-13.3-10.7-24-24-24s-24 10.7-24 24v8H32C14.3 64 0 78.3 0 96V208c0 17.7 14.3 32 32 32H42c20.8 0 36.1 19.6 31 39.8L33 440.2c-2.4 9.6-.2 19.7 5.8 27.5S54.1 480 64 480h96c14.7 0 27.5-10 31-24.2L217 352H321.4c23.7 0 44.8-14.9 52.7-37.2L400.9 240H432c8.5 0 16.6-3.4 22.6-9.4L477.3 208H544c17.7 0 32-14.3 32-32V96c0-17.7-14.3-32-32-32H528V56zM321.4 304H229l16-64h105l-21 58.7c-1.1 3.2-4.2 5.3-7.5 5.3zM80 128H464c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],\"square-phone\":[448,512,[\"phone-square\"],\"f098\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm90.7 96.7c9.7-2.6 19.9 2.3 23.7 11.6l20 48c3.4 8.2 1 17.6-5.8 23.2L168 231.7c16.6 35.2 45.1 63.7 80.3 80.3l20.2-24.7c5.6-6.8 15-9.2 23.2-5.8l48 20c9.3 3.9 14.2 14 11.6 23.7l-12 44C336.9 378 329 384 320 384C196.3 384 96 283.7 96 160c0-9 6-16.9 14.7-19.3l44-12z\"],plus:[448,512,[10133,61543,\"add\"],\"2b\",\"M256 80c0-17.7-14.3-32-32-32s-32 14.3-32 32V224H48c-17.7 0-32 14.3-32 32s14.3 32 32 32H192V432c0 17.7 14.3 32 32 32s32-14.3 32-32V288H400c17.7 0 32-14.3 32-32s-14.3-32-32-32H256V80z\"],expand:[448,512,[],\"f065\",\"M32 32C14.3 32 0 46.3 0 64v96c0 17.7 14.3 32 32 32s32-14.3 32-32V96h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H32zM64 352c0-17.7-14.3-32-32-32s-32 14.3-32 32v96c0 17.7 14.3 32 32 32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H64V352zM320 32c-17.7 0-32 14.3-32 32s14.3 32 32 32h64v64c0 17.7 14.3 32 32 32s32-14.3 32-32V64c0-17.7-14.3-32-32-32H320zM448 352c0-17.7-14.3-32-32-32s-32 14.3-32 32v64H320c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c17.7 0 32-14.3 32-32V352z\"],computer:[640,512,[],\"e4e5\",\"M384 96V320H64L64 96H384zM64 32C28.7 32 0 60.7 0 96V320c0 35.3 28.7 64 64 64H181.3l-10.7 32H96c-17.7 0-32 14.3-32 32s14.3 32 32 32H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H277.3l-10.7-32H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm464 0c-26.5 0-48 21.5-48 48V432c0 26.5 21.5 48 48 48h64c26.5 0 48-21.5 48-48V80c0-26.5-21.5-48-48-48H528zm16 64h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H544c-8.8 0-16-7.2-16-16s7.2-16 16-16zm-16 80c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16s-7.2 16-16 16H544c-8.8 0-16-7.2-16-16zm32 160a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],xmark:[384,512,[128473,10005,10006,10060,215,\"close\",\"multiply\",\"remove\",\"times\"],\"f00d\",\"M342.6 150.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L192 210.7 86.6 105.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L146.7 256 41.4 361.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L192 301.3 297.4 406.6c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L237.3 256 342.6 150.6z\"],\"arrows-up-down-left-right\":[512,512,[\"arrows\"],\"f047\",\"M278.6 9.4c-12.5-12.5-32.8-12.5-45.3 0l-64 64c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l9.4-9.4V224H109.3l9.4-9.4c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-64 64c-12.5 12.5-12.5 32.8 0 45.3l64 64c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-9.4-9.4H224V402.7l-9.4-9.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l64 64c12.5 12.5 32.8 12.5 45.3 0l64-64c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-9.4 9.4V288H402.7l-9.4 9.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l64-64c12.5-12.5 12.5-32.8 0-45.3l-64-64c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l9.4 9.4H288V109.3l9.4 9.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-64-64z\"],\"chalkboard-user\":[640,512,[\"chalkboard-teacher\"],\"f51c\",\"M160 64c0-35.3 28.7-64 64-64H576c35.3 0 64 28.7 64 64V352c0 35.3-28.7 64-64 64H336.8c-11.8-25.5-29.9-47.5-52.4-64H384V320c0-17.7 14.3-32 32-32h64c17.7 0 32 14.3 32 32v32h64V64L224 64v49.1C205.2 102.2 183.3 96 160 96V64zm0 64a96 96 0 1 1 0 192 96 96 0 1 1 0-192zM133.3 352h53.3C260.3 352 320 411.7 320 485.3c0 14.7-11.9 26.7-26.7 26.7H26.7C11.9 512 0 500.1 0 485.3C0 411.7 59.7 352 133.3 352z\"],\"peso-sign\":[384,512,[],\"e222\",\"M64 32C46.3 32 32 46.3 32 64v64c-17.7 0-32 14.3-32 32s14.3 32 32 32l0 32c-17.7 0-32 14.3-32 32s14.3 32 32 32l0 64v96c0 17.7 14.3 32 32 32s32-14.3 32-32V384h80c68.4 0 127.7-39 156.8-96H352c17.7 0 32-14.3 32-32s-14.3-32-32-32h-.7c.5-5.3 .7-10.6 .7-16s-.2-10.7-.7-16h.7c17.7 0 32-14.3 32-32s-14.3-32-32-32H332.8C303.7 71 244.4 32 176 32H64zm190.4 96H96V96h80c30.5 0 58.2 12.2 78.4 32zM96 192H286.9c.7 5.2 1.1 10.6 1.1 16s-.4 10.8-1.1 16H96V192zm158.4 96c-20.2 19.8-47.9 32-78.4 32H96V288H254.4z\"],\"building-shield\":[576,512,[],\"e4d8\",\"M0 48C0 21.5 21.5 0 48 0H336c26.5 0 48 21.5 48 48V207l-42.4 17H304 272c-8.8 0-16 7.2-16 16v32 24.2V304c0 .9 .1 1.7 .2 2.6c2.3 58.1 24.1 144.8 98.7 201.5c-5.8 2.5-12.2 3.9-18.9 3.9H240V432c0-26.5-21.5-48-48-48s-48 21.5-48 48v80H48c-26.5 0-48-21.5-48-48V48zM80 224c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H80zm80 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H176c-8.8 0-16 7.2-16 16zM64 112v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H80c-8.8 0-16 7.2-16 16zM176 96c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H176zm80 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H272c-8.8 0-16 7.2-16 16zM423.1 225.7c5.7-2.3 12.1-2.3 17.8 0l120 48C570 277.4 576 286.2 576 296c0 63.3-25.9 168.8-134.8 214.2c-5.9 2.5-12.6 2.5-18.5 0C313.9 464.8 288 359.3 288 296c0-9.8 6-18.6 15.1-22.3l120-48zM527.4 312L432 273.8V461.7c68.2-33 91.5-99 95.4-149.7z\"],baby:[448,512,[],\"f77c\",\"M152 88a72 72 0 1 1 144 0A72 72 0 1 1 152 88zM39.7 144.5c13-17.9 38-21.8 55.9-8.8L131.8 162c26.8 19.5 59.1 30 92.2 30s65.4-10.5 92.2-30l36.2-26.4c17.9-13 42.9-9 55.9 8.8s9 42.9-8.8 55.9l-36.2 26.4c-13.6 9.9-28.1 18.2-43.3 25V288H128V251.7c-15.2-6.7-29.7-15.1-43.3-25L48.5 200.3c-17.9-13-21.8-38-8.8-55.9zm89.8 184.8l60.6 53-26 37.2 24.3 24.3c15.6 15.6 15.6 40.9 0 56.6s-40.9 15.6-56.6 0l-48-48C70 438.6 68.1 417 79.2 401.1l50.2-71.8zm128.5 53l60.6-53 50.2 71.8c11.1 15.9 9.2 37.5-4.5 51.2l-48 48c-15.6 15.6-40.9 15.6-56.6 0s-15.6-40.9 0-56.6L284 419.4l-26-37.2z\"],\"users-line\":[640,512,[],\"e592\",\"M211.2 96a64 64 0 1 0 -128 0 64 64 0 1 0 128 0zM32 256c0 17.7 14.3 32 32 32h85.6c10.1-39.4 38.6-71.5 75.8-86.6c-9.7-6-21.2-9.4-33.4-9.4H96c-35.3 0-64 28.7-64 64zm461.6 32H576c17.7 0 32-14.3 32-32c0-35.3-28.7-64-64-64H448c-11.7 0-22.7 3.1-32.1 8.6c38.1 14.8 67.4 47.3 77.7 87.4zM391.2 226.4c-6.9-1.6-14.2-2.4-21.6-2.4h-96c-8.5 0-16.7 1.1-24.5 3.1c-30.8 8.1-55.6 31.1-66.1 60.9c-3.5 10-5.5 20.8-5.5 32c0 17.7 14.3 32 32 32h224c17.7 0 32-14.3 32-32c0-11.2-1.9-22-5.5-32c-10.8-30.7-36.8-54.2-68.9-61.6zM563.2 96a64 64 0 1 0 -128 0 64 64 0 1 0 128 0zM321.6 192a80 80 0 1 0 0-160 80 80 0 1 0 0 160zM32 416c-17.7 0-32 14.3-32 32s14.3 32 32 32H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H32z\"],\"quote-left\":[448,512,[8220,\"quote-left-alt\"],\"f10d\",\"M0 216C0 149.7 53.7 96 120 96h8c17.7 0 32 14.3 32 32s-14.3 32-32 32h-8c-30.9 0-56 25.1-56 56v8h64c35.3 0 64 28.7 64 64v64c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V320 288 216zm256 0c0-66.3 53.7-120 120-120h8c17.7 0 32 14.3 32 32s-14.3 32-32 32h-8c-30.9 0-56 25.1-56 56v8h64c35.3 0 64 28.7 64 64v64c0 35.3-28.7 64-64 64H320c-35.3 0-64-28.7-64-64V320 288 216z\"],tractor:[640,512,[128668],\"f722\",\"M96 64c0-35.3 28.7-64 64-64H266.3c26.2 0 49.7 15.9 59.4 40.2L373.7 160H480V126.2c0-24.8 5.8-49.3 16.9-71.6l2.5-5c7.9-15.8 27.1-22.2 42.9-14.3s22.2 27.1 14.3 42.9l-2.5 5c-6.7 13.3-10.1 28-10.1 42.9V160h56c22.1 0 40 17.9 40 40v45.4c0 16.5-8.5 31.9-22.6 40.7l-43.3 27.1c-14.2-5.9-29.8-9.2-46.1-9.2c-39.3 0-74.1 18.9-96 48H352c0 17.7-14.3 32-32 32h-8.2c-1.7 4.8-3.7 9.5-5.8 14.1l5.8 5.8c12.5 12.5 12.5 32.8 0 45.3l-22.6 22.6c-12.5 12.5-32.8 12.5-45.3 0l-5.8-5.8c-4.6 2.2-9.3 4.1-14.1 5.8V480c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32v-8.2c-4.8-1.7-9.5-3.7-14.1-5.8l-5.8 5.8c-12.5 12.5-32.8 12.5-45.3 0L40.2 449.1c-12.5-12.5-12.5-32.8 0-45.3l5.8-5.8c-2.2-4.6-4.1-9.3-5.8-14.1H32c-17.7 0-32-14.3-32-32V320c0-17.7 14.3-32 32-32h8.2c1.7-4.8 3.7-9.5 5.8-14.1l-5.8-5.8c-12.5-12.5-12.5-32.8 0-45.3l22.6-22.6c9-9 21.9-11.5 33.1-7.6V192 160 64zm170.3 0H160v96h32H304.7L266.3 64zM176 256a80 80 0 1 0 0 160 80 80 0 1 0 0-160zM528 448a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm0 64c-48.6 0-88-39.4-88-88c0-29.8 14.8-56.1 37.4-72c14.3-10.1 31.8-16 50.6-16c2.7 0 5.3 .1 7.9 .3c44.9 4 80.1 41.7 80.1 87.7c0 48.6-39.4 88-88 88z\"],\"trash-arrow-up\":[448,512,[\"trash-restore\"],\"f829\",\"M163.8 0H284.2c12.1 0 23.2 6.8 28.6 17.7L320 32h96c17.7 0 32 14.3 32 32s-14.3 32-32 32H32C14.3 96 0 81.7 0 64S14.3 32 32 32h96l7.2-14.3C140.6 6.8 151.7 0 163.8 0zM32 128H416L394.8 467c-1.6 25.3-22.6 45-47.9 45H101.1c-25.3 0-46.3-19.7-47.9-45L32 128zm192 64c-6.4 0-12.5 2.5-17 7l-80 80c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l39-39V408c0 13.3 10.7 24 24 24s24-10.7 24-24V273.9l39 39c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-80-80c-4.5-4.5-10.6-7-17-7z\"],\"arrow-down-up-lock\":[640,512,[],\"e4b0\",\"M150.6 502.6l96-96c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L160 402.7V288H416V272c0-17.2 3.9-33.5 10.8-48H352V109.3l41.4 41.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-96-96c-6-6-14.1-9.4-22.6-9.4s-16.6 3.4-22.6 9.4l-96 96c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L288 109.3V224l-128 0H96l-64 0c-17.7 0-32 14.3-32 32s14.3 32 32 32H96V402.7L54.6 361.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l96 96c12.5 12.5 32.8 12.5 45.3 0zM160 192V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V192h64zM288 320V448c0 17.7 14.3 32 32 32s32-14.3 32-32V320H288zm240-80c17.7 0 32 14.3 32 32v48H496V272c0-17.7 14.3-32 32-32zm-80 32v48c-17.7 0-32 14.3-32 32V480c0 17.7 14.3 32 32 32H608c17.7 0 32-14.3 32-32V352c0-17.7-14.3-32-32-32V272c0-44.2-35.8-80-80-80s-80 35.8-80 80z\"],\"lines-leaning\":[384,512,[],\"e51e\",\"M190.4 74.1c5.6-16.8-3.5-34.9-20.2-40.5s-34.9 3.5-40.5 20.2l-128 384c-5.6 16.8 3.5 34.9 20.2 40.5s34.9-3.5 40.5-20.2l128-384zm70.9-41.7c-17.4-2.9-33.9 8.9-36.8 26.3l-64 384c-2.9 17.4 8.9 33.9 26.3 36.8s33.9-8.9 36.8-26.3l64-384c2.9-17.4-8.9-33.9-26.3-36.8zM352 32c-17.7 0-32 14.3-32 32V448c0 17.7 14.3 32 32 32s32-14.3 32-32V64c0-17.7-14.3-32-32-32z\"],\"ruler-combined\":[512,512,[],\"f546\",\"M.2 468.9C2.7 493.1 23.1 512 48 512l96 0 320 0c26.5 0 48-21.5 48-48l0-96c0-26.5-21.5-48-48-48l-48 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-64 0 0 80c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-80-80 0c-8.8 0-16-7.2-16-16s7.2-16 16-16l80 0 0-64-80 0c-8.8 0-16-7.2-16-16s7.2-16 16-16l80 0 0-64-80 0c-8.8 0-16-7.2-16-16s7.2-16 16-16l80 0 0-48c0-26.5-21.5-48-48-48L48 0C21.5 0 0 21.5 0 48L0 368l0 96c0 1.7 .1 3.3 .2 4.9z\"],copyright:[512,512,[169],\"f1f9\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM199.4 312.6c31.2 31.2 81.9 31.2 113.1 0c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9c-50 50-131 50-181 0s-50-131 0-181s131-50 181 0c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0c-31.2-31.2-81.9-31.2-113.1 0s-31.2 81.9 0 113.1z\"],equals:[448,512,[62764],\"3d\",\"M48 128c-17.7 0-32 14.3-32 32s14.3 32 32 32H400c17.7 0 32-14.3 32-32s-14.3-32-32-32H48zm0 192c-17.7 0-32 14.3-32 32s14.3 32 32 32H400c17.7 0 32-14.3 32-32s-14.3-32-32-32H48z\"],blender:[512,512,[],\"f517\",\"M0 64C0 28.7 28.7 0 64 0h64 32H470.1c21.1 0 36.4 20.1 30.9 40.4L494.5 64H336c-8.8 0-16 7.2-16 16s7.2 16 16 16H485.8l-17.5 64H336c-8.8 0-16 7.2-16 16s7.2 16 16 16H459.6l-17.5 64H336c-8.8 0-16 7.2-16 16s7.2 16 16 16h97.5L416 352H160l-8.7-96H64c-35.3 0-64-28.7-64-64V64zM145.5 192L133.8 64H64V192h81.5zM144 384H432c26.5 0 48 21.5 48 48v32c0 26.5-21.5 48-48 48H144c-26.5 0-48-21.5-48-48V432c0-26.5 21.5-48 48-48zm144 96a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],teeth:[576,512,[],\"f62e\",\"M0 128C0 75 43 32 96 32H480c53 0 96 43 96 96V384c0 53-43 96-96 96H96c-53 0-96-43-96-96V128zm176 48v56c0 13.3 10.7 24 24 24h48c13.3 0 24-10.7 24-24V176c0-26.5-21.5-48-48-48s-48 21.5-48 48zm176-48c-26.5 0-48 21.5-48 48v56c0 13.3 10.7 24 24 24h48c13.3 0 24-10.7 24-24V176c0-26.5-21.5-48-48-48zM48 208v24c0 13.3 10.7 24 24 24h48c13.3 0 24-10.7 24-24V208c0-26.5-21.5-48-48-48s-48 21.5-48 48zM96 384c26.5 0 48-21.5 48-48V312c0-13.3-10.7-24-24-24H72c-13.3 0-24 10.7-24 24v24c0 26.5 21.5 48 48 48zm80-48c0 26.5 21.5 48 48 48s48-21.5 48-48V312c0-13.3-10.7-24-24-24H200c-13.3 0-24 10.7-24 24v24zm176 48c26.5 0 48-21.5 48-48V312c0-13.3-10.7-24-24-24H328c-13.3 0-24 10.7-24 24v24c0 26.5 21.5 48 48 48zm80-176v24c0 13.3 10.7 24 24 24h48c13.3 0 24-10.7 24-24V208c0-26.5-21.5-48-48-48s-48 21.5-48 48zm48 176c26.5 0 48-21.5 48-48V312c0-13.3-10.7-24-24-24H456c-13.3 0-24 10.7-24 24v24c0 26.5 21.5 48 48 48z\"],\"shekel-sign\":[448,512,[8362,\"ils\",\"shekel\",\"sheqel\",\"sheqel-sign\"],\"f20b\",\"M32 32C14.3 32 0 46.3 0 64V448c0 17.7 14.3 32 32 32s32-14.3 32-32V96H192c35.3 0 64 28.7 64 64V320c0 17.7 14.3 32 32 32s32-14.3 32-32V160c0-70.7-57.3-128-128-128H32zM320 480c70.7 0 128-57.3 128-128V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V352c0 35.3-28.7 64-64 64H192V192c0-17.7-14.3-32-32-32s-32 14.3-32 32V448c0 17.7 14.3 32 32 32H320z\"],map:[576,512,[128506,62072],\"f279\",\"M384 476.1L192 421.2V35.9L384 90.8V476.1zm32-1.2V88.4L543.1 37.5c15.8-6.3 32.9 5.3 32.9 22.3V394.6c0 9.8-6 18.6-15.1 22.3L416 474.8zM15.1 95.1L160 37.2V423.6L32.9 474.5C17.1 480.8 0 469.2 0 452.2V117.4c0-9.8 6-18.6 15.1-22.3z\"],rocket:[512,512,[],\"f135\",\"M156.6 384.9L125.7 354c-8.5-8.5-11.5-20.8-7.7-32.2c3-8.9 7-20.5 11.8-33.8L24 288c-8.6 0-16.6-4.6-20.9-12.1s-4.2-16.7 .2-24.1l52.5-88.5c13-21.9 36.5-35.3 61.9-35.3l82.3 0c2.4-4 4.8-7.7 7.2-11.3C289.1-4.1 411.1-8.1 483.9 5.3c11.6 2.1 20.6 11.2 22.8 22.8c13.4 72.9 9.3 194.8-111.4 276.7c-3.5 2.4-7.3 4.8-11.3 7.2v82.3c0 25.4-13.4 49-35.3 61.9l-88.5 52.5c-7.4 4.4-16.6 4.5-24.1 .2s-12.1-12.2-12.1-20.9V380.8c-14.1 4.9-26.4 8.9-35.7 11.9c-11.2 3.6-23.4 .5-31.8-7.8zM384 168a40 40 0 1 0 0-80 40 40 0 1 0 0 80z\"],\"photo-film\":[640,512,[\"photo-video\"],\"f87c\",\"M256 0H576c35.3 0 64 28.7 64 64V288c0 35.3-28.7 64-64 64H256c-35.3 0-64-28.7-64-64V64c0-35.3 28.7-64 64-64zM476 106.7C471.5 100 464 96 456 96s-15.5 4-20 10.7l-56 84L362.7 169c-4.6-5.7-11.5-9-18.7-9s-14.2 3.3-18.7 9l-64 80c-5.8 7.2-6.9 17.1-2.9 25.4s12.4 13.6 21.6 13.6h80 48H552c8.9 0 17-4.9 21.2-12.7s3.7-17.3-1.2-24.6l-96-144zM336 96a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM64 128h96V384v32c0 17.7 14.3 32 32 32H320c17.7 0 32-14.3 32-32V384H512v64c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V192c0-35.3 28.7-64 64-64zm8 64c-8.8 0-16 7.2-16 16v16c0 8.8 7.2 16 16 16H88c8.8 0 16-7.2 16-16V208c0-8.8-7.2-16-16-16H72zm0 104c-8.8 0-16 7.2-16 16v16c0 8.8 7.2 16 16 16H88c8.8 0 16-7.2 16-16V312c0-8.8-7.2-16-16-16H72zm0 104c-8.8 0-16 7.2-16 16v16c0 8.8 7.2 16 16 16H88c8.8 0 16-7.2 16-16V416c0-8.8-7.2-16-16-16H72zm336 16v16c0 8.8 7.2 16 16 16h16c8.8 0 16-7.2 16-16V416c0-8.8-7.2-16-16-16H424c-8.8 0-16 7.2-16 16z\"],\"folder-minus\":[512,512,[],\"f65d\",\"M448 480H64c-35.3 0-64-28.7-64-64V96C0 60.7 28.7 32 64 32H192c20.1 0 39.1 9.5 51.2 25.6l19.2 25.6c6 8.1 15.5 12.8 25.6 12.8H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64zM184 272c-13.3 0-24 10.7-24 24s10.7 24 24 24H328c13.3 0 24-10.7 24-24s-10.7-24-24-24H184z\"],store:[576,512,[],\"f54e\",\"M547.6 103.8L490.3 13.1C485.2 5 476.1 0 466.4 0H109.6C99.9 0 90.8 5 85.7 13.1L28.3 103.8c-29.6 46.8-3.4 111.9 51.9 119.4c4 .5 8.1 .8 12.1 .8c26.1 0 49.3-11.4 65.2-29c15.9 17.6 39.1 29 65.2 29c26.1 0 49.3-11.4 65.2-29c15.9 17.6 39.1 29 65.2 29c26.2 0 49.3-11.4 65.2-29c16 17.6 39.1 29 65.2 29c4.1 0 8.1-.3 12.1-.8c55.5-7.4 81.8-72.5 52.1-119.4zM499.7 254.9l-.1 0c-5.3 .7-10.7 1.1-16.2 1.1c-12.4 0-24.3-1.9-35.4-5.3V384H128V250.6c-11.2 3.5-23.2 5.4-35.6 5.4c-5.5 0-11-.4-16.3-1.1l-.1 0c-4.1-.6-8.1-1.3-12-2.3V384v64c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V384 252.6c-4 1-8 1.8-12.3 2.3z\"],\"arrow-trend-up\":[576,512,[],\"e098\",\"M384 160c-17.7 0-32-14.3-32-32s14.3-32 32-32H544c17.7 0 32 14.3 32 32V288c0 17.7-14.3 32-32 32s-32-14.3-32-32V205.3L342.6 374.6c-12.5 12.5-32.8 12.5-45.3 0L192 269.3 54.6 406.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l160-160c12.5-12.5 32.8-12.5 45.3 0L320 306.7 466.7 160H384z\"],\"plug-circle-minus\":[576,512,[],\"e55e\",\"M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8c12.3-2.5 24.1-6.4 35.1-11.5c-2.1-10.8-3.1-21.9-3.1-33.3c0-80.3 53.8-148 127.3-169.2c.5-2.2 .7-4.5 .7-6.8c0-17.7-14.3-32-32-32H32zM576 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-64 0c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16s7.2-16 16-16H496c8.8 0 16 7.2 16 16z\"],\"sign-hanging\":[512,512,[\"sign\"],\"f4d9\",\"M96 0c17.7 0 32 14.3 32 32V64l352 0c17.7 0 32 14.3 32 32s-14.3 32-32 32l-352 0V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V128H32C14.3 128 0 113.7 0 96S14.3 64 32 64H64V32C64 14.3 78.3 0 96 0zm96 160H448c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32V192c0-17.7 14.3-32 32-32z\"],\"bezier-curve\":[640,512,[],\"f55b\",\"M296 136V88h48v48H296zM288 32c-26.5 0-48 21.5-48 48v4H121.6C111.2 62.7 89.3 48 64 48C28.7 48 0 76.7 0 112s28.7 64 64 64c25.3 0 47.2-14.7 57.6-36h66.9c-58.9 39.6-98.9 105-104 180H80c-26.5 0-48 21.5-48 48v64c0 26.5 21.5 48 48 48h64c26.5 0 48-21.5 48-48V368c0-26.5-21.5-48-48-48h-3.3c5.9-67 48.5-123.4 107.5-149.1c8.6 12.7 23.2 21.1 39.8 21.1h64c16.6 0 31.1-8.4 39.8-21.1c59 25.7 101.6 82.1 107.5 149.1H496c-26.5 0-48 21.5-48 48v64c0 26.5 21.5 48 48 48h64c26.5 0 48-21.5 48-48V368c0-26.5-21.5-48-48-48h-4.5c-5-75-45.1-140.4-104-180h66.9c10.4 21.3 32.3 36 57.6 36c35.3 0 64-28.7 64-64s-28.7-64-64-64c-25.3 0-47.2 14.7-57.6 36H400V80c0-26.5-21.5-48-48-48H288zM88 376h48v48H88V376zm416 48V376h48v48H504z\"],\"bell-slash\":[640,512,[128277,61943],\"f1f6\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-90.2-70.7c.2-.4 .4-.9 .6-1.3c5.2-11.5 3.1-25-5.3-34.4l-7.4-8.3C497.3 319.2 480 273.9 480 226.8V208c0-77.4-55-142-128-156.8V32c0-17.7-14.3-32-32-32s-32 14.3-32 32V51.2c-42.6 8.6-79 34.2-102 69.3L38.8 5.1zM406.2 416L160 222.1v4.8c0 47-17.3 92.4-48.5 127.6l-7.4 8.3c-8.4 9.4-10.4 22.9-5.3 34.4S115.4 416 128 416H406.2zm-40.9 77.3c12-12 18.7-28.3 18.7-45.3H320 256c0 17 6.7 33.3 18.7 45.3s28.3 18.7 45.3 18.7s33.3-6.7 45.3-18.7z\"],tablet:[448,512,[\"tablet-android\"],\"f3fb\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zM176 432h96c8.8 0 16 7.2 16 16s-7.2 16-16 16H176c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],\"school-flag\":[576,512,[],\"e56e\",\"M288 0H400c8.8 0 16 7.2 16 16V80c0 8.8-7.2 16-16 16H320.7l89.6 64H512c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H336V400c0-26.5-21.5-48-48-48s-48 21.5-48 48V512H64c-35.3 0-64-28.7-64-64V224c0-35.3 28.7-64 64-64H165.7L256 95.5V32c0-17.7 14.3-32 32-32zm48 240a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM80 224c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H80zm368 16v64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V240c0-8.8-7.2-16-16-16H464c-8.8 0-16 7.2-16 16zM80 352c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H80zm384 0c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V368c0-8.8-7.2-16-16-16H464z\"],fill:[512,512,[],\"f575\",\"M86.6 9.4C74.1-3.1 53.9-3.1 41.4 9.4s-12.5 32.8 0 45.3L122.7 136 30.6 228.1c-37.5 37.5-37.5 98.3 0 135.8L148.1 481.4c37.5 37.5 98.3 37.5 135.8 0L474.3 290.9c28.1-28.1 28.1-73.7 0-101.8L322.9 37.7c-28.1-28.1-73.7-28.1-101.8 0L168 90.7 86.6 9.4zM168 181.3l49.4 49.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L213.3 136l53.1-53.1c3.1-3.1 8.2-3.1 11.3 0L429.1 234.3c3.1 3.1 3.1 8.2 0 11.3L386.7 288H67.5c1.4-5.4 4.2-10.4 8.4-14.6L168 181.3z\"],\"angle-up\":[448,512,[8963],\"f106\",\"M201.4 137.4c12.5-12.5 32.8-12.5 45.3 0l160 160c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L224 205.3 86.6 342.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l160-160z\"],\"drumstick-bite\":[512,512,[],\"f6d7\",\"M160 265.2c0 8.5-3.4 16.6-9.4 22.6l-26.8 26.8c-12.3 12.3-32.5 11.4-49.4 7.2C69.8 320.6 65 320 60 320c-33.1 0-60 26.9-60 60s26.9 60 60 60c6.3 0 12 5.7 12 12c0 33.1 26.9 60 60 60s60-26.9 60-60c0-5-.6-9.8-1.8-14.5c-4.2-16.9-5.2-37.1 7.2-49.4l26.8-26.8c6-6 14.1-9.4 22.6-9.4H336c6.3 0 12.4-.3 18.5-1c11.9-1.2 16.4-15.5 10.8-26c-8.5-15.8-13.3-33.8-13.3-53c0-61.9 50.1-112 112-112c8 0 15.7 .8 23.2 2.4c11.7 2.5 24.1-5.9 22-17.6C494.5 62.5 422.5 0 336 0C238.8 0 160 78.8 160 176v89.2z\"],\"holly-berry\":[512,512,[],\"f7aa\",\"M256 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-80 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM276.8 383.8c1 .1 2.1 .2 3.2 .2c39.8 0 72 32.2 72 72v22.7c0 16.4 16 27.9 31.6 22.8l12.8-4.3c18-6 37.3-6.5 55.6-1.5l19.4 5.3c17.9 4.9 34.4-11.6 29.5-29.5L495.6 452c-5-18.3-4.4-37.6 1.5-55.6l4.3-12.8c5.2-15.5-6.4-31.6-22.8-31.6c-34.6 0-62.7-28.1-62.7-62.7v-32c0-16.4-16-27.9-31.6-22.8l-12.8 4.3c-18 6-37.3 6.5-55.6 1.5l-29.6-8.1c-2.9-.8-5.9-1-8.7-.7c4.2 9.7 5.8 20.8 3.7 32.3L275 298.7c-1.5 8.4-1.4 17 .5 25.3l5.3 23.9c2.8 12.7 1.1 25.2-4 35.9zM127.6 234.5c-15.5-5.2-31.6 6.4-31.6 22.8v32C96 323.9 67.9 352 33.3 352c-16.4 0-27.9 16-22.8 31.6l4.3 12.8c6 18 6.5 37.3 1.5 55.6l-5.3 19.4C6.2 489.4 22.6 505.8 40.5 501L60 495.6c18.3-5 37.6-4.5 55.6 1.5l12.8 4.3c15.5 5.2 31.6-6.4 31.6-22.8v-32c0-34.6 28.1-62.7 62.7-62.7c16.4 0 27.9-16 22.8-31.6l-4.3-12.8c-6-18-6.5-37.3-1.5-55.6l5.3-19.4c4.9-17.9-11.6-34.4-29.5-29.5L196 240.4c-18.3 5-37.6 4.4-55.6-1.5l-12.8-4.3zM384 144a48 48 0 1 0 -96 0 48 48 0 1 0 96 0z\"],\"chevron-left\":[320,512,[9001],\"f053\",\"M9.4 233.4c-12.5 12.5-12.5 32.8 0 45.3l192 192c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L77.3 256 246.6 86.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-192 192z\"],bacteria:[640,512,[],\"e059\",\"M304.9 .7c-9.6-2.7-19.5 2.8-22.3 12.4l-4.3 15.2c-8.3-.6-16.8 0-25.2 1.9c-7.3 1.7-14.3 3.5-21.1 5.5l-5.5-12.7c-3.9-9.1-14.5-13.4-23.6-9.5s-13.4 14.5-9.5 23.6l4.4 10.4c-16.6 6.7-31.7 14.4-45.4 22.8L147 62c-5.5-8.3-16.7-10.5-25-5s-10.5 16.7-5 25l6 9c-13.7 11-25.5 22.8-35.8 34.9l-10-8c-7.8-6.2-19.1-5-25.3 2.8s-5 19.1 2.8 25.3L65.9 155c-1.8 2.8-3.5 5.7-5.1 8.5c-6.6 11.4-11.8 22.6-16 33l-8-3.2c-9.2-3.7-19.7 .8-23.4 10s.8 19.7 10 23.4l10.4 4.2c-.2 .8-.4 1.5-.5 2.3c-2.2 9.3-3.4 17.3-4.1 23.4c-.4 3.1-.6 5.7-.8 7.8c-.1 1.1-.1 2-.2 2.8l-.1 1.1 0 .5 0 .2 0 .1c0 0 0 .1 29.1 1l-.1 0L28 269.3c-.1 3.1 0 6.1 .2 9.1l-15.2 4.3C3.5 285.4-2 295.4 .7 304.9s12.7 15.1 22.3 12.4l15.6-4.5c7.6 13.6 18.9 25 32.6 32.6L66.7 361c-2.7 9.6 2.8 19.5 12.4 22.3s19.5-2.8 22.3-12.4l4.3-15.2c1.2 .1 2.4 .2 3.6 .2c15.6 .5 30.3-3.3 43-10.2l9 9c7 7 18.4 7 25.5 0s7-18.4 0-25.5l-7.2-7.2c9.3-12.6 15.2-27.8 16.3-44.5l7.1 3c9.1 3.9 19.7-.3 23.6-9.5s-.3-19.7-9.5-23.6l-8.6-3.7c6.4-9.9 17.3-22.4 36.9-33.3l1.3 4.4c2.7 9.6 12.7 15.1 22.3 12.4s15.1-12.7 12.4-22.3l-2.3-8.1c3.8-1.1 7.7-2.1 11.9-3.1c11.6-2.7 22.1-7.7 31.1-14.4l7.2 7.2c7 7 18.4 7 25.5 0s7-18.4 0-25.5l-9-9c7.6-13.9 11.3-30.1 10.1-46.6l15.2-4.3c9.6-2.7 15.1-12.7 12.4-22.3S370.6 64 361 66.7l-15.6 4.5c-7.7-13.9-19.1-25.1-32.6-32.6l4.5-15.6c2.7-9.6-2.8-19.5-12.4-22.3zM112 272l-48-1.5 0 0c11.7 .4 27.3 .9 48 1.6zm16-80a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm64-48a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zM322.7 489c-2.7 9.6 2.8 19.5 12.4 22.3s19.5-2.8 22.2-12.4l4.3-15.2c8.3 .6 16.8 0 25.2-1.9c7.3-1.7 14.3-3.5 21.1-5.5l5.5 12.7c3.9 9.1 14.5 13.4 23.6 9.5s13.4-14.5 9.5-23.6l-4.4-10.4c16.6-6.7 31.7-14.4 45.4-22.8L493 450c5.5 8.3 16.7 10.5 25 5s10.5-16.7 5-25l-6-9c13.7-11 25.5-22.8 35.8-34.9l10 8c7.8 6.2 19.1 5 25.3-2.8s5-19.1-2.8-25.3L574.1 357c1.8-2.8 3.5-5.7 5.1-8.5c6.6-11.4 11.8-22.6 16-33l8 3.2c9.2 3.7 19.7-.8 23.4-10s-.8-19.7-10-23.4l-10.4-4.2c.2-.8 .4-1.5 .5-2.3c2.2-9.3 3.4-17.3 4.1-23.4c.4-3.1 .6-5.7 .8-7.8c.1-1.1 .1-2 .2-2.8l.1-1.1 0-.5 0-.2 0-.1c0 0 0-.1-29.1-1l.1 0 29.1 .9c.1-3.1 0-6.1-.2-9.1l15.2-4.3c9.6-2.7 15.1-12.7 12.4-22.3s-12.7-15.1-22.3-12.4l-15.6 4.5c-7.6-13.6-18.9-25-32.6-32.6l4.5-15.6c2.7-9.6-2.8-19.5-12.4-22.3s-19.5 2.8-22.3 12.4l-4.3 15.2c-1.2-.1-2.4-.2-3.6-.2c-15.6-.5-30.3 3.3-43 10.2l-9-9c-7-7-18.4-7-25.5 0s-7 18.4 0 25.5l7.2 7.2c-9.3 12.6-15.2 27.8-16.3 44.5l-7.1-3c-9.1-3.9-19.7 .3-23.6 9.5s.3 19.7 9.5 23.6l8.6 3.7c-6.4 9.9-17.3 22.4-36.9 33.3l-1.3-4.4c-2.7-9.6-12.7-15.1-22.3-12.4s-15.1 12.7-12.4 22.3l2.3 8.1c-3.8 1.1-7.7 2.1-11.9 3.1c-11.6 2.7-22.1 7.7-31.1 14.4l-7.2-7.2c-7-7-18.4-7-25.5 0s-7 18.4 0 25.5l9 9c-7.6 13.9-11.3 30.1-10.1 46.6l-15.2 4.3c-9.6 2.7-15.1 12.7-12.4 22.2s12.7 15.1 22.3 12.4l15.6-4.5c7.7 13.9 19.1 25.1 32.6 32.6L322.7 489zM576 241.5l0 0c-11.7-.4-27.3-.9-48-1.6l48 1.5zM448 384a32 32 0 1 1 -64 0 32 32 0 1 1 64 0z\"],\"hand-lizard\":[512,512,[],\"f258\",\"M0 112C0 85.5 21.5 64 48 64H160h80 46.5c36.8 0 71.2 18 92.1 48.2l113.5 164c13 18.7 19.9 41 19.9 63.8v12 16 48c0 17.7-14.3 32-32 32H384c-17.7 0-32-14.3-32-32V402.2L273.9 352H240 160 112c-26.5 0-48-21.5-48-48s21.5-48 48-48h48 80c26.5 0 48-21.5 48-48s-21.5-48-48-48H160 48c-26.5 0-48-21.5-48-48z\"],notdef:[384,512,[],\"e1fe\",\"M64 390.3L153.5 256 64 121.7V390.3zM102.5 448H281.5L192 313.7 102.5 448zm128-192L320 390.3V121.7L230.5 256zM281.5 64H102.5L192 198.3 281.5 64zM0 48C0 21.5 21.5 0 48 0H336c26.5 0 48 21.5 48 48V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V48z\"],disease:[512,512,[],\"f7fa\",\"M236.4 61.4L227 75.5c-21.3 32-59.4 48.5-97.3 42.1l-59.6-9.9C33.4 101.6 0 129.9 .1 167.1c0 15.9 6.4 31.2 17.6 42.5l29.2 29.2c11 11 17.2 25.9 17.2 41.5c0 15.8-6.4 30.9-17.7 42L33.3 335.1C22.2 345.9 16 360.7 16 376.2c0 36.8 34.1 64.2 70.1 56.2l62.3-13.8c7.7-1.7 15.7-2.6 23.6-2.6h10c27.2 0 53.7 9.3 75 26.3L287.8 467c10.5 8.4 23.6 13 37 13c32.7 0 59.3-26.5 59.3-59.3l0-25.2c0-34.9 21.4-66.2 53.9-78.8l36.9-14.3c22.4-8.7 37.2-30.3 37.2-54.3c0-28.1-20.1-52.3-47.8-57.3l-28-5.1c-36.5-6.7-65.4-34.5-73.6-70.7l-7.1-31.5C348.9 53.4 322.1 32 291.3 32c-22 0-42.6 11-54.9 29.4zM160 192a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm128 16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zm0 80a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"briefcase-medical\":[512,512,[],\"f469\",\"M184 48H328c4.4 0 8 3.6 8 8V96H176V56c0-4.4 3.6-8 8-8zm-56 8V96H64C28.7 96 0 124.7 0 160V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H384V56c0-30.9-25.1-56-56-56H184c-30.9 0-56 25.1-56 56zm96 152c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H288v48c0 8.8-7.2 16-16 16H240c-8.8 0-16-7.2-16-16V320H176c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h48V208z\"],genderless:[384,512,[],\"f22d\",\"M192 144a112 112 0 1 1 0 224 112 112 0 1 1 0-224zm0 288a176 176 0 1 0 0-352 176 176 0 1 0 0 352z\"],\"chevron-right\":[320,512,[9002],\"f054\",\"M310.6 233.4c12.5 12.5 12.5 32.8 0 45.3l-192 192c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3L242.7 256 73.4 86.6c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0l192 192z\"],retweet:[576,512,[],\"f079\",\"M272 416c17.7 0 32-14.3 32-32s-14.3-32-32-32H160c-17.7 0-32-14.3-32-32V192h32c12.9 0 24.6-7.8 29.6-19.8s2.2-25.7-6.9-34.9l-64-64c-12.5-12.5-32.8-12.5-45.3 0l-64 64c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8l32 0 0 128c0 53 43 96 96 96H272zM304 96c-17.7 0-32 14.3-32 32s14.3 32 32 32l112 0c17.7 0 32 14.3 32 32l0 128H416c-12.9 0-24.6 7.8-29.6 19.8s-2.2 25.7 6.9 34.9l64 64c12.5 12.5 32.8 12.5 45.3 0l64-64c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8l-32 0V192c0-53-43-96-96-96L304 96z\"],\"car-rear\":[512,512,[\"car-alt\"],\"f5de\",\"M165.4 96H346.6c13.6 0 25.7 8.6 30.2 21.4L402.9 192H109.1l26.1-74.6c4.5-12.8 16.6-21.4 30.2-21.4zm-90.6 .3L39.6 196.8C16.4 206.4 0 229.3 0 256v80c0 23.7 12.9 44.4 32 55.4V448c0 17.7 14.3 32 32 32H96c17.7 0 32-14.3 32-32V400H384v48c0 17.7 14.3 32 32 32h32c17.7 0 32-14.3 32-32V391.4c19.1-11.1 32-31.7 32-55.4V256c0-26.7-16.4-49.6-39.6-59.2L437.2 96.3C423.7 57.8 387.4 32 346.6 32H165.4c-40.8 0-77.1 25.8-90.6 64.3zM208 272h96c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H208c-8.8 0-16-7.2-16-16V288c0-8.8 7.2-16 16-16zM48 280c0-13.3 10.7-24 24-24h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H72c-13.3 0-24-10.7-24-24zm360-24h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H408c-13.3 0-24-10.7-24-24s10.7-24 24-24z\"],\"pump-soap\":[448,512,[],\"e06b\",\"M128 32v96H256V96h60.1c4.2 0 8.3 1.7 11.3 4.7l33.9 33.9c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L372.7 55.4c-15-15-35.4-23.4-56.6-23.4H256c0-17.7-14.3-32-32-32H160c-17.7 0-32 14.3-32 32zM117.4 160c-33.3 0-61 25.5-63.8 58.7L35 442.7C31.9 480 61.3 512 98.8 512H285.2c37.4 0 66.9-32 63.8-69.3l-18.7-224c-2.8-33.2-30.5-58.7-63.8-58.7H117.4zM256 360c0 35.3-28.7 56-64 56s-64-20.7-64-56c0-32.5 37-80.9 50.9-97.9c3.2-3.9 8.1-6.1 13.1-6.1s9.9 2.2 13.1 6.1C219 279.1 256 327.5 256 360z\"],\"video-slash\":[640,512,[],\"f4e2\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7l-86.4-67.7 13.8 9.2c9.8 6.5 22.4 7.2 32.9 1.6s16.9-16.4 16.9-28.2V128c0-11.8-6.5-22.6-16.9-28.2s-23-5-32.9 1.6l-96 64L448 174.9V192 320v5.8l-32-25.1V128c0-35.3-28.7-64-64-64H113.9L38.8 5.1zM407 416.7L32.3 121.5c-.2 2.1-.3 4.3-.3 6.5V384c0 35.3 28.7 64 64 64H352c23.4 0 43.9-12.6 55-31.3z\"],\"battery-quarter\":[576,512,[\"battery-2\"],\"f243\",\"M464 160c8.8 0 16 7.2 16 16V336c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16H464zM80 96C35.8 96 0 131.8 0 176V336c0 44.2 35.8 80 80 80H464c44.2 0 80-35.8 80-80V320c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32V176c0-44.2-35.8-80-80-80H80zm112 96H96V320h96V192z\"],radio:[512,512,[128251],\"f8d7\",\"M494.8 47c12.7-3.7 20-17.1 16.3-29.8S494-2.8 481.2 1L51.7 126.9c-9.4 2.7-17.9 7.3-25.1 13.2C10.5 151.7 0 170.6 0 192v4V304 448c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V192c0-35.3-28.7-64-64-64H218.5L494.8 47zM368 240a80 80 0 1 1 0 160 80 80 0 1 1 0-160zM80 256c0-8.8 7.2-16 16-16h96c8.8 0 16 7.2 16 16s-7.2 16-16 16H96c-8.8 0-16-7.2-16-16zM64 320c0-8.8 7.2-16 16-16H208c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16zm16 64c0-8.8 7.2-16 16-16h96c8.8 0 16 7.2 16 16s-7.2 16-16 16H96c-8.8 0-16-7.2-16-16z\"],\"baby-carriage\":[512,512,[\"carriage-baby\"],\"f77d\",\"M256 192H.1C2.7 117.9 41.3 52.9 99 14.1c13.3-8.9 30.8-4.3 39.9 8.8L256 192zm128-32c0-35.3 28.7-64 64-64h32c17.7 0 32 14.3 32 32s-14.3 32-32 32l-32 0v64c0 25.2-5.8 50.2-17 73.5s-27.8 44.5-48.6 62.3s-45.5 32-72.7 41.6S253.4 416 224 416s-58.5-5-85.7-14.6s-51.9-23.8-72.7-41.6s-37.3-39-48.6-62.3S0 249.2 0 224l224 0 160 0V160zM80 416a48 48 0 1 1 0 96 48 48 0 1 1 0-96zm240 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0z\"],\"traffic-light\":[320,512,[128678],\"f637\",\"M64 0C28.7 0 0 28.7 0 64V352c0 88.4 71.6 160 160 160s160-71.6 160-160V64c0-35.3-28.7-64-64-64H64zm96 416a48 48 0 1 1 0-96 48 48 0 1 1 0 96zm48-176a48 48 0 1 1 -96 0 48 48 0 1 1 96 0zm-48-80a48 48 0 1 1 0-96 48 48 0 1 1 0 96z\"],thermometer:[512,512,[],\"f491\",\"M96 382.1V293.3c0-14.9 5.9-29.1 16.4-39.6l27.3-27.3 57 57c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-57-57 41.4-41.4 57 57c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-57-57 41.4-41.4 57 57c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-57-57 45.5-45.5C355.2 10.9 381.4 0 408.8 0C465.8 0 512 46.2 512 103.2c0 27.4-10.9 53.6-30.2 73L258.3 399.6c-10.5 10.5-24.7 16.4-39.6 16.4H129.9L41 505c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l89-89z\"],\"vr-cardboard\":[640,512,[],\"f729\",\"M576 64H64C28.7 64 0 92.7 0 128V384c0 35.3 28.7 64 64 64H184.4c24.2 0 46.4-13.7 57.2-35.4l32-64c8.8-17.5 26.7-28.6 46.3-28.6s37.5 11.1 46.3 28.6l32 64c10.8 21.7 33 35.4 57.2 35.4H576c35.3 0 64-28.7 64-64V128c0-35.3-28.7-64-64-64zM96 240a64 64 0 1 1 128 0A64 64 0 1 1 96 240zm384-64a64 64 0 1 1 0 128 64 64 0 1 1 0-128z\"],\"hand-middle-finger\":[448,512,[128405],\"f806\",\"M232 0c-22.1 0-40 17.9-40 40V204.2c-8.5-7.6-19.7-12.2-32-12.2c-26.5 0-48 21.5-48 48v7 73c0 8.8-7.2 16-16 16s-16-7.2-16-16V264.3c-2 1.4-3.9 3-5.8 4.5L55 284.8C40.4 297 32 315 32 334V372c0 38 16.9 74 46.1 98.3l5.4 4.5c28.8 24 65 37.1 102.4 37.1H304c70.7 0 128-57.3 128-128V320 288c0-26.5-21.5-48-48-48c-12.4 0-23.6 4.7-32.1 12.3C350 227.5 329.3 208 304 208c-12.3 0-23.5 4.6-32 12.2V40c0-22.1-17.9-40-40-40z\"],percent:[384,512,[62101,62785,\"percentage\"],\"25\",\"M374.6 118.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-320 320c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l320-320zM128 128A64 64 0 1 0 0 128a64 64 0 1 0 128 0zM384 384a64 64 0 1 0 -128 0 64 64 0 1 0 128 0z\"],\"truck-moving\":[640,512,[],\"f4df\",\"M64 32C28.7 32 0 60.7 0 96V304v80 16c0 44.2 35.8 80 80 80c26.2 0 49.4-12.6 64-32c14.6 19.4 37.8 32 64 32c44.2 0 80-35.8 80-80c0-5.5-.6-10.8-1.6-16H416h33.6c-1 5.2-1.6 10.5-1.6 16c0 44.2 35.8 80 80 80s80-35.8 80-80c0-5.5-.6-10.8-1.6-16H608c17.7 0 32-14.3 32-32V288 272 261.7c0-9.2-3.2-18.2-9-25.3l-58.8-71.8c-10.6-13-26.5-20.5-43.3-20.5H480V96c0-35.3-28.7-64-64-64H64zM585 256H480V192h48.8c2.4 0 4.7 1.1 6.2 2.9L585 256zM528 368a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM176 400a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM80 368a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"glass-water-droplet\":[384,512,[],\"e4f5\",\"M32 0C23.1 0 14.6 3.7 8.6 10.2S-.6 25.4 .1 34.3L28.9 437.7c3 41.9 37.8 74.3 79.8 74.3H275.3c42 0 76.8-32.4 79.8-74.3L383.9 34.3c.6-8.9-2.4-17.6-8.5-24.1S360.9 0 352 0H32zM83 297.5L66.4 64H317.6L301 297.5 288 304c-20.1 10.1-43.9 10.1-64 0s-43.9-10.1-64 0s-43.9 10.1-64 0l-13-6.5zM256 196c0-24-33.7-70.1-52.2-93.5c-6.1-7.7-17.5-7.7-23.6 0C161.7 125.9 128 172 128 196c0 33.1 28.7 60 64 60s64-26.9 64-60z\"],display:[576,512,[],\"e163\",\"M64 0C28.7 0 0 28.7 0 64V352c0 35.3 28.7 64 64 64H240l-10.7 32H160c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H346.7L336 416H512c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zM512 64V352H64V64H512z\"],\"face-smile\":[512,512,[128578,\"smile\"],\"f118\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM164.1 325.5C182 346.2 212.6 368 256 368s74-21.8 91.9-42.5c5.8-6.7 15.9-7.4 22.6-1.6s7.4 15.9 1.6 22.6C349.8 372.1 311.1 400 256 400s-93.8-27.9-116.1-53.5c-5.8-6.7-5.1-16.8 1.6-22.6s16.8-5.1 22.6 1.6zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],thumbtack:[384,512,[128204,128392,\"thumb-tack\"],\"f08d\",\"M32 32C32 14.3 46.3 0 64 0H320c17.7 0 32 14.3 32 32s-14.3 32-32 32H290.5l11.4 148.2c36.7 19.9 65.7 53.2 79.5 94.7l1 3c3.3 9.8 1.6 20.5-4.4 28.8s-15.7 13.3-26 13.3H32c-10.3 0-19.9-4.9-26-13.3s-7.7-19.1-4.4-28.8l1-3c13.8-41.5 42.8-74.8 79.5-94.7L93.5 64H64C46.3 64 32 49.7 32 32zM160 384h64v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V384z\"],trophy:[576,512,[127942],\"f091\",\"M400 0H176c-26.5 0-48.1 21.8-47.1 48.2c.2 5.3 .4 10.6 .7 15.8H24C10.7 64 0 74.7 0 88c0 92.6 33.5 157 78.5 200.7c44.3 43.1 98.3 64.8 138.1 75.8c23.4 6.5 39.4 26 39.4 45.6c0 20.9-17 37.9-37.9 37.9H192c-17.7 0-32 14.3-32 32s14.3 32 32 32H384c17.7 0 32-14.3 32-32s-14.3-32-32-32H357.9C337 448 320 431 320 410.1c0-19.6 15.9-39.2 39.4-45.6c39.9-11 93.9-32.7 138.2-75.8C542.5 245 576 180.6 576 88c0-13.3-10.7-24-24-24H446.4c.3-5.2 .5-10.4 .7-15.8C448.1 21.8 426.5 0 400 0zM48.9 112h84.4c9.1 90.1 29.2 150.3 51.9 190.6c-24.9-11-50.8-26.5-73.2-48.3c-32-31.1-58-76-63-142.3zM464.1 254.3c-22.4 21.8-48.3 37.3-73.2 48.3c22.7-40.3 42.8-100.5 51.9-190.6h84.4c-5.1 66.3-31.1 111.2-63 142.3z\"],\"person-praying\":[448,512,[128720,\"pray\"],\"f683\",\"M352 64A64 64 0 1 0 224 64a64 64 0 1 0 128 0zM232.7 264l22.9 31.5c6.5 8.9 16.3 14.7 27.2 16.1s21.9-1.7 30.4-8.7l88-72c17.1-14 19.6-39.2 5.6-56.3s-39.2-19.6-56.3-5.6l-55.2 45.2-26.2-36C253.6 156.7 228.6 144 202 144c-30.9 0-59.2 17.1-73.6 44.4L79.8 280.9c-20.2 38.5-9.4 85.9 25.6 111.8L158.6 432H72c-22.1 0-40 17.9-40 40s17.9 40 40 40H280c17.3 0 32.6-11.1 38-27.5s-.3-34.4-14.2-44.7L187.7 354l45-90z\"],hammer:[576,512,[128296],\"f6e3\",\"M413.5 237.5c-28.2 4.8-58.2-3.6-80-25.4l-38.1-38.1C280.4 159 272 138.8 272 117.6V105.5L192.3 62c-5.3-2.9-8.6-8.6-8.3-14.7s3.9-11.5 9.5-14l47.2-21C259.1 4.2 279 0 299.2 0h18.1c36.7 0 72 14 98.7 39.1l44.6 42c24.2 22.8 33.2 55.7 26.6 86L503 183l8-8c9.4-9.4 24.6-9.4 33.9 0l24 24c9.4 9.4 9.4 24.6 0 33.9l-88 88c-9.4 9.4-24.6 9.4-33.9 0l-24-24c-9.4-9.4-9.4-24.6 0-33.9l8-8-17.5-17.5zM27.4 377.1L260.9 182.6c3.5 4.9 7.5 9.6 11.8 14l38.1 38.1c6 6 12.4 11.2 19.2 15.7L134.9 484.6c-14.5 17.4-36 27.4-58.6 27.4C34.1 512 0 477.8 0 435.7c0-22.6 10.1-44.1 27.4-58.6z\"],\"hand-peace\":[512,512,[9996],\"f25b\",\"M224 0c17.7 0 32 14.3 32 32V240H192V32c0-17.7 14.3-32 32-32zm96 160c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V192c0-17.7 14.3-32 32-32zm64 64c0-17.7 14.3-32 32-32s32 14.3 32 32v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V224zM93.3 51.2L175.9 240H106.1L34.7 76.8C27.6 60.6 35 41.8 51.2 34.7s35.1 .3 42.1 16.5zm27 221.3l-.2-.5h69.9H216c22.1 0 40 17.9 40 40s-17.9 40-40 40H160c-8.8 0-16 7.2-16 16s7.2 16 16 16h56c39.8 0 72-32.2 72-72l0-.6c9.4 5.4 20.3 8.6 32 8.6c13.2 0 25.4-4 35.6-10.8c8.7 24.9 32.5 42.8 60.4 42.8c11.7 0 22.6-3.1 32-8.6V352c0 88.4-71.6 160-160 160H226.3c-42.4 0-83.1-16.9-113.1-46.9l-11.6-11.6C77.5 429.5 64 396.9 64 363V336c0-32.7 24.6-59.7 56.3-63.5z\"],rotate:[512,512,[128260,\"sync-alt\"],\"f2f1\",\"M142.9 142.9c62.2-62.2 162.7-62.5 225.3-1L327 183c-6.9 6.9-8.9 17.2-5.2 26.2s12.5 14.8 22.2 14.8H463.5c0 0 0 0 0 0H472c13.3 0 24-10.7 24-24V72c0-9.7-5.8-18.5-14.8-22.2s-19.3-1.7-26.2 5.2L413.4 96.6c-87.6-86.5-228.7-86.2-315.8 1C73.2 122 55.6 150.7 44.8 181.4c-5.9 16.7 2.9 34.9 19.5 40.8s34.9-2.9 40.8-19.5c7.7-21.8 20.2-42.3 37.8-59.8zM16 312v7.6 .7V440c0 9.7 5.8 18.5 14.8 22.2s19.3 1.7 26.2-5.2l41.6-41.6c87.6 86.5 228.7 86.2 315.8-1c24.4-24.4 42.1-53.1 52.9-83.7c5.9-16.7-2.9-34.9-19.5-40.8s-34.9 2.9-40.8 19.5c-7.7 21.8-20.2 42.3-37.8 59.8c-62.2 62.2-162.7 62.5-225.3 1L185 329c6.9-6.9 8.9-17.2 5.2-26.2s-12.5-14.8-22.2-14.8H48.4h-.7H40c-13.3 0-24 10.7-24 24z\"],spinner:[512,512,[],\"f110\",\"M304 48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zm0 416a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM48 304a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm464-48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM142.9 437A48 48 0 1 0 75 369.1 48 48 0 1 0 142.9 437zm0-294.2A48 48 0 1 0 75 75a48 48 0 1 0 67.9 67.9zM369.1 437A48 48 0 1 0 437 369.1 48 48 0 1 0 369.1 437z\"],robot:[640,512,[129302],\"f544\",\"M320 0c17.7 0 32 14.3 32 32V96H472c39.8 0 72 32.2 72 72V440c0 39.8-32.2 72-72 72H168c-39.8 0-72-32.2-72-72V168c0-39.8 32.2-72 72-72H288V32c0-17.7 14.3-32 32-32zM208 384c-8.8 0-16 7.2-16 16s7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H208zm96 0c-8.8 0-16 7.2-16 16s7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H304zm96 0c-8.8 0-16 7.2-16 16s7.2 16 16 16h32c8.8 0 16-7.2 16-16s-7.2-16-16-16H400zM264 256a40 40 0 1 0 -80 0 40 40 0 1 0 80 0zm152 40a40 40 0 1 0 0-80 40 40 0 1 0 0 80zM48 224H64V416H48c-26.5 0-48-21.5-48-48V272c0-26.5 21.5-48 48-48zm544 0c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H576V224h16z\"],peace:[512,512,[9774],\"f67c\",\"M224 445.3V323.5l-94.3 77.1c26.1 22.8 58.5 38.7 94.3 44.7zM89.2 351.1L224 240.8V66.7C133.2 81.9 64 160.9 64 256c0 34.6 9.2 67.1 25.2 95.1zm293.1 49.5L288 323.5V445.3c35.7-6 68.1-21.9 94.3-44.7zm40.6-49.5c16-28 25.2-60.5 25.2-95.1c0-95.1-69.2-174.1-160-189.3V240.8L422.8 351.1zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256z\"],gears:[640,512,[\"cogs\"],\"f085\",\"M308.5 135.3c7.1-6.3 9.9-16.2 6.2-25c-2.3-5.3-4.8-10.5-7.6-15.5L304 89.4c-3-5-6.3-9.9-9.8-14.6c-5.7-7.6-15.7-10.1-24.7-7.1l-28.2 9.3c-10.7-8.8-23-16-36.2-20.9L199 27.1c-1.9-9.3-9.1-16.7-18.5-17.8C173.9 8.4 167.2 8 160.4 8h-.7c-6.8 0-13.5 .4-20.1 1.2c-9.4 1.1-16.6 8.6-18.5 17.8L115 56.1c-13.3 5-25.5 12.1-36.2 20.9L50.5 67.8c-9-3-19-.5-24.7 7.1c-3.5 4.7-6.8 9.6-9.9 14.6l-3 5.3c-2.8 5-5.3 10.2-7.6 15.6c-3.7 8.7-.9 18.6 6.2 25l22.2 19.8C32.6 161.9 32 168.9 32 176s.6 14.1 1.7 20.9L11.5 216.7c-7.1 6.3-9.9 16.2-6.2 25c2.3 5.3 4.8 10.5 7.6 15.6l3 5.2c3 5.1 6.3 9.9 9.9 14.6c5.7 7.6 15.7 10.1 24.7 7.1l28.2-9.3c10.7 8.8 23 16 36.2 20.9l6.1 29.1c1.9 9.3 9.1 16.7 18.5 17.8c6.7 .8 13.5 1.2 20.4 1.2s13.7-.4 20.4-1.2c9.4-1.1 16.6-8.6 18.5-17.8l6.1-29.1c13.3-5 25.5-12.1 36.2-20.9l28.2 9.3c9 3 19 .5 24.7-7.1c3.5-4.7 6.8-9.5 9.8-14.6l3.1-5.4c2.8-5 5.3-10.2 7.6-15.5c3.7-8.7 .9-18.6-6.2-25l-22.2-19.8c1.1-6.8 1.7-13.8 1.7-20.9s-.6-14.1-1.7-20.9l22.2-19.8zM112 176a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM504.7 500.5c6.3 7.1 16.2 9.9 25 6.2c5.3-2.3 10.5-4.8 15.5-7.6l5.4-3.1c5-3 9.9-6.3 14.6-9.8c7.6-5.7 10.1-15.7 7.1-24.7l-9.3-28.2c8.8-10.7 16-23 20.9-36.2l29.1-6.1c9.3-1.9 16.7-9.1 17.8-18.5c.8-6.7 1.2-13.5 1.2-20.4s-.4-13.7-1.2-20.4c-1.1-9.4-8.6-16.6-17.8-18.5L583.9 307c-5-13.3-12.1-25.5-20.9-36.2l9.3-28.2c3-9 .5-19-7.1-24.7c-4.7-3.5-9.6-6.8-14.6-9.9l-5.3-3c-5-2.8-10.2-5.3-15.6-7.6c-8.7-3.7-18.6-.9-25 6.2l-19.8 22.2c-6.8-1.1-13.8-1.7-20.9-1.7s-14.1 .6-20.9 1.7l-19.8-22.2c-6.3-7.1-16.2-9.9-25-6.2c-5.3 2.3-10.5 4.8-15.6 7.6l-5.2 3c-5.1 3-9.9 6.3-14.6 9.9c-7.6 5.7-10.1 15.7-7.1 24.7l9.3 28.2c-8.8 10.7-16 23-20.9 36.2L315.1 313c-9.3 1.9-16.7 9.1-17.8 18.5c-.8 6.7-1.2 13.5-1.2 20.4s.4 13.7 1.2 20.4c1.1 9.4 8.6 16.6 17.8 18.5l29.1 6.1c5 13.3 12.1 25.5 20.9 36.2l-9.3 28.2c-3 9-.5 19 7.1 24.7c4.7 3.5 9.5 6.8 14.6 9.8l5.4 3.1c5 2.8 10.2 5.3 15.5 7.6c8.7 3.7 18.6 .9 25-6.2l19.8-22.2c6.8 1.1 13.8 1.7 20.9 1.7s14.1-.6 20.9-1.7l19.8 22.2zM464 304a48 48 0 1 1 0 96 48 48 0 1 1 0-96z\"],warehouse:[640,512,[],\"f494\",\"M0 488V171.3c0-26.2 15.9-49.7 40.2-59.4L308.1 4.8c7.6-3.1 16.1-3.1 23.8 0L599.8 111.9c24.3 9.7 40.2 33.3 40.2 59.4V488c0 13.3-10.7 24-24 24H568c-13.3 0-24-10.7-24-24V224c0-17.7-14.3-32-32-32H128c-17.7 0-32 14.3-32 32V488c0 13.3-10.7 24-24 24H24c-13.3 0-24-10.7-24-24zm488 24l-336 0c-13.3 0-24-10.7-24-24V432H512l0 56c0 13.3-10.7 24-24 24zM128 400V336H512v64H128zm0-96V224H512l0 80H128z\"],\"arrow-up-right-dots\":[576,512,[],\"e4b7\",\"M160 0c-17.7 0-32 14.3-32 32s14.3 32 32 32h50.7L9.4 265.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L256 109.3V160c0 17.7 14.3 32 32 32s32-14.3 32-32V32c0-17.7-14.3-32-32-32H160zM576 80a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM448 208a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM400 384a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm48 80a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zm128 0a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM272 384a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm48 80a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM144 512a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM576 336a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zm-48-80a48 48 0 1 0 0-96 48 48 0 1 0 0 96z\"],splotch:[512,512,[],\"f5bc\",\"M208.5 62.3l28.1-36.9C248.8 9.4 267.8 0 288 0c28.5 0 53.6 18.7 61.8 46l17.8 59.4c10.3 34.4 36.1 62 69.8 74.6l39.8 14.9c20.9 7.9 34.8 27.9 34.8 50.2c0 16.9-7.9 32.8-21.5 42.9l-67.3 50.5c-24.3 18.2-37.2 47.9-33.8 78.1l2.5 22.7c4.3 38.7-26 72.6-65 72.6c-14.8 0-29.3-5.1-40.8-14.3l-55.4-44.3c-4.5-3.6-9.3-6.7-14.5-9.2c-15.8-7.9-33.7-10.4-51-7.3L82.4 451.9C47.8 458.2 16 431.6 16 396.5c0-13.2 4.7-26 13.1-36.2l11.2-13.4c14.6-17.4 22.6-39.4 22.6-62.1c0-18.8-5.5-37.2-15.8-53L8.8 173.5C3.1 164.7 0 154.4 0 143.9c0-33.4 30.1-58.8 63-53.2l51.3 8.7c35.9 6.1 72.2-8.2 94.2-37.1z\"],\"face-grin-hearts\":[512,512,[128525,\"grin-hearts\"],\"f584\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM388.1 312.8c12.3-3.8 24.3 6.9 19.3 18.7C382.4 390.6 324.2 432 256.3 432s-126.2-41.4-151.1-100.5c-5-11.8 7-22.5 19.3-18.7c39.7 12.2 84.5 19 131.8 19s92.1-6.8 131.8-19zM199.3 129.1c17.8 4.8 28.4 23.1 23.6 40.8l-17.4 65c-2.3 8.5-11.1 13.6-19.6 11.3l-65.1-17.4c-17.8-4.8-28.4-23.1-23.6-40.8s23.1-28.4 40.8-23.6l16.1 4.3 4.3-16.1c4.8-17.8 23.1-28.4 40.8-23.6zm154.3 23.6l4.3 16.1 16.1-4.3c17.8-4.8 36.1 5.8 40.8 23.6s-5.8 36.1-23.6 40.8l-65.1 17.4c-8.5 2.3-17.3-2.8-19.6-11.3l-17.4-65c-4.8-17.8 5.8-36.1 23.6-40.8s36.1 5.8 40.9 23.6z\"],\"dice-four\":[448,512,[9859],\"f524\",\"M0 96C0 60.7 28.7 32 64 32H384c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm160 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM128 384a32 32 0 1 0 0-64 32 32 0 1 0 0 64zM352 160a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM320 384a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"sim-card\":[384,512,[],\"f7c4\",\"M64 0H242.7c17 0 33.3 6.7 45.3 18.7L365.3 96c12 12 18.7 28.3 18.7 45.3V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64C0 28.7 28.7 0 64 0zM96 192c-17.7 0-32 14.3-32 32v32h64V192H96zM64 352h80 96 80V288H240 144 64v64zM320 224c0-17.7-14.3-32-32-32H256v64h64V224zM160 192v64h64V192H160zM288 448c17.7 0 32-14.3 32-32V384H256v64h32zM160 384v64h64V384H160zM64 416c0 17.7 14.3 32 32 32h32V384H64v32z\"],transgender:[512,512,[9895,\"transgender-alt\"],\"f225\",\"M112 0c6.5 0 12.3 3.9 14.8 9.9s1.1 12.9-3.5 17.4l-31 31L112 78.1l7-7c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-7 7 15.2 15.2C187.7 107.6 220.5 96 256 96s68.3 11.6 94.9 31.2l68.8-68.8-31-31c-4.6-4.6-5.9-11.5-3.5-17.4s8.3-9.9 14.8-9.9h96c8.8 0 16 7.2 16 16v96c0 6.5-3.9 12.3-9.9 14.8s-12.9 1.1-17.4-3.5l-31-31-68.8 68.8C404.4 187.7 416 220.5 416 256c0 80.2-59 146.6-136 158.2V432h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H280v8c0 13.3-10.7 24-24 24s-24-10.7-24-24v-8H216c-13.3 0-24-10.7-24-24s10.7-24 24-24h16V414.2C155 402.6 96 336.2 96 256c0-35.5 11.6-68.3 31.2-94.9L112 145.9l-7 7c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l7-7L58.3 92.3l-31 31c-4.6 4.6-11.5 5.9-17.4 3.5S0 118.5 0 112V16C0 7.2 7.2 0 16 0h96zM352 256a96 96 0 1 0 -192 0 96 96 0 1 0 192 0z\"],mercury:[384,512,[9791],\"f223\",\"M72.1 7C85.8-4 106-1.8 117 12c17.6 22 44.7 36 75 36s57.3-14 75-36c11.1-13.8 31.2-16 45-5s16 31.2 5 45c-7.8 9.7-16.6 18.4-26.4 26.1C337.3 109.7 368 163.3 368 224c0 89.1-66.2 162.7-152 174.4V424h32c13.3 0 24 10.7 24 24s-10.7 24-24 24H216v16c0 13.3-10.7 24-24 24s-24-10.7-24-24V472H136c-13.3 0-24-10.7-24-24s10.7-24 24-24h32V398.4C82.2 386.7 16 313.1 16 224c0-60.7 30.7-114.3 77.5-145.9C83.7 70.5 74.9 61.7 67.1 52c-11.1-13.8-8.8-33.9 5-45zM80 224a112 112 0 1 0 224 0A112 112 0 1 0 80 224z\"],\"arrow-turn-down\":[384,512,[\"level-down\"],\"f149\",\"M32 64C14.3 64 0 49.7 0 32S14.3 0 32 0l96 0c53 0 96 43 96 96l0 306.7 73.4-73.4c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3l-128 128c-12.5 12.5-32.8 12.5-45.3 0l-128-128c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0L160 402.7 160 96c0-17.7-14.3-32-32-32L32 64z\"],\"person-falling-burst\":[640,512,[],\"e547\",\"M256 32c0-17.7-14.3-32-32-32s-32 14.3-32 32l0 9.8c0 39-23.7 74-59.9 88.4C71.6 154.5 32 213 32 278.2V352c0 17.7 14.3 32 32 32s32-14.3 32-32l0-73.8c0-10 1.6-19.8 4.5-29L261.1 497.4c9.6 14.8 29.4 19.1 44.3 9.5s19.1-29.4 9.5-44.3L222.6 320H224l80 0 38.4 51.2c10.6 14.1 30.7 17 44.8 6.4s17-30.7 6.4-44.8l-43.2-57.6C341.3 263.1 327.1 256 312 256l-71.5 0-56.8-80.2-.2-.3c44.7-29 72.5-79 72.5-133.6l0-9.8zM96 80A48 48 0 1 0 0 80a48 48 0 1 0 96 0zM464 286.1l58.6 53.9c4.8 4.4 11.9 5.5 17.8 2.6s9.5-9 9-15.5l-5.6-79.4 78.7-12.2c6.5-1 11.7-5.9 13.1-12.2s-1.1-13-6.5-16.7l-65.6-45.1L603 92.2c3.3-5.7 2.7-12.8-1.4-17.9s-10.9-7.2-17.2-5.3L508.3 92.1l-29.4-74C476.4 12 470.6 8 464 8s-12.4 4-14.9 10.1l-29.4 74L343.6 68.9c-6.3-1.9-13.1 .2-17.2 5.3s-4.6 12.2-1.4 17.9l39.5 69.1-65.6 45.1c-5.4 3.7-8 10.3-6.5 16.7c.1 .3 .1 .6 .2 .8l19.4 0c20.1 0 39.2 7.5 53.8 20.8l18.4 2.9L383 265.3l36.2 48.3c2.1 2.8 3.9 5.7 5.5 8.6L464 286.1z\"],award:[384,512,[],\"f559\",\"M173.8 5.5c11-7.3 25.4-7.3 36.4 0L228 17.2c6 3.9 13 5.8 20.1 5.4l21.3-1.3c13.2-.8 25.6 6.4 31.5 18.2l9.6 19.1c3.2 6.4 8.4 11.5 14.7 14.7L344.5 83c11.8 5.9 19 18.3 18.2 31.5l-1.3 21.3c-.4 7.1 1.5 14.2 5.4 20.1l11.8 17.8c7.3 11 7.3 25.4 0 36.4L366.8 228c-3.9 6-5.8 13-5.4 20.1l1.3 21.3c.8 13.2-6.4 25.6-18.2 31.5l-19.1 9.6c-6.4 3.2-11.5 8.4-14.7 14.7L301 344.5c-5.9 11.8-18.3 19-31.5 18.2l-21.3-1.3c-7.1-.4-14.2 1.5-20.1 5.4l-17.8 11.8c-11 7.3-25.4 7.3-36.4 0L156 366.8c-6-3.9-13-5.8-20.1-5.4l-21.3 1.3c-13.2 .8-25.6-6.4-31.5-18.2l-9.6-19.1c-3.2-6.4-8.4-11.5-14.7-14.7L39.5 301c-11.8-5.9-19-18.3-18.2-31.5l1.3-21.3c.4-7.1-1.5-14.2-5.4-20.1L5.5 210.2c-7.3-11-7.3-25.4 0-36.4L17.2 156c3.9-6 5.8-13 5.4-20.1l-1.3-21.3c-.8-13.2 6.4-25.6 18.2-31.5l19.1-9.6C65 70.2 70.2 65 73.4 58.6L83 39.5c5.9-11.8 18.3-19 31.5-18.2l21.3 1.3c7.1 .4 14.2-1.5 20.1-5.4L173.8 5.5zM272 192a80 80 0 1 0 -160 0 80 80 0 1 0 160 0zM1.3 441.8L44.4 339.3c.2 .1 .3 .2 .4 .4l9.6 19.1c11.7 23.2 36 37.3 62 35.8l21.3-1.3c.2 0 .5 0 .7 .2l17.8 11.8c5.1 3.3 10.5 5.9 16.1 7.7l-37.6 89.3c-2.3 5.5-7.4 9.2-13.3 9.7s-11.6-2.2-14.8-7.2L74.4 455.5l-56.1 8.3c-5.7 .8-11.4-1.5-15-6s-4.3-10.7-2.1-16zm248 60.4L211.7 413c5.6-1.8 11-4.3 16.1-7.7l17.8-11.8c.2-.1 .4-.2 .7-.2l21.3 1.3c26 1.5 50.3-12.6 62-35.8l9.6-19.1c.1-.2 .2-.3 .4-.4l43.2 102.5c2.2 5.3 1.4 11.4-2.1 16s-9.3 6.9-15 6l-56.1-8.3-32.2 49.2c-3.2 5-8.9 7.7-14.8 7.2s-11-4.3-13.3-9.7z\"],\"ticket-simple\":[576,512,[\"ticket-alt\"],\"f3ff\",\"M0 128C0 92.7 28.7 64 64 64H512c35.3 0 64 28.7 64 64v64c0 8.8-7.4 15.7-15.7 18.6C541.5 217.1 528 235 528 256s13.5 38.9 32.3 45.4c8.3 2.9 15.7 9.8 15.7 18.6v64c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V320c0-8.8 7.4-15.7 15.7-18.6C34.5 294.9 48 277 48 256s-13.5-38.9-32.3-45.4C7.4 207.7 0 200.8 0 192V128z\"],building:[384,512,[127970,61687],\"f1ad\",\"M48 0C21.5 0 0 21.5 0 48V464c0 26.5 21.5 48 48 48h96V432c0-26.5 21.5-48 48-48s48 21.5 48 48v80h96c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48H48zM64 240c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V240zm112-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V240c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V240zM80 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16zm80 16c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H176c-8.8 0-16-7.2-16-16V112zM272 96h32c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V112c0-8.8 7.2-16 16-16z\"],\"angles-left\":[512,512,[171,\"angle-double-left\"],\"f100\",\"M41.4 233.4c-12.5 12.5-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L109.3 256 246.6 118.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-160 160zm352-160l-160 160c-12.5 12.5-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L301.3 256 438.6 118.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0z\"],qrcode:[448,512,[],\"f029\",\"M0 80C0 53.5 21.5 32 48 32h96c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V80zM64 96v64h64V96H64zM0 336c0-26.5 21.5-48 48-48h96c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V336zm64 16v64h64V352H64zM304 32h96c26.5 0 48 21.5 48 48v96c0 26.5-21.5 48-48 48H304c-26.5 0-48-21.5-48-48V80c0-26.5 21.5-48 48-48zm80 64H320v64h64V96zM256 304c0-8.8 7.2-16 16-16h64c8.8 0 16 7.2 16 16s7.2 16 16 16h32c8.8 0 16-7.2 16-16s7.2-16 16-16s16 7.2 16 16v96c0 8.8-7.2 16-16 16H368c-8.8 0-16-7.2-16-16s-7.2-16-16-16s-16 7.2-16 16v64c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V304zM368 480a16 16 0 1 1 0-32 16 16 0 1 1 0 32zm64 0a16 16 0 1 1 0-32 16 16 0 1 1 0 32z\"],\"clock-rotate-left\":[512,512,[\"history\"],\"f1da\",\"M75 75L41 41C25.9 25.9 0 36.6 0 57.9V168c0 13.3 10.7 24 24 24H134.1c21.4 0 32.1-25.9 17-41l-30.8-30.8C155 85.5 203 64 256 64c106 0 192 86 192 192s-86 192-192 192c-40.8 0-78.6-12.7-109.7-34.4c-14.5-10.1-34.4-6.6-44.6 7.9s-6.6 34.4 7.9 44.6C151.2 495 201.7 512 256 512c141.4 0 256-114.6 256-256S397.4 0 256 0C185.3 0 121.3 28.7 75 75zm181 53c-13.3 0-24 10.7-24 24V256c0 6.4 2.5 12.5 7 17l72 72c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-65-65V152c0-13.3-10.7-24-24-24z\"],\"face-grin-beam-sweat\":[512,512,[128517,\"grin-beam-sweat\"],\"f583\",\"M476.8 126.3c-4.1 1.1-8.4 1.7-12.8 1.7c-26.5 0-48-21-48-47c0-5 1.8-11.3 4.6-18.1c.3-.7 .6-1.4 .9-2.1c9-20.2 26.5-44.9 36-57.5c3.2-4.4 9.6-4.4 12.8 0C483.4 20.6 512 61 512 81c0 21.7-14.9 39.8-35.2 45.3zM256 0c51.4 0 99.3 15.2 139.4 41.2c-1.5 3.1-3 6.2-4.3 9.3c-3.4 8-7.1 19-7.1 30.5c0 44.3 36.6 79 80 79c9.6 0 18.8-1.7 27.4-4.8c13.3 30.9 20.6 65 20.6 100.8c0 141.4-114.6 256-256 256S0 397.4 0 256S114.6 0 256 0zM388.1 312.8c-39.7 12.2-84.5 19-131.8 19s-92.1-6.8-131.8-19c-12.3-3.8-24.3 6.9-19.3 18.7c25 59.1 83.2 100.5 151.1 100.5s126.2-41.4 151.1-100.5c5-11.8-7-22.5-19.3-18.7zm-170.5-84l0 0 0 0c2.1 2.8 5.7 3.9 8.9 2.8s5.5-4.1 5.5-7.6c0-17.9-6.7-35.6-16.6-48.8c-9.8-13-23.9-23.2-39.4-23.2s-29.6 10.2-39.4 23.2C126.7 188.4 120 206.1 120 224c0 3.4 2.2 6.5 5.5 7.6s6.9 0 8.9-2.8l0 0 0 0 0 0 .2-.2c.2-.2 .4-.5 .7-.9c.6-.8 1.6-2 2.8-3.4c2.5-2.8 6-6.6 10.2-10.3c8.8-7.8 18.8-14 27.7-14s18.9 6.2 27.7 14c4.2 3.7 7.7 7.5 10.2 10.3c1.2 1.4 2.2 2.6 2.8 3.4c.3 .4 .6 .7 .7 .9l.2 .2 0 0 0 0zm160 0l0 0 0 0 0 0c2.1 2.8 5.7 3.9 8.9 2.8s5.5-4.1 5.5-7.6c0-17.9-6.7-35.6-16.6-48.8c-9.8-13-23.9-23.2-39.4-23.2s-29.6 10.2-39.4 23.2C286.7 188.4 280 206.1 280 224c0 3.4 2.2 6.5 5.5 7.6s6.9 0 8.9-2.8l0 0 0 0 0 0 .2-.2c.2-.2 .4-.5 .7-.9c.6-.8 1.6-2 2.8-3.4c2.5-2.8 6-6.6 10.2-10.3c8.8-7.8 18.8-14 27.7-14s18.9 6.2 27.7 14c4.2 3.7 7.7 7.5 10.2 10.3c1.2 1.4 2.2 2.6 2.8 3.4c.3 .4 .6 .7 .7 .9l.2 .2 0 0z\"],\"file-export\":[576,512,[\"arrow-right-from-file\"],\"f56e\",\"M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384V288H216c-13.3 0-24 10.7-24 24s10.7 24 24 24H384V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V64zM384 336V288H494.1l-39-39c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l80 80c9.4 9.4 9.4 24.6 0 33.9l-80 80c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l39-39H384zm0-208H256V0L384 128z\"],shield:[512,512,[128737,\"shield-blank\"],\"f132\",\"M256 0c4.6 0 9.2 1 13.4 2.9L457.7 82.8c22 9.3 38.4 31 38.3 57.2c-.5 99.2-41.3 280.7-213.6 363.2c-16.7 8-36.1 8-52.8 0C57.3 420.7 16.5 239.2 16 140c-.1-26.2 16.3-47.9 38.3-57.2L242.7 2.9C246.8 1 251.4 0 256 0z\"],\"arrow-up-short-wide\":[576,512,[\"sort-amount-up-alt\"],\"f885\",\"M151.6 42.4C145.5 35.8 137 32 128 32s-17.5 3.8-23.6 10.4l-88 96c-11.9 13-11.1 33.3 2 45.2s33.3 11.1 45.2-2L96 146.3V448c0 17.7 14.3 32 32 32s32-14.3 32-32V146.3l32.4 35.4c11.9 13 32.2 13.9 45.2 2s13.9-32.2 2-45.2l-88-96zM320 32c-17.7 0-32 14.3-32 32s14.3 32 32 32h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H320zm0 128c-17.7 0-32 14.3-32 32s14.3 32 32 32h96c17.7 0 32-14.3 32-32s-14.3-32-32-32H320zm0 128c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H320zm0 128c-17.7 0-32 14.3-32 32s14.3 32 32 32H544c17.7 0 32-14.3 32-32s-14.3-32-32-32H320z\"],\"house-medical\":[576,512,[],\"e3b2\",\"M543.8 287.6c17 0 32-14 32-32.1c1-9-3-17-11-24L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32V448c0 35.3 28.7 64 64 64H448.5c35.5 0 64.2-28.8 64-64.3l-.7-160.2h32zM256 208c0-8.8 7.2-16 16-16h32c8.8 0 16 7.2 16 16v48h48c8.8 0 16 7.2 16 16v32c0 8.8-7.2 16-16 16H320v48c0 8.8-7.2 16-16 16H272c-8.8 0-16-7.2-16-16V320H208c-8.8 0-16-7.2-16-16V272c0-8.8 7.2-16 16-16h48V208z\"],\"golf-ball-tee\":[384,512,[\"golf-ball\"],\"f450\",\"M384 192c0 66.8-34.1 125.6-85.8 160H85.8C34.1 317.6 0 258.8 0 192C0 86 86 0 192 0S384 86 384 192zM242.1 256.6c0 18.5-15 33.5-33.5 33.5c-4.9 0-9.1 5.1-5.4 8.4c5.9 5.2 13.7 8.4 22.1 8.4c18.5 0 33.5-15 33.5-33.5c0-8.5-3.2-16.2-8.4-22.1c-3.3-3.7-8.4 .5-8.4 5.4zm-52.3-49.3c-4.9 0-9.1 5.1-5.4 8.4c5.9 5.2 13.7 8.4 22.1 8.4c18.5 0 33.5-15 33.5-33.5c0-8.5-3.2-16.2-8.4-22.1c-3.3-3.7-8.4 .5-8.4 5.4c0 18.5-15 33.5-33.5 33.5zm113.5-17.5c0 18.5-15 33.5-33.5 33.5c-4.9 0-9.1 5.1-5.4 8.4c5.9 5.2 13.7 8.4 22.1 8.4c18.5 0 33.5-15 33.5-33.5c0-8.5-3.2-16.2-8.4-22.1c-3.3-3.7-8.4 .5-8.4 5.4zM96 416c0-17.7 14.3-32 32-32h64 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H240c-8.8 0-16 7.2-16 16v16c0 17.7-14.3 32-32 32s-32-14.3-32-32V464c0-8.8-7.2-16-16-16H128c-17.7 0-32-14.3-32-32z\"],\"circle-chevron-left\":[512,512,[\"chevron-circle-left\"],\"f137\",\"M512 256A256 256 0 1 0 0 256a256 256 0 1 0 512 0zM271 135c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-87 87 87 87c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0L167 273c-9.4-9.4-9.4-24.6 0-33.9L271 135z\"],\"house-chimney-window\":[576,512,[],\"e00d\",\"M575.8 255.5c0 18-15 32.1-32 32.1h-32l.7 160.2c.2 35.5-28.5 64.3-64 64.3H128.1c-35.3 0-64-28.7-64-64V287.6H32c-18 0-32-14-32-32.1c0-9 3-17 10-24L266.4 8c7-7 15-8 22-8s15 2 21 7L416 100.7V64c0-17.7 14.3-32 32-32h32c17.7 0 32 14.3 32 32V185l52.8 46.4c8 7 12 15 11 24zM248 192c-13.3 0-24 10.7-24 24v80c0 13.3 10.7 24 24 24h80c13.3 0 24-10.7 24-24V216c0-13.3-10.7-24-24-24H248z\"],\"pen-nib\":[512,512,[10001],\"f5ad\",\"M368.4 18.3L312.7 74.1 437.9 199.3l55.7-55.7c21.9-21.9 21.9-57.3 0-79.2L447.6 18.3c-21.9-21.9-57.3-21.9-79.2 0zM288 94.6l-9.2 2.8L134.7 140.6c-19.9 6-35.7 21.2-42.3 41L3.8 445.8c-3.8 11.3-1 23.9 7.3 32.4L164.7 324.7c-3-6.3-4.7-13.3-4.7-20.7c0-26.5 21.5-48 48-48s48 21.5 48 48s-21.5 48-48 48c-7.4 0-14.4-1.7-20.7-4.7L33.7 500.9c8.6 8.3 21.1 11.2 32.4 7.3l264.3-88.6c19.7-6.6 35-22.4 41-42.3l43.2-144.1 2.8-9.2L288 94.6z\"],\"tent-arrow-turn-left\":[576,512,[],\"e580\",\"M120.1 41.8c9.9-8.9 10.7-24 1.8-33.9S97.8-2.7 87.9 6.2l-80 72C2.9 82.7 0 89.2 0 96s2.9 13.3 7.9 17.8l80 72c9.9 8.9 25 8.1 33.9-1.8s8.1-25-1.8-33.9L86.5 120 456 120c39.8 0 72 32.2 72 72v40c0 13.3 10.7 24 24 24s24-10.7 24-24V192c0-66.3-53.7-120-120-120L86.5 72l33.5-30.2zM307.4 166.5c-11.5-8.7-27.3-8.7-38.8 0l-168 128c-6.6 5-11 12.5-12.3 20.7l-24 160c-1.4 9.2 1.3 18.6 7.4 25.6S86.7 512 96 512H240h16c17.7 0 32-14.3 32-32V361.9c0-5.5 4.4-9.9 9.9-9.9c3.7 0 7.2 2.1 8.8 5.5l68.4 136.8c5.4 10.8 16.5 17.7 28.6 17.7H464h16c9.3 0 18.2-4.1 24.2-11.1s8.8-16.4 7.4-25.6l-24-160c-1.2-8.2-5.6-15.7-12.3-20.7l-168-128z\"],tents:[640,512,[],\"e582\",\"M396.6 6.5L235.8 129.1c9.6 1.8 18.9 5.8 27 12l168 128c13.2 10.1 22 24.9 24.5 41.4l6.2 41.5H608c9.3 0 18.2-4.1 24.2-11.1s8.8-16.4 7.4-25.6l-24-160c-1.2-8.2-5.6-15.7-12.3-20.7l-168-128c-11.5-8.7-27.3-8.7-38.8 0zm-153.2 160c-11.5-8.7-27.3-8.7-38.8 0l-168 128c-6.6 5-11 12.5-12.3 20.7l-24 160c-1.4 9.2 1.3 18.6 7.4 25.6S22.7 512 32 512H176h16c17.7 0 32-14.3 32-32V361.9c0-5.5 4.4-9.9 9.9-9.9c3.7 0 7.2 2.1 8.8 5.5l68.4 136.8c5.4 10.8 16.5 17.7 28.6 17.7H400h16c9.3 0 18.2-4.1 24.2-11.1s8.8-16.4 7.4-25.6l-24-160c-1.2-8.2-5.6-15.7-12.3-20.7l-168-128z\"],\"wand-magic\":[512,512,[\"magic\"],\"f0d0\",\"M14.1 463.3c-18.7-18.7-18.7-49.1 0-67.9L395.4 14.1c18.7-18.7 49.1-18.7 67.9 0l34.6 34.6c18.7 18.7 18.7 49.1 0 67.9L116.5 497.9c-18.7 18.7-49.1 18.7-67.9 0L14.1 463.3zM347.6 187.6l105-105L429.4 59.3l-105 105 23.3 23.3z\"],dog:[576,512,[128021],\"f6d3\",\"M309.6 158.5L332.7 19.8C334.6 8.4 344.5 0 356.1 0c7.5 0 14.5 3.5 19 9.5L392 32h52.1c12.7 0 24.9 5.1 33.9 14.1L496 64h56c13.3 0 24 10.7 24 24v24c0 44.2-35.8 80-80 80H464 448 426.7l-5.1 30.5-112-64zM416 256.1L416 480c0 17.7-14.3 32-32 32H352c-17.7 0-32-14.3-32-32V364.8c-24 12.3-51.2 19.2-80 19.2s-56-6.9-80-19.2V480c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V249.8c-28.8-10.9-51.4-35.3-59.2-66.5L1 167.8c-4.3-17.1 6.1-34.5 23.3-38.8s34.5 6.1 38.8 23.3l3.9 15.5C70.5 182 83.3 192 98 192h30 16H303.8L416 256.1zM464 80a16 16 0 1 0 -32 0 16 16 0 1 0 32 0z\"],carrot:[512,512,[129365],\"f787\",\"M346.7 6C337.6 17 320 42.3 320 72c0 40 15.3 55.3 40 80s40 40 80 40c29.7 0 55-17.6 66-26.7c4-3.3 6-8.2 6-13.3s-2-10-6-13.2c-11.4-9.1-38.3-26.8-74-26.8c-32 0-40 8-40 8s8-8 8-40c0-35.7-17.7-62.6-26.8-74C370 2 365.1 0 360 0s-10 2-13.3 6zM244.6 136c-40 0-77.1 18.1-101.7 48.2l60.5 60.5c6.2 6.2 6.2 16.4 0 22.6s-16.4 6.2-22.6 0l-55.3-55.3 0 .1L2.2 477.9C-2 487-.1 497.8 7 505s17.9 9 27.1 4.8l134.7-62.4-52.1-52.1c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L199.7 433l100.2-46.4c46.4-21.5 76.2-68 76.2-119.2C376 194.8 317.2 136 244.6 136z\"],moon:[384,512,[127769,9214],\"f186\",\"M223.5 32C100 32 0 132.3 0 256S100 480 223.5 480c60.6 0 115.5-24.2 155.8-63.4c5-4.9 6.3-12.5 3.1-18.7s-10.1-9.7-17-8.5c-9.8 1.7-19.8 2.6-30.1 2.6c-96.9 0-175.5-78.8-175.5-176c0-65.8 36-123.1 89.3-153.3c6.1-3.5 9.2-10.5 7.7-17.3s-7.3-11.9-14.3-12.5c-6.3-.5-12.6-.8-19-.8z\"],\"wine-glass-empty\":[320,512,[\"wine-glass-alt\"],\"f5ce\",\"M64 0C47.4 0 33.5 12.8 32.1 29.3l-14 168.4c-6 72 42.5 135.2 109.9 150.6V448H80c-17.7 0-32 14.3-32 32s14.3 32 32 32h80 80c17.7 0 32-14.3 32-32s-14.3-32-32-32H192V348.4c67.4-15.4 115.9-78.6 109.9-150.6l-14-168.4C286.5 12.8 272.6 0 256 0H64zM81.9 203.1L93.4 64H226.6l11.6 139.1C242 248.8 205.9 288 160 288s-82-39.2-78.1-84.9z\"],cheese:[512,512,[],\"f7ef\",\"M512 240.2V256H0c0-20 10-38.7 26.6-49.8L274.9 40.7c8.6-5.7 18.6-8.7 28.9-8.7C418.8 32 512 125.2 512 240.2zm0 47.8V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V288H512z\"],\"yin-yang\":[512,512,[9775],\"f6ad\",\"M256 64c53 0 96 43 96 96s-43 96-96 96s-96 43-96 96s43 96 96 96C150 448 64 362 64 256S150 64 256 64zm0 448A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm32-352a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM224 352a32 32 0 1 1 64 0 32 32 0 1 1 -64 0z\"],music:[512,512,[127925],\"f001\",\"M499.1 6.3c8.1 6 12.9 15.6 12.9 25.7v72V368c0 44.2-43 80-96 80s-96-35.8-96-80s43-80 96-80c11.2 0 22 1.6 32 4.6V147L192 223.8V432c0 44.2-43 80-96 80s-96-35.8-96-80s43-80 96-80c11.2 0 22 1.6 32 4.6V200 128c0-14.1 9.3-26.6 22.8-30.7l320-96c9.7-2.9 20.2-1.1 28.3 5z\"],\"code-commit\":[640,512,[],\"f386\",\"M320 336a80 80 0 1 0 0-160 80 80 0 1 0 0 160zm156.8-48C462 361 397.4 416 320 416s-142-55-156.8-128H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H163.2C178 151 242.6 96 320 96s142 55 156.8 128H608c17.7 0 32 14.3 32 32s-14.3 32-32 32H476.8z\"],\"temperature-low\":[512,512,[],\"f76b\",\"M448 96a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM320 96a96 96 0 1 1 192 0A96 96 0 1 1 320 96zM144 64c-26.5 0-48 21.5-48 48V276.5c0 17.3-7.1 31.9-15.3 42.5C70.2 332.6 64 349.5 64 368c0 44.2 35.8 80 80 80s80-35.8 80-80c0-18.5-6.2-35.4-16.7-48.9c-8.2-10.6-15.3-25.2-15.3-42.5V112c0-26.5-21.5-48-48-48zM32 112C32 50.2 82.1 0 144 0s112 50.1 112 112V276.5c0 .1 .1 .3 .2 .6c.2 .6 .8 1.6 1.7 2.8c18.9 24.4 30.1 55 30.1 88.1c0 79.5-64.5 144-144 144S0 447.5 0 368c0-33.2 11.2-63.8 30.1-88.1c.9-1.2 1.5-2.2 1.7-2.8c.1-.3 .2-.5 .2-.6V112zM192 368c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-20.9 13.4-38.7 32-45.3V272c0-8.8 7.2-16 16-16s16 7.2 16 16v50.7c18.6 6.6 32 24.4 32 45.3z\"],\"person-biking\":[640,512,[128692,\"biking\"],\"f84a\",\"M400 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm27.2 64l-61.8-48.8c-17.3-13.6-41.7-13.8-59.1-.3l-83.1 64.2c-30.7 23.8-28.5 70.8 4.3 91.6L288 305.1V416c0 17.7 14.3 32 32 32s32-14.3 32-32V288c0-10.7-5.3-20.7-14.2-26.6L295 232.9l60.3-48.5L396 217c5.7 4.5 12.7 7 20 7h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H427.2zM56 384a72 72 0 1 1 144 0A72 72 0 1 1 56 384zm200 0A128 128 0 1 0 0 384a128 128 0 1 0 256 0zm184 0a72 72 0 1 1 144 0 72 72 0 1 1 -144 0zm200 0a128 128 0 1 0 -256 0 128 128 0 1 0 256 0z\"],broom:[576,512,[129529],\"f51a\",\"M566.6 54.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-192 192-34.7-34.7c-4.2-4.2-10-6.6-16-6.6c-12.5 0-22.6 10.1-22.6 22.6v29.1L364.3 320h29.1c12.5 0 22.6-10.1 22.6-22.6c0-6-2.4-11.8-6.6-16l-34.7-34.7 192-192zM341.1 353.4L222.6 234.9c-42.7-3.7-85.2 11.7-115.8 42.3l-8 8C76.5 307.5 64 337.7 64 369.2c0 6.8 7.1 11.2 13.2 8.2l51.1-25.5c5-2.5 9.5 4.1 5.4 7.9L7.3 473.4C2.7 477.6 0 483.6 0 489.9C0 502.1 9.9 512 22.1 512l173.3 0c38.8 0 75.9-15.4 103.4-42.8c30.6-30.6 45.9-73.1 42.3-115.8z\"],\"shield-heart\":[512,512,[],\"e574\",\"M269.4 2.9C265.2 1 260.7 0 256 0s-9.2 1-13.4 2.9L54.3 82.8c-22 9.3-38.4 31-38.3 57.2c.5 99.2 41.3 280.7 213.6 363.2c16.7 8 36.1 8 52.8 0C454.7 420.7 495.5 239.2 496 140c.1-26.2-16.3-47.9-38.3-57.2L269.4 2.9zM144 221.3c0-33.8 27.4-61.3 61.3-61.3c16.2 0 31.8 6.5 43.3 17.9l7.4 7.4 7.4-7.4c11.5-11.5 27.1-17.9 43.3-17.9c33.8 0 61.3 27.4 61.3 61.3c0 16.2-6.5 31.8-17.9 43.3l-82.7 82.7c-6.2 6.2-16.4 6.2-22.6 0l-82.7-82.7c-11.5-11.5-17.9-27.1-17.9-43.3z\"],gopuram:[512,512,[],\"f664\",\"M120 0c13.3 0 24 10.7 24 24v8h40V24c0-13.3 10.7-24 24-24s24 10.7 24 24v8h48V24c0-13.3 10.7-24 24-24s24 10.7 24 24v8h40V24c0-13.3 10.7-24 24-24s24 10.7 24 24v8V64v64c17.7 0 32 14.3 32 32v64c17.7 0 32 14.3 32 32v96c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32H416V352H384V224H352V128H320v96h32V352h32V512H304V464c0-26.5-21.5-48-48-48s-48 21.5-48 48v48H128V352h32V224h32V128H160v96H128V352H96V512H32c-17.7 0-32-14.3-32-32V384c0-17.7 14.3-32 32-32V256c0-17.7 14.3-32 32-32V160c0-17.7 14.3-32 32-32V64 32 24c0-13.3 10.7-24 24-24zM256 272c-17.7 0-32 14.3-32 32v48h64V304c0-17.7-14.3-32-32-32zm-32-80v32h64V192c0-17.7-14.3-32-32-32s-32 14.3-32 32z\"],\"earth-oceania\":[512,512,[\"globe-oceania\"],\"e47b\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM208.6 357.3l-39-13.5c-6.5-2.2-13.6-2.3-20.1-.3l-15.3 4.9c-18.5 5.9-38.5-2.4-47.5-19.5l-3.3-6.2c-10.6-20.1-2.3-45 18.2-54.7l35.3-16.8c2.3-1.1 4.4-2.8 5.9-4.8l5.3-7c7.2-9.6 18.6-15.3 30.6-15.3s23.4 5.7 30.6 15.3l4.6 6.1c2 2.6 4.9 4.5 8.1 5.1c7.8 1.6 15.7-1.5 20.4-7.9l10.4-14.2c2-2.8 5.3-4.4 8.7-4.4c4.4 0 8.4 2.7 10 6.8l10.1 25.9c2.8 7.2 6.7 14 11.5 20.2L311 299.8c5.8 7.4 9 16.6 9 26s-3.2 18.6-9 26L299 367.2c-8.3 10.6-21 16.8-34.4 16.8c-8.4 0-16.6-2.4-23.7-7l-25.4-16.4c-2.2-1.4-4.5-2.5-6.9-3.4zm65.2-214.8L296 164.7c10.1 10.1 2.9 27.3-11.3 27.3H254.8c-5.6 0-11.1-1.2-16.2-3.4l-42.8-19c-14.3-6.3-11.9-27.3 3.4-30.3l38.5-7.7c13.1-2.6 26.7 1.5 36.1 10.9zM248 432c0-8.8 7.2-16 16-16h16c8.8 0 16 7.2 16 16s-7.2 16-16 16H264c-8.8 0-16-7.2-16-16zM431.2 298.9l8 24c2.8 8.4-1.7 17.4-10.1 20.2s-17.4-1.7-20.2-10.1l-8-24c-2.8-8.4 1.7-17.4 10.1-20.2s17.4 1.7 20.2 10.1zm-19.9 80.4l-32 32c-6.2 6.2-16.4 6.2-22.6 0s-6.2-16.4 0-22.6l32-32c6.2-6.2 16.4-6.2 22.6 0s6.2 16.4 0 22.6z\"],\"square-xmark\":[448,512,[10062,\"times-square\",\"xmark-square\"],\"f2d3\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zm79 143c9.4-9.4 24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9z\"],hashtag:[448,512,[62098],\"23\",\"M181.3 32.4c17.4 2.9 29.2 19.4 26.3 36.8L197.8 128h95.1l11.5-69.3c2.9-17.4 19.4-29.2 36.8-26.3s29.2 19.4 26.3 36.8L357.8 128H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H347.1L325.8 320H384c17.7 0 32 14.3 32 32s-14.3 32-32 32H315.1l-11.5 69.3c-2.9 17.4-19.4 29.2-36.8 26.3s-29.2-19.4-26.3-36.8l9.8-58.7H155.1l-11.5 69.3c-2.9 17.4-19.4 29.2-36.8 26.3s-29.2-19.4-26.3-36.8L90.2 384H32c-17.7 0-32-14.3-32-32s14.3-32 32-32h68.9l21.3-128H64c-17.7 0-32-14.3-32-32s14.3-32 32-32h68.9l11.5-69.3c2.9-17.4 19.4-29.2 36.8-26.3zM187.1 192L165.8 320h95.1l21.3-128H187.1z\"],\"up-right-and-down-left-from-center\":[512,512,[\"expand-alt\"],\"f424\",\"M344 0H488c13.3 0 24 10.7 24 24V168c0 9.7-5.8 18.5-14.8 22.2s-19.3 1.7-26.2-5.2l-39-39-87 87c-9.4 9.4-24.6 9.4-33.9 0l-32-32c-9.4-9.4-9.4-24.6 0-33.9l87-87L327 41c-6.9-6.9-8.9-17.2-5.2-26.2S334.3 0 344 0zM168 512H24c-13.3 0-24-10.7-24-24V344c0-9.7 5.8-18.5 14.8-22.2s19.3-1.7 26.2 5.2l39 39 87-87c9.4-9.4 24.6-9.4 33.9 0l32 32c9.4 9.4 9.4 24.6 0 33.9l-87 87 39 39c6.9 6.9 8.9 17.2 5.2 26.2s-12.5 14.8-22.2 14.8z\"],\"oil-can\":[640,512,[],\"f613\",\"M320 128c17.7 0 32-14.3 32-32s-14.3-32-32-32H192c-17.7 0-32 14.3-32 32s14.3 32 32 32h32v32H144 96 48c-26.5 0-48 21.5-48 48v64.8c0 19 11.2 36.2 28.5 43.9l67.5 30V368c0 26.5 21.5 48 48 48H403.1c18.4 0 35.8-7.9 48-21.7L633.5 187.7c12.3-13.9-.3-35.4-18.4-31.5L448 192l-50.5-25.2c-8.9-4.4-18.7-6.8-28.6-6.8H288V128h32zM96 208v86.1L48 272.8V208H96z\"],t:[384,512,[116],\"54\",\"M32 32C14.3 32 0 46.3 0 64S14.3 96 32 96H160V448c0 17.7 14.3 32 32 32s32-14.3 32-32V96H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H192 32z\"],hippo:[640,512,[129435],\"f6ed\",\"M407 47c9.4-9.4 24.6-9.4 33.9 0l17.2 17.2c1.9-.1 3.9-.2 5.8-.2h32c11.2 0 21.9 2.3 31.6 6.5L543 55c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9L564 101.9c7.6 12.2 12 26.7 12 42.1c0 10.2 7.4 18.8 16.7 23c27.9 12.5 47.3 40.5 47.3 73c0 26.2-12.6 49.4-32 64v32c0 8.8-7.2 16-16 16H560c-8.8 0-16-7.2-16-16V320H480v16c0 8.8-7.2 16-16 16H432c-8.8 0-16-7.2-16-16V318.4c-11.8-2.4-22.7-7.4-32-14.4c-1.5-1.1-2.9-2.3-4.3-3.5c-17-14.7-27.7-36.4-27.7-60.5c0-8.8-7.2-16-16-16s-16 7.2-16 16c0 44.7 26.2 83.2 64 101.2V352c0 17.7 14.3 32 32 32h32v64c0 17.7-14.3 32-32 32H352c-17.7 0-32-14.3-32-32V372c-19.8 7.7-41.4 12-64 12s-44.2-4.3-64-12v76c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V329.1L45.9 369.7c-5.4 12.1-19.6 17.6-31.7 12.2S-3.3 362.4 2.1 350.3L24 300.9c5.3-11.9 8-24.7 8-37.7C32 155.7 117.2 68 223.8 64.1l.2-.1h7.2H256h32c41.7 0 83.4 12.1 117.2 25.7c1.7-1.8 3.5-3.6 5.3-5.2L407 81c-9.4-9.4-9.4-24.6 0-33.9zm73 185a24 24 0 1 0 -48 0 24 24 0 1 0 48 0zm88 24a24 24 0 1 0 0-48 24 24 0 1 0 0 48zM480 144a16 16 0 1 0 -32 0 16 16 0 1 0 32 0zm48 16a16 16 0 1 0 0-32 16 16 0 1 0 0 32z\"],\"chart-column\":[512,512,[],\"e0e3\",\"M32 32c17.7 0 32 14.3 32 32V400c0 8.8 7.2 16 16 16H480c17.7 0 32 14.3 32 32s-14.3 32-32 32H80c-44.2 0-80-35.8-80-80V64C0 46.3 14.3 32 32 32zM160 224c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V256c0-17.7 14.3-32 32-32zm128-64V320c0 17.7-14.3 32-32 32s-32-14.3-32-32V160c0-17.7 14.3-32 32-32s32 14.3 32 32zm64 32c17.7 0 32 14.3 32 32v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V224c0-17.7 14.3-32 32-32zM480 96V320c0 17.7-14.3 32-32 32s-32-14.3-32-32V96c0-17.7 14.3-32 32-32s32 14.3 32 32z\"],infinity:[640,512,[8734,9854],\"f534\",\"M0 241.1C0 161 65 96 145.1 96c38.5 0 75.4 15.3 102.6 42.5L320 210.7l72.2-72.2C419.5 111.3 456.4 96 494.9 96C575 96 640 161 640 241.1v29.7C640 351 575 416 494.9 416c-38.5 0-75.4-15.3-102.6-42.5L320 301.3l-72.2 72.2C220.5 400.7 183.6 416 145.1 416C65 416 0 351 0 270.9V241.1zM274.7 256l-72.2-72.2c-15.2-15.2-35.9-23.8-57.4-23.8C100.3 160 64 196.3 64 241.1v29.7c0 44.8 36.3 81.1 81.1 81.1c21.5 0 42.2-8.5 57.4-23.8L274.7 256zm90.5 0l72.2 72.2c15.2 15.2 35.9 23.8 57.4 23.8c44.8 0 81.1-36.3 81.1-81.1V241.1c0-44.8-36.3-81.1-81.1-81.1c-21.5 0-42.2 8.5-57.4 23.8L365.3 256z\"],\"vial-circle-check\":[512,512,[],\"e596\",\"M0 64C0 46.3 14.3 32 32 32H96h64 64c17.7 0 32 14.3 32 32s-14.3 32-32 32V266.8c-20.2 28.6-32 63.5-32 101.2c0 25.2 5.3 49.1 14.8 70.8C189.5 463.7 160.6 480 128 480c-53 0-96-43-96-96V96C14.3 96 0 81.7 0 64zM96 96v96h64V96H96zM224 368a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm211.3-43.3c-6.2-6.2-16.4-6.2-22.6 0L352 385.4l-28.7-28.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l40 40c6.2 6.2 16.4 6.2 22.6 0l72-72c6.2-6.2 6.2-16.4 0-22.6z\"],\"person-arrow-down-to-line\":[640,512,[],\"e538\",\"M192 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm-8 352V352h16v96H184zm-64 0H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H152h80H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H264V256.9l28.6 47.5c9.1 15.1 28.8 20 43.9 10.9s20-28.8 10.9-43.9l-58.3-97c-17.4-28.9-48.6-46.6-82.3-46.6H177.1c-33.7 0-64.9 17.7-82.3 46.6l-58.3 97c-9.1 15.1-4.2 34.8 10.9 43.9s34.8 4.2 43.9-10.9L120 256.9V448zM464 64V306.7l-25.4-25.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l80 80c12.5 12.5 32.8 12.5 45.3 0l80-80c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L528 306.7V64c0-17.7-14.3-32-32-32s-32 14.3-32 32z\"],voicemail:[640,512,[],\"f897\",\"M144 320a80 80 0 1 1 0-160 80 80 0 1 1 0 160zm119.8 0c15.3-22.9 24.2-50.4 24.2-80c0-79.5-64.5-144-144-144S0 160.5 0 240s64.5 144 144 144H496c79.5 0 144-64.5 144-144s-64.5-144-144-144s-144 64.5-144 144c0 29.6 8.9 57.1 24.2 80H263.8zM496 160a80 80 0 1 1 0 160 80 80 0 1 1 0-160z\"],fan:[512,512,[],\"f863\",\"M258.6 0c-1.7 0-3.4 .1-5.1 .5C168 17 115.6 102.3 130.5 189.3c2.9 17 8.4 32.9 15.9 47.4L32 224H29.4C13.2 224 0 237.2 0 253.4c0 1.7 .1 3.4 .5 5.1C17 344 102.3 396.4 189.3 381.5c17-2.9 32.9-8.4 47.4-15.9L224 480v2.6c0 16.2 13.2 29.4 29.4 29.4c1.7 0 3.4-.1 5.1-.5C344 495 396.4 409.7 381.5 322.7c-2.9-17-8.4-32.9-15.9-47.4L480 288h2.6c16.2 0 29.4-13.2 29.4-29.4c0-1.7-.1-3.4-.5-5.1C495 168 409.7 115.6 322.7 130.5c-17 2.9-32.9 8.4-47.4 15.9L288 32V29.4C288 13.2 274.8 0 258.6 0zM256 224a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"person-walking-luggage\":[576,512,[],\"e554\",\"M432 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM347.7 200.5c1-.4 1.9-.8 2.9-1.2l-16.9 63.5c-5.6 21.1-.1 43.6 14.7 59.7l70.7 77.1 22 88.1c4.3 17.1 21.7 27.6 38.8 23.3s27.6-21.7 23.3-38.8l-23-92.1c-1.9-7.8-5.8-14.9-11.2-20.8l-49.5-54 19.3-65.5 9.6 23c4.4 10.6 12.5 19.3 22.8 24.5l26.7 13.3c15.8 7.9 35 1.5 42.9-14.3s1.5-35-14.3-42.9L505 232.7l-15.3-36.8C472.5 154.8 432.3 128 387.7 128c-22.8 0-45.3 4.8-66.1 14l-8 3.5c-32.9 14.6-58.1 42.4-69.4 76.5l-2.6 7.8c-5.6 16.8 3.5 34.9 20.2 40.5s34.9-3.5 40.5-20.2l2.6-7.8c5.7-17.1 18.3-30.9 34.7-38.2l8-3.5zm-30 135.1l-25 62.4-59.4 59.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L340.3 441c4.6-4.6 8.2-10.1 10.6-16.1l14.5-36.2-40.7-44.4c-2.5-2.7-4.8-5.6-7-8.6zM256 274.1c-7.7-4.4-17.4-1.8-21.9 5.9l-32 55.4L147.7 304c-15.3-8.8-34.9-3.6-43.7 11.7L40 426.6c-8.8 15.3-3.6 34.9 11.7 43.7l55.4 32c15.3 8.8 34.9 3.6 43.7-11.7l64-110.9c1.5-2.6 2.6-5.2 3.3-8L261.9 296c4.4-7.7 1.8-17.4-5.9-21.9z\"],\"up-down\":[256,512,[8597,11021,\"arrows-alt-v\"],\"f338\",\"M145.6 7.7C141 2.8 134.7 0 128 0s-13 2.8-17.6 7.7l-104 112c-6.5 7-8.2 17.2-4.4 25.9S14.5 160 24 160H80V352H24c-9.5 0-18.2 5.7-22 14.4s-2.1 18.9 4.4 25.9l104 112c4.5 4.9 10.9 7.7 17.6 7.7s13-2.8 17.6-7.7l104-112c6.5-7 8.2-17.2 4.4-25.9s-12.5-14.4-22-14.4H176V160h56c9.5 0 18.2-5.7 22-14.4s2.1-18.9-4.4-25.9l-104-112z\"],\"cloud-moon-rain\":[576,512,[],\"f73c\",\"M481.2 0C417 0 363.5 46.5 353.7 107.6c35.4 17.6 60.2 53.3 62.1 95.1c23.2 11 42 29.7 53.1 52.7c4 .4 8.1 .6 12.3 .6c34.9 0 66.7-13.8 89.9-36.1c5.1-4.9 6.4-12.5 3.2-18.7s-10.1-9.7-17-8.6c-4.9 .8-10 1.3-15.2 1.3c-49 0-88.4-39.3-88.4-87.4c0-32.6 18-61.1 44.9-76.1c6.1-3.4 9.3-10.5 7.8-17.4s-7.3-12-14.3-12.6c-3.6-.3-7.3-.5-10.9-.5zM367.9 383.9c44.2 0 80-35.8 80-80c0-39.3-28.4-72.1-65.8-78.7c1.2-5.6 1.9-11.3 1.9-17.2c0-44.2-35.8-80-80-80c-17 0-32.8 5.3-45.8 14.4C241.3 114.6 210.8 96 176 96c-53 0-96 43-96 96l0 1.3c-45.4 7.6-80 47.1-80 94.6c0 53 43 96 96 96H367.9zM85.4 420.1c-11-7.4-25.9-4.4-33.3 6.7l-32 48c-7.4 11-4.4 25.9 6.7 33.3s25.9 4.4 33.3-6.7l32-48c7.4-11 4.4-25.9-6.7-33.3zm96 0c-11-7.4-25.9-4.4-33.3 6.7l-32 48c-7.4 11-4.4 25.9 6.7 33.3s25.9 4.4 33.3-6.7l32-48c7.4-11 4.4-25.9-6.7-33.3zm96 0c-11-7.4-25.9-4.4-33.3 6.7l-32 48c-7.4 11-4.4 25.9 6.7 33.3s25.9 4.4 33.3-6.7l32-48c7.4-11 4.4-25.9-6.7-33.3zm96 0c-11-7.4-25.9-4.4-33.3 6.7l-32 48c-7.4 11-4.4 25.9 6.7 33.3s25.9 4.4 33.3-6.7l32-48c7.4-11 4.4-25.9-6.7-33.3z\"],calendar:[448,512,[128197,128198],\"f133\",\"M96 32V64H48C21.5 64 0 85.5 0 112v48H448V112c0-26.5-21.5-48-48-48H352V32c0-17.7-14.3-32-32-32s-32 14.3-32 32V64H160V32c0-17.7-14.3-32-32-32S96 14.3 96 32zM448 192H0V464c0 26.5 21.5 48 48 48H400c26.5 0 48-21.5 48-48V192z\"],trailer:[640,512,[],\"e041\",\"M48 32C21.5 32 0 53.5 0 80V336c0 26.5 21.5 48 48 48H65.1c7.8-54.3 54.4-96 110.9-96s103.1 41.7 110.9 96H488h8H608c17.7 0 32-14.3 32-32s-14.3-32-32-32H544V80c0-26.5-21.5-48-48-48H48zM80 96c8.8 0 16 7.2 16 16l0 131.2c-11.4 5.9-22.2 12.9-32 21V112c0-8.8 7.2-16 16-16zm96 128c-5.4 0-10.7 .2-16 .7L160 112c0-8.8 7.2-16 16-16s16 7.2 16 16l0 112.7c-5.3-.5-10.6-.7-16-.7zm80 19.2L256 112c0-8.8 7.2-16 16-16s16 7.2 16 16l0 152.2c-9.8-8.1-20.6-15.2-32-21zM368 96c8.8 0 16 7.2 16 16l0 192c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-192c0-8.8 7.2-16 16-16zm112 16l0 192c0 8.8-7.2 16-16 16s-16-7.2-16-16l0-192c0-8.8 7.2-16 16-16s16 7.2 16 16zM176 480a80 80 0 1 0 0-160 80 80 0 1 0 0 160zm0-112a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],bahai:[576,512,[\"haykal\"],\"f666\",\"M288 0c14.5 0 27.2 9.7 30.9 23.8l23.9 89.6 75.9-53.3c11.9-8.3 27.8-7.6 39 1.7s14.6 24.9 8.4 38.1l-39.3 84 92.4 8c14.4 1.2 26.2 12 28.8 26.3s-4.9 28.5-18 34.6l-84.1 39.1 65.7 65.5c10.3 10.2 12.4 26.1 5.1 38.7s-22 18.7-36 14.9L391 386.8l8.2 92.4c1.3 14.4-7.3 27.9-20.9 32.9s-28.9 .1-37.2-11.7l-53.1-76-53.1 76c-8.3 11.9-23.6 16.7-37.2 11.7s-22.2-18.5-20.9-32.9l8.2-92.4L95.4 410.9c-14 3.8-28.8-2.3-36-14.9s-5.2-28.4 5.1-38.7l65.7-65.5L46 252.7c-13.1-6.1-20.5-20.3-18-34.6s14.3-25.1 28.8-26.3l92.4-8-39.3-84c-6.1-13.1-2.7-28.8 8.4-38.1s27.1-10 39-1.7l75.9 53.3 23.9-89.6C260.8 9.7 273.5 0 288 0zm0 156.2l-4.8 18c-2.7 10.1-10.2 18.2-20 21.8s-20.8 2.1-29.3-3.9l-15.2-10.7 7.9 16.8c4.4 9.5 4 20.5-1.3 29.6s-14.5 15-25 15.9l-18.5 1.6 16.8 7.8c9.5 4.4 16.2 13.2 18 23.5s-1.5 20.8-8.9 28.2l-13.2 13.1 17.9-4.8c10.1-2.7 20.9-.3 28.9 6.4s12.2 16.9 11.3 27.3l-1.6 18.5 10.6-15.2c6-8.6 15.8-13.7 26.2-13.7s20.2 5.1 26.2 13.7l10.6 15.2-1.6-18.5c-.9-10.4 3.3-20.6 11.3-27.3s18.8-9.1 28.9-6.4l17.9 4.8-13.2-13.1c-7.4-7.4-10.7-17.9-8.9-28.2s8.5-19.1 18-23.5l16.8-7.8-18.5-1.6c-10.4-.9-19.7-6.8-25-15.9s-5.7-20.1-1.3-29.6l7.9-16.8-15.2 10.7c-8.6 6-19.5 7.5-29.3 3.9s-17.3-11.7-20-21.8l-4.8-18z\"],\"sd-card\":[384,512,[],\"f7c2\",\"M320 0H141.3C124.3 0 108 6.7 96 18.7L18.7 96C6.7 108 0 124.3 0 141.3V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64zM160 88v48c0 13.3-10.7 24-24 24s-24-10.7-24-24V88c0-13.3 10.7-24 24-24s24 10.7 24 24zm80 0v48c0 13.3-10.7 24-24 24s-24-10.7-24-24V88c0-13.3 10.7-24 24-24s24 10.7 24 24zm80 0v48c0 13.3-10.7 24-24 24s-24-10.7-24-24V88c0-13.3 10.7-24 24-24s24 10.7 24 24z\"],dragon:[640,512,[128009],\"f6d5\",\"M352 124.5l-51.9-13c-6.5-1.6-11.3-7.1-12-13.8s2.8-13.1 8.7-16.1l40.8-20.4L294.4 28.8c-5.5-4.1-7.8-11.3-5.6-17.9S297.1 0 304 0H416h32 16c30.2 0 58.7 14.2 76.8 38.4l57.6 76.8c6.2 8.3 9.6 18.4 9.6 28.8c0 26.5-21.5 48-48 48H538.5c-17 0-33.3-6.7-45.3-18.7L480 160H448v21.5c0 24.8 12.8 47.9 33.8 61.1l106.6 66.6c32.1 20.1 51.6 55.2 51.6 93.1C640 462.9 590.9 512 530.2 512H496 432 32.3c-3.3 0-6.6-.4-9.6-1.4C13.5 507.8 6 501 2.4 492.1C1 488.7 .2 485.2 0 481.4c-.2-3.7 .3-7.3 1.3-10.7c2.8-9.2 9.6-16.7 18.6-20.4c3-1.2 6.2-2 9.5-2.2L433.3 412c8.3-.7 14.7-7.7 14.7-16.1c0-4.3-1.7-8.4-4.7-11.4l-44.4-44.4c-30-30-46.9-70.7-46.9-113.1V181.5v-57zM512 72.3c0-.1 0-.2 0-.3s0-.2 0-.3v.6zm-1.3 7.4L464.3 68.1c-.2 1.3-.3 2.6-.3 3.9c0 13.3 10.7 24 24 24c10.6 0 19.5-6.8 22.7-16.3zM130.9 116.5c16.3-14.5 40.4-16.2 58.5-4.1l130.6 87V227c0 32.8 8.4 64.8 24 93H112c-6.7 0-12.7-4.2-15-10.4s-.5-13.3 4.6-17.7L171 232.3 18.4 255.8c-7 1.1-13.9-2.6-16.9-9s-1.5-14.1 3.8-18.8L130.9 116.5z\"],\"shoe-prints\":[640,512,[],\"f54b\",\"M416 0C352.3 0 256 32 256 32V160c48 0 76 16 104 32s56 32 104 32c56.4 0 176-16 176-96S512 0 416 0zM128 96c0 35.3 28.7 64 64 64h32V32H192c-35.3 0-64 28.7-64 64zM288 512c96 0 224-48 224-128s-119.6-96-176-96c-48 0-76 16-104 32s-56 32-104 32V480s96.3 32 160 32zM0 416c0 35.3 28.7 64 64 64H96V352H64c-35.3 0-64 28.7-64 64z\"],\"circle-plus\":[512,512,[\"plus-circle\"],\"f055\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM232 344V280H168c-13.3 0-24-10.7-24-24s10.7-24 24-24h64V168c0-13.3 10.7-24 24-24s24 10.7 24 24v64h64c13.3 0 24 10.7 24 24s-10.7 24-24 24H280v64c0 13.3-10.7 24-24 24s-24-10.7-24-24z\"],\"face-grin-tongue-wink\":[512,512,[128540,\"grin-tongue-wink\"],\"f58b\",\"M174.5 498.8C73.1 464.7 0 368.9 0 256C0 114.6 114.6 0 256 0S512 114.6 512 256c0 112.9-73.1 208.7-174.5 242.8C346.7 484 352 466.6 352 448V401.1c24.3-17.5 43.6-41.6 55.4-69.6c5-11.8-7-22.5-19.3-18.7c-39.7 12.2-84.5 19-131.8 19s-92.1-6.8-131.8-19c-12.3-3.8-24.3 6.9-19.3 18.7c11.7 27.8 30.8 51.7 54.8 69.2V448c0 18.6 5.3 36 14.5 50.8zm20.7-265.2c5.3 7.1 15.3 8.5 22.4 3.2s8.5-15.3 3.2-22.4c-30.4-40.5-91.2-40.5-121.6 0c-5.3 7.1-3.9 17.1 3.2 22.4s17.1 3.9 22.4-3.2c17.6-23.5 52.8-23.5 70.4 0zM336 272a64 64 0 1 0 0-128 64 64 0 1 0 0 128zM320 402.6V448c0 35.3-28.7 64-64 64s-64-28.7-64-64V402.6c0-14.7 11.9-26.6 26.6-26.6h2c11.3 0 21.1 7.9 23.6 18.9c2.8 12.6 20.8 12.6 23.6 0c2.5-11.1 12.3-18.9 23.6-18.9h2c14.7 0 26.6 11.9 26.6 26.6zM336 184a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],\"hand-holding\":[576,512,[],\"f4bd\",\"M559.7 392.2c17.8-13.1 21.6-38.1 8.5-55.9s-38.1-21.6-55.9-8.5L392.6 416H272c-8.8 0-16-7.2-16-16s7.2-16 16-16h16 64c17.7 0 32-14.3 32-32s-14.3-32-32-32H288 272 193.7c-29.1 0-57.3 9.9-80 28L68.8 384H32c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H192 352.5c29 0 57.3-9.3 80.7-26.5l126.6-93.3zm-366.1-8.3a.5 .5 0 1 1 -.9 .2 .5 .5 0 1 1 .9-.2z\"],\"plug-circle-exclamation\":[576,512,[],\"e55d\",\"M96 0C78.3 0 64 14.3 64 32v96h64V32c0-17.7-14.3-32-32-32zM288 0c-17.7 0-32 14.3-32 32v96h64V32c0-17.7-14.3-32-32-32zM32 160c-17.7 0-32 14.3-32 32s14.3 32 32 32v32c0 77.4 55 142 128 156.8V480c0 17.7 14.3 32 32 32s32-14.3 32-32V412.8c12.3-2.5 24.1-6.4 35.1-11.5c-2.1-10.8-3.1-21.9-3.1-33.3c0-80.3 53.8-148 127.3-169.2c.5-2.2 .7-4.5 .7-6.8c0-17.7-14.3-32-32-32H32zM432 512a144 144 0 1 0 0-288 144 144 0 1 0 0 288zm0-96a24 24 0 1 1 0 48 24 24 0 1 1 0-48zm0-144c8.8 0 16 7.2 16 16v80c0 8.8-7.2 16-16 16s-16-7.2-16-16V288c0-8.8 7.2-16 16-16z\"],\"link-slash\":[640,512,[\"chain-broken\",\"chain-slash\",\"unlink\"],\"f127\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L489.3 358.2l90.5-90.5c56.5-56.5 56.5-148 0-204.5c-50-50-128.8-56.5-186.3-15.4l-1.6 1.1c-14.4 10.3-17.7 30.3-7.4 44.6s30.3 17.7 44.6 7.4l1.6-1.1c32.1-22.9 76-19.3 103.8 8.6c31.5 31.5 31.5 82.5 0 114l-96 96-31.9-25C430.9 239.6 420.1 175.1 377 132c-52.2-52.3-134.5-56.2-191.3-11.7L38.8 5.1zM239 162c30.1-14.9 67.7-9.9 92.8 15.3c20 20 27.5 48.3 21.7 74.5L239 162zM116.6 187.9L60.2 244.3c-56.5 56.5-56.5 148 0 204.5c50 50 128.8 56.5 186.3 15.4l1.6-1.1c14.4-10.3 17.7-30.3 7.4-44.6s-30.3-17.7-44.6-7.4l-1.6 1.1c-32.1 22.9-76 19.3-103.8-8.6C74 372 74 321 105.5 289.5l61.8-61.8-50.6-39.9zM220.9 270c-2.1 39.8 12.2 80.1 42.2 110c38.9 38.9 94.4 51 143.6 36.3L220.9 270z\"],clone:[512,512,[],\"f24d\",\"M288 448H64V224h64V160H64c-35.3 0-64 28.7-64 64V448c0 35.3 28.7 64 64 64H288c35.3 0 64-28.7 64-64V384H288v64zm-64-96H448c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H224c-35.3 0-64 28.7-64 64V288c0 35.3 28.7 64 64 64z\"],\"person-walking-arrow-loop-left\":[640,512,[],\"e551\",\"M208 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM123.7 200.5c1-.4 1.9-.8 2.9-1.2l-16.9 63.5c-5.6 21.1-.1 43.6 14.7 59.7l70.7 77.1 22 88.1c4.3 17.1 21.7 27.6 38.8 23.3s27.6-21.7 23.3-38.8l-23-92.1c-1.9-7.8-5.8-14.9-11.2-20.8l-49.5-54 19.3-65.5 9.6 23c4.4 10.6 12.5 19.3 22.8 24.5l26.7 13.3c15.8 7.9 35 1.5 42.9-14.3s1.5-35-14.3-42.9L281 232.7l-15.3-36.8C248.5 154.8 208.3 128 163.7 128c-22.8 0-45.3 4.8-66.1 14l-8 3.5c-32.9 14.6-58.1 42.4-69.4 76.5l-2.6 7.8c-5.6 16.8 3.5 34.9 20.2 40.5s34.9-3.5 40.5-20.2l2.6-7.8c5.7-17.1 18.3-30.9 34.7-38.2l8-3.5zm-30 135.1L68.7 398 9.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L116.3 441c4.6-4.6 8.2-10.1 10.6-16.1l14.5-36.2-40.7-44.4c-2.5-2.7-4.8-5.6-7-8.6zm347.7 119c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L461.3 384H480c88.4 0 160-71.6 160-160s-71.6-160-160-160L352 64c-17.7 0-32 14.3-32 32s14.3 32 32 32l128 0c53 0 96 43 96 96s-43 96-96 96H461.3l25.4-25.4c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-80 80c-12.5 12.5-12.5 32.8 0 45.3l80 80z\"],\"arrow-up-z-a\":[576,512,[\"sort-alpha-up-alt\"],\"f882\",\"M183.6 42.4C177.5 35.8 169 32 160 32s-17.5 3.8-23.6 10.4l-88 96c-11.9 13-11.1 33.3 2 45.2s33.3 11.1 45.2-2L128 146.3V448c0 17.7 14.3 32 32 32s32-14.3 32-32V146.3l32.4 35.4c11.9 13 32.2 13.9 45.2 2s13.9-32.2 2-45.2l-88-96zM320 64c0 17.7 14.3 32 32 32h50.7l-73.4 73.4c-9.2 9.2-11.9 22.9-6.9 34.9s16.6 19.8 29.6 19.8H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H429.3l73.4-73.4c9.2-9.2 11.9-22.9 6.9-34.9s-16.6-19.8-29.6-19.8H352c-17.7 0-32 14.3-32 32zm96 192c-12.1 0-23.2 6.8-28.6 17.7l-64 128-16 32c-7.9 15.8-1.5 35 14.3 42.9s35 1.5 42.9-14.3l7.2-14.3h88.4l7.2 14.3c7.9 15.8 27.1 22.2 42.9 14.3s22.2-27.1 14.3-42.9l-16-32-64-128C439.2 262.8 428.1 256 416 256zM395.8 400L416 359.6 436.2 400H395.8z\"],\"fire-flame-curved\":[384,512,[\"fire-alt\"],\"f7e4\",\"M153.6 29.9l16-21.3C173.6 3.2 180 0 186.7 0C198.4 0 208 9.6 208 21.3V43.5c0 13.1 5.4 25.7 14.9 34.7L307.6 159C356.4 205.6 384 270.2 384 337.7C384 434 306 512 209.7 512H192C86 512 0 426 0 320v-3.8c0-48.8 19.4-95.6 53.9-130.1l3.5-3.5c4.2-4.2 10-6.6 16-6.6C85.9 176 96 186.1 96 198.6V288c0 35.3 28.7 64 64 64s64-28.7 64-64v-3.9c0-18-7.2-35.3-19.9-48l-38.6-38.6c-24-24-37.5-56.7-37.5-90.7c0-27.7 9-54.8 25.6-76.9z\"],tornado:[448,512,[127786],\"f76f\",\"M0 32V45.6C0 62.7 1.7 79.6 5 96H357.8c3.2-6.9 7.5-13.3 13-18.8l38.6-38.6c4.2-4.2 6.6-10 6.6-16C416 10.1 405.9 0 393.4 0H32C14.3 0 0 14.3 0 32zm352.2 96H13.6c12.2 35.9 32.3 68.7 58.8 96H412l-47.2-62.9c-7.3-9.7-11.6-21.2-12.6-33.1zm-226 138.2l116.4 68.5c8.2 4.8 15.8 10.7 22.5 17.3H445c2-9.8 3-19.9 3-30.1c0-23-5.3-45.5-15.3-65.9H110.2c5.2 3.6 10.5 7 16 10.2zM288 384c10.3 21.4 13.8 45.5 9.9 69l-5.9 35.7c-2 12.2 7.4 23.4 19.8 23.4c5.3 0 10.4-2.1 14.2-5.9l78.2-78.2c12.8-12.8 23.1-27.7 30.4-43.9H288z\"],\"file-circle-plus\":[576,512,[58606],\"e494\",\"M0 64C0 28.7 28.7 0 64 0H224V128c0 17.7 14.3 32 32 32H384v38.6C310.1 219.5 256 287.4 256 368c0 59.1 29.1 111.3 73.7 143.3c-3.2 .5-6.4 .7-9.7 .7H64c-35.3 0-64-28.7-64-64V64zm384 64H256V0L384 128zm48 96a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm16 80c0-8.8-7.2-16-16-16s-16 7.2-16 16v48H368c-8.8 0-16 7.2-16 16s7.2 16 16 16h48v48c0 8.8 7.2 16 16 16s16-7.2 16-16V384h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H448V304z\"],\"book-quran\":[448,512,[\"quran\"],\"f687\",\"M352 0c53 0 96 43 96 96V416c0 53-43 96-96 96H64 32c-17.7 0-32-14.3-32-32s14.3-32 32-32V384c-17.7 0-32-14.3-32-32V32C0 14.3 14.3 0 32 0H64 352zm0 384H96v64H352c17.7 0 32-14.3 32-32s-14.3-32-32-32zM274.1 150.2l-8.9 21.4-23.1 1.9c-5.7 .5-8 7.5-3.7 11.2L256 199.8l-5.4 22.6c-1.3 5.5 4.7 9.9 9.6 6.9L280 217.2l19.8 12.1c4.9 3 10.9-1.4 9.6-6.9L304 199.8l17.6-15.1c4.3-3.7 2-10.8-3.7-11.2l-23.1-1.9-8.9-21.4c-2.2-5.3-9.6-5.3-11.8 0zM96 192c0 70.7 57.3 128 128 128c25.6 0 49.5-7.5 69.5-20.5c3.2-2.1 4.5-6.2 3.1-9.7s-5.2-5.6-9-4.8c-6.1 1.2-12.5 1.9-19 1.9c-52.4 0-94.9-42.5-94.9-94.9s42.5-94.9 94.9-94.9c6.5 0 12.8 .7 19 1.9c3.8 .8 7.5-1.3 9-4.8s.2-7.6-3.1-9.7C273.5 71.5 249.6 64 224 64C153.3 64 96 121.3 96 192z\"],anchor:[576,512,[9875],\"f13d\",\"M320 96a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm21.1 80C367 158.8 384 129.4 384 96c0-53-43-96-96-96s-96 43-96 96c0 33.4 17 62.8 42.9 80H224c-17.7 0-32 14.3-32 32s14.3 32 32 32h32V448H208c-53 0-96-43-96-96v-6.1l7 7c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9L97 263c-9.4-9.4-24.6-9.4-33.9 0L7 319c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l7-7V352c0 88.4 71.6 160 160 160h80 80c88.4 0 160-71.6 160-160v-6.1l7 7c9.4 9.4 24.6 9.4 33.9 0s9.4-24.6 0-33.9l-56-56c-9.4-9.4-24.6-9.4-33.9 0l-56 56c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l7-7V352c0 53-43 96-96 96H320V240h32c17.7 0 32-14.3 32-32s-14.3-32-32-32H341.1z\"],\"border-all\":[448,512,[],\"f84c\",\"M384 96V224H256V96H384zm0 192V416H256V288H384zM192 224H64V96H192V224zM64 288H192V416H64V288zM64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64z\"],\"face-angry\":[512,512,[128544,\"angry\"],\"f556\",\"M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM338.7 395.9c6.6-5.9 7.1-16 1.2-22.6C323.8 355.4 295.7 336 256 336s-67.8 19.4-83.9 37.3c-5.9 6.6-5.4 16.7 1.2 22.6s16.7 5.4 22.6-1.2c11.7-13 31.6-26.7 60.1-26.7s48.4 13.7 60.1 26.7c5.9 6.6 16 7.1 22.6 1.2zM176.4 272c17.7 0 32-14.3 32-32c0-1.5-.1-3-.3-4.4l10.9 3.6c8.4 2.8 17.4-1.7 20.2-10.1s-1.7-17.4-10.1-20.2l-96-32c-8.4-2.8-17.4 1.7-20.2 10.1s1.7 17.4 10.1 20.2l30.7 10.2c-5.8 5.8-9.3 13.8-9.3 22.6c0 17.7 14.3 32 32 32zm192-32c0-8.9-3.6-17-9.5-22.8l30.2-10.1c8.4-2.8 12.9-11.9 10.1-20.2s-11.9-12.9-20.2-10.1l-96 32c-8.4 2.8-12.9 11.9-10.1 20.2s11.9 12.9 20.2 10.1l11.7-3.9c-.2 1.5-.3 3.1-.3 4.7c0 17.7 14.3 32 32 32s32-14.3 32-32z\"],\"cookie-bite\":[512,512,[],\"f564\",\"M257.5 27.6c-.8-5.4-4.9-9.8-10.3-10.6v0c-22.1-3.1-44.6 .9-64.4 11.4l-74 39.5C89.1 78.4 73.2 94.9 63.4 115L26.7 190.6c-9.8 20.1-13 42.9-9.1 64.9l14.5 82.8c3.9 22.1 14.6 42.3 30.7 57.9l60.3 58.4c16.1 15.6 36.6 25.6 58.7 28.7l83 11.7c22.1 3.1 44.6-.9 64.4-11.4l74-39.5c19.7-10.5 35.6-27 45.4-47.2l36.7-75.5c9.8-20.1 13-42.9 9.1-64.9v0c-.9-5.3-5.3-9.3-10.6-10.1c-51.5-8.2-92.8-47.1-104.5-97.4c-1.8-7.6-8-13.4-15.7-14.6c-54.6-8.7-97.7-52-106.2-106.8zM208 144a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM144 336a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm224-64a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"arrow-trend-down\":[576,512,[],\"e097\",\"M384 352c-17.7 0-32 14.3-32 32s14.3 32 32 32H544c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32s-32 14.3-32 32v82.7L342.6 137.4c-12.5-12.5-32.8-12.5-45.3 0L192 242.7 54.6 105.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0L320 205.3 466.7 352H384z\"],rss:[448,512,[\"feed\"],\"f09e\",\"M0 64C0 46.3 14.3 32 32 32c229.8 0 416 186.2 416 416c0 17.7-14.3 32-32 32s-32-14.3-32-32C384 253.6 226.4 96 32 96C14.3 96 0 81.7 0 64zM0 416a64 64 0 1 1 128 0A64 64 0 1 1 0 416zM32 160c159.1 0 288 128.9 288 288c0 17.7-14.3 32-32 32s-32-14.3-32-32c0-123.7-100.3-224-224-224c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],\"draw-polygon\":[448,512,[],\"f5ee\",\"M96 151.4V360.6c9.7 5.6 17.8 13.7 23.4 23.4H328.6c0-.1 .1-.2 .1-.3l-4.5-7.9-32-56 0 0c-1.4 .1-2.8 .1-4.2 .1c-35.3 0-64-28.7-64-64s28.7-64 64-64c1.4 0 2.8 0 4.2 .1l0 0 32-56 4.5-7.9-.1-.3H119.4c-5.6 9.7-13.7 17.8-23.4 23.4zM384.3 352c35.2 .2 63.7 28.7 63.7 64c0 35.3-28.7 64-64 64c-23.7 0-44.4-12.9-55.4-32H119.4c-11.1 19.1-31.7 32-55.4 32c-35.3 0-64-28.7-64-64c0-23.7 12.9-44.4 32-55.4V151.4C12.9 140.4 0 119.7 0 96C0 60.7 28.7 32 64 32c23.7 0 44.4 12.9 55.4 32H328.6c11.1-19.1 31.7-32 55.4-32c35.3 0 64 28.7 64 64c0 35.3-28.5 63.8-63.7 64l-4.5 7.9-32 56-2.3 4c4.2 8.5 6.5 18 6.5 28.1s-2.3 19.6-6.5 28.1l2.3 4 32 56 4.5 7.9z\"],\"scale-balanced\":[640,512,[9878,\"balance-scale\"],\"f24e\",\"M384 32H512c17.7 0 32 14.3 32 32s-14.3 32-32 32H398.4c-5.2 25.8-22.9 47.1-46.4 57.3V448H512c17.7 0 32 14.3 32 32s-14.3 32-32 32H320 128c-17.7 0-32-14.3-32-32s14.3-32 32-32H288V153.3c-23.5-10.3-41.2-31.6-46.4-57.3H128c-17.7 0-32-14.3-32-32s14.3-32 32-32H256c14.6-19.4 37.8-32 64-32s49.4 12.6 64 32zm55.6 288H584.4L512 195.8 439.6 320zM512 416c-62.9 0-115.2-34-126-78.9c-2.6-11 1-22.3 6.7-32.1l95.2-163.2c5-8.6 14.2-13.8 24.1-13.8s19.1 5.3 24.1 13.8l95.2 163.2c5.7 9.8 9.3 21.1 6.7 32.1C627.2 382 574.9 416 512 416zM126.8 195.8L54.4 320H199.3L126.8 195.8zM.9 337.1c-2.6-11 1-22.3 6.7-32.1l95.2-163.2c5-8.6 14.2-13.8 24.1-13.8s19.1 5.3 24.1 13.8l95.2 163.2c5.7 9.8 9.3 21.1 6.7 32.1C242 382 189.7 416 126.8 416S11.7 382 .9 337.1z\"],\"gauge-simple-high\":[512,512,[61668,\"tachometer\",\"tachometer-fast\"],\"f62a\",\"M0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm320 96c0-15.9-5.8-30.4-15.3-41.6l76.6-147.4c6.1-11.8 1.5-26.3-10.2-32.4s-26.2-1.5-32.4 10.2L262.1 288.3c-2-.2-4-.3-6.1-.3c-35.3 0-64 28.7-64 64s28.7 64 64 64s64-28.7 64-64z\"],shower:[512,512,[128703],\"f2cc\",\"M64 131.9C64 112.1 80.1 96 99.9 96c9.5 0 18.6 3.8 25.4 10.5l16.2 16.2c-21 38.9-17.4 87.5 10.9 123L151 247c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0L345 121c9.4-9.4 9.4-24.6 0-33.9s-24.6-9.4-33.9 0l-1.3 1.3c-35.5-28.3-84.2-31.9-123-10.9L170.5 61.3C151.8 42.5 126.4 32 99.9 32C44.7 32 0 76.7 0 131.9V448c0 17.7 14.3 32 32 32s32-14.3 32-32V131.9zM256 352a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm64 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm0-128a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm64 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm0-128a32 32 0 1 0 0-64 32 32 0 1 0 0 64zm64 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm32-32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],desktop:[576,512,[128421,61704,\"desktop-alt\"],\"f390\",\"M64 0C28.7 0 0 28.7 0 64V352c0 35.3 28.7 64 64 64H240l-10.7 32H160c-17.7 0-32 14.3-32 32s14.3 32 32 32H416c17.7 0 32-14.3 32-32s-14.3-32-32-32H346.7L336 416H512c35.3 0 64-28.7 64-64V64c0-35.3-28.7-64-64-64H64zM512 64V288H64V64H512z\"],m:[448,512,[109],\"4d\",\"M22.7 33.4c13.5-4.1 28.1 1.1 35.9 12.9L224 294.3 389.4 46.2c7.8-11.7 22.4-17 35.9-12.9S448 49.9 448 64V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V169.7L250.6 369.8c-5.9 8.9-15.9 14.2-26.6 14.2s-20.7-5.3-26.6-14.2L64 169.7V448c0 17.7-14.3 32-32 32s-32-14.3-32-32V64C0 49.9 9.2 37.5 22.7 33.4z\"],\"table-list\":[512,512,[\"th-list\"],\"f00b\",\"M0 96C0 60.7 28.7 32 64 32H448c35.3 0 64 28.7 64 64V416c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V96zm64 0v64h64V96H64zm384 0H192v64H448V96zM64 224v64h64V224H64zm384 0H192v64H448V224zM64 352v64h64V352H64zm384 0H192v64H448V352z\"],\"comment-sms\":[512,512,[\"sms\"],\"f7cd\",\"M256 448c141.4 0 256-93.1 256-208S397.4 32 256 32S0 125.1 0 240c0 45.1 17.7 86.8 47.7 120.9c-1.9 24.5-11.4 46.3-21.4 62.9c-5.5 9.2-11.1 16.6-15.2 21.6c-2.1 2.5-3.7 4.4-4.9 5.7c-.6 .6-1 1.1-1.3 1.4l-.3 .3 0 0 0 0 0 0 0 0c-4.6 4.6-5.9 11.4-3.4 17.4c2.5 6 8.3 9.9 14.8 9.9c28.7 0 57.6-8.9 81.6-19.3c22.9-10 42.4-21.9 54.3-30.6c31.8 11.5 67 17.9 104.1 17.9zM96 212.8c0-20.3 16.5-36.8 36.8-36.8H152c8.8 0 16 7.2 16 16s-7.2 16-16 16H132.8c-2.7 0-4.8 2.2-4.8 4.8c0 1.6 .8 3.1 2.2 4l29.4 19.6c10.3 6.8 16.4 18.3 16.4 30.7c0 20.3-16.5 36.8-36.8 36.8H112c-8.8 0-16-7.2-16-16s7.2-16 16-16h27.2c2.7 0 4.8-2.2 4.8-4.8c0-1.6-.8-3.1-2.2-4l-29.4-19.6C102.2 236.7 96 225.2 96 212.8zM372.8 176H392c8.8 0 16 7.2 16 16s-7.2 16-16 16H372.8c-2.7 0-4.8 2.2-4.8 4.8c0 1.6 .8 3.1 2.2 4l29.4 19.6c10.2 6.8 16.4 18.3 16.4 30.7c0 20.3-16.5 36.8-36.8 36.8H352c-8.8 0-16-7.2-16-16s7.2-16 16-16h27.2c2.7 0 4.8-2.2 4.8-4.8c0-1.6-.8-3.1-2.2-4l-29.4-19.6c-10.2-6.8-16.4-18.3-16.4-30.7c0-20.3 16.5-36.8 36.8-36.8zm-152 6.4L256 229.3l35.2-46.9c4.1-5.5 11.3-7.8 17.9-5.6s10.9 8.3 10.9 15.2v96c0 8.8-7.2 16-16 16s-16-7.2-16-16V240l-19.2 25.6c-3 4-7.8 6.4-12.8 6.4s-9.8-2.4-12.8-6.4L224 240v48c0 8.8-7.2 16-16 16s-16-7.2-16-16V192c0-6.9 4.4-13 10.9-15.2s13.7 .1 17.9 5.6z\"],book:[448,512,[128212],\"f02d\",\"M96 0C43 0 0 43 0 96V416c0 53 43 96 96 96H384h32c17.7 0 32-14.3 32-32s-14.3-32-32-32V384c17.7 0 32-14.3 32-32V32c0-17.7-14.3-32-32-32H384 96zm0 384H352v64H96c-17.7 0-32-14.3-32-32s14.3-32 32-32zm32-240c0-8.8 7.2-16 16-16H336c8.8 0 16 7.2 16 16s-7.2 16-16 16H144c-8.8 0-16-7.2-16-16zm16 48H336c8.8 0 16 7.2 16 16s-7.2 16-16 16H144c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],\"user-plus\":[640,512,[],\"f234\",\"M96 128a128 128 0 1 1 256 0A128 128 0 1 1 96 128zM0 482.3C0 383.8 79.8 304 178.3 304h91.4C368.2 304 448 383.8 448 482.3c0 16.4-13.3 29.7-29.7 29.7H29.7C13.3 512 0 498.7 0 482.3zM504 312V248H440c-13.3 0-24-10.7-24-24s10.7-24 24-24h64V136c0-13.3 10.7-24 24-24s24 10.7 24 24v64h64c13.3 0 24 10.7 24 24s-10.7 24-24 24H552v64c0 13.3-10.7 24-24 24s-24-10.7-24-24z\"],check:[448,512,[10003,10004],\"f00c\",\"M438.6 105.4c12.5 12.5 12.5 32.8 0 45.3l-256 256c-12.5 12.5-32.8 12.5-45.3 0l-128-128c-12.5-12.5-12.5-32.8 0-45.3s32.8-12.5 45.3 0L160 338.7 393.4 105.4c12.5-12.5 32.8-12.5 45.3 0z\"],\"battery-three-quarters\":[576,512,[\"battery-4\"],\"f241\",\"M464 160c8.8 0 16 7.2 16 16V336c0 8.8-7.2 16-16 16H80c-8.8 0-16-7.2-16-16V176c0-8.8 7.2-16 16-16H464zM80 96C35.8 96 0 131.8 0 176V336c0 44.2 35.8 80 80 80H464c44.2 0 80-35.8 80-80V320c17.7 0 32-14.3 32-32V224c0-17.7-14.3-32-32-32V176c0-44.2-35.8-80-80-80H80zm272 96H96V320H352V192z\"],\"house-circle-check\":[640,512,[],\"e509\",\"M320.7 352c8.1-89.7 83.5-160 175.3-160c8.9 0 17.6 .7 26.1 1.9L309.5 7c-6-5-14-7-21-7s-15 1-22 8L10 231.5c-7 7-10 15-10 24c0 18 14 32.1 32 32.1h32v69.7c-.1 .9-.1 1.8-.1 2.8V472c0 22.1 17.9 40 40 40h16c1.2 0 2.4-.1 3.6-.2c1.5 .1 3 .2 4.5 .2H160h24c22.1 0 40-17.9 40-40V448 384c0-17.7 14.3-32 32-32h64l.7 0zM640 368a144 144 0 1 0 -288 0 144 144 0 1 0 288 0zm-76.7-43.3c6.2 6.2 6.2 16.4 0 22.6l-72 72c-6.2 6.2-16.4 6.2-22.6 0l-40-40c-6.2-6.2-6.2-16.4 0-22.6s16.4-6.2 22.6 0L480 385.4l60.7-60.7c6.2-6.2 16.4-6.2 22.6 0z\"],\"angle-left\":[320,512,[8249],\"f104\",\"M41.4 233.4c-12.5 12.5-12.5 32.8 0 45.3l160 160c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3L109.3 256 246.6 118.6c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-160 160z\"],\"diagram-successor\":[512,512,[],\"e47a\",\"M512 416l0-64c0-35.3-28.7-64-64-64L64 288c-35.3 0-64 28.7-64 64l0 64c0 35.3 28.7 64 64 64l384 0c35.3 0 64-28.7 64-64zM64 160l0-64 144 0 16 0 0 64L64 160zm224 0l0-64 80 0c8.8 0 16 7.2 16 16l0 16-38.1 0c-21.4 0-32.1 25.9-17 41L399 239c9.4 9.4 24.6 9.4 33.9 0L503 169c15.1-15.1 4.4-41-17-41L448 128l0-16c0-44.2-35.8-80-80-80L224 32l-16 0L64 32C28.7 32 0 60.7 0 96l0 64c0 35.3 28.7 64 64 64l160 0c35.3 0 64-28.7 64-64z\"],\"truck-arrow-right\":[640,512,[],\"e58b\",\"M0 48C0 21.5 21.5 0 48 0H368c26.5 0 48 21.5 48 48V96h50.7c17 0 33.3 6.7 45.3 18.7L589.3 192c12 12 18.7 28.3 18.7 45.3V256v32 64c17.7 0 32 14.3 32 32s-14.3 32-32 32H576c0 53-43 96-96 96s-96-43-96-96H256c0 53-43 96-96 96s-96-43-96-96H48c-26.5 0-48-21.5-48-48V48zM416 256H544V237.3L466.7 160H416v96zM160 464a48 48 0 1 0 0-96 48 48 0 1 0 0 96zm368-48a48 48 0 1 0 -96 0 48 48 0 1 0 96 0zM257 95c-9.4-9.4-24.6-9.4-33.9 0s-9.4 24.6 0 33.9l39 39H96c-13.3 0-24 10.7-24 24s10.7 24 24 24H262.1l-39 39c-9.4 9.4-9.4 24.6 0 33.9s24.6 9.4 33.9 0l80-80c9.4-9.4 9.4-24.6 0-33.9L257 95z\"],\"arrows-split-up-and-left\":[512,512,[],\"e4bc\",\"M246.6 150.6c-12.5 12.5-32.8 12.5-45.3 0s-12.5-32.8 0-45.3l96-96c12.5-12.5 32.8-12.5 45.3 0l96 96c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L352 109.3V384c0 35.3 28.7 64 64 64h64c17.7 0 32 14.3 32 32s-14.3 32-32 32H416c-70.7 0-128-57.3-128-128c0-35.3-28.7-64-64-64H109.3l41.4 41.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0l-96-96c-12.5-12.5-12.5-32.8 0-45.3l96-96c12.5-12.5 32.8-12.5 45.3 0s12.5 32.8 0 45.3L109.3 256H224c23.3 0 45.2 6.2 64 17.1V109.3l-41.4 41.4z\"],\"hand-fist\":[448,512,[9994,\"fist-raised\"],\"f6de\",\"M192 0c17.7 0 32 14.3 32 32V144H160V32c0-17.7 14.3-32 32-32zM64 64c0-17.7 14.3-32 32-32s32 14.3 32 32v80H64V64zm192 0c0-17.7 14.3-32 32-32s32 14.3 32 32v96c0 17.7-14.3 32-32 32s-32-14.3-32-32V64zm96 64c0-17.7 14.3-32 32-32s32 14.3 32 32v64c0 17.7-14.3 32-32 32s-32-14.3-32-32V128zm-96 88l0-.6c9.4 5.4 20.3 8.6 32 8.6c13.2 0 25.4-4 35.6-10.8c8.7 24.9 32.5 42.8 60.4 42.8c11.7 0 22.6-3.1 32-8.6V256c0 52.3-25.1 98.8-64 128v96c0 17.7-14.3 32-32 32H160c-17.7 0-32-14.3-32-32V401.6c-17.3-7.9-33.2-18.8-46.9-32.5L69.5 357.5C45.5 333.5 32 300.9 32 267V240c0-35.3 28.7-64 64-64h88c22.1 0 40 17.9 40 40s-17.9 40-40 40H128c-8.8 0-16 7.2-16 16s7.2 16 16 16h56c39.8 0 72-32.2 72-72z\"],\"cloud-moon\":[640,512,[],\"f6c3\",\"M495.8 0c5.5 0 10.9 .2 16.3 .7c7 .6 12.8 5.7 14.3 12.5s-1.6 13.9-7.7 17.3c-44.4 25.2-74.4 73-74.4 127.8c0 81 65.5 146.6 146.2 146.6c8.6 0 17-.7 25.1-2.1c6.9-1.2 13.8 2.2 17 8.5s1.9 13.8-3.1 18.7c-34.5 33.6-81.7 54.4-133.6 54.4c-9.3 0-18.4-.7-27.4-1.9c-11.2-22.6-29.8-40.9-52.6-51.7c-2.7-58.5-50.3-105.3-109.2-106.7c-1.7-10.4-2.6-21-2.6-31.8C304 86.1 389.8 0 495.8 0zM447.9 431.9c0 44.2-35.8 80-80 80H96c-53 0-96-43-96-96c0-47.6 34.6-87 80-94.6l0-1.3c0-53 43-96 96-96c34.9 0 65.4 18.6 82.2 46.4c13-9.1 28.8-14.4 45.8-14.4c44.2 0 80 35.8 80 80c0 5.9-.6 11.7-1.9 17.2c37.4 6.7 65.8 39.4 65.8 78.7z\"],briefcase:[512,512,[128188],\"f0b1\",\"M184 48H328c4.4 0 8 3.6 8 8V96H176V56c0-4.4 3.6-8 8-8zm-56 8V96H64C28.7 96 0 124.7 0 160v96H192 320 512V160c0-35.3-28.7-64-64-64H384V56c0-30.9-25.1-56-56-56H184c-30.9 0-56 25.1-56 56zM512 288H320v32c0 17.7-14.3 32-32 32H224c-17.7 0-32-14.3-32-32V288H0V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V288z\"],\"person-falling\":[512,512,[],\"e546\",\"M288 0c17.7 0 32 14.3 32 32l0 9.8c0 54.6-27.9 104.6-72.5 133.6l.2 .3L304.5 256l87.5 0c15.1 0 29.3 7.1 38.4 19.2l43.2 57.6c10.6 14.1 7.7 34.2-6.4 44.8s-34.2 7.7-44.8-6.4L384 320l-96 0h-1.4l92.3 142.6c9.6 14.8 5.4 34.6-9.5 44.3s-34.6 5.4-44.3-9.5L164.5 249.2c-2.9 9.2-4.5 19-4.5 29l0 73.8c0 17.7-14.3 32-32 32s-32-14.3-32-32V278.2c0-65.1 39.6-123.7 100.1-147.9C232.3 115.8 256 80.8 256 41.8l0-9.8c0-17.7 14.3-32 32-32zM112 32a48 48 0 1 1 0 96 48 48 0 1 1 0-96z\"],\"image-portrait\":[384,512,[\"portrait\"],\"f3e0\",\"M384 64c0-35.3-28.7-64-64-64H64C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64l0-384zM128 192a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM80 356.6c0-37.9 30.7-68.6 68.6-68.6h86.9c37.9 0 68.6 30.7 68.6 68.6c0 15.1-12.3 27.4-27.4 27.4H107.4C92.3 384 80 371.7 80 356.6z\"],\"user-tag\":[640,512,[],\"f507\",\"M224 256A128 128 0 1 0 224 0a128 128 0 1 0 0 256zm-45.7 48C79.8 304 0 383.8 0 482.3C0 498.7 13.3 512 29.7 512H418.3c10 0 18.8-4.9 24.2-12.5l-99.2-99.2c-14.9-14.9-23.3-35.1-23.3-56.1v-33c-15.9-4.7-32.8-7.2-50.3-7.2H178.3zM384 224c-17.7 0-32 14.3-32 32v82.7c0 17 6.7 33.3 18.7 45.3L478.1 491.3c18.7 18.7 49.1 18.7 67.9 0l73.4-73.4c18.7-18.7 18.7-49.1 0-67.9L512 242.7c-12-12-28.3-18.7-45.3-18.7H384zm24 80a24 24 0 1 1 48 0 24 24 0 1 1 -48 0z\"],rug:[640,512,[],\"e569\",\"M24 64H56 80V88v88 80 80 88 24H56 24c-13.3 0-24-10.7-24-24s10.7-24 24-24h8V360H24c-13.3 0-24-10.7-24-24s10.7-24 24-24h8V280H24c-13.3 0-24-10.7-24-24s10.7-24 24-24h8V200H24c-13.3 0-24-10.7-24-24s10.7-24 24-24h8V112H24C10.7 112 0 101.3 0 88S10.7 64 24 64zm88 0H528V448H112V64zM640 88c0 13.3-10.7 24-24 24h-8v40h8c13.3 0 24 10.7 24 24s-10.7 24-24 24h-8v32h8c13.3 0 24 10.7 24 24s-10.7 24-24 24h-8v32h8c13.3 0 24 10.7 24 24s-10.7 24-24 24h-8v40h8c13.3 0 24 10.7 24 24s-10.7 24-24 24H584 560V424 336 256 176 88 64h24 32c13.3 0 24 10.7 24 24z\"],\"earth-europe\":[512,512,[\"globe-europe\"],\"f7a2\",\"M266.3 48.3L232.5 73.6c-5.4 4-8.5 10.4-8.5 17.1v9.1c0 6.8 5.5 12.3 12.3 12.3c2.4 0 4.8-.7 6.8-2.1l41.8-27.9c2-1.3 4.4-2.1 6.8-2.1h1c6.2 0 11.3 5.1 11.3 11.3c0 3-1.2 5.9-3.3 8l-19.9 19.9c-5.8 5.8-12.9 10.2-20.7 12.8l-26.5 8.8c-5.8 1.9-9.6 7.3-9.6 13.4c0 3.7-1.5 7.3-4.1 10l-17.9 17.9c-6.4 6.4-9.9 15-9.9 24v4.3c0 16.4 13.6 29.7 29.9 29.7c11 0 21.2-6.2 26.1-16l4-8.1c2.4-4.8 7.4-7.9 12.8-7.9c4.5 0 8.7 2.1 11.4 5.7l16.3 21.7c2.1 2.9 5.5 4.5 9.1 4.5c8.4 0 13.9-8.9 10.1-16.4l-1.1-2.3c-3.5-7 0-15.5 7.5-18l21.2-7.1c7.6-2.5 12.7-9.6 12.7-17.6c0-10.3 8.3-18.6 18.6-18.6H400c8.8 0 16 7.2 16 16s-7.2 16-16 16H379.3c-7.2 0-14.2 2.9-19.3 8l-4.7 4.7c-2.1 2.1-3.3 5-3.3 8c0 6.2 5.1 11.3 11.3 11.3h11.3c6 0 11.8 2.4 16 6.6l6.5 6.5c1.8 1.8 2.8 4.3 2.8 6.8s-1 5-2.8 6.8l-7.5 7.5C386 262 384 266.9 384 272s2 10 5.7 13.7L408 304c10.2 10.2 24.1 16 38.6 16H454c6.5-20.2 10-41.7 10-64c0-111.4-87.6-202.4-197.7-207.7zm172 307.9c-3.7-2.6-8.2-4.1-13-4.1c-6 0-11.8-2.4-16-6.6L396 332c-7.7-7.7-18-12-28.9-12c-9.7 0-19.2-3.5-26.6-9.8L314 287.4c-11.6-9.9-26.4-15.4-41.7-15.4H251.4c-12.6 0-25 3.7-35.5 10.7L188.5 301c-17.8 11.9-28.5 31.9-28.5 53.3v3.2c0 17 6.7 33.3 18.7 45.3l16 16c8.5 8.5 20 13.3 32 13.3H248c13.3 0 24 10.7 24 24c0 2.5 .4 5 1.1 7.3c71.3-5.8 132.5-47.6 165.2-107.2zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zM187.3 100.7c-6.2-6.2-16.4-6.2-22.6 0l-32 32c-6.2 6.2-6.2 16.4 0 22.6s16.4 6.2 22.6 0l32-32c6.2-6.2 6.2-16.4 0-22.6z\"],\"cart-flatbed-suitcase\":[640,512,[\"luggage-cart\"],\"f59d\",\"M0 32C0 14.3 14.3 0 32 0H48c44.2 0 80 35.8 80 80V368c0 8.8 7.2 16 16 16H608c17.7 0 32 14.3 32 32s-14.3 32-32 32H541.3c1.8 5 2.7 10.4 2.7 16c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-5.6 1-11 2.7-16H253.3c1.8 5 2.7 10.4 2.7 16c0 26.5-21.5 48-48 48s-48-21.5-48-48c0-5.6 1-11 2.7-16H144c-44.2 0-80-35.8-80-80V80c0-8.8-7.2-16-16-16H32C14.3 64 0 49.7 0 32zM432 96V56c0-4.4-3.6-8-8-8H344c-4.4 0-8 3.6-8 8V96h96zM288 96V56c0-30.9 25.1-56 56-56h80c30.9 0 56 25.1 56 56V96 320H288V96zM512 320V96h16c26.5 0 48 21.5 48 48V272c0 26.5-21.5 48-48 48H512zM240 96h16V320H240c-26.5 0-48-21.5-48-48V144c0-26.5 21.5-48 48-48z\"],\"rectangle-xmark\":[512,512,[62164,\"rectangle-times\",\"times-rectangle\",\"window-close\"],\"f410\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H448c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM175 175c9.4-9.4 24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9z\"],\"baht-sign\":[320,512,[],\"e0ac\",\"M144 0c-17.7 0-32 14.3-32 32V64H37.6C16.8 64 0 80.8 0 101.6V224v41.7V288 406.3c0 23 18.7 41.7 41.7 41.7H112v32c0 17.7 14.3 32 32 32s32-14.3 32-32V448h32c61.9 0 112-50.1 112-112c0-40.1-21.1-75.3-52.7-95.1C280.3 222.6 288 200.2 288 176c0-61.9-50.1-112-112-112V32c0-17.7-14.3-32-32-32zM112 128v96H64V128h48zm64 96V128c26.5 0 48 21.5 48 48s-21.5 48-48 48zm-64 64v96H64V288h48zm64 96V288h32c26.5 0 48 21.5 48 48s-21.5 48-48 48H176z\"],\"book-open\":[576,512,[128214,128366],\"f518\",\"M249.6 471.5c10.8 3.8 22.4-4.1 22.4-15.5V78.6c0-4.2-1.6-8.4-5-11C247.4 52 202.4 32 144 32C93.5 32 46.3 45.3 18.1 56.1C6.8 60.5 0 71.7 0 83.8V454.1c0 11.9 12.8 20.2 24.1 16.5C55.6 460.1 105.5 448 144 448c33.9 0 79 14 105.6 23.5zm76.8 0C353 462 398.1 448 432 448c38.5 0 88.4 12.1 119.9 22.6c11.3 3.8 24.1-4.6 24.1-16.5V83.8c0-12.1-6.8-23.3-18.1-27.6C529.7 45.3 482.5 32 432 32c-58.4 0-103.4 20-123 35.6c-3.3 2.6-5 6.8-5 11V456c0 11.4 11.7 19.3 22.4 15.5z\"],\"book-journal-whills\":[448,512,[\"journal-whills\"],\"f66a\",\"M0 96C0 43 43 0 96 0H384h32c17.7 0 32 14.3 32 32V352c0 17.7-14.3 32-32 32v64c17.7 0 32 14.3 32 32s-14.3 32-32 32H384 96c-53 0-96-43-96-96V96zM64 416c0 17.7 14.3 32 32 32H352V384H96c-17.7 0-32 14.3-32 32zm90.4-234.4l-21.2-21.2c-3 10.1-5.1 20.6-5.1 31.6c0 .2 0 .5 .1 .8s.1 .5 .1 .8L165.2 226c2.5 2.1 3.4 5.8 2.3 8.9c-1.3 3-4.1 5.1-7.5 5.1c-1.9-.1-3.8-.8-5.2-2l-23.6-20.6C142.8 267 186.9 304 240 304s97.3-37 108.9-86.6L325.3 238c-1.4 1.2-3.3 2-5.3 2c-2.2-.1-4.4-1.1-6-2.8c-1.2-1.5-1.9-3.4-2-5.2c.1-2.2 1.1-4.4 2.8-6l37.1-32.5c0-.3 0-.5 .1-.8s.1-.5 .1-.8c0-11-2.1-21.5-5.1-31.6l-21.2 21.2c-3.1 3.1-8.1 3.1-11.3 0s-3.1-8.1 0-11.2l26.4-26.5c-8.2-17-20.5-31.7-35.9-42.6c-2.7-1.9-6.2 1.4-5 4.5c8.5 22.4 3.6 48-13 65.6c-3.2 3.4-3.6 8.9-.9 12.7c9.8 14 12.7 31.9 7.5 48.5c-5.9 19.4-22 34.1-41.9 38.3l-1.4-34.3 12.6 8.6c.6 .4 1.5 .6 2.3 .6c1.5 0 2.7-.8 3.5-2s.6-2.8-.1-4L260 225.4l18-3.6c1.8-.4 3.1-2.1 3.1-4s-1.4-3.5-3.1-3.9l-18-3.7 8.5-14.3c.8-1.2 .9-2.9 .1-4.1s-2-2-3.5-2l-.1 0c-.7 .1-1.5 .3-2.1 .7l-14.1 9.6L244 87.9c-.1-2.2-1.9-3.9-4-3.9s-3.9 1.6-4 3.9l-4.6 110.8-12-8.1c-1.5-1.1-3.6-.9-5 .4s-1.6 3.4-.8 5l8.6 14.3-18 3.7c-1.8 .4-3.1 2-3.1 3.9s1.4 3.6 3.1 4l18 3.8-8.6 14.2c-.2 .6-.5 1.4-.5 2c0 1.1 .5 2.1 1.2 3c.8 .6 1.8 1 2.8 1c.7 0 1.6-.2 2.2-.6l10.4-7.1-1.4 32.8c-19.9-4.1-36-18.9-41.9-38.3c-5.1-16.6-2.2-34.4 7.6-48.5c2.7-3.9 2.3-9.3-.9-12.7c-16.6-17.5-21.6-43.1-13.1-65.5c1.2-3.1-2.3-6.4-5-4.5c-15.3 10.9-27.6 25.6-35.8 42.6l26.4 26.5c3.1 3.1 3.1 8.1 0 11.2s-8.1 3.1-11.2 0z\"],handcuffs:[640,512,[],\"e4f8\",\"M240 32a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zM192 48a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm-32 80c17.7 0 32 14.3 32 32h8c13.3 0 24 10.7 24 24v16c0 1.7-.2 3.4-.5 5.1C280.3 229.6 320 286.2 320 352c0 88.4-71.6 160-160 160S0 440.4 0 352c0-65.8 39.7-122.4 96.5-146.9c-.4-1.6-.5-3.3-.5-5.1V184c0-13.3 10.7-24 24-24h8c0-17.7 14.3-32 32-32zm0 320a96 96 0 1 0 0-192 96 96 0 1 0 0 192zm192-96c0-25.9-5.1-50.5-14.4-73.1c16.9-32.9 44.8-59.1 78.9-73.9c-.4-1.6-.5-3.3-.5-5.1V184c0-13.3 10.7-24 24-24h8c0-17.7 14.3-32 32-32s32 14.3 32 32h8c13.3 0 24 10.7 24 24v16c0 1.7-.2 3.4-.5 5.1C600.3 229.6 640 286.2 640 352c0 88.4-71.6 160-160 160c-62 0-115.8-35.3-142.4-86.9c9.3-22.5 14.4-47.2 14.4-73.1zm224 0a96 96 0 1 0 -192 0 96 96 0 1 0 192 0zM368 0a32 32 0 1 1 0 64 32 32 0 1 1 0-64zm80 48a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"triangle-exclamation\":[512,512,[9888,\"exclamation-triangle\",\"warning\"],\"f071\",\"M256 32c14.2 0 27.3 7.5 34.5 19.8l216 368c7.3 12.4 7.3 27.7 .2 40.1S486.3 480 472 480H40c-14.3 0-27.6-7.7-34.7-20.1s-7-27.8 .2-40.1l216-368C228.7 39.5 241.8 32 256 32zm0 128c-13.3 0-24 10.7-24 24V296c0 13.3 10.7 24 24 24s24-10.7 24-24V184c0-13.3-10.7-24-24-24zm32 224a32 32 0 1 0 -64 0 32 32 0 1 0 64 0z\"],database:[448,512,[],\"f1c0\",\"M448 80v48c0 44.2-100.3 80-224 80S0 172.2 0 128V80C0 35.8 100.3 0 224 0S448 35.8 448 80zM393.2 214.7c20.8-7.4 39.9-16.9 54.8-28.6V288c0 44.2-100.3 80-224 80S0 332.2 0 288V186.1c14.9 11.8 34 21.2 54.8 28.6C99.7 230.7 159.5 240 224 240s124.3-9.3 169.2-25.3zM0 346.1c14.9 11.8 34 21.2 54.8 28.6C99.7 390.7 159.5 400 224 400s124.3-9.3 169.2-25.3c20.8-7.4 39.9-16.9 54.8-28.6V432c0 44.2-100.3 80-224 80S0 476.2 0 432V346.1z\"],share:[512,512,[\"mail-forward\"],\"f064\",\"M307 34.8c-11.5 5.1-19 16.6-19 29.2v64H176C78.8 128 0 206.8 0 304C0 417.3 81.5 467.9 100.2 478.1c2.5 1.4 5.3 1.9 8.1 1.9c10.9 0 19.7-8.9 19.7-19.7c0-7.5-4.3-14.4-9.8-19.5C108.8 431.9 96 414.4 96 384c0-53 43-96 96-96h96v64c0 12.6 7.4 24.1 19 29.2s25 3 34.4-5.4l160-144c6.7-6.1 10.6-14.7 10.6-23.8s-3.8-17.7-10.6-23.8l-160-144c-9.4-8.5-22.9-10.6-34.4-5.4z\"],\"bottle-droplet\":[320,512,[],\"e4c4\",\"M96 0C82.7 0 72 10.7 72 24s10.7 24 24 24c4.4 0 8 3.6 8 8v64.9c0 12.2-7.2 23.1-17.2 30.1C53.7 174.1 32 212.5 32 256V448c0 35.3 28.7 64 64 64H224c35.3 0 64-28.7 64-64V256c0-43.5-21.7-81.9-54.8-105c-10-7-17.2-17.9-17.2-30.1V56c0-4.4 3.6-8 8-8c13.3 0 24-10.7 24-24s-10.7-24-24-24l-8 0 0 0 0 0H104l0 0 0 0L96 0zm64 382c-26.5 0-48-20.1-48-45c0-16.8 22.1-48.1 36.3-66.4c6-7.8 17.5-7.8 23.5 0C185.9 288.9 208 320.2 208 337c0 24.9-21.5 45-48 45z\"],\"mask-face\":[640,512,[],\"e1d7\",\"M320 64c-27.2 0-53.8 8-76.4 23.1l-37.1 24.8c-15.8 10.5-34.3 16.1-53.3 16.1H144 128 56c-30.9 0-56 25.1-56 56v85c0 55.1 37.5 103.1 90.9 116.4l108 27C233.8 435 275.4 448 320 448s86.2-13 121.1-35.5l108-27C602.5 372.1 640 324.1 640 269V184c0-30.9-25.1-56-56-56H512 496h-9.2c-19 0-37.5-5.6-53.3-16.1L396.4 87.1C373.8 72 347.2 64 320 64zM132.3 346.3l-29.8-7.4C70.5 330.9 48 302.1 48 269V184c0-4.4 3.6-8 8-8H96v48c0 45.1 13.4 87.2 36.3 122.3zm405.1-7.4l-29.8 7.4c23-35.2 36.3-77.2 36.3-122.3V176h40c4.4 0 8 3.6 8 8v85c0 33-22.5 61.8-54.5 69.9zM192 208c0-8.8 7.2-16 16-16H432c8.8 0 16 7.2 16 16s-7.2 16-16 16H208c-8.8 0-16-7.2-16-16zm16 48H432c8.8 0 16 7.2 16 16s-7.2 16-16 16H208c-8.8 0-16-7.2-16-16s7.2-16 16-16zm16 80c0-8.8 7.2-16 16-16H400c8.8 0 16 7.2 16 16s-7.2 16-16 16H240c-8.8 0-16-7.2-16-16z\"],\"hill-rockslide\":[576,512,[],\"e508\",\"M252.4 103.8l27 48c2.8 5 8.2 8.2 13.9 8.2l53.3 0c5.8 0 11.1-3.1 13.9-8.2l27-48c2.7-4.9 2.7-10.8 0-15.7l-27-48c-2.8-5-8.2-8.2-13.9-8.2H293.4c-5.8 0-11.1 3.1-13.9 8.2l-27 48c-2.7 4.9-2.7 10.8 0 15.7zM68.3 87C43.1 61.8 0 79.7 0 115.3V432c0 44.2 35.8 80 80 80H396.7c35.6 0 53.5-43.1 28.3-68.3L68.3 87zM504.2 403.6c4.9 2.7 10.8 2.7 15.7 0l48-27c5-2.8 8.2-8.2 8.2-13.9V309.4c0-5.8-3.1-11.1-8.2-13.9l-48-27c-4.9-2.7-10.8-2.7-15.7 0l-48 27c-5 2.8-8.2 8.2-8.2 13.9v53.3c0 5.8 3.1 11.1 8.2 13.9l48 27zM192 64a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zM384 288a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"right-left\":[512,512,[\"exchange-alt\"],\"f362\",\"M32 96l320 0V32c0-12.9 7.8-24.6 19.8-29.6s25.7-2.2 34.9 6.9l96 96c6 6 9.4 14.1 9.4 22.6s-3.4 16.6-9.4 22.6l-96 96c-9.2 9.2-22.9 11.9-34.9 6.9s-19.8-16.6-19.8-29.6V160L32 160c-17.7 0-32-14.3-32-32s14.3-32 32-32zM480 352c17.7 0 32 14.3 32 32s-14.3 32-32 32H160v64c0 12.9-7.8 24.6-19.8 29.6s-25.7 2.2-34.9-6.9l-96-96c-6-6-9.4-14.1-9.4-22.6s3.4-16.6 9.4-22.6l96-96c9.2-9.2 22.9-11.9 34.9-6.9s19.8 16.6 19.8 29.6l0 64H480z\"],\"paper-plane\":[512,512,[61913],\"f1d8\",\"M498.1 5.6c10.1 7 15.4 19.1 13.5 31.2l-64 416c-1.5 9.7-7.4 18.2-16 23s-18.9 5.4-28 1.6L284 427.7l-68.5 74.1c-8.9 9.7-22.9 12.9-35.2 8.1S160 493.2 160 480V396.4c0-4 1.5-7.8 4.2-10.7L331.8 202.8c5.8-6.3 5.6-16-.4-22s-15.7-6.4-22-.7L106 360.8 17.7 316.6C7.1 311.3 .3 300.7 0 288.9s5.9-22.8 16.1-28.7l448-256c10.7-6.1 23.9-5.5 34 1.4z\"],\"road-circle-exclamation\":[640,512,[],\"e565\",\"M213.2 32H288V96c0 17.7 14.3 32 32 32s32-14.3 32-32V32h74.8c27.1 0 51.3 17.1 60.3 42.6l42.7 120.6c-10.9-2.1-22.2-3.2-33.8-3.2c-59.5 0-112.1 29.6-144 74.8V224c0-17.7-14.3-32-32-32s-32 14.3-32 32v64c0 17.7 14.3 32 32 32c2.3 0 4.6-.3 6.8-.7c-4.5 15.5-6.8 31.8-6.8 48.7c0 5.4 .2 10.7 .7 16l-.7 0c-17.7 0-32 14.3-32 32v64H86.6C56.5 480 32 455.5 32 425.4c0-6.2 1.1-12.4 3.1-18.2L152.9 74.6C162 49.1 186.1 32 213.2 32zM496 224a144 144 0 1 1 0 288 144 144 0 1 1 0-288zm0 240a24 24 0 1 0 0-48 24 24 0 1 0 0 48zm0-192c-8.8 0-16 7.2-16 16v80c0 8.8 7.2 16 16 16s16-7.2 16-16V288c0-8.8-7.2-16-16-16z\"],dungeon:[512,512,[],\"f6d9\",\"M336.6 156.5c1.3 1.1 2.7 2.2 3.9 3.3c9.3 8.2 23 10.5 33.4 3.6l67.6-45.1c11.4-7.6 14.2-23.2 5.1-33.4C430 66.6 410.9 50.6 389.7 37.6c-11.9-7.3-26.9-1.4-32.1 11.6l-30.5 76.2c-4.5 11.1 .2 23.6 9.5 31.2zM328 36.8c5.1-12.8-1.6-27.4-15-30.5C294.7 2.2 275.6 0 256 0s-38.7 2.2-57 6.4C185.5 9.4 178.8 24 184 36.8l30.3 75.8c4.5 11.3 16.8 17.2 29 16c4.2-.4 8.4-.6 12.7-.6s8.6 .2 12.7 .6c12.1 1.2 24.4-4.7 29-16L328 36.8zM65.5 85c-9.1 10.2-6.3 25.8 5.1 33.4l67.6 45.1c10.3 6.9 24.1 4.6 33.4-3.6c1.3-1.1 2.6-2.3 4-3.3c9.3-7.5 13.9-20.1 9.5-31.2L154.4 49.2c-5.2-12.9-20.3-18.8-32.1-11.6C101.1 50.6 82 66.6 65.5 85zm314 137.1c.9 3.3 1.7 6.6 2.3 10c2.5 13 13 23.9 26.2 23.9h80c13.3 0 24.1-10.8 22.9-24c-2.5-27.2-9.3-53.2-19.7-77.3c-5.5-12.9-21.4-16.6-33.1-8.9l-68.6 45.7c-9.8 6.5-13.2 19.2-10 30.5zM53.9 145.8c-11.6-7.8-27.6-4-33.1 8.9C10.4 178.8 3.6 204.8 1.1 232c-1.2 13.2 9.6 24 22.9 24h80c13.3 0 23.8-10.8 26.2-23.9c.6-3.4 1.4-6.7 2.3-10c3.1-11.4-.2-24-10-30.5L53.9 145.8zM104 288H24c-13.3 0-24 10.7-24 24v48c0 13.3 10.7 24 24 24h80c13.3 0 24-10.7 24-24V312c0-13.3-10.7-24-24-24zm304 0c-13.3 0-24 10.7-24 24v48c0 13.3 10.7 24 24 24h80c13.3 0 24-10.7 24-24V312c0-13.3-10.7-24-24-24H408zM24 416c-13.3 0-24 10.7-24 24v48c0 13.3 10.7 24 24 24h80c13.3 0 24-10.7 24-24V440c0-13.3-10.7-24-24-24H24zm384 0c-13.3 0-24 10.7-24 24v48c0 13.3 10.7 24 24 24h80c13.3 0 24-10.7 24-24V440c0-13.3-10.7-24-24-24H408zM272 192c0-8.8-7.2-16-16-16s-16 7.2-16 16V464c0 8.8 7.2 16 16 16s16-7.2 16-16V192zm-64 32c0-8.8-7.2-16-16-16s-16 7.2-16 16V464c0 8.8 7.2 16 16 16s16-7.2 16-16V224zm128 0c0-8.8-7.2-16-16-16s-16 7.2-16 16V464c0 8.8 7.2 16 16 16s16-7.2 16-16V224z\"],\"align-right\":[448,512,[],\"f038\",\"M448 64c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32s14.3-32 32-32H416c17.7 0 32 14.3 32 32zm0 256c0 17.7-14.3 32-32 32H192c-17.7 0-32-14.3-32-32s14.3-32 32-32H416c17.7 0 32 14.3 32 32zM0 192c0-17.7 14.3-32 32-32H416c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32zM448 448c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32H416c17.7 0 32 14.3 32 32z\"],\"money-bill-1-wave\":[576,512,[\"money-bill-wave-alt\"],\"f53b\",\"M0 112.5V422.3c0 18 10.1 35 27 41.3c87 32.5 174 10.3 261-11.9c79.8-20.3 159.6-40.7 239.3-18.9c23 6.3 48.7-9.5 48.7-33.4V89.7c0-18-10.1-35-27-41.3C462 15.9 375 38.1 288 60.3C208.2 80.6 128.4 100.9 48.7 79.1C25.6 72.8 0 88.6 0 112.5zM128 416H64V352c35.3 0 64 28.7 64 64zM64 224V160h64c0 35.3-28.7 64-64 64zM448 352c0-35.3 28.7-64 64-64v64H448zm64-192c-35.3 0-64-28.7-64-64h64v64zM384 256c0 61.9-43 112-96 112s-96-50.1-96-112s43-112 96-112s96 50.1 96 112zM252 208c0 9.7 6.9 17.7 16 19.6V276h-4c-11 0-20 9-20 20s9 20 20 20h24 24c11 0 20-9 20-20s-9-20-20-20h-4V208c0-11-9-20-20-20H272c-11 0-20 9-20 20z\"],\"life-ring\":[512,512,[],\"f1cd\",\"M367.2 412.5C335.9 434.9 297.5 448 256 448s-79.9-13.1-111.2-35.5l58-58c15.8 8.6 34 13.5 53.3 13.5s37.4-4.9 53.3-13.5l58 58zm90.7 .8c33.8-43.4 54-98 54-157.3s-20.2-113.9-54-157.3c9-12.5 7.9-30.1-3.4-41.3S425.8 45 413.3 54C369.9 20.2 315.3 0 256 0S142.1 20.2 98.7 54c-12.5-9-30.1-7.9-41.3 3.4S45 86.2 54 98.7C20.2 142.1 0 196.7 0 256s20.2 113.9 54 157.3c-9 12.5-7.9 30.1 3.4 41.3S86.2 467 98.7 458c43.4 33.8 98 54 157.3 54s113.9-20.2 157.3-54c12.5 9 30.1 7.9 41.3-3.4s12.4-28.8 3.4-41.3zm-45.5-46.1l-58-58c8.6-15.8 13.5-34 13.5-53.3s-4.9-37.4-13.5-53.3l58-58C434.9 176.1 448 214.5 448 256s-13.1 79.9-35.5 111.2zM367.2 99.5l-58 58c-15.8-8.6-34-13.5-53.3-13.5s-37.4 4.9-53.3 13.5l-58-58C176.1 77.1 214.5 64 256 64s79.9 13.1 111.2 35.5zM157.5 309.3l-58 58C77.1 335.9 64 297.5 64 256s13.1-79.9 35.5-111.2l58 58c-8.6 15.8-13.5 34-13.5 53.3s4.9 37.4 13.5 53.3zM208 256a48 48 0 1 1 96 0 48 48 0 1 1 -96 0z\"],hands:[576,512,[\"sign-language\",\"signing\"],\"f2a7\",\"M544 160l-.1 72.6c-.1 52.2-24 101-64 133.1c.1-1.9 .1-3.8 .1-5.7v-8c0-71.8-37-138.6-97.9-176.7l-60.2-37.6c-8.6-5.4-17.9-8.4-27.3-9.4L248.7 48.8c-6.6-11.5-2.7-26.2 8.8-32.8s26.2-2.7 32.8 8.8l78 135.1c3.3 5.7 10.7 7.7 16.4 4.4s7.7-10.7 4.4-16.4l-62-107.4c-6.6-11.5-2.7-26.2 8.8-32.8S362 5 368.6 16.5l68 117.8 0 0 0 0 43.3 75L480 160c0-17.7 14.4-32 32-32s32 14.4 32 32zM243.9 88.5L268.5 131c-13.9 4.5-26.4 13.7-34.7 27c-.9 1.4-1.7 2.9-2.5 4.4l-28.9-50c-6.6-11.5-2.7-26.2 8.8-32.8s26.2-2.7 32.8 8.8zm-46.4 63.7l26.8 46.4c.6 6 2.1 11.8 4.3 17.4H224 210.7l0 0H179l-23-39.8c-6.6-11.5-2.7-26.2 8.8-32.8s26.2-2.7 32.8 8.8zM260.9 175c9.4-15 29.1-19.5 44.1-10.2l60.2 37.6C416.7 234.7 448 291.2 448 352v8c0 83.9-68.1 152-152 152H120c-13.3 0-24-10.7-24-24s10.7-24 24-24h92c6.6 0 12-5.4 12-12s-5.4-12-12-12H88c-13.3 0-24-10.7-24-24s10.7-24 24-24H212c6.6 0 12-5.4 12-12s-5.4-12-12-12H56c-13.3 0-24-10.7-24-24s10.7-24 24-24H212c6.6 0 12-5.4 12-12s-5.4-12-12-12H88c-13.3 0-24-10.7-24-24s10.7-24 24-24H224l0 0 0 0h93.2L271 219.1c-15-9.4-19.5-29.1-10.2-44.1z\"],\"calendar-day\":[448,512,[],\"f783\",\"M128 0c17.7 0 32 14.3 32 32V64H288V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64h48c26.5 0 48 21.5 48 48v48H0V112C0 85.5 21.5 64 48 64H96V32c0-17.7 14.3-32 32-32zM0 192H448V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V192zm80 64c-8.8 0-16 7.2-16 16v96c0 8.8 7.2 16 16 16h96c8.8 0 16-7.2 16-16V272c0-8.8-7.2-16-16-16H80z\"],\"water-ladder\":[576,512,[\"ladder-water\",\"swimming-pool\"],\"f5c5\",\"M128 127.7C128 74.9 170.9 32 223.7 32c48.3 0 89 36 95 83.9l1 8.2c2.2 17.5-10.2 33.5-27.8 35.7s-33.5-10.2-35.7-27.8l-1-8.2c-2-15.9-15.5-27.8-31.5-27.8c-17.5 0-31.7 14.2-31.7 31.7V224H384V127.7C384 74.9 426.9 32 479.7 32c48.3 0 89 36 95 83.9l1 8.2c2.2 17.5-10.2 33.5-27.8 35.7s-33.5-10.2-35.7-27.8l-1-8.2c-2-15.9-15.5-27.8-31.5-27.8c-17.5 0-31.7 14.2-31.7 31.7V361c-1.6 1-3.3 2-4.8 3.1c-18 12.4-40.1 20.3-59.2 20.3h0V288H192v96.5c-19 0-41.2-7.9-59.1-20.3c-1.6-1.1-3.2-2.2-4.9-3.1V127.7zM306.5 389.9C329 405.4 356.5 416 384 416c26.9 0 55.4-10.8 77.4-26.1l0 0c11.9-8.5 28.1-7.8 39.2 1.7c14.4 11.9 32.5 21 50.6 25.2c17.2 4 27.9 21.2 23.9 38.4s-21.2 27.9-38.4 23.9c-24.5-5.7-44.9-16.5-58.2-25C449.5 469.7 417 480 384 480c-31.9 0-60.6-9.9-80.4-18.9c-5.8-2.7-11.1-5.3-15.6-7.7c-4.5 2.4-9.7 5.1-15.6 7.7c-19.8 9-48.5 18.9-80.4 18.9c-33 0-65.5-10.3-94.5-25.8c-13.4 8.4-33.7 19.3-58.2 25c-17.2 4-34.4-6.7-38.4-23.9s6.7-34.4 23.9-38.4c18.1-4.2 36.2-13.3 50.6-25.2c11.1-9.4 27.3-10.1 39.2-1.7l0 0C136.7 405.2 165.1 416 192 416c27.5 0 55-10.6 77.5-26.1c11.1-7.9 25.9-7.9 37 0z\"],\"arrows-up-down\":[320,512,[\"arrows-v\"],\"f07d\",\"M182.6 9.4c-12.5-12.5-32.8-12.5-45.3 0l-96 96c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L128 109.3V402.7L86.6 361.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3l96 96c12.5 12.5 32.8 12.5 45.3 0l96-96c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L192 402.7V109.3l41.4 41.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-96-96z\"],\"face-grimace\":[512,512,[128556,\"grimace\"],\"f57f\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zm96-112h-8V360l55.3 0c-3.8 22.7-23.6 40-47.3 40zm47.3-56L344 344V304h8c23.8 0 43.5 17.3 47.3 40zM328 344H264V304h64v40zm0 56H264V360h64v40zm-80-96v40l-64 0V304h64zm0 56v40H184V360l64 0zm-80-16H112.7c3.8-22.7 23.6-40 47.3-40h8v40zm0 56h-8c-23.8 0-43.5-17.3-47.3-40H168v40zM144.4 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],\"wheelchair-move\":[448,512,[\"wheelchair-alt\"],\"e2ce\",\"M320 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zM204.5 121.3c-5.4-2.5-11.7-1.9-16.4 1.7l-40.9 30.7c-14.1 10.6-34.2 7.7-44.8-6.4s-7.7-34.2 6.4-44.8l40.9-30.7c23.7-17.8 55.3-21 82.1-8.4l90.4 42.5c29.1 13.7 36.8 51.6 15.2 75.5L299.1 224h97.4c30.3 0 53 27.7 47.1 57.4L415.4 422.3c-3.5 17.3-20.3 28.6-37.7 25.1s-28.6-20.3-25.1-37.7L377 288H306.7c8.6 19.6 13.3 41.2 13.3 64c0 88.4-71.6 160-160 160S0 440.4 0 352s71.6-160 160-160c11.1 0 22 1.1 32.4 3.3l54.2-54.2-42.1-19.8zM160 448a96 96 0 1 0 0-192 96 96 0 1 0 0 192z\"],\"turn-down\":[384,512,[10549,\"level-down-alt\"],\"f3be\",\"M350 334.5c3.8 8.8 2 19-4.6 26l-136 144c-4.5 4.8-10.8 7.5-17.4 7.5s-12.9-2.7-17.4-7.5l-136-144c-6.6-7-8.4-17.2-4.6-26s12.5-14.5 22-14.5h88l0-192c0-17.7-14.3-32-32-32H32C14.3 96 0 81.7 0 64V32C0 14.3 14.3 0 32 0l80 0c70.7 0 128 57.3 128 128l0 192h88c9.6 0 18.2 5.7 22 14.5z\"],\"person-walking-arrow-right\":[640,512,[],\"e552\",\"M208 96a48 48 0 1 0 0-96 48 48 0 1 0 0 96zM123.7 200.5c1-.4 1.9-.8 2.9-1.2l-16.9 63.5c-5.6 21.1-.1 43.6 14.7 59.7l70.7 77.1 22 88.1c4.3 17.1 21.7 27.6 38.8 23.3s27.6-21.7 23.3-38.8l-23-92.1c-1.9-7.8-5.8-14.9-11.2-20.8l-49.5-54 19.3-65.5 9.6 23c4.4 10.6 12.5 19.3 22.8 24.5l26.7 13.3c15.8 7.9 35 1.5 42.9-14.3s1.5-35-14.3-42.9L281 232.7l-15.3-36.8C248.5 154.8 208.3 128 163.7 128c-22.8 0-45.3 4.8-66.1 14l-8 3.5c-32.9 14.6-58.1 42.4-69.4 76.5l-2.6 7.8c-5.6 16.8 3.5 34.9 20.2 40.5s34.9-3.5 40.5-20.2l2.6-7.8c5.7-17.1 18.3-30.9 34.7-38.2l8-3.5zm-30 135.1L68.7 398 9.4 457.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L116.3 441c4.6-4.6 8.2-10.1 10.6-16.1l14.5-36.2-40.7-44.4c-2.5-2.7-4.8-5.6-7-8.6zM550.6 153.4c-12.5-12.5-32.8-12.5-45.3 0s-12.5 32.8 0 45.3L530.7 224H384c-17.7 0-32 14.3-32 32s14.3 32 32 32H530.7l-25.4 25.4c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l80-80c12.5-12.5 12.5-32.8 0-45.3l-80-80z\"],\"square-envelope\":[448,512,[\"envelope-square\"],\"f199\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM218 271.7L64.2 172.4C66 156.4 79.5 144 96 144H352c16.5 0 30 12.4 31.8 28.4L230 271.7c-1.8 1.2-3.9 1.8-6 1.8s-4.2-.6-6-1.8zm29.4 26.9L384 210.4V336c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V210.4l136.6 88.2c7 4.5 15.1 6.9 23.4 6.9s16.4-2.4 23.4-6.9z\"],dice:[640,512,[127922],\"f522\",\"M274.9 34.3c-28.1-28.1-73.7-28.1-101.8 0L34.3 173.1c-28.1 28.1-28.1 73.7 0 101.8L173.1 413.7c28.1 28.1 73.7 28.1 101.8 0L413.7 274.9c28.1-28.1 28.1-73.7 0-101.8L274.9 34.3zM200 224a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zM96 200a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM224 376a24 24 0 1 1 0-48 24 24 0 1 1 0 48zM352 200a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM224 120a24 24 0 1 1 0-48 24 24 0 1 1 0 48zm96 328c0 35.3 28.7 64 64 64H576c35.3 0 64-28.7 64-64V256c0-35.3-28.7-64-64-64H461.7c11.6 36 3.1 77-25.4 105.5L320 413.8V448zM480 328a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],\"bowling-ball\":[512,512,[],\"f436\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM240 80a32 32 0 1 1 0 64 32 32 0 1 1 0-64zM208 208a32 32 0 1 1 64 0 32 32 0 1 1 -64 0zm-64-64a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],brain:[512,512,[129504],\"f5dc\",\"M184 0c30.9 0 56 25.1 56 56V456c0 30.9-25.1 56-56 56c-28.9 0-52.7-21.9-55.7-50.1c-5.2 1.4-10.7 2.1-16.3 2.1c-35.3 0-64-28.7-64-64c0-7.4 1.3-14.6 3.6-21.2C21.4 367.4 0 338.2 0 304c0-31.9 18.7-59.5 45.8-72.3C37.1 220.8 32 207 32 192c0-30.7 21.6-56.3 50.4-62.6C80.8 123.9 80 118 80 112c0-29.9 20.6-55.1 48.3-62.1C131.3 21.9 155.1 0 184 0zM328 0c28.9 0 52.6 21.9 55.7 49.9c27.8 7 48.3 32.1 48.3 62.1c0 6-.8 11.9-2.4 17.4c28.8 6.2 50.4 31.9 50.4 62.6c0 15-5.1 28.8-13.8 39.7C493.3 244.5 512 272.1 512 304c0 34.2-21.4 63.4-51.6 74.8c2.3 6.6 3.6 13.8 3.6 21.2c0 35.3-28.7 64-64 64c-5.6 0-11.1-.7-16.3-2.1c-3 28.2-26.8 50.1-55.7 50.1c-30.9 0-56-25.1-56-56V56c0-30.9 25.1-56 56-56z\"],bandage:[640,512,[129657,\"band-aid\"],\"f462\",\"M480 416h96c35.3 0 64-28.7 64-64V160c0-35.3-28.7-64-64-64H480V416zM448 96H192V416H448V96zM64 96C28.7 96 0 124.7 0 160V352c0 35.3 28.7 64 64 64h96V96H64zM248 208a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm120-24a24 24 0 1 1 0 48 24 24 0 1 1 0-48zM248 304a24 24 0 1 1 48 0 24 24 0 1 1 -48 0zm120-24a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],\"calendar-minus\":[448,512,[],\"f272\",\"M128 0c17.7 0 32 14.3 32 32V64H288V32c0-17.7 14.3-32 32-32s32 14.3 32 32V64h48c26.5 0 48 21.5 48 48v48H0V112C0 85.5 21.5 64 48 64H96V32c0-17.7 14.3-32 32-32zM0 192H448V464c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V192zM312 376c13.3 0 24-10.7 24-24s-10.7-24-24-24H136c-13.3 0-24 10.7-24 24s10.7 24 24 24H312z\"],\"circle-xmark\":[512,512,[61532,\"times-circle\",\"xmark-circle\"],\"f057\",\"M256 512A256 256 0 1 0 256 0a256 256 0 1 0 0 512zM175 175c9.4-9.4 24.6-9.4 33.9 0l47 47 47-47c9.4-9.4 24.6-9.4 33.9 0s9.4 24.6 0 33.9l-47 47 47 47c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-47-47-47 47c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l47-47-47-47c-9.4-9.4-9.4-24.6 0-33.9z\"],gifts:[640,512,[],\"f79c\",\"M200.6 32C205 19.5 198.5 5.8 186 1.4S159.8 3.5 155.4 16L144.7 46.2l-9.9-29.8C130.6 3.8 117-3 104.4 1.2S85 19 89.2 31.6l8.3 25-27.4-20c-10.7-7.8-25.7-5.4-33.5 5.3s-5.4 25.7 5.3 33.5L70.2 96H48C21.5 96 0 117.5 0 144V464c0 26.5 21.5 48 48 48H200.6c-5.4-9.4-8.6-20.3-8.6-32V256c0-29.9 20.5-55 48.2-62c1.8-31 17.1-58.2 40.1-76.1C271.7 104.7 256.9 96 240 96H217.8l28.3-20.6c10.7-7.8 13.1-22.8 5.3-33.5s-22.8-13.1-33.5-5.3L192.5 55.1 200.6 32zM363.5 185.5L393.1 224H344c-13.3 0-24-10.7-24-24c0-13.1 10.8-24 24.2-24c7.6 0 14.7 3.5 19.3 9.5zM272 200c0 8.4 1.4 16.5 4.1 24H272c-26.5 0-48 21.5-48 48v80H416V256h32v96H640V272c0-26.5-21.5-48-48-48h-4.1c2.7-7.5 4.1-15.6 4.1-24c0-39.9-32.5-72-72.2-72c-22.4 0-43.6 10.4-57.3 28.2L432 195.8l-30.5-39.6c-13.7-17.8-35-28.2-57.3-28.2c-39.7 0-72.2 32.1-72.2 72zM224 464c0 26.5 21.5 48 48 48H416V384H224v80zm224 48H592c26.5 0 48-21.5 48-48V384H448V512zm96-312c0 13.3-10.7 24-24 24H470.9l29.6-38.5c4.6-5.9 11.7-9.5 19.3-9.5c13.4 0 24.2 10.9 24.2 24z\"],hotel:[512,512,[127976],\"f594\",\"M0 32C0 14.3 14.3 0 32 0H480c17.7 0 32 14.3 32 32s-14.3 32-32 32V448c17.7 0 32 14.3 32 32s-14.3 32-32 32H304V464c0-26.5-21.5-48-48-48s-48 21.5-48 48v48H32c-17.7 0-32-14.3-32-32s14.3-32 32-32V64C14.3 64 0 49.7 0 32zm96 80v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H112c-8.8 0-16 7.2-16 16zM240 96c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H240zm112 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V112c0-8.8-7.2-16-16-16H368c-8.8 0-16 7.2-16 16zM112 192c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V208c0-8.8-7.2-16-16-16H112zm112 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V208c0-8.8-7.2-16-16-16H240c-8.8 0-16 7.2-16 16zm144-16c-8.8 0-16 7.2-16 16v32c0 8.8 7.2 16 16 16h32c8.8 0 16-7.2 16-16V208c0-8.8-7.2-16-16-16H368zM328 384c13.3 0 24.3-10.9 21-23.8c-10.6-41.5-48.2-72.2-93-72.2s-82.5 30.7-93 72.2c-3.3 12.8 7.8 23.8 21 23.8H328z\"],\"earth-asia\":[512,512,[127759,\"globe-asia\"],\"f57e\",\"M51.7 295.1l31.7 6.3c7.9 1.6 16-.9 21.7-6.6l15.4-15.4c11.6-11.6 31.1-8.4 38.4 6.2l9.3 18.5c4.8 9.6 14.6 15.7 25.4 15.7c15.2 0 26.1-14.6 21.7-29.2l-6-19.9c-4.6-15.4 6.9-30.9 23-30.9h2.3c13.4 0 25.9-6.7 33.3-17.8l10.7-16.1c5.6-8.5 5.3-19.6-.8-27.7l-16.1-21.5c-10.3-13.7-3.3-33.5 13.4-37.7l17-4.3c7.5-1.9 13.6-7.2 16.5-14.4l16.4-40.9C303.4 52.1 280.2 48 256 48C141.1 48 48 141.1 48 256c0 13.4 1.3 26.5 3.7 39.1zm407.7 4.6c-3-.3-6-.1-9 .8l-15.8 4.4c-6.7 1.9-13.8-.9-17.5-6.7l-2-3.1c-6-9.4-16.4-15.1-27.6-15.1s-21.6 5.7-27.6 15.1l-6.1 9.5c-1.4 2.2-3.4 4.1-5.7 5.3L312 330.1c-18.1 10.1-25.5 32.4-17 51.3l5.5 12.4c8.6 19.2 30.7 28.5 50.5 21.1l2.6-1c10-3.7 21.3-2.2 29.9 4.1l1.5 1.1c37.2-29.5 64.1-71.4 74.4-119.5zM0 256a256 256 0 1 1 512 0A256 256 0 1 1 0 256zm144.5 92.1c-2.1 8.6 3.1 17.3 11.6 19.4l32 8c8.6 2.1 17.3-3.1 19.4-11.6s-3.1-17.3-11.6-19.4l-32-8c-8.6-2.1-17.3 3.1-19.4 11.6zm92-20c-2.1 8.6 3.1 17.3 11.6 19.4s17.3-3.1 19.4-11.6l8-32c2.1-8.6-3.1-17.3-11.6-19.4s-17.3 3.1-19.4 11.6l-8 32zM343.2 113.7c-7.9-4-17.5-.7-21.5 7.2l-16 32c-4 7.9-.7 17.5 7.2 21.5s17.5 .7 21.5-7.2l16-32c4-7.9 .7-17.5-7.2-21.5z\"],\"id-card-clip\":[576,512,[\"id-card-alt\"],\"f47f\",\"M256 0h64c17.7 0 32 14.3 32 32V96c0 17.7-14.3 32-32 32H256c-17.7 0-32-14.3-32-32V32c0-17.7 14.3-32 32-32zM64 64H192v48c0 26.5 21.5 48 48 48h96c26.5 0 48-21.5 48-48V64H512c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128C0 92.7 28.7 64 64 64zM176 437.3c0 5.9 4.8 10.7 10.7 10.7H389.3c5.9 0 10.7-4.8 10.7-10.7c0-29.5-23.9-53.3-53.3-53.3H229.3c-29.5 0-53.3 23.9-53.3 53.3zM288 352a64 64 0 1 0 0-128 64 64 0 1 0 0 128z\"],\"magnifying-glass-plus\":[512,512,[\"search-plus\"],\"f00e\",\"M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zM184 296c0 13.3 10.7 24 24 24s24-10.7 24-24V232h64c13.3 0 24-10.7 24-24s-10.7-24-24-24H232V120c0-13.3-10.7-24-24-24s-24 10.7-24 24v64H120c-13.3 0-24 10.7-24 24s10.7 24 24 24h64v64z\"],\"thumbs-up\":[512,512,[128077,61575],\"f164\",\"M313.4 32.9c26 5.2 42.9 30.5 37.7 56.5l-2.3 11.4c-5.3 26.7-15.1 52.1-28.8 75.2H464c26.5 0 48 21.5 48 48c0 18.5-10.5 34.6-25.9 42.6C497 275.4 504 288.9 504 304c0 23.4-16.8 42.9-38.9 47.1c4.4 7.3 6.9 15.8 6.9 24.9c0 21.3-13.9 39.4-33.1 45.6c.7 3.3 1.1 6.8 1.1 10.4c0 26.5-21.5 48-48 48H294.5c-19 0-37.5-5.6-53.3-16.1l-38.5-25.7C176 420.4 160 390.4 160 358.3V320 272 247.1c0-29.2 13.3-56.7 36-75l7.4-5.9c26.5-21.2 44.6-51 51.2-84.2l2.3-11.4c5.2-26 30.5-42.9 56.5-37.7zM32 192H96c17.7 0 32 14.3 32 32V448c0 17.7-14.3 32-32 32H32c-17.7 0-32-14.3-32-32V224c0-17.7 14.3-32 32-32z\"],\"user-clock\":[640,512,[],\"f4fd\",\"M224 0a128 128 0 1 1 0 256A128 128 0 1 1 224 0zM178.3 304h91.4c20.6 0 40.4 3.5 58.8 9.9C323 331 320 349.1 320 368c0 59.5 29.5 112.1 74.8 144H29.7C13.3 512 0 498.7 0 482.3C0 383.8 79.8 304 178.3 304zM352 368a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm144-80c-8.8 0-16 7.2-16 16v64c0 8.8 7.2 16 16 16h48c8.8 0 16-7.2 16-16s-7.2-16-16-16H512V304c0-8.8-7.2-16-16-16z\"],\"hand-dots\":[512,512,[\"allergies\"],\"f461\",\"M288 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V336c0 1.5 0 3.1 .1 4.6L67.6 283c-16-15.2-41.3-14.6-56.6 1.4s-14.6 41.3 1.4 56.6L124.8 448c43.1 41.1 100.4 64 160 64H304c97.2 0 176-78.8 176-176V128c0-17.7-14.3-32-32-32s-32 14.3-32 32V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V64c0-17.7-14.3-32-32-32s-32 14.3-32 32V240c0 8.8-7.2 16-16 16s-16-7.2-16-16V32zM240 336a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zm80 16a16 16 0 1 1 0 32 16 16 0 1 1 0-32zm48-16a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zm-16 80a16 16 0 1 1 0 32 16 16 0 1 1 0-32zM240 432a16 16 0 1 1 32 0 16 16 0 1 1 -32 0zm-48-48a16 16 0 1 1 0 32 16 16 0 1 1 0-32z\"],\"file-invoice\":[384,512,[],\"f570\",\"M64 0C28.7 0 0 28.7 0 64V448c0 35.3 28.7 64 64 64H320c35.3 0 64-28.7 64-64V160H256c-17.7 0-32-14.3-32-32V0H64zM256 0V128H384L256 0zM80 64h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16s7.2-16 16-16zm0 64h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H80c-8.8 0-16-7.2-16-16s7.2-16 16-16zm16 96H288c17.7 0 32 14.3 32 32v64c0 17.7-14.3 32-32 32H96c-17.7 0-32-14.3-32-32V256c0-17.7 14.3-32 32-32zm0 32v64H288V256H96zM240 416h64c8.8 0 16 7.2 16 16s-7.2 16-16 16H240c-8.8 0-16-7.2-16-16s7.2-16 16-16z\"],\"window-minimize\":[512,512,[128469],\"f2d1\",\"M32 416c-17.7 0-32 14.3-32 32s14.3 32 32 32H480c17.7 0 32-14.3 32-32s-14.3-32-32-32H32z\"],\"mug-saucer\":[640,512,[\"coffee\"],\"f0f4\",\"M96 64c0-17.7 14.3-32 32-32H448h64c70.7 0 128 57.3 128 128s-57.3 128-128 128H480c0 53-43 96-96 96H192c-53 0-96-43-96-96V64zM480 224h32c35.3 0 64-28.7 64-64s-28.7-64-64-64H480V224zM32 416H544c17.7 0 32 14.3 32 32s-14.3 32-32 32H32c-17.7 0-32-14.3-32-32s14.3-32 32-32z\"],brush:[384,512,[],\"f55d\",\"M162.4 6c-1.5-3.6-5-6-8.9-6h-19c-3.9 0-7.5 2.4-8.9 6L104.9 57.7c-3.2 8-14.6 8-17.8 0L66.4 6c-1.5-3.6-5-6-8.9-6H48C21.5 0 0 21.5 0 48V224v22.4V256H9.6 374.4 384v-9.6V224 48c0-26.5-21.5-48-48-48H230.5c-3.9 0-7.5 2.4-8.9 6L200.9 57.7c-3.2 8-14.6 8-17.8 0L162.4 6zM0 288v32c0 35.3 28.7 64 64 64h64v64c0 35.3 28.7 64 64 64s64-28.7 64-64V384h64c35.3 0 64-28.7 64-64V288H0zM192 432a16 16 0 1 1 0 32 16 16 0 1 1 0-32z\"],mask:[576,512,[],\"f6fa\",\"M288 64C64 64 0 160 0 272S80 448 176 448h8.4c24.2 0 46.4-13.7 57.2-35.4l23.2-46.3c4.4-8.8 13.3-14.3 23.2-14.3s18.8 5.5 23.2 14.3l23.2 46.3c10.8 21.7 33 35.4 57.2 35.4H400c96 0 176-64 176-176s-64-208-288-208zM96 256a64 64 0 1 1 128 0A64 64 0 1 1 96 256zm320-64a64 64 0 1 1 0 128 64 64 0 1 1 0-128z\"],\"magnifying-glass-minus\":[512,512,[\"search-minus\"],\"f010\",\"M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zM136 184c-13.3 0-24 10.7-24 24s10.7 24 24 24H280c13.3 0 24-10.7 24-24s-10.7-24-24-24H136z\"],\"ruler-vertical\":[256,512,[],\"f548\",\"M0 48C0 21.5 21.5 0 48 0H208c26.5 0 48 21.5 48 48V96H176c-8.8 0-16 7.2-16 16s7.2 16 16 16h80v64H176c-8.8 0-16 7.2-16 16s7.2 16 16 16h80v64H176c-8.8 0-16 7.2-16 16s7.2 16 16 16h80v64H176c-8.8 0-16 7.2-16 16s7.2 16 16 16h80v48c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V48z\"],\"user-large\":[512,512,[\"user-alt\"],\"f406\",\"M256 288A144 144 0 1 0 256 0a144 144 0 1 0 0 288zm-94.7 32C72.2 320 0 392.2 0 481.3c0 17 13.8 30.7 30.7 30.7H481.3c17 0 30.7-13.8 30.7-30.7C512 392.2 439.8 320 350.7 320H161.3z\"],\"train-tram\":[448,512,[128650],\"e5b4\",\"M86.8 48c-12.2 0-23.6 5.5-31.2 15L42.7 79C34.5 89.3 19.4 91 9 82.7S-3 59.4 5.3 49L18 33C34.7 12.2 60 0 86.8 0H361.2c26.7 0 52 12.2 68.7 33l12.8 16c8.3 10.4 6.6 25.5-3.8 33.7s-25.5 6.6-33.7-3.7L392.5 63c-7.6-9.5-19.1-15-31.2-15H248V96h40c53 0 96 43 96 96V352c0 30.6-14.3 57.8-36.6 75.4l65.5 65.5c7.1 7.1 2.1 19.1-7.9 19.1H365.3c-8.5 0-16.6-3.4-22.6-9.4L288 448H160l-54.6 54.6c-6 6-14.1 9.4-22.6 9.4H43c-10 0-15-12.1-7.9-19.1l65.5-65.5C78.3 409.8 64 382.6 64 352V192c0-53 43-96 96-96h40V48H86.8zM160 160c-17.7 0-32 14.3-32 32v32c0 17.7 14.3 32 32 32H288c17.7 0 32-14.3 32-32V192c0-17.7-14.3-32-32-32H160zm32 192a32 32 0 1 0 -64 0 32 32 0 1 0 64 0zm96 32a32 32 0 1 0 0-64 32 32 0 1 0 0 64z\"],\"user-nurse\":[448,512,[],\"f82f\",\"M96 128V70.2c0-13.3 8.3-25.3 20.8-30l96-36c7.2-2.7 15.2-2.7 22.5 0l96 36c12.5 4.7 20.8 16.6 20.8 30V128h-.3c.2 2.6 .3 5.3 .3 8v40c0 70.7-57.3 128-128 128s-128-57.3-128-128V136c0-2.7 .1-5.4 .3-8H96zm48 48c0 44.2 35.8 80 80 80s80-35.8 80-80V160H144v16zM111.9 327.7c10.5-3.4 21.8 .4 29.4 8.5l71 75.5c6.3 6.7 17 6.7 23.3 0l71-75.5c7.6-8.1 18.9-11.9 29.4-8.5C401 348.6 448 409.4 448 481.3c0 17-13.8 30.7-30.7 30.7H30.7C13.8 512 0 498.2 0 481.3c0-71.9 47-132.7 111.9-153.6zM208 48V64H192c-4.4 0-8 3.6-8 8V88c0 4.4 3.6 8 8 8h16v16c0 4.4 3.6 8 8 8h16c4.4 0 8-3.6 8-8V96h16c4.4 0 8-3.6 8-8V72c0-4.4-3.6-8-8-8H240V48c0-4.4-3.6-8-8-8H216c-4.4 0-8 3.6-8 8z\"],syringe:[512,512,[128137],\"f48e\",\"M441 7l32 32 32 32c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-15-15L417.9 128l55 55c9.4 9.4 9.4 24.6 0 33.9s-24.6 9.4-33.9 0l-72-72L295 73c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0l55 55L422.1 56 407 41c-9.4-9.4-9.4-24.6 0-33.9s24.6-9.4 33.9 0zM210.3 155.7l61.1-61.1c.3 .3 .6 .7 1 1l16 16 56 56 56 56 16 16c.3 .3 .6 .6 1 1l-191 191c-10.5 10.5-24.7 16.4-39.6 16.4H97.9L41 505c-9.4 9.4-24.6 9.4-33.9 0s-9.4-24.6 0-33.9l57-57V325.3c0-14.9 5.9-29.1 16.4-39.6l43.3-43.3 57 57c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-57-57 41.4-41.4 57 57c6.2 6.2 16.4 6.2 22.6 0s6.2-16.4 0-22.6l-57-57z\"],\"cloud-sun\":[640,512,[9925],\"f6c4\",\"M294.2 1.2c5.1 2.1 8.7 6.7 9.6 12.1l14.1 84.7 84.7 14.1c5.4 .9 10 4.5 12.1 9.6s1.5 10.9-1.6 15.4l-38.5 55c-2.2-.1-4.4-.2-6.7-.2c-23.3 0-45.1 6.2-64 17.1l0-1.1c0-53-43-96-96-96s-96 43-96 96s43 96 96 96c8.1 0 15.9-1 23.4-2.9c-36.6 18.1-63.3 53.1-69.8 94.9l-24.4 17c-4.5 3.1-10.3 3.8-15.4 1.6s-8.7-6.7-9.6-12.1L98.1 317.9 13.4 303.8c-5.4-.9-10-4.5-12.1-9.6s-1.5-10.9 1.6-15.4L52.5 208 2.9 137.2c-3.2-4.5-3.8-10.3-1.6-15.4s6.7-8.7 12.1-9.6L98.1 98.1l14.1-84.7c.9-5.4 4.5-10 9.6-12.1s10.9-1.5 15.4 1.6L208 52.5 278.8 2.9c4.5-3.2 10.3-3.8 15.4-1.6zM144 208a64 64 0 1 1 128 0 64 64 0 1 1 -128 0zM639.9 431.9c0 44.2-35.8 80-80 80H288c-53 0-96-43-96-96c0-47.6 34.6-87 80-94.6l0-1.3c0-53 43-96 96-96c34.9 0 65.4 18.6 82.2 46.4c13-9.1 28.8-14.4 45.8-14.4c44.2 0 80 35.8 80 80c0 5.9-.6 11.7-1.9 17.2c37.4 6.7 65.8 39.4 65.8 78.7z\"],\"stopwatch-20\":[448,512,[],\"e06f\",\"M176 0c-17.7 0-32 14.3-32 32s14.3 32 32 32h16V98.4C92.3 113.8 16 200 16 304c0 114.9 93.1 208 208 208s208-93.1 208-208c0-41.8-12.3-80.7-33.5-113.2l24.1-24.1c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0L355.7 143c-28.1-23-62.2-38.8-99.7-44.6V64h16c17.7 0 32-14.3 32-32s-14.3-32-32-32H176zM288 204c28.7 0 52 23.3 52 52v96c0 28.7-23.3 52-52 52s-52-23.3-52-52V256c0-28.7 23.3-52 52-52zm-12 52v96c0 6.6 5.4 12 12 12s12-5.4 12-12V256c0-6.6-5.4-12-12-12s-12 5.4-12 12zM159.5 244c-5.4 0-10.2 3.5-11.9 8.6l-.6 1.7c-3.5 10.5-14.8 16.1-25.3 12.6s-16.1-14.8-12.6-25.3l.6-1.7c7.2-21.5 27.2-35.9 49.8-35.9c29 0 52.5 23.5 52.5 52.5v2.2c0 13.4-4.9 26.4-13.8 36.4l-39 43.9c-6.2 7-10 15.7-10.9 24.9H192c11 0 20 9 20 20s-9 20-20 20H128c-11 0-20-9-20-20V368.3c0-20.6 7.5-40.4 21.2-55.8l39-43.9c2.4-2.7 3.7-6.2 3.7-9.8v-2.2c0-6.9-5.6-12.5-12.5-12.5z\"],\"square-full\":[512,512,[128997,128998,128999,129e3,129001,129002,129003,11035,11036],\"f45c\",\"M0 0H512V512H0V0z\"],magnet:[448,512,[129522],\"f076\",\"M0 160v96C0 379.7 100.3 480 224 480s224-100.3 224-224V160H320v96c0 53-43 96-96 96s-96-43-96-96V160H0zm0-32H128V64c0-17.7-14.3-32-32-32H32C14.3 32 0 46.3 0 64v64zm320 0H448V64c0-17.7-14.3-32-32-32H352c-17.7 0-32 14.3-32 32v64z\"],jar:[320,512,[],\"e516\",\"M32 32C32 14.3 46.3 0 64 0H256c17.7 0 32 14.3 32 32s-14.3 32-32 32H64C46.3 64 32 49.7 32 32zM0 160c0-35.3 28.7-64 64-64H256c35.3 0 64 28.7 64 64V448c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V160zm96 64c-17.7 0-32 14.3-32 32v96c0 17.7 14.3 32 32 32H224c17.7 0 32-14.3 32-32V256c0-17.7-14.3-32-32-32H96z\"],\"note-sticky\":[448,512,[62026,\"sticky-note\"],\"f249\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H288V368c0-26.5 21.5-48 48-48H448V96c0-35.3-28.7-64-64-64H64zM448 352H402.7 336c-8.8 0-16 7.2-16 16v66.7V480l32-32 64-64 32-32z\"],\"bug-slash\":[640,512,[],\"e490\",\"M38.8 5.1C28.4-3.1 13.3-1.2 5.1 9.2S-1.2 34.7 9.2 42.9l592 464c10.4 8.2 25.5 6.3 33.7-4.1s6.3-25.5-4.1-33.7L477.4 348.9c1.7-9.4 2.6-19 2.6-28.9h64c17.7 0 32-14.3 32-32s-14.3-32-32-32H479.7c-1.1-14.1-5-27.5-11.1-39.5c.7-.6 1.4-1.2 2.1-1.9l64-64c12.5-12.5 12.5-32.8 0-45.3s-32.8-12.5-45.3 0l-64 64c-.7 .7-1.3 1.4-1.9 2.1C409.2 164.1 393.1 160 376 160H264c-8.3 0-16.3 1-24 2.8L38.8 5.1zm392 430.3L336 360.7V479.2c36.6-3.6 69.7-19.6 94.8-43.8zM166.7 227.3c-3.4 9-5.6 18.7-6.4 28.7H96c-17.7 0-32 14.3-32 32s14.3 32 32 32h64c0 24.6 5.5 47.8 15.4 68.6c-2.2 1.3-4.2 2.9-6 4.8l-64 64c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0l63.1-63.1c24.5 21.8 55.8 36.2 90.3 39.6V335.5L166.7 227.3zM320 0c-53 0-96 43-96 96v3.6c0 15.7 12.7 28.4 28.4 28.4H387.6c15.7 0 28.4-12.7 28.4-28.4V96c0-53-43-96-96-96z\"],\"arrow-up-from-water-pump\":[576,512,[],\"e4b6\",\"M112 0C85.5 0 64 21.5 64 48V256H48c-26.5 0-48 21.5-48 48v96c0 8 2 15.6 5.4 22.2c3.8-1.7 7.8-3.1 12-4.1c13.1-3.1 26.7-9.8 37.3-18.6c22.2-18.7 54.3-20.1 78.1-3.4c18 12.4 40.1 20.3 59.2 20.3c21.1 0 42-8.5 59.2-20.3c22.1-15.5 51.6-15.5 73.7 0c18.4 12.7 39.6 20.3 59.2 20.3c19 0 41.2-7.9 59.2-20.3c23.8-16.7 55.8-15.3 78.1 3.4c10.6 8.8 24.2 15.6 37.3 18.6c4.2 1 8.2 2.4 12 4.1C574 415.6 576 408 576 400V304c0-26.5-21.5-48-48-48H480l0-146.7 25.4 25.4c12.5 12.5 32.8 12.5 45.3 0s12.5-32.8 0-45.3l-80-80c-12.5-12.5-32.8-12.5-45.3 0l-80 80c-12.5 12.5-12.5 32.8 0 45.3s32.8 12.5 45.3 0L416 109.3 416 256H288V48c0-26.5-21.5-48-48-48H112zM306.5 421.9c-11.1-7.9-25.9-7.9-37 0C247 437.4 219.5 448 192 448c-26.9 0-55.3-10.8-77.4-26.1l0 0c-11.9-8.5-28.1-7.8-39.2 1.7c-14.4 11.9-32.5 21-50.6 25.2c-17.2 4-27.9 21.2-23.9 38.4s21.2 27.9 38.4 23.9c24.5-5.7 44.9-16.5 58.2-25C126.5 501.7 159 512 192 512c31.9 0 60.6-9.9 80.4-18.9c5.8-2.7 11.1-5.3 15.6-7.7c4.5 2.4 9.7 5.1 15.6 7.7c19.8 9 48.5 18.9 80.4 18.9c33 0 65.5-10.3 94.5-25.8c13.4 8.4 33.7 19.3 58.2 25c17.2 4 34.4-6.7 38.4-23.9s-6.7-34.4-23.9-38.4c-18.1-4.2-36.2-13.3-50.6-25.2c-11.1-9.4-27.3-10.1-39.2-1.7l0 0C439.4 437.2 410.9 448 384 448c-27.5 0-55-10.6-77.5-26.1z\"],bone:[576,512,[129460],\"f5d7\",\"M153.7 144.8c6.9 16.3 20.6 31.2 38.3 31.2H384c17.7 0 31.4-14.9 38.3-31.2C434.4 116.1 462.9 96 496 96c44.2 0 80 35.8 80 80c0 30.4-17 56.9-42 70.4c-3.6 1.9-6 5.5-6 9.6s2.4 7.7 6 9.6c25 13.5 42 40 42 70.4c0 44.2-35.8 80-80 80c-33.1 0-61.6-20.1-73.7-48.8C415.4 350.9 401.7 336 384 336H192c-17.7 0-31.4 14.9-38.3 31.2C141.6 395.9 113.1 416 80 416c-44.2 0-80-35.8-80-80c0-30.4 17-56.9 42-70.4c3.6-1.9 6-5.5 6-9.6s-2.4-7.7-6-9.6C17 232.9 0 206.4 0 176c0-44.2 35.8-80 80-80c33.1 0 61.6 20.1 73.7 48.8z\"],\"user-injured\":[448,512,[],\"f728\",\"M240 80H342.7c-7.9-19.5-20.4-36.5-36.2-49.9L240 80zm37.7-68.2C261.3 4.2 243.2 0 224 0c-53.7 0-99.7 33.1-118.7 80h81.4l91-68.2zM224 256c70.7 0 128-57.3 128-128c0-5.4-.3-10.8-1-16H97c-.7 5.2-1 10.6-1 16c0 70.7 57.3 128 128 128zM124 312.4c-9.7 3.1-19.1 7-28 11.7V512H243.7L181.5 408.2 124 312.4zm33-7.2L204.3 384H272c44.2 0 80 35.8 80 80c0 18-6 34.6-16 48h82.3c16.4 0 29.7-13.3 29.7-29.7C448 383.8 368.2 304 269.7 304H178.3c-7.2 0-14.3 .4-21.3 1.3zM0 482.3C0 498.7 13.3 512 29.7 512H64V345.4C24.9 378.1 0 427.3 0 482.3zM320 464c0-26.5-21.5-48-48-48H223.5l57.1 95.2C303 507.2 320 487.6 320 464z\"],\"face-sad-tear\":[512,512,[128546,\"sad-tear\"],\"f5b4\",\"M0 256a256 256 0 1 0 512 0A256 256 0 1 0 0 256zm240 80c0-8.8 7.2-16 16-16c45 0 85.6 20.5 115.7 53.1c6 6.5 5.6 16.6-.9 22.6s-16.6 5.6-22.6-.9c-25-27.1-57.4-42.9-92.3-42.9c-8.8 0-16-7.2-16-16zm-80 80c-26.5 0-48-21-48-47c0-20 28.6-60.4 41.6-77.7c3.2-4.4 9.6-4.4 12.8 0C179.6 308.6 208 349 208 369c0 26-21.5 47-48 47zM367.6 208a32 32 0 1 1 -64 0 32 32 0 1 1 64 0zm-192-32a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],plane:[576,512,[],\"f072\",\"M482.3 192c34.2 0 93.7 29 93.7 64c0 36-59.5 64-93.7 64l-116.6 0L265.2 495.9c-5.7 10-16.3 16.1-27.8 16.1l-56.2 0c-10.6 0-18.3-10.2-15.4-20.4l49-171.6L112 320 68.8 377.6c-3 4-7.8 6.4-12.8 6.4l-42 0c-7.8 0-14-6.3-14-14c0-1.3 .2-2.6 .5-3.9L32 256 .5 145.9c-.4-1.3-.5-2.6-.5-3.9c0-7.8 6.3-14 14-14l42 0c5 0 9.8 2.4 12.8 6.4L112 192l102.9 0-49-171.6C162.9 10.2 170.6 0 181.2 0l56.2 0c11.5 0 22.1 6.2 27.8 16.1L365.7 192l116.6 0z\"],\"tent-arrows-down\":[576,512,[],\"e581\",\"M209.8 111.9c-8.9-9.9-24-10.7-33.9-1.8l-39.9 36L136 24c0-13.3-10.7-24-24-24S88 10.7 88 24l0 122.1-39.9-36c-9.9-8.9-25-8.1-33.9 1.8s-8.1 25 1.8 33.9l80 72c9.1 8.2 23 8.2 32.1 0l80-72c9.9-8.9 10.7-24 1.8-33.9zm352 0c-8.9-9.9-24-10.7-33.9-1.8l-39.9 36V24c0-13.3-10.7-24-24-24s-24 10.7-24 24V146.1l-39.9-36c-9.9-8.9-25-8.1-33.9 1.8s-8.1 25 1.8 33.9l80 72c9.1 8.2 23 8.2 32.1 0l80-72c9.9-8.9 10.7-24 1.8-33.9zM307.4 166.5c-11.5-8.7-27.3-8.7-38.8 0l-168 128c-6.6 5-11 12.5-12.3 20.7l-24 160c-1.4 9.2 1.3 18.6 7.4 25.6S86.7 512 96 512H240h16c17.7 0 32-14.3 32-32V361.9c0-5.5 4.4-9.9 9.9-9.9c3.7 0 7.2 2.1 8.8 5.5l68.4 136.8c5.4 10.8 16.5 17.7 28.6 17.7H464h16c9.3 0 18.2-4.1 24.2-11.1s8.8-16.4 7.4-25.6l-24-160c-1.2-8.2-5.6-15.7-12.3-20.7l-168-128z\"],exclamation:[64,512,[10069,10071,61738],\"21\",\"M64 64c0-17.7-14.3-32-32-32S0 46.3 0 64V320c0 17.7 14.3 32 32 32s32-14.3 32-32V64zM32 480a40 40 0 1 0 0-80 40 40 0 1 0 0 80z\"],\"arrows-spin\":[512,512,[],\"e4bb\",\"M256 96c38.4 0 73.7 13.5 101.3 36.1l-32.6 32.6c-4.6 4.6-5.9 11.5-3.5 17.4s8.3 9.9 14.8 9.9H448c8.8 0 16-7.2 16-16V64c0-6.5-3.9-12.3-9.9-14.8s-12.9-1.1-17.4 3.5l-34 34C363.4 52.6 312.1 32 256 32c-10.9 0-21.5 .8-32 2.3V99.2c10.3-2.1 21-3.2 32-3.2zM132.1 154.7l32.6 32.6c4.6 4.6 11.5 5.9 17.4 3.5s9.9-8.3 9.9-14.8V64c0-8.8-7.2-16-16-16H64c-6.5 0-12.3 3.9-14.8 9.9s-1.1 12.9 3.5 17.4l34 34C52.6 148.6 32 199.9 32 256c0 10.9 .8 21.5 2.3 32H99.2c-2.1-10.3-3.2-21-3.2-32c0-38.4 13.5-73.7 36.1-101.3zM477.7 224H412.8c2.1 10.3 3.2 21 3.2 32c0 38.4-13.5 73.7-36.1 101.3l-32.6-32.6c-4.6-4.6-11.5-5.9-17.4-3.5s-9.9 8.3-9.9 14.8V448c0 8.8 7.2 16 16 16H448c6.5 0 12.3-3.9 14.8-9.9s1.1-12.9-3.5-17.4l-34-34C459.4 363.4 480 312.1 480 256c0-10.9-.8-21.5-2.3-32zM256 416c-38.4 0-73.7-13.5-101.3-36.1l32.6-32.6c4.6-4.6 5.9-11.5 3.5-17.4s-8.3-9.9-14.8-9.9H64c-8.8 0-16 7.2-16 16l0 112c0 6.5 3.9 12.3 9.9 14.8s12.9 1.1 17.4-3.5l34-34C148.6 459.4 199.9 480 256 480c10.9 0 21.5-.8 32-2.3V412.8c-10.3 2.1-21 3.2-32 3.2z\"],print:[512,512,[128424,128438,9113],\"f02f\",\"M128 0C92.7 0 64 28.7 64 64v96h64V64H354.7L384 93.3V160h64V93.3c0-17-6.7-33.3-18.7-45.3L400 18.7C388 6.7 371.7 0 354.7 0H128zM384 352v32 64H128V384 368 352H384zm64 32h32c17.7 0 32-14.3 32-32V256c0-35.3-28.7-64-64-64H64c-35.3 0-64 28.7-64 64v96c0 17.7 14.3 32 32 32H64v64c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V384zM432 248a24 24 0 1 1 0 48 24 24 0 1 1 0-48z\"],\"turkish-lira-sign\":[384,512,[\"try\",\"turkish-lira\"],\"e2bb\",\"M96 32c17.7 0 32 14.3 32 32V99.3L247.2 65.2c17-4.9 34.7 5 39.6 22s-5 34.7-22 39.6L128 165.9v29.4l119.2-34.1c17-4.9 34.7 5 39.6 22s-5 34.7-22 39.6L128 261.9V416h63.8c68.2 0 124.4-53.5 127.8-121.6l.4-8c.9-17.7 15.9-31.2 33.6-30.4s31.2 15.9 30.4 33.6l-.4 8C378.5 399.8 294.1 480 191.8 480H96c-17.7 0-32-14.3-32-32V280.1l-23.2 6.6c-17 4.9-34.7-5-39.6-22s5-34.7 22-39.6L64 213.6V184.1l-23.2 6.6c-17 4.9-34.7-5-39.6-22s5-34.7 22-39.6L64 117.6V64c0-17.7 14.3-32 32-32z\"],\"dollar-sign\":[320,512,[128178,61781,\"dollar\",\"usd\"],\"24\",\"M160 0c17.7 0 32 14.3 32 32V67.7c1.6 .2 3.1 .4 4.7 .7c.4 .1 .7 .1 1.1 .2l48 8.8c17.4 3.2 28.9 19.9 25.7 37.2s-19.9 28.9-37.2 25.7l-47.5-8.7c-31.3-4.6-58.9-1.5-78.3 6.2s-27.2 18.3-29 28.1c-2 10.7-.5 16.7 1.2 20.4c1.8 3.9 5.5 8.3 12.8 13.2c16.3 10.7 41.3 17.7 73.7 26.3l2.9 .8c28.6 7.6 63.6 16.8 89.6 33.8c14.2 9.3 27.6 21.9 35.9 39.5c8.5 17.9 10.3 37.9 6.4 59.2c-6.9 38-33.1 63.4-65.6 76.7c-13.7 5.6-28.6 9.2-44.4 11V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V445.1c-.4-.1-.9-.1-1.3-.2l-.2 0 0 0c-24.4-3.8-64.5-14.3-91.5-26.3c-16.1-7.2-23.4-26.1-16.2-42.2s26.1-23.4 42.2-16.2c20.9 9.3 55.3 18.5 75.2 21.6c31.9 4.7 58.2 2 76-5.3c16.9-6.9 24.6-16.9 26.8-28.9c1.9-10.6 .4-16.7-1.3-20.4c-1.9-4-5.6-8.4-13-13.3c-16.4-10.7-41.5-17.7-74-26.3l-2.8-.7 0 0C119.4 279.3 84.4 270 58.4 253c-14.2-9.3-27.5-22-35.8-39.6c-8.4-17.9-10.1-37.9-6.1-59.2C23.7 116 52.3 91.2 84.8 78.3c13.3-5.3 27.9-8.9 43.2-11V32c0-17.7 14.3-32 32-32z\"],x:[384,512,[120],\"58\",\"M376.6 84.5c11.3-13.6 9.5-33.8-4.1-45.1s-33.8-9.5-45.1 4.1L192 206 56.6 43.5C45.3 29.9 25.1 28.1 11.5 39.4S-3.9 70.9 7.4 84.5L150.3 256 7.4 427.5c-11.3 13.6-9.5 33.8 4.1 45.1s33.8 9.5 45.1-4.1L192 306 327.4 468.5c11.3 13.6 31.5 15.4 45.1 4.1s15.4-31.5 4.1-45.1L233.7 256 376.6 84.5z\"],\"magnifying-glass-dollar\":[512,512,[\"search-dollar\"],\"f688\",\"M416 208c0 45.9-14.9 88.3-40 122.7L502.6 457.4c12.5 12.5 12.5 32.8 0 45.3s-32.8 12.5-45.3 0L330.7 376c-34.4 25.2-76.8 40-122.7 40C93.1 416 0 322.9 0 208S93.1 0 208 0S416 93.1 416 208zM228 104c0-11-9-20-20-20s-20 9-20 20v14c-7.6 1.7-15.2 4.4-22.2 8.5c-13.9 8.3-25.9 22.8-25.8 43.9c.1 20.3 12 33.1 24.7 40.7c11 6.6 24.7 10.8 35.6 14l1.7 .5c12.6 3.8 21.8 6.8 28 10.7c5.1 3.2 5.8 5.4 5.9 8.2c.1 5-1.8 8-5.9 10.5c-5 3.1-12.9 5-21.4 4.7c-11.1-.4-21.5-3.9-35.1-8.5c-2.3-.8-4.7-1.6-7.2-2.4c-10.5-3.5-21.8 2.2-25.3 12.6s2.2 21.8 12.6 25.3c1.9 .6 4 1.3 6.1 2.1l0 0 0 0c8.3 2.9 17.9 6.2 28.2 8.4V312c0 11 9 20 20 20s20-9 20-20V298.2c8-1.7 16-4.5 23.2-9c14.3-8.9 25.1-24.1 24.8-45c-.3-20.3-11.7-33.4-24.6-41.6c-11.5-7.2-25.9-11.6-37.1-15l-.7-.2c-12.8-3.9-21.9-6.7-28.3-10.5c-5.2-3.1-5.3-4.9-5.3-6.7c0-3.7 1.4-6.5 6.2-9.3c5.4-3.2 13.6-5.1 21.5-5c9.6 .1 20.2 2.2 31.2 5.2c10.7 2.8 21.6-3.5 24.5-14.2s-3.5-21.6-14.2-24.5c-6.5-1.7-13.7-3.4-21.1-4.7V104z\"],\"users-gear\":[640,512,[\"users-cog\"],\"f509\",\"M144 160A80 80 0 1 0 144 0a80 80 0 1 0 0 160zm368 0A80 80 0 1 0 512 0a80 80 0 1 0 0 160zM0 298.7C0 310.4 9.6 320 21.3 320H234.7c.2 0 .4 0 .7 0c-26.6-23.5-43.3-57.8-43.3-96c0-7.6 .7-15 1.9-22.3c-13.6-6.3-28.7-9.7-44.6-9.7H106.7C47.8 192 0 239.8 0 298.7zM320 320c24 0 45.9-8.8 62.7-23.3c2.5-3.7 5.2-7.3 8-10.7c2.7-3.3 5.7-6.1 9-8.3C410 262.3 416 243.9 416 224c0-53-43-96-96-96s-96 43-96 96s43 96 96 96zm65.4 60.2c-10.3-5.9-18.1-16.2-20.8-28.2H261.3C187.7 352 128 411.7 128 485.3c0 14.7 11.9 26.7 26.7 26.7H455.2c-2.1-5.2-3.2-10.9-3.2-16.4v-3c-1.3-.7-2.7-1.5-4-2.3l-2.6 1.5c-16.8 9.7-40.5 8-54.7-9.7c-4.5-5.6-8.6-11.5-12.4-17.6l-.1-.2-.1-.2-2.4-4.1-.1-.2-.1-.2c-3.4-6.2-6.4-12.6-9-19.3c-8.2-21.2 2.2-42.6 19-52.3l2.7-1.5c0-.8 0-1.5 0-2.3s0-1.5 0-2.3l-2.7-1.5zM533.3 192H490.7c-15.9 0-31 3.5-44.6 9.7c1.3 7.2 1.9 14.7 1.9 22.3c0 17.4-3.5 33.9-9.7 49c2.5 .9 4.9 2 7.1 3.3l2.6 1.5c1.3-.8 2.6-1.6 4-2.3v-3c0-19.4 13.3-39.1 35.8-42.6c7.9-1.2 16-1.9 24.2-1.9s16.3 .6 24.2 1.9c22.5 3.5 35.8 23.2 35.8 42.6v3c1.3 .7 2.7 1.5 4 2.3l2.6-1.5c16.8-9.7 40.5-8 54.7 9.7c2.3 2.8 4.5 5.8 6.6 8.7c-2.1-57.1-49-102.7-106.6-102.7zm91.3 163.9c6.3-3.6 9.5-11.1 6.8-18c-2.1-5.5-4.6-10.8-7.4-15.9l-2.3-4c-3.1-5.1-6.5-9.9-10.2-14.5c-4.6-5.7-12.7-6.7-19-3l-2.9 1.7c-9.2 5.3-20.4 4-29.6-1.3s-16.1-14.5-16.1-25.1v-3.4c0-7.3-4.9-13.8-12.1-14.9c-6.5-1-13.1-1.5-19.9-1.5s-13.4 .5-19.9 1.5c-7.2 1.1-12.1 7.6-12.1 14.9v3.4c0 10.6-6.9 19.8-16.1 25.1s-20.4 6.6-29.6 1.3l-2.9-1.7c-6.3-3.6-14.4-2.6-19 3c-3.7 4.6-7.1 9.5-10.2 14.6l-2.3 3.9c-2.8 5.1-5.3 10.4-7.4 15.9c-2.6 6.8 .5 14.3 6.8 17.9l2.9 1.7c9.2 5.3 13.7 15.8 13.7 26.4s-4.5 21.1-13.7 26.4l-3 1.7c-6.3 3.6-9.5 11.1-6.8 17.9c2.1 5.5 4.6 10.7 7.4 15.8l2.4 4.1c3 5.1 6.4 9.9 10.1 14.5c4.6 5.7 12.7 6.7 19 3l2.9-1.7c9.2-5.3 20.4-4 29.6 1.3s16.1 14.5 16.1 25.1v3.4c0 7.3 4.9 13.8 12.1 14.9c6.5 1 13.1 1.5 19.9 1.5s13.4-.5 19.9-1.5c7.2-1.1 12.1-7.6 12.1-14.9v-3.4c0-10.6 6.9-19.8 16.1-25.1s20.4-6.6 29.6-1.3l2.9 1.7c6.3 3.6 14.4 2.6 19-3c3.7-4.6 7.1-9.4 10.1-14.5l2.4-4.2c2.8-5.1 5.3-10.3 7.4-15.8c2.6-6.8-.5-14.3-6.8-17.9l-3-1.7c-9.2-5.3-13.7-15.8-13.7-26.4s4.5-21.1 13.7-26.4l3-1.7zM472 384a40 40 0 1 1 80 0 40 40 0 1 1 -80 0z\"],\"person-military-pointing\":[576,512,[],\"e54a\",\"M246.9 14.1C234 15.2 224 26 224 39c0 13.8 11.2 25 25 25H400c8.8 0 16-7.2 16-16V17.4C416 8 408 .7 398.7 1.4L246.9 14.1zM240 112c0 44.2 35.8 80 80 80s80-35.8 80-80c0-5.5-.6-10.8-1.6-16H241.6c-1 5.2-1.6 10.5-1.6 16zM72 224c-22.1 0-40 17.9-40 40s17.9 40 40 40H224v89.4L386.8 230.5c-13.3-4.3-27.3-6.5-41.6-6.5H240 72zm345.7 20.9L246.6 416H416V369.7l53.6 90.6c11.2 19 35.8 25.3 54.8 14.1s25.3-35.8 14.1-54.8L462.3 290.8c-11.2-18.9-26.6-34.5-44.6-45.9zM224 448v32c0 17.7 14.3 32 32 32H384c17.7 0 32-14.3 32-32V448H224z\"],\"building-columns\":[512,512,[\"bank\",\"institution\",\"museum\",\"university\"],\"f19c\",\"M243.4 2.6l-224 96c-14 6-21.8 21-18.7 35.8S16.8 160 32 160v8c0 13.3 10.7 24 24 24H456c13.3 0 24-10.7 24-24v-8c15.2 0 28.3-10.7 31.3-25.6s-4.8-29.9-18.7-35.8l-224-96c-8-3.4-17.2-3.4-25.2 0zM128 224H64V420.3c-.6 .3-1.2 .7-1.8 1.1l-48 32c-11.7 7.8-17 22.4-12.9 35.9S17.9 512 32 512H480c14.1 0 26.5-9.2 30.6-22.7s-1.1-28.1-12.9-35.9l-48-32c-.6-.4-1.2-.7-1.8-1.1V224H384V416H344V224H280V416H232V224H168V416H128V224zM256 64a32 32 0 1 1 0 64 32 32 0 1 1 0-64z\"],umbrella:[576,512,[],\"f0e9\",\"M288 0c17.7 0 32 14.3 32 32V49.7C451.8 63.4 557.7 161 573.9 285.9c2 15.6-17.3 24.4-27.8 12.7C532.1 283 504.8 272 480 272c-38.7 0-71 27.5-78.4 64.1c-1.7 8.7-8.7 15.9-17.6 15.9s-15.8-7.2-17.6-15.9C359 299.5 326.7 272 288 272s-71 27.5-78.4 64.1c-1.7 8.7-8.7 15.9-17.6 15.9s-15.8-7.2-17.6-15.9C167 299.5 134.7 272 96 272c-24.8 0-52.1 11-66.1 26.7C19.4 310.4 .1 301.5 2.1 285.9C18.3 161 124.2 63.4 256 49.7V32c0-17.7 14.3-32 32-32zm0 304c12.3 0 23.5 4.6 32 12.2V430.6c0 45-36.5 81.4-81.4 81.4c-30.8 0-59-17.4-72.8-45l-2.3-4.7c-7.9-15.8-1.5-35 14.3-42.9s35-1.5 42.9 14.3l2.3 4.7c3 5.9 9 9.6 15.6 9.6c9.6 0 17.4-7.8 17.4-17.4V316.2c8.5-7.6 19.7-12.2 32-12.2z\"],trowel:[512,512,[],\"e589\",\"M343.9 213.4L245.3 312l65.4 65.4c7.9 7.9 11.1 19.4 8.4 30.3s-10.8 19.6-21.5 22.9l-256 80c-11.4 3.5-23.8 .5-32.2-7.9S-2.1 481.8 1.5 470.5l80-256c3.3-10.7 12-18.9 22.9-21.5s22.4 .5 30.3 8.4L200 266.7l98.6-98.6c-14.3-14.6-14.2-38 .3-52.5l95.4-95.4c26.9-26.9 70.5-26.9 97.5 0s26.9 70.5 0 97.5l-95.4 95.4c-14.5 14.5-37.9 14.6-52.5 .3z\"],d:[384,512,[100],\"44\",\"M0 96C0 60.7 28.7 32 64 32h96c123.7 0 224 100.3 224 224s-100.3 224-224 224H64c-35.3 0-64-28.7-64-64V96zm160 0H64V416h96c88.4 0 160-71.6 160-160s-71.6-160-160-160z\"],stapler:[640,512,[],\"e5af\",\"M640 299.3V304 432c0 26.5-21.5 48-48 48H512 448 64c-17.7 0-32-14.3-32-32s14.3-32 32-32H448V368H96c-17.7 0-32-14.3-32-32V219.4L33.8 214C14.2 210.5 0 193.5 0 173.7c0-8.9 2.9-17.5 8.2-24.6l35.6-47.5C76.7 57.8 128.2 32 182.9 32c27 0 53.6 6.3 77.8 18.4L586.9 213.5C619.5 229.7 640 263 640 299.3zM448 304V288L128 230.9V304H448z\"],\"masks-theater\":[640,512,[127917,\"theater-masks\"],\"f630\",\"M74.6 373.2c41.7 36.1 108 82.5 166.1 73.7c6.1-.9 12.1-2.5 18-4.5c-9.2-12.3-17.3-24.4-24.2-35.4c-21.9-35-28.8-75.2-25.9-113.6c-20.6 4.1-39.2 13-54.7 25.4c-6.5 5.2-16.3 1.3-14.8-7c6.4-33.5 33-60.9 68.2-66.3c2.6-.4 5.3-.7 7.9-.8l19.4-131.3c2-13.8 8-32.7 25-45.9C278.2 53.2 310.5 37 363.2 32.2c-.8-.7-1.6-1.4-2.4-2.1C340.6 14.5 288.4-11.5 175.7 5.6S20.5 63 5.7 83.9C0 91.9-.8 102 .6 111.8L24.8 276.1c5.5 37.3 21.5 72.6 49.8 97.2zm87.7-219.6c4.4-3.1 10.8-2 11.8 3.3c.1 .5 .2 1.1 .3 1.6c3.2 21.8-11.6 42-33.1 45.3s-41.5-11.8-44.7-33.5c-.1-.5-.1-1.1-.2-1.6c-.6-5.4 5.2-8.4 10.3-6.7c9 3 18.8 3.9 28.7 2.4s19.1-5.3 26.8-10.8zM261.6 390c29.4 46.9 79.5 110.9 137.6 119.7s124.5-37.5 166.1-73.7c28.3-24.5 44.3-59.8 49.8-97.2l24.2-164.3c1.4-9.8 .6-19.9-5.1-27.9c-14.8-20.9-57.3-61.2-170-78.3S299.4 77.2 279.2 92.8c-7.8 6-11.5 15.4-12.9 25.2L242.1 282.3c-5.5 37.3-.4 75.8 19.6 107.7zM404.5 235.3c-7.7-5.5-16.8-9.3-26.8-10.8s-19.8-.6-28.7 2.4c-5.1 1.7-10.9-1.3-10.3-6.7c.1-.5 .1-1.1 .2-1.6c3.2-21.8 23.2-36.8 44.7-33.5s36.3 23.5 33.1 45.3c-.1 .5-.2 1.1-.3 1.6c-1 5.3-7.4 6.4-11.8 3.3zm136.2 15.5c-1 5.3-7.4 6.4-11.8 3.3c-7.7-5.5-16.8-9.3-26.8-10.8s-19.8-.6-28.7 2.4c-5.1 1.7-10.9-1.3-10.3-6.7c.1-.5 .1-1.1 .2-1.6c3.2-21.8 23.2-36.8 44.7-33.5s36.3 23.5 33.1 45.3c-.1 .5-.2 1.1-.3 1.6zM530 350.2c-19.6 44.7-66.8 72.5-116.8 64.9s-87.1-48.2-93-96.7c-1-8.3 8.9-12.1 15.2-6.7c23.9 20.8 53.6 35.3 87 40.3s66.1 .1 94.9-12.8c7.6-3.4 16 3.2 12.6 10.9z\"],\"kip-sign\":[384,512,[],\"e1c4\",\"M340.8 88.3c13.4-11.5 15-31.7 3.5-45.1s-31.7-15-45.1-3.5L128 186.4V64c0-17.7-14.3-32-32-32S64 46.3 64 64V224H32c-17.7 0-32 14.3-32 32s14.3 32 32 32H64V448c0 17.7 14.3 32 32 32s32-14.3 32-32V325.6L299.2 472.3c13.4 11.5 33.6 9.9 45.1-3.5s9.9-33.6-3.5-45.1L182.5 288H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H182.5L340.8 88.3z\"],\"hand-point-left\":[512,512,[],\"f0a5\",\"M32 96C14.3 96 0 110.3 0 128s14.3 32 32 32l208 0 0-64L32 96zM192 288c-17.7 0-32 14.3-32 32s14.3 32 32 32l64 0c17.7 0 32-14.3 32-32s-14.3-32-32-32l-64 0zm-64-64c0 17.7 14.3 32 32 32l48 0c17.7 0 32-14.3 32-32s-14.3-32-32-32l-48 0c-17.7 0-32 14.3-32 32zm96 160c-17.7 0-32 14.3-32 32s14.3 32 32 32l64 0c17.7 0 32-14.3 32-32s-14.3-32-32-32l-64 0zm88-96l-.6 0c5.4 9.4 8.6 20.3 8.6 32c0 13.2-4 25.4-10.8 35.6c24.9 8.7 42.8 32.5 42.8 60.4c0 11.7-3.1 22.6-8.6 32l8.6 0c88.4 0 160-71.6 160-160l0-61.7c0-42.4-16.9-83.1-46.9-113.1l-11.6-11.6C429.5 77.5 396.9 64 363 64l-27 0c-35.3 0-64 28.7-64 64l0 88c0 22.1 17.9 40 40 40s40-17.9 40-40l0-56c0-8.8 7.2-16 16-16s16 7.2 16 16l0 56c0 39.8-32.2 72-72 72z\"],\"handshake-simple\":[640,512,[129309,\"handshake-alt\"],\"f4c6\",\"M323.4 85.2l-96.8 78.4c-16.1 13-19.2 36.4-7 53.1c12.9 17.8 38 21.3 55.3 7.8l99.3-77.2c7-5.4 17-4.2 22.5 2.8s4.2 17-2.8 22.5l-20.9 16.2L550.2 352H592c26.5 0 48-21.5 48-48V176c0-26.5-21.5-48-48-48H516h-4-.7l-3.9-2.5L434.8 79c-15.3-9.8-33.2-15-51.4-15c-21.8 0-43 7.5-60 21.2zm22.8 124.4l-51.7 40.2C263 274.4 217.3 268 193.7 235.6c-22.2-30.5-16.6-73.1 12.7-96.8l83.2-67.3c-11.6-4.9-24.1-7.4-36.8-7.4C234 64 215.7 69.6 200 80l-72 48H48c-26.5 0-48 21.5-48 48V304c0 26.5 21.5 48 48 48H156.2l91.4 83.4c19.6 17.9 49.9 16.5 67.8-3.1c5.5-6.1 9.2-13.2 11.1-20.6l17 15.6c19.5 17.9 49.9 16.6 67.8-2.9c4.5-4.9 7.8-10.6 9.9-16.5c19.4 13 45.8 10.3 62.1-7.5c17.9-19.5 16.6-49.9-2.9-67.8l-134.2-123z\"],\"jet-fighter\":[640,512,[\"fighter-jet\"],\"f0fb\",\"M160 24c0-13.3 10.7-24 24-24H296c13.3 0 24 10.7 24 24s-10.7 24-24 24H280L384 192H500.4c7.7 0 15.3 1.4 22.5 4.1L625 234.4c9 3.4 15 12 15 21.6s-6 18.2-15 21.6L522.9 315.9c-7.2 2.7-14.8 4.1-22.5 4.1H384L280 464h16c13.3 0 24 10.7 24 24s-10.7 24-24 24H184c-13.3 0-24-10.7-24-24s10.7-24 24-24h8V320H160l-54.6 54.6c-6 6-14.1 9.4-22.6 9.4H64c-17.7 0-32-14.3-32-32V288c-17.7 0-32-14.3-32-32s14.3-32 32-32V160c0-17.7 14.3-32 32-32H82.7c8.5 0 16.6 3.4 22.6 9.4L160 192h32V48h-8c-13.3 0-24-10.7-24-24zM80 240c-8.8 0-16 7.2-16 16s7.2 16 16 16h64c8.8 0 16-7.2 16-16s-7.2-16-16-16H80z\"],\"square-share-nodes\":[448,512,[\"share-alt-square\"],\"f1e1\",\"M64 32C28.7 32 0 60.7 0 96V416c0 35.3 28.7 64 64 64H384c35.3 0 64-28.7 64-64V96c0-35.3-28.7-64-64-64H64zM384 160c0 35.3-28.7 64-64 64c-15.4 0-29.5-5.4-40.6-14.5L194.1 256l85.3 46.5c11-9.1 25.2-14.5 40.6-14.5c35.3 0 64 28.7 64 64s-28.7 64-64 64s-64-28.7-64-64c0-2.5 .1-4.9 .4-7.3L174.5 300c-11.7 12.3-28.2 20-46.5 20c-35.3 0-64-28.7-64-64s28.7-64 64-64c18.3 0 34.8 7.7 46.5 20l81.9-44.7c-.3-2.4-.4-4.9-.4-7.3c0-35.3 28.7-64 64-64s64 28.7 64 64z\"],barcode:[512,512,[],\"f02a\",\"M24 32C10.7 32 0 42.7 0 56V456c0 13.3 10.7 24 24 24H40c13.3 0 24-10.7 24-24V56c0-13.3-10.7-24-24-24H24zm88 0c-8.8 0-16 7.2-16 16V464c0 8.8 7.2 16 16 16s16-7.2 16-16V48c0-8.8-7.2-16-16-16zm72 0c-13.3 0-24 10.7-24 24V456c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V56c0-13.3-10.7-24-24-24H184zm96 0c-13.3 0-24 10.7-24 24V456c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V56c0-13.3-10.7-24-24-24H280zM448 56V456c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V56c0-13.3-10.7-24-24-24H472c-13.3 0-24 10.7-24 24zm-64-8V464c0 8.8 7.2 16 16 16s16-7.2 16-16V48c0-8.8-7.2-16-16-16s-16 7.2-16 16z\"],\"plus-minus\":[384,512,[],\"e43c\",\"M224 32c0-17.7-14.3-32-32-32s-32 14.3-32 32V144H48c-17.7 0-32 14.3-32 32s14.3 32 32 32H160V320c0 17.7 14.3 32 32 32s32-14.3 32-32V208H336c17.7 0 32-14.3 32-32s-14.3-32-32-32H224V32zM0 480c0 17.7 14.3 32 32 32H352c17.7 0 32-14.3 32-32s-14.3-32-32-32H32c-17.7 0-32 14.3-32 32z\"],video:[576,512,[\"video-camera\"],\"f03d\",\"M0 128C0 92.7 28.7 64 64 64H320c35.3 0 64 28.7 64 64V384c0 35.3-28.7 64-64 64H64c-35.3 0-64-28.7-64-64V128zM559.1 99.8c10.4 5.6 16.9 16.4 16.9 28.2V384c0 11.8-6.5 22.6-16.9 28.2s-23 5-32.9-1.6l-96-64L416 337.1V320 192 174.9l14.2-9.5 96-64c9.8-6.5 22.4-7.2 32.9-1.6z\"],\"graduation-cap\":[640,512,[127891,\"mortar-board\"],\"f19d\",\"M320 32c-8.1 0-16.1 1.4-23.7 4.1L15.8 137.4C6.3 140.9 0 149.9 0 160s6.3 19.1 15.8 22.6l57.9 20.9C57.3 229.3 48 259.8 48 291.9v28.1c0 28.4-10.8 57.7-22.3 80.8c-6.5 13-13.9 25.8-22.5 37.6C0 442.7-.9 448.3 .9 453.4s6 8.9 11.2 10.2l64 16c4.2 1.1 8.7 .3 12.4-2s6.3-6.1 7.1-10.4c8.6-42.8 4.3-81.2-2.1-108.7C90.3 344.3 86 329.8 80 316.5V291.9c0-30.2 10.2-58.7 27.9-81.5c12.9-15.5 29.6-28 49.2-35.7l157-61.7c8.2-3.2 17.5 .8 20.7 9s-.8 17.5-9 20.7l-157 61.7c-12.4 4.9-23.3 12.4-32.2 21.6l159.6 57.6c7.6 2.7 15.6 4.1 23.7 4.1s16.1-1.4 23.7-4.1L624.2 182.6c9.5-3.4 15.8-12.5 15.8-22.6s-6.3-19.1-15.8-22.6L343.7 36.1C336.1 33.4 328.1 32 320 32zM128 408c0 35.3 86 72 192 72s192-36.7 192-72L496.7 262.6 354.5 314c-11.1 4-22.8 6-34.5 6s-23.5-2-34.5-6L143.3 262.6 128 408z\"],\"hand-holding-medical\":[576,512,[],\"e05c\",\"M224 24V80H168c-13.3 0-24 10.7-24 24v48c0 13.3 10.7 24 24 24h56v56c0 13.3 10.7 24 24 24h48c13.3 0 24-10.7 24-24V176h56c13.3 0 24-10.7 24-24V104c0-13.3-10.7-24-24-24H320V24c0-13.3-10.7-24-24-24H248c-13.3 0-24 10.7-24 24zM559.7 392.2c17.8-13.1 21.6-38.1 8.5-55.9s-38.1-21.6-55.9-8.5L392.6 416H272c-8.8 0-16-7.2-16-16s7.2-16 16-16h16 64c17.7 0 32-14.3 32-32s-14.3-32-32-32H288 272 193.7c-29.1 0-57.3 9.9-80 28L68.8 384H32c-17.7 0-32 14.3-32 32v64c0 17.7 14.3 32 32 32H192 352.5c29 0 57.3-9.3 80.7-26.5l126.6-93.3zm-367-8.2l.9 0 0 0c-.3 0-.6 0-.9 0z\"],\"person-circle-check\":[576,512,[],\"e53e\",\"M112 48a48 48 0 1 1 96 0 48 48 0 1 1 -96 0zm40 304V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V256.9L59.4 304.5c-9.1 15.1-28.8 20-43.9 10.9s-20-28.8-10.9-43.9l58.3-97c17.4-28.9 48.6-46.6 82.3-46.6h29.7c33.7 0 64.9 17.7 82.3 46.6l44.9 74.7c-16.1 17.6-28.6 38.5-36.6 61.5c-1.9-1.8-3.5-3.9-4.9-6.3L232 256.9V480c0 17.7-14.3 32-32 32s-32-14.3-32-32V352H152zm136 16a144 144 0 1 1 288 0 144 144 0 1 1 -288 0zm211.3-43.3c-6.2-6.2-16.4-6.2-22.6 0L416 385.4l-28.7-28.7c-6.2-6.2-16.4-6.2-22.6 0s-6.2 16.4 0 22.6l40 40c6.2 6.2 16.4 6.2 22.6 0l72-72c6.2-6.2 6.2-16.4 0-22.6z\"],\"turn-up\":[384,512,[10548,\"level-up-alt\"],\"f3bf\",\"M350 177.5c3.8-8.8 2-19-4.6-26l-136-144C204.9 2.7 198.6 0 192 0s-12.9 2.7-17.4 7.5l-136 144c-6.6 7-8.4 17.2-4.6 26s12.5 14.5 22 14.5h88l0 192c0 17.7-14.3 32-32 32H32c-17.7 0-32 14.3-32 32v32c0 17.7 14.3 32 32 32l80 0c70.7 0 128-57.3 128-128l0-192h88c9.6 0 18.2-5.7 22-14.5z\"]};!function(c){try{for(var l=arguments.length,s=new Array(1<l?l-1:0),a=1;a<l;a++)s[a-1]=arguments[a];c.apply(void 0,s)}catch(c){if(!h)throw c}}(function(){d(\"fas\",p),d(\"fa-solid\",p)})}(),function(){\"use strict\";function a(l,c){var s,a=Object.keys(l);return Object.getOwnPropertySymbols&&(s=Object.getOwnPropertySymbols(l),c&&(s=s.filter(function(c){return Object.getOwnPropertyDescriptor(l,c).enumerable})),a.push.apply(a,s)),a}function u(l){for(var c=1;c<arguments.length;c++){var s=null!=arguments[c]?arguments[c]:{};c%2?a(Object(s),!0).forEach(function(c){t(l,c,s[c])}):Object.getOwnPropertyDescriptors?Object.defineProperties(l,Object.getOwnPropertyDescriptors(s)):a(Object(s)).forEach(function(c){Object.defineProperty(l,c,Object.getOwnPropertyDescriptor(s,c))})}return l}function z(c){return(z=\"function\"==typeof Symbol&&\"symbol\"==typeof Symbol.iterator?function(c){return typeof c}:function(c){return c&&\"function\"==typeof Symbol&&c.constructor===Symbol&&c!==Symbol.prototype?\"symbol\":typeof c})(c)}function e(c,l){for(var s=0;s<l.length;s++){var a=l[s];a.enumerable=a.enumerable||!1,a.configurable=!0,\"value\"in a&&(a.writable=!0),Object.defineProperty(c,a.key,a)}}function t(c,l,s){return l in c?Object.defineProperty(c,l,{value:s,enumerable:!0,configurable:!0,writable:!0}):c[l]=s,c}function n(c,l){return function(c){if(Array.isArray(c))return c}(c)||function(c,l){var s=null==c?null:\"undefined\"!=typeof Symbol&&c[Symbol.iterator]||c[\"@@iterator\"];if(null!=s){var a,z,e=[],H=!0,t=!1;try{for(s=s.call(c);!(H=(a=s.next()).done)&&(e.push(a.value),!l||e.length!==l);H=!0);}catch(c){t=!0,z=c}finally{try{H||null==s.return||s.return()}finally{if(t)throw z}}return e}}(c,l)||s(c,l)||function(){throw new TypeError(\"Invalid attempt to destructure non-iterable instance.\\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method.\")}()}function M(c){return function(c){if(Array.isArray(c))return H(c)}(c)||function(c){if(\"undefined\"!=typeof Symbol&&null!=c[Symbol.iterator]||null!=c[\"@@iterator\"])return Array.from(c)}(c)||s(c)||function(){throw new TypeError(\"Invalid attempt to spread non-iterable instance.\\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method.\")}()}function s(c,l){if(c){if(\"string\"==typeof c)return H(c,l);var s=Object.prototype.toString.call(c).slice(8,-1);return\"Map\"===(s=\"Object\"===s&&c.constructor?c.constructor.name:s)||\"Set\"===s?Array.from(c):\"Arguments\"===s||/^(?:Ui|I)nt(?:8|16|32)(?:Clamped)?Array$/.test(s)?H(c,l):void 0}}function H(c,l){(null==l||l>c.length)&&(l=c.length);for(var s=0,a=new Array(l);s<l;s++)a[s]=c[s];return a}function c(){}var l={},V={},r=null,h={mark:c,measure:c};try{\"undefined\"!=typeof window&&(l=window),\"undefined\"!=typeof document&&(V=document),\"undefined\"!=typeof MutationObserver&&(r=MutationObserver),\"undefined\"!=typeof performance&&(h=performance)}catch(c){}var i=(l.navigator||{}).userAgent,m=void 0===i?\"\":i,C=l,v=V,o=r,i=h,f=!!C.document,L=!!v.documentElement&&!!v.head&&\"function\"==typeof v.addEventListener&&\"function\"==typeof v.createElement,d=~m.indexOf(\"MSIE\")||~m.indexOf(\"Trident/\"),l=\"___FONT_AWESOME___\",p=16,b=\"svg-inline--fa\",g=\"data-fa-i2svg\",w=\"data-fa-pseudo-element\",k=\"data-fa-pseudo-element-pending\",y=\"data-prefix\",S=\"data-icon\",A=\"fontawesome-i2svg\",x=\"async\",q=[\"HTML\",\"HEAD\",\"STYLE\",\"SCRIPT\"],Z=function(){try{return!0}catch(c){return!1}}(),O=\"classic\",j=\"sharp\",P=[O,j];function N(c){return new Proxy(c,{get:function(c,l){return l in c?c[l]:c[O]}})}var E=N((t(V={},O,{fa:\"solid\",fas:\"solid\",\"fa-solid\":\"solid\",far:\"regular\",\"fa-regular\":\"regular\",fal:\"light\",\"fa-light\":\"light\",fat:\"thin\",\"fa-thin\":\"thin\",fad:\"duotone\",\"fa-duotone\":\"duotone\",fab:\"brands\",\"fa-brands\":\"brands\",fak:\"kit\",fakd:\"kit\",\"fa-kit\":\"kit\",\"fa-kit-duotone\":\"kit\"}),t(V,j,{fa:\"solid\",fass:\"solid\",\"fa-solid\":\"solid\",fasr:\"regular\",\"fa-regular\":\"regular\",fasl:\"light\",\"fa-light\":\"light\",fast:\"thin\",\"fa-thin\":\"thin\"}),V)),I=N((t(r={},O,{solid:\"fas\",regular:\"far\",light:\"fal\",thin:\"fat\",duotone:\"fad\",brands:\"fab\",kit:\"fak\"}),t(r,j,{solid:\"fass\",regular:\"fasr\",light:\"fasl\",thin:\"fast\"}),r)),T=N((t(h={},O,{fab:\"fa-brands\",fad:\"fa-duotone\",fak:\"fa-kit\",fal:\"fa-light\",far:\"fa-regular\",fas:\"fa-solid\",fat:\"fa-thin\"}),t(h,j,{fass:\"fa-solid\",fasr:\"fa-regular\",fasl:\"fa-light\",fast:\"fa-thin\"}),h)),D=N((t(m={},O,{\"fa-brands\":\"fab\",\"fa-duotone\":\"fad\",\"fa-kit\":\"fak\",\"fa-light\":\"fal\",\"fa-regular\":\"far\",\"fa-solid\":\"fas\",\"fa-thin\":\"fat\"}),t(m,j,{\"fa-solid\":\"fass\",\"fa-regular\":\"fasr\",\"fa-light\":\"fasl\",\"fa-thin\":\"fast\"}),m)),Y=/fa(s|r|l|t|d|b|k|ss|sr|sl|st)?[\\-\\ ]/,R=\"fa-layers-text\",F=/Font ?Awesome ?([56 ]*)(Solid|Regular|Light|Thin|Duotone|Brands|Free|Pro|Sharp|Kit)?.*/i,_=N((t(V={},O,{900:\"fas\",400:\"far\",normal:\"far\",300:\"fal\",100:\"fat\"}),t(V,j,{900:\"fass\",400:\"fasr\",300:\"fasl\",100:\"fast\"}),V)),r=[1,2,3,4,5,6,7,8,9,10],h=r.concat([11,12,13,14,15,16,17,18,19,20]),W=[\"class\",\"data-prefix\",\"data-icon\",\"data-fa-transform\",\"data-fa-mask\"],U={GROUP:\"duotone-group\",SWAP_OPACITY:\"swap-opacity\",PRIMARY:\"primary\",SECONDARY:\"secondary\"},m=new Set;Object.keys(I[O]).map(m.add.bind(m)),Object.keys(I[j]).map(m.add.bind(m));var X=[].concat(P,M(m),[\"2xs\",\"xs\",\"sm\",\"lg\",\"xl\",\"2xl\",\"beat\",\"border\",\"fade\",\"beat-fade\",\"bounce\",\"flip-both\",\"flip-horizontal\",\"flip-vertical\",\"flip\",\"fw\",\"inverse\",\"layers-counter\",\"layers-text\",\"layers\",\"li\",\"pull-left\",\"pull-right\",\"pulse\",\"rotate-180\",\"rotate-270\",\"rotate-90\",\"rotate-by\",\"shake\",\"spin-pulse\",\"spin-reverse\",\"spin\",\"stack-1x\",\"stack-2x\",\"stack\",\"ul\",U.GROUP,U.SWAP_OPACITY,U.PRIMARY,U.SECONDARY]).concat(r.map(function(c){return\"\".concat(c,\"x\")})).concat(h.map(function(c){return\"w-\".concat(c)})),B=C.FontAwesomeConfig||{};v&&\"function\"==typeof v.querySelector&&[[\"data-family-prefix\",\"familyPrefix\"],[\"data-css-prefix\",\"cssPrefix\"],[\"data-family-default\",\"familyDefault\"],[\"data-style-default\",\"styleDefault\"],[\"data-replacement-class\",\"replacementClass\"],[\"data-auto-replace-svg\",\"autoReplaceSvg\"],[\"data-auto-add-css\",\"autoAddCss\"],[\"data-auto-a11y\",\"autoA11y\"],[\"data-search-pseudo-elements\",\"searchPseudoElements\"],[\"data-observe-mutations\",\"observeMutations\"],[\"data-mutate-approach\",\"mutateApproach\"],[\"data-keep-original-source\",\"keepOriginalSource\"],[\"data-measure-performance\",\"measurePerformance\"],[\"data-show-missing-icons\",\"showMissingIcons\"]].forEach(function(c){var l=n(c,2),c=l[0],l=l[1],c=\"\"===(c=function(c){var l=v.querySelector(\"script[\"+c+\"]\");if(l)return l.getAttribute(c)}(c))||\"false\"!==c&&(\"true\"===c||c);null!=c&&(B[l]=c)});V={styleDefault:\"solid\",familyDefault:\"classic\",cssPrefix:\"fa\",replacementClass:b,autoReplaceSvg:!0,autoAddCss:!0,autoA11y:!0,searchPseudoElements:!1,observeMutations:!0,mutateApproach:\"async\",keepOriginalSource:!0,measurePerformance:!1,showMissingIcons:!0};B.familyPrefix&&(B.cssPrefix=B.familyPrefix);var G=u(u({},V),B);G.autoReplaceSvg||(G.observeMutations=!1);var Q={};Object.keys(V).forEach(function(l){Object.defineProperty(Q,l,{enumerable:!0,set:function(c){G[l]=c,K.forEach(function(c){return c(Q)})},get:function(){return G[l]}})}),Object.defineProperty(Q,\"familyPrefix\",{enumerable:!0,set:function(c){G.cssPrefix=c,K.forEach(function(c){return c(Q)})},get:function(){return G.cssPrefix}}),C.FontAwesomeConfig=Q;var K=[];var $=p,J={size:16,x:0,y:0,rotate:0,flipX:!1,flipY:!1};var c1=\"0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ\";function l1(){for(var c=12,l=\"\";0<c--;)l+=c1[62*Math.random()|0];return l}function s1(c){for(var l=[],s=(c||[]).length>>>0;s--;)l[s]=c[s];return l}function a1(c){return c.classList?s1(c.classList):(c.getAttribute(\"class\")||\"\").split(\" \").filter(function(c){return c})}function z1(c){return\"\".concat(c).replace(/&/g,\"&amp;\").replace(/\"/g,\"&quot;\").replace(/'/g,\"&#39;\").replace(/</g,\"&lt;\").replace(/>/g,\"&gt;\")}function e1(s){return Object.keys(s||{}).reduce(function(c,l){return c+\"\".concat(l,\": \").concat(s[l].trim(),\";\")},\"\")}function H1(c){return c.size!==J.size||c.x!==J.x||c.y!==J.y||c.rotate!==J.rotate||c.flipX||c.flipY}function t1(){var c,l,s=b,a=Q.cssPrefix,z=Q.replacementClass,e=':host,:root{--fa-font-solid:normal 900 1em/1 \"Font Awesome 6 Solid\";--fa-font-regular:normal 400 1em/1 \"Font Awesome 6 Regular\";--fa-font-light:normal 300 1em/1 \"Font Awesome 6 Light\";--fa-font-thin:normal 100 1em/1 \"Font Awesome 6 Thin\";--fa-font-duotone:normal 900 1em/1 \"Font Awesome 6 Duotone\";--fa-font-sharp-solid:normal 900 1em/1 \"Font Awesome 6 Sharp\";--fa-font-sharp-regular:normal 400 1em/1 \"Font Awesome 6 Sharp\";--fa-font-sharp-light:normal 300 1em/1 \"Font Awesome 6 Sharp\";--fa-font-sharp-thin:normal 100 1em/1 \"Font Awesome 6 Sharp\";--fa-font-brands:normal 400 1em/1 \"Font Awesome 6 Brands\"}svg:not(:host).svg-inline--fa,svg:not(:root).svg-inline--fa{overflow:visible;box-sizing:content-box}.svg-inline--fa{display:var(--fa-display,inline-block);height:1em;overflow:visible;vertical-align:-.125em}.svg-inline--fa.fa-2xs{vertical-align:.1em}.svg-inline--fa.fa-xs{vertical-align:0}.svg-inline--fa.fa-sm{vertical-align:-.0714285705em}.svg-inline--fa.fa-lg{vertical-align:-.2em}.svg-inline--fa.fa-xl{vertical-align:-.25em}.svg-inline--fa.fa-2xl{vertical-align:-.3125em}.svg-inline--fa.fa-pull-left{margin-right:var(--fa-pull-margin,.3em);width:auto}.svg-inline--fa.fa-pull-right{margin-left:var(--fa-pull-margin,.3em);width:auto}.svg-inline--fa.fa-li{width:var(--fa-li-width,2em);top:.25em}.svg-inline--fa.fa-fw{width:var(--fa-fw-width,1.25em)}.fa-layers svg.svg-inline--fa{bottom:0;left:0;margin:auto;position:absolute;right:0;top:0}.fa-layers-counter,.fa-layers-text{display:inline-block;position:absolute;text-align:center}.fa-layers{display:inline-block;height:1em;position:relative;text-align:center;vertical-align:-.125em;width:1em}.fa-layers svg.svg-inline--fa{-webkit-transform-origin:center center;transform-origin:center center}.fa-layers-text{left:50%;top:50%;-webkit-transform:translate(-50%,-50%);transform:translate(-50%,-50%);-webkit-transform-origin:center center;transform-origin:center center}.fa-layers-counter{background-color:var(--fa-counter-background-color,#ff253a);border-radius:var(--fa-counter-border-radius,1em);box-sizing:border-box;color:var(--fa-inverse,#fff);line-height:var(--fa-counter-line-height,1);max-width:var(--fa-counter-max-width,5em);min-width:var(--fa-counter-min-width,1.5em);overflow:hidden;padding:var(--fa-counter-padding,.25em .5em);right:var(--fa-right,0);text-overflow:ellipsis;top:var(--fa-top,0);-webkit-transform:scale(var(--fa-counter-scale,.25));transform:scale(var(--fa-counter-scale,.25));-webkit-transform-origin:top right;transform-origin:top right}.fa-layers-bottom-right{bottom:var(--fa-bottom,0);right:var(--fa-right,0);top:auto;-webkit-transform:scale(var(--fa-layers-scale,.25));transform:scale(var(--fa-layers-scale,.25));-webkit-transform-origin:bottom right;transform-origin:bottom right}.fa-layers-bottom-left{bottom:var(--fa-bottom,0);left:var(--fa-left,0);right:auto;top:auto;-webkit-transform:scale(var(--fa-layers-scale,.25));transform:scale(var(--fa-layers-scale,.25));-webkit-transform-origin:bottom left;transform-origin:bottom left}.fa-layers-top-right{top:var(--fa-top,0);right:var(--fa-right,0);-webkit-transform:scale(var(--fa-layers-scale,.25));transform:scale(var(--fa-layers-scale,.25));-webkit-transform-origin:top right;transform-origin:top right}.fa-layers-top-left{left:var(--fa-left,0);right:auto;top:var(--fa-top,0);-webkit-transform:scale(var(--fa-layers-scale,.25));transform:scale(var(--fa-layers-scale,.25));-webkit-transform-origin:top left;transform-origin:top left}.fa-1x{font-size:1em}.fa-2x{font-size:2em}.fa-3x{font-size:3em}.fa-4x{font-size:4em}.fa-5x{font-size:5em}.fa-6x{font-size:6em}.fa-7x{font-size:7em}.fa-8x{font-size:8em}.fa-9x{font-size:9em}.fa-10x{font-size:10em}.fa-2xs{font-size:.625em;line-height:.1em;vertical-align:.225em}.fa-xs{font-size:.75em;line-height:.0833333337em;vertical-align:.125em}.fa-sm{font-size:.875em;line-height:.0714285718em;vertical-align:.0535714295em}.fa-lg{font-size:1.25em;line-height:.05em;vertical-align:-.075em}.fa-xl{font-size:1.5em;line-height:.0416666682em;vertical-align:-.125em}.fa-2xl{font-size:2em;line-height:.03125em;vertical-align:-.1875em}.fa-fw{text-align:center;width:1.25em}.fa-ul{list-style-type:none;margin-left:var(--fa-li-margin,2.5em);padding-left:0}.fa-ul>li{position:relative}.fa-li{left:calc(var(--fa-li-width,2em) * -1);position:absolute;text-align:center;width:var(--fa-li-width,2em);line-height:inherit}.fa-border{border-color:var(--fa-border-color,#eee);border-radius:var(--fa-border-radius,.1em);border-style:var(--fa-border-style,solid);border-width:var(--fa-border-width,.08em);padding:var(--fa-border-padding,.2em .25em .15em)}.fa-pull-left{float:left;margin-right:var(--fa-pull-margin,.3em)}.fa-pull-right{float:right;margin-left:var(--fa-pull-margin,.3em)}.fa-beat{-webkit-animation-name:fa-beat;animation-name:fa-beat;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,ease-in-out);animation-timing-function:var(--fa-animation-timing,ease-in-out)}.fa-bounce{-webkit-animation-name:fa-bounce;animation-name:fa-bounce;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,cubic-bezier(.28,.84,.42,1));animation-timing-function:var(--fa-animation-timing,cubic-bezier(.28,.84,.42,1))}.fa-fade{-webkit-animation-name:fa-fade;animation-name:fa-fade;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,cubic-bezier(.4,0,.6,1));animation-timing-function:var(--fa-animation-timing,cubic-bezier(.4,0,.6,1))}.fa-beat-fade{-webkit-animation-name:fa-beat-fade;animation-name:fa-beat-fade;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,cubic-bezier(.4,0,.6,1));animation-timing-function:var(--fa-animation-timing,cubic-bezier(.4,0,.6,1))}.fa-flip{-webkit-animation-name:fa-flip;animation-name:fa-flip;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,ease-in-out);animation-timing-function:var(--fa-animation-timing,ease-in-out)}.fa-shake{-webkit-animation-name:fa-shake;animation-name:fa-shake;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,linear);animation-timing-function:var(--fa-animation-timing,linear)}.fa-spin{-webkit-animation-name:fa-spin;animation-name:fa-spin;-webkit-animation-delay:var(--fa-animation-delay,0s);animation-delay:var(--fa-animation-delay,0s);-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,2s);animation-duration:var(--fa-animation-duration,2s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,linear);animation-timing-function:var(--fa-animation-timing,linear)}.fa-spin-reverse{--fa-animation-direction:reverse}.fa-pulse,.fa-spin-pulse{-webkit-animation-name:fa-spin;animation-name:fa-spin;-webkit-animation-direction:var(--fa-animation-direction,normal);animation-direction:var(--fa-animation-direction,normal);-webkit-animation-duration:var(--fa-animation-duration,1s);animation-duration:var(--fa-animation-duration,1s);-webkit-animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-iteration-count:var(--fa-animation-iteration-count,infinite);-webkit-animation-timing-function:var(--fa-animation-timing,steps(8));animation-timing-function:var(--fa-animation-timing,steps(8))}@media (prefers-reduced-motion:reduce){.fa-beat,.fa-beat-fade,.fa-bounce,.fa-fade,.fa-flip,.fa-pulse,.fa-shake,.fa-spin,.fa-spin-pulse{-webkit-animation-delay:-1ms;animation-delay:-1ms;-webkit-animation-duration:1ms;animation-duration:1ms;-webkit-animation-iteration-count:1;animation-iteration-count:1;-webkit-transition-delay:0s;transition-delay:0s;-webkit-transition-duration:0s;transition-duration:0s}}@-webkit-keyframes fa-beat{0%,90%{-webkit-transform:scale(1);transform:scale(1)}45%{-webkit-transform:scale(var(--fa-beat-scale,1.25));transform:scale(var(--fa-beat-scale,1.25))}}@keyframes fa-beat{0%,90%{-webkit-transform:scale(1);transform:scale(1)}45%{-webkit-transform:scale(var(--fa-beat-scale,1.25));transform:scale(var(--fa-beat-scale,1.25))}}@-webkit-keyframes fa-bounce{0%{-webkit-transform:scale(1,1) translateY(0);transform:scale(1,1) translateY(0)}10%{-webkit-transform:scale(var(--fa-bounce-start-scale-x,1.1),var(--fa-bounce-start-scale-y,.9)) translateY(0);transform:scale(var(--fa-bounce-start-scale-x,1.1),var(--fa-bounce-start-scale-y,.9)) translateY(0)}30%{-webkit-transform:scale(var(--fa-bounce-jump-scale-x,.9),var(--fa-bounce-jump-scale-y,1.1)) translateY(var(--fa-bounce-height,-.5em));transform:scale(var(--fa-bounce-jump-scale-x,.9),var(--fa-bounce-jump-scale-y,1.1)) translateY(var(--fa-bounce-height,-.5em))}50%{-webkit-transform:scale(var(--fa-bounce-land-scale-x,1.05),var(--fa-bounce-land-scale-y,.95)) translateY(0);transform:scale(var(--fa-bounce-land-scale-x,1.05),var(--fa-bounce-land-scale-y,.95)) translateY(0)}57%{-webkit-transform:scale(1,1) translateY(var(--fa-bounce-rebound,-.125em));transform:scale(1,1) translateY(var(--fa-bounce-rebound,-.125em))}64%{-webkit-transform:scale(1,1) translateY(0);transform:scale(1,1) translateY(0)}100%{-webkit-transform:scale(1,1) translateY(0);transform:scale(1,1) translateY(0)}}@keyframes fa-bounce{0%{-webkit-transform:scale(1,1) translateY(0);transform:scale(1,1) translateY(0)}10%{-webkit-transform:scale(var(--fa-bounce-start-scale-x,1.1),var(--fa-bounce-start-scale-y,.9)) translateY(0);transform:scale(var(--fa-bounce-start-scale-x,1.1),var(--fa-bounce-start-scale-y,.9)) translateY(0)}30%{-webkit-transform:scale(var(--fa-bounce-jump-scale-x,.9),var(--fa-bounce-jump-scale-y,1.1)) translateY(var(--fa-bounce-height,-.5em));transform:scale(var(--fa-bounce-jump-scale-x,.9),var(--fa-bounce-jump-scale-y,1.1)) translateY(var(--fa-bounce-height,-.5em))}50%{-webkit-transform:scale(var(--fa-bounce-land-scale-x,1.05),var(--fa-bounce-land-scale-y,.95)) translateY(0);transform:scale(var(--fa-bounce-land-scale-x,1.05),var(--fa-bounce-land-scale-y,.95)) translateY(0)}57%{-webkit-transform:scale(1,1) translateY(var(--fa-bounce-rebound,-.125em));transform:scale(1,1) translateY(var(--fa-bounce-rebound,-.125em))}64%{-webkit-transform:scale(1,1) translateY(0);transform:scale(1,1) translateY(0)}100%{-webkit-transform:scale(1,1) translateY(0);transform:scale(1,1) translateY(0)}}@-webkit-keyframes fa-fade{50%{opacity:var(--fa-fade-opacity,.4)}}@keyframes fa-fade{50%{opacity:var(--fa-fade-opacity,.4)}}@-webkit-keyframes fa-beat-fade{0%,100%{opacity:var(--fa-beat-fade-opacity,.4);-webkit-transform:scale(1);transform:scale(1)}50%{opacity:1;-webkit-transform:scale(var(--fa-beat-fade-scale,1.125));transform:scale(var(--fa-beat-fade-scale,1.125))}}@keyframes fa-beat-fade{0%,100%{opacity:var(--fa-beat-fade-opacity,.4);-webkit-transform:scale(1);transform:scale(1)}50%{opacity:1;-webkit-transform:scale(var(--fa-beat-fade-scale,1.125));transform:scale(var(--fa-beat-fade-scale,1.125))}}@-webkit-keyframes fa-flip{50%{-webkit-transform:rotate3d(var(--fa-flip-x,0),var(--fa-flip-y,1),var(--fa-flip-z,0),var(--fa-flip-angle,-180deg));transform:rotate3d(var(--fa-flip-x,0),var(--fa-flip-y,1),var(--fa-flip-z,0),var(--fa-flip-angle,-180deg))}}@keyframes fa-flip{50%{-webkit-transform:rotate3d(var(--fa-flip-x,0),var(--fa-flip-y,1),var(--fa-flip-z,0),var(--fa-flip-angle,-180deg));transform:rotate3d(var(--fa-flip-x,0),var(--fa-flip-y,1),var(--fa-flip-z,0),var(--fa-flip-angle,-180deg))}}@-webkit-keyframes fa-shake{0%{-webkit-transform:rotate(-15deg);transform:rotate(-15deg)}4%{-webkit-transform:rotate(15deg);transform:rotate(15deg)}24%,8%{-webkit-transform:rotate(-18deg);transform:rotate(-18deg)}12%,28%{-webkit-transform:rotate(18deg);transform:rotate(18deg)}16%{-webkit-transform:rotate(-22deg);transform:rotate(-22deg)}20%{-webkit-transform:rotate(22deg);transform:rotate(22deg)}32%{-webkit-transform:rotate(-12deg);transform:rotate(-12deg)}36%{-webkit-transform:rotate(12deg);transform:rotate(12deg)}100%,40%{-webkit-transform:rotate(0);transform:rotate(0)}}@keyframes fa-shake{0%{-webkit-transform:rotate(-15deg);transform:rotate(-15deg)}4%{-webkit-transform:rotate(15deg);transform:rotate(15deg)}24%,8%{-webkit-transform:rotate(-18deg);transform:rotate(-18deg)}12%,28%{-webkit-transform:rotate(18deg);transform:rotate(18deg)}16%{-webkit-transform:rotate(-22deg);transform:rotate(-22deg)}20%{-webkit-transform:rotate(22deg);transform:rotate(22deg)}32%{-webkit-transform:rotate(-12deg);transform:rotate(-12deg)}36%{-webkit-transform:rotate(12deg);transform:rotate(12deg)}100%,40%{-webkit-transform:rotate(0);transform:rotate(0)}}@-webkit-keyframes fa-spin{0%{-webkit-transform:rotate(0);transform:rotate(0)}100%{-webkit-transform:rotate(360deg);transform:rotate(360deg)}}@keyframes fa-spin{0%{-webkit-transform:rotate(0);transform:rotate(0)}100%{-webkit-transform:rotate(360deg);transform:rotate(360deg)}}.fa-rotate-90{-webkit-transform:rotate(90deg);transform:rotate(90deg)}.fa-rotate-180{-webkit-transform:rotate(180deg);transform:rotate(180deg)}.fa-rotate-270{-webkit-transform:rotate(270deg);transform:rotate(270deg)}.fa-flip-horizontal{-webkit-transform:scale(-1,1);transform:scale(-1,1)}.fa-flip-vertical{-webkit-transform:scale(1,-1);transform:scale(1,-1)}.fa-flip-both,.fa-flip-horizontal.fa-flip-vertical{-webkit-transform:scale(-1,-1);transform:scale(-1,-1)}.fa-rotate-by{-webkit-transform:rotate(var(--fa-rotate-angle,0));transform:rotate(var(--fa-rotate-angle,0))}.fa-stack{display:inline-block;vertical-align:middle;height:2em;position:relative;width:2.5em}.fa-stack-1x,.fa-stack-2x{bottom:0;left:0;margin:auto;position:absolute;right:0;top:0;z-index:var(--fa-stack-z-index,auto)}.svg-inline--fa.fa-stack-1x{height:1em;width:1.25em}.svg-inline--fa.fa-stack-2x{height:2em;width:2.5em}.fa-inverse{color:var(--fa-inverse,#fff)}.fa-sr-only,.sr-only{position:absolute;width:1px;height:1px;padding:0;margin:-1px;overflow:hidden;clip:rect(0,0,0,0);white-space:nowrap;border-width:0}.fa-sr-only-focusable:not(:focus),.sr-only-focusable:not(:focus){position:absolute;width:1px;height:1px;padding:0;margin:-1px;overflow:hidden;clip:rect(0,0,0,0);white-space:nowrap;border-width:0}.svg-inline--fa .fa-primary{fill:var(--fa-primary-color,currentColor);opacity:var(--fa-primary-opacity,1)}.svg-inline--fa .fa-secondary{fill:var(--fa-secondary-color,currentColor);opacity:var(--fa-secondary-opacity,.4)}.svg-inline--fa.fa-swap-opacity .fa-primary{opacity:var(--fa-secondary-opacity,.4)}.svg-inline--fa.fa-swap-opacity .fa-secondary{opacity:var(--fa-primary-opacity,1)}.svg-inline--fa mask .fa-primary,.svg-inline--fa mask .fa-secondary{fill:#000}.fa-duotone.fa-inverse,.fad.fa-inverse{color:var(--fa-inverse,#fff)}';return\"fa\"===a&&z===s||(c=new RegExp(\"\\\\.\".concat(\"fa\",\"\\\\-\"),\"g\"),l=new RegExp(\"\\\\--\".concat(\"fa\",\"\\\\-\"),\"g\"),s=new RegExp(\"\\\\.\".concat(s),\"g\"),e=e.replace(c,\".\".concat(a,\"-\")).replace(l,\"--\".concat(a,\"-\")).replace(s,\".\".concat(z))),e}var V1=!1;function r1(){Q.autoAddCss&&!V1&&(function(c){if(c&&L){var l=v.createElement(\"style\");l.setAttribute(\"type\",\"text/css\"),l.innerHTML=c;for(var s=v.head.childNodes,a=null,z=s.length-1;-1<z;z--){var e=s[z],H=(e.tagName||\"\").toUpperCase();-1<[\"STYLE\",\"LINK\"].indexOf(H)&&(a=e)}v.head.insertBefore(l,a)}}(t1()),V1=!0)}m={mixout:function(){return{dom:{css:t1,insertCss:r1}}},hooks:function(){return{beforeDOMElementCreation:function(){r1()},beforeI2svg:function(){r1()}}}},r=C||{};r[l]||(r[l]={}),r[l].styles||(r[l].styles={}),r[l].hooks||(r[l].hooks={}),r[l].shims||(r[l].shims=[]);function M1(){v.removeEventListener(\"DOMContentLoaded\",M1),i1=1,n1.map(function(c){return c()})}var h1=r[l],n1=[],i1=!1;function m1(c){L&&(i1?setTimeout(c,0):n1.push(c))}function o1(c){var s,l=c.tag,a=c.attributes,z=void 0===a?{}:a,a=c.children,a=void 0===a?[]:a;return\"string\"==typeof c?z1(c):\"<\".concat(l,\" \").concat((s=z,Object.keys(s||{}).reduce(function(c,l){return c+\"\".concat(l,'=\"').concat(z1(s[l]),'\" ')},\"\").trim()),\">\").concat(a.map(o1).join(\"\"),\"</\").concat(l,\">\")}function f1(c,l,s){if(c&&c[l]&&c[l][s])return{prefix:l,iconName:s,icon:c[l][s]}}L&&((i1=(v.documentElement.doScroll?/^loaded|^c/:/^loaded|^i|^c/).test(v.readyState))||v.addEventListener(\"DOMContentLoaded\",M1));function C1(c,l,s,a){for(var z,e,H=Object.keys(c),t=H.length,V=void 0!==a?v1(l,a):l,r=void 0===s?(z=1,c[H[0]]):(z=0,s);z<t;z++)r=V(r,c[e=H[z]],e,c);return r}var v1=function(z,e){return function(c,l,s,a){return z.call(e,c,l,s,a)}};function L1(c){c=function(c){for(var l=[],s=0,a=c.length;s<a;){var z,e=c.charCodeAt(s++);55296<=e&&e<=56319&&s<a?56320==(64512&(z=c.charCodeAt(s++)))?l.push(((1023&e)<<10)+(1023&z)+65536):(l.push(e),s--):l.push(e)}return l}(c);return 1===c.length?c[0].toString(16):null}function u1(a){return Object.keys(a).reduce(function(c,l){var s=a[l];return!!s.icon?c[s.iconName]=s.icon:c[l]=s,c},{})}function d1(c,l,s){var a=(2<arguments.length&&void 0!==s?s:{}).skipHooks,s=void 0!==a&&a,a=u1(l);\"function\"!=typeof h1.hooks.addPack||s?h1.styles[c]=u(u({},h1.styles[c]||{}),a):h1.hooks.addPack(c,u1(l)),\"fas\"===c&&d1(\"fa\",l)}var p1=h1.styles,b1=h1.shims,g1=(t(h={},O,Object.values(T[O])),t(h,j,Object.values(T[j])),h),w1=null,k1={},y1={},S1={},A1={},x1={},q1=(t(V={},O,Object.keys(E[O])),t(V,j,Object.keys(E[j])),V);function Z1(c,l){var s=l.split(\"-\"),l=s[0],s=s.slice(1).join(\"-\");return l!==c||\"\"===s||~X.indexOf(s)?null:s}function O1(){function c(a){return C1(p1,function(c,l,s){return c[s]=C1(l,a,{}),c},{})}k1=c(function(l,c,s){return c[3]&&(l[c[3]]=s),c[2]&&c[2].filter(function(c){return\"number\"==typeof c}).forEach(function(c){l[c.toString(16)]=s}),l}),y1=c(function(l,c,s){return l[s]=s,c[2]&&c[2].filter(function(c){return\"string\"==typeof c}).forEach(function(c){l[c]=s}),l}),x1=c(function(l,c,s){c=c[2];return l[s]=s,c.forEach(function(c){l[c]=s}),l});var z=\"far\"in p1||Q.autoFetchSvg,l=C1(b1,function(c,l){var s=l[0],a=l[1],l=l[2];return\"far\"!==a||z||(a=\"fas\"),\"string\"==typeof s&&(c.names[s]={prefix:a,iconName:l}),\"number\"==typeof s&&(c.unicodes[s.toString(16)]={prefix:a,iconName:l}),c},{names:{},unicodes:{}});S1=l.names,A1=l.unicodes,w1=I1(Q.styleDefault,{family:Q.familyDefault})}function j1(c,l){return(k1[c]||{})[l]}function P1(c,l){return(x1[c]||{})[l]}function N1(c){return S1[c]||{prefix:null,iconName:null}}r=function(c){w1=I1(c.styleDefault,{family:Q.familyDefault})},K.push(r),O1();function E1(){return{prefix:null,iconName:null,rest:[]}}function I1(c,l){var s=(1<arguments.length&&void 0!==l?l:{}).family,l=void 0===s?O:s,s=E[l][c],s=I[l][c]||I[l][s],c=c in h1.styles?c:null;return s||c||null}var T1=(t(l={},O,Object.keys(T[O])),t(l,j,Object.keys(T[j])),l);function D1(c,l){var l=(1<arguments.length&&void 0!==l?l:{}).skipLookups,a=void 0!==l&&l,z=(t(l={},O,\"\".concat(Q.cssPrefix,\"-\").concat(O)),t(l,j,\"\".concat(Q.cssPrefix,\"-\").concat(j)),l),e=null,H=O;(c.includes(z[O])||c.some(function(c){return T1[O].includes(c)}))&&(H=O),(c.includes(z[j])||c.some(function(c){return T1[j].includes(c)}))&&(H=j);l=c.reduce(function(c,l){var s=Z1(Q.cssPrefix,l);return p1[l]?(l=g1[H].includes(l)?D[H][l]:l,e=l,c.prefix=l):-1<q1[H].indexOf(l)?(e=l,c.prefix=I1(l,{family:H})):s?c.iconName=s:l!==Q.replacementClass&&l!==z[O]&&l!==z[j]&&c.rest.push(l),!a&&c.prefix&&c.iconName&&(s=\"fa\"===e?N1(c.iconName):{},l=P1(c.prefix,c.iconName),s.prefix&&(e=null),c.iconName=s.iconName||l||c.iconName,c.prefix=s.prefix||c.prefix,\"far\"!==c.prefix||p1.far||!p1.fas||Q.autoFetchSvg||(c.prefix=\"fas\")),c},E1());return(c.includes(\"fa-brands\")||c.includes(\"fab\"))&&(l.prefix=\"fab\"),(c.includes(\"fa-duotone\")||c.includes(\"fad\"))&&(l.prefix=\"fad\"),l.prefix||H!==j||!p1.fass&&!Q.autoFetchSvg||(l.prefix=\"fass\",l.iconName=P1(l.prefix,l.iconName)||l.iconName),\"fa\"!==l.prefix&&\"fa\"!==e||(l.prefix=w1||\"fas\"),l}var h=function(){function c(){!function(c,l){if(!(c instanceof l))throw new TypeError(\"Cannot call a class as a function\")}(this,c),this.definitions={}}var l,s,a;return l=c,(s=[{key:\"add\",value:function(){for(var s=this,c=arguments.length,l=new Array(c),a=0;a<c;a++)l[a]=arguments[a];var z=l.reduce(this._pullDefinitions,{});Object.keys(z).forEach(function(c){s.definitions[c]=u(u({},s.definitions[c]||{}),z[c]),d1(c,z[c]);var l=T[O][c];l&&d1(l,z[c]),O1()})}},{key:\"reset\",value:function(){this.definitions={}}},{key:\"_pullDefinitions\",value:function(z,c){var e=c.prefix&&c.iconName&&c.icon?{0:c}:c;return Object.keys(e).map(function(c){var l=e[c],s=l.prefix,c=l.iconName,a=l.icon,l=a[2];z[s]||(z[s]={}),0<l.length&&l.forEach(function(c){\"string\"==typeof c&&(z[s][c]=a)}),z[s][c]=a}),z}}])&&e(l.prototype,s),a&&e(l,a),Object.defineProperty(l,\"prototype\",{writable:!1}),c}(),V=[],Y1={},R1={},F1=Object.keys(R1);function _1(c,l){for(var s=arguments.length,a=new Array(2<s?s-2:0),z=2;z<s;z++)a[z-2]=arguments[z];return(Y1[c]||[]).forEach(function(c){l=c.apply(null,[l].concat(a))}),l}function W1(c){for(var l=arguments.length,s=new Array(1<l?l-1:0),a=1;a<l;a++)s[a-1]=arguments[a];(Y1[c]||[]).forEach(function(c){c.apply(null,s)})}function U1(c){var l=c,c=Array.prototype.slice.call(arguments,1);return R1[l]?R1[l].apply(null,c):void 0}function X1(c){\"fa\"===c.prefix&&(c.prefix=\"fas\");var l=c.iconName,c=c.prefix||w1;if(l)return l=P1(c,l)||l,f1(B1.definitions,c,l)||f1(h1.styles,c,l)}var B1=new h,G1={noAuto:function(){Q.autoReplaceSvg=!1,Q.observeMutations=!1,W1(\"noAuto\")},config:Q,dom:{i2svg:function(){var c=0<arguments.length&&void 0!==arguments[0]?arguments[0]:{};return L?(W1(\"beforeI2svg\",c),U1(\"pseudoElements2svg\",c),U1(\"i2svg\",c)):Promise.reject(\"Operation requires a DOM of some kind.\")},watch:function(){var c=0<arguments.length&&void 0!==arguments[0]?arguments[0]:{},l=c.autoReplaceSvgRoot;!1===Q.autoReplaceSvg&&(Q.autoReplaceSvg=!0),Q.observeMutations=!0,m1(function(){Q1({autoReplaceSvgRoot:l}),W1(\"watch\",c)})}},parse:{icon:function(c){if(null===c)return null;if(\"object\"===z(c)&&c.prefix&&c.iconName)return{prefix:c.prefix,iconName:P1(c.prefix,c.iconName)||c.iconName};if(Array.isArray(c)&&2===c.length){var l=0===c[1].indexOf(\"fa-\")?c[1].slice(3):c[1],s=I1(c[0]);return{prefix:s,iconName:P1(s,l)||l}}if(\"string\"==typeof c&&(-1<c.indexOf(\"\".concat(Q.cssPrefix,\"-\"))||c.match(Y))){l=D1(c.split(\" \"),{skipLookups:!0});return{prefix:l.prefix||w1,iconName:P1(l.prefix,l.iconName)||l.iconName}}return\"string\"==typeof c?{prefix:w1,iconName:P1(w1,c)||c}:void 0}},library:B1,findIconDefinition:X1,toHtml:o1},Q1=function(){var c=(0<arguments.length&&void 0!==arguments[0]?arguments[0]:{}).autoReplaceSvgRoot,c=void 0===c?v:c;(0<Object.keys(h1.styles).length||Q.autoFetchSvg)&&L&&Q.autoReplaceSvg&&G1.dom.i2svg({node:c})};function K1(l,c){return Object.defineProperty(l,\"abstract\",{get:c}),Object.defineProperty(l,\"html\",{get:function(){return l.abstract.map(o1)}}),Object.defineProperty(l,\"node\",{get:function(){if(L){var c=v.createElement(\"div\");return c.innerHTML=l.html,c.children}}}),l}function $1(c){var l=c.icons,s=l.main,a=l.mask,z=c.prefix,e=c.iconName,H=c.transform,t=c.symbol,V=c.title,r=c.maskId,M=c.titleId,h=c.extra,n=c.watchable,i=void 0!==n&&n,m=a.found?a:s,l=m.width,c=m.height,n=\"fak\"===z,m=[Q.replacementClass,e?\"\".concat(Q.cssPrefix,\"-\").concat(e):\"\"].filter(function(c){return-1===h.classes.indexOf(c)}).filter(function(c){return\"\"!==c||!!c}).concat(h.classes).join(\" \"),m={children:[],attributes:u(u({},h.attributes),{},{\"data-prefix\":z,\"data-icon\":e,class:m,role:h.attributes.role||\"img\",xmlns:\"http://www.w3.org/2000/svg\",viewBox:\"0 0 \".concat(l,\" \").concat(c)})},c=n&&!~h.classes.indexOf(\"fa-fw\")?{width:\"\".concat(l/c*16*.0625,\"em\")}:{};i&&(m.attributes[g]=\"\"),V&&(m.children.push({tag:\"title\",attributes:{id:m.attributes[\"aria-labelledby\"]||\"title-\".concat(M||l1())},children:[V]}),delete m.attributes.title);var o,f,C,v,L,c=u(u({},m),{},{prefix:z,iconName:e,main:s,mask:a,maskId:r,transform:H,symbol:t,styles:u(u({},c),h.styles)}),a=a.found&&s.found?U1(\"generateAbstractMask\",c)||{children:[],attributes:{}}:U1(\"generateAbstractIcon\",c)||{children:[],attributes:{}},s=a.children,a=a.attributes;return c.children=s,c.attributes=a,t?(f=(o=c).prefix,C=o.iconName,v=o.children,L=o.attributes,o=!0===(o=o.symbol)?\"\".concat(f,\"-\").concat(Q.cssPrefix,\"-\").concat(C):o,[{tag:\"svg\",attributes:{style:\"display: none;\"},children:[{tag:\"symbol\",attributes:u(u({},L),{},{id:o}),children:v}]}]):(C=(f=c).children,L=f.main,o=f.mask,v=f.attributes,c=f.styles,H1(f=f.transform)&&L.found&&!o.found&&(o=L.width/L.height/2,L=.5,v.style=e1(u(u({},c),{},{\"transform-origin\":\"\".concat(o+f.x/16,\"em \").concat(L+f.y/16,\"em\")}))),[{tag:\"svg\",attributes:v,children:C}])}function J1(c){var l=c.content,s=c.width,a=c.height,z=c.transform,e=c.title,H=c.extra,t=c.watchable,V=void 0!==t&&t,c=u(u(u({},H.attributes),e?{title:e}:{}),{},{class:H.classes.join(\" \")});V&&(c[g]=\"\");t=u({},H.styles);H1(z)&&(t.transform=(H=(V={transform:z,startCentered:!0,width:s,height:a}).transform,z=V.width,a=void 0===(s=V.height)?p:s,V=void 0!==(s=V.startCentered)&&s,s=\"\",s+=V&&d?\"translate(\".concat(H.x/$-(void 0===z?p:z)/2,\"em, \").concat(H.y/$-a/2,\"em) \"):V?\"translate(calc(-50% + \".concat(H.x/$,\"em), calc(-50% + \").concat(H.y/$,\"em)) \"):\"translate(\".concat(H.x/$,\"em, \").concat(H.y/$,\"em) \"),s+=\"scale(\".concat(H.size/$*(H.flipX?-1:1),\", \").concat(H.size/$*(H.flipY?-1:1),\") \"),s+=\"rotate(\".concat(H.rotate,\"deg) \")),t[\"-webkit-transform\"]=t.transform);t=e1(t);0<t.length&&(c.style=t);t=[];return t.push({tag:\"span\",attributes:c,children:[l]}),e&&t.push({tag:\"span\",attributes:{class:\"sr-only\"},children:[e]}),t}var c2=h1.styles;function l2(c){var l=c[0],s=c[1],c=n(c.slice(4),1)[0];return{found:!0,width:l,height:s,icon:Array.isArray(c)?{tag:\"g\",attributes:{class:\"\".concat(Q.cssPrefix,\"-\").concat(U.GROUP)},children:[{tag:\"path\",attributes:{class:\"\".concat(Q.cssPrefix,\"-\").concat(U.SECONDARY),fill:\"currentColor\",d:c[0]}},{tag:\"path\",attributes:{class:\"\".concat(Q.cssPrefix,\"-\").concat(U.PRIMARY),fill:\"currentColor\",d:c[1]}}]}:{tag:\"path\",attributes:{fill:\"currentColor\",d:c}}}}var s2={found:!1,width:512,height:512};function a2(z,e){var H=e;return\"fa\"===e&&null!==Q.styleDefault&&(e=w1),new Promise(function(c,l){var s,a;U1(\"missingIconAbstract\");if(\"fa\"===H&&(a=N1(z)||{},z=a.iconName||z,e=a.prefix||e),z&&e&&c2[e]&&c2[e][z])return c(l2(c2[e][z]));s=z,a=e,Z||Q.showMissingIcons||!s||console.error('Icon with name \"'.concat(s,'\" and prefix \"').concat(a,'\" is missing.')),c(u(u({},s2),{},{icon:Q.showMissingIcons&&z&&U1(\"missingIconAbstract\")||{}}))})}function z2(){}function e2(c){H2.mark(\"\".concat(t2,\" \").concat(c,\" ends\")),H2.measure(\"\".concat(t2,\" \").concat(c),\"\".concat(t2,\" \").concat(c,\" begins\"),\"\".concat(t2,\" \").concat(c,\" ends\"))}var H2=Q.measurePerformance&&i&&i.mark&&i.measure?i:{mark:z2,measure:z2},t2='FA \"6.5.2\"',V2={begin:function(c){return H2.mark(\"\".concat(t2,\" \").concat(c,\" begins\")),function(){return e2(c)}},end:e2},r2=function(){};function M2(c){return\"string\"==typeof(c.getAttribute?c.getAttribute(g):null)}function h2(c){return v.createElementNS(\"http://www.w3.org/2000/svg\",c)}function n2(c){return v.createElement(c)}var i2={replace:function(c){var l=c[0];l.parentNode&&(c[1].forEach(function(c){l.parentNode.insertBefore(function l(s,c){var c=(1<arguments.length&&void 0!==c?c:{}).ceFn,a=void 0===c?\"svg\"===s.tag?h2:n2:c;if(\"string\"==typeof s)return v.createTextNode(s);var z=a(s.tag);return Object.keys(s.attributes||[]).forEach(function(c){z.setAttribute(c,s.attributes[c])}),(s.children||[]).forEach(function(c){z.appendChild(l(c,{ceFn:a}))}),z}(c),l)}),null===l.getAttribute(g)&&Q.keepOriginalSource?(c=v.createComment((c=\" \".concat((c=l).outerHTML,\" \"),c=\"\".concat(c,\"Font Awesome fontawesome.com \"))),l.parentNode.replaceChild(c,l)):l.remove())},nest:function(c){var l=c[0],s=c[1];if(~a1(l).indexOf(Q.replacementClass))return i2.replace(c);var a=new RegExp(\"\".concat(Q.cssPrefix,\"-.*\"));delete s[0].attributes.id,s[0].attributes.class&&(c=s[0].attributes.class.split(\" \").reduce(function(c,l){return(l===Q.replacementClass||l.match(a)?c.toSvg:c.toNode).push(l),c},{toNode:[],toSvg:[]}),s[0].attributes.class=c.toSvg.join(\" \"),0===c.toNode.length?l.removeAttribute(\"class\"):l.setAttribute(\"class\",c.toNode.join(\" \")));s=s.map(o1).join(\"\\n\");l.setAttribute(g,\"\"),l.innerHTML=s}};function m2(c){c()}function o2(s,c){var a=\"function\"==typeof c?c:r2;0===s.length?a():(Q.mutateApproach===x?C.requestAnimationFrame||m2:m2)(function(){var c=!0!==Q.autoReplaceSvg&&i2[Q.autoReplaceSvg]||i2.replace,l=V2.begin(\"mutate\");s.map(c),l(),a()})}var f2=!1;function C2(){f2=!0}function v2(){f2=!1}var L2=null;function u2(c){var e,H,l,t;o&&Q.observeMutations&&(l=c.treeCallback,e=void 0===l?r2:l,l=c.nodeCallback,H=void 0===l?r2:l,l=c.pseudoElementsCallback,t=void 0===l?r2:l,c=void 0===(c=c.observeMutationsRoot)?v:c,L2=new o(function(c){var z;f2||(z=w1,s1(c).forEach(function(c){var l,s,a;\"childList\"===c.type&&0<c.addedNodes.length&&!M2(c.addedNodes[0])&&(Q.searchPseudoElements&&t(c.target),e(c.target)),\"attributes\"===c.type&&c.target.parentNode&&Q.searchPseudoElements&&t(c.target.parentNode),\"attributes\"===c.type&&M2(c.target)&&~W.indexOf(c.attributeName)&&(\"class\"===c.attributeName&&(s=c.target,a=s.getAttribute?s.getAttribute(y):null,s=s.getAttribute?s.getAttribute(S):null,a&&s)?(s=(l=D1(a1(c.target))).prefix,l=l.iconName,c.target.setAttribute(y,s||z),l&&c.target.setAttribute(S,l)):(l=c.target)&&l.classList&&l.classList.contains&&l.classList.contains(Q.replacementClass)&&H(c.target))}))}),L&&L2.observe(c,{childList:!0,attributes:!0,characterData:!0,subtree:!0}))}function d2(c){var l=c.getAttribute(\"data-prefix\"),s=c.getAttribute(\"data-icon\"),a=void 0!==c.innerText?c.innerText.trim():\"\",z=D1(a1(c));return z.prefix||(z.prefix=w1),l&&s&&(z.prefix=l,z.iconName=s),z.iconName&&z.prefix||(z.prefix&&0<a.length&&(z.iconName=(s=z.prefix,a=c.innerText,(y1[s]||{})[a]||j1(z.prefix,L1(c.innerText)))),!z.iconName&&Q.autoFetchSvg&&c.firstChild&&c.firstChild.nodeType===Node.TEXT_NODE&&(z.iconName=c.firstChild.data)),z}function p2(c,l){var s=1<arguments.length&&void 0!==l?l:{styleParser:!0},a=d2(c),z=a.iconName,e=a.prefix,H=a.rest,t=(l=s1((t=c).attributes).reduce(function(c,l){return\"class\"!==c.name&&\"style\"!==c.name&&(c[l.name]=l.value),c},{}),a=t.getAttribute(\"title\"),t=t.getAttribute(\"data-fa-title-id\"),Q.autoA11y&&(a?l[\"aria-labelledby\"]=\"\".concat(Q.replacementClass,\"-title-\").concat(t||l1()):(l[\"aria-hidden\"]=\"true\",l.focusable=\"false\")),l),l=_1(\"parseNodeAttributes\",{},c),V=s.styleParser?(s=(V=c).getAttribute(\"style\"),V=[],V=s?s.split(\";\").reduce(function(c,l){var s=l.split(\":\"),l=s[0],s=s.slice(1);return l&&0<s.length&&(c[l]=s.join(\":\").trim()),c},{}):V):[];return u({iconName:z,title:c.getAttribute(\"title\"),titleId:c.getAttribute(\"data-fa-title-id\"),prefix:e,transform:J,mask:{iconName:null,prefix:null,rest:[]},maskId:null,symbol:!1,extra:{classes:H,styles:V,attributes:t}},l)}var b2=h1.styles;function g2(c){var l=\"nest\"===Q.autoReplaceSvg?p2(c,{styleParser:!1}):p2(c);return~l.extra.classes.indexOf(R)?U1(\"generateLayersText\",c,l):U1(\"generateSvgReplacementMutation\",c,l)}var w2=new Set;function k2(c){var a=1<arguments.length&&void 0!==arguments[1]?arguments[1]:null;if(!L)return Promise.resolve();function z(c){return l.add(\"\".concat(A,\"-\").concat(c))}function e(c){return l.remove(\"\".concat(A,\"-\").concat(c))}var l=v.documentElement.classList,s=Q.autoFetchSvg?w2:P.map(function(c){return\"fa-\".concat(c)}).concat(Object.keys(b2));s.includes(\"fa\")||s.push(\"fa\");var H=[\".\".concat(R,\":not([\").concat(g,\"])\")].concat(s.map(function(c){return\".\".concat(c,\":not([\").concat(g,\"])\")})).join(\", \");if(0===H.length)return Promise.resolve();s=[];try{s=s1(c.querySelectorAll(H))}catch(c){}if(!(0<s.length))return Promise.resolve();z(\"pending\"),e(\"complete\");var t=V2.begin(\"onTree\"),V=s.reduce(function(c,l){try{var s=g2(l);s&&c.push(s)}catch(c){Z||\"MissingIcon\"===c.name&&console.error(c)}return c},[]);return new Promise(function(l,s){Promise.all(V).then(function(c){o2(c,function(){z(\"active\"),z(\"complete\"),e(\"pending\"),\"function\"==typeof a&&a(),t(),l()})}).catch(function(c){t(),s(c)})})}function y2(c){var l=1<arguments.length&&void 0!==arguments[1]?arguments[1]:null;g2(c).then(function(c){c&&o2([c],l)})}P.map(function(c){w2.add(\"fa-\".concat(c))}),Object.keys(E[O]).map(w2.add.bind(w2)),Object.keys(E[j]).map(w2.add.bind(w2));function S2(c){var l=1<arguments.length&&void 0!==arguments[1]?arguments[1]:{},s=l.transform,a=void 0===s?J:s,z=void 0!==(s=l.symbol)&&s,e=void 0===(s=l.mask)?null:s,H=void 0===(s=l.maskId)?null:s,t=void 0===(s=l.title)?null:s,V=void 0===(s=l.titleId)?null:s,r=void 0===(s=l.classes)?[]:s,M=void 0===(s=l.attributes)?{}:s,h=void 0===(s=l.styles)?{}:s;if(c){var n=c.prefix,i=c.iconName,m=c.icon;return K1(u({type:\"icon\"},c),function(){return W1(\"beforeDOMElementCreation\",{iconDefinition:c,params:l}),Q.autoA11y&&(t?M[\"aria-labelledby\"]=\"\".concat(Q.replacementClass,\"-title-\").concat(V||l1()):(M[\"aria-hidden\"]=\"true\",M.focusable=\"false\")),$1({icons:{main:l2(m),mask:e?l2(e.icon):{found:!1,width:null,height:null,icon:{}}},prefix:n,iconName:i,transform:u(u({},J),a),symbol:z,title:t,maskId:H,titleId:V,extra:{attributes:M,styles:h,classes:r}})})}}var w2=M(w2),r={mixout:function(){return{icon:(a=S2,function(c){var l=1<arguments.length&&void 0!==arguments[1]?arguments[1]:{},s=(c||{}).icon?c:X1(c||{}),c=(c=l.mask)&&((c||{}).icon?c:X1(c||{}));return a(s,u(u({},l),{},{mask:c}))})};var a},hooks:function(){return{mutationObserverCallbacks:function(c){return c.treeCallback=k2,c.nodeCallback=y2,c}}},provides:function(c){c.i2svg=function(c){var l=c.node,c=c.callback;return k2(void 0===l?v:l,void 0===c?function(){}:c)},c.generateSvgReplacementMutation=function(a,c){var z=c.iconName,e=c.title,H=c.titleId,t=c.prefix,V=c.transform,r=c.symbol,l=c.mask,M=c.maskId,h=c.extra;return new Promise(function(s,c){Promise.all([a2(z,t),l.iconName?a2(l.iconName,l.prefix):Promise.resolve({found:!1,width:512,height:512,icon:{}})]).then(function(c){var l=n(c,2),c=l[0],l=l[1];s([a,$1({icons:{main:c,mask:l},prefix:t,iconName:z,transform:V,symbol:r,maskId:M,title:e,titleId:H,extra:h,watchable:!0})])}).catch(c)})},c.generateAbstractIcon=function(c){var l,s=c.children,a=c.attributes,z=c.main,e=c.transform,c=e1(c.styles);return 0<c.length&&(a.style=c),H1(e)&&(l=U1(\"generateAbstractTransformGrouping\",{main:z,transform:e,containerWidth:z.width,iconWidth:z.width})),s.push(l||z.icon),{children:s,attributes:a}}}},l={mixout:function(){return{layer:function(c){var s=1<arguments.length&&void 0!==arguments[1]?arguments[1]:{},l=s.classes,a=void 0===l?[]:l;return K1({type:\"layer\"},function(){W1(\"beforeDOMElementCreation\",{assembler:c,params:s});var l=[];return c(function(c){Array.isArray(c)?c.map(function(c){l=l.concat(c.abstract)}):l=l.concat(c.abstract)}),[{tag:\"span\",attributes:{class:[\"\".concat(Q.cssPrefix,\"-layers\")].concat(M(a)).join(\" \")},children:l}]})}}}},h={mixout:function(){return{counter:function(z){var e=1<arguments.length&&void 0!==arguments[1]?arguments[1]:{},c=e.title,H=void 0===c?null:c,c=e.classes,t=void 0===c?[]:c,c=e.attributes,V=void 0===c?{}:c,c=e.styles,r=void 0===c?{}:c;return K1({type:\"counter\",content:z},function(){return W1(\"beforeDOMElementCreation\",{content:z,params:e}),c={content:z.toString(),title:H,extra:{attributes:V,styles:r,classes:[\"\".concat(Q.cssPrefix,\"-layers-counter\")].concat(M(t))}},l=c.content,s=c.title,a=c.extra,c=u(u(u({},a.attributes),s?{title:s}:{}),{},{class:a.classes.join(\" \")}),0<(a=e1(a.styles)).length&&(c.style=a),(a=[]).push({tag:\"span\",attributes:c,children:[l]}),s&&a.push({tag:\"span\",attributes:{class:\"sr-only\"},children:[s]}),a;var c,l,s,a})}}}},i={mixout:function(){return{text:function(c){var l=1<arguments.length&&void 0!==arguments[1]?arguments[1]:{},s=l.transform,a=void 0===s?J:s,s=l.title,z=void 0===s?null:s,s=l.classes,e=void 0===s?[]:s,s=l.attributes,H=void 0===s?{}:s,s=l.styles,t=void 0===s?{}:s;return K1({type:\"text\",content:c},function(){return W1(\"beforeDOMElementCreation\",{content:c,params:l}),J1({content:c,transform:u(u({},J),a),title:z,extra:{attributes:H,styles:t,classes:[\"\".concat(Q.cssPrefix,\"-layers-text\")].concat(M(e))}})})}}},provides:function(c){c.generateLayersText=function(c,l){var s,a=l.title,z=l.transform,e=l.extra,H=null,t=null;return d&&(s=parseInt(getComputedStyle(c).fontSize,10),H=(l=c.getBoundingClientRect()).width/s,t=l.height/s),Q.autoA11y&&!a&&(e.attributes[\"aria-hidden\"]=\"true\"),Promise.resolve([c,J1({content:c.innerHTML,width:H,height:t,transform:z,title:a,extra:e,watchable:!0})])}}},A2=new RegExp('\"',\"ug\"),x2=[1105920,1112319];function q2(m,o){var f=\"\".concat(k).concat(o.replace(\":\",\"-\"));return new Promise(function(s,c){if(null!==m.getAttribute(f))return s();var a,z,e,H,l,t,V,r=s1(m.children).filter(function(c){return c.getAttribute(w)===o})[0],M=C.getComputedStyle(m,o),h=M.getPropertyValue(\"font-family\").match(F),n=M.getPropertyValue(\"font-weight\"),i=M.getPropertyValue(\"content\");if(r&&!h)return m.removeChild(r),s();h&&\"none\"!==i&&\"\"!==i?(t=M.getPropertyValue(\"content\"),l=~[\"Sharp\"].indexOf(h[2])?j:O,a=~[\"Solid\",\"Regular\",\"Light\",\"Thin\",\"Duotone\",\"Brands\",\"Kit\"].indexOf(h[2])?I[l][h[2].toLowerCase()]:_[l][n],i=(M=(M=(i=t).replace(A2,\"\"),l=0,t=(n=M).length,V=55296<=(i=n.charCodeAt(l))&&i<=56319&&l+1<t&&56320<=(V=n.charCodeAt(l+1))&&V<=57343?1024*(i-55296)+V-56320+65536:i,i=x2[0]<=V&&V<=x2[1],{value:L1((V=2===M.length&&M[0]===M[1])?M[0]:M),isSecondary:i||V})).value,V=M.isSecondary,M=h[0].startsWith(\"FontAwesome\"),h=j1(a,i),z=h,M&&(i=A1[M=i],M=j1(\"fas\",M),(M=i||(M?{prefix:\"fas\",iconName:M}:null)||{prefix:null,iconName:null}).iconName&&M.prefix&&(h=M.iconName,a=M.prefix)),!h||V||r&&r.getAttribute(y)===a&&r.getAttribute(S)===z?s():(m.setAttribute(f,z),r&&m.removeChild(r),(H=(e={iconName:null,title:null,titleId:null,prefix:null,transform:J,symbol:!1,mask:{iconName:null,prefix:null,rest:[]},maskId:null,extra:{classes:[],styles:{},attributes:{}}}).extra).attributes[w]=o,a2(h,a).then(function(c){var l=$1(u(u({},e),{},{icons:{main:c,mask:E1()},prefix:a,iconName:z,extra:H,watchable:!0})),c=v.createElementNS(\"http://www.w3.org/2000/svg\",\"svg\");\"::before\"===o?m.insertBefore(c,m.firstChild):m.appendChild(c),c.outerHTML=l.map(o1).join(\"\\n\"),m.removeAttribute(f),s()}).catch(c))):s()})}function Z2(c){return Promise.all([q2(c,\"::before\"),q2(c,\"::after\")])}function O2(c){return!(c.parentNode===document.head||~q.indexOf(c.tagName.toUpperCase())||c.getAttribute(w)||c.parentNode&&\"svg\"===c.parentNode.tagName)}function j2(z){if(L)return new Promise(function(c,l){var s=s1(z.querySelectorAll(\"*\")).filter(O2).map(Z2),a=V2.begin(\"searchPseudoElements\");C2(),Promise.all(s).then(function(){a(),v2(),c()}).catch(function(){a(),v2(),l()})})}function P2(c){return c.toLowerCase().split(\" \").reduce(function(c,l){var s=l.toLowerCase().split(\"-\"),l=s[0],a=s.slice(1).join(\"-\");if(l&&\"h\"===a)return c.flipX=!0,c;if(l&&\"v\"===a)return c.flipY=!0,c;if(a=parseFloat(a),isNaN(a))return c;switch(l){case\"grow\":c.size=c.size+a;break;case\"shrink\":c.size=c.size-a;break;case\"left\":c.x=c.x-a;break;case\"right\":c.x=c.x+a;break;case\"up\":c.y=c.y-a;break;case\"down\":c.y=c.y+a;break;case\"rotate\":c.rotate=c.rotate+a}return c},{size:16,x:0,y:0,flipX:!1,flipY:!1,rotate:0})}var N2=!1,E2={x:0,y:0,width:\"100%\",height:\"100%\"};function I2(c){return c.attributes&&(c.attributes.fill||(!(1<arguments.length&&void 0!==arguments[1])||arguments[1]))&&(c.attributes.fill=\"black\"),c}var T2;T2={mixoutsTo:G1}.mixoutsTo,V=[m,r,l,h,i,{hooks:function(){return{mutationObserverCallbacks:function(c){return c.pseudoElementsCallback=j2,c}}},provides:function(c){c.pseudoElements2svg=function(c){c=c.node;Q.searchPseudoElements&&j2(void 0===c?v:c)}}},{mixout:function(){return{dom:{unwatch:function(){C2(),N2=!0}}}},hooks:function(){return{bootstrap:function(){u2(_1(\"mutationObserverCallbacks\",{}))},noAuto:function(){L2&&L2.disconnect()},watch:function(c){c=c.observeMutationsRoot;N2?v2():u2(_1(\"mutationObserverCallbacks\",{observeMutationsRoot:c}))}}}},{mixout:function(){return{parse:{transform:P2}}},hooks:function(){return{parseNodeAttributes:function(c,l){l=l.getAttribute(\"data-fa-transform\");return l&&(c.transform=P2(l)),c}}},provides:function(c){c.generateAbstractTransformGrouping=function(c){var l=c.main,s=c.transform,a=c.containerWidth,z=c.iconWidth,e={transform:\"translate(\".concat(a/2,\" 256)\")},c=\"translate(\".concat(32*s.x,\", \").concat(32*s.y,\") \"),a=\"scale(\".concat(s.size/16*(s.flipX?-1:1),\", \").concat(s.size/16*(s.flipY?-1:1),\") \"),s=\"rotate(\".concat(s.rotate,\" 0 0)\"),z={outer:e,inner:{transform:\"\".concat(c,\" \").concat(a,\" \").concat(s)},path:{transform:\"translate(\".concat(z/2*-1,\" -256)\")}};return{tag:\"g\",attributes:u({},z.outer),children:[{tag:\"g\",attributes:u({},z.inner),children:[{tag:l.icon.tag,children:l.icon.children,attributes:u(u({},l.icon.attributes),z.path)}]}]}}}},{hooks:function(){return{parseNodeAttributes:function(c,l){var s=l.getAttribute(\"data-fa-mask\"),s=s?D1(s.split(\" \").map(function(c){return c.trim()})):E1();return s.prefix||(s.prefix=w1),c.mask=s,c.maskId=l.getAttribute(\"data-fa-mask-id\"),c}}},provides:function(c){c.generateAbstractMask=function(c){var l=c.children,s=c.attributes,a=c.main,z=c.mask,e=c.maskId,H=c.transform,t=a.width,V=a.icon,r=z.width,c=z.icon,H=(z=(a={transform:H,containerWidth:r,iconWidth:t}).transform,H=a.containerWidth,r=a.iconWidth,t={transform:\"translate(\".concat(H/2,\" 256)\")},a=\"translate(\".concat(32*z.x,\", \").concat(32*z.y,\") \"),H=\"scale(\".concat(z.size/16*(z.flipX?-1:1),\", \").concat(z.size/16*(z.flipY?-1:1),\") \"),z=\"rotate(\".concat(z.rotate,\" 0 0)\"),{outer:t,inner:{transform:\"\".concat(a,\" \").concat(H,\" \").concat(z)},path:{transform:\"translate(\".concat(r/2*-1,\" -256)\")}}),z={tag:\"rect\",attributes:u(u({},E2),{},{fill:\"white\"})},r=V.children?{children:V.children.map(I2)}:{},r={tag:\"g\",attributes:u({},H.inner),children:[I2(u({tag:V.tag,attributes:u(u({},V.attributes),H.path)},r))]},H={tag:\"g\",attributes:u({},H.outer),children:[r]},r=\"mask-\".concat(e||l1()),e=\"clip-\".concat(e||l1()),H={tag:\"mask\",attributes:u(u({},E2),{},{id:r,maskUnits:\"userSpaceOnUse\",maskContentUnits:\"userSpaceOnUse\"}),children:[z,H]},H={tag:\"defs\",children:[{tag:\"clipPath\",attributes:{id:e},children:\"g\"===(c=c).tag?c.children:[c]},H]};return l.push(H,{tag:\"rect\",attributes:u({fill:\"currentColor\",\"clip-path\":\"url(#\".concat(e,\")\"),mask:\"url(#\".concat(r,\")\")},E2)}),{children:l,attributes:s}}}},{provides:function(c){var e=!1;C.matchMedia&&(e=C.matchMedia(\"(prefers-reduced-motion: reduce)\").matches),c.missingIconAbstract=function(){var c=[],l={fill:\"currentColor\"},s={attributeType:\"XML\",repeatCount:\"indefinite\",dur:\"2s\"};c.push({tag:\"path\",attributes:u(u({},l),{},{d:\"M156.5,447.7l-12.6,29.5c-18.7-9.5-35.9-21.2-51.5-34.9l22.7-22.7C127.6,430.5,141.5,440,156.5,447.7z M40.6,272H8.5 c1.4,21.2,5.4,41.7,11.7,61.1L50,321.2C45.1,305.5,41.8,289,40.6,272z M40.6,240c1.4-18.8,5.2-37,11.1-54.1l-29.5-12.6 C14.7,194.3,10,216.7,8.5,240H40.6z M64.3,156.5c7.8-14.9,17.2-28.8,28.1-41.5L69.7,92.3c-13.7,15.6-25.5,32.8-34.9,51.5 L64.3,156.5z M397,419.6c-13.9,12-29.4,22.3-46.1,30.4l11.9,29.8c20.7-9.9,39.8-22.6,56.9-37.6L397,419.6z M115,92.4 c13.9-12,29.4-22.3,46.1-30.4l-11.9-29.8c-20.7,9.9-39.8,22.6-56.8,37.6L115,92.4z M447.7,355.5c-7.8,14.9-17.2,28.8-28.1,41.5 l22.7,22.7c13.7-15.6,25.5-32.9,34.9-51.5L447.7,355.5z M471.4,272c-1.4,18.8-5.2,37-11.1,54.1l29.5,12.6 c7.5-21.1,12.2-43.5,13.6-66.8H471.4z M321.2,462c-15.7,5-32.2,8.2-49.2,9.4v32.1c21.2-1.4,41.7-5.4,61.1-11.7L321.2,462z M240,471.4c-18.8-1.4-37-5.2-54.1-11.1l-12.6,29.5c21.1,7.5,43.5,12.2,66.8,13.6V471.4z M462,190.8c5,15.7,8.2,32.2,9.4,49.2h32.1 c-1.4-21.2-5.4-41.7-11.7-61.1L462,190.8z M92.4,397c-12-13.9-22.3-29.4-30.4-46.1l-29.8,11.9c9.9,20.7,22.6,39.8,37.6,56.9 L92.4,397z M272,40.6c18.8,1.4,36.9,5.2,54.1,11.1l12.6-29.5C317.7,14.7,295.3,10,272,8.5V40.6z M190.8,50 c15.7-5,32.2-8.2,49.2-9.4V8.5c-21.2,1.4-41.7,5.4-61.1,11.7L190.8,50z M442.3,92.3L419.6,115c12,13.9,22.3,29.4,30.5,46.1 l29.8-11.9C470,128.5,457.3,109.4,442.3,92.3z M397,92.4l22.7-22.7c-15.6-13.7-32.8-25.5-51.5-34.9l-12.6,29.5 C370.4,72.1,384.4,81.5,397,92.4z\"})});var a=u(u({},s),{},{attributeName:\"opacity\"}),z={tag:\"circle\",attributes:u(u({},l),{},{cx:\"256\",cy:\"364\",r:\"28\"}),children:[]};return e||z.children.push({tag:\"animate\",attributes:u(u({},s),{},{attributeName:\"r\",values:\"28;14;28;28;14;28;\"})},{tag:\"animate\",attributes:u(u({},a),{},{values:\"1;0;1;1;0;1;\"})}),c.push(z),c.push({tag:\"path\",attributes:u(u({},l),{},{opacity:\"1\",d:\"M263.7,312h-16c-6.6,0-12-5.4-12-12c0-71,77.4-63.9,77.4-107.8c0-20-17.8-40.2-57.4-40.2c-29.1,0-44.3,9.6-59.2,28.7 c-3.9,5-11.1,6-16.2,2.4l-13.1-9.2c-5.6-3.9-6.9-11.8-2.6-17.2c21.2-27.2,46.4-44.7,91.2-44.7c52.3,0,97.4,29.8,97.4,80.2 c0,67.6-77.4,63.5-77.4,107.8C275.7,306.6,270.3,312,263.7,312z\"}),children:e?[]:[{tag:\"animate\",attributes:u(u({},a),{},{values:\"1;0;0;0;0;1;\"})}]}),e||c.push({tag:\"path\",attributes:u(u({},l),{},{opacity:\"0\",d:\"M232.5,134.5l7,168c0.3,6.4,5.6,11.5,12,11.5h9c6.4,0,11.7-5.1,12-11.5l7-168c0.3-6.8-5.2-12.5-12-12.5h-23 C237.7,122,232.2,127.7,232.5,134.5z\"}),children:[{tag:\"animate\",attributes:u(u({},a),{},{values:\"0;0;1;1;0;0;\"})}]}),{tag:\"g\",attributes:{class:\"missing\"},children:c}}}},{hooks:function(){return{parseNodeAttributes:function(c,l){l=l.getAttribute(\"data-fa-symbol\");return c.symbol=null!==l&&(\"\"===l||l),c}}}}],Y1={},Object.keys(R1).forEach(function(c){-1===F1.indexOf(c)&&delete R1[c]}),V.forEach(function(c){var l,s=c.mixout?c.mixout():{};Object.keys(s).forEach(function(l){\"function\"==typeof s[l]&&(T2[l]=s[l]),\"object\"===z(s[l])&&Object.keys(s[l]).forEach(function(c){T2[l]||(T2[l]={}),T2[l][c]=s[l][c]})}),c.hooks&&(l=c.hooks(),Object.keys(l).forEach(function(c){Y1[c]||(Y1[c]=[]),Y1[c].push(l[c])})),c.provides&&c.provides(R1)}),function(c){try{for(var l=arguments.length,s=new Array(1<l?l-1:0),a=1;a<l;a++)s[a-1]=arguments[a];c.apply(void 0,s)}catch(c){if(!Z)throw c}}(function(c){f&&(C.FontAwesome||(C.FontAwesome=G1),m1(function(){Q1(),W1(\"bootstrap\")})),h1.hooks=u(u({},h1.hooks),{},{addPack:function(c,l){h1.styles[c]=u(u({},h1.styles[c]||{}),l),O1(),Q1()},addPacks:function(c){c.forEach(function(c){var l=n(c,2),c=l[0],l=l[1];h1.styles[c]=u(u({},h1.styles[c]||{}),l)}),O1(),Q1()},addShims:function(c){var l;(l=h1.shims).push.apply(l,M(c)),O1(),Q1()}})})}();","// The module cache\nvar __webpack_module_cache__ = {};\n\n// The require function\nfunction __webpack_require__(moduleId) {\n\t// Check if module is in cache\n\tvar cachedModule = __webpack_module_cache__[moduleId];\n\tif (cachedModule !== undefined) {\n\t\treturn cachedModule.exports;\n\t}\n\t// Create a new module (and put it into the cache)\n\tvar module = __webpack_module_cache__[moduleId] = {\n\t\t// no module.id needed\n\t\t// no module.loaded needed\n\t\texports: {}\n\t};\n\n\t// Execute the module function\n\t__webpack_modules__[moduleId](module, module.exports, __webpack_require__);\n\n\t// Return the exports of the module\n\treturn module.exports;\n}\n\n","// getDefaultExport function for compatibility with non-harmony modules\n__webpack_require__.n = (module) => {\n\tvar getter = module && module.__esModule ?\n\t\t() => (module['default']) :\n\t\t() => (module);\n\t__webpack_require__.d(getter, { a: getter });\n\treturn getter;\n};","// define getter functions for harmony exports\n__webpack_require__.d = (exports, definition) => {\n\tfor(var key in definition) {\n\t\tif(__webpack_require__.o(definition, key) && !__webpack_require__.o(exports, key)) {\n\t\t\tObject.defineProperty(exports, key, { enumerable: true, get: definition[key] });\n\t\t}\n\t}\n};","__webpack_require__.o = (obj, prop) => (Object.prototype.hasOwnProperty.call(obj, prop))"],"names":["c","l","window","document","a","s","navigator","userAgent","z","e","H","Object","keys","getOwnPropertySymbols","filter","getOwnPropertyDescriptor","enumerable","push","apply","t","arguments","length","forEach","V","getOwnPropertyDescriptors","defineProperties","defineProperty","value","configurable","writable","r","Array","documentElement","head","addEventListener","createElement","indexOf","M","h","n","i","m","o","Proxy","get","C","fa","fas","far","fal","fat","fad","fab","fak","fakd","fass","fasr","fasl","fast","f","v","solid","regular","light","thin","duotone","brands","kit","normal","concat","Set","map","add","bind","isArray","Symbol","iterator","from","prototype","toString","call","slice","constructor","name","test","TypeError","styles","hooks","shims","L","u","reduce","icon","iconName","d","skipHooks","addPack","p","monero","hooli","yelp","lastfm","shopware","aws","redhat","yoast","cloudflare","ups","pixiv","wpexplorer","dyalog","bity","stackpath","buysellads","modx","guilded","vnv","microsoft","qq","orcid","java","invision","centercode","drupal","jxl","unity","whmcs","rocketchat","vk","untappd","mailchimp","contao","deskpro","brave","sistrix","edge","threads","napster","artstation","markdown","sourcetree","diaspora","foursquare","pagelines","algolia","safari","google","atlassian","nimblr","chromecast","evernote","adversal","fonticons","weixin","shirtsinbulk","codepen","lyft","rev","windows","meetup","centos","adn","cloudsmith","opensuse","codiepie","node","mix","steam","scribd","debian","openid","instalod","expeditedssl","sellcast","delicious","freebsd","vuejs","accusoft","ioxhost","golang","kickstarter","grav","weibo","uncharted","firstdraft","wpressr","angellist","skype","joget","fedora","meta","laravel","hotjar","hips","behance","reddit","discord","chrome","wpbeginner","confluence","shoelace","mdb","dochub","ebay","amazon","unsplash","yarn","asymmetrik","gratipay","apple","hive","gitkraken","keybase","padlet","stumbleupon","fedex","shopify","neos","hackerrank","researchgate","swift","angular","speakap","angrycreative","empire","envira","studiovinari","wordpress","firefox","linode","goodreads","jsfiddle","sith","themeisle","page4","hashnode","react","squarespace","bitcoin","keycdn","opera","umbraco","ubuntu","draft2digital","stripe","houzz","gg","dhl","xing","blackberry","playstation","quinscape","less","opencart","vine","paypal","gitlab","typo3","yahoo","dailymotion","affiliatetheme","bootstrap","odnoklassniki","mintbit","ethereum","patreon","avianex","ello","gofore","bimobject","mandalorian","osi","periscope","fulcrum","cloudscale","forumbee","mizuni","schlix","bandcamp","wpforms","cloudversify","usps","megaport","magento","spotify","fly","aviato","itunes","cuttlefish","blogger","flickr","viber","soundcloud","digg","letterboxd","symfony","maxcdn","etsy","audible","bilibili","erlang","dashcube","elementor","palfed","superpowers","resolving","xbox","searchengin","tiktok","renren","linux","glide","linkedin","hubspot","deploydog","twitch","ravelry","mixer","vimeo","mendeley","uniregistry","figma","dropbox","instagram","cmplid","upwork","facebook","gripfire","uikit","phabricator","ussunnah","earlybirds","autoprefixer","whatsapp","slideshare","viadeo","line","servicestack","simplybuilt","bitbucket","imdb","deezer","jira","docker","screenpal","bluetooth","gitter","microblog","yandex","readme","html5","sellsy","sass","wirsindhandwerk","buromobelexperte","salesforce","medapps","ns8","apper","waze","bluesky","snapchat","rust","wix","supple","webflow","rebel","css3","staylinked","kaggle","deviantart","cpanel","trello","perbyte","grunt","weebly","connectdevelop","leanpub","themeco","python","android","bots","hornbill","js","ideal","git","dev","sketch","uber","github","php","alipay","youtube","skyatlas","replyd","suse","jenkins","twitter","rockrms","pinterest","buffer","npm","yammer","btc","dribbble","stubber","telegram","odysee","slack","medrt","usb","tumblr","vaadin","quora","reacteurope","medium","amilia","mixcloud","flipboard","viacoin","sitrox","discourse","joomla","mastodon","airbnb","gulp","strava","ember","teamspeak","pushed","nutritionix","wodu","intercom","zhihu","korvue","pix","message","comments","paste","compass","lightbulb","flag","futbol","hand","bookmark","folder","user","star","clipboard","image","lemon","handshake","gem","registered","square","snowflake","newspaper","heart","circle","eye","comment","envelope","hourglass","clock","keyboard","images","sun","bell","file","hospital","copy","copyright","building","moon","calendar","clone","at","stethoscope","info","explosion","ring","volleyball","atom","soap","icons","fingerprint","football","crop","person","laptop","menorah","bong","spoon","pager","strikethrough","k","pencil","backward","blog","w","rainbow","paw","cloud","gavel","binoculars","motorcycle","scissors","table","clover","reply","helicopter","swatchbook","bars","film","sitemap","memory","hanukiah","feather","compress","ankh","asterisk","heading","ghost","list","gamepad","egg","campground","paintbrush","lock","tree","microscope","sink","mitten","users","om","worm","plug","stopwatch","stamp","stairs","pills","tooth","bicycle","snowman","school","igloo","joint","horse","q","g","capsules","bullseye","bacon","radiation","vial","gauge","dumpster","highlighter","key","bullhorn","globe","synagogue","ban","repeat","cross","box","maximize","shapes","shuffle","spider","slash","server","venus","passport","microchip","crown","wheelchair","fire","city","unlock","headset","wifi","bath","underline","signature","stroopwafel","bold","jedi","gift","glasses","train","crow","sailboat","frog","bucket","microphone","cow","screwdriver","dolly","smoking","minimize","monument","snowplow","cannabis","tablets","ethernet","chair","icicles","neuter","marker","volcano","viruses","certificate","suitcase","scroll","spa","pause","bomb","subscript","burst","smog","crutch","palette","vest","ferry","seedling","children","chalkboard","virus","archway","couch","italic","church","democrat","locust","sort","language","question","code","leaf","road","taxi","poop","kaaba","eject","tarp","cube","elevator","recycle","trademark","basketball","wallet","burger","wrench","bugs","bridge","cat","route","panorama","tags","terminal","tape","pen","signal","bus","prescription","shop","vihara","diamond","bacterium","biohazard","phone","trash","poo","shirt","cubes","divide","headphones","republican","ruler","restroom","j","otter","child","satellite","tag","paperclip","ribbon","lungs","indent","mountain","camera","meteor","sleigh","water","braille","landmark","truck","crosshairs","tent","cookie","dumbbell","dna","minus","chess","gear","mosque","mosquito","vials","pallet","faucet","timeline","coins","khanda","sliders","hamsa","flask","ticket","tty","calculator","ship","download","forward","mobile","outdent","house","b","utensils","skull","stop","upload","hurricane","mound","caravan","bolt","vault","mars","toilet","guitar","industry","ellipsis","toolbox","bug","car","medal","bed","podcast","superscript","droplet","eraser","dove","socks","inbox","section","dharmachakra","hotdog","drum","fax","paragraph","link","play","font","receipt","tv","shrimp","wind","y","fish","clapperboard","baseball","grip","gun","plus","expand","computer","xmark","baby","tractor","equals","blender","teeth","rocket","store","tablet","fill","bacteria","notdef","disease","genderless","retweet","radio","thermometer","percent","display","thumbtack","trophy","hammer","rotate","spinner","robot","peace","gears","warehouse","splotch","transgender","mercury","award","qrcode","shield","tents","dog","carrot","cheese","music","broom","gopuram","hashtag","hippo","infinity","voicemail","fan","trailer","bahai","dragon","tornado","anchor","rss","shower","desktop","book","check","briefcase","rug","handcuffs","database","share","dungeon","hands","dice","brain","bandage","gifts","hotel","brush","mask","syringe","magnet","jar","bone","plane","exclamation","print","x","umbrella","trowel","stapler","barcode","video","next","done","return","mark","measure","MutationObserver","performance","S","A","Z","O","P","N","E","I","T","D","Y","R","F","_","W","U","GROUP","SWAP_OPACITY","PRIMARY","SECONDARY","X","B","FontAwesomeConfig","querySelector","getAttribute","styleDefault","familyDefault","cssPrefix","replacementClass","autoReplaceSvg","autoAddCss","autoA11y","searchPseudoElements","observeMutations","mutateApproach","keepOriginalSource","measurePerformance","showMissingIcons","familyPrefix","G","Q","set","K","$","J","size","flipX","flipY","c1","l1","Math","random","s1","a1","classList","split","z1","replace","e1","trim","H1","t1","RegExp","V1","r1","setAttribute","innerHTML","childNodes","tagName","toUpperCase","insertBefore","mixout","dom","css","insertCss","beforeDOMElementCreation","beforeI2svg","h1","n1","i1","m1","setTimeout","o1","attributes","join","f1","prefix","C1","v1","doScroll","readyState","M1","removeEventListener","L1","charCodeAt","u1","d1","p1","b1","g1","values","w1","k1","y1","S1","A1","x1","q1","O1","autoFetchSvg","names","unicodes","I1","family","j1","P1","N1","T1","D1","skipLookups","includes","some","Z1","rest","this","definitions","_pullDefinitions","Y1","R1","F1","_1","W1","U1","X1","B1","G1","noAuto","config","i2svg","Promise","reject","watch","autoReplaceSvgRoot","Q1","parse","match","library","findIconDefinition","toHtml","K1","abstract","html","$1","main","transform","symbol","title","maskId","titleId","extra","watchable","found","width","height","classes","class","role","xmlns","viewBox","id","style","J1","content","startCentered","c2","l2","s2","a2","console","error","z2","e2","H2","t2","V2","begin","end","r2","M2","h2","createElementNS","n2","i2","parentNode","ceFn","createTextNode","appendChild","createComment","outerHTML","replaceChild","remove","nest","toSvg","toNode","removeAttribute","m2","o2","requestAnimationFrame","f2","C2","v2","L2","u2","treeCallback","nodeCallback","pseudoElementsCallback","observeMutationsRoot","type","addedNodes","target","attributeName","contains","observe","childList","characterData","subtree","p2","styleParser","innerText","firstChild","nodeType","Node","TEXT_NODE","data","d2","focusable","b2","g2","w2","k2","resolve","querySelectorAll","all","then","catch","y2","S2","iconDefinition","params","mutationObserverCallbacks","provides","callback","generateSvgReplacementMutation","generateAbstractIcon","containerWidth","iconWidth","layer","assembler","counter","text","generateLayersText","parseInt","getComputedStyle","fontSize","getBoundingClientRect","A2","x2","q2","getPropertyValue","removeChild","toLowerCase","isSecondary","startsWith","Z2","O2","j2","P2","parseFloat","isNaN","T2","N2","E2","I2","pseudoElements2svg","unwatch","disconnect","parseNodeAttributes","generateAbstractTransformGrouping","outer","inner","path","generateAbstractMask","maskUnits","maskContentUnits","matchMedia","matches","missingIconAbstract","attributeType","repeatCount","dur","cx","cy","opacity","FontAwesome","addPacks","addShims","__webpack_module_cache__","__webpack_require__","moduleId","cachedModule","undefined","exports","module","__webpack_modules__","getter","__esModule","definition","obj","prop","hasOwnProperty"],"sourceRoot":""}
\ No newline at end of file
diff --git a/_static/scripts/pydata-sphinx-theme.js b/_static/scripts/pydata-sphinx-theme.js
new file mode 100644
index 000000000..771fed0d2
--- /dev/null
+++ b/_static/scripts/pydata-sphinx-theme.js
@@ -0,0 +1,2 @@
+(()=>{"use strict";function e(e){"loading"!=document.readyState?e():document.addEventListener("DOMContentLoaded",e)}const t=e=>"string"==typeof e&&/^[v\d]/.test(e)&&o.test(e),n=(e,t,n)=>{u(n);const o=((e,t)=>{const n=r(e),o=r(t),a=n.pop(),s=o.pop(),i=c(n,o);return 0!==i?i:a&&s?c(a.split("."),s.split(".")):a||s?a?-1:1:0})(e,t);return l[n].includes(o)},o=/^[v^~<>=]*?(\d+)(?:\.([x*]|\d+)(?:\.([x*]|\d+)(?:\.([x*]|\d+))?(?:-([\da-z\-]+(?:\.[\da-z\-]+)*))?(?:\+[\da-z\-]+(?:\.[\da-z\-]+)*)?)?)?$/i,r=e=>{if("string"!=typeof e)throw new TypeError("Invalid argument expected string");const t=e.match(o);if(!t)throw new Error(`Invalid argument not valid semver ('${e}' received)`);return t.shift(),t},a=e=>"*"===e||"x"===e||"X"===e,s=e=>{const t=parseInt(e,10);return isNaN(t)?e:t},i=(e,t)=>{if(a(e)||a(t))return 0;const[n,o]=((e,t)=>typeof e!=typeof t?[String(e),String(t)]:[e,t])(s(e),s(t));return n>o?1:n<o?-1:0},c=(e,t)=>{for(let n=0;n<Math.max(e.length,t.length);n++){const o=i(e[n]||"0",t[n]||"0");if(0!==o)return o}return 0},l={">":[1],">=":[0,1],"=":[0],"<=":[-1,0],"<":[-1]},d=Object.keys(l),u=e=>{if("string"!=typeof e)throw new TypeError("Invalid operator type, expected string but got "+typeof e);if(-1===d.indexOf(e))throw new Error(`Invalid operator, expected one of ${d.join("|")}`)};var m=window.matchMedia("(prefers-color-scheme: dark)");function h(e){document.documentElement.dataset.theme=m.matches?"dark":"light"}function p(e){"light"!==e&&"dark"!==e&&"auto"!==e&&(console.error(`Got invalid theme mode: ${e}. Resetting to auto.`),e="auto");var t=m.matches?"dark":"light";document.documentElement.dataset.mode=e;var n="auto"==e?t:e;document.documentElement.dataset.theme=n,document.querySelectorAll(".dropdown-menu").forEach((e=>{"dark"===n?e.classList.add("dropdown-menu-dark"):e.classList.remove("dropdown-menu-dark")})),localStorage.setItem("mode",e),localStorage.setItem("theme",n),console.log(`[PST]: Changed to ${e} mode using the ${n} theme.`),m.onchange="auto"==e?h:""}function f(){const e=document.documentElement.dataset.defaultMode||"auto",t=localStorage.getItem("mode")||e;var n,o;p(((o=(n=m.matches?["auto","light","dark"]:["auto","dark","light"]).indexOf(t)+1)===n.length&&(o=0),n[o]))}var g=()=>{let e=document.querySelectorAll("form.bd-search");return e.length?(1==e.length?e[0]:document.querySelector(":not(#pst-search-dialog) > form.bd-search")).querySelector("input"):void 0},v=()=>{const e=g(),t=document.getElementById("pst-search-dialog");e===t.querySelector("input")?t.open?t.close():t.showModal():document.activeElement===e?e.blur():(e.focus(),e.select(),e.scrollIntoView({block:"center"}))},y=0===navigator.platform.indexOf("Mac")||"iPhone"===navigator.platform;const w=({currentTarget:e,clientX:t,clientY:n})=>{if(!e.open)return;const{left:o,right:r,top:a,bottom:s}=e.getBoundingClientRect();(t<o||r<t||n<a||s<n)&&e.close()};var E=()=>"dirhtml"==DOCUMENTATION_OPTIONS.BUILDER?DOCUMENTATION_OPTIONS.pagename.endsWith("index")?`${DOCUMENTATION_OPTIONS.pagename.substring(0,DOCUMENTATION_OPTIONS.pagename.length-5)}`:`${DOCUMENTATION_OPTIONS.pagename}/`:`${DOCUMENTATION_OPTIONS.pagename}.html`;async function T(e){document.querySelector("#bd-header-version-warning").remove();const t=DOCUMENTATION_OPTIONS.VERSION,n=new Date,o=JSON.parse(localStorage.getItem("pst_banner_pref")||"{}");console.debug(`[PST] Dismissing the version warning banner on ${t} starting ${n}.`),o[t]=n,localStorage.setItem("pst_banner_pref",JSON.stringify(o))}async function S(e){e.preventDefault();const t=E();let n=e.currentTarget.getAttribute("href"),o=n.replace(t,"");try{(await fetch(n,{method:"HEAD"})).ok?location.href=n:location.href=o}catch(e){location.href=o}}async function b(){var e=document.querySelectorAll(".version-switcher__button");const o=e.length>0,r=DOCUMENTATION_OPTIONS.hasOwnProperty("theme_switcher_json_url"),a=DOCUMENTATION_OPTIONS.show_version_warning_banner;if(r&&(o||a)){const o=await async function(e){const t=E();try{var n=new URL(e)}catch(o){if(!(o instanceof TypeError))throw o;{"file:"==window.location.protocol&&console.info("[PST] looks like you're viewing this site from a local filesystem, so the version switcher won't work unless you've disabled CORS. See https://pydata-sphinx-theme.readthedocs.io/en/stable/user_guide/version-dropdown.html");const o=window.location.href.indexOf(t),r=-1==o?window.location.href:window.location.href.substring(0,o);n=new URL(e,r)}}const o=await fetch(n);return await o.json()}(DOCUMENTATION_OPTIONS.theme_switcher_json_url);o&&(function(e,t){const n=E();t.forEach((e=>{e.dataset.activeVersionName="",e.dataset.activeVersion=""}));const o=(e=e.map((e=>(e.match=e.version==DOCUMENTATION_OPTIONS.theme_switcher_version_match,e.preferred=e.preferred||!1,"name"in e||(e.name=e.version),e)))).map((e=>e.preferred&&e.match)).some(Boolean);var r=!1;e.forEach((e=>{const a=document.createElement("a");a.setAttribute("class","dropdown-item list-group-item list-group-item-action py-1"),a.setAttribute("href",`${e.url}${n}`),a.setAttribute("role","option");const s=document.createElement("span");s.textContent=`${e.name}`,a.appendChild(s),a.dataset.versionName=e.name,a.dataset.version=e.version;let i=o&&e.preferred,c=!o&&!r&&e.match;(i||c)&&(a.classList.add("active"),t.forEach((t=>{t.innerText=e.name,t.dataset.activeVersionName=e.name,t.dataset.activeVersion=e.version})),r=!0),document.querySelectorAll(".version-switcher__menu").forEach((e=>{let t=a.cloneNode(!0);t.onclick=S,e.append(t)}))}))}(o,e),a&&function(e){var o=DOCUMENTATION_OPTIONS.VERSION,r=e.filter((e=>e.preferred));if(1!==r.length){const e=0==r.length?"No":"Multiple";return void console.log(`[PST] ${e} versions marked "preferred" found in versions JSON, ignoring.`)}const a=r[0].version,s=r[0].url,i=t(o)&&t(a);if(i&&n(o,a,"="))return void console.log("[PST]: This is the preferred version of the docs, not showing the warning banner.");const c=JSON.parse(localStorage.getItem("pst_banner_pref")||"{}")[o];if(null!=c){const e=new Date(c),t=(new Date-e)/864e5;if(t<14)return void console.info(`[PST] Suppressing version warning banner; was dismissed ${Math.floor(t)} day(s) ago`)}const l=document.querySelector("#bd-header-version-warning"),d=document.createElement("div"),u=document.createElement("div"),m=document.createElement("strong"),h=document.createElement("a"),p=document.createElement("a");d.classList="bd-header-announcement__content  ms-auto me-auto",u.classList="sidebar-message",h.classList="btn text-wrap font-weight-bold ms-3 my-1 align-baseline pst-button-link-to-stable-version",h.href=`${s}${E()}`,h.innerText="Switch to stable version",h.onclick=S,p.classList="ms-3 my-1 align-baseline";const f=document.createElement("i");p.append(f),f.classList="fa-solid fa-xmark",p.onclick=T,u.innerText="This is documentation for ";const g=o.includes("dev")||o.includes("rc")||o.includes("pre"),v=i&&n(o,a,">");g||v?m.innerText="an unstable development version":i&&n(o,a,"<")?m.innerText=`an old version (${o})`:m.innerText=o?`version ${o}`:"an unknown version",l.appendChild(d),l.append(p),d.appendChild(u),u.appendChild(m),u.appendChild(document.createTextNode(".")),u.appendChild(h),l.classList.remove("d-none")}(o))}}function O(){const e=()=>{document.querySelectorAll(["pre",".nboutput > .output_area",".cell_output > .output",".jp-RenderedHTMLCommon",".pst-scrollable-table-container"].join(", ")).forEach((e=>{e.tabIndex=e.scrollWidth>e.clientWidth||e.scrollHeight>e.clientHeight?0:-1}))},t=function(e,t){let n=null;return(...t)=>{clearTimeout(n),n=setTimeout((()=>{e(...t)}),300)}}(e);window.addEventListener("resize",t),new MutationObserver(t).observe(document.getElementById("main-content"),{subtree:!0,childList:!0}),e()}async function N(){const e=document.querySelector(".bd-header-announcement"),{pstAnnouncementUrl:t}=e?e.dataset:null;if(t)try{const n=await fetch(t);if(!n.ok)throw new Error(`[PST]: HTTP response status not ok: ${n.status} ${n.statusText}`);const o=await n.text();if(0===o.length)return void console.log(`[PST]: Empty announcement at: ${t}`);e.innerHTML=`<div class="bd-header-announcement__content">${o}</div>`,e.classList.remove("d-none")}catch(e){console.log(`[PST]: Failed to load announcement at: ${t}`),console.error(e)}}e((async function(){await Promise.allSettled([b(),N()]);const e=document.querySelector(".pst-async-banner-revealer");if(!e)return;e.classList.remove("d-none");const t=Array.from(e.children).reduce(((e,t)=>e+t.offsetHeight),0);e.style.setProperty("height",`${t}px`),setTimeout((()=>{e.style.setProperty("height","auto")}),320)})),e((function(){p(document.documentElement.dataset.mode),document.querySelectorAll(".theme-switch-button").forEach((e=>{e.addEventListener("click",f)}))})),e((function(){if(!document.querySelector(".bd-docs-nav"))return;var e=document.querySelector("div.bd-sidebar");let t=parseInt(sessionStorage.getItem("sidebar-scroll-top"),10);if(isNaN(t)){var n=document.querySelector(".bd-docs-nav").querySelectorAll(".active");if(n.length>0){var o=n[n.length-1],r=o.getBoundingClientRect().y-e.getBoundingClientRect().y;if(o.getBoundingClientRect().y>.5*window.innerHeight){let t=.25;e.scrollTop=r-e.clientHeight*t,console.log("[PST]: Scrolled sidebar using last active link...")}}}else e.scrollTop=t,console.log("[PST]: Scrolled sidebar using stored browser position...");window.addEventListener("beforeunload",(()=>{sessionStorage.setItem("sidebar-scroll-top",e.scrollTop)}))})),e((function(){window.addEventListener("activate.bs.scrollspy",(function(){document.querySelectorAll(".bd-toc-nav a").forEach((e=>{e.parentElement.classList.remove("active")})),document.querySelectorAll(".bd-toc-nav a.active").forEach((e=>{e.parentElement.classList.add("active")}))}))})),e((()=>{(()=>{let e=document.querySelectorAll(".search-button__kbd-shortcut");y&&e.forEach((e=>e.querySelector("kbd.kbd-shortcut__modifier").innerText="⌘"))})(),window.addEventListener("keydown",(e=>{let t=g();e.shiftKey||e.altKey||(y?!e.metaKey||e.ctrlKey:e.metaKey||!e.ctrlKey)||!/^k$/i.test(e.key)?document.activeElement===t&&/Escape/i.test(e.key)&&v():(e.preventDefault(),v())}),!0),document.querySelectorAll(".search-button__button").forEach((e=>{e.onclick=v})),document.getElementById("pst-search-dialog").addEventListener("click",w)})),e((function(){const e=document.getElementById("pst-primary-sidebar"),t=document.getElementById("pst-secondary-sidebar"),n=document.getElementById("pst-primary-sidebar-modal"),o=document.getElementById("pst-secondary-sidebar-modal"),r=document.querySelector(".primary-toggle"),a=document.querySelector(".secondary-toggle"),s=(e,t)=>{Array.from(e.childNodes).forEach((e=>t.appendChild(e))),Array.from(e.classList).forEach((n=>{e.classList.remove(n),t.classList.add(n)}))};[[r,n,e],[a,o,t]].forEach((([e,t,n])=>{e&&t&&n&&(e.addEventListener("click",(e=>{e.preventDefault(),e.stopPropagation(),s(n,t),t.showModal()})),t.addEventListener("click",w),t.addEventListener("keydown",(e=>{"Escape"===e.key&&(e.preventDefault(),e.stopPropagation(),t.close())})),t.addEventListener("close",(()=>{s(t,n)})))}))})),"complete"===document.readyState?O():window.addEventListener("load",O)})();
+//# sourceMappingURL=pydata-sphinx-theme.js.map
\ No newline at end of file
diff --git a/_static/scripts/pydata-sphinx-theme.js.map b/_static/scripts/pydata-sphinx-theme.js.map
new file mode 100644
index 000000000..0ed1d5e5a
--- /dev/null
+++ b/_static/scripts/pydata-sphinx-theme.js.map
@@ -0,0 +1 @@
+{"version":3,"file":"scripts/pydata-sphinx-theme.js","mappings":"mBASO,SAASA,EAAcC,GACD,WAAvBC,SAASC,WAAyBF,IACjCC,SAASE,iBAAiB,mBAAoBH,EACrD,CCLO,MAiCMI,EAAYC,GAA+B,iBAAZA,GAAwB,SAASC,KAAKD,IAAYE,EAAOD,KAAKD,GAkB7FG,EAAU,CAACC,EAAIC,EAAIC,KAE5BC,EAAoBD,GAGpB,MAAME,EAxDqB,EAACJ,EAAIC,KAEhC,MAAMI,EAAKC,EAAiBN,GACtBO,EAAKD,EAAiBL,GAEtBO,EAAKH,EAAGI,MACRC,EAAKH,EAAGE,MAERE,EAAIC,EAAgBP,EAAIE,GAC9B,OAAU,IAANI,EACOA,EAEPH,GAAME,EACCE,EAAgBJ,EAAGK,MAAM,KAAMH,EAAGG,MAAM,MAE1CL,GAAME,EACJF,GAAM,EAAI,EAEd,CAAC,EAsCIM,CAAgBd,EAAIC,GAChC,OAAOc,EAAeb,GAAUc,SAASZ,EAAI,EAgD3CN,EAAS,6IACTQ,EAAoBV,IACtB,GAAuB,iBAAZA,EACP,MAAM,IAAIqB,UAAU,oCAExB,MAAMC,EAAQtB,EAAQsB,MAAMpB,GAC5B,IAAKoB,EACD,MAAM,IAAIC,MAAM,uCAAuCvB,gBAG3D,OADAsB,EAAME,QACCF,CAAK,EAEVG,EAAcC,GAAY,MAANA,GAAmB,MAANA,GAAmB,MAANA,EAC9CC,EAAYC,IACd,MAAMC,EAAIC,SAASF,EAAG,IACtB,OAAOG,MAAMF,GAAKD,EAAIC,CAAC,EAGrBG,EAAiB,CAACC,EAAGC,KACvB,GAAIT,EAAWQ,IAAMR,EAAWS,GAC5B,OAAO,EACX,MAAOC,EAAIC,GAJG,EAACH,EAAGC,WAAaD,UAAaC,EAAI,CAACG,OAAOJ,GAAII,OAAOH,IAAM,CAACD,EAAGC,GAI5DI,CAAUX,EAASM,GAAIN,EAASO,IACjD,OAAIC,EAAKC,EACE,EACPD,EAAKC,GACG,EACL,CAAC,EAENpB,EAAkB,CAACiB,EAAGC,KACxB,IAAK,IAAIK,EAAI,EAAGA,EAAIC,KAAKC,IAAIR,EAAES,OAAQR,EAAEQ,QAASH,IAAK,CACnD,MAAMxB,EAAIiB,EAAeC,EAAEM,IAAM,IAAKL,EAAEK,IAAM,KAC9C,GAAU,IAANxB,EACA,OAAOA,CACf,CACA,OAAO,CAAC,EAENI,EAAiB,CACnB,IAAK,CAAC,GACN,KAAM,CAAC,EAAG,GACV,IAAK,CAAC,GACN,KAAM,EAAE,EAAG,GACX,IAAK,EAAE,IAELwB,EAAmBC,OAAOC,KAAK1B,GAC/BZ,EAAuBuC,IACzB,GAAkB,iBAAPA,EACP,MAAM,IAAIzB,UAAU,yDAAyDyB,GAEjF,IAAsC,IAAlCH,EAAiBI,QAAQD,GACzB,MAAM,IAAIvB,MAAM,qCAAqCoB,EAAiBK,KAAK,OAC/E,ECxJJ,IAAIC,EAAcC,OAAOC,WAAW,gCAOpC,SAASC,EAAUC,GACjBzD,SAAS0D,gBAAgBC,QAAQC,MAAQP,EAAYQ,QACjD,OACA,OACN,CAQA,SAASC,EAASC,GACH,UAATA,GAA6B,SAATA,GAA4B,SAATA,IACzCC,QAAQC,MAAM,2BAA2BF,yBACzCA,EAAO,QAIT,IAAIG,EAAcb,EAAYQ,QAAU,OAAS,QACjD7D,SAAS0D,gBAAgBC,QAAQI,KAAOA,EACxC,IAAIH,EAAgB,QAARG,EAAiBG,EAAcH,EAC3C/D,SAAS0D,gBAAgBC,QAAQC,MAAQA,EAGzC5D,SAASmE,iBAAiB,kBAAkBC,SAASC,IACrC,SAAVT,EACFS,EAAGC,UAAUC,IAAI,sBAEjBF,EAAGC,UAAUE,OAAO,qBACtB,IAIFC,aAAaC,QAAQ,OAAQX,GAC7BU,aAAaC,QAAQ,QAASd,GAC9BI,QAAQW,IAAI,qBAAqBZ,oBAAuBH,YAGxDP,EAAYuB,SAAmB,QAARb,EAAiBP,EAAY,EACtD,CAMA,SAASqB,IACP,MAAMC,EAAc9E,SAAS0D,gBAAgBC,QAAQmB,aAAe,OAC9DC,EAAcN,aAAaO,QAAQ,SAAWF,EAEpD,IAAiBG,EACXC,EAYNpB,IAZMoB,GADWD,EASF5B,EAAYQ,QACvB,CAAC,OAAQ,QAAS,QAClB,CAAC,OAAQ,OAAQ,UAVIV,QAWS4B,GAXU,KACrBE,EAAInC,SACvBoC,EAAe,GAEVD,EAAIC,IASf,CAsGA,IAAIC,EAAkB,KACpB,IAAIC,EAAQpF,SAASmE,iBAAiB,kBACtC,OAAKiB,EAAMtC,QAKW,GAAhBsC,EAAMtC,OAEDsC,EAAM,GAGNpF,SAASqF,cACd,8CAGQA,cAAc,cAZ1B,CAaF,EAQEC,EAAoB,KAEtB,MAAMC,EAAQJ,IAIRK,EAAexF,SAASyF,eAAe,qBAEzCF,IADgBC,EAAaH,cAAc,SAEzCG,EAAaE,KACfF,EAAaG,QAIbH,EAAaI,YAKX5F,SAAS6F,gBAAkBN,EAC7BA,EAAMO,QAENP,EAAMQ,QACNR,EAAMS,SACNT,EAAMU,eAAe,CAAEC,MAAO,WAElC,EA2CEC,EACoC,IAAtCC,UAAUC,SAASlD,QAAQ,QAAuC,WAAvBiD,UAAUC,SAevD,MAAMC,EAA6B,EACjCC,cAAeC,EACfC,UACAC,cAEA,IAAKF,EAAOd,KACV,OAMF,MAAM,KAAEiB,EAAI,MAAEC,EAAK,IAAEC,EAAG,OAAEC,GAAWN,EAAOO,yBAI1CN,EAAUE,GAAQC,EAAQH,GAAWC,EAAUG,GAAOC,EAASJ,IAG/DF,EAAOb,OACT,EAMF,IA2BIqB,EAAoB,IACe,WAAjCC,sBAAsBC,QACjBD,sBAAsBE,SAASC,SAAS,SAC3C,GAAGH,sBAAsBE,SAASE,UAAU,EAAGJ,sBAAsBE,SAASrE,OAAS,KACvF,GAAGmE,sBAAsBE,YAExB,GAAGF,sBAAsBE,gBAUlCG,eAAeC,EAA0BC,GACxBxH,SAASqF,cAAc,8BAC/Bb,SACP,MAAMpE,EAAU6G,sBAAsBQ,QAChCC,EAAM,IAAIC,KACVC,EAAcC,KAAKC,MACvBrD,aAAaO,QAAQ,oBAAsB,MAE7ChB,QAAQ+D,MACN,kDAAkD3H,cAAoBsH,MAExEE,EAAYxH,GAAWsH,EACvBjD,aAAaC,QAAQ,kBAAmBmD,KAAKG,UAAUJ,GACzD,CAQAN,eAAeW,EAA2BT,GAExCA,EAAMU,iBACN,MAAMC,EAAkBnB,IACxB,IAAIoB,EAASZ,EAAMjB,cAAc8B,aAAa,QAC1CC,EAAoBF,EAAOG,QAAQJ,EAAiB,IACxD,WACmBK,MAAMJ,EAAQ,CAAEK,OAAQ,UAChCC,GACPC,SAASC,KAAOR,EAEhBO,SAASC,KAAON,CAEpB,CAAE,MAAOO,GAEPF,SAASC,KAAON,CAClB,CACF,CA2NAhB,eAAewB,IAGb,IAAIC,EAAsB/I,SAASmE,iBACjC,6BAEF,MAAM6E,EAAkBD,EAAoBjG,OAAS,EAC/CmG,EAAkBhC,sBAAsBiC,eAC5C,2BAEIC,EAAqBlC,sBAAsBmC,4BAEjD,GAAIH,IAAoBD,GAAmBG,GAAqB,CAC9D,MAAME,QAjOV/B,eAAwCgC,GACtC,MAAMC,EAAcvC,IAEpB,IACE,IAAIwC,EAAS,IAAIC,IAAIH,EACvB,CAAE,MAAOT,GACP,KAAIA,aAAepH,WAsBjB,MAAMoH,EAtBsB,CAEI,SAA5BvF,OAAOqF,SAASe,UAIlB1F,QAAQ2F,KACN,gOAKJ,MAAMC,EAAStG,OAAOqF,SAASC,KAAKzF,QAAQoG,GAGtCM,GACO,GAAXD,EACItG,OAAOqF,SAASC,KAChBtF,OAAOqF,SAASC,KAAKvB,UAAU,EAAGuC,GACxCJ,EAAS,IAAIC,IAAIH,EAAKO,EACxB,CAIF,CAEA,MAAMC,QAAiBtB,MAAMgB,GAE7B,aADmBM,EAASC,MAE9B,CA8LuBC,CACjB/C,sBAAsBgD,yBAIpBZ,IAhMR,SAAiCA,EAAMN,GACrC,MAAMZ,EAAkBnB,IACxB+B,EAAoB3E,SAAS8F,IAE3BA,EAAIvG,QAA2B,kBAAI,GACnCuG,EAAIvG,QAAuB,cAAI,EAAE,IAkBnC,MAAMwG,GAXNd,EAAOA,EAAKe,KAAKC,IAEfA,EAAM3I,MACJ2I,EAAMjK,SAAW6G,sBAAsBqD,6BACzCD,EAAME,UAAYF,EAAME,YAAa,EAE/B,SAAUF,IACdA,EAAMG,KAAOH,EAAMjK,SAEdiK,MAGND,KAAKC,GAAUA,EAAME,WAAaF,EAAM3I,QACxC+I,KAAKC,SACR,IAAIC,GAAa,EAEjBtB,EAAKjF,SAASiG,IAEZ,MAAMO,EAAS5K,SAAS6K,cAAc,KACtCD,EAAOE,aACL,QACA,6DAEFF,EAAOE,aAAa,OAAQ,GAAGT,EAAMf,MAAMnB,KAC3CyC,EAAOE,aAAa,OAAQ,UAC5B,MAAMC,EAAO/K,SAAS6K,cAAc,QACpCE,EAAKC,YAAc,GAAGX,EAAMG,OAC5BI,EAAOK,YAAYF,GAGnBH,EAAOjH,QAAqB,YAAI0G,EAAMG,KACtCI,EAAOjH,QAAiB,QAAI0G,EAAMjK,QAIlC,IAAI8K,EAAwBf,GAA6BE,EAAME,UAC3DY,GACDhB,IAA8BQ,GAAcN,EAAM3I,OACjDwJ,GAAyBC,KAC3BP,EAAOtG,UAAUC,IAAI,UACrBwE,EAAoB3E,SAAS8F,IAC3BA,EAAIkB,UAAYf,EAAMG,KACtBN,EAAIvG,QAA2B,kBAAI0G,EAAMG,KACzCN,EAAIvG,QAAuB,cAAI0G,EAAMjK,OAAO,IAE9CuK,GAAa,GAIf3K,SAASmE,iBAAiB,2BAA2BC,SAASiH,IAG5D,IAAIC,EAAOV,EAAOW,WAAU,GAC5BD,EAAKE,QAAUvD,EAIfoD,EAAKI,OAAOH,EAAK,GACjB,GAEN,CAyHMI,CAAwBrC,EAAMN,GAC1BI,GA5GV,SAAkCE,GAChC,IAAIjJ,EAAU6G,sBAAsBQ,QAEhCkE,EAAmBtC,EAAKuC,QAAQvB,GAAUA,EAAME,YACpD,GAAgC,IAA5BoB,EAAiB7I,OAAc,CACjC,MAAM+I,EAAqC,GAA3BF,EAAiB7I,OAAc,KAAO,WAItD,YAHAkB,QAAQW,IACN,SAASkH,kEAGb,CACA,MAAMC,EAAmBH,EAAiB,GAAGvL,QACvC2L,EAAeJ,EAAiB,GAAGrC,IAEnC0C,EAAwB7L,EAASC,IAAYD,EAAS2L,GAC5D,GAAIE,GAAyBzL,EAAQH,EAAS0L,EAAkB,KAI9D,YAHA9H,QAAQW,IACN,qFAKJ,MAAMsH,EAAmBpE,KAAKC,MAC5BrD,aAAaO,QAAQ,oBAAsB,MAC3C5E,GACF,GAAwB,MAApB6L,EAA0B,CAC5B,MAAMC,EAAe,IAAIvE,KAAKsE,GAGxBE,GAFM,IAAIxE,KAEWuE,GADG,MAG9B,GAAIC,EADoB,GAOtB,YALAnI,QAAQ2F,KACN,2DAA2D/G,KAAKwJ,MAC9DD,gBAKR,CAGA,MAAME,EAASrM,SAASqF,cAAc,8BAChCiH,EAAStM,SAAS6K,cAAc,OAChC0B,EAAQvM,SAAS6K,cAAc,OAC/B2B,EAAOxM,SAAS6K,cAAc,UAC9B4B,EAASzM,SAAS6K,cAAc,KAChC6B,EAAY1M,SAAS6K,cAAc,KAGzCyB,EAAOhI,UAAY,mDACnBiI,EAAMjI,UAAY,kBAClBmI,EAAOnI,UACL,4FACFmI,EAAO7D,KAAO,GAAGmD,IAAe/E,MAChCyF,EAAOrB,UAAY,2BACnBqB,EAAOjB,QAAUvD,EACjByE,EAAUpI,UAAY,2BACtB,MAAMqI,EAAU3M,SAAS6K,cAAc,KACvC6B,EAAUjB,OAAOkB,GACjBA,EAAQrI,UAAY,oBACpBoI,EAAUlB,QAAUjE,EAEpBgF,EAAMnB,UAAY,6BAClB,MAAMwB,EACJxM,EAAQoB,SAAS,QACjBpB,EAAQoB,SAAS,OACjBpB,EAAQoB,SAAS,OACbqL,EACJb,GAAyBzL,EAAQH,EAAS0L,EAAkB,KAC1Dc,GAASC,EACXL,EAAKpB,UAAY,kCACRY,GAAyBzL,EAAQH,EAAS0L,EAAkB,KACrEU,EAAKpB,UAAY,mBAAmBhL,KAIpCoM,EAAKpB,UAHKhL,EAGO,WAAWA,IAFX,qBAInBiM,EAAOpB,YAAYqB,GACnBD,EAAOZ,OAAOiB,GACdJ,EAAOrB,YAAYsB,GACnBA,EAAMtB,YAAYuB,GAClBD,EAAMtB,YAAYjL,SAAS8M,eAAe,MAC1CP,EAAMtB,YAAYwB,GAClBJ,EAAO/H,UAAUE,OAAO,SAC1B,CAuBQuI,CAAyB1D,GAG/B,CACF,CAkFA,SAAS2D,IACP,MAAMC,EAAiB,KACrBjN,SACGmE,iBACC,CAEE,MAEA,2BAEA,yBAEA,yBAEA,mCACAf,KAAK,OAERgB,SAASC,IACRA,EAAG6I,SACD7I,EAAG8I,YAAc9I,EAAG+I,aAAe/I,EAAGgJ,aAAehJ,EAAGiJ,aACpD,GACC,CAAC,GACR,EAEAC,EAqBR,SAAkBxN,EAAUyN,GAC1B,IAAIC,EAAY,KAChB,MAAO,IAAIC,KACTC,aAAaF,GACbA,EAAYG,YAAW,KACrB7N,KAAY2N,EAAK,GA1BoC,IA2B/C,CAEZ,CA7BkCG,CAASZ,GAGzC3J,OAAOpD,iBAAiB,SAAUqN,GAOb,IAAIO,iBAAiBP,GAG7BQ,QAAQ/N,SAASyF,eAAe,gBAAiB,CAC5DuI,SAAS,EACTC,WAAW,IAIbhB,GACF,CAcA3F,eAAe4G,IACb,MAAM7B,EAASrM,SAASqF,cAAc,4BAChC,mBAAE8I,GAAuB9B,EAASA,EAAO1I,QAAU,KAEzD,GAAKwK,EAIL,IACE,MAAMrE,QAAiBtB,MAAM2F,GAC7B,IAAKrE,EAASpB,GACZ,MAAM,IAAI/G,MACR,uCAAuCmI,EAASsE,UAAUtE,EAASuE,cAGvE,MAAMhF,QAAaS,EAASwE,OAC5B,GAAoB,IAAhBjF,EAAKvG,OAEP,YADAkB,QAAQW,IAAI,iCAAiCwJ,KAG/C9B,EAAOkC,UAAY,gDAAgDlF,UACnEgD,EAAO/H,UAAUE,OAAO,SAC1B,CAAE,MAAOgK,GACPxK,QAAQW,IAAI,0CAA0CwJ,KACtDnK,QAAQC,MAAMuK,EAChB,CACF,CA2CA1O,GAtCAwH,uBAEQmH,QAAQC,WAAW,CAAC5F,IAAuBoF,MAKjD,MAAMS,EAAW3O,SAASqF,cAAc,8BACxC,IAAKsJ,EACH,OAIFA,EAASrK,UAAUE,OAAO,UAG1B,MAAMoK,EAASC,MAAMC,KAAKH,EAASI,UAAUC,QAC3C,CAACJ,EAAQvK,IAAOuK,EAASvK,EAAG4K,cAC5B,GAKFN,EAASO,MAAMC,YAAY,SAAU,GAAGP,OAIxChB,YAAW,KACTe,EAASO,MAAMC,YAAY,SAAU,OAAO,GAC3C,IACL,IAUArP,GAjwBA,WAGEgE,EAAS9D,SAAS0D,gBAAgBC,QAAQI,MAG1C/D,SAASmE,iBAAiB,wBAAwBC,SAASC,IACzDA,EAAGnE,iBAAiB,QAAS2E,EAAU,GAE3C,IAyvBA/E,GAptBA,WAEE,IAAKE,SAASqF,cAAc,gBAC1B,OAGF,IAAI+J,EAAUpP,SAASqF,cAAc,kBAIrC,IAAIgK,EAAkBnN,SACpBoN,eAAetK,QAAQ,sBACvB,IAGF,GAAK7C,MAAMkN,GAIJ,CAEL,IACIE,EADavP,SAASqF,cAAc,gBACVlB,iBAAiB,WAC/C,GAAIoL,EAAazM,OAAS,EAAG,CAE3B,IAAI0M,EAAgBD,EAAaA,EAAazM,OAAS,GACnD2M,EACFD,EAAczI,wBAAwB2I,EACtCN,EAAQrI,wBAAwB2I,EAElC,GAAIF,EAAczI,wBAAwB2I,EAAyB,GAArBpM,OAAOqM,YAAmB,CACtE,IAAIC,EAAS,IACbR,EAAQS,UAAYJ,EAASL,EAAQ9B,aAAesC,EACpD5L,QAAQW,IAAI,oDACd,CACF,CACF,MAnBEyK,EAAQS,UAAYR,EACpBrL,QAAQW,IAAI,4DAqBdrB,OAAOpD,iBAAiB,gBAAgB,KACtCoP,eAAe5K,QAAQ,qBAAsB0K,EAAQS,UAAU,GAEnE,IA2qBA/P,GA3uBA,WACEwD,OAAOpD,iBAAiB,yBAAyB,WAC9BF,SAASmE,iBAAiB,iBAElCC,SAAS0L,IAChBA,EAAQC,cAAczL,UAAUE,OAAO,SAAS,IAG3BxE,SAASmE,iBAAiB,wBAClCC,SAAS0L,IACtBA,EAAQC,cAAczL,UAAUC,IAAI,SAAS,GAEjD,GACF,IA+tBAzE,GA3hByB,KAnCK,MAC5B,IAAIkQ,EAAYhQ,SAASmE,iBAAiB,gCACtCgC,GACF6J,EAAU5L,SACP6L,GAAOA,EAAE5K,cAAc,8BAA8B+F,UAAY,KAEtE,EA8BA8E,GA/EA5M,OAAOpD,iBACL,WACCsH,IACC,IAAIjC,EAAQJ,IAITqC,EAAM2I,UACN3I,EAAM4I,SAENjK,GACGqB,EAAM6I,SAAY7I,EAAM8I,QACvB9I,EAAM6I,UAAW7I,EAAM8I,WAE5B,OAAOjQ,KAAKmH,EAAM+I,KAMXvQ,SAAS6F,gBAAkBN,GAAS,UAAUlF,KAAKmH,EAAM+I,MAChEjL,KALAkC,EAAMU,iBACN5C,IAKF,IAEF,GA2DFtF,SAASmE,iBAAiB,0BAA0BC,SAAS8F,IAC3DA,EAAIsB,QAAUlG,CAAiB,IAIZtF,SAASyF,eAAe,qBAEhCvF,iBAAiB,QAASoG,EAA2B,IAghBpExG,GAlNA,WAIE,MAAM0Q,EAAiBxQ,SAASyF,eAAe,uBACzCgL,EAAmBzQ,SAASyF,eAAe,yBAI3CiL,EAAgB1Q,SAASyF,eAAe,6BACxCkL,EAAkB3Q,SAASyF,eAC/B,+BAKImL,EAAgB5Q,SAASqF,cAAc,mBACvCwL,EAAkB7Q,SAASqF,cAAc,qBAGzCyL,EAA6B,CAAChC,EAAMiC,KACxClC,MAAMC,KAAKA,EAAKkC,YAAY5M,SAASkH,GAASyF,EAAG9F,YAAYK,KAC7DuD,MAAMC,KAAKA,EAAKxK,WAAWF,SAAS6M,IAClCnC,EAAKxK,UAAUE,OAAOyM,GACtBF,EAAGzM,UAAUC,IAAI0M,EAAI,GACrB,EAIJ,CACE,CAACL,EAAeF,EAAeF,GAC/B,CAACK,EAAiBF,EAAiBF,IACnCrM,SAAQ,EAAE8M,EAAc1K,EAAQ4I,MAC3B8B,GAAiB1K,GAAW4I,IAMjC8B,EAAahR,iBAAiB,SAAUsH,IACtCA,EAAMU,iBACNV,EAAM2J,kBAINL,EAA2B1B,EAAS5I,GAEpCA,EAAOZ,WAAW,IAIpBY,EAAOtG,iBAAiB,QAASoG,GAKjCE,EAAOtG,iBAAiB,WAAYsH,IAChB,WAAdA,EAAM+I,MACR/I,EAAMU,iBACNV,EAAM2J,kBACN3K,EAAOb,QACT,IAKFa,EAAOtG,iBAAiB,SAAS,KAC/B4Q,EAA2BtK,EAAQ4I,EAAQ,IAC3C,GAEN,IAgJ4B,aAAxBpP,SAASC,WACX+M,IAEA1J,OAAOpD,iBAAiB,OAAQ8M,E","sources":["webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/scripts/mixin.js","webpack://pydata_sphinx_theme/./node_modules/compare-versions/lib/esm/index.js","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/scripts/pydata-sphinx-theme.js"],"sourcesContent":["/* define several functions to replace jQuery methods\n * inspired by https://tobiasahlin.com/blog/move-from-jquery-to-vanilla-javascript/\n */\n\n/**\n * Execute a method if DOM has finished loading\n *\n * @param {function} callback the method to execute\n */\nexport function documentReady(callback) {\n  if (document.readyState != \"loading\") callback();\n  else document.addEventListener(\"DOMContentLoaded\", callback);\n}\n","/**\n * Compare [semver](https://semver.org/) version strings to find greater, equal or lesser.\n * This library supports the full semver specification, including comparing versions with different number of digits like `1.0.0`, `1.0`, `1`, and pre-release versions like `1.0.0-alpha`.\n * @param v1 - First version to compare\n * @param v2 - Second version to compare\n * @returns Numeric value compatible with the [Array.sort(fn) interface](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Array/sort#Parameters).\n */\nexport const compareVersions = (v1, v2) => {\n    // validate input and split into segments\n    const n1 = validateAndParse(v1);\n    const n2 = validateAndParse(v2);\n    // pop off the patch\n    const p1 = n1.pop();\n    const p2 = n2.pop();\n    // validate numbers\n    const r = compareSegments(n1, n2);\n    if (r !== 0)\n        return r;\n    // validate pre-release\n    if (p1 && p2) {\n        return compareSegments(p1.split('.'), p2.split('.'));\n    }\n    else if (p1 || p2) {\n        return p1 ? -1 : 1;\n    }\n    return 0;\n};\n/**\n * Validate [semver](https://semver.org/) version strings.\n *\n * @param version Version number to validate\n * @returns `true` if the version number is a valid semver version number, `false` otherwise.\n *\n * @example\n * ```\n * validate('1.0.0-rc.1'); // return true\n * validate('1.0-rc.1'); // return false\n * validate('foo'); // return false\n * ```\n */\nexport const validate = (version) => typeof version === 'string' && /^[v\\d]/.test(version) && semver.test(version);\n/**\n * Compare [semver](https://semver.org/) version strings using the specified operator.\n *\n * @param v1 First version to compare\n * @param v2 Second version to compare\n * @param operator Allowed arithmetic operator to use\n * @returns `true` if the comparison between the firstVersion and the secondVersion satisfies the operator, `false` otherwise.\n *\n * @example\n * ```\n * compare('10.1.8', '10.0.4', '>'); // return true\n * compare('10.0.1', '10.0.1', '='); // return true\n * compare('10.1.1', '10.2.2', '<'); // return true\n * compare('10.1.1', '10.2.2', '<='); // return true\n * compare('10.1.1', '10.2.2', '>='); // return false\n * ```\n */\nexport const compare = (v1, v2, operator) => {\n    // validate input operator\n    assertValidOperator(operator);\n    // since result of compareVersions can only be -1 or 0 or 1\n    // a simple map can be used to replace switch\n    const res = compareVersions(v1, v2);\n    return operatorResMap[operator].includes(res);\n};\n/**\n * Match [npm semver](https://docs.npmjs.com/cli/v6/using-npm/semver) version range.\n *\n * @param version Version number to match\n * @param range Range pattern for version\n * @returns `true` if the version number is within the range, `false` otherwise.\n *\n * @example\n * ```\n * satisfies('1.1.0', '^1.0.0'); // return true\n * satisfies('1.1.0', '~1.0.0'); // return false\n * ```\n */\nexport const satisfies = (version, range) => {\n    // if no range operator then \"=\"\n    const m = range.match(/^([<>=~^]+)/);\n    const op = m ? m[1] : '=';\n    // if gt/lt/eq then operator compare\n    if (op !== '^' && op !== '~')\n        return compare(version, range, op);\n    // else range of either \"~\" or \"^\" is assumed\n    const [v1, v2, v3, , vp] = validateAndParse(version);\n    const [r1, r2, r3, , rp] = validateAndParse(range);\n    const v = [v1, v2, v3];\n    const r = [r1, r2 !== null && r2 !== void 0 ? r2 : 'x', r3 !== null && r3 !== void 0 ? r3 : 'x'];\n    // validate pre-release\n    if (rp) {\n        if (!vp)\n            return false;\n        if (compareSegments(v, r) !== 0)\n            return false;\n        if (compareSegments(vp.split('.'), rp.split('.')) === -1)\n            return false;\n    }\n    // first non-zero number\n    const nonZero = r.findIndex((v) => v !== '0') + 1;\n    // pointer to where segments can be >=\n    const i = op === '~' ? 2 : nonZero > 1 ? nonZero : 1;\n    // before pointer must be equal\n    if (compareSegments(v.slice(0, i), r.slice(0, i)) !== 0)\n        return false;\n    // after pointer must be >=\n    if (compareSegments(v.slice(i), r.slice(i)) === -1)\n        return false;\n    return true;\n};\nconst semver = /^[v^~<>=]*?(\\d+)(?:\\.([x*]|\\d+)(?:\\.([x*]|\\d+)(?:\\.([x*]|\\d+))?(?:-([\\da-z\\-]+(?:\\.[\\da-z\\-]+)*))?(?:\\+[\\da-z\\-]+(?:\\.[\\da-z\\-]+)*)?)?)?$/i;\nconst validateAndParse = (version) => {\n    if (typeof version !== 'string') {\n        throw new TypeError('Invalid argument expected string');\n    }\n    const match = version.match(semver);\n    if (!match) {\n        throw new Error(`Invalid argument not valid semver ('${version}' received)`);\n    }\n    match.shift();\n    return match;\n};\nconst isWildcard = (s) => s === '*' || s === 'x' || s === 'X';\nconst tryParse = (v) => {\n    const n = parseInt(v, 10);\n    return isNaN(n) ? v : n;\n};\nconst forceType = (a, b) => typeof a !== typeof b ? [String(a), String(b)] : [a, b];\nconst compareStrings = (a, b) => {\n    if (isWildcard(a) || isWildcard(b))\n        return 0;\n    const [ap, bp] = forceType(tryParse(a), tryParse(b));\n    if (ap > bp)\n        return 1;\n    if (ap < bp)\n        return -1;\n    return 0;\n};\nconst compareSegments = (a, b) => {\n    for (let i = 0; i < Math.max(a.length, b.length); i++) {\n        const r = compareStrings(a[i] || '0', b[i] || '0');\n        if (r !== 0)\n            return r;\n    }\n    return 0;\n};\nconst operatorResMap = {\n    '>': [1],\n    '>=': [0, 1],\n    '=': [0],\n    '<=': [-1, 0],\n    '<': [-1],\n};\nconst allowedOperators = Object.keys(operatorResMap);\nconst assertValidOperator = (op) => {\n    if (typeof op !== 'string') {\n        throw new TypeError(`Invalid operator type, expected string but got ${typeof op}`);\n    }\n    if (allowedOperators.indexOf(op) === -1) {\n        throw new Error(`Invalid operator, expected one of ${allowedOperators.join('|')}`);\n    }\n};\n//# sourceMappingURL=index.js.map","// Define the custom behavior of the page\nimport { documentReady } from \"./mixin\";\nimport { compare, validate } from \"compare-versions\";\n\nimport \"../styles/pydata-sphinx-theme.scss\";\n\n/*******************************************************************************\n * Theme interaction\n */\n\nvar prefersDark = window.matchMedia(\"(prefers-color-scheme: dark)\");\n\n/**\n * set the the body theme to the one specified by the user browser\n *\n * @param {event} e\n */\nfunction autoTheme(e) {\n  document.documentElement.dataset.theme = prefersDark.matches\n    ? \"dark\"\n    : \"light\";\n}\n\n/**\n * Set the theme using the specified mode.\n * It can be one of [\"auto\", \"dark\", \"light\"]\n *\n * @param {str} mode\n */\nfunction setTheme(mode) {\n  if (mode !== \"light\" && mode !== \"dark\" && mode !== \"auto\") {\n    console.error(`Got invalid theme mode: ${mode}. Resetting to auto.`);\n    mode = \"auto\";\n  }\n\n  // get the theme\n  var colorScheme = prefersDark.matches ? \"dark\" : \"light\";\n  document.documentElement.dataset.mode = mode;\n  var theme = mode == \"auto\" ? colorScheme : mode;\n  document.documentElement.dataset.theme = theme;\n  // TODO: remove this line after Bootstrap upgrade\n  // v5.3 has a colors mode: https://getbootstrap.com/docs/5.3/customize/color-modes/\n  document.querySelectorAll(\".dropdown-menu\").forEach((el) => {\n    if (theme === \"dark\") {\n      el.classList.add(\"dropdown-menu-dark\");\n    } else {\n      el.classList.remove(\"dropdown-menu-dark\");\n    }\n  });\n\n  // save mode and theme\n  localStorage.setItem(\"mode\", mode);\n  localStorage.setItem(\"theme\", theme);\n  console.log(`[PST]: Changed to ${mode} mode using the ${theme} theme.`);\n\n  // add a listener if set on auto\n  prefersDark.onchange = mode == \"auto\" ? autoTheme : \"\";\n}\n\n/**\n * Change the theme option order so that clicking on the btn is always a change\n * from \"auto\"\n */\nfunction cycleMode() {\n  const defaultMode = document.documentElement.dataset.defaultMode || \"auto\";\n  const currentMode = localStorage.getItem(\"mode\") || defaultMode;\n\n  var loopArray = (arr, current) => {\n    var nextPosition = arr.indexOf(current) + 1;\n    if (nextPosition === arr.length) {\n      nextPosition = 0;\n    }\n    return arr[nextPosition];\n  };\n\n  // make sure the next theme after auto is always a change\n  var modeList = prefersDark.matches\n    ? [\"auto\", \"light\", \"dark\"]\n    : [\"auto\", \"dark\", \"light\"];\n  var newMode = loopArray(modeList, currentMode);\n  setTheme(newMode);\n}\n\n/**\n * add the theme listener on the btns of the navbar\n */\nfunction addModeListener() {\n  // the theme was set a first time using the initial mini-script\n  // running setMode will ensure the use of the dark mode if auto is selected\n  setTheme(document.documentElement.dataset.mode);\n\n  // Attach event handlers for toggling themes colors\n  document.querySelectorAll(\".theme-switch-button\").forEach((el) => {\n    el.addEventListener(\"click\", cycleMode);\n  });\n}\n\n/*******************************************************************************\n * TOC interactivity\n */\n\n/**\n * TOC sidebar - add \"active\" class to parent list\n *\n * Bootstrap's scrollspy adds the active class to the <a> link,\n * but for the automatic collapsing we need this on the parent list item.\n *\n * The event is triggered on \"window\" (and not the nav item as documented),\n * see https://github.com/twbs/bootstrap/issues/20086\n */\nfunction addTOCInteractivity() {\n  window.addEventListener(\"activate.bs.scrollspy\", function () {\n    const navLinks = document.querySelectorAll(\".bd-toc-nav a\");\n\n    navLinks.forEach((navLink) => {\n      navLink.parentElement.classList.remove(\"active\");\n    });\n\n    const activeNavLinks = document.querySelectorAll(\".bd-toc-nav a.active\");\n    activeNavLinks.forEach((navLink) => {\n      navLink.parentElement.classList.add(\"active\");\n    });\n  });\n}\n\n/*******************************************************************************\n * Scroll\n */\n\n/**\n * Navigation sidebar scrolling to active page\n */\nfunction scrollToActive() {\n  // If the docs nav doesn't exist, do nothing (e.g., on search page)\n  if (!document.querySelector(\".bd-docs-nav\")) {\n    return;\n  }\n\n  var sidebar = document.querySelector(\"div.bd-sidebar\");\n\n  // Remember the sidebar scroll position between page loads\n  // Inspired on source of revealjs.com\n  let storedScrollTop = parseInt(\n    sessionStorage.getItem(\"sidebar-scroll-top\"),\n    10,\n  );\n\n  if (!isNaN(storedScrollTop)) {\n    // If we've got a saved scroll position, just use that\n    sidebar.scrollTop = storedScrollTop;\n    console.log(\"[PST]: Scrolled sidebar using stored browser position...\");\n  } else {\n    // Otherwise, calculate a position to scroll to based on the lowest `active` link\n    var sidebarNav = document.querySelector(\".bd-docs-nav\");\n    var active_pages = sidebarNav.querySelectorAll(\".active\");\n    if (active_pages.length > 0) {\n      // Use the last active page as the offset since it's the page we're on\n      var latest_active = active_pages[active_pages.length - 1];\n      var offset =\n        latest_active.getBoundingClientRect().y -\n        sidebar.getBoundingClientRect().y;\n      // Only scroll the navbar if the active link is lower than 50% of the page\n      if (latest_active.getBoundingClientRect().y > window.innerHeight * 0.5) {\n        let buffer = 0.25; // Buffer so we have some space above the scrolled item\n        sidebar.scrollTop = offset - sidebar.clientHeight * buffer;\n        console.log(\"[PST]: Scrolled sidebar using last active link...\");\n      }\n    }\n  }\n\n  // Store the sidebar scroll position\n  window.addEventListener(\"beforeunload\", () => {\n    sessionStorage.setItem(\"sidebar-scroll-top\", sidebar.scrollTop);\n  });\n}\n\n/*******************************************************************************\n * Search\n */\n\n/**\n * Find any search forms on the page and return their input element\n */\nvar findSearchInput = () => {\n  let forms = document.querySelectorAll(\"form.bd-search\");\n  if (!forms.length) {\n    // no search form found\n    return;\n  } else {\n    var form;\n    if (forms.length == 1) {\n      // there is exactly one search form (persistent or hidden)\n      form = forms[0];\n    } else {\n      // must be at least one persistent form, use the first persistent one\n      form = document.querySelector(\n        \":not(#pst-search-dialog) > form.bd-search\",\n      );\n    }\n    return form.querySelector(\"input\");\n  }\n};\n\n/**\n * Activate the search field on the page.\n * - If there is a search field already visible it will be activated.\n * - If not, then a search field will pop up.\n */\nvar toggleSearchField = () => {\n  // Find the search input to highlight\n  const input = findSearchInput();\n\n  // if the input field is the hidden one (the one associated with the\n  // search button) then toggle the button state (to show/hide the field)\n  const searchDialog = document.getElementById(\"pst-search-dialog\");\n  const hiddenInput = searchDialog.querySelector(\"input\");\n  if (input === hiddenInput) {\n    if (searchDialog.open) {\n      searchDialog.close();\n    } else {\n      // Note: browsers should focus the input field inside the modal dialog\n      // automatically when it is opened.\n      searchDialog.showModal();\n    }\n  } else {\n    // if the input field is not the hidden one, then toggle its focus state\n\n    if (document.activeElement === input) {\n      input.blur();\n    } else {\n      input.focus();\n      input.select();\n      input.scrollIntoView({ block: \"center\" });\n    }\n  }\n};\n\n/**\n * Add an event listener for toggleSearchField() for Ctrl/Cmd + K\n */\nvar addEventListenerForSearchKeyboard = () => {\n  window.addEventListener(\n    \"keydown\",\n    (event) => {\n      let input = findSearchInput();\n      // toggle on Ctrl+k or ⌘+k\n      if (\n        // Ignore if shift or alt are pressed\n        !event.shiftKey &&\n        !event.altKey &&\n        // On Mac use ⌘, all other OS use Ctrl\n        (useCommandKey\n          ? event.metaKey && !event.ctrlKey\n          : !event.metaKey && event.ctrlKey) &&\n        // Case-insensitive so the shortcut still works with caps lock\n        /^k$/i.test(event.key)\n      ) {\n        event.preventDefault();\n        toggleSearchField();\n      }\n      // also allow Escape key to hide (but not show) the dynamic search field\n      else if (document.activeElement === input && /Escape/i.test(event.key)) {\n        toggleSearchField();\n      }\n    },\n    true,\n  );\n};\n\n/**\n * If the user is on a Mac, use command (⌘) instead of control (ctrl) key\n *\n * Note: `navigator.platform` is deprecated; however MDN still recommends using\n * it for the one specific use case of detecting whether a keyboard shortcut\n * should use control or command:\n * https://developer.mozilla.org/en-US/docs/Web/API/Navigator/platform#examples\n */\nvar useCommandKey =\n  navigator.platform.indexOf(\"Mac\") === 0 || navigator.platform === \"iPhone\";\n\n/**\n * Change the search hint to `meta key` if we are a Mac\n */\n\nvar changeSearchShortcutKey = () => {\n  let shortcuts = document.querySelectorAll(\".search-button__kbd-shortcut\");\n  if (useCommandKey) {\n    shortcuts.forEach(\n      (f) => (f.querySelector(\"kbd.kbd-shortcut__modifier\").innerText = \"⌘\"),\n    );\n  }\n};\n\nconst closeDialogOnBackdropClick = ({\n  currentTarget: dialog,\n  clientX,\n  clientY,\n}) => {\n  if (!dialog.open) {\n    return;\n  }\n\n  // Dialog.getBoundingClientRect() does not include ::backdrop. (This is the\n  // trick that allows us to determine if click was inside or outside of the\n  // dialog: click handler includes backdrop, getBoundingClientRect does not.)\n  const { left, right, top, bottom } = dialog.getBoundingClientRect();\n\n  // 0, 0 means top left\n  const clickWasOutsideDialog =\n    clientX < left || right < clientX || clientY < top || bottom < clientY;\n\n  if (clickWasOutsideDialog) {\n    dialog.close();\n  }\n};\n\n/**\n * Activate callbacks for search button popup\n */\nvar setupSearchButtons = () => {\n  changeSearchShortcutKey();\n  addEventListenerForSearchKeyboard();\n\n  // Add the search button trigger event callback\n  document.querySelectorAll(\".search-button__button\").forEach((btn) => {\n    btn.onclick = toggleSearchField;\n  });\n\n  // If user clicks outside the search modal dialog, then close it.\n  const searchDialog = document.getElementById(\"pst-search-dialog\");\n  // Dialog click handler includes clicks on dialog ::backdrop.\n  searchDialog.addEventListener(\"click\", closeDialogOnBackdropClick);\n};\n\n/*******************************************************************************\n * Version Switcher\n * Note that this depends on two variables existing that are defined in\n * and `html-page-context` hook:\n *\n * - DOCUMENTATION_OPTIONS.pagename\n * - DOCUMENTATION_OPTIONS.theme_switcher_url\n */\n\n/**\n * path component of URL\n */\nvar getCurrentUrlPath = () => {\n  if (DOCUMENTATION_OPTIONS.BUILDER == \"dirhtml\") {\n    return DOCUMENTATION_OPTIONS.pagename.endsWith(\"index\")\n      ? `${DOCUMENTATION_OPTIONS.pagename.substring(0, DOCUMENTATION_OPTIONS.pagename.length - \"index\".length)}`\n      : `${DOCUMENTATION_OPTIONS.pagename}/`;\n  }\n  return `${DOCUMENTATION_OPTIONS.pagename}.html`;\n};\n\n/**\n * Allow user to dismiss the warning banner about the docs version being dev / old.\n * We store the dismissal date and version, to give us flexibility about making the\n * dismissal last for longer than one browser session, if we decide to do that.\n *\n * @param {event} event the event that trigger the check\n */\nasync function DismissBannerAndStorePref(event) {\n  const banner = document.querySelector(\"#bd-header-version-warning\");\n  banner.remove();\n  const version = DOCUMENTATION_OPTIONS.VERSION;\n  const now = new Date();\n  const banner_pref = JSON.parse(\n    localStorage.getItem(\"pst_banner_pref\") || \"{}\",\n  );\n  console.debug(\n    `[PST] Dismissing the version warning banner on ${version} starting ${now}.`,\n  );\n  banner_pref[version] = now;\n  localStorage.setItem(\"pst_banner_pref\", JSON.stringify(banner_pref));\n}\n\n/**\n * Check if corresponding page path exists in other version of docs\n * and, if so, go there instead of the homepage of the other docs version\n *\n * @param {event} event the event that trigger the check\n */\nasync function checkPageExistsAndRedirect(event) {\n  // ensure we don't follow the initial link\n  event.preventDefault();\n  const currentFilePath = getCurrentUrlPath();\n  let tryUrl = event.currentTarget.getAttribute(\"href\");\n  let otherDocsHomepage = tryUrl.replace(currentFilePath, \"\");\n  try {\n    let head = await fetch(tryUrl, { method: \"HEAD\" });\n    if (head.ok) {\n      location.href = tryUrl; // the page exists, go there\n    } else {\n      location.href = otherDocsHomepage;\n    }\n  } catch (err) {\n    // something went wrong, probably CORS restriction, fallback to other docs homepage\n    location.href = otherDocsHomepage;\n  }\n}\n\n/**\n * Load and parse the version switcher JSON file from an absolute or relative URL.\n *\n * @param {string} url The URL to load version switcher entries from.\n */\nasync function fetchVersionSwitcherJSON(url) {\n  const currentPath = getCurrentUrlPath();\n  // first check if it's a valid URL\n  try {\n    var result = new URL(url);\n  } catch (err) {\n    if (err instanceof TypeError) {\n      // Assume we got a relative path, and fix accordingly.\n      if (window.location.protocol == \"file:\") {\n        // Here instead of returning `null` we work out what the file path would be\n        // anyway (same code path as for served docs), as a convenience to folks who\n        // routinely disable CORS when they boot up their browser.\n        console.info(\n          \"[PST] looks like you're viewing this site from a local filesystem, so \" +\n            \"the version switcher won't work unless you've disabled CORS. See \" +\n            \"https://pydata-sphinx-theme.readthedocs.io/en/stable/user_guide/version-dropdown.html\",\n        );\n      }\n      const cutoff = window.location.href.indexOf(currentPath);\n      // cutoff == -1 can happen e.g. on the homepage of locally served docs, where you\n      // get something like http://127.0.0.1:8000/ (no trailing `index.html`)\n      const origin =\n        cutoff == -1\n          ? window.location.href\n          : window.location.href.substring(0, cutoff);\n      result = new URL(url, origin);\n    } else {\n      // something unexpected happened\n      throw err;\n    }\n  }\n  // load and return the JSON\n  const response = await fetch(result);\n  const data = await response.json();\n  return data;\n}\n\n// Populate the version switcher from the JSON data\nfunction populateVersionSwitcher(data, versionSwitcherBtns) {\n  const currentFilePath = getCurrentUrlPath();\n  versionSwitcherBtns.forEach((btn) => {\n    // Set empty strings by default so that these attributes exist and can be used in CSS selectors\n    btn.dataset[\"activeVersionName\"] = \"\";\n    btn.dataset[\"activeVersion\"] = \"\";\n  });\n  // in case there are multiple entries with the same version string, this helps us\n  // decide which entry's `name` to put on the button itself. Without this, it would\n  // always be the *last* version-matching entry; now it will be either the\n  // version-matching entry that is also marked as `\"preferred\": true`, or if that\n  // doesn't exist: the *first* version-matching entry.\n  data = data.map((entry) => {\n    // does this entry match the version that we're currently building/viewing?\n    entry.match =\n      entry.version == DOCUMENTATION_OPTIONS.theme_switcher_version_match;\n    entry.preferred = entry.preferred || false;\n    // if no custom name specified (e.g., \"latest\"), use version string\n    if (!(\"name\" in entry)) {\n      entry.name = entry.version;\n    }\n    return entry;\n  });\n  const hasMatchingPreferredEntry = data\n    .map((entry) => entry.preferred && entry.match)\n    .some(Boolean);\n  var foundMatch = false;\n  // create links to the corresponding page in the other docs versions\n  data.forEach((entry) => {\n    // create the node\n    const anchor = document.createElement(\"a\");\n    anchor.setAttribute(\n      \"class\",\n      \"dropdown-item list-group-item list-group-item-action py-1\",\n    );\n    anchor.setAttribute(\"href\", `${entry.url}${currentFilePath}`);\n    anchor.setAttribute(\"role\", \"option\");\n    const span = document.createElement(\"span\");\n    span.textContent = `${entry.name}`;\n    anchor.appendChild(span);\n    // Add dataset values for the version and name in case people want\n    // to apply CSS styling based on this information.\n    anchor.dataset[\"versionName\"] = entry.name;\n    anchor.dataset[\"version\"] = entry.version;\n    // replace dropdown button text with the preferred display name of the\n    // currently-viewed version, rather than using sphinx's {{ version }} variable.\n    // also highlight the dropdown entry for the currently-viewed version's entry\n    let matchesAndIsPreferred = hasMatchingPreferredEntry && entry.preferred;\n    let matchesAndIsFirst =\n      !hasMatchingPreferredEntry && !foundMatch && entry.match;\n    if (matchesAndIsPreferred || matchesAndIsFirst) {\n      anchor.classList.add(\"active\");\n      versionSwitcherBtns.forEach((btn) => {\n        btn.innerText = entry.name;\n        btn.dataset[\"activeVersionName\"] = entry.name;\n        btn.dataset[\"activeVersion\"] = entry.version;\n      });\n      foundMatch = true;\n    }\n    // There may be multiple version-switcher elements, e.g. one\n    // in a slide-over panel displayed on smaller screens.\n    document.querySelectorAll(\".version-switcher__menu\").forEach((menu) => {\n      // we need to clone the node for each menu, but onclick attributes are not\n      // preserved by `.cloneNode()` so we add onclick here after cloning.\n      let node = anchor.cloneNode(true);\n      node.onclick = checkPageExistsAndRedirect;\n      // on click, AJAX calls will check if the linked page exists before\n      // trying to redirect, and if not, will redirect to the homepage\n      // for that version of the docs.\n      menu.append(node);\n    });\n  });\n}\n\n/*******************************************************************************\n * Warning banner when viewing non-stable version of the docs.\n */\n\n/**\n * Show a warning banner when viewing a non-stable version of the docs.\n *\n * adapted 2023-06 from https://mne.tools/versionwarning.js, which was\n * originally adapted 2020-05 from https://scikit-learn.org/versionwarning.js\n *\n * @param {Array} data The version data used to populate the switcher menu.\n */\nfunction showVersionWarningBanner(data) {\n  var version = DOCUMENTATION_OPTIONS.VERSION;\n  // figure out what latest stable version is\n  var preferredEntries = data.filter((entry) => entry.preferred);\n  if (preferredEntries.length !== 1) {\n    const howMany = preferredEntries.length == 0 ? \"No\" : \"Multiple\";\n    console.log(\n      `[PST] ${howMany} versions marked \"preferred\" found in versions JSON, ignoring.`,\n    );\n    return;\n  }\n  const preferredVersion = preferredEntries[0].version;\n  const preferredURL = preferredEntries[0].url;\n  // if already on preferred version, nothing to do\n  const versionsAreComparable = validate(version) && validate(preferredVersion);\n  if (versionsAreComparable && compare(version, preferredVersion, \"=\")) {\n    console.log(\n      \"[PST]: This is the preferred version of the docs, not showing the warning banner.\",\n    );\n    return;\n  }\n  // check if banner has been dismissed recently\n  const dismiss_date_str = JSON.parse(\n    localStorage.getItem(\"pst_banner_pref\") || \"{}\",\n  )[version];\n  if (dismiss_date_str != null) {\n    const dismiss_date = new Date(dismiss_date_str);\n    const now = new Date();\n    const milliseconds_in_a_day = 24 * 60 * 60 * 1000;\n    const days_passed = (now - dismiss_date) / milliseconds_in_a_day;\n    const timeout_in_days = 14;\n    if (days_passed < timeout_in_days) {\n      console.info(\n        `[PST] Suppressing version warning banner; was dismissed ${Math.floor(\n          days_passed,\n        )} day(s) ago`,\n      );\n      return;\n    }\n  }\n\n  // now construct the warning banner\n  const banner = document.querySelector(\"#bd-header-version-warning\");\n  const middle = document.createElement(\"div\");\n  const inner = document.createElement(\"div\");\n  const bold = document.createElement(\"strong\");\n  const button = document.createElement(\"a\");\n  const close_btn = document.createElement(\"a\");\n  // these classes exist since pydata-sphinx-theme v0.10.0\n  // the init class is used for animation\n  middle.classList = \"bd-header-announcement__content  ms-auto me-auto\";\n  inner.classList = \"sidebar-message\";\n  button.classList =\n    \"btn text-wrap font-weight-bold ms-3 my-1 align-baseline pst-button-link-to-stable-version\";\n  button.href = `${preferredURL}${getCurrentUrlPath()}`;\n  button.innerText = \"Switch to stable version\";\n  button.onclick = checkPageExistsAndRedirect;\n  close_btn.classList = \"ms-3 my-1 align-baseline\";\n  const close_x = document.createElement(\"i\");\n  close_btn.append(close_x);\n  close_x.classList = \"fa-solid fa-xmark\";\n  close_btn.onclick = DismissBannerAndStorePref;\n  // add the version-dependent text\n  inner.innerText = \"This is documentation for \";\n  const isDev =\n    version.includes(\"dev\") ||\n    version.includes(\"rc\") ||\n    version.includes(\"pre\");\n  const newerThanPreferred =\n    versionsAreComparable && compare(version, preferredVersion, \">\");\n  if (isDev || newerThanPreferred) {\n    bold.innerText = \"an unstable development version\";\n  } else if (versionsAreComparable && compare(version, preferredVersion, \"<\")) {\n    bold.innerText = `an old version (${version})`;\n  } else if (!version) {\n    bold.innerText = \"an unknown version\"; // e.g., an empty string\n  } else {\n    bold.innerText = `version ${version}`;\n  }\n  banner.appendChild(middle);\n  banner.append(close_btn);\n  middle.appendChild(inner);\n  inner.appendChild(bold);\n  inner.appendChild(document.createTextNode(\".\"));\n  inner.appendChild(button);\n  banner.classList.remove(\"d-none\");\n}\n\nasync function fetchAndUseVersions() {\n  // fetch the JSON version data (only once), then use it to populate the version\n  // switcher and maybe show the version warning bar\n  var versionSwitcherBtns = document.querySelectorAll(\n    \".version-switcher__button\",\n  );\n  const hasSwitcherMenu = versionSwitcherBtns.length > 0;\n  const hasVersionsJSON = DOCUMENTATION_OPTIONS.hasOwnProperty(\n    \"theme_switcher_json_url\",\n  );\n  const wantsWarningBanner = DOCUMENTATION_OPTIONS.show_version_warning_banner;\n\n  if (hasVersionsJSON && (hasSwitcherMenu || wantsWarningBanner)) {\n    const data = await fetchVersionSwitcherJSON(\n      DOCUMENTATION_OPTIONS.theme_switcher_json_url,\n    );\n    // TODO: remove the `if(data)` once the `return null` is fixed within fetchVersionSwitcherJSON.\n    // We don't really want the switcher and warning bar to silently not work.\n    if (data) {\n      populateVersionSwitcher(data, versionSwitcherBtns);\n      if (wantsWarningBanner) {\n        showVersionWarningBanner(data);\n      }\n    }\n  }\n}\n\n/*******************************************************************************\n * Sidebar modals (for mobile / narrow screens)\n */\nfunction setupMobileSidebarKeyboardHandlers() {\n  // These are the left and right sidebars for wider screens. We cut and paste\n  // the content from these widescreen sidebars into the mobile dialogs, when\n  // the user clicks the hamburger icon button\n  const primarySidebar = document.getElementById(\"pst-primary-sidebar\");\n  const secondarySidebar = document.getElementById(\"pst-secondary-sidebar\");\n\n  // These are the corresponding left/right <dialog> elements, which are empty\n  // until the user clicks the hamburger icon\n  const primaryDialog = document.getElementById(\"pst-primary-sidebar-modal\");\n  const secondaryDialog = document.getElementById(\n    \"pst-secondary-sidebar-modal\",\n  );\n\n  // These are the hamburger-style buttons in the header nav bar. They only\n  // appear at narrow screen width.\n  const primaryToggle = document.querySelector(\".primary-toggle\");\n  const secondaryToggle = document.querySelector(\".secondary-toggle\");\n\n  // Cut nodes and classes from `from`, paste into/onto `to`\n  const cutAndPasteNodesAndClasses = (from, to) => {\n    Array.from(from.childNodes).forEach((node) => to.appendChild(node));\n    Array.from(from.classList).forEach((cls) => {\n      from.classList.remove(cls);\n      to.classList.add(cls);\n    });\n  };\n\n  // Hook up the ways to open and close the dialog\n  [\n    [primaryToggle, primaryDialog, primarySidebar],\n    [secondaryToggle, secondaryDialog, secondarySidebar],\n  ].forEach(([toggleButton, dialog, sidebar]) => {\n    if (!toggleButton || !dialog || !sidebar) {\n      return;\n    }\n\n    // Clicking the button can only open the sidebar, not close it.\n    // Clicking the button is also the *only* way to open the sidebar.\n    toggleButton.addEventListener(\"click\", (event) => {\n      event.preventDefault();\n      event.stopPropagation();\n\n      // When we open the dialog, we cut and paste the nodes and classes from\n      // the widescreen sidebar into the dialog\n      cutAndPasteNodesAndClasses(sidebar, dialog);\n\n      dialog.showModal();\n    });\n\n    // Listen for clicks on the backdrop in order to close the dialog\n    dialog.addEventListener(\"click\", closeDialogOnBackdropClick);\n\n    // We have to manually attach the escape key because there's some code in\n    // Sphinx's Sphinx_highlight.js that prevents the default behavior of the\n    // escape key\n    dialog.addEventListener(\"keydown\", (event) => {\n      if (event.key === \"Escape\") {\n        event.preventDefault();\n        event.stopPropagation();\n        dialog.close();\n      }\n    });\n\n    // When the dialog is closed, move the nodes (and classes) back to their\n    // original place\n    dialog.addEventListener(\"close\", () => {\n      cutAndPasteNodesAndClasses(dialog, sidebar);\n    });\n  });\n}\n\n/**\n * When the page loads, or the window resizes, or descendant nodes are added or\n * removed from the main element, check all code blocks and Jupyter notebook\n * outputs, and for each one that has scrollable overflow, set tabIndex = 0.\n */\nfunction addTabStopsToScrollableElements() {\n  const updateTabStops = () => {\n    document\n      .querySelectorAll(\n        [\n          // code blocks\n          \"pre\",\n          // NBSphinx notebook output\n          \".nboutput > .output_area\",\n          // Myst-NB\n          \".cell_output > .output\",\n          // ipywidgets\n          \".jp-RenderedHTMLCommon\",\n          // [rST table nodes](https://www.docutils.org/docs/ref/doctree.html#table)\n          \".pst-scrollable-table-container\",\n        ].join(\", \"),\n      )\n      .forEach((el) => {\n        el.tabIndex =\n          el.scrollWidth > el.clientWidth || el.scrollHeight > el.clientHeight\n            ? 0\n            : -1;\n      });\n  };\n  const debouncedUpdateTabStops = debounce(updateTabStops, 300);\n\n  // On window resize\n  window.addEventListener(\"resize\", debouncedUpdateTabStops);\n\n  // The following MutationObserver is for ipywidgets, which take some time to\n  // finish loading and rendering on the page (so even after the \"load\" event is\n  // fired, they still have not finished rendering). Would be nice to replace\n  // the MutationObserver if there is a way to hook into the ipywidgets code to\n  // know when it is done.\n  const mainObserver = new MutationObserver(debouncedUpdateTabStops);\n\n  // On descendant nodes added/removed from main element\n  mainObserver.observe(document.getElementById(\"main-content\"), {\n    subtree: true,\n    childList: true,\n  });\n\n  // On page load (when this function gets called)\n  updateTabStops();\n}\nfunction debounce(callback, wait) {\n  let timeoutId = null;\n  return (...args) => {\n    clearTimeout(timeoutId);\n    timeoutId = setTimeout(() => {\n      callback(...args);\n    }, wait);\n  };\n}\n\n/*******************************************************************************\n * Announcement banner - fetch and load remote HTML\n */\nasync function setupAnnouncementBanner() {\n  const banner = document.querySelector(\".bd-header-announcement\");\n  const { pstAnnouncementUrl } = banner ? banner.dataset : null;\n\n  if (!pstAnnouncementUrl) {\n    return;\n  }\n\n  try {\n    const response = await fetch(pstAnnouncementUrl);\n    if (!response.ok) {\n      throw new Error(\n        `[PST]: HTTP response status not ok: ${response.status} ${response.statusText}`,\n      );\n    }\n    const data = await response.text();\n    if (data.length === 0) {\n      console.log(`[PST]: Empty announcement at: ${pstAnnouncementUrl}`);\n      return;\n    }\n    banner.innerHTML = `<div class=\"bd-header-announcement__content\">${data}</div>`;\n    banner.classList.remove(\"d-none\");\n  } catch (_error) {\n    console.log(`[PST]: Failed to load announcement at: ${pstAnnouncementUrl}`);\n    console.error(_error);\n  }\n}\n\n/*******************************************************************************\n * Reveal (and animate) the banners (version warning, announcement) together\n */\nasync function fetchRevealBannersTogether() {\n  // Wait until finished fetching and loading banners\n  await Promise.allSettled([fetchAndUseVersions(), setupAnnouncementBanner()]);\n\n  // The revealer element should have CSS rules that set height to 0, overflow\n  // to hidden, and an animation transition on the height (unless the user has\n  // turned off animations)\n  const revealer = document.querySelector(\".pst-async-banner-revealer\");\n  if (!revealer) {\n    return;\n  }\n\n  // Remove the d-none (display-none) class to calculate the children heights.\n  revealer.classList.remove(\"d-none\");\n\n  // Add together the heights of the element's children\n  const height = Array.from(revealer.children).reduce(\n    (height, el) => height + el.offsetHeight,\n    0,\n  );\n\n  // Use the calculated height to give the revealer a non-zero height (if\n  // animations allowed, the height change will animate)\n  revealer.style.setProperty(\"height\", `${height}px`);\n\n  // Wait for a bit more than 300ms (the transition duration), then set height\n  // to auto so the banner can resize if the window is resized.\n  setTimeout(() => {\n    revealer.style.setProperty(\"height\", \"auto\");\n  }, 320);\n}\n\n/*******************************************************************************\n * Call functions after document loading.\n */\n\n// This one first to kick off the network request for the version warning\n// and announcement banner data as early as possible.\ndocumentReady(fetchRevealBannersTogether);\n\ndocumentReady(addModeListener);\ndocumentReady(scrollToActive);\ndocumentReady(addTOCInteractivity);\ndocumentReady(setupSearchButtons);\ndocumentReady(setupMobileSidebarKeyboardHandlers);\n\n// Determining whether an element has scrollable content depends on stylesheets,\n// so we're checking for the \"load\" event rather than \"DOMContentLoaded\"\nif (document.readyState === \"complete\") {\n  addTabStopsToScrollableElements();\n} else {\n  window.addEventListener(\"load\", addTabStopsToScrollableElements);\n}\n"],"names":["documentReady","callback","document","readyState","addEventListener","validate","version","test","semver","compare","v1","v2","operator","assertValidOperator","res","n1","validateAndParse","n2","p1","pop","p2","r","compareSegments","split","compareVersions","operatorResMap","includes","TypeError","match","Error","shift","isWildcard","s","tryParse","v","n","parseInt","isNaN","compareStrings","a","b","ap","bp","String","forceType","i","Math","max","length","allowedOperators","Object","keys","op","indexOf","join","prefersDark","window","matchMedia","autoTheme","e","documentElement","dataset","theme","matches","setTheme","mode","console","error","colorScheme","querySelectorAll","forEach","el","classList","add","remove","localStorage","setItem","log","onchange","cycleMode","defaultMode","currentMode","getItem","arr","nextPosition","findSearchInput","forms","querySelector","toggleSearchField","input","searchDialog","getElementById","open","close","showModal","activeElement","blur","focus","select","scrollIntoView","block","useCommandKey","navigator","platform","closeDialogOnBackdropClick","currentTarget","dialog","clientX","clientY","left","right","top","bottom","getBoundingClientRect","getCurrentUrlPath","DOCUMENTATION_OPTIONS","BUILDER","pagename","endsWith","substring","async","DismissBannerAndStorePref","event","VERSION","now","Date","banner_pref","JSON","parse","debug","stringify","checkPageExistsAndRedirect","preventDefault","currentFilePath","tryUrl","getAttribute","otherDocsHomepage","replace","fetch","method","ok","location","href","err","fetchAndUseVersions","versionSwitcherBtns","hasSwitcherMenu","hasVersionsJSON","hasOwnProperty","wantsWarningBanner","show_version_warning_banner","data","url","currentPath","result","URL","protocol","info","cutoff","origin","response","json","fetchVersionSwitcherJSON","theme_switcher_json_url","btn","hasMatchingPreferredEntry","map","entry","theme_switcher_version_match","preferred","name","some","Boolean","foundMatch","anchor","createElement","setAttribute","span","textContent","appendChild","matchesAndIsPreferred","matchesAndIsFirst","innerText","menu","node","cloneNode","onclick","append","populateVersionSwitcher","preferredEntries","filter","howMany","preferredVersion","preferredURL","versionsAreComparable","dismiss_date_str","dismiss_date","days_passed","floor","banner","middle","inner","bold","button","close_btn","close_x","isDev","newerThanPreferred","createTextNode","showVersionWarningBanner","addTabStopsToScrollableElements","updateTabStops","tabIndex","scrollWidth","clientWidth","scrollHeight","clientHeight","debouncedUpdateTabStops","wait","timeoutId","args","clearTimeout","setTimeout","debounce","MutationObserver","observe","subtree","childList","setupAnnouncementBanner","pstAnnouncementUrl","status","statusText","text","innerHTML","_error","Promise","allSettled","revealer","height","Array","from","children","reduce","offsetHeight","style","setProperty","sidebar","storedScrollTop","sessionStorage","active_pages","latest_active","offset","y","innerHeight","buffer","scrollTop","navLink","parentElement","shortcuts","f","changeSearchShortcutKey","shiftKey","altKey","metaKey","ctrlKey","key","primarySidebar","secondarySidebar","primaryDialog","secondaryDialog","primaryToggle","secondaryToggle","cutAndPasteNodesAndClasses","to","childNodes","cls","toggleButton","stopPropagation"],"sourceRoot":""}
\ No newline at end of file
diff --git a/_static/searchtools.js b/_static/searchtools.js
new file mode 100644
index 000000000..2c774d17a
--- /dev/null
+++ b/_static/searchtools.js
@@ -0,0 +1,632 @@
+/*
+ * Sphinx JavaScript utilities for the full-text search.
+ */
+"use strict";
+
+/**
+ * Simple result scoring code.
+ */
+if (typeof Scorer === "undefined") {
+  var Scorer = {
+    // Implement the following function to further tweak the score for each result
+    // The function takes a result array [docname, title, anchor, descr, score, filename]
+    // and returns the new score.
+    /*
+    score: result => {
+      const [docname, title, anchor, descr, score, filename, kind] = result
+      return score
+    },
+    */
+
+    // query matches the full name of an object
+    objNameMatch: 11,
+    // or matches in the last dotted part of the object name
+    objPartialMatch: 6,
+    // Additive scores depending on the priority of the object
+    objPrio: {
+      0: 15, // used to be importantResults
+      1: 5, // used to be objectResults
+      2: -5, // used to be unimportantResults
+    },
+    //  Used when the priority is not in the mapping.
+    objPrioDefault: 0,
+
+    // query found in title
+    title: 15,
+    partialTitle: 7,
+    // query found in terms
+    term: 5,
+    partialTerm: 2,
+  };
+}
+
+// Global search result kind enum, used by themes to style search results.
+class SearchResultKind {
+    static get index() { return  "index"; }
+    static get object() { return "object"; }
+    static get text() { return "text"; }
+    static get title() { return "title"; }
+}
+
+const _removeChildren = (element) => {
+  while (element && element.lastChild) element.removeChild(element.lastChild);
+};
+
+/**
+ * See https://developer.mozilla.org/en-US/docs/Web/JavaScript/Guide/Regular_Expressions#escaping
+ */
+const _escapeRegExp = (string) =>
+  string.replace(/[.*+\-?^${}()|[\]\\]/g, "\\$&"); // $& means the whole matched string
+
+const _displayItem = (item, searchTerms, highlightTerms) => {
+  const docBuilder = DOCUMENTATION_OPTIONS.BUILDER;
+  const docFileSuffix = DOCUMENTATION_OPTIONS.FILE_SUFFIX;
+  const docLinkSuffix = DOCUMENTATION_OPTIONS.LINK_SUFFIX;
+  const showSearchSummary = DOCUMENTATION_OPTIONS.SHOW_SEARCH_SUMMARY;
+  const contentRoot = document.documentElement.dataset.content_root;
+
+  const [docName, title, anchor, descr, score, _filename, kind] = item;
+
+  let listItem = document.createElement("li");
+  // Add a class representing the item's type:
+  // can be used by a theme's CSS selector for styling
+  // See SearchResultKind for the class names.
+  listItem.classList.add(`kind-${kind}`);
+  let requestUrl;
+  let linkUrl;
+  if (docBuilder === "dirhtml") {
+    // dirhtml builder
+    let dirname = docName + "/";
+    if (dirname.match(/\/index\/$/))
+      dirname = dirname.substring(0, dirname.length - 6);
+    else if (dirname === "index/") dirname = "";
+    requestUrl = contentRoot + dirname;
+    linkUrl = requestUrl;
+  } else {
+    // normal html builders
+    requestUrl = contentRoot + docName + docFileSuffix;
+    linkUrl = docName + docLinkSuffix;
+  }
+  let linkEl = listItem.appendChild(document.createElement("a"));
+  linkEl.href = linkUrl + anchor;
+  linkEl.dataset.score = score;
+  linkEl.innerHTML = title;
+  if (descr) {
+    listItem.appendChild(document.createElement("span")).innerHTML =
+      " (" + descr + ")";
+    // highlight search terms in the description
+    if (SPHINX_HIGHLIGHT_ENABLED)  // set in sphinx_highlight.js
+      highlightTerms.forEach((term) => _highlightText(listItem, term, "highlighted"));
+  }
+  else if (showSearchSummary)
+    fetch(requestUrl)
+      .then((responseData) => responseData.text())
+      .then((data) => {
+        if (data)
+          listItem.appendChild(
+            Search.makeSearchSummary(data, searchTerms, anchor)
+          );
+        // highlight search terms in the summary
+        if (SPHINX_HIGHLIGHT_ENABLED)  // set in sphinx_highlight.js
+          highlightTerms.forEach((term) => _highlightText(listItem, term, "highlighted"));
+      });
+  Search.output.appendChild(listItem);
+};
+const _finishSearch = (resultCount) => {
+  Search.stopPulse();
+  Search.title.innerText = _("Search Results");
+  if (!resultCount)
+    Search.status.innerText = Documentation.gettext(
+      "Your search did not match any documents. Please make sure that all words are spelled correctly and that you've selected enough categories."
+    );
+  else
+    Search.status.innerText = Documentation.ngettext(
+      "Search finished, found one page matching the search query.",
+      "Search finished, found ${resultCount} pages matching the search query.",
+      resultCount,
+    ).replace('${resultCount}', resultCount);
+};
+const _displayNextItem = (
+  results,
+  resultCount,
+  searchTerms,
+  highlightTerms,
+) => {
+  // results left, load the summary and display it
+  // this is intended to be dynamic (don't sub resultsCount)
+  if (results.length) {
+    _displayItem(results.pop(), searchTerms, highlightTerms);
+    setTimeout(
+      () => _displayNextItem(results, resultCount, searchTerms, highlightTerms),
+      5
+    );
+  }
+  // search finished, update title and status message
+  else _finishSearch(resultCount);
+};
+// Helper function used by query() to order search results.
+// Each input is an array of [docname, title, anchor, descr, score, filename, kind].
+// Order the results by score (in opposite order of appearance, since the
+// `_displayNextItem` function uses pop() to retrieve items) and then alphabetically.
+const _orderResultsByScoreThenName = (a, b) => {
+  const leftScore = a[4];
+  const rightScore = b[4];
+  if (leftScore === rightScore) {
+    // same score: sort alphabetically
+    const leftTitle = a[1].toLowerCase();
+    const rightTitle = b[1].toLowerCase();
+    if (leftTitle === rightTitle) return 0;
+    return leftTitle > rightTitle ? -1 : 1; // inverted is intentional
+  }
+  return leftScore > rightScore ? 1 : -1;
+};
+
+/**
+ * Default splitQuery function. Can be overridden in ``sphinx.search`` with a
+ * custom function per language.
+ *
+ * The regular expression works by splitting the string on consecutive characters
+ * that are not Unicode letters, numbers, underscores, or emoji characters.
+ * This is the same as ``\W+`` in Python, preserving the surrogate pair area.
+ */
+if (typeof splitQuery === "undefined") {
+  var splitQuery = (query) => query
+      .split(/[^\p{Letter}\p{Number}_\p{Emoji_Presentation}]+/gu)
+      .filter(term => term)  // remove remaining empty strings
+}
+
+/**
+ * Search Module
+ */
+const Search = {
+  _index: null,
+  _queued_query: null,
+  _pulse_status: -1,
+
+  htmlToText: (htmlString, anchor) => {
+    const htmlElement = new DOMParser().parseFromString(htmlString, 'text/html');
+    for (const removalQuery of [".headerlink", "script", "style"]) {
+      htmlElement.querySelectorAll(removalQuery).forEach((el) => { el.remove() });
+    }
+    if (anchor) {
+      const anchorContent = htmlElement.querySelector(`[role="main"] ${anchor}`);
+      if (anchorContent) return anchorContent.textContent;
+
+      console.warn(
+        `Anchored content block not found. Sphinx search tries to obtain it via DOM query '[role=main] ${anchor}'. Check your theme or template.`
+      );
+    }
+
+    // if anchor not specified or not found, fall back to main content
+    const docContent = htmlElement.querySelector('[role="main"]');
+    if (docContent) return docContent.textContent;
+
+    console.warn(
+      "Content block not found. Sphinx search tries to obtain it via DOM query '[role=main]'. Check your theme or template."
+    );
+    return "";
+  },
+
+  init: () => {
+    const query = new URLSearchParams(window.location.search).get("q");
+    document
+      .querySelectorAll('input[name="q"]')
+      .forEach((el) => (el.value = query));
+    if (query) Search.performSearch(query);
+  },
+
+  loadIndex: (url) =>
+    (document.body.appendChild(document.createElement("script")).src = url),
+
+  setIndex: (index) => {
+    Search._index = index;
+    if (Search._queued_query !== null) {
+      const query = Search._queued_query;
+      Search._queued_query = null;
+      Search.query(query);
+    }
+  },
+
+  hasIndex: () => Search._index !== null,
+
+  deferQuery: (query) => (Search._queued_query = query),
+
+  stopPulse: () => (Search._pulse_status = -1),
+
+  startPulse: () => {
+    if (Search._pulse_status >= 0) return;
+
+    const pulse = () => {
+      Search._pulse_status = (Search._pulse_status + 1) % 4;
+      Search.dots.innerText = ".".repeat(Search._pulse_status);
+      if (Search._pulse_status >= 0) window.setTimeout(pulse, 500);
+    };
+    pulse();
+  },
+
+  /**
+   * perform a search for something (or wait until index is loaded)
+   */
+  performSearch: (query) => {
+    // create the required interface elements
+    const searchText = document.createElement("h2");
+    searchText.textContent = _("Searching");
+    const searchSummary = document.createElement("p");
+    searchSummary.classList.add("search-summary");
+    searchSummary.innerText = "";
+    const searchList = document.createElement("ul");
+    searchList.setAttribute("role", "list");
+    searchList.classList.add("search");
+
+    const out = document.getElementById("search-results");
+    Search.title = out.appendChild(searchText);
+    Search.dots = Search.title.appendChild(document.createElement("span"));
+    Search.status = out.appendChild(searchSummary);
+    Search.output = out.appendChild(searchList);
+
+    const searchProgress = document.getElementById("search-progress");
+    // Some themes don't use the search progress node
+    if (searchProgress) {
+      searchProgress.innerText = _("Preparing search...");
+    }
+    Search.startPulse();
+
+    // index already loaded, the browser was quick!
+    if (Search.hasIndex()) Search.query(query);
+    else Search.deferQuery(query);
+  },
+
+  _parseQuery: (query) => {
+    // stem the search terms and add them to the correct list
+    const stemmer = new Stemmer();
+    const searchTerms = new Set();
+    const excludedTerms = new Set();
+    const highlightTerms = new Set();
+    const objectTerms = new Set(splitQuery(query.toLowerCase().trim()));
+    splitQuery(query.trim()).forEach((queryTerm) => {
+      const queryTermLower = queryTerm.toLowerCase();
+
+      // maybe skip this "word"
+      // stopwords array is from language_data.js
+      if (
+        stopwords.indexOf(queryTermLower) !== -1 ||
+        queryTerm.match(/^\d+$/)
+      )
+        return;
+
+      // stem the word
+      let word = stemmer.stemWord(queryTermLower);
+      // select the correct list
+      if (word[0] === "-") excludedTerms.add(word.substr(1));
+      else {
+        searchTerms.add(word);
+        highlightTerms.add(queryTermLower);
+      }
+    });
+
+    if (SPHINX_HIGHLIGHT_ENABLED) {  // set in sphinx_highlight.js
+      localStorage.setItem("sphinx_highlight_terms", [...highlightTerms].join(" "))
+    }
+
+    // console.debug("SEARCH: searching for:");
+    // console.info("required: ", [...searchTerms]);
+    // console.info("excluded: ", [...excludedTerms]);
+
+    return [query, searchTerms, excludedTerms, highlightTerms, objectTerms];
+  },
+
+  /**
+   * execute search (requires search index to be loaded)
+   */
+  _performSearch: (query, searchTerms, excludedTerms, highlightTerms, objectTerms) => {
+    const filenames = Search._index.filenames;
+    const docNames = Search._index.docnames;
+    const titles = Search._index.titles;
+    const allTitles = Search._index.alltitles;
+    const indexEntries = Search._index.indexentries;
+
+    // Collect multiple result groups to be sorted separately and then ordered.
+    // Each is an array of [docname, title, anchor, descr, score, filename, kind].
+    const normalResults = [];
+    const nonMainIndexResults = [];
+
+    _removeChildren(document.getElementById("search-progress"));
+
+    const queryLower = query.toLowerCase().trim();
+    for (const [title, foundTitles] of Object.entries(allTitles)) {
+      if (title.toLowerCase().trim().includes(queryLower) && (queryLower.length >= title.length/2)) {
+        for (const [file, id] of foundTitles) {
+          const score = Math.round(Scorer.title * queryLower.length / title.length);
+          const boost = titles[file] === title ? 1 : 0;  // add a boost for document titles
+          normalResults.push([
+            docNames[file],
+            titles[file] !== title ? `${titles[file]} > ${title}` : title,
+            id !== null ? "#" + id : "",
+            null,
+            score + boost,
+            filenames[file],
+            SearchResultKind.title,
+          ]);
+        }
+      }
+    }
+
+    // search for explicit entries in index directives
+    for (const [entry, foundEntries] of Object.entries(indexEntries)) {
+      if (entry.includes(queryLower) && (queryLower.length >= entry.length/2)) {
+        for (const [file, id, isMain] of foundEntries) {
+          const score = Math.round(100 * queryLower.length / entry.length);
+          const result = [
+            docNames[file],
+            titles[file],
+            id ? "#" + id : "",
+            null,
+            score,
+            filenames[file],
+            SearchResultKind.index,
+          ];
+          if (isMain) {
+            normalResults.push(result);
+          } else {
+            nonMainIndexResults.push(result);
+          }
+        }
+      }
+    }
+
+    // lookup as object
+    objectTerms.forEach((term) =>
+      normalResults.push(...Search.performObjectSearch(term, objectTerms))
+    );
+
+    // lookup as search terms in fulltext
+    normalResults.push(...Search.performTermsSearch(searchTerms, excludedTerms));
+
+    // let the scorer override scores with a custom scoring function
+    if (Scorer.score) {
+      normalResults.forEach((item) => (item[4] = Scorer.score(item)));
+      nonMainIndexResults.forEach((item) => (item[4] = Scorer.score(item)));
+    }
+
+    // Sort each group of results by score and then alphabetically by name.
+    normalResults.sort(_orderResultsByScoreThenName);
+    nonMainIndexResults.sort(_orderResultsByScoreThenName);
+
+    // Combine the result groups in (reverse) order.
+    // Non-main index entries are typically arbitrary cross-references,
+    // so display them after other results.
+    let results = [...nonMainIndexResults, ...normalResults];
+
+    // remove duplicate search results
+    // note the reversing of results, so that in the case of duplicates, the highest-scoring entry is kept
+    let seen = new Set();
+    results = results.reverse().reduce((acc, result) => {
+      let resultStr = result.slice(0, 4).concat([result[5]]).map(v => String(v)).join(',');
+      if (!seen.has(resultStr)) {
+        acc.push(result);
+        seen.add(resultStr);
+      }
+      return acc;
+    }, []);
+
+    return results.reverse();
+  },
+
+  query: (query) => {
+    const [searchQuery, searchTerms, excludedTerms, highlightTerms, objectTerms] = Search._parseQuery(query);
+    const results = Search._performSearch(searchQuery, searchTerms, excludedTerms, highlightTerms, objectTerms);
+
+    // for debugging
+    //Search.lastresults = results.slice();  // a copy
+    // console.info("search results:", Search.lastresults);
+
+    // print the results
+    _displayNextItem(results, results.length, searchTerms, highlightTerms);
+  },
+
+  /**
+   * search for object names
+   */
+  performObjectSearch: (object, objectTerms) => {
+    const filenames = Search._index.filenames;
+    const docNames = Search._index.docnames;
+    const objects = Search._index.objects;
+    const objNames = Search._index.objnames;
+    const titles = Search._index.titles;
+
+    const results = [];
+
+    const objectSearchCallback = (prefix, match) => {
+      const name = match[4]
+      const fullname = (prefix ? prefix + "." : "") + name;
+      const fullnameLower = fullname.toLowerCase();
+      if (fullnameLower.indexOf(object) < 0) return;
+
+      let score = 0;
+      const parts = fullnameLower.split(".");
+
+      // check for different match types: exact matches of full name or
+      // "last name" (i.e. last dotted part)
+      if (fullnameLower === object || parts.slice(-1)[0] === object)
+        score += Scorer.objNameMatch;
+      else if (parts.slice(-1)[0].indexOf(object) > -1)
+        score += Scorer.objPartialMatch; // matches in last name
+
+      const objName = objNames[match[1]][2];
+      const title = titles[match[0]];
+
+      // If more than one term searched for, we require other words to be
+      // found in the name/title/description
+      const otherTerms = new Set(objectTerms);
+      otherTerms.delete(object);
+      if (otherTerms.size > 0) {
+        const haystack = `${prefix} ${name} ${objName} ${title}`.toLowerCase();
+        if (
+          [...otherTerms].some((otherTerm) => haystack.indexOf(otherTerm) < 0)
+        )
+          return;
+      }
+
+      let anchor = match[3];
+      if (anchor === "") anchor = fullname;
+      else if (anchor === "-") anchor = objNames[match[1]][1] + "-" + fullname;
+
+      const descr = objName + _(", in ") + title;
+
+      // add custom score for some objects according to scorer
+      if (Scorer.objPrio.hasOwnProperty(match[2]))
+        score += Scorer.objPrio[match[2]];
+      else score += Scorer.objPrioDefault;
+
+      results.push([
+        docNames[match[0]],
+        fullname,
+        "#" + anchor,
+        descr,
+        score,
+        filenames[match[0]],
+        SearchResultKind.object,
+      ]);
+    };
+    Object.keys(objects).forEach((prefix) =>
+      objects[prefix].forEach((array) =>
+        objectSearchCallback(prefix, array)
+      )
+    );
+    return results;
+  },
+
+  /**
+   * search for full-text terms in the index
+   */
+  performTermsSearch: (searchTerms, excludedTerms) => {
+    // prepare search
+    const terms = Search._index.terms;
+    const titleTerms = Search._index.titleterms;
+    const filenames = Search._index.filenames;
+    const docNames = Search._index.docnames;
+    const titles = Search._index.titles;
+
+    const scoreMap = new Map();
+    const fileMap = new Map();
+
+    // perform the search on the required terms
+    searchTerms.forEach((word) => {
+      const files = [];
+      const arr = [
+        { files: terms[word], score: Scorer.term },
+        { files: titleTerms[word], score: Scorer.title },
+      ];
+      // add support for partial matches
+      if (word.length > 2) {
+        const escapedWord = _escapeRegExp(word);
+        if (!terms.hasOwnProperty(word)) {
+          Object.keys(terms).forEach((term) => {
+            if (term.match(escapedWord))
+              arr.push({ files: terms[term], score: Scorer.partialTerm });
+          });
+        }
+        if (!titleTerms.hasOwnProperty(word)) {
+          Object.keys(titleTerms).forEach((term) => {
+            if (term.match(escapedWord))
+              arr.push({ files: titleTerms[term], score: Scorer.partialTitle });
+          });
+        }
+      }
+
+      // no match but word was a required one
+      if (arr.every((record) => record.files === undefined)) return;
+
+      // found search word in contents
+      arr.forEach((record) => {
+        if (record.files === undefined) return;
+
+        let recordFiles = record.files;
+        if (recordFiles.length === undefined) recordFiles = [recordFiles];
+        files.push(...recordFiles);
+
+        // set score for the word in each file
+        recordFiles.forEach((file) => {
+          if (!scoreMap.has(file)) scoreMap.set(file, {});
+          scoreMap.get(file)[word] = record.score;
+        });
+      });
+
+      // create the mapping
+      files.forEach((file) => {
+        if (!fileMap.has(file)) fileMap.set(file, [word]);
+        else if (fileMap.get(file).indexOf(word) === -1) fileMap.get(file).push(word);
+      });
+    });
+
+    // now check if the files don't contain excluded terms
+    const results = [];
+    for (const [file, wordList] of fileMap) {
+      // check if all requirements are matched
+
+      // as search terms with length < 3 are discarded
+      const filteredTermCount = [...searchTerms].filter(
+        (term) => term.length > 2
+      ).length;
+      if (
+        wordList.length !== searchTerms.size &&
+        wordList.length !== filteredTermCount
+      )
+        continue;
+
+      // ensure that none of the excluded terms is in the search result
+      if (
+        [...excludedTerms].some(
+          (term) =>
+            terms[term] === file ||
+            titleTerms[term] === file ||
+            (terms[term] || []).includes(file) ||
+            (titleTerms[term] || []).includes(file)
+        )
+      )
+        break;
+
+      // select one (max) score for the file.
+      const score = Math.max(...wordList.map((w) => scoreMap.get(file)[w]));
+      // add result to the result list
+      results.push([
+        docNames[file],
+        titles[file],
+        "",
+        null,
+        score,
+        filenames[file],
+        SearchResultKind.text,
+      ]);
+    }
+    return results;
+  },
+
+  /**
+   * helper function to return a node containing the
+   * search summary for a given text. keywords is a list
+   * of stemmed words.
+   */
+  makeSearchSummary: (htmlText, keywords, anchor) => {
+    const text = Search.htmlToText(htmlText, anchor);
+    if (text === "") return null;
+
+    const textLower = text.toLowerCase();
+    const actualStartPosition = [...keywords]
+      .map((k) => textLower.indexOf(k.toLowerCase()))
+      .filter((i) => i > -1)
+      .slice(-1)[0];
+    const startWithContext = Math.max(actualStartPosition - 120, 0);
+
+    const top = startWithContext === 0 ? "" : "...";
+    const tail = startWithContext + 240 < text.length ? "..." : "";
+
+    let summary = document.createElement("p");
+    summary.classList.add("context");
+    summary.textContent = top + text.substr(startWithContext, 240).trim() + tail;
+
+    return summary;
+  },
+};
+
+_ready(Search.init);
diff --git a/_static/sphinx-design.min.css b/_static/sphinx-design.min.css
new file mode 100644
index 000000000..860c36da0
--- /dev/null
+++ b/_static/sphinx-design.min.css
@@ -0,0 +1 @@
+.sd-bg-primary{background-color:var(--sd-color-primary) !important}.sd-bg-text-primary{color:var(--sd-color-primary-text) !important}button.sd-bg-primary:focus,button.sd-bg-primary:hover{background-color:var(--sd-color-primary-highlight) !important}a.sd-bg-primary:focus,a.sd-bg-primary:hover{background-color:var(--sd-color-primary-highlight) !important}.sd-bg-secondary{background-color:var(--sd-color-secondary) !important}.sd-bg-text-secondary{color:var(--sd-color-secondary-text) !important}button.sd-bg-secondary:focus,button.sd-bg-secondary:hover{background-color:var(--sd-color-secondary-highlight) !important}a.sd-bg-secondary:focus,a.sd-bg-secondary:hover{background-color:var(--sd-color-secondary-highlight) !important}.sd-bg-success{background-color:var(--sd-color-success) !important}.sd-bg-text-success{color:var(--sd-color-success-text) !important}button.sd-bg-success:focus,button.sd-bg-success:hover{background-color:var(--sd-color-success-highlight) !important}a.sd-bg-success:focus,a.sd-bg-success:hover{background-color:var(--sd-color-success-highlight) !important}.sd-bg-info{background-color:var(--sd-color-info) !important}.sd-bg-text-info{color:var(--sd-color-info-text) !important}button.sd-bg-info:focus,button.sd-bg-info:hover{background-color:var(--sd-color-info-highlight) !important}a.sd-bg-info:focus,a.sd-bg-info:hover{background-color:var(--sd-color-info-highlight) !important}.sd-bg-warning{background-color:var(--sd-color-warning) !important}.sd-bg-text-warning{color:var(--sd-color-warning-text) !important}button.sd-bg-warning:focus,button.sd-bg-warning:hover{background-color:var(--sd-color-warning-highlight) !important}a.sd-bg-warning:focus,a.sd-bg-warning:hover{background-color:var(--sd-color-warning-highlight) !important}.sd-bg-danger{background-color:var(--sd-color-danger) !important}.sd-bg-text-danger{color:var(--sd-color-danger-text) !important}button.sd-bg-danger:focus,button.sd-bg-danger:hover{background-color:var(--sd-color-danger-highlight) !important}a.sd-bg-danger:focus,a.sd-bg-danger:hover{background-color:var(--sd-color-danger-highlight) !important}.sd-bg-light{background-color:var(--sd-color-light) !important}.sd-bg-text-light{color:var(--sd-color-light-text) !important}button.sd-bg-light:focus,button.sd-bg-light:hover{background-color:var(--sd-color-light-highlight) !important}a.sd-bg-light:focus,a.sd-bg-light:hover{background-color:var(--sd-color-light-highlight) !important}.sd-bg-muted{background-color:var(--sd-color-muted) !important}.sd-bg-text-muted{color:var(--sd-color-muted-text) !important}button.sd-bg-muted:focus,button.sd-bg-muted:hover{background-color:var(--sd-color-muted-highlight) !important}a.sd-bg-muted:focus,a.sd-bg-muted:hover{background-color:var(--sd-color-muted-highlight) !important}.sd-bg-dark{background-color:var(--sd-color-dark) !important}.sd-bg-text-dark{color:var(--sd-color-dark-text) !important}button.sd-bg-dark:focus,button.sd-bg-dark:hover{background-color:var(--sd-color-dark-highlight) !important}a.sd-bg-dark:focus,a.sd-bg-dark:hover{background-color:var(--sd-color-dark-highlight) !important}.sd-bg-black{background-color:var(--sd-color-black) !important}.sd-bg-text-black{color:var(--sd-color-black-text) !important}button.sd-bg-black:focus,button.sd-bg-black:hover{background-color:var(--sd-color-black-highlight) !important}a.sd-bg-black:focus,a.sd-bg-black:hover{background-color:var(--sd-color-black-highlight) !important}.sd-bg-white{background-color:var(--sd-color-white) !important}.sd-bg-text-white{color:var(--sd-color-white-text) !important}button.sd-bg-white:focus,button.sd-bg-white:hover{background-color:var(--sd-color-white-highlight) !important}a.sd-bg-white:focus,a.sd-bg-white:hover{background-color:var(--sd-color-white-highlight) !important}.sd-text-primary,.sd-text-primary>p{color:var(--sd-color-primary) !important}a.sd-text-primary:focus,a.sd-text-primary:hover{color:var(--sd-color-primary-highlight) !important}.sd-text-secondary,.sd-text-secondary>p{color:var(--sd-color-secondary) !important}a.sd-text-secondary:focus,a.sd-text-secondary:hover{color:var(--sd-color-secondary-highlight) !important}.sd-text-success,.sd-text-success>p{color:var(--sd-color-success) !important}a.sd-text-success:focus,a.sd-text-success:hover{color:var(--sd-color-success-highlight) !important}.sd-text-info,.sd-text-info>p{color:var(--sd-color-info) !important}a.sd-text-info:focus,a.sd-text-info:hover{color:var(--sd-color-info-highlight) !important}.sd-text-warning,.sd-text-warning>p{color:var(--sd-color-warning) !important}a.sd-text-warning:focus,a.sd-text-warning:hover{color:var(--sd-color-warning-highlight) !important}.sd-text-danger,.sd-text-danger>p{color:var(--sd-color-danger) !important}a.sd-text-danger:focus,a.sd-text-danger:hover{color:var(--sd-color-danger-highlight) !important}.sd-text-light,.sd-text-light>p{color:var(--sd-color-light) !important}a.sd-text-light:focus,a.sd-text-light:hover{color:var(--sd-color-light-highlight) !important}.sd-text-muted,.sd-text-muted>p{color:var(--sd-color-muted) !important}a.sd-text-muted:focus,a.sd-text-muted:hover{color:var(--sd-color-muted-highlight) !important}.sd-text-dark,.sd-text-dark>p{color:var(--sd-color-dark) !important}a.sd-text-dark:focus,a.sd-text-dark:hover{color:var(--sd-color-dark-highlight) !important}.sd-text-black,.sd-text-black>p{color:var(--sd-color-black) !important}a.sd-text-black:focus,a.sd-text-black:hover{color:var(--sd-color-black-highlight) !important}.sd-text-white,.sd-text-white>p{color:var(--sd-color-white) !important}a.sd-text-white:focus,a.sd-text-white:hover{color:var(--sd-color-white-highlight) !important}.sd-outline-primary{border-color:var(--sd-color-primary) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-primary:focus,a.sd-outline-primary:hover{border-color:var(--sd-color-primary-highlight) !important}.sd-outline-secondary{border-color:var(--sd-color-secondary) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-secondary:focus,a.sd-outline-secondary:hover{border-color:var(--sd-color-secondary-highlight) !important}.sd-outline-success{border-color:var(--sd-color-success) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-success:focus,a.sd-outline-success:hover{border-color:var(--sd-color-success-highlight) !important}.sd-outline-info{border-color:var(--sd-color-info) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-info:focus,a.sd-outline-info:hover{border-color:var(--sd-color-info-highlight) !important}.sd-outline-warning{border-color:var(--sd-color-warning) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-warning:focus,a.sd-outline-warning:hover{border-color:var(--sd-color-warning-highlight) !important}.sd-outline-danger{border-color:var(--sd-color-danger) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-danger:focus,a.sd-outline-danger:hover{border-color:var(--sd-color-danger-highlight) !important}.sd-outline-light{border-color:var(--sd-color-light) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-light:focus,a.sd-outline-light:hover{border-color:var(--sd-color-light-highlight) !important}.sd-outline-muted{border-color:var(--sd-color-muted) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-muted:focus,a.sd-outline-muted:hover{border-color:var(--sd-color-muted-highlight) !important}.sd-outline-dark{border-color:var(--sd-color-dark) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-dark:focus,a.sd-outline-dark:hover{border-color:var(--sd-color-dark-highlight) !important}.sd-outline-black{border-color:var(--sd-color-black) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-black:focus,a.sd-outline-black:hover{border-color:var(--sd-color-black-highlight) !important}.sd-outline-white{border-color:var(--sd-color-white) !important;border-style:solid !important;border-width:1px !important}a.sd-outline-white:focus,a.sd-outline-white:hover{border-color:var(--sd-color-white-highlight) !important}.sd-bg-transparent{background-color:transparent !important}.sd-outline-transparent{border-color:transparent !important}.sd-text-transparent{color:transparent !important}.sd-p-0{padding:0 !important}.sd-pt-0,.sd-py-0{padding-top:0 !important}.sd-pr-0,.sd-px-0{padding-right:0 !important}.sd-pb-0,.sd-py-0{padding-bottom:0 !important}.sd-pl-0,.sd-px-0{padding-left:0 !important}.sd-p-1{padding:.25rem !important}.sd-pt-1,.sd-py-1{padding-top:.25rem !important}.sd-pr-1,.sd-px-1{padding-right:.25rem !important}.sd-pb-1,.sd-py-1{padding-bottom:.25rem !important}.sd-pl-1,.sd-px-1{padding-left:.25rem !important}.sd-p-2{padding:.5rem !important}.sd-pt-2,.sd-py-2{padding-top:.5rem !important}.sd-pr-2,.sd-px-2{padding-right:.5rem !important}.sd-pb-2,.sd-py-2{padding-bottom:.5rem !important}.sd-pl-2,.sd-px-2{padding-left:.5rem !important}.sd-p-3{padding:1rem !important}.sd-pt-3,.sd-py-3{padding-top:1rem !important}.sd-pr-3,.sd-px-3{padding-right:1rem !important}.sd-pb-3,.sd-py-3{padding-bottom:1rem !important}.sd-pl-3,.sd-px-3{padding-left:1rem !important}.sd-p-4{padding:1.5rem !important}.sd-pt-4,.sd-py-4{padding-top:1.5rem !important}.sd-pr-4,.sd-px-4{padding-right:1.5rem !important}.sd-pb-4,.sd-py-4{padding-bottom:1.5rem !important}.sd-pl-4,.sd-px-4{padding-left:1.5rem !important}.sd-p-5{padding:3rem !important}.sd-pt-5,.sd-py-5{padding-top:3rem !important}.sd-pr-5,.sd-px-5{padding-right:3rem !important}.sd-pb-5,.sd-py-5{padding-bottom:3rem !important}.sd-pl-5,.sd-px-5{padding-left:3rem !important}.sd-m-auto{margin:auto !important}.sd-mt-auto,.sd-my-auto{margin-top:auto !important}.sd-mr-auto,.sd-mx-auto{margin-right:auto !important}.sd-mb-auto,.sd-my-auto{margin-bottom:auto !important}.sd-ml-auto,.sd-mx-auto{margin-left:auto !important}.sd-m-0{margin:0 !important}.sd-mt-0,.sd-my-0{margin-top:0 !important}.sd-mr-0,.sd-mx-0{margin-right:0 !important}.sd-mb-0,.sd-my-0{margin-bottom:0 !important}.sd-ml-0,.sd-mx-0{margin-left:0 !important}.sd-m-1{margin:.25rem !important}.sd-mt-1,.sd-my-1{margin-top:.25rem !important}.sd-mr-1,.sd-mx-1{margin-right:.25rem !important}.sd-mb-1,.sd-my-1{margin-bottom:.25rem !important}.sd-ml-1,.sd-mx-1{margin-left:.25rem !important}.sd-m-2{margin:.5rem !important}.sd-mt-2,.sd-my-2{margin-top:.5rem !important}.sd-mr-2,.sd-mx-2{margin-right:.5rem !important}.sd-mb-2,.sd-my-2{margin-bottom:.5rem !important}.sd-ml-2,.sd-mx-2{margin-left:.5rem !important}.sd-m-3{margin:1rem !important}.sd-mt-3,.sd-my-3{margin-top:1rem !important}.sd-mr-3,.sd-mx-3{margin-right:1rem !important}.sd-mb-3,.sd-my-3{margin-bottom:1rem !important}.sd-ml-3,.sd-mx-3{margin-left:1rem !important}.sd-m-4{margin:1.5rem !important}.sd-mt-4,.sd-my-4{margin-top:1.5rem !important}.sd-mr-4,.sd-mx-4{margin-right:1.5rem !important}.sd-mb-4,.sd-my-4{margin-bottom:1.5rem !important}.sd-ml-4,.sd-mx-4{margin-left:1.5rem !important}.sd-m-5{margin:3rem !important}.sd-mt-5,.sd-my-5{margin-top:3rem !important}.sd-mr-5,.sd-mx-5{margin-right:3rem !important}.sd-mb-5,.sd-my-5{margin-bottom:3rem !important}.sd-ml-5,.sd-mx-5{margin-left:3rem !important}.sd-w-25{width:25% !important}.sd-w-50{width:50% !important}.sd-w-75{width:75% !important}.sd-w-100{width:100% !important}.sd-w-auto{width:auto !important}.sd-h-25{height:25% !important}.sd-h-50{height:50% !important}.sd-h-75{height:75% !important}.sd-h-100{height:100% !important}.sd-h-auto{height:auto !important}.sd-d-none{display:none !important}.sd-d-inline{display:inline !important}.sd-d-inline-block{display:inline-block !important}.sd-d-block{display:block !important}.sd-d-grid{display:grid !important}.sd-d-flex-row{display:-ms-flexbox !important;display:flex !important;flex-direction:row !important}.sd-d-flex-column{display:-ms-flexbox !important;display:flex !important;flex-direction:column !important}.sd-d-inline-flex{display:-ms-inline-flexbox !important;display:inline-flex !important}@media(min-width: 576px){.sd-d-sm-none{display:none !important}.sd-d-sm-inline{display:inline !important}.sd-d-sm-inline-block{display:inline-block !important}.sd-d-sm-block{display:block !important}.sd-d-sm-grid{display:grid !important}.sd-d-sm-flex{display:-ms-flexbox !important;display:flex !important}.sd-d-sm-inline-flex{display:-ms-inline-flexbox !important;display:inline-flex !important}}@media(min-width: 768px){.sd-d-md-none{display:none !important}.sd-d-md-inline{display:inline !important}.sd-d-md-inline-block{display:inline-block !important}.sd-d-md-block{display:block !important}.sd-d-md-grid{display:grid !important}.sd-d-md-flex{display:-ms-flexbox !important;display:flex !important}.sd-d-md-inline-flex{display:-ms-inline-flexbox !important;display:inline-flex !important}}@media(min-width: 992px){.sd-d-lg-none{display:none !important}.sd-d-lg-inline{display:inline !important}.sd-d-lg-inline-block{display:inline-block !important}.sd-d-lg-block{display:block !important}.sd-d-lg-grid{display:grid !important}.sd-d-lg-flex{display:-ms-flexbox !important;display:flex !important}.sd-d-lg-inline-flex{display:-ms-inline-flexbox !important;display:inline-flex !important}}@media(min-width: 1200px){.sd-d-xl-none{display:none !important}.sd-d-xl-inline{display:inline !important}.sd-d-xl-inline-block{display:inline-block !important}.sd-d-xl-block{display:block !important}.sd-d-xl-grid{display:grid !important}.sd-d-xl-flex{display:-ms-flexbox !important;display:flex !important}.sd-d-xl-inline-flex{display:-ms-inline-flexbox !important;display:inline-flex !important}}.sd-align-major-start{justify-content:flex-start !important}.sd-align-major-end{justify-content:flex-end !important}.sd-align-major-center{justify-content:center !important}.sd-align-major-justify{justify-content:space-between !important}.sd-align-major-spaced{justify-content:space-evenly !important}.sd-align-minor-start{align-items:flex-start !important}.sd-align-minor-end{align-items:flex-end !important}.sd-align-minor-center{align-items:center !important}.sd-align-minor-stretch{align-items:stretch !important}.sd-text-justify{text-align:justify !important}.sd-text-left{text-align:left !important}.sd-text-right{text-align:right !important}.sd-text-center{text-align:center !important}.sd-font-weight-light{font-weight:300 !important}.sd-font-weight-lighter{font-weight:lighter !important}.sd-font-weight-normal{font-weight:400 !important}.sd-font-weight-bold{font-weight:700 !important}.sd-font-weight-bolder{font-weight:bolder !important}.sd-font-italic{font-style:italic !important}.sd-text-decoration-none{text-decoration:none !important}.sd-text-lowercase{text-transform:lowercase !important}.sd-text-uppercase{text-transform:uppercase !important}.sd-text-capitalize{text-transform:capitalize !important}.sd-text-wrap{white-space:normal !important}.sd-text-nowrap{white-space:nowrap !important}.sd-text-truncate{overflow:hidden;text-overflow:ellipsis;white-space:nowrap}.sd-fs-1,.sd-fs-1>p{font-size:calc(1.375rem + 1.5vw) !important;line-height:unset !important}.sd-fs-2,.sd-fs-2>p{font-size:calc(1.325rem + 0.9vw) !important;line-height:unset !important}.sd-fs-3,.sd-fs-3>p{font-size:calc(1.3rem + 0.6vw) !important;line-height:unset !important}.sd-fs-4,.sd-fs-4>p{font-size:calc(1.275rem + 0.3vw) !important;line-height:unset !important}.sd-fs-5,.sd-fs-5>p{font-size:1.25rem !important;line-height:unset !important}.sd-fs-6,.sd-fs-6>p{font-size:1rem !important;line-height:unset !important}.sd-border-0{border:0 solid !important}.sd-border-top-0{border-top:0 solid !important}.sd-border-bottom-0{border-bottom:0 solid !important}.sd-border-right-0{border-right:0 solid !important}.sd-border-left-0{border-left:0 solid !important}.sd-border-1{border:1px solid !important}.sd-border-top-1{border-top:1px solid !important}.sd-border-bottom-1{border-bottom:1px solid !important}.sd-border-right-1{border-right:1px solid !important}.sd-border-left-1{border-left:1px solid !important}.sd-border-2{border:2px solid !important}.sd-border-top-2{border-top:2px solid !important}.sd-border-bottom-2{border-bottom:2px solid !important}.sd-border-right-2{border-right:2px solid !important}.sd-border-left-2{border-left:2px solid !important}.sd-border-3{border:3px solid !important}.sd-border-top-3{border-top:3px solid !important}.sd-border-bottom-3{border-bottom:3px solid !important}.sd-border-right-3{border-right:3px solid !important}.sd-border-left-3{border-left:3px solid !important}.sd-border-4{border:4px solid !important}.sd-border-top-4{border-top:4px solid !important}.sd-border-bottom-4{border-bottom:4px solid !important}.sd-border-right-4{border-right:4px solid !important}.sd-border-left-4{border-left:4px solid !important}.sd-border-5{border:5px solid !important}.sd-border-top-5{border-top:5px solid !important}.sd-border-bottom-5{border-bottom:5px solid !important}.sd-border-right-5{border-right:5px solid !important}.sd-border-left-5{border-left:5px solid !important}.sd-rounded-0{border-radius:0 !important}.sd-rounded-1{border-radius:.2rem !important}.sd-rounded-2{border-radius:.3rem !important}.sd-rounded-3{border-radius:.5rem !important}.sd-rounded-pill{border-radius:50rem !important}.sd-rounded-circle{border-radius:50% !important}.shadow-none{box-shadow:none !important}.sd-shadow-sm{box-shadow:0 .125rem .25rem var(--sd-color-shadow) !important}.sd-shadow-md{box-shadow:0 .5rem 1rem var(--sd-color-shadow) !important}.sd-shadow-lg{box-shadow:0 1rem 3rem var(--sd-color-shadow) !important}@keyframes sd-slide-from-left{0%{transform:translateX(-100%)}100%{transform:translateX(0)}}@keyframes sd-slide-from-right{0%{transform:translateX(200%)}100%{transform:translateX(0)}}@keyframes sd-grow100{0%{transform:scale(0);opacity:.5}100%{transform:scale(1);opacity:1}}@keyframes sd-grow50{0%{transform:scale(0.5);opacity:.5}100%{transform:scale(1);opacity:1}}@keyframes sd-grow50-rot20{0%{transform:scale(0.5) rotateZ(-20deg);opacity:.5}75%{transform:scale(1) rotateZ(5deg);opacity:1}95%{transform:scale(1) rotateZ(-1deg);opacity:1}100%{transform:scale(1) rotateZ(0);opacity:1}}.sd-animate-slide-from-left{animation:1s ease-out 0s 1 normal none running sd-slide-from-left}.sd-animate-slide-from-right{animation:1s ease-out 0s 1 normal none running sd-slide-from-right}.sd-animate-grow100{animation:1s ease-out 0s 1 normal none running sd-grow100}.sd-animate-grow50{animation:1s ease-out 0s 1 normal none running sd-grow50}.sd-animate-grow50-rot20{animation:1s ease-out 0s 1 normal none running sd-grow50-rot20}.sd-badge{display:inline-block;padding:.35em .65em;font-size:.75em;font-weight:700;line-height:1;text-align:center;white-space:nowrap;vertical-align:baseline;border-radius:.25rem}.sd-badge:empty{display:none}a.sd-badge{text-decoration:none}.sd-btn .sd-badge{position:relative;top:-1px}.sd-btn{background-color:transparent;border:1px solid transparent;border-radius:.25rem;cursor:pointer;display:inline-block;font-weight:400;font-size:1rem;line-height:1.5;padding:.375rem .75rem;text-align:center;text-decoration:none;transition:color .15s ease-in-out,background-color .15s ease-in-out,border-color .15s ease-in-out,box-shadow .15s ease-in-out;vertical-align:middle;user-select:none;-moz-user-select:none;-ms-user-select:none;-webkit-user-select:none}.sd-btn:hover{text-decoration:none}@media(prefers-reduced-motion: reduce){.sd-btn{transition:none}}.sd-btn-primary,.sd-btn-outline-primary:hover,.sd-btn-outline-primary:focus{color:var(--sd-color-primary-text) !important;background-color:var(--sd-color-primary) !important;border-color:var(--sd-color-primary) !important;border-width:1px !important;border-style:solid !important}.sd-btn-primary:hover,.sd-btn-primary:focus{color:var(--sd-color-primary-text) !important;background-color:var(--sd-color-primary-highlight) !important;border-color:var(--sd-color-primary-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-primary{color:var(--sd-color-primary) !important;border-color:var(--sd-color-primary) !important;border-width:1px !important;border-style:solid !important}.sd-btn-secondary,.sd-btn-outline-secondary:hover,.sd-btn-outline-secondary:focus{color:var(--sd-color-secondary-text) !important;background-color:var(--sd-color-secondary) !important;border-color:var(--sd-color-secondary) !important;border-width:1px !important;border-style:solid !important}.sd-btn-secondary:hover,.sd-btn-secondary:focus{color:var(--sd-color-secondary-text) !important;background-color:var(--sd-color-secondary-highlight) !important;border-color:var(--sd-color-secondary-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-secondary{color:var(--sd-color-secondary) !important;border-color:var(--sd-color-secondary) !important;border-width:1px !important;border-style:solid !important}.sd-btn-success,.sd-btn-outline-success:hover,.sd-btn-outline-success:focus{color:var(--sd-color-success-text) !important;background-color:var(--sd-color-success) !important;border-color:var(--sd-color-success) !important;border-width:1px !important;border-style:solid !important}.sd-btn-success:hover,.sd-btn-success:focus{color:var(--sd-color-success-text) !important;background-color:var(--sd-color-success-highlight) !important;border-color:var(--sd-color-success-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-success{color:var(--sd-color-success) !important;border-color:var(--sd-color-success) !important;border-width:1px !important;border-style:solid !important}.sd-btn-info,.sd-btn-outline-info:hover,.sd-btn-outline-info:focus{color:var(--sd-color-info-text) !important;background-color:var(--sd-color-info) !important;border-color:var(--sd-color-info) !important;border-width:1px !important;border-style:solid !important}.sd-btn-info:hover,.sd-btn-info:focus{color:var(--sd-color-info-text) !important;background-color:var(--sd-color-info-highlight) !important;border-color:var(--sd-color-info-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-info{color:var(--sd-color-info) !important;border-color:var(--sd-color-info) !important;border-width:1px !important;border-style:solid !important}.sd-btn-warning,.sd-btn-outline-warning:hover,.sd-btn-outline-warning:focus{color:var(--sd-color-warning-text) !important;background-color:var(--sd-color-warning) !important;border-color:var(--sd-color-warning) !important;border-width:1px !important;border-style:solid !important}.sd-btn-warning:hover,.sd-btn-warning:focus{color:var(--sd-color-warning-text) !important;background-color:var(--sd-color-warning-highlight) !important;border-color:var(--sd-color-warning-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-warning{color:var(--sd-color-warning) !important;border-color:var(--sd-color-warning) !important;border-width:1px !important;border-style:solid !important}.sd-btn-danger,.sd-btn-outline-danger:hover,.sd-btn-outline-danger:focus{color:var(--sd-color-danger-text) !important;background-color:var(--sd-color-danger) !important;border-color:var(--sd-color-danger) !important;border-width:1px !important;border-style:solid !important}.sd-btn-danger:hover,.sd-btn-danger:focus{color:var(--sd-color-danger-text) !important;background-color:var(--sd-color-danger-highlight) !important;border-color:var(--sd-color-danger-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-danger{color:var(--sd-color-danger) !important;border-color:var(--sd-color-danger) !important;border-width:1px !important;border-style:solid !important}.sd-btn-light,.sd-btn-outline-light:hover,.sd-btn-outline-light:focus{color:var(--sd-color-light-text) !important;background-color:var(--sd-color-light) !important;border-color:var(--sd-color-light) !important;border-width:1px !important;border-style:solid !important}.sd-btn-light:hover,.sd-btn-light:focus{color:var(--sd-color-light-text) !important;background-color:var(--sd-color-light-highlight) !important;border-color:var(--sd-color-light-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-light{color:var(--sd-color-light) !important;border-color:var(--sd-color-light) !important;border-width:1px !important;border-style:solid !important}.sd-btn-muted,.sd-btn-outline-muted:hover,.sd-btn-outline-muted:focus{color:var(--sd-color-muted-text) !important;background-color:var(--sd-color-muted) !important;border-color:var(--sd-color-muted) !important;border-width:1px !important;border-style:solid !important}.sd-btn-muted:hover,.sd-btn-muted:focus{color:var(--sd-color-muted-text) !important;background-color:var(--sd-color-muted-highlight) !important;border-color:var(--sd-color-muted-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-muted{color:var(--sd-color-muted) !important;border-color:var(--sd-color-muted) !important;border-width:1px !important;border-style:solid !important}.sd-btn-dark,.sd-btn-outline-dark:hover,.sd-btn-outline-dark:focus{color:var(--sd-color-dark-text) !important;background-color:var(--sd-color-dark) !important;border-color:var(--sd-color-dark) !important;border-width:1px !important;border-style:solid !important}.sd-btn-dark:hover,.sd-btn-dark:focus{color:var(--sd-color-dark-text) !important;background-color:var(--sd-color-dark-highlight) !important;border-color:var(--sd-color-dark-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-dark{color:var(--sd-color-dark) !important;border-color:var(--sd-color-dark) !important;border-width:1px !important;border-style:solid !important}.sd-btn-black,.sd-btn-outline-black:hover,.sd-btn-outline-black:focus{color:var(--sd-color-black-text) !important;background-color:var(--sd-color-black) !important;border-color:var(--sd-color-black) !important;border-width:1px !important;border-style:solid !important}.sd-btn-black:hover,.sd-btn-black:focus{color:var(--sd-color-black-text) !important;background-color:var(--sd-color-black-highlight) !important;border-color:var(--sd-color-black-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-black{color:var(--sd-color-black) !important;border-color:var(--sd-color-black) !important;border-width:1px !important;border-style:solid !important}.sd-btn-white,.sd-btn-outline-white:hover,.sd-btn-outline-white:focus{color:var(--sd-color-white-text) !important;background-color:var(--sd-color-white) !important;border-color:var(--sd-color-white) !important;border-width:1px !important;border-style:solid !important}.sd-btn-white:hover,.sd-btn-white:focus{color:var(--sd-color-white-text) !important;background-color:var(--sd-color-white-highlight) !important;border-color:var(--sd-color-white-highlight) !important;border-width:1px !important;border-style:solid !important}.sd-btn-outline-white{color:var(--sd-color-white) !important;border-color:var(--sd-color-white) !important;border-width:1px !important;border-style:solid !important}.sd-stretched-link::after{position:absolute;top:0;right:0;bottom:0;left:0;z-index:1;content:""}.sd-hide-link-text{font-size:0}.sd-octicon,.sd-material-icon{display:inline-block;fill:currentColor;vertical-align:middle}.sd-avatar-xs{border-radius:50%;object-fit:cover;object-position:center;width:1rem;height:1rem}.sd-avatar-sm{border-radius:50%;object-fit:cover;object-position:center;width:3rem;height:3rem}.sd-avatar-md{border-radius:50%;object-fit:cover;object-position:center;width:5rem;height:5rem}.sd-avatar-lg{border-radius:50%;object-fit:cover;object-position:center;width:7rem;height:7rem}.sd-avatar-xl{border-radius:50%;object-fit:cover;object-position:center;width:10rem;height:10rem}.sd-avatar-inherit{border-radius:50%;object-fit:cover;object-position:center;width:inherit;height:inherit}.sd-avatar-initial{border-radius:50%;object-fit:cover;object-position:center;width:initial;height:initial}.sd-card{background-clip:border-box;background-color:var(--sd-color-card-background);border:1px solid var(--sd-color-card-border);border-radius:.25rem;color:var(--sd-color-card-text);display:-ms-flexbox;display:flex;-ms-flex-direction:column;flex-direction:column;min-width:0;position:relative;word-wrap:break-word}.sd-card>hr{margin-left:0;margin-right:0}.sd-card-hover:hover{border-color:var(--sd-color-card-border-hover);transform:scale(1.01)}.sd-card-body{-ms-flex:1 1 auto;flex:1 1 auto;padding:1rem 1rem}.sd-card-title{margin-bottom:.5rem}.sd-card-subtitle{margin-top:-0.25rem;margin-bottom:0}.sd-card-text:last-child{margin-bottom:0}.sd-card-link:hover{text-decoration:none}.sd-card-link+.card-link{margin-left:1rem}.sd-card-header{padding:.5rem 1rem;margin-bottom:0;background-color:var(--sd-color-card-header);border-bottom:1px solid var(--sd-color-card-border)}.sd-card-header:first-child{border-radius:calc(0.25rem - 1px) calc(0.25rem - 1px) 0 0}.sd-card-footer{padding:.5rem 1rem;background-color:var(--sd-color-card-footer);border-top:1px solid var(--sd-color-card-border)}.sd-card-footer:last-child{border-radius:0 0 calc(0.25rem - 1px) calc(0.25rem - 1px)}.sd-card-header-tabs{margin-right:-0.5rem;margin-bottom:-0.5rem;margin-left:-0.5rem;border-bottom:0}.sd-card-header-pills{margin-right:-0.5rem;margin-left:-0.5rem}.sd-card-img-overlay{position:absolute;top:0;right:0;bottom:0;left:0;padding:1rem;border-radius:calc(0.25rem - 1px)}.sd-card-img,.sd-card-img-bottom,.sd-card-img-top{width:100%}.sd-card-img,.sd-card-img-top{border-top-left-radius:calc(0.25rem - 1px);border-top-right-radius:calc(0.25rem - 1px)}.sd-card-img,.sd-card-img-bottom{border-bottom-left-radius:calc(0.25rem - 1px);border-bottom-right-radius:calc(0.25rem - 1px)}.sd-cards-carousel{width:100%;display:flex;flex-wrap:nowrap;-ms-flex-direction:row;flex-direction:row;overflow-x:hidden;scroll-snap-type:x mandatory}.sd-cards-carousel.sd-show-scrollbar{overflow-x:auto}.sd-cards-carousel:hover,.sd-cards-carousel:focus{overflow-x:auto}.sd-cards-carousel>.sd-card{flex-shrink:0;scroll-snap-align:start}.sd-cards-carousel>.sd-card:not(:last-child){margin-right:3px}.sd-card-cols-1>.sd-card{width:90%}.sd-card-cols-2>.sd-card{width:45%}.sd-card-cols-3>.sd-card{width:30%}.sd-card-cols-4>.sd-card{width:22.5%}.sd-card-cols-5>.sd-card{width:18%}.sd-card-cols-6>.sd-card{width:15%}.sd-card-cols-7>.sd-card{width:12.8571428571%}.sd-card-cols-8>.sd-card{width:11.25%}.sd-card-cols-9>.sd-card{width:10%}.sd-card-cols-10>.sd-card{width:9%}.sd-card-cols-11>.sd-card{width:8.1818181818%}.sd-card-cols-12>.sd-card{width:7.5%}.sd-container,.sd-container-fluid,.sd-container-lg,.sd-container-md,.sd-container-sm,.sd-container-xl{margin-left:auto;margin-right:auto;padding-left:var(--sd-gutter-x, 0.75rem);padding-right:var(--sd-gutter-x, 0.75rem);width:100%}@media(min-width: 576px){.sd-container-sm,.sd-container{max-width:540px}}@media(min-width: 768px){.sd-container-md,.sd-container-sm,.sd-container{max-width:720px}}@media(min-width: 992px){.sd-container-lg,.sd-container-md,.sd-container-sm,.sd-container{max-width:960px}}@media(min-width: 1200px){.sd-container-xl,.sd-container-lg,.sd-container-md,.sd-container-sm,.sd-container{max-width:1140px}}.sd-row{--sd-gutter-x: 1.5rem;--sd-gutter-y: 0;display:-ms-flexbox;display:flex;-ms-flex-wrap:wrap;flex-wrap:wrap;margin-top:calc(var(--sd-gutter-y) * -1);margin-right:calc(var(--sd-gutter-x) * -0.5);margin-left:calc(var(--sd-gutter-x) * -0.5)}.sd-row>*{box-sizing:border-box;flex-shrink:0;width:100%;max-width:100%;padding-right:calc(var(--sd-gutter-x) * 0.5);padding-left:calc(var(--sd-gutter-x) * 0.5);margin-top:var(--sd-gutter-y)}.sd-col{flex:1 0 0%;-ms-flex:1 0 0%}.sd-row-cols-auto>*{flex:0 0 auto;width:auto}.sd-row-cols-1>*{flex:0 0 auto;-ms-flex:0 0 auto;width:100%}.sd-row-cols-2>*{flex:0 0 auto;-ms-flex:0 0 auto;width:50%}.sd-row-cols-3>*{flex:0 0 auto;-ms-flex:0 0 auto;width:33.3333333333%}.sd-row-cols-4>*{flex:0 0 auto;-ms-flex:0 0 auto;width:25%}.sd-row-cols-5>*{flex:0 0 auto;-ms-flex:0 0 auto;width:20%}.sd-row-cols-6>*{flex:0 0 auto;-ms-flex:0 0 auto;width:16.6666666667%}.sd-row-cols-7>*{flex:0 0 auto;-ms-flex:0 0 auto;width:14.2857142857%}.sd-row-cols-8>*{flex:0 0 auto;-ms-flex:0 0 auto;width:12.5%}.sd-row-cols-9>*{flex:0 0 auto;-ms-flex:0 0 auto;width:11.1111111111%}.sd-row-cols-10>*{flex:0 0 auto;-ms-flex:0 0 auto;width:10%}.sd-row-cols-11>*{flex:0 0 auto;-ms-flex:0 0 auto;width:9.0909090909%}.sd-row-cols-12>*{flex:0 0 auto;-ms-flex:0 0 auto;width:8.3333333333%}@media(min-width: 576px){.sd-col-sm{flex:1 0 0%;-ms-flex:1 0 0%}.sd-row-cols-sm-auto{flex:1 0 auto;-ms-flex:1 0 auto;width:100%}.sd-row-cols-sm-1>*{flex:0 0 auto;-ms-flex:0 0 auto;width:100%}.sd-row-cols-sm-2>*{flex:0 0 auto;-ms-flex:0 0 auto;width:50%}.sd-row-cols-sm-3>*{flex:0 0 auto;-ms-flex:0 0 auto;width:33.3333333333%}.sd-row-cols-sm-4>*{flex:0 0 auto;-ms-flex:0 0 auto;width:25%}.sd-row-cols-sm-5>*{flex:0 0 auto;-ms-flex:0 0 auto;width:20%}.sd-row-cols-sm-6>*{flex:0 0 auto;-ms-flex:0 0 auto;width:16.6666666667%}.sd-row-cols-sm-7>*{flex:0 0 auto;-ms-flex:0 0 auto;width:14.2857142857%}.sd-row-cols-sm-8>*{flex:0 0 auto;-ms-flex:0 0 auto;width:12.5%}.sd-row-cols-sm-9>*{flex:0 0 auto;-ms-flex:0 0 auto;width:11.1111111111%}.sd-row-cols-sm-10>*{flex:0 0 auto;-ms-flex:0 0 auto;width:10%}.sd-row-cols-sm-11>*{flex:0 0 auto;-ms-flex:0 0 auto;width:9.0909090909%}.sd-row-cols-sm-12>*{flex:0 0 auto;-ms-flex:0 0 auto;width:8.3333333333%}}@media(min-width: 768px){.sd-col-md{flex:1 0 0%;-ms-flex:1 0 0%}.sd-row-cols-md-auto{flex:1 0 auto;-ms-flex:1 0 auto;width:100%}.sd-row-cols-md-1>*{flex:0 0 auto;-ms-flex:0 0 auto;width:100%}.sd-row-cols-md-2>*{flex:0 0 auto;-ms-flex:0 0 auto;width:50%}.sd-row-cols-md-3>*{flex:0 0 auto;-ms-flex:0 0 auto;width:33.3333333333%}.sd-row-cols-md-4>*{flex:0 0 auto;-ms-flex:0 0 auto;width:25%}.sd-row-cols-md-5>*{flex:0 0 auto;-ms-flex:0 0 auto;width:20%}.sd-row-cols-md-6>*{flex:0 0 auto;-ms-flex:0 0 auto;width:16.6666666667%}.sd-row-cols-md-7>*{flex:0 0 auto;-ms-flex:0 0 auto;width:14.2857142857%}.sd-row-cols-md-8>*{flex:0 0 auto;-ms-flex:0 0 auto;width:12.5%}.sd-row-cols-md-9>*{flex:0 0 auto;-ms-flex:0 0 auto;width:11.1111111111%}.sd-row-cols-md-10>*{flex:0 0 auto;-ms-flex:0 0 auto;width:10%}.sd-row-cols-md-11>*{flex:0 0 auto;-ms-flex:0 0 auto;width:9.0909090909%}.sd-row-cols-md-12>*{flex:0 0 auto;-ms-flex:0 0 auto;width:8.3333333333%}}@media(min-width: 992px){.sd-col-lg{flex:1 0 0%;-ms-flex:1 0 0%}.sd-row-cols-lg-auto{flex:1 0 auto;-ms-flex:1 0 auto;width:100%}.sd-row-cols-lg-1>*{flex:0 0 auto;-ms-flex:0 0 auto;width:100%}.sd-row-cols-lg-2>*{flex:0 0 auto;-ms-flex:0 0 auto;width:50%}.sd-row-cols-lg-3>*{flex:0 0 auto;-ms-flex:0 0 auto;width:33.3333333333%}.sd-row-cols-lg-4>*{flex:0 0 auto;-ms-flex:0 0 auto;width:25%}.sd-row-cols-lg-5>*{flex:0 0 auto;-ms-flex:0 0 auto;width:20%}.sd-row-cols-lg-6>*{flex:0 0 auto;-ms-flex:0 0 auto;width:16.6666666667%}.sd-row-cols-lg-7>*{flex:0 0 auto;-ms-flex:0 0 auto;width:14.2857142857%}.sd-row-cols-lg-8>*{flex:0 0 auto;-ms-flex:0 0 auto;width:12.5%}.sd-row-cols-lg-9>*{flex:0 0 auto;-ms-flex:0 0 auto;width:11.1111111111%}.sd-row-cols-lg-10>*{flex:0 0 auto;-ms-flex:0 0 auto;width:10%}.sd-row-cols-lg-11>*{flex:0 0 auto;-ms-flex:0 0 auto;width:9.0909090909%}.sd-row-cols-lg-12>*{flex:0 0 auto;-ms-flex:0 0 auto;width:8.3333333333%}}@media(min-width: 1200px){.sd-col-xl{flex:1 0 0%;-ms-flex:1 0 0%}.sd-row-cols-xl-auto{flex:1 0 auto;-ms-flex:1 0 auto;width:100%}.sd-row-cols-xl-1>*{flex:0 0 auto;-ms-flex:0 0 auto;width:100%}.sd-row-cols-xl-2>*{flex:0 0 auto;-ms-flex:0 0 auto;width:50%}.sd-row-cols-xl-3>*{flex:0 0 auto;-ms-flex:0 0 auto;width:33.3333333333%}.sd-row-cols-xl-4>*{flex:0 0 auto;-ms-flex:0 0 auto;width:25%}.sd-row-cols-xl-5>*{flex:0 0 auto;-ms-flex:0 0 auto;width:20%}.sd-row-cols-xl-6>*{flex:0 0 auto;-ms-flex:0 0 auto;width:16.6666666667%}.sd-row-cols-xl-7>*{flex:0 0 auto;-ms-flex:0 0 auto;width:14.2857142857%}.sd-row-cols-xl-8>*{flex:0 0 auto;-ms-flex:0 0 auto;width:12.5%}.sd-row-cols-xl-9>*{flex:0 0 auto;-ms-flex:0 0 auto;width:11.1111111111%}.sd-row-cols-xl-10>*{flex:0 0 auto;-ms-flex:0 0 auto;width:10%}.sd-row-cols-xl-11>*{flex:0 0 auto;-ms-flex:0 0 auto;width:9.0909090909%}.sd-row-cols-xl-12>*{flex:0 0 auto;-ms-flex:0 0 auto;width:8.3333333333%}}.sd-col-auto{flex:0 0 auto;-ms-flex:0 0 auto;width:auto}.sd-col-1{flex:0 0 auto;-ms-flex:0 0 auto;width:8.3333333333%}.sd-col-2{flex:0 0 auto;-ms-flex:0 0 auto;width:16.6666666667%}.sd-col-3{flex:0 0 auto;-ms-flex:0 0 auto;width:25%}.sd-col-4{flex:0 0 auto;-ms-flex:0 0 auto;width:33.3333333333%}.sd-col-5{flex:0 0 auto;-ms-flex:0 0 auto;width:41.6666666667%}.sd-col-6{flex:0 0 auto;-ms-flex:0 0 auto;width:50%}.sd-col-7{flex:0 0 auto;-ms-flex:0 0 auto;width:58.3333333333%}.sd-col-8{flex:0 0 auto;-ms-flex:0 0 auto;width:66.6666666667%}.sd-col-9{flex:0 0 auto;-ms-flex:0 0 auto;width:75%}.sd-col-10{flex:0 0 auto;-ms-flex:0 0 auto;width:83.3333333333%}.sd-col-11{flex:0 0 auto;-ms-flex:0 0 auto;width:91.6666666667%}.sd-col-12{flex:0 0 auto;-ms-flex:0 0 auto;width:100%}.sd-g-0,.sd-gy-0{--sd-gutter-y: 0}.sd-g-0,.sd-gx-0{--sd-gutter-x: 0}.sd-g-1,.sd-gy-1{--sd-gutter-y: 0.25rem}.sd-g-1,.sd-gx-1{--sd-gutter-x: 0.25rem}.sd-g-2,.sd-gy-2{--sd-gutter-y: 0.5rem}.sd-g-2,.sd-gx-2{--sd-gutter-x: 0.5rem}.sd-g-3,.sd-gy-3{--sd-gutter-y: 1rem}.sd-g-3,.sd-gx-3{--sd-gutter-x: 1rem}.sd-g-4,.sd-gy-4{--sd-gutter-y: 1.5rem}.sd-g-4,.sd-gx-4{--sd-gutter-x: 1.5rem}.sd-g-5,.sd-gy-5{--sd-gutter-y: 3rem}.sd-g-5,.sd-gx-5{--sd-gutter-x: 3rem}@media(min-width: 576px){.sd-col-sm-auto{-ms-flex:0 0 auto;flex:0 0 auto;width:auto}.sd-col-sm-1{-ms-flex:0 0 auto;flex:0 0 auto;width:8.3333333333%}.sd-col-sm-2{-ms-flex:0 0 auto;flex:0 0 auto;width:16.6666666667%}.sd-col-sm-3{-ms-flex:0 0 auto;flex:0 0 auto;width:25%}.sd-col-sm-4{-ms-flex:0 0 auto;flex:0 0 auto;width:33.3333333333%}.sd-col-sm-5{-ms-flex:0 0 auto;flex:0 0 auto;width:41.6666666667%}.sd-col-sm-6{-ms-flex:0 0 auto;flex:0 0 auto;width:50%}.sd-col-sm-7{-ms-flex:0 0 auto;flex:0 0 auto;width:58.3333333333%}.sd-col-sm-8{-ms-flex:0 0 auto;flex:0 0 auto;width:66.6666666667%}.sd-col-sm-9{-ms-flex:0 0 auto;flex:0 0 auto;width:75%}.sd-col-sm-10{-ms-flex:0 0 auto;flex:0 0 auto;width:83.3333333333%}.sd-col-sm-11{-ms-flex:0 0 auto;flex:0 0 auto;width:91.6666666667%}.sd-col-sm-12{-ms-flex:0 0 auto;flex:0 0 auto;width:100%}.sd-g-sm-0,.sd-gy-sm-0{--sd-gutter-y: 0}.sd-g-sm-0,.sd-gx-sm-0{--sd-gutter-x: 0}.sd-g-sm-1,.sd-gy-sm-1{--sd-gutter-y: 0.25rem}.sd-g-sm-1,.sd-gx-sm-1{--sd-gutter-x: 0.25rem}.sd-g-sm-2,.sd-gy-sm-2{--sd-gutter-y: 0.5rem}.sd-g-sm-2,.sd-gx-sm-2{--sd-gutter-x: 0.5rem}.sd-g-sm-3,.sd-gy-sm-3{--sd-gutter-y: 1rem}.sd-g-sm-3,.sd-gx-sm-3{--sd-gutter-x: 1rem}.sd-g-sm-4,.sd-gy-sm-4{--sd-gutter-y: 1.5rem}.sd-g-sm-4,.sd-gx-sm-4{--sd-gutter-x: 1.5rem}.sd-g-sm-5,.sd-gy-sm-5{--sd-gutter-y: 3rem}.sd-g-sm-5,.sd-gx-sm-5{--sd-gutter-x: 3rem}}@media(min-width: 768px){.sd-col-md-auto{-ms-flex:0 0 auto;flex:0 0 auto;width:auto}.sd-col-md-1{-ms-flex:0 0 auto;flex:0 0 auto;width:8.3333333333%}.sd-col-md-2{-ms-flex:0 0 auto;flex:0 0 auto;width:16.6666666667%}.sd-col-md-3{-ms-flex:0 0 auto;flex:0 0 auto;width:25%}.sd-col-md-4{-ms-flex:0 0 auto;flex:0 0 auto;width:33.3333333333%}.sd-col-md-5{-ms-flex:0 0 auto;flex:0 0 auto;width:41.6666666667%}.sd-col-md-6{-ms-flex:0 0 auto;flex:0 0 auto;width:50%}.sd-col-md-7{-ms-flex:0 0 auto;flex:0 0 auto;width:58.3333333333%}.sd-col-md-8{-ms-flex:0 0 auto;flex:0 0 auto;width:66.6666666667%}.sd-col-md-9{-ms-flex:0 0 auto;flex:0 0 auto;width:75%}.sd-col-md-10{-ms-flex:0 0 auto;flex:0 0 auto;width:83.3333333333%}.sd-col-md-11{-ms-flex:0 0 auto;flex:0 0 auto;width:91.6666666667%}.sd-col-md-12{-ms-flex:0 0 auto;flex:0 0 auto;width:100%}.sd-g-md-0,.sd-gy-md-0{--sd-gutter-y: 0}.sd-g-md-0,.sd-gx-md-0{--sd-gutter-x: 0}.sd-g-md-1,.sd-gy-md-1{--sd-gutter-y: 0.25rem}.sd-g-md-1,.sd-gx-md-1{--sd-gutter-x: 0.25rem}.sd-g-md-2,.sd-gy-md-2{--sd-gutter-y: 0.5rem}.sd-g-md-2,.sd-gx-md-2{--sd-gutter-x: 0.5rem}.sd-g-md-3,.sd-gy-md-3{--sd-gutter-y: 1rem}.sd-g-md-3,.sd-gx-md-3{--sd-gutter-x: 1rem}.sd-g-md-4,.sd-gy-md-4{--sd-gutter-y: 1.5rem}.sd-g-md-4,.sd-gx-md-4{--sd-gutter-x: 1.5rem}.sd-g-md-5,.sd-gy-md-5{--sd-gutter-y: 3rem}.sd-g-md-5,.sd-gx-md-5{--sd-gutter-x: 3rem}}@media(min-width: 992px){.sd-col-lg-auto{-ms-flex:0 0 auto;flex:0 0 auto;width:auto}.sd-col-lg-1{-ms-flex:0 0 auto;flex:0 0 auto;width:8.3333333333%}.sd-col-lg-2{-ms-flex:0 0 auto;flex:0 0 auto;width:16.6666666667%}.sd-col-lg-3{-ms-flex:0 0 auto;flex:0 0 auto;width:25%}.sd-col-lg-4{-ms-flex:0 0 auto;flex:0 0 auto;width:33.3333333333%}.sd-col-lg-5{-ms-flex:0 0 auto;flex:0 0 auto;width:41.6666666667%}.sd-col-lg-6{-ms-flex:0 0 auto;flex:0 0 auto;width:50%}.sd-col-lg-7{-ms-flex:0 0 auto;flex:0 0 auto;width:58.3333333333%}.sd-col-lg-8{-ms-flex:0 0 auto;flex:0 0 auto;width:66.6666666667%}.sd-col-lg-9{-ms-flex:0 0 auto;flex:0 0 auto;width:75%}.sd-col-lg-10{-ms-flex:0 0 auto;flex:0 0 auto;width:83.3333333333%}.sd-col-lg-11{-ms-flex:0 0 auto;flex:0 0 auto;width:91.6666666667%}.sd-col-lg-12{-ms-flex:0 0 auto;flex:0 0 auto;width:100%}.sd-g-lg-0,.sd-gy-lg-0{--sd-gutter-y: 0}.sd-g-lg-0,.sd-gx-lg-0{--sd-gutter-x: 0}.sd-g-lg-1,.sd-gy-lg-1{--sd-gutter-y: 0.25rem}.sd-g-lg-1,.sd-gx-lg-1{--sd-gutter-x: 0.25rem}.sd-g-lg-2,.sd-gy-lg-2{--sd-gutter-y: 0.5rem}.sd-g-lg-2,.sd-gx-lg-2{--sd-gutter-x: 0.5rem}.sd-g-lg-3,.sd-gy-lg-3{--sd-gutter-y: 1rem}.sd-g-lg-3,.sd-gx-lg-3{--sd-gutter-x: 1rem}.sd-g-lg-4,.sd-gy-lg-4{--sd-gutter-y: 1.5rem}.sd-g-lg-4,.sd-gx-lg-4{--sd-gutter-x: 1.5rem}.sd-g-lg-5,.sd-gy-lg-5{--sd-gutter-y: 3rem}.sd-g-lg-5,.sd-gx-lg-5{--sd-gutter-x: 3rem}}@media(min-width: 1200px){.sd-col-xl-auto{-ms-flex:0 0 auto;flex:0 0 auto;width:auto}.sd-col-xl-1{-ms-flex:0 0 auto;flex:0 0 auto;width:8.3333333333%}.sd-col-xl-2{-ms-flex:0 0 auto;flex:0 0 auto;width:16.6666666667%}.sd-col-xl-3{-ms-flex:0 0 auto;flex:0 0 auto;width:25%}.sd-col-xl-4{-ms-flex:0 0 auto;flex:0 0 auto;width:33.3333333333%}.sd-col-xl-5{-ms-flex:0 0 auto;flex:0 0 auto;width:41.6666666667%}.sd-col-xl-6{-ms-flex:0 0 auto;flex:0 0 auto;width:50%}.sd-col-xl-7{-ms-flex:0 0 auto;flex:0 0 auto;width:58.3333333333%}.sd-col-xl-8{-ms-flex:0 0 auto;flex:0 0 auto;width:66.6666666667%}.sd-col-xl-9{-ms-flex:0 0 auto;flex:0 0 auto;width:75%}.sd-col-xl-10{-ms-flex:0 0 auto;flex:0 0 auto;width:83.3333333333%}.sd-col-xl-11{-ms-flex:0 0 auto;flex:0 0 auto;width:91.6666666667%}.sd-col-xl-12{-ms-flex:0 0 auto;flex:0 0 auto;width:100%}.sd-g-xl-0,.sd-gy-xl-0{--sd-gutter-y: 0}.sd-g-xl-0,.sd-gx-xl-0{--sd-gutter-x: 0}.sd-g-xl-1,.sd-gy-xl-1{--sd-gutter-y: 0.25rem}.sd-g-xl-1,.sd-gx-xl-1{--sd-gutter-x: 0.25rem}.sd-g-xl-2,.sd-gy-xl-2{--sd-gutter-y: 0.5rem}.sd-g-xl-2,.sd-gx-xl-2{--sd-gutter-x: 0.5rem}.sd-g-xl-3,.sd-gy-xl-3{--sd-gutter-y: 1rem}.sd-g-xl-3,.sd-gx-xl-3{--sd-gutter-x: 1rem}.sd-g-xl-4,.sd-gy-xl-4{--sd-gutter-y: 1.5rem}.sd-g-xl-4,.sd-gx-xl-4{--sd-gutter-x: 1.5rem}.sd-g-xl-5,.sd-gy-xl-5{--sd-gutter-y: 3rem}.sd-g-xl-5,.sd-gx-xl-5{--sd-gutter-x: 3rem}}.sd-flex-row-reverse{flex-direction:row-reverse !important}details.sd-dropdown{position:relative;font-size:var(--sd-fontsize-dropdown)}details.sd-dropdown:hover{cursor:pointer}details.sd-dropdown .sd-summary-content{cursor:default}details.sd-dropdown summary.sd-summary-title{padding:.5em .6em .5em 1em;font-size:var(--sd-fontsize-dropdown-title);font-weight:var(--sd-fontweight-dropdown-title);user-select:none;-moz-user-select:none;-ms-user-select:none;-webkit-user-select:none;list-style:none;display:inline-flex;justify-content:space-between}details.sd-dropdown summary.sd-summary-title::-webkit-details-marker{display:none}details.sd-dropdown summary.sd-summary-title:focus{outline:none}details.sd-dropdown summary.sd-summary-title .sd-summary-icon{margin-right:.6em;display:inline-flex;align-items:center}details.sd-dropdown summary.sd-summary-title .sd-summary-icon svg{opacity:.8}details.sd-dropdown summary.sd-summary-title .sd-summary-text{flex-grow:1;line-height:1.5;padding-right:.5rem}details.sd-dropdown summary.sd-summary-title .sd-summary-state-marker{pointer-events:none;display:inline-flex;align-items:center}details.sd-dropdown summary.sd-summary-title .sd-summary-state-marker svg{opacity:.6}details.sd-dropdown summary.sd-summary-title:hover .sd-summary-state-marker svg{opacity:1;transform:scale(1.1)}details.sd-dropdown[open] summary .sd-octicon.no-title{visibility:hidden}details.sd-dropdown .sd-summary-chevron-right{transition:.25s}details.sd-dropdown[open]>.sd-summary-title .sd-summary-chevron-right{transform:rotate(90deg)}details.sd-dropdown[open]>.sd-summary-title .sd-summary-chevron-down{transform:rotate(180deg)}details.sd-dropdown:not([open]).sd-card{border:none}details.sd-dropdown:not([open])>.sd-card-header{border:1px solid var(--sd-color-card-border);border-radius:.25rem}details.sd-dropdown.sd-fade-in[open] summary~*{-moz-animation:sd-fade-in .5s ease-in-out;-webkit-animation:sd-fade-in .5s ease-in-out;animation:sd-fade-in .5s ease-in-out}details.sd-dropdown.sd-fade-in-slide-down[open] summary~*{-moz-animation:sd-fade-in .5s ease-in-out,sd-slide-down .5s ease-in-out;-webkit-animation:sd-fade-in .5s ease-in-out,sd-slide-down .5s ease-in-out;animation:sd-fade-in .5s ease-in-out,sd-slide-down .5s ease-in-out}.sd-col>.sd-dropdown{width:100%}.sd-summary-content>.sd-tab-set:first-child{margin-top:0}@keyframes sd-fade-in{0%{opacity:0}100%{opacity:1}}@keyframes sd-slide-down{0%{transform:translate(0, -10px)}100%{transform:translate(0, 0)}}.sd-tab-set{border-radius:.125rem;display:flex;flex-wrap:wrap;margin:1em 0;position:relative}.sd-tab-set>input{opacity:0;position:absolute}.sd-tab-set>input:checked+label{border-color:var(--sd-color-tabs-underline-active);color:var(--sd-color-tabs-label-active)}.sd-tab-set>input:checked+label+.sd-tab-content{display:block}.sd-tab-set>input:not(:checked)+label:hover{color:var(--sd-color-tabs-label-hover);border-color:var(--sd-color-tabs-underline-hover)}.sd-tab-set>input:focus+label{outline-style:auto}.sd-tab-set>input:not(.focus-visible)+label{outline:none;-webkit-tap-highlight-color:transparent}.sd-tab-set>label{border-bottom:.125rem solid transparent;margin-bottom:0;color:var(--sd-color-tabs-label-inactive);border-color:var(--sd-color-tabs-underline-inactive);cursor:pointer;font-size:var(--sd-fontsize-tabs-label);font-weight:700;padding:1em 1.25em .5em;transition:color 250ms;width:auto;z-index:1}html .sd-tab-set>label:hover{color:var(--sd-color-tabs-label-active)}.sd-col>.sd-tab-set{width:100%}.sd-tab-content{box-shadow:0 -0.0625rem var(--sd-color-tabs-overline),0 .0625rem var(--sd-color-tabs-underline);display:none;order:99;padding-bottom:.75rem;padding-top:.75rem;width:100%}.sd-tab-content>:first-child{margin-top:0 !important}.sd-tab-content>:last-child{margin-bottom:0 !important}.sd-tab-content>.sd-tab-set{margin:0}.sd-sphinx-override,.sd-sphinx-override *{-moz-box-sizing:border-box;-webkit-box-sizing:border-box;box-sizing:border-box}.sd-sphinx-override p{margin-top:0}:root{--sd-color-primary: #0071bc;--sd-color-secondary: #6c757d;--sd-color-success: #28a745;--sd-color-info: #17a2b8;--sd-color-warning: #f0b37e;--sd-color-danger: #dc3545;--sd-color-light: #f8f9fa;--sd-color-muted: #6c757d;--sd-color-dark: #212529;--sd-color-black: black;--sd-color-white: white;--sd-color-primary-highlight: #0060a0;--sd-color-secondary-highlight: #5c636a;--sd-color-success-highlight: #228e3b;--sd-color-info-highlight: #148a9c;--sd-color-warning-highlight: #cc986b;--sd-color-danger-highlight: #bb2d3b;--sd-color-light-highlight: #d3d4d5;--sd-color-muted-highlight: #5c636a;--sd-color-dark-highlight: #1c1f23;--sd-color-black-highlight: black;--sd-color-white-highlight: #d9d9d9;--sd-color-primary-bg: rgba(0, 113, 188, 0.2);--sd-color-secondary-bg: rgba(108, 117, 125, 0.2);--sd-color-success-bg: rgba(40, 167, 69, 0.2);--sd-color-info-bg: rgba(23, 162, 184, 0.2);--sd-color-warning-bg: rgba(240, 179, 126, 0.2);--sd-color-danger-bg: rgba(220, 53, 69, 0.2);--sd-color-light-bg: rgba(248, 249, 250, 0.2);--sd-color-muted-bg: rgba(108, 117, 125, 0.2);--sd-color-dark-bg: rgba(33, 37, 41, 0.2);--sd-color-black-bg: rgba(0, 0, 0, 0.2);--sd-color-white-bg: rgba(255, 255, 255, 0.2);--sd-color-primary-text: #fff;--sd-color-secondary-text: #fff;--sd-color-success-text: #fff;--sd-color-info-text: #fff;--sd-color-warning-text: #212529;--sd-color-danger-text: #fff;--sd-color-light-text: #212529;--sd-color-muted-text: #fff;--sd-color-dark-text: #fff;--sd-color-black-text: #fff;--sd-color-white-text: #212529;--sd-color-shadow: rgba(0, 0, 0, 0.15);--sd-color-card-border: rgba(0, 0, 0, 0.125);--sd-color-card-border-hover: hsla(231, 99%, 66%, 1);--sd-color-card-background: transparent;--sd-color-card-text: inherit;--sd-color-card-header: transparent;--sd-color-card-footer: transparent;--sd-color-tabs-label-active: hsla(231, 99%, 66%, 1);--sd-color-tabs-label-hover: hsla(231, 99%, 66%, 1);--sd-color-tabs-label-inactive: hsl(0, 0%, 66%);--sd-color-tabs-underline-active: hsla(231, 99%, 66%, 1);--sd-color-tabs-underline-hover: rgba(178, 206, 245, 0.62);--sd-color-tabs-underline-inactive: transparent;--sd-color-tabs-overline: rgb(222, 222, 222);--sd-color-tabs-underline: rgb(222, 222, 222);--sd-fontsize-tabs-label: 1rem;--sd-fontsize-dropdown: inherit;--sd-fontsize-dropdown-title: 1rem;--sd-fontweight-dropdown-title: 700}
diff --git a/_static/sphinx_highlight.js b/_static/sphinx_highlight.js
new file mode 100644
index 000000000..8a96c69a1
--- /dev/null
+++ b/_static/sphinx_highlight.js
@@ -0,0 +1,154 @@
+/* Highlighting utilities for Sphinx HTML documentation. */
+"use strict";
+
+const SPHINX_HIGHLIGHT_ENABLED = true
+
+/**
+ * highlight a given string on a node by wrapping it in
+ * span elements with the given class name.
+ */
+const _highlight = (node, addItems, text, className) => {
+  if (node.nodeType === Node.TEXT_NODE) {
+    const val = node.nodeValue;
+    const parent = node.parentNode;
+    const pos = val.toLowerCase().indexOf(text);
+    if (
+      pos >= 0 &&
+      !parent.classList.contains(className) &&
+      !parent.classList.contains("nohighlight")
+    ) {
+      let span;
+
+      const closestNode = parent.closest("body, svg, foreignObject");
+      const isInSVG = closestNode && closestNode.matches("svg");
+      if (isInSVG) {
+        span = document.createElementNS("http://www.w3.org/2000/svg", "tspan");
+      } else {
+        span = document.createElement("span");
+        span.classList.add(className);
+      }
+
+      span.appendChild(document.createTextNode(val.substr(pos, text.length)));
+      const rest = document.createTextNode(val.substr(pos + text.length));
+      parent.insertBefore(
+        span,
+        parent.insertBefore(
+          rest,
+          node.nextSibling
+        )
+      );
+      node.nodeValue = val.substr(0, pos);
+      /* There may be more occurrences of search term in this node. So call this
+       * function recursively on the remaining fragment.
+       */
+      _highlight(rest, addItems, text, className);
+
+      if (isInSVG) {
+        const rect = document.createElementNS(
+          "http://www.w3.org/2000/svg",
+          "rect"
+        );
+        const bbox = parent.getBBox();
+        rect.x.baseVal.value = bbox.x;
+        rect.y.baseVal.value = bbox.y;
+        rect.width.baseVal.value = bbox.width;
+        rect.height.baseVal.value = bbox.height;
+        rect.setAttribute("class", className);
+        addItems.push({ parent: parent, target: rect });
+      }
+    }
+  } else if (node.matches && !node.matches("button, select, textarea")) {
+    node.childNodes.forEach((el) => _highlight(el, addItems, text, className));
+  }
+};
+const _highlightText = (thisNode, text, className) => {
+  let addItems = [];
+  _highlight(thisNode, addItems, text, className);
+  addItems.forEach((obj) =>
+    obj.parent.insertAdjacentElement("beforebegin", obj.target)
+  );
+};
+
+/**
+ * Small JavaScript module for the documentation.
+ */
+const SphinxHighlight = {
+
+  /**
+   * highlight the search words provided in localstorage in the text
+   */
+  highlightSearchWords: () => {
+    if (!SPHINX_HIGHLIGHT_ENABLED) return;  // bail if no highlight
+
+    // get and clear terms from localstorage
+    const url = new URL(window.location);
+    const highlight =
+        localStorage.getItem("sphinx_highlight_terms")
+        || url.searchParams.get("highlight")
+        || "";
+    localStorage.removeItem("sphinx_highlight_terms")
+    url.searchParams.delete("highlight");
+    window.history.replaceState({}, "", url);
+
+    // get individual terms from highlight string
+    const terms = highlight.toLowerCase().split(/\s+/).filter(x => x);
+    if (terms.length === 0) return; // nothing to do
+
+    // There should never be more than one element matching "div.body"
+    const divBody = document.querySelectorAll("div.body");
+    const body = divBody.length ? divBody[0] : document.querySelector("body");
+    window.setTimeout(() => {
+      terms.forEach((term) => _highlightText(body, term, "highlighted"));
+    }, 10);
+
+    const searchBox = document.getElementById("searchbox");
+    if (searchBox === null) return;
+    searchBox.appendChild(
+      document
+        .createRange()
+        .createContextualFragment(
+          '<p class="highlight-link">' +
+            '<a href="javascript:SphinxHighlight.hideSearchWords()">' +
+            _("Hide Search Matches") +
+            "</a></p>"
+        )
+    );
+  },
+
+  /**
+   * helper function to hide the search marks again
+   */
+  hideSearchWords: () => {
+    document
+      .querySelectorAll("#searchbox .highlight-link")
+      .forEach((el) => el.remove());
+    document
+      .querySelectorAll("span.highlighted")
+      .forEach((el) => el.classList.remove("highlighted"));
+    localStorage.removeItem("sphinx_highlight_terms")
+  },
+
+  initEscapeListener: () => {
+    // only install a listener if it is really needed
+    if (!DOCUMENTATION_OPTIONS.ENABLE_SEARCH_SHORTCUTS) return;
+
+    document.addEventListener("keydown", (event) => {
+      // bail for input elements
+      if (BLACKLISTED_KEY_CONTROL_ELEMENTS.has(document.activeElement.tagName)) return;
+      // bail with special keys
+      if (event.shiftKey || event.altKey || event.ctrlKey || event.metaKey) return;
+      if (DOCUMENTATION_OPTIONS.ENABLE_SEARCH_SHORTCUTS && (event.key === "Escape")) {
+        SphinxHighlight.hideSearchWords();
+        event.preventDefault();
+      }
+    });
+  },
+};
+
+_ready(() => {
+  /* Do not call highlightSearchWords() when we are on the search page.
+   * It will highlight words from the *previous* search query.
+   */
+  if (typeof Search === "undefined") SphinxHighlight.highlightSearchWords();
+  SphinxHighlight.initEscapeListener();
+});
diff --git a/_static/styles/pydata-sphinx-theme.css b/_static/styles/pydata-sphinx-theme.css
new file mode 100644
index 000000000..d26dd09d8
--- /dev/null
+++ b/_static/styles/pydata-sphinx-theme.css
@@ -0,0 +1,32 @@
+@charset "UTF-8";
+/*!
+ * Bootstrap  v5.3.3 (https://getbootstrap.com/)
+ * Copyright 2011-2024 The Bootstrap Authors
+ * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)
+ */:root,[data-bs-theme=light]{--bs-blue:#0d6efd;--bs-indigo:#6610f2;--bs-purple:#6f42c1;--bs-pink:#d63384;--bs-red:#dc3545;--bs-orange:#fd7e14;--bs-yellow:#ffc107;--bs-green:#198754;--bs-teal:#20c997;--bs-cyan:#0dcaf0;--bs-black:#000;--bs-white:#fff;--bs-gray:#6c757d;--bs-gray-dark:#343a40;--bs-gray-100:#f8f9fa;--bs-gray-200:#e9ecef;--bs-gray-300:#dee2e6;--bs-gray-400:#ced4da;--bs-gray-500:#adb5bd;--bs-gray-600:#6c757d;--bs-gray-700:#495057;--bs-gray-800:#343a40;--bs-gray-900:#212529;--bs-primary:#0d6efd;--bs-secondary:#6c757d;--bs-success:#198754;--bs-info:#0dcaf0;--bs-warning:#ffc107;--bs-danger:#dc3545;--bs-light:#f8f9fa;--bs-dark:#212529;--bs-primary-rgb:13,110,253;--bs-secondary-rgb:108,117,125;--bs-success-rgb:25,135,84;--bs-info-rgb:13,202,240;--bs-warning-rgb:255,193,7;--bs-danger-rgb:220,53,69;--bs-light-rgb:248,249,250;--bs-dark-rgb:33,37,41;--bs-primary-text-emphasis:#052c65;--bs-secondary-text-emphasis:#2b2f32;--bs-success-text-emphasis:#0a3622;--bs-info-text-emphasis:#055160;--bs-warning-text-emphasis:#664d03;--bs-danger-text-emphasis:#58151c;--bs-light-text-emphasis:#495057;--bs-dark-text-emphasis:#495057;--bs-primary-bg-subtle:#cfe2ff;--bs-secondary-bg-subtle:#e2e3e5;--bs-success-bg-subtle:#d1e7dd;--bs-info-bg-subtle:#cff4fc;--bs-warning-bg-subtle:#fff3cd;--bs-danger-bg-subtle:#f8d7da;--bs-light-bg-subtle:#fcfcfd;--bs-dark-bg-subtle:#ced4da;--bs-primary-border-subtle:#9ec5fe;--bs-secondary-border-subtle:#c4c8cb;--bs-success-border-subtle:#a3cfbb;--bs-info-border-subtle:#9eeaf9;--bs-warning-border-subtle:#ffe69c;--bs-danger-border-subtle:#f1aeb5;--bs-light-border-subtle:#e9ecef;--bs-dark-border-subtle:#adb5bd;--bs-white-rgb:255,255,255;--bs-black-rgb:0,0,0;--bs-font-sans-serif:system-ui,-apple-system,"Segoe UI",Roboto,"Helvetica Neue","Noto Sans","Liberation Sans",Arial,sans-serif,"Apple Color Emoji","Segoe UI Emoji","Segoe UI Symbol","Noto Color Emoji";--bs-font-monospace:SFMono-Regular,Menlo,Monaco,Consolas,"Liberation Mono","Courier New",monospace;--bs-gradient:linear-gradient(180deg,hsla(0,0%,100%,.15),hsla(0,0%,100%,0));--bs-body-font-family:var(--bs-font-sans-serif);--bs-body-font-size:1rem;--bs-body-font-weight:400;--bs-body-line-height:1.5;--bs-body-color:#212529;--bs-body-color-rgb:33,37,41;--bs-body-bg:#fff;--bs-body-bg-rgb:255,255,255;--bs-emphasis-color:#000;--bs-emphasis-color-rgb:0,0,0;--bs-secondary-color:rgba(33,37,41,.75);--bs-secondary-color-rgb:33,37,41;--bs-secondary-bg:#e9ecef;--bs-secondary-bg-rgb:233,236,239;--bs-tertiary-color:rgba(33,37,41,.5);--bs-tertiary-color-rgb:33,37,41;--bs-tertiary-bg:#f8f9fa;--bs-tertiary-bg-rgb:248,249,250;--bs-heading-color:inherit;--bs-link-color:#0d6efd;--bs-link-color-rgb:13,110,253;--bs-link-decoration:underline;--bs-link-hover-color:#0a58ca;--bs-link-hover-color-rgb:10,88,202;--bs-code-color:#d63384;--bs-highlight-color:#212529;--bs-highlight-bg:#fff3cd;--bs-border-width:1px;--bs-border-style:solid;--bs-border-color:#dee2e6;--bs-border-color-translucent:rgba(0,0,0,.175);--bs-border-radius:0.375rem;--bs-border-radius-sm:0.25rem;--bs-border-radius-lg:0.5rem;--bs-border-radius-xl:1rem;--bs-border-radius-xxl:2rem;--bs-border-radius-2xl:var(--bs-border-radius-xxl);--bs-border-radius-pill:50rem;--bs-box-shadow:0 0.5rem 1rem rgba(0,0,0,.15);--bs-box-shadow-sm:0 0.125rem 0.25rem rgba(0,0,0,.075);--bs-box-shadow-lg:0 1rem 3rem rgba(0,0,0,.175);--bs-box-shadow-inset:inset 0 1px 2px rgba(0,0,0,.075);--bs-focus-ring-width:0.1875rem;--bs-focus-ring-opacity:1;--bs-focus-ring-color:var(--pst-color-accent);--bs-form-valid-color:#198754;--bs-form-valid-border-color:#198754;--bs-form-invalid-color:#dc3545;--bs-form-invalid-border-color:#dc3545}[data-bs-theme=dark]{--bs-body-color:#dee2e6;--bs-body-color-rgb:222,226,230;--bs-body-bg:#212529;--bs-body-bg-rgb:33,37,41;--bs-emphasis-color:#fff;--bs-emphasis-color-rgb:255,255,255;--bs-secondary-color:rgba(222,226,230,.75);--bs-secondary-color-rgb:222,226,230;--bs-secondary-bg:#343a40;--bs-secondary-bg-rgb:52,58,64;--bs-tertiary-color:rgba(222,226,230,.5);--bs-tertiary-color-rgb:222,226,230;--bs-tertiary-bg:#2b3035;--bs-tertiary-bg-rgb:43,48,53;--bs-primary-text-emphasis:#6ea8fe;--bs-secondary-text-emphasis:#a7acb1;--bs-success-text-emphasis:#75b798;--bs-info-text-emphasis:#6edff6;--bs-warning-text-emphasis:#ffda6a;--bs-danger-text-emphasis:#ea868f;--bs-light-text-emphasis:#f8f9fa;--bs-dark-text-emphasis:#dee2e6;--bs-primary-bg-subtle:#031633;--bs-secondary-bg-subtle:#161719;--bs-success-bg-subtle:#051b11;--bs-info-bg-subtle:#032830;--bs-warning-bg-subtle:#332701;--bs-danger-bg-subtle:#2c0b0e;--bs-light-bg-subtle:#343a40;--bs-dark-bg-subtle:#1a1d20;--bs-primary-border-subtle:#084298;--bs-secondary-border-subtle:#41464b;--bs-success-border-subtle:#0f5132;--bs-info-border-subtle:#087990;--bs-warning-border-subtle:#997404;--bs-danger-border-subtle:#842029;--bs-light-border-subtle:#495057;--bs-dark-border-subtle:#343a40;--bs-heading-color:inherit;--bs-link-color:#6ea8fe;--bs-link-hover-color:#8bb9fe;--bs-link-color-rgb:110,168,254;--bs-link-hover-color-rgb:139,185,254;--bs-code-color:#e685b5;--bs-highlight-color:#dee2e6;--bs-highlight-bg:#664d03;--bs-border-color:#495057;--bs-border-color-translucent:hsla(0,0%,100%,.15);--bs-form-valid-color:#75b798;--bs-form-valid-border-color:#75b798;--bs-form-invalid-color:#ea868f;--bs-form-invalid-border-color:#ea868f;color-scheme:dark}*,:after,:before{box-sizing:border-box}@media (prefers-reduced-motion:no-preference){:root{scroll-behavior:smooth}}body{-webkit-text-size-adjust:100%;-webkit-tap-highlight-color:rgba(0,0,0,0);background-color:var(--bs-body-bg);color:var(--bs-body-color);font-family:var(--bs-body-font-family);font-size:var(--bs-body-font-size);font-weight:var(--bs-body-font-weight);line-height:var(--bs-body-line-height);margin:0;text-align:var(--bs-body-text-align)}hr{border-top:var(--bs-border-width) solid;color:inherit;margin:1rem 0;opacity:.25}.h1,.h2,.h3,.h4,.h5,.h6,h1,h2,h3,h4,h5,h6{color:var(--bs-heading-color);font-weight:500;line-height:1.2;margin-bottom:.5rem;margin-top:0}.h1,h1{font-size:calc(1.375rem + 1.5vw)}@media (min-width:1200px){.h1,h1{font-size:2.5rem}}.h2,h2{font-size:calc(1.325rem + .9vw)}@media (min-width:1200px){.h2,h2{font-size:2rem}}.h3,h3{font-size:calc(1.3rem + .6vw)}@media (min-width:1200px){.h3,h3{font-size:1.75rem}}.h4,h4{font-size:calc(1.275rem + .3vw)}@media (min-width:1200px){.h4,h4{font-size:1.5rem}}.h5,h5{font-size:1.25rem}.h6,h6{font-size:1rem}p{margin-bottom:1rem;margin-top:0}abbr[title]{cursor:help;text-decoration:underline dotted;text-decoration-skip-ink:none}address{font-style:normal;line-height:inherit;margin-bottom:1rem}ol,ul{padding-left:2rem}dl,ol,ul{margin-bottom:1rem;margin-top:0}ol ol,ol ul,ul ol,ul ul{margin-bottom:0}dt{font-weight:700}dd{margin-bottom:.5rem;margin-left:0}blockquote{margin:0 0 1rem}b,strong{font-weight:bolder}.small,small{font-size:.875em}.mark,mark{background-color:var(--bs-highlight-bg);color:var(--bs-highlight-color);padding:.1875em}sub,sup{font-size:.75em;line-height:0;position:relative;vertical-align:baseline}sub{bottom:-.25em}sup{top:-.5em}a{color:rgba(var(--bs-link-color-rgb),var(--bs-link-opacity,1))}a:hover{--bs-link-color-rgb:var(--bs-link-hover-color-rgb)}a:not([href]):not([class]),a:not([href]):not([class]):hover{color:inherit;text-decoration:none}code,kbd,pre,samp{font-family:var(--bs-font-monospace);font-size:1em}pre{display:block;font-size:.875em;margin-bottom:1rem;margin-top:0;overflow:auto}pre code{color:inherit;font-size:inherit;word-break:normal}code{word-wrap:break-word;color:var(--bs-code-color);font-size:.875em}a>code{color:inherit}kbd{background-color:var(--bs-body-color);border-radius:.25rem;color:var(--bs-body-bg);font-size:.875em;padding:.1875rem .375rem}kbd kbd{font-size:1em;padding:0}figure{margin:0 0 1rem}img,svg{vertical-align:middle}table{border-collapse:collapse;caption-side:bottom}caption{color:var(--bs-secondary-color);padding-bottom:.5rem;padding-top:.5rem;text-align:left}th{text-align:inherit;text-align:-webkit-match-parent}tbody,td,tfoot,th,thead,tr{border:0 solid;border-color:inherit}label{display:inline-block}button{border-radius:0}button:focus:not(:focus-visible){outline:0}button,input,optgroup,select,textarea{font-family:inherit;font-size:inherit;line-height:inherit;margin:0}button,select{text-transform:none}[role=button]{cursor:pointer}select{word-wrap:normal}select:disabled{opacity:1}[list]:not([type=date]):not([type=datetime-local]):not([type=month]):not([type=week]):not([type=time])::-webkit-calendar-picker-indicator{display:none!important}[type=button],[type=reset],[type=submit],button{-webkit-appearance:button}[type=button]:not(:disabled),[type=reset]:not(:disabled),[type=submit]:not(:disabled),button:not(:disabled){cursor:pointer}::-moz-focus-inner{border-style:none;padding:0}textarea{resize:vertical}fieldset{border:0;margin:0;min-width:0;padding:0}legend{float:left;font-size:calc(1.275rem + .3vw);line-height:inherit;margin-bottom:.5rem;padding:0;width:100%}@media (min-width:1200px){legend{font-size:1.5rem}}legend+*{clear:left}::-webkit-datetime-edit-day-field,::-webkit-datetime-edit-fields-wrapper,::-webkit-datetime-edit-hour-field,::-webkit-datetime-edit-minute,::-webkit-datetime-edit-month-field,::-webkit-datetime-edit-text,::-webkit-datetime-edit-year-field{padding:0}::-webkit-inner-spin-button{height:auto}[type=search]{-webkit-appearance:textfield;outline-offset:-2px}::-webkit-search-decoration{-webkit-appearance:none}::-webkit-color-swatch-wrapper{padding:0}::file-selector-button{-webkit-appearance:button;font:inherit}output{display:inline-block}iframe{border:0}summary{cursor:pointer;display:list-item}progress{vertical-align:baseline}[hidden]{display:none!important}.lead{font-size:1.25rem;font-weight:300}.display-1{font-size:calc(1.625rem + 4.5vw);font-weight:300;line-height:1.2}@media (min-width:1200px){.display-1{font-size:5rem}}.display-2{font-size:calc(1.575rem + 3.9vw);font-weight:300;line-height:1.2}@media (min-width:1200px){.display-2{font-size:4.5rem}}.display-3{font-size:calc(1.525rem + 3.3vw);font-weight:300;line-height:1.2}@media (min-width:1200px){.display-3{font-size:4rem}}.display-4{font-size:calc(1.475rem + 2.7vw);font-weight:300;line-height:1.2}@media (min-width:1200px){.display-4{font-size:3.5rem}}.display-5{font-size:calc(1.425rem + 2.1vw);font-weight:300;line-height:1.2}@media (min-width:1200px){.display-5{font-size:3rem}}.display-6{font-size:calc(1.375rem + 1.5vw);font-weight:300;line-height:1.2}@media (min-width:1200px){.display-6{font-size:2.5rem}}.list-inline,.list-unstyled{list-style:none;padding-left:0}.list-inline-item{display:inline-block}.list-inline-item:not(:last-child){margin-right:.5rem}.initialism{font-size:.875em;text-transform:uppercase}.blockquote{font-size:1.25rem;margin-bottom:1rem}.blockquote>:last-child{margin-bottom:0}.blockquote-footer{color:#6c757d;font-size:.875em;margin-bottom:1rem;margin-top:-1rem}.blockquote-footer:before{content:"— "}.img-fluid,.img-thumbnail{height:auto;max-width:100%}.img-thumbnail{background-color:var(--bs-body-bg);border:var(--bs-border-width) solid var(--bs-border-color);border-radius:var(--bs-border-radius);padding:.25rem}.figure{display:inline-block}.figure-img{line-height:1;margin-bottom:.5rem}.figure-caption{color:var(--bs-secondary-color);font-size:.875em}.container,.container-fluid,.container-lg,.container-md,.container-sm,.container-xl{--bs-gutter-x:1.5rem;--bs-gutter-y:0;margin-left:auto;margin-right:auto;padding-left:calc(var(--bs-gutter-x)*.5);padding-right:calc(var(--bs-gutter-x)*.5);width:100%}@media (min-width:540px){.container,.container-sm{max-width:540px}}@media (min-width:720px){.container,.container-md,.container-sm{max-width:720px}}@media (min-width:960px){.container,.container-lg,.container-md,.container-sm{max-width:960px}}@media (min-width:1200px){.container,.container-lg,.container-md,.container-sm,.container-xl{max-width:1400px}}:root{--bs-breakpoint-xs:0;--bs-breakpoint-sm:540px;--bs-breakpoint-md:720px;--bs-breakpoint-lg:960px;--bs-breakpoint-xl:1200px}.row{--bs-gutter-x:1.5rem;--bs-gutter-y:0;display:flex;flex-wrap:wrap;margin-left:calc(var(--bs-gutter-x)*-.5);margin-right:calc(var(--bs-gutter-x)*-.5);margin-top:calc(var(--bs-gutter-y)*-1)}.row>*{flex-shrink:0;margin-top:var(--bs-gutter-y);max-width:100%;padding-left:calc(var(--bs-gutter-x)*.5);padding-right:calc(var(--bs-gutter-x)*.5);width:100%}.col{flex:1 0 0%}.row-cols-auto>*{flex:0 0 auto;width:auto}.row-cols-1>*{flex:0 0 auto;width:100%}.row-cols-2>*{flex:0 0 auto;width:50%}.row-cols-3>*{flex:0 0 auto;width:33.33333333%}.row-cols-4>*{flex:0 0 auto;width:25%}.row-cols-5>*{flex:0 0 auto;width:20%}.row-cols-6>*{flex:0 0 auto;width:16.66666667%}.col-auto{flex:0 0 auto;width:auto}.col-1{flex:0 0 auto;width:8.33333333%}.col-2{flex:0 0 auto;width:16.66666667%}.col-3{flex:0 0 auto;width:25%}.col-4{flex:0 0 auto;width:33.33333333%}.col-5{flex:0 0 auto;width:41.66666667%}.col-6{flex:0 0 auto;width:50%}.col-7{flex:0 0 auto;width:58.33333333%}.col-8{flex:0 0 auto;width:66.66666667%}.col-9{flex:0 0 auto;width:75%}.col-10{flex:0 0 auto;width:83.33333333%}.col-11{flex:0 0 auto;width:91.66666667%}.col-12{flex:0 0 auto;width:100%}.offset-1{margin-left:8.33333333%}.offset-2{margin-left:16.66666667%}.offset-3{margin-left:25%}.offset-4{margin-left:33.33333333%}.offset-5{margin-left:41.66666667%}.offset-6{margin-left:50%}.offset-7{margin-left:58.33333333%}.offset-8{margin-left:66.66666667%}.offset-9{margin-left:75%}.offset-10{margin-left:83.33333333%}.offset-11{margin-left:91.66666667%}.g-0,.gx-0{--bs-gutter-x:0}.g-0,.gy-0{--bs-gutter-y:0}.g-1,.gx-1{--bs-gutter-x:0.25rem}.g-1,.gy-1{--bs-gutter-y:0.25rem}.g-2,.gx-2{--bs-gutter-x:0.5rem}.g-2,.gy-2{--bs-gutter-y:0.5rem}.g-3,.gx-3{--bs-gutter-x:1rem}.g-3,.gy-3{--bs-gutter-y:1rem}.g-4,.gx-4{--bs-gutter-x:1.5rem}.g-4,.gy-4{--bs-gutter-y:1.5rem}.g-5,.gx-5{--bs-gutter-x:3rem}.g-5,.gy-5{--bs-gutter-y:3rem}@media (min-width:540px){.col-sm{flex:1 0 0%}.row-cols-sm-auto>*{flex:0 0 auto;width:auto}.row-cols-sm-1>*{flex:0 0 auto;width:100%}.row-cols-sm-2>*{flex:0 0 auto;width:50%}.row-cols-sm-3>*{flex:0 0 auto;width:33.33333333%}.row-cols-sm-4>*{flex:0 0 auto;width:25%}.row-cols-sm-5>*{flex:0 0 auto;width:20%}.row-cols-sm-6>*{flex:0 0 auto;width:16.66666667%}.col-sm-auto{flex:0 0 auto;width:auto}.col-sm-1{flex:0 0 auto;width:8.33333333%}.col-sm-2{flex:0 0 auto;width:16.66666667%}.col-sm-3{flex:0 0 auto;width:25%}.col-sm-4{flex:0 0 auto;width:33.33333333%}.col-sm-5{flex:0 0 auto;width:41.66666667%}.col-sm-6{flex:0 0 auto;width:50%}.col-sm-7{flex:0 0 auto;width:58.33333333%}.col-sm-8{flex:0 0 auto;width:66.66666667%}.col-sm-9{flex:0 0 auto;width:75%}.col-sm-10{flex:0 0 auto;width:83.33333333%}.col-sm-11{flex:0 0 auto;width:91.66666667%}.col-sm-12{flex:0 0 auto;width:100%}.offset-sm-0{margin-left:0}.offset-sm-1{margin-left:8.33333333%}.offset-sm-2{margin-left:16.66666667%}.offset-sm-3{margin-left:25%}.offset-sm-4{margin-left:33.33333333%}.offset-sm-5{margin-left:41.66666667%}.offset-sm-6{margin-left:50%}.offset-sm-7{margin-left:58.33333333%}.offset-sm-8{margin-left:66.66666667%}.offset-sm-9{margin-left:75%}.offset-sm-10{margin-left:83.33333333%}.offset-sm-11{margin-left:91.66666667%}.g-sm-0,.gx-sm-0{--bs-gutter-x:0}.g-sm-0,.gy-sm-0{--bs-gutter-y:0}.g-sm-1,.gx-sm-1{--bs-gutter-x:0.25rem}.g-sm-1,.gy-sm-1{--bs-gutter-y:0.25rem}.g-sm-2,.gx-sm-2{--bs-gutter-x:0.5rem}.g-sm-2,.gy-sm-2{--bs-gutter-y:0.5rem}.g-sm-3,.gx-sm-3{--bs-gutter-x:1rem}.g-sm-3,.gy-sm-3{--bs-gutter-y:1rem}.g-sm-4,.gx-sm-4{--bs-gutter-x:1.5rem}.g-sm-4,.gy-sm-4{--bs-gutter-y:1.5rem}.g-sm-5,.gx-sm-5{--bs-gutter-x:3rem}.g-sm-5,.gy-sm-5{--bs-gutter-y:3rem}}@media (min-width:720px){.col-md{flex:1 0 0%}.row-cols-md-auto>*{flex:0 0 auto;width:auto}.row-cols-md-1>*{flex:0 0 auto;width:100%}.row-cols-md-2>*{flex:0 0 auto;width:50%}.row-cols-md-3>*{flex:0 0 auto;width:33.33333333%}.row-cols-md-4>*{flex:0 0 auto;width:25%}.row-cols-md-5>*{flex:0 0 auto;width:20%}.row-cols-md-6>*{flex:0 0 auto;width:16.66666667%}.col-md-auto{flex:0 0 auto;width:auto}.col-md-1{flex:0 0 auto;width:8.33333333%}.col-md-2{flex:0 0 auto;width:16.66666667%}.col-md-3{flex:0 0 auto;width:25%}.col-md-4{flex:0 0 auto;width:33.33333333%}.col-md-5{flex:0 0 auto;width:41.66666667%}.col-md-6{flex:0 0 auto;width:50%}.col-md-7{flex:0 0 auto;width:58.33333333%}.col-md-8{flex:0 0 auto;width:66.66666667%}.col-md-9{flex:0 0 auto;width:75%}.col-md-10{flex:0 0 auto;width:83.33333333%}.col-md-11{flex:0 0 auto;width:91.66666667%}.col-md-12{flex:0 0 auto;width:100%}.offset-md-0{margin-left:0}.offset-md-1{margin-left:8.33333333%}.offset-md-2{margin-left:16.66666667%}.offset-md-3{margin-left:25%}.offset-md-4{margin-left:33.33333333%}.offset-md-5{margin-left:41.66666667%}.offset-md-6{margin-left:50%}.offset-md-7{margin-left:58.33333333%}.offset-md-8{margin-left:66.66666667%}.offset-md-9{margin-left:75%}.offset-md-10{margin-left:83.33333333%}.offset-md-11{margin-left:91.66666667%}.g-md-0,.gx-md-0{--bs-gutter-x:0}.g-md-0,.gy-md-0{--bs-gutter-y:0}.g-md-1,.gx-md-1{--bs-gutter-x:0.25rem}.g-md-1,.gy-md-1{--bs-gutter-y:0.25rem}.g-md-2,.gx-md-2{--bs-gutter-x:0.5rem}.g-md-2,.gy-md-2{--bs-gutter-y:0.5rem}.g-md-3,.gx-md-3{--bs-gutter-x:1rem}.g-md-3,.gy-md-3{--bs-gutter-y:1rem}.g-md-4,.gx-md-4{--bs-gutter-x:1.5rem}.g-md-4,.gy-md-4{--bs-gutter-y:1.5rem}.g-md-5,.gx-md-5{--bs-gutter-x:3rem}.g-md-5,.gy-md-5{--bs-gutter-y:3rem}}@media (min-width:960px){.col-lg{flex:1 0 0%}.row-cols-lg-auto>*{flex:0 0 auto;width:auto}.row-cols-lg-1>*{flex:0 0 auto;width:100%}.row-cols-lg-2>*{flex:0 0 auto;width:50%}.row-cols-lg-3>*{flex:0 0 auto;width:33.33333333%}.row-cols-lg-4>*{flex:0 0 auto;width:25%}.row-cols-lg-5>*{flex:0 0 auto;width:20%}.row-cols-lg-6>*{flex:0 0 auto;width:16.66666667%}.col-lg-auto{flex:0 0 auto;width:auto}.col-lg-1{flex:0 0 auto;width:8.33333333%}.col-lg-2{flex:0 0 auto;width:16.66666667%}.col-lg-3{flex:0 0 auto;width:25%}.col-lg-4{flex:0 0 auto;width:33.33333333%}.col-lg-5{flex:0 0 auto;width:41.66666667%}.col-lg-6{flex:0 0 auto;width:50%}.col-lg-7{flex:0 0 auto;width:58.33333333%}.col-lg-8{flex:0 0 auto;width:66.66666667%}.col-lg-9{flex:0 0 auto;width:75%}.col-lg-10{flex:0 0 auto;width:83.33333333%}.col-lg-11{flex:0 0 auto;width:91.66666667%}.col-lg-12{flex:0 0 auto;width:100%}.offset-lg-0{margin-left:0}.offset-lg-1{margin-left:8.33333333%}.offset-lg-2{margin-left:16.66666667%}.offset-lg-3{margin-left:25%}.offset-lg-4{margin-left:33.33333333%}.offset-lg-5{margin-left:41.66666667%}.offset-lg-6{margin-left:50%}.offset-lg-7{margin-left:58.33333333%}.offset-lg-8{margin-left:66.66666667%}.offset-lg-9{margin-left:75%}.offset-lg-10{margin-left:83.33333333%}.offset-lg-11{margin-left:91.66666667%}.g-lg-0,.gx-lg-0{--bs-gutter-x:0}.g-lg-0,.gy-lg-0{--bs-gutter-y:0}.g-lg-1,.gx-lg-1{--bs-gutter-x:0.25rem}.g-lg-1,.gy-lg-1{--bs-gutter-y:0.25rem}.g-lg-2,.gx-lg-2{--bs-gutter-x:0.5rem}.g-lg-2,.gy-lg-2{--bs-gutter-y:0.5rem}.g-lg-3,.gx-lg-3{--bs-gutter-x:1rem}.g-lg-3,.gy-lg-3{--bs-gutter-y:1rem}.g-lg-4,.gx-lg-4{--bs-gutter-x:1.5rem}.g-lg-4,.gy-lg-4{--bs-gutter-y:1.5rem}.g-lg-5,.gx-lg-5{--bs-gutter-x:3rem}.g-lg-5,.gy-lg-5{--bs-gutter-y:3rem}}@media (min-width:1200px){.col-xl{flex:1 0 0%}.row-cols-xl-auto>*{flex:0 0 auto;width:auto}.row-cols-xl-1>*{flex:0 0 auto;width:100%}.row-cols-xl-2>*{flex:0 0 auto;width:50%}.row-cols-xl-3>*{flex:0 0 auto;width:33.33333333%}.row-cols-xl-4>*{flex:0 0 auto;width:25%}.row-cols-xl-5>*{flex:0 0 auto;width:20%}.row-cols-xl-6>*{flex:0 0 auto;width:16.66666667%}.col-xl-auto{flex:0 0 auto;width:auto}.col-xl-1{flex:0 0 auto;width:8.33333333%}.col-xl-2{flex:0 0 auto;width:16.66666667%}.col-xl-3{flex:0 0 auto;width:25%}.col-xl-4{flex:0 0 auto;width:33.33333333%}.col-xl-5{flex:0 0 auto;width:41.66666667%}.col-xl-6{flex:0 0 auto;width:50%}.col-xl-7{flex:0 0 auto;width:58.33333333%}.col-xl-8{flex:0 0 auto;width:66.66666667%}.col-xl-9{flex:0 0 auto;width:75%}.col-xl-10{flex:0 0 auto;width:83.33333333%}.col-xl-11{flex:0 0 auto;width:91.66666667%}.col-xl-12{flex:0 0 auto;width:100%}.offset-xl-0{margin-left:0}.offset-xl-1{margin-left:8.33333333%}.offset-xl-2{margin-left:16.66666667%}.offset-xl-3{margin-left:25%}.offset-xl-4{margin-left:33.33333333%}.offset-xl-5{margin-left:41.66666667%}.offset-xl-6{margin-left:50%}.offset-xl-7{margin-left:58.33333333%}.offset-xl-8{margin-left:66.66666667%}.offset-xl-9{margin-left:75%}.offset-xl-10{margin-left:83.33333333%}.offset-xl-11{margin-left:91.66666667%}.g-xl-0,.gx-xl-0{--bs-gutter-x:0}.g-xl-0,.gy-xl-0{--bs-gutter-y:0}.g-xl-1,.gx-xl-1{--bs-gutter-x:0.25rem}.g-xl-1,.gy-xl-1{--bs-gutter-y:0.25rem}.g-xl-2,.gx-xl-2{--bs-gutter-x:0.5rem}.g-xl-2,.gy-xl-2{--bs-gutter-y:0.5rem}.g-xl-3,.gx-xl-3{--bs-gutter-x:1rem}.g-xl-3,.gy-xl-3{--bs-gutter-y:1rem}.g-xl-4,.gx-xl-4{--bs-gutter-x:1.5rem}.g-xl-4,.gy-xl-4{--bs-gutter-y:1.5rem}.g-xl-5,.gx-xl-5{--bs-gutter-x:3rem}.g-xl-5,.gy-xl-5{--bs-gutter-y:3rem}}.table{--bs-table-color-type:initial;--bs-table-bg-type:initial;--bs-table-color-state:initial;--bs-table-bg-state:initial;--bs-table-color:var(--bs-emphasis-color);--bs-table-bg:var(--bs-body-bg);--bs-table-border-color:var(--bs-border-color);--bs-table-accent-bg:transparent;--bs-table-striped-color:var(--bs-emphasis-color);--bs-table-striped-bg:rgba(var(--bs-emphasis-color-rgb),0.05);--bs-table-active-color:var(--bs-emphasis-color);--bs-table-active-bg:rgba(var(--bs-emphasis-color-rgb),0.1);--bs-table-hover-color:var(--bs-emphasis-color);--bs-table-hover-bg:rgba(var(--bs-emphasis-color-rgb),0.075);border-color:var(--bs-table-border-color);margin-bottom:1rem;vertical-align:top;width:100%}.table>:not(caption)>*>*{background-color:var(--bs-table-bg);border-bottom-width:var(--bs-border-width);box-shadow:inset 0 0 0 9999px var(--bs-table-bg-state,var(--bs-table-bg-type,var(--bs-table-accent-bg)));color:var(--bs-table-color-state,var(--bs-table-color-type,var(--bs-table-color)));padding:.5rem}.table>tbody{vertical-align:inherit}.table>thead{vertical-align:bottom}.table-group-divider{border-top:calc(var(--bs-border-width)*2) solid}.caption-top{caption-side:top}.table-sm>:not(caption)>*>*{padding:.25rem}.table-bordered>:not(caption)>*{border-width:var(--bs-border-width) 0}.table-bordered>:not(caption)>*>*{border-width:0 var(--bs-border-width)}.table-borderless>:not(caption)>*>*{border-bottom-width:0}.table-borderless>:not(:first-child){border-top-width:0}.table-striped-columns>:not(caption)>tr>:nth-child(2n),.table-striped>tbody>tr:nth-of-type(odd)>*{--bs-table-color-type:var(--bs-table-striped-color);--bs-table-bg-type:var(--bs-table-striped-bg)}.table-active{--bs-table-color-state:var(--bs-table-active-color);--bs-table-bg-state:var(--bs-table-active-bg)}.table-hover>tbody>tr:hover>*{--bs-table-color-state:var(--bs-table-hover-color);--bs-table-bg-state:var(--bs-table-hover-bg)}.table-primary{--bs-table-color:#000;--bs-table-bg:#cfe2ff;--bs-table-border-color:#a6b5cc;--bs-table-striped-bg:#c5d7f2;--bs-table-striped-color:#000;--bs-table-active-bg:#bacbe6;--bs-table-active-color:#000;--bs-table-hover-bg:#bfd1ec;--bs-table-hover-color:#000}.table-primary,.table-secondary{border-color:var(--bs-table-border-color);color:var(--bs-table-color)}.table-secondary{--bs-table-color:#000;--bs-table-bg:#e2e3e5;--bs-table-border-color:#b5b6b7;--bs-table-striped-bg:#d7d8da;--bs-table-striped-color:#000;--bs-table-active-bg:#cbccce;--bs-table-active-color:#000;--bs-table-hover-bg:#d1d2d4;--bs-table-hover-color:#000}.table-success{--bs-table-color:#000;--bs-table-bg:#d1e7dd;--bs-table-border-color:#a7b9b1;--bs-table-striped-bg:#c7dbd2;--bs-table-striped-color:#000;--bs-table-active-bg:#bcd0c7;--bs-table-active-color:#000;--bs-table-hover-bg:#c1d6cc;--bs-table-hover-color:#000}.table-info,.table-success{border-color:var(--bs-table-border-color);color:var(--bs-table-color)}.table-info{--bs-table-color:#000;--bs-table-bg:#cff4fc;--bs-table-border-color:#a6c3ca;--bs-table-striped-bg:#c5e8ef;--bs-table-striped-color:#000;--bs-table-active-bg:#badce3;--bs-table-active-color:#000;--bs-table-hover-bg:#bfe2e9;--bs-table-hover-color:#000}.table-warning{--bs-table-color:#000;--bs-table-bg:#fff3cd;--bs-table-border-color:#ccc2a4;--bs-table-striped-bg:#f2e7c3;--bs-table-striped-color:#000;--bs-table-active-bg:#e6dbb9;--bs-table-active-color:#000;--bs-table-hover-bg:#ece1be;--bs-table-hover-color:#000}.table-danger,.table-warning{border-color:var(--bs-table-border-color);color:var(--bs-table-color)}.table-danger{--bs-table-color:#000;--bs-table-bg:#f8d7da;--bs-table-border-color:#c6acae;--bs-table-striped-bg:#eccccf;--bs-table-striped-color:#000;--bs-table-active-bg:#dfc2c4;--bs-table-active-color:#000;--bs-table-hover-bg:#e5c7ca;--bs-table-hover-color:#000}.table-light{--bs-table-color:#000;--bs-table-bg:#f8f9fa;--bs-table-border-color:#c6c7c8;--bs-table-striped-bg:#ecedee;--bs-table-striped-color:#000;--bs-table-active-bg:#dfe0e1;--bs-table-active-color:#000;--bs-table-hover-bg:#e5e6e7;--bs-table-hover-color:#000}.table-dark,.table-light{border-color:var(--bs-table-border-color);color:var(--bs-table-color)}.table-dark{--bs-table-color:#fff;--bs-table-bg:#212529;--bs-table-border-color:#4d5154;--bs-table-striped-bg:#2c3034;--bs-table-striped-color:#fff;--bs-table-active-bg:#373b3e;--bs-table-active-color:#fff;--bs-table-hover-bg:#323539;--bs-table-hover-color:#fff}.table-responsive{-webkit-overflow-scrolling:touch;overflow-x:auto}@media (max-width:539.98px){.table-responsive-sm{-webkit-overflow-scrolling:touch;overflow-x:auto}}@media (max-width:719.98px){.table-responsive-md{-webkit-overflow-scrolling:touch;overflow-x:auto}}@media (max-width:959.98px){.table-responsive-lg{-webkit-overflow-scrolling:touch;overflow-x:auto}}@media (max-width:1199.98px){.table-responsive-xl{-webkit-overflow-scrolling:touch;overflow-x:auto}}.form-label{margin-bottom:.5rem}.col-form-label{font-size:inherit;line-height:1.5;margin-bottom:0;padding-bottom:calc(.375rem + var(--bs-border-width));padding-top:calc(.375rem + var(--bs-border-width))}.col-form-label-lg{font-size:1.25rem;padding-bottom:calc(.5rem + var(--bs-border-width));padding-top:calc(.5rem + var(--bs-border-width))}.col-form-label-sm{font-size:.875rem;padding-bottom:calc(.25rem + var(--bs-border-width));padding-top:calc(.25rem + var(--bs-border-width))}.form-text{color:var(--bs-secondary-color);font-size:.875em;margin-top:.25rem}.form-control{appearance:none;background-clip:padding-box;background-color:var(--bs-body-bg);border:var(--bs-border-width) solid var(--bs-border-color);border-radius:var(--bs-border-radius);color:var(--bs-body-color);display:block;font-size:1rem;font-weight:400;line-height:1.5;padding:.375rem .75rem;transition:border-color .15s ease-in-out,box-shadow .15s ease-in-out;width:100%}@media (prefers-reduced-motion:reduce){.form-control{transition:none}}.form-control[type=file]{overflow:hidden}.form-control[type=file]:not(:disabled):not([readonly]){cursor:pointer}.form-control:focus{background-color:var(--bs-body-bg);border-color:#86b7fe;box-shadow:0 0 0 .1875rem var(--pst-color-accent);color:var(--bs-body-color);outline:0}.form-control::-webkit-date-and-time-value{height:1.5em;margin:0;min-width:85px}.form-control::-webkit-datetime-edit{display:block;padding:0}.form-control::placeholder{color:var(--bs-secondary-color);opacity:1}.form-control:disabled{background-color:var(--bs-secondary-bg);opacity:1}.form-control::file-selector-button{background-color:var(--bs-tertiary-bg);border:0 solid;border-color:inherit;border-inline-end-width:var(--bs-border-width);border-radius:0;color:var(--bs-body-color);margin:-.375rem -.75rem;margin-inline-end:.75rem;padding:.375rem .75rem;pointer-events:none;transition:color .15s ease-in-out,background-color .15s ease-in-out,border-color .15s ease-in-out,box-shadow .15s ease-in-out}@media (prefers-reduced-motion:reduce){.form-control::file-selector-button{transition:none}}.form-control:hover:not(:disabled):not([readonly])::file-selector-button{background-color:var(--bs-secondary-bg)}.form-control-plaintext{background-color:transparent;border:solid transparent;border-width:var(--bs-border-width) 0;color:var(--bs-body-color);display:block;line-height:1.5;margin-bottom:0;padding:.375rem 0;width:100%}.form-control-plaintext:focus{outline:0}.form-control-plaintext.form-control-lg,.form-control-plaintext.form-control-sm{padding-left:0;padding-right:0}.form-control-sm{border-radius:var(--bs-border-radius-sm);font-size:.875rem;min-height:calc(1.5em + .5rem + var(--bs-border-width)*2);padding:.25rem .5rem}.form-control-sm::file-selector-button{margin:-.25rem -.5rem;margin-inline-end:.5rem;padding:.25rem .5rem}.form-control-lg{border-radius:var(--bs-border-radius-lg);font-size:1.25rem;min-height:calc(1.5em + 1rem + var(--bs-border-width)*2);padding:.5rem 1rem}.form-control-lg::file-selector-button{margin:-.5rem -1rem;margin-inline-end:1rem;padding:.5rem 1rem}textarea.form-control{min-height:calc(1.5em + .75rem + var(--bs-border-width)*2)}textarea.form-control-sm{min-height:calc(1.5em + .5rem + var(--bs-border-width)*2)}textarea.form-control-lg{min-height:calc(1.5em + 1rem + var(--bs-border-width)*2)}.form-control-color{height:calc(1.5em + .75rem + var(--bs-border-width)*2);padding:.375rem;width:3rem}.form-control-color:not(:disabled):not([readonly]){cursor:pointer}.form-control-color::-moz-color-swatch{border:0!important;border-radius:var(--bs-border-radius)}.form-control-color::-webkit-color-swatch{border:0!important;border-radius:var(--bs-border-radius)}.form-control-color.form-control-sm{height:calc(1.5em + .5rem + var(--bs-border-width)*2)}.form-control-color.form-control-lg{height:calc(1.5em + 1rem + var(--bs-border-width)*2)}.form-select{--bs-form-select-bg-img:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16'%3E%3Cpath fill='none' stroke='%23343a40' stroke-linecap='round' stroke-linejoin='round' stroke-width='2' d='m2 5 6 6 6-6'/%3E%3C/svg%3E");appearance:none;background-color:var(--bs-body-bg);background-image:var(--bs-form-select-bg-img),var(--bs-form-select-bg-icon,none);background-position:right .75rem center;background-repeat:no-repeat;background-size:16px 12px;border:var(--bs-border-width) solid var(--bs-border-color);border-radius:var(--bs-border-radius);color:var(--bs-body-color);display:block;font-size:1rem;font-weight:400;line-height:1.5;padding:.375rem 2.25rem .375rem .75rem;transition:border-color .15s ease-in-out,box-shadow .15s ease-in-out;width:100%}@media (prefers-reduced-motion:reduce){.form-select{transition:none}}.form-select:focus{border-color:#86b7fe;box-shadow:0 0 0 .1875rem var(--pst-color-accent);outline:0}.form-select[multiple],.form-select[size]:not([size="1"]){background-image:none;padding-right:.75rem}.form-select:disabled{background-color:var(--bs-secondary-bg)}.form-select:-moz-focusring{color:transparent;text-shadow:0 0 0 var(--bs-body-color)}.form-select-sm{border-radius:var(--bs-border-radius-sm);font-size:.875rem;padding-bottom:.25rem;padding-left:.5rem;padding-top:.25rem}.form-select-lg{border-radius:var(--bs-border-radius-lg);font-size:1.25rem;padding-bottom:.5rem;padding-left:1rem;padding-top:.5rem}[data-bs-theme=dark] .form-select{--bs-form-select-bg-img:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16'%3E%3Cpath fill='none' stroke='%23dee2e6' stroke-linecap='round' stroke-linejoin='round' stroke-width='2' d='m2 5 6 6 6-6'/%3E%3C/svg%3E")}.form-check{display:block;margin-bottom:.125rem;min-height:1.5rem;padding-left:1.5em}.form-check .form-check-input{float:left;margin-left:-1.5em}.form-check-reverse{padding-left:0;padding-right:1.5em;text-align:right}.form-check-reverse .form-check-input{float:right;margin-left:0;margin-right:-1.5em}.form-check-input{--bs-form-check-bg:var(--bs-body-bg);appearance:none;background-color:var(--bs-form-check-bg);background-image:var(--bs-form-check-bg-image);background-position:50%;background-repeat:no-repeat;background-size:contain;border:var(--bs-border-width) solid var(--bs-border-color);flex-shrink:0;height:1em;margin-top:.25em;print-color-adjust:exact;vertical-align:top;width:1em}.form-check-input[type=checkbox]{border-radius:.25em}.form-check-input[type=radio]{border-radius:50%}.form-check-input:active{filter:brightness(90%)}.form-check-input:focus{border-color:#86b7fe;box-shadow:0 0 0 .1875rem var(--pst-color-accent);outline:0}.form-check-input:checked{background-color:#0d6efd;border-color:#0d6efd}.form-check-input:checked[type=checkbox]{--bs-form-check-bg-image:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 20 20'%3E%3Cpath fill='none' stroke='%23fff' stroke-linecap='round' stroke-linejoin='round' stroke-width='3' d='m6 10 3 3 6-6'/%3E%3C/svg%3E")}.form-check-input:checked[type=radio]{--bs-form-check-bg-image:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'%3E%3Ccircle r='2' fill='%23fff'/%3E%3C/svg%3E")}.form-check-input[type=checkbox]:indeterminate{--bs-form-check-bg-image:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 20 20'%3E%3Cpath fill='none' stroke='%23fff' stroke-linecap='round' stroke-linejoin='round' stroke-width='3' d='M6 10h8'/%3E%3C/svg%3E");background-color:#0d6efd;border-color:#0d6efd}.form-check-input:disabled{filter:none;opacity:.5;pointer-events:none}.form-check-input:disabled~.form-check-label,.form-check-input[disabled]~.form-check-label{cursor:default;opacity:.5}.form-switch{padding-left:2.5em}.form-switch .form-check-input{--bs-form-switch-bg:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'%3E%3Ccircle r='3' fill='rgba(0, 0, 0, 0.25)'/%3E%3C/svg%3E");background-image:var(--bs-form-switch-bg);background-position:0;border-radius:2em;margin-left:-2.5em;transition:background-position .15s ease-in-out;width:2em}@media (prefers-reduced-motion:reduce){.form-switch .form-check-input{transition:none}}.form-switch .form-check-input:focus{--bs-form-switch-bg:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'%3E%3Ccircle r='3' fill='%2386b7fe'/%3E%3C/svg%3E")}.form-switch .form-check-input:checked{--bs-form-switch-bg:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'%3E%3Ccircle r='3' fill='%23fff'/%3E%3C/svg%3E");background-position:100%}.form-switch.form-check-reverse{padding-left:0;padding-right:2.5em}.form-switch.form-check-reverse .form-check-input{margin-left:0;margin-right:-2.5em}.form-check-inline{display:inline-block;margin-right:1rem}.btn-check{clip:rect(0,0,0,0);pointer-events:none;position:absolute}.btn-check:disabled+.btn,.btn-check[disabled]+.btn{filter:none;opacity:.65;pointer-events:none}[data-bs-theme=dark] .form-switch .form-check-input:not(:checked):not(:focus){--bs-form-switch-bg:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'%3E%3Ccircle r='3' fill='rgba(255, 255, 255, 0.25)'/%3E%3C/svg%3E")}.form-range{appearance:none;background-color:transparent;height:1.375rem;padding:0;width:100%}.form-range:focus{outline:0}.form-range:focus::-webkit-slider-thumb{box-shadow:0 0 0 1px #fff,0 0 0 .1875rem var(--pst-color-accent)}.form-range:focus::-moz-range-thumb{box-shadow:0 0 0 1px #fff,0 0 0 .1875rem var(--pst-color-accent)}.form-range::-moz-focus-outer{border:0}.form-range::-webkit-slider-thumb{appearance:none;background-color:#0d6efd;border:0;border-radius:1rem;height:1rem;margin-top:-.25rem;transition:background-color .15s ease-in-out,border-color .15s ease-in-out,box-shadow .15s ease-in-out;width:1rem}@media (prefers-reduced-motion:reduce){.form-range::-webkit-slider-thumb{transition:none}}.form-range::-webkit-slider-thumb:active{background-color:#b6d4fe}.form-range::-webkit-slider-runnable-track{background-color:var(--bs-secondary-bg);border-color:transparent;border-radius:1rem;color:transparent;cursor:pointer;height:.5rem;width:100%}.form-range::-moz-range-thumb{appearance:none;background-color:#0d6efd;border:0;border-radius:1rem;height:1rem;transition:background-color .15s ease-in-out,border-color .15s ease-in-out,box-shadow .15s ease-in-out;width:1rem}@media (prefers-reduced-motion:reduce){.form-range::-moz-range-thumb{transition:none}}.form-range::-moz-range-thumb:active{background-color:#b6d4fe}.form-range::-moz-range-track{background-color:var(--bs-secondary-bg);border-color:transparent;border-radius:1rem;color:transparent;cursor:pointer;height:.5rem;width:100%}.form-range:disabled{pointer-events:none}.form-range:disabled::-webkit-slider-thumb{background-color:var(--bs-secondary-color)}.form-range:disabled::-moz-range-thumb{background-color:var(--bs-secondary-color)}.form-floating{position:relative}.form-floating>.form-control,.form-floating>.form-control-plaintext,.form-floating>.form-select{height:calc(3.5rem + var(--bs-border-width)*2);line-height:1.25;min-height:calc(3.5rem + var(--bs-border-width)*2)}.form-floating>label{border:var(--bs-border-width) solid transparent;height:100%;left:0;overflow:hidden;padding:1rem .75rem;pointer-events:none;position:absolute;text-align:start;text-overflow:ellipsis;top:0;transform-origin:0 0;transition:opacity .1s ease-in-out,transform .1s ease-in-out;white-space:nowrap;z-index:2}@media (prefers-reduced-motion:reduce){.form-floating>label{transition:none}}.form-floating>.form-control,.form-floating>.form-control-plaintext{padding:1rem .75rem}.form-floating>.form-control-plaintext::placeholder,.form-floating>.form-control::placeholder{color:transparent}.form-floating>.form-control-plaintext:focus,.form-floating>.form-control-plaintext:not(:placeholder-shown),.form-floating>.form-control:focus,.form-floating>.form-control:not(:placeholder-shown){padding-bottom:.625rem;padding-top:1.625rem}.form-floating>.form-control-plaintext:-webkit-autofill,.form-floating>.form-control:-webkit-autofill{padding-bottom:.625rem;padding-top:1.625rem}.form-floating>.form-select{padding-bottom:.625rem;padding-top:1.625rem}.form-floating>.form-control-plaintext~label,.form-floating>.form-control:focus~label,.form-floating>.form-control:not(:placeholder-shown)~label,.form-floating>.form-select~label{color:rgba(var(--bs-body-color-rgb),.65);transform:scale(.85) translateY(-.5rem) translateX(.15rem)}.form-floating>.form-control-plaintext~label:after,.form-floating>.form-control:focus~label:after,.form-floating>.form-control:not(:placeholder-shown)~label:after,.form-floating>.form-select~label:after{background-color:var(--bs-body-bg);border-radius:var(--bs-border-radius);content:"";height:1.5em;inset:1rem .375rem;position:absolute;z-index:-1}.form-floating>.form-control:-webkit-autofill~label{color:rgba(var(--bs-body-color-rgb),.65);transform:scale(.85) translateY(-.5rem) translateX(.15rem)}.form-floating>.form-control-plaintext~label{border-width:var(--bs-border-width) 0}.form-floating>.form-control:disabled~label,.form-floating>:disabled~label{color:#6c757d}.form-floating>.form-control:disabled~label:after,.form-floating>:disabled~label:after{background-color:var(--bs-secondary-bg)}.input-group{align-items:stretch;display:flex;flex-wrap:wrap;position:relative;width:100%}.input-group>.form-control,.input-group>.form-floating,.input-group>.form-select{flex:1 1 auto;min-width:0;position:relative;width:1%}.input-group>.form-control:focus,.input-group>.form-floating:focus-within,.input-group>.form-select:focus{z-index:5}.input-group .btn{position:relative;z-index:2}.input-group .btn:focus{z-index:5}.input-group-text{align-items:center;background-color:var(--bs-tertiary-bg);border:var(--bs-border-width) solid var(--bs-border-color);border-radius:var(--bs-border-radius);color:var(--bs-body-color);display:flex;font-size:1rem;font-weight:400;line-height:1.5;padding:.375rem .75rem;text-align:center;white-space:nowrap}.input-group-lg>.btn,.input-group-lg>.form-control,.input-group-lg>.form-select,.input-group-lg>.input-group-text{border-radius:var(--bs-border-radius-lg);font-size:1.25rem;padding:.5rem 1rem}.input-group-sm>.btn,.input-group-sm>.form-control,.input-group-sm>.form-select,.input-group-sm>.input-group-text{border-radius:var(--bs-border-radius-sm);font-size:.875rem;padding:.25rem .5rem}.input-group-lg>.form-select,.input-group-sm>.form-select{padding-right:3rem}.input-group.has-validation>.dropdown-toggle:nth-last-child(n+4),.input-group.has-validation>.form-floating:nth-last-child(n+3)>.form-control,.input-group.has-validation>.form-floating:nth-last-child(n+3)>.form-select,.input-group.has-validation>:nth-last-child(n+3):not(.dropdown-toggle):not(.dropdown-menu):not(.form-floating),.input-group:not(.has-validation)>.dropdown-toggle:nth-last-child(n+3),.input-group:not(.has-validation)>.form-floating:not(:last-child)>.form-control,.input-group:not(.has-validation)>.form-floating:not(:last-child)>.form-select,.input-group:not(.has-validation)>:not(:last-child):not(.dropdown-toggle):not(.dropdown-menu):not(.form-floating){border-bottom-right-radius:0;border-top-right-radius:0}.input-group>:not(:first-child):not(.dropdown-menu):not(.valid-tooltip):not(.valid-feedback):not(.invalid-tooltip):not(.invalid-feedback){border-bottom-left-radius:0;border-top-left-radius:0;margin-left:calc(var(--bs-border-width)*-1)}.input-group>.form-floating:not(:first-child)>.form-control,.input-group>.form-floating:not(:first-child)>.form-select{border-bottom-left-radius:0;border-top-left-radius:0}.valid-feedback{color:var(--bs-form-valid-color);display:none;font-size:.875em;margin-top:.25rem;width:100%}.valid-tooltip{background-color:var(--bs-success);border-radius:var(--bs-border-radius);color:#fff;display:none;font-size:.875rem;margin-top:.1rem;max-width:100%;padding:.25rem .5rem;position:absolute;top:100%;z-index:5}.is-valid~.valid-feedback,.is-valid~.valid-tooltip,.was-validated :valid~.valid-feedback,.was-validated :valid~.valid-tooltip{display:block}.form-control.is-valid,.was-validated .form-control:valid{background-image:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 8 8'%3E%3Cpath fill='%23198754' d='M2.3 6.73.6 4.53c-.4-1.04.46-1.4 1.1-.8l1.1 1.4 3.4-3.8c.6-.63 1.6-.27 1.2.7l-4 4.6c-.43.5-.8.4-1.1.1z'/%3E%3C/svg%3E");background-position:right calc(.375em + .1875rem) center;background-repeat:no-repeat;background-size:calc(.75em + .375rem) calc(.75em + .375rem);border-color:var(--bs-form-valid-border-color);padding-right:calc(1.5em + .75rem)}.form-control.is-valid:focus,.was-validated .form-control:valid:focus{border-color:var(--bs-form-valid-border-color);box-shadow:0 0 0 .1875rem rgba(var(--bs-success-rgb),1)}.was-validated textarea.form-control:valid,textarea.form-control.is-valid{background-position:top calc(.375em + .1875rem) right calc(.375em + .1875rem);padding-right:calc(1.5em + .75rem)}.form-select.is-valid,.was-validated .form-select:valid{border-color:var(--bs-form-valid-border-color)}.form-select.is-valid:not([multiple]):not([size]),.form-select.is-valid:not([multiple])[size="1"],.was-validated .form-select:valid:not([multiple]):not([size]),.was-validated .form-select:valid:not([multiple])[size="1"]{--bs-form-select-bg-icon:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 8 8'%3E%3Cpath fill='%23198754' d='M2.3 6.73.6 4.53c-.4-1.04.46-1.4 1.1-.8l1.1 1.4 3.4-3.8c.6-.63 1.6-.27 1.2.7l-4 4.6c-.43.5-.8.4-1.1.1z'/%3E%3C/svg%3E");background-position:right .75rem center,center right 2.25rem;background-size:16px 12px,calc(.75em + .375rem) calc(.75em + .375rem);padding-right:4.125rem}.form-select.is-valid:focus,.was-validated .form-select:valid:focus{border-color:var(--bs-form-valid-border-color);box-shadow:0 0 0 .1875rem rgba(var(--bs-success-rgb),1)}.form-control-color.is-valid,.was-validated .form-control-color:valid{width:calc(3.75rem + 1.5em)}.form-check-input.is-valid,.was-validated .form-check-input:valid{border-color:var(--bs-form-valid-border-color)}.form-check-input.is-valid:checked,.was-validated .form-check-input:valid:checked{background-color:var(--bs-form-valid-color)}.form-check-input.is-valid:focus,.was-validated .form-check-input:valid:focus{box-shadow:0 0 0 .1875rem rgba(var(--bs-success-rgb),1)}.form-check-input.is-valid~.form-check-label,.was-validated .form-check-input:valid~.form-check-label{color:var(--bs-form-valid-color)}.form-check-inline .form-check-input~.valid-feedback{margin-left:.5em}.input-group>.form-control:not(:focus).is-valid,.input-group>.form-floating:not(:focus-within).is-valid,.input-group>.form-select:not(:focus).is-valid,.was-validated .input-group>.form-control:not(:focus):valid,.was-validated .input-group>.form-floating:not(:focus-within):valid,.was-validated .input-group>.form-select:not(:focus):valid{z-index:3}.invalid-feedback{color:var(--bs-form-invalid-color);display:none;font-size:.875em;margin-top:.25rem;width:100%}.invalid-tooltip{background-color:var(--bs-danger);border-radius:var(--bs-border-radius);color:#fff;display:none;font-size:.875rem;margin-top:.1rem;max-width:100%;padding:.25rem .5rem;position:absolute;top:100%;z-index:5}.is-invalid~.invalid-feedback,.is-invalid~.invalid-tooltip,.was-validated :invalid~.invalid-feedback,.was-validated :invalid~.invalid-tooltip{display:block}.form-control.is-invalid,.was-validated .form-control:invalid{background-image:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' width='12' height='12' fill='none' stroke='%23dc3545'%3E%3Ccircle cx='6' cy='6' r='4.5'/%3E%3Cpath stroke-linejoin='round' d='M5.8 3.6h.4L6 6.5z'/%3E%3Ccircle cx='6' cy='8.2' r='.6' fill='%23dc3545' stroke='none'/%3E%3C/svg%3E");background-position:right calc(.375em + .1875rem) center;background-repeat:no-repeat;background-size:calc(.75em + .375rem) calc(.75em + .375rem);border-color:var(--bs-form-invalid-border-color);padding-right:calc(1.5em + .75rem)}.form-control.is-invalid:focus,.was-validated .form-control:invalid:focus{border-color:var(--bs-form-invalid-border-color);box-shadow:0 0 0 .1875rem rgba(var(--bs-danger-rgb),1)}.was-validated textarea.form-control:invalid,textarea.form-control.is-invalid{background-position:top calc(.375em + .1875rem) right calc(.375em + .1875rem);padding-right:calc(1.5em + .75rem)}.form-select.is-invalid,.was-validated .form-select:invalid{border-color:var(--bs-form-invalid-border-color)}.form-select.is-invalid:not([multiple]):not([size]),.form-select.is-invalid:not([multiple])[size="1"],.was-validated .form-select:invalid:not([multiple]):not([size]),.was-validated .form-select:invalid:not([multiple])[size="1"]{--bs-form-select-bg-icon:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' width='12' height='12' fill='none' stroke='%23dc3545'%3E%3Ccircle cx='6' cy='6' r='4.5'/%3E%3Cpath stroke-linejoin='round' d='M5.8 3.6h.4L6 6.5z'/%3E%3Ccircle cx='6' cy='8.2' r='.6' fill='%23dc3545' stroke='none'/%3E%3C/svg%3E");background-position:right .75rem center,center right 2.25rem;background-size:16px 12px,calc(.75em + .375rem) calc(.75em + .375rem);padding-right:4.125rem}.form-select.is-invalid:focus,.was-validated .form-select:invalid:focus{border-color:var(--bs-form-invalid-border-color);box-shadow:0 0 0 .1875rem rgba(var(--bs-danger-rgb),1)}.form-control-color.is-invalid,.was-validated .form-control-color:invalid{width:calc(3.75rem + 1.5em)}.form-check-input.is-invalid,.was-validated .form-check-input:invalid{border-color:var(--bs-form-invalid-border-color)}.form-check-input.is-invalid:checked,.was-validated .form-check-input:invalid:checked{background-color:var(--bs-form-invalid-color)}.form-check-input.is-invalid:focus,.was-validated .form-check-input:invalid:focus{box-shadow:0 0 0 .1875rem rgba(var(--bs-danger-rgb),1)}.form-check-input.is-invalid~.form-check-label,.was-validated .form-check-input:invalid~.form-check-label{color:var(--bs-form-invalid-color)}.form-check-inline .form-check-input~.invalid-feedback{margin-left:.5em}.input-group>.form-control:not(:focus).is-invalid,.input-group>.form-floating:not(:focus-within).is-invalid,.input-group>.form-select:not(:focus).is-invalid,.was-validated .input-group>.form-control:not(:focus):invalid,.was-validated .input-group>.form-floating:not(:focus-within):invalid,.was-validated .input-group>.form-select:not(:focus):invalid{z-index:4}.btn{--bs-btn-padding-x:0.75rem;--bs-btn-padding-y:0.375rem;--bs-btn-font-family: ;--bs-btn-font-size:1rem;--bs-btn-font-weight:400;--bs-btn-line-height:1.5;--bs-btn-color:var(--bs-body-color);--bs-btn-bg:transparent;--bs-btn-border-width:var(--bs-border-width);--bs-btn-border-color:transparent;--bs-btn-border-radius:var(--bs-border-radius);--bs-btn-hover-border-color:transparent;--bs-btn-box-shadow:inset 0 1px 0 hsla(0,0%,100%,.15),0 1px 1px rgba(0,0,0,.075);--bs-btn-disabled-opacity:0.65;--bs-btn-focus-box-shadow:0 0 0 0.1875rem rgba(var(--bs-btn-focus-shadow-rgb),.5);background-color:var(--bs-btn-bg);border:var(--bs-btn-border-width) solid var(--bs-btn-border-color);border-radius:var(--bs-btn-border-radius);color:var(--bs-btn-color);cursor:pointer;display:inline-block;font-family:var(--bs-btn-font-family);font-size:var(--bs-btn-font-size);font-weight:var(--bs-btn-font-weight);line-height:var(--bs-btn-line-height);padding:var(--bs-btn-padding-y) var(--bs-btn-padding-x);text-align:center;text-decoration:none;transition:color .15s ease-in-out,background-color .15s ease-in-out,border-color .15s ease-in-out,box-shadow .15s ease-in-out;user-select:none;vertical-align:middle}@media (prefers-reduced-motion:reduce){.btn{transition:none}}.btn:hover{background-color:var(--bs-btn-hover-bg);border-color:var(--bs-btn-hover-border-color);color:var(--bs-btn-hover-color)}.btn-check+.btn:hover{background-color:var(--bs-btn-bg);border-color:var(--bs-btn-border-color);color:var(--bs-btn-color)}.btn:focus-visible{background-color:var(--bs-btn-hover-bg);border-color:var(--bs-btn-hover-border-color);box-shadow:var(--bs-btn-focus-box-shadow);color:var(--bs-btn-hover-color);outline:0}.btn-check:focus-visible+.btn{border-color:var(--bs-btn-hover-border-color);box-shadow:var(--bs-btn-focus-box-shadow);outline:0}.btn-check:checked+.btn,.btn.active,.btn.show,.btn:first-child:active,:not(.btn-check)+.btn:active{background-color:var(--bs-btn-active-bg);border-color:var(--bs-btn-active-border-color);color:var(--bs-btn-active-color)}.btn-check:checked+.btn:focus-visible,.btn.active:focus-visible,.btn.show:focus-visible,.btn:first-child:active:focus-visible,:not(.btn-check)+.btn:active:focus-visible{box-shadow:var(--bs-btn-focus-box-shadow)}.btn-check:checked:focus-visible+.btn{box-shadow:var(--bs-btn-focus-box-shadow)}.btn.disabled,.btn:disabled,fieldset:disabled .btn{background-color:var(--bs-btn-disabled-bg);border-color:var(--bs-btn-disabled-border-color);color:var(--bs-btn-disabled-color);opacity:var(--bs-btn-disabled-opacity);pointer-events:none}.btn-primary{--bs-btn-color:#fff;--bs-btn-bg:#0d6efd;--bs-btn-border-color:#0d6efd;--bs-btn-hover-color:#fff;--bs-btn-hover-bg:#0b5ed7;--bs-btn-hover-border-color:#0a58ca;--bs-btn-focus-shadow-rgb:49,132,253;--bs-btn-active-color:#fff;--bs-btn-active-bg:#0a58ca;--bs-btn-active-border-color:#0a53be;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#fff;--bs-btn-disabled-bg:#0d6efd;--bs-btn-disabled-border-color:#0d6efd}.btn-secondary{--bs-btn-color:#fff;--bs-btn-bg:#6c757d;--bs-btn-border-color:#6c757d;--bs-btn-hover-color:#fff;--bs-btn-hover-bg:#5c636a;--bs-btn-hover-border-color:#565e64;--bs-btn-focus-shadow-rgb:130,138,145;--bs-btn-active-color:#fff;--bs-btn-active-bg:#565e64;--bs-btn-active-border-color:#51585e;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#fff;--bs-btn-disabled-bg:#6c757d;--bs-btn-disabled-border-color:#6c757d}.btn-success{--bs-btn-color:#fff;--bs-btn-bg:#198754;--bs-btn-border-color:#198754;--bs-btn-hover-color:#fff;--bs-btn-hover-bg:#157347;--bs-btn-hover-border-color:#146c43;--bs-btn-focus-shadow-rgb:60,153,110;--bs-btn-active-color:#fff;--bs-btn-active-bg:#146c43;--bs-btn-active-border-color:#13653f;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#fff;--bs-btn-disabled-bg:#198754;--bs-btn-disabled-border-color:#198754}.btn-info{--bs-btn-color:#000;--bs-btn-bg:#0dcaf0;--bs-btn-border-color:#0dcaf0;--bs-btn-hover-color:#000;--bs-btn-hover-bg:#31d2f2;--bs-btn-hover-border-color:#25cff2;--bs-btn-focus-shadow-rgb:11,172,204;--bs-btn-active-color:#000;--bs-btn-active-bg:#3dd5f3;--bs-btn-active-border-color:#25cff2;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#000;--bs-btn-disabled-bg:#0dcaf0;--bs-btn-disabled-border-color:#0dcaf0}.btn-warning{--bs-btn-color:#000;--bs-btn-bg:#ffc107;--bs-btn-border-color:#ffc107;--bs-btn-hover-color:#000;--bs-btn-hover-bg:#ffca2c;--bs-btn-hover-border-color:#ffc720;--bs-btn-focus-shadow-rgb:217,164,6;--bs-btn-active-color:#000;--bs-btn-active-bg:#ffcd39;--bs-btn-active-border-color:#ffc720;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#000;--bs-btn-disabled-bg:#ffc107;--bs-btn-disabled-border-color:#ffc107}.btn-danger{--bs-btn-color:#fff;--bs-btn-bg:#dc3545;--bs-btn-border-color:#dc3545;--bs-btn-hover-color:#fff;--bs-btn-hover-bg:#bb2d3b;--bs-btn-hover-border-color:#b02a37;--bs-btn-focus-shadow-rgb:225,83,97;--bs-btn-active-color:#fff;--bs-btn-active-bg:#b02a37;--bs-btn-active-border-color:#a52834;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#fff;--bs-btn-disabled-bg:#dc3545;--bs-btn-disabled-border-color:#dc3545}.btn-light{--bs-btn-color:#000;--bs-btn-bg:#f8f9fa;--bs-btn-border-color:#f8f9fa;--bs-btn-hover-color:#000;--bs-btn-hover-bg:#d3d4d5;--bs-btn-hover-border-color:#c6c7c8;--bs-btn-focus-shadow-rgb:211,212,213;--bs-btn-active-color:#000;--bs-btn-active-bg:#c6c7c8;--bs-btn-active-border-color:#babbbc;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#000;--bs-btn-disabled-bg:#f8f9fa;--bs-btn-disabled-border-color:#f8f9fa}.btn-dark{--bs-btn-color:#fff;--bs-btn-bg:#212529;--bs-btn-border-color:#212529;--bs-btn-hover-color:#fff;--bs-btn-hover-bg:#424649;--bs-btn-hover-border-color:#373b3e;--bs-btn-focus-shadow-rgb:66,70,73;--bs-btn-active-color:#fff;--bs-btn-active-bg:#4d5154;--bs-btn-active-border-color:#373b3e;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#fff;--bs-btn-disabled-bg:#212529;--bs-btn-disabled-border-color:#212529}.btn-outline-primary{--bs-btn-color:#0d6efd;--bs-btn-border-color:#0d6efd;--bs-btn-hover-color:#fff;--bs-btn-hover-bg:#0d6efd;--bs-btn-hover-border-color:#0d6efd;--bs-btn-focus-shadow-rgb:13,110,253;--bs-btn-active-color:#fff;--bs-btn-active-bg:#0d6efd;--bs-btn-active-border-color:#0d6efd;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#0d6efd;--bs-btn-disabled-bg:transparent;--bs-btn-disabled-border-color:#0d6efd;--bs-gradient:none}.btn-outline-secondary{--bs-btn-color:#6c757d;--bs-btn-border-color:#6c757d;--bs-btn-hover-color:#fff;--bs-btn-hover-bg:#6c757d;--bs-btn-hover-border-color:#6c757d;--bs-btn-focus-shadow-rgb:108,117,125;--bs-btn-active-color:#fff;--bs-btn-active-bg:#6c757d;--bs-btn-active-border-color:#6c757d;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#6c757d;--bs-btn-disabled-bg:transparent;--bs-btn-disabled-border-color:#6c757d;--bs-gradient:none}.btn-outline-success{--bs-btn-color:#198754;--bs-btn-border-color:#198754;--bs-btn-hover-color:#fff;--bs-btn-hover-bg:#198754;--bs-btn-hover-border-color:#198754;--bs-btn-focus-shadow-rgb:25,135,84;--bs-btn-active-color:#fff;--bs-btn-active-bg:#198754;--bs-btn-active-border-color:#198754;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#198754;--bs-btn-disabled-bg:transparent;--bs-btn-disabled-border-color:#198754;--bs-gradient:none}.btn-outline-info{--bs-btn-color:#0dcaf0;--bs-btn-border-color:#0dcaf0;--bs-btn-hover-color:#000;--bs-btn-hover-bg:#0dcaf0;--bs-btn-hover-border-color:#0dcaf0;--bs-btn-focus-shadow-rgb:13,202,240;--bs-btn-active-color:#000;--bs-btn-active-bg:#0dcaf0;--bs-btn-active-border-color:#0dcaf0;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#0dcaf0;--bs-btn-disabled-bg:transparent;--bs-btn-disabled-border-color:#0dcaf0;--bs-gradient:none}.btn-outline-warning{--bs-btn-color:#ffc107;--bs-btn-border-color:#ffc107;--bs-btn-hover-color:#000;--bs-btn-hover-bg:#ffc107;--bs-btn-hover-border-color:#ffc107;--bs-btn-focus-shadow-rgb:255,193,7;--bs-btn-active-color:#000;--bs-btn-active-bg:#ffc107;--bs-btn-active-border-color:#ffc107;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#ffc107;--bs-btn-disabled-bg:transparent;--bs-btn-disabled-border-color:#ffc107;--bs-gradient:none}.btn-outline-danger{--bs-btn-color:#dc3545;--bs-btn-border-color:#dc3545;--bs-btn-hover-color:#fff;--bs-btn-hover-bg:#dc3545;--bs-btn-hover-border-color:#dc3545;--bs-btn-focus-shadow-rgb:220,53,69;--bs-btn-active-color:#fff;--bs-btn-active-bg:#dc3545;--bs-btn-active-border-color:#dc3545;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#dc3545;--bs-btn-disabled-bg:transparent;--bs-btn-disabled-border-color:#dc3545;--bs-gradient:none}.btn-outline-light{--bs-btn-color:#f8f9fa;--bs-btn-border-color:#f8f9fa;--bs-btn-hover-color:#000;--bs-btn-hover-bg:#f8f9fa;--bs-btn-hover-border-color:#f8f9fa;--bs-btn-focus-shadow-rgb:248,249,250;--bs-btn-active-color:#000;--bs-btn-active-bg:#f8f9fa;--bs-btn-active-border-color:#f8f9fa;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#f8f9fa;--bs-btn-disabled-bg:transparent;--bs-btn-disabled-border-color:#f8f9fa;--bs-gradient:none}.btn-outline-dark{--bs-btn-color:#212529;--bs-btn-border-color:#212529;--bs-btn-hover-color:#fff;--bs-btn-hover-bg:#212529;--bs-btn-hover-border-color:#212529;--bs-btn-focus-shadow-rgb:33,37,41;--bs-btn-active-color:#fff;--bs-btn-active-bg:#212529;--bs-btn-active-border-color:#212529;--bs-btn-active-shadow:inset 0 3px 5px rgba(0,0,0,.125);--bs-btn-disabled-color:#212529;--bs-btn-disabled-bg:transparent;--bs-btn-disabled-border-color:#212529;--bs-gradient:none}.btn-link{--bs-btn-font-weight:400;--bs-btn-color:var(--bs-link-color);--bs-btn-bg:transparent;--bs-btn-border-color:transparent;--bs-btn-hover-color:var(--bs-link-hover-color);--bs-btn-hover-border-color:transparent;--bs-btn-active-color:var(--bs-link-hover-color);--bs-btn-active-border-color:transparent;--bs-btn-disabled-color:#6c757d;--bs-btn-disabled-border-color:transparent;--bs-btn-box-shadow:0 0 0 #000;--bs-btn-focus-shadow-rgb:49,132,253;text-decoration:underline}.btn-link:focus-visible{color:var(--bs-btn-color)}.btn-link:hover{color:var(--bs-btn-hover-color)}.btn-group-lg>.btn,.btn-lg{--bs-btn-padding-y:0.5rem;--bs-btn-padding-x:1rem;--bs-btn-font-size:1.25rem;--bs-btn-border-radius:var(--bs-border-radius-lg)}.btn-group-sm>.btn,.btn-sm{--bs-btn-padding-y:0.25rem;--bs-btn-padding-x:0.5rem;--bs-btn-font-size:0.875rem;--bs-btn-border-radius:var(--bs-border-radius-sm)}.fade{transition:opacity .15s linear}@media (prefers-reduced-motion:reduce){.fade{transition:none}}.fade:not(.show){opacity:0}.collapse:not(.show){display:none}.collapsing{height:0;overflow:hidden;transition:height .35s ease}@media (prefers-reduced-motion:reduce){.collapsing{transition:none}}.collapsing.collapse-horizontal{height:auto;transition:width .35s ease;width:0}@media (prefers-reduced-motion:reduce){.collapsing.collapse-horizontal{transition:none}}.dropdown,.dropdown-center,.dropend,.dropstart,.dropup,.dropup-center{position:relative}.dropdown-toggle{white-space:nowrap}.dropdown-toggle:after{border-bottom:0;border-left:.3em solid transparent;border-right:.3em solid transparent;border-top:.3em solid;content:"";display:inline-block;margin-left:.255em;vertical-align:.255em}.dropdown-toggle:empty:after{margin-left:0}.dropdown-menu{--bs-dropdown-zindex:1000;--bs-dropdown-min-width:10rem;--bs-dropdown-padding-x:0;--bs-dropdown-padding-y:0.5rem;--bs-dropdown-spacer:0.125rem;--bs-dropdown-font-size:1rem;--bs-dropdown-color:var(--bs-body-color);--bs-dropdown-bg:var(--bs-body-bg);--bs-dropdown-border-color:var(--bs-border-color-translucent);--bs-dropdown-border-radius:var(--bs-border-radius);--bs-dropdown-border-width:var(--bs-border-width);--bs-dropdown-inner-border-radius:calc(var(--bs-border-radius) - var(--bs-border-width));--bs-dropdown-divider-bg:var(--bs-border-color-translucent);--bs-dropdown-divider-margin-y:0.5rem;--bs-dropdown-box-shadow:var(--bs-box-shadow);--bs-dropdown-link-color:var(--bs-body-color);--bs-dropdown-link-hover-color:var(--bs-body-color);--bs-dropdown-link-hover-bg:var(--pst-color-surface);--bs-dropdown-link-active-color:#fff;--bs-dropdown-link-active-bg:var(--pst-color-surface);--bs-dropdown-link-disabled-color:var(--bs-tertiary-color);--bs-dropdown-item-padding-x:1rem;--bs-dropdown-item-padding-y:0.25rem;--bs-dropdown-header-color:#6c757d;--bs-dropdown-header-padding-x:1rem;--bs-dropdown-header-padding-y:0.5rem;background-clip:padding-box;background-color:var(--bs-dropdown-bg);border:var(--bs-dropdown-border-width) solid var(--bs-dropdown-border-color);border-radius:var(--bs-dropdown-border-radius);color:var(--bs-dropdown-color);display:none;font-size:var(--bs-dropdown-font-size);list-style:none;margin:0;min-width:var(--bs-dropdown-min-width);padding:var(--bs-dropdown-padding-y) var(--bs-dropdown-padding-x);position:absolute;text-align:left;z-index:var(--bs-dropdown-zindex)}.dropdown-menu[data-bs-popper]{left:0;margin-top:var(--bs-dropdown-spacer);top:100%}.dropdown-menu-start{--bs-position:start}.dropdown-menu-start[data-bs-popper]{left:0;right:auto}.dropdown-menu-end{--bs-position:end}.dropdown-menu-end[data-bs-popper]{left:auto;right:0}@media (min-width:540px){.dropdown-menu-sm-start{--bs-position:start}.dropdown-menu-sm-start[data-bs-popper]{left:0;right:auto}.dropdown-menu-sm-end{--bs-position:end}.dropdown-menu-sm-end[data-bs-popper]{left:auto;right:0}}@media (min-width:720px){.dropdown-menu-md-start{--bs-position:start}.dropdown-menu-md-start[data-bs-popper]{left:0;right:auto}.dropdown-menu-md-end{--bs-position:end}.dropdown-menu-md-end[data-bs-popper]{left:auto;right:0}}@media (min-width:960px){.dropdown-menu-lg-start{--bs-position:start}.dropdown-menu-lg-start[data-bs-popper]{left:0;right:auto}.dropdown-menu-lg-end{--bs-position:end}.dropdown-menu-lg-end[data-bs-popper]{left:auto;right:0}}@media (min-width:1200px){.dropdown-menu-xl-start{--bs-position:start}.dropdown-menu-xl-start[data-bs-popper]{left:0;right:auto}.dropdown-menu-xl-end{--bs-position:end}.dropdown-menu-xl-end[data-bs-popper]{left:auto;right:0}}.dropup .dropdown-menu[data-bs-popper]{bottom:100%;margin-bottom:var(--bs-dropdown-spacer);margin-top:0;top:auto}.dropup .dropdown-toggle:after{border-bottom:.3em solid;border-left:.3em solid transparent;border-right:.3em solid transparent;border-top:0;content:"";display:inline-block;margin-left:.255em;vertical-align:.255em}.dropup .dropdown-toggle:empty:after{margin-left:0}.dropend .dropdown-menu[data-bs-popper]{left:100%;margin-left:var(--bs-dropdown-spacer);margin-top:0;right:auto;top:0}.dropend .dropdown-toggle:after{border-bottom:.3em solid transparent;border-left:.3em solid;border-right:0;border-top:.3em solid transparent;content:"";display:inline-block;margin-left:.255em;vertical-align:.255em}.dropend .dropdown-toggle:empty:after{margin-left:0}.dropend .dropdown-toggle:after{vertical-align:0}.dropstart .dropdown-menu[data-bs-popper]{left:auto;margin-right:var(--bs-dropdown-spacer);margin-top:0;right:100%;top:0}.dropstart .dropdown-toggle:after{content:"";display:inline-block;display:none;margin-left:.255em;vertical-align:.255em}.dropstart .dropdown-toggle:before{border-bottom:.3em solid transparent;border-right:.3em solid;border-top:.3em solid transparent;content:"";display:inline-block;margin-right:.255em;vertical-align:.255em}.dropstart .dropdown-toggle:empty:after{margin-left:0}.dropstart .dropdown-toggle:before{vertical-align:0}.dropdown-divider{border-top:1px solid var(--bs-dropdown-divider-bg);height:0;margin:var(--bs-dropdown-divider-margin-y) 0;opacity:1;overflow:hidden}.dropdown-item{background-color:transparent;border:0;border-radius:var(--bs-dropdown-item-border-radius,0);clear:both;color:var(--bs-dropdown-link-color);display:block;font-weight:400;padding:var(--bs-dropdown-item-padding-y) var(--bs-dropdown-item-padding-x);text-align:inherit;text-decoration:none;white-space:nowrap;width:100%}.dropdown-item:focus,.dropdown-item:hover{background-color:var(--bs-dropdown-link-hover-bg);color:var(--bs-dropdown-link-hover-color)}.dropdown-item.active,.dropdown-item:active{background-color:var(--bs-dropdown-link-active-bg);color:var(--bs-dropdown-link-active-color);text-decoration:none}.dropdown-item.disabled,.dropdown-item:disabled{background-color:transparent;color:var(--bs-dropdown-link-disabled-color);pointer-events:none}.dropdown-menu.show{display:block}.dropdown-header{color:var(--bs-dropdown-header-color);display:block;font-size:.875rem;margin-bottom:0;padding:var(--bs-dropdown-header-padding-y) var(--bs-dropdown-header-padding-x);white-space:nowrap}.dropdown-item-text{color:var(--bs-dropdown-link-color);display:block;padding:var(--bs-dropdown-item-padding-y) var(--bs-dropdown-item-padding-x)}.dropdown-menu-dark{--bs-dropdown-color:#dee2e6;--bs-dropdown-bg:#343a40;--bs-dropdown-border-color:var(--bs-border-color-translucent);--bs-dropdown-box-shadow: ;--bs-dropdown-link-color:#dee2e6;--bs-dropdown-link-hover-color:#fff;--bs-dropdown-divider-bg:var(--bs-border-color-translucent);--bs-dropdown-link-hover-bg:var(--pst-color-surface);--bs-dropdown-link-active-color:#fff;--bs-dropdown-link-active-bg:var(--pst-color-surface);--bs-dropdown-link-disabled-color:#adb5bd;--bs-dropdown-header-color:#adb5bd}.btn-group,.btn-group-vertical{display:inline-flex;position:relative;vertical-align:middle}.btn-group-vertical>.btn,.btn-group>.btn{flex:1 1 auto;position:relative}.btn-group-vertical>.btn-check:checked+.btn,.btn-group-vertical>.btn-check:focus+.btn,.btn-group-vertical>.btn.active,.btn-group-vertical>.btn:active,.btn-group-vertical>.btn:focus,.btn-group-vertical>.btn:hover,.btn-group>.btn-check:checked+.btn,.btn-group>.btn-check:focus+.btn,.btn-group>.btn.active,.btn-group>.btn:active,.btn-group>.btn:focus,.btn-group>.btn:hover{z-index:1}.btn-toolbar{display:flex;flex-wrap:wrap;justify-content:flex-start}.btn-toolbar .input-group{width:auto}.btn-group{border-radius:var(--bs-border-radius)}.btn-group>.btn-group:not(:first-child),.btn-group>:not(.btn-check:first-child)+.btn{margin-left:calc(var(--bs-border-width)*-1)}.btn-group>.btn-group:not(:last-child)>.btn,.btn-group>.btn.dropdown-toggle-split:first-child,.btn-group>.btn:not(:last-child):not(.dropdown-toggle){border-bottom-right-radius:0;border-top-right-radius:0}.btn-group>.btn-group:not(:first-child)>.btn,.btn-group>.btn:nth-child(n+3),.btn-group>:not(.btn-check)+.btn{border-bottom-left-radius:0;border-top-left-radius:0}.dropdown-toggle-split{padding-left:.5625rem;padding-right:.5625rem}.dropdown-toggle-split:after,.dropend .dropdown-toggle-split:after,.dropup .dropdown-toggle-split:after{margin-left:0}.dropstart .dropdown-toggle-split:before{margin-right:0}.btn-group-sm>.btn+.dropdown-toggle-split,.btn-sm+.dropdown-toggle-split{padding-left:.375rem;padding-right:.375rem}.btn-group-lg>.btn+.dropdown-toggle-split,.btn-lg+.dropdown-toggle-split{padding-left:.75rem;padding-right:.75rem}.btn-group-vertical{align-items:flex-start;flex-direction:column;justify-content:center}.btn-group-vertical>.btn,.btn-group-vertical>.btn-group{width:100%}.btn-group-vertical>.btn-group:not(:first-child),.btn-group-vertical>.btn:not(:first-child){margin-top:calc(var(--bs-border-width)*-1)}.btn-group-vertical>.btn-group:not(:last-child)>.btn,.btn-group-vertical>.btn:not(:last-child):not(.dropdown-toggle){border-bottom-left-radius:0;border-bottom-right-radius:0}.btn-group-vertical>.btn-group:not(:first-child)>.btn,.btn-group-vertical>.btn~.btn{border-top-left-radius:0;border-top-right-radius:0}.nav{--bs-nav-link-padding-x:1rem;--bs-nav-link-padding-y:0.5rem;--bs-nav-link-font-weight: ;--bs-nav-link-color:var(--bs-link-color);--bs-nav-link-hover-color:var(--bs-link-hover-color);--bs-nav-link-disabled-color:var(--bs-secondary-color);display:flex;flex-wrap:wrap;list-style:none;margin-bottom:0;padding-left:0}.nav-link{background:none;border:0;color:var(--bs-nav-link-color);display:block;font-size:var(--bs-nav-link-font-size);font-weight:var(--bs-nav-link-font-weight);padding:var(--bs-nav-link-padding-y) var(--bs-nav-link-padding-x);text-decoration:none;transition:color .15s ease-in-out,background-color .15s ease-in-out,border-color .15s ease-in-out}@media (prefers-reduced-motion:reduce){.nav-link{transition:none}}.nav-link:focus,.nav-link:hover{color:var(--bs-nav-link-hover-color)}.nav-link:focus-visible{box-shadow:0 0 0 .1875rem var(--pst-color-accent);outline:0}.nav-link.disabled,.nav-link:disabled{color:var(--bs-nav-link-disabled-color);cursor:default;pointer-events:none}.nav-tabs{--bs-nav-tabs-border-width:var(--bs-border-width);--bs-nav-tabs-border-color:var(--bs-border-color);--bs-nav-tabs-border-radius:var(--bs-border-radius);--bs-nav-tabs-link-hover-border-color:var(--bs-secondary-bg) var(--bs-secondary-bg) var(--bs-border-color);--bs-nav-tabs-link-active-color:var(--bs-emphasis-color);--bs-nav-tabs-link-active-bg:var(--bs-body-bg);--bs-nav-tabs-link-active-border-color:var(--bs-border-color) var(--bs-border-color) var(--bs-body-bg);border-bottom:var(--bs-nav-tabs-border-width) solid var(--bs-nav-tabs-border-color)}.nav-tabs .nav-link{border:var(--bs-nav-tabs-border-width) solid transparent;border-top-left-radius:var(--bs-nav-tabs-border-radius);border-top-right-radius:var(--bs-nav-tabs-border-radius);margin-bottom:calc(var(--bs-nav-tabs-border-width)*-1)}.nav-tabs .nav-link:focus,.nav-tabs .nav-link:hover{border-color:var(--bs-nav-tabs-link-hover-border-color);isolation:isolate}.nav-tabs .nav-item.show .nav-link,.nav-tabs .nav-link.active{background-color:var(--bs-nav-tabs-link-active-bg);border-color:var(--bs-nav-tabs-link-active-border-color);color:var(--bs-nav-tabs-link-active-color)}.nav-tabs .dropdown-menu{border-top-left-radius:0;border-top-right-radius:0;margin-top:calc(var(--bs-nav-tabs-border-width)*-1)}.nav-pills{--bs-nav-pills-border-radius:var(--bs-border-radius);--bs-nav-pills-link-active-color:#fff;--bs-nav-pills-link-active-bg:#0d6efd}.nav-pills .nav-link{border-radius:var(--bs-nav-pills-border-radius)}.nav-pills .nav-link.active,.nav-pills .show>.nav-link{background-color:var(--bs-nav-pills-link-active-bg);color:var(--bs-nav-pills-link-active-color)}.nav-underline{--bs-nav-underline-gap:1rem;--bs-nav-underline-border-width:0.125rem;--bs-nav-underline-link-active-color:var(--bs-emphasis-color);gap:var(--bs-nav-underline-gap)}.nav-underline .nav-link{border-bottom:var(--bs-nav-underline-border-width) solid transparent;padding-left:0;padding-right:0}.nav-underline .nav-link:focus,.nav-underline .nav-link:hover{border-bottom-color:currentcolor}.nav-underline .nav-link.active,.nav-underline .show>.nav-link{border-bottom-color:currentcolor;color:var(--bs-nav-underline-link-active-color);font-weight:700}.nav-fill .nav-item,.nav-fill>.nav-link{flex:1 1 auto;text-align:center}.nav-justified .nav-item,.nav-justified>.nav-link{flex-basis:0;flex-grow:1;text-align:center}.nav-fill .nav-item .nav-link,.nav-justified .nav-item .nav-link{width:100%}.tab-content>.tab-pane{display:none}.tab-content>.active{display:block}.navbar{--bs-navbar-padding-x:0;--bs-navbar-padding-y:0.5rem;--bs-navbar-color:rgba(var(--bs-emphasis-color-rgb),0.65);--bs-navbar-hover-color:rgba(var(--bs-emphasis-color-rgb),0.8);--bs-navbar-disabled-color:rgba(var(--bs-emphasis-color-rgb),0.3);--bs-navbar-active-color:rgba(var(--bs-emphasis-color-rgb),1);--bs-navbar-brand-padding-y:0.3125rem;--bs-navbar-brand-margin-end:1rem;--bs-navbar-brand-font-size:1.25rem;--bs-navbar-brand-color:rgba(var(--bs-emphasis-color-rgb),1);--bs-navbar-brand-hover-color:rgba(var(--bs-emphasis-color-rgb),1);--bs-navbar-nav-link-padding-x:0.5rem;--bs-navbar-toggler-padding-y:0.25rem;--bs-navbar-toggler-padding-x:0.75rem;--bs-navbar-toggler-font-size:1.25rem;--bs-navbar-toggler-icon-bg:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 30 30'%3E%3Cpath stroke='rgba(33, 37, 41, 0.75)' stroke-linecap='round' stroke-miterlimit='10' stroke-width='2' d='M4 7h22M4 15h22M4 23h22'/%3E%3C/svg%3E");--bs-navbar-toggler-border-color:rgba(var(--bs-emphasis-color-rgb),0.15);--bs-navbar-toggler-border-radius:var(--bs-border-radius);--bs-navbar-toggler-focus-width:0.1875rem;--bs-navbar-toggler-transition:box-shadow 0.15s ease-in-out;align-items:center;display:flex;flex-wrap:wrap;justify-content:space-between;padding:var(--bs-navbar-padding-y) var(--bs-navbar-padding-x);position:relative}.navbar>.container,.navbar>.container-fluid,.navbar>.container-lg,.navbar>.container-md,.navbar>.container-sm,.navbar>.container-xl{align-items:center;display:flex;flex-wrap:inherit;justify-content:space-between}.navbar-brand{color:var(--bs-navbar-brand-color);font-size:var(--bs-navbar-brand-font-size);margin-right:var(--bs-navbar-brand-margin-end);padding-bottom:var(--bs-navbar-brand-padding-y);padding-top:var(--bs-navbar-brand-padding-y);text-decoration:none;white-space:nowrap}.navbar-brand:focus,.navbar-brand:hover{color:var(--bs-navbar-brand-hover-color)}.navbar-nav{--bs-nav-link-padding-x:0;--bs-nav-link-padding-y:0.5rem;--bs-nav-link-font-weight: ;--bs-nav-link-color:var(--bs-navbar-color);--bs-nav-link-hover-color:var(--bs-navbar-hover-color);--bs-nav-link-disabled-color:var(--bs-navbar-disabled-color);display:flex;flex-direction:column;list-style:none;margin-bottom:0;padding-left:0}.navbar-nav .nav-link.active,.navbar-nav .nav-link.show{color:var(--bs-navbar-active-color)}.navbar-nav .dropdown-menu{position:static}.navbar-text{color:var(--bs-navbar-color);padding-bottom:.5rem;padding-top:.5rem}.navbar-text a,.navbar-text a:focus,.navbar-text a:hover{color:var(--bs-navbar-active-color)}.navbar-collapse{align-items:center;flex-basis:100%;flex-grow:1}.navbar-toggler{background-color:transparent;border:var(--bs-border-width) solid var(--bs-navbar-toggler-border-color);border-radius:var(--bs-navbar-toggler-border-radius);color:var(--bs-navbar-color);font-size:var(--bs-navbar-toggler-font-size);line-height:1;padding:var(--bs-navbar-toggler-padding-y) var(--bs-navbar-toggler-padding-x);transition:var(--bs-navbar-toggler-transition)}@media (prefers-reduced-motion:reduce){.navbar-toggler{transition:none}}.navbar-toggler:hover{text-decoration:none}.navbar-toggler:focus{box-shadow:0 0 0 var(--bs-navbar-toggler-focus-width);outline:0;text-decoration:none}.navbar-toggler-icon{background-image:var(--bs-navbar-toggler-icon-bg);background-position:50%;background-repeat:no-repeat;background-size:100%;display:inline-block;height:1.5em;vertical-align:middle;width:1.5em}.navbar-nav-scroll{max-height:var(--bs-scroll-height,75vh);overflow-y:auto}@media (min-width:540px){.navbar-expand-sm{flex-wrap:nowrap;justify-content:flex-start}.navbar-expand-sm .navbar-nav{flex-direction:row}.navbar-expand-sm .navbar-nav .dropdown-menu{position:absolute}.navbar-expand-sm .navbar-nav .nav-link{padding-left:var(--bs-navbar-nav-link-padding-x);padding-right:var(--bs-navbar-nav-link-padding-x)}.navbar-expand-sm .navbar-nav-scroll{overflow:visible}.navbar-expand-sm .navbar-collapse{display:flex!important;flex-basis:auto}.navbar-expand-sm .navbar-toggler{display:none}.navbar-expand-sm .offcanvas{background-color:transparent!important;border:0!important;flex-grow:1;height:auto!important;position:static;transform:none!important;transition:none;visibility:visible!important;width:auto!important;z-index:auto}.navbar-expand-sm .offcanvas .offcanvas-header{display:none}.navbar-expand-sm .offcanvas .offcanvas-body{display:flex;flex-grow:0;overflow-y:visible;padding:0}}@media (min-width:720px){.navbar-expand-md{flex-wrap:nowrap;justify-content:flex-start}.navbar-expand-md .navbar-nav{flex-direction:row}.navbar-expand-md .navbar-nav .dropdown-menu{position:absolute}.navbar-expand-md .navbar-nav .nav-link{padding-left:var(--bs-navbar-nav-link-padding-x);padding-right:var(--bs-navbar-nav-link-padding-x)}.navbar-expand-md .navbar-nav-scroll{overflow:visible}.navbar-expand-md .navbar-collapse{display:flex!important;flex-basis:auto}.navbar-expand-md .navbar-toggler{display:none}.navbar-expand-md .offcanvas{background-color:transparent!important;border:0!important;flex-grow:1;height:auto!important;position:static;transform:none!important;transition:none;visibility:visible!important;width:auto!important;z-index:auto}.navbar-expand-md .offcanvas .offcanvas-header{display:none}.navbar-expand-md .offcanvas .offcanvas-body{display:flex;flex-grow:0;overflow-y:visible;padding:0}}@media (min-width:960px){.navbar-expand-lg{flex-wrap:nowrap;justify-content:flex-start}.navbar-expand-lg .navbar-nav{flex-direction:row}.navbar-expand-lg .navbar-nav .dropdown-menu{position:absolute}.navbar-expand-lg .navbar-nav .nav-link{padding-left:var(--bs-navbar-nav-link-padding-x);padding-right:var(--bs-navbar-nav-link-padding-x)}.navbar-expand-lg .navbar-nav-scroll{overflow:visible}.navbar-expand-lg .navbar-collapse{display:flex!important;flex-basis:auto}.navbar-expand-lg .navbar-toggler{display:none}.navbar-expand-lg .offcanvas{background-color:transparent!important;border:0!important;flex-grow:1;height:auto!important;position:static;transform:none!important;transition:none;visibility:visible!important;width:auto!important;z-index:auto}.navbar-expand-lg .offcanvas .offcanvas-header{display:none}.navbar-expand-lg .offcanvas .offcanvas-body{display:flex;flex-grow:0;overflow-y:visible;padding:0}}@media (min-width:1200px){.navbar-expand-xl{flex-wrap:nowrap;justify-content:flex-start}.navbar-expand-xl .navbar-nav{flex-direction:row}.navbar-expand-xl .navbar-nav .dropdown-menu{position:absolute}.navbar-expand-xl .navbar-nav .nav-link{padding-left:var(--bs-navbar-nav-link-padding-x);padding-right:var(--bs-navbar-nav-link-padding-x)}.navbar-expand-xl .navbar-nav-scroll{overflow:visible}.navbar-expand-xl .navbar-collapse{display:flex!important;flex-basis:auto}.navbar-expand-xl .navbar-toggler{display:none}.navbar-expand-xl .offcanvas{background-color:transparent!important;border:0!important;flex-grow:1;height:auto!important;position:static;transform:none!important;transition:none;visibility:visible!important;width:auto!important;z-index:auto}.navbar-expand-xl .offcanvas .offcanvas-header{display:none}.navbar-expand-xl .offcanvas .offcanvas-body{display:flex;flex-grow:0;overflow-y:visible;padding:0}}.navbar-expand{flex-wrap:nowrap;justify-content:flex-start}.navbar-expand .navbar-nav{flex-direction:row}.navbar-expand .navbar-nav .dropdown-menu{position:absolute}.navbar-expand .navbar-nav .nav-link{padding-left:var(--bs-navbar-nav-link-padding-x);padding-right:var(--bs-navbar-nav-link-padding-x)}.navbar-expand .navbar-nav-scroll{overflow:visible}.navbar-expand .navbar-collapse{display:flex!important;flex-basis:auto}.navbar-expand .navbar-toggler{display:none}.navbar-expand .offcanvas{background-color:transparent!important;border:0!important;flex-grow:1;height:auto!important;position:static;transform:none!important;transition:none;visibility:visible!important;width:auto!important;z-index:auto}.navbar-expand .offcanvas .offcanvas-header{display:none}.navbar-expand .offcanvas .offcanvas-body{display:flex;flex-grow:0;overflow-y:visible;padding:0}.navbar-dark,.navbar[data-bs-theme=dark]{--bs-navbar-color:hsla(0,0%,100%,.55);--bs-navbar-hover-color:hsla(0,0%,100%,.75);--bs-navbar-disabled-color:hsla(0,0%,100%,.25);--bs-navbar-active-color:#fff;--bs-navbar-brand-color:#fff;--bs-navbar-brand-hover-color:#fff;--bs-navbar-toggler-border-color:hsla(0,0%,100%,.1)}.navbar-dark,.navbar[data-bs-theme=dark],[data-bs-theme=dark] .navbar-toggler-icon{--bs-navbar-toggler-icon-bg:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 30 30'%3E%3Cpath stroke='rgba(255, 255, 255, 0.55)' stroke-linecap='round' stroke-miterlimit='10' stroke-width='2' d='M4 7h22M4 15h22M4 23h22'/%3E%3C/svg%3E")}.card{--bs-card-spacer-y:1rem;--bs-card-spacer-x:1rem;--bs-card-title-spacer-y:0.5rem;--bs-card-title-color: ;--bs-card-subtitle-color: ;--bs-card-border-width:var(--bs-border-width);--bs-card-border-color:var(--bs-border-color-translucent);--bs-card-border-radius:var(--bs-border-radius);--bs-card-box-shadow: ;--bs-card-inner-border-radius:calc(var(--bs-border-radius) - var(--bs-border-width));--bs-card-cap-padding-y:0.5rem;--bs-card-cap-padding-x:1rem;--bs-card-cap-bg:rgba(var(--bs-body-color-rgb),0.03);--bs-card-cap-color: ;--bs-card-height: ;--bs-card-color: ;--bs-card-bg:var(--bs-body-bg);--bs-card-img-overlay-padding:1rem;--bs-card-group-margin:0.75rem;word-wrap:break-word;background-clip:border-box;background-color:var(--bs-card-bg);border:var(--bs-card-border-width) solid var(--bs-card-border-color);border-radius:var(--bs-card-border-radius);color:var(--bs-body-color);display:flex;flex-direction:column;height:var(--bs-card-height);min-width:0;position:relative}.card>hr{margin-left:0;margin-right:0}.card>.list-group{border-bottom:inherit;border-top:inherit}.card>.list-group:first-child{border-top-left-radius:var(--bs-card-inner-border-radius);border-top-right-radius:var(--bs-card-inner-border-radius);border-top-width:0}.card>.list-group:last-child{border-bottom-left-radius:var(--bs-card-inner-border-radius);border-bottom-right-radius:var(--bs-card-inner-border-radius);border-bottom-width:0}.card>.card-header+.list-group,.card>.list-group+.card-footer{border-top:0}.card-body{color:var(--bs-card-color);flex:1 1 auto;padding:var(--bs-card-spacer-y) var(--bs-card-spacer-x)}.card-title{color:var(--bs-card-title-color);margin-bottom:var(--bs-card-title-spacer-y)}.card-subtitle{color:var(--bs-card-subtitle-color);margin-top:calc(var(--bs-card-title-spacer-y)*-.5)}.card-subtitle,.card-text:last-child{margin-bottom:0}.card-link+.card-link{margin-left:var(--bs-card-spacer-x)}.card-header{background-color:var(--bs-card-cap-bg);border-bottom:var(--bs-card-border-width) solid var(--bs-card-border-color);color:var(--bs-card-cap-color);margin-bottom:0;padding:var(--bs-card-cap-padding-y) var(--bs-card-cap-padding-x)}.card-header:first-child{border-radius:var(--bs-card-inner-border-radius) var(--bs-card-inner-border-radius) 0 0}.card-footer{background-color:var(--bs-card-cap-bg);border-top:var(--bs-card-border-width) solid var(--bs-card-border-color);color:var(--bs-card-cap-color);padding:var(--bs-card-cap-padding-y) var(--bs-card-cap-padding-x)}.card-footer:last-child{border-radius:0 0 var(--bs-card-inner-border-radius) var(--bs-card-inner-border-radius)}.card-header-tabs{border-bottom:0;margin-bottom:calc(var(--bs-card-cap-padding-y)*-1);margin-left:calc(var(--bs-card-cap-padding-x)*-.5);margin-right:calc(var(--bs-card-cap-padding-x)*-.5)}.card-header-tabs .nav-link.active{background-color:var(--bs-card-bg);border-bottom-color:var(--bs-card-bg)}.card-header-pills{margin-left:calc(var(--bs-card-cap-padding-x)*-.5);margin-right:calc(var(--bs-card-cap-padding-x)*-.5)}.card-img-overlay{border-radius:var(--bs-card-inner-border-radius);bottom:0;left:0;padding:var(--bs-card-img-overlay-padding);position:absolute;right:0;top:0}.card-img,.card-img-bottom,.card-img-top{width:100%}.card-img,.card-img-top{border-top-left-radius:var(--bs-card-inner-border-radius);border-top-right-radius:var(--bs-card-inner-border-radius)}.card-img,.card-img-bottom{border-bottom-left-radius:var(--bs-card-inner-border-radius);border-bottom-right-radius:var(--bs-card-inner-border-radius)}.card-group>.card{margin-bottom:var(--bs-card-group-margin)}@media (min-width:540px){.card-group{display:flex;flex-flow:row wrap}.card-group>.card{flex:1 0 0%;margin-bottom:0}.card-group>.card+.card{border-left:0;margin-left:0}.card-group>.card:not(:last-child){border-bottom-right-radius:0;border-top-right-radius:0}.card-group>.card:not(:last-child) .card-header,.card-group>.card:not(:last-child) .card-img-top{border-top-right-radius:0}.card-group>.card:not(:last-child) .card-footer,.card-group>.card:not(:last-child) .card-img-bottom{border-bottom-right-radius:0}.card-group>.card:not(:first-child){border-bottom-left-radius:0;border-top-left-radius:0}.card-group>.card:not(:first-child) .card-header,.card-group>.card:not(:first-child) .card-img-top{border-top-left-radius:0}.card-group>.card:not(:first-child) .card-footer,.card-group>.card:not(:first-child) .card-img-bottom{border-bottom-left-radius:0}}.accordion{--bs-accordion-color:var(--bs-body-color);--bs-accordion-bg:var(--bs-body-bg);--bs-accordion-transition:color 0.15s ease-in-out,background-color 0.15s ease-in-out,border-color 0.15s ease-in-out,box-shadow 0.15s ease-in-out,border-radius 0.15s ease;--bs-accordion-border-color:var(--bs-border-color);--bs-accordion-border-width:var(--bs-border-width);--bs-accordion-border-radius:var(--bs-border-radius);--bs-accordion-inner-border-radius:calc(var(--bs-border-radius) - var(--bs-border-width));--bs-accordion-btn-padding-x:1.25rem;--bs-accordion-btn-padding-y:1rem;--bs-accordion-btn-color:var(--bs-body-color);--bs-accordion-btn-bg:var(--bs-accordion-bg);--bs-accordion-btn-icon:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='none' stroke='%23212529' stroke-linecap='round' stroke-linejoin='round'%3E%3Cpath d='m2 5 6 6 6-6'/%3E%3C/svg%3E");--bs-accordion-btn-icon-width:1.25rem;--bs-accordion-btn-icon-transform:rotate(-180deg);--bs-accordion-btn-icon-transition:transform 0.2s ease-in-out;--bs-accordion-btn-active-icon:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='none' stroke='%23052c65' stroke-linecap='round' stroke-linejoin='round'%3E%3Cpath d='m2 5 6 6 6-6'/%3E%3C/svg%3E");--bs-accordion-btn-focus-box-shadow:0 0 0 0.1875rem var(--pst-color-accent);--bs-accordion-body-padding-x:1.25rem;--bs-accordion-body-padding-y:1rem;--bs-accordion-active-color:var(--bs-primary-text-emphasis);--bs-accordion-active-bg:var(--bs-primary-bg-subtle)}.accordion-button{align-items:center;background-color:var(--bs-accordion-btn-bg);border:0;border-radius:0;color:var(--bs-accordion-btn-color);display:flex;font-size:1rem;overflow-anchor:none;padding:var(--bs-accordion-btn-padding-y) var(--bs-accordion-btn-padding-x);position:relative;text-align:left;transition:var(--bs-accordion-transition);width:100%}@media (prefers-reduced-motion:reduce){.accordion-button{transition:none}}.accordion-button:not(.collapsed){background-color:var(--bs-accordion-active-bg);box-shadow:inset 0 calc(var(--bs-accordion-border-width)*-1) 0 var(--bs-accordion-border-color);color:var(--bs-accordion-active-color)}.accordion-button:not(.collapsed):after{background-image:var(--bs-accordion-btn-active-icon);transform:var(--bs-accordion-btn-icon-transform)}.accordion-button:after{background-image:var(--bs-accordion-btn-icon);background-repeat:no-repeat;background-size:var(--bs-accordion-btn-icon-width);content:"";flex-shrink:0;height:var(--bs-accordion-btn-icon-width);margin-left:auto;transition:var(--bs-accordion-btn-icon-transition);width:var(--bs-accordion-btn-icon-width)}@media (prefers-reduced-motion:reduce){.accordion-button:after{transition:none}}.accordion-button:hover{z-index:2}.accordion-button:focus{box-shadow:var(--bs-accordion-btn-focus-box-shadow);outline:0;z-index:3}.accordion-header{margin-bottom:0}.accordion-item{background-color:var(--bs-accordion-bg);border:var(--bs-accordion-border-width) solid var(--bs-accordion-border-color);color:var(--bs-accordion-color)}.accordion-item:first-of-type{border-top-left-radius:var(--bs-accordion-border-radius);border-top-right-radius:var(--bs-accordion-border-radius)}.accordion-item:first-of-type>.accordion-header .accordion-button{border-top-left-radius:var(--bs-accordion-inner-border-radius);border-top-right-radius:var(--bs-accordion-inner-border-radius)}.accordion-item:not(:first-of-type){border-top:0}.accordion-item:last-of-type{border-bottom-left-radius:var(--bs-accordion-border-radius);border-bottom-right-radius:var(--bs-accordion-border-radius)}.accordion-item:last-of-type>.accordion-header .accordion-button.collapsed{border-bottom-left-radius:var(--bs-accordion-inner-border-radius);border-bottom-right-radius:var(--bs-accordion-inner-border-radius)}.accordion-item:last-of-type>.accordion-collapse{border-bottom-left-radius:var(--bs-accordion-border-radius);border-bottom-right-radius:var(--bs-accordion-border-radius)}.accordion-body{padding:var(--bs-accordion-body-padding-y) var(--bs-accordion-body-padding-x)}.accordion-flush>.accordion-item{border-left:0;border-radius:0;border-right:0}.accordion-flush>.accordion-item:first-child{border-top:0}.accordion-flush>.accordion-item:last-child{border-bottom:0}.accordion-flush>.accordion-item>.accordion-collapse,.accordion-flush>.accordion-item>.accordion-header .accordion-button,.accordion-flush>.accordion-item>.accordion-header .accordion-button.collapsed{border-radius:0}[data-bs-theme=dark] .accordion-button:after{--bs-accordion-btn-icon:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='%236ea8fe'%3E%3Cpath fill-rule='evenodd' d='M1.646 4.646a.5.5 0 0 1 .708 0L8 10.293l5.646-5.647a.5.5 0 0 1 .708.708l-6 6a.5.5 0 0 1-.708 0l-6-6a.5.5 0 0 1 0-.708z'/%3E%3C/svg%3E");--bs-accordion-btn-active-icon:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='%236ea8fe'%3E%3Cpath fill-rule='evenodd' d='M1.646 4.646a.5.5 0 0 1 .708 0L8 10.293l5.646-5.647a.5.5 0 0 1 .708.708l-6 6a.5.5 0 0 1-.708 0l-6-6a.5.5 0 0 1 0-.708z'/%3E%3C/svg%3E")}.breadcrumb{--bs-breadcrumb-padding-x:0;--bs-breadcrumb-padding-y:0;--bs-breadcrumb-margin-bottom:1rem;--bs-breadcrumb-bg: ;--bs-breadcrumb-border-radius: ;--bs-breadcrumb-divider-color:var(--bs-secondary-color);--bs-breadcrumb-item-padding-x:0.5rem;--bs-breadcrumb-item-active-color:var(--bs-secondary-color);background-color:var(--bs-breadcrumb-bg);border-radius:var(--bs-breadcrumb-border-radius);display:flex;flex-wrap:wrap;font-size:var(--bs-breadcrumb-font-size);list-style:none;margin-bottom:var(--bs-breadcrumb-margin-bottom);padding:var(--bs-breadcrumb-padding-y) var(--bs-breadcrumb-padding-x)}.breadcrumb-item+.breadcrumb-item{padding-left:var(--bs-breadcrumb-item-padding-x)}.breadcrumb-item+.breadcrumb-item:before{color:var(--bs-breadcrumb-divider-color);content:var(--bs-breadcrumb-divider,"/");float:left;padding-right:var(--bs-breadcrumb-item-padding-x)}.breadcrumb-item.active{color:var(--bs-breadcrumb-item-active-color)}.pagination{--bs-pagination-padding-x:0.75rem;--bs-pagination-padding-y:0.375rem;--bs-pagination-font-size:1rem;--bs-pagination-color:var(--bs-link-color);--bs-pagination-bg:var(--bs-body-bg);--bs-pagination-border-width:var(--bs-border-width);--bs-pagination-border-color:var(--bs-border-color);--bs-pagination-border-radius:var(--bs-border-radius);--bs-pagination-hover-color:var(--bs-link-hover-color);--bs-pagination-hover-bg:var(--bs-tertiary-bg);--bs-pagination-hover-border-color:var(--bs-border-color);--bs-pagination-focus-color:var(--bs-link-hover-color);--bs-pagination-focus-bg:var(--bs-secondary-bg);--bs-pagination-focus-box-shadow:0 0 0 0.1875rem var(--pst-color-accent);--bs-pagination-active-color:#fff;--bs-pagination-active-bg:#0d6efd;--bs-pagination-active-border-color:#0d6efd;--bs-pagination-disabled-color:var(--bs-secondary-color);--bs-pagination-disabled-bg:var(--bs-secondary-bg);--bs-pagination-disabled-border-color:var(--bs-border-color);display:flex;list-style:none;padding-left:0}.page-link{background-color:var(--bs-pagination-bg);border:var(--bs-pagination-border-width) solid var(--bs-pagination-border-color);color:var(--bs-pagination-color);display:block;font-size:var(--bs-pagination-font-size);padding:var(--bs-pagination-padding-y) var(--bs-pagination-padding-x);position:relative;text-decoration:none;transition:color .15s ease-in-out,background-color .15s ease-in-out,border-color .15s ease-in-out,box-shadow .15s ease-in-out}@media (prefers-reduced-motion:reduce){.page-link{transition:none}}.page-link:hover{background-color:var(--bs-pagination-hover-bg);border-color:var(--bs-pagination-hover-border-color);color:var(--bs-pagination-hover-color);z-index:2}.page-link:focus{background-color:var(--bs-pagination-focus-bg);box-shadow:var(--bs-pagination-focus-box-shadow);color:var(--bs-pagination-focus-color);outline:0;z-index:3}.active>.page-link,.page-link.active{background-color:var(--bs-pagination-active-bg);border-color:var(--bs-pagination-active-border-color);color:var(--bs-pagination-active-color);z-index:3}.disabled>.page-link,.page-link.disabled{background-color:var(--bs-pagination-disabled-bg);border-color:var(--bs-pagination-disabled-border-color);color:var(--bs-pagination-disabled-color);pointer-events:none}.page-item:not(:first-child) .page-link{margin-left:calc(var(--bs-border-width)*-1)}.page-item:first-child .page-link{border-bottom-left-radius:var(--bs-pagination-border-radius);border-top-left-radius:var(--bs-pagination-border-radius)}.page-item:last-child .page-link{border-bottom-right-radius:var(--bs-pagination-border-radius);border-top-right-radius:var(--bs-pagination-border-radius)}.pagination-lg{--bs-pagination-padding-x:1.5rem;--bs-pagination-padding-y:0.75rem;--bs-pagination-font-size:1.25rem;--bs-pagination-border-radius:var(--bs-border-radius-lg)}.pagination-sm{--bs-pagination-padding-x:0.5rem;--bs-pagination-padding-y:0.25rem;--bs-pagination-font-size:0.875rem;--bs-pagination-border-radius:var(--bs-border-radius-sm)}.badge{--bs-badge-padding-x:0.65em;--bs-badge-padding-y:0.35em;--bs-badge-font-size:0.75em;--bs-badge-font-weight:700;--bs-badge-color:#fff;--bs-badge-border-radius:var(--bs-border-radius);border-radius:var(--bs-badge-border-radius);color:var(--bs-badge-color);display:inline-block;font-size:var(--bs-badge-font-size);font-weight:var(--bs-badge-font-weight);line-height:1;padding:var(--bs-badge-padding-y) var(--bs-badge-padding-x);text-align:center;vertical-align:baseline;white-space:nowrap}.badge:empty{display:none}.btn .badge{position:relative;top:-1px}.alert{--bs-alert-bg:transparent;--bs-alert-padding-x:1rem;--bs-alert-padding-y:1rem;--bs-alert-margin-bottom:1rem;--bs-alert-color:inherit;--bs-alert-border-color:transparent;--bs-alert-border:var(--bs-border-width) solid var(--bs-alert-border-color);--bs-alert-border-radius:var(--bs-border-radius);--bs-alert-link-color:inherit;background-color:var(--bs-alert-bg);border:var(--bs-alert-border);border-radius:var(--bs-alert-border-radius);color:var(--bs-alert-color);margin-bottom:var(--bs-alert-margin-bottom);padding:var(--bs-alert-padding-y) var(--bs-alert-padding-x);position:relative}.alert-heading{color:inherit}.alert-link{color:var(--bs-alert-link-color);font-weight:700}.alert-dismissible{padding-right:3rem}.alert-dismissible .btn-close{padding:1.25rem 1rem;position:absolute;right:0;top:0;z-index:2}.alert-primary{--bs-alert-color:var(--bs-primary-text-emphasis);--bs-alert-bg:var(--bs-primary-bg-subtle);--bs-alert-border-color:var(--bs-primary-border-subtle);--bs-alert-link-color:var(--bs-primary-text-emphasis)}.alert-secondary{--bs-alert-color:var(--bs-secondary-text-emphasis);--bs-alert-bg:var(--bs-secondary-bg-subtle);--bs-alert-border-color:var(--bs-secondary-border-subtle);--bs-alert-link-color:var(--bs-secondary-text-emphasis)}.alert-success{--bs-alert-color:var(--bs-success-text-emphasis);--bs-alert-bg:var(--bs-success-bg-subtle);--bs-alert-border-color:var(--bs-success-border-subtle);--bs-alert-link-color:var(--bs-success-text-emphasis)}.alert-info{--bs-alert-color:var(--bs-info-text-emphasis);--bs-alert-bg:var(--bs-info-bg-subtle);--bs-alert-border-color:var(--bs-info-border-subtle);--bs-alert-link-color:var(--bs-info-text-emphasis)}.alert-warning{--bs-alert-color:var(--bs-warning-text-emphasis);--bs-alert-bg:var(--bs-warning-bg-subtle);--bs-alert-border-color:var(--bs-warning-border-subtle);--bs-alert-link-color:var(--bs-warning-text-emphasis)}.alert-danger{--bs-alert-color:var(--bs-danger-text-emphasis);--bs-alert-bg:var(--bs-danger-bg-subtle);--bs-alert-border-color:var(--bs-danger-border-subtle);--bs-alert-link-color:var(--bs-danger-text-emphasis)}.alert-light{--bs-alert-color:var(--bs-light-text-emphasis);--bs-alert-bg:var(--bs-light-bg-subtle);--bs-alert-border-color:var(--bs-light-border-subtle);--bs-alert-link-color:var(--bs-light-text-emphasis)}.alert-dark{--bs-alert-color:var(--bs-dark-text-emphasis);--bs-alert-bg:var(--bs-dark-bg-subtle);--bs-alert-border-color:var(--bs-dark-border-subtle);--bs-alert-link-color:var(--bs-dark-text-emphasis)}@keyframes progress-bar-stripes{0%{background-position-x:1rem}}.progress,.progress-stacked{--bs-progress-height:1rem;--bs-progress-font-size:0.75rem;--bs-progress-bg:var(--bs-secondary-bg);--bs-progress-border-radius:var(--bs-border-radius);--bs-progress-box-shadow:var(--bs-box-shadow-inset);--bs-progress-bar-color:#fff;--bs-progress-bar-bg:#0d6efd;--bs-progress-bar-transition:width 0.6s ease;background-color:var(--bs-progress-bg);border-radius:var(--bs-progress-border-radius);display:flex;font-size:var(--bs-progress-font-size);height:var(--bs-progress-height);overflow:hidden}.progress-bar{background-color:var(--bs-progress-bar-bg);color:var(--bs-progress-bar-color);display:flex;flex-direction:column;justify-content:center;overflow:hidden;text-align:center;transition:var(--bs-progress-bar-transition);white-space:nowrap}@media (prefers-reduced-motion:reduce){.progress-bar{transition:none}}.progress-bar-striped{background-image:linear-gradient(45deg,hsla(0,0%,100%,.15) 25%,transparent 0,transparent 50%,hsla(0,0%,100%,.15) 0,hsla(0,0%,100%,.15) 75%,transparent 0,transparent);background-size:var(--bs-progress-height) var(--bs-progress-height)}.progress-stacked>.progress{overflow:visible}.progress-stacked>.progress>.progress-bar{width:100%}.progress-bar-animated{animation:progress-bar-stripes 1s linear infinite}@media (prefers-reduced-motion:reduce){.progress-bar-animated{animation:none}}.list-group{--bs-list-group-color:var(--bs-body-color);--bs-list-group-bg:var(--bs-body-bg);--bs-list-group-border-color:var(--bs-border-color);--bs-list-group-border-width:var(--bs-border-width);--bs-list-group-border-radius:var(--bs-border-radius);--bs-list-group-item-padding-x:1rem;--bs-list-group-item-padding-y:0.5rem;--bs-list-group-action-color:var(--bs-secondary-color);--bs-list-group-action-hover-color:var(--bs-emphasis-color);--bs-list-group-action-hover-bg:var(--bs-tertiary-bg);--bs-list-group-action-active-color:var(--bs-body-color);--bs-list-group-action-active-bg:var(--bs-secondary-bg);--bs-list-group-disabled-color:var(--bs-secondary-color);--bs-list-group-disabled-bg:var(--bs-body-bg);--bs-list-group-active-color:#fff;--bs-list-group-active-bg:#0d6efd;--bs-list-group-active-border-color:#0d6efd;border-radius:var(--bs-list-group-border-radius);display:flex;flex-direction:column;margin-bottom:0;padding-left:0}.list-group-numbered{counter-reset:section;list-style-type:none}.list-group-numbered>.list-group-item:before{content:counters(section,".") ". ";counter-increment:section}.list-group-item-action{color:var(--bs-list-group-action-color);text-align:inherit;width:100%}.list-group-item-action:focus,.list-group-item-action:hover{background-color:var(--bs-list-group-action-hover-bg);color:var(--bs-list-group-action-hover-color);text-decoration:none;z-index:1}.list-group-item-action:active{background-color:var(--bs-list-group-action-active-bg);color:var(--bs-list-group-action-active-color)}.list-group-item{background-color:var(--bs-list-group-bg);border:var(--bs-list-group-border-width) solid var(--bs-list-group-border-color);color:var(--bs-list-group-color);display:block;padding:var(--bs-list-group-item-padding-y) var(--bs-list-group-item-padding-x);position:relative;text-decoration:none}.list-group-item:first-child{border-top-left-radius:inherit;border-top-right-radius:inherit}.list-group-item:last-child{border-bottom-left-radius:inherit;border-bottom-right-radius:inherit}.list-group-item.disabled,.list-group-item:disabled{background-color:var(--bs-list-group-disabled-bg);color:var(--bs-list-group-disabled-color);pointer-events:none}.list-group-item.active{background-color:var(--bs-list-group-active-bg);border-color:var(--bs-list-group-active-border-color);color:var(--bs-list-group-active-color);z-index:2}.list-group-item+.list-group-item{border-top-width:0}.list-group-item+.list-group-item.active{border-top-width:var(--bs-list-group-border-width);margin-top:calc(var(--bs-list-group-border-width)*-1)}.list-group-horizontal{flex-direction:row}.list-group-horizontal>.list-group-item:first-child:not(:last-child){border-bottom-left-radius:var(--bs-list-group-border-radius);border-top-right-radius:0}.list-group-horizontal>.list-group-item:last-child:not(:first-child){border-bottom-left-radius:0;border-top-right-radius:var(--bs-list-group-border-radius)}.list-group-horizontal>.list-group-item.active{margin-top:0}.list-group-horizontal>.list-group-item+.list-group-item{border-left-width:0;border-top-width:var(--bs-list-group-border-width)}.list-group-horizontal>.list-group-item+.list-group-item.active{border-left-width:var(--bs-list-group-border-width);margin-left:calc(var(--bs-list-group-border-width)*-1)}@media (min-width:540px){.list-group-horizontal-sm{flex-direction:row}.list-group-horizontal-sm>.list-group-item:first-child:not(:last-child){border-bottom-left-radius:var(--bs-list-group-border-radius);border-top-right-radius:0}.list-group-horizontal-sm>.list-group-item:last-child:not(:first-child){border-bottom-left-radius:0;border-top-right-radius:var(--bs-list-group-border-radius)}.list-group-horizontal-sm>.list-group-item.active{margin-top:0}.list-group-horizontal-sm>.list-group-item+.list-group-item{border-left-width:0;border-top-width:var(--bs-list-group-border-width)}.list-group-horizontal-sm>.list-group-item+.list-group-item.active{border-left-width:var(--bs-list-group-border-width);margin-left:calc(var(--bs-list-group-border-width)*-1)}}@media (min-width:720px){.list-group-horizontal-md{flex-direction:row}.list-group-horizontal-md>.list-group-item:first-child:not(:last-child){border-bottom-left-radius:var(--bs-list-group-border-radius);border-top-right-radius:0}.list-group-horizontal-md>.list-group-item:last-child:not(:first-child){border-bottom-left-radius:0;border-top-right-radius:var(--bs-list-group-border-radius)}.list-group-horizontal-md>.list-group-item.active{margin-top:0}.list-group-horizontal-md>.list-group-item+.list-group-item{border-left-width:0;border-top-width:var(--bs-list-group-border-width)}.list-group-horizontal-md>.list-group-item+.list-group-item.active{border-left-width:var(--bs-list-group-border-width);margin-left:calc(var(--bs-list-group-border-width)*-1)}}@media (min-width:960px){.list-group-horizontal-lg{flex-direction:row}.list-group-horizontal-lg>.list-group-item:first-child:not(:last-child){border-bottom-left-radius:var(--bs-list-group-border-radius);border-top-right-radius:0}.list-group-horizontal-lg>.list-group-item:last-child:not(:first-child){border-bottom-left-radius:0;border-top-right-radius:var(--bs-list-group-border-radius)}.list-group-horizontal-lg>.list-group-item.active{margin-top:0}.list-group-horizontal-lg>.list-group-item+.list-group-item{border-left-width:0;border-top-width:var(--bs-list-group-border-width)}.list-group-horizontal-lg>.list-group-item+.list-group-item.active{border-left-width:var(--bs-list-group-border-width);margin-left:calc(var(--bs-list-group-border-width)*-1)}}@media (min-width:1200px){.list-group-horizontal-xl{flex-direction:row}.list-group-horizontal-xl>.list-group-item:first-child:not(:last-child){border-bottom-left-radius:var(--bs-list-group-border-radius);border-top-right-radius:0}.list-group-horizontal-xl>.list-group-item:last-child:not(:first-child){border-bottom-left-radius:0;border-top-right-radius:var(--bs-list-group-border-radius)}.list-group-horizontal-xl>.list-group-item.active{margin-top:0}.list-group-horizontal-xl>.list-group-item+.list-group-item{border-left-width:0;border-top-width:var(--bs-list-group-border-width)}.list-group-horizontal-xl>.list-group-item+.list-group-item.active{border-left-width:var(--bs-list-group-border-width);margin-left:calc(var(--bs-list-group-border-width)*-1)}}.list-group-flush{border-radius:0}.list-group-flush>.list-group-item{border-width:0 0 var(--bs-list-group-border-width)}.list-group-flush>.list-group-item:last-child{border-bottom-width:0}.list-group-item-primary{--bs-list-group-color:var(--bs-primary-text-emphasis);--bs-list-group-bg:var(--bs-primary-bg-subtle);--bs-list-group-border-color:var(--bs-primary-border-subtle);--bs-list-group-action-hover-color:var(--bs-emphasis-color);--bs-list-group-action-hover-bg:var(--bs-primary-border-subtle);--bs-list-group-action-active-color:var(--bs-emphasis-color);--bs-list-group-action-active-bg:var(--bs-primary-border-subtle);--bs-list-group-active-color:var(--bs-primary-bg-subtle);--bs-list-group-active-bg:var(--bs-primary-text-emphasis);--bs-list-group-active-border-color:var(--bs-primary-text-emphasis)}.list-group-item-secondary{--bs-list-group-color:var(--bs-secondary-text-emphasis);--bs-list-group-bg:var(--bs-secondary-bg-subtle);--bs-list-group-border-color:var(--bs-secondary-border-subtle);--bs-list-group-action-hover-color:var(--bs-emphasis-color);--bs-list-group-action-hover-bg:var(--bs-secondary-border-subtle);--bs-list-group-action-active-color:var(--bs-emphasis-color);--bs-list-group-action-active-bg:var(--bs-secondary-border-subtle);--bs-list-group-active-color:var(--bs-secondary-bg-subtle);--bs-list-group-active-bg:var(--bs-secondary-text-emphasis);--bs-list-group-active-border-color:var(--bs-secondary-text-emphasis)}.list-group-item-success{--bs-list-group-color:var(--bs-success-text-emphasis);--bs-list-group-bg:var(--bs-success-bg-subtle);--bs-list-group-border-color:var(--bs-success-border-subtle);--bs-list-group-action-hover-color:var(--bs-emphasis-color);--bs-list-group-action-hover-bg:var(--bs-success-border-subtle);--bs-list-group-action-active-color:var(--bs-emphasis-color);--bs-list-group-action-active-bg:var(--bs-success-border-subtle);--bs-list-group-active-color:var(--bs-success-bg-subtle);--bs-list-group-active-bg:var(--bs-success-text-emphasis);--bs-list-group-active-border-color:var(--bs-success-text-emphasis)}.list-group-item-info{--bs-list-group-color:var(--bs-info-text-emphasis);--bs-list-group-bg:var(--bs-info-bg-subtle);--bs-list-group-border-color:var(--bs-info-border-subtle);--bs-list-group-action-hover-color:var(--bs-emphasis-color);--bs-list-group-action-hover-bg:var(--bs-info-border-subtle);--bs-list-group-action-active-color:var(--bs-emphasis-color);--bs-list-group-action-active-bg:var(--bs-info-border-subtle);--bs-list-group-active-color:var(--bs-info-bg-subtle);--bs-list-group-active-bg:var(--bs-info-text-emphasis);--bs-list-group-active-border-color:var(--bs-info-text-emphasis)}.list-group-item-warning{--bs-list-group-color:var(--bs-warning-text-emphasis);--bs-list-group-bg:var(--bs-warning-bg-subtle);--bs-list-group-border-color:var(--bs-warning-border-subtle);--bs-list-group-action-hover-color:var(--bs-emphasis-color);--bs-list-group-action-hover-bg:var(--bs-warning-border-subtle);--bs-list-group-action-active-color:var(--bs-emphasis-color);--bs-list-group-action-active-bg:var(--bs-warning-border-subtle);--bs-list-group-active-color:var(--bs-warning-bg-subtle);--bs-list-group-active-bg:var(--bs-warning-text-emphasis);--bs-list-group-active-border-color:var(--bs-warning-text-emphasis)}.list-group-item-danger{--bs-list-group-color:var(--bs-danger-text-emphasis);--bs-list-group-bg:var(--bs-danger-bg-subtle);--bs-list-group-border-color:var(--bs-danger-border-subtle);--bs-list-group-action-hover-color:var(--bs-emphasis-color);--bs-list-group-action-hover-bg:var(--bs-danger-border-subtle);--bs-list-group-action-active-color:var(--bs-emphasis-color);--bs-list-group-action-active-bg:var(--bs-danger-border-subtle);--bs-list-group-active-color:var(--bs-danger-bg-subtle);--bs-list-group-active-bg:var(--bs-danger-text-emphasis);--bs-list-group-active-border-color:var(--bs-danger-text-emphasis)}.list-group-item-light{--bs-list-group-color:var(--bs-light-text-emphasis);--bs-list-group-bg:var(--bs-light-bg-subtle);--bs-list-group-border-color:var(--bs-light-border-subtle);--bs-list-group-action-hover-color:var(--bs-emphasis-color);--bs-list-group-action-hover-bg:var(--bs-light-border-subtle);--bs-list-group-action-active-color:var(--bs-emphasis-color);--bs-list-group-action-active-bg:var(--bs-light-border-subtle);--bs-list-group-active-color:var(--bs-light-bg-subtle);--bs-list-group-active-bg:var(--bs-light-text-emphasis);--bs-list-group-active-border-color:var(--bs-light-text-emphasis)}.list-group-item-dark{--bs-list-group-color:var(--bs-dark-text-emphasis);--bs-list-group-bg:var(--bs-dark-bg-subtle);--bs-list-group-border-color:var(--bs-dark-border-subtle);--bs-list-group-action-hover-color:var(--bs-emphasis-color);--bs-list-group-action-hover-bg:var(--bs-dark-border-subtle);--bs-list-group-action-active-color:var(--bs-emphasis-color);--bs-list-group-action-active-bg:var(--bs-dark-border-subtle);--bs-list-group-active-color:var(--bs-dark-bg-subtle);--bs-list-group-active-bg:var(--bs-dark-text-emphasis);--bs-list-group-active-border-color:var(--bs-dark-text-emphasis)}.btn-close{--bs-btn-close-color:#000;--bs-btn-close-bg:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16'%3E%3Cpath d='M.293.293a1 1 0 0 1 1.414 0L8 6.586 14.293.293a1 1 0 1 1 1.414 1.414L9.414 8l6.293 6.293a1 1 0 0 1-1.414 1.414L8 9.414l-6.293 6.293a1 1 0 0 1-1.414-1.414L6.586 8 .293 1.707a1 1 0 0 1 0-1.414z'/%3E%3C/svg%3E");--bs-btn-close-opacity:0.5;--bs-btn-close-hover-opacity:0.75;--bs-btn-close-focus-shadow:0 0 0 0.1875rem var(--pst-color-accent);--bs-btn-close-focus-opacity:1;--bs-btn-close-disabled-opacity:0.25;--bs-btn-close-white-filter:invert(1) grayscale(100%) brightness(200%);background:transparent var(--bs-btn-close-bg) center/1em auto no-repeat;border:0;border-radius:.375rem;box-sizing:content-box;height:1em;opacity:var(--bs-btn-close-opacity);padding:.25em;width:1em}.btn-close,.btn-close:hover{color:var(--bs-btn-close-color)}.btn-close:hover{opacity:var(--bs-btn-close-hover-opacity);text-decoration:none}.btn-close:focus{box-shadow:var(--bs-btn-close-focus-shadow);opacity:var(--bs-btn-close-focus-opacity);outline:0}.btn-close.disabled,.btn-close:disabled{opacity:var(--bs-btn-close-disabled-opacity);pointer-events:none;user-select:none}.btn-close-white,[data-bs-theme=dark] .btn-close{filter:var(--bs-btn-close-white-filter)}.toast{--bs-toast-zindex:1090;--bs-toast-padding-x:0.75rem;--bs-toast-padding-y:0.5rem;--bs-toast-spacing:1.5rem;--bs-toast-max-width:350px;--bs-toast-font-size:0.875rem;--bs-toast-color: ;--bs-toast-bg:rgba(var(--bs-body-bg-rgb),0.85);--bs-toast-border-width:var(--bs-border-width);--bs-toast-border-color:var(--bs-border-color-translucent);--bs-toast-border-radius:var(--bs-border-radius);--bs-toast-box-shadow:var(--bs-box-shadow);--bs-toast-header-color:var(--bs-secondary-color);--bs-toast-header-bg:rgba(var(--bs-body-bg-rgb),0.85);--bs-toast-header-border-color:var(--bs-border-color-translucent);background-clip:padding-box;background-color:var(--bs-toast-bg);border:var(--bs-toast-border-width) solid var(--bs-toast-border-color);border-radius:var(--bs-toast-border-radius);box-shadow:var(--bs-toast-box-shadow);color:var(--bs-toast-color);font-size:var(--bs-toast-font-size);max-width:100%;pointer-events:auto;width:var(--bs-toast-max-width)}.toast.showing{opacity:0}.toast:not(.show){display:none}.toast-container{--bs-toast-zindex:1090;max-width:100%;pointer-events:none;position:absolute;width:max-content;z-index:var(--bs-toast-zindex)}.toast-container>:not(:last-child){margin-bottom:var(--bs-toast-spacing)}.toast-header{align-items:center;background-clip:padding-box;background-color:var(--bs-toast-header-bg);border-bottom:var(--bs-toast-border-width) solid var(--bs-toast-header-border-color);border-top-left-radius:calc(var(--bs-toast-border-radius) - var(--bs-toast-border-width));border-top-right-radius:calc(var(--bs-toast-border-radius) - var(--bs-toast-border-width));color:var(--bs-toast-header-color);display:flex;padding:var(--bs-toast-padding-y) var(--bs-toast-padding-x)}.toast-header .btn-close{margin-left:var(--bs-toast-padding-x);margin-right:calc(var(--bs-toast-padding-x)*-.5)}.toast-body{word-wrap:break-word;padding:var(--bs-toast-padding-x)}.modal{--bs-modal-zindex:1055;--bs-modal-width:500px;--bs-modal-padding:1rem;--bs-modal-margin:0.5rem;--bs-modal-color: ;--bs-modal-bg:var(--bs-body-bg);--bs-modal-border-color:var(--bs-border-color-translucent);--bs-modal-border-width:var(--bs-border-width);--bs-modal-border-radius:var(--bs-border-radius-lg);--bs-modal-box-shadow:var(--bs-box-shadow-sm);--bs-modal-inner-border-radius:calc(var(--bs-border-radius-lg) - var(--bs-border-width));--bs-modal-header-padding-x:1rem;--bs-modal-header-padding-y:1rem;--bs-modal-header-padding:1rem 1rem;--bs-modal-header-border-color:var(--bs-border-color);--bs-modal-header-border-width:var(--bs-border-width);--bs-modal-title-line-height:1.5;--bs-modal-footer-gap:0.5rem;--bs-modal-footer-bg: ;--bs-modal-footer-border-color:var(--bs-border-color);--bs-modal-footer-border-width:var(--bs-border-width);display:none;height:100%;left:0;outline:0;overflow-x:hidden;overflow-y:auto;position:fixed;top:0;width:100%;z-index:var(--bs-modal-zindex)}.modal-dialog{margin:var(--bs-modal-margin);pointer-events:none;position:relative;width:auto}.modal.fade .modal-dialog{transform:translateY(-50px);transition:transform .3s ease-out}@media (prefers-reduced-motion:reduce){.modal.fade .modal-dialog{transition:none}}.modal.show .modal-dialog{transform:none}.modal.modal-static .modal-dialog{transform:scale(1.02)}.modal-dialog-scrollable{height:calc(100% - var(--bs-modal-margin)*2)}.modal-dialog-scrollable .modal-content{max-height:100%;overflow:hidden}.modal-dialog-scrollable .modal-body{overflow-y:auto}.modal-dialog-centered{align-items:center;display:flex;min-height:calc(100% - var(--bs-modal-margin)*2)}.modal-content{background-clip:padding-box;background-color:var(--bs-modal-bg);border:var(--bs-modal-border-width) solid var(--bs-modal-border-color);border-radius:var(--bs-modal-border-radius);color:var(--bs-modal-color);display:flex;flex-direction:column;outline:0;pointer-events:auto;position:relative;width:100%}.modal-backdrop{--bs-backdrop-zindex:1050;--bs-backdrop-bg:#000;--bs-backdrop-opacity:0.5;background-color:var(--bs-backdrop-bg);height:100vh;left:0;position:fixed;top:0;width:100vw;z-index:var(--bs-backdrop-zindex)}.modal-backdrop.fade{opacity:0}.modal-backdrop.show{opacity:var(--bs-backdrop-opacity)}.modal-header{align-items:center;border-bottom:var(--bs-modal-header-border-width) solid var(--bs-modal-header-border-color);border-top-left-radius:var(--bs-modal-inner-border-radius);border-top-right-radius:var(--bs-modal-inner-border-radius);display:flex;flex-shrink:0;padding:var(--bs-modal-header-padding)}.modal-header .btn-close{margin:calc(var(--bs-modal-header-padding-y)*-.5) calc(var(--bs-modal-header-padding-x)*-.5) calc(var(--bs-modal-header-padding-y)*-.5) auto;padding:calc(var(--bs-modal-header-padding-y)*.5) calc(var(--bs-modal-header-padding-x)*.5)}.modal-title{line-height:var(--bs-modal-title-line-height);margin-bottom:0}.modal-body{flex:1 1 auto;padding:var(--bs-modal-padding);position:relative}.modal-footer{align-items:center;background-color:var(--bs-modal-footer-bg);border-bottom-left-radius:var(--bs-modal-inner-border-radius);border-bottom-right-radius:var(--bs-modal-inner-border-radius);border-top:var(--bs-modal-footer-border-width) solid var(--bs-modal-footer-border-color);display:flex;flex-shrink:0;flex-wrap:wrap;justify-content:flex-end;padding:calc(var(--bs-modal-padding) - var(--bs-modal-footer-gap)*.5)}.modal-footer>*{margin:calc(var(--bs-modal-footer-gap)*.5)}@media (min-width:540px){.modal{--bs-modal-margin:1.75rem;--bs-modal-box-shadow:var(--bs-box-shadow)}.modal-dialog{margin-left:auto;margin-right:auto;max-width:var(--bs-modal-width)}.modal-sm{--bs-modal-width:300px}}@media (min-width:960px){.modal-lg,.modal-xl{--bs-modal-width:800px}}@media (min-width:1200px){.modal-xl{--bs-modal-width:1140px}}.modal-fullscreen{height:100%;margin:0;max-width:none;width:100vw}.modal-fullscreen .modal-content{border:0;border-radius:0;height:100%}.modal-fullscreen .modal-footer,.modal-fullscreen .modal-header{border-radius:0}.modal-fullscreen .modal-body{overflow-y:auto}@media (max-width:539.98px){.modal-fullscreen-sm-down{height:100%;margin:0;max-width:none;width:100vw}.modal-fullscreen-sm-down .modal-content{border:0;border-radius:0;height:100%}.modal-fullscreen-sm-down .modal-footer,.modal-fullscreen-sm-down .modal-header{border-radius:0}.modal-fullscreen-sm-down .modal-body{overflow-y:auto}}@media (max-width:719.98px){.modal-fullscreen-md-down{height:100%;margin:0;max-width:none;width:100vw}.modal-fullscreen-md-down .modal-content{border:0;border-radius:0;height:100%}.modal-fullscreen-md-down .modal-footer,.modal-fullscreen-md-down .modal-header{border-radius:0}.modal-fullscreen-md-down .modal-body{overflow-y:auto}}@media (max-width:959.98px){.modal-fullscreen-lg-down{height:100%;margin:0;max-width:none;width:100vw}.modal-fullscreen-lg-down .modal-content{border:0;border-radius:0;height:100%}.modal-fullscreen-lg-down .modal-footer,.modal-fullscreen-lg-down .modal-header{border-radius:0}.modal-fullscreen-lg-down .modal-body{overflow-y:auto}}@media (max-width:1199.98px){.modal-fullscreen-xl-down{height:100%;margin:0;max-width:none;width:100vw}.modal-fullscreen-xl-down .modal-content{border:0;border-radius:0;height:100%}.modal-fullscreen-xl-down .modal-footer,.modal-fullscreen-xl-down .modal-header{border-radius:0}.modal-fullscreen-xl-down .modal-body{overflow-y:auto}}.tooltip{--bs-tooltip-zindex:1080;--bs-tooltip-max-width:200px;--bs-tooltip-padding-x:0.5rem;--bs-tooltip-padding-y:0.25rem;--bs-tooltip-margin: ;--bs-tooltip-font-size:0.875rem;--bs-tooltip-color:var(--bs-body-bg);--bs-tooltip-bg:var(--bs-emphasis-color);--bs-tooltip-border-radius:var(--bs-border-radius);--bs-tooltip-opacity:0.9;--bs-tooltip-arrow-width:0.8rem;--bs-tooltip-arrow-height:0.4rem;word-wrap:break-word;display:block;font-family:var(--bs-font-sans-serif);font-size:var(--bs-tooltip-font-size);font-style:normal;font-weight:400;letter-spacing:normal;line-break:auto;line-height:1.5;margin:var(--bs-tooltip-margin);opacity:0;text-align:left;text-align:start;text-decoration:none;text-shadow:none;text-transform:none;white-space:normal;word-break:normal;word-spacing:normal;z-index:var(--bs-tooltip-zindex)}.tooltip.show{opacity:var(--bs-tooltip-opacity)}.tooltip .tooltip-arrow{display:block;height:var(--bs-tooltip-arrow-height);width:var(--bs-tooltip-arrow-width)}.tooltip .tooltip-arrow:before{border-color:transparent;border-style:solid;content:"";position:absolute}.bs-tooltip-auto[data-popper-placement^=top] .tooltip-arrow,.bs-tooltip-top .tooltip-arrow{bottom:calc(var(--bs-tooltip-arrow-height)*-1)}.bs-tooltip-auto[data-popper-placement^=top] .tooltip-arrow:before,.bs-tooltip-top .tooltip-arrow:before{border-top-color:var(--bs-tooltip-bg);border-width:var(--bs-tooltip-arrow-height) calc(var(--bs-tooltip-arrow-width)*.5) 0;top:-1px}.bs-tooltip-auto[data-popper-placement^=right] .tooltip-arrow,.bs-tooltip-end .tooltip-arrow{height:var(--bs-tooltip-arrow-width);left:calc(var(--bs-tooltip-arrow-height)*-1);width:var(--bs-tooltip-arrow-height)}.bs-tooltip-auto[data-popper-placement^=right] .tooltip-arrow:before,.bs-tooltip-end .tooltip-arrow:before{border-right-color:var(--bs-tooltip-bg);border-width:calc(var(--bs-tooltip-arrow-width)*.5) var(--bs-tooltip-arrow-height) calc(var(--bs-tooltip-arrow-width)*.5) 0;right:-1px}.bs-tooltip-auto[data-popper-placement^=bottom] .tooltip-arrow,.bs-tooltip-bottom .tooltip-arrow{top:calc(var(--bs-tooltip-arrow-height)*-1)}.bs-tooltip-auto[data-popper-placement^=bottom] .tooltip-arrow:before,.bs-tooltip-bottom .tooltip-arrow:before{border-bottom-color:var(--bs-tooltip-bg);border-width:0 calc(var(--bs-tooltip-arrow-width)*.5) var(--bs-tooltip-arrow-height);bottom:-1px}.bs-tooltip-auto[data-popper-placement^=left] .tooltip-arrow,.bs-tooltip-start .tooltip-arrow{height:var(--bs-tooltip-arrow-width);right:calc(var(--bs-tooltip-arrow-height)*-1);width:var(--bs-tooltip-arrow-height)}.bs-tooltip-auto[data-popper-placement^=left] .tooltip-arrow:before,.bs-tooltip-start .tooltip-arrow:before{border-left-color:var(--bs-tooltip-bg);border-width:calc(var(--bs-tooltip-arrow-width)*.5) 0 calc(var(--bs-tooltip-arrow-width)*.5) var(--bs-tooltip-arrow-height);left:-1px}.tooltip-inner{background-color:var(--bs-tooltip-bg);border-radius:var(--bs-tooltip-border-radius);color:var(--bs-tooltip-color);max-width:var(--bs-tooltip-max-width);padding:var(--bs-tooltip-padding-y) var(--bs-tooltip-padding-x);text-align:center}.popover{--bs-popover-zindex:1070;--bs-popover-max-width:276px;--bs-popover-font-size:0.875rem;--bs-popover-bg:var(--bs-body-bg);--bs-popover-border-width:var(--bs-border-width);--bs-popover-border-color:var(--bs-border-color-translucent);--bs-popover-border-radius:var(--bs-border-radius-lg);--bs-popover-inner-border-radius:calc(var(--bs-border-radius-lg) - var(--bs-border-width));--bs-popover-box-shadow:var(--bs-box-shadow);--bs-popover-header-padding-x:1rem;--bs-popover-header-padding-y:0.5rem;--bs-popover-header-font-size:1rem;--bs-popover-header-color:inherit;--bs-popover-header-bg:var(--bs-secondary-bg);--bs-popover-body-padding-x:1rem;--bs-popover-body-padding-y:1rem;--bs-popover-body-color:var(--bs-body-color);--bs-popover-arrow-width:1rem;--bs-popover-arrow-height:0.5rem;--bs-popover-arrow-border:var(--bs-popover-border-color);word-wrap:break-word;background-clip:padding-box;background-color:var(--bs-popover-bg);border:var(--bs-popover-border-width) solid var(--bs-popover-border-color);border-radius:var(--bs-popover-border-radius);display:block;font-family:var(--bs-font-sans-serif);font-size:var(--bs-popover-font-size);font-style:normal;font-weight:400;letter-spacing:normal;line-break:auto;line-height:1.5;max-width:var(--bs-popover-max-width);text-align:left;text-align:start;text-decoration:none;text-shadow:none;text-transform:none;white-space:normal;word-break:normal;word-spacing:normal;z-index:var(--bs-popover-zindex)}.popover .popover-arrow{display:block;height:var(--bs-popover-arrow-height);width:var(--bs-popover-arrow-width)}.popover .popover-arrow:after,.popover .popover-arrow:before{border:0 solid transparent;content:"";display:block;position:absolute}.bs-popover-auto[data-popper-placement^=top]>.popover-arrow,.bs-popover-top>.popover-arrow{bottom:calc((var(--bs-popover-arrow-height))*-1 - var(--bs-popover-border-width))}.bs-popover-auto[data-popper-placement^=top]>.popover-arrow:after,.bs-popover-auto[data-popper-placement^=top]>.popover-arrow:before,.bs-popover-top>.popover-arrow:after,.bs-popover-top>.popover-arrow:before{border-width:var(--bs-popover-arrow-height) calc(var(--bs-popover-arrow-width)*.5) 0}.bs-popover-auto[data-popper-placement^=top]>.popover-arrow:before,.bs-popover-top>.popover-arrow:before{border-top-color:var(--bs-popover-arrow-border);bottom:0}.bs-popover-auto[data-popper-placement^=top]>.popover-arrow:after,.bs-popover-top>.popover-arrow:after{border-top-color:var(--bs-popover-bg);bottom:var(--bs-popover-border-width)}.bs-popover-auto[data-popper-placement^=right]>.popover-arrow,.bs-popover-end>.popover-arrow{height:var(--bs-popover-arrow-width);left:calc((var(--bs-popover-arrow-height))*-1 - var(--bs-popover-border-width));width:var(--bs-popover-arrow-height)}.bs-popover-auto[data-popper-placement^=right]>.popover-arrow:after,.bs-popover-auto[data-popper-placement^=right]>.popover-arrow:before,.bs-popover-end>.popover-arrow:after,.bs-popover-end>.popover-arrow:before{border-width:calc(var(--bs-popover-arrow-width)*.5) var(--bs-popover-arrow-height) calc(var(--bs-popover-arrow-width)*.5) 0}.bs-popover-auto[data-popper-placement^=right]>.popover-arrow:before,.bs-popover-end>.popover-arrow:before{border-right-color:var(--bs-popover-arrow-border);left:0}.bs-popover-auto[data-popper-placement^=right]>.popover-arrow:after,.bs-popover-end>.popover-arrow:after{border-right-color:var(--bs-popover-bg);left:var(--bs-popover-border-width)}.bs-popover-auto[data-popper-placement^=bottom]>.popover-arrow,.bs-popover-bottom>.popover-arrow{top:calc((var(--bs-popover-arrow-height))*-1 - var(--bs-popover-border-width))}.bs-popover-auto[data-popper-placement^=bottom]>.popover-arrow:after,.bs-popover-auto[data-popper-placement^=bottom]>.popover-arrow:before,.bs-popover-bottom>.popover-arrow:after,.bs-popover-bottom>.popover-arrow:before{border-width:0 calc(var(--bs-popover-arrow-width)*.5) var(--bs-popover-arrow-height)}.bs-popover-auto[data-popper-placement^=bottom]>.popover-arrow:before,.bs-popover-bottom>.popover-arrow:before{border-bottom-color:var(--bs-popover-arrow-border);top:0}.bs-popover-auto[data-popper-placement^=bottom]>.popover-arrow:after,.bs-popover-bottom>.popover-arrow:after{border-bottom-color:var(--bs-popover-bg);top:var(--bs-popover-border-width)}.bs-popover-auto[data-popper-placement^=bottom] .popover-header:before,.bs-popover-bottom .popover-header:before{border-bottom:var(--bs-popover-border-width) solid var(--bs-popover-header-bg);content:"";display:block;left:50%;margin-left:calc(var(--bs-popover-arrow-width)*-.5);position:absolute;top:0;width:var(--bs-popover-arrow-width)}.bs-popover-auto[data-popper-placement^=left]>.popover-arrow,.bs-popover-start>.popover-arrow{height:var(--bs-popover-arrow-width);right:calc((var(--bs-popover-arrow-height))*-1 - var(--bs-popover-border-width));width:var(--bs-popover-arrow-height)}.bs-popover-auto[data-popper-placement^=left]>.popover-arrow:after,.bs-popover-auto[data-popper-placement^=left]>.popover-arrow:before,.bs-popover-start>.popover-arrow:after,.bs-popover-start>.popover-arrow:before{border-width:calc(var(--bs-popover-arrow-width)*.5) 0 calc(var(--bs-popover-arrow-width)*.5) var(--bs-popover-arrow-height)}.bs-popover-auto[data-popper-placement^=left]>.popover-arrow:before,.bs-popover-start>.popover-arrow:before{border-left-color:var(--bs-popover-arrow-border);right:0}.bs-popover-auto[data-popper-placement^=left]>.popover-arrow:after,.bs-popover-start>.popover-arrow:after{border-left-color:var(--bs-popover-bg);right:var(--bs-popover-border-width)}.popover-header{background-color:var(--bs-popover-header-bg);border-bottom:var(--bs-popover-border-width) solid var(--bs-popover-border-color);border-top-left-radius:var(--bs-popover-inner-border-radius);border-top-right-radius:var(--bs-popover-inner-border-radius);color:var(--bs-popover-header-color);font-size:var(--bs-popover-header-font-size);margin-bottom:0;padding:var(--bs-popover-header-padding-y) var(--bs-popover-header-padding-x)}.popover-header:empty{display:none}.popover-body{color:var(--bs-popover-body-color);padding:var(--bs-popover-body-padding-y) var(--bs-popover-body-padding-x)}.carousel{position:relative}.carousel.pointer-event{touch-action:pan-y}.carousel-inner{overflow:hidden;position:relative;width:100%}.carousel-inner:after{clear:both;content:"";display:block}.carousel-item{backface-visibility:hidden;display:none;float:left;margin-right:-100%;position:relative;transition:transform .6s ease-in-out;width:100%}@media (prefers-reduced-motion:reduce){.carousel-item{transition:none}}.carousel-item-next,.carousel-item-prev,.carousel-item.active{display:block}.active.carousel-item-end,.carousel-item-next:not(.carousel-item-start){transform:translateX(100%)}.active.carousel-item-start,.carousel-item-prev:not(.carousel-item-end){transform:translateX(-100%)}.carousel-fade .carousel-item{opacity:0;transform:none;transition-property:opacity}.carousel-fade .carousel-item-next.carousel-item-start,.carousel-fade .carousel-item-prev.carousel-item-end,.carousel-fade .carousel-item.active{opacity:1;z-index:1}.carousel-fade .active.carousel-item-end,.carousel-fade .active.carousel-item-start{opacity:0;transition:opacity 0s .6s;z-index:0}@media (prefers-reduced-motion:reduce){.carousel-fade .active.carousel-item-end,.carousel-fade .active.carousel-item-start{transition:none}}.carousel-control-next,.carousel-control-prev{align-items:center;background:none;border:0;bottom:0;color:#fff;display:flex;justify-content:center;opacity:.5;padding:0;position:absolute;text-align:center;top:0;transition:opacity .15s ease;width:15%;z-index:1}@media (prefers-reduced-motion:reduce){.carousel-control-next,.carousel-control-prev{transition:none}}.carousel-control-next:focus,.carousel-control-next:hover,.carousel-control-prev:focus,.carousel-control-prev:hover{color:#fff;opacity:.9;outline:0;text-decoration:none}.carousel-control-prev{left:0}.carousel-control-next{right:0}.carousel-control-next-icon,.carousel-control-prev-icon{background-position:50%;background-repeat:no-repeat;background-size:100% 100%;display:inline-block;height:2rem;width:2rem}.carousel-control-prev-icon{background-image:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='%23fff'%3E%3Cpath d='M11.354 1.646a.5.5 0 0 1 0 .708L5.707 8l5.647 5.646a.5.5 0 0 1-.708.708l-6-6a.5.5 0 0 1 0-.708l6-6a.5.5 0 0 1 .708 0z'/%3E%3C/svg%3E")}.carousel-control-next-icon{background-image:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='%23fff'%3E%3Cpath d='M4.646 1.646a.5.5 0 0 1 .708 0l6 6a.5.5 0 0 1 0 .708l-6 6a.5.5 0 0 1-.708-.708L10.293 8 4.646 2.354a.5.5 0 0 1 0-.708z'/%3E%3C/svg%3E")}.carousel-indicators{bottom:0;display:flex;justify-content:center;left:0;margin-bottom:1rem;margin-left:15%;margin-right:15%;padding:0;position:absolute;right:0;z-index:2}.carousel-indicators [data-bs-target]{background-clip:padding-box;background-color:#fff;border:0;border-bottom:10px solid transparent;border-top:10px solid transparent;box-sizing:content-box;cursor:pointer;flex:0 1 auto;height:3px;margin-left:3px;margin-right:3px;opacity:.5;padding:0;text-indent:-999px;transition:opacity .6s ease;width:30px}@media (prefers-reduced-motion:reduce){.carousel-indicators [data-bs-target]{transition:none}}.carousel-indicators .active{opacity:1}.carousel-caption{bottom:1.25rem;color:#fff;left:15%;padding-bottom:1.25rem;padding-top:1.25rem;position:absolute;right:15%;text-align:center}.carousel-dark .carousel-control-next-icon,.carousel-dark .carousel-control-prev-icon{filter:invert(1) grayscale(100)}.carousel-dark .carousel-indicators [data-bs-target]{background-color:#000}.carousel-dark .carousel-caption{color:#000}[data-bs-theme=dark] .carousel .carousel-control-next-icon,[data-bs-theme=dark] .carousel .carousel-control-prev-icon,[data-bs-theme=dark].carousel .carousel-control-next-icon,[data-bs-theme=dark].carousel .carousel-control-prev-icon{filter:invert(1) grayscale(100)}[data-bs-theme=dark] .carousel .carousel-indicators [data-bs-target],[data-bs-theme=dark].carousel .carousel-indicators [data-bs-target]{background-color:#000}[data-bs-theme=dark] .carousel .carousel-caption,[data-bs-theme=dark].carousel .carousel-caption{color:#000}.spinner-border,.spinner-grow{animation:var(--bs-spinner-animation-speed) linear infinite var(--bs-spinner-animation-name);border-radius:50%;display:inline-block;height:var(--bs-spinner-height);vertical-align:var(--bs-spinner-vertical-align);width:var(--bs-spinner-width)}@keyframes spinner-border{to{transform:rotate(1turn)}}.spinner-border{--bs-spinner-width:2rem;--bs-spinner-height:2rem;--bs-spinner-vertical-align:-0.125em;--bs-spinner-border-width:0.25em;--bs-spinner-animation-speed:0.75s;--bs-spinner-animation-name:spinner-border;border-right-color:currentcolor;border:var(--bs-spinner-border-width) solid;border-right:var(--bs-spinner-border-width) solid transparent}.spinner-border-sm{--bs-spinner-width:1rem;--bs-spinner-height:1rem;--bs-spinner-border-width:0.2em}@keyframes spinner-grow{0%{transform:scale(0)}50%{opacity:1;transform:none}}.spinner-grow{--bs-spinner-width:2rem;--bs-spinner-height:2rem;--bs-spinner-vertical-align:-0.125em;--bs-spinner-animation-speed:0.75s;--bs-spinner-animation-name:spinner-grow;background-color:currentcolor;opacity:0}.spinner-grow-sm{--bs-spinner-width:1rem;--bs-spinner-height:1rem}@media (prefers-reduced-motion:reduce){.spinner-border,.spinner-grow{--bs-spinner-animation-speed:1.5s}}.offcanvas,.offcanvas-lg,.offcanvas-md,.offcanvas-sm,.offcanvas-xl{--bs-offcanvas-zindex:1045;--bs-offcanvas-width:400px;--bs-offcanvas-height:30vh;--bs-offcanvas-padding-x:1rem;--bs-offcanvas-padding-y:1rem;--bs-offcanvas-color:var(--bs-body-color);--bs-offcanvas-bg:var(--bs-body-bg);--bs-offcanvas-border-width:var(--bs-border-width);--bs-offcanvas-border-color:var(--bs-border-color-translucent);--bs-offcanvas-box-shadow:var(--bs-box-shadow-sm);--bs-offcanvas-transition:transform 0.3s ease-in-out;--bs-offcanvas-title-line-height:1.5}@media (max-width:539.98px){.offcanvas-sm{background-clip:padding-box;background-color:var(--bs-offcanvas-bg);bottom:0;color:var(--bs-offcanvas-color);display:flex;flex-direction:column;max-width:100%;outline:0;position:fixed;transition:var(--bs-offcanvas-transition);visibility:hidden;z-index:var(--bs-offcanvas-zindex)}}@media (max-width:539.98px) and (prefers-reduced-motion:reduce){.offcanvas-sm{transition:none}}@media (max-width:539.98px){.offcanvas-sm.offcanvas-start{border-right:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);left:0;top:0;transform:translateX(-100%);width:var(--bs-offcanvas-width)}.offcanvas-sm.offcanvas-end{border-left:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);right:0;top:0;transform:translateX(100%);width:var(--bs-offcanvas-width)}.offcanvas-sm.offcanvas-top{border-bottom:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);top:0;transform:translateY(-100%)}.offcanvas-sm.offcanvas-bottom,.offcanvas-sm.offcanvas-top{height:var(--bs-offcanvas-height);left:0;max-height:100%;right:0}.offcanvas-sm.offcanvas-bottom{border-top:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);transform:translateY(100%)}.offcanvas-sm.show:not(.hiding),.offcanvas-sm.showing{transform:none}.offcanvas-sm.hiding,.offcanvas-sm.show,.offcanvas-sm.showing{visibility:visible}}@media (min-width:540px){.offcanvas-sm{--bs-offcanvas-height:auto;--bs-offcanvas-border-width:0;background-color:transparent!important}.offcanvas-sm .offcanvas-header{display:none}.offcanvas-sm .offcanvas-body{background-color:transparent!important;display:flex;flex-grow:0;overflow-y:visible;padding:0}}@media (max-width:719.98px){.offcanvas-md{background-clip:padding-box;background-color:var(--bs-offcanvas-bg);bottom:0;color:var(--bs-offcanvas-color);display:flex;flex-direction:column;max-width:100%;outline:0;position:fixed;transition:var(--bs-offcanvas-transition);visibility:hidden;z-index:var(--bs-offcanvas-zindex)}}@media (max-width:719.98px) and (prefers-reduced-motion:reduce){.offcanvas-md{transition:none}}@media (max-width:719.98px){.offcanvas-md.offcanvas-start{border-right:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);left:0;top:0;transform:translateX(-100%);width:var(--bs-offcanvas-width)}.offcanvas-md.offcanvas-end{border-left:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);right:0;top:0;transform:translateX(100%);width:var(--bs-offcanvas-width)}.offcanvas-md.offcanvas-top{border-bottom:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);top:0;transform:translateY(-100%)}.offcanvas-md.offcanvas-bottom,.offcanvas-md.offcanvas-top{height:var(--bs-offcanvas-height);left:0;max-height:100%;right:0}.offcanvas-md.offcanvas-bottom{border-top:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);transform:translateY(100%)}.offcanvas-md.show:not(.hiding),.offcanvas-md.showing{transform:none}.offcanvas-md.hiding,.offcanvas-md.show,.offcanvas-md.showing{visibility:visible}}@media (min-width:720px){.offcanvas-md{--bs-offcanvas-height:auto;--bs-offcanvas-border-width:0;background-color:transparent!important}.offcanvas-md .offcanvas-header{display:none}.offcanvas-md .offcanvas-body{background-color:transparent!important;display:flex;flex-grow:0;overflow-y:visible;padding:0}}@media (max-width:959.98px){.offcanvas-lg{background-clip:padding-box;background-color:var(--bs-offcanvas-bg);bottom:0;color:var(--bs-offcanvas-color);display:flex;flex-direction:column;max-width:100%;outline:0;position:fixed;transition:var(--bs-offcanvas-transition);visibility:hidden;z-index:var(--bs-offcanvas-zindex)}}@media (max-width:959.98px) and (prefers-reduced-motion:reduce){.offcanvas-lg{transition:none}}@media (max-width:959.98px){.offcanvas-lg.offcanvas-start{border-right:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);left:0;top:0;transform:translateX(-100%);width:var(--bs-offcanvas-width)}.offcanvas-lg.offcanvas-end{border-left:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);right:0;top:0;transform:translateX(100%);width:var(--bs-offcanvas-width)}.offcanvas-lg.offcanvas-top{border-bottom:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);top:0;transform:translateY(-100%)}.offcanvas-lg.offcanvas-bottom,.offcanvas-lg.offcanvas-top{height:var(--bs-offcanvas-height);left:0;max-height:100%;right:0}.offcanvas-lg.offcanvas-bottom{border-top:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);transform:translateY(100%)}.offcanvas-lg.show:not(.hiding),.offcanvas-lg.showing{transform:none}.offcanvas-lg.hiding,.offcanvas-lg.show,.offcanvas-lg.showing{visibility:visible}}@media (min-width:960px){.offcanvas-lg{--bs-offcanvas-height:auto;--bs-offcanvas-border-width:0;background-color:transparent!important}.offcanvas-lg .offcanvas-header{display:none}.offcanvas-lg .offcanvas-body{background-color:transparent!important;display:flex;flex-grow:0;overflow-y:visible;padding:0}}@media (max-width:1199.98px){.offcanvas-xl{background-clip:padding-box;background-color:var(--bs-offcanvas-bg);bottom:0;color:var(--bs-offcanvas-color);display:flex;flex-direction:column;max-width:100%;outline:0;position:fixed;transition:var(--bs-offcanvas-transition);visibility:hidden;z-index:var(--bs-offcanvas-zindex)}}@media (max-width:1199.98px) and (prefers-reduced-motion:reduce){.offcanvas-xl{transition:none}}@media (max-width:1199.98px){.offcanvas-xl.offcanvas-start{border-right:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);left:0;top:0;transform:translateX(-100%);width:var(--bs-offcanvas-width)}.offcanvas-xl.offcanvas-end{border-left:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);right:0;top:0;transform:translateX(100%);width:var(--bs-offcanvas-width)}.offcanvas-xl.offcanvas-top{border-bottom:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);top:0;transform:translateY(-100%)}.offcanvas-xl.offcanvas-bottom,.offcanvas-xl.offcanvas-top{height:var(--bs-offcanvas-height);left:0;max-height:100%;right:0}.offcanvas-xl.offcanvas-bottom{border-top:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);transform:translateY(100%)}.offcanvas-xl.show:not(.hiding),.offcanvas-xl.showing{transform:none}.offcanvas-xl.hiding,.offcanvas-xl.show,.offcanvas-xl.showing{visibility:visible}}@media (min-width:1200px){.offcanvas-xl{--bs-offcanvas-height:auto;--bs-offcanvas-border-width:0;background-color:transparent!important}.offcanvas-xl .offcanvas-header{display:none}.offcanvas-xl .offcanvas-body{background-color:transparent!important;display:flex;flex-grow:0;overflow-y:visible;padding:0}}.offcanvas{background-clip:padding-box;background-color:var(--bs-offcanvas-bg);bottom:0;color:var(--bs-offcanvas-color);display:flex;flex-direction:column;max-width:100%;outline:0;position:fixed;transition:var(--bs-offcanvas-transition);visibility:hidden;z-index:var(--bs-offcanvas-zindex)}@media (prefers-reduced-motion:reduce){.offcanvas{transition:none}}.offcanvas.offcanvas-start{border-right:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);left:0;top:0;transform:translateX(-100%);width:var(--bs-offcanvas-width)}.offcanvas.offcanvas-end{border-left:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);right:0;top:0;transform:translateX(100%);width:var(--bs-offcanvas-width)}.offcanvas.offcanvas-top{border-bottom:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);top:0;transform:translateY(-100%)}.offcanvas.offcanvas-bottom,.offcanvas.offcanvas-top{height:var(--bs-offcanvas-height);left:0;max-height:100%;right:0}.offcanvas.offcanvas-bottom{border-top:var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);transform:translateY(100%)}.offcanvas.show:not(.hiding),.offcanvas.showing{transform:none}.offcanvas.hiding,.offcanvas.show,.offcanvas.showing{visibility:visible}.offcanvas-backdrop{background-color:#000;height:100vh;left:0;position:fixed;top:0;width:100vw;z-index:1040}.offcanvas-backdrop.fade{opacity:0}.offcanvas-backdrop.show{opacity:.5}.offcanvas-header{align-items:center;display:flex;padding:var(--bs-offcanvas-padding-y) var(--bs-offcanvas-padding-x)}.offcanvas-header .btn-close{margin:calc(var(--bs-offcanvas-padding-y)*-.5) calc(var(--bs-offcanvas-padding-x)*-.5) calc(var(--bs-offcanvas-padding-y)*-.5) auto;padding:calc(var(--bs-offcanvas-padding-y)*.5) calc(var(--bs-offcanvas-padding-x)*.5)}.offcanvas-title{line-height:var(--bs-offcanvas-title-line-height);margin-bottom:0}.offcanvas-body{flex-grow:1;overflow-y:auto;padding:var(--bs-offcanvas-padding-y) var(--bs-offcanvas-padding-x)}.placeholder{background-color:currentcolor;cursor:wait;display:inline-block;min-height:1em;opacity:.5;vertical-align:middle}.placeholder.btn:before{content:"";display:inline-block}.placeholder-xs{min-height:.6em}.placeholder-sm{min-height:.8em}.placeholder-lg{min-height:1.2em}.placeholder-glow .placeholder{animation:placeholder-glow 2s ease-in-out infinite}@keyframes placeholder-glow{50%{opacity:.2}}.placeholder-wave{animation:placeholder-wave 2s linear infinite;mask-image:linear-gradient(130deg,#000 55%,rgba(0,0,0,.8) 75%,#000 95%);mask-size:200% 100%}@keyframes placeholder-wave{to{mask-position:-200% 0}}.clearfix:after{clear:both;content:"";display:block}.text-bg-primary{background-color:RGBA(var(--bs-primary-rgb),var(--bs-bg-opacity,1))!important;color:#fff!important}.text-bg-secondary{background-color:RGBA(var(--bs-secondary-rgb),var(--bs-bg-opacity,1))!important;color:#fff!important}.text-bg-success{background-color:RGBA(var(--bs-success-rgb),var(--bs-bg-opacity,1))!important;color:#fff!important}.text-bg-info{background-color:RGBA(var(--bs-info-rgb),var(--bs-bg-opacity,1))!important;color:#000!important}.text-bg-warning{background-color:RGBA(var(--bs-warning-rgb),var(--bs-bg-opacity,1))!important;color:#000!important}.text-bg-danger{background-color:RGBA(var(--bs-danger-rgb),var(--bs-bg-opacity,1))!important;color:#fff!important}.text-bg-light{background-color:RGBA(var(--bs-light-rgb),var(--bs-bg-opacity,1))!important;color:#000!important}.text-bg-dark{background-color:RGBA(var(--bs-dark-rgb),var(--bs-bg-opacity,1))!important;color:#fff!important}.link-primary{color:RGBA(var(--bs-primary-rgb),var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(var(--bs-primary-rgb),var(--bs-link-underline-opacity,1))!important}.link-primary:focus,.link-primary:hover{color:RGBA(10,88,202,var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(10,88,202,var(--bs-link-underline-opacity,1))!important}.link-secondary{color:RGBA(var(--bs-secondary-rgb),var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(var(--bs-secondary-rgb),var(--bs-link-underline-opacity,1))!important}.link-secondary:focus,.link-secondary:hover{color:RGBA(86,94,100,var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(86,94,100,var(--bs-link-underline-opacity,1))!important}.link-success{color:RGBA(var(--bs-success-rgb),var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(var(--bs-success-rgb),var(--bs-link-underline-opacity,1))!important}.link-success:focus,.link-success:hover{color:RGBA(20,108,67,var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(20,108,67,var(--bs-link-underline-opacity,1))!important}.link-info{color:RGBA(var(--bs-info-rgb),var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(var(--bs-info-rgb),var(--bs-link-underline-opacity,1))!important}.link-info:focus,.link-info:hover{color:RGBA(61,213,243,var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(61,213,243,var(--bs-link-underline-opacity,1))!important}.link-warning{color:RGBA(var(--bs-warning-rgb),var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(var(--bs-warning-rgb),var(--bs-link-underline-opacity,1))!important}.link-warning:focus,.link-warning:hover{color:RGBA(255,205,57,var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(255,205,57,var(--bs-link-underline-opacity,1))!important}.link-danger{color:RGBA(var(--bs-danger-rgb),var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(var(--bs-danger-rgb),var(--bs-link-underline-opacity,1))!important}.link-danger:focus,.link-danger:hover{color:RGBA(176,42,55,var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(176,42,55,var(--bs-link-underline-opacity,1))!important}.link-light{color:RGBA(var(--bs-light-rgb),var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(var(--bs-light-rgb),var(--bs-link-underline-opacity,1))!important}.link-light:focus,.link-light:hover{color:RGBA(249,250,251,var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(249,250,251,var(--bs-link-underline-opacity,1))!important}.link-dark{color:RGBA(var(--bs-dark-rgb),var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(var(--bs-dark-rgb),var(--bs-link-underline-opacity,1))!important}.link-dark:focus,.link-dark:hover{color:RGBA(26,30,33,var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(26,30,33,var(--bs-link-underline-opacity,1))!important}.link-body-emphasis{color:RGBA(var(--bs-emphasis-color-rgb),var(--bs-link-opacity,1))!important;text-decoration-color:RGBA(var(--bs-emphasis-color-rgb),var(--bs-link-underline-opacity,1))!important}.link-body-emphasis:focus,.link-body-emphasis:hover{color:RGBA(var(--bs-emphasis-color-rgb),var(--bs-link-opacity,.75))!important;text-decoration-color:RGBA(var(--bs-emphasis-color-rgb),var(--bs-link-underline-opacity,.75))!important}.focus-ring:focus{box-shadow:var(--bs-focus-ring-x,0) var(--bs-focus-ring-y,0) var(--bs-focus-ring-blur,0) var(--bs-focus-ring-width) var(--bs-focus-ring-color);outline:0}.icon-link{align-items:center;backface-visibility:hidden;display:inline-flex;gap:.375rem;text-decoration-color:rgba(var(--bs-link-color-rgb),var(--bs-link-opacity,.5));text-underline-offset:.25em}.icon-link>.bi{fill:currentcolor;flex-shrink:0;height:1em;transition:transform .2s ease-in-out;width:1em}@media (prefers-reduced-motion:reduce){.icon-link>.bi{transition:none}}.icon-link-hover:focus-visible>.bi,.icon-link-hover:hover>.bi{transform:var(--bs-icon-link-transform,translate3d(.25em,0,0))}.ratio{position:relative;width:100%}.ratio:before{content:"";display:block;padding-top:var(--bs-aspect-ratio)}.ratio>*{height:100%;left:0;position:absolute;top:0;width:100%}.ratio-1x1{--bs-aspect-ratio:100%}.ratio-4x3{--bs-aspect-ratio:75%}.ratio-16x9{--bs-aspect-ratio:56.25%}.ratio-21x9{--bs-aspect-ratio:42.8571428571%}.fixed-top{top:0}.fixed-bottom,.fixed-top{left:0;position:fixed;right:0;z-index:1030}.fixed-bottom{bottom:0}.sticky-top{top:0}.sticky-bottom,.sticky-top{position:sticky;z-index:1020}.sticky-bottom{bottom:0}@media (min-width:540px){.sticky-sm-top{position:sticky;top:0;z-index:1020}.sticky-sm-bottom{bottom:0;position:sticky;z-index:1020}}@media (min-width:720px){.sticky-md-top{position:sticky;top:0;z-index:1020}.sticky-md-bottom{bottom:0;position:sticky;z-index:1020}}@media (min-width:960px){.sticky-lg-top{position:sticky;top:0;z-index:1020}.sticky-lg-bottom{bottom:0;position:sticky;z-index:1020}}@media (min-width:1200px){.sticky-xl-top{position:sticky;top:0;z-index:1020}.sticky-xl-bottom{bottom:0;position:sticky;z-index:1020}}.hstack{align-items:center;flex-direction:row}.hstack,.vstack{align-self:stretch;display:flex}.vstack{flex:1 1 auto;flex-direction:column}.visually-hidden,.visually-hidden-focusable:not(:focus):not(:focus-within){clip:rect(0,0,0,0)!important;border:0!important;height:1px!important;margin:-1px!important;overflow:hidden!important;padding:0!important;white-space:nowrap!important;width:1px!important}.visually-hidden-focusable:not(:focus):not(:focus-within):not(caption),.visually-hidden:not(caption){position:absolute!important}.stretched-link:after{bottom:0;content:"";left:0;position:absolute;right:0;top:0;z-index:1}.text-truncate{overflow:hidden;text-overflow:ellipsis;white-space:nowrap}.vr{align-self:stretch;background-color:currentcolor;display:inline-block;min-height:1em;opacity:.25;width:var(--bs-border-width)}.align-baseline{vertical-align:baseline!important}.align-top{vertical-align:top!important}.align-middle{vertical-align:middle!important}.align-bottom{vertical-align:bottom!important}.align-text-bottom{vertical-align:text-bottom!important}.align-text-top{vertical-align:text-top!important}.float-start{float:left!important}.float-end{float:right!important}.float-none{float:none!important}.object-fit-contain{object-fit:contain!important}.object-fit-cover{object-fit:cover!important}.object-fit-fill{object-fit:fill!important}.object-fit-scale{object-fit:scale-down!important}.object-fit-none{object-fit:none!important}.opacity-0{opacity:0!important}.opacity-25{opacity:.25!important}.opacity-50{opacity:.5!important}.opacity-75{opacity:.75!important}.opacity-100{opacity:1!important}.overflow-auto{overflow:auto!important}.overflow-hidden{overflow:hidden!important}.overflow-visible{overflow:visible!important}.overflow-scroll{overflow:scroll!important}.overflow-x-auto{overflow-x:auto!important}.overflow-x-hidden{overflow-x:hidden!important}.overflow-x-visible{overflow-x:visible!important}.overflow-x-scroll{overflow-x:scroll!important}.overflow-y-auto{overflow-y:auto!important}.overflow-y-hidden{overflow-y:hidden!important}.overflow-y-visible{overflow-y:visible!important}.overflow-y-scroll{overflow-y:scroll!important}.d-inline{display:inline!important}.d-inline-block{display:inline-block!important}.d-block{display:block!important}.d-grid{display:grid!important}.d-inline-grid{display:inline-grid!important}.d-table{display:table!important}.d-table-row{display:table-row!important}.d-table-cell{display:table-cell!important}.d-flex{display:flex!important}.d-inline-flex{display:inline-flex!important}.d-none{display:none!important}.shadow{box-shadow:var(--bs-box-shadow)!important}.shadow-sm{box-shadow:var(--bs-box-shadow-sm)!important}.shadow-lg{box-shadow:var(--bs-box-shadow-lg)!important}.shadow-none{box-shadow:none!important}.focus-ring-primary{--bs-focus-ring-color:rgba(var(--bs-primary-rgb),var(--bs-focus-ring-opacity))}.focus-ring-secondary{--bs-focus-ring-color:rgba(var(--bs-secondary-rgb),var(--bs-focus-ring-opacity))}.focus-ring-success{--bs-focus-ring-color:rgba(var(--bs-success-rgb),var(--bs-focus-ring-opacity))}.focus-ring-info{--bs-focus-ring-color:rgba(var(--bs-info-rgb),var(--bs-focus-ring-opacity))}.focus-ring-warning{--bs-focus-ring-color:rgba(var(--bs-warning-rgb),var(--bs-focus-ring-opacity))}.focus-ring-danger{--bs-focus-ring-color:rgba(var(--bs-danger-rgb),var(--bs-focus-ring-opacity))}.focus-ring-light{--bs-focus-ring-color:rgba(var(--bs-light-rgb),var(--bs-focus-ring-opacity))}.focus-ring-dark{--bs-focus-ring-color:rgba(var(--bs-dark-rgb),var(--bs-focus-ring-opacity))}.position-static{position:static!important}.position-relative{position:relative!important}.position-absolute{position:absolute!important}.position-fixed{position:fixed!important}.position-sticky{position:sticky!important}.top-0{top:0!important}.top-50{top:50%!important}.top-100{top:100%!important}.bottom-0{bottom:0!important}.bottom-50{bottom:50%!important}.bottom-100{bottom:100%!important}.start-0{left:0!important}.start-50{left:50%!important}.start-100{left:100%!important}.end-0{right:0!important}.end-50{right:50%!important}.end-100{right:100%!important}.translate-middle{transform:translate(-50%,-50%)!important}.translate-middle-x{transform:translateX(-50%)!important}.translate-middle-y{transform:translateY(-50%)!important}.border{border:var(--bs-border-width) var(--bs-border-style) var(--bs-border-color)!important}.border-0{border:0!important}.border-top{border-top:var(--bs-border-width) var(--bs-border-style) var(--bs-border-color)!important}.border-top-0{border-top:0!important}.border-end{border-right:var(--bs-border-width) var(--bs-border-style) var(--bs-border-color)!important}.border-end-0{border-right:0!important}.border-bottom{border-bottom:var(--bs-border-width) var(--bs-border-style) var(--bs-border-color)!important}.border-bottom-0{border-bottom:0!important}.border-start{border-left:var(--bs-border-width) var(--bs-border-style) var(--bs-border-color)!important}.border-start-0{border-left:0!important}.border-primary{--bs-border-opacity:1;border-color:rgba(var(--bs-primary-rgb),var(--bs-border-opacity))!important}.border-secondary{--bs-border-opacity:1;border-color:rgba(var(--bs-secondary-rgb),var(--bs-border-opacity))!important}.border-success{--bs-border-opacity:1;border-color:rgba(var(--bs-success-rgb),var(--bs-border-opacity))!important}.border-info{--bs-border-opacity:1;border-color:rgba(var(--bs-info-rgb),var(--bs-border-opacity))!important}.border-warning{--bs-border-opacity:1;border-color:rgba(var(--bs-warning-rgb),var(--bs-border-opacity))!important}.border-danger{--bs-border-opacity:1;border-color:rgba(var(--bs-danger-rgb),var(--bs-border-opacity))!important}.border-light{--bs-border-opacity:1;border-color:rgba(var(--bs-light-rgb),var(--bs-border-opacity))!important}.border-dark{--bs-border-opacity:1;border-color:rgba(var(--bs-dark-rgb),var(--bs-border-opacity))!important}.border-black{--bs-border-opacity:1;border-color:rgba(var(--bs-black-rgb),var(--bs-border-opacity))!important}.border-white{--bs-border-opacity:1;border-color:rgba(var(--bs-white-rgb),var(--bs-border-opacity))!important}.border-primary-subtle{border-color:var(--bs-primary-border-subtle)!important}.border-secondary-subtle{border-color:var(--bs-secondary-border-subtle)!important}.border-success-subtle{border-color:var(--bs-success-border-subtle)!important}.border-info-subtle{border-color:var(--bs-info-border-subtle)!important}.border-warning-subtle{border-color:var(--bs-warning-border-subtle)!important}.border-danger-subtle{border-color:var(--bs-danger-border-subtle)!important}.border-light-subtle{border-color:var(--bs-light-border-subtle)!important}.border-dark-subtle{border-color:var(--bs-dark-border-subtle)!important}.border-1{border-width:1px!important}.border-2{border-width:2px!important}.border-3{border-width:3px!important}.border-4{border-width:4px!important}.border-5{border-width:5px!important}.border-opacity-10{--bs-border-opacity:0.1}.border-opacity-25{--bs-border-opacity:0.25}.border-opacity-50{--bs-border-opacity:0.5}.border-opacity-75{--bs-border-opacity:0.75}.border-opacity-100{--bs-border-opacity:1}.w-25{width:25%!important}.w-50{width:50%!important}.w-75{width:75%!important}.w-100{width:100%!important}.w-auto{width:auto!important}.mw-100{max-width:100%!important}.vw-100{width:100vw!important}.min-vw-100{min-width:100vw!important}.h-25{height:25%!important}.h-50{height:50%!important}.h-75{height:75%!important}.h-100{height:100%!important}.h-auto{height:auto!important}.mh-100{max-height:100%!important}.vh-100{height:100vh!important}.min-vh-100{min-height:100vh!important}.flex-fill{flex:1 1 auto!important}.flex-row{flex-direction:row!important}.flex-column{flex-direction:column!important}.flex-row-reverse{flex-direction:row-reverse!important}.flex-column-reverse{flex-direction:column-reverse!important}.flex-grow-0{flex-grow:0!important}.flex-grow-1{flex-grow:1!important}.flex-shrink-0{flex-shrink:0!important}.flex-shrink-1{flex-shrink:1!important}.flex-wrap{flex-wrap:wrap!important}.flex-nowrap{flex-wrap:nowrap!important}.flex-wrap-reverse{flex-wrap:wrap-reverse!important}.justify-content-start{justify-content:flex-start!important}.justify-content-end{justify-content:flex-end!important}.justify-content-center{justify-content:center!important}.justify-content-between{justify-content:space-between!important}.justify-content-around{justify-content:space-around!important}.justify-content-evenly{justify-content:space-evenly!important}.align-items-start{align-items:flex-start!important}.align-items-end{align-items:flex-end!important}.align-items-center{align-items:center!important}.align-items-baseline{align-items:baseline!important}.align-items-stretch{align-items:stretch!important}.align-content-start{align-content:flex-start!important}.align-content-end{align-content:flex-end!important}.align-content-center{align-content:center!important}.align-content-between{align-content:space-between!important}.align-content-around{align-content:space-around!important}.align-content-stretch{align-content:stretch!important}.align-self-auto{align-self:auto!important}.align-self-start{align-self:flex-start!important}.align-self-end{align-self:flex-end!important}.align-self-center{align-self:center!important}.align-self-baseline{align-self:baseline!important}.align-self-stretch{align-self:stretch!important}.order-first{order:-1!important}.order-0{order:0!important}.order-1{order:1!important}.order-2{order:2!important}.order-3{order:3!important}.order-4{order:4!important}.order-5{order:5!important}.order-last{order:6!important}.m-0{margin:0!important}.m-1{margin:.25rem!important}.m-2{margin:.5rem!important}.m-3{margin:1rem!important}.m-4{margin:1.5rem!important}.m-5{margin:3rem!important}.m-auto{margin:auto!important}.mx-0{margin-left:0!important;margin-right:0!important}.mx-1{margin-left:.25rem!important;margin-right:.25rem!important}.mx-2{margin-left:.5rem!important;margin-right:.5rem!important}.mx-3{margin-left:1rem!important;margin-right:1rem!important}.mx-4{margin-left:1.5rem!important;margin-right:1.5rem!important}.mx-5{margin-left:3rem!important;margin-right:3rem!important}.mx-auto{margin-left:auto!important;margin-right:auto!important}.my-0{margin-bottom:0!important;margin-top:0!important}.my-1{margin-bottom:.25rem!important;margin-top:.25rem!important}.my-2{margin-bottom:.5rem!important;margin-top:.5rem!important}.my-3{margin-bottom:1rem!important;margin-top:1rem!important}.my-4{margin-bottom:1.5rem!important;margin-top:1.5rem!important}.my-5{margin-bottom:3rem!important;margin-top:3rem!important}.my-auto{margin-bottom:auto!important;margin-top:auto!important}.mt-0{margin-top:0!important}.mt-1{margin-top:.25rem!important}.mt-2{margin-top:.5rem!important}.mt-3{margin-top:1rem!important}.mt-4{margin-top:1.5rem!important}.mt-5{margin-top:3rem!important}.mt-auto{margin-top:auto!important}.me-0{margin-right:0!important}.me-1{margin-right:.25rem!important}.me-2{margin-right:.5rem!important}.me-3{margin-right:1rem!important}.me-4{margin-right:1.5rem!important}.me-5{margin-right:3rem!important}.me-auto{margin-right:auto!important}.mb-0{margin-bottom:0!important}.mb-1{margin-bottom:.25rem!important}.mb-2{margin-bottom:.5rem!important}.mb-3{margin-bottom:1rem!important}.mb-4{margin-bottom:1.5rem!important}.mb-5{margin-bottom:3rem!important}.mb-auto{margin-bottom:auto!important}.ms-0{margin-left:0!important}.ms-1{margin-left:.25rem!important}.ms-2{margin-left:.5rem!important}.ms-3{margin-left:1rem!important}.ms-4{margin-left:1.5rem!important}.ms-5{margin-left:3rem!important}.ms-auto{margin-left:auto!important}.p-0{padding:0!important}.p-1{padding:.25rem!important}.p-2{padding:.5rem!important}.p-3{padding:1rem!important}.p-4{padding:1.5rem!important}.p-5{padding:3rem!important}.px-0{padding-left:0!important;padding-right:0!important}.px-1{padding-left:.25rem!important;padding-right:.25rem!important}.px-2{padding-left:.5rem!important;padding-right:.5rem!important}.px-3{padding-left:1rem!important;padding-right:1rem!important}.px-4{padding-left:1.5rem!important;padding-right:1.5rem!important}.px-5{padding-left:3rem!important;padding-right:3rem!important}.py-0{padding-bottom:0!important;padding-top:0!important}.py-1{padding-bottom:.25rem!important;padding-top:.25rem!important}.py-2{padding-bottom:.5rem!important;padding-top:.5rem!important}.py-3{padding-bottom:1rem!important;padding-top:1rem!important}.py-4{padding-bottom:1.5rem!important;padding-top:1.5rem!important}.py-5{padding-bottom:3rem!important;padding-top:3rem!important}.pt-0{padding-top:0!important}.pt-1{padding-top:.25rem!important}.pt-2{padding-top:.5rem!important}.pt-3{padding-top:1rem!important}.pt-4{padding-top:1.5rem!important}.pt-5{padding-top:3rem!important}.pe-0{padding-right:0!important}.pe-1{padding-right:.25rem!important}.pe-2{padding-right:.5rem!important}.pe-3{padding-right:1rem!important}.pe-4{padding-right:1.5rem!important}.pe-5{padding-right:3rem!important}.pb-0{padding-bottom:0!important}.pb-1{padding-bottom:.25rem!important}.pb-2{padding-bottom:.5rem!important}.pb-3{padding-bottom:1rem!important}.pb-4{padding-bottom:1.5rem!important}.pb-5{padding-bottom:3rem!important}.ps-0{padding-left:0!important}.ps-1{padding-left:.25rem!important}.ps-2{padding-left:.5rem!important}.ps-3{padding-left:1rem!important}.ps-4{padding-left:1.5rem!important}.ps-5{padding-left:3rem!important}.gap-0{gap:0!important}.gap-1{gap:.25rem!important}.gap-2{gap:.5rem!important}.gap-3{gap:1rem!important}.gap-4{gap:1.5rem!important}.gap-5{gap:3rem!important}.row-gap-0{row-gap:0!important}.row-gap-1{row-gap:.25rem!important}.row-gap-2{row-gap:.5rem!important}.row-gap-3{row-gap:1rem!important}.row-gap-4{row-gap:1.5rem!important}.row-gap-5{row-gap:3rem!important}.column-gap-0{column-gap:0!important}.column-gap-1{column-gap:.25rem!important}.column-gap-2{column-gap:.5rem!important}.column-gap-3{column-gap:1rem!important}.column-gap-4{column-gap:1.5rem!important}.column-gap-5{column-gap:3rem!important}.font-monospace{font-family:var(--bs-font-monospace)!important}.fs-1{font-size:calc(1.375rem + 1.5vw)!important}.fs-2{font-size:calc(1.325rem + .9vw)!important}.fs-3{font-size:calc(1.3rem + .6vw)!important}.fs-4{font-size:calc(1.275rem + .3vw)!important}.fs-5{font-size:1.25rem!important}.fs-6{font-size:1rem!important}.fst-italic{font-style:italic!important}.fst-normal{font-style:normal!important}.fw-lighter{font-weight:lighter!important}.fw-light{font-weight:300!important}.fw-normal{font-weight:400!important}.fw-medium{font-weight:500!important}.fw-semibold{font-weight:600!important}.fw-bold{font-weight:700!important}.fw-bolder{font-weight:bolder!important}.lh-1{line-height:1!important}.lh-sm{line-height:1.25!important}.lh-base{line-height:1.5!important}.lh-lg{line-height:2!important}.text-start{text-align:left!important}.text-end{text-align:right!important}.text-center{text-align:center!important}.text-decoration-none{text-decoration:none!important}.text-decoration-underline{text-decoration:underline!important}.text-decoration-line-through{text-decoration:line-through!important}.text-lowercase{text-transform:lowercase!important}.text-uppercase{text-transform:uppercase!important}.text-capitalize{text-transform:capitalize!important}.text-wrap{white-space:normal!important}.text-nowrap{white-space:nowrap!important}.text-break{word-wrap:break-word!important;word-break:break-word!important}.text-primary{--bs-text-opacity:1;color:rgba(var(--bs-primary-rgb),var(--bs-text-opacity))!important}.text-secondary{--bs-text-opacity:1;color:rgba(var(--bs-secondary-rgb),var(--bs-text-opacity))!important}.text-success{--bs-text-opacity:1;color:rgba(var(--bs-success-rgb),var(--bs-text-opacity))!important}.text-info{--bs-text-opacity:1;color:rgba(var(--bs-info-rgb),var(--bs-text-opacity))!important}.text-warning{--bs-text-opacity:1;color:rgba(var(--bs-warning-rgb),var(--bs-text-opacity))!important}.text-danger{--bs-text-opacity:1;color:rgba(var(--bs-danger-rgb),var(--bs-text-opacity))!important}.text-light{--bs-text-opacity:1;color:rgba(var(--bs-light-rgb),var(--bs-text-opacity))!important}.text-dark{--bs-text-opacity:1;color:rgba(var(--bs-dark-rgb),var(--bs-text-opacity))!important}.text-black{--bs-text-opacity:1;color:rgba(var(--bs-black-rgb),var(--bs-text-opacity))!important}.text-white{--bs-text-opacity:1;color:rgba(var(--bs-white-rgb),var(--bs-text-opacity))!important}.text-body{--bs-text-opacity:1;color:rgba(var(--bs-body-color-rgb),var(--bs-text-opacity))!important}.text-muted{--bs-text-opacity:1;color:var(--bs-secondary-color)!important}.text-black-50{--bs-text-opacity:1;color:rgba(0,0,0,.5)!important}.text-white-50{--bs-text-opacity:1;color:hsla(0,0%,100%,.5)!important}.text-body-secondary{--bs-text-opacity:1;color:var(--bs-secondary-color)!important}.text-body-tertiary{--bs-text-opacity:1;color:var(--bs-tertiary-color)!important}.text-body-emphasis{--bs-text-opacity:1;color:var(--bs-emphasis-color)!important}.text-reset{--bs-text-opacity:1;color:inherit!important}.text-opacity-25{--bs-text-opacity:0.25}.text-opacity-50{--bs-text-opacity:0.5}.text-opacity-75{--bs-text-opacity:0.75}.text-opacity-100{--bs-text-opacity:1}.text-primary-emphasis{color:var(--bs-primary-text-emphasis)!important}.text-secondary-emphasis{color:var(--bs-secondary-text-emphasis)!important}.text-success-emphasis{color:var(--bs-success-text-emphasis)!important}.text-info-emphasis{color:var(--bs-info-text-emphasis)!important}.text-warning-emphasis{color:var(--bs-warning-text-emphasis)!important}.text-danger-emphasis{color:var(--bs-danger-text-emphasis)!important}.text-light-emphasis{color:var(--bs-light-text-emphasis)!important}.text-dark-emphasis{color:var(--bs-dark-text-emphasis)!important}.link-opacity-10,.link-opacity-10-hover:hover{--bs-link-opacity:0.1}.link-opacity-25,.link-opacity-25-hover:hover{--bs-link-opacity:0.25}.link-opacity-50,.link-opacity-50-hover:hover{--bs-link-opacity:0.5}.link-opacity-75,.link-opacity-75-hover:hover{--bs-link-opacity:0.75}.link-opacity-100,.link-opacity-100-hover:hover{--bs-link-opacity:1}.link-offset-1,.link-offset-1-hover:hover{text-underline-offset:.125em!important}.link-offset-2,.link-offset-2-hover:hover{text-underline-offset:.25em!important}.link-offset-3,.link-offset-3-hover:hover{text-underline-offset:.375em!important}.link-underline-primary{--bs-link-underline-opacity:1;text-decoration-color:rgba(var(--bs-primary-rgb),var(--bs-link-underline-opacity))!important}.link-underline-secondary{--bs-link-underline-opacity:1;text-decoration-color:rgba(var(--bs-secondary-rgb),var(--bs-link-underline-opacity))!important}.link-underline-success{--bs-link-underline-opacity:1;text-decoration-color:rgba(var(--bs-success-rgb),var(--bs-link-underline-opacity))!important}.link-underline-info{--bs-link-underline-opacity:1;text-decoration-color:rgba(var(--bs-info-rgb),var(--bs-link-underline-opacity))!important}.link-underline-warning{--bs-link-underline-opacity:1;text-decoration-color:rgba(var(--bs-warning-rgb),var(--bs-link-underline-opacity))!important}.link-underline-danger{--bs-link-underline-opacity:1;text-decoration-color:rgba(var(--bs-danger-rgb),var(--bs-link-underline-opacity))!important}.link-underline-light{--bs-link-underline-opacity:1;text-decoration-color:rgba(var(--bs-light-rgb),var(--bs-link-underline-opacity))!important}.link-underline-dark{--bs-link-underline-opacity:1;text-decoration-color:rgba(var(--bs-dark-rgb),var(--bs-link-underline-opacity))!important}.link-underline{--bs-link-underline-opacity:1;text-decoration-color:rgba(var(--bs-link-color-rgb),var(--bs-link-underline-opacity,1))!important}.link-underline-opacity-0,.link-underline-opacity-0-hover:hover{--bs-link-underline-opacity:0}.link-underline-opacity-10,.link-underline-opacity-10-hover:hover{--bs-link-underline-opacity:0.1}.link-underline-opacity-25,.link-underline-opacity-25-hover:hover{--bs-link-underline-opacity:0.25}.link-underline-opacity-50,.link-underline-opacity-50-hover:hover{--bs-link-underline-opacity:0.5}.link-underline-opacity-75,.link-underline-opacity-75-hover:hover{--bs-link-underline-opacity:0.75}.link-underline-opacity-100,.link-underline-opacity-100-hover:hover{--bs-link-underline-opacity:1}.bg-primary{--bs-bg-opacity:1;background-color:rgba(var(--bs-primary-rgb),var(--bs-bg-opacity))!important}.bg-secondary{--bs-bg-opacity:1;background-color:rgba(var(--bs-secondary-rgb),var(--bs-bg-opacity))!important}.bg-success{--bs-bg-opacity:1;background-color:rgba(var(--bs-success-rgb),var(--bs-bg-opacity))!important}.bg-info{--bs-bg-opacity:1;background-color:rgba(var(--bs-info-rgb),var(--bs-bg-opacity))!important}.bg-warning{--bs-bg-opacity:1;background-color:rgba(var(--bs-warning-rgb),var(--bs-bg-opacity))!important}.bg-danger{--bs-bg-opacity:1;background-color:rgba(var(--bs-danger-rgb),var(--bs-bg-opacity))!important}.bg-light{--bs-bg-opacity:1;background-color:rgba(var(--bs-light-rgb),var(--bs-bg-opacity))!important}.bg-dark{--bs-bg-opacity:1;background-color:rgba(var(--bs-dark-rgb),var(--bs-bg-opacity))!important}.bg-black{--bs-bg-opacity:1;background-color:rgba(var(--bs-black-rgb),var(--bs-bg-opacity))!important}.bg-white{--bs-bg-opacity:1;background-color:rgba(var(--bs-white-rgb),var(--bs-bg-opacity))!important}.bg-body{--bs-bg-opacity:1;background-color:rgba(var(--bs-body-bg-rgb),var(--bs-bg-opacity))!important}.bg-transparent{--bs-bg-opacity:1;background-color:transparent!important}.bg-body-secondary{--bs-bg-opacity:1;background-color:rgba(var(--bs-secondary-bg-rgb),var(--bs-bg-opacity))!important}.bg-body-tertiary{--bs-bg-opacity:1;background-color:rgba(var(--bs-tertiary-bg-rgb),var(--bs-bg-opacity))!important}.bg-opacity-10{--bs-bg-opacity:0.1}.bg-opacity-25{--bs-bg-opacity:0.25}.bg-opacity-50{--bs-bg-opacity:0.5}.bg-opacity-75{--bs-bg-opacity:0.75}.bg-opacity-100{--bs-bg-opacity:1}.bg-primary-subtle{background-color:var(--bs-primary-bg-subtle)!important}.bg-secondary-subtle{background-color:var(--bs-secondary-bg-subtle)!important}.bg-success-subtle{background-color:var(--bs-success-bg-subtle)!important}.bg-info-subtle{background-color:var(--bs-info-bg-subtle)!important}.bg-warning-subtle{background-color:var(--bs-warning-bg-subtle)!important}.bg-danger-subtle{background-color:var(--bs-danger-bg-subtle)!important}.bg-light-subtle{background-color:var(--bs-light-bg-subtle)!important}.bg-dark-subtle{background-color:var(--bs-dark-bg-subtle)!important}.bg-gradient{background-image:var(--bs-gradient)!important}.user-select-all{user-select:all!important}.user-select-auto{user-select:auto!important}.user-select-none{user-select:none!important}.pe-none{pointer-events:none!important}.pe-auto{pointer-events:auto!important}.rounded{border-radius:var(--bs-border-radius)!important}.rounded-0{border-radius:0!important}.rounded-1{border-radius:var(--bs-border-radius-sm)!important}.rounded-2{border-radius:var(--bs-border-radius)!important}.rounded-3{border-radius:var(--bs-border-radius-lg)!important}.rounded-4{border-radius:var(--bs-border-radius-xl)!important}.rounded-5{border-radius:var(--bs-border-radius-xxl)!important}.rounded-circle{border-radius:50%!important}.rounded-pill{border-radius:var(--bs-border-radius-pill)!important}.rounded-top{border-top-left-radius:var(--bs-border-radius)!important;border-top-right-radius:var(--bs-border-radius)!important}.rounded-top-0{border-top-left-radius:0!important;border-top-right-radius:0!important}.rounded-top-1{border-top-left-radius:var(--bs-border-radius-sm)!important;border-top-right-radius:var(--bs-border-radius-sm)!important}.rounded-top-2{border-top-left-radius:var(--bs-border-radius)!important;border-top-right-radius:var(--bs-border-radius)!important}.rounded-top-3{border-top-left-radius:var(--bs-border-radius-lg)!important;border-top-right-radius:var(--bs-border-radius-lg)!important}.rounded-top-4{border-top-left-radius:var(--bs-border-radius-xl)!important;border-top-right-radius:var(--bs-border-radius-xl)!important}.rounded-top-5{border-top-left-radius:var(--bs-border-radius-xxl)!important;border-top-right-radius:var(--bs-border-radius-xxl)!important}.rounded-top-circle{border-top-left-radius:50%!important;border-top-right-radius:50%!important}.rounded-top-pill{border-top-left-radius:var(--bs-border-radius-pill)!important;border-top-right-radius:var(--bs-border-radius-pill)!important}.rounded-end{border-bottom-right-radius:var(--bs-border-radius)!important;border-top-right-radius:var(--bs-border-radius)!important}.rounded-end-0{border-bottom-right-radius:0!important;border-top-right-radius:0!important}.rounded-end-1{border-bottom-right-radius:var(--bs-border-radius-sm)!important;border-top-right-radius:var(--bs-border-radius-sm)!important}.rounded-end-2{border-bottom-right-radius:var(--bs-border-radius)!important;border-top-right-radius:var(--bs-border-radius)!important}.rounded-end-3{border-bottom-right-radius:var(--bs-border-radius-lg)!important;border-top-right-radius:var(--bs-border-radius-lg)!important}.rounded-end-4{border-bottom-right-radius:var(--bs-border-radius-xl)!important;border-top-right-radius:var(--bs-border-radius-xl)!important}.rounded-end-5{border-bottom-right-radius:var(--bs-border-radius-xxl)!important;border-top-right-radius:var(--bs-border-radius-xxl)!important}.rounded-end-circle{border-bottom-right-radius:50%!important;border-top-right-radius:50%!important}.rounded-end-pill{border-bottom-right-radius:var(--bs-border-radius-pill)!important;border-top-right-radius:var(--bs-border-radius-pill)!important}.rounded-bottom{border-bottom-left-radius:var(--bs-border-radius)!important;border-bottom-right-radius:var(--bs-border-radius)!important}.rounded-bottom-0{border-bottom-left-radius:0!important;border-bottom-right-radius:0!important}.rounded-bottom-1{border-bottom-left-radius:var(--bs-border-radius-sm)!important;border-bottom-right-radius:var(--bs-border-radius-sm)!important}.rounded-bottom-2{border-bottom-left-radius:var(--bs-border-radius)!important;border-bottom-right-radius:var(--bs-border-radius)!important}.rounded-bottom-3{border-bottom-left-radius:var(--bs-border-radius-lg)!important;border-bottom-right-radius:var(--bs-border-radius-lg)!important}.rounded-bottom-4{border-bottom-left-radius:var(--bs-border-radius-xl)!important;border-bottom-right-radius:var(--bs-border-radius-xl)!important}.rounded-bottom-5{border-bottom-left-radius:var(--bs-border-radius-xxl)!important;border-bottom-right-radius:var(--bs-border-radius-xxl)!important}.rounded-bottom-circle{border-bottom-left-radius:50%!important;border-bottom-right-radius:50%!important}.rounded-bottom-pill{border-bottom-left-radius:var(--bs-border-radius-pill)!important;border-bottom-right-radius:var(--bs-border-radius-pill)!important}.rounded-start{border-bottom-left-radius:var(--bs-border-radius)!important;border-top-left-radius:var(--bs-border-radius)!important}.rounded-start-0{border-bottom-left-radius:0!important;border-top-left-radius:0!important}.rounded-start-1{border-bottom-left-radius:var(--bs-border-radius-sm)!important;border-top-left-radius:var(--bs-border-radius-sm)!important}.rounded-start-2{border-bottom-left-radius:var(--bs-border-radius)!important;border-top-left-radius:var(--bs-border-radius)!important}.rounded-start-3{border-bottom-left-radius:var(--bs-border-radius-lg)!important;border-top-left-radius:var(--bs-border-radius-lg)!important}.rounded-start-4{border-bottom-left-radius:var(--bs-border-radius-xl)!important;border-top-left-radius:var(--bs-border-radius-xl)!important}.rounded-start-5{border-bottom-left-radius:var(--bs-border-radius-xxl)!important;border-top-left-radius:var(--bs-border-radius-xxl)!important}.rounded-start-circle{border-bottom-left-radius:50%!important;border-top-left-radius:50%!important}.rounded-start-pill{border-bottom-left-radius:var(--bs-border-radius-pill)!important;border-top-left-radius:var(--bs-border-radius-pill)!important}.visible{visibility:visible!important}.invisible{visibility:hidden!important}.z-n1{z-index:-1!important}.z-0{z-index:0!important}.z-1{z-index:1!important}.z-2{z-index:2!important}.z-3{z-index:3!important}@media (min-width:540px){.float-sm-start{float:left!important}.float-sm-end{float:right!important}.float-sm-none{float:none!important}.object-fit-sm-contain{object-fit:contain!important}.object-fit-sm-cover{object-fit:cover!important}.object-fit-sm-fill{object-fit:fill!important}.object-fit-sm-scale{object-fit:scale-down!important}.object-fit-sm-none{object-fit:none!important}.d-sm-inline{display:inline!important}.d-sm-inline-block{display:inline-block!important}.d-sm-block{display:block!important}.d-sm-grid{display:grid!important}.d-sm-inline-grid{display:inline-grid!important}.d-sm-table{display:table!important}.d-sm-table-row{display:table-row!important}.d-sm-table-cell{display:table-cell!important}.d-sm-flex{display:flex!important}.d-sm-inline-flex{display:inline-flex!important}.d-sm-none{display:none!important}.flex-sm-fill{flex:1 1 auto!important}.flex-sm-row{flex-direction:row!important}.flex-sm-column{flex-direction:column!important}.flex-sm-row-reverse{flex-direction:row-reverse!important}.flex-sm-column-reverse{flex-direction:column-reverse!important}.flex-sm-grow-0{flex-grow:0!important}.flex-sm-grow-1{flex-grow:1!important}.flex-sm-shrink-0{flex-shrink:0!important}.flex-sm-shrink-1{flex-shrink:1!important}.flex-sm-wrap{flex-wrap:wrap!important}.flex-sm-nowrap{flex-wrap:nowrap!important}.flex-sm-wrap-reverse{flex-wrap:wrap-reverse!important}.justify-content-sm-start{justify-content:flex-start!important}.justify-content-sm-end{justify-content:flex-end!important}.justify-content-sm-center{justify-content:center!important}.justify-content-sm-between{justify-content:space-between!important}.justify-content-sm-around{justify-content:space-around!important}.justify-content-sm-evenly{justify-content:space-evenly!important}.align-items-sm-start{align-items:flex-start!important}.align-items-sm-end{align-items:flex-end!important}.align-items-sm-center{align-items:center!important}.align-items-sm-baseline{align-items:baseline!important}.align-items-sm-stretch{align-items:stretch!important}.align-content-sm-start{align-content:flex-start!important}.align-content-sm-end{align-content:flex-end!important}.align-content-sm-center{align-content:center!important}.align-content-sm-between{align-content:space-between!important}.align-content-sm-around{align-content:space-around!important}.align-content-sm-stretch{align-content:stretch!important}.align-self-sm-auto{align-self:auto!important}.align-self-sm-start{align-self:flex-start!important}.align-self-sm-end{align-self:flex-end!important}.align-self-sm-center{align-self:center!important}.align-self-sm-baseline{align-self:baseline!important}.align-self-sm-stretch{align-self:stretch!important}.order-sm-first{order:-1!important}.order-sm-0{order:0!important}.order-sm-1{order:1!important}.order-sm-2{order:2!important}.order-sm-3{order:3!important}.order-sm-4{order:4!important}.order-sm-5{order:5!important}.order-sm-last{order:6!important}.m-sm-0{margin:0!important}.m-sm-1{margin:.25rem!important}.m-sm-2{margin:.5rem!important}.m-sm-3{margin:1rem!important}.m-sm-4{margin:1.5rem!important}.m-sm-5{margin:3rem!important}.m-sm-auto{margin:auto!important}.mx-sm-0{margin-left:0!important;margin-right:0!important}.mx-sm-1{margin-left:.25rem!important;margin-right:.25rem!important}.mx-sm-2{margin-left:.5rem!important;margin-right:.5rem!important}.mx-sm-3{margin-left:1rem!important;margin-right:1rem!important}.mx-sm-4{margin-left:1.5rem!important;margin-right:1.5rem!important}.mx-sm-5{margin-left:3rem!important;margin-right:3rem!important}.mx-sm-auto{margin-left:auto!important;margin-right:auto!important}.my-sm-0{margin-bottom:0!important;margin-top:0!important}.my-sm-1{margin-bottom:.25rem!important;margin-top:.25rem!important}.my-sm-2{margin-bottom:.5rem!important;margin-top:.5rem!important}.my-sm-3{margin-bottom:1rem!important;margin-top:1rem!important}.my-sm-4{margin-bottom:1.5rem!important;margin-top:1.5rem!important}.my-sm-5{margin-bottom:3rem!important;margin-top:3rem!important}.my-sm-auto{margin-bottom:auto!important;margin-top:auto!important}.mt-sm-0{margin-top:0!important}.mt-sm-1{margin-top:.25rem!important}.mt-sm-2{margin-top:.5rem!important}.mt-sm-3{margin-top:1rem!important}.mt-sm-4{margin-top:1.5rem!important}.mt-sm-5{margin-top:3rem!important}.mt-sm-auto{margin-top:auto!important}.me-sm-0{margin-right:0!important}.me-sm-1{margin-right:.25rem!important}.me-sm-2{margin-right:.5rem!important}.me-sm-3{margin-right:1rem!important}.me-sm-4{margin-right:1.5rem!important}.me-sm-5{margin-right:3rem!important}.me-sm-auto{margin-right:auto!important}.mb-sm-0{margin-bottom:0!important}.mb-sm-1{margin-bottom:.25rem!important}.mb-sm-2{margin-bottom:.5rem!important}.mb-sm-3{margin-bottom:1rem!important}.mb-sm-4{margin-bottom:1.5rem!important}.mb-sm-5{margin-bottom:3rem!important}.mb-sm-auto{margin-bottom:auto!important}.ms-sm-0{margin-left:0!important}.ms-sm-1{margin-left:.25rem!important}.ms-sm-2{margin-left:.5rem!important}.ms-sm-3{margin-left:1rem!important}.ms-sm-4{margin-left:1.5rem!important}.ms-sm-5{margin-left:3rem!important}.ms-sm-auto{margin-left:auto!important}.p-sm-0{padding:0!important}.p-sm-1{padding:.25rem!important}.p-sm-2{padding:.5rem!important}.p-sm-3{padding:1rem!important}.p-sm-4{padding:1.5rem!important}.p-sm-5{padding:3rem!important}.px-sm-0{padding-left:0!important;padding-right:0!important}.px-sm-1{padding-left:.25rem!important;padding-right:.25rem!important}.px-sm-2{padding-left:.5rem!important;padding-right:.5rem!important}.px-sm-3{padding-left:1rem!important;padding-right:1rem!important}.px-sm-4{padding-left:1.5rem!important;padding-right:1.5rem!important}.px-sm-5{padding-left:3rem!important;padding-right:3rem!important}.py-sm-0{padding-bottom:0!important;padding-top:0!important}.py-sm-1{padding-bottom:.25rem!important;padding-top:.25rem!important}.py-sm-2{padding-bottom:.5rem!important;padding-top:.5rem!important}.py-sm-3{padding-bottom:1rem!important;padding-top:1rem!important}.py-sm-4{padding-bottom:1.5rem!important;padding-top:1.5rem!important}.py-sm-5{padding-bottom:3rem!important;padding-top:3rem!important}.pt-sm-0{padding-top:0!important}.pt-sm-1{padding-top:.25rem!important}.pt-sm-2{padding-top:.5rem!important}.pt-sm-3{padding-top:1rem!important}.pt-sm-4{padding-top:1.5rem!important}.pt-sm-5{padding-top:3rem!important}.pe-sm-0{padding-right:0!important}.pe-sm-1{padding-right:.25rem!important}.pe-sm-2{padding-right:.5rem!important}.pe-sm-3{padding-right:1rem!important}.pe-sm-4{padding-right:1.5rem!important}.pe-sm-5{padding-right:3rem!important}.pb-sm-0{padding-bottom:0!important}.pb-sm-1{padding-bottom:.25rem!important}.pb-sm-2{padding-bottom:.5rem!important}.pb-sm-3{padding-bottom:1rem!important}.pb-sm-4{padding-bottom:1.5rem!important}.pb-sm-5{padding-bottom:3rem!important}.ps-sm-0{padding-left:0!important}.ps-sm-1{padding-left:.25rem!important}.ps-sm-2{padding-left:.5rem!important}.ps-sm-3{padding-left:1rem!important}.ps-sm-4{padding-left:1.5rem!important}.ps-sm-5{padding-left:3rem!important}.gap-sm-0{gap:0!important}.gap-sm-1{gap:.25rem!important}.gap-sm-2{gap:.5rem!important}.gap-sm-3{gap:1rem!important}.gap-sm-4{gap:1.5rem!important}.gap-sm-5{gap:3rem!important}.row-gap-sm-0{row-gap:0!important}.row-gap-sm-1{row-gap:.25rem!important}.row-gap-sm-2{row-gap:.5rem!important}.row-gap-sm-3{row-gap:1rem!important}.row-gap-sm-4{row-gap:1.5rem!important}.row-gap-sm-5{row-gap:3rem!important}.column-gap-sm-0{column-gap:0!important}.column-gap-sm-1{column-gap:.25rem!important}.column-gap-sm-2{column-gap:.5rem!important}.column-gap-sm-3{column-gap:1rem!important}.column-gap-sm-4{column-gap:1.5rem!important}.column-gap-sm-5{column-gap:3rem!important}.text-sm-start{text-align:left!important}.text-sm-end{text-align:right!important}.text-sm-center{text-align:center!important}}@media (min-width:720px){.float-md-start{float:left!important}.float-md-end{float:right!important}.float-md-none{float:none!important}.object-fit-md-contain{object-fit:contain!important}.object-fit-md-cover{object-fit:cover!important}.object-fit-md-fill{object-fit:fill!important}.object-fit-md-scale{object-fit:scale-down!important}.object-fit-md-none{object-fit:none!important}.d-md-inline{display:inline!important}.d-md-inline-block{display:inline-block!important}.d-md-block{display:block!important}.d-md-grid{display:grid!important}.d-md-inline-grid{display:inline-grid!important}.d-md-table{display:table!important}.d-md-table-row{display:table-row!important}.d-md-table-cell{display:table-cell!important}.d-md-flex{display:flex!important}.d-md-inline-flex{display:inline-flex!important}.d-md-none{display:none!important}.flex-md-fill{flex:1 1 auto!important}.flex-md-row{flex-direction:row!important}.flex-md-column{flex-direction:column!important}.flex-md-row-reverse{flex-direction:row-reverse!important}.flex-md-column-reverse{flex-direction:column-reverse!important}.flex-md-grow-0{flex-grow:0!important}.flex-md-grow-1{flex-grow:1!important}.flex-md-shrink-0{flex-shrink:0!important}.flex-md-shrink-1{flex-shrink:1!important}.flex-md-wrap{flex-wrap:wrap!important}.flex-md-nowrap{flex-wrap:nowrap!important}.flex-md-wrap-reverse{flex-wrap:wrap-reverse!important}.justify-content-md-start{justify-content:flex-start!important}.justify-content-md-end{justify-content:flex-end!important}.justify-content-md-center{justify-content:center!important}.justify-content-md-between{justify-content:space-between!important}.justify-content-md-around{justify-content:space-around!important}.justify-content-md-evenly{justify-content:space-evenly!important}.align-items-md-start{align-items:flex-start!important}.align-items-md-end{align-items:flex-end!important}.align-items-md-center{align-items:center!important}.align-items-md-baseline{align-items:baseline!important}.align-items-md-stretch{align-items:stretch!important}.align-content-md-start{align-content:flex-start!important}.align-content-md-end{align-content:flex-end!important}.align-content-md-center{align-content:center!important}.align-content-md-between{align-content:space-between!important}.align-content-md-around{align-content:space-around!important}.align-content-md-stretch{align-content:stretch!important}.align-self-md-auto{align-self:auto!important}.align-self-md-start{align-self:flex-start!important}.align-self-md-end{align-self:flex-end!important}.align-self-md-center{align-self:center!important}.align-self-md-baseline{align-self:baseline!important}.align-self-md-stretch{align-self:stretch!important}.order-md-first{order:-1!important}.order-md-0{order:0!important}.order-md-1{order:1!important}.order-md-2{order:2!important}.order-md-3{order:3!important}.order-md-4{order:4!important}.order-md-5{order:5!important}.order-md-last{order:6!important}.m-md-0{margin:0!important}.m-md-1{margin:.25rem!important}.m-md-2{margin:.5rem!important}.m-md-3{margin:1rem!important}.m-md-4{margin:1.5rem!important}.m-md-5{margin:3rem!important}.m-md-auto{margin:auto!important}.mx-md-0{margin-left:0!important;margin-right:0!important}.mx-md-1{margin-left:.25rem!important;margin-right:.25rem!important}.mx-md-2{margin-left:.5rem!important;margin-right:.5rem!important}.mx-md-3{margin-left:1rem!important;margin-right:1rem!important}.mx-md-4{margin-left:1.5rem!important;margin-right:1.5rem!important}.mx-md-5{margin-left:3rem!important;margin-right:3rem!important}.mx-md-auto{margin-left:auto!important;margin-right:auto!important}.my-md-0{margin-bottom:0!important;margin-top:0!important}.my-md-1{margin-bottom:.25rem!important;margin-top:.25rem!important}.my-md-2{margin-bottom:.5rem!important;margin-top:.5rem!important}.my-md-3{margin-bottom:1rem!important;margin-top:1rem!important}.my-md-4{margin-bottom:1.5rem!important;margin-top:1.5rem!important}.my-md-5{margin-bottom:3rem!important;margin-top:3rem!important}.my-md-auto{margin-bottom:auto!important;margin-top:auto!important}.mt-md-0{margin-top:0!important}.mt-md-1{margin-top:.25rem!important}.mt-md-2{margin-top:.5rem!important}.mt-md-3{margin-top:1rem!important}.mt-md-4{margin-top:1.5rem!important}.mt-md-5{margin-top:3rem!important}.mt-md-auto{margin-top:auto!important}.me-md-0{margin-right:0!important}.me-md-1{margin-right:.25rem!important}.me-md-2{margin-right:.5rem!important}.me-md-3{margin-right:1rem!important}.me-md-4{margin-right:1.5rem!important}.me-md-5{margin-right:3rem!important}.me-md-auto{margin-right:auto!important}.mb-md-0{margin-bottom:0!important}.mb-md-1{margin-bottom:.25rem!important}.mb-md-2{margin-bottom:.5rem!important}.mb-md-3{margin-bottom:1rem!important}.mb-md-4{margin-bottom:1.5rem!important}.mb-md-5{margin-bottom:3rem!important}.mb-md-auto{margin-bottom:auto!important}.ms-md-0{margin-left:0!important}.ms-md-1{margin-left:.25rem!important}.ms-md-2{margin-left:.5rem!important}.ms-md-3{margin-left:1rem!important}.ms-md-4{margin-left:1.5rem!important}.ms-md-5{margin-left:3rem!important}.ms-md-auto{margin-left:auto!important}.p-md-0{padding:0!important}.p-md-1{padding:.25rem!important}.p-md-2{padding:.5rem!important}.p-md-3{padding:1rem!important}.p-md-4{padding:1.5rem!important}.p-md-5{padding:3rem!important}.px-md-0{padding-left:0!important;padding-right:0!important}.px-md-1{padding-left:.25rem!important;padding-right:.25rem!important}.px-md-2{padding-left:.5rem!important;padding-right:.5rem!important}.px-md-3{padding-left:1rem!important;padding-right:1rem!important}.px-md-4{padding-left:1.5rem!important;padding-right:1.5rem!important}.px-md-5{padding-left:3rem!important;padding-right:3rem!important}.py-md-0{padding-bottom:0!important;padding-top:0!important}.py-md-1{padding-bottom:.25rem!important;padding-top:.25rem!important}.py-md-2{padding-bottom:.5rem!important;padding-top:.5rem!important}.py-md-3{padding-bottom:1rem!important;padding-top:1rem!important}.py-md-4{padding-bottom:1.5rem!important;padding-top:1.5rem!important}.py-md-5{padding-bottom:3rem!important;padding-top:3rem!important}.pt-md-0{padding-top:0!important}.pt-md-1{padding-top:.25rem!important}.pt-md-2{padding-top:.5rem!important}.pt-md-3{padding-top:1rem!important}.pt-md-4{padding-top:1.5rem!important}.pt-md-5{padding-top:3rem!important}.pe-md-0{padding-right:0!important}.pe-md-1{padding-right:.25rem!important}.pe-md-2{padding-right:.5rem!important}.pe-md-3{padding-right:1rem!important}.pe-md-4{padding-right:1.5rem!important}.pe-md-5{padding-right:3rem!important}.pb-md-0{padding-bottom:0!important}.pb-md-1{padding-bottom:.25rem!important}.pb-md-2{padding-bottom:.5rem!important}.pb-md-3{padding-bottom:1rem!important}.pb-md-4{padding-bottom:1.5rem!important}.pb-md-5{padding-bottom:3rem!important}.ps-md-0{padding-left:0!important}.ps-md-1{padding-left:.25rem!important}.ps-md-2{padding-left:.5rem!important}.ps-md-3{padding-left:1rem!important}.ps-md-4{padding-left:1.5rem!important}.ps-md-5{padding-left:3rem!important}.gap-md-0{gap:0!important}.gap-md-1{gap:.25rem!important}.gap-md-2{gap:.5rem!important}.gap-md-3{gap:1rem!important}.gap-md-4{gap:1.5rem!important}.gap-md-5{gap:3rem!important}.row-gap-md-0{row-gap:0!important}.row-gap-md-1{row-gap:.25rem!important}.row-gap-md-2{row-gap:.5rem!important}.row-gap-md-3{row-gap:1rem!important}.row-gap-md-4{row-gap:1.5rem!important}.row-gap-md-5{row-gap:3rem!important}.column-gap-md-0{column-gap:0!important}.column-gap-md-1{column-gap:.25rem!important}.column-gap-md-2{column-gap:.5rem!important}.column-gap-md-3{column-gap:1rem!important}.column-gap-md-4{column-gap:1.5rem!important}.column-gap-md-5{column-gap:3rem!important}.text-md-start{text-align:left!important}.text-md-end{text-align:right!important}.text-md-center{text-align:center!important}}@media (min-width:960px){.float-lg-start{float:left!important}.float-lg-end{float:right!important}.float-lg-none{float:none!important}.object-fit-lg-contain{object-fit:contain!important}.object-fit-lg-cover{object-fit:cover!important}.object-fit-lg-fill{object-fit:fill!important}.object-fit-lg-scale{object-fit:scale-down!important}.object-fit-lg-none{object-fit:none!important}.d-lg-inline{display:inline!important}.d-lg-inline-block{display:inline-block!important}.d-lg-block{display:block!important}.d-lg-grid{display:grid!important}.d-lg-inline-grid{display:inline-grid!important}.d-lg-table{display:table!important}.d-lg-table-row{display:table-row!important}.d-lg-table-cell{display:table-cell!important}.d-lg-flex{display:flex!important}.d-lg-inline-flex{display:inline-flex!important}.d-lg-none{display:none!important}.flex-lg-fill{flex:1 1 auto!important}.flex-lg-row{flex-direction:row!important}.flex-lg-column{flex-direction:column!important}.flex-lg-row-reverse{flex-direction:row-reverse!important}.flex-lg-column-reverse{flex-direction:column-reverse!important}.flex-lg-grow-0{flex-grow:0!important}.flex-lg-grow-1{flex-grow:1!important}.flex-lg-shrink-0{flex-shrink:0!important}.flex-lg-shrink-1{flex-shrink:1!important}.flex-lg-wrap{flex-wrap:wrap!important}.flex-lg-nowrap{flex-wrap:nowrap!important}.flex-lg-wrap-reverse{flex-wrap:wrap-reverse!important}.justify-content-lg-start{justify-content:flex-start!important}.justify-content-lg-end{justify-content:flex-end!important}.justify-content-lg-center{justify-content:center!important}.justify-content-lg-between{justify-content:space-between!important}.justify-content-lg-around{justify-content:space-around!important}.justify-content-lg-evenly{justify-content:space-evenly!important}.align-items-lg-start{align-items:flex-start!important}.align-items-lg-end{align-items:flex-end!important}.align-items-lg-center{align-items:center!important}.align-items-lg-baseline{align-items:baseline!important}.align-items-lg-stretch{align-items:stretch!important}.align-content-lg-start{align-content:flex-start!important}.align-content-lg-end{align-content:flex-end!important}.align-content-lg-center{align-content:center!important}.align-content-lg-between{align-content:space-between!important}.align-content-lg-around{align-content:space-around!important}.align-content-lg-stretch{align-content:stretch!important}.align-self-lg-auto{align-self:auto!important}.align-self-lg-start{align-self:flex-start!important}.align-self-lg-end{align-self:flex-end!important}.align-self-lg-center{align-self:center!important}.align-self-lg-baseline{align-self:baseline!important}.align-self-lg-stretch{align-self:stretch!important}.order-lg-first{order:-1!important}.order-lg-0{order:0!important}.order-lg-1{order:1!important}.order-lg-2{order:2!important}.order-lg-3{order:3!important}.order-lg-4{order:4!important}.order-lg-5{order:5!important}.order-lg-last{order:6!important}.m-lg-0{margin:0!important}.m-lg-1{margin:.25rem!important}.m-lg-2{margin:.5rem!important}.m-lg-3{margin:1rem!important}.m-lg-4{margin:1.5rem!important}.m-lg-5{margin:3rem!important}.m-lg-auto{margin:auto!important}.mx-lg-0{margin-left:0!important;margin-right:0!important}.mx-lg-1{margin-left:.25rem!important;margin-right:.25rem!important}.mx-lg-2{margin-left:.5rem!important;margin-right:.5rem!important}.mx-lg-3{margin-left:1rem!important;margin-right:1rem!important}.mx-lg-4{margin-left:1.5rem!important;margin-right:1.5rem!important}.mx-lg-5{margin-left:3rem!important;margin-right:3rem!important}.mx-lg-auto{margin-left:auto!important;margin-right:auto!important}.my-lg-0{margin-bottom:0!important;margin-top:0!important}.my-lg-1{margin-bottom:.25rem!important;margin-top:.25rem!important}.my-lg-2{margin-bottom:.5rem!important;margin-top:.5rem!important}.my-lg-3{margin-bottom:1rem!important;margin-top:1rem!important}.my-lg-4{margin-bottom:1.5rem!important;margin-top:1.5rem!important}.my-lg-5{margin-bottom:3rem!important;margin-top:3rem!important}.my-lg-auto{margin-bottom:auto!important;margin-top:auto!important}.mt-lg-0{margin-top:0!important}.mt-lg-1{margin-top:.25rem!important}.mt-lg-2{margin-top:.5rem!important}.mt-lg-3{margin-top:1rem!important}.mt-lg-4{margin-top:1.5rem!important}.mt-lg-5{margin-top:3rem!important}.mt-lg-auto{margin-top:auto!important}.me-lg-0{margin-right:0!important}.me-lg-1{margin-right:.25rem!important}.me-lg-2{margin-right:.5rem!important}.me-lg-3{margin-right:1rem!important}.me-lg-4{margin-right:1.5rem!important}.me-lg-5{margin-right:3rem!important}.me-lg-auto{margin-right:auto!important}.mb-lg-0{margin-bottom:0!important}.mb-lg-1{margin-bottom:.25rem!important}.mb-lg-2{margin-bottom:.5rem!important}.mb-lg-3{margin-bottom:1rem!important}.mb-lg-4{margin-bottom:1.5rem!important}.mb-lg-5{margin-bottom:3rem!important}.mb-lg-auto{margin-bottom:auto!important}.ms-lg-0{margin-left:0!important}.ms-lg-1{margin-left:.25rem!important}.ms-lg-2{margin-left:.5rem!important}.ms-lg-3{margin-left:1rem!important}.ms-lg-4{margin-left:1.5rem!important}.ms-lg-5{margin-left:3rem!important}.ms-lg-auto{margin-left:auto!important}.p-lg-0{padding:0!important}.p-lg-1{padding:.25rem!important}.p-lg-2{padding:.5rem!important}.p-lg-3{padding:1rem!important}.p-lg-4{padding:1.5rem!important}.p-lg-5{padding:3rem!important}.px-lg-0{padding-left:0!important;padding-right:0!important}.px-lg-1{padding-left:.25rem!important;padding-right:.25rem!important}.px-lg-2{padding-left:.5rem!important;padding-right:.5rem!important}.px-lg-3{padding-left:1rem!important;padding-right:1rem!important}.px-lg-4{padding-left:1.5rem!important;padding-right:1.5rem!important}.px-lg-5{padding-left:3rem!important;padding-right:3rem!important}.py-lg-0{padding-bottom:0!important;padding-top:0!important}.py-lg-1{padding-bottom:.25rem!important;padding-top:.25rem!important}.py-lg-2{padding-bottom:.5rem!important;padding-top:.5rem!important}.py-lg-3{padding-bottom:1rem!important;padding-top:1rem!important}.py-lg-4{padding-bottom:1.5rem!important;padding-top:1.5rem!important}.py-lg-5{padding-bottom:3rem!important;padding-top:3rem!important}.pt-lg-0{padding-top:0!important}.pt-lg-1{padding-top:.25rem!important}.pt-lg-2{padding-top:.5rem!important}.pt-lg-3{padding-top:1rem!important}.pt-lg-4{padding-top:1.5rem!important}.pt-lg-5{padding-top:3rem!important}.pe-lg-0{padding-right:0!important}.pe-lg-1{padding-right:.25rem!important}.pe-lg-2{padding-right:.5rem!important}.pe-lg-3{padding-right:1rem!important}.pe-lg-4{padding-right:1.5rem!important}.pe-lg-5{padding-right:3rem!important}.pb-lg-0{padding-bottom:0!important}.pb-lg-1{padding-bottom:.25rem!important}.pb-lg-2{padding-bottom:.5rem!important}.pb-lg-3{padding-bottom:1rem!important}.pb-lg-4{padding-bottom:1.5rem!important}.pb-lg-5{padding-bottom:3rem!important}.ps-lg-0{padding-left:0!important}.ps-lg-1{padding-left:.25rem!important}.ps-lg-2{padding-left:.5rem!important}.ps-lg-3{padding-left:1rem!important}.ps-lg-4{padding-left:1.5rem!important}.ps-lg-5{padding-left:3rem!important}.gap-lg-0{gap:0!important}.gap-lg-1{gap:.25rem!important}.gap-lg-2{gap:.5rem!important}.gap-lg-3{gap:1rem!important}.gap-lg-4{gap:1.5rem!important}.gap-lg-5{gap:3rem!important}.row-gap-lg-0{row-gap:0!important}.row-gap-lg-1{row-gap:.25rem!important}.row-gap-lg-2{row-gap:.5rem!important}.row-gap-lg-3{row-gap:1rem!important}.row-gap-lg-4{row-gap:1.5rem!important}.row-gap-lg-5{row-gap:3rem!important}.column-gap-lg-0{column-gap:0!important}.column-gap-lg-1{column-gap:.25rem!important}.column-gap-lg-2{column-gap:.5rem!important}.column-gap-lg-3{column-gap:1rem!important}.column-gap-lg-4{column-gap:1.5rem!important}.column-gap-lg-5{column-gap:3rem!important}.text-lg-start{text-align:left!important}.text-lg-end{text-align:right!important}.text-lg-center{text-align:center!important}}@media (min-width:1200px){.float-xl-start{float:left!important}.float-xl-end{float:right!important}.float-xl-none{float:none!important}.object-fit-xl-contain{object-fit:contain!important}.object-fit-xl-cover{object-fit:cover!important}.object-fit-xl-fill{object-fit:fill!important}.object-fit-xl-scale{object-fit:scale-down!important}.object-fit-xl-none{object-fit:none!important}.d-xl-inline{display:inline!important}.d-xl-inline-block{display:inline-block!important}.d-xl-block{display:block!important}.d-xl-grid{display:grid!important}.d-xl-inline-grid{display:inline-grid!important}.d-xl-table{display:table!important}.d-xl-table-row{display:table-row!important}.d-xl-table-cell{display:table-cell!important}.d-xl-flex{display:flex!important}.d-xl-inline-flex{display:inline-flex!important}.d-xl-none{display:none!important}.flex-xl-fill{flex:1 1 auto!important}.flex-xl-row{flex-direction:row!important}.flex-xl-column{flex-direction:column!important}.flex-xl-row-reverse{flex-direction:row-reverse!important}.flex-xl-column-reverse{flex-direction:column-reverse!important}.flex-xl-grow-0{flex-grow:0!important}.flex-xl-grow-1{flex-grow:1!important}.flex-xl-shrink-0{flex-shrink:0!important}.flex-xl-shrink-1{flex-shrink:1!important}.flex-xl-wrap{flex-wrap:wrap!important}.flex-xl-nowrap{flex-wrap:nowrap!important}.flex-xl-wrap-reverse{flex-wrap:wrap-reverse!important}.justify-content-xl-start{justify-content:flex-start!important}.justify-content-xl-end{justify-content:flex-end!important}.justify-content-xl-center{justify-content:center!important}.justify-content-xl-between{justify-content:space-between!important}.justify-content-xl-around{justify-content:space-around!important}.justify-content-xl-evenly{justify-content:space-evenly!important}.align-items-xl-start{align-items:flex-start!important}.align-items-xl-end{align-items:flex-end!important}.align-items-xl-center{align-items:center!important}.align-items-xl-baseline{align-items:baseline!important}.align-items-xl-stretch{align-items:stretch!important}.align-content-xl-start{align-content:flex-start!important}.align-content-xl-end{align-content:flex-end!important}.align-content-xl-center{align-content:center!important}.align-content-xl-between{align-content:space-between!important}.align-content-xl-around{align-content:space-around!important}.align-content-xl-stretch{align-content:stretch!important}.align-self-xl-auto{align-self:auto!important}.align-self-xl-start{align-self:flex-start!important}.align-self-xl-end{align-self:flex-end!important}.align-self-xl-center{align-self:center!important}.align-self-xl-baseline{align-self:baseline!important}.align-self-xl-stretch{align-self:stretch!important}.order-xl-first{order:-1!important}.order-xl-0{order:0!important}.order-xl-1{order:1!important}.order-xl-2{order:2!important}.order-xl-3{order:3!important}.order-xl-4{order:4!important}.order-xl-5{order:5!important}.order-xl-last{order:6!important}.m-xl-0{margin:0!important}.m-xl-1{margin:.25rem!important}.m-xl-2{margin:.5rem!important}.m-xl-3{margin:1rem!important}.m-xl-4{margin:1.5rem!important}.m-xl-5{margin:3rem!important}.m-xl-auto{margin:auto!important}.mx-xl-0{margin-left:0!important;margin-right:0!important}.mx-xl-1{margin-left:.25rem!important;margin-right:.25rem!important}.mx-xl-2{margin-left:.5rem!important;margin-right:.5rem!important}.mx-xl-3{margin-left:1rem!important;margin-right:1rem!important}.mx-xl-4{margin-left:1.5rem!important;margin-right:1.5rem!important}.mx-xl-5{margin-left:3rem!important;margin-right:3rem!important}.mx-xl-auto{margin-left:auto!important;margin-right:auto!important}.my-xl-0{margin-bottom:0!important;margin-top:0!important}.my-xl-1{margin-bottom:.25rem!important;margin-top:.25rem!important}.my-xl-2{margin-bottom:.5rem!important;margin-top:.5rem!important}.my-xl-3{margin-bottom:1rem!important;margin-top:1rem!important}.my-xl-4{margin-bottom:1.5rem!important;margin-top:1.5rem!important}.my-xl-5{margin-bottom:3rem!important;margin-top:3rem!important}.my-xl-auto{margin-bottom:auto!important;margin-top:auto!important}.mt-xl-0{margin-top:0!important}.mt-xl-1{margin-top:.25rem!important}.mt-xl-2{margin-top:.5rem!important}.mt-xl-3{margin-top:1rem!important}.mt-xl-4{margin-top:1.5rem!important}.mt-xl-5{margin-top:3rem!important}.mt-xl-auto{margin-top:auto!important}.me-xl-0{margin-right:0!important}.me-xl-1{margin-right:.25rem!important}.me-xl-2{margin-right:.5rem!important}.me-xl-3{margin-right:1rem!important}.me-xl-4{margin-right:1.5rem!important}.me-xl-5{margin-right:3rem!important}.me-xl-auto{margin-right:auto!important}.mb-xl-0{margin-bottom:0!important}.mb-xl-1{margin-bottom:.25rem!important}.mb-xl-2{margin-bottom:.5rem!important}.mb-xl-3{margin-bottom:1rem!important}.mb-xl-4{margin-bottom:1.5rem!important}.mb-xl-5{margin-bottom:3rem!important}.mb-xl-auto{margin-bottom:auto!important}.ms-xl-0{margin-left:0!important}.ms-xl-1{margin-left:.25rem!important}.ms-xl-2{margin-left:.5rem!important}.ms-xl-3{margin-left:1rem!important}.ms-xl-4{margin-left:1.5rem!important}.ms-xl-5{margin-left:3rem!important}.ms-xl-auto{margin-left:auto!important}.p-xl-0{padding:0!important}.p-xl-1{padding:.25rem!important}.p-xl-2{padding:.5rem!important}.p-xl-3{padding:1rem!important}.p-xl-4{padding:1.5rem!important}.p-xl-5{padding:3rem!important}.px-xl-0{padding-left:0!important;padding-right:0!important}.px-xl-1{padding-left:.25rem!important;padding-right:.25rem!important}.px-xl-2{padding-left:.5rem!important;padding-right:.5rem!important}.px-xl-3{padding-left:1rem!important;padding-right:1rem!important}.px-xl-4{padding-left:1.5rem!important;padding-right:1.5rem!important}.px-xl-5{padding-left:3rem!important;padding-right:3rem!important}.py-xl-0{padding-bottom:0!important;padding-top:0!important}.py-xl-1{padding-bottom:.25rem!important;padding-top:.25rem!important}.py-xl-2{padding-bottom:.5rem!important;padding-top:.5rem!important}.py-xl-3{padding-bottom:1rem!important;padding-top:1rem!important}.py-xl-4{padding-bottom:1.5rem!important;padding-top:1.5rem!important}.py-xl-5{padding-bottom:3rem!important;padding-top:3rem!important}.pt-xl-0{padding-top:0!important}.pt-xl-1{padding-top:.25rem!important}.pt-xl-2{padding-top:.5rem!important}.pt-xl-3{padding-top:1rem!important}.pt-xl-4{padding-top:1.5rem!important}.pt-xl-5{padding-top:3rem!important}.pe-xl-0{padding-right:0!important}.pe-xl-1{padding-right:.25rem!important}.pe-xl-2{padding-right:.5rem!important}.pe-xl-3{padding-right:1rem!important}.pe-xl-4{padding-right:1.5rem!important}.pe-xl-5{padding-right:3rem!important}.pb-xl-0{padding-bottom:0!important}.pb-xl-1{padding-bottom:.25rem!important}.pb-xl-2{padding-bottom:.5rem!important}.pb-xl-3{padding-bottom:1rem!important}.pb-xl-4{padding-bottom:1.5rem!important}.pb-xl-5{padding-bottom:3rem!important}.ps-xl-0{padding-left:0!important}.ps-xl-1{padding-left:.25rem!important}.ps-xl-2{padding-left:.5rem!important}.ps-xl-3{padding-left:1rem!important}.ps-xl-4{padding-left:1.5rem!important}.ps-xl-5{padding-left:3rem!important}.gap-xl-0{gap:0!important}.gap-xl-1{gap:.25rem!important}.gap-xl-2{gap:.5rem!important}.gap-xl-3{gap:1rem!important}.gap-xl-4{gap:1.5rem!important}.gap-xl-5{gap:3rem!important}.row-gap-xl-0{row-gap:0!important}.row-gap-xl-1{row-gap:.25rem!important}.row-gap-xl-2{row-gap:.5rem!important}.row-gap-xl-3{row-gap:1rem!important}.row-gap-xl-4{row-gap:1.5rem!important}.row-gap-xl-5{row-gap:3rem!important}.column-gap-xl-0{column-gap:0!important}.column-gap-xl-1{column-gap:.25rem!important}.column-gap-xl-2{column-gap:.5rem!important}.column-gap-xl-3{column-gap:1rem!important}.column-gap-xl-4{column-gap:1.5rem!important}.column-gap-xl-5{column-gap:3rem!important}.text-xl-start{text-align:left!important}.text-xl-end{text-align:right!important}.text-xl-center{text-align:center!important}.fs-1{font-size:2.5rem!important}.fs-2{font-size:2rem!important}.fs-3{font-size:1.75rem!important}.fs-4{font-size:1.5rem!important}}@media print{.d-print-inline{display:inline!important}.d-print-inline-block{display:inline-block!important}.d-print-block{display:block!important}.d-print-grid{display:grid!important}.d-print-inline-grid{display:inline-grid!important}.d-print-table{display:table!important}.d-print-table-row{display:table-row!important}.d-print-table-cell{display:table-cell!important}.d-print-flex{display:flex!important}.d-print-inline-flex{display:inline-flex!important}.d-print-none{display:none!important}}
+/*!
+ * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com
+ * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)
+ * Copyright 2024 Fonticons, Inc.
+ */.fa{font-family:var(--fa-style-family,"Font Awesome 6 Free");font-weight:var(--fa-style,900)}.fa,.fa-brands,.fa-classic,.fa-regular,.fa-sharp,.fa-solid,.fab,.far,.fas{-moz-osx-font-smoothing:grayscale;-webkit-font-smoothing:antialiased;display:var(--fa-display,inline-block);font-style:normal;font-variant:normal;line-height:1;text-rendering:auto}.fa-classic,.fa-regular,.fa-solid,.far,.fas{font-family:Font Awesome\ 6 Free}.fa-brands,.fab{font-family:Font Awesome\ 6 Brands}.fa-1x{font-size:1em}.fa-2x{font-size:2em}.fa-3x{font-size:3em}.fa-4x{font-size:4em}.fa-5x{font-size:5em}.fa-6x{font-size:6em}.fa-7x{font-size:7em}.fa-8x{font-size:8em}.fa-9x{font-size:9em}.fa-10x{font-size:10em}.fa-2xs{font-size:.625em;line-height:.1em;vertical-align:.225em}.fa-xs{font-size:.75em;line-height:.0833333337em;vertical-align:.125em}.fa-sm{font-size:.875em;line-height:.0714285718em;vertical-align:.0535714295em}.fa-lg{font-size:1.25em;line-height:.05em;vertical-align:-.075em}.fa-xl{font-size:1.5em;line-height:.0416666682em;vertical-align:-.125em}.fa-2xl{font-size:2em;line-height:.03125em;vertical-align:-.1875em}.fa-fw{text-align:center;width:1.25em}.fa-ul{list-style-type:none;margin-left:var(--fa-li-margin,2.5em);padding-left:0}.fa-ul>li{position:relative}.fa-li{left:calc(var(--fa-li-width, 2em)*-1);line-height:inherit;position:absolute;text-align:center;width:var(--fa-li-width,2em)}.fa-border{border-color:var(--fa-border-color,#eee);border-radius:var(--fa-border-radius,.1em);border-style:var(--fa-border-style,solid);border-width:var(--fa-border-width,.08em);padding:var(--fa-border-padding,.2em .25em .15em)}.fa-pull-left{float:left;margin-right:var(--fa-pull-margin,.3em)}.fa-pull-right{float:right;margin-left:var(--fa-pull-margin,.3em)}.fa-beat{animation-delay:var(--fa-animation-delay,0s);animation-direction:var(--fa-animation-direction,normal);animation-duration:var(--fa-animation-duration,1s);animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-name:fa-beat;animation-timing-function:var(--fa-animation-timing,ease-in-out)}.fa-bounce{animation-delay:var(--fa-animation-delay,0s);animation-direction:var(--fa-animation-direction,normal);animation-duration:var(--fa-animation-duration,1s);animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-name:fa-bounce;animation-timing-function:var(--fa-animation-timing,cubic-bezier(.28,.84,.42,1))}.fa-fade{animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-name:fa-fade;animation-timing-function:var(--fa-animation-timing,cubic-bezier(.4,0,.6,1))}.fa-beat-fade,.fa-fade{animation-delay:var(--fa-animation-delay,0s);animation-direction:var(--fa-animation-direction,normal);animation-duration:var(--fa-animation-duration,1s)}.fa-beat-fade{animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-name:fa-beat-fade;animation-timing-function:var(--fa-animation-timing,cubic-bezier(.4,0,.6,1))}.fa-flip{animation-delay:var(--fa-animation-delay,0s);animation-direction:var(--fa-animation-direction,normal);animation-duration:var(--fa-animation-duration,1s);animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-name:fa-flip;animation-timing-function:var(--fa-animation-timing,ease-in-out)}.fa-shake{animation-duration:var(--fa-animation-duration,1s);animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-name:fa-shake;animation-timing-function:var(--fa-animation-timing,linear)}.fa-shake,.fa-spin{animation-delay:var(--fa-animation-delay,0s);animation-direction:var(--fa-animation-direction,normal)}.fa-spin{animation-duration:var(--fa-animation-duration,2s);animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-name:fa-spin;animation-timing-function:var(--fa-animation-timing,linear)}.fa-spin-reverse{--fa-animation-direction:reverse}.fa-pulse,.fa-spin-pulse{animation-direction:var(--fa-animation-direction,normal);animation-duration:var(--fa-animation-duration,1s);animation-iteration-count:var(--fa-animation-iteration-count,infinite);animation-name:fa-spin;animation-timing-function:var(--fa-animation-timing,steps(8))}@media (prefers-reduced-motion:reduce){.fa-beat,.fa-beat-fade,.fa-bounce,.fa-fade,.fa-flip,.fa-pulse,.fa-shake,.fa-spin,.fa-spin-pulse{animation-delay:-1ms;animation-duration:1ms;animation-iteration-count:1;transition-delay:0s;transition-duration:0s}}@keyframes fa-beat{0%,90%{transform:scale(1)}45%{transform:scale(var(--fa-beat-scale,1.25))}}@keyframes fa-bounce{0%{transform:scale(1) translateY(0)}10%{transform:scale(var(--fa-bounce-start-scale-x,1.1),var(--fa-bounce-start-scale-y,.9)) translateY(0)}30%{transform:scale(var(--fa-bounce-jump-scale-x,.9),var(--fa-bounce-jump-scale-y,1.1)) translateY(var(--fa-bounce-height,-.5em))}50%{transform:scale(var(--fa-bounce-land-scale-x,1.05),var(--fa-bounce-land-scale-y,.95)) translateY(0)}57%{transform:scale(1) translateY(var(--fa-bounce-rebound,-.125em))}64%{transform:scale(1) translateY(0)}to{transform:scale(1) translateY(0)}}@keyframes fa-fade{50%{opacity:var(--fa-fade-opacity,.4)}}@keyframes fa-beat-fade{0%,to{opacity:var(--fa-beat-fade-opacity,.4);transform:scale(1)}50%{opacity:1;transform:scale(var(--fa-beat-fade-scale,1.125))}}@keyframes fa-flip{50%{transform:rotate3d(var(--fa-flip-x,0),var(--fa-flip-y,1),var(--fa-flip-z,0),var(--fa-flip-angle,-180deg))}}@keyframes fa-shake{0%{transform:rotate(-15deg)}4%{transform:rotate(15deg)}24%,8%{transform:rotate(-18deg)}12%,28%{transform:rotate(18deg)}16%{transform:rotate(-22deg)}20%{transform:rotate(22deg)}32%{transform:rotate(-12deg)}36%{transform:rotate(12deg)}40%,to{transform:rotate(0deg)}}@keyframes fa-spin{0%{transform:rotate(0deg)}to{transform:rotate(1turn)}}.fa-rotate-90{transform:rotate(90deg)}.fa-rotate-180{transform:rotate(180deg)}.fa-rotate-270{transform:rotate(270deg)}.fa-flip-horizontal{transform:scaleX(-1)}.fa-flip-vertical{transform:scaleY(-1)}.fa-flip-both,.fa-flip-horizontal.fa-flip-vertical{transform:scale(-1)}.fa-rotate-by{transform:rotate(var(--fa-rotate-angle,0))}.fa-stack{display:inline-block;height:2em;line-height:2em;position:relative;vertical-align:middle;width:2.5em}.fa-stack-1x,.fa-stack-2x{left:0;position:absolute;text-align:center;width:100%;z-index:var(--fa-stack-z-index,auto)}.fa-stack-1x{line-height:inherit}.fa-stack-2x{font-size:2em}.fa-inverse{color:var(--fa-inverse,#fff)}.fa-0:before{content:"\30 "}.fa-1:before{content:"\31 "}.fa-2:before{content:"\32 "}.fa-3:before{content:"\33 "}.fa-4:before{content:"\34 "}.fa-5:before{content:"\35 "}.fa-6:before{content:"\36 "}.fa-7:before{content:"\37 "}.fa-8:before{content:"\38 "}.fa-9:before{content:"\39 "}.fa-fill-drip:before{content:"\f576"}.fa-arrows-to-circle:before{content:"\e4bd"}.fa-chevron-circle-right:before,.fa-circle-chevron-right:before{content:"\f138"}.fa-at:before{content:"\@"}.fa-trash-alt:before,.fa-trash-can:before{content:"\f2ed"}.fa-text-height:before{content:"\f034"}.fa-user-times:before,.fa-user-xmark:before{content:"\f235"}.fa-stethoscope:before{content:"\f0f1"}.fa-comment-alt:before,.fa-message:before{content:"\f27a"}.fa-info:before{content:"\f129"}.fa-compress-alt:before,.fa-down-left-and-up-right-to-center:before{content:"\f422"}.fa-explosion:before{content:"\e4e9"}.fa-file-alt:before,.fa-file-lines:before,.fa-file-text:before{content:"\f15c"}.fa-wave-square:before{content:"\f83e"}.fa-ring:before{content:"\f70b"}.fa-building-un:before{content:"\e4d9"}.fa-dice-three:before{content:"\f527"}.fa-calendar-alt:before,.fa-calendar-days:before{content:"\f073"}.fa-anchor-circle-check:before{content:"\e4aa"}.fa-building-circle-arrow-right:before{content:"\e4d1"}.fa-volleyball-ball:before,.fa-volleyball:before{content:"\f45f"}.fa-arrows-up-to-line:before{content:"\e4c2"}.fa-sort-desc:before,.fa-sort-down:before{content:"\f0dd"}.fa-circle-minus:before,.fa-minus-circle:before{content:"\f056"}.fa-door-open:before{content:"\f52b"}.fa-right-from-bracket:before,.fa-sign-out-alt:before{content:"\f2f5"}.fa-atom:before{content:"\f5d2"}.fa-soap:before{content:"\e06e"}.fa-heart-music-camera-bolt:before,.fa-icons:before{content:"\f86d"}.fa-microphone-alt-slash:before,.fa-microphone-lines-slash:before{content:"\f539"}.fa-bridge-circle-check:before{content:"\e4c9"}.fa-pump-medical:before{content:"\e06a"}.fa-fingerprint:before{content:"\f577"}.fa-hand-point-right:before{content:"\f0a4"}.fa-magnifying-glass-location:before,.fa-search-location:before{content:"\f689"}.fa-forward-step:before,.fa-step-forward:before{content:"\f051"}.fa-face-smile-beam:before,.fa-smile-beam:before{content:"\f5b8"}.fa-flag-checkered:before{content:"\f11e"}.fa-football-ball:before,.fa-football:before{content:"\f44e"}.fa-school-circle-exclamation:before{content:"\e56c"}.fa-crop:before{content:"\f125"}.fa-angle-double-down:before,.fa-angles-down:before{content:"\f103"}.fa-users-rectangle:before{content:"\e594"}.fa-people-roof:before{content:"\e537"}.fa-people-line:before{content:"\e534"}.fa-beer-mug-empty:before,.fa-beer:before{content:"\f0fc"}.fa-diagram-predecessor:before{content:"\e477"}.fa-arrow-up-long:before,.fa-long-arrow-up:before{content:"\f176"}.fa-burn:before,.fa-fire-flame-simple:before{content:"\f46a"}.fa-male:before,.fa-person:before{content:"\f183"}.fa-laptop:before{content:"\f109"}.fa-file-csv:before{content:"\f6dd"}.fa-menorah:before{content:"\f676"}.fa-truck-plane:before{content:"\e58f"}.fa-record-vinyl:before{content:"\f8d9"}.fa-face-grin-stars:before,.fa-grin-stars:before{content:"\f587"}.fa-bong:before{content:"\f55c"}.fa-pastafarianism:before,.fa-spaghetti-monster-flying:before{content:"\f67b"}.fa-arrow-down-up-across-line:before{content:"\e4af"}.fa-spoon:before,.fa-utensil-spoon:before{content:"\f2e5"}.fa-jar-wheat:before{content:"\e517"}.fa-envelopes-bulk:before,.fa-mail-bulk:before{content:"\f674"}.fa-file-circle-exclamation:before{content:"\e4eb"}.fa-circle-h:before,.fa-hospital-symbol:before{content:"\f47e"}.fa-pager:before{content:"\f815"}.fa-address-book:before,.fa-contact-book:before{content:"\f2b9"}.fa-strikethrough:before{content:"\f0cc"}.fa-k:before{content:"K"}.fa-landmark-flag:before{content:"\e51c"}.fa-pencil-alt:before,.fa-pencil:before{content:"\f303"}.fa-backward:before{content:"\f04a"}.fa-caret-right:before{content:"\f0da"}.fa-comments:before{content:"\f086"}.fa-file-clipboard:before,.fa-paste:before{content:"\f0ea"}.fa-code-pull-request:before{content:"\e13c"}.fa-clipboard-list:before{content:"\f46d"}.fa-truck-loading:before,.fa-truck-ramp-box:before{content:"\f4de"}.fa-user-check:before{content:"\f4fc"}.fa-vial-virus:before{content:"\e597"}.fa-sheet-plastic:before{content:"\e571"}.fa-blog:before{content:"\f781"}.fa-user-ninja:before{content:"\f504"}.fa-person-arrow-up-from-line:before{content:"\e539"}.fa-scroll-torah:before,.fa-torah:before{content:"\f6a0"}.fa-broom-ball:before,.fa-quidditch-broom-ball:before,.fa-quidditch:before{content:"\f458"}.fa-toggle-off:before{content:"\f204"}.fa-archive:before,.fa-box-archive:before{content:"\f187"}.fa-person-drowning:before{content:"\e545"}.fa-arrow-down-9-1:before,.fa-sort-numeric-desc:before,.fa-sort-numeric-down-alt:before{content:"\f886"}.fa-face-grin-tongue-squint:before,.fa-grin-tongue-squint:before{content:"\f58a"}.fa-spray-can:before{content:"\f5bd"}.fa-truck-monster:before{content:"\f63b"}.fa-w:before{content:"W"}.fa-earth-africa:before,.fa-globe-africa:before{content:"\f57c"}.fa-rainbow:before{content:"\f75b"}.fa-circle-notch:before{content:"\f1ce"}.fa-tablet-alt:before,.fa-tablet-screen-button:before{content:"\f3fa"}.fa-paw:before{content:"\f1b0"}.fa-cloud:before{content:"\f0c2"}.fa-trowel-bricks:before{content:"\e58a"}.fa-face-flushed:before,.fa-flushed:before{content:"\f579"}.fa-hospital-user:before{content:"\f80d"}.fa-tent-arrow-left-right:before{content:"\e57f"}.fa-gavel:before,.fa-legal:before{content:"\f0e3"}.fa-binoculars:before{content:"\f1e5"}.fa-microphone-slash:before{content:"\f131"}.fa-box-tissue:before{content:"\e05b"}.fa-motorcycle:before{content:"\f21c"}.fa-bell-concierge:before,.fa-concierge-bell:before{content:"\f562"}.fa-pen-ruler:before,.fa-pencil-ruler:before{content:"\f5ae"}.fa-people-arrows-left-right:before,.fa-people-arrows:before{content:"\e068"}.fa-mars-and-venus-burst:before{content:"\e523"}.fa-caret-square-right:before,.fa-square-caret-right:before{content:"\f152"}.fa-cut:before,.fa-scissors:before{content:"\f0c4"}.fa-sun-plant-wilt:before{content:"\e57a"}.fa-toilets-portable:before{content:"\e584"}.fa-hockey-puck:before{content:"\f453"}.fa-table:before{content:"\f0ce"}.fa-magnifying-glass-arrow-right:before{content:"\e521"}.fa-digital-tachograph:before,.fa-tachograph-digital:before{content:"\f566"}.fa-users-slash:before{content:"\e073"}.fa-clover:before{content:"\e139"}.fa-mail-reply:before,.fa-reply:before{content:"\f3e5"}.fa-star-and-crescent:before{content:"\f699"}.fa-house-fire:before{content:"\e50c"}.fa-minus-square:before,.fa-square-minus:before{content:"\f146"}.fa-helicopter:before{content:"\f533"}.fa-compass:before{content:"\f14e"}.fa-caret-square-down:before,.fa-square-caret-down:before{content:"\f150"}.fa-file-circle-question:before{content:"\e4ef"}.fa-laptop-code:before{content:"\f5fc"}.fa-swatchbook:before{content:"\f5c3"}.fa-prescription-bottle:before{content:"\f485"}.fa-bars:before,.fa-navicon:before{content:"\f0c9"}.fa-people-group:before{content:"\e533"}.fa-hourglass-3:before,.fa-hourglass-end:before{content:"\f253"}.fa-heart-broken:before,.fa-heart-crack:before{content:"\f7a9"}.fa-external-link-square-alt:before,.fa-square-up-right:before{content:"\f360"}.fa-face-kiss-beam:before,.fa-kiss-beam:before{content:"\f597"}.fa-film:before{content:"\f008"}.fa-ruler-horizontal:before{content:"\f547"}.fa-people-robbery:before{content:"\e536"}.fa-lightbulb:before{content:"\f0eb"}.fa-caret-left:before{content:"\f0d9"}.fa-circle-exclamation:before,.fa-exclamation-circle:before{content:"\f06a"}.fa-school-circle-xmark:before{content:"\e56d"}.fa-arrow-right-from-bracket:before,.fa-sign-out:before{content:"\f08b"}.fa-chevron-circle-down:before,.fa-circle-chevron-down:before{content:"\f13a"}.fa-unlock-alt:before,.fa-unlock-keyhole:before{content:"\f13e"}.fa-cloud-showers-heavy:before{content:"\f740"}.fa-headphones-alt:before,.fa-headphones-simple:before{content:"\f58f"}.fa-sitemap:before{content:"\f0e8"}.fa-circle-dollar-to-slot:before,.fa-donate:before{content:"\f4b9"}.fa-memory:before{content:"\f538"}.fa-road-spikes:before{content:"\e568"}.fa-fire-burner:before{content:"\e4f1"}.fa-flag:before{content:"\f024"}.fa-hanukiah:before{content:"\f6e6"}.fa-feather:before{content:"\f52d"}.fa-volume-down:before,.fa-volume-low:before{content:"\f027"}.fa-comment-slash:before{content:"\f4b3"}.fa-cloud-sun-rain:before{content:"\f743"}.fa-compress:before{content:"\f066"}.fa-wheat-alt:before,.fa-wheat-awn:before{content:"\e2cd"}.fa-ankh:before{content:"\f644"}.fa-hands-holding-child:before{content:"\e4fa"}.fa-asterisk:before{content:"\*"}.fa-check-square:before,.fa-square-check:before{content:"\f14a"}.fa-peseta-sign:before{content:"\e221"}.fa-header:before,.fa-heading:before{content:"\f1dc"}.fa-ghost:before{content:"\f6e2"}.fa-list-squares:before,.fa-list:before{content:"\f03a"}.fa-phone-square-alt:before,.fa-square-phone-flip:before{content:"\f87b"}.fa-cart-plus:before{content:"\f217"}.fa-gamepad:before{content:"\f11b"}.fa-circle-dot:before,.fa-dot-circle:before{content:"\f192"}.fa-dizzy:before,.fa-face-dizzy:before{content:"\f567"}.fa-egg:before{content:"\f7fb"}.fa-house-medical-circle-xmark:before{content:"\e513"}.fa-campground:before{content:"\f6bb"}.fa-folder-plus:before{content:"\f65e"}.fa-futbol-ball:before,.fa-futbol:before,.fa-soccer-ball:before{content:"\f1e3"}.fa-paint-brush:before,.fa-paintbrush:before{content:"\f1fc"}.fa-lock:before{content:"\f023"}.fa-gas-pump:before{content:"\f52f"}.fa-hot-tub-person:before,.fa-hot-tub:before{content:"\f593"}.fa-map-location:before,.fa-map-marked:before{content:"\f59f"}.fa-house-flood-water:before{content:"\e50e"}.fa-tree:before{content:"\f1bb"}.fa-bridge-lock:before{content:"\e4cc"}.fa-sack-dollar:before{content:"\f81d"}.fa-edit:before,.fa-pen-to-square:before{content:"\f044"}.fa-car-side:before{content:"\f5e4"}.fa-share-alt:before,.fa-share-nodes:before{content:"\f1e0"}.fa-heart-circle-minus:before{content:"\e4ff"}.fa-hourglass-2:before,.fa-hourglass-half:before{content:"\f252"}.fa-microscope:before{content:"\f610"}.fa-sink:before{content:"\e06d"}.fa-bag-shopping:before,.fa-shopping-bag:before{content:"\f290"}.fa-arrow-down-z-a:before,.fa-sort-alpha-desc:before,.fa-sort-alpha-down-alt:before{content:"\f881"}.fa-mitten:before{content:"\f7b5"}.fa-person-rays:before{content:"\e54d"}.fa-users:before{content:"\f0c0"}.fa-eye-slash:before{content:"\f070"}.fa-flask-vial:before{content:"\e4f3"}.fa-hand-paper:before,.fa-hand:before{content:"\f256"}.fa-om:before{content:"\f679"}.fa-worm:before{content:"\e599"}.fa-house-circle-xmark:before{content:"\e50b"}.fa-plug:before{content:"\f1e6"}.fa-chevron-up:before{content:"\f077"}.fa-hand-spock:before{content:"\f259"}.fa-stopwatch:before{content:"\f2f2"}.fa-face-kiss:before,.fa-kiss:before{content:"\f596"}.fa-bridge-circle-xmark:before{content:"\e4cb"}.fa-face-grin-tongue:before,.fa-grin-tongue:before{content:"\f589"}.fa-chess-bishop:before{content:"\f43a"}.fa-face-grin-wink:before,.fa-grin-wink:before{content:"\f58c"}.fa-deaf:before,.fa-deafness:before,.fa-ear-deaf:before,.fa-hard-of-hearing:before{content:"\f2a4"}.fa-road-circle-check:before{content:"\e564"}.fa-dice-five:before{content:"\f523"}.fa-rss-square:before,.fa-square-rss:before{content:"\f143"}.fa-land-mine-on:before{content:"\e51b"}.fa-i-cursor:before{content:"\f246"}.fa-stamp:before{content:"\f5bf"}.fa-stairs:before{content:"\e289"}.fa-i:before{content:"I"}.fa-hryvnia-sign:before,.fa-hryvnia:before{content:"\f6f2"}.fa-pills:before{content:"\f484"}.fa-face-grin-wide:before,.fa-grin-alt:before{content:"\f581"}.fa-tooth:before{content:"\f5c9"}.fa-v:before{content:"V"}.fa-bangladeshi-taka-sign:before{content:"\e2e6"}.fa-bicycle:before{content:"\f206"}.fa-rod-asclepius:before,.fa-rod-snake:before,.fa-staff-aesculapius:before,.fa-staff-snake:before{content:"\e579"}.fa-head-side-cough-slash:before{content:"\e062"}.fa-ambulance:before,.fa-truck-medical:before{content:"\f0f9"}.fa-wheat-awn-circle-exclamation:before{content:"\e598"}.fa-snowman:before{content:"\f7d0"}.fa-mortar-pestle:before{content:"\f5a7"}.fa-road-barrier:before{content:"\e562"}.fa-school:before{content:"\f549"}.fa-igloo:before{content:"\f7ae"}.fa-joint:before{content:"\f595"}.fa-angle-right:before{content:"\f105"}.fa-horse:before{content:"\f6f0"}.fa-q:before{content:"Q"}.fa-g:before{content:"G"}.fa-notes-medical:before{content:"\f481"}.fa-temperature-2:before,.fa-temperature-half:before,.fa-thermometer-2:before,.fa-thermometer-half:before{content:"\f2c9"}.fa-dong-sign:before{content:"\e169"}.fa-capsules:before{content:"\f46b"}.fa-poo-bolt:before,.fa-poo-storm:before{content:"\f75a"}.fa-face-frown-open:before,.fa-frown-open:before{content:"\f57a"}.fa-hand-point-up:before{content:"\f0a6"}.fa-money-bill:before{content:"\f0d6"}.fa-bookmark:before{content:"\f02e"}.fa-align-justify:before{content:"\f039"}.fa-umbrella-beach:before{content:"\f5ca"}.fa-helmet-un:before{content:"\e503"}.fa-bullseye:before{content:"\f140"}.fa-bacon:before{content:"\f7e5"}.fa-hand-point-down:before{content:"\f0a7"}.fa-arrow-up-from-bracket:before{content:"\e09a"}.fa-folder-blank:before,.fa-folder:before{content:"\f07b"}.fa-file-medical-alt:before,.fa-file-waveform:before{content:"\f478"}.fa-radiation:before{content:"\f7b9"}.fa-chart-simple:before{content:"\e473"}.fa-mars-stroke:before{content:"\f229"}.fa-vial:before{content:"\f492"}.fa-dashboard:before,.fa-gauge-med:before,.fa-gauge:before,.fa-tachometer-alt-average:before{content:"\f624"}.fa-magic-wand-sparkles:before,.fa-wand-magic-sparkles:before{content:"\e2ca"}.fa-e:before{content:"E"}.fa-pen-alt:before,.fa-pen-clip:before{content:"\f305"}.fa-bridge-circle-exclamation:before{content:"\e4ca"}.fa-user:before{content:"\f007"}.fa-school-circle-check:before{content:"\e56b"}.fa-dumpster:before{content:"\f793"}.fa-shuttle-van:before,.fa-van-shuttle:before{content:"\f5b6"}.fa-building-user:before{content:"\e4da"}.fa-caret-square-left:before,.fa-square-caret-left:before{content:"\f191"}.fa-highlighter:before{content:"\f591"}.fa-key:before{content:"\f084"}.fa-bullhorn:before{content:"\f0a1"}.fa-globe:before{content:"\f0ac"}.fa-synagogue:before{content:"\f69b"}.fa-person-half-dress:before{content:"\e548"}.fa-road-bridge:before{content:"\e563"}.fa-location-arrow:before{content:"\f124"}.fa-c:before{content:"C"}.fa-tablet-button:before{content:"\f10a"}.fa-building-lock:before{content:"\e4d6"}.fa-pizza-slice:before{content:"\f818"}.fa-money-bill-wave:before{content:"\f53a"}.fa-area-chart:before,.fa-chart-area:before{content:"\f1fe"}.fa-house-flag:before{content:"\e50d"}.fa-person-circle-minus:before{content:"\e540"}.fa-ban:before,.fa-cancel:before{content:"\f05e"}.fa-camera-rotate:before{content:"\e0d8"}.fa-air-freshener:before,.fa-spray-can-sparkles:before{content:"\f5d0"}.fa-star:before{content:"\f005"}.fa-repeat:before{content:"\f363"}.fa-cross:before{content:"\f654"}.fa-box:before{content:"\f466"}.fa-venus-mars:before{content:"\f228"}.fa-arrow-pointer:before,.fa-mouse-pointer:before{content:"\f245"}.fa-expand-arrows-alt:before,.fa-maximize:before{content:"\f31e"}.fa-charging-station:before{content:"\f5e7"}.fa-shapes:before,.fa-triangle-circle-square:before{content:"\f61f"}.fa-random:before,.fa-shuffle:before{content:"\f074"}.fa-person-running:before,.fa-running:before{content:"\f70c"}.fa-mobile-retro:before{content:"\e527"}.fa-grip-lines-vertical:before{content:"\f7a5"}.fa-spider:before{content:"\f717"}.fa-hands-bound:before{content:"\e4f9"}.fa-file-invoice-dollar:before{content:"\f571"}.fa-plane-circle-exclamation:before{content:"\e556"}.fa-x-ray:before{content:"\f497"}.fa-spell-check:before{content:"\f891"}.fa-slash:before{content:"\f715"}.fa-computer-mouse:before,.fa-mouse:before{content:"\f8cc"}.fa-arrow-right-to-bracket:before,.fa-sign-in:before{content:"\f090"}.fa-shop-slash:before,.fa-store-alt-slash:before{content:"\e070"}.fa-server:before{content:"\f233"}.fa-virus-covid-slash:before{content:"\e4a9"}.fa-shop-lock:before{content:"\e4a5"}.fa-hourglass-1:before,.fa-hourglass-start:before{content:"\f251"}.fa-blender-phone:before{content:"\f6b6"}.fa-building-wheat:before{content:"\e4db"}.fa-person-breastfeeding:before{content:"\e53a"}.fa-right-to-bracket:before,.fa-sign-in-alt:before{content:"\f2f6"}.fa-venus:before{content:"\f221"}.fa-passport:before{content:"\f5ab"}.fa-heart-pulse:before,.fa-heartbeat:before{content:"\f21e"}.fa-people-carry-box:before,.fa-people-carry:before{content:"\f4ce"}.fa-temperature-high:before{content:"\f769"}.fa-microchip:before{content:"\f2db"}.fa-crown:before{content:"\f521"}.fa-weight-hanging:before{content:"\f5cd"}.fa-xmarks-lines:before{content:"\e59a"}.fa-file-prescription:before{content:"\f572"}.fa-weight-scale:before,.fa-weight:before{content:"\f496"}.fa-user-friends:before,.fa-user-group:before{content:"\f500"}.fa-arrow-up-a-z:before,.fa-sort-alpha-up:before{content:"\f15e"}.fa-chess-knight:before{content:"\f441"}.fa-face-laugh-squint:before,.fa-laugh-squint:before{content:"\f59b"}.fa-wheelchair:before{content:"\f193"}.fa-arrow-circle-up:before,.fa-circle-arrow-up:before{content:"\f0aa"}.fa-toggle-on:before{content:"\f205"}.fa-person-walking:before,.fa-walking:before{content:"\f554"}.fa-l:before{content:"L"}.fa-fire:before{content:"\f06d"}.fa-bed-pulse:before,.fa-procedures:before{content:"\f487"}.fa-shuttle-space:before,.fa-space-shuttle:before{content:"\f197"}.fa-face-laugh:before,.fa-laugh:before{content:"\f599"}.fa-folder-open:before{content:"\f07c"}.fa-heart-circle-plus:before{content:"\e500"}.fa-code-fork:before{content:"\e13b"}.fa-city:before{content:"\f64f"}.fa-microphone-alt:before,.fa-microphone-lines:before{content:"\f3c9"}.fa-pepper-hot:before{content:"\f816"}.fa-unlock:before{content:"\f09c"}.fa-colon-sign:before{content:"\e140"}.fa-headset:before{content:"\f590"}.fa-store-slash:before{content:"\e071"}.fa-road-circle-xmark:before{content:"\e566"}.fa-user-minus:before{content:"\f503"}.fa-mars-stroke-up:before,.fa-mars-stroke-v:before{content:"\f22a"}.fa-champagne-glasses:before,.fa-glass-cheers:before{content:"\f79f"}.fa-clipboard:before{content:"\f328"}.fa-house-circle-exclamation:before{content:"\e50a"}.fa-file-arrow-up:before,.fa-file-upload:before{content:"\f574"}.fa-wifi-3:before,.fa-wifi-strong:before,.fa-wifi:before{content:"\f1eb"}.fa-bath:before,.fa-bathtub:before{content:"\f2cd"}.fa-underline:before{content:"\f0cd"}.fa-user-edit:before,.fa-user-pen:before{content:"\f4ff"}.fa-signature:before{content:"\f5b7"}.fa-stroopwafel:before{content:"\f551"}.fa-bold:before{content:"\f032"}.fa-anchor-lock:before{content:"\e4ad"}.fa-building-ngo:before{content:"\e4d7"}.fa-manat-sign:before{content:"\e1d5"}.fa-not-equal:before{content:"\f53e"}.fa-border-style:before,.fa-border-top-left:before{content:"\f853"}.fa-map-location-dot:before,.fa-map-marked-alt:before{content:"\f5a0"}.fa-jedi:before{content:"\f669"}.fa-poll:before,.fa-square-poll-vertical:before{content:"\f681"}.fa-mug-hot:before{content:"\f7b6"}.fa-battery-car:before,.fa-car-battery:before{content:"\f5df"}.fa-gift:before{content:"\f06b"}.fa-dice-two:before{content:"\f528"}.fa-chess-queen:before{content:"\f445"}.fa-glasses:before{content:"\f530"}.fa-chess-board:before{content:"\f43c"}.fa-building-circle-check:before{content:"\e4d2"}.fa-person-chalkboard:before{content:"\e53d"}.fa-mars-stroke-h:before,.fa-mars-stroke-right:before{content:"\f22b"}.fa-hand-back-fist:before,.fa-hand-rock:before{content:"\f255"}.fa-caret-square-up:before,.fa-square-caret-up:before{content:"\f151"}.fa-cloud-showers-water:before{content:"\e4e4"}.fa-bar-chart:before,.fa-chart-bar:before{content:"\f080"}.fa-hands-bubbles:before,.fa-hands-wash:before{content:"\e05e"}.fa-less-than-equal:before{content:"\f537"}.fa-train:before{content:"\f238"}.fa-eye-low-vision:before,.fa-low-vision:before{content:"\f2a8"}.fa-crow:before{content:"\f520"}.fa-sailboat:before{content:"\e445"}.fa-window-restore:before{content:"\f2d2"}.fa-plus-square:before,.fa-square-plus:before{content:"\f0fe"}.fa-torii-gate:before{content:"\f6a1"}.fa-frog:before{content:"\f52e"}.fa-bucket:before{content:"\e4cf"}.fa-image:before{content:"\f03e"}.fa-microphone:before{content:"\f130"}.fa-cow:before{content:"\f6c8"}.fa-caret-up:before{content:"\f0d8"}.fa-screwdriver:before{content:"\f54a"}.fa-folder-closed:before{content:"\e185"}.fa-house-tsunami:before{content:"\e515"}.fa-square-nfi:before{content:"\e576"}.fa-arrow-up-from-ground-water:before{content:"\e4b5"}.fa-glass-martini-alt:before,.fa-martini-glass:before{content:"\f57b"}.fa-rotate-back:before,.fa-rotate-backward:before,.fa-rotate-left:before,.fa-undo-alt:before{content:"\f2ea"}.fa-columns:before,.fa-table-columns:before{content:"\f0db"}.fa-lemon:before{content:"\f094"}.fa-head-side-mask:before{content:"\e063"}.fa-handshake:before{content:"\f2b5"}.fa-gem:before{content:"\f3a5"}.fa-dolly-box:before,.fa-dolly:before{content:"\f472"}.fa-smoking:before{content:"\f48d"}.fa-compress-arrows-alt:before,.fa-minimize:before{content:"\f78c"}.fa-monument:before{content:"\f5a6"}.fa-snowplow:before{content:"\f7d2"}.fa-angle-double-right:before,.fa-angles-right:before{content:"\f101"}.fa-cannabis:before{content:"\f55f"}.fa-circle-play:before,.fa-play-circle:before{content:"\f144"}.fa-tablets:before{content:"\f490"}.fa-ethernet:before{content:"\f796"}.fa-eur:before,.fa-euro-sign:before,.fa-euro:before{content:"\f153"}.fa-chair:before{content:"\f6c0"}.fa-check-circle:before,.fa-circle-check:before{content:"\f058"}.fa-circle-stop:before,.fa-stop-circle:before{content:"\f28d"}.fa-compass-drafting:before,.fa-drafting-compass:before{content:"\f568"}.fa-plate-wheat:before{content:"\e55a"}.fa-icicles:before{content:"\f7ad"}.fa-person-shelter:before{content:"\e54f"}.fa-neuter:before{content:"\f22c"}.fa-id-badge:before{content:"\f2c1"}.fa-marker:before{content:"\f5a1"}.fa-face-laugh-beam:before,.fa-laugh-beam:before{content:"\f59a"}.fa-helicopter-symbol:before{content:"\e502"}.fa-universal-access:before{content:"\f29a"}.fa-chevron-circle-up:before,.fa-circle-chevron-up:before{content:"\f139"}.fa-lari-sign:before{content:"\e1c8"}.fa-volcano:before{content:"\f770"}.fa-person-walking-dashed-line-arrow-right:before{content:"\e553"}.fa-gbp:before,.fa-pound-sign:before,.fa-sterling-sign:before{content:"\f154"}.fa-viruses:before{content:"\e076"}.fa-square-person-confined:before{content:"\e577"}.fa-user-tie:before{content:"\f508"}.fa-arrow-down-long:before,.fa-long-arrow-down:before{content:"\f175"}.fa-tent-arrow-down-to-line:before{content:"\e57e"}.fa-certificate:before{content:"\f0a3"}.fa-mail-reply-all:before,.fa-reply-all:before{content:"\f122"}.fa-suitcase:before{content:"\f0f2"}.fa-person-skating:before,.fa-skating:before{content:"\f7c5"}.fa-filter-circle-dollar:before,.fa-funnel-dollar:before{content:"\f662"}.fa-camera-retro:before{content:"\f083"}.fa-arrow-circle-down:before,.fa-circle-arrow-down:before{content:"\f0ab"}.fa-arrow-right-to-file:before,.fa-file-import:before{content:"\f56f"}.fa-external-link-square:before,.fa-square-arrow-up-right:before{content:"\f14c"}.fa-box-open:before{content:"\f49e"}.fa-scroll:before{content:"\f70e"}.fa-spa:before{content:"\f5bb"}.fa-location-pin-lock:before{content:"\e51f"}.fa-pause:before{content:"\f04c"}.fa-hill-avalanche:before{content:"\e507"}.fa-temperature-0:before,.fa-temperature-empty:before,.fa-thermometer-0:before,.fa-thermometer-empty:before{content:"\f2cb"}.fa-bomb:before{content:"\f1e2"}.fa-registered:before{content:"\f25d"}.fa-address-card:before,.fa-contact-card:before,.fa-vcard:before{content:"\f2bb"}.fa-balance-scale-right:before,.fa-scale-unbalanced-flip:before{content:"\f516"}.fa-subscript:before{content:"\f12c"}.fa-diamond-turn-right:before,.fa-directions:before{content:"\f5eb"}.fa-burst:before{content:"\e4dc"}.fa-house-laptop:before,.fa-laptop-house:before{content:"\e066"}.fa-face-tired:before,.fa-tired:before{content:"\f5c8"}.fa-money-bills:before{content:"\e1f3"}.fa-smog:before{content:"\f75f"}.fa-crutch:before{content:"\f7f7"}.fa-cloud-arrow-up:before,.fa-cloud-upload-alt:before,.fa-cloud-upload:before{content:"\f0ee"}.fa-palette:before{content:"\f53f"}.fa-arrows-turn-right:before{content:"\e4c0"}.fa-vest:before{content:"\e085"}.fa-ferry:before{content:"\e4ea"}.fa-arrows-down-to-people:before{content:"\e4b9"}.fa-seedling:before,.fa-sprout:before{content:"\f4d8"}.fa-arrows-alt-h:before,.fa-left-right:before{content:"\f337"}.fa-boxes-packing:before{content:"\e4c7"}.fa-arrow-circle-left:before,.fa-circle-arrow-left:before{content:"\f0a8"}.fa-group-arrows-rotate:before{content:"\e4f6"}.fa-bowl-food:before{content:"\e4c6"}.fa-candy-cane:before{content:"\f786"}.fa-arrow-down-wide-short:before,.fa-sort-amount-asc:before,.fa-sort-amount-down:before{content:"\f160"}.fa-cloud-bolt:before,.fa-thunderstorm:before{content:"\f76c"}.fa-remove-format:before,.fa-text-slash:before{content:"\f87d"}.fa-face-smile-wink:before,.fa-smile-wink:before{content:"\f4da"}.fa-file-word:before{content:"\f1c2"}.fa-file-powerpoint:before{content:"\f1c4"}.fa-arrows-h:before,.fa-arrows-left-right:before{content:"\f07e"}.fa-house-lock:before{content:"\e510"}.fa-cloud-arrow-down:before,.fa-cloud-download-alt:before,.fa-cloud-download:before{content:"\f0ed"}.fa-children:before{content:"\e4e1"}.fa-blackboard:before,.fa-chalkboard:before{content:"\f51b"}.fa-user-alt-slash:before,.fa-user-large-slash:before{content:"\f4fa"}.fa-envelope-open:before{content:"\f2b6"}.fa-handshake-alt-slash:before,.fa-handshake-simple-slash:before{content:"\e05f"}.fa-mattress-pillow:before{content:"\e525"}.fa-guarani-sign:before{content:"\e19a"}.fa-arrows-rotate:before,.fa-refresh:before,.fa-sync:before{content:"\f021"}.fa-fire-extinguisher:before{content:"\f134"}.fa-cruzeiro-sign:before{content:"\e152"}.fa-greater-than-equal:before{content:"\f532"}.fa-shield-alt:before,.fa-shield-halved:before{content:"\f3ed"}.fa-atlas:before,.fa-book-atlas:before{content:"\f558"}.fa-virus:before{content:"\e074"}.fa-envelope-circle-check:before{content:"\e4e8"}.fa-layer-group:before{content:"\f5fd"}.fa-arrows-to-dot:before{content:"\e4be"}.fa-archway:before{content:"\f557"}.fa-heart-circle-check:before{content:"\e4fd"}.fa-house-chimney-crack:before,.fa-house-damage:before{content:"\f6f1"}.fa-file-archive:before,.fa-file-zipper:before{content:"\f1c6"}.fa-square:before{content:"\f0c8"}.fa-glass-martini:before,.fa-martini-glass-empty:before{content:"\f000"}.fa-couch:before{content:"\f4b8"}.fa-cedi-sign:before{content:"\e0df"}.fa-italic:before{content:"\f033"}.fa-table-cells-column-lock:before{content:"\e678"}.fa-church:before{content:"\f51d"}.fa-comments-dollar:before{content:"\f653"}.fa-democrat:before{content:"\f747"}.fa-z:before{content:"Z"}.fa-person-skiing:before,.fa-skiing:before{content:"\f7c9"}.fa-road-lock:before{content:"\e567"}.fa-a:before{content:"A"}.fa-temperature-arrow-down:before,.fa-temperature-down:before{content:"\e03f"}.fa-feather-alt:before,.fa-feather-pointed:before{content:"\f56b"}.fa-p:before{content:"P"}.fa-snowflake:before{content:"\f2dc"}.fa-newspaper:before{content:"\f1ea"}.fa-ad:before,.fa-rectangle-ad:before{content:"\f641"}.fa-arrow-circle-right:before,.fa-circle-arrow-right:before{content:"\f0a9"}.fa-filter-circle-xmark:before{content:"\e17b"}.fa-locust:before{content:"\e520"}.fa-sort:before,.fa-unsorted:before{content:"\f0dc"}.fa-list-1-2:before,.fa-list-numeric:before,.fa-list-ol:before{content:"\f0cb"}.fa-person-dress-burst:before{content:"\e544"}.fa-money-check-alt:before,.fa-money-check-dollar:before{content:"\f53d"}.fa-vector-square:before{content:"\f5cb"}.fa-bread-slice:before{content:"\f7ec"}.fa-language:before{content:"\f1ab"}.fa-face-kiss-wink-heart:before,.fa-kiss-wink-heart:before{content:"\f598"}.fa-filter:before{content:"\f0b0"}.fa-question:before{content:"\?"}.fa-file-signature:before{content:"\f573"}.fa-arrows-alt:before,.fa-up-down-left-right:before{content:"\f0b2"}.fa-house-chimney-user:before{content:"\e065"}.fa-hand-holding-heart:before{content:"\f4be"}.fa-puzzle-piece:before{content:"\f12e"}.fa-money-check:before{content:"\f53c"}.fa-star-half-alt:before,.fa-star-half-stroke:before{content:"\f5c0"}.fa-code:before{content:"\f121"}.fa-glass-whiskey:before,.fa-whiskey-glass:before{content:"\f7a0"}.fa-building-circle-exclamation:before{content:"\e4d3"}.fa-magnifying-glass-chart:before{content:"\e522"}.fa-arrow-up-right-from-square:before,.fa-external-link:before{content:"\f08e"}.fa-cubes-stacked:before{content:"\e4e6"}.fa-krw:before,.fa-won-sign:before,.fa-won:before{content:"\f159"}.fa-virus-covid:before{content:"\e4a8"}.fa-austral-sign:before{content:"\e0a9"}.fa-f:before{content:"F"}.fa-leaf:before{content:"\f06c"}.fa-road:before{content:"\f018"}.fa-cab:before,.fa-taxi:before{content:"\f1ba"}.fa-person-circle-plus:before{content:"\e541"}.fa-chart-pie:before,.fa-pie-chart:before{content:"\f200"}.fa-bolt-lightning:before{content:"\e0b7"}.fa-sack-xmark:before{content:"\e56a"}.fa-file-excel:before{content:"\f1c3"}.fa-file-contract:before{content:"\f56c"}.fa-fish-fins:before{content:"\e4f2"}.fa-building-flag:before{content:"\e4d5"}.fa-face-grin-beam:before,.fa-grin-beam:before{content:"\f582"}.fa-object-ungroup:before{content:"\f248"}.fa-poop:before{content:"\f619"}.fa-location-pin:before,.fa-map-marker:before{content:"\f041"}.fa-kaaba:before{content:"\f66b"}.fa-toilet-paper:before{content:"\f71e"}.fa-hard-hat:before,.fa-hat-hard:before,.fa-helmet-safety:before{content:"\f807"}.fa-eject:before{content:"\f052"}.fa-arrow-alt-circle-right:before,.fa-circle-right:before{content:"\f35a"}.fa-plane-circle-check:before{content:"\e555"}.fa-face-rolling-eyes:before,.fa-meh-rolling-eyes:before{content:"\f5a5"}.fa-object-group:before{content:"\f247"}.fa-chart-line:before,.fa-line-chart:before{content:"\f201"}.fa-mask-ventilator:before{content:"\e524"}.fa-arrow-right:before{content:"\f061"}.fa-map-signs:before,.fa-signs-post:before{content:"\f277"}.fa-cash-register:before{content:"\f788"}.fa-person-circle-question:before{content:"\e542"}.fa-h:before{content:"H"}.fa-tarp:before{content:"\e57b"}.fa-screwdriver-wrench:before,.fa-tools:before{content:"\f7d9"}.fa-arrows-to-eye:before{content:"\e4bf"}.fa-plug-circle-bolt:before{content:"\e55b"}.fa-heart:before{content:"\f004"}.fa-mars-and-venus:before{content:"\f224"}.fa-home-user:before,.fa-house-user:before{content:"\e1b0"}.fa-dumpster-fire:before{content:"\f794"}.fa-house-crack:before{content:"\e3b1"}.fa-cocktail:before,.fa-martini-glass-citrus:before{content:"\f561"}.fa-face-surprise:before,.fa-surprise:before{content:"\f5c2"}.fa-bottle-water:before{content:"\e4c5"}.fa-circle-pause:before,.fa-pause-circle:before{content:"\f28b"}.fa-toilet-paper-slash:before{content:"\e072"}.fa-apple-alt:before,.fa-apple-whole:before{content:"\f5d1"}.fa-kitchen-set:before{content:"\e51a"}.fa-r:before{content:"R"}.fa-temperature-1:before,.fa-temperature-quarter:before,.fa-thermometer-1:before,.fa-thermometer-quarter:before{content:"\f2ca"}.fa-cube:before{content:"\f1b2"}.fa-bitcoin-sign:before{content:"\e0b4"}.fa-shield-dog:before{content:"\e573"}.fa-solar-panel:before{content:"\f5ba"}.fa-lock-open:before{content:"\f3c1"}.fa-elevator:before{content:"\e16d"}.fa-money-bill-transfer:before{content:"\e528"}.fa-money-bill-trend-up:before{content:"\e529"}.fa-house-flood-water-circle-arrow-right:before{content:"\e50f"}.fa-poll-h:before,.fa-square-poll-horizontal:before{content:"\f682"}.fa-circle:before{content:"\f111"}.fa-backward-fast:before,.fa-fast-backward:before{content:"\f049"}.fa-recycle:before{content:"\f1b8"}.fa-user-astronaut:before{content:"\f4fb"}.fa-plane-slash:before{content:"\e069"}.fa-trademark:before{content:"\f25c"}.fa-basketball-ball:before,.fa-basketball:before{content:"\f434"}.fa-satellite-dish:before{content:"\f7c0"}.fa-arrow-alt-circle-up:before,.fa-circle-up:before{content:"\f35b"}.fa-mobile-alt:before,.fa-mobile-screen-button:before{content:"\f3cd"}.fa-volume-high:before,.fa-volume-up:before{content:"\f028"}.fa-users-rays:before{content:"\e593"}.fa-wallet:before{content:"\f555"}.fa-clipboard-check:before{content:"\f46c"}.fa-file-audio:before{content:"\f1c7"}.fa-burger:before,.fa-hamburger:before{content:"\f805"}.fa-wrench:before{content:"\f0ad"}.fa-bugs:before{content:"\e4d0"}.fa-rupee-sign:before,.fa-rupee:before{content:"\f156"}.fa-file-image:before{content:"\f1c5"}.fa-circle-question:before,.fa-question-circle:before{content:"\f059"}.fa-plane-departure:before{content:"\f5b0"}.fa-handshake-slash:before{content:"\e060"}.fa-book-bookmark:before{content:"\e0bb"}.fa-code-branch:before{content:"\f126"}.fa-hat-cowboy:before{content:"\f8c0"}.fa-bridge:before{content:"\e4c8"}.fa-phone-alt:before,.fa-phone-flip:before{content:"\f879"}.fa-truck-front:before{content:"\e2b7"}.fa-cat:before{content:"\f6be"}.fa-anchor-circle-exclamation:before{content:"\e4ab"}.fa-truck-field:before{content:"\e58d"}.fa-route:before{content:"\f4d7"}.fa-clipboard-question:before{content:"\e4e3"}.fa-panorama:before{content:"\e209"}.fa-comment-medical:before{content:"\f7f5"}.fa-teeth-open:before{content:"\f62f"}.fa-file-circle-minus:before{content:"\e4ed"}.fa-tags:before{content:"\f02c"}.fa-wine-glass:before{content:"\f4e3"}.fa-fast-forward:before,.fa-forward-fast:before{content:"\f050"}.fa-face-meh-blank:before,.fa-meh-blank:before{content:"\f5a4"}.fa-parking:before,.fa-square-parking:before{content:"\f540"}.fa-house-signal:before{content:"\e012"}.fa-bars-progress:before,.fa-tasks-alt:before{content:"\f828"}.fa-faucet-drip:before{content:"\e006"}.fa-cart-flatbed:before,.fa-dolly-flatbed:before{content:"\f474"}.fa-ban-smoking:before,.fa-smoking-ban:before{content:"\f54d"}.fa-terminal:before{content:"\f120"}.fa-mobile-button:before{content:"\f10b"}.fa-house-medical-flag:before{content:"\e514"}.fa-basket-shopping:before,.fa-shopping-basket:before{content:"\f291"}.fa-tape:before{content:"\f4db"}.fa-bus-alt:before,.fa-bus-simple:before{content:"\f55e"}.fa-eye:before{content:"\f06e"}.fa-face-sad-cry:before,.fa-sad-cry:before{content:"\f5b3"}.fa-audio-description:before{content:"\f29e"}.fa-person-military-to-person:before{content:"\e54c"}.fa-file-shield:before{content:"\e4f0"}.fa-user-slash:before{content:"\f506"}.fa-pen:before{content:"\f304"}.fa-tower-observation:before{content:"\e586"}.fa-file-code:before{content:"\f1c9"}.fa-signal-5:before,.fa-signal-perfect:before,.fa-signal:before{content:"\f012"}.fa-bus:before{content:"\f207"}.fa-heart-circle-xmark:before{content:"\e501"}.fa-home-lg:before,.fa-house-chimney:before{content:"\e3af"}.fa-window-maximize:before{content:"\f2d0"}.fa-face-frown:before,.fa-frown:before{content:"\f119"}.fa-prescription:before{content:"\f5b1"}.fa-shop:before,.fa-store-alt:before{content:"\f54f"}.fa-floppy-disk:before,.fa-save:before{content:"\f0c7"}.fa-vihara:before{content:"\f6a7"}.fa-balance-scale-left:before,.fa-scale-unbalanced:before{content:"\f515"}.fa-sort-asc:before,.fa-sort-up:before{content:"\f0de"}.fa-comment-dots:before,.fa-commenting:before{content:"\f4ad"}.fa-plant-wilt:before{content:"\e5aa"}.fa-diamond:before{content:"\f219"}.fa-face-grin-squint:before,.fa-grin-squint:before{content:"\f585"}.fa-hand-holding-dollar:before,.fa-hand-holding-usd:before{content:"\f4c0"}.fa-bacterium:before{content:"\e05a"}.fa-hand-pointer:before{content:"\f25a"}.fa-drum-steelpan:before{content:"\f56a"}.fa-hand-scissors:before{content:"\f257"}.fa-hands-praying:before,.fa-praying-hands:before{content:"\f684"}.fa-arrow-right-rotate:before,.fa-arrow-rotate-forward:before,.fa-arrow-rotate-right:before,.fa-redo:before{content:"\f01e"}.fa-biohazard:before{content:"\f780"}.fa-location-crosshairs:before,.fa-location:before{content:"\f601"}.fa-mars-double:before{content:"\f227"}.fa-child-dress:before{content:"\e59c"}.fa-users-between-lines:before{content:"\e591"}.fa-lungs-virus:before{content:"\e067"}.fa-face-grin-tears:before,.fa-grin-tears:before{content:"\f588"}.fa-phone:before{content:"\f095"}.fa-calendar-times:before,.fa-calendar-xmark:before{content:"\f273"}.fa-child-reaching:before{content:"\e59d"}.fa-head-side-virus:before{content:"\e064"}.fa-user-cog:before,.fa-user-gear:before{content:"\f4fe"}.fa-arrow-up-1-9:before,.fa-sort-numeric-up:before{content:"\f163"}.fa-door-closed:before{content:"\f52a"}.fa-shield-virus:before{content:"\e06c"}.fa-dice-six:before{content:"\f526"}.fa-mosquito-net:before{content:"\e52c"}.fa-bridge-water:before{content:"\e4ce"}.fa-person-booth:before{content:"\f756"}.fa-text-width:before{content:"\f035"}.fa-hat-wizard:before{content:"\f6e8"}.fa-pen-fancy:before{content:"\f5ac"}.fa-digging:before,.fa-person-digging:before{content:"\f85e"}.fa-trash:before{content:"\f1f8"}.fa-gauge-simple-med:before,.fa-gauge-simple:before,.fa-tachometer-average:before{content:"\f629"}.fa-book-medical:before{content:"\f7e6"}.fa-poo:before{content:"\f2fe"}.fa-quote-right-alt:before,.fa-quote-right:before{content:"\f10e"}.fa-shirt:before,.fa-t-shirt:before,.fa-tshirt:before{content:"\f553"}.fa-cubes:before{content:"\f1b3"}.fa-divide:before{content:"\f529"}.fa-tenge-sign:before,.fa-tenge:before{content:"\f7d7"}.fa-headphones:before{content:"\f025"}.fa-hands-holding:before{content:"\f4c2"}.fa-hands-clapping:before{content:"\e1a8"}.fa-republican:before{content:"\f75e"}.fa-arrow-left:before{content:"\f060"}.fa-person-circle-xmark:before{content:"\e543"}.fa-ruler:before{content:"\f545"}.fa-align-left:before{content:"\f036"}.fa-dice-d6:before{content:"\f6d1"}.fa-restroom:before{content:"\f7bd"}.fa-j:before{content:"J"}.fa-users-viewfinder:before{content:"\e595"}.fa-file-video:before{content:"\f1c8"}.fa-external-link-alt:before,.fa-up-right-from-square:before{content:"\f35d"}.fa-table-cells:before,.fa-th:before{content:"\f00a"}.fa-file-pdf:before{content:"\f1c1"}.fa-bible:before,.fa-book-bible:before{content:"\f647"}.fa-o:before{content:"O"}.fa-medkit:before,.fa-suitcase-medical:before{content:"\f0fa"}.fa-user-secret:before{content:"\f21b"}.fa-otter:before{content:"\f700"}.fa-female:before,.fa-person-dress:before{content:"\f182"}.fa-comment-dollar:before{content:"\f651"}.fa-briefcase-clock:before,.fa-business-time:before{content:"\f64a"}.fa-table-cells-large:before,.fa-th-large:before{content:"\f009"}.fa-book-tanakh:before,.fa-tanakh:before{content:"\f827"}.fa-phone-volume:before,.fa-volume-control-phone:before{content:"\f2a0"}.fa-hat-cowboy-side:before{content:"\f8c1"}.fa-clipboard-user:before{content:"\f7f3"}.fa-child:before{content:"\f1ae"}.fa-lira-sign:before{content:"\f195"}.fa-satellite:before{content:"\f7bf"}.fa-plane-lock:before{content:"\e558"}.fa-tag:before{content:"\f02b"}.fa-comment:before{content:"\f075"}.fa-birthday-cake:before,.fa-cake-candles:before,.fa-cake:before{content:"\f1fd"}.fa-envelope:before{content:"\f0e0"}.fa-angle-double-up:before,.fa-angles-up:before{content:"\f102"}.fa-paperclip:before{content:"\f0c6"}.fa-arrow-right-to-city:before{content:"\e4b3"}.fa-ribbon:before{content:"\f4d6"}.fa-lungs:before{content:"\f604"}.fa-arrow-up-9-1:before,.fa-sort-numeric-up-alt:before{content:"\f887"}.fa-litecoin-sign:before{content:"\e1d3"}.fa-border-none:before{content:"\f850"}.fa-circle-nodes:before{content:"\e4e2"}.fa-parachute-box:before{content:"\f4cd"}.fa-indent:before{content:"\f03c"}.fa-truck-field-un:before{content:"\e58e"}.fa-hourglass-empty:before,.fa-hourglass:before{content:"\f254"}.fa-mountain:before{content:"\f6fc"}.fa-user-doctor:before,.fa-user-md:before{content:"\f0f0"}.fa-circle-info:before,.fa-info-circle:before{content:"\f05a"}.fa-cloud-meatball:before{content:"\f73b"}.fa-camera-alt:before,.fa-camera:before{content:"\f030"}.fa-square-virus:before{content:"\e578"}.fa-meteor:before{content:"\f753"}.fa-car-on:before{content:"\e4dd"}.fa-sleigh:before{content:"\f7cc"}.fa-arrow-down-1-9:before,.fa-sort-numeric-asc:before,.fa-sort-numeric-down:before{content:"\f162"}.fa-hand-holding-droplet:before,.fa-hand-holding-water:before{content:"\f4c1"}.fa-water:before{content:"\f773"}.fa-calendar-check:before{content:"\f274"}.fa-braille:before{content:"\f2a1"}.fa-prescription-bottle-alt:before,.fa-prescription-bottle-medical:before{content:"\f486"}.fa-landmark:before{content:"\f66f"}.fa-truck:before{content:"\f0d1"}.fa-crosshairs:before{content:"\f05b"}.fa-person-cane:before{content:"\e53c"}.fa-tent:before{content:"\e57d"}.fa-vest-patches:before{content:"\e086"}.fa-check-double:before{content:"\f560"}.fa-arrow-down-a-z:before,.fa-sort-alpha-asc:before,.fa-sort-alpha-down:before{content:"\f15d"}.fa-money-bill-wheat:before{content:"\e52a"}.fa-cookie:before{content:"\f563"}.fa-arrow-left-rotate:before,.fa-arrow-rotate-back:before,.fa-arrow-rotate-backward:before,.fa-arrow-rotate-left:before,.fa-undo:before{content:"\f0e2"}.fa-hard-drive:before,.fa-hdd:before{content:"\f0a0"}.fa-face-grin-squint-tears:before,.fa-grin-squint-tears:before{content:"\f586"}.fa-dumbbell:before{content:"\f44b"}.fa-list-alt:before,.fa-rectangle-list:before{content:"\f022"}.fa-tarp-droplet:before{content:"\e57c"}.fa-house-medical-circle-check:before{content:"\e511"}.fa-person-skiing-nordic:before,.fa-skiing-nordic:before{content:"\f7ca"}.fa-calendar-plus:before{content:"\f271"}.fa-plane-arrival:before{content:"\f5af"}.fa-arrow-alt-circle-left:before,.fa-circle-left:before{content:"\f359"}.fa-subway:before,.fa-train-subway:before{content:"\f239"}.fa-chart-gantt:before{content:"\e0e4"}.fa-indian-rupee-sign:before,.fa-indian-rupee:before,.fa-inr:before{content:"\e1bc"}.fa-crop-alt:before,.fa-crop-simple:before{content:"\f565"}.fa-money-bill-1:before,.fa-money-bill-alt:before{content:"\f3d1"}.fa-left-long:before,.fa-long-arrow-alt-left:before{content:"\f30a"}.fa-dna:before{content:"\f471"}.fa-virus-slash:before{content:"\e075"}.fa-minus:before,.fa-subtract:before{content:"\f068"}.fa-chess:before{content:"\f439"}.fa-arrow-left-long:before,.fa-long-arrow-left:before{content:"\f177"}.fa-plug-circle-check:before{content:"\e55c"}.fa-street-view:before{content:"\f21d"}.fa-franc-sign:before{content:"\e18f"}.fa-volume-off:before{content:"\f026"}.fa-american-sign-language-interpreting:before,.fa-asl-interpreting:before,.fa-hands-american-sign-language-interpreting:before,.fa-hands-asl-interpreting:before{content:"\f2a3"}.fa-cog:before,.fa-gear:before{content:"\f013"}.fa-droplet-slash:before,.fa-tint-slash:before{content:"\f5c7"}.fa-mosque:before{content:"\f678"}.fa-mosquito:before{content:"\e52b"}.fa-star-of-david:before{content:"\f69a"}.fa-person-military-rifle:before{content:"\e54b"}.fa-cart-shopping:before,.fa-shopping-cart:before{content:"\f07a"}.fa-vials:before{content:"\f493"}.fa-plug-circle-plus:before{content:"\e55f"}.fa-place-of-worship:before{content:"\f67f"}.fa-grip-vertical:before{content:"\f58e"}.fa-arrow-turn-up:before,.fa-level-up:before{content:"\f148"}.fa-u:before{content:"U"}.fa-square-root-alt:before,.fa-square-root-variable:before{content:"\f698"}.fa-clock-four:before,.fa-clock:before{content:"\f017"}.fa-backward-step:before,.fa-step-backward:before{content:"\f048"}.fa-pallet:before{content:"\f482"}.fa-faucet:before{content:"\e005"}.fa-baseball-bat-ball:before{content:"\f432"}.fa-s:before{content:"S"}.fa-timeline:before{content:"\e29c"}.fa-keyboard:before{content:"\f11c"}.fa-caret-down:before{content:"\f0d7"}.fa-clinic-medical:before,.fa-house-chimney-medical:before{content:"\f7f2"}.fa-temperature-3:before,.fa-temperature-three-quarters:before,.fa-thermometer-3:before,.fa-thermometer-three-quarters:before{content:"\f2c8"}.fa-mobile-android-alt:before,.fa-mobile-screen:before{content:"\f3cf"}.fa-plane-up:before{content:"\e22d"}.fa-piggy-bank:before{content:"\f4d3"}.fa-battery-3:before,.fa-battery-half:before{content:"\f242"}.fa-mountain-city:before{content:"\e52e"}.fa-coins:before{content:"\f51e"}.fa-khanda:before{content:"\f66d"}.fa-sliders-h:before,.fa-sliders:before{content:"\f1de"}.fa-folder-tree:before{content:"\f802"}.fa-network-wired:before{content:"\f6ff"}.fa-map-pin:before{content:"\f276"}.fa-hamsa:before{content:"\f665"}.fa-cent-sign:before{content:"\e3f5"}.fa-flask:before{content:"\f0c3"}.fa-person-pregnant:before{content:"\e31e"}.fa-wand-sparkles:before{content:"\f72b"}.fa-ellipsis-v:before,.fa-ellipsis-vertical:before{content:"\f142"}.fa-ticket:before{content:"\f145"}.fa-power-off:before{content:"\f011"}.fa-long-arrow-alt-right:before,.fa-right-long:before{content:"\f30b"}.fa-flag-usa:before{content:"\f74d"}.fa-laptop-file:before{content:"\e51d"}.fa-teletype:before,.fa-tty:before{content:"\f1e4"}.fa-diagram-next:before{content:"\e476"}.fa-person-rifle:before{content:"\e54e"}.fa-house-medical-circle-exclamation:before{content:"\e512"}.fa-closed-captioning:before{content:"\f20a"}.fa-hiking:before,.fa-person-hiking:before{content:"\f6ec"}.fa-venus-double:before{content:"\f226"}.fa-images:before{content:"\f302"}.fa-calculator:before{content:"\f1ec"}.fa-people-pulling:before{content:"\e535"}.fa-n:before{content:"N"}.fa-cable-car:before,.fa-tram:before{content:"\f7da"}.fa-cloud-rain:before{content:"\f73d"}.fa-building-circle-xmark:before{content:"\e4d4"}.fa-ship:before{content:"\f21a"}.fa-arrows-down-to-line:before{content:"\e4b8"}.fa-download:before{content:"\f019"}.fa-face-grin:before,.fa-grin:before{content:"\f580"}.fa-backspace:before,.fa-delete-left:before{content:"\f55a"}.fa-eye-dropper-empty:before,.fa-eye-dropper:before,.fa-eyedropper:before{content:"\f1fb"}.fa-file-circle-check:before{content:"\e5a0"}.fa-forward:before{content:"\f04e"}.fa-mobile-android:before,.fa-mobile-phone:before,.fa-mobile:before{content:"\f3ce"}.fa-face-meh:before,.fa-meh:before{content:"\f11a"}.fa-align-center:before{content:"\f037"}.fa-book-dead:before,.fa-book-skull:before{content:"\f6b7"}.fa-drivers-license:before,.fa-id-card:before{content:"\f2c2"}.fa-dedent:before,.fa-outdent:before{content:"\f03b"}.fa-heart-circle-exclamation:before{content:"\e4fe"}.fa-home-alt:before,.fa-home-lg-alt:before,.fa-home:before,.fa-house:before{content:"\f015"}.fa-calendar-week:before{content:"\f784"}.fa-laptop-medical:before{content:"\f812"}.fa-b:before{content:"B"}.fa-file-medical:before{content:"\f477"}.fa-dice-one:before{content:"\f525"}.fa-kiwi-bird:before{content:"\f535"}.fa-arrow-right-arrow-left:before,.fa-exchange:before{content:"\f0ec"}.fa-redo-alt:before,.fa-rotate-forward:before,.fa-rotate-right:before{content:"\f2f9"}.fa-cutlery:before,.fa-utensils:before{content:"\f2e7"}.fa-arrow-up-wide-short:before,.fa-sort-amount-up:before{content:"\f161"}.fa-mill-sign:before{content:"\e1ed"}.fa-bowl-rice:before{content:"\e2eb"}.fa-skull:before{content:"\f54c"}.fa-broadcast-tower:before,.fa-tower-broadcast:before{content:"\f519"}.fa-truck-pickup:before{content:"\f63c"}.fa-long-arrow-alt-up:before,.fa-up-long:before{content:"\f30c"}.fa-stop:before{content:"\f04d"}.fa-code-merge:before{content:"\f387"}.fa-upload:before{content:"\f093"}.fa-hurricane:before{content:"\f751"}.fa-mound:before{content:"\e52d"}.fa-toilet-portable:before{content:"\e583"}.fa-compact-disc:before{content:"\f51f"}.fa-file-arrow-down:before,.fa-file-download:before{content:"\f56d"}.fa-caravan:before{content:"\f8ff"}.fa-shield-cat:before{content:"\e572"}.fa-bolt:before,.fa-zap:before{content:"\f0e7"}.fa-glass-water:before{content:"\e4f4"}.fa-oil-well:before{content:"\e532"}.fa-vault:before{content:"\e2c5"}.fa-mars:before{content:"\f222"}.fa-toilet:before{content:"\f7d8"}.fa-plane-circle-xmark:before{content:"\e557"}.fa-cny:before,.fa-jpy:before,.fa-rmb:before,.fa-yen-sign:before,.fa-yen:before{content:"\f157"}.fa-rouble:before,.fa-rub:before,.fa-ruble-sign:before,.fa-ruble:before{content:"\f158"}.fa-sun:before{content:"\f185"}.fa-guitar:before{content:"\f7a6"}.fa-face-laugh-wink:before,.fa-laugh-wink:before{content:"\f59c"}.fa-horse-head:before{content:"\f7ab"}.fa-bore-hole:before{content:"\e4c3"}.fa-industry:before{content:"\f275"}.fa-arrow-alt-circle-down:before,.fa-circle-down:before{content:"\f358"}.fa-arrows-turn-to-dots:before{content:"\e4c1"}.fa-florin-sign:before{content:"\e184"}.fa-arrow-down-short-wide:before,.fa-sort-amount-desc:before,.fa-sort-amount-down-alt:before{content:"\f884"}.fa-less-than:before{content:"\<"}.fa-angle-down:before{content:"\f107"}.fa-car-tunnel:before{content:"\e4de"}.fa-head-side-cough:before{content:"\e061"}.fa-grip-lines:before{content:"\f7a4"}.fa-thumbs-down:before{content:"\f165"}.fa-user-lock:before{content:"\f502"}.fa-arrow-right-long:before,.fa-long-arrow-right:before{content:"\f178"}.fa-anchor-circle-xmark:before{content:"\e4ac"}.fa-ellipsis-h:before,.fa-ellipsis:before{content:"\f141"}.fa-chess-pawn:before{content:"\f443"}.fa-first-aid:before,.fa-kit-medical:before{content:"\f479"}.fa-person-through-window:before{content:"\e5a9"}.fa-toolbox:before{content:"\f552"}.fa-hands-holding-circle:before{content:"\e4fb"}.fa-bug:before{content:"\f188"}.fa-credit-card-alt:before,.fa-credit-card:before{content:"\f09d"}.fa-automobile:before,.fa-car:before{content:"\f1b9"}.fa-hand-holding-hand:before{content:"\e4f7"}.fa-book-open-reader:before,.fa-book-reader:before{content:"\f5da"}.fa-mountain-sun:before{content:"\e52f"}.fa-arrows-left-right-to-line:before{content:"\e4ba"}.fa-dice-d20:before{content:"\f6cf"}.fa-truck-droplet:before{content:"\e58c"}.fa-file-circle-xmark:before{content:"\e5a1"}.fa-temperature-arrow-up:before,.fa-temperature-up:before{content:"\e040"}.fa-medal:before{content:"\f5a2"}.fa-bed:before{content:"\f236"}.fa-h-square:before,.fa-square-h:before{content:"\f0fd"}.fa-podcast:before{content:"\f2ce"}.fa-temperature-4:before,.fa-temperature-full:before,.fa-thermometer-4:before,.fa-thermometer-full:before{content:"\f2c7"}.fa-bell:before{content:"\f0f3"}.fa-superscript:before{content:"\f12b"}.fa-plug-circle-xmark:before{content:"\e560"}.fa-star-of-life:before{content:"\f621"}.fa-phone-slash:before{content:"\f3dd"}.fa-paint-roller:before{content:"\f5aa"}.fa-hands-helping:before,.fa-handshake-angle:before{content:"\f4c4"}.fa-location-dot:before,.fa-map-marker-alt:before{content:"\f3c5"}.fa-file:before{content:"\f15b"}.fa-greater-than:before{content:"\>"}.fa-person-swimming:before,.fa-swimmer:before{content:"\f5c4"}.fa-arrow-down:before{content:"\f063"}.fa-droplet:before,.fa-tint:before{content:"\f043"}.fa-eraser:before{content:"\f12d"}.fa-earth-america:before,.fa-earth-americas:before,.fa-earth:before,.fa-globe-americas:before{content:"\f57d"}.fa-person-burst:before{content:"\e53b"}.fa-dove:before{content:"\f4ba"}.fa-battery-0:before,.fa-battery-empty:before{content:"\f244"}.fa-socks:before{content:"\f696"}.fa-inbox:before{content:"\f01c"}.fa-section:before{content:"\e447"}.fa-gauge-high:before,.fa-tachometer-alt-fast:before,.fa-tachometer-alt:before{content:"\f625"}.fa-envelope-open-text:before{content:"\f658"}.fa-hospital-alt:before,.fa-hospital-wide:before,.fa-hospital:before{content:"\f0f8"}.fa-wine-bottle:before{content:"\f72f"}.fa-chess-rook:before{content:"\f447"}.fa-bars-staggered:before,.fa-reorder:before,.fa-stream:before{content:"\f550"}.fa-dharmachakra:before{content:"\f655"}.fa-hotdog:before{content:"\f80f"}.fa-blind:before,.fa-person-walking-with-cane:before{content:"\f29d"}.fa-drum:before{content:"\f569"}.fa-ice-cream:before{content:"\f810"}.fa-heart-circle-bolt:before{content:"\e4fc"}.fa-fax:before{content:"\f1ac"}.fa-paragraph:before{content:"\f1dd"}.fa-check-to-slot:before,.fa-vote-yea:before{content:"\f772"}.fa-star-half:before{content:"\f089"}.fa-boxes-alt:before,.fa-boxes-stacked:before,.fa-boxes:before{content:"\f468"}.fa-chain:before,.fa-link:before{content:"\f0c1"}.fa-assistive-listening-systems:before,.fa-ear-listen:before{content:"\f2a2"}.fa-tree-city:before{content:"\e587"}.fa-play:before{content:"\f04b"}.fa-font:before{content:"\f031"}.fa-table-cells-row-lock:before{content:"\e67a"}.fa-rupiah-sign:before{content:"\e23d"}.fa-magnifying-glass:before,.fa-search:before{content:"\f002"}.fa-ping-pong-paddle-ball:before,.fa-table-tennis-paddle-ball:before,.fa-table-tennis:before{content:"\f45d"}.fa-diagnoses:before,.fa-person-dots-from-line:before{content:"\f470"}.fa-trash-can-arrow-up:before,.fa-trash-restore-alt:before{content:"\f82a"}.fa-naira-sign:before{content:"\e1f6"}.fa-cart-arrow-down:before{content:"\f218"}.fa-walkie-talkie:before{content:"\f8ef"}.fa-file-edit:before,.fa-file-pen:before{content:"\f31c"}.fa-receipt:before{content:"\f543"}.fa-pen-square:before,.fa-pencil-square:before,.fa-square-pen:before{content:"\f14b"}.fa-suitcase-rolling:before{content:"\f5c1"}.fa-person-circle-exclamation:before{content:"\e53f"}.fa-chevron-down:before{content:"\f078"}.fa-battery-5:before,.fa-battery-full:before,.fa-battery:before{content:"\f240"}.fa-skull-crossbones:before{content:"\f714"}.fa-code-compare:before{content:"\e13a"}.fa-list-dots:before,.fa-list-ul:before{content:"\f0ca"}.fa-school-lock:before{content:"\e56f"}.fa-tower-cell:before{content:"\e585"}.fa-down-long:before,.fa-long-arrow-alt-down:before{content:"\f309"}.fa-ranking-star:before{content:"\e561"}.fa-chess-king:before{content:"\f43f"}.fa-person-harassing:before{content:"\e549"}.fa-brazilian-real-sign:before{content:"\e46c"}.fa-landmark-alt:before,.fa-landmark-dome:before{content:"\f752"}.fa-arrow-up:before{content:"\f062"}.fa-television:before,.fa-tv-alt:before,.fa-tv:before{content:"\f26c"}.fa-shrimp:before{content:"\e448"}.fa-list-check:before,.fa-tasks:before{content:"\f0ae"}.fa-jug-detergent:before{content:"\e519"}.fa-circle-user:before,.fa-user-circle:before{content:"\f2bd"}.fa-user-shield:before{content:"\f505"}.fa-wind:before{content:"\f72e"}.fa-car-burst:before,.fa-car-crash:before{content:"\f5e1"}.fa-y:before{content:"Y"}.fa-person-snowboarding:before,.fa-snowboarding:before{content:"\f7ce"}.fa-shipping-fast:before,.fa-truck-fast:before{content:"\f48b"}.fa-fish:before{content:"\f578"}.fa-user-graduate:before{content:"\f501"}.fa-adjust:before,.fa-circle-half-stroke:before{content:"\f042"}.fa-clapperboard:before{content:"\e131"}.fa-circle-radiation:before,.fa-radiation-alt:before{content:"\f7ba"}.fa-baseball-ball:before,.fa-baseball:before{content:"\f433"}.fa-jet-fighter-up:before{content:"\e518"}.fa-diagram-project:before,.fa-project-diagram:before{content:"\f542"}.fa-copy:before{content:"\f0c5"}.fa-volume-mute:before,.fa-volume-times:before,.fa-volume-xmark:before{content:"\f6a9"}.fa-hand-sparkles:before{content:"\e05d"}.fa-grip-horizontal:before,.fa-grip:before{content:"\f58d"}.fa-share-from-square:before,.fa-share-square:before{content:"\f14d"}.fa-child-combatant:before,.fa-child-rifle:before{content:"\e4e0"}.fa-gun:before{content:"\e19b"}.fa-phone-square:before,.fa-square-phone:before{content:"\f098"}.fa-add:before,.fa-plus:before{content:"\+"}.fa-expand:before{content:"\f065"}.fa-computer:before{content:"\e4e5"}.fa-close:before,.fa-multiply:before,.fa-remove:before,.fa-times:before,.fa-xmark:before{content:"\f00d"}.fa-arrows-up-down-left-right:before,.fa-arrows:before{content:"\f047"}.fa-chalkboard-teacher:before,.fa-chalkboard-user:before{content:"\f51c"}.fa-peso-sign:before{content:"\e222"}.fa-building-shield:before{content:"\e4d8"}.fa-baby:before{content:"\f77c"}.fa-users-line:before{content:"\e592"}.fa-quote-left-alt:before,.fa-quote-left:before{content:"\f10d"}.fa-tractor:before{content:"\f722"}.fa-trash-arrow-up:before,.fa-trash-restore:before{content:"\f829"}.fa-arrow-down-up-lock:before{content:"\e4b0"}.fa-lines-leaning:before{content:"\e51e"}.fa-ruler-combined:before{content:"\f546"}.fa-copyright:before{content:"\f1f9"}.fa-equals:before{content:"\="}.fa-blender:before{content:"\f517"}.fa-teeth:before{content:"\f62e"}.fa-ils:before,.fa-shekel-sign:before,.fa-shekel:before,.fa-sheqel-sign:before,.fa-sheqel:before{content:"\f20b"}.fa-map:before{content:"\f279"}.fa-rocket:before{content:"\f135"}.fa-photo-film:before,.fa-photo-video:before{content:"\f87c"}.fa-folder-minus:before{content:"\f65d"}.fa-store:before{content:"\f54e"}.fa-arrow-trend-up:before{content:"\e098"}.fa-plug-circle-minus:before{content:"\e55e"}.fa-sign-hanging:before,.fa-sign:before{content:"\f4d9"}.fa-bezier-curve:before{content:"\f55b"}.fa-bell-slash:before{content:"\f1f6"}.fa-tablet-android:before,.fa-tablet:before{content:"\f3fb"}.fa-school-flag:before{content:"\e56e"}.fa-fill:before{content:"\f575"}.fa-angle-up:before{content:"\f106"}.fa-drumstick-bite:before{content:"\f6d7"}.fa-holly-berry:before{content:"\f7aa"}.fa-chevron-left:before{content:"\f053"}.fa-bacteria:before{content:"\e059"}.fa-hand-lizard:before{content:"\f258"}.fa-notdef:before{content:"\e1fe"}.fa-disease:before{content:"\f7fa"}.fa-briefcase-medical:before{content:"\f469"}.fa-genderless:before{content:"\f22d"}.fa-chevron-right:before{content:"\f054"}.fa-retweet:before{content:"\f079"}.fa-car-alt:before,.fa-car-rear:before{content:"\f5de"}.fa-pump-soap:before{content:"\e06b"}.fa-video-slash:before{content:"\f4e2"}.fa-battery-2:before,.fa-battery-quarter:before{content:"\f243"}.fa-radio:before{content:"\f8d7"}.fa-baby-carriage:before,.fa-carriage-baby:before{content:"\f77d"}.fa-traffic-light:before{content:"\f637"}.fa-thermometer:before{content:"\f491"}.fa-vr-cardboard:before{content:"\f729"}.fa-hand-middle-finger:before{content:"\f806"}.fa-percent:before,.fa-percentage:before{content:"\%"}.fa-truck-moving:before{content:"\f4df"}.fa-glass-water-droplet:before{content:"\e4f5"}.fa-display:before{content:"\e163"}.fa-face-smile:before,.fa-smile:before{content:"\f118"}.fa-thumb-tack:before,.fa-thumbtack:before{content:"\f08d"}.fa-trophy:before{content:"\f091"}.fa-person-praying:before,.fa-pray:before{content:"\f683"}.fa-hammer:before{content:"\f6e3"}.fa-hand-peace:before{content:"\f25b"}.fa-rotate:before,.fa-sync-alt:before{content:"\f2f1"}.fa-spinner:before{content:"\f110"}.fa-robot:before{content:"\f544"}.fa-peace:before{content:"\f67c"}.fa-cogs:before,.fa-gears:before{content:"\f085"}.fa-warehouse:before{content:"\f494"}.fa-arrow-up-right-dots:before{content:"\e4b7"}.fa-splotch:before{content:"\f5bc"}.fa-face-grin-hearts:before,.fa-grin-hearts:before{content:"\f584"}.fa-dice-four:before{content:"\f524"}.fa-sim-card:before{content:"\f7c4"}.fa-transgender-alt:before,.fa-transgender:before{content:"\f225"}.fa-mercury:before{content:"\f223"}.fa-arrow-turn-down:before,.fa-level-down:before{content:"\f149"}.fa-person-falling-burst:before{content:"\e547"}.fa-award:before{content:"\f559"}.fa-ticket-alt:before,.fa-ticket-simple:before{content:"\f3ff"}.fa-building:before{content:"\f1ad"}.fa-angle-double-left:before,.fa-angles-left:before{content:"\f100"}.fa-qrcode:before{content:"\f029"}.fa-clock-rotate-left:before,.fa-history:before{content:"\f1da"}.fa-face-grin-beam-sweat:before,.fa-grin-beam-sweat:before{content:"\f583"}.fa-arrow-right-from-file:before,.fa-file-export:before{content:"\f56e"}.fa-shield-blank:before,.fa-shield:before{content:"\f132"}.fa-arrow-up-short-wide:before,.fa-sort-amount-up-alt:before{content:"\f885"}.fa-house-medical:before{content:"\e3b2"}.fa-golf-ball-tee:before,.fa-golf-ball:before{content:"\f450"}.fa-chevron-circle-left:before,.fa-circle-chevron-left:before{content:"\f137"}.fa-house-chimney-window:before{content:"\e00d"}.fa-pen-nib:before{content:"\f5ad"}.fa-tent-arrow-turn-left:before{content:"\e580"}.fa-tents:before{content:"\e582"}.fa-magic:before,.fa-wand-magic:before{content:"\f0d0"}.fa-dog:before{content:"\f6d3"}.fa-carrot:before{content:"\f787"}.fa-moon:before{content:"\f186"}.fa-wine-glass-alt:before,.fa-wine-glass-empty:before{content:"\f5ce"}.fa-cheese:before{content:"\f7ef"}.fa-yin-yang:before{content:"\f6ad"}.fa-music:before{content:"\f001"}.fa-code-commit:before{content:"\f386"}.fa-temperature-low:before{content:"\f76b"}.fa-biking:before,.fa-person-biking:before{content:"\f84a"}.fa-broom:before{content:"\f51a"}.fa-shield-heart:before{content:"\e574"}.fa-gopuram:before{content:"\f664"}.fa-earth-oceania:before,.fa-globe-oceania:before{content:"\e47b"}.fa-square-xmark:before,.fa-times-square:before,.fa-xmark-square:before{content:"\f2d3"}.fa-hashtag:before{content:"\#"}.fa-expand-alt:before,.fa-up-right-and-down-left-from-center:before{content:"\f424"}.fa-oil-can:before{content:"\f613"}.fa-t:before{content:"T"}.fa-hippo:before{content:"\f6ed"}.fa-chart-column:before{content:"\e0e3"}.fa-infinity:before{content:"\f534"}.fa-vial-circle-check:before{content:"\e596"}.fa-person-arrow-down-to-line:before{content:"\e538"}.fa-voicemail:before{content:"\f897"}.fa-fan:before{content:"\f863"}.fa-person-walking-luggage:before{content:"\e554"}.fa-arrows-alt-v:before,.fa-up-down:before{content:"\f338"}.fa-cloud-moon-rain:before{content:"\f73c"}.fa-calendar:before{content:"\f133"}.fa-trailer:before{content:"\e041"}.fa-bahai:before,.fa-haykal:before{content:"\f666"}.fa-sd-card:before{content:"\f7c2"}.fa-dragon:before{content:"\f6d5"}.fa-shoe-prints:before{content:"\f54b"}.fa-circle-plus:before,.fa-plus-circle:before{content:"\f055"}.fa-face-grin-tongue-wink:before,.fa-grin-tongue-wink:before{content:"\f58b"}.fa-hand-holding:before{content:"\f4bd"}.fa-plug-circle-exclamation:before{content:"\e55d"}.fa-chain-broken:before,.fa-chain-slash:before,.fa-link-slash:before,.fa-unlink:before{content:"\f127"}.fa-clone:before{content:"\f24d"}.fa-person-walking-arrow-loop-left:before{content:"\e551"}.fa-arrow-up-z-a:before,.fa-sort-alpha-up-alt:before{content:"\f882"}.fa-fire-alt:before,.fa-fire-flame-curved:before{content:"\f7e4"}.fa-tornado:before{content:"\f76f"}.fa-file-circle-plus:before{content:"\e494"}.fa-book-quran:before,.fa-quran:before{content:"\f687"}.fa-anchor:before{content:"\f13d"}.fa-border-all:before{content:"\f84c"}.fa-angry:before,.fa-face-angry:before{content:"\f556"}.fa-cookie-bite:before{content:"\f564"}.fa-arrow-trend-down:before{content:"\e097"}.fa-feed:before,.fa-rss:before{content:"\f09e"}.fa-draw-polygon:before{content:"\f5ee"}.fa-balance-scale:before,.fa-scale-balanced:before{content:"\f24e"}.fa-gauge-simple-high:before,.fa-tachometer-fast:before,.fa-tachometer:before{content:"\f62a"}.fa-shower:before{content:"\f2cc"}.fa-desktop-alt:before,.fa-desktop:before{content:"\f390"}.fa-m:before{content:"M"}.fa-table-list:before,.fa-th-list:before{content:"\f00b"}.fa-comment-sms:before,.fa-sms:before{content:"\f7cd"}.fa-book:before{content:"\f02d"}.fa-user-plus:before{content:"\f234"}.fa-check:before{content:"\f00c"}.fa-battery-4:before,.fa-battery-three-quarters:before{content:"\f241"}.fa-house-circle-check:before{content:"\e509"}.fa-angle-left:before{content:"\f104"}.fa-diagram-successor:before{content:"\e47a"}.fa-truck-arrow-right:before{content:"\e58b"}.fa-arrows-split-up-and-left:before{content:"\e4bc"}.fa-fist-raised:before,.fa-hand-fist:before{content:"\f6de"}.fa-cloud-moon:before{content:"\f6c3"}.fa-briefcase:before{content:"\f0b1"}.fa-person-falling:before{content:"\e546"}.fa-image-portrait:before,.fa-portrait:before{content:"\f3e0"}.fa-user-tag:before{content:"\f507"}.fa-rug:before{content:"\e569"}.fa-earth-europe:before,.fa-globe-europe:before{content:"\f7a2"}.fa-cart-flatbed-suitcase:before,.fa-luggage-cart:before{content:"\f59d"}.fa-rectangle-times:before,.fa-rectangle-xmark:before,.fa-times-rectangle:before,.fa-window-close:before{content:"\f410"}.fa-baht-sign:before{content:"\e0ac"}.fa-book-open:before{content:"\f518"}.fa-book-journal-whills:before,.fa-journal-whills:before{content:"\f66a"}.fa-handcuffs:before{content:"\e4f8"}.fa-exclamation-triangle:before,.fa-triangle-exclamation:before,.fa-warning:before{content:"\f071"}.fa-database:before{content:"\f1c0"}.fa-mail-forward:before,.fa-share:before{content:"\f064"}.fa-bottle-droplet:before{content:"\e4c4"}.fa-mask-face:before{content:"\e1d7"}.fa-hill-rockslide:before{content:"\e508"}.fa-exchange-alt:before,.fa-right-left:before{content:"\f362"}.fa-paper-plane:before{content:"\f1d8"}.fa-road-circle-exclamation:before{content:"\e565"}.fa-dungeon:before{content:"\f6d9"}.fa-align-right:before{content:"\f038"}.fa-money-bill-1-wave:before,.fa-money-bill-wave-alt:before{content:"\f53b"}.fa-life-ring:before{content:"\f1cd"}.fa-hands:before,.fa-sign-language:before,.fa-signing:before{content:"\f2a7"}.fa-calendar-day:before{content:"\f783"}.fa-ladder-water:before,.fa-swimming-pool:before,.fa-water-ladder:before{content:"\f5c5"}.fa-arrows-up-down:before,.fa-arrows-v:before{content:"\f07d"}.fa-face-grimace:before,.fa-grimace:before{content:"\f57f"}.fa-wheelchair-alt:before,.fa-wheelchair-move:before{content:"\e2ce"}.fa-level-down-alt:before,.fa-turn-down:before{content:"\f3be"}.fa-person-walking-arrow-right:before{content:"\e552"}.fa-envelope-square:before,.fa-square-envelope:before{content:"\f199"}.fa-dice:before{content:"\f522"}.fa-bowling-ball:before{content:"\f436"}.fa-brain:before{content:"\f5dc"}.fa-band-aid:before,.fa-bandage:before{content:"\f462"}.fa-calendar-minus:before{content:"\f272"}.fa-circle-xmark:before,.fa-times-circle:before,.fa-xmark-circle:before{content:"\f057"}.fa-gifts:before{content:"\f79c"}.fa-hotel:before{content:"\f594"}.fa-earth-asia:before,.fa-globe-asia:before{content:"\f57e"}.fa-id-card-alt:before,.fa-id-card-clip:before{content:"\f47f"}.fa-magnifying-glass-plus:before,.fa-search-plus:before{content:"\f00e"}.fa-thumbs-up:before{content:"\f164"}.fa-user-clock:before{content:"\f4fd"}.fa-allergies:before,.fa-hand-dots:before{content:"\f461"}.fa-file-invoice:before{content:"\f570"}.fa-window-minimize:before{content:"\f2d1"}.fa-coffee:before,.fa-mug-saucer:before{content:"\f0f4"}.fa-brush:before{content:"\f55d"}.fa-mask:before{content:"\f6fa"}.fa-magnifying-glass-minus:before,.fa-search-minus:before{content:"\f010"}.fa-ruler-vertical:before{content:"\f548"}.fa-user-alt:before,.fa-user-large:before{content:"\f406"}.fa-train-tram:before{content:"\e5b4"}.fa-user-nurse:before{content:"\f82f"}.fa-syringe:before{content:"\f48e"}.fa-cloud-sun:before{content:"\f6c4"}.fa-stopwatch-20:before{content:"\e06f"}.fa-square-full:before{content:"\f45c"}.fa-magnet:before{content:"\f076"}.fa-jar:before{content:"\e516"}.fa-note-sticky:before,.fa-sticky-note:before{content:"\f249"}.fa-bug-slash:before{content:"\e490"}.fa-arrow-up-from-water-pump:before{content:"\e4b6"}.fa-bone:before{content:"\f5d7"}.fa-user-injured:before{content:"\f728"}.fa-face-sad-tear:before,.fa-sad-tear:before{content:"\f5b4"}.fa-plane:before{content:"\f072"}.fa-tent-arrows-down:before{content:"\e581"}.fa-exclamation:before{content:"\!"}.fa-arrows-spin:before{content:"\e4bb"}.fa-print:before{content:"\f02f"}.fa-try:before,.fa-turkish-lira-sign:before,.fa-turkish-lira:before{content:"\e2bb"}.fa-dollar-sign:before,.fa-dollar:before,.fa-usd:before{content:"\$"}.fa-x:before{content:"X"}.fa-magnifying-glass-dollar:before,.fa-search-dollar:before{content:"\f688"}.fa-users-cog:before,.fa-users-gear:before{content:"\f509"}.fa-person-military-pointing:before{content:"\e54a"}.fa-bank:before,.fa-building-columns:before,.fa-institution:before,.fa-museum:before,.fa-university:before{content:"\f19c"}.fa-umbrella:before{content:"\f0e9"}.fa-trowel:before{content:"\e589"}.fa-d:before{content:"D"}.fa-stapler:before{content:"\e5af"}.fa-masks-theater:before,.fa-theater-masks:before{content:"\f630"}.fa-kip-sign:before{content:"\e1c4"}.fa-hand-point-left:before{content:"\f0a5"}.fa-handshake-alt:before,.fa-handshake-simple:before{content:"\f4c6"}.fa-fighter-jet:before,.fa-jet-fighter:before{content:"\f0fb"}.fa-share-alt-square:before,.fa-square-share-nodes:before{content:"\f1e1"}.fa-barcode:before{content:"\f02a"}.fa-plus-minus:before{content:"\e43c"}.fa-video-camera:before,.fa-video:before{content:"\f03d"}.fa-graduation-cap:before,.fa-mortar-board:before{content:"\f19d"}.fa-hand-holding-medical:before{content:"\e05c"}.fa-person-circle-check:before{content:"\e53e"}.fa-level-up-alt:before,.fa-turn-up:before{content:"\f3bf"}.fa-sr-only,.fa-sr-only-focusable:not(:focus),.sr-only,.sr-only-focusable:not(:focus){clip:rect(0,0,0,0);border-width:0;height:1px;margin:-1px;overflow:hidden;padding:0;position:absolute;white-space:nowrap;width:1px}
+
+/*!
+ * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com
+ * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)
+ * Copyright 2024 Fonticons, Inc.
+ */:host,:root{--fa-font-solid:normal 900 1em/1 "Font Awesome 6 Free"}@font-face{font-display:block;font-family:Font Awesome\ 6 Free;font-style:normal;font-weight:900;src:url(../vendor/fontawesome/webfonts/fa-solid-900.woff2) format("woff2"),url(../vendor/fontawesome/webfonts/fa-solid-900.ttf) format("truetype")}.fa-solid,.fas{font-weight:900}
+
+/*!
+ * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com
+ * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)
+ * Copyright 2024 Fonticons, Inc.
+ */:host,:root{--fa-style-family-classic:"Font Awesome 6 Free";--fa-font-regular:normal 400 1em/1 "Font Awesome 6 Free"}@font-face{font-display:block;font-family:Font Awesome\ 6 Free;font-style:normal;font-weight:400;src:url(../vendor/fontawesome/webfonts/fa-regular-400.woff2) format("woff2"),url(../vendor/fontawesome/webfonts/fa-regular-400.ttf) format("truetype")}.fa-regular,.far{font-weight:400}
+
+/*!
+ * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com
+ * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)
+ * Copyright 2024 Fonticons, Inc.
+ */:host,:root{--fa-style-family-brands:"Font Awesome 6 Brands";--fa-font-brands:normal 400 1em/1 "Font Awesome 6 Brands"}@font-face{font-display:block;font-family:Font Awesome\ 6 Brands;font-style:normal;font-weight:400;src:url(../vendor/fontawesome/webfonts/fa-brands-400.woff2) format("woff2"),url(../vendor/fontawesome/webfonts/fa-brands-400.ttf) format("truetype")}.fa-brands,.fab{font-weight:400}.fa-monero:before{content:"\f3d0"}.fa-hooli:before{content:"\f427"}.fa-yelp:before{content:"\f1e9"}.fa-cc-visa:before{content:"\f1f0"}.fa-lastfm:before{content:"\f202"}.fa-shopware:before{content:"\f5b5"}.fa-creative-commons-nc:before{content:"\f4e8"}.fa-aws:before{content:"\f375"}.fa-redhat:before{content:"\f7bc"}.fa-yoast:before{content:"\f2b1"}.fa-cloudflare:before{content:"\e07d"}.fa-ups:before{content:"\f7e0"}.fa-pixiv:before{content:"\e640"}.fa-wpexplorer:before{content:"\f2de"}.fa-dyalog:before{content:"\f399"}.fa-bity:before{content:"\f37a"}.fa-stackpath:before{content:"\f842"}.fa-buysellads:before{content:"\f20d"}.fa-first-order:before{content:"\f2b0"}.fa-modx:before{content:"\f285"}.fa-guilded:before{content:"\e07e"}.fa-vnv:before{content:"\f40b"}.fa-js-square:before,.fa-square-js:before{content:"\f3b9"}.fa-microsoft:before{content:"\f3ca"}.fa-qq:before{content:"\f1d6"}.fa-orcid:before{content:"\f8d2"}.fa-java:before{content:"\f4e4"}.fa-invision:before{content:"\f7b0"}.fa-creative-commons-pd-alt:before{content:"\f4ed"}.fa-centercode:before{content:"\f380"}.fa-glide-g:before{content:"\f2a6"}.fa-drupal:before{content:"\f1a9"}.fa-jxl:before{content:"\e67b"}.fa-hire-a-helper:before{content:"\f3b0"}.fa-creative-commons-by:before{content:"\f4e7"}.fa-unity:before{content:"\e049"}.fa-whmcs:before{content:"\f40d"}.fa-rocketchat:before{content:"\f3e8"}.fa-vk:before{content:"\f189"}.fa-untappd:before{content:"\f405"}.fa-mailchimp:before{content:"\f59e"}.fa-css3-alt:before{content:"\f38b"}.fa-reddit-square:before,.fa-square-reddit:before{content:"\f1a2"}.fa-vimeo-v:before{content:"\f27d"}.fa-contao:before{content:"\f26d"}.fa-square-font-awesome:before{content:"\e5ad"}.fa-deskpro:before{content:"\f38f"}.fa-brave:before{content:"\e63c"}.fa-sistrix:before{content:"\f3ee"}.fa-instagram-square:before,.fa-square-instagram:before{content:"\e055"}.fa-battle-net:before{content:"\f835"}.fa-the-red-yeti:before{content:"\f69d"}.fa-hacker-news-square:before,.fa-square-hacker-news:before{content:"\f3af"}.fa-edge:before{content:"\f282"}.fa-threads:before{content:"\e618"}.fa-napster:before{content:"\f3d2"}.fa-snapchat-square:before,.fa-square-snapchat:before{content:"\f2ad"}.fa-google-plus-g:before{content:"\f0d5"}.fa-artstation:before{content:"\f77a"}.fa-markdown:before{content:"\f60f"}.fa-sourcetree:before{content:"\f7d3"}.fa-google-plus:before{content:"\f2b3"}.fa-diaspora:before{content:"\f791"}.fa-foursquare:before{content:"\f180"}.fa-stack-overflow:before{content:"\f16c"}.fa-github-alt:before{content:"\f113"}.fa-phoenix-squadron:before{content:"\f511"}.fa-pagelines:before{content:"\f18c"}.fa-algolia:before{content:"\f36c"}.fa-red-river:before{content:"\f3e3"}.fa-creative-commons-sa:before{content:"\f4ef"}.fa-safari:before{content:"\f267"}.fa-google:before{content:"\f1a0"}.fa-font-awesome-alt:before,.fa-square-font-awesome-stroke:before{content:"\f35c"}.fa-atlassian:before{content:"\f77b"}.fa-linkedin-in:before{content:"\f0e1"}.fa-digital-ocean:before{content:"\f391"}.fa-nimblr:before{content:"\f5a8"}.fa-chromecast:before{content:"\f838"}.fa-evernote:before{content:"\f839"}.fa-hacker-news:before{content:"\f1d4"}.fa-creative-commons-sampling:before{content:"\f4f0"}.fa-adversal:before{content:"\f36a"}.fa-creative-commons:before{content:"\f25e"}.fa-watchman-monitoring:before{content:"\e087"}.fa-fonticons:before{content:"\f280"}.fa-weixin:before{content:"\f1d7"}.fa-shirtsinbulk:before{content:"\f214"}.fa-codepen:before{content:"\f1cb"}.fa-git-alt:before{content:"\f841"}.fa-lyft:before{content:"\f3c3"}.fa-rev:before{content:"\f5b2"}.fa-windows:before{content:"\f17a"}.fa-wizards-of-the-coast:before{content:"\f730"}.fa-square-viadeo:before,.fa-viadeo-square:before{content:"\f2aa"}.fa-meetup:before{content:"\f2e0"}.fa-centos:before{content:"\f789"}.fa-adn:before{content:"\f170"}.fa-cloudsmith:before{content:"\f384"}.fa-opensuse:before{content:"\e62b"}.fa-pied-piper-alt:before{content:"\f1a8"}.fa-dribbble-square:before,.fa-square-dribbble:before{content:"\f397"}.fa-codiepie:before{content:"\f284"}.fa-node:before{content:"\f419"}.fa-mix:before{content:"\f3cb"}.fa-steam:before{content:"\f1b6"}.fa-cc-apple-pay:before{content:"\f416"}.fa-scribd:before{content:"\f28a"}.fa-debian:before{content:"\e60b"}.fa-openid:before{content:"\f19b"}.fa-instalod:before{content:"\e081"}.fa-expeditedssl:before{content:"\f23e"}.fa-sellcast:before{content:"\f2da"}.fa-square-twitter:before,.fa-twitter-square:before{content:"\f081"}.fa-r-project:before{content:"\f4f7"}.fa-delicious:before{content:"\f1a5"}.fa-freebsd:before{content:"\f3a4"}.fa-vuejs:before{content:"\f41f"}.fa-accusoft:before{content:"\f369"}.fa-ioxhost:before{content:"\f208"}.fa-fonticons-fi:before{content:"\f3a2"}.fa-app-store:before{content:"\f36f"}.fa-cc-mastercard:before{content:"\f1f1"}.fa-itunes-note:before{content:"\f3b5"}.fa-golang:before{content:"\e40f"}.fa-kickstarter:before,.fa-square-kickstarter:before{content:"\f3bb"}.fa-grav:before{content:"\f2d6"}.fa-weibo:before{content:"\f18a"}.fa-uncharted:before{content:"\e084"}.fa-firstdraft:before{content:"\f3a1"}.fa-square-youtube:before,.fa-youtube-square:before{content:"\f431"}.fa-wikipedia-w:before{content:"\f266"}.fa-rendact:before,.fa-wpressr:before{content:"\f3e4"}.fa-angellist:before{content:"\f209"}.fa-galactic-republic:before{content:"\f50c"}.fa-nfc-directional:before{content:"\e530"}.fa-skype:before{content:"\f17e"}.fa-joget:before{content:"\f3b7"}.fa-fedora:before{content:"\f798"}.fa-stripe-s:before{content:"\f42a"}.fa-meta:before{content:"\e49b"}.fa-laravel:before{content:"\f3bd"}.fa-hotjar:before{content:"\f3b1"}.fa-bluetooth-b:before{content:"\f294"}.fa-square-letterboxd:before{content:"\e62e"}.fa-sticker-mule:before{content:"\f3f7"}.fa-creative-commons-zero:before{content:"\f4f3"}.fa-hips:before{content:"\f452"}.fa-behance:before{content:"\f1b4"}.fa-reddit:before{content:"\f1a1"}.fa-discord:before{content:"\f392"}.fa-chrome:before{content:"\f268"}.fa-app-store-ios:before{content:"\f370"}.fa-cc-discover:before{content:"\f1f2"}.fa-wpbeginner:before{content:"\f297"}.fa-confluence:before{content:"\f78d"}.fa-shoelace:before{content:"\e60c"}.fa-mdb:before{content:"\f8ca"}.fa-dochub:before{content:"\f394"}.fa-accessible-icon:before{content:"\f368"}.fa-ebay:before{content:"\f4f4"}.fa-amazon:before{content:"\f270"}.fa-unsplash:before{content:"\e07c"}.fa-yarn:before{content:"\f7e3"}.fa-square-steam:before,.fa-steam-square:before{content:"\f1b7"}.fa-500px:before{content:"\f26e"}.fa-square-vimeo:before,.fa-vimeo-square:before{content:"\f194"}.fa-asymmetrik:before{content:"\f372"}.fa-font-awesome-flag:before,.fa-font-awesome-logo-full:before,.fa-font-awesome:before{content:"\f2b4"}.fa-gratipay:before{content:"\f184"}.fa-apple:before{content:"\f179"}.fa-hive:before{content:"\e07f"}.fa-gitkraken:before{content:"\f3a6"}.fa-keybase:before{content:"\f4f5"}.fa-apple-pay:before{content:"\f415"}.fa-padlet:before{content:"\e4a0"}.fa-amazon-pay:before{content:"\f42c"}.fa-github-square:before,.fa-square-github:before{content:"\f092"}.fa-stumbleupon:before{content:"\f1a4"}.fa-fedex:before{content:"\f797"}.fa-phoenix-framework:before{content:"\f3dc"}.fa-shopify:before{content:"\e057"}.fa-neos:before{content:"\f612"}.fa-square-threads:before{content:"\e619"}.fa-hackerrank:before{content:"\f5f7"}.fa-researchgate:before{content:"\f4f8"}.fa-swift:before{content:"\f8e1"}.fa-angular:before{content:"\f420"}.fa-speakap:before{content:"\f3f3"}.fa-angrycreative:before{content:"\f36e"}.fa-y-combinator:before{content:"\f23b"}.fa-empire:before{content:"\f1d1"}.fa-envira:before{content:"\f299"}.fa-google-scholar:before{content:"\e63b"}.fa-gitlab-square:before,.fa-square-gitlab:before{content:"\e5ae"}.fa-studiovinari:before{content:"\f3f8"}.fa-pied-piper:before{content:"\f2ae"}.fa-wordpress:before{content:"\f19a"}.fa-product-hunt:before{content:"\f288"}.fa-firefox:before{content:"\f269"}.fa-linode:before{content:"\f2b8"}.fa-goodreads:before{content:"\f3a8"}.fa-odnoklassniki-square:before,.fa-square-odnoklassniki:before{content:"\f264"}.fa-jsfiddle:before{content:"\f1cc"}.fa-sith:before{content:"\f512"}.fa-themeisle:before{content:"\f2b2"}.fa-page4:before{content:"\f3d7"}.fa-hashnode:before{content:"\e499"}.fa-react:before{content:"\f41b"}.fa-cc-paypal:before{content:"\f1f4"}.fa-squarespace:before{content:"\f5be"}.fa-cc-stripe:before{content:"\f1f5"}.fa-creative-commons-share:before{content:"\f4f2"}.fa-bitcoin:before{content:"\f379"}.fa-keycdn:before{content:"\f3ba"}.fa-opera:before{content:"\f26a"}.fa-itch-io:before{content:"\f83a"}.fa-umbraco:before{content:"\f8e8"}.fa-galactic-senate:before{content:"\f50d"}.fa-ubuntu:before{content:"\f7df"}.fa-draft2digital:before{content:"\f396"}.fa-stripe:before{content:"\f429"}.fa-houzz:before{content:"\f27c"}.fa-gg:before{content:"\f260"}.fa-dhl:before{content:"\f790"}.fa-pinterest-square:before,.fa-square-pinterest:before{content:"\f0d3"}.fa-xing:before{content:"\f168"}.fa-blackberry:before{content:"\f37b"}.fa-creative-commons-pd:before{content:"\f4ec"}.fa-playstation:before{content:"\f3df"}.fa-quinscape:before{content:"\f459"}.fa-less:before{content:"\f41d"}.fa-blogger-b:before{content:"\f37d"}.fa-opencart:before{content:"\f23d"}.fa-vine:before{content:"\f1ca"}.fa-signal-messenger:before{content:"\e663"}.fa-paypal:before{content:"\f1ed"}.fa-gitlab:before{content:"\f296"}.fa-typo3:before{content:"\f42b"}.fa-reddit-alien:before{content:"\f281"}.fa-yahoo:before{content:"\f19e"}.fa-dailymotion:before{content:"\e052"}.fa-affiliatetheme:before{content:"\f36b"}.fa-pied-piper-pp:before{content:"\f1a7"}.fa-bootstrap:before{content:"\f836"}.fa-odnoklassniki:before{content:"\f263"}.fa-nfc-symbol:before{content:"\e531"}.fa-mintbit:before{content:"\e62f"}.fa-ethereum:before{content:"\f42e"}.fa-speaker-deck:before{content:"\f83c"}.fa-creative-commons-nc-eu:before{content:"\f4e9"}.fa-patreon:before{content:"\f3d9"}.fa-avianex:before{content:"\f374"}.fa-ello:before{content:"\f5f1"}.fa-gofore:before{content:"\f3a7"}.fa-bimobject:before{content:"\f378"}.fa-brave-reverse:before{content:"\e63d"}.fa-facebook-f:before{content:"\f39e"}.fa-google-plus-square:before,.fa-square-google-plus:before{content:"\f0d4"}.fa-web-awesome:before{content:"\e682"}.fa-mandalorian:before{content:"\f50f"}.fa-first-order-alt:before{content:"\f50a"}.fa-osi:before{content:"\f41a"}.fa-google-wallet:before{content:"\f1ee"}.fa-d-and-d-beyond:before{content:"\f6ca"}.fa-periscope:before{content:"\f3da"}.fa-fulcrum:before{content:"\f50b"}.fa-cloudscale:before{content:"\f383"}.fa-forumbee:before{content:"\f211"}.fa-mizuni:before{content:"\f3cc"}.fa-schlix:before{content:"\f3ea"}.fa-square-xing:before,.fa-xing-square:before{content:"\f169"}.fa-bandcamp:before{content:"\f2d5"}.fa-wpforms:before{content:"\f298"}.fa-cloudversify:before{content:"\f385"}.fa-usps:before{content:"\f7e1"}.fa-megaport:before{content:"\f5a3"}.fa-magento:before{content:"\f3c4"}.fa-spotify:before{content:"\f1bc"}.fa-optin-monster:before{content:"\f23c"}.fa-fly:before{content:"\f417"}.fa-aviato:before{content:"\f421"}.fa-itunes:before{content:"\f3b4"}.fa-cuttlefish:before{content:"\f38c"}.fa-blogger:before{content:"\f37c"}.fa-flickr:before{content:"\f16e"}.fa-viber:before{content:"\f409"}.fa-soundcloud:before{content:"\f1be"}.fa-digg:before{content:"\f1a6"}.fa-tencent-weibo:before{content:"\f1d5"}.fa-letterboxd:before{content:"\e62d"}.fa-symfony:before{content:"\f83d"}.fa-maxcdn:before{content:"\f136"}.fa-etsy:before{content:"\f2d7"}.fa-facebook-messenger:before{content:"\f39f"}.fa-audible:before{content:"\f373"}.fa-think-peaks:before{content:"\f731"}.fa-bilibili:before{content:"\e3d9"}.fa-erlang:before{content:"\f39d"}.fa-x-twitter:before{content:"\e61b"}.fa-cotton-bureau:before{content:"\f89e"}.fa-dashcube:before{content:"\f210"}.fa-42-group:before,.fa-innosoft:before{content:"\e080"}.fa-stack-exchange:before{content:"\f18d"}.fa-elementor:before{content:"\f430"}.fa-pied-piper-square:before,.fa-square-pied-piper:before{content:"\e01e"}.fa-creative-commons-nd:before{content:"\f4eb"}.fa-palfed:before{content:"\f3d8"}.fa-superpowers:before{content:"\f2dd"}.fa-resolving:before{content:"\f3e7"}.fa-xbox:before{content:"\f412"}.fa-square-web-awesome-stroke:before{content:"\e684"}.fa-searchengin:before{content:"\f3eb"}.fa-tiktok:before{content:"\e07b"}.fa-facebook-square:before,.fa-square-facebook:before{content:"\f082"}.fa-renren:before{content:"\f18b"}.fa-linux:before{content:"\f17c"}.fa-glide:before{content:"\f2a5"}.fa-linkedin:before{content:"\f08c"}.fa-hubspot:before{content:"\f3b2"}.fa-deploydog:before{content:"\f38e"}.fa-twitch:before{content:"\f1e8"}.fa-ravelry:before{content:"\f2d9"}.fa-mixer:before{content:"\e056"}.fa-lastfm-square:before,.fa-square-lastfm:before{content:"\f203"}.fa-vimeo:before{content:"\f40a"}.fa-mendeley:before{content:"\f7b3"}.fa-uniregistry:before{content:"\f404"}.fa-figma:before{content:"\f799"}.fa-creative-commons-remix:before{content:"\f4ee"}.fa-cc-amazon-pay:before{content:"\f42d"}.fa-dropbox:before{content:"\f16b"}.fa-instagram:before{content:"\f16d"}.fa-cmplid:before{content:"\e360"}.fa-upwork:before{content:"\e641"}.fa-facebook:before{content:"\f09a"}.fa-gripfire:before{content:"\f3ac"}.fa-jedi-order:before{content:"\f50e"}.fa-uikit:before{content:"\f403"}.fa-fort-awesome-alt:before{content:"\f3a3"}.fa-phabricator:before{content:"\f3db"}.fa-ussunnah:before{content:"\f407"}.fa-earlybirds:before{content:"\f39a"}.fa-trade-federation:before{content:"\f513"}.fa-autoprefixer:before{content:"\f41c"}.fa-whatsapp:before{content:"\f232"}.fa-square-upwork:before{content:"\e67c"}.fa-slideshare:before{content:"\f1e7"}.fa-google-play:before{content:"\f3ab"}.fa-viadeo:before{content:"\f2a9"}.fa-line:before{content:"\f3c0"}.fa-google-drive:before{content:"\f3aa"}.fa-servicestack:before{content:"\f3ec"}.fa-simplybuilt:before{content:"\f215"}.fa-bitbucket:before{content:"\f171"}.fa-imdb:before{content:"\f2d8"}.fa-deezer:before{content:"\e077"}.fa-raspberry-pi:before{content:"\f7bb"}.fa-jira:before{content:"\f7b1"}.fa-docker:before{content:"\f395"}.fa-screenpal:before{content:"\e570"}.fa-bluetooth:before{content:"\f293"}.fa-gitter:before{content:"\f426"}.fa-d-and-d:before{content:"\f38d"}.fa-microblog:before{content:"\e01a"}.fa-cc-diners-club:before{content:"\f24c"}.fa-gg-circle:before{content:"\f261"}.fa-pied-piper-hat:before{content:"\f4e5"}.fa-kickstarter-k:before{content:"\f3bc"}.fa-yandex:before{content:"\f413"}.fa-readme:before{content:"\f4d5"}.fa-html5:before{content:"\f13b"}.fa-sellsy:before{content:"\f213"}.fa-square-web-awesome:before{content:"\e683"}.fa-sass:before{content:"\f41e"}.fa-wirsindhandwerk:before,.fa-wsh:before{content:"\e2d0"}.fa-buromobelexperte:before{content:"\f37f"}.fa-salesforce:before{content:"\f83b"}.fa-octopus-deploy:before{content:"\e082"}.fa-medapps:before{content:"\f3c6"}.fa-ns8:before{content:"\f3d5"}.fa-pinterest-p:before{content:"\f231"}.fa-apper:before{content:"\f371"}.fa-fort-awesome:before{content:"\f286"}.fa-waze:before{content:"\f83f"}.fa-bluesky:before{content:"\e671"}.fa-cc-jcb:before{content:"\f24b"}.fa-snapchat-ghost:before,.fa-snapchat:before{content:"\f2ab"}.fa-fantasy-flight-games:before{content:"\f6dc"}.fa-rust:before{content:"\e07a"}.fa-wix:before{content:"\f5cf"}.fa-behance-square:before,.fa-square-behance:before{content:"\f1b5"}.fa-supple:before{content:"\f3f9"}.fa-webflow:before{content:"\e65c"}.fa-rebel:before{content:"\f1d0"}.fa-css3:before{content:"\f13c"}.fa-staylinked:before{content:"\f3f5"}.fa-kaggle:before{content:"\f5fa"}.fa-space-awesome:before{content:"\e5ac"}.fa-deviantart:before{content:"\f1bd"}.fa-cpanel:before{content:"\f388"}.fa-goodreads-g:before{content:"\f3a9"}.fa-git-square:before,.fa-square-git:before{content:"\f1d2"}.fa-square-tumblr:before,.fa-tumblr-square:before{content:"\f174"}.fa-trello:before{content:"\f181"}.fa-creative-commons-nc-jp:before{content:"\f4ea"}.fa-get-pocket:before{content:"\f265"}.fa-perbyte:before{content:"\e083"}.fa-grunt:before{content:"\f3ad"}.fa-weebly:before{content:"\f5cc"}.fa-connectdevelop:before{content:"\f20e"}.fa-leanpub:before{content:"\f212"}.fa-black-tie:before{content:"\f27e"}.fa-themeco:before{content:"\f5c6"}.fa-python:before{content:"\f3e2"}.fa-android:before{content:"\f17b"}.fa-bots:before{content:"\e340"}.fa-free-code-camp:before{content:"\f2c5"}.fa-hornbill:before{content:"\f592"}.fa-js:before{content:"\f3b8"}.fa-ideal:before{content:"\e013"}.fa-git:before{content:"\f1d3"}.fa-dev:before{content:"\f6cc"}.fa-sketch:before{content:"\f7c6"}.fa-yandex-international:before{content:"\f414"}.fa-cc-amex:before{content:"\f1f3"}.fa-uber:before{content:"\f402"}.fa-github:before{content:"\f09b"}.fa-php:before{content:"\f457"}.fa-alipay:before{content:"\f642"}.fa-youtube:before{content:"\f167"}.fa-skyatlas:before{content:"\f216"}.fa-firefox-browser:before{content:"\e007"}.fa-replyd:before{content:"\f3e6"}.fa-suse:before{content:"\f7d6"}.fa-jenkins:before{content:"\f3b6"}.fa-twitter:before{content:"\f099"}.fa-rockrms:before{content:"\f3e9"}.fa-pinterest:before{content:"\f0d2"}.fa-buffer:before{content:"\f837"}.fa-npm:before{content:"\f3d4"}.fa-yammer:before{content:"\f840"}.fa-btc:before{content:"\f15a"}.fa-dribbble:before{content:"\f17d"}.fa-stumbleupon-circle:before{content:"\f1a3"}.fa-internet-explorer:before{content:"\f26b"}.fa-stubber:before{content:"\e5c7"}.fa-telegram-plane:before,.fa-telegram:before{content:"\f2c6"}.fa-old-republic:before{content:"\f510"}.fa-odysee:before{content:"\e5c6"}.fa-square-whatsapp:before,.fa-whatsapp-square:before{content:"\f40c"}.fa-node-js:before{content:"\f3d3"}.fa-edge-legacy:before{content:"\e078"}.fa-slack-hash:before,.fa-slack:before{content:"\f198"}.fa-medrt:before{content:"\f3c8"}.fa-usb:before{content:"\f287"}.fa-tumblr:before{content:"\f173"}.fa-vaadin:before{content:"\f408"}.fa-quora:before{content:"\f2c4"}.fa-square-x-twitter:before{content:"\e61a"}.fa-reacteurope:before{content:"\f75d"}.fa-medium-m:before,.fa-medium:before{content:"\f23a"}.fa-amilia:before{content:"\f36d"}.fa-mixcloud:before{content:"\f289"}.fa-flipboard:before{content:"\f44d"}.fa-viacoin:before{content:"\f237"}.fa-critical-role:before{content:"\f6c9"}.fa-sitrox:before{content:"\e44a"}.fa-discourse:before{content:"\f393"}.fa-joomla:before{content:"\f1aa"}.fa-mastodon:before{content:"\f4f6"}.fa-airbnb:before{content:"\f834"}.fa-wolf-pack-battalion:before{content:"\f514"}.fa-buy-n-large:before{content:"\f8a6"}.fa-gulp:before{content:"\f3ae"}.fa-creative-commons-sampling-plus:before{content:"\f4f1"}.fa-strava:before{content:"\f428"}.fa-ember:before{content:"\f423"}.fa-canadian-maple-leaf:before{content:"\f785"}.fa-teamspeak:before{content:"\f4f9"}.fa-pushed:before{content:"\f3e1"}.fa-wordpress-simple:before{content:"\f411"}.fa-nutritionix:before{content:"\f3d6"}.fa-wodu:before{content:"\e088"}.fa-google-pay:before{content:"\e079"}.fa-intercom:before{content:"\f7af"}.fa-zhihu:before{content:"\f63f"}.fa-korvue:before{content:"\f42f"}.fa-pix:before{content:"\e43a"}.fa-steam-symbol:before{content:"\f3f6"}html{--pst-header-height:4rem;--pst-header-article-height:calc(var(--pst-header-height)*2/3);--pst-sidebar-secondary:17rem;--pst-font-size-base:1rem;--pst-font-size-h1:2.625rem;--pst-font-size-h2:2.125rem;--pst-font-size-h3:1.75rem;--pst-font-size-h4:1.5rem;--pst-font-size-h5:1.25rem;--pst-font-size-h6:1rem;--pst-font-size-milli:0.9rem;--pst-sidebar-font-size:0.9rem;--pst-sidebar-font-size-mobile:1.1rem;--pst-sidebar-header-font-size:1.2rem;--pst-sidebar-header-font-weight:600;--pst-admonition-font-weight-heading:600;--pst-font-weight-caption:300;--pst-font-weight-heading:600;--pst-font-family-base-system:-apple-system,"BlinkMacSystemFont","Segoe UI","Helvetica Neue","Arial",sans-serif,"Apple Color Emoji","Segoe UI Emoji","Segoe UI Symbol";--pst-font-family-monospace-system:"SFMono-Regular","Menlo","Consolas","Monaco","Liberation Mono","Lucida Console",monospace;--pst-font-family-base:var(--pst-font-family-base-system);--pst-font-family-heading:var(--pst-font-family-base-system);--pst-font-family-monospace:var(--pst-font-family-monospace-system);--pst-font-size-icon:1.5rem;--pst-icon-check-circle:"";--pst-icon-info-circle:"";--pst-icon-exclamation-triangle:"";--pst-icon-exclamation-circle:"";--pst-icon-times-circle:"";--pst-icon-lightbulb:"";--pst-icon-download:"";--pst-icon-angle-left:"";--pst-icon-angle-right:"";--pst-icon-external-link:"";--pst-icon-search-minus:"";--pst-icon-github:"";--pst-icon-gitlab:"";--pst-icon-share:"";--pst-icon-bell:"";--pst-icon-pencil:"";--pst-breadcrumb-divider:"";--pst-icon-admonition-default:var(--pst-icon-bell);--pst-icon-admonition-note:var(--pst-icon-info-circle);--pst-icon-admonition-attention:var(--pst-icon-exclamation-circle);--pst-icon-admonition-caution:var(--pst-icon-exclamation-triangle);--pst-icon-admonition-warning:var(--pst-icon-exclamation-triangle);--pst-icon-admonition-danger:var(--pst-icon-exclamation-triangle);--pst-icon-admonition-error:var(--pst-icon-times-circle);--pst-icon-admonition-hint:var(--pst-icon-lightbulb);--pst-icon-admonition-tip:var(--pst-icon-lightbulb);--pst-icon-admonition-important:var(--pst-icon-exclamation-circle);--pst-icon-admonition-seealso:var(--pst-icon-share);--pst-icon-admonition-todo:var(--pst-icon-pencil);--pst-icon-versionmodified-default:var(--pst-icon-exclamation-circle);--pst-icon-versionmodified-added:var(--pst-icon-exclamation-circle);--pst-icon-versionmodified-changed:var(--pst-icon-exclamation-circle);--pst-icon-versionmodified-deprecated:var(--pst-icon-exclamation-circle);font-size:var(--pst-font-size-base);scroll-padding-top:calc(var(--pst-header-height) + 1rem)}body{background-color:var(--pst-color-background);color:var(--pst-color-text-base);display:flex;flex-direction:column;font-family:var(--pst-font-family-base);font-weight:400;line-height:1.65;min-height:100vh}body::-webkit-scrollbar-track{background-color:var(--pst-color-background)}p{font-size:1em;margin-bottom:1.15rem}p.rubric{border-bottom:1px solid var(--pst-color-border)}p.centered{text-align:center}a{word-wrap:break-word;color:var(--pst-color-link);text-decoration:underline;text-decoration-thickness:max(1px,.0625rem);text-underline-offset:.1578em}a:hover{color:var(--pst-color-link-hover);text-decoration-skip-ink:none;text-decoration-thickness:max(3px,.1875rem,.12em)}a:active,a:visited{color:var(--pst-color-link)}a:visited:hover{color:var(--pst-color-link-hover)}a.headerlink{color:var(--pst-color-secondary);font-size:.8em;margin-left:.2em;padding:0 4px;text-decoration:none;user-select:none}a.github:before,a.gitlab:before{color:var(--pst-color-text-muted);font:var(--fa-font-brands);margin-right:.25rem}a.github:before{content:var(--pst-icon-github)}a.gitlab:before{content:var(--pst-icon-gitlab)}.h1,.h2,.h3,.h4,.h5,.h6,h1,h2,h3,h4,h5,h6{font-family:var(--pst-font-family-heading);font-weight:var(--pst-font-weight-heading);line-height:1.15;margin:2.75rem 0 1.05rem}.h1,h1{font-size:var(--pst-font-size-h1);margin-top:0}.h1,.h2,h1,h2{color:var(--pst-color-heading)}.h2,h2{font-size:var(--pst-font-size-h2)}.h3,h3{font-size:var(--pst-font-size-h3)}.h3,.h4,h3,h4{color:var(--pst-color-heading)}.h4,h4{font-size:var(--pst-font-size-h4)}.h5,h5{font-size:var(--pst-font-size-h5)}.h5,.h6,h5,h6{color:var(--pst-color-text-base)}.h6,h6{font-size:var(--pst-font-size-h6)}.small,.text_small,small{font-size:var(--pst-font-size-milli)}hr{border:0;border-top:1px solid var(--pst-color-border)}code,kbd,pre,samp{font-family:var(--pst-font-family-monospace)}kbd{background-color:var(--pst-color-on-background);color:var(--pst-color-text-muted)}kbd:not(.compound){border:1px solid var(--pst-color-border);box-shadow:1px 1px 1px var(--pst-color-shadow);margin:0 .1rem;padding:.1rem .4rem}code{color:var(--pst-color-inline-code)}pre{background-color:var(--pst-color-surface);border:1px solid var(--pst-color-border);border-radius:.25rem;color:var(--pst-color-text-base);line-height:1.2em;margin:1.5em 0;padding:1rem}pre .linenos{opacity:.8;padding-right:10px}#pst-back-to-top{background-color:var(--pst-color-secondary);border:none;color:var(--pst-color-secondary-text);display:none;left:50vw;position:fixed;top:90vh;transform:translate(-50%);z-index:1080}#pst-back-to-top .fa-arrow-up{margin-inline-end:.5em}#pst-back-to-top:hover{background-color:var(--pst-color-secondary-highlight);color:var(--pst-color-link-hover);color:var(--pst-color-secondary-highlight-text);text-decoration:underline;text-decoration-skip-ink:none;text-decoration-thickness:max(1px,.0625rem);text-decoration-thickness:max(3px,.1875rem,.12em);text-decoration-thickness:1px;text-underline-offset:.1578em}#pst-back-to-top:focus-visible{box-shadow:none;outline:var(--pst-color-accent) solid .1875rem;outline-color:var(--pst-color-secondary);outline-offset:.1875rem}:focus-visible{box-shadow:none;outline:var(--pst-color-accent) solid .1875rem}.skip-link{background-color:var(--pst-color-warning);border-bottom:1px solid var(--pst-color-border);left:0;padding:.5rem;position:fixed;right:0;text-align:center;top:0;transform:translateY(-100%);transition:transform .15s ease-in-out;z-index:1055}.skip-link:focus-within{transform:translateY(0)}.skip-link a{color:var(--pst-color-warning-text)!important}.skip-link a:focus-visible{outline-color:#14181e}.bd-container{display:flex;flex-grow:1;justify-content:center}.bd-container .bd-container__inner{display:flex}.bd-page-width{width:100%}@media (min-width:960px){.bd-page-width{max-width:88rem}}.pst-async-banner-revealer{height:0;overflow:hidden;transition:height .3s ease-in-out}@media (prefers-reduced-motion){.pst-async-banner-revealer{transition:none}}#bd-header-version-warning,.bd-header-announcement{align-items:center;display:flex;justify-content:center;min-height:3rem;padding:.5rem 12.5%;position:relative;text-align:center;width:100%}@media (max-width:959.98px){#bd-header-version-warning,.bd-header-announcement{padding:.5rem 2%}}#bd-header-version-warning p,.bd-header-announcement p{font-weight:700;margin:0}#bd-header-version-warning a,.bd-header-announcement a{color:var(--pst-color-inline-code-links)}#bd-header-version-warning .pst-button-link-to-stable-version,.bd-header-announcement .pst-button-link-to-stable-version{background-color:var(--pst-color-danger);border-color:var(--pst-color-danger);border-radius:.25rem;box-shadow:0 .2rem .5rem var(--pst-color-shadow),0 0 .0625rem var(--pst-color-shadow)!important;color:var(--pst-color-danger-text)}#bd-header-version-warning .pst-button-link-to-stable-version:hover,.bd-header-announcement .pst-button-link-to-stable-version:hover{background-color:var(--pst-color-danger-highlight);border-color:var(--pst-color-danger-highlight);color:var(--pst-color-link-hover);color:var(--pst-color-danger-highlight-text);text-decoration:underline;text-decoration-skip-ink:none;text-decoration-thickness:max(1px,.0625rem);text-decoration-thickness:max(3px,.1875rem,.12em);text-decoration-thickness:1px;text-underline-offset:.1578em}#bd-header-version-warning .pst-button-link-to-stable-version:focus-visible,.bd-header-announcement .pst-button-link-to-stable-version:focus-visible{outline:.1875rem solid var(--pst-color-danger);outline-offset:.1875rem}.bd-header-announcement{background-color:var(--pst-color-secondary-bg)}#bd-header-version-warning{background-color:var(--pst-color-danger-bg)}.bd-main{display:flex;flex-direction:column;flex-grow:1;min-width:0}.bd-main .bd-content{display:flex;height:100%;justify-content:center}.bd-main .bd-content .bd-article-container{display:flex;flex-direction:column;justify-content:start;max-width:60em;overflow-x:auto;padding:1rem;width:100%}@media (min-width:1200px){.bd-main .bd-content .bd-article-container .bd-article{padding-left:2rem;padding-top:1.5rem}}.bd-footer{border-top:1px solid var(--pst-color-border);width:100%}.bd-footer .bd-footer__inner{display:flex;flex-grow:1;margin:auto;padding:1rem}.bd-footer .footer-items__center,.bd-footer .footer-items__end,.bd-footer .footer-items__start{display:flex;flex-direction:column;flex-grow:1;gap:.5rem;justify-content:center}.bd-footer .footer-items__center{text-align:center}.bd-footer .footer-items__end{text-align:end}.bd-footer .footer-item p{margin-bottom:0}.bd-footer-article{margin-top:auto}.bd-footer-article .footer-article-items{display:flex;flex-direction:column}.bd-footer-content .footer-content-items{display:flex;flex-direction:column;margin-top:auto}.bd-header{background-color:var(--pst-color-on-background)!important;box-shadow:0 .125rem .25rem 0 var(--pst-color-shadow);justify-content:center;max-width:100vw;padding:0;position:sticky;top:0;width:100%;z-index:1030}.bd-header .bd-header__inner{align-items:center;display:flex;height:fit-content;padding-left:1rem;padding-right:1rem}.bd-header :focus-visible{border-radius:.125rem}.bd-header .navbar-item{align-items:center;display:flex;height:var(--pst-header-height);max-height:var(--pst-header-height)}.bd-header .navbar-header-items{display:none;flex-shrink:1}@media (min-width:960px){.bd-header .navbar-header-items{display:inherit;flex-grow:1;padding:0 0 0 .5rem}}.bd-header .navbar-header-items__center,.bd-header .navbar-header-items__end,.bd-header .navbar-header-items__start{align-items:center;display:flex;flex-flow:wrap;row-gap:0}.bd-header .navbar-header-items__center,.bd-header .navbar-header-items__end{column-gap:1.12rem}.bd-header .navbar-header-items__start{flex-shrink:0;gap:.5rem;margin-right:auto}.bd-header .navbar-header-items__end{justify-content:end}.bd-header ul.navbar-nav{display:flex}@media (min-width:960px){.bd-header ul.navbar-nav{align-items:baseline}}.bd-header ul.navbar-nav>li.nav-item{margin-inline:2px}.bd-header ul.navbar-nav>li.nav-item>.nav-link{color:var(--pst-color-text-muted);padding-inline:6px;position:relative}.bd-header ul.navbar-nav>li.nav-item>.nav-link:before{background-color:transparent;content:"";display:block;inset:0;bottom:calc(max(3px, .1875rem, .12em)*-1);margin:max(3px,.1875rem,.12em) 0;position:absolute}.bd-header ul.navbar-nav>li.nav-item>.nav-link:hover{color:var(--pst-color-secondary);text-decoration:none}.bd-header ul.navbar-nav>li.nav-item>.nav-link:hover:before{border-bottom:max(3px,.1875rem,.12em) solid var(--pst-color-secondary)}.bd-header ul.navbar-nav>li.nav-item>.nav-link:focus-visible{box-shadow:none;outline:3px solid var(--pst-color-accent);outline-offset:.1875rem}.bd-header ul.navbar-nav>li.nav-item.current>.nav-link{color:var(--pst-color-primary)}.bd-header ul.navbar-nav>li.nav-item.current>.nav-link:before{border-bottom:3px solid var(--pst-color-primary)}.bd-header ul.navbar-nav>li.nav-item.dropdown{margin-inline:4px}.bd-header ul.navbar-nav>li.nav-item.dropdown button{padding-inline:8px}.bd-header ul.navbar-nav>li.nav-item.dropdown>.dropdown-toggle{border-radius:.125rem;color:var(--pst-color-text-muted)}.bd-header ul.navbar-nav>li.nav-item.dropdown>.dropdown-toggle:focus-visible{box-shadow:0 0 0 .1875rem var(--pst-color-accent)}.bd-header ul.navbar-nav>li.nav-item.dropdown>.dropdown-toggle:hover{box-shadow:0 0 0 .1875rem var(--pst-color-link-hover);color:var(--pst-color-text-base);text-decoration:none}.bd-header ul.navbar-nav li a.nav-link.dropdown-item{color:var(--pst-color-text-muted);text-decoration:none}.bd-header ul.navbar-nav li a.nav-link.dropdown-item:hover{color:var(--pst-color-link-hover);text-decoration:underline;text-decoration-skip-ink:none;text-decoration-thickness:max(1px,.0625rem);text-decoration-thickness:max(3px,.1875rem,.12em);text-underline-offset:.1578em}.bd-header ul.navbar-nav .dropdown button{border:none;display:unset}.bd-header ul.navbar-nav .dropdown button:hover{color:var(--pst-color-link-hover);text-decoration:underline;text-decoration-skip-ink:none;text-decoration-thickness:max(1px,.0625rem);text-decoration-thickness:max(3px,.1875rem,.12em);text-underline-offset:.1578em}.bd-header ul.navbar-nav .dropdown .dropdown-menu{background-color:var(--pst-color-on-background);border:1px solid var(--pst-color-border);box-shadow:0 0 .3rem .1rem var(--pst-color-shadow);margin:.5rem 0;min-width:20rem;padding:.5rem 0;z-index:1070}.bd-header ul.navbar-nav .dropdown .dropdown-menu .dropdown-item{padding:.25rem 1.5rem}.bd-header ul.navbar-nav .dropdown .dropdown-menu .dropdown-item:focus:not(:hover,:active){background-color:inherit}.bd-header ul.navbar-nav .dropdown .dropdown-menu .dropdown-item:focus-visible{z-index:10}.bd-header ul.navbar-nav .dropdown .dropdown-menu:not(.show){display:none}.bd-header button.sidebar-toggle{background-color:inherit;color:var(--pst-color-muted);font-size:var(--pst-font-size-icon);margin-bottom:0;padding:.5rem}.bd-header button.primary-toggle{margin-right:1rem}@media (min-width:960px){.bd-header button.primary-toggle{display:none}}.bd-header button.secondary-toggle{margin-left:1rem}@media (min-width:1200px){.bd-header button.secondary-toggle{display:none}}@media (min-width:960px){.navbar-center-items .navbar-item{display:inline-block}}.nav-link{transition:none}.nav-link:hover{color:var(--pst-color-link-hover);text-decoration:underline;text-decoration-skip-ink:none;text-decoration-thickness:max(1px,.0625rem);text-decoration-thickness:max(3px,.1875rem,.12em);text-underline-offset:.1578em}.nav-link.nav-external:after{content:var(--pst-icon-external-link);font:var(--fa-font-solid);font-size:.75em;margin-left:.3em}.bd-navbar-elements li.nav-item i{font-size:.7rem;padding-left:2px;vertical-align:middle}.navbar-persistent--mobile{margin-left:auto}@media (min-width:960px){.navbar-persistent--mobile{display:none}}.navbar-persistent--container{display:none}@media (min-width:960px){.navbar-persistent--container{display:flex}}.header-article__inner{padding:0 .5rem}.header-article__inner .header-article-item{min-height:var(--pst-header-article-height)}.header-article__inner .header-article-items__end{margin-left:auto}.bd-sidebar-primary{background-color:var(--pst-color-background);border-right:1px solid var(--pst-color-border);display:flex;flex:0 0 auto;flex-direction:column;font-size:var(--pst-sidebar-font-size-mobile);gap:1rem;max-height:calc(100vh - var(--pst-header-height));overflow-y:auto;padding:2rem 1rem 1rem;position:sticky;top:var(--pst-header-height);width:25%}@media (min-width:960px){.bd-sidebar-primary{font-size:var(--pst-sidebar-font-size)}}.bd-sidebar-primary :focus-visible{border-radius:.125rem}.bd-sidebar-primary .nav-link{font-size:var(--pst-sidebar-font-size-mobile)}.bd-sidebar-primary.no-sidebar{border-right:0}@media (min-width:960px){.bd-sidebar-primary.hide-on-wide{display:none}}.bd-sidebar-primary .h1,.bd-sidebar-primary .h2,.bd-sidebar-primary .h3,.bd-sidebar-primary .h4,.bd-sidebar-primary h1,.bd-sidebar-primary h2,.bd-sidebar-primary h3,.bd-sidebar-primary h4{color:var(--pst-color-text-base)}.bd-sidebar-primary .sidebar-primary-items__end .sidebar-primary-item,.bd-sidebar-primary .sidebar-primary-items__start .sidebar-primary-item{padding:.5rem 0}.bd-sidebar-primary .sidebar-header-items{display:flex;flex-direction:column}.bd-sidebar-primary .sidebar-header-items .sidebar-header-items__title{color:var(--pst-color-text-base);font-size:var(--pst-sidebar-header-font-size);font-weight:var(--pst-sidebar-header-font-weight);margin-bottom:.5rem}.bd-sidebar-primary .sidebar-header-items .bd-navbar-elements .nav-link:focus-visible{box-shadow:none;outline:var(--pst-color-accent) solid .1875rem;outline-offset:.1875rem}.bd-sidebar-primary .sidebar-header-items .sidebar-header-items__center{display:flex;flex-direction:column}.bd-sidebar-primary .sidebar-header-items .sidebar-header-items__end{align-items:center;display:flex;gap:1rem}@media (min-width:960px){.bd-sidebar-primary .sidebar-header-items{display:none}}.bd-sidebar-primary .sidebar-primary-items__start{border-top:1px solid var(--pst-color-border)}@media (min-width:960px){.bd-sidebar-primary .sidebar-primary-items__start{border-top:none}}.bd-sidebar-primary .sidebar-primary-items__end{margin-bottom:1em;margin-top:auto}.bd-sidebar-primary .list-caption{list-style:none;padding-left:0}.bd-sidebar-primary .list-caption li.toctree-l0.has-children>details>summary{align-items:baseline;display:flex;height:auto;justify-content:space-between;position:relative;width:auto}.bd-sidebar-primary .list-caption li.toctree-l0.has-children>details>summary .toctree-toggle{flex:0 0 auto}.bd-sidebar-primary .list-caption li.toctree-l0.has-children>details>summary .toctree-toggle .fa-chevron-down{font-size:1rem}.bd-sidebar-primary li.has-children{position:relative}.bd-sidebar-primary li.has-children .caption,.bd-sidebar-primary li.has-children>.reference{margin-right:calc(30px + .1875rem);padding-top:.25rem}.bd-sidebar-primary li.has-children>details>summary{list-style:none}.bd-sidebar-primary li.has-children>details>summary::-webkit-details-marker{display:none}.bd-sidebar-primary li.has-children>details>summary:focus-visible{outline:none}.bd-sidebar-primary li.has-children>details>summary:focus-visible>.toctree-toggle{outline:var(--pst-color-accent) solid .1875rem;outline-offset:-.1875rem}.bd-sidebar-primary li.has-children>details>summary .toctree-toggle{align-items:center;cursor:pointer;display:inline-flex;height:30px;justify-content:center;position:absolute;right:0;top:0;width:30px}.bd-sidebar-primary li.has-children>details>summary .toctree-toggle .fa-chevron-down{font-size:.75rem}.bd-sidebar-primary li.has-children>details[open]>summary .fa-chevron-down{transform:rotate(180deg)}nav.bd-links{margin-right:-1rem}@media (min-width:960px){nav.bd-links{display:block}}nav.bd-links ul{display:block;list-style:none}nav.bd-links ul ul{padding:0 0 0 1rem}nav.bd-links li>a{box-shadow:none;color:var(--pst-color-text-muted);display:block;margin-right:.1875rem;padding:.25rem .65rem;text-decoration:none}nav.bd-links li>a:hover{background-color:transparent;text-decoration:underline;text-decoration-skip-ink:none;text-decoration-thickness:max(3px,.1875rem,.12em)}nav.bd-links li>a:active,nav.bd-links li>a:hover{color:var(--pst-color-link-hover)}nav.bd-links li>a:focus-visible{box-shadow:0 0 0 .1875rem var(--pst-color-accent);outline:none;z-index:10}nav.bd-links li>a.reference.external:after{content:var(--pst-icon-external-link);font:var(--fa-font-solid);font-size:.75em;margin-left:.3em}nav.bd-links .current>a{background-color:transparent;box-shadow:inset max(3px,.1875rem,.12em) 0 0 var(--pst-color-primary);color:var(--pst-color-primary);font-weight:600}nav.bd-links .current>a:focus-visible{box-shadow:inset max(3px,.1875rem,.12em) 0 0 var(--pst-color-primary),0 0 0 .1875rem var(--pst-color-accent);outline:none}nav.bd-links p.bd-links__title{font-size:var(--pst-sidebar-header-font-size)}nav.bd-links p.bd-links__title,nav.bd-links p.caption{font-weight:var(--pst-sidebar-header-font-weight);margin-bottom:.5rem}nav.bd-links p.caption{color:var(--pst-color-text-base);font-size:var(--pst-sidebar-font-size-mobile);margin-top:1.25rem;position:relative}nav.bd-links p.caption:first-child{margin-top:0}@media (min-width:960px){nav.bd-links p.caption{font-size:var(--pst-sidebar-font-size)}}.bd-sidebar-secondary{background-color:var(--pst-color-background);display:flex;flex-direction:column;flex-shrink:0;font-size:var(--pst-sidebar-font-size-mobile);max-height:calc(100vh - var(--pst-header-height));order:2;overflow-y:auto;padding:2rem 1rem 1rem;position:sticky;top:var(--pst-header-height);width:var(--pst-sidebar-secondary)}@media (min-width:1200px){.bd-sidebar-secondary{font-size:var(--pst-sidebar-font-size)}}.sidebar-secondary-item{padding:.5rem}@media (min-width:1200px){.sidebar-secondary-item{border-left:1px solid var(--pst-color-border);padding-left:1rem}}.sidebar-secondary-item i{padding-right:.5rem}.bd-sidebar::backdrop{background-color:#000;opacity:.5}@media (max-width:959.98px){.bd-sidebar-primary{border:0;flex-grow:0.75;height:100vh;left:0;margin-left:-75%;max-height:100vh;max-width:350px;position:fixed;top:0;transition:visibility .2s ease-out,margin .2s ease-out;visibility:hidden;width:75%;z-index:1055}}.bd-sidebar-primary[open]{margin-left:0;visibility:visible}@media (max-width:1199.98px){.bd-sidebar-secondary{border:0;flex-grow:0.75;height:100vh;margin-right:-75%;max-height:100vh;max-width:350px;position:fixed;right:0;top:0;transition:visibility .2s ease-out,margin .2s ease-out;visibility:hidden;width:75%;z-index:1055}}.bd-sidebar-secondary[open]{margin-right:0;visibility:visible}ul.bd-breadcrumbs{display:flex;flex-wrap:wrap;font-size:.8rem;list-style:none;padding-left:0}ul.bd-breadcrumbs li.breadcrumb-item{align-items:center;display:flex;font-weight:700;overflow-x:hidden;white-space:nowrap}ul.bd-breadcrumbs li.breadcrumb-item .ellipsis,ul.bd-breadcrumbs li.breadcrumb-item a{margin:.1875rem;overflow-x:hidden;text-overflow:ellipsis}ul.bd-breadcrumbs li.breadcrumb-item a{color:var(--pst-color-text-muted);text-decoration:none}ul.bd-breadcrumbs li.breadcrumb-item a:hover{color:var(--pst-color-link-hover);text-decoration:underline;text-decoration-skip-ink:none;text-decoration-thickness:max(1px,.0625rem);text-decoration-thickness:max(3px,.1875rem,.12em);text-underline-offset:.1578em}ul.bd-breadcrumbs li.breadcrumb-item:not(.breadcrumb-home):before{color:var(--pst-color-text-muted);content:var(--pst-breadcrumb-divider);font:var(--fa-font-solid);font-size:.8rem;padding:0 .5rem}html .pst-navbar-icon{align-items:center;border:none;border-radius:0;box-sizing:border-box;color:var(--pst-color-text-muted);display:flex;font-size:1rem;height:2.15rem;justify-content:center;line-height:1.65;min-height:24px;min-width:24px;padding:.25rem 0;position:relative}html .pst-navbar-icon:before{background-color:transparent;content:"";display:block;inset:0;bottom:calc(max(3px, .1875rem, .12em)*-1);margin:max(3px,.1875rem,.12em) 0;position:absolute}html .pst-navbar-icon:hover{color:var(--pst-color-secondary);text-decoration:none}html .pst-navbar-icon:hover:before{border-bottom:max(3px,.1875rem,.12em) solid var(--pst-color-secondary)}html .pst-navbar-icon:focus-visible{box-shadow:none;outline:3px solid var(--pst-color-accent);outline-offset:.1875rem}ul.navbar-icon-links{align-items:center;column-gap:1.12rem;display:flex;flex-flow:row wrap;justify-content:space-evenly;list-style:none;margin-bottom:0;padding-left:0}ul.navbar-icon-links i.fa-brands,ul.navbar-icon-links i.fa-regular,ul.navbar-icon-links i.fa-solid{font-size:var(--pst-font-size-icon);font-style:normal;vertical-align:middle}ul.navbar-icon-links i.fa-square-twitter:before{color:#55acee}ul.navbar-icon-links i.fa-square-gitlab:before{color:#548}ul.navbar-icon-links i.fa-bitbucket:before{color:#0052cc}ul.navbar-icon-links img.icon-link-image{border-radius:.2rem;height:1.5em}ul.navbar-icon-links .fa-pydata{stroke:var(--pst-color-background);stroke-linejoin:round;stroke-width:.35}.navbar-brand{align-items:center;display:flex;flex-shrink:0;gap:.5rem;height:var(--pst-header-height);margin:0;max-height:var(--pst-header-height);padding:.5rem 0;position:relative;width:auto}.navbar-brand p{color:var(--pst-color-text-base);margin-bottom:0}.navbar-brand img{height:100%;max-width:100%;width:auto}.navbar-brand:hover,.navbar-brand:visited:hover{color:var(--pst-color-link-hover);color:var(--pst-color-text-base);text-decoration:underline;text-decoration-skip-ink:none;text-decoration-thickness:max(1px,.0625rem);text-decoration-thickness:max(3px,.1875rem,.12em);text-underline-offset:.1578em}ul.navbar-nav ul{padding:0 0 0 1rem}ul.navbar-nav li{display:flex;flex-direction:column}ul.navbar-nav li a{align-items:center;color:var(--pst-color-text-muted);display:flex;height:100%;padding-bottom:.25rem;padding-top:.25rem;text-decoration:none}ul.navbar-nav li a:hover{color:var(--pst-color-link-hover);text-decoration:underline;text-decoration-skip-ink:none;text-decoration-thickness:max(1px,.0625rem);text-decoration-thickness:max(3px,.1875rem,.12em);text-underline-offset:.1578em}.page-toc .section-nav{border-bottom:none;padding-left:0}.page-toc .section-nav ul{padding-left:1rem}.page-toc .nav-link{font-size:var(--pst-sidebar-font-size-mobile)}@media (min-width:1200px){.page-toc .nav-link{font-size:var(--pst-sidebar-font-size)}}.page-toc .onthispage{color:var(--pst-color-text-base);font-weight:var(--pst-sidebar-header-font-weight);margin-bottom:.5rem}.prev-next-area{width:100%}.prev-next-area p{color:var(--pst-color-text-muted);line-height:1.3em;margin:0 .3em}.prev-next-area i{font-size:1.2em}.prev-next-area a{align-items:center;border:none;color:var(--pst-color-text-muted);display:flex;max-width:45%;overflow-x:hidden;padding:10px;text-decoration:none}.prev-next-area a p.prev-next-title{word-wrap:break-word;color:var(--pst-color-link);font-size:1.1em;font-weight:var(--pst-admonition-font-weight-heading);text-decoration:underline;text-decoration-thickness:max(1px,.0625rem);text-underline-offset:.1578em}.prev-next-area a p.prev-next-title:hover{color:var(--pst-color-link-hover);text-decoration-skip-ink:none;text-decoration-thickness:max(3px,.1875rem,.12em)}.prev-next-area a p.prev-next-title:active,.prev-next-area a p.prev-next-title:visited{color:var(--pst-color-link)}.prev-next-area a p.prev-next-title:visited:hover{color:var(--pst-color-link-hover)}.prev-next-area a:hover p.prev-next-title,.prev-next-area a:visited:hover p.prev-next-title{color:var(--pst-color-link-hover);text-decoration:underline;text-decoration-skip-ink:none;text-decoration-thickness:max(1px,.0625rem);text-decoration-thickness:max(3px,.1875rem,.12em);text-underline-offset:.1578em}.prev-next-area a .prev-next-info{flex-direction:column;margin:0 .5em}.prev-next-area a .prev-next-info .prev-next-subtitle{text-transform:capitalize}.prev-next-area a.left-prev{float:left}.prev-next-area a.right-next{float:right}.prev-next-area a.right-next div.prev-next-info{text-align:right}.bd-search{border:1px solid var(--pst-color-border);border-radius:.25rem;color:var(--pst-color-text-base);gap:.5rem;padding-left:.5rem;position:relative}.bd-search,.bd-search:active{background-color:var(--pst-color-background)}.bd-search:active{color:var(--pst-color-text-muted)}.bd-search:focus-within{box-shadow:0 0 0 .1875rem var(--pst-color-accent)}.bd-search:focus-within input:focus{box-shadow:none}.bd-search .icon{color:var(--pst-color-border);left:25px;position:absolute}.bd-search .fa-solid.fa-magnifying-glass{color:var(--pst-color-text-muted);left:calc(1.25rem - .35em);position:absolute}.bd-search input.form-control{background-color:var(--pst-color-background);border:none;color:var(--pst-color-text-base)}.bd-search input.form-control::placeholder{color:var(--pst-color-text-muted)}.bd-search input.form-control::-webkit-search-cancel-button,.bd-search input.form-control::-webkit-search-decoration{appearance:none}.bd-search input.form-control:focus,.bd-search input.form-control:focus-visible{color:var(--pst-color-text-muted)}.bd-search .search-button__kbd-shortcut{color:var(--pst-color-border);display:none;margin-inline-end:.5rem}@media (min-width:720px){.bd-search .search-button__kbd-shortcut{display:flex}}.search-button i{font-size:1.3rem}#pst-search-dialog{display:none}#pst-search-dialog[open]{background-color:transparent;border:none;display:flex;left:50%;margin-top:.5rem;max-width:800px;padding:.1875rem;position:fixed;right:1rem;top:30%;transform:translate(-50%,-50%);width:90%;z-index:1055}#pst-search-dialog[open]::backdrop{background-color:#000;opacity:.5}#pst-search-dialog[open] form.bd-search{flex-grow:1}#pst-search-dialog[open] form.bd-search input,#pst-search-dialog[open] form.bd-search svg{font-size:var(--pst-font-size-icon)}.search-button-field{align-items:center;background-color:var(--pst-color-surface);border:1px solid var(--pst-color-border);border-radius:1.5em;color:var(--pst-color-text-muted);display:inline-flex;padding:.5em}.search-button-field:hover{box-shadow:0 0 0 .1875rem var(--pst-color-link-hover)}.search-button-field:focus-visible{border-radius:1.5em}.search-button-field .search-button__default-text{font-size:var(--bs-nav-link-font-size);font-weight:var(--bs-nav-link-font-weight);margin-left:.5em;margin-right:.5em}.search-button-field .kbd-shortcut__modifier{font-size:.75em}.search-button-field>*{align-items:center}.search-button-field>:not(svg){display:none}@media (min-width:960px){.search-button-field>:not(svg){display:flex}}div#searchbox p.highlight-link{box-shadow:0 .2rem .5rem var(--pst-color-shadow),0 0 .0625rem var(--pst-color-shadow)!important;margin:1rem 0;width:fit-content}@media (min-width:1200px){div#searchbox p.highlight-link{margin-left:2rem}}div#searchbox p.highlight-link a{background-color:var(--pst-color-primary);border-radius:.25rem;color:var(--pst-color-primary-text);font-size:1.25rem;padding:.75rem;text-decoration:none;transition:box-shadow .25s ease-out}div#searchbox p.highlight-link a:hover{box-shadow:inset 0 0 50px 50px rgba(0,0,0,.25)}div#searchbox p.highlight-link a:before{color:unset;content:var(--pst-icon-search-minus);font:var(--fa-font-solid);margin-right:.5rem}.theme-switch-button .theme-switch{display:none}.theme-switch-button .theme-switch:active{color:var(--pst-color-link-hover);text-decoration:none}.theme-switch-button .theme-switch .fa-lg{aspect-ratio:1/1}html[data-mode=auto] .theme-switch-button .theme-switch[data-mode=auto],html[data-mode=dark] .theme-switch-button .theme-switch[data-mode=dark],html[data-mode=light] .theme-switch-button .theme-switch[data-mode=light]{display:inline}button.btn.version-switcher__button{border-color:var(--pst-color-border);color:var(--pst-color-text-base);margin-bottom:1em}@media (min-width:960px){button.btn.version-switcher__button{margin-bottom:unset}}button.btn.version-switcher__button:hover{border-color:transparent;box-shadow:0 0 0 .1875rem var(--pst-color-secondary)}button.btn.version-switcher__button:active{border-color:var(--pst-color-border);color:var(--pst-color-text-base)}button.btn.version-switcher__button:focus-visible{border-color:transparent}.version-switcher__menu{border-color:var(--pst-color-border);border-radius:var(--bs-dropdown-border-radius)}.version-switcher__menu a.list-group-item{background-color:var(--pst-color-on-background);color:var(--pst-color-text-base);padding:.75rem 1.25rem}.version-switcher__menu a.list-group-item:not(:last-child){border-bottom:1px solid var(--pst-color-border)}.version-switcher__menu a.list-group-item:hover{background-color:var(--pst-color-surface);color:var(--pst-color-link-hover);text-decoration:underline;text-decoration-skip-ink:none;text-decoration-thickness:max(1px,.0625rem);text-decoration-thickness:max(3px,.1875rem,.12em);text-underline-offset:.1578em}.version-switcher__menu a.list-group-item.active{box-shadow:inset max(3px,.1875rem,.12em) 0 0 var(--pst-color-primary);color:var(--pst-color-primary);font-weight:600;position:relative;z-index:1}.version-switcher__menu a.list-group-item.active:focus-visible{box-shadow:inset max(3px,.1875rem,.12em) 0 0 var(--pst-color-primary),0 0 0 .1875rem var(--pst-color-accent);outline:none}.version-switcher__menu a.list-group-item.active span:before{content:"";height:100%;left:0;position:absolute;top:0;width:100%;z-index:-1}.version-switcher__menu a.list-group-item:focus-visible{z-index:10}.version-switcher__menu,button.version-switcher__button{box-sizing:border-box;font-size:1.1em;min-height:24px;min-width:24px;z-index:1055}@media (min-width:960px){.version-switcher__menu,button.version-switcher__button{font-size:unset}}nav.page-toc{margin-bottom:1rem}.bd-toc .nav .nav{display:none}.bd-toc .nav .nav.visible,.bd-toc .nav>.active>ul,.toc-entry{display:block}.toc-entry a.nav-link,.toc-entry a>code{color:var(--pst-color-text-muted)}.toc-entry a.nav-link{display:block;margin-left:-1rem;padding:.25rem 0 .25rem 1rem;text-decoration:none}.toc-entry a.nav-link:hover{background-color:transparent;text-decoration:underline;text-decoration-skip-ink:none;text-decoration-thickness:max(3px,.1875rem,.12em)}.toc-entry a.nav-link:active,.toc-entry a.nav-link:hover{color:var(--pst-color-link-hover)}.toc-entry a.nav-link:focus-visible{box-shadow:0 0 0 .1875rem var(--pst-color-accent);outline:none;z-index:10}.toc-entry a.nav-link.active{background-color:transparent;box-shadow:inset max(3px,.1875rem,.12em) 0 0 var(--pst-color-primary);color:var(--pst-color-primary);font-weight:600}.toc-entry a.nav-link.active:focus-visible{box-shadow:inset max(3px,.1875rem,.12em) 0 0 var(--pst-color-primary),0 0 0 .1875rem var(--pst-color-accent);outline:none}.toc-entry a.nav-link.active:hover{color:var(--pst-color-link-hover)}.toc-entry a.nav-link:focus-visible{border-radius:.125rem}div.deprecated,div.versionadded,div.versionchanged{background-color:var(--pst-color-on-background);border-left:.2rem solid;border-color:var(--pst-color-info);border-radius:.25rem;box-shadow:0 .2rem .5rem var(--pst-color-shadow),0 0 .0625rem var(--pst-color-shadow)!important;break-inside:avoid;margin:1.5625em auto;overflow:hidden;padding:0 .6rem;page-break-inside:avoid;position:relative;vertical-align:middle}div.deprecated>p,div.versionadded>p,div.versionchanged>p{margin-bottom:.6rem;margin-top:.6rem}div.versionadded{background-color:var(--pst-color-success-bg);border-color:var(--pst-color-success)}div.versionchanged{background-color:var(--pst-color-warning-bg);border-color:var(--pst-color-warning)}div.deprecated{background-color:var(--pst-color-danger-bg);border-color:var(--pst-color-danger)}span.versionmodified{font-weight:600}span.versionmodified:before{color:var(--pst-color-info);content:var(--pst-icon-versionmodified-default);font:var(--fa-font-solid);margin-right:.6rem}span.versionmodified.added:before{color:var(--pst-color-success);content:var(--pst-icon-versionmodified-added)}span.versionmodified.changed:before{color:var(--pst-color-warning);content:var(--pst-icon-versionmodified-changed)}span.versionmodified.deprecated:before{color:var(--pst-color-danger);content:var(--pst-icon-versionmodified-deprecated)}.sidebar-indices-items{border-top:1px solid var(--pst-color-border);display:flex;flex-direction:column}@media (min-width:960px){.sidebar-indices-items{border-top:none}}.sidebar-indices-items .sidebar-indices-items__title{color:var(--pst-color-text-base);font-size:var(--pst-sidebar-header-font-size);font-weight:var(--pst-sidebar-header-font-weight);margin-bottom:.5rem}.sidebar-indices-items ul.indices-link{list-style:none;margin-right:-1rem;padding:0}.sidebar-indices-items ul.indices-link li>a{color:var(--pst-color-text-muted);display:block;padding:.25rem 0}.sidebar-indices-items ul.indices-link li>a:hover{background-color:transparent;color:var(--pst-color-primary);text-decoration:none}.admonition,div.admonition{background-color:var(--pst-color-on-background);border-left:.2rem solid;border-color:var(--pst-color-info);border-radius:.25rem;box-shadow:0 .2rem .5rem var(--pst-color-shadow),0 0 .0625rem var(--pst-color-shadow)!important;break-inside:avoid;margin:1.5625em auto;overflow:hidden;padding:0 .6rem .8rem;page-break-inside:avoid}.admonition :last-child,div.admonition :last-child{margin-bottom:0}.admonition p.admonition-title~*,div.admonition p.admonition-title~*{margin-left:1.4rem;margin-right:1.4rem}.admonition>ol,.admonition>ul,div.admonition>ol,div.admonition>ul{margin-left:1em}.admonition>.admonition-title,div.admonition>.admonition-title{background-color:var(--pst-color-info-bg);font-weight:var(--pst-admonition-font-weight-heading);margin:0 -.6rem;padding:.4rem .6rem .4rem 2rem;position:relative;z-index:1}.admonition>.admonition-title:before,div.admonition>.admonition-title:before{content:"";height:100%;left:0;pointer-events:none;position:absolute;top:0;width:100%;z-index:-1}.admonition>.admonition-title:after,div.admonition>.admonition-title:after{color:var(--pst-color-info);content:var(--pst-icon-admonition-default);font:var(--fa-font-solid);height:1rem;left:.5rem;line-height:inherit;opacity:1;position:absolute;width:1rem}.admonition>.admonition-title+*,div.admonition>.admonition-title+*{margin-top:.4em}.admonition.attention,div.admonition.attention{border-color:var(--pst-color-attention)}.admonition.attention>.admonition-title,div.admonition.attention>.admonition-title{background-color:var(--pst-color-attention-bg)}.admonition.attention>.admonition-title:after,div.admonition.attention>.admonition-title:after{color:var(--pst-color-attention);content:var(--pst-icon-admonition-attention)}.admonition.caution,div.admonition.caution{border-color:var(--pst-color-warning)}.admonition.caution>.admonition-title,div.admonition.caution>.admonition-title{background-color:var(--pst-color-warning-bg)}.admonition.caution>.admonition-title:after,div.admonition.caution>.admonition-title:after{color:var(--pst-color-warning);content:var(--pst-icon-admonition-caution)}.admonition.warning,div.admonition.warning{border-color:var(--pst-color-warning)}.admonition.warning>.admonition-title,div.admonition.warning>.admonition-title{background-color:var(--pst-color-warning-bg)}.admonition.warning>.admonition-title:after,div.admonition.warning>.admonition-title:after{color:var(--pst-color-warning);content:var(--pst-icon-admonition-warning)}.admonition.danger,div.admonition.danger{border-color:var(--pst-color-danger)}.admonition.danger>.admonition-title,div.admonition.danger>.admonition-title{background-color:var(--pst-color-danger-bg)}.admonition.danger>.admonition-title:after,div.admonition.danger>.admonition-title:after{color:var(--pst-color-danger);content:var(--pst-icon-admonition-danger)}.admonition.error,div.admonition.error{border-color:var(--pst-color-danger)}.admonition.error>.admonition-title,div.admonition.error>.admonition-title{background-color:var(--pst-color-danger-bg)}.admonition.error>.admonition-title:after,div.admonition.error>.admonition-title:after{color:var(--pst-color-danger);content:var(--pst-icon-admonition-error)}.admonition.hint,div.admonition.hint{border-color:var(--pst-color-success)}.admonition.hint>.admonition-title,div.admonition.hint>.admonition-title{background-color:var(--pst-color-success-bg)}.admonition.hint>.admonition-title:after,div.admonition.hint>.admonition-title:after{color:var(--pst-color-success);content:var(--pst-icon-admonition-hint)}.admonition.tip,div.admonition.tip{border-color:var(--pst-color-success)}.admonition.tip>.admonition-title,div.admonition.tip>.admonition-title{background-color:var(--pst-color-success-bg)}.admonition.tip>.admonition-title:after,div.admonition.tip>.admonition-title:after{color:var(--pst-color-success);content:var(--pst-icon-admonition-tip)}.admonition.important,div.admonition.important{border-color:var(--pst-color-attention)}.admonition.important>.admonition-title,div.admonition.important>.admonition-title{background-color:var(--pst-color-attention-bg)}.admonition.important>.admonition-title:after,div.admonition.important>.admonition-title:after{color:var(--pst-color-attention);content:var(--pst-icon-admonition-important)}.admonition.note,div.admonition.note{border-color:var(--pst-color-info)}.admonition.note>.admonition-title,div.admonition.note>.admonition-title{background-color:var(--pst-color-info-bg)}.admonition.note>.admonition-title:after,div.admonition.note>.admonition-title:after{color:var(--pst-color-info);content:var(--pst-icon-admonition-note)}.admonition.seealso,div.admonition.seealso{border-color:var(--pst-color-success)}.admonition.seealso>.admonition-title,div.admonition.seealso>.admonition-title{background-color:var(--pst-color-success-bg)}.admonition.seealso>.admonition-title:after,div.admonition.seealso>.admonition-title:after{color:var(--pst-color-success);content:var(--pst-icon-admonition-seealso)}.admonition.admonition-todo,div.admonition.admonition-todo{border-color:var(--pst-color-secondary)}.admonition.admonition-todo>.admonition-title,div.admonition.admonition-todo>.admonition-title{background-color:var(--pst-color-secondary-bg)}.admonition.admonition-todo>.admonition-title:after,div.admonition.admonition-todo>.admonition-title:after{color:var(--pst-color-secondary);content:var(--pst-icon-admonition-todo)}.admonition.sidebar,div.admonition.sidebar{border-width:0 0 0 .2rem;clear:both;float:right;margin-left:.5rem;margin-top:0;max-width:40%}.admonition.sidebar.attention,.admonition.sidebar.important,div.admonition.sidebar.attention,div.admonition.sidebar.important{border-color:var(--pst-color-attention)}.admonition.sidebar.caution,.admonition.sidebar.warning,div.admonition.sidebar.caution,div.admonition.sidebar.warning{border-color:var(--pst-color-warning)}.admonition.sidebar.danger,.admonition.sidebar.error,div.admonition.sidebar.danger,div.admonition.sidebar.error{border-color:var(--pst-color-danger)}.admonition.sidebar.hint,.admonition.sidebar.seealso,.admonition.sidebar.tip,div.admonition.sidebar.hint,div.admonition.sidebar.seealso,div.admonition.sidebar.tip{border-color:var(--pst-color-success)}.admonition.sidebar.note,.admonition.sidebar.todo,div.admonition.sidebar.note,div.admonition.sidebar.todo{border-color:var(--pst-color-info)}.admonition.sidebar p.admonition-title~*,div.admonition.sidebar p.admonition-title~*{margin-left:0;margin-right:0}aside.topic,div.topic,div.topic.contents,nav.contents{background-color:var(--pst-color-surface);border-color:var(--pst-color-border);border-radius:.25rem;box-shadow:0 .2rem .5rem var(--pst-color-shadow),0 0 .0625rem var(--pst-color-shadow)!important;display:flex;flex-direction:column;padding:1rem 1.25rem}aside.topic .topic-title,div.topic .topic-title,div.topic.contents .topic-title,nav.contents .topic-title{margin:0 0 .5rem}aside.topic p,div.topic p,div.topic.contents p,nav.contents p{color:var(--pst-color-on-surface)!important}aside.topic ul.simple,div.topic ul.simple,div.topic.contents ul.simple,nav.contents ul.simple{padding-left:1rem}aside.topic ul.simple ul,div.topic ul.simple ul,div.topic.contents ul.simple ul,nav.contents ul.simple ul{padding-left:2em}aside.sidebar{background-color:var(--pst-color-surface);border:1px solid var(--pst-color-border);border-radius:.25rem;margin-left:.5rem;padding:0}aside.sidebar>:last-child{padding-bottom:1rem}aside.sidebar p.sidebar-title{border-bottom:1px solid var(--pst-color-border);font-family:var(--pst-font-family-heading);font-weight:var(--pst-admonition-font-weight-heading);margin-bottom:0;padding-bottom:.5rem;padding-top:.5rem;position:relative}aside.sidebar>:not(.sidebar-title):first-child,aside.sidebar>p.sidebar-title+*{margin-top:1rem}aside.sidebar>*{padding-left:1rem;padding-right:1rem}p.rubric{display:flex;flex-direction:column}.seealso dd{margin-bottom:0;margin-top:0}table.field-list{border-collapse:separate;border-spacing:10px;margin-left:1px}table.field-list th.field-name{background-color:var(--pst-color-surface);padding:1px 8px 1px 5px;white-space:nowrap}table.field-list td.field-body p{font-style:italic}table.field-list td.field-body p>strong{font-style:normal}table.field-list td.field-body blockquote{border-left:none;margin:0 0 .3em;padding-left:30px}.table.autosummary td:first-child{white-space:nowrap}.sig{font-family:var(--pst-font-family-monospace)}.sig-inline.c-texpr,.sig-inline.cpp-texpr{font-family:unset}.sig.c .k,.sig.c .kt,.sig.c .m,.sig.c .s,.sig.c .sc,.sig.cpp .k,.sig.cpp .kt,.sig.cpp .m,.sig.cpp .s,.sig.cpp .sc{color:var(--pst-color-text-base)}.sig-name{color:var(--pst-color-inline-code)}.sig-param .default_value,.sig-param .o{color:var(--pst-color-text-muted);font-weight:400}dt:target,span.highlighted{background-color:var(--pst-color-target)}.viewcode-back{font-family:var(--pst-font-family-base)}.viewcode-block:target{background-color:var(--pst-color-target);border-bottom:1px solid var(--pst-color-border);border-top:1px solid var(--pst-color-border);position:relative}dl>dt>a:has(.viewcode-link){float:right}dl>dt>a:has(.viewcode-link) .viewcode-link{float:none}dl[class]:not(.option-list,.field-list,.footnote,.glossary,.simple){margin-bottom:3rem}dl[class]:not(.option-list,.field-list,.footnote,.glossary,.simple) dd{margin-left:2rem}dl[class]:not(.option-list,.field-list,.footnote,.glossary,.simple) dd>dl.simple>dt{display:flex}dl[class]:not(.option-list,.field-list,.footnote,.glossary,.simple) dl.field-list{display:grid;grid-template-columns:unset}dl[class]:not(.option-list,.field-list,.footnote,.glossary,.simple) dt.field-even,dl[class]:not(.option-list,.field-list,.footnote,.glossary,.simple) dt.field-odd{background-color:var(--pst-color-surface);margin-bottom:.1rem;margin-top:.2rem}div.highlight,div.literal-block-wrapper,div[class*=highlight-]{border-radius:.25rem;break-inside:avoid;display:flex;flex-direction:column;width:unset}div.literal-block-wrapper{border:1px solid var(--pst-color-border);border-radius:.25rem}div.literal-block-wrapper div.code-block-caption{border-bottom:1px solid var(--pst-color-border);font-size:1rem;font-weight:var(--pst-font-weight-caption);margin:0;padding:.5rem}div.literal-block-wrapper div.code-block-caption a.headerlink{font-size:inherit}div.literal-block-wrapper div[class*=highlight-]{border-radius:0;margin:0}div.literal-block-wrapper div[class*=highlight-] pre{border:none;box-shadow:none}code.literal{background-color:var(--pst-color-surface);border:1px solid var(--pst-color-border);border-radius:.25rem;padding:.1rem .25rem}a>code{color:var(--pst-color-inline-code-links)}a>code:hover{color:var(--pst-color-link-hover)}span.linenos{opacity:.8!important}figure>a,figure>a>img,figure>img,figure>video{display:block;margin-left:auto;margin-right:auto}figure a.headerlink{font-size:inherit;position:absolute}figure:hover a.headerlink{visibility:visible}figure figcaption{color:var(--pst-color-text-muted);font-family:var(--pst-font-family-heading);font-weight:var(--pst-font-weight-caption);margin-left:auto;margin-right:auto;margin-top:.3rem;text-align:center}figure figcaption>p:last-child{margin-bottom:0}figure figcaption p{display:inline-block;text-align:start}figure figcaption table.table{margin-left:auto;margin-right:auto;width:fit-content}dt.label>span.brackets:not(:only-child):before{content:"["}dt.label>span.brackets:not(:only-child):after{content:"]"}a.footnote-reference{font-size:small;vertical-align:super}aside.footnote{margin-bottom:.5rem}aside.footnote:last-child{margin-bottom:1rem}aside.footnote span.backrefs,aside.footnote span.label{font-weight:700}aside.footnote:target{background-color:var(--pst-color-target)}div.doctest>div.highlight span.gp,span.linenos,table.highlighttable td.linenos{user-select:none}dd{margin-bottom:10px;margin-left:30px;margin-top:3px}ol,ul{padding-inline-start:2rem}ol li>p:first-child,ul li>p:first-child{margin-bottom:.25rem;margin-top:.25rem}blockquote{background-color:var(--pst-color-surface);border-left:.25em solid var(--pst-color-blockquote-notch);border-radius:.25rem;color:var(--pst-color-text-muted);padding:1em;position:relative}blockquote p{color:var(--pst-color-text-base)}blockquote .line-block{margin:0}blockquote p:last-child{margin-bottom:0}blockquote:before{content:"";height:100%;left:0;pointer-events:none;position:absolute;top:0;width:100%}blockquote a{color:var(--pst-color-inline-code-links)}blockquote:before{z-index:-1}span.guilabel{background-color:var(--pst-color-info-bg);border:1px solid var(--pst-color-info);border-radius:4px;font-size:80%;font-weight:700;margin:auto 2px;padding:2.4px 6px;position:relative}span.guilabel:before{content:"";height:100%;left:0;pointer-events:none;position:absolute;top:0;width:100%;z-index:-1}a.reference.download:before{color:var(--pst-color-text-muted);content:var(--pst-icon-download);font:var(--fa-font-solid);font-size:.8em;padding:0 .25em}table{margin-left:auto;margin-right:auto}table.table-right{margin-right:0}table.table-left{margin-left:0}table caption{caption-side:top;color:var(--pst-color-text-muted);text-align:center}td.text-left,th.text-left{text-align:left}td.text-right,th.text-right{text-align:right}td.text-center,th.text-center{text-align:center}.table{--bs-table-bg:transparent;--bs-table-color:var(
+    --pst-color-text-base
+  );border:1px solid var(--pst-color-table-outer-border);color:var(--pst-color-table)}.table td~td,.table td~th,.table th~td,.table th~th{border-left:1px solid var(--pst-color-table-inner-border)}.table thead tr{background-color:var(--pst-color-table-heading-bg);border-bottom:2px solid var(--pst-color-primary)}.table tbody tr:nth-child(odd){background-color:var(--pst-color-table-row-zebra-low-bg)}.table tbody tr:nth-child(2n){background-color:var(--pst-color-table-row-zebra-high-bg)}.pst-scrollable-table-container{overflow-x:auto}.toctree-wrapper p.caption{font-size:1.5em;margin-bottom:0}.toctree-wrapper>ul{padding-left:0}.toctree-wrapper li[class^=toctree-l]{list-style:none;margin-bottom:.2em}.toctree-wrapper li[class^=toctree-l]>a{font-size:1.1em;list-style:none}.toctree-wrapper li[class^=toctree-l]>ul{list-style:none;padding-inline-start:1.5em}.toctree-wrapper .toctree-l1>a{font-size:1.3em}div.topic.contents ul.simple,nav.contents ul.simple{list-style:none;padding-left:0}div.math,span.math{align-items:center;display:flex;max-width:100%;overflow:hidden}span.math{display:inline-flex}div.math{flex-direction:row-reverse;gap:.5em}div.math span.eqno a.headerlink{font-size:1em;position:relative}div.math mjx-container{flex-grow:1;overflow:auto;padding-bottom:.2rem}div.math mjx-container mjx-assistive-mml{height:0}.ablog-sidebar-item .h2,.ablog-sidebar-item .h3,.ablog-sidebar-item h2,.ablog-sidebar-item h3{font-size:var(--pst-sidebar-header-font-size);margin-top:.5rem}.ablog-sidebar-item .h2 a,.ablog-sidebar-item .h3 a,.ablog-sidebar-item h2 a,.ablog-sidebar-item h3 a{color:var(--pst-color-text-base)}.ablog-sidebar-item ul{display:flex;flex-direction:column;gap:.5em;list-style:none;margin-bottom:0;overflow-y:hidden;padding-left:0}.ablog-sidebar-item ul.ablog-cloud{flex-flow:row wrap;gap:.5rem}.ablog-sidebar-item ul.ablog-cloud li{align-items:center;display:flex}.ablog__prev-next{display:flex;font-size:1.2em;padding:1rem 0}.ablog__prev-next>span{display:flex;max-width:45%}.ablog__prev-next>span a{align-items:center;display:flex;gap:1rem;line-height:1.5rem;margin-left:auto}.ablog__prev-next>span a i:before{color:var(--pst-color-text-base)}.ablog__prev-next span.ablog__prev i.fa-arrow-circle-left:before{content:var(--pst-icon-angle-left)}.ablog__prev-next span.ablog__spacer{display:none}.ablog__prev-next span.ablog__next{margin-left:auto;text-align:right}.ablog__prev-next span.ablog__next i.fa-arrow-circle-right:before{content:var(--pst-icon-angle-right)}.ablog__collection,.postlist{padding-left:0}.ablog__collection .ablog-post,.postlist .ablog-post{list-style:none}.ablog__collection .ablog-post .ablog-archive,.postlist .ablog-post .ablog-archive{display:flex;flex-flow:row wrap;font-size:.75rem;gap:1rem;list-style:none;padding-left:0}.ablog__collection .ablog-post .ablog-post-title,.postlist .ablog-post .ablog-post-title{font-size:1.25rem;margin-top:0}.ablog__collection .ablog-post .ablog-post-title a,.postlist .ablog-post .ablog-post-title a{font-weight:700}.ablog__collection .ablog-post .ablog-post-expand,.postlist .ablog-post .ablog-post-expand{margin-bottom:.5rem}.docutils.container{margin-left:unset;margin-right:unset;max-width:unset;padding-left:unset;padding-right:unset;width:unset}.btn{--bs-btn-focus-box-shadow:0 0 0 0.1875rem var(--pst-color-accent)}div.highlight button.copybtn{align-items:center;background-color:var(--pst-color-surface);border:none;display:flex;justify-content:center}div.highlight button.copybtn:not(.success){color:var(--pst-color-muted)}div.highlight button.copybtn:hover:not(.success){background-color:var(--pst-color-shadow);color:var(--pst-color-text)}div.highlight button.copybtn.o-tooltip--left:after{background-color:var(--pst-color-surface);color:var(--pst-color-text)}div.highlight button.copybtn:focus{opacity:1}div.highlight button.copybtn:focus-visible{outline:var(--pst-color-accent) solid .1875rem}div.highlight:has(button.copybtn){min-height:2em}#ethical-ad-placement .ethical-footer a,#ethical-ad-placement .ethical-footer a:active,#ethical-ad-placement .ethical-footer a:hover,#ethical-ad-placement .ethical-footer a:visited,#ethical-ad-placement .ethical-sidebar a,#ethical-ad-placement .ethical-sidebar a:active,#ethical-ad-placement .ethical-sidebar a:hover,#ethical-ad-placement .ethical-sidebar a:visited{color:var(--pst-color-text-base)}#ethical-ad-placement .ethical-footer,#ethical-ad-placement .ethical-sidebar{background-color:var(--pst-color-background);border:1px solid var(--pst-color-border);border-radius:5px;color:var(--pst-color-text-base);font-size:14px;line-height:20px}.bd-content div.jupyter_container{background-color:unset;border:none;box-shadow:none}.bd-content div.jupyter_container div.highlight,.bd-content div.jupyter_container div.output{border-radius:.25rem}.bd-content div.jupyter_container div.highlight{background-color:var(--pst-color-surface)}.bd-content div.jupyter_container .cell_input,.bd-content div.jupyter_container .cell_output{border-radius:.25rem}.bd-content div.jupyter_container .cell_input pre,.bd-content div.jupyter_container .cell_output pre{padding:1rem}html[data-theme=dark] div.graphviz>object.inheritance{color-scheme:normal;filter:brightness(.8) invert(.82) contrast(1.2)}.xr-wrap[hidden]{display:block!important}.jp-OutputArea-output.lm-Widget{overflow:auto}:root{--pst-teal-50:#f4fbfc;--pst-teal-100:#e9f6f8;--pst-teal-200:#d0ecf1;--pst-teal-300:#abdde6;--pst-teal-400:#3fb1c5;--pst-teal-500:#0a7d91;--pst-teal-600:#085d6c;--pst-teal-700:#064752;--pst-teal-800:#042c33;--pst-teal-900:#021b1f;--pst-violet-50:#f4eefb;--pst-violet-100:#e0c7ff;--pst-violet-200:#d5b4fd;--pst-violet-300:#b780ff;--pst-violet-400:#9c5ffd;--pst-violet-500:#8045e5;--pst-violet-600:#6432bd;--pst-violet-700:#4b258f;--pst-violet-800:#341a61;--pst-violet-900:#1e0e39;--pst-gray-50:#f9f9fa;--pst-gray-100:#f3f4f5;--pst-gray-200:#e5e7ea;--pst-gray-300:#d1d5da;--pst-gray-400:#9ca4af;--pst-gray-500:#677384;--pst-gray-600:#48566b;--pst-gray-700:#29313d;--pst-gray-800:#222832;--pst-gray-900:#14181e;--pst-pink-50:#fcf8fd;--pst-pink-100:#fcf0fa;--pst-pink-200:#f8dff5;--pst-pink-300:#f3c7ee;--pst-pink-400:#e47fd7;--pst-pink-500:#c132af;--pst-pink-600:#912583;--pst-pink-700:#6e1c64;--pst-pink-800:#46123f;--pst-pink-900:#2b0b27;--pst-foundation-white:#fff;--pst-foundation-black:#14181e}html:not([data-theme]){--pst-color-primary:#0a7d91;--pst-color-primary-bg:#d0ecf1;--pst-color-secondary:#8045e5;--pst-color-secondary-bg:#e0c7ff;--pst-color-accent:#c132af;--pst-color-accent-bg:#f8dff5;--pst-color-info:#276be9;--pst-color-info-bg:#dce7fc;--pst-color-warning:#f66a0a;--pst-color-warning-bg:#f8e3d0;--pst-color-success:#00843f;--pst-color-success-bg:#d6ece1;--pst-color-attention:var(--pst-color-warning);--pst-color-attention-bg:var(--pst-color-warning-bg);--pst-color-danger:#d72d47;--pst-color-danger-bg:#f9e1e4;--pst-color-text-base:#222832;--pst-color-text-muted:#48566b;--pst-color-shadow:rgba(0,0,0,.1);--pst-color-border:#d1d5da;--pst-color-border-muted:rgba(23,23,26,.2);--pst-color-blockquote-notch:#677384;--pst-color-inline-code:#912583;--pst-color-inline-code-links:#085d6c;--pst-color-target:#f3cf95;--pst-color-table:#14181e;--pst-color-table-row-hover-bg:#d5b4fd;--pst-color-table-inner-border:#e5e7ea;--pst-color-background:#fff;--pst-color-on-background:#fff;--pst-color-surface:#f3f4f5;--pst-color-on-surface:#222832;--pst-color-heading:var(--pst-color-text-base);--pst-color-link:var(--pst-color-primary);--pst-color-link-hover:var(--pst-color-secondary);--pst-color-table-outer-border:var(--pst-color-surface);--pst-color-table-heading-bg:var(--pst-color-surface);--pst-color-table-row-zebra-high-bg:var(--pst-color-on-background);--pst-color-table-row-zebra-low-bg:var(--pst-color-surface)}html:not([data-theme]) .only-dark,html:not([data-theme]) .only-dark~figcaption{display:none!important}html[data-theme=light]{--pst-color-primary-bg:#d0ecf1;--pst-color-secondary-bg:#e0c7ff;--pst-color-accent:#c132af;--pst-color-accent-bg:#f8dff5;--pst-color-info-bg:#dce7fc;--pst-color-warning-bg:#f8e3d0;--pst-color-success-bg:#d6ece1;--pst-color-attention:var(--pst-color-warning);--pst-color-attention-bg:var(--pst-color-warning-bg);--pst-color-danger-bg:#f9e1e4;--pst-color-text-base:#222832;--pst-color-text-muted:#48566b;--pst-color-shadow:rgba(0,0,0,.1);--pst-color-border:#d1d5da;--pst-color-border-muted:rgba(23,23,26,.2);--pst-color-blockquote-notch:#677384;--pst-color-inline-code:#912583;--pst-color-inline-code-links:#085d6c;--pst-color-target:#f3cf95;--pst-color-table:#14181e;--pst-color-table-row-hover-bg:#d5b4fd;--pst-color-table-inner-border:#e5e7ea;--pst-color-background:#fff;--pst-color-on-background:#fff;--pst-color-surface:#f3f4f5;--pst-color-on-surface:#222832;--pst-color-heading:var(--pst-color-text-base);--pst-color-link:var(--pst-color-primary);--pst-color-link-hover:var(--pst-color-secondary);--pst-color-table-outer-border:var(--pst-color-surface);--pst-color-table-heading-bg:var(--pst-color-surface);--pst-color-table-row-zebra-high-bg:var(--pst-color-on-background);--pst-color-table-row-zebra-low-bg:var(--pst-color-surface);color-scheme:light}html[data-theme=light] .only-dark,html[data-theme=light] .only-dark~figcaption{display:none!important}html[data-theme=dark]{--pst-color-primary-bg:#042c33;--pst-color-secondary-bg:#341a61;--pst-color-accent:#e47fd7;--pst-color-accent-bg:#46123f;--pst-color-info-bg:#06245d;--pst-color-warning-bg:#652a02;--pst-color-success-bg:#002f17;--pst-color-attention:var(--pst-color-warning);--pst-color-attention-bg:var(--pst-color-warning-bg);--pst-color-danger-bg:#4e111b;--pst-color-text-base:#ced6dd;--pst-color-text-muted:#9ca4af;--pst-color-shadow:rgba(0,0,0,.2);--pst-color-border:#48566b;--pst-color-border-muted:#29313d;--pst-color-blockquote-notch:#9ca4af;--pst-color-inline-code:#f3c7ee;--pst-color-inline-code-links:#3fb1c5;--pst-color-target:#675c04;--pst-color-table:#fff;--pst-color-table-row-hover-bg:#4b258f;--pst-color-table-inner-border:#364150;--pst-color-background:#14181e;--pst-color-on-background:#222832;--pst-color-surface:#29313d;--pst-color-on-surface:#f3f4f5;--pst-color-heading:var(--pst-color-text-base);--pst-color-link:var(--pst-color-primary);--pst-color-link-hover:var(--pst-color-secondary);--pst-color-table-outer-border:var(--pst-color-surface);--pst-color-table-heading-bg:var(--pst-color-surface);--pst-color-table-row-zebra-high-bg:var(--pst-color-on-background);--pst-color-table-row-zebra-low-bg:var(--pst-color-surface);color-scheme:dark}html[data-theme=dark] .only-light,html[data-theme=dark] .only-light~figcaption{display:none!important}html[data-theme=dark] img:not(.only-dark,.dark-light){filter:brightness(.8) contrast(1.2)}html[data-theme=dark] .bd-content img:not(.only-dark,.dark-light){background-color:#fff;border-radius:.25rem}html[data-theme=dark] .MathJax_SVG *{fill:var(--pst-color-text-base)}.pst-color-primary{color:var(--pst-color-primary)}.pst-color-secondary{color:var(--pst-color-secondary)}.pst-color-accent{color:var(--pst-color-accent)}.pst-color-info{color:var(--pst-color-info)}.pst-color-warning{color:var(--pst-color-warning)}.pst-color-success{color:var(--pst-color-success)}.pst-color-attention{color:var(--pst-color-attention)}.pst-color-danger{color:var(--pst-color-danger)}.pst-color-text-base{color:var(--pst-color-text-base)}.pst-color-text-muted{color:var(--pst-color-text-muted)}.pst-color-shadow{color:var(--pst-color-shadow)}.pst-color-border{color:var(--pst-color-border)}.pst-color-border-muted{color:var(--pst-color-border-muted)}.pst-color-blockquote-notch{color:var(--pst-color-blockquote-notch)}.pst-color-inline-code{color:var(--pst-color-inline-code)}.pst-color-inline-code-links{color:var(--pst-color-inline-code-links)}.pst-color-target{color:var(--pst-color-target)}.pst-color-table{color:var(--pst-color-table)}.pst-color-table-row-hover{color:var(--pst-color-table-row-hover)}.pst-color-table-inner-border{color:var(--pst-color-table-inner-border)}.pst-color-background{color:var(--pst-color-background)}.pst-color-on-background{color:var(--pst-color-on-background)}.pst-color-surface{color:var(--pst-color-surface)}.pst-color-on-surface{color:var(--pst-color-on-surface)}html[data-theme=light]{--pst-color-primary:#0a7d91;--pst-color-primary-text:#fff;--pst-color-primary-highlight:#086b7c;--sd-color-primary:var(--pst-color-primary);--sd-color-primary-text:var(--pst-color-primary-text);--sd-color-primary-highlight:var(--pst-color-primary-highlight);--pst-color-primary-highlight-text:#fff;--sd-color-primary-bg:#d0ecf1;--sd-color-primary-bg-text:#14181e;--pst-color-secondary:#8045e5;--pst-color-secondary-text:#fff;--pst-color-secondary-highlight:#651ee0;--sd-color-secondary:var(--pst-color-secondary);--sd-color-secondary-text:var(--pst-color-secondary-text);--sd-color-secondary-highlight:var(--pst-color-secondary-highlight);--pst-color-secondary-highlight-text:#fff;--sd-color-secondary-bg:#e0c7ff;--sd-color-secondary-bg-text:#14181e;--pst-color-success:#00843f;--pst-color-success-text:#fff;--pst-color-success-highlight:#007036;--sd-color-success:var(--pst-color-success);--sd-color-success-text:var(--pst-color-success-text);--sd-color-success-highlight:var(--pst-color-success-highlight);--pst-color-success-highlight-text:#fff;--sd-color-success-bg:#d6ece1;--sd-color-success-bg-text:#14181e;--pst-color-info:#276be9;--pst-color-info-text:#fff;--pst-color-info-highlight:#1457d3;--sd-color-info:var(--pst-color-info);--sd-color-info-text:var(--pst-color-info-text);--sd-color-info-highlight:var(--pst-color-info-highlight);--pst-color-info-highlight-text:#fff;--sd-color-info-bg:#dce7fc;--sd-color-info-bg-text:#14181e;--pst-color-warning:#f66a0a;--pst-color-warning-text:#14181e;--pst-color-warning-highlight:#d25a07;--sd-color-warning:var(--pst-color-warning);--sd-color-warning-text:var(--pst-color-warning-text);--sd-color-warning-highlight:var(--pst-color-warning-highlight);--pst-color-warning-highlight-text:#14181e;--sd-color-warning-bg:#f8e3d0;--sd-color-warning-bg-text:#14181e;--pst-color-danger:#d72d47;--pst-color-danger-text:#fff;--pst-color-danger-highlight:#bb2239;--sd-color-danger:var(--pst-color-danger);--sd-color-danger-text:var(--pst-color-danger-text);--sd-color-danger-highlight:var(--pst-color-danger-highlight);--pst-color-danger-highlight-text:#fff;--sd-color-danger-bg:#f9e1e4;--sd-color-danger-bg-text:#14181e;--pst-color-light:#f3f4f5;--pst-color-light-text:#14181e;--pst-color-light-highlight:#c9cfd6;--sd-color-light:var(--pst-color-light);--sd-color-light-text:var(--pst-color-light-text);--sd-color-light-highlight:var(--pst-color-light-highlight);--pst-color-light-highlight-text:#14181e;--sd-color-light-bg:#f7f7f8;--sd-color-light-bg-text:#14181e;--pst-color-muted:#29313d;--pst-color-muted-text:#fff;--pst-color-muted-highlight:#212936;--sd-color-muted:var(--pst-color-muted);--sd-color-muted-text:var(--pst-color-muted-text);--sd-color-muted-highlight:var(--pst-color-muted-highlight);--pst-color-muted-highlight-text:#fff;--sd-color-muted-bg:#5a6c86;--sd-color-muted-bg-text:#fff;--pst-color-dark:#222832;--pst-color-dark-text:#fff;--pst-color-dark-highlight:#1b222c;--sd-color-dark:var(--pst-color-dark);--sd-color-dark-text:var(--pst-color-dark-text);--sd-color-dark-highlight:var(--pst-color-dark-highlight);--pst-color-dark-highlight-text:#fff;--pst-color-black:#14181e;--pst-color-black-text:#fff;--pst-color-black-highlight:#10141a;--sd-color-black:var(--pst-color-black);--sd-color-black-text:var(--pst-color-black-text);--sd-color-black-highlight:var(--pst-color-black-highlight);--pst-color-black-highlight-text:#fff;--pst-color-white:#fff;--pst-color-white-text:#14181e;--pst-color-white-highlight:#dbd7d7;--sd-color-white:var(--pst-color-white);--sd-color-white-text:var(--pst-color-white-text);--sd-color-white-highlight:var(--pst-color-white-highlight);--pst-color-white-highlight-text:#14181e}html[data-theme=dark]{--pst-color-primary:#3fb1c5;--pst-color-primary-text:#14181e;--pst-color-primary-highlight:#319aac;--sd-color-primary:var(--pst-color-primary);--sd-color-primary-text:var(--pst-color-primary-text);--sd-color-primary-highlight:var(--pst-color-primary-highlight);--pst-color-primary-highlight-text:#14181e;--sd-color-primary-bg:#042c33;--sd-color-primary-bg-text:#fff;--pst-color-secondary:#9c5ffd;--pst-color-secondary-text:#14181e;--pst-color-secondary-highlight:#7c2bfc;--sd-color-secondary:var(--pst-color-secondary);--sd-color-secondary-text:var(--pst-color-secondary-text);--sd-color-secondary-highlight:var(--pst-color-secondary-highlight);--pst-color-secondary-highlight-text:#fff;--sd-color-secondary-bg:#341a61;--sd-color-secondary-bg-text:#fff;--pst-color-success:#5fb488;--pst-color-success-text:#14181e;--pst-color-success-highlight:#47a373;--sd-color-success:var(--pst-color-success);--sd-color-success-text:var(--pst-color-success-text);--sd-color-success-highlight:var(--pst-color-success-highlight);--pst-color-success-highlight-text:#14181e;--sd-color-success-bg:#002f17;--sd-color-success-bg-text:#fff;--pst-color-info:#79a3f2;--pst-color-info-text:#14181e;--pst-color-info-highlight:#4681ee;--sd-color-info:var(--pst-color-info);--sd-color-info-text:var(--pst-color-info-text);--sd-color-info-highlight:var(--pst-color-info-highlight);--pst-color-info-highlight-text:#14181e;--sd-color-info-bg:#06245d;--sd-color-info-bg-text:#fff;--pst-color-warning:#ff9245;--pst-color-warning-text:#14181e;--pst-color-warning-highlight:#ff7614;--sd-color-warning:var(--pst-color-warning);--sd-color-warning-text:var(--pst-color-warning-text);--sd-color-warning-highlight:var(--pst-color-warning-highlight);--pst-color-warning-highlight-text:#14181e;--sd-color-warning-bg:#652a02;--sd-color-warning-bg-text:#fff;--pst-color-danger:#e78894;--pst-color-danger-text:#14181e;--pst-color-danger-highlight:#df596a;--sd-color-danger:var(--pst-color-danger);--sd-color-danger-text:var(--pst-color-danger-text);--sd-color-danger-highlight:var(--pst-color-danger-highlight);--pst-color-danger-highlight-text:#14181e;--sd-color-danger-bg:#4e111b;--sd-color-danger-bg-text:#fff;--pst-color-light:#f3f4f5;--pst-color-light-text:#14181e;--pst-color-light-highlight:#c9cfd6;--sd-color-light:var(--pst-color-light);--sd-color-light-text:var(--pst-color-light-text);--sd-color-light-highlight:var(--pst-color-light-highlight);--pst-color-light-highlight-text:#14181e;--sd-color-light-bg:#a3abb2;--sd-color-light-bg-text:#14181e;--pst-color-muted:#f3f4f5;--pst-color-muted-text:#14181e;--pst-color-muted-highlight:#c9cfd6;--sd-color-muted:var(--pst-color-muted);--sd-color-muted-text:var(--pst-color-muted-text);--sd-color-muted-highlight:var(--pst-color-muted-highlight);--pst-color-muted-highlight-text:#14181e;--sd-color-muted-bg:#1d222b;--sd-color-muted-bg-text:#fff;--pst-color-dark:#222832;--pst-color-dark-text:#fff;--pst-color-dark-highlight:#1b222c;--sd-color-dark:var(--pst-color-dark);--sd-color-dark-text:var(--pst-color-dark-text);--sd-color-dark-highlight:var(--pst-color-dark-highlight);--pst-color-dark-highlight-text:#fff;--pst-color-black:#14181e;--pst-color-black-text:#fff;--pst-color-black-highlight:#10141a;--sd-color-black:var(--pst-color-black);--sd-color-black-text:var(--pst-color-black-text);--sd-color-black-highlight:var(--pst-color-black-highlight);--pst-color-black-highlight-text:#fff;--pst-color-white:#fff;--pst-color-white-text:#14181e;--pst-color-white-highlight:#dbd7d7;--sd-color-white:var(--pst-color-white);--sd-color-white-text:var(--pst-color-white-text);--sd-color-white-highlight:var(--pst-color-white-highlight);--pst-color-white-highlight-text:#14181e}html[data-theme=dark],html[data-theme=light]{--sd-color-card-border:var(--pst-color-border)}html[data-theme=light] .sd-shadow-lg,html[data-theme=light] .sd-shadow-md,html[data-theme=light] .sd-shadow-sm,html[data-theme=light] .sd-shadow-xs{box-shadow:0 .2rem .5rem var(--pst-color-shadow),0 0 .0625rem var(--pst-color-shadow)!important}.bd-content .sd-card{border:1px solid var(--pst-color-border)}.bd-content .sd-card .sd-card-header{background-color:var(--pst-color-on-background);border-bottom:1px solid var(--pst-color-border)}.bd-content .sd-card .sd-card-footer{border-top:1px solid var(--pst-color-border)}.bd-content .sd-card .sd-card-body,.bd-content .sd-card .sd-card-footer{background-color:var(--pst-color-on-background)}.bd-content .sd-card .sd-stretched-link:focus-visible{outline:none}.bd-content .sd-card .sd-stretched-link:focus-visible:after{border-radius:.25rem;outline:var(--pst-color-accent) solid .1875rem}.bd-content .sd-card.sd-card-hover:hover{border-color:var(--pst-color-link-hover)}.bd-content .sd-tab-set>input:checked+label{background-color:var(--pst-color-on-background);border-color:var(--pst-color-primary) var(--pst-color-primary) transparent;border-radius:.125rem .125rem 0 0;border-style:solid solid none;border-width:.125rem .125rem 0;color:var(--pst-color-primary);transform:translateY(.125rem)}.bd-content .sd-tab-set>input:focus-visible+label{background-color:var(--pst-color-accent-bg);border:.125rem solid var(--pst-color-accent);border-radius:.125rem;color:var(--pst-color-on-surface)}.bd-content .sd-tab-set>input:not(:checked,:focus-visible)+label:hover{border-color:transparent;color:var(--pst-color-secondary);text-decoration-line:underline;text-decoration-thickness:max(3px,.1875rem,.12em)}.bd-content .sd-tab-set>label{background-color:var(--pst-color-surface);border:.125rem solid transparent;border-radius:.125rem .125rem 0 0;color:var(--pst-color-on-surface);line-height:1.95;margin-inline-end:.25rem;padding:0 .75em}.bd-content .sd-tab-set .sd-tab-content,html[data-theme=dark] .bd-content .sd-tab-set>label{background-color:var(--pst-color-on-background)}.bd-content .sd-tab-set .sd-tab-content{border:.125rem solid var(--pst-color-primary);border-radius:.1875rem;box-shadow:unset;padding:.625rem}details.sd-dropdown{border:0!important;box-shadow:0 .2rem .5rem var(--pst-color-shadow),0 0 .0625rem var(--pst-color-shadow)!important}details.sd-dropdown summary.sd-card-header{--pst-sd-dropdown-color:var(--pst-gray-500);--pst-sd-dropdown-bg-color:var(--pst-color-surface);align-items:center;background-color:var(--pst-sd-dropdown-bg-color)!important;border:0!important;border-left:.2rem solid var(--pst-sd-dropdown-color)!important;display:flex;font-weight:600;padding-bottom:.5rem;padding-top:.5rem;position:relative}details.sd-dropdown summary.sd-card-header+div.sd-summary-content{--pst-sd-dropdown-color:var(--sd-color-card-border);border:0}details.sd-dropdown summary.sd-card-header.sd-bg-primary,details.sd-dropdown summary.sd-card-header.sd-bg-primary+div.sd-summary-content{--pst-sd-dropdown-color:var(--sd-color-primary);--pst-sd-dropdown-bg-color:var(--sd-color-primary-bg)}details.sd-dropdown summary.sd-card-header.sd-bg-text-primary{color:var(--sd-color-primary-bg-text)!important}details.sd-dropdown summary.sd-card-header.sd-bg-secondary,details.sd-dropdown summary.sd-card-header.sd-bg-secondary+div.sd-summary-content{--pst-sd-dropdown-color:var(--sd-color-secondary);--pst-sd-dropdown-bg-color:var(--sd-color-secondary-bg)}details.sd-dropdown summary.sd-card-header.sd-bg-text-secondary{color:var(--sd-color-secondary-bg-text)!important}details.sd-dropdown summary.sd-card-header.sd-bg-success,details.sd-dropdown summary.sd-card-header.sd-bg-success+div.sd-summary-content{--pst-sd-dropdown-color:var(--sd-color-success);--pst-sd-dropdown-bg-color:var(--sd-color-success-bg)}details.sd-dropdown summary.sd-card-header.sd-bg-text-success{color:var(--sd-color-success-bg-text)!important}details.sd-dropdown summary.sd-card-header.sd-bg-info,details.sd-dropdown summary.sd-card-header.sd-bg-info+div.sd-summary-content{--pst-sd-dropdown-color:var(--sd-color-info);--pst-sd-dropdown-bg-color:var(--sd-color-info-bg)}details.sd-dropdown summary.sd-card-header.sd-bg-text-info{color:var(--sd-color-info-bg-text)!important}details.sd-dropdown summary.sd-card-header.sd-bg-warning,details.sd-dropdown summary.sd-card-header.sd-bg-warning+div.sd-summary-content{--pst-sd-dropdown-color:var(--sd-color-warning);--pst-sd-dropdown-bg-color:var(--sd-color-warning-bg)}details.sd-dropdown summary.sd-card-header.sd-bg-text-warning{color:var(--sd-color-warning-bg-text)!important}details.sd-dropdown summary.sd-card-header.sd-bg-danger,details.sd-dropdown summary.sd-card-header.sd-bg-danger+div.sd-summary-content{--pst-sd-dropdown-color:var(--sd-color-danger);--pst-sd-dropdown-bg-color:var(--sd-color-danger-bg)}details.sd-dropdown summary.sd-card-header.sd-bg-text-danger{color:var(--sd-color-danger-bg-text)!important}details.sd-dropdown summary.sd-card-header.sd-bg-light,details.sd-dropdown summary.sd-card-header.sd-bg-light+div.sd-summary-content{--pst-sd-dropdown-color:var(--sd-color-light);--pst-sd-dropdown-bg-color:var(--sd-color-light-bg)}details.sd-dropdown summary.sd-card-header.sd-bg-text-light{color:var(--sd-color-light-bg-text)!important}details.sd-dropdown summary.sd-card-header.sd-bg-muted,details.sd-dropdown summary.sd-card-header.sd-bg-muted+div.sd-summary-content{--pst-sd-dropdown-color:var(--sd-color-muted);--pst-sd-dropdown-bg-color:var(--sd-color-muted-bg)}details.sd-dropdown summary.sd-card-header.sd-bg-text-muted{color:var(--sd-color-muted-bg-text)!important}details.sd-dropdown summary.sd-card-header.sd-bg-dark,details.sd-dropdown summary.sd-card-header.sd-bg-dark+div.sd-summary-content{--pst-sd-dropdown-color:var(--sd-color-dark);--pst-sd-dropdown-bg-color:var(--sd-color-dark-bg)}details.sd-dropdown summary.sd-card-header.sd-bg-text-dark{color:var(--sd-color-dark-bg-text)!important}details.sd-dropdown summary.sd-card-header.sd-bg-black,details.sd-dropdown summary.sd-card-header.sd-bg-black+div.sd-summary-content{--pst-sd-dropdown-color:var(--sd-color-black);--pst-sd-dropdown-bg-color:var(--sd-color-black-bg)}details.sd-dropdown summary.sd-card-header.sd-bg-text-black{color:var(--sd-color-black-bg-text)!important}details.sd-dropdown summary.sd-card-header.sd-bg-white,details.sd-dropdown summary.sd-card-header.sd-bg-white+div.sd-summary-content{--pst-sd-dropdown-color:var(--sd-color-white);--pst-sd-dropdown-bg-color:var(--sd-color-white-bg)}details.sd-dropdown summary.sd-card-header.sd-bg-text-white{color:var(--sd-color-white-bg-text)!important}details.sd-dropdown summary.sd-card-header:before{content:"";height:100%;left:0;pointer-events:none;position:absolute;top:0;width:100%;z-index:-1}details.sd-dropdown summary.sd-card-header+div.sd-summary-content{background-color:var(--pst-color-on-background);border-bottom-left-radius:calc(.25rem - 1px);border-left:.2rem solid var(--pst-sd-dropdown-color)!important}details.sd-dropdown summary.sd-card-header span.sd-summary-icon{align-items:center;color:var(--pst-sd-dropdown-color)!important;display:inline-flex}details.sd-dropdown summary.sd-card-header span.sd-summary-icon svg{opacity:1}details.sd-dropdown summary.sd-card-header .sd-summary-down,details.sd-dropdown summary.sd-card-header .sd-summary-up{top:.7rem}details.sd-dropdown summary.sd-card-header:hover{filter:brightness(.9)}html[data-theme=dark] details.sd-dropdown summary.sd-card-header:hover{filter:brightness(1.1)}details.sd-dropdown summary.sd-card-header:focus:focus-visible{border-radius:.1875rem;outline:var(--pst-color-accent) solid .1875rem;outline-offset:.125rem}details.sd-dropdown[open] summary.sd-card-header:focus:focus-visible{border-bottom-left-radius:0;border-bottom-right-radius:0}html .sd-btn{min-width:2.25rem;padding:.3125rem .75rem .4375rem}html .sd-btn:hover{color:var(--pst-color-link-hover);text-decoration:underline;text-decoration-skip-ink:none;text-decoration-thickness:max(1px,.0625rem);text-decoration-thickness:max(3px,.1875rem,.12em);text-decoration-thickness:1px;text-underline-offset:.1578em}html .sd-btn-primary:hover{color:var(--pst-color-primary-highlight-text)!important}html .sd-btn-secondary:hover{color:var(--pst-color-secondary-highlight-text)!important}html .sd-btn-success:hover{color:var(--pst-color-success-highlight-text)!important}html .sd-btn-info:hover{color:var(--pst-color-info-highlight-text)!important}html .sd-btn-warning:hover{color:var(--pst-color-warning-highlight-text)!important}html .sd-btn-danger:hover{color:var(--pst-color-danger-highlight-text)!important}html .sd-btn-light:hover{color:var(--pst-color-light-highlight-text)!important}html .sd-btn-muted:hover{color:var(--pst-color-muted-highlight-text)!important}html .sd-btn-dark:hover{color:var(--pst-color-dark-highlight-text)!important}html .sd-btn-black:hover{color:var(--pst-color-black-highlight-text)!important}html .sd-btn-white:hover{color:var(--pst-color-white-highlight-text)!important}html .sd-btn-outline-primary:focus-visible,html .sd-btn-primary:focus-visible{background-color:var(--sd-color-primary)!important;border-color:var(--sd-color-primary)!important;border-radius:.1875rem;outline:var(--sd-color-primary) solid .1875rem;outline-offset:.125rem}html .sd-btn-outline-secondary:focus-visible,html .sd-btn-secondary:focus-visible{background-color:var(--sd-color-secondary)!important;border-color:var(--sd-color-secondary)!important;border-radius:.1875rem;outline:var(--sd-color-secondary) solid .1875rem;outline-offset:.125rem}html .sd-btn-outline-success:focus-visible,html .sd-btn-success:focus-visible{background-color:var(--sd-color-success)!important;border-color:var(--sd-color-success)!important;border-radius:.1875rem;outline:var(--sd-color-success) solid .1875rem;outline-offset:.125rem}html .sd-btn-info:focus-visible,html .sd-btn-outline-info:focus-visible{background-color:var(--sd-color-info)!important;border-color:var(--sd-color-info)!important;border-radius:.1875rem;outline:var(--sd-color-info) solid .1875rem;outline-offset:.125rem}html .sd-btn-outline-warning:focus-visible,html .sd-btn-warning:focus-visible{background-color:var(--sd-color-warning)!important;border-color:var(--sd-color-warning)!important;border-radius:.1875rem;outline:var(--sd-color-warning) solid .1875rem;outline-offset:.125rem}html .sd-btn-danger:focus-visible,html .sd-btn-outline-danger:focus-visible{background-color:var(--sd-color-danger)!important;border-color:var(--sd-color-danger)!important;border-radius:.1875rem;outline:var(--sd-color-danger) solid .1875rem;outline-offset:.125rem}html .sd-btn-light:focus-visible,html .sd-btn-outline-light:focus-visible{background-color:var(--sd-color-light)!important;border-color:var(--sd-color-light)!important;border-radius:.1875rem;outline:var(--sd-color-light) solid .1875rem;outline-offset:.125rem}html .sd-btn-muted:focus-visible,html .sd-btn-outline-muted:focus-visible{background-color:var(--sd-color-muted)!important;border-color:var(--sd-color-muted)!important;border-radius:.1875rem;outline:var(--sd-color-muted) solid .1875rem;outline-offset:.125rem}html .sd-btn-dark:focus-visible,html .sd-btn-outline-dark:focus-visible{background-color:var(--sd-color-dark)!important;border-color:var(--sd-color-dark)!important;border-radius:.1875rem;outline:var(--sd-color-dark) solid .1875rem;outline-offset:.125rem}html .sd-btn-black:focus-visible,html .sd-btn-outline-black:focus-visible{background-color:var(--sd-color-black)!important;border-color:var(--sd-color-black)!important;border-radius:.1875rem;outline:var(--sd-color-black) solid .1875rem;outline-offset:.125rem}html .sd-btn-outline-white:focus-visible,html .sd-btn-white:focus-visible{background-color:var(--sd-color-white)!important;border-color:var(--sd-color-white)!important;border-radius:.1875rem;outline:var(--sd-color-white) solid .1875rem;outline-offset:.125rem}.bd-content .dropdown.admonition.toggle .admonition-title:hover .tb-icon{opacity:1;scale:1.1}.bd-content .dropdown.admonition.toggle .admonition-title .tb-icon{opacity:.6}.bd-content .dropdown.admonition.toggle .admonition-title:hover{filter:brightness(.9)}html[data-theme=dark] .bd-content .dropdown.admonition.toggle .admonition-title:hover{filter:brightness(1.1)}.bd-content .dropdown.admonition.toggle .admonition-title:hover{box-shadow:none}.bd-content .dropdown.admonition.toggle button.toggle-button{color:inherit}.bd-content .dropdown.admonition.toggle button.toggle-button.toggle-button-hidden .toggle-chevron-right{transform:rotate(90deg);transition:none}.bd-content .dropdown.admonition.toggle button.toggle-button .toggle-chevron-right{transform:rotate(-90deg);transition:none}.bd-content .dropdown.admonition.toggle:has(:focus-visible){overflow:visible}.bd-content .dropdown.admonition.toggle:has(:focus-visible) .admonition-title:before{border-radius:.1875rem;content:"";height:100%;left:-.2rem;outline:var(--pst-color-accent) solid .1875rem;outline-offset:.125rem;width:calc(100% + .2rem)}.bd-content .dropdown.admonition.toggle:has(:focus-visible):not(.toggle-hidden) .admonition-title:before{border-bottom-left-radius:0;border-bottom-right-radius:0}.bd-content details.toggle-details summary{border-left:3px solid var(--pst-color-primary)}.bd-content details.toggle-details summary .toggle-chevron-right{transform:rotate(90deg);transition:none}.bd-content details.toggle-details summary:hover .tb-icon{opacity:1;scale:1.1}.bd-content details.toggle-details summary .tb-icon{opacity:.6}.bd-content details.toggle-details summary:hover{filter:brightness(.9)}html[data-theme=dark] .bd-content details.toggle-details summary:hover{filter:brightness(1.1)}.bd-content details.toggle-details summary:hover{box-shadow:none}.bd-content details.toggle-details summary:focus-visible{outline-offset:.125rem}.bd-content details.toggle-details[open] .toggle-chevron-right{transform:rotate(-90deg);transition:none}.bd-content details.toggle-details[open] :focus-visible{border-bottom-left-radius:0;border-bottom-right-radius:0}html .jp-RenderedHTMLCommon,html div.rendered_html{margin:.1875rem}html .jp-RenderedHTMLCommon table,html div.rendered_html table{table-layout:auto}.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe,.bd-content .nboutput .output_area.rendered_html table.dataframe{border:1px solid var(--pst-color-table-outer-border);color:var(--pst-color-table)}.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe td~td,.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe td~th,.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe th~td,.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe th~th,.bd-content .nboutput .output_area.rendered_html table.dataframe td~td,.bd-content .nboutput .output_area.rendered_html table.dataframe td~th,.bd-content .nboutput .output_area.rendered_html table.dataframe th~td,.bd-content .nboutput .output_area.rendered_html table.dataframe th~th{border-left:1px solid var(--pst-color-table-inner-border)}.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe thead tr,.bd-content .nboutput .output_area.rendered_html table.dataframe thead tr{background-color:var(--pst-color-table-heading-bg);border-bottom:2px solid var(--pst-color-primary)}.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe tbody tr:nth-child(odd),.bd-content .nboutput .output_area.rendered_html table.dataframe tbody tr:nth-child(odd){background-color:var(--pst-color-table-row-zebra-low-bg)}.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe tbody tr:nth-child(2n),.bd-content .nboutput .output_area.rendered_html table.dataframe tbody tr:nth-child(2n){background-color:var(--pst-color-table-row-zebra-high-bg)}.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe tbody tr:hover,.bd-content .nboutput .output_area.rendered_html table.dataframe tbody tr:hover{background-color:var(--pst-color-table-row-hover-bg)}html[data-theme=dark] .bd-content .nboutput .output_area .widget-subarea,html[data-theme=dark] .bd-content .nboutput .output_area.rendered_html:not(:has(table.dataframe)){background-color:var(--pst-color-text-base);border-radius:.25rem;color:var(--pst-color-on-background);padding:.5rem}html[data-theme=dark] .bd-content .nboutput .output_area.stderr{background-color:var(--pst-color-danger)}div.nblast.container{margin-bottom:1rem}html[data-theme=light] .nbinput.container .prompt pre{color:#005b82}html[data-theme=dark] .nbinput.container .prompt pre{color:#00e0e0}html[data-theme=light] .nboutput.container .prompt pre{color:#a12236}html[data-theme=dark] .nboutput.container .prompt pre{color:#ffa07a}div.cell_output .output{max-width:100%;overflow-x:auto}.bd-content div.cell_output table.dataframe{border:1px solid var(--pst-color-table-outer-border);color:var(--pst-color-table)}.bd-content div.cell_output table.dataframe td~td,.bd-content div.cell_output table.dataframe td~th,.bd-content div.cell_output table.dataframe th~td,.bd-content div.cell_output table.dataframe th~th{border-left:1px solid var(--pst-color-table-inner-border)}.bd-content div.cell_output table.dataframe thead tr{background-color:var(--pst-color-table-heading-bg);border-bottom:2px solid var(--pst-color-primary)}.bd-content div.cell_output table.dataframe tbody tr:nth-child(odd){background-color:var(--pst-color-table-row-zebra-low-bg)}.bd-content div.cell_output table.dataframe tbody tr:nth-child(2n){background-color:var(--pst-color-table-row-zebra-high-bg)}.bd-content div.cell_output table.dataframe tbody tr:hover{background-color:var(--pst-color-table-row-hover-bg)}html[data-theme=dark] .bd-content div.cell_output .text_html:not(:has(table.dataframe)),html[data-theme=dark] .bd-content div.cell_output .widget-subarea,html[data-theme=dark] .bd-content div.cell_output img{background-color:var(--pst-color-text-base);border-radius:.25rem;color:var(--pst-color-on-background);padding:.5rem}.bd-content div.cell_input{display:flex;flex-direction:column;justify-content:stretch}.bd-content div.cell_input,.bd-content div.output{border-radius:.25rem}.bd-content div.output table{table-layout:auto}html[data-theme=dark] .bd-content img.leaflet-tile.leaflet-tile-loaded{border-radius:0;padding:0}.bd-search-container div#search-results>.h2,.bd-search-container div#search-results>h2{font-size:var(--pst-font-size-icon);margin-top:1rem}.bd-search-container div#search-results p.search-summary{color:var(--pst-color-text-muted)}.bd-search-container ul.search{list-style:none;margin:0}.bd-search-container ul.search li{background-image:none;border-top:1px solid var(--pst-color-text-muted);margin:1rem 0;padding:1rem 0}.bd-search-container ul.search li>a{font-size:1.2em}.bd-search-container ul.search li div.context,.bd-search-container ul.search li p.context{color:var(--pst-color-text-base);margin:.5em 0 0}.bd-search-container ul.search li div.context a:before,.bd-search-container ul.search li p.context a:before{color:var(--pst-color-text-muted);content:"#";padding-right:.2em}
+/*# sourceMappingURL=pydata-sphinx-theme.css.map*/
\ No newline at end of file
diff --git a/_static/styles/pydata-sphinx-theme.css.map b/_static/styles/pydata-sphinx-theme.css.map
new file mode 100644
index 000000000..1142bcdf1
--- /dev/null
+++ b/_static/styles/pydata-sphinx-theme.css.map
@@ -0,0 +1 @@
+{"version":3,"file":"styles/pydata-sphinx-theme.css","mappings":"AAsGE,gBCqlBF;AC1rBE;;;;EAAA,CCDF,4BASI,qQAIA,sMAIA,iKAIA,sNAIA,iRAIA,iPAIA,iRAGF,2BACA,qBAMA,yMACA,mGACA,4EAOA,gDC2OI,wBALI,CDpOR,0BACA,0BAKA,wBACA,6BACA,kBACA,6BAEA,yBACA,8BAEA,wCACA,kCACA,0BACA,kCAEA,sCACA,iCACA,yBACA,iCAGA,2BAEA,wBACA,+BACA,+BAEA,8BACA,oCAMA,wBACA,6BACA,0BAGA,sBACA,wBACA,0BACA,+CAEA,4BACA,8BACA,6BACA,2BACA,4BACA,mDACA,8BAGA,8CACA,uDACA,gDACA,uDAIA,gCACA,0BACA,8CAIA,8BACA,qCACA,gCACA,sCFFF,CI9GI,qBFyHA,wBACA,gCACA,qBACA,0BAEA,yBACA,oCAEA,2CACA,qCACA,0BACA,+BAEA,yCACA,oCACA,yBACA,8BAGE,iRAIA,iPAIA,iRAGF,2BAEA,wBACA,8BACA,gCACA,sCAEA,wBACA,6BACA,0BAEA,0BACA,kDAEA,8BACA,qCACA,gCACA,uCAlDA,iBFgDJ,CKtKA,iBAGE,qBLyKF,CK1JI,8CANJ,MAOM,sBL8JJ,CACF,CKjJA,KASE,6BAA8B,CAC9B,0CAFA,mCAFA,2BAJA,uCF6OI,kCALI,CEtOR,uCACA,uCAJA,QAAS,CAMT,oCLuJF,CK3IA,GAIE,wCAFA,aCmnB4B,CDpnB5B,cAIA,WL8IF,CKpIA,0CAOE,8BAFA,eCwjB4B,CDvjB5B,eCwjB4B,CD5jB5B,mBCwjB4B,CDzjB5B,YL2IF,CKlIA,OFuMQ,gCHjER,CG3FI,0BE3CJ,OF8MQ,gBHpEN,CACF,CKtIA,OFkMQ,+BHxDR,CGpGI,0BEtCJ,OFyMQ,cH3DN,CACF,CK1IA,OF6LQ,6BH/CR,CG7GI,0BEjCJ,OFoMQ,iBHlDN,CACF,CK9IA,OFwLQ,+BHtCR,CGtHI,0BE5BJ,OF+LQ,gBHzCN,CACF,CKlJA,OF+KM,iBHzBN,CKjJA,OF0KM,cHrBN,CK1IA,EAEE,kBCwV0B,CDzV1B,YL8IF,CKnIA,YAEE,WAAY,CADZ,gCAAiC,CAEjC,6BLsIF,CKhIA,QAEE,iBAAkB,CAClB,mBAAoB,CAFpB,kBLqIF,CK7HA,MAEE,iBLgIF,CK7HA,SAIE,kBAAmB,CADnB,YLiIF,CK7HA,wBAIE,eLgIF,CK7HA,GACE,eLgIF,CK3HA,GACE,oBACA,aL8HF,CKxHA,WACE,eL2HF,CKnHA,SAEE,kBLsHF,CK9GA,aF6EM,gBHqCN,CK3GA,WAGE,wCADA,gCADA,eLgHF,CKrGA,QF0DM,eALI,CEjDR,aAAc,CAFd,iBAAkB,CAGlB,uBLwGF,CKrGA,IAAM,aLyGN,CKxGA,IAAM,SL4GN,CKvGA,EACE,6DL2GF,CKxGE,QACE,kDL0GJ,CK/FE,4DAEE,aAAc,CACd,oBLiGJ,CK1FA,kBAIE,oCCgV4B,CHlUxB,aHgFN,CKtFA,IACE,aAAc,CFKV,gBALI,CEER,kBAAmB,CADnB,YAAa,CAEb,aL0FF,CKrFE,SAEE,aAAc,CFLZ,iBALI,CEWN,iBLuFJ,CKnFA,KAGE,oBAAqB,CADrB,2BFZI,gBHmGN,CKnFE,OACE,aLqFJ,CKjFA,IAIE,qCCy5CkC,CC9rDhC,qBFoSF,uBCy5CkC,CHj7C9B,gBALI,CE2BR,wBLwFF,CKlFE,QF5BI,aALI,CEkCN,SLqFJ,CK1EA,OACE,eL6EF,CKvEA,QAEE,qBL0EF,CKlEA,MAEE,wBAAyB,CADzB,mBLsEF,CKlEA,QAGE,+BC4Z4B,CD7Z5B,oBC2X4B,CD5X5B,iBC4X4B,CDzX5B,eLqEF,CK9DA,GAEE,kBAAmB,CACnB,+BLgEF,CK7DA,2BAQE,cAAe,CAFf,oBLkEF,CKxDA,MACE,oBL2DF,CKrDA,OAEE,eLuDF,CK/CA,iCACE,SLkDF,CK7CA,sCAME,mBAAoB,CF5HhB,iBALI,CEmIR,mBAAoB,CAHpB,QLmDF,CK5CA,cAEE,mBL+CF,CK1CA,cACE,cL6CF,CK1CA,OAGE,gBL2CF,CKxCE,gBACE,SL0CJ,CKnCA,0IACE,sBLsCF,CK9BA,gDAIE,yBLiCF,CK9BI,4GACE,cLmCN,CK5BA,mBAEE,iBAAkB,CADlB,SLgCF,CK1BA,SACE,eL6BF,CKnBA,SAIE,QAAS,CADT,QAAS,CAFT,WAAY,CACZ,SLwBF,CKdA,OACE,UAAW,CF9ML,gCEoNN,mBAAoB,CAHpB,mBCmN4B,CDpN5B,SAAU,CADV,ULqBF,CGhYI,0BEyWJ,OFtMQ,gBHiON,CACF,CKnBE,SACE,ULqBJ,CKdA,+OAOE,SLiBF,CKdA,4BACE,WLiBF,CKRA,cACE,4BAA6B,CAC7B,mBLWF,CKQA,4BACE,uBLGF,CKEA,+BACE,SLCF,CKMA,uBAEE,yBAA0B,CAD1B,YLFF,CKQA,OACE,oBLLF,CKUA,OACE,QLPF,CKcA,QAEE,cAAe,CADf,iBLVF,CKmBA,SACE,uBLhBF,CKwBA,SACE,sBLrBF,CDhjBA,MImQM,iBALI,CJ5PR,eCmjBF,CD9iBE,WIgQM,iCJ5PJ,eOynBkB,CPxnBlB,eC+iBJ,CGhdI,0BJpGF,WIuQM,cHiTN,CACF,CDzjBE,WIgQM,iCJ5PJ,eOynBkB,CPxnBlB,eC0jBJ,CG3dI,0BJpGF,WIuQM,gBH4TN,CACF,CDpkBE,WIgQM,iCJ5PJ,eOynBkB,CPxnBlB,eCqkBJ,CGteI,0BJpGF,WIuQM,cHuUN,CACF,CD/kBE,WIgQM,iCJ5PJ,eOynBkB,CPxnBlB,eCglBJ,CGjfI,0BJpGF,WIuQM,gBHkVN,CACF,CD1lBE,WIgQM,iCJ5PJ,eOynBkB,CPxnBlB,eC2lBJ,CG5fI,0BJpGF,WIuQM,cH6VN,CACF,CDrmBE,WIgQM,iCJ5PJ,eOynBkB,CPxnBlB,eCsmBJ,CGvgBI,0BJpGF,WIuQM,gBHwWN,CACF,CDnlBA,4BS3DE,eAAgB,CADhB,cRypBF,CD1lBA,kBACE,oBC6lBF,CD3lBE,mCACE,kBC6lBJ,CDnlBA,YI8MM,gBALI,CJvMR,wBCslBF,CDllBA,YIwMM,iBALI,CJlMR,kBCslBF,CDnlBE,wBACE,eCqlBJ,CDjlBA,mBAIE,aOtFS,CHiRL,gBALI,CJxLR,kBUjGO,CVgGP,gBCulBF,CDllBE,0BACE,YColBJ,CU9qBA,0BCCE,WAAY,CAHZ,cX8rBF,CU5rBA,eAEE,kCJ+jDkC,CI9jDlC,2DHGE,sCGLF,cV2rBF,CU7qBA,QAEE,oBV+qBF,CU5qBA,YAEE,aAAc,CADd,mBVgrBF,CU5qBA,gBAEE,+BJkjDkC,CH3zC9B,gBHwbN,CYjtBE,oFCHA,qBACA,gBAKA,gBAAiB,CADjB,iBAAkB,CADlB,yCADA,0CADA,UbiuBF,CcvqBI,yBF5CE,yBACE,eZutBN,CACF,Cc7qBI,yBF5CE,uCACE,eZ4tBN,CACF,CclrBI,yBF5CE,qDACE,eZiuBN,CACF,CcvrBI,0BF5CE,mEACE,gBZsuBN,CACF,CevvBA,MAEI,yHf4vBJ,CevvBE,KCNA,qBACA,gBACA,YAAa,CACb,cAAe,CAIf,yCADA,0CADA,sChBkwBF,Ce9vBI,OCOF,aAAc,CAKd,8BAHA,cAAe,CAEf,yCADA,0CAFA,UhB8vBF,CgB3sBM,KACE,WhB8sBR,CgB3sBM,iBApCJ,aAAc,CACd,UhBmvBF,CgBruBE,cACE,aAAc,CACd,UhBwuBJ,CgB1uBE,cACE,aAAc,CACd,ShB6uBJ,CgB/uBE,cACE,aAAc,CACd,kBhBkvBJ,CgBpvBE,cACE,aAAc,CACd,ShBuvBJ,CgBzvBE,cACE,aAAc,CACd,ShB4vBJ,CgB9vBE,cACE,aAAc,CACd,kBhBiwBJ,CgBluBM,UAhDJ,aAAc,CACd,UhBsxBF,CgBjuBU,OAhEN,aAAc,CACd,iBhBqyBJ,CgBtuBU,OAhEN,aAAc,CACd,kBhB0yBJ,CgB3uBU,OAhEN,aAAc,CACd,ShB+yBJ,CgBhvBU,OAhEN,aAAc,CACd,kBhBozBJ,CgBrvBU,OAhEN,aAAc,CACd,kBhByzBJ,CgB1vBU,OAhEN,aAAc,CACd,ShB8zBJ,CgB/vBU,OAhEN,aAAc,CACd,kBhBm0BJ,CgBpwBU,OAhEN,aAAc,CACd,kBhBw0BJ,CgBzwBU,OAhEN,aAAc,CACd,ShB60BJ,CgB9wBU,QAhEN,aAAc,CACd,kBhBk1BJ,CgBnxBU,QAhEN,aAAc,CACd,kBhBu1BJ,CgBxxBU,QAhEN,aAAc,CACd,UhB41BJ,CgBrxBY,UAxDV,uBhBi1BF,CgBzxBY,UAxDV,wBhBq1BF,CgB7xBY,UAxDV,ehBy1BF,CgBjyBY,UAxDV,wBhB61BF,CgBryBY,UAxDV,wBhBi2BF,CgBzyBY,UAxDV,ehBq2BF,CgB7yBY,UAxDV,wBhBy2BF,CgBjzBY,UAxDV,wBhB62BF,CgBrzBY,UAxDV,ehBi3BF,CgBzzBY,WAxDV,wBhBq3BF,CgB7zBY,WAxDV,wBhBy3BF,CgBtzBQ,WAEE,ehByzBV,CgBtzBQ,WAEE,ehByzBV,CgBh0BQ,WAEE,qBhBm0BV,CgBh0BQ,WAEE,qBhBm0BV,CgB10BQ,WAEE,oBhB60BV,CgB10BQ,WAEE,oBhB60BV,CgBp1BQ,WAEE,kBhBu1BV,CgBp1BQ,WAEE,kBhBu1BV,CgB91BQ,WAEE,oBhBi2BV,CgB91BQ,WAEE,oBhBi2BV,CgBx2BQ,WAEE,kBhB22BV,CgBx2BQ,WAEE,kBhB22BV,Ccr6BI,yBEUE,QACE,WhB+5BN,CgB55BI,oBApCJ,aAAc,CACd,UhBm8BA,CgBr7BA,iBACE,aAAc,CACd,UhBu7BF,CgBz7BA,iBACE,aAAc,CACd,ShB27BF,CgB77BA,iBACE,aAAc,CACd,kBhB+7BF,CgBj8BA,iBACE,aAAc,CACd,ShBm8BF,CgBr8BA,iBACE,aAAc,CACd,ShBu8BF,CgBz8BA,iBACE,aAAc,CACd,kBhB28BF,CgB56BI,aAhDJ,aAAc,CACd,UhB+9BA,CgB16BQ,UAhEN,aAAc,CACd,iBhB6+BF,CgB96BQ,UAhEN,aAAc,CACd,kBhBi/BF,CgBl7BQ,UAhEN,aAAc,CACd,ShBq/BF,CgBt7BQ,UAhEN,aAAc,CACd,kBhBy/BF,CgB17BQ,UAhEN,aAAc,CACd,kBhB6/BF,CgB97BQ,UAhEN,aAAc,CACd,ShBigCF,CgBl8BQ,UAhEN,aAAc,CACd,kBhBqgCF,CgBt8BQ,UAhEN,aAAc,CACd,kBhBygCF,CgB18BQ,UAhEN,aAAc,CACd,ShB6gCF,CgB98BQ,WAhEN,aAAc,CACd,kBhBihCF,CgBl9BQ,WAhEN,aAAc,CACd,kBhBqhCF,CgBt9BQ,WAhEN,aAAc,CACd,UhByhCF,CgBl9BU,aAxDV,ahB6gCA,CgBr9BU,aAxDV,uBhBghCA,CgBx9BU,aAxDV,wBhBmhCA,CgB39BU,aAxDV,ehBshCA,CgB99BU,aAxDV,wBhByhCA,CgBj+BU,aAxDV,wBhB4hCA,CgBp+BU,aAxDV,ehB+hCA,CgBv+BU,aAxDV,wBhBkiCA,CgB1+BU,aAxDV,wBhBqiCA,CgB7+BU,aAxDV,ehBwiCA,CgBh/BU,cAxDV,wBhB2iCA,CgBn/BU,cAxDV,wBhB8iCA,CgB3+BM,iBAEE,ehB6+BR,CgB1+BM,iBAEE,ehB4+BR,CgBn/BM,iBAEE,qBhBq/BR,CgBl/BM,iBAEE,qBhBo/BR,CgB3/BM,iBAEE,oBhB6/BR,CgB1/BM,iBAEE,oBhB4/BR,CgBngCM,iBAEE,kBhBqgCR,CgBlgCM,iBAEE,kBhBogCR,CgB3gCM,iBAEE,oBhB6gCR,CgB1gCM,iBAEE,oBhB4gCR,CgBnhCM,iBAEE,kBhBqhCR,CgBlhCM,iBAEE,kBhBohCR,CACF,Cc/kCI,yBEUE,QACE,WhBwkCN,CgBrkCI,oBApCJ,aAAc,CACd,UhB4mCA,CgB9lCA,iBACE,aAAc,CACd,UhBgmCF,CgBlmCA,iBACE,aAAc,CACd,ShBomCF,CgBtmCA,iBACE,aAAc,CACd,kBhBwmCF,CgB1mCA,iBACE,aAAc,CACd,ShB4mCF,CgB9mCA,iBACE,aAAc,CACd,ShBgnCF,CgBlnCA,iBACE,aAAc,CACd,kBhBonCF,CgBrlCI,aAhDJ,aAAc,CACd,UhBwoCA,CgBnlCQ,UAhEN,aAAc,CACd,iBhBspCF,CgBvlCQ,UAhEN,aAAc,CACd,kBhB0pCF,CgB3lCQ,UAhEN,aAAc,CACd,ShB8pCF,CgB/lCQ,UAhEN,aAAc,CACd,kBhBkqCF,CgBnmCQ,UAhEN,aAAc,CACd,kBhBsqCF,CgBvmCQ,UAhEN,aAAc,CACd,ShB0qCF,CgB3mCQ,UAhEN,aAAc,CACd,kBhB8qCF,CgB/mCQ,UAhEN,aAAc,CACd,kBhBkrCF,CgBnnCQ,UAhEN,aAAc,CACd,ShBsrCF,CgBvnCQ,WAhEN,aAAc,CACd,kBhB0rCF,CgB3nCQ,WAhEN,aAAc,CACd,kBhB8rCF,CgB/nCQ,WAhEN,aAAc,CACd,UhBksCF,CgB3nCU,aAxDV,ahBsrCA,CgB9nCU,aAxDV,uBhByrCA,CgBjoCU,aAxDV,wBhB4rCA,CgBpoCU,aAxDV,ehB+rCA,CgBvoCU,aAxDV,wBhBksCA,CgB1oCU,aAxDV,wBhBqsCA,CgB7oCU,aAxDV,ehBwsCA,CgBhpCU,aAxDV,wBhB2sCA,CgBnpCU,aAxDV,wBhB8sCA,CgBtpCU,aAxDV,ehBitCA,CgBzpCU,cAxDV,wBhBotCA,CgB5pCU,cAxDV,wBhButCA,CgBppCM,iBAEE,ehBspCR,CgBnpCM,iBAEE,ehBqpCR,CgB5pCM,iBAEE,qBhB8pCR,CgB3pCM,iBAEE,qBhB6pCR,CgBpqCM,iBAEE,oBhBsqCR,CgBnqCM,iBAEE,oBhBqqCR,CgB5qCM,iBAEE,kBhB8qCR,CgB3qCM,iBAEE,kBhB6qCR,CgBprCM,iBAEE,oBhBsrCR,CgBnrCM,iBAEE,oBhBqrCR,CgB5rCM,iBAEE,kBhB8rCR,CgB3rCM,iBAEE,kBhB6rCR,CACF,CcxvCI,yBEUE,QACE,WhBivCN,CgB9uCI,oBApCJ,aAAc,CACd,UhBqxCA,CgBvwCA,iBACE,aAAc,CACd,UhBywCF,CgB3wCA,iBACE,aAAc,CACd,ShB6wCF,CgB/wCA,iBACE,aAAc,CACd,kBhBixCF,CgBnxCA,iBACE,aAAc,CACd,ShBqxCF,CgBvxCA,iBACE,aAAc,CACd,ShByxCF,CgB3xCA,iBACE,aAAc,CACd,kBhB6xCF,CgB9vCI,aAhDJ,aAAc,CACd,UhBizCA,CgB5vCQ,UAhEN,aAAc,CACd,iBhB+zCF,CgBhwCQ,UAhEN,aAAc,CACd,kBhBm0CF,CgBpwCQ,UAhEN,aAAc,CACd,ShBu0CF,CgBxwCQ,UAhEN,aAAc,CACd,kBhB20CF,CgB5wCQ,UAhEN,aAAc,CACd,kBhB+0CF,CgBhxCQ,UAhEN,aAAc,CACd,ShBm1CF,CgBpxCQ,UAhEN,aAAc,CACd,kBhBu1CF,CgBxxCQ,UAhEN,aAAc,CACd,kBhB21CF,CgB5xCQ,UAhEN,aAAc,CACd,ShB+1CF,CgBhyCQ,WAhEN,aAAc,CACd,kBhBm2CF,CgBpyCQ,WAhEN,aAAc,CACd,kBhBu2CF,CgBxyCQ,WAhEN,aAAc,CACd,UhB22CF,CgBpyCU,aAxDV,ahB+1CA,CgBvyCU,aAxDV,uBhBk2CA,CgB1yCU,aAxDV,wBhBq2CA,CgB7yCU,aAxDV,ehBw2CA,CgBhzCU,aAxDV,wBhB22CA,CgBnzCU,aAxDV,wBhB82CA,CgBtzCU,aAxDV,ehBi3CA,CgBzzCU,aAxDV,wBhBo3CA,CgB5zCU,aAxDV,wBhBu3CA,CgB/zCU,aAxDV,ehB03CA,CgBl0CU,cAxDV,wBhB63CA,CgBr0CU,cAxDV,wBhBg4CA,CgB7zCM,iBAEE,ehB+zCR,CgB5zCM,iBAEE,ehB8zCR,CgBr0CM,iBAEE,qBhBu0CR,CgBp0CM,iBAEE,qBhBs0CR,CgB70CM,iBAEE,oBhB+0CR,CgB50CM,iBAEE,oBhB80CR,CgBr1CM,iBAEE,kBhBu1CR,CgBp1CM,iBAEE,kBhBs1CR,CgB71CM,iBAEE,oBhB+1CR,CgB51CM,iBAEE,oBhB81CR,CgBr2CM,iBAEE,kBhBu2CR,CgBp2CM,iBAEE,kBhBs2CR,CACF,Ccj6CI,0BEUE,QACE,WhB05CN,CgBv5CI,oBApCJ,aAAc,CACd,UhB87CA,CgBh7CA,iBACE,aAAc,CACd,UhBk7CF,CgBp7CA,iBACE,aAAc,CACd,ShBs7CF,CgBx7CA,iBACE,aAAc,CACd,kBhB07CF,CgB57CA,iBACE,aAAc,CACd,ShB87CF,CgBh8CA,iBACE,aAAc,CACd,ShBk8CF,CgBp8CA,iBACE,aAAc,CACd,kBhBs8CF,CgBv6CI,aAhDJ,aAAc,CACd,UhB09CA,CgBr6CQ,UAhEN,aAAc,CACd,iBhBw+CF,CgBz6CQ,UAhEN,aAAc,CACd,kBhB4+CF,CgB76CQ,UAhEN,aAAc,CACd,ShBg/CF,CgBj7CQ,UAhEN,aAAc,CACd,kBhBo/CF,CgBr7CQ,UAhEN,aAAc,CACd,kBhBw/CF,CgBz7CQ,UAhEN,aAAc,CACd,ShB4/CF,CgB77CQ,UAhEN,aAAc,CACd,kBhBggDF,CgBj8CQ,UAhEN,aAAc,CACd,kBhBogDF,CgBr8CQ,UAhEN,aAAc,CACd,ShBwgDF,CgBz8CQ,WAhEN,aAAc,CACd,kBhB4gDF,CgB78CQ,WAhEN,aAAc,CACd,kBhBghDF,CgBj9CQ,WAhEN,aAAc,CACd,UhBohDF,CgB78CU,aAxDV,ahBwgDA,CgBh9CU,aAxDV,uBhB2gDA,CgBn9CU,aAxDV,wBhB8gDA,CgBt9CU,aAxDV,ehBihDA,CgBz9CU,aAxDV,wBhBohDA,CgB59CU,aAxDV,wBhBuhDA,CgB/9CU,aAxDV,ehB0hDA,CgBl+CU,aAxDV,wBhB6hDA,CgBr+CU,aAxDV,wBhBgiDA,CgBx+CU,aAxDV,ehBmiDA,CgB3+CU,cAxDV,wBhBsiDA,CgB9+CU,cAxDV,wBhByiDA,CgBt+CM,iBAEE,ehBw+CR,CgBr+CM,iBAEE,ehBu+CR,CgB9+CM,iBAEE,qBhBg/CR,CgB7+CM,iBAEE,qBhB++CR,CgBt/CM,iBAEE,oBhBw/CR,CgBr/CM,iBAEE,oBhBu/CR,CgB9/CM,iBAEE,kBhBggDR,CgB7/CM,iBAEE,kBhB+/CR,CgBtgDM,iBAEE,oBhBwgDR,CgBrgDM,iBAEE,oBhBugDR,CgB9gDM,iBAEE,kBhBghDR,CgB7gDM,iBAEE,kBhB+gDR,CACF,CiBroDA,OAEE,8BACA,2BACA,+BACA,4BAEA,0CACA,gCACA,+CACA,iCACA,kDACA,8DACA,iDACA,4DACA,gDACA,6DAKA,0CAFA,kBRtBO,CQuBP,kBXusB4B,CWzsB5B,UjBuoDF,CiB7nDE,yBAIE,oCACA,0CX+sB0B,CW9sB1B,yGAHA,mFAFA,ajBmoDJ,CiB3nDE,aACE,sBjB6nDJ,CiB1nDE,aACE,qBjB4nDJ,CiBxnDA,qBACE,+CjB2nDF,CiBpnDA,aACE,gBjBunDF,CiB7mDE,4BACE,cjBgnDJ,CiBjmDE,gCACE,qCjBomDJ,CiBjmDI,kCACE,qCjBmmDN,CiB5lDE,oCACE,qBjB+lDJ,CiB5lDE,qCACE,kBjB8lDJ,CiB5kDE,kGACE,oDACA,6CjBolDJ,CiB5kDA,cACE,oDACA,6CjB+kDF,CiBvkDE,8BACE,mDACA,4CjB0kDJ,CkBttDE,eAOE,sBACA,sBACA,gCACA,8BACA,8BACA,6BACA,6BACA,4BACA,2BlBqtDJ,CkBpuDE,gCAkBE,0CADA,2BlBiuDJ,CkBlvDE,iBAOE,sBACA,sBACA,gCACA,8BACA,8BACA,6BACA,6BACA,4BACA,2BlBmuDJ,CkBlvDE,eAOE,sBACA,sBACA,gCACA,8BACA,8BACA,6BACA,6BACA,4BACA,2BlBivDJ,CkBhwDE,2BAkBE,0CADA,2BlB6vDJ,CkB9wDE,YAOE,sBACA,sBACA,gCACA,8BACA,8BACA,6BACA,6BACA,4BACA,2BlB+vDJ,CkB9wDE,eAOE,sBACA,sBACA,gCACA,8BACA,8BACA,6BACA,6BACA,4BACA,2BlB6wDJ,CkB5xDE,6BAkBE,0CADA,2BlByxDJ,CkB1yDE,cAOE,sBACA,sBACA,gCACA,8BACA,8BACA,6BACA,6BACA,4BACA,2BlB2xDJ,CkB1yDE,aAOE,sBACA,sBACA,gCACA,8BACA,8BACA,6BACA,6BACA,4BACA,2BlByyDJ,CkBxzDE,yBAkBE,0CADA,2BlBqzDJ,CkBt0DE,YAOE,sBACA,sBACA,gCACA,8BACA,8BACA,6BACA,6BACA,4BACA,2BlBuzDJ,CiBnqDI,kBAEE,gCAAiC,CADjC,ejBuqDN,CcjwDI,4BGyFA,qBAEE,gCAAiC,CADjC,ejB6qDJ,CACF,CcxwDI,4BGyFA,qBAEE,gCAAiC,CADjC,ejBmrDJ,CACF,Cc9wDI,4BGyFA,qBAEE,gCAAiC,CADjC,ejByrDJ,CACF,CcpxDI,6BGyFA,qBAEE,gCAAiC,CADjC,ejB+rDJ,CACF,CmBl2DA,YACE,mBnBo2DF,CmB31DA,gBhBiRM,iBALI,CgBrQR,eb+lB4B,CanmB5B,eAAgB,CADhB,sDADA,kDnBk2DF,CmBx1DA,mBhBsQM,iBALI,CgB/PR,oDADA,gDnB61DF,CmBx1DA,mBhBgQM,iBALI,CgBzPR,qDADA,iDnB61DF,CoBz3DA,WAKE,+Bd+1BsC,CHzkBlC,gBALI,CiBrRR,iBpB83DF,CqB/3DA,cASE,eAAgB,CAEhB,2BAA4B,CAD5B,kCfq3BsC,Cen3BtC,2DdGE,sCcPF,0Bf43BsC,Cen4BtC,aAAc,ClB0RV,cALI,CkBhRR,efkmB4B,CejmB5B,efymB4B,Ce7mB5B,uBCSI,oEDMJ,CAhBA,UrB64DF,CsB/3DM,uCDhBN,cCiBQ,etBk4DN,CACF,CqBh4DE,yBACE,erBk4DJ,CqBh4DI,wDACE,crBk4DN,CqB73DE,oBAEE,kCfg2BoC,Ce/1BpC,oBf82BoC,Cex2BlC,iDZhBkB,CYQpB,0Bfs2BoC,Cen2BpC,SrBg4DJ,CqBv3DE,2CAYE,aAKA,QAAS,CAXT,crBs3DJ,CqBt2DE,qCACE,aAAc,CACd,SrBw2DJ,CqBp2DE,2BACE,+Bf40BoC,Ce10BpC,SrBq2DJ,CqB71DE,uBAEE,uCf8yBoC,Ce3yBpC,SrB41DJ,CqBx1DE,oCE1FA,sCjBqiCgC,Cel8B9B,cAAe,CAFf,oBAAqB,CAGrB,8CfgsB0B,Ce/rB1B,eAAgB,CAPhB,0BfsyBoC,CexyBpC,wBACA,wBforB0B,CetrB1B,uBAKA,mBAAoB,CCpFlB,6HtBo7DN,CsBh7DM,uCD0EJ,oCCzEM,etBm7DN,CACF,CqB51DE,yEACE,uCrB81DJ,CqBr1DA,wBAOE,4BAA6B,CAC7B,wBAAyB,CACzB,sCAHA,0Bf2xBsC,CehyBtC,aAAc,CAId,efwf4B,Cezf5B,eAAgB,CADhB,kBADA,UrB+1DF,CqBt1DE,8BACE,SrBw1DJ,CqBr1DE,gFAGE,cAAe,CADf,erBu1DJ,CqB30DA,iBdjII,yCJ4QE,iBALI,CkBrIR,yDf4wBsC,Ce3wBtC,oBrBg1DF,CqB50DE,uCAEE,sBACA,uBfooB0B,CetoB1B,oBrBg1DJ,CqB10DA,iBd9II,yCJ4QE,iBALI,CkBxHR,wDfgwBsC,Ce/vBtC,kBrB+0DF,CqB30DE,uCAEE,oBACA,sBf2nB0B,Ce7nB1B,kBrB+0DJ,CqBr0DE,sBACE,0DrBw0DJ,CqBr0DE,yBACE,yDrBu0DJ,CqBp0DE,yBACE,wDrBs0DJ,CqBj0DA,oBAEE,sDf8tBsC,Ce7tBtC,efilB4B,CenlB5B,UrBs0DF,CqBl0DE,mDACE,crBo0DJ,CqBj0DE,uCACE,kBAAoB,CdvLpB,qCP2/DJ,CqBh0DE,0CACE,kBAAoB,Cd5LpB,qCP+/DJ,CqB/zDE,oCAAoB,qDrBk0DtB,CqBj0DE,oCAAoB,oDrBo0DtB,CwBnhEA,aACE,sQAUA,eAAgB,CAChB,kClBk3BsC,CkBj3BtC,iFAEA,uClB+9BkC,CkBh+BlC,2BAA4B,CAE5B,yBlB+9BkC,CkB99BlC,2DjBHE,sCiBJF,0BlBy3BsC,CkBh4BtC,aAAc,CrBuRV,cALI,CqB7QR,elB+lB4B,CkB9lB5B,elBsmB4B,CkB1mB5B,uCFMI,oEESJ,CAhBA,UxBmiEF,CsBxhEM,uCEfN,aFgBQ,etB2hEN,CACF,CwBthEE,mBACE,oBlBs3BoC,CkBh3BlC,iDlBi+B4B,CkBt+B9B,SxByhEJ,CwBhhEE,0DAGE,qBAAsB,CADtB,oBxBkhEJ,CwB9gEE,sBAEE,uCxB+gEJ,CwB1gEE,4BACE,iBAAkB,CAClB,sCxB4gEJ,CwBxgEA,gBjBtCI,yCJ4QE,iBALI,CqB/NR,qBlBquB4B,CkBpuB5B,kBlBquB4B,CkBvuB5B,kBxB+gEF,CwBxgEA,gBjB9CI,yCJ4QE,iBALI,CqBvNR,oBlBiuB4B,CkBhuB5B,iBlBiuB4B,CkBnuB5B,iBxB+gEF,CwBtgEI,kCACE,qQxBygEN,CyBjlEA,YACE,aAAc,CAGd,qBnBq6BwC,CmBv6BxC,iBnBq6BwC,CmBp6BxC,kBzBqlEF,CyBllEE,8BACE,UAAW,CACX,kBzBolEJ,CyBhlEA,oBAEE,cAAe,CADf,mBnB25BwC,CmBz5BxC,gBzBmlEF,CyBjlEE,sCACE,WAAY,CAEZ,aAAc,CADd,mBzBolEJ,CyB/kEA,kBACE,qCAOA,eAAgB,CAChB,yCACA,+CAEA,uBAA2B,CAD3B,2BAA4B,CAE5B,uBAAwB,CACxB,0DnB24BwC,CmBt5BxC,aAAc,CAEd,UnBy4BwC,CmBx4BxC,iBASA,wBAAyB,CARzB,kBAAmB,CAHnB,SzB4lEF,CyB9kEE,iClB3BE,mBP4mEJ,CyB7kEE,8BAEE,iBzB8kEJ,CyB3kEE,yBACE,sBzB6kEJ,CyB1kEE,wBACE,oBnBs1BoC,CmBp1BpC,iDhBpCoB,CgBmCpB,SzB6kEJ,CyBzkEE,0BACE,wBnB5BM,CmB6BN,oBzB2kEJ,CyBzkEI,yCAII,oQzBwkER,CyBpkEI,sCAII,4KzBmkER,CyB9jEE,+CAOI,+PANF,wBnBjDM,CmBkDN,oBzBikEJ,CyBxjEE,2BAEE,WAAY,CACZ,UnBk2BuC,CmBp2BvC,mBzB4jEJ,CyBnjEI,2FACE,cAAe,CACf,UzBqjEN,CyBviEA,aACE,kBzB0iEF,CyBxiEE,+BACE,qLAIA,0CACA,qBAAgC,ClBjHhC,kBkB+GA,mBHlHE,+CGsHF,CALA,SzB8iEJ,CsB3pEM,uCG0GJ,+BHzGM,etB8pEN,CACF,CyB5iEI,qCACE,0KzB8iEN,CyB3iEI,uCAMI,wKALF,wBzB8iEN,CyBpiEE,gCAEE,cAAe,CADf,mBzBuiEJ,CyBpiEI,kDAEE,aAAc,CADd,mBzBuiEN,CyBjiEA,mBACE,oBAAqB,CACrB,iBzBoiEF,CyBjiEA,WAEE,kBAAsB,CACtB,mBAAoB,CAFpB,iBzBsiEF,CyBhiEI,mDAEE,WAAY,CACZ,WnBspBwB,CmBxpBxB,mBzBoiEN,CyB3hEI,8EACE,0LzB8hEN,C0BjtEA,YAIE,eAAgB,CAChB,4BAA6B,CAH7B,gBACA,SAAU,CAFV,U1BwtEF,C0BltEE,kBACE,S1BotEJ,C0BhtEI,wCAA0B,gE1BmtE9B,C0BltEI,oCAA0B,gE1BqtE9B,C0BltEE,8BACE,Q1BotEJ,C0BjtEE,kCAIE,eAAgB,CH1BlB,wBjBkCQ,CoBNN,QpB6/BuC,CC1gCvC,mBmBSA,WpB8/BuC,CoB7/BvC,mBJbE,sGImBF,CARA,U1B0tEJ,CsBjuEM,uCIMJ,kCJLM,etBouEN,CACF,C0BrtEI,yCHjCF,wBvByvEF,C0BntEE,2CAKE,uCpBu+B8B,CoBt+B9B,wBAAyB,CnB7BzB,mBmB0BA,iBAAkB,CAClB,cpBu+B8B,CoBz+B9B,YpBw+B8B,CoBz+B9B,U1B2tEJ,C0BjtEE,8BAGE,eAAgB,CHpDlB,wBjBkCQ,CoBoBN,QpBm+BuC,CC1gCvC,mBmBoCA,WpBm+BuC,CgB1gCrC,sGI6CF,CAPA,U1BytEJ,CsB3vEM,uCIiCJ,8BJhCM,etB8vEN,CACF,C0BrtEI,qCH3DF,wBvBmxEF,C0BntEE,8BAKE,uCpB68B8B,CoB58B9B,wBAAyB,CnBvDzB,mBmBoDA,iBAAkB,CAClB,cpB68B8B,CoB/8B9B,YpB88B8B,CoB/8B9B,U1B2tEJ,C0BjtEE,qBACE,mB1BmtEJ,C0BjtEI,2CACE,0C1BmtEN,C0BhtEI,uCACE,0C1BktEN,C2BzyEA,eACE,iB3B4yEF,C2B1yEE,gGAGE,8CrBwiCoC,CqBtiCpC,gBrBuiCoC,CqBxiCpC,kD3B6yEJ,C2BzyEE,qBAYE,gDAPA,WAAY,CAFZ,MAAO,CAIP,eAAgB,CADhB,oBAKA,mBAAoB,CAVpB,iBAAkB,CAOlB,gBAAiB,CACjB,sBAAuB,CAPvB,KAAM,CAWN,oBAAqB,CLRnB,4DKSF,CAJA,kBAAmB,CANnB,S3BqzEJ,CsBhzEM,uCKTJ,qBLUM,etBmzEN,CACF,C2B7yEE,oEAEE,mB3B+yEJ,C2B7yEI,8FACE,iB3BgzEN,C2B7yEI,oMAGE,sBrB4gCkC,CqB7gClC,oB3BizEN,C2B7yEI,sGAEE,sBrBugCkC,CqBxgClC,oB3BizEN,C2B5yEE,4BAEE,sBrBigCoC,CqBlgCpC,oB3B+yEJ,C2BvyEI,mLACE,yCACA,0D3B4yEN,C2B1yEM,2MAME,kCrBg0BgC,CCh3BpC,sCoB+CI,UAAW,CADX,YrBm/BgC,CqBr/BhC,mBADA,iBAAkB,CAElB,U3BmzER,C2BzyEI,oDACE,yCACA,0D3B2yEN,C2BtyEI,6CACE,qC3BwyEN,C2BpyEE,2EAEE,a3BsyEJ,C2BpyEI,uFACE,uC3BuyEN,C4B93EA,aAIE,mBAAoB,CAFpB,YAAa,CACb,cAAe,CAFf,iBAAkB,CAIlB,U5Bi4EF,C4B/3EE,iFAIE,aAAc,CAEd,WAAY,CAHZ,iBAAkB,CAElB,Q5Bk4EJ,C4B73EE,0GAGE,S5B+3EJ,C4Bz3EE,kBACE,iBAAkB,CAClB,S5B23EJ,C4Bz3EI,wBACE,S5B23EN,C4Bh3EA,kBAEE,kBAAmB,CAQnB,sCtB06BsC,CsBz6BtC,2DrBtCE,sCqBkCF,0BtBm1BsC,CsBz1BtC,YAAa,CzBgPT,cALI,CyBvOR,etByjB4B,CsBxjB5B,etBgkB4B,CsBnkB5B,uBAKA,iBAAkB,CAClB,kB5Bs3EF,C4B12EA,kHrBhDI,yCJ4QE,iBALI,CyBnNR,kB5B+2EF,C4B12EA,kHrBzDI,yCJ4QE,iBALI,CyB1MR,oB5B+2EF,C4B12EA,0DAEE,kB5B62EF,C4Bv1EI,iqBrBzEA,6BADA,yBPg7EJ,C4Bz1EE,0IrBxEE,4BADA,yBqB0EA,2C5B61EJ,C4Bz1EE,uHrB7EE,4BADA,wBP46EJ,C6Bn8EE,gBAME,gCvBkjCqB,CuBvjCrB,YAAa,C1BoQX,gBALI,C0B7PN,iBvBu0BoC,CuBx0BpC,U7By8EJ,C6Bl8EE,eAWE,kCvBoiCqB,CC/jCrB,sCsB0BA,UvBqiCqB,CuB3iCrB,YAAa,C1BwPX,iBALI,C0BhPN,iBAFA,cAAe,CACf,qBALA,iBAAkB,CAClB,QAAS,CACT,S7B68EJ,C6Bh8EI,8HAEE,a7Bq8EN,C6Bp/EI,0DAyDI,yQAEA,yDADA,2BAA4B,CAE5B,4DAPF,8CvBuhCmB,CuBphCjB,kC7Bq8ER,C6B97EM,sEACE,8CvB4gCiB,CuBvgCf,uD7B47EV,C6BjgFI,0EAgFI,8EADA,kC7Bu7ER,C6BtgFI,wDAuFE,8C7Bm7EN,C6Bh7EQ,4NAEE,iRAEA,6DACA,sEAFA,sB7Bm7EV,C6B76EM,oEACE,8CvBw+BiB,CuBn+Bf,uD7B26EV,C6BphFI,sEAkHI,2B7Bs6ER,C6BxhFI,kEAyHE,8C7Bm6EN,C6Bj6EM,kFACE,2C7Bm6ER,C6Bh6EM,8EACE,uD7Bk6ER,C6B/5EM,sGACE,gC7Bi6ER,C6B55EI,qDACE,gB7B+5EN,C6BziFI,kVAoJM,S7B65EV,C6B7hFE,kBAME,kCvBkjCqB,CuBvjCrB,YAAa,C1BoQX,gBALI,C0B7PN,iBvBu0BoC,CuBx0BpC,U7BmiFJ,C6B5hFE,iBAWE,iCvBoiCqB,CC/jCrB,sCsB0BA,UvBqiCqB,CuB3iCrB,YAAa,C1BwPX,iBALI,C0BhPN,iBAFA,cAAe,CACf,qBALA,iBAAkB,CAClB,QAAS,CACT,S7BuiFJ,C6B1hFI,8IAEE,a7B+hFN,C6B9kFI,8DAyDI,sUAEA,yDADA,2BAA4B,CAE5B,4DAPF,gDvBuhCmB,CuBphCjB,kC7B+hFR,C6BxhFM,0EACE,gDvB4gCiB,CuBvgCf,sD7BshFV,C6B3lFI,8EAgFI,8EADA,kC7BihFR,C6BhmFI,4DAuFE,gD7B6gFN,C6B1gFQ,oOAEE,8UAEA,6DACA,sEAFA,sB7B6gFV,C6BvgFM,wEACE,gDvBw+BiB,CuBn+Bf,sD7BqgFV,C6B9mFI,0EAkHI,2B7BggFR,C6BlnFI,sEAyHE,gD7B6/EN,C6B3/EM,sFACE,6C7B6/ER,C6B1/EM,kFACE,sD7B4/ER,C6Bz/EM,0GACE,kC7B2/ER,C6Bt/EI,uDACE,gB7By/EN,C6BnoFI,8VAsJM,S7Bq/EV,C8B7oFA,KAEE,2BACA,4BACA,uB3BuRI,uBALI,C2BhRR,yBACA,yBACA,oCACA,wBACA,6CACA,kCACA,+CACA,wCACA,iFACA,+BACA,kFPhBA,iCOkCqB,CAFrB,mEvBjBE,0CuBUF,0BAKA,eAXA,oBAAqB,CAErB,sC3BsQI,iCALI,C2B/PR,sCACA,sCAJA,wDAMA,iBAAkB,CAClB,qBRfI,6HQwBJ,CALA,gBAAiB,CAFjB,qB9BkpFF,CsB/pFM,uCQhBN,KRiBQ,etBkqFN,CACF,C8B9oFE,WAGE,wCACA,8CAHA,+B9BkpFJ,C8B5oFE,sBAGE,kCACA,wCAFA,yB9B+oFJ,C8B1oFE,mBPpDA,uCOsDuB,CACrB,8CAME,0CARF,gCAGA,S9B6oFJ,C8BpoFE,8BACE,8CAME,0CALF,S9BuoFJ,C8B9nFE,mGAME,yCAGA,+CAJA,gC9B8nFJ,C8BvnFI,yKAKI,yC9BqnFR,C8BhnFE,sCAKI,yC9B8mFN,C8B1mFE,mDAKE,2CAEA,iDAJA,mCAKA,uCAJA,mB9B6mFJ,C8B7lFE,aC/GA,oBACA,oBACA,8BACA,0BACA,0BACA,oCACA,qCACA,2BACA,2BACA,qCACA,wDACA,6BACA,6BACA,sC/BgtFF,C8B9mFE,eC/GA,oBACA,oBACA,8BACA,0BACA,0BACA,oCACA,sCACA,2BACA,2BACA,qCACA,wDACA,6BACA,6BACA,sC/BiuFF,C8B/nFE,aC/GA,oBACA,oBACA,8BACA,0BACA,0BACA,oCACA,qCACA,2BACA,2BACA,qCACA,wDACA,6BACA,6BACA,sC/BkvFF,C8BhpFE,UC/GA,oBACA,oBACA,8BACA,0BACA,0BACA,oCACA,qCACA,2BACA,2BACA,qCACA,wDACA,6BACA,6BACA,sC/BmwFF,C8BjqFE,aC/GA,oBACA,oBACA,8BACA,0BACA,0BACA,oCACA,oCACA,2BACA,2BACA,qCACA,wDACA,6BACA,6BACA,sC/BoxFF,C8BlrFE,YC/GA,oBACA,oBACA,8BACA,0BACA,0BACA,oCACA,oCACA,2BACA,2BACA,qCACA,wDACA,6BACA,6BACA,sC/BqyFF,C8BnsFE,WC/GA,oBACA,oBACA,8BACA,0BACA,0BACA,oCACA,sCACA,2BACA,2BACA,qCACA,wDACA,6BACA,6BACA,sC/BszFF,C8BptFE,UC/GA,oBACA,oBACA,8BACA,0BACA,0BACA,oCACA,mCACA,2BACA,2BACA,qCACA,wDACA,6BACA,6BACA,sC/Bu0FF,C8B3sFE,qBChHA,uBACA,8BACA,0BACA,0BACA,oCACA,qCACA,2BACA,2BACA,qCACA,wDACA,gCACA,iCACA,uCACA,kB/B+zFF,C8B5tFE,uBChHA,uBACA,8BACA,0BACA,0BACA,oCACA,sCACA,2BACA,2BACA,qCACA,wDACA,gCACA,iCACA,uCACA,kB/Bg1FF,C8B7uFE,qBChHA,uBACA,8BACA,0BACA,0BACA,oCACA,oCACA,2BACA,2BACA,qCACA,wDACA,gCACA,iCACA,uCACA,kB/Bi2FF,C8B9vFE,kBChHA,uBACA,8BACA,0BACA,0BACA,oCACA,qCACA,2BACA,2BACA,qCACA,wDACA,gCACA,iCACA,uCACA,kB/Bk3FF,C8B/wFE,qBChHA,uBACA,8BACA,0BACA,0BACA,oCACA,oCACA,2BACA,2BACA,qCACA,wDACA,gCACA,iCACA,uCACA,kB/Bm4FF,C8BhyFE,oBChHA,uBACA,8BACA,0BACA,0BACA,oCACA,oCACA,2BACA,2BACA,qCACA,wDACA,gCACA,iCACA,uCACA,kB/Bo5FF,C8BjzFE,mBChHA,uBACA,8BACA,0BACA,0BACA,oCACA,sCACA,2BACA,2BACA,qCACA,wDACA,gCACA,iCACA,uCACA,kB/Bq6FF,C8Bl0FE,kBChHA,uBACA,8BACA,0BACA,0BACA,oCACA,mCACA,2BACA,2BACA,qCACA,wDACA,gCACA,iCACA,uCACA,kB/Bs7FF,C8Bv0FA,UACE,yBACA,oCACA,wBACA,kCACA,gDACA,wCACA,iDACA,yCACA,gCACA,2CACA,+BACA,qCAEA,yB9By0FF,C8B/zFE,wBACE,yB9Bi0FJ,C8B9zFE,gBACE,+B9Bg0FJ,C8BrzFA,2BCjJE,0BACA,wB5B8NI,0BALI,C4BvNR,iD/B08FF,C8BxzFA,2BCrJE,2BACA,0B5B8NI,2BALI,C4BvNR,iD/Bi9FF,CgCphGA,MVgBM,8BtBwgGN,CsBpgGM,uCUpBN,MVqBQ,etBugGN,CACF,CgC1hGE,iBACE,ShC4hGJ,CgCthGE,qBACE,YhCyhGJ,CgCrhGA,YACE,QAAS,CACT,eAAgB,CVDZ,2BtB0hGN,CsBthGM,uCULN,YVMQ,etByhGN,CACF,CgC3hGE,gCAEE,WAAY,CVNV,0BUOF,CAFA,OhC+hGJ,CsBhiGM,uCUAJ,gCVCM,etBmiGN,CACF,CiCxjGA,sEAME,iBjC2jGF,CiCxjGA,iBACE,kBjC2jGF,CkCniGI,uBA/BF,eAAgB,CAChB,mCAFA,oCADA,sBAqCI,UAAW,CAHX,oBAAqB,CACrB,kB5B6hBwB,C4B5hBxB,qBlC0iGN,CkCjhGI,6BACE,alCmhGN,CiCjkGA,eAEE,0BACA,8BACA,0BACA,+BACA,8B9BuQI,4BALI,C8BhQR,yCACA,mCACA,8DACA,oDACA,kDACA,yFACA,4DACA,sCACA,8CACA,8CACA,oDACA,qDACA,qCACA,sDACA,2DACA,kCACA,qCACA,mCACA,oCACA,sCAcA,2BAA4B,CAD5B,uCAEA,6E1BzCE,+C0BoCF,+BALA,YAAa,C9B6OT,sCALI,C8BjOR,eAAgB,CAJhB,QAAS,CAFT,uCACA,kEAJA,iBAAkB,CAQlB,eAAgB,CAPhB,iCjC6kGF,CiC9jGE,+BAEE,MAAO,CACP,qCAFA,QjCkkGJ,CiCxiGI,qBACE,mBjC2iGN,CiCziGM,qCAEE,MAAO,CADP,UjC4iGR,CiCviGI,mBACE,iBjC0iGN,CiCxiGM,mCAEE,SAAU,CADV,OjC2iGR,CcplGI,yBmB4BA,wBACE,mBjC4jGJ,CiC1jGI,wCAEE,MAAO,CADP,UjC6jGN,CiCxjGE,sBACE,iBjC0jGJ,CiCxjGI,sCAEE,SAAU,CADV,OjC2jGN,CACF,CcrmGI,yBmB4BA,wBACE,mBjC4kGJ,CiC1kGI,wCAEE,MAAO,CADP,UjC6kGN,CiCxkGE,sBACE,iBjC0kGJ,CiCxkGI,sCAEE,SAAU,CADV,OjC2kGN,CACF,CcrnGI,yBmB4BA,wBACE,mBjC4lGJ,CiC1lGI,wCAEE,MAAO,CADP,UjC6lGN,CiCxlGE,sBACE,iBjC0lGJ,CiCxlGI,sCAEE,SAAU,CADV,OjC2lGN,CACF,CcroGI,0BmB4BA,wBACE,mBjC4mGJ,CiC1mGI,wCAEE,MAAO,CADP,UjC6mGN,CiCxmGE,sBACE,iBjC0mGJ,CiCxmGI,sCAEE,SAAU,CADV,OjC2mGN,CACF,CiCjmGE,uCAEE,WAAY,CAEZ,wCADA,YAAa,CAFb,QjCsmGJ,CkCvrGI,+BAxBF,yBACA,mCAFA,oCADA,YAAa,CA8BT,UAAW,CAHX,oBAAqB,CACrB,kB5B6hBwB,C4B5hBxB,qBlC8rGN,CkCrqGI,qCACE,alCuqGN,CiCvmGE,wCAGE,SAAU,CAEV,sCADA,YAAa,CAFb,UAAW,CADX,KjC8mGJ,CkC5sGI,gCAjBF,qCACA,uBAFA,cAAe,CADf,kCAuBI,UAAW,CAHX,oBAAqB,CACrB,kB5B6hBwB,C4B5hBxB,qBlCmtGN,CkC1rGI,sCACE,alC4rGN,CiClnGI,gCACE,gBjConGN,CiC9mGE,0CAGE,SAAU,CAEV,uCADA,YAAa,CAFb,UAAW,CADX,KjCqnGJ,CkCpuGI,kCAIE,UAAW,CAHX,oBAAqB,CAenB,YAAa,CAdf,kB5B6hBwB,C4B5hBxB,qBlCuuGN,CkCvtGM,mCA7BJ,qCADA,wBADA,kCAmCM,UAAW,CAHX,oBAAqB,CACrB,mB5B0gBsB,C4BzgBtB,qBlCguGR,CkC1tGI,wCACE,alC4tGN,CiCjoGI,mCACE,gBjCmoGN,CiC5nGA,kBAIE,mDAHA,QAAS,CACT,6CAGA,SAAU,CAFV,ejCioGF,CiCznGA,eAUE,4BAA6B,CAC7B,QAAS,C1BtKP,sD0B+JF,UAAW,CAEX,oCALA,aAAc,CAId,e3Byb4B,C2B3b5B,4EAIA,kBAAmB,CACnB,qBACA,kBAAmB,CAPnB,UjCsoGF,CiC1nGE,0CVxLA,iDU4LuB,CAFrB,yCjC4nGJ,CiCvnGE,4CV/LA,kDUmMuB,CAFrB,2CACA,oBjCynGJ,CiCrnGE,gDAIE,4BAA6B,CAF7B,6CACA,mBjCunGJ,CiChnGA,oBACE,ajCmnGF,CiC/mGA,iBAKE,sCAJA,aAAc,C9BqEV,iBALI,C8B9DR,eAAgB,CADhB,gFAIA,kBjCknGF,CiC9mGA,oBAGE,oCAFA,aAAc,CACd,2EjCknGF,CiC7mGA,oBAEE,4BACA,yBACA,8DACA,2BACA,iCACA,oCACA,4DACA,qDACA,qCACA,sDACA,0CACA,kCjC+mGF,CmCr2GA,+BAGE,mBAAoB,CADpB,iBAAkB,CAElB,qBnCw2GF,CmCt2GE,yCAEE,aAAc,CADd,iBnC02GJ,CmCp2GE,kXAME,SnC42GJ,CmCv2GA,aACE,YAAa,CACb,cAAe,CACf,0BnC02GF,CmCx2GE,0BACE,UnC02GJ,CmCt2GA,W5BhBI,qCP03GJ,CmCt2GE,qFAEE,2CnCw2GJ,CmCp2GE,qJ5BTE,6BADA,yBPo3GJ,CmCh2GE,6G5BLE,4BADA,wBP42GJ,CmCn1GA,uBAEE,sBADA,sBnCu1GF,CmCp1GE,wGAGE,anCo1GJ,CmCj1GE,yCACE,cnCm1GJ,CmC/0GA,yEAEE,qBADA,qBnCm1GF,CmC/0GA,yEAEE,oBADA,oBnCm1GF,CmC9zGA,oBAEE,sBAAuB,CADvB,qBAAsB,CAEtB,sBnCi0GF,CmC/zGE,wDAEE,UnCi0GJ,CmC9zGE,4FAEE,0CnCg0GJ,CmC5zGE,qH5BzFE,4BADA,4BP25GJ,CmC5zGE,oF5B7GE,yBACA,yBP66GJ,CoCr8GA,KAEE,6BACA,+BAEA,4BACA,yCACA,qDACA,uDAGA,YAAa,CACb,cAAe,CAGf,eAAgB,CADhB,eAAgB,CADhB,cpCs8GF,CoCj8GA,UAOE,eAAgB,CAChB,QAAS,CAHT,+BAJA,aAAc,CjCuQV,sCALI,CiC/PR,2CAFA,kEAIA,qBdbI,iGtBo9GN,CsBh9GM,uCcGN,UdFQ,etBm9GN,CACF,CoCv8GE,gCAEE,oCpCw8GJ,CoCp8GE,wBAEE,iD3BhBoB,C2BepB,SpCu8GJ,CoCl8GE,sCAEE,wCAEA,cAAe,CADf,mBpCo8GJ,CoC37GA,UAEE,kDACA,kDACA,oDACA,2GACA,yDACA,+CACA,uGAGA,mFpC27GF,CoCz7GE,oBAEE,yD7B7CA,wDACA,yD6B2CA,sDpC87GJ,CoC17GI,oDAIE,wDADA,iBpC27GN,CoCt7GE,8DAGE,mDACA,yDAFA,0CpC07GJ,CoCr7GE,yB7B/DE,yBACA,0B6BgEA,mDpCw7GJ,CoC76GA,WAEE,qDACA,sCACA,qCpC+6GF,CoC56GE,qB7B5FE,+CP2gHJ,CoC36GE,uDb/GA,mDakHuB,CADrB,2CpC86GJ,CoCp6GA,eAEE,4BACA,yCACA,8DAGA,+BpCo6GF,CoCl6GE,yBAGE,qEADA,cAAe,CADf,epCs6GJ,CoCl6GI,8DAEE,gCpCm6GN,CoC/5GE,+DAIE,gCAAiC,CADjC,gDADA,epCm6GJ,CoCv5GE,wCAEE,aAAc,CACd,iBpC05GJ,CoCr5GE,kDAEE,YAAa,CACb,WAAY,CACZ,iBpCw5GJ,CoCl5GE,iEACE,UpCs5GJ,CoC54GE,uBACE,YpC+4GJ,CoC74GE,qBACE,apC+4GJ,CqC5kHA,QAEE,wBACA,6BACA,0DACA,+DACA,kEACA,8DACA,sCACA,kCACA,oCACA,6DACA,mEACA,sCACA,sCACA,sCACA,sCACA,qRACA,yEACA,0DACA,0CACA,4DAMA,kBAAmB,CAFnB,YAAa,CACb,cAAe,CAEf,6BAA8B,CAC9B,8DALA,iBrCilHF,CqCtkHE,oIAGE,kBAAmB,CAFnB,YAAa,CACb,iBAAkB,CAElB,6BrC6kHJ,CqCzjHA,cAKE,mClC0NI,0CALI,CkCvNR,+CADA,gDADA,6CAKA,qBACA,kBrC2jHF,CqCzjHE,wCAEE,wCrC0jHJ,CqChjHA,YAEE,0BACA,+BAEA,4BACA,2CACA,uDACA,6DAGA,YAAa,CACb,qBAAsB,CAGtB,eAAgB,CADhB,eAAgB,CADhB,crCijHF,CqC5iHI,wDAEE,mCrC6iHN,CqCziHE,2BACE,erC2iHJ,CqCliHA,aAGE,6BADA,oB/B6gCkC,C+B9gClC,iBrCuiHF,CqCniHE,yDAGE,mCrCqiHJ,CqCxhHA,iBAKE,kBAAmB,CAJnB,eAAgB,CAChB,WrC4hHF,CqCrhHA,gBAKE,4BAA6B,CAC7B,0E9BxIE,qD8BsIF,6BlCsII,4CALI,CkClIR,aAAc,CAFd,8EftII,8CtBqqHN,CsBjqHM,uCeiIN,gBfhIQ,etBoqHN,CACF,CqC3hHE,sBACE,oBrC6hHJ,CqC1hHE,sBAGE,sDADA,SAAU,CADV,oBrC8hHJ,CqCthHA,qBAKE,kDAEA,uBAA2B,CAD3B,2BAA4B,CAE5B,oBAAqB,CAPrB,oBAAqB,CAErB,YAAa,CACb,qBAAsB,CAFtB,WrC+hHF,CqCthHA,mBACE,wCACA,erCyhHF,CcnpHI,yBuBsIA,kBAEI,gBAAiB,CACjB,0BrCghHN,CqC9gHM,8BACE,kBrCghHR,CqC9gHQ,6CACE,iBrCghHV,CqC7gHQ,wCAEE,iDADA,iDrCghHV,CqC3gHM,qCACE,gBrC6gHR,CqC1gHM,mCACE,sBAAwB,CACxB,erC4gHR,CqCzgHM,kCACE,YrC2gHR,CqCxgHM,6BAQE,sCAAwC,CACxC,kBAAoB,CALpB,WAAY,CAEZ,qBAAuB,CAJvB,eAAgB,CAQhB,wBAA0B,Cf9N9B,eegOI,CALA,4BAA8B,CAF9B,oBAAsB,CAFtB,YrCihHR,CqCrgHQ,+CACE,YrCugHV,CqCpgHQ,6CACE,YAAa,CACb,WAAY,CAEZ,kBAAmB,CADnB,SrCugHV,CACF,CcnsHI,yBuBsIA,kBAEI,gBAAiB,CACjB,0BrC+jHN,CqC7jHM,8BACE,kBrC+jHR,CqC7jHQ,6CACE,iBrC+jHV,CqC5jHQ,wCAEE,iDADA,iDrC+jHV,CqC1jHM,qCACE,gBrC4jHR,CqCzjHM,mCACE,sBAAwB,CACxB,erC2jHR,CqCxjHM,kCACE,YrC0jHR,CqCvjHM,6BAQE,sCAAwC,CACxC,kBAAoB,CALpB,WAAY,CAEZ,qBAAuB,CAJvB,eAAgB,CAQhB,wBAA0B,Cf9N9B,eegOI,CALA,4BAA8B,CAF9B,oBAAsB,CAFtB,YrCgkHR,CqCpjHQ,+CACE,YrCsjHV,CqCnjHQ,6CACE,YAAa,CACb,WAAY,CAEZ,kBAAmB,CADnB,SrCsjHV,CACF,CclvHI,yBuBsIA,kBAEI,gBAAiB,CACjB,0BrC8mHN,CqC5mHM,8BACE,kBrC8mHR,CqC5mHQ,6CACE,iBrC8mHV,CqC3mHQ,wCAEE,iDADA,iDrC8mHV,CqCzmHM,qCACE,gBrC2mHR,CqCxmHM,mCACE,sBAAwB,CACxB,erC0mHR,CqCvmHM,kCACE,YrCymHR,CqCtmHM,6BAQE,sCAAwC,CACxC,kBAAoB,CALpB,WAAY,CAEZ,qBAAuB,CAJvB,eAAgB,CAQhB,wBAA0B,Cf9N9B,eegOI,CALA,4BAA8B,CAF9B,oBAAsB,CAFtB,YrC+mHR,CqCnmHQ,+CACE,YrCqmHV,CqClmHQ,6CACE,YAAa,CACb,WAAY,CAEZ,kBAAmB,CADnB,SrCqmHV,CACF,CcjyHI,0BuBsIA,kBAEI,gBAAiB,CACjB,0BrC6pHN,CqC3pHM,8BACE,kBrC6pHR,CqC3pHQ,6CACE,iBrC6pHV,CqC1pHQ,wCAEE,iDADA,iDrC6pHV,CqCxpHM,qCACE,gBrC0pHR,CqCvpHM,mCACE,sBAAwB,CACxB,erCypHR,CqCtpHM,kCACE,YrCwpHR,CqCrpHM,6BAQE,sCAAwC,CACxC,kBAAoB,CALpB,WAAY,CAEZ,qBAAuB,CAJvB,eAAgB,CAQhB,wBAA0B,Cf9N9B,eegOI,CALA,4BAA8B,CAF9B,oBAAsB,CAFtB,YrC8pHR,CqClpHQ,+CACE,YrCopHV,CqCjpHQ,6CACE,YAAa,CACb,WAAY,CAEZ,kBAAmB,CADnB,SrCopHV,CACF,CqC1sHI,eAEI,gBAAiB,CACjB,0BrC2sHR,CqCzsHQ,2BACE,kBrC2sHV,CqCzsHU,0CACE,iBrC2sHZ,CqCxsHU,qCAEE,iDADA,iDrC2sHZ,CqCtsHQ,kCACE,gBrCwsHV,CqCrsHQ,gCACE,sBAAwB,CACxB,erCusHV,CqCpsHQ,+BACE,YrCssHV,CqCnsHQ,0BAQE,sCAAwC,CACxC,kBAAoB,CALpB,WAAY,CAEZ,qBAAuB,CAJvB,eAAgB,CAQhB,wBAA0B,Cf9N9B,eegOI,CALA,4BAA8B,CAF9B,oBAAsB,CAFtB,YrC4sHV,CqChsHU,4CACE,YrCksHZ,CqC/rHU,0CACE,YAAa,CACb,WAAY,CAEZ,kBAAmB,CADnB,SrCksHZ,CqChrHA,yCAGE,sCACA,4CACA,+CACA,8BACA,6BACA,mCACA,mDrCmrHF,CqC5qHI,mFANF,uRrCsrHF,CsCx8HA,MAEE,wBACA,wBACA,gCACA,wBACA,2BACA,8CACA,0DACA,gDACA,uBACA,qFACA,+BACA,6BACA,qDACA,sBACA,mBACA,kBACA,+BACA,mCACA,+BASA,oBAAqB,CAErB,0BAA2B,CAD3B,mCAEA,qE/BjBE,2C+BaF,2BAJA,YAAa,CACb,qBAAsB,CAEtB,6BADA,WAAY,CAHZ,iBtCk9HF,CsCr8HE,SAEE,aAAc,CADd,ctCw8HJ,CsCp8HE,kBAEE,qBAAsB,CADtB,kBtCu8HJ,CsCp8HI,8B/BrBA,0DACA,2D+BqBE,kBtCw8HN,CsCp8HI,6B/BXA,6DADA,8D+BaE,qBtCw8HN,CsCj8HE,8DAEE,YtCm8HJ,CsC/7HA,WAKE,2BAFA,aAAc,CACd,uDtCi8HF,CsC77HA,YAEE,iCADA,2CtCi8HF,CsC77HA,eAGE,oCAFA,kDtCk8HF,CsC77HA,qCAJE,etCq8HF,CsCx7HE,sBACE,mCtC27HJ,CsCn7HA,aAIE,uCACA,4EAFA,+BADA,eAAgB,CADhB,iEtC07HF,CsCp7HE,yB/B7FE,uFPohIJ,CsCl7HA,aAGE,uCACA,yEAFA,+BADA,iEtCw7HF,CsCn7HE,wB/BxGE,uFP8hIJ,CsC56HA,kBAIE,eAAgB,CAFhB,oDACA,mDAFA,mDtCk7HF,CsC76HE,mCACE,mCACA,qCtC+6HJ,CsC36HA,mBAEE,mDADA,mDtC+6HF,CsC16HA,kB/BpII,iD+BwIF,QAAS,CACT,MAAO,CACP,2CALA,iBAAkB,CAElB,OAAQ,CADR,KtCk7HF,CsC16HA,yCAGE,UtC66HF,CsC16HA,wB/B3II,0DACA,0DP0jIJ,CsC36HA,2B/BjII,6DADA,6DPmjIJ,CsCp6HE,kBACE,yCtCu6HJ,CcliII,yBwBuHJ,YAQI,YAAa,CACb,kBtCu6HF,CsCp6HE,kBAEE,WAAY,CACZ,etCq6HJ,CsCn6HI,wBAEE,aAAc,CADd,atCs6HN,CsCh6HM,mC/B1KJ,6BADA,yBP+kIF,CsCj6HQ,iGAGE,yBtCk6HV,CsCh6HQ,oGAGE,4BtCi6HV,CsC75HM,oC/B3KJ,4BADA,wBP6kIF,CsC95HQ,mGAGE,wBtC+5HV,CsC75HQ,sGAGE,2BtC85HV,CACF,CuCnoIA,WAEE,0CACA,oCACA,0KACA,mDACA,mDACA,qDACA,0FACA,qCACA,kCACA,8CACA,6CACA,qPACA,sCACA,kDACA,8DACA,4PACA,4EACA,sCACA,mCACA,4DACA,oDvCqoIF,CuCjoIA,kBAGE,kBAAmB,CAMnB,4CACA,QAAS,ChCrBP,gBgCkBF,oCALA,YAAa,CpC+PT,cALI,CoChPR,oBAAqB,CAPrB,4EAJA,iBAAkB,CAOlB,eAAgB,CjBtBZ,yCiB2BJ,CATA,UvC6oIF,CsB3pIM,uCiBUN,kBjBTQ,etB8pIN,CACF,CuCvoIE,kCAEE,+CACA,gGAFA,sCvC2oIJ,CuCvoII,wCACE,qDACA,gDvCyoIN,CuCpoIE,wBAME,8CACA,2BAA4B,CAC5B,mDAHA,UAAW,CAJX,aAAc,CAEd,0CACA,gBAAiB,CjB7Cf,kDiBkDF,CAPA,wCvC6oIJ,CsBprIM,uCiBqCJ,wBjBpCM,etBurIN,CACF,CuCxoIE,wBACE,SvC0oIJ,CuCvoIE,wBAGE,oDADA,SAAU,CADV,SvC2oIJ,CuCroIA,kBACE,evCwoIF,CuCroIA,gBAEE,wCACA,+EAFA,+BvC0oIF,CuCtoIE,8BhC7DE,yDACA,yDPssIJ,CuCvoII,kEhChEA,+DACA,+DP0sIJ,CuCtoIE,oCACE,YvCwoIJ,CuCpoIE,6BhC3DE,4DADA,4DPosIJ,CuCpoIM,2EhC/DF,kEADA,kEPwsIJ,CuCnoII,iDhCpEA,4DADA,4DP4sIJ,CuCjoIA,gBACE,6EvCooIF,CuC3nIE,iCAEE,aAAc,ChC9Gd,gBgC6GA,cvCgoIJ,CuC5nII,6CAAgB,YvC+nIpB,CuC9nII,4CAAe,evCioInB,CuCtnII,yMhC7HA,ePyvIJ,CuCpnII,6CACE,sTACA,4TvCunIN,CwCjxIA,YAEE,4BACA,4BACA,mCAEA,qBACA,gCACA,wDACA,sCACA,4DASA,yCjCAE,iDiCNF,YAAa,CACb,cAAe,CrCiRX,wCALI,CqCxQR,eAAgB,CAFhB,iDADA,qExCqxIF,CwC3wIE,kCACE,gDxC8wIJ,CwC5wII,yCAGE,yCACA,yCAHA,UAAW,CACX,iDxCgxIN,CwC1wIE,wBACE,4CxC4wIJ,CyCjzIA,YAEE,kCACA,mCtC4RI,8BALI,CsCrRR,2CACA,qCACA,oDACA,oDACA,sDACA,uDACA,+CACA,0DACA,uDACA,gDACA,yEACA,kCACA,kCACA,4CACA,yDACA,mDACA,6DAGA,YAAa,CjCnBb,eAAgB,CADhB,cRu0IF,CyC/yIA,WAOE,yCACA,iFAHA,iCAHA,aAAc,CtCiQV,wCALI,CsC3PR,sEAFA,iBAAkB,CAKlB,qBnBlBI,6HtBu0IN,CsBn0IM,uCmBQN,WnBPQ,etBs0IN,CACF,CyCrzIE,iBAIE,+CACA,qDAHA,uCADA,SzC0zIJ,CyCnzIE,iBAGE,+CAEA,iDAHA,uCAEA,SnC2uCgC,CmC9uChC,SzCyzIJ,CyClzIE,qClBnDA,+CkBuDuB,CACrB,sDAFA,wCADA,SzCszIJ,CyChzIE,yCAIE,kDACA,wDAHA,0CACA,mBzCmzIJ,CyC5yIE,wCACE,2CzC+yIJ,CyC1yIM,kClC7BF,6DADA,yDP40IJ,CyCxyIM,iClCjDF,8DADA,0DP81IJ,CyC3xIA,eClGE,iCACA,kCvC0RI,iCALI,CuCnRR,wD1Ci4IF,CyC9xIA,eCtGE,iCACA,kCvC0RI,kCALI,CuCnRR,wD1Cw4IF,C2C14IA,OAEE,4BACA,4BxCuRI,2BALI,CwChRR,2BACA,sBACA,iDpCOE,4CoCCF,4BALA,oBAAqB,CxCgRjB,mCALI,CwCxQR,wCACA,aAAc,CAHd,4DAKA,iBAAkB,CAElB,uBAAwB,CADxB,kB3C44IF,C2Ct4IE,aACE,Y3Cw4IJ,C2Cn4IA,YACE,iBAAkB,CAClB,Q3Cs4IF,C4Ct6IA,OAEE,0BACA,0BACA,0BACA,8BACA,yBACA,oCACA,4EACA,iDACA,8BAOA,oCACA,8BrCHE,4CqCCF,4BADA,4CADA,4DADA,iB5C46IF,C4Cl6IA,eAEE,a5Co6IF,C4Ch6IA,YAEE,iCADA,e5Co6IF,C4C35IA,mBACE,kB5C85IF,C4C35IE,8BAKE,qBAJA,iBAAkB,CAElB,OAAQ,CADR,KAAM,CAEN,S5C85IJ,C4Cr5IE,eACE,iDACA,0CACA,wDACA,qD5Cw5IJ,C4C55IE,iBACE,mDACA,4CACA,0DACA,uD5C+5IJ,C4Cn6IE,eACE,iDACA,0CACA,wDACA,qD5Cs6IJ,C4C16IE,YACE,8CACA,uCACA,qDACA,kD5C66IJ,C4Cj7IE,eACE,iDACA,0CACA,wDACA,qD5Co7IJ,C4Cx7IE,cACE,gDACA,yCACA,uDACA,oD5C27IJ,C4C/7IE,aACE,+CACA,wCACA,sDACA,mD5Ck8IJ,C4Ct8IE,YACE,8CACA,uCACA,qDACA,kD5Cy8IJ,C6CrgJE,gCACE,GAAK,0B7CygJP,CACF,C6CrgJA,4BAGE,0B1CkRI,+BALI,C0C3QR,wCACA,oDACA,oDACA,6BACA,6BACA,6CAOA,uCtCRE,+CsCIF,YAAa,C1CwQT,sCALI,C0ClQR,iCACA,e7CugJF,C6ChgJA,cAQE,2CAHA,mCAJA,YAAa,CACb,qBAAsB,CACtB,sBAAuB,CACvB,eAAgB,CAEhB,iBAAkB,CvBtBd,4CuByBJ,CAFA,kB7CqgJF,CsBxhJM,uCuBYN,cvBXQ,etB2hJN,CACF,C6CrgJA,sBtBAE,sKsBEA,mE7CwgJF,C6CrgJA,4BACE,gB7CwgJF,C6CrgJA,0CACE,U7CwgJF,C6CpgJE,uBACE,iD7CugJJ,C6CpgJM,uCAJJ,uBAKM,c7CugJN,CACF,C8CnkJA,YAEE,2CACA,qCACA,oDACA,oDACA,sDACA,oCACA,sCACA,uDACA,4DACA,sDACA,yDACA,wDACA,yDACA,8CACA,kCACA,kCACA,4CvCHE,iDuCMF,YAAa,CACb,qBAAsB,CAItB,eAAgB,CADhB,c9CmkJF,C8C9jJA,qBAEE,qBAAsB,CADtB,oB9CkkJF,C8C/jJE,6CAEE,kCAAoC,CACpC,yB9CgkJJ,C8CvjJA,wBAEE,wCACA,kBAAmB,CAFnB,U9C4jJF,C8CvjJE,4DAKE,sDAFA,8CACA,oBAAqB,CAFrB,S9C2jJJ,C8CrjJE,+BAEE,uDADA,8C9CwjJJ,C8C/iJA,iBAME,yCACA,iFAHA,iCAFA,aAAc,CACd,gFAFA,iBAAkB,CAIlB,oB9CojJF,C8ChjJE,6BvCvDE,+BACA,+BP0mJJ,C8ChjJE,4BvC5CE,kCADA,kCPimJJ,C8ChjJE,oDAIE,kDAFA,0CACA,mB9CkjJJ,C8C7iJE,wBAGE,gDACA,sDAFA,wCADA,S9CkjJJ,C8C3iJE,kCACE,kB9C6iJJ,C8C3iJI,yCAEE,mDADA,qD9C8iJN,C8ChiJI,uBACE,kB9CmiJN,C8ChiJQ,qEvCvDJ,6DAZA,yBPumJJ,C8C/hJQ,qEvC5DJ,4BAZA,0DP2mJJ,C8C9hJQ,+CACE,Y9CgiJV,C8C7hJQ,yDAEE,mBAAoB,CADpB,kD9CgiJV,C8C7hJU,gEAEE,oDADA,sD9CgiJZ,CcrnJI,yBgC8DA,0BACE,kB9C2jJJ,C8CxjJM,wEvCvDJ,6DAZA,yBP+nJF,C8CvjJM,wEvC5DJ,4BAZA,0DPmoJF,C8CtjJM,kDACE,Y9CwjJR,C8CrjJM,4DAEE,mBAAoB,CADpB,kD9CwjJR,C8CrjJQ,mEAEE,oDADA,sD9CwjJV,CACF,Cc9oJI,yBgC8DA,0BACE,kB9CmlJJ,C8ChlJM,wEvCvDJ,6DAZA,yBPupJF,C8C/kJM,wEvC5DJ,4BAZA,0DP2pJF,C8C9kJM,kDACE,Y9CglJR,C8C7kJM,4DAEE,mBAAoB,CADpB,kD9CglJR,C8C7kJQ,mEAEE,oDADA,sD9CglJV,CACF,CctqJI,yBgC8DA,0BACE,kB9C2mJJ,C8CxmJM,wEvCvDJ,6DAZA,yBP+qJF,C8CvmJM,wEvC5DJ,4BAZA,0DPmrJF,C8CtmJM,kDACE,Y9CwmJR,C8CrmJM,4DAEE,mBAAoB,CADpB,kD9CwmJR,C8CrmJQ,mEAEE,oDADA,sD9CwmJV,CACF,Cc9rJI,0BgC8DA,0BACE,kB9CmoJJ,C8ChoJM,wEvCvDJ,6DAZA,yBPusJF,C8C/nJM,wEvC5DJ,4BAZA,0DP2sJF,C8C9nJM,kDACE,Y9CgoJR,C8C7nJM,4DAEE,mBAAoB,CADpB,kD9CgoJR,C8C7nJQ,mEAEE,oDADA,sD9CgoJV,CACF,C8ClnJA,kBvChJI,ePqwJJ,C8ClnJE,mCACE,kD9ConJJ,C8ClnJI,8CACE,qB9ConJN,C8CvmJE,yBACE,sDACA,+CACA,6DACA,4DACA,gEACA,6DACA,iEACA,yDACA,0DACA,mE9C0mJJ,C8CpnJE,2BACE,wDACA,iDACA,+DACA,4DACA,kEACA,6DACA,mEACA,2DACA,4DACA,qE9CunJJ,C8CjoJE,yBACE,sDACA,+CACA,6DACA,4DACA,gEACA,6DACA,iEACA,yDACA,0DACA,mE9CooJJ,C8C9oJE,sBACE,mDACA,4CACA,0DACA,4DACA,6DACA,6DACA,8DACA,sDACA,uDACA,gE9CipJJ,C8C3pJE,yBACE,sDACA,+CACA,6DACA,4DACA,gEACA,6DACA,iEACA,yDACA,0DACA,mE9C8pJJ,C8CxqJE,wBACE,qDACA,8CACA,4DACA,4DACA,+DACA,6DACA,gEACA,wDACA,yDACA,kE9C2qJJ,C8CrrJE,uBACE,oDACA,6CACA,2DACA,4DACA,8DACA,6DACA,+DACA,uDACA,wDACA,iE9CwrJJ,C8ClsJE,sBACE,mDACA,4CACA,0DACA,4DACA,6DACA,6DACA,8DACA,sDACA,uDACA,gE9CqsJJ,C+Cj4JA,WAEE,0BACA,oVACA,2BACA,kCACA,oEACA,+BACA,qCACA,uEAQA,wEACA,QAAS,CxCJP,sBwCFF,sBAAuB,CAEvB,UzCopD2B,CyC9oD3B,oCALA,cAFA,S/Cw4JF,C+C93JE,4BAPA,+B/C04JF,C+Cn4JE,iBAGE,0CADA,oB/Ci4JJ,C+C73JE,iBAEE,4CACA,0CAFA,S/Ci4JJ,C+C53JE,wCAIE,6CAFA,mBAAoB,CACpB,gB/C83JJ,C+C/2JI,iDATF,uC/Cg4JF,CgDj7JA,OAEE,uBACA,6BACA,4BACA,0BACA,2B7CyRI,6BALI,C6ClRR,mBACA,+CACA,+CACA,2DACA,iDACA,2CACA,kDACA,sDACA,kEASA,2BAA4B,CAD5B,oCAEA,uEzCPE,4CyCQF,sCALA,4B7CyQI,mCALI,C6CtQR,cAAe,CAGf,mBAAoB,CAJpB,+BhD07JF,CgD/6JE,eACE,ShDi7JJ,CgD96JE,kBACE,YhDg7JJ,CgD56JA,iBACE,uBAKA,cAAe,CACf,mBAAoB,CAJpB,iBAAkB,CAElB,iBAAkB,CADlB,8BhDi7JF,CgD56JE,mCACE,qChD86JJ,CgD16JA,cAEE,kBAAmB,CAInB,2BAA4B,CAD5B,2CAEA,qFzChCE,0FACA,2FyC4BF,mCAHA,YAAa,CAEb,2DhDm7JF,CgD56JE,yBAEE,sCADA,gDhD+6JJ,CgD16JA,YAEE,oBAAqB,CADrB,iChD86JF,CiD3+JA,OAEE,uBACA,uBACA,wBACA,yBACA,mBACA,gCACA,2DACA,+CACA,oDACA,8CACA,yFACA,iCACA,iCACA,oCACA,sDACA,sDACA,iCACA,6BACA,uBACA,sDACA,sDAOA,YAAa,CAEb,WAAY,CAJZ,MAAO,CASP,SAAU,CAJV,iBAAkB,CAClB,eAAgB,CARhB,cAAe,CACf,KAAM,CAIN,UAAW,CAFX,8BjDi/JF,CiDl+JA,cAGE,8BAEA,mBAAoB,CAJpB,iBAAkB,CAClB,UjDu+JF,CiDj+JE,0BAEE,2B3Ck8CgC,CgBh/C9B,iCtBihKN,CsB7gKM,uC2BwCJ,0B3BvCM,etBghKN,CACF,CiDt+JE,0BACE,cjDw+JJ,CiDp+JE,kCACE,qBjDs+JJ,CiDl+JA,yBACE,4CjDq+JF,CiDn+JE,wCACE,eAAgB,CAChB,ejDq+JJ,CiDl+JE,qCACE,ejDo+JJ,CiDh+JA,uBAEE,kBAAmB,CADnB,YAAa,CAEb,gDjDm+JF,CiD/9JA,eASE,2BAA4B,CAD5B,oCAEA,uE1CrFE,4C0CiFF,4BAJA,YAAa,CACb,qBAAsB,CAWtB,SAAU,CAPV,mBAAoB,CANpB,iBAAkB,CAGlB,UjDy+JF,CiD39JA,gBAEE,0BACA,sBACA,0BC5GA,sCD+G4D,CChH5D,YAAa,CAHb,MAAO,CAFP,cAAe,CACf,KAAM,CAGN,WAAY,CADZ,iClDmlKF,CkD7kKE,qBAAS,SlDglKX,CkD/kKE,qBAAS,kClDklKX,CiDl+JA,cAGE,kBAAmB,CAEnB,4F1CrGE,2DACA,4D0CgGF,YAAa,CACb,aAAc,CAEd,sCjDw+JF,CiDp+JE,yBAEE,6IADA,2FjDu+JJ,CiDj+JA,aAEE,8CADA,ejDq+JF,CiD/9JA,YAIE,aAAc,CACd,gCAJA,iBjDo+JF,CiD59JA,cAIE,kBAAmB,CAGnB,2C1CvHE,8DADA,+D0CyHF,yFAPA,YAAa,CACb,aAAc,CACd,cAAe,CAEf,wBAAyB,CACzB,qEjDm+JF,CiD39JE,gBACE,0CjD69JJ,CcxkKI,yBmCiHF,OACE,0BACA,0CjD29JF,CiDv9JA,cAGE,gBAAiB,CADjB,iBAAkB,CADlB,+BjD29JF,CiDt9JA,UACE,sBjDw9JF,CACF,CcvlKI,yBmCmIF,oBAEE,sBjDu9JF,CACF,Cc7lKI,0BmC0IF,UACE,uBjDs9JF,CACF,CiD78JI,kBAGE,WAAY,CACZ,QAAS,CAFT,cAAe,CADf,WjDk9JN,CiD78JM,iCAEE,QAAS,C1CzMb,gB0CwMI,WjDi9JR,CiD58JM,gE1C7MF,eP6pKJ,CiD38JM,8BACE,ejD68JR,CcvmKI,4BmCwIA,0BAGE,WAAY,CACZ,QAAS,CAFT,cAAe,CADf,WjDs+JJ,CiDj+JI,yCAEE,QAAS,C1CzMb,gB0CwMI,WjDq+JN,CiDh+JI,gF1C7MF,ePirKF,CiD/9JI,sCACE,ejDi+JN,CACF,Cc5nKI,4BmCwIA,0BAGE,WAAY,CACZ,QAAS,CAFT,cAAe,CADf,WjD0/JJ,CiDr/JI,yCAEE,QAAS,C1CzMb,gB0CwMI,WjDy/JN,CiDp/JI,gF1C7MF,ePqsKF,CiDn/JI,sCACE,ejDq/JN,CACF,CchpKI,4BmCwIA,0BAGE,WAAY,CACZ,QAAS,CAFT,cAAe,CADf,WjD8gKJ,CiDzgKI,yCAEE,QAAS,C1CzMb,gB0CwMI,WjD6gKN,CiDxgKI,gF1C7MF,ePytKF,CiDvgKI,sCACE,ejDygKN,CACF,CcpqKI,6BmCwIA,0BAGE,WAAY,CACZ,QAAS,CAFT,cAAe,CADf,WjDkiKJ,CiD7hKI,yCAEE,QAAS,C1CzMb,gB0CwMI,WjDiiKN,CiD5hKI,gF1C7MF,eP6uKF,CiD3hKI,sCACE,ejD6hKN,CACF,CmDnwKA,SAEE,yBACA,6BACA,8BACA,+BACA,sBhDwRI,+BALI,CgDjRR,qCACA,yCACA,mDACA,yBACA,gCACA,iCAYA,oBAAqB,CARrB,aAAc,CCjBd,qC9C+lB4B,CHjUxB,qCALI,CiDvRR,iBAAkB,CAClB,e9CwmB4B,C8CjmB5B,qBAAsB,CAItB,eAAgB,CAVhB,e9C+mB4B,C6CjmB5B,gCAQA,SAAU,CCrBV,eAAgB,CAChB,gBAAiB,CACjB,oBAAqB,CACrB,gBAAiB,CACjB,mBAAoB,CAGpB,kBAAmB,CADnB,iBAAkB,CAElB,mBAAoB,CDGpB,gCnDqxKF,CmDzwKE,cAAS,iCnD4wKX,CmD1wKE,wBACE,aAAc,CAEd,sCADA,mCnD6wKJ,CmD1wKI,+BAGE,wBAAyB,CACzB,kBAAmB,CAFnB,UAAW,CADX,iBnD+wKN,CmDvwKA,2FACE,8CnD0wKF,CmDxwKE,yGAGE,sCADA,qFADA,QnD4wKJ,CmDrwKA,6FAGE,qCAFA,6CACA,oCnD0wKF,CmDvwKE,2GAGE,wCADA,4HADA,UnD2wKJ,CmDnwKA,iGACE,2CnDuwKF,CmDrwKE,+GAGE,yCADA,qFADA,WnDywKJ,CmDlwKA,8FAGE,qCAFA,8CACA,oCnDuwKF,CmDpwKE,4GAGE,uCADA,4HADA,SnDwwKJ,CmDhvKA,eAKE,sC5CjGE,8C4C+FF,8BAFA,sCACA,gEAEA,iBnDsvKF,CqDz2KA,SAEE,yBACA,6BlD4RI,+BALI,CkDrRR,kCACA,iDACA,6DACA,sDACA,2FACA,6CACA,mCACA,qClDmRI,kCALI,CkD5QR,kCACA,8CACA,iCACA,iCACA,6CACA,8BACA,iCACA,yDAWA,oBAAqB,CAErB,2BAA4B,CAD5B,sCAEA,2E9ChBE,8C8CMF,aAAc,CDxBd,qC9C+lB4B,CHjUxB,qCALI,CiDvRR,iBAAkB,CAClB,e9CwmB4B,C8CjmB5B,qBAAsB,CAItB,eAAgB,CAVhB,e9C+mB4B,C+C1lB5B,sCDpBA,eAAgB,CAChB,gBAAiB,CACjB,oBAAqB,CACrB,gBAAiB,CACjB,mBAAoB,CAGpB,kBAAmB,CADnB,iBAAkB,CAElB,mBAAoB,CCUpB,gCrD+3KF,CqDh3KE,wBACE,aAAc,CAEd,sCADA,mCrDm3KJ,CqDh3KI,6DAOE,0BAAe,CAHf,UAAW,CADX,aAAc,CADd,iBrDs3KN,CqD32KE,2FACE,iFrD82KJ,CqD52KI,gNAEE,oFrD62KN,CqD12KI,yGAEE,gDADA,QrD62KN,CqDz2KI,uGAEE,sCADA,qCrD42KN,CqDp2KE,6FAGE,qCAFA,gFACA,oCrDy2KJ,CqDt2KI,oNAEE,2HrDu2KN,CqDp2KI,2GAEE,kDADA,MrDu2KN,CqDn2KI,yGAEE,wCADA,mCrDs2KN,CqD71KE,iGACE,8ErDi2KJ,CqD/1KI,4NAEE,oFrDg2KN,CqD71KI,+GAEE,mDADA,KrDg2KN,CqD51KI,6GAEE,yCADA,kCrD+1KN,CqDz1KE,iHAQE,+EADA,UAAW,CAHX,aAAc,CADd,QAAS,CAGT,oDALA,iBAAkB,CAClB,KAAM,CAGN,mCrD81KJ,CqDr1KE,8FAGE,qCAFA,iFACA,oCrD01KJ,CqDv1KI,sNAEE,2HrDw1KN,CqDr1KI,4GAEE,iDADA,OrDw1KN,CqDp1KI,0GAEE,uCADA,oCrDu1KN,CqD/zKA,gBAKE,6CACA,kF9C5JE,6DACA,8D8CyJF,qClDyGI,4CALI,CkDtGR,eAAgB,CADhB,6ErD00KF,CqDl0KE,sBACE,YrDo0KJ,CqDh0KA,cAEE,mCADA,yErDo0KF,CsDx/KA,UACE,iBtD2/KF,CsDx/KA,wBACE,kBtD2/KF,CsDx/KA,gBAGE,eAAgB,CAFhB,iBAAkB,CAClB,UtD4/KF,CuDjhLE,sBAEE,UAAW,CACX,UAAW,CAFX,avDqhLJ,CsD5/KA,eAME,0BAA2B,CAJ3B,YAAa,CACb,UAAW,CAEX,kBAAmB,CAJnB,iBAAkB,ChCbd,oCgCmBJ,CAHA,UtDkgLF,CsB9gLM,uCgCQN,ehCPQ,etBihLN,CACF,CsDjgLA,8DAGE,atDogLF,CsDjgLA,wEAEE,0BtDogLF,CsDjgLA,wEAEE,2BtDogLF,CsD3/KE,8BACE,SAAU,CAEV,cAAe,CADf,2BtD+/KJ,CsD3/KE,iJAIE,SAAU,CADV,StD8/KJ,CsD1/KE,oFAGE,SAAU,ChC5DR,yBgC6DF,CAFA,StD8/KJ,CsBrjLM,uCgCqDJ,oFhCpDM,etByjLN,CACF,CsDz/KA,8CAQE,kBAAmB,CAMnB,eAAgB,CAChB,QAAS,CAXT,QAAS,CAQT,UhD1FS,CgDqFT,YAAa,CAEb,sBAAuB,CAOvB,UhD6gDmC,CgDlhDnC,SAAU,CATV,iBAAkB,CAWlB,iBAAkB,CAVlB,KAAM,ChCzEF,4BgCuFJ,CAPA,ShDkhDmC,CgDvhDnC,StDugLF,CsB9kLM,uCgCkEN,8ChCjEQ,etBklLN,CACF,CsD9/KE,oHAEE,UhDpGO,CgDuGP,UhDqgDiC,CgDtgDjC,SAAU,CADV,oBtDmgLJ,CsD9/KA,uBACE,MtDigLF,CsD9/KA,uBACE,OtDigLF,CsD5/KA,wDAME,uBAAwB,CADxB,2BAA4B,CAE5B,yBAA0B,CAL1B,oBAAqB,CAErB,WhDqgDmC,CgDtgDnC,UtDmgLF,CsD5/KA,4BACE,sRtD+/KF,CsD7/KA,4BACE,uRtDggLF,CsDx/KA,qBAGE,QAAS,CAGT,YAAa,CACb,sBAAuB,CAHvB,MAAO,CAOP,kBAAmB,CACnB,ehDo9CmC,CgDt9CnC,gBhDs9CmC,CgDx9CnC,SAAU,CAPV,iBAAkB,CAClB,OAAQ,CAGR,StDigLF,CsDx/KE,sCAWE,2BAA4B,CAD5B,qBhDlKO,CgDoKP,QAAS,CAGT,qCADA,kCAbA,sBAAuB,CAQvB,cAAe,CAPf,aAAc,CAEd,UhDo9CiC,CgDj9CjC,ehDm9CiC,CgDp9CjC,gBhDo9CiC,CgD18CjC,UhD28CiC,CgDt9CjC,SAAU,CAGV,kBAAmB,ChCxJjB,2BgCiKF,CAdA,UtDugLJ,CsBtpLM,uCgC4IJ,sChC3IM,etBypLN,CACF,CsD3/KE,6BACE,StD6/KJ,CsDp/KA,kBAGE,chDk8CmC,CgD97CnC,UhD7LS,CgD0LT,SAEA,sBhD87CmC,CgD/7CnC,mBhD+7CmC,CgDn8CnC,iBAAkB,CAClB,UAMA,iBtDu/KF,CsDj/KE,sFAEE,+BtDo/KJ,CsDj/KE,qDACE,qBtDm/KJ,CsDh/KE,iCACE,UtDk/KJ,CsD5/KE,0OAEE,+BtDggLJ,CsD7/KE,yIACE,qBtD+/KJ,CsD5/KE,iGACE,UtD8/KJ,CwDhtLA,8BAQE,6FADA,iBAAkB,CALlB,oBAAqB,CAErB,gCACA,gDAFA,6BxDutLF,CwD9sLA,0BACE,GAAK,uBxDktLL,CACF,CwD/sLA,gBAEE,wBACA,yBACA,qCACA,iCACA,mCACA,2CAGA,gCACA,2CAA+B,CAA/B,6DxD8sLF,CwD3sLA,mBAEE,wBACA,yBACA,+BxD6sLF,CwDpsLA,wBACE,GACE,kBxDusLF,CwDrsLA,IACE,SAAU,CACV,cxDusLF,CACF,CwDnsLA,cAEE,wBACA,yBACA,qCACA,mCACA,yCAGA,6BAA8B,CAC9B,SxDksLF,CwD/rLA,iBACE,wBACA,wBxDksLF,CwD9rLE,uCACE,8BAEE,iCxDisLJ,CACF,CyDjxLA,mEAEE,2BACA,2BACA,2BACA,8BACA,8BACA,0CACA,oCACA,mDACA,+DACA,kDACA,qDACA,oCzDkxLF,CcrtLI,4B2C5CF,cAWI,2BAA4B,CAD5B,wCAPA,QAAS,CAKT,gCAHA,YAAa,CACb,qBAAsB,CACtB,cAAe,CAKf,SAAU,CAVV,cAAe,CnClBf,yCmC8BA,CALA,iBAAkB,CALlB,kCzD6wLJ,CACF,CsB9xLM,gEmCYJ,cnCXM,etBiyLN,CACF,Cc3uLI,4B2C5BE,8BAIE,qFAFA,MAAO,CADP,KAAM,CAIN,2BAA4B,CAF5B,+BzD4wLN,CyDvwLI,4BAIE,oFAFA,OAAQ,CADR,KAAM,CAIN,0BAA2B,CAF3B,+BzD2wLN,CyDtwLI,4BAME,sFALA,KAAM,CAMN,2BzDwwLN,CyDrwLI,2DANE,kCADA,MAAO,CAEP,eAAgB,CAHhB,OzDqxLN,CyD7wLI,+BAKE,mFACA,0BzDuwLN,CyDpwLI,sDAEE,czDqwLN,CyDlwLI,8DAGE,kBzDkwLN,CACF,Cc/xLI,yB2C/BF,cAiEM,2BACA,8BACA,sCzDiwLN,CyD/vLM,gCACE,YzDiwLR,CyD9vLM,8BAME,sCAAwC,CALxC,YAAa,CACb,WAAY,CAEZ,kBAAmB,CADnB,SzDkwLR,CACF,CcnyLI,4B2C5CF,cAWI,2BAA4B,CAD5B,wCAPA,QAAS,CAKT,gCAHA,YAAa,CACb,qBAAsB,CACtB,cAAe,CAKf,SAAU,CAVV,cAAe,CnClBf,yCmC8BA,CALA,iBAAkB,CALlB,kCzD21LJ,CACF,CsB52LM,gEmCYJ,cnCXM,etB+2LN,CACF,CczzLI,4B2C5BE,8BAIE,qFAFA,MAAO,CADP,KAAM,CAIN,2BAA4B,CAF5B,+BzD01LN,CyDr1LI,4BAIE,oFAFA,OAAQ,CADR,KAAM,CAIN,0BAA2B,CAF3B,+BzDy1LN,CyDp1LI,4BAME,sFALA,KAAM,CAMN,2BzDs1LN,CyDn1LI,2DANE,kCADA,MAAO,CAEP,eAAgB,CAHhB,OzDm2LN,CyD31LI,+BAKE,mFACA,0BzDq1LN,CyDl1LI,sDAEE,czDm1LN,CyDh1LI,8DAGE,kBzDg1LN,CACF,Cc72LI,yB2C/BF,cAiEM,2BACA,8BACA,sCzD+0LN,CyD70LM,gCACE,YzD+0LR,CyD50LM,8BAME,sCAAwC,CALxC,YAAa,CACb,WAAY,CAEZ,kBAAmB,CADnB,SzDg1LR,CACF,Ccj3LI,4B2C5CF,cAWI,2BAA4B,CAD5B,wCAPA,QAAS,CAKT,gCAHA,YAAa,CACb,qBAAsB,CACtB,cAAe,CAKf,SAAU,CAVV,cAAe,CnClBf,yCmC8BA,CALA,iBAAkB,CALlB,kCzDy6LJ,CACF,CsB17LM,gEmCYJ,cnCXM,etB67LN,CACF,Ccv4LI,4B2C5BE,8BAIE,qFAFA,MAAO,CADP,KAAM,CAIN,2BAA4B,CAF5B,+BzDw6LN,CyDn6LI,4BAIE,oFAFA,OAAQ,CADR,KAAM,CAIN,0BAA2B,CAF3B,+BzDu6LN,CyDl6LI,4BAME,sFALA,KAAM,CAMN,2BzDo6LN,CyDj6LI,2DANE,kCADA,MAAO,CAEP,eAAgB,CAHhB,OzDi7LN,CyDz6LI,+BAKE,mFACA,0BzDm6LN,CyDh6LI,sDAEE,czDi6LN,CyD95LI,8DAGE,kBzD85LN,CACF,Cc37LI,yB2C/BF,cAiEM,2BACA,8BACA,sCzD65LN,CyD35LM,gCACE,YzD65LR,CyD15LM,8BAME,sCAAwC,CALxC,YAAa,CACb,WAAY,CAEZ,kBAAmB,CADnB,SzD85LR,CACF,Cc/7LI,6B2C5CF,cAWI,2BAA4B,CAD5B,wCAPA,QAAS,CAKT,gCAHA,YAAa,CACb,qBAAsB,CACtB,cAAe,CAKf,SAAU,CAVV,cAAe,CnClBf,yCmC8BA,CALA,iBAAkB,CALlB,kCzDu/LJ,CACF,CsBxgMM,iEmCYJ,cnCXM,etB2gMN,CACF,Ccr9LI,6B2C5BE,8BAIE,qFAFA,MAAO,CADP,KAAM,CAIN,2BAA4B,CAF5B,+BzDs/LN,CyDj/LI,4BAIE,oFAFA,OAAQ,CADR,KAAM,CAIN,0BAA2B,CAF3B,+BzDq/LN,CyDh/LI,4BAME,sFALA,KAAM,CAMN,2BzDk/LN,CyD/+LI,2DANE,kCADA,MAAO,CAEP,eAAgB,CAHhB,OzD+/LN,CyDv/LI,+BAKE,mFACA,0BzDi/LN,CyD9+LI,sDAEE,czD++LN,CyD5+LI,8DAGE,kBzD4+LN,CACF,CczgMI,0B2C/BF,cAiEM,2BACA,8BACA,sCzD2+LN,CyDz+LM,gCACE,YzD2+LR,CyDx+LM,8BAME,sCAAwC,CALxC,YAAa,CACb,WAAY,CAEZ,kBAAmB,CADnB,SzD4+LR,CACF,CyDzjME,WAWI,2BAA4B,CAD5B,wCAPA,QAAS,CAKT,gCAHA,YAAa,CACb,qBAAsB,CACtB,cAAe,CAKf,SAAU,CAVV,cAAe,CnClBf,yCmC8BA,CALA,iBAAkB,CALlB,kCzDokMN,CsBplMM,uCmCYJ,WnCXM,etBulMN,CACF,CyD7jMM,2BAIE,qFAFA,MAAO,CADP,KAAM,CAIN,2BAA4B,CAF5B,+BzDikMR,CyD5jMM,yBAIE,oFAFA,OAAQ,CADR,KAAM,CAIN,0BAA2B,CAF3B,+BzDgkMR,CyD3jMM,yBAME,sFALA,KAAM,CAMN,2BzD6jMR,CyD1jMM,qDANE,kCADA,MAAO,CAEP,eAAgB,CAHhB,OzD0kMR,CyDlkMM,4BAKE,mFACA,0BzD4jMR,CyDzjMM,gDAEE,czD0jMR,CyDvjMM,qDAGE,kBzDujMR,CyD5hMA,oBP9GE,qB5CUS,C4CXT,YAAa,CAHb,MAAO,CAFP,cAAe,CACf,KAAM,CAGN,WAAY,CADZ,YlDupMF,CkDjpME,yBAAS,SlDopMX,CkDnpME,yBAAS,UlDspMX,CyDxiMA,kBAEE,kBAAmB,CADnB,YAAa,CAEb,mEzD2iMF,CyDziME,6BAEE,oIADA,qFzD4iMJ,CyDviMA,iBAEE,kDADA,ezD2iMF,CyDviMA,gBACE,WAAY,CAEZ,eAAgB,CADhB,mEzD2iMF,C0DvrMA,aAKE,6BAA8B,CAD9B,WAAY,CAHZ,oBAAqB,CACrB,cAAe,CAIf,UpDgzCkC,CoDnzClC,qB1D6rMF,C0DxrME,wBAEE,UAAW,CADX,oB1D2rMJ,C0DrrMA,gBACE,e1DwrMF,C0DrrMA,gBACE,e1DwrMF,C0DrrMA,gBACE,gB1DwrMF,C0DnrME,+BACE,kD1DsrMJ,C0DlrMA,4BACE,IACE,U1DqrMF,CACF,C0DlrMA,kBAGE,6CAA8C,CAF9C,wEACA,mB1DqrMF,C0DjrMA,4BACE,GACE,qB1DorMF,CACF,CuDnuME,gBAEE,UAAW,CACX,UAAW,CAFX,avDuuMJ,C2DxuME,iBAEE,8EADA,oB3D4uMJ,C2D7uME,mBAEE,gFADA,oB3DivMJ,C2DlvME,iBAEE,8EADA,oB3DsvMJ,C2DvvME,cAEE,2EADA,oB3D2vMJ,C2D5vME,iBAEE,8EADA,oB3DgwMJ,C2DjwME,gBAEE,6EADA,oB3DqwMJ,C2DtwME,eAEE,4EADA,oB3D0wMJ,C2D3wME,cAEE,2EADA,oB3D+wMJ,C4DhxME,cACE,qEACA,8F5DmxMJ,C4DhxMM,wCAGE,yDACA,kF5DgxMR,C4DzxME,gBACE,uEACA,gG5D4xMJ,C4DzxMM,4CAGE,yDACA,kF5DyxMR,C4DlyME,cACE,qEACA,8F5DqyMJ,C4DlyMM,wCAGE,yDACA,kF5DkyMR,C4D3yME,WACE,kEACA,2F5D8yMJ,C4D3yMM,kCAGE,0DACA,mF5D2yMR,C4DpzME,cACE,qEACA,8F5DuzMJ,C4DpzMM,wCAGE,0DACA,mF5DozMR,C4D7zME,aACE,oEACA,6F5Dg0MJ,C4D7zMM,sCAGE,yDACA,kF5D6zMR,C4Dt0ME,YACE,mEACA,4F5Dy0MJ,C4Dt0MM,oCAGE,2DACA,oF5Ds0MR,C4D/0ME,WACE,kEACA,2F5Dk1MJ,C4D/0MM,kCAGE,wDACA,iF5D+0MR,C4Dx0MA,oBACE,4EACA,qG5D20MF,C4Dx0MI,oDAEE,8EACA,uG5Dy0MN,C6Dn2MA,kBAGE,+IAFA,S7Du2MF,C8Dx2MA,WAGE,kBAAmB,CAGnB,0BAA2B,CAL3B,mBAAoB,CACpB,WxD6c4B,CwD3c5B,+EACA,2B9D42MF,C8Dz2ME,eAIE,iBAAkB,CAHlB,aAAc,CAEd,UxDsc0B,CgBjcxB,oCwCHF,CAHA,S9D82MJ,CsBp2MM,uCwCZJ,exCaM,etBu2MN,CACF,C8Dz2MI,8DACE,8D9D42MN,C+D/3MA,OACE,iBAAkB,CAClB,U/Dk4MF,C+Dh4ME,cAGE,UAAW,CAFX,aAAc,CACd,kC/Dm4MJ,C+D/3ME,SAKE,WAAY,CAFZ,MAAO,CAFP,iBAAkB,CAClB,KAAM,CAEN,U/Dk4MJ,C+D53ME,WACE,sB/D+3MJ,C+Dh4ME,WACE,qB/Dm4MJ,C+Dp4ME,YACE,wB/Du4MJ,C+Dx4ME,YACE,gC/D24MJ,CgEh6MA,WAEE,KhEs6MF,CgEh6MA,yBAJE,MAAO,CAHP,cAAe,CAEf,OAAQ,CAER,YhE26MF,CgEx6MA,cAGE,QhEq6MF,CgE35MI,YAEE,KhE+5MN,CgE35MI,2BALE,eAAgB,CAEhB,YhEo6MN,CgEj6MI,eAEE,QhE+5MN,Cc/3MI,yBkDxCA,eACE,eAAgB,CAChB,KAAM,CACN,YhE26MJ,CgEx6ME,kBAEE,QAAS,CADT,eAAgB,CAEhB,YhE06MJ,CACF,Cc54MI,yBkDxCA,eACE,eAAgB,CAChB,KAAM,CACN,YhEu7MJ,CgEp7ME,kBAEE,QAAS,CADT,eAAgB,CAEhB,YhEs7MJ,CACF,Ccx5MI,yBkDxCA,eACE,eAAgB,CAChB,KAAM,CACN,YhEm8MJ,CgEh8ME,kBAEE,QAAS,CADT,eAAgB,CAEhB,YhEk8MJ,CACF,Ccp6MI,0BkDxCA,eACE,eAAgB,CAChB,KAAM,CACN,YhE+8MJ,CgE58ME,kBAEE,QAAS,CADT,eAAgB,CAEhB,YhE88MJ,CACF,CiE9+MA,QAGE,kBAAmB,CADnB,kBjEk/MF,CiE7+MA,gBAHE,kBAAmB,CAHnB,YjE0/MF,CiEp/MA,QAEE,aAAc,CACd,qBjEi/MF,CkEx/MA,2ECSE,4BAAiC,CAEjC,kBAAoB,CANpB,oBAAsB,CAEtB,qBAAuB,CACvB,yBAA2B,CAF3B,mBAAqB,CAIrB,4BAA8B,CAN9B,mBnEggNF,CmEt/ME,qGACE,2BnEy/MJ,CoEvgNE,sBAIE,QAAS,CAGT,UAAW,CAFX,MAAO,CAJP,iBAAkB,CAElB,OAAQ,CADR,KAAM,CAIN,SpE2gNJ,CqElhNA,eCAE,eAAgB,CAChB,sBAAuB,CACvB,kBtEshNF,CuE5hNA,IAEE,kBAAmB,CAGnB,6BAA8B,CAJ9B,oBAAqB,CAGrB,cAAe,CAEf,WjE2rB4B,CiE9rB5B,4BvEkiNF,CwEn+MQ,gBAOI,iCxEg+MZ,CwEv+MQ,WAOI,4BxEo+MZ,CwE3+MQ,cAOI,+BxEw+MZ,CwE/+MQ,cAOI,+BxE4+MZ,CwEn/MQ,mBAOI,oCxEg/MZ,CwEv/MQ,gBAOI,iCxEo/MZ,CwE3/MQ,aAOI,oBxEw/MZ,CwE//MQ,WAOI,qBxE4/MZ,CwEngNQ,YAOI,oBxEggNZ,CwEvgNQ,oBAOI,4BxEogNZ,CwE3gNQ,kBAOI,0BxEwgNZ,CwE/gNQ,iBAOI,yBxE4gNZ,CwEnhNQ,kBAOI,+BxEghNZ,CwEvhNQ,iBAOI,yBxEohNZ,CwE3hNQ,WAOI,mBxEwhNZ,CwE/hNQ,YAOI,qBxE4hNZ,CwEniNQ,YAOI,oBxEgiNZ,CwEviNQ,YAOI,qBxEoiNZ,CwE3iNQ,aAOI,mBxEwiNZ,CwE/iNQ,eAOI,uBxE4iNZ,CwEnjNQ,iBAOI,yBxEgjNZ,CwEvjNQ,kBAOI,0BxEojNZ,CwE3jNQ,iBAOI,yBxEwjNZ,CwE/jNQ,iBAOI,yBxE4jNZ,CwEnkNQ,mBAOI,2BxEgkNZ,CwEvkNQ,oBAOI,4BxEokNZ,CwE3kNQ,mBAOI,2BxEwkNZ,CwE/kNQ,iBAOI,yBxE4kNZ,CwEnlNQ,mBAOI,2BxEglNZ,CwEvlNQ,oBAOI,4BxEolNZ,CwE3lNQ,mBAOI,2BxEwlNZ,CwE/lNQ,UAOI,wBxE4lNZ,CwEnmNQ,gBAOI,8BxEgmNZ,CwEvmNQ,SAOI,uBxEomNZ,CwE3mNQ,QAOI,sBxEwmNZ,CwE/mNQ,eAOI,6BxE4mNZ,CwEnnNQ,SAOI,uBxEgnNZ,CwEvnNQ,aAOI,2BxEonNZ,CwE3nNQ,cAOI,4BxEwnNZ,CwE/nNQ,QAOI,sBxE4nNZ,CwEnoNQ,eAOI,6BxEgoNZ,CwEvoNQ,QAOI,sBxEooNZ,CwE3oNQ,QAOI,yCxEwoNZ,CwE/oNQ,WAOI,4CxE4oNZ,CwEnpNQ,WAOI,4CxEgpNZ,CwEvpNQ,aAOI,yBxEopNZ,CwErqNQ,oBACE,8ExEwqNV,CwEzqNQ,sBACE,gFxE4qNV,CwE7qNQ,oBACE,8ExEgrNV,CwEjrNQ,iBACE,2ExEorNV,CwErrNQ,oBACE,8ExEwrNV,CwEzrNQ,mBACE,6ExE4rNV,CwE7rNQ,kBACE,4ExEgsNV,CwEjsNQ,iBACE,2ExEosNV,CwE3rNQ,iBAOI,yBxEwrNZ,CwE/rNQ,mBAOI,2BxE4rNZ,CwEnsNQ,mBAOI,2BxEgsNZ,CwEvsNQ,gBAOI,wBxEosNZ,CwE3sNQ,iBAOI,yBxEwsNZ,CwE/sNQ,OAOI,exE4sNZ,CwEntNQ,QAOI,iBxEgtNZ,CwEvtNQ,SAOI,kBxEotNZ,CwE3tNQ,UAOI,kBxEwtNZ,CwE/tNQ,WAOI,oBxE4tNZ,CwEnuNQ,YAOI,qBxEguNZ,CwEvuNQ,SAOI,gBxEouNZ,CwE3uNQ,UAOI,kBxEwuNZ,CwE/uNQ,WAOI,mBxE4uNZ,CwEnvNQ,OAOI,iBxEgvNZ,CwEvvNQ,QAOI,mBxEovNZ,CwE3vNQ,SAOI,oBxEwvNZ,CwE/vNQ,kBAOI,wCxE4vNZ,CwEnwNQ,oBAOI,oCxEgwNZ,CwEvwNQ,oBAOI,oCxEowNZ,CwE3wNQ,QAOI,qFxEwwNZ,CwE/wNQ,UAOI,kBxE4wNZ,CwEnxNQ,YAOI,yFxEgxNZ,CwEvxNQ,cAOI,sBxEoxNZ,CwE3xNQ,YAOI,2FxEwxNZ,CwE/xNQ,cAOI,wBxE4xNZ,CwEnyNQ,eAOI,4FxEgyNZ,CwEvyNQ,iBAOI,yBxEoyNZ,CwE3yNQ,cAOI,0FxEwyNZ,CwE/yNQ,gBAOI,uBxE4yNZ,CwEnzNQ,gBAIQ,sBAGJ,2ExEizNZ,CwExzNQ,kBAIQ,sBAGJ,6ExEszNZ,CwE7zNQ,gBAIQ,sBAGJ,2ExE2zNZ,CwEl0NQ,aAIQ,sBAGJ,wExEg0NZ,CwEv0NQ,gBAIQ,sBAGJ,2ExEq0NZ,CwE50NQ,eAIQ,sBAGJ,0ExE00NZ,CwEj1NQ,cAIQ,sBAGJ,yExE+0NZ,CwEt1NQ,aAIQ,sBAGJ,wExEo1NZ,CwE31NQ,cAIQ,sBAGJ,yExEy1NZ,CwEh2NQ,cAIQ,sBAGJ,yExE81NZ,CwEr2NQ,uBAOI,sDxEk2NZ,CwEz2NQ,yBAOI,wDxEs2NZ,CwE72NQ,uBAOI,sDxE02NZ,CwEj3NQ,oBAOI,mDxE82NZ,CwEr3NQ,uBAOI,sDxEk3NZ,CwEz3NQ,sBAOI,qDxEs3NZ,CwE73NQ,qBAOI,oDxE03NZ,CwEj4NQ,oBAOI,mDxE83NZ,CwEr4NQ,UAOI,0BxEk4NZ,CwEz4NQ,UAOI,0BxEs4NZ,CwE74NQ,UAOI,0BxE04NZ,CwEj5NQ,UAOI,0BxE84NZ,CwEr5NQ,UAOI,0BxEk5NZ,CwEn6NQ,mBACE,uBxEs6NV,CwEv6NQ,mBACE,wBxE06NV,CwE36NQ,mBACE,uBxE86NV,CwE/6NQ,mBACE,wBxEk7NV,CwEn7NQ,oBACE,qBxEs7NV,CwE76NQ,MAOI,mBxE06NZ,CwEj7NQ,MAOI,mBxE86NZ,CwEr7NQ,MAOI,mBxEk7NZ,CwEz7NQ,OAOI,oBxEs7NZ,CwE77NQ,QAOI,oBxE07NZ,CwEj8NQ,QAOI,wBxE87NZ,CwEr8NQ,QAOI,qBxEk8NZ,CwEz8NQ,YAOI,yBxEs8NZ,CwE78NQ,MAOI,oBxE08NZ,CwEj9NQ,MAOI,oBxE88NZ,CwEr9NQ,MAOI,oBxEk9NZ,CwEz9NQ,OAOI,qBxEs9NZ,CwE79NQ,QAOI,qBxE09NZ,CwEj+NQ,QAOI,yBxE89NZ,CwEr+NQ,QAOI,sBxEk+NZ,CwEz+NQ,YAOI,0BxEs+NZ,CwE7+NQ,WAOI,uBxE0+NZ,CwEj/NQ,UAOI,4BxE8+NZ,CwEr/NQ,aAOI,+BxEk/NZ,CwEz/NQ,kBAOI,oCxEs/NZ,CwE7/NQ,qBAOI,uCxE0/NZ,CwEjgOQ,aAOI,qBxE8/NZ,CwErgOQ,aAOI,qBxEkgOZ,CwEzgOQ,eAOI,uBxEsgOZ,CwE7gOQ,eAOI,uBxE0gOZ,CwEjhOQ,WAOI,wBxE8gOZ,CwErhOQ,aAOI,0BxEkhOZ,CwEzhOQ,mBAOI,gCxEshOZ,CwE7hOQ,uBAOI,oCxE0hOZ,CwEjiOQ,qBAOI,kCxE8hOZ,CwEriOQ,wBAOI,gCxEkiOZ,CwEziOQ,yBAOI,uCxEsiOZ,CwE7iOQ,wBAOI,sCxE0iOZ,CwEjjOQ,wBAOI,sCxE8iOZ,CwErjOQ,mBAOI,gCxEkjOZ,CwEzjOQ,iBAOI,8BxEsjOZ,CwE7jOQ,oBAOI,4BxE0jOZ,CwEjkOQ,sBAOI,8BxE8jOZ,CwErkOQ,qBAOI,6BxEkkOZ,CwEzkOQ,qBAOI,kCxEskOZ,CwE7kOQ,mBAOI,gCxE0kOZ,CwEjlOQ,sBAOI,8BxE8kOZ,CwErlOQ,uBAOI,qCxEklOZ,CwEzlOQ,sBAOI,oCxEslOZ,CwE7lOQ,uBAOI,+BxE0lOZ,CwEjmOQ,iBAOI,yBxE8lOZ,CwErmOQ,kBAOI,+BxEkmOZ,CwEzmOQ,gBAOI,6BxEsmOZ,CwE7mOQ,mBAOI,2BxE0mOZ,CwEjnOQ,qBAOI,6BxE8mOZ,CwErnOQ,oBAOI,4BxEknOZ,CwEznOQ,aAOI,kBxEsnOZ,CwE7nOQ,SAOI,iBxE0nOZ,CwEjoOQ,SAOI,iBxE8nOZ,CwEroOQ,SAOI,iBxEkoOZ,CwEzoOQ,SAOI,iBxEsoOZ,CwE7oOQ,SAOI,iBxE0oOZ,CwEjpOQ,SAOI,iBxE8oOZ,CwErpOQ,YAOI,iBxEkpOZ,CwEzpOQ,KAOI,kBxEspOZ,CwE7pOQ,KAOI,uBxE0pOZ,CwEjqOQ,KAOI,sBxE8pOZ,CwErqOQ,KAOI,qBxEkqOZ,CwEzqOQ,KAOI,uBxEsqOZ,CwE7qOQ,KAOI,qBxE0qOZ,CwEjrOQ,QAOI,qBxE8qOZ,CwErrOQ,MAOI,gDxEmrOZ,CwE1rOQ,MAOI,0DxEwrOZ,CwE/rOQ,MAOI,wDxE6rOZ,CwEpsOQ,MAOI,sDxEksOZ,CwEzsOQ,MAOI,0DxEusOZ,CwE9sOQ,MAOI,sDxE4sOZ,CwEntOQ,SAOI,sDxEitOZ,CwExtOQ,MAOI,gDxEstOZ,CwE7tOQ,MAOI,0DxE2tOZ,CwEluOQ,MAOI,wDxEguOZ,CwEvuOQ,MAOI,sDxEquOZ,CwE5uOQ,MAOI,0DxE0uOZ,CwEjvOQ,MAOI,sDxE+uOZ,CwEtvOQ,SAOI,sDxEovOZ,CwE3vOQ,MAOI,sBxEwvOZ,CwE/vOQ,MAOI,2BxE4vOZ,CwEnwOQ,MAOI,0BxEgwOZ,CwEvwOQ,MAOI,yBxEowOZ,CwE3wOQ,MAOI,2BxEwwOZ,CwE/wOQ,MAOI,yBxE4wOZ,CwEnxOQ,SAOI,yBxEgxOZ,CwEvxOQ,MAOI,wBxEoxOZ,CwE3xOQ,MAOI,6BxEwxOZ,CwE/xOQ,MAOI,4BxE4xOZ,CwEnyOQ,MAOI,2BxEgyOZ,CwEvyOQ,MAOI,6BxEoyOZ,CwE3yOQ,MAOI,2BxEwyOZ,CwE/yOQ,SAOI,2BxE4yOZ,CwEnzOQ,MAOI,yBxEgzOZ,CwEvzOQ,MAOI,8BxEozOZ,CwE3zOQ,MAOI,6BxEwzOZ,CwE/zOQ,MAOI,4BxE4zOZ,CwEn0OQ,MAOI,8BxEg0OZ,CwEv0OQ,MAOI,4BxEo0OZ,CwE30OQ,SAOI,4BxEw0OZ,CwE/0OQ,MAOI,uBxE40OZ,CwEn1OQ,MAOI,4BxEg1OZ,CwEv1OQ,MAOI,2BxEo1OZ,CwE31OQ,MAOI,0BxEw1OZ,CwE/1OQ,MAOI,4BxE41OZ,CwEn2OQ,MAOI,0BxEg2OZ,CwEv2OQ,SAOI,0BxEo2OZ,CwE32OQ,KAOI,mBxEw2OZ,CwE/2OQ,KAOI,wBxE42OZ,CwEn3OQ,KAOI,uBxEg3OZ,CwEv3OQ,KAOI,sBxEo3OZ,CwE33OQ,KAOI,wBxEw3OZ,CwE/3OQ,KAOI,sBxE43OZ,CwEn4OQ,MAOI,kDxEi4OZ,CwEx4OQ,MAOI,4DxEs4OZ,CwE74OQ,MAOI,0DxE24OZ,CwEl5OQ,MAOI,wDxEg5OZ,CwEv5OQ,MAOI,4DxEq5OZ,CwE55OQ,MAOI,wDxE05OZ,CwEj6OQ,MAOI,kDxE+5OZ,CwEt6OQ,MAOI,4DxEo6OZ,CwE36OQ,MAOI,0DxEy6OZ,CwEh7OQ,MAOI,wDxE86OZ,CwEr7OQ,MAOI,4DxEm7OZ,CwE17OQ,MAOI,wDxEw7OZ,CwE/7OQ,MAOI,uBxE47OZ,CwEn8OQ,MAOI,4BxEg8OZ,CwEv8OQ,MAOI,2BxEo8OZ,CwE38OQ,MAOI,0BxEw8OZ,CwE/8OQ,MAOI,4BxE48OZ,CwEn9OQ,MAOI,0BxEg9OZ,CwEv9OQ,MAOI,yBxEo9OZ,CwE39OQ,MAOI,8BxEw9OZ,CwE/9OQ,MAOI,6BxE49OZ,CwEn+OQ,MAOI,4BxEg+OZ,CwEv+OQ,MAOI,8BxEo+OZ,CwE3+OQ,MAOI,4BxEw+OZ,CwE/+OQ,MAOI,0BxE4+OZ,CwEn/OQ,MAOI,+BxEg/OZ,CwEv/OQ,MAOI,8BxEo/OZ,CwE3/OQ,MAOI,6BxEw/OZ,CwE//OQ,MAOI,+BxE4/OZ,CwEngPQ,MAOI,6BxEggPZ,CwEvgPQ,MAOI,wBxEogPZ,CwE3gPQ,MAOI,6BxEwgPZ,CwE/gPQ,MAOI,4BxE4gPZ,CwEnhPQ,MAOI,2BxEghPZ,CwEvhPQ,MAOI,6BxEohPZ,CwE3hPQ,MAOI,2BxEwhPZ,CwE/hPQ,OAOI,exE4hPZ,CwEniPQ,OAOI,oBxEgiPZ,CwEviPQ,OAOI,mBxEoiPZ,CwE3iPQ,OAOI,kBxEwiPZ,CwE/iPQ,OAOI,oBxE4iPZ,CwEnjPQ,OAOI,kBxEgjPZ,CwEvjPQ,WAOI,mBxEojPZ,CwE3jPQ,WAOI,wBxEwjPZ,CwE/jPQ,WAOI,uBxE4jPZ,CwEnkPQ,WAOI,sBxEgkPZ,CwEvkPQ,WAOI,wBxEokPZ,CwE3kPQ,WAOI,sBxEwkPZ,CwE/kPQ,cAOI,sBxE4kPZ,CwEnlPQ,cAOI,2BxEglPZ,CwEvlPQ,cAOI,0BxEolPZ,CwE3lPQ,cAOI,yBxEwlPZ,CwE/lPQ,cAOI,2BxE4lPZ,CwEnmPQ,cAOI,yBxEgmPZ,CwEvmPQ,gBAOI,8CxEomPZ,CwE3mPQ,MAOI,0CxEwmPZ,CwE/mPQ,MAOI,yCxE4mPZ,CwEnnPQ,MAOI,uCxEgnPZ,CwEvnPQ,MAOI,yCxEonPZ,CwE3nPQ,MAOI,2BxEwnPZ,CwE/nPQ,MAOI,wBxE4nPZ,CwEnoPQ,YAOI,2BxEgoPZ,CwEvoPQ,YAOI,2BxEooPZ,CwE3oPQ,YAOI,6BxEwoPZ,CwE/oPQ,UAOI,yBxE4oPZ,CwEnpPQ,WAOI,yBxEgpPZ,CwEvpPQ,WAOI,yBxEopPZ,CwE3pPQ,aAOI,yBxEwpPZ,CwE/pPQ,SAOI,yBxE4pPZ,CwEnqPQ,WAOI,4BxEgqPZ,CwEvqPQ,MAOI,uBxEoqPZ,CwE3qPQ,OAOI,0BxEwqPZ,CwE/qPQ,SAOI,yBxE4qPZ,CwEnrPQ,OAOI,uBxEgrPZ,CwEvrPQ,YAOI,yBxEorPZ,CwE3rPQ,UAOI,0BxEwrPZ,CwE/rPQ,aAOI,2BxE4rPZ,CwEnsPQ,sBAOI,8BxEgsPZ,CwEvsPQ,2BAOI,mCxEosPZ,CwE3sPQ,8BAOI,sCxEwsPZ,CwE/sPQ,gBAOI,kCxE4sPZ,CwEntPQ,gBAOI,kCxEgtPZ,CwEvtPQ,iBAOI,mCxEotPZ,CwE3tPQ,WAOI,4BxEwtPZ,CwE/tPQ,aAOI,4BxE4tPZ,CwEnuPQ,YAOI,8DxEkuPZ,CwEzuPQ,cAIQ,oBAGJ,kExEwuPZ,CwE/uPQ,gBAIQ,oBAGJ,oExE6uPZ,CwEpvPQ,cAIQ,oBAGJ,kExEkvPZ,CwEzvPQ,WAIQ,oBAGJ,+DxEuvPZ,CwE9vPQ,cAIQ,oBAGJ,kExE4vPZ,CwEnwPQ,aAIQ,oBAGJ,iExEiwPZ,CwExwPQ,YAIQ,oBAGJ,gExEswPZ,CwE7wPQ,WAIQ,oBAGJ,+DxE2wPZ,CwElxPQ,YAIQ,oBAGJ,gExEgxPZ,CwEvxPQ,YAIQ,oBAGJ,gExEqxPZ,CwE5xPQ,WAIQ,oBAGJ,qExE0xPZ,CwEjyPQ,YAIQ,oBAGJ,yCxE+xPZ,CwEtyPQ,eAIQ,oBAGJ,8BxEoyPZ,CwE3yPQ,eAIQ,oBAGJ,kCxEyyPZ,CwEhzPQ,qBAIQ,oBAGJ,yCxE8yPZ,CwErzPQ,oBAIQ,oBAGJ,wCxEmzPZ,CwE1zPQ,oBAIQ,oBAGJ,wCxEwzPZ,CwE/zPQ,YAIQ,oBAGJ,uBxE6zPZ,CwE90PQ,iBACE,sBxEi1PV,CwEl1PQ,iBACE,qBxEq1PV,CwEt1PQ,iBACE,sBxEy1PV,CwE11PQ,kBACE,mBxE61PV,CwEp1PQ,uBAOI,+CxEi1PZ,CwEx1PQ,yBAOI,iDxEq1PZ,CwE51PQ,uBAOI,+CxEy1PZ,CwEh2PQ,oBAOI,4CxE61PZ,CwEp2PQ,uBAOI,+CxEi2PZ,CwEx2PQ,sBAOI,8CxEq2PZ,CwE52PQ,qBAOI,6CxEy2PZ,CwEh3PQ,oBAOI,4CxE62PZ,CwEz3PU,8CACE,qBxEg4PZ,CwEj4PU,8CACE,sBxEw4PZ,CwEz4PU,8CACE,qBxEg5PZ,CwEj5PU,8CACE,sBxEw5PZ,CwEz5PU,gDACE,mBxEg6PZ,CwEh5PU,0CAOI,sCxEi5Pd,CwEx5PU,0CAOI,qCxEy5Pd,CwEh6PU,0CAOI,sCxEi6Pd,CwEp7PQ,wBAIQ,8BAGJ,4FxEk7PZ,CwEz7PQ,0BAIQ,8BAGJ,8FxEu7PZ,CwE97PQ,wBAIQ,8BAGJ,4FxE47PZ,CwEn8PQ,qBAIQ,8BAGJ,yFxEi8PZ,CwEx8PQ,wBAIQ,8BAGJ,4FxEs8PZ,CwE78PQ,uBAIQ,8BAGJ,2FxE28PZ,CwEl9PQ,sBAIQ,8BAGJ,0FxEg9PZ,CwEv9PQ,qBAIQ,8BAGJ,yFxEq9PZ,CwE59PQ,gBAIQ,8BAGJ,iGxE09PZ,CwEt+PU,gEACE,6BxE6+PZ,CwE9+PU,kEACE,+BxEq/PZ,CwEt/PU,kEACE,gCxE6/PZ,CwE9/PU,kEACE,+BxEqgQZ,CwEtgQU,kEACE,gCxE6gQZ,CwE9gQU,oEACE,6BxEqhQZ,CwEjhQQ,YAIQ,kBAGJ,2ExE+gQZ,CwEthQQ,cAIQ,kBAGJ,6ExEohQZ,CwE3hQQ,YAIQ,kBAGJ,2ExEyhQZ,CwEhiQQ,SAIQ,kBAGJ,wExE8hQZ,CwEriQQ,YAIQ,kBAGJ,2ExEmiQZ,CwE1iQQ,WAIQ,kBAGJ,0ExEwiQZ,CwE/iQQ,UAIQ,kBAGJ,yExE6iQZ,CwEpjQQ,SAIQ,kBAGJ,wExEkjQZ,CwEzjQQ,UAIQ,kBAGJ,yExEujQZ,CwE9jQQ,UAIQ,kBAGJ,yExE4jQZ,CwEnkQQ,SAIQ,kBAGJ,2ExEikQZ,CwExkQQ,gBAIQ,kBAGJ,sCxEskQZ,CwE7kQQ,mBAIQ,kBAGJ,gFxE2kQZ,CwEllQQ,kBAIQ,kBAGJ,+ExEglQZ,CwEjmQQ,eACE,mBxEomQV,CwErmQQ,eACE,oBxEwmQV,CwEzmQQ,eACE,mBxE4mQV,CwE7mQQ,eACE,oBxEgnQV,CwEjnQQ,gBACE,iBxEonQV,CwE3mQQ,mBAOI,sDxEwmQZ,CwE/mQQ,qBAOI,wDxE4mQZ,CwEnnQQ,mBAOI,sDxEgnQZ,CwEvnQQ,gBAOI,mDxEonQZ,CwE3nQQ,mBAOI,sDxEwnQZ,CwE/nQQ,kBAOI,qDxE4nQZ,CwEnoQQ,iBAOI,oDxEgoQZ,CwEvoQQ,gBAOI,mDxEooQZ,CwE3oQQ,aAOI,6CxEwoQZ,CwE/oQQ,iBAOI,yBxE4oQZ,CwEnpQQ,kBAOI,0BxEgpQZ,CwEvpQQ,kBAOI,0BxEopQZ,CwE3pQQ,SAOI,6BxEwpQZ,CwE/pQQ,SAOI,6BxE4pQZ,CwEnqQQ,SAOI,+CxEgqQZ,CwEvqQQ,WAOI,yBxEoqQZ,CwE3qQQ,WAOI,kDxEwqQZ,CwE/qQQ,WAOI,+CxE4qQZ,CwEnrQQ,WAOI,kDxEgrQZ,CwEvrQQ,WAOI,kDxEorQZ,CwE3rQQ,WAOI,mDxEwrQZ,CwE/rQQ,gBAOI,2BxE4rQZ,CwEnsQQ,cAOI,oDxEgsQZ,CwEvsQQ,aAOI,kHxEqsQZ,CwE5sQQ,eAOI,sExE0sQZ,CwEjtQQ,eAOI,wHxE+sQZ,CwEttQQ,eAOI,kHxEotQZ,CwE3tQQ,eAOI,wHxEytQZ,CwEhuQQ,eAOI,wHxE8tQZ,CwEruQQ,eAOI,0HxEmuQZ,CwE1uQQ,oBAOI,0ExEwuQZ,CwE/uQQ,kBAOI,4HxE6uQZ,CwEpvQQ,aAOI,sHxEkvQZ,CwEzvQQ,eAOI,0ExEuvQZ,CwE9vQQ,eAOI,4HxE4vQZ,CwEnwQQ,eAOI,sHxEiwQZ,CwExwQQ,eAOI,4HxEswQZ,CwE7wQQ,eAOI,4HxE2wQZ,CwElxQQ,eAOI,8HxEgxQZ,CwEvxQQ,oBAOI,8ExEqxQZ,CwE5xQQ,kBAOI,gIxE0xQZ,CwEjyQQ,gBAOI,wHxE+xQZ,CwEtyQQ,kBAOI,4ExEoyQZ,CwE3yQQ,kBAOI,8HxEyyQZ,CwEhzQQ,kBAOI,wHxE8yQZ,CwErzQQ,kBAOI,8HxEmzQZ,CwE1zQQ,kBAOI,8HxEwzQZ,CwE/zQQ,kBAOI,gIxE6zQZ,CwEp0QQ,uBAOI,gFxEk0QZ,CwEz0QQ,qBAOI,kIxEu0QZ,CwE90QQ,eAOI,oHxE40QZ,CwEn1QQ,iBAOI,wExEi1QZ,CwEx1QQ,iBAOI,0HxEs1QZ,CwE71QQ,iBAOI,oHxE21QZ,CwEl2QQ,iBAOI,0HxEg2QZ,CwEv2QQ,iBAOI,0HxEq2QZ,CwE52QQ,iBAOI,4HxE02QZ,CwEj3QQ,sBAOI,4ExE+2QZ,CwEt3QQ,oBAOI,8HxEo3QZ,CwE33QQ,SAOI,4BxEw3QZ,CwE/3QQ,WAOI,2BxE43QZ,CwEn4QQ,MAOI,oBxEg4QZ,CwEv4QQ,KAOI,mBxEo4QZ,CwE34QQ,KAOI,mBxEw4QZ,CwE/4QQ,KAOI,mBxE44QZ,CwEn5QQ,KAOI,mBxEg5QZ,Cc15QI,yB0DGI,gBAOI,oBxEq5QV,CwE55QM,cAOI,qBxEw5QV,CwE/5QM,eAOI,oBxE25QV,CwEl6QM,uBAOI,4BxE85QV,CwEr6QM,qBAOI,0BxEi6QV,CwEx6QM,oBAOI,yBxEo6QV,CwE36QM,qBAOI,+BxEu6QV,CwE96QM,oBAOI,yBxE06QV,CwEj7QM,aAOI,wBxE66QV,CwEp7QM,mBAOI,8BxEg7QV,CwEv7QM,YAOI,uBxEm7QV,CwE17QM,WAOI,sBxEs7QV,CwE77QM,kBAOI,6BxEy7QV,CwEh8QM,YAOI,uBxE47QV,CwEn8QM,gBAOI,2BxE+7QV,CwEt8QM,iBAOI,4BxEk8QV,CwEz8QM,WAOI,sBxEq8QV,CwE58QM,kBAOI,6BxEw8QV,CwE/8QM,WAOI,sBxE28QV,CwEl9QM,cAOI,uBxE88QV,CwEr9QM,aAOI,4BxEi9QV,CwEx9QM,gBAOI,+BxEo9QV,CwE39QM,qBAOI,oCxEu9QV,CwE99QM,wBAOI,uCxE09QV,CwEj+QM,gBAOI,qBxE69QV,CwEp+QM,gBAOI,qBxEg+QV,CwEv+QM,kBAOI,uBxEm+QV,CwE1+QM,kBAOI,uBxEs+QV,CwE7+QM,cAOI,wBxEy+QV,CwEh/QM,gBAOI,0BxE4+QV,CwEn/QM,sBAOI,gCxE++QV,CwEt/QM,0BAOI,oCxEk/QV,CwEz/QM,wBAOI,kCxEq/QV,CwE5/QM,2BAOI,gCxEw/QV,CwE//QM,4BAOI,uCxE2/QV,CwElgRM,2BAOI,sCxE8/QV,CwErgRM,2BAOI,sCxEigRV,CwExgRM,sBAOI,gCxEogRV,CwE3gRM,oBAOI,8BxEugRV,CwE9gRM,uBAOI,4BxE0gRV,CwEjhRM,yBAOI,8BxE6gRV,CwEphRM,wBAOI,6BxEghRV,CwEvhRM,wBAOI,kCxEmhRV,CwE1hRM,sBAOI,gCxEshRV,CwE7hRM,yBAOI,8BxEyhRV,CwEhiRM,0BAOI,qCxE4hRV,CwEniRM,yBAOI,oCxE+hRV,CwEtiRM,0BAOI,+BxEkiRV,CwEziRM,oBAOI,yBxEqiRV,CwE5iRM,qBAOI,+BxEwiRV,CwE/iRM,mBAOI,6BxE2iRV,CwEljRM,sBAOI,2BxE8iRV,CwErjRM,wBAOI,6BxEijRV,CwExjRM,uBAOI,4BxEojRV,CwE3jRM,gBAOI,kBxEujRV,CwE9jRM,YAOI,iBxE0jRV,CwEjkRM,YAOI,iBxE6jRV,CwEpkRM,YAOI,iBxEgkRV,CwEvkRM,YAOI,iBxEmkRV,CwE1kRM,YAOI,iBxEskRV,CwE7kRM,YAOI,iBxEykRV,CwEhlRM,eAOI,iBxE4kRV,CwEnlRM,QAOI,kBxE+kRV,CwEtlRM,QAOI,uBxEklRV,CwEzlRM,QAOI,sBxEqlRV,CwE5lRM,QAOI,qBxEwlRV,CwE/lRM,QAOI,uBxE2lRV,CwElmRM,QAOI,qBxE8lRV,CwErmRM,WAOI,qBxEimRV,CwExmRM,SAOI,gDxEqmRV,CwE5mRM,SAOI,0DxEymRV,CwEhnRM,SAOI,wDxE6mRV,CwEpnRM,SAOI,sDxEinRV,CwExnRM,SAOI,0DxEqnRV,CwE5nRM,SAOI,sDxEynRV,CwEhoRM,YAOI,sDxE6nRV,CwEpoRM,SAOI,gDxEioRV,CwExoRM,SAOI,0DxEqoRV,CwE5oRM,SAOI,wDxEyoRV,CwEhpRM,SAOI,sDxE6oRV,CwEppRM,SAOI,0DxEipRV,CwExpRM,SAOI,sDxEqpRV,CwE5pRM,YAOI,sDxEypRV,CwEhqRM,SAOI,sBxE4pRV,CwEnqRM,SAOI,2BxE+pRV,CwEtqRM,SAOI,0BxEkqRV,CwEzqRM,SAOI,yBxEqqRV,CwE5qRM,SAOI,2BxEwqRV,CwE/qRM,SAOI,yBxE2qRV,CwElrRM,YAOI,yBxE8qRV,CwErrRM,SAOI,wBxEirRV,CwExrRM,SAOI,6BxEorRV,CwE3rRM,SAOI,4BxEurRV,CwE9rRM,SAOI,2BxE0rRV,CwEjsRM,SAOI,6BxE6rRV,CwEpsRM,SAOI,2BxEgsRV,CwEvsRM,YAOI,2BxEmsRV,CwE1sRM,SAOI,yBxEssRV,CwE7sRM,SAOI,8BxEysRV,CwEhtRM,SAOI,6BxE4sRV,CwEntRM,SAOI,4BxE+sRV,CwEttRM,SAOI,8BxEktRV,CwEztRM,SAOI,4BxEqtRV,CwE5tRM,YAOI,4BxEwtRV,CwE/tRM,SAOI,uBxE2tRV,CwEluRM,SAOI,4BxE8tRV,CwEruRM,SAOI,2BxEiuRV,CwExuRM,SAOI,0BxEouRV,CwE3uRM,SAOI,4BxEuuRV,CwE9uRM,SAOI,0BxE0uRV,CwEjvRM,YAOI,0BxE6uRV,CwEpvRM,QAOI,mBxEgvRV,CwEvvRM,QAOI,wBxEmvRV,CwE1vRM,QAOI,uBxEsvRV,CwE7vRM,QAOI,sBxEyvRV,CwEhwRM,QAOI,wBxE4vRV,CwEnwRM,QAOI,sBxE+vRV,CwEtwRM,SAOI,kDxEmwRV,CwE1wRM,SAOI,4DxEuwRV,CwE9wRM,SAOI,0DxE2wRV,CwElxRM,SAOI,wDxE+wRV,CwEtxRM,SAOI,4DxEmxRV,CwE1xRM,SAOI,wDxEuxRV,CwE9xRM,SAOI,kDxE2xRV,CwElyRM,SAOI,4DxE+xRV,CwEtyRM,SAOI,0DxEmyRV,CwE1yRM,SAOI,wDxEuyRV,CwE9yRM,SAOI,4DxE2yRV,CwElzRM,SAOI,wDxE+yRV,CwEtzRM,SAOI,uBxEkzRV,CwEzzRM,SAOI,4BxEqzRV,CwE5zRM,SAOI,2BxEwzRV,CwE/zRM,SAOI,0BxE2zRV,CwEl0RM,SAOI,4BxE8zRV,CwEr0RM,SAOI,0BxEi0RV,CwEx0RM,SAOI,yBxEo0RV,CwE30RM,SAOI,8BxEu0RV,CwE90RM,SAOI,6BxE00RV,CwEj1RM,SAOI,4BxE60RV,CwEp1RM,SAOI,8BxEg1RV,CwEv1RM,SAOI,4BxEm1RV,CwE11RM,SAOI,0BxEs1RV,CwE71RM,SAOI,+BxEy1RV,CwEh2RM,SAOI,8BxE41RV,CwEn2RM,SAOI,6BxE+1RV,CwEt2RM,SAOI,+BxEk2RV,CwEz2RM,SAOI,6BxEq2RV,CwE52RM,SAOI,wBxEw2RV,CwE/2RM,SAOI,6BxE22RV,CwEl3RM,SAOI,4BxE82RV,CwEr3RM,SAOI,2BxEi3RV,CwEx3RM,SAOI,6BxEo3RV,CwE33RM,SAOI,2BxEu3RV,CwE93RM,UAOI,exE03RV,CwEj4RM,UAOI,oBxE63RV,CwEp4RM,UAOI,mBxEg4RV,CwEv4RM,UAOI,kBxEm4RV,CwE14RM,UAOI,oBxEs4RV,CwE74RM,UAOI,kBxEy4RV,CwEh5RM,cAOI,mBxE44RV,CwEn5RM,cAOI,wBxE+4RV,CwEt5RM,cAOI,uBxEk5RV,CwEz5RM,cAOI,sBxEq5RV,CwE55RM,cAOI,wBxEw5RV,CwE/5RM,cAOI,sBxE25RV,CwEl6RM,iBAOI,sBxE85RV,CwEr6RM,iBAOI,2BxEi6RV,CwEx6RM,iBAOI,0BxEo6RV,CwE36RM,iBAOI,yBxEu6RV,CwE96RM,iBAOI,2BxE06RV,CwEj7RM,iBAOI,yBxE66RV,CwEp7RM,eAOI,yBxEg7RV,CwEv7RM,aAOI,0BxEm7RV,CwE17RM,gBAOI,2BxEs7RV,CACF,Ccj8RI,yB0DGI,gBAOI,oBxE27RV,CwEl8RM,cAOI,qBxE87RV,CwEr8RM,eAOI,oBxEi8RV,CwEx8RM,uBAOI,4BxEo8RV,CwE38RM,qBAOI,0BxEu8RV,CwE98RM,oBAOI,yBxE08RV,CwEj9RM,qBAOI,+BxE68RV,CwEp9RM,oBAOI,yBxEg9RV,CwEv9RM,aAOI,wBxEm9RV,CwE19RM,mBAOI,8BxEs9RV,CwE79RM,YAOI,uBxEy9RV,CwEh+RM,WAOI,sBxE49RV,CwEn+RM,kBAOI,6BxE+9RV,CwEt+RM,YAOI,uBxEk+RV,CwEz+RM,gBAOI,2BxEq+RV,CwE5+RM,iBAOI,4BxEw+RV,CwE/+RM,WAOI,sBxE2+RV,CwEl/RM,kBAOI,6BxE8+RV,CwEr/RM,WAOI,sBxEi/RV,CwEx/RM,cAOI,uBxEo/RV,CwE3/RM,aAOI,4BxEu/RV,CwE9/RM,gBAOI,+BxE0/RV,CwEjgSM,qBAOI,oCxE6/RV,CwEpgSM,wBAOI,uCxEggSV,CwEvgSM,gBAOI,qBxEmgSV,CwE1gSM,gBAOI,qBxEsgSV,CwE7gSM,kBAOI,uBxEygSV,CwEhhSM,kBAOI,uBxE4gSV,CwEnhSM,cAOI,wBxE+gSV,CwEthSM,gBAOI,0BxEkhSV,CwEzhSM,sBAOI,gCxEqhSV,CwE5hSM,0BAOI,oCxEwhSV,CwE/hSM,wBAOI,kCxE2hSV,CwEliSM,2BAOI,gCxE8hSV,CwEriSM,4BAOI,uCxEiiSV,CwExiSM,2BAOI,sCxEoiSV,CwE3iSM,2BAOI,sCxEuiSV,CwE9iSM,sBAOI,gCxE0iSV,CwEjjSM,oBAOI,8BxE6iSV,CwEpjSM,uBAOI,4BxEgjSV,CwEvjSM,yBAOI,8BxEmjSV,CwE1jSM,wBAOI,6BxEsjSV,CwE7jSM,wBAOI,kCxEyjSV,CwEhkSM,sBAOI,gCxE4jSV,CwEnkSM,yBAOI,8BxE+jSV,CwEtkSM,0BAOI,qCxEkkSV,CwEzkSM,yBAOI,oCxEqkSV,CwE5kSM,0BAOI,+BxEwkSV,CwE/kSM,oBAOI,yBxE2kSV,CwEllSM,qBAOI,+BxE8kSV,CwErlSM,mBAOI,6BxEilSV,CwExlSM,sBAOI,2BxEolSV,CwE3lSM,wBAOI,6BxEulSV,CwE9lSM,uBAOI,4BxE0lSV,CwEjmSM,gBAOI,kBxE6lSV,CwEpmSM,YAOI,iBxEgmSV,CwEvmSM,YAOI,iBxEmmSV,CwE1mSM,YAOI,iBxEsmSV,CwE7mSM,YAOI,iBxEymSV,CwEhnSM,YAOI,iBxE4mSV,CwEnnSM,YAOI,iBxE+mSV,CwEtnSM,eAOI,iBxEknSV,CwEznSM,QAOI,kBxEqnSV,CwE5nSM,QAOI,uBxEwnSV,CwE/nSM,QAOI,sBxE2nSV,CwEloSM,QAOI,qBxE8nSV,CwEroSM,QAOI,uBxEioSV,CwExoSM,QAOI,qBxEooSV,CwE3oSM,WAOI,qBxEuoSV,CwE9oSM,SAOI,gDxE2oSV,CwElpSM,SAOI,0DxE+oSV,CwEtpSM,SAOI,wDxEmpSV,CwE1pSM,SAOI,sDxEupSV,CwE9pSM,SAOI,0DxE2pSV,CwElqSM,SAOI,sDxE+pSV,CwEtqSM,YAOI,sDxEmqSV,CwE1qSM,SAOI,gDxEuqSV,CwE9qSM,SAOI,0DxE2qSV,CwElrSM,SAOI,wDxE+qSV,CwEtrSM,SAOI,sDxEmrSV,CwE1rSM,SAOI,0DxEurSV,CwE9rSM,SAOI,sDxE2rSV,CwElsSM,YAOI,sDxE+rSV,CwEtsSM,SAOI,sBxEksSV,CwEzsSM,SAOI,2BxEqsSV,CwE5sSM,SAOI,0BxEwsSV,CwE/sSM,SAOI,yBxE2sSV,CwEltSM,SAOI,2BxE8sSV,CwErtSM,SAOI,yBxEitSV,CwExtSM,YAOI,yBxEotSV,CwE3tSM,SAOI,wBxEutSV,CwE9tSM,SAOI,6BxE0tSV,CwEjuSM,SAOI,4BxE6tSV,CwEpuSM,SAOI,2BxEguSV,CwEvuSM,SAOI,6BxEmuSV,CwE1uSM,SAOI,2BxEsuSV,CwE7uSM,YAOI,2BxEyuSV,CwEhvSM,SAOI,yBxE4uSV,CwEnvSM,SAOI,8BxE+uSV,CwEtvSM,SAOI,6BxEkvSV,CwEzvSM,SAOI,4BxEqvSV,CwE5vSM,SAOI,8BxEwvSV,CwE/vSM,SAOI,4BxE2vSV,CwElwSM,YAOI,4BxE8vSV,CwErwSM,SAOI,uBxEiwSV,CwExwSM,SAOI,4BxEowSV,CwE3wSM,SAOI,2BxEuwSV,CwE9wSM,SAOI,0BxE0wSV,CwEjxSM,SAOI,4BxE6wSV,CwEpxSM,SAOI,0BxEgxSV,CwEvxSM,YAOI,0BxEmxSV,CwE1xSM,QAOI,mBxEsxSV,CwE7xSM,QAOI,wBxEyxSV,CwEhySM,QAOI,uBxE4xSV,CwEnySM,QAOI,sBxE+xSV,CwEtySM,QAOI,wBxEkySV,CwEzySM,QAOI,sBxEqySV,CwE5ySM,SAOI,kDxEyySV,CwEhzSM,SAOI,4DxE6ySV,CwEpzSM,SAOI,0DxEizSV,CwExzSM,SAOI,wDxEqzSV,CwE5zSM,SAOI,4DxEyzSV,CwEh0SM,SAOI,wDxE6zSV,CwEp0SM,SAOI,kDxEi0SV,CwEx0SM,SAOI,4DxEq0SV,CwE50SM,SAOI,0DxEy0SV,CwEh1SM,SAOI,wDxE60SV,CwEp1SM,SAOI,4DxEi1SV,CwEx1SM,SAOI,wDxEq1SV,CwE51SM,SAOI,uBxEw1SV,CwE/1SM,SAOI,4BxE21SV,CwEl2SM,SAOI,2BxE81SV,CwEr2SM,SAOI,0BxEi2SV,CwEx2SM,SAOI,4BxEo2SV,CwE32SM,SAOI,0BxEu2SV,CwE92SM,SAOI,yBxE02SV,CwEj3SM,SAOI,8BxE62SV,CwEp3SM,SAOI,6BxEg3SV,CwEv3SM,SAOI,4BxEm3SV,CwE13SM,SAOI,8BxEs3SV,CwE73SM,SAOI,4BxEy3SV,CwEh4SM,SAOI,0BxE43SV,CwEn4SM,SAOI,+BxE+3SV,CwEt4SM,SAOI,8BxEk4SV,CwEz4SM,SAOI,6BxEq4SV,CwE54SM,SAOI,+BxEw4SV,CwE/4SM,SAOI,6BxE24SV,CwEl5SM,SAOI,wBxE84SV,CwEr5SM,SAOI,6BxEi5SV,CwEx5SM,SAOI,4BxEo5SV,CwE35SM,SAOI,2BxEu5SV,CwE95SM,SAOI,6BxE05SV,CwEj6SM,SAOI,2BxE65SV,CwEp6SM,UAOI,exEg6SV,CwEv6SM,UAOI,oBxEm6SV,CwE16SM,UAOI,mBxEs6SV,CwE76SM,UAOI,kBxEy6SV,CwEh7SM,UAOI,oBxE46SV,CwEn7SM,UAOI,kBxE+6SV,CwEt7SM,cAOI,mBxEk7SV,CwEz7SM,cAOI,wBxEq7SV,CwE57SM,cAOI,uBxEw7SV,CwE/7SM,cAOI,sBxE27SV,CwEl8SM,cAOI,wBxE87SV,CwEr8SM,cAOI,sBxEi8SV,CwEx8SM,iBAOI,sBxEo8SV,CwE38SM,iBAOI,2BxEu8SV,CwE98SM,iBAOI,0BxE08SV,CwEj9SM,iBAOI,yBxE68SV,CwEp9SM,iBAOI,2BxEg9SV,CwEv9SM,iBAOI,yBxEm9SV,CwE19SM,eAOI,yBxEs9SV,CwE79SM,aAOI,0BxEy9SV,CwEh+SM,gBAOI,2BxE49SV,CACF,Ccv+SI,yB0DGI,gBAOI,oBxEi+SV,CwEx+SM,cAOI,qBxEo+SV,CwE3+SM,eAOI,oBxEu+SV,CwE9+SM,uBAOI,4BxE0+SV,CwEj/SM,qBAOI,0BxE6+SV,CwEp/SM,oBAOI,yBxEg/SV,CwEv/SM,qBAOI,+BxEm/SV,CwE1/SM,oBAOI,yBxEs/SV,CwE7/SM,aAOI,wBxEy/SV,CwEhgTM,mBAOI,8BxE4/SV,CwEngTM,YAOI,uBxE+/SV,CwEtgTM,WAOI,sBxEkgTV,CwEzgTM,kBAOI,6BxEqgTV,CwE5gTM,YAOI,uBxEwgTV,CwE/gTM,gBAOI,2BxE2gTV,CwElhTM,iBAOI,4BxE8gTV,CwErhTM,WAOI,sBxEihTV,CwExhTM,kBAOI,6BxEohTV,CwE3hTM,WAOI,sBxEuhTV,CwE9hTM,cAOI,uBxE0hTV,CwEjiTM,aAOI,4BxE6hTV,CwEpiTM,gBAOI,+BxEgiTV,CwEviTM,qBAOI,oCxEmiTV,CwE1iTM,wBAOI,uCxEsiTV,CwE7iTM,gBAOI,qBxEyiTV,CwEhjTM,gBAOI,qBxE4iTV,CwEnjTM,kBAOI,uBxE+iTV,CwEtjTM,kBAOI,uBxEkjTV,CwEzjTM,cAOI,wBxEqjTV,CwE5jTM,gBAOI,0BxEwjTV,CwE/jTM,sBAOI,gCxE2jTV,CwElkTM,0BAOI,oCxE8jTV,CwErkTM,wBAOI,kCxEikTV,CwExkTM,2BAOI,gCxEokTV,CwE3kTM,4BAOI,uCxEukTV,CwE9kTM,2BAOI,sCxE0kTV,CwEjlTM,2BAOI,sCxE6kTV,CwEplTM,sBAOI,gCxEglTV,CwEvlTM,oBAOI,8BxEmlTV,CwE1lTM,uBAOI,4BxEslTV,CwE7lTM,yBAOI,8BxEylTV,CwEhmTM,wBAOI,6BxE4lTV,CwEnmTM,wBAOI,kCxE+lTV,CwEtmTM,sBAOI,gCxEkmTV,CwEzmTM,yBAOI,8BxEqmTV,CwE5mTM,0BAOI,qCxEwmTV,CwE/mTM,yBAOI,oCxE2mTV,CwElnTM,0BAOI,+BxE8mTV,CwErnTM,oBAOI,yBxEinTV,CwExnTM,qBAOI,+BxEonTV,CwE3nTM,mBAOI,6BxEunTV,CwE9nTM,sBAOI,2BxE0nTV,CwEjoTM,wBAOI,6BxE6nTV,CwEpoTM,uBAOI,4BxEgoTV,CwEvoTM,gBAOI,kBxEmoTV,CwE1oTM,YAOI,iBxEsoTV,CwE7oTM,YAOI,iBxEyoTV,CwEhpTM,YAOI,iBxE4oTV,CwEnpTM,YAOI,iBxE+oTV,CwEtpTM,YAOI,iBxEkpTV,CwEzpTM,YAOI,iBxEqpTV,CwE5pTM,eAOI,iBxEwpTV,CwE/pTM,QAOI,kBxE2pTV,CwElqTM,QAOI,uBxE8pTV,CwErqTM,QAOI,sBxEiqTV,CwExqTM,QAOI,qBxEoqTV,CwE3qTM,QAOI,uBxEuqTV,CwE9qTM,QAOI,qBxE0qTV,CwEjrTM,WAOI,qBxE6qTV,CwEprTM,SAOI,gDxEirTV,CwExrTM,SAOI,0DxEqrTV,CwE5rTM,SAOI,wDxEyrTV,CwEhsTM,SAOI,sDxE6rTV,CwEpsTM,SAOI,0DxEisTV,CwExsTM,SAOI,sDxEqsTV,CwE5sTM,YAOI,sDxEysTV,CwEhtTM,SAOI,gDxE6sTV,CwEptTM,SAOI,0DxEitTV,CwExtTM,SAOI,wDxEqtTV,CwE5tTM,SAOI,sDxEytTV,CwEhuTM,SAOI,0DxE6tTV,CwEpuTM,SAOI,sDxEiuTV,CwExuTM,YAOI,sDxEquTV,CwE5uTM,SAOI,sBxEwuTV,CwE/uTM,SAOI,2BxE2uTV,CwElvTM,SAOI,0BxE8uTV,CwErvTM,SAOI,yBxEivTV,CwExvTM,SAOI,2BxEovTV,CwE3vTM,SAOI,yBxEuvTV,CwE9vTM,YAOI,yBxE0vTV,CwEjwTM,SAOI,wBxE6vTV,CwEpwTM,SAOI,6BxEgwTV,CwEvwTM,SAOI,4BxEmwTV,CwE1wTM,SAOI,2BxEswTV,CwE7wTM,SAOI,6BxEywTV,CwEhxTM,SAOI,2BxE4wTV,CwEnxTM,YAOI,2BxE+wTV,CwEtxTM,SAOI,yBxEkxTV,CwEzxTM,SAOI,8BxEqxTV,CwE5xTM,SAOI,6BxEwxTV,CwE/xTM,SAOI,4BxE2xTV,CwElyTM,SAOI,8BxE8xTV,CwEryTM,SAOI,4BxEiyTV,CwExyTM,YAOI,4BxEoyTV,CwE3yTM,SAOI,uBxEuyTV,CwE9yTM,SAOI,4BxE0yTV,CwEjzTM,SAOI,2BxE6yTV,CwEpzTM,SAOI,0BxEgzTV,CwEvzTM,SAOI,4BxEmzTV,CwE1zTM,SAOI,0BxEszTV,CwE7zTM,YAOI,0BxEyzTV,CwEh0TM,QAOI,mBxE4zTV,CwEn0TM,QAOI,wBxE+zTV,CwEt0TM,QAOI,uBxEk0TV,CwEz0TM,QAOI,sBxEq0TV,CwE50TM,QAOI,wBxEw0TV,CwE/0TM,QAOI,sBxE20TV,CwEl1TM,SAOI,kDxE+0TV,CwEt1TM,SAOI,4DxEm1TV,CwE11TM,SAOI,0DxEu1TV,CwE91TM,SAOI,wDxE21TV,CwEl2TM,SAOI,4DxE+1TV,CwEt2TM,SAOI,wDxEm2TV,CwE12TM,SAOI,kDxEu2TV,CwE92TM,SAOI,4DxE22TV,CwEl3TM,SAOI,0DxE+2TV,CwEt3TM,SAOI,wDxEm3TV,CwE13TM,SAOI,4DxEu3TV,CwE93TM,SAOI,wDxE23TV,CwEl4TM,SAOI,uBxE83TV,CwEr4TM,SAOI,4BxEi4TV,CwEx4TM,SAOI,2BxEo4TV,CwE34TM,SAOI,0BxEu4TV,CwE94TM,SAOI,4BxE04TV,CwEj5TM,SAOI,0BxE64TV,CwEp5TM,SAOI,yBxEg5TV,CwEv5TM,SAOI,8BxEm5TV,CwE15TM,SAOI,6BxEs5TV,CwE75TM,SAOI,4BxEy5TV,CwEh6TM,SAOI,8BxE45TV,CwEn6TM,SAOI,4BxE+5TV,CwEt6TM,SAOI,0BxEk6TV,CwEz6TM,SAOI,+BxEq6TV,CwE56TM,SAOI,8BxEw6TV,CwE/6TM,SAOI,6BxE26TV,CwEl7TM,SAOI,+BxE86TV,CwEr7TM,SAOI,6BxEi7TV,CwEx7TM,SAOI,wBxEo7TV,CwE37TM,SAOI,6BxEu7TV,CwE97TM,SAOI,4BxE07TV,CwEj8TM,SAOI,2BxE67TV,CwEp8TM,SAOI,6BxEg8TV,CwEv8TM,SAOI,2BxEm8TV,CwE18TM,UAOI,exEs8TV,CwE78TM,UAOI,oBxEy8TV,CwEh9TM,UAOI,mBxE48TV,CwEn9TM,UAOI,kBxE+8TV,CwEt9TM,UAOI,oBxEk9TV,CwEz9TM,UAOI,kBxEq9TV,CwE59TM,cAOI,mBxEw9TV,CwE/9TM,cAOI,wBxE29TV,CwEl+TM,cAOI,uBxE89TV,CwEr+TM,cAOI,sBxEi+TV,CwEx+TM,cAOI,wBxEo+TV,CwE3+TM,cAOI,sBxEu+TV,CwE9+TM,iBAOI,sBxE0+TV,CwEj/TM,iBAOI,2BxE6+TV,CwEp/TM,iBAOI,0BxEg/TV,CwEv/TM,iBAOI,yBxEm/TV,CwE1/TM,iBAOI,2BxEs/TV,CwE7/TM,iBAOI,yBxEy/TV,CwEhgUM,eAOI,yBxE4/TV,CwEngUM,aAOI,0BxE+/TV,CwEtgUM,gBAOI,2BxEkgUV,CACF,Cc7gUI,0B0DGI,gBAOI,oBxEugUV,CwE9gUM,cAOI,qBxE0gUV,CwEjhUM,eAOI,oBxE6gUV,CwEphUM,uBAOI,4BxEghUV,CwEvhUM,qBAOI,0BxEmhUV,CwE1hUM,oBAOI,yBxEshUV,CwE7hUM,qBAOI,+BxEyhUV,CwEhiUM,oBAOI,yBxE4hUV,CwEniUM,aAOI,wBxE+hUV,CwEtiUM,mBAOI,8BxEkiUV,CwEziUM,YAOI,uBxEqiUV,CwE5iUM,WAOI,sBxEwiUV,CwE/iUM,kBAOI,6BxE2iUV,CwEljUM,YAOI,uBxE8iUV,CwErjUM,gBAOI,2BxEijUV,CwExjUM,iBAOI,4BxEojUV,CwE3jUM,WAOI,sBxEujUV,CwE9jUM,kBAOI,6BxE0jUV,CwEjkUM,WAOI,sBxE6jUV,CwEpkUM,cAOI,uBxEgkUV,CwEvkUM,aAOI,4BxEmkUV,CwE1kUM,gBAOI,+BxEskUV,CwE7kUM,qBAOI,oCxEykUV,CwEhlUM,wBAOI,uCxE4kUV,CwEnlUM,gBAOI,qBxE+kUV,CwEtlUM,gBAOI,qBxEklUV,CwEzlUM,kBAOI,uBxEqlUV,CwE5lUM,kBAOI,uBxEwlUV,CwE/lUM,cAOI,wBxE2lUV,CwElmUM,gBAOI,0BxE8lUV,CwErmUM,sBAOI,gCxEimUV,CwExmUM,0BAOI,oCxEomUV,CwE3mUM,wBAOI,kCxEumUV,CwE9mUM,2BAOI,gCxE0mUV,CwEjnUM,4BAOI,uCxE6mUV,CwEpnUM,2BAOI,sCxEgnUV,CwEvnUM,2BAOI,sCxEmnUV,CwE1nUM,sBAOI,gCxEsnUV,CwE7nUM,oBAOI,8BxEynUV,CwEhoUM,uBAOI,4BxE4nUV,CwEnoUM,yBAOI,8BxE+nUV,CwEtoUM,wBAOI,6BxEkoUV,CwEzoUM,wBAOI,kCxEqoUV,CwE5oUM,sBAOI,gCxEwoUV,CwE/oUM,yBAOI,8BxE2oUV,CwElpUM,0BAOI,qCxE8oUV,CwErpUM,yBAOI,oCxEipUV,CwExpUM,0BAOI,+BxEopUV,CwE3pUM,oBAOI,yBxEupUV,CwE9pUM,qBAOI,+BxE0pUV,CwEjqUM,mBAOI,6BxE6pUV,CwEpqUM,sBAOI,2BxEgqUV,CwEvqUM,wBAOI,6BxEmqUV,CwE1qUM,uBAOI,4BxEsqUV,CwE7qUM,gBAOI,kBxEyqUV,CwEhrUM,YAOI,iBxE4qUV,CwEnrUM,YAOI,iBxE+qUV,CwEtrUM,YAOI,iBxEkrUV,CwEzrUM,YAOI,iBxEqrUV,CwE5rUM,YAOI,iBxEwrUV,CwE/rUM,YAOI,iBxE2rUV,CwElsUM,eAOI,iBxE8rUV,CwErsUM,QAOI,kBxEisUV,CwExsUM,QAOI,uBxEosUV,CwE3sUM,QAOI,sBxEusUV,CwE9sUM,QAOI,qBxE0sUV,CwEjtUM,QAOI,uBxE6sUV,CwEptUM,QAOI,qBxEgtUV,CwEvtUM,WAOI,qBxEmtUV,CwE1tUM,SAOI,gDxEutUV,CwE9tUM,SAOI,0DxE2tUV,CwEluUM,SAOI,wDxE+tUV,CwEtuUM,SAOI,sDxEmuUV,CwE1uUM,SAOI,0DxEuuUV,CwE9uUM,SAOI,sDxE2uUV,CwElvUM,YAOI,sDxE+uUV,CwEtvUM,SAOI,gDxEmvUV,CwE1vUM,SAOI,0DxEuvUV,CwE9vUM,SAOI,wDxE2vUV,CwElwUM,SAOI,sDxE+vUV,CwEtwUM,SAOI,0DxEmwUV,CwE1wUM,SAOI,sDxEuwUV,CwE9wUM,YAOI,sDxE2wUV,CwElxUM,SAOI,sBxE8wUV,CwErxUM,SAOI,2BxEixUV,CwExxUM,SAOI,0BxEoxUV,CwE3xUM,SAOI,yBxEuxUV,CwE9xUM,SAOI,2BxE0xUV,CwEjyUM,SAOI,yBxE6xUV,CwEpyUM,YAOI,yBxEgyUV,CwEvyUM,SAOI,wBxEmyUV,CwE1yUM,SAOI,6BxEsyUV,CwE7yUM,SAOI,4BxEyyUV,CwEhzUM,SAOI,2BxE4yUV,CwEnzUM,SAOI,6BxE+yUV,CwEtzUM,SAOI,2BxEkzUV,CwEzzUM,YAOI,2BxEqzUV,CwE5zUM,SAOI,yBxEwzUV,CwE/zUM,SAOI,8BxE2zUV,CwEl0UM,SAOI,6BxE8zUV,CwEr0UM,SAOI,4BxEi0UV,CwEx0UM,SAOI,8BxEo0UV,CwE30UM,SAOI,4BxEu0UV,CwE90UM,YAOI,4BxE00UV,CwEj1UM,SAOI,uBxE60UV,CwEp1UM,SAOI,4BxEg1UV,CwEv1UM,SAOI,2BxEm1UV,CwE11UM,SAOI,0BxEs1UV,CwE71UM,SAOI,4BxEy1UV,CwEh2UM,SAOI,0BxE41UV,CwEn2UM,YAOI,0BxE+1UV,CwEt2UM,QAOI,mBxEk2UV,CwEz2UM,QAOI,wBxEq2UV,CwE52UM,QAOI,uBxEw2UV,CwE/2UM,QAOI,sBxE22UV,CwEl3UM,QAOI,wBxE82UV,CwEr3UM,QAOI,sBxEi3UV,CwEx3UM,SAOI,kDxEq3UV,CwE53UM,SAOI,4DxEy3UV,CwEh4UM,SAOI,0DxE63UV,CwEp4UM,SAOI,wDxEi4UV,CwEx4UM,SAOI,4DxEq4UV,CwE54UM,SAOI,wDxEy4UV,CwEh5UM,SAOI,kDxE64UV,CwEp5UM,SAOI,4DxEi5UV,CwEx5UM,SAOI,0DxEq5UV,CwE55UM,SAOI,wDxEy5UV,CwEh6UM,SAOI,4DxE65UV,CwEp6UM,SAOI,wDxEi6UV,CwEx6UM,SAOI,uBxEo6UV,CwE36UM,SAOI,4BxEu6UV,CwE96UM,SAOI,2BxE06UV,CwEj7UM,SAOI,0BxE66UV,CwEp7UM,SAOI,4BxEg7UV,CwEv7UM,SAOI,0BxEm7UV,CwE17UM,SAOI,yBxEs7UV,CwE77UM,SAOI,8BxEy7UV,CwEh8UM,SAOI,6BxE47UV,CwEn8UM,SAOI,4BxE+7UV,CwEt8UM,SAOI,8BxEk8UV,CwEz8UM,SAOI,4BxEq8UV,CwE58UM,SAOI,0BxEw8UV,CwE/8UM,SAOI,+BxE28UV,CwEl9UM,SAOI,8BxE88UV,CwEr9UM,SAOI,6BxEi9UV,CwEx9UM,SAOI,+BxEo9UV,CwE39UM,SAOI,6BxEu9UV,CwE99UM,SAOI,wBxE09UV,CwEj+UM,SAOI,6BxE69UV,CwEp+UM,SAOI,4BxEg+UV,CwEv+UM,SAOI,2BxEm+UV,CwE1+UM,SAOI,6BxEs+UV,CwE7+UM,SAOI,2BxEy+UV,CwEh/UM,UAOI,exE4+UV,CwEn/UM,UAOI,oBxE++UV,CwEt/UM,UAOI,mBxEk/UV,CwEz/UM,UAOI,kBxEq/UV,CwE5/UM,UAOI,oBxEw/UV,CwE//UM,UAOI,kBxE2/UV,CwElgVM,cAOI,mBxE8/UV,CwErgVM,cAOI,wBxEigVV,CwExgVM,cAOI,uBxEogVV,CwE3gVM,cAOI,sBxEugVV,CwE9gVM,cAOI,wBxE0gVV,CwEjhVM,cAOI,sBxE6gVV,CwEphVM,iBAOI,sBxEghVV,CwEvhVM,iBAOI,2BxEmhVV,CwE1hVM,iBAOI,0BxEshVV,CwE7hVM,iBAOI,yBxEyhVV,CwEhiVM,iBAOI,2BxE4hVV,CwEniVM,iBAOI,yBxE+hVV,CwEtiVM,eAOI,yBxEkiVV,CwEziVM,aAOI,0BxEqiVV,CwE5iVM,gBAOI,2BxEwiVV,CwE/iVM,MAOI,0BxE6iVV,CwEpjVM,MAOI,wBxEgjVV,CwEvjVM,MAOI,2BxEmjVV,CwE1jVM,MAOI,0BxEsjVV,CAbF,CyE5kVA,aD4BQ,gBAOI,wBxE2jVV,CwElkVM,sBAOI,8BxE8jVV,CwErkVM,eAOI,uBxEikVV,CwExkVM,cAOI,sBxEokVV,CwE3kVM,qBAOI,6BxEukVV,CwE9kVM,eAOI,uBxE0kVV,CwEjlVM,mBAOI,2BxE6kVV,CwEplVM,oBAOI,4BxEglVV,CwEvlVM,cAOI,sBxEmlVV,CwE1lVM,qBAOI,6BxEslVV,CwE7lVM,cAOI,sBxEylVV,CACF;A0EnqVA;;;;EAAA,CCGA,IACE,yDACA,+B3EuqVF,C2EpqVA,0EASE,iCAAkC,CAClC,kCAAmC,CACnC,uCACA,iBAAkB,CAClB,mBAAoB,CACpB,aAAc,CACd,mB3EuqVF,C2EpqVA,4CAKE,gC3EuqVF,C2EpqVA,gBAEE,kC3EuqVF,C4EtsVE,OACE,a5EysVJ,C4E1sVE,OACE,a5E6sVJ,C4E9sVE,OACE,a5EitVJ,C4EltVE,OACE,a5EqtVJ,C4EttVE,OACE,a5EytVJ,C4E1tVE,OACE,a5E6tVJ,C4E9tVE,OACE,a5EiuVJ,C4EluVE,OACE,a5EquVJ,C4EtuVE,OACE,a5EyuVJ,C4E1uVE,QACE,c5E6uVJ,C4EvuVE,QCIA,iBACA,iBACA,qB7EuuVF,C4E7uVE,OCIA,gBACA,0BACA,qB7E6uVF,C4EnvVE,OCIA,iBACA,0BACA,4B7EmvVF,C4EzvVE,OCIA,iBACA,kBACA,sB7EyvVF,C4E/vVE,OCIA,gBACA,0BACA,sB7E+vVF,C4ErwVE,QCIA,cACA,qBACA,uB7EqwVF,C8EpxVA,OACE,iBAAkB,CAClB,Y9EuxVF,C+EzxVA,OACE,oBAAqB,CACrB,sCACA,c/E4xVF,C+E1xVE,UAAO,iB/E6xVT,C+E1xVA,OACE,sCAIA,mBAAoB,CAHpB,iBAAkB,CAClB,iBAAkB,CAClB,4B/E8xVF,CgF1yVA,WACE,yCACA,2CACA,0CACA,0CACA,iDhF6yVF,CgF1yVA,cACE,UAAW,CACX,uChF6yVF,CgF1yVA,eACE,WAAY,CACZ,sChF6yVF,CiF5zVA,SAEE,6CACA,yDACA,mDACA,uEAJA,uBAKA,gEjF+zVF,CiF5zVA,WAEE,6CACA,yDACA,mDACA,uEAJA,yBAKA,gFjF+zVF,CiF5zVA,SAKE,uEAJA,uBAKA,4EjF+zVF,CiF5zVA,uBAPE,6CACA,yDACA,kDjF00VF,CiFr0VA,cAKE,uEAJA,4BAKA,4EjF+zVF,CiF5zVA,SAEE,6CACA,yDACA,mDACA,uEAJA,uBAKA,gEjF+zVF,CiF5zVA,UAIE,mDACA,uEAJA,wBAKA,2DjF+zVF,CiF5zVA,mBAPE,6CACA,wDjF20VF,CiFr0VA,SAIE,mDACA,uEAJA,uBAKA,2DjF+zVF,CiF5zVA,iBACE,gCjF+zVF,CiF5zVA,yBAGE,yDACA,mDACA,uEAHA,uBAIA,6DjF+zVF,CiFzzVA,uCACE,gGASE,oBAAqB,CACrB,sBAAuB,CACvB,2BAA4B,CAC5B,mBAAoB,CACpB,sBjF4zVF,CACF,CiFzzVA,mBACE,OAAU,kBjF4zVV,CiF3zVA,IAAM,0CjF8zVN,CACF,CiF5zVA,qBACE,GAAO,gCjF+zVP,CiF9zVA,IAAO,mGjFi0VP,CiFh0VA,IAAO,6HjFm0VP,CiFl0VA,IAAO,mGjFq0VP,CiFp0VA,IAAO,+DjFu0VP,CiFt0VA,IAAO,gCjFy0VP,CiFx0VA,GAAO,gCjF20VP,CACF,CiFz0VA,mBACE,IAAM,iCjF40VN,CACF,CiF10VA,wBACE,MACE,uCACA,kBjF40VF,CiF10VA,IACE,SAAU,CACV,gDjF40VF,CACF,CiFz0VA,mBACE,IACE,yGjF20VF,CACF,CiFx0VA,oBACE,GAAK,wBjF20VL,CiF10VA,GAAK,uBjF60VL,CiF50VA,OAAU,wBjF+0VV,CiF90VA,QAAW,uBjFi1VX,CiFh1VA,IAAM,wBjFm1VN,CiFl1VA,IAAM,uBjFq1VN,CiFp1VA,IAAM,wBjFu1VN,CiFt1VA,IAAM,uBjFy1VN,CiFx1VA,OAAY,sBjF21VZ,CACF,CiFz1VA,mBACE,GAAK,sBjF41VL,CiF31VA,GAAO,uBjF81VP,CACF,CkFl/VA,cACE,uBlFo/VF,CkFj/VA,eACE,wBlFo/VF,CkFj/VA,eACE,wBlFo/VF,CkFj/VA,oBACE,oBlFo/VF,CkFj/VA,kBACE,oBlFo/VF,CkFj/VA,mDAEE,mBlFo/VF,CkFj/VA,cACE,0ClFo/VF,CmF9gWA,UACE,oBAAqB,CACrB,UAAW,CACX,eAAgB,CAChB,iBAAkB,CAClB,qBCmCwB,CDlCxB,WnFihWF,CmF9gWA,0BAEE,MAAO,CACP,iBAAkB,CAClB,iBAAkB,CAClB,UAAW,CACX,oCnFihWF,CmF9gWA,aACE,mBnFihWF,CmF9gWA,aACE,anFihWF,CmF9gWA,YACE,4BnFihWF,CqFxiWE,aAAsC,crF8iWxC,CqF9iWE,aAAsC,crFkjWxC,CqFljWE,aAAsC,crFsjWxC,CqFtjWE,aAAsC,crF0jWxC,CqF1jWE,aAAsC,crF8jWxC,CqF9jWE,aAAsC,crFkkWxC,CqFlkWE,aAAsC,crFskWxC,CqFtkWE,aAAsC,crF0kWxC,CqF1kWE,aAAsC,crF8kWxC,CqF9kWE,aAAsC,crFklWxC,CqFllWE,qBAAsC,erFslWxC,CqFtlWE,4BAAsC,erF0lWxC,CqF1lWE,gEAAsC,erFkmWxC,CqFlmWE,cAAsC,YrFsmWxC,CqFtmWE,0CAAsC,erF8mWxC,CqF9mWE,uBAAsC,erFknWxC,CqFlnWE,4CAAsC,erF0nWxC,CqF1nWE,uBAAsC,erF8nWxC,CqF9nWE,0CAAsC,erFsoWxC,CqFtoWE,gBAAsC,erF0oWxC,CqF1oWE,oEAAsC,erFkpWxC,CqFlpWE,qBAAsC,erFspWxC,CqFtpWE,+DAAsC,erFkqWxC,CqFlqWE,uBAAsC,erFsqWxC,CqFtqWE,gBAAsC,erF0qWxC,CqF1qWE,uBAAsC,erF8qWxC,CqF9qWE,sBAAsC,erFkrWxC,CqFlrWE,iDAAsC,erF0rWxC,CqF1rWE,+BAAsC,erF8rWxC,CqF9rWE,uCAAsC,erFksWxC,CqFlsWE,iDAAsC,erF0sWxC,CqF1sWE,6BAAsC,erF8sWxC,CqF9sWE,0CAAsC,erFstWxC,CqFttWE,gDAAsC,erF8tWxC,CqF9tWE,qBAAsC,erFkuWxC,CqFluWE,sDAAsC,erF0uWxC,CqF1uWE,gBAAsC,erF8uWxC,CqF9uWE,gBAAsC,erFkvWxC,CqFlvWE,oDAAsC,erF0vWxC,CqF1vWE,kEAAsC,erFkwWxC,CqFlwWE,+BAAsC,erFswWxC,CqFtwWE,wBAAsC,erF0wWxC,CqF1wWE,uBAAsC,erF8wWxC,CqF9wWE,4BAAsC,erFkxWxC,CqFlxWE,gEAAsC,erF0xWxC,CqF1xWE,gDAAsC,erFkyWxC,CqFlyWE,iDAAsC,erF0yWxC,CqF1yWE,0BAAsC,erF8yWxC,CqF9yWE,6CAAsC,erFszWxC,CqFtzWE,qCAAsC,erF0zWxC,CqF1zWE,gBAAsC,erF8zWxC,CqF9zWE,oDAAsC,erFs0WxC,CqFt0WE,2BAAsC,erF00WxC,CqF10WE,uBAAsC,erF80WxC,CqF90WE,uBAAsC,erFk1WxC,CqFl1WE,0CAAsC,erF01WxC,CqF11WE,+BAAsC,erF81WxC,CqF91WE,kDAAsC,erFs2WxC,CqFt2WE,6CAAsC,erF82WxC,CqF92WE,kCAAsC,erFs3WxC,CqFt3WE,kBAAsC,erF03WxC,CqF13WE,oBAAsC,erF83WxC,CqF93WE,mBAAsC,erFk4WxC,CqFl4WE,uBAAsC,erFs4WxC,CqFt4WE,wBAAsC,erF04WxC,CqF14WE,iDAAsC,erFk5WxC,CqFl5WE,gBAAsC,erFs5WxC,CqFt5WE,8DAAsC,erF85WxC,CqF95WE,qCAAsC,erFk6WxC,CqFl6WE,0CAAsC,erF06WxC,CqF16WE,qBAAsC,erF86WxC,CqF96WE,+CAAsC,erFs7WxC,CqFt7WE,mCAAsC,erF07WxC,CqF17WE,+CAAsC,erFk8WxC,CqFl8WE,iBAAsC,erFs8WxC,CqFt8WE,gDAAsC,erF88WxC,CqF98WE,yBAAsC,erFk9WxC,CqFl9WE,aAAsC,WrFs9WxC,CqFt9WE,yBAAsC,erF09WxC,CqF19WE,wCAAsC,erFk+WxC,CqFl+WE,oBAAsC,erFs+WxC,CqFt+WE,uBAAsC,erF0+WxC,CqF1+WE,oBAAsC,erF8+WxC,CqF9+WE,2CAAsC,erFs/WxC,CqFt/WE,6BAAsC,erF0/WxC,CqF1/WE,0BAAsC,erF8/WxC,CqF9/WE,mDAAsC,erFsgXxC,CqFtgXE,sBAAsC,erF0gXxC,CqF1gXE,sBAAsC,erF8gXxC,CqF9gXE,yBAAsC,erFkhXxC,CqFlhXE,gBAAsC,erFshXxC,CqFthXE,sBAAsC,erF0hXxC,CqF1hXE,qCAAsC,erF8hXxC,CqF9hXE,yCAAsC,erFsiXxC,CqFtiXE,2EAAsC,erFkjXxC,CqFljXE,sBAAsC,erFsjXxC,CqFtjXE,0CAAsC,erF8jXxC,CqF9jXE,2BAAsC,erFkkXxC,CqFlkXE,wFAAsC,erF8kXxC,CqF9kXE,iEAAsC,erFslXxC,CqFtlXE,qBAAsC,erF0lXxC,CqF1lXE,yBAAsC,erF8lXxC,CqF9lXE,aAAsC,WrFkmXxC,CqFlmXE,gDAAsC,erF0mXxC,CqF1mXE,mBAAsC,erF8mXxC,CqF9mXE,wBAAsC,erFknXxC,CqFlnXE,sDAAsC,erF0nXxC,CqF1nXE,eAAsC,erF8nXxC,CqF9nXE,iBAAsC,erFkoXxC,CqFloXE,yBAAsC,erFsoXxC,CqFtoXE,2CAAsC,erF8oXxC,CqF9oXE,yBAAsC,erFkpXxC,CqFlpXE,iCAAsC,erFspXxC,CqFtpXE,kCAAsC,erF8pXxC,CqF9pXE,sBAAsC,erFkqXxC,CqFlqXE,4BAAsC,erFsqXxC,CqFtqXE,sBAAsC,erF0qXxC,CqF1qXE,sBAAsC,erF8qXxC,CqF9qXE,oDAAsC,erFsrXxC,CqFtrXE,6CAAsC,erF8rXxC,CqF9rXE,6DAAsC,erFssXxC,CqFtsXE,gCAAsC,erF0sXxC,CqF1sXE,4DAAsC,erFktXxC,CqFltXE,mCAAsC,erF0tXxC,CqF1tXE,0BAAsC,erF8tXxC,CqF9tXE,4BAAsC,erFkuXxC,CqFluXE,uBAAsC,erFsuXxC,CqFtuXE,iBAAsC,erF0uXxC,CqF1uXE,wCAAsC,erF8uXxC,CqF9uXE,4DAAsC,erFsvXxC,CqFtvXE,uBAAsC,erF0vXxC,CqF1vXE,kBAAsC,erF8vXxC,CqF9vXE,uCAAsC,erFswXxC,CqFtwXE,6BAAsC,erF0wXxC,CqF1wXE,sBAAsC,erF8wXxC,CqF9wXE,gDAAsC,erFsxXxC,CqFtxXE,sBAAsC,erF0xXxC,CqF1xXE,mBAAsC,erF8xXxC,CqF9xXE,0DAAsC,erFsyXxC,CqFtyXE,gCAAsC,erF0yXxC,CqF1yXE,uBAAsC,erF8yXxC,CqF9yXE,sBAAsC,erFkzXxC,CqFlzXE,+BAAsC,erFszXxC,CqFtzXE,mCAAsC,erF8zXxC,CqF9zXE,wBAAsC,erFk0XxC,CqFl0XE,gDAAsC,erF00XxC,CqF10XE,+CAAsC,erFk1XxC,CqFl1XE,+DAAsC,erF01XxC,CqF11XE,+CAAsC,erFk2XxC,CqFl2XE,gBAAsC,erFs2XxC,CqFt2XE,4BAAsC,erF02XxC,CqF12XE,0BAAsC,erF82XxC,CqF92XE,qBAAsC,erFk3XxC,CqFl3XE,sBAAsC,erFs3XxC,CqFt3XE,4DAAsC,erF83XxC,CqF93XE,+BAAsC,erFk4XxC,CqFl4XE,wDAAsC,erF04XxC,CqF14XE,8DAAsC,erFk5XxC,CqFl5XE,gDAAsC,erF05XxC,CqF15XE,+BAAsC,erF85XxC,CqF95XE,uDAAsC,erFs6XxC,CqFt6XE,mBAAsC,erF06XxC,CqF16XE,mDAAsC,erFk7XxC,CqFl7XE,kBAAsC,erFs7XxC,CqFt7XE,uBAAsC,erF07XxC,CqF17XE,uBAAsC,erF87XxC,CqF97XE,gBAAsC,erFk8XxC,CqFl8XE,oBAAsC,erFs8XxC,CqFt8XE,mBAAsC,erF08XxC,CqF18XE,6CAAsC,erFk9XxC,CqFl9XE,yBAAsC,erFs9XxC,CqFt9XE,0BAAsC,erF09XxC,CqF19XE,oBAAsC,erF89XxC,CqF99XE,0CAAsC,erFs+XxC,CqFt+XE,gBAAsC,erF0+XxC,CqF1+XE,+BAAsC,erF8+XxC,CqF9+XE,oBAAsC,YrFk/XxC,CqFl/XE,gDAAsC,erF0/XxC,CqF1/XE,uBAAsC,erF8/XxC,CqF9/XE,qCAAsC,erFsgYxC,CqFtgYE,iBAAsC,erF0gYxC,CqF1gYE,wCAAsC,erFkhYxC,CqFlhYE,yDAAsC,erF0hYxC,CqF1hYE,qBAAsC,erF8hYxC,CqF9hYE,mBAAsC,erFkiYxC,CqFliYE,4CAAsC,erF0iYxC,CqF1iYE,uCAAsC,erFkjYxC,CqFljYE,eAAsC,erFsjYxC,CqFtjYE,sCAAsC,erF0jYxC,CqF1jYE,sBAAsC,erF8jYxC,CqF9jYE,uBAAsC,erFkkYxC,CqFlkYE,gEAAsC,erF8kYxC,CqF9kYE,6CAAsC,erFslYxC,CqFtlYE,gBAAsC,erF0lYxC,CqF1lYE,oBAAsC,erF8lYxC,CqF9lYE,6CAAsC,erFsmYxC,CqFtmYE,8CAAsC,erF8mYxC,CqF9mYE,6BAAsC,erFknYxC,CqFlnYE,gBAAsC,erFsnYxC,CqFtnYE,uBAAsC,erF0nYxC,CqF1nYE,uBAAsC,erF8nYxC,CqF9nYE,yCAAsC,erFsoYxC,CqFtoYE,oBAAsC,erF0oYxC,CqF1oYE,4CAAsC,erFkpYxC,CqFlpYE,8BAAsC,erFspYxC,CqFtpYE,iDAAsC,erF8pYxC,CqF9pYE,sBAAsC,erFkqYxC,CqFlqYE,gBAAsC,erFsqYxC,CqFtqYE,gDAAsC,erF8qYxC,CqF9qYE,oFAAsC,erF0rYxC,CqF1rYE,kBAAsC,erF8rYxC,CqF9rYE,uBAAsC,erFksYxC,CqFlsYE,iBAAsC,erFssYxC,CqFtsYE,qBAAsC,erF0sYxC,CqF1sYE,sBAAsC,erF8sYxC,CqF9sYE,sCAAsC,erFstYxC,CqFttYE,cAAsC,erF0tYxC,CqF1tYE,gBAAsC,erF8tYxC,CqF9tYE,8BAAsC,erFkuYxC,CqFluYE,gBAAsC,erFsuYxC,CqFtuYE,sBAAsC,erF0uYxC,CqF1uYE,sBAAsC,erF8uYxC,CqF9uYE,qBAAsC,erFkvYxC,CqFlvYE,qCAAsC,erF0vYxC,CqF1vYE,+BAAsC,erF8vYxC,CqF9vYE,mDAAsC,erFswYxC,CqFtwYE,wBAAsC,erF0wYxC,CqF1wYE,+CAAsC,erFkxYxC,CqFlxYE,mFAAsC,erFkyYxC,CqFlyYE,6BAAsC,erFsyYxC,CqFtyYE,qBAAsC,erF0yYxC,CqF1yYE,4CAAsC,erFkzYxC,CqFlzYE,wBAAsC,erFszYxC,CqFtzYE,oBAAsC,erF0zYxC,CqF1zYE,iBAAsC,erF8zYxC,CqF9zYE,kBAAsC,erFk0YxC,CqFl0YE,aAAsC,WrFs0YxC,CqFt0YE,2CAAsC,erF80YxC,CqF90YE,iBAAsC,erFk1YxC,CqFl1YE,8CAAsC,erF01YxC,CqF11YE,iBAAsC,erF81YxC,CqF91YE,aAAsC,WrFk2YxC,CqFl2YE,iCAAsC,erFs2YxC,CqFt2YE,mBAAsC,erF02YxC,CqF12YE,kGAAsC,erF03YxC,CqF13YE,iCAAsC,erF83YxC,CqF93YE,8CAAsC,erFs4YxC,CqFt4YE,wCAAsC,erF04YxC,CqF14YE,mBAAsC,erF84YxC,CqF94YE,yBAAsC,erFk5YxC,CqFl5YE,wBAAsC,erFs5YxC,CqFt5YE,kBAAsC,erF05YxC,CqF15YE,iBAAsC,erF85YxC,CqF95YE,iBAAsC,erFk6YxC,CqFl6YE,uBAAsC,erFs6YxC,CqFt6YE,iBAAsC,erF06YxC,CqF16YE,aAAsC,WrF86YxC,CqF96YE,aAAsC,WrFk7YxC,CqFl7YE,yBAAsC,erFs7YxC,CqFt7YE,0GAAsC,erFs8YxC,CqFt8YE,qBAAsC,erF08YxC,CqF18YE,oBAAsC,erF88YxC,CqF98YE,yCAAsC,erFs9YxC,CqFt9YE,iDAAsC,erF89YxC,CqF99YE,yBAAsC,erFk+YxC,CqFl+YE,sBAAsC,erFs+YxC,CqFt+YE,oBAAsC,erF0+YxC,CqF1+YE,yBAAsC,erF8+YxC,CqF9+YE,0BAAsC,erFk/YxC,CqFl/YE,qBAAsC,erFs/YxC,CqFt/YE,oBAAsC,erF0/YxC,CqF1/YE,iBAAsC,erF8/YxC,CqF9/YE,2BAAsC,erFkgZxC,CqFlgZE,iCAAsC,erFsgZxC,CqFtgZE,0CAAsC,erF8gZxC,CqF9gZE,qDAAsC,erFshZxC,CqFthZE,qBAAsC,erF0hZxC,CqF1hZE,wBAAsC,erF8hZxC,CqF9hZE,uBAAsC,erFkiZxC,CqFliZE,gBAAsC,erFsiZxC,CqFtiZE,6FAAsC,erFsjZxC,CqFtjZE,8DAAsC,erF8jZxC,CqF9jZE,aAAsC,WrFkkZxC,CqFlkZE,uCAAsC,erF0kZxC,CqF1kZE,qCAAsC,erF8kZxC,CqF9kZE,gBAAsC,erFklZxC,CqFllZE,+BAAsC,erFslZxC,CqFtlZE,oBAAsC,erF0lZxC,CqF1lZE,8CAAsC,erFkmZxC,CqFlmZE,yBAAsC,erFsmZxC,CqFtmZE,0DAAsC,erF8mZxC,CqF9mZE,uBAAsC,erFknZxC,CqFlnZE,eAAsC,erFsnZxC,CqFtnZE,oBAAsC,erF0nZxC,CqF1nZE,iBAAsC,erF8nZxC,CqF9nZE,qBAAsC,erFkoZxC,CqFloZE,6BAAsC,erFsoZxC,CqFtoZE,uBAAsC,erF0oZxC,CqF1oZE,0BAAsC,erF8oZxC,CqF9oZE,aAAsC,WrFkpZxC,CqFlpZE,yBAAsC,erFspZxC,CqFtpZE,yBAAsC,erF0pZxC,CqF1pZE,uBAAsC,erF8pZxC,CqF9pZE,2BAAsC,erFkqZxC,CqFlqZE,4CAAsC,erF0qZxC,CqF1qZE,sBAAsC,erF8qZxC,CqF9qZE,+BAAsC,erFkrZxC,CqFlrZE,iCAAsC,erF0rZxC,CqF1rZE,yBAAsC,erF8rZxC,CqF9rZE,uDAAsC,erFssZxC,CqFtsZE,gBAAsC,erF0sZxC,CqF1sZE,kBAAsC,erF8sZxC,CqF9sZE,iBAAsC,erFktZxC,CqFltZE,eAAsC,erFstZxC,CqFttZE,sBAAsC,erF0tZxC,CqF1tZE,kDAAsC,erFkuZxC,CqFluZE,iDAAsC,erF0uZxC,CqF1uZE,4BAAsC,erF8uZxC,CqF9uZE,oDAAsC,erFsvZxC,CqFtvZE,qCAAsC,erF8vZxC,CqF9vZE,6CAAsC,erFswZxC,CqFtwZE,wBAAsC,erF0wZxC,CqF1wZE,+BAAsC,erF8wZxC,CqF9wZE,kBAAsC,erFkxZxC,CqFlxZE,uBAAsC,erFsxZxC,CqFtxZE,+BAAsC,erF0xZxC,CqF1xZE,oCAAsC,erF8xZxC,CqF9xZE,iBAAsC,erFkyZxC,CqFlyZE,uBAAsC,erFsyZxC,CqFtyZE,iBAAsC,erF0yZxC,CqF1yZE,2CAAsC,erFkzZxC,CqFlzZE,qDAAsC,erF0zZxC,CqF1zZE,iDAAsC,erFk0ZxC,CqFl0ZE,kBAAsC,erFs0ZxC,CqFt0ZE,6BAAsC,erF00ZxC,CqF10ZE,qBAAsC,erF80ZxC,CqF90ZE,kDAAsC,erFs1ZxC,CqFt1ZE,yBAAsC,erF01ZxC,CqF11ZE,0BAAsC,erF81ZxC,CqF91ZE,gCAAsC,erFk2ZxC,CqFl2ZE,mDAAsC,erF02ZxC,CqF12ZE,iBAAsC,erF82ZxC,CqF92ZE,oBAAsC,erFk3ZxC,CqFl3ZE,4CAAsC,erF03ZxC,CqF13ZE,oDAAsC,erFk4ZxC,CqFl4ZE,4BAAsC,erFs4ZxC,CqFt4ZE,qBAAsC,erF04ZxC,CqF14ZE,iBAAsC,erF84ZxC,CqF94ZE,0BAAsC,erFk5ZxC,CqFl5ZE,wBAAsC,erFs5ZxC,CqFt5ZE,6BAAsC,erF05ZxC,CqF15ZE,0CAAsC,erFk6ZxC,CqFl6ZE,8CAAsC,erF06ZxC,CqF16ZE,iDAAsC,erFk7ZxC,CqFl7ZE,wBAAsC,erFs7ZxC,CqFt7ZE,qDAAsC,erF87ZxC,CqF97ZE,sBAAsC,erFk8ZxC,CqFl8ZE,sDAAsC,erF08ZxC,CqF18ZE,qBAAsC,erF88ZxC,CqF98ZE,6CAAsC,erFs9ZxC,CqFt9ZE,aAAsC,WrF09ZxC,CqF19ZE,gBAAsC,erF89ZxC,CqF99ZE,2CAAsC,erFs+ZxC,CqFt+ZE,kDAAsC,erF8+ZxC,CqF9+ZE,uCAAsC,erFs/ZxC,CqFt/ZE,uBAAsC,erF0/ZxC,CqF1/ZE,6BAAsC,erF8/ZxC,CqF9/ZE,qBAAsC,erFkgaxC,CqFlgaE,gBAAsC,erFsgaxC,CqFtgaE,sDAAsC,erF8gaxC,CqF9gaE,sBAAsC,erFkhaxC,CqFlhaE,kBAAsC,erFshaxC,CqFthaE,sBAAsC,erF0haxC,CqF1haE,mBAAsC,erF8haxC,CqF9haE,uBAAsC,erFkiaxC,CqFliaE,6BAAsC,erFsiaxC,CqFtiaE,sBAAsC,erF0iaxC,CqF1iaE,mDAAsC,erFkjaxC,CqFljaE,qDAAsC,erF0jaxC,CqF1jaE,qBAAsC,erF8jaxC,CqF9jaE,oCAAsC,erFkkaxC,CqFlkaE,gDAAsC,erF0kaxC,CqF1kaE,yDAAsC,erFslaxC,CqFtlaE,mCAAsC,erF8laxC,CqF9laE,qBAAsC,erFkmaxC,CqFlmaE,yCAAsC,erF0maxC,CqF1maE,qBAAsC,erF8maxC,CqF9maE,uBAAsC,erFknaxC,CqFlnaE,gBAAsC,erFsnaxC,CqFtnaE,uBAAsC,erF0naxC,CqF1naE,wBAAsC,erF8naxC,CqF9naE,sBAAsC,erFkoaxC,CqFloaE,qBAAsC,erFsoaxC,CqFtoaE,mDAAsC,erF8oaxC,CqF9oaE,sDAAsC,erFspaxC,CqFtpaE,gBAAsC,erF0paxC,CqF1paE,gDAAsC,erFkqaxC,CqFlqaE,mBAAsC,erFsqaxC,CqFtqaE,8CAAsC,erF8qaxC,CqF9qaE,gBAAsC,erFkraxC,CqFlraE,oBAAsC,erFsraxC,CqFtraE,uBAAsC,erF0raxC,CqF1raE,mBAAsC,erF8raxC,CqF9raE,uBAAsC,erFksaxC,CqFlsaE,iCAAsC,erFssaxC,CqFtsaE,6BAAsC,erF0saxC,CqF1saE,sDAAsC,erFktaxC,CqFltaE,+CAAsC,erF0taxC,CqF1taE,sDAAsC,erFkuaxC,CqFluaE,+BAAsC,erFsuaxC,CqFtuaE,0CAAsC,erF8uaxC,CqF9uaE,+CAAsC,erFsvaxC,CqFtvaE,2BAAsC,erF0vaxC,CqF1vaE,iBAAsC,erF8vaxC,CqF9vaE,gDAAsC,erFswaxC,CqFtwaE,gBAAsC,erF0waxC,CqF1waE,oBAAsC,erF8waxC,CqF9waE,0BAAsC,erFkxaxC,CqFlxaE,8CAAsC,erF0xaxC,CqF1xaE,sBAAsC,erF8xaxC,CqF9xaE,gBAAsC,erFkyaxC,CqFlyaE,kBAAsC,erFsyaxC,CqFtyaE,iBAAsC,erF0yaxC,CqF1yaE,sBAAsC,erF8yaxC,CqF9yaE,eAAsC,erFkzaxC,CqFlzaE,oBAAsC,erFszaxC,CqFtzaE,uBAAsC,erF0zaxC,CqF1zaE,yBAAsC,erF8zaxC,CqF9zaE,yBAAsC,erFk0axC,CqFl0aE,sBAAsC,erFs0axC,CqFt0aE,sCAAsC,erF00axC,CqF10aE,sDAAsC,erFk1axC,CqFl1aE,6FAAsC,erFk2axC,CqFl2aE,4CAAsC,erF02axC,CqF12aE,iBAAsC,erF82axC,CqF92aE,0BAAsC,erFk3axC,CqFl3aE,qBAAsC,erFs3axC,CqFt3aE,eAAsC,erF03axC,CqF13aE,sCAAsC,erFk4axC,CqFl4aE,mBAAsC,erFs4axC,CqFt4aE,mDAAsC,erF84axC,CqF94aE,oBAAsC,erFk5axC,CqFl5aE,oBAAsC,erFs5axC,CqFt5aE,sDAAsC,erF85axC,CqF95aE,oBAAsC,erFk6axC,CqFl6aE,8CAAsC,erF06axC,CqF16aE,mBAAsC,erF86axC,CqF96aE,oBAAsC,erFk7axC,CqFl7aE,oDAAsC,erF87axC,CqF97aE,iBAAsC,erFk8axC,CqFl8aE,gDAAsC,erF08axC,CqF18aE,8CAAsC,erFk9axC,CqFl9aE,wDAAsC,erF09axC,CqF19aE,uBAAsC,erF89axC,CqF99aE,mBAAsC,erFk+axC,CqFl+aE,0BAAsC,erFs+axC,CqFt+aE,kBAAsC,erF0+axC,CqF1+aE,oBAAsC,erF8+axC,CqF9+aE,kBAAsC,erFk/axC,CqFl/aE,iDAAsC,erF0/axC,CqF1/aE,6BAAsC,erF8/axC,CqF9/aE,4BAAsC,erFkgbxC,CqFlgbE,0DAAsC,erF0gbxC,CqF1gbE,qBAAsC,erF8gbxC,CqF9gbE,mBAAsC,erFkhbxC,CqFlhbE,kDAAsC,erFshbxC,CqFthbE,8DAAsC,erFkibxC,CqFlibE,mBAAsC,erFsibxC,CqFtibE,kCAAsC,erF0ibxC,CqF1ibE,oBAAsC,erF8ibxC,CqF9ibE,sDAAsC,erFsjbxC,CqFtjbE,mCAAsC,erF0jbxC,CqF1jbE,uBAAsC,erF8jbxC,CqF9jbE,+CAAsC,erFskbxC,CqFtkbE,oBAAsC,erF0kbxC,CqF1kbE,6CAAsC,erFklbxC,CqFllbE,yDAAsC,erF0lbxC,CqF1lbE,wBAAsC,erF8lbxC,CqF9lbE,0DAAsC,erFsmbxC,CqFtmbE,sDAAsC,erF8mbxC,CqF9mbE,iEAAsC,erFsnbxC,CqFtnbE,oBAAsC,erF0nbxC,CqF1nbE,kBAAsC,erF8nbxC,CqF9nbE,eAAsC,erFkobxC,CqFlobE,6BAAsC,erFsobxC,CqFtobE,iBAAsC,erF0obxC,CqF1obE,0BAAsC,erF8obxC,CqF9obE,4GAAsC,erF8pbxC,CqF9pbE,gBAAsC,erFkqbxC,CqFlqbE,sBAAsC,erFsqbxC,CqFtqbE,iEAAsC,erFkrbxC,CqFlrbE,gEAAsC,erF0rbxC,CqF1rbE,qBAAsC,erF8rbxC,CqF9rbE,oDAAsC,erFssbxC,CqFtsbE,iBAAsC,erF0sbxC,CqF1sbE,gDAAsC,erFktbxC,CqFltbE,uCAAsC,erF0tbxC,CqF1tbE,uBAAsC,erF8tbxC,CqF9tbE,gBAAsC,erFkubxC,CqFlubE,kBAAsC,erFsubxC,CqFtubE,8EAAsC,erFkvbxC,CqFlvbE,mBAAsC,erFsvbxC,CqFtvbE,6BAAsC,erF0vbxC,CqF1vbE,gBAAsC,erF8vbxC,CqF9vbE,iBAAsC,erFkwbxC,CqFlwbE,iCAAsC,erFswbxC,CqFtwbE,sCAAsC,erF8wbxC,CqF9wbE,8CAAsC,erFsxbxC,CqFtxbE,yBAAsC,erF0xbxC,CqF1xbE,0DAAsC,erFkybxC,CqFlybE,+BAAsC,erFsybxC,CqFtybE,qBAAsC,erF0ybxC,CqF1ybE,sBAAsC,erF8ybxC,CqF9ybE,wFAAsC,erF0zbxC,CqF1zbE,8CAAsC,erFk0bxC,CqFl0bE,+CAAsC,erF00bxC,CqF10bE,iDAAsC,erFk1bxC,CqFl1bE,qBAAsC,erFs1bxC,CqFt1bE,2BAAsC,erF01bxC,CqF11bE,iDAAsC,erFk2bxC,CqFl2bE,sBAAsC,erFs2bxC,CqFt2bE,oFAAsC,erFk3bxC,CqFl3bE,oBAAsC,erFs3bxC,CqFt3bE,4CAAsC,erF83bxC,CqF93bE,sDAAsC,erFs4bxC,CqFt4bE,yBAAsC,erF04bxC,CqF14bE,iEAAsC,erFk5bxC,CqFl5bE,2BAAsC,erFs5bxC,CqFt5bE,wBAAsC,erF05bxC,CqF15bE,4DAAsC,erFs6bxC,CqFt6bE,6BAAsC,erF06bxC,CqF16bE,yBAAsC,erF86bxC,CqF96bE,8BAAsC,erFk7bxC,CqFl7bE,+CAAsC,erF07bxC,CqF17bE,uCAAsC,erFk8bxC,CqFl8bE,iBAAsC,erFs8bxC,CqFt8bE,iCAAsC,erF08bxC,CqF18bE,uBAAsC,erF88bxC,CqF98bE,yBAAsC,erFk9bxC,CqFl9bE,mBAAsC,erFs9bxC,CqFt9bE,8BAAsC,erF09bxC,CqF19bE,uDAAsC,erFk+bxC,CqFl+bE,+CAAsC,erF0+bxC,CqF1+bE,kBAAsC,erF8+bxC,CqF9+bE,wDAAsC,erFs/bxC,CqFt/bE,iBAAsC,erF0/bxC,CqF1/bE,qBAAsC,erF8/bxC,CqF9/bE,kBAAsC,erFkgcxC,CqFlgcE,mCAAsC,erFsgcxC,CqFtgcE,kBAAsC,erF0gcxC,CqF1gcE,2BAAsC,erF8gcxC,CqF9gcE,oBAAsC,erFkhcxC,CqFlhcE,aAAsC,WrFshcxC,CqFthcE,2CAAsC,erF8hcxC,CqF9hcE,qBAAsC,erFkicxC,CqFlicE,aAAsC,WrFsicxC,CqFticE,8DAAsC,erF8icxC,CqF9icE,kDAAsC,erFsjcxC,CqFtjcE,aAAsC,WrF0jcxC,CqF1jcE,qBAAsC,erF8jcxC,CqF9jcE,qBAAsC,erFkkcxC,CqFlkcE,sCAAsC,erF0kcxC,CqF1kcE,4DAAsC,erFklcxC,CqFllcE,+BAAsC,erFslcxC,CqFtlcE,kBAAsC,erF0lcxC,CqF1lcE,oCAAsC,erFkmcxC,CqFlmcE,+DAAsC,erF8mcxC,CqF9mcE,8BAAsC,erFkncxC,CqFlncE,yDAAsC,erF0ncxC,CqF1ncE,yBAAsC,erF8ncxC,CqF9ncE,uBAAsC,erFkocxC,CqFlocE,oBAAsC,erFsocxC,CqFtocE,2DAAsC,erF8ocxC,CqF9ocE,kBAAsC,erFkpcxC,CqFlpcE,oBAAsC,YrFspcxC,CqFtpcE,0BAAsC,erF0pcxC,CqF1pcE,oDAAsC,erFkqcxC,CqFlqcE,8BAAsC,erFsqcxC,CqFtqcE,8BAAsC,erF0qcxC,CqF1qcE,wBAAsC,erF8qcxC,CqF9qcE,uBAAsC,erFkrcxC,CqFlrcE,qDAAsC,erF0rcxC,CqF1rcE,gBAAsC,erF8rcxC,CqF9rcE,kDAAsC,erFsscxC,CqFtscE,uCAAsC,erF0scxC,CqF1scE,kCAAsC,erF8scxC,CqF9scE,+DAAsC,erFstcxC,CqFttcE,yBAAsC,erF0tcxC,CqF1tcE,kDAAsC,erFsucxC,CqFtucE,uBAAsC,erF0ucxC,CqF1ucE,wBAAsC,erF8ucxC,CqF9ucE,aAAsC,WrFkvcxC,CqFlvcE,gBAAsC,erFsvcxC,CqFtvcE,gBAAsC,erF0vcxC,CqF1vcE,+BAAsC,erFkwcxC,CqFlwcE,8BAAsC,erFswcxC,CqFtwcE,0CAAsC,erF8wcxC,CqF9wcE,0BAAsC,erFkxcxC,CqFlxcE,sBAAsC,erFsxcxC,CqFtxcE,sBAAsC,erF0xcxC,CqF1xcE,yBAAsC,erF8xcxC,CqF9xcE,qBAAsC,erFkycxC,CqFlycE,yBAAsC,erFsycxC,CqFtycE,+CAAsC,erF8ycxC,CqF9ycE,0BAAsC,erFkzcxC,CqFlzcE,gBAAsC,erFszcxC,CqFtzcE,8CAAsC,erF8zcxC,CqF9zcE,iBAAsC,erFk0cxC,CqFl0cE,wBAAsC,erFs0cxC,CqFt0cE,iEAAsC,erFk1cxC,CqFl1cE,iBAAsC,erFs1cxC,CqFt1cE,0DAAsC,erF81cxC,CqF91cE,8BAAsC,erFk2cxC,CqFl2cE,yDAAsC,erF02cxC,CqF12cE,wBAAsC,erF82cxC,CqF92cE,4CAAsC,erFs3cxC,CqFt3cE,2BAAsC,erF03cxC,CqF13cE,uBAAsC,erF83cxC,CqF93cE,2CAAsC,erFs4cxC,CqFt4cE,yBAAsC,erF04cxC,CqF14cE,kCAAsC,erF84cxC,CqF94cE,aAAsC,WrFk5cxC,CqFl5cE,gBAAsC,erFs5cxC,CqFt5cE,+CAAsC,erF85cxC,CqF95cE,yBAAsC,erFk6cxC,CqFl6cE,4BAAsC,erFs6cxC,CqFt6cE,iBAAsC,erF06cxC,CqF16cE,0BAAsC,erF86cxC,CqF96cE,2CAAsC,erFs7cxC,CqFt7cE,yBAAsC,erF07cxC,CqF17cE,uBAAsC,erF87cxC,CqF97cE,oDAAsC,erFs8cxC,CqFt8cE,6CAAsC,erF88cxC,CqF98cE,wBAAsC,erFk9cxC,CqFl9cE,gDAAsC,erF09cxC,CqF19cE,8BAAsC,erF89cxC,CqF99cE,4CAAsC,erFs+cxC,CqFt+cE,uBAAsC,erF0+cxC,CqF1+cE,aAAsC,WrF8+cxC,CqF9+cE,gHAAsC,erF8/cxC,CqF9/cE,gBAAsC,erFkgdxC,CqFlgdE,wBAAsC,erFsgdxC,CqFtgdE,sBAAsC,erF0gdxC,CqF1gdE,uBAAsC,erF8gdxC,CqF9gdE,qBAAsC,erFkhdxC,CqFlhdE,oBAAsC,erFshdxC,CqFthdE,+BAAsC,erF0hdxC,CqF1hdE,+BAAsC,erF8hdxC,CqF9hdE,gDAAsC,erFkidxC,CqFlidE,oDAAsC,erF0idxC,CqF1idE,kBAAsC,erF8idxC,CqF9idE,kDAAsC,erFsjdxC,CqFtjdE,mBAAsC,erF0jdxC,CqF1jdE,0BAAsC,erF8jdxC,CqF9jdE,uBAAsC,erFkkdxC,CqFlkdE,qBAAsC,erFskdxC,CqFtkdE,iDAAsC,erF8kdxC,CqF9kdE,0BAAsC,erFkldxC,CqFlldE,oDAAsC,erF0ldxC,CqF1ldE,sDAAsC,erFkmdxC,CqFlmdE,4CAAsC,erF0mdxC,CqF1mdE,sBAAsC,erF8mdxC,CqF9mdE,kBAAsC,erFkndxC,CqFlndE,2BAAsC,erFsndxC,CqFtndE,sBAAsC,erF0ndxC,CqF1ndE,uCAAsC,erFkodxC,CqFlodE,kBAAsC,erFsodxC,CqFtodE,gBAAsC,erF0odxC,CqF1odE,uCAAsC,erFkpdxC,CqFlpdE,sBAAsC,erFspdxC,CqFtpdE,sDAAsC,erF8pdxC,CqF9pdE,2BAAsC,erFkqdxC,CqFlqdE,2BAAsC,erFsqdxC,CqFtqdE,yBAAsC,erF0qdxC,CqF1qdE,uBAAsC,erF8qdxC,CqF9qdE,sBAAsC,erFkrdxC,CqFlrdE,kBAAsC,erFsrdxC,CqFtrdE,2CAAsC,erF8rdxC,CqF9rdE,uBAAsC,erFksdxC,CqFlsdE,eAAsC,erFssdxC,CqFtsdE,qCAAsC,erF0sdxC,CqF1sdE,uBAAsC,erF8sdxC,CqF9sdE,iBAAsC,erFktdxC,CqFltdE,8BAAsC,erFstdxC,CqFttdE,oBAAsC,erF0tdxC,CqF1tdE,2BAAsC,erF8tdxC,CqF9tdE,sBAAsC,erFkudxC,CqFludE,6BAAsC,erFsudxC,CqFtudE,gBAAsC,erF0udxC,CqF1udE,sBAAsC,erF8udxC,CqF9udE,gDAAsC,erFsvdxC,CqFtvdE,+CAAsC,erF8vdxC,CqF9vdE,6CAAsC,erFswdxC,CqFtwdE,wBAAsC,erF0wdxC,CqF1wdE,8CAAsC,erFkxdxC,CqFlxdE,uBAAsC,erFsxdxC,CqFtxdE,iDAAsC,erF8xdxC,CqF9xdE,8CAAsC,erFsydxC,CqFtydE,oBAAsC,erF0ydxC,CqF1ydE,yBAAsC,erF8ydxC,CqF9ydE,8BAAsC,erFkzdxC,CqFlzdE,sDAAsC,erF0zdxC,CqF1zdE,gBAAsC,erF8zdxC,CqF9zdE,yCAAsC,erFs0dxC,CqFt0dE,eAAsC,erF00dxC,CqF10dE,2CAAsC,erFk1dxC,CqFl1dE,6BAAsC,erFs1dxC,CqFt1dE,qCAAsC,erF01dxC,CqF11dE,uBAAsC,erF81dxC,CqF91dE,sBAAsC,erFk2dxC,CqFl2dE,eAAsC,erFs2dxC,CqFt2dE,6BAAsC,erF02dxC,CqF12dE,qBAAsC,erF82dxC,CqF92dE,gEAAsC,erF03dxC,CqF13dE,eAAsC,erF83dxC,CqF93dE,8BAAsC,erFk4dxC,CqFl4dE,4CAAsC,erF04dxC,CqF14dE,2BAAsC,erF84dxC,CqF94dE,uCAAsC,erFs5dxC,CqFt5dE,wBAAsC,erF05dxC,CqF15dE,qCAAsC,erFk6dxC,CqFl6dE,uCAAsC,erF06dxC,CqF16dE,kBAAsC,erF86dxC,CqF96dE,0DAAsC,erFs7dxC,CqFt7dE,uCAAsC,erF87dxC,CqF97dE,8CAAsC,erFs8dxC,CqFt8dE,sBAAsC,erF08dxC,CqF18dE,mBAAsC,erF88dxC,CqF98dE,mDAAsC,erFs9dxC,CqFt9dE,2DAAsC,erF89dxC,CqF99dE,qBAAsC,erFk+dxC,CqFl+dE,wBAAsC,erFs+dxC,CqFt+dE,yBAAsC,erF0+dxC,CqF1+dE,yBAAsC,erF8+dxC,CqF9+dE,kDAAsC,erFs/dxC,CqFt/dE,4GAAsC,erFsgexC,CqFtgeE,qBAAsC,erF0gexC,CqF1geE,mDAAsC,erFkhexC,CqFlheE,uBAAsC,erFshexC,CqFtheE,uBAAsC,erF0hexC,CqF1heE,+BAAsC,erF8hexC,CqF9heE,uBAAsC,erFkiexC,CqFlieE,iDAAsC,erF0iexC,CqF1ieE,iBAAsC,erF8iexC,CqF9ieE,oDAAsC,erFsjexC,CqFtjeE,0BAAsC,erF0jexC,CqF1jeE,2BAAsC,erF8jexC,CqF9jeE,yCAAsC,erFskexC,CqFtkeE,mDAAsC,erF8kexC,CqF9keE,uBAAsC,erFklexC,CqFlleE,wBAAsC,erFslexC,CqFtleE,oBAAsC,erF0lexC,CqF1leE,wBAAsC,erF8lexC,CqF9leE,wBAAsC,erFkmexC,CqFlmeE,wBAAsC,erFsmexC,CqFtmeE,sBAAsC,erF0mexC,CqF1meE,sBAAsC,erF8mexC,CqF9meE,qBAAsC,erFknexC,CqFlneE,6CAAsC,erF0nexC,CqF1neE,iBAAsC,erF8nexC,CqF9neE,kFAAsC,erF0oexC,CqF1oeE,wBAAsC,erF8oexC,CqF9oeE,eAAsC,erFkpexC,CqFlpeE,kDAAsC,erF0pexC,CqF1peE,sDAAsC,erFsqexC,CqFtqeE,iBAAsC,erF0qexC,CqF1qeE,kBAAsC,erF8qexC,CqF9qeE,uCAAsC,erFsrexC,CqFtreE,sBAAsC,erF0rexC,CqF1reE,yBAAsC,erF8rexC,CqF9reE,0BAAsC,erFksexC,CqFlseE,sBAAsC,erFssexC,CqFtseE,sBAAsC,erF0sexC,CqF1seE,+BAAsC,erF8sexC,CqF9seE,iBAAsC,erFktexC,CqFlteE,sBAAsC,erFstexC,CqFtteE,mBAAsC,erF0texC,CqF1teE,oBAAsC,erF8texC,CqF9teE,aAAsC,WrFkuexC,CqFlueE,4BAAsC,erFsuexC,CqFtueE,sBAAsC,erF0uexC,CqF1ueE,6DAAsC,erFkvexC,CqFlveE,qCAAsC,erF0vexC,CqF1veE,oBAAsC,erF8vexC,CqF9veE,uCAAsC,erFswexC,CqFtweE,aAAsC,WrF0wexC,CqF1weE,8CAAsC,erFkxexC,CqFlxeE,uBAAsC,erFsxexC,CqFtxeE,iBAAsC,erF0xexC,CqF1xeE,0CAAsC,erFkyexC,CqFlyeE,0BAAsC,erFsyexC,CqFtyeE,oDAAsC,erF8yexC,CqF9yeE,iDAAsC,erFszexC,CqFtzeE,yCAAsC,erF8zexC,CqF9zeE,wDAAsC,erFs0exC,CqFt0eE,2BAAsC,erF00exC,CqF10eE,0BAAsC,erF80exC,CqF90eE,iBAAsC,erFk1exC,CqFl1eE,qBAAsC,erFs1exC,CqFt1eE,qBAAsC,erF01exC,CqF11eE,sBAAsC,erF81exC,CqF91eE,eAAsC,erFk2exC,CqFl2eE,mBAAsC,erFs2exC,CqFt2eE,iEAAsC,erFk3exC,CqFl3eE,oBAAsC,erFs3exC,CqFt3eE,gDAAsC,erF83exC,CqF93eE,qBAAsC,erFk4exC,CqFl4eE,+BAAsC,erFs4exC,CqFt4eE,kBAAsC,erF04exC,CqF14eE,iBAAsC,erF84exC,CqF94eE,uDAAsC,erFs5exC,CqFt5eE,yBAAsC,erF05exC,CqF15eE,uBAAsC,erF85exC,CqF95eE,wBAAsC,erFk6exC,CqFl6eE,yBAAsC,erFs6exC,CqFt6eE,kBAAsC,erF06exC,CqF16eE,0BAAsC,erF86exC,CqF96eE,gDAAsC,erFs7exC,CqFt7eE,oBAAsC,erF07exC,CqF17eE,0CAAsC,erFk8exC,CqFl8eE,8CAAsC,erF08exC,CqF18eE,0BAAsC,erF88exC,CqF98eE,wCAAsC,erFs9exC,CqFt9eE,wBAAsC,erF09exC,CqF19eE,kBAAsC,erF89exC,CqF99eE,kBAAsC,erFk+exC,CqFl+eE,kBAAsC,erFs+exC,CqFt+eE,mFAAsC,erFk/exC,CqFl/eE,8DAAsC,erF0/exC,CqF1/eE,iBAAsC,erF8/exC,CqF9/eE,0BAAsC,erFkgfxC,CqFlgfE,mBAAsC,erFsgfxC,CqFtgfE,0EAAsC,erF8gfxC,CqF9gfE,oBAAsC,erFkhfxC,CqFlhfE,iBAAsC,erFshfxC,CqFthfE,sBAAsC,erF0hfxC,CqF1hfE,uBAAsC,erF8hfxC,CqF9hfE,gBAAsC,erFkifxC,CqFlifE,wBAAsC,erFsifxC,CqFtifE,wBAAsC,erF0ifxC,CqF1ifE,+EAAsC,erFsjfxC,CqFtjfE,4BAAsC,erF0jfxC,CqF1jfE,kBAAsC,erF8jfxC,CqF9jfE,wIAAsC,erFklfxC,CqFllfE,qCAAsC,erF0lfxC,CqF1lfE,+DAAsC,erFkmfxC,CqFlmfE,oBAAsC,erFsmfxC,CqFtmfE,8CAAsC,erF8mfxC,CqF9mfE,wBAAsC,erFknfxC,CqFlnfE,sCAAsC,erFsnfxC,CqFtnfE,yDAAsC,erF8nfxC,CqF9nfE,yBAAsC,erFkofxC,CqFlofE,yBAAsC,erFsofxC,CqFtofE,wDAAsC,erF8ofxC,CqF9ofE,0CAAsC,erFspfxC,CqFtpfE,uBAAsC,erF0pfxC,CqF1pfE,oEAAsC,erFsqfxC,CqFtqfE,2CAAsC,erF8qfxC,CqF9qfE,kDAAsC,erFsrfxC,CqFtrfE,oDAAsC,erF8rfxC,CqF9rfE,eAAsC,erFksfxC,CqFlsfE,uBAAsC,erFssfxC,CqFtsfE,qCAAsC,erF8sfxC,CqF9sfE,iBAAsC,erFktfxC,CqFltfE,sDAAsC,erF0tfxC,CqF1tfE,6BAAsC,erF8tfxC,CqF9tfE,uBAAsC,erFkufxC,CqFlufE,sBAAsC,erFsufxC,CqFtufE,sBAAsC,erF0ufxC,CqF1ufE,kKAAsC,erF0vfxC,CqF1vfE,+BAAsC,erFkwfxC,CqFlwfE,+CAAsC,erF0wfxC,CqF1wfE,kBAAsC,erF8wfxC,CqF9wfE,oBAAsC,erFkxfxC,CqFlxfE,yBAAsC,erFsxfxC,CqFtxfE,iCAAsC,erF0xfxC,CqF1xfE,kDAAsC,erFkyfxC,CqFlyfE,iBAAsC,erFsyfxC,CqFtyfE,4BAAsC,erF0yfxC,CqF1yfE,4BAAsC,erF8yfxC,CqF9yfE,yBAAsC,erFkzfxC,CqFlzfE,6CAAsC,erF0zfxC,CqF1zfE,aAAsC,WrF8zfxC,CqF9zfE,2DAAsC,erFs0fxC,CqFt0fE,uCAAsC,erF80fxC,CqF90fE,kDAAsC,erFs1fxC,CqFt1fE,kBAAsC,erF01fxC,CqF11fE,kBAAsC,erF81fxC,CqF91fE,6BAAsC,erFk2fxC,CqFl2fE,aAAsC,WrFs2fxC,CqFt2fE,oBAAsC,erF02fxC,CqF12fE,oBAAsC,erF82fxC,CqF92fE,sBAAsC,erFk3fxC,CqFl3fE,2DAAsC,erF03fxC,CqF13fE,8HAAsC,erF04fxC,CqF14fE,uDAAsC,erFk5fxC,CqFl5fE,oBAAsC,erFs5fxC,CqFt5fE,sBAAsC,erF05fxC,CqF15fE,6CAAsC,erFk6fxC,CqFl6fE,yBAAsC,erFs6fxC,CqFt6fE,iBAAsC,erF06fxC,CqF16fE,kBAAsC,erF86fxC,CqF96fE,wCAAsC,erFs7fxC,CqFt7fE,uBAAsC,erF07fxC,CqF17fE,yBAAsC,erF87fxC,CqF97fE,mBAAsC,erFk8fxC,CqFl8fE,iBAAsC,erFs8fxC,CqFt8fE,qBAAsC,erF08fxC,CqF18fE,iBAAsC,erF88fxC,CqF98fE,2BAAsC,erFk9fxC,CqFl9fE,yBAAsC,erFs9fxC,CqFt9fE,mDAAsC,erF89fxC,CqF99fE,kBAAsC,erFk+fxC,CqFl+fE,qBAAsC,erFs+fxC,CqFt+fE,sDAAsC,erF8+fxC,CqF9+fE,oBAAsC,erFk/fxC,CqFl/fE,uBAAsC,erFs/fxC,CqFt/fE,mCAAsC,erF8/fxC,CqF9/fE,wBAAsC,erFkggBxC,CqFlggBE,wBAAsC,erFsggBxC,CqFtggBE,4CAAsC,erF0ggBxC,CqF1ggBE,6BAAsC,erF8ggBxC,CqF9ggBE,2CAAsC,erFshgBxC,CqFthgBE,wBAAsC,erF0hgBxC,CqF1hgBE,kBAAsC,erF8hgBxC,CqF9hgBE,sBAAsC,erFkigBxC,CqFligBE,0BAAsC,erFsigBxC,CqFtigBE,aAAsC,WrF0igBxC,CqF1igBE,qCAAsC,erFkjgBxC,CqFljgBE,sBAAsC,erFsjgBxC,CqFtjgBE,iCAAsC,erF0jgBxC,CqF1jgBE,gBAAsC,erF8jgBxC,CqF9jgBE,+BAAsC,erFkkgBxC,CqFlkgBE,oBAAsC,erFskgBxC,CqFtkgBE,qCAAsC,erF8kgBxC,CqF9kgBE,4CAAsC,erFslgBxC,CqFtlgBE,0EAAsC,erFkmgBxC,CqFlmgBE,6BAAsC,erFsmgBxC,CqFtmgBE,mBAAsC,erF0mgBxC,CqF1mgBE,oEAAsC,erFsngBxC,CqFtngBE,mCAAsC,erF8ngBxC,CqF9ngBE,wBAAsC,erFkogBxC,CqFlogBE,2CAAsC,erF0ogBxC,CqF1ogBE,8CAAsC,erFkpgBxC,CqFlpgBE,qCAAsC,erF0pgBxC,CqF1pgBE,oCAAsC,erF8pgBxC,CqF9pgBE,4EAAsC,erF8qgBxC,CqF9qgBE,yBAAsC,erFkrgBxC,CqFlrgBE,0BAAsC,erFsrgBxC,CqFtrgBE,aAAsC,WrF0rgBxC,CqF1rgBE,wBAAsC,erF8rgBxC,CqF9rgBE,oBAAsC,erFksgBxC,CqFlsgBE,qBAAsC,erFssgBxC,CqFtsgBE,sDAAsC,erF8sgBxC,CqF9sgBE,sEAAsC,erF0tgBxC,CqF1tgBE,uCAAsC,erFkugBxC,CqFlugBE,yDAAsC,erF0ugBxC,CqF1ugBE,qBAAsC,erF8ugBxC,CqF9ugBE,qBAAsC,erFkvgBxC,CqFlvgBE,iBAAsC,erFsvgBxC,CqFtvgBE,sDAAsC,erF8vgBxC,CqF9vgBE,wBAAsC,erFkwgBxC,CqFlwgBE,gDAAsC,erF0wgBxC,CqF1wgBE,gBAAsC,erF8wgBxC,CqF9wgBE,sBAAsC,erFkxgBxC,CqFlxgBE,kBAAsC,erFsxgBxC,CqFtxgBE,qBAAsC,erF0xgBxC,CqF1xgBE,iBAAsC,erF8xgBxC,CqF9xgBE,2BAAsC,erFkygBxC,CqFlygBE,wBAAsC,erFsygBxC,CqFtygBE,oDAAsC,erF8ygBxC,CqF9ygBE,mBAAsC,erFkzgBxC,CqFlzgBE,sBAAsC,erFszgBxC,CqFtzgBE,+BAAsC,erF8zgBxC,CqF9zgBE,uBAAsC,erFk0gBxC,CqFl0gBE,oBAAsC,erFs0gBxC,CqFt0gBE,iBAAsC,erF00gBxC,CqF10gBE,gBAAsC,erF80gBxC,CqF90gBE,kBAAsC,erFk1gBxC,CqFl1gBE,8BAAsC,erFs1gBxC,CqFt1gBE,gFAAsC,erF02gBxC,CqF12gBE,wEAAsC,erF03gBxC,CqF13gBE,eAAsC,erF83gBxC,CqF93gBE,kBAAsC,erFk4gBxC,CqFl4gBE,iDAAsC,erF04gBxC,CqF14gBE,sBAAsC,erF84gBxC,CqF94gBE,qBAAsC,erFk5gBxC,CqFl5gBE,oBAAsC,erFs5gBxC,CqFt5gBE,wDAAsC,erF85gBxC,CqF95gBE,+BAAsC,erFk6gBxC,CqFl6gBE,uBAAsC,erFs6gBxC,CqFt6gBE,6FAAsC,erFk7gBxC,CqFl7gBE,qBAAsC,YrFs7gBxC,CqFt7gBE,sBAAsC,erF07gBxC,CqF17gBE,sBAAsC,erF87gBxC,CqF97gBE,2BAAsC,erFk8gBxC,CqFl8gBE,sBAAsC,erFs8gBxC,CqFt8gBE,uBAAsC,erF08gBxC,CqF18gBE,qBAAsC,erF88gBxC,CqF98gBE,wDAAsC,erFs9gBxC,CqFt9gBE,+BAAsC,erF09gBxC,CqF19gBE,0CAAsC,erFk+gBxC,CqFl+gBE,sBAAsC,erFs+gBxC,CqFt+gBE,4CAAsC,erF8+gBxC,CqF9+gBE,iCAAsC,erFk/gBxC,CqFl/gBE,mBAAsC,erFs/gBxC,CqFt/gBE,gCAAsC,erF0/gBxC,CqF1/gBE,eAAsC,erF8/gBxC,CqF9/gBE,kDAAsC,erFsghBxC,CqFtghBE,qCAAsC,erF8ghBxC,CqF9ghBE,6BAAsC,erFkhhBxC,CqFlhhBE,mDAAsC,erF0hhBxC,CqF1hhBE,wBAAsC,erF8hhBxC,CqF9hhBE,qCAAsC,erFkihBxC,CqFlihBE,oBAAsC,erFsihBxC,CqFtihBE,yBAAsC,erF0ihBxC,CqF1ihBE,6BAAsC,erF8ihBxC,CqF9ihBE,0DAAsC,erFsjhBxC,CqFtjhBE,iBAAsC,erF0jhBxC,CqF1jhBE,eAAsC,erF8jhBxC,CqF9jhBE,wCAAsC,erFskhBxC,CqFtkhBE,mBAAsC,erF0khBxC,CqF1khBE,0GAAsC,erF0lhBxC,CqF1lhBE,gBAAsC,erF8lhBxC,CqF9lhBE,uBAAsC,erFkmhBxC,CqFlmhBE,6BAAsC,erFsmhBxC,CqFtmhBE,wBAAsC,erF0mhBxC,CqF1mhBE,uBAAsC,erF8mhBxC,CqF9mhBE,wBAAsC,erFknhBxC,CqFlnhBE,oDAAsC,erF0nhBxC,CqF1nhBE,kDAAsC,erFkohBxC,CqFlohBE,gBAAsC,erFsohBxC,CqFtohBE,wBAAsC,YrF0ohBxC,CqF1ohBE,8CAAsC,erFkphBxC,CqFlphBE,sBAAsC,erFsphBxC,CqFtphBE,mCAAsC,erF8phBxC,CqF9phBE,kBAAsC,erFkqhBxC,CqFlqhBE,8FAAsC,erFkrhBxC,CqFlrhBE,wBAAsC,erFsrhBxC,CqFtrhBE,gBAAsC,erF0rhBxC,CqF1rhBE,8CAAsC,erFkshBxC,CqFlshBE,iBAAsC,erFsshBxC,CqFtshBE,iBAAsC,erF0shBxC,CqF1shBE,mBAAsC,erF8shBxC,CqF9shBE,+EAAsC,erF0thBxC,CqF1thBE,8BAAsC,erF8thBxC,CqF9thBE,qEAAsC,erF0uhBxC,CqF1uhBE,uBAAsC,erF8uhBxC,CqF9uhBE,sBAAsC,erFkvhBxC,CqFlvhBE,+DAAsC,erF8vhBxC,CqF9vhBE,wBAAsC,erFkwhBxC,CqFlwhBE,kBAAsC,erFswhBxC,CqFtwhBE,qDAAsC,erF8whBxC,CqF9whBE,gBAAsC,erFkxhBxC,CqFlxhBE,qBAAsC,erFsxhBxC,CqFtxhBE,6BAAsC,erF0xhBxC,CqF1xhBE,eAAsC,erF8xhBxC,CqF9xhBE,qBAAsC,erFkyhBxC,CqFlyhBE,6CAAsC,erF0yhBxC,CqF1yhBE,qBAAsC,erF8yhBxC,CqF9yhBE,+DAAsC,erF0zhBxC,CqF1zhBE,iCAAsC,erFk0hBxC,CqFl0hBE,6DAAsC,erF00hBxC,CqF10hBE,qBAAsC,erF80hBxC,CqF90hBE,gBAAsC,erFk1hBxC,CqFl1hBE,gBAAsC,erFs1hBxC,CqFt1hBE,gCAAsC,erF01hBxC,CqF11hBE,uBAAsC,erF81hBxC,CqF91hBE,8CAAsC,erFs2hBxC,CqFt2hBE,6FAAsC,erFk3hBxC,CqFl3hBE,sDAAsC,erF03hBxC,CqF13hBE,2DAAsC,erFk4hBxC,CqFl4hBE,sBAAsC,erFs4hBxC,CqFt4hBE,2BAAsC,erF04hBxC,CqF14hBE,yBAAsC,erF84hBxC,CqF94hBE,yCAAsC,erFs5hBxC,CqFt5hBE,mBAAsC,erF05hBxC,CqF15hBE,qEAAsC,erFs6hBxC,CqFt6hBE,4BAAsC,erF06hBxC,CqF16hBE,qCAAsC,erF86hBxC,CqF96hBE,wBAAsC,erFk7hBxC,CqFl7hBE,gEAAsC,erF87hBxC,CqF97hBE,4BAAsC,erFk8hBxC,CqFl8hBE,wBAAsC,erFs8hBxC,CqFt8hBE,wCAAsC,erF88hBxC,CqF98hBE,uBAAsC,erFk9hBxC,CqFl9hBE,sBAAsC,erFs9hBxC,CqFt9hBE,oDAAsC,erF89hBxC,CqF99hBE,wBAAsC,erFk+hBxC,CqFl+hBE,sBAAsC,erFs+hBxC,CqFt+hBE,4BAAsC,erF0+hBxC,CqF1+hBE,+BAAsC,erF8+hBxC,CqF9+hBE,iDAAsC,erFs/hBxC,CqFt/hBE,oBAAsC,erF0/hBxC,CqF1/hBE,sDAAsC,erFsgiBxC,CqFtgiBE,kBAAsC,erF0giBxC,CqF1giBE,uCAAsC,erFkhiBxC,CqFlhiBE,yBAAsC,erFshiBxC,CqFthiBE,8CAAsC,erF8hiBxC,CqF9hiBE,uBAAsC,erFkiiBxC,CqFliiBE,gBAAsC,erFsiiBxC,CqFtiiBE,0CAAsC,erF8iiBxC,CqF9iiBE,aAAsC,WrFkjiBxC,CqFljiBE,uDAAsC,erF0jiBxC,CqF1jiBE,+CAAsC,erFkkiBxC,CqFlkiBE,gBAAsC,erFskiBxC,CqFtkiBE,yBAAsC,erF0kiBxC,CqF1kiBE,gDAAsC,erFkliBxC,CqFlliBE,wBAAsC,erFsliBxC,CqFtliBE,qDAAsC,erF8liBxC,CqF9liBE,6CAAsC,erFsmiBxC,CqFtmiBE,0BAAsC,erF0miBxC,CqF1miBE,sDAAsC,erFkniBxC,CqFlniBE,gBAAsC,erFsniBxC,CqFtniBE,uEAAsC,erFkoiBxC,CqFloiBE,yBAAsC,erFsoiBxC,CqFtoiBE,2CAAsC,erF8oiBxC,CqF9oiBE,qDAAsC,erFspiBxC,CqFtpiBE,kDAAsC,erF8piBxC,CqF9piBE,eAAsC,erFkqiBxC,CqFlqiBE,gDAAsC,erF0qiBxC,CqF1qiBE,+BAAsC,YrFkriBxC,CqFlriBE,kBAAsC,erFsriBxC,CqFtriBE,oBAAsC,erF0riBxC,CqF1riBE,yFAAsC,erF8siBxC,CqF9siBE,uDAAsC,erFstiBxC,CqFttiBE,yDAAsC,erF8tiBxC,CqF9tiBE,qBAAsC,erFkuiBxC,CqFluiBE,2BAAsC,erFsuiBxC,CqFtuiBE,gBAAsC,erF0uiBxC,CqF1uiBE,sBAAsC,erF8uiBxC,CqF9uiBE,gDAAsC,erFsviBxC,CqFtviBE,mBAAsC,erF0viBxC,CqF1viBE,mDAAsC,erFkwiBxC,CqFlwiBE,8BAAsC,erFswiBxC,CqFtwiBE,yBAAsC,erF0wiBxC,CqF1wiBE,0BAAsC,erF8wiBxC,CqF9wiBE,qBAAsC,erFkxiBxC,CqFlxiBE,kBAAsC,YrFsxiBxC,CqFtxiBE,mBAAsC,erF0xiBxC,CqF1xiBE,iBAAsC,erF8xiBxC,CqF9xiBE,iGAAsC,erFkziBxC,CqFlziBE,eAAsC,erFsziBxC,CqFtziBE,kBAAsC,erF0ziBxC,CqF1ziBE,6CAAsC,erFk0iBxC,CqFl0iBE,wBAAsC,erFs0iBxC,CqFt0iBE,iBAAsC,erF00iBxC,CqF10iBE,0BAAsC,erF80iBxC,CqF90iBE,6BAAsC,erFk1iBxC,CqFl1iBE,wCAAsC,erF01iBxC,CqF11iBE,wBAAsC,erF81iBxC,CqF91iBE,sBAAsC,erFk2iBxC,CqFl2iBE,4CAAsC,erF02iBxC,CqF12iBE,uBAAsC,erF82iBxC,CqF92iBE,gBAAsC,erFk3iBxC,CqFl3iBE,oBAAsC,erFs3iBxC,CqFt3iBE,0BAAsC,erF03iBxC,CqF13iBE,uBAAsC,erF83iBxC,CqF93iBE,wBAAsC,erFk4iBxC,CqFl4iBE,oBAAsC,erFs4iBxC,CqFt4iBE,uBAAsC,erF04iBxC,CqF14iBE,kBAAsC,erF84iBxC,CqF94iBE,mBAAsC,erFk5iBxC,CqFl5iBE,6BAAsC,erFs5iBxC,CqFt5iBE,sBAAsC,erF05iBxC,CqF15iBE,yBAAsC,erF85iBxC,CqF95iBE,mBAAsC,erFk6iBxC,CqFl6iBE,uCAAsC,erF06iBxC,CqF16iBE,qBAAsC,erF86iBxC,CqF96iBE,uBAAsC,erFk7iBxC,CqFl7iBE,gDAAsC,erF07iBxC,CqF17iBE,iBAAsC,erF87iBxC,CqF97iBE,kDAAsC,erFs8iBxC,CqFt8iBE,yBAAsC,erF08iBxC,CqF18iBE,uBAAsC,erF88iBxC,CqF98iBE,wBAAsC,erFk9iBxC,CqFl9iBE,8BAAsC,erFs9iBxC,CqFt9iBE,yCAAsC,YrF89iBxC,CqF99iBE,wBAAsC,erFk+iBxC,CqFl+iBE,+BAAsC,erFs+iBxC,CqFt+iBE,mBAAsC,erF0+iBxC,CqF1+iBE,uCAAsC,erFk/iBxC,CqFl/iBE,2CAAsC,erF0/iBxC,CqF1/iBE,kBAAsC,erF8/iBxC,CqF9/iBE,0CAAsC,erFsgjBxC,CqFtgjBE,kBAAsC,erF0gjBxC,CqF1gjBE,sBAAsC,erF8gjBxC,CqF9gjBE,sCAAsC,erFshjBxC,CqFthjBE,mBAAsC,erF0hjBxC,CqF1hjBE,iBAAsC,erF8hjBxC,CqF9hjBE,iBAAsC,erFkijBxC,CqFlijBE,iCAAsC,erF0ijBxC,CqF1ijBE,qBAAsC,erF8ijBxC,CqF9ijBE,+BAAsC,erFkjjBxC,CqFljjBE,mBAAsC,erFsjjBxC,CqFtjjBE,mDAAsC,erF8jjBxC,CqF9jjBE,qBAAsC,erFkkjBxC,CqFlkjBE,oBAAsC,erFskjBxC,CqFtkjBE,kDAAsC,erF8kjBxC,CqF9kjBE,mBAAsC,erFkljBxC,CqFlljBE,iDAAsC,erF0ljBxC,CqF1ljBE,gCAAsC,erF8ljBxC,CqF9ljBE,iBAAsC,erFkmjBxC,CqFlmjBE,+CAAsC,erF0mjBxC,CqF1mjBE,oBAAsC,erF8mjBxC,CqF9mjBE,oDAAsC,erFsnjBxC,CqFtnjBE,kBAAsC,erF0njBxC,CqF1njBE,gDAAsC,erFkojBxC,CqFlojBE,2DAAsC,erF0ojBxC,CqF1ojBE,wDAAsC,erFkpjBxC,CqFlpjBE,0CAAsC,erF0pjBxC,CqF1pjBE,6DAAsC,erFkqjBxC,CqFlqjBE,yBAAsC,erFsqjBxC,CqFtqjBE,8CAAsC,erF8qjBxC,CqF9qjBE,8DAAsC,erFsrjBxC,CqFtrjBE,gCAAsC,erF0rjBxC,CqF1rjBE,mBAAsC,erF8rjBxC,CqF9rjBE,gCAAsC,erFksjBxC,CqFlsjBE,iBAAsC,erFssjBxC,CqFtsjBE,uCAAsC,erF8sjBxC,CqF9sjBE,eAAsC,erFktjBxC,CqFltjBE,kBAAsC,erFstjBxC,CqFttjBE,gBAAsC,erF0tjBxC,CqF1tjBE,sDAAsC,erFkujBxC,CqFlujBE,kBAAsC,erFsujBxC,CqFtujBE,oBAAsC,erF0ujBxC,CqF1ujBE,iBAAsC,erF8ujBxC,CqF9ujBE,uBAAsC,erFkvjBxC,CqFlvjBE,2BAAsC,erFsvjBxC,CqFtvjBE,2CAAsC,erF8vjBxC,CqF9vjBE,iBAAsC,erFkwjBxC,CqFlwjBE,wBAAsC,erFswjBxC,CqFtwjBE,mBAAsC,erF0wjBxC,CqF1wjBE,kDAAsC,erFkxjBxC,CqFlxjBE,wEAAsC,erF8xjBxC,CqF9xjBE,mBAAsC,YrFkyjBxC,CqFlyjBE,oEAAsC,erF0yjBxC,CqF1yjBE,mBAAsC,erF8yjBxC,CqF9yjBE,aAAsC,WrFkzjBxC,CqFlzjBE,iBAAsC,erFszjBxC,CqFtzjBE,wBAAsC,erF0zjBxC,CqF1zjBE,oBAAsC,erF8zjBxC,CqF9zjBE,6BAAsC,erFk0jBxC,CqFl0jBE,qCAAsC,erFs0jBxC,CqFt0jBE,qBAAsC,erF00jBxC,CqF10jBE,eAAsC,erF80jBxC,CqF90jBE,kCAAsC,erFk1jBxC,CqFl1jBE,2CAAsC,erF01jBxC,CqF11jBE,2BAAsC,erF81jBxC,CqF91jBE,oBAAsC,erFk2jBxC,CqFl2jBE,mBAAsC,erFs2jBxC,CqFt2jBE,mCAAsC,erF82jBxC,CqF92jBE,mBAAsC,erFk3jBxC,CqFl3jBE,kBAAsC,erFs3jBxC,CqFt3jBE,uBAAsC,erF03jBxC,CqF13jBE,8CAAsC,erFk4jBxC,CqFl4jBE,6DAAsC,erF04jBxC,CqF14jBE,wBAAsC,erF84jBxC,CqF94jBE,mCAAsC,erFk5jBxC,CqFl5jBE,uFAAsC,erFk6jBxC,CqFl6jBE,iBAAsC,erFs6jBxC,CqFt6jBE,0CAAsC,erF06jBxC,CqF16jBE,qDAAsC,erFk7jBxC,CqFl7jBE,iDAAsC,erF07jBxC,CqF17jBE,mBAAsC,erF87jBxC,CqF97jBE,4BAAsC,erFk8jBxC,CqFl8jBE,uCAAsC,erF08jBxC,CqF18jBE,kBAAsC,erF88jBxC,CqF98jBE,sBAAsC,erFk9jBxC,CqFl9jBE,uCAAsC,erF09jBxC,CqF19jBE,uBAAsC,erF89jBxC,CqF99jBE,4BAAsC,erFk+jBxC,CqFl+jBE,+BAAsC,erF0+jBxC,CqF1+jBE,wBAAsC,erF8+jBxC,CqF9+jBE,mDAAsC,erFs/jBxC,CqFt/jBE,8EAAsC,erFkgkBxC,CqFlgkBE,kBAAsC,erFsgkBxC,CqFtgkBE,0CAAsC,erF8gkBxC,CqF9gkBE,aAAsC,WrFkhkBxC,CqFlhkBE,yCAAsC,erF0hkBxC,CqF1hkBE,sCAAsC,erFkikBxC,CqFlikBE,gBAAsC,erFsikBxC,CqFtikBE,qBAAsC,erF0ikBxC,CqF1ikBE,iBAAsC,erF8ikBxC,CqF9ikBE,uDAAsC,erFsjkBxC,CqFtjkBE,8BAAsC,erF0jkBxC,CqF1jkBE,sBAAsC,erF8jkBxC,CqF9jkBE,6BAAsC,erFkkkBxC,CqFlkkBE,6BAAsC,erFskkBxC,CqFtkkBE,oCAAsC,erF0kkBxC,CqF1kkBE,4CAAsC,erFklkBxC,CqFllkBE,sBAAsC,erFslkBxC,CqFtlkBE,qBAAsC,erF0lkBxC,CqF1lkBE,0BAAsC,erF8lkBxC,CqF9lkBE,8CAAsC,erFsmkBxC,CqFtmkBE,oBAAsC,erF0mkBxC,CqF1mkBE,eAAsC,erF8mkBxC,CqF9mkBE,gDAAsC,erFsnkBxC,CqFtnkBE,yDAAsC,erF8nkBxC,CqF9nkBE,yGAAsC,erF8okBxC,CqF9okBE,qBAAsC,erFkpkBxC,CqFlpkBE,qBAAsC,erFspkBxC,CqFtpkBE,yDAAsC,erF8pkBxC,CqF9pkBE,qBAAsC,erFkqkBxC,CqFlqkBE,mFAAsC,erF8qkBxC,CqF9qkBE,oBAAsC,erFkrkBxC,CqFlrkBE,yCAAsC,erF0rkBxC,CqF1rkBE,0BAAsC,erF8rkBxC,CqF9rkBE,qBAAsC,erFkskBxC,CqFlskBE,0BAAsC,erFsskBxC,CqFtskBE,8CAAsC,erF8skBxC,CqF9skBE,uBAAsC,erFktkBxC,CqFltkBE,mCAAsC,erFstkBxC,CqFttkBE,mBAAsC,erF0tkBxC,CqF1tkBE,uBAAsC,erF8tkBxC,CqF9tkBE,4DAAsC,erFsukBxC,CqFtukBE,qBAAsC,erF0ukBxC,CqF1ukBE,6DAAsC,erFsvkBxC,CqFtvkBE,wBAAsC,erF0vkBxC,CqF1vkBE,yEAAsC,erFswkBxC,CqFtwkBE,8CAAsC,erF8wkBxC,CqF9wkBE,2CAAsC,erFsxkBxC,CqFtxkBE,qDAAsC,erF8xkBxC,CqF9xkBE,+CAAsC,erFsykBxC,CqFtykBE,sCAAsC,erF0ykBxC,CqF1ykBE,sDAAsC,erFkzkBxC,CqFlzkBE,gBAAsC,erFszkBxC,CqFtzkBE,wBAAsC,erF0zkBxC,CqF1zkBE,iBAAsC,erF8zkBxC,CqF9zkBE,uCAAsC,erFs0kBxC,CqFt0kBE,0BAAsC,erF00kBxC,CqF10kBE,wEAAsC,erFs1kBxC,CqFt1kBE,iBAAsC,erF01kBxC,CqF11kBE,iBAAsC,erF81kBxC,CqF91kBE,4CAAsC,erFs2kBxC,CqFt2kBE,+CAAsC,erF82kBxC,CqF92kBE,wDAAsC,erFs3kBxC,CqFt3kBE,qBAAsC,erF03kBxC,CqF13kBE,sBAAsC,erF83kBxC,CqF93kBE,0CAAsC,erFs4kBxC,CqFt4kBE,wBAAsC,erF04kBxC,CqF14kBE,2BAAsC,erF84kBxC,CqF94kBE,wCAAsC,erFs5kBxC,CqFt5kBE,iBAAsC,erF05kBxC,CqF15kBE,gBAAsC,erF85kBxC,CqF95kBE,0DAAsC,erFs6kBxC,CqFt6kBE,0BAAsC,erF06kBxC,CqF16kBE,0CAAsC,erFk7kBxC,CqFl7kBE,sBAAsC,erFs7kBxC,CqFt7kBE,sBAAsC,erF07kBxC,CqF17kBE,mBAAsC,erF87kBxC,CqF97kBE,qBAAsC,erFk8kBxC,CqFl8kBE,wBAAsC,erFs8kBxC,CqFt8kBE,uBAAsC,erF08kBxC,CqF18kBE,kBAAsC,erF88kBxC,CqF98kBE,eAAsC,erFk9kBxC,CqFl9kBE,8CAAsC,erF09kBxC,CqF19kBE,qBAAsC,erF89kBxC,CqF99kBE,oCAAsC,erFk+kBxC,CqFl+kBE,gBAAsC,erFs+kBxC,CqFt+kBE,wBAAsC,erF0+kBxC,CqF1+kBE,6CAAsC,erFk/kBxC,CqFl/kBE,iBAAsC,erFs/kBxC,CqFt/kBE,4BAAsC,erF0/kBxC,CqF1/kBE,uBAAsC,YrF8/kBxC,CqF9/kBE,uBAAsC,erFkglBxC,CqFlglBE,iBAAsC,erFsglBxC,CqFtglBE,oEAAsC,erFkhlBxC,CqFlhlBE,wDAAsC,YrF8hlBxC,CqF9hlBE,aAAsC,WrFkilBxC,CqFlilBE,4DAAsC,erF0ilBxC,CqF1ilBE,2CAAsC,erFkjlBxC,CqFljlBE,oCAAsC,erFsjlBxC,CqFtjlBE,2GAAsC,erF0klBxC,CqF1klBE,oBAAsC,erF8klBxC,CqF9klBE,kBAAsC,erFkllBxC,CqFlllBE,aAAsC,WrFsllBxC,CqFtllBE,mBAAsC,erF0llBxC,CqF1llBE,kDAAsC,erFkmlBxC,CqFlmlBE,oBAAsC,erFsmlBxC,CqFtmlBE,2BAAsC,erF0mlBxC,CqF1mlBE,qDAAsC,erFknlBxC,CqFlnlBE,8CAAsC,erF0nlBxC,CqF1nlBE,0DAAsC,erFkolBxC,CqFlolBE,mBAAsC,erFsolBxC,CqFtolBE,sBAAsC,erF0olBxC,CqF1olBE,yCAAsC,erFkplBxC,CqFlplBE,kDAAsC,erF0plBxC,CqF1plBE,gCAAsC,erF8plBxC,CqF9plBE,+BAAsC,erFkqlBxC,CqFlqlBE,2CAAsC,erF0qlBxC,C6E3olBE,sFAPA,kBAAsB,CAEtB,cAAe,CANf,UAAW,CAEX,WAAY,CACZ,eAAgB,CAFhB,SAAU,CAHV,iBAAkB,CAOlB,kBAAmB,CANnB,S7EirlBF;;AsF3slBA;;;;EAAA,CAQA,YAEE,sDtF2slBF,CsFxslBA,WAIE,kBF8BwB,CEjCxB,iCACA,iBAAkB,CAClB,eAAgB,CAEhB,kJtF2slBF,CsFvslBA,eAEE,etFyslBF;;AuFjulBA;;;;EAAA,CAQA,YACE,gDACA,wDvFiulBF,CuF9tlBA,WAIE,kBH8BwB,CGjCxB,iCACA,iBAAkB,CAClB,eAAgB,CAEhB,sJvFiulBF,CuF7tlBA,iBAEE,evF+tlBF;;AwFvvlBA;;;;EAAA,CAQA,YACE,iDACA,yDxFuvlBF,CwFpvlBA,WAIE,kBJ8BwB,CIjCxB,mCACA,iBAAkB,CAClB,eAAgB,CAEhB,oJxFuvlBF,CwFnvlBA,gBAEE,exFqvlBF,CwFjvlBE,kBAAqC,exFqvlBvC,CwFrvlBE,iBAAqC,exFyvlBvC,CwFzvlBE,gBAAqC,exF6vlBvC,CwF7vlBE,mBAAqC,exFiwlBvC,CwFjwlBE,kBAAqC,exFqwlBvC,CwFrwlBE,oBAAqC,exFywlBvC,CwFzwlBE,+BAAqC,exF6wlBvC,CwF7wlBE,eAAqC,exFixlBvC,CwFjxlBE,kBAAqC,exFqxlBvC,CwFrxlBE,iBAAqC,exFyxlBvC,CwFzxlBE,sBAAqC,exF6xlBvC,CwF7xlBE,eAAqC,exFiylBvC,CwFjylBE,iBAAqC,exFqylBvC,CwFrylBE,sBAAqC,exFyylBvC,CwFzylBE,kBAAqC,exF6ylBvC,CwF7ylBE,gBAAqC,exFizlBvC,CwFjzlBE,qBAAqC,exFqzlBvC,CwFrzlBE,sBAAqC,exFyzlBvC,CwFzzlBE,uBAAqC,exF6zlBvC,CwF7zlBE,gBAAqC,exFi0lBvC,CwFj0lBE,mBAAqC,exFq0lBvC,CwFr0lBE,eAAqC,exFy0lBvC,CwFz0lBE,0CAAqC,exFi1lBvC,CwFj1lBE,qBAAqC,exFq1lBvC,CwFr1lBE,cAAqC,exFy1lBvC,CwFz1lBE,iBAAqC,exF61lBvC,CwF71lBE,gBAAqC,exFi2lBvC,CwFj2lBE,oBAAqC,exFq2lBvC,CwFr2lBE,mCAAqC,exFy2lBvC,CwFz2lBE,sBAAqC,exF62lBvC,CwF72lBE,mBAAqC,exFi3lBvC,CwFj3lBE,kBAAqC,exFq3lBvC,CwFr3lBE,eAAqC,exFy3lBvC,CwFz3lBE,yBAAqC,exF63lBvC,CwF73lBE,+BAAqC,exFi4lBvC,CwFj4lBE,iBAAqC,exFq4lBvC,CwFr4lBE,iBAAqC,exFy4lBvC,CwFz4lBE,sBAAqC,exF64lBvC,CwF74lBE,cAAqC,exFi5lBvC,CwFj5lBE,mBAAqC,exFq5lBvC,CwFr5lBE,qBAAqC,exFy5lBvC,CwFz5lBE,oBAAqC,exF65lBvC,CwF75lBE,kDAAqC,exFq6lBvC,CwFr6lBE,mBAAqC,exFy6lBvC,CwFz6lBE,kBAAqC,exF66lBvC,CwF76lBE,+BAAqC,exFi7lBvC,CwFj7lBE,mBAAqC,exFq7lBvC,CwFr7lBE,iBAAqC,exFy7lBvC,CwFz7lBE,mBAAqC,exF67lBvC,CwF77lBE,wDAAqC,exFq8lBvC,CwFr8lBE,sBAAqC,exFy8lBvC,CwFz8lBE,wBAAqC,exF68lBvC,CwF78lBE,4DAAqC,exFq9lBvC,CwFr9lBE,gBAAqC,exFy9lBvC,CwFz9lBE,mBAAqC,exF69lBvC,CwF79lBE,mBAAqC,exFi+lBvC,CwFj+lBE,sDAAqC,exFy+lBvC,CwFz+lBE,yBAAqC,exF6+lBvC,CwF7+lBE,sBAAqC,exFi/lBvC,CwFj/lBE,oBAAqC,exFq/lBvC,CwFr/lBE,sBAAqC,exFy/lBvC,CwFz/lBE,uBAAqC,exF6/lBvC,CwF7/lBE,oBAAqC,exFigmBvC,CwFjgmBE,sBAAqC,exFqgmBvC,CwFrgmBE,0BAAqC,exFygmBvC,CwFzgmBE,sBAAqC,exF6gmBvC,CwF7gmBE,4BAAqC,exFihmBvC,CwFjhmBE,qBAAqC,exFqhmBvC,CwFrhmBE,mBAAqC,exFyhmBvC,CwFzhmBE,qBAAqC,exF6hmBvC,CwF7hmBE,+BAAqC,exFiimBvC,CwFjimBE,kBAAqC,exFqimBvC,CwFrimBE,kBAAqC,exFyimBvC,CwFzimBE,kEAAqC,exFijmBvC,CwFjjmBE,qBAAqC,exFqjmBvC,CwFrjmBE,uBAAqC,exFyjmBvC,CwFzjmBE,yBAAqC,exF6jmBvC,CwF7jmBE,kBAAqC,exFikmBvC,CwFjkmBE,sBAAqC,exFqkmBvC,CwFrkmBE,oBAAqC,exFykmBvC,CwFzkmBE,uBAAqC,exF6kmBvC,CwF7kmBE,qCAAqC,exFilmBvC,CwFjlmBE,oBAAqC,exFqlmBvC,CwFrlmBE,4BAAqC,exFylmBvC,CwFzlmBE,+BAAqC,exF6lmBvC,CwF7lmBE,qBAAqC,exFimmBvC,CwFjmmBE,kBAAqC,exFqmmBvC,CwFrmmBE,wBAAqC,exFymmBvC,CwFzmmBE,mBAAqC,exF6mmBvC,CwF7mmBE,mBAAqC,exFinmBvC,CwFjnmBE,gBAAqC,exFqnmBvC,CwFrnmBE,eAAqC,exFynmBvC,CwFznmBE,mBAAqC,exF6nmBvC,CwF7nmBE,gCAAqC,exFiomBvC,CwFjomBE,kDAAqC,exFyomBvC,CwFzomBE,kBAAqC,exF6omBvC,CwF7omBE,kBAAqC,exFipmBvC,CwFjpmBE,eAAqC,exFqpmBvC,CwFrpmBE,sBAAqC,exFypmBvC,CwFzpmBE,oBAAqC,exF6pmBvC,CwF7pmBE,0BAAqC,exFiqmBvC,CwFjqmBE,sDAAqC,exFyqmBvC,CwFzqmBE,oBAAqC,exF6qmBvC,CwF7qmBE,gBAAqC,exFirmBvC,CwFjrmBE,eAAqC,exFqrmBvC,CwFrrmBE,iBAAqC,exFyrmBvC,CwFzrmBE,wBAAqC,exF6rmBvC,CwF7rmBE,kBAAqC,exFismBvC,CwFjsmBE,kBAAqC,exFqsmBvC,CwFrsmBE,kBAAqC,exFysmBvC,CwFzsmBE,oBAAqC,exF6smBvC,CwF7smBE,wBAAqC,exFitmBvC,CwFjtmBE,oBAAqC,exFqtmBvC,CwFrtmBE,oDAAqC,exF6tmBvC,CwF7tmBE,qBAAqC,exFiumBvC,CwFjumBE,qBAAqC,exFqumBvC,CwFrumBE,mBAAqC,exFyumBvC,CwFzumBE,iBAAqC,exF6umBvC,CwF7umBE,oBAAqC,exFivmBvC,CwFjvmBE,mBAAqC,exFqvmBvC,CwFrvmBE,wBAAqC,exFyvmBvC,CwFzvmBE,qBAAqC,exF6vmBvC,CwF7vmBE,yBAAqC,exFiwmBvC,CwFjwmBE,uBAAqC,exFqwmBvC,CwFrwmBE,kBAAqC,exFywmBvC,CwFzwmBE,qDAAqC,exFixmBvC,CwFjxmBE,gBAAqC,exFqxmBvC,CwFrxmBE,iBAAqC,exFyxmBvC,CwFzxmBE,qBAAqC,exF6xmBvC,CwF7xmBE,sBAAqC,exFiymBvC,CwFjymBE,oDAAqC,exFyymBvC,CwFzymBE,uBAAqC,exF6ymBvC,CwF7ymBE,sCAAqC,exFqzmBvC,CwFrzmBE,qBAAqC,exFyzmBvC,CwFzzmBE,6BAAqC,exF6zmBvC,CwF7zmBE,2BAAqC,exFi0mBvC,CwFj0mBE,iBAAqC,exFq0mBvC,CwFr0mBE,iBAAqC,exFy0mBvC,CwFz0mBE,kBAAqC,exF60mBvC,CwF70mBE,oBAAqC,exFi1mBvC,CwFj1mBE,gBAAqC,exFq1mBvC,CwFr1mBE,mBAAqC,exFy1mBvC,CwFz1mBE,kBAAqC,exF61mBvC,CwF71mBE,uBAAqC,exFi2mBvC,CwFj2mBE,6BAAqC,exFq2mBvC,CwFr2mBE,wBAAqC,exFy2mBvC,CwFz2mBE,iCAAqC,exF62mBvC,CwF72mBE,gBAAqC,exFi3mBvC,CwFj3mBE,mBAAqC,exFq3mBvC,CwFr3mBE,kBAAqC,exFy3mBvC,CwFz3mBE,mBAAqC,exF63mBvC,CwF73mBE,kBAAqC,exFi4mBvC,CwFj4mBE,yBAAqC,exFq4mBvC,CwFr4mBE,uBAAqC,exFy4mBvC,CwFz4mBE,sBAAqC,exF64mBvC,CwF74mBE,sBAAqC,exFi5mBvC,CwFj5mBE,oBAAqC,exFq5mBvC,CwFr5mBE,eAAqC,exFy5mBvC,CwFz5mBE,kBAAqC,exF65mBvC,CwF75mBE,2BAAqC,exFi6mBvC,CwFj6mBE,gBAAqC,exFq6mBvC,CwFr6mBE,kBAAqC,exFy6mBvC,CwFz6mBE,oBAAqC,exF66mBvC,CwF76mBE,gBAAqC,exFi7mBvC,CwFj7mBE,gDAAqC,exFy7mBvC,CwFz7mBE,iBAAqC,exF67mBvC,CwF77mBE,gDAAqC,exFq8mBvC,CwFr8mBE,sBAAqC,exFy8mBvC,CwFz8mBE,uFAAqC,exFq9mBvC,CwFr9mBE,oBAAqC,exFy9mBvC,CwFz9mBE,iBAAqC,exF69mBvC,CwF79mBE,gBAAqC,exFi+mBvC,CwFj+mBE,qBAAqC,exFq+mBvC,CwFr+mBE,mBAAqC,exFy+mBvC,CwFz+mBE,qBAAqC,exF6+mBvC,CwF7+mBE,kBAAqC,exFi/mBvC,CwFj/mBE,sBAAqC,exFq/mBvC,CwFr/mBE,kDAAqC,exF6/mBvC,CwF7/mBE,uBAAqC,exFignBvC,CwFjgnBE,iBAAqC,exFqgnBvC,CwFrgnBE,6BAAqC,exFygnBvC,CwFzgnBE,mBAAqC,exF6gnBvC,CwF7gnBE,gBAAqC,exFihnBvC,CwFjhnBE,0BAAqC,exFqhnBvC,CwFrhnBE,sBAAqC,exFyhnBvC,CwFzhnBE,wBAAqC,exF6hnBvC,CwF7hnBE,iBAAqC,exFiinBvC,CwFjinBE,mBAAqC,exFqinBvC,CwFrinBE,mBAAqC,exFyinBvC,CwFzinBE,yBAAqC,exF6inBvC,CwF7inBE,wBAAqC,exFijnBvC,CwFjjnBE,kBAAqC,exFqjnBvC,CwFrjnBE,kBAAqC,exFyjnBvC,CwFzjnBE,0BAAqC,exF6jnBvC,CwF7jnBE,kDAAqC,exFqknBvC,CwFrknBE,wBAAqC,exFyknBvC,CwFzknBE,sBAAqC,exF6knBvC,CwF7knBE,qBAAqC,exFilnBvC,CwFjlnBE,wBAAqC,exFqlnBvC,CwFrlnBE,mBAAqC,exFylnBvC,CwFzlnBE,kBAAqC,exF6lnBvC,CwF7lnBE,qBAAqC,exFimnBvC,CwFjmnBE,gEAAqC,exFymnBvC,CwFzmnBE,oBAAqC,exF6mnBvC,CwF7mnBE,gBAAqC,exFinnBvC,CwFjnnBE,qBAAqC,exFqnnBvC,CwFrnnBE,iBAAqC,exFynnBvC,CwFznnBE,oBAAqC,exF6nnBvC,CwF7nnBE,iBAAqC,exFionBvC,CwFjonBE,qBAAqC,exFqonBvC,CwFronBE,uBAAqC,exFyonBvC,CwFzonBE,qBAAqC,exF6onBvC,CwF7onBE,kCAAqC,exFipnBvC,CwFjpnBE,mBAAqC,exFqpnBvC,CwFrpnBE,kBAAqC,exFypnBvC,CwFzpnBE,iBAAqC,exF6pnBvC,CwF7pnBE,mBAAqC,exFiqnBvC,CwFjqnBE,mBAAqC,exFqqnBvC,CwFrqnBE,2BAAqC,exFyqnBvC,CwFzqnBE,kBAAqC,exF6qnBvC,CwF7qnBE,yBAAqC,exFirnBvC,CwFjrnBE,kBAAqC,exFqrnBvC,CwFrrnBE,iBAAqC,exFyrnBvC,CwFzrnBE,cAAqC,exF6rnBvC,CwF7rnBE,eAAqC,exFisnBvC,CwFjsnBE,wDAAqC,exFysnBvC,CwFzsnBE,gBAAqC,exF6snBvC,CwF7snBE,sBAAqC,exFitnBvC,CwFjtnBE,+BAAqC,exFqtnBvC,CwFrtnBE,uBAAqC,exFytnBvC,CwFztnBE,qBAAqC,exF6tnBvC,CwF7tnBE,gBAAqC,exFiunBvC,CwFjunBE,qBAAqC,exFqunBvC,CwFrunBE,oBAAqC,exFyunBvC,CwFzunBE,gBAAqC,exF6unBvC,CwF7unBE,4BAAqC,exFivnBvC,CwFjvnBE,kBAAqC,exFqvnBvC,CwFrvnBE,kBAAqC,exFyvnBvC,CwFzvnBE,iBAAqC,exF6vnBvC,CwF7vnBE,wBAAqC,exFiwnBvC,CwFjwnBE,iBAAqC,exFqwnBvC,CwFrwnBE,uBAAqC,exFywnBvC,CwFzwnBE,0BAAqC,exF6wnBvC,CwF7wnBE,yBAAqC,exFixnBvC,CwFjxnBE,qBAAqC,exFqxnBvC,CwFrxnBE,yBAAqC,exFyxnBvC,CwFzxnBE,sBAAqC,exF6xnBvC,CwF7xnBE,mBAAqC,exFiynBvC,CwFjynBE,oBAAqC,exFqynBvC,CwFrynBE,wBAAqC,exFyynBvC,CwFzynBE,kCAAqC,exF6ynBvC,CwF7ynBE,mBAAqC,exFiznBvC,CwFjznBE,mBAAqC,exFqznBvC,CwFrznBE,gBAAqC,exFyznBvC,CwFzznBE,kBAAqC,exF6znBvC,CwF7znBE,qBAAqC,exFi0nBvC,CwFj0nBE,yBAAqC,exFq0nBvC,CwFr0nBE,sBAAqC,exFy0nBvC,CwFz0nBE,4DAAqC,exFi1nBvC,CwFj1nBE,uBAAqC,exFq1nBvC,CwFr1nBE,uBAAqC,exFy1nBvC,CwFz1nBE,2BAAqC,exF61nBvC,CwF71nBE,eAAqC,exFi2nBvC,CwFj2nBE,yBAAqC,exFq2nBvC,CwFr2nBE,0BAAqC,exFy2nBvC,CwFz2nBE,qBAAqC,exF62nBvC,CwF72nBE,mBAAqC,exFi3nBvC,CwFj3nBE,sBAAqC,exFq3nBvC,CwFr3nBE,oBAAqC,exFy3nBvC,CwFz3nBE,kBAAqC,exF63nBvC,CwF73nBE,kBAAqC,exFi4nBvC,CwFj4nBE,8CAAqC,exFy4nBvC,CwFz4nBE,oBAAqC,exF64nBvC,CwF74nBE,mBAAqC,exFi5nBvC,CwFj5nBE,wBAAqC,exFq5nBvC,CwFr5nBE,gBAAqC,exFy5nBvC,CwFz5nBE,oBAAqC,exF65nBvC,CwF75nBE,mBAAqC,exFi6nBvC,CwFj6nBE,mBAAqC,exFq6nBvC,CwFr6nBE,yBAAqC,exFy6nBvC,CwFz6nBE,eAAqC,exF66nBvC,CwF76nBE,kBAAqC,exFi7nBvC,CwFj7nBE,kBAAqC,exFq7nBvC,CwFr7nBE,sBAAqC,exFy7nBvC,CwFz7nBE,mBAAqC,exF67nBvC,CwF77nBE,kBAAqC,exFi8nBvC,CwFj8nBE,iBAAqC,exFq8nBvC,CwFr8nBE,sBAAqC,exFy8nBvC,CwFz8nBE,gBAAqC,exF68nBvC,CwF78nBE,yBAAqC,exFi9nBvC,CwFj9nBE,sBAAqC,exFq9nBvC,CwFr9nBE,mBAAqC,exFy9nBvC,CwFz9nBE,kBAAqC,exF69nBvC,CwF79nBE,gBAAqC,exFi+nBvC,CwFj+nBE,8BAAqC,exFq+nBvC,CwFr+nBE,mBAAqC,exFy+nBvC,CwFz+nBE,uBAAqC,exF6+nBvC,CwF7+nBE,oBAAqC,exFi/nBvC,CwFj/nBE,kBAAqC,exFq/nBvC,CwFr/nBE,qBAAqC,exFy/nBvC,CwFz/nBE,yBAAqC,exF6/nBvC,CwF7/nBE,oBAAqC,exFigoBvC,CwFjgoBE,wCAAqC,exFygoBvC,CwFzgoBE,0BAAqC,exF6goBvC,CwF7goBE,qBAAqC,exFihoBvC,CwFjhoBE,0DAAqC,exFyhoBvC,CwFzhoBE,+BAAqC,exF6hoBvC,CwF7hoBE,kBAAqC,exFiioBvC,CwFjioBE,uBAAqC,exFqioBvC,CwFrioBE,qBAAqC,exFyioBvC,CwFzioBE,gBAAqC,exF6ioBvC,CwF7ioBE,qCAAqC,exFijoBvC,CwFjjoBE,uBAAqC,exFqjoBvC,CwFrjoBE,kBAAqC,exFyjoBvC,CwFzjoBE,sDAAqC,exFikoBvC,CwFjkoBE,kBAAqC,exFqkoBvC,CwFrkoBE,iBAAqC,exFykoBvC,CwFzkoBE,iBAAqC,exF6koBvC,CwF7koBE,oBAAqC,exFiloBvC,CwFjloBE,mBAAqC,exFqloBvC,CwFrloBE,qBAAqC,exFyloBvC,CwFzloBE,kBAAqC,exF6loBvC,CwF7loBE,mBAAqC,exFimoBvC,CwFjmoBE,iBAAqC,exFqmoBvC,CwFrmoBE,kDAAqC,exF6moBvC,CwF7moBE,iBAAqC,exFinoBvC,CwFjnoBE,oBAAqC,exFqnoBvC,CwFrnoBE,uBAAqC,exFynoBvC,CwFznoBE,iBAAqC,exF6noBvC,CwF7noBE,kCAAqC,exFiooBvC,CwFjooBE,yBAAqC,exFqooBvC,CwFrooBE,mBAAqC,exFyooBvC,CwFzooBE,qBAAqC,exF6ooBvC,CwF7ooBE,kBAAqC,exFipoBvC,CwFjpoBE,kBAAqC,exFqpoBvC,CwFrpoBE,oBAAqC,exFypoBvC,CwFzpoBE,oBAAqC,exF6poBvC,CwF7poBE,sBAAqC,exFiqoBvC,CwFjqoBE,iBAAqC,exFqqoBvC,CwFrqoBE,4BAAqC,exFyqoBvC,CwFzqoBE,uBAAqC,exF6qoBvC,CwF7qoBE,oBAAqC,exFiroBvC,CwFjroBE,sBAAqC,exFqroBvC,CwFrroBE,4BAAqC,exFyroBvC,CwFzroBE,wBAAqC,exF6roBvC,CwF7roBE,oBAAqC,exFisoBvC,CwFjsoBE,yBAAqC,exFqsoBvC,CwFrsoBE,sBAAqC,exFysoBvC,CwFzsoBE,uBAAqC,exF6soBvC,CwF7soBE,kBAAqC,exFitoBvC,CwFjtoBE,gBAAqC,exFqtoBvC,CwFrtoBE,wBAAqC,exFytoBvC,CwFztoBE,wBAAqC,exF6toBvC,CwF7toBE,uBAAqC,exFiuoBvC,CwFjuoBE,qBAAqC,exFquoBvC,CwFruoBE,gBAAqC,exFyuoBvC,CwFzuoBE,kBAAqC,exF6uoBvC,CwF7uoBE,wBAAqC,exFivoBvC,CwFjvoBE,gBAAqC,exFqvoBvC,CwFrvoBE,kBAAqC,exFyvoBvC,CwFzvoBE,qBAAqC,exF6voBvC,CwF7voBE,qBAAqC,exFiwoBvC,CwFjwoBE,kBAAqC,exFqwoBvC,CwFrwoBE,mBAAqC,exFywoBvC,CwFzwoBE,qBAAqC,exF6woBvC,CwF7woBE,0BAAqC,exFixoBvC,CwFjxoBE,qBAAqC,exFqxoBvC,CwFrxoBE,0BAAqC,exFyxoBvC,CwFzxoBE,yBAAqC,exF6xoBvC,CwF7xoBE,kBAAqC,exFiyoBvC,CwFjyoBE,kBAAqC,exFqyoBvC,CwFryoBE,iBAAqC,exFyyoBvC,CwFzyoBE,kBAAqC,exF6yoBvC,CwF7yoBE,8BAAqC,exFizoBvC,CwFjzoBE,gBAAqC,exFqzoBvC,CwFrzoBE,0CAAqC,exF6zoBvC,CwF7zoBE,4BAAqC,exFi0oBvC,CwFj0oBE,sBAAqC,exFq0oBvC,CwFr0oBE,0BAAqC,exFy0oBvC,CwFz0oBE,mBAAqC,exF60oBvC,CwF70oBE,eAAqC,exFi1oBvC,CwFj1oBE,uBAAqC,exFq1oBvC,CwFr1oBE,iBAAqC,exFy1oBvC,CwFz1oBE,wBAAqC,exF61oBvC,CwF71oBE,gBAAqC,exFi2oBvC,CwFj2oBE,mBAAqC,exFq2oBvC,CwFr2oBE,kBAAqC,exFy2oBvC,CwFz2oBE,8CAAqC,exFi3oBvC,CwFj3oBE,gCAAqC,exFq3oBvC,CwFr3oBE,gBAAqC,exFy3oBvC,CwFz3oBE,eAAqC,exF63oBvC,CwF73oBE,oDAAqC,exFq4oBvC,CwFr4oBE,kBAAqC,exFy4oBvC,CwFz4oBE,mBAAqC,exF64oBvC,CwF74oBE,iBAAqC,exFi5oBvC,CwFj5oBE,gBAAqC,exFq5oBvC,CwFr5oBE,sBAAqC,exFy5oBvC,CwFz5oBE,kBAAqC,exF65oBvC,CwF75oBE,yBAAqC,exFi6oBvC,CwFj6oBE,sBAAqC,exFq6oBvC,CwFr6oBE,kBAAqC,exFy6oBvC,CwFz6oBE,uBAAqC,exF66oBvC,CwF76oBE,4CAAqC,exFq7oBvC,CwFr7oBE,kDAAqC,exF67oBvC,CwF77oBE,kBAAqC,exFi8oBvC,CwFj8oBE,kCAAqC,exFq8oBvC,CwFr8oBE,sBAAqC,exFy8oBvC,CwFz8oBE,mBAAqC,exF68oBvC,CwF78oBE,iBAAqC,exFi9oBvC,CwFj9oBE,kBAAqC,exFq9oBvC,CwFr9oBE,0BAAqC,exFy9oBvC,CwFz9oBE,mBAAqC,exF69oBvC,CwF79oBE,qBAAqC,exFi+oBvC,CwFj+oBE,mBAAqC,exFq+oBvC,CwFr+oBE,kBAAqC,exFy+oBvC,CwFz+oBE,mBAAqC,exF6+oBvC,CwF7+oBE,gBAAqC,exFi/oBvC,CwFj/oBE,0BAAqC,exFq/oBvC,CwFr/oBE,oBAAqC,exFy/oBvC,CwFz/oBE,cAAqC,exF6/oBvC,CwF7/oBE,iBAAqC,exFigpBvC,CwFjgpBE,eAAqC,exFqgpBvC,CwFrgpBE,eAAqC,exFygpBvC,CwFzgpBE,kBAAqC,exF6gpBvC,CwF7gpBE,gCAAqC,exFihpBvC,CwFjhpBE,mBAAqC,exFqhpBvC,CwFrhpBE,gBAAqC,exFyhpBvC,CwFzhpBE,kBAAqC,exF6hpBvC,CwF7hpBE,eAAqC,exFiipBvC,CwFjipBE,kBAAqC,exFqipBvC,CwFripBE,mBAAqC,exFyipBvC,CwFzipBE,oBAAqC,exF6ipBvC,CwF7ipBE,2BAAqC,exFijpBvC,CwFjjpBE,kBAAqC,exFqjpBvC,CwFrjpBE,gBAAqC,exFyjpBvC,CwFzjpBE,mBAAqC,exF6jpBvC,CwF7jpBE,mBAAqC,exFikpBvC,CwFjkpBE,mBAAqC,exFqkpBvC,CwFrkpBE,qBAAqC,exFykpBvC,CwFzkpBE,kBAAqC,exF6kpBvC,CwF7kpBE,eAAqC,exFilpBvC,CwFjlpBE,kBAAqC,exFqlpBvC,CwFrlpBE,eAAqC,exFylpBvC,CwFzlpBE,oBAAqC,exF6lpBvC,CwF7lpBE,8BAAqC,exFimpBvC,CwFjmpBE,6BAAqC,exFqmpBvC,CwFrmpBE,mBAAqC,exFympBvC,CwFzmpBE,8CAAqC,exFinpBvC,CwFjnpBE,wBAAqC,exFqnpBvC,CwFrnpBE,kBAAqC,exFynpBvC,CwFznpBE,sDAAqC,exFiopBvC,CwFjopBE,mBAAqC,exFqopBvC,CwFropBE,uBAAqC,exFyopBvC,CwFzopBE,uCAAqC,exFippBvC,CwFjppBE,iBAAqC,exFqppBvC,CwFrppBE,eAAqC,exFyppBvC,CwFzppBE,kBAAqC,exF6ppBvC,CwF7ppBE,kBAAqC,exFiqpBvC,CwFjqpBE,iBAAqC,exFqqpBvC,CwFrqpBE,4BAAqC,exFyqpBvC,CwFzqpBE,uBAAqC,exF6qpBvC,CwF7qpBE,sCAAqC,exFqrpBvC,CwFrrpBE,kBAAqC,exFyrpBvC,CwFzrpBE,oBAAqC,exF6rpBvC,CwF7rpBE,qBAAqC,exFispBvC,CwFjspBE,mBAAqC,exFqspBvC,CwFrspBE,yBAAqC,exFyspBvC,CwFzspBE,kBAAqC,exF6spBvC,CwF7spBE,qBAAqC,exFitpBvC,CwFjtpBE,kBAAqC,exFqtpBvC,CwFrtpBE,oBAAqC,exFytpBvC,CwFztpBE,kBAAqC,exF6tpBvC,CwF7tpBE,+BAAqC,exFiupBvC,CwFjupBE,uBAAqC,exFqupBvC,CwFrupBE,gBAAqC,exFyupBvC,CwFzupBE,0CAAqC,exF6upBvC,CwF7upBE,kBAAqC,exFivpBvC,CwFjvpBE,iBAAqC,exFqvpBvC,CwFrvpBE,+BAAqC,exFyvpBvC,CwFzvpBE,qBAAqC,exF6vpBvC,CwF7vpBE,kBAAqC,exFiwpBvC,CwFjwpBE,4BAAqC,exFqwpBvC,CwFrwpBE,uBAAqC,exFywpBvC,CwFzwpBE,gBAAqC,exF6wpBvC,CwF7wpBE,sBAAqC,exFixpBvC,CwFjxpBE,oBAAqC,exFqxpBvC,CwFrxpBE,iBAAqC,exFyxpBvC,CwFzxpBE,kBAAqC,exF6xpBvC,CwF7xpBE,eAAqC,exFiypBvC,CwFjypBE,wBAAqC,exFqypBvC,CyFj0pBA,KAOE,wBAAyB,CACzB,8DAAmE,CACnE,6BAA8B,CCH9B,yBAA0B,CAI1B,2BAA4B,CAC5B,2BAA4B,CAC5B,0BAA2B,CAC3B,yBAA0B,CAC1B,0BAA2B,CAC3B,uBAAwB,CAGxB,4BAA6B,CAG7B,8BAA+B,CAC/B,qCAAsC,CACtC,qCAAsC,CACtC,oCAAqC,CAGrC,wCAAyC,CAGzC,6BAA8B,CAC9B,6BAA8B,CAI9B,uKAGA,6HAEA,yDAA0D,CAC1D,4DAA6D,CAC7D,mEAAoE,CCpCpE,2BAA4B,CAG5B,4BACA,2BACA,oCACA,kCACA,4BACA,yBACA,wBACA,0BACA,2BACA,6BACA,4BACA,sBACA,sBACA,qBACA,oBACA,sBAGA,6BCtBA,kDAAmD,CACnD,sDAAuD,CACvD,kEAAmE,CACnE,kEAAmE,CACnE,kEAAmE,CACnE,iEAAkE,CAClE,wDAAyD,CACzD,oDAAqD,CACrD,mDAAoD,CACpD,kEAAmE,CACnE,mDAAoD,CACpD,iDAAkD,CCXlD,qEAAsE,CACtE,mEAAoE,CACpE,qEAAsE,CACtE,wEAAyE,CCPzE,mCAAoC,CACpC,wD9Fw0pBF,C8Fr0pBA,KACE,4CAA6C,CAI7C,gCAAiC,CAEjC,YAAa,CACb,qBAAsB,CANtB,uCAAwC,CACxC,eAAgB,CAChB,gBJoCiB,CIlCjB,gB9F8wqBF,C8FzwqBE,8BACE,4C9F2wqBJ,C8FvwqBA,EAEE,aAAc,CADd,qB9F4wqBF,C8FxwqBE,SACE,+C9F0wqBJ,C8FvwqBE,WACE,iB9FywqBJ,C8FrwqBA,ECkCE,oBAAqB,CACrB,2BAA4B,CAvC5B,yBAA0B,CAGxB,2CAvBuB,CA2BvB,6B/F0wqBJ,C+FtuqBE,QACE,iCAAkC,CA1BlC,6BAA8B,CAH9B,iD/FuwqBJ,C+F/tqBE,mBACE,2B/FouqBJ,C+FluqBI,gBACE,iC/FouqBN,C8FvxqBE,aACE,gCAAiC,CACjC,cAAgB,CAEhB,gBAAkB,CADlB,aAAc,CAEd,oBAAqB,CACrB,gB9FyxqBJ,C8FnxqBI,gCACE,iCAAkC,CAClC,0BAA2B,CAC3B,mB9FqxqBN,C8FjxqBE,gBACE,8B9FmxqBJ,C8FhxqBE,gBACE,8B9FkxqBJ,C8F9wqBA,0CAEE,0CAA2C,CAC3C,0CAA2C,CAC3C,gBAAiB,CAHjB,wB9FoxqBF,C8F9wqBA,OAIE,iCAAkC,CADlC,Y9FixqBF,C8F5wqBA,cAHE,8B9FoxqBF,C8FjxqBA,OAGE,iC9F8wqBF,C8F1wqBA,OAGE,iC9F4wqBF,C8FxwqBA,cAHE,8B9FgxqBF,C8F7wqBA,OAGE,iC9F0wqBF,C8FtwqBA,OAGE,iC9FwwqBF,C8FpwqBA,cAHE,gC9F4wqBF,C8FzwqBA,OAGE,iC9FswqBF,C8FlwqBA,yBAEE,oC9FqwqBF,C8FlwqBA,GACE,QAAS,CACT,4C9FqwqBF,C8FlwqBA,kBAIE,4C9FqwqBF,C8FlwqBA,IAEE,+CAAgD,CAChD,iC9FowqBF,C8FjwqBE,mBACE,wCAAyC,CAGzC,8CAA+C,CAF/C,cAAgB,CAChB,mB9FowqBJ,C8F/vqBA,KACE,kC9FkwqBF,C8F/vqBA,IAGE,yCAA0C,CAG1C,wCAAyC,CACzC,oBL/HyB,CK4HzB,gCAAiC,CACjC,iBAAkB,CAJlB,cAAe,CACf,Y9FuwqBF,C8FhwqBE,aAEE,UAAY,CACZ,kB9FiwqBJ,C8F5vqBA,iBASE,2CAA4C,CAC5C,WAAY,CAFZ,qCAAsC,CAJtC,YAAa,CAEb,SAAU,CAHV,cAAe,CAEf,QAAS,CAET,yBAA0B,CAL1B,Y9FswqBF,C8F5vqBE,8BAGE,sB9F4vqBJ,C8FzvqBE,uBAIE,qDAAsD,CCnIxD,iCAAkC,CDoIhC,+CAAgD,CCjKlD,yBAA0B,CAkBxB,6BAA8B,CAf9B,2CAvBuB,CAmCvB,iDAtB8B,CDsK9B,6BAA8B,CCxJ9B,6B/Fy5qBJ,C8F5vqBE,+BACE,eAAgB,CAChB,8CrFnKiB,CqFoKjB,wCAAyC,CACzC,uB9F8vqBJ,C8FrvqBA,eAEE,eAAgB,CADhB,8C9FyvqBF,CgGn8qBA,WAME,yCAA0C,CAG1C,+CAAgD,CANhD,MAAO,CAIP,aAAe,CANf,cAAe,CAGf,OAAQ,CACR,iBAAkB,CAHlB,KAAM,CAUN,2BAA4B,CAC5B,qCAAuC,CALvC,YhG+8qBF,CgGx8qBE,wBACE,uBhG08qBJ,CgGv8qBE,aAEE,6ChGw8qBJ,CgGt8qBI,2BAEE,qBhGu8qBN,CiGv+qBA,cAEE,YAAa,CADb,WAAY,CAEZ,sBjG0+qBF,CiGx+qBE,mCACE,YjG0+qBJ,CiGt+qBA,eACE,UjGy+qBF,Ccr7qBI,yBmFrDJ,eAII,ejG0+qBF,CACF,CkGz/qBA,2BAGE,QAAS,CACT,eAAgB,CAIhB,iClGu/qBF,CkGr/qBE,gCAVF,2BAWI,elGw/qBF,CACF,CkGr/qBA,mDAME,kBAAmB,CAFnB,YAAa,CAGb,sBAAuB,CALvB,eAAgB,CAOhB,mBAAqB,CAJrB,iBAAkB,CAGlB,iBAAkB,CALlB,UlG8/qBF,Ccp8qBI,4BoF7DJ,mDAaI,gBlGy/qBF,CACF,CkGv/qBE,uDACE,eAAiB,CACjB,QlG0/qBJ,CkGt/qBE,uDACE,wClGy/qBJ,CkGr/qBE,yHAOE,wCAJmB,CAKnB,oCALmB,CAOnB,oBAAsB,CC5CxB,+FACE,CD0CA,kClGo/qBJ,CkGj/qBI,qIAKE,kDAbuB,CAcvB,8CAduB,CHc3B,iCAAkC,CGC9B,4CAA6C,CH9BjD,yBAA0B,CAkBxB,6BAA8B,CAf9B,2CAvBuB,CAmCvB,iDAtB8B,CGkC5B,6BAA8B,CHpBhC,6B/F8grBJ,CkGp/qBI,qJACE,+CACA,uBlGu/qBN,CkGj/qBA,wBACE,8ClGo/qBF,CkGj/qBA,2BACE,2ClGo/qBF,CoG9jrBA,SAGE,YAAa,CADb,qBAAsB,CADtB,WAAY,CAGZ,WpGokrBF,CoGlkrBE,qBACE,YAAa,CAEb,WAAY,CADZ,sBpGqkrBJ,CoGlkrBI,2CAEE,YAAa,CACb,qBAAsB,CAFtB,qBAAsB,CAQtB,cAAe,CACf,eAAgB,CAChB,YAAa,CAHb,UpGmkrBN,Cc1hrBI,0BsFpCE,uDAII,iBAAkB,CADlB,kBpGgkrBR,CACF,CqG/lrBA,WAEE,4CAA6C,CAD7C,UrGmmrBF,CqGhmrBE,6BACE,YAAa,CACb,WAAY,CAEZ,WAAY,CADZ,YrGmmrBJ,CqG/lrBE,+FAGE,YAAa,CACb,qBAAsB,CAGtB,WAAY,CAFZ,SAAW,CACX,sBrGkmrBJ,CqG9lrBE,iCACE,iBrGgmrBJ,CqG7lrBE,8BACE,crG+lrBJ,CqG3lrBE,0BACE,erG6lrBJ,CsG5nrBA,mBACE,etG+nrBF,CsG7nrBE,yCACE,YAAa,CACb,qBtG+nrBJ,CuGnorBE,yCACE,YAAa,CACb,qBAAsB,CACtB,evGsorBJ,CwGjorBA,WAME,yDAA2D,CAC3D,qDAAwD,CAIxD,sBAAuB,CADvB,eAAgB,CADhB,SAAU,CARV,eAAgB,CAChB,KAAM,CAMN,UAAW,CALX,YxG8orBF,CwGporBE,6BAEE,kBAAmB,CADnB,YAAa,CAEb,kBAAmB,CACnB,iBAAkB,CAClB,kBxGsorBJ,CwGnorBE,0BACE,qBxGqorBJ,CwGjorBE,wBAIE,kBAAmB,CADnB,YAAa,CAFb,+BAAgC,CAChC,mCxGqorBJ,CwG/nrBE,gCACE,YAAa,CACb,axGiorBJ,Cc/mrBI,yB0FpBF,gCAKI,eAAgB,CAChB,WAAY,CACZ,mBxGkorBJ,CACF,CwG/nrBE,oHAIE,kBAAmB,CADnB,YAAa,CAEb,cAAe,CAGf,SxG+nrBJ,CwG5nrBE,6EAEE,kBxG8nrBJ,CwG1nrBE,uCACE,aAAc,CAEd,SAAW,CADX,iBxG6nrBJ,CwGznrBE,qCAEE,mBxG0nrBJ,CwGtnrBE,yBACE,YxGwnrBJ,Cc7orBI,yB0FoBF,yBAKI,oBxGwnrBJ,CACF,CwGtnrBI,qCACE,iBxGwnrBN,CwGtnrBM,+CT0FJ,iCAAkC,CSvF5B,kBAAmB,CT2FzB,iB/F6hrBF,C+F1hrBE,sDAKE,4BAA6B,CAJ7B,UAAW,CACX,aAAc,CAEd,OAAQ,CAIN,0CACA,iCANF,iB/FgirBJ,C+FthrBE,qDACE,gCAAiC,CACjC,oB/FwhrBJ,C+FvhrBI,4DAEI,sE/FwhrBR,C+FjhrBE,6DACE,eAAgB,CAChB,yCAA0C,CAC1C,uB/FmhrBJ,CwGzorBQ,uDACE,8BxG2orBV,CwGxorBU,8DACE,gDxG0orBZ,CwGrorBM,8CACE,iBxGuorBR,CwGrorBQ,qDACE,kBxGuorBV,CwGporBQ,+DACE,qBfnFU,CeoFV,iCxGsorBV,CwGporBU,6EACE,iDxGsorBZ,CwGnorBU,qEAEE,sDAEA,gCAAiC,CAHjC,oBxGuorBZ,CwG9nrBI,qDTpCF,iCAAkC,CAClC,oB/FqqrBF,C+FnqrBE,2DACE,iCAAkC,CAzEpC,yBAA0B,CAkBxB,6BAA8B,CAf9B,2CAvBuB,CAmCvB,iDAtB8B,CAc9B,6B/F4urBJ,CwGporBM,0CAEE,WAAY,CADZ,axGuorBR,CwGporBQ,gDTtFN,iCAAkC,CA7BlC,yBAA0B,CAkBxB,6BAA8B,CAf9B,2CAvBuB,CAmCvB,iDAtB8B,CAc9B,6B/FwvrBJ,CwGvorBM,kDAIE,+CAAgD,CAFhD,wCAAyC,CACzC,kDAAqD,CAGrD,cAAgB,CAChB,eAAgB,CAFhB,eAAiB,CAJjB,YxG+orBR,CwGvorBQ,iEAGE,qBxGuorBV,CwGporBU,2FACE,wBxGsorBZ,CwGnorBU,+EACE,UxGqorBZ,CwGhorBQ,6DACE,YxGkorBV,CwGvnrBE,iCAIE,wBAAyB,CAFzB,4BAA6B,CAD7B,mCAAoC,CAEpC,eAAgB,CAEhB,axGynrBJ,CwGtnrBE,iCACE,iBxGwnrBJ,Cc9vrBI,yB0FqIF,iCAII,YxGynrBJ,CACF,CwGtnrBE,mCACE,gBxGwnrBJ,CctwrBI,0B0F6IF,mCAII,YxGynrBJ,CACF,Cc3wrBI,yB0FwJF,kCACE,oBxGunrBF,CACF,CwGpnrBA,UAME,exGinrBF,CwGtnrBE,gBTjKA,iCAAkC,CA7BlC,yBAA0B,CAkBxB,6BAA8B,CAf9B,2CAvBuB,CAmCvB,iDAtB8B,CAc9B,6B/FqzrBJ,CwGvnrBE,6BAEE,qCAAsC,CADtC,yBAA0B,CAE1B,eAAiB,CACjB,gBxGynrBJ,CwGrnrBA,kCACE,eAAiB,CACjB,gBAAiB,CACjB,qBxGwnrBF,CwGpnrBA,2BACE,gBxGunrBF,Cc5yrBI,yB0FoLJ,2BAII,YxGwnrBF,CACF,CwGpnrBA,8BACE,YxGunrBF,CcrzrBI,yB0F6LJ,8BAII,YxGwnrBF,CACF,CyGz3rBA,uBACE,ezG43rBF,CyGz3rBE,4CACE,2CzG23rBJ,CyGx3rBE,kDACE,gBzG03rBJ,C0G53rBA,oBAaE,4CAA6C,CAD7C,8CAA+C,CAX/C,YAAa,C1FuBX,aAAc,C0FtBhB,qBAAsB,CAatB,6CAA8C,CAZ9C,QAAS,CACT,iDAAkD,CAUlD,eAAgB,CAHhB,uBANA,eAAgB,CAChB,4BAA6B,C1FmB3B,ShBu3rBJ,Ccx1rBI,yB4FxDJ,oBAkBI,sC1Gk4rBF,CACF,C0Gh4rBE,mCACE,qB1Gk4rBJ,C0G93rBE,8BACE,6C1Gg4rBJ,C0G73rBE,+BACE,c1G+3rBJ,Cct2rBI,yB4FtBF,iCAEI,Y1G83rBJ,CACF,C0G13rBE,4LAIE,gC1G+3rBJ,C0G13rBI,8IACE,e1G63rBN,C0Gx3rBE,0CACE,YAAa,CACb,qB1G03rBJ,C0Gx3rBI,uEAGE,gCAAiC,CADjC,6CAA8C,CAD9C,iDAAkD,CAGlD,mB1G03rBN,C0Gr3rBQ,sFACE,eAAgB,CAChB,8CjG5CW,CiG6CX,uB1Gu3rBV,C0Gj3rBI,wEACE,YAAa,CACb,qB1Gm3rBN,C0G/2rBI,qEAEE,kBAAmB,CADnB,YAAa,CAEb,Q1Gi3rBN,Cch5rBI,yB4FAF,0CAmCI,Y1Gi3rBJ,CACF,C0G92rBE,kDAEE,4C1G+2rBJ,Ccx5rBI,yB4FuCF,kDAKI,e1Gg3rBJ,CACF,C0G72rBE,gDAEE,iBAAkB,CADlB,e1Gg3rBJ,C0G52rBE,kCACE,eAAgB,CAChB,c1G82rBJ,C0Gx2rBQ,6EAME,oBAAqB,CAFrB,YAAa,CAFb,WAAY,CAGZ,6BAA8B,CAJ9B,iBAAkB,CAElB,U1G62rBV,C0Gx2rBU,6FAGE,a1Gw2rBZ,C0Gp2rBY,8GACE,c1Gs2rBd,C0G91rBE,oCAGE,iB1G81rBJ,C0G51rBI,4FAEE,mCAIA,kB1G21rBN,C0Gv1rBM,oDAEE,e1Gw1rBR,C0Gt1rBQ,4EACE,Y1Gw1rBV,C0Gp1rBQ,kEACE,Y1Gs1rBV,C0Gp1rBU,kFACE,8CjG9IS,CiG+IT,wB1Gs1rBZ,C0Gj1rBQ,oEAgBE,kBAAmB,CAfnB,cAAe,CAaf,mBAAoB,CAHpB,WA5CiB,CAgDjB,sBAAuB,CAVvB,iBAAkB,CAElB,OAAQ,CADR,KAAM,CAIN,U1Gk1rBV,C0G10rBU,qFACE,gB1G40rBZ,C0Gn0rBU,2EACE,wB1Gq0rBZ,C0G5zrBA,aACE,kB1Gg0rBF,Cc99rBI,yB4F6JJ,aAII,a1Gi0rBF,CACF,C0G/zrBE,gBACE,aAAc,CACd,e1Gi0rBJ,C0G9zrBI,mBACE,kB1Gg0rBN,C0G5zrBE,kBAME,eAAgB,CX9HlB,iCAAkC,CWyHhC,aAAc,CAMd,qBjG9Ne,CiGyNf,sBXzHF,oB/F27rBF,C+Fz7rBE,wBAEE,4BAA6B,CAD7B,yBAA0B,CAxE1B,6BAA8B,CAH9B,iD/F0gsBJ,C+Fv7rBE,iDANE,iC/Fg8rBJ,C+Ft7rBE,gCACE,iDtF5GoB,CsF6GpB,YAAa,CACb,U/Fw7rBJ,C0Gz0rBM,2CAEE,qCAAsC,CADtC,yBAA0B,CAE1B,eAAiB,CACjB,gB1G20rBR,C0Gt0rBE,wBAGE,4BAA6B,CXtG7B,qEANe,CAHjB,8BAA+B,CAD/B,e/Fy7rBF,C+F76rBI,sCACE,6GACA,Y/F+6rBN,C0Gz0rBE,+BACE,6C1G60rBJ,C0Gv0rBE,sDALE,iDAAkD,CAClD,mB1Gm1rBJ,C0G/0rBE,uBAKE,gCAAiC,CAMjC,6CAA8C,CAR9C,kBAAmB,CADnB,iB1G60rBJ,C0Gx0rBI,mCACE,Y1G00rBN,CcjisBI,yB4F+MF,uBAcI,sC1Gw0rBJ,CACF,C2GjmsBA,sBAiBE,4CAA6C,CAhB7C,YAAa,CAGb,qBAAsB,CADtB,aAAc,CAOd,6CAA8C,CAH9C,iDAAkD,CALlD,OAAQ,CAgBR,eAAgB,CAVhB,sBAAuB,CAHvB,eAAgB,CAChB,4BAA6B,CAG7B,kC3G2msBF,CczjsBI,0B6F3DJ,sBAaI,sC3G2msBF,CACF,C2GpmsBA,wBACE,a3GumsBF,CclksBI,0B6FtCJ,wBAII,6CAA8C,CAC9C,iB3GwmsBF,CACF,C2GtmsBE,0BACE,mB3GwmsBJ,C4GjmsBA,sBACE,qBAAuB,CACvB,U5GmnsBF,CcllsBI,4B8F9BJ,oBAhBE,QAAS,CANT,cAAe,CAHf,YAAa,CAgBX,MAAO,CADP,gBAAiB,CAdnB,gBAAiB,CAGjB,eAAgB,CAPhB,cAAe,CACf,KAAM,CAON,sDACE,CAEF,iBAAkB,CANlB,SAAU,CAHV,Y5G2psBA,CACF,C4G7nsBE,0BACE,aAAc,CACd,kB5G+nsBJ,CcxmsBI,6B8FnBJ,sBA3BE,QAAS,CANT,cAAe,CAHf,YAAa,CAYX,iBAAkB,CAXpB,gBAAiB,CAGjB,eAAgB,CAPhB,cAAe,CAgBb,OAAQ,CAfV,KAAM,CAON,sDACE,CAEF,iBAAkB,CANlB,SAAU,CAHV,Y5GirsBA,CACF,C4GxosBE,4BACE,cAAe,CACf,kB5G0osBJ,C6GvssBA,kBAGE,YAAa,CACb,cAAe,CAGf,eAAiB,CANjB,eAAgB,CAChB,c7GgtsBF,C6GzssBE,qCAEE,kBAAmB,CADnB,YAAa,CAiBb,eAAiB,CAdjB,iBAAkB,CADlB,kB7G6ssBJ,C6G1ssBI,sFAOE,epGHa,CoGFb,iBAAkB,CAClB,sB7G6ssBN,C6GlssBI,uCdoEF,iCAAkC,CAClC,oB/FiosBF,C+F/nsBE,6CACE,iCAAkC,CAzEpC,yBAA0B,CAkBxB,6BAA8B,CAf9B,2CAvBuB,CAmCvB,iDAtB8B,CAc9B,6B/FwssBJ,C6GzssBI,kEAIE,iCAAkC,CADlC,qCAAsC,CAFtC,yBAA0B,CAC1B,eAAiB,CAGjB,e7G2ssBN,C8G/usBE,sBAKE,kBAAmB,CAKnB,WAAY,CADZ,eAAgB,CX+ClB,qBAAsB,CJ0HtB,iCAAkC,Ce9KhC,YAAa,CAOb,cAAe,CAKf,eAVA,sBAAuB,CAMvB,gBpB2Be,CSmBjB,eAAgB,CADhB,cAAe,CW5Cb,iBfyKF,iB/F6ksBF,C+F1ksBE,6BAKE,4BAA6B,CAJ7B,UAAW,CACX,aAAc,CAEd,OAAQ,CAIN,0CACA,iCANF,iB/FglsBJ,C+FtksBE,4BACE,gCAAiC,CACjC,oB/FwksBJ,C+FvksBI,mCAEI,sE/FwksBR,C+FjksBE,oCACE,eAAgB,CAChB,yCAA0C,CAC1C,uB/FmksBJ,C8GpwsBA,qBAKE,kBAAmB,CAFnB,kBrBcoB,CqBhBpB,YAAa,CACb,kBAAmB,CAEnB,4BAA6B,CAI7B,eAAgB,CADhB,eAAgB,CADhB,c9GywsBF,C8GnwsBI,mGAKE,mCAAoC,CADpC,iBAAkB,CADlB,qB9GwwsBN,C8GlwsBI,gDACE,a9GowsBN,C8GjwsBI,+CACE,U9GmwsBN,C8GhwsBI,2CACE,a9GkwsBN,C8G7vsBE,yCAEE,mBAAqB,CADrB,Y9GgwsBJ,C8G5vsBE,gCACE,kCAAmC,CACnC,qBAAsB,CACtB,gB9G8vsBJ,C+G/zsBA,cAWE,kBAAmB,CAJnB,YAAa,CAGb,aAAc,CAEd,SAAW,CAVX,+BAAgC,CAIhC,QAAS,CAHT,mCAAoC,CACpC,eAAiB,CAHjB,iBAAkB,CAIlB,U/G00sBF,C+Gh0sBE,gBACE,gCAAiC,CACjC,e/Gk0sBJ,C+G9zsBE,kBAEE,WAAY,CADZ,cAAe,CAEf,U/Gg0sBJ,C+G7zsBE,gDhB6BA,iCAAkC,CgBzBhC,gCAAiC,ChBJnC,yBAA0B,CAkBxB,6BAA8B,CAf9B,2CAvBuB,CAmCvB,iDAtB8B,CAc9B,6B/F+zsBJ,CgHh2sBE,iBACE,kBhHs2sBJ,CgHl2sBE,iBACE,YAAa,CACb,qBhHo2sBJ,CgHl2sBI,mBAEE,kBAAmB,CjBoFvB,iCAAkC,CiBrF9B,YAAa,CAEb,WAAY,CAEZ,qBvBmBkB,CuBpBlB,kBvBoBkB,CM+DtB,oB/FoxsBF,C+FlxsBE,yBACE,iCAAkC,CAzEpC,yBAA0B,CAkBxB,6BAA8B,CAf9B,2CAvBuB,CAmCvB,iDAtB8B,CAc9B,6B/F21sBJ,CiH73sBE,uBAEE,kBAAmB,CADnB,cjHo4sBJ,CiHj4sBI,0BACE,iBjHm4sBN,CiH93sBE,oBACE,6CjHg4sBJ,Cch1sBI,0BmGjDF,oBAII,sCjHi4sBJ,CACF,CiH93sBE,sBACE,gCAAiC,CACjC,iDAAkD,CAClD,mBjHg4sBJ,CkHt5sBA,gBACE,UlH45sBF,CkH15sBE,kBACE,iCAAkC,CAElC,iBAAkB,CADlB,alH65sBJ,CkHz5sBE,kBACE,elH25sBJ,CkHx5sBE,kBAGE,kBAAmB,CACnB,WAAY,CAIZ,iCAAkC,CANlC,YAAa,CAIb,aAAc,CACd,iBAAkB,CAFlB,YAAa,CAIb,oBlHy5sBJ,CkHv5sBI,oCnB0CF,oBAAqB,CACrB,2BAA4B,CmBvCxB,eAAgB,CADhB,qDAAsD,CnBC1D,yBAA0B,CAGxB,2CAvBuB,CA2BvB,6B/Fq5sBJ,C+Fj3sBE,0CACE,iCAAkC,CA1BlC,6BAA8B,CAH9B,iD/Fk5sBJ,C+F12sBE,uFACE,2B/F+2sBJ,C+F72sBI,kDACE,iC/F+2sBN,CkHr6sBM,4FnBwBJ,iCAAkC,CA7BlC,yBAA0B,CAkBxB,6BAA8B,CAf9B,2CAvBuB,CAmCvB,iDAtB8B,CAc9B,6B/F26sBJ,CkHx6sBI,kCACE,qBAAsB,CACtB,alH06sBN,CkHx6sBM,sDACE,yBlH06sBR,CkHt6sBI,4BACE,UlHw6sBN,CkHr6sBI,6BACE,WlHu6sBN,CkHr6sBM,gDACE,gBlHu6sBR,CmH99sBA,WAME,wCAAyC,CADzC,oB1BwByB,C0BtBzB,gCAAiC,CAJjC,SAAW,CADX,kBAAoB,CADpB,iBnH0+sBF,CmHj+sBE,6BANA,4CnH2+sBF,CmHr+sBE,kBAEE,iCnHm+sBJ,CmH/9sBE,wBACE,iDnHi+sBJ,CmH/9sBI,oCACE,enHi+sBN,CmH79sBE,iBAEE,6BAA8B,CAC9B,SAAU,CAFV,iBnHi+sBJ,CmH59sBE,yCAGE,iCAAkC,CADlC,0BAAgC,CADhC,iBnHg+sBJ,CmH39sBE,8BACE,4CAA6C,CAE7C,WAAY,CADZ,gCnH89sBJ,CmH19sBI,2CACE,iCnH49sBN,CmHx9sBI,qHAEE,enHy9sBN,CmHt9sBI,gFAEE,iCnHu9sBN,CmHl9sBE,wCAGE,6BAA8B,CAF9B,YAAa,CACb,uBnHq9sBJ,Cct9sBI,yBqGDF,wCAMI,YnHq9sBJ,CACF,CmH98sBA,iBAEE,gBnHm9sBF,CmH78sBA,mBACE,YnHm9sBF,CmHj9sBE,yBAaE,4BAA6B,CAE7B,WAAY,CAdZ,YAAa,CAMb,QAAS,CAGT,gBAAkB,CAElB,eAAgB,CAEhB,gB1G/Ee,C0GqEf,cAAe,CAKf,UAAW,CAHX,OAAQ,CAER,8BAAgC,CAGhC,SAAU,CANV,YnH29sBJ,CmH/8sBI,mCACE,qBAAuB,CACvB,UnHi9sBN,CmH98sBI,wCACE,WnHg9sBN,CmH78sBM,0FAEE,mCnH+8sBR,CmHr8sBA,qBAIE,kBAAmB,CAKnB,yCAA0C,CAJ1C,wCAAyC,CACzC,mBAL8B,CAM9B,iCAAkC,CAJlC,mBAAoB,CAKpB,YnH28sBF,CmHx8sBE,2BACE,qDnH08sBJ,CmHv8sBE,mCACE,mBnHy8sBJ,CmHr8sBE,kDACE,sCAAuC,CACvC,0CAA2C,CAE3C,gBAAkB,CADlB,iBnHw8sBJ,CmHp8sBE,6CACE,enHs8sBJ,CmHl8sBE,uBACE,kBnHo8sBJ,CmHh8sBE,+BACE,YnHk8sBJ,CctitBI,yBqGmGF,+BAII,YnHm8sBJ,CACF,CoHlmtBE,+BjBAA,+FACE,CiBAA,aAAc,CACd,iBpH4mtBJ,CcvjtBI,0BsGvDF,+BAMI,gBpH4mtBJ,CACF,CoHtmtBI,iCAIE,yCAA0C,CAH1C,oBAAsB,CAItB,mCAAoC,CAHpC,iBAAkB,CAClB,cAAgB,CAGhB,oBAAqB,CAGrB,mCpHsmtBN,CoHpmtBM,uCACE,8CpHsmtBR,CoHnmtBM,wCAEE,WAAY,CADZ,oCAAqC,CAErC,yBAA0B,CAC1B,kBpHqmtBR,CqHxotBE,mCACE,YrH8otBJ,CqH5otBI,0CAEE,iCAAkC,CADlC,oBrH+otBN,CqH3otBI,0CACE,gBrH6otBN,CqHvotBE,0NAGE,crHgptBJ,CsHvqtBA,oCACE,oCAAqC,CACrC,gCAAiC,CAGjC,iBtHwqtBF,Cc9mtBI,yBwG/DJ,oCAQI,mBtHyqtBF,CACF,CsHvqtBE,0CAEE,wBAAyB,CADzB,oDtH0qtBJ,CsHtqtBE,2CAEE,oCAAqC,CADrC,gCtHyqtBJ,CsHrqtBE,kDACE,wBtHuqtBJ,CsHnqtBA,wBACE,oCAAqC,CACrC,8CtHsqtBF,CsHpqtBE,0CACE,+CAAgD,CAChD,gCAAiC,CACjC,sBtHsqtBJ,CsHpqtBI,2DACE,+CtHsqtBN,CsHnqtBI,gDAGE,yCAA0C,CvBkB9C,iCAAkC,CA7BlC,yBAA0B,CAkBxB,6BAA8B,CAf9B,2CAvBuB,CAmCvB,iDAtB8B,CAc9B,6B/F6qtBJ,CsHtqtBI,iDvBgHA,qEANe,CAHjB,8BAA+B,CAD/B,eAAgB,CuBnGZ,iBAAkB,CAClB,StHyqtBN,C+F3jtBI,+DACE,6GACA,Y/F6jtBN,CsH3qtBM,6DACE,UAAW,CAEX,WAAY,CAGZ,MAAO,CAFP,iBAAkB,CAGlB,KAAM,CALN,UAAW,CAGX,UtH+qtBR,CsHzqtBI,wDACE,UtH2qtBN,CsHrqtBA,wDnBPE,qBAAsB,CmBStB,eAAgB,CnBPhB,eAAgB,CADhB,cAAe,CmBSf,YtH2qtBF,CcprtBI,yBwGMJ,wDAUI,etHyqtBF,CACF,CuHrvtBA,aAEE,kBvHyvtBF,CuHrvtBE,kBACE,YvHwvtBJ,CuH1utBA,6DACE,avHmvtBF,CuH7utBE,wCAHE,iCvH2vtBJ,CuHxvtBE,sBACE,aAAc,CAKd,iBAAkB,CADlB,4BAAkB,CxBiFpB,oB/FkqtBF,C+FhqtBE,4BAEE,4BAA6B,CAD7B,yBAA0B,CAxE1B,6BAA8B,CAH9B,iD/FivtBJ,C+F9ptBE,yDANE,iC/FuqtBJ,C+F7ptBE,oCACE,iDtF5GoB,CsF6GpB,YAAa,CACb,U/F+ptBJ,CuH7vtBI,6BAGE,4BAA6B,CxBgH/B,qEANe,CAHjB,8BAA+B,CAD/B,e/F0ptBF,C+F9otBI,2CACE,6GACA,Y/FgptBN,CuHlwtBM,mCACE,iCvHowtBR,CuHhwtBI,oCACE,qBvHkwtBN,CwHvztBA,mDAcE,+CAAgD,CAHhD,uBAAyB,CACzB,kCAAmC,CACnC,oB/BmByB,CUxBzB,+FACE,CqBCF,kBAAmB,CANnB,oBAAqB,CAErB,eAAgB,CADhB,eAAiB,CAIjB,uBAAwB,CASxB,iBAAkB,CAflB,qBxHs0tBF,CwHrztBE,yDACE,mBAAqB,CACrB,gBxHyztBJ,CwHrztBA,iBAEE,4CAA6C,CAD7C,qCxHyztBF,CwHrztBA,mBAEE,4CAA6C,CAD7C,qCxHyztBF,CwHrztBA,eAEE,2CAA4C,CAD5C,oCxHyztBF,CwHrztBA,qBACE,exHwztBF,CwHtztBE,4BAEE,2BAA4B,CAE5B,+CAAgD,CADhD,yBAA0B,CAF1B,kBxH2ztBJ,CwHnztBE,kCACE,8BAA+B,CAC/B,6CxHsztBJ,CwHjztBE,oCACE,8BAA+B,CAC/B,+CxHoztBJ,CwH/ytBE,uCACE,6BAA8B,CAC9B,kDxHkztBJ,CyHv3tBA,uBAGE,4CAA6C,CAF7C,YAAa,CACb,qBzH23tBF,Cc9ztBI,yB2G/DJ,uBAMI,ezH23tBF,CACF,CyHz3tBE,qDAGE,gCAAiC,CADjC,6CAA8C,CAD9C,iDAAkD,CAGlD,mBzH23tBJ,CyHx3tBE,uCAEE,eAAgB,CADhB,kBAAmB,CAEnB,SzH03tBJ,CyHx3tBI,4CAGE,iCAAkC,CAFlC,aAAc,CACd,gBzH23tBN,CyHx3tBM,kDAGE,4BAA6B,CAF7B,8BAA+B,CAC/B,oBzH23tBR,C0Hh5tBA,2BAYE,+CAAgD,CAHhD,wBACA,kCAAmC,CACnC,oBAdyB,CvBIzB,+FACE,CuBMF,kBAAmB,CANnB,oBAAqB,CAErB,eAAgB,CADhB,qBAAwB,CAIxB,uB1Hg6tBF,C0Ht5tBE,mDACE,e1Hy5tBJ,C0Hr5tBE,qEACE,kBAAmB,CACnB,mB1Hw5tBJ,C0Hp5tBE,kEAEE,e1Hw5tBJ,C0Hp5tBE,+DAQE,yCAA0C,CAL1C,qDAAsD,CAFtD,eAAiB,CACjB,8BAAkC,CAElC,iBAAkB,CAOlB,S1Hk5tBJ,C2Hz7tBE,6EACE,UAAW,CAEX,WAAY,CAEZ,MAAO,CAOP,mBAAoB,CARpB,iBAAkB,CAElB,KAAM,CAJN,UAAW,CAKX,U3H67tBJ,C0H35tBI,2EAKE,2BAA4B,CAG5B,0CAA2C,CAF3C,yBAA0B,CAF1B,WAAY,CAFZ,UAAY,CAKZ,mBAAoB,CAEpB,SAAU,CARV,iBAAkB,CAElB,U1Ho6tBN,C0H15tBI,mEACE,e1H65tBN,C0Hz5tBE,+CACE,uC1H45tBJ,C0H15tBI,mFACE,8C1H65tBN,C0H35tBM,+FACE,gCAAiC,CACjC,4C1H85tBR,C0Hz5tBE,2CACE,qC1H45tBJ,C0H15tBI,+EACE,4C1H65tBN,C0H35tBM,2FACE,8BAA+B,CAC/B,0C1H85tBR,C0Hz5tBE,2CACE,qC1H45tBJ,C0H15tBI,+EACE,4C1H65tBN,C0H35tBM,2FACE,8BAA+B,CAC/B,0C1H85tBR,C0Hz5tBE,yCACE,oC1H45tBJ,C0H15tBI,6EACE,2C1H65tBN,C0H35tBM,yFACE,6BAA8B,CAC9B,yC1H85tBR,C0Hz5tBE,uCACE,oC1H45tBJ,C0H15tBI,2EACE,2C1H65tBN,C0H35tBM,uFACE,6BAA8B,CAC9B,wC1H85tBR,C0Hz5tBE,qCACE,qC1H45tBJ,C0H15tBI,yEACE,4C1H65tBN,C0H35tBM,qFACE,8BAA+B,CAC/B,uC1H85tBR,C0Hz5tBE,mCACE,qC1H45tBJ,C0H15tBI,uEACE,4C1H65tBN,C0H35tBM,mFACE,8BAA+B,CAC/B,sC1H85tBR,C0Hz5tBE,+CACE,uC1H45tBJ,C0H15tBI,mFACE,8C1H65tBN,C0H35tBM,+FACE,gCAAiC,CACjC,4C1H85tBR,C0Hz5tBE,qCACE,kC1H45tBJ,C0H15tBI,yEACE,yC1H65tBN,C0H35tBM,qFACE,2BAA4B,CAC5B,uC1H85tBR,C0Hz5tBE,2CACE,qC1H45tBJ,C0H15tBI,+EACE,4C1H65tBN,C0H35tBM,2FACE,8BAA+B,CAC/B,0C1H85tBR,C0Hz5tBE,2DACE,uC1H45tBJ,C0H15tBI,+FACE,8C1H65tBN,C0H35tBM,2GACE,gCAAiC,CACjC,uC1H85tBR,C0Hr5tBE,2CAQE,yBALA,UAAW,CADX,WAAY,CAEZ,iBAAmB,CACnB,YAAa,CAJb,a1H65tBJ,C0Hl5tBI,8HAEE,uC1Hq5tBN,C0Hl5tBI,sHAEE,qC1Hq5tBN,C0Hl5tBI,gHAEE,oC1Hq5tBN,C0Hl5tBI,mKAGE,qC1Hq5tBN,C0Hl5tBI,0GAEE,kC1Hq5tBN,C0Hj5tBI,qFACE,aAAc,CACd,c1Ho5tBN,C0Hv4tBA,sDAOE,yCAA0C,CAC1C,oCAAqC,CACrC,oBAtRyB,CvBIzB,+FACE,CuB6QF,YAAa,CACb,qBAAsB,CAItB,oB1Hg5tBF,C0H54tBE,0GACE,gB1Hi5tBJ,C0H74tBE,8DACE,2C1Hk5tBJ,C0H94tBE,8FACE,iB1Hm5tBJ,C0Hj5tBI,0GAEE,gB1Hq5tBN,C0H74tBA,cAEE,yCAA0C,CAD1C,wCAAyC,CAEzC,oBArTyB,CAwTzB,iBAAmB,CACnB,S1Hi5tBF,C0H/4tBE,0BACE,mB1Hi5tBJ,C0H94tBE,8BAKE,+CAAgD,CAChD,0CAA2C,CAC3C,qDAAsD,CALtD,eAAgB,CAEhB,oBAAsB,CADtB,iBAAmB,CAFnB,iB1Hs5tBJ,C0H54tBE,+EAEE,e1H84tBJ,C0H34tBE,gBACE,iBAAkB,CAClB,kB1H64tBJ,C0Ht4tBA,SACE,YAAa,CACb,qB1H44tBF,C0Ht4tBA,YAEE,eAAgB,CADhB,Y1H64tBF,C4H3uuBA,iBACE,wBAAyB,CACzB,mBAAoB,CACpB,e5HmvuBF,C4HjvuBE,+BAGE,yCAA0C,CAF1C,uBAAwB,CACxB,kB5HovuBJ,C4H9uuBI,iCACE,iB5HmvuBN,C4HjvuBM,wCACE,iB5HmvuBR,C4H9uuBI,0CACE,gBAAiB,CACjB,eAAiB,CACjB,iB5HgvuBN,C4HruuBE,kCACE,kB5H2uuBJ,C4HtuuBA,KACE,4C5H0uuBF,C4HruuBA,0CAEE,iB5HyuuBF,C4H1tuBA,kHAIE,gC5HyuuBF,C4H/tuBA,UACE,kC5HkuuBF,C4H/tuBA,wCAEE,iCAAkC,CAClC,e5HkuuBF,C4H9tuBA,2BAEE,wC5HiuuBF,C4H9tuBA,eACE,uC5HiuuBF,C4H9tuBA,uBAIE,wCAAyC,CAFzC,+CAAgD,CADhD,4CAA6C,CAE7C,iB5HkuuBF,C4H9tuBA,4BASE,W5HytuBF,C4HvtuBE,2CACE,U5HytuBJ,C4H/suBA,oEAEE,kB5HotuBF,C4HltuBE,uEACE,gB5HotuBJ,C4HjtuBI,oFACE,Y5HmtuBN,C4H/suBE,kFACE,YAAa,CACb,2B5HituBJ,C4H9suBE,mKAIE,yCAA0C,CAD1C,mBAAqB,CADrB,gB5HktuBJ,C6Ht2uBA,+DAME,oBHTyB,CGUzB,kBAAmB,CAJnB,YAAa,CACb,qBAAsB,CACtB,W7H+2uBF,C6Hx2uBA,0BACE,wCAAyC,CACzC,oB7H22uBF,C6Hx2uBE,iDAEE,+CAAgD,CAEhD,cAAe,CACf,0CAA2C,CAJ3C,QAAS,CAET,a7H42uBJ,C6Hx2uBI,8DACE,iB7H02uBN,C6Hp2uBE,iDAEE,eAAgB,CADhB,Q7Hu2uBJ,C6Hp2uBI,qDACE,WAAY,CACZ,e7Hs2uBN,C6H91uBA,aAEE,yCAA0C,CAC1C,wCAAyC,CACzC,oBAAsB,CAHtB,oB7Hu2uBF,C6Hj2uBA,OACE,wC7Ho2uBF,C6Hl2uBE,aACE,iC7Ho2uBJ,C6H/1uBA,aACE,oB7Hk2uBF,C8Hv6uBA,8CAIE,aAAc,CACd,gBAAiB,CACjB,iB9H06uBF,C8Ht6uBE,oBAGE,iBAAkB,CADlB,iB9Hy6uBJ,C8Hp6uBE,0BACE,kB9Hs6uBJ,C8Hn6uBE,kBAGE,iCAAkC,CAFlC,0CAA2C,CAC3C,0CAA2C,CAE3C,gBAAiB,CACjB,iBAAkB,CAClB,gBAAkB,CAClB,iB9Hq6uBJ,C8Hn6uBI,+BAEE,e9Ho6uBN,C8Hj6uBI,oBAEE,oBAAqB,CADrB,gB9Ho6uBN,C8Hh6uBI,8BAEE,gBAAiB,CACjB,iBAAkB,CAFlB,iB9Ho6uBN,C+H38uBA,+CACE,W/H88uBF,C+H38uBA,8CACE,W/H88uBF,C+H18uBA,qBAEE,eAAgB,CADhB,oB/H88uBF,C+Hz8uBA,eACE,mB/H48uBF,C+H18uBE,0BACE,kB/H48uBJ,C+Hz8uBE,uDAEE,e/H28uBJ,C+Hx8uBE,sBACE,wC/H08uBJ,CgIl+uBA,+EAIE,gBhIw+uBF,CiIj/uBA,GAEE,kBAAmB,CACnB,gBAAiB,CAFjB,cjIs/uBF,CiIj/uBA,MAEE,yBjIo/uBF,CiIl/uBE,wCACE,oBAAsB,CACtB,iBjIq/uBJ,CkIjgvBA,WAuBE,yCAA0C,CApB1C,yDAA2D,CAC3D,oBRDyB,CQDzB,iCAAkC,CADlC,WAAY,CAIZ,iBlIqgvBF,CkIngvBE,aACE,gClIqgvBJ,CkIjgvBE,uBACE,QlImgvBJ,CkI//uBE,wBACE,elIigvBJ,C2HvgvBE,kBACE,UAAW,CAEX,WAAY,CAEZ,MAAO,CAOP,mBAAoB,CARpB,iBAAkB,CAElB,KAAM,CAJN,U3H+gvBJ,CkIngvBE,aACE,wClIqgvBJ,CkIjgvBE,kBACE,UlImgvBJ,CmIhivBA,cAWE,yCAA0C,CAV1C,sCAAuC,CAGvC,iBAAkB,CAFlB,aAAc,CACd,eAAgB,CAGhB,eAAgB,CADhB,iBAAkB,CAElB,iBnIuivBF,C2HrivBE,qBACE,UAAW,CAEX,WAAY,CAEZ,MAAO,CAOP,mBAAoB,CARpB,iBAAkB,CAElB,KAAM,CAJN,UAAW,CAKX,U3HwivBJ,CmI1ivBA,4BAKE,iCAAkC,CAJlC,gCAAiC,CACjC,yBAA0B,CAC1B,cAAgB,CAChB,enI8ivBF,CoIhkvBA,MAEE,gBAAiB,CACjB,iBpIqkvBF,CoInkvBE,kBACE,cpIqkvBJ,CoIlkvBE,iBACE,apIokvBJ,CoI9jvBA,cAEE,gBAAiB,CACjB,iCAAkC,CAFlC,iBpImkvBF,CoI3jvBE,0BACE,epI+jvBJ,CoI5jvBE,4BACE,gBpI+jvBJ,CoI5jvBE,8BACE,iBpI+jvBJ,CoI1jvBA,OAGE,yBAA0B,CAC1B;;GAAA,CjCrBA,oDAAqD,CADrD,4BnGqlvBF,CmGhlvBI,oDAEE,yDnGolvBN,CmG/kvBI,gBACE,kDAAmD,CACnD,gDnGilvBN,CmG3kvBM,+BACE,wDnG6kvBR,CmG1kvBM,8BACE,yDnG4kvBR,CoI1kvBA,gCAGE,epI2kvBF,CqI9nvBE,2BACE,eAAgB,CAChB,erIoovBJ,CqIjovBE,oBACE,crImovBJ,CqIhovBE,sCACE,eAAgB,CAChB,kBrIkovBJ,CqIhovBI,wCAEE,eAAgB,CADhB,erImovBN,CqI/nvBI,yCACE,eAAgB,CAChB,0BrIiovBN,CqI5nvBE,+BACE,erI8nvBJ,CqItnvBE,oDACE,eAAgB,CAChB,crI0nvBJ,CsI1pvBA,mBAEE,kBAAmB,CACnB,YAAa,CACb,cAAe,CAGf,etIiqvBF,CsI7pvBA,UACE,mBtIgqvBF,CsI5pvBA,SAIE,0BAA2B,CAH3B,QtIgqvBF,CsI1pvBE,gCAEE,aAAc,CADd,iBtI6pvBJ,CsIxpvBE,uBACE,WAAY,CAEZ,aAAc,CADd,oBtI2pvBJ,CsItpvBI,yCACE,QtIwpvBN,CuI5rvBE,8FAEE,6CAA8C,CAG9C,gBvIqsvBJ,CuIjsvBI,sGACE,gCvIqsvBN,CuIjsvBE,uBASE,YAAa,CACb,qBAAsB,CACtB,QAAU,CATV,eAAgB,CAUhB,eAAgB,CANhB,iBAAkB,CAHlB,cvIusvBJ,CuI3rvBI,mCACE,kBAAmB,CACnB,SvI6rvBN,CuI1rvBM,sCAGE,kBAAmB,CADnB,YvI4rvBR,CuIlrvBA,kBAEE,YAAa,CADb,eAAgB,CAEhB,cvIwrvBF,CuIrrvBE,uBAEE,YAAa,CACb,avIsrvBJ,CuInrvBI,yBAEE,kBAAmB,CADnB,YAAa,CAGb,QAAS,CACT,kBAAmB,CAFnB,gBvIurvBN,CuInrvBM,kCACE,gCvIqrvBR,CuI9qvBI,iEACE,kCvIgrvBN,CuI1qvBE,qCACE,YvI4qvBJ,CuIxqvBE,mCACE,gBAAiB,CACjB,gBvI0qvBJ,CuIxqvBI,kEACE,mCvI0qvBN,CuIlqvBA,6BAEE,cvIwqvBF,CuItqvBE,qDACE,evIyqvBJ,CuItqvBI,mFACE,YAAa,CACb,kBAAmB,CAGnB,gBAAkB,CAFlB,QAAS,CACT,eAAgB,CAEhB,cvIyqvBN,CuIrqvBI,yFAEE,iBAAkB,CADlB,YvIyqvBN,CuItqvBM,6FACE,evIyqvBR,CuIpqvBI,2FACE,mBvIuqvBN,CwIxyvBA,oBAGE,iBAAkB,CAClB,kBAAmB,CACnB,eAAgB,CAJhB,kBAAmB,CACnB,mBAAoB,CAIpB,WxI8yvBF,CwI3yvBA,KACE,iExI8yvBF,CyI1zvBA,6BAGE,kBAAmB,CASnB,yCAA0C,CAD1C,WAAY,CATZ,YAAa,CAEb,sBzIk0vBF,CyI/zvBE,2CACE,4BzIi0vBJ,CyI1zvBI,iDAEE,wCAAyC,CADzC,2BzI6zvBN,CyIvzvBE,mDAEE,yCAA0C,CAD1C,2BzI0zvBJ,CyItzvBE,mCAEE,SzIuzvBJ,CyIpzvBE,2CACE,8CzIszvBJ,CyIlzvBA,kCAIE,czIkzvBF,C0I91vBE,8WAQE,gC1Ii2vBJ,C0I91vBE,6EAEE,4CAA6C,CAC7C,wCAAyC,CACzC,iBAAkB,CAClB,gCAAiC,CACjC,cAAe,CACf,gB1Ig2vBJ,C2Ix2vBA,kCAGE,sBAAuB,CADvB,WAAY,CAEZ,e3Io3vBF,C2Ij3vBE,6FAEE,oB3Im3vBJ,C2Ih3vBE,gDACE,yC3Ik3vBJ,C2I92vBE,6FAEE,oB3Ig3vBJ,C2I92vBI,qGACE,Y3Ii3vBN,C4Ih5vBA,sDAEE,mBAAoB,CADpB,+C5Is5vBF,C6Ir5vBA,iBACE,uB7I25vBF,C6Iv5vBA,gCAEE,a7Iy5vBF,C8I30vBA,MAIM,k+B9I83vBN,C8I9ovBA,uBAzDY,4BAFA,+BAEA,8BAFA,iCAEA,2BAFA,8BAEA,yBAFA,4BAEA,4BAFA,+BAEA,4BAFA,+BAEA,+CAFA,qDAEA,2BAFA,8BAEA,qUAFA,uCAEA,6JAWR,8CAA+C,CAC/C,yCAA0C,CAC1C,iDAAkD,CAClD,uDAAwD,CACxD,qDAAsD,CACtD,kEAAmE,CACnE,2D9ImuvBJ,C8I9tvBI,+EAEE,sB9I2uvBN,C8IlsvBE,uBAnEU,gEAEA,2BAFA,wHAEA,+CAFA,mFAEA,qUAFA,uCAEA,6JAWR,8CAA+C,CAC/C,yCAA0C,CAC1C,iDAAkD,CAClD,uDAAwD,CACxD,qDAAsD,CACtD,kEAAmE,CACnE,2DAA4D,CAmD5D,kB9IsuvBJ,C8IpxvBI,+EAEE,sB9IiyvBN,C8IxvvBE,sBAnEU,gEAEA,2BAFA,wHAEA,+CAFA,mFAEA,wTAFA,uCAEA,mKAWR,8CAA+C,CAC/C,yCAA0C,CAC1C,iDAAkD,CAClD,uDAAwD,CACxD,qDAAsD,CACtD,kEAAmE,CACnE,2DAA4D,CAmD5D,iB9IiyvBJ,C8I10vBI,+EAEE,sB9Iu1vBN,C8Ih1vBI,sDACE,mC9Im1vBN,C8I50vBI,kEACE,sBACA,oB9I+0vBN,C8I30vBI,qCACE,+B9I80vBN,C8IvzvBE,mBACE,8B9I0zvBJ,C8I3zvBE,qBACE,gC9I8zvBJ,C8I/zvBE,kBACE,6B9Ik0vBJ,C8In0vBE,gBACE,2B9Is0vBJ,C8Iv0vBE,mBACE,8B9I00vBJ,C8I30vBE,mBACE,8B9I80vBJ,C8I/0vBE,qBACE,gC9Ik1vBJ,C8In1vBE,kBACE,6B9Is1vBJ,C8Iv1vBE,qBACE,gC9I01vBJ,C8I31vBE,sBACE,iC9I81vBJ,C8I/1vBE,kBACE,6B9Ik2vBJ,C8In2vBE,kBACE,6B9Is2vBJ,C8Iv2vBE,wBACE,mC9I02vBJ,C8I32vBE,4BACE,uC9I82vBJ,C8I/2vBE,uBACE,kC9Ik3vBJ,C8In3vBE,6BACE,wC9Is3vBJ,C8Iv3vBE,kBACE,6B9I03vBJ,C8I33vBE,iBACE,4B9I83vBJ,C8I/3vBE,2BACE,sC9Ik4vBJ,C8In4vBE,8BACE,yC9Is4vBJ,C8Iv4vBE,sBACE,iC9I04vBJ,C8I34vBE,yBACE,oC9I84vBJ,C8I/4vBE,mBACE,8B9Ik5vBJ,C8In5vBE,sBACE,iC9Is5vBJ,C+IzpwBE,uBAzBA,4BAIA,8BAOA,sCAGA,4CACA,sDACA,gEAGA,wCAkBY,8BAKA,mCA1CZ,8BAIA,gCAOA,wCAGA,gDACA,0DACA,oEAGA,0CAkBY,gCAKA,qCA1CZ,4BAIA,8BAOA,sCAGA,4CACA,sDACA,gEAGA,wCAkBY,8BAKA,mCA1CZ,yBAIA,2BAOA,mCAGA,sCACA,gDACA,0DAGA,qCAkBY,2BAKA,gCA1CZ,4BAIA,iCAOA,sCAGA,4CACA,sDACA,gEAGA,2CAkBY,8BAKA,mCA1CZ,2BAIA,6BAOA,qCAGA,0CACA,oDACA,8DAGA,uCAkBY,6BAKA,kCA1CZ,0BAIA,+BAOA,oCAGA,wCACA,kDACA,4DAGA,yCAkBY,4BAKA,iCA1CZ,0BAIA,4BAOA,oCAGA,wCACA,kDACA,4DAGA,sCAkBY,4BAKA,8BA1CZ,yBAIA,2BAOA,mCAGA,sCACA,gDACA,0DAGA,qCAnBA,0BAIA,4BAOA,oCAGA,wCACA,kDACA,4DAGA,sCAnBA,uBAIA,+BAOA,oCAGA,wCACA,kDACA,4DAGA,wC/ImxwBF,C+I7wwBE,sBAzBA,4BAIA,iCAOA,sCAGA,4CACA,sDACA,gEAGA,2CAkBY,8BAKA,gCA1CZ,8BAIA,mCAOA,wCAGA,gDACA,0DACA,oEAGA,0CAkBY,gCAKA,kCA1CZ,4BAIA,iCAOA,sCAGA,4CACA,sDACA,gEAGA,2CAkBY,8BAKA,gCA1CZ,yBAIA,8BAOA,mCAGA,sCACA,gDACA,0DAGA,wCAkBY,2BAKA,6BA1CZ,4BAIA,iCAOA,sCAGA,4CACA,sDACA,gEAGA,2CAkBY,8BAKA,gCA1CZ,2BAIA,gCAOA,qCAGA,0CACA,oDACA,8DAGA,0CAkBY,6BAKA,+BA1CZ,0BAIA,+BAOA,oCAGA,wCACA,kDACA,4DAGA,yCAkBY,4BAKA,iCA1CZ,0BAIA,+BAOA,oCAGA,wCACA,kDACA,4DAGA,yCAkBY,4BAKA,8BA1CZ,yBAIA,2BAOA,mCAGA,sCACA,gDACA,0DAGA,qCAnBA,0BAIA,4BAOA,oCAGA,wCACA,kDACA,4DAGA,sCAnBA,uBAIA,+BAOA,oCAGA,wCACA,kDACA,4DAGA,wC/Im3wBF,C+Iz0wBE,6CACE,8C/Ig1wBJ,C+Ix0wBE,oJ5ClIA,+FnGo9wBF,C+It0wBA,qBACE,wC/I40wBF,C+I10wBE,qCACE,+CAAgD,CAChD,+C/I40wBJ,C+Iz0wBE,qCAEE,4C/I20wBJ,C+Ix0wBE,wEAJE,+C/I+0wBJ,C+It0wBE,sDAEE,Y/Iu0wBJ,C+Ip0wBI,4DAEE,oBAAsB,CADtB,8C/Iu0wBN,C+Il0wBE,yCACE,wC/Io0wBJ,C+IzzwBI,4CAOE,+CAAgD,CALhD,2EAIA,iCAAoC,CALpC,6BAA8B,CAI9B,8BAAiC,CAIjC,8BAA+B,CAD/B,6B/I8zwBN,C+I1zwBI,kDAGE,2CAA4C,CAF5C,4CAA8C,CAC9C,qBAAuB,CAEvB,iC/I4zwBN,C+IxzwBI,uEACE,wBAAyB,CACzB,gCAAiC,CACjC,8BAA+B,CAC/B,iD/I0zwBN,C+IrzwBE,8BAIE,yCAA0C,CAF1C,gCAAkC,CAClC,iCAAoC,CAFpC,iCAAkC,CAMlC,gBAAiB,CADjB,wBAA0B,CAD1B,e/IyzwBJ,C+I/ywBE,4FALI,+C/I8zwBN,C+IzzwBE,wCACE,6CAA+C,CAC/C,sBAAwB,CACxB,gBAAiB,CACjB,e/IqzwBJ,C+I5ywBA,oBAEE,kBAAoB,C5ClPpB,+FnGoixBF,C+I5ywBE,2CAYE,2CAA4C,CAC5C,mDAAoD,CAVpD,kBAAmB,CAqCnB,0DAA4D,CAvC5D,kBAAoB,CA0CpB,8DAAiE,CAzCjE,YAAa,CAGb,eAAgB,CAEhB,oBAAsB,CADtB,iBAAmB,CAFnB,iB/IqzwBJ,C+I1ywBI,kEAGE,mDAAoD,CAFpD,Q/I6ywBN,C+IlywBQ,yIACE,gDACA,qD/IwywBV,C+IrywBM,8DAEE,+C/IsywBR,C+I7ywBQ,6IACE,kDACA,uD/ImzwBV,C+IhzwBM,gEAEE,iD/IizwBR,C+IxzwBQ,yIACE,gDACA,qD/I8zwBV,C+I3zwBM,8DAEE,+C/I4zwBR,C+In0wBQ,mIACE,6CACA,kD/Iy0wBV,C+It0wBM,2DAEE,4C/Iu0wBR,C+I90wBQ,yIACE,gDACA,qD/Io1wBV,C+Ij1wBM,8DAEE,+C/Ik1wBR,C+Iz1wBQ,uIACE,+CACA,oD/I+1wBV,C+I51wBM,6DAEE,8C/I61wBR,C+Ip2wBQ,qIACE,8CACA,mD/I02wBV,C+Iv2wBM,4DAEE,6C/Iw2wBR,C+I/2wBQ,qIACE,8CACA,mD/Iq3wBV,C+Il3wBM,4DAEE,6C/Im3wBR,C+I13wBQ,mIACE,6CACA,kD/Ig4wBV,C+I73wBM,2DAEE,4C/I83wBR,C+Ir4wBQ,qIACE,8CACA,mD/I24wBV,C+Ix4wBM,4DAEE,6C/Iy4wBR,C+Ih5wBQ,qIACE,8CACA,mD/Is5wBV,C+In5wBM,4DAEE,6C/Io5wBR,C2HzqxBE,kDACE,UAAW,CAEX,WAAY,CAEZ,MAAO,CAOP,mBAAoB,CARpB,iBAAkB,CAElB,KAAM,CAJN,UAAW,CAKX,U3H4qxBJ,C+Il5wBI,kEAGE,+CAAgD,CADhD,4CAA8C,CAD9C,8D/Is5wBN,C+Ij5wBI,gEAEE,kBAAmB,CACnB,4CAA8C,CAF9C,mB/Iq5wBN,C+Ij5wBM,oEACE,S/Im5wBR,C+I94wBI,sHAEE,S/Ig5wBN,CmG3oxBE,iDACE,qBnG6oxBJ,CmG3oxBI,uEACE,sBnG6oxBN,C+Ih5wBI,+DAGE,sBtIpTa,CsIkTb,8CtItSe,CsIuSf,sB/Im5wBN,C+I94wBE,qEACE,2BAA4B,CAC5B,4B/Ig5wBJ,C+Iv4wBE,aACE,iBAAkB,CAClB,gC/I84wBJ,C+I54wBI,mBhDjSF,iCAAkC,CA7BlC,yBAA0B,CAkBxB,6BAA8B,CAf9B,2CAvBuB,CAmCvB,iDAtB8B,CgDwU5B,6BAA8B,ChD1ThC,6B/F4sxBJ,C+I54wBM,2BACE,uD/I84wBR,C+I/4wBM,6BACE,yD/Ii5wBR,C+Il5wBM,2BACE,uD/Io5wBR,C+Ir5wBM,wBACE,oD/Iu5wBR,C+Ix5wBM,2BACE,uD/I05wBR,C+I35wBM,0BACE,sD/I65wBR,C+I95wBM,yBACE,qD/Ig6wBR,C+Ij6wBM,yBACE,qD/Im6wBR,C+Ip6wBM,wBACE,oD/Is6wBR,C+Iv6wBM,yBACE,qD/Iy6wBR,C+I16wBM,yBACE,qD/I46wBR,C+Ip6wBM,8EASE,mDACA,+CAPA,sBtI5VW,CsI0VX,+CACA,sB/I06wBR,C+I56wBM,kFASE,qDACA,iDAPA,sBtI5VW,CsI0VX,iDACA,sB/Ik7wBR,C+Ip7wBM,8EASE,mDACA,+CAPA,sBtI5VW,CsI0VX,+CACA,sB/I07wBR,C+I57wBM,wEASE,gDACA,4CAPA,sBtI5VW,CsI0VX,4CACA,sB/Ik8wBR,C+Ip8wBM,8EASE,mDACA,+CAPA,sBtI5VW,CsI0VX,+CACA,sB/I08wBR,C+I58wBM,4EASE,kDACA,8CAPA,sBtI5VW,CsI0VX,8CACA,sB/Ik9wBR,C+Ip9wBM,0EASE,iDACA,6CAPA,sBtI5VW,CsI0VX,6CACA,sB/I09wBR,C+I59wBM,0EASE,iDACA,6CAPA,sBtI5VW,CsI0VX,6CACA,sB/Ik+wBR,C+Ip+wBM,wEASE,gDACA,4CAPA,sBtI5VW,CsI0VX,4CACA,sB/I0+wBR,C+I5+wBM,0EASE,iDACA,6CAPA,sBtI5VW,CsI0VX,6CACA,sB/Ik/wBR,C+Ip/wBM,0EASE,iDACA,6CAPA,sBtI5VW,CsI0VX,6CACA,sB/I0/wBR,CgJj1xBI,yEACE,SAAU,CACV,ShJ21xBN,CgJx1xBI,mEACE,UhJ01xBN,CmGpzxBE,gEACE,qBnGszxBJ,CmGpzxBI,sFACE,sBnGszxBN,CgJr1xBM,gEACE,ehJu1xBR,CgJn1xBI,6DACE,ahJq1xBN,CgJ53xBI,wGACE,uBAAwB,CACxB,ehJ83xBN,CgJz3xBI,mFACE,wBAAyB,CACzB,ehJ23xBN,CgJ50xBI,4DAGE,gBhJq2xBN,CgJ50xBM,qFAUE,sBvIlFW,CuIyEX,UAAW,CAMX,WAAY,CAFZ,YAGA,8CvIpEa,CuIqEb,sBvIxEY,CuIqEZ,wBhJ+0xBR,CgJv0xBM,yGACE,2BAA4B,CAC5B,4BhJy0xBR,CgJj0xBI,2CACE,8ChJm0xBN,CgJh7xBI,iEACE,uBAAwB,CACxB,ehJk7xBN,CgJp6xBI,0DACE,SAAU,CACV,ShJs6xBN,CgJn6xBI,oDACE,UhJq6xBN,CmG/3xBE,iDACE,qBnGi4xBJ,CmG/3xBI,uEACE,sBnGi4xBN,CgJ90xBM,iDACE,ehJg1xBR,CgJ70xBM,yDACE,sBhJ+0xBR,CgJh8xBI,+DACE,wBAAyB,CACzB,ehJk8xBN,CgJ30xBM,wDACE,2BAA4B,CAC5B,4BhJ60xBR,CiJ98xBA,mDAME,ejJu9xBF,CiJr9xBE,+DACE,iBjJw9xBJ,CiJ/8xBM,2I9CHJ,oDAAqD,CADrD,4BnGy9xBF,CmGp9xBI,4lBAEE,yDnG49xBN,CmGv9xBI,6JACE,kDAAmD,CACnD,gDnG09xBN,CmGp9xBM,2LACE,wDnGu9xBR,CmGp9xBM,yLACE,yDnGu9xBR,CiJx+xBU,yKACE,oDjJ2+xBZ,CiJn+xBM,2K9CvBJ,2CAA4C,CAC5C,oBAAsB,CAFtB,oCAAqC,CAGrC,anG+/xBF,CiJp+xBM,gEACE,wCjJs+xBR,CiJ/9xBA,qBACE,kBjJk+xBF,CiJx9xBE,sDAEE,ajJ09xBJ,CiJv9xBE,qDAEE,ajJw9xBJ,CiJj9xBE,uDAEE,ajJm9xBJ,CiJh9xBE,sDAEE,ajJi9xBJ,CiJz8xBA,wBACE,cAAe,CACf,ejJ+8xBF,CiJ18xBE,4C9C7EA,oDAAqD,CADrD,4BnG6hyBF,CmGxhyBI,wMAEE,yDnG4hyBN,CmGvhyBI,qDACE,kDAAmD,CACnD,gDnGyhyBN,CmGnhyBM,oEACE,wDnGqhyBR,CmGlhyBM,mEACE,yDnGohyBR,CiJ39xBM,2DACE,oDjJ69xBR,CiJv9xBI,gN9C/FF,2CAA4C,CAC5C,oBAAsB,CAFtB,oCAAqC,CAGrC,anG4jyBF,CiJp9xBE,2BACE,YAAa,CACb,qBAAsB,CACtB,uBjJu9xBJ,CiJp9xBE,kDAEE,oBjJs9xBJ,CiJl9xBI,6BACE,iBjJo9xBN,CkJplyBA,uEACE,eAAgB,CAChB,SlJgmyBF,CmJlmyBI,uFACE,mCAAoC,CACpC,enJ6myBN,CmJ1myBI,yDACE,iCnJ4myBN,CmJxmyBE,+BAEE,eAAgB,CADhB,QnJ2myBJ,CmJxmyBI,kCACE,qBAAsB,CAGtB,gDAAiD,CADjD,aAAc,CADd,cnJ4myBN,CmJvmyBM,oCACE,enJymyBR,CmJtmyBM,0FAEE,gCAAiC,CACjC,enJwmyBR,CmJrmyBQ,4GAGE,iCAAkC,CAFlC,WAAY,CACZ,kBnJymyBV","sources":["webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_type.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/pydata-sphinx-theme.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_banner.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_root.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/vendor/_rfs.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_color-mode.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_reboot.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_variables.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_border-radius.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_lists.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/variables/_bootstrap.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_images.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_image.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_containers.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_container.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_breakpoints.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_grid.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_grid.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_tables.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_table-variants.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/forms/_labels.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/forms/_form-text.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/forms/_form-control.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_transition.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_gradients.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/forms/_form-select.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/forms/_form-check.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/forms/_form-range.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/forms/_floating-labels.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/forms/_input-group.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_forms.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_buttons.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_buttons.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_transitions.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_dropdown.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_caret.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_button-group.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_nav.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_navbar.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_card.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_accordion.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_breadcrumb.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_pagination.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_pagination.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_badge.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_alert.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_progress.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_list-group.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_close.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_toasts.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_modal.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_backdrop.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_tooltip.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_reset-text.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_popover.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_carousel.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_clearfix.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_spinners.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_offcanvas.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/_placeholders.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/helpers/_color-bg.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/helpers/_colored-links.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/helpers/_focus-ring.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/helpers/_icon-link.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/helpers/_ratio.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/helpers/_position.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/helpers/_stacks.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/helpers/_visually-hidden.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_visually-hidden.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/helpers/_stretched-link.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/helpers/_text-truncation.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_text-truncate.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/helpers/_vr.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/mixins/_utilities.scss","webpack://pydata_sphinx_theme/./node_modules/bootstrap/scss/utilities/_api.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/fontawesome.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/_core.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/_sizing.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/_mixins.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/_fixed-width.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/_list.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/_bordered-pulled.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/_animated.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/_rotated-flipped.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/_stacked.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/_variables.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/_icons.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/solid.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/regular.scss","webpack://pydata_sphinx_theme/./node_modules/@fortawesome/fontawesome-free/scss/brands.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/variables/_layout.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/variables/_fonts.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/variables/_icons.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/variables/_admonitions.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/variables/_versionmodified.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/base/_base.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/abstracts/_links.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/sections/_skip-link.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/sections/_container.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/sections/_announcement.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/abstracts/_mixins.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/sections/_article.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/sections/_footer.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/sections/_footer-article.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/sections/_footer-content.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/sections/_header.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/sections/_header-article.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/sections/_sidebar-primary.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/sections/_sidebar-secondary.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/sections/_sidebar-toggle.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/components/_breadcrumbs.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/components/_icon-links.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/components/header/_header-logo.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/components/_navbar-links.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/components/_page-toc.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/components/_prev-next.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/components/_search.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/components/_searchbox.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/components/_switcher-theme.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/components/_switcher-version.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/components/_toc-inpage.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/components/_versionmodified.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/components/_indices.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/content/_admonitions.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/abstracts/_color.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/content/_api.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/content/_code.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/content/_figures.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/content/_footnotes.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/content/_hacks.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/content/_lists.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/content/_quotes.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/content/_spans.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/content/_tables.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/content/_toctree.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/content/_math.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/extensions/_ablog.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/extensions/_bootstrap.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/extensions/_copybutton.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/extensions/_ethical-ads.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/extensions/_execution.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/extensions/_graphviz.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/extensions/_pydata.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/variables/_color.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/extensions/_sphinx_design.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/extensions/_togglebutton.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/extensions/_notebooks.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/extensions/_leaflet.scss","webpack://pydata_sphinx_theme/./src/pydata_sphinx_theme/assets/styles/pages/_search.scss"],"sourcesContent":["//\n// Headings\n//\n.h1 {\n  @extend h1;\n}\n\n.h2 {\n  @extend h2;\n}\n\n.h3 {\n  @extend h3;\n}\n\n.h4 {\n  @extend h4;\n}\n\n.h5 {\n  @extend h5;\n}\n\n.h6 {\n  @extend h6;\n}\n\n\n.lead {\n  @include font-size($lead-font-size);\n  font-weight: $lead-font-weight;\n}\n\n// Type display classes\n@each $display, $font-size in $display-font-sizes {\n  .display-#{$display} {\n    @include font-size($font-size);\n    font-family: $display-font-family;\n    font-style: $display-font-style;\n    font-weight: $display-font-weight;\n    line-height: $display-line-height;\n  }\n}\n\n//\n// Emphasis\n//\n.small {\n  @extend small;\n}\n\n.mark {\n  @extend mark;\n}\n\n//\n// Lists\n//\n\n.list-unstyled {\n  @include list-unstyled();\n}\n\n// Inline turns list items into inline-block\n.list-inline {\n  @include list-unstyled();\n}\n.list-inline-item {\n  display: inline-block;\n\n  &:not(:last-child) {\n    margin-right: $list-inline-padding;\n  }\n}\n\n\n//\n// Misc\n//\n\n// Builds on `abbr`\n.initialism {\n  @include font-size($initialism-font-size);\n  text-transform: uppercase;\n}\n\n// Blockquotes\n.blockquote {\n  margin-bottom: $blockquote-margin-y;\n  @include font-size($blockquote-font-size);\n\n  > :last-child {\n    margin-bottom: 0;\n  }\n}\n\n.blockquote-footer {\n  margin-top: -$blockquote-margin-y;\n  margin-bottom: $blockquote-margin-y;\n  @include font-size($blockquote-footer-font-size);\n  color: $blockquote-footer-color;\n\n  &::before {\n    content: \"\\2014\\00A0\"; // em dash, nbsp\n  }\n}\n","@charset \"UTF-8\";\n/*!\n * Bootstrap  v5.3.3 (https://getbootstrap.com/)\n * Copyright 2011-2024 The Bootstrap Authors\n * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n */\n:root,\n[data-bs-theme=light] {\n  --bs-blue: #0d6efd;\n  --bs-indigo: #6610f2;\n  --bs-purple: #6f42c1;\n  --bs-pink: #d63384;\n  --bs-red: #dc3545;\n  --bs-orange: #fd7e14;\n  --bs-yellow: #ffc107;\n  --bs-green: #198754;\n  --bs-teal: #20c997;\n  --bs-cyan: #0dcaf0;\n  --bs-black: #000;\n  --bs-white: #fff;\n  --bs-gray: #6c757d;\n  --bs-gray-dark: #343a40;\n  --bs-gray-100: #f8f9fa;\n  --bs-gray-200: #e9ecef;\n  --bs-gray-300: #dee2e6;\n  --bs-gray-400: #ced4da;\n  --bs-gray-500: #adb5bd;\n  --bs-gray-600: #6c757d;\n  --bs-gray-700: #495057;\n  --bs-gray-800: #343a40;\n  --bs-gray-900: #212529;\n  --bs-primary: #0d6efd;\n  --bs-secondary: #6c757d;\n  --bs-success: #198754;\n  --bs-info: #0dcaf0;\n  --bs-warning: #ffc107;\n  --bs-danger: #dc3545;\n  --bs-light: #f8f9fa;\n  --bs-dark: #212529;\n  --bs-primary-rgb: 13, 110, 253;\n  --bs-secondary-rgb: 108, 117, 125;\n  --bs-success-rgb: 25, 135, 84;\n  --bs-info-rgb: 13, 202, 240;\n  --bs-warning-rgb: 255, 193, 7;\n  --bs-danger-rgb: 220, 53, 69;\n  --bs-light-rgb: 248, 249, 250;\n  --bs-dark-rgb: 33, 37, 41;\n  --bs-primary-text-emphasis: #052c65;\n  --bs-secondary-text-emphasis: #2b2f32;\n  --bs-success-text-emphasis: #0a3622;\n  --bs-info-text-emphasis: #055160;\n  --bs-warning-text-emphasis: #664d03;\n  --bs-danger-text-emphasis: #58151c;\n  --bs-light-text-emphasis: #495057;\n  --bs-dark-text-emphasis: #495057;\n  --bs-primary-bg-subtle: #cfe2ff;\n  --bs-secondary-bg-subtle: #e2e3e5;\n  --bs-success-bg-subtle: #d1e7dd;\n  --bs-info-bg-subtle: #cff4fc;\n  --bs-warning-bg-subtle: #fff3cd;\n  --bs-danger-bg-subtle: #f8d7da;\n  --bs-light-bg-subtle: #fcfcfd;\n  --bs-dark-bg-subtle: #ced4da;\n  --bs-primary-border-subtle: #9ec5fe;\n  --bs-secondary-border-subtle: #c4c8cb;\n  --bs-success-border-subtle: #a3cfbb;\n  --bs-info-border-subtle: #9eeaf9;\n  --bs-warning-border-subtle: #ffe69c;\n  --bs-danger-border-subtle: #f1aeb5;\n  --bs-light-border-subtle: #e9ecef;\n  --bs-dark-border-subtle: #adb5bd;\n  --bs-white-rgb: 255, 255, 255;\n  --bs-black-rgb: 0, 0, 0;\n  --bs-font-sans-serif: system-ui, -apple-system, \"Segoe UI\", Roboto, \"Helvetica Neue\", \"Noto Sans\", \"Liberation Sans\", Arial, sans-serif, \"Apple Color Emoji\", \"Segoe UI Emoji\", \"Segoe UI Symbol\", \"Noto Color Emoji\";\n  --bs-font-monospace: SFMono-Regular, Menlo, Monaco, Consolas, \"Liberation Mono\", \"Courier New\", monospace;\n  --bs-gradient: linear-gradient(180deg, rgba(255, 255, 255, 0.15), rgba(255, 255, 255, 0));\n  --bs-body-font-family: var(--bs-font-sans-serif);\n  --bs-body-font-size: 1rem;\n  --bs-body-font-weight: 400;\n  --bs-body-line-height: 1.5;\n  --bs-body-color: #212529;\n  --bs-body-color-rgb: 33, 37, 41;\n  --bs-body-bg: #fff;\n  --bs-body-bg-rgb: 255, 255, 255;\n  --bs-emphasis-color: #000;\n  --bs-emphasis-color-rgb: 0, 0, 0;\n  --bs-secondary-color: rgba(33, 37, 41, 0.75);\n  --bs-secondary-color-rgb: 33, 37, 41;\n  --bs-secondary-bg: #e9ecef;\n  --bs-secondary-bg-rgb: 233, 236, 239;\n  --bs-tertiary-color: rgba(33, 37, 41, 0.5);\n  --bs-tertiary-color-rgb: 33, 37, 41;\n  --bs-tertiary-bg: #f8f9fa;\n  --bs-tertiary-bg-rgb: 248, 249, 250;\n  --bs-heading-color: inherit;\n  --bs-link-color: #0d6efd;\n  --bs-link-color-rgb: 13, 110, 253;\n  --bs-link-decoration: underline;\n  --bs-link-hover-color: #0a58ca;\n  --bs-link-hover-color-rgb: 10, 88, 202;\n  --bs-code-color: #d63384;\n  --bs-highlight-color: #212529;\n  --bs-highlight-bg: #fff3cd;\n  --bs-border-width: 1px;\n  --bs-border-style: solid;\n  --bs-border-color: #dee2e6;\n  --bs-border-color-translucent: rgba(0, 0, 0, 0.175);\n  --bs-border-radius: 0.375rem;\n  --bs-border-radius-sm: 0.25rem;\n  --bs-border-radius-lg: 0.5rem;\n  --bs-border-radius-xl: 1rem;\n  --bs-border-radius-xxl: 2rem;\n  --bs-border-radius-2xl: var(--bs-border-radius-xxl);\n  --bs-border-radius-pill: 50rem;\n  --bs-box-shadow: 0 0.5rem 1rem rgba(0, 0, 0, 0.15);\n  --bs-box-shadow-sm: 0 0.125rem 0.25rem rgba(0, 0, 0, 0.075);\n  --bs-box-shadow-lg: 0 1rem 3rem rgba(0, 0, 0, 0.175);\n  --bs-box-shadow-inset: inset 0 1px 2px rgba(0, 0, 0, 0.075);\n  --bs-focus-ring-width: 0.1875rem;\n  --bs-focus-ring-opacity: 1;\n  --bs-focus-ring-color: var(--pst-color-accent);\n  --bs-form-valid-color: #198754;\n  --bs-form-valid-border-color: #198754;\n  --bs-form-invalid-color: #dc3545;\n  --bs-form-invalid-border-color: #dc3545;\n}\n\n[data-bs-theme=dark] {\n  color-scheme: dark;\n  --bs-body-color: #dee2e6;\n  --bs-body-color-rgb: 222, 226, 230;\n  --bs-body-bg: #212529;\n  --bs-body-bg-rgb: 33, 37, 41;\n  --bs-emphasis-color: #fff;\n  --bs-emphasis-color-rgb: 255, 255, 255;\n  --bs-secondary-color: rgba(222, 226, 230, 0.75);\n  --bs-secondary-color-rgb: 222, 226, 230;\n  --bs-secondary-bg: #343a40;\n  --bs-secondary-bg-rgb: 52, 58, 64;\n  --bs-tertiary-color: rgba(222, 226, 230, 0.5);\n  --bs-tertiary-color-rgb: 222, 226, 230;\n  --bs-tertiary-bg: #2b3035;\n  --bs-tertiary-bg-rgb: 43, 48, 53;\n  --bs-primary-text-emphasis: #6ea8fe;\n  --bs-secondary-text-emphasis: #a7acb1;\n  --bs-success-text-emphasis: #75b798;\n  --bs-info-text-emphasis: #6edff6;\n  --bs-warning-text-emphasis: #ffda6a;\n  --bs-danger-text-emphasis: #ea868f;\n  --bs-light-text-emphasis: #f8f9fa;\n  --bs-dark-text-emphasis: #dee2e6;\n  --bs-primary-bg-subtle: #031633;\n  --bs-secondary-bg-subtle: #161719;\n  --bs-success-bg-subtle: #051b11;\n  --bs-info-bg-subtle: #032830;\n  --bs-warning-bg-subtle: #332701;\n  --bs-danger-bg-subtle: #2c0b0e;\n  --bs-light-bg-subtle: #343a40;\n  --bs-dark-bg-subtle: #1a1d20;\n  --bs-primary-border-subtle: #084298;\n  --bs-secondary-border-subtle: #41464b;\n  --bs-success-border-subtle: #0f5132;\n  --bs-info-border-subtle: #087990;\n  --bs-warning-border-subtle: #997404;\n  --bs-danger-border-subtle: #842029;\n  --bs-light-border-subtle: #495057;\n  --bs-dark-border-subtle: #343a40;\n  --bs-heading-color: inherit;\n  --bs-link-color: #6ea8fe;\n  --bs-link-hover-color: #8bb9fe;\n  --bs-link-color-rgb: 110, 168, 254;\n  --bs-link-hover-color-rgb: 139, 185, 254;\n  --bs-code-color: #e685b5;\n  --bs-highlight-color: #dee2e6;\n  --bs-highlight-bg: #664d03;\n  --bs-border-color: #495057;\n  --bs-border-color-translucent: rgba(255, 255, 255, 0.15);\n  --bs-form-valid-color: #75b798;\n  --bs-form-valid-border-color: #75b798;\n  --bs-form-invalid-color: #ea868f;\n  --bs-form-invalid-border-color: #ea868f;\n}\n\n*,\n*::before,\n*::after {\n  box-sizing: border-box;\n}\n\n@media (prefers-reduced-motion: no-preference) {\n  :root {\n    scroll-behavior: smooth;\n  }\n}\n\nbody {\n  margin: 0;\n  font-family: var(--bs-body-font-family);\n  font-size: var(--bs-body-font-size);\n  font-weight: var(--bs-body-font-weight);\n  line-height: var(--bs-body-line-height);\n  color: var(--bs-body-color);\n  text-align: var(--bs-body-text-align);\n  background-color: var(--bs-body-bg);\n  -webkit-text-size-adjust: 100%;\n  -webkit-tap-highlight-color: rgba(0, 0, 0, 0);\n}\n\nhr {\n  margin: 1rem 0;\n  color: inherit;\n  border: 0;\n  border-top: var(--bs-border-width) solid;\n  opacity: 0.25;\n}\n\nh6, .h6, h5, .h5, h4, .h4, h3, .h3, h2, .h2, h1, .h1 {\n  margin-top: 0;\n  margin-bottom: 0.5rem;\n  font-weight: 500;\n  line-height: 1.2;\n  color: var(--bs-heading-color);\n}\n\nh1, .h1 {\n  font-size: calc(1.375rem + 1.5vw);\n}\n@media (min-width: 1200px) {\n  h1, .h1 {\n    font-size: 2.5rem;\n  }\n}\n\nh2, .h2 {\n  font-size: calc(1.325rem + 0.9vw);\n}\n@media (min-width: 1200px) {\n  h2, .h2 {\n    font-size: 2rem;\n  }\n}\n\nh3, .h3 {\n  font-size: calc(1.3rem + 0.6vw);\n}\n@media (min-width: 1200px) {\n  h3, .h3 {\n    font-size: 1.75rem;\n  }\n}\n\nh4, .h4 {\n  font-size: calc(1.275rem + 0.3vw);\n}\n@media (min-width: 1200px) {\n  h4, .h4 {\n    font-size: 1.5rem;\n  }\n}\n\nh5, .h5 {\n  font-size: 1.25rem;\n}\n\nh6, .h6 {\n  font-size: 1rem;\n}\n\np {\n  margin-top: 0;\n  margin-bottom: 1rem;\n}\n\nabbr[title] {\n  text-decoration: underline dotted;\n  cursor: help;\n  text-decoration-skip-ink: none;\n}\n\naddress {\n  margin-bottom: 1rem;\n  font-style: normal;\n  line-height: inherit;\n}\n\nol,\nul {\n  padding-left: 2rem;\n}\n\nol,\nul,\ndl {\n  margin-top: 0;\n  margin-bottom: 1rem;\n}\n\nol ol,\nul ul,\nol ul,\nul ol {\n  margin-bottom: 0;\n}\n\ndt {\n  font-weight: 700;\n}\n\ndd {\n  margin-bottom: 0.5rem;\n  margin-left: 0;\n}\n\nblockquote {\n  margin: 0 0 1rem;\n}\n\nb,\nstrong {\n  font-weight: bolder;\n}\n\nsmall, .small {\n  font-size: 0.875em;\n}\n\nmark, .mark {\n  padding: 0.1875em;\n  color: var(--bs-highlight-color);\n  background-color: var(--bs-highlight-bg);\n}\n\nsub,\nsup {\n  position: relative;\n  font-size: 0.75em;\n  line-height: 0;\n  vertical-align: baseline;\n}\n\nsub {\n  bottom: -0.25em;\n}\n\nsup {\n  top: -0.5em;\n}\n\na {\n  color: rgba(var(--bs-link-color-rgb), var(--bs-link-opacity, 1));\n  text-decoration: underline;\n}\na:hover {\n  --bs-link-color-rgb: var(--bs-link-hover-color-rgb);\n}\n\na:not([href]):not([class]), a:not([href]):not([class]):hover {\n  color: inherit;\n  text-decoration: none;\n}\n\npre,\ncode,\nkbd,\nsamp {\n  font-family: var(--bs-font-monospace);\n  font-size: 1em;\n}\n\npre {\n  display: block;\n  margin-top: 0;\n  margin-bottom: 1rem;\n  overflow: auto;\n  font-size: 0.875em;\n}\npre code {\n  font-size: inherit;\n  color: inherit;\n  word-break: normal;\n}\n\ncode {\n  font-size: 0.875em;\n  color: var(--bs-code-color);\n  word-wrap: break-word;\n}\na > code {\n  color: inherit;\n}\n\nkbd {\n  padding: 0.1875rem 0.375rem;\n  font-size: 0.875em;\n  color: var(--bs-body-bg);\n  background-color: var(--bs-body-color);\n  border-radius: 0.25rem;\n}\nkbd kbd {\n  padding: 0;\n  font-size: 1em;\n}\n\nfigure {\n  margin: 0 0 1rem;\n}\n\nimg,\nsvg {\n  vertical-align: middle;\n}\n\ntable {\n  caption-side: bottom;\n  border-collapse: collapse;\n}\n\ncaption {\n  padding-top: 0.5rem;\n  padding-bottom: 0.5rem;\n  color: var(--bs-secondary-color);\n  text-align: left;\n}\n\nth {\n  text-align: inherit;\n  text-align: -webkit-match-parent;\n}\n\nthead,\ntbody,\ntfoot,\ntr,\ntd,\nth {\n  border-color: inherit;\n  border-style: solid;\n  border-width: 0;\n}\n\nlabel {\n  display: inline-block;\n}\n\nbutton {\n  border-radius: 0;\n}\n\nbutton:focus:not(:focus-visible) {\n  outline: 0;\n}\n\ninput,\nbutton,\nselect,\noptgroup,\ntextarea {\n  margin: 0;\n  font-family: inherit;\n  font-size: inherit;\n  line-height: inherit;\n}\n\nbutton,\nselect {\n  text-transform: none;\n}\n\n[role=button] {\n  cursor: pointer;\n}\n\nselect {\n  word-wrap: normal;\n}\nselect:disabled {\n  opacity: 1;\n}\n\n[list]:not([type=date]):not([type=datetime-local]):not([type=month]):not([type=week]):not([type=time])::-webkit-calendar-picker-indicator {\n  display: none !important;\n}\n\nbutton,\n[type=button],\n[type=reset],\n[type=submit] {\n  -webkit-appearance: button;\n}\nbutton:not(:disabled),\n[type=button]:not(:disabled),\n[type=reset]:not(:disabled),\n[type=submit]:not(:disabled) {\n  cursor: pointer;\n}\n\n::-moz-focus-inner {\n  padding: 0;\n  border-style: none;\n}\n\ntextarea {\n  resize: vertical;\n}\n\nfieldset {\n  min-width: 0;\n  padding: 0;\n  margin: 0;\n  border: 0;\n}\n\nlegend {\n  float: left;\n  width: 100%;\n  padding: 0;\n  margin-bottom: 0.5rem;\n  font-size: calc(1.275rem + 0.3vw);\n  line-height: inherit;\n}\n@media (min-width: 1200px) {\n  legend {\n    font-size: 1.5rem;\n  }\n}\nlegend + * {\n  clear: left;\n}\n\n::-webkit-datetime-edit-fields-wrapper,\n::-webkit-datetime-edit-text,\n::-webkit-datetime-edit-minute,\n::-webkit-datetime-edit-hour-field,\n::-webkit-datetime-edit-day-field,\n::-webkit-datetime-edit-month-field,\n::-webkit-datetime-edit-year-field {\n  padding: 0;\n}\n\n::-webkit-inner-spin-button {\n  height: auto;\n}\n\n[type=search] {\n  -webkit-appearance: textfield;\n  outline-offset: -2px;\n}\n\n/* rtl:raw:\n[type=\"tel\"],\n[type=\"url\"],\n[type=\"email\"],\n[type=\"number\"] {\n  direction: ltr;\n}\n*/\n::-webkit-search-decoration {\n  -webkit-appearance: none;\n}\n\n::-webkit-color-swatch-wrapper {\n  padding: 0;\n}\n\n::file-selector-button {\n  font: inherit;\n  -webkit-appearance: button;\n}\n\noutput {\n  display: inline-block;\n}\n\niframe {\n  border: 0;\n}\n\nsummary {\n  display: list-item;\n  cursor: pointer;\n}\n\nprogress {\n  vertical-align: baseline;\n}\n\n[hidden] {\n  display: none !important;\n}\n\n.lead {\n  font-size: 1.25rem;\n  font-weight: 300;\n}\n\n.display-1 {\n  font-size: calc(1.625rem + 4.5vw);\n  font-weight: 300;\n  line-height: 1.2;\n}\n@media (min-width: 1200px) {\n  .display-1 {\n    font-size: 5rem;\n  }\n}\n\n.display-2 {\n  font-size: calc(1.575rem + 3.9vw);\n  font-weight: 300;\n  line-height: 1.2;\n}\n@media (min-width: 1200px) {\n  .display-2 {\n    font-size: 4.5rem;\n  }\n}\n\n.display-3 {\n  font-size: calc(1.525rem + 3.3vw);\n  font-weight: 300;\n  line-height: 1.2;\n}\n@media (min-width: 1200px) {\n  .display-3 {\n    font-size: 4rem;\n  }\n}\n\n.display-4 {\n  font-size: calc(1.475rem + 2.7vw);\n  font-weight: 300;\n  line-height: 1.2;\n}\n@media (min-width: 1200px) {\n  .display-4 {\n    font-size: 3.5rem;\n  }\n}\n\n.display-5 {\n  font-size: calc(1.425rem + 2.1vw);\n  font-weight: 300;\n  line-height: 1.2;\n}\n@media (min-width: 1200px) {\n  .display-5 {\n    font-size: 3rem;\n  }\n}\n\n.display-6 {\n  font-size: calc(1.375rem + 1.5vw);\n  font-weight: 300;\n  line-height: 1.2;\n}\n@media (min-width: 1200px) {\n  .display-6 {\n    font-size: 2.5rem;\n  }\n}\n\n.list-unstyled {\n  padding-left: 0;\n  list-style: none;\n}\n\n.list-inline {\n  padding-left: 0;\n  list-style: none;\n}\n\n.list-inline-item {\n  display: inline-block;\n}\n.list-inline-item:not(:last-child) {\n  margin-right: 0.5rem;\n}\n\n.initialism {\n  font-size: 0.875em;\n  text-transform: uppercase;\n}\n\n.blockquote {\n  margin-bottom: 1rem;\n  font-size: 1.25rem;\n}\n.blockquote > :last-child {\n  margin-bottom: 0;\n}\n\n.blockquote-footer {\n  margin-top: -1rem;\n  margin-bottom: 1rem;\n  font-size: 0.875em;\n  color: #6c757d;\n}\n.blockquote-footer::before {\n  content: \"— \";\n}\n\n.img-fluid {\n  max-width: 100%;\n  height: auto;\n}\n\n.img-thumbnail {\n  padding: 0.25rem;\n  background-color: var(--bs-body-bg);\n  border: var(--bs-border-width) solid var(--bs-border-color);\n  border-radius: var(--bs-border-radius);\n  max-width: 100%;\n  height: auto;\n}\n\n.figure {\n  display: inline-block;\n}\n\n.figure-img {\n  margin-bottom: 0.5rem;\n  line-height: 1;\n}\n\n.figure-caption {\n  font-size: 0.875em;\n  color: var(--bs-secondary-color);\n}\n\n.container,\n.container-fluid,\n.container-xl,\n.container-lg,\n.container-md,\n.container-sm {\n  --bs-gutter-x: 1.5rem;\n  --bs-gutter-y: 0;\n  width: 100%;\n  padding-right: calc(var(--bs-gutter-x) * 0.5);\n  padding-left: calc(var(--bs-gutter-x) * 0.5);\n  margin-right: auto;\n  margin-left: auto;\n}\n\n@media (min-width: 540px) {\n  .container-sm, .container {\n    max-width: 540px;\n  }\n}\n@media (min-width: 720px) {\n  .container-md, .container-sm, .container {\n    max-width: 720px;\n  }\n}\n@media (min-width: 960px) {\n  .container-lg, .container-md, .container-sm, .container {\n    max-width: 960px;\n  }\n}\n@media (min-width: 1200px) {\n  .container-xl, .container-lg, .container-md, .container-sm, .container {\n    max-width: 1400px;\n  }\n}\n:root {\n  --bs-breakpoint-xs: 0;\n  --bs-breakpoint-sm: 540px;\n  --bs-breakpoint-md: 720px;\n  --bs-breakpoint-lg: 960px;\n  --bs-breakpoint-xl: 1200px;\n}\n\n.row {\n  --bs-gutter-x: 1.5rem;\n  --bs-gutter-y: 0;\n  display: flex;\n  flex-wrap: wrap;\n  margin-top: calc(-1 * var(--bs-gutter-y));\n  margin-right: calc(-0.5 * var(--bs-gutter-x));\n  margin-left: calc(-0.5 * var(--bs-gutter-x));\n}\n.row > * {\n  flex-shrink: 0;\n  width: 100%;\n  max-width: 100%;\n  padding-right: calc(var(--bs-gutter-x) * 0.5);\n  padding-left: calc(var(--bs-gutter-x) * 0.5);\n  margin-top: var(--bs-gutter-y);\n}\n\n.col {\n  flex: 1 0 0%;\n}\n\n.row-cols-auto > * {\n  flex: 0 0 auto;\n  width: auto;\n}\n\n.row-cols-1 > * {\n  flex: 0 0 auto;\n  width: 100%;\n}\n\n.row-cols-2 > * {\n  flex: 0 0 auto;\n  width: 50%;\n}\n\n.row-cols-3 > * {\n  flex: 0 0 auto;\n  width: 33.33333333%;\n}\n\n.row-cols-4 > * {\n  flex: 0 0 auto;\n  width: 25%;\n}\n\n.row-cols-5 > * {\n  flex: 0 0 auto;\n  width: 20%;\n}\n\n.row-cols-6 > * {\n  flex: 0 0 auto;\n  width: 16.66666667%;\n}\n\n.col-auto {\n  flex: 0 0 auto;\n  width: auto;\n}\n\n.col-1 {\n  flex: 0 0 auto;\n  width: 8.33333333%;\n}\n\n.col-2 {\n  flex: 0 0 auto;\n  width: 16.66666667%;\n}\n\n.col-3 {\n  flex: 0 0 auto;\n  width: 25%;\n}\n\n.col-4 {\n  flex: 0 0 auto;\n  width: 33.33333333%;\n}\n\n.col-5 {\n  flex: 0 0 auto;\n  width: 41.66666667%;\n}\n\n.col-6 {\n  flex: 0 0 auto;\n  width: 50%;\n}\n\n.col-7 {\n  flex: 0 0 auto;\n  width: 58.33333333%;\n}\n\n.col-8 {\n  flex: 0 0 auto;\n  width: 66.66666667%;\n}\n\n.col-9 {\n  flex: 0 0 auto;\n  width: 75%;\n}\n\n.col-10 {\n  flex: 0 0 auto;\n  width: 83.33333333%;\n}\n\n.col-11 {\n  flex: 0 0 auto;\n  width: 91.66666667%;\n}\n\n.col-12 {\n  flex: 0 0 auto;\n  width: 100%;\n}\n\n.offset-1 {\n  margin-left: 8.33333333%;\n}\n\n.offset-2 {\n  margin-left: 16.66666667%;\n}\n\n.offset-3 {\n  margin-left: 25%;\n}\n\n.offset-4 {\n  margin-left: 33.33333333%;\n}\n\n.offset-5 {\n  margin-left: 41.66666667%;\n}\n\n.offset-6 {\n  margin-left: 50%;\n}\n\n.offset-7 {\n  margin-left: 58.33333333%;\n}\n\n.offset-8 {\n  margin-left: 66.66666667%;\n}\n\n.offset-9 {\n  margin-left: 75%;\n}\n\n.offset-10 {\n  margin-left: 83.33333333%;\n}\n\n.offset-11 {\n  margin-left: 91.66666667%;\n}\n\n.g-0,\n.gx-0 {\n  --bs-gutter-x: 0;\n}\n\n.g-0,\n.gy-0 {\n  --bs-gutter-y: 0;\n}\n\n.g-1,\n.gx-1 {\n  --bs-gutter-x: 0.25rem;\n}\n\n.g-1,\n.gy-1 {\n  --bs-gutter-y: 0.25rem;\n}\n\n.g-2,\n.gx-2 {\n  --bs-gutter-x: 0.5rem;\n}\n\n.g-2,\n.gy-2 {\n  --bs-gutter-y: 0.5rem;\n}\n\n.g-3,\n.gx-3 {\n  --bs-gutter-x: 1rem;\n}\n\n.g-3,\n.gy-3 {\n  --bs-gutter-y: 1rem;\n}\n\n.g-4,\n.gx-4 {\n  --bs-gutter-x: 1.5rem;\n}\n\n.g-4,\n.gy-4 {\n  --bs-gutter-y: 1.5rem;\n}\n\n.g-5,\n.gx-5 {\n  --bs-gutter-x: 3rem;\n}\n\n.g-5,\n.gy-5 {\n  --bs-gutter-y: 3rem;\n}\n\n@media (min-width: 540px) {\n  .col-sm {\n    flex: 1 0 0%;\n  }\n  .row-cols-sm-auto > * {\n    flex: 0 0 auto;\n    width: auto;\n  }\n  .row-cols-sm-1 > * {\n    flex: 0 0 auto;\n    width: 100%;\n  }\n  .row-cols-sm-2 > * {\n    flex: 0 0 auto;\n    width: 50%;\n  }\n  .row-cols-sm-3 > * {\n    flex: 0 0 auto;\n    width: 33.33333333%;\n  }\n  .row-cols-sm-4 > * {\n    flex: 0 0 auto;\n    width: 25%;\n  }\n  .row-cols-sm-5 > * {\n    flex: 0 0 auto;\n    width: 20%;\n  }\n  .row-cols-sm-6 > * {\n    flex: 0 0 auto;\n    width: 16.66666667%;\n  }\n  .col-sm-auto {\n    flex: 0 0 auto;\n    width: auto;\n  }\n  .col-sm-1 {\n    flex: 0 0 auto;\n    width: 8.33333333%;\n  }\n  .col-sm-2 {\n    flex: 0 0 auto;\n    width: 16.66666667%;\n  }\n  .col-sm-3 {\n    flex: 0 0 auto;\n    width: 25%;\n  }\n  .col-sm-4 {\n    flex: 0 0 auto;\n    width: 33.33333333%;\n  }\n  .col-sm-5 {\n    flex: 0 0 auto;\n    width: 41.66666667%;\n  }\n  .col-sm-6 {\n    flex: 0 0 auto;\n    width: 50%;\n  }\n  .col-sm-7 {\n    flex: 0 0 auto;\n    width: 58.33333333%;\n  }\n  .col-sm-8 {\n    flex: 0 0 auto;\n    width: 66.66666667%;\n  }\n  .col-sm-9 {\n    flex: 0 0 auto;\n    width: 75%;\n  }\n  .col-sm-10 {\n    flex: 0 0 auto;\n    width: 83.33333333%;\n  }\n  .col-sm-11 {\n    flex: 0 0 auto;\n    width: 91.66666667%;\n  }\n  .col-sm-12 {\n    flex: 0 0 auto;\n    width: 100%;\n  }\n  .offset-sm-0 {\n    margin-left: 0;\n  }\n  .offset-sm-1 {\n    margin-left: 8.33333333%;\n  }\n  .offset-sm-2 {\n    margin-left: 16.66666667%;\n  }\n  .offset-sm-3 {\n    margin-left: 25%;\n  }\n  .offset-sm-4 {\n    margin-left: 33.33333333%;\n  }\n  .offset-sm-5 {\n    margin-left: 41.66666667%;\n  }\n  .offset-sm-6 {\n    margin-left: 50%;\n  }\n  .offset-sm-7 {\n    margin-left: 58.33333333%;\n  }\n  .offset-sm-8 {\n    margin-left: 66.66666667%;\n  }\n  .offset-sm-9 {\n    margin-left: 75%;\n  }\n  .offset-sm-10 {\n    margin-left: 83.33333333%;\n  }\n  .offset-sm-11 {\n    margin-left: 91.66666667%;\n  }\n  .g-sm-0,\n  .gx-sm-0 {\n    --bs-gutter-x: 0;\n  }\n  .g-sm-0,\n  .gy-sm-0 {\n    --bs-gutter-y: 0;\n  }\n  .g-sm-1,\n  .gx-sm-1 {\n    --bs-gutter-x: 0.25rem;\n  }\n  .g-sm-1,\n  .gy-sm-1 {\n    --bs-gutter-y: 0.25rem;\n  }\n  .g-sm-2,\n  .gx-sm-2 {\n    --bs-gutter-x: 0.5rem;\n  }\n  .g-sm-2,\n  .gy-sm-2 {\n    --bs-gutter-y: 0.5rem;\n  }\n  .g-sm-3,\n  .gx-sm-3 {\n    --bs-gutter-x: 1rem;\n  }\n  .g-sm-3,\n  .gy-sm-3 {\n    --bs-gutter-y: 1rem;\n  }\n  .g-sm-4,\n  .gx-sm-4 {\n    --bs-gutter-x: 1.5rem;\n  }\n  .g-sm-4,\n  .gy-sm-4 {\n    --bs-gutter-y: 1.5rem;\n  }\n  .g-sm-5,\n  .gx-sm-5 {\n    --bs-gutter-x: 3rem;\n  }\n  .g-sm-5,\n  .gy-sm-5 {\n    --bs-gutter-y: 3rem;\n  }\n}\n@media (min-width: 720px) {\n  .col-md {\n    flex: 1 0 0%;\n  }\n  .row-cols-md-auto > * {\n    flex: 0 0 auto;\n    width: auto;\n  }\n  .row-cols-md-1 > * {\n    flex: 0 0 auto;\n    width: 100%;\n  }\n  .row-cols-md-2 > * {\n    flex: 0 0 auto;\n    width: 50%;\n  }\n  .row-cols-md-3 > * {\n    flex: 0 0 auto;\n    width: 33.33333333%;\n  }\n  .row-cols-md-4 > * {\n    flex: 0 0 auto;\n    width: 25%;\n  }\n  .row-cols-md-5 > * {\n    flex: 0 0 auto;\n    width: 20%;\n  }\n  .row-cols-md-6 > * {\n    flex: 0 0 auto;\n    width: 16.66666667%;\n  }\n  .col-md-auto {\n    flex: 0 0 auto;\n    width: auto;\n  }\n  .col-md-1 {\n    flex: 0 0 auto;\n    width: 8.33333333%;\n  }\n  .col-md-2 {\n    flex: 0 0 auto;\n    width: 16.66666667%;\n  }\n  .col-md-3 {\n    flex: 0 0 auto;\n    width: 25%;\n  }\n  .col-md-4 {\n    flex: 0 0 auto;\n    width: 33.33333333%;\n  }\n  .col-md-5 {\n    flex: 0 0 auto;\n    width: 41.66666667%;\n  }\n  .col-md-6 {\n    flex: 0 0 auto;\n    width: 50%;\n  }\n  .col-md-7 {\n    flex: 0 0 auto;\n    width: 58.33333333%;\n  }\n  .col-md-8 {\n    flex: 0 0 auto;\n    width: 66.66666667%;\n  }\n  .col-md-9 {\n    flex: 0 0 auto;\n    width: 75%;\n  }\n  .col-md-10 {\n    flex: 0 0 auto;\n    width: 83.33333333%;\n  }\n  .col-md-11 {\n    flex: 0 0 auto;\n    width: 91.66666667%;\n  }\n  .col-md-12 {\n    flex: 0 0 auto;\n    width: 100%;\n  }\n  .offset-md-0 {\n    margin-left: 0;\n  }\n  .offset-md-1 {\n    margin-left: 8.33333333%;\n  }\n  .offset-md-2 {\n    margin-left: 16.66666667%;\n  }\n  .offset-md-3 {\n    margin-left: 25%;\n  }\n  .offset-md-4 {\n    margin-left: 33.33333333%;\n  }\n  .offset-md-5 {\n    margin-left: 41.66666667%;\n  }\n  .offset-md-6 {\n    margin-left: 50%;\n  }\n  .offset-md-7 {\n    margin-left: 58.33333333%;\n  }\n  .offset-md-8 {\n    margin-left: 66.66666667%;\n  }\n  .offset-md-9 {\n    margin-left: 75%;\n  }\n  .offset-md-10 {\n    margin-left: 83.33333333%;\n  }\n  .offset-md-11 {\n    margin-left: 91.66666667%;\n  }\n  .g-md-0,\n  .gx-md-0 {\n    --bs-gutter-x: 0;\n  }\n  .g-md-0,\n  .gy-md-0 {\n    --bs-gutter-y: 0;\n  }\n  .g-md-1,\n  .gx-md-1 {\n    --bs-gutter-x: 0.25rem;\n  }\n  .g-md-1,\n  .gy-md-1 {\n    --bs-gutter-y: 0.25rem;\n  }\n  .g-md-2,\n  .gx-md-2 {\n    --bs-gutter-x: 0.5rem;\n  }\n  .g-md-2,\n  .gy-md-2 {\n    --bs-gutter-y: 0.5rem;\n  }\n  .g-md-3,\n  .gx-md-3 {\n    --bs-gutter-x: 1rem;\n  }\n  .g-md-3,\n  .gy-md-3 {\n    --bs-gutter-y: 1rem;\n  }\n  .g-md-4,\n  .gx-md-4 {\n    --bs-gutter-x: 1.5rem;\n  }\n  .g-md-4,\n  .gy-md-4 {\n    --bs-gutter-y: 1.5rem;\n  }\n  .g-md-5,\n  .gx-md-5 {\n    --bs-gutter-x: 3rem;\n  }\n  .g-md-5,\n  .gy-md-5 {\n    --bs-gutter-y: 3rem;\n  }\n}\n@media (min-width: 960px) {\n  .col-lg {\n    flex: 1 0 0%;\n  }\n  .row-cols-lg-auto > * {\n    flex: 0 0 auto;\n    width: auto;\n  }\n  .row-cols-lg-1 > * {\n    flex: 0 0 auto;\n    width: 100%;\n  }\n  .row-cols-lg-2 > * {\n    flex: 0 0 auto;\n    width: 50%;\n  }\n  .row-cols-lg-3 > * {\n    flex: 0 0 auto;\n    width: 33.33333333%;\n  }\n  .row-cols-lg-4 > * {\n    flex: 0 0 auto;\n    width: 25%;\n  }\n  .row-cols-lg-5 > * {\n    flex: 0 0 auto;\n    width: 20%;\n  }\n  .row-cols-lg-6 > * {\n    flex: 0 0 auto;\n    width: 16.66666667%;\n  }\n  .col-lg-auto {\n    flex: 0 0 auto;\n    width: auto;\n  }\n  .col-lg-1 {\n    flex: 0 0 auto;\n    width: 8.33333333%;\n  }\n  .col-lg-2 {\n    flex: 0 0 auto;\n    width: 16.66666667%;\n  }\n  .col-lg-3 {\n    flex: 0 0 auto;\n    width: 25%;\n  }\n  .col-lg-4 {\n    flex: 0 0 auto;\n    width: 33.33333333%;\n  }\n  .col-lg-5 {\n    flex: 0 0 auto;\n    width: 41.66666667%;\n  }\n  .col-lg-6 {\n    flex: 0 0 auto;\n    width: 50%;\n  }\n  .col-lg-7 {\n    flex: 0 0 auto;\n    width: 58.33333333%;\n  }\n  .col-lg-8 {\n    flex: 0 0 auto;\n    width: 66.66666667%;\n  }\n  .col-lg-9 {\n    flex: 0 0 auto;\n    width: 75%;\n  }\n  .col-lg-10 {\n    flex: 0 0 auto;\n    width: 83.33333333%;\n  }\n  .col-lg-11 {\n    flex: 0 0 auto;\n    width: 91.66666667%;\n  }\n  .col-lg-12 {\n    flex: 0 0 auto;\n    width: 100%;\n  }\n  .offset-lg-0 {\n    margin-left: 0;\n  }\n  .offset-lg-1 {\n    margin-left: 8.33333333%;\n  }\n  .offset-lg-2 {\n    margin-left: 16.66666667%;\n  }\n  .offset-lg-3 {\n    margin-left: 25%;\n  }\n  .offset-lg-4 {\n    margin-left: 33.33333333%;\n  }\n  .offset-lg-5 {\n    margin-left: 41.66666667%;\n  }\n  .offset-lg-6 {\n    margin-left: 50%;\n  }\n  .offset-lg-7 {\n    margin-left: 58.33333333%;\n  }\n  .offset-lg-8 {\n    margin-left: 66.66666667%;\n  }\n  .offset-lg-9 {\n    margin-left: 75%;\n  }\n  .offset-lg-10 {\n    margin-left: 83.33333333%;\n  }\n  .offset-lg-11 {\n    margin-left: 91.66666667%;\n  }\n  .g-lg-0,\n  .gx-lg-0 {\n    --bs-gutter-x: 0;\n  }\n  .g-lg-0,\n  .gy-lg-0 {\n    --bs-gutter-y: 0;\n  }\n  .g-lg-1,\n  .gx-lg-1 {\n    --bs-gutter-x: 0.25rem;\n  }\n  .g-lg-1,\n  .gy-lg-1 {\n    --bs-gutter-y: 0.25rem;\n  }\n  .g-lg-2,\n  .gx-lg-2 {\n    --bs-gutter-x: 0.5rem;\n  }\n  .g-lg-2,\n  .gy-lg-2 {\n    --bs-gutter-y: 0.5rem;\n  }\n  .g-lg-3,\n  .gx-lg-3 {\n    --bs-gutter-x: 1rem;\n  }\n  .g-lg-3,\n  .gy-lg-3 {\n    --bs-gutter-y: 1rem;\n  }\n  .g-lg-4,\n  .gx-lg-4 {\n    --bs-gutter-x: 1.5rem;\n  }\n  .g-lg-4,\n  .gy-lg-4 {\n    --bs-gutter-y: 1.5rem;\n  }\n  .g-lg-5,\n  .gx-lg-5 {\n    --bs-gutter-x: 3rem;\n  }\n  .g-lg-5,\n  .gy-lg-5 {\n    --bs-gutter-y: 3rem;\n  }\n}\n@media (min-width: 1200px) {\n  .col-xl {\n    flex: 1 0 0%;\n  }\n  .row-cols-xl-auto > * {\n    flex: 0 0 auto;\n    width: auto;\n  }\n  .row-cols-xl-1 > * {\n    flex: 0 0 auto;\n    width: 100%;\n  }\n  .row-cols-xl-2 > * {\n    flex: 0 0 auto;\n    width: 50%;\n  }\n  .row-cols-xl-3 > * {\n    flex: 0 0 auto;\n    width: 33.33333333%;\n  }\n  .row-cols-xl-4 > * {\n    flex: 0 0 auto;\n    width: 25%;\n  }\n  .row-cols-xl-5 > * {\n    flex: 0 0 auto;\n    width: 20%;\n  }\n  .row-cols-xl-6 > * {\n    flex: 0 0 auto;\n    width: 16.66666667%;\n  }\n  .col-xl-auto {\n    flex: 0 0 auto;\n    width: auto;\n  }\n  .col-xl-1 {\n    flex: 0 0 auto;\n    width: 8.33333333%;\n  }\n  .col-xl-2 {\n    flex: 0 0 auto;\n    width: 16.66666667%;\n  }\n  .col-xl-3 {\n    flex: 0 0 auto;\n    width: 25%;\n  }\n  .col-xl-4 {\n    flex: 0 0 auto;\n    width: 33.33333333%;\n  }\n  .col-xl-5 {\n    flex: 0 0 auto;\n    width: 41.66666667%;\n  }\n  .col-xl-6 {\n    flex: 0 0 auto;\n    width: 50%;\n  }\n  .col-xl-7 {\n    flex: 0 0 auto;\n    width: 58.33333333%;\n  }\n  .col-xl-8 {\n    flex: 0 0 auto;\n    width: 66.66666667%;\n  }\n  .col-xl-9 {\n    flex: 0 0 auto;\n    width: 75%;\n  }\n  .col-xl-10 {\n    flex: 0 0 auto;\n    width: 83.33333333%;\n  }\n  .col-xl-11 {\n    flex: 0 0 auto;\n    width: 91.66666667%;\n  }\n  .col-xl-12 {\n    flex: 0 0 auto;\n    width: 100%;\n  }\n  .offset-xl-0 {\n    margin-left: 0;\n  }\n  .offset-xl-1 {\n    margin-left: 8.33333333%;\n  }\n  .offset-xl-2 {\n    margin-left: 16.66666667%;\n  }\n  .offset-xl-3 {\n    margin-left: 25%;\n  }\n  .offset-xl-4 {\n    margin-left: 33.33333333%;\n  }\n  .offset-xl-5 {\n    margin-left: 41.66666667%;\n  }\n  .offset-xl-6 {\n    margin-left: 50%;\n  }\n  .offset-xl-7 {\n    margin-left: 58.33333333%;\n  }\n  .offset-xl-8 {\n    margin-left: 66.66666667%;\n  }\n  .offset-xl-9 {\n    margin-left: 75%;\n  }\n  .offset-xl-10 {\n    margin-left: 83.33333333%;\n  }\n  .offset-xl-11 {\n    margin-left: 91.66666667%;\n  }\n  .g-xl-0,\n  .gx-xl-0 {\n    --bs-gutter-x: 0;\n  }\n  .g-xl-0,\n  .gy-xl-0 {\n    --bs-gutter-y: 0;\n  }\n  .g-xl-1,\n  .gx-xl-1 {\n    --bs-gutter-x: 0.25rem;\n  }\n  .g-xl-1,\n  .gy-xl-1 {\n    --bs-gutter-y: 0.25rem;\n  }\n  .g-xl-2,\n  .gx-xl-2 {\n    --bs-gutter-x: 0.5rem;\n  }\n  .g-xl-2,\n  .gy-xl-2 {\n    --bs-gutter-y: 0.5rem;\n  }\n  .g-xl-3,\n  .gx-xl-3 {\n    --bs-gutter-x: 1rem;\n  }\n  .g-xl-3,\n  .gy-xl-3 {\n    --bs-gutter-y: 1rem;\n  }\n  .g-xl-4,\n  .gx-xl-4 {\n    --bs-gutter-x: 1.5rem;\n  }\n  .g-xl-4,\n  .gy-xl-4 {\n    --bs-gutter-y: 1.5rem;\n  }\n  .g-xl-5,\n  .gx-xl-5 {\n    --bs-gutter-x: 3rem;\n  }\n  .g-xl-5,\n  .gy-xl-5 {\n    --bs-gutter-y: 3rem;\n  }\n}\n.table {\n  --bs-table-color-type: initial;\n  --bs-table-bg-type: initial;\n  --bs-table-color-state: initial;\n  --bs-table-bg-state: initial;\n  --bs-table-color: var(--bs-emphasis-color);\n  --bs-table-bg: var(--bs-body-bg);\n  --bs-table-border-color: var(--bs-border-color);\n  --bs-table-accent-bg: transparent;\n  --bs-table-striped-color: var(--bs-emphasis-color);\n  --bs-table-striped-bg: rgba(var(--bs-emphasis-color-rgb), 0.05);\n  --bs-table-active-color: var(--bs-emphasis-color);\n  --bs-table-active-bg: rgba(var(--bs-emphasis-color-rgb), 0.1);\n  --bs-table-hover-color: var(--bs-emphasis-color);\n  --bs-table-hover-bg: rgba(var(--bs-emphasis-color-rgb), 0.075);\n  width: 100%;\n  margin-bottom: 1rem;\n  vertical-align: top;\n  border-color: var(--bs-table-border-color);\n}\n.table > :not(caption) > * > * {\n  padding: 0.5rem 0.5rem;\n  color: var(--bs-table-color-state, var(--bs-table-color-type, var(--bs-table-color)));\n  background-color: var(--bs-table-bg);\n  border-bottom-width: var(--bs-border-width);\n  box-shadow: inset 0 0 0 9999px var(--bs-table-bg-state, var(--bs-table-bg-type, var(--bs-table-accent-bg)));\n}\n.table > tbody {\n  vertical-align: inherit;\n}\n.table > thead {\n  vertical-align: bottom;\n}\n\n.table-group-divider {\n  border-top: calc(var(--bs-border-width) * 2) solid currentcolor;\n}\n\n.caption-top {\n  caption-side: top;\n}\n\n.table-sm > :not(caption) > * > * {\n  padding: 0.25rem 0.25rem;\n}\n\n.table-bordered > :not(caption) > * {\n  border-width: var(--bs-border-width) 0;\n}\n.table-bordered > :not(caption) > * > * {\n  border-width: 0 var(--bs-border-width);\n}\n\n.table-borderless > :not(caption) > * > * {\n  border-bottom-width: 0;\n}\n.table-borderless > :not(:first-child) {\n  border-top-width: 0;\n}\n\n.table-striped > tbody > tr:nth-of-type(odd) > * {\n  --bs-table-color-type: var(--bs-table-striped-color);\n  --bs-table-bg-type: var(--bs-table-striped-bg);\n}\n\n.table-striped-columns > :not(caption) > tr > :nth-child(even) {\n  --bs-table-color-type: var(--bs-table-striped-color);\n  --bs-table-bg-type: var(--bs-table-striped-bg);\n}\n\n.table-active {\n  --bs-table-color-state: var(--bs-table-active-color);\n  --bs-table-bg-state: var(--bs-table-active-bg);\n}\n\n.table-hover > tbody > tr:hover > * {\n  --bs-table-color-state: var(--bs-table-hover-color);\n  --bs-table-bg-state: var(--bs-table-hover-bg);\n}\n\n.table-primary {\n  --bs-table-color: #000;\n  --bs-table-bg: #cfe2ff;\n  --bs-table-border-color: #a6b5cc;\n  --bs-table-striped-bg: #c5d7f2;\n  --bs-table-striped-color: #000;\n  --bs-table-active-bg: #bacbe6;\n  --bs-table-active-color: #000;\n  --bs-table-hover-bg: #bfd1ec;\n  --bs-table-hover-color: #000;\n  color: var(--bs-table-color);\n  border-color: var(--bs-table-border-color);\n}\n\n.table-secondary {\n  --bs-table-color: #000;\n  --bs-table-bg: #e2e3e5;\n  --bs-table-border-color: #b5b6b7;\n  --bs-table-striped-bg: #d7d8da;\n  --bs-table-striped-color: #000;\n  --bs-table-active-bg: #cbccce;\n  --bs-table-active-color: #000;\n  --bs-table-hover-bg: #d1d2d4;\n  --bs-table-hover-color: #000;\n  color: var(--bs-table-color);\n  border-color: var(--bs-table-border-color);\n}\n\n.table-success {\n  --bs-table-color: #000;\n  --bs-table-bg: #d1e7dd;\n  --bs-table-border-color: #a7b9b1;\n  --bs-table-striped-bg: #c7dbd2;\n  --bs-table-striped-color: #000;\n  --bs-table-active-bg: #bcd0c7;\n  --bs-table-active-color: #000;\n  --bs-table-hover-bg: #c1d6cc;\n  --bs-table-hover-color: #000;\n  color: var(--bs-table-color);\n  border-color: var(--bs-table-border-color);\n}\n\n.table-info {\n  --bs-table-color: #000;\n  --bs-table-bg: #cff4fc;\n  --bs-table-border-color: #a6c3ca;\n  --bs-table-striped-bg: #c5e8ef;\n  --bs-table-striped-color: #000;\n  --bs-table-active-bg: #badce3;\n  --bs-table-active-color: #000;\n  --bs-table-hover-bg: #bfe2e9;\n  --bs-table-hover-color: #000;\n  color: var(--bs-table-color);\n  border-color: var(--bs-table-border-color);\n}\n\n.table-warning {\n  --bs-table-color: #000;\n  --bs-table-bg: #fff3cd;\n  --bs-table-border-color: #ccc2a4;\n  --bs-table-striped-bg: #f2e7c3;\n  --bs-table-striped-color: #000;\n  --bs-table-active-bg: #e6dbb9;\n  --bs-table-active-color: #000;\n  --bs-table-hover-bg: #ece1be;\n  --bs-table-hover-color: #000;\n  color: var(--bs-table-color);\n  border-color: var(--bs-table-border-color);\n}\n\n.table-danger {\n  --bs-table-color: #000;\n  --bs-table-bg: #f8d7da;\n  --bs-table-border-color: #c6acae;\n  --bs-table-striped-bg: #eccccf;\n  --bs-table-striped-color: #000;\n  --bs-table-active-bg: #dfc2c4;\n  --bs-table-active-color: #000;\n  --bs-table-hover-bg: #e5c7ca;\n  --bs-table-hover-color: #000;\n  color: var(--bs-table-color);\n  border-color: var(--bs-table-border-color);\n}\n\n.table-light {\n  --bs-table-color: #000;\n  --bs-table-bg: #f8f9fa;\n  --bs-table-border-color: #c6c7c8;\n  --bs-table-striped-bg: #ecedee;\n  --bs-table-striped-color: #000;\n  --bs-table-active-bg: #dfe0e1;\n  --bs-table-active-color: #000;\n  --bs-table-hover-bg: #e5e6e7;\n  --bs-table-hover-color: #000;\n  color: var(--bs-table-color);\n  border-color: var(--bs-table-border-color);\n}\n\n.table-dark {\n  --bs-table-color: #fff;\n  --bs-table-bg: #212529;\n  --bs-table-border-color: #4d5154;\n  --bs-table-striped-bg: #2c3034;\n  --bs-table-striped-color: #fff;\n  --bs-table-active-bg: #373b3e;\n  --bs-table-active-color: #fff;\n  --bs-table-hover-bg: #323539;\n  --bs-table-hover-color: #fff;\n  color: var(--bs-table-color);\n  border-color: var(--bs-table-border-color);\n}\n\n.table-responsive {\n  overflow-x: auto;\n  -webkit-overflow-scrolling: touch;\n}\n\n@media (max-width: 539.98px) {\n  .table-responsive-sm {\n    overflow-x: auto;\n    -webkit-overflow-scrolling: touch;\n  }\n}\n@media (max-width: 719.98px) {\n  .table-responsive-md {\n    overflow-x: auto;\n    -webkit-overflow-scrolling: touch;\n  }\n}\n@media (max-width: 959.98px) {\n  .table-responsive-lg {\n    overflow-x: auto;\n    -webkit-overflow-scrolling: touch;\n  }\n}\n@media (max-width: 1199.98px) {\n  .table-responsive-xl {\n    overflow-x: auto;\n    -webkit-overflow-scrolling: touch;\n  }\n}\n.form-label {\n  margin-bottom: 0.5rem;\n}\n\n.col-form-label {\n  padding-top: calc(0.375rem + var(--bs-border-width));\n  padding-bottom: calc(0.375rem + var(--bs-border-width));\n  margin-bottom: 0;\n  font-size: inherit;\n  line-height: 1.5;\n}\n\n.col-form-label-lg {\n  padding-top: calc(0.5rem + var(--bs-border-width));\n  padding-bottom: calc(0.5rem + var(--bs-border-width));\n  font-size: 1.25rem;\n}\n\n.col-form-label-sm {\n  padding-top: calc(0.25rem + var(--bs-border-width));\n  padding-bottom: calc(0.25rem + var(--bs-border-width));\n  font-size: 0.875rem;\n}\n\n.form-text {\n  margin-top: 0.25rem;\n  font-size: 0.875em;\n  color: var(--bs-secondary-color);\n}\n\n.form-control {\n  display: block;\n  width: 100%;\n  padding: 0.375rem 0.75rem;\n  font-size: 1rem;\n  font-weight: 400;\n  line-height: 1.5;\n  color: var(--bs-body-color);\n  appearance: none;\n  background-color: var(--bs-body-bg);\n  background-clip: padding-box;\n  border: var(--bs-border-width) solid var(--bs-border-color);\n  border-radius: var(--bs-border-radius);\n  transition: border-color 0.15s ease-in-out, box-shadow 0.15s ease-in-out;\n}\n@media (prefers-reduced-motion: reduce) {\n  .form-control {\n    transition: none;\n  }\n}\n.form-control[type=file] {\n  overflow: hidden;\n}\n.form-control[type=file]:not(:disabled):not([readonly]) {\n  cursor: pointer;\n}\n.form-control:focus {\n  color: var(--bs-body-color);\n  background-color: var(--bs-body-bg);\n  border-color: #86b7fe;\n  outline: 0;\n  box-shadow: 0 0 0 0.1875rem var(--pst-color-accent);\n}\n.form-control::-webkit-date-and-time-value {\n  min-width: 85px;\n  height: 1.5em;\n  margin: 0;\n}\n.form-control::-webkit-datetime-edit {\n  display: block;\n  padding: 0;\n}\n.form-control::placeholder {\n  color: var(--bs-secondary-color);\n  opacity: 1;\n}\n.form-control:disabled {\n  background-color: var(--bs-secondary-bg);\n  opacity: 1;\n}\n.form-control::file-selector-button {\n  padding: 0.375rem 0.75rem;\n  margin: -0.375rem -0.75rem;\n  margin-inline-end: 0.75rem;\n  color: var(--bs-body-color);\n  background-color: var(--bs-tertiary-bg);\n  pointer-events: none;\n  border-color: inherit;\n  border-style: solid;\n  border-width: 0;\n  border-inline-end-width: var(--bs-border-width);\n  border-radius: 0;\n  transition: color 0.15s ease-in-out, background-color 0.15s ease-in-out, border-color 0.15s ease-in-out, box-shadow 0.15s ease-in-out;\n}\n@media (prefers-reduced-motion: reduce) {\n  .form-control::file-selector-button {\n    transition: none;\n  }\n}\n.form-control:hover:not(:disabled):not([readonly])::file-selector-button {\n  background-color: var(--bs-secondary-bg);\n}\n\n.form-control-plaintext {\n  display: block;\n  width: 100%;\n  padding: 0.375rem 0;\n  margin-bottom: 0;\n  line-height: 1.5;\n  color: var(--bs-body-color);\n  background-color: transparent;\n  border: solid transparent;\n  border-width: var(--bs-border-width) 0;\n}\n.form-control-plaintext:focus {\n  outline: 0;\n}\n.form-control-plaintext.form-control-sm, .form-control-plaintext.form-control-lg {\n  padding-right: 0;\n  padding-left: 0;\n}\n\n.form-control-sm {\n  min-height: calc(1.5em + 0.5rem + calc(var(--bs-border-width) * 2));\n  padding: 0.25rem 0.5rem;\n  font-size: 0.875rem;\n  border-radius: var(--bs-border-radius-sm);\n}\n.form-control-sm::file-selector-button {\n  padding: 0.25rem 0.5rem;\n  margin: -0.25rem -0.5rem;\n  margin-inline-end: 0.5rem;\n}\n\n.form-control-lg {\n  min-height: calc(1.5em + 1rem + calc(var(--bs-border-width) * 2));\n  padding: 0.5rem 1rem;\n  font-size: 1.25rem;\n  border-radius: var(--bs-border-radius-lg);\n}\n.form-control-lg::file-selector-button {\n  padding: 0.5rem 1rem;\n  margin: -0.5rem -1rem;\n  margin-inline-end: 1rem;\n}\n\ntextarea.form-control {\n  min-height: calc(1.5em + 0.75rem + calc(var(--bs-border-width) * 2));\n}\ntextarea.form-control-sm {\n  min-height: calc(1.5em + 0.5rem + calc(var(--bs-border-width) * 2));\n}\ntextarea.form-control-lg {\n  min-height: calc(1.5em + 1rem + calc(var(--bs-border-width) * 2));\n}\n\n.form-control-color {\n  width: 3rem;\n  height: calc(1.5em + 0.75rem + calc(var(--bs-border-width) * 2));\n  padding: 0.375rem;\n}\n.form-control-color:not(:disabled):not([readonly]) {\n  cursor: pointer;\n}\n.form-control-color::-moz-color-swatch {\n  border: 0 !important;\n  border-radius: var(--bs-border-radius);\n}\n.form-control-color::-webkit-color-swatch {\n  border: 0 !important;\n  border-radius: var(--bs-border-radius);\n}\n.form-control-color.form-control-sm {\n  height: calc(1.5em + 0.5rem + calc(var(--bs-border-width) * 2));\n}\n.form-control-color.form-control-lg {\n  height: calc(1.5em + 1rem + calc(var(--bs-border-width) * 2));\n}\n\n.form-select {\n  --bs-form-select-bg-img: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16'%3e%3cpath fill='none' stroke='%23343a40' stroke-linecap='round' stroke-linejoin='round' stroke-width='2' d='m2 5 6 6 6-6'/%3e%3c/svg%3e\");\n  display: block;\n  width: 100%;\n  padding: 0.375rem 2.25rem 0.375rem 0.75rem;\n  font-size: 1rem;\n  font-weight: 400;\n  line-height: 1.5;\n  color: var(--bs-body-color);\n  appearance: none;\n  background-color: var(--bs-body-bg);\n  background-image: var(--bs-form-select-bg-img), var(--bs-form-select-bg-icon, none);\n  background-repeat: no-repeat;\n  background-position: right 0.75rem center;\n  background-size: 16px 12px;\n  border: var(--bs-border-width) solid var(--bs-border-color);\n  border-radius: var(--bs-border-radius);\n  transition: border-color 0.15s ease-in-out, box-shadow 0.15s ease-in-out;\n}\n@media (prefers-reduced-motion: reduce) {\n  .form-select {\n    transition: none;\n  }\n}\n.form-select:focus {\n  border-color: #86b7fe;\n  outline: 0;\n  box-shadow: 0 0 0 0.1875rem var(--pst-color-accent);\n}\n.form-select[multiple], .form-select[size]:not([size=\"1\"]) {\n  padding-right: 0.75rem;\n  background-image: none;\n}\n.form-select:disabled {\n  background-color: var(--bs-secondary-bg);\n}\n.form-select:-moz-focusring {\n  color: transparent;\n  text-shadow: 0 0 0 var(--bs-body-color);\n}\n\n.form-select-sm {\n  padding-top: 0.25rem;\n  padding-bottom: 0.25rem;\n  padding-left: 0.5rem;\n  font-size: 0.875rem;\n  border-radius: var(--bs-border-radius-sm);\n}\n\n.form-select-lg {\n  padding-top: 0.5rem;\n  padding-bottom: 0.5rem;\n  padding-left: 1rem;\n  font-size: 1.25rem;\n  border-radius: var(--bs-border-radius-lg);\n}\n\n[data-bs-theme=dark] .form-select {\n  --bs-form-select-bg-img: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16'%3e%3cpath fill='none' stroke='%23dee2e6' stroke-linecap='round' stroke-linejoin='round' stroke-width='2' d='m2 5 6 6 6-6'/%3e%3c/svg%3e\");\n}\n\n.form-check {\n  display: block;\n  min-height: 1.5rem;\n  padding-left: 1.5em;\n  margin-bottom: 0.125rem;\n}\n.form-check .form-check-input {\n  float: left;\n  margin-left: -1.5em;\n}\n\n.form-check-reverse {\n  padding-right: 1.5em;\n  padding-left: 0;\n  text-align: right;\n}\n.form-check-reverse .form-check-input {\n  float: right;\n  margin-right: -1.5em;\n  margin-left: 0;\n}\n\n.form-check-input {\n  --bs-form-check-bg: var(--bs-body-bg);\n  flex-shrink: 0;\n  width: 1em;\n  height: 1em;\n  margin-top: 0.25em;\n  vertical-align: top;\n  appearance: none;\n  background-color: var(--bs-form-check-bg);\n  background-image: var(--bs-form-check-bg-image);\n  background-repeat: no-repeat;\n  background-position: center;\n  background-size: contain;\n  border: var(--bs-border-width) solid var(--bs-border-color);\n  print-color-adjust: exact;\n}\n.form-check-input[type=checkbox] {\n  border-radius: 0.25em;\n}\n.form-check-input[type=radio] {\n  border-radius: 50%;\n}\n.form-check-input:active {\n  filter: brightness(90%);\n}\n.form-check-input:focus {\n  border-color: #86b7fe;\n  outline: 0;\n  box-shadow: 0 0 0 0.1875rem var(--pst-color-accent);\n}\n.form-check-input:checked {\n  background-color: #0d6efd;\n  border-color: #0d6efd;\n}\n.form-check-input:checked[type=checkbox] {\n  --bs-form-check-bg-image: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 20 20'%3e%3cpath fill='none' stroke='%23fff' stroke-linecap='round' stroke-linejoin='round' stroke-width='3' d='m6 10 3 3 6-6'/%3e%3c/svg%3e\");\n}\n.form-check-input:checked[type=radio] {\n  --bs-form-check-bg-image: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'%3e%3ccircle r='2' fill='%23fff'/%3e%3c/svg%3e\");\n}\n.form-check-input[type=checkbox]:indeterminate {\n  background-color: #0d6efd;\n  border-color: #0d6efd;\n  --bs-form-check-bg-image: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 20 20'%3e%3cpath fill='none' stroke='%23fff' stroke-linecap='round' stroke-linejoin='round' stroke-width='3' d='M6 10h8'/%3e%3c/svg%3e\");\n}\n.form-check-input:disabled {\n  pointer-events: none;\n  filter: none;\n  opacity: 0.5;\n}\n.form-check-input[disabled] ~ .form-check-label, .form-check-input:disabled ~ .form-check-label {\n  cursor: default;\n  opacity: 0.5;\n}\n\n.form-switch {\n  padding-left: 2.5em;\n}\n.form-switch .form-check-input {\n  --bs-form-switch-bg: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'%3e%3ccircle r='3' fill='rgba%280, 0, 0, 0.25%29'/%3e%3c/svg%3e\");\n  width: 2em;\n  margin-left: -2.5em;\n  background-image: var(--bs-form-switch-bg);\n  background-position: left center;\n  border-radius: 2em;\n  transition: background-position 0.15s ease-in-out;\n}\n@media (prefers-reduced-motion: reduce) {\n  .form-switch .form-check-input {\n    transition: none;\n  }\n}\n.form-switch .form-check-input:focus {\n  --bs-form-switch-bg: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'%3e%3ccircle r='3' fill='%2386b7fe'/%3e%3c/svg%3e\");\n}\n.form-switch .form-check-input:checked {\n  background-position: right center;\n  --bs-form-switch-bg: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'%3e%3ccircle r='3' fill='%23fff'/%3e%3c/svg%3e\");\n}\n.form-switch.form-check-reverse {\n  padding-right: 2.5em;\n  padding-left: 0;\n}\n.form-switch.form-check-reverse .form-check-input {\n  margin-right: -2.5em;\n  margin-left: 0;\n}\n\n.form-check-inline {\n  display: inline-block;\n  margin-right: 1rem;\n}\n\n.btn-check {\n  position: absolute;\n  clip: rect(0, 0, 0, 0);\n  pointer-events: none;\n}\n.btn-check[disabled] + .btn, .btn-check:disabled + .btn {\n  pointer-events: none;\n  filter: none;\n  opacity: 0.65;\n}\n\n[data-bs-theme=dark] .form-switch .form-check-input:not(:checked):not(:focus) {\n  --bs-form-switch-bg: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'%3e%3ccircle r='3' fill='rgba%28255, 255, 255, 0.25%29'/%3e%3c/svg%3e\");\n}\n\n.form-range {\n  width: 100%;\n  height: 1.375rem;\n  padding: 0;\n  appearance: none;\n  background-color: transparent;\n}\n.form-range:focus {\n  outline: 0;\n}\n.form-range:focus::-webkit-slider-thumb {\n  box-shadow: 0 0 0 1px #fff, 0 0 0 0.1875rem var(--pst-color-accent);\n}\n.form-range:focus::-moz-range-thumb {\n  box-shadow: 0 0 0 1px #fff, 0 0 0 0.1875rem var(--pst-color-accent);\n}\n.form-range::-moz-focus-outer {\n  border: 0;\n}\n.form-range::-webkit-slider-thumb {\n  width: 1rem;\n  height: 1rem;\n  margin-top: -0.25rem;\n  appearance: none;\n  background-color: #0d6efd;\n  border: 0;\n  border-radius: 1rem;\n  transition: background-color 0.15s ease-in-out, border-color 0.15s ease-in-out, box-shadow 0.15s ease-in-out;\n}\n@media (prefers-reduced-motion: reduce) {\n  .form-range::-webkit-slider-thumb {\n    transition: none;\n  }\n}\n.form-range::-webkit-slider-thumb:active {\n  background-color: #b6d4fe;\n}\n.form-range::-webkit-slider-runnable-track {\n  width: 100%;\n  height: 0.5rem;\n  color: transparent;\n  cursor: pointer;\n  background-color: var(--bs-secondary-bg);\n  border-color: transparent;\n  border-radius: 1rem;\n}\n.form-range::-moz-range-thumb {\n  width: 1rem;\n  height: 1rem;\n  appearance: none;\n  background-color: #0d6efd;\n  border: 0;\n  border-radius: 1rem;\n  transition: background-color 0.15s ease-in-out, border-color 0.15s ease-in-out, box-shadow 0.15s ease-in-out;\n}\n@media (prefers-reduced-motion: reduce) {\n  .form-range::-moz-range-thumb {\n    transition: none;\n  }\n}\n.form-range::-moz-range-thumb:active {\n  background-color: #b6d4fe;\n}\n.form-range::-moz-range-track {\n  width: 100%;\n  height: 0.5rem;\n  color: transparent;\n  cursor: pointer;\n  background-color: var(--bs-secondary-bg);\n  border-color: transparent;\n  border-radius: 1rem;\n}\n.form-range:disabled {\n  pointer-events: none;\n}\n.form-range:disabled::-webkit-slider-thumb {\n  background-color: var(--bs-secondary-color);\n}\n.form-range:disabled::-moz-range-thumb {\n  background-color: var(--bs-secondary-color);\n}\n\n.form-floating {\n  position: relative;\n}\n.form-floating > .form-control,\n.form-floating > .form-control-plaintext,\n.form-floating > .form-select {\n  height: calc(3.5rem + calc(var(--bs-border-width) * 2));\n  min-height: calc(3.5rem + calc(var(--bs-border-width) * 2));\n  line-height: 1.25;\n}\n.form-floating > label {\n  position: absolute;\n  top: 0;\n  left: 0;\n  z-index: 2;\n  height: 100%;\n  padding: 1rem 0.75rem;\n  overflow: hidden;\n  text-align: start;\n  text-overflow: ellipsis;\n  white-space: nowrap;\n  pointer-events: none;\n  border: var(--bs-border-width) solid transparent;\n  transform-origin: 0 0;\n  transition: opacity 0.1s ease-in-out, transform 0.1s ease-in-out;\n}\n@media (prefers-reduced-motion: reduce) {\n  .form-floating > label {\n    transition: none;\n  }\n}\n.form-floating > .form-control,\n.form-floating > .form-control-plaintext {\n  padding: 1rem 0.75rem;\n}\n.form-floating > .form-control::placeholder,\n.form-floating > .form-control-plaintext::placeholder {\n  color: transparent;\n}\n.form-floating > .form-control:focus, .form-floating > .form-control:not(:placeholder-shown),\n.form-floating > .form-control-plaintext:focus,\n.form-floating > .form-control-plaintext:not(:placeholder-shown) {\n  padding-top: 1.625rem;\n  padding-bottom: 0.625rem;\n}\n.form-floating > .form-control:-webkit-autofill,\n.form-floating > .form-control-plaintext:-webkit-autofill {\n  padding-top: 1.625rem;\n  padding-bottom: 0.625rem;\n}\n.form-floating > .form-select {\n  padding-top: 1.625rem;\n  padding-bottom: 0.625rem;\n}\n.form-floating > .form-control:focus ~ label,\n.form-floating > .form-control:not(:placeholder-shown) ~ label,\n.form-floating > .form-control-plaintext ~ label,\n.form-floating > .form-select ~ label {\n  color: rgba(var(--bs-body-color-rgb), 0.65);\n  transform: scale(0.85) translateY(-0.5rem) translateX(0.15rem);\n}\n.form-floating > .form-control:focus ~ label::after,\n.form-floating > .form-control:not(:placeholder-shown) ~ label::after,\n.form-floating > .form-control-plaintext ~ label::after,\n.form-floating > .form-select ~ label::after {\n  position: absolute;\n  inset: 1rem 0.375rem;\n  z-index: -1;\n  height: 1.5em;\n  content: \"\";\n  background-color: var(--bs-body-bg);\n  border-radius: var(--bs-border-radius);\n}\n.form-floating > .form-control:-webkit-autofill ~ label {\n  color: rgba(var(--bs-body-color-rgb), 0.65);\n  transform: scale(0.85) translateY(-0.5rem) translateX(0.15rem);\n}\n.form-floating > .form-control-plaintext ~ label {\n  border-width: var(--bs-border-width) 0;\n}\n.form-floating > :disabled ~ label,\n.form-floating > .form-control:disabled ~ label {\n  color: #6c757d;\n}\n.form-floating > :disabled ~ label::after,\n.form-floating > .form-control:disabled ~ label::after {\n  background-color: var(--bs-secondary-bg);\n}\n\n.input-group {\n  position: relative;\n  display: flex;\n  flex-wrap: wrap;\n  align-items: stretch;\n  width: 100%;\n}\n.input-group > .form-control,\n.input-group > .form-select,\n.input-group > .form-floating {\n  position: relative;\n  flex: 1 1 auto;\n  width: 1%;\n  min-width: 0;\n}\n.input-group > .form-control:focus,\n.input-group > .form-select:focus,\n.input-group > .form-floating:focus-within {\n  z-index: 5;\n}\n.input-group .btn {\n  position: relative;\n  z-index: 2;\n}\n.input-group .btn:focus {\n  z-index: 5;\n}\n\n.input-group-text {\n  display: flex;\n  align-items: center;\n  padding: 0.375rem 0.75rem;\n  font-size: 1rem;\n  font-weight: 400;\n  line-height: 1.5;\n  color: var(--bs-body-color);\n  text-align: center;\n  white-space: nowrap;\n  background-color: var(--bs-tertiary-bg);\n  border: var(--bs-border-width) solid var(--bs-border-color);\n  border-radius: var(--bs-border-radius);\n}\n\n.input-group-lg > .form-control,\n.input-group-lg > .form-select,\n.input-group-lg > .input-group-text,\n.input-group-lg > .btn {\n  padding: 0.5rem 1rem;\n  font-size: 1.25rem;\n  border-radius: var(--bs-border-radius-lg);\n}\n\n.input-group-sm > .form-control,\n.input-group-sm > .form-select,\n.input-group-sm > .input-group-text,\n.input-group-sm > .btn {\n  padding: 0.25rem 0.5rem;\n  font-size: 0.875rem;\n  border-radius: var(--bs-border-radius-sm);\n}\n\n.input-group-lg > .form-select,\n.input-group-sm > .form-select {\n  padding-right: 3rem;\n}\n\n.input-group:not(.has-validation) > :not(:last-child):not(.dropdown-toggle):not(.dropdown-menu):not(.form-floating),\n.input-group:not(.has-validation) > .dropdown-toggle:nth-last-child(n+3),\n.input-group:not(.has-validation) > .form-floating:not(:last-child) > .form-control,\n.input-group:not(.has-validation) > .form-floating:not(:last-child) > .form-select {\n  border-top-right-radius: 0;\n  border-bottom-right-radius: 0;\n}\n.input-group.has-validation > :nth-last-child(n+3):not(.dropdown-toggle):not(.dropdown-menu):not(.form-floating),\n.input-group.has-validation > .dropdown-toggle:nth-last-child(n+4),\n.input-group.has-validation > .form-floating:nth-last-child(n+3) > .form-control,\n.input-group.has-validation > .form-floating:nth-last-child(n+3) > .form-select {\n  border-top-right-radius: 0;\n  border-bottom-right-radius: 0;\n}\n.input-group > :not(:first-child):not(.dropdown-menu):not(.valid-tooltip):not(.valid-feedback):not(.invalid-tooltip):not(.invalid-feedback) {\n  margin-left: calc(var(--bs-border-width) * -1);\n  border-top-left-radius: 0;\n  border-bottom-left-radius: 0;\n}\n.input-group > .form-floating:not(:first-child) > .form-control,\n.input-group > .form-floating:not(:first-child) > .form-select {\n  border-top-left-radius: 0;\n  border-bottom-left-radius: 0;\n}\n\n.valid-feedback {\n  display: none;\n  width: 100%;\n  margin-top: 0.25rem;\n  font-size: 0.875em;\n  color: var(--bs-form-valid-color);\n}\n\n.valid-tooltip {\n  position: absolute;\n  top: 100%;\n  z-index: 5;\n  display: none;\n  max-width: 100%;\n  padding: 0.25rem 0.5rem;\n  margin-top: 0.1rem;\n  font-size: 0.875rem;\n  color: #fff;\n  background-color: var(--bs-success);\n  border-radius: var(--bs-border-radius);\n}\n\n.was-validated :valid ~ .valid-feedback,\n.was-validated :valid ~ .valid-tooltip,\n.is-valid ~ .valid-feedback,\n.is-valid ~ .valid-tooltip {\n  display: block;\n}\n\n.was-validated .form-control:valid, .form-control.is-valid {\n  border-color: var(--bs-form-valid-border-color);\n  padding-right: calc(1.5em + 0.75rem);\n  background-image: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 8 8'%3e%3cpath fill='%23198754' d='M2.3 6.73.6 4.53c-.4-1.04.46-1.4 1.1-.8l1.1 1.4 3.4-3.8c.6-.63 1.6-.27 1.2.7l-4 4.6c-.43.5-.8.4-1.1.1z'/%3e%3c/svg%3e\");\n  background-repeat: no-repeat;\n  background-position: right calc(0.375em + 0.1875rem) center;\n  background-size: calc(0.75em + 0.375rem) calc(0.75em + 0.375rem);\n}\n.was-validated .form-control:valid:focus, .form-control.is-valid:focus {\n  border-color: var(--bs-form-valid-border-color);\n  box-shadow: 0 0 0 0.1875rem rgba(var(--bs-success-rgb), 1);\n}\n\n.was-validated textarea.form-control:valid, textarea.form-control.is-valid {\n  padding-right: calc(1.5em + 0.75rem);\n  background-position: top calc(0.375em + 0.1875rem) right calc(0.375em + 0.1875rem);\n}\n\n.was-validated .form-select:valid, .form-select.is-valid {\n  border-color: var(--bs-form-valid-border-color);\n}\n.was-validated .form-select:valid:not([multiple]):not([size]), .was-validated .form-select:valid:not([multiple])[size=\"1\"], .form-select.is-valid:not([multiple]):not([size]), .form-select.is-valid:not([multiple])[size=\"1\"] {\n  --bs-form-select-bg-icon: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 8 8'%3e%3cpath fill='%23198754' d='M2.3 6.73.6 4.53c-.4-1.04.46-1.4 1.1-.8l1.1 1.4 3.4-3.8c.6-.63 1.6-.27 1.2.7l-4 4.6c-.43.5-.8.4-1.1.1z'/%3e%3c/svg%3e\");\n  padding-right: 4.125rem;\n  background-position: right 0.75rem center, center right 2.25rem;\n  background-size: 16px 12px, calc(0.75em + 0.375rem) calc(0.75em + 0.375rem);\n}\n.was-validated .form-select:valid:focus, .form-select.is-valid:focus {\n  border-color: var(--bs-form-valid-border-color);\n  box-shadow: 0 0 0 0.1875rem rgba(var(--bs-success-rgb), 1);\n}\n\n.was-validated .form-control-color:valid, .form-control-color.is-valid {\n  width: calc(3rem + calc(1.5em + 0.75rem));\n}\n\n.was-validated .form-check-input:valid, .form-check-input.is-valid {\n  border-color: var(--bs-form-valid-border-color);\n}\n.was-validated .form-check-input:valid:checked, .form-check-input.is-valid:checked {\n  background-color: var(--bs-form-valid-color);\n}\n.was-validated .form-check-input:valid:focus, .form-check-input.is-valid:focus {\n  box-shadow: 0 0 0 0.1875rem rgba(var(--bs-success-rgb), 1);\n}\n.was-validated .form-check-input:valid ~ .form-check-label, .form-check-input.is-valid ~ .form-check-label {\n  color: var(--bs-form-valid-color);\n}\n\n.form-check-inline .form-check-input ~ .valid-feedback {\n  margin-left: 0.5em;\n}\n\n.was-validated .input-group > .form-control:not(:focus):valid, .input-group > .form-control:not(:focus).is-valid,\n.was-validated .input-group > .form-select:not(:focus):valid,\n.input-group > .form-select:not(:focus).is-valid,\n.was-validated .input-group > .form-floating:not(:focus-within):valid,\n.input-group > .form-floating:not(:focus-within).is-valid {\n  z-index: 3;\n}\n\n.invalid-feedback {\n  display: none;\n  width: 100%;\n  margin-top: 0.25rem;\n  font-size: 0.875em;\n  color: var(--bs-form-invalid-color);\n}\n\n.invalid-tooltip {\n  position: absolute;\n  top: 100%;\n  z-index: 5;\n  display: none;\n  max-width: 100%;\n  padding: 0.25rem 0.5rem;\n  margin-top: 0.1rem;\n  font-size: 0.875rem;\n  color: #fff;\n  background-color: var(--bs-danger);\n  border-radius: var(--bs-border-radius);\n}\n\n.was-validated :invalid ~ .invalid-feedback,\n.was-validated :invalid ~ .invalid-tooltip,\n.is-invalid ~ .invalid-feedback,\n.is-invalid ~ .invalid-tooltip {\n  display: block;\n}\n\n.was-validated .form-control:invalid, .form-control.is-invalid {\n  border-color: var(--bs-form-invalid-border-color);\n  padding-right: calc(1.5em + 0.75rem);\n  background-image: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 12 12' width='12' height='12' fill='none' stroke='%23dc3545'%3e%3ccircle cx='6' cy='6' r='4.5'/%3e%3cpath stroke-linejoin='round' d='M5.8 3.6h.4L6 6.5z'/%3e%3ccircle cx='6' cy='8.2' r='.6' fill='%23dc3545' stroke='none'/%3e%3c/svg%3e\");\n  background-repeat: no-repeat;\n  background-position: right calc(0.375em + 0.1875rem) center;\n  background-size: calc(0.75em + 0.375rem) calc(0.75em + 0.375rem);\n}\n.was-validated .form-control:invalid:focus, .form-control.is-invalid:focus {\n  border-color: var(--bs-form-invalid-border-color);\n  box-shadow: 0 0 0 0.1875rem rgba(var(--bs-danger-rgb), 1);\n}\n\n.was-validated textarea.form-control:invalid, textarea.form-control.is-invalid {\n  padding-right: calc(1.5em + 0.75rem);\n  background-position: top calc(0.375em + 0.1875rem) right calc(0.375em + 0.1875rem);\n}\n\n.was-validated .form-select:invalid, .form-select.is-invalid {\n  border-color: var(--bs-form-invalid-border-color);\n}\n.was-validated .form-select:invalid:not([multiple]):not([size]), .was-validated .form-select:invalid:not([multiple])[size=\"1\"], .form-select.is-invalid:not([multiple]):not([size]), .form-select.is-invalid:not([multiple])[size=\"1\"] {\n  --bs-form-select-bg-icon: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 12 12' width='12' height='12' fill='none' stroke='%23dc3545'%3e%3ccircle cx='6' cy='6' r='4.5'/%3e%3cpath stroke-linejoin='round' d='M5.8 3.6h.4L6 6.5z'/%3e%3ccircle cx='6' cy='8.2' r='.6' fill='%23dc3545' stroke='none'/%3e%3c/svg%3e\");\n  padding-right: 4.125rem;\n  background-position: right 0.75rem center, center right 2.25rem;\n  background-size: 16px 12px, calc(0.75em + 0.375rem) calc(0.75em + 0.375rem);\n}\n.was-validated .form-select:invalid:focus, .form-select.is-invalid:focus {\n  border-color: var(--bs-form-invalid-border-color);\n  box-shadow: 0 0 0 0.1875rem rgba(var(--bs-danger-rgb), 1);\n}\n\n.was-validated .form-control-color:invalid, .form-control-color.is-invalid {\n  width: calc(3rem + calc(1.5em + 0.75rem));\n}\n\n.was-validated .form-check-input:invalid, .form-check-input.is-invalid {\n  border-color: var(--bs-form-invalid-border-color);\n}\n.was-validated .form-check-input:invalid:checked, .form-check-input.is-invalid:checked {\n  background-color: var(--bs-form-invalid-color);\n}\n.was-validated .form-check-input:invalid:focus, .form-check-input.is-invalid:focus {\n  box-shadow: 0 0 0 0.1875rem rgba(var(--bs-danger-rgb), 1);\n}\n.was-validated .form-check-input:invalid ~ .form-check-label, .form-check-input.is-invalid ~ .form-check-label {\n  color: var(--bs-form-invalid-color);\n}\n\n.form-check-inline .form-check-input ~ .invalid-feedback {\n  margin-left: 0.5em;\n}\n\n.was-validated .input-group > .form-control:not(:focus):invalid, .input-group > .form-control:not(:focus).is-invalid,\n.was-validated .input-group > .form-select:not(:focus):invalid,\n.input-group > .form-select:not(:focus).is-invalid,\n.was-validated .input-group > .form-floating:not(:focus-within):invalid,\n.input-group > .form-floating:not(:focus-within).is-invalid {\n  z-index: 4;\n}\n\n.btn {\n  --bs-btn-padding-x: 0.75rem;\n  --bs-btn-padding-y: 0.375rem;\n  --bs-btn-font-family: ;\n  --bs-btn-font-size: 1rem;\n  --bs-btn-font-weight: 400;\n  --bs-btn-line-height: 1.5;\n  --bs-btn-color: var(--bs-body-color);\n  --bs-btn-bg: transparent;\n  --bs-btn-border-width: var(--bs-border-width);\n  --bs-btn-border-color: transparent;\n  --bs-btn-border-radius: var(--bs-border-radius);\n  --bs-btn-hover-border-color: transparent;\n  --bs-btn-box-shadow: inset 0 1px 0 rgba(255, 255, 255, 0.15), 0 1px 1px rgba(0, 0, 0, 0.075);\n  --bs-btn-disabled-opacity: 0.65;\n  --bs-btn-focus-box-shadow: 0 0 0 0.1875rem rgba(var(--bs-btn-focus-shadow-rgb), .5);\n  display: inline-block;\n  padding: var(--bs-btn-padding-y) var(--bs-btn-padding-x);\n  font-family: var(--bs-btn-font-family);\n  font-size: var(--bs-btn-font-size);\n  font-weight: var(--bs-btn-font-weight);\n  line-height: var(--bs-btn-line-height);\n  color: var(--bs-btn-color);\n  text-align: center;\n  text-decoration: none;\n  vertical-align: middle;\n  cursor: pointer;\n  user-select: none;\n  border: var(--bs-btn-border-width) solid var(--bs-btn-border-color);\n  border-radius: var(--bs-btn-border-radius);\n  background-color: var(--bs-btn-bg);\n  transition: color 0.15s ease-in-out, background-color 0.15s ease-in-out, border-color 0.15s ease-in-out, box-shadow 0.15s ease-in-out;\n}\n@media (prefers-reduced-motion: reduce) {\n  .btn {\n    transition: none;\n  }\n}\n.btn:hover {\n  color: var(--bs-btn-hover-color);\n  background-color: var(--bs-btn-hover-bg);\n  border-color: var(--bs-btn-hover-border-color);\n}\n.btn-check + .btn:hover {\n  color: var(--bs-btn-color);\n  background-color: var(--bs-btn-bg);\n  border-color: var(--bs-btn-border-color);\n}\n.btn:focus-visible {\n  color: var(--bs-btn-hover-color);\n  background-color: var(--bs-btn-hover-bg);\n  border-color: var(--bs-btn-hover-border-color);\n  outline: 0;\n  box-shadow: var(--bs-btn-focus-box-shadow);\n}\n.btn-check:focus-visible + .btn {\n  border-color: var(--bs-btn-hover-border-color);\n  outline: 0;\n  box-shadow: var(--bs-btn-focus-box-shadow);\n}\n.btn-check:checked + .btn, :not(.btn-check) + .btn:active, .btn:first-child:active, .btn.active, .btn.show {\n  color: var(--bs-btn-active-color);\n  background-color: var(--bs-btn-active-bg);\n  border-color: var(--bs-btn-active-border-color);\n}\n.btn-check:checked + .btn:focus-visible, :not(.btn-check) + .btn:active:focus-visible, .btn:first-child:active:focus-visible, .btn.active:focus-visible, .btn.show:focus-visible {\n  box-shadow: var(--bs-btn-focus-box-shadow);\n}\n.btn-check:checked:focus-visible + .btn {\n  box-shadow: var(--bs-btn-focus-box-shadow);\n}\n.btn:disabled, .btn.disabled, fieldset:disabled .btn {\n  color: var(--bs-btn-disabled-color);\n  pointer-events: none;\n  background-color: var(--bs-btn-disabled-bg);\n  border-color: var(--bs-btn-disabled-border-color);\n  opacity: var(--bs-btn-disabled-opacity);\n}\n\n.btn-primary {\n  --bs-btn-color: #fff;\n  --bs-btn-bg: #0d6efd;\n  --bs-btn-border-color: #0d6efd;\n  --bs-btn-hover-color: #fff;\n  --bs-btn-hover-bg: #0b5ed7;\n  --bs-btn-hover-border-color: #0a58ca;\n  --bs-btn-focus-shadow-rgb: 49, 132, 253;\n  --bs-btn-active-color: #fff;\n  --bs-btn-active-bg: #0a58ca;\n  --bs-btn-active-border-color: #0a53be;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #fff;\n  --bs-btn-disabled-bg: #0d6efd;\n  --bs-btn-disabled-border-color: #0d6efd;\n}\n\n.btn-secondary {\n  --bs-btn-color: #fff;\n  --bs-btn-bg: #6c757d;\n  --bs-btn-border-color: #6c757d;\n  --bs-btn-hover-color: #fff;\n  --bs-btn-hover-bg: #5c636a;\n  --bs-btn-hover-border-color: #565e64;\n  --bs-btn-focus-shadow-rgb: 130, 138, 145;\n  --bs-btn-active-color: #fff;\n  --bs-btn-active-bg: #565e64;\n  --bs-btn-active-border-color: #51585e;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #fff;\n  --bs-btn-disabled-bg: #6c757d;\n  --bs-btn-disabled-border-color: #6c757d;\n}\n\n.btn-success {\n  --bs-btn-color: #fff;\n  --bs-btn-bg: #198754;\n  --bs-btn-border-color: #198754;\n  --bs-btn-hover-color: #fff;\n  --bs-btn-hover-bg: #157347;\n  --bs-btn-hover-border-color: #146c43;\n  --bs-btn-focus-shadow-rgb: 60, 153, 110;\n  --bs-btn-active-color: #fff;\n  --bs-btn-active-bg: #146c43;\n  --bs-btn-active-border-color: #13653f;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #fff;\n  --bs-btn-disabled-bg: #198754;\n  --bs-btn-disabled-border-color: #198754;\n}\n\n.btn-info {\n  --bs-btn-color: #000;\n  --bs-btn-bg: #0dcaf0;\n  --bs-btn-border-color: #0dcaf0;\n  --bs-btn-hover-color: #000;\n  --bs-btn-hover-bg: #31d2f2;\n  --bs-btn-hover-border-color: #25cff2;\n  --bs-btn-focus-shadow-rgb: 11, 172, 204;\n  --bs-btn-active-color: #000;\n  --bs-btn-active-bg: #3dd5f3;\n  --bs-btn-active-border-color: #25cff2;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #000;\n  --bs-btn-disabled-bg: #0dcaf0;\n  --bs-btn-disabled-border-color: #0dcaf0;\n}\n\n.btn-warning {\n  --bs-btn-color: #000;\n  --bs-btn-bg: #ffc107;\n  --bs-btn-border-color: #ffc107;\n  --bs-btn-hover-color: #000;\n  --bs-btn-hover-bg: #ffca2c;\n  --bs-btn-hover-border-color: #ffc720;\n  --bs-btn-focus-shadow-rgb: 217, 164, 6;\n  --bs-btn-active-color: #000;\n  --bs-btn-active-bg: #ffcd39;\n  --bs-btn-active-border-color: #ffc720;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #000;\n  --bs-btn-disabled-bg: #ffc107;\n  --bs-btn-disabled-border-color: #ffc107;\n}\n\n.btn-danger {\n  --bs-btn-color: #fff;\n  --bs-btn-bg: #dc3545;\n  --bs-btn-border-color: #dc3545;\n  --bs-btn-hover-color: #fff;\n  --bs-btn-hover-bg: #bb2d3b;\n  --bs-btn-hover-border-color: #b02a37;\n  --bs-btn-focus-shadow-rgb: 225, 83, 97;\n  --bs-btn-active-color: #fff;\n  --bs-btn-active-bg: #b02a37;\n  --bs-btn-active-border-color: #a52834;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #fff;\n  --bs-btn-disabled-bg: #dc3545;\n  --bs-btn-disabled-border-color: #dc3545;\n}\n\n.btn-light {\n  --bs-btn-color: #000;\n  --bs-btn-bg: #f8f9fa;\n  --bs-btn-border-color: #f8f9fa;\n  --bs-btn-hover-color: #000;\n  --bs-btn-hover-bg: #d3d4d5;\n  --bs-btn-hover-border-color: #c6c7c8;\n  --bs-btn-focus-shadow-rgb: 211, 212, 213;\n  --bs-btn-active-color: #000;\n  --bs-btn-active-bg: #c6c7c8;\n  --bs-btn-active-border-color: #babbbc;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #000;\n  --bs-btn-disabled-bg: #f8f9fa;\n  --bs-btn-disabled-border-color: #f8f9fa;\n}\n\n.btn-dark {\n  --bs-btn-color: #fff;\n  --bs-btn-bg: #212529;\n  --bs-btn-border-color: #212529;\n  --bs-btn-hover-color: #fff;\n  --bs-btn-hover-bg: #424649;\n  --bs-btn-hover-border-color: #373b3e;\n  --bs-btn-focus-shadow-rgb: 66, 70, 73;\n  --bs-btn-active-color: #fff;\n  --bs-btn-active-bg: #4d5154;\n  --bs-btn-active-border-color: #373b3e;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #fff;\n  --bs-btn-disabled-bg: #212529;\n  --bs-btn-disabled-border-color: #212529;\n}\n\n.btn-outline-primary {\n  --bs-btn-color: #0d6efd;\n  --bs-btn-border-color: #0d6efd;\n  --bs-btn-hover-color: #fff;\n  --bs-btn-hover-bg: #0d6efd;\n  --bs-btn-hover-border-color: #0d6efd;\n  --bs-btn-focus-shadow-rgb: 13, 110, 253;\n  --bs-btn-active-color: #fff;\n  --bs-btn-active-bg: #0d6efd;\n  --bs-btn-active-border-color: #0d6efd;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #0d6efd;\n  --bs-btn-disabled-bg: transparent;\n  --bs-btn-disabled-border-color: #0d6efd;\n  --bs-gradient: none;\n}\n\n.btn-outline-secondary {\n  --bs-btn-color: #6c757d;\n  --bs-btn-border-color: #6c757d;\n  --bs-btn-hover-color: #fff;\n  --bs-btn-hover-bg: #6c757d;\n  --bs-btn-hover-border-color: #6c757d;\n  --bs-btn-focus-shadow-rgb: 108, 117, 125;\n  --bs-btn-active-color: #fff;\n  --bs-btn-active-bg: #6c757d;\n  --bs-btn-active-border-color: #6c757d;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #6c757d;\n  --bs-btn-disabled-bg: transparent;\n  --bs-btn-disabled-border-color: #6c757d;\n  --bs-gradient: none;\n}\n\n.btn-outline-success {\n  --bs-btn-color: #198754;\n  --bs-btn-border-color: #198754;\n  --bs-btn-hover-color: #fff;\n  --bs-btn-hover-bg: #198754;\n  --bs-btn-hover-border-color: #198754;\n  --bs-btn-focus-shadow-rgb: 25, 135, 84;\n  --bs-btn-active-color: #fff;\n  --bs-btn-active-bg: #198754;\n  --bs-btn-active-border-color: #198754;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #198754;\n  --bs-btn-disabled-bg: transparent;\n  --bs-btn-disabled-border-color: #198754;\n  --bs-gradient: none;\n}\n\n.btn-outline-info {\n  --bs-btn-color: #0dcaf0;\n  --bs-btn-border-color: #0dcaf0;\n  --bs-btn-hover-color: #000;\n  --bs-btn-hover-bg: #0dcaf0;\n  --bs-btn-hover-border-color: #0dcaf0;\n  --bs-btn-focus-shadow-rgb: 13, 202, 240;\n  --bs-btn-active-color: #000;\n  --bs-btn-active-bg: #0dcaf0;\n  --bs-btn-active-border-color: #0dcaf0;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #0dcaf0;\n  --bs-btn-disabled-bg: transparent;\n  --bs-btn-disabled-border-color: #0dcaf0;\n  --bs-gradient: none;\n}\n\n.btn-outline-warning {\n  --bs-btn-color: #ffc107;\n  --bs-btn-border-color: #ffc107;\n  --bs-btn-hover-color: #000;\n  --bs-btn-hover-bg: #ffc107;\n  --bs-btn-hover-border-color: #ffc107;\n  --bs-btn-focus-shadow-rgb: 255, 193, 7;\n  --bs-btn-active-color: #000;\n  --bs-btn-active-bg: #ffc107;\n  --bs-btn-active-border-color: #ffc107;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #ffc107;\n  --bs-btn-disabled-bg: transparent;\n  --bs-btn-disabled-border-color: #ffc107;\n  --bs-gradient: none;\n}\n\n.btn-outline-danger {\n  --bs-btn-color: #dc3545;\n  --bs-btn-border-color: #dc3545;\n  --bs-btn-hover-color: #fff;\n  --bs-btn-hover-bg: #dc3545;\n  --bs-btn-hover-border-color: #dc3545;\n  --bs-btn-focus-shadow-rgb: 220, 53, 69;\n  --bs-btn-active-color: #fff;\n  --bs-btn-active-bg: #dc3545;\n  --bs-btn-active-border-color: #dc3545;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #dc3545;\n  --bs-btn-disabled-bg: transparent;\n  --bs-btn-disabled-border-color: #dc3545;\n  --bs-gradient: none;\n}\n\n.btn-outline-light {\n  --bs-btn-color: #f8f9fa;\n  --bs-btn-border-color: #f8f9fa;\n  --bs-btn-hover-color: #000;\n  --bs-btn-hover-bg: #f8f9fa;\n  --bs-btn-hover-border-color: #f8f9fa;\n  --bs-btn-focus-shadow-rgb: 248, 249, 250;\n  --bs-btn-active-color: #000;\n  --bs-btn-active-bg: #f8f9fa;\n  --bs-btn-active-border-color: #f8f9fa;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #f8f9fa;\n  --bs-btn-disabled-bg: transparent;\n  --bs-btn-disabled-border-color: #f8f9fa;\n  --bs-gradient: none;\n}\n\n.btn-outline-dark {\n  --bs-btn-color: #212529;\n  --bs-btn-border-color: #212529;\n  --bs-btn-hover-color: #fff;\n  --bs-btn-hover-bg: #212529;\n  --bs-btn-hover-border-color: #212529;\n  --bs-btn-focus-shadow-rgb: 33, 37, 41;\n  --bs-btn-active-color: #fff;\n  --bs-btn-active-bg: #212529;\n  --bs-btn-active-border-color: #212529;\n  --bs-btn-active-shadow: inset 0 3px 5px rgba(0, 0, 0, 0.125);\n  --bs-btn-disabled-color: #212529;\n  --bs-btn-disabled-bg: transparent;\n  --bs-btn-disabled-border-color: #212529;\n  --bs-gradient: none;\n}\n\n.btn-link {\n  --bs-btn-font-weight: 400;\n  --bs-btn-color: var(--bs-link-color);\n  --bs-btn-bg: transparent;\n  --bs-btn-border-color: transparent;\n  --bs-btn-hover-color: var(--bs-link-hover-color);\n  --bs-btn-hover-border-color: transparent;\n  --bs-btn-active-color: var(--bs-link-hover-color);\n  --bs-btn-active-border-color: transparent;\n  --bs-btn-disabled-color: #6c757d;\n  --bs-btn-disabled-border-color: transparent;\n  --bs-btn-box-shadow: 0 0 0 #000;\n  --bs-btn-focus-shadow-rgb: 49, 132, 253;\n  text-decoration: underline;\n}\n.btn-link:focus-visible {\n  color: var(--bs-btn-color);\n}\n.btn-link:hover {\n  color: var(--bs-btn-hover-color);\n}\n\n.btn-lg, .btn-group-lg > .btn {\n  --bs-btn-padding-y: 0.5rem;\n  --bs-btn-padding-x: 1rem;\n  --bs-btn-font-size: 1.25rem;\n  --bs-btn-border-radius: var(--bs-border-radius-lg);\n}\n\n.btn-sm, .btn-group-sm > .btn {\n  --bs-btn-padding-y: 0.25rem;\n  --bs-btn-padding-x: 0.5rem;\n  --bs-btn-font-size: 0.875rem;\n  --bs-btn-border-radius: var(--bs-border-radius-sm);\n}\n\n.fade {\n  transition: opacity 0.15s linear;\n}\n@media (prefers-reduced-motion: reduce) {\n  .fade {\n    transition: none;\n  }\n}\n.fade:not(.show) {\n  opacity: 0;\n}\n\n.collapse:not(.show) {\n  display: none;\n}\n\n.collapsing {\n  height: 0;\n  overflow: hidden;\n  transition: height 0.35s ease;\n}\n@media (prefers-reduced-motion: reduce) {\n  .collapsing {\n    transition: none;\n  }\n}\n.collapsing.collapse-horizontal {\n  width: 0;\n  height: auto;\n  transition: width 0.35s ease;\n}\n@media (prefers-reduced-motion: reduce) {\n  .collapsing.collapse-horizontal {\n    transition: none;\n  }\n}\n\n.dropup,\n.dropend,\n.dropdown,\n.dropstart,\n.dropup-center,\n.dropdown-center {\n  position: relative;\n}\n\n.dropdown-toggle {\n  white-space: nowrap;\n}\n.dropdown-toggle::after {\n  display: inline-block;\n  margin-left: 0.255em;\n  vertical-align: 0.255em;\n  content: \"\";\n  border-top: 0.3em solid;\n  border-right: 0.3em solid transparent;\n  border-bottom: 0;\n  border-left: 0.3em solid transparent;\n}\n.dropdown-toggle:empty::after {\n  margin-left: 0;\n}\n\n.dropdown-menu {\n  --bs-dropdown-zindex: 1000;\n  --bs-dropdown-min-width: 10rem;\n  --bs-dropdown-padding-x: 0;\n  --bs-dropdown-padding-y: 0.5rem;\n  --bs-dropdown-spacer: 0.125rem;\n  --bs-dropdown-font-size: 1rem;\n  --bs-dropdown-color: var(--bs-body-color);\n  --bs-dropdown-bg: var(--bs-body-bg);\n  --bs-dropdown-border-color: var(--bs-border-color-translucent);\n  --bs-dropdown-border-radius: var(--bs-border-radius);\n  --bs-dropdown-border-width: var(--bs-border-width);\n  --bs-dropdown-inner-border-radius: calc(var(--bs-border-radius) - var(--bs-border-width));\n  --bs-dropdown-divider-bg: var(--bs-border-color-translucent);\n  --bs-dropdown-divider-margin-y: 0.5rem;\n  --bs-dropdown-box-shadow: var(--bs-box-shadow);\n  --bs-dropdown-link-color: var(--bs-body-color);\n  --bs-dropdown-link-hover-color: var(--bs-body-color);\n  --bs-dropdown-link-hover-bg: var(--pst-color-surface);\n  --bs-dropdown-link-active-color: #fff;\n  --bs-dropdown-link-active-bg: var(--pst-color-surface);\n  --bs-dropdown-link-disabled-color: var(--bs-tertiary-color);\n  --bs-dropdown-item-padding-x: 1rem;\n  --bs-dropdown-item-padding-y: 0.25rem;\n  --bs-dropdown-header-color: #6c757d;\n  --bs-dropdown-header-padding-x: 1rem;\n  --bs-dropdown-header-padding-y: 0.5rem;\n  position: absolute;\n  z-index: var(--bs-dropdown-zindex);\n  display: none;\n  min-width: var(--bs-dropdown-min-width);\n  padding: var(--bs-dropdown-padding-y) var(--bs-dropdown-padding-x);\n  margin: 0;\n  font-size: var(--bs-dropdown-font-size);\n  color: var(--bs-dropdown-color);\n  text-align: left;\n  list-style: none;\n  background-color: var(--bs-dropdown-bg);\n  background-clip: padding-box;\n  border: var(--bs-dropdown-border-width) solid var(--bs-dropdown-border-color);\n  border-radius: var(--bs-dropdown-border-radius);\n}\n.dropdown-menu[data-bs-popper] {\n  top: 100%;\n  left: 0;\n  margin-top: var(--bs-dropdown-spacer);\n}\n\n.dropdown-menu-start {\n  --bs-position: start;\n}\n.dropdown-menu-start[data-bs-popper] {\n  right: auto;\n  left: 0;\n}\n\n.dropdown-menu-end {\n  --bs-position: end;\n}\n.dropdown-menu-end[data-bs-popper] {\n  right: 0;\n  left: auto;\n}\n\n@media (min-width: 540px) {\n  .dropdown-menu-sm-start {\n    --bs-position: start;\n  }\n  .dropdown-menu-sm-start[data-bs-popper] {\n    right: auto;\n    left: 0;\n  }\n  .dropdown-menu-sm-end {\n    --bs-position: end;\n  }\n  .dropdown-menu-sm-end[data-bs-popper] {\n    right: 0;\n    left: auto;\n  }\n}\n@media (min-width: 720px) {\n  .dropdown-menu-md-start {\n    --bs-position: start;\n  }\n  .dropdown-menu-md-start[data-bs-popper] {\n    right: auto;\n    left: 0;\n  }\n  .dropdown-menu-md-end {\n    --bs-position: end;\n  }\n  .dropdown-menu-md-end[data-bs-popper] {\n    right: 0;\n    left: auto;\n  }\n}\n@media (min-width: 960px) {\n  .dropdown-menu-lg-start {\n    --bs-position: start;\n  }\n  .dropdown-menu-lg-start[data-bs-popper] {\n    right: auto;\n    left: 0;\n  }\n  .dropdown-menu-lg-end {\n    --bs-position: end;\n  }\n  .dropdown-menu-lg-end[data-bs-popper] {\n    right: 0;\n    left: auto;\n  }\n}\n@media (min-width: 1200px) {\n  .dropdown-menu-xl-start {\n    --bs-position: start;\n  }\n  .dropdown-menu-xl-start[data-bs-popper] {\n    right: auto;\n    left: 0;\n  }\n  .dropdown-menu-xl-end {\n    --bs-position: end;\n  }\n  .dropdown-menu-xl-end[data-bs-popper] {\n    right: 0;\n    left: auto;\n  }\n}\n.dropup .dropdown-menu[data-bs-popper] {\n  top: auto;\n  bottom: 100%;\n  margin-top: 0;\n  margin-bottom: var(--bs-dropdown-spacer);\n}\n.dropup .dropdown-toggle::after {\n  display: inline-block;\n  margin-left: 0.255em;\n  vertical-align: 0.255em;\n  content: \"\";\n  border-top: 0;\n  border-right: 0.3em solid transparent;\n  border-bottom: 0.3em solid;\n  border-left: 0.3em solid transparent;\n}\n.dropup .dropdown-toggle:empty::after {\n  margin-left: 0;\n}\n\n.dropend .dropdown-menu[data-bs-popper] {\n  top: 0;\n  right: auto;\n  left: 100%;\n  margin-top: 0;\n  margin-left: var(--bs-dropdown-spacer);\n}\n.dropend .dropdown-toggle::after {\n  display: inline-block;\n  margin-left: 0.255em;\n  vertical-align: 0.255em;\n  content: \"\";\n  border-top: 0.3em solid transparent;\n  border-right: 0;\n  border-bottom: 0.3em solid transparent;\n  border-left: 0.3em solid;\n}\n.dropend .dropdown-toggle:empty::after {\n  margin-left: 0;\n}\n.dropend .dropdown-toggle::after {\n  vertical-align: 0;\n}\n\n.dropstart .dropdown-menu[data-bs-popper] {\n  top: 0;\n  right: 100%;\n  left: auto;\n  margin-top: 0;\n  margin-right: var(--bs-dropdown-spacer);\n}\n.dropstart .dropdown-toggle::after {\n  display: inline-block;\n  margin-left: 0.255em;\n  vertical-align: 0.255em;\n  content: \"\";\n}\n.dropstart .dropdown-toggle::after {\n  display: none;\n}\n.dropstart .dropdown-toggle::before {\n  display: inline-block;\n  margin-right: 0.255em;\n  vertical-align: 0.255em;\n  content: \"\";\n  border-top: 0.3em solid transparent;\n  border-right: 0.3em solid;\n  border-bottom: 0.3em solid transparent;\n}\n.dropstart .dropdown-toggle:empty::after {\n  margin-left: 0;\n}\n.dropstart .dropdown-toggle::before {\n  vertical-align: 0;\n}\n\n.dropdown-divider {\n  height: 0;\n  margin: var(--bs-dropdown-divider-margin-y) 0;\n  overflow: hidden;\n  border-top: 1px solid var(--bs-dropdown-divider-bg);\n  opacity: 1;\n}\n\n.dropdown-item {\n  display: block;\n  width: 100%;\n  padding: var(--bs-dropdown-item-padding-y) var(--bs-dropdown-item-padding-x);\n  clear: both;\n  font-weight: 400;\n  color: var(--bs-dropdown-link-color);\n  text-align: inherit;\n  text-decoration: none;\n  white-space: nowrap;\n  background-color: transparent;\n  border: 0;\n  border-radius: var(--bs-dropdown-item-border-radius, 0);\n}\n.dropdown-item:hover, .dropdown-item:focus {\n  color: var(--bs-dropdown-link-hover-color);\n  background-color: var(--bs-dropdown-link-hover-bg);\n}\n.dropdown-item.active, .dropdown-item:active {\n  color: var(--bs-dropdown-link-active-color);\n  text-decoration: none;\n  background-color: var(--bs-dropdown-link-active-bg);\n}\n.dropdown-item.disabled, .dropdown-item:disabled {\n  color: var(--bs-dropdown-link-disabled-color);\n  pointer-events: none;\n  background-color: transparent;\n}\n\n.dropdown-menu.show {\n  display: block;\n}\n\n.dropdown-header {\n  display: block;\n  padding: var(--bs-dropdown-header-padding-y) var(--bs-dropdown-header-padding-x);\n  margin-bottom: 0;\n  font-size: 0.875rem;\n  color: var(--bs-dropdown-header-color);\n  white-space: nowrap;\n}\n\n.dropdown-item-text {\n  display: block;\n  padding: var(--bs-dropdown-item-padding-y) var(--bs-dropdown-item-padding-x);\n  color: var(--bs-dropdown-link-color);\n}\n\n.dropdown-menu-dark {\n  --bs-dropdown-color: #dee2e6;\n  --bs-dropdown-bg: #343a40;\n  --bs-dropdown-border-color: var(--bs-border-color-translucent);\n  --bs-dropdown-box-shadow: ;\n  --bs-dropdown-link-color: #dee2e6;\n  --bs-dropdown-link-hover-color: #fff;\n  --bs-dropdown-divider-bg: var(--bs-border-color-translucent);\n  --bs-dropdown-link-hover-bg: var(--pst-color-surface);\n  --bs-dropdown-link-active-color: #fff;\n  --bs-dropdown-link-active-bg: var(--pst-color-surface);\n  --bs-dropdown-link-disabled-color: #adb5bd;\n  --bs-dropdown-header-color: #adb5bd;\n}\n\n.btn-group,\n.btn-group-vertical {\n  position: relative;\n  display: inline-flex;\n  vertical-align: middle;\n}\n.btn-group > .btn,\n.btn-group-vertical > .btn {\n  position: relative;\n  flex: 1 1 auto;\n}\n.btn-group > .btn-check:checked + .btn,\n.btn-group > .btn-check:focus + .btn,\n.btn-group > .btn:hover,\n.btn-group > .btn:focus,\n.btn-group > .btn:active,\n.btn-group > .btn.active,\n.btn-group-vertical > .btn-check:checked + .btn,\n.btn-group-vertical > .btn-check:focus + .btn,\n.btn-group-vertical > .btn:hover,\n.btn-group-vertical > .btn:focus,\n.btn-group-vertical > .btn:active,\n.btn-group-vertical > .btn.active {\n  z-index: 1;\n}\n\n.btn-toolbar {\n  display: flex;\n  flex-wrap: wrap;\n  justify-content: flex-start;\n}\n.btn-toolbar .input-group {\n  width: auto;\n}\n\n.btn-group {\n  border-radius: var(--bs-border-radius);\n}\n.btn-group > :not(.btn-check:first-child) + .btn,\n.btn-group > .btn-group:not(:first-child) {\n  margin-left: calc(var(--bs-border-width) * -1);\n}\n.btn-group > .btn:not(:last-child):not(.dropdown-toggle),\n.btn-group > .btn.dropdown-toggle-split:first-child,\n.btn-group > .btn-group:not(:last-child) > .btn {\n  border-top-right-radius: 0;\n  border-bottom-right-radius: 0;\n}\n.btn-group > .btn:nth-child(n+3),\n.btn-group > :not(.btn-check) + .btn,\n.btn-group > .btn-group:not(:first-child) > .btn {\n  border-top-left-radius: 0;\n  border-bottom-left-radius: 0;\n}\n\n.dropdown-toggle-split {\n  padding-right: 0.5625rem;\n  padding-left: 0.5625rem;\n}\n.dropdown-toggle-split::after, .dropup .dropdown-toggle-split::after, .dropend .dropdown-toggle-split::after {\n  margin-left: 0;\n}\n.dropstart .dropdown-toggle-split::before {\n  margin-right: 0;\n}\n\n.btn-sm + .dropdown-toggle-split, .btn-group-sm > .btn + .dropdown-toggle-split {\n  padding-right: 0.375rem;\n  padding-left: 0.375rem;\n}\n\n.btn-lg + .dropdown-toggle-split, .btn-group-lg > .btn + .dropdown-toggle-split {\n  padding-right: 0.75rem;\n  padding-left: 0.75rem;\n}\n\n.btn-group-vertical {\n  flex-direction: column;\n  align-items: flex-start;\n  justify-content: center;\n}\n.btn-group-vertical > .btn,\n.btn-group-vertical > .btn-group {\n  width: 100%;\n}\n.btn-group-vertical > .btn:not(:first-child),\n.btn-group-vertical > .btn-group:not(:first-child) {\n  margin-top: calc(var(--bs-border-width) * -1);\n}\n.btn-group-vertical > .btn:not(:last-child):not(.dropdown-toggle),\n.btn-group-vertical > .btn-group:not(:last-child) > .btn {\n  border-bottom-right-radius: 0;\n  border-bottom-left-radius: 0;\n}\n.btn-group-vertical > .btn ~ .btn,\n.btn-group-vertical > .btn-group:not(:first-child) > .btn {\n  border-top-left-radius: 0;\n  border-top-right-radius: 0;\n}\n\n.nav {\n  --bs-nav-link-padding-x: 1rem;\n  --bs-nav-link-padding-y: 0.5rem;\n  --bs-nav-link-font-weight: ;\n  --bs-nav-link-color: var(--bs-link-color);\n  --bs-nav-link-hover-color: var(--bs-link-hover-color);\n  --bs-nav-link-disabled-color: var(--bs-secondary-color);\n  display: flex;\n  flex-wrap: wrap;\n  padding-left: 0;\n  margin-bottom: 0;\n  list-style: none;\n}\n\n.nav-link {\n  display: block;\n  padding: var(--bs-nav-link-padding-y) var(--bs-nav-link-padding-x);\n  font-size: var(--bs-nav-link-font-size);\n  font-weight: var(--bs-nav-link-font-weight);\n  color: var(--bs-nav-link-color);\n  text-decoration: none;\n  background: none;\n  border: 0;\n  transition: color 0.15s ease-in-out, background-color 0.15s ease-in-out, border-color 0.15s ease-in-out;\n}\n@media (prefers-reduced-motion: reduce) {\n  .nav-link {\n    transition: none;\n  }\n}\n.nav-link:hover, .nav-link:focus {\n  color: var(--bs-nav-link-hover-color);\n}\n.nav-link:focus-visible {\n  outline: 0;\n  box-shadow: 0 0 0 0.1875rem var(--pst-color-accent);\n}\n.nav-link.disabled, .nav-link:disabled {\n  color: var(--bs-nav-link-disabled-color);\n  pointer-events: none;\n  cursor: default;\n}\n\n.nav-tabs {\n  --bs-nav-tabs-border-width: var(--bs-border-width);\n  --bs-nav-tabs-border-color: var(--bs-border-color);\n  --bs-nav-tabs-border-radius: var(--bs-border-radius);\n  --bs-nav-tabs-link-hover-border-color: var(--bs-secondary-bg) var(--bs-secondary-bg) var(--bs-border-color);\n  --bs-nav-tabs-link-active-color: var(--bs-emphasis-color);\n  --bs-nav-tabs-link-active-bg: var(--bs-body-bg);\n  --bs-nav-tabs-link-active-border-color: var(--bs-border-color) var(--bs-border-color) var(--bs-body-bg);\n  border-bottom: var(--bs-nav-tabs-border-width) solid var(--bs-nav-tabs-border-color);\n}\n.nav-tabs .nav-link {\n  margin-bottom: calc(-1 * var(--bs-nav-tabs-border-width));\n  border: var(--bs-nav-tabs-border-width) solid transparent;\n  border-top-left-radius: var(--bs-nav-tabs-border-radius);\n  border-top-right-radius: var(--bs-nav-tabs-border-radius);\n}\n.nav-tabs .nav-link:hover, .nav-tabs .nav-link:focus {\n  isolation: isolate;\n  border-color: var(--bs-nav-tabs-link-hover-border-color);\n}\n.nav-tabs .nav-link.active,\n.nav-tabs .nav-item.show .nav-link {\n  color: var(--bs-nav-tabs-link-active-color);\n  background-color: var(--bs-nav-tabs-link-active-bg);\n  border-color: var(--bs-nav-tabs-link-active-border-color);\n}\n.nav-tabs .dropdown-menu {\n  margin-top: calc(-1 * var(--bs-nav-tabs-border-width));\n  border-top-left-radius: 0;\n  border-top-right-radius: 0;\n}\n\n.nav-pills {\n  --bs-nav-pills-border-radius: var(--bs-border-radius);\n  --bs-nav-pills-link-active-color: #fff;\n  --bs-nav-pills-link-active-bg: #0d6efd;\n}\n.nav-pills .nav-link {\n  border-radius: var(--bs-nav-pills-border-radius);\n}\n.nav-pills .nav-link.active,\n.nav-pills .show > .nav-link {\n  color: var(--bs-nav-pills-link-active-color);\n  background-color: var(--bs-nav-pills-link-active-bg);\n}\n\n.nav-underline {\n  --bs-nav-underline-gap: 1rem;\n  --bs-nav-underline-border-width: 0.125rem;\n  --bs-nav-underline-link-active-color: var(--bs-emphasis-color);\n  gap: var(--bs-nav-underline-gap);\n}\n.nav-underline .nav-link {\n  padding-right: 0;\n  padding-left: 0;\n  border-bottom: var(--bs-nav-underline-border-width) solid transparent;\n}\n.nav-underline .nav-link:hover, .nav-underline .nav-link:focus {\n  border-bottom-color: currentcolor;\n}\n.nav-underline .nav-link.active,\n.nav-underline .show > .nav-link {\n  font-weight: 700;\n  color: var(--bs-nav-underline-link-active-color);\n  border-bottom-color: currentcolor;\n}\n\n.nav-fill > .nav-link,\n.nav-fill .nav-item {\n  flex: 1 1 auto;\n  text-align: center;\n}\n\n.nav-justified > .nav-link,\n.nav-justified .nav-item {\n  flex-basis: 0;\n  flex-grow: 1;\n  text-align: center;\n}\n\n.nav-fill .nav-item .nav-link,\n.nav-justified .nav-item .nav-link {\n  width: 100%;\n}\n\n.tab-content > .tab-pane {\n  display: none;\n}\n.tab-content > .active {\n  display: block;\n}\n\n.navbar {\n  --bs-navbar-padding-x: 0;\n  --bs-navbar-padding-y: 0.5rem;\n  --bs-navbar-color: rgba(var(--bs-emphasis-color-rgb), 0.65);\n  --bs-navbar-hover-color: rgba(var(--bs-emphasis-color-rgb), 0.8);\n  --bs-navbar-disabled-color: rgba(var(--bs-emphasis-color-rgb), 0.3);\n  --bs-navbar-active-color: rgba(var(--bs-emphasis-color-rgb), 1);\n  --bs-navbar-brand-padding-y: 0.3125rem;\n  --bs-navbar-brand-margin-end: 1rem;\n  --bs-navbar-brand-font-size: 1.25rem;\n  --bs-navbar-brand-color: rgba(var(--bs-emphasis-color-rgb), 1);\n  --bs-navbar-brand-hover-color: rgba(var(--bs-emphasis-color-rgb), 1);\n  --bs-navbar-nav-link-padding-x: 0.5rem;\n  --bs-navbar-toggler-padding-y: 0.25rem;\n  --bs-navbar-toggler-padding-x: 0.75rem;\n  --bs-navbar-toggler-font-size: 1.25rem;\n  --bs-navbar-toggler-icon-bg: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 30 30'%3e%3cpath stroke='rgba%2833, 37, 41, 0.75%29' stroke-linecap='round' stroke-miterlimit='10' stroke-width='2' d='M4 7h22M4 15h22M4 23h22'/%3e%3c/svg%3e\");\n  --bs-navbar-toggler-border-color: rgba(var(--bs-emphasis-color-rgb), 0.15);\n  --bs-navbar-toggler-border-radius: var(--bs-border-radius);\n  --bs-navbar-toggler-focus-width: 0.1875rem;\n  --bs-navbar-toggler-transition: box-shadow 0.15s ease-in-out;\n  position: relative;\n  display: flex;\n  flex-wrap: wrap;\n  align-items: center;\n  justify-content: space-between;\n  padding: var(--bs-navbar-padding-y) var(--bs-navbar-padding-x);\n}\n.navbar > .container,\n.navbar > .container-fluid,\n.navbar > .container-sm,\n.navbar > .container-md,\n.navbar > .container-lg,\n.navbar > .container-xl {\n  display: flex;\n  flex-wrap: inherit;\n  align-items: center;\n  justify-content: space-between;\n}\n.navbar-brand {\n  padding-top: var(--bs-navbar-brand-padding-y);\n  padding-bottom: var(--bs-navbar-brand-padding-y);\n  margin-right: var(--bs-navbar-brand-margin-end);\n  font-size: var(--bs-navbar-brand-font-size);\n  color: var(--bs-navbar-brand-color);\n  text-decoration: none;\n  white-space: nowrap;\n}\n.navbar-brand:hover, .navbar-brand:focus {\n  color: var(--bs-navbar-brand-hover-color);\n}\n\n.navbar-nav {\n  --bs-nav-link-padding-x: 0;\n  --bs-nav-link-padding-y: 0.5rem;\n  --bs-nav-link-font-weight: ;\n  --bs-nav-link-color: var(--bs-navbar-color);\n  --bs-nav-link-hover-color: var(--bs-navbar-hover-color);\n  --bs-nav-link-disabled-color: var(--bs-navbar-disabled-color);\n  display: flex;\n  flex-direction: column;\n  padding-left: 0;\n  margin-bottom: 0;\n  list-style: none;\n}\n.navbar-nav .nav-link.active, .navbar-nav .nav-link.show {\n  color: var(--bs-navbar-active-color);\n}\n.navbar-nav .dropdown-menu {\n  position: static;\n}\n\n.navbar-text {\n  padding-top: 0.5rem;\n  padding-bottom: 0.5rem;\n  color: var(--bs-navbar-color);\n}\n.navbar-text a,\n.navbar-text a:hover,\n.navbar-text a:focus {\n  color: var(--bs-navbar-active-color);\n}\n\n.navbar-collapse {\n  flex-basis: 100%;\n  flex-grow: 1;\n  align-items: center;\n}\n\n.navbar-toggler {\n  padding: var(--bs-navbar-toggler-padding-y) var(--bs-navbar-toggler-padding-x);\n  font-size: var(--bs-navbar-toggler-font-size);\n  line-height: 1;\n  color: var(--bs-navbar-color);\n  background-color: transparent;\n  border: var(--bs-border-width) solid var(--bs-navbar-toggler-border-color);\n  border-radius: var(--bs-navbar-toggler-border-radius);\n  transition: var(--bs-navbar-toggler-transition);\n}\n@media (prefers-reduced-motion: reduce) {\n  .navbar-toggler {\n    transition: none;\n  }\n}\n.navbar-toggler:hover {\n  text-decoration: none;\n}\n.navbar-toggler:focus {\n  text-decoration: none;\n  outline: 0;\n  box-shadow: 0 0 0 var(--bs-navbar-toggler-focus-width);\n}\n\n.navbar-toggler-icon {\n  display: inline-block;\n  width: 1.5em;\n  height: 1.5em;\n  vertical-align: middle;\n  background-image: var(--bs-navbar-toggler-icon-bg);\n  background-repeat: no-repeat;\n  background-position: center;\n  background-size: 100%;\n}\n\n.navbar-nav-scroll {\n  max-height: var(--bs-scroll-height, 75vh);\n  overflow-y: auto;\n}\n\n@media (min-width: 540px) {\n  .navbar-expand-sm {\n    flex-wrap: nowrap;\n    justify-content: flex-start;\n  }\n  .navbar-expand-sm .navbar-nav {\n    flex-direction: row;\n  }\n  .navbar-expand-sm .navbar-nav .dropdown-menu {\n    position: absolute;\n  }\n  .navbar-expand-sm .navbar-nav .nav-link {\n    padding-right: var(--bs-navbar-nav-link-padding-x);\n    padding-left: var(--bs-navbar-nav-link-padding-x);\n  }\n  .navbar-expand-sm .navbar-nav-scroll {\n    overflow: visible;\n  }\n  .navbar-expand-sm .navbar-collapse {\n    display: flex !important;\n    flex-basis: auto;\n  }\n  .navbar-expand-sm .navbar-toggler {\n    display: none;\n  }\n  .navbar-expand-sm .offcanvas {\n    position: static;\n    z-index: auto;\n    flex-grow: 1;\n    width: auto !important;\n    height: auto !important;\n    visibility: visible !important;\n    background-color: transparent !important;\n    border: 0 !important;\n    transform: none !important;\n    transition: none;\n  }\n  .navbar-expand-sm .offcanvas .offcanvas-header {\n    display: none;\n  }\n  .navbar-expand-sm .offcanvas .offcanvas-body {\n    display: flex;\n    flex-grow: 0;\n    padding: 0;\n    overflow-y: visible;\n  }\n}\n@media (min-width: 720px) {\n  .navbar-expand-md {\n    flex-wrap: nowrap;\n    justify-content: flex-start;\n  }\n  .navbar-expand-md .navbar-nav {\n    flex-direction: row;\n  }\n  .navbar-expand-md .navbar-nav .dropdown-menu {\n    position: absolute;\n  }\n  .navbar-expand-md .navbar-nav .nav-link {\n    padding-right: var(--bs-navbar-nav-link-padding-x);\n    padding-left: var(--bs-navbar-nav-link-padding-x);\n  }\n  .navbar-expand-md .navbar-nav-scroll {\n    overflow: visible;\n  }\n  .navbar-expand-md .navbar-collapse {\n    display: flex !important;\n    flex-basis: auto;\n  }\n  .navbar-expand-md .navbar-toggler {\n    display: none;\n  }\n  .navbar-expand-md .offcanvas {\n    position: static;\n    z-index: auto;\n    flex-grow: 1;\n    width: auto !important;\n    height: auto !important;\n    visibility: visible !important;\n    background-color: transparent !important;\n    border: 0 !important;\n    transform: none !important;\n    transition: none;\n  }\n  .navbar-expand-md .offcanvas .offcanvas-header {\n    display: none;\n  }\n  .navbar-expand-md .offcanvas .offcanvas-body {\n    display: flex;\n    flex-grow: 0;\n    padding: 0;\n    overflow-y: visible;\n  }\n}\n@media (min-width: 960px) {\n  .navbar-expand-lg {\n    flex-wrap: nowrap;\n    justify-content: flex-start;\n  }\n  .navbar-expand-lg .navbar-nav {\n    flex-direction: row;\n  }\n  .navbar-expand-lg .navbar-nav .dropdown-menu {\n    position: absolute;\n  }\n  .navbar-expand-lg .navbar-nav .nav-link {\n    padding-right: var(--bs-navbar-nav-link-padding-x);\n    padding-left: var(--bs-navbar-nav-link-padding-x);\n  }\n  .navbar-expand-lg .navbar-nav-scroll {\n    overflow: visible;\n  }\n  .navbar-expand-lg .navbar-collapse {\n    display: flex !important;\n    flex-basis: auto;\n  }\n  .navbar-expand-lg .navbar-toggler {\n    display: none;\n  }\n  .navbar-expand-lg .offcanvas {\n    position: static;\n    z-index: auto;\n    flex-grow: 1;\n    width: auto !important;\n    height: auto !important;\n    visibility: visible !important;\n    background-color: transparent !important;\n    border: 0 !important;\n    transform: none !important;\n    transition: none;\n  }\n  .navbar-expand-lg .offcanvas .offcanvas-header {\n    display: none;\n  }\n  .navbar-expand-lg .offcanvas .offcanvas-body {\n    display: flex;\n    flex-grow: 0;\n    padding: 0;\n    overflow-y: visible;\n  }\n}\n@media (min-width: 1200px) {\n  .navbar-expand-xl {\n    flex-wrap: nowrap;\n    justify-content: flex-start;\n  }\n  .navbar-expand-xl .navbar-nav {\n    flex-direction: row;\n  }\n  .navbar-expand-xl .navbar-nav .dropdown-menu {\n    position: absolute;\n  }\n  .navbar-expand-xl .navbar-nav .nav-link {\n    padding-right: var(--bs-navbar-nav-link-padding-x);\n    padding-left: var(--bs-navbar-nav-link-padding-x);\n  }\n  .navbar-expand-xl .navbar-nav-scroll {\n    overflow: visible;\n  }\n  .navbar-expand-xl .navbar-collapse {\n    display: flex !important;\n    flex-basis: auto;\n  }\n  .navbar-expand-xl .navbar-toggler {\n    display: none;\n  }\n  .navbar-expand-xl .offcanvas {\n    position: static;\n    z-index: auto;\n    flex-grow: 1;\n    width: auto !important;\n    height: auto !important;\n    visibility: visible !important;\n    background-color: transparent !important;\n    border: 0 !important;\n    transform: none !important;\n    transition: none;\n  }\n  .navbar-expand-xl .offcanvas .offcanvas-header {\n    display: none;\n  }\n  .navbar-expand-xl .offcanvas .offcanvas-body {\n    display: flex;\n    flex-grow: 0;\n    padding: 0;\n    overflow-y: visible;\n  }\n}\n.navbar-expand {\n  flex-wrap: nowrap;\n  justify-content: flex-start;\n}\n.navbar-expand .navbar-nav {\n  flex-direction: row;\n}\n.navbar-expand .navbar-nav .dropdown-menu {\n  position: absolute;\n}\n.navbar-expand .navbar-nav .nav-link {\n  padding-right: var(--bs-navbar-nav-link-padding-x);\n  padding-left: var(--bs-navbar-nav-link-padding-x);\n}\n.navbar-expand .navbar-nav-scroll {\n  overflow: visible;\n}\n.navbar-expand .navbar-collapse {\n  display: flex !important;\n  flex-basis: auto;\n}\n.navbar-expand .navbar-toggler {\n  display: none;\n}\n.navbar-expand .offcanvas {\n  position: static;\n  z-index: auto;\n  flex-grow: 1;\n  width: auto !important;\n  height: auto !important;\n  visibility: visible !important;\n  background-color: transparent !important;\n  border: 0 !important;\n  transform: none !important;\n  transition: none;\n}\n.navbar-expand .offcanvas .offcanvas-header {\n  display: none;\n}\n.navbar-expand .offcanvas .offcanvas-body {\n  display: flex;\n  flex-grow: 0;\n  padding: 0;\n  overflow-y: visible;\n}\n\n.navbar-dark,\n.navbar[data-bs-theme=dark] {\n  --bs-navbar-color: rgba(255, 255, 255, 0.55);\n  --bs-navbar-hover-color: rgba(255, 255, 255, 0.75);\n  --bs-navbar-disabled-color: rgba(255, 255, 255, 0.25);\n  --bs-navbar-active-color: #fff;\n  --bs-navbar-brand-color: #fff;\n  --bs-navbar-brand-hover-color: #fff;\n  --bs-navbar-toggler-border-color: rgba(255, 255, 255, 0.1);\n  --bs-navbar-toggler-icon-bg: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 30 30'%3e%3cpath stroke='rgba%28255, 255, 255, 0.55%29' stroke-linecap='round' stroke-miterlimit='10' stroke-width='2' d='M4 7h22M4 15h22M4 23h22'/%3e%3c/svg%3e\");\n}\n\n[data-bs-theme=dark] .navbar-toggler-icon {\n  --bs-navbar-toggler-icon-bg: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 30 30'%3e%3cpath stroke='rgba%28255, 255, 255, 0.55%29' stroke-linecap='round' stroke-miterlimit='10' stroke-width='2' d='M4 7h22M4 15h22M4 23h22'/%3e%3c/svg%3e\");\n}\n\n.card {\n  --bs-card-spacer-y: 1rem;\n  --bs-card-spacer-x: 1rem;\n  --bs-card-title-spacer-y: 0.5rem;\n  --bs-card-title-color: ;\n  --bs-card-subtitle-color: ;\n  --bs-card-border-width: var(--bs-border-width);\n  --bs-card-border-color: var(--bs-border-color-translucent);\n  --bs-card-border-radius: var(--bs-border-radius);\n  --bs-card-box-shadow: ;\n  --bs-card-inner-border-radius: calc(var(--bs-border-radius) - (var(--bs-border-width)));\n  --bs-card-cap-padding-y: 0.5rem;\n  --bs-card-cap-padding-x: 1rem;\n  --bs-card-cap-bg: rgba(var(--bs-body-color-rgb), 0.03);\n  --bs-card-cap-color: ;\n  --bs-card-height: ;\n  --bs-card-color: ;\n  --bs-card-bg: var(--bs-body-bg);\n  --bs-card-img-overlay-padding: 1rem;\n  --bs-card-group-margin: 0.75rem;\n  position: relative;\n  display: flex;\n  flex-direction: column;\n  min-width: 0;\n  height: var(--bs-card-height);\n  color: var(--bs-body-color);\n  word-wrap: break-word;\n  background-color: var(--bs-card-bg);\n  background-clip: border-box;\n  border: var(--bs-card-border-width) solid var(--bs-card-border-color);\n  border-radius: var(--bs-card-border-radius);\n}\n.card > hr {\n  margin-right: 0;\n  margin-left: 0;\n}\n.card > .list-group {\n  border-top: inherit;\n  border-bottom: inherit;\n}\n.card > .list-group:first-child {\n  border-top-width: 0;\n  border-top-left-radius: var(--bs-card-inner-border-radius);\n  border-top-right-radius: var(--bs-card-inner-border-radius);\n}\n.card > .list-group:last-child {\n  border-bottom-width: 0;\n  border-bottom-right-radius: var(--bs-card-inner-border-radius);\n  border-bottom-left-radius: var(--bs-card-inner-border-radius);\n}\n.card > .card-header + .list-group,\n.card > .list-group + .card-footer {\n  border-top: 0;\n}\n\n.card-body {\n  flex: 1 1 auto;\n  padding: var(--bs-card-spacer-y) var(--bs-card-spacer-x);\n  color: var(--bs-card-color);\n}\n\n.card-title {\n  margin-bottom: var(--bs-card-title-spacer-y);\n  color: var(--bs-card-title-color);\n}\n\n.card-subtitle {\n  margin-top: calc(-0.5 * var(--bs-card-title-spacer-y));\n  margin-bottom: 0;\n  color: var(--bs-card-subtitle-color);\n}\n\n.card-text:last-child {\n  margin-bottom: 0;\n}\n\n.card-link + .card-link {\n  margin-left: var(--bs-card-spacer-x);\n}\n\n.card-header {\n  padding: var(--bs-card-cap-padding-y) var(--bs-card-cap-padding-x);\n  margin-bottom: 0;\n  color: var(--bs-card-cap-color);\n  background-color: var(--bs-card-cap-bg);\n  border-bottom: var(--bs-card-border-width) solid var(--bs-card-border-color);\n}\n.card-header:first-child {\n  border-radius: var(--bs-card-inner-border-radius) var(--bs-card-inner-border-radius) 0 0;\n}\n\n.card-footer {\n  padding: var(--bs-card-cap-padding-y) var(--bs-card-cap-padding-x);\n  color: var(--bs-card-cap-color);\n  background-color: var(--bs-card-cap-bg);\n  border-top: var(--bs-card-border-width) solid var(--bs-card-border-color);\n}\n.card-footer:last-child {\n  border-radius: 0 0 var(--bs-card-inner-border-radius) var(--bs-card-inner-border-radius);\n}\n\n.card-header-tabs {\n  margin-right: calc(-0.5 * var(--bs-card-cap-padding-x));\n  margin-bottom: calc(-1 * var(--bs-card-cap-padding-y));\n  margin-left: calc(-0.5 * var(--bs-card-cap-padding-x));\n  border-bottom: 0;\n}\n.card-header-tabs .nav-link.active {\n  background-color: var(--bs-card-bg);\n  border-bottom-color: var(--bs-card-bg);\n}\n\n.card-header-pills {\n  margin-right: calc(-0.5 * var(--bs-card-cap-padding-x));\n  margin-left: calc(-0.5 * var(--bs-card-cap-padding-x));\n}\n\n.card-img-overlay {\n  position: absolute;\n  top: 0;\n  right: 0;\n  bottom: 0;\n  left: 0;\n  padding: var(--bs-card-img-overlay-padding);\n  border-radius: var(--bs-card-inner-border-radius);\n}\n\n.card-img,\n.card-img-top,\n.card-img-bottom {\n  width: 100%;\n}\n\n.card-img,\n.card-img-top {\n  border-top-left-radius: var(--bs-card-inner-border-radius);\n  border-top-right-radius: var(--bs-card-inner-border-radius);\n}\n\n.card-img,\n.card-img-bottom {\n  border-bottom-right-radius: var(--bs-card-inner-border-radius);\n  border-bottom-left-radius: var(--bs-card-inner-border-radius);\n}\n\n.card-group > .card {\n  margin-bottom: var(--bs-card-group-margin);\n}\n@media (min-width: 540px) {\n  .card-group {\n    display: flex;\n    flex-flow: row wrap;\n  }\n  .card-group > .card {\n    flex: 1 0 0%;\n    margin-bottom: 0;\n  }\n  .card-group > .card + .card {\n    margin-left: 0;\n    border-left: 0;\n  }\n  .card-group > .card:not(:last-child) {\n    border-top-right-radius: 0;\n    border-bottom-right-radius: 0;\n  }\n  .card-group > .card:not(:last-child) .card-img-top,\n  .card-group > .card:not(:last-child) .card-header {\n    border-top-right-radius: 0;\n  }\n  .card-group > .card:not(:last-child) .card-img-bottom,\n  .card-group > .card:not(:last-child) .card-footer {\n    border-bottom-right-radius: 0;\n  }\n  .card-group > .card:not(:first-child) {\n    border-top-left-radius: 0;\n    border-bottom-left-radius: 0;\n  }\n  .card-group > .card:not(:first-child) .card-img-top,\n  .card-group > .card:not(:first-child) .card-header {\n    border-top-left-radius: 0;\n  }\n  .card-group > .card:not(:first-child) .card-img-bottom,\n  .card-group > .card:not(:first-child) .card-footer {\n    border-bottom-left-radius: 0;\n  }\n}\n\n.accordion {\n  --bs-accordion-color: var(--bs-body-color);\n  --bs-accordion-bg: var(--bs-body-bg);\n  --bs-accordion-transition: color 0.15s ease-in-out, background-color 0.15s ease-in-out, border-color 0.15s ease-in-out, box-shadow 0.15s ease-in-out, border-radius 0.15s ease;\n  --bs-accordion-border-color: var(--bs-border-color);\n  --bs-accordion-border-width: var(--bs-border-width);\n  --bs-accordion-border-radius: var(--bs-border-radius);\n  --bs-accordion-inner-border-radius: calc(var(--bs-border-radius) - (var(--bs-border-width)));\n  --bs-accordion-btn-padding-x: 1.25rem;\n  --bs-accordion-btn-padding-y: 1rem;\n  --bs-accordion-btn-color: var(--bs-body-color);\n  --bs-accordion-btn-bg: var(--bs-accordion-bg);\n  --bs-accordion-btn-icon: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='none' stroke='%23212529' stroke-linecap='round' stroke-linejoin='round'%3e%3cpath d='M2 5L8 11L14 5'/%3e%3c/svg%3e\");\n  --bs-accordion-btn-icon-width: 1.25rem;\n  --bs-accordion-btn-icon-transform: rotate(-180deg);\n  --bs-accordion-btn-icon-transition: transform 0.2s ease-in-out;\n  --bs-accordion-btn-active-icon: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='none' stroke='%23052c65' stroke-linecap='round' stroke-linejoin='round'%3e%3cpath d='M2 5L8 11L14 5'/%3e%3c/svg%3e\");\n  --bs-accordion-btn-focus-box-shadow: 0 0 0 0.1875rem var(--pst-color-accent);\n  --bs-accordion-body-padding-x: 1.25rem;\n  --bs-accordion-body-padding-y: 1rem;\n  --bs-accordion-active-color: var(--bs-primary-text-emphasis);\n  --bs-accordion-active-bg: var(--bs-primary-bg-subtle);\n}\n\n.accordion-button {\n  position: relative;\n  display: flex;\n  align-items: center;\n  width: 100%;\n  padding: var(--bs-accordion-btn-padding-y) var(--bs-accordion-btn-padding-x);\n  font-size: 1rem;\n  color: var(--bs-accordion-btn-color);\n  text-align: left;\n  background-color: var(--bs-accordion-btn-bg);\n  border: 0;\n  border-radius: 0;\n  overflow-anchor: none;\n  transition: var(--bs-accordion-transition);\n}\n@media (prefers-reduced-motion: reduce) {\n  .accordion-button {\n    transition: none;\n  }\n}\n.accordion-button:not(.collapsed) {\n  color: var(--bs-accordion-active-color);\n  background-color: var(--bs-accordion-active-bg);\n  box-shadow: inset 0 calc(-1 * var(--bs-accordion-border-width)) 0 var(--bs-accordion-border-color);\n}\n.accordion-button:not(.collapsed)::after {\n  background-image: var(--bs-accordion-btn-active-icon);\n  transform: var(--bs-accordion-btn-icon-transform);\n}\n.accordion-button::after {\n  flex-shrink: 0;\n  width: var(--bs-accordion-btn-icon-width);\n  height: var(--bs-accordion-btn-icon-width);\n  margin-left: auto;\n  content: \"\";\n  background-image: var(--bs-accordion-btn-icon);\n  background-repeat: no-repeat;\n  background-size: var(--bs-accordion-btn-icon-width);\n  transition: var(--bs-accordion-btn-icon-transition);\n}\n@media (prefers-reduced-motion: reduce) {\n  .accordion-button::after {\n    transition: none;\n  }\n}\n.accordion-button:hover {\n  z-index: 2;\n}\n.accordion-button:focus {\n  z-index: 3;\n  outline: 0;\n  box-shadow: var(--bs-accordion-btn-focus-box-shadow);\n}\n\n.accordion-header {\n  margin-bottom: 0;\n}\n\n.accordion-item {\n  color: var(--bs-accordion-color);\n  background-color: var(--bs-accordion-bg);\n  border: var(--bs-accordion-border-width) solid var(--bs-accordion-border-color);\n}\n.accordion-item:first-of-type {\n  border-top-left-radius: var(--bs-accordion-border-radius);\n  border-top-right-radius: var(--bs-accordion-border-radius);\n}\n.accordion-item:first-of-type > .accordion-header .accordion-button {\n  border-top-left-radius: var(--bs-accordion-inner-border-radius);\n  border-top-right-radius: var(--bs-accordion-inner-border-radius);\n}\n.accordion-item:not(:first-of-type) {\n  border-top: 0;\n}\n.accordion-item:last-of-type {\n  border-bottom-right-radius: var(--bs-accordion-border-radius);\n  border-bottom-left-radius: var(--bs-accordion-border-radius);\n}\n.accordion-item:last-of-type > .accordion-header .accordion-button.collapsed {\n  border-bottom-right-radius: var(--bs-accordion-inner-border-radius);\n  border-bottom-left-radius: var(--bs-accordion-inner-border-radius);\n}\n.accordion-item:last-of-type > .accordion-collapse {\n  border-bottom-right-radius: var(--bs-accordion-border-radius);\n  border-bottom-left-radius: var(--bs-accordion-border-radius);\n}\n\n.accordion-body {\n  padding: var(--bs-accordion-body-padding-y) var(--bs-accordion-body-padding-x);\n}\n\n.accordion-flush > .accordion-item {\n  border-right: 0;\n  border-left: 0;\n  border-radius: 0;\n}\n.accordion-flush > .accordion-item:first-child {\n  border-top: 0;\n}\n.accordion-flush > .accordion-item:last-child {\n  border-bottom: 0;\n}\n.accordion-flush > .accordion-item > .accordion-header .accordion-button, .accordion-flush > .accordion-item > .accordion-header .accordion-button.collapsed {\n  border-radius: 0;\n}\n.accordion-flush > .accordion-item > .accordion-collapse {\n  border-radius: 0;\n}\n\n[data-bs-theme=dark] .accordion-button::after {\n  --bs-accordion-btn-icon: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='%236ea8fe'%3e%3cpath fill-rule='evenodd' d='M1.646 4.646a.5.5 0 0 1 .708 0L8 10.293l5.646-5.647a.5.5 0 0 1 .708.708l-6 6a.5.5 0 0 1-.708 0l-6-6a.5.5 0 0 1 0-.708z'/%3e%3c/svg%3e\");\n  --bs-accordion-btn-active-icon: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='%236ea8fe'%3e%3cpath fill-rule='evenodd' d='M1.646 4.646a.5.5 0 0 1 .708 0L8 10.293l5.646-5.647a.5.5 0 0 1 .708.708l-6 6a.5.5 0 0 1-.708 0l-6-6a.5.5 0 0 1 0-.708z'/%3e%3c/svg%3e\");\n}\n\n.breadcrumb {\n  --bs-breadcrumb-padding-x: 0;\n  --bs-breadcrumb-padding-y: 0;\n  --bs-breadcrumb-margin-bottom: 1rem;\n  --bs-breadcrumb-bg: ;\n  --bs-breadcrumb-border-radius: ;\n  --bs-breadcrumb-divider-color: var(--bs-secondary-color);\n  --bs-breadcrumb-item-padding-x: 0.5rem;\n  --bs-breadcrumb-item-active-color: var(--bs-secondary-color);\n  display: flex;\n  flex-wrap: wrap;\n  padding: var(--bs-breadcrumb-padding-y) var(--bs-breadcrumb-padding-x);\n  margin-bottom: var(--bs-breadcrumb-margin-bottom);\n  font-size: var(--bs-breadcrumb-font-size);\n  list-style: none;\n  background-color: var(--bs-breadcrumb-bg);\n  border-radius: var(--bs-breadcrumb-border-radius);\n}\n\n.breadcrumb-item + .breadcrumb-item {\n  padding-left: var(--bs-breadcrumb-item-padding-x);\n}\n.breadcrumb-item + .breadcrumb-item::before {\n  float: left;\n  padding-right: var(--bs-breadcrumb-item-padding-x);\n  color: var(--bs-breadcrumb-divider-color);\n  content: var(--bs-breadcrumb-divider, \"/\") /* rtl: var(--bs-breadcrumb-divider, \"/\") */;\n}\n.breadcrumb-item.active {\n  color: var(--bs-breadcrumb-item-active-color);\n}\n\n.pagination {\n  --bs-pagination-padding-x: 0.75rem;\n  --bs-pagination-padding-y: 0.375rem;\n  --bs-pagination-font-size: 1rem;\n  --bs-pagination-color: var(--bs-link-color);\n  --bs-pagination-bg: var(--bs-body-bg);\n  --bs-pagination-border-width: var(--bs-border-width);\n  --bs-pagination-border-color: var(--bs-border-color);\n  --bs-pagination-border-radius: var(--bs-border-radius);\n  --bs-pagination-hover-color: var(--bs-link-hover-color);\n  --bs-pagination-hover-bg: var(--bs-tertiary-bg);\n  --bs-pagination-hover-border-color: var(--bs-border-color);\n  --bs-pagination-focus-color: var(--bs-link-hover-color);\n  --bs-pagination-focus-bg: var(--bs-secondary-bg);\n  --bs-pagination-focus-box-shadow: 0 0 0 0.1875rem var(--pst-color-accent);\n  --bs-pagination-active-color: #fff;\n  --bs-pagination-active-bg: #0d6efd;\n  --bs-pagination-active-border-color: #0d6efd;\n  --bs-pagination-disabled-color: var(--bs-secondary-color);\n  --bs-pagination-disabled-bg: var(--bs-secondary-bg);\n  --bs-pagination-disabled-border-color: var(--bs-border-color);\n  display: flex;\n  padding-left: 0;\n  list-style: none;\n}\n\n.page-link {\n  position: relative;\n  display: block;\n  padding: var(--bs-pagination-padding-y) var(--bs-pagination-padding-x);\n  font-size: var(--bs-pagination-font-size);\n  color: var(--bs-pagination-color);\n  text-decoration: none;\n  background-color: var(--bs-pagination-bg);\n  border: var(--bs-pagination-border-width) solid var(--bs-pagination-border-color);\n  transition: color 0.15s ease-in-out, background-color 0.15s ease-in-out, border-color 0.15s ease-in-out, box-shadow 0.15s ease-in-out;\n}\n@media (prefers-reduced-motion: reduce) {\n  .page-link {\n    transition: none;\n  }\n}\n.page-link:hover {\n  z-index: 2;\n  color: var(--bs-pagination-hover-color);\n  background-color: var(--bs-pagination-hover-bg);\n  border-color: var(--bs-pagination-hover-border-color);\n}\n.page-link:focus {\n  z-index: 3;\n  color: var(--bs-pagination-focus-color);\n  background-color: var(--bs-pagination-focus-bg);\n  outline: 0;\n  box-shadow: var(--bs-pagination-focus-box-shadow);\n}\n.page-link.active, .active > .page-link {\n  z-index: 3;\n  color: var(--bs-pagination-active-color);\n  background-color: var(--bs-pagination-active-bg);\n  border-color: var(--bs-pagination-active-border-color);\n}\n.page-link.disabled, .disabled > .page-link {\n  color: var(--bs-pagination-disabled-color);\n  pointer-events: none;\n  background-color: var(--bs-pagination-disabled-bg);\n  border-color: var(--bs-pagination-disabled-border-color);\n}\n\n.page-item:not(:first-child) .page-link {\n  margin-left: calc(var(--bs-border-width) * -1);\n}\n.page-item:first-child .page-link {\n  border-top-left-radius: var(--bs-pagination-border-radius);\n  border-bottom-left-radius: var(--bs-pagination-border-radius);\n}\n.page-item:last-child .page-link {\n  border-top-right-radius: var(--bs-pagination-border-radius);\n  border-bottom-right-radius: var(--bs-pagination-border-radius);\n}\n\n.pagination-lg {\n  --bs-pagination-padding-x: 1.5rem;\n  --bs-pagination-padding-y: 0.75rem;\n  --bs-pagination-font-size: 1.25rem;\n  --bs-pagination-border-radius: var(--bs-border-radius-lg);\n}\n\n.pagination-sm {\n  --bs-pagination-padding-x: 0.5rem;\n  --bs-pagination-padding-y: 0.25rem;\n  --bs-pagination-font-size: 0.875rem;\n  --bs-pagination-border-radius: var(--bs-border-radius-sm);\n}\n\n.badge {\n  --bs-badge-padding-x: 0.65em;\n  --bs-badge-padding-y: 0.35em;\n  --bs-badge-font-size: 0.75em;\n  --bs-badge-font-weight: 700;\n  --bs-badge-color: #fff;\n  --bs-badge-border-radius: var(--bs-border-radius);\n  display: inline-block;\n  padding: var(--bs-badge-padding-y) var(--bs-badge-padding-x);\n  font-size: var(--bs-badge-font-size);\n  font-weight: var(--bs-badge-font-weight);\n  line-height: 1;\n  color: var(--bs-badge-color);\n  text-align: center;\n  white-space: nowrap;\n  vertical-align: baseline;\n  border-radius: var(--bs-badge-border-radius);\n}\n.badge:empty {\n  display: none;\n}\n\n.btn .badge {\n  position: relative;\n  top: -1px;\n}\n\n.alert {\n  --bs-alert-bg: transparent;\n  --bs-alert-padding-x: 1rem;\n  --bs-alert-padding-y: 1rem;\n  --bs-alert-margin-bottom: 1rem;\n  --bs-alert-color: inherit;\n  --bs-alert-border-color: transparent;\n  --bs-alert-border: var(--bs-border-width) solid var(--bs-alert-border-color);\n  --bs-alert-border-radius: var(--bs-border-radius);\n  --bs-alert-link-color: inherit;\n  position: relative;\n  padding: var(--bs-alert-padding-y) var(--bs-alert-padding-x);\n  margin-bottom: var(--bs-alert-margin-bottom);\n  color: var(--bs-alert-color);\n  background-color: var(--bs-alert-bg);\n  border: var(--bs-alert-border);\n  border-radius: var(--bs-alert-border-radius);\n}\n\n.alert-heading {\n  color: inherit;\n}\n\n.alert-link {\n  font-weight: 700;\n  color: var(--bs-alert-link-color);\n}\n\n.alert-dismissible {\n  padding-right: 3rem;\n}\n.alert-dismissible .btn-close {\n  position: absolute;\n  top: 0;\n  right: 0;\n  z-index: 2;\n  padding: 1.25rem 1rem;\n}\n\n.alert-primary {\n  --bs-alert-color: var(--bs-primary-text-emphasis);\n  --bs-alert-bg: var(--bs-primary-bg-subtle);\n  --bs-alert-border-color: var(--bs-primary-border-subtle);\n  --bs-alert-link-color: var(--bs-primary-text-emphasis);\n}\n\n.alert-secondary {\n  --bs-alert-color: var(--bs-secondary-text-emphasis);\n  --bs-alert-bg: var(--bs-secondary-bg-subtle);\n  --bs-alert-border-color: var(--bs-secondary-border-subtle);\n  --bs-alert-link-color: var(--bs-secondary-text-emphasis);\n}\n\n.alert-success {\n  --bs-alert-color: var(--bs-success-text-emphasis);\n  --bs-alert-bg: var(--bs-success-bg-subtle);\n  --bs-alert-border-color: var(--bs-success-border-subtle);\n  --bs-alert-link-color: var(--bs-success-text-emphasis);\n}\n\n.alert-info {\n  --bs-alert-color: var(--bs-info-text-emphasis);\n  --bs-alert-bg: var(--bs-info-bg-subtle);\n  --bs-alert-border-color: var(--bs-info-border-subtle);\n  --bs-alert-link-color: var(--bs-info-text-emphasis);\n}\n\n.alert-warning {\n  --bs-alert-color: var(--bs-warning-text-emphasis);\n  --bs-alert-bg: var(--bs-warning-bg-subtle);\n  --bs-alert-border-color: var(--bs-warning-border-subtle);\n  --bs-alert-link-color: var(--bs-warning-text-emphasis);\n}\n\n.alert-danger {\n  --bs-alert-color: var(--bs-danger-text-emphasis);\n  --bs-alert-bg: var(--bs-danger-bg-subtle);\n  --bs-alert-border-color: var(--bs-danger-border-subtle);\n  --bs-alert-link-color: var(--bs-danger-text-emphasis);\n}\n\n.alert-light {\n  --bs-alert-color: var(--bs-light-text-emphasis);\n  --bs-alert-bg: var(--bs-light-bg-subtle);\n  --bs-alert-border-color: var(--bs-light-border-subtle);\n  --bs-alert-link-color: var(--bs-light-text-emphasis);\n}\n\n.alert-dark {\n  --bs-alert-color: var(--bs-dark-text-emphasis);\n  --bs-alert-bg: var(--bs-dark-bg-subtle);\n  --bs-alert-border-color: var(--bs-dark-border-subtle);\n  --bs-alert-link-color: var(--bs-dark-text-emphasis);\n}\n\n@keyframes progress-bar-stripes {\n  0% {\n    background-position-x: 1rem;\n  }\n}\n.progress,\n.progress-stacked {\n  --bs-progress-height: 1rem;\n  --bs-progress-font-size: 0.75rem;\n  --bs-progress-bg: var(--bs-secondary-bg);\n  --bs-progress-border-radius: var(--bs-border-radius);\n  --bs-progress-box-shadow: var(--bs-box-shadow-inset);\n  --bs-progress-bar-color: #fff;\n  --bs-progress-bar-bg: #0d6efd;\n  --bs-progress-bar-transition: width 0.6s ease;\n  display: flex;\n  height: var(--bs-progress-height);\n  overflow: hidden;\n  font-size: var(--bs-progress-font-size);\n  background-color: var(--bs-progress-bg);\n  border-radius: var(--bs-progress-border-radius);\n}\n\n.progress-bar {\n  display: flex;\n  flex-direction: column;\n  justify-content: center;\n  overflow: hidden;\n  color: var(--bs-progress-bar-color);\n  text-align: center;\n  white-space: nowrap;\n  background-color: var(--bs-progress-bar-bg);\n  transition: var(--bs-progress-bar-transition);\n}\n@media (prefers-reduced-motion: reduce) {\n  .progress-bar {\n    transition: none;\n  }\n}\n\n.progress-bar-striped {\n  background-image: linear-gradient(45deg, rgba(255, 255, 255, 0.15) 25%, transparent 25%, transparent 50%, rgba(255, 255, 255, 0.15) 50%, rgba(255, 255, 255, 0.15) 75%, transparent 75%, transparent);\n  background-size: var(--bs-progress-height) var(--bs-progress-height);\n}\n\n.progress-stacked > .progress {\n  overflow: visible;\n}\n\n.progress-stacked > .progress > .progress-bar {\n  width: 100%;\n}\n\n.progress-bar-animated {\n  animation: 1s linear infinite progress-bar-stripes;\n}\n@media (prefers-reduced-motion: reduce) {\n  .progress-bar-animated {\n    animation: none;\n  }\n}\n\n.list-group {\n  --bs-list-group-color: var(--bs-body-color);\n  --bs-list-group-bg: var(--bs-body-bg);\n  --bs-list-group-border-color: var(--bs-border-color);\n  --bs-list-group-border-width: var(--bs-border-width);\n  --bs-list-group-border-radius: var(--bs-border-radius);\n  --bs-list-group-item-padding-x: 1rem;\n  --bs-list-group-item-padding-y: 0.5rem;\n  --bs-list-group-action-color: var(--bs-secondary-color);\n  --bs-list-group-action-hover-color: var(--bs-emphasis-color);\n  --bs-list-group-action-hover-bg: var(--bs-tertiary-bg);\n  --bs-list-group-action-active-color: var(--bs-body-color);\n  --bs-list-group-action-active-bg: var(--bs-secondary-bg);\n  --bs-list-group-disabled-color: var(--bs-secondary-color);\n  --bs-list-group-disabled-bg: var(--bs-body-bg);\n  --bs-list-group-active-color: #fff;\n  --bs-list-group-active-bg: #0d6efd;\n  --bs-list-group-active-border-color: #0d6efd;\n  display: flex;\n  flex-direction: column;\n  padding-left: 0;\n  margin-bottom: 0;\n  border-radius: var(--bs-list-group-border-radius);\n}\n\n.list-group-numbered {\n  list-style-type: none;\n  counter-reset: section;\n}\n.list-group-numbered > .list-group-item::before {\n  content: counters(section, \".\") \". \";\n  counter-increment: section;\n}\n\n.list-group-item-action {\n  width: 100%;\n  color: var(--bs-list-group-action-color);\n  text-align: inherit;\n}\n.list-group-item-action:hover, .list-group-item-action:focus {\n  z-index: 1;\n  color: var(--bs-list-group-action-hover-color);\n  text-decoration: none;\n  background-color: var(--bs-list-group-action-hover-bg);\n}\n.list-group-item-action:active {\n  color: var(--bs-list-group-action-active-color);\n  background-color: var(--bs-list-group-action-active-bg);\n}\n\n.list-group-item {\n  position: relative;\n  display: block;\n  padding: var(--bs-list-group-item-padding-y) var(--bs-list-group-item-padding-x);\n  color: var(--bs-list-group-color);\n  text-decoration: none;\n  background-color: var(--bs-list-group-bg);\n  border: var(--bs-list-group-border-width) solid var(--bs-list-group-border-color);\n}\n.list-group-item:first-child {\n  border-top-left-radius: inherit;\n  border-top-right-radius: inherit;\n}\n.list-group-item:last-child {\n  border-bottom-right-radius: inherit;\n  border-bottom-left-radius: inherit;\n}\n.list-group-item.disabled, .list-group-item:disabled {\n  color: var(--bs-list-group-disabled-color);\n  pointer-events: none;\n  background-color: var(--bs-list-group-disabled-bg);\n}\n.list-group-item.active {\n  z-index: 2;\n  color: var(--bs-list-group-active-color);\n  background-color: var(--bs-list-group-active-bg);\n  border-color: var(--bs-list-group-active-border-color);\n}\n.list-group-item + .list-group-item {\n  border-top-width: 0;\n}\n.list-group-item + .list-group-item.active {\n  margin-top: calc(-1 * var(--bs-list-group-border-width));\n  border-top-width: var(--bs-list-group-border-width);\n}\n\n.list-group-horizontal {\n  flex-direction: row;\n}\n.list-group-horizontal > .list-group-item:first-child:not(:last-child) {\n  border-bottom-left-radius: var(--bs-list-group-border-radius);\n  border-top-right-radius: 0;\n}\n.list-group-horizontal > .list-group-item:last-child:not(:first-child) {\n  border-top-right-radius: var(--bs-list-group-border-radius);\n  border-bottom-left-radius: 0;\n}\n.list-group-horizontal > .list-group-item.active {\n  margin-top: 0;\n}\n.list-group-horizontal > .list-group-item + .list-group-item {\n  border-top-width: var(--bs-list-group-border-width);\n  border-left-width: 0;\n}\n.list-group-horizontal > .list-group-item + .list-group-item.active {\n  margin-left: calc(-1 * var(--bs-list-group-border-width));\n  border-left-width: var(--bs-list-group-border-width);\n}\n\n@media (min-width: 540px) {\n  .list-group-horizontal-sm {\n    flex-direction: row;\n  }\n  .list-group-horizontal-sm > .list-group-item:first-child:not(:last-child) {\n    border-bottom-left-radius: var(--bs-list-group-border-radius);\n    border-top-right-radius: 0;\n  }\n  .list-group-horizontal-sm > .list-group-item:last-child:not(:first-child) {\n    border-top-right-radius: var(--bs-list-group-border-radius);\n    border-bottom-left-radius: 0;\n  }\n  .list-group-horizontal-sm > .list-group-item.active {\n    margin-top: 0;\n  }\n  .list-group-horizontal-sm > .list-group-item + .list-group-item {\n    border-top-width: var(--bs-list-group-border-width);\n    border-left-width: 0;\n  }\n  .list-group-horizontal-sm > .list-group-item + .list-group-item.active {\n    margin-left: calc(-1 * var(--bs-list-group-border-width));\n    border-left-width: var(--bs-list-group-border-width);\n  }\n}\n@media (min-width: 720px) {\n  .list-group-horizontal-md {\n    flex-direction: row;\n  }\n  .list-group-horizontal-md > .list-group-item:first-child:not(:last-child) {\n    border-bottom-left-radius: var(--bs-list-group-border-radius);\n    border-top-right-radius: 0;\n  }\n  .list-group-horizontal-md > .list-group-item:last-child:not(:first-child) {\n    border-top-right-radius: var(--bs-list-group-border-radius);\n    border-bottom-left-radius: 0;\n  }\n  .list-group-horizontal-md > .list-group-item.active {\n    margin-top: 0;\n  }\n  .list-group-horizontal-md > .list-group-item + .list-group-item {\n    border-top-width: var(--bs-list-group-border-width);\n    border-left-width: 0;\n  }\n  .list-group-horizontal-md > .list-group-item + .list-group-item.active {\n    margin-left: calc(-1 * var(--bs-list-group-border-width));\n    border-left-width: var(--bs-list-group-border-width);\n  }\n}\n@media (min-width: 960px) {\n  .list-group-horizontal-lg {\n    flex-direction: row;\n  }\n  .list-group-horizontal-lg > .list-group-item:first-child:not(:last-child) {\n    border-bottom-left-radius: var(--bs-list-group-border-radius);\n    border-top-right-radius: 0;\n  }\n  .list-group-horizontal-lg > .list-group-item:last-child:not(:first-child) {\n    border-top-right-radius: var(--bs-list-group-border-radius);\n    border-bottom-left-radius: 0;\n  }\n  .list-group-horizontal-lg > .list-group-item.active {\n    margin-top: 0;\n  }\n  .list-group-horizontal-lg > .list-group-item + .list-group-item {\n    border-top-width: var(--bs-list-group-border-width);\n    border-left-width: 0;\n  }\n  .list-group-horizontal-lg > .list-group-item + .list-group-item.active {\n    margin-left: calc(-1 * var(--bs-list-group-border-width));\n    border-left-width: var(--bs-list-group-border-width);\n  }\n}\n@media (min-width: 1200px) {\n  .list-group-horizontal-xl {\n    flex-direction: row;\n  }\n  .list-group-horizontal-xl > .list-group-item:first-child:not(:last-child) {\n    border-bottom-left-radius: var(--bs-list-group-border-radius);\n    border-top-right-radius: 0;\n  }\n  .list-group-horizontal-xl > .list-group-item:last-child:not(:first-child) {\n    border-top-right-radius: var(--bs-list-group-border-radius);\n    border-bottom-left-radius: 0;\n  }\n  .list-group-horizontal-xl > .list-group-item.active {\n    margin-top: 0;\n  }\n  .list-group-horizontal-xl > .list-group-item + .list-group-item {\n    border-top-width: var(--bs-list-group-border-width);\n    border-left-width: 0;\n  }\n  .list-group-horizontal-xl > .list-group-item + .list-group-item.active {\n    margin-left: calc(-1 * var(--bs-list-group-border-width));\n    border-left-width: var(--bs-list-group-border-width);\n  }\n}\n.list-group-flush {\n  border-radius: 0;\n}\n.list-group-flush > .list-group-item {\n  border-width: 0 0 var(--bs-list-group-border-width);\n}\n.list-group-flush > .list-group-item:last-child {\n  border-bottom-width: 0;\n}\n\n.list-group-item-primary {\n  --bs-list-group-color: var(--bs-primary-text-emphasis);\n  --bs-list-group-bg: var(--bs-primary-bg-subtle);\n  --bs-list-group-border-color: var(--bs-primary-border-subtle);\n  --bs-list-group-action-hover-color: var(--bs-emphasis-color);\n  --bs-list-group-action-hover-bg: var(--bs-primary-border-subtle);\n  --bs-list-group-action-active-color: var(--bs-emphasis-color);\n  --bs-list-group-action-active-bg: var(--bs-primary-border-subtle);\n  --bs-list-group-active-color: var(--bs-primary-bg-subtle);\n  --bs-list-group-active-bg: var(--bs-primary-text-emphasis);\n  --bs-list-group-active-border-color: var(--bs-primary-text-emphasis);\n}\n\n.list-group-item-secondary {\n  --bs-list-group-color: var(--bs-secondary-text-emphasis);\n  --bs-list-group-bg: var(--bs-secondary-bg-subtle);\n  --bs-list-group-border-color: var(--bs-secondary-border-subtle);\n  --bs-list-group-action-hover-color: var(--bs-emphasis-color);\n  --bs-list-group-action-hover-bg: var(--bs-secondary-border-subtle);\n  --bs-list-group-action-active-color: var(--bs-emphasis-color);\n  --bs-list-group-action-active-bg: var(--bs-secondary-border-subtle);\n  --bs-list-group-active-color: var(--bs-secondary-bg-subtle);\n  --bs-list-group-active-bg: var(--bs-secondary-text-emphasis);\n  --bs-list-group-active-border-color: var(--bs-secondary-text-emphasis);\n}\n\n.list-group-item-success {\n  --bs-list-group-color: var(--bs-success-text-emphasis);\n  --bs-list-group-bg: var(--bs-success-bg-subtle);\n  --bs-list-group-border-color: var(--bs-success-border-subtle);\n  --bs-list-group-action-hover-color: var(--bs-emphasis-color);\n  --bs-list-group-action-hover-bg: var(--bs-success-border-subtle);\n  --bs-list-group-action-active-color: var(--bs-emphasis-color);\n  --bs-list-group-action-active-bg: var(--bs-success-border-subtle);\n  --bs-list-group-active-color: var(--bs-success-bg-subtle);\n  --bs-list-group-active-bg: var(--bs-success-text-emphasis);\n  --bs-list-group-active-border-color: var(--bs-success-text-emphasis);\n}\n\n.list-group-item-info {\n  --bs-list-group-color: var(--bs-info-text-emphasis);\n  --bs-list-group-bg: var(--bs-info-bg-subtle);\n  --bs-list-group-border-color: var(--bs-info-border-subtle);\n  --bs-list-group-action-hover-color: var(--bs-emphasis-color);\n  --bs-list-group-action-hover-bg: var(--bs-info-border-subtle);\n  --bs-list-group-action-active-color: var(--bs-emphasis-color);\n  --bs-list-group-action-active-bg: var(--bs-info-border-subtle);\n  --bs-list-group-active-color: var(--bs-info-bg-subtle);\n  --bs-list-group-active-bg: var(--bs-info-text-emphasis);\n  --bs-list-group-active-border-color: var(--bs-info-text-emphasis);\n}\n\n.list-group-item-warning {\n  --bs-list-group-color: var(--bs-warning-text-emphasis);\n  --bs-list-group-bg: var(--bs-warning-bg-subtle);\n  --bs-list-group-border-color: var(--bs-warning-border-subtle);\n  --bs-list-group-action-hover-color: var(--bs-emphasis-color);\n  --bs-list-group-action-hover-bg: var(--bs-warning-border-subtle);\n  --bs-list-group-action-active-color: var(--bs-emphasis-color);\n  --bs-list-group-action-active-bg: var(--bs-warning-border-subtle);\n  --bs-list-group-active-color: var(--bs-warning-bg-subtle);\n  --bs-list-group-active-bg: var(--bs-warning-text-emphasis);\n  --bs-list-group-active-border-color: var(--bs-warning-text-emphasis);\n}\n\n.list-group-item-danger {\n  --bs-list-group-color: var(--bs-danger-text-emphasis);\n  --bs-list-group-bg: var(--bs-danger-bg-subtle);\n  --bs-list-group-border-color: var(--bs-danger-border-subtle);\n  --bs-list-group-action-hover-color: var(--bs-emphasis-color);\n  --bs-list-group-action-hover-bg: var(--bs-danger-border-subtle);\n  --bs-list-group-action-active-color: var(--bs-emphasis-color);\n  --bs-list-group-action-active-bg: var(--bs-danger-border-subtle);\n  --bs-list-group-active-color: var(--bs-danger-bg-subtle);\n  --bs-list-group-active-bg: var(--bs-danger-text-emphasis);\n  --bs-list-group-active-border-color: var(--bs-danger-text-emphasis);\n}\n\n.list-group-item-light {\n  --bs-list-group-color: var(--bs-light-text-emphasis);\n  --bs-list-group-bg: var(--bs-light-bg-subtle);\n  --bs-list-group-border-color: var(--bs-light-border-subtle);\n  --bs-list-group-action-hover-color: var(--bs-emphasis-color);\n  --bs-list-group-action-hover-bg: var(--bs-light-border-subtle);\n  --bs-list-group-action-active-color: var(--bs-emphasis-color);\n  --bs-list-group-action-active-bg: var(--bs-light-border-subtle);\n  --bs-list-group-active-color: var(--bs-light-bg-subtle);\n  --bs-list-group-active-bg: var(--bs-light-text-emphasis);\n  --bs-list-group-active-border-color: var(--bs-light-text-emphasis);\n}\n\n.list-group-item-dark {\n  --bs-list-group-color: var(--bs-dark-text-emphasis);\n  --bs-list-group-bg: var(--bs-dark-bg-subtle);\n  --bs-list-group-border-color: var(--bs-dark-border-subtle);\n  --bs-list-group-action-hover-color: var(--bs-emphasis-color);\n  --bs-list-group-action-hover-bg: var(--bs-dark-border-subtle);\n  --bs-list-group-action-active-color: var(--bs-emphasis-color);\n  --bs-list-group-action-active-bg: var(--bs-dark-border-subtle);\n  --bs-list-group-active-color: var(--bs-dark-bg-subtle);\n  --bs-list-group-active-bg: var(--bs-dark-text-emphasis);\n  --bs-list-group-active-border-color: var(--bs-dark-text-emphasis);\n}\n\n.btn-close {\n  --bs-btn-close-color: #000;\n  --bs-btn-close-bg: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='%23000'%3e%3cpath d='M.293.293a1 1 0 0 1 1.414 0L8 6.586 14.293.293a1 1 0 1 1 1.414 1.414L9.414 8l6.293 6.293a1 1 0 0 1-1.414 1.414L8 9.414l-6.293 6.293a1 1 0 0 1-1.414-1.414L6.586 8 .293 1.707a1 1 0 0 1 0-1.414z'/%3e%3c/svg%3e\");\n  --bs-btn-close-opacity: 0.5;\n  --bs-btn-close-hover-opacity: 0.75;\n  --bs-btn-close-focus-shadow: 0 0 0 0.1875rem var(--pst-color-accent);\n  --bs-btn-close-focus-opacity: 1;\n  --bs-btn-close-disabled-opacity: 0.25;\n  --bs-btn-close-white-filter: invert(1) grayscale(100%) brightness(200%);\n  box-sizing: content-box;\n  width: 1em;\n  height: 1em;\n  padding: 0.25em 0.25em;\n  color: var(--bs-btn-close-color);\n  background: transparent var(--bs-btn-close-bg) center/1em auto no-repeat;\n  border: 0;\n  border-radius: 0.375rem;\n  opacity: var(--bs-btn-close-opacity);\n}\n.btn-close:hover {\n  color: var(--bs-btn-close-color);\n  text-decoration: none;\n  opacity: var(--bs-btn-close-hover-opacity);\n}\n.btn-close:focus {\n  outline: 0;\n  box-shadow: var(--bs-btn-close-focus-shadow);\n  opacity: var(--bs-btn-close-focus-opacity);\n}\n.btn-close:disabled, .btn-close.disabled {\n  pointer-events: none;\n  user-select: none;\n  opacity: var(--bs-btn-close-disabled-opacity);\n}\n\n.btn-close-white {\n  filter: var(--bs-btn-close-white-filter);\n}\n\n[data-bs-theme=dark] .btn-close {\n  filter: var(--bs-btn-close-white-filter);\n}\n\n.toast {\n  --bs-toast-zindex: 1090;\n  --bs-toast-padding-x: 0.75rem;\n  --bs-toast-padding-y: 0.5rem;\n  --bs-toast-spacing: 1.5rem;\n  --bs-toast-max-width: 350px;\n  --bs-toast-font-size: 0.875rem;\n  --bs-toast-color: ;\n  --bs-toast-bg: rgba(var(--bs-body-bg-rgb), 0.85);\n  --bs-toast-border-width: var(--bs-border-width);\n  --bs-toast-border-color: var(--bs-border-color-translucent);\n  --bs-toast-border-radius: var(--bs-border-radius);\n  --bs-toast-box-shadow: var(--bs-box-shadow);\n  --bs-toast-header-color: var(--bs-secondary-color);\n  --bs-toast-header-bg: rgba(var(--bs-body-bg-rgb), 0.85);\n  --bs-toast-header-border-color: var(--bs-border-color-translucent);\n  width: var(--bs-toast-max-width);\n  max-width: 100%;\n  font-size: var(--bs-toast-font-size);\n  color: var(--bs-toast-color);\n  pointer-events: auto;\n  background-color: var(--bs-toast-bg);\n  background-clip: padding-box;\n  border: var(--bs-toast-border-width) solid var(--bs-toast-border-color);\n  box-shadow: var(--bs-toast-box-shadow);\n  border-radius: var(--bs-toast-border-radius);\n}\n.toast.showing {\n  opacity: 0;\n}\n.toast:not(.show) {\n  display: none;\n}\n\n.toast-container {\n  --bs-toast-zindex: 1090;\n  position: absolute;\n  z-index: var(--bs-toast-zindex);\n  width: max-content;\n  max-width: 100%;\n  pointer-events: none;\n}\n.toast-container > :not(:last-child) {\n  margin-bottom: var(--bs-toast-spacing);\n}\n\n.toast-header {\n  display: flex;\n  align-items: center;\n  padding: var(--bs-toast-padding-y) var(--bs-toast-padding-x);\n  color: var(--bs-toast-header-color);\n  background-color: var(--bs-toast-header-bg);\n  background-clip: padding-box;\n  border-bottom: var(--bs-toast-border-width) solid var(--bs-toast-header-border-color);\n  border-top-left-radius: calc(var(--bs-toast-border-radius) - var(--bs-toast-border-width));\n  border-top-right-radius: calc(var(--bs-toast-border-radius) - var(--bs-toast-border-width));\n}\n.toast-header .btn-close {\n  margin-right: calc(-0.5 * var(--bs-toast-padding-x));\n  margin-left: var(--bs-toast-padding-x);\n}\n\n.toast-body {\n  padding: var(--bs-toast-padding-x);\n  word-wrap: break-word;\n}\n\n.modal {\n  --bs-modal-zindex: 1055;\n  --bs-modal-width: 500px;\n  --bs-modal-padding: 1rem;\n  --bs-modal-margin: 0.5rem;\n  --bs-modal-color: ;\n  --bs-modal-bg: var(--bs-body-bg);\n  --bs-modal-border-color: var(--bs-border-color-translucent);\n  --bs-modal-border-width: var(--bs-border-width);\n  --bs-modal-border-radius: var(--bs-border-radius-lg);\n  --bs-modal-box-shadow: var(--bs-box-shadow-sm);\n  --bs-modal-inner-border-radius: calc(var(--bs-border-radius-lg) - (var(--bs-border-width)));\n  --bs-modal-header-padding-x: 1rem;\n  --bs-modal-header-padding-y: 1rem;\n  --bs-modal-header-padding: 1rem 1rem;\n  --bs-modal-header-border-color: var(--bs-border-color);\n  --bs-modal-header-border-width: var(--bs-border-width);\n  --bs-modal-title-line-height: 1.5;\n  --bs-modal-footer-gap: 0.5rem;\n  --bs-modal-footer-bg: ;\n  --bs-modal-footer-border-color: var(--bs-border-color);\n  --bs-modal-footer-border-width: var(--bs-border-width);\n  position: fixed;\n  top: 0;\n  left: 0;\n  z-index: var(--bs-modal-zindex);\n  display: none;\n  width: 100%;\n  height: 100%;\n  overflow-x: hidden;\n  overflow-y: auto;\n  outline: 0;\n}\n\n.modal-dialog {\n  position: relative;\n  width: auto;\n  margin: var(--bs-modal-margin);\n  pointer-events: none;\n}\n.modal.fade .modal-dialog {\n  transition: transform 0.3s ease-out;\n  transform: translate(0, -50px);\n}\n@media (prefers-reduced-motion: reduce) {\n  .modal.fade .modal-dialog {\n    transition: none;\n  }\n}\n.modal.show .modal-dialog {\n  transform: none;\n}\n.modal.modal-static .modal-dialog {\n  transform: scale(1.02);\n}\n\n.modal-dialog-scrollable {\n  height: calc(100% - var(--bs-modal-margin) * 2);\n}\n.modal-dialog-scrollable .modal-content {\n  max-height: 100%;\n  overflow: hidden;\n}\n.modal-dialog-scrollable .modal-body {\n  overflow-y: auto;\n}\n\n.modal-dialog-centered {\n  display: flex;\n  align-items: center;\n  min-height: calc(100% - var(--bs-modal-margin) * 2);\n}\n\n.modal-content {\n  position: relative;\n  display: flex;\n  flex-direction: column;\n  width: 100%;\n  color: var(--bs-modal-color);\n  pointer-events: auto;\n  background-color: var(--bs-modal-bg);\n  background-clip: padding-box;\n  border: var(--bs-modal-border-width) solid var(--bs-modal-border-color);\n  border-radius: var(--bs-modal-border-radius);\n  outline: 0;\n}\n\n.modal-backdrop {\n  --bs-backdrop-zindex: 1050;\n  --bs-backdrop-bg: #000;\n  --bs-backdrop-opacity: 0.5;\n  position: fixed;\n  top: 0;\n  left: 0;\n  z-index: var(--bs-backdrop-zindex);\n  width: 100vw;\n  height: 100vh;\n  background-color: var(--bs-backdrop-bg);\n}\n.modal-backdrop.fade {\n  opacity: 0;\n}\n.modal-backdrop.show {\n  opacity: var(--bs-backdrop-opacity);\n}\n\n.modal-header {\n  display: flex;\n  flex-shrink: 0;\n  align-items: center;\n  padding: var(--bs-modal-header-padding);\n  border-bottom: var(--bs-modal-header-border-width) solid var(--bs-modal-header-border-color);\n  border-top-left-radius: var(--bs-modal-inner-border-radius);\n  border-top-right-radius: var(--bs-modal-inner-border-radius);\n}\n.modal-header .btn-close {\n  padding: calc(var(--bs-modal-header-padding-y) * 0.5) calc(var(--bs-modal-header-padding-x) * 0.5);\n  margin: calc(-0.5 * var(--bs-modal-header-padding-y)) calc(-0.5 * var(--bs-modal-header-padding-x)) calc(-0.5 * var(--bs-modal-header-padding-y)) auto;\n}\n\n.modal-title {\n  margin-bottom: 0;\n  line-height: var(--bs-modal-title-line-height);\n}\n\n.modal-body {\n  position: relative;\n  flex: 1 1 auto;\n  padding: var(--bs-modal-padding);\n}\n\n.modal-footer {\n  display: flex;\n  flex-shrink: 0;\n  flex-wrap: wrap;\n  align-items: center;\n  justify-content: flex-end;\n  padding: calc(var(--bs-modal-padding) - var(--bs-modal-footer-gap) * 0.5);\n  background-color: var(--bs-modal-footer-bg);\n  border-top: var(--bs-modal-footer-border-width) solid var(--bs-modal-footer-border-color);\n  border-bottom-right-radius: var(--bs-modal-inner-border-radius);\n  border-bottom-left-radius: var(--bs-modal-inner-border-radius);\n}\n.modal-footer > * {\n  margin: calc(var(--bs-modal-footer-gap) * 0.5);\n}\n\n@media (min-width: 540px) {\n  .modal {\n    --bs-modal-margin: 1.75rem;\n    --bs-modal-box-shadow: var(--bs-box-shadow);\n  }\n  .modal-dialog {\n    max-width: var(--bs-modal-width);\n    margin-right: auto;\n    margin-left: auto;\n  }\n  .modal-sm {\n    --bs-modal-width: 300px;\n  }\n}\n@media (min-width: 960px) {\n  .modal-lg,\n  .modal-xl {\n    --bs-modal-width: 800px;\n  }\n}\n@media (min-width: 1200px) {\n  .modal-xl {\n    --bs-modal-width: 1140px;\n  }\n}\n.modal-fullscreen {\n  width: 100vw;\n  max-width: none;\n  height: 100%;\n  margin: 0;\n}\n.modal-fullscreen .modal-content {\n  height: 100%;\n  border: 0;\n  border-radius: 0;\n}\n.modal-fullscreen .modal-header,\n.modal-fullscreen .modal-footer {\n  border-radius: 0;\n}\n.modal-fullscreen .modal-body {\n  overflow-y: auto;\n}\n\n@media (max-width: 539.98px) {\n  .modal-fullscreen-sm-down {\n    width: 100vw;\n    max-width: none;\n    height: 100%;\n    margin: 0;\n  }\n  .modal-fullscreen-sm-down .modal-content {\n    height: 100%;\n    border: 0;\n    border-radius: 0;\n  }\n  .modal-fullscreen-sm-down .modal-header,\n  .modal-fullscreen-sm-down .modal-footer {\n    border-radius: 0;\n  }\n  .modal-fullscreen-sm-down .modal-body {\n    overflow-y: auto;\n  }\n}\n@media (max-width: 719.98px) {\n  .modal-fullscreen-md-down {\n    width: 100vw;\n    max-width: none;\n    height: 100%;\n    margin: 0;\n  }\n  .modal-fullscreen-md-down .modal-content {\n    height: 100%;\n    border: 0;\n    border-radius: 0;\n  }\n  .modal-fullscreen-md-down .modal-header,\n  .modal-fullscreen-md-down .modal-footer {\n    border-radius: 0;\n  }\n  .modal-fullscreen-md-down .modal-body {\n    overflow-y: auto;\n  }\n}\n@media (max-width: 959.98px) {\n  .modal-fullscreen-lg-down {\n    width: 100vw;\n    max-width: none;\n    height: 100%;\n    margin: 0;\n  }\n  .modal-fullscreen-lg-down .modal-content {\n    height: 100%;\n    border: 0;\n    border-radius: 0;\n  }\n  .modal-fullscreen-lg-down .modal-header,\n  .modal-fullscreen-lg-down .modal-footer {\n    border-radius: 0;\n  }\n  .modal-fullscreen-lg-down .modal-body {\n    overflow-y: auto;\n  }\n}\n@media (max-width: 1199.98px) {\n  .modal-fullscreen-xl-down {\n    width: 100vw;\n    max-width: none;\n    height: 100%;\n    margin: 0;\n  }\n  .modal-fullscreen-xl-down .modal-content {\n    height: 100%;\n    border: 0;\n    border-radius: 0;\n  }\n  .modal-fullscreen-xl-down .modal-header,\n  .modal-fullscreen-xl-down .modal-footer {\n    border-radius: 0;\n  }\n  .modal-fullscreen-xl-down .modal-body {\n    overflow-y: auto;\n  }\n}\n.tooltip {\n  --bs-tooltip-zindex: 1080;\n  --bs-tooltip-max-width: 200px;\n  --bs-tooltip-padding-x: 0.5rem;\n  --bs-tooltip-padding-y: 0.25rem;\n  --bs-tooltip-margin: ;\n  --bs-tooltip-font-size: 0.875rem;\n  --bs-tooltip-color: var(--bs-body-bg);\n  --bs-tooltip-bg: var(--bs-emphasis-color);\n  --bs-tooltip-border-radius: var(--bs-border-radius);\n  --bs-tooltip-opacity: 0.9;\n  --bs-tooltip-arrow-width: 0.8rem;\n  --bs-tooltip-arrow-height: 0.4rem;\n  z-index: var(--bs-tooltip-zindex);\n  display: block;\n  margin: var(--bs-tooltip-margin);\n  font-family: var(--bs-font-sans-serif);\n  font-style: normal;\n  font-weight: 400;\n  line-height: 1.5;\n  text-align: left;\n  text-align: start;\n  text-decoration: none;\n  text-shadow: none;\n  text-transform: none;\n  letter-spacing: normal;\n  word-break: normal;\n  white-space: normal;\n  word-spacing: normal;\n  line-break: auto;\n  font-size: var(--bs-tooltip-font-size);\n  word-wrap: break-word;\n  opacity: 0;\n}\n.tooltip.show {\n  opacity: var(--bs-tooltip-opacity);\n}\n.tooltip .tooltip-arrow {\n  display: block;\n  width: var(--bs-tooltip-arrow-width);\n  height: var(--bs-tooltip-arrow-height);\n}\n.tooltip .tooltip-arrow::before {\n  position: absolute;\n  content: \"\";\n  border-color: transparent;\n  border-style: solid;\n}\n\n.bs-tooltip-top .tooltip-arrow, .bs-tooltip-auto[data-popper-placement^=top] .tooltip-arrow {\n  bottom: calc(-1 * var(--bs-tooltip-arrow-height));\n}\n.bs-tooltip-top .tooltip-arrow::before, .bs-tooltip-auto[data-popper-placement^=top] .tooltip-arrow::before {\n  top: -1px;\n  border-width: var(--bs-tooltip-arrow-height) calc(var(--bs-tooltip-arrow-width) * 0.5) 0;\n  border-top-color: var(--bs-tooltip-bg);\n}\n\n/* rtl:begin:ignore */\n.bs-tooltip-end .tooltip-arrow, .bs-tooltip-auto[data-popper-placement^=right] .tooltip-arrow {\n  left: calc(-1 * var(--bs-tooltip-arrow-height));\n  width: var(--bs-tooltip-arrow-height);\n  height: var(--bs-tooltip-arrow-width);\n}\n.bs-tooltip-end .tooltip-arrow::before, .bs-tooltip-auto[data-popper-placement^=right] .tooltip-arrow::before {\n  right: -1px;\n  border-width: calc(var(--bs-tooltip-arrow-width) * 0.5) var(--bs-tooltip-arrow-height) calc(var(--bs-tooltip-arrow-width) * 0.5) 0;\n  border-right-color: var(--bs-tooltip-bg);\n}\n\n/* rtl:end:ignore */\n.bs-tooltip-bottom .tooltip-arrow, .bs-tooltip-auto[data-popper-placement^=bottom] .tooltip-arrow {\n  top: calc(-1 * var(--bs-tooltip-arrow-height));\n}\n.bs-tooltip-bottom .tooltip-arrow::before, .bs-tooltip-auto[data-popper-placement^=bottom] .tooltip-arrow::before {\n  bottom: -1px;\n  border-width: 0 calc(var(--bs-tooltip-arrow-width) * 0.5) var(--bs-tooltip-arrow-height);\n  border-bottom-color: var(--bs-tooltip-bg);\n}\n\n/* rtl:begin:ignore */\n.bs-tooltip-start .tooltip-arrow, .bs-tooltip-auto[data-popper-placement^=left] .tooltip-arrow {\n  right: calc(-1 * var(--bs-tooltip-arrow-height));\n  width: var(--bs-tooltip-arrow-height);\n  height: var(--bs-tooltip-arrow-width);\n}\n.bs-tooltip-start .tooltip-arrow::before, .bs-tooltip-auto[data-popper-placement^=left] .tooltip-arrow::before {\n  left: -1px;\n  border-width: calc(var(--bs-tooltip-arrow-width) * 0.5) 0 calc(var(--bs-tooltip-arrow-width) * 0.5) var(--bs-tooltip-arrow-height);\n  border-left-color: var(--bs-tooltip-bg);\n}\n\n/* rtl:end:ignore */\n.tooltip-inner {\n  max-width: var(--bs-tooltip-max-width);\n  padding: var(--bs-tooltip-padding-y) var(--bs-tooltip-padding-x);\n  color: var(--bs-tooltip-color);\n  text-align: center;\n  background-color: var(--bs-tooltip-bg);\n  border-radius: var(--bs-tooltip-border-radius);\n}\n\n.popover {\n  --bs-popover-zindex: 1070;\n  --bs-popover-max-width: 276px;\n  --bs-popover-font-size: 0.875rem;\n  --bs-popover-bg: var(--bs-body-bg);\n  --bs-popover-border-width: var(--bs-border-width);\n  --bs-popover-border-color: var(--bs-border-color-translucent);\n  --bs-popover-border-radius: var(--bs-border-radius-lg);\n  --bs-popover-inner-border-radius: calc(var(--bs-border-radius-lg) - var(--bs-border-width));\n  --bs-popover-box-shadow: var(--bs-box-shadow);\n  --bs-popover-header-padding-x: 1rem;\n  --bs-popover-header-padding-y: 0.5rem;\n  --bs-popover-header-font-size: 1rem;\n  --bs-popover-header-color: inherit;\n  --bs-popover-header-bg: var(--bs-secondary-bg);\n  --bs-popover-body-padding-x: 1rem;\n  --bs-popover-body-padding-y: 1rem;\n  --bs-popover-body-color: var(--bs-body-color);\n  --bs-popover-arrow-width: 1rem;\n  --bs-popover-arrow-height: 0.5rem;\n  --bs-popover-arrow-border: var(--bs-popover-border-color);\n  z-index: var(--bs-popover-zindex);\n  display: block;\n  max-width: var(--bs-popover-max-width);\n  font-family: var(--bs-font-sans-serif);\n  font-style: normal;\n  font-weight: 400;\n  line-height: 1.5;\n  text-align: left;\n  text-align: start;\n  text-decoration: none;\n  text-shadow: none;\n  text-transform: none;\n  letter-spacing: normal;\n  word-break: normal;\n  white-space: normal;\n  word-spacing: normal;\n  line-break: auto;\n  font-size: var(--bs-popover-font-size);\n  word-wrap: break-word;\n  background-color: var(--bs-popover-bg);\n  background-clip: padding-box;\n  border: var(--bs-popover-border-width) solid var(--bs-popover-border-color);\n  border-radius: var(--bs-popover-border-radius);\n}\n.popover .popover-arrow {\n  display: block;\n  width: var(--bs-popover-arrow-width);\n  height: var(--bs-popover-arrow-height);\n}\n.popover .popover-arrow::before, .popover .popover-arrow::after {\n  position: absolute;\n  display: block;\n  content: \"\";\n  border-color: transparent;\n  border-style: solid;\n  border-width: 0;\n}\n\n.bs-popover-top > .popover-arrow, .bs-popover-auto[data-popper-placement^=top] > .popover-arrow {\n  bottom: calc(-1 * (var(--bs-popover-arrow-height)) - var(--bs-popover-border-width));\n}\n.bs-popover-top > .popover-arrow::before, .bs-popover-auto[data-popper-placement^=top] > .popover-arrow::before, .bs-popover-top > .popover-arrow::after, .bs-popover-auto[data-popper-placement^=top] > .popover-arrow::after {\n  border-width: var(--bs-popover-arrow-height) calc(var(--bs-popover-arrow-width) * 0.5) 0;\n}\n.bs-popover-top > .popover-arrow::before, .bs-popover-auto[data-popper-placement^=top] > .popover-arrow::before {\n  bottom: 0;\n  border-top-color: var(--bs-popover-arrow-border);\n}\n.bs-popover-top > .popover-arrow::after, .bs-popover-auto[data-popper-placement^=top] > .popover-arrow::after {\n  bottom: var(--bs-popover-border-width);\n  border-top-color: var(--bs-popover-bg);\n}\n\n/* rtl:begin:ignore */\n.bs-popover-end > .popover-arrow, .bs-popover-auto[data-popper-placement^=right] > .popover-arrow {\n  left: calc(-1 * (var(--bs-popover-arrow-height)) - var(--bs-popover-border-width));\n  width: var(--bs-popover-arrow-height);\n  height: var(--bs-popover-arrow-width);\n}\n.bs-popover-end > .popover-arrow::before, .bs-popover-auto[data-popper-placement^=right] > .popover-arrow::before, .bs-popover-end > .popover-arrow::after, .bs-popover-auto[data-popper-placement^=right] > .popover-arrow::after {\n  border-width: calc(var(--bs-popover-arrow-width) * 0.5) var(--bs-popover-arrow-height) calc(var(--bs-popover-arrow-width) * 0.5) 0;\n}\n.bs-popover-end > .popover-arrow::before, .bs-popover-auto[data-popper-placement^=right] > .popover-arrow::before {\n  left: 0;\n  border-right-color: var(--bs-popover-arrow-border);\n}\n.bs-popover-end > .popover-arrow::after, .bs-popover-auto[data-popper-placement^=right] > .popover-arrow::after {\n  left: var(--bs-popover-border-width);\n  border-right-color: var(--bs-popover-bg);\n}\n\n/* rtl:end:ignore */\n.bs-popover-bottom > .popover-arrow, .bs-popover-auto[data-popper-placement^=bottom] > .popover-arrow {\n  top: calc(-1 * (var(--bs-popover-arrow-height)) - var(--bs-popover-border-width));\n}\n.bs-popover-bottom > .popover-arrow::before, .bs-popover-auto[data-popper-placement^=bottom] > .popover-arrow::before, .bs-popover-bottom > .popover-arrow::after, .bs-popover-auto[data-popper-placement^=bottom] > .popover-arrow::after {\n  border-width: 0 calc(var(--bs-popover-arrow-width) * 0.5) var(--bs-popover-arrow-height);\n}\n.bs-popover-bottom > .popover-arrow::before, .bs-popover-auto[data-popper-placement^=bottom] > .popover-arrow::before {\n  top: 0;\n  border-bottom-color: var(--bs-popover-arrow-border);\n}\n.bs-popover-bottom > .popover-arrow::after, .bs-popover-auto[data-popper-placement^=bottom] > .popover-arrow::after {\n  top: var(--bs-popover-border-width);\n  border-bottom-color: var(--bs-popover-bg);\n}\n.bs-popover-bottom .popover-header::before, .bs-popover-auto[data-popper-placement^=bottom] .popover-header::before {\n  position: absolute;\n  top: 0;\n  left: 50%;\n  display: block;\n  width: var(--bs-popover-arrow-width);\n  margin-left: calc(-0.5 * var(--bs-popover-arrow-width));\n  content: \"\";\n  border-bottom: var(--bs-popover-border-width) solid var(--bs-popover-header-bg);\n}\n\n/* rtl:begin:ignore */\n.bs-popover-start > .popover-arrow, .bs-popover-auto[data-popper-placement^=left] > .popover-arrow {\n  right: calc(-1 * (var(--bs-popover-arrow-height)) - var(--bs-popover-border-width));\n  width: var(--bs-popover-arrow-height);\n  height: var(--bs-popover-arrow-width);\n}\n.bs-popover-start > .popover-arrow::before, .bs-popover-auto[data-popper-placement^=left] > .popover-arrow::before, .bs-popover-start > .popover-arrow::after, .bs-popover-auto[data-popper-placement^=left] > .popover-arrow::after {\n  border-width: calc(var(--bs-popover-arrow-width) * 0.5) 0 calc(var(--bs-popover-arrow-width) * 0.5) var(--bs-popover-arrow-height);\n}\n.bs-popover-start > .popover-arrow::before, .bs-popover-auto[data-popper-placement^=left] > .popover-arrow::before {\n  right: 0;\n  border-left-color: var(--bs-popover-arrow-border);\n}\n.bs-popover-start > .popover-arrow::after, .bs-popover-auto[data-popper-placement^=left] > .popover-arrow::after {\n  right: var(--bs-popover-border-width);\n  border-left-color: var(--bs-popover-bg);\n}\n\n/* rtl:end:ignore */\n.popover-header {\n  padding: var(--bs-popover-header-padding-y) var(--bs-popover-header-padding-x);\n  margin-bottom: 0;\n  font-size: var(--bs-popover-header-font-size);\n  color: var(--bs-popover-header-color);\n  background-color: var(--bs-popover-header-bg);\n  border-bottom: var(--bs-popover-border-width) solid var(--bs-popover-border-color);\n  border-top-left-radius: var(--bs-popover-inner-border-radius);\n  border-top-right-radius: var(--bs-popover-inner-border-radius);\n}\n.popover-header:empty {\n  display: none;\n}\n\n.popover-body {\n  padding: var(--bs-popover-body-padding-y) var(--bs-popover-body-padding-x);\n  color: var(--bs-popover-body-color);\n}\n\n.carousel {\n  position: relative;\n}\n\n.carousel.pointer-event {\n  touch-action: pan-y;\n}\n\n.carousel-inner {\n  position: relative;\n  width: 100%;\n  overflow: hidden;\n}\n.carousel-inner::after {\n  display: block;\n  clear: both;\n  content: \"\";\n}\n\n.carousel-item {\n  position: relative;\n  display: none;\n  float: left;\n  width: 100%;\n  margin-right: -100%;\n  backface-visibility: hidden;\n  transition: transform 0.6s ease-in-out;\n}\n@media (prefers-reduced-motion: reduce) {\n  .carousel-item {\n    transition: none;\n  }\n}\n\n.carousel-item.active,\n.carousel-item-next,\n.carousel-item-prev {\n  display: block;\n}\n\n.carousel-item-next:not(.carousel-item-start),\n.active.carousel-item-end {\n  transform: translateX(100%);\n}\n\n.carousel-item-prev:not(.carousel-item-end),\n.active.carousel-item-start {\n  transform: translateX(-100%);\n}\n\n.carousel-fade .carousel-item {\n  opacity: 0;\n  transition-property: opacity;\n  transform: none;\n}\n.carousel-fade .carousel-item.active,\n.carousel-fade .carousel-item-next.carousel-item-start,\n.carousel-fade .carousel-item-prev.carousel-item-end {\n  z-index: 1;\n  opacity: 1;\n}\n.carousel-fade .active.carousel-item-start,\n.carousel-fade .active.carousel-item-end {\n  z-index: 0;\n  opacity: 0;\n  transition: opacity 0s 0.6s;\n}\n@media (prefers-reduced-motion: reduce) {\n  .carousel-fade .active.carousel-item-start,\n  .carousel-fade .active.carousel-item-end {\n    transition: none;\n  }\n}\n\n.carousel-control-prev,\n.carousel-control-next {\n  position: absolute;\n  top: 0;\n  bottom: 0;\n  z-index: 1;\n  display: flex;\n  align-items: center;\n  justify-content: center;\n  width: 15%;\n  padding: 0;\n  color: #fff;\n  text-align: center;\n  background: none;\n  border: 0;\n  opacity: 0.5;\n  transition: opacity 0.15s ease;\n}\n@media (prefers-reduced-motion: reduce) {\n  .carousel-control-prev,\n  .carousel-control-next {\n    transition: none;\n  }\n}\n.carousel-control-prev:hover, .carousel-control-prev:focus,\n.carousel-control-next:hover,\n.carousel-control-next:focus {\n  color: #fff;\n  text-decoration: none;\n  outline: 0;\n  opacity: 0.9;\n}\n\n.carousel-control-prev {\n  left: 0;\n}\n\n.carousel-control-next {\n  right: 0;\n}\n\n.carousel-control-prev-icon,\n.carousel-control-next-icon {\n  display: inline-block;\n  width: 2rem;\n  height: 2rem;\n  background-repeat: no-repeat;\n  background-position: 50%;\n  background-size: 100% 100%;\n}\n\n.carousel-control-prev-icon {\n  background-image: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='%23fff'%3e%3cpath d='M11.354 1.646a.5.5 0 0 1 0 .708L5.707 8l5.647 5.646a.5.5 0 0 1-.708.708l-6-6a.5.5 0 0 1 0-.708l6-6a.5.5 0 0 1 .708 0z'/%3e%3c/svg%3e\") /*rtl:url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='%23fff'%3e%3cpath d='M4.646 1.646a.5.5 0 0 1 .708 0l6 6a.5.5 0 0 1 0 .708l-6 6a.5.5 0 0 1-.708-.708L10.293 8 4.646 2.354a.5.5 0 0 1 0-.708z'/%3e%3c/svg%3e\")*/;\n}\n\n.carousel-control-next-icon {\n  background-image: url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='%23fff'%3e%3cpath d='M4.646 1.646a.5.5 0 0 1 .708 0l6 6a.5.5 0 0 1 0 .708l-6 6a.5.5 0 0 1-.708-.708L10.293 8 4.646 2.354a.5.5 0 0 1 0-.708z'/%3e%3c/svg%3e\") /*rtl:url(\"data:image/svg+xml,%3csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='%23fff'%3e%3cpath d='M11.354 1.646a.5.5 0 0 1 0 .708L5.707 8l5.647 5.646a.5.5 0 0 1-.708.708l-6-6a.5.5 0 0 1 0-.708l6-6a.5.5 0 0 1 .708 0z'/%3e%3c/svg%3e\")*/;\n}\n\n.carousel-indicators {\n  position: absolute;\n  right: 0;\n  bottom: 0;\n  left: 0;\n  z-index: 2;\n  display: flex;\n  justify-content: center;\n  padding: 0;\n  margin-right: 15%;\n  margin-bottom: 1rem;\n  margin-left: 15%;\n}\n.carousel-indicators [data-bs-target] {\n  box-sizing: content-box;\n  flex: 0 1 auto;\n  width: 30px;\n  height: 3px;\n  padding: 0;\n  margin-right: 3px;\n  margin-left: 3px;\n  text-indent: -999px;\n  cursor: pointer;\n  background-color: #fff;\n  background-clip: padding-box;\n  border: 0;\n  border-top: 10px solid transparent;\n  border-bottom: 10px solid transparent;\n  opacity: 0.5;\n  transition: opacity 0.6s ease;\n}\n@media (prefers-reduced-motion: reduce) {\n  .carousel-indicators [data-bs-target] {\n    transition: none;\n  }\n}\n.carousel-indicators .active {\n  opacity: 1;\n}\n\n.carousel-caption {\n  position: absolute;\n  right: 15%;\n  bottom: 1.25rem;\n  left: 15%;\n  padding-top: 1.25rem;\n  padding-bottom: 1.25rem;\n  color: #fff;\n  text-align: center;\n}\n\n.carousel-dark .carousel-control-prev-icon,\n.carousel-dark .carousel-control-next-icon {\n  filter: invert(1) grayscale(100);\n}\n.carousel-dark .carousel-indicators [data-bs-target] {\n  background-color: #000;\n}\n.carousel-dark .carousel-caption {\n  color: #000;\n}\n\n[data-bs-theme=dark] .carousel .carousel-control-prev-icon,\n[data-bs-theme=dark] .carousel .carousel-control-next-icon, [data-bs-theme=dark].carousel .carousel-control-prev-icon,\n[data-bs-theme=dark].carousel .carousel-control-next-icon {\n  filter: invert(1) grayscale(100);\n}\n[data-bs-theme=dark] .carousel .carousel-indicators [data-bs-target], [data-bs-theme=dark].carousel .carousel-indicators [data-bs-target] {\n  background-color: #000;\n}\n[data-bs-theme=dark] .carousel .carousel-caption, [data-bs-theme=dark].carousel .carousel-caption {\n  color: #000;\n}\n\n.spinner-grow,\n.spinner-border {\n  display: inline-block;\n  width: var(--bs-spinner-width);\n  height: var(--bs-spinner-height);\n  vertical-align: var(--bs-spinner-vertical-align);\n  border-radius: 50%;\n  animation: var(--bs-spinner-animation-speed) linear infinite var(--bs-spinner-animation-name);\n}\n\n@keyframes spinner-border {\n  to {\n    transform: rotate(360deg) /* rtl:ignore */;\n  }\n}\n.spinner-border {\n  --bs-spinner-width: 2rem;\n  --bs-spinner-height: 2rem;\n  --bs-spinner-vertical-align: -0.125em;\n  --bs-spinner-border-width: 0.25em;\n  --bs-spinner-animation-speed: 0.75s;\n  --bs-spinner-animation-name: spinner-border;\n  border: var(--bs-spinner-border-width) solid currentcolor;\n  border-right-color: transparent;\n}\n\n.spinner-border-sm {\n  --bs-spinner-width: 1rem;\n  --bs-spinner-height: 1rem;\n  --bs-spinner-border-width: 0.2em;\n}\n\n@keyframes spinner-grow {\n  0% {\n    transform: scale(0);\n  }\n  50% {\n    opacity: 1;\n    transform: none;\n  }\n}\n.spinner-grow {\n  --bs-spinner-width: 2rem;\n  --bs-spinner-height: 2rem;\n  --bs-spinner-vertical-align: -0.125em;\n  --bs-spinner-animation-speed: 0.75s;\n  --bs-spinner-animation-name: spinner-grow;\n  background-color: currentcolor;\n  opacity: 0;\n}\n\n.spinner-grow-sm {\n  --bs-spinner-width: 1rem;\n  --bs-spinner-height: 1rem;\n}\n\n@media (prefers-reduced-motion: reduce) {\n  .spinner-border,\n  .spinner-grow {\n    --bs-spinner-animation-speed: 1.5s;\n  }\n}\n.offcanvas, .offcanvas-xl, .offcanvas-lg, .offcanvas-md, .offcanvas-sm {\n  --bs-offcanvas-zindex: 1045;\n  --bs-offcanvas-width: 400px;\n  --bs-offcanvas-height: 30vh;\n  --bs-offcanvas-padding-x: 1rem;\n  --bs-offcanvas-padding-y: 1rem;\n  --bs-offcanvas-color: var(--bs-body-color);\n  --bs-offcanvas-bg: var(--bs-body-bg);\n  --bs-offcanvas-border-width: var(--bs-border-width);\n  --bs-offcanvas-border-color: var(--bs-border-color-translucent);\n  --bs-offcanvas-box-shadow: var(--bs-box-shadow-sm);\n  --bs-offcanvas-transition: transform 0.3s ease-in-out;\n  --bs-offcanvas-title-line-height: 1.5;\n}\n\n@media (max-width: 539.98px) {\n  .offcanvas-sm {\n    position: fixed;\n    bottom: 0;\n    z-index: var(--bs-offcanvas-zindex);\n    display: flex;\n    flex-direction: column;\n    max-width: 100%;\n    color: var(--bs-offcanvas-color);\n    visibility: hidden;\n    background-color: var(--bs-offcanvas-bg);\n    background-clip: padding-box;\n    outline: 0;\n    transition: var(--bs-offcanvas-transition);\n  }\n}\n@media (max-width: 539.98px) and (prefers-reduced-motion: reduce) {\n  .offcanvas-sm {\n    transition: none;\n  }\n}\n@media (max-width: 539.98px) {\n  .offcanvas-sm.offcanvas-start {\n    top: 0;\n    left: 0;\n    width: var(--bs-offcanvas-width);\n    border-right: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateX(-100%);\n  }\n  .offcanvas-sm.offcanvas-end {\n    top: 0;\n    right: 0;\n    width: var(--bs-offcanvas-width);\n    border-left: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateX(100%);\n  }\n  .offcanvas-sm.offcanvas-top {\n    top: 0;\n    right: 0;\n    left: 0;\n    height: var(--bs-offcanvas-height);\n    max-height: 100%;\n    border-bottom: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateY(-100%);\n  }\n  .offcanvas-sm.offcanvas-bottom {\n    right: 0;\n    left: 0;\n    height: var(--bs-offcanvas-height);\n    max-height: 100%;\n    border-top: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateY(100%);\n  }\n  .offcanvas-sm.showing, .offcanvas-sm.show:not(.hiding) {\n    transform: none;\n  }\n  .offcanvas-sm.showing, .offcanvas-sm.hiding, .offcanvas-sm.show {\n    visibility: visible;\n  }\n}\n@media (min-width: 540px) {\n  .offcanvas-sm {\n    --bs-offcanvas-height: auto;\n    --bs-offcanvas-border-width: 0;\n    background-color: transparent !important;\n  }\n  .offcanvas-sm .offcanvas-header {\n    display: none;\n  }\n  .offcanvas-sm .offcanvas-body {\n    display: flex;\n    flex-grow: 0;\n    padding: 0;\n    overflow-y: visible;\n    background-color: transparent !important;\n  }\n}\n\n@media (max-width: 719.98px) {\n  .offcanvas-md {\n    position: fixed;\n    bottom: 0;\n    z-index: var(--bs-offcanvas-zindex);\n    display: flex;\n    flex-direction: column;\n    max-width: 100%;\n    color: var(--bs-offcanvas-color);\n    visibility: hidden;\n    background-color: var(--bs-offcanvas-bg);\n    background-clip: padding-box;\n    outline: 0;\n    transition: var(--bs-offcanvas-transition);\n  }\n}\n@media (max-width: 719.98px) and (prefers-reduced-motion: reduce) {\n  .offcanvas-md {\n    transition: none;\n  }\n}\n@media (max-width: 719.98px) {\n  .offcanvas-md.offcanvas-start {\n    top: 0;\n    left: 0;\n    width: var(--bs-offcanvas-width);\n    border-right: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateX(-100%);\n  }\n  .offcanvas-md.offcanvas-end {\n    top: 0;\n    right: 0;\n    width: var(--bs-offcanvas-width);\n    border-left: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateX(100%);\n  }\n  .offcanvas-md.offcanvas-top {\n    top: 0;\n    right: 0;\n    left: 0;\n    height: var(--bs-offcanvas-height);\n    max-height: 100%;\n    border-bottom: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateY(-100%);\n  }\n  .offcanvas-md.offcanvas-bottom {\n    right: 0;\n    left: 0;\n    height: var(--bs-offcanvas-height);\n    max-height: 100%;\n    border-top: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateY(100%);\n  }\n  .offcanvas-md.showing, .offcanvas-md.show:not(.hiding) {\n    transform: none;\n  }\n  .offcanvas-md.showing, .offcanvas-md.hiding, .offcanvas-md.show {\n    visibility: visible;\n  }\n}\n@media (min-width: 720px) {\n  .offcanvas-md {\n    --bs-offcanvas-height: auto;\n    --bs-offcanvas-border-width: 0;\n    background-color: transparent !important;\n  }\n  .offcanvas-md .offcanvas-header {\n    display: none;\n  }\n  .offcanvas-md .offcanvas-body {\n    display: flex;\n    flex-grow: 0;\n    padding: 0;\n    overflow-y: visible;\n    background-color: transparent !important;\n  }\n}\n\n@media (max-width: 959.98px) {\n  .offcanvas-lg {\n    position: fixed;\n    bottom: 0;\n    z-index: var(--bs-offcanvas-zindex);\n    display: flex;\n    flex-direction: column;\n    max-width: 100%;\n    color: var(--bs-offcanvas-color);\n    visibility: hidden;\n    background-color: var(--bs-offcanvas-bg);\n    background-clip: padding-box;\n    outline: 0;\n    transition: var(--bs-offcanvas-transition);\n  }\n}\n@media (max-width: 959.98px) and (prefers-reduced-motion: reduce) {\n  .offcanvas-lg {\n    transition: none;\n  }\n}\n@media (max-width: 959.98px) {\n  .offcanvas-lg.offcanvas-start {\n    top: 0;\n    left: 0;\n    width: var(--bs-offcanvas-width);\n    border-right: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateX(-100%);\n  }\n  .offcanvas-lg.offcanvas-end {\n    top: 0;\n    right: 0;\n    width: var(--bs-offcanvas-width);\n    border-left: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateX(100%);\n  }\n  .offcanvas-lg.offcanvas-top {\n    top: 0;\n    right: 0;\n    left: 0;\n    height: var(--bs-offcanvas-height);\n    max-height: 100%;\n    border-bottom: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateY(-100%);\n  }\n  .offcanvas-lg.offcanvas-bottom {\n    right: 0;\n    left: 0;\n    height: var(--bs-offcanvas-height);\n    max-height: 100%;\n    border-top: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateY(100%);\n  }\n  .offcanvas-lg.showing, .offcanvas-lg.show:not(.hiding) {\n    transform: none;\n  }\n  .offcanvas-lg.showing, .offcanvas-lg.hiding, .offcanvas-lg.show {\n    visibility: visible;\n  }\n}\n@media (min-width: 960px) {\n  .offcanvas-lg {\n    --bs-offcanvas-height: auto;\n    --bs-offcanvas-border-width: 0;\n    background-color: transparent !important;\n  }\n  .offcanvas-lg .offcanvas-header {\n    display: none;\n  }\n  .offcanvas-lg .offcanvas-body {\n    display: flex;\n    flex-grow: 0;\n    padding: 0;\n    overflow-y: visible;\n    background-color: transparent !important;\n  }\n}\n\n@media (max-width: 1199.98px) {\n  .offcanvas-xl {\n    position: fixed;\n    bottom: 0;\n    z-index: var(--bs-offcanvas-zindex);\n    display: flex;\n    flex-direction: column;\n    max-width: 100%;\n    color: var(--bs-offcanvas-color);\n    visibility: hidden;\n    background-color: var(--bs-offcanvas-bg);\n    background-clip: padding-box;\n    outline: 0;\n    transition: var(--bs-offcanvas-transition);\n  }\n}\n@media (max-width: 1199.98px) and (prefers-reduced-motion: reduce) {\n  .offcanvas-xl {\n    transition: none;\n  }\n}\n@media (max-width: 1199.98px) {\n  .offcanvas-xl.offcanvas-start {\n    top: 0;\n    left: 0;\n    width: var(--bs-offcanvas-width);\n    border-right: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateX(-100%);\n  }\n  .offcanvas-xl.offcanvas-end {\n    top: 0;\n    right: 0;\n    width: var(--bs-offcanvas-width);\n    border-left: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateX(100%);\n  }\n  .offcanvas-xl.offcanvas-top {\n    top: 0;\n    right: 0;\n    left: 0;\n    height: var(--bs-offcanvas-height);\n    max-height: 100%;\n    border-bottom: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateY(-100%);\n  }\n  .offcanvas-xl.offcanvas-bottom {\n    right: 0;\n    left: 0;\n    height: var(--bs-offcanvas-height);\n    max-height: 100%;\n    border-top: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n    transform: translateY(100%);\n  }\n  .offcanvas-xl.showing, .offcanvas-xl.show:not(.hiding) {\n    transform: none;\n  }\n  .offcanvas-xl.showing, .offcanvas-xl.hiding, .offcanvas-xl.show {\n    visibility: visible;\n  }\n}\n@media (min-width: 1200px) {\n  .offcanvas-xl {\n    --bs-offcanvas-height: auto;\n    --bs-offcanvas-border-width: 0;\n    background-color: transparent !important;\n  }\n  .offcanvas-xl .offcanvas-header {\n    display: none;\n  }\n  .offcanvas-xl .offcanvas-body {\n    display: flex;\n    flex-grow: 0;\n    padding: 0;\n    overflow-y: visible;\n    background-color: transparent !important;\n  }\n}\n\n.offcanvas {\n  position: fixed;\n  bottom: 0;\n  z-index: var(--bs-offcanvas-zindex);\n  display: flex;\n  flex-direction: column;\n  max-width: 100%;\n  color: var(--bs-offcanvas-color);\n  visibility: hidden;\n  background-color: var(--bs-offcanvas-bg);\n  background-clip: padding-box;\n  outline: 0;\n  transition: var(--bs-offcanvas-transition);\n}\n@media (prefers-reduced-motion: reduce) {\n  .offcanvas {\n    transition: none;\n  }\n}\n.offcanvas.offcanvas-start {\n  top: 0;\n  left: 0;\n  width: var(--bs-offcanvas-width);\n  border-right: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n  transform: translateX(-100%);\n}\n.offcanvas.offcanvas-end {\n  top: 0;\n  right: 0;\n  width: var(--bs-offcanvas-width);\n  border-left: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n  transform: translateX(100%);\n}\n.offcanvas.offcanvas-top {\n  top: 0;\n  right: 0;\n  left: 0;\n  height: var(--bs-offcanvas-height);\n  max-height: 100%;\n  border-bottom: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n  transform: translateY(-100%);\n}\n.offcanvas.offcanvas-bottom {\n  right: 0;\n  left: 0;\n  height: var(--bs-offcanvas-height);\n  max-height: 100%;\n  border-top: var(--bs-offcanvas-border-width) solid var(--bs-offcanvas-border-color);\n  transform: translateY(100%);\n}\n.offcanvas.showing, .offcanvas.show:not(.hiding) {\n  transform: none;\n}\n.offcanvas.showing, .offcanvas.hiding, .offcanvas.show {\n  visibility: visible;\n}\n\n.offcanvas-backdrop {\n  position: fixed;\n  top: 0;\n  left: 0;\n  z-index: 1040;\n  width: 100vw;\n  height: 100vh;\n  background-color: #000;\n}\n.offcanvas-backdrop.fade {\n  opacity: 0;\n}\n.offcanvas-backdrop.show {\n  opacity: 0.5;\n}\n\n.offcanvas-header {\n  display: flex;\n  align-items: center;\n  padding: var(--bs-offcanvas-padding-y) var(--bs-offcanvas-padding-x);\n}\n.offcanvas-header .btn-close {\n  padding: calc(var(--bs-offcanvas-padding-y) * 0.5) calc(var(--bs-offcanvas-padding-x) * 0.5);\n  margin: calc(-0.5 * var(--bs-offcanvas-padding-y)) calc(-0.5 * var(--bs-offcanvas-padding-x)) calc(-0.5 * var(--bs-offcanvas-padding-y)) auto;\n}\n\n.offcanvas-title {\n  margin-bottom: 0;\n  line-height: var(--bs-offcanvas-title-line-height);\n}\n\n.offcanvas-body {\n  flex-grow: 1;\n  padding: var(--bs-offcanvas-padding-y) var(--bs-offcanvas-padding-x);\n  overflow-y: auto;\n}\n\n.placeholder {\n  display: inline-block;\n  min-height: 1em;\n  vertical-align: middle;\n  cursor: wait;\n  background-color: currentcolor;\n  opacity: 0.5;\n}\n.placeholder.btn::before {\n  display: inline-block;\n  content: \"\";\n}\n\n.placeholder-xs {\n  min-height: 0.6em;\n}\n\n.placeholder-sm {\n  min-height: 0.8em;\n}\n\n.placeholder-lg {\n  min-height: 1.2em;\n}\n\n.placeholder-glow .placeholder {\n  animation: placeholder-glow 2s ease-in-out infinite;\n}\n\n@keyframes placeholder-glow {\n  50% {\n    opacity: 0.2;\n  }\n}\n.placeholder-wave {\n  mask-image: linear-gradient(130deg, #000 55%, rgba(0, 0, 0, 0.8) 75%, #000 95%);\n  mask-size: 200% 100%;\n  animation: placeholder-wave 2s linear infinite;\n}\n\n@keyframes placeholder-wave {\n  100% {\n    mask-position: -200% 0%;\n  }\n}\n.clearfix::after {\n  display: block;\n  clear: both;\n  content: \"\";\n}\n\n.text-bg-primary {\n  color: #fff !important;\n  background-color: RGBA(var(--bs-primary-rgb), var(--bs-bg-opacity, 1)) !important;\n}\n\n.text-bg-secondary {\n  color: #fff !important;\n  background-color: RGBA(var(--bs-secondary-rgb), var(--bs-bg-opacity, 1)) !important;\n}\n\n.text-bg-success {\n  color: #fff !important;\n  background-color: RGBA(var(--bs-success-rgb), var(--bs-bg-opacity, 1)) !important;\n}\n\n.text-bg-info {\n  color: #000 !important;\n  background-color: RGBA(var(--bs-info-rgb), var(--bs-bg-opacity, 1)) !important;\n}\n\n.text-bg-warning {\n  color: #000 !important;\n  background-color: RGBA(var(--bs-warning-rgb), var(--bs-bg-opacity, 1)) !important;\n}\n\n.text-bg-danger {\n  color: #fff !important;\n  background-color: RGBA(var(--bs-danger-rgb), var(--bs-bg-opacity, 1)) !important;\n}\n\n.text-bg-light {\n  color: #000 !important;\n  background-color: RGBA(var(--bs-light-rgb), var(--bs-bg-opacity, 1)) !important;\n}\n\n.text-bg-dark {\n  color: #fff !important;\n  background-color: RGBA(var(--bs-dark-rgb), var(--bs-bg-opacity, 1)) !important;\n}\n\n.link-primary {\n  color: RGBA(var(--bs-primary-rgb), var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(var(--bs-primary-rgb), var(--bs-link-underline-opacity, 1)) !important;\n}\n.link-primary:hover, .link-primary:focus {\n  color: RGBA(10, 88, 202, var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(10, 88, 202, var(--bs-link-underline-opacity, 1)) !important;\n}\n\n.link-secondary {\n  color: RGBA(var(--bs-secondary-rgb), var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(var(--bs-secondary-rgb), var(--bs-link-underline-opacity, 1)) !important;\n}\n.link-secondary:hover, .link-secondary:focus {\n  color: RGBA(86, 94, 100, var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(86, 94, 100, var(--bs-link-underline-opacity, 1)) !important;\n}\n\n.link-success {\n  color: RGBA(var(--bs-success-rgb), var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(var(--bs-success-rgb), var(--bs-link-underline-opacity, 1)) !important;\n}\n.link-success:hover, .link-success:focus {\n  color: RGBA(20, 108, 67, var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(20, 108, 67, var(--bs-link-underline-opacity, 1)) !important;\n}\n\n.link-info {\n  color: RGBA(var(--bs-info-rgb), var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(var(--bs-info-rgb), var(--bs-link-underline-opacity, 1)) !important;\n}\n.link-info:hover, .link-info:focus {\n  color: RGBA(61, 213, 243, var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(61, 213, 243, var(--bs-link-underline-opacity, 1)) !important;\n}\n\n.link-warning {\n  color: RGBA(var(--bs-warning-rgb), var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(var(--bs-warning-rgb), var(--bs-link-underline-opacity, 1)) !important;\n}\n.link-warning:hover, .link-warning:focus {\n  color: RGBA(255, 205, 57, var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(255, 205, 57, var(--bs-link-underline-opacity, 1)) !important;\n}\n\n.link-danger {\n  color: RGBA(var(--bs-danger-rgb), var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(var(--bs-danger-rgb), var(--bs-link-underline-opacity, 1)) !important;\n}\n.link-danger:hover, .link-danger:focus {\n  color: RGBA(176, 42, 55, var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(176, 42, 55, var(--bs-link-underline-opacity, 1)) !important;\n}\n\n.link-light {\n  color: RGBA(var(--bs-light-rgb), var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(var(--bs-light-rgb), var(--bs-link-underline-opacity, 1)) !important;\n}\n.link-light:hover, .link-light:focus {\n  color: RGBA(249, 250, 251, var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(249, 250, 251, var(--bs-link-underline-opacity, 1)) !important;\n}\n\n.link-dark {\n  color: RGBA(var(--bs-dark-rgb), var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(var(--bs-dark-rgb), var(--bs-link-underline-opacity, 1)) !important;\n}\n.link-dark:hover, .link-dark:focus {\n  color: RGBA(26, 30, 33, var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(26, 30, 33, var(--bs-link-underline-opacity, 1)) !important;\n}\n\n.link-body-emphasis {\n  color: RGBA(var(--bs-emphasis-color-rgb), var(--bs-link-opacity, 1)) !important;\n  text-decoration-color: RGBA(var(--bs-emphasis-color-rgb), var(--bs-link-underline-opacity, 1)) !important;\n}\n.link-body-emphasis:hover, .link-body-emphasis:focus {\n  color: RGBA(var(--bs-emphasis-color-rgb), var(--bs-link-opacity, 0.75)) !important;\n  text-decoration-color: RGBA(var(--bs-emphasis-color-rgb), var(--bs-link-underline-opacity, 0.75)) !important;\n}\n\n.focus-ring:focus {\n  outline: 0;\n  box-shadow: var(--bs-focus-ring-x, 0) var(--bs-focus-ring-y, 0) var(--bs-focus-ring-blur, 0) var(--bs-focus-ring-width) var(--bs-focus-ring-color);\n}\n\n.icon-link {\n  display: inline-flex;\n  gap: 0.375rem;\n  align-items: center;\n  text-decoration-color: rgba(var(--bs-link-color-rgb), var(--bs-link-opacity, 0.5));\n  text-underline-offset: 0.25em;\n  backface-visibility: hidden;\n}\n.icon-link > .bi {\n  flex-shrink: 0;\n  width: 1em;\n  height: 1em;\n  fill: currentcolor;\n  transition: 0.2s ease-in-out transform;\n}\n@media (prefers-reduced-motion: reduce) {\n  .icon-link > .bi {\n    transition: none;\n  }\n}\n\n.icon-link-hover:hover > .bi, .icon-link-hover:focus-visible > .bi {\n  transform: var(--bs-icon-link-transform, translate3d(0.25em, 0, 0));\n}\n\n.ratio {\n  position: relative;\n  width: 100%;\n}\n.ratio::before {\n  display: block;\n  padding-top: var(--bs-aspect-ratio);\n  content: \"\";\n}\n.ratio > * {\n  position: absolute;\n  top: 0;\n  left: 0;\n  width: 100%;\n  height: 100%;\n}\n\n.ratio-1x1 {\n  --bs-aspect-ratio: 100%;\n}\n\n.ratio-4x3 {\n  --bs-aspect-ratio: 75%;\n}\n\n.ratio-16x9 {\n  --bs-aspect-ratio: 56.25%;\n}\n\n.ratio-21x9 {\n  --bs-aspect-ratio: 42.8571428571%;\n}\n\n.fixed-top {\n  position: fixed;\n  top: 0;\n  right: 0;\n  left: 0;\n  z-index: 1030;\n}\n\n.fixed-bottom {\n  position: fixed;\n  right: 0;\n  bottom: 0;\n  left: 0;\n  z-index: 1030;\n}\n\n.sticky-top {\n  position: sticky;\n  top: 0;\n  z-index: 1020;\n}\n\n.sticky-bottom {\n  position: sticky;\n  bottom: 0;\n  z-index: 1020;\n}\n\n@media (min-width: 540px) {\n  .sticky-sm-top {\n    position: sticky;\n    top: 0;\n    z-index: 1020;\n  }\n  .sticky-sm-bottom {\n    position: sticky;\n    bottom: 0;\n    z-index: 1020;\n  }\n}\n@media (min-width: 720px) {\n  .sticky-md-top {\n    position: sticky;\n    top: 0;\n    z-index: 1020;\n  }\n  .sticky-md-bottom {\n    position: sticky;\n    bottom: 0;\n    z-index: 1020;\n  }\n}\n@media (min-width: 960px) {\n  .sticky-lg-top {\n    position: sticky;\n    top: 0;\n    z-index: 1020;\n  }\n  .sticky-lg-bottom {\n    position: sticky;\n    bottom: 0;\n    z-index: 1020;\n  }\n}\n@media (min-width: 1200px) {\n  .sticky-xl-top {\n    position: sticky;\n    top: 0;\n    z-index: 1020;\n  }\n  .sticky-xl-bottom {\n    position: sticky;\n    bottom: 0;\n    z-index: 1020;\n  }\n}\n.hstack {\n  display: flex;\n  flex-direction: row;\n  align-items: center;\n  align-self: stretch;\n}\n\n.vstack {\n  display: flex;\n  flex: 1 1 auto;\n  flex-direction: column;\n  align-self: stretch;\n}\n\n.visually-hidden,\n.visually-hidden-focusable:not(:focus):not(:focus-within) {\n  width: 1px !important;\n  height: 1px !important;\n  padding: 0 !important;\n  margin: -1px !important;\n  overflow: hidden !important;\n  clip: rect(0, 0, 0, 0) !important;\n  white-space: nowrap !important;\n  border: 0 !important;\n}\n.visually-hidden:not(caption),\n.visually-hidden-focusable:not(:focus):not(:focus-within):not(caption) {\n  position: absolute !important;\n}\n\n.stretched-link::after {\n  position: absolute;\n  top: 0;\n  right: 0;\n  bottom: 0;\n  left: 0;\n  z-index: 1;\n  content: \"\";\n}\n\n.text-truncate {\n  overflow: hidden;\n  text-overflow: ellipsis;\n  white-space: nowrap;\n}\n\n.vr {\n  display: inline-block;\n  align-self: stretch;\n  width: var(--bs-border-width);\n  min-height: 1em;\n  background-color: currentcolor;\n  opacity: 0.25;\n}\n\n.align-baseline {\n  vertical-align: baseline !important;\n}\n\n.align-top {\n  vertical-align: top !important;\n}\n\n.align-middle {\n  vertical-align: middle !important;\n}\n\n.align-bottom {\n  vertical-align: bottom !important;\n}\n\n.align-text-bottom {\n  vertical-align: text-bottom !important;\n}\n\n.align-text-top {\n  vertical-align: text-top !important;\n}\n\n.float-start {\n  float: left !important;\n}\n\n.float-end {\n  float: right !important;\n}\n\n.float-none {\n  float: none !important;\n}\n\n.object-fit-contain {\n  object-fit: contain !important;\n}\n\n.object-fit-cover {\n  object-fit: cover !important;\n}\n\n.object-fit-fill {\n  object-fit: fill !important;\n}\n\n.object-fit-scale {\n  object-fit: scale-down !important;\n}\n\n.object-fit-none {\n  object-fit: none !important;\n}\n\n.opacity-0 {\n  opacity: 0 !important;\n}\n\n.opacity-25 {\n  opacity: 0.25 !important;\n}\n\n.opacity-50 {\n  opacity: 0.5 !important;\n}\n\n.opacity-75 {\n  opacity: 0.75 !important;\n}\n\n.opacity-100 {\n  opacity: 1 !important;\n}\n\n.overflow-auto {\n  overflow: auto !important;\n}\n\n.overflow-hidden {\n  overflow: hidden !important;\n}\n\n.overflow-visible {\n  overflow: visible !important;\n}\n\n.overflow-scroll {\n  overflow: scroll !important;\n}\n\n.overflow-x-auto {\n  overflow-x: auto !important;\n}\n\n.overflow-x-hidden {\n  overflow-x: hidden !important;\n}\n\n.overflow-x-visible {\n  overflow-x: visible !important;\n}\n\n.overflow-x-scroll {\n  overflow-x: scroll !important;\n}\n\n.overflow-y-auto {\n  overflow-y: auto !important;\n}\n\n.overflow-y-hidden {\n  overflow-y: hidden !important;\n}\n\n.overflow-y-visible {\n  overflow-y: visible !important;\n}\n\n.overflow-y-scroll {\n  overflow-y: scroll !important;\n}\n\n.d-inline {\n  display: inline !important;\n}\n\n.d-inline-block {\n  display: inline-block !important;\n}\n\n.d-block {\n  display: block !important;\n}\n\n.d-grid {\n  display: grid !important;\n}\n\n.d-inline-grid {\n  display: inline-grid !important;\n}\n\n.d-table {\n  display: table !important;\n}\n\n.d-table-row {\n  display: table-row !important;\n}\n\n.d-table-cell {\n  display: table-cell !important;\n}\n\n.d-flex {\n  display: flex !important;\n}\n\n.d-inline-flex {\n  display: inline-flex !important;\n}\n\n.d-none {\n  display: none !important;\n}\n\n.shadow {\n  box-shadow: var(--bs-box-shadow) !important;\n}\n\n.shadow-sm {\n  box-shadow: var(--bs-box-shadow-sm) !important;\n}\n\n.shadow-lg {\n  box-shadow: var(--bs-box-shadow-lg) !important;\n}\n\n.shadow-none {\n  box-shadow: none !important;\n}\n\n.focus-ring-primary {\n  --bs-focus-ring-color: rgba(var(--bs-primary-rgb), var(--bs-focus-ring-opacity));\n}\n\n.focus-ring-secondary {\n  --bs-focus-ring-color: rgba(var(--bs-secondary-rgb), var(--bs-focus-ring-opacity));\n}\n\n.focus-ring-success {\n  --bs-focus-ring-color: rgba(var(--bs-success-rgb), var(--bs-focus-ring-opacity));\n}\n\n.focus-ring-info {\n  --bs-focus-ring-color: rgba(var(--bs-info-rgb), var(--bs-focus-ring-opacity));\n}\n\n.focus-ring-warning {\n  --bs-focus-ring-color: rgba(var(--bs-warning-rgb), var(--bs-focus-ring-opacity));\n}\n\n.focus-ring-danger {\n  --bs-focus-ring-color: rgba(var(--bs-danger-rgb), var(--bs-focus-ring-opacity));\n}\n\n.focus-ring-light {\n  --bs-focus-ring-color: rgba(var(--bs-light-rgb), var(--bs-focus-ring-opacity));\n}\n\n.focus-ring-dark {\n  --bs-focus-ring-color: rgba(var(--bs-dark-rgb), var(--bs-focus-ring-opacity));\n}\n\n.position-static {\n  position: static !important;\n}\n\n.position-relative {\n  position: relative !important;\n}\n\n.position-absolute {\n  position: absolute !important;\n}\n\n.position-fixed {\n  position: fixed !important;\n}\n\n.position-sticky {\n  position: sticky !important;\n}\n\n.top-0 {\n  top: 0 !important;\n}\n\n.top-50 {\n  top: 50% !important;\n}\n\n.top-100 {\n  top: 100% !important;\n}\n\n.bottom-0 {\n  bottom: 0 !important;\n}\n\n.bottom-50 {\n  bottom: 50% !important;\n}\n\n.bottom-100 {\n  bottom: 100% !important;\n}\n\n.start-0 {\n  left: 0 !important;\n}\n\n.start-50 {\n  left: 50% !important;\n}\n\n.start-100 {\n  left: 100% !important;\n}\n\n.end-0 {\n  right: 0 !important;\n}\n\n.end-50 {\n  right: 50% !important;\n}\n\n.end-100 {\n  right: 100% !important;\n}\n\n.translate-middle {\n  transform: translate(-50%, -50%) !important;\n}\n\n.translate-middle-x {\n  transform: translateX(-50%) !important;\n}\n\n.translate-middle-y {\n  transform: translateY(-50%) !important;\n}\n\n.border {\n  border: var(--bs-border-width) var(--bs-border-style) var(--bs-border-color) !important;\n}\n\n.border-0 {\n  border: 0 !important;\n}\n\n.border-top {\n  border-top: var(--bs-border-width) var(--bs-border-style) var(--bs-border-color) !important;\n}\n\n.border-top-0 {\n  border-top: 0 !important;\n}\n\n.border-end {\n  border-right: var(--bs-border-width) var(--bs-border-style) var(--bs-border-color) !important;\n}\n\n.border-end-0 {\n  border-right: 0 !important;\n}\n\n.border-bottom {\n  border-bottom: var(--bs-border-width) var(--bs-border-style) var(--bs-border-color) !important;\n}\n\n.border-bottom-0 {\n  border-bottom: 0 !important;\n}\n\n.border-start {\n  border-left: var(--bs-border-width) var(--bs-border-style) var(--bs-border-color) !important;\n}\n\n.border-start-0 {\n  border-left: 0 !important;\n}\n\n.border-primary {\n  --bs-border-opacity: 1;\n  border-color: rgba(var(--bs-primary-rgb), var(--bs-border-opacity)) !important;\n}\n\n.border-secondary {\n  --bs-border-opacity: 1;\n  border-color: rgba(var(--bs-secondary-rgb), var(--bs-border-opacity)) !important;\n}\n\n.border-success {\n  --bs-border-opacity: 1;\n  border-color: rgba(var(--bs-success-rgb), var(--bs-border-opacity)) !important;\n}\n\n.border-info {\n  --bs-border-opacity: 1;\n  border-color: rgba(var(--bs-info-rgb), var(--bs-border-opacity)) !important;\n}\n\n.border-warning {\n  --bs-border-opacity: 1;\n  border-color: rgba(var(--bs-warning-rgb), var(--bs-border-opacity)) !important;\n}\n\n.border-danger {\n  --bs-border-opacity: 1;\n  border-color: rgba(var(--bs-danger-rgb), var(--bs-border-opacity)) !important;\n}\n\n.border-light {\n  --bs-border-opacity: 1;\n  border-color: rgba(var(--bs-light-rgb), var(--bs-border-opacity)) !important;\n}\n\n.border-dark {\n  --bs-border-opacity: 1;\n  border-color: rgba(var(--bs-dark-rgb), var(--bs-border-opacity)) !important;\n}\n\n.border-black {\n  --bs-border-opacity: 1;\n  border-color: rgba(var(--bs-black-rgb), var(--bs-border-opacity)) !important;\n}\n\n.border-white {\n  --bs-border-opacity: 1;\n  border-color: rgba(var(--bs-white-rgb), var(--bs-border-opacity)) !important;\n}\n\n.border-primary-subtle {\n  border-color: var(--bs-primary-border-subtle) !important;\n}\n\n.border-secondary-subtle {\n  border-color: var(--bs-secondary-border-subtle) !important;\n}\n\n.border-success-subtle {\n  border-color: var(--bs-success-border-subtle) !important;\n}\n\n.border-info-subtle {\n  border-color: var(--bs-info-border-subtle) !important;\n}\n\n.border-warning-subtle {\n  border-color: var(--bs-warning-border-subtle) !important;\n}\n\n.border-danger-subtle {\n  border-color: var(--bs-danger-border-subtle) !important;\n}\n\n.border-light-subtle {\n  border-color: var(--bs-light-border-subtle) !important;\n}\n\n.border-dark-subtle {\n  border-color: var(--bs-dark-border-subtle) !important;\n}\n\n.border-1 {\n  border-width: 1px !important;\n}\n\n.border-2 {\n  border-width: 2px !important;\n}\n\n.border-3 {\n  border-width: 3px !important;\n}\n\n.border-4 {\n  border-width: 4px !important;\n}\n\n.border-5 {\n  border-width: 5px !important;\n}\n\n.border-opacity-10 {\n  --bs-border-opacity: 0.1;\n}\n\n.border-opacity-25 {\n  --bs-border-opacity: 0.25;\n}\n\n.border-opacity-50 {\n  --bs-border-opacity: 0.5;\n}\n\n.border-opacity-75 {\n  --bs-border-opacity: 0.75;\n}\n\n.border-opacity-100 {\n  --bs-border-opacity: 1;\n}\n\n.w-25 {\n  width: 25% !important;\n}\n\n.w-50 {\n  width: 50% !important;\n}\n\n.w-75 {\n  width: 75% !important;\n}\n\n.w-100 {\n  width: 100% !important;\n}\n\n.w-auto {\n  width: auto !important;\n}\n\n.mw-100 {\n  max-width: 100% !important;\n}\n\n.vw-100 {\n  width: 100vw !important;\n}\n\n.min-vw-100 {\n  min-width: 100vw !important;\n}\n\n.h-25 {\n  height: 25% !important;\n}\n\n.h-50 {\n  height: 50% !important;\n}\n\n.h-75 {\n  height: 75% !important;\n}\n\n.h-100 {\n  height: 100% !important;\n}\n\n.h-auto {\n  height: auto !important;\n}\n\n.mh-100 {\n  max-height: 100% !important;\n}\n\n.vh-100 {\n  height: 100vh !important;\n}\n\n.min-vh-100 {\n  min-height: 100vh !important;\n}\n\n.flex-fill {\n  flex: 1 1 auto !important;\n}\n\n.flex-row {\n  flex-direction: row !important;\n}\n\n.flex-column {\n  flex-direction: column !important;\n}\n\n.flex-row-reverse {\n  flex-direction: row-reverse !important;\n}\n\n.flex-column-reverse {\n  flex-direction: column-reverse !important;\n}\n\n.flex-grow-0 {\n  flex-grow: 0 !important;\n}\n\n.flex-grow-1 {\n  flex-grow: 1 !important;\n}\n\n.flex-shrink-0 {\n  flex-shrink: 0 !important;\n}\n\n.flex-shrink-1 {\n  flex-shrink: 1 !important;\n}\n\n.flex-wrap {\n  flex-wrap: wrap !important;\n}\n\n.flex-nowrap {\n  flex-wrap: nowrap !important;\n}\n\n.flex-wrap-reverse {\n  flex-wrap: wrap-reverse !important;\n}\n\n.justify-content-start {\n  justify-content: flex-start !important;\n}\n\n.justify-content-end {\n  justify-content: flex-end !important;\n}\n\n.justify-content-center {\n  justify-content: center !important;\n}\n\n.justify-content-between {\n  justify-content: space-between !important;\n}\n\n.justify-content-around {\n  justify-content: space-around !important;\n}\n\n.justify-content-evenly {\n  justify-content: space-evenly !important;\n}\n\n.align-items-start {\n  align-items: flex-start !important;\n}\n\n.align-items-end {\n  align-items: flex-end !important;\n}\n\n.align-items-center {\n  align-items: center !important;\n}\n\n.align-items-baseline {\n  align-items: baseline !important;\n}\n\n.align-items-stretch {\n  align-items: stretch !important;\n}\n\n.align-content-start {\n  align-content: flex-start !important;\n}\n\n.align-content-end {\n  align-content: flex-end !important;\n}\n\n.align-content-center {\n  align-content: center !important;\n}\n\n.align-content-between {\n  align-content: space-between !important;\n}\n\n.align-content-around {\n  align-content: space-around !important;\n}\n\n.align-content-stretch {\n  align-content: stretch !important;\n}\n\n.align-self-auto {\n  align-self: auto !important;\n}\n\n.align-self-start {\n  align-self: flex-start !important;\n}\n\n.align-self-end {\n  align-self: flex-end !important;\n}\n\n.align-self-center {\n  align-self: center !important;\n}\n\n.align-self-baseline {\n  align-self: baseline !important;\n}\n\n.align-self-stretch {\n  align-self: stretch !important;\n}\n\n.order-first {\n  order: -1 !important;\n}\n\n.order-0 {\n  order: 0 !important;\n}\n\n.order-1 {\n  order: 1 !important;\n}\n\n.order-2 {\n  order: 2 !important;\n}\n\n.order-3 {\n  order: 3 !important;\n}\n\n.order-4 {\n  order: 4 !important;\n}\n\n.order-5 {\n  order: 5 !important;\n}\n\n.order-last {\n  order: 6 !important;\n}\n\n.m-0 {\n  margin: 0 !important;\n}\n\n.m-1 {\n  margin: 0.25rem !important;\n}\n\n.m-2 {\n  margin: 0.5rem !important;\n}\n\n.m-3 {\n  margin: 1rem !important;\n}\n\n.m-4 {\n  margin: 1.5rem !important;\n}\n\n.m-5 {\n  margin: 3rem !important;\n}\n\n.m-auto {\n  margin: auto !important;\n}\n\n.mx-0 {\n  margin-right: 0 !important;\n  margin-left: 0 !important;\n}\n\n.mx-1 {\n  margin-right: 0.25rem !important;\n  margin-left: 0.25rem !important;\n}\n\n.mx-2 {\n  margin-right: 0.5rem !important;\n  margin-left: 0.5rem !important;\n}\n\n.mx-3 {\n  margin-right: 1rem !important;\n  margin-left: 1rem !important;\n}\n\n.mx-4 {\n  margin-right: 1.5rem !important;\n  margin-left: 1.5rem !important;\n}\n\n.mx-5 {\n  margin-right: 3rem !important;\n  margin-left: 3rem !important;\n}\n\n.mx-auto {\n  margin-right: auto !important;\n  margin-left: auto !important;\n}\n\n.my-0 {\n  margin-top: 0 !important;\n  margin-bottom: 0 !important;\n}\n\n.my-1 {\n  margin-top: 0.25rem !important;\n  margin-bottom: 0.25rem !important;\n}\n\n.my-2 {\n  margin-top: 0.5rem !important;\n  margin-bottom: 0.5rem !important;\n}\n\n.my-3 {\n  margin-top: 1rem !important;\n  margin-bottom: 1rem !important;\n}\n\n.my-4 {\n  margin-top: 1.5rem !important;\n  margin-bottom: 1.5rem !important;\n}\n\n.my-5 {\n  margin-top: 3rem !important;\n  margin-bottom: 3rem !important;\n}\n\n.my-auto {\n  margin-top: auto !important;\n  margin-bottom: auto !important;\n}\n\n.mt-0 {\n  margin-top: 0 !important;\n}\n\n.mt-1 {\n  margin-top: 0.25rem !important;\n}\n\n.mt-2 {\n  margin-top: 0.5rem !important;\n}\n\n.mt-3 {\n  margin-top: 1rem !important;\n}\n\n.mt-4 {\n  margin-top: 1.5rem !important;\n}\n\n.mt-5 {\n  margin-top: 3rem !important;\n}\n\n.mt-auto {\n  margin-top: auto !important;\n}\n\n.me-0 {\n  margin-right: 0 !important;\n}\n\n.me-1 {\n  margin-right: 0.25rem !important;\n}\n\n.me-2 {\n  margin-right: 0.5rem !important;\n}\n\n.me-3 {\n  margin-right: 1rem !important;\n}\n\n.me-4 {\n  margin-right: 1.5rem !important;\n}\n\n.me-5 {\n  margin-right: 3rem !important;\n}\n\n.me-auto {\n  margin-right: auto !important;\n}\n\n.mb-0 {\n  margin-bottom: 0 !important;\n}\n\n.mb-1 {\n  margin-bottom: 0.25rem !important;\n}\n\n.mb-2 {\n  margin-bottom: 0.5rem !important;\n}\n\n.mb-3 {\n  margin-bottom: 1rem !important;\n}\n\n.mb-4 {\n  margin-bottom: 1.5rem !important;\n}\n\n.mb-5 {\n  margin-bottom: 3rem !important;\n}\n\n.mb-auto {\n  margin-bottom: auto !important;\n}\n\n.ms-0 {\n  margin-left: 0 !important;\n}\n\n.ms-1 {\n  margin-left: 0.25rem !important;\n}\n\n.ms-2 {\n  margin-left: 0.5rem !important;\n}\n\n.ms-3 {\n  margin-left: 1rem !important;\n}\n\n.ms-4 {\n  margin-left: 1.5rem !important;\n}\n\n.ms-5 {\n  margin-left: 3rem !important;\n}\n\n.ms-auto {\n  margin-left: auto !important;\n}\n\n.p-0 {\n  padding: 0 !important;\n}\n\n.p-1 {\n  padding: 0.25rem !important;\n}\n\n.p-2 {\n  padding: 0.5rem !important;\n}\n\n.p-3 {\n  padding: 1rem !important;\n}\n\n.p-4 {\n  padding: 1.5rem !important;\n}\n\n.p-5 {\n  padding: 3rem !important;\n}\n\n.px-0 {\n  padding-right: 0 !important;\n  padding-left: 0 !important;\n}\n\n.px-1 {\n  padding-right: 0.25rem !important;\n  padding-left: 0.25rem !important;\n}\n\n.px-2 {\n  padding-right: 0.5rem !important;\n  padding-left: 0.5rem !important;\n}\n\n.px-3 {\n  padding-right: 1rem !important;\n  padding-left: 1rem !important;\n}\n\n.px-4 {\n  padding-right: 1.5rem !important;\n  padding-left: 1.5rem !important;\n}\n\n.px-5 {\n  padding-right: 3rem !important;\n  padding-left: 3rem !important;\n}\n\n.py-0 {\n  padding-top: 0 !important;\n  padding-bottom: 0 !important;\n}\n\n.py-1 {\n  padding-top: 0.25rem !important;\n  padding-bottom: 0.25rem !important;\n}\n\n.py-2 {\n  padding-top: 0.5rem !important;\n  padding-bottom: 0.5rem !important;\n}\n\n.py-3 {\n  padding-top: 1rem !important;\n  padding-bottom: 1rem !important;\n}\n\n.py-4 {\n  padding-top: 1.5rem !important;\n  padding-bottom: 1.5rem !important;\n}\n\n.py-5 {\n  padding-top: 3rem !important;\n  padding-bottom: 3rem !important;\n}\n\n.pt-0 {\n  padding-top: 0 !important;\n}\n\n.pt-1 {\n  padding-top: 0.25rem !important;\n}\n\n.pt-2 {\n  padding-top: 0.5rem !important;\n}\n\n.pt-3 {\n  padding-top: 1rem !important;\n}\n\n.pt-4 {\n  padding-top: 1.5rem !important;\n}\n\n.pt-5 {\n  padding-top: 3rem !important;\n}\n\n.pe-0 {\n  padding-right: 0 !important;\n}\n\n.pe-1 {\n  padding-right: 0.25rem !important;\n}\n\n.pe-2 {\n  padding-right: 0.5rem !important;\n}\n\n.pe-3 {\n  padding-right: 1rem !important;\n}\n\n.pe-4 {\n  padding-right: 1.5rem !important;\n}\n\n.pe-5 {\n  padding-right: 3rem !important;\n}\n\n.pb-0 {\n  padding-bottom: 0 !important;\n}\n\n.pb-1 {\n  padding-bottom: 0.25rem !important;\n}\n\n.pb-2 {\n  padding-bottom: 0.5rem !important;\n}\n\n.pb-3 {\n  padding-bottom: 1rem !important;\n}\n\n.pb-4 {\n  padding-bottom: 1.5rem !important;\n}\n\n.pb-5 {\n  padding-bottom: 3rem !important;\n}\n\n.ps-0 {\n  padding-left: 0 !important;\n}\n\n.ps-1 {\n  padding-left: 0.25rem !important;\n}\n\n.ps-2 {\n  padding-left: 0.5rem !important;\n}\n\n.ps-3 {\n  padding-left: 1rem !important;\n}\n\n.ps-4 {\n  padding-left: 1.5rem !important;\n}\n\n.ps-5 {\n  padding-left: 3rem !important;\n}\n\n.gap-0 {\n  gap: 0 !important;\n}\n\n.gap-1 {\n  gap: 0.25rem !important;\n}\n\n.gap-2 {\n  gap: 0.5rem !important;\n}\n\n.gap-3 {\n  gap: 1rem !important;\n}\n\n.gap-4 {\n  gap: 1.5rem !important;\n}\n\n.gap-5 {\n  gap: 3rem !important;\n}\n\n.row-gap-0 {\n  row-gap: 0 !important;\n}\n\n.row-gap-1 {\n  row-gap: 0.25rem !important;\n}\n\n.row-gap-2 {\n  row-gap: 0.5rem !important;\n}\n\n.row-gap-3 {\n  row-gap: 1rem !important;\n}\n\n.row-gap-4 {\n  row-gap: 1.5rem !important;\n}\n\n.row-gap-5 {\n  row-gap: 3rem !important;\n}\n\n.column-gap-0 {\n  column-gap: 0 !important;\n}\n\n.column-gap-1 {\n  column-gap: 0.25rem !important;\n}\n\n.column-gap-2 {\n  column-gap: 0.5rem !important;\n}\n\n.column-gap-3 {\n  column-gap: 1rem !important;\n}\n\n.column-gap-4 {\n  column-gap: 1.5rem !important;\n}\n\n.column-gap-5 {\n  column-gap: 3rem !important;\n}\n\n.font-monospace {\n  font-family: var(--bs-font-monospace) !important;\n}\n\n.fs-1 {\n  font-size: calc(1.375rem + 1.5vw) !important;\n}\n\n.fs-2 {\n  font-size: calc(1.325rem + 0.9vw) !important;\n}\n\n.fs-3 {\n  font-size: calc(1.3rem + 0.6vw) !important;\n}\n\n.fs-4 {\n  font-size: calc(1.275rem + 0.3vw) !important;\n}\n\n.fs-5 {\n  font-size: 1.25rem !important;\n}\n\n.fs-6 {\n  font-size: 1rem !important;\n}\n\n.fst-italic {\n  font-style: italic !important;\n}\n\n.fst-normal {\n  font-style: normal !important;\n}\n\n.fw-lighter {\n  font-weight: lighter !important;\n}\n\n.fw-light {\n  font-weight: 300 !important;\n}\n\n.fw-normal {\n  font-weight: 400 !important;\n}\n\n.fw-medium {\n  font-weight: 500 !important;\n}\n\n.fw-semibold {\n  font-weight: 600 !important;\n}\n\n.fw-bold {\n  font-weight: 700 !important;\n}\n\n.fw-bolder {\n  font-weight: bolder !important;\n}\n\n.lh-1 {\n  line-height: 1 !important;\n}\n\n.lh-sm {\n  line-height: 1.25 !important;\n}\n\n.lh-base {\n  line-height: 1.5 !important;\n}\n\n.lh-lg {\n  line-height: 2 !important;\n}\n\n.text-start {\n  text-align: left !important;\n}\n\n.text-end {\n  text-align: right !important;\n}\n\n.text-center {\n  text-align: center !important;\n}\n\n.text-decoration-none {\n  text-decoration: none !important;\n}\n\n.text-decoration-underline {\n  text-decoration: underline !important;\n}\n\n.text-decoration-line-through {\n  text-decoration: line-through !important;\n}\n\n.text-lowercase {\n  text-transform: lowercase !important;\n}\n\n.text-uppercase {\n  text-transform: uppercase !important;\n}\n\n.text-capitalize {\n  text-transform: capitalize !important;\n}\n\n.text-wrap {\n  white-space: normal !important;\n}\n\n.text-nowrap {\n  white-space: nowrap !important;\n}\n\n/* rtl:begin:remove */\n.text-break {\n  word-wrap: break-word !important;\n  word-break: break-word !important;\n}\n\n/* rtl:end:remove */\n.text-primary {\n  --bs-text-opacity: 1;\n  color: rgba(var(--bs-primary-rgb), var(--bs-text-opacity)) !important;\n}\n\n.text-secondary {\n  --bs-text-opacity: 1;\n  color: rgba(var(--bs-secondary-rgb), var(--bs-text-opacity)) !important;\n}\n\n.text-success {\n  --bs-text-opacity: 1;\n  color: rgba(var(--bs-success-rgb), var(--bs-text-opacity)) !important;\n}\n\n.text-info {\n  --bs-text-opacity: 1;\n  color: rgba(var(--bs-info-rgb), var(--bs-text-opacity)) !important;\n}\n\n.text-warning {\n  --bs-text-opacity: 1;\n  color: rgba(var(--bs-warning-rgb), var(--bs-text-opacity)) !important;\n}\n\n.text-danger {\n  --bs-text-opacity: 1;\n  color: rgba(var(--bs-danger-rgb), var(--bs-text-opacity)) !important;\n}\n\n.text-light {\n  --bs-text-opacity: 1;\n  color: rgba(var(--bs-light-rgb), var(--bs-text-opacity)) !important;\n}\n\n.text-dark {\n  --bs-text-opacity: 1;\n  color: rgba(var(--bs-dark-rgb), var(--bs-text-opacity)) !important;\n}\n\n.text-black {\n  --bs-text-opacity: 1;\n  color: rgba(var(--bs-black-rgb), var(--bs-text-opacity)) !important;\n}\n\n.text-white {\n  --bs-text-opacity: 1;\n  color: rgba(var(--bs-white-rgb), var(--bs-text-opacity)) !important;\n}\n\n.text-body {\n  --bs-text-opacity: 1;\n  color: rgba(var(--bs-body-color-rgb), var(--bs-text-opacity)) !important;\n}\n\n.text-muted {\n  --bs-text-opacity: 1;\n  color: var(--bs-secondary-color) !important;\n}\n\n.text-black-50 {\n  --bs-text-opacity: 1;\n  color: rgba(0, 0, 0, 0.5) !important;\n}\n\n.text-white-50 {\n  --bs-text-opacity: 1;\n  color: rgba(255, 255, 255, 0.5) !important;\n}\n\n.text-body-secondary {\n  --bs-text-opacity: 1;\n  color: var(--bs-secondary-color) !important;\n}\n\n.text-body-tertiary {\n  --bs-text-opacity: 1;\n  color: var(--bs-tertiary-color) !important;\n}\n\n.text-body-emphasis {\n  --bs-text-opacity: 1;\n  color: var(--bs-emphasis-color) !important;\n}\n\n.text-reset {\n  --bs-text-opacity: 1;\n  color: inherit !important;\n}\n\n.text-opacity-25 {\n  --bs-text-opacity: 0.25;\n}\n\n.text-opacity-50 {\n  --bs-text-opacity: 0.5;\n}\n\n.text-opacity-75 {\n  --bs-text-opacity: 0.75;\n}\n\n.text-opacity-100 {\n  --bs-text-opacity: 1;\n}\n\n.text-primary-emphasis {\n  color: var(--bs-primary-text-emphasis) !important;\n}\n\n.text-secondary-emphasis {\n  color: var(--bs-secondary-text-emphasis) !important;\n}\n\n.text-success-emphasis {\n  color: var(--bs-success-text-emphasis) !important;\n}\n\n.text-info-emphasis {\n  color: var(--bs-info-text-emphasis) !important;\n}\n\n.text-warning-emphasis {\n  color: var(--bs-warning-text-emphasis) !important;\n}\n\n.text-danger-emphasis {\n  color: var(--bs-danger-text-emphasis) !important;\n}\n\n.text-light-emphasis {\n  color: var(--bs-light-text-emphasis) !important;\n}\n\n.text-dark-emphasis {\n  color: var(--bs-dark-text-emphasis) !important;\n}\n\n.link-opacity-10 {\n  --bs-link-opacity: 0.1;\n}\n\n.link-opacity-10-hover:hover {\n  --bs-link-opacity: 0.1;\n}\n\n.link-opacity-25 {\n  --bs-link-opacity: 0.25;\n}\n\n.link-opacity-25-hover:hover {\n  --bs-link-opacity: 0.25;\n}\n\n.link-opacity-50 {\n  --bs-link-opacity: 0.5;\n}\n\n.link-opacity-50-hover:hover {\n  --bs-link-opacity: 0.5;\n}\n\n.link-opacity-75 {\n  --bs-link-opacity: 0.75;\n}\n\n.link-opacity-75-hover:hover {\n  --bs-link-opacity: 0.75;\n}\n\n.link-opacity-100 {\n  --bs-link-opacity: 1;\n}\n\n.link-opacity-100-hover:hover {\n  --bs-link-opacity: 1;\n}\n\n.link-offset-1 {\n  text-underline-offset: 0.125em !important;\n}\n\n.link-offset-1-hover:hover {\n  text-underline-offset: 0.125em !important;\n}\n\n.link-offset-2 {\n  text-underline-offset: 0.25em !important;\n}\n\n.link-offset-2-hover:hover {\n  text-underline-offset: 0.25em !important;\n}\n\n.link-offset-3 {\n  text-underline-offset: 0.375em !important;\n}\n\n.link-offset-3-hover:hover {\n  text-underline-offset: 0.375em !important;\n}\n\n.link-underline-primary {\n  --bs-link-underline-opacity: 1;\n  text-decoration-color: rgba(var(--bs-primary-rgb), var(--bs-link-underline-opacity)) !important;\n}\n\n.link-underline-secondary {\n  --bs-link-underline-opacity: 1;\n  text-decoration-color: rgba(var(--bs-secondary-rgb), var(--bs-link-underline-opacity)) !important;\n}\n\n.link-underline-success {\n  --bs-link-underline-opacity: 1;\n  text-decoration-color: rgba(var(--bs-success-rgb), var(--bs-link-underline-opacity)) !important;\n}\n\n.link-underline-info {\n  --bs-link-underline-opacity: 1;\n  text-decoration-color: rgba(var(--bs-info-rgb), var(--bs-link-underline-opacity)) !important;\n}\n\n.link-underline-warning {\n  --bs-link-underline-opacity: 1;\n  text-decoration-color: rgba(var(--bs-warning-rgb), var(--bs-link-underline-opacity)) !important;\n}\n\n.link-underline-danger {\n  --bs-link-underline-opacity: 1;\n  text-decoration-color: rgba(var(--bs-danger-rgb), var(--bs-link-underline-opacity)) !important;\n}\n\n.link-underline-light {\n  --bs-link-underline-opacity: 1;\n  text-decoration-color: rgba(var(--bs-light-rgb), var(--bs-link-underline-opacity)) !important;\n}\n\n.link-underline-dark {\n  --bs-link-underline-opacity: 1;\n  text-decoration-color: rgba(var(--bs-dark-rgb), var(--bs-link-underline-opacity)) !important;\n}\n\n.link-underline {\n  --bs-link-underline-opacity: 1;\n  text-decoration-color: rgba(var(--bs-link-color-rgb), var(--bs-link-underline-opacity, 1)) !important;\n}\n\n.link-underline-opacity-0 {\n  --bs-link-underline-opacity: 0;\n}\n\n.link-underline-opacity-0-hover:hover {\n  --bs-link-underline-opacity: 0;\n}\n\n.link-underline-opacity-10 {\n  --bs-link-underline-opacity: 0.1;\n}\n\n.link-underline-opacity-10-hover:hover {\n  --bs-link-underline-opacity: 0.1;\n}\n\n.link-underline-opacity-25 {\n  --bs-link-underline-opacity: 0.25;\n}\n\n.link-underline-opacity-25-hover:hover {\n  --bs-link-underline-opacity: 0.25;\n}\n\n.link-underline-opacity-50 {\n  --bs-link-underline-opacity: 0.5;\n}\n\n.link-underline-opacity-50-hover:hover {\n  --bs-link-underline-opacity: 0.5;\n}\n\n.link-underline-opacity-75 {\n  --bs-link-underline-opacity: 0.75;\n}\n\n.link-underline-opacity-75-hover:hover {\n  --bs-link-underline-opacity: 0.75;\n}\n\n.link-underline-opacity-100 {\n  --bs-link-underline-opacity: 1;\n}\n\n.link-underline-opacity-100-hover:hover {\n  --bs-link-underline-opacity: 1;\n}\n\n.bg-primary {\n  --bs-bg-opacity: 1;\n  background-color: rgba(var(--bs-primary-rgb), var(--bs-bg-opacity)) !important;\n}\n\n.bg-secondary {\n  --bs-bg-opacity: 1;\n  background-color: rgba(var(--bs-secondary-rgb), var(--bs-bg-opacity)) !important;\n}\n\n.bg-success {\n  --bs-bg-opacity: 1;\n  background-color: rgba(var(--bs-success-rgb), var(--bs-bg-opacity)) !important;\n}\n\n.bg-info {\n  --bs-bg-opacity: 1;\n  background-color: rgba(var(--bs-info-rgb), var(--bs-bg-opacity)) !important;\n}\n\n.bg-warning {\n  --bs-bg-opacity: 1;\n  background-color: rgba(var(--bs-warning-rgb), var(--bs-bg-opacity)) !important;\n}\n\n.bg-danger {\n  --bs-bg-opacity: 1;\n  background-color: rgba(var(--bs-danger-rgb), var(--bs-bg-opacity)) !important;\n}\n\n.bg-light {\n  --bs-bg-opacity: 1;\n  background-color: rgba(var(--bs-light-rgb), var(--bs-bg-opacity)) !important;\n}\n\n.bg-dark {\n  --bs-bg-opacity: 1;\n  background-color: rgba(var(--bs-dark-rgb), var(--bs-bg-opacity)) !important;\n}\n\n.bg-black {\n  --bs-bg-opacity: 1;\n  background-color: rgba(var(--bs-black-rgb), var(--bs-bg-opacity)) !important;\n}\n\n.bg-white {\n  --bs-bg-opacity: 1;\n  background-color: rgba(var(--bs-white-rgb), var(--bs-bg-opacity)) !important;\n}\n\n.bg-body {\n  --bs-bg-opacity: 1;\n  background-color: rgba(var(--bs-body-bg-rgb), var(--bs-bg-opacity)) !important;\n}\n\n.bg-transparent {\n  --bs-bg-opacity: 1;\n  background-color: transparent !important;\n}\n\n.bg-body-secondary {\n  --bs-bg-opacity: 1;\n  background-color: rgba(var(--bs-secondary-bg-rgb), var(--bs-bg-opacity)) !important;\n}\n\n.bg-body-tertiary {\n  --bs-bg-opacity: 1;\n  background-color: rgba(var(--bs-tertiary-bg-rgb), var(--bs-bg-opacity)) !important;\n}\n\n.bg-opacity-10 {\n  --bs-bg-opacity: 0.1;\n}\n\n.bg-opacity-25 {\n  --bs-bg-opacity: 0.25;\n}\n\n.bg-opacity-50 {\n  --bs-bg-opacity: 0.5;\n}\n\n.bg-opacity-75 {\n  --bs-bg-opacity: 0.75;\n}\n\n.bg-opacity-100 {\n  --bs-bg-opacity: 1;\n}\n\n.bg-primary-subtle {\n  background-color: var(--bs-primary-bg-subtle) !important;\n}\n\n.bg-secondary-subtle {\n  background-color: var(--bs-secondary-bg-subtle) !important;\n}\n\n.bg-success-subtle {\n  background-color: var(--bs-success-bg-subtle) !important;\n}\n\n.bg-info-subtle {\n  background-color: var(--bs-info-bg-subtle) !important;\n}\n\n.bg-warning-subtle {\n  background-color: var(--bs-warning-bg-subtle) !important;\n}\n\n.bg-danger-subtle {\n  background-color: var(--bs-danger-bg-subtle) !important;\n}\n\n.bg-light-subtle {\n  background-color: var(--bs-light-bg-subtle) !important;\n}\n\n.bg-dark-subtle {\n  background-color: var(--bs-dark-bg-subtle) !important;\n}\n\n.bg-gradient {\n  background-image: var(--bs-gradient) !important;\n}\n\n.user-select-all {\n  user-select: all !important;\n}\n\n.user-select-auto {\n  user-select: auto !important;\n}\n\n.user-select-none {\n  user-select: none !important;\n}\n\n.pe-none {\n  pointer-events: none !important;\n}\n\n.pe-auto {\n  pointer-events: auto !important;\n}\n\n.rounded {\n  border-radius: var(--bs-border-radius) !important;\n}\n\n.rounded-0 {\n  border-radius: 0 !important;\n}\n\n.rounded-1 {\n  border-radius: var(--bs-border-radius-sm) !important;\n}\n\n.rounded-2 {\n  border-radius: var(--bs-border-radius) !important;\n}\n\n.rounded-3 {\n  border-radius: var(--bs-border-radius-lg) !important;\n}\n\n.rounded-4 {\n  border-radius: var(--bs-border-radius-xl) !important;\n}\n\n.rounded-5 {\n  border-radius: var(--bs-border-radius-xxl) !important;\n}\n\n.rounded-circle {\n  border-radius: 50% !important;\n}\n\n.rounded-pill {\n  border-radius: var(--bs-border-radius-pill) !important;\n}\n\n.rounded-top {\n  border-top-left-radius: var(--bs-border-radius) !important;\n  border-top-right-radius: var(--bs-border-radius) !important;\n}\n\n.rounded-top-0 {\n  border-top-left-radius: 0 !important;\n  border-top-right-radius: 0 !important;\n}\n\n.rounded-top-1 {\n  border-top-left-radius: var(--bs-border-radius-sm) !important;\n  border-top-right-radius: var(--bs-border-radius-sm) !important;\n}\n\n.rounded-top-2 {\n  border-top-left-radius: var(--bs-border-radius) !important;\n  border-top-right-radius: var(--bs-border-radius) !important;\n}\n\n.rounded-top-3 {\n  border-top-left-radius: var(--bs-border-radius-lg) !important;\n  border-top-right-radius: var(--bs-border-radius-lg) !important;\n}\n\n.rounded-top-4 {\n  border-top-left-radius: var(--bs-border-radius-xl) !important;\n  border-top-right-radius: var(--bs-border-radius-xl) !important;\n}\n\n.rounded-top-5 {\n  border-top-left-radius: var(--bs-border-radius-xxl) !important;\n  border-top-right-radius: var(--bs-border-radius-xxl) !important;\n}\n\n.rounded-top-circle {\n  border-top-left-radius: 50% !important;\n  border-top-right-radius: 50% !important;\n}\n\n.rounded-top-pill {\n  border-top-left-radius: var(--bs-border-radius-pill) !important;\n  border-top-right-radius: var(--bs-border-radius-pill) !important;\n}\n\n.rounded-end {\n  border-top-right-radius: var(--bs-border-radius) !important;\n  border-bottom-right-radius: var(--bs-border-radius) !important;\n}\n\n.rounded-end-0 {\n  border-top-right-radius: 0 !important;\n  border-bottom-right-radius: 0 !important;\n}\n\n.rounded-end-1 {\n  border-top-right-radius: var(--bs-border-radius-sm) !important;\n  border-bottom-right-radius: var(--bs-border-radius-sm) !important;\n}\n\n.rounded-end-2 {\n  border-top-right-radius: var(--bs-border-radius) !important;\n  border-bottom-right-radius: var(--bs-border-radius) !important;\n}\n\n.rounded-end-3 {\n  border-top-right-radius: var(--bs-border-radius-lg) !important;\n  border-bottom-right-radius: var(--bs-border-radius-lg) !important;\n}\n\n.rounded-end-4 {\n  border-top-right-radius: var(--bs-border-radius-xl) !important;\n  border-bottom-right-radius: var(--bs-border-radius-xl) !important;\n}\n\n.rounded-end-5 {\n  border-top-right-radius: var(--bs-border-radius-xxl) !important;\n  border-bottom-right-radius: var(--bs-border-radius-xxl) !important;\n}\n\n.rounded-end-circle {\n  border-top-right-radius: 50% !important;\n  border-bottom-right-radius: 50% !important;\n}\n\n.rounded-end-pill {\n  border-top-right-radius: var(--bs-border-radius-pill) !important;\n  border-bottom-right-radius: var(--bs-border-radius-pill) !important;\n}\n\n.rounded-bottom {\n  border-bottom-right-radius: var(--bs-border-radius) !important;\n  border-bottom-left-radius: var(--bs-border-radius) !important;\n}\n\n.rounded-bottom-0 {\n  border-bottom-right-radius: 0 !important;\n  border-bottom-left-radius: 0 !important;\n}\n\n.rounded-bottom-1 {\n  border-bottom-right-radius: var(--bs-border-radius-sm) !important;\n  border-bottom-left-radius: var(--bs-border-radius-sm) !important;\n}\n\n.rounded-bottom-2 {\n  border-bottom-right-radius: var(--bs-border-radius) !important;\n  border-bottom-left-radius: var(--bs-border-radius) !important;\n}\n\n.rounded-bottom-3 {\n  border-bottom-right-radius: var(--bs-border-radius-lg) !important;\n  border-bottom-left-radius: var(--bs-border-radius-lg) !important;\n}\n\n.rounded-bottom-4 {\n  border-bottom-right-radius: var(--bs-border-radius-xl) !important;\n  border-bottom-left-radius: var(--bs-border-radius-xl) !important;\n}\n\n.rounded-bottom-5 {\n  border-bottom-right-radius: var(--bs-border-radius-xxl) !important;\n  border-bottom-left-radius: var(--bs-border-radius-xxl) !important;\n}\n\n.rounded-bottom-circle {\n  border-bottom-right-radius: 50% !important;\n  border-bottom-left-radius: 50% !important;\n}\n\n.rounded-bottom-pill {\n  border-bottom-right-radius: var(--bs-border-radius-pill) !important;\n  border-bottom-left-radius: var(--bs-border-radius-pill) !important;\n}\n\n.rounded-start {\n  border-bottom-left-radius: var(--bs-border-radius) !important;\n  border-top-left-radius: var(--bs-border-radius) !important;\n}\n\n.rounded-start-0 {\n  border-bottom-left-radius: 0 !important;\n  border-top-left-radius: 0 !important;\n}\n\n.rounded-start-1 {\n  border-bottom-left-radius: var(--bs-border-radius-sm) !important;\n  border-top-left-radius: var(--bs-border-radius-sm) !important;\n}\n\n.rounded-start-2 {\n  border-bottom-left-radius: var(--bs-border-radius) !important;\n  border-top-left-radius: var(--bs-border-radius) !important;\n}\n\n.rounded-start-3 {\n  border-bottom-left-radius: var(--bs-border-radius-lg) !important;\n  border-top-left-radius: var(--bs-border-radius-lg) !important;\n}\n\n.rounded-start-4 {\n  border-bottom-left-radius: var(--bs-border-radius-xl) !important;\n  border-top-left-radius: var(--bs-border-radius-xl) !important;\n}\n\n.rounded-start-5 {\n  border-bottom-left-radius: var(--bs-border-radius-xxl) !important;\n  border-top-left-radius: var(--bs-border-radius-xxl) !important;\n}\n\n.rounded-start-circle {\n  border-bottom-left-radius: 50% !important;\n  border-top-left-radius: 50% !important;\n}\n\n.rounded-start-pill {\n  border-bottom-left-radius: var(--bs-border-radius-pill) !important;\n  border-top-left-radius: var(--bs-border-radius-pill) !important;\n}\n\n.visible {\n  visibility: visible !important;\n}\n\n.invisible {\n  visibility: hidden !important;\n}\n\n.z-n1 {\n  z-index: -1 !important;\n}\n\n.z-0 {\n  z-index: 0 !important;\n}\n\n.z-1 {\n  z-index: 1 !important;\n}\n\n.z-2 {\n  z-index: 2 !important;\n}\n\n.z-3 {\n  z-index: 3 !important;\n}\n\n@media (min-width: 540px) {\n  .float-sm-start {\n    float: left !important;\n  }\n  .float-sm-end {\n    float: right !important;\n  }\n  .float-sm-none {\n    float: none !important;\n  }\n  .object-fit-sm-contain {\n    object-fit: contain !important;\n  }\n  .object-fit-sm-cover {\n    object-fit: cover !important;\n  }\n  .object-fit-sm-fill {\n    object-fit: fill !important;\n  }\n  .object-fit-sm-scale {\n    object-fit: scale-down !important;\n  }\n  .object-fit-sm-none {\n    object-fit: none !important;\n  }\n  .d-sm-inline {\n    display: inline !important;\n  }\n  .d-sm-inline-block {\n    display: inline-block !important;\n  }\n  .d-sm-block {\n    display: block !important;\n  }\n  .d-sm-grid {\n    display: grid !important;\n  }\n  .d-sm-inline-grid {\n    display: inline-grid !important;\n  }\n  .d-sm-table {\n    display: table !important;\n  }\n  .d-sm-table-row {\n    display: table-row !important;\n  }\n  .d-sm-table-cell {\n    display: table-cell !important;\n  }\n  .d-sm-flex {\n    display: flex !important;\n  }\n  .d-sm-inline-flex {\n    display: inline-flex !important;\n  }\n  .d-sm-none {\n    display: none !important;\n  }\n  .flex-sm-fill {\n    flex: 1 1 auto !important;\n  }\n  .flex-sm-row {\n    flex-direction: row !important;\n  }\n  .flex-sm-column {\n    flex-direction: column !important;\n  }\n  .flex-sm-row-reverse {\n    flex-direction: row-reverse !important;\n  }\n  .flex-sm-column-reverse {\n    flex-direction: column-reverse !important;\n  }\n  .flex-sm-grow-0 {\n    flex-grow: 0 !important;\n  }\n  .flex-sm-grow-1 {\n    flex-grow: 1 !important;\n  }\n  .flex-sm-shrink-0 {\n    flex-shrink: 0 !important;\n  }\n  .flex-sm-shrink-1 {\n    flex-shrink: 1 !important;\n  }\n  .flex-sm-wrap {\n    flex-wrap: wrap !important;\n  }\n  .flex-sm-nowrap {\n    flex-wrap: nowrap !important;\n  }\n  .flex-sm-wrap-reverse {\n    flex-wrap: wrap-reverse !important;\n  }\n  .justify-content-sm-start {\n    justify-content: flex-start !important;\n  }\n  .justify-content-sm-end {\n    justify-content: flex-end !important;\n  }\n  .justify-content-sm-center {\n    justify-content: center !important;\n  }\n  .justify-content-sm-between {\n    justify-content: space-between !important;\n  }\n  .justify-content-sm-around {\n    justify-content: space-around !important;\n  }\n  .justify-content-sm-evenly {\n    justify-content: space-evenly !important;\n  }\n  .align-items-sm-start {\n    align-items: flex-start !important;\n  }\n  .align-items-sm-end {\n    align-items: flex-end !important;\n  }\n  .align-items-sm-center {\n    align-items: center !important;\n  }\n  .align-items-sm-baseline {\n    align-items: baseline !important;\n  }\n  .align-items-sm-stretch {\n    align-items: stretch !important;\n  }\n  .align-content-sm-start {\n    align-content: flex-start !important;\n  }\n  .align-content-sm-end {\n    align-content: flex-end !important;\n  }\n  .align-content-sm-center {\n    align-content: center !important;\n  }\n  .align-content-sm-between {\n    align-content: space-between !important;\n  }\n  .align-content-sm-around {\n    align-content: space-around !important;\n  }\n  .align-content-sm-stretch {\n    align-content: stretch !important;\n  }\n  .align-self-sm-auto {\n    align-self: auto !important;\n  }\n  .align-self-sm-start {\n    align-self: flex-start !important;\n  }\n  .align-self-sm-end {\n    align-self: flex-end !important;\n  }\n  .align-self-sm-center {\n    align-self: center !important;\n  }\n  .align-self-sm-baseline {\n    align-self: baseline !important;\n  }\n  .align-self-sm-stretch {\n    align-self: stretch !important;\n  }\n  .order-sm-first {\n    order: -1 !important;\n  }\n  .order-sm-0 {\n    order: 0 !important;\n  }\n  .order-sm-1 {\n    order: 1 !important;\n  }\n  .order-sm-2 {\n    order: 2 !important;\n  }\n  .order-sm-3 {\n    order: 3 !important;\n  }\n  .order-sm-4 {\n    order: 4 !important;\n  }\n  .order-sm-5 {\n    order: 5 !important;\n  }\n  .order-sm-last {\n    order: 6 !important;\n  }\n  .m-sm-0 {\n    margin: 0 !important;\n  }\n  .m-sm-1 {\n    margin: 0.25rem !important;\n  }\n  .m-sm-2 {\n    margin: 0.5rem !important;\n  }\n  .m-sm-3 {\n    margin: 1rem !important;\n  }\n  .m-sm-4 {\n    margin: 1.5rem !important;\n  }\n  .m-sm-5 {\n    margin: 3rem !important;\n  }\n  .m-sm-auto {\n    margin: auto !important;\n  }\n  .mx-sm-0 {\n    margin-right: 0 !important;\n    margin-left: 0 !important;\n  }\n  .mx-sm-1 {\n    margin-right: 0.25rem !important;\n    margin-left: 0.25rem !important;\n  }\n  .mx-sm-2 {\n    margin-right: 0.5rem !important;\n    margin-left: 0.5rem !important;\n  }\n  .mx-sm-3 {\n    margin-right: 1rem !important;\n    margin-left: 1rem !important;\n  }\n  .mx-sm-4 {\n    margin-right: 1.5rem !important;\n    margin-left: 1.5rem !important;\n  }\n  .mx-sm-5 {\n    margin-right: 3rem !important;\n    margin-left: 3rem !important;\n  }\n  .mx-sm-auto {\n    margin-right: auto !important;\n    margin-left: auto !important;\n  }\n  .my-sm-0 {\n    margin-top: 0 !important;\n    margin-bottom: 0 !important;\n  }\n  .my-sm-1 {\n    margin-top: 0.25rem !important;\n    margin-bottom: 0.25rem !important;\n  }\n  .my-sm-2 {\n    margin-top: 0.5rem !important;\n    margin-bottom: 0.5rem !important;\n  }\n  .my-sm-3 {\n    margin-top: 1rem !important;\n    margin-bottom: 1rem !important;\n  }\n  .my-sm-4 {\n    margin-top: 1.5rem !important;\n    margin-bottom: 1.5rem !important;\n  }\n  .my-sm-5 {\n    margin-top: 3rem !important;\n    margin-bottom: 3rem !important;\n  }\n  .my-sm-auto {\n    margin-top: auto !important;\n    margin-bottom: auto !important;\n  }\n  .mt-sm-0 {\n    margin-top: 0 !important;\n  }\n  .mt-sm-1 {\n    margin-top: 0.25rem !important;\n  }\n  .mt-sm-2 {\n    margin-top: 0.5rem !important;\n  }\n  .mt-sm-3 {\n    margin-top: 1rem !important;\n  }\n  .mt-sm-4 {\n    margin-top: 1.5rem !important;\n  }\n  .mt-sm-5 {\n    margin-top: 3rem !important;\n  }\n  .mt-sm-auto {\n    margin-top: auto !important;\n  }\n  .me-sm-0 {\n    margin-right: 0 !important;\n  }\n  .me-sm-1 {\n    margin-right: 0.25rem !important;\n  }\n  .me-sm-2 {\n    margin-right: 0.5rem !important;\n  }\n  .me-sm-3 {\n    margin-right: 1rem !important;\n  }\n  .me-sm-4 {\n    margin-right: 1.5rem !important;\n  }\n  .me-sm-5 {\n    margin-right: 3rem !important;\n  }\n  .me-sm-auto {\n    margin-right: auto !important;\n  }\n  .mb-sm-0 {\n    margin-bottom: 0 !important;\n  }\n  .mb-sm-1 {\n    margin-bottom: 0.25rem !important;\n  }\n  .mb-sm-2 {\n    margin-bottom: 0.5rem !important;\n  }\n  .mb-sm-3 {\n    margin-bottom: 1rem !important;\n  }\n  .mb-sm-4 {\n    margin-bottom: 1.5rem !important;\n  }\n  .mb-sm-5 {\n    margin-bottom: 3rem !important;\n  }\n  .mb-sm-auto {\n    margin-bottom: auto !important;\n  }\n  .ms-sm-0 {\n    margin-left: 0 !important;\n  }\n  .ms-sm-1 {\n    margin-left: 0.25rem !important;\n  }\n  .ms-sm-2 {\n    margin-left: 0.5rem !important;\n  }\n  .ms-sm-3 {\n    margin-left: 1rem !important;\n  }\n  .ms-sm-4 {\n    margin-left: 1.5rem !important;\n  }\n  .ms-sm-5 {\n    margin-left: 3rem !important;\n  }\n  .ms-sm-auto {\n    margin-left: auto !important;\n  }\n  .p-sm-0 {\n    padding: 0 !important;\n  }\n  .p-sm-1 {\n    padding: 0.25rem !important;\n  }\n  .p-sm-2 {\n    padding: 0.5rem !important;\n  }\n  .p-sm-3 {\n    padding: 1rem !important;\n  }\n  .p-sm-4 {\n    padding: 1.5rem !important;\n  }\n  .p-sm-5 {\n    padding: 3rem !important;\n  }\n  .px-sm-0 {\n    padding-right: 0 !important;\n    padding-left: 0 !important;\n  }\n  .px-sm-1 {\n    padding-right: 0.25rem !important;\n    padding-left: 0.25rem !important;\n  }\n  .px-sm-2 {\n    padding-right: 0.5rem !important;\n    padding-left: 0.5rem !important;\n  }\n  .px-sm-3 {\n    padding-right: 1rem !important;\n    padding-left: 1rem !important;\n  }\n  .px-sm-4 {\n    padding-right: 1.5rem !important;\n    padding-left: 1.5rem !important;\n  }\n  .px-sm-5 {\n    padding-right: 3rem !important;\n    padding-left: 3rem !important;\n  }\n  .py-sm-0 {\n    padding-top: 0 !important;\n    padding-bottom: 0 !important;\n  }\n  .py-sm-1 {\n    padding-top: 0.25rem !important;\n    padding-bottom: 0.25rem !important;\n  }\n  .py-sm-2 {\n    padding-top: 0.5rem !important;\n    padding-bottom: 0.5rem !important;\n  }\n  .py-sm-3 {\n    padding-top: 1rem !important;\n    padding-bottom: 1rem !important;\n  }\n  .py-sm-4 {\n    padding-top: 1.5rem !important;\n    padding-bottom: 1.5rem !important;\n  }\n  .py-sm-5 {\n    padding-top: 3rem !important;\n    padding-bottom: 3rem !important;\n  }\n  .pt-sm-0 {\n    padding-top: 0 !important;\n  }\n  .pt-sm-1 {\n    padding-top: 0.25rem !important;\n  }\n  .pt-sm-2 {\n    padding-top: 0.5rem !important;\n  }\n  .pt-sm-3 {\n    padding-top: 1rem !important;\n  }\n  .pt-sm-4 {\n    padding-top: 1.5rem !important;\n  }\n  .pt-sm-5 {\n    padding-top: 3rem !important;\n  }\n  .pe-sm-0 {\n    padding-right: 0 !important;\n  }\n  .pe-sm-1 {\n    padding-right: 0.25rem !important;\n  }\n  .pe-sm-2 {\n    padding-right: 0.5rem !important;\n  }\n  .pe-sm-3 {\n    padding-right: 1rem !important;\n  }\n  .pe-sm-4 {\n    padding-right: 1.5rem !important;\n  }\n  .pe-sm-5 {\n    padding-right: 3rem !important;\n  }\n  .pb-sm-0 {\n    padding-bottom: 0 !important;\n  }\n  .pb-sm-1 {\n    padding-bottom: 0.25rem !important;\n  }\n  .pb-sm-2 {\n    padding-bottom: 0.5rem !important;\n  }\n  .pb-sm-3 {\n    padding-bottom: 1rem !important;\n  }\n  .pb-sm-4 {\n    padding-bottom: 1.5rem !important;\n  }\n  .pb-sm-5 {\n    padding-bottom: 3rem !important;\n  }\n  .ps-sm-0 {\n    padding-left: 0 !important;\n  }\n  .ps-sm-1 {\n    padding-left: 0.25rem !important;\n  }\n  .ps-sm-2 {\n    padding-left: 0.5rem !important;\n  }\n  .ps-sm-3 {\n    padding-left: 1rem !important;\n  }\n  .ps-sm-4 {\n    padding-left: 1.5rem !important;\n  }\n  .ps-sm-5 {\n    padding-left: 3rem !important;\n  }\n  .gap-sm-0 {\n    gap: 0 !important;\n  }\n  .gap-sm-1 {\n    gap: 0.25rem !important;\n  }\n  .gap-sm-2 {\n    gap: 0.5rem !important;\n  }\n  .gap-sm-3 {\n    gap: 1rem !important;\n  }\n  .gap-sm-4 {\n    gap: 1.5rem !important;\n  }\n  .gap-sm-5 {\n    gap: 3rem !important;\n  }\n  .row-gap-sm-0 {\n    row-gap: 0 !important;\n  }\n  .row-gap-sm-1 {\n    row-gap: 0.25rem !important;\n  }\n  .row-gap-sm-2 {\n    row-gap: 0.5rem !important;\n  }\n  .row-gap-sm-3 {\n    row-gap: 1rem !important;\n  }\n  .row-gap-sm-4 {\n    row-gap: 1.5rem !important;\n  }\n  .row-gap-sm-5 {\n    row-gap: 3rem !important;\n  }\n  .column-gap-sm-0 {\n    column-gap: 0 !important;\n  }\n  .column-gap-sm-1 {\n    column-gap: 0.25rem !important;\n  }\n  .column-gap-sm-2 {\n    column-gap: 0.5rem !important;\n  }\n  .column-gap-sm-3 {\n    column-gap: 1rem !important;\n  }\n  .column-gap-sm-4 {\n    column-gap: 1.5rem !important;\n  }\n  .column-gap-sm-5 {\n    column-gap: 3rem !important;\n  }\n  .text-sm-start {\n    text-align: left !important;\n  }\n  .text-sm-end {\n    text-align: right !important;\n  }\n  .text-sm-center {\n    text-align: center !important;\n  }\n}\n@media (min-width: 720px) {\n  .float-md-start {\n    float: left !important;\n  }\n  .float-md-end {\n    float: right !important;\n  }\n  .float-md-none {\n    float: none !important;\n  }\n  .object-fit-md-contain {\n    object-fit: contain !important;\n  }\n  .object-fit-md-cover {\n    object-fit: cover !important;\n  }\n  .object-fit-md-fill {\n    object-fit: fill !important;\n  }\n  .object-fit-md-scale {\n    object-fit: scale-down !important;\n  }\n  .object-fit-md-none {\n    object-fit: none !important;\n  }\n  .d-md-inline {\n    display: inline !important;\n  }\n  .d-md-inline-block {\n    display: inline-block !important;\n  }\n  .d-md-block {\n    display: block !important;\n  }\n  .d-md-grid {\n    display: grid !important;\n  }\n  .d-md-inline-grid {\n    display: inline-grid !important;\n  }\n  .d-md-table {\n    display: table !important;\n  }\n  .d-md-table-row {\n    display: table-row !important;\n  }\n  .d-md-table-cell {\n    display: table-cell !important;\n  }\n  .d-md-flex {\n    display: flex !important;\n  }\n  .d-md-inline-flex {\n    display: inline-flex !important;\n  }\n  .d-md-none {\n    display: none !important;\n  }\n  .flex-md-fill {\n    flex: 1 1 auto !important;\n  }\n  .flex-md-row {\n    flex-direction: row !important;\n  }\n  .flex-md-column {\n    flex-direction: column !important;\n  }\n  .flex-md-row-reverse {\n    flex-direction: row-reverse !important;\n  }\n  .flex-md-column-reverse {\n    flex-direction: column-reverse !important;\n  }\n  .flex-md-grow-0 {\n    flex-grow: 0 !important;\n  }\n  .flex-md-grow-1 {\n    flex-grow: 1 !important;\n  }\n  .flex-md-shrink-0 {\n    flex-shrink: 0 !important;\n  }\n  .flex-md-shrink-1 {\n    flex-shrink: 1 !important;\n  }\n  .flex-md-wrap {\n    flex-wrap: wrap !important;\n  }\n  .flex-md-nowrap {\n    flex-wrap: nowrap !important;\n  }\n  .flex-md-wrap-reverse {\n    flex-wrap: wrap-reverse !important;\n  }\n  .justify-content-md-start {\n    justify-content: flex-start !important;\n  }\n  .justify-content-md-end {\n    justify-content: flex-end !important;\n  }\n  .justify-content-md-center {\n    justify-content: center !important;\n  }\n  .justify-content-md-between {\n    justify-content: space-between !important;\n  }\n  .justify-content-md-around {\n    justify-content: space-around !important;\n  }\n  .justify-content-md-evenly {\n    justify-content: space-evenly !important;\n  }\n  .align-items-md-start {\n    align-items: flex-start !important;\n  }\n  .align-items-md-end {\n    align-items: flex-end !important;\n  }\n  .align-items-md-center {\n    align-items: center !important;\n  }\n  .align-items-md-baseline {\n    align-items: baseline !important;\n  }\n  .align-items-md-stretch {\n    align-items: stretch !important;\n  }\n  .align-content-md-start {\n    align-content: flex-start !important;\n  }\n  .align-content-md-end {\n    align-content: flex-end !important;\n  }\n  .align-content-md-center {\n    align-content: center !important;\n  }\n  .align-content-md-between {\n    align-content: space-between !important;\n  }\n  .align-content-md-around {\n    align-content: space-around !important;\n  }\n  .align-content-md-stretch {\n    align-content: stretch !important;\n  }\n  .align-self-md-auto {\n    align-self: auto !important;\n  }\n  .align-self-md-start {\n    align-self: flex-start !important;\n  }\n  .align-self-md-end {\n    align-self: flex-end !important;\n  }\n  .align-self-md-center {\n    align-self: center !important;\n  }\n  .align-self-md-baseline {\n    align-self: baseline !important;\n  }\n  .align-self-md-stretch {\n    align-self: stretch !important;\n  }\n  .order-md-first {\n    order: -1 !important;\n  }\n  .order-md-0 {\n    order: 0 !important;\n  }\n  .order-md-1 {\n    order: 1 !important;\n  }\n  .order-md-2 {\n    order: 2 !important;\n  }\n  .order-md-3 {\n    order: 3 !important;\n  }\n  .order-md-4 {\n    order: 4 !important;\n  }\n  .order-md-5 {\n    order: 5 !important;\n  }\n  .order-md-last {\n    order: 6 !important;\n  }\n  .m-md-0 {\n    margin: 0 !important;\n  }\n  .m-md-1 {\n    margin: 0.25rem !important;\n  }\n  .m-md-2 {\n    margin: 0.5rem !important;\n  }\n  .m-md-3 {\n    margin: 1rem !important;\n  }\n  .m-md-4 {\n    margin: 1.5rem !important;\n  }\n  .m-md-5 {\n    margin: 3rem !important;\n  }\n  .m-md-auto {\n    margin: auto !important;\n  }\n  .mx-md-0 {\n    margin-right: 0 !important;\n    margin-left: 0 !important;\n  }\n  .mx-md-1 {\n    margin-right: 0.25rem !important;\n    margin-left: 0.25rem !important;\n  }\n  .mx-md-2 {\n    margin-right: 0.5rem !important;\n    margin-left: 0.5rem !important;\n  }\n  .mx-md-3 {\n    margin-right: 1rem !important;\n    margin-left: 1rem !important;\n  }\n  .mx-md-4 {\n    margin-right: 1.5rem !important;\n    margin-left: 1.5rem !important;\n  }\n  .mx-md-5 {\n    margin-right: 3rem !important;\n    margin-left: 3rem !important;\n  }\n  .mx-md-auto {\n    margin-right: auto !important;\n    margin-left: auto !important;\n  }\n  .my-md-0 {\n    margin-top: 0 !important;\n    margin-bottom: 0 !important;\n  }\n  .my-md-1 {\n    margin-top: 0.25rem !important;\n    margin-bottom: 0.25rem !important;\n  }\n  .my-md-2 {\n    margin-top: 0.5rem !important;\n    margin-bottom: 0.5rem !important;\n  }\n  .my-md-3 {\n    margin-top: 1rem !important;\n    margin-bottom: 1rem !important;\n  }\n  .my-md-4 {\n    margin-top: 1.5rem !important;\n    margin-bottom: 1.5rem !important;\n  }\n  .my-md-5 {\n    margin-top: 3rem !important;\n    margin-bottom: 3rem !important;\n  }\n  .my-md-auto {\n    margin-top: auto !important;\n    margin-bottom: auto !important;\n  }\n  .mt-md-0 {\n    margin-top: 0 !important;\n  }\n  .mt-md-1 {\n    margin-top: 0.25rem !important;\n  }\n  .mt-md-2 {\n    margin-top: 0.5rem !important;\n  }\n  .mt-md-3 {\n    margin-top: 1rem !important;\n  }\n  .mt-md-4 {\n    margin-top: 1.5rem !important;\n  }\n  .mt-md-5 {\n    margin-top: 3rem !important;\n  }\n  .mt-md-auto {\n    margin-top: auto !important;\n  }\n  .me-md-0 {\n    margin-right: 0 !important;\n  }\n  .me-md-1 {\n    margin-right: 0.25rem !important;\n  }\n  .me-md-2 {\n    margin-right: 0.5rem !important;\n  }\n  .me-md-3 {\n    margin-right: 1rem !important;\n  }\n  .me-md-4 {\n    margin-right: 1.5rem !important;\n  }\n  .me-md-5 {\n    margin-right: 3rem !important;\n  }\n  .me-md-auto {\n    margin-right: auto !important;\n  }\n  .mb-md-0 {\n    margin-bottom: 0 !important;\n  }\n  .mb-md-1 {\n    margin-bottom: 0.25rem !important;\n  }\n  .mb-md-2 {\n    margin-bottom: 0.5rem !important;\n  }\n  .mb-md-3 {\n    margin-bottom: 1rem !important;\n  }\n  .mb-md-4 {\n    margin-bottom: 1.5rem !important;\n  }\n  .mb-md-5 {\n    margin-bottom: 3rem !important;\n  }\n  .mb-md-auto {\n    margin-bottom: auto !important;\n  }\n  .ms-md-0 {\n    margin-left: 0 !important;\n  }\n  .ms-md-1 {\n    margin-left: 0.25rem !important;\n  }\n  .ms-md-2 {\n    margin-left: 0.5rem !important;\n  }\n  .ms-md-3 {\n    margin-left: 1rem !important;\n  }\n  .ms-md-4 {\n    margin-left: 1.5rem !important;\n  }\n  .ms-md-5 {\n    margin-left: 3rem !important;\n  }\n  .ms-md-auto {\n    margin-left: auto !important;\n  }\n  .p-md-0 {\n    padding: 0 !important;\n  }\n  .p-md-1 {\n    padding: 0.25rem !important;\n  }\n  .p-md-2 {\n    padding: 0.5rem !important;\n  }\n  .p-md-3 {\n    padding: 1rem !important;\n  }\n  .p-md-4 {\n    padding: 1.5rem !important;\n  }\n  .p-md-5 {\n    padding: 3rem !important;\n  }\n  .px-md-0 {\n    padding-right: 0 !important;\n    padding-left: 0 !important;\n  }\n  .px-md-1 {\n    padding-right: 0.25rem !important;\n    padding-left: 0.25rem !important;\n  }\n  .px-md-2 {\n    padding-right: 0.5rem !important;\n    padding-left: 0.5rem !important;\n  }\n  .px-md-3 {\n    padding-right: 1rem !important;\n    padding-left: 1rem !important;\n  }\n  .px-md-4 {\n    padding-right: 1.5rem !important;\n    padding-left: 1.5rem !important;\n  }\n  .px-md-5 {\n    padding-right: 3rem !important;\n    padding-left: 3rem !important;\n  }\n  .py-md-0 {\n    padding-top: 0 !important;\n    padding-bottom: 0 !important;\n  }\n  .py-md-1 {\n    padding-top: 0.25rem !important;\n    padding-bottom: 0.25rem !important;\n  }\n  .py-md-2 {\n    padding-top: 0.5rem !important;\n    padding-bottom: 0.5rem !important;\n  }\n  .py-md-3 {\n    padding-top: 1rem !important;\n    padding-bottom: 1rem !important;\n  }\n  .py-md-4 {\n    padding-top: 1.5rem !important;\n    padding-bottom: 1.5rem !important;\n  }\n  .py-md-5 {\n    padding-top: 3rem !important;\n    padding-bottom: 3rem !important;\n  }\n  .pt-md-0 {\n    padding-top: 0 !important;\n  }\n  .pt-md-1 {\n    padding-top: 0.25rem !important;\n  }\n  .pt-md-2 {\n    padding-top: 0.5rem !important;\n  }\n  .pt-md-3 {\n    padding-top: 1rem !important;\n  }\n  .pt-md-4 {\n    padding-top: 1.5rem !important;\n  }\n  .pt-md-5 {\n    padding-top: 3rem !important;\n  }\n  .pe-md-0 {\n    padding-right: 0 !important;\n  }\n  .pe-md-1 {\n    padding-right: 0.25rem !important;\n  }\n  .pe-md-2 {\n    padding-right: 0.5rem !important;\n  }\n  .pe-md-3 {\n    padding-right: 1rem !important;\n  }\n  .pe-md-4 {\n    padding-right: 1.5rem !important;\n  }\n  .pe-md-5 {\n    padding-right: 3rem !important;\n  }\n  .pb-md-0 {\n    padding-bottom: 0 !important;\n  }\n  .pb-md-1 {\n    padding-bottom: 0.25rem !important;\n  }\n  .pb-md-2 {\n    padding-bottom: 0.5rem !important;\n  }\n  .pb-md-3 {\n    padding-bottom: 1rem !important;\n  }\n  .pb-md-4 {\n    padding-bottom: 1.5rem !important;\n  }\n  .pb-md-5 {\n    padding-bottom: 3rem !important;\n  }\n  .ps-md-0 {\n    padding-left: 0 !important;\n  }\n  .ps-md-1 {\n    padding-left: 0.25rem !important;\n  }\n  .ps-md-2 {\n    padding-left: 0.5rem !important;\n  }\n  .ps-md-3 {\n    padding-left: 1rem !important;\n  }\n  .ps-md-4 {\n    padding-left: 1.5rem !important;\n  }\n  .ps-md-5 {\n    padding-left: 3rem !important;\n  }\n  .gap-md-0 {\n    gap: 0 !important;\n  }\n  .gap-md-1 {\n    gap: 0.25rem !important;\n  }\n  .gap-md-2 {\n    gap: 0.5rem !important;\n  }\n  .gap-md-3 {\n    gap: 1rem !important;\n  }\n  .gap-md-4 {\n    gap: 1.5rem !important;\n  }\n  .gap-md-5 {\n    gap: 3rem !important;\n  }\n  .row-gap-md-0 {\n    row-gap: 0 !important;\n  }\n  .row-gap-md-1 {\n    row-gap: 0.25rem !important;\n  }\n  .row-gap-md-2 {\n    row-gap: 0.5rem !important;\n  }\n  .row-gap-md-3 {\n    row-gap: 1rem !important;\n  }\n  .row-gap-md-4 {\n    row-gap: 1.5rem !important;\n  }\n  .row-gap-md-5 {\n    row-gap: 3rem !important;\n  }\n  .column-gap-md-0 {\n    column-gap: 0 !important;\n  }\n  .column-gap-md-1 {\n    column-gap: 0.25rem !important;\n  }\n  .column-gap-md-2 {\n    column-gap: 0.5rem !important;\n  }\n  .column-gap-md-3 {\n    column-gap: 1rem !important;\n  }\n  .column-gap-md-4 {\n    column-gap: 1.5rem !important;\n  }\n  .column-gap-md-5 {\n    column-gap: 3rem !important;\n  }\n  .text-md-start {\n    text-align: left !important;\n  }\n  .text-md-end {\n    text-align: right !important;\n  }\n  .text-md-center {\n    text-align: center !important;\n  }\n}\n@media (min-width: 960px) {\n  .float-lg-start {\n    float: left !important;\n  }\n  .float-lg-end {\n    float: right !important;\n  }\n  .float-lg-none {\n    float: none !important;\n  }\n  .object-fit-lg-contain {\n    object-fit: contain !important;\n  }\n  .object-fit-lg-cover {\n    object-fit: cover !important;\n  }\n  .object-fit-lg-fill {\n    object-fit: fill !important;\n  }\n  .object-fit-lg-scale {\n    object-fit: scale-down !important;\n  }\n  .object-fit-lg-none {\n    object-fit: none !important;\n  }\n  .d-lg-inline {\n    display: inline !important;\n  }\n  .d-lg-inline-block {\n    display: inline-block !important;\n  }\n  .d-lg-block {\n    display: block !important;\n  }\n  .d-lg-grid {\n    display: grid !important;\n  }\n  .d-lg-inline-grid {\n    display: inline-grid !important;\n  }\n  .d-lg-table {\n    display: table !important;\n  }\n  .d-lg-table-row {\n    display: table-row !important;\n  }\n  .d-lg-table-cell {\n    display: table-cell !important;\n  }\n  .d-lg-flex {\n    display: flex !important;\n  }\n  .d-lg-inline-flex {\n    display: inline-flex !important;\n  }\n  .d-lg-none {\n    display: none !important;\n  }\n  .flex-lg-fill {\n    flex: 1 1 auto !important;\n  }\n  .flex-lg-row {\n    flex-direction: row !important;\n  }\n  .flex-lg-column {\n    flex-direction: column !important;\n  }\n  .flex-lg-row-reverse {\n    flex-direction: row-reverse !important;\n  }\n  .flex-lg-column-reverse {\n    flex-direction: column-reverse !important;\n  }\n  .flex-lg-grow-0 {\n    flex-grow: 0 !important;\n  }\n  .flex-lg-grow-1 {\n    flex-grow: 1 !important;\n  }\n  .flex-lg-shrink-0 {\n    flex-shrink: 0 !important;\n  }\n  .flex-lg-shrink-1 {\n    flex-shrink: 1 !important;\n  }\n  .flex-lg-wrap {\n    flex-wrap: wrap !important;\n  }\n  .flex-lg-nowrap {\n    flex-wrap: nowrap !important;\n  }\n  .flex-lg-wrap-reverse {\n    flex-wrap: wrap-reverse !important;\n  }\n  .justify-content-lg-start {\n    justify-content: flex-start !important;\n  }\n  .justify-content-lg-end {\n    justify-content: flex-end !important;\n  }\n  .justify-content-lg-center {\n    justify-content: center !important;\n  }\n  .justify-content-lg-between {\n    justify-content: space-between !important;\n  }\n  .justify-content-lg-around {\n    justify-content: space-around !important;\n  }\n  .justify-content-lg-evenly {\n    justify-content: space-evenly !important;\n  }\n  .align-items-lg-start {\n    align-items: flex-start !important;\n  }\n  .align-items-lg-end {\n    align-items: flex-end !important;\n  }\n  .align-items-lg-center {\n    align-items: center !important;\n  }\n  .align-items-lg-baseline {\n    align-items: baseline !important;\n  }\n  .align-items-lg-stretch {\n    align-items: stretch !important;\n  }\n  .align-content-lg-start {\n    align-content: flex-start !important;\n  }\n  .align-content-lg-end {\n    align-content: flex-end !important;\n  }\n  .align-content-lg-center {\n    align-content: center !important;\n  }\n  .align-content-lg-between {\n    align-content: space-between !important;\n  }\n  .align-content-lg-around {\n    align-content: space-around !important;\n  }\n  .align-content-lg-stretch {\n    align-content: stretch !important;\n  }\n  .align-self-lg-auto {\n    align-self: auto !important;\n  }\n  .align-self-lg-start {\n    align-self: flex-start !important;\n  }\n  .align-self-lg-end {\n    align-self: flex-end !important;\n  }\n  .align-self-lg-center {\n    align-self: center !important;\n  }\n  .align-self-lg-baseline {\n    align-self: baseline !important;\n  }\n  .align-self-lg-stretch {\n    align-self: stretch !important;\n  }\n  .order-lg-first {\n    order: -1 !important;\n  }\n  .order-lg-0 {\n    order: 0 !important;\n  }\n  .order-lg-1 {\n    order: 1 !important;\n  }\n  .order-lg-2 {\n    order: 2 !important;\n  }\n  .order-lg-3 {\n    order: 3 !important;\n  }\n  .order-lg-4 {\n    order: 4 !important;\n  }\n  .order-lg-5 {\n    order: 5 !important;\n  }\n  .order-lg-last {\n    order: 6 !important;\n  }\n  .m-lg-0 {\n    margin: 0 !important;\n  }\n  .m-lg-1 {\n    margin: 0.25rem !important;\n  }\n  .m-lg-2 {\n    margin: 0.5rem !important;\n  }\n  .m-lg-3 {\n    margin: 1rem !important;\n  }\n  .m-lg-4 {\n    margin: 1.5rem !important;\n  }\n  .m-lg-5 {\n    margin: 3rem !important;\n  }\n  .m-lg-auto {\n    margin: auto !important;\n  }\n  .mx-lg-0 {\n    margin-right: 0 !important;\n    margin-left: 0 !important;\n  }\n  .mx-lg-1 {\n    margin-right: 0.25rem !important;\n    margin-left: 0.25rem !important;\n  }\n  .mx-lg-2 {\n    margin-right: 0.5rem !important;\n    margin-left: 0.5rem !important;\n  }\n  .mx-lg-3 {\n    margin-right: 1rem !important;\n    margin-left: 1rem !important;\n  }\n  .mx-lg-4 {\n    margin-right: 1.5rem !important;\n    margin-left: 1.5rem !important;\n  }\n  .mx-lg-5 {\n    margin-right: 3rem !important;\n    margin-left: 3rem !important;\n  }\n  .mx-lg-auto {\n    margin-right: auto !important;\n    margin-left: auto !important;\n  }\n  .my-lg-0 {\n    margin-top: 0 !important;\n    margin-bottom: 0 !important;\n  }\n  .my-lg-1 {\n    margin-top: 0.25rem !important;\n    margin-bottom: 0.25rem !important;\n  }\n  .my-lg-2 {\n    margin-top: 0.5rem !important;\n    margin-bottom: 0.5rem !important;\n  }\n  .my-lg-3 {\n    margin-top: 1rem !important;\n    margin-bottom: 1rem !important;\n  }\n  .my-lg-4 {\n    margin-top: 1.5rem !important;\n    margin-bottom: 1.5rem !important;\n  }\n  .my-lg-5 {\n    margin-top: 3rem !important;\n    margin-bottom: 3rem !important;\n  }\n  .my-lg-auto {\n    margin-top: auto !important;\n    margin-bottom: auto !important;\n  }\n  .mt-lg-0 {\n    margin-top: 0 !important;\n  }\n  .mt-lg-1 {\n    margin-top: 0.25rem !important;\n  }\n  .mt-lg-2 {\n    margin-top: 0.5rem !important;\n  }\n  .mt-lg-3 {\n    margin-top: 1rem !important;\n  }\n  .mt-lg-4 {\n    margin-top: 1.5rem !important;\n  }\n  .mt-lg-5 {\n    margin-top: 3rem !important;\n  }\n  .mt-lg-auto {\n    margin-top: auto !important;\n  }\n  .me-lg-0 {\n    margin-right: 0 !important;\n  }\n  .me-lg-1 {\n    margin-right: 0.25rem !important;\n  }\n  .me-lg-2 {\n    margin-right: 0.5rem !important;\n  }\n  .me-lg-3 {\n    margin-right: 1rem !important;\n  }\n  .me-lg-4 {\n    margin-right: 1.5rem !important;\n  }\n  .me-lg-5 {\n    margin-right: 3rem !important;\n  }\n  .me-lg-auto {\n    margin-right: auto !important;\n  }\n  .mb-lg-0 {\n    margin-bottom: 0 !important;\n  }\n  .mb-lg-1 {\n    margin-bottom: 0.25rem !important;\n  }\n  .mb-lg-2 {\n    margin-bottom: 0.5rem !important;\n  }\n  .mb-lg-3 {\n    margin-bottom: 1rem !important;\n  }\n  .mb-lg-4 {\n    margin-bottom: 1.5rem !important;\n  }\n  .mb-lg-5 {\n    margin-bottom: 3rem !important;\n  }\n  .mb-lg-auto {\n    margin-bottom: auto !important;\n  }\n  .ms-lg-0 {\n    margin-left: 0 !important;\n  }\n  .ms-lg-1 {\n    margin-left: 0.25rem !important;\n  }\n  .ms-lg-2 {\n    margin-left: 0.5rem !important;\n  }\n  .ms-lg-3 {\n    margin-left: 1rem !important;\n  }\n  .ms-lg-4 {\n    margin-left: 1.5rem !important;\n  }\n  .ms-lg-5 {\n    margin-left: 3rem !important;\n  }\n  .ms-lg-auto {\n    margin-left: auto !important;\n  }\n  .p-lg-0 {\n    padding: 0 !important;\n  }\n  .p-lg-1 {\n    padding: 0.25rem !important;\n  }\n  .p-lg-2 {\n    padding: 0.5rem !important;\n  }\n  .p-lg-3 {\n    padding: 1rem !important;\n  }\n  .p-lg-4 {\n    padding: 1.5rem !important;\n  }\n  .p-lg-5 {\n    padding: 3rem !important;\n  }\n  .px-lg-0 {\n    padding-right: 0 !important;\n    padding-left: 0 !important;\n  }\n  .px-lg-1 {\n    padding-right: 0.25rem !important;\n    padding-left: 0.25rem !important;\n  }\n  .px-lg-2 {\n    padding-right: 0.5rem !important;\n    padding-left: 0.5rem !important;\n  }\n  .px-lg-3 {\n    padding-right: 1rem !important;\n    padding-left: 1rem !important;\n  }\n  .px-lg-4 {\n    padding-right: 1.5rem !important;\n    padding-left: 1.5rem !important;\n  }\n  .px-lg-5 {\n    padding-right: 3rem !important;\n    padding-left: 3rem !important;\n  }\n  .py-lg-0 {\n    padding-top: 0 !important;\n    padding-bottom: 0 !important;\n  }\n  .py-lg-1 {\n    padding-top: 0.25rem !important;\n    padding-bottom: 0.25rem !important;\n  }\n  .py-lg-2 {\n    padding-top: 0.5rem !important;\n    padding-bottom: 0.5rem !important;\n  }\n  .py-lg-3 {\n    padding-top: 1rem !important;\n    padding-bottom: 1rem !important;\n  }\n  .py-lg-4 {\n    padding-top: 1.5rem !important;\n    padding-bottom: 1.5rem !important;\n  }\n  .py-lg-5 {\n    padding-top: 3rem !important;\n    padding-bottom: 3rem !important;\n  }\n  .pt-lg-0 {\n    padding-top: 0 !important;\n  }\n  .pt-lg-1 {\n    padding-top: 0.25rem !important;\n  }\n  .pt-lg-2 {\n    padding-top: 0.5rem !important;\n  }\n  .pt-lg-3 {\n    padding-top: 1rem !important;\n  }\n  .pt-lg-4 {\n    padding-top: 1.5rem !important;\n  }\n  .pt-lg-5 {\n    padding-top: 3rem !important;\n  }\n  .pe-lg-0 {\n    padding-right: 0 !important;\n  }\n  .pe-lg-1 {\n    padding-right: 0.25rem !important;\n  }\n  .pe-lg-2 {\n    padding-right: 0.5rem !important;\n  }\n  .pe-lg-3 {\n    padding-right: 1rem !important;\n  }\n  .pe-lg-4 {\n    padding-right: 1.5rem !important;\n  }\n  .pe-lg-5 {\n    padding-right: 3rem !important;\n  }\n  .pb-lg-0 {\n    padding-bottom: 0 !important;\n  }\n  .pb-lg-1 {\n    padding-bottom: 0.25rem !important;\n  }\n  .pb-lg-2 {\n    padding-bottom: 0.5rem !important;\n  }\n  .pb-lg-3 {\n    padding-bottom: 1rem !important;\n  }\n  .pb-lg-4 {\n    padding-bottom: 1.5rem !important;\n  }\n  .pb-lg-5 {\n    padding-bottom: 3rem !important;\n  }\n  .ps-lg-0 {\n    padding-left: 0 !important;\n  }\n  .ps-lg-1 {\n    padding-left: 0.25rem !important;\n  }\n  .ps-lg-2 {\n    padding-left: 0.5rem !important;\n  }\n  .ps-lg-3 {\n    padding-left: 1rem !important;\n  }\n  .ps-lg-4 {\n    padding-left: 1.5rem !important;\n  }\n  .ps-lg-5 {\n    padding-left: 3rem !important;\n  }\n  .gap-lg-0 {\n    gap: 0 !important;\n  }\n  .gap-lg-1 {\n    gap: 0.25rem !important;\n  }\n  .gap-lg-2 {\n    gap: 0.5rem !important;\n  }\n  .gap-lg-3 {\n    gap: 1rem !important;\n  }\n  .gap-lg-4 {\n    gap: 1.5rem !important;\n  }\n  .gap-lg-5 {\n    gap: 3rem !important;\n  }\n  .row-gap-lg-0 {\n    row-gap: 0 !important;\n  }\n  .row-gap-lg-1 {\n    row-gap: 0.25rem !important;\n  }\n  .row-gap-lg-2 {\n    row-gap: 0.5rem !important;\n  }\n  .row-gap-lg-3 {\n    row-gap: 1rem !important;\n  }\n  .row-gap-lg-4 {\n    row-gap: 1.5rem !important;\n  }\n  .row-gap-lg-5 {\n    row-gap: 3rem !important;\n  }\n  .column-gap-lg-0 {\n    column-gap: 0 !important;\n  }\n  .column-gap-lg-1 {\n    column-gap: 0.25rem !important;\n  }\n  .column-gap-lg-2 {\n    column-gap: 0.5rem !important;\n  }\n  .column-gap-lg-3 {\n    column-gap: 1rem !important;\n  }\n  .column-gap-lg-4 {\n    column-gap: 1.5rem !important;\n  }\n  .column-gap-lg-5 {\n    column-gap: 3rem !important;\n  }\n  .text-lg-start {\n    text-align: left !important;\n  }\n  .text-lg-end {\n    text-align: right !important;\n  }\n  .text-lg-center {\n    text-align: center !important;\n  }\n}\n@media (min-width: 1200px) {\n  .float-xl-start {\n    float: left !important;\n  }\n  .float-xl-end {\n    float: right !important;\n  }\n  .float-xl-none {\n    float: none !important;\n  }\n  .object-fit-xl-contain {\n    object-fit: contain !important;\n  }\n  .object-fit-xl-cover {\n    object-fit: cover !important;\n  }\n  .object-fit-xl-fill {\n    object-fit: fill !important;\n  }\n  .object-fit-xl-scale {\n    object-fit: scale-down !important;\n  }\n  .object-fit-xl-none {\n    object-fit: none !important;\n  }\n  .d-xl-inline {\n    display: inline !important;\n  }\n  .d-xl-inline-block {\n    display: inline-block !important;\n  }\n  .d-xl-block {\n    display: block !important;\n  }\n  .d-xl-grid {\n    display: grid !important;\n  }\n  .d-xl-inline-grid {\n    display: inline-grid !important;\n  }\n  .d-xl-table {\n    display: table !important;\n  }\n  .d-xl-table-row {\n    display: table-row !important;\n  }\n  .d-xl-table-cell {\n    display: table-cell !important;\n  }\n  .d-xl-flex {\n    display: flex !important;\n  }\n  .d-xl-inline-flex {\n    display: inline-flex !important;\n  }\n  .d-xl-none {\n    display: none !important;\n  }\n  .flex-xl-fill {\n    flex: 1 1 auto !important;\n  }\n  .flex-xl-row {\n    flex-direction: row !important;\n  }\n  .flex-xl-column {\n    flex-direction: column !important;\n  }\n  .flex-xl-row-reverse {\n    flex-direction: row-reverse !important;\n  }\n  .flex-xl-column-reverse {\n    flex-direction: column-reverse !important;\n  }\n  .flex-xl-grow-0 {\n    flex-grow: 0 !important;\n  }\n  .flex-xl-grow-1 {\n    flex-grow: 1 !important;\n  }\n  .flex-xl-shrink-0 {\n    flex-shrink: 0 !important;\n  }\n  .flex-xl-shrink-1 {\n    flex-shrink: 1 !important;\n  }\n  .flex-xl-wrap {\n    flex-wrap: wrap !important;\n  }\n  .flex-xl-nowrap {\n    flex-wrap: nowrap !important;\n  }\n  .flex-xl-wrap-reverse {\n    flex-wrap: wrap-reverse !important;\n  }\n  .justify-content-xl-start {\n    justify-content: flex-start !important;\n  }\n  .justify-content-xl-end {\n    justify-content: flex-end !important;\n  }\n  .justify-content-xl-center {\n    justify-content: center !important;\n  }\n  .justify-content-xl-between {\n    justify-content: space-between !important;\n  }\n  .justify-content-xl-around {\n    justify-content: space-around !important;\n  }\n  .justify-content-xl-evenly {\n    justify-content: space-evenly !important;\n  }\n  .align-items-xl-start {\n    align-items: flex-start !important;\n  }\n  .align-items-xl-end {\n    align-items: flex-end !important;\n  }\n  .align-items-xl-center {\n    align-items: center !important;\n  }\n  .align-items-xl-baseline {\n    align-items: baseline !important;\n  }\n  .align-items-xl-stretch {\n    align-items: stretch !important;\n  }\n  .align-content-xl-start {\n    align-content: flex-start !important;\n  }\n  .align-content-xl-end {\n    align-content: flex-end !important;\n  }\n  .align-content-xl-center {\n    align-content: center !important;\n  }\n  .align-content-xl-between {\n    align-content: space-between !important;\n  }\n  .align-content-xl-around {\n    align-content: space-around !important;\n  }\n  .align-content-xl-stretch {\n    align-content: stretch !important;\n  }\n  .align-self-xl-auto {\n    align-self: auto !important;\n  }\n  .align-self-xl-start {\n    align-self: flex-start !important;\n  }\n  .align-self-xl-end {\n    align-self: flex-end !important;\n  }\n  .align-self-xl-center {\n    align-self: center !important;\n  }\n  .align-self-xl-baseline {\n    align-self: baseline !important;\n  }\n  .align-self-xl-stretch {\n    align-self: stretch !important;\n  }\n  .order-xl-first {\n    order: -1 !important;\n  }\n  .order-xl-0 {\n    order: 0 !important;\n  }\n  .order-xl-1 {\n    order: 1 !important;\n  }\n  .order-xl-2 {\n    order: 2 !important;\n  }\n  .order-xl-3 {\n    order: 3 !important;\n  }\n  .order-xl-4 {\n    order: 4 !important;\n  }\n  .order-xl-5 {\n    order: 5 !important;\n  }\n  .order-xl-last {\n    order: 6 !important;\n  }\n  .m-xl-0 {\n    margin: 0 !important;\n  }\n  .m-xl-1 {\n    margin: 0.25rem !important;\n  }\n  .m-xl-2 {\n    margin: 0.5rem !important;\n  }\n  .m-xl-3 {\n    margin: 1rem !important;\n  }\n  .m-xl-4 {\n    margin: 1.5rem !important;\n  }\n  .m-xl-5 {\n    margin: 3rem !important;\n  }\n  .m-xl-auto {\n    margin: auto !important;\n  }\n  .mx-xl-0 {\n    margin-right: 0 !important;\n    margin-left: 0 !important;\n  }\n  .mx-xl-1 {\n    margin-right: 0.25rem !important;\n    margin-left: 0.25rem !important;\n  }\n  .mx-xl-2 {\n    margin-right: 0.5rem !important;\n    margin-left: 0.5rem !important;\n  }\n  .mx-xl-3 {\n    margin-right: 1rem !important;\n    margin-left: 1rem !important;\n  }\n  .mx-xl-4 {\n    margin-right: 1.5rem !important;\n    margin-left: 1.5rem !important;\n  }\n  .mx-xl-5 {\n    margin-right: 3rem !important;\n    margin-left: 3rem !important;\n  }\n  .mx-xl-auto {\n    margin-right: auto !important;\n    margin-left: auto !important;\n  }\n  .my-xl-0 {\n    margin-top: 0 !important;\n    margin-bottom: 0 !important;\n  }\n  .my-xl-1 {\n    margin-top: 0.25rem !important;\n    margin-bottom: 0.25rem !important;\n  }\n  .my-xl-2 {\n    margin-top: 0.5rem !important;\n    margin-bottom: 0.5rem !important;\n  }\n  .my-xl-3 {\n    margin-top: 1rem !important;\n    margin-bottom: 1rem !important;\n  }\n  .my-xl-4 {\n    margin-top: 1.5rem !important;\n    margin-bottom: 1.5rem !important;\n  }\n  .my-xl-5 {\n    margin-top: 3rem !important;\n    margin-bottom: 3rem !important;\n  }\n  .my-xl-auto {\n    margin-top: auto !important;\n    margin-bottom: auto !important;\n  }\n  .mt-xl-0 {\n    margin-top: 0 !important;\n  }\n  .mt-xl-1 {\n    margin-top: 0.25rem !important;\n  }\n  .mt-xl-2 {\n    margin-top: 0.5rem !important;\n  }\n  .mt-xl-3 {\n    margin-top: 1rem !important;\n  }\n  .mt-xl-4 {\n    margin-top: 1.5rem !important;\n  }\n  .mt-xl-5 {\n    margin-top: 3rem !important;\n  }\n  .mt-xl-auto {\n    margin-top: auto !important;\n  }\n  .me-xl-0 {\n    margin-right: 0 !important;\n  }\n  .me-xl-1 {\n    margin-right: 0.25rem !important;\n  }\n  .me-xl-2 {\n    margin-right: 0.5rem !important;\n  }\n  .me-xl-3 {\n    margin-right: 1rem !important;\n  }\n  .me-xl-4 {\n    margin-right: 1.5rem !important;\n  }\n  .me-xl-5 {\n    margin-right: 3rem !important;\n  }\n  .me-xl-auto {\n    margin-right: auto !important;\n  }\n  .mb-xl-0 {\n    margin-bottom: 0 !important;\n  }\n  .mb-xl-1 {\n    margin-bottom: 0.25rem !important;\n  }\n  .mb-xl-2 {\n    margin-bottom: 0.5rem !important;\n  }\n  .mb-xl-3 {\n    margin-bottom: 1rem !important;\n  }\n  .mb-xl-4 {\n    margin-bottom: 1.5rem !important;\n  }\n  .mb-xl-5 {\n    margin-bottom: 3rem !important;\n  }\n  .mb-xl-auto {\n    margin-bottom: auto !important;\n  }\n  .ms-xl-0 {\n    margin-left: 0 !important;\n  }\n  .ms-xl-1 {\n    margin-left: 0.25rem !important;\n  }\n  .ms-xl-2 {\n    margin-left: 0.5rem !important;\n  }\n  .ms-xl-3 {\n    margin-left: 1rem !important;\n  }\n  .ms-xl-4 {\n    margin-left: 1.5rem !important;\n  }\n  .ms-xl-5 {\n    margin-left: 3rem !important;\n  }\n  .ms-xl-auto {\n    margin-left: auto !important;\n  }\n  .p-xl-0 {\n    padding: 0 !important;\n  }\n  .p-xl-1 {\n    padding: 0.25rem !important;\n  }\n  .p-xl-2 {\n    padding: 0.5rem !important;\n  }\n  .p-xl-3 {\n    padding: 1rem !important;\n  }\n  .p-xl-4 {\n    padding: 1.5rem !important;\n  }\n  .p-xl-5 {\n    padding: 3rem !important;\n  }\n  .px-xl-0 {\n    padding-right: 0 !important;\n    padding-left: 0 !important;\n  }\n  .px-xl-1 {\n    padding-right: 0.25rem !important;\n    padding-left: 0.25rem !important;\n  }\n  .px-xl-2 {\n    padding-right: 0.5rem !important;\n    padding-left: 0.5rem !important;\n  }\n  .px-xl-3 {\n    padding-right: 1rem !important;\n    padding-left: 1rem !important;\n  }\n  .px-xl-4 {\n    padding-right: 1.5rem !important;\n    padding-left: 1.5rem !important;\n  }\n  .px-xl-5 {\n    padding-right: 3rem !important;\n    padding-left: 3rem !important;\n  }\n  .py-xl-0 {\n    padding-top: 0 !important;\n    padding-bottom: 0 !important;\n  }\n  .py-xl-1 {\n    padding-top: 0.25rem !important;\n    padding-bottom: 0.25rem !important;\n  }\n  .py-xl-2 {\n    padding-top: 0.5rem !important;\n    padding-bottom: 0.5rem !important;\n  }\n  .py-xl-3 {\n    padding-top: 1rem !important;\n    padding-bottom: 1rem !important;\n  }\n  .py-xl-4 {\n    padding-top: 1.5rem !important;\n    padding-bottom: 1.5rem !important;\n  }\n  .py-xl-5 {\n    padding-top: 3rem !important;\n    padding-bottom: 3rem !important;\n  }\n  .pt-xl-0 {\n    padding-top: 0 !important;\n  }\n  .pt-xl-1 {\n    padding-top: 0.25rem !important;\n  }\n  .pt-xl-2 {\n    padding-top: 0.5rem !important;\n  }\n  .pt-xl-3 {\n    padding-top: 1rem !important;\n  }\n  .pt-xl-4 {\n    padding-top: 1.5rem !important;\n  }\n  .pt-xl-5 {\n    padding-top: 3rem !important;\n  }\n  .pe-xl-0 {\n    padding-right: 0 !important;\n  }\n  .pe-xl-1 {\n    padding-right: 0.25rem !important;\n  }\n  .pe-xl-2 {\n    padding-right: 0.5rem !important;\n  }\n  .pe-xl-3 {\n    padding-right: 1rem !important;\n  }\n  .pe-xl-4 {\n    padding-right: 1.5rem !important;\n  }\n  .pe-xl-5 {\n    padding-right: 3rem !important;\n  }\n  .pb-xl-0 {\n    padding-bottom: 0 !important;\n  }\n  .pb-xl-1 {\n    padding-bottom: 0.25rem !important;\n  }\n  .pb-xl-2 {\n    padding-bottom: 0.5rem !important;\n  }\n  .pb-xl-3 {\n    padding-bottom: 1rem !important;\n  }\n  .pb-xl-4 {\n    padding-bottom: 1.5rem !important;\n  }\n  .pb-xl-5 {\n    padding-bottom: 3rem !important;\n  }\n  .ps-xl-0 {\n    padding-left: 0 !important;\n  }\n  .ps-xl-1 {\n    padding-left: 0.25rem !important;\n  }\n  .ps-xl-2 {\n    padding-left: 0.5rem !important;\n  }\n  .ps-xl-3 {\n    padding-left: 1rem !important;\n  }\n  .ps-xl-4 {\n    padding-left: 1.5rem !important;\n  }\n  .ps-xl-5 {\n    padding-left: 3rem !important;\n  }\n  .gap-xl-0 {\n    gap: 0 !important;\n  }\n  .gap-xl-1 {\n    gap: 0.25rem !important;\n  }\n  .gap-xl-2 {\n    gap: 0.5rem !important;\n  }\n  .gap-xl-3 {\n    gap: 1rem !important;\n  }\n  .gap-xl-4 {\n    gap: 1.5rem !important;\n  }\n  .gap-xl-5 {\n    gap: 3rem !important;\n  }\n  .row-gap-xl-0 {\n    row-gap: 0 !important;\n  }\n  .row-gap-xl-1 {\n    row-gap: 0.25rem !important;\n  }\n  .row-gap-xl-2 {\n    row-gap: 0.5rem !important;\n  }\n  .row-gap-xl-3 {\n    row-gap: 1rem !important;\n  }\n  .row-gap-xl-4 {\n    row-gap: 1.5rem !important;\n  }\n  .row-gap-xl-5 {\n    row-gap: 3rem !important;\n  }\n  .column-gap-xl-0 {\n    column-gap: 0 !important;\n  }\n  .column-gap-xl-1 {\n    column-gap: 0.25rem !important;\n  }\n  .column-gap-xl-2 {\n    column-gap: 0.5rem !important;\n  }\n  .column-gap-xl-3 {\n    column-gap: 1rem !important;\n  }\n  .column-gap-xl-4 {\n    column-gap: 1.5rem !important;\n  }\n  .column-gap-xl-5 {\n    column-gap: 3rem !important;\n  }\n  .text-xl-start {\n    text-align: left !important;\n  }\n  .text-xl-end {\n    text-align: right !important;\n  }\n  .text-xl-center {\n    text-align: center !important;\n  }\n}\n@media (min-width: 1200px) {\n  .fs-1 {\n    font-size: 2.5rem !important;\n  }\n  .fs-2 {\n    font-size: 2rem !important;\n  }\n  .fs-3 {\n    font-size: 1.75rem !important;\n  }\n  .fs-4 {\n    font-size: 1.5rem !important;\n  }\n}\n@media print {\n  .d-print-inline {\n    display: inline !important;\n  }\n  .d-print-inline-block {\n    display: inline-block !important;\n  }\n  .d-print-block {\n    display: block !important;\n  }\n  .d-print-grid {\n    display: grid !important;\n  }\n  .d-print-inline-grid {\n    display: inline-grid !important;\n  }\n  .d-print-table {\n    display: table !important;\n  }\n  .d-print-table-row {\n    display: table-row !important;\n  }\n  .d-print-table-cell {\n    display: table-cell !important;\n  }\n  .d-print-flex {\n    display: flex !important;\n  }\n  .d-print-inline-flex {\n    display: inline-flex !important;\n  }\n  .d-print-none {\n    display: none !important;\n  }\n}\n/*!\n * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com\n * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)\n * Copyright 2024 Fonticons, Inc.\n */\n.fa {\n  font-family: var(--fa-style-family, \"Font Awesome 6 Free\");\n  font-weight: var(--fa-style, 900);\n}\n\n.fa,\n.fa-classic,\n.fa-sharp,\n.fas,\n.fa-solid,\n.far,\n.fa-regular,\n.fab,\n.fa-brands {\n  -moz-osx-font-smoothing: grayscale;\n  -webkit-font-smoothing: antialiased;\n  display: var(--fa-display, inline-block);\n  font-style: normal;\n  font-variant: normal;\n  line-height: 1;\n  text-rendering: auto;\n}\n\n.fas,\n.fa-classic,\n.fa-solid,\n.far,\n.fa-regular {\n  font-family: \"Font Awesome 6 Free\";\n}\n\n.fab,\n.fa-brands {\n  font-family: \"Font Awesome 6 Brands\";\n}\n\n.fa-1x {\n  font-size: 1em;\n}\n\n.fa-2x {\n  font-size: 2em;\n}\n\n.fa-3x {\n  font-size: 3em;\n}\n\n.fa-4x {\n  font-size: 4em;\n}\n\n.fa-5x {\n  font-size: 5em;\n}\n\n.fa-6x {\n  font-size: 6em;\n}\n\n.fa-7x {\n  font-size: 7em;\n}\n\n.fa-8x {\n  font-size: 8em;\n}\n\n.fa-9x {\n  font-size: 9em;\n}\n\n.fa-10x {\n  font-size: 10em;\n}\n\n.fa-2xs {\n  font-size: 0.625em;\n  line-height: 0.1em;\n  vertical-align: 0.225em;\n}\n\n.fa-xs {\n  font-size: 0.75em;\n  line-height: 0.0833333337em;\n  vertical-align: 0.125em;\n}\n\n.fa-sm {\n  font-size: 0.875em;\n  line-height: 0.0714285718em;\n  vertical-align: 0.0535714295em;\n}\n\n.fa-lg {\n  font-size: 1.25em;\n  line-height: 0.05em;\n  vertical-align: -0.075em;\n}\n\n.fa-xl {\n  font-size: 1.5em;\n  line-height: 0.0416666682em;\n  vertical-align: -0.125em;\n}\n\n.fa-2xl {\n  font-size: 2em;\n  line-height: 0.03125em;\n  vertical-align: -0.1875em;\n}\n\n.fa-fw {\n  text-align: center;\n  width: 1.25em;\n}\n\n.fa-ul {\n  list-style-type: none;\n  margin-left: var(--fa-li-margin, 2.5em);\n  padding-left: 0;\n}\n.fa-ul > li {\n  position: relative;\n}\n\n.fa-li {\n  left: calc(var(--fa-li-width, 2em) * -1);\n  position: absolute;\n  text-align: center;\n  width: var(--fa-li-width, 2em);\n  line-height: inherit;\n}\n\n.fa-border {\n  border-color: var(--fa-border-color, #eee);\n  border-radius: var(--fa-border-radius, 0.1em);\n  border-style: var(--fa-border-style, solid);\n  border-width: var(--fa-border-width, 0.08em);\n  padding: var(--fa-border-padding, 0.2em 0.25em 0.15em);\n}\n\n.fa-pull-left {\n  float: left;\n  margin-right: var(--fa-pull-margin, 0.3em);\n}\n\n.fa-pull-right {\n  float: right;\n  margin-left: var(--fa-pull-margin, 0.3em);\n}\n\n.fa-beat {\n  animation-name: fa-beat;\n  animation-delay: var(--fa-animation-delay, 0s);\n  animation-direction: var(--fa-animation-direction, normal);\n  animation-duration: var(--fa-animation-duration, 1s);\n  animation-iteration-count: var(--fa-animation-iteration-count, infinite);\n  animation-timing-function: var(--fa-animation-timing, ease-in-out);\n}\n\n.fa-bounce {\n  animation-name: fa-bounce;\n  animation-delay: var(--fa-animation-delay, 0s);\n  animation-direction: var(--fa-animation-direction, normal);\n  animation-duration: var(--fa-animation-duration, 1s);\n  animation-iteration-count: var(--fa-animation-iteration-count, infinite);\n  animation-timing-function: var(--fa-animation-timing, cubic-bezier(0.28, 0.84, 0.42, 1));\n}\n\n.fa-fade {\n  animation-name: fa-fade;\n  animation-delay: var(--fa-animation-delay, 0s);\n  animation-direction: var(--fa-animation-direction, normal);\n  animation-duration: var(--fa-animation-duration, 1s);\n  animation-iteration-count: var(--fa-animation-iteration-count, infinite);\n  animation-timing-function: var(--fa-animation-timing, cubic-bezier(0.4, 0, 0.6, 1));\n}\n\n.fa-beat-fade {\n  animation-name: fa-beat-fade;\n  animation-delay: var(--fa-animation-delay, 0s);\n  animation-direction: var(--fa-animation-direction, normal);\n  animation-duration: var(--fa-animation-duration, 1s);\n  animation-iteration-count: var(--fa-animation-iteration-count, infinite);\n  animation-timing-function: var(--fa-animation-timing, cubic-bezier(0.4, 0, 0.6, 1));\n}\n\n.fa-flip {\n  animation-name: fa-flip;\n  animation-delay: var(--fa-animation-delay, 0s);\n  animation-direction: var(--fa-animation-direction, normal);\n  animation-duration: var(--fa-animation-duration, 1s);\n  animation-iteration-count: var(--fa-animation-iteration-count, infinite);\n  animation-timing-function: var(--fa-animation-timing, ease-in-out);\n}\n\n.fa-shake {\n  animation-name: fa-shake;\n  animation-delay: var(--fa-animation-delay, 0s);\n  animation-direction: var(--fa-animation-direction, normal);\n  animation-duration: var(--fa-animation-duration, 1s);\n  animation-iteration-count: var(--fa-animation-iteration-count, infinite);\n  animation-timing-function: var(--fa-animation-timing, linear);\n}\n\n.fa-spin {\n  animation-name: fa-spin;\n  animation-delay: var(--fa-animation-delay, 0s);\n  animation-direction: var(--fa-animation-direction, normal);\n  animation-duration: var(--fa-animation-duration, 2s);\n  animation-iteration-count: var(--fa-animation-iteration-count, infinite);\n  animation-timing-function: var(--fa-animation-timing, linear);\n}\n\n.fa-spin-reverse {\n  --fa-animation-direction: reverse;\n}\n\n.fa-pulse,\n.fa-spin-pulse {\n  animation-name: fa-spin;\n  animation-direction: var(--fa-animation-direction, normal);\n  animation-duration: var(--fa-animation-duration, 1s);\n  animation-iteration-count: var(--fa-animation-iteration-count, infinite);\n  animation-timing-function: var(--fa-animation-timing, steps(8));\n}\n\n@media (prefers-reduced-motion: reduce) {\n  .fa-beat,\n  .fa-bounce,\n  .fa-fade,\n  .fa-beat-fade,\n  .fa-flip,\n  .fa-pulse,\n  .fa-shake,\n  .fa-spin,\n  .fa-spin-pulse {\n    animation-delay: -1ms;\n    animation-duration: 1ms;\n    animation-iteration-count: 1;\n    transition-delay: 0s;\n    transition-duration: 0s;\n  }\n}\n@keyframes fa-beat {\n  0%, 90% {\n    transform: scale(1);\n  }\n  45% {\n    transform: scale(var(--fa-beat-scale, 1.25));\n  }\n}\n@keyframes fa-bounce {\n  0% {\n    transform: scale(1, 1) translateY(0);\n  }\n  10% {\n    transform: scale(var(--fa-bounce-start-scale-x, 1.1), var(--fa-bounce-start-scale-y, 0.9)) translateY(0);\n  }\n  30% {\n    transform: scale(var(--fa-bounce-jump-scale-x, 0.9), var(--fa-bounce-jump-scale-y, 1.1)) translateY(var(--fa-bounce-height, -0.5em));\n  }\n  50% {\n    transform: scale(var(--fa-bounce-land-scale-x, 1.05), var(--fa-bounce-land-scale-y, 0.95)) translateY(0);\n  }\n  57% {\n    transform: scale(1, 1) translateY(var(--fa-bounce-rebound, -0.125em));\n  }\n  64% {\n    transform: scale(1, 1) translateY(0);\n  }\n  100% {\n    transform: scale(1, 1) translateY(0);\n  }\n}\n@keyframes fa-fade {\n  50% {\n    opacity: var(--fa-fade-opacity, 0.4);\n  }\n}\n@keyframes fa-beat-fade {\n  0%, 100% {\n    opacity: var(--fa-beat-fade-opacity, 0.4);\n    transform: scale(1);\n  }\n  50% {\n    opacity: 1;\n    transform: scale(var(--fa-beat-fade-scale, 1.125));\n  }\n}\n@keyframes fa-flip {\n  50% {\n    transform: rotate3d(var(--fa-flip-x, 0), var(--fa-flip-y, 1), var(--fa-flip-z, 0), var(--fa-flip-angle, -180deg));\n  }\n}\n@keyframes fa-shake {\n  0% {\n    transform: rotate(-15deg);\n  }\n  4% {\n    transform: rotate(15deg);\n  }\n  8%, 24% {\n    transform: rotate(-18deg);\n  }\n  12%, 28% {\n    transform: rotate(18deg);\n  }\n  16% {\n    transform: rotate(-22deg);\n  }\n  20% {\n    transform: rotate(22deg);\n  }\n  32% {\n    transform: rotate(-12deg);\n  }\n  36% {\n    transform: rotate(12deg);\n  }\n  40%, 100% {\n    transform: rotate(0deg);\n  }\n}\n@keyframes fa-spin {\n  0% {\n    transform: rotate(0deg);\n  }\n  100% {\n    transform: rotate(360deg);\n  }\n}\n.fa-rotate-90 {\n  transform: rotate(90deg);\n}\n\n.fa-rotate-180 {\n  transform: rotate(180deg);\n}\n\n.fa-rotate-270 {\n  transform: rotate(270deg);\n}\n\n.fa-flip-horizontal {\n  transform: scale(-1, 1);\n}\n\n.fa-flip-vertical {\n  transform: scale(1, -1);\n}\n\n.fa-flip-both,\n.fa-flip-horizontal.fa-flip-vertical {\n  transform: scale(-1, -1);\n}\n\n.fa-rotate-by {\n  transform: rotate(var(--fa-rotate-angle, 0));\n}\n\n.fa-stack {\n  display: inline-block;\n  height: 2em;\n  line-height: 2em;\n  position: relative;\n  vertical-align: middle;\n  width: 2.5em;\n}\n\n.fa-stack-1x,\n.fa-stack-2x {\n  left: 0;\n  position: absolute;\n  text-align: center;\n  width: 100%;\n  z-index: var(--fa-stack-z-index, auto);\n}\n\n.fa-stack-1x {\n  line-height: inherit;\n}\n\n.fa-stack-2x {\n  font-size: 2em;\n}\n\n.fa-inverse {\n  color: var(--fa-inverse, #fff);\n}\n\n/* Font Awesome uses the Unicode Private Use Area (PUA) to ensure screen\nreaders do not read off random characters that represent icons */\n.fa-0::before {\n  content: \"\\30 \";\n}\n\n.fa-1::before {\n  content: \"\\31 \";\n}\n\n.fa-2::before {\n  content: \"\\32 \";\n}\n\n.fa-3::before {\n  content: \"\\33 \";\n}\n\n.fa-4::before {\n  content: \"\\34 \";\n}\n\n.fa-5::before {\n  content: \"\\35 \";\n}\n\n.fa-6::before {\n  content: \"\\36 \";\n}\n\n.fa-7::before {\n  content: \"\\37 \";\n}\n\n.fa-8::before {\n  content: \"\\38 \";\n}\n\n.fa-9::before {\n  content: \"\\39 \";\n}\n\n.fa-fill-drip::before {\n  content: \"\\f576\";\n}\n\n.fa-arrows-to-circle::before {\n  content: \"\\e4bd\";\n}\n\n.fa-circle-chevron-right::before {\n  content: \"\\f138\";\n}\n\n.fa-chevron-circle-right::before {\n  content: \"\\f138\";\n}\n\n.fa-at::before {\n  content: \"\\@\";\n}\n\n.fa-trash-can::before {\n  content: \"\\f2ed\";\n}\n\n.fa-trash-alt::before {\n  content: \"\\f2ed\";\n}\n\n.fa-text-height::before {\n  content: \"\\f034\";\n}\n\n.fa-user-xmark::before {\n  content: \"\\f235\";\n}\n\n.fa-user-times::before {\n  content: \"\\f235\";\n}\n\n.fa-stethoscope::before {\n  content: \"\\f0f1\";\n}\n\n.fa-message::before {\n  content: \"\\f27a\";\n}\n\n.fa-comment-alt::before {\n  content: \"\\f27a\";\n}\n\n.fa-info::before {\n  content: \"\\f129\";\n}\n\n.fa-down-left-and-up-right-to-center::before {\n  content: \"\\f422\";\n}\n\n.fa-compress-alt::before {\n  content: \"\\f422\";\n}\n\n.fa-explosion::before {\n  content: \"\\e4e9\";\n}\n\n.fa-file-lines::before {\n  content: \"\\f15c\";\n}\n\n.fa-file-alt::before {\n  content: \"\\f15c\";\n}\n\n.fa-file-text::before {\n  content: \"\\f15c\";\n}\n\n.fa-wave-square::before {\n  content: \"\\f83e\";\n}\n\n.fa-ring::before {\n  content: \"\\f70b\";\n}\n\n.fa-building-un::before {\n  content: \"\\e4d9\";\n}\n\n.fa-dice-three::before {\n  content: \"\\f527\";\n}\n\n.fa-calendar-days::before {\n  content: \"\\f073\";\n}\n\n.fa-calendar-alt::before {\n  content: \"\\f073\";\n}\n\n.fa-anchor-circle-check::before {\n  content: \"\\e4aa\";\n}\n\n.fa-building-circle-arrow-right::before {\n  content: \"\\e4d1\";\n}\n\n.fa-volleyball::before {\n  content: \"\\f45f\";\n}\n\n.fa-volleyball-ball::before {\n  content: \"\\f45f\";\n}\n\n.fa-arrows-up-to-line::before {\n  content: \"\\e4c2\";\n}\n\n.fa-sort-down::before {\n  content: \"\\f0dd\";\n}\n\n.fa-sort-desc::before {\n  content: \"\\f0dd\";\n}\n\n.fa-circle-minus::before {\n  content: \"\\f056\";\n}\n\n.fa-minus-circle::before {\n  content: \"\\f056\";\n}\n\n.fa-door-open::before {\n  content: \"\\f52b\";\n}\n\n.fa-right-from-bracket::before {\n  content: \"\\f2f5\";\n}\n\n.fa-sign-out-alt::before {\n  content: \"\\f2f5\";\n}\n\n.fa-atom::before {\n  content: \"\\f5d2\";\n}\n\n.fa-soap::before {\n  content: \"\\e06e\";\n}\n\n.fa-icons::before {\n  content: \"\\f86d\";\n}\n\n.fa-heart-music-camera-bolt::before {\n  content: \"\\f86d\";\n}\n\n.fa-microphone-lines-slash::before {\n  content: \"\\f539\";\n}\n\n.fa-microphone-alt-slash::before {\n  content: \"\\f539\";\n}\n\n.fa-bridge-circle-check::before {\n  content: \"\\e4c9\";\n}\n\n.fa-pump-medical::before {\n  content: \"\\e06a\";\n}\n\n.fa-fingerprint::before {\n  content: \"\\f577\";\n}\n\n.fa-hand-point-right::before {\n  content: \"\\f0a4\";\n}\n\n.fa-magnifying-glass-location::before {\n  content: \"\\f689\";\n}\n\n.fa-search-location::before {\n  content: \"\\f689\";\n}\n\n.fa-forward-step::before {\n  content: \"\\f051\";\n}\n\n.fa-step-forward::before {\n  content: \"\\f051\";\n}\n\n.fa-face-smile-beam::before {\n  content: \"\\f5b8\";\n}\n\n.fa-smile-beam::before {\n  content: \"\\f5b8\";\n}\n\n.fa-flag-checkered::before {\n  content: \"\\f11e\";\n}\n\n.fa-football::before {\n  content: \"\\f44e\";\n}\n\n.fa-football-ball::before {\n  content: \"\\f44e\";\n}\n\n.fa-school-circle-exclamation::before {\n  content: \"\\e56c\";\n}\n\n.fa-crop::before {\n  content: \"\\f125\";\n}\n\n.fa-angles-down::before {\n  content: \"\\f103\";\n}\n\n.fa-angle-double-down::before {\n  content: \"\\f103\";\n}\n\n.fa-users-rectangle::before {\n  content: \"\\e594\";\n}\n\n.fa-people-roof::before {\n  content: \"\\e537\";\n}\n\n.fa-people-line::before {\n  content: \"\\e534\";\n}\n\n.fa-beer-mug-empty::before {\n  content: \"\\f0fc\";\n}\n\n.fa-beer::before {\n  content: \"\\f0fc\";\n}\n\n.fa-diagram-predecessor::before {\n  content: \"\\e477\";\n}\n\n.fa-arrow-up-long::before {\n  content: \"\\f176\";\n}\n\n.fa-long-arrow-up::before {\n  content: \"\\f176\";\n}\n\n.fa-fire-flame-simple::before {\n  content: \"\\f46a\";\n}\n\n.fa-burn::before {\n  content: \"\\f46a\";\n}\n\n.fa-person::before {\n  content: \"\\f183\";\n}\n\n.fa-male::before {\n  content: \"\\f183\";\n}\n\n.fa-laptop::before {\n  content: \"\\f109\";\n}\n\n.fa-file-csv::before {\n  content: \"\\f6dd\";\n}\n\n.fa-menorah::before {\n  content: \"\\f676\";\n}\n\n.fa-truck-plane::before {\n  content: \"\\e58f\";\n}\n\n.fa-record-vinyl::before {\n  content: \"\\f8d9\";\n}\n\n.fa-face-grin-stars::before {\n  content: \"\\f587\";\n}\n\n.fa-grin-stars::before {\n  content: \"\\f587\";\n}\n\n.fa-bong::before {\n  content: \"\\f55c\";\n}\n\n.fa-spaghetti-monster-flying::before {\n  content: \"\\f67b\";\n}\n\n.fa-pastafarianism::before {\n  content: \"\\f67b\";\n}\n\n.fa-arrow-down-up-across-line::before {\n  content: \"\\e4af\";\n}\n\n.fa-spoon::before {\n  content: \"\\f2e5\";\n}\n\n.fa-utensil-spoon::before {\n  content: \"\\f2e5\";\n}\n\n.fa-jar-wheat::before {\n  content: \"\\e517\";\n}\n\n.fa-envelopes-bulk::before {\n  content: \"\\f674\";\n}\n\n.fa-mail-bulk::before {\n  content: \"\\f674\";\n}\n\n.fa-file-circle-exclamation::before {\n  content: \"\\e4eb\";\n}\n\n.fa-circle-h::before {\n  content: \"\\f47e\";\n}\n\n.fa-hospital-symbol::before {\n  content: \"\\f47e\";\n}\n\n.fa-pager::before {\n  content: \"\\f815\";\n}\n\n.fa-address-book::before {\n  content: \"\\f2b9\";\n}\n\n.fa-contact-book::before {\n  content: \"\\f2b9\";\n}\n\n.fa-strikethrough::before {\n  content: \"\\f0cc\";\n}\n\n.fa-k::before {\n  content: \"K\";\n}\n\n.fa-landmark-flag::before {\n  content: \"\\e51c\";\n}\n\n.fa-pencil::before {\n  content: \"\\f303\";\n}\n\n.fa-pencil-alt::before {\n  content: \"\\f303\";\n}\n\n.fa-backward::before {\n  content: \"\\f04a\";\n}\n\n.fa-caret-right::before {\n  content: \"\\f0da\";\n}\n\n.fa-comments::before {\n  content: \"\\f086\";\n}\n\n.fa-paste::before {\n  content: \"\\f0ea\";\n}\n\n.fa-file-clipboard::before {\n  content: \"\\f0ea\";\n}\n\n.fa-code-pull-request::before {\n  content: \"\\e13c\";\n}\n\n.fa-clipboard-list::before {\n  content: \"\\f46d\";\n}\n\n.fa-truck-ramp-box::before {\n  content: \"\\f4de\";\n}\n\n.fa-truck-loading::before {\n  content: \"\\f4de\";\n}\n\n.fa-user-check::before {\n  content: \"\\f4fc\";\n}\n\n.fa-vial-virus::before {\n  content: \"\\e597\";\n}\n\n.fa-sheet-plastic::before {\n  content: \"\\e571\";\n}\n\n.fa-blog::before {\n  content: \"\\f781\";\n}\n\n.fa-user-ninja::before {\n  content: \"\\f504\";\n}\n\n.fa-person-arrow-up-from-line::before {\n  content: \"\\e539\";\n}\n\n.fa-scroll-torah::before {\n  content: \"\\f6a0\";\n}\n\n.fa-torah::before {\n  content: \"\\f6a0\";\n}\n\n.fa-broom-ball::before {\n  content: \"\\f458\";\n}\n\n.fa-quidditch::before {\n  content: \"\\f458\";\n}\n\n.fa-quidditch-broom-ball::before {\n  content: \"\\f458\";\n}\n\n.fa-toggle-off::before {\n  content: \"\\f204\";\n}\n\n.fa-box-archive::before {\n  content: \"\\f187\";\n}\n\n.fa-archive::before {\n  content: \"\\f187\";\n}\n\n.fa-person-drowning::before {\n  content: \"\\e545\";\n}\n\n.fa-arrow-down-9-1::before {\n  content: \"\\f886\";\n}\n\n.fa-sort-numeric-desc::before {\n  content: \"\\f886\";\n}\n\n.fa-sort-numeric-down-alt::before {\n  content: \"\\f886\";\n}\n\n.fa-face-grin-tongue-squint::before {\n  content: \"\\f58a\";\n}\n\n.fa-grin-tongue-squint::before {\n  content: \"\\f58a\";\n}\n\n.fa-spray-can::before {\n  content: \"\\f5bd\";\n}\n\n.fa-truck-monster::before {\n  content: \"\\f63b\";\n}\n\n.fa-w::before {\n  content: \"W\";\n}\n\n.fa-earth-africa::before {\n  content: \"\\f57c\";\n}\n\n.fa-globe-africa::before {\n  content: \"\\f57c\";\n}\n\n.fa-rainbow::before {\n  content: \"\\f75b\";\n}\n\n.fa-circle-notch::before {\n  content: \"\\f1ce\";\n}\n\n.fa-tablet-screen-button::before {\n  content: \"\\f3fa\";\n}\n\n.fa-tablet-alt::before {\n  content: \"\\f3fa\";\n}\n\n.fa-paw::before {\n  content: \"\\f1b0\";\n}\n\n.fa-cloud::before {\n  content: \"\\f0c2\";\n}\n\n.fa-trowel-bricks::before {\n  content: \"\\e58a\";\n}\n\n.fa-face-flushed::before {\n  content: \"\\f579\";\n}\n\n.fa-flushed::before {\n  content: \"\\f579\";\n}\n\n.fa-hospital-user::before {\n  content: \"\\f80d\";\n}\n\n.fa-tent-arrow-left-right::before {\n  content: \"\\e57f\";\n}\n\n.fa-gavel::before {\n  content: \"\\f0e3\";\n}\n\n.fa-legal::before {\n  content: \"\\f0e3\";\n}\n\n.fa-binoculars::before {\n  content: \"\\f1e5\";\n}\n\n.fa-microphone-slash::before {\n  content: \"\\f131\";\n}\n\n.fa-box-tissue::before {\n  content: \"\\e05b\";\n}\n\n.fa-motorcycle::before {\n  content: \"\\f21c\";\n}\n\n.fa-bell-concierge::before {\n  content: \"\\f562\";\n}\n\n.fa-concierge-bell::before {\n  content: \"\\f562\";\n}\n\n.fa-pen-ruler::before {\n  content: \"\\f5ae\";\n}\n\n.fa-pencil-ruler::before {\n  content: \"\\f5ae\";\n}\n\n.fa-people-arrows::before {\n  content: \"\\e068\";\n}\n\n.fa-people-arrows-left-right::before {\n  content: \"\\e068\";\n}\n\n.fa-mars-and-venus-burst::before {\n  content: \"\\e523\";\n}\n\n.fa-square-caret-right::before {\n  content: \"\\f152\";\n}\n\n.fa-caret-square-right::before {\n  content: \"\\f152\";\n}\n\n.fa-scissors::before {\n  content: \"\\f0c4\";\n}\n\n.fa-cut::before {\n  content: \"\\f0c4\";\n}\n\n.fa-sun-plant-wilt::before {\n  content: \"\\e57a\";\n}\n\n.fa-toilets-portable::before {\n  content: \"\\e584\";\n}\n\n.fa-hockey-puck::before {\n  content: \"\\f453\";\n}\n\n.fa-table::before {\n  content: \"\\f0ce\";\n}\n\n.fa-magnifying-glass-arrow-right::before {\n  content: \"\\e521\";\n}\n\n.fa-tachograph-digital::before {\n  content: \"\\f566\";\n}\n\n.fa-digital-tachograph::before {\n  content: \"\\f566\";\n}\n\n.fa-users-slash::before {\n  content: \"\\e073\";\n}\n\n.fa-clover::before {\n  content: \"\\e139\";\n}\n\n.fa-reply::before {\n  content: \"\\f3e5\";\n}\n\n.fa-mail-reply::before {\n  content: \"\\f3e5\";\n}\n\n.fa-star-and-crescent::before {\n  content: \"\\f699\";\n}\n\n.fa-house-fire::before {\n  content: \"\\e50c\";\n}\n\n.fa-square-minus::before {\n  content: \"\\f146\";\n}\n\n.fa-minus-square::before {\n  content: \"\\f146\";\n}\n\n.fa-helicopter::before {\n  content: \"\\f533\";\n}\n\n.fa-compass::before {\n  content: \"\\f14e\";\n}\n\n.fa-square-caret-down::before {\n  content: \"\\f150\";\n}\n\n.fa-caret-square-down::before {\n  content: \"\\f150\";\n}\n\n.fa-file-circle-question::before {\n  content: \"\\e4ef\";\n}\n\n.fa-laptop-code::before {\n  content: \"\\f5fc\";\n}\n\n.fa-swatchbook::before {\n  content: \"\\f5c3\";\n}\n\n.fa-prescription-bottle::before {\n  content: \"\\f485\";\n}\n\n.fa-bars::before {\n  content: \"\\f0c9\";\n}\n\n.fa-navicon::before {\n  content: \"\\f0c9\";\n}\n\n.fa-people-group::before {\n  content: \"\\e533\";\n}\n\n.fa-hourglass-end::before {\n  content: \"\\f253\";\n}\n\n.fa-hourglass-3::before {\n  content: \"\\f253\";\n}\n\n.fa-heart-crack::before {\n  content: \"\\f7a9\";\n}\n\n.fa-heart-broken::before {\n  content: \"\\f7a9\";\n}\n\n.fa-square-up-right::before {\n  content: \"\\f360\";\n}\n\n.fa-external-link-square-alt::before {\n  content: \"\\f360\";\n}\n\n.fa-face-kiss-beam::before {\n  content: \"\\f597\";\n}\n\n.fa-kiss-beam::before {\n  content: \"\\f597\";\n}\n\n.fa-film::before {\n  content: \"\\f008\";\n}\n\n.fa-ruler-horizontal::before {\n  content: \"\\f547\";\n}\n\n.fa-people-robbery::before {\n  content: \"\\e536\";\n}\n\n.fa-lightbulb::before {\n  content: \"\\f0eb\";\n}\n\n.fa-caret-left::before {\n  content: \"\\f0d9\";\n}\n\n.fa-circle-exclamation::before {\n  content: \"\\f06a\";\n}\n\n.fa-exclamation-circle::before {\n  content: \"\\f06a\";\n}\n\n.fa-school-circle-xmark::before {\n  content: \"\\e56d\";\n}\n\n.fa-arrow-right-from-bracket::before {\n  content: \"\\f08b\";\n}\n\n.fa-sign-out::before {\n  content: \"\\f08b\";\n}\n\n.fa-circle-chevron-down::before {\n  content: \"\\f13a\";\n}\n\n.fa-chevron-circle-down::before {\n  content: \"\\f13a\";\n}\n\n.fa-unlock-keyhole::before {\n  content: \"\\f13e\";\n}\n\n.fa-unlock-alt::before {\n  content: \"\\f13e\";\n}\n\n.fa-cloud-showers-heavy::before {\n  content: \"\\f740\";\n}\n\n.fa-headphones-simple::before {\n  content: \"\\f58f\";\n}\n\n.fa-headphones-alt::before {\n  content: \"\\f58f\";\n}\n\n.fa-sitemap::before {\n  content: \"\\f0e8\";\n}\n\n.fa-circle-dollar-to-slot::before {\n  content: \"\\f4b9\";\n}\n\n.fa-donate::before {\n  content: \"\\f4b9\";\n}\n\n.fa-memory::before {\n  content: \"\\f538\";\n}\n\n.fa-road-spikes::before {\n  content: \"\\e568\";\n}\n\n.fa-fire-burner::before {\n  content: \"\\e4f1\";\n}\n\n.fa-flag::before {\n  content: \"\\f024\";\n}\n\n.fa-hanukiah::before {\n  content: \"\\f6e6\";\n}\n\n.fa-feather::before {\n  content: \"\\f52d\";\n}\n\n.fa-volume-low::before {\n  content: \"\\f027\";\n}\n\n.fa-volume-down::before {\n  content: \"\\f027\";\n}\n\n.fa-comment-slash::before {\n  content: \"\\f4b3\";\n}\n\n.fa-cloud-sun-rain::before {\n  content: \"\\f743\";\n}\n\n.fa-compress::before {\n  content: \"\\f066\";\n}\n\n.fa-wheat-awn::before {\n  content: \"\\e2cd\";\n}\n\n.fa-wheat-alt::before {\n  content: \"\\e2cd\";\n}\n\n.fa-ankh::before {\n  content: \"\\f644\";\n}\n\n.fa-hands-holding-child::before {\n  content: \"\\e4fa\";\n}\n\n.fa-asterisk::before {\n  content: \"\\*\";\n}\n\n.fa-square-check::before {\n  content: \"\\f14a\";\n}\n\n.fa-check-square::before {\n  content: \"\\f14a\";\n}\n\n.fa-peseta-sign::before {\n  content: \"\\e221\";\n}\n\n.fa-heading::before {\n  content: \"\\f1dc\";\n}\n\n.fa-header::before {\n  content: \"\\f1dc\";\n}\n\n.fa-ghost::before {\n  content: \"\\f6e2\";\n}\n\n.fa-list::before {\n  content: \"\\f03a\";\n}\n\n.fa-list-squares::before {\n  content: \"\\f03a\";\n}\n\n.fa-square-phone-flip::before {\n  content: \"\\f87b\";\n}\n\n.fa-phone-square-alt::before {\n  content: \"\\f87b\";\n}\n\n.fa-cart-plus::before {\n  content: \"\\f217\";\n}\n\n.fa-gamepad::before {\n  content: \"\\f11b\";\n}\n\n.fa-circle-dot::before {\n  content: \"\\f192\";\n}\n\n.fa-dot-circle::before {\n  content: \"\\f192\";\n}\n\n.fa-face-dizzy::before {\n  content: \"\\f567\";\n}\n\n.fa-dizzy::before {\n  content: \"\\f567\";\n}\n\n.fa-egg::before {\n  content: \"\\f7fb\";\n}\n\n.fa-house-medical-circle-xmark::before {\n  content: \"\\e513\";\n}\n\n.fa-campground::before {\n  content: \"\\f6bb\";\n}\n\n.fa-folder-plus::before {\n  content: \"\\f65e\";\n}\n\n.fa-futbol::before {\n  content: \"\\f1e3\";\n}\n\n.fa-futbol-ball::before {\n  content: \"\\f1e3\";\n}\n\n.fa-soccer-ball::before {\n  content: \"\\f1e3\";\n}\n\n.fa-paintbrush::before {\n  content: \"\\f1fc\";\n}\n\n.fa-paint-brush::before {\n  content: \"\\f1fc\";\n}\n\n.fa-lock::before {\n  content: \"\\f023\";\n}\n\n.fa-gas-pump::before {\n  content: \"\\f52f\";\n}\n\n.fa-hot-tub-person::before {\n  content: \"\\f593\";\n}\n\n.fa-hot-tub::before {\n  content: \"\\f593\";\n}\n\n.fa-map-location::before {\n  content: \"\\f59f\";\n}\n\n.fa-map-marked::before {\n  content: \"\\f59f\";\n}\n\n.fa-house-flood-water::before {\n  content: \"\\e50e\";\n}\n\n.fa-tree::before {\n  content: \"\\f1bb\";\n}\n\n.fa-bridge-lock::before {\n  content: \"\\e4cc\";\n}\n\n.fa-sack-dollar::before {\n  content: \"\\f81d\";\n}\n\n.fa-pen-to-square::before {\n  content: \"\\f044\";\n}\n\n.fa-edit::before {\n  content: \"\\f044\";\n}\n\n.fa-car-side::before {\n  content: \"\\f5e4\";\n}\n\n.fa-share-nodes::before {\n  content: \"\\f1e0\";\n}\n\n.fa-share-alt::before {\n  content: \"\\f1e0\";\n}\n\n.fa-heart-circle-minus::before {\n  content: \"\\e4ff\";\n}\n\n.fa-hourglass-half::before {\n  content: \"\\f252\";\n}\n\n.fa-hourglass-2::before {\n  content: \"\\f252\";\n}\n\n.fa-microscope::before {\n  content: \"\\f610\";\n}\n\n.fa-sink::before {\n  content: \"\\e06d\";\n}\n\n.fa-bag-shopping::before {\n  content: \"\\f290\";\n}\n\n.fa-shopping-bag::before {\n  content: \"\\f290\";\n}\n\n.fa-arrow-down-z-a::before {\n  content: \"\\f881\";\n}\n\n.fa-sort-alpha-desc::before {\n  content: \"\\f881\";\n}\n\n.fa-sort-alpha-down-alt::before {\n  content: \"\\f881\";\n}\n\n.fa-mitten::before {\n  content: \"\\f7b5\";\n}\n\n.fa-person-rays::before {\n  content: \"\\e54d\";\n}\n\n.fa-users::before {\n  content: \"\\f0c0\";\n}\n\n.fa-eye-slash::before {\n  content: \"\\f070\";\n}\n\n.fa-flask-vial::before {\n  content: \"\\e4f3\";\n}\n\n.fa-hand::before {\n  content: \"\\f256\";\n}\n\n.fa-hand-paper::before {\n  content: \"\\f256\";\n}\n\n.fa-om::before {\n  content: \"\\f679\";\n}\n\n.fa-worm::before {\n  content: \"\\e599\";\n}\n\n.fa-house-circle-xmark::before {\n  content: \"\\e50b\";\n}\n\n.fa-plug::before {\n  content: \"\\f1e6\";\n}\n\n.fa-chevron-up::before {\n  content: \"\\f077\";\n}\n\n.fa-hand-spock::before {\n  content: \"\\f259\";\n}\n\n.fa-stopwatch::before {\n  content: \"\\f2f2\";\n}\n\n.fa-face-kiss::before {\n  content: \"\\f596\";\n}\n\n.fa-kiss::before {\n  content: \"\\f596\";\n}\n\n.fa-bridge-circle-xmark::before {\n  content: \"\\e4cb\";\n}\n\n.fa-face-grin-tongue::before {\n  content: \"\\f589\";\n}\n\n.fa-grin-tongue::before {\n  content: \"\\f589\";\n}\n\n.fa-chess-bishop::before {\n  content: \"\\f43a\";\n}\n\n.fa-face-grin-wink::before {\n  content: \"\\f58c\";\n}\n\n.fa-grin-wink::before {\n  content: \"\\f58c\";\n}\n\n.fa-ear-deaf::before {\n  content: \"\\f2a4\";\n}\n\n.fa-deaf::before {\n  content: \"\\f2a4\";\n}\n\n.fa-deafness::before {\n  content: \"\\f2a4\";\n}\n\n.fa-hard-of-hearing::before {\n  content: \"\\f2a4\";\n}\n\n.fa-road-circle-check::before {\n  content: \"\\e564\";\n}\n\n.fa-dice-five::before {\n  content: \"\\f523\";\n}\n\n.fa-square-rss::before {\n  content: \"\\f143\";\n}\n\n.fa-rss-square::before {\n  content: \"\\f143\";\n}\n\n.fa-land-mine-on::before {\n  content: \"\\e51b\";\n}\n\n.fa-i-cursor::before {\n  content: \"\\f246\";\n}\n\n.fa-stamp::before {\n  content: \"\\f5bf\";\n}\n\n.fa-stairs::before {\n  content: \"\\e289\";\n}\n\n.fa-i::before {\n  content: \"I\";\n}\n\n.fa-hryvnia-sign::before {\n  content: \"\\f6f2\";\n}\n\n.fa-hryvnia::before {\n  content: \"\\f6f2\";\n}\n\n.fa-pills::before {\n  content: \"\\f484\";\n}\n\n.fa-face-grin-wide::before {\n  content: \"\\f581\";\n}\n\n.fa-grin-alt::before {\n  content: \"\\f581\";\n}\n\n.fa-tooth::before {\n  content: \"\\f5c9\";\n}\n\n.fa-v::before {\n  content: \"V\";\n}\n\n.fa-bangladeshi-taka-sign::before {\n  content: \"\\e2e6\";\n}\n\n.fa-bicycle::before {\n  content: \"\\f206\";\n}\n\n.fa-staff-snake::before {\n  content: \"\\e579\";\n}\n\n.fa-rod-asclepius::before {\n  content: \"\\e579\";\n}\n\n.fa-rod-snake::before {\n  content: \"\\e579\";\n}\n\n.fa-staff-aesculapius::before {\n  content: \"\\e579\";\n}\n\n.fa-head-side-cough-slash::before {\n  content: \"\\e062\";\n}\n\n.fa-truck-medical::before {\n  content: \"\\f0f9\";\n}\n\n.fa-ambulance::before {\n  content: \"\\f0f9\";\n}\n\n.fa-wheat-awn-circle-exclamation::before {\n  content: \"\\e598\";\n}\n\n.fa-snowman::before {\n  content: \"\\f7d0\";\n}\n\n.fa-mortar-pestle::before {\n  content: \"\\f5a7\";\n}\n\n.fa-road-barrier::before {\n  content: \"\\e562\";\n}\n\n.fa-school::before {\n  content: \"\\f549\";\n}\n\n.fa-igloo::before {\n  content: \"\\f7ae\";\n}\n\n.fa-joint::before {\n  content: \"\\f595\";\n}\n\n.fa-angle-right::before {\n  content: \"\\f105\";\n}\n\n.fa-horse::before {\n  content: \"\\f6f0\";\n}\n\n.fa-q::before {\n  content: \"Q\";\n}\n\n.fa-g::before {\n  content: \"G\";\n}\n\n.fa-notes-medical::before {\n  content: \"\\f481\";\n}\n\n.fa-temperature-half::before {\n  content: \"\\f2c9\";\n}\n\n.fa-temperature-2::before {\n  content: \"\\f2c9\";\n}\n\n.fa-thermometer-2::before {\n  content: \"\\f2c9\";\n}\n\n.fa-thermometer-half::before {\n  content: \"\\f2c9\";\n}\n\n.fa-dong-sign::before {\n  content: \"\\e169\";\n}\n\n.fa-capsules::before {\n  content: \"\\f46b\";\n}\n\n.fa-poo-storm::before {\n  content: \"\\f75a\";\n}\n\n.fa-poo-bolt::before {\n  content: \"\\f75a\";\n}\n\n.fa-face-frown-open::before {\n  content: \"\\f57a\";\n}\n\n.fa-frown-open::before {\n  content: \"\\f57a\";\n}\n\n.fa-hand-point-up::before {\n  content: \"\\f0a6\";\n}\n\n.fa-money-bill::before {\n  content: \"\\f0d6\";\n}\n\n.fa-bookmark::before {\n  content: \"\\f02e\";\n}\n\n.fa-align-justify::before {\n  content: \"\\f039\";\n}\n\n.fa-umbrella-beach::before {\n  content: \"\\f5ca\";\n}\n\n.fa-helmet-un::before {\n  content: \"\\e503\";\n}\n\n.fa-bullseye::before {\n  content: \"\\f140\";\n}\n\n.fa-bacon::before {\n  content: \"\\f7e5\";\n}\n\n.fa-hand-point-down::before {\n  content: \"\\f0a7\";\n}\n\n.fa-arrow-up-from-bracket::before {\n  content: \"\\e09a\";\n}\n\n.fa-folder::before {\n  content: \"\\f07b\";\n}\n\n.fa-folder-blank::before {\n  content: \"\\f07b\";\n}\n\n.fa-file-waveform::before {\n  content: \"\\f478\";\n}\n\n.fa-file-medical-alt::before {\n  content: \"\\f478\";\n}\n\n.fa-radiation::before {\n  content: \"\\f7b9\";\n}\n\n.fa-chart-simple::before {\n  content: \"\\e473\";\n}\n\n.fa-mars-stroke::before {\n  content: \"\\f229\";\n}\n\n.fa-vial::before {\n  content: \"\\f492\";\n}\n\n.fa-gauge::before {\n  content: \"\\f624\";\n}\n\n.fa-dashboard::before {\n  content: \"\\f624\";\n}\n\n.fa-gauge-med::before {\n  content: \"\\f624\";\n}\n\n.fa-tachometer-alt-average::before {\n  content: \"\\f624\";\n}\n\n.fa-wand-magic-sparkles::before {\n  content: \"\\e2ca\";\n}\n\n.fa-magic-wand-sparkles::before {\n  content: \"\\e2ca\";\n}\n\n.fa-e::before {\n  content: \"E\";\n}\n\n.fa-pen-clip::before {\n  content: \"\\f305\";\n}\n\n.fa-pen-alt::before {\n  content: \"\\f305\";\n}\n\n.fa-bridge-circle-exclamation::before {\n  content: \"\\e4ca\";\n}\n\n.fa-user::before {\n  content: \"\\f007\";\n}\n\n.fa-school-circle-check::before {\n  content: \"\\e56b\";\n}\n\n.fa-dumpster::before {\n  content: \"\\f793\";\n}\n\n.fa-van-shuttle::before {\n  content: \"\\f5b6\";\n}\n\n.fa-shuttle-van::before {\n  content: \"\\f5b6\";\n}\n\n.fa-building-user::before {\n  content: \"\\e4da\";\n}\n\n.fa-square-caret-left::before {\n  content: \"\\f191\";\n}\n\n.fa-caret-square-left::before {\n  content: \"\\f191\";\n}\n\n.fa-highlighter::before {\n  content: \"\\f591\";\n}\n\n.fa-key::before {\n  content: \"\\f084\";\n}\n\n.fa-bullhorn::before {\n  content: \"\\f0a1\";\n}\n\n.fa-globe::before {\n  content: \"\\f0ac\";\n}\n\n.fa-synagogue::before {\n  content: \"\\f69b\";\n}\n\n.fa-person-half-dress::before {\n  content: \"\\e548\";\n}\n\n.fa-road-bridge::before {\n  content: \"\\e563\";\n}\n\n.fa-location-arrow::before {\n  content: \"\\f124\";\n}\n\n.fa-c::before {\n  content: \"C\";\n}\n\n.fa-tablet-button::before {\n  content: \"\\f10a\";\n}\n\n.fa-building-lock::before {\n  content: \"\\e4d6\";\n}\n\n.fa-pizza-slice::before {\n  content: \"\\f818\";\n}\n\n.fa-money-bill-wave::before {\n  content: \"\\f53a\";\n}\n\n.fa-chart-area::before {\n  content: \"\\f1fe\";\n}\n\n.fa-area-chart::before {\n  content: \"\\f1fe\";\n}\n\n.fa-house-flag::before {\n  content: \"\\e50d\";\n}\n\n.fa-person-circle-minus::before {\n  content: \"\\e540\";\n}\n\n.fa-ban::before {\n  content: \"\\f05e\";\n}\n\n.fa-cancel::before {\n  content: \"\\f05e\";\n}\n\n.fa-camera-rotate::before {\n  content: \"\\e0d8\";\n}\n\n.fa-spray-can-sparkles::before {\n  content: \"\\f5d0\";\n}\n\n.fa-air-freshener::before {\n  content: \"\\f5d0\";\n}\n\n.fa-star::before {\n  content: \"\\f005\";\n}\n\n.fa-repeat::before {\n  content: \"\\f363\";\n}\n\n.fa-cross::before {\n  content: \"\\f654\";\n}\n\n.fa-box::before {\n  content: \"\\f466\";\n}\n\n.fa-venus-mars::before {\n  content: \"\\f228\";\n}\n\n.fa-arrow-pointer::before {\n  content: \"\\f245\";\n}\n\n.fa-mouse-pointer::before {\n  content: \"\\f245\";\n}\n\n.fa-maximize::before {\n  content: \"\\f31e\";\n}\n\n.fa-expand-arrows-alt::before {\n  content: \"\\f31e\";\n}\n\n.fa-charging-station::before {\n  content: \"\\f5e7\";\n}\n\n.fa-shapes::before {\n  content: \"\\f61f\";\n}\n\n.fa-triangle-circle-square::before {\n  content: \"\\f61f\";\n}\n\n.fa-shuffle::before {\n  content: \"\\f074\";\n}\n\n.fa-random::before {\n  content: \"\\f074\";\n}\n\n.fa-person-running::before {\n  content: \"\\f70c\";\n}\n\n.fa-running::before {\n  content: \"\\f70c\";\n}\n\n.fa-mobile-retro::before {\n  content: \"\\e527\";\n}\n\n.fa-grip-lines-vertical::before {\n  content: \"\\f7a5\";\n}\n\n.fa-spider::before {\n  content: \"\\f717\";\n}\n\n.fa-hands-bound::before {\n  content: \"\\e4f9\";\n}\n\n.fa-file-invoice-dollar::before {\n  content: \"\\f571\";\n}\n\n.fa-plane-circle-exclamation::before {\n  content: \"\\e556\";\n}\n\n.fa-x-ray::before {\n  content: \"\\f497\";\n}\n\n.fa-spell-check::before {\n  content: \"\\f891\";\n}\n\n.fa-slash::before {\n  content: \"\\f715\";\n}\n\n.fa-computer-mouse::before {\n  content: \"\\f8cc\";\n}\n\n.fa-mouse::before {\n  content: \"\\f8cc\";\n}\n\n.fa-arrow-right-to-bracket::before {\n  content: \"\\f090\";\n}\n\n.fa-sign-in::before {\n  content: \"\\f090\";\n}\n\n.fa-shop-slash::before {\n  content: \"\\e070\";\n}\n\n.fa-store-alt-slash::before {\n  content: \"\\e070\";\n}\n\n.fa-server::before {\n  content: \"\\f233\";\n}\n\n.fa-virus-covid-slash::before {\n  content: \"\\e4a9\";\n}\n\n.fa-shop-lock::before {\n  content: \"\\e4a5\";\n}\n\n.fa-hourglass-start::before {\n  content: \"\\f251\";\n}\n\n.fa-hourglass-1::before {\n  content: \"\\f251\";\n}\n\n.fa-blender-phone::before {\n  content: \"\\f6b6\";\n}\n\n.fa-building-wheat::before {\n  content: \"\\e4db\";\n}\n\n.fa-person-breastfeeding::before {\n  content: \"\\e53a\";\n}\n\n.fa-right-to-bracket::before {\n  content: \"\\f2f6\";\n}\n\n.fa-sign-in-alt::before {\n  content: \"\\f2f6\";\n}\n\n.fa-venus::before {\n  content: \"\\f221\";\n}\n\n.fa-passport::before {\n  content: \"\\f5ab\";\n}\n\n.fa-heart-pulse::before {\n  content: \"\\f21e\";\n}\n\n.fa-heartbeat::before {\n  content: \"\\f21e\";\n}\n\n.fa-people-carry-box::before {\n  content: \"\\f4ce\";\n}\n\n.fa-people-carry::before {\n  content: \"\\f4ce\";\n}\n\n.fa-temperature-high::before {\n  content: \"\\f769\";\n}\n\n.fa-microchip::before {\n  content: \"\\f2db\";\n}\n\n.fa-crown::before {\n  content: \"\\f521\";\n}\n\n.fa-weight-hanging::before {\n  content: \"\\f5cd\";\n}\n\n.fa-xmarks-lines::before {\n  content: \"\\e59a\";\n}\n\n.fa-file-prescription::before {\n  content: \"\\f572\";\n}\n\n.fa-weight-scale::before {\n  content: \"\\f496\";\n}\n\n.fa-weight::before {\n  content: \"\\f496\";\n}\n\n.fa-user-group::before {\n  content: \"\\f500\";\n}\n\n.fa-user-friends::before {\n  content: \"\\f500\";\n}\n\n.fa-arrow-up-a-z::before {\n  content: \"\\f15e\";\n}\n\n.fa-sort-alpha-up::before {\n  content: \"\\f15e\";\n}\n\n.fa-chess-knight::before {\n  content: \"\\f441\";\n}\n\n.fa-face-laugh-squint::before {\n  content: \"\\f59b\";\n}\n\n.fa-laugh-squint::before {\n  content: \"\\f59b\";\n}\n\n.fa-wheelchair::before {\n  content: \"\\f193\";\n}\n\n.fa-circle-arrow-up::before {\n  content: \"\\f0aa\";\n}\n\n.fa-arrow-circle-up::before {\n  content: \"\\f0aa\";\n}\n\n.fa-toggle-on::before {\n  content: \"\\f205\";\n}\n\n.fa-person-walking::before {\n  content: \"\\f554\";\n}\n\n.fa-walking::before {\n  content: \"\\f554\";\n}\n\n.fa-l::before {\n  content: \"L\";\n}\n\n.fa-fire::before {\n  content: \"\\f06d\";\n}\n\n.fa-bed-pulse::before {\n  content: \"\\f487\";\n}\n\n.fa-procedures::before {\n  content: \"\\f487\";\n}\n\n.fa-shuttle-space::before {\n  content: \"\\f197\";\n}\n\n.fa-space-shuttle::before {\n  content: \"\\f197\";\n}\n\n.fa-face-laugh::before {\n  content: \"\\f599\";\n}\n\n.fa-laugh::before {\n  content: \"\\f599\";\n}\n\n.fa-folder-open::before {\n  content: \"\\f07c\";\n}\n\n.fa-heart-circle-plus::before {\n  content: \"\\e500\";\n}\n\n.fa-code-fork::before {\n  content: \"\\e13b\";\n}\n\n.fa-city::before {\n  content: \"\\f64f\";\n}\n\n.fa-microphone-lines::before {\n  content: \"\\f3c9\";\n}\n\n.fa-microphone-alt::before {\n  content: \"\\f3c9\";\n}\n\n.fa-pepper-hot::before {\n  content: \"\\f816\";\n}\n\n.fa-unlock::before {\n  content: \"\\f09c\";\n}\n\n.fa-colon-sign::before {\n  content: \"\\e140\";\n}\n\n.fa-headset::before {\n  content: \"\\f590\";\n}\n\n.fa-store-slash::before {\n  content: \"\\e071\";\n}\n\n.fa-road-circle-xmark::before {\n  content: \"\\e566\";\n}\n\n.fa-user-minus::before {\n  content: \"\\f503\";\n}\n\n.fa-mars-stroke-up::before {\n  content: \"\\f22a\";\n}\n\n.fa-mars-stroke-v::before {\n  content: \"\\f22a\";\n}\n\n.fa-champagne-glasses::before {\n  content: \"\\f79f\";\n}\n\n.fa-glass-cheers::before {\n  content: \"\\f79f\";\n}\n\n.fa-clipboard::before {\n  content: \"\\f328\";\n}\n\n.fa-house-circle-exclamation::before {\n  content: \"\\e50a\";\n}\n\n.fa-file-arrow-up::before {\n  content: \"\\f574\";\n}\n\n.fa-file-upload::before {\n  content: \"\\f574\";\n}\n\n.fa-wifi::before {\n  content: \"\\f1eb\";\n}\n\n.fa-wifi-3::before {\n  content: \"\\f1eb\";\n}\n\n.fa-wifi-strong::before {\n  content: \"\\f1eb\";\n}\n\n.fa-bath::before {\n  content: \"\\f2cd\";\n}\n\n.fa-bathtub::before {\n  content: \"\\f2cd\";\n}\n\n.fa-underline::before {\n  content: \"\\f0cd\";\n}\n\n.fa-user-pen::before {\n  content: \"\\f4ff\";\n}\n\n.fa-user-edit::before {\n  content: \"\\f4ff\";\n}\n\n.fa-signature::before {\n  content: \"\\f5b7\";\n}\n\n.fa-stroopwafel::before {\n  content: \"\\f551\";\n}\n\n.fa-bold::before {\n  content: \"\\f032\";\n}\n\n.fa-anchor-lock::before {\n  content: \"\\e4ad\";\n}\n\n.fa-building-ngo::before {\n  content: \"\\e4d7\";\n}\n\n.fa-manat-sign::before {\n  content: \"\\e1d5\";\n}\n\n.fa-not-equal::before {\n  content: \"\\f53e\";\n}\n\n.fa-border-top-left::before {\n  content: \"\\f853\";\n}\n\n.fa-border-style::before {\n  content: \"\\f853\";\n}\n\n.fa-map-location-dot::before {\n  content: \"\\f5a0\";\n}\n\n.fa-map-marked-alt::before {\n  content: \"\\f5a0\";\n}\n\n.fa-jedi::before {\n  content: \"\\f669\";\n}\n\n.fa-square-poll-vertical::before {\n  content: \"\\f681\";\n}\n\n.fa-poll::before {\n  content: \"\\f681\";\n}\n\n.fa-mug-hot::before {\n  content: \"\\f7b6\";\n}\n\n.fa-car-battery::before {\n  content: \"\\f5df\";\n}\n\n.fa-battery-car::before {\n  content: \"\\f5df\";\n}\n\n.fa-gift::before {\n  content: \"\\f06b\";\n}\n\n.fa-dice-two::before {\n  content: \"\\f528\";\n}\n\n.fa-chess-queen::before {\n  content: \"\\f445\";\n}\n\n.fa-glasses::before {\n  content: \"\\f530\";\n}\n\n.fa-chess-board::before {\n  content: \"\\f43c\";\n}\n\n.fa-building-circle-check::before {\n  content: \"\\e4d2\";\n}\n\n.fa-person-chalkboard::before {\n  content: \"\\e53d\";\n}\n\n.fa-mars-stroke-right::before {\n  content: \"\\f22b\";\n}\n\n.fa-mars-stroke-h::before {\n  content: \"\\f22b\";\n}\n\n.fa-hand-back-fist::before {\n  content: \"\\f255\";\n}\n\n.fa-hand-rock::before {\n  content: \"\\f255\";\n}\n\n.fa-square-caret-up::before {\n  content: \"\\f151\";\n}\n\n.fa-caret-square-up::before {\n  content: \"\\f151\";\n}\n\n.fa-cloud-showers-water::before {\n  content: \"\\e4e4\";\n}\n\n.fa-chart-bar::before {\n  content: \"\\f080\";\n}\n\n.fa-bar-chart::before {\n  content: \"\\f080\";\n}\n\n.fa-hands-bubbles::before {\n  content: \"\\e05e\";\n}\n\n.fa-hands-wash::before {\n  content: \"\\e05e\";\n}\n\n.fa-less-than-equal::before {\n  content: \"\\f537\";\n}\n\n.fa-train::before {\n  content: \"\\f238\";\n}\n\n.fa-eye-low-vision::before {\n  content: \"\\f2a8\";\n}\n\n.fa-low-vision::before {\n  content: \"\\f2a8\";\n}\n\n.fa-crow::before {\n  content: \"\\f520\";\n}\n\n.fa-sailboat::before {\n  content: \"\\e445\";\n}\n\n.fa-window-restore::before {\n  content: \"\\f2d2\";\n}\n\n.fa-square-plus::before {\n  content: \"\\f0fe\";\n}\n\n.fa-plus-square::before {\n  content: \"\\f0fe\";\n}\n\n.fa-torii-gate::before {\n  content: \"\\f6a1\";\n}\n\n.fa-frog::before {\n  content: \"\\f52e\";\n}\n\n.fa-bucket::before {\n  content: \"\\e4cf\";\n}\n\n.fa-image::before {\n  content: \"\\f03e\";\n}\n\n.fa-microphone::before {\n  content: \"\\f130\";\n}\n\n.fa-cow::before {\n  content: \"\\f6c8\";\n}\n\n.fa-caret-up::before {\n  content: \"\\f0d8\";\n}\n\n.fa-screwdriver::before {\n  content: \"\\f54a\";\n}\n\n.fa-folder-closed::before {\n  content: \"\\e185\";\n}\n\n.fa-house-tsunami::before {\n  content: \"\\e515\";\n}\n\n.fa-square-nfi::before {\n  content: \"\\e576\";\n}\n\n.fa-arrow-up-from-ground-water::before {\n  content: \"\\e4b5\";\n}\n\n.fa-martini-glass::before {\n  content: \"\\f57b\";\n}\n\n.fa-glass-martini-alt::before {\n  content: \"\\f57b\";\n}\n\n.fa-rotate-left::before {\n  content: \"\\f2ea\";\n}\n\n.fa-rotate-back::before {\n  content: \"\\f2ea\";\n}\n\n.fa-rotate-backward::before {\n  content: \"\\f2ea\";\n}\n\n.fa-undo-alt::before {\n  content: \"\\f2ea\";\n}\n\n.fa-table-columns::before {\n  content: \"\\f0db\";\n}\n\n.fa-columns::before {\n  content: \"\\f0db\";\n}\n\n.fa-lemon::before {\n  content: \"\\f094\";\n}\n\n.fa-head-side-mask::before {\n  content: \"\\e063\";\n}\n\n.fa-handshake::before {\n  content: \"\\f2b5\";\n}\n\n.fa-gem::before {\n  content: \"\\f3a5\";\n}\n\n.fa-dolly::before {\n  content: \"\\f472\";\n}\n\n.fa-dolly-box::before {\n  content: \"\\f472\";\n}\n\n.fa-smoking::before {\n  content: \"\\f48d\";\n}\n\n.fa-minimize::before {\n  content: \"\\f78c\";\n}\n\n.fa-compress-arrows-alt::before {\n  content: \"\\f78c\";\n}\n\n.fa-monument::before {\n  content: \"\\f5a6\";\n}\n\n.fa-snowplow::before {\n  content: \"\\f7d2\";\n}\n\n.fa-angles-right::before {\n  content: \"\\f101\";\n}\n\n.fa-angle-double-right::before {\n  content: \"\\f101\";\n}\n\n.fa-cannabis::before {\n  content: \"\\f55f\";\n}\n\n.fa-circle-play::before {\n  content: \"\\f144\";\n}\n\n.fa-play-circle::before {\n  content: \"\\f144\";\n}\n\n.fa-tablets::before {\n  content: \"\\f490\";\n}\n\n.fa-ethernet::before {\n  content: \"\\f796\";\n}\n\n.fa-euro-sign::before {\n  content: \"\\f153\";\n}\n\n.fa-eur::before {\n  content: \"\\f153\";\n}\n\n.fa-euro::before {\n  content: \"\\f153\";\n}\n\n.fa-chair::before {\n  content: \"\\f6c0\";\n}\n\n.fa-circle-check::before {\n  content: \"\\f058\";\n}\n\n.fa-check-circle::before {\n  content: \"\\f058\";\n}\n\n.fa-circle-stop::before {\n  content: \"\\f28d\";\n}\n\n.fa-stop-circle::before {\n  content: \"\\f28d\";\n}\n\n.fa-compass-drafting::before {\n  content: \"\\f568\";\n}\n\n.fa-drafting-compass::before {\n  content: \"\\f568\";\n}\n\n.fa-plate-wheat::before {\n  content: \"\\e55a\";\n}\n\n.fa-icicles::before {\n  content: \"\\f7ad\";\n}\n\n.fa-person-shelter::before {\n  content: \"\\e54f\";\n}\n\n.fa-neuter::before {\n  content: \"\\f22c\";\n}\n\n.fa-id-badge::before {\n  content: \"\\f2c1\";\n}\n\n.fa-marker::before {\n  content: \"\\f5a1\";\n}\n\n.fa-face-laugh-beam::before {\n  content: \"\\f59a\";\n}\n\n.fa-laugh-beam::before {\n  content: \"\\f59a\";\n}\n\n.fa-helicopter-symbol::before {\n  content: \"\\e502\";\n}\n\n.fa-universal-access::before {\n  content: \"\\f29a\";\n}\n\n.fa-circle-chevron-up::before {\n  content: \"\\f139\";\n}\n\n.fa-chevron-circle-up::before {\n  content: \"\\f139\";\n}\n\n.fa-lari-sign::before {\n  content: \"\\e1c8\";\n}\n\n.fa-volcano::before {\n  content: \"\\f770\";\n}\n\n.fa-person-walking-dashed-line-arrow-right::before {\n  content: \"\\e553\";\n}\n\n.fa-sterling-sign::before {\n  content: \"\\f154\";\n}\n\n.fa-gbp::before {\n  content: \"\\f154\";\n}\n\n.fa-pound-sign::before {\n  content: \"\\f154\";\n}\n\n.fa-viruses::before {\n  content: \"\\e076\";\n}\n\n.fa-square-person-confined::before {\n  content: \"\\e577\";\n}\n\n.fa-user-tie::before {\n  content: \"\\f508\";\n}\n\n.fa-arrow-down-long::before {\n  content: \"\\f175\";\n}\n\n.fa-long-arrow-down::before {\n  content: \"\\f175\";\n}\n\n.fa-tent-arrow-down-to-line::before {\n  content: \"\\e57e\";\n}\n\n.fa-certificate::before {\n  content: \"\\f0a3\";\n}\n\n.fa-reply-all::before {\n  content: \"\\f122\";\n}\n\n.fa-mail-reply-all::before {\n  content: \"\\f122\";\n}\n\n.fa-suitcase::before {\n  content: \"\\f0f2\";\n}\n\n.fa-person-skating::before {\n  content: \"\\f7c5\";\n}\n\n.fa-skating::before {\n  content: \"\\f7c5\";\n}\n\n.fa-filter-circle-dollar::before {\n  content: \"\\f662\";\n}\n\n.fa-funnel-dollar::before {\n  content: \"\\f662\";\n}\n\n.fa-camera-retro::before {\n  content: \"\\f083\";\n}\n\n.fa-circle-arrow-down::before {\n  content: \"\\f0ab\";\n}\n\n.fa-arrow-circle-down::before {\n  content: \"\\f0ab\";\n}\n\n.fa-file-import::before {\n  content: \"\\f56f\";\n}\n\n.fa-arrow-right-to-file::before {\n  content: \"\\f56f\";\n}\n\n.fa-square-arrow-up-right::before {\n  content: \"\\f14c\";\n}\n\n.fa-external-link-square::before {\n  content: \"\\f14c\";\n}\n\n.fa-box-open::before {\n  content: \"\\f49e\";\n}\n\n.fa-scroll::before {\n  content: \"\\f70e\";\n}\n\n.fa-spa::before {\n  content: \"\\f5bb\";\n}\n\n.fa-location-pin-lock::before {\n  content: \"\\e51f\";\n}\n\n.fa-pause::before {\n  content: \"\\f04c\";\n}\n\n.fa-hill-avalanche::before {\n  content: \"\\e507\";\n}\n\n.fa-temperature-empty::before {\n  content: \"\\f2cb\";\n}\n\n.fa-temperature-0::before {\n  content: \"\\f2cb\";\n}\n\n.fa-thermometer-0::before {\n  content: \"\\f2cb\";\n}\n\n.fa-thermometer-empty::before {\n  content: \"\\f2cb\";\n}\n\n.fa-bomb::before {\n  content: \"\\f1e2\";\n}\n\n.fa-registered::before {\n  content: \"\\f25d\";\n}\n\n.fa-address-card::before {\n  content: \"\\f2bb\";\n}\n\n.fa-contact-card::before {\n  content: \"\\f2bb\";\n}\n\n.fa-vcard::before {\n  content: \"\\f2bb\";\n}\n\n.fa-scale-unbalanced-flip::before {\n  content: \"\\f516\";\n}\n\n.fa-balance-scale-right::before {\n  content: \"\\f516\";\n}\n\n.fa-subscript::before {\n  content: \"\\f12c\";\n}\n\n.fa-diamond-turn-right::before {\n  content: \"\\f5eb\";\n}\n\n.fa-directions::before {\n  content: \"\\f5eb\";\n}\n\n.fa-burst::before {\n  content: \"\\e4dc\";\n}\n\n.fa-house-laptop::before {\n  content: \"\\e066\";\n}\n\n.fa-laptop-house::before {\n  content: \"\\e066\";\n}\n\n.fa-face-tired::before {\n  content: \"\\f5c8\";\n}\n\n.fa-tired::before {\n  content: \"\\f5c8\";\n}\n\n.fa-money-bills::before {\n  content: \"\\e1f3\";\n}\n\n.fa-smog::before {\n  content: \"\\f75f\";\n}\n\n.fa-crutch::before {\n  content: \"\\f7f7\";\n}\n\n.fa-cloud-arrow-up::before {\n  content: \"\\f0ee\";\n}\n\n.fa-cloud-upload::before {\n  content: \"\\f0ee\";\n}\n\n.fa-cloud-upload-alt::before {\n  content: \"\\f0ee\";\n}\n\n.fa-palette::before {\n  content: \"\\f53f\";\n}\n\n.fa-arrows-turn-right::before {\n  content: \"\\e4c0\";\n}\n\n.fa-vest::before {\n  content: \"\\e085\";\n}\n\n.fa-ferry::before {\n  content: \"\\e4ea\";\n}\n\n.fa-arrows-down-to-people::before {\n  content: \"\\e4b9\";\n}\n\n.fa-seedling::before {\n  content: \"\\f4d8\";\n}\n\n.fa-sprout::before {\n  content: \"\\f4d8\";\n}\n\n.fa-left-right::before {\n  content: \"\\f337\";\n}\n\n.fa-arrows-alt-h::before {\n  content: \"\\f337\";\n}\n\n.fa-boxes-packing::before {\n  content: \"\\e4c7\";\n}\n\n.fa-circle-arrow-left::before {\n  content: \"\\f0a8\";\n}\n\n.fa-arrow-circle-left::before {\n  content: \"\\f0a8\";\n}\n\n.fa-group-arrows-rotate::before {\n  content: \"\\e4f6\";\n}\n\n.fa-bowl-food::before {\n  content: \"\\e4c6\";\n}\n\n.fa-candy-cane::before {\n  content: \"\\f786\";\n}\n\n.fa-arrow-down-wide-short::before {\n  content: \"\\f160\";\n}\n\n.fa-sort-amount-asc::before {\n  content: \"\\f160\";\n}\n\n.fa-sort-amount-down::before {\n  content: \"\\f160\";\n}\n\n.fa-cloud-bolt::before {\n  content: \"\\f76c\";\n}\n\n.fa-thunderstorm::before {\n  content: \"\\f76c\";\n}\n\n.fa-text-slash::before {\n  content: \"\\f87d\";\n}\n\n.fa-remove-format::before {\n  content: \"\\f87d\";\n}\n\n.fa-face-smile-wink::before {\n  content: \"\\f4da\";\n}\n\n.fa-smile-wink::before {\n  content: \"\\f4da\";\n}\n\n.fa-file-word::before {\n  content: \"\\f1c2\";\n}\n\n.fa-file-powerpoint::before {\n  content: \"\\f1c4\";\n}\n\n.fa-arrows-left-right::before {\n  content: \"\\f07e\";\n}\n\n.fa-arrows-h::before {\n  content: \"\\f07e\";\n}\n\n.fa-house-lock::before {\n  content: \"\\e510\";\n}\n\n.fa-cloud-arrow-down::before {\n  content: \"\\f0ed\";\n}\n\n.fa-cloud-download::before {\n  content: \"\\f0ed\";\n}\n\n.fa-cloud-download-alt::before {\n  content: \"\\f0ed\";\n}\n\n.fa-children::before {\n  content: \"\\e4e1\";\n}\n\n.fa-chalkboard::before {\n  content: \"\\f51b\";\n}\n\n.fa-blackboard::before {\n  content: \"\\f51b\";\n}\n\n.fa-user-large-slash::before {\n  content: \"\\f4fa\";\n}\n\n.fa-user-alt-slash::before {\n  content: \"\\f4fa\";\n}\n\n.fa-envelope-open::before {\n  content: \"\\f2b6\";\n}\n\n.fa-handshake-simple-slash::before {\n  content: \"\\e05f\";\n}\n\n.fa-handshake-alt-slash::before {\n  content: \"\\e05f\";\n}\n\n.fa-mattress-pillow::before {\n  content: \"\\e525\";\n}\n\n.fa-guarani-sign::before {\n  content: \"\\e19a\";\n}\n\n.fa-arrows-rotate::before {\n  content: \"\\f021\";\n}\n\n.fa-refresh::before {\n  content: \"\\f021\";\n}\n\n.fa-sync::before {\n  content: \"\\f021\";\n}\n\n.fa-fire-extinguisher::before {\n  content: \"\\f134\";\n}\n\n.fa-cruzeiro-sign::before {\n  content: \"\\e152\";\n}\n\n.fa-greater-than-equal::before {\n  content: \"\\f532\";\n}\n\n.fa-shield-halved::before {\n  content: \"\\f3ed\";\n}\n\n.fa-shield-alt::before {\n  content: \"\\f3ed\";\n}\n\n.fa-book-atlas::before {\n  content: \"\\f558\";\n}\n\n.fa-atlas::before {\n  content: \"\\f558\";\n}\n\n.fa-virus::before {\n  content: \"\\e074\";\n}\n\n.fa-envelope-circle-check::before {\n  content: \"\\e4e8\";\n}\n\n.fa-layer-group::before {\n  content: \"\\f5fd\";\n}\n\n.fa-arrows-to-dot::before {\n  content: \"\\e4be\";\n}\n\n.fa-archway::before {\n  content: \"\\f557\";\n}\n\n.fa-heart-circle-check::before {\n  content: \"\\e4fd\";\n}\n\n.fa-house-chimney-crack::before {\n  content: \"\\f6f1\";\n}\n\n.fa-house-damage::before {\n  content: \"\\f6f1\";\n}\n\n.fa-file-zipper::before {\n  content: \"\\f1c6\";\n}\n\n.fa-file-archive::before {\n  content: \"\\f1c6\";\n}\n\n.fa-square::before {\n  content: \"\\f0c8\";\n}\n\n.fa-martini-glass-empty::before {\n  content: \"\\f000\";\n}\n\n.fa-glass-martini::before {\n  content: \"\\f000\";\n}\n\n.fa-couch::before {\n  content: \"\\f4b8\";\n}\n\n.fa-cedi-sign::before {\n  content: \"\\e0df\";\n}\n\n.fa-italic::before {\n  content: \"\\f033\";\n}\n\n.fa-table-cells-column-lock::before {\n  content: \"\\e678\";\n}\n\n.fa-church::before {\n  content: \"\\f51d\";\n}\n\n.fa-comments-dollar::before {\n  content: \"\\f653\";\n}\n\n.fa-democrat::before {\n  content: \"\\f747\";\n}\n\n.fa-z::before {\n  content: \"Z\";\n}\n\n.fa-person-skiing::before {\n  content: \"\\f7c9\";\n}\n\n.fa-skiing::before {\n  content: \"\\f7c9\";\n}\n\n.fa-road-lock::before {\n  content: \"\\e567\";\n}\n\n.fa-a::before {\n  content: \"A\";\n}\n\n.fa-temperature-arrow-down::before {\n  content: \"\\e03f\";\n}\n\n.fa-temperature-down::before {\n  content: \"\\e03f\";\n}\n\n.fa-feather-pointed::before {\n  content: \"\\f56b\";\n}\n\n.fa-feather-alt::before {\n  content: \"\\f56b\";\n}\n\n.fa-p::before {\n  content: \"P\";\n}\n\n.fa-snowflake::before {\n  content: \"\\f2dc\";\n}\n\n.fa-newspaper::before {\n  content: \"\\f1ea\";\n}\n\n.fa-rectangle-ad::before {\n  content: \"\\f641\";\n}\n\n.fa-ad::before {\n  content: \"\\f641\";\n}\n\n.fa-circle-arrow-right::before {\n  content: \"\\f0a9\";\n}\n\n.fa-arrow-circle-right::before {\n  content: \"\\f0a9\";\n}\n\n.fa-filter-circle-xmark::before {\n  content: \"\\e17b\";\n}\n\n.fa-locust::before {\n  content: \"\\e520\";\n}\n\n.fa-sort::before {\n  content: \"\\f0dc\";\n}\n\n.fa-unsorted::before {\n  content: \"\\f0dc\";\n}\n\n.fa-list-ol::before {\n  content: \"\\f0cb\";\n}\n\n.fa-list-1-2::before {\n  content: \"\\f0cb\";\n}\n\n.fa-list-numeric::before {\n  content: \"\\f0cb\";\n}\n\n.fa-person-dress-burst::before {\n  content: \"\\e544\";\n}\n\n.fa-money-check-dollar::before {\n  content: \"\\f53d\";\n}\n\n.fa-money-check-alt::before {\n  content: \"\\f53d\";\n}\n\n.fa-vector-square::before {\n  content: \"\\f5cb\";\n}\n\n.fa-bread-slice::before {\n  content: \"\\f7ec\";\n}\n\n.fa-language::before {\n  content: \"\\f1ab\";\n}\n\n.fa-face-kiss-wink-heart::before {\n  content: \"\\f598\";\n}\n\n.fa-kiss-wink-heart::before {\n  content: \"\\f598\";\n}\n\n.fa-filter::before {\n  content: \"\\f0b0\";\n}\n\n.fa-question::before {\n  content: \"\\?\";\n}\n\n.fa-file-signature::before {\n  content: \"\\f573\";\n}\n\n.fa-up-down-left-right::before {\n  content: \"\\f0b2\";\n}\n\n.fa-arrows-alt::before {\n  content: \"\\f0b2\";\n}\n\n.fa-house-chimney-user::before {\n  content: \"\\e065\";\n}\n\n.fa-hand-holding-heart::before {\n  content: \"\\f4be\";\n}\n\n.fa-puzzle-piece::before {\n  content: \"\\f12e\";\n}\n\n.fa-money-check::before {\n  content: \"\\f53c\";\n}\n\n.fa-star-half-stroke::before {\n  content: \"\\f5c0\";\n}\n\n.fa-star-half-alt::before {\n  content: \"\\f5c0\";\n}\n\n.fa-code::before {\n  content: \"\\f121\";\n}\n\n.fa-whiskey-glass::before {\n  content: \"\\f7a0\";\n}\n\n.fa-glass-whiskey::before {\n  content: \"\\f7a0\";\n}\n\n.fa-building-circle-exclamation::before {\n  content: \"\\e4d3\";\n}\n\n.fa-magnifying-glass-chart::before {\n  content: \"\\e522\";\n}\n\n.fa-arrow-up-right-from-square::before {\n  content: \"\\f08e\";\n}\n\n.fa-external-link::before {\n  content: \"\\f08e\";\n}\n\n.fa-cubes-stacked::before {\n  content: \"\\e4e6\";\n}\n\n.fa-won-sign::before {\n  content: \"\\f159\";\n}\n\n.fa-krw::before {\n  content: \"\\f159\";\n}\n\n.fa-won::before {\n  content: \"\\f159\";\n}\n\n.fa-virus-covid::before {\n  content: \"\\e4a8\";\n}\n\n.fa-austral-sign::before {\n  content: \"\\e0a9\";\n}\n\n.fa-f::before {\n  content: \"F\";\n}\n\n.fa-leaf::before {\n  content: \"\\f06c\";\n}\n\n.fa-road::before {\n  content: \"\\f018\";\n}\n\n.fa-taxi::before {\n  content: \"\\f1ba\";\n}\n\n.fa-cab::before {\n  content: \"\\f1ba\";\n}\n\n.fa-person-circle-plus::before {\n  content: \"\\e541\";\n}\n\n.fa-chart-pie::before {\n  content: \"\\f200\";\n}\n\n.fa-pie-chart::before {\n  content: \"\\f200\";\n}\n\n.fa-bolt-lightning::before {\n  content: \"\\e0b7\";\n}\n\n.fa-sack-xmark::before {\n  content: \"\\e56a\";\n}\n\n.fa-file-excel::before {\n  content: \"\\f1c3\";\n}\n\n.fa-file-contract::before {\n  content: \"\\f56c\";\n}\n\n.fa-fish-fins::before {\n  content: \"\\e4f2\";\n}\n\n.fa-building-flag::before {\n  content: \"\\e4d5\";\n}\n\n.fa-face-grin-beam::before {\n  content: \"\\f582\";\n}\n\n.fa-grin-beam::before {\n  content: \"\\f582\";\n}\n\n.fa-object-ungroup::before {\n  content: \"\\f248\";\n}\n\n.fa-poop::before {\n  content: \"\\f619\";\n}\n\n.fa-location-pin::before {\n  content: \"\\f041\";\n}\n\n.fa-map-marker::before {\n  content: \"\\f041\";\n}\n\n.fa-kaaba::before {\n  content: \"\\f66b\";\n}\n\n.fa-toilet-paper::before {\n  content: \"\\f71e\";\n}\n\n.fa-helmet-safety::before {\n  content: \"\\f807\";\n}\n\n.fa-hard-hat::before {\n  content: \"\\f807\";\n}\n\n.fa-hat-hard::before {\n  content: \"\\f807\";\n}\n\n.fa-eject::before {\n  content: \"\\f052\";\n}\n\n.fa-circle-right::before {\n  content: \"\\f35a\";\n}\n\n.fa-arrow-alt-circle-right::before {\n  content: \"\\f35a\";\n}\n\n.fa-plane-circle-check::before {\n  content: \"\\e555\";\n}\n\n.fa-face-rolling-eyes::before {\n  content: \"\\f5a5\";\n}\n\n.fa-meh-rolling-eyes::before {\n  content: \"\\f5a5\";\n}\n\n.fa-object-group::before {\n  content: \"\\f247\";\n}\n\n.fa-chart-line::before {\n  content: \"\\f201\";\n}\n\n.fa-line-chart::before {\n  content: \"\\f201\";\n}\n\n.fa-mask-ventilator::before {\n  content: \"\\e524\";\n}\n\n.fa-arrow-right::before {\n  content: \"\\f061\";\n}\n\n.fa-signs-post::before {\n  content: \"\\f277\";\n}\n\n.fa-map-signs::before {\n  content: \"\\f277\";\n}\n\n.fa-cash-register::before {\n  content: \"\\f788\";\n}\n\n.fa-person-circle-question::before {\n  content: \"\\e542\";\n}\n\n.fa-h::before {\n  content: \"H\";\n}\n\n.fa-tarp::before {\n  content: \"\\e57b\";\n}\n\n.fa-screwdriver-wrench::before {\n  content: \"\\f7d9\";\n}\n\n.fa-tools::before {\n  content: \"\\f7d9\";\n}\n\n.fa-arrows-to-eye::before {\n  content: \"\\e4bf\";\n}\n\n.fa-plug-circle-bolt::before {\n  content: \"\\e55b\";\n}\n\n.fa-heart::before {\n  content: \"\\f004\";\n}\n\n.fa-mars-and-venus::before {\n  content: \"\\f224\";\n}\n\n.fa-house-user::before {\n  content: \"\\e1b0\";\n}\n\n.fa-home-user::before {\n  content: \"\\e1b0\";\n}\n\n.fa-dumpster-fire::before {\n  content: \"\\f794\";\n}\n\n.fa-house-crack::before {\n  content: \"\\e3b1\";\n}\n\n.fa-martini-glass-citrus::before {\n  content: \"\\f561\";\n}\n\n.fa-cocktail::before {\n  content: \"\\f561\";\n}\n\n.fa-face-surprise::before {\n  content: \"\\f5c2\";\n}\n\n.fa-surprise::before {\n  content: \"\\f5c2\";\n}\n\n.fa-bottle-water::before {\n  content: \"\\e4c5\";\n}\n\n.fa-circle-pause::before {\n  content: \"\\f28b\";\n}\n\n.fa-pause-circle::before {\n  content: \"\\f28b\";\n}\n\n.fa-toilet-paper-slash::before {\n  content: \"\\e072\";\n}\n\n.fa-apple-whole::before {\n  content: \"\\f5d1\";\n}\n\n.fa-apple-alt::before {\n  content: \"\\f5d1\";\n}\n\n.fa-kitchen-set::before {\n  content: \"\\e51a\";\n}\n\n.fa-r::before {\n  content: \"R\";\n}\n\n.fa-temperature-quarter::before {\n  content: \"\\f2ca\";\n}\n\n.fa-temperature-1::before {\n  content: \"\\f2ca\";\n}\n\n.fa-thermometer-1::before {\n  content: \"\\f2ca\";\n}\n\n.fa-thermometer-quarter::before {\n  content: \"\\f2ca\";\n}\n\n.fa-cube::before {\n  content: \"\\f1b2\";\n}\n\n.fa-bitcoin-sign::before {\n  content: \"\\e0b4\";\n}\n\n.fa-shield-dog::before {\n  content: \"\\e573\";\n}\n\n.fa-solar-panel::before {\n  content: \"\\f5ba\";\n}\n\n.fa-lock-open::before {\n  content: \"\\f3c1\";\n}\n\n.fa-elevator::before {\n  content: \"\\e16d\";\n}\n\n.fa-money-bill-transfer::before {\n  content: \"\\e528\";\n}\n\n.fa-money-bill-trend-up::before {\n  content: \"\\e529\";\n}\n\n.fa-house-flood-water-circle-arrow-right::before {\n  content: \"\\e50f\";\n}\n\n.fa-square-poll-horizontal::before {\n  content: \"\\f682\";\n}\n\n.fa-poll-h::before {\n  content: \"\\f682\";\n}\n\n.fa-circle::before {\n  content: \"\\f111\";\n}\n\n.fa-backward-fast::before {\n  content: \"\\f049\";\n}\n\n.fa-fast-backward::before {\n  content: \"\\f049\";\n}\n\n.fa-recycle::before {\n  content: \"\\f1b8\";\n}\n\n.fa-user-astronaut::before {\n  content: \"\\f4fb\";\n}\n\n.fa-plane-slash::before {\n  content: \"\\e069\";\n}\n\n.fa-trademark::before {\n  content: \"\\f25c\";\n}\n\n.fa-basketball::before {\n  content: \"\\f434\";\n}\n\n.fa-basketball-ball::before {\n  content: \"\\f434\";\n}\n\n.fa-satellite-dish::before {\n  content: \"\\f7c0\";\n}\n\n.fa-circle-up::before {\n  content: \"\\f35b\";\n}\n\n.fa-arrow-alt-circle-up::before {\n  content: \"\\f35b\";\n}\n\n.fa-mobile-screen-button::before {\n  content: \"\\f3cd\";\n}\n\n.fa-mobile-alt::before {\n  content: \"\\f3cd\";\n}\n\n.fa-volume-high::before {\n  content: \"\\f028\";\n}\n\n.fa-volume-up::before {\n  content: \"\\f028\";\n}\n\n.fa-users-rays::before {\n  content: \"\\e593\";\n}\n\n.fa-wallet::before {\n  content: \"\\f555\";\n}\n\n.fa-clipboard-check::before {\n  content: \"\\f46c\";\n}\n\n.fa-file-audio::before {\n  content: \"\\f1c7\";\n}\n\n.fa-burger::before {\n  content: \"\\f805\";\n}\n\n.fa-hamburger::before {\n  content: \"\\f805\";\n}\n\n.fa-wrench::before {\n  content: \"\\f0ad\";\n}\n\n.fa-bugs::before {\n  content: \"\\e4d0\";\n}\n\n.fa-rupee-sign::before {\n  content: \"\\f156\";\n}\n\n.fa-rupee::before {\n  content: \"\\f156\";\n}\n\n.fa-file-image::before {\n  content: \"\\f1c5\";\n}\n\n.fa-circle-question::before {\n  content: \"\\f059\";\n}\n\n.fa-question-circle::before {\n  content: \"\\f059\";\n}\n\n.fa-plane-departure::before {\n  content: \"\\f5b0\";\n}\n\n.fa-handshake-slash::before {\n  content: \"\\e060\";\n}\n\n.fa-book-bookmark::before {\n  content: \"\\e0bb\";\n}\n\n.fa-code-branch::before {\n  content: \"\\f126\";\n}\n\n.fa-hat-cowboy::before {\n  content: \"\\f8c0\";\n}\n\n.fa-bridge::before {\n  content: \"\\e4c8\";\n}\n\n.fa-phone-flip::before {\n  content: \"\\f879\";\n}\n\n.fa-phone-alt::before {\n  content: \"\\f879\";\n}\n\n.fa-truck-front::before {\n  content: \"\\e2b7\";\n}\n\n.fa-cat::before {\n  content: \"\\f6be\";\n}\n\n.fa-anchor-circle-exclamation::before {\n  content: \"\\e4ab\";\n}\n\n.fa-truck-field::before {\n  content: \"\\e58d\";\n}\n\n.fa-route::before {\n  content: \"\\f4d7\";\n}\n\n.fa-clipboard-question::before {\n  content: \"\\e4e3\";\n}\n\n.fa-panorama::before {\n  content: \"\\e209\";\n}\n\n.fa-comment-medical::before {\n  content: \"\\f7f5\";\n}\n\n.fa-teeth-open::before {\n  content: \"\\f62f\";\n}\n\n.fa-file-circle-minus::before {\n  content: \"\\e4ed\";\n}\n\n.fa-tags::before {\n  content: \"\\f02c\";\n}\n\n.fa-wine-glass::before {\n  content: \"\\f4e3\";\n}\n\n.fa-forward-fast::before {\n  content: \"\\f050\";\n}\n\n.fa-fast-forward::before {\n  content: \"\\f050\";\n}\n\n.fa-face-meh-blank::before {\n  content: \"\\f5a4\";\n}\n\n.fa-meh-blank::before {\n  content: \"\\f5a4\";\n}\n\n.fa-square-parking::before {\n  content: \"\\f540\";\n}\n\n.fa-parking::before {\n  content: \"\\f540\";\n}\n\n.fa-house-signal::before {\n  content: \"\\e012\";\n}\n\n.fa-bars-progress::before {\n  content: \"\\f828\";\n}\n\n.fa-tasks-alt::before {\n  content: \"\\f828\";\n}\n\n.fa-faucet-drip::before {\n  content: \"\\e006\";\n}\n\n.fa-cart-flatbed::before {\n  content: \"\\f474\";\n}\n\n.fa-dolly-flatbed::before {\n  content: \"\\f474\";\n}\n\n.fa-ban-smoking::before {\n  content: \"\\f54d\";\n}\n\n.fa-smoking-ban::before {\n  content: \"\\f54d\";\n}\n\n.fa-terminal::before {\n  content: \"\\f120\";\n}\n\n.fa-mobile-button::before {\n  content: \"\\f10b\";\n}\n\n.fa-house-medical-flag::before {\n  content: \"\\e514\";\n}\n\n.fa-basket-shopping::before {\n  content: \"\\f291\";\n}\n\n.fa-shopping-basket::before {\n  content: \"\\f291\";\n}\n\n.fa-tape::before {\n  content: \"\\f4db\";\n}\n\n.fa-bus-simple::before {\n  content: \"\\f55e\";\n}\n\n.fa-bus-alt::before {\n  content: \"\\f55e\";\n}\n\n.fa-eye::before {\n  content: \"\\f06e\";\n}\n\n.fa-face-sad-cry::before {\n  content: \"\\f5b3\";\n}\n\n.fa-sad-cry::before {\n  content: \"\\f5b3\";\n}\n\n.fa-audio-description::before {\n  content: \"\\f29e\";\n}\n\n.fa-person-military-to-person::before {\n  content: \"\\e54c\";\n}\n\n.fa-file-shield::before {\n  content: \"\\e4f0\";\n}\n\n.fa-user-slash::before {\n  content: \"\\f506\";\n}\n\n.fa-pen::before {\n  content: \"\\f304\";\n}\n\n.fa-tower-observation::before {\n  content: \"\\e586\";\n}\n\n.fa-file-code::before {\n  content: \"\\f1c9\";\n}\n\n.fa-signal::before {\n  content: \"\\f012\";\n}\n\n.fa-signal-5::before {\n  content: \"\\f012\";\n}\n\n.fa-signal-perfect::before {\n  content: \"\\f012\";\n}\n\n.fa-bus::before {\n  content: \"\\f207\";\n}\n\n.fa-heart-circle-xmark::before {\n  content: \"\\e501\";\n}\n\n.fa-house-chimney::before {\n  content: \"\\e3af\";\n}\n\n.fa-home-lg::before {\n  content: \"\\e3af\";\n}\n\n.fa-window-maximize::before {\n  content: \"\\f2d0\";\n}\n\n.fa-face-frown::before {\n  content: \"\\f119\";\n}\n\n.fa-frown::before {\n  content: \"\\f119\";\n}\n\n.fa-prescription::before {\n  content: \"\\f5b1\";\n}\n\n.fa-shop::before {\n  content: \"\\f54f\";\n}\n\n.fa-store-alt::before {\n  content: \"\\f54f\";\n}\n\n.fa-floppy-disk::before {\n  content: \"\\f0c7\";\n}\n\n.fa-save::before {\n  content: \"\\f0c7\";\n}\n\n.fa-vihara::before {\n  content: \"\\f6a7\";\n}\n\n.fa-scale-unbalanced::before {\n  content: \"\\f515\";\n}\n\n.fa-balance-scale-left::before {\n  content: \"\\f515\";\n}\n\n.fa-sort-up::before {\n  content: \"\\f0de\";\n}\n\n.fa-sort-asc::before {\n  content: \"\\f0de\";\n}\n\n.fa-comment-dots::before {\n  content: \"\\f4ad\";\n}\n\n.fa-commenting::before {\n  content: \"\\f4ad\";\n}\n\n.fa-plant-wilt::before {\n  content: \"\\e5aa\";\n}\n\n.fa-diamond::before {\n  content: \"\\f219\";\n}\n\n.fa-face-grin-squint::before {\n  content: \"\\f585\";\n}\n\n.fa-grin-squint::before {\n  content: \"\\f585\";\n}\n\n.fa-hand-holding-dollar::before {\n  content: \"\\f4c0\";\n}\n\n.fa-hand-holding-usd::before {\n  content: \"\\f4c0\";\n}\n\n.fa-bacterium::before {\n  content: \"\\e05a\";\n}\n\n.fa-hand-pointer::before {\n  content: \"\\f25a\";\n}\n\n.fa-drum-steelpan::before {\n  content: \"\\f56a\";\n}\n\n.fa-hand-scissors::before {\n  content: \"\\f257\";\n}\n\n.fa-hands-praying::before {\n  content: \"\\f684\";\n}\n\n.fa-praying-hands::before {\n  content: \"\\f684\";\n}\n\n.fa-arrow-rotate-right::before {\n  content: \"\\f01e\";\n}\n\n.fa-arrow-right-rotate::before {\n  content: \"\\f01e\";\n}\n\n.fa-arrow-rotate-forward::before {\n  content: \"\\f01e\";\n}\n\n.fa-redo::before {\n  content: \"\\f01e\";\n}\n\n.fa-biohazard::before {\n  content: \"\\f780\";\n}\n\n.fa-location-crosshairs::before {\n  content: \"\\f601\";\n}\n\n.fa-location::before {\n  content: \"\\f601\";\n}\n\n.fa-mars-double::before {\n  content: \"\\f227\";\n}\n\n.fa-child-dress::before {\n  content: \"\\e59c\";\n}\n\n.fa-users-between-lines::before {\n  content: \"\\e591\";\n}\n\n.fa-lungs-virus::before {\n  content: \"\\e067\";\n}\n\n.fa-face-grin-tears::before {\n  content: \"\\f588\";\n}\n\n.fa-grin-tears::before {\n  content: \"\\f588\";\n}\n\n.fa-phone::before {\n  content: \"\\f095\";\n}\n\n.fa-calendar-xmark::before {\n  content: \"\\f273\";\n}\n\n.fa-calendar-times::before {\n  content: \"\\f273\";\n}\n\n.fa-child-reaching::before {\n  content: \"\\e59d\";\n}\n\n.fa-head-side-virus::before {\n  content: \"\\e064\";\n}\n\n.fa-user-gear::before {\n  content: \"\\f4fe\";\n}\n\n.fa-user-cog::before {\n  content: \"\\f4fe\";\n}\n\n.fa-arrow-up-1-9::before {\n  content: \"\\f163\";\n}\n\n.fa-sort-numeric-up::before {\n  content: \"\\f163\";\n}\n\n.fa-door-closed::before {\n  content: \"\\f52a\";\n}\n\n.fa-shield-virus::before {\n  content: \"\\e06c\";\n}\n\n.fa-dice-six::before {\n  content: \"\\f526\";\n}\n\n.fa-mosquito-net::before {\n  content: \"\\e52c\";\n}\n\n.fa-bridge-water::before {\n  content: \"\\e4ce\";\n}\n\n.fa-person-booth::before {\n  content: \"\\f756\";\n}\n\n.fa-text-width::before {\n  content: \"\\f035\";\n}\n\n.fa-hat-wizard::before {\n  content: \"\\f6e8\";\n}\n\n.fa-pen-fancy::before {\n  content: \"\\f5ac\";\n}\n\n.fa-person-digging::before {\n  content: \"\\f85e\";\n}\n\n.fa-digging::before {\n  content: \"\\f85e\";\n}\n\n.fa-trash::before {\n  content: \"\\f1f8\";\n}\n\n.fa-gauge-simple::before {\n  content: \"\\f629\";\n}\n\n.fa-gauge-simple-med::before {\n  content: \"\\f629\";\n}\n\n.fa-tachometer-average::before {\n  content: \"\\f629\";\n}\n\n.fa-book-medical::before {\n  content: \"\\f7e6\";\n}\n\n.fa-poo::before {\n  content: \"\\f2fe\";\n}\n\n.fa-quote-right::before {\n  content: \"\\f10e\";\n}\n\n.fa-quote-right-alt::before {\n  content: \"\\f10e\";\n}\n\n.fa-shirt::before {\n  content: \"\\f553\";\n}\n\n.fa-t-shirt::before {\n  content: \"\\f553\";\n}\n\n.fa-tshirt::before {\n  content: \"\\f553\";\n}\n\n.fa-cubes::before {\n  content: \"\\f1b3\";\n}\n\n.fa-divide::before {\n  content: \"\\f529\";\n}\n\n.fa-tenge-sign::before {\n  content: \"\\f7d7\";\n}\n\n.fa-tenge::before {\n  content: \"\\f7d7\";\n}\n\n.fa-headphones::before {\n  content: \"\\f025\";\n}\n\n.fa-hands-holding::before {\n  content: \"\\f4c2\";\n}\n\n.fa-hands-clapping::before {\n  content: \"\\e1a8\";\n}\n\n.fa-republican::before {\n  content: \"\\f75e\";\n}\n\n.fa-arrow-left::before {\n  content: \"\\f060\";\n}\n\n.fa-person-circle-xmark::before {\n  content: \"\\e543\";\n}\n\n.fa-ruler::before {\n  content: \"\\f545\";\n}\n\n.fa-align-left::before {\n  content: \"\\f036\";\n}\n\n.fa-dice-d6::before {\n  content: \"\\f6d1\";\n}\n\n.fa-restroom::before {\n  content: \"\\f7bd\";\n}\n\n.fa-j::before {\n  content: \"J\";\n}\n\n.fa-users-viewfinder::before {\n  content: \"\\e595\";\n}\n\n.fa-file-video::before {\n  content: \"\\f1c8\";\n}\n\n.fa-up-right-from-square::before {\n  content: \"\\f35d\";\n}\n\n.fa-external-link-alt::before {\n  content: \"\\f35d\";\n}\n\n.fa-table-cells::before {\n  content: \"\\f00a\";\n}\n\n.fa-th::before {\n  content: \"\\f00a\";\n}\n\n.fa-file-pdf::before {\n  content: \"\\f1c1\";\n}\n\n.fa-book-bible::before {\n  content: \"\\f647\";\n}\n\n.fa-bible::before {\n  content: \"\\f647\";\n}\n\n.fa-o::before {\n  content: \"O\";\n}\n\n.fa-suitcase-medical::before {\n  content: \"\\f0fa\";\n}\n\n.fa-medkit::before {\n  content: \"\\f0fa\";\n}\n\n.fa-user-secret::before {\n  content: \"\\f21b\";\n}\n\n.fa-otter::before {\n  content: \"\\f700\";\n}\n\n.fa-person-dress::before {\n  content: \"\\f182\";\n}\n\n.fa-female::before {\n  content: \"\\f182\";\n}\n\n.fa-comment-dollar::before {\n  content: \"\\f651\";\n}\n\n.fa-business-time::before {\n  content: \"\\f64a\";\n}\n\n.fa-briefcase-clock::before {\n  content: \"\\f64a\";\n}\n\n.fa-table-cells-large::before {\n  content: \"\\f009\";\n}\n\n.fa-th-large::before {\n  content: \"\\f009\";\n}\n\n.fa-book-tanakh::before {\n  content: \"\\f827\";\n}\n\n.fa-tanakh::before {\n  content: \"\\f827\";\n}\n\n.fa-phone-volume::before {\n  content: \"\\f2a0\";\n}\n\n.fa-volume-control-phone::before {\n  content: \"\\f2a0\";\n}\n\n.fa-hat-cowboy-side::before {\n  content: \"\\f8c1\";\n}\n\n.fa-clipboard-user::before {\n  content: \"\\f7f3\";\n}\n\n.fa-child::before {\n  content: \"\\f1ae\";\n}\n\n.fa-lira-sign::before {\n  content: \"\\f195\";\n}\n\n.fa-satellite::before {\n  content: \"\\f7bf\";\n}\n\n.fa-plane-lock::before {\n  content: \"\\e558\";\n}\n\n.fa-tag::before {\n  content: \"\\f02b\";\n}\n\n.fa-comment::before {\n  content: \"\\f075\";\n}\n\n.fa-cake-candles::before {\n  content: \"\\f1fd\";\n}\n\n.fa-birthday-cake::before {\n  content: \"\\f1fd\";\n}\n\n.fa-cake::before {\n  content: \"\\f1fd\";\n}\n\n.fa-envelope::before {\n  content: \"\\f0e0\";\n}\n\n.fa-angles-up::before {\n  content: \"\\f102\";\n}\n\n.fa-angle-double-up::before {\n  content: \"\\f102\";\n}\n\n.fa-paperclip::before {\n  content: \"\\f0c6\";\n}\n\n.fa-arrow-right-to-city::before {\n  content: \"\\e4b3\";\n}\n\n.fa-ribbon::before {\n  content: \"\\f4d6\";\n}\n\n.fa-lungs::before {\n  content: \"\\f604\";\n}\n\n.fa-arrow-up-9-1::before {\n  content: \"\\f887\";\n}\n\n.fa-sort-numeric-up-alt::before {\n  content: \"\\f887\";\n}\n\n.fa-litecoin-sign::before {\n  content: \"\\e1d3\";\n}\n\n.fa-border-none::before {\n  content: \"\\f850\";\n}\n\n.fa-circle-nodes::before {\n  content: \"\\e4e2\";\n}\n\n.fa-parachute-box::before {\n  content: \"\\f4cd\";\n}\n\n.fa-indent::before {\n  content: \"\\f03c\";\n}\n\n.fa-truck-field-un::before {\n  content: \"\\e58e\";\n}\n\n.fa-hourglass::before {\n  content: \"\\f254\";\n}\n\n.fa-hourglass-empty::before {\n  content: \"\\f254\";\n}\n\n.fa-mountain::before {\n  content: \"\\f6fc\";\n}\n\n.fa-user-doctor::before {\n  content: \"\\f0f0\";\n}\n\n.fa-user-md::before {\n  content: \"\\f0f0\";\n}\n\n.fa-circle-info::before {\n  content: \"\\f05a\";\n}\n\n.fa-info-circle::before {\n  content: \"\\f05a\";\n}\n\n.fa-cloud-meatball::before {\n  content: \"\\f73b\";\n}\n\n.fa-camera::before {\n  content: \"\\f030\";\n}\n\n.fa-camera-alt::before {\n  content: \"\\f030\";\n}\n\n.fa-square-virus::before {\n  content: \"\\e578\";\n}\n\n.fa-meteor::before {\n  content: \"\\f753\";\n}\n\n.fa-car-on::before {\n  content: \"\\e4dd\";\n}\n\n.fa-sleigh::before {\n  content: \"\\f7cc\";\n}\n\n.fa-arrow-down-1-9::before {\n  content: \"\\f162\";\n}\n\n.fa-sort-numeric-asc::before {\n  content: \"\\f162\";\n}\n\n.fa-sort-numeric-down::before {\n  content: \"\\f162\";\n}\n\n.fa-hand-holding-droplet::before {\n  content: \"\\f4c1\";\n}\n\n.fa-hand-holding-water::before {\n  content: \"\\f4c1\";\n}\n\n.fa-water::before {\n  content: \"\\f773\";\n}\n\n.fa-calendar-check::before {\n  content: \"\\f274\";\n}\n\n.fa-braille::before {\n  content: \"\\f2a1\";\n}\n\n.fa-prescription-bottle-medical::before {\n  content: \"\\f486\";\n}\n\n.fa-prescription-bottle-alt::before {\n  content: \"\\f486\";\n}\n\n.fa-landmark::before {\n  content: \"\\f66f\";\n}\n\n.fa-truck::before {\n  content: \"\\f0d1\";\n}\n\n.fa-crosshairs::before {\n  content: \"\\f05b\";\n}\n\n.fa-person-cane::before {\n  content: \"\\e53c\";\n}\n\n.fa-tent::before {\n  content: \"\\e57d\";\n}\n\n.fa-vest-patches::before {\n  content: \"\\e086\";\n}\n\n.fa-check-double::before {\n  content: \"\\f560\";\n}\n\n.fa-arrow-down-a-z::before {\n  content: \"\\f15d\";\n}\n\n.fa-sort-alpha-asc::before {\n  content: \"\\f15d\";\n}\n\n.fa-sort-alpha-down::before {\n  content: \"\\f15d\";\n}\n\n.fa-money-bill-wheat::before {\n  content: \"\\e52a\";\n}\n\n.fa-cookie::before {\n  content: \"\\f563\";\n}\n\n.fa-arrow-rotate-left::before {\n  content: \"\\f0e2\";\n}\n\n.fa-arrow-left-rotate::before {\n  content: \"\\f0e2\";\n}\n\n.fa-arrow-rotate-back::before {\n  content: \"\\f0e2\";\n}\n\n.fa-arrow-rotate-backward::before {\n  content: \"\\f0e2\";\n}\n\n.fa-undo::before {\n  content: \"\\f0e2\";\n}\n\n.fa-hard-drive::before {\n  content: \"\\f0a0\";\n}\n\n.fa-hdd::before {\n  content: \"\\f0a0\";\n}\n\n.fa-face-grin-squint-tears::before {\n  content: \"\\f586\";\n}\n\n.fa-grin-squint-tears::before {\n  content: \"\\f586\";\n}\n\n.fa-dumbbell::before {\n  content: \"\\f44b\";\n}\n\n.fa-rectangle-list::before {\n  content: \"\\f022\";\n}\n\n.fa-list-alt::before {\n  content: \"\\f022\";\n}\n\n.fa-tarp-droplet::before {\n  content: \"\\e57c\";\n}\n\n.fa-house-medical-circle-check::before {\n  content: \"\\e511\";\n}\n\n.fa-person-skiing-nordic::before {\n  content: \"\\f7ca\";\n}\n\n.fa-skiing-nordic::before {\n  content: \"\\f7ca\";\n}\n\n.fa-calendar-plus::before {\n  content: \"\\f271\";\n}\n\n.fa-plane-arrival::before {\n  content: \"\\f5af\";\n}\n\n.fa-circle-left::before {\n  content: \"\\f359\";\n}\n\n.fa-arrow-alt-circle-left::before {\n  content: \"\\f359\";\n}\n\n.fa-train-subway::before {\n  content: \"\\f239\";\n}\n\n.fa-subway::before {\n  content: \"\\f239\";\n}\n\n.fa-chart-gantt::before {\n  content: \"\\e0e4\";\n}\n\n.fa-indian-rupee-sign::before {\n  content: \"\\e1bc\";\n}\n\n.fa-indian-rupee::before {\n  content: \"\\e1bc\";\n}\n\n.fa-inr::before {\n  content: \"\\e1bc\";\n}\n\n.fa-crop-simple::before {\n  content: \"\\f565\";\n}\n\n.fa-crop-alt::before {\n  content: \"\\f565\";\n}\n\n.fa-money-bill-1::before {\n  content: \"\\f3d1\";\n}\n\n.fa-money-bill-alt::before {\n  content: \"\\f3d1\";\n}\n\n.fa-left-long::before {\n  content: \"\\f30a\";\n}\n\n.fa-long-arrow-alt-left::before {\n  content: \"\\f30a\";\n}\n\n.fa-dna::before {\n  content: \"\\f471\";\n}\n\n.fa-virus-slash::before {\n  content: \"\\e075\";\n}\n\n.fa-minus::before {\n  content: \"\\f068\";\n}\n\n.fa-subtract::before {\n  content: \"\\f068\";\n}\n\n.fa-chess::before {\n  content: \"\\f439\";\n}\n\n.fa-arrow-left-long::before {\n  content: \"\\f177\";\n}\n\n.fa-long-arrow-left::before {\n  content: \"\\f177\";\n}\n\n.fa-plug-circle-check::before {\n  content: \"\\e55c\";\n}\n\n.fa-street-view::before {\n  content: \"\\f21d\";\n}\n\n.fa-franc-sign::before {\n  content: \"\\e18f\";\n}\n\n.fa-volume-off::before {\n  content: \"\\f026\";\n}\n\n.fa-hands-asl-interpreting::before {\n  content: \"\\f2a3\";\n}\n\n.fa-american-sign-language-interpreting::before {\n  content: \"\\f2a3\";\n}\n\n.fa-asl-interpreting::before {\n  content: \"\\f2a3\";\n}\n\n.fa-hands-american-sign-language-interpreting::before {\n  content: \"\\f2a3\";\n}\n\n.fa-gear::before {\n  content: \"\\f013\";\n}\n\n.fa-cog::before {\n  content: \"\\f013\";\n}\n\n.fa-droplet-slash::before {\n  content: \"\\f5c7\";\n}\n\n.fa-tint-slash::before {\n  content: \"\\f5c7\";\n}\n\n.fa-mosque::before {\n  content: \"\\f678\";\n}\n\n.fa-mosquito::before {\n  content: \"\\e52b\";\n}\n\n.fa-star-of-david::before {\n  content: \"\\f69a\";\n}\n\n.fa-person-military-rifle::before {\n  content: \"\\e54b\";\n}\n\n.fa-cart-shopping::before {\n  content: \"\\f07a\";\n}\n\n.fa-shopping-cart::before {\n  content: \"\\f07a\";\n}\n\n.fa-vials::before {\n  content: \"\\f493\";\n}\n\n.fa-plug-circle-plus::before {\n  content: \"\\e55f\";\n}\n\n.fa-place-of-worship::before {\n  content: \"\\f67f\";\n}\n\n.fa-grip-vertical::before {\n  content: \"\\f58e\";\n}\n\n.fa-arrow-turn-up::before {\n  content: \"\\f148\";\n}\n\n.fa-level-up::before {\n  content: \"\\f148\";\n}\n\n.fa-u::before {\n  content: \"U\";\n}\n\n.fa-square-root-variable::before {\n  content: \"\\f698\";\n}\n\n.fa-square-root-alt::before {\n  content: \"\\f698\";\n}\n\n.fa-clock::before {\n  content: \"\\f017\";\n}\n\n.fa-clock-four::before {\n  content: \"\\f017\";\n}\n\n.fa-backward-step::before {\n  content: \"\\f048\";\n}\n\n.fa-step-backward::before {\n  content: \"\\f048\";\n}\n\n.fa-pallet::before {\n  content: \"\\f482\";\n}\n\n.fa-faucet::before {\n  content: \"\\e005\";\n}\n\n.fa-baseball-bat-ball::before {\n  content: \"\\f432\";\n}\n\n.fa-s::before {\n  content: \"S\";\n}\n\n.fa-timeline::before {\n  content: \"\\e29c\";\n}\n\n.fa-keyboard::before {\n  content: \"\\f11c\";\n}\n\n.fa-caret-down::before {\n  content: \"\\f0d7\";\n}\n\n.fa-house-chimney-medical::before {\n  content: \"\\f7f2\";\n}\n\n.fa-clinic-medical::before {\n  content: \"\\f7f2\";\n}\n\n.fa-temperature-three-quarters::before {\n  content: \"\\f2c8\";\n}\n\n.fa-temperature-3::before {\n  content: \"\\f2c8\";\n}\n\n.fa-thermometer-3::before {\n  content: \"\\f2c8\";\n}\n\n.fa-thermometer-three-quarters::before {\n  content: \"\\f2c8\";\n}\n\n.fa-mobile-screen::before {\n  content: \"\\f3cf\";\n}\n\n.fa-mobile-android-alt::before {\n  content: \"\\f3cf\";\n}\n\n.fa-plane-up::before {\n  content: \"\\e22d\";\n}\n\n.fa-piggy-bank::before {\n  content: \"\\f4d3\";\n}\n\n.fa-battery-half::before {\n  content: \"\\f242\";\n}\n\n.fa-battery-3::before {\n  content: \"\\f242\";\n}\n\n.fa-mountain-city::before {\n  content: \"\\e52e\";\n}\n\n.fa-coins::before {\n  content: \"\\f51e\";\n}\n\n.fa-khanda::before {\n  content: \"\\f66d\";\n}\n\n.fa-sliders::before {\n  content: \"\\f1de\";\n}\n\n.fa-sliders-h::before {\n  content: \"\\f1de\";\n}\n\n.fa-folder-tree::before {\n  content: \"\\f802\";\n}\n\n.fa-network-wired::before {\n  content: \"\\f6ff\";\n}\n\n.fa-map-pin::before {\n  content: \"\\f276\";\n}\n\n.fa-hamsa::before {\n  content: \"\\f665\";\n}\n\n.fa-cent-sign::before {\n  content: \"\\e3f5\";\n}\n\n.fa-flask::before {\n  content: \"\\f0c3\";\n}\n\n.fa-person-pregnant::before {\n  content: \"\\e31e\";\n}\n\n.fa-wand-sparkles::before {\n  content: \"\\f72b\";\n}\n\n.fa-ellipsis-vertical::before {\n  content: \"\\f142\";\n}\n\n.fa-ellipsis-v::before {\n  content: \"\\f142\";\n}\n\n.fa-ticket::before {\n  content: \"\\f145\";\n}\n\n.fa-power-off::before {\n  content: \"\\f011\";\n}\n\n.fa-right-long::before {\n  content: \"\\f30b\";\n}\n\n.fa-long-arrow-alt-right::before {\n  content: \"\\f30b\";\n}\n\n.fa-flag-usa::before {\n  content: \"\\f74d\";\n}\n\n.fa-laptop-file::before {\n  content: \"\\e51d\";\n}\n\n.fa-tty::before {\n  content: \"\\f1e4\";\n}\n\n.fa-teletype::before {\n  content: \"\\f1e4\";\n}\n\n.fa-diagram-next::before {\n  content: \"\\e476\";\n}\n\n.fa-person-rifle::before {\n  content: \"\\e54e\";\n}\n\n.fa-house-medical-circle-exclamation::before {\n  content: \"\\e512\";\n}\n\n.fa-closed-captioning::before {\n  content: \"\\f20a\";\n}\n\n.fa-person-hiking::before {\n  content: \"\\f6ec\";\n}\n\n.fa-hiking::before {\n  content: \"\\f6ec\";\n}\n\n.fa-venus-double::before {\n  content: \"\\f226\";\n}\n\n.fa-images::before {\n  content: \"\\f302\";\n}\n\n.fa-calculator::before {\n  content: \"\\f1ec\";\n}\n\n.fa-people-pulling::before {\n  content: \"\\e535\";\n}\n\n.fa-n::before {\n  content: \"N\";\n}\n\n.fa-cable-car::before {\n  content: \"\\f7da\";\n}\n\n.fa-tram::before {\n  content: \"\\f7da\";\n}\n\n.fa-cloud-rain::before {\n  content: \"\\f73d\";\n}\n\n.fa-building-circle-xmark::before {\n  content: \"\\e4d4\";\n}\n\n.fa-ship::before {\n  content: \"\\f21a\";\n}\n\n.fa-arrows-down-to-line::before {\n  content: \"\\e4b8\";\n}\n\n.fa-download::before {\n  content: \"\\f019\";\n}\n\n.fa-face-grin::before {\n  content: \"\\f580\";\n}\n\n.fa-grin::before {\n  content: \"\\f580\";\n}\n\n.fa-delete-left::before {\n  content: \"\\f55a\";\n}\n\n.fa-backspace::before {\n  content: \"\\f55a\";\n}\n\n.fa-eye-dropper::before {\n  content: \"\\f1fb\";\n}\n\n.fa-eye-dropper-empty::before {\n  content: \"\\f1fb\";\n}\n\n.fa-eyedropper::before {\n  content: \"\\f1fb\";\n}\n\n.fa-file-circle-check::before {\n  content: \"\\e5a0\";\n}\n\n.fa-forward::before {\n  content: \"\\f04e\";\n}\n\n.fa-mobile::before {\n  content: \"\\f3ce\";\n}\n\n.fa-mobile-android::before {\n  content: \"\\f3ce\";\n}\n\n.fa-mobile-phone::before {\n  content: \"\\f3ce\";\n}\n\n.fa-face-meh::before {\n  content: \"\\f11a\";\n}\n\n.fa-meh::before {\n  content: \"\\f11a\";\n}\n\n.fa-align-center::before {\n  content: \"\\f037\";\n}\n\n.fa-book-skull::before {\n  content: \"\\f6b7\";\n}\n\n.fa-book-dead::before {\n  content: \"\\f6b7\";\n}\n\n.fa-id-card::before {\n  content: \"\\f2c2\";\n}\n\n.fa-drivers-license::before {\n  content: \"\\f2c2\";\n}\n\n.fa-outdent::before {\n  content: \"\\f03b\";\n}\n\n.fa-dedent::before {\n  content: \"\\f03b\";\n}\n\n.fa-heart-circle-exclamation::before {\n  content: \"\\e4fe\";\n}\n\n.fa-house::before {\n  content: \"\\f015\";\n}\n\n.fa-home::before {\n  content: \"\\f015\";\n}\n\n.fa-home-alt::before {\n  content: \"\\f015\";\n}\n\n.fa-home-lg-alt::before {\n  content: \"\\f015\";\n}\n\n.fa-calendar-week::before {\n  content: \"\\f784\";\n}\n\n.fa-laptop-medical::before {\n  content: \"\\f812\";\n}\n\n.fa-b::before {\n  content: \"B\";\n}\n\n.fa-file-medical::before {\n  content: \"\\f477\";\n}\n\n.fa-dice-one::before {\n  content: \"\\f525\";\n}\n\n.fa-kiwi-bird::before {\n  content: \"\\f535\";\n}\n\n.fa-arrow-right-arrow-left::before {\n  content: \"\\f0ec\";\n}\n\n.fa-exchange::before {\n  content: \"\\f0ec\";\n}\n\n.fa-rotate-right::before {\n  content: \"\\f2f9\";\n}\n\n.fa-redo-alt::before {\n  content: \"\\f2f9\";\n}\n\n.fa-rotate-forward::before {\n  content: \"\\f2f9\";\n}\n\n.fa-utensils::before {\n  content: \"\\f2e7\";\n}\n\n.fa-cutlery::before {\n  content: \"\\f2e7\";\n}\n\n.fa-arrow-up-wide-short::before {\n  content: \"\\f161\";\n}\n\n.fa-sort-amount-up::before {\n  content: \"\\f161\";\n}\n\n.fa-mill-sign::before {\n  content: \"\\e1ed\";\n}\n\n.fa-bowl-rice::before {\n  content: \"\\e2eb\";\n}\n\n.fa-skull::before {\n  content: \"\\f54c\";\n}\n\n.fa-tower-broadcast::before {\n  content: \"\\f519\";\n}\n\n.fa-broadcast-tower::before {\n  content: \"\\f519\";\n}\n\n.fa-truck-pickup::before {\n  content: \"\\f63c\";\n}\n\n.fa-up-long::before {\n  content: \"\\f30c\";\n}\n\n.fa-long-arrow-alt-up::before {\n  content: \"\\f30c\";\n}\n\n.fa-stop::before {\n  content: \"\\f04d\";\n}\n\n.fa-code-merge::before {\n  content: \"\\f387\";\n}\n\n.fa-upload::before {\n  content: \"\\f093\";\n}\n\n.fa-hurricane::before {\n  content: \"\\f751\";\n}\n\n.fa-mound::before {\n  content: \"\\e52d\";\n}\n\n.fa-toilet-portable::before {\n  content: \"\\e583\";\n}\n\n.fa-compact-disc::before {\n  content: \"\\f51f\";\n}\n\n.fa-file-arrow-down::before {\n  content: \"\\f56d\";\n}\n\n.fa-file-download::before {\n  content: \"\\f56d\";\n}\n\n.fa-caravan::before {\n  content: \"\\f8ff\";\n}\n\n.fa-shield-cat::before {\n  content: \"\\e572\";\n}\n\n.fa-bolt::before {\n  content: \"\\f0e7\";\n}\n\n.fa-zap::before {\n  content: \"\\f0e7\";\n}\n\n.fa-glass-water::before {\n  content: \"\\e4f4\";\n}\n\n.fa-oil-well::before {\n  content: \"\\e532\";\n}\n\n.fa-vault::before {\n  content: \"\\e2c5\";\n}\n\n.fa-mars::before {\n  content: \"\\f222\";\n}\n\n.fa-toilet::before {\n  content: \"\\f7d8\";\n}\n\n.fa-plane-circle-xmark::before {\n  content: \"\\e557\";\n}\n\n.fa-yen-sign::before {\n  content: \"\\f157\";\n}\n\n.fa-cny::before {\n  content: \"\\f157\";\n}\n\n.fa-jpy::before {\n  content: \"\\f157\";\n}\n\n.fa-rmb::before {\n  content: \"\\f157\";\n}\n\n.fa-yen::before {\n  content: \"\\f157\";\n}\n\n.fa-ruble-sign::before {\n  content: \"\\f158\";\n}\n\n.fa-rouble::before {\n  content: \"\\f158\";\n}\n\n.fa-rub::before {\n  content: \"\\f158\";\n}\n\n.fa-ruble::before {\n  content: \"\\f158\";\n}\n\n.fa-sun::before {\n  content: \"\\f185\";\n}\n\n.fa-guitar::before {\n  content: \"\\f7a6\";\n}\n\n.fa-face-laugh-wink::before {\n  content: \"\\f59c\";\n}\n\n.fa-laugh-wink::before {\n  content: \"\\f59c\";\n}\n\n.fa-horse-head::before {\n  content: \"\\f7ab\";\n}\n\n.fa-bore-hole::before {\n  content: \"\\e4c3\";\n}\n\n.fa-industry::before {\n  content: \"\\f275\";\n}\n\n.fa-circle-down::before {\n  content: \"\\f358\";\n}\n\n.fa-arrow-alt-circle-down::before {\n  content: \"\\f358\";\n}\n\n.fa-arrows-turn-to-dots::before {\n  content: \"\\e4c1\";\n}\n\n.fa-florin-sign::before {\n  content: \"\\e184\";\n}\n\n.fa-arrow-down-short-wide::before {\n  content: \"\\f884\";\n}\n\n.fa-sort-amount-desc::before {\n  content: \"\\f884\";\n}\n\n.fa-sort-amount-down-alt::before {\n  content: \"\\f884\";\n}\n\n.fa-less-than::before {\n  content: \"\\<\";\n}\n\n.fa-angle-down::before {\n  content: \"\\f107\";\n}\n\n.fa-car-tunnel::before {\n  content: \"\\e4de\";\n}\n\n.fa-head-side-cough::before {\n  content: \"\\e061\";\n}\n\n.fa-grip-lines::before {\n  content: \"\\f7a4\";\n}\n\n.fa-thumbs-down::before {\n  content: \"\\f165\";\n}\n\n.fa-user-lock::before {\n  content: \"\\f502\";\n}\n\n.fa-arrow-right-long::before {\n  content: \"\\f178\";\n}\n\n.fa-long-arrow-right::before {\n  content: \"\\f178\";\n}\n\n.fa-anchor-circle-xmark::before {\n  content: \"\\e4ac\";\n}\n\n.fa-ellipsis::before {\n  content: \"\\f141\";\n}\n\n.fa-ellipsis-h::before {\n  content: \"\\f141\";\n}\n\n.fa-chess-pawn::before {\n  content: \"\\f443\";\n}\n\n.fa-kit-medical::before {\n  content: \"\\f479\";\n}\n\n.fa-first-aid::before {\n  content: \"\\f479\";\n}\n\n.fa-person-through-window::before {\n  content: \"\\e5a9\";\n}\n\n.fa-toolbox::before {\n  content: \"\\f552\";\n}\n\n.fa-hands-holding-circle::before {\n  content: \"\\e4fb\";\n}\n\n.fa-bug::before {\n  content: \"\\f188\";\n}\n\n.fa-credit-card::before {\n  content: \"\\f09d\";\n}\n\n.fa-credit-card-alt::before {\n  content: \"\\f09d\";\n}\n\n.fa-car::before {\n  content: \"\\f1b9\";\n}\n\n.fa-automobile::before {\n  content: \"\\f1b9\";\n}\n\n.fa-hand-holding-hand::before {\n  content: \"\\e4f7\";\n}\n\n.fa-book-open-reader::before {\n  content: \"\\f5da\";\n}\n\n.fa-book-reader::before {\n  content: \"\\f5da\";\n}\n\n.fa-mountain-sun::before {\n  content: \"\\e52f\";\n}\n\n.fa-arrows-left-right-to-line::before {\n  content: \"\\e4ba\";\n}\n\n.fa-dice-d20::before {\n  content: \"\\f6cf\";\n}\n\n.fa-truck-droplet::before {\n  content: \"\\e58c\";\n}\n\n.fa-file-circle-xmark::before {\n  content: \"\\e5a1\";\n}\n\n.fa-temperature-arrow-up::before {\n  content: \"\\e040\";\n}\n\n.fa-temperature-up::before {\n  content: \"\\e040\";\n}\n\n.fa-medal::before {\n  content: \"\\f5a2\";\n}\n\n.fa-bed::before {\n  content: \"\\f236\";\n}\n\n.fa-square-h::before {\n  content: \"\\f0fd\";\n}\n\n.fa-h-square::before {\n  content: \"\\f0fd\";\n}\n\n.fa-podcast::before {\n  content: \"\\f2ce\";\n}\n\n.fa-temperature-full::before {\n  content: \"\\f2c7\";\n}\n\n.fa-temperature-4::before {\n  content: \"\\f2c7\";\n}\n\n.fa-thermometer-4::before {\n  content: \"\\f2c7\";\n}\n\n.fa-thermometer-full::before {\n  content: \"\\f2c7\";\n}\n\n.fa-bell::before {\n  content: \"\\f0f3\";\n}\n\n.fa-superscript::before {\n  content: \"\\f12b\";\n}\n\n.fa-plug-circle-xmark::before {\n  content: \"\\e560\";\n}\n\n.fa-star-of-life::before {\n  content: \"\\f621\";\n}\n\n.fa-phone-slash::before {\n  content: \"\\f3dd\";\n}\n\n.fa-paint-roller::before {\n  content: \"\\f5aa\";\n}\n\n.fa-handshake-angle::before {\n  content: \"\\f4c4\";\n}\n\n.fa-hands-helping::before {\n  content: \"\\f4c4\";\n}\n\n.fa-location-dot::before {\n  content: \"\\f3c5\";\n}\n\n.fa-map-marker-alt::before {\n  content: \"\\f3c5\";\n}\n\n.fa-file::before {\n  content: \"\\f15b\";\n}\n\n.fa-greater-than::before {\n  content: \"\\>\";\n}\n\n.fa-person-swimming::before {\n  content: \"\\f5c4\";\n}\n\n.fa-swimmer::before {\n  content: \"\\f5c4\";\n}\n\n.fa-arrow-down::before {\n  content: \"\\f063\";\n}\n\n.fa-droplet::before {\n  content: \"\\f043\";\n}\n\n.fa-tint::before {\n  content: \"\\f043\";\n}\n\n.fa-eraser::before {\n  content: \"\\f12d\";\n}\n\n.fa-earth-americas::before {\n  content: \"\\f57d\";\n}\n\n.fa-earth::before {\n  content: \"\\f57d\";\n}\n\n.fa-earth-america::before {\n  content: \"\\f57d\";\n}\n\n.fa-globe-americas::before {\n  content: \"\\f57d\";\n}\n\n.fa-person-burst::before {\n  content: \"\\e53b\";\n}\n\n.fa-dove::before {\n  content: \"\\f4ba\";\n}\n\n.fa-battery-empty::before {\n  content: \"\\f244\";\n}\n\n.fa-battery-0::before {\n  content: \"\\f244\";\n}\n\n.fa-socks::before {\n  content: \"\\f696\";\n}\n\n.fa-inbox::before {\n  content: \"\\f01c\";\n}\n\n.fa-section::before {\n  content: \"\\e447\";\n}\n\n.fa-gauge-high::before {\n  content: \"\\f625\";\n}\n\n.fa-tachometer-alt::before {\n  content: \"\\f625\";\n}\n\n.fa-tachometer-alt-fast::before {\n  content: \"\\f625\";\n}\n\n.fa-envelope-open-text::before {\n  content: \"\\f658\";\n}\n\n.fa-hospital::before {\n  content: \"\\f0f8\";\n}\n\n.fa-hospital-alt::before {\n  content: \"\\f0f8\";\n}\n\n.fa-hospital-wide::before {\n  content: \"\\f0f8\";\n}\n\n.fa-wine-bottle::before {\n  content: \"\\f72f\";\n}\n\n.fa-chess-rook::before {\n  content: \"\\f447\";\n}\n\n.fa-bars-staggered::before {\n  content: \"\\f550\";\n}\n\n.fa-reorder::before {\n  content: \"\\f550\";\n}\n\n.fa-stream::before {\n  content: \"\\f550\";\n}\n\n.fa-dharmachakra::before {\n  content: \"\\f655\";\n}\n\n.fa-hotdog::before {\n  content: \"\\f80f\";\n}\n\n.fa-person-walking-with-cane::before {\n  content: \"\\f29d\";\n}\n\n.fa-blind::before {\n  content: \"\\f29d\";\n}\n\n.fa-drum::before {\n  content: \"\\f569\";\n}\n\n.fa-ice-cream::before {\n  content: \"\\f810\";\n}\n\n.fa-heart-circle-bolt::before {\n  content: \"\\e4fc\";\n}\n\n.fa-fax::before {\n  content: \"\\f1ac\";\n}\n\n.fa-paragraph::before {\n  content: \"\\f1dd\";\n}\n\n.fa-check-to-slot::before {\n  content: \"\\f772\";\n}\n\n.fa-vote-yea::before {\n  content: \"\\f772\";\n}\n\n.fa-star-half::before {\n  content: \"\\f089\";\n}\n\n.fa-boxes-stacked::before {\n  content: \"\\f468\";\n}\n\n.fa-boxes::before {\n  content: \"\\f468\";\n}\n\n.fa-boxes-alt::before {\n  content: \"\\f468\";\n}\n\n.fa-link::before {\n  content: \"\\f0c1\";\n}\n\n.fa-chain::before {\n  content: \"\\f0c1\";\n}\n\n.fa-ear-listen::before {\n  content: \"\\f2a2\";\n}\n\n.fa-assistive-listening-systems::before {\n  content: \"\\f2a2\";\n}\n\n.fa-tree-city::before {\n  content: \"\\e587\";\n}\n\n.fa-play::before {\n  content: \"\\f04b\";\n}\n\n.fa-font::before {\n  content: \"\\f031\";\n}\n\n.fa-table-cells-row-lock::before {\n  content: \"\\e67a\";\n}\n\n.fa-rupiah-sign::before {\n  content: \"\\e23d\";\n}\n\n.fa-magnifying-glass::before {\n  content: \"\\f002\";\n}\n\n.fa-search::before {\n  content: \"\\f002\";\n}\n\n.fa-table-tennis-paddle-ball::before {\n  content: \"\\f45d\";\n}\n\n.fa-ping-pong-paddle-ball::before {\n  content: \"\\f45d\";\n}\n\n.fa-table-tennis::before {\n  content: \"\\f45d\";\n}\n\n.fa-person-dots-from-line::before {\n  content: \"\\f470\";\n}\n\n.fa-diagnoses::before {\n  content: \"\\f470\";\n}\n\n.fa-trash-can-arrow-up::before {\n  content: \"\\f82a\";\n}\n\n.fa-trash-restore-alt::before {\n  content: \"\\f82a\";\n}\n\n.fa-naira-sign::before {\n  content: \"\\e1f6\";\n}\n\n.fa-cart-arrow-down::before {\n  content: \"\\f218\";\n}\n\n.fa-walkie-talkie::before {\n  content: \"\\f8ef\";\n}\n\n.fa-file-pen::before {\n  content: \"\\f31c\";\n}\n\n.fa-file-edit::before {\n  content: \"\\f31c\";\n}\n\n.fa-receipt::before {\n  content: \"\\f543\";\n}\n\n.fa-square-pen::before {\n  content: \"\\f14b\";\n}\n\n.fa-pen-square::before {\n  content: \"\\f14b\";\n}\n\n.fa-pencil-square::before {\n  content: \"\\f14b\";\n}\n\n.fa-suitcase-rolling::before {\n  content: \"\\f5c1\";\n}\n\n.fa-person-circle-exclamation::before {\n  content: \"\\e53f\";\n}\n\n.fa-chevron-down::before {\n  content: \"\\f078\";\n}\n\n.fa-battery-full::before {\n  content: \"\\f240\";\n}\n\n.fa-battery::before {\n  content: \"\\f240\";\n}\n\n.fa-battery-5::before {\n  content: \"\\f240\";\n}\n\n.fa-skull-crossbones::before {\n  content: \"\\f714\";\n}\n\n.fa-code-compare::before {\n  content: \"\\e13a\";\n}\n\n.fa-list-ul::before {\n  content: \"\\f0ca\";\n}\n\n.fa-list-dots::before {\n  content: \"\\f0ca\";\n}\n\n.fa-school-lock::before {\n  content: \"\\e56f\";\n}\n\n.fa-tower-cell::before {\n  content: \"\\e585\";\n}\n\n.fa-down-long::before {\n  content: \"\\f309\";\n}\n\n.fa-long-arrow-alt-down::before {\n  content: \"\\f309\";\n}\n\n.fa-ranking-star::before {\n  content: \"\\e561\";\n}\n\n.fa-chess-king::before {\n  content: \"\\f43f\";\n}\n\n.fa-person-harassing::before {\n  content: \"\\e549\";\n}\n\n.fa-brazilian-real-sign::before {\n  content: \"\\e46c\";\n}\n\n.fa-landmark-dome::before {\n  content: \"\\f752\";\n}\n\n.fa-landmark-alt::before {\n  content: \"\\f752\";\n}\n\n.fa-arrow-up::before {\n  content: \"\\f062\";\n}\n\n.fa-tv::before {\n  content: \"\\f26c\";\n}\n\n.fa-television::before {\n  content: \"\\f26c\";\n}\n\n.fa-tv-alt::before {\n  content: \"\\f26c\";\n}\n\n.fa-shrimp::before {\n  content: \"\\e448\";\n}\n\n.fa-list-check::before {\n  content: \"\\f0ae\";\n}\n\n.fa-tasks::before {\n  content: \"\\f0ae\";\n}\n\n.fa-jug-detergent::before {\n  content: \"\\e519\";\n}\n\n.fa-circle-user::before {\n  content: \"\\f2bd\";\n}\n\n.fa-user-circle::before {\n  content: \"\\f2bd\";\n}\n\n.fa-user-shield::before {\n  content: \"\\f505\";\n}\n\n.fa-wind::before {\n  content: \"\\f72e\";\n}\n\n.fa-car-burst::before {\n  content: \"\\f5e1\";\n}\n\n.fa-car-crash::before {\n  content: \"\\f5e1\";\n}\n\n.fa-y::before {\n  content: \"Y\";\n}\n\n.fa-person-snowboarding::before {\n  content: \"\\f7ce\";\n}\n\n.fa-snowboarding::before {\n  content: \"\\f7ce\";\n}\n\n.fa-truck-fast::before {\n  content: \"\\f48b\";\n}\n\n.fa-shipping-fast::before {\n  content: \"\\f48b\";\n}\n\n.fa-fish::before {\n  content: \"\\f578\";\n}\n\n.fa-user-graduate::before {\n  content: \"\\f501\";\n}\n\n.fa-circle-half-stroke::before {\n  content: \"\\f042\";\n}\n\n.fa-adjust::before {\n  content: \"\\f042\";\n}\n\n.fa-clapperboard::before {\n  content: \"\\e131\";\n}\n\n.fa-circle-radiation::before {\n  content: \"\\f7ba\";\n}\n\n.fa-radiation-alt::before {\n  content: \"\\f7ba\";\n}\n\n.fa-baseball::before {\n  content: \"\\f433\";\n}\n\n.fa-baseball-ball::before {\n  content: \"\\f433\";\n}\n\n.fa-jet-fighter-up::before {\n  content: \"\\e518\";\n}\n\n.fa-diagram-project::before {\n  content: \"\\f542\";\n}\n\n.fa-project-diagram::before {\n  content: \"\\f542\";\n}\n\n.fa-copy::before {\n  content: \"\\f0c5\";\n}\n\n.fa-volume-xmark::before {\n  content: \"\\f6a9\";\n}\n\n.fa-volume-mute::before {\n  content: \"\\f6a9\";\n}\n\n.fa-volume-times::before {\n  content: \"\\f6a9\";\n}\n\n.fa-hand-sparkles::before {\n  content: \"\\e05d\";\n}\n\n.fa-grip::before {\n  content: \"\\f58d\";\n}\n\n.fa-grip-horizontal::before {\n  content: \"\\f58d\";\n}\n\n.fa-share-from-square::before {\n  content: \"\\f14d\";\n}\n\n.fa-share-square::before {\n  content: \"\\f14d\";\n}\n\n.fa-child-combatant::before {\n  content: \"\\e4e0\";\n}\n\n.fa-child-rifle::before {\n  content: \"\\e4e0\";\n}\n\n.fa-gun::before {\n  content: \"\\e19b\";\n}\n\n.fa-square-phone::before {\n  content: \"\\f098\";\n}\n\n.fa-phone-square::before {\n  content: \"\\f098\";\n}\n\n.fa-plus::before {\n  content: \"\\+\";\n}\n\n.fa-add::before {\n  content: \"\\+\";\n}\n\n.fa-expand::before {\n  content: \"\\f065\";\n}\n\n.fa-computer::before {\n  content: \"\\e4e5\";\n}\n\n.fa-xmark::before {\n  content: \"\\f00d\";\n}\n\n.fa-close::before {\n  content: \"\\f00d\";\n}\n\n.fa-multiply::before {\n  content: \"\\f00d\";\n}\n\n.fa-remove::before {\n  content: \"\\f00d\";\n}\n\n.fa-times::before {\n  content: \"\\f00d\";\n}\n\n.fa-arrows-up-down-left-right::before {\n  content: \"\\f047\";\n}\n\n.fa-arrows::before {\n  content: \"\\f047\";\n}\n\n.fa-chalkboard-user::before {\n  content: \"\\f51c\";\n}\n\n.fa-chalkboard-teacher::before {\n  content: \"\\f51c\";\n}\n\n.fa-peso-sign::before {\n  content: \"\\e222\";\n}\n\n.fa-building-shield::before {\n  content: \"\\e4d8\";\n}\n\n.fa-baby::before {\n  content: \"\\f77c\";\n}\n\n.fa-users-line::before {\n  content: \"\\e592\";\n}\n\n.fa-quote-left::before {\n  content: \"\\f10d\";\n}\n\n.fa-quote-left-alt::before {\n  content: \"\\f10d\";\n}\n\n.fa-tractor::before {\n  content: \"\\f722\";\n}\n\n.fa-trash-arrow-up::before {\n  content: \"\\f829\";\n}\n\n.fa-trash-restore::before {\n  content: \"\\f829\";\n}\n\n.fa-arrow-down-up-lock::before {\n  content: \"\\e4b0\";\n}\n\n.fa-lines-leaning::before {\n  content: \"\\e51e\";\n}\n\n.fa-ruler-combined::before {\n  content: \"\\f546\";\n}\n\n.fa-copyright::before {\n  content: \"\\f1f9\";\n}\n\n.fa-equals::before {\n  content: \"\\=\";\n}\n\n.fa-blender::before {\n  content: \"\\f517\";\n}\n\n.fa-teeth::before {\n  content: \"\\f62e\";\n}\n\n.fa-shekel-sign::before {\n  content: \"\\f20b\";\n}\n\n.fa-ils::before {\n  content: \"\\f20b\";\n}\n\n.fa-shekel::before {\n  content: \"\\f20b\";\n}\n\n.fa-sheqel::before {\n  content: \"\\f20b\";\n}\n\n.fa-sheqel-sign::before {\n  content: \"\\f20b\";\n}\n\n.fa-map::before {\n  content: \"\\f279\";\n}\n\n.fa-rocket::before {\n  content: \"\\f135\";\n}\n\n.fa-photo-film::before {\n  content: \"\\f87c\";\n}\n\n.fa-photo-video::before {\n  content: \"\\f87c\";\n}\n\n.fa-folder-minus::before {\n  content: \"\\f65d\";\n}\n\n.fa-store::before {\n  content: \"\\f54e\";\n}\n\n.fa-arrow-trend-up::before {\n  content: \"\\e098\";\n}\n\n.fa-plug-circle-minus::before {\n  content: \"\\e55e\";\n}\n\n.fa-sign-hanging::before {\n  content: \"\\f4d9\";\n}\n\n.fa-sign::before {\n  content: \"\\f4d9\";\n}\n\n.fa-bezier-curve::before {\n  content: \"\\f55b\";\n}\n\n.fa-bell-slash::before {\n  content: \"\\f1f6\";\n}\n\n.fa-tablet::before {\n  content: \"\\f3fb\";\n}\n\n.fa-tablet-android::before {\n  content: \"\\f3fb\";\n}\n\n.fa-school-flag::before {\n  content: \"\\e56e\";\n}\n\n.fa-fill::before {\n  content: \"\\f575\";\n}\n\n.fa-angle-up::before {\n  content: \"\\f106\";\n}\n\n.fa-drumstick-bite::before {\n  content: \"\\f6d7\";\n}\n\n.fa-holly-berry::before {\n  content: \"\\f7aa\";\n}\n\n.fa-chevron-left::before {\n  content: \"\\f053\";\n}\n\n.fa-bacteria::before {\n  content: \"\\e059\";\n}\n\n.fa-hand-lizard::before {\n  content: \"\\f258\";\n}\n\n.fa-notdef::before {\n  content: \"\\e1fe\";\n}\n\n.fa-disease::before {\n  content: \"\\f7fa\";\n}\n\n.fa-briefcase-medical::before {\n  content: \"\\f469\";\n}\n\n.fa-genderless::before {\n  content: \"\\f22d\";\n}\n\n.fa-chevron-right::before {\n  content: \"\\f054\";\n}\n\n.fa-retweet::before {\n  content: \"\\f079\";\n}\n\n.fa-car-rear::before {\n  content: \"\\f5de\";\n}\n\n.fa-car-alt::before {\n  content: \"\\f5de\";\n}\n\n.fa-pump-soap::before {\n  content: \"\\e06b\";\n}\n\n.fa-video-slash::before {\n  content: \"\\f4e2\";\n}\n\n.fa-battery-quarter::before {\n  content: \"\\f243\";\n}\n\n.fa-battery-2::before {\n  content: \"\\f243\";\n}\n\n.fa-radio::before {\n  content: \"\\f8d7\";\n}\n\n.fa-baby-carriage::before {\n  content: \"\\f77d\";\n}\n\n.fa-carriage-baby::before {\n  content: \"\\f77d\";\n}\n\n.fa-traffic-light::before {\n  content: \"\\f637\";\n}\n\n.fa-thermometer::before {\n  content: \"\\f491\";\n}\n\n.fa-vr-cardboard::before {\n  content: \"\\f729\";\n}\n\n.fa-hand-middle-finger::before {\n  content: \"\\f806\";\n}\n\n.fa-percent::before {\n  content: \"\\%\";\n}\n\n.fa-percentage::before {\n  content: \"\\%\";\n}\n\n.fa-truck-moving::before {\n  content: \"\\f4df\";\n}\n\n.fa-glass-water-droplet::before {\n  content: \"\\e4f5\";\n}\n\n.fa-display::before {\n  content: \"\\e163\";\n}\n\n.fa-face-smile::before {\n  content: \"\\f118\";\n}\n\n.fa-smile::before {\n  content: \"\\f118\";\n}\n\n.fa-thumbtack::before {\n  content: \"\\f08d\";\n}\n\n.fa-thumb-tack::before {\n  content: \"\\f08d\";\n}\n\n.fa-trophy::before {\n  content: \"\\f091\";\n}\n\n.fa-person-praying::before {\n  content: \"\\f683\";\n}\n\n.fa-pray::before {\n  content: \"\\f683\";\n}\n\n.fa-hammer::before {\n  content: \"\\f6e3\";\n}\n\n.fa-hand-peace::before {\n  content: \"\\f25b\";\n}\n\n.fa-rotate::before {\n  content: \"\\f2f1\";\n}\n\n.fa-sync-alt::before {\n  content: \"\\f2f1\";\n}\n\n.fa-spinner::before {\n  content: \"\\f110\";\n}\n\n.fa-robot::before {\n  content: \"\\f544\";\n}\n\n.fa-peace::before {\n  content: \"\\f67c\";\n}\n\n.fa-gears::before {\n  content: \"\\f085\";\n}\n\n.fa-cogs::before {\n  content: \"\\f085\";\n}\n\n.fa-warehouse::before {\n  content: \"\\f494\";\n}\n\n.fa-arrow-up-right-dots::before {\n  content: \"\\e4b7\";\n}\n\n.fa-splotch::before {\n  content: \"\\f5bc\";\n}\n\n.fa-face-grin-hearts::before {\n  content: \"\\f584\";\n}\n\n.fa-grin-hearts::before {\n  content: \"\\f584\";\n}\n\n.fa-dice-four::before {\n  content: \"\\f524\";\n}\n\n.fa-sim-card::before {\n  content: \"\\f7c4\";\n}\n\n.fa-transgender::before {\n  content: \"\\f225\";\n}\n\n.fa-transgender-alt::before {\n  content: \"\\f225\";\n}\n\n.fa-mercury::before {\n  content: \"\\f223\";\n}\n\n.fa-arrow-turn-down::before {\n  content: \"\\f149\";\n}\n\n.fa-level-down::before {\n  content: \"\\f149\";\n}\n\n.fa-person-falling-burst::before {\n  content: \"\\e547\";\n}\n\n.fa-award::before {\n  content: \"\\f559\";\n}\n\n.fa-ticket-simple::before {\n  content: \"\\f3ff\";\n}\n\n.fa-ticket-alt::before {\n  content: \"\\f3ff\";\n}\n\n.fa-building::before {\n  content: \"\\f1ad\";\n}\n\n.fa-angles-left::before {\n  content: \"\\f100\";\n}\n\n.fa-angle-double-left::before {\n  content: \"\\f100\";\n}\n\n.fa-qrcode::before {\n  content: \"\\f029\";\n}\n\n.fa-clock-rotate-left::before {\n  content: \"\\f1da\";\n}\n\n.fa-history::before {\n  content: \"\\f1da\";\n}\n\n.fa-face-grin-beam-sweat::before {\n  content: \"\\f583\";\n}\n\n.fa-grin-beam-sweat::before {\n  content: \"\\f583\";\n}\n\n.fa-file-export::before {\n  content: \"\\f56e\";\n}\n\n.fa-arrow-right-from-file::before {\n  content: \"\\f56e\";\n}\n\n.fa-shield::before {\n  content: \"\\f132\";\n}\n\n.fa-shield-blank::before {\n  content: \"\\f132\";\n}\n\n.fa-arrow-up-short-wide::before {\n  content: \"\\f885\";\n}\n\n.fa-sort-amount-up-alt::before {\n  content: \"\\f885\";\n}\n\n.fa-house-medical::before {\n  content: \"\\e3b2\";\n}\n\n.fa-golf-ball-tee::before {\n  content: \"\\f450\";\n}\n\n.fa-golf-ball::before {\n  content: \"\\f450\";\n}\n\n.fa-circle-chevron-left::before {\n  content: \"\\f137\";\n}\n\n.fa-chevron-circle-left::before {\n  content: \"\\f137\";\n}\n\n.fa-house-chimney-window::before {\n  content: \"\\e00d\";\n}\n\n.fa-pen-nib::before {\n  content: \"\\f5ad\";\n}\n\n.fa-tent-arrow-turn-left::before {\n  content: \"\\e580\";\n}\n\n.fa-tents::before {\n  content: \"\\e582\";\n}\n\n.fa-wand-magic::before {\n  content: \"\\f0d0\";\n}\n\n.fa-magic::before {\n  content: \"\\f0d0\";\n}\n\n.fa-dog::before {\n  content: \"\\f6d3\";\n}\n\n.fa-carrot::before {\n  content: \"\\f787\";\n}\n\n.fa-moon::before {\n  content: \"\\f186\";\n}\n\n.fa-wine-glass-empty::before {\n  content: \"\\f5ce\";\n}\n\n.fa-wine-glass-alt::before {\n  content: \"\\f5ce\";\n}\n\n.fa-cheese::before {\n  content: \"\\f7ef\";\n}\n\n.fa-yin-yang::before {\n  content: \"\\f6ad\";\n}\n\n.fa-music::before {\n  content: \"\\f001\";\n}\n\n.fa-code-commit::before {\n  content: \"\\f386\";\n}\n\n.fa-temperature-low::before {\n  content: \"\\f76b\";\n}\n\n.fa-person-biking::before {\n  content: \"\\f84a\";\n}\n\n.fa-biking::before {\n  content: \"\\f84a\";\n}\n\n.fa-broom::before {\n  content: \"\\f51a\";\n}\n\n.fa-shield-heart::before {\n  content: \"\\e574\";\n}\n\n.fa-gopuram::before {\n  content: \"\\f664\";\n}\n\n.fa-earth-oceania::before {\n  content: \"\\e47b\";\n}\n\n.fa-globe-oceania::before {\n  content: \"\\e47b\";\n}\n\n.fa-square-xmark::before {\n  content: \"\\f2d3\";\n}\n\n.fa-times-square::before {\n  content: \"\\f2d3\";\n}\n\n.fa-xmark-square::before {\n  content: \"\\f2d3\";\n}\n\n.fa-hashtag::before {\n  content: \"\\#\";\n}\n\n.fa-up-right-and-down-left-from-center::before {\n  content: \"\\f424\";\n}\n\n.fa-expand-alt::before {\n  content: \"\\f424\";\n}\n\n.fa-oil-can::before {\n  content: \"\\f613\";\n}\n\n.fa-t::before {\n  content: \"T\";\n}\n\n.fa-hippo::before {\n  content: \"\\f6ed\";\n}\n\n.fa-chart-column::before {\n  content: \"\\e0e3\";\n}\n\n.fa-infinity::before {\n  content: \"\\f534\";\n}\n\n.fa-vial-circle-check::before {\n  content: \"\\e596\";\n}\n\n.fa-person-arrow-down-to-line::before {\n  content: \"\\e538\";\n}\n\n.fa-voicemail::before {\n  content: \"\\f897\";\n}\n\n.fa-fan::before {\n  content: \"\\f863\";\n}\n\n.fa-person-walking-luggage::before {\n  content: \"\\e554\";\n}\n\n.fa-up-down::before {\n  content: \"\\f338\";\n}\n\n.fa-arrows-alt-v::before {\n  content: \"\\f338\";\n}\n\n.fa-cloud-moon-rain::before {\n  content: \"\\f73c\";\n}\n\n.fa-calendar::before {\n  content: \"\\f133\";\n}\n\n.fa-trailer::before {\n  content: \"\\e041\";\n}\n\n.fa-bahai::before {\n  content: \"\\f666\";\n}\n\n.fa-haykal::before {\n  content: \"\\f666\";\n}\n\n.fa-sd-card::before {\n  content: \"\\f7c2\";\n}\n\n.fa-dragon::before {\n  content: \"\\f6d5\";\n}\n\n.fa-shoe-prints::before {\n  content: \"\\f54b\";\n}\n\n.fa-circle-plus::before {\n  content: \"\\f055\";\n}\n\n.fa-plus-circle::before {\n  content: \"\\f055\";\n}\n\n.fa-face-grin-tongue-wink::before {\n  content: \"\\f58b\";\n}\n\n.fa-grin-tongue-wink::before {\n  content: \"\\f58b\";\n}\n\n.fa-hand-holding::before {\n  content: \"\\f4bd\";\n}\n\n.fa-plug-circle-exclamation::before {\n  content: \"\\e55d\";\n}\n\n.fa-link-slash::before {\n  content: \"\\f127\";\n}\n\n.fa-chain-broken::before {\n  content: \"\\f127\";\n}\n\n.fa-chain-slash::before {\n  content: \"\\f127\";\n}\n\n.fa-unlink::before {\n  content: \"\\f127\";\n}\n\n.fa-clone::before {\n  content: \"\\f24d\";\n}\n\n.fa-person-walking-arrow-loop-left::before {\n  content: \"\\e551\";\n}\n\n.fa-arrow-up-z-a::before {\n  content: \"\\f882\";\n}\n\n.fa-sort-alpha-up-alt::before {\n  content: \"\\f882\";\n}\n\n.fa-fire-flame-curved::before {\n  content: \"\\f7e4\";\n}\n\n.fa-fire-alt::before {\n  content: \"\\f7e4\";\n}\n\n.fa-tornado::before {\n  content: \"\\f76f\";\n}\n\n.fa-file-circle-plus::before {\n  content: \"\\e494\";\n}\n\n.fa-book-quran::before {\n  content: \"\\f687\";\n}\n\n.fa-quran::before {\n  content: \"\\f687\";\n}\n\n.fa-anchor::before {\n  content: \"\\f13d\";\n}\n\n.fa-border-all::before {\n  content: \"\\f84c\";\n}\n\n.fa-face-angry::before {\n  content: \"\\f556\";\n}\n\n.fa-angry::before {\n  content: \"\\f556\";\n}\n\n.fa-cookie-bite::before {\n  content: \"\\f564\";\n}\n\n.fa-arrow-trend-down::before {\n  content: \"\\e097\";\n}\n\n.fa-rss::before {\n  content: \"\\f09e\";\n}\n\n.fa-feed::before {\n  content: \"\\f09e\";\n}\n\n.fa-draw-polygon::before {\n  content: \"\\f5ee\";\n}\n\n.fa-scale-balanced::before {\n  content: \"\\f24e\";\n}\n\n.fa-balance-scale::before {\n  content: \"\\f24e\";\n}\n\n.fa-gauge-simple-high::before {\n  content: \"\\f62a\";\n}\n\n.fa-tachometer::before {\n  content: \"\\f62a\";\n}\n\n.fa-tachometer-fast::before {\n  content: \"\\f62a\";\n}\n\n.fa-shower::before {\n  content: \"\\f2cc\";\n}\n\n.fa-desktop::before {\n  content: \"\\f390\";\n}\n\n.fa-desktop-alt::before {\n  content: \"\\f390\";\n}\n\n.fa-m::before {\n  content: \"M\";\n}\n\n.fa-table-list::before {\n  content: \"\\f00b\";\n}\n\n.fa-th-list::before {\n  content: \"\\f00b\";\n}\n\n.fa-comment-sms::before {\n  content: \"\\f7cd\";\n}\n\n.fa-sms::before {\n  content: \"\\f7cd\";\n}\n\n.fa-book::before {\n  content: \"\\f02d\";\n}\n\n.fa-user-plus::before {\n  content: \"\\f234\";\n}\n\n.fa-check::before {\n  content: \"\\f00c\";\n}\n\n.fa-battery-three-quarters::before {\n  content: \"\\f241\";\n}\n\n.fa-battery-4::before {\n  content: \"\\f241\";\n}\n\n.fa-house-circle-check::before {\n  content: \"\\e509\";\n}\n\n.fa-angle-left::before {\n  content: \"\\f104\";\n}\n\n.fa-diagram-successor::before {\n  content: \"\\e47a\";\n}\n\n.fa-truck-arrow-right::before {\n  content: \"\\e58b\";\n}\n\n.fa-arrows-split-up-and-left::before {\n  content: \"\\e4bc\";\n}\n\n.fa-hand-fist::before {\n  content: \"\\f6de\";\n}\n\n.fa-fist-raised::before {\n  content: \"\\f6de\";\n}\n\n.fa-cloud-moon::before {\n  content: \"\\f6c3\";\n}\n\n.fa-briefcase::before {\n  content: \"\\f0b1\";\n}\n\n.fa-person-falling::before {\n  content: \"\\e546\";\n}\n\n.fa-image-portrait::before {\n  content: \"\\f3e0\";\n}\n\n.fa-portrait::before {\n  content: \"\\f3e0\";\n}\n\n.fa-user-tag::before {\n  content: \"\\f507\";\n}\n\n.fa-rug::before {\n  content: \"\\e569\";\n}\n\n.fa-earth-europe::before {\n  content: \"\\f7a2\";\n}\n\n.fa-globe-europe::before {\n  content: \"\\f7a2\";\n}\n\n.fa-cart-flatbed-suitcase::before {\n  content: \"\\f59d\";\n}\n\n.fa-luggage-cart::before {\n  content: \"\\f59d\";\n}\n\n.fa-rectangle-xmark::before {\n  content: \"\\f410\";\n}\n\n.fa-rectangle-times::before {\n  content: \"\\f410\";\n}\n\n.fa-times-rectangle::before {\n  content: \"\\f410\";\n}\n\n.fa-window-close::before {\n  content: \"\\f410\";\n}\n\n.fa-baht-sign::before {\n  content: \"\\e0ac\";\n}\n\n.fa-book-open::before {\n  content: \"\\f518\";\n}\n\n.fa-book-journal-whills::before {\n  content: \"\\f66a\";\n}\n\n.fa-journal-whills::before {\n  content: \"\\f66a\";\n}\n\n.fa-handcuffs::before {\n  content: \"\\e4f8\";\n}\n\n.fa-triangle-exclamation::before {\n  content: \"\\f071\";\n}\n\n.fa-exclamation-triangle::before {\n  content: \"\\f071\";\n}\n\n.fa-warning::before {\n  content: \"\\f071\";\n}\n\n.fa-database::before {\n  content: \"\\f1c0\";\n}\n\n.fa-share::before {\n  content: \"\\f064\";\n}\n\n.fa-mail-forward::before {\n  content: \"\\f064\";\n}\n\n.fa-bottle-droplet::before {\n  content: \"\\e4c4\";\n}\n\n.fa-mask-face::before {\n  content: \"\\e1d7\";\n}\n\n.fa-hill-rockslide::before {\n  content: \"\\e508\";\n}\n\n.fa-right-left::before {\n  content: \"\\f362\";\n}\n\n.fa-exchange-alt::before {\n  content: \"\\f362\";\n}\n\n.fa-paper-plane::before {\n  content: \"\\f1d8\";\n}\n\n.fa-road-circle-exclamation::before {\n  content: \"\\e565\";\n}\n\n.fa-dungeon::before {\n  content: \"\\f6d9\";\n}\n\n.fa-align-right::before {\n  content: \"\\f038\";\n}\n\n.fa-money-bill-1-wave::before {\n  content: \"\\f53b\";\n}\n\n.fa-money-bill-wave-alt::before {\n  content: \"\\f53b\";\n}\n\n.fa-life-ring::before {\n  content: \"\\f1cd\";\n}\n\n.fa-hands::before {\n  content: \"\\f2a7\";\n}\n\n.fa-sign-language::before {\n  content: \"\\f2a7\";\n}\n\n.fa-signing::before {\n  content: \"\\f2a7\";\n}\n\n.fa-calendar-day::before {\n  content: \"\\f783\";\n}\n\n.fa-water-ladder::before {\n  content: \"\\f5c5\";\n}\n\n.fa-ladder-water::before {\n  content: \"\\f5c5\";\n}\n\n.fa-swimming-pool::before {\n  content: \"\\f5c5\";\n}\n\n.fa-arrows-up-down::before {\n  content: \"\\f07d\";\n}\n\n.fa-arrows-v::before {\n  content: \"\\f07d\";\n}\n\n.fa-face-grimace::before {\n  content: \"\\f57f\";\n}\n\n.fa-grimace::before {\n  content: \"\\f57f\";\n}\n\n.fa-wheelchair-move::before {\n  content: \"\\e2ce\";\n}\n\n.fa-wheelchair-alt::before {\n  content: \"\\e2ce\";\n}\n\n.fa-turn-down::before {\n  content: \"\\f3be\";\n}\n\n.fa-level-down-alt::before {\n  content: \"\\f3be\";\n}\n\n.fa-person-walking-arrow-right::before {\n  content: \"\\e552\";\n}\n\n.fa-square-envelope::before {\n  content: \"\\f199\";\n}\n\n.fa-envelope-square::before {\n  content: \"\\f199\";\n}\n\n.fa-dice::before {\n  content: \"\\f522\";\n}\n\n.fa-bowling-ball::before {\n  content: \"\\f436\";\n}\n\n.fa-brain::before {\n  content: \"\\f5dc\";\n}\n\n.fa-bandage::before {\n  content: \"\\f462\";\n}\n\n.fa-band-aid::before {\n  content: \"\\f462\";\n}\n\n.fa-calendar-minus::before {\n  content: \"\\f272\";\n}\n\n.fa-circle-xmark::before {\n  content: \"\\f057\";\n}\n\n.fa-times-circle::before {\n  content: \"\\f057\";\n}\n\n.fa-xmark-circle::before {\n  content: \"\\f057\";\n}\n\n.fa-gifts::before {\n  content: \"\\f79c\";\n}\n\n.fa-hotel::before {\n  content: \"\\f594\";\n}\n\n.fa-earth-asia::before {\n  content: \"\\f57e\";\n}\n\n.fa-globe-asia::before {\n  content: \"\\f57e\";\n}\n\n.fa-id-card-clip::before {\n  content: \"\\f47f\";\n}\n\n.fa-id-card-alt::before {\n  content: \"\\f47f\";\n}\n\n.fa-magnifying-glass-plus::before {\n  content: \"\\f00e\";\n}\n\n.fa-search-plus::before {\n  content: \"\\f00e\";\n}\n\n.fa-thumbs-up::before {\n  content: \"\\f164\";\n}\n\n.fa-user-clock::before {\n  content: \"\\f4fd\";\n}\n\n.fa-hand-dots::before {\n  content: \"\\f461\";\n}\n\n.fa-allergies::before {\n  content: \"\\f461\";\n}\n\n.fa-file-invoice::before {\n  content: \"\\f570\";\n}\n\n.fa-window-minimize::before {\n  content: \"\\f2d1\";\n}\n\n.fa-mug-saucer::before {\n  content: \"\\f0f4\";\n}\n\n.fa-coffee::before {\n  content: \"\\f0f4\";\n}\n\n.fa-brush::before {\n  content: \"\\f55d\";\n}\n\n.fa-mask::before {\n  content: \"\\f6fa\";\n}\n\n.fa-magnifying-glass-minus::before {\n  content: \"\\f010\";\n}\n\n.fa-search-minus::before {\n  content: \"\\f010\";\n}\n\n.fa-ruler-vertical::before {\n  content: \"\\f548\";\n}\n\n.fa-user-large::before {\n  content: \"\\f406\";\n}\n\n.fa-user-alt::before {\n  content: \"\\f406\";\n}\n\n.fa-train-tram::before {\n  content: \"\\e5b4\";\n}\n\n.fa-user-nurse::before {\n  content: \"\\f82f\";\n}\n\n.fa-syringe::before {\n  content: \"\\f48e\";\n}\n\n.fa-cloud-sun::before {\n  content: \"\\f6c4\";\n}\n\n.fa-stopwatch-20::before {\n  content: \"\\e06f\";\n}\n\n.fa-square-full::before {\n  content: \"\\f45c\";\n}\n\n.fa-magnet::before {\n  content: \"\\f076\";\n}\n\n.fa-jar::before {\n  content: \"\\e516\";\n}\n\n.fa-note-sticky::before {\n  content: \"\\f249\";\n}\n\n.fa-sticky-note::before {\n  content: \"\\f249\";\n}\n\n.fa-bug-slash::before {\n  content: \"\\e490\";\n}\n\n.fa-arrow-up-from-water-pump::before {\n  content: \"\\e4b6\";\n}\n\n.fa-bone::before {\n  content: \"\\f5d7\";\n}\n\n.fa-user-injured::before {\n  content: \"\\f728\";\n}\n\n.fa-face-sad-tear::before {\n  content: \"\\f5b4\";\n}\n\n.fa-sad-tear::before {\n  content: \"\\f5b4\";\n}\n\n.fa-plane::before {\n  content: \"\\f072\";\n}\n\n.fa-tent-arrows-down::before {\n  content: \"\\e581\";\n}\n\n.fa-exclamation::before {\n  content: \"\\!\";\n}\n\n.fa-arrows-spin::before {\n  content: \"\\e4bb\";\n}\n\n.fa-print::before {\n  content: \"\\f02f\";\n}\n\n.fa-turkish-lira-sign::before {\n  content: \"\\e2bb\";\n}\n\n.fa-try::before {\n  content: \"\\e2bb\";\n}\n\n.fa-turkish-lira::before {\n  content: \"\\e2bb\";\n}\n\n.fa-dollar-sign::before {\n  content: \"\\$\";\n}\n\n.fa-dollar::before {\n  content: \"\\$\";\n}\n\n.fa-usd::before {\n  content: \"\\$\";\n}\n\n.fa-x::before {\n  content: \"X\";\n}\n\n.fa-magnifying-glass-dollar::before {\n  content: \"\\f688\";\n}\n\n.fa-search-dollar::before {\n  content: \"\\f688\";\n}\n\n.fa-users-gear::before {\n  content: \"\\f509\";\n}\n\n.fa-users-cog::before {\n  content: \"\\f509\";\n}\n\n.fa-person-military-pointing::before {\n  content: \"\\e54a\";\n}\n\n.fa-building-columns::before {\n  content: \"\\f19c\";\n}\n\n.fa-bank::before {\n  content: \"\\f19c\";\n}\n\n.fa-institution::before {\n  content: \"\\f19c\";\n}\n\n.fa-museum::before {\n  content: \"\\f19c\";\n}\n\n.fa-university::before {\n  content: \"\\f19c\";\n}\n\n.fa-umbrella::before {\n  content: \"\\f0e9\";\n}\n\n.fa-trowel::before {\n  content: \"\\e589\";\n}\n\n.fa-d::before {\n  content: \"D\";\n}\n\n.fa-stapler::before {\n  content: \"\\e5af\";\n}\n\n.fa-masks-theater::before {\n  content: \"\\f630\";\n}\n\n.fa-theater-masks::before {\n  content: \"\\f630\";\n}\n\n.fa-kip-sign::before {\n  content: \"\\e1c4\";\n}\n\n.fa-hand-point-left::before {\n  content: \"\\f0a5\";\n}\n\n.fa-handshake-simple::before {\n  content: \"\\f4c6\";\n}\n\n.fa-handshake-alt::before {\n  content: \"\\f4c6\";\n}\n\n.fa-jet-fighter::before {\n  content: \"\\f0fb\";\n}\n\n.fa-fighter-jet::before {\n  content: \"\\f0fb\";\n}\n\n.fa-square-share-nodes::before {\n  content: \"\\f1e1\";\n}\n\n.fa-share-alt-square::before {\n  content: \"\\f1e1\";\n}\n\n.fa-barcode::before {\n  content: \"\\f02a\";\n}\n\n.fa-plus-minus::before {\n  content: \"\\e43c\";\n}\n\n.fa-video::before {\n  content: \"\\f03d\";\n}\n\n.fa-video-camera::before {\n  content: \"\\f03d\";\n}\n\n.fa-graduation-cap::before {\n  content: \"\\f19d\";\n}\n\n.fa-mortar-board::before {\n  content: \"\\f19d\";\n}\n\n.fa-hand-holding-medical::before {\n  content: \"\\e05c\";\n}\n\n.fa-person-circle-check::before {\n  content: \"\\e53e\";\n}\n\n.fa-turn-up::before {\n  content: \"\\f3bf\";\n}\n\n.fa-level-up-alt::before {\n  content: \"\\f3bf\";\n}\n\n.sr-only,\n.fa-sr-only {\n  position: absolute;\n  width: 1px;\n  height: 1px;\n  padding: 0;\n  margin: -1px;\n  overflow: hidden;\n  clip: rect(0, 0, 0, 0);\n  white-space: nowrap;\n  border-width: 0;\n}\n\n.sr-only-focusable:not(:focus),\n.fa-sr-only-focusable:not(:focus) {\n  position: absolute;\n  width: 1px;\n  height: 1px;\n  padding: 0;\n  margin: -1px;\n  overflow: hidden;\n  clip: rect(0, 0, 0, 0);\n  white-space: nowrap;\n  border-width: 0;\n}\n\n/*!\n * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com\n * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)\n * Copyright 2024 Fonticons, Inc.\n */\n:root, :host {\n  --fa-style-family-classic: \"Font Awesome 6 Free\";\n  --fa-font-solid: normal 900 1em/1 \"Font Awesome 6 Free\";\n}\n\n@font-face {\n  font-family: \"Font Awesome 6 Free\";\n  font-style: normal;\n  font-weight: 900;\n  font-display: block;\n  src: url(\"../webfonts/fa-solid-900.woff2\") format(\"woff2\"), url(\"../webfonts/fa-solid-900.ttf\") format(\"truetype\");\n}\n.fas,\n.fa-solid {\n  font-weight: 900;\n}\n\n/*!\n * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com\n * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)\n * Copyright 2024 Fonticons, Inc.\n */\n:root, :host {\n  --fa-style-family-classic: \"Font Awesome 6 Free\";\n  --fa-font-regular: normal 400 1em/1 \"Font Awesome 6 Free\";\n}\n\n@font-face {\n  font-family: \"Font Awesome 6 Free\";\n  font-style: normal;\n  font-weight: 400;\n  font-display: block;\n  src: url(\"../webfonts/fa-regular-400.woff2\") format(\"woff2\"), url(\"../webfonts/fa-regular-400.ttf\") format(\"truetype\");\n}\n.far,\n.fa-regular {\n  font-weight: 400;\n}\n\n/*!\n * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com\n * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)\n * Copyright 2024 Fonticons, Inc.\n */\n:root, :host {\n  --fa-style-family-brands: \"Font Awesome 6 Brands\";\n  --fa-font-brands: normal 400 1em/1 \"Font Awesome 6 Brands\";\n}\n\n@font-face {\n  font-family: \"Font Awesome 6 Brands\";\n  font-style: normal;\n  font-weight: 400;\n  font-display: block;\n  src: url(\"../webfonts/fa-brands-400.woff2\") format(\"woff2\"), url(\"../webfonts/fa-brands-400.ttf\") format(\"truetype\");\n}\n.fab,\n.fa-brands {\n  font-weight: 400;\n}\n\n.fa-monero:before {\n  content: \"\\f3d0\";\n}\n\n.fa-hooli:before {\n  content: \"\\f427\";\n}\n\n.fa-yelp:before {\n  content: \"\\f1e9\";\n}\n\n.fa-cc-visa:before {\n  content: \"\\f1f0\";\n}\n\n.fa-lastfm:before {\n  content: \"\\f202\";\n}\n\n.fa-shopware:before {\n  content: \"\\f5b5\";\n}\n\n.fa-creative-commons-nc:before {\n  content: \"\\f4e8\";\n}\n\n.fa-aws:before {\n  content: \"\\f375\";\n}\n\n.fa-redhat:before {\n  content: \"\\f7bc\";\n}\n\n.fa-yoast:before {\n  content: \"\\f2b1\";\n}\n\n.fa-cloudflare:before {\n  content: \"\\e07d\";\n}\n\n.fa-ups:before {\n  content: \"\\f7e0\";\n}\n\n.fa-pixiv:before {\n  content: \"\\e640\";\n}\n\n.fa-wpexplorer:before {\n  content: \"\\f2de\";\n}\n\n.fa-dyalog:before {\n  content: \"\\f399\";\n}\n\n.fa-bity:before {\n  content: \"\\f37a\";\n}\n\n.fa-stackpath:before {\n  content: \"\\f842\";\n}\n\n.fa-buysellads:before {\n  content: \"\\f20d\";\n}\n\n.fa-first-order:before {\n  content: \"\\f2b0\";\n}\n\n.fa-modx:before {\n  content: \"\\f285\";\n}\n\n.fa-guilded:before {\n  content: \"\\e07e\";\n}\n\n.fa-vnv:before {\n  content: \"\\f40b\";\n}\n\n.fa-square-js:before {\n  content: \"\\f3b9\";\n}\n\n.fa-js-square:before {\n  content: \"\\f3b9\";\n}\n\n.fa-microsoft:before {\n  content: \"\\f3ca\";\n}\n\n.fa-qq:before {\n  content: \"\\f1d6\";\n}\n\n.fa-orcid:before {\n  content: \"\\f8d2\";\n}\n\n.fa-java:before {\n  content: \"\\f4e4\";\n}\n\n.fa-invision:before {\n  content: \"\\f7b0\";\n}\n\n.fa-creative-commons-pd-alt:before {\n  content: \"\\f4ed\";\n}\n\n.fa-centercode:before {\n  content: \"\\f380\";\n}\n\n.fa-glide-g:before {\n  content: \"\\f2a6\";\n}\n\n.fa-drupal:before {\n  content: \"\\f1a9\";\n}\n\n.fa-jxl:before {\n  content: \"\\e67b\";\n}\n\n.fa-hire-a-helper:before {\n  content: \"\\f3b0\";\n}\n\n.fa-creative-commons-by:before {\n  content: \"\\f4e7\";\n}\n\n.fa-unity:before {\n  content: \"\\e049\";\n}\n\n.fa-whmcs:before {\n  content: \"\\f40d\";\n}\n\n.fa-rocketchat:before {\n  content: \"\\f3e8\";\n}\n\n.fa-vk:before {\n  content: \"\\f189\";\n}\n\n.fa-untappd:before {\n  content: \"\\f405\";\n}\n\n.fa-mailchimp:before {\n  content: \"\\f59e\";\n}\n\n.fa-css3-alt:before {\n  content: \"\\f38b\";\n}\n\n.fa-square-reddit:before {\n  content: \"\\f1a2\";\n}\n\n.fa-reddit-square:before {\n  content: \"\\f1a2\";\n}\n\n.fa-vimeo-v:before {\n  content: \"\\f27d\";\n}\n\n.fa-contao:before {\n  content: \"\\f26d\";\n}\n\n.fa-square-font-awesome:before {\n  content: \"\\e5ad\";\n}\n\n.fa-deskpro:before {\n  content: \"\\f38f\";\n}\n\n.fa-brave:before {\n  content: \"\\e63c\";\n}\n\n.fa-sistrix:before {\n  content: \"\\f3ee\";\n}\n\n.fa-square-instagram:before {\n  content: \"\\e055\";\n}\n\n.fa-instagram-square:before {\n  content: \"\\e055\";\n}\n\n.fa-battle-net:before {\n  content: \"\\f835\";\n}\n\n.fa-the-red-yeti:before {\n  content: \"\\f69d\";\n}\n\n.fa-square-hacker-news:before {\n  content: \"\\f3af\";\n}\n\n.fa-hacker-news-square:before {\n  content: \"\\f3af\";\n}\n\n.fa-edge:before {\n  content: \"\\f282\";\n}\n\n.fa-threads:before {\n  content: \"\\e618\";\n}\n\n.fa-napster:before {\n  content: \"\\f3d2\";\n}\n\n.fa-square-snapchat:before {\n  content: \"\\f2ad\";\n}\n\n.fa-snapchat-square:before {\n  content: \"\\f2ad\";\n}\n\n.fa-google-plus-g:before {\n  content: \"\\f0d5\";\n}\n\n.fa-artstation:before {\n  content: \"\\f77a\";\n}\n\n.fa-markdown:before {\n  content: \"\\f60f\";\n}\n\n.fa-sourcetree:before {\n  content: \"\\f7d3\";\n}\n\n.fa-google-plus:before {\n  content: \"\\f2b3\";\n}\n\n.fa-diaspora:before {\n  content: \"\\f791\";\n}\n\n.fa-foursquare:before {\n  content: \"\\f180\";\n}\n\n.fa-stack-overflow:before {\n  content: \"\\f16c\";\n}\n\n.fa-github-alt:before {\n  content: \"\\f113\";\n}\n\n.fa-phoenix-squadron:before {\n  content: \"\\f511\";\n}\n\n.fa-pagelines:before {\n  content: \"\\f18c\";\n}\n\n.fa-algolia:before {\n  content: \"\\f36c\";\n}\n\n.fa-red-river:before {\n  content: \"\\f3e3\";\n}\n\n.fa-creative-commons-sa:before {\n  content: \"\\f4ef\";\n}\n\n.fa-safari:before {\n  content: \"\\f267\";\n}\n\n.fa-google:before {\n  content: \"\\f1a0\";\n}\n\n.fa-square-font-awesome-stroke:before {\n  content: \"\\f35c\";\n}\n\n.fa-font-awesome-alt:before {\n  content: \"\\f35c\";\n}\n\n.fa-atlassian:before {\n  content: \"\\f77b\";\n}\n\n.fa-linkedin-in:before {\n  content: \"\\f0e1\";\n}\n\n.fa-digital-ocean:before {\n  content: \"\\f391\";\n}\n\n.fa-nimblr:before {\n  content: \"\\f5a8\";\n}\n\n.fa-chromecast:before {\n  content: \"\\f838\";\n}\n\n.fa-evernote:before {\n  content: \"\\f839\";\n}\n\n.fa-hacker-news:before {\n  content: \"\\f1d4\";\n}\n\n.fa-creative-commons-sampling:before {\n  content: \"\\f4f0\";\n}\n\n.fa-adversal:before {\n  content: \"\\f36a\";\n}\n\n.fa-creative-commons:before {\n  content: \"\\f25e\";\n}\n\n.fa-watchman-monitoring:before {\n  content: \"\\e087\";\n}\n\n.fa-fonticons:before {\n  content: \"\\f280\";\n}\n\n.fa-weixin:before {\n  content: \"\\f1d7\";\n}\n\n.fa-shirtsinbulk:before {\n  content: \"\\f214\";\n}\n\n.fa-codepen:before {\n  content: \"\\f1cb\";\n}\n\n.fa-git-alt:before {\n  content: \"\\f841\";\n}\n\n.fa-lyft:before {\n  content: \"\\f3c3\";\n}\n\n.fa-rev:before {\n  content: \"\\f5b2\";\n}\n\n.fa-windows:before {\n  content: \"\\f17a\";\n}\n\n.fa-wizards-of-the-coast:before {\n  content: \"\\f730\";\n}\n\n.fa-square-viadeo:before {\n  content: \"\\f2aa\";\n}\n\n.fa-viadeo-square:before {\n  content: \"\\f2aa\";\n}\n\n.fa-meetup:before {\n  content: \"\\f2e0\";\n}\n\n.fa-centos:before {\n  content: \"\\f789\";\n}\n\n.fa-adn:before {\n  content: \"\\f170\";\n}\n\n.fa-cloudsmith:before {\n  content: \"\\f384\";\n}\n\n.fa-opensuse:before {\n  content: \"\\e62b\";\n}\n\n.fa-pied-piper-alt:before {\n  content: \"\\f1a8\";\n}\n\n.fa-square-dribbble:before {\n  content: \"\\f397\";\n}\n\n.fa-dribbble-square:before {\n  content: \"\\f397\";\n}\n\n.fa-codiepie:before {\n  content: \"\\f284\";\n}\n\n.fa-node:before {\n  content: \"\\f419\";\n}\n\n.fa-mix:before {\n  content: \"\\f3cb\";\n}\n\n.fa-steam:before {\n  content: \"\\f1b6\";\n}\n\n.fa-cc-apple-pay:before {\n  content: \"\\f416\";\n}\n\n.fa-scribd:before {\n  content: \"\\f28a\";\n}\n\n.fa-debian:before {\n  content: \"\\e60b\";\n}\n\n.fa-openid:before {\n  content: \"\\f19b\";\n}\n\n.fa-instalod:before {\n  content: \"\\e081\";\n}\n\n.fa-expeditedssl:before {\n  content: \"\\f23e\";\n}\n\n.fa-sellcast:before {\n  content: \"\\f2da\";\n}\n\n.fa-square-twitter:before {\n  content: \"\\f081\";\n}\n\n.fa-twitter-square:before {\n  content: \"\\f081\";\n}\n\n.fa-r-project:before {\n  content: \"\\f4f7\";\n}\n\n.fa-delicious:before {\n  content: \"\\f1a5\";\n}\n\n.fa-freebsd:before {\n  content: \"\\f3a4\";\n}\n\n.fa-vuejs:before {\n  content: \"\\f41f\";\n}\n\n.fa-accusoft:before {\n  content: \"\\f369\";\n}\n\n.fa-ioxhost:before {\n  content: \"\\f208\";\n}\n\n.fa-fonticons-fi:before {\n  content: \"\\f3a2\";\n}\n\n.fa-app-store:before {\n  content: \"\\f36f\";\n}\n\n.fa-cc-mastercard:before {\n  content: \"\\f1f1\";\n}\n\n.fa-itunes-note:before {\n  content: \"\\f3b5\";\n}\n\n.fa-golang:before {\n  content: \"\\e40f\";\n}\n\n.fa-kickstarter:before {\n  content: \"\\f3bb\";\n}\n\n.fa-square-kickstarter:before {\n  content: \"\\f3bb\";\n}\n\n.fa-grav:before {\n  content: \"\\f2d6\";\n}\n\n.fa-weibo:before {\n  content: \"\\f18a\";\n}\n\n.fa-uncharted:before {\n  content: \"\\e084\";\n}\n\n.fa-firstdraft:before {\n  content: \"\\f3a1\";\n}\n\n.fa-square-youtube:before {\n  content: \"\\f431\";\n}\n\n.fa-youtube-square:before {\n  content: \"\\f431\";\n}\n\n.fa-wikipedia-w:before {\n  content: \"\\f266\";\n}\n\n.fa-wpressr:before {\n  content: \"\\f3e4\";\n}\n\n.fa-rendact:before {\n  content: \"\\f3e4\";\n}\n\n.fa-angellist:before {\n  content: \"\\f209\";\n}\n\n.fa-galactic-republic:before {\n  content: \"\\f50c\";\n}\n\n.fa-nfc-directional:before {\n  content: \"\\e530\";\n}\n\n.fa-skype:before {\n  content: \"\\f17e\";\n}\n\n.fa-joget:before {\n  content: \"\\f3b7\";\n}\n\n.fa-fedora:before {\n  content: \"\\f798\";\n}\n\n.fa-stripe-s:before {\n  content: \"\\f42a\";\n}\n\n.fa-meta:before {\n  content: \"\\e49b\";\n}\n\n.fa-laravel:before {\n  content: \"\\f3bd\";\n}\n\n.fa-hotjar:before {\n  content: \"\\f3b1\";\n}\n\n.fa-bluetooth-b:before {\n  content: \"\\f294\";\n}\n\n.fa-square-letterboxd:before {\n  content: \"\\e62e\";\n}\n\n.fa-sticker-mule:before {\n  content: \"\\f3f7\";\n}\n\n.fa-creative-commons-zero:before {\n  content: \"\\f4f3\";\n}\n\n.fa-hips:before {\n  content: \"\\f452\";\n}\n\n.fa-behance:before {\n  content: \"\\f1b4\";\n}\n\n.fa-reddit:before {\n  content: \"\\f1a1\";\n}\n\n.fa-discord:before {\n  content: \"\\f392\";\n}\n\n.fa-chrome:before {\n  content: \"\\f268\";\n}\n\n.fa-app-store-ios:before {\n  content: \"\\f370\";\n}\n\n.fa-cc-discover:before {\n  content: \"\\f1f2\";\n}\n\n.fa-wpbeginner:before {\n  content: \"\\f297\";\n}\n\n.fa-confluence:before {\n  content: \"\\f78d\";\n}\n\n.fa-shoelace:before {\n  content: \"\\e60c\";\n}\n\n.fa-mdb:before {\n  content: \"\\f8ca\";\n}\n\n.fa-dochub:before {\n  content: \"\\f394\";\n}\n\n.fa-accessible-icon:before {\n  content: \"\\f368\";\n}\n\n.fa-ebay:before {\n  content: \"\\f4f4\";\n}\n\n.fa-amazon:before {\n  content: \"\\f270\";\n}\n\n.fa-unsplash:before {\n  content: \"\\e07c\";\n}\n\n.fa-yarn:before {\n  content: \"\\f7e3\";\n}\n\n.fa-square-steam:before {\n  content: \"\\f1b7\";\n}\n\n.fa-steam-square:before {\n  content: \"\\f1b7\";\n}\n\n.fa-500px:before {\n  content: \"\\f26e\";\n}\n\n.fa-square-vimeo:before {\n  content: \"\\f194\";\n}\n\n.fa-vimeo-square:before {\n  content: \"\\f194\";\n}\n\n.fa-asymmetrik:before {\n  content: \"\\f372\";\n}\n\n.fa-font-awesome:before {\n  content: \"\\f2b4\";\n}\n\n.fa-font-awesome-flag:before {\n  content: \"\\f2b4\";\n}\n\n.fa-font-awesome-logo-full:before {\n  content: \"\\f2b4\";\n}\n\n.fa-gratipay:before {\n  content: \"\\f184\";\n}\n\n.fa-apple:before {\n  content: \"\\f179\";\n}\n\n.fa-hive:before {\n  content: \"\\e07f\";\n}\n\n.fa-gitkraken:before {\n  content: \"\\f3a6\";\n}\n\n.fa-keybase:before {\n  content: \"\\f4f5\";\n}\n\n.fa-apple-pay:before {\n  content: \"\\f415\";\n}\n\n.fa-padlet:before {\n  content: \"\\e4a0\";\n}\n\n.fa-amazon-pay:before {\n  content: \"\\f42c\";\n}\n\n.fa-square-github:before {\n  content: \"\\f092\";\n}\n\n.fa-github-square:before {\n  content: \"\\f092\";\n}\n\n.fa-stumbleupon:before {\n  content: \"\\f1a4\";\n}\n\n.fa-fedex:before {\n  content: \"\\f797\";\n}\n\n.fa-phoenix-framework:before {\n  content: \"\\f3dc\";\n}\n\n.fa-shopify:before {\n  content: \"\\e057\";\n}\n\n.fa-neos:before {\n  content: \"\\f612\";\n}\n\n.fa-square-threads:before {\n  content: \"\\e619\";\n}\n\n.fa-hackerrank:before {\n  content: \"\\f5f7\";\n}\n\n.fa-researchgate:before {\n  content: \"\\f4f8\";\n}\n\n.fa-swift:before {\n  content: \"\\f8e1\";\n}\n\n.fa-angular:before {\n  content: \"\\f420\";\n}\n\n.fa-speakap:before {\n  content: \"\\f3f3\";\n}\n\n.fa-angrycreative:before {\n  content: \"\\f36e\";\n}\n\n.fa-y-combinator:before {\n  content: \"\\f23b\";\n}\n\n.fa-empire:before {\n  content: \"\\f1d1\";\n}\n\n.fa-envira:before {\n  content: \"\\f299\";\n}\n\n.fa-google-scholar:before {\n  content: \"\\e63b\";\n}\n\n.fa-square-gitlab:before {\n  content: \"\\e5ae\";\n}\n\n.fa-gitlab-square:before {\n  content: \"\\e5ae\";\n}\n\n.fa-studiovinari:before {\n  content: \"\\f3f8\";\n}\n\n.fa-pied-piper:before {\n  content: \"\\f2ae\";\n}\n\n.fa-wordpress:before {\n  content: \"\\f19a\";\n}\n\n.fa-product-hunt:before {\n  content: \"\\f288\";\n}\n\n.fa-firefox:before {\n  content: \"\\f269\";\n}\n\n.fa-linode:before {\n  content: \"\\f2b8\";\n}\n\n.fa-goodreads:before {\n  content: \"\\f3a8\";\n}\n\n.fa-square-odnoklassniki:before {\n  content: \"\\f264\";\n}\n\n.fa-odnoklassniki-square:before {\n  content: \"\\f264\";\n}\n\n.fa-jsfiddle:before {\n  content: \"\\f1cc\";\n}\n\n.fa-sith:before {\n  content: \"\\f512\";\n}\n\n.fa-themeisle:before {\n  content: \"\\f2b2\";\n}\n\n.fa-page4:before {\n  content: \"\\f3d7\";\n}\n\n.fa-hashnode:before {\n  content: \"\\e499\";\n}\n\n.fa-react:before {\n  content: \"\\f41b\";\n}\n\n.fa-cc-paypal:before {\n  content: \"\\f1f4\";\n}\n\n.fa-squarespace:before {\n  content: \"\\f5be\";\n}\n\n.fa-cc-stripe:before {\n  content: \"\\f1f5\";\n}\n\n.fa-creative-commons-share:before {\n  content: \"\\f4f2\";\n}\n\n.fa-bitcoin:before {\n  content: \"\\f379\";\n}\n\n.fa-keycdn:before {\n  content: \"\\f3ba\";\n}\n\n.fa-opera:before {\n  content: \"\\f26a\";\n}\n\n.fa-itch-io:before {\n  content: \"\\f83a\";\n}\n\n.fa-umbraco:before {\n  content: \"\\f8e8\";\n}\n\n.fa-galactic-senate:before {\n  content: \"\\f50d\";\n}\n\n.fa-ubuntu:before {\n  content: \"\\f7df\";\n}\n\n.fa-draft2digital:before {\n  content: \"\\f396\";\n}\n\n.fa-stripe:before {\n  content: \"\\f429\";\n}\n\n.fa-houzz:before {\n  content: \"\\f27c\";\n}\n\n.fa-gg:before {\n  content: \"\\f260\";\n}\n\n.fa-dhl:before {\n  content: \"\\f790\";\n}\n\n.fa-square-pinterest:before {\n  content: \"\\f0d3\";\n}\n\n.fa-pinterest-square:before {\n  content: \"\\f0d3\";\n}\n\n.fa-xing:before {\n  content: \"\\f168\";\n}\n\n.fa-blackberry:before {\n  content: \"\\f37b\";\n}\n\n.fa-creative-commons-pd:before {\n  content: \"\\f4ec\";\n}\n\n.fa-playstation:before {\n  content: \"\\f3df\";\n}\n\n.fa-quinscape:before {\n  content: \"\\f459\";\n}\n\n.fa-less:before {\n  content: \"\\f41d\";\n}\n\n.fa-blogger-b:before {\n  content: \"\\f37d\";\n}\n\n.fa-opencart:before {\n  content: \"\\f23d\";\n}\n\n.fa-vine:before {\n  content: \"\\f1ca\";\n}\n\n.fa-signal-messenger:before {\n  content: \"\\e663\";\n}\n\n.fa-paypal:before {\n  content: \"\\f1ed\";\n}\n\n.fa-gitlab:before {\n  content: \"\\f296\";\n}\n\n.fa-typo3:before {\n  content: \"\\f42b\";\n}\n\n.fa-reddit-alien:before {\n  content: \"\\f281\";\n}\n\n.fa-yahoo:before {\n  content: \"\\f19e\";\n}\n\n.fa-dailymotion:before {\n  content: \"\\e052\";\n}\n\n.fa-affiliatetheme:before {\n  content: \"\\f36b\";\n}\n\n.fa-pied-piper-pp:before {\n  content: \"\\f1a7\";\n}\n\n.fa-bootstrap:before {\n  content: \"\\f836\";\n}\n\n.fa-odnoklassniki:before {\n  content: \"\\f263\";\n}\n\n.fa-nfc-symbol:before {\n  content: \"\\e531\";\n}\n\n.fa-mintbit:before {\n  content: \"\\e62f\";\n}\n\n.fa-ethereum:before {\n  content: \"\\f42e\";\n}\n\n.fa-speaker-deck:before {\n  content: \"\\f83c\";\n}\n\n.fa-creative-commons-nc-eu:before {\n  content: \"\\f4e9\";\n}\n\n.fa-patreon:before {\n  content: \"\\f3d9\";\n}\n\n.fa-avianex:before {\n  content: \"\\f374\";\n}\n\n.fa-ello:before {\n  content: \"\\f5f1\";\n}\n\n.fa-gofore:before {\n  content: \"\\f3a7\";\n}\n\n.fa-bimobject:before {\n  content: \"\\f378\";\n}\n\n.fa-brave-reverse:before {\n  content: \"\\e63d\";\n}\n\n.fa-facebook-f:before {\n  content: \"\\f39e\";\n}\n\n.fa-square-google-plus:before {\n  content: \"\\f0d4\";\n}\n\n.fa-google-plus-square:before {\n  content: \"\\f0d4\";\n}\n\n.fa-web-awesome:before {\n  content: \"\\e682\";\n}\n\n.fa-mandalorian:before {\n  content: \"\\f50f\";\n}\n\n.fa-first-order-alt:before {\n  content: \"\\f50a\";\n}\n\n.fa-osi:before {\n  content: \"\\f41a\";\n}\n\n.fa-google-wallet:before {\n  content: \"\\f1ee\";\n}\n\n.fa-d-and-d-beyond:before {\n  content: \"\\f6ca\";\n}\n\n.fa-periscope:before {\n  content: \"\\f3da\";\n}\n\n.fa-fulcrum:before {\n  content: \"\\f50b\";\n}\n\n.fa-cloudscale:before {\n  content: \"\\f383\";\n}\n\n.fa-forumbee:before {\n  content: \"\\f211\";\n}\n\n.fa-mizuni:before {\n  content: \"\\f3cc\";\n}\n\n.fa-schlix:before {\n  content: \"\\f3ea\";\n}\n\n.fa-square-xing:before {\n  content: \"\\f169\";\n}\n\n.fa-xing-square:before {\n  content: \"\\f169\";\n}\n\n.fa-bandcamp:before {\n  content: \"\\f2d5\";\n}\n\n.fa-wpforms:before {\n  content: \"\\f298\";\n}\n\n.fa-cloudversify:before {\n  content: \"\\f385\";\n}\n\n.fa-usps:before {\n  content: \"\\f7e1\";\n}\n\n.fa-megaport:before {\n  content: \"\\f5a3\";\n}\n\n.fa-magento:before {\n  content: \"\\f3c4\";\n}\n\n.fa-spotify:before {\n  content: \"\\f1bc\";\n}\n\n.fa-optin-monster:before {\n  content: \"\\f23c\";\n}\n\n.fa-fly:before {\n  content: \"\\f417\";\n}\n\n.fa-aviato:before {\n  content: \"\\f421\";\n}\n\n.fa-itunes:before {\n  content: \"\\f3b4\";\n}\n\n.fa-cuttlefish:before {\n  content: \"\\f38c\";\n}\n\n.fa-blogger:before {\n  content: \"\\f37c\";\n}\n\n.fa-flickr:before {\n  content: \"\\f16e\";\n}\n\n.fa-viber:before {\n  content: \"\\f409\";\n}\n\n.fa-soundcloud:before {\n  content: \"\\f1be\";\n}\n\n.fa-digg:before {\n  content: \"\\f1a6\";\n}\n\n.fa-tencent-weibo:before {\n  content: \"\\f1d5\";\n}\n\n.fa-letterboxd:before {\n  content: \"\\e62d\";\n}\n\n.fa-symfony:before {\n  content: \"\\f83d\";\n}\n\n.fa-maxcdn:before {\n  content: \"\\f136\";\n}\n\n.fa-etsy:before {\n  content: \"\\f2d7\";\n}\n\n.fa-facebook-messenger:before {\n  content: \"\\f39f\";\n}\n\n.fa-audible:before {\n  content: \"\\f373\";\n}\n\n.fa-think-peaks:before {\n  content: \"\\f731\";\n}\n\n.fa-bilibili:before {\n  content: \"\\e3d9\";\n}\n\n.fa-erlang:before {\n  content: \"\\f39d\";\n}\n\n.fa-x-twitter:before {\n  content: \"\\e61b\";\n}\n\n.fa-cotton-bureau:before {\n  content: \"\\f89e\";\n}\n\n.fa-dashcube:before {\n  content: \"\\f210\";\n}\n\n.fa-42-group:before {\n  content: \"\\e080\";\n}\n\n.fa-innosoft:before {\n  content: \"\\e080\";\n}\n\n.fa-stack-exchange:before {\n  content: \"\\f18d\";\n}\n\n.fa-elementor:before {\n  content: \"\\f430\";\n}\n\n.fa-square-pied-piper:before {\n  content: \"\\e01e\";\n}\n\n.fa-pied-piper-square:before {\n  content: \"\\e01e\";\n}\n\n.fa-creative-commons-nd:before {\n  content: \"\\f4eb\";\n}\n\n.fa-palfed:before {\n  content: \"\\f3d8\";\n}\n\n.fa-superpowers:before {\n  content: \"\\f2dd\";\n}\n\n.fa-resolving:before {\n  content: \"\\f3e7\";\n}\n\n.fa-xbox:before {\n  content: \"\\f412\";\n}\n\n.fa-square-web-awesome-stroke:before {\n  content: \"\\e684\";\n}\n\n.fa-searchengin:before {\n  content: \"\\f3eb\";\n}\n\n.fa-tiktok:before {\n  content: \"\\e07b\";\n}\n\n.fa-square-facebook:before {\n  content: \"\\f082\";\n}\n\n.fa-facebook-square:before {\n  content: \"\\f082\";\n}\n\n.fa-renren:before {\n  content: \"\\f18b\";\n}\n\n.fa-linux:before {\n  content: \"\\f17c\";\n}\n\n.fa-glide:before {\n  content: \"\\f2a5\";\n}\n\n.fa-linkedin:before {\n  content: \"\\f08c\";\n}\n\n.fa-hubspot:before {\n  content: \"\\f3b2\";\n}\n\n.fa-deploydog:before {\n  content: \"\\f38e\";\n}\n\n.fa-twitch:before {\n  content: \"\\f1e8\";\n}\n\n.fa-ravelry:before {\n  content: \"\\f2d9\";\n}\n\n.fa-mixer:before {\n  content: \"\\e056\";\n}\n\n.fa-square-lastfm:before {\n  content: \"\\f203\";\n}\n\n.fa-lastfm-square:before {\n  content: \"\\f203\";\n}\n\n.fa-vimeo:before {\n  content: \"\\f40a\";\n}\n\n.fa-mendeley:before {\n  content: \"\\f7b3\";\n}\n\n.fa-uniregistry:before {\n  content: \"\\f404\";\n}\n\n.fa-figma:before {\n  content: \"\\f799\";\n}\n\n.fa-creative-commons-remix:before {\n  content: \"\\f4ee\";\n}\n\n.fa-cc-amazon-pay:before {\n  content: \"\\f42d\";\n}\n\n.fa-dropbox:before {\n  content: \"\\f16b\";\n}\n\n.fa-instagram:before {\n  content: \"\\f16d\";\n}\n\n.fa-cmplid:before {\n  content: \"\\e360\";\n}\n\n.fa-upwork:before {\n  content: \"\\e641\";\n}\n\n.fa-facebook:before {\n  content: \"\\f09a\";\n}\n\n.fa-gripfire:before {\n  content: \"\\f3ac\";\n}\n\n.fa-jedi-order:before {\n  content: \"\\f50e\";\n}\n\n.fa-uikit:before {\n  content: \"\\f403\";\n}\n\n.fa-fort-awesome-alt:before {\n  content: \"\\f3a3\";\n}\n\n.fa-phabricator:before {\n  content: \"\\f3db\";\n}\n\n.fa-ussunnah:before {\n  content: \"\\f407\";\n}\n\n.fa-earlybirds:before {\n  content: \"\\f39a\";\n}\n\n.fa-trade-federation:before {\n  content: \"\\f513\";\n}\n\n.fa-autoprefixer:before {\n  content: \"\\f41c\";\n}\n\n.fa-whatsapp:before {\n  content: \"\\f232\";\n}\n\n.fa-square-upwork:before {\n  content: \"\\e67c\";\n}\n\n.fa-slideshare:before {\n  content: \"\\f1e7\";\n}\n\n.fa-google-play:before {\n  content: \"\\f3ab\";\n}\n\n.fa-viadeo:before {\n  content: \"\\f2a9\";\n}\n\n.fa-line:before {\n  content: \"\\f3c0\";\n}\n\n.fa-google-drive:before {\n  content: \"\\f3aa\";\n}\n\n.fa-servicestack:before {\n  content: \"\\f3ec\";\n}\n\n.fa-simplybuilt:before {\n  content: \"\\f215\";\n}\n\n.fa-bitbucket:before {\n  content: \"\\f171\";\n}\n\n.fa-imdb:before {\n  content: \"\\f2d8\";\n}\n\n.fa-deezer:before {\n  content: \"\\e077\";\n}\n\n.fa-raspberry-pi:before {\n  content: \"\\f7bb\";\n}\n\n.fa-jira:before {\n  content: \"\\f7b1\";\n}\n\n.fa-docker:before {\n  content: \"\\f395\";\n}\n\n.fa-screenpal:before {\n  content: \"\\e570\";\n}\n\n.fa-bluetooth:before {\n  content: \"\\f293\";\n}\n\n.fa-gitter:before {\n  content: \"\\f426\";\n}\n\n.fa-d-and-d:before {\n  content: \"\\f38d\";\n}\n\n.fa-microblog:before {\n  content: \"\\e01a\";\n}\n\n.fa-cc-diners-club:before {\n  content: \"\\f24c\";\n}\n\n.fa-gg-circle:before {\n  content: \"\\f261\";\n}\n\n.fa-pied-piper-hat:before {\n  content: \"\\f4e5\";\n}\n\n.fa-kickstarter-k:before {\n  content: \"\\f3bc\";\n}\n\n.fa-yandex:before {\n  content: \"\\f413\";\n}\n\n.fa-readme:before {\n  content: \"\\f4d5\";\n}\n\n.fa-html5:before {\n  content: \"\\f13b\";\n}\n\n.fa-sellsy:before {\n  content: \"\\f213\";\n}\n\n.fa-square-web-awesome:before {\n  content: \"\\e683\";\n}\n\n.fa-sass:before {\n  content: \"\\f41e\";\n}\n\n.fa-wirsindhandwerk:before {\n  content: \"\\e2d0\";\n}\n\n.fa-wsh:before {\n  content: \"\\e2d0\";\n}\n\n.fa-buromobelexperte:before {\n  content: \"\\f37f\";\n}\n\n.fa-salesforce:before {\n  content: \"\\f83b\";\n}\n\n.fa-octopus-deploy:before {\n  content: \"\\e082\";\n}\n\n.fa-medapps:before {\n  content: \"\\f3c6\";\n}\n\n.fa-ns8:before {\n  content: \"\\f3d5\";\n}\n\n.fa-pinterest-p:before {\n  content: \"\\f231\";\n}\n\n.fa-apper:before {\n  content: \"\\f371\";\n}\n\n.fa-fort-awesome:before {\n  content: \"\\f286\";\n}\n\n.fa-waze:before {\n  content: \"\\f83f\";\n}\n\n.fa-bluesky:before {\n  content: \"\\e671\";\n}\n\n.fa-cc-jcb:before {\n  content: \"\\f24b\";\n}\n\n.fa-snapchat:before {\n  content: \"\\f2ab\";\n}\n\n.fa-snapchat-ghost:before {\n  content: \"\\f2ab\";\n}\n\n.fa-fantasy-flight-games:before {\n  content: \"\\f6dc\";\n}\n\n.fa-rust:before {\n  content: \"\\e07a\";\n}\n\n.fa-wix:before {\n  content: \"\\f5cf\";\n}\n\n.fa-square-behance:before {\n  content: \"\\f1b5\";\n}\n\n.fa-behance-square:before {\n  content: \"\\f1b5\";\n}\n\n.fa-supple:before {\n  content: \"\\f3f9\";\n}\n\n.fa-webflow:before {\n  content: \"\\e65c\";\n}\n\n.fa-rebel:before {\n  content: \"\\f1d0\";\n}\n\n.fa-css3:before {\n  content: \"\\f13c\";\n}\n\n.fa-staylinked:before {\n  content: \"\\f3f5\";\n}\n\n.fa-kaggle:before {\n  content: \"\\f5fa\";\n}\n\n.fa-space-awesome:before {\n  content: \"\\e5ac\";\n}\n\n.fa-deviantart:before {\n  content: \"\\f1bd\";\n}\n\n.fa-cpanel:before {\n  content: \"\\f388\";\n}\n\n.fa-goodreads-g:before {\n  content: \"\\f3a9\";\n}\n\n.fa-square-git:before {\n  content: \"\\f1d2\";\n}\n\n.fa-git-square:before {\n  content: \"\\f1d2\";\n}\n\n.fa-square-tumblr:before {\n  content: \"\\f174\";\n}\n\n.fa-tumblr-square:before {\n  content: \"\\f174\";\n}\n\n.fa-trello:before {\n  content: \"\\f181\";\n}\n\n.fa-creative-commons-nc-jp:before {\n  content: \"\\f4ea\";\n}\n\n.fa-get-pocket:before {\n  content: \"\\f265\";\n}\n\n.fa-perbyte:before {\n  content: \"\\e083\";\n}\n\n.fa-grunt:before {\n  content: \"\\f3ad\";\n}\n\n.fa-weebly:before {\n  content: \"\\f5cc\";\n}\n\n.fa-connectdevelop:before {\n  content: \"\\f20e\";\n}\n\n.fa-leanpub:before {\n  content: \"\\f212\";\n}\n\n.fa-black-tie:before {\n  content: \"\\f27e\";\n}\n\n.fa-themeco:before {\n  content: \"\\f5c6\";\n}\n\n.fa-python:before {\n  content: \"\\f3e2\";\n}\n\n.fa-android:before {\n  content: \"\\f17b\";\n}\n\n.fa-bots:before {\n  content: \"\\e340\";\n}\n\n.fa-free-code-camp:before {\n  content: \"\\f2c5\";\n}\n\n.fa-hornbill:before {\n  content: \"\\f592\";\n}\n\n.fa-js:before {\n  content: \"\\f3b8\";\n}\n\n.fa-ideal:before {\n  content: \"\\e013\";\n}\n\n.fa-git:before {\n  content: \"\\f1d3\";\n}\n\n.fa-dev:before {\n  content: \"\\f6cc\";\n}\n\n.fa-sketch:before {\n  content: \"\\f7c6\";\n}\n\n.fa-yandex-international:before {\n  content: \"\\f414\";\n}\n\n.fa-cc-amex:before {\n  content: \"\\f1f3\";\n}\n\n.fa-uber:before {\n  content: \"\\f402\";\n}\n\n.fa-github:before {\n  content: \"\\f09b\";\n}\n\n.fa-php:before {\n  content: \"\\f457\";\n}\n\n.fa-alipay:before {\n  content: \"\\f642\";\n}\n\n.fa-youtube:before {\n  content: \"\\f167\";\n}\n\n.fa-skyatlas:before {\n  content: \"\\f216\";\n}\n\n.fa-firefox-browser:before {\n  content: \"\\e007\";\n}\n\n.fa-replyd:before {\n  content: \"\\f3e6\";\n}\n\n.fa-suse:before {\n  content: \"\\f7d6\";\n}\n\n.fa-jenkins:before {\n  content: \"\\f3b6\";\n}\n\n.fa-twitter:before {\n  content: \"\\f099\";\n}\n\n.fa-rockrms:before {\n  content: \"\\f3e9\";\n}\n\n.fa-pinterest:before {\n  content: \"\\f0d2\";\n}\n\n.fa-buffer:before {\n  content: \"\\f837\";\n}\n\n.fa-npm:before {\n  content: \"\\f3d4\";\n}\n\n.fa-yammer:before {\n  content: \"\\f840\";\n}\n\n.fa-btc:before {\n  content: \"\\f15a\";\n}\n\n.fa-dribbble:before {\n  content: \"\\f17d\";\n}\n\n.fa-stumbleupon-circle:before {\n  content: \"\\f1a3\";\n}\n\n.fa-internet-explorer:before {\n  content: \"\\f26b\";\n}\n\n.fa-stubber:before {\n  content: \"\\e5c7\";\n}\n\n.fa-telegram:before {\n  content: \"\\f2c6\";\n}\n\n.fa-telegram-plane:before {\n  content: \"\\f2c6\";\n}\n\n.fa-old-republic:before {\n  content: \"\\f510\";\n}\n\n.fa-odysee:before {\n  content: \"\\e5c6\";\n}\n\n.fa-square-whatsapp:before {\n  content: \"\\f40c\";\n}\n\n.fa-whatsapp-square:before {\n  content: \"\\f40c\";\n}\n\n.fa-node-js:before {\n  content: \"\\f3d3\";\n}\n\n.fa-edge-legacy:before {\n  content: \"\\e078\";\n}\n\n.fa-slack:before {\n  content: \"\\f198\";\n}\n\n.fa-slack-hash:before {\n  content: \"\\f198\";\n}\n\n.fa-medrt:before {\n  content: \"\\f3c8\";\n}\n\n.fa-usb:before {\n  content: \"\\f287\";\n}\n\n.fa-tumblr:before {\n  content: \"\\f173\";\n}\n\n.fa-vaadin:before {\n  content: \"\\f408\";\n}\n\n.fa-quora:before {\n  content: \"\\f2c4\";\n}\n\n.fa-square-x-twitter:before {\n  content: \"\\e61a\";\n}\n\n.fa-reacteurope:before {\n  content: \"\\f75d\";\n}\n\n.fa-medium:before {\n  content: \"\\f23a\";\n}\n\n.fa-medium-m:before {\n  content: \"\\f23a\";\n}\n\n.fa-amilia:before {\n  content: \"\\f36d\";\n}\n\n.fa-mixcloud:before {\n  content: \"\\f289\";\n}\n\n.fa-flipboard:before {\n  content: \"\\f44d\";\n}\n\n.fa-viacoin:before {\n  content: \"\\f237\";\n}\n\n.fa-critical-role:before {\n  content: \"\\f6c9\";\n}\n\n.fa-sitrox:before {\n  content: \"\\e44a\";\n}\n\n.fa-discourse:before {\n  content: \"\\f393\";\n}\n\n.fa-joomla:before {\n  content: \"\\f1aa\";\n}\n\n.fa-mastodon:before {\n  content: \"\\f4f6\";\n}\n\n.fa-airbnb:before {\n  content: \"\\f834\";\n}\n\n.fa-wolf-pack-battalion:before {\n  content: \"\\f514\";\n}\n\n.fa-buy-n-large:before {\n  content: \"\\f8a6\";\n}\n\n.fa-gulp:before {\n  content: \"\\f3ae\";\n}\n\n.fa-creative-commons-sampling-plus:before {\n  content: \"\\f4f1\";\n}\n\n.fa-strava:before {\n  content: \"\\f428\";\n}\n\n.fa-ember:before {\n  content: \"\\f423\";\n}\n\n.fa-canadian-maple-leaf:before {\n  content: \"\\f785\";\n}\n\n.fa-teamspeak:before {\n  content: \"\\f4f9\";\n}\n\n.fa-pushed:before {\n  content: \"\\f3e1\";\n}\n\n.fa-wordpress-simple:before {\n  content: \"\\f411\";\n}\n\n.fa-nutritionix:before {\n  content: \"\\f3d6\";\n}\n\n.fa-wodu:before {\n  content: \"\\e088\";\n}\n\n.fa-google-pay:before {\n  content: \"\\e079\";\n}\n\n.fa-intercom:before {\n  content: \"\\f7af\";\n}\n\n.fa-zhihu:before {\n  content: \"\\f63f\";\n}\n\n.fa-korvue:before {\n  content: \"\\f42f\";\n}\n\n.fa-pix:before {\n  content: \"\\e43a\";\n}\n\n.fa-steam-symbol:before {\n  content: \"\\f3f6\";\n}\n\nhtml {\n  /*****************************************************************************\n  * Overall Layout Variables\n  */\n  --pst-header-height: 4rem;\n  --pst-header-article-height: calc(var(--pst-header-height) * 2 / 3);\n  --pst-sidebar-secondary: 17rem;\n}\n\n/*******************************************************************************\n* Breakpoints that trigger UI changes\n*\n* Note that media-breakpoint-down begins at the next highest level!\n* So we should choose a media-breakpoint-down one *lower* than when we want to start\n* example: media-breakpoint-up(md) and media-breakpoint-down(sm) trigger at the same time\n* ref: https://github.com/twbs/bootstrap/issues/31214\n*/\n/*******************************************************************************\n* Define the animation behaviour\n*/\n/*******************************************************************************\n* UI shaping and padding\n*/\nhtml {\n  /*****************************************************************************\n  * Font features used in this theme\n  */\n  --pst-font-size-base: 1rem;\n  --pst-font-size-h1: 2.625rem;\n  --pst-font-size-h2: 2.125rem;\n  --pst-font-size-h3: 1.75rem;\n  --pst-font-size-h4: 1.5rem;\n  --pst-font-size-h5: 1.25rem;\n  --pst-font-size-h6: 1rem;\n  --pst-font-size-milli: 0.9rem;\n  --pst-sidebar-font-size: 0.9rem;\n  --pst-sidebar-font-size-mobile: 1.1rem;\n  --pst-sidebar-header-font-size: 1.2rem;\n  --pst-sidebar-header-font-weight: 600;\n  --pst-admonition-font-weight-heading: 600;\n  --pst-font-weight-caption: 300;\n  --pst-font-weight-heading: 600;\n  --pst-font-family-base-system: -apple-system, \"BlinkMacSystemFont\", \"Segoe UI\",\n    \"Helvetica Neue\", \"Arial\", sans-serif, \"Apple Color Emoji\", \"Segoe UI Emoji\",\n    \"Segoe UI Symbol\";\n  --pst-font-family-monospace-system: \"SFMono-Regular\", \"Menlo\", \"Consolas\",\n    \"Monaco\", \"Liberation Mono\", \"Lucida Console\", monospace;\n  --pst-font-family-base: var(--pst-font-family-base-system);\n  --pst-font-family-heading: var(--pst-font-family-base-system);\n  --pst-font-family-monospace: var(--pst-font-family-monospace-system);\n}\n\nhtml {\n  /*****************************************************************************\n  * Icon\n  */\n  --pst-font-size-icon: 1.5rem;\n  --pst-icon-check-circle: \"\";\n  --pst-icon-info-circle: \"\";\n  --pst-icon-exclamation-triangle: \"\";\n  --pst-icon-exclamation-circle: \"\";\n  --pst-icon-times-circle: \"\";\n  --pst-icon-lightbulb: \"\";\n  --pst-icon-download: \"\";\n  --pst-icon-angle-left: \"\";\n  --pst-icon-angle-right: \"\";\n  --pst-icon-external-link: \"\";\n  --pst-icon-search-minus: \"\";\n  --pst-icon-github: \"\";\n  --pst-icon-gitlab: \"\";\n  --pst-icon-share: \"\";\n  --pst-icon-bell: \"\";\n  --pst-icon-pencil: \"\";\n  --pst-breadcrumb-divider: \"\";\n}\n\nhtml {\n  /*****************************************************************************\n  * Admonitions\n  **/\n  --pst-icon-admonition-default: var(--pst-icon-bell);\n  --pst-icon-admonition-note: var(--pst-icon-info-circle);\n  --pst-icon-admonition-attention: var(--pst-icon-exclamation-circle);\n  --pst-icon-admonition-caution: var(--pst-icon-exclamation-triangle);\n  --pst-icon-admonition-warning: var(--pst-icon-exclamation-triangle);\n  --pst-icon-admonition-danger: var(--pst-icon-exclamation-triangle);\n  --pst-icon-admonition-error: var(--pst-icon-times-circle);\n  --pst-icon-admonition-hint: var(--pst-icon-lightbulb);\n  --pst-icon-admonition-tip: var(--pst-icon-lightbulb);\n  --pst-icon-admonition-important: var(--pst-icon-exclamation-circle);\n  --pst-icon-admonition-seealso: var(--pst-icon-share);\n  --pst-icon-admonition-todo: var(--pst-icon-pencil);\n}\n\nhtml {\n  /*****************************************************************************\n  * versionmodified\n  **/\n  --pst-icon-versionmodified-default: var(--pst-icon-exclamation-circle);\n  --pst-icon-versionmodified-added: var(--pst-icon-exclamation-circle);\n  --pst-icon-versionmodified-changed: var(--pst-icon-exclamation-circle);\n  --pst-icon-versionmodified-deprecated: var(--pst-icon-exclamation-circle);\n}\n\n/*******************************************************************************\n* master color map. Only the colors that actually differ between light and dark\n* themes are specified separately.\n*\n* To see the full list of colors see https://www.figma.com/file/rUrrHGhUBBIAAjQ82x6pz9/PyData-Design-system---proposal-for-implementation-(2)?node-id=1234%3A765&t=ifcFT1JtnrSshGfi-1\n*/\n/**\n* Function to get items from nested maps\n*/\n/* Assign base colors for the PyData theme */\n:root {\n  --pst-teal-50: #f4fbfc;\n  --pst-teal-100: #e9f6f8;\n  --pst-teal-200: #d0ecf1;\n  --pst-teal-300: #abdde6;\n  --pst-teal-400: #3fb1c5;\n  --pst-teal-500: #0a7d91;\n  --pst-teal-600: #085d6c;\n  --pst-teal-700: #064752;\n  --pst-teal-800: #042c33;\n  --pst-teal-900: #021b1f;\n  --pst-violet-50: #f4eefb;\n  --pst-violet-100: #e0c7ff;\n  --pst-violet-200: #d5b4fd;\n  --pst-violet-300: #b780ff;\n  --pst-violet-400: #9c5ffd;\n  --pst-violet-500: #8045e5;\n  --pst-violet-600: #6432bd;\n  --pst-violet-700: #4b258f;\n  --pst-violet-800: #341a61;\n  --pst-violet-900: #1e0e39;\n  --pst-gray-50: #f9f9fa;\n  --pst-gray-100: #f3f4f5;\n  --pst-gray-200: #e5e7ea;\n  --pst-gray-300: #d1d5da;\n  --pst-gray-400: #9ca4af;\n  --pst-gray-500: #677384;\n  --pst-gray-600: #48566b;\n  --pst-gray-700: #29313d;\n  --pst-gray-800: #222832;\n  --pst-gray-900: #14181e;\n  --pst-pink-50: #fcf8fd;\n  --pst-pink-100: #fcf0fa;\n  --pst-pink-200: #f8dff5;\n  --pst-pink-300: #f3c7ee;\n  --pst-pink-400: #e47fd7;\n  --pst-pink-500: #c132af;\n  --pst-pink-600: #912583;\n  --pst-pink-700: #6e1c64;\n  --pst-pink-800: #46123f;\n  --pst-pink-900: #2b0b27;\n  --pst-foundation-white: #ffffff;\n  --pst-foundation-black: #14181e;\n}\n\n/*******************************************************************************\n* write the color rules for each theme (light/dark)\n*/\n/* NOTE:\n * Mixins enable us to reuse the same definitions for the different modes\n * https://sass-lang.com/documentation/at-rules/mixin\n * something inserts a variable into a CSS selector or property name\n * https://sass-lang.com/documentation/interpolation\n */\n/* Defaults to light mode if data-theme is not set */\nhtml:not([data-theme]) {\n  --pst-color-primary: #0a7d91;\n  --pst-color-primary-bg: #d0ecf1;\n  --pst-color-secondary: #8045e5;\n  --pst-color-secondary-bg: #e0c7ff;\n  --pst-color-accent: #c132af;\n  --pst-color-accent-bg: #f8dff5;\n  --pst-color-info: #276be9;\n  --pst-color-info-bg: #dce7fc;\n  --pst-color-warning: #f66a0a;\n  --pst-color-warning-bg: #f8e3d0;\n  --pst-color-success: #00843f;\n  --pst-color-success-bg: #d6ece1;\n  --pst-color-attention: var(--pst-color-warning);\n  --pst-color-attention-bg: var(--pst-color-warning-bg);\n  --pst-color-danger: #d72d47;\n  --pst-color-danger-bg: #f9e1e4;\n  --pst-color-text-base: #222832;\n  --pst-color-text-muted: #48566b;\n  --pst-color-shadow: rgba(0, 0, 0, 0.1);\n  --pst-color-border: #d1d5da;\n  --pst-color-border-muted: rgba(23, 23, 26, 0.2);\n  --pst-color-blockquote-notch: #677384;\n  --pst-color-inline-code: #912583;\n  --pst-color-inline-code-links: #085d6c;\n  --pst-color-target: #f3cf95;\n  --pst-color-table: #14181e;\n  --pst-color-table-row-hover-bg: #d5b4fd;\n  --pst-color-table-inner-border: #e5e7ea;\n  --pst-color-background: #ffffff;\n  --pst-color-on-background: #ffffff;\n  --pst-color-surface: #f3f4f5;\n  --pst-color-on-surface: #222832;\n}\nhtml:not([data-theme]) {\n  --pst-color-heading: var(--pst-color-text-base);\n  --pst-color-link: var(--pst-color-primary);\n  --pst-color-link-hover: var(--pst-color-secondary);\n  --pst-color-table-outer-border: var(--pst-color-surface);\n  --pst-color-table-heading-bg: var(--pst-color-surface);\n  --pst-color-table-row-zebra-high-bg: var(--pst-color-on-background);\n  --pst-color-table-row-zebra-low-bg: var(--pst-color-surface);\n}\nhtml:not([data-theme]) .only-dark,\nhtml:not([data-theme]) .only-dark ~ figcaption {\n  display: none !important;\n}\n\n/* NOTE: @each {...} is like a for-loop\n * https://sass-lang.com/documentation/at-rules/control/each\n */\nhtml[data-theme=light] {\n  --pst-color-primary: #0a7d91;\n  --pst-color-primary-bg: #d0ecf1;\n  --pst-color-secondary: #8045e5;\n  --pst-color-secondary-bg: #e0c7ff;\n  --pst-color-accent: #c132af;\n  --pst-color-accent-bg: #f8dff5;\n  --pst-color-info: #276be9;\n  --pst-color-info-bg: #dce7fc;\n  --pst-color-warning: #f66a0a;\n  --pst-color-warning-bg: #f8e3d0;\n  --pst-color-success: #00843f;\n  --pst-color-success-bg: #d6ece1;\n  --pst-color-attention: var(--pst-color-warning);\n  --pst-color-attention-bg: var(--pst-color-warning-bg);\n  --pst-color-danger: #d72d47;\n  --pst-color-danger-bg: #f9e1e4;\n  --pst-color-text-base: #222832;\n  --pst-color-text-muted: #48566b;\n  --pst-color-shadow: rgba(0, 0, 0, 0.1);\n  --pst-color-border: #d1d5da;\n  --pst-color-border-muted: rgba(23, 23, 26, 0.2);\n  --pst-color-blockquote-notch: #677384;\n  --pst-color-inline-code: #912583;\n  --pst-color-inline-code-links: #085d6c;\n  --pst-color-target: #f3cf95;\n  --pst-color-table: #14181e;\n  --pst-color-table-row-hover-bg: #d5b4fd;\n  --pst-color-table-inner-border: #e5e7ea;\n  --pst-color-background: #ffffff;\n  --pst-color-on-background: #ffffff;\n  --pst-color-surface: #f3f4f5;\n  --pst-color-on-surface: #222832;\n  color-scheme: light;\n}\nhtml[data-theme=light] {\n  --pst-color-heading: var(--pst-color-text-base);\n  --pst-color-link: var(--pst-color-primary);\n  --pst-color-link-hover: var(--pst-color-secondary);\n  --pst-color-table-outer-border: var(--pst-color-surface);\n  --pst-color-table-heading-bg: var(--pst-color-surface);\n  --pst-color-table-row-zebra-high-bg: var(--pst-color-on-background);\n  --pst-color-table-row-zebra-low-bg: var(--pst-color-surface);\n}\nhtml[data-theme=light] .only-dark,\nhtml[data-theme=light] .only-dark ~ figcaption {\n  display: none !important;\n}\n\nhtml[data-theme=dark] {\n  --pst-color-primary: #3fb1c5;\n  --pst-color-primary-bg: #042c33;\n  --pst-color-secondary: #9c5ffd;\n  --pst-color-secondary-bg: #341a61;\n  --pst-color-accent: #e47fd7;\n  --pst-color-accent-bg: #46123f;\n  --pst-color-info: #79a3f2;\n  --pst-color-info-bg: #06245d;\n  --pst-color-warning: #ff9245;\n  --pst-color-warning-bg: #652a02;\n  --pst-color-success: #5fb488;\n  --pst-color-success-bg: #002f17;\n  --pst-color-attention: var(--pst-color-warning);\n  --pst-color-attention-bg: var(--pst-color-warning-bg);\n  --pst-color-danger: #e78894;\n  --pst-color-danger-bg: #4e111b;\n  --pst-color-text-base: #ced6dd;\n  --pst-color-text-muted: #9ca4af;\n  --pst-color-shadow: rgba(0, 0, 0, 0.2);\n  --pst-color-border: #48566b;\n  --pst-color-border-muted: #29313d;\n  --pst-color-blockquote-notch: #9ca4af;\n  --pst-color-inline-code: #f3c7ee;\n  --pst-color-inline-code-links: #3fb1c5;\n  --pst-color-target: #675c04;\n  --pst-color-table: #ffffff;\n  --pst-color-table-row-hover-bg: #4b258f;\n  --pst-color-table-inner-border: #364150;\n  --pst-color-background: #14181e;\n  --pst-color-on-background: #222832;\n  --pst-color-surface: #29313d;\n  --pst-color-on-surface: #f3f4f5;\n  /* Adjust images in dark mode (unless they have class .only-dark or\n   * .dark-light, in which case assume they're already optimized for dark\n   * mode).\n   */\n  /* Give images a light background in dark mode in case they have\n  *  transparency and black text (unless they have class .only-dark or .dark-light, in\n  *  which case assume they're already optimized for dark mode).\n  */\n  color-scheme: dark;\n}\nhtml[data-theme=dark] {\n  --pst-color-heading: var(--pst-color-text-base);\n  --pst-color-link: var(--pst-color-primary);\n  --pst-color-link-hover: var(--pst-color-secondary);\n  --pst-color-table-outer-border: var(--pst-color-surface);\n  --pst-color-table-heading-bg: var(--pst-color-surface);\n  --pst-color-table-row-zebra-high-bg: var(--pst-color-on-background);\n  --pst-color-table-row-zebra-low-bg: var(--pst-color-surface);\n}\nhtml[data-theme=dark] .only-light,\nhtml[data-theme=dark] .only-light ~ figcaption {\n  display: none !important;\n}\nhtml[data-theme=dark] img:not(.only-dark, .dark-light) {\n  filter: brightness(0.8) contrast(1.2);\n}\nhtml[data-theme=dark] .bd-content img:not(.only-dark, .dark-light) {\n  background-color: rgb(255, 255, 255);\n  border-radius: 0.25rem;\n}\nhtml[data-theme=dark] .MathJax_SVG * {\n  fill: var(--pst-color-text-base);\n}\n\n.pst-color-primary {\n  color: var(--pst-color-primary);\n}\n\n.pst-color-secondary {\n  color: var(--pst-color-secondary);\n}\n\n.pst-color-accent {\n  color: var(--pst-color-accent);\n}\n\n.pst-color-info {\n  color: var(--pst-color-info);\n}\n\n.pst-color-warning {\n  color: var(--pst-color-warning);\n}\n\n.pst-color-success {\n  color: var(--pst-color-success);\n}\n\n.pst-color-attention {\n  color: var(--pst-color-attention);\n}\n\n.pst-color-danger {\n  color: var(--pst-color-danger);\n}\n\n.pst-color-text-base {\n  color: var(--pst-color-text-base);\n}\n\n.pst-color-text-muted {\n  color: var(--pst-color-text-muted);\n}\n\n.pst-color-shadow {\n  color: var(--pst-color-shadow);\n}\n\n.pst-color-border {\n  color: var(--pst-color-border);\n}\n\n.pst-color-border-muted {\n  color: var(--pst-color-border-muted);\n}\n\n.pst-color-blockquote-notch {\n  color: var(--pst-color-blockquote-notch);\n}\n\n.pst-color-inline-code {\n  color: var(--pst-color-inline-code);\n}\n\n.pst-color-inline-code-links {\n  color: var(--pst-color-inline-code-links);\n}\n\n.pst-color-target {\n  color: var(--pst-color-target);\n}\n\n.pst-color-table {\n  color: var(--pst-color-table);\n}\n\n.pst-color-table-row-hover {\n  color: var(--pst-color-table-row-hover);\n}\n\n.pst-color-table-inner-border {\n  color: var(--pst-color-table-inner-border);\n}\n\n.pst-color-background {\n  color: var(--pst-color-background);\n}\n\n.pst-color-on-background {\n  color: var(--pst-color-on-background);\n}\n\n.pst-color-surface {\n  color: var(--pst-color-surface);\n}\n\n.pst-color-on-surface {\n  color: var(--pst-color-on-surface);\n}\n\n/**\n * Miscellaneous color functions and mixins\n**/\n/**\n* Function to get items from nested maps\n*/\n/**\n  * Function to check if the color needs converting to a \"color\" type\n  * if it is a string we cannot use other color manipulation functions\n  * It is used to create the sphinx-design colours as these are often interpolated\n  */\n/**\n  * Function to convert the string representation of a color to a color type (hex)\n  */\n/*********************************************\n* SASS Mixins\n*********************************************/\n/**\n * A consistent box shadow style we apply across elements.\n */\n/**\n   * Set background of some cell outputs to white-ish to make sure colors work\n   * This is because many libraries make output that only looks good on white\n   */\nhtml {\n  font-size: var(--pst-font-size-base);\n  scroll-padding-top: calc(var(--pst-header-height) + 1rem);\n}\n\nbody {\n  background-color: var(--pst-color-background);\n  font-family: var(--pst-font-family-base);\n  font-weight: 400;\n  line-height: 1.65;\n  color: var(--pst-color-text-base);\n  min-height: 100vh;\n  display: flex;\n  flex-direction: column;\n}\nbody::-webkit-scrollbar-track {\n  background-color: var(--pst-color-background);\n}\n\np {\n  margin-bottom: 1.15rem;\n  font-size: 1em;\n  /* section header in docstring pages */\n}\np.rubric {\n  border-bottom: 1px solid var(--pst-color-border);\n}\np.centered {\n  text-align: center;\n}\n\na {\n  word-wrap: break-word;\n  color: var(--pst-color-link);\n  text-decoration: underline;\n  text-decoration-thickness: max(1px, .0625rem);\n  text-underline-offset: 0.1578em;\n}\na:hover {\n  color: var(--pst-color-link-hover);\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n}\na:active {\n  color: var(--pst-color-link);\n}\na:visited {\n  color: var(--pst-color-link);\n}\na:visited:hover {\n  color: var(--pst-color-link-hover);\n}\na.headerlink {\n  color: var(--pst-color-secondary);\n  font-size: 0.8em;\n  padding: 0 4px;\n  margin-left: 0.2em;\n  text-decoration: none;\n  user-select: none;\n}\na.github::before, a.gitlab::before {\n  color: var(--pst-color-text-muted);\n  font: var(--fa-font-brands);\n  margin-right: 0.25rem;\n}\na.github::before {\n  content: var(--pst-icon-github);\n}\na.gitlab::before {\n  content: var(--pst-icon-gitlab);\n}\n\nh6, .h6, h5, .h5, h4, .h4, h3, .h3, h2, .h2, h1, .h1 {\n  margin: 2.75rem 0 1.05rem;\n  font-family: var(--pst-font-family-heading);\n  font-weight: var(--pst-font-weight-heading);\n  line-height: 1.15;\n}\n\nh1, .h1 {\n  margin-top: 0;\n  font-size: var(--pst-font-size-h1);\n  color: var(--pst-color-heading);\n}\n\nh2, .h2 {\n  font-size: var(--pst-font-size-h2);\n  color: var(--pst-color-heading);\n}\n\nh3, .h3 {\n  font-size: var(--pst-font-size-h3);\n  color: var(--pst-color-heading);\n}\n\nh4, .h4 {\n  font-size: var(--pst-font-size-h4);\n  color: var(--pst-color-heading);\n}\n\nh5, .h5 {\n  font-size: var(--pst-font-size-h5);\n  color: var(--pst-color-text-base);\n}\n\nh6, .h6 {\n  font-size: var(--pst-font-size-h6);\n  color: var(--pst-color-text-base);\n}\n\nsmall, .small,\n.text_small {\n  font-size: var(--pst-font-size-milli);\n}\n\nhr {\n  border: 0;\n  border-top: 1px solid var(--pst-color-border);\n}\n\npre,\ncode,\nkbd,\nsamp {\n  font-family: var(--pst-font-family-monospace);\n}\n\nkbd {\n  background-color: var(--pst-color-on-background);\n  color: var(--pst-color-text-muted);\n}\nkbd:not(.compound) {\n  border: 1px solid var(--pst-color-border);\n  margin: 0 0.1rem;\n  padding: 0.1rem 0.4rem;\n  box-shadow: 1px 1px 1px var(--pst-color-shadow);\n}\n\ncode {\n  color: var(--pst-color-inline-code);\n}\n\npre {\n  margin: 1.5em 0;\n  padding: 1rem;\n  background-color: var(--pst-color-surface);\n  color: var(--pst-color-text-base);\n  line-height: 1.2em;\n  border: 1px solid var(--pst-color-border);\n  border-radius: 0.25rem;\n}\npre .linenos {\n  opacity: 0.8;\n  padding-right: 10px;\n}\n\n#pst-back-to-top {\n  z-index: 1080;\n  position: fixed;\n  display: none;\n  top: 90vh;\n  left: 50vw;\n  transform: translate(-50%);\n  color: var(--pst-color-secondary-text);\n  background-color: var(--pst-color-secondary);\n  border: none;\n}\n#pst-back-to-top .fa-arrow-up {\n  margin-inline-end: 0.5em;\n}\n#pst-back-to-top:hover {\n  text-decoration: underline;\n  text-decoration-thickness: max(1px, .0625rem);\n  text-underline-offset: 0.1578em;\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n  color: var(--pst-color-link-hover);\n  text-decoration-thickness: 1px;\n  background-color: var(--pst-color-secondary-highlight);\n  color: var(--pst-color-secondary-highlight-text);\n}\n#pst-back-to-top:focus-visible {\n  box-shadow: none;\n  outline: var(--pst-color-accent) solid 0.1875rem;\n  outline-color: var(--pst-color-secondary);\n  outline-offset: 0.1875rem;\n}\n\n:focus-visible {\n  outline: var(--pst-color-accent) solid 0.1875rem;\n  box-shadow: none;\n}\n\n/***\n * Rules for the UX/UI of skip navigation link btn.\n * It's only visible to people\n * navigating with keyboard for accessibility purposes\n * ref: https://www.youtube.com/watch?v=VUR0I5mqq7I\n ***/\n.skip-link {\n  position: fixed;\n  top: 0;\n  left: 0;\n  right: 0;\n  text-align: center;\n  background-color: var(--pst-color-warning);\n  padding: 0.5rem;\n  z-index: 1055;\n  border-bottom: 1px solid var(--pst-color-border);\n  transform: translateY(-100%);\n  transition: transform 150ms ease-in-out;\n}\n.skip-link:focus-within {\n  transform: translateY(0%);\n}\n.skip-link a {\n  color: var(--pst-color-warning-text) !important;\n}\n.skip-link a:focus-visible {\n  outline-color: #14181e;\n}\n\n.bd-container {\n  flex-grow: 1;\n  display: flex;\n  justify-content: center;\n}\n.bd-container .bd-container__inner {\n  display: flex;\n}\n\n.bd-page-width {\n  width: 100%;\n}\n@media (min-width: 960px) {\n  .bd-page-width {\n    max-width: 88rem;\n  }\n}\n\n.pst-async-banner-revealer {\n  height: 0;\n  overflow: hidden;\n  transition: height 300ms ease-in-out;\n}\n@media (prefers-reduced-motion) {\n  .pst-async-banner-revealer {\n    transition: none;\n  }\n}\n\n#bd-header-version-warning,\n.bd-header-announcement {\n  min-height: 3rem;\n  width: 100%;\n  display: flex;\n  position: relative;\n  align-items: center;\n  justify-content: center;\n  text-align: center;\n  padding: 0.5rem 12.5%;\n}\n@media (max-width: 959.98px) {\n  #bd-header-version-warning,\n  .bd-header-announcement {\n    padding: 0.5rem 2%;\n  }\n}\n#bd-header-version-warning p,\n.bd-header-announcement p {\n  font-weight: bold;\n  margin: 0;\n}\n#bd-header-version-warning a,\n.bd-header-announcement a {\n  color: var(--pst-color-inline-code-links);\n}\n#bd-header-version-warning .pst-button-link-to-stable-version,\n.bd-header-announcement .pst-button-link-to-stable-version {\n  box-shadow: 0 0.2rem 0.5rem var(--pst-color-shadow), 0 0 0.0625rem var(--pst-color-shadow) !important;\n  background-color: var(--pst-color-danger);\n  border-color: var(--pst-color-danger);\n  color: var(--pst-color-danger-text);\n  border-radius: 0.25rem;\n}\n#bd-header-version-warning .pst-button-link-to-stable-version:hover,\n.bd-header-announcement .pst-button-link-to-stable-version:hover {\n  text-decoration: underline;\n  text-decoration-thickness: max(1px, .0625rem);\n  text-underline-offset: 0.1578em;\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n  color: var(--pst-color-link-hover);\n  text-decoration-thickness: 1px;\n  background-color: var(--pst-color-danger-highlight);\n  border-color: var(--pst-color-danger-highlight);\n  color: var(--pst-color-danger-highlight-text);\n}\n#bd-header-version-warning .pst-button-link-to-stable-version:focus-visible,\n.bd-header-announcement .pst-button-link-to-stable-version:focus-visible {\n  outline: 0.1875rem solid var(--pst-color-danger);\n  outline-offset: 0.1875rem;\n}\n\n.bd-header-announcement {\n  background-color: var(--pst-color-secondary-bg);\n}\n\n#bd-header-version-warning {\n  background-color: var(--pst-color-danger-bg);\n}\n\n/**\n * Main content area\n */\n.bd-main {\n  flex-grow: 1;\n  flex-direction: column;\n  display: flex;\n  min-width: 0;\n}\n.bd-main .bd-content {\n  display: flex;\n  justify-content: center;\n  height: 100%;\n}\n.bd-main .bd-content .bd-article-container {\n  justify-content: start;\n  display: flex;\n  flex-direction: column;\n  width: 100%;\n  max-width: 60em;\n  overflow-x: auto;\n  padding: 1rem;\n}\n@media (min-width: 1200px) {\n  .bd-main .bd-content .bd-article-container .bd-article {\n    padding-top: 1.5rem;\n    padding-left: 2rem;\n  }\n}\n\n.bd-footer {\n  width: 100%;\n  border-top: 1px solid var(--pst-color-border);\n}\n.bd-footer .bd-footer__inner {\n  display: flex;\n  flex-grow: 1;\n  padding: 1rem;\n  margin: auto;\n}\n.bd-footer .footer-items__start,\n.bd-footer .footer-items__center,\n.bd-footer .footer-items__end {\n  display: flex;\n  flex-direction: column;\n  gap: 0.5rem;\n  justify-content: center;\n  flex-grow: 1;\n}\n.bd-footer .footer-items__center {\n  text-align: center;\n}\n.bd-footer .footer-items__end {\n  text-align: end;\n}\n.bd-footer .footer-item p {\n  margin-bottom: 0;\n}\n\n.bd-footer-article {\n  margin-top: auto;\n}\n.bd-footer-article .footer-article-items {\n  display: flex;\n  flex-direction: column;\n}\n\n.bd-footer-content .footer-content-items {\n  display: flex;\n  flex-direction: column;\n  margin-top: auto;\n}\n\n/**\n * Header at the top of the page\n * It includes the announcement bar and the navigation bar.\n */\n.bd-header {\n  position: sticky;\n  top: 0;\n  z-index: 1030;\n  background-color: var(--pst-color-on-background) !important;\n  box-shadow: 0 0.125rem 0.25rem 0 var(--pst-color-shadow);\n  width: 100%;\n  padding: 0;\n  max-width: 100vw;\n  justify-content: center;\n}\n.bd-header .bd-header__inner {\n  display: flex;\n  align-items: center;\n  height: fit-content;\n  padding-left: 1rem;\n  padding-right: 1rem;\n}\n.bd-header :focus-visible {\n  border-radius: 0.125rem;\n}\n.bd-header .navbar-item {\n  height: var(--pst-header-height);\n  max-height: var(--pst-header-height);\n  display: flex;\n  align-items: center;\n}\n.bd-header .navbar-header-items {\n  display: none;\n  flex-shrink: 1;\n}\n@media (min-width: 960px) {\n  .bd-header .navbar-header-items {\n    display: inherit;\n    flex-grow: 1;\n    padding: 0 0 0 0.5rem;\n  }\n}\n.bd-header .navbar-header-items__end,\n.bd-header .navbar-header-items__center,\n.bd-header .navbar-header-items__start {\n  display: flex;\n  align-items: center;\n  flex-flow: wrap;\n  row-gap: 0;\n}\n.bd-header .navbar-header-items__end,\n.bd-header .navbar-header-items__center {\n  column-gap: 1.12rem;\n}\n.bd-header .navbar-header-items__start {\n  flex-shrink: 0;\n  margin-right: auto;\n  gap: 0.5rem;\n}\n.bd-header .navbar-header-items__end {\n  justify-content: end;\n}\n.bd-header ul.navbar-nav {\n  display: flex;\n}\n@media (min-width: 960px) {\n  .bd-header ul.navbar-nav {\n    align-items: baseline;\n  }\n}\n.bd-header ul.navbar-nav > li.nav-item {\n  margin-inline: 2px;\n}\n.bd-header ul.navbar-nav > li.nav-item > .nav-link {\n  color: var(--pst-color-text-muted);\n  position: relative;\n  padding-inline: 6px;\n}\n.bd-header ul.navbar-nav > li.nav-item > .nav-link::before {\n  content: \"\";\n  display: block;\n  position: absolute;\n  inset: 0;\n  background-color: transparent;\n  bottom: calc(-1 * max(3px, .1875rem, .12em));\n  margin: max(3px, .1875rem, .12em) 0;\n}\n.bd-header ul.navbar-nav > li.nav-item > .nav-link:hover {\n  color: var(--pst-color-secondary);\n  text-decoration: none;\n}\n.bd-header ul.navbar-nav > li.nav-item > .nav-link:hover::before {\n  border-bottom: max(3px, .1875rem, .12em) solid var(--pst-color-secondary);\n}\n.bd-header ul.navbar-nav > li.nav-item > .nav-link:focus-visible {\n  box-shadow: none;\n  outline: 3px solid var(--pst-color-accent);\n  outline-offset: 0.1875rem;\n}\n.bd-header ul.navbar-nav > li.nav-item.current > .nav-link {\n  color: var(--pst-color-primary);\n}\n.bd-header ul.navbar-nav > li.nav-item.current > .nav-link::before {\n  border-bottom: 3px solid var(--pst-color-primary);\n}\n.bd-header ul.navbar-nav > li.nav-item.dropdown {\n  margin-inline: 4px;\n}\n.bd-header ul.navbar-nav > li.nav-item.dropdown button {\n  padding-inline: 8px;\n}\n.bd-header ul.navbar-nav > li.nav-item.dropdown > .dropdown-toggle {\n  border-radius: 0.125rem;\n  color: var(--pst-color-text-muted);\n}\n.bd-header ul.navbar-nav > li.nav-item.dropdown > .dropdown-toggle:focus-visible {\n  box-shadow: 0 0 0 0.1875rem var(--pst-color-accent);\n}\n.bd-header ul.navbar-nav > li.nav-item.dropdown > .dropdown-toggle:hover {\n  text-decoration: none;\n  box-shadow: 0 0 0 0.1875rem var(--pst-color-link-hover);\n  color: var(--pst-color-text-base);\n}\n.bd-header ul.navbar-nav li a.nav-link.dropdown-item {\n  color: var(--pst-color-text-muted);\n  text-decoration: none;\n}\n.bd-header ul.navbar-nav li a.nav-link.dropdown-item:hover {\n  color: var(--pst-color-link-hover);\n  text-decoration: underline;\n  text-decoration-thickness: max(1px, .0625rem);\n  text-underline-offset: 0.1578em;\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n}\n.bd-header ul.navbar-nav .dropdown button {\n  display: unset;\n  border: none;\n}\n.bd-header ul.navbar-nav .dropdown button:hover {\n  text-decoration: underline;\n  text-decoration-thickness: max(1px, .0625rem);\n  text-underline-offset: 0.1578em;\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n  color: var(--pst-color-link-hover);\n}\n.bd-header ul.navbar-nav .dropdown .dropdown-menu {\n  z-index: 1070;\n  border: 1px solid var(--pst-color-border);\n  box-shadow: 0 0 0.3rem 0.1rem var(--pst-color-shadow);\n  background-color: var(--pst-color-on-background);\n  padding: 0.5rem 0;\n  margin: 0.5rem 0;\n  min-width: 20rem;\n}\n.bd-header ul.navbar-nav .dropdown .dropdown-menu .dropdown-item {\n  padding: 0.25rem 1.5rem;\n}\n.bd-header ul.navbar-nav .dropdown .dropdown-menu .dropdown-item:focus:not(:hover, :active) {\n  background-color: inherit;\n}\n.bd-header ul.navbar-nav .dropdown .dropdown-menu .dropdown-item:focus-visible {\n  z-index: 10;\n}\n.bd-header ul.navbar-nav .dropdown .dropdown-menu:not(.show) {\n  display: none;\n}\n.bd-header button.sidebar-toggle {\n  font-size: var(--pst-font-size-icon);\n  color: var(--pst-color-muted);\n  margin-bottom: 0;\n  background-color: inherit;\n  padding: 0.5rem;\n}\n.bd-header button.primary-toggle {\n  margin-right: 1rem;\n}\n@media (min-width: 960px) {\n  .bd-header button.primary-toggle {\n    display: none;\n  }\n}\n.bd-header button.secondary-toggle {\n  margin-left: 1rem;\n}\n@media (min-width: 1200px) {\n  .bd-header button.secondary-toggle {\n    display: none;\n  }\n}\n\n@media (min-width: 960px) {\n  .navbar-center-items .navbar-item {\n    display: inline-block;\n  }\n}\n.nav-link {\n  transition: none;\n}\n.nav-link:hover {\n  text-decoration: underline;\n  text-decoration-thickness: max(1px, .0625rem);\n  text-underline-offset: 0.1578em;\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n  color: var(--pst-color-link-hover);\n}\n.nav-link.nav-external::after {\n  font: var(--fa-font-solid);\n  content: var(--pst-icon-external-link);\n  font-size: 0.75em;\n  margin-left: 0.3em;\n}\n\n.bd-navbar-elements li.nav-item i {\n  font-size: 0.7rem;\n  padding-left: 2px;\n  vertical-align: middle;\n}\n\n.navbar-persistent--mobile {\n  margin-left: auto;\n}\n@media (min-width: 960px) {\n  .navbar-persistent--mobile {\n    display: none;\n  }\n}\n\n.navbar-persistent--container {\n  display: none;\n}\n@media (min-width: 960px) {\n  .navbar-persistent--container {\n    display: flex;\n  }\n}\n\n.header-article__inner {\n  padding: 0 0.5rem;\n}\n.header-article__inner .header-article-item {\n  min-height: var(--pst-header-article-height);\n}\n.header-article__inner .header-article-items__end {\n  margin-left: auto;\n}\n\n/**\n * The primary sidebar on the left.\n * e.g., between-pages navigation.\n */\n.bd-sidebar-primary {\n  display: flex;\n  flex-direction: column;\n  gap: 1rem;\n  max-height: calc(100vh - var(--pst-header-height));\n  position: sticky;\n  top: var(--pst-header-height);\n  flex: 0 0 auto;\n  width: 25%;\n  padding: 2rem 1rem 1rem 1rem;\n  border-right: 1px solid var(--pst-color-border);\n  background-color: var(--pst-color-background);\n  overflow-y: auto;\n  font-size: var(--pst-sidebar-font-size-mobile);\n}\n@media (min-width: 960px) {\n  .bd-sidebar-primary {\n    font-size: var(--pst-sidebar-font-size);\n  }\n}\n.bd-sidebar-primary :focus-visible {\n  border-radius: 0.125rem;\n}\n.bd-sidebar-primary .nav-link {\n  font-size: var(--pst-sidebar-font-size-mobile);\n}\n.bd-sidebar-primary.no-sidebar {\n  border-right: 0;\n}\n@media (min-width: 960px) {\n  .bd-sidebar-primary.hide-on-wide {\n    display: none;\n  }\n}\n.bd-sidebar-primary h1, .bd-sidebar-primary .h1,\n.bd-sidebar-primary h2,\n.bd-sidebar-primary .h2,\n.bd-sidebar-primary h3,\n.bd-sidebar-primary .h3,\n.bd-sidebar-primary h4,\n.bd-sidebar-primary .h4 {\n  color: var(--pst-color-text-base);\n}\n.bd-sidebar-primary .sidebar-primary-items__start .sidebar-primary-item,\n.bd-sidebar-primary .sidebar-primary-items__end .sidebar-primary-item {\n  padding: 0.5rem 0;\n}\n.bd-sidebar-primary .sidebar-header-items {\n  display: flex;\n  flex-direction: column;\n}\n.bd-sidebar-primary .sidebar-header-items .sidebar-header-items__title {\n  font-weight: var(--pst-sidebar-header-font-weight);\n  font-size: var(--pst-sidebar-header-font-size);\n  color: var(--pst-color-text-base);\n  margin-bottom: 0.5rem;\n}\n.bd-sidebar-primary .sidebar-header-items .bd-navbar-elements .nav-link:focus-visible {\n  box-shadow: none;\n  outline: var(--pst-color-accent) solid 0.1875rem;\n  outline-offset: 0.1875rem;\n}\n.bd-sidebar-primary .sidebar-header-items .sidebar-header-items__center {\n  display: flex;\n  flex-direction: column;\n}\n.bd-sidebar-primary .sidebar-header-items .sidebar-header-items__end {\n  display: flex;\n  align-items: center;\n  gap: 1rem;\n}\n@media (min-width: 960px) {\n  .bd-sidebar-primary .sidebar-header-items {\n    display: none;\n  }\n}\n.bd-sidebar-primary .sidebar-primary-items__start {\n  border-top: 1px solid var(--pst-color-border);\n}\n@media (min-width: 960px) {\n  .bd-sidebar-primary .sidebar-primary-items__start {\n    border-top: none;\n  }\n}\n.bd-sidebar-primary .sidebar-primary-items__end {\n  margin-top: auto;\n  margin-bottom: 1em;\n}\n.bd-sidebar-primary .list-caption {\n  list-style: none;\n  padding-left: 0;\n}\n.bd-sidebar-primary .list-caption li.toctree-l0.has-children > details > summary {\n  position: relative;\n  height: auto;\n  width: auto;\n  display: flex;\n  justify-content: space-between;\n  align-items: baseline;\n}\n.bd-sidebar-primary .list-caption li.toctree-l0.has-children > details > summary .toctree-toggle {\n  flex: 0 0 auto;\n}\n.bd-sidebar-primary .list-caption li.toctree-l0.has-children > details > summary .toctree-toggle .fa-chevron-down {\n  font-size: 1rem;\n}\n.bd-sidebar-primary li.has-children {\n  position: relative;\n}\n.bd-sidebar-primary li.has-children > .reference,\n.bd-sidebar-primary li.has-children .caption {\n  margin-right: calc(30px + 0.1875rem);\n  padding-top: 0.25rem;\n}\n.bd-sidebar-primary li.has-children > details > summary {\n  list-style: none;\n}\n.bd-sidebar-primary li.has-children > details > summary::-webkit-details-marker {\n  display: none;\n}\n.bd-sidebar-primary li.has-children > details > summary:focus-visible {\n  outline: none;\n}\n.bd-sidebar-primary li.has-children > details > summary:focus-visible > .toctree-toggle {\n  outline: var(--pst-color-accent) solid 0.1875rem;\n  outline-offset: -0.1875rem;\n}\n.bd-sidebar-primary li.has-children > details > summary .toctree-toggle {\n  cursor: pointer;\n  position: absolute;\n  top: 0;\n  right: 0;\n  width: 30px;\n  height: 30px;\n  display: inline-flex;\n  justify-content: center;\n  align-items: center;\n}\n.bd-sidebar-primary li.has-children > details > summary .toctree-toggle .fa-chevron-down {\n  font-size: 0.75rem;\n}\n.bd-sidebar-primary li.has-children > details[open] > summary .fa-chevron-down {\n  transform: rotate(180deg);\n}\n\n/* Between-page links and captions */\nnav.bd-links {\n  margin-right: -1rem;\n}\n@media (min-width: 960px) {\n  nav.bd-links {\n    display: block;\n  }\n}\nnav.bd-links ul {\n  display: block;\n  list-style: none;\n}\nnav.bd-links ul ul {\n  padding: 0 0 0 1rem;\n}\nnav.bd-links li > a {\n  display: block;\n  padding: 0.25rem 0.65rem;\n  color: var(--pst-color-text-muted);\n  text-decoration: none;\n  box-shadow: none;\n  margin-right: 0.1875rem;\n}\nnav.bd-links li > a:hover {\n  text-decoration: underline;\n  background-color: transparent;\n  color: var(--pst-color-link-hover);\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n}\nnav.bd-links li > a:active {\n  color: var(--pst-color-link-hover);\n}\nnav.bd-links li > a:focus-visible {\n  box-shadow: 0 0 0 0.1875rem var(--pst-color-accent);\n  outline: none;\n  z-index: 10;\n}\nnav.bd-links li > a.reference.external::after {\n  font: var(--fa-font-solid);\n  content: var(--pst-icon-external-link);\n  font-size: 0.75em;\n  margin-left: 0.3em;\n}\nnav.bd-links .current > a {\n  font-weight: 600;\n  color: var(--pst-color-primary);\n  box-shadow: inset max(3px, .1875rem, .12em) 0 0 var(--pst-color-primary);\n  background-color: transparent;\n}\nnav.bd-links .current > a:focus-visible {\n  box-shadow: inset max(3px, .1875rem, .12em) 0 0 var(--pst-color-primary), 0 0 0 0.1875rem var(--pst-color-accent);\n  outline: none;\n}\nnav.bd-links p.bd-links__title {\n  font-size: var(--pst-sidebar-header-font-size);\n  font-weight: var(--pst-sidebar-header-font-weight);\n  margin-bottom: 0.5rem;\n}\nnav.bd-links p.caption {\n  font-weight: var(--pst-sidebar-header-font-weight);\n  position: relative;\n  margin-top: 1.25rem;\n  margin-bottom: 0.5rem;\n  color: var(--pst-color-text-base);\n  font-size: var(--pst-sidebar-font-size-mobile);\n}\nnav.bd-links p.caption:first-child {\n  margin-top: 0;\n}\n@media (min-width: 960px) {\n  nav.bd-links p.caption {\n    font-size: var(--pst-sidebar-font-size);\n  }\n}\n\n/**\n * Secondary sidebar on the right.\n * e.g., in-page table of contents.\n */\n.bd-sidebar-secondary {\n  display: flex;\n  order: 2;\n  flex-shrink: 0;\n  flex-direction: column;\n  position: sticky;\n  top: var(--pst-header-height);\n  max-height: calc(100vh - var(--pst-header-height));\n  padding: 2rem 1rem 1rem;\n  width: var(--pst-sidebar-secondary);\n  font-size: var(--pst-sidebar-font-size-mobile);\n  background-color: var(--pst-color-background);\n  overflow-y: auto;\n}\n@media (min-width: 1200px) {\n  .bd-sidebar-secondary {\n    font-size: var(--pst-sidebar-font-size);\n  }\n}\n\n.sidebar-secondary-item {\n  padding: 0.5rem;\n}\n@media (min-width: 1200px) {\n  .sidebar-secondary-item {\n    border-left: 1px solid var(--pst-color-border);\n    padding-left: 1rem;\n  }\n}\n.sidebar-secondary-item i {\n  padding-right: 0.5rem;\n}\n\n/*******************************************************************************\n* Rules for the UX/UI of sidebar sliding drawers on mobile\n* Note that this sheet controls styles across many parts of the theme\n* It is aggregated into this one sheet instead of being split across\n* components in order to keep it easier to debug in one place.\n* It is broken up into major sections below.\n*/\n/*******************************************************************************\n* Sidebar drawer behavior\n*/\n/**\n * NOTE: We use this mixin to define the toggle behavior on narrow screens,\n * And the wide-screen behavior of the sections is defined in their own section\n * .scss files.\n */\n.bd-sidebar::backdrop {\n  background-color: black;\n  opacity: 0.5;\n}\n\n@media (max-width: 959.98px) {\n  .bd-sidebar-primary {\n    position: fixed;\n    top: 0;\n    z-index: 1055;\n    height: 100vh;\n    max-height: 100vh;\n    width: 75%;\n    flex-grow: 0.75;\n    max-width: 350px;\n    transition: visibility 200ms ease-out, margin 200ms ease-out;\n    visibility: hidden;\n    border: 0;\n    margin-left: -75%;\n    left: 0;\n  }\n}\n.bd-sidebar-primary[open] {\n  margin-left: 0;\n  visibility: visible;\n}\n\n@media (max-width: 1199.98px) {\n  .bd-sidebar-secondary {\n    position: fixed;\n    top: 0;\n    z-index: 1055;\n    height: 100vh;\n    max-height: 100vh;\n    width: 75%;\n    flex-grow: 0.75;\n    max-width: 350px;\n    transition: visibility 200ms ease-out, margin 200ms ease-out;\n    visibility: hidden;\n    border: 0;\n    margin-right: -75%;\n    right: 0;\n  }\n}\n.bd-sidebar-secondary[open] {\n  margin-right: 0;\n  visibility: visible;\n}\n\n/**\n * Breadcrumbs for parent pages meant for the article header\n */\nul.bd-breadcrumbs {\n  list-style: none;\n  padding-left: 0;\n  display: flex;\n  flex-wrap: wrap;\n  font-size: 0.8rem;\n}\nul.bd-breadcrumbs li.breadcrumb-item {\n  display: flex;\n  align-items: center;\n  white-space: nowrap;\n  overflow-x: hidden;\n  font-weight: bold;\n}\nul.bd-breadcrumbs li.breadcrumb-item a,\nul.bd-breadcrumbs li.breadcrumb-item .ellipsis {\n  overflow-x: hidden;\n  text-overflow: ellipsis;\n  margin: 0.1875rem;\n}\nul.bd-breadcrumbs li.breadcrumb-item a {\n  color: var(--pst-color-text-muted);\n  text-decoration: none;\n}\nul.bd-breadcrumbs li.breadcrumb-item a:hover {\n  color: var(--pst-color-link-hover);\n  text-decoration: underline;\n  text-decoration-thickness: max(1px, .0625rem);\n  text-underline-offset: 0.1578em;\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n}\nul.bd-breadcrumbs li.breadcrumb-item:not(.breadcrumb-home)::before {\n  font: var(--fa-font-solid);\n  font-size: 0.8rem;\n  content: var(--pst-breadcrumb-divider);\n  color: var(--pst-color-text-muted);\n  padding: 0 0.5rem;\n}\n\n/**\n * Icon links in the navbar\n */\nhtml .pst-navbar-icon {\n  box-sizing: border-box;\n  min-width: 24px;\n  min-height: 24px;\n  color: var(--pst-color-text-muted);\n  position: relative;\n  display: flex;\n  align-items: center;\n  justify-content: center;\n  border-radius: 0;\n  border: none;\n  font-size: 1rem;\n  line-height: 1.65;\n  padding: 0.25rem 0;\n  height: 2.15rem;\n}\nhtml .pst-navbar-icon::before {\n  content: \"\";\n  display: block;\n  position: absolute;\n  inset: 0;\n  background-color: transparent;\n  bottom: calc(-1 * max(3px, .1875rem, .12em));\n  margin: max(3px, .1875rem, .12em) 0;\n}\nhtml .pst-navbar-icon:hover {\n  color: var(--pst-color-secondary);\n  text-decoration: none;\n}\nhtml .pst-navbar-icon:hover::before {\n  border-bottom: max(3px, .1875rem, .12em) solid var(--pst-color-secondary);\n}\nhtml .pst-navbar-icon:focus-visible {\n  box-shadow: none;\n  outline: 3px solid var(--pst-color-accent);\n  outline-offset: 0.1875rem;\n}\n\nul.navbar-icon-links {\n  display: flex;\n  flex-flow: row wrap;\n  column-gap: 1.12rem;\n  justify-content: space-evenly;\n  align-items: center;\n  padding-left: 0;\n  margin-bottom: 0;\n  list-style: none;\n}\nul.navbar-icon-links i {\n  /* Social media buttons hard-code the brand color */\n}\nul.navbar-icon-links i.fa-brands, ul.navbar-icon-links i.fa-regular, ul.navbar-icon-links i.fa-solid {\n  vertical-align: middle;\n  font-style: normal;\n  font-size: var(--pst-font-size-icon);\n}\nul.navbar-icon-links i.fa-square-twitter::before {\n  color: #55acee;\n}\nul.navbar-icon-links i.fa-square-gitlab::before {\n  color: #548;\n}\nul.navbar-icon-links i.fa-bitbucket::before {\n  color: #0052cc;\n}\nul.navbar-icon-links img.icon-link-image {\n  height: 1.5em;\n  border-radius: 0.2rem;\n}\nul.navbar-icon-links .fa-pydata {\n  stroke: var(--pst-color-background);\n  stroke-linejoin: round;\n  stroke-width: 0.35;\n}\n\n/**\n * Logo in the navbar\n */\n.navbar-brand {\n  position: relative;\n  height: var(--pst-header-height);\n  max-height: var(--pst-header-height);\n  padding: 0.5rem 0;\n  width: auto;\n  margin: 0;\n  display: flex;\n  flex-shrink: 0;\n  align-items: center;\n  gap: 0.5rem;\n}\n.navbar-brand p {\n  color: var(--pst-color-text-base);\n  margin-bottom: 0;\n}\n.navbar-brand img {\n  max-width: 100%;\n  height: 100%;\n  width: auto;\n}\n.navbar-brand:hover, .navbar-brand:visited:hover {\n  text-decoration: underline;\n  text-decoration-thickness: max(1px, .0625rem);\n  text-underline-offset: 0.1578em;\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n  color: var(--pst-color-link-hover);\n  color: var(--pst-color-text-base);\n}\n\n/**\n * Navigation links in the navbar and icon links\n */\nul.navbar-nav ul {\n  padding: 0 0 0 1rem;\n}\nul.navbar-nav li {\n  display: flex;\n  flex-direction: column;\n}\nul.navbar-nav li a {\n  display: flex;\n  align-items: center;\n  height: 100%;\n  padding-top: 0.25rem;\n  padding-bottom: 0.25rem;\n  color: var(--pst-color-text-muted);\n  text-decoration: none;\n}\nul.navbar-nav li a:hover {\n  color: var(--pst-color-link-hover);\n  text-decoration: underline;\n  text-decoration-thickness: max(1px, .0625rem);\n  text-underline-offset: 0.1578em;\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n}\n\n/**\n * The list of in-page TOC links\n */\n.page-toc .section-nav {\n  padding-left: 0;\n  border-bottom: none;\n}\n.page-toc .section-nav ul {\n  padding-left: 1rem;\n}\n.page-toc .nav-link {\n  font-size: var(--pst-sidebar-font-size-mobile);\n}\n@media (min-width: 1200px) {\n  .page-toc .nav-link {\n    font-size: var(--pst-sidebar-font-size);\n  }\n}\n.page-toc .onthispage {\n  color: var(--pst-color-text-base);\n  font-weight: var(--pst-sidebar-header-font-weight);\n  margin-bottom: 0.5rem;\n}\n\n/**\n* Previous / Next navigation buttons\n**/\n.prev-next-area {\n  width: 100%;\n}\n.prev-next-area p {\n  color: var(--pst-color-text-muted);\n  margin: 0 0.3em;\n  line-height: 1.3em;\n}\n.prev-next-area i {\n  font-size: 1.2em;\n}\n.prev-next-area a {\n  display: flex;\n  align-items: center;\n  border: none;\n  padding: 10px;\n  max-width: 45%;\n  overflow-x: hidden;\n  color: var(--pst-color-text-muted);\n  text-decoration: none;\n}\n.prev-next-area a p.prev-next-title {\n  word-wrap: break-word;\n  color: var(--pst-color-link);\n  text-decoration: underline;\n  text-decoration-thickness: max(1px, .0625rem);\n  text-underline-offset: 0.1578em;\n  font-weight: var(--pst-admonition-font-weight-heading);\n  font-size: 1.1em;\n}\n.prev-next-area a p.prev-next-title:hover {\n  color: var(--pst-color-link-hover);\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n}\n.prev-next-area a p.prev-next-title:active {\n  color: var(--pst-color-link);\n}\n.prev-next-area a p.prev-next-title:visited {\n  color: var(--pst-color-link);\n}\n.prev-next-area a p.prev-next-title:visited:hover {\n  color: var(--pst-color-link-hover);\n}\n.prev-next-area a:hover p.prev-next-title, .prev-next-area a:visited:hover p.prev-next-title {\n  text-decoration: underline;\n  text-decoration-thickness: max(1px, .0625rem);\n  text-underline-offset: 0.1578em;\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n  color: var(--pst-color-link-hover);\n}\n.prev-next-area a .prev-next-info {\n  flex-direction: column;\n  margin: 0 0.5em;\n}\n.prev-next-area a .prev-next-info .prev-next-subtitle {\n  text-transform: capitalize;\n}\n.prev-next-area a.left-prev {\n  float: left;\n}\n.prev-next-area a.right-next {\n  float: right;\n}\n.prev-next-area a.right-next div.prev-next-info {\n  text-align: right;\n}\n\n/**\n * Search field\n **/\n.bd-search {\n  position: relative;\n  padding-left: 0.5rem;\n  gap: 0.5rem;\n  background-color: var(--pst-color-background);\n  border-radius: 0.25rem;\n  border: 1px solid var(--pst-color-border);\n  color: var(--pst-color-text-base);\n}\n.bd-search:active {\n  background-color: var(--pst-color-background);\n  color: var(--pst-color-text-muted);\n}\n.bd-search:focus-within {\n  box-shadow: 0 0 0 0.1875rem var(--pst-color-accent);\n}\n.bd-search:focus-within input:focus {\n  box-shadow: none;\n}\n.bd-search .icon {\n  position: absolute;\n  color: var(--pst-color-border);\n  left: 25px;\n}\n.bd-search .fa-solid.fa-magnifying-glass {\n  position: absolute;\n  left: calc((2.5rem - 0.7em) / 2);\n  color: var(--pst-color-text-muted);\n}\n.bd-search input.form-control {\n  background-color: var(--pst-color-background);\n  color: var(--pst-color-text-base);\n  border: none;\n}\n.bd-search input.form-control::placeholder {\n  color: var(--pst-color-text-muted);\n}\n.bd-search input.form-control::-webkit-search-cancel-button, .bd-search input.form-control::-webkit-search-decoration {\n  appearance: none;\n}\n.bd-search input.form-control:focus, .bd-search input.form-control:focus-visible {\n  color: var(--pst-color-text-muted);\n}\n.bd-search .search-button__kbd-shortcut {\n  display: none;\n  margin-inline-end: 0.5rem;\n  color: var(--pst-color-border);\n}\n@media (min-width: 720px) {\n  .bd-search .search-button__kbd-shortcut {\n    display: flex;\n  }\n}\n\n/**\n * Search button - located in the navbar\n */\n.search-button i {\n  font-size: 1.3rem;\n}\n\n/**\n * The search modal <dialog>\n */\n#pst-search-dialog {\n  display: none;\n}\n#pst-search-dialog[open] {\n  display: flex;\n  position: fixed;\n  z-index: 1055;\n  top: 30%;\n  left: 50%;\n  transform: translate(-50%, -50%);\n  right: 1rem;\n  margin-top: 0.5rem;\n  width: 90%;\n  max-width: 800px;\n  background-color: transparent;\n  padding: 0.1875rem;\n  border: none;\n}\n#pst-search-dialog[open]::backdrop {\n  background-color: black;\n  opacity: 0.5;\n}\n#pst-search-dialog[open] form.bd-search {\n  flex-grow: 1;\n}\n#pst-search-dialog[open] form.bd-search svg,\n#pst-search-dialog[open] form.bd-search input {\n  font-size: var(--pst-font-size-icon);\n}\n\n/**\n * The search button component that looks like a field.\n * Lives at components/search-button-field.html\n */\n.search-button-field {\n  display: inline-flex;\n  align-items: center;\n  border: var(--pst-color-border) solid 1px;\n  border-radius: 1.5em;\n  color: var(--pst-color-text-muted);\n  padding: 0.5em;\n  background-color: var(--pst-color-surface);\n}\n.search-button-field:hover {\n  box-shadow: 0 0 0 0.1875rem var(--pst-color-link-hover);\n}\n.search-button-field:focus-visible {\n  border-radius: 1.5em;\n}\n.search-button-field .search-button__default-text {\n  font-size: var(--bs-nav-link-font-size);\n  font-weight: var(--bs-nav-link-font-weight);\n  margin-right: 0.5em;\n  margin-left: 0.5em;\n}\n.search-button-field .kbd-shortcut__modifier {\n  font-size: 0.75em;\n}\n.search-button-field > * {\n  align-items: center;\n}\n.search-button-field > :not(svg) {\n  display: none;\n}\n@media (min-width: 960px) {\n  .search-button-field > :not(svg) {\n    display: flex;\n  }\n}\n\n/**\n * The 'Hide Search Matches' button.\n * This only shows up when a person lands on a page after clicking a search result.\n * Clicking it removes the highlighting of the search term from the page.\n * We want it to behave like a button.\n */\ndiv#searchbox p.highlight-link {\n  margin: 1rem 0;\n  width: fit-content;\n  box-shadow: 0 0.2rem 0.5rem var(--pst-color-shadow), 0 0 0.0625rem var(--pst-color-shadow) !important;\n}\n@media (min-width: 1200px) {\n  div#searchbox p.highlight-link {\n    margin-left: 2rem;\n  }\n}\ndiv#searchbox p.highlight-link a {\n  border-radius: 0.25rem;\n  font-size: 1.25rem;\n  padding: 0.75rem;\n  background-color: var(--pst-color-primary);\n  color: var(--pst-color-primary-text);\n  text-decoration: none;\n  transition: box-shadow 0.25s ease-out;\n}\ndiv#searchbox p.highlight-link a:hover {\n  box-shadow: inset 0 0 50px 50px rgba(0, 0, 0, 0.25);\n}\ndiv#searchbox p.highlight-link a::before {\n  content: var(--pst-icon-search-minus);\n  color: unset;\n  font: var(--fa-font-solid);\n  margin-right: 0.5rem;\n}\n\n/**\n * Light/dark theme switcher\n */\n.theme-switch-button .theme-switch {\n  display: none;\n}\n.theme-switch-button .theme-switch:active {\n  text-decoration: none;\n  color: var(--pst-color-link-hover);\n}\n.theme-switch-button .theme-switch .fa-lg {\n  aspect-ratio: 1/1;\n}\n\nhtml[data-mode=auto] .theme-switch-button .theme-switch[data-mode=auto] {\n  display: inline;\n}\n\nhtml[data-mode=light] .theme-switch-button .theme-switch[data-mode=light] {\n  display: inline;\n}\n\nhtml[data-mode=dark] .theme-switch-button .theme-switch[data-mode=dark] {\n  display: inline;\n}\n\nbutton.btn.version-switcher__button {\n  border-color: var(--pst-color-border);\n  color: var(--pst-color-text-base);\n  margin-bottom: 1em;\n}\n@media (min-width: 960px) {\n  button.btn.version-switcher__button {\n    margin-bottom: unset;\n  }\n}\nbutton.btn.version-switcher__button:hover {\n  box-shadow: 0 0 0 0.1875rem var(--pst-color-secondary);\n  border-color: transparent;\n}\nbutton.btn.version-switcher__button:active {\n  color: var(--pst-color-text-base);\n  border-color: var(--pst-color-border);\n}\nbutton.btn.version-switcher__button:focus-visible {\n  border-color: transparent;\n}\n\n.version-switcher__menu {\n  border-color: var(--pst-color-border);\n  border-radius: var(--bs-dropdown-border-radius);\n}\n.version-switcher__menu a.list-group-item {\n  background-color: var(--pst-color-on-background);\n  color: var(--pst-color-text-base);\n  padding: 0.75rem 1.25rem;\n}\n.version-switcher__menu a.list-group-item:not(:last-child) {\n  border-bottom: 1px solid var(--pst-color-border);\n}\n.version-switcher__menu a.list-group-item:hover {\n  text-decoration: underline;\n  text-decoration-thickness: max(1px, .0625rem);\n  text-underline-offset: 0.1578em;\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n  color: var(--pst-color-link-hover);\n  background-color: var(--pst-color-surface);\n}\n.version-switcher__menu a.list-group-item.active {\n  font-weight: 600;\n  color: var(--pst-color-primary);\n  box-shadow: inset max(3px, .1875rem, .12em) 0 0 var(--pst-color-primary);\n  position: relative;\n  z-index: 1;\n}\n.version-switcher__menu a.list-group-item.active:focus-visible {\n  box-shadow: inset max(3px, .1875rem, .12em) 0 0 var(--pst-color-primary), 0 0 0 0.1875rem var(--pst-color-accent);\n  outline: none;\n}\n.version-switcher__menu a.list-group-item.active span::before {\n  content: \"\";\n  width: 100%;\n  height: 100%;\n  position: absolute;\n  z-index: -1;\n  left: 0;\n  top: 0;\n}\n.version-switcher__menu a.list-group-item:focus-visible {\n  z-index: 10;\n}\n\nbutton.version-switcher__button,\n.version-switcher__menu {\n  font-size: 1.1em;\n  z-index: 1055;\n  box-sizing: border-box;\n  min-width: 24px;\n  min-height: 24px;\n}\n@media (min-width: 960px) {\n  button.version-switcher__button,\n  .version-switcher__menu {\n    font-size: unset;\n  }\n}\n\n/* Collapsing of the TOC sidebar while scrolling */\n/* Nav: hide second level (shown on .active) */\nnav.page-toc {\n  margin-bottom: 1rem;\n}\n\n.bd-toc .nav .nav {\n  display: none;\n}\n.bd-toc .nav .nav.visible {\n  display: block;\n}\n.bd-toc .nav > .active > ul {\n  display: block;\n}\n\n.toc-entry {\n  display: block;\n}\n.toc-entry a > code {\n  color: var(--pst-color-text-muted);\n}\n.toc-entry a.nav-link {\n  display: block;\n  padding: 0.25rem 0;\n  padding-left: 1rem;\n  margin-left: -1rem;\n  color: var(--pst-color-text-muted);\n  text-decoration: none;\n}\n.toc-entry a.nav-link:hover {\n  text-decoration: underline;\n  background-color: transparent;\n  color: var(--pst-color-link-hover);\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n}\n.toc-entry a.nav-link:active {\n  color: var(--pst-color-link-hover);\n}\n.toc-entry a.nav-link:focus-visible {\n  box-shadow: 0 0 0 0.1875rem var(--pst-color-accent);\n  outline: none;\n  z-index: 10;\n}\n.toc-entry a.nav-link.active {\n  font-weight: 600;\n  color: var(--pst-color-primary);\n  box-shadow: inset max(3px, .1875rem, .12em) 0 0 var(--pst-color-primary);\n  background-color: transparent;\n}\n.toc-entry a.nav-link.active:focus-visible {\n  box-shadow: inset max(3px, .1875rem, .12em) 0 0 var(--pst-color-primary), 0 0 0 0.1875rem var(--pst-color-accent);\n  outline: none;\n}\n.toc-entry a.nav-link.active:hover {\n  color: var(--pst-color-link-hover);\n}\n.toc-entry a.nav-link:focus-visible {\n  border-radius: 0.125rem;\n}\n\ndiv.versionadded,\ndiv.versionchanged,\ndiv.deprecated {\n  vertical-align: middle;\n  margin: 1.5625em auto;\n  padding: 0 0.6rem;\n  overflow: hidden;\n  /* break-inside has replaced page-break-inside and is widely usable since 2019 */\n  page-break-inside: avoid;\n  break-inside: avoid;\n  border-left: 0.2rem solid;\n  border-color: var(--pst-color-info);\n  border-radius: 0.25rem;\n  background-color: var(--pst-color-on-background);\n  box-shadow: 0 0.2rem 0.5rem var(--pst-color-shadow), 0 0 0.0625rem var(--pst-color-shadow) !important;\n  position: relative;\n}\ndiv.versionadded > p,\ndiv.versionchanged > p,\ndiv.deprecated > p {\n  margin-bottom: 0.6rem;\n  margin-top: 0.6rem;\n}\n\ndiv.versionadded {\n  border-color: var(--pst-color-success);\n  background-color: var(--pst-color-success-bg);\n}\n\ndiv.versionchanged {\n  border-color: var(--pst-color-warning);\n  background-color: var(--pst-color-warning-bg);\n}\n\ndiv.deprecated {\n  border-color: var(--pst-color-danger);\n  background-color: var(--pst-color-danger-bg);\n}\n\nspan.versionmodified {\n  font-weight: 600;\n}\nspan.versionmodified::before {\n  margin-right: 0.6rem;\n  color: var(--pst-color-info);\n  font: var(--fa-font-solid);\n  content: var(--pst-icon-versionmodified-default);\n}\n\nspan.versionmodified.added::before {\n  color: var(--pst-color-success);\n  content: var(--pst-icon-versionmodified-added);\n}\n\nspan.versionmodified.changed::before {\n  color: var(--pst-color-warning);\n  content: var(--pst-icon-versionmodified-changed);\n}\n\nspan.versionmodified.deprecated::before {\n  color: var(--pst-color-danger);\n  content: var(--pst-icon-versionmodified-deprecated);\n}\n\n.sidebar-indices-items {\n  display: flex;\n  flex-direction: column;\n  border-top: 1px solid var(--pst-color-border);\n}\n@media (min-width: 960px) {\n  .sidebar-indices-items {\n    border-top: none;\n  }\n}\n.sidebar-indices-items .sidebar-indices-items__title {\n  font-weight: var(--pst-sidebar-header-font-weight);\n  font-size: var(--pst-sidebar-header-font-size);\n  color: var(--pst-color-text-base);\n  margin-bottom: 0.5rem;\n}\n.sidebar-indices-items ul.indices-link {\n  margin-right: -1rem;\n  list-style: none;\n  padding: 0;\n}\n.sidebar-indices-items ul.indices-link li > a {\n  display: block;\n  padding: 0.25rem 0;\n  color: var(--pst-color-text-muted);\n}\n.sidebar-indices-items ul.indices-link li > a:hover {\n  color: var(--pst-color-primary);\n  text-decoration: none;\n  background-color: transparent;\n}\n\n/**\n * Admonitions and blocks of styled content.\n * Admonitions CSS originally inspired by https://squidfunk.github.io/mkdocs-material/getting-started/\n */\ndiv.admonition,\n.admonition {\n  margin: 1.5625em auto;\n  padding: 0 0.6rem 0.8rem;\n  overflow: hidden;\n  /* break-inside has replaced page-break-inside and is widely usable since 2019 */\n  page-break-inside: avoid;\n  break-inside: avoid;\n  border-left: 0.2rem solid;\n  border-color: var(--pst-color-info);\n  border-radius: 0.25rem;\n  background-color: var(--pst-color-on-background);\n  box-shadow: 0 0.2rem 0.5rem var(--pst-color-shadow), 0 0 0.0625rem var(--pst-color-shadow) !important;\n  /**\n   * Special-case for a `sidebar` class that makes the admonition float to\n   * the right like the {sidebar} directive.\n   */\n}\ndiv.admonition *:last-child,\n.admonition *:last-child {\n  margin-bottom: 0;\n}\ndiv.admonition p.admonition-title ~ *,\n.admonition p.admonition-title ~ * {\n  margin-left: 1.4rem;\n  margin-right: 1.4rem;\n}\ndiv.admonition > ol,\ndiv.admonition > ul,\n.admonition > ol,\n.admonition > ul {\n  margin-left: 1em;\n}\ndiv.admonition > .admonition-title,\n.admonition > .admonition-title {\n  margin: 0 -0.6rem;\n  padding: 0.4rem 0.6rem 0.4rem 2rem;\n  font-weight: var(--pst-admonition-font-weight-heading);\n  position: relative;\n  background-color: var(--pst-color-info-bg);\n  z-index: 1;\n}\ndiv.admonition > .admonition-title::before,\n.admonition > .admonition-title::before {\n  content: \"\";\n  width: 100%;\n  height: 100%;\n  position: absolute;\n  left: 0;\n  top: 0;\n  z-index: -1;\n  pointer-events: none;\n}\ndiv.admonition > .admonition-title::after,\n.admonition > .admonition-title::after {\n  position: absolute;\n  left: 0.5rem;\n  width: 1rem;\n  height: 1rem;\n  color: var(--pst-color-info);\n  font: var(--fa-font-solid);\n  line-height: inherit;\n  content: var(--pst-icon-admonition-default);\n  opacity: 1;\n}\ndiv.admonition > .admonition-title + *,\n.admonition > .admonition-title + * {\n  margin-top: 0.4em;\n}\ndiv.admonition.attention,\n.admonition.attention {\n  border-color: var(--pst-color-attention);\n}\ndiv.admonition.attention > .admonition-title,\n.admonition.attention > .admonition-title {\n  background-color: var(--pst-color-attention-bg);\n}\ndiv.admonition.attention > .admonition-title::after,\n.admonition.attention > .admonition-title::after {\n  color: var(--pst-color-attention);\n  content: var(--pst-icon-admonition-attention);\n}\ndiv.admonition.caution,\n.admonition.caution {\n  border-color: var(--pst-color-warning);\n}\ndiv.admonition.caution > .admonition-title,\n.admonition.caution > .admonition-title {\n  background-color: var(--pst-color-warning-bg);\n}\ndiv.admonition.caution > .admonition-title::after,\n.admonition.caution > .admonition-title::after {\n  color: var(--pst-color-warning);\n  content: var(--pst-icon-admonition-caution);\n}\ndiv.admonition.warning,\n.admonition.warning {\n  border-color: var(--pst-color-warning);\n}\ndiv.admonition.warning > .admonition-title,\n.admonition.warning > .admonition-title {\n  background-color: var(--pst-color-warning-bg);\n}\ndiv.admonition.warning > .admonition-title::after,\n.admonition.warning > .admonition-title::after {\n  color: var(--pst-color-warning);\n  content: var(--pst-icon-admonition-warning);\n}\ndiv.admonition.danger,\n.admonition.danger {\n  border-color: var(--pst-color-danger);\n}\ndiv.admonition.danger > .admonition-title,\n.admonition.danger > .admonition-title {\n  background-color: var(--pst-color-danger-bg);\n}\ndiv.admonition.danger > .admonition-title::after,\n.admonition.danger > .admonition-title::after {\n  color: var(--pst-color-danger);\n  content: var(--pst-icon-admonition-danger);\n}\ndiv.admonition.error,\n.admonition.error {\n  border-color: var(--pst-color-danger);\n}\ndiv.admonition.error > .admonition-title,\n.admonition.error > .admonition-title {\n  background-color: var(--pst-color-danger-bg);\n}\ndiv.admonition.error > .admonition-title::after,\n.admonition.error > .admonition-title::after {\n  color: var(--pst-color-danger);\n  content: var(--pst-icon-admonition-error);\n}\ndiv.admonition.hint,\n.admonition.hint {\n  border-color: var(--pst-color-success);\n}\ndiv.admonition.hint > .admonition-title,\n.admonition.hint > .admonition-title {\n  background-color: var(--pst-color-success-bg);\n}\ndiv.admonition.hint > .admonition-title::after,\n.admonition.hint > .admonition-title::after {\n  color: var(--pst-color-success);\n  content: var(--pst-icon-admonition-hint);\n}\ndiv.admonition.tip,\n.admonition.tip {\n  border-color: var(--pst-color-success);\n}\ndiv.admonition.tip > .admonition-title,\n.admonition.tip > .admonition-title {\n  background-color: var(--pst-color-success-bg);\n}\ndiv.admonition.tip > .admonition-title::after,\n.admonition.tip > .admonition-title::after {\n  color: var(--pst-color-success);\n  content: var(--pst-icon-admonition-tip);\n}\ndiv.admonition.important,\n.admonition.important {\n  border-color: var(--pst-color-attention);\n}\ndiv.admonition.important > .admonition-title,\n.admonition.important > .admonition-title {\n  background-color: var(--pst-color-attention-bg);\n}\ndiv.admonition.important > .admonition-title::after,\n.admonition.important > .admonition-title::after {\n  color: var(--pst-color-attention);\n  content: var(--pst-icon-admonition-important);\n}\ndiv.admonition.note,\n.admonition.note {\n  border-color: var(--pst-color-info);\n}\ndiv.admonition.note > .admonition-title,\n.admonition.note > .admonition-title {\n  background-color: var(--pst-color-info-bg);\n}\ndiv.admonition.note > .admonition-title::after,\n.admonition.note > .admonition-title::after {\n  color: var(--pst-color-info);\n  content: var(--pst-icon-admonition-note);\n}\ndiv.admonition.seealso,\n.admonition.seealso {\n  border-color: var(--pst-color-success);\n}\ndiv.admonition.seealso > .admonition-title,\n.admonition.seealso > .admonition-title {\n  background-color: var(--pst-color-success-bg);\n}\ndiv.admonition.seealso > .admonition-title::after,\n.admonition.seealso > .admonition-title::after {\n  color: var(--pst-color-success);\n  content: var(--pst-icon-admonition-seealso);\n}\ndiv.admonition.admonition-todo,\n.admonition.admonition-todo {\n  border-color: var(--pst-color-secondary);\n}\ndiv.admonition.admonition-todo > .admonition-title,\n.admonition.admonition-todo > .admonition-title {\n  background-color: var(--pst-color-secondary-bg);\n}\ndiv.admonition.admonition-todo > .admonition-title::after,\n.admonition.admonition-todo > .admonition-title::after {\n  color: var(--pst-color-secondary);\n  content: var(--pst-icon-admonition-todo);\n}\ndiv.admonition.sidebar,\n.admonition.sidebar {\n  max-width: 40%;\n  float: right;\n  clear: both;\n  margin-left: 0.5rem;\n  margin-top: 0;\n  border-width: 0 0 0 0.2rem;\n}\ndiv.admonition.sidebar.attention, div.admonition.sidebar.important,\n.admonition.sidebar.attention,\n.admonition.sidebar.important {\n  border-color: var(--pst-color-attention);\n}\ndiv.admonition.sidebar.caution, div.admonition.sidebar.warning,\n.admonition.sidebar.caution,\n.admonition.sidebar.warning {\n  border-color: var(--pst-color-warning);\n}\ndiv.admonition.sidebar.danger, div.admonition.sidebar.error,\n.admonition.sidebar.danger,\n.admonition.sidebar.error {\n  border-color: var(--pst-color-danger);\n}\ndiv.admonition.sidebar.hint, div.admonition.sidebar.tip, div.admonition.sidebar.seealso,\n.admonition.sidebar.hint,\n.admonition.sidebar.tip,\n.admonition.sidebar.seealso {\n  border-color: var(--pst-color-success);\n}\ndiv.admonition.sidebar.note, div.admonition.sidebar.todo,\n.admonition.sidebar.note,\n.admonition.sidebar.todo {\n  border-color: var(--pst-color-info);\n}\ndiv.admonition.sidebar p.admonition-title ~ *,\n.admonition.sidebar p.admonition-title ~ * {\n  margin-left: 0;\n  margin-right: 0;\n}\n\n/**************************************************************\n * Similar content blocks that are not technically admonitions.\n */\n/**\n * Topics and the {contents} directive\n */\ndiv.topic,\ndiv.topic.contents,\nnav.contents,\naside.topic {\n  display: flex;\n  flex-direction: column;\n  background-color: var(--pst-color-surface);\n  border-color: var(--pst-color-border);\n  border-radius: 0.25rem;\n  padding: 1rem 1.25rem;\n  box-shadow: 0 0.2rem 0.5rem var(--pst-color-shadow), 0 0 0.0625rem var(--pst-color-shadow) !important;\n}\ndiv.topic .topic-title,\ndiv.topic.contents .topic-title,\nnav.contents .topic-title,\naside.topic .topic-title {\n  margin: 0 0 0.5rem;\n}\ndiv.topic p,\ndiv.topic.contents p,\nnav.contents p,\naside.topic p {\n  color: var(--pst-color-on-surface) !important;\n}\ndiv.topic ul.simple,\ndiv.topic.contents ul.simple,\nnav.contents ul.simple,\naside.topic ul.simple {\n  padding-left: 1rem;\n}\ndiv.topic ul.simple ul,\ndiv.topic.contents ul.simple ul,\nnav.contents ul.simple ul,\naside.topic ul.simple ul {\n  padding-left: 2em;\n}\n\n/**\n * Sidebar directive\n */\naside.sidebar {\n  border: 1px solid var(--pst-color-border);\n  background-color: var(--pst-color-surface);\n  border-radius: 0.25rem;\n  margin-left: 0.5rem;\n  padding: 0;\n}\naside.sidebar > *:last-child {\n  padding-bottom: 1rem;\n}\naside.sidebar p.sidebar-title {\n  position: relative;\n  margin-bottom: 0;\n  padding-top: 0.5rem;\n  padding-bottom: 0.5rem;\n  border-bottom: 1px solid var(--pst-color-border);\n  font-family: var(--pst-font-family-heading);\n  font-weight: var(--pst-admonition-font-weight-heading);\n}\naside.sidebar > *:not(.sidebar-title):first-child,\naside.sidebar > p.sidebar-title + * {\n  margin-top: 1rem;\n}\naside.sidebar > * {\n  padding-left: 1rem;\n  padding-right: 1rem;\n}\n\n/**\n * Rubrics look kind of like section headers\n */\np.rubric {\n  display: flex;\n  flex-direction: column;\n}\n\n/**\n * Seealso is kind of like a vertically-collapsed admonition\n */\n.seealso dd {\n  margin-top: 0;\n  margin-bottom: 0;\n}\n\n/*******************************************************************************\n* Styling for field lists\n*/\n/* grey highlighting of 'parameter' and 'returns' field */\ntable.field-list {\n  border-collapse: separate;\n  border-spacing: 10px;\n  margin-left: 1px;\n  /* italic font for parameter types */\n}\ntable.field-list th.field-name {\n  padding: 1px 8px 1px 5px;\n  white-space: nowrap;\n  background-color: var(--pst-color-surface);\n}\ntable.field-list td.field-body {\n  /* reduced space around parameter description */\n}\ntable.field-list td.field-body p {\n  font-style: italic;\n}\ntable.field-list td.field-body p > strong {\n  font-style: normal;\n}\ntable.field-list td.field-body blockquote {\n  border-left: none;\n  margin: 0 0 0.3em;\n  padding-left: 30px;\n}\n\n/*******************************************************************************\n* Styling for autosummary tables\n*/\n.table.autosummary td:first-child {\n  white-space: nowrap;\n}\n\n/* overriding basic.css to use our own monospace font */\n.sig {\n  font-family: var(--pst-font-family-monospace);\n}\n\n/* C++ specific styling - overriding the basic.css to avoid custom colors */\n.sig-inline.c-texpr,\n.sig-inline.cpp-texpr {\n  font-family: unset;\n}\n\n.sig.c .k,\n.sig.c .kt,\n.sig.cpp .k,\n.sig.cpp .kt {\n  color: var(--pst-color-text-base);\n}\n\n.sig.c .m,\n.sig.cpp .m {\n  color: var(--pst-color-text-base);\n}\n\n.sig.c .s,\n.sig.c .sc,\n.sig.cpp .s,\n.sig.cpp .sc {\n  color: var(--pst-color-text-base);\n}\n\n.sig-name {\n  color: var(--pst-color-inline-code);\n}\n\n.sig-param .o,\n.sig-param .default_value {\n  color: var(--pst-color-text-muted);\n  font-weight: normal;\n}\n\ndt:target,\nspan.highlighted {\n  background-color: var(--pst-color-target);\n}\n\n.viewcode-back {\n  font-family: var(--pst-font-family-base);\n}\n\n.viewcode-block:target {\n  border-top: 1px solid var(--pst-color-border);\n  border-bottom: 1px solid var(--pst-color-border);\n  position: relative;\n  background-color: var(--pst-color-target);\n}\n\ndl > dt > a:has(.viewcode-link) {\n  float: right;\n}\ndl > dt > a:has(.viewcode-link) .viewcode-link {\n  float: none;\n}\n\n/*******************************************************************************\n* Styling for autosummary titles like \"parameters\" and \"returns\"\n*/\ndl[class]:not(.option-list, .field-list, .footnote, .glossary, .simple) {\n  margin-bottom: 3rem;\n}\ndl[class]:not(.option-list, .field-list, .footnote, .glossary, .simple) dd {\n  margin-left: 2rem;\n}\ndl[class]:not(.option-list, .field-list, .footnote, .glossary, .simple) dd > dl.simple > dt {\n  display: flex;\n}\ndl[class]:not(.option-list, .field-list, .footnote, .glossary, .simple) dl.field-list {\n  display: grid;\n  grid-template-columns: unset;\n}\ndl[class]:not(.option-list, .field-list, .footnote, .glossary, .simple) dt.field-odd,\ndl[class]:not(.option-list, .field-list, .footnote, .glossary, .simple) dt.field-even {\n  margin-top: 0.2rem;\n  margin-bottom: 0.1rem;\n  background-color: var(--pst-color-surface);\n}\n\n/**\n * Code block styling\n * Note that we inherit a lot of styling from Bootstrap so not many rules here.\n */\ndiv[class*=highlight-],\ndiv.highlight,\ndiv.literal-block-wrapper {\n  display: flex;\n  flex-direction: column;\n  width: unset;\n  border-radius: 0.25rem;\n  break-inside: avoid;\n}\n\ndiv.literal-block-wrapper {\n  border: 1px solid var(--pst-color-border);\n  border-radius: 0.25rem;\n}\ndiv.literal-block-wrapper div.code-block-caption {\n  margin: 0;\n  border-bottom: 1px solid var(--pst-color-border);\n  padding: 0.5rem;\n  font-size: 1rem;\n  font-weight: var(--pst-font-weight-caption);\n}\ndiv.literal-block-wrapper div.code-block-caption a.headerlink {\n  font-size: inherit;\n}\ndiv.literal-block-wrapper div[class*=highlight-] {\n  margin: 0;\n  border-radius: 0;\n}\ndiv.literal-block-wrapper div[class*=highlight-] pre {\n  border: none;\n  box-shadow: none;\n}\n\n/**\n * In-line code\n */\ncode.literal {\n  padding: 0.1rem 0.25rem;\n  background-color: var(--pst-color-surface);\n  border: 1px solid var(--pst-color-border);\n  border-radius: 0.25rem;\n}\n\na > code {\n  color: var(--pst-color-inline-code-links);\n}\na > code:hover {\n  color: var(--pst-color-link-hover);\n}\n\nspan.linenos {\n  opacity: 0.8 !important;\n}\n\nfigure > a,\nfigure > a > img,\nfigure > img,\nfigure > video {\n  display: block;\n  margin-left: auto;\n  margin-right: auto;\n}\n\nfigure a.headerlink {\n  position: absolute;\n  font-size: inherit;\n}\nfigure:hover a.headerlink {\n  visibility: visible;\n}\nfigure figcaption {\n  font-family: var(--pst-font-family-heading);\n  font-weight: var(--pst-font-weight-caption);\n  color: var(--pst-color-text-muted);\n  margin-left: auto;\n  margin-right: auto;\n  margin-top: 0.3rem;\n  text-align: center;\n}\nfigure figcaption > p:last-child {\n  margin-bottom: 0;\n}\nfigure figcaption p {\n  text-align: start;\n  display: inline-block;\n}\nfigure figcaption table.table {\n  width: fit-content;\n  margin-left: auto;\n  margin-right: auto;\n}\n\ndt.label > span.brackets:not(:only-child)::before {\n  content: \"[\";\n}\n\ndt.label > span.brackets:not(:only-child)::after {\n  content: \"]\";\n}\n\na.footnote-reference {\n  vertical-align: super;\n  font-size: small;\n}\n\naside.footnote {\n  margin-bottom: 0.5rem;\n}\naside.footnote:last-child {\n  margin-bottom: 1rem;\n}\naside.footnote span.label,\naside.footnote span.backrefs {\n  font-weight: bold;\n}\naside.footnote:target {\n  background-color: var(--pst-color-target);\n}\n\n/**\n * Hacky fixes that don't fit cleanly into other sections\n */\ntable.highlighttable td.linenos,\nspan.linenos,\ndiv.doctest > div.highlight span.gp {\n  /* gp: Generic.Prompt */\n  user-select: none;\n}\n\ndd {\n  margin-top: 3px;\n  margin-bottom: 10px;\n  margin-left: 30px;\n}\n\nol,\nul {\n  padding-inline-start: 2rem;\n}\nol li > p:first-child,\nul li > p:first-child {\n  margin-bottom: 0.25rem;\n  margin-top: 0.25rem;\n}\n\nblockquote {\n  padding: 1em;\n  color: var(--pst-color-text-muted);\n  border-left: 0.25em solid var(--pst-color-blockquote-notch);\n  border-radius: 0.25rem;\n  position: relative;\n  background-color: var(--pst-color-surface);\n}\nblockquote p {\n  color: var(--pst-color-text-base);\n}\nblockquote .line-block {\n  margin: 0;\n}\nblockquote p:last-child {\n  margin-bottom: 0;\n}\nblockquote::before {\n  content: \"\";\n  width: 100%;\n  height: 100%;\n  position: absolute;\n  left: 0;\n  top: 0;\n  z-index: -1;\n  pointer-events: none;\n}\nblockquote a {\n  color: var(--pst-color-inline-code-links);\n}\nblockquote::before {\n  z-index: -1;\n}\n\n/**\n * Span-level styling within content\n */\nspan.guilabel {\n  border: 1px solid var(--pst-color-info);\n  font-size: 80%;\n  font-weight: 700;\n  border-radius: 4px;\n  padding: 2.4px 6px;\n  margin: auto 2px;\n  position: relative;\n  background-color: var(--pst-color-info-bg);\n}\nspan.guilabel::before {\n  content: \"\";\n  width: 100%;\n  height: 100%;\n  position: absolute;\n  left: 0;\n  top: 0;\n  z-index: -1;\n  pointer-events: none;\n}\n\na.reference.download::before {\n  content: var(--pst-icon-download);\n  font: var(--fa-font-solid);\n  font-size: 0.8em;\n  padding: 0 0.25em;\n  color: var(--pst-color-text-muted);\n}\n\n/**\n * Tables\n */\ntable {\n  margin-left: auto;\n  margin-right: auto;\n}\ntable.table-right {\n  margin-right: 0;\n}\ntable.table-left {\n  margin-left: 0;\n}\n\ntable caption {\n  text-align: center;\n  caption-side: top;\n  color: var(--pst-color-text-muted);\n}\n\nth.text-left,\ntd.text-left {\n  text-align: left;\n}\nth.text-right,\ntd.text-right {\n  text-align: right;\n}\nth.text-center,\ntd.text-center {\n  text-align: center;\n}\n\n.table {\n  color: var(--pst-color-table);\n  border: 1px solid var(--pst-color-table-outer-border);\n  --bs-table-bg: transparent;\n  --bs-table-color: var(\n    --pst-color-text-base\n  );\n}\n.table th ~ th,\n.table th ~ td,\n.table td ~ th,\n.table td ~ td {\n  border-left: 1px solid var(--pst-color-table-inner-border);\n}\n.table thead tr {\n  background-color: var(--pst-color-table-heading-bg);\n  border-bottom: 2px solid var(--pst-color-primary);\n}\n.table tbody tr:nth-child(odd) {\n  background-color: var(--pst-color-table-row-zebra-low-bg);\n}\n.table tbody tr:nth-child(even) {\n  background-color: var(--pst-color-table-row-zebra-high-bg);\n}\n\n.pst-scrollable-table-container {\n  overflow-x: auto;\n}\n\n/**\n * Style the toctree component in pages (avoid modifying the navbars)\n */\n.toctree-wrapper p.caption {\n  font-size: 1.5em;\n  margin-bottom: 0;\n}\n.toctree-wrapper > ul {\n  padding-left: 0;\n}\n.toctree-wrapper li[class^=toctree-l] {\n  list-style: none;\n  margin-bottom: 0.2em;\n}\n.toctree-wrapper li[class^=toctree-l] > a {\n  list-style: none;\n  font-size: 1.1em;\n}\n.toctree-wrapper li[class^=toctree-l] > ul {\n  list-style: none;\n  padding-inline-start: 1.5em;\n}\n.toctree-wrapper .toctree-l1 > a {\n  font-size: 1.3em;\n}\n\ndiv.topic.contents ul.simple,\nnav.contents ul.simple {\n  list-style: none;\n  padding-left: 0;\n}\n\n/**\n * Mathematics via MathJax.\n *\n * This is designed for MathJax v3\n * ref: https://www.sphinx-doc.org/en/master/usage/extensions/math.html#module-sphinx.ext.mathjax\n */\nspan.math,\ndiv.math {\n  align-items: center;\n  display: flex;\n  max-width: 100%;\n  overflow: hidden;\n}\n\nspan.math {\n  display: inline-flex;\n}\n\ndiv.math {\n  gap: 0.5em;\n  flex-direction: row-reverse;\n}\ndiv.math span.eqno a.headerlink {\n  position: relative;\n  font-size: 1em;\n}\ndiv.math mjx-container {\n  flex-grow: 1;\n  padding-bottom: 0.2rem;\n  overflow: auto;\n}\ndiv.math mjx-container mjx-assistive-mml {\n  height: 0;\n}\n\n/**\n * ABlog\n * ref: https://ablog.readthedocs.io/\n */\n/**\n * Sidebar template components\n */\n.ablog-sidebar-item h2, .ablog-sidebar-item .h2,\n.ablog-sidebar-item h3,\n.ablog-sidebar-item .h3 {\n  font-size: var(--pst-sidebar-header-font-size);\n  margin-top: 0.5rem;\n}\n.ablog-sidebar-item h2 a, .ablog-sidebar-item .h2 a,\n.ablog-sidebar-item h3 a,\n.ablog-sidebar-item .h3 a {\n  color: var(--pst-color-text-base);\n}\n.ablog-sidebar-item ul {\n  list-style: none;\n  padding-left: 0;\n  overflow-y: hidden;\n  display: flex;\n  flex-direction: column;\n  gap: 0.5em;\n  margin-bottom: 0;\n}\n.ablog-sidebar-item ul.ablog-cloud {\n  flex-flow: row wrap;\n  gap: 0.5rem;\n}\n.ablog-sidebar-item ul.ablog-cloud li {\n  display: flex;\n  align-items: center;\n}\n\n/**\n * Previous / next buttons at the bottom\n */\n.ablog__prev-next {\n  font-size: 1.2em;\n  display: flex;\n  padding: 1rem 0;\n}\n.ablog__prev-next > span {\n  display: flex;\n  max-width: 45%;\n}\n.ablog__prev-next > span a {\n  display: flex;\n  align-items: center;\n  margin-left: auto;\n  gap: 1rem;\n  line-height: 1.5rem;\n}\n.ablog__prev-next > span a i::before {\n  color: var(--pst-color-text-base);\n}\n.ablog__prev-next span.ablog__prev i.fa-arrow-circle-left::before {\n  content: var(--pst-icon-angle-left);\n}\n.ablog__prev-next span.ablog__spacer {\n  display: none;\n}\n.ablog__prev-next span.ablog__next {\n  margin-left: auto;\n  text-align: right;\n}\n.ablog__prev-next span.ablog__next i.fa-arrow-circle-right::before {\n  content: var(--pst-icon-angle-right);\n}\n\n/**\n * {postlist} directive and posts page\n */\n.ablog__collection,\n.postlist {\n  padding-left: 0;\n}\n.ablog__collection .ablog-post,\n.postlist .ablog-post {\n  list-style: none;\n}\n.ablog__collection .ablog-post .ablog-archive,\n.postlist .ablog-post .ablog-archive {\n  display: flex;\n  flex-flow: row wrap;\n  gap: 1rem;\n  list-style: none;\n  font-size: 0.75rem;\n  padding-left: 0;\n}\n.ablog__collection .ablog-post .ablog-post-title,\n.postlist .ablog-post .ablog-post-title {\n  margin-top: 0;\n  font-size: 1.25rem;\n}\n.ablog__collection .ablog-post .ablog-post-title a,\n.postlist .ablog-post .ablog-post-title a {\n  font-weight: bold;\n}\n.ablog__collection .ablog-post .ablog-post-expand,\n.postlist .ablog-post .ablog-post-expand {\n  margin-bottom: 0.5rem;\n}\n\n/**\n * Special cases for Bootstrap functionality\n */\n.docutils.container {\n  padding-left: unset;\n  padding-right: unset;\n  margin-left: unset;\n  margin-right: unset;\n  max-width: unset;\n  width: unset;\n}\n\n.btn {\n  --bs-btn-focus-box-shadow: 0 0 0 0.1875rem var(--pst-color-accent);\n}\n\n/**\n * Sphinx Copybutton\n * ref: https://sphinx-copybutton.readthedocs.io/\n */\ndiv.highlight button.copybtn {\n  display: flex;\n  align-items: center;\n  justify-content: center;\n  border: none;\n  background-color: var(--pst-color-surface);\n}\ndiv.highlight button.copybtn:not(.success) {\n  color: var(--pst-color-muted);\n}\ndiv.highlight button.copybtn:hover:not(.success) {\n  color: var(--pst-color-text);\n  background-color: var(--pst-color-shadow);\n}\ndiv.highlight button.copybtn.o-tooltip--left::after {\n  color: var(--pst-color-text);\n  background-color: var(--pst-color-surface);\n}\ndiv.highlight button.copybtn:focus {\n  opacity: 1;\n}\ndiv.highlight button.copybtn:focus-visible {\n  outline: var(--pst-color-accent) solid 0.1875rem;\n}\n\ndiv.highlight:has(button.copybtn) {\n  min-height: 2em;\n}\n\n#ethical-ad-placement .ethical-sidebar a,\n#ethical-ad-placement .ethical-sidebar a:visited,\n#ethical-ad-placement .ethical-sidebar a:hover,\n#ethical-ad-placement .ethical-sidebar a:active,\n#ethical-ad-placement .ethical-footer a,\n#ethical-ad-placement .ethical-footer a:visited,\n#ethical-ad-placement .ethical-footer a:hover,\n#ethical-ad-placement .ethical-footer a:active {\n  color: var(--pst-color-text-base);\n}\n#ethical-ad-placement .ethical-sidebar,\n#ethical-ad-placement .ethical-footer {\n  background-color: var(--pst-color-background);\n  border: 1px solid var(--pst-color-border);\n  border-radius: 5px;\n  color: var(--pst-color-text-base);\n  font-size: 14px;\n  line-height: 20px;\n}\n\n/**\n * Styles for various Sphinx execution libraries.\n * For now, where these define output sections, we simply revert their background\n * to be a \"light theme\" background. This ensures that inputs/outputs behave similarly,\n * because the CSS is often controlled by each package.\n * In the future, we might add dark theme support for specific packages.\n */\n/******************************************************************************\n * Jupyter Sphinx\n */\n.bd-content div.jupyter_container {\n  border: none;\n  background-color: unset;\n  box-shadow: none;\n}\n.bd-content div.jupyter_container div.output,\n.bd-content div.jupyter_container div.highlight {\n  border-radius: 0.25rem;\n}\n.bd-content div.jupyter_container div.highlight {\n  background-color: var(--pst-color-surface);\n}\n.bd-content div.jupyter_container .cell_input,\n.bd-content div.jupyter_container .cell_output {\n  border-radius: 0.25rem;\n}\n.bd-content div.jupyter_container .cell_input pre,\n.bd-content div.jupyter_container .cell_output pre {\n  padding: 1rem;\n}\n\n/* Styles for graphviz generated output from Sphinx */\n/* Style the inheritance diagram such that it has a dark mode */\nhtml[data-theme=dark] div.graphviz > object.inheritance {\n  filter: brightness(0.8) invert(0.82) contrast(1.2);\n  color-scheme: normal;\n}\n\n/**\n * Special-cases for packages in the PyData ecosystem\n */\n.xr-wrap[hidden] {\n  display: block !important;\n}\n\n.jp-OutputArea-output.lm-Widget {\n  overflow: auto;\n}\n\n/*******************************************************************************\n* master color map. Only the colors that actually differ between light and dark\n* themes are specified separately.\n*\n* To see the full list of colors see https://www.figma.com/file/rUrrHGhUBBIAAjQ82x6pz9/PyData-Design-system---proposal-for-implementation-(2)?node-id=1234%3A765&t=ifcFT1JtnrSshGfi-1\n*/\n/**\n* Function to get items from nested maps\n*/\n/* Assign base colors for the PyData theme */\n:root {\n  --pst-teal-50: #f4fbfc;\n  --pst-teal-100: #e9f6f8;\n  --pst-teal-200: #d0ecf1;\n  --pst-teal-300: #abdde6;\n  --pst-teal-400: #3fb1c5;\n  --pst-teal-500: #0a7d91;\n  --pst-teal-600: #085d6c;\n  --pst-teal-700: #064752;\n  --pst-teal-800: #042c33;\n  --pst-teal-900: #021b1f;\n  --pst-violet-50: #f4eefb;\n  --pst-violet-100: #e0c7ff;\n  --pst-violet-200: #d5b4fd;\n  --pst-violet-300: #b780ff;\n  --pst-violet-400: #9c5ffd;\n  --pst-violet-500: #8045e5;\n  --pst-violet-600: #6432bd;\n  --pst-violet-700: #4b258f;\n  --pst-violet-800: #341a61;\n  --pst-violet-900: #1e0e39;\n  --pst-gray-50: #f9f9fa;\n  --pst-gray-100: #f3f4f5;\n  --pst-gray-200: #e5e7ea;\n  --pst-gray-300: #d1d5da;\n  --pst-gray-400: #9ca4af;\n  --pst-gray-500: #677384;\n  --pst-gray-600: #48566b;\n  --pst-gray-700: #29313d;\n  --pst-gray-800: #222832;\n  --pst-gray-900: #14181e;\n  --pst-pink-50: #fcf8fd;\n  --pst-pink-100: #fcf0fa;\n  --pst-pink-200: #f8dff5;\n  --pst-pink-300: #f3c7ee;\n  --pst-pink-400: #e47fd7;\n  --pst-pink-500: #c132af;\n  --pst-pink-600: #912583;\n  --pst-pink-700: #6e1c64;\n  --pst-pink-800: #46123f;\n  --pst-pink-900: #2b0b27;\n  --pst-foundation-white: #ffffff;\n  --pst-foundation-black: #14181e;\n}\n\n/*******************************************************************************\n* write the color rules for each theme (light/dark)\n*/\n/* NOTE:\n * Mixins enable us to reuse the same definitions for the different modes\n * https://sass-lang.com/documentation/at-rules/mixin\n * something inserts a variable into a CSS selector or property name\n * https://sass-lang.com/documentation/interpolation\n */\n/* Defaults to light mode if data-theme is not set */\nhtml:not([data-theme]) {\n  --pst-color-primary: #0a7d91;\n  --pst-color-primary-bg: #d0ecf1;\n  --pst-color-secondary: #8045e5;\n  --pst-color-secondary-bg: #e0c7ff;\n  --pst-color-accent: #c132af;\n  --pst-color-accent-bg: #f8dff5;\n  --pst-color-info: #276be9;\n  --pst-color-info-bg: #dce7fc;\n  --pst-color-warning: #f66a0a;\n  --pst-color-warning-bg: #f8e3d0;\n  --pst-color-success: #00843f;\n  --pst-color-success-bg: #d6ece1;\n  --pst-color-attention: var(--pst-color-warning);\n  --pst-color-attention-bg: var(--pst-color-warning-bg);\n  --pst-color-danger: #d72d47;\n  --pst-color-danger-bg: #f9e1e4;\n  --pst-color-text-base: #222832;\n  --pst-color-text-muted: #48566b;\n  --pst-color-shadow: rgba(0, 0, 0, 0.1);\n  --pst-color-border: #d1d5da;\n  --pst-color-border-muted: rgba(23, 23, 26, 0.2);\n  --pst-color-blockquote-notch: #677384;\n  --pst-color-inline-code: #912583;\n  --pst-color-inline-code-links: #085d6c;\n  --pst-color-target: #f3cf95;\n  --pst-color-table: #14181e;\n  --pst-color-table-row-hover-bg: #d5b4fd;\n  --pst-color-table-inner-border: #e5e7ea;\n  --pst-color-background: #ffffff;\n  --pst-color-on-background: #ffffff;\n  --pst-color-surface: #f3f4f5;\n  --pst-color-on-surface: #222832;\n}\n\nhtml:not([data-theme]) {\n  --pst-color-heading: var(--pst-color-text-base);\n  --pst-color-link: var(--pst-color-primary);\n  --pst-color-link-hover: var(--pst-color-secondary);\n  --pst-color-table-outer-border: var(--pst-color-surface);\n  --pst-color-table-heading-bg: var(--pst-color-surface);\n  --pst-color-table-row-zebra-high-bg: var(--pst-color-on-background);\n  --pst-color-table-row-zebra-low-bg: var(--pst-color-surface);\n}\n\nhtml:not([data-theme]) .only-dark,\nhtml:not([data-theme]) .only-dark ~ figcaption {\n  display: none !important;\n}\n\n/* NOTE: @each {...} is like a for-loop\n * https://sass-lang.com/documentation/at-rules/control/each\n */\nhtml[data-theme=light] {\n  --pst-color-primary: #0a7d91;\n  --pst-color-primary-bg: #d0ecf1;\n  --pst-color-secondary: #8045e5;\n  --pst-color-secondary-bg: #e0c7ff;\n  --pst-color-accent: #c132af;\n  --pst-color-accent-bg: #f8dff5;\n  --pst-color-info: #276be9;\n  --pst-color-info-bg: #dce7fc;\n  --pst-color-warning: #f66a0a;\n  --pst-color-warning-bg: #f8e3d0;\n  --pst-color-success: #00843f;\n  --pst-color-success-bg: #d6ece1;\n  --pst-color-attention: var(--pst-color-warning);\n  --pst-color-attention-bg: var(--pst-color-warning-bg);\n  --pst-color-danger: #d72d47;\n  --pst-color-danger-bg: #f9e1e4;\n  --pst-color-text-base: #222832;\n  --pst-color-text-muted: #48566b;\n  --pst-color-shadow: rgba(0, 0, 0, 0.1);\n  --pst-color-border: #d1d5da;\n  --pst-color-border-muted: rgba(23, 23, 26, 0.2);\n  --pst-color-blockquote-notch: #677384;\n  --pst-color-inline-code: #912583;\n  --pst-color-inline-code-links: #085d6c;\n  --pst-color-target: #f3cf95;\n  --pst-color-table: #14181e;\n  --pst-color-table-row-hover-bg: #d5b4fd;\n  --pst-color-table-inner-border: #e5e7ea;\n  --pst-color-background: #ffffff;\n  --pst-color-on-background: #ffffff;\n  --pst-color-surface: #f3f4f5;\n  --pst-color-on-surface: #222832;\n  color-scheme: light;\n}\n\nhtml[data-theme=light] {\n  --pst-color-heading: var(--pst-color-text-base);\n  --pst-color-link: var(--pst-color-primary);\n  --pst-color-link-hover: var(--pst-color-secondary);\n  --pst-color-table-outer-border: var(--pst-color-surface);\n  --pst-color-table-heading-bg: var(--pst-color-surface);\n  --pst-color-table-row-zebra-high-bg: var(--pst-color-on-background);\n  --pst-color-table-row-zebra-low-bg: var(--pst-color-surface);\n}\n\nhtml[data-theme=light] .only-dark,\nhtml[data-theme=light] .only-dark ~ figcaption {\n  display: none !important;\n}\n\nhtml[data-theme=dark] {\n  --pst-color-primary: #3fb1c5;\n  --pst-color-primary-bg: #042c33;\n  --pst-color-secondary: #9c5ffd;\n  --pst-color-secondary-bg: #341a61;\n  --pst-color-accent: #e47fd7;\n  --pst-color-accent-bg: #46123f;\n  --pst-color-info: #79a3f2;\n  --pst-color-info-bg: #06245d;\n  --pst-color-warning: #ff9245;\n  --pst-color-warning-bg: #652a02;\n  --pst-color-success: #5fb488;\n  --pst-color-success-bg: #002f17;\n  --pst-color-attention: var(--pst-color-warning);\n  --pst-color-attention-bg: var(--pst-color-warning-bg);\n  --pst-color-danger: #e78894;\n  --pst-color-danger-bg: #4e111b;\n  --pst-color-text-base: #ced6dd;\n  --pst-color-text-muted: #9ca4af;\n  --pst-color-shadow: rgba(0, 0, 0, 0.2);\n  --pst-color-border: #48566b;\n  --pst-color-border-muted: #29313d;\n  --pst-color-blockquote-notch: #9ca4af;\n  --pst-color-inline-code: #f3c7ee;\n  --pst-color-inline-code-links: #3fb1c5;\n  --pst-color-target: #675c04;\n  --pst-color-table: #ffffff;\n  --pst-color-table-row-hover-bg: #4b258f;\n  --pst-color-table-inner-border: #364150;\n  --pst-color-background: #14181e;\n  --pst-color-on-background: #222832;\n  --pst-color-surface: #29313d;\n  --pst-color-on-surface: #f3f4f5;\n  /* Adjust images in dark mode (unless they have class .only-dark or\n   * .dark-light, in which case assume they're already optimized for dark\n   * mode).\n   */\n  /* Give images a light background in dark mode in case they have\n  *  transparency and black text (unless they have class .only-dark or .dark-light, in\n  *  which case assume they're already optimized for dark mode).\n  */\n  color-scheme: dark;\n}\n\nhtml[data-theme=dark] {\n  --pst-color-heading: var(--pst-color-text-base);\n  --pst-color-link: var(--pst-color-primary);\n  --pst-color-link-hover: var(--pst-color-secondary);\n  --pst-color-table-outer-border: var(--pst-color-surface);\n  --pst-color-table-heading-bg: var(--pst-color-surface);\n  --pst-color-table-row-zebra-high-bg: var(--pst-color-on-background);\n  --pst-color-table-row-zebra-low-bg: var(--pst-color-surface);\n}\n\nhtml[data-theme=dark] .only-light,\nhtml[data-theme=dark] .only-light ~ figcaption {\n  display: none !important;\n}\n\nhtml[data-theme=dark] img:not(.only-dark, .dark-light) {\n  filter: brightness(0.8) contrast(1.2);\n}\n\nhtml[data-theme=dark] .bd-content img:not(.only-dark, .dark-light) {\n  background-color: rgb(255, 255, 255);\n  border-radius: 0.25rem;\n}\n\nhtml[data-theme=dark] .MathJax_SVG * {\n  fill: var(--pst-color-text-base);\n}\n\n.pst-color-primary {\n  color: var(--pst-color-primary);\n}\n\n.pst-color-secondary {\n  color: var(--pst-color-secondary);\n}\n\n.pst-color-accent {\n  color: var(--pst-color-accent);\n}\n\n.pst-color-info {\n  color: var(--pst-color-info);\n}\n\n.pst-color-warning {\n  color: var(--pst-color-warning);\n}\n\n.pst-color-success {\n  color: var(--pst-color-success);\n}\n\n.pst-color-attention {\n  color: var(--pst-color-attention);\n}\n\n.pst-color-danger {\n  color: var(--pst-color-danger);\n}\n\n.pst-color-text-base {\n  color: var(--pst-color-text-base);\n}\n\n.pst-color-text-muted {\n  color: var(--pst-color-text-muted);\n}\n\n.pst-color-shadow {\n  color: var(--pst-color-shadow);\n}\n\n.pst-color-border {\n  color: var(--pst-color-border);\n}\n\n.pst-color-border-muted {\n  color: var(--pst-color-border-muted);\n}\n\n.pst-color-blockquote-notch {\n  color: var(--pst-color-blockquote-notch);\n}\n\n.pst-color-inline-code {\n  color: var(--pst-color-inline-code);\n}\n\n.pst-color-inline-code-links {\n  color: var(--pst-color-inline-code-links);\n}\n\n.pst-color-target {\n  color: var(--pst-color-target);\n}\n\n.pst-color-table {\n  color: var(--pst-color-table);\n}\n\n.pst-color-table-row-hover {\n  color: var(--pst-color-table-row-hover);\n}\n\n.pst-color-table-inner-border {\n  color: var(--pst-color-table-inner-border);\n}\n\n.pst-color-background {\n  color: var(--pst-color-background);\n}\n\n.pst-color-on-background {\n  color: var(--pst-color-on-background);\n}\n\n.pst-color-surface {\n  color: var(--pst-color-surface);\n}\n\n.pst-color-on-surface {\n  color: var(--pst-color-on-surface);\n}\n\n/*******************************************************************************\n  * Color and variables\n  *\n  * This is a list of the semantic color names from sphinx-design (we only\n  * need to override variables that sphinx-design has actually defined).\n  * https://github.com/executablebooks/sphinx-design/blob/9226a12a/style/_colors.scss#L31-L43\n  */\n/**\n  * Here we create some extra --pst-color-* variables and use\n  * them to override the value of the corresponding sphinx-design variables.\n  * This is easier than re-writing the sphinx-design rules. Even easier would be\n  * directly assigning our values to the --sd-color-* variables, but then our\n  * downstream users couldn't override *our* colors and have it affect buttons\n  * and badges.\n  *\n  * First, define the extra keys needed to cover the full range of semantic\n  * color names used in sphinx-design, then merge them with the names we\n  * already define for our own needs.\n  * see https://sphinx-design.readthedocs.io/en/latest/css_variables.html\n  */\nhtml[data-theme=light] {\n  --pst-color-primary: rgb(10, 125, 145);\n  --pst-color-primary-text: #fff;\n  --pst-color-primary-highlight: #086b7c;\n  --sd-color-primary: var(--pst-color-primary);\n  --sd-color-primary-text: var(--pst-color-primary-text);\n  --sd-color-primary-highlight: var(--pst-color-primary-highlight);\n  --pst-color-primary-highlight-text: #fff;\n  --sd-color-primary-bg: #d0ecf1;\n  --sd-color-primary-bg-text: #14181e;\n  --pst-color-secondary: rgb(128, 69, 229);\n  --pst-color-secondary-text: #fff;\n  --pst-color-secondary-highlight: #651ee0;\n  --sd-color-secondary: var(--pst-color-secondary);\n  --sd-color-secondary-text: var(--pst-color-secondary-text);\n  --sd-color-secondary-highlight: var(--pst-color-secondary-highlight);\n  --pst-color-secondary-highlight-text: #fff;\n  --sd-color-secondary-bg: #e0c7ff;\n  --sd-color-secondary-bg-text: #14181e;\n  --pst-color-success: #00843f;\n  --pst-color-success-text: #fff;\n  --pst-color-success-highlight: #007036;\n  --sd-color-success: var(--pst-color-success);\n  --sd-color-success-text: var(--pst-color-success-text);\n  --sd-color-success-highlight: var(--pst-color-success-highlight);\n  --pst-color-success-highlight-text: #fff;\n  --sd-color-success-bg: #d6ece1;\n  --sd-color-success-bg-text: #14181e;\n  --pst-color-info: #276be9;\n  --pst-color-info-text: #fff;\n  --pst-color-info-highlight: #1457d3;\n  --sd-color-info: var(--pst-color-info);\n  --sd-color-info-text: var(--pst-color-info-text);\n  --sd-color-info-highlight: var(--pst-color-info-highlight);\n  --pst-color-info-highlight-text: #fff;\n  --sd-color-info-bg: #dce7fc;\n  --sd-color-info-bg-text: #14181e;\n  --pst-color-warning: #f66a0a;\n  --pst-color-warning-text: #14181e;\n  --pst-color-warning-highlight: #d25a07;\n  --sd-color-warning: var(--pst-color-warning);\n  --sd-color-warning-text: var(--pst-color-warning-text);\n  --sd-color-warning-highlight: var(--pst-color-warning-highlight);\n  --pst-color-warning-highlight-text: #14181e;\n  --sd-color-warning-bg: #f8e3d0;\n  --sd-color-warning-bg-text: #14181e;\n  --pst-color-danger: #d72d47;\n  --pst-color-danger-text: #fff;\n  --pst-color-danger-highlight: #bb2239;\n  --sd-color-danger: var(--pst-color-danger);\n  --sd-color-danger-text: var(--pst-color-danger-text);\n  --sd-color-danger-highlight: var(--pst-color-danger-highlight);\n  --pst-color-danger-highlight-text: #fff;\n  --sd-color-danger-bg: #f9e1e4;\n  --sd-color-danger-bg-text: #14181e;\n  --pst-color-light: #f3f4f5;\n  --pst-color-light-text: #14181e;\n  --pst-color-light-highlight: #c9cfd6;\n  --sd-color-light: var(--pst-color-light);\n  --sd-color-light-text: var(--pst-color-light-text);\n  --sd-color-light-highlight: var(--pst-color-light-highlight);\n  --pst-color-light-highlight-text: #14181e;\n  --sd-color-light-bg: #f7f7f8;\n  --sd-color-light-bg-text: #14181e;\n  --pst-color-muted: #29313d;\n  --pst-color-muted-text: #fff;\n  --pst-color-muted-highlight: #212936;\n  --sd-color-muted: var(--pst-color-muted);\n  --sd-color-muted-text: var(--pst-color-muted-text);\n  --sd-color-muted-highlight: var(--pst-color-muted-highlight);\n  --pst-color-muted-highlight-text: #fff;\n  --sd-color-muted-bg: #5a6c86;\n  --sd-color-muted-bg-text: #fff;\n  --pst-color-dark: #222832;\n  --pst-color-dark-text: #fff;\n  --pst-color-dark-highlight: #1b222c;\n  --sd-color-dark: var(--pst-color-dark);\n  --sd-color-dark-text: var(--pst-color-dark-text);\n  --sd-color-dark-highlight: var(--pst-color-dark-highlight);\n  --pst-color-dark-highlight-text: #fff;\n  --pst-color-black: #14181e;\n  --pst-color-black-text: #fff;\n  --pst-color-black-highlight: #10141a;\n  --sd-color-black: var(--pst-color-black);\n  --sd-color-black-text: var(--pst-color-black-text);\n  --sd-color-black-highlight: var(--pst-color-black-highlight);\n  --pst-color-black-highlight-text: #fff;\n  --pst-color-white: #fff;\n  --pst-color-white-text: #14181e;\n  --pst-color-white-highlight: #dbd7d7;\n  --sd-color-white: var(--pst-color-white);\n  --sd-color-white-text: var(--pst-color-white-text);\n  --sd-color-white-highlight: var(--pst-color-white-highlight);\n  --pst-color-white-highlight-text: #14181e;\n}\n\nhtml[data-theme=dark] {\n  --pst-color-primary: rgb(63, 177, 197);\n  --pst-color-primary-text: #14181e;\n  --pst-color-primary-highlight: #319aac;\n  --sd-color-primary: var(--pst-color-primary);\n  --sd-color-primary-text: var(--pst-color-primary-text);\n  --sd-color-primary-highlight: var(--pst-color-primary-highlight);\n  --pst-color-primary-highlight-text: #14181e;\n  --sd-color-primary-bg: #042c33;\n  --sd-color-primary-bg-text: #fff;\n  --pst-color-secondary: rgb(156, 95, 253);\n  --pst-color-secondary-text: #14181e;\n  --pst-color-secondary-highlight: #7c2bfc;\n  --sd-color-secondary: var(--pst-color-secondary);\n  --sd-color-secondary-text: var(--pst-color-secondary-text);\n  --sd-color-secondary-highlight: var(--pst-color-secondary-highlight);\n  --pst-color-secondary-highlight-text: #fff;\n  --sd-color-secondary-bg: #341a61;\n  --sd-color-secondary-bg-text: #fff;\n  --pst-color-success: #5fb488;\n  --pst-color-success-text: #14181e;\n  --pst-color-success-highlight: #47a373;\n  --sd-color-success: var(--pst-color-success);\n  --sd-color-success-text: var(--pst-color-success-text);\n  --sd-color-success-highlight: var(--pst-color-success-highlight);\n  --pst-color-success-highlight-text: #14181e;\n  --sd-color-success-bg: #002f17;\n  --sd-color-success-bg-text: #fff;\n  --pst-color-info: #79a3f2;\n  --pst-color-info-text: #14181e;\n  --pst-color-info-highlight: #4681ee;\n  --sd-color-info: var(--pst-color-info);\n  --sd-color-info-text: var(--pst-color-info-text);\n  --sd-color-info-highlight: var(--pst-color-info-highlight);\n  --pst-color-info-highlight-text: #14181e;\n  --sd-color-info-bg: #06245d;\n  --sd-color-info-bg-text: #fff;\n  --pst-color-warning: #ff9245;\n  --pst-color-warning-text: #14181e;\n  --pst-color-warning-highlight: #ff7614;\n  --sd-color-warning: var(--pst-color-warning);\n  --sd-color-warning-text: var(--pst-color-warning-text);\n  --sd-color-warning-highlight: var(--pst-color-warning-highlight);\n  --pst-color-warning-highlight-text: #14181e;\n  --sd-color-warning-bg: #652a02;\n  --sd-color-warning-bg-text: #fff;\n  --pst-color-danger: #e78894;\n  --pst-color-danger-text: #14181e;\n  --pst-color-danger-highlight: #df596a;\n  --sd-color-danger: var(--pst-color-danger);\n  --sd-color-danger-text: var(--pst-color-danger-text);\n  --sd-color-danger-highlight: var(--pst-color-danger-highlight);\n  --pst-color-danger-highlight-text: #14181e;\n  --sd-color-danger-bg: #4e111b;\n  --sd-color-danger-bg-text: #fff;\n  --pst-color-light: #f3f4f5;\n  --pst-color-light-text: #14181e;\n  --pst-color-light-highlight: #c9cfd6;\n  --sd-color-light: var(--pst-color-light);\n  --sd-color-light-text: var(--pst-color-light-text);\n  --sd-color-light-highlight: var(--pst-color-light-highlight);\n  --pst-color-light-highlight-text: #14181e;\n  --sd-color-light-bg: #a3abb2;\n  --sd-color-light-bg-text: #14181e;\n  --pst-color-muted: #f3f4f5;\n  --pst-color-muted-text: #14181e;\n  --pst-color-muted-highlight: #c9cfd6;\n  --sd-color-muted: var(--pst-color-muted);\n  --sd-color-muted-text: var(--pst-color-muted-text);\n  --sd-color-muted-highlight: var(--pst-color-muted-highlight);\n  --pst-color-muted-highlight-text: #14181e;\n  --sd-color-muted-bg: #1d222b;\n  --sd-color-muted-bg-text: #fff;\n  --pst-color-dark: #222832;\n  --pst-color-dark-text: #fff;\n  --pst-color-dark-highlight: #1b222c;\n  --sd-color-dark: var(--pst-color-dark);\n  --sd-color-dark-text: var(--pst-color-dark-text);\n  --sd-color-dark-highlight: var(--pst-color-dark-highlight);\n  --pst-color-dark-highlight-text: #fff;\n  --pst-color-black: #14181e;\n  --pst-color-black-text: #fff;\n  --pst-color-black-highlight: #10141a;\n  --sd-color-black: var(--pst-color-black);\n  --sd-color-black-text: var(--pst-color-black-text);\n  --sd-color-black-highlight: var(--pst-color-black-highlight);\n  --pst-color-black-highlight-text: #fff;\n  --pst-color-white: #fff;\n  --pst-color-white-text: #14181e;\n  --pst-color-white-highlight: #dbd7d7;\n  --sd-color-white: var(--pst-color-white);\n  --sd-color-white-text: var(--pst-color-white-text);\n  --sd-color-white-highlight: var(--pst-color-white-highlight);\n  --pst-color-white-highlight-text: #14181e;\n}\n\nhtml[data-theme=light] {\n  --sd-color-card-border: var(--pst-color-border);\n}\n\nhtml[data-theme=dark] {\n  --sd-color-card-border: var(--pst-color-border);\n}\n\n/*******************************************************************************\n  * shadows\n  */\nhtml[data-theme=light] .sd-shadow-xs,\nhtml[data-theme=light] .sd-shadow-sm,\nhtml[data-theme=light] .sd-shadow-md,\nhtml[data-theme=light] .sd-shadow-lg {\n  box-shadow: 0 0.2rem 0.5rem var(--pst-color-shadow), 0 0 0.0625rem var(--pst-color-shadow) !important;\n}\n\n/*******************************************************************************\n  * cards\n  */\n.bd-content .sd-card {\n  border: 1px solid var(--pst-color-border);\n}\n.bd-content .sd-card .sd-card-header {\n  background-color: var(--pst-color-on-background);\n  border-bottom: 1px solid var(--pst-color-border);\n}\n.bd-content .sd-card .sd-card-footer {\n  background-color: var(--pst-color-on-background);\n  border-top: 1px solid var(--pst-color-border);\n}\n.bd-content .sd-card .sd-card-body {\n  background-color: var(--pst-color-on-background);\n}\n.bd-content .sd-card .sd-stretched-link:focus-visible {\n  outline: none;\n}\n.bd-content .sd-card .sd-stretched-link:focus-visible::after {\n  outline: var(--pst-color-accent) solid 0.1875rem;\n  border-radius: 0.25rem;\n}\n.bd-content .sd-card.sd-card-hover:hover {\n  border-color: var(--pst-color-link-hover);\n}\n\n/*******************************************************************************\n  * tabs\n  */\n.bd-content .sd-tab-set > input:checked + label {\n  border-style: solid solid none;\n  border-color: var(--pst-color-primary) var(--pst-color-primary) transparent;\n  border-width: 0.125rem 0.125rem 0;\n  border-radius: 0.125rem 0.125rem 0 0;\n  background-color: var(--pst-color-on-background);\n  transform: translateY(0.125rem);\n  color: var(--pst-color-primary);\n}\n.bd-content .sd-tab-set > input:focus-visible + label {\n  border: 0.125rem solid var(--pst-color-accent);\n  border-radius: 0.125rem;\n  background-color: var(--pst-color-accent-bg);\n  color: var(--pst-color-on-surface);\n}\n.bd-content .sd-tab-set > input:not(:checked, :focus-visible) + label:hover {\n  border-color: transparent;\n  color: var(--pst-color-secondary);\n  text-decoration-line: underline;\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n}\n.bd-content .sd-tab-set > label {\n  color: var(--pst-color-on-surface);\n  border: 0.125rem solid transparent;\n  border-radius: 0.125rem 0.125rem 0 0;\n  background-color: var(--pst-color-surface);\n  padding: 0 0.75em;\n  margin-inline-end: 0.25rem;\n  line-height: 1.95;\n}\nhtml[data-theme=dark] .bd-content .sd-tab-set > label {\n  background-color: var(--pst-color-on-background);\n}\n.bd-content .sd-tab-set .sd-tab-content {\n  border: 0.125rem solid var(--pst-color-primary);\n  border-radius: 0.1875rem;\n  box-shadow: unset;\n  padding: 0.625rem;\n  background-color: var(--pst-color-on-background);\n}\n\n/*******************************************************************************\n* Dropdowns\n*/\ndetails.sd-dropdown {\n  border: 0 !important;\n  box-shadow: 0 0.2rem 0.5rem var(--pst-color-shadow), 0 0 0.0625rem var(--pst-color-shadow) !important;\n}\ndetails.sd-dropdown summary.sd-card-header {\n  border: 0 !important;\n  display: flex;\n  align-items: center;\n  position: relative;\n  font-weight: 600;\n  padding-top: 0.5rem;\n  padding-bottom: 0.5rem;\n  --pst-sd-dropdown-color: var(--pst-gray-500);\n  --pst-sd-dropdown-bg-color: var(--pst-color-surface);\n  background-color: var(--pst-sd-dropdown-bg-color) !important;\n  border-left: 0.2rem solid var(--pst-sd-dropdown-color) !important;\n}\ndetails.sd-dropdown summary.sd-card-header + div.sd-summary-content {\n  border: 0;\n  --pst-sd-dropdown-color: var(--sd-color-card-border);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-primary {\n  --pst-sd-dropdown-color: var(--sd-color-primary);\n  --pst-sd-dropdown-bg-color: var(--sd-color-primary-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-primary + div.sd-summary-content {\n  --pst-sd-dropdown-color: var(--sd-color-primary);\n  --pst-sd-dropdown-bg-color: var(--sd-color-primary-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-text-primary {\n  color: var(--sd-color-primary-bg-text) !important;\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-secondary {\n  --pst-sd-dropdown-color: var(--sd-color-secondary);\n  --pst-sd-dropdown-bg-color: var(--sd-color-secondary-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-secondary + div.sd-summary-content {\n  --pst-sd-dropdown-color: var(--sd-color-secondary);\n  --pst-sd-dropdown-bg-color: var(--sd-color-secondary-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-text-secondary {\n  color: var(--sd-color-secondary-bg-text) !important;\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-success {\n  --pst-sd-dropdown-color: var(--sd-color-success);\n  --pst-sd-dropdown-bg-color: var(--sd-color-success-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-success + div.sd-summary-content {\n  --pst-sd-dropdown-color: var(--sd-color-success);\n  --pst-sd-dropdown-bg-color: var(--sd-color-success-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-text-success {\n  color: var(--sd-color-success-bg-text) !important;\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-info {\n  --pst-sd-dropdown-color: var(--sd-color-info);\n  --pst-sd-dropdown-bg-color: var(--sd-color-info-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-info + div.sd-summary-content {\n  --pst-sd-dropdown-color: var(--sd-color-info);\n  --pst-sd-dropdown-bg-color: var(--sd-color-info-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-text-info {\n  color: var(--sd-color-info-bg-text) !important;\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-warning {\n  --pst-sd-dropdown-color: var(--sd-color-warning);\n  --pst-sd-dropdown-bg-color: var(--sd-color-warning-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-warning + div.sd-summary-content {\n  --pst-sd-dropdown-color: var(--sd-color-warning);\n  --pst-sd-dropdown-bg-color: var(--sd-color-warning-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-text-warning {\n  color: var(--sd-color-warning-bg-text) !important;\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-danger {\n  --pst-sd-dropdown-color: var(--sd-color-danger);\n  --pst-sd-dropdown-bg-color: var(--sd-color-danger-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-danger + div.sd-summary-content {\n  --pst-sd-dropdown-color: var(--sd-color-danger);\n  --pst-sd-dropdown-bg-color: var(--sd-color-danger-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-text-danger {\n  color: var(--sd-color-danger-bg-text) !important;\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-light {\n  --pst-sd-dropdown-color: var(--sd-color-light);\n  --pst-sd-dropdown-bg-color: var(--sd-color-light-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-light + div.sd-summary-content {\n  --pst-sd-dropdown-color: var(--sd-color-light);\n  --pst-sd-dropdown-bg-color: var(--sd-color-light-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-text-light {\n  color: var(--sd-color-light-bg-text) !important;\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-muted {\n  --pst-sd-dropdown-color: var(--sd-color-muted);\n  --pst-sd-dropdown-bg-color: var(--sd-color-muted-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-muted + div.sd-summary-content {\n  --pst-sd-dropdown-color: var(--sd-color-muted);\n  --pst-sd-dropdown-bg-color: var(--sd-color-muted-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-text-muted {\n  color: var(--sd-color-muted-bg-text) !important;\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-dark {\n  --pst-sd-dropdown-color: var(--sd-color-dark);\n  --pst-sd-dropdown-bg-color: var(--sd-color-dark-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-dark + div.sd-summary-content {\n  --pst-sd-dropdown-color: var(--sd-color-dark);\n  --pst-sd-dropdown-bg-color: var(--sd-color-dark-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-text-dark {\n  color: var(--sd-color-dark-bg-text) !important;\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-black {\n  --pst-sd-dropdown-color: var(--sd-color-black);\n  --pst-sd-dropdown-bg-color: var(--sd-color-black-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-black + div.sd-summary-content {\n  --pst-sd-dropdown-color: var(--sd-color-black);\n  --pst-sd-dropdown-bg-color: var(--sd-color-black-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-text-black {\n  color: var(--sd-color-black-bg-text) !important;\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-white {\n  --pst-sd-dropdown-color: var(--sd-color-white);\n  --pst-sd-dropdown-bg-color: var(--sd-color-white-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-white + div.sd-summary-content {\n  --pst-sd-dropdown-color: var(--sd-color-white);\n  --pst-sd-dropdown-bg-color: var(--sd-color-white-bg);\n}\ndetails.sd-dropdown summary.sd-card-header.sd-bg-text-white {\n  color: var(--sd-color-white-bg-text) !important;\n}\ndetails.sd-dropdown summary.sd-card-header::before {\n  content: \"\";\n  width: 100%;\n  height: 100%;\n  position: absolute;\n  left: 0;\n  top: 0;\n  z-index: -1;\n  pointer-events: none;\n}\ndetails.sd-dropdown summary.sd-card-header + div.sd-summary-content {\n  border-left: 0.2rem solid var(--pst-sd-dropdown-color) !important;\n  border-bottom-left-radius: calc(0.25rem - 1px);\n  background-color: var(--pst-color-on-background);\n}\ndetails.sd-dropdown summary.sd-card-header span.sd-summary-icon {\n  display: inline-flex;\n  align-items: center;\n  color: var(--pst-sd-dropdown-color) !important;\n}\ndetails.sd-dropdown summary.sd-card-header span.sd-summary-icon svg {\n  opacity: 1;\n}\ndetails.sd-dropdown summary.sd-card-header .sd-summary-up,\ndetails.sd-dropdown summary.sd-card-header .sd-summary-down {\n  top: 0.7rem;\n}\ndetails.sd-dropdown summary.sd-card-header:hover {\n  filter: brightness(0.9);\n}\nhtml[data-theme=dark] details.sd-dropdown summary.sd-card-header:hover {\n  filter: brightness(1.1);\n}\ndetails.sd-dropdown summary.sd-card-header:focus:focus-visible {\n  outline: var(--pst-color-accent) solid 0.1875rem;\n  outline-offset: 0.125rem;\n  border-radius: 0.1875rem;\n}\ndetails.sd-dropdown[open] summary.sd-card-header:focus:focus-visible {\n  border-bottom-left-radius: 0;\n  border-bottom-right-radius: 0;\n}\n\n/*******************************************************************************\n* Buttons (which in Sphinx Design are actually links that look like buttons)\n* ref: https://sphinx-design.readthedocs.io/en/pydata-theme/badges_buttons.html#buttons\n*/\nhtml .sd-btn {\n  min-width: 2.25rem;\n  padding: 0.3125rem 0.75rem 0.4375rem;\n}\nhtml .sd-btn:hover {\n  text-decoration: underline;\n  text-decoration-thickness: max(1px, .0625rem);\n  text-underline-offset: 0.1578em;\n  text-decoration-thickness: max(3px, .1875rem, .12em);\n  text-decoration-skip-ink: none;\n  color: var(--pst-color-link-hover);\n  text-decoration-thickness: 1px;\n}\nhtml .sd-btn-primary:hover {\n  color: var(--pst-color-primary-highlight-text) !important;\n}\nhtml .sd-btn-secondary:hover {\n  color: var(--pst-color-secondary-highlight-text) !important;\n}\nhtml .sd-btn-success:hover {\n  color: var(--pst-color-success-highlight-text) !important;\n}\nhtml .sd-btn-info:hover {\n  color: var(--pst-color-info-highlight-text) !important;\n}\nhtml .sd-btn-warning:hover {\n  color: var(--pst-color-warning-highlight-text) !important;\n}\nhtml .sd-btn-danger:hover {\n  color: var(--pst-color-danger-highlight-text) !important;\n}\nhtml .sd-btn-light:hover {\n  color: var(--pst-color-light-highlight-text) !important;\n}\nhtml .sd-btn-muted:hover {\n  color: var(--pst-color-muted-highlight-text) !important;\n}\nhtml .sd-btn-dark:hover {\n  color: var(--pst-color-dark-highlight-text) !important;\n}\nhtml .sd-btn-black:hover {\n  color: var(--pst-color-black-highlight-text) !important;\n}\nhtml .sd-btn-white:hover {\n  color: var(--pst-color-white-highlight-text) !important;\n}\nhtml .sd-btn-primary:focus-visible,\nhtml .sd-btn-outline-primary:focus-visible {\n  outline: var(--sd-color-primary) solid 0.1875rem;\n  outline-offset: 0.125rem;\n  border-radius: 0.1875rem;\n  background-color: var(--sd-color-primary) !important;\n  border-color: var(--sd-color-primary) !important;\n}\nhtml .sd-btn-secondary:focus-visible,\nhtml .sd-btn-outline-secondary:focus-visible {\n  outline: var(--sd-color-secondary) solid 0.1875rem;\n  outline-offset: 0.125rem;\n  border-radius: 0.1875rem;\n  background-color: var(--sd-color-secondary) !important;\n  border-color: var(--sd-color-secondary) !important;\n}\nhtml .sd-btn-success:focus-visible,\nhtml .sd-btn-outline-success:focus-visible {\n  outline: var(--sd-color-success) solid 0.1875rem;\n  outline-offset: 0.125rem;\n  border-radius: 0.1875rem;\n  background-color: var(--sd-color-success) !important;\n  border-color: var(--sd-color-success) !important;\n}\nhtml .sd-btn-info:focus-visible,\nhtml .sd-btn-outline-info:focus-visible {\n  outline: var(--sd-color-info) solid 0.1875rem;\n  outline-offset: 0.125rem;\n  border-radius: 0.1875rem;\n  background-color: var(--sd-color-info) !important;\n  border-color: var(--sd-color-info) !important;\n}\nhtml .sd-btn-warning:focus-visible,\nhtml .sd-btn-outline-warning:focus-visible {\n  outline: var(--sd-color-warning) solid 0.1875rem;\n  outline-offset: 0.125rem;\n  border-radius: 0.1875rem;\n  background-color: var(--sd-color-warning) !important;\n  border-color: var(--sd-color-warning) !important;\n}\nhtml .sd-btn-danger:focus-visible,\nhtml .sd-btn-outline-danger:focus-visible {\n  outline: var(--sd-color-danger) solid 0.1875rem;\n  outline-offset: 0.125rem;\n  border-radius: 0.1875rem;\n  background-color: var(--sd-color-danger) !important;\n  border-color: var(--sd-color-danger) !important;\n}\nhtml .sd-btn-light:focus-visible,\nhtml .sd-btn-outline-light:focus-visible {\n  outline: var(--sd-color-light) solid 0.1875rem;\n  outline-offset: 0.125rem;\n  border-radius: 0.1875rem;\n  background-color: var(--sd-color-light) !important;\n  border-color: var(--sd-color-light) !important;\n}\nhtml .sd-btn-muted:focus-visible,\nhtml .sd-btn-outline-muted:focus-visible {\n  outline: var(--sd-color-muted) solid 0.1875rem;\n  outline-offset: 0.125rem;\n  border-radius: 0.1875rem;\n  background-color: var(--sd-color-muted) !important;\n  border-color: var(--sd-color-muted) !important;\n}\nhtml .sd-btn-dark:focus-visible,\nhtml .sd-btn-outline-dark:focus-visible {\n  outline: var(--sd-color-dark) solid 0.1875rem;\n  outline-offset: 0.125rem;\n  border-radius: 0.1875rem;\n  background-color: var(--sd-color-dark) !important;\n  border-color: var(--sd-color-dark) !important;\n}\nhtml .sd-btn-black:focus-visible,\nhtml .sd-btn-outline-black:focus-visible {\n  outline: var(--sd-color-black) solid 0.1875rem;\n  outline-offset: 0.125rem;\n  border-radius: 0.1875rem;\n  background-color: var(--sd-color-black) !important;\n  border-color: var(--sd-color-black) !important;\n}\nhtml .sd-btn-white:focus-visible,\nhtml .sd-btn-outline-white:focus-visible {\n  outline: var(--sd-color-white) solid 0.1875rem;\n  outline-offset: 0.125rem;\n  border-radius: 0.1875rem;\n  background-color: var(--sd-color-white) !important;\n  border-color: var(--sd-color-white) !important;\n}\n\n/**\n * Sphinx togglebutton\n *\n * The rules in this style sheet are meant to tweak the\n * [sphinx-togglebutton](https://sphinx-togglebutton.readthedocs.io/en/latest/index.html)\n * extension so that it matches the look and feel of this theme.\n */\n.bd-content .dropdown.admonition.toggle .admonition-title:hover .tb-icon {\n  opacity: 1;\n  scale: 1.1;\n}\n.bd-content .dropdown.admonition.toggle .admonition-title .tb-icon {\n  opacity: 0.6;\n}\n.bd-content .dropdown.admonition.toggle .admonition-title:hover {\n  filter: brightness(0.9);\n}\nhtml[data-theme=dark] .bd-content .dropdown.admonition.toggle .admonition-title:hover {\n  filter: brightness(1.1);\n}\n.bd-content .dropdown.admonition.toggle .admonition-title:hover {\n  box-shadow: none;\n}\n.bd-content .dropdown.admonition.toggle button.toggle-button {\n  color: inherit;\n}\n.bd-content .dropdown.admonition.toggle button.toggle-button.toggle-button-hidden .toggle-chevron-right {\n  transform: rotate(90deg);\n  transition: none;\n}\n.bd-content .dropdown.admonition.toggle button.toggle-button .toggle-chevron-right {\n  transform: rotate(-90deg);\n  transition: none;\n}\n.bd-content .dropdown.admonition.toggle:has(:focus-visible) {\n  /* Override Sphinx Toggle Button. Make the overflow visible, otherwise the\n  focus ring is hidden. */\n  overflow: visible;\n  /*\n  Why not just do the following?\n\n  ```\n  .admonition-title {\n    outline: $focus-ring-outline;\n  }\n  ```\n\n  Why use ::before? If we put the focus ring on the ::before pseudo-element,\n  we can reposition the focus ring by repositioning the pseudo-element.\n\n  Why reposition? The left edge of the admonition title box does not align\n  with the left edge of the overall admonition box. There is a left border\n  that belongs to the overall box. The border is outside of the admonition\n  title, which means it is also outside of a focus ring around the title. We\n  can make the focus ring bigger, with `outline-offset`, but this will\n  result in a ring that looks off-centered. So we have to pull the ring left\n  and stretch it right. That's what the pseudo-element allows us to do.\n\n  Note: we do not have to do this for collapsible admonitions made with\n  Sphinx Design (as opposed to sphinx-togglebutton).\n  */\n}\n.bd-content .dropdown.admonition.toggle:has(:focus-visible) .admonition-title::before {\n  content: \"\";\n  left: -0.2rem;\n  width: calc(100% + 0.2rem);\n  height: 100%;\n  outline: var(--pst-color-accent) solid 0.1875rem;\n  outline-offset: 0.125rem;\n  border-radius: 0.1875rem;\n}\n.bd-content .dropdown.admonition.toggle:has(:focus-visible):not(.toggle-hidden) .admonition-title::before {\n  border-bottom-left-radius: 0;\n  border-bottom-right-radius: 0;\n}\n.bd-content details.toggle-details summary {\n  border-left: 3px solid var(--pst-color-primary);\n}\n.bd-content details.toggle-details summary .toggle-chevron-right {\n  transform: rotate(90deg);\n  transition: none;\n}\n.bd-content details.toggle-details summary:hover .tb-icon {\n  opacity: 1;\n  scale: 1.1;\n}\n.bd-content details.toggle-details summary .tb-icon {\n  opacity: 0.6;\n}\n.bd-content details.toggle-details summary:hover {\n  filter: brightness(0.9);\n}\nhtml[data-theme=dark] .bd-content details.toggle-details summary:hover {\n  filter: brightness(1.1);\n}\n.bd-content details.toggle-details summary:hover {\n  box-shadow: none;\n}\n.bd-content details.toggle-details summary:focus-visible {\n  outline-offset: 0.125rem;\n}\n.bd-content details.toggle-details[open] .toggle-chevron-right {\n  transform: rotate(-90deg);\n  transition: none;\n}\n.bd-content details.toggle-details[open] :focus-visible {\n  border-bottom-left-radius: 0;\n  border-bottom-right-radius: 0;\n}\n\n/**\n * Styles for various Sphinx execution libraries to display pre-executed notebooks.\n * For now, where these define output sections, we simply revert their background\n * to be a \"light theme\" background. This ensures that inputs/outputs behave similarly,\n * because the CSS is often controlled by each package.\n * In the future, we might add dark theme support for specific packages.\n */\n/*******************************************************************************\n * nbsphinx\n */\nhtml div.rendered_html,\nhtml .jp-RenderedHTMLCommon {\n  margin: 0.1875rem;\n}\nhtml div.rendered_html table,\nhtml .jp-RenderedHTMLCommon table {\n  table-layout: auto;\n}\n\n.bd-content .nboutput .output_area.rendered_html table.dataframe,\n.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe {\n  color: var(--pst-color-table);\n  border: 1px solid var(--pst-color-table-outer-border);\n}\n.bd-content .nboutput .output_area.rendered_html table.dataframe th ~ th,\n.bd-content .nboutput .output_area.rendered_html table.dataframe th ~ td,\n.bd-content .nboutput .output_area.rendered_html table.dataframe td ~ th,\n.bd-content .nboutput .output_area.rendered_html table.dataframe td ~ td,\n.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe th ~ th,\n.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe th ~ td,\n.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe td ~ th,\n.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe td ~ td {\n  border-left: 1px solid var(--pst-color-table-inner-border);\n}\n.bd-content .nboutput .output_area.rendered_html table.dataframe thead tr,\n.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe thead tr {\n  background-color: var(--pst-color-table-heading-bg);\n  border-bottom: 2px solid var(--pst-color-primary);\n}\n.bd-content .nboutput .output_area.rendered_html table.dataframe tbody tr:nth-child(odd),\n.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe tbody tr:nth-child(odd) {\n  background-color: var(--pst-color-table-row-zebra-low-bg);\n}\n.bd-content .nboutput .output_area.rendered_html table.dataframe tbody tr:nth-child(even),\n.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe tbody tr:nth-child(even) {\n  background-color: var(--pst-color-table-row-zebra-high-bg);\n}\n.bd-content .nboutput .output_area.rendered_html table.dataframe tbody tr:hover,\n.bd-content .nboutput .output_area .jp-RenderedHTMLCommon table.dataframe tbody tr:hover {\n  background-color: var(--pst-color-table-row-hover-bg);\n}\nhtml[data-theme=dark] .bd-content .nboutput .output_area.rendered_html:not(:has(table.dataframe)),\nhtml[data-theme=dark] .bd-content .nboutput .output_area .widget-subarea {\n  color: var(--pst-color-on-background);\n  background-color: var(--pst-color-text-base);\n  border-radius: 0.25rem;\n  padding: 0.5rem;\n}\nhtml[data-theme=dark] .bd-content .nboutput .output_area.stderr {\n  background-color: var(--pst-color-danger);\n}\n\ndiv.nblast.container {\n  margin-bottom: 1rem;\n}\n\nhtml[data-theme=light] .nbinput.container .prompt pre {\n  color: #005b82;\n}\nhtml[data-theme=dark] .nbinput.container .prompt pre {\n  color: #00e0e0;\n}\n\nhtml[data-theme=light] .nboutput.container .prompt pre {\n  color: #a12236;\n}\nhtml[data-theme=dark] .nboutput.container .prompt pre {\n  color: #ffa07a;\n}\n\n/*******************************************************************************\n * myst NB\n */\ndiv.cell_output .output {\n  max-width: 100%;\n  overflow-x: auto;\n}\n\n.bd-content div.cell_output table.dataframe {\n  color: var(--pst-color-table);\n  border: 1px solid var(--pst-color-table-outer-border);\n}\n.bd-content div.cell_output table.dataframe th ~ th,\n.bd-content div.cell_output table.dataframe th ~ td,\n.bd-content div.cell_output table.dataframe td ~ th,\n.bd-content div.cell_output table.dataframe td ~ td {\n  border-left: 1px solid var(--pst-color-table-inner-border);\n}\n.bd-content div.cell_output table.dataframe thead tr {\n  background-color: var(--pst-color-table-heading-bg);\n  border-bottom: 2px solid var(--pst-color-primary);\n}\n.bd-content div.cell_output table.dataframe tbody tr:nth-child(odd) {\n  background-color: var(--pst-color-table-row-zebra-low-bg);\n}\n.bd-content div.cell_output table.dataframe tbody tr:nth-child(even) {\n  background-color: var(--pst-color-table-row-zebra-high-bg);\n}\n.bd-content div.cell_output table.dataframe tbody tr:hover {\n  background-color: var(--pst-color-table-row-hover-bg);\n}\nhtml[data-theme=dark] .bd-content div.cell_output img,\nhtml[data-theme=dark] .bd-content div.cell_output .text_html:not(:has(table.dataframe)),\nhtml[data-theme=dark] .bd-content div.cell_output .widget-subarea {\n  color: var(--pst-color-on-background);\n  background-color: var(--pst-color-text-base);\n  border-radius: 0.25rem;\n  padding: 0.5rem;\n}\n\n.bd-content div.cell_input {\n  display: flex;\n  flex-direction: column;\n  justify-content: stretch;\n}\n.bd-content div.cell_input,\n.bd-content div.output {\n  border-radius: 0.25rem;\n}\n.bd-content div.output table {\n  table-layout: auto;\n}\n\n/**\n * style for the various mapping libs based on leaflet (folium, geemap, ipyleaflet)\n * mainly ensure the good display of the maps in both themes and avoid the customization\n * of the tiles\n */\n/**\n * avoid border override from pydata-sphinx-theme\n * minimal selctor to get the priority\n */\nhtml[data-theme=dark] .bd-content img.leaflet-tile.leaflet-tile-loaded {\n  border-radius: 0;\n  padding: 0;\n}\n\n/**\n * /search.html page special-cases\n */\n.bd-search-container {\n  /*******************************************\n  * Search results\n  */\n}\n.bd-search-container div#search-results > h2, .bd-search-container div#search-results > .h2 {\n  font-size: var(--pst-font-size-icon);\n  margin-top: 1rem;\n}\n.bd-search-container div#search-results p.search-summary {\n  color: var(--pst-color-text-muted);\n}\n.bd-search-container ul.search {\n  margin: 0;\n  list-style: none;\n}\n.bd-search-container ul.search li {\n  background-image: none;\n  padding: 1rem 0;\n  margin: 1rem 0;\n  border-top: 1px solid var(--pst-color-text-muted);\n}\n.bd-search-container ul.search li > a {\n  font-size: 1.2em;\n}\n.bd-search-container ul.search li div.context,\n.bd-search-container ul.search li p.context {\n  color: var(--pst-color-text-base);\n  margin: 0.5em 0 0;\n}\n.bd-search-container ul.search li div.context a::before,\n.bd-search-container ul.search li p.context a::before {\n  content: \"#\";\n  padding-right: 0.2em;\n  color: var(--pst-color-text-muted);\n}","@mixin bsBanner($file) {\n  /*!\n   * Bootstrap #{$file} v5.3.3 (https://getbootstrap.com/)\n   * Copyright 2011-2024 The Bootstrap Authors\n   * Licensed under MIT (https://github.com/twbs/bootstrap/blob/main/LICENSE)\n   */\n}\n",":root,\n[data-bs-theme=\"light\"] {\n  // Note: Custom variable values only support SassScript inside `#{}`.\n\n  // Colors\n  //\n  // Generate palettes for full colors, grays, and theme colors.\n\n  @each $color, $value in $colors {\n    --#{$prefix}#{$color}: #{$value};\n  }\n\n  @each $color, $value in $grays {\n    --#{$prefix}gray-#{$color}: #{$value};\n  }\n\n  @each $color, $value in $theme-colors {\n    --#{$prefix}#{$color}: #{$value};\n  }\n\n  @each $color, $value in $theme-colors-rgb {\n    --#{$prefix}#{$color}-rgb: #{$value};\n  }\n\n  @each $color, $value in $theme-colors-text {\n    --#{$prefix}#{$color}-text-emphasis: #{$value};\n  }\n\n  @each $color, $value in $theme-colors-bg-subtle {\n    --#{$prefix}#{$color}-bg-subtle: #{$value};\n  }\n\n  @each $color, $value in $theme-colors-border-subtle {\n    --#{$prefix}#{$color}-border-subtle: #{$value};\n  }\n\n  --#{$prefix}white-rgb: #{to-rgb($white)};\n  --#{$prefix}black-rgb: #{to-rgb($black)};\n\n  // Fonts\n\n  // Note: Use `inspect` for lists so that quoted items keep the quotes.\n  // See https://github.com/sass/sass/issues/2383#issuecomment-336349172\n  --#{$prefix}font-sans-serif: #{inspect($font-family-sans-serif)};\n  --#{$prefix}font-monospace: #{inspect($font-family-monospace)};\n  --#{$prefix}gradient: #{$gradient};\n\n  // Root and body\n  // scss-docs-start root-body-variables\n  @if $font-size-root != null {\n    --#{$prefix}root-font-size: #{$font-size-root};\n  }\n  --#{$prefix}body-font-family: #{inspect($font-family-base)};\n  @include rfs($font-size-base, --#{$prefix}body-font-size);\n  --#{$prefix}body-font-weight: #{$font-weight-base};\n  --#{$prefix}body-line-height: #{$line-height-base};\n  @if $body-text-align != null {\n    --#{$prefix}body-text-align: #{$body-text-align};\n  }\n\n  --#{$prefix}body-color: #{$body-color};\n  --#{$prefix}body-color-rgb: #{to-rgb($body-color)};\n  --#{$prefix}body-bg: #{$body-bg};\n  --#{$prefix}body-bg-rgb: #{to-rgb($body-bg)};\n\n  --#{$prefix}emphasis-color: #{$body-emphasis-color};\n  --#{$prefix}emphasis-color-rgb: #{to-rgb($body-emphasis-color)};\n\n  --#{$prefix}secondary-color: #{$body-secondary-color};\n  --#{$prefix}secondary-color-rgb: #{to-rgb($body-secondary-color)};\n  --#{$prefix}secondary-bg: #{$body-secondary-bg};\n  --#{$prefix}secondary-bg-rgb: #{to-rgb($body-secondary-bg)};\n\n  --#{$prefix}tertiary-color: #{$body-tertiary-color};\n  --#{$prefix}tertiary-color-rgb: #{to-rgb($body-tertiary-color)};\n  --#{$prefix}tertiary-bg: #{$body-tertiary-bg};\n  --#{$prefix}tertiary-bg-rgb: #{to-rgb($body-tertiary-bg)};\n  // scss-docs-end root-body-variables\n\n  --#{$prefix}heading-color: #{$headings-color};\n\n  --#{$prefix}link-color: #{$link-color};\n  --#{$prefix}link-color-rgb: #{to-rgb($link-color)};\n  --#{$prefix}link-decoration: #{$link-decoration};\n\n  --#{$prefix}link-hover-color: #{$link-hover-color};\n  --#{$prefix}link-hover-color-rgb: #{to-rgb($link-hover-color)};\n\n  @if $link-hover-decoration != null {\n    --#{$prefix}link-hover-decoration: #{$link-hover-decoration};\n  }\n\n  --#{$prefix}code-color: #{$code-color};\n  --#{$prefix}highlight-color: #{$mark-color};\n  --#{$prefix}highlight-bg: #{$mark-bg};\n\n  // scss-docs-start root-border-var\n  --#{$prefix}border-width: #{$border-width};\n  --#{$prefix}border-style: #{$border-style};\n  --#{$prefix}border-color: #{$border-color};\n  --#{$prefix}border-color-translucent: #{$border-color-translucent};\n\n  --#{$prefix}border-radius: #{$border-radius};\n  --#{$prefix}border-radius-sm: #{$border-radius-sm};\n  --#{$prefix}border-radius-lg: #{$border-radius-lg};\n  --#{$prefix}border-radius-xl: #{$border-radius-xl};\n  --#{$prefix}border-radius-xxl: #{$border-radius-xxl};\n  --#{$prefix}border-radius-2xl: var(--#{$prefix}border-radius-xxl); // Deprecated in v5.3.0 for consistency\n  --#{$prefix}border-radius-pill: #{$border-radius-pill};\n  // scss-docs-end root-border-var\n\n  --#{$prefix}box-shadow: #{$box-shadow};\n  --#{$prefix}box-shadow-sm: #{$box-shadow-sm};\n  --#{$prefix}box-shadow-lg: #{$box-shadow-lg};\n  --#{$prefix}box-shadow-inset: #{$box-shadow-inset};\n\n  // Focus styles\n  // scss-docs-start root-focus-variables\n  --#{$prefix}focus-ring-width: #{$focus-ring-width};\n  --#{$prefix}focus-ring-opacity: #{$focus-ring-opacity};\n  --#{$prefix}focus-ring-color: #{$focus-ring-color};\n  // scss-docs-end root-focus-variables\n\n  // scss-docs-start root-form-validation-variables\n  --#{$prefix}form-valid-color: #{$form-valid-color};\n  --#{$prefix}form-valid-border-color: #{$form-valid-border-color};\n  --#{$prefix}form-invalid-color: #{$form-invalid-color};\n  --#{$prefix}form-invalid-border-color: #{$form-invalid-border-color};\n  // scss-docs-end root-form-validation-variables\n}\n\n@if $enable-dark-mode {\n  @include color-mode(dark, true) {\n    color-scheme: dark;\n\n    // scss-docs-start root-dark-mode-vars\n    --#{$prefix}body-color: #{$body-color-dark};\n    --#{$prefix}body-color-rgb: #{to-rgb($body-color-dark)};\n    --#{$prefix}body-bg: #{$body-bg-dark};\n    --#{$prefix}body-bg-rgb: #{to-rgb($body-bg-dark)};\n\n    --#{$prefix}emphasis-color: #{$body-emphasis-color-dark};\n    --#{$prefix}emphasis-color-rgb: #{to-rgb($body-emphasis-color-dark)};\n\n    --#{$prefix}secondary-color: #{$body-secondary-color-dark};\n    --#{$prefix}secondary-color-rgb: #{to-rgb($body-secondary-color-dark)};\n    --#{$prefix}secondary-bg: #{$body-secondary-bg-dark};\n    --#{$prefix}secondary-bg-rgb: #{to-rgb($body-secondary-bg-dark)};\n\n    --#{$prefix}tertiary-color: #{$body-tertiary-color-dark};\n    --#{$prefix}tertiary-color-rgb: #{to-rgb($body-tertiary-color-dark)};\n    --#{$prefix}tertiary-bg: #{$body-tertiary-bg-dark};\n    --#{$prefix}tertiary-bg-rgb: #{to-rgb($body-tertiary-bg-dark)};\n\n    @each $color, $value in $theme-colors-text-dark {\n      --#{$prefix}#{$color}-text-emphasis: #{$value};\n    }\n\n    @each $color, $value in $theme-colors-bg-subtle-dark {\n      --#{$prefix}#{$color}-bg-subtle: #{$value};\n    }\n\n    @each $color, $value in $theme-colors-border-subtle-dark {\n      --#{$prefix}#{$color}-border-subtle: #{$value};\n    }\n\n    --#{$prefix}heading-color: #{$headings-color-dark};\n\n    --#{$prefix}link-color: #{$link-color-dark};\n    --#{$prefix}link-hover-color: #{$link-hover-color-dark};\n    --#{$prefix}link-color-rgb: #{to-rgb($link-color-dark)};\n    --#{$prefix}link-hover-color-rgb: #{to-rgb($link-hover-color-dark)};\n\n    --#{$prefix}code-color: #{$code-color-dark};\n    --#{$prefix}highlight-color: #{$mark-color-dark};\n    --#{$prefix}highlight-bg: #{$mark-bg-dark};\n\n    --#{$prefix}border-color: #{$border-color-dark};\n    --#{$prefix}border-color-translucent: #{$border-color-translucent-dark};\n\n    --#{$prefix}form-valid-color: #{$form-valid-color-dark};\n    --#{$prefix}form-valid-border-color: #{$form-valid-border-color-dark};\n    --#{$prefix}form-invalid-color: #{$form-invalid-color-dark};\n    --#{$prefix}form-invalid-border-color: #{$form-invalid-border-color-dark};\n    // scss-docs-end root-dark-mode-vars\n  }\n}\n","// stylelint-disable scss/dimension-no-non-numeric-values\n\n// SCSS RFS mixin\n//\n// Automated responsive values for font sizes, paddings, margins and much more\n//\n// Licensed under MIT (https://github.com/twbs/rfs/blob/main/LICENSE)\n\n// Configuration\n\n// Base value\n$rfs-base-value: 1.25rem !default;\n$rfs-unit: rem !default;\n\n@if $rfs-unit != rem and $rfs-unit != px {\n  @error \"`#{$rfs-unit}` is not a valid unit for $rfs-unit. Use `px` or `rem`.\";\n}\n\n// Breakpoint at where values start decreasing if screen width is smaller\n$rfs-breakpoint: 1200px !default;\n$rfs-breakpoint-unit: px !default;\n\n@if $rfs-breakpoint-unit != px and $rfs-breakpoint-unit != em and $rfs-breakpoint-unit != rem {\n  @error \"`#{$rfs-breakpoint-unit}` is not a valid unit for $rfs-breakpoint-unit. Use `px`, `em` or `rem`.\";\n}\n\n// Resize values based on screen height and width\n$rfs-two-dimensional: false !default;\n\n// Factor of decrease\n$rfs-factor: 10 !default;\n\n@if type-of($rfs-factor) != number or $rfs-factor <= 1 {\n  @error \"`#{$rfs-factor}` is not a valid  $rfs-factor, it must be greater than 1.\";\n}\n\n// Mode. Possibilities: \"min-media-query\", \"max-media-query\"\n$rfs-mode: min-media-query !default;\n\n// Generate enable or disable classes. Possibilities: false, \"enable\" or \"disable\"\n$rfs-class: false !default;\n\n// 1 rem = $rfs-rem-value px\n$rfs-rem-value: 16 !default;\n\n// Safari iframe resize bug: https://github.com/twbs/rfs/issues/14\n$rfs-safari-iframe-resize-bug-fix: false !default;\n\n// Disable RFS by setting $enable-rfs to false\n$enable-rfs: true !default;\n\n// Cache $rfs-base-value unit\n$rfs-base-value-unit: unit($rfs-base-value);\n\n@function divide($dividend, $divisor, $precision: 10) {\n  $sign: if($dividend > 0 and $divisor > 0 or $dividend < 0 and $divisor < 0, 1, -1);\n  $dividend: abs($dividend);\n  $divisor: abs($divisor);\n  @if $dividend == 0 {\n    @return 0;\n  }\n  @if $divisor == 0 {\n    @error \"Cannot divide by 0\";\n  }\n  $remainder: $dividend;\n  $result: 0;\n  $factor: 10;\n  @while ($remainder > 0 and $precision >= 0) {\n    $quotient: 0;\n    @while ($remainder >= $divisor) {\n      $remainder: $remainder - $divisor;\n      $quotient: $quotient + 1;\n    }\n    $result: $result * 10 + $quotient;\n    $factor: $factor * .1;\n    $remainder: $remainder * 10;\n    $precision: $precision - 1;\n    @if ($precision < 0 and $remainder >= $divisor * 5) {\n      $result: $result + 1;\n    }\n  }\n  $result: $result * $factor * $sign;\n  $dividend-unit: unit($dividend);\n  $divisor-unit: unit($divisor);\n  $unit-map: (\n    \"px\": 1px,\n    \"rem\": 1rem,\n    \"em\": 1em,\n    \"%\": 1%\n  );\n  @if ($dividend-unit != $divisor-unit and map-has-key($unit-map, $dividend-unit)) {\n    $result: $result * map-get($unit-map, $dividend-unit);\n  }\n  @return $result;\n}\n\n// Remove px-unit from $rfs-base-value for calculations\n@if $rfs-base-value-unit == px {\n  $rfs-base-value: divide($rfs-base-value, $rfs-base-value * 0 + 1);\n}\n@else if $rfs-base-value-unit == rem {\n  $rfs-base-value: divide($rfs-base-value, divide($rfs-base-value * 0 + 1, $rfs-rem-value));\n}\n\n// Cache $rfs-breakpoint unit to prevent multiple calls\n$rfs-breakpoint-unit-cache: unit($rfs-breakpoint);\n\n// Remove unit from $rfs-breakpoint for calculations\n@if $rfs-breakpoint-unit-cache == px {\n  $rfs-breakpoint: divide($rfs-breakpoint, $rfs-breakpoint * 0 + 1);\n}\n@else if $rfs-breakpoint-unit-cache == rem or $rfs-breakpoint-unit-cache == \"em\" {\n  $rfs-breakpoint: divide($rfs-breakpoint, divide($rfs-breakpoint * 0 + 1, $rfs-rem-value));\n}\n\n// Calculate the media query value\n$rfs-mq-value: if($rfs-breakpoint-unit == px, #{$rfs-breakpoint}px, #{divide($rfs-breakpoint, $rfs-rem-value)}#{$rfs-breakpoint-unit});\n$rfs-mq-property-width: if($rfs-mode == max-media-query, max-width, min-width);\n$rfs-mq-property-height: if($rfs-mode == max-media-query, max-height, min-height);\n\n// Internal mixin used to determine which media query needs to be used\n@mixin _rfs-media-query {\n  @if $rfs-two-dimensional {\n    @if $rfs-mode == max-media-query {\n      @media (#{$rfs-mq-property-width}: #{$rfs-mq-value}), (#{$rfs-mq-property-height}: #{$rfs-mq-value}) {\n        @content;\n      }\n    }\n    @else {\n      @media (#{$rfs-mq-property-width}: #{$rfs-mq-value}) and (#{$rfs-mq-property-height}: #{$rfs-mq-value}) {\n        @content;\n      }\n    }\n  }\n  @else {\n    @media (#{$rfs-mq-property-width}: #{$rfs-mq-value}) {\n      @content;\n    }\n  }\n}\n\n// Internal mixin that adds disable classes to the selector if needed.\n@mixin _rfs-rule {\n  @if $rfs-class == disable and $rfs-mode == max-media-query {\n    // Adding an extra class increases specificity, which prevents the media query to override the property\n    &,\n    .disable-rfs &,\n    &.disable-rfs {\n      @content;\n    }\n  }\n  @else if $rfs-class == enable and $rfs-mode == min-media-query {\n    .enable-rfs &,\n    &.enable-rfs {\n      @content;\n    }\n  } @else {\n    @content;\n  }\n}\n\n// Internal mixin that adds enable classes to the selector if needed.\n@mixin _rfs-media-query-rule {\n\n  @if $rfs-class == enable {\n    @if $rfs-mode == min-media-query {\n      @content;\n    }\n\n    @include _rfs-media-query () {\n      .enable-rfs &,\n      &.enable-rfs {\n        @content;\n      }\n    }\n  }\n  @else {\n    @if $rfs-class == disable and $rfs-mode == min-media-query {\n      .disable-rfs &,\n      &.disable-rfs {\n        @content;\n      }\n    }\n    @include _rfs-media-query () {\n      @content;\n    }\n  }\n}\n\n// Helper function to get the formatted non-responsive value\n@function rfs-value($values) {\n  // Convert to list\n  $values: if(type-of($values) != list, ($values,), $values);\n\n  $val: \"\";\n\n  // Loop over each value and calculate value\n  @each $value in $values {\n    @if $value == 0 {\n      $val: $val + \" 0\";\n    }\n    @else {\n      // Cache $value unit\n      $unit: if(type-of($value) == \"number\", unit($value), false);\n\n      @if $unit == px {\n        // Convert to rem if needed\n        $val: $val + \" \" + if($rfs-unit == rem, #{divide($value, $value * 0 + $rfs-rem-value)}rem, $value);\n      }\n      @else if $unit == rem {\n        // Convert to px if needed\n        $val: $val + \" \" + if($rfs-unit == px, #{divide($value, $value * 0 + 1) * $rfs-rem-value}px, $value);\n      } @else {\n        // If $value isn't a number (like inherit) or $value has a unit (not px or rem, like 1.5em) or $ is 0, just print the value\n        $val: $val + \" \" + $value;\n      }\n    }\n  }\n\n  // Remove first space\n  @return unquote(str-slice($val, 2));\n}\n\n// Helper function to get the responsive value calculated by RFS\n@function rfs-fluid-value($values) {\n  // Convert to list\n  $values: if(type-of($values) != list, ($values,), $values);\n\n  $val: \"\";\n\n  // Loop over each value and calculate value\n  @each $value in $values {\n    @if $value == 0 {\n      $val: $val + \" 0\";\n    } @else {\n      // Cache $value unit\n      $unit: if(type-of($value) == \"number\", unit($value), false);\n\n      // If $value isn't a number (like inherit) or $value has a unit (not px or rem, like 1.5em) or $ is 0, just print the value\n      @if not $unit or $unit != px and $unit != rem {\n        $val: $val + \" \" + $value;\n      } @else {\n        // Remove unit from $value for calculations\n        $value: divide($value, $value * 0 + if($unit == px, 1, divide(1, $rfs-rem-value)));\n\n        // Only add the media query if the value is greater than the minimum value\n        @if abs($value) <= $rfs-base-value or not $enable-rfs {\n          $val: $val + \" \" + if($rfs-unit == rem, #{divide($value, $rfs-rem-value)}rem, #{$value}px);\n        }\n        @else {\n          // Calculate the minimum value\n          $value-min: $rfs-base-value + divide(abs($value) - $rfs-base-value, $rfs-factor);\n\n          // Calculate difference between $value and the minimum value\n          $value-diff: abs($value) - $value-min;\n\n          // Base value formatting\n          $min-width: if($rfs-unit == rem, #{divide($value-min, $rfs-rem-value)}rem, #{$value-min}px);\n\n          // Use negative value if needed\n          $min-width: if($value < 0, -$min-width, $min-width);\n\n          // Use `vmin` if two-dimensional is enabled\n          $variable-unit: if($rfs-two-dimensional, vmin, vw);\n\n          // Calculate the variable width between 0 and $rfs-breakpoint\n          $variable-width: #{divide($value-diff * 100, $rfs-breakpoint)}#{$variable-unit};\n\n          // Return the calculated value\n          $val: $val + \" calc(\" + $min-width + if($value < 0, \" - \", \" + \") + $variable-width + \")\";\n        }\n      }\n    }\n  }\n\n  // Remove first space\n  @return unquote(str-slice($val, 2));\n}\n\n// RFS mixin\n@mixin rfs($values, $property: font-size) {\n  @if $values != null {\n    $val: rfs-value($values);\n    $fluid-val: rfs-fluid-value($values);\n\n    // Do not print the media query if responsive & non-responsive values are the same\n    @if $val == $fluid-val {\n      #{$property}: $val;\n    }\n    @else {\n      @include _rfs-rule () {\n        #{$property}: if($rfs-mode == max-media-query, $val, $fluid-val);\n\n        // Include safari iframe resize fix if needed\n        min-width: if($rfs-safari-iframe-resize-bug-fix, (0 * 1vw), null);\n      }\n\n      @include _rfs-media-query-rule () {\n        #{$property}: if($rfs-mode == max-media-query, $fluid-val, $val);\n      }\n    }\n  }\n}\n\n// Shorthand helper mixins\n@mixin font-size($value) {\n  @include rfs($value);\n}\n\n@mixin padding($value) {\n  @include rfs($value, padding);\n}\n\n@mixin padding-top($value) {\n  @include rfs($value, padding-top);\n}\n\n@mixin padding-right($value) {\n  @include rfs($value, padding-right);\n}\n\n@mixin padding-bottom($value) {\n  @include rfs($value, padding-bottom);\n}\n\n@mixin padding-left($value) {\n  @include rfs($value, padding-left);\n}\n\n@mixin margin($value) {\n  @include rfs($value, margin);\n}\n\n@mixin margin-top($value) {\n  @include rfs($value, margin-top);\n}\n\n@mixin margin-right($value) {\n  @include rfs($value, margin-right);\n}\n\n@mixin margin-bottom($value) {\n  @include rfs($value, margin-bottom);\n}\n\n@mixin margin-left($value) {\n  @include rfs($value, margin-left);\n}\n","// scss-docs-start color-mode-mixin\n@mixin color-mode($mode: light, $root: false) {\n  @if $color-mode-type == \"media-query\" {\n    @if $root == true {\n      @media (prefers-color-scheme: $mode) {\n        :root {\n          @content;\n        }\n      }\n    } @else {\n      @media (prefers-color-scheme: $mode) {\n        @content;\n      }\n    }\n  } @else {\n    [data-bs-theme=\"#{$mode}\"] {\n      @content;\n    }\n  }\n}\n// scss-docs-end color-mode-mixin\n","// stylelint-disable declaration-no-important, selector-no-qualifying-type, property-no-vendor-prefix\n\n\n// Reboot\n//\n// Normalization of HTML elements, manually forked from Normalize.css to remove\n// styles targeting irrelevant browsers while applying new styles.\n//\n// Normalize is licensed MIT. https://github.com/necolas/normalize.css\n\n\n// Document\n//\n// Change from `box-sizing: content-box` so that `width` is not affected by `padding` or `border`.\n\n*,\n*::before,\n*::after {\n  box-sizing: border-box;\n}\n\n\n// Root\n//\n// Ability to the value of the root font sizes, affecting the value of `rem`.\n// null by default, thus nothing is generated.\n\n:root {\n  @if $font-size-root != null {\n    @include font-size(var(--#{$prefix}root-font-size));\n  }\n\n  @if $enable-smooth-scroll {\n    @media (prefers-reduced-motion: no-preference) {\n      scroll-behavior: smooth;\n    }\n  }\n}\n\n\n// Body\n//\n// 1. Remove the margin in all browsers.\n// 2. As a best practice, apply a default `background-color`.\n// 3. Prevent adjustments of font size after orientation changes in iOS.\n// 4. Change the default tap highlight to be completely transparent in iOS.\n\n// scss-docs-start reboot-body-rules\nbody {\n  margin: 0; // 1\n  font-family: var(--#{$prefix}body-font-family);\n  @include font-size(var(--#{$prefix}body-font-size));\n  font-weight: var(--#{$prefix}body-font-weight);\n  line-height: var(--#{$prefix}body-line-height);\n  color: var(--#{$prefix}body-color);\n  text-align: var(--#{$prefix}body-text-align);\n  background-color: var(--#{$prefix}body-bg); // 2\n  -webkit-text-size-adjust: 100%; // 3\n  -webkit-tap-highlight-color: rgba($black, 0); // 4\n}\n// scss-docs-end reboot-body-rules\n\n\n// Content grouping\n//\n// 1. Reset Firefox's gray color\n\nhr {\n  margin: $hr-margin-y 0;\n  color: $hr-color; // 1\n  border: 0;\n  border-top: $hr-border-width solid $hr-border-color;\n  opacity: $hr-opacity;\n}\n\n\n// Typography\n//\n// 1. Remove top margins from headings\n//    By default, `<h1>`-`<h6>` all receive top and bottom margins. We nuke the top\n//    margin for easier control within type scales as it avoids margin collapsing.\n\n%heading {\n  margin-top: 0; // 1\n  margin-bottom: $headings-margin-bottom;\n  font-family: $headings-font-family;\n  font-style: $headings-font-style;\n  font-weight: $headings-font-weight;\n  line-height: $headings-line-height;\n  color: var(--#{$prefix}heading-color);\n}\n\nh1 {\n  @extend %heading;\n  @include font-size($h1-font-size);\n}\n\nh2 {\n  @extend %heading;\n  @include font-size($h2-font-size);\n}\n\nh3 {\n  @extend %heading;\n  @include font-size($h3-font-size);\n}\n\nh4 {\n  @extend %heading;\n  @include font-size($h4-font-size);\n}\n\nh5 {\n  @extend %heading;\n  @include font-size($h5-font-size);\n}\n\nh6 {\n  @extend %heading;\n  @include font-size($h6-font-size);\n}\n\n\n// Reset margins on paragraphs\n//\n// Similarly, the top margin on `<p>`s get reset. However, we also reset the\n// bottom margin to use `rem` units instead of `em`.\n\np {\n  margin-top: 0;\n  margin-bottom: $paragraph-margin-bottom;\n}\n\n\n// Abbreviations\n//\n// 1. Add the correct text decoration in Chrome, Edge, Opera, and Safari.\n// 2. Add explicit cursor to indicate changed behavior.\n// 3. Prevent the text-decoration to be skipped.\n\nabbr[title] {\n  text-decoration: underline dotted; // 1\n  cursor: help; // 2\n  text-decoration-skip-ink: none; // 3\n}\n\n\n// Address\n\naddress {\n  margin-bottom: 1rem;\n  font-style: normal;\n  line-height: inherit;\n}\n\n\n// Lists\n\nol,\nul {\n  padding-left: 2rem;\n}\n\nol,\nul,\ndl {\n  margin-top: 0;\n  margin-bottom: 1rem;\n}\n\nol ol,\nul ul,\nol ul,\nul ol {\n  margin-bottom: 0;\n}\n\ndt {\n  font-weight: $dt-font-weight;\n}\n\n// 1. Undo browser default\n\ndd {\n  margin-bottom: .5rem;\n  margin-left: 0; // 1\n}\n\n\n// Blockquote\n\nblockquote {\n  margin: 0 0 1rem;\n}\n\n\n// Strong\n//\n// Add the correct font weight in Chrome, Edge, and Safari\n\nb,\nstrong {\n  font-weight: $font-weight-bolder;\n}\n\n\n// Small\n//\n// Add the correct font size in all browsers\n\nsmall {\n  @include font-size($small-font-size);\n}\n\n\n// Mark\n\nmark {\n  padding: $mark-padding;\n  color: var(--#{$prefix}highlight-color);\n  background-color: var(--#{$prefix}highlight-bg);\n}\n\n\n// Sub and Sup\n//\n// Prevent `sub` and `sup` elements from affecting the line height in\n// all browsers.\n\nsub,\nsup {\n  position: relative;\n  @include font-size($sub-sup-font-size);\n  line-height: 0;\n  vertical-align: baseline;\n}\n\nsub { bottom: -.25em; }\nsup { top: -.5em; }\n\n\n// Links\n\na {\n  color: rgba(var(--#{$prefix}link-color-rgb), var(--#{$prefix}link-opacity, 1));\n  text-decoration: $link-decoration;\n\n  &:hover {\n    --#{$prefix}link-color-rgb: var(--#{$prefix}link-hover-color-rgb);\n    text-decoration: $link-hover-decoration;\n  }\n}\n\n// And undo these styles for placeholder links/named anchors (without href).\n// It would be more straightforward to just use a[href] in previous block, but that\n// causes specificity issues in many other styles that are too complex to fix.\n// See https://github.com/twbs/bootstrap/issues/19402\n\na:not([href]):not([class]) {\n  &,\n  &:hover {\n    color: inherit;\n    text-decoration: none;\n  }\n}\n\n\n// Code\n\npre,\ncode,\nkbd,\nsamp {\n  font-family: $font-family-code;\n  @include font-size(1em); // Correct the odd `em` font sizing in all browsers.\n}\n\n// 1. Remove browser default top margin\n// 2. Reset browser default of `1em` to use `rem`s\n// 3. Don't allow content to break outside\n\npre {\n  display: block;\n  margin-top: 0; // 1\n  margin-bottom: 1rem; // 2\n  overflow: auto; // 3\n  @include font-size($code-font-size);\n  color: $pre-color;\n\n  // Account for some code outputs that place code tags in pre tags\n  code {\n    @include font-size(inherit);\n    color: inherit;\n    word-break: normal;\n  }\n}\n\ncode {\n  @include font-size($code-font-size);\n  color: var(--#{$prefix}code-color);\n  word-wrap: break-word;\n\n  // Streamline the style when inside anchors to avoid broken underline and more\n  a > & {\n    color: inherit;\n  }\n}\n\nkbd {\n  padding: $kbd-padding-y $kbd-padding-x;\n  @include font-size($kbd-font-size);\n  color: $kbd-color;\n  background-color: $kbd-bg;\n  @include border-radius($border-radius-sm);\n\n  kbd {\n    padding: 0;\n    @include font-size(1em);\n    font-weight: $nested-kbd-font-weight;\n  }\n}\n\n\n// Figures\n//\n// Apply a consistent margin strategy (matches our type styles).\n\nfigure {\n  margin: 0 0 1rem;\n}\n\n\n// Images and content\n\nimg,\nsvg {\n  vertical-align: middle;\n}\n\n\n// Tables\n//\n// Prevent double borders\n\ntable {\n  caption-side: bottom;\n  border-collapse: collapse;\n}\n\ncaption {\n  padding-top: $table-cell-padding-y;\n  padding-bottom: $table-cell-padding-y;\n  color: $table-caption-color;\n  text-align: left;\n}\n\n// 1. Removes font-weight bold by inheriting\n// 2. Matches default `<td>` alignment by inheriting `text-align`.\n// 3. Fix alignment for Safari\n\nth {\n  font-weight: $table-th-font-weight; // 1\n  text-align: inherit; // 2\n  text-align: -webkit-match-parent; // 3\n}\n\nthead,\ntbody,\ntfoot,\ntr,\ntd,\nth {\n  border-color: inherit;\n  border-style: solid;\n  border-width: 0;\n}\n\n\n// Forms\n//\n// 1. Allow labels to use `margin` for spacing.\n\nlabel {\n  display: inline-block; // 1\n}\n\n// Remove the default `border-radius` that macOS Chrome adds.\n// See https://github.com/twbs/bootstrap/issues/24093\n\nbutton {\n  // stylelint-disable-next-line property-disallowed-list\n  border-radius: 0;\n}\n\n// Explicitly remove focus outline in Chromium when it shouldn't be\n// visible (e.g. as result of mouse click or touch tap). It already\n// should be doing this automatically, but seems to currently be\n// confused and applies its very visible two-tone outline anyway.\n\nbutton:focus:not(:focus-visible) {\n  outline: 0;\n}\n\n// 1. Remove the margin in Firefox and Safari\n\ninput,\nbutton,\nselect,\noptgroup,\ntextarea {\n  margin: 0; // 1\n  font-family: inherit;\n  @include font-size(inherit);\n  line-height: inherit;\n}\n\n// Remove the inheritance of text transform in Firefox\nbutton,\nselect {\n  text-transform: none;\n}\n// Set the cursor for non-`<button>` buttons\n//\n// Details at https://github.com/twbs/bootstrap/pull/30562\n[role=\"button\"] {\n  cursor: pointer;\n}\n\nselect {\n  // Remove the inheritance of word-wrap in Safari.\n  // See https://github.com/twbs/bootstrap/issues/24990\n  word-wrap: normal;\n\n  // Undo the opacity change from Chrome\n  &:disabled {\n    opacity: 1;\n  }\n}\n\n// Remove the dropdown arrow only from text type inputs built with datalists in Chrome.\n// See https://stackoverflow.com/a/54997118\n\n[list]:not([type=\"date\"]):not([type=\"datetime-local\"]):not([type=\"month\"]):not([type=\"week\"]):not([type=\"time\"])::-webkit-calendar-picker-indicator {\n  display: none !important;\n}\n\n// 1. Prevent a WebKit bug where (2) destroys native `audio` and `video`\n//    controls in Android 4.\n// 2. Correct the inability to style clickable types in iOS and Safari.\n// 3. Opinionated: add \"hand\" cursor to non-disabled button elements.\n\nbutton,\n[type=\"button\"], // 1\n[type=\"reset\"],\n[type=\"submit\"] {\n  -webkit-appearance: button; // 2\n\n  @if $enable-button-pointers {\n    &:not(:disabled) {\n      cursor: pointer; // 3\n    }\n  }\n}\n\n// Remove inner border and padding from Firefox, but don't restore the outline like Normalize.\n\n::-moz-focus-inner {\n  padding: 0;\n  border-style: none;\n}\n\n// 1. Textareas should really only resize vertically so they don't break their (horizontal) containers.\n\ntextarea {\n  resize: vertical; // 1\n}\n\n// 1. Browsers set a default `min-width: min-content;` on fieldsets,\n//    unlike e.g. `<div>`s, which have `min-width: 0;` by default.\n//    So we reset that to ensure fieldsets behave more like a standard block element.\n//    See https://github.com/twbs/bootstrap/issues/12359\n//    and https://html.spec.whatwg.org/multipage/#the-fieldset-and-legend-elements\n// 2. Reset the default outline behavior of fieldsets so they don't affect page layout.\n\nfieldset {\n  min-width: 0; // 1\n  padding: 0; // 2\n  margin: 0; // 2\n  border: 0; // 2\n}\n\n// 1. By using `float: left`, the legend will behave like a block element.\n//    This way the border of a fieldset wraps around the legend if present.\n// 2. Fix wrapping bug.\n//    See https://github.com/twbs/bootstrap/issues/29712\n\nlegend {\n  float: left; // 1\n  width: 100%;\n  padding: 0;\n  margin-bottom: $legend-margin-bottom;\n  @include font-size($legend-font-size);\n  font-weight: $legend-font-weight;\n  line-height: inherit;\n\n  + * {\n    clear: left; // 2\n  }\n}\n\n// Fix height of inputs with a type of datetime-local, date, month, week, or time\n// See https://github.com/twbs/bootstrap/issues/18842\n\n::-webkit-datetime-edit-fields-wrapper,\n::-webkit-datetime-edit-text,\n::-webkit-datetime-edit-minute,\n::-webkit-datetime-edit-hour-field,\n::-webkit-datetime-edit-day-field,\n::-webkit-datetime-edit-month-field,\n::-webkit-datetime-edit-year-field {\n  padding: 0;\n}\n\n::-webkit-inner-spin-button {\n  height: auto;\n}\n\n// 1. This overrides the extra rounded corners on search inputs in iOS so that our\n//    `.form-control` class can properly style them. Note that this cannot simply\n//    be added to `.form-control` as it's not specific enough. For details, see\n//    https://github.com/twbs/bootstrap/issues/11586.\n// 2. Correct the outline style in Safari.\n\n[type=\"search\"] {\n  -webkit-appearance: textfield; // 1\n  outline-offset: -2px; // 2\n}\n\n// 1. A few input types should stay LTR\n// See https://rtlstyling.com/posts/rtl-styling#form-inputs\n// 2. RTL only output\n// See https://rtlcss.com/learn/usage-guide/control-directives/#raw\n\n/* rtl:raw:\n[type=\"tel\"],\n[type=\"url\"],\n[type=\"email\"],\n[type=\"number\"] {\n  direction: ltr;\n}\n*/\n\n// Remove the inner padding in Chrome and Safari on macOS.\n\n::-webkit-search-decoration {\n  -webkit-appearance: none;\n}\n\n// Remove padding around color pickers in webkit browsers\n\n::-webkit-color-swatch-wrapper {\n  padding: 0;\n}\n\n\n// 1. Inherit font family and line height for file input buttons\n// 2. Correct the inability to style clickable types in iOS and Safari.\n\n::file-selector-button {\n  font: inherit; // 1\n  -webkit-appearance: button; // 2\n}\n\n// Correct element displays\n\noutput {\n  display: inline-block;\n}\n\n// Remove border from iframe\n\niframe {\n  border: 0;\n}\n\n// Summary\n//\n// 1. Add the correct display in all browsers\n\nsummary {\n  display: list-item; // 1\n  cursor: pointer;\n}\n\n\n// Progress\n//\n// Add the correct vertical alignment in Chrome, Firefox, and Opera.\n\nprogress {\n  vertical-align: baseline;\n}\n\n\n// Hidden attribute\n//\n// Always hide an element with the `hidden` HTML attribute.\n\n[hidden] {\n  display: none !important;\n}\n","// Variables\n//\n// Variables should follow the `$component-state-property-size` formula for\n// consistent naming. Ex: $nav-link-disabled-color and $modal-content-box-shadow-xs.\n\n// Color system\n\n// scss-docs-start gray-color-variables\n$white:    #fff !default;\n$gray-100: #f8f9fa !default;\n$gray-200: #e9ecef !default;\n$gray-300: #dee2e6 !default;\n$gray-400: #ced4da !default;\n$gray-500: #adb5bd !default;\n$gray-600: #6c757d !default;\n$gray-700: #495057 !default;\n$gray-800: #343a40 !default;\n$gray-900: #212529 !default;\n$black:    #000 !default;\n// scss-docs-end gray-color-variables\n\n// fusv-disable\n// scss-docs-start gray-colors-map\n$grays: (\n  \"100\": $gray-100,\n  \"200\": $gray-200,\n  \"300\": $gray-300,\n  \"400\": $gray-400,\n  \"500\": $gray-500,\n  \"600\": $gray-600,\n  \"700\": $gray-700,\n  \"800\": $gray-800,\n  \"900\": $gray-900\n) !default;\n// scss-docs-end gray-colors-map\n// fusv-enable\n\n// scss-docs-start color-variables\n$blue:    #0d6efd !default;\n$indigo:  #6610f2 !default;\n$purple:  #6f42c1 !default;\n$pink:    #d63384 !default;\n$red:     #dc3545 !default;\n$orange:  #fd7e14 !default;\n$yellow:  #ffc107 !default;\n$green:   #198754 !default;\n$teal:    #20c997 !default;\n$cyan:    #0dcaf0 !default;\n// scss-docs-end color-variables\n\n// scss-docs-start colors-map\n$colors: (\n  \"blue\":       $blue,\n  \"indigo\":     $indigo,\n  \"purple\":     $purple,\n  \"pink\":       $pink,\n  \"red\":        $red,\n  \"orange\":     $orange,\n  \"yellow\":     $yellow,\n  \"green\":      $green,\n  \"teal\":       $teal,\n  \"cyan\":       $cyan,\n  \"black\":      $black,\n  \"white\":      $white,\n  \"gray\":       $gray-600,\n  \"gray-dark\":  $gray-800\n) !default;\n// scss-docs-end colors-map\n\n// The contrast ratio to reach against white, to determine if color changes from \"light\" to \"dark\". Acceptable values for WCAG 2.0 are 3, 4.5 and 7.\n// See https://www.w3.org/TR/WCAG20/#visual-audio-contrast-contrast\n$min-contrast-ratio:   4.5 !default;\n\n// Customize the light and dark text colors for use in our color contrast function.\n$color-contrast-dark:      $black !default;\n$color-contrast-light:     $white !default;\n\n// fusv-disable\n$blue-100: tint-color($blue, 80%) !default;\n$blue-200: tint-color($blue, 60%) !default;\n$blue-300: tint-color($blue, 40%) !default;\n$blue-400: tint-color($blue, 20%) !default;\n$blue-500: $blue !default;\n$blue-600: shade-color($blue, 20%) !default;\n$blue-700: shade-color($blue, 40%) !default;\n$blue-800: shade-color($blue, 60%) !default;\n$blue-900: shade-color($blue, 80%) !default;\n\n$indigo-100: tint-color($indigo, 80%) !default;\n$indigo-200: tint-color($indigo, 60%) !default;\n$indigo-300: tint-color($indigo, 40%) !default;\n$indigo-400: tint-color($indigo, 20%) !default;\n$indigo-500: $indigo !default;\n$indigo-600: shade-color($indigo, 20%) !default;\n$indigo-700: shade-color($indigo, 40%) !default;\n$indigo-800: shade-color($indigo, 60%) !default;\n$indigo-900: shade-color($indigo, 80%) !default;\n\n$purple-100: tint-color($purple, 80%) !default;\n$purple-200: tint-color($purple, 60%) !default;\n$purple-300: tint-color($purple, 40%) !default;\n$purple-400: tint-color($purple, 20%) !default;\n$purple-500: $purple !default;\n$purple-600: shade-color($purple, 20%) !default;\n$purple-700: shade-color($purple, 40%) !default;\n$purple-800: shade-color($purple, 60%) !default;\n$purple-900: shade-color($purple, 80%) !default;\n\n$pink-100: tint-color($pink, 80%) !default;\n$pink-200: tint-color($pink, 60%) !default;\n$pink-300: tint-color($pink, 40%) !default;\n$pink-400: tint-color($pink, 20%) !default;\n$pink-500: $pink !default;\n$pink-600: shade-color($pink, 20%) !default;\n$pink-700: shade-color($pink, 40%) !default;\n$pink-800: shade-color($pink, 60%) !default;\n$pink-900: shade-color($pink, 80%) !default;\n\n$red-100: tint-color($red, 80%) !default;\n$red-200: tint-color($red, 60%) !default;\n$red-300: tint-color($red, 40%) !default;\n$red-400: tint-color($red, 20%) !default;\n$red-500: $red !default;\n$red-600: shade-color($red, 20%) !default;\n$red-700: shade-color($red, 40%) !default;\n$red-800: shade-color($red, 60%) !default;\n$red-900: shade-color($red, 80%) !default;\n\n$orange-100: tint-color($orange, 80%) !default;\n$orange-200: tint-color($orange, 60%) !default;\n$orange-300: tint-color($orange, 40%) !default;\n$orange-400: tint-color($orange, 20%) !default;\n$orange-500: $orange !default;\n$orange-600: shade-color($orange, 20%) !default;\n$orange-700: shade-color($orange, 40%) !default;\n$orange-800: shade-color($orange, 60%) !default;\n$orange-900: shade-color($orange, 80%) !default;\n\n$yellow-100: tint-color($yellow, 80%) !default;\n$yellow-200: tint-color($yellow, 60%) !default;\n$yellow-300: tint-color($yellow, 40%) !default;\n$yellow-400: tint-color($yellow, 20%) !default;\n$yellow-500: $yellow !default;\n$yellow-600: shade-color($yellow, 20%) !default;\n$yellow-700: shade-color($yellow, 40%) !default;\n$yellow-800: shade-color($yellow, 60%) !default;\n$yellow-900: shade-color($yellow, 80%) !default;\n\n$green-100: tint-color($green, 80%) !default;\n$green-200: tint-color($green, 60%) !default;\n$green-300: tint-color($green, 40%) !default;\n$green-400: tint-color($green, 20%) !default;\n$green-500: $green !default;\n$green-600: shade-color($green, 20%) !default;\n$green-700: shade-color($green, 40%) !default;\n$green-800: shade-color($green, 60%) !default;\n$green-900: shade-color($green, 80%) !default;\n\n$teal-100: tint-color($teal, 80%) !default;\n$teal-200: tint-color($teal, 60%) !default;\n$teal-300: tint-color($teal, 40%) !default;\n$teal-400: tint-color($teal, 20%) !default;\n$teal-500: $teal !default;\n$teal-600: shade-color($teal, 20%) !default;\n$teal-700: shade-color($teal, 40%) !default;\n$teal-800: shade-color($teal, 60%) !default;\n$teal-900: shade-color($teal, 80%) !default;\n\n$cyan-100: tint-color($cyan, 80%) !default;\n$cyan-200: tint-color($cyan, 60%) !default;\n$cyan-300: tint-color($cyan, 40%) !default;\n$cyan-400: tint-color($cyan, 20%) !default;\n$cyan-500: $cyan !default;\n$cyan-600: shade-color($cyan, 20%) !default;\n$cyan-700: shade-color($cyan, 40%) !default;\n$cyan-800: shade-color($cyan, 60%) !default;\n$cyan-900: shade-color($cyan, 80%) !default;\n\n$blues: (\n  \"blue-100\": $blue-100,\n  \"blue-200\": $blue-200,\n  \"blue-300\": $blue-300,\n  \"blue-400\": $blue-400,\n  \"blue-500\": $blue-500,\n  \"blue-600\": $blue-600,\n  \"blue-700\": $blue-700,\n  \"blue-800\": $blue-800,\n  \"blue-900\": $blue-900\n) !default;\n\n$indigos: (\n  \"indigo-100\": $indigo-100,\n  \"indigo-200\": $indigo-200,\n  \"indigo-300\": $indigo-300,\n  \"indigo-400\": $indigo-400,\n  \"indigo-500\": $indigo-500,\n  \"indigo-600\": $indigo-600,\n  \"indigo-700\": $indigo-700,\n  \"indigo-800\": $indigo-800,\n  \"indigo-900\": $indigo-900\n) !default;\n\n$purples: (\n  \"purple-100\": $purple-100,\n  \"purple-200\": $purple-200,\n  \"purple-300\": $purple-300,\n  \"purple-400\": $purple-400,\n  \"purple-500\": $purple-500,\n  \"purple-600\": $purple-600,\n  \"purple-700\": $purple-700,\n  \"purple-800\": $purple-800,\n  \"purple-900\": $purple-900\n) !default;\n\n$pinks: (\n  \"pink-100\": $pink-100,\n  \"pink-200\": $pink-200,\n  \"pink-300\": $pink-300,\n  \"pink-400\": $pink-400,\n  \"pink-500\": $pink-500,\n  \"pink-600\": $pink-600,\n  \"pink-700\": $pink-700,\n  \"pink-800\": $pink-800,\n  \"pink-900\": $pink-900\n) !default;\n\n$reds: (\n  \"red-100\": $red-100,\n  \"red-200\": $red-200,\n  \"red-300\": $red-300,\n  \"red-400\": $red-400,\n  \"red-500\": $red-500,\n  \"red-600\": $red-600,\n  \"red-700\": $red-700,\n  \"red-800\": $red-800,\n  \"red-900\": $red-900\n) !default;\n\n$oranges: (\n  \"orange-100\": $orange-100,\n  \"orange-200\": $orange-200,\n  \"orange-300\": $orange-300,\n  \"orange-400\": $orange-400,\n  \"orange-500\": $orange-500,\n  \"orange-600\": $orange-600,\n  \"orange-700\": $orange-700,\n  \"orange-800\": $orange-800,\n  \"orange-900\": $orange-900\n) !default;\n\n$yellows: (\n  \"yellow-100\": $yellow-100,\n  \"yellow-200\": $yellow-200,\n  \"yellow-300\": $yellow-300,\n  \"yellow-400\": $yellow-400,\n  \"yellow-500\": $yellow-500,\n  \"yellow-600\": $yellow-600,\n  \"yellow-700\": $yellow-700,\n  \"yellow-800\": $yellow-800,\n  \"yellow-900\": $yellow-900\n) !default;\n\n$greens: (\n  \"green-100\": $green-100,\n  \"green-200\": $green-200,\n  \"green-300\": $green-300,\n  \"green-400\": $green-400,\n  \"green-500\": $green-500,\n  \"green-600\": $green-600,\n  \"green-700\": $green-700,\n  \"green-800\": $green-800,\n  \"green-900\": $green-900\n) !default;\n\n$teals: (\n  \"teal-100\": $teal-100,\n  \"teal-200\": $teal-200,\n  \"teal-300\": $teal-300,\n  \"teal-400\": $teal-400,\n  \"teal-500\": $teal-500,\n  \"teal-600\": $teal-600,\n  \"teal-700\": $teal-700,\n  \"teal-800\": $teal-800,\n  \"teal-900\": $teal-900\n) !default;\n\n$cyans: (\n  \"cyan-100\": $cyan-100,\n  \"cyan-200\": $cyan-200,\n  \"cyan-300\": $cyan-300,\n  \"cyan-400\": $cyan-400,\n  \"cyan-500\": $cyan-500,\n  \"cyan-600\": $cyan-600,\n  \"cyan-700\": $cyan-700,\n  \"cyan-800\": $cyan-800,\n  \"cyan-900\": $cyan-900\n) !default;\n// fusv-enable\n\n// scss-docs-start theme-color-variables\n$primary:       $blue !default;\n$secondary:     $gray-600 !default;\n$success:       $green !default;\n$info:          $cyan !default;\n$warning:       $yellow !default;\n$danger:        $red !default;\n$light:         $gray-100 !default;\n$dark:          $gray-900 !default;\n// scss-docs-end theme-color-variables\n\n// scss-docs-start theme-colors-map\n$theme-colors: (\n  \"primary\":    $primary,\n  \"secondary\":  $secondary,\n  \"success\":    $success,\n  \"info\":       $info,\n  \"warning\":    $warning,\n  \"danger\":     $danger,\n  \"light\":      $light,\n  \"dark\":       $dark\n) !default;\n// scss-docs-end theme-colors-map\n\n// scss-docs-start theme-text-variables\n$primary-text-emphasis:   shade-color($primary, 60%) !default;\n$secondary-text-emphasis: shade-color($secondary, 60%) !default;\n$success-text-emphasis:   shade-color($success, 60%) !default;\n$info-text-emphasis:      shade-color($info, 60%) !default;\n$warning-text-emphasis:   shade-color($warning, 60%) !default;\n$danger-text-emphasis:    shade-color($danger, 60%) !default;\n$light-text-emphasis:     $gray-700 !default;\n$dark-text-emphasis:      $gray-700 !default;\n// scss-docs-end theme-text-variables\n\n// scss-docs-start theme-bg-subtle-variables\n$primary-bg-subtle:       tint-color($primary, 80%) !default;\n$secondary-bg-subtle:     tint-color($secondary, 80%) !default;\n$success-bg-subtle:       tint-color($success, 80%) !default;\n$info-bg-subtle:          tint-color($info, 80%) !default;\n$warning-bg-subtle:       tint-color($warning, 80%) !default;\n$danger-bg-subtle:        tint-color($danger, 80%) !default;\n$light-bg-subtle:         mix($gray-100, $white) !default;\n$dark-bg-subtle:          $gray-400 !default;\n// scss-docs-end theme-bg-subtle-variables\n\n// scss-docs-start theme-border-subtle-variables\n$primary-border-subtle:   tint-color($primary, 60%) !default;\n$secondary-border-subtle: tint-color($secondary, 60%) !default;\n$success-border-subtle:   tint-color($success, 60%) !default;\n$info-border-subtle:      tint-color($info, 60%) !default;\n$warning-border-subtle:   tint-color($warning, 60%) !default;\n$danger-border-subtle:    tint-color($danger, 60%) !default;\n$light-border-subtle:     $gray-200 !default;\n$dark-border-subtle:      $gray-500 !default;\n// scss-docs-end theme-border-subtle-variables\n\n// Characters which are escaped by the escape-svg function\n$escaped-characters: (\n  (\"<\", \"%3c\"),\n  (\">\", \"%3e\"),\n  (\"#\", \"%23\"),\n  (\"(\", \"%28\"),\n  (\")\", \"%29\"),\n) !default;\n\n// Options\n//\n// Quickly modify global styling by enabling or disabling optional features.\n\n$enable-caret:                true !default;\n$enable-rounded:              true !default;\n$enable-shadows:              false !default;\n$enable-gradients:            false !default;\n$enable-transitions:          true !default;\n$enable-reduced-motion:       true !default;\n$enable-smooth-scroll:        true !default;\n$enable-grid-classes:         true !default;\n$enable-container-classes:    true !default;\n$enable-cssgrid:              false !default;\n$enable-button-pointers:      true !default;\n$enable-rfs:                  true !default;\n$enable-validation-icons:     true !default;\n$enable-negative-margins:     false !default;\n$enable-deprecation-messages: true !default;\n$enable-important-utilities:  true !default;\n\n$enable-dark-mode:            true !default;\n$color-mode-type:             data !default; // `data` or `media-query`\n\n// Prefix for :root CSS variables\n\n$variable-prefix:             bs- !default; // Deprecated in v5.2.0 for the shorter `$prefix`\n$prefix:                      $variable-prefix !default;\n\n// Gradient\n//\n// The gradient which is added to components if `$enable-gradients` is `true`\n// This gradient is also added to elements with `.bg-gradient`\n// scss-docs-start variable-gradient\n$gradient: linear-gradient(180deg, rgba($white, .15), rgba($white, 0)) !default;\n// scss-docs-end variable-gradient\n\n// Spacing\n//\n// Control the default styling of most Bootstrap elements by modifying these\n// variables. Mostly focused on spacing.\n// You can add more entries to the $spacers map, should you need more variation.\n\n// scss-docs-start spacer-variables-maps\n$spacer: 1rem !default;\n$spacers: (\n  0: 0,\n  1: $spacer * .25,\n  2: $spacer * .5,\n  3: $spacer,\n  4: $spacer * 1.5,\n  5: $spacer * 3,\n) !default;\n// scss-docs-end spacer-variables-maps\n\n// Position\n//\n// Define the edge positioning anchors of the position utilities.\n\n// scss-docs-start position-map\n$position-values: (\n  0: 0,\n  50: 50%,\n  100: 100%\n) !default;\n// scss-docs-end position-map\n\n// Body\n//\n// Settings for the `<body>` element.\n\n$body-text-align:           null !default;\n$body-color:                $gray-900 !default;\n$body-bg:                   $white !default;\n\n$body-secondary-color:      rgba($body-color, .75) !default;\n$body-secondary-bg:         $gray-200 !default;\n\n$body-tertiary-color:       rgba($body-color, .5) !default;\n$body-tertiary-bg:          $gray-100 !default;\n\n$body-emphasis-color:       $black !default;\n\n// Links\n//\n// Style anchor elements.\n\n$link-color:                              $primary !default;\n$link-decoration:                         underline !default;\n$link-shade-percentage:                   20% !default;\n$link-hover-color:                        shift-color($link-color, $link-shade-percentage) !default;\n$link-hover-decoration:                   null !default;\n\n$stretched-link-pseudo-element:           after !default;\n$stretched-link-z-index:                  1 !default;\n\n// Icon links\n// scss-docs-start icon-link-variables\n$icon-link-gap:               .375rem !default;\n$icon-link-underline-offset:  .25em !default;\n$icon-link-icon-size:         1em !default;\n$icon-link-icon-transition:   .2s ease-in-out transform !default;\n$icon-link-icon-transform:    translate3d(.25em, 0, 0) !default;\n// scss-docs-end icon-link-variables\n\n// Paragraphs\n//\n// Style p element.\n\n$paragraph-margin-bottom:   1rem !default;\n\n\n// Grid breakpoints\n//\n// Define the minimum dimensions at which your layout will change,\n// adapting to different screen sizes, for use in media queries.\n\n// scss-docs-start grid-breakpoints\n$grid-breakpoints: (\n  xs: 0,\n  sm: 576px,\n  md: 768px,\n  lg: 992px,\n  xl: 1200px,\n  xxl: 1400px\n) !default;\n// scss-docs-end grid-breakpoints\n\n@include _assert-ascending($grid-breakpoints, \"$grid-breakpoints\");\n@include _assert-starts-at-zero($grid-breakpoints, \"$grid-breakpoints\");\n\n\n// Grid containers\n//\n// Define the maximum width of `.container` for different screen sizes.\n\n// scss-docs-start container-max-widths\n$container-max-widths: (\n  sm: 540px,\n  md: 720px,\n  lg: 960px,\n  xl: 1140px,\n  xxl: 1320px\n) !default;\n// scss-docs-end container-max-widths\n\n@include _assert-ascending($container-max-widths, \"$container-max-widths\");\n\n\n// Grid columns\n//\n// Set the number of columns and specify the width of the gutters.\n\n$grid-columns:                12 !default;\n$grid-gutter-width:           1.5rem !default;\n$grid-row-columns:            6 !default;\n\n// Container padding\n\n$container-padding-x: $grid-gutter-width !default;\n\n\n// Components\n//\n// Define common padding and border radius sizes and more.\n\n// scss-docs-start border-variables\n$border-width:                1px !default;\n$border-widths: (\n  1: 1px,\n  2: 2px,\n  3: 3px,\n  4: 4px,\n  5: 5px\n) !default;\n$border-style:                solid !default;\n$border-color:                $gray-300 !default;\n$border-color-translucent:    rgba($black, .175) !default;\n// scss-docs-end border-variables\n\n// scss-docs-start border-radius-variables\n$border-radius:               .375rem !default;\n$border-radius-sm:            .25rem !default;\n$border-radius-lg:            .5rem !default;\n$border-radius-xl:            1rem !default;\n$border-radius-xxl:           2rem !default;\n$border-radius-pill:          50rem !default;\n// scss-docs-end border-radius-variables\n// fusv-disable\n$border-radius-2xl:           $border-radius-xxl !default; // Deprecated in v5.3.0\n// fusv-enable\n\n// scss-docs-start box-shadow-variables\n$box-shadow:                  0 .5rem 1rem rgba($black, .15) !default;\n$box-shadow-sm:               0 .125rem .25rem rgba($black, .075) !default;\n$box-shadow-lg:               0 1rem 3rem rgba($black, .175) !default;\n$box-shadow-inset:            inset 0 1px 2px rgba($black, .075) !default;\n// scss-docs-end box-shadow-variables\n\n$component-active-color:      $white !default;\n$component-active-bg:         $primary !default;\n\n// scss-docs-start focus-ring-variables\n$focus-ring-width:      .25rem !default;\n$focus-ring-opacity:    .25 !default;\n$focus-ring-color:      rgba($primary, $focus-ring-opacity) !default;\n$focus-ring-blur:       0 !default;\n$focus-ring-box-shadow: 0 0 $focus-ring-blur $focus-ring-width $focus-ring-color !default;\n// scss-docs-end focus-ring-variables\n\n// scss-docs-start caret-variables\n$caret-width:                 .3em !default;\n$caret-vertical-align:        $caret-width * .85 !default;\n$caret-spacing:               $caret-width * .85 !default;\n// scss-docs-end caret-variables\n\n$transition-base:             all .2s ease-in-out !default;\n$transition-fade:             opacity .15s linear !default;\n// scss-docs-start collapse-transition\n$transition-collapse:         height .35s ease !default;\n$transition-collapse-width:   width .35s ease !default;\n// scss-docs-end collapse-transition\n\n// stylelint-disable function-disallowed-list\n// scss-docs-start aspect-ratios\n$aspect-ratios: (\n  \"1x1\": 100%,\n  \"4x3\": calc(3 / 4 * 100%),\n  \"16x9\": calc(9 / 16 * 100%),\n  \"21x9\": calc(9 / 21 * 100%)\n) !default;\n// scss-docs-end aspect-ratios\n// stylelint-enable function-disallowed-list\n\n// Typography\n//\n// Font, line-height, and color for body text, headings, and more.\n\n// scss-docs-start font-variables\n// stylelint-disable value-keyword-case\n$font-family-sans-serif:      system-ui, -apple-system, \"Segoe UI\", Roboto, \"Helvetica Neue\", \"Noto Sans\", \"Liberation Sans\", Arial, sans-serif, \"Apple Color Emoji\", \"Segoe UI Emoji\", \"Segoe UI Symbol\", \"Noto Color Emoji\" !default;\n$font-family-monospace:       SFMono-Regular, Menlo, Monaco, Consolas, \"Liberation Mono\", \"Courier New\", monospace !default;\n// stylelint-enable value-keyword-case\n$font-family-base:            var(--#{$prefix}font-sans-serif) !default;\n$font-family-code:            var(--#{$prefix}font-monospace) !default;\n\n// $font-size-root affects the value of `rem`, which is used for as well font sizes, paddings, and margins\n// $font-size-base affects the font size of the body text\n$font-size-root:              null !default;\n$font-size-base:              1rem !default; // Assumes the browser default, typically `16px`\n$font-size-sm:                $font-size-base * .875 !default;\n$font-size-lg:                $font-size-base * 1.25 !default;\n\n$font-weight-lighter:         lighter !default;\n$font-weight-light:           300 !default;\n$font-weight-normal:          400 !default;\n$font-weight-medium:          500 !default;\n$font-weight-semibold:        600 !default;\n$font-weight-bold:            700 !default;\n$font-weight-bolder:          bolder !default;\n\n$font-weight-base:            $font-weight-normal !default;\n\n$line-height-base:            1.5 !default;\n$line-height-sm:              1.25 !default;\n$line-height-lg:              2 !default;\n\n$h1-font-size:                $font-size-base * 2.5 !default;\n$h2-font-size:                $font-size-base * 2 !default;\n$h3-font-size:                $font-size-base * 1.75 !default;\n$h4-font-size:                $font-size-base * 1.5 !default;\n$h5-font-size:                $font-size-base * 1.25 !default;\n$h6-font-size:                $font-size-base !default;\n// scss-docs-end font-variables\n\n// scss-docs-start font-sizes\n$font-sizes: (\n  1: $h1-font-size,\n  2: $h2-font-size,\n  3: $h3-font-size,\n  4: $h4-font-size,\n  5: $h5-font-size,\n  6: $h6-font-size\n) !default;\n// scss-docs-end font-sizes\n\n// scss-docs-start headings-variables\n$headings-margin-bottom:      $spacer * .5 !default;\n$headings-font-family:        null !default;\n$headings-font-style:         null !default;\n$headings-font-weight:        500 !default;\n$headings-line-height:        1.2 !default;\n$headings-color:              inherit !default;\n// scss-docs-end headings-variables\n\n// scss-docs-start display-headings\n$display-font-sizes: (\n  1: 5rem,\n  2: 4.5rem,\n  3: 4rem,\n  4: 3.5rem,\n  5: 3rem,\n  6: 2.5rem\n) !default;\n\n$display-font-family: null !default;\n$display-font-style:  null !default;\n$display-font-weight: 300 !default;\n$display-line-height: $headings-line-height !default;\n// scss-docs-end display-headings\n\n// scss-docs-start type-variables\n$lead-font-size:              $font-size-base * 1.25 !default;\n$lead-font-weight:            300 !default;\n\n$small-font-size:             .875em !default;\n\n$sub-sup-font-size:           .75em !default;\n\n// fusv-disable\n$text-muted:                  var(--#{$prefix}secondary-color) !default; // Deprecated in 5.3.0\n// fusv-enable\n\n$initialism-font-size:        $small-font-size !default;\n\n$blockquote-margin-y:         $spacer !default;\n$blockquote-font-size:        $font-size-base * 1.25 !default;\n$blockquote-footer-color:     $gray-600 !default;\n$blockquote-footer-font-size: $small-font-size !default;\n\n$hr-margin-y:                 $spacer !default;\n$hr-color:                    inherit !default;\n\n// fusv-disable\n$hr-bg-color:                 null !default; // Deprecated in v5.2.0\n$hr-height:                   null !default; // Deprecated in v5.2.0\n// fusv-enable\n\n$hr-border-color:             null !default; // Allows for inherited colors\n$hr-border-width:             var(--#{$prefix}border-width) !default;\n$hr-opacity:                  .25 !default;\n\n// scss-docs-start vr-variables\n$vr-border-width:             var(--#{$prefix}border-width) !default;\n// scss-docs-end vr-variables\n\n$legend-margin-bottom:        .5rem !default;\n$legend-font-size:            1.5rem !default;\n$legend-font-weight:          null !default;\n\n$dt-font-weight:              $font-weight-bold !default;\n\n$list-inline-padding:         .5rem !default;\n\n$mark-padding:                .1875em !default;\n$mark-color:                  $body-color !default;\n$mark-bg:                     $yellow-100 !default;\n// scss-docs-end type-variables\n\n\n// Tables\n//\n// Customizes the `.table` component with basic values, each used across all table variations.\n\n// scss-docs-start table-variables\n$table-cell-padding-y:        .5rem !default;\n$table-cell-padding-x:        .5rem !default;\n$table-cell-padding-y-sm:     .25rem !default;\n$table-cell-padding-x-sm:     .25rem !default;\n\n$table-cell-vertical-align:   top !default;\n\n$table-color:                 var(--#{$prefix}emphasis-color) !default;\n$table-bg:                    var(--#{$prefix}body-bg) !default;\n$table-accent-bg:             transparent !default;\n\n$table-th-font-weight:        null !default;\n\n$table-striped-color:         $table-color !default;\n$table-striped-bg-factor:     .05 !default;\n$table-striped-bg:            rgba(var(--#{$prefix}emphasis-color-rgb), $table-striped-bg-factor) !default;\n\n$table-active-color:          $table-color !default;\n$table-active-bg-factor:      .1 !default;\n$table-active-bg:             rgba(var(--#{$prefix}emphasis-color-rgb), $table-active-bg-factor) !default;\n\n$table-hover-color:           $table-color !default;\n$table-hover-bg-factor:       .075 !default;\n$table-hover-bg:              rgba(var(--#{$prefix}emphasis-color-rgb), $table-hover-bg-factor) !default;\n\n$table-border-factor:         .2 !default;\n$table-border-width:          var(--#{$prefix}border-width) !default;\n$table-border-color:          var(--#{$prefix}border-color) !default;\n\n$table-striped-order:         odd !default;\n$table-striped-columns-order: even !default;\n\n$table-group-separator-color: currentcolor !default;\n\n$table-caption-color:         var(--#{$prefix}secondary-color) !default;\n\n$table-bg-scale:              -80% !default;\n// scss-docs-end table-variables\n\n// scss-docs-start table-loop\n$table-variants: (\n  \"primary\":    shift-color($primary, $table-bg-scale),\n  \"secondary\":  shift-color($secondary, $table-bg-scale),\n  \"success\":    shift-color($success, $table-bg-scale),\n  \"info\":       shift-color($info, $table-bg-scale),\n  \"warning\":    shift-color($warning, $table-bg-scale),\n  \"danger\":     shift-color($danger, $table-bg-scale),\n  \"light\":      $light,\n  \"dark\":       $dark,\n) !default;\n// scss-docs-end table-loop\n\n\n// Buttons + Forms\n//\n// Shared variables that are reassigned to `$input-` and `$btn-` specific variables.\n\n// scss-docs-start input-btn-variables\n$input-btn-padding-y:         .375rem !default;\n$input-btn-padding-x:         .75rem !default;\n$input-btn-font-family:       null !default;\n$input-btn-font-size:         $font-size-base !default;\n$input-btn-line-height:       $line-height-base !default;\n\n$input-btn-focus-width:         $focus-ring-width !default;\n$input-btn-focus-color-opacity: $focus-ring-opacity !default;\n$input-btn-focus-color:         $focus-ring-color !default;\n$input-btn-focus-blur:          $focus-ring-blur !default;\n$input-btn-focus-box-shadow:    $focus-ring-box-shadow !default;\n\n$input-btn-padding-y-sm:      .25rem !default;\n$input-btn-padding-x-sm:      .5rem !default;\n$input-btn-font-size-sm:      $font-size-sm !default;\n\n$input-btn-padding-y-lg:      .5rem !default;\n$input-btn-padding-x-lg:      1rem !default;\n$input-btn-font-size-lg:      $font-size-lg !default;\n\n$input-btn-border-width:      var(--#{$prefix}border-width) !default;\n// scss-docs-end input-btn-variables\n\n\n// Buttons\n//\n// For each of Bootstrap's buttons, define text, background, and border color.\n\n// scss-docs-start btn-variables\n$btn-color:                   var(--#{$prefix}body-color) !default;\n$btn-padding-y:               $input-btn-padding-y !default;\n$btn-padding-x:               $input-btn-padding-x !default;\n$btn-font-family:             $input-btn-font-family !default;\n$btn-font-size:               $input-btn-font-size !default;\n$btn-line-height:             $input-btn-line-height !default;\n$btn-white-space:             null !default; // Set to `nowrap` to prevent text wrapping\n\n$btn-padding-y-sm:            $input-btn-padding-y-sm !default;\n$btn-padding-x-sm:            $input-btn-padding-x-sm !default;\n$btn-font-size-sm:            $input-btn-font-size-sm !default;\n\n$btn-padding-y-lg:            $input-btn-padding-y-lg !default;\n$btn-padding-x-lg:            $input-btn-padding-x-lg !default;\n$btn-font-size-lg:            $input-btn-font-size-lg !default;\n\n$btn-border-width:            $input-btn-border-width !default;\n\n$btn-font-weight:             $font-weight-normal !default;\n$btn-box-shadow:              inset 0 1px 0 rgba($white, .15), 0 1px 1px rgba($black, .075) !default;\n$btn-focus-width:             $input-btn-focus-width !default;\n$btn-focus-box-shadow:        $input-btn-focus-box-shadow !default;\n$btn-disabled-opacity:        .65 !default;\n$btn-active-box-shadow:       inset 0 3px 5px rgba($black, .125) !default;\n\n$btn-link-color:              var(--#{$prefix}link-color) !default;\n$btn-link-hover-color:        var(--#{$prefix}link-hover-color) !default;\n$btn-link-disabled-color:     $gray-600 !default;\n$btn-link-focus-shadow-rgb:   to-rgb(mix(color-contrast($link-color), $link-color, 15%)) !default;\n\n// Allows for customizing button radius independently from global border radius\n$btn-border-radius:           var(--#{$prefix}border-radius) !default;\n$btn-border-radius-sm:        var(--#{$prefix}border-radius-sm) !default;\n$btn-border-radius-lg:        var(--#{$prefix}border-radius-lg) !default;\n\n$btn-transition:              color .15s ease-in-out, background-color .15s ease-in-out, border-color .15s ease-in-out, box-shadow .15s ease-in-out !default;\n\n$btn-hover-bg-shade-amount:       15% !default;\n$btn-hover-bg-tint-amount:        15% !default;\n$btn-hover-border-shade-amount:   20% !default;\n$btn-hover-border-tint-amount:    10% !default;\n$btn-active-bg-shade-amount:      20% !default;\n$btn-active-bg-tint-amount:       20% !default;\n$btn-active-border-shade-amount:  25% !default;\n$btn-active-border-tint-amount:   10% !default;\n// scss-docs-end btn-variables\n\n\n// Forms\n\n// scss-docs-start form-text-variables\n$form-text-margin-top:                  .25rem !default;\n$form-text-font-size:                   $small-font-size !default;\n$form-text-font-style:                  null !default;\n$form-text-font-weight:                 null !default;\n$form-text-color:                       var(--#{$prefix}secondary-color) !default;\n// scss-docs-end form-text-variables\n\n// scss-docs-start form-label-variables\n$form-label-margin-bottom:              .5rem !default;\n$form-label-font-size:                  null !default;\n$form-label-font-style:                 null !default;\n$form-label-font-weight:                null !default;\n$form-label-color:                      null !default;\n// scss-docs-end form-label-variables\n\n// scss-docs-start form-input-variables\n$input-padding-y:                       $input-btn-padding-y !default;\n$input-padding-x:                       $input-btn-padding-x !default;\n$input-font-family:                     $input-btn-font-family !default;\n$input-font-size:                       $input-btn-font-size !default;\n$input-font-weight:                     $font-weight-base !default;\n$input-line-height:                     $input-btn-line-height !default;\n\n$input-padding-y-sm:                    $input-btn-padding-y-sm !default;\n$input-padding-x-sm:                    $input-btn-padding-x-sm !default;\n$input-font-size-sm:                    $input-btn-font-size-sm !default;\n\n$input-padding-y-lg:                    $input-btn-padding-y-lg !default;\n$input-padding-x-lg:                    $input-btn-padding-x-lg !default;\n$input-font-size-lg:                    $input-btn-font-size-lg !default;\n\n$input-bg:                              var(--#{$prefix}body-bg) !default;\n$input-disabled-color:                  null !default;\n$input-disabled-bg:                     var(--#{$prefix}secondary-bg) !default;\n$input-disabled-border-color:           null !default;\n\n$input-color:                           var(--#{$prefix}body-color) !default;\n$input-border-color:                    var(--#{$prefix}border-color) !default;\n$input-border-width:                    $input-btn-border-width !default;\n$input-box-shadow:                      var(--#{$prefix}box-shadow-inset) !default;\n\n$input-border-radius:                   var(--#{$prefix}border-radius) !default;\n$input-border-radius-sm:                var(--#{$prefix}border-radius-sm) !default;\n$input-border-radius-lg:                var(--#{$prefix}border-radius-lg) !default;\n\n$input-focus-bg:                        $input-bg !default;\n$input-focus-border-color:              tint-color($component-active-bg, 50%) !default;\n$input-focus-color:                     $input-color !default;\n$input-focus-width:                     $input-btn-focus-width !default;\n$input-focus-box-shadow:                $input-btn-focus-box-shadow !default;\n\n$input-placeholder-color:               var(--#{$prefix}secondary-color) !default;\n$input-plaintext-color:                 var(--#{$prefix}body-color) !default;\n\n$input-height-border:                   calc(#{$input-border-width} * 2) !default; // stylelint-disable-line function-disallowed-list\n\n$input-height-inner:                    add($input-line-height * 1em, $input-padding-y * 2) !default;\n$input-height-inner-half:               add($input-line-height * .5em, $input-padding-y) !default;\n$input-height-inner-quarter:            add($input-line-height * .25em, $input-padding-y * .5) !default;\n\n$input-height:                          add($input-line-height * 1em, add($input-padding-y * 2, $input-height-border, false)) !default;\n$input-height-sm:                       add($input-line-height * 1em, add($input-padding-y-sm * 2, $input-height-border, false)) !default;\n$input-height-lg:                       add($input-line-height * 1em, add($input-padding-y-lg * 2, $input-height-border, false)) !default;\n\n$input-transition:                      border-color .15s ease-in-out, box-shadow .15s ease-in-out !default;\n\n$form-color-width:                      3rem !default;\n// scss-docs-end form-input-variables\n\n// scss-docs-start form-check-variables\n$form-check-input-width:                  1em !default;\n$form-check-min-height:                   $font-size-base * $line-height-base !default;\n$form-check-padding-start:                $form-check-input-width + .5em !default;\n$form-check-margin-bottom:                .125rem !default;\n$form-check-label-color:                  null !default;\n$form-check-label-cursor:                 null !default;\n$form-check-transition:                   null !default;\n\n$form-check-input-active-filter:          brightness(90%) !default;\n\n$form-check-input-bg:                     $input-bg !default;\n$form-check-input-border:                 var(--#{$prefix}border-width) solid var(--#{$prefix}border-color) !default;\n$form-check-input-border-radius:          .25em !default;\n$form-check-radio-border-radius:          50% !default;\n$form-check-input-focus-border:           $input-focus-border-color !default;\n$form-check-input-focus-box-shadow:       $focus-ring-box-shadow !default;\n\n$form-check-input-checked-color:          $component-active-color !default;\n$form-check-input-checked-bg-color:       $component-active-bg !default;\n$form-check-input-checked-border-color:   $form-check-input-checked-bg-color !default;\n$form-check-input-checked-bg-image:       url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 20 20'><path fill='none' stroke='#{$form-check-input-checked-color}' stroke-linecap='round' stroke-linejoin='round' stroke-width='3' d='m6 10 3 3 6-6'/></svg>\") !default;\n$form-check-radio-checked-bg-image:       url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'><circle r='2' fill='#{$form-check-input-checked-color}'/></svg>\") !default;\n\n$form-check-input-indeterminate-color:          $component-active-color !default;\n$form-check-input-indeterminate-bg-color:       $component-active-bg !default;\n$form-check-input-indeterminate-border-color:   $form-check-input-indeterminate-bg-color !default;\n$form-check-input-indeterminate-bg-image:       url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 20 20'><path fill='none' stroke='#{$form-check-input-indeterminate-color}' stroke-linecap='round' stroke-linejoin='round' stroke-width='3' d='M6 10h8'/></svg>\") !default;\n\n$form-check-input-disabled-opacity:        .5 !default;\n$form-check-label-disabled-opacity:        $form-check-input-disabled-opacity !default;\n$form-check-btn-check-disabled-opacity:    $btn-disabled-opacity !default;\n\n$form-check-inline-margin-end:    1rem !default;\n// scss-docs-end form-check-variables\n\n// scss-docs-start form-switch-variables\n$form-switch-color:               rgba($black, .25) !default;\n$form-switch-width:               2em !default;\n$form-switch-padding-start:       $form-switch-width + .5em !default;\n$form-switch-bg-image:            url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'><circle r='3' fill='#{$form-switch-color}'/></svg>\") !default;\n$form-switch-border-radius:       $form-switch-width !default;\n$form-switch-transition:          background-position .15s ease-in-out !default;\n\n$form-switch-focus-color:         $input-focus-border-color !default;\n$form-switch-focus-bg-image:      url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'><circle r='3' fill='#{$form-switch-focus-color}'/></svg>\") !default;\n\n$form-switch-checked-color:       $component-active-color !default;\n$form-switch-checked-bg-image:    url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='-4 -4 8 8'><circle r='3' fill='#{$form-switch-checked-color}'/></svg>\") !default;\n$form-switch-checked-bg-position: right center !default;\n// scss-docs-end form-switch-variables\n\n// scss-docs-start input-group-variables\n$input-group-addon-padding-y:           $input-padding-y !default;\n$input-group-addon-padding-x:           $input-padding-x !default;\n$input-group-addon-font-weight:         $input-font-weight !default;\n$input-group-addon-color:               $input-color !default;\n$input-group-addon-bg:                  var(--#{$prefix}tertiary-bg) !default;\n$input-group-addon-border-color:        $input-border-color !default;\n// scss-docs-end input-group-variables\n\n// scss-docs-start form-select-variables\n$form-select-padding-y:             $input-padding-y !default;\n$form-select-padding-x:             $input-padding-x !default;\n$form-select-font-family:           $input-font-family !default;\n$form-select-font-size:             $input-font-size !default;\n$form-select-indicator-padding:     $form-select-padding-x * 3 !default; // Extra padding for background-image\n$form-select-font-weight:           $input-font-weight !default;\n$form-select-line-height:           $input-line-height !default;\n$form-select-color:                 $input-color !default;\n$form-select-bg:                    $input-bg !default;\n$form-select-disabled-color:        null !default;\n$form-select-disabled-bg:           $input-disabled-bg !default;\n$form-select-disabled-border-color: $input-disabled-border-color !default;\n$form-select-bg-position:           right $form-select-padding-x center !default;\n$form-select-bg-size:               16px 12px !default; // In pixels because image dimensions\n$form-select-indicator-color:       $gray-800 !default;\n$form-select-indicator:             url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16'><path fill='none' stroke='#{$form-select-indicator-color}' stroke-linecap='round' stroke-linejoin='round' stroke-width='2' d='m2 5 6 6 6-6'/></svg>\") !default;\n\n$form-select-feedback-icon-padding-end: $form-select-padding-x * 2.5 + $form-select-indicator-padding !default;\n$form-select-feedback-icon-position:    center right $form-select-indicator-padding !default;\n$form-select-feedback-icon-size:        $input-height-inner-half $input-height-inner-half !default;\n\n$form-select-border-width:        $input-border-width !default;\n$form-select-border-color:        $input-border-color !default;\n$form-select-border-radius:       $input-border-radius !default;\n$form-select-box-shadow:          var(--#{$prefix}box-shadow-inset) !default;\n\n$form-select-focus-border-color:  $input-focus-border-color !default;\n$form-select-focus-width:         $input-focus-width !default;\n$form-select-focus-box-shadow:    0 0 0 $form-select-focus-width $input-btn-focus-color !default;\n\n$form-select-padding-y-sm:        $input-padding-y-sm !default;\n$form-select-padding-x-sm:        $input-padding-x-sm !default;\n$form-select-font-size-sm:        $input-font-size-sm !default;\n$form-select-border-radius-sm:    $input-border-radius-sm !default;\n\n$form-select-padding-y-lg:        $input-padding-y-lg !default;\n$form-select-padding-x-lg:        $input-padding-x-lg !default;\n$form-select-font-size-lg:        $input-font-size-lg !default;\n$form-select-border-radius-lg:    $input-border-radius-lg !default;\n\n$form-select-transition:          $input-transition !default;\n// scss-docs-end form-select-variables\n\n// scss-docs-start form-range-variables\n$form-range-track-width:          100% !default;\n$form-range-track-height:         .5rem !default;\n$form-range-track-cursor:         pointer !default;\n$form-range-track-bg:             var(--#{$prefix}secondary-bg) !default;\n$form-range-track-border-radius:  1rem !default;\n$form-range-track-box-shadow:     var(--#{$prefix}box-shadow-inset) !default;\n\n$form-range-thumb-width:                   1rem !default;\n$form-range-thumb-height:                  $form-range-thumb-width !default;\n$form-range-thumb-bg:                      $component-active-bg !default;\n$form-range-thumb-border:                  0 !default;\n$form-range-thumb-border-radius:           1rem !default;\n$form-range-thumb-box-shadow:              0 .1rem .25rem rgba($black, .1) !default;\n$form-range-thumb-focus-box-shadow:        0 0 0 1px $body-bg, $input-focus-box-shadow !default;\n$form-range-thumb-focus-box-shadow-width:  $input-focus-width !default; // For focus box shadow issue in Edge\n$form-range-thumb-active-bg:               tint-color($component-active-bg, 70%) !default;\n$form-range-thumb-disabled-bg:             var(--#{$prefix}secondary-color) !default;\n$form-range-thumb-transition:              background-color .15s ease-in-out, border-color .15s ease-in-out, box-shadow .15s ease-in-out !default;\n// scss-docs-end form-range-variables\n\n// scss-docs-start form-file-variables\n$form-file-button-color:          $input-color !default;\n$form-file-button-bg:             var(--#{$prefix}tertiary-bg) !default;\n$form-file-button-hover-bg:       var(--#{$prefix}secondary-bg) !default;\n// scss-docs-end form-file-variables\n\n// scss-docs-start form-floating-variables\n$form-floating-height:                  add(3.5rem, $input-height-border) !default;\n$form-floating-line-height:             1.25 !default;\n$form-floating-padding-x:               $input-padding-x !default;\n$form-floating-padding-y:               1rem !default;\n$form-floating-input-padding-t:         1.625rem !default;\n$form-floating-input-padding-b:         .625rem !default;\n$form-floating-label-height:            1.5em !default;\n$form-floating-label-opacity:           .65 !default;\n$form-floating-label-transform:         scale(.85) translateY(-.5rem) translateX(.15rem) !default;\n$form-floating-label-disabled-color:    $gray-600 !default;\n$form-floating-transition:              opacity .1s ease-in-out, transform .1s ease-in-out !default;\n// scss-docs-end form-floating-variables\n\n// Form validation\n\n// scss-docs-start form-feedback-variables\n$form-feedback-margin-top:          $form-text-margin-top !default;\n$form-feedback-font-size:           $form-text-font-size !default;\n$form-feedback-font-style:          $form-text-font-style !default;\n$form-feedback-valid-color:         $success !default;\n$form-feedback-invalid-color:       $danger !default;\n\n$form-feedback-icon-valid-color:    $form-feedback-valid-color !default;\n$form-feedback-icon-valid:          url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 8 8'><path fill='#{$form-feedback-icon-valid-color}' d='M2.3 6.73.6 4.53c-.4-1.04.46-1.4 1.1-.8l1.1 1.4 3.4-3.8c.6-.63 1.6-.27 1.2.7l-4 4.6c-.43.5-.8.4-1.1.1z'/></svg>\") !default;\n$form-feedback-icon-invalid-color:  $form-feedback-invalid-color !default;\n$form-feedback-icon-invalid:        url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 12 12' width='12' height='12' fill='none' stroke='#{$form-feedback-icon-invalid-color}'><circle cx='6' cy='6' r='4.5'/><path stroke-linejoin='round' d='M5.8 3.6h.4L6 6.5z'/><circle cx='6' cy='8.2' r='.6' fill='#{$form-feedback-icon-invalid-color}' stroke='none'/></svg>\") !default;\n// scss-docs-end form-feedback-variables\n\n// scss-docs-start form-validation-colors\n$form-valid-color:                  $form-feedback-valid-color !default;\n$form-valid-border-color:           $form-feedback-valid-color !default;\n$form-invalid-color:                $form-feedback-invalid-color !default;\n$form-invalid-border-color:         $form-feedback-invalid-color !default;\n// scss-docs-end form-validation-colors\n\n// scss-docs-start form-validation-states\n$form-validation-states: (\n  \"valid\": (\n    \"color\": var(--#{$prefix}form-valid-color),\n    \"icon\": $form-feedback-icon-valid,\n    \"tooltip-color\": #fff,\n    \"tooltip-bg-color\": var(--#{$prefix}success),\n    \"focus-box-shadow\": 0 0 $input-btn-focus-blur $input-focus-width rgba(var(--#{$prefix}success-rgb), $input-btn-focus-color-opacity),\n    \"border-color\": var(--#{$prefix}form-valid-border-color),\n  ),\n  \"invalid\": (\n    \"color\": var(--#{$prefix}form-invalid-color),\n    \"icon\": $form-feedback-icon-invalid,\n    \"tooltip-color\": #fff,\n    \"tooltip-bg-color\": var(--#{$prefix}danger),\n    \"focus-box-shadow\": 0 0 $input-btn-focus-blur $input-focus-width rgba(var(--#{$prefix}danger-rgb), $input-btn-focus-color-opacity),\n    \"border-color\": var(--#{$prefix}form-invalid-border-color),\n  )\n) !default;\n// scss-docs-end form-validation-states\n\n// Z-index master list\n//\n// Warning: Avoid customizing these values. They're used for a bird's eye view\n// of components dependent on the z-axis and are designed to all work together.\n\n// scss-docs-start zindex-stack\n$zindex-dropdown:                   1000 !default;\n$zindex-sticky:                     1020 !default;\n$zindex-fixed:                      1030 !default;\n$zindex-offcanvas-backdrop:         1040 !default;\n$zindex-offcanvas:                  1045 !default;\n$zindex-modal-backdrop:             1050 !default;\n$zindex-modal:                      1055 !default;\n$zindex-popover:                    1070 !default;\n$zindex-tooltip:                    1080 !default;\n$zindex-toast:                      1090 !default;\n// scss-docs-end zindex-stack\n\n// scss-docs-start zindex-levels-map\n$zindex-levels: (\n  n1: -1,\n  0: 0,\n  1: 1,\n  2: 2,\n  3: 3\n) !default;\n// scss-docs-end zindex-levels-map\n\n\n// Navs\n\n// scss-docs-start nav-variables\n$nav-link-padding-y:                .5rem !default;\n$nav-link-padding-x:                1rem !default;\n$nav-link-font-size:                null !default;\n$nav-link-font-weight:              null !default;\n$nav-link-color:                    var(--#{$prefix}link-color) !default;\n$nav-link-hover-color:              var(--#{$prefix}link-hover-color) !default;\n$nav-link-transition:               color .15s ease-in-out, background-color .15s ease-in-out, border-color .15s ease-in-out !default;\n$nav-link-disabled-color:           var(--#{$prefix}secondary-color) !default;\n$nav-link-focus-box-shadow:         $focus-ring-box-shadow !default;\n\n$nav-tabs-border-color:             var(--#{$prefix}border-color) !default;\n$nav-tabs-border-width:             var(--#{$prefix}border-width) !default;\n$nav-tabs-border-radius:            var(--#{$prefix}border-radius) !default;\n$nav-tabs-link-hover-border-color:  var(--#{$prefix}secondary-bg) var(--#{$prefix}secondary-bg) $nav-tabs-border-color !default;\n$nav-tabs-link-active-color:        var(--#{$prefix}emphasis-color) !default;\n$nav-tabs-link-active-bg:           var(--#{$prefix}body-bg) !default;\n$nav-tabs-link-active-border-color: var(--#{$prefix}border-color) var(--#{$prefix}border-color) $nav-tabs-link-active-bg !default;\n\n$nav-pills-border-radius:           var(--#{$prefix}border-radius) !default;\n$nav-pills-link-active-color:       $component-active-color !default;\n$nav-pills-link-active-bg:          $component-active-bg !default;\n\n$nav-underline-gap:                 1rem !default;\n$nav-underline-border-width:        .125rem !default;\n$nav-underline-link-active-color:   var(--#{$prefix}emphasis-color) !default;\n// scss-docs-end nav-variables\n\n\n// Navbar\n\n// scss-docs-start navbar-variables\n$navbar-padding-y:                  $spacer * .5 !default;\n$navbar-padding-x:                  null !default;\n\n$navbar-nav-link-padding-x:         .5rem !default;\n\n$navbar-brand-font-size:            $font-size-lg !default;\n// Compute the navbar-brand padding-y so the navbar-brand will have the same height as navbar-text and nav-link\n$nav-link-height:                   $font-size-base * $line-height-base + $nav-link-padding-y * 2 !default;\n$navbar-brand-height:               $navbar-brand-font-size * $line-height-base !default;\n$navbar-brand-padding-y:            ($nav-link-height - $navbar-brand-height) * .5 !default;\n$navbar-brand-margin-end:           1rem !default;\n\n$navbar-toggler-padding-y:          .25rem !default;\n$navbar-toggler-padding-x:          .75rem !default;\n$navbar-toggler-font-size:          $font-size-lg !default;\n$navbar-toggler-border-radius:      $btn-border-radius !default;\n$navbar-toggler-focus-width:        $btn-focus-width !default;\n$navbar-toggler-transition:         box-shadow .15s ease-in-out !default;\n\n$navbar-light-color:                rgba(var(--#{$prefix}emphasis-color-rgb), .65) !default;\n$navbar-light-hover-color:          rgba(var(--#{$prefix}emphasis-color-rgb), .8) !default;\n$navbar-light-active-color:         rgba(var(--#{$prefix}emphasis-color-rgb), 1) !default;\n$navbar-light-disabled-color:       rgba(var(--#{$prefix}emphasis-color-rgb), .3) !default;\n$navbar-light-icon-color:           rgba($body-color, .75) !default;\n$navbar-light-toggler-icon-bg:      url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 30 30'><path stroke='#{$navbar-light-icon-color}' stroke-linecap='round' stroke-miterlimit='10' stroke-width='2' d='M4 7h22M4 15h22M4 23h22'/></svg>\") !default;\n$navbar-light-toggler-border-color: rgba(var(--#{$prefix}emphasis-color-rgb), .15) !default;\n$navbar-light-brand-color:          $navbar-light-active-color !default;\n$navbar-light-brand-hover-color:    $navbar-light-active-color !default;\n// scss-docs-end navbar-variables\n\n// scss-docs-start navbar-dark-variables\n$navbar-dark-color:                 rgba($white, .55) !default;\n$navbar-dark-hover-color:           rgba($white, .75) !default;\n$navbar-dark-active-color:          $white !default;\n$navbar-dark-disabled-color:        rgba($white, .25) !default;\n$navbar-dark-icon-color:            $navbar-dark-color !default;\n$navbar-dark-toggler-icon-bg:       url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 30 30'><path stroke='#{$navbar-dark-icon-color}' stroke-linecap='round' stroke-miterlimit='10' stroke-width='2' d='M4 7h22M4 15h22M4 23h22'/></svg>\") !default;\n$navbar-dark-toggler-border-color:  rgba($white, .1) !default;\n$navbar-dark-brand-color:           $navbar-dark-active-color !default;\n$navbar-dark-brand-hover-color:     $navbar-dark-active-color !default;\n// scss-docs-end navbar-dark-variables\n\n\n// Dropdowns\n//\n// Dropdown menu container and contents.\n\n// scss-docs-start dropdown-variables\n$dropdown-min-width:                10rem !default;\n$dropdown-padding-x:                0 !default;\n$dropdown-padding-y:                .5rem !default;\n$dropdown-spacer:                   .125rem !default;\n$dropdown-font-size:                $font-size-base !default;\n$dropdown-color:                    var(--#{$prefix}body-color) !default;\n$dropdown-bg:                       var(--#{$prefix}body-bg) !default;\n$dropdown-border-color:             var(--#{$prefix}border-color-translucent) !default;\n$dropdown-border-radius:            var(--#{$prefix}border-radius) !default;\n$dropdown-border-width:             var(--#{$prefix}border-width) !default;\n$dropdown-inner-border-radius:      calc(#{$dropdown-border-radius} - #{$dropdown-border-width}) !default; // stylelint-disable-line function-disallowed-list\n$dropdown-divider-bg:               $dropdown-border-color !default;\n$dropdown-divider-margin-y:         $spacer * .5 !default;\n$dropdown-box-shadow:               var(--#{$prefix}box-shadow) !default;\n\n$dropdown-link-color:               var(--#{$prefix}body-color) !default;\n$dropdown-link-hover-color:         $dropdown-link-color !default;\n$dropdown-link-hover-bg:            var(--#{$prefix}tertiary-bg) !default;\n\n$dropdown-link-active-color:        $component-active-color !default;\n$dropdown-link-active-bg:           $component-active-bg !default;\n\n$dropdown-link-disabled-color:      var(--#{$prefix}tertiary-color) !default;\n\n$dropdown-item-padding-y:           $spacer * .25 !default;\n$dropdown-item-padding-x:           $spacer !default;\n\n$dropdown-header-color:             $gray-600 !default;\n$dropdown-header-padding-x:         $dropdown-item-padding-x !default;\n$dropdown-header-padding-y:         $dropdown-padding-y !default;\n// fusv-disable\n$dropdown-header-padding:           $dropdown-header-padding-y $dropdown-header-padding-x !default; // Deprecated in v5.2.0\n// fusv-enable\n// scss-docs-end dropdown-variables\n\n// scss-docs-start dropdown-dark-variables\n$dropdown-dark-color:               $gray-300 !default;\n$dropdown-dark-bg:                  $gray-800 !default;\n$dropdown-dark-border-color:        $dropdown-border-color !default;\n$dropdown-dark-divider-bg:          $dropdown-divider-bg !default;\n$dropdown-dark-box-shadow:          null !default;\n$dropdown-dark-link-color:          $dropdown-dark-color !default;\n$dropdown-dark-link-hover-color:    $white !default;\n$dropdown-dark-link-hover-bg:       rgba($white, .15) !default;\n$dropdown-dark-link-active-color:   $dropdown-link-active-color !default;\n$dropdown-dark-link-active-bg:      $dropdown-link-active-bg !default;\n$dropdown-dark-link-disabled-color: $gray-500 !default;\n$dropdown-dark-header-color:        $gray-500 !default;\n// scss-docs-end dropdown-dark-variables\n\n\n// Pagination\n\n// scss-docs-start pagination-variables\n$pagination-padding-y:              .375rem !default;\n$pagination-padding-x:              .75rem !default;\n$pagination-padding-y-sm:           .25rem !default;\n$pagination-padding-x-sm:           .5rem !default;\n$pagination-padding-y-lg:           .75rem !default;\n$pagination-padding-x-lg:           1.5rem !default;\n\n$pagination-font-size:              $font-size-base !default;\n\n$pagination-color:                  var(--#{$prefix}link-color) !default;\n$pagination-bg:                     var(--#{$prefix}body-bg) !default;\n$pagination-border-radius:          var(--#{$prefix}border-radius) !default;\n$pagination-border-width:           var(--#{$prefix}border-width) !default;\n$pagination-margin-start:           calc(#{$pagination-border-width} * -1) !default; // stylelint-disable-line function-disallowed-list\n$pagination-border-color:           var(--#{$prefix}border-color) !default;\n\n$pagination-focus-color:            var(--#{$prefix}link-hover-color) !default;\n$pagination-focus-bg:               var(--#{$prefix}secondary-bg) !default;\n$pagination-focus-box-shadow:       $focus-ring-box-shadow !default;\n$pagination-focus-outline:          0 !default;\n\n$pagination-hover-color:            var(--#{$prefix}link-hover-color) !default;\n$pagination-hover-bg:               var(--#{$prefix}tertiary-bg) !default;\n$pagination-hover-border-color:     var(--#{$prefix}border-color) !default; // Todo in v6: remove this?\n\n$pagination-active-color:           $component-active-color !default;\n$pagination-active-bg:              $component-active-bg !default;\n$pagination-active-border-color:    $component-active-bg !default;\n\n$pagination-disabled-color:         var(--#{$prefix}secondary-color) !default;\n$pagination-disabled-bg:            var(--#{$prefix}secondary-bg) !default;\n$pagination-disabled-border-color:  var(--#{$prefix}border-color) !default;\n\n$pagination-transition:              color .15s ease-in-out, background-color .15s ease-in-out, border-color .15s ease-in-out, box-shadow .15s ease-in-out !default;\n\n$pagination-border-radius-sm:       var(--#{$prefix}border-radius-sm) !default;\n$pagination-border-radius-lg:       var(--#{$prefix}border-radius-lg) !default;\n// scss-docs-end pagination-variables\n\n\n// Placeholders\n\n// scss-docs-start placeholders\n$placeholder-opacity-max:           .5 !default;\n$placeholder-opacity-min:           .2 !default;\n// scss-docs-end placeholders\n\n// Cards\n\n// scss-docs-start card-variables\n$card-spacer-y:                     $spacer !default;\n$card-spacer-x:                     $spacer !default;\n$card-title-spacer-y:               $spacer * .5 !default;\n$card-title-color:                  null !default;\n$card-subtitle-color:               null !default;\n$card-border-width:                 var(--#{$prefix}border-width) !default;\n$card-border-color:                 var(--#{$prefix}border-color-translucent) !default;\n$card-border-radius:                var(--#{$prefix}border-radius) !default;\n$card-box-shadow:                   null !default;\n$card-inner-border-radius:          subtract($card-border-radius, $card-border-width) !default;\n$card-cap-padding-y:                $card-spacer-y * .5 !default;\n$card-cap-padding-x:                $card-spacer-x !default;\n$card-cap-bg:                       rgba(var(--#{$prefix}body-color-rgb), .03) !default;\n$card-cap-color:                    null !default;\n$card-height:                       null !default;\n$card-color:                        null !default;\n$card-bg:                           var(--#{$prefix}body-bg) !default;\n$card-img-overlay-padding:          $spacer !default;\n$card-group-margin:                 $grid-gutter-width * .5 !default;\n// scss-docs-end card-variables\n\n// Accordion\n\n// scss-docs-start accordion-variables\n$accordion-padding-y:                     1rem !default;\n$accordion-padding-x:                     1.25rem !default;\n$accordion-color:                         var(--#{$prefix}body-color) !default;\n$accordion-bg:                            var(--#{$prefix}body-bg) !default;\n$accordion-border-width:                  var(--#{$prefix}border-width) !default;\n$accordion-border-color:                  var(--#{$prefix}border-color) !default;\n$accordion-border-radius:                 var(--#{$prefix}border-radius) !default;\n$accordion-inner-border-radius:           subtract($accordion-border-radius, $accordion-border-width) !default;\n\n$accordion-body-padding-y:                $accordion-padding-y !default;\n$accordion-body-padding-x:                $accordion-padding-x !default;\n\n$accordion-button-padding-y:              $accordion-padding-y !default;\n$accordion-button-padding-x:              $accordion-padding-x !default;\n$accordion-button-color:                  var(--#{$prefix}body-color) !default;\n$accordion-button-bg:                     var(--#{$prefix}accordion-bg) !default;\n$accordion-transition:                    $btn-transition, border-radius .15s ease !default;\n$accordion-button-active-bg:              var(--#{$prefix}primary-bg-subtle) !default;\n$accordion-button-active-color:           var(--#{$prefix}primary-text-emphasis) !default;\n\n// fusv-disable\n$accordion-button-focus-border-color:     $input-focus-border-color !default; // Deprecated in v5.3.3\n// fusv-enable\n$accordion-button-focus-box-shadow:       $btn-focus-box-shadow !default;\n\n$accordion-icon-width:                    1.25rem !default;\n$accordion-icon-color:                    $body-color !default;\n$accordion-icon-active-color:             $primary-text-emphasis !default;\n$accordion-icon-transition:               transform .2s ease-in-out !default;\n$accordion-icon-transform:                rotate(-180deg) !default;\n\n$accordion-button-icon:         url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='none' stroke='#{$accordion-icon-color}' stroke-linecap='round' stroke-linejoin='round'><path d='M2 5L8 11L14 5'/></svg>\") !default;\n$accordion-button-active-icon:  url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='none' stroke='#{$accordion-icon-active-color}' stroke-linecap='round' stroke-linejoin='round'><path d='M2 5L8 11L14 5'/></svg>\") !default;\n// scss-docs-end accordion-variables\n\n// Tooltips\n\n// scss-docs-start tooltip-variables\n$tooltip-font-size:                 $font-size-sm !default;\n$tooltip-max-width:                 200px !default;\n$tooltip-color:                     var(--#{$prefix}body-bg) !default;\n$tooltip-bg:                        var(--#{$prefix}emphasis-color) !default;\n$tooltip-border-radius:             var(--#{$prefix}border-radius) !default;\n$tooltip-opacity:                   .9 !default;\n$tooltip-padding-y:                 $spacer * .25 !default;\n$tooltip-padding-x:                 $spacer * .5 !default;\n$tooltip-margin:                    null !default; // TODO: remove this in v6\n\n$tooltip-arrow-width:               .8rem !default;\n$tooltip-arrow-height:              .4rem !default;\n// fusv-disable\n$tooltip-arrow-color:               null !default; // Deprecated in Bootstrap 5.2.0 for CSS variables\n// fusv-enable\n// scss-docs-end tooltip-variables\n\n// Form tooltips must come after regular tooltips\n// scss-docs-start tooltip-feedback-variables\n$form-feedback-tooltip-padding-y:     $tooltip-padding-y !default;\n$form-feedback-tooltip-padding-x:     $tooltip-padding-x !default;\n$form-feedback-tooltip-font-size:     $tooltip-font-size !default;\n$form-feedback-tooltip-line-height:   null !default;\n$form-feedback-tooltip-opacity:       $tooltip-opacity !default;\n$form-feedback-tooltip-border-radius: $tooltip-border-radius !default;\n// scss-docs-end tooltip-feedback-variables\n\n\n// Popovers\n\n// scss-docs-start popover-variables\n$popover-font-size:                 $font-size-sm !default;\n$popover-bg:                        var(--#{$prefix}body-bg) !default;\n$popover-max-width:                 276px !default;\n$popover-border-width:              var(--#{$prefix}border-width) !default;\n$popover-border-color:              var(--#{$prefix}border-color-translucent) !default;\n$popover-border-radius:             var(--#{$prefix}border-radius-lg) !default;\n$popover-inner-border-radius:       calc(#{$popover-border-radius} - #{$popover-border-width}) !default; // stylelint-disable-line function-disallowed-list\n$popover-box-shadow:                var(--#{$prefix}box-shadow) !default;\n\n$popover-header-font-size:          $font-size-base !default;\n$popover-header-bg:                 var(--#{$prefix}secondary-bg) !default;\n$popover-header-color:              $headings-color !default;\n$popover-header-padding-y:          .5rem !default;\n$popover-header-padding-x:          $spacer !default;\n\n$popover-body-color:                var(--#{$prefix}body-color) !default;\n$popover-body-padding-y:            $spacer !default;\n$popover-body-padding-x:            $spacer !default;\n\n$popover-arrow-width:               1rem !default;\n$popover-arrow-height:              .5rem !default;\n// scss-docs-end popover-variables\n\n// fusv-disable\n// Deprecated in Bootstrap 5.2.0 for CSS variables\n$popover-arrow-color:               $popover-bg !default;\n$popover-arrow-outer-color:         var(--#{$prefix}border-color-translucent) !default;\n// fusv-enable\n\n\n// Toasts\n\n// scss-docs-start toast-variables\n$toast-max-width:                   350px !default;\n$toast-padding-x:                   .75rem !default;\n$toast-padding-y:                   .5rem !default;\n$toast-font-size:                   .875rem !default;\n$toast-color:                       null !default;\n$toast-background-color:            rgba(var(--#{$prefix}body-bg-rgb), .85) !default;\n$toast-border-width:                var(--#{$prefix}border-width) !default;\n$toast-border-color:                var(--#{$prefix}border-color-translucent) !default;\n$toast-border-radius:               var(--#{$prefix}border-radius) !default;\n$toast-box-shadow:                  var(--#{$prefix}box-shadow) !default;\n$toast-spacing:                     $container-padding-x !default;\n\n$toast-header-color:                var(--#{$prefix}secondary-color) !default;\n$toast-header-background-color:     rgba(var(--#{$prefix}body-bg-rgb), .85) !default;\n$toast-header-border-color:         $toast-border-color !default;\n// scss-docs-end toast-variables\n\n\n// Badges\n\n// scss-docs-start badge-variables\n$badge-font-size:                   .75em !default;\n$badge-font-weight:                 $font-weight-bold !default;\n$badge-color:                       $white !default;\n$badge-padding-y:                   .35em !default;\n$badge-padding-x:                   .65em !default;\n$badge-border-radius:               var(--#{$prefix}border-radius) !default;\n// scss-docs-end badge-variables\n\n\n// Modals\n\n// scss-docs-start modal-variables\n$modal-inner-padding:               $spacer !default;\n\n$modal-footer-margin-between:       .5rem !default;\n\n$modal-dialog-margin:               .5rem !default;\n$modal-dialog-margin-y-sm-up:       1.75rem !default;\n\n$modal-title-line-height:           $line-height-base !default;\n\n$modal-content-color:               null !default;\n$modal-content-bg:                  var(--#{$prefix}body-bg) !default;\n$modal-content-border-color:        var(--#{$prefix}border-color-translucent) !default;\n$modal-content-border-width:        var(--#{$prefix}border-width) !default;\n$modal-content-border-radius:       var(--#{$prefix}border-radius-lg) !default;\n$modal-content-inner-border-radius: subtract($modal-content-border-radius, $modal-content-border-width) !default;\n$modal-content-box-shadow-xs:       var(--#{$prefix}box-shadow-sm) !default;\n$modal-content-box-shadow-sm-up:    var(--#{$prefix}box-shadow) !default;\n\n$modal-backdrop-bg:                 $black !default;\n$modal-backdrop-opacity:            .5 !default;\n\n$modal-header-border-color:         var(--#{$prefix}border-color) !default;\n$modal-header-border-width:         $modal-content-border-width !default;\n$modal-header-padding-y:            $modal-inner-padding !default;\n$modal-header-padding-x:            $modal-inner-padding !default;\n$modal-header-padding:              $modal-header-padding-y $modal-header-padding-x !default; // Keep this for backwards compatibility\n\n$modal-footer-bg:                   null !default;\n$modal-footer-border-color:         $modal-header-border-color !default;\n$modal-footer-border-width:         $modal-header-border-width !default;\n\n$modal-sm:                          300px !default;\n$modal-md:                          500px !default;\n$modal-lg:                          800px !default;\n$modal-xl:                          1140px !default;\n\n$modal-fade-transform:              translate(0, -50px) !default;\n$modal-show-transform:              none !default;\n$modal-transition:                  transform .3s ease-out !default;\n$modal-scale-transform:             scale(1.02) !default;\n// scss-docs-end modal-variables\n\n\n// Alerts\n//\n// Define alert colors, border radius, and padding.\n\n// scss-docs-start alert-variables\n$alert-padding-y:               $spacer !default;\n$alert-padding-x:               $spacer !default;\n$alert-margin-bottom:           1rem !default;\n$alert-border-radius:           var(--#{$prefix}border-radius) !default;\n$alert-link-font-weight:        $font-weight-bold !default;\n$alert-border-width:            var(--#{$prefix}border-width) !default;\n$alert-dismissible-padding-r:   $alert-padding-x * 3 !default; // 3x covers width of x plus default padding on either side\n// scss-docs-end alert-variables\n\n// fusv-disable\n$alert-bg-scale:                -80% !default; // Deprecated in v5.2.0, to be removed in v6\n$alert-border-scale:            -70% !default; // Deprecated in v5.2.0, to be removed in v6\n$alert-color-scale:             40% !default; // Deprecated in v5.2.0, to be removed in v6\n// fusv-enable\n\n// Progress bars\n\n// scss-docs-start progress-variables\n$progress-height:                   1rem !default;\n$progress-font-size:                $font-size-base * .75 !default;\n$progress-bg:                       var(--#{$prefix}secondary-bg) !default;\n$progress-border-radius:            var(--#{$prefix}border-radius) !default;\n$progress-box-shadow:               var(--#{$prefix}box-shadow-inset) !default;\n$progress-bar-color:                $white !default;\n$progress-bar-bg:                   $primary !default;\n$progress-bar-animation-timing:     1s linear infinite !default;\n$progress-bar-transition:           width .6s ease !default;\n// scss-docs-end progress-variables\n\n\n// List group\n\n// scss-docs-start list-group-variables\n$list-group-color:                  var(--#{$prefix}body-color) !default;\n$list-group-bg:                     var(--#{$prefix}body-bg) !default;\n$list-group-border-color:           var(--#{$prefix}border-color) !default;\n$list-group-border-width:           var(--#{$prefix}border-width) !default;\n$list-group-border-radius:          var(--#{$prefix}border-radius) !default;\n\n$list-group-item-padding-y:         $spacer * .5 !default;\n$list-group-item-padding-x:         $spacer !default;\n// fusv-disable\n$list-group-item-bg-scale:          -80% !default; // Deprecated in v5.3.0\n$list-group-item-color-scale:       40% !default; // Deprecated in v5.3.0\n// fusv-enable\n\n$list-group-hover-bg:               var(--#{$prefix}tertiary-bg) !default;\n$list-group-active-color:           $component-active-color !default;\n$list-group-active-bg:              $component-active-bg !default;\n$list-group-active-border-color:    $list-group-active-bg !default;\n\n$list-group-disabled-color:         var(--#{$prefix}secondary-color) !default;\n$list-group-disabled-bg:            $list-group-bg !default;\n\n$list-group-action-color:           var(--#{$prefix}secondary-color) !default;\n$list-group-action-hover-color:     var(--#{$prefix}emphasis-color) !default;\n\n$list-group-action-active-color:    var(--#{$prefix}body-color) !default;\n$list-group-action-active-bg:       var(--#{$prefix}secondary-bg) !default;\n// scss-docs-end list-group-variables\n\n\n// Image thumbnails\n\n// scss-docs-start thumbnail-variables\n$thumbnail-padding:                 .25rem !default;\n$thumbnail-bg:                      var(--#{$prefix}body-bg) !default;\n$thumbnail-border-width:            var(--#{$prefix}border-width) !default;\n$thumbnail-border-color:            var(--#{$prefix}border-color) !default;\n$thumbnail-border-radius:           var(--#{$prefix}border-radius) !default;\n$thumbnail-box-shadow:              var(--#{$prefix}box-shadow-sm) !default;\n// scss-docs-end thumbnail-variables\n\n\n// Figures\n\n// scss-docs-start figure-variables\n$figure-caption-font-size:          $small-font-size !default;\n$figure-caption-color:              var(--#{$prefix}secondary-color) !default;\n// scss-docs-end figure-variables\n\n\n// Breadcrumbs\n\n// scss-docs-start breadcrumb-variables\n$breadcrumb-font-size:              null !default;\n$breadcrumb-padding-y:              0 !default;\n$breadcrumb-padding-x:              0 !default;\n$breadcrumb-item-padding-x:         .5rem !default;\n$breadcrumb-margin-bottom:          1rem !default;\n$breadcrumb-bg:                     null !default;\n$breadcrumb-divider-color:          var(--#{$prefix}secondary-color) !default;\n$breadcrumb-active-color:           var(--#{$prefix}secondary-color) !default;\n$breadcrumb-divider:                quote(\"/\") !default;\n$breadcrumb-divider-flipped:        $breadcrumb-divider !default;\n$breadcrumb-border-radius:          null !default;\n// scss-docs-end breadcrumb-variables\n\n// Carousel\n\n// scss-docs-start carousel-variables\n$carousel-control-color:             $white !default;\n$carousel-control-width:             15% !default;\n$carousel-control-opacity:           .5 !default;\n$carousel-control-hover-opacity:     .9 !default;\n$carousel-control-transition:        opacity .15s ease !default;\n\n$carousel-indicator-width:           30px !default;\n$carousel-indicator-height:          3px !default;\n$carousel-indicator-hit-area-height: 10px !default;\n$carousel-indicator-spacer:          3px !default;\n$carousel-indicator-opacity:         .5 !default;\n$carousel-indicator-active-bg:       $white !default;\n$carousel-indicator-active-opacity:  1 !default;\n$carousel-indicator-transition:      opacity .6s ease !default;\n\n$carousel-caption-width:             70% !default;\n$carousel-caption-color:             $white !default;\n$carousel-caption-padding-y:         1.25rem !default;\n$carousel-caption-spacer:            1.25rem !default;\n\n$carousel-control-icon-width:        2rem !default;\n\n$carousel-control-prev-icon-bg:      url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='#{$carousel-control-color}'><path d='M11.354 1.646a.5.5 0 0 1 0 .708L5.707 8l5.647 5.646a.5.5 0 0 1-.708.708l-6-6a.5.5 0 0 1 0-.708l6-6a.5.5 0 0 1 .708 0z'/></svg>\") !default;\n$carousel-control-next-icon-bg:      url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='#{$carousel-control-color}'><path d='M4.646 1.646a.5.5 0 0 1 .708 0l6 6a.5.5 0 0 1 0 .708l-6 6a.5.5 0 0 1-.708-.708L10.293 8 4.646 2.354a.5.5 0 0 1 0-.708z'/></svg>\") !default;\n\n$carousel-transition-duration:       .6s !default;\n$carousel-transition:                transform $carousel-transition-duration ease-in-out !default; // Define transform transition first if using multiple transitions (e.g., `transform 2s ease, opacity .5s ease-out`)\n// scss-docs-end carousel-variables\n\n// scss-docs-start carousel-dark-variables\n$carousel-dark-indicator-active-bg:  $black !default;\n$carousel-dark-caption-color:        $black !default;\n$carousel-dark-control-icon-filter:  invert(1) grayscale(100) !default;\n// scss-docs-end carousel-dark-variables\n\n\n// Spinners\n\n// scss-docs-start spinner-variables\n$spinner-width:           2rem !default;\n$spinner-height:          $spinner-width !default;\n$spinner-vertical-align:  -.125em !default;\n$spinner-border-width:    .25em !default;\n$spinner-animation-speed: .75s !default;\n\n$spinner-width-sm:        1rem !default;\n$spinner-height-sm:       $spinner-width-sm !default;\n$spinner-border-width-sm: .2em !default;\n// scss-docs-end spinner-variables\n\n\n// Close\n\n// scss-docs-start close-variables\n$btn-close-width:            1em !default;\n$btn-close-height:           $btn-close-width !default;\n$btn-close-padding-x:        .25em !default;\n$btn-close-padding-y:        $btn-close-padding-x !default;\n$btn-close-color:            $black !default;\n$btn-close-bg:               url(\"data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 16 16' fill='#{$btn-close-color}'><path d='M.293.293a1 1 0 0 1 1.414 0L8 6.586 14.293.293a1 1 0 1 1 1.414 1.414L9.414 8l6.293 6.293a1 1 0 0 1-1.414 1.414L8 9.414l-6.293 6.293a1 1 0 0 1-1.414-1.414L6.586 8 .293 1.707a1 1 0 0 1 0-1.414z'/></svg>\") !default;\n$btn-close-focus-shadow:     $focus-ring-box-shadow !default;\n$btn-close-opacity:          .5 !default;\n$btn-close-hover-opacity:    .75 !default;\n$btn-close-focus-opacity:    1 !default;\n$btn-close-disabled-opacity: .25 !default;\n$btn-close-white-filter:     invert(1) grayscale(100%) brightness(200%) !default;\n// scss-docs-end close-variables\n\n\n// Offcanvas\n\n// scss-docs-start offcanvas-variables\n$offcanvas-padding-y:               $modal-inner-padding !default;\n$offcanvas-padding-x:               $modal-inner-padding !default;\n$offcanvas-horizontal-width:        400px !default;\n$offcanvas-vertical-height:         30vh !default;\n$offcanvas-transition-duration:     .3s !default;\n$offcanvas-border-color:            $modal-content-border-color !default;\n$offcanvas-border-width:            $modal-content-border-width !default;\n$offcanvas-title-line-height:       $modal-title-line-height !default;\n$offcanvas-bg-color:                var(--#{$prefix}body-bg) !default;\n$offcanvas-color:                   var(--#{$prefix}body-color) !default;\n$offcanvas-box-shadow:              $modal-content-box-shadow-xs !default;\n$offcanvas-backdrop-bg:             $modal-backdrop-bg !default;\n$offcanvas-backdrop-opacity:        $modal-backdrop-opacity !default;\n// scss-docs-end offcanvas-variables\n\n// Code\n\n$code-font-size:                    $small-font-size !default;\n$code-color:                        $pink !default;\n\n$kbd-padding-y:                     .1875rem !default;\n$kbd-padding-x:                     .375rem !default;\n$kbd-font-size:                     $code-font-size !default;\n$kbd-color:                         var(--#{$prefix}body-bg) !default;\n$kbd-bg:                            var(--#{$prefix}body-color) !default;\n$nested-kbd-font-weight:            null !default; // Deprecated in v5.2.0, removing in v6\n\n$pre-color:                         null !default;\n\n@import \"variables-dark\"; // TODO: can be removed safely in v6, only here to avoid breaking changes in v5.3\n","// stylelint-disable property-disallowed-list\n// Single side border-radius\n\n// Helper function to replace negative values with 0\n@function valid-radius($radius) {\n  $return: ();\n  @each $value in $radius {\n    @if type-of($value) == number {\n      $return: append($return, max($value, 0));\n    } @else {\n      $return: append($return, $value);\n    }\n  }\n  @return $return;\n}\n\n// scss-docs-start border-radius-mixins\n@mixin border-radius($radius: $border-radius, $fallback-border-radius: false) {\n  @if $enable-rounded {\n    border-radius: valid-radius($radius);\n  }\n  @else if $fallback-border-radius != false {\n    border-radius: $fallback-border-radius;\n  }\n}\n\n@mixin border-top-radius($radius: $border-radius) {\n  @if $enable-rounded {\n    border-top-left-radius: valid-radius($radius);\n    border-top-right-radius: valid-radius($radius);\n  }\n}\n\n@mixin border-end-radius($radius: $border-radius) {\n  @if $enable-rounded {\n    border-top-right-radius: valid-radius($radius);\n    border-bottom-right-radius: valid-radius($radius);\n  }\n}\n\n@mixin border-bottom-radius($radius: $border-radius) {\n  @if $enable-rounded {\n    border-bottom-right-radius: valid-radius($radius);\n    border-bottom-left-radius: valid-radius($radius);\n  }\n}\n\n@mixin border-start-radius($radius: $border-radius) {\n  @if $enable-rounded {\n    border-top-left-radius: valid-radius($radius);\n    border-bottom-left-radius: valid-radius($radius);\n  }\n}\n\n@mixin border-top-start-radius($radius: $border-radius) {\n  @if $enable-rounded {\n    border-top-left-radius: valid-radius($radius);\n  }\n}\n\n@mixin border-top-end-radius($radius: $border-radius) {\n  @if $enable-rounded {\n    border-top-right-radius: valid-radius($radius);\n  }\n}\n\n@mixin border-bottom-end-radius($radius: $border-radius) {\n  @if $enable-rounded {\n    border-bottom-right-radius: valid-radius($radius);\n  }\n}\n\n@mixin border-bottom-start-radius($radius: $border-radius) {\n  @if $enable-rounded {\n    border-bottom-left-radius: valid-radius($radius);\n  }\n}\n// scss-docs-end border-radius-mixins\n","// Lists\n\n// Unstyled keeps list items block level, just removes default browser padding and list-style\n@mixin list-unstyled {\n  padding-left: 0;\n  list-style: none;\n}\n","// Override bootstrap variables\n$spacer: 1rem;\n$container-max-widths: (\n  sm: 540px,\n  md: 720px,\n  lg: 960px,\n  xl: 1400px,\n);\n$grid-breakpoints: (\n  xs: 0,\n  sm: 540px,\n  md: 720px,\n  lg: 960px,\n  xl: 1200px,\n);\n$dropdown-link-hover-bg: var(--pst-color-surface);\n\n// --pst-color-surface can also be assigned to the dark variant because it is\n// scoped to different values depending on light/dark theme\n$dropdown-dark-link-hover-bg: var(--pst-color-surface);\n$dropdown-link-active-bg: var(--pst-color-surface);\n$dropdown-dark-link-active-bg: var(--pst-color-surface);\n$focus-ring-width: 0.1875rem; // 3px at 100% zoom (0.1875 * 16px base font = 3px)\n$focus-ring-opacity: 1;\n$focus-ring-color: var(--pst-color-accent);\n$focus-ring-blur: 0;\n$focus-ring-box-shadow: 0 0 $focus-ring-blur $focus-ring-width $focus-ring-color;\n\n// For many elements, we do not use `outline-offset`. For some we set the offset\n// equal to the focus ring width (either outwards or inwards). But for some\n// other elements (e.g., collapsible admonitions) we set it to this value.\n$focus-ring-offset: 0.125rem; // 2px at 100% zoom (0.125 * 16px base font = 2px)\n\n// outline creates the same style of focus ring, it just uses CSS outline instead of box shadow\n$focus-ring-outline: $focus-ring-color solid $focus-ring-width;\n$btn-focus-box-shadow: $focus-ring-box-shadow;\n","// Responsive images (ensure images don't scale beyond their parents)\n//\n// This is purposefully opt-in via an explicit class rather than being the default for all `<img>`s.\n// We previously tried the \"images are responsive by default\" approach in Bootstrap v2,\n// and abandoned it in Bootstrap v3 because it breaks lots of third-party widgets (including Google Maps)\n// which weren't expecting the images within themselves to be involuntarily resized.\n// See also https://github.com/twbs/bootstrap/issues/18178\n.img-fluid {\n  @include img-fluid();\n}\n\n\n// Image thumbnails\n.img-thumbnail {\n  padding: $thumbnail-padding;\n  background-color: $thumbnail-bg;\n  border: $thumbnail-border-width solid $thumbnail-border-color;\n  @include border-radius($thumbnail-border-radius);\n  @include box-shadow($thumbnail-box-shadow);\n\n  // Keep them at most 100% wide\n  @include img-fluid();\n}\n\n//\n// Figures\n//\n\n.figure {\n  // Ensures the caption's text aligns with the image.\n  display: inline-block;\n}\n\n.figure-img {\n  margin-bottom: $spacer * .5;\n  line-height: 1;\n}\n\n.figure-caption {\n  @include font-size($figure-caption-font-size);\n  color: $figure-caption-color;\n}\n","// Image Mixins\n// - Responsive image\n// - Retina image\n\n\n// Responsive image\n//\n// Keep images from scaling beyond the width of their parents.\n\n@mixin img-fluid {\n  // Part 1: Set a maximum relative to the parent\n  max-width: 100%;\n  // Part 2: Override the height to auto, otherwise images will be stretched\n  // when setting a width and height attribute on the img element.\n  height: auto;\n}\n","// Container widths\n//\n// Set the container width, and override it for fixed navbars in media queries.\n\n@if $enable-container-classes {\n  // Single container class with breakpoint max-widths\n  .container,\n  // 100% wide container at all breakpoints\n  .container-fluid {\n    @include make-container();\n  }\n\n  // Responsive containers that are 100% wide until a breakpoint\n  @each $breakpoint, $container-max-width in $container-max-widths {\n    .container-#{$breakpoint} {\n      @extend .container-fluid;\n    }\n\n    @include media-breakpoint-up($breakpoint, $grid-breakpoints) {\n      %responsive-container-#{$breakpoint} {\n        max-width: $container-max-width;\n      }\n\n      // Extend each breakpoint which is smaller or equal to the current breakpoint\n      $extend-breakpoint: true;\n\n      @each $name, $width in $grid-breakpoints {\n        @if ($extend-breakpoint) {\n          .container#{breakpoint-infix($name, $grid-breakpoints)} {\n            @extend %responsive-container-#{$breakpoint};\n          }\n\n          // Once the current breakpoint is reached, stop extending\n          @if ($breakpoint == $name) {\n            $extend-breakpoint: false;\n          }\n        }\n      }\n    }\n  }\n}\n","// Container mixins\n\n@mixin make-container($gutter: $container-padding-x) {\n  --#{$prefix}gutter-x: #{$gutter};\n  --#{$prefix}gutter-y: 0;\n  width: 100%;\n  padding-right: calc(var(--#{$prefix}gutter-x) * .5); // stylelint-disable-line function-disallowed-list\n  padding-left: calc(var(--#{$prefix}gutter-x) * .5); // stylelint-disable-line function-disallowed-list\n  margin-right: auto;\n  margin-left: auto;\n}\n","// Breakpoint viewport sizes and media queries.\n//\n// Breakpoints are defined as a map of (name: minimum width), order from small to large:\n//\n//    (xs: 0, sm: 576px, md: 768px, lg: 992px, xl: 1200px, xxl: 1400px)\n//\n// The map defined in the `$grid-breakpoints` global variable is used as the `$breakpoints` argument by default.\n\n// Name of the next breakpoint, or null for the last breakpoint.\n//\n//    >> breakpoint-next(sm)\n//    md\n//    >> breakpoint-next(sm, (xs: 0, sm: 576px, md: 768px, lg: 992px, xl: 1200px, xxl: 1400px))\n//    md\n//    >> breakpoint-next(sm, $breakpoint-names: (xs sm md lg xl xxl))\n//    md\n@function breakpoint-next($name, $breakpoints: $grid-breakpoints, $breakpoint-names: map-keys($breakpoints)) {\n  $n: index($breakpoint-names, $name);\n  @if not $n {\n    @error \"breakpoint `#{$name}` not found in `#{$breakpoints}`\";\n  }\n  @return if($n < length($breakpoint-names), nth($breakpoint-names, $n + 1), null);\n}\n\n// Minimum breakpoint width. Null for the smallest (first) breakpoint.\n//\n//    >> breakpoint-min(sm, (xs: 0, sm: 576px, md: 768px, lg: 992px, xl: 1200px, xxl: 1400px))\n//    576px\n@function breakpoint-min($name, $breakpoints: $grid-breakpoints) {\n  $min: map-get($breakpoints, $name);\n  @return if($min != 0, $min, null);\n}\n\n// Maximum breakpoint width.\n// The maximum value is reduced by 0.02px to work around the limitations of\n// `min-` and `max-` prefixes and viewports with fractional widths.\n// See https://www.w3.org/TR/mediaqueries-4/#mq-min-max\n// Uses 0.02px rather than 0.01px to work around a current rounding bug in Safari.\n// See https://bugs.webkit.org/show_bug.cgi?id=178261\n//\n//    >> breakpoint-max(md, (xs: 0, sm: 576px, md: 768px, lg: 992px, xl: 1200px, xxl: 1400px))\n//    767.98px\n@function breakpoint-max($name, $breakpoints: $grid-breakpoints) {\n  $max: map-get($breakpoints, $name);\n  @return if($max and $max > 0, $max - .02, null);\n}\n\n// Returns a blank string if smallest breakpoint, otherwise returns the name with a dash in front.\n// Useful for making responsive utilities.\n//\n//    >> breakpoint-infix(xs, (xs: 0, sm: 576px, md: 768px, lg: 992px, xl: 1200px, xxl: 1400px))\n//    \"\"  (Returns a blank string)\n//    >> breakpoint-infix(sm, (xs: 0, sm: 576px, md: 768px, lg: 992px, xl: 1200px, xxl: 1400px))\n//    \"-sm\"\n@function breakpoint-infix($name, $breakpoints: $grid-breakpoints) {\n  @return if(breakpoint-min($name, $breakpoints) == null, \"\", \"-#{$name}\");\n}\n\n// Media of at least the minimum breakpoint width. No query for the smallest breakpoint.\n// Makes the @content apply to the given breakpoint and wider.\n@mixin media-breakpoint-up($name, $breakpoints: $grid-breakpoints) {\n  $min: breakpoint-min($name, $breakpoints);\n  @if $min {\n    @media (min-width: $min) {\n      @content;\n    }\n  } @else {\n    @content;\n  }\n}\n\n// Media of at most the maximum breakpoint width. No query for the largest breakpoint.\n// Makes the @content apply to the given breakpoint and narrower.\n@mixin media-breakpoint-down($name, $breakpoints: $grid-breakpoints) {\n  $max: breakpoint-max($name, $breakpoints);\n  @if $max {\n    @media (max-width: $max) {\n      @content;\n    }\n  } @else {\n    @content;\n  }\n}\n\n// Media that spans multiple breakpoint widths.\n// Makes the @content apply between the min and max breakpoints\n@mixin media-breakpoint-between($lower, $upper, $breakpoints: $grid-breakpoints) {\n  $min: breakpoint-min($lower, $breakpoints);\n  $max: breakpoint-max($upper, $breakpoints);\n\n  @if $min != null and $max != null {\n    @media (min-width: $min) and (max-width: $max) {\n      @content;\n    }\n  } @else if $max == null {\n    @include media-breakpoint-up($lower, $breakpoints) {\n      @content;\n    }\n  } @else if $min == null {\n    @include media-breakpoint-down($upper, $breakpoints) {\n      @content;\n    }\n  }\n}\n\n// Media between the breakpoint's minimum and maximum widths.\n// No minimum for the smallest breakpoint, and no maximum for the largest one.\n// Makes the @content apply only to the given breakpoint, not viewports any wider or narrower.\n@mixin media-breakpoint-only($name, $breakpoints: $grid-breakpoints) {\n  $min:  breakpoint-min($name, $breakpoints);\n  $next: breakpoint-next($name, $breakpoints);\n  $max:  breakpoint-max($next, $breakpoints);\n\n  @if $min != null and $max != null {\n    @media (min-width: $min) and (max-width: $max) {\n      @content;\n    }\n  } @else if $max == null {\n    @include media-breakpoint-up($name, $breakpoints) {\n      @content;\n    }\n  } @else if $min == null {\n    @include media-breakpoint-down($next, $breakpoints) {\n      @content;\n    }\n  }\n}\n","// Row\n//\n// Rows contain your columns.\n\n:root {\n  @each $name, $value in $grid-breakpoints {\n    --#{$prefix}breakpoint-#{$name}: #{$value};\n  }\n}\n\n@if $enable-grid-classes {\n  .row {\n    @include make-row();\n\n    > * {\n      @include make-col-ready();\n    }\n  }\n}\n\n@if $enable-cssgrid {\n  .grid {\n    display: grid;\n    grid-template-rows: repeat(var(--#{$prefix}rows, 1), 1fr);\n    grid-template-columns: repeat(var(--#{$prefix}columns, #{$grid-columns}), 1fr);\n    gap: var(--#{$prefix}gap, #{$grid-gutter-width});\n\n    @include make-cssgrid();\n  }\n}\n\n\n// Columns\n//\n// Common styles for small and large grid columns\n\n@if $enable-grid-classes {\n  @include make-grid-columns();\n}\n","// Grid system\n//\n// Generate semantic grid columns with these mixins.\n\n@mixin make-row($gutter: $grid-gutter-width) {\n  --#{$prefix}gutter-x: #{$gutter};\n  --#{$prefix}gutter-y: 0;\n  display: flex;\n  flex-wrap: wrap;\n  // TODO: Revisit calc order after https://github.com/react-bootstrap/react-bootstrap/issues/6039 is fixed\n  margin-top: calc(-1 * var(--#{$prefix}gutter-y)); // stylelint-disable-line function-disallowed-list\n  margin-right: calc(-.5 * var(--#{$prefix}gutter-x)); // stylelint-disable-line function-disallowed-list\n  margin-left: calc(-.5 * var(--#{$prefix}gutter-x)); // stylelint-disable-line function-disallowed-list\n}\n\n@mixin make-col-ready() {\n  // Add box sizing if only the grid is loaded\n  box-sizing: if(variable-exists(include-column-box-sizing) and $include-column-box-sizing, border-box, null);\n  // Prevent columns from becoming too narrow when at smaller grid tiers by\n  // always setting `width: 100%;`. This works because we set the width\n  // later on to override this initial width.\n  flex-shrink: 0;\n  width: 100%;\n  max-width: 100%; // Prevent `.col-auto`, `.col` (& responsive variants) from breaking out the grid\n  padding-right: calc(var(--#{$prefix}gutter-x) * .5); // stylelint-disable-line function-disallowed-list\n  padding-left: calc(var(--#{$prefix}gutter-x) * .5); // stylelint-disable-line function-disallowed-list\n  margin-top: var(--#{$prefix}gutter-y);\n}\n\n@mixin make-col($size: false, $columns: $grid-columns) {\n  @if $size {\n    flex: 0 0 auto;\n    width: percentage(divide($size, $columns));\n\n  } @else {\n    flex: 1 1 0;\n    max-width: 100%;\n  }\n}\n\n@mixin make-col-auto() {\n  flex: 0 0 auto;\n  width: auto;\n}\n\n@mixin make-col-offset($size, $columns: $grid-columns) {\n  $num: divide($size, $columns);\n  margin-left: if($num == 0, 0, percentage($num));\n}\n\n// Row columns\n//\n// Specify on a parent element(e.g., .row) to force immediate children into NN\n// number of columns. Supports wrapping to new lines, but does not do a Masonry\n// style grid.\n@mixin row-cols($count) {\n  > * {\n    flex: 0 0 auto;\n    width: percentage(divide(1, $count));\n  }\n}\n\n// Framework grid generation\n//\n// Used only by Bootstrap to generate the correct number of grid classes given\n// any value of `$grid-columns`.\n\n@mixin make-grid-columns($columns: $grid-columns, $gutter: $grid-gutter-width, $breakpoints: $grid-breakpoints) {\n  @each $breakpoint in map-keys($breakpoints) {\n    $infix: breakpoint-infix($breakpoint, $breakpoints);\n\n    @include media-breakpoint-up($breakpoint, $breakpoints) {\n      // Provide basic `.col-{bp}` classes for equal-width flexbox columns\n      .col#{$infix} {\n        flex: 1 0 0%; // Flexbugs #4: https://github.com/philipwalton/flexbugs#flexbug-4\n      }\n\n      .row-cols#{$infix}-auto > * {\n        @include make-col-auto();\n      }\n\n      @if $grid-row-columns > 0 {\n        @for $i from 1 through $grid-row-columns {\n          .row-cols#{$infix}-#{$i} {\n            @include row-cols($i);\n          }\n        }\n      }\n\n      .col#{$infix}-auto {\n        @include make-col-auto();\n      }\n\n      @if $columns > 0 {\n        @for $i from 1 through $columns {\n          .col#{$infix}-#{$i} {\n            @include make-col($i, $columns);\n          }\n        }\n\n        // `$columns - 1` because offsetting by the width of an entire row isn't possible\n        @for $i from 0 through ($columns - 1) {\n          @if not ($infix == \"\" and $i == 0) { // Avoid emitting useless .offset-0\n            .offset#{$infix}-#{$i} {\n              @include make-col-offset($i, $columns);\n            }\n          }\n        }\n      }\n\n      // Gutters\n      //\n      // Make use of `.g-*`, `.gx-*` or `.gy-*` utilities to change spacing between the columns.\n      @each $key, $value in $gutters {\n        .g#{$infix}-#{$key},\n        .gx#{$infix}-#{$key} {\n          --#{$prefix}gutter-x: #{$value};\n        }\n\n        .g#{$infix}-#{$key},\n        .gy#{$infix}-#{$key} {\n          --#{$prefix}gutter-y: #{$value};\n        }\n      }\n    }\n  }\n}\n\n@mixin make-cssgrid($columns: $grid-columns, $breakpoints: $grid-breakpoints) {\n  @each $breakpoint in map-keys($breakpoints) {\n    $infix: breakpoint-infix($breakpoint, $breakpoints);\n\n    @include media-breakpoint-up($breakpoint, $breakpoints) {\n      @if $columns > 0 {\n        @for $i from 1 through $columns {\n          .g-col#{$infix}-#{$i} {\n            grid-column: auto / span $i;\n          }\n        }\n\n        // Start with `1` because `0` is an invalid value.\n        // Ends with `$columns - 1` because offsetting by the width of an entire row isn't possible.\n        @for $i from 1 through ($columns - 1) {\n          .g-start#{$infix}-#{$i} {\n            grid-column-start: $i;\n          }\n        }\n      }\n    }\n  }\n}\n","//\n// Basic Bootstrap table\n//\n\n.table {\n  // Reset needed for nesting tables\n  --#{$prefix}table-color-type: initial;\n  --#{$prefix}table-bg-type: initial;\n  --#{$prefix}table-color-state: initial;\n  --#{$prefix}table-bg-state: initial;\n  // End of reset\n  --#{$prefix}table-color: #{$table-color};\n  --#{$prefix}table-bg: #{$table-bg};\n  --#{$prefix}table-border-color: #{$table-border-color};\n  --#{$prefix}table-accent-bg: #{$table-accent-bg};\n  --#{$prefix}table-striped-color: #{$table-striped-color};\n  --#{$prefix}table-striped-bg: #{$table-striped-bg};\n  --#{$prefix}table-active-color: #{$table-active-color};\n  --#{$prefix}table-active-bg: #{$table-active-bg};\n  --#{$prefix}table-hover-color: #{$table-hover-color};\n  --#{$prefix}table-hover-bg: #{$table-hover-bg};\n\n  width: 100%;\n  margin-bottom: $spacer;\n  vertical-align: $table-cell-vertical-align;\n  border-color: var(--#{$prefix}table-border-color);\n\n  // Target th & td\n  // We need the child combinator to prevent styles leaking to nested tables which doesn't have a `.table` class.\n  // We use the universal selectors here to simplify the selector (else we would need 6 different selectors).\n  // Another advantage is that this generates less code and makes the selector less specific making it easier to override.\n  // stylelint-disable-next-line selector-max-universal\n  > :not(caption) > * > * {\n    padding: $table-cell-padding-y $table-cell-padding-x;\n    // Following the precept of cascades: https://codepen.io/miriamsuzanne/full/vYNgodb\n    color: var(--#{$prefix}table-color-state, var(--#{$prefix}table-color-type, var(--#{$prefix}table-color)));\n    background-color: var(--#{$prefix}table-bg);\n    border-bottom-width: $table-border-width;\n    box-shadow: inset 0 0 0 9999px var(--#{$prefix}table-bg-state, var(--#{$prefix}table-bg-type, var(--#{$prefix}table-accent-bg)));\n  }\n\n  > tbody {\n    vertical-align: inherit;\n  }\n\n  > thead {\n    vertical-align: bottom;\n  }\n}\n\n.table-group-divider {\n  border-top: calc(#{$table-border-width} * 2) solid $table-group-separator-color; // stylelint-disable-line function-disallowed-list\n}\n\n//\n// Change placement of captions with a class\n//\n\n.caption-top {\n  caption-side: top;\n}\n\n\n//\n// Condensed table w/ half padding\n//\n\n.table-sm {\n  // stylelint-disable-next-line selector-max-universal\n  > :not(caption) > * > * {\n    padding: $table-cell-padding-y-sm $table-cell-padding-x-sm;\n  }\n}\n\n\n// Border versions\n//\n// Add or remove borders all around the table and between all the columns.\n//\n// When borders are added on all sides of the cells, the corners can render odd when\n// these borders do not have the same color or if they are semi-transparent.\n// Therefore we add top and border bottoms to the `tr`s and left and right borders\n// to the `td`s or `th`s\n\n.table-bordered {\n  > :not(caption) > * {\n    border-width: $table-border-width 0;\n\n    // stylelint-disable-next-line selector-max-universal\n    > * {\n      border-width: 0 $table-border-width;\n    }\n  }\n}\n\n.table-borderless {\n  // stylelint-disable-next-line selector-max-universal\n  > :not(caption) > * > * {\n    border-bottom-width: 0;\n  }\n\n  > :not(:first-child) {\n    border-top-width: 0;\n  }\n}\n\n// Zebra-striping\n//\n// Default zebra-stripe styles (alternating gray and transparent backgrounds)\n\n// For rows\n.table-striped {\n  > tbody > tr:nth-of-type(#{$table-striped-order}) > * {\n    --#{$prefix}table-color-type: var(--#{$prefix}table-striped-color);\n    --#{$prefix}table-bg-type: var(--#{$prefix}table-striped-bg);\n  }\n}\n\n// For columns\n.table-striped-columns {\n  > :not(caption) > tr > :nth-child(#{$table-striped-columns-order}) {\n    --#{$prefix}table-color-type: var(--#{$prefix}table-striped-color);\n    --#{$prefix}table-bg-type: var(--#{$prefix}table-striped-bg);\n  }\n}\n\n// Active table\n//\n// The `.table-active` class can be added to highlight rows or cells\n\n.table-active {\n  --#{$prefix}table-color-state: var(--#{$prefix}table-active-color);\n  --#{$prefix}table-bg-state: var(--#{$prefix}table-active-bg);\n}\n\n// Hover effect\n//\n// Placed here since it has to come after the potential zebra striping\n\n.table-hover {\n  > tbody > tr:hover > * {\n    --#{$prefix}table-color-state: var(--#{$prefix}table-hover-color);\n    --#{$prefix}table-bg-state: var(--#{$prefix}table-hover-bg);\n  }\n}\n\n\n// Table variants\n//\n// Table variants set the table cell backgrounds, border colors\n// and the colors of the striped, hovered & active tables\n\n@each $color, $value in $table-variants {\n  @include table-variant($color, $value);\n}\n\n// Responsive tables\n//\n// Generate series of `.table-responsive-*` classes for configuring the screen\n// size of where your table will overflow.\n\n@each $breakpoint in map-keys($grid-breakpoints) {\n  $infix: breakpoint-infix($breakpoint, $grid-breakpoints);\n\n  @include media-breakpoint-down($breakpoint) {\n    .table-responsive#{$infix} {\n      overflow-x: auto;\n      -webkit-overflow-scrolling: touch;\n    }\n  }\n}\n","// scss-docs-start table-variant\n@mixin table-variant($state, $background) {\n  .table-#{$state} {\n    $color: color-contrast(opaque($body-bg, $background));\n    $hover-bg: mix($color, $background, percentage($table-hover-bg-factor));\n    $striped-bg: mix($color, $background, percentage($table-striped-bg-factor));\n    $active-bg: mix($color, $background, percentage($table-active-bg-factor));\n    $table-border-color: mix($color, $background, percentage($table-border-factor));\n\n    --#{$prefix}table-color: #{$color};\n    --#{$prefix}table-bg: #{$background};\n    --#{$prefix}table-border-color: #{$table-border-color};\n    --#{$prefix}table-striped-bg: #{$striped-bg};\n    --#{$prefix}table-striped-color: #{color-contrast($striped-bg)};\n    --#{$prefix}table-active-bg: #{$active-bg};\n    --#{$prefix}table-active-color: #{color-contrast($active-bg)};\n    --#{$prefix}table-hover-bg: #{$hover-bg};\n    --#{$prefix}table-hover-color: #{color-contrast($hover-bg)};\n\n    color: var(--#{$prefix}table-color);\n    border-color: var(--#{$prefix}table-border-color);\n  }\n}\n// scss-docs-end table-variant\n","//\n// Labels\n//\n\n.form-label {\n  margin-bottom: $form-label-margin-bottom;\n  @include font-size($form-label-font-size);\n  font-style: $form-label-font-style;\n  font-weight: $form-label-font-weight;\n  color: $form-label-color;\n}\n\n// For use with horizontal and inline forms, when you need the label (or legend)\n// text to align with the form controls.\n.col-form-label {\n  padding-top: add($input-padding-y, $input-border-width);\n  padding-bottom: add($input-padding-y, $input-border-width);\n  margin-bottom: 0; // Override the `<legend>` default\n  @include font-size(inherit); // Override the `<legend>` default\n  font-style: $form-label-font-style;\n  font-weight: $form-label-font-weight;\n  line-height: $input-line-height;\n  color: $form-label-color;\n}\n\n.col-form-label-lg {\n  padding-top: add($input-padding-y-lg, $input-border-width);\n  padding-bottom: add($input-padding-y-lg, $input-border-width);\n  @include font-size($input-font-size-lg);\n}\n\n.col-form-label-sm {\n  padding-top: add($input-padding-y-sm, $input-border-width);\n  padding-bottom: add($input-padding-y-sm, $input-border-width);\n  @include font-size($input-font-size-sm);\n}\n","//\n// Form text\n//\n\n.form-text {\n  margin-top: $form-text-margin-top;\n  @include font-size($form-text-font-size);\n  font-style: $form-text-font-style;\n  font-weight: $form-text-font-weight;\n  color: $form-text-color;\n}\n","//\n// General form controls (plus a few specific high-level interventions)\n//\n\n.form-control {\n  display: block;\n  width: 100%;\n  padding: $input-padding-y $input-padding-x;\n  font-family: $input-font-family;\n  @include font-size($input-font-size);\n  font-weight: $input-font-weight;\n  line-height: $input-line-height;\n  color: $input-color;\n  appearance: none; // Fix appearance for date inputs in Safari\n  background-color: $input-bg;\n  background-clip: padding-box;\n  border: $input-border-width solid $input-border-color;\n\n  // Note: This has no effect on <select>s in some browsers, due to the limited stylability of `<select>`s in CSS.\n  @include border-radius($input-border-radius, 0);\n\n  @include box-shadow($input-box-shadow);\n  @include transition($input-transition);\n\n  &[type=\"file\"] {\n    overflow: hidden; // prevent pseudo element button overlap\n\n    &:not(:disabled):not([readonly]) {\n      cursor: pointer;\n    }\n  }\n\n  // Customize the `:focus` state to imitate native WebKit styles.\n  &:focus {\n    color: $input-focus-color;\n    background-color: $input-focus-bg;\n    border-color: $input-focus-border-color;\n    outline: 0;\n    @if $enable-shadows {\n      @include box-shadow($input-box-shadow, $input-focus-box-shadow);\n    } @else {\n      // Avoid using mixin so we can pass custom focus shadow properly\n      box-shadow: $input-focus-box-shadow;\n    }\n  }\n\n  &::-webkit-date-and-time-value {\n    // On Android Chrome, form-control's \"width: 100%\" makes the input width too small\n    // Tested under Android 11 / Chrome 89, Android 12 / Chrome 100, Android 13 / Chrome 109\n    //\n    // On iOS Safari, form-control's \"appearance: none\" + \"width: 100%\" makes the input width too small\n    // Tested under iOS 16.2 / Safari 16.2\n    min-width: 85px; // Seems to be a good minimum safe width\n\n    // Add some height to date inputs on iOS\n    // https://github.com/twbs/bootstrap/issues/23307\n    // TODO: we can remove this workaround once https://bugs.webkit.org/show_bug.cgi?id=198959 is resolved\n    // Multiply line-height by 1em if it has no unit\n    height: if(unit($input-line-height) == \"\", $input-line-height * 1em, $input-line-height);\n\n    // Android Chrome type=\"date\" is taller than the other inputs\n    // because of \"margin: 1px 24px 1px 4px\" inside the shadow DOM\n    // Tested under Android 11 / Chrome 89, Android 12 / Chrome 100, Android 13 / Chrome 109\n    margin: 0;\n  }\n\n  // Prevent excessive date input height in Webkit\n  // https://github.com/twbs/bootstrap/issues/34433\n  &::-webkit-datetime-edit {\n    display: block;\n    padding: 0;\n  }\n\n  // Placeholder\n  &::placeholder {\n    color: $input-placeholder-color;\n    // Override Firefox's unusual default opacity; see https://github.com/twbs/bootstrap/pull/11526.\n    opacity: 1;\n  }\n\n  // Disabled inputs\n  //\n  // HTML5 says that controls under a fieldset > legend:first-child won't be\n  // disabled if the fieldset is disabled. Due to implementation difficulty, we\n  // don't honor that edge case; we style them as disabled anyway.\n  &:disabled {\n    color: $input-disabled-color;\n    background-color: $input-disabled-bg;\n    border-color: $input-disabled-border-color;\n    // iOS fix for unreadable disabled content; see https://github.com/twbs/bootstrap/issues/11655.\n    opacity: 1;\n  }\n\n  // File input buttons theming\n  &::file-selector-button {\n    padding: $input-padding-y $input-padding-x;\n    margin: (-$input-padding-y) (-$input-padding-x);\n    margin-inline-end: $input-padding-x;\n    color: $form-file-button-color;\n    @include gradient-bg($form-file-button-bg);\n    pointer-events: none;\n    border-color: inherit;\n    border-style: solid;\n    border-width: 0;\n    border-inline-end-width: $input-border-width;\n    border-radius: 0; // stylelint-disable-line property-disallowed-list\n    @include transition($btn-transition);\n  }\n\n  &:hover:not(:disabled):not([readonly])::file-selector-button {\n    background-color: $form-file-button-hover-bg;\n  }\n}\n\n// Readonly controls as plain text\n//\n// Apply class to a readonly input to make it appear like regular plain\n// text (without any border, background color, focus indicator)\n\n.form-control-plaintext {\n  display: block;\n  width: 100%;\n  padding: $input-padding-y 0;\n  margin-bottom: 0; // match inputs if this class comes on inputs with default margins\n  line-height: $input-line-height;\n  color: $input-plaintext-color;\n  background-color: transparent;\n  border: solid transparent;\n  border-width: $input-border-width 0;\n\n  &:focus {\n    outline: 0;\n  }\n\n  &.form-control-sm,\n  &.form-control-lg {\n    padding-right: 0;\n    padding-left: 0;\n  }\n}\n\n// Form control sizing\n//\n// Build on `.form-control` with modifier classes to decrease or increase the\n// height and font-size of form controls.\n//\n// Repeated in `_input_group.scss` to avoid Sass extend issues.\n\n.form-control-sm {\n  min-height: $input-height-sm;\n  padding: $input-padding-y-sm $input-padding-x-sm;\n  @include font-size($input-font-size-sm);\n  @include border-radius($input-border-radius-sm);\n\n  &::file-selector-button {\n    padding: $input-padding-y-sm $input-padding-x-sm;\n    margin: (-$input-padding-y-sm) (-$input-padding-x-sm);\n    margin-inline-end: $input-padding-x-sm;\n  }\n}\n\n.form-control-lg {\n  min-height: $input-height-lg;\n  padding: $input-padding-y-lg $input-padding-x-lg;\n  @include font-size($input-font-size-lg);\n  @include border-radius($input-border-radius-lg);\n\n  &::file-selector-button {\n    padding: $input-padding-y-lg $input-padding-x-lg;\n    margin: (-$input-padding-y-lg) (-$input-padding-x-lg);\n    margin-inline-end: $input-padding-x-lg;\n  }\n}\n\n// Make sure textareas don't shrink too much when resized\n// https://github.com/twbs/bootstrap/pull/29124\n// stylelint-disable selector-no-qualifying-type\ntextarea {\n  &.form-control {\n    min-height: $input-height;\n  }\n\n  &.form-control-sm {\n    min-height: $input-height-sm;\n  }\n\n  &.form-control-lg {\n    min-height: $input-height-lg;\n  }\n}\n// stylelint-enable selector-no-qualifying-type\n\n.form-control-color {\n  width: $form-color-width;\n  height: $input-height;\n  padding: $input-padding-y;\n\n  &:not(:disabled):not([readonly]) {\n    cursor: pointer;\n  }\n\n  &::-moz-color-swatch {\n    border: 0 !important; // stylelint-disable-line declaration-no-important\n    @include border-radius($input-border-radius);\n  }\n\n  &::-webkit-color-swatch {\n    border: 0 !important; // stylelint-disable-line declaration-no-important\n    @include border-radius($input-border-radius);\n  }\n\n  &.form-control-sm { height: $input-height-sm; }\n  &.form-control-lg { height: $input-height-lg; }\n}\n","// stylelint-disable property-disallowed-list\n@mixin transition($transition...) {\n  @if length($transition) == 0 {\n    $transition: $transition-base;\n  }\n\n  @if length($transition) > 1 {\n    @each $value in $transition {\n      @if $value == null or $value == none {\n        @warn \"The keyword 'none' or 'null' must be used as a single argument.\";\n      }\n    }\n  }\n\n  @if $enable-transitions {\n    @if nth($transition, 1) != null {\n      transition: $transition;\n    }\n\n    @if $enable-reduced-motion and nth($transition, 1) != null and nth($transition, 1) != none {\n      @media (prefers-reduced-motion: reduce) {\n        transition: none;\n      }\n    }\n  }\n}\n","// Gradients\n\n// scss-docs-start gradient-bg-mixin\n@mixin gradient-bg($color: null) {\n  background-color: $color;\n\n  @if $enable-gradients {\n    background-image: var(--#{$prefix}gradient);\n  }\n}\n// scss-docs-end gradient-bg-mixin\n\n// scss-docs-start gradient-mixins\n// Horizontal gradient, from left to right\n//\n// Creates two color stops, start and end, by specifying a color and position for each color stop.\n@mixin gradient-x($start-color: $gray-700, $end-color: $gray-800, $start-percent: 0%, $end-percent: 100%) {\n  background-image: linear-gradient(to right, $start-color $start-percent, $end-color $end-percent);\n}\n\n// Vertical gradient, from top to bottom\n//\n// Creates two color stops, start and end, by specifying a color and position for each color stop.\n@mixin gradient-y($start-color: $gray-700, $end-color: $gray-800, $start-percent: null, $end-percent: null) {\n  background-image: linear-gradient(to bottom, $start-color $start-percent, $end-color $end-percent);\n}\n\n@mixin gradient-directional($start-color: $gray-700, $end-color: $gray-800, $deg: 45deg) {\n  background-image: linear-gradient($deg, $start-color, $end-color);\n}\n\n@mixin gradient-x-three-colors($start-color: $blue, $mid-color: $purple, $color-stop: 50%, $end-color: $red) {\n  background-image: linear-gradient(to right, $start-color, $mid-color $color-stop, $end-color);\n}\n\n@mixin gradient-y-three-colors($start-color: $blue, $mid-color: $purple, $color-stop: 50%, $end-color: $red) {\n  background-image: linear-gradient($start-color, $mid-color $color-stop, $end-color);\n}\n\n@mixin gradient-radial($inner-color: $gray-700, $outer-color: $gray-800) {\n  background-image: radial-gradient(circle, $inner-color, $outer-color);\n}\n\n@mixin gradient-striped($color: rgba($white, .15), $angle: 45deg) {\n  background-image: linear-gradient($angle, $color 25%, transparent 25%, transparent 50%, $color 50%, $color 75%, transparent 75%, transparent);\n}\n// scss-docs-end gradient-mixins\n","// Select\n//\n// Replaces the browser default select with a custom one, mostly pulled from\n// https://primer.github.io/.\n\n.form-select {\n  --#{$prefix}form-select-bg-img: #{escape-svg($form-select-indicator)};\n\n  display: block;\n  width: 100%;\n  padding: $form-select-padding-y $form-select-indicator-padding $form-select-padding-y $form-select-padding-x;\n  font-family: $form-select-font-family;\n  @include font-size($form-select-font-size);\n  font-weight: $form-select-font-weight;\n  line-height: $form-select-line-height;\n  color: $form-select-color;\n  appearance: none;\n  background-color: $form-select-bg;\n  background-image: var(--#{$prefix}form-select-bg-img), var(--#{$prefix}form-select-bg-icon, none);\n  background-repeat: no-repeat;\n  background-position: $form-select-bg-position;\n  background-size: $form-select-bg-size;\n  border: $form-select-border-width solid $form-select-border-color;\n  @include border-radius($form-select-border-radius, 0);\n  @include box-shadow($form-select-box-shadow);\n  @include transition($form-select-transition);\n\n  &:focus {\n    border-color: $form-select-focus-border-color;\n    outline: 0;\n    @if $enable-shadows {\n      @include box-shadow($form-select-box-shadow, $form-select-focus-box-shadow);\n    } @else {\n      // Avoid using mixin so we can pass custom focus shadow properly\n      box-shadow: $form-select-focus-box-shadow;\n    }\n  }\n\n  &[multiple],\n  &[size]:not([size=\"1\"]) {\n    padding-right: $form-select-padding-x;\n    background-image: none;\n  }\n\n  &:disabled {\n    color: $form-select-disabled-color;\n    background-color: $form-select-disabled-bg;\n    border-color: $form-select-disabled-border-color;\n  }\n\n  // Remove outline from select box in FF\n  &:-moz-focusring {\n    color: transparent;\n    text-shadow: 0 0 0 $form-select-color;\n  }\n}\n\n.form-select-sm {\n  padding-top: $form-select-padding-y-sm;\n  padding-bottom: $form-select-padding-y-sm;\n  padding-left: $form-select-padding-x-sm;\n  @include font-size($form-select-font-size-sm);\n  @include border-radius($form-select-border-radius-sm);\n}\n\n.form-select-lg {\n  padding-top: $form-select-padding-y-lg;\n  padding-bottom: $form-select-padding-y-lg;\n  padding-left: $form-select-padding-x-lg;\n  @include font-size($form-select-font-size-lg);\n  @include border-radius($form-select-border-radius-lg);\n}\n\n@if $enable-dark-mode {\n  @include color-mode(dark) {\n    .form-select {\n      --#{$prefix}form-select-bg-img: #{escape-svg($form-select-indicator-dark)};\n    }\n  }\n}\n","//\n// Check/radio\n//\n\n.form-check {\n  display: block;\n  min-height: $form-check-min-height;\n  padding-left: $form-check-padding-start;\n  margin-bottom: $form-check-margin-bottom;\n\n  .form-check-input {\n    float: left;\n    margin-left: $form-check-padding-start * -1;\n  }\n}\n\n.form-check-reverse {\n  padding-right: $form-check-padding-start;\n  padding-left: 0;\n  text-align: right;\n\n  .form-check-input {\n    float: right;\n    margin-right: $form-check-padding-start * -1;\n    margin-left: 0;\n  }\n}\n\n.form-check-input {\n  --#{$prefix}form-check-bg: #{$form-check-input-bg};\n\n  flex-shrink: 0;\n  width: $form-check-input-width;\n  height: $form-check-input-width;\n  margin-top: ($line-height-base - $form-check-input-width) * .5; // line-height minus check height\n  vertical-align: top;\n  appearance: none;\n  background-color: var(--#{$prefix}form-check-bg);\n  background-image: var(--#{$prefix}form-check-bg-image);\n  background-repeat: no-repeat;\n  background-position: center;\n  background-size: contain;\n  border: $form-check-input-border;\n  print-color-adjust: exact; // Keep themed appearance for print\n  @include transition($form-check-transition);\n\n  &[type=\"checkbox\"] {\n    @include border-radius($form-check-input-border-radius);\n  }\n\n  &[type=\"radio\"] {\n    // stylelint-disable-next-line property-disallowed-list\n    border-radius: $form-check-radio-border-radius;\n  }\n\n  &:active {\n    filter: $form-check-input-active-filter;\n  }\n\n  &:focus {\n    border-color: $form-check-input-focus-border;\n    outline: 0;\n    box-shadow: $form-check-input-focus-box-shadow;\n  }\n\n  &:checked {\n    background-color: $form-check-input-checked-bg-color;\n    border-color: $form-check-input-checked-border-color;\n\n    &[type=\"checkbox\"] {\n      @if $enable-gradients {\n        --#{$prefix}form-check-bg-image: #{escape-svg($form-check-input-checked-bg-image)}, var(--#{$prefix}gradient);\n      } @else {\n        --#{$prefix}form-check-bg-image: #{escape-svg($form-check-input-checked-bg-image)};\n      }\n    }\n\n    &[type=\"radio\"] {\n      @if $enable-gradients {\n        --#{$prefix}form-check-bg-image: #{escape-svg($form-check-radio-checked-bg-image)}, var(--#{$prefix}gradient);\n      } @else {\n        --#{$prefix}form-check-bg-image: #{escape-svg($form-check-radio-checked-bg-image)};\n      }\n    }\n  }\n\n  &[type=\"checkbox\"]:indeterminate {\n    background-color: $form-check-input-indeterminate-bg-color;\n    border-color: $form-check-input-indeterminate-border-color;\n\n    @if $enable-gradients {\n      --#{$prefix}form-check-bg-image: #{escape-svg($form-check-input-indeterminate-bg-image)}, var(--#{$prefix}gradient);\n    } @else {\n      --#{$prefix}form-check-bg-image: #{escape-svg($form-check-input-indeterminate-bg-image)};\n    }\n  }\n\n  &:disabled {\n    pointer-events: none;\n    filter: none;\n    opacity: $form-check-input-disabled-opacity;\n  }\n\n  // Use disabled attribute in addition of :disabled pseudo-class\n  // See: https://github.com/twbs/bootstrap/issues/28247\n  &[disabled],\n  &:disabled {\n    ~ .form-check-label {\n      cursor: default;\n      opacity: $form-check-label-disabled-opacity;\n    }\n  }\n}\n\n.form-check-label {\n  color: $form-check-label-color;\n  cursor: $form-check-label-cursor;\n}\n\n//\n// Switch\n//\n\n.form-switch {\n  padding-left: $form-switch-padding-start;\n\n  .form-check-input {\n    --#{$prefix}form-switch-bg: #{escape-svg($form-switch-bg-image)};\n\n    width: $form-switch-width;\n    margin-left: $form-switch-padding-start * -1;\n    background-image: var(--#{$prefix}form-switch-bg);\n    background-position: left center;\n    @include border-radius($form-switch-border-radius, 0);\n    @include transition($form-switch-transition);\n\n    &:focus {\n      --#{$prefix}form-switch-bg: #{escape-svg($form-switch-focus-bg-image)};\n    }\n\n    &:checked {\n      background-position: $form-switch-checked-bg-position;\n\n      @if $enable-gradients {\n        --#{$prefix}form-switch-bg: #{escape-svg($form-switch-checked-bg-image)}, var(--#{$prefix}gradient);\n      } @else {\n        --#{$prefix}form-switch-bg: #{escape-svg($form-switch-checked-bg-image)};\n      }\n    }\n  }\n\n  &.form-check-reverse {\n    padding-right: $form-switch-padding-start;\n    padding-left: 0;\n\n    .form-check-input {\n      margin-right: $form-switch-padding-start * -1;\n      margin-left: 0;\n    }\n  }\n}\n\n.form-check-inline {\n  display: inline-block;\n  margin-right: $form-check-inline-margin-end;\n}\n\n.btn-check {\n  position: absolute;\n  clip: rect(0, 0, 0, 0);\n  pointer-events: none;\n\n  &[disabled],\n  &:disabled {\n    + .btn {\n      pointer-events: none;\n      filter: none;\n      opacity: $form-check-btn-check-disabled-opacity;\n    }\n  }\n}\n\n@if $enable-dark-mode {\n  @include color-mode(dark) {\n    .form-switch .form-check-input:not(:checked):not(:focus) {\n      --#{$prefix}form-switch-bg: #{escape-svg($form-switch-bg-image-dark)};\n    }\n  }\n}\n","// Range\n//\n// Style range inputs the same across browsers. Vendor-specific rules for pseudo\n// elements cannot be mixed. As such, there are no shared styles for focus or\n// active states on prefixed selectors.\n\n.form-range {\n  width: 100%;\n  height: add($form-range-thumb-height, $form-range-thumb-focus-box-shadow-width * 2);\n  padding: 0; // Need to reset padding\n  appearance: none;\n  background-color: transparent;\n\n  &:focus {\n    outline: 0;\n\n    // Pseudo-elements must be split across multiple rulesets to have an effect.\n    // No box-shadow() mixin for focus accessibility.\n    &::-webkit-slider-thumb { box-shadow: $form-range-thumb-focus-box-shadow; }\n    &::-moz-range-thumb     { box-shadow: $form-range-thumb-focus-box-shadow; }\n  }\n\n  &::-moz-focus-outer {\n    border: 0;\n  }\n\n  &::-webkit-slider-thumb {\n    width: $form-range-thumb-width;\n    height: $form-range-thumb-height;\n    margin-top: ($form-range-track-height - $form-range-thumb-height) * .5; // Webkit specific\n    appearance: none;\n    @include gradient-bg($form-range-thumb-bg);\n    border: $form-range-thumb-border;\n    @include border-radius($form-range-thumb-border-radius);\n    @include box-shadow($form-range-thumb-box-shadow);\n    @include transition($form-range-thumb-transition);\n\n    &:active {\n      @include gradient-bg($form-range-thumb-active-bg);\n    }\n  }\n\n  &::-webkit-slider-runnable-track {\n    width: $form-range-track-width;\n    height: $form-range-track-height;\n    color: transparent; // Why?\n    cursor: $form-range-track-cursor;\n    background-color: $form-range-track-bg;\n    border-color: transparent;\n    @include border-radius($form-range-track-border-radius);\n    @include box-shadow($form-range-track-box-shadow);\n  }\n\n  &::-moz-range-thumb {\n    width: $form-range-thumb-width;\n    height: $form-range-thumb-height;\n    appearance: none;\n    @include gradient-bg($form-range-thumb-bg);\n    border: $form-range-thumb-border;\n    @include border-radius($form-range-thumb-border-radius);\n    @include box-shadow($form-range-thumb-box-shadow);\n    @include transition($form-range-thumb-transition);\n\n    &:active {\n      @include gradient-bg($form-range-thumb-active-bg);\n    }\n  }\n\n  &::-moz-range-track {\n    width: $form-range-track-width;\n    height: $form-range-track-height;\n    color: transparent;\n    cursor: $form-range-track-cursor;\n    background-color: $form-range-track-bg;\n    border-color: transparent; // Firefox specific?\n    @include border-radius($form-range-track-border-radius);\n    @include box-shadow($form-range-track-box-shadow);\n  }\n\n  &:disabled {\n    pointer-events: none;\n\n    &::-webkit-slider-thumb {\n      background-color: $form-range-thumb-disabled-bg;\n    }\n\n    &::-moz-range-thumb {\n      background-color: $form-range-thumb-disabled-bg;\n    }\n  }\n}\n",".form-floating {\n  position: relative;\n\n  > .form-control,\n  > .form-control-plaintext,\n  > .form-select {\n    height: $form-floating-height;\n    min-height: $form-floating-height;\n    line-height: $form-floating-line-height;\n  }\n\n  > label {\n    position: absolute;\n    top: 0;\n    left: 0;\n    z-index: 2;\n    height: 100%; // allow textareas\n    padding: $form-floating-padding-y $form-floating-padding-x;\n    overflow: hidden;\n    text-align: start;\n    text-overflow: ellipsis;\n    white-space: nowrap;\n    pointer-events: none;\n    border: $input-border-width solid transparent; // Required for aligning label's text with the input as it affects inner box model\n    transform-origin: 0 0;\n    @include transition($form-floating-transition);\n  }\n\n  > .form-control,\n  > .form-control-plaintext {\n    padding: $form-floating-padding-y $form-floating-padding-x;\n\n    &::placeholder {\n      color: transparent;\n    }\n\n    &:focus,\n    &:not(:placeholder-shown) {\n      padding-top: $form-floating-input-padding-t;\n      padding-bottom: $form-floating-input-padding-b;\n    }\n    // Duplicated because `:-webkit-autofill` invalidates other selectors when grouped\n    &:-webkit-autofill {\n      padding-top: $form-floating-input-padding-t;\n      padding-bottom: $form-floating-input-padding-b;\n    }\n  }\n\n  > .form-select {\n    padding-top: $form-floating-input-padding-t;\n    padding-bottom: $form-floating-input-padding-b;\n  }\n\n  > .form-control:focus,\n  > .form-control:not(:placeholder-shown),\n  > .form-control-plaintext,\n  > .form-select {\n    ~ label {\n      color: rgba(var(--#{$prefix}body-color-rgb), #{$form-floating-label-opacity});\n      transform: $form-floating-label-transform;\n\n      &::after {\n        position: absolute;\n        inset: $form-floating-padding-y ($form-floating-padding-x * .5);\n        z-index: -1;\n        height: $form-floating-label-height;\n        content: \"\";\n        background-color: $input-bg;\n        @include border-radius($input-border-radius);\n      }\n    }\n  }\n  // Duplicated because `:-webkit-autofill` invalidates other selectors when grouped\n  > .form-control:-webkit-autofill {\n    ~ label {\n      color: rgba(var(--#{$prefix}body-color-rgb), #{$form-floating-label-opacity});\n      transform: $form-floating-label-transform;\n    }\n  }\n\n  > .form-control-plaintext {\n    ~ label {\n      border-width: $input-border-width 0; // Required to properly position label text - as explained above\n    }\n  }\n\n  > :disabled ~ label,\n  > .form-control:disabled ~ label { // Required for `.form-control`s because of specificity\n    color: $form-floating-label-disabled-color;\n\n    &::after {\n      background-color: $input-disabled-bg;\n    }\n  }\n}\n","//\n// Base styles\n//\n\n.input-group {\n  position: relative;\n  display: flex;\n  flex-wrap: wrap; // For form validation feedback\n  align-items: stretch;\n  width: 100%;\n\n  > .form-control,\n  > .form-select,\n  > .form-floating {\n    position: relative; // For focus state's z-index\n    flex: 1 1 auto;\n    width: 1%;\n    min-width: 0; // https://stackoverflow.com/questions/36247140/why-dont-flex-items-shrink-past-content-size\n  }\n\n  // Bring the \"active\" form control to the top of surrounding elements\n  > .form-control:focus,\n  > .form-select:focus,\n  > .form-floating:focus-within {\n    z-index: 5;\n  }\n\n  // Ensure buttons are always above inputs for more visually pleasing borders.\n  // This isn't needed for `.input-group-text` since it shares the same border-color\n  // as our inputs.\n  .btn {\n    position: relative;\n    z-index: 2;\n\n    &:focus {\n      z-index: 5;\n    }\n  }\n}\n\n\n// Textual addons\n//\n// Serves as a catch-all element for any text or radio/checkbox input you wish\n// to prepend or append to an input.\n\n.input-group-text {\n  display: flex;\n  align-items: center;\n  padding: $input-group-addon-padding-y $input-group-addon-padding-x;\n  @include font-size($input-font-size); // Match inputs\n  font-weight: $input-group-addon-font-weight;\n  line-height: $input-line-height;\n  color: $input-group-addon-color;\n  text-align: center;\n  white-space: nowrap;\n  background-color: $input-group-addon-bg;\n  border: $input-border-width solid $input-group-addon-border-color;\n  @include border-radius($input-border-radius);\n}\n\n\n// Sizing\n//\n// Remix the default form control sizing classes into new ones for easier\n// manipulation.\n\n.input-group-lg > .form-control,\n.input-group-lg > .form-select,\n.input-group-lg > .input-group-text,\n.input-group-lg > .btn {\n  padding: $input-padding-y-lg $input-padding-x-lg;\n  @include font-size($input-font-size-lg);\n  @include border-radius($input-border-radius-lg);\n}\n\n.input-group-sm > .form-control,\n.input-group-sm > .form-select,\n.input-group-sm > .input-group-text,\n.input-group-sm > .btn {\n  padding: $input-padding-y-sm $input-padding-x-sm;\n  @include font-size($input-font-size-sm);\n  @include border-radius($input-border-radius-sm);\n}\n\n.input-group-lg > .form-select,\n.input-group-sm > .form-select {\n  padding-right: $form-select-padding-x + $form-select-indicator-padding;\n}\n\n\n// Rounded corners\n//\n// These rulesets must come after the sizing ones to properly override sm and lg\n// border-radius values when extending. They're more specific than we'd like\n// with the `.input-group >` part, but without it, we cannot override the sizing.\n\n// stylelint-disable-next-line no-duplicate-selectors\n.input-group {\n  &:not(.has-validation) {\n    > :not(:last-child):not(.dropdown-toggle):not(.dropdown-menu):not(.form-floating),\n    > .dropdown-toggle:nth-last-child(n + 3),\n    > .form-floating:not(:last-child) > .form-control,\n    > .form-floating:not(:last-child) > .form-select {\n      @include border-end-radius(0);\n    }\n  }\n\n  &.has-validation {\n    > :nth-last-child(n + 3):not(.dropdown-toggle):not(.dropdown-menu):not(.form-floating),\n    > .dropdown-toggle:nth-last-child(n + 4),\n    > .form-floating:nth-last-child(n + 3) > .form-control,\n    > .form-floating:nth-last-child(n + 3) > .form-select {\n      @include border-end-radius(0);\n    }\n  }\n\n  $validation-messages: \"\";\n  @each $state in map-keys($form-validation-states) {\n    $validation-messages: $validation-messages + \":not(.\" + unquote($state) + \"-tooltip)\" + \":not(.\" + unquote($state) + \"-feedback)\";\n  }\n\n  > :not(:first-child):not(.dropdown-menu)#{$validation-messages} {\n    margin-left: calc(#{$input-border-width} * -1); // stylelint-disable-line function-disallowed-list\n    @include border-start-radius(0);\n  }\n\n  > .form-floating:not(:first-child) > .form-control,\n  > .form-floating:not(:first-child) > .form-select {\n    @include border-start-radius(0);\n  }\n}\n","// This mixin uses an `if()` technique to be compatible with Dart Sass\n// See https://github.com/sass/sass/issues/1873#issuecomment-152293725 for more details\n\n// scss-docs-start form-validation-mixins\n@mixin form-validation-state-selector($state) {\n  @if ($state == \"valid\" or $state == \"invalid\") {\n    .was-validated #{if(&, \"&\", \"\")}:#{$state},\n    #{if(&, \"&\", \"\")}.is-#{$state} {\n      @content;\n    }\n  } @else {\n    #{if(&, \"&\", \"\")}.is-#{$state} {\n      @content;\n    }\n  }\n}\n\n@mixin form-validation-state(\n  $state,\n  $color,\n  $icon,\n  $tooltip-color: color-contrast($color),\n  $tooltip-bg-color: rgba($color, $form-feedback-tooltip-opacity),\n  $focus-box-shadow: 0 0 $input-btn-focus-blur $input-focus-width rgba($color, $input-btn-focus-color-opacity),\n  $border-color: $color\n) {\n  .#{$state}-feedback {\n    display: none;\n    width: 100%;\n    margin-top: $form-feedback-margin-top;\n    @include font-size($form-feedback-font-size);\n    font-style: $form-feedback-font-style;\n    color: $color;\n  }\n\n  .#{$state}-tooltip {\n    position: absolute;\n    top: 100%;\n    z-index: 5;\n    display: none;\n    max-width: 100%; // Contain to parent when possible\n    padding: $form-feedback-tooltip-padding-y $form-feedback-tooltip-padding-x;\n    margin-top: .1rem;\n    @include font-size($form-feedback-tooltip-font-size);\n    line-height: $form-feedback-tooltip-line-height;\n    color: $tooltip-color;\n    background-color: $tooltip-bg-color;\n    @include border-radius($form-feedback-tooltip-border-radius);\n  }\n\n  @include form-validation-state-selector($state) {\n    ~ .#{$state}-feedback,\n    ~ .#{$state}-tooltip {\n      display: block;\n    }\n  }\n\n  .form-control {\n    @include form-validation-state-selector($state) {\n      border-color: $border-color;\n\n      @if $enable-validation-icons {\n        padding-right: $input-height-inner;\n        background-image: escape-svg($icon);\n        background-repeat: no-repeat;\n        background-position: right $input-height-inner-quarter center;\n        background-size: $input-height-inner-half $input-height-inner-half;\n      }\n\n      &:focus {\n        border-color: $border-color;\n        @if $enable-shadows {\n          @include box-shadow($input-box-shadow, $focus-box-shadow);\n        } @else {\n          // Avoid using mixin so we can pass custom focus shadow properly\n          box-shadow: $focus-box-shadow;\n        }\n      }\n    }\n  }\n\n  // stylelint-disable-next-line selector-no-qualifying-type\n  textarea.form-control {\n    @include form-validation-state-selector($state) {\n      @if $enable-validation-icons {\n        padding-right: $input-height-inner;\n        background-position: top $input-height-inner-quarter right $input-height-inner-quarter;\n      }\n    }\n  }\n\n  .form-select {\n    @include form-validation-state-selector($state) {\n      border-color: $border-color;\n\n      @if $enable-validation-icons {\n        &:not([multiple]):not([size]),\n        &:not([multiple])[size=\"1\"] {\n          --#{$prefix}form-select-bg-icon: #{escape-svg($icon)};\n          padding-right: $form-select-feedback-icon-padding-end;\n          background-position: $form-select-bg-position, $form-select-feedback-icon-position;\n          background-size: $form-select-bg-size, $form-select-feedback-icon-size;\n        }\n      }\n\n      &:focus {\n        border-color: $border-color;\n        @if $enable-shadows {\n          @include box-shadow($form-select-box-shadow, $focus-box-shadow);\n        } @else {\n          // Avoid using mixin so we can pass custom focus shadow properly\n          box-shadow: $focus-box-shadow;\n        }\n      }\n    }\n  }\n\n  .form-control-color {\n    @include form-validation-state-selector($state) {\n      @if $enable-validation-icons {\n        width: add($form-color-width, $input-height-inner);\n      }\n    }\n  }\n\n  .form-check-input {\n    @include form-validation-state-selector($state) {\n      border-color: $border-color;\n\n      &:checked {\n        background-color: $color;\n      }\n\n      &:focus {\n        box-shadow: $focus-box-shadow;\n      }\n\n      ~ .form-check-label {\n        color: $color;\n      }\n    }\n  }\n  .form-check-inline .form-check-input {\n    ~ .#{$state}-feedback {\n      margin-left: .5em;\n    }\n  }\n\n  .input-group {\n    > .form-control:not(:focus),\n    > .form-select:not(:focus),\n    > .form-floating:not(:focus-within) {\n      @include form-validation-state-selector($state) {\n        @if $state == \"valid\" {\n          z-index: 3;\n        } @else if $state == \"invalid\" {\n          z-index: 4;\n        }\n      }\n    }\n  }\n}\n// scss-docs-end form-validation-mixins\n","//\n// Base styles\n//\n\n.btn {\n  // scss-docs-start btn-css-vars\n  --#{$prefix}btn-padding-x: #{$btn-padding-x};\n  --#{$prefix}btn-padding-y: #{$btn-padding-y};\n  --#{$prefix}btn-font-family: #{$btn-font-family};\n  @include rfs($btn-font-size, --#{$prefix}btn-font-size);\n  --#{$prefix}btn-font-weight: #{$btn-font-weight};\n  --#{$prefix}btn-line-height: #{$btn-line-height};\n  --#{$prefix}btn-color: #{$btn-color};\n  --#{$prefix}btn-bg: transparent;\n  --#{$prefix}btn-border-width: #{$btn-border-width};\n  --#{$prefix}btn-border-color: transparent;\n  --#{$prefix}btn-border-radius: #{$btn-border-radius};\n  --#{$prefix}btn-hover-border-color: transparent;\n  --#{$prefix}btn-box-shadow: #{$btn-box-shadow};\n  --#{$prefix}btn-disabled-opacity: #{$btn-disabled-opacity};\n  --#{$prefix}btn-focus-box-shadow: 0 0 0 #{$btn-focus-width} rgba(var(--#{$prefix}btn-focus-shadow-rgb), .5);\n  // scss-docs-end btn-css-vars\n\n  display: inline-block;\n  padding: var(--#{$prefix}btn-padding-y) var(--#{$prefix}btn-padding-x);\n  font-family: var(--#{$prefix}btn-font-family);\n  @include font-size(var(--#{$prefix}btn-font-size));\n  font-weight: var(--#{$prefix}btn-font-weight);\n  line-height: var(--#{$prefix}btn-line-height);\n  color: var(--#{$prefix}btn-color);\n  text-align: center;\n  text-decoration: if($link-decoration == none, null, none);\n  white-space: $btn-white-space;\n  vertical-align: middle;\n  cursor: if($enable-button-pointers, pointer, null);\n  user-select: none;\n  border: var(--#{$prefix}btn-border-width) solid var(--#{$prefix}btn-border-color);\n  @include border-radius(var(--#{$prefix}btn-border-radius));\n  @include gradient-bg(var(--#{$prefix}btn-bg));\n  @include box-shadow(var(--#{$prefix}btn-box-shadow));\n  @include transition($btn-transition);\n\n  &:hover {\n    color: var(--#{$prefix}btn-hover-color);\n    text-decoration: if($link-hover-decoration == underline, none, null);\n    background-color: var(--#{$prefix}btn-hover-bg);\n    border-color: var(--#{$prefix}btn-hover-border-color);\n  }\n\n  .btn-check + &:hover {\n    // override for the checkbox/radio buttons\n    color: var(--#{$prefix}btn-color);\n    background-color: var(--#{$prefix}btn-bg);\n    border-color: var(--#{$prefix}btn-border-color);\n  }\n\n  &:focus-visible {\n    color: var(--#{$prefix}btn-hover-color);\n    @include gradient-bg(var(--#{$prefix}btn-hover-bg));\n    border-color: var(--#{$prefix}btn-hover-border-color);\n    outline: 0;\n    // Avoid using mixin so we can pass custom focus shadow properly\n    @if $enable-shadows {\n      box-shadow: var(--#{$prefix}btn-box-shadow), var(--#{$prefix}btn-focus-box-shadow);\n    } @else {\n      box-shadow: var(--#{$prefix}btn-focus-box-shadow);\n    }\n  }\n\n  .btn-check:focus-visible + & {\n    border-color: var(--#{$prefix}btn-hover-border-color);\n    outline: 0;\n    // Avoid using mixin so we can pass custom focus shadow properly\n    @if $enable-shadows {\n      box-shadow: var(--#{$prefix}btn-box-shadow), var(--#{$prefix}btn-focus-box-shadow);\n    } @else {\n      box-shadow: var(--#{$prefix}btn-focus-box-shadow);\n    }\n  }\n\n  .btn-check:checked + &,\n  :not(.btn-check) + &:active,\n  &:first-child:active,\n  &.active,\n  &.show {\n    color: var(--#{$prefix}btn-active-color);\n    background-color: var(--#{$prefix}btn-active-bg);\n    // Remove CSS gradients if they're enabled\n    background-image: if($enable-gradients, none, null);\n    border-color: var(--#{$prefix}btn-active-border-color);\n    @include box-shadow(var(--#{$prefix}btn-active-shadow));\n\n    &:focus-visible {\n      // Avoid using mixin so we can pass custom focus shadow properly\n      @if $enable-shadows {\n        box-shadow: var(--#{$prefix}btn-active-shadow), var(--#{$prefix}btn-focus-box-shadow);\n      } @else {\n        box-shadow: var(--#{$prefix}btn-focus-box-shadow);\n      }\n    }\n  }\n\n  .btn-check:checked:focus-visible + & {\n    // Avoid using mixin so we can pass custom focus shadow properly\n    @if $enable-shadows {\n      box-shadow: var(--#{$prefix}btn-active-shadow), var(--#{$prefix}btn-focus-box-shadow);\n    } @else {\n      box-shadow: var(--#{$prefix}btn-focus-box-shadow);\n    }\n  }\n\n  &:disabled,\n  &.disabled,\n  fieldset:disabled & {\n    color: var(--#{$prefix}btn-disabled-color);\n    pointer-events: none;\n    background-color: var(--#{$prefix}btn-disabled-bg);\n    background-image: if($enable-gradients, none, null);\n    border-color: var(--#{$prefix}btn-disabled-border-color);\n    opacity: var(--#{$prefix}btn-disabled-opacity);\n    @include box-shadow(none);\n  }\n}\n\n\n//\n// Alternate buttons\n//\n\n// scss-docs-start btn-variant-loops\n@each $color, $value in $theme-colors {\n  .btn-#{$color} {\n    @if $color == \"light\" {\n      @include button-variant(\n        $value,\n        $value,\n        $hover-background: shade-color($value, $btn-hover-bg-shade-amount),\n        $hover-border: shade-color($value, $btn-hover-border-shade-amount),\n        $active-background: shade-color($value, $btn-active-bg-shade-amount),\n        $active-border: shade-color($value, $btn-active-border-shade-amount)\n      );\n    } @else if $color == \"dark\" {\n      @include button-variant(\n        $value,\n        $value,\n        $hover-background: tint-color($value, $btn-hover-bg-tint-amount),\n        $hover-border: tint-color($value, $btn-hover-border-tint-amount),\n        $active-background: tint-color($value, $btn-active-bg-tint-amount),\n        $active-border: tint-color($value, $btn-active-border-tint-amount)\n      );\n    } @else {\n      @include button-variant($value, $value);\n    }\n  }\n}\n\n@each $color, $value in $theme-colors {\n  .btn-outline-#{$color} {\n    @include button-outline-variant($value);\n  }\n}\n// scss-docs-end btn-variant-loops\n\n\n//\n// Link buttons\n//\n\n// Make a button look and behave like a link\n.btn-link {\n  --#{$prefix}btn-font-weight: #{$font-weight-normal};\n  --#{$prefix}btn-color: #{$btn-link-color};\n  --#{$prefix}btn-bg: transparent;\n  --#{$prefix}btn-border-color: transparent;\n  --#{$prefix}btn-hover-color: #{$btn-link-hover-color};\n  --#{$prefix}btn-hover-border-color: transparent;\n  --#{$prefix}btn-active-color: #{$btn-link-hover-color};\n  --#{$prefix}btn-active-border-color: transparent;\n  --#{$prefix}btn-disabled-color: #{$btn-link-disabled-color};\n  --#{$prefix}btn-disabled-border-color: transparent;\n  --#{$prefix}btn-box-shadow: 0 0 0 #000; // Can't use `none` as keyword negates all values when used with multiple shadows\n  --#{$prefix}btn-focus-shadow-rgb: #{$btn-link-focus-shadow-rgb};\n\n  text-decoration: $link-decoration;\n  @if $enable-gradients {\n    background-image: none;\n  }\n\n  &:hover,\n  &:focus-visible {\n    text-decoration: $link-hover-decoration;\n  }\n\n  &:focus-visible {\n    color: var(--#{$prefix}btn-color);\n  }\n\n  &:hover {\n    color: var(--#{$prefix}btn-hover-color);\n  }\n\n  // No need for an active state here\n}\n\n\n//\n// Button Sizes\n//\n\n.btn-lg {\n  @include button-size($btn-padding-y-lg, $btn-padding-x-lg, $btn-font-size-lg, $btn-border-radius-lg);\n}\n\n.btn-sm {\n  @include button-size($btn-padding-y-sm, $btn-padding-x-sm, $btn-font-size-sm, $btn-border-radius-sm);\n}\n","// Button variants\n//\n// Easily pump out default styles, as well as :hover, :focus, :active,\n// and disabled options for all buttons\n\n// scss-docs-start btn-variant-mixin\n@mixin button-variant(\n  $background,\n  $border,\n  $color: color-contrast($background),\n  $hover-background: if($color == $color-contrast-light, shade-color($background, $btn-hover-bg-shade-amount), tint-color($background, $btn-hover-bg-tint-amount)),\n  $hover-border: if($color == $color-contrast-light, shade-color($border, $btn-hover-border-shade-amount), tint-color($border, $btn-hover-border-tint-amount)),\n  $hover-color: color-contrast($hover-background),\n  $active-background: if($color == $color-contrast-light, shade-color($background, $btn-active-bg-shade-amount), tint-color($background, $btn-active-bg-tint-amount)),\n  $active-border: if($color == $color-contrast-light, shade-color($border, $btn-active-border-shade-amount), tint-color($border, $btn-active-border-tint-amount)),\n  $active-color: color-contrast($active-background),\n  $disabled-background: $background,\n  $disabled-border: $border,\n  $disabled-color: color-contrast($disabled-background)\n) {\n  --#{$prefix}btn-color: #{$color};\n  --#{$prefix}btn-bg: #{$background};\n  --#{$prefix}btn-border-color: #{$border};\n  --#{$prefix}btn-hover-color: #{$hover-color};\n  --#{$prefix}btn-hover-bg: #{$hover-background};\n  --#{$prefix}btn-hover-border-color: #{$hover-border};\n  --#{$prefix}btn-focus-shadow-rgb: #{to-rgb(mix($color, $border, 15%))};\n  --#{$prefix}btn-active-color: #{$active-color};\n  --#{$prefix}btn-active-bg: #{$active-background};\n  --#{$prefix}btn-active-border-color: #{$active-border};\n  --#{$prefix}btn-active-shadow: #{$btn-active-box-shadow};\n  --#{$prefix}btn-disabled-color: #{$disabled-color};\n  --#{$prefix}btn-disabled-bg: #{$disabled-background};\n  --#{$prefix}btn-disabled-border-color: #{$disabled-border};\n}\n// scss-docs-end btn-variant-mixin\n\n// scss-docs-start btn-outline-variant-mixin\n@mixin button-outline-variant(\n  $color,\n  $color-hover: color-contrast($color),\n  $active-background: $color,\n  $active-border: $color,\n  $active-color: color-contrast($active-background)\n) {\n  --#{$prefix}btn-color: #{$color};\n  --#{$prefix}btn-border-color: #{$color};\n  --#{$prefix}btn-hover-color: #{$color-hover};\n  --#{$prefix}btn-hover-bg: #{$active-background};\n  --#{$prefix}btn-hover-border-color: #{$active-border};\n  --#{$prefix}btn-focus-shadow-rgb: #{to-rgb($color)};\n  --#{$prefix}btn-active-color: #{$active-color};\n  --#{$prefix}btn-active-bg: #{$active-background};\n  --#{$prefix}btn-active-border-color: #{$active-border};\n  --#{$prefix}btn-active-shadow: #{$btn-active-box-shadow};\n  --#{$prefix}btn-disabled-color: #{$color};\n  --#{$prefix}btn-disabled-bg: transparent;\n  --#{$prefix}btn-disabled-border-color: #{$color};\n  --#{$prefix}gradient: none;\n}\n// scss-docs-end btn-outline-variant-mixin\n\n// scss-docs-start btn-size-mixin\n@mixin button-size($padding-y, $padding-x, $font-size, $border-radius) {\n  --#{$prefix}btn-padding-y: #{$padding-y};\n  --#{$prefix}btn-padding-x: #{$padding-x};\n  @include rfs($font-size, --#{$prefix}btn-font-size);\n  --#{$prefix}btn-border-radius: #{$border-radius};\n}\n// scss-docs-end btn-size-mixin\n",".fade {\n  @include transition($transition-fade);\n\n  &:not(.show) {\n    opacity: 0;\n  }\n}\n\n// scss-docs-start collapse-classes\n.collapse {\n  &:not(.show) {\n    display: none;\n  }\n}\n\n.collapsing {\n  height: 0;\n  overflow: hidden;\n  @include transition($transition-collapse);\n\n  &.collapse-horizontal {\n    width: 0;\n    height: auto;\n    @include transition($transition-collapse-width);\n  }\n}\n// scss-docs-end collapse-classes\n","// The dropdown wrapper (`<div>`)\n.dropup,\n.dropend,\n.dropdown,\n.dropstart,\n.dropup-center,\n.dropdown-center {\n  position: relative;\n}\n\n.dropdown-toggle {\n  white-space: nowrap;\n\n  // Generate the caret automatically\n  @include caret();\n}\n\n// The dropdown menu\n.dropdown-menu {\n  // scss-docs-start dropdown-css-vars\n  --#{$prefix}dropdown-zindex: #{$zindex-dropdown};\n  --#{$prefix}dropdown-min-width: #{$dropdown-min-width};\n  --#{$prefix}dropdown-padding-x: #{$dropdown-padding-x};\n  --#{$prefix}dropdown-padding-y: #{$dropdown-padding-y};\n  --#{$prefix}dropdown-spacer: #{$dropdown-spacer};\n  @include rfs($dropdown-font-size, --#{$prefix}dropdown-font-size);\n  --#{$prefix}dropdown-color: #{$dropdown-color};\n  --#{$prefix}dropdown-bg: #{$dropdown-bg};\n  --#{$prefix}dropdown-border-color: #{$dropdown-border-color};\n  --#{$prefix}dropdown-border-radius: #{$dropdown-border-radius};\n  --#{$prefix}dropdown-border-width: #{$dropdown-border-width};\n  --#{$prefix}dropdown-inner-border-radius: #{$dropdown-inner-border-radius};\n  --#{$prefix}dropdown-divider-bg: #{$dropdown-divider-bg};\n  --#{$prefix}dropdown-divider-margin-y: #{$dropdown-divider-margin-y};\n  --#{$prefix}dropdown-box-shadow: #{$dropdown-box-shadow};\n  --#{$prefix}dropdown-link-color: #{$dropdown-link-color};\n  --#{$prefix}dropdown-link-hover-color: #{$dropdown-link-hover-color};\n  --#{$prefix}dropdown-link-hover-bg: #{$dropdown-link-hover-bg};\n  --#{$prefix}dropdown-link-active-color: #{$dropdown-link-active-color};\n  --#{$prefix}dropdown-link-active-bg: #{$dropdown-link-active-bg};\n  --#{$prefix}dropdown-link-disabled-color: #{$dropdown-link-disabled-color};\n  --#{$prefix}dropdown-item-padding-x: #{$dropdown-item-padding-x};\n  --#{$prefix}dropdown-item-padding-y: #{$dropdown-item-padding-y};\n  --#{$prefix}dropdown-header-color: #{$dropdown-header-color};\n  --#{$prefix}dropdown-header-padding-x: #{$dropdown-header-padding-x};\n  --#{$prefix}dropdown-header-padding-y: #{$dropdown-header-padding-y};\n  // scss-docs-end dropdown-css-vars\n\n  position: absolute;\n  z-index: var(--#{$prefix}dropdown-zindex);\n  display: none; // none by default, but block on \"open\" of the menu\n  min-width: var(--#{$prefix}dropdown-min-width);\n  padding: var(--#{$prefix}dropdown-padding-y) var(--#{$prefix}dropdown-padding-x);\n  margin: 0; // Override default margin of ul\n  @include font-size(var(--#{$prefix}dropdown-font-size));\n  color: var(--#{$prefix}dropdown-color);\n  text-align: left; // Ensures proper alignment if parent has it changed (e.g., modal footer)\n  list-style: none;\n  background-color: var(--#{$prefix}dropdown-bg);\n  background-clip: padding-box;\n  border: var(--#{$prefix}dropdown-border-width) solid var(--#{$prefix}dropdown-border-color);\n  @include border-radius(var(--#{$prefix}dropdown-border-radius));\n  @include box-shadow(var(--#{$prefix}dropdown-box-shadow));\n\n  &[data-bs-popper] {\n    top: 100%;\n    left: 0;\n    margin-top: var(--#{$prefix}dropdown-spacer);\n  }\n\n  @if $dropdown-padding-y == 0 {\n    > .dropdown-item:first-child,\n    > li:first-child .dropdown-item {\n      @include border-top-radius(var(--#{$prefix}dropdown-inner-border-radius));\n    }\n    > .dropdown-item:last-child,\n    > li:last-child .dropdown-item {\n      @include border-bottom-radius(var(--#{$prefix}dropdown-inner-border-radius));\n    }\n\n  }\n}\n\n// scss-docs-start responsive-breakpoints\n// We deliberately hardcode the `bs-` prefix because we check\n// this custom property in JS to determine Popper's positioning\n\n@each $breakpoint in map-keys($grid-breakpoints) {\n  @include media-breakpoint-up($breakpoint) {\n    $infix: breakpoint-infix($breakpoint, $grid-breakpoints);\n\n    .dropdown-menu#{$infix}-start {\n      --bs-position: start;\n\n      &[data-bs-popper] {\n        right: auto;\n        left: 0;\n      }\n    }\n\n    .dropdown-menu#{$infix}-end {\n      --bs-position: end;\n\n      &[data-bs-popper] {\n        right: 0;\n        left: auto;\n      }\n    }\n  }\n}\n// scss-docs-end responsive-breakpoints\n\n// Allow for dropdowns to go bottom up (aka, dropup-menu)\n// Just add .dropup after the standard .dropdown class and you're set.\n.dropup {\n  .dropdown-menu[data-bs-popper] {\n    top: auto;\n    bottom: 100%;\n    margin-top: 0;\n    margin-bottom: var(--#{$prefix}dropdown-spacer);\n  }\n\n  .dropdown-toggle {\n    @include caret(up);\n  }\n}\n\n.dropend {\n  .dropdown-menu[data-bs-popper] {\n    top: 0;\n    right: auto;\n    left: 100%;\n    margin-top: 0;\n    margin-left: var(--#{$prefix}dropdown-spacer);\n  }\n\n  .dropdown-toggle {\n    @include caret(end);\n    &::after {\n      vertical-align: 0;\n    }\n  }\n}\n\n.dropstart {\n  .dropdown-menu[data-bs-popper] {\n    top: 0;\n    right: 100%;\n    left: auto;\n    margin-top: 0;\n    margin-right: var(--#{$prefix}dropdown-spacer);\n  }\n\n  .dropdown-toggle {\n    @include caret(start);\n    &::before {\n      vertical-align: 0;\n    }\n  }\n}\n\n\n// Dividers (basically an `<hr>`) within the dropdown\n.dropdown-divider {\n  height: 0;\n  margin: var(--#{$prefix}dropdown-divider-margin-y) 0;\n  overflow: hidden;\n  border-top: 1px solid var(--#{$prefix}dropdown-divider-bg);\n  opacity: 1; // Revisit in v6 to de-dupe styles that conflict with <hr> element\n}\n\n// Links, buttons, and more within the dropdown menu\n//\n// `<button>`-specific styles are denoted with `// For <button>s`\n.dropdown-item {\n  display: block;\n  width: 100%; // For `<button>`s\n  padding: var(--#{$prefix}dropdown-item-padding-y) var(--#{$prefix}dropdown-item-padding-x);\n  clear: both;\n  font-weight: $font-weight-normal;\n  color: var(--#{$prefix}dropdown-link-color);\n  text-align: inherit; // For `<button>`s\n  text-decoration: if($link-decoration == none, null, none);\n  white-space: nowrap; // prevent links from randomly breaking onto new lines\n  background-color: transparent; // For `<button>`s\n  border: 0; // For `<button>`s\n  @include border-radius(var(--#{$prefix}dropdown-item-border-radius, 0));\n\n  &:hover,\n  &:focus {\n    color: var(--#{$prefix}dropdown-link-hover-color);\n    text-decoration: if($link-hover-decoration == underline, none, null);\n    @include gradient-bg(var(--#{$prefix}dropdown-link-hover-bg));\n  }\n\n  &.active,\n  &:active {\n    color: var(--#{$prefix}dropdown-link-active-color);\n    text-decoration: none;\n    @include gradient-bg(var(--#{$prefix}dropdown-link-active-bg));\n  }\n\n  &.disabled,\n  &:disabled {\n    color: var(--#{$prefix}dropdown-link-disabled-color);\n    pointer-events: none;\n    background-color: transparent;\n    // Remove CSS gradients if they're enabled\n    background-image: if($enable-gradients, none, null);\n  }\n}\n\n.dropdown-menu.show {\n  display: block;\n}\n\n// Dropdown section headers\n.dropdown-header {\n  display: block;\n  padding: var(--#{$prefix}dropdown-header-padding-y) var(--#{$prefix}dropdown-header-padding-x);\n  margin-bottom: 0; // for use with heading elements\n  @include font-size($font-size-sm);\n  color: var(--#{$prefix}dropdown-header-color);\n  white-space: nowrap; // as with > li > a\n}\n\n// Dropdown text\n.dropdown-item-text {\n  display: block;\n  padding: var(--#{$prefix}dropdown-item-padding-y) var(--#{$prefix}dropdown-item-padding-x);\n  color: var(--#{$prefix}dropdown-link-color);\n}\n\n// Dark dropdowns\n.dropdown-menu-dark {\n  // scss-docs-start dropdown-dark-css-vars\n  --#{$prefix}dropdown-color: #{$dropdown-dark-color};\n  --#{$prefix}dropdown-bg: #{$dropdown-dark-bg};\n  --#{$prefix}dropdown-border-color: #{$dropdown-dark-border-color};\n  --#{$prefix}dropdown-box-shadow: #{$dropdown-dark-box-shadow};\n  --#{$prefix}dropdown-link-color: #{$dropdown-dark-link-color};\n  --#{$prefix}dropdown-link-hover-color: #{$dropdown-dark-link-hover-color};\n  --#{$prefix}dropdown-divider-bg: #{$dropdown-dark-divider-bg};\n  --#{$prefix}dropdown-link-hover-bg: #{$dropdown-dark-link-hover-bg};\n  --#{$prefix}dropdown-link-active-color: #{$dropdown-dark-link-active-color};\n  --#{$prefix}dropdown-link-active-bg: #{$dropdown-dark-link-active-bg};\n  --#{$prefix}dropdown-link-disabled-color: #{$dropdown-dark-link-disabled-color};\n  --#{$prefix}dropdown-header-color: #{$dropdown-dark-header-color};\n  // scss-docs-end dropdown-dark-css-vars\n}\n","// scss-docs-start caret-mixins\n@mixin caret-down($width: $caret-width) {\n  border-top: $width solid;\n  border-right: $width solid transparent;\n  border-bottom: 0;\n  border-left: $width solid transparent;\n}\n\n@mixin caret-up($width: $caret-width) {\n  border-top: 0;\n  border-right: $width solid transparent;\n  border-bottom: $width solid;\n  border-left: $width solid transparent;\n}\n\n@mixin caret-end($width: $caret-width) {\n  border-top: $width solid transparent;\n  border-right: 0;\n  border-bottom: $width solid transparent;\n  border-left: $width solid;\n}\n\n@mixin caret-start($width: $caret-width) {\n  border-top: $width solid transparent;\n  border-right: $width solid;\n  border-bottom: $width solid transparent;\n}\n\n@mixin caret(\n  $direction: down,\n  $width: $caret-width,\n  $spacing: $caret-spacing,\n  $vertical-align: $caret-vertical-align\n) {\n  @if $enable-caret {\n    &::after {\n      display: inline-block;\n      margin-left: $spacing;\n      vertical-align: $vertical-align;\n      content: \"\";\n      @if $direction == down {\n        @include caret-down($width);\n      } @else if $direction == up {\n        @include caret-up($width);\n      } @else if $direction == end {\n        @include caret-end($width);\n      }\n    }\n\n    @if $direction == start {\n      &::after {\n        display: none;\n      }\n\n      &::before {\n        display: inline-block;\n        margin-right: $spacing;\n        vertical-align: $vertical-align;\n        content: \"\";\n        @include caret-start($width);\n      }\n    }\n\n    &:empty::after {\n      margin-left: 0;\n    }\n  }\n}\n// scss-docs-end caret-mixins\n","// Make the div behave like a button\n.btn-group,\n.btn-group-vertical {\n  position: relative;\n  display: inline-flex;\n  vertical-align: middle; // match .btn alignment given font-size hack above\n\n  > .btn {\n    position: relative;\n    flex: 1 1 auto;\n  }\n\n  // Bring the hover, focused, and \"active\" buttons to the front to overlay\n  // the borders properly\n  > .btn-check:checked + .btn,\n  > .btn-check:focus + .btn,\n  > .btn:hover,\n  > .btn:focus,\n  > .btn:active,\n  > .btn.active {\n    z-index: 1;\n  }\n}\n\n// Optional: Group multiple button groups together for a toolbar\n.btn-toolbar {\n  display: flex;\n  flex-wrap: wrap;\n  justify-content: flex-start;\n\n  .input-group {\n    width: auto;\n  }\n}\n\n.btn-group {\n  @include border-radius($btn-border-radius);\n\n  // Prevent double borders when buttons are next to each other\n  > :not(.btn-check:first-child) + .btn,\n  > .btn-group:not(:first-child) {\n    margin-left: calc(#{$btn-border-width} * -1); // stylelint-disable-line function-disallowed-list\n  }\n\n  // Reset rounded corners\n  > .btn:not(:last-child):not(.dropdown-toggle),\n  > .btn.dropdown-toggle-split:first-child,\n  > .btn-group:not(:last-child) > .btn {\n    @include border-end-radius(0);\n  }\n\n  // The left radius should be 0 if the button is:\n  // - the \"third or more\" child\n  // - the second child and the previous element isn't `.btn-check` (making it the first child visually)\n  // - part of a btn-group which isn't the first child\n  > .btn:nth-child(n + 3),\n  > :not(.btn-check) + .btn,\n  > .btn-group:not(:first-child) > .btn {\n    @include border-start-radius(0);\n  }\n}\n\n// Sizing\n//\n// Remix the default button sizing classes into new ones for easier manipulation.\n\n.btn-group-sm > .btn { @extend .btn-sm; }\n.btn-group-lg > .btn { @extend .btn-lg; }\n\n\n//\n// Split button dropdowns\n//\n\n.dropdown-toggle-split {\n  padding-right: $btn-padding-x * .75;\n  padding-left: $btn-padding-x * .75;\n\n  &::after,\n  .dropup &::after,\n  .dropend &::after {\n    margin-left: 0;\n  }\n\n  .dropstart &::before {\n    margin-right: 0;\n  }\n}\n\n.btn-sm + .dropdown-toggle-split {\n  padding-right: $btn-padding-x-sm * .75;\n  padding-left: $btn-padding-x-sm * .75;\n}\n\n.btn-lg + .dropdown-toggle-split {\n  padding-right: $btn-padding-x-lg * .75;\n  padding-left: $btn-padding-x-lg * .75;\n}\n\n\n// The clickable button for toggling the menu\n// Set the same inset shadow as the :active state\n.btn-group.show .dropdown-toggle {\n  @include box-shadow($btn-active-box-shadow);\n\n  // Show no shadow for `.btn-link` since it has no other button styles.\n  &.btn-link {\n    @include box-shadow(none);\n  }\n}\n\n\n//\n// Vertical button groups\n//\n\n.btn-group-vertical {\n  flex-direction: column;\n  align-items: flex-start;\n  justify-content: center;\n\n  > .btn,\n  > .btn-group {\n    width: 100%;\n  }\n\n  > .btn:not(:first-child),\n  > .btn-group:not(:first-child) {\n    margin-top: calc(#{$btn-border-width} * -1); // stylelint-disable-line function-disallowed-list\n  }\n\n  // Reset rounded corners\n  > .btn:not(:last-child):not(.dropdown-toggle),\n  > .btn-group:not(:last-child) > .btn {\n    @include border-bottom-radius(0);\n  }\n\n  > .btn ~ .btn,\n  > .btn-group:not(:first-child) > .btn {\n    @include border-top-radius(0);\n  }\n}\n","// Base class\n//\n// Kickstart any navigation component with a set of style resets. Works with\n// `<nav>`s, `<ul>`s or `<ol>`s.\n\n.nav {\n  // scss-docs-start nav-css-vars\n  --#{$prefix}nav-link-padding-x: #{$nav-link-padding-x};\n  --#{$prefix}nav-link-padding-y: #{$nav-link-padding-y};\n  @include rfs($nav-link-font-size, --#{$prefix}nav-link-font-size);\n  --#{$prefix}nav-link-font-weight: #{$nav-link-font-weight};\n  --#{$prefix}nav-link-color: #{$nav-link-color};\n  --#{$prefix}nav-link-hover-color: #{$nav-link-hover-color};\n  --#{$prefix}nav-link-disabled-color: #{$nav-link-disabled-color};\n  // scss-docs-end nav-css-vars\n\n  display: flex;\n  flex-wrap: wrap;\n  padding-left: 0;\n  margin-bottom: 0;\n  list-style: none;\n}\n\n.nav-link {\n  display: block;\n  padding: var(--#{$prefix}nav-link-padding-y) var(--#{$prefix}nav-link-padding-x);\n  @include font-size(var(--#{$prefix}nav-link-font-size));\n  font-weight: var(--#{$prefix}nav-link-font-weight);\n  color: var(--#{$prefix}nav-link-color);\n  text-decoration: if($link-decoration == none, null, none);\n  background: none;\n  border: 0;\n  @include transition($nav-link-transition);\n\n  &:hover,\n  &:focus {\n    color: var(--#{$prefix}nav-link-hover-color);\n    text-decoration: if($link-hover-decoration == underline, none, null);\n  }\n\n  &:focus-visible {\n    outline: 0;\n    box-shadow: $nav-link-focus-box-shadow;\n  }\n\n  // Disabled state lightens text\n  &.disabled,\n  &:disabled {\n    color: var(--#{$prefix}nav-link-disabled-color);\n    pointer-events: none;\n    cursor: default;\n  }\n}\n\n//\n// Tabs\n//\n\n.nav-tabs {\n  // scss-docs-start nav-tabs-css-vars\n  --#{$prefix}nav-tabs-border-width: #{$nav-tabs-border-width};\n  --#{$prefix}nav-tabs-border-color: #{$nav-tabs-border-color};\n  --#{$prefix}nav-tabs-border-radius: #{$nav-tabs-border-radius};\n  --#{$prefix}nav-tabs-link-hover-border-color: #{$nav-tabs-link-hover-border-color};\n  --#{$prefix}nav-tabs-link-active-color: #{$nav-tabs-link-active-color};\n  --#{$prefix}nav-tabs-link-active-bg: #{$nav-tabs-link-active-bg};\n  --#{$prefix}nav-tabs-link-active-border-color: #{$nav-tabs-link-active-border-color};\n  // scss-docs-end nav-tabs-css-vars\n\n  border-bottom: var(--#{$prefix}nav-tabs-border-width) solid var(--#{$prefix}nav-tabs-border-color);\n\n  .nav-link {\n    margin-bottom: calc(-1 * var(--#{$prefix}nav-tabs-border-width)); // stylelint-disable-line function-disallowed-list\n    border: var(--#{$prefix}nav-tabs-border-width) solid transparent;\n    @include border-top-radius(var(--#{$prefix}nav-tabs-border-radius));\n\n    &:hover,\n    &:focus {\n      // Prevents active .nav-link tab overlapping focus outline of previous/next .nav-link\n      isolation: isolate;\n      border-color: var(--#{$prefix}nav-tabs-link-hover-border-color);\n    }\n  }\n\n  .nav-link.active,\n  .nav-item.show .nav-link {\n    color: var(--#{$prefix}nav-tabs-link-active-color);\n    background-color: var(--#{$prefix}nav-tabs-link-active-bg);\n    border-color: var(--#{$prefix}nav-tabs-link-active-border-color);\n  }\n\n  .dropdown-menu {\n    // Make dropdown border overlap tab border\n    margin-top: calc(-1 * var(--#{$prefix}nav-tabs-border-width)); // stylelint-disable-line function-disallowed-list\n    // Remove the top rounded corners here since there is a hard edge above the menu\n    @include border-top-radius(0);\n  }\n}\n\n\n//\n// Pills\n//\n\n.nav-pills {\n  // scss-docs-start nav-pills-css-vars\n  --#{$prefix}nav-pills-border-radius: #{$nav-pills-border-radius};\n  --#{$prefix}nav-pills-link-active-color: #{$nav-pills-link-active-color};\n  --#{$prefix}nav-pills-link-active-bg: #{$nav-pills-link-active-bg};\n  // scss-docs-end nav-pills-css-vars\n\n  .nav-link {\n    @include border-radius(var(--#{$prefix}nav-pills-border-radius));\n  }\n\n  .nav-link.active,\n  .show > .nav-link {\n    color: var(--#{$prefix}nav-pills-link-active-color);\n    @include gradient-bg(var(--#{$prefix}nav-pills-link-active-bg));\n  }\n}\n\n\n//\n// Underline\n//\n\n.nav-underline {\n  // scss-docs-start nav-underline-css-vars\n  --#{$prefix}nav-underline-gap: #{$nav-underline-gap};\n  --#{$prefix}nav-underline-border-width: #{$nav-underline-border-width};\n  --#{$prefix}nav-underline-link-active-color: #{$nav-underline-link-active-color};\n  // scss-docs-end nav-underline-css-vars\n\n  gap: var(--#{$prefix}nav-underline-gap);\n\n  .nav-link {\n    padding-right: 0;\n    padding-left: 0;\n    border-bottom: var(--#{$prefix}nav-underline-border-width) solid transparent;\n\n    &:hover,\n    &:focus {\n      border-bottom-color: currentcolor;\n    }\n  }\n\n  .nav-link.active,\n  .show > .nav-link {\n    font-weight: $font-weight-bold;\n    color: var(--#{$prefix}nav-underline-link-active-color);\n    border-bottom-color: currentcolor;\n  }\n}\n\n\n//\n// Justified variants\n//\n\n.nav-fill {\n  > .nav-link,\n  .nav-item {\n    flex: 1 1 auto;\n    text-align: center;\n  }\n}\n\n.nav-justified {\n  > .nav-link,\n  .nav-item {\n    flex-basis: 0;\n    flex-grow: 1;\n    text-align: center;\n  }\n}\n\n.nav-fill,\n.nav-justified {\n  .nav-item .nav-link {\n    width: 100%; // Make sure button will grow\n  }\n}\n\n\n// Tabbable tabs\n//\n// Hide tabbable panes to start, show them when `.active`\n\n.tab-content {\n  > .tab-pane {\n    display: none;\n  }\n  > .active {\n    display: block;\n  }\n}\n","// Navbar\n//\n// Provide a static navbar from which we expand to create full-width, fixed, and\n// other navbar variations.\n\n.navbar {\n  // scss-docs-start navbar-css-vars\n  --#{$prefix}navbar-padding-x: #{if($navbar-padding-x == null, 0, $navbar-padding-x)};\n  --#{$prefix}navbar-padding-y: #{$navbar-padding-y};\n  --#{$prefix}navbar-color: #{$navbar-light-color};\n  --#{$prefix}navbar-hover-color: #{$navbar-light-hover-color};\n  --#{$prefix}navbar-disabled-color: #{$navbar-light-disabled-color};\n  --#{$prefix}navbar-active-color: #{$navbar-light-active-color};\n  --#{$prefix}navbar-brand-padding-y: #{$navbar-brand-padding-y};\n  --#{$prefix}navbar-brand-margin-end: #{$navbar-brand-margin-end};\n  --#{$prefix}navbar-brand-font-size: #{$navbar-brand-font-size};\n  --#{$prefix}navbar-brand-color: #{$navbar-light-brand-color};\n  --#{$prefix}navbar-brand-hover-color: #{$navbar-light-brand-hover-color};\n  --#{$prefix}navbar-nav-link-padding-x: #{$navbar-nav-link-padding-x};\n  --#{$prefix}navbar-toggler-padding-y: #{$navbar-toggler-padding-y};\n  --#{$prefix}navbar-toggler-padding-x: #{$navbar-toggler-padding-x};\n  --#{$prefix}navbar-toggler-font-size: #{$navbar-toggler-font-size};\n  --#{$prefix}navbar-toggler-icon-bg: #{escape-svg($navbar-light-toggler-icon-bg)};\n  --#{$prefix}navbar-toggler-border-color: #{$navbar-light-toggler-border-color};\n  --#{$prefix}navbar-toggler-border-radius: #{$navbar-toggler-border-radius};\n  --#{$prefix}navbar-toggler-focus-width: #{$navbar-toggler-focus-width};\n  --#{$prefix}navbar-toggler-transition: #{$navbar-toggler-transition};\n  // scss-docs-end navbar-css-vars\n\n  position: relative;\n  display: flex;\n  flex-wrap: wrap; // allow us to do the line break for collapsing content\n  align-items: center;\n  justify-content: space-between; // space out brand from logo\n  padding: var(--#{$prefix}navbar-padding-y) var(--#{$prefix}navbar-padding-x);\n  @include gradient-bg();\n\n  // Because flex properties aren't inherited, we need to redeclare these first\n  // few properties so that content nested within behave properly.\n  // The `flex-wrap` property is inherited to simplify the expanded navbars\n  %container-flex-properties {\n    display: flex;\n    flex-wrap: inherit;\n    align-items: center;\n    justify-content: space-between;\n  }\n\n  > .container,\n  > .container-fluid {\n    @extend %container-flex-properties;\n  }\n\n  @each $breakpoint, $container-max-width in $container-max-widths {\n    > .container#{breakpoint-infix($breakpoint, $container-max-widths)} {\n      @extend %container-flex-properties;\n    }\n  }\n}\n\n\n// Navbar brand\n//\n// Used for brand, project, or site names.\n\n.navbar-brand {\n  padding-top: var(--#{$prefix}navbar-brand-padding-y);\n  padding-bottom: var(--#{$prefix}navbar-brand-padding-y);\n  margin-right: var(--#{$prefix}navbar-brand-margin-end);\n  @include font-size(var(--#{$prefix}navbar-brand-font-size));\n  color: var(--#{$prefix}navbar-brand-color);\n  text-decoration: if($link-decoration == none, null, none);\n  white-space: nowrap;\n\n  &:hover,\n  &:focus {\n    color: var(--#{$prefix}navbar-brand-hover-color);\n    text-decoration: if($link-hover-decoration == underline, none, null);\n  }\n}\n\n\n// Navbar nav\n//\n// Custom navbar navigation (doesn't require `.nav`, but does make use of `.nav-link`).\n\n.navbar-nav {\n  // scss-docs-start navbar-nav-css-vars\n  --#{$prefix}nav-link-padding-x: 0;\n  --#{$prefix}nav-link-padding-y: #{$nav-link-padding-y};\n  @include rfs($nav-link-font-size, --#{$prefix}nav-link-font-size);\n  --#{$prefix}nav-link-font-weight: #{$nav-link-font-weight};\n  --#{$prefix}nav-link-color: var(--#{$prefix}navbar-color);\n  --#{$prefix}nav-link-hover-color: var(--#{$prefix}navbar-hover-color);\n  --#{$prefix}nav-link-disabled-color: var(--#{$prefix}navbar-disabled-color);\n  // scss-docs-end navbar-nav-css-vars\n\n  display: flex;\n  flex-direction: column; // cannot use `inherit` to get the `.navbar`s value\n  padding-left: 0;\n  margin-bottom: 0;\n  list-style: none;\n\n  .nav-link {\n    &.active,\n    &.show {\n      color: var(--#{$prefix}navbar-active-color);\n    }\n  }\n\n  .dropdown-menu {\n    position: static;\n  }\n}\n\n\n// Navbar text\n//\n//\n\n.navbar-text {\n  padding-top: $nav-link-padding-y;\n  padding-bottom: $nav-link-padding-y;\n  color: var(--#{$prefix}navbar-color);\n\n  a,\n  a:hover,\n  a:focus  {\n    color: var(--#{$prefix}navbar-active-color);\n  }\n}\n\n\n// Responsive navbar\n//\n// Custom styles for responsive collapsing and toggling of navbar contents.\n// Powered by the collapse Bootstrap JavaScript plugin.\n\n// When collapsed, prevent the toggleable navbar contents from appearing in\n// the default flexbox row orientation. Requires the use of `flex-wrap: wrap`\n// on the `.navbar` parent.\n.navbar-collapse {\n  flex-basis: 100%;\n  flex-grow: 1;\n  // For always expanded or extra full navbars, ensure content aligns itself\n  // properly vertically. Can be easily overridden with flex utilities.\n  align-items: center;\n}\n\n// Button for toggling the navbar when in its collapsed state\n.navbar-toggler {\n  padding: var(--#{$prefix}navbar-toggler-padding-y) var(--#{$prefix}navbar-toggler-padding-x);\n  @include font-size(var(--#{$prefix}navbar-toggler-font-size));\n  line-height: 1;\n  color: var(--#{$prefix}navbar-color);\n  background-color: transparent; // remove default button style\n  border: var(--#{$prefix}border-width) solid var(--#{$prefix}navbar-toggler-border-color); // remove default button style\n  @include border-radius(var(--#{$prefix}navbar-toggler-border-radius));\n  @include transition(var(--#{$prefix}navbar-toggler-transition));\n\n  &:hover {\n    text-decoration: none;\n  }\n\n  &:focus {\n    text-decoration: none;\n    outline: 0;\n    box-shadow: 0 0 0 var(--#{$prefix}navbar-toggler-focus-width);\n  }\n}\n\n// Keep as a separate element so folks can easily override it with another icon\n// or image file as needed.\n.navbar-toggler-icon {\n  display: inline-block;\n  width: 1.5em;\n  height: 1.5em;\n  vertical-align: middle;\n  background-image: var(--#{$prefix}navbar-toggler-icon-bg);\n  background-repeat: no-repeat;\n  background-position: center;\n  background-size: 100%;\n}\n\n.navbar-nav-scroll {\n  max-height: var(--#{$prefix}scroll-height, 75vh);\n  overflow-y: auto;\n}\n\n// scss-docs-start navbar-expand-loop\n// Generate series of `.navbar-expand-*` responsive classes for configuring\n// where your navbar collapses.\n.navbar-expand {\n  @each $breakpoint in map-keys($grid-breakpoints) {\n    $next: breakpoint-next($breakpoint, $grid-breakpoints);\n    $infix: breakpoint-infix($next, $grid-breakpoints);\n\n    // stylelint-disable-next-line scss/selector-no-union-class-name\n    &#{$infix} {\n      @include media-breakpoint-up($next) {\n        flex-wrap: nowrap;\n        justify-content: flex-start;\n\n        .navbar-nav {\n          flex-direction: row;\n\n          .dropdown-menu {\n            position: absolute;\n          }\n\n          .nav-link {\n            padding-right: var(--#{$prefix}navbar-nav-link-padding-x);\n            padding-left: var(--#{$prefix}navbar-nav-link-padding-x);\n          }\n        }\n\n        .navbar-nav-scroll {\n          overflow: visible;\n        }\n\n        .navbar-collapse {\n          display: flex !important; // stylelint-disable-line declaration-no-important\n          flex-basis: auto;\n        }\n\n        .navbar-toggler {\n          display: none;\n        }\n\n        .offcanvas {\n          // stylelint-disable declaration-no-important\n          position: static;\n          z-index: auto;\n          flex-grow: 1;\n          width: auto !important;\n          height: auto !important;\n          visibility: visible !important;\n          background-color: transparent !important;\n          border: 0 !important;\n          transform: none !important;\n          @include box-shadow(none);\n          @include transition(none);\n          // stylelint-enable declaration-no-important\n\n          .offcanvas-header {\n            display: none;\n          }\n\n          .offcanvas-body {\n            display: flex;\n            flex-grow: 0;\n            padding: 0;\n            overflow-y: visible;\n          }\n        }\n      }\n    }\n  }\n}\n// scss-docs-end navbar-expand-loop\n\n// Navbar themes\n//\n// Styles for switching between navbars with light or dark background.\n\n.navbar-light {\n  @include deprecate(\"`.navbar-light`\", \"v5.2.0\", \"v6.0.0\", true);\n}\n\n.navbar-dark,\n.navbar[data-bs-theme=\"dark\"] {\n  // scss-docs-start navbar-dark-css-vars\n  --#{$prefix}navbar-color: #{$navbar-dark-color};\n  --#{$prefix}navbar-hover-color: #{$navbar-dark-hover-color};\n  --#{$prefix}navbar-disabled-color: #{$navbar-dark-disabled-color};\n  --#{$prefix}navbar-active-color: #{$navbar-dark-active-color};\n  --#{$prefix}navbar-brand-color: #{$navbar-dark-brand-color};\n  --#{$prefix}navbar-brand-hover-color: #{$navbar-dark-brand-hover-color};\n  --#{$prefix}navbar-toggler-border-color: #{$navbar-dark-toggler-border-color};\n  --#{$prefix}navbar-toggler-icon-bg: #{escape-svg($navbar-dark-toggler-icon-bg)};\n  // scss-docs-end navbar-dark-css-vars\n}\n\n@if $enable-dark-mode {\n  @include color-mode(dark) {\n    .navbar-toggler-icon {\n      --#{$prefix}navbar-toggler-icon-bg: #{escape-svg($navbar-dark-toggler-icon-bg)};\n    }\n  }\n}\n","//\n// Base styles\n//\n\n.card {\n  // scss-docs-start card-css-vars\n  --#{$prefix}card-spacer-y: #{$card-spacer-y};\n  --#{$prefix}card-spacer-x: #{$card-spacer-x};\n  --#{$prefix}card-title-spacer-y: #{$card-title-spacer-y};\n  --#{$prefix}card-title-color: #{$card-title-color};\n  --#{$prefix}card-subtitle-color: #{$card-subtitle-color};\n  --#{$prefix}card-border-width: #{$card-border-width};\n  --#{$prefix}card-border-color: #{$card-border-color};\n  --#{$prefix}card-border-radius: #{$card-border-radius};\n  --#{$prefix}card-box-shadow: #{$card-box-shadow};\n  --#{$prefix}card-inner-border-radius: #{$card-inner-border-radius};\n  --#{$prefix}card-cap-padding-y: #{$card-cap-padding-y};\n  --#{$prefix}card-cap-padding-x: #{$card-cap-padding-x};\n  --#{$prefix}card-cap-bg: #{$card-cap-bg};\n  --#{$prefix}card-cap-color: #{$card-cap-color};\n  --#{$prefix}card-height: #{$card-height};\n  --#{$prefix}card-color: #{$card-color};\n  --#{$prefix}card-bg: #{$card-bg};\n  --#{$prefix}card-img-overlay-padding: #{$card-img-overlay-padding};\n  --#{$prefix}card-group-margin: #{$card-group-margin};\n  // scss-docs-end card-css-vars\n\n  position: relative;\n  display: flex;\n  flex-direction: column;\n  min-width: 0; // See https://github.com/twbs/bootstrap/pull/22740#issuecomment-305868106\n  height: var(--#{$prefix}card-height);\n  color: var(--#{$prefix}body-color);\n  word-wrap: break-word;\n  background-color: var(--#{$prefix}card-bg);\n  background-clip: border-box;\n  border: var(--#{$prefix}card-border-width) solid var(--#{$prefix}card-border-color);\n  @include border-radius(var(--#{$prefix}card-border-radius));\n  @include box-shadow(var(--#{$prefix}card-box-shadow));\n\n  > hr {\n    margin-right: 0;\n    margin-left: 0;\n  }\n\n  > .list-group {\n    border-top: inherit;\n    border-bottom: inherit;\n\n    &:first-child {\n      border-top-width: 0;\n      @include border-top-radius(var(--#{$prefix}card-inner-border-radius));\n    }\n\n    &:last-child  {\n      border-bottom-width: 0;\n      @include border-bottom-radius(var(--#{$prefix}card-inner-border-radius));\n    }\n  }\n\n  // Due to specificity of the above selector (`.card > .list-group`), we must\n  // use a child selector here to prevent double borders.\n  > .card-header + .list-group,\n  > .list-group + .card-footer {\n    border-top: 0;\n  }\n}\n\n.card-body {\n  // Enable `flex-grow: 1` for decks and groups so that card blocks take up\n  // as much space as possible, ensuring footers are aligned to the bottom.\n  flex: 1 1 auto;\n  padding: var(--#{$prefix}card-spacer-y) var(--#{$prefix}card-spacer-x);\n  color: var(--#{$prefix}card-color);\n}\n\n.card-title {\n  margin-bottom: var(--#{$prefix}card-title-spacer-y);\n  color: var(--#{$prefix}card-title-color);\n}\n\n.card-subtitle {\n  margin-top: calc(-.5 * var(--#{$prefix}card-title-spacer-y)); // stylelint-disable-line function-disallowed-list\n  margin-bottom: 0;\n  color: var(--#{$prefix}card-subtitle-color);\n}\n\n.card-text:last-child {\n  margin-bottom: 0;\n}\n\n.card-link {\n  &:hover {\n    text-decoration: if($link-hover-decoration == underline, none, null);\n  }\n\n  + .card-link {\n    margin-left: var(--#{$prefix}card-spacer-x);\n  }\n}\n\n//\n// Optional textual caps\n//\n\n.card-header {\n  padding: var(--#{$prefix}card-cap-padding-y) var(--#{$prefix}card-cap-padding-x);\n  margin-bottom: 0; // Removes the default margin-bottom of <hN>\n  color: var(--#{$prefix}card-cap-color);\n  background-color: var(--#{$prefix}card-cap-bg);\n  border-bottom: var(--#{$prefix}card-border-width) solid var(--#{$prefix}card-border-color);\n\n  &:first-child {\n    @include border-radius(var(--#{$prefix}card-inner-border-radius) var(--#{$prefix}card-inner-border-radius) 0 0);\n  }\n}\n\n.card-footer {\n  padding: var(--#{$prefix}card-cap-padding-y) var(--#{$prefix}card-cap-padding-x);\n  color: var(--#{$prefix}card-cap-color);\n  background-color: var(--#{$prefix}card-cap-bg);\n  border-top: var(--#{$prefix}card-border-width) solid var(--#{$prefix}card-border-color);\n\n  &:last-child {\n    @include border-radius(0 0 var(--#{$prefix}card-inner-border-radius) var(--#{$prefix}card-inner-border-radius));\n  }\n}\n\n\n//\n// Header navs\n//\n\n.card-header-tabs {\n  margin-right: calc(-.5 * var(--#{$prefix}card-cap-padding-x)); // stylelint-disable-line function-disallowed-list\n  margin-bottom: calc(-1 * var(--#{$prefix}card-cap-padding-y)); // stylelint-disable-line function-disallowed-list\n  margin-left: calc(-.5 * var(--#{$prefix}card-cap-padding-x)); // stylelint-disable-line function-disallowed-list\n  border-bottom: 0;\n\n  .nav-link.active {\n    background-color: var(--#{$prefix}card-bg);\n    border-bottom-color: var(--#{$prefix}card-bg);\n  }\n}\n\n.card-header-pills {\n  margin-right: calc(-.5 * var(--#{$prefix}card-cap-padding-x)); // stylelint-disable-line function-disallowed-list\n  margin-left: calc(-.5 * var(--#{$prefix}card-cap-padding-x)); // stylelint-disable-line function-disallowed-list\n}\n\n// Card image\n.card-img-overlay {\n  position: absolute;\n  top: 0;\n  right: 0;\n  bottom: 0;\n  left: 0;\n  padding: var(--#{$prefix}card-img-overlay-padding);\n  @include border-radius(var(--#{$prefix}card-inner-border-radius));\n}\n\n.card-img,\n.card-img-top,\n.card-img-bottom {\n  width: 100%; // Required because we use flexbox and this inherently applies align-self: stretch\n}\n\n.card-img,\n.card-img-top {\n  @include border-top-radius(var(--#{$prefix}card-inner-border-radius));\n}\n\n.card-img,\n.card-img-bottom {\n  @include border-bottom-radius(var(--#{$prefix}card-inner-border-radius));\n}\n\n\n//\n// Card groups\n//\n\n.card-group {\n  // The child selector allows nested `.card` within `.card-group`\n  // to display properly.\n  > .card {\n    margin-bottom: var(--#{$prefix}card-group-margin);\n  }\n\n  @include media-breakpoint-up(sm) {\n    display: flex;\n    flex-flow: row wrap;\n    // The child selector allows nested `.card` within `.card-group`\n    // to display properly.\n    > .card {\n      // Flexbugs #4: https://github.com/philipwalton/flexbugs#flexbug-4\n      flex: 1 0 0%;\n      margin-bottom: 0;\n\n      + .card {\n        margin-left: 0;\n        border-left: 0;\n      }\n\n      // Handle rounded corners\n      @if $enable-rounded {\n        &:not(:last-child) {\n          @include border-end-radius(0);\n\n          .card-img-top,\n          .card-header {\n            // stylelint-disable-next-line property-disallowed-list\n            border-top-right-radius: 0;\n          }\n          .card-img-bottom,\n          .card-footer {\n            // stylelint-disable-next-line property-disallowed-list\n            border-bottom-right-radius: 0;\n          }\n        }\n\n        &:not(:first-child) {\n          @include border-start-radius(0);\n\n          .card-img-top,\n          .card-header {\n            // stylelint-disable-next-line property-disallowed-list\n            border-top-left-radius: 0;\n          }\n          .card-img-bottom,\n          .card-footer {\n            // stylelint-disable-next-line property-disallowed-list\n            border-bottom-left-radius: 0;\n          }\n        }\n      }\n    }\n  }\n}\n","//\n// Base styles\n//\n\n.accordion {\n  // scss-docs-start accordion-css-vars\n  --#{$prefix}accordion-color: #{$accordion-color};\n  --#{$prefix}accordion-bg: #{$accordion-bg};\n  --#{$prefix}accordion-transition: #{$accordion-transition};\n  --#{$prefix}accordion-border-color: #{$accordion-border-color};\n  --#{$prefix}accordion-border-width: #{$accordion-border-width};\n  --#{$prefix}accordion-border-radius: #{$accordion-border-radius};\n  --#{$prefix}accordion-inner-border-radius: #{$accordion-inner-border-radius};\n  --#{$prefix}accordion-btn-padding-x: #{$accordion-button-padding-x};\n  --#{$prefix}accordion-btn-padding-y: #{$accordion-button-padding-y};\n  --#{$prefix}accordion-btn-color: #{$accordion-button-color};\n  --#{$prefix}accordion-btn-bg: #{$accordion-button-bg};\n  --#{$prefix}accordion-btn-icon: #{escape-svg($accordion-button-icon)};\n  --#{$prefix}accordion-btn-icon-width: #{$accordion-icon-width};\n  --#{$prefix}accordion-btn-icon-transform: #{$accordion-icon-transform};\n  --#{$prefix}accordion-btn-icon-transition: #{$accordion-icon-transition};\n  --#{$prefix}accordion-btn-active-icon: #{escape-svg($accordion-button-active-icon)};\n  --#{$prefix}accordion-btn-focus-box-shadow: #{$accordion-button-focus-box-shadow};\n  --#{$prefix}accordion-body-padding-x: #{$accordion-body-padding-x};\n  --#{$prefix}accordion-body-padding-y: #{$accordion-body-padding-y};\n  --#{$prefix}accordion-active-color: #{$accordion-button-active-color};\n  --#{$prefix}accordion-active-bg: #{$accordion-button-active-bg};\n  // scss-docs-end accordion-css-vars\n}\n\n.accordion-button {\n  position: relative;\n  display: flex;\n  align-items: center;\n  width: 100%;\n  padding: var(--#{$prefix}accordion-btn-padding-y) var(--#{$prefix}accordion-btn-padding-x);\n  @include font-size($font-size-base);\n  color: var(--#{$prefix}accordion-btn-color);\n  text-align: left; // Reset button style\n  background-color: var(--#{$prefix}accordion-btn-bg);\n  border: 0;\n  @include border-radius(0);\n  overflow-anchor: none;\n  @include transition(var(--#{$prefix}accordion-transition));\n\n  &:not(.collapsed) {\n    color: var(--#{$prefix}accordion-active-color);\n    background-color: var(--#{$prefix}accordion-active-bg);\n    box-shadow: inset 0 calc(-1 * var(--#{$prefix}accordion-border-width)) 0 var(--#{$prefix}accordion-border-color); // stylelint-disable-line function-disallowed-list\n\n    &::after {\n      background-image: var(--#{$prefix}accordion-btn-active-icon);\n      transform: var(--#{$prefix}accordion-btn-icon-transform);\n    }\n  }\n\n  // Accordion icon\n  &::after {\n    flex-shrink: 0;\n    width: var(--#{$prefix}accordion-btn-icon-width);\n    height: var(--#{$prefix}accordion-btn-icon-width);\n    margin-left: auto;\n    content: \"\";\n    background-image: var(--#{$prefix}accordion-btn-icon);\n    background-repeat: no-repeat;\n    background-size: var(--#{$prefix}accordion-btn-icon-width);\n    @include transition(var(--#{$prefix}accordion-btn-icon-transition));\n  }\n\n  &:hover {\n    z-index: 2;\n  }\n\n  &:focus {\n    z-index: 3;\n    outline: 0;\n    box-shadow: var(--#{$prefix}accordion-btn-focus-box-shadow);\n  }\n}\n\n.accordion-header {\n  margin-bottom: 0;\n}\n\n.accordion-item {\n  color: var(--#{$prefix}accordion-color);\n  background-color: var(--#{$prefix}accordion-bg);\n  border: var(--#{$prefix}accordion-border-width) solid var(--#{$prefix}accordion-border-color);\n\n  &:first-of-type {\n    @include border-top-radius(var(--#{$prefix}accordion-border-radius));\n\n    > .accordion-header .accordion-button {\n      @include border-top-radius(var(--#{$prefix}accordion-inner-border-radius));\n    }\n  }\n\n  &:not(:first-of-type) {\n    border-top: 0;\n  }\n\n  // Only set a border-radius on the last item if the accordion is collapsed\n  &:last-of-type {\n    @include border-bottom-radius(var(--#{$prefix}accordion-border-radius));\n\n    > .accordion-header .accordion-button {\n      &.collapsed {\n        @include border-bottom-radius(var(--#{$prefix}accordion-inner-border-radius));\n      }\n    }\n\n    > .accordion-collapse {\n      @include border-bottom-radius(var(--#{$prefix}accordion-border-radius));\n    }\n  }\n}\n\n.accordion-body {\n  padding: var(--#{$prefix}accordion-body-padding-y) var(--#{$prefix}accordion-body-padding-x);\n}\n\n\n// Flush accordion items\n//\n// Remove borders and border-radius to keep accordion items edge-to-edge.\n\n.accordion-flush {\n  > .accordion-item {\n    border-right: 0;\n    border-left: 0;\n    @include border-radius(0);\n\n    &:first-child { border-top: 0; }\n    &:last-child { border-bottom: 0; }\n\n    // stylelint-disable selector-max-class\n    > .accordion-header .accordion-button {\n      &,\n      &.collapsed {\n        @include border-radius(0);\n      }\n    }\n    // stylelint-enable selector-max-class\n\n    > .accordion-collapse {\n      @include border-radius(0);\n    }\n  }\n}\n\n@if $enable-dark-mode {\n  @include color-mode(dark) {\n    .accordion-button::after {\n      --#{$prefix}accordion-btn-icon: #{escape-svg($accordion-button-icon-dark)};\n      --#{$prefix}accordion-btn-active-icon: #{escape-svg($accordion-button-active-icon-dark)};\n    }\n  }\n}\n",".breadcrumb {\n  // scss-docs-start breadcrumb-css-vars\n  --#{$prefix}breadcrumb-padding-x: #{$breadcrumb-padding-x};\n  --#{$prefix}breadcrumb-padding-y: #{$breadcrumb-padding-y};\n  --#{$prefix}breadcrumb-margin-bottom: #{$breadcrumb-margin-bottom};\n  @include rfs($breadcrumb-font-size, --#{$prefix}breadcrumb-font-size);\n  --#{$prefix}breadcrumb-bg: #{$breadcrumb-bg};\n  --#{$prefix}breadcrumb-border-radius: #{$breadcrumb-border-radius};\n  --#{$prefix}breadcrumb-divider-color: #{$breadcrumb-divider-color};\n  --#{$prefix}breadcrumb-item-padding-x: #{$breadcrumb-item-padding-x};\n  --#{$prefix}breadcrumb-item-active-color: #{$breadcrumb-active-color};\n  // scss-docs-end breadcrumb-css-vars\n\n  display: flex;\n  flex-wrap: wrap;\n  padding: var(--#{$prefix}breadcrumb-padding-y) var(--#{$prefix}breadcrumb-padding-x);\n  margin-bottom: var(--#{$prefix}breadcrumb-margin-bottom);\n  @include font-size(var(--#{$prefix}breadcrumb-font-size));\n  list-style: none;\n  background-color: var(--#{$prefix}breadcrumb-bg);\n  @include border-radius(var(--#{$prefix}breadcrumb-border-radius));\n}\n\n.breadcrumb-item {\n  // The separator between breadcrumbs (by default, a forward-slash: \"/\")\n  + .breadcrumb-item {\n    padding-left: var(--#{$prefix}breadcrumb-item-padding-x);\n\n    &::before {\n      float: left; // Suppress inline spacings and underlining of the separator\n      padding-right: var(--#{$prefix}breadcrumb-item-padding-x);\n      color: var(--#{$prefix}breadcrumb-divider-color);\n      content: var(--#{$prefix}breadcrumb-divider, escape-svg($breadcrumb-divider)) #{\"/* rtl:\"} var(--#{$prefix}breadcrumb-divider, escape-svg($breadcrumb-divider-flipped)) #{\"*/\"};\n    }\n  }\n\n  &.active {\n    color: var(--#{$prefix}breadcrumb-item-active-color);\n  }\n}\n",".pagination {\n  // scss-docs-start pagination-css-vars\n  --#{$prefix}pagination-padding-x: #{$pagination-padding-x};\n  --#{$prefix}pagination-padding-y: #{$pagination-padding-y};\n  @include rfs($pagination-font-size, --#{$prefix}pagination-font-size);\n  --#{$prefix}pagination-color: #{$pagination-color};\n  --#{$prefix}pagination-bg: #{$pagination-bg};\n  --#{$prefix}pagination-border-width: #{$pagination-border-width};\n  --#{$prefix}pagination-border-color: #{$pagination-border-color};\n  --#{$prefix}pagination-border-radius: #{$pagination-border-radius};\n  --#{$prefix}pagination-hover-color: #{$pagination-hover-color};\n  --#{$prefix}pagination-hover-bg: #{$pagination-hover-bg};\n  --#{$prefix}pagination-hover-border-color: #{$pagination-hover-border-color};\n  --#{$prefix}pagination-focus-color: #{$pagination-focus-color};\n  --#{$prefix}pagination-focus-bg: #{$pagination-focus-bg};\n  --#{$prefix}pagination-focus-box-shadow: #{$pagination-focus-box-shadow};\n  --#{$prefix}pagination-active-color: #{$pagination-active-color};\n  --#{$prefix}pagination-active-bg: #{$pagination-active-bg};\n  --#{$prefix}pagination-active-border-color: #{$pagination-active-border-color};\n  --#{$prefix}pagination-disabled-color: #{$pagination-disabled-color};\n  --#{$prefix}pagination-disabled-bg: #{$pagination-disabled-bg};\n  --#{$prefix}pagination-disabled-border-color: #{$pagination-disabled-border-color};\n  // scss-docs-end pagination-css-vars\n\n  display: flex;\n  @include list-unstyled();\n}\n\n.page-link {\n  position: relative;\n  display: block;\n  padding: var(--#{$prefix}pagination-padding-y) var(--#{$prefix}pagination-padding-x);\n  @include font-size(var(--#{$prefix}pagination-font-size));\n  color: var(--#{$prefix}pagination-color);\n  text-decoration: if($link-decoration == none, null, none);\n  background-color: var(--#{$prefix}pagination-bg);\n  border: var(--#{$prefix}pagination-border-width) solid var(--#{$prefix}pagination-border-color);\n  @include transition($pagination-transition);\n\n  &:hover {\n    z-index: 2;\n    color: var(--#{$prefix}pagination-hover-color);\n    text-decoration: if($link-hover-decoration == underline, none, null);\n    background-color: var(--#{$prefix}pagination-hover-bg);\n    border-color: var(--#{$prefix}pagination-hover-border-color);\n  }\n\n  &:focus {\n    z-index: 3;\n    color: var(--#{$prefix}pagination-focus-color);\n    background-color: var(--#{$prefix}pagination-focus-bg);\n    outline: $pagination-focus-outline;\n    box-shadow: var(--#{$prefix}pagination-focus-box-shadow);\n  }\n\n  &.active,\n  .active > & {\n    z-index: 3;\n    color: var(--#{$prefix}pagination-active-color);\n    @include gradient-bg(var(--#{$prefix}pagination-active-bg));\n    border-color: var(--#{$prefix}pagination-active-border-color);\n  }\n\n  &.disabled,\n  .disabled > & {\n    color: var(--#{$prefix}pagination-disabled-color);\n    pointer-events: none;\n    background-color: var(--#{$prefix}pagination-disabled-bg);\n    border-color: var(--#{$prefix}pagination-disabled-border-color);\n  }\n}\n\n.page-item {\n  &:not(:first-child) .page-link {\n    margin-left: $pagination-margin-start;\n  }\n\n  @if $pagination-margin-start == calc(#{$pagination-border-width} * -1) {\n    &:first-child {\n      .page-link {\n        @include border-start-radius(var(--#{$prefix}pagination-border-radius));\n      }\n    }\n\n    &:last-child {\n      .page-link {\n        @include border-end-radius(var(--#{$prefix}pagination-border-radius));\n      }\n    }\n  } @else {\n    // Add border-radius to all pageLinks in case they have left margin\n    .page-link {\n      @include border-radius(var(--#{$prefix}pagination-border-radius));\n    }\n  }\n}\n\n\n//\n// Sizing\n//\n\n.pagination-lg {\n  @include pagination-size($pagination-padding-y-lg, $pagination-padding-x-lg, $font-size-lg, $pagination-border-radius-lg);\n}\n\n.pagination-sm {\n  @include pagination-size($pagination-padding-y-sm, $pagination-padding-x-sm, $font-size-sm, $pagination-border-radius-sm);\n}\n","// Pagination\n\n// scss-docs-start pagination-mixin\n@mixin pagination-size($padding-y, $padding-x, $font-size, $border-radius) {\n  --#{$prefix}pagination-padding-x: #{$padding-x};\n  --#{$prefix}pagination-padding-y: #{$padding-y};\n  @include rfs($font-size, --#{$prefix}pagination-font-size);\n  --#{$prefix}pagination-border-radius: #{$border-radius};\n}\n// scss-docs-end pagination-mixin\n","// Base class\n//\n// Requires one of the contextual, color modifier classes for `color` and\n// `background-color`.\n\n.badge {\n  // scss-docs-start badge-css-vars\n  --#{$prefix}badge-padding-x: #{$badge-padding-x};\n  --#{$prefix}badge-padding-y: #{$badge-padding-y};\n  @include rfs($badge-font-size, --#{$prefix}badge-font-size);\n  --#{$prefix}badge-font-weight: #{$badge-font-weight};\n  --#{$prefix}badge-color: #{$badge-color};\n  --#{$prefix}badge-border-radius: #{$badge-border-radius};\n  // scss-docs-end badge-css-vars\n\n  display: inline-block;\n  padding: var(--#{$prefix}badge-padding-y) var(--#{$prefix}badge-padding-x);\n  @include font-size(var(--#{$prefix}badge-font-size));\n  font-weight: var(--#{$prefix}badge-font-weight);\n  line-height: 1;\n  color: var(--#{$prefix}badge-color);\n  text-align: center;\n  white-space: nowrap;\n  vertical-align: baseline;\n  @include border-radius(var(--#{$prefix}badge-border-radius));\n  @include gradient-bg();\n\n  // Empty badges collapse automatically\n  &:empty {\n    display: none;\n  }\n}\n\n// Quick fix for badges in buttons\n.btn .badge {\n  position: relative;\n  top: -1px;\n}\n","//\n// Base styles\n//\n\n.alert {\n  // scss-docs-start alert-css-vars\n  --#{$prefix}alert-bg: transparent;\n  --#{$prefix}alert-padding-x: #{$alert-padding-x};\n  --#{$prefix}alert-padding-y: #{$alert-padding-y};\n  --#{$prefix}alert-margin-bottom: #{$alert-margin-bottom};\n  --#{$prefix}alert-color: inherit;\n  --#{$prefix}alert-border-color: transparent;\n  --#{$prefix}alert-border: #{$alert-border-width} solid var(--#{$prefix}alert-border-color);\n  --#{$prefix}alert-border-radius: #{$alert-border-radius};\n  --#{$prefix}alert-link-color: inherit;\n  // scss-docs-end alert-css-vars\n\n  position: relative;\n  padding: var(--#{$prefix}alert-padding-y) var(--#{$prefix}alert-padding-x);\n  margin-bottom: var(--#{$prefix}alert-margin-bottom);\n  color: var(--#{$prefix}alert-color);\n  background-color: var(--#{$prefix}alert-bg);\n  border: var(--#{$prefix}alert-border);\n  @include border-radius(var(--#{$prefix}alert-border-radius));\n}\n\n// Headings for larger alerts\n.alert-heading {\n  // Specified to prevent conflicts of changing $headings-color\n  color: inherit;\n}\n\n// Provide class for links that match alerts\n.alert-link {\n  font-weight: $alert-link-font-weight;\n  color: var(--#{$prefix}alert-link-color);\n}\n\n\n// Dismissible alerts\n//\n// Expand the right padding and account for the close button's positioning.\n\n.alert-dismissible {\n  padding-right: $alert-dismissible-padding-r;\n\n  // Adjust close link position\n  .btn-close {\n    position: absolute;\n    top: 0;\n    right: 0;\n    z-index: $stretched-link-z-index + 1;\n    padding: $alert-padding-y * 1.25 $alert-padding-x;\n  }\n}\n\n\n// scss-docs-start alert-modifiers\n// Generate contextual modifier classes for colorizing the alert\n@each $state in map-keys($theme-colors) {\n  .alert-#{$state} {\n    --#{$prefix}alert-color: var(--#{$prefix}#{$state}-text-emphasis);\n    --#{$prefix}alert-bg: var(--#{$prefix}#{$state}-bg-subtle);\n    --#{$prefix}alert-border-color: var(--#{$prefix}#{$state}-border-subtle);\n    --#{$prefix}alert-link-color: var(--#{$prefix}#{$state}-text-emphasis);\n  }\n}\n// scss-docs-end alert-modifiers\n","// Disable animation if transitions are disabled\n\n// scss-docs-start progress-keyframes\n@if $enable-transitions {\n  @keyframes progress-bar-stripes {\n    0% { background-position-x: $progress-height; }\n  }\n}\n// scss-docs-end progress-keyframes\n\n.progress,\n.progress-stacked {\n  // scss-docs-start progress-css-vars\n  --#{$prefix}progress-height: #{$progress-height};\n  @include rfs($progress-font-size, --#{$prefix}progress-font-size);\n  --#{$prefix}progress-bg: #{$progress-bg};\n  --#{$prefix}progress-border-radius: #{$progress-border-radius};\n  --#{$prefix}progress-box-shadow: #{$progress-box-shadow};\n  --#{$prefix}progress-bar-color: #{$progress-bar-color};\n  --#{$prefix}progress-bar-bg: #{$progress-bar-bg};\n  --#{$prefix}progress-bar-transition: #{$progress-bar-transition};\n  // scss-docs-end progress-css-vars\n\n  display: flex;\n  height: var(--#{$prefix}progress-height);\n  overflow: hidden; // force rounded corners by cropping it\n  @include font-size(var(--#{$prefix}progress-font-size));\n  background-color: var(--#{$prefix}progress-bg);\n  @include border-radius(var(--#{$prefix}progress-border-radius));\n  @include box-shadow(var(--#{$prefix}progress-box-shadow));\n}\n\n.progress-bar {\n  display: flex;\n  flex-direction: column;\n  justify-content: center;\n  overflow: hidden;\n  color: var(--#{$prefix}progress-bar-color);\n  text-align: center;\n  white-space: nowrap;\n  background-color: var(--#{$prefix}progress-bar-bg);\n  @include transition(var(--#{$prefix}progress-bar-transition));\n}\n\n.progress-bar-striped {\n  @include gradient-striped();\n  background-size: var(--#{$prefix}progress-height) var(--#{$prefix}progress-height);\n}\n\n.progress-stacked > .progress {\n  overflow: visible;\n}\n\n.progress-stacked > .progress > .progress-bar {\n  width: 100%;\n}\n\n@if $enable-transitions {\n  .progress-bar-animated {\n    animation: $progress-bar-animation-timing progress-bar-stripes;\n\n    @if $enable-reduced-motion {\n      @media (prefers-reduced-motion: reduce) {\n        animation: none;\n      }\n    }\n  }\n}\n","// Base class\n//\n// Easily usable on <ul>, <ol>, or <div>.\n\n.list-group {\n  // scss-docs-start list-group-css-vars\n  --#{$prefix}list-group-color: #{$list-group-color};\n  --#{$prefix}list-group-bg: #{$list-group-bg};\n  --#{$prefix}list-group-border-color: #{$list-group-border-color};\n  --#{$prefix}list-group-border-width: #{$list-group-border-width};\n  --#{$prefix}list-group-border-radius: #{$list-group-border-radius};\n  --#{$prefix}list-group-item-padding-x: #{$list-group-item-padding-x};\n  --#{$prefix}list-group-item-padding-y: #{$list-group-item-padding-y};\n  --#{$prefix}list-group-action-color: #{$list-group-action-color};\n  --#{$prefix}list-group-action-hover-color: #{$list-group-action-hover-color};\n  --#{$prefix}list-group-action-hover-bg: #{$list-group-hover-bg};\n  --#{$prefix}list-group-action-active-color: #{$list-group-action-active-color};\n  --#{$prefix}list-group-action-active-bg: #{$list-group-action-active-bg};\n  --#{$prefix}list-group-disabled-color: #{$list-group-disabled-color};\n  --#{$prefix}list-group-disabled-bg: #{$list-group-disabled-bg};\n  --#{$prefix}list-group-active-color: #{$list-group-active-color};\n  --#{$prefix}list-group-active-bg: #{$list-group-active-bg};\n  --#{$prefix}list-group-active-border-color: #{$list-group-active-border-color};\n  // scss-docs-end list-group-css-vars\n\n  display: flex;\n  flex-direction: column;\n\n  // No need to set list-style: none; since .list-group-item is block level\n  padding-left: 0; // reset padding because ul and ol\n  margin-bottom: 0;\n  @include border-radius(var(--#{$prefix}list-group-border-radius));\n}\n\n.list-group-numbered {\n  list-style-type: none;\n  counter-reset: section;\n\n  > .list-group-item::before {\n    // Increments only this instance of the section counter\n    content: counters(section, \".\") \". \";\n    counter-increment: section;\n  }\n}\n\n// Interactive list items\n//\n// Use anchor or button elements instead of `li`s or `div`s to create interactive\n// list items. Includes an extra `.active` modifier class for selected items.\n\n.list-group-item-action {\n  width: 100%; // For `<button>`s (anchors become 100% by default though)\n  color: var(--#{$prefix}list-group-action-color);\n  text-align: inherit; // For `<button>`s (anchors inherit)\n\n  // Hover state\n  &:hover,\n  &:focus {\n    z-index: 1; // Place hover/focus items above their siblings for proper border styling\n    color: var(--#{$prefix}list-group-action-hover-color);\n    text-decoration: none;\n    background-color: var(--#{$prefix}list-group-action-hover-bg);\n  }\n\n  &:active {\n    color: var(--#{$prefix}list-group-action-active-color);\n    background-color: var(--#{$prefix}list-group-action-active-bg);\n  }\n}\n\n// Individual list items\n//\n// Use on `li`s or `div`s within the `.list-group` parent.\n\n.list-group-item {\n  position: relative;\n  display: block;\n  padding: var(--#{$prefix}list-group-item-padding-y) var(--#{$prefix}list-group-item-padding-x);\n  color: var(--#{$prefix}list-group-color);\n  text-decoration: if($link-decoration == none, null, none);\n  background-color: var(--#{$prefix}list-group-bg);\n  border: var(--#{$prefix}list-group-border-width) solid var(--#{$prefix}list-group-border-color);\n\n  &:first-child {\n    @include border-top-radius(inherit);\n  }\n\n  &:last-child {\n    @include border-bottom-radius(inherit);\n  }\n\n  &.disabled,\n  &:disabled {\n    color: var(--#{$prefix}list-group-disabled-color);\n    pointer-events: none;\n    background-color: var(--#{$prefix}list-group-disabled-bg);\n  }\n\n  // Include both here for `<a>`s and `<button>`s\n  &.active {\n    z-index: 2; // Place active items above their siblings for proper border styling\n    color: var(--#{$prefix}list-group-active-color);\n    background-color: var(--#{$prefix}list-group-active-bg);\n    border-color: var(--#{$prefix}list-group-active-border-color);\n  }\n\n  // stylelint-disable-next-line scss/selector-no-redundant-nesting-selector\n  & + .list-group-item {\n    border-top-width: 0;\n\n    &.active {\n      margin-top: calc(-1 * var(--#{$prefix}list-group-border-width)); // stylelint-disable-line function-disallowed-list\n      border-top-width: var(--#{$prefix}list-group-border-width);\n    }\n  }\n}\n\n// Horizontal\n//\n// Change the layout of list group items from vertical (default) to horizontal.\n\n@each $breakpoint in map-keys($grid-breakpoints) {\n  @include media-breakpoint-up($breakpoint) {\n    $infix: breakpoint-infix($breakpoint, $grid-breakpoints);\n\n    .list-group-horizontal#{$infix} {\n      flex-direction: row;\n\n      > .list-group-item {\n        &:first-child:not(:last-child) {\n          @include border-bottom-start-radius(var(--#{$prefix}list-group-border-radius));\n          @include border-top-end-radius(0);\n        }\n\n        &:last-child:not(:first-child) {\n          @include border-top-end-radius(var(--#{$prefix}list-group-border-radius));\n          @include border-bottom-start-radius(0);\n        }\n\n        &.active {\n          margin-top: 0;\n        }\n\n        + .list-group-item {\n          border-top-width: var(--#{$prefix}list-group-border-width);\n          border-left-width: 0;\n\n          &.active {\n            margin-left: calc(-1 * var(--#{$prefix}list-group-border-width)); // stylelint-disable-line function-disallowed-list\n            border-left-width: var(--#{$prefix}list-group-border-width);\n          }\n        }\n      }\n    }\n  }\n}\n\n\n// Flush list items\n//\n// Remove borders and border-radius to keep list group items edge-to-edge. Most\n// useful within other components (e.g., cards).\n\n.list-group-flush {\n  @include border-radius(0);\n\n  > .list-group-item {\n    border-width: 0 0 var(--#{$prefix}list-group-border-width);\n\n    &:last-child {\n      border-bottom-width: 0;\n    }\n  }\n}\n\n\n// scss-docs-start list-group-modifiers\n// List group contextual variants\n//\n// Add modifier classes to change text and background color on individual items.\n// Organizationally, this must come after the `:hover` states.\n\n@each $state in map-keys($theme-colors) {\n  .list-group-item-#{$state} {\n    --#{$prefix}list-group-color: var(--#{$prefix}#{$state}-text-emphasis);\n    --#{$prefix}list-group-bg: var(--#{$prefix}#{$state}-bg-subtle);\n    --#{$prefix}list-group-border-color: var(--#{$prefix}#{$state}-border-subtle);\n    --#{$prefix}list-group-action-hover-color: var(--#{$prefix}emphasis-color);\n    --#{$prefix}list-group-action-hover-bg: var(--#{$prefix}#{$state}-border-subtle);\n    --#{$prefix}list-group-action-active-color: var(--#{$prefix}emphasis-color);\n    --#{$prefix}list-group-action-active-bg: var(--#{$prefix}#{$state}-border-subtle);\n    --#{$prefix}list-group-active-color: var(--#{$prefix}#{$state}-bg-subtle);\n    --#{$prefix}list-group-active-bg: var(--#{$prefix}#{$state}-text-emphasis);\n    --#{$prefix}list-group-active-border-color: var(--#{$prefix}#{$state}-text-emphasis);\n  }\n}\n// scss-docs-end list-group-modifiers\n","// Transparent background and border properties included for button version.\n// iOS requires the button element instead of an anchor tag.\n// If you want the anchor version, it requires `href=\"#\"`.\n// See https://developer.mozilla.org/en-US/docs/Web/Events/click#Safari_Mobile\n\n.btn-close {\n  // scss-docs-start close-css-vars\n  --#{$prefix}btn-close-color: #{$btn-close-color};\n  --#{$prefix}btn-close-bg: #{ escape-svg($btn-close-bg) };\n  --#{$prefix}btn-close-opacity: #{$btn-close-opacity};\n  --#{$prefix}btn-close-hover-opacity: #{$btn-close-hover-opacity};\n  --#{$prefix}btn-close-focus-shadow: #{$btn-close-focus-shadow};\n  --#{$prefix}btn-close-focus-opacity: #{$btn-close-focus-opacity};\n  --#{$prefix}btn-close-disabled-opacity: #{$btn-close-disabled-opacity};\n  --#{$prefix}btn-close-white-filter: #{$btn-close-white-filter};\n  // scss-docs-end close-css-vars\n\n  box-sizing: content-box;\n  width: $btn-close-width;\n  height: $btn-close-height;\n  padding: $btn-close-padding-y $btn-close-padding-x;\n  color: var(--#{$prefix}btn-close-color);\n  background: transparent var(--#{$prefix}btn-close-bg) center / $btn-close-width auto no-repeat; // include transparent for button elements\n  border: 0; // for button elements\n  @include border-radius();\n  opacity: var(--#{$prefix}btn-close-opacity);\n\n  // Override <a>'s hover style\n  &:hover {\n    color: var(--#{$prefix}btn-close-color);\n    text-decoration: none;\n    opacity: var(--#{$prefix}btn-close-hover-opacity);\n  }\n\n  &:focus {\n    outline: 0;\n    box-shadow: var(--#{$prefix}btn-close-focus-shadow);\n    opacity: var(--#{$prefix}btn-close-focus-opacity);\n  }\n\n  &:disabled,\n  &.disabled {\n    pointer-events: none;\n    user-select: none;\n    opacity: var(--#{$prefix}btn-close-disabled-opacity);\n  }\n}\n\n@mixin btn-close-white() {\n  filter: var(--#{$prefix}btn-close-white-filter);\n}\n\n.btn-close-white {\n  @include btn-close-white();\n}\n\n@if $enable-dark-mode {\n  @include color-mode(dark) {\n    .btn-close {\n      @include btn-close-white();\n    }\n  }\n}\n",".toast {\n  // scss-docs-start toast-css-vars\n  --#{$prefix}toast-zindex: #{$zindex-toast};\n  --#{$prefix}toast-padding-x: #{$toast-padding-x};\n  --#{$prefix}toast-padding-y: #{$toast-padding-y};\n  --#{$prefix}toast-spacing: #{$toast-spacing};\n  --#{$prefix}toast-max-width: #{$toast-max-width};\n  @include rfs($toast-font-size, --#{$prefix}toast-font-size);\n  --#{$prefix}toast-color: #{$toast-color};\n  --#{$prefix}toast-bg: #{$toast-background-color};\n  --#{$prefix}toast-border-width: #{$toast-border-width};\n  --#{$prefix}toast-border-color: #{$toast-border-color};\n  --#{$prefix}toast-border-radius: #{$toast-border-radius};\n  --#{$prefix}toast-box-shadow: #{$toast-box-shadow};\n  --#{$prefix}toast-header-color: #{$toast-header-color};\n  --#{$prefix}toast-header-bg: #{$toast-header-background-color};\n  --#{$prefix}toast-header-border-color: #{$toast-header-border-color};\n  // scss-docs-end toast-css-vars\n\n  width: var(--#{$prefix}toast-max-width);\n  max-width: 100%;\n  @include font-size(var(--#{$prefix}toast-font-size));\n  color: var(--#{$prefix}toast-color);\n  pointer-events: auto;\n  background-color: var(--#{$prefix}toast-bg);\n  background-clip: padding-box;\n  border: var(--#{$prefix}toast-border-width) solid var(--#{$prefix}toast-border-color);\n  box-shadow: var(--#{$prefix}toast-box-shadow);\n  @include border-radius(var(--#{$prefix}toast-border-radius));\n\n  &.showing {\n    opacity: 0;\n  }\n\n  &:not(.show) {\n    display: none;\n  }\n}\n\n.toast-container {\n  --#{$prefix}toast-zindex: #{$zindex-toast};\n\n  position: absolute;\n  z-index: var(--#{$prefix}toast-zindex);\n  width: max-content;\n  max-width: 100%;\n  pointer-events: none;\n\n  > :not(:last-child) {\n    margin-bottom: var(--#{$prefix}toast-spacing);\n  }\n}\n\n.toast-header {\n  display: flex;\n  align-items: center;\n  padding: var(--#{$prefix}toast-padding-y) var(--#{$prefix}toast-padding-x);\n  color: var(--#{$prefix}toast-header-color);\n  background-color: var(--#{$prefix}toast-header-bg);\n  background-clip: padding-box;\n  border-bottom: var(--#{$prefix}toast-border-width) solid var(--#{$prefix}toast-header-border-color);\n  @include border-top-radius(calc(var(--#{$prefix}toast-border-radius) - var(--#{$prefix}toast-border-width)));\n\n  .btn-close {\n    margin-right: calc(-.5 * var(--#{$prefix}toast-padding-x)); // stylelint-disable-line function-disallowed-list\n    margin-left: var(--#{$prefix}toast-padding-x);\n  }\n}\n\n.toast-body {\n  padding: var(--#{$prefix}toast-padding-x);\n  word-wrap: break-word;\n}\n","// stylelint-disable function-disallowed-list\n\n// .modal-open      - body class for killing the scroll\n// .modal           - container to scroll within\n// .modal-dialog    - positioning shell for the actual modal\n// .modal-content   - actual modal w/ bg and corners and stuff\n\n\n// Container that the modal scrolls within\n.modal {\n  // scss-docs-start modal-css-vars\n  --#{$prefix}modal-zindex: #{$zindex-modal};\n  --#{$prefix}modal-width: #{$modal-md};\n  --#{$prefix}modal-padding: #{$modal-inner-padding};\n  --#{$prefix}modal-margin: #{$modal-dialog-margin};\n  --#{$prefix}modal-color: #{$modal-content-color};\n  --#{$prefix}modal-bg: #{$modal-content-bg};\n  --#{$prefix}modal-border-color: #{$modal-content-border-color};\n  --#{$prefix}modal-border-width: #{$modal-content-border-width};\n  --#{$prefix}modal-border-radius: #{$modal-content-border-radius};\n  --#{$prefix}modal-box-shadow: #{$modal-content-box-shadow-xs};\n  --#{$prefix}modal-inner-border-radius: #{$modal-content-inner-border-radius};\n  --#{$prefix}modal-header-padding-x: #{$modal-header-padding-x};\n  --#{$prefix}modal-header-padding-y: #{$modal-header-padding-y};\n  --#{$prefix}modal-header-padding: #{$modal-header-padding}; // Todo in v6: Split this padding into x and y\n  --#{$prefix}modal-header-border-color: #{$modal-header-border-color};\n  --#{$prefix}modal-header-border-width: #{$modal-header-border-width};\n  --#{$prefix}modal-title-line-height: #{$modal-title-line-height};\n  --#{$prefix}modal-footer-gap: #{$modal-footer-margin-between};\n  --#{$prefix}modal-footer-bg: #{$modal-footer-bg};\n  --#{$prefix}modal-footer-border-color: #{$modal-footer-border-color};\n  --#{$prefix}modal-footer-border-width: #{$modal-footer-border-width};\n  // scss-docs-end modal-css-vars\n\n  position: fixed;\n  top: 0;\n  left: 0;\n  z-index: var(--#{$prefix}modal-zindex);\n  display: none;\n  width: 100%;\n  height: 100%;\n  overflow-x: hidden;\n  overflow-y: auto;\n  // Prevent Chrome on Windows from adding a focus outline. For details, see\n  // https://github.com/twbs/bootstrap/pull/10951.\n  outline: 0;\n  // We deliberately don't use `-webkit-overflow-scrolling: touch;` due to a\n  // gnarly iOS Safari bug: https://bugs.webkit.org/show_bug.cgi?id=158342\n  // See also https://github.com/twbs/bootstrap/issues/17695\n}\n\n// Shell div to position the modal with bottom padding\n.modal-dialog {\n  position: relative;\n  width: auto;\n  margin: var(--#{$prefix}modal-margin);\n  // allow clicks to pass through for custom click handling to close modal\n  pointer-events: none;\n\n  // When fading in the modal, animate it to slide down\n  .modal.fade & {\n    @include transition($modal-transition);\n    transform: $modal-fade-transform;\n  }\n  .modal.show & {\n    transform: $modal-show-transform;\n  }\n\n  // When trying to close, animate focus to scale\n  .modal.modal-static & {\n    transform: $modal-scale-transform;\n  }\n}\n\n.modal-dialog-scrollable {\n  height: calc(100% - var(--#{$prefix}modal-margin) * 2);\n\n  .modal-content {\n    max-height: 100%;\n    overflow: hidden;\n  }\n\n  .modal-body {\n    overflow-y: auto;\n  }\n}\n\n.modal-dialog-centered {\n  display: flex;\n  align-items: center;\n  min-height: calc(100% - var(--#{$prefix}modal-margin) * 2);\n}\n\n// Actual modal\n.modal-content {\n  position: relative;\n  display: flex;\n  flex-direction: column;\n  width: 100%; // Ensure `.modal-content` extends the full width of the parent `.modal-dialog`\n  // counteract the pointer-events: none; in the .modal-dialog\n  color: var(--#{$prefix}modal-color);\n  pointer-events: auto;\n  background-color: var(--#{$prefix}modal-bg);\n  background-clip: padding-box;\n  border: var(--#{$prefix}modal-border-width) solid var(--#{$prefix}modal-border-color);\n  @include border-radius(var(--#{$prefix}modal-border-radius));\n  @include box-shadow(var(--#{$prefix}modal-box-shadow));\n  // Remove focus outline from opened modal\n  outline: 0;\n}\n\n// Modal background\n.modal-backdrop {\n  // scss-docs-start modal-backdrop-css-vars\n  --#{$prefix}backdrop-zindex: #{$zindex-modal-backdrop};\n  --#{$prefix}backdrop-bg: #{$modal-backdrop-bg};\n  --#{$prefix}backdrop-opacity: #{$modal-backdrop-opacity};\n  // scss-docs-end modal-backdrop-css-vars\n\n  @include overlay-backdrop(var(--#{$prefix}backdrop-zindex), var(--#{$prefix}backdrop-bg), var(--#{$prefix}backdrop-opacity));\n}\n\n// Modal header\n// Top section of the modal w/ title and dismiss\n.modal-header {\n  display: flex;\n  flex-shrink: 0;\n  align-items: center;\n  padding: var(--#{$prefix}modal-header-padding);\n  border-bottom: var(--#{$prefix}modal-header-border-width) solid var(--#{$prefix}modal-header-border-color);\n  @include border-top-radius(var(--#{$prefix}modal-inner-border-radius));\n\n  .btn-close {\n    padding: calc(var(--#{$prefix}modal-header-padding-y) * .5) calc(var(--#{$prefix}modal-header-padding-x) * .5);\n    margin: calc(-.5 * var(--#{$prefix}modal-header-padding-y)) calc(-.5 * var(--#{$prefix}modal-header-padding-x)) calc(-.5 * var(--#{$prefix}modal-header-padding-y)) auto;\n  }\n}\n\n// Title text within header\n.modal-title {\n  margin-bottom: 0;\n  line-height: var(--#{$prefix}modal-title-line-height);\n}\n\n// Modal body\n// Where all modal content resides (sibling of .modal-header and .modal-footer)\n.modal-body {\n  position: relative;\n  // Enable `flex-grow: 1` so that the body take up as much space as possible\n  // when there should be a fixed height on `.modal-dialog`.\n  flex: 1 1 auto;\n  padding: var(--#{$prefix}modal-padding);\n}\n\n// Footer (for actions)\n.modal-footer {\n  display: flex;\n  flex-shrink: 0;\n  flex-wrap: wrap;\n  align-items: center; // vertically center\n  justify-content: flex-end; // Right align buttons with flex property because text-align doesn't work on flex items\n  padding: calc(var(--#{$prefix}modal-padding) - var(--#{$prefix}modal-footer-gap) * .5);\n  background-color: var(--#{$prefix}modal-footer-bg);\n  border-top: var(--#{$prefix}modal-footer-border-width) solid var(--#{$prefix}modal-footer-border-color);\n  @include border-bottom-radius(var(--#{$prefix}modal-inner-border-radius));\n\n  // Place margin between footer elements\n  // This solution is far from ideal because of the universal selector usage,\n  // but is needed to fix https://github.com/twbs/bootstrap/issues/24800\n  > * {\n    margin: calc(var(--#{$prefix}modal-footer-gap) * .5); // Todo in v6: replace with gap on parent class\n  }\n}\n\n// Scale up the modal\n@include media-breakpoint-up(sm) {\n  .modal {\n    --#{$prefix}modal-margin: #{$modal-dialog-margin-y-sm-up};\n    --#{$prefix}modal-box-shadow: #{$modal-content-box-shadow-sm-up};\n  }\n\n  // Automatically set modal's width for larger viewports\n  .modal-dialog {\n    max-width: var(--#{$prefix}modal-width);\n    margin-right: auto;\n    margin-left: auto;\n  }\n\n  .modal-sm {\n    --#{$prefix}modal-width: #{$modal-sm};\n  }\n}\n\n@include media-breakpoint-up(lg) {\n  .modal-lg,\n  .modal-xl {\n    --#{$prefix}modal-width: #{$modal-lg};\n  }\n}\n\n@include media-breakpoint-up(xl) {\n  .modal-xl {\n    --#{$prefix}modal-width: #{$modal-xl};\n  }\n}\n\n// scss-docs-start modal-fullscreen-loop\n@each $breakpoint in map-keys($grid-breakpoints) {\n  $infix: breakpoint-infix($breakpoint, $grid-breakpoints);\n  $postfix: if($infix != \"\", $infix + \"-down\", \"\");\n\n  @include media-breakpoint-down($breakpoint) {\n    .modal-fullscreen#{$postfix} {\n      width: 100vw;\n      max-width: none;\n      height: 100%;\n      margin: 0;\n\n      .modal-content {\n        height: 100%;\n        border: 0;\n        @include border-radius(0);\n      }\n\n      .modal-header,\n      .modal-footer {\n        @include border-radius(0);\n      }\n\n      .modal-body {\n        overflow-y: auto;\n      }\n    }\n  }\n}\n// scss-docs-end modal-fullscreen-loop\n","// Shared between modals and offcanvases\n@mixin overlay-backdrop($zindex, $backdrop-bg, $backdrop-opacity) {\n  position: fixed;\n  top: 0;\n  left: 0;\n  z-index: $zindex;\n  width: 100vw;\n  height: 100vh;\n  background-color: $backdrop-bg;\n\n  // Fade for backdrop\n  &.fade { opacity: 0; }\n  &.show { opacity: $backdrop-opacity; }\n}\n","// Base class\n.tooltip {\n  // scss-docs-start tooltip-css-vars\n  --#{$prefix}tooltip-zindex: #{$zindex-tooltip};\n  --#{$prefix}tooltip-max-width: #{$tooltip-max-width};\n  --#{$prefix}tooltip-padding-x: #{$tooltip-padding-x};\n  --#{$prefix}tooltip-padding-y: #{$tooltip-padding-y};\n  --#{$prefix}tooltip-margin: #{$tooltip-margin};\n  @include rfs($tooltip-font-size, --#{$prefix}tooltip-font-size);\n  --#{$prefix}tooltip-color: #{$tooltip-color};\n  --#{$prefix}tooltip-bg: #{$tooltip-bg};\n  --#{$prefix}tooltip-border-radius: #{$tooltip-border-radius};\n  --#{$prefix}tooltip-opacity: #{$tooltip-opacity};\n  --#{$prefix}tooltip-arrow-width: #{$tooltip-arrow-width};\n  --#{$prefix}tooltip-arrow-height: #{$tooltip-arrow-height};\n  // scss-docs-end tooltip-css-vars\n\n  z-index: var(--#{$prefix}tooltip-zindex);\n  display: block;\n  margin: var(--#{$prefix}tooltip-margin);\n  @include deprecate(\"`$tooltip-margin`\", \"v5\", \"v5.x\", true);\n  // Our parent element can be arbitrary since tooltips are by default inserted as a sibling of their target element.\n  // So reset our font and text properties to avoid inheriting weird values.\n  @include reset-text();\n  @include font-size(var(--#{$prefix}tooltip-font-size));\n  // Allow breaking very long words so they don't overflow the tooltip's bounds\n  word-wrap: break-word;\n  opacity: 0;\n\n  &.show { opacity: var(--#{$prefix}tooltip-opacity); }\n\n  .tooltip-arrow {\n    display: block;\n    width: var(--#{$prefix}tooltip-arrow-width);\n    height: var(--#{$prefix}tooltip-arrow-height);\n\n    &::before {\n      position: absolute;\n      content: \"\";\n      border-color: transparent;\n      border-style: solid;\n    }\n  }\n}\n\n.bs-tooltip-top .tooltip-arrow {\n  bottom: calc(-1 * var(--#{$prefix}tooltip-arrow-height)); // stylelint-disable-line function-disallowed-list\n\n  &::before {\n    top: -1px;\n    border-width: var(--#{$prefix}tooltip-arrow-height) calc(var(--#{$prefix}tooltip-arrow-width) * .5) 0; // stylelint-disable-line function-disallowed-list\n    border-top-color: var(--#{$prefix}tooltip-bg);\n  }\n}\n\n/* rtl:begin:ignore */\n.bs-tooltip-end .tooltip-arrow {\n  left: calc(-1 * var(--#{$prefix}tooltip-arrow-height)); // stylelint-disable-line function-disallowed-list\n  width: var(--#{$prefix}tooltip-arrow-height);\n  height: var(--#{$prefix}tooltip-arrow-width);\n\n  &::before {\n    right: -1px;\n    border-width: calc(var(--#{$prefix}tooltip-arrow-width) * .5) var(--#{$prefix}tooltip-arrow-height) calc(var(--#{$prefix}tooltip-arrow-width) * .5) 0; // stylelint-disable-line function-disallowed-list\n    border-right-color: var(--#{$prefix}tooltip-bg);\n  }\n}\n\n/* rtl:end:ignore */\n\n.bs-tooltip-bottom .tooltip-arrow {\n  top: calc(-1 * var(--#{$prefix}tooltip-arrow-height)); // stylelint-disable-line function-disallowed-list\n\n  &::before {\n    bottom: -1px;\n    border-width: 0 calc(var(--#{$prefix}tooltip-arrow-width) * .5) var(--#{$prefix}tooltip-arrow-height); // stylelint-disable-line function-disallowed-list\n    border-bottom-color: var(--#{$prefix}tooltip-bg);\n  }\n}\n\n/* rtl:begin:ignore */\n.bs-tooltip-start .tooltip-arrow {\n  right: calc(-1 * var(--#{$prefix}tooltip-arrow-height)); // stylelint-disable-line function-disallowed-list\n  width: var(--#{$prefix}tooltip-arrow-height);\n  height: var(--#{$prefix}tooltip-arrow-width);\n\n  &::before {\n    left: -1px;\n    border-width: calc(var(--#{$prefix}tooltip-arrow-width) * .5) 0 calc(var(--#{$prefix}tooltip-arrow-width) * .5) var(--#{$prefix}tooltip-arrow-height); // stylelint-disable-line function-disallowed-list\n    border-left-color: var(--#{$prefix}tooltip-bg);\n  }\n}\n\n/* rtl:end:ignore */\n\n.bs-tooltip-auto {\n  &[data-popper-placement^=\"top\"] {\n    @extend .bs-tooltip-top;\n  }\n  &[data-popper-placement^=\"right\"] {\n    @extend .bs-tooltip-end;\n  }\n  &[data-popper-placement^=\"bottom\"] {\n    @extend .bs-tooltip-bottom;\n  }\n  &[data-popper-placement^=\"left\"] {\n    @extend .bs-tooltip-start;\n  }\n}\n\n// Wrapper for the tooltip content\n.tooltip-inner {\n  max-width: var(--#{$prefix}tooltip-max-width);\n  padding: var(--#{$prefix}tooltip-padding-y) var(--#{$prefix}tooltip-padding-x);\n  color: var(--#{$prefix}tooltip-color);\n  text-align: center;\n  background-color: var(--#{$prefix}tooltip-bg);\n  @include border-radius(var(--#{$prefix}tooltip-border-radius));\n}\n","@mixin reset-text {\n  font-family: $font-family-base;\n  // We deliberately do NOT reset font-size or overflow-wrap / word-wrap.\n  font-style: normal;\n  font-weight: $font-weight-normal;\n  line-height: $line-height-base;\n  text-align: left; // Fallback for where `start` is not supported\n  text-align: start;\n  text-decoration: none;\n  text-shadow: none;\n  text-transform: none;\n  letter-spacing: normal;\n  word-break: normal;\n  white-space: normal;\n  word-spacing: normal;\n  line-break: auto;\n}\n",".popover {\n  // scss-docs-start popover-css-vars\n  --#{$prefix}popover-zindex: #{$zindex-popover};\n  --#{$prefix}popover-max-width: #{$popover-max-width};\n  @include rfs($popover-font-size, --#{$prefix}popover-font-size);\n  --#{$prefix}popover-bg: #{$popover-bg};\n  --#{$prefix}popover-border-width: #{$popover-border-width};\n  --#{$prefix}popover-border-color: #{$popover-border-color};\n  --#{$prefix}popover-border-radius: #{$popover-border-radius};\n  --#{$prefix}popover-inner-border-radius: #{$popover-inner-border-radius};\n  --#{$prefix}popover-box-shadow: #{$popover-box-shadow};\n  --#{$prefix}popover-header-padding-x: #{$popover-header-padding-x};\n  --#{$prefix}popover-header-padding-y: #{$popover-header-padding-y};\n  @include rfs($popover-header-font-size, --#{$prefix}popover-header-font-size);\n  --#{$prefix}popover-header-color: #{$popover-header-color};\n  --#{$prefix}popover-header-bg: #{$popover-header-bg};\n  --#{$prefix}popover-body-padding-x: #{$popover-body-padding-x};\n  --#{$prefix}popover-body-padding-y: #{$popover-body-padding-y};\n  --#{$prefix}popover-body-color: #{$popover-body-color};\n  --#{$prefix}popover-arrow-width: #{$popover-arrow-width};\n  --#{$prefix}popover-arrow-height: #{$popover-arrow-height};\n  --#{$prefix}popover-arrow-border: var(--#{$prefix}popover-border-color);\n  // scss-docs-end popover-css-vars\n\n  z-index: var(--#{$prefix}popover-zindex);\n  display: block;\n  max-width: var(--#{$prefix}popover-max-width);\n  // Our parent element can be arbitrary since tooltips are by default inserted as a sibling of their target element.\n  // So reset our font and text properties to avoid inheriting weird values.\n  @include reset-text();\n  @include font-size(var(--#{$prefix}popover-font-size));\n  // Allow breaking very long words so they don't overflow the popover's bounds\n  word-wrap: break-word;\n  background-color: var(--#{$prefix}popover-bg);\n  background-clip: padding-box;\n  border: var(--#{$prefix}popover-border-width) solid var(--#{$prefix}popover-border-color);\n  @include border-radius(var(--#{$prefix}popover-border-radius));\n  @include box-shadow(var(--#{$prefix}popover-box-shadow));\n\n  .popover-arrow {\n    display: block;\n    width: var(--#{$prefix}popover-arrow-width);\n    height: var(--#{$prefix}popover-arrow-height);\n\n    &::before,\n    &::after {\n      position: absolute;\n      display: block;\n      content: \"\";\n      border-color: transparent;\n      border-style: solid;\n      border-width: 0;\n    }\n  }\n}\n\n.bs-popover-top {\n  > .popover-arrow {\n    bottom: calc(-1 * (var(--#{$prefix}popover-arrow-height)) - var(--#{$prefix}popover-border-width)); // stylelint-disable-line function-disallowed-list\n\n    &::before,\n    &::after {\n      border-width: var(--#{$prefix}popover-arrow-height) calc(var(--#{$prefix}popover-arrow-width) * .5) 0; // stylelint-disable-line function-disallowed-list\n    }\n\n    &::before {\n      bottom: 0;\n      border-top-color: var(--#{$prefix}popover-arrow-border);\n    }\n\n    &::after {\n      bottom: var(--#{$prefix}popover-border-width);\n      border-top-color: var(--#{$prefix}popover-bg);\n    }\n  }\n}\n\n/* rtl:begin:ignore */\n.bs-popover-end {\n  > .popover-arrow {\n    left: calc(-1 * (var(--#{$prefix}popover-arrow-height)) - var(--#{$prefix}popover-border-width)); // stylelint-disable-line function-disallowed-list\n    width: var(--#{$prefix}popover-arrow-height);\n    height: var(--#{$prefix}popover-arrow-width);\n\n    &::before,\n    &::after {\n      border-width: calc(var(--#{$prefix}popover-arrow-width) * .5) var(--#{$prefix}popover-arrow-height) calc(var(--#{$prefix}popover-arrow-width) * .5) 0; // stylelint-disable-line function-disallowed-list\n    }\n\n    &::before {\n      left: 0;\n      border-right-color: var(--#{$prefix}popover-arrow-border);\n    }\n\n    &::after {\n      left: var(--#{$prefix}popover-border-width);\n      border-right-color: var(--#{$prefix}popover-bg);\n    }\n  }\n}\n\n/* rtl:end:ignore */\n\n.bs-popover-bottom {\n  > .popover-arrow {\n    top: calc(-1 * (var(--#{$prefix}popover-arrow-height)) - var(--#{$prefix}popover-border-width)); // stylelint-disable-line function-disallowed-list\n\n    &::before,\n    &::after {\n      border-width: 0 calc(var(--#{$prefix}popover-arrow-width) * .5) var(--#{$prefix}popover-arrow-height); // stylelint-disable-line function-disallowed-list\n    }\n\n    &::before {\n      top: 0;\n      border-bottom-color: var(--#{$prefix}popover-arrow-border);\n    }\n\n    &::after {\n      top: var(--#{$prefix}popover-border-width);\n      border-bottom-color: var(--#{$prefix}popover-bg);\n    }\n  }\n\n  // This will remove the popover-header's border just below the arrow\n  .popover-header::before {\n    position: absolute;\n    top: 0;\n    left: 50%;\n    display: block;\n    width: var(--#{$prefix}popover-arrow-width);\n    margin-left: calc(-.5 * var(--#{$prefix}popover-arrow-width)); // stylelint-disable-line function-disallowed-list\n    content: \"\";\n    border-bottom: var(--#{$prefix}popover-border-width) solid var(--#{$prefix}popover-header-bg);\n  }\n}\n\n/* rtl:begin:ignore */\n.bs-popover-start {\n  > .popover-arrow {\n    right: calc(-1 * (var(--#{$prefix}popover-arrow-height)) - var(--#{$prefix}popover-border-width)); // stylelint-disable-line function-disallowed-list\n    width: var(--#{$prefix}popover-arrow-height);\n    height: var(--#{$prefix}popover-arrow-width);\n\n    &::before,\n    &::after {\n      border-width: calc(var(--#{$prefix}popover-arrow-width) * .5) 0 calc(var(--#{$prefix}popover-arrow-width) * .5) var(--#{$prefix}popover-arrow-height); // stylelint-disable-line function-disallowed-list\n    }\n\n    &::before {\n      right: 0;\n      border-left-color: var(--#{$prefix}popover-arrow-border);\n    }\n\n    &::after {\n      right: var(--#{$prefix}popover-border-width);\n      border-left-color: var(--#{$prefix}popover-bg);\n    }\n  }\n}\n\n/* rtl:end:ignore */\n\n.bs-popover-auto {\n  &[data-popper-placement^=\"top\"] {\n    @extend .bs-popover-top;\n  }\n  &[data-popper-placement^=\"right\"] {\n    @extend .bs-popover-end;\n  }\n  &[data-popper-placement^=\"bottom\"] {\n    @extend .bs-popover-bottom;\n  }\n  &[data-popper-placement^=\"left\"] {\n    @extend .bs-popover-start;\n  }\n}\n\n// Offset the popover to account for the popover arrow\n.popover-header {\n  padding: var(--#{$prefix}popover-header-padding-y) var(--#{$prefix}popover-header-padding-x);\n  margin-bottom: 0; // Reset the default from Reboot\n  @include font-size(var(--#{$prefix}popover-header-font-size));\n  color: var(--#{$prefix}popover-header-color);\n  background-color: var(--#{$prefix}popover-header-bg);\n  border-bottom: var(--#{$prefix}popover-border-width) solid var(--#{$prefix}popover-border-color);\n  @include border-top-radius(var(--#{$prefix}popover-inner-border-radius));\n\n  &:empty {\n    display: none;\n  }\n}\n\n.popover-body {\n  padding: var(--#{$prefix}popover-body-padding-y) var(--#{$prefix}popover-body-padding-x);\n  color: var(--#{$prefix}popover-body-color);\n}\n","// Notes on the classes:\n//\n// 1. .carousel.pointer-event should ideally be pan-y (to allow for users to scroll vertically)\n//    even when their scroll action started on a carousel, but for compatibility (with Firefox)\n//    we're preventing all actions instead\n// 2. The .carousel-item-start and .carousel-item-end is used to indicate where\n//    the active slide is heading.\n// 3. .active.carousel-item is the current slide.\n// 4. .active.carousel-item-start and .active.carousel-item-end is the current\n//    slide in its in-transition state. Only one of these occurs at a time.\n// 5. .carousel-item-next.carousel-item-start and .carousel-item-prev.carousel-item-end\n//    is the upcoming slide in transition.\n\n.carousel {\n  position: relative;\n}\n\n.carousel.pointer-event {\n  touch-action: pan-y;\n}\n\n.carousel-inner {\n  position: relative;\n  width: 100%;\n  overflow: hidden;\n  @include clearfix();\n}\n\n.carousel-item {\n  position: relative;\n  display: none;\n  float: left;\n  width: 100%;\n  margin-right: -100%;\n  backface-visibility: hidden;\n  @include transition($carousel-transition);\n}\n\n.carousel-item.active,\n.carousel-item-next,\n.carousel-item-prev {\n  display: block;\n}\n\n.carousel-item-next:not(.carousel-item-start),\n.active.carousel-item-end {\n  transform: translateX(100%);\n}\n\n.carousel-item-prev:not(.carousel-item-end),\n.active.carousel-item-start {\n  transform: translateX(-100%);\n}\n\n\n//\n// Alternate transitions\n//\n\n.carousel-fade {\n  .carousel-item {\n    opacity: 0;\n    transition-property: opacity;\n    transform: none;\n  }\n\n  .carousel-item.active,\n  .carousel-item-next.carousel-item-start,\n  .carousel-item-prev.carousel-item-end {\n    z-index: 1;\n    opacity: 1;\n  }\n\n  .active.carousel-item-start,\n  .active.carousel-item-end {\n    z-index: 0;\n    opacity: 0;\n    @include transition(opacity 0s $carousel-transition-duration);\n  }\n}\n\n\n//\n// Left/right controls for nav\n//\n\n.carousel-control-prev,\n.carousel-control-next {\n  position: absolute;\n  top: 0;\n  bottom: 0;\n  z-index: 1;\n  // Use flex for alignment (1-3)\n  display: flex; // 1. allow flex styles\n  align-items: center; // 2. vertically center contents\n  justify-content: center; // 3. horizontally center contents\n  width: $carousel-control-width;\n  padding: 0;\n  color: $carousel-control-color;\n  text-align: center;\n  background: none;\n  border: 0;\n  opacity: $carousel-control-opacity;\n  @include transition($carousel-control-transition);\n\n  // Hover/focus state\n  &:hover,\n  &:focus {\n    color: $carousel-control-color;\n    text-decoration: none;\n    outline: 0;\n    opacity: $carousel-control-hover-opacity;\n  }\n}\n.carousel-control-prev {\n  left: 0;\n  background-image: if($enable-gradients, linear-gradient(90deg, rgba($black, .25), rgba($black, .001)), null);\n}\n.carousel-control-next {\n  right: 0;\n  background-image: if($enable-gradients, linear-gradient(270deg, rgba($black, .25), rgba($black, .001)), null);\n}\n\n// Icons for within\n.carousel-control-prev-icon,\n.carousel-control-next-icon {\n  display: inline-block;\n  width: $carousel-control-icon-width;\n  height: $carousel-control-icon-width;\n  background-repeat: no-repeat;\n  background-position: 50%;\n  background-size: 100% 100%;\n}\n\n.carousel-control-prev-icon {\n  background-image: escape-svg($carousel-control-prev-icon-bg) #{\"/*rtl:\" + escape-svg($carousel-control-next-icon-bg) + \"*/\"};\n}\n.carousel-control-next-icon {\n  background-image: escape-svg($carousel-control-next-icon-bg) #{\"/*rtl:\" + escape-svg($carousel-control-prev-icon-bg) + \"*/\"};\n}\n\n// Optional indicator pips/controls\n//\n// Add a container (such as a list) with the following class and add an item (ideally a focusable control,\n// like a button) with data-bs-target for each slide your carousel holds.\n\n.carousel-indicators {\n  position: absolute;\n  right: 0;\n  bottom: 0;\n  left: 0;\n  z-index: 2;\n  display: flex;\n  justify-content: center;\n  padding: 0;\n  // Use the .carousel-control's width as margin so we don't overlay those\n  margin-right: $carousel-control-width;\n  margin-bottom: 1rem;\n  margin-left: $carousel-control-width;\n\n  [data-bs-target] {\n    box-sizing: content-box;\n    flex: 0 1 auto;\n    width: $carousel-indicator-width;\n    height: $carousel-indicator-height;\n    padding: 0;\n    margin-right: $carousel-indicator-spacer;\n    margin-left: $carousel-indicator-spacer;\n    text-indent: -999px;\n    cursor: pointer;\n    background-color: $carousel-indicator-active-bg;\n    background-clip: padding-box;\n    border: 0;\n    // Use transparent borders to increase the hit area by 10px on top and bottom.\n    border-top: $carousel-indicator-hit-area-height solid transparent;\n    border-bottom: $carousel-indicator-hit-area-height solid transparent;\n    opacity: $carousel-indicator-opacity;\n    @include transition($carousel-indicator-transition);\n  }\n\n  .active {\n    opacity: $carousel-indicator-active-opacity;\n  }\n}\n\n\n// Optional captions\n//\n//\n\n.carousel-caption {\n  position: absolute;\n  right: (100% - $carousel-caption-width) * .5;\n  bottom: $carousel-caption-spacer;\n  left: (100% - $carousel-caption-width) * .5;\n  padding-top: $carousel-caption-padding-y;\n  padding-bottom: $carousel-caption-padding-y;\n  color: $carousel-caption-color;\n  text-align: center;\n}\n\n// Dark mode carousel\n\n@mixin carousel-dark() {\n  .carousel-control-prev-icon,\n  .carousel-control-next-icon {\n    filter: $carousel-dark-control-icon-filter;\n  }\n\n  .carousel-indicators [data-bs-target] {\n    background-color: $carousel-dark-indicator-active-bg;\n  }\n\n  .carousel-caption {\n    color: $carousel-dark-caption-color;\n  }\n}\n\n.carousel-dark {\n  @include carousel-dark();\n}\n\n@if $enable-dark-mode {\n  @include color-mode(dark) {\n    @if $color-mode-type == \"media-query\" {\n      .carousel {\n        @include carousel-dark();\n      }\n    } @else {\n      .carousel,\n      &.carousel {\n        @include carousel-dark();\n      }\n    }\n  }\n}\n","// scss-docs-start clearfix\n@mixin clearfix() {\n  &::after {\n    display: block;\n    clear: both;\n    content: \"\";\n  }\n}\n// scss-docs-end clearfix\n","//\n// Rotating border\n//\n\n.spinner-grow,\n.spinner-border {\n  display: inline-block;\n  width: var(--#{$prefix}spinner-width);\n  height: var(--#{$prefix}spinner-height);\n  vertical-align: var(--#{$prefix}spinner-vertical-align);\n  // stylelint-disable-next-line property-disallowed-list\n  border-radius: 50%;\n  animation: var(--#{$prefix}spinner-animation-speed) linear infinite var(--#{$prefix}spinner-animation-name);\n}\n\n// scss-docs-start spinner-border-keyframes\n@keyframes spinner-border {\n  to { transform: rotate(360deg) #{\"/* rtl:ignore */\"}; }\n}\n// scss-docs-end spinner-border-keyframes\n\n.spinner-border {\n  // scss-docs-start spinner-border-css-vars\n  --#{$prefix}spinner-width: #{$spinner-width};\n  --#{$prefix}spinner-height: #{$spinner-height};\n  --#{$prefix}spinner-vertical-align: #{$spinner-vertical-align};\n  --#{$prefix}spinner-border-width: #{$spinner-border-width};\n  --#{$prefix}spinner-animation-speed: #{$spinner-animation-speed};\n  --#{$prefix}spinner-animation-name: spinner-border;\n  // scss-docs-end spinner-border-css-vars\n\n  border: var(--#{$prefix}spinner-border-width) solid currentcolor;\n  border-right-color: transparent;\n}\n\n.spinner-border-sm {\n  // scss-docs-start spinner-border-sm-css-vars\n  --#{$prefix}spinner-width: #{$spinner-width-sm};\n  --#{$prefix}spinner-height: #{$spinner-height-sm};\n  --#{$prefix}spinner-border-width: #{$spinner-border-width-sm};\n  // scss-docs-end spinner-border-sm-css-vars\n}\n\n//\n// Growing circle\n//\n\n// scss-docs-start spinner-grow-keyframes\n@keyframes spinner-grow {\n  0% {\n    transform: scale(0);\n  }\n  50% {\n    opacity: 1;\n    transform: none;\n  }\n}\n// scss-docs-end spinner-grow-keyframes\n\n.spinner-grow {\n  // scss-docs-start spinner-grow-css-vars\n  --#{$prefix}spinner-width: #{$spinner-width};\n  --#{$prefix}spinner-height: #{$spinner-height};\n  --#{$prefix}spinner-vertical-align: #{$spinner-vertical-align};\n  --#{$prefix}spinner-animation-speed: #{$spinner-animation-speed};\n  --#{$prefix}spinner-animation-name: spinner-grow;\n  // scss-docs-end spinner-grow-css-vars\n\n  background-color: currentcolor;\n  opacity: 0;\n}\n\n.spinner-grow-sm {\n  --#{$prefix}spinner-width: #{$spinner-width-sm};\n  --#{$prefix}spinner-height: #{$spinner-height-sm};\n}\n\n@if $enable-reduced-motion {\n  @media (prefers-reduced-motion: reduce) {\n    .spinner-border,\n    .spinner-grow {\n      --#{$prefix}spinner-animation-speed: #{$spinner-animation-speed * 2};\n    }\n  }\n}\n","// stylelint-disable function-disallowed-list\n\n%offcanvas-css-vars {\n  // scss-docs-start offcanvas-css-vars\n  --#{$prefix}offcanvas-zindex: #{$zindex-offcanvas};\n  --#{$prefix}offcanvas-width: #{$offcanvas-horizontal-width};\n  --#{$prefix}offcanvas-height: #{$offcanvas-vertical-height};\n  --#{$prefix}offcanvas-padding-x: #{$offcanvas-padding-x};\n  --#{$prefix}offcanvas-padding-y: #{$offcanvas-padding-y};\n  --#{$prefix}offcanvas-color: #{$offcanvas-color};\n  --#{$prefix}offcanvas-bg: #{$offcanvas-bg-color};\n  --#{$prefix}offcanvas-border-width: #{$offcanvas-border-width};\n  --#{$prefix}offcanvas-border-color: #{$offcanvas-border-color};\n  --#{$prefix}offcanvas-box-shadow: #{$offcanvas-box-shadow};\n  --#{$prefix}offcanvas-transition: #{transform $offcanvas-transition-duration ease-in-out};\n  --#{$prefix}offcanvas-title-line-height: #{$offcanvas-title-line-height};\n  // scss-docs-end offcanvas-css-vars\n}\n\n@each $breakpoint in map-keys($grid-breakpoints) {\n  $next: breakpoint-next($breakpoint, $grid-breakpoints);\n  $infix: breakpoint-infix($next, $grid-breakpoints);\n\n  .offcanvas#{$infix} {\n    @extend %offcanvas-css-vars;\n  }\n}\n\n@each $breakpoint in map-keys($grid-breakpoints) {\n  $next: breakpoint-next($breakpoint, $grid-breakpoints);\n  $infix: breakpoint-infix($next, $grid-breakpoints);\n\n  .offcanvas#{$infix} {\n    @include media-breakpoint-down($next) {\n      position: fixed;\n      bottom: 0;\n      z-index: var(--#{$prefix}offcanvas-zindex);\n      display: flex;\n      flex-direction: column;\n      max-width: 100%;\n      color: var(--#{$prefix}offcanvas-color);\n      visibility: hidden;\n      background-color: var(--#{$prefix}offcanvas-bg);\n      background-clip: padding-box;\n      outline: 0;\n      @include box-shadow(var(--#{$prefix}offcanvas-box-shadow));\n      @include transition(var(--#{$prefix}offcanvas-transition));\n\n      &.offcanvas-start {\n        top: 0;\n        left: 0;\n        width: var(--#{$prefix}offcanvas-width);\n        border-right: var(--#{$prefix}offcanvas-border-width) solid var(--#{$prefix}offcanvas-border-color);\n        transform: translateX(-100%);\n      }\n\n      &.offcanvas-end {\n        top: 0;\n        right: 0;\n        width: var(--#{$prefix}offcanvas-width);\n        border-left: var(--#{$prefix}offcanvas-border-width) solid var(--#{$prefix}offcanvas-border-color);\n        transform: translateX(100%);\n      }\n\n      &.offcanvas-top {\n        top: 0;\n        right: 0;\n        left: 0;\n        height: var(--#{$prefix}offcanvas-height);\n        max-height: 100%;\n        border-bottom: var(--#{$prefix}offcanvas-border-width) solid var(--#{$prefix}offcanvas-border-color);\n        transform: translateY(-100%);\n      }\n\n      &.offcanvas-bottom {\n        right: 0;\n        left: 0;\n        height: var(--#{$prefix}offcanvas-height);\n        max-height: 100%;\n        border-top: var(--#{$prefix}offcanvas-border-width) solid var(--#{$prefix}offcanvas-border-color);\n        transform: translateY(100%);\n      }\n\n      &.showing,\n      &.show:not(.hiding) {\n        transform: none;\n      }\n\n      &.showing,\n      &.hiding,\n      &.show {\n        visibility: visible;\n      }\n    }\n\n    @if not ($infix == \"\") {\n      @include media-breakpoint-up($next) {\n        --#{$prefix}offcanvas-height: auto;\n        --#{$prefix}offcanvas-border-width: 0;\n        background-color: transparent !important; // stylelint-disable-line declaration-no-important\n\n        .offcanvas-header {\n          display: none;\n        }\n\n        .offcanvas-body {\n          display: flex;\n          flex-grow: 0;\n          padding: 0;\n          overflow-y: visible;\n          // Reset `background-color` in case `.bg-*` classes are used in offcanvas\n          background-color: transparent !important; // stylelint-disable-line declaration-no-important\n        }\n      }\n    }\n  }\n}\n\n.offcanvas-backdrop {\n  @include overlay-backdrop($zindex-offcanvas-backdrop, $offcanvas-backdrop-bg, $offcanvas-backdrop-opacity);\n}\n\n.offcanvas-header {\n  display: flex;\n  align-items: center;\n  padding: var(--#{$prefix}offcanvas-padding-y) var(--#{$prefix}offcanvas-padding-x);\n\n  .btn-close {\n    padding: calc(var(--#{$prefix}offcanvas-padding-y) * .5) calc(var(--#{$prefix}offcanvas-padding-x) * .5);\n    margin: calc(-.5 * var(--#{$prefix}offcanvas-padding-y)) calc(-.5 * var(--#{$prefix}offcanvas-padding-x)) calc(-.5 * var(--#{$prefix}offcanvas-padding-y)) auto;\n  }\n}\n\n.offcanvas-title {\n  margin-bottom: 0;\n  line-height: var(--#{$prefix}offcanvas-title-line-height);\n}\n\n.offcanvas-body {\n  flex-grow: 1;\n  padding: var(--#{$prefix}offcanvas-padding-y) var(--#{$prefix}offcanvas-padding-x);\n  overflow-y: auto;\n}\n",".placeholder {\n  display: inline-block;\n  min-height: 1em;\n  vertical-align: middle;\n  cursor: wait;\n  background-color: currentcolor;\n  opacity: $placeholder-opacity-max;\n\n  &.btn::before {\n    display: inline-block;\n    content: \"\";\n  }\n}\n\n// Sizing\n.placeholder-xs {\n  min-height: .6em;\n}\n\n.placeholder-sm {\n  min-height: .8em;\n}\n\n.placeholder-lg {\n  min-height: 1.2em;\n}\n\n// Animation\n.placeholder-glow {\n  .placeholder {\n    animation: placeholder-glow 2s ease-in-out infinite;\n  }\n}\n\n@keyframes placeholder-glow {\n  50% {\n    opacity: $placeholder-opacity-min;\n  }\n}\n\n.placeholder-wave {\n  mask-image: linear-gradient(130deg, $black 55%, rgba(0, 0, 0, (1 - $placeholder-opacity-min)) 75%, $black 95%);\n  mask-size: 200% 100%;\n  animation: placeholder-wave 2s linear infinite;\n}\n\n@keyframes placeholder-wave {\n  100% {\n    mask-position: -200% 0%;\n  }\n}\n","// All-caps `RGBA()` function used because of this Sass bug: https://github.com/sass/node-sass/issues/2251\n@each $color, $value in $theme-colors {\n  .text-bg-#{$color} {\n    color: color-contrast($value) if($enable-important-utilities, !important, null);\n    background-color: RGBA(var(--#{$prefix}#{$color}-rgb), var(--#{$prefix}bg-opacity, 1)) if($enable-important-utilities, !important, null);\n  }\n}\n","// All-caps `RGBA()` function used because of this Sass bug: https://github.com/sass/node-sass/issues/2251\n@each $color, $value in $theme-colors {\n  .link-#{$color} {\n    color: RGBA(var(--#{$prefix}#{$color}-rgb), var(--#{$prefix}link-opacity, 1)) if($enable-important-utilities, !important, null);\n    text-decoration-color: RGBA(var(--#{$prefix}#{$color}-rgb), var(--#{$prefix}link-underline-opacity, 1)) if($enable-important-utilities, !important, null);\n\n    @if $link-shade-percentage != 0 {\n      &:hover,\n      &:focus {\n        $hover-color: if(color-contrast($value) == $color-contrast-light, shade-color($value, $link-shade-percentage), tint-color($value, $link-shade-percentage));\n        color: RGBA(#{to-rgb($hover-color)}, var(--#{$prefix}link-opacity, 1)) if($enable-important-utilities, !important, null);\n        text-decoration-color: RGBA(to-rgb($hover-color), var(--#{$prefix}link-underline-opacity, 1)) if($enable-important-utilities, !important, null);\n      }\n    }\n  }\n}\n\n// One-off special link helper as a bridge until v6\n.link-body-emphasis {\n  color: RGBA(var(--#{$prefix}emphasis-color-rgb), var(--#{$prefix}link-opacity, 1)) if($enable-important-utilities, !important, null);\n  text-decoration-color: RGBA(var(--#{$prefix}emphasis-color-rgb), var(--#{$prefix}link-underline-opacity, 1)) if($enable-important-utilities, !important, null);\n\n  @if $link-shade-percentage != 0 {\n    &:hover,\n    &:focus {\n      color: RGBA(var(--#{$prefix}emphasis-color-rgb), var(--#{$prefix}link-opacity, .75)) if($enable-important-utilities, !important, null);\n      text-decoration-color: RGBA(var(--#{$prefix}emphasis-color-rgb), var(--#{$prefix}link-underline-opacity, .75)) if($enable-important-utilities, !important, null);\n    }\n  }\n}\n",".focus-ring:focus {\n  outline: 0;\n  // By default, there is no `--bs-focus-ring-x`, `--bs-focus-ring-y`, or `--bs-focus-ring-blur`, but we provide CSS variables with fallbacks to initial `0` values\n  box-shadow: var(--#{$prefix}focus-ring-x, 0) var(--#{$prefix}focus-ring-y, 0) var(--#{$prefix}focus-ring-blur, 0) var(--#{$prefix}focus-ring-width) var(--#{$prefix}focus-ring-color);\n}\n",".icon-link {\n  display: inline-flex;\n  gap: $icon-link-gap;\n  align-items: center;\n  text-decoration-color: rgba(var(--#{$prefix}link-color-rgb), var(--#{$prefix}link-opacity, .5));\n  text-underline-offset: $icon-link-underline-offset;\n  backface-visibility: hidden;\n\n  > .bi {\n    flex-shrink: 0;\n    width: $icon-link-icon-size;\n    height: $icon-link-icon-size;\n    fill: currentcolor;\n    @include transition($icon-link-icon-transition);\n  }\n}\n\n.icon-link-hover {\n  &:hover,\n  &:focus-visible {\n    > .bi {\n      transform: var(--#{$prefix}icon-link-transform, $icon-link-icon-transform);\n    }\n  }\n}\n","// Credit: Nicolas Gallagher and SUIT CSS.\n\n.ratio {\n  position: relative;\n  width: 100%;\n\n  &::before {\n    display: block;\n    padding-top: var(--#{$prefix}aspect-ratio);\n    content: \"\";\n  }\n\n  > * {\n    position: absolute;\n    top: 0;\n    left: 0;\n    width: 100%;\n    height: 100%;\n  }\n}\n\n@each $key, $ratio in $aspect-ratios {\n  .ratio-#{$key} {\n    --#{$prefix}aspect-ratio: #{$ratio};\n  }\n}\n","// Shorthand\n\n.fixed-top {\n  position: fixed;\n  top: 0;\n  right: 0;\n  left: 0;\n  z-index: $zindex-fixed;\n}\n\n.fixed-bottom {\n  position: fixed;\n  right: 0;\n  bottom: 0;\n  left: 0;\n  z-index: $zindex-fixed;\n}\n\n// Responsive sticky top and bottom\n@each $breakpoint in map-keys($grid-breakpoints) {\n  @include media-breakpoint-up($breakpoint) {\n    $infix: breakpoint-infix($breakpoint, $grid-breakpoints);\n\n    .sticky#{$infix}-top {\n      position: sticky;\n      top: 0;\n      z-index: $zindex-sticky;\n    }\n\n    .sticky#{$infix}-bottom {\n      position: sticky;\n      bottom: 0;\n      z-index: $zindex-sticky;\n    }\n  }\n}\n","// scss-docs-start stacks\n.hstack {\n  display: flex;\n  flex-direction: row;\n  align-items: center;\n  align-self: stretch;\n}\n\n.vstack {\n  display: flex;\n  flex: 1 1 auto;\n  flex-direction: column;\n  align-self: stretch;\n}\n// scss-docs-end stacks\n","//\n// Visually hidden\n//\n\n.visually-hidden,\n.visually-hidden-focusable:not(:focus):not(:focus-within) {\n  @include visually-hidden();\n}\n","// stylelint-disable declaration-no-important\n\n// Hide content visually while keeping it accessible to assistive technologies\n//\n// See: https://www.a11yproject.com/posts/2013-01-11-how-to-hide-content/\n// See: https://kittygiraudel.com/2016/10/13/css-hide-and-seek/\n\n@mixin visually-hidden() {\n  width: 1px !important;\n  height: 1px !important;\n  padding: 0 !important;\n  margin: -1px !important; // Fix for https://github.com/twbs/bootstrap/issues/25686\n  overflow: hidden !important;\n  clip: rect(0, 0, 0, 0) !important;\n  white-space: nowrap !important;\n  border: 0 !important;\n\n  // Fix for positioned table caption that could become anonymous cells\n  &:not(caption) {\n    position: absolute !important;\n  }\n}\n\n// Use to only display content when it's focused, or one of its child elements is focused\n// (i.e. when focus is within the element/container that the class was applied to)\n//\n// Useful for \"Skip to main content\" links; see https://www.w3.org/TR/2013/NOTE-WCAG20-TECHS-20130905/G1\n\n@mixin visually-hidden-focusable() {\n  &:not(:focus):not(:focus-within) {\n    @include visually-hidden();\n  }\n}\n","//\n// Stretched link\n//\n\n.stretched-link {\n  &::#{$stretched-link-pseudo-element} {\n    position: absolute;\n    top: 0;\n    right: 0;\n    bottom: 0;\n    left: 0;\n    z-index: $stretched-link-z-index;\n    content: \"\";\n  }\n}\n","//\n// Text truncation\n//\n\n.text-truncate {\n  @include text-truncate();\n}\n","// Text truncate\n// Requires inline-block or block for proper styling\n\n@mixin text-truncate() {\n  overflow: hidden;\n  text-overflow: ellipsis;\n  white-space: nowrap;\n}\n",".vr {\n  display: inline-block;\n  align-self: stretch;\n  width: $vr-border-width;\n  min-height: 1em;\n  background-color: currentcolor;\n  opacity: $hr-opacity;\n}\n","// Utility generator\n// Used to generate utilities & print utilities\n@mixin generate-utility($utility, $infix: \"\", $is-rfs-media-query: false) {\n  $values: map-get($utility, values);\n\n  // If the values are a list or string, convert it into a map\n  @if type-of($values) == \"string\" or type-of(nth($values, 1)) != \"list\" {\n    $values: zip($values, $values);\n  }\n\n  @each $key, $value in $values {\n    $properties: map-get($utility, property);\n\n    // Multiple properties are possible, for example with vertical or horizontal margins or paddings\n    @if type-of($properties) == \"string\" {\n      $properties: append((), $properties);\n    }\n\n    // Use custom class if present\n    $property-class: if(map-has-key($utility, class), map-get($utility, class), nth($properties, 1));\n    $property-class: if($property-class == null, \"\", $property-class);\n\n    // Use custom CSS variable name if present, otherwise default to `class`\n    $css-variable-name: if(map-has-key($utility, css-variable-name), map-get($utility, css-variable-name), map-get($utility, class));\n\n    // State params to generate pseudo-classes\n    $state: if(map-has-key($utility, state), map-get($utility, state), ());\n\n    $infix: if($property-class == \"\" and str-slice($infix, 1, 1) == \"-\", str-slice($infix, 2), $infix);\n\n    // Don't prefix if value key is null (e.g. with shadow class)\n    $property-class-modifier: if($key, if($property-class == \"\" and $infix == \"\", \"\", \"-\") + $key, \"\");\n\n    @if map-get($utility, rfs) {\n      // Inside the media query\n      @if $is-rfs-media-query {\n        $val: rfs-value($value);\n\n        // Do not render anything if fluid and non fluid values are the same\n        $value: if($val == rfs-fluid-value($value), null, $val);\n      }\n      @else {\n        $value: rfs-fluid-value($value);\n      }\n    }\n\n    $is-css-var: map-get($utility, css-var);\n    $is-local-vars: map-get($utility, local-vars);\n    $is-rtl: map-get($utility, rtl);\n\n    @if $value != null {\n      @if $is-rtl == false {\n        /* rtl:begin:remove */\n      }\n\n      @if $is-css-var {\n        .#{$property-class + $infix + $property-class-modifier} {\n          --#{$prefix}#{$css-variable-name}: #{$value};\n        }\n\n        @each $pseudo in $state {\n          .#{$property-class + $infix + $property-class-modifier}-#{$pseudo}:#{$pseudo} {\n            --#{$prefix}#{$css-variable-name}: #{$value};\n          }\n        }\n      } @else {\n        .#{$property-class + $infix + $property-class-modifier} {\n          @each $property in $properties {\n            @if $is-local-vars {\n              @each $local-var, $variable in $is-local-vars {\n                --#{$prefix}#{$local-var}: #{$variable};\n              }\n            }\n            #{$property}: $value if($enable-important-utilities, !important, null);\n          }\n        }\n\n        @each $pseudo in $state {\n          .#{$property-class + $infix + $property-class-modifier}-#{$pseudo}:#{$pseudo} {\n            @each $property in $properties {\n              @if $is-local-vars {\n                @each $local-var, $variable in $is-local-vars {\n                  --#{$prefix}#{$local-var}: #{$variable};\n                }\n              }\n              #{$property}: $value if($enable-important-utilities, !important, null);\n            }\n          }\n        }\n      }\n\n      @if $is-rtl == false {\n        /* rtl:end:remove */\n      }\n    }\n  }\n}\n","// Loop over each breakpoint\n@each $breakpoint in map-keys($grid-breakpoints) {\n\n  // Generate media query if needed\n  @include media-breakpoint-up($breakpoint) {\n    $infix: breakpoint-infix($breakpoint, $grid-breakpoints);\n\n    // Loop over each utility property\n    @each $key, $utility in $utilities {\n      // The utility can be disabled with `false`, thus check if the utility is a map first\n      // Only proceed if responsive media queries are enabled or if it's the base media query\n      @if type-of($utility) == \"map\" and (map-get($utility, responsive) or $infix == \"\") {\n        @include generate-utility($utility, $infix);\n      }\n    }\n  }\n}\n\n// RFS rescaling\n@media (min-width: $rfs-mq-value) {\n  @each $breakpoint in map-keys($grid-breakpoints) {\n    $infix: breakpoint-infix($breakpoint, $grid-breakpoints);\n\n    @if (map-get($grid-breakpoints, $breakpoint) < $rfs-breakpoint) {\n      // Loop over each utility property\n      @each $key, $utility in $utilities {\n        // The utility can be disabled with `false`, thus check if the utility is a map first\n        // Only proceed if responsive media queries are enabled or if it's the base media query\n        @if type-of($utility) == \"map\" and map-get($utility, rfs) and (map-get($utility, responsive) or $infix == \"\") {\n          @include generate-utility($utility, $infix, true);\n        }\n      }\n    }\n  }\n}\n\n\n// Print utilities\n@media print {\n  @each $key, $utility in $utilities {\n    // The utility can be disabled with `false`, thus check if the utility is a map first\n    // Then check if the utility needs print styles\n    @if type-of($utility) == \"map\" and map-get($utility, print) == true {\n      @include generate-utility($utility, \"-print\");\n    }\n  }\n}\n","/*!\n * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com\n * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)\n * Copyright 2024 Fonticons, Inc.\n */\n// Font Awesome core compile (Web Fonts-based)\n// -------------------------\n\n@import 'functions';\n@import 'variables';\n@import 'mixins';\n@import 'core';\n@import 'sizing';\n@import 'fixed-width';\n@import 'list';\n@import 'bordered-pulled';\n@import 'animated';\n@import 'rotated-flipped';\n@import 'stacked';\n@import 'icons';\n@import 'screen-reader';\n","// base icon class definition\n// -------------------------\n\n.#{$fa-css-prefix} {\n  font-family: var(--#{$fa-css-prefix}-style-family, '#{$fa-style-family}');\n  font-weight: var(--#{$fa-css-prefix}-style, #{$fa-style});\n}\n\n.#{$fa-css-prefix},\n.#{$fa-css-prefix}-classic,\n.#{$fa-css-prefix}-sharp,\n.fas,\n.#{$fa-css-prefix}-solid,\n.far,\n.#{$fa-css-prefix}-regular,\n.fab,\n.#{$fa-css-prefix}-brands {\n  -moz-osx-font-smoothing: grayscale;\n  -webkit-font-smoothing: antialiased;\n  display: var(--#{$fa-css-prefix}-display, #{$fa-display});\n  font-style: normal;\n  font-variant: normal;\n  line-height: 1;\n  text-rendering: auto;\n}\n\n.fas,\n.#{$fa-css-prefix}-classic,\n.#{$fa-css-prefix}-solid,\n.far,\n.#{$fa-css-prefix}-regular {\n  font-family: 'Font Awesome 6 Free';\n}\n\n.fab,\n.#{$fa-css-prefix}-brands {\n  font-family: 'Font Awesome 6 Brands';\n}\n\n\n%fa-icon {\n  @include fa-icon;\n}\n","// sizing icons\n// -------------------------\n\n// literal magnification scale\n@for $i from 1 through 10 {\n  .#{$fa-css-prefix}-#{$i}x {\n    font-size: $i * 1em;\n  }\n}\n\n// step-based scale (with alignment)\n@each $size, $value in $fa-sizes {\n  .#{$fa-css-prefix}-#{$size} {\n     @include fa-size($value);\n  }\n}\n","// mixins\n// --------------------------\n\n// base rendering for an icon\n@mixin fa-icon {\n  -webkit-font-smoothing: antialiased;\n  -moz-osx-font-smoothing: grayscale;\n  display: inline-block;\n  font-style: normal;\n  font-variant: normal;\n  font-weight: normal;\n  line-height: 1;\n}\n\n// sets relative font-sizing and alignment (in _sizing)\n@mixin fa-size ($font-size) {\n  font-size: fa-divide($font-size, $fa-size-scale-base) * 1em; // converts step in sizing scale into an em-based value that's relative to the scale's base\n  line-height: fa-divide(1, $font-size) * 1em; // sets the line-height of the icon back to that of it's parent\n  vertical-align: (fa-divide(6, $font-size) - fa-divide(3, 8)) * 1em; // vertically centers the icon taking into account the surrounding text's descender\n}\n\n// only display content to screen readers\n// see: https://www.a11yproject.com/posts/2013-01-11-how-to-hide-content/\n// see: https://hugogiraudel.com/2016/10/13/css-hide-and-seek/\n@mixin fa-sr-only() {\n  position: absolute;\n  width: 1px;\n  height: 1px;\n  padding: 0;\n  margin: -1px;\n  overflow: hidden;\n  clip: rect(0, 0, 0, 0);\n  white-space: nowrap;\n  border-width: 0;\n}\n\n// use in conjunction with .sr-only to only display content when it's focused\n@mixin fa-sr-only-focusable() {\n  &:not(:focus) {\n    @include fa-sr-only();\n  }\n}\n\n// sets a specific icon family to use alongside style + icon mixins\n\n// convenience mixins for declaring pseudo-elements by CSS variable,\n// including all style-specific font properties, and both the ::before\n// and ::after elements in the duotone case.\n@mixin fa-icon-solid($fa-var) {\n  @extend %fa-icon;\n  @extend .fa-solid;\n\n  &::before {\n    content: unquote(\"\\\"#{ $fa-var }\\\"\");\n  }\n}\n@mixin fa-icon-regular($fa-var) {\n  @extend %fa-icon;\n  @extend .fa-regular;\n\n  &::before {\n    content: unquote(\"\\\"#{ $fa-var }\\\"\");\n  }\n}\n@mixin fa-icon-brands($fa-var) {\n  @extend %fa-icon;\n  @extend .fa-brands;\n\n  &::before {\n    content: unquote(\"\\\"#{ $fa-var }\\\"\");\n  }\n}\n","// fixed-width icons\n// -------------------------\n\n.#{$fa-css-prefix}-fw {\n  text-align: center;\n  width: $fa-fw-width;\n}\n","// icons in a list\n// -------------------------\n\n.#{$fa-css-prefix}-ul {\n  list-style-type: none;\n  margin-left: var(--#{$fa-css-prefix}-li-margin, #{$fa-li-margin});\n  padding-left: 0;\n\n  > li { position: relative; }\n}\n\n.#{$fa-css-prefix}-li {\n  left: calc(var(--#{$fa-css-prefix}-li-width, #{$fa-li-width}) * -1);\n  position: absolute;\n  text-align: center;\n  width: var(--#{$fa-css-prefix}-li-width, #{$fa-li-width});\n  line-height: inherit;\n}\n","// bordered + pulled icons\n// -------------------------\n\n.#{$fa-css-prefix}-border {\n  border-color: var(--#{$fa-css-prefix}-border-color, #{$fa-border-color});\n  border-radius: var(--#{$fa-css-prefix}-border-radius, #{$fa-border-radius});\n  border-style: var(--#{$fa-css-prefix}-border-style, #{$fa-border-style});\n  border-width: var(--#{$fa-css-prefix}-border-width, #{$fa-border-width});\n  padding: var(--#{$fa-css-prefix}-border-padding, #{$fa-border-padding});\n}\n\n.#{$fa-css-prefix}-pull-left {\n  float: left;\n  margin-right: var(--#{$fa-css-prefix}-pull-margin, #{$fa-pull-margin}); \n}\n\n.#{$fa-css-prefix}-pull-right {\n  float: right;\n  margin-left: var(--#{$fa-css-prefix}-pull-margin, #{$fa-pull-margin}); \n}\n","// animating icons\n// --------------------------\n\n.#{$fa-css-prefix}-beat {\n  animation-name: #{$fa-css-prefix}-beat;\n  animation-delay: var(--#{$fa-css-prefix}-animation-delay, 0s);\n  animation-direction: var(--#{$fa-css-prefix}-animation-direction, normal);\n  animation-duration: var(--#{$fa-css-prefix}-animation-duration, 1s);\n  animation-iteration-count: var(--#{$fa-css-prefix}-animation-iteration-count, infinite);\n  animation-timing-function: var(--#{$fa-css-prefix}-animation-timing, ease-in-out);\n}\n\n.#{$fa-css-prefix}-bounce {\n  animation-name: #{$fa-css-prefix}-bounce;\n  animation-delay: var(--#{$fa-css-prefix}-animation-delay, 0s);\n  animation-direction: var(--#{$fa-css-prefix}-animation-direction, normal);\n  animation-duration: var(--#{$fa-css-prefix}-animation-duration, 1s);\n  animation-iteration-count: var(--#{$fa-css-prefix}-animation-iteration-count, infinite);\n  animation-timing-function: var(--#{$fa-css-prefix}-animation-timing, cubic-bezier(0.280, 0.840, 0.420, 1));\n}\n\n.#{$fa-css-prefix}-fade {\n  animation-name: #{$fa-css-prefix}-fade;\n  animation-delay: var(--#{$fa-css-prefix}-animation-delay, 0s);\n  animation-direction: var(--#{$fa-css-prefix}-animation-direction, normal);\n  animation-duration: var(--#{$fa-css-prefix}-animation-duration, 1s);\n  animation-iteration-count: var(--#{$fa-css-prefix}-animation-iteration-count, infinite);\n  animation-timing-function: var(--#{$fa-css-prefix}-animation-timing, cubic-bezier(.4,0,.6,1));\n}\n\n.#{$fa-css-prefix}-beat-fade {\n  animation-name: #{$fa-css-prefix}-beat-fade;\n  animation-delay: var(--#{$fa-css-prefix}-animation-delay, 0s);\n  animation-direction: var(--#{$fa-css-prefix}-animation-direction, normal);\n  animation-duration: var(--#{$fa-css-prefix}-animation-duration, 1s);\n  animation-iteration-count: var(--#{$fa-css-prefix}-animation-iteration-count, infinite);\n  animation-timing-function: var(--#{$fa-css-prefix}-animation-timing, cubic-bezier(.4,0,.6,1));\n}\n\n.#{$fa-css-prefix}-flip {\n  animation-name: #{$fa-css-prefix}-flip;\n  animation-delay: var(--#{$fa-css-prefix}-animation-delay, 0s);\n  animation-direction: var(--#{$fa-css-prefix}-animation-direction, normal);\n  animation-duration: var(--#{$fa-css-prefix}-animation-duration, 1s);\n  animation-iteration-count: var(--#{$fa-css-prefix}-animation-iteration-count, infinite);\n  animation-timing-function: var(--#{$fa-css-prefix}-animation-timing, ease-in-out);\n}\n\n.#{$fa-css-prefix}-shake {\n  animation-name: #{$fa-css-prefix}-shake;\n  animation-delay: var(--#{$fa-css-prefix}-animation-delay, 0s);\n  animation-direction: var(--#{$fa-css-prefix}-animation-direction, normal);\n  animation-duration: var(--#{$fa-css-prefix}-animation-duration, 1s);\n  animation-iteration-count: var(--#{$fa-css-prefix}-animation-iteration-count, infinite);\n  animation-timing-function: var(--#{$fa-css-prefix}-animation-timing, linear);\n}\n\n.#{$fa-css-prefix}-spin {\n  animation-name: #{$fa-css-prefix}-spin;\n  animation-delay: var(--#{$fa-css-prefix}-animation-delay, 0s);\n  animation-direction: var(--#{$fa-css-prefix}-animation-direction, normal);\n  animation-duration: var(--#{$fa-css-prefix}-animation-duration, 2s);\n  animation-iteration-count: var(--#{$fa-css-prefix}-animation-iteration-count, infinite);\n  animation-timing-function: var(--#{$fa-css-prefix}-animation-timing, linear);\n}\n\n.#{$fa-css-prefix}-spin-reverse {\n  --#{$fa-css-prefix}-animation-direction: reverse;\n}\n\n.#{$fa-css-prefix}-pulse,\n.#{$fa-css-prefix}-spin-pulse {\n  animation-name: #{$fa-css-prefix}-spin;\n  animation-direction: var(--#{$fa-css-prefix}-animation-direction, normal);\n  animation-duration: var(--#{$fa-css-prefix}-animation-duration, 1s);\n  animation-iteration-count: var(--#{$fa-css-prefix}-animation-iteration-count, infinite);\n  animation-timing-function: var(--#{$fa-css-prefix}-animation-timing, steps(8));\n}\n\n// if agent or operating system prefers reduced motion, disable animations\n// see: https://www.smashingmagazine.com/2020/09/design-reduced-motion-sensitivities/\n// see: https://developer.mozilla.org/en-US/docs/Web/CSS/@media/prefers-reduced-motion\n@media (prefers-reduced-motion: reduce) {\n  .#{$fa-css-prefix}-beat,\n  .#{$fa-css-prefix}-bounce,\n  .#{$fa-css-prefix}-fade,\n  .#{$fa-css-prefix}-beat-fade,\n  .#{$fa-css-prefix}-flip,\n  .#{$fa-css-prefix}-pulse,\n  .#{$fa-css-prefix}-shake,\n  .#{$fa-css-prefix}-spin,\n  .#{$fa-css-prefix}-spin-pulse {\n    animation-delay: -1ms;\n    animation-duration: 1ms;\n    animation-iteration-count: 1;\n    transition-delay: 0s;\n    transition-duration: 0s;\n  }\n}\n\n@keyframes #{$fa-css-prefix}-beat {\n  0%, 90% { transform: scale(1); }\n  45% { transform: scale(var(--#{$fa-css-prefix}-beat-scale, 1.25)); }\n}\n\n@keyframes #{$fa-css-prefix}-bounce {\n  0%   { transform: scale(1,1) translateY(0); }\n  10%  { transform: scale(var(--#{$fa-css-prefix}-bounce-start-scale-x, 1.1),var(--#{$fa-css-prefix}-bounce-start-scale-y, 0.9)) translateY(0); }\n  30%  { transform: scale(var(--#{$fa-css-prefix}-bounce-jump-scale-x, 0.9),var(--#{$fa-css-prefix}-bounce-jump-scale-y, 1.1)) translateY(var(--#{$fa-css-prefix}-bounce-height, -0.5em)); }\n  50%  { transform: scale(var(--#{$fa-css-prefix}-bounce-land-scale-x, 1.05),var(--#{$fa-css-prefix}-bounce-land-scale-y, 0.95)) translateY(0); }\n  57%  { transform: scale(1,1) translateY(var(--#{$fa-css-prefix}-bounce-rebound, -0.125em)); }\n  64%  { transform: scale(1,1) translateY(0); }\n  100% { transform: scale(1,1) translateY(0); }\n}\n\n@keyframes #{$fa-css-prefix}-fade {\n  50% { opacity: var(--#{$fa-css-prefix}-fade-opacity, 0.4); }\n}\n\n@keyframes #{$fa-css-prefix}-beat-fade {\n  0%, 100% {\n    opacity: var(--#{$fa-css-prefix}-beat-fade-opacity, 0.4);\n    transform: scale(1);\n  }\n  50% {\n    opacity: 1;\n    transform: scale(var(--#{$fa-css-prefix}-beat-fade-scale, 1.125));\n  }\n}\n\n@keyframes #{$fa-css-prefix}-flip {\n  50% {\n    transform: rotate3d(var(--#{$fa-css-prefix}-flip-x, 0), var(--#{$fa-css-prefix}-flip-y, 1), var(--#{$fa-css-prefix}-flip-z, 0), var(--#{$fa-css-prefix}-flip-angle, -180deg));\n  }\n}\n\n@keyframes #{$fa-css-prefix}-shake {\n  0% { transform: rotate(-15deg); }\n  4% { transform: rotate(15deg); }\n  8%, 24% { transform: rotate(-18deg); }\n  12%, 28% { transform: rotate(18deg); }\n  16% { transform: rotate(-22deg); }\n  20% { transform: rotate(22deg); }\n  32% { transform: rotate(-12deg); }\n  36% { transform: rotate(12deg); }\n  40%, 100% { transform: rotate(0deg); }\n}\n\n@keyframes #{$fa-css-prefix}-spin {\n  0% { transform: rotate(0deg); }\n  100% { transform: rotate(360deg); }\n}\n","// rotating + flipping icons\n// -------------------------\n\n.#{$fa-css-prefix}-rotate-90 {\n  transform: rotate(90deg);\n}\n\n.#{$fa-css-prefix}-rotate-180 {\n  transform: rotate(180deg);\n}\n\n.#{$fa-css-prefix}-rotate-270 {\n  transform: rotate(270deg);\n}\n\n.#{$fa-css-prefix}-flip-horizontal {\n  transform: scale(-1, 1);\n}\n\n.#{$fa-css-prefix}-flip-vertical {\n  transform: scale(1, -1);\n}\n\n.#{$fa-css-prefix}-flip-both,\n.#{$fa-css-prefix}-flip-horizontal.#{$fa-css-prefix}-flip-vertical {\n  transform: scale(-1, -1);\n}\n\n.#{$fa-css-prefix}-rotate-by {\n  transform: rotate(var(--#{$fa-css-prefix}-rotate-angle, 0));\n}\n","// stacking icons\n// -------------------------\n\n.#{$fa-css-prefix}-stack {\n  display: inline-block;\n  height: 2em;\n  line-height: 2em;\n  position: relative;\n  vertical-align: $fa-stack-vertical-align;\n  width: $fa-stack-width;\n}\n\n.#{$fa-css-prefix}-stack-1x,\n.#{$fa-css-prefix}-stack-2x {\n  left: 0;\n  position: absolute;\n  text-align: center;\n  width: 100%;\n  z-index: var(--#{$fa-css-prefix}-stack-z-index, #{$fa-stack-z-index});\n}\n\n.#{$fa-css-prefix}-stack-1x {\n  line-height: inherit;\n}\n\n.#{$fa-css-prefix}-stack-2x {\n  font-size: 2em;\n}\n\n.#{$fa-css-prefix}-inverse {\n  color: var(--#{$fa-css-prefix}-inverse, #{$fa-inverse});\n}\n","// variables\n// --------------------------\n\n$fa-css-prefix          : fa !default;\n$fa-style               : 900 !default;\n$fa-style-family        : \"Font Awesome 6 Free\" !default;\n\n$fa-display             : inline-block !default;\n\n$fa-fw-width            : fa-divide(20em, 16) !default;\n$fa-inverse             : #fff !default;\n\n$fa-border-color        : #eee !default;\n$fa-border-padding      : .2em .25em .15em !default;\n$fa-border-radius       : .1em !default;\n$fa-border-style        : solid !default;\n$fa-border-width        : .08em !default;\n\n$fa-size-scale-2xs      : 10 !default;\n$fa-size-scale-xs       : 12 !default;\n$fa-size-scale-sm       : 14 !default;\n$fa-size-scale-base     : 16 !default;\n$fa-size-scale-lg       : 20 !default;\n$fa-size-scale-xl       : 24 !default;\n$fa-size-scale-2xl      : 32 !default;\n\n$fa-sizes: (\n  \"2xs\"                 : $fa-size-scale-2xs,\n  \"xs\"                  : $fa-size-scale-xs,\n  \"sm\"                  : $fa-size-scale-sm,\n  \"lg\"                  : $fa-size-scale-lg,\n  \"xl\"                  : $fa-size-scale-xl,\n  \"2xl\"                 : $fa-size-scale-2xl\n) !default;\n\n$fa-li-width            : 2em !default;\n$fa-li-margin           : $fa-li-width * fa-divide(5, 4) !default;\n\n$fa-pull-margin         : .3em !default;\n\n$fa-primary-opacity     : 1 !default;\n$fa-secondary-opacity   : .4 !default;\n\n$fa-stack-vertical-align: middle !default;\n$fa-stack-width         : ($fa-fw-width * 2) !default;\n$fa-stack-z-index       : auto !default;\n\n$fa-font-display        : block !default;\n$fa-font-path           : \"../webfonts\" !default;\n\n$fa-var-0: \\30;\n$fa-var-1: \\31;\n$fa-var-2: \\32;\n$fa-var-3: \\33;\n$fa-var-4: \\34;\n$fa-var-5: \\35;\n$fa-var-6: \\36;\n$fa-var-7: \\37;\n$fa-var-8: \\38;\n$fa-var-9: \\39;\n$fa-var-fill-drip: \\f576;\n$fa-var-arrows-to-circle: \\e4bd;\n$fa-var-circle-chevron-right: \\f138;\n$fa-var-chevron-circle-right: \\f138;\n$fa-var-at: \\40;\n$fa-var-trash-can: \\f2ed;\n$fa-var-trash-alt: \\f2ed;\n$fa-var-text-height: \\f034;\n$fa-var-user-xmark: \\f235;\n$fa-var-user-times: \\f235;\n$fa-var-stethoscope: \\f0f1;\n$fa-var-message: \\f27a;\n$fa-var-comment-alt: \\f27a;\n$fa-var-info: \\f129;\n$fa-var-down-left-and-up-right-to-center: \\f422;\n$fa-var-compress-alt: \\f422;\n$fa-var-explosion: \\e4e9;\n$fa-var-file-lines: \\f15c;\n$fa-var-file-alt: \\f15c;\n$fa-var-file-text: \\f15c;\n$fa-var-wave-square: \\f83e;\n$fa-var-ring: \\f70b;\n$fa-var-building-un: \\e4d9;\n$fa-var-dice-three: \\f527;\n$fa-var-calendar-days: \\f073;\n$fa-var-calendar-alt: \\f073;\n$fa-var-anchor-circle-check: \\e4aa;\n$fa-var-building-circle-arrow-right: \\e4d1;\n$fa-var-volleyball: \\f45f;\n$fa-var-volleyball-ball: \\f45f;\n$fa-var-arrows-up-to-line: \\e4c2;\n$fa-var-sort-down: \\f0dd;\n$fa-var-sort-desc: \\f0dd;\n$fa-var-circle-minus: \\f056;\n$fa-var-minus-circle: \\f056;\n$fa-var-door-open: \\f52b;\n$fa-var-right-from-bracket: \\f2f5;\n$fa-var-sign-out-alt: \\f2f5;\n$fa-var-atom: \\f5d2;\n$fa-var-soap: \\e06e;\n$fa-var-icons: \\f86d;\n$fa-var-heart-music-camera-bolt: \\f86d;\n$fa-var-microphone-lines-slash: \\f539;\n$fa-var-microphone-alt-slash: \\f539;\n$fa-var-bridge-circle-check: \\e4c9;\n$fa-var-pump-medical: \\e06a;\n$fa-var-fingerprint: \\f577;\n$fa-var-hand-point-right: \\f0a4;\n$fa-var-magnifying-glass-location: \\f689;\n$fa-var-search-location: \\f689;\n$fa-var-forward-step: \\f051;\n$fa-var-step-forward: \\f051;\n$fa-var-face-smile-beam: \\f5b8;\n$fa-var-smile-beam: \\f5b8;\n$fa-var-flag-checkered: \\f11e;\n$fa-var-football: \\f44e;\n$fa-var-football-ball: \\f44e;\n$fa-var-school-circle-exclamation: \\e56c;\n$fa-var-crop: \\f125;\n$fa-var-angles-down: \\f103;\n$fa-var-angle-double-down: \\f103;\n$fa-var-users-rectangle: \\e594;\n$fa-var-people-roof: \\e537;\n$fa-var-people-line: \\e534;\n$fa-var-beer-mug-empty: \\f0fc;\n$fa-var-beer: \\f0fc;\n$fa-var-diagram-predecessor: \\e477;\n$fa-var-arrow-up-long: \\f176;\n$fa-var-long-arrow-up: \\f176;\n$fa-var-fire-flame-simple: \\f46a;\n$fa-var-burn: \\f46a;\n$fa-var-person: \\f183;\n$fa-var-male: \\f183;\n$fa-var-laptop: \\f109;\n$fa-var-file-csv: \\f6dd;\n$fa-var-menorah: \\f676;\n$fa-var-truck-plane: \\e58f;\n$fa-var-record-vinyl: \\f8d9;\n$fa-var-face-grin-stars: \\f587;\n$fa-var-grin-stars: \\f587;\n$fa-var-bong: \\f55c;\n$fa-var-spaghetti-monster-flying: \\f67b;\n$fa-var-pastafarianism: \\f67b;\n$fa-var-arrow-down-up-across-line: \\e4af;\n$fa-var-spoon: \\f2e5;\n$fa-var-utensil-spoon: \\f2e5;\n$fa-var-jar-wheat: \\e517;\n$fa-var-envelopes-bulk: \\f674;\n$fa-var-mail-bulk: \\f674;\n$fa-var-file-circle-exclamation: \\e4eb;\n$fa-var-circle-h: \\f47e;\n$fa-var-hospital-symbol: \\f47e;\n$fa-var-pager: \\f815;\n$fa-var-address-book: \\f2b9;\n$fa-var-contact-book: \\f2b9;\n$fa-var-strikethrough: \\f0cc;\n$fa-var-k: \\4b;\n$fa-var-landmark-flag: \\e51c;\n$fa-var-pencil: \\f303;\n$fa-var-pencil-alt: \\f303;\n$fa-var-backward: \\f04a;\n$fa-var-caret-right: \\f0da;\n$fa-var-comments: \\f086;\n$fa-var-paste: \\f0ea;\n$fa-var-file-clipboard: \\f0ea;\n$fa-var-code-pull-request: \\e13c;\n$fa-var-clipboard-list: \\f46d;\n$fa-var-truck-ramp-box: \\f4de;\n$fa-var-truck-loading: \\f4de;\n$fa-var-user-check: \\f4fc;\n$fa-var-vial-virus: \\e597;\n$fa-var-sheet-plastic: \\e571;\n$fa-var-blog: \\f781;\n$fa-var-user-ninja: \\f504;\n$fa-var-person-arrow-up-from-line: \\e539;\n$fa-var-scroll-torah: \\f6a0;\n$fa-var-torah: \\f6a0;\n$fa-var-broom-ball: \\f458;\n$fa-var-quidditch: \\f458;\n$fa-var-quidditch-broom-ball: \\f458;\n$fa-var-toggle-off: \\f204;\n$fa-var-box-archive: \\f187;\n$fa-var-archive: \\f187;\n$fa-var-person-drowning: \\e545;\n$fa-var-arrow-down-9-1: \\f886;\n$fa-var-sort-numeric-desc: \\f886;\n$fa-var-sort-numeric-down-alt: \\f886;\n$fa-var-face-grin-tongue-squint: \\f58a;\n$fa-var-grin-tongue-squint: \\f58a;\n$fa-var-spray-can: \\f5bd;\n$fa-var-truck-monster: \\f63b;\n$fa-var-w: \\57;\n$fa-var-earth-africa: \\f57c;\n$fa-var-globe-africa: \\f57c;\n$fa-var-rainbow: \\f75b;\n$fa-var-circle-notch: \\f1ce;\n$fa-var-tablet-screen-button: \\f3fa;\n$fa-var-tablet-alt: \\f3fa;\n$fa-var-paw: \\f1b0;\n$fa-var-cloud: \\f0c2;\n$fa-var-trowel-bricks: \\e58a;\n$fa-var-face-flushed: \\f579;\n$fa-var-flushed: \\f579;\n$fa-var-hospital-user: \\f80d;\n$fa-var-tent-arrow-left-right: \\e57f;\n$fa-var-gavel: \\f0e3;\n$fa-var-legal: \\f0e3;\n$fa-var-binoculars: \\f1e5;\n$fa-var-microphone-slash: \\f131;\n$fa-var-box-tissue: \\e05b;\n$fa-var-motorcycle: \\f21c;\n$fa-var-bell-concierge: \\f562;\n$fa-var-concierge-bell: \\f562;\n$fa-var-pen-ruler: \\f5ae;\n$fa-var-pencil-ruler: \\f5ae;\n$fa-var-people-arrows: \\e068;\n$fa-var-people-arrows-left-right: \\e068;\n$fa-var-mars-and-venus-burst: \\e523;\n$fa-var-square-caret-right: \\f152;\n$fa-var-caret-square-right: \\f152;\n$fa-var-scissors: \\f0c4;\n$fa-var-cut: \\f0c4;\n$fa-var-sun-plant-wilt: \\e57a;\n$fa-var-toilets-portable: \\e584;\n$fa-var-hockey-puck: \\f453;\n$fa-var-table: \\f0ce;\n$fa-var-magnifying-glass-arrow-right: \\e521;\n$fa-var-tachograph-digital: \\f566;\n$fa-var-digital-tachograph: \\f566;\n$fa-var-users-slash: \\e073;\n$fa-var-clover: \\e139;\n$fa-var-reply: \\f3e5;\n$fa-var-mail-reply: \\f3e5;\n$fa-var-star-and-crescent: \\f699;\n$fa-var-house-fire: \\e50c;\n$fa-var-square-minus: \\f146;\n$fa-var-minus-square: \\f146;\n$fa-var-helicopter: \\f533;\n$fa-var-compass: \\f14e;\n$fa-var-square-caret-down: \\f150;\n$fa-var-caret-square-down: \\f150;\n$fa-var-file-circle-question: \\e4ef;\n$fa-var-laptop-code: \\f5fc;\n$fa-var-swatchbook: \\f5c3;\n$fa-var-prescription-bottle: \\f485;\n$fa-var-bars: \\f0c9;\n$fa-var-navicon: \\f0c9;\n$fa-var-people-group: \\e533;\n$fa-var-hourglass-end: \\f253;\n$fa-var-hourglass-3: \\f253;\n$fa-var-heart-crack: \\f7a9;\n$fa-var-heart-broken: \\f7a9;\n$fa-var-square-up-right: \\f360;\n$fa-var-external-link-square-alt: \\f360;\n$fa-var-face-kiss-beam: \\f597;\n$fa-var-kiss-beam: \\f597;\n$fa-var-film: \\f008;\n$fa-var-ruler-horizontal: \\f547;\n$fa-var-people-robbery: \\e536;\n$fa-var-lightbulb: \\f0eb;\n$fa-var-caret-left: \\f0d9;\n$fa-var-circle-exclamation: \\f06a;\n$fa-var-exclamation-circle: \\f06a;\n$fa-var-school-circle-xmark: \\e56d;\n$fa-var-arrow-right-from-bracket: \\f08b;\n$fa-var-sign-out: \\f08b;\n$fa-var-circle-chevron-down: \\f13a;\n$fa-var-chevron-circle-down: \\f13a;\n$fa-var-unlock-keyhole: \\f13e;\n$fa-var-unlock-alt: \\f13e;\n$fa-var-cloud-showers-heavy: \\f740;\n$fa-var-headphones-simple: \\f58f;\n$fa-var-headphones-alt: \\f58f;\n$fa-var-sitemap: \\f0e8;\n$fa-var-circle-dollar-to-slot: \\f4b9;\n$fa-var-donate: \\f4b9;\n$fa-var-memory: \\f538;\n$fa-var-road-spikes: \\e568;\n$fa-var-fire-burner: \\e4f1;\n$fa-var-flag: \\f024;\n$fa-var-hanukiah: \\f6e6;\n$fa-var-feather: \\f52d;\n$fa-var-volume-low: \\f027;\n$fa-var-volume-down: \\f027;\n$fa-var-comment-slash: \\f4b3;\n$fa-var-cloud-sun-rain: \\f743;\n$fa-var-compress: \\f066;\n$fa-var-wheat-awn: \\e2cd;\n$fa-var-wheat-alt: \\e2cd;\n$fa-var-ankh: \\f644;\n$fa-var-hands-holding-child: \\e4fa;\n$fa-var-asterisk: \\2a;\n$fa-var-square-check: \\f14a;\n$fa-var-check-square: \\f14a;\n$fa-var-peseta-sign: \\e221;\n$fa-var-heading: \\f1dc;\n$fa-var-header: \\f1dc;\n$fa-var-ghost: \\f6e2;\n$fa-var-list: \\f03a;\n$fa-var-list-squares: \\f03a;\n$fa-var-square-phone-flip: \\f87b;\n$fa-var-phone-square-alt: \\f87b;\n$fa-var-cart-plus: \\f217;\n$fa-var-gamepad: \\f11b;\n$fa-var-circle-dot: \\f192;\n$fa-var-dot-circle: \\f192;\n$fa-var-face-dizzy: \\f567;\n$fa-var-dizzy: \\f567;\n$fa-var-egg: \\f7fb;\n$fa-var-house-medical-circle-xmark: \\e513;\n$fa-var-campground: \\f6bb;\n$fa-var-folder-plus: \\f65e;\n$fa-var-futbol: \\f1e3;\n$fa-var-futbol-ball: \\f1e3;\n$fa-var-soccer-ball: \\f1e3;\n$fa-var-paintbrush: \\f1fc;\n$fa-var-paint-brush: \\f1fc;\n$fa-var-lock: \\f023;\n$fa-var-gas-pump: \\f52f;\n$fa-var-hot-tub-person: \\f593;\n$fa-var-hot-tub: \\f593;\n$fa-var-map-location: \\f59f;\n$fa-var-map-marked: \\f59f;\n$fa-var-house-flood-water: \\e50e;\n$fa-var-tree: \\f1bb;\n$fa-var-bridge-lock: \\e4cc;\n$fa-var-sack-dollar: \\f81d;\n$fa-var-pen-to-square: \\f044;\n$fa-var-edit: \\f044;\n$fa-var-car-side: \\f5e4;\n$fa-var-share-nodes: \\f1e0;\n$fa-var-share-alt: \\f1e0;\n$fa-var-heart-circle-minus: \\e4ff;\n$fa-var-hourglass-half: \\f252;\n$fa-var-hourglass-2: \\f252;\n$fa-var-microscope: \\f610;\n$fa-var-sink: \\e06d;\n$fa-var-bag-shopping: \\f290;\n$fa-var-shopping-bag: \\f290;\n$fa-var-arrow-down-z-a: \\f881;\n$fa-var-sort-alpha-desc: \\f881;\n$fa-var-sort-alpha-down-alt: \\f881;\n$fa-var-mitten: \\f7b5;\n$fa-var-person-rays: \\e54d;\n$fa-var-users: \\f0c0;\n$fa-var-eye-slash: \\f070;\n$fa-var-flask-vial: \\e4f3;\n$fa-var-hand: \\f256;\n$fa-var-hand-paper: \\f256;\n$fa-var-om: \\f679;\n$fa-var-worm: \\e599;\n$fa-var-house-circle-xmark: \\e50b;\n$fa-var-plug: \\f1e6;\n$fa-var-chevron-up: \\f077;\n$fa-var-hand-spock: \\f259;\n$fa-var-stopwatch: \\f2f2;\n$fa-var-face-kiss: \\f596;\n$fa-var-kiss: \\f596;\n$fa-var-bridge-circle-xmark: \\e4cb;\n$fa-var-face-grin-tongue: \\f589;\n$fa-var-grin-tongue: \\f589;\n$fa-var-chess-bishop: \\f43a;\n$fa-var-face-grin-wink: \\f58c;\n$fa-var-grin-wink: \\f58c;\n$fa-var-ear-deaf: \\f2a4;\n$fa-var-deaf: \\f2a4;\n$fa-var-deafness: \\f2a4;\n$fa-var-hard-of-hearing: \\f2a4;\n$fa-var-road-circle-check: \\e564;\n$fa-var-dice-five: \\f523;\n$fa-var-square-rss: \\f143;\n$fa-var-rss-square: \\f143;\n$fa-var-land-mine-on: \\e51b;\n$fa-var-i-cursor: \\f246;\n$fa-var-stamp: \\f5bf;\n$fa-var-stairs: \\e289;\n$fa-var-i: \\49;\n$fa-var-hryvnia-sign: \\f6f2;\n$fa-var-hryvnia: \\f6f2;\n$fa-var-pills: \\f484;\n$fa-var-face-grin-wide: \\f581;\n$fa-var-grin-alt: \\f581;\n$fa-var-tooth: \\f5c9;\n$fa-var-v: \\56;\n$fa-var-bangladeshi-taka-sign: \\e2e6;\n$fa-var-bicycle: \\f206;\n$fa-var-staff-snake: \\e579;\n$fa-var-rod-asclepius: \\e579;\n$fa-var-rod-snake: \\e579;\n$fa-var-staff-aesculapius: \\e579;\n$fa-var-head-side-cough-slash: \\e062;\n$fa-var-truck-medical: \\f0f9;\n$fa-var-ambulance: \\f0f9;\n$fa-var-wheat-awn-circle-exclamation: \\e598;\n$fa-var-snowman: \\f7d0;\n$fa-var-mortar-pestle: \\f5a7;\n$fa-var-road-barrier: \\e562;\n$fa-var-school: \\f549;\n$fa-var-igloo: \\f7ae;\n$fa-var-joint: \\f595;\n$fa-var-angle-right: \\f105;\n$fa-var-horse: \\f6f0;\n$fa-var-q: \\51;\n$fa-var-g: \\47;\n$fa-var-notes-medical: \\f481;\n$fa-var-temperature-half: \\f2c9;\n$fa-var-temperature-2: \\f2c9;\n$fa-var-thermometer-2: \\f2c9;\n$fa-var-thermometer-half: \\f2c9;\n$fa-var-dong-sign: \\e169;\n$fa-var-capsules: \\f46b;\n$fa-var-poo-storm: \\f75a;\n$fa-var-poo-bolt: \\f75a;\n$fa-var-face-frown-open: \\f57a;\n$fa-var-frown-open: \\f57a;\n$fa-var-hand-point-up: \\f0a6;\n$fa-var-money-bill: \\f0d6;\n$fa-var-bookmark: \\f02e;\n$fa-var-align-justify: \\f039;\n$fa-var-umbrella-beach: \\f5ca;\n$fa-var-helmet-un: \\e503;\n$fa-var-bullseye: \\f140;\n$fa-var-bacon: \\f7e5;\n$fa-var-hand-point-down: \\f0a7;\n$fa-var-arrow-up-from-bracket: \\e09a;\n$fa-var-folder: \\f07b;\n$fa-var-folder-blank: \\f07b;\n$fa-var-file-waveform: \\f478;\n$fa-var-file-medical-alt: \\f478;\n$fa-var-radiation: \\f7b9;\n$fa-var-chart-simple: \\e473;\n$fa-var-mars-stroke: \\f229;\n$fa-var-vial: \\f492;\n$fa-var-gauge: \\f624;\n$fa-var-dashboard: \\f624;\n$fa-var-gauge-med: \\f624;\n$fa-var-tachometer-alt-average: \\f624;\n$fa-var-wand-magic-sparkles: \\e2ca;\n$fa-var-magic-wand-sparkles: \\e2ca;\n$fa-var-e: \\45;\n$fa-var-pen-clip: \\f305;\n$fa-var-pen-alt: \\f305;\n$fa-var-bridge-circle-exclamation: \\e4ca;\n$fa-var-user: \\f007;\n$fa-var-school-circle-check: \\e56b;\n$fa-var-dumpster: \\f793;\n$fa-var-van-shuttle: \\f5b6;\n$fa-var-shuttle-van: \\f5b6;\n$fa-var-building-user: \\e4da;\n$fa-var-square-caret-left: \\f191;\n$fa-var-caret-square-left: \\f191;\n$fa-var-highlighter: \\f591;\n$fa-var-key: \\f084;\n$fa-var-bullhorn: \\f0a1;\n$fa-var-globe: \\f0ac;\n$fa-var-synagogue: \\f69b;\n$fa-var-person-half-dress: \\e548;\n$fa-var-road-bridge: \\e563;\n$fa-var-location-arrow: \\f124;\n$fa-var-c: \\43;\n$fa-var-tablet-button: \\f10a;\n$fa-var-building-lock: \\e4d6;\n$fa-var-pizza-slice: \\f818;\n$fa-var-money-bill-wave: \\f53a;\n$fa-var-chart-area: \\f1fe;\n$fa-var-area-chart: \\f1fe;\n$fa-var-house-flag: \\e50d;\n$fa-var-person-circle-minus: \\e540;\n$fa-var-ban: \\f05e;\n$fa-var-cancel: \\f05e;\n$fa-var-camera-rotate: \\e0d8;\n$fa-var-spray-can-sparkles: \\f5d0;\n$fa-var-air-freshener: \\f5d0;\n$fa-var-star: \\f005;\n$fa-var-repeat: \\f363;\n$fa-var-cross: \\f654;\n$fa-var-box: \\f466;\n$fa-var-venus-mars: \\f228;\n$fa-var-arrow-pointer: \\f245;\n$fa-var-mouse-pointer: \\f245;\n$fa-var-maximize: \\f31e;\n$fa-var-expand-arrows-alt: \\f31e;\n$fa-var-charging-station: \\f5e7;\n$fa-var-shapes: \\f61f;\n$fa-var-triangle-circle-square: \\f61f;\n$fa-var-shuffle: \\f074;\n$fa-var-random: \\f074;\n$fa-var-person-running: \\f70c;\n$fa-var-running: \\f70c;\n$fa-var-mobile-retro: \\e527;\n$fa-var-grip-lines-vertical: \\f7a5;\n$fa-var-spider: \\f717;\n$fa-var-hands-bound: \\e4f9;\n$fa-var-file-invoice-dollar: \\f571;\n$fa-var-plane-circle-exclamation: \\e556;\n$fa-var-x-ray: \\f497;\n$fa-var-spell-check: \\f891;\n$fa-var-slash: \\f715;\n$fa-var-computer-mouse: \\f8cc;\n$fa-var-mouse: \\f8cc;\n$fa-var-arrow-right-to-bracket: \\f090;\n$fa-var-sign-in: \\f090;\n$fa-var-shop-slash: \\e070;\n$fa-var-store-alt-slash: \\e070;\n$fa-var-server: \\f233;\n$fa-var-virus-covid-slash: \\e4a9;\n$fa-var-shop-lock: \\e4a5;\n$fa-var-hourglass-start: \\f251;\n$fa-var-hourglass-1: \\f251;\n$fa-var-blender-phone: \\f6b6;\n$fa-var-building-wheat: \\e4db;\n$fa-var-person-breastfeeding: \\e53a;\n$fa-var-right-to-bracket: \\f2f6;\n$fa-var-sign-in-alt: \\f2f6;\n$fa-var-venus: \\f221;\n$fa-var-passport: \\f5ab;\n$fa-var-heart-pulse: \\f21e;\n$fa-var-heartbeat: \\f21e;\n$fa-var-people-carry-box: \\f4ce;\n$fa-var-people-carry: \\f4ce;\n$fa-var-temperature-high: \\f769;\n$fa-var-microchip: \\f2db;\n$fa-var-crown: \\f521;\n$fa-var-weight-hanging: \\f5cd;\n$fa-var-xmarks-lines: \\e59a;\n$fa-var-file-prescription: \\f572;\n$fa-var-weight-scale: \\f496;\n$fa-var-weight: \\f496;\n$fa-var-user-group: \\f500;\n$fa-var-user-friends: \\f500;\n$fa-var-arrow-up-a-z: \\f15e;\n$fa-var-sort-alpha-up: \\f15e;\n$fa-var-chess-knight: \\f441;\n$fa-var-face-laugh-squint: \\f59b;\n$fa-var-laugh-squint: \\f59b;\n$fa-var-wheelchair: \\f193;\n$fa-var-circle-arrow-up: \\f0aa;\n$fa-var-arrow-circle-up: \\f0aa;\n$fa-var-toggle-on: \\f205;\n$fa-var-person-walking: \\f554;\n$fa-var-walking: \\f554;\n$fa-var-l: \\4c;\n$fa-var-fire: \\f06d;\n$fa-var-bed-pulse: \\f487;\n$fa-var-procedures: \\f487;\n$fa-var-shuttle-space: \\f197;\n$fa-var-space-shuttle: \\f197;\n$fa-var-face-laugh: \\f599;\n$fa-var-laugh: \\f599;\n$fa-var-folder-open: \\f07c;\n$fa-var-heart-circle-plus: \\e500;\n$fa-var-code-fork: \\e13b;\n$fa-var-city: \\f64f;\n$fa-var-microphone-lines: \\f3c9;\n$fa-var-microphone-alt: \\f3c9;\n$fa-var-pepper-hot: \\f816;\n$fa-var-unlock: \\f09c;\n$fa-var-colon-sign: \\e140;\n$fa-var-headset: \\f590;\n$fa-var-store-slash: \\e071;\n$fa-var-road-circle-xmark: \\e566;\n$fa-var-user-minus: \\f503;\n$fa-var-mars-stroke-up: \\f22a;\n$fa-var-mars-stroke-v: \\f22a;\n$fa-var-champagne-glasses: \\f79f;\n$fa-var-glass-cheers: \\f79f;\n$fa-var-clipboard: \\f328;\n$fa-var-house-circle-exclamation: \\e50a;\n$fa-var-file-arrow-up: \\f574;\n$fa-var-file-upload: \\f574;\n$fa-var-wifi: \\f1eb;\n$fa-var-wifi-3: \\f1eb;\n$fa-var-wifi-strong: \\f1eb;\n$fa-var-bath: \\f2cd;\n$fa-var-bathtub: \\f2cd;\n$fa-var-underline: \\f0cd;\n$fa-var-user-pen: \\f4ff;\n$fa-var-user-edit: \\f4ff;\n$fa-var-signature: \\f5b7;\n$fa-var-stroopwafel: \\f551;\n$fa-var-bold: \\f032;\n$fa-var-anchor-lock: \\e4ad;\n$fa-var-building-ngo: \\e4d7;\n$fa-var-manat-sign: \\e1d5;\n$fa-var-not-equal: \\f53e;\n$fa-var-border-top-left: \\f853;\n$fa-var-border-style: \\f853;\n$fa-var-map-location-dot: \\f5a0;\n$fa-var-map-marked-alt: \\f5a0;\n$fa-var-jedi: \\f669;\n$fa-var-square-poll-vertical: \\f681;\n$fa-var-poll: \\f681;\n$fa-var-mug-hot: \\f7b6;\n$fa-var-car-battery: \\f5df;\n$fa-var-battery-car: \\f5df;\n$fa-var-gift: \\f06b;\n$fa-var-dice-two: \\f528;\n$fa-var-chess-queen: \\f445;\n$fa-var-glasses: \\f530;\n$fa-var-chess-board: \\f43c;\n$fa-var-building-circle-check: \\e4d2;\n$fa-var-person-chalkboard: \\e53d;\n$fa-var-mars-stroke-right: \\f22b;\n$fa-var-mars-stroke-h: \\f22b;\n$fa-var-hand-back-fist: \\f255;\n$fa-var-hand-rock: \\f255;\n$fa-var-square-caret-up: \\f151;\n$fa-var-caret-square-up: \\f151;\n$fa-var-cloud-showers-water: \\e4e4;\n$fa-var-chart-bar: \\f080;\n$fa-var-bar-chart: \\f080;\n$fa-var-hands-bubbles: \\e05e;\n$fa-var-hands-wash: \\e05e;\n$fa-var-less-than-equal: \\f537;\n$fa-var-train: \\f238;\n$fa-var-eye-low-vision: \\f2a8;\n$fa-var-low-vision: \\f2a8;\n$fa-var-crow: \\f520;\n$fa-var-sailboat: \\e445;\n$fa-var-window-restore: \\f2d2;\n$fa-var-square-plus: \\f0fe;\n$fa-var-plus-square: \\f0fe;\n$fa-var-torii-gate: \\f6a1;\n$fa-var-frog: \\f52e;\n$fa-var-bucket: \\e4cf;\n$fa-var-image: \\f03e;\n$fa-var-microphone: \\f130;\n$fa-var-cow: \\f6c8;\n$fa-var-caret-up: \\f0d8;\n$fa-var-screwdriver: \\f54a;\n$fa-var-folder-closed: \\e185;\n$fa-var-house-tsunami: \\e515;\n$fa-var-square-nfi: \\e576;\n$fa-var-arrow-up-from-ground-water: \\e4b5;\n$fa-var-martini-glass: \\f57b;\n$fa-var-glass-martini-alt: \\f57b;\n$fa-var-rotate-left: \\f2ea;\n$fa-var-rotate-back: \\f2ea;\n$fa-var-rotate-backward: \\f2ea;\n$fa-var-undo-alt: \\f2ea;\n$fa-var-table-columns: \\f0db;\n$fa-var-columns: \\f0db;\n$fa-var-lemon: \\f094;\n$fa-var-head-side-mask: \\e063;\n$fa-var-handshake: \\f2b5;\n$fa-var-gem: \\f3a5;\n$fa-var-dolly: \\f472;\n$fa-var-dolly-box: \\f472;\n$fa-var-smoking: \\f48d;\n$fa-var-minimize: \\f78c;\n$fa-var-compress-arrows-alt: \\f78c;\n$fa-var-monument: \\f5a6;\n$fa-var-snowplow: \\f7d2;\n$fa-var-angles-right: \\f101;\n$fa-var-angle-double-right: \\f101;\n$fa-var-cannabis: \\f55f;\n$fa-var-circle-play: \\f144;\n$fa-var-play-circle: \\f144;\n$fa-var-tablets: \\f490;\n$fa-var-ethernet: \\f796;\n$fa-var-euro-sign: \\f153;\n$fa-var-eur: \\f153;\n$fa-var-euro: \\f153;\n$fa-var-chair: \\f6c0;\n$fa-var-circle-check: \\f058;\n$fa-var-check-circle: \\f058;\n$fa-var-circle-stop: \\f28d;\n$fa-var-stop-circle: \\f28d;\n$fa-var-compass-drafting: \\f568;\n$fa-var-drafting-compass: \\f568;\n$fa-var-plate-wheat: \\e55a;\n$fa-var-icicles: \\f7ad;\n$fa-var-person-shelter: \\e54f;\n$fa-var-neuter: \\f22c;\n$fa-var-id-badge: \\f2c1;\n$fa-var-marker: \\f5a1;\n$fa-var-face-laugh-beam: \\f59a;\n$fa-var-laugh-beam: \\f59a;\n$fa-var-helicopter-symbol: \\e502;\n$fa-var-universal-access: \\f29a;\n$fa-var-circle-chevron-up: \\f139;\n$fa-var-chevron-circle-up: \\f139;\n$fa-var-lari-sign: \\e1c8;\n$fa-var-volcano: \\f770;\n$fa-var-person-walking-dashed-line-arrow-right: \\e553;\n$fa-var-sterling-sign: \\f154;\n$fa-var-gbp: \\f154;\n$fa-var-pound-sign: \\f154;\n$fa-var-viruses: \\e076;\n$fa-var-square-person-confined: \\e577;\n$fa-var-user-tie: \\f508;\n$fa-var-arrow-down-long: \\f175;\n$fa-var-long-arrow-down: \\f175;\n$fa-var-tent-arrow-down-to-line: \\e57e;\n$fa-var-certificate: \\f0a3;\n$fa-var-reply-all: \\f122;\n$fa-var-mail-reply-all: \\f122;\n$fa-var-suitcase: \\f0f2;\n$fa-var-person-skating: \\f7c5;\n$fa-var-skating: \\f7c5;\n$fa-var-filter-circle-dollar: \\f662;\n$fa-var-funnel-dollar: \\f662;\n$fa-var-camera-retro: \\f083;\n$fa-var-circle-arrow-down: \\f0ab;\n$fa-var-arrow-circle-down: \\f0ab;\n$fa-var-file-import: \\f56f;\n$fa-var-arrow-right-to-file: \\f56f;\n$fa-var-square-arrow-up-right: \\f14c;\n$fa-var-external-link-square: \\f14c;\n$fa-var-box-open: \\f49e;\n$fa-var-scroll: \\f70e;\n$fa-var-spa: \\f5bb;\n$fa-var-location-pin-lock: \\e51f;\n$fa-var-pause: \\f04c;\n$fa-var-hill-avalanche: \\e507;\n$fa-var-temperature-empty: \\f2cb;\n$fa-var-temperature-0: \\f2cb;\n$fa-var-thermometer-0: \\f2cb;\n$fa-var-thermometer-empty: \\f2cb;\n$fa-var-bomb: \\f1e2;\n$fa-var-registered: \\f25d;\n$fa-var-address-card: \\f2bb;\n$fa-var-contact-card: \\f2bb;\n$fa-var-vcard: \\f2bb;\n$fa-var-scale-unbalanced-flip: \\f516;\n$fa-var-balance-scale-right: \\f516;\n$fa-var-subscript: \\f12c;\n$fa-var-diamond-turn-right: \\f5eb;\n$fa-var-directions: \\f5eb;\n$fa-var-burst: \\e4dc;\n$fa-var-house-laptop: \\e066;\n$fa-var-laptop-house: \\e066;\n$fa-var-face-tired: \\f5c8;\n$fa-var-tired: \\f5c8;\n$fa-var-money-bills: \\e1f3;\n$fa-var-smog: \\f75f;\n$fa-var-crutch: \\f7f7;\n$fa-var-cloud-arrow-up: \\f0ee;\n$fa-var-cloud-upload: \\f0ee;\n$fa-var-cloud-upload-alt: \\f0ee;\n$fa-var-palette: \\f53f;\n$fa-var-arrows-turn-right: \\e4c0;\n$fa-var-vest: \\e085;\n$fa-var-ferry: \\e4ea;\n$fa-var-arrows-down-to-people: \\e4b9;\n$fa-var-seedling: \\f4d8;\n$fa-var-sprout: \\f4d8;\n$fa-var-left-right: \\f337;\n$fa-var-arrows-alt-h: \\f337;\n$fa-var-boxes-packing: \\e4c7;\n$fa-var-circle-arrow-left: \\f0a8;\n$fa-var-arrow-circle-left: \\f0a8;\n$fa-var-group-arrows-rotate: \\e4f6;\n$fa-var-bowl-food: \\e4c6;\n$fa-var-candy-cane: \\f786;\n$fa-var-arrow-down-wide-short: \\f160;\n$fa-var-sort-amount-asc: \\f160;\n$fa-var-sort-amount-down: \\f160;\n$fa-var-cloud-bolt: \\f76c;\n$fa-var-thunderstorm: \\f76c;\n$fa-var-text-slash: \\f87d;\n$fa-var-remove-format: \\f87d;\n$fa-var-face-smile-wink: \\f4da;\n$fa-var-smile-wink: \\f4da;\n$fa-var-file-word: \\f1c2;\n$fa-var-file-powerpoint: \\f1c4;\n$fa-var-arrows-left-right: \\f07e;\n$fa-var-arrows-h: \\f07e;\n$fa-var-house-lock: \\e510;\n$fa-var-cloud-arrow-down: \\f0ed;\n$fa-var-cloud-download: \\f0ed;\n$fa-var-cloud-download-alt: \\f0ed;\n$fa-var-children: \\e4e1;\n$fa-var-chalkboard: \\f51b;\n$fa-var-blackboard: \\f51b;\n$fa-var-user-large-slash: \\f4fa;\n$fa-var-user-alt-slash: \\f4fa;\n$fa-var-envelope-open: \\f2b6;\n$fa-var-handshake-simple-slash: \\e05f;\n$fa-var-handshake-alt-slash: \\e05f;\n$fa-var-mattress-pillow: \\e525;\n$fa-var-guarani-sign: \\e19a;\n$fa-var-arrows-rotate: \\f021;\n$fa-var-refresh: \\f021;\n$fa-var-sync: \\f021;\n$fa-var-fire-extinguisher: \\f134;\n$fa-var-cruzeiro-sign: \\e152;\n$fa-var-greater-than-equal: \\f532;\n$fa-var-shield-halved: \\f3ed;\n$fa-var-shield-alt: \\f3ed;\n$fa-var-book-atlas: \\f558;\n$fa-var-atlas: \\f558;\n$fa-var-virus: \\e074;\n$fa-var-envelope-circle-check: \\e4e8;\n$fa-var-layer-group: \\f5fd;\n$fa-var-arrows-to-dot: \\e4be;\n$fa-var-archway: \\f557;\n$fa-var-heart-circle-check: \\e4fd;\n$fa-var-house-chimney-crack: \\f6f1;\n$fa-var-house-damage: \\f6f1;\n$fa-var-file-zipper: \\f1c6;\n$fa-var-file-archive: \\f1c6;\n$fa-var-square: \\f0c8;\n$fa-var-martini-glass-empty: \\f000;\n$fa-var-glass-martini: \\f000;\n$fa-var-couch: \\f4b8;\n$fa-var-cedi-sign: \\e0df;\n$fa-var-italic: \\f033;\n$fa-var-table-cells-column-lock: \\e678;\n$fa-var-church: \\f51d;\n$fa-var-comments-dollar: \\f653;\n$fa-var-democrat: \\f747;\n$fa-var-z: \\5a;\n$fa-var-person-skiing: \\f7c9;\n$fa-var-skiing: \\f7c9;\n$fa-var-road-lock: \\e567;\n$fa-var-a: \\41;\n$fa-var-temperature-arrow-down: \\e03f;\n$fa-var-temperature-down: \\e03f;\n$fa-var-feather-pointed: \\f56b;\n$fa-var-feather-alt: \\f56b;\n$fa-var-p: \\50;\n$fa-var-snowflake: \\f2dc;\n$fa-var-newspaper: \\f1ea;\n$fa-var-rectangle-ad: \\f641;\n$fa-var-ad: \\f641;\n$fa-var-circle-arrow-right: \\f0a9;\n$fa-var-arrow-circle-right: \\f0a9;\n$fa-var-filter-circle-xmark: \\e17b;\n$fa-var-locust: \\e520;\n$fa-var-sort: \\f0dc;\n$fa-var-unsorted: \\f0dc;\n$fa-var-list-ol: \\f0cb;\n$fa-var-list-1-2: \\f0cb;\n$fa-var-list-numeric: \\f0cb;\n$fa-var-person-dress-burst: \\e544;\n$fa-var-money-check-dollar: \\f53d;\n$fa-var-money-check-alt: \\f53d;\n$fa-var-vector-square: \\f5cb;\n$fa-var-bread-slice: \\f7ec;\n$fa-var-language: \\f1ab;\n$fa-var-face-kiss-wink-heart: \\f598;\n$fa-var-kiss-wink-heart: \\f598;\n$fa-var-filter: \\f0b0;\n$fa-var-question: \\3f;\n$fa-var-file-signature: \\f573;\n$fa-var-up-down-left-right: \\f0b2;\n$fa-var-arrows-alt: \\f0b2;\n$fa-var-house-chimney-user: \\e065;\n$fa-var-hand-holding-heart: \\f4be;\n$fa-var-puzzle-piece: \\f12e;\n$fa-var-money-check: \\f53c;\n$fa-var-star-half-stroke: \\f5c0;\n$fa-var-star-half-alt: \\f5c0;\n$fa-var-code: \\f121;\n$fa-var-whiskey-glass: \\f7a0;\n$fa-var-glass-whiskey: \\f7a0;\n$fa-var-building-circle-exclamation: \\e4d3;\n$fa-var-magnifying-glass-chart: \\e522;\n$fa-var-arrow-up-right-from-square: \\f08e;\n$fa-var-external-link: \\f08e;\n$fa-var-cubes-stacked: \\e4e6;\n$fa-var-won-sign: \\f159;\n$fa-var-krw: \\f159;\n$fa-var-won: \\f159;\n$fa-var-virus-covid: \\e4a8;\n$fa-var-austral-sign: \\e0a9;\n$fa-var-f: \\46;\n$fa-var-leaf: \\f06c;\n$fa-var-road: \\f018;\n$fa-var-taxi: \\f1ba;\n$fa-var-cab: \\f1ba;\n$fa-var-person-circle-plus: \\e541;\n$fa-var-chart-pie: \\f200;\n$fa-var-pie-chart: \\f200;\n$fa-var-bolt-lightning: \\e0b7;\n$fa-var-sack-xmark: \\e56a;\n$fa-var-file-excel: \\f1c3;\n$fa-var-file-contract: \\f56c;\n$fa-var-fish-fins: \\e4f2;\n$fa-var-building-flag: \\e4d5;\n$fa-var-face-grin-beam: \\f582;\n$fa-var-grin-beam: \\f582;\n$fa-var-object-ungroup: \\f248;\n$fa-var-poop: \\f619;\n$fa-var-location-pin: \\f041;\n$fa-var-map-marker: \\f041;\n$fa-var-kaaba: \\f66b;\n$fa-var-toilet-paper: \\f71e;\n$fa-var-helmet-safety: \\f807;\n$fa-var-hard-hat: \\f807;\n$fa-var-hat-hard: \\f807;\n$fa-var-eject: \\f052;\n$fa-var-circle-right: \\f35a;\n$fa-var-arrow-alt-circle-right: \\f35a;\n$fa-var-plane-circle-check: \\e555;\n$fa-var-face-rolling-eyes: \\f5a5;\n$fa-var-meh-rolling-eyes: \\f5a5;\n$fa-var-object-group: \\f247;\n$fa-var-chart-line: \\f201;\n$fa-var-line-chart: \\f201;\n$fa-var-mask-ventilator: \\e524;\n$fa-var-arrow-right: \\f061;\n$fa-var-signs-post: \\f277;\n$fa-var-map-signs: \\f277;\n$fa-var-cash-register: \\f788;\n$fa-var-person-circle-question: \\e542;\n$fa-var-h: \\48;\n$fa-var-tarp: \\e57b;\n$fa-var-screwdriver-wrench: \\f7d9;\n$fa-var-tools: \\f7d9;\n$fa-var-arrows-to-eye: \\e4bf;\n$fa-var-plug-circle-bolt: \\e55b;\n$fa-var-heart: \\f004;\n$fa-var-mars-and-venus: \\f224;\n$fa-var-house-user: \\e1b0;\n$fa-var-home-user: \\e1b0;\n$fa-var-dumpster-fire: \\f794;\n$fa-var-house-crack: \\e3b1;\n$fa-var-martini-glass-citrus: \\f561;\n$fa-var-cocktail: \\f561;\n$fa-var-face-surprise: \\f5c2;\n$fa-var-surprise: \\f5c2;\n$fa-var-bottle-water: \\e4c5;\n$fa-var-circle-pause: \\f28b;\n$fa-var-pause-circle: \\f28b;\n$fa-var-toilet-paper-slash: \\e072;\n$fa-var-apple-whole: \\f5d1;\n$fa-var-apple-alt: \\f5d1;\n$fa-var-kitchen-set: \\e51a;\n$fa-var-r: \\52;\n$fa-var-temperature-quarter: \\f2ca;\n$fa-var-temperature-1: \\f2ca;\n$fa-var-thermometer-1: \\f2ca;\n$fa-var-thermometer-quarter: \\f2ca;\n$fa-var-cube: \\f1b2;\n$fa-var-bitcoin-sign: \\e0b4;\n$fa-var-shield-dog: \\e573;\n$fa-var-solar-panel: \\f5ba;\n$fa-var-lock-open: \\f3c1;\n$fa-var-elevator: \\e16d;\n$fa-var-money-bill-transfer: \\e528;\n$fa-var-money-bill-trend-up: \\e529;\n$fa-var-house-flood-water-circle-arrow-right: \\e50f;\n$fa-var-square-poll-horizontal: \\f682;\n$fa-var-poll-h: \\f682;\n$fa-var-circle: \\f111;\n$fa-var-backward-fast: \\f049;\n$fa-var-fast-backward: \\f049;\n$fa-var-recycle: \\f1b8;\n$fa-var-user-astronaut: \\f4fb;\n$fa-var-plane-slash: \\e069;\n$fa-var-trademark: \\f25c;\n$fa-var-basketball: \\f434;\n$fa-var-basketball-ball: \\f434;\n$fa-var-satellite-dish: \\f7c0;\n$fa-var-circle-up: \\f35b;\n$fa-var-arrow-alt-circle-up: \\f35b;\n$fa-var-mobile-screen-button: \\f3cd;\n$fa-var-mobile-alt: \\f3cd;\n$fa-var-volume-high: \\f028;\n$fa-var-volume-up: \\f028;\n$fa-var-users-rays: \\e593;\n$fa-var-wallet: \\f555;\n$fa-var-clipboard-check: \\f46c;\n$fa-var-file-audio: \\f1c7;\n$fa-var-burger: \\f805;\n$fa-var-hamburger: \\f805;\n$fa-var-wrench: \\f0ad;\n$fa-var-bugs: \\e4d0;\n$fa-var-rupee-sign: \\f156;\n$fa-var-rupee: \\f156;\n$fa-var-file-image: \\f1c5;\n$fa-var-circle-question: \\f059;\n$fa-var-question-circle: \\f059;\n$fa-var-plane-departure: \\f5b0;\n$fa-var-handshake-slash: \\e060;\n$fa-var-book-bookmark: \\e0bb;\n$fa-var-code-branch: \\f126;\n$fa-var-hat-cowboy: \\f8c0;\n$fa-var-bridge: \\e4c8;\n$fa-var-phone-flip: \\f879;\n$fa-var-phone-alt: \\f879;\n$fa-var-truck-front: \\e2b7;\n$fa-var-cat: \\f6be;\n$fa-var-anchor-circle-exclamation: \\e4ab;\n$fa-var-truck-field: \\e58d;\n$fa-var-route: \\f4d7;\n$fa-var-clipboard-question: \\e4e3;\n$fa-var-panorama: \\e209;\n$fa-var-comment-medical: \\f7f5;\n$fa-var-teeth-open: \\f62f;\n$fa-var-file-circle-minus: \\e4ed;\n$fa-var-tags: \\f02c;\n$fa-var-wine-glass: \\f4e3;\n$fa-var-forward-fast: \\f050;\n$fa-var-fast-forward: \\f050;\n$fa-var-face-meh-blank: \\f5a4;\n$fa-var-meh-blank: \\f5a4;\n$fa-var-square-parking: \\f540;\n$fa-var-parking: \\f540;\n$fa-var-house-signal: \\e012;\n$fa-var-bars-progress: \\f828;\n$fa-var-tasks-alt: \\f828;\n$fa-var-faucet-drip: \\e006;\n$fa-var-cart-flatbed: \\f474;\n$fa-var-dolly-flatbed: \\f474;\n$fa-var-ban-smoking: \\f54d;\n$fa-var-smoking-ban: \\f54d;\n$fa-var-terminal: \\f120;\n$fa-var-mobile-button: \\f10b;\n$fa-var-house-medical-flag: \\e514;\n$fa-var-basket-shopping: \\f291;\n$fa-var-shopping-basket: \\f291;\n$fa-var-tape: \\f4db;\n$fa-var-bus-simple: \\f55e;\n$fa-var-bus-alt: \\f55e;\n$fa-var-eye: \\f06e;\n$fa-var-face-sad-cry: \\f5b3;\n$fa-var-sad-cry: \\f5b3;\n$fa-var-audio-description: \\f29e;\n$fa-var-person-military-to-person: \\e54c;\n$fa-var-file-shield: \\e4f0;\n$fa-var-user-slash: \\f506;\n$fa-var-pen: \\f304;\n$fa-var-tower-observation: \\e586;\n$fa-var-file-code: \\f1c9;\n$fa-var-signal: \\f012;\n$fa-var-signal-5: \\f012;\n$fa-var-signal-perfect: \\f012;\n$fa-var-bus: \\f207;\n$fa-var-heart-circle-xmark: \\e501;\n$fa-var-house-chimney: \\e3af;\n$fa-var-home-lg: \\e3af;\n$fa-var-window-maximize: \\f2d0;\n$fa-var-face-frown: \\f119;\n$fa-var-frown: \\f119;\n$fa-var-prescription: \\f5b1;\n$fa-var-shop: \\f54f;\n$fa-var-store-alt: \\f54f;\n$fa-var-floppy-disk: \\f0c7;\n$fa-var-save: \\f0c7;\n$fa-var-vihara: \\f6a7;\n$fa-var-scale-unbalanced: \\f515;\n$fa-var-balance-scale-left: \\f515;\n$fa-var-sort-up: \\f0de;\n$fa-var-sort-asc: \\f0de;\n$fa-var-comment-dots: \\f4ad;\n$fa-var-commenting: \\f4ad;\n$fa-var-plant-wilt: \\e5aa;\n$fa-var-diamond: \\f219;\n$fa-var-face-grin-squint: \\f585;\n$fa-var-grin-squint: \\f585;\n$fa-var-hand-holding-dollar: \\f4c0;\n$fa-var-hand-holding-usd: \\f4c0;\n$fa-var-bacterium: \\e05a;\n$fa-var-hand-pointer: \\f25a;\n$fa-var-drum-steelpan: \\f56a;\n$fa-var-hand-scissors: \\f257;\n$fa-var-hands-praying: \\f684;\n$fa-var-praying-hands: \\f684;\n$fa-var-arrow-rotate-right: \\f01e;\n$fa-var-arrow-right-rotate: \\f01e;\n$fa-var-arrow-rotate-forward: \\f01e;\n$fa-var-redo: \\f01e;\n$fa-var-biohazard: \\f780;\n$fa-var-location-crosshairs: \\f601;\n$fa-var-location: \\f601;\n$fa-var-mars-double: \\f227;\n$fa-var-child-dress: \\e59c;\n$fa-var-users-between-lines: \\e591;\n$fa-var-lungs-virus: \\e067;\n$fa-var-face-grin-tears: \\f588;\n$fa-var-grin-tears: \\f588;\n$fa-var-phone: \\f095;\n$fa-var-calendar-xmark: \\f273;\n$fa-var-calendar-times: \\f273;\n$fa-var-child-reaching: \\e59d;\n$fa-var-head-side-virus: \\e064;\n$fa-var-user-gear: \\f4fe;\n$fa-var-user-cog: \\f4fe;\n$fa-var-arrow-up-1-9: \\f163;\n$fa-var-sort-numeric-up: \\f163;\n$fa-var-door-closed: \\f52a;\n$fa-var-shield-virus: \\e06c;\n$fa-var-dice-six: \\f526;\n$fa-var-mosquito-net: \\e52c;\n$fa-var-bridge-water: \\e4ce;\n$fa-var-person-booth: \\f756;\n$fa-var-text-width: \\f035;\n$fa-var-hat-wizard: \\f6e8;\n$fa-var-pen-fancy: \\f5ac;\n$fa-var-person-digging: \\f85e;\n$fa-var-digging: \\f85e;\n$fa-var-trash: \\f1f8;\n$fa-var-gauge-simple: \\f629;\n$fa-var-gauge-simple-med: \\f629;\n$fa-var-tachometer-average: \\f629;\n$fa-var-book-medical: \\f7e6;\n$fa-var-poo: \\f2fe;\n$fa-var-quote-right: \\f10e;\n$fa-var-quote-right-alt: \\f10e;\n$fa-var-shirt: \\f553;\n$fa-var-t-shirt: \\f553;\n$fa-var-tshirt: \\f553;\n$fa-var-cubes: \\f1b3;\n$fa-var-divide: \\f529;\n$fa-var-tenge-sign: \\f7d7;\n$fa-var-tenge: \\f7d7;\n$fa-var-headphones: \\f025;\n$fa-var-hands-holding: \\f4c2;\n$fa-var-hands-clapping: \\e1a8;\n$fa-var-republican: \\f75e;\n$fa-var-arrow-left: \\f060;\n$fa-var-person-circle-xmark: \\e543;\n$fa-var-ruler: \\f545;\n$fa-var-align-left: \\f036;\n$fa-var-dice-d6: \\f6d1;\n$fa-var-restroom: \\f7bd;\n$fa-var-j: \\4a;\n$fa-var-users-viewfinder: \\e595;\n$fa-var-file-video: \\f1c8;\n$fa-var-up-right-from-square: \\f35d;\n$fa-var-external-link-alt: \\f35d;\n$fa-var-table-cells: \\f00a;\n$fa-var-th: \\f00a;\n$fa-var-file-pdf: \\f1c1;\n$fa-var-book-bible: \\f647;\n$fa-var-bible: \\f647;\n$fa-var-o: \\4f;\n$fa-var-suitcase-medical: \\f0fa;\n$fa-var-medkit: \\f0fa;\n$fa-var-user-secret: \\f21b;\n$fa-var-otter: \\f700;\n$fa-var-person-dress: \\f182;\n$fa-var-female: \\f182;\n$fa-var-comment-dollar: \\f651;\n$fa-var-business-time: \\f64a;\n$fa-var-briefcase-clock: \\f64a;\n$fa-var-table-cells-large: \\f009;\n$fa-var-th-large: \\f009;\n$fa-var-book-tanakh: \\f827;\n$fa-var-tanakh: \\f827;\n$fa-var-phone-volume: \\f2a0;\n$fa-var-volume-control-phone: \\f2a0;\n$fa-var-hat-cowboy-side: \\f8c1;\n$fa-var-clipboard-user: \\f7f3;\n$fa-var-child: \\f1ae;\n$fa-var-lira-sign: \\f195;\n$fa-var-satellite: \\f7bf;\n$fa-var-plane-lock: \\e558;\n$fa-var-tag: \\f02b;\n$fa-var-comment: \\f075;\n$fa-var-cake-candles: \\f1fd;\n$fa-var-birthday-cake: \\f1fd;\n$fa-var-cake: \\f1fd;\n$fa-var-envelope: \\f0e0;\n$fa-var-angles-up: \\f102;\n$fa-var-angle-double-up: \\f102;\n$fa-var-paperclip: \\f0c6;\n$fa-var-arrow-right-to-city: \\e4b3;\n$fa-var-ribbon: \\f4d6;\n$fa-var-lungs: \\f604;\n$fa-var-arrow-up-9-1: \\f887;\n$fa-var-sort-numeric-up-alt: \\f887;\n$fa-var-litecoin-sign: \\e1d3;\n$fa-var-border-none: \\f850;\n$fa-var-circle-nodes: \\e4e2;\n$fa-var-parachute-box: \\f4cd;\n$fa-var-indent: \\f03c;\n$fa-var-truck-field-un: \\e58e;\n$fa-var-hourglass: \\f254;\n$fa-var-hourglass-empty: \\f254;\n$fa-var-mountain: \\f6fc;\n$fa-var-user-doctor: \\f0f0;\n$fa-var-user-md: \\f0f0;\n$fa-var-circle-info: \\f05a;\n$fa-var-info-circle: \\f05a;\n$fa-var-cloud-meatball: \\f73b;\n$fa-var-camera: \\f030;\n$fa-var-camera-alt: \\f030;\n$fa-var-square-virus: \\e578;\n$fa-var-meteor: \\f753;\n$fa-var-car-on: \\e4dd;\n$fa-var-sleigh: \\f7cc;\n$fa-var-arrow-down-1-9: \\f162;\n$fa-var-sort-numeric-asc: \\f162;\n$fa-var-sort-numeric-down: \\f162;\n$fa-var-hand-holding-droplet: \\f4c1;\n$fa-var-hand-holding-water: \\f4c1;\n$fa-var-water: \\f773;\n$fa-var-calendar-check: \\f274;\n$fa-var-braille: \\f2a1;\n$fa-var-prescription-bottle-medical: \\f486;\n$fa-var-prescription-bottle-alt: \\f486;\n$fa-var-landmark: \\f66f;\n$fa-var-truck: \\f0d1;\n$fa-var-crosshairs: \\f05b;\n$fa-var-person-cane: \\e53c;\n$fa-var-tent: \\e57d;\n$fa-var-vest-patches: \\e086;\n$fa-var-check-double: \\f560;\n$fa-var-arrow-down-a-z: \\f15d;\n$fa-var-sort-alpha-asc: \\f15d;\n$fa-var-sort-alpha-down: \\f15d;\n$fa-var-money-bill-wheat: \\e52a;\n$fa-var-cookie: \\f563;\n$fa-var-arrow-rotate-left: \\f0e2;\n$fa-var-arrow-left-rotate: \\f0e2;\n$fa-var-arrow-rotate-back: \\f0e2;\n$fa-var-arrow-rotate-backward: \\f0e2;\n$fa-var-undo: \\f0e2;\n$fa-var-hard-drive: \\f0a0;\n$fa-var-hdd: \\f0a0;\n$fa-var-face-grin-squint-tears: \\f586;\n$fa-var-grin-squint-tears: \\f586;\n$fa-var-dumbbell: \\f44b;\n$fa-var-rectangle-list: \\f022;\n$fa-var-list-alt: \\f022;\n$fa-var-tarp-droplet: \\e57c;\n$fa-var-house-medical-circle-check: \\e511;\n$fa-var-person-skiing-nordic: \\f7ca;\n$fa-var-skiing-nordic: \\f7ca;\n$fa-var-calendar-plus: \\f271;\n$fa-var-plane-arrival: \\f5af;\n$fa-var-circle-left: \\f359;\n$fa-var-arrow-alt-circle-left: \\f359;\n$fa-var-train-subway: \\f239;\n$fa-var-subway: \\f239;\n$fa-var-chart-gantt: \\e0e4;\n$fa-var-indian-rupee-sign: \\e1bc;\n$fa-var-indian-rupee: \\e1bc;\n$fa-var-inr: \\e1bc;\n$fa-var-crop-simple: \\f565;\n$fa-var-crop-alt: \\f565;\n$fa-var-money-bill-1: \\f3d1;\n$fa-var-money-bill-alt: \\f3d1;\n$fa-var-left-long: \\f30a;\n$fa-var-long-arrow-alt-left: \\f30a;\n$fa-var-dna: \\f471;\n$fa-var-virus-slash: \\e075;\n$fa-var-minus: \\f068;\n$fa-var-subtract: \\f068;\n$fa-var-chess: \\f439;\n$fa-var-arrow-left-long: \\f177;\n$fa-var-long-arrow-left: \\f177;\n$fa-var-plug-circle-check: \\e55c;\n$fa-var-street-view: \\f21d;\n$fa-var-franc-sign: \\e18f;\n$fa-var-volume-off: \\f026;\n$fa-var-hands-asl-interpreting: \\f2a3;\n$fa-var-american-sign-language-interpreting: \\f2a3;\n$fa-var-asl-interpreting: \\f2a3;\n$fa-var-hands-american-sign-language-interpreting: \\f2a3;\n$fa-var-gear: \\f013;\n$fa-var-cog: \\f013;\n$fa-var-droplet-slash: \\f5c7;\n$fa-var-tint-slash: \\f5c7;\n$fa-var-mosque: \\f678;\n$fa-var-mosquito: \\e52b;\n$fa-var-star-of-david: \\f69a;\n$fa-var-person-military-rifle: \\e54b;\n$fa-var-cart-shopping: \\f07a;\n$fa-var-shopping-cart: \\f07a;\n$fa-var-vials: \\f493;\n$fa-var-plug-circle-plus: \\e55f;\n$fa-var-place-of-worship: \\f67f;\n$fa-var-grip-vertical: \\f58e;\n$fa-var-arrow-turn-up: \\f148;\n$fa-var-level-up: \\f148;\n$fa-var-u: \\55;\n$fa-var-square-root-variable: \\f698;\n$fa-var-square-root-alt: \\f698;\n$fa-var-clock: \\f017;\n$fa-var-clock-four: \\f017;\n$fa-var-backward-step: \\f048;\n$fa-var-step-backward: \\f048;\n$fa-var-pallet: \\f482;\n$fa-var-faucet: \\e005;\n$fa-var-baseball-bat-ball: \\f432;\n$fa-var-s: \\53;\n$fa-var-timeline: \\e29c;\n$fa-var-keyboard: \\f11c;\n$fa-var-caret-down: \\f0d7;\n$fa-var-house-chimney-medical: \\f7f2;\n$fa-var-clinic-medical: \\f7f2;\n$fa-var-temperature-three-quarters: \\f2c8;\n$fa-var-temperature-3: \\f2c8;\n$fa-var-thermometer-3: \\f2c8;\n$fa-var-thermometer-three-quarters: \\f2c8;\n$fa-var-mobile-screen: \\f3cf;\n$fa-var-mobile-android-alt: \\f3cf;\n$fa-var-plane-up: \\e22d;\n$fa-var-piggy-bank: \\f4d3;\n$fa-var-battery-half: \\f242;\n$fa-var-battery-3: \\f242;\n$fa-var-mountain-city: \\e52e;\n$fa-var-coins: \\f51e;\n$fa-var-khanda: \\f66d;\n$fa-var-sliders: \\f1de;\n$fa-var-sliders-h: \\f1de;\n$fa-var-folder-tree: \\f802;\n$fa-var-network-wired: \\f6ff;\n$fa-var-map-pin: \\f276;\n$fa-var-hamsa: \\f665;\n$fa-var-cent-sign: \\e3f5;\n$fa-var-flask: \\f0c3;\n$fa-var-person-pregnant: \\e31e;\n$fa-var-wand-sparkles: \\f72b;\n$fa-var-ellipsis-vertical: \\f142;\n$fa-var-ellipsis-v: \\f142;\n$fa-var-ticket: \\f145;\n$fa-var-power-off: \\f011;\n$fa-var-right-long: \\f30b;\n$fa-var-long-arrow-alt-right: \\f30b;\n$fa-var-flag-usa: \\f74d;\n$fa-var-laptop-file: \\e51d;\n$fa-var-tty: \\f1e4;\n$fa-var-teletype: \\f1e4;\n$fa-var-diagram-next: \\e476;\n$fa-var-person-rifle: \\e54e;\n$fa-var-house-medical-circle-exclamation: \\e512;\n$fa-var-closed-captioning: \\f20a;\n$fa-var-person-hiking: \\f6ec;\n$fa-var-hiking: \\f6ec;\n$fa-var-venus-double: \\f226;\n$fa-var-images: \\f302;\n$fa-var-calculator: \\f1ec;\n$fa-var-people-pulling: \\e535;\n$fa-var-n: \\4e;\n$fa-var-cable-car: \\f7da;\n$fa-var-tram: \\f7da;\n$fa-var-cloud-rain: \\f73d;\n$fa-var-building-circle-xmark: \\e4d4;\n$fa-var-ship: \\f21a;\n$fa-var-arrows-down-to-line: \\e4b8;\n$fa-var-download: \\f019;\n$fa-var-face-grin: \\f580;\n$fa-var-grin: \\f580;\n$fa-var-delete-left: \\f55a;\n$fa-var-backspace: \\f55a;\n$fa-var-eye-dropper: \\f1fb;\n$fa-var-eye-dropper-empty: \\f1fb;\n$fa-var-eyedropper: \\f1fb;\n$fa-var-file-circle-check: \\e5a0;\n$fa-var-forward: \\f04e;\n$fa-var-mobile: \\f3ce;\n$fa-var-mobile-android: \\f3ce;\n$fa-var-mobile-phone: \\f3ce;\n$fa-var-face-meh: \\f11a;\n$fa-var-meh: \\f11a;\n$fa-var-align-center: \\f037;\n$fa-var-book-skull: \\f6b7;\n$fa-var-book-dead: \\f6b7;\n$fa-var-id-card: \\f2c2;\n$fa-var-drivers-license: \\f2c2;\n$fa-var-outdent: \\f03b;\n$fa-var-dedent: \\f03b;\n$fa-var-heart-circle-exclamation: \\e4fe;\n$fa-var-house: \\f015;\n$fa-var-home: \\f015;\n$fa-var-home-alt: \\f015;\n$fa-var-home-lg-alt: \\f015;\n$fa-var-calendar-week: \\f784;\n$fa-var-laptop-medical: \\f812;\n$fa-var-b: \\42;\n$fa-var-file-medical: \\f477;\n$fa-var-dice-one: \\f525;\n$fa-var-kiwi-bird: \\f535;\n$fa-var-arrow-right-arrow-left: \\f0ec;\n$fa-var-exchange: \\f0ec;\n$fa-var-rotate-right: \\f2f9;\n$fa-var-redo-alt: \\f2f9;\n$fa-var-rotate-forward: \\f2f9;\n$fa-var-utensils: \\f2e7;\n$fa-var-cutlery: \\f2e7;\n$fa-var-arrow-up-wide-short: \\f161;\n$fa-var-sort-amount-up: \\f161;\n$fa-var-mill-sign: \\e1ed;\n$fa-var-bowl-rice: \\e2eb;\n$fa-var-skull: \\f54c;\n$fa-var-tower-broadcast: \\f519;\n$fa-var-broadcast-tower: \\f519;\n$fa-var-truck-pickup: \\f63c;\n$fa-var-up-long: \\f30c;\n$fa-var-long-arrow-alt-up: \\f30c;\n$fa-var-stop: \\f04d;\n$fa-var-code-merge: \\f387;\n$fa-var-upload: \\f093;\n$fa-var-hurricane: \\f751;\n$fa-var-mound: \\e52d;\n$fa-var-toilet-portable: \\e583;\n$fa-var-compact-disc: \\f51f;\n$fa-var-file-arrow-down: \\f56d;\n$fa-var-file-download: \\f56d;\n$fa-var-caravan: \\f8ff;\n$fa-var-shield-cat: \\e572;\n$fa-var-bolt: \\f0e7;\n$fa-var-zap: \\f0e7;\n$fa-var-glass-water: \\e4f4;\n$fa-var-oil-well: \\e532;\n$fa-var-vault: \\e2c5;\n$fa-var-mars: \\f222;\n$fa-var-toilet: \\f7d8;\n$fa-var-plane-circle-xmark: \\e557;\n$fa-var-yen-sign: \\f157;\n$fa-var-cny: \\f157;\n$fa-var-jpy: \\f157;\n$fa-var-rmb: \\f157;\n$fa-var-yen: \\f157;\n$fa-var-ruble-sign: \\f158;\n$fa-var-rouble: \\f158;\n$fa-var-rub: \\f158;\n$fa-var-ruble: \\f158;\n$fa-var-sun: \\f185;\n$fa-var-guitar: \\f7a6;\n$fa-var-face-laugh-wink: \\f59c;\n$fa-var-laugh-wink: \\f59c;\n$fa-var-horse-head: \\f7ab;\n$fa-var-bore-hole: \\e4c3;\n$fa-var-industry: \\f275;\n$fa-var-circle-down: \\f358;\n$fa-var-arrow-alt-circle-down: \\f358;\n$fa-var-arrows-turn-to-dots: \\e4c1;\n$fa-var-florin-sign: \\e184;\n$fa-var-arrow-down-short-wide: \\f884;\n$fa-var-sort-amount-desc: \\f884;\n$fa-var-sort-amount-down-alt: \\f884;\n$fa-var-less-than: \\3c;\n$fa-var-angle-down: \\f107;\n$fa-var-car-tunnel: \\e4de;\n$fa-var-head-side-cough: \\e061;\n$fa-var-grip-lines: \\f7a4;\n$fa-var-thumbs-down: \\f165;\n$fa-var-user-lock: \\f502;\n$fa-var-arrow-right-long: \\f178;\n$fa-var-long-arrow-right: \\f178;\n$fa-var-anchor-circle-xmark: \\e4ac;\n$fa-var-ellipsis: \\f141;\n$fa-var-ellipsis-h: \\f141;\n$fa-var-chess-pawn: \\f443;\n$fa-var-kit-medical: \\f479;\n$fa-var-first-aid: \\f479;\n$fa-var-person-through-window: \\e5a9;\n$fa-var-toolbox: \\f552;\n$fa-var-hands-holding-circle: \\e4fb;\n$fa-var-bug: \\f188;\n$fa-var-credit-card: \\f09d;\n$fa-var-credit-card-alt: \\f09d;\n$fa-var-car: \\f1b9;\n$fa-var-automobile: \\f1b9;\n$fa-var-hand-holding-hand: \\e4f7;\n$fa-var-book-open-reader: \\f5da;\n$fa-var-book-reader: \\f5da;\n$fa-var-mountain-sun: \\e52f;\n$fa-var-arrows-left-right-to-line: \\e4ba;\n$fa-var-dice-d20: \\f6cf;\n$fa-var-truck-droplet: \\e58c;\n$fa-var-file-circle-xmark: \\e5a1;\n$fa-var-temperature-arrow-up: \\e040;\n$fa-var-temperature-up: \\e040;\n$fa-var-medal: \\f5a2;\n$fa-var-bed: \\f236;\n$fa-var-square-h: \\f0fd;\n$fa-var-h-square: \\f0fd;\n$fa-var-podcast: \\f2ce;\n$fa-var-temperature-full: \\f2c7;\n$fa-var-temperature-4: \\f2c7;\n$fa-var-thermometer-4: \\f2c7;\n$fa-var-thermometer-full: \\f2c7;\n$fa-var-bell: \\f0f3;\n$fa-var-superscript: \\f12b;\n$fa-var-plug-circle-xmark: \\e560;\n$fa-var-star-of-life: \\f621;\n$fa-var-phone-slash: \\f3dd;\n$fa-var-paint-roller: \\f5aa;\n$fa-var-handshake-angle: \\f4c4;\n$fa-var-hands-helping: \\f4c4;\n$fa-var-location-dot: \\f3c5;\n$fa-var-map-marker-alt: \\f3c5;\n$fa-var-file: \\f15b;\n$fa-var-greater-than: \\3e;\n$fa-var-person-swimming: \\f5c4;\n$fa-var-swimmer: \\f5c4;\n$fa-var-arrow-down: \\f063;\n$fa-var-droplet: \\f043;\n$fa-var-tint: \\f043;\n$fa-var-eraser: \\f12d;\n$fa-var-earth-americas: \\f57d;\n$fa-var-earth: \\f57d;\n$fa-var-earth-america: \\f57d;\n$fa-var-globe-americas: \\f57d;\n$fa-var-person-burst: \\e53b;\n$fa-var-dove: \\f4ba;\n$fa-var-battery-empty: \\f244;\n$fa-var-battery-0: \\f244;\n$fa-var-socks: \\f696;\n$fa-var-inbox: \\f01c;\n$fa-var-section: \\e447;\n$fa-var-gauge-high: \\f625;\n$fa-var-tachometer-alt: \\f625;\n$fa-var-tachometer-alt-fast: \\f625;\n$fa-var-envelope-open-text: \\f658;\n$fa-var-hospital: \\f0f8;\n$fa-var-hospital-alt: \\f0f8;\n$fa-var-hospital-wide: \\f0f8;\n$fa-var-wine-bottle: \\f72f;\n$fa-var-chess-rook: \\f447;\n$fa-var-bars-staggered: \\f550;\n$fa-var-reorder: \\f550;\n$fa-var-stream: \\f550;\n$fa-var-dharmachakra: \\f655;\n$fa-var-hotdog: \\f80f;\n$fa-var-person-walking-with-cane: \\f29d;\n$fa-var-blind: \\f29d;\n$fa-var-drum: \\f569;\n$fa-var-ice-cream: \\f810;\n$fa-var-heart-circle-bolt: \\e4fc;\n$fa-var-fax: \\f1ac;\n$fa-var-paragraph: \\f1dd;\n$fa-var-check-to-slot: \\f772;\n$fa-var-vote-yea: \\f772;\n$fa-var-star-half: \\f089;\n$fa-var-boxes-stacked: \\f468;\n$fa-var-boxes: \\f468;\n$fa-var-boxes-alt: \\f468;\n$fa-var-link: \\f0c1;\n$fa-var-chain: \\f0c1;\n$fa-var-ear-listen: \\f2a2;\n$fa-var-assistive-listening-systems: \\f2a2;\n$fa-var-tree-city: \\e587;\n$fa-var-play: \\f04b;\n$fa-var-font: \\f031;\n$fa-var-table-cells-row-lock: \\e67a;\n$fa-var-rupiah-sign: \\e23d;\n$fa-var-magnifying-glass: \\f002;\n$fa-var-search: \\f002;\n$fa-var-table-tennis-paddle-ball: \\f45d;\n$fa-var-ping-pong-paddle-ball: \\f45d;\n$fa-var-table-tennis: \\f45d;\n$fa-var-person-dots-from-line: \\f470;\n$fa-var-diagnoses: \\f470;\n$fa-var-trash-can-arrow-up: \\f82a;\n$fa-var-trash-restore-alt: \\f82a;\n$fa-var-naira-sign: \\e1f6;\n$fa-var-cart-arrow-down: \\f218;\n$fa-var-walkie-talkie: \\f8ef;\n$fa-var-file-pen: \\f31c;\n$fa-var-file-edit: \\f31c;\n$fa-var-receipt: \\f543;\n$fa-var-square-pen: \\f14b;\n$fa-var-pen-square: \\f14b;\n$fa-var-pencil-square: \\f14b;\n$fa-var-suitcase-rolling: \\f5c1;\n$fa-var-person-circle-exclamation: \\e53f;\n$fa-var-chevron-down: \\f078;\n$fa-var-battery-full: \\f240;\n$fa-var-battery: \\f240;\n$fa-var-battery-5: \\f240;\n$fa-var-skull-crossbones: \\f714;\n$fa-var-code-compare: \\e13a;\n$fa-var-list-ul: \\f0ca;\n$fa-var-list-dots: \\f0ca;\n$fa-var-school-lock: \\e56f;\n$fa-var-tower-cell: \\e585;\n$fa-var-down-long: \\f309;\n$fa-var-long-arrow-alt-down: \\f309;\n$fa-var-ranking-star: \\e561;\n$fa-var-chess-king: \\f43f;\n$fa-var-person-harassing: \\e549;\n$fa-var-brazilian-real-sign: \\e46c;\n$fa-var-landmark-dome: \\f752;\n$fa-var-landmark-alt: \\f752;\n$fa-var-arrow-up: \\f062;\n$fa-var-tv: \\f26c;\n$fa-var-television: \\f26c;\n$fa-var-tv-alt: \\f26c;\n$fa-var-shrimp: \\e448;\n$fa-var-list-check: \\f0ae;\n$fa-var-tasks: \\f0ae;\n$fa-var-jug-detergent: \\e519;\n$fa-var-circle-user: \\f2bd;\n$fa-var-user-circle: \\f2bd;\n$fa-var-user-shield: \\f505;\n$fa-var-wind: \\f72e;\n$fa-var-car-burst: \\f5e1;\n$fa-var-car-crash: \\f5e1;\n$fa-var-y: \\59;\n$fa-var-person-snowboarding: \\f7ce;\n$fa-var-snowboarding: \\f7ce;\n$fa-var-truck-fast: \\f48b;\n$fa-var-shipping-fast: \\f48b;\n$fa-var-fish: \\f578;\n$fa-var-user-graduate: \\f501;\n$fa-var-circle-half-stroke: \\f042;\n$fa-var-adjust: \\f042;\n$fa-var-clapperboard: \\e131;\n$fa-var-circle-radiation: \\f7ba;\n$fa-var-radiation-alt: \\f7ba;\n$fa-var-baseball: \\f433;\n$fa-var-baseball-ball: \\f433;\n$fa-var-jet-fighter-up: \\e518;\n$fa-var-diagram-project: \\f542;\n$fa-var-project-diagram: \\f542;\n$fa-var-copy: \\f0c5;\n$fa-var-volume-xmark: \\f6a9;\n$fa-var-volume-mute: \\f6a9;\n$fa-var-volume-times: \\f6a9;\n$fa-var-hand-sparkles: \\e05d;\n$fa-var-grip: \\f58d;\n$fa-var-grip-horizontal: \\f58d;\n$fa-var-share-from-square: \\f14d;\n$fa-var-share-square: \\f14d;\n$fa-var-child-combatant: \\e4e0;\n$fa-var-child-rifle: \\e4e0;\n$fa-var-gun: \\e19b;\n$fa-var-square-phone: \\f098;\n$fa-var-phone-square: \\f098;\n$fa-var-plus: \\2b;\n$fa-var-add: \\2b;\n$fa-var-expand: \\f065;\n$fa-var-computer: \\e4e5;\n$fa-var-xmark: \\f00d;\n$fa-var-close: \\f00d;\n$fa-var-multiply: \\f00d;\n$fa-var-remove: \\f00d;\n$fa-var-times: \\f00d;\n$fa-var-arrows-up-down-left-right: \\f047;\n$fa-var-arrows: \\f047;\n$fa-var-chalkboard-user: \\f51c;\n$fa-var-chalkboard-teacher: \\f51c;\n$fa-var-peso-sign: \\e222;\n$fa-var-building-shield: \\e4d8;\n$fa-var-baby: \\f77c;\n$fa-var-users-line: \\e592;\n$fa-var-quote-left: \\f10d;\n$fa-var-quote-left-alt: \\f10d;\n$fa-var-tractor: \\f722;\n$fa-var-trash-arrow-up: \\f829;\n$fa-var-trash-restore: \\f829;\n$fa-var-arrow-down-up-lock: \\e4b0;\n$fa-var-lines-leaning: \\e51e;\n$fa-var-ruler-combined: \\f546;\n$fa-var-copyright: \\f1f9;\n$fa-var-equals: \\3d;\n$fa-var-blender: \\f517;\n$fa-var-teeth: \\f62e;\n$fa-var-shekel-sign: \\f20b;\n$fa-var-ils: \\f20b;\n$fa-var-shekel: \\f20b;\n$fa-var-sheqel: \\f20b;\n$fa-var-sheqel-sign: \\f20b;\n$fa-var-map: \\f279;\n$fa-var-rocket: \\f135;\n$fa-var-photo-film: \\f87c;\n$fa-var-photo-video: \\f87c;\n$fa-var-folder-minus: \\f65d;\n$fa-var-store: \\f54e;\n$fa-var-arrow-trend-up: \\e098;\n$fa-var-plug-circle-minus: \\e55e;\n$fa-var-sign-hanging: \\f4d9;\n$fa-var-sign: \\f4d9;\n$fa-var-bezier-curve: \\f55b;\n$fa-var-bell-slash: \\f1f6;\n$fa-var-tablet: \\f3fb;\n$fa-var-tablet-android: \\f3fb;\n$fa-var-school-flag: \\e56e;\n$fa-var-fill: \\f575;\n$fa-var-angle-up: \\f106;\n$fa-var-drumstick-bite: \\f6d7;\n$fa-var-holly-berry: \\f7aa;\n$fa-var-chevron-left: \\f053;\n$fa-var-bacteria: \\e059;\n$fa-var-hand-lizard: \\f258;\n$fa-var-notdef: \\e1fe;\n$fa-var-disease: \\f7fa;\n$fa-var-briefcase-medical: \\f469;\n$fa-var-genderless: \\f22d;\n$fa-var-chevron-right: \\f054;\n$fa-var-retweet: \\f079;\n$fa-var-car-rear: \\f5de;\n$fa-var-car-alt: \\f5de;\n$fa-var-pump-soap: \\e06b;\n$fa-var-video-slash: \\f4e2;\n$fa-var-battery-quarter: \\f243;\n$fa-var-battery-2: \\f243;\n$fa-var-radio: \\f8d7;\n$fa-var-baby-carriage: \\f77d;\n$fa-var-carriage-baby: \\f77d;\n$fa-var-traffic-light: \\f637;\n$fa-var-thermometer: \\f491;\n$fa-var-vr-cardboard: \\f729;\n$fa-var-hand-middle-finger: \\f806;\n$fa-var-percent: \\25;\n$fa-var-percentage: \\25;\n$fa-var-truck-moving: \\f4df;\n$fa-var-glass-water-droplet: \\e4f5;\n$fa-var-display: \\e163;\n$fa-var-face-smile: \\f118;\n$fa-var-smile: \\f118;\n$fa-var-thumbtack: \\f08d;\n$fa-var-thumb-tack: \\f08d;\n$fa-var-trophy: \\f091;\n$fa-var-person-praying: \\f683;\n$fa-var-pray: \\f683;\n$fa-var-hammer: \\f6e3;\n$fa-var-hand-peace: \\f25b;\n$fa-var-rotate: \\f2f1;\n$fa-var-sync-alt: \\f2f1;\n$fa-var-spinner: \\f110;\n$fa-var-robot: \\f544;\n$fa-var-peace: \\f67c;\n$fa-var-gears: \\f085;\n$fa-var-cogs: \\f085;\n$fa-var-warehouse: \\f494;\n$fa-var-arrow-up-right-dots: \\e4b7;\n$fa-var-splotch: \\f5bc;\n$fa-var-face-grin-hearts: \\f584;\n$fa-var-grin-hearts: \\f584;\n$fa-var-dice-four: \\f524;\n$fa-var-sim-card: \\f7c4;\n$fa-var-transgender: \\f225;\n$fa-var-transgender-alt: \\f225;\n$fa-var-mercury: \\f223;\n$fa-var-arrow-turn-down: \\f149;\n$fa-var-level-down: \\f149;\n$fa-var-person-falling-burst: \\e547;\n$fa-var-award: \\f559;\n$fa-var-ticket-simple: \\f3ff;\n$fa-var-ticket-alt: \\f3ff;\n$fa-var-building: \\f1ad;\n$fa-var-angles-left: \\f100;\n$fa-var-angle-double-left: \\f100;\n$fa-var-qrcode: \\f029;\n$fa-var-clock-rotate-left: \\f1da;\n$fa-var-history: \\f1da;\n$fa-var-face-grin-beam-sweat: \\f583;\n$fa-var-grin-beam-sweat: \\f583;\n$fa-var-file-export: \\f56e;\n$fa-var-arrow-right-from-file: \\f56e;\n$fa-var-shield: \\f132;\n$fa-var-shield-blank: \\f132;\n$fa-var-arrow-up-short-wide: \\f885;\n$fa-var-sort-amount-up-alt: \\f885;\n$fa-var-house-medical: \\e3b2;\n$fa-var-golf-ball-tee: \\f450;\n$fa-var-golf-ball: \\f450;\n$fa-var-circle-chevron-left: \\f137;\n$fa-var-chevron-circle-left: \\f137;\n$fa-var-house-chimney-window: \\e00d;\n$fa-var-pen-nib: \\f5ad;\n$fa-var-tent-arrow-turn-left: \\e580;\n$fa-var-tents: \\e582;\n$fa-var-wand-magic: \\f0d0;\n$fa-var-magic: \\f0d0;\n$fa-var-dog: \\f6d3;\n$fa-var-carrot: \\f787;\n$fa-var-moon: \\f186;\n$fa-var-wine-glass-empty: \\f5ce;\n$fa-var-wine-glass-alt: \\f5ce;\n$fa-var-cheese: \\f7ef;\n$fa-var-yin-yang: \\f6ad;\n$fa-var-music: \\f001;\n$fa-var-code-commit: \\f386;\n$fa-var-temperature-low: \\f76b;\n$fa-var-person-biking: \\f84a;\n$fa-var-biking: \\f84a;\n$fa-var-broom: \\f51a;\n$fa-var-shield-heart: \\e574;\n$fa-var-gopuram: \\f664;\n$fa-var-earth-oceania: \\e47b;\n$fa-var-globe-oceania: \\e47b;\n$fa-var-square-xmark: \\f2d3;\n$fa-var-times-square: \\f2d3;\n$fa-var-xmark-square: \\f2d3;\n$fa-var-hashtag: \\23;\n$fa-var-up-right-and-down-left-from-center: \\f424;\n$fa-var-expand-alt: \\f424;\n$fa-var-oil-can: \\f613;\n$fa-var-t: \\54;\n$fa-var-hippo: \\f6ed;\n$fa-var-chart-column: \\e0e3;\n$fa-var-infinity: \\f534;\n$fa-var-vial-circle-check: \\e596;\n$fa-var-person-arrow-down-to-line: \\e538;\n$fa-var-voicemail: \\f897;\n$fa-var-fan: \\f863;\n$fa-var-person-walking-luggage: \\e554;\n$fa-var-up-down: \\f338;\n$fa-var-arrows-alt-v: \\f338;\n$fa-var-cloud-moon-rain: \\f73c;\n$fa-var-calendar: \\f133;\n$fa-var-trailer: \\e041;\n$fa-var-bahai: \\f666;\n$fa-var-haykal: \\f666;\n$fa-var-sd-card: \\f7c2;\n$fa-var-dragon: \\f6d5;\n$fa-var-shoe-prints: \\f54b;\n$fa-var-circle-plus: \\f055;\n$fa-var-plus-circle: \\f055;\n$fa-var-face-grin-tongue-wink: \\f58b;\n$fa-var-grin-tongue-wink: \\f58b;\n$fa-var-hand-holding: \\f4bd;\n$fa-var-plug-circle-exclamation: \\e55d;\n$fa-var-link-slash: \\f127;\n$fa-var-chain-broken: \\f127;\n$fa-var-chain-slash: \\f127;\n$fa-var-unlink: \\f127;\n$fa-var-clone: \\f24d;\n$fa-var-person-walking-arrow-loop-left: \\e551;\n$fa-var-arrow-up-z-a: \\f882;\n$fa-var-sort-alpha-up-alt: \\f882;\n$fa-var-fire-flame-curved: \\f7e4;\n$fa-var-fire-alt: \\f7e4;\n$fa-var-tornado: \\f76f;\n$fa-var-file-circle-plus: \\e494;\n$fa-var-book-quran: \\f687;\n$fa-var-quran: \\f687;\n$fa-var-anchor: \\f13d;\n$fa-var-border-all: \\f84c;\n$fa-var-face-angry: \\f556;\n$fa-var-angry: \\f556;\n$fa-var-cookie-bite: \\f564;\n$fa-var-arrow-trend-down: \\e097;\n$fa-var-rss: \\f09e;\n$fa-var-feed: \\f09e;\n$fa-var-draw-polygon: \\f5ee;\n$fa-var-scale-balanced: \\f24e;\n$fa-var-balance-scale: \\f24e;\n$fa-var-gauge-simple-high: \\f62a;\n$fa-var-tachometer: \\f62a;\n$fa-var-tachometer-fast: \\f62a;\n$fa-var-shower: \\f2cc;\n$fa-var-desktop: \\f390;\n$fa-var-desktop-alt: \\f390;\n$fa-var-m: \\4d;\n$fa-var-table-list: \\f00b;\n$fa-var-th-list: \\f00b;\n$fa-var-comment-sms: \\f7cd;\n$fa-var-sms: \\f7cd;\n$fa-var-book: \\f02d;\n$fa-var-user-plus: \\f234;\n$fa-var-check: \\f00c;\n$fa-var-battery-three-quarters: \\f241;\n$fa-var-battery-4: \\f241;\n$fa-var-house-circle-check: \\e509;\n$fa-var-angle-left: \\f104;\n$fa-var-diagram-successor: \\e47a;\n$fa-var-truck-arrow-right: \\e58b;\n$fa-var-arrows-split-up-and-left: \\e4bc;\n$fa-var-hand-fist: \\f6de;\n$fa-var-fist-raised: \\f6de;\n$fa-var-cloud-moon: \\f6c3;\n$fa-var-briefcase: \\f0b1;\n$fa-var-person-falling: \\e546;\n$fa-var-image-portrait: \\f3e0;\n$fa-var-portrait: \\f3e0;\n$fa-var-user-tag: \\f507;\n$fa-var-rug: \\e569;\n$fa-var-earth-europe: \\f7a2;\n$fa-var-globe-europe: \\f7a2;\n$fa-var-cart-flatbed-suitcase: \\f59d;\n$fa-var-luggage-cart: \\f59d;\n$fa-var-rectangle-xmark: \\f410;\n$fa-var-rectangle-times: \\f410;\n$fa-var-times-rectangle: \\f410;\n$fa-var-window-close: \\f410;\n$fa-var-baht-sign: \\e0ac;\n$fa-var-book-open: \\f518;\n$fa-var-book-journal-whills: \\f66a;\n$fa-var-journal-whills: \\f66a;\n$fa-var-handcuffs: \\e4f8;\n$fa-var-triangle-exclamation: \\f071;\n$fa-var-exclamation-triangle: \\f071;\n$fa-var-warning: \\f071;\n$fa-var-database: \\f1c0;\n$fa-var-share: \\f064;\n$fa-var-mail-forward: \\f064;\n$fa-var-bottle-droplet: \\e4c4;\n$fa-var-mask-face: \\e1d7;\n$fa-var-hill-rockslide: \\e508;\n$fa-var-right-left: \\f362;\n$fa-var-exchange-alt: \\f362;\n$fa-var-paper-plane: \\f1d8;\n$fa-var-road-circle-exclamation: \\e565;\n$fa-var-dungeon: \\f6d9;\n$fa-var-align-right: \\f038;\n$fa-var-money-bill-1-wave: \\f53b;\n$fa-var-money-bill-wave-alt: \\f53b;\n$fa-var-life-ring: \\f1cd;\n$fa-var-hands: \\f2a7;\n$fa-var-sign-language: \\f2a7;\n$fa-var-signing: \\f2a7;\n$fa-var-calendar-day: \\f783;\n$fa-var-water-ladder: \\f5c5;\n$fa-var-ladder-water: \\f5c5;\n$fa-var-swimming-pool: \\f5c5;\n$fa-var-arrows-up-down: \\f07d;\n$fa-var-arrows-v: \\f07d;\n$fa-var-face-grimace: \\f57f;\n$fa-var-grimace: \\f57f;\n$fa-var-wheelchair-move: \\e2ce;\n$fa-var-wheelchair-alt: \\e2ce;\n$fa-var-turn-down: \\f3be;\n$fa-var-level-down-alt: \\f3be;\n$fa-var-person-walking-arrow-right: \\e552;\n$fa-var-square-envelope: \\f199;\n$fa-var-envelope-square: \\f199;\n$fa-var-dice: \\f522;\n$fa-var-bowling-ball: \\f436;\n$fa-var-brain: \\f5dc;\n$fa-var-bandage: \\f462;\n$fa-var-band-aid: \\f462;\n$fa-var-calendar-minus: \\f272;\n$fa-var-circle-xmark: \\f057;\n$fa-var-times-circle: \\f057;\n$fa-var-xmark-circle: \\f057;\n$fa-var-gifts: \\f79c;\n$fa-var-hotel: \\f594;\n$fa-var-earth-asia: \\f57e;\n$fa-var-globe-asia: \\f57e;\n$fa-var-id-card-clip: \\f47f;\n$fa-var-id-card-alt: \\f47f;\n$fa-var-magnifying-glass-plus: \\f00e;\n$fa-var-search-plus: \\f00e;\n$fa-var-thumbs-up: \\f164;\n$fa-var-user-clock: \\f4fd;\n$fa-var-hand-dots: \\f461;\n$fa-var-allergies: \\f461;\n$fa-var-file-invoice: \\f570;\n$fa-var-window-minimize: \\f2d1;\n$fa-var-mug-saucer: \\f0f4;\n$fa-var-coffee: \\f0f4;\n$fa-var-brush: \\f55d;\n$fa-var-mask: \\f6fa;\n$fa-var-magnifying-glass-minus: \\f010;\n$fa-var-search-minus: \\f010;\n$fa-var-ruler-vertical: \\f548;\n$fa-var-user-large: \\f406;\n$fa-var-user-alt: \\f406;\n$fa-var-train-tram: \\e5b4;\n$fa-var-user-nurse: \\f82f;\n$fa-var-syringe: \\f48e;\n$fa-var-cloud-sun: \\f6c4;\n$fa-var-stopwatch-20: \\e06f;\n$fa-var-square-full: \\f45c;\n$fa-var-magnet: \\f076;\n$fa-var-jar: \\e516;\n$fa-var-note-sticky: \\f249;\n$fa-var-sticky-note: \\f249;\n$fa-var-bug-slash: \\e490;\n$fa-var-arrow-up-from-water-pump: \\e4b6;\n$fa-var-bone: \\f5d7;\n$fa-var-user-injured: \\f728;\n$fa-var-face-sad-tear: \\f5b4;\n$fa-var-sad-tear: \\f5b4;\n$fa-var-plane: \\f072;\n$fa-var-tent-arrows-down: \\e581;\n$fa-var-exclamation: \\21;\n$fa-var-arrows-spin: \\e4bb;\n$fa-var-print: \\f02f;\n$fa-var-turkish-lira-sign: \\e2bb;\n$fa-var-try: \\e2bb;\n$fa-var-turkish-lira: \\e2bb;\n$fa-var-dollar-sign: \\24;\n$fa-var-dollar: \\24;\n$fa-var-usd: \\24;\n$fa-var-x: \\58;\n$fa-var-magnifying-glass-dollar: \\f688;\n$fa-var-search-dollar: \\f688;\n$fa-var-users-gear: \\f509;\n$fa-var-users-cog: \\f509;\n$fa-var-person-military-pointing: \\e54a;\n$fa-var-building-columns: \\f19c;\n$fa-var-bank: \\f19c;\n$fa-var-institution: \\f19c;\n$fa-var-museum: \\f19c;\n$fa-var-university: \\f19c;\n$fa-var-umbrella: \\f0e9;\n$fa-var-trowel: \\e589;\n$fa-var-d: \\44;\n$fa-var-stapler: \\e5af;\n$fa-var-masks-theater: \\f630;\n$fa-var-theater-masks: \\f630;\n$fa-var-kip-sign: \\e1c4;\n$fa-var-hand-point-left: \\f0a5;\n$fa-var-handshake-simple: \\f4c6;\n$fa-var-handshake-alt: \\f4c6;\n$fa-var-jet-fighter: \\f0fb;\n$fa-var-fighter-jet: \\f0fb;\n$fa-var-square-share-nodes: \\f1e1;\n$fa-var-share-alt-square: \\f1e1;\n$fa-var-barcode: \\f02a;\n$fa-var-plus-minus: \\e43c;\n$fa-var-video: \\f03d;\n$fa-var-video-camera: \\f03d;\n$fa-var-graduation-cap: \\f19d;\n$fa-var-mortar-board: \\f19d;\n$fa-var-hand-holding-medical: \\e05c;\n$fa-var-person-circle-check: \\e53e;\n$fa-var-turn-up: \\f3bf;\n$fa-var-level-up-alt: \\f3bf;\n\n$fa-var-monero: \\f3d0;\n$fa-var-hooli: \\f427;\n$fa-var-yelp: \\f1e9;\n$fa-var-cc-visa: \\f1f0;\n$fa-var-lastfm: \\f202;\n$fa-var-shopware: \\f5b5;\n$fa-var-creative-commons-nc: \\f4e8;\n$fa-var-aws: \\f375;\n$fa-var-redhat: \\f7bc;\n$fa-var-yoast: \\f2b1;\n$fa-var-cloudflare: \\e07d;\n$fa-var-ups: \\f7e0;\n$fa-var-pixiv: \\e640;\n$fa-var-wpexplorer: \\f2de;\n$fa-var-dyalog: \\f399;\n$fa-var-bity: \\f37a;\n$fa-var-stackpath: \\f842;\n$fa-var-buysellads: \\f20d;\n$fa-var-first-order: \\f2b0;\n$fa-var-modx: \\f285;\n$fa-var-guilded: \\e07e;\n$fa-var-vnv: \\f40b;\n$fa-var-square-js: \\f3b9;\n$fa-var-js-square: \\f3b9;\n$fa-var-microsoft: \\f3ca;\n$fa-var-qq: \\f1d6;\n$fa-var-orcid: \\f8d2;\n$fa-var-java: \\f4e4;\n$fa-var-invision: \\f7b0;\n$fa-var-creative-commons-pd-alt: \\f4ed;\n$fa-var-centercode: \\f380;\n$fa-var-glide-g: \\f2a6;\n$fa-var-drupal: \\f1a9;\n$fa-var-jxl: \\e67b;\n$fa-var-hire-a-helper: \\f3b0;\n$fa-var-creative-commons-by: \\f4e7;\n$fa-var-unity: \\e049;\n$fa-var-whmcs: \\f40d;\n$fa-var-rocketchat: \\f3e8;\n$fa-var-vk: \\f189;\n$fa-var-untappd: \\f405;\n$fa-var-mailchimp: \\f59e;\n$fa-var-css3-alt: \\f38b;\n$fa-var-square-reddit: \\f1a2;\n$fa-var-reddit-square: \\f1a2;\n$fa-var-vimeo-v: \\f27d;\n$fa-var-contao: \\f26d;\n$fa-var-square-font-awesome: \\e5ad;\n$fa-var-deskpro: \\f38f;\n$fa-var-brave: \\e63c;\n$fa-var-sistrix: \\f3ee;\n$fa-var-square-instagram: \\e055;\n$fa-var-instagram-square: \\e055;\n$fa-var-battle-net: \\f835;\n$fa-var-the-red-yeti: \\f69d;\n$fa-var-square-hacker-news: \\f3af;\n$fa-var-hacker-news-square: \\f3af;\n$fa-var-edge: \\f282;\n$fa-var-threads: \\e618;\n$fa-var-napster: \\f3d2;\n$fa-var-square-snapchat: \\f2ad;\n$fa-var-snapchat-square: \\f2ad;\n$fa-var-google-plus-g: \\f0d5;\n$fa-var-artstation: \\f77a;\n$fa-var-markdown: \\f60f;\n$fa-var-sourcetree: \\f7d3;\n$fa-var-google-plus: \\f2b3;\n$fa-var-diaspora: \\f791;\n$fa-var-foursquare: \\f180;\n$fa-var-stack-overflow: \\f16c;\n$fa-var-github-alt: \\f113;\n$fa-var-phoenix-squadron: \\f511;\n$fa-var-pagelines: \\f18c;\n$fa-var-algolia: \\f36c;\n$fa-var-red-river: \\f3e3;\n$fa-var-creative-commons-sa: \\f4ef;\n$fa-var-safari: \\f267;\n$fa-var-google: \\f1a0;\n$fa-var-square-font-awesome-stroke: \\f35c;\n$fa-var-font-awesome-alt: \\f35c;\n$fa-var-atlassian: \\f77b;\n$fa-var-linkedin-in: \\f0e1;\n$fa-var-digital-ocean: \\f391;\n$fa-var-nimblr: \\f5a8;\n$fa-var-chromecast: \\f838;\n$fa-var-evernote: \\f839;\n$fa-var-hacker-news: \\f1d4;\n$fa-var-creative-commons-sampling: \\f4f0;\n$fa-var-adversal: \\f36a;\n$fa-var-creative-commons: \\f25e;\n$fa-var-watchman-monitoring: \\e087;\n$fa-var-fonticons: \\f280;\n$fa-var-weixin: \\f1d7;\n$fa-var-shirtsinbulk: \\f214;\n$fa-var-codepen: \\f1cb;\n$fa-var-git-alt: \\f841;\n$fa-var-lyft: \\f3c3;\n$fa-var-rev: \\f5b2;\n$fa-var-windows: \\f17a;\n$fa-var-wizards-of-the-coast: \\f730;\n$fa-var-square-viadeo: \\f2aa;\n$fa-var-viadeo-square: \\f2aa;\n$fa-var-meetup: \\f2e0;\n$fa-var-centos: \\f789;\n$fa-var-adn: \\f170;\n$fa-var-cloudsmith: \\f384;\n$fa-var-opensuse: \\e62b;\n$fa-var-pied-piper-alt: \\f1a8;\n$fa-var-square-dribbble: \\f397;\n$fa-var-dribbble-square: \\f397;\n$fa-var-codiepie: \\f284;\n$fa-var-node: \\f419;\n$fa-var-mix: \\f3cb;\n$fa-var-steam: \\f1b6;\n$fa-var-cc-apple-pay: \\f416;\n$fa-var-scribd: \\f28a;\n$fa-var-debian: \\e60b;\n$fa-var-openid: \\f19b;\n$fa-var-instalod: \\e081;\n$fa-var-expeditedssl: \\f23e;\n$fa-var-sellcast: \\f2da;\n$fa-var-square-twitter: \\f081;\n$fa-var-twitter-square: \\f081;\n$fa-var-r-project: \\f4f7;\n$fa-var-delicious: \\f1a5;\n$fa-var-freebsd: \\f3a4;\n$fa-var-vuejs: \\f41f;\n$fa-var-accusoft: \\f369;\n$fa-var-ioxhost: \\f208;\n$fa-var-fonticons-fi: \\f3a2;\n$fa-var-app-store: \\f36f;\n$fa-var-cc-mastercard: \\f1f1;\n$fa-var-itunes-note: \\f3b5;\n$fa-var-golang: \\e40f;\n$fa-var-kickstarter: \\f3bb;\n$fa-var-square-kickstarter: \\f3bb;\n$fa-var-grav: \\f2d6;\n$fa-var-weibo: \\f18a;\n$fa-var-uncharted: \\e084;\n$fa-var-firstdraft: \\f3a1;\n$fa-var-square-youtube: \\f431;\n$fa-var-youtube-square: \\f431;\n$fa-var-wikipedia-w: \\f266;\n$fa-var-wpressr: \\f3e4;\n$fa-var-rendact: \\f3e4;\n$fa-var-angellist: \\f209;\n$fa-var-galactic-republic: \\f50c;\n$fa-var-nfc-directional: \\e530;\n$fa-var-skype: \\f17e;\n$fa-var-joget: \\f3b7;\n$fa-var-fedora: \\f798;\n$fa-var-stripe-s: \\f42a;\n$fa-var-meta: \\e49b;\n$fa-var-laravel: \\f3bd;\n$fa-var-hotjar: \\f3b1;\n$fa-var-bluetooth-b: \\f294;\n$fa-var-square-letterboxd: \\e62e;\n$fa-var-sticker-mule: \\f3f7;\n$fa-var-creative-commons-zero: \\f4f3;\n$fa-var-hips: \\f452;\n$fa-var-behance: \\f1b4;\n$fa-var-reddit: \\f1a1;\n$fa-var-discord: \\f392;\n$fa-var-chrome: \\f268;\n$fa-var-app-store-ios: \\f370;\n$fa-var-cc-discover: \\f1f2;\n$fa-var-wpbeginner: \\f297;\n$fa-var-confluence: \\f78d;\n$fa-var-shoelace: \\e60c;\n$fa-var-mdb: \\f8ca;\n$fa-var-dochub: \\f394;\n$fa-var-accessible-icon: \\f368;\n$fa-var-ebay: \\f4f4;\n$fa-var-amazon: \\f270;\n$fa-var-unsplash: \\e07c;\n$fa-var-yarn: \\f7e3;\n$fa-var-square-steam: \\f1b7;\n$fa-var-steam-square: \\f1b7;\n$fa-var-500px: \\f26e;\n$fa-var-square-vimeo: \\f194;\n$fa-var-vimeo-square: \\f194;\n$fa-var-asymmetrik: \\f372;\n$fa-var-font-awesome: \\f2b4;\n$fa-var-font-awesome-flag: \\f2b4;\n$fa-var-font-awesome-logo-full: \\f2b4;\n$fa-var-gratipay: \\f184;\n$fa-var-apple: \\f179;\n$fa-var-hive: \\e07f;\n$fa-var-gitkraken: \\f3a6;\n$fa-var-keybase: \\f4f5;\n$fa-var-apple-pay: \\f415;\n$fa-var-padlet: \\e4a0;\n$fa-var-amazon-pay: \\f42c;\n$fa-var-square-github: \\f092;\n$fa-var-github-square: \\f092;\n$fa-var-stumbleupon: \\f1a4;\n$fa-var-fedex: \\f797;\n$fa-var-phoenix-framework: \\f3dc;\n$fa-var-shopify: \\e057;\n$fa-var-neos: \\f612;\n$fa-var-square-threads: \\e619;\n$fa-var-hackerrank: \\f5f7;\n$fa-var-researchgate: \\f4f8;\n$fa-var-swift: \\f8e1;\n$fa-var-angular: \\f420;\n$fa-var-speakap: \\f3f3;\n$fa-var-angrycreative: \\f36e;\n$fa-var-y-combinator: \\f23b;\n$fa-var-empire: \\f1d1;\n$fa-var-envira: \\f299;\n$fa-var-google-scholar: \\e63b;\n$fa-var-square-gitlab: \\e5ae;\n$fa-var-gitlab-square: \\e5ae;\n$fa-var-studiovinari: \\f3f8;\n$fa-var-pied-piper: \\f2ae;\n$fa-var-wordpress: \\f19a;\n$fa-var-product-hunt: \\f288;\n$fa-var-firefox: \\f269;\n$fa-var-linode: \\f2b8;\n$fa-var-goodreads: \\f3a8;\n$fa-var-square-odnoklassniki: \\f264;\n$fa-var-odnoklassniki-square: \\f264;\n$fa-var-jsfiddle: \\f1cc;\n$fa-var-sith: \\f512;\n$fa-var-themeisle: \\f2b2;\n$fa-var-page4: \\f3d7;\n$fa-var-hashnode: \\e499;\n$fa-var-react: \\f41b;\n$fa-var-cc-paypal: \\f1f4;\n$fa-var-squarespace: \\f5be;\n$fa-var-cc-stripe: \\f1f5;\n$fa-var-creative-commons-share: \\f4f2;\n$fa-var-bitcoin: \\f379;\n$fa-var-keycdn: \\f3ba;\n$fa-var-opera: \\f26a;\n$fa-var-itch-io: \\f83a;\n$fa-var-umbraco: \\f8e8;\n$fa-var-galactic-senate: \\f50d;\n$fa-var-ubuntu: \\f7df;\n$fa-var-draft2digital: \\f396;\n$fa-var-stripe: \\f429;\n$fa-var-houzz: \\f27c;\n$fa-var-gg: \\f260;\n$fa-var-dhl: \\f790;\n$fa-var-square-pinterest: \\f0d3;\n$fa-var-pinterest-square: \\f0d3;\n$fa-var-xing: \\f168;\n$fa-var-blackberry: \\f37b;\n$fa-var-creative-commons-pd: \\f4ec;\n$fa-var-playstation: \\f3df;\n$fa-var-quinscape: \\f459;\n$fa-var-less: \\f41d;\n$fa-var-blogger-b: \\f37d;\n$fa-var-opencart: \\f23d;\n$fa-var-vine: \\f1ca;\n$fa-var-signal-messenger: \\e663;\n$fa-var-paypal: \\f1ed;\n$fa-var-gitlab: \\f296;\n$fa-var-typo3: \\f42b;\n$fa-var-reddit-alien: \\f281;\n$fa-var-yahoo: \\f19e;\n$fa-var-dailymotion: \\e052;\n$fa-var-affiliatetheme: \\f36b;\n$fa-var-pied-piper-pp: \\f1a7;\n$fa-var-bootstrap: \\f836;\n$fa-var-odnoklassniki: \\f263;\n$fa-var-nfc-symbol: \\e531;\n$fa-var-mintbit: \\e62f;\n$fa-var-ethereum: \\f42e;\n$fa-var-speaker-deck: \\f83c;\n$fa-var-creative-commons-nc-eu: \\f4e9;\n$fa-var-patreon: \\f3d9;\n$fa-var-avianex: \\f374;\n$fa-var-ello: \\f5f1;\n$fa-var-gofore: \\f3a7;\n$fa-var-bimobject: \\f378;\n$fa-var-brave-reverse: \\e63d;\n$fa-var-facebook-f: \\f39e;\n$fa-var-square-google-plus: \\f0d4;\n$fa-var-google-plus-square: \\f0d4;\n$fa-var-web-awesome: \\e682;\n$fa-var-mandalorian: \\f50f;\n$fa-var-first-order-alt: \\f50a;\n$fa-var-osi: \\f41a;\n$fa-var-google-wallet: \\f1ee;\n$fa-var-d-and-d-beyond: \\f6ca;\n$fa-var-periscope: \\f3da;\n$fa-var-fulcrum: \\f50b;\n$fa-var-cloudscale: \\f383;\n$fa-var-forumbee: \\f211;\n$fa-var-mizuni: \\f3cc;\n$fa-var-schlix: \\f3ea;\n$fa-var-square-xing: \\f169;\n$fa-var-xing-square: \\f169;\n$fa-var-bandcamp: \\f2d5;\n$fa-var-wpforms: \\f298;\n$fa-var-cloudversify: \\f385;\n$fa-var-usps: \\f7e1;\n$fa-var-megaport: \\f5a3;\n$fa-var-magento: \\f3c4;\n$fa-var-spotify: \\f1bc;\n$fa-var-optin-monster: \\f23c;\n$fa-var-fly: \\f417;\n$fa-var-aviato: \\f421;\n$fa-var-itunes: \\f3b4;\n$fa-var-cuttlefish: \\f38c;\n$fa-var-blogger: \\f37c;\n$fa-var-flickr: \\f16e;\n$fa-var-viber: \\f409;\n$fa-var-soundcloud: \\f1be;\n$fa-var-digg: \\f1a6;\n$fa-var-tencent-weibo: \\f1d5;\n$fa-var-letterboxd: \\e62d;\n$fa-var-symfony: \\f83d;\n$fa-var-maxcdn: \\f136;\n$fa-var-etsy: \\f2d7;\n$fa-var-facebook-messenger: \\f39f;\n$fa-var-audible: \\f373;\n$fa-var-think-peaks: \\f731;\n$fa-var-bilibili: \\e3d9;\n$fa-var-erlang: \\f39d;\n$fa-var-x-twitter: \\e61b;\n$fa-var-cotton-bureau: \\f89e;\n$fa-var-dashcube: \\f210;\n$fa-var-42-group: \\e080;\n$fa-var-innosoft: \\e080;\n$fa-var-stack-exchange: \\f18d;\n$fa-var-elementor: \\f430;\n$fa-var-square-pied-piper: \\e01e;\n$fa-var-pied-piper-square: \\e01e;\n$fa-var-creative-commons-nd: \\f4eb;\n$fa-var-palfed: \\f3d8;\n$fa-var-superpowers: \\f2dd;\n$fa-var-resolving: \\f3e7;\n$fa-var-xbox: \\f412;\n$fa-var-square-web-awesome-stroke: \\e684;\n$fa-var-searchengin: \\f3eb;\n$fa-var-tiktok: \\e07b;\n$fa-var-square-facebook: \\f082;\n$fa-var-facebook-square: \\f082;\n$fa-var-renren: \\f18b;\n$fa-var-linux: \\f17c;\n$fa-var-glide: \\f2a5;\n$fa-var-linkedin: \\f08c;\n$fa-var-hubspot: \\f3b2;\n$fa-var-deploydog: \\f38e;\n$fa-var-twitch: \\f1e8;\n$fa-var-ravelry: \\f2d9;\n$fa-var-mixer: \\e056;\n$fa-var-square-lastfm: \\f203;\n$fa-var-lastfm-square: \\f203;\n$fa-var-vimeo: \\f40a;\n$fa-var-mendeley: \\f7b3;\n$fa-var-uniregistry: \\f404;\n$fa-var-figma: \\f799;\n$fa-var-creative-commons-remix: \\f4ee;\n$fa-var-cc-amazon-pay: \\f42d;\n$fa-var-dropbox: \\f16b;\n$fa-var-instagram: \\f16d;\n$fa-var-cmplid: \\e360;\n$fa-var-upwork: \\e641;\n$fa-var-facebook: \\f09a;\n$fa-var-gripfire: \\f3ac;\n$fa-var-jedi-order: \\f50e;\n$fa-var-uikit: \\f403;\n$fa-var-fort-awesome-alt: \\f3a3;\n$fa-var-phabricator: \\f3db;\n$fa-var-ussunnah: \\f407;\n$fa-var-earlybirds: \\f39a;\n$fa-var-trade-federation: \\f513;\n$fa-var-autoprefixer: \\f41c;\n$fa-var-whatsapp: \\f232;\n$fa-var-square-upwork: \\e67c;\n$fa-var-slideshare: \\f1e7;\n$fa-var-google-play: \\f3ab;\n$fa-var-viadeo: \\f2a9;\n$fa-var-line: \\f3c0;\n$fa-var-google-drive: \\f3aa;\n$fa-var-servicestack: \\f3ec;\n$fa-var-simplybuilt: \\f215;\n$fa-var-bitbucket: \\f171;\n$fa-var-imdb: \\f2d8;\n$fa-var-deezer: \\e077;\n$fa-var-raspberry-pi: \\f7bb;\n$fa-var-jira: \\f7b1;\n$fa-var-docker: \\f395;\n$fa-var-screenpal: \\e570;\n$fa-var-bluetooth: \\f293;\n$fa-var-gitter: \\f426;\n$fa-var-d-and-d: \\f38d;\n$fa-var-microblog: \\e01a;\n$fa-var-cc-diners-club: \\f24c;\n$fa-var-gg-circle: \\f261;\n$fa-var-pied-piper-hat: \\f4e5;\n$fa-var-kickstarter-k: \\f3bc;\n$fa-var-yandex: \\f413;\n$fa-var-readme: \\f4d5;\n$fa-var-html5: \\f13b;\n$fa-var-sellsy: \\f213;\n$fa-var-square-web-awesome: \\e683;\n$fa-var-sass: \\f41e;\n$fa-var-wirsindhandwerk: \\e2d0;\n$fa-var-wsh: \\e2d0;\n$fa-var-buromobelexperte: \\f37f;\n$fa-var-salesforce: \\f83b;\n$fa-var-octopus-deploy: \\e082;\n$fa-var-medapps: \\f3c6;\n$fa-var-ns8: \\f3d5;\n$fa-var-pinterest-p: \\f231;\n$fa-var-apper: \\f371;\n$fa-var-fort-awesome: \\f286;\n$fa-var-waze: \\f83f;\n$fa-var-bluesky: \\e671;\n$fa-var-cc-jcb: \\f24b;\n$fa-var-snapchat: \\f2ab;\n$fa-var-snapchat-ghost: \\f2ab;\n$fa-var-fantasy-flight-games: \\f6dc;\n$fa-var-rust: \\e07a;\n$fa-var-wix: \\f5cf;\n$fa-var-square-behance: \\f1b5;\n$fa-var-behance-square: \\f1b5;\n$fa-var-supple: \\f3f9;\n$fa-var-webflow: \\e65c;\n$fa-var-rebel: \\f1d0;\n$fa-var-css3: \\f13c;\n$fa-var-staylinked: \\f3f5;\n$fa-var-kaggle: \\f5fa;\n$fa-var-space-awesome: \\e5ac;\n$fa-var-deviantart: \\f1bd;\n$fa-var-cpanel: \\f388;\n$fa-var-goodreads-g: \\f3a9;\n$fa-var-square-git: \\f1d2;\n$fa-var-git-square: \\f1d2;\n$fa-var-square-tumblr: \\f174;\n$fa-var-tumblr-square: \\f174;\n$fa-var-trello: \\f181;\n$fa-var-creative-commons-nc-jp: \\f4ea;\n$fa-var-get-pocket: \\f265;\n$fa-var-perbyte: \\e083;\n$fa-var-grunt: \\f3ad;\n$fa-var-weebly: \\f5cc;\n$fa-var-connectdevelop: \\f20e;\n$fa-var-leanpub: \\f212;\n$fa-var-black-tie: \\f27e;\n$fa-var-themeco: \\f5c6;\n$fa-var-python: \\f3e2;\n$fa-var-android: \\f17b;\n$fa-var-bots: \\e340;\n$fa-var-free-code-camp: \\f2c5;\n$fa-var-hornbill: \\f592;\n$fa-var-js: \\f3b8;\n$fa-var-ideal: \\e013;\n$fa-var-git: \\f1d3;\n$fa-var-dev: \\f6cc;\n$fa-var-sketch: \\f7c6;\n$fa-var-yandex-international: \\f414;\n$fa-var-cc-amex: \\f1f3;\n$fa-var-uber: \\f402;\n$fa-var-github: \\f09b;\n$fa-var-php: \\f457;\n$fa-var-alipay: \\f642;\n$fa-var-youtube: \\f167;\n$fa-var-skyatlas: \\f216;\n$fa-var-firefox-browser: \\e007;\n$fa-var-replyd: \\f3e6;\n$fa-var-suse: \\f7d6;\n$fa-var-jenkins: \\f3b6;\n$fa-var-twitter: \\f099;\n$fa-var-rockrms: \\f3e9;\n$fa-var-pinterest: \\f0d2;\n$fa-var-buffer: \\f837;\n$fa-var-npm: \\f3d4;\n$fa-var-yammer: \\f840;\n$fa-var-btc: \\f15a;\n$fa-var-dribbble: \\f17d;\n$fa-var-stumbleupon-circle: \\f1a3;\n$fa-var-internet-explorer: \\f26b;\n$fa-var-stubber: \\e5c7;\n$fa-var-telegram: \\f2c6;\n$fa-var-telegram-plane: \\f2c6;\n$fa-var-old-republic: \\f510;\n$fa-var-odysee: \\e5c6;\n$fa-var-square-whatsapp: \\f40c;\n$fa-var-whatsapp-square: \\f40c;\n$fa-var-node-js: \\f3d3;\n$fa-var-edge-legacy: \\e078;\n$fa-var-slack: \\f198;\n$fa-var-slack-hash: \\f198;\n$fa-var-medrt: \\f3c8;\n$fa-var-usb: \\f287;\n$fa-var-tumblr: \\f173;\n$fa-var-vaadin: \\f408;\n$fa-var-quora: \\f2c4;\n$fa-var-square-x-twitter: \\e61a;\n$fa-var-reacteurope: \\f75d;\n$fa-var-medium: \\f23a;\n$fa-var-medium-m: \\f23a;\n$fa-var-amilia: \\f36d;\n$fa-var-mixcloud: \\f289;\n$fa-var-flipboard: \\f44d;\n$fa-var-viacoin: \\f237;\n$fa-var-critical-role: \\f6c9;\n$fa-var-sitrox: \\e44a;\n$fa-var-discourse: \\f393;\n$fa-var-joomla: \\f1aa;\n$fa-var-mastodon: \\f4f6;\n$fa-var-airbnb: \\f834;\n$fa-var-wolf-pack-battalion: \\f514;\n$fa-var-buy-n-large: \\f8a6;\n$fa-var-gulp: \\f3ae;\n$fa-var-creative-commons-sampling-plus: \\f4f1;\n$fa-var-strava: \\f428;\n$fa-var-ember: \\f423;\n$fa-var-canadian-maple-leaf: \\f785;\n$fa-var-teamspeak: \\f4f9;\n$fa-var-pushed: \\f3e1;\n$fa-var-wordpress-simple: \\f411;\n$fa-var-nutritionix: \\f3d6;\n$fa-var-wodu: \\e088;\n$fa-var-google-pay: \\e079;\n$fa-var-intercom: \\f7af;\n$fa-var-zhihu: \\f63f;\n$fa-var-korvue: \\f42f;\n$fa-var-pix: \\e43a;\n$fa-var-steam-symbol: \\f3f6;\n\n$fa-icons: (\n  \"0\": $fa-var-0,\n  \"1\": $fa-var-1,\n  \"2\": $fa-var-2,\n  \"3\": $fa-var-3,\n  \"4\": $fa-var-4,\n  \"5\": $fa-var-5,\n  \"6\": $fa-var-6,\n  \"7\": $fa-var-7,\n  \"8\": $fa-var-8,\n  \"9\": $fa-var-9,\n  \"fill-drip\": $fa-var-fill-drip,\n  \"arrows-to-circle\": $fa-var-arrows-to-circle,\n  \"circle-chevron-right\": $fa-var-circle-chevron-right,\n  \"chevron-circle-right\": $fa-var-chevron-circle-right,\n  \"at\": $fa-var-at,\n  \"trash-can\": $fa-var-trash-can,\n  \"trash-alt\": $fa-var-trash-alt,\n  \"text-height\": $fa-var-text-height,\n  \"user-xmark\": $fa-var-user-xmark,\n  \"user-times\": $fa-var-user-times,\n  \"stethoscope\": $fa-var-stethoscope,\n  \"message\": $fa-var-message,\n  \"comment-alt\": $fa-var-comment-alt,\n  \"info\": $fa-var-info,\n  \"down-left-and-up-right-to-center\": $fa-var-down-left-and-up-right-to-center,\n  \"compress-alt\": $fa-var-compress-alt,\n  \"explosion\": $fa-var-explosion,\n  \"file-lines\": $fa-var-file-lines,\n  \"file-alt\": $fa-var-file-alt,\n  \"file-text\": $fa-var-file-text,\n  \"wave-square\": $fa-var-wave-square,\n  \"ring\": $fa-var-ring,\n  \"building-un\": $fa-var-building-un,\n  \"dice-three\": $fa-var-dice-three,\n  \"calendar-days\": $fa-var-calendar-days,\n  \"calendar-alt\": $fa-var-calendar-alt,\n  \"anchor-circle-check\": $fa-var-anchor-circle-check,\n  \"building-circle-arrow-right\": $fa-var-building-circle-arrow-right,\n  \"volleyball\": $fa-var-volleyball,\n  \"volleyball-ball\": $fa-var-volleyball-ball,\n  \"arrows-up-to-line\": $fa-var-arrows-up-to-line,\n  \"sort-down\": $fa-var-sort-down,\n  \"sort-desc\": $fa-var-sort-desc,\n  \"circle-minus\": $fa-var-circle-minus,\n  \"minus-circle\": $fa-var-minus-circle,\n  \"door-open\": $fa-var-door-open,\n  \"right-from-bracket\": $fa-var-right-from-bracket,\n  \"sign-out-alt\": $fa-var-sign-out-alt,\n  \"atom\": $fa-var-atom,\n  \"soap\": $fa-var-soap,\n  \"icons\": $fa-var-icons,\n  \"heart-music-camera-bolt\": $fa-var-heart-music-camera-bolt,\n  \"microphone-lines-slash\": $fa-var-microphone-lines-slash,\n  \"microphone-alt-slash\": $fa-var-microphone-alt-slash,\n  \"bridge-circle-check\": $fa-var-bridge-circle-check,\n  \"pump-medical\": $fa-var-pump-medical,\n  \"fingerprint\": $fa-var-fingerprint,\n  \"hand-point-right\": $fa-var-hand-point-right,\n  \"magnifying-glass-location\": $fa-var-magnifying-glass-location,\n  \"search-location\": $fa-var-search-location,\n  \"forward-step\": $fa-var-forward-step,\n  \"step-forward\": $fa-var-step-forward,\n  \"face-smile-beam\": $fa-var-face-smile-beam,\n  \"smile-beam\": $fa-var-smile-beam,\n  \"flag-checkered\": $fa-var-flag-checkered,\n  \"football\": $fa-var-football,\n  \"football-ball\": $fa-var-football-ball,\n  \"school-circle-exclamation\": $fa-var-school-circle-exclamation,\n  \"crop\": $fa-var-crop,\n  \"angles-down\": $fa-var-angles-down,\n  \"angle-double-down\": $fa-var-angle-double-down,\n  \"users-rectangle\": $fa-var-users-rectangle,\n  \"people-roof\": $fa-var-people-roof,\n  \"people-line\": $fa-var-people-line,\n  \"beer-mug-empty\": $fa-var-beer-mug-empty,\n  \"beer\": $fa-var-beer,\n  \"diagram-predecessor\": $fa-var-diagram-predecessor,\n  \"arrow-up-long\": $fa-var-arrow-up-long,\n  \"long-arrow-up\": $fa-var-long-arrow-up,\n  \"fire-flame-simple\": $fa-var-fire-flame-simple,\n  \"burn\": $fa-var-burn,\n  \"person\": $fa-var-person,\n  \"male\": $fa-var-male,\n  \"laptop\": $fa-var-laptop,\n  \"file-csv\": $fa-var-file-csv,\n  \"menorah\": $fa-var-menorah,\n  \"truck-plane\": $fa-var-truck-plane,\n  \"record-vinyl\": $fa-var-record-vinyl,\n  \"face-grin-stars\": $fa-var-face-grin-stars,\n  \"grin-stars\": $fa-var-grin-stars,\n  \"bong\": $fa-var-bong,\n  \"spaghetti-monster-flying\": $fa-var-spaghetti-monster-flying,\n  \"pastafarianism\": $fa-var-pastafarianism,\n  \"arrow-down-up-across-line\": $fa-var-arrow-down-up-across-line,\n  \"spoon\": $fa-var-spoon,\n  \"utensil-spoon\": $fa-var-utensil-spoon,\n  \"jar-wheat\": $fa-var-jar-wheat,\n  \"envelopes-bulk\": $fa-var-envelopes-bulk,\n  \"mail-bulk\": $fa-var-mail-bulk,\n  \"file-circle-exclamation\": $fa-var-file-circle-exclamation,\n  \"circle-h\": $fa-var-circle-h,\n  \"hospital-symbol\": $fa-var-hospital-symbol,\n  \"pager\": $fa-var-pager,\n  \"address-book\": $fa-var-address-book,\n  \"contact-book\": $fa-var-contact-book,\n  \"strikethrough\": $fa-var-strikethrough,\n  \"k\": $fa-var-k,\n  \"landmark-flag\": $fa-var-landmark-flag,\n  \"pencil\": $fa-var-pencil,\n  \"pencil-alt\": $fa-var-pencil-alt,\n  \"backward\": $fa-var-backward,\n  \"caret-right\": $fa-var-caret-right,\n  \"comments\": $fa-var-comments,\n  \"paste\": $fa-var-paste,\n  \"file-clipboard\": $fa-var-file-clipboard,\n  \"code-pull-request\": $fa-var-code-pull-request,\n  \"clipboard-list\": $fa-var-clipboard-list,\n  \"truck-ramp-box\": $fa-var-truck-ramp-box,\n  \"truck-loading\": $fa-var-truck-loading,\n  \"user-check\": $fa-var-user-check,\n  \"vial-virus\": $fa-var-vial-virus,\n  \"sheet-plastic\": $fa-var-sheet-plastic,\n  \"blog\": $fa-var-blog,\n  \"user-ninja\": $fa-var-user-ninja,\n  \"person-arrow-up-from-line\": $fa-var-person-arrow-up-from-line,\n  \"scroll-torah\": $fa-var-scroll-torah,\n  \"torah\": $fa-var-torah,\n  \"broom-ball\": $fa-var-broom-ball,\n  \"quidditch\": $fa-var-quidditch,\n  \"quidditch-broom-ball\": $fa-var-quidditch-broom-ball,\n  \"toggle-off\": $fa-var-toggle-off,\n  \"box-archive\": $fa-var-box-archive,\n  \"archive\": $fa-var-archive,\n  \"person-drowning\": $fa-var-person-drowning,\n  \"arrow-down-9-1\": $fa-var-arrow-down-9-1,\n  \"sort-numeric-desc\": $fa-var-sort-numeric-desc,\n  \"sort-numeric-down-alt\": $fa-var-sort-numeric-down-alt,\n  \"face-grin-tongue-squint\": $fa-var-face-grin-tongue-squint,\n  \"grin-tongue-squint\": $fa-var-grin-tongue-squint,\n  \"spray-can\": $fa-var-spray-can,\n  \"truck-monster\": $fa-var-truck-monster,\n  \"w\": $fa-var-w,\n  \"earth-africa\": $fa-var-earth-africa,\n  \"globe-africa\": $fa-var-globe-africa,\n  \"rainbow\": $fa-var-rainbow,\n  \"circle-notch\": $fa-var-circle-notch,\n  \"tablet-screen-button\": $fa-var-tablet-screen-button,\n  \"tablet-alt\": $fa-var-tablet-alt,\n  \"paw\": $fa-var-paw,\n  \"cloud\": $fa-var-cloud,\n  \"trowel-bricks\": $fa-var-trowel-bricks,\n  \"face-flushed\": $fa-var-face-flushed,\n  \"flushed\": $fa-var-flushed,\n  \"hospital-user\": $fa-var-hospital-user,\n  \"tent-arrow-left-right\": $fa-var-tent-arrow-left-right,\n  \"gavel\": $fa-var-gavel,\n  \"legal\": $fa-var-legal,\n  \"binoculars\": $fa-var-binoculars,\n  \"microphone-slash\": $fa-var-microphone-slash,\n  \"box-tissue\": $fa-var-box-tissue,\n  \"motorcycle\": $fa-var-motorcycle,\n  \"bell-concierge\": $fa-var-bell-concierge,\n  \"concierge-bell\": $fa-var-concierge-bell,\n  \"pen-ruler\": $fa-var-pen-ruler,\n  \"pencil-ruler\": $fa-var-pencil-ruler,\n  \"people-arrows\": $fa-var-people-arrows,\n  \"people-arrows-left-right\": $fa-var-people-arrows-left-right,\n  \"mars-and-venus-burst\": $fa-var-mars-and-venus-burst,\n  \"square-caret-right\": $fa-var-square-caret-right,\n  \"caret-square-right\": $fa-var-caret-square-right,\n  \"scissors\": $fa-var-scissors,\n  \"cut\": $fa-var-cut,\n  \"sun-plant-wilt\": $fa-var-sun-plant-wilt,\n  \"toilets-portable\": $fa-var-toilets-portable,\n  \"hockey-puck\": $fa-var-hockey-puck,\n  \"table\": $fa-var-table,\n  \"magnifying-glass-arrow-right\": $fa-var-magnifying-glass-arrow-right,\n  \"tachograph-digital\": $fa-var-tachograph-digital,\n  \"digital-tachograph\": $fa-var-digital-tachograph,\n  \"users-slash\": $fa-var-users-slash,\n  \"clover\": $fa-var-clover,\n  \"reply\": $fa-var-reply,\n  \"mail-reply\": $fa-var-mail-reply,\n  \"star-and-crescent\": $fa-var-star-and-crescent,\n  \"house-fire\": $fa-var-house-fire,\n  \"square-minus\": $fa-var-square-minus,\n  \"minus-square\": $fa-var-minus-square,\n  \"helicopter\": $fa-var-helicopter,\n  \"compass\": $fa-var-compass,\n  \"square-caret-down\": $fa-var-square-caret-down,\n  \"caret-square-down\": $fa-var-caret-square-down,\n  \"file-circle-question\": $fa-var-file-circle-question,\n  \"laptop-code\": $fa-var-laptop-code,\n  \"swatchbook\": $fa-var-swatchbook,\n  \"prescription-bottle\": $fa-var-prescription-bottle,\n  \"bars\": $fa-var-bars,\n  \"navicon\": $fa-var-navicon,\n  \"people-group\": $fa-var-people-group,\n  \"hourglass-end\": $fa-var-hourglass-end,\n  \"hourglass-3\": $fa-var-hourglass-3,\n  \"heart-crack\": $fa-var-heart-crack,\n  \"heart-broken\": $fa-var-heart-broken,\n  \"square-up-right\": $fa-var-square-up-right,\n  \"external-link-square-alt\": $fa-var-external-link-square-alt,\n  \"face-kiss-beam\": $fa-var-face-kiss-beam,\n  \"kiss-beam\": $fa-var-kiss-beam,\n  \"film\": $fa-var-film,\n  \"ruler-horizontal\": $fa-var-ruler-horizontal,\n  \"people-robbery\": $fa-var-people-robbery,\n  \"lightbulb\": $fa-var-lightbulb,\n  \"caret-left\": $fa-var-caret-left,\n  \"circle-exclamation\": $fa-var-circle-exclamation,\n  \"exclamation-circle\": $fa-var-exclamation-circle,\n  \"school-circle-xmark\": $fa-var-school-circle-xmark,\n  \"arrow-right-from-bracket\": $fa-var-arrow-right-from-bracket,\n  \"sign-out\": $fa-var-sign-out,\n  \"circle-chevron-down\": $fa-var-circle-chevron-down,\n  \"chevron-circle-down\": $fa-var-chevron-circle-down,\n  \"unlock-keyhole\": $fa-var-unlock-keyhole,\n  \"unlock-alt\": $fa-var-unlock-alt,\n  \"cloud-showers-heavy\": $fa-var-cloud-showers-heavy,\n  \"headphones-simple\": $fa-var-headphones-simple,\n  \"headphones-alt\": $fa-var-headphones-alt,\n  \"sitemap\": $fa-var-sitemap,\n  \"circle-dollar-to-slot\": $fa-var-circle-dollar-to-slot,\n  \"donate\": $fa-var-donate,\n  \"memory\": $fa-var-memory,\n  \"road-spikes\": $fa-var-road-spikes,\n  \"fire-burner\": $fa-var-fire-burner,\n  \"flag\": $fa-var-flag,\n  \"hanukiah\": $fa-var-hanukiah,\n  \"feather\": $fa-var-feather,\n  \"volume-low\": $fa-var-volume-low,\n  \"volume-down\": $fa-var-volume-down,\n  \"comment-slash\": $fa-var-comment-slash,\n  \"cloud-sun-rain\": $fa-var-cloud-sun-rain,\n  \"compress\": $fa-var-compress,\n  \"wheat-awn\": $fa-var-wheat-awn,\n  \"wheat-alt\": $fa-var-wheat-alt,\n  \"ankh\": $fa-var-ankh,\n  \"hands-holding-child\": $fa-var-hands-holding-child,\n  \"asterisk\": $fa-var-asterisk,\n  \"square-check\": $fa-var-square-check,\n  \"check-square\": $fa-var-check-square,\n  \"peseta-sign\": $fa-var-peseta-sign,\n  \"heading\": $fa-var-heading,\n  \"header\": $fa-var-header,\n  \"ghost\": $fa-var-ghost,\n  \"list\": $fa-var-list,\n  \"list-squares\": $fa-var-list-squares,\n  \"square-phone-flip\": $fa-var-square-phone-flip,\n  \"phone-square-alt\": $fa-var-phone-square-alt,\n  \"cart-plus\": $fa-var-cart-plus,\n  \"gamepad\": $fa-var-gamepad,\n  \"circle-dot\": $fa-var-circle-dot,\n  \"dot-circle\": $fa-var-dot-circle,\n  \"face-dizzy\": $fa-var-face-dizzy,\n  \"dizzy\": $fa-var-dizzy,\n  \"egg\": $fa-var-egg,\n  \"house-medical-circle-xmark\": $fa-var-house-medical-circle-xmark,\n  \"campground\": $fa-var-campground,\n  \"folder-plus\": $fa-var-folder-plus,\n  \"futbol\": $fa-var-futbol,\n  \"futbol-ball\": $fa-var-futbol-ball,\n  \"soccer-ball\": $fa-var-soccer-ball,\n  \"paintbrush\": $fa-var-paintbrush,\n  \"paint-brush\": $fa-var-paint-brush,\n  \"lock\": $fa-var-lock,\n  \"gas-pump\": $fa-var-gas-pump,\n  \"hot-tub-person\": $fa-var-hot-tub-person,\n  \"hot-tub\": $fa-var-hot-tub,\n  \"map-location\": $fa-var-map-location,\n  \"map-marked\": $fa-var-map-marked,\n  \"house-flood-water\": $fa-var-house-flood-water,\n  \"tree\": $fa-var-tree,\n  \"bridge-lock\": $fa-var-bridge-lock,\n  \"sack-dollar\": $fa-var-sack-dollar,\n  \"pen-to-square\": $fa-var-pen-to-square,\n  \"edit\": $fa-var-edit,\n  \"car-side\": $fa-var-car-side,\n  \"share-nodes\": $fa-var-share-nodes,\n  \"share-alt\": $fa-var-share-alt,\n  \"heart-circle-minus\": $fa-var-heart-circle-minus,\n  \"hourglass-half\": $fa-var-hourglass-half,\n  \"hourglass-2\": $fa-var-hourglass-2,\n  \"microscope\": $fa-var-microscope,\n  \"sink\": $fa-var-sink,\n  \"bag-shopping\": $fa-var-bag-shopping,\n  \"shopping-bag\": $fa-var-shopping-bag,\n  \"arrow-down-z-a\": $fa-var-arrow-down-z-a,\n  \"sort-alpha-desc\": $fa-var-sort-alpha-desc,\n  \"sort-alpha-down-alt\": $fa-var-sort-alpha-down-alt,\n  \"mitten\": $fa-var-mitten,\n  \"person-rays\": $fa-var-person-rays,\n  \"users\": $fa-var-users,\n  \"eye-slash\": $fa-var-eye-slash,\n  \"flask-vial\": $fa-var-flask-vial,\n  \"hand\": $fa-var-hand,\n  \"hand-paper\": $fa-var-hand-paper,\n  \"om\": $fa-var-om,\n  \"worm\": $fa-var-worm,\n  \"house-circle-xmark\": $fa-var-house-circle-xmark,\n  \"plug\": $fa-var-plug,\n  \"chevron-up\": $fa-var-chevron-up,\n  \"hand-spock\": $fa-var-hand-spock,\n  \"stopwatch\": $fa-var-stopwatch,\n  \"face-kiss\": $fa-var-face-kiss,\n  \"kiss\": $fa-var-kiss,\n  \"bridge-circle-xmark\": $fa-var-bridge-circle-xmark,\n  \"face-grin-tongue\": $fa-var-face-grin-tongue,\n  \"grin-tongue\": $fa-var-grin-tongue,\n  \"chess-bishop\": $fa-var-chess-bishop,\n  \"face-grin-wink\": $fa-var-face-grin-wink,\n  \"grin-wink\": $fa-var-grin-wink,\n  \"ear-deaf\": $fa-var-ear-deaf,\n  \"deaf\": $fa-var-deaf,\n  \"deafness\": $fa-var-deafness,\n  \"hard-of-hearing\": $fa-var-hard-of-hearing,\n  \"road-circle-check\": $fa-var-road-circle-check,\n  \"dice-five\": $fa-var-dice-five,\n  \"square-rss\": $fa-var-square-rss,\n  \"rss-square\": $fa-var-rss-square,\n  \"land-mine-on\": $fa-var-land-mine-on,\n  \"i-cursor\": $fa-var-i-cursor,\n  \"stamp\": $fa-var-stamp,\n  \"stairs\": $fa-var-stairs,\n  \"i\": $fa-var-i,\n  \"hryvnia-sign\": $fa-var-hryvnia-sign,\n  \"hryvnia\": $fa-var-hryvnia,\n  \"pills\": $fa-var-pills,\n  \"face-grin-wide\": $fa-var-face-grin-wide,\n  \"grin-alt\": $fa-var-grin-alt,\n  \"tooth\": $fa-var-tooth,\n  \"v\": $fa-var-v,\n  \"bangladeshi-taka-sign\": $fa-var-bangladeshi-taka-sign,\n  \"bicycle\": $fa-var-bicycle,\n  \"staff-snake\": $fa-var-staff-snake,\n  \"rod-asclepius\": $fa-var-rod-asclepius,\n  \"rod-snake\": $fa-var-rod-snake,\n  \"staff-aesculapius\": $fa-var-staff-aesculapius,\n  \"head-side-cough-slash\": $fa-var-head-side-cough-slash,\n  \"truck-medical\": $fa-var-truck-medical,\n  \"ambulance\": $fa-var-ambulance,\n  \"wheat-awn-circle-exclamation\": $fa-var-wheat-awn-circle-exclamation,\n  \"snowman\": $fa-var-snowman,\n  \"mortar-pestle\": $fa-var-mortar-pestle,\n  \"road-barrier\": $fa-var-road-barrier,\n  \"school\": $fa-var-school,\n  \"igloo\": $fa-var-igloo,\n  \"joint\": $fa-var-joint,\n  \"angle-right\": $fa-var-angle-right,\n  \"horse\": $fa-var-horse,\n  \"q\": $fa-var-q,\n  \"g\": $fa-var-g,\n  \"notes-medical\": $fa-var-notes-medical,\n  \"temperature-half\": $fa-var-temperature-half,\n  \"temperature-2\": $fa-var-temperature-2,\n  \"thermometer-2\": $fa-var-thermometer-2,\n  \"thermometer-half\": $fa-var-thermometer-half,\n  \"dong-sign\": $fa-var-dong-sign,\n  \"capsules\": $fa-var-capsules,\n  \"poo-storm\": $fa-var-poo-storm,\n  \"poo-bolt\": $fa-var-poo-bolt,\n  \"face-frown-open\": $fa-var-face-frown-open,\n  \"frown-open\": $fa-var-frown-open,\n  \"hand-point-up\": $fa-var-hand-point-up,\n  \"money-bill\": $fa-var-money-bill,\n  \"bookmark\": $fa-var-bookmark,\n  \"align-justify\": $fa-var-align-justify,\n  \"umbrella-beach\": $fa-var-umbrella-beach,\n  \"helmet-un\": $fa-var-helmet-un,\n  \"bullseye\": $fa-var-bullseye,\n  \"bacon\": $fa-var-bacon,\n  \"hand-point-down\": $fa-var-hand-point-down,\n  \"arrow-up-from-bracket\": $fa-var-arrow-up-from-bracket,\n  \"folder\": $fa-var-folder,\n  \"folder-blank\": $fa-var-folder-blank,\n  \"file-waveform\": $fa-var-file-waveform,\n  \"file-medical-alt\": $fa-var-file-medical-alt,\n  \"radiation\": $fa-var-radiation,\n  \"chart-simple\": $fa-var-chart-simple,\n  \"mars-stroke\": $fa-var-mars-stroke,\n  \"vial\": $fa-var-vial,\n  \"gauge\": $fa-var-gauge,\n  \"dashboard\": $fa-var-dashboard,\n  \"gauge-med\": $fa-var-gauge-med,\n  \"tachometer-alt-average\": $fa-var-tachometer-alt-average,\n  \"wand-magic-sparkles\": $fa-var-wand-magic-sparkles,\n  \"magic-wand-sparkles\": $fa-var-magic-wand-sparkles,\n  \"e\": $fa-var-e,\n  \"pen-clip\": $fa-var-pen-clip,\n  \"pen-alt\": $fa-var-pen-alt,\n  \"bridge-circle-exclamation\": $fa-var-bridge-circle-exclamation,\n  \"user\": $fa-var-user,\n  \"school-circle-check\": $fa-var-school-circle-check,\n  \"dumpster\": $fa-var-dumpster,\n  \"van-shuttle\": $fa-var-van-shuttle,\n  \"shuttle-van\": $fa-var-shuttle-van,\n  \"building-user\": $fa-var-building-user,\n  \"square-caret-left\": $fa-var-square-caret-left,\n  \"caret-square-left\": $fa-var-caret-square-left,\n  \"highlighter\": $fa-var-highlighter,\n  \"key\": $fa-var-key,\n  \"bullhorn\": $fa-var-bullhorn,\n  \"globe\": $fa-var-globe,\n  \"synagogue\": $fa-var-synagogue,\n  \"person-half-dress\": $fa-var-person-half-dress,\n  \"road-bridge\": $fa-var-road-bridge,\n  \"location-arrow\": $fa-var-location-arrow,\n  \"c\": $fa-var-c,\n  \"tablet-button\": $fa-var-tablet-button,\n  \"building-lock\": $fa-var-building-lock,\n  \"pizza-slice\": $fa-var-pizza-slice,\n  \"money-bill-wave\": $fa-var-money-bill-wave,\n  \"chart-area\": $fa-var-chart-area,\n  \"area-chart\": $fa-var-area-chart,\n  \"house-flag\": $fa-var-house-flag,\n  \"person-circle-minus\": $fa-var-person-circle-minus,\n  \"ban\": $fa-var-ban,\n  \"cancel\": $fa-var-cancel,\n  \"camera-rotate\": $fa-var-camera-rotate,\n  \"spray-can-sparkles\": $fa-var-spray-can-sparkles,\n  \"air-freshener\": $fa-var-air-freshener,\n  \"star\": $fa-var-star,\n  \"repeat\": $fa-var-repeat,\n  \"cross\": $fa-var-cross,\n  \"box\": $fa-var-box,\n  \"venus-mars\": $fa-var-venus-mars,\n  \"arrow-pointer\": $fa-var-arrow-pointer,\n  \"mouse-pointer\": $fa-var-mouse-pointer,\n  \"maximize\": $fa-var-maximize,\n  \"expand-arrows-alt\": $fa-var-expand-arrows-alt,\n  \"charging-station\": $fa-var-charging-station,\n  \"shapes\": $fa-var-shapes,\n  \"triangle-circle-square\": $fa-var-triangle-circle-square,\n  \"shuffle\": $fa-var-shuffle,\n  \"random\": $fa-var-random,\n  \"person-running\": $fa-var-person-running,\n  \"running\": $fa-var-running,\n  \"mobile-retro\": $fa-var-mobile-retro,\n  \"grip-lines-vertical\": $fa-var-grip-lines-vertical,\n  \"spider\": $fa-var-spider,\n  \"hands-bound\": $fa-var-hands-bound,\n  \"file-invoice-dollar\": $fa-var-file-invoice-dollar,\n  \"plane-circle-exclamation\": $fa-var-plane-circle-exclamation,\n  \"x-ray\": $fa-var-x-ray,\n  \"spell-check\": $fa-var-spell-check,\n  \"slash\": $fa-var-slash,\n  \"computer-mouse\": $fa-var-computer-mouse,\n  \"mouse\": $fa-var-mouse,\n  \"arrow-right-to-bracket\": $fa-var-arrow-right-to-bracket,\n  \"sign-in\": $fa-var-sign-in,\n  \"shop-slash\": $fa-var-shop-slash,\n  \"store-alt-slash\": $fa-var-store-alt-slash,\n  \"server\": $fa-var-server,\n  \"virus-covid-slash\": $fa-var-virus-covid-slash,\n  \"shop-lock\": $fa-var-shop-lock,\n  \"hourglass-start\": $fa-var-hourglass-start,\n  \"hourglass-1\": $fa-var-hourglass-1,\n  \"blender-phone\": $fa-var-blender-phone,\n  \"building-wheat\": $fa-var-building-wheat,\n  \"person-breastfeeding\": $fa-var-person-breastfeeding,\n  \"right-to-bracket\": $fa-var-right-to-bracket,\n  \"sign-in-alt\": $fa-var-sign-in-alt,\n  \"venus\": $fa-var-venus,\n  \"passport\": $fa-var-passport,\n  \"heart-pulse\": $fa-var-heart-pulse,\n  \"heartbeat\": $fa-var-heartbeat,\n  \"people-carry-box\": $fa-var-people-carry-box,\n  \"people-carry\": $fa-var-people-carry,\n  \"temperature-high\": $fa-var-temperature-high,\n  \"microchip\": $fa-var-microchip,\n  \"crown\": $fa-var-crown,\n  \"weight-hanging\": $fa-var-weight-hanging,\n  \"xmarks-lines\": $fa-var-xmarks-lines,\n  \"file-prescription\": $fa-var-file-prescription,\n  \"weight-scale\": $fa-var-weight-scale,\n  \"weight\": $fa-var-weight,\n  \"user-group\": $fa-var-user-group,\n  \"user-friends\": $fa-var-user-friends,\n  \"arrow-up-a-z\": $fa-var-arrow-up-a-z,\n  \"sort-alpha-up\": $fa-var-sort-alpha-up,\n  \"chess-knight\": $fa-var-chess-knight,\n  \"face-laugh-squint\": $fa-var-face-laugh-squint,\n  \"laugh-squint\": $fa-var-laugh-squint,\n  \"wheelchair\": $fa-var-wheelchair,\n  \"circle-arrow-up\": $fa-var-circle-arrow-up,\n  \"arrow-circle-up\": $fa-var-arrow-circle-up,\n  \"toggle-on\": $fa-var-toggle-on,\n  \"person-walking\": $fa-var-person-walking,\n  \"walking\": $fa-var-walking,\n  \"l\": $fa-var-l,\n  \"fire\": $fa-var-fire,\n  \"bed-pulse\": $fa-var-bed-pulse,\n  \"procedures\": $fa-var-procedures,\n  \"shuttle-space\": $fa-var-shuttle-space,\n  \"space-shuttle\": $fa-var-space-shuttle,\n  \"face-laugh\": $fa-var-face-laugh,\n  \"laugh\": $fa-var-laugh,\n  \"folder-open\": $fa-var-folder-open,\n  \"heart-circle-plus\": $fa-var-heart-circle-plus,\n  \"code-fork\": $fa-var-code-fork,\n  \"city\": $fa-var-city,\n  \"microphone-lines\": $fa-var-microphone-lines,\n  \"microphone-alt\": $fa-var-microphone-alt,\n  \"pepper-hot\": $fa-var-pepper-hot,\n  \"unlock\": $fa-var-unlock,\n  \"colon-sign\": $fa-var-colon-sign,\n  \"headset\": $fa-var-headset,\n  \"store-slash\": $fa-var-store-slash,\n  \"road-circle-xmark\": $fa-var-road-circle-xmark,\n  \"user-minus\": $fa-var-user-minus,\n  \"mars-stroke-up\": $fa-var-mars-stroke-up,\n  \"mars-stroke-v\": $fa-var-mars-stroke-v,\n  \"champagne-glasses\": $fa-var-champagne-glasses,\n  \"glass-cheers\": $fa-var-glass-cheers,\n  \"clipboard\": $fa-var-clipboard,\n  \"house-circle-exclamation\": $fa-var-house-circle-exclamation,\n  \"file-arrow-up\": $fa-var-file-arrow-up,\n  \"file-upload\": $fa-var-file-upload,\n  \"wifi\": $fa-var-wifi,\n  \"wifi-3\": $fa-var-wifi-3,\n  \"wifi-strong\": $fa-var-wifi-strong,\n  \"bath\": $fa-var-bath,\n  \"bathtub\": $fa-var-bathtub,\n  \"underline\": $fa-var-underline,\n  \"user-pen\": $fa-var-user-pen,\n  \"user-edit\": $fa-var-user-edit,\n  \"signature\": $fa-var-signature,\n  \"stroopwafel\": $fa-var-stroopwafel,\n  \"bold\": $fa-var-bold,\n  \"anchor-lock\": $fa-var-anchor-lock,\n  \"building-ngo\": $fa-var-building-ngo,\n  \"manat-sign\": $fa-var-manat-sign,\n  \"not-equal\": $fa-var-not-equal,\n  \"border-top-left\": $fa-var-border-top-left,\n  \"border-style\": $fa-var-border-style,\n  \"map-location-dot\": $fa-var-map-location-dot,\n  \"map-marked-alt\": $fa-var-map-marked-alt,\n  \"jedi\": $fa-var-jedi,\n  \"square-poll-vertical\": $fa-var-square-poll-vertical,\n  \"poll\": $fa-var-poll,\n  \"mug-hot\": $fa-var-mug-hot,\n  \"car-battery\": $fa-var-car-battery,\n  \"battery-car\": $fa-var-battery-car,\n  \"gift\": $fa-var-gift,\n  \"dice-two\": $fa-var-dice-two,\n  \"chess-queen\": $fa-var-chess-queen,\n  \"glasses\": $fa-var-glasses,\n  \"chess-board\": $fa-var-chess-board,\n  \"building-circle-check\": $fa-var-building-circle-check,\n  \"person-chalkboard\": $fa-var-person-chalkboard,\n  \"mars-stroke-right\": $fa-var-mars-stroke-right,\n  \"mars-stroke-h\": $fa-var-mars-stroke-h,\n  \"hand-back-fist\": $fa-var-hand-back-fist,\n  \"hand-rock\": $fa-var-hand-rock,\n  \"square-caret-up\": $fa-var-square-caret-up,\n  \"caret-square-up\": $fa-var-caret-square-up,\n  \"cloud-showers-water\": $fa-var-cloud-showers-water,\n  \"chart-bar\": $fa-var-chart-bar,\n  \"bar-chart\": $fa-var-bar-chart,\n  \"hands-bubbles\": $fa-var-hands-bubbles,\n  \"hands-wash\": $fa-var-hands-wash,\n  \"less-than-equal\": $fa-var-less-than-equal,\n  \"train\": $fa-var-train,\n  \"eye-low-vision\": $fa-var-eye-low-vision,\n  \"low-vision\": $fa-var-low-vision,\n  \"crow\": $fa-var-crow,\n  \"sailboat\": $fa-var-sailboat,\n  \"window-restore\": $fa-var-window-restore,\n  \"square-plus\": $fa-var-square-plus,\n  \"plus-square\": $fa-var-plus-square,\n  \"torii-gate\": $fa-var-torii-gate,\n  \"frog\": $fa-var-frog,\n  \"bucket\": $fa-var-bucket,\n  \"image\": $fa-var-image,\n  \"microphone\": $fa-var-microphone,\n  \"cow\": $fa-var-cow,\n  \"caret-up\": $fa-var-caret-up,\n  \"screwdriver\": $fa-var-screwdriver,\n  \"folder-closed\": $fa-var-folder-closed,\n  \"house-tsunami\": $fa-var-house-tsunami,\n  \"square-nfi\": $fa-var-square-nfi,\n  \"arrow-up-from-ground-water\": $fa-var-arrow-up-from-ground-water,\n  \"martini-glass\": $fa-var-martini-glass,\n  \"glass-martini-alt\": $fa-var-glass-martini-alt,\n  \"rotate-left\": $fa-var-rotate-left,\n  \"rotate-back\": $fa-var-rotate-back,\n  \"rotate-backward\": $fa-var-rotate-backward,\n  \"undo-alt\": $fa-var-undo-alt,\n  \"table-columns\": $fa-var-table-columns,\n  \"columns\": $fa-var-columns,\n  \"lemon\": $fa-var-lemon,\n  \"head-side-mask\": $fa-var-head-side-mask,\n  \"handshake\": $fa-var-handshake,\n  \"gem\": $fa-var-gem,\n  \"dolly\": $fa-var-dolly,\n  \"dolly-box\": $fa-var-dolly-box,\n  \"smoking\": $fa-var-smoking,\n  \"minimize\": $fa-var-minimize,\n  \"compress-arrows-alt\": $fa-var-compress-arrows-alt,\n  \"monument\": $fa-var-monument,\n  \"snowplow\": $fa-var-snowplow,\n  \"angles-right\": $fa-var-angles-right,\n  \"angle-double-right\": $fa-var-angle-double-right,\n  \"cannabis\": $fa-var-cannabis,\n  \"circle-play\": $fa-var-circle-play,\n  \"play-circle\": $fa-var-play-circle,\n  \"tablets\": $fa-var-tablets,\n  \"ethernet\": $fa-var-ethernet,\n  \"euro-sign\": $fa-var-euro-sign,\n  \"eur\": $fa-var-eur,\n  \"euro\": $fa-var-euro,\n  \"chair\": $fa-var-chair,\n  \"circle-check\": $fa-var-circle-check,\n  \"check-circle\": $fa-var-check-circle,\n  \"circle-stop\": $fa-var-circle-stop,\n  \"stop-circle\": $fa-var-stop-circle,\n  \"compass-drafting\": $fa-var-compass-drafting,\n  \"drafting-compass\": $fa-var-drafting-compass,\n  \"plate-wheat\": $fa-var-plate-wheat,\n  \"icicles\": $fa-var-icicles,\n  \"person-shelter\": $fa-var-person-shelter,\n  \"neuter\": $fa-var-neuter,\n  \"id-badge\": $fa-var-id-badge,\n  \"marker\": $fa-var-marker,\n  \"face-laugh-beam\": $fa-var-face-laugh-beam,\n  \"laugh-beam\": $fa-var-laugh-beam,\n  \"helicopter-symbol\": $fa-var-helicopter-symbol,\n  \"universal-access\": $fa-var-universal-access,\n  \"circle-chevron-up\": $fa-var-circle-chevron-up,\n  \"chevron-circle-up\": $fa-var-chevron-circle-up,\n  \"lari-sign\": $fa-var-lari-sign,\n  \"volcano\": $fa-var-volcano,\n  \"person-walking-dashed-line-arrow-right\": $fa-var-person-walking-dashed-line-arrow-right,\n  \"sterling-sign\": $fa-var-sterling-sign,\n  \"gbp\": $fa-var-gbp,\n  \"pound-sign\": $fa-var-pound-sign,\n  \"viruses\": $fa-var-viruses,\n  \"square-person-confined\": $fa-var-square-person-confined,\n  \"user-tie\": $fa-var-user-tie,\n  \"arrow-down-long\": $fa-var-arrow-down-long,\n  \"long-arrow-down\": $fa-var-long-arrow-down,\n  \"tent-arrow-down-to-line\": $fa-var-tent-arrow-down-to-line,\n  \"certificate\": $fa-var-certificate,\n  \"reply-all\": $fa-var-reply-all,\n  \"mail-reply-all\": $fa-var-mail-reply-all,\n  \"suitcase\": $fa-var-suitcase,\n  \"person-skating\": $fa-var-person-skating,\n  \"skating\": $fa-var-skating,\n  \"filter-circle-dollar\": $fa-var-filter-circle-dollar,\n  \"funnel-dollar\": $fa-var-funnel-dollar,\n  \"camera-retro\": $fa-var-camera-retro,\n  \"circle-arrow-down\": $fa-var-circle-arrow-down,\n  \"arrow-circle-down\": $fa-var-arrow-circle-down,\n  \"file-import\": $fa-var-file-import,\n  \"arrow-right-to-file\": $fa-var-arrow-right-to-file,\n  \"square-arrow-up-right\": $fa-var-square-arrow-up-right,\n  \"external-link-square\": $fa-var-external-link-square,\n  \"box-open\": $fa-var-box-open,\n  \"scroll\": $fa-var-scroll,\n  \"spa\": $fa-var-spa,\n  \"location-pin-lock\": $fa-var-location-pin-lock,\n  \"pause\": $fa-var-pause,\n  \"hill-avalanche\": $fa-var-hill-avalanche,\n  \"temperature-empty\": $fa-var-temperature-empty,\n  \"temperature-0\": $fa-var-temperature-0,\n  \"thermometer-0\": $fa-var-thermometer-0,\n  \"thermometer-empty\": $fa-var-thermometer-empty,\n  \"bomb\": $fa-var-bomb,\n  \"registered\": $fa-var-registered,\n  \"address-card\": $fa-var-address-card,\n  \"contact-card\": $fa-var-contact-card,\n  \"vcard\": $fa-var-vcard,\n  \"scale-unbalanced-flip\": $fa-var-scale-unbalanced-flip,\n  \"balance-scale-right\": $fa-var-balance-scale-right,\n  \"subscript\": $fa-var-subscript,\n  \"diamond-turn-right\": $fa-var-diamond-turn-right,\n  \"directions\": $fa-var-directions,\n  \"burst\": $fa-var-burst,\n  \"house-laptop\": $fa-var-house-laptop,\n  \"laptop-house\": $fa-var-laptop-house,\n  \"face-tired\": $fa-var-face-tired,\n  \"tired\": $fa-var-tired,\n  \"money-bills\": $fa-var-money-bills,\n  \"smog\": $fa-var-smog,\n  \"crutch\": $fa-var-crutch,\n  \"cloud-arrow-up\": $fa-var-cloud-arrow-up,\n  \"cloud-upload\": $fa-var-cloud-upload,\n  \"cloud-upload-alt\": $fa-var-cloud-upload-alt,\n  \"palette\": $fa-var-palette,\n  \"arrows-turn-right\": $fa-var-arrows-turn-right,\n  \"vest\": $fa-var-vest,\n  \"ferry\": $fa-var-ferry,\n  \"arrows-down-to-people\": $fa-var-arrows-down-to-people,\n  \"seedling\": $fa-var-seedling,\n  \"sprout\": $fa-var-sprout,\n  \"left-right\": $fa-var-left-right,\n  \"arrows-alt-h\": $fa-var-arrows-alt-h,\n  \"boxes-packing\": $fa-var-boxes-packing,\n  \"circle-arrow-left\": $fa-var-circle-arrow-left,\n  \"arrow-circle-left\": $fa-var-arrow-circle-left,\n  \"group-arrows-rotate\": $fa-var-group-arrows-rotate,\n  \"bowl-food\": $fa-var-bowl-food,\n  \"candy-cane\": $fa-var-candy-cane,\n  \"arrow-down-wide-short\": $fa-var-arrow-down-wide-short,\n  \"sort-amount-asc\": $fa-var-sort-amount-asc,\n  \"sort-amount-down\": $fa-var-sort-amount-down,\n  \"cloud-bolt\": $fa-var-cloud-bolt,\n  \"thunderstorm\": $fa-var-thunderstorm,\n  \"text-slash\": $fa-var-text-slash,\n  \"remove-format\": $fa-var-remove-format,\n  \"face-smile-wink\": $fa-var-face-smile-wink,\n  \"smile-wink\": $fa-var-smile-wink,\n  \"file-word\": $fa-var-file-word,\n  \"file-powerpoint\": $fa-var-file-powerpoint,\n  \"arrows-left-right\": $fa-var-arrows-left-right,\n  \"arrows-h\": $fa-var-arrows-h,\n  \"house-lock\": $fa-var-house-lock,\n  \"cloud-arrow-down\": $fa-var-cloud-arrow-down,\n  \"cloud-download\": $fa-var-cloud-download,\n  \"cloud-download-alt\": $fa-var-cloud-download-alt,\n  \"children\": $fa-var-children,\n  \"chalkboard\": $fa-var-chalkboard,\n  \"blackboard\": $fa-var-blackboard,\n  \"user-large-slash\": $fa-var-user-large-slash,\n  \"user-alt-slash\": $fa-var-user-alt-slash,\n  \"envelope-open\": $fa-var-envelope-open,\n  \"handshake-simple-slash\": $fa-var-handshake-simple-slash,\n  \"handshake-alt-slash\": $fa-var-handshake-alt-slash,\n  \"mattress-pillow\": $fa-var-mattress-pillow,\n  \"guarani-sign\": $fa-var-guarani-sign,\n  \"arrows-rotate\": $fa-var-arrows-rotate,\n  \"refresh\": $fa-var-refresh,\n  \"sync\": $fa-var-sync,\n  \"fire-extinguisher\": $fa-var-fire-extinguisher,\n  \"cruzeiro-sign\": $fa-var-cruzeiro-sign,\n  \"greater-than-equal\": $fa-var-greater-than-equal,\n  \"shield-halved\": $fa-var-shield-halved,\n  \"shield-alt\": $fa-var-shield-alt,\n  \"book-atlas\": $fa-var-book-atlas,\n  \"atlas\": $fa-var-atlas,\n  \"virus\": $fa-var-virus,\n  \"envelope-circle-check\": $fa-var-envelope-circle-check,\n  \"layer-group\": $fa-var-layer-group,\n  \"arrows-to-dot\": $fa-var-arrows-to-dot,\n  \"archway\": $fa-var-archway,\n  \"heart-circle-check\": $fa-var-heart-circle-check,\n  \"house-chimney-crack\": $fa-var-house-chimney-crack,\n  \"house-damage\": $fa-var-house-damage,\n  \"file-zipper\": $fa-var-file-zipper,\n  \"file-archive\": $fa-var-file-archive,\n  \"square\": $fa-var-square,\n  \"martini-glass-empty\": $fa-var-martini-glass-empty,\n  \"glass-martini\": $fa-var-glass-martini,\n  \"couch\": $fa-var-couch,\n  \"cedi-sign\": $fa-var-cedi-sign,\n  \"italic\": $fa-var-italic,\n  \"table-cells-column-lock\": $fa-var-table-cells-column-lock,\n  \"church\": $fa-var-church,\n  \"comments-dollar\": $fa-var-comments-dollar,\n  \"democrat\": $fa-var-democrat,\n  \"z\": $fa-var-z,\n  \"person-skiing\": $fa-var-person-skiing,\n  \"skiing\": $fa-var-skiing,\n  \"road-lock\": $fa-var-road-lock,\n  \"a\": $fa-var-a,\n  \"temperature-arrow-down\": $fa-var-temperature-arrow-down,\n  \"temperature-down\": $fa-var-temperature-down,\n  \"feather-pointed\": $fa-var-feather-pointed,\n  \"feather-alt\": $fa-var-feather-alt,\n  \"p\": $fa-var-p,\n  \"snowflake\": $fa-var-snowflake,\n  \"newspaper\": $fa-var-newspaper,\n  \"rectangle-ad\": $fa-var-rectangle-ad,\n  \"ad\": $fa-var-ad,\n  \"circle-arrow-right\": $fa-var-circle-arrow-right,\n  \"arrow-circle-right\": $fa-var-arrow-circle-right,\n  \"filter-circle-xmark\": $fa-var-filter-circle-xmark,\n  \"locust\": $fa-var-locust,\n  \"sort\": $fa-var-sort,\n  \"unsorted\": $fa-var-unsorted,\n  \"list-ol\": $fa-var-list-ol,\n  \"list-1-2\": $fa-var-list-1-2,\n  \"list-numeric\": $fa-var-list-numeric,\n  \"person-dress-burst\": $fa-var-person-dress-burst,\n  \"money-check-dollar\": $fa-var-money-check-dollar,\n  \"money-check-alt\": $fa-var-money-check-alt,\n  \"vector-square\": $fa-var-vector-square,\n  \"bread-slice\": $fa-var-bread-slice,\n  \"language\": $fa-var-language,\n  \"face-kiss-wink-heart\": $fa-var-face-kiss-wink-heart,\n  \"kiss-wink-heart\": $fa-var-kiss-wink-heart,\n  \"filter\": $fa-var-filter,\n  \"question\": $fa-var-question,\n  \"file-signature\": $fa-var-file-signature,\n  \"up-down-left-right\": $fa-var-up-down-left-right,\n  \"arrows-alt\": $fa-var-arrows-alt,\n  \"house-chimney-user\": $fa-var-house-chimney-user,\n  \"hand-holding-heart\": $fa-var-hand-holding-heart,\n  \"puzzle-piece\": $fa-var-puzzle-piece,\n  \"money-check\": $fa-var-money-check,\n  \"star-half-stroke\": $fa-var-star-half-stroke,\n  \"star-half-alt\": $fa-var-star-half-alt,\n  \"code\": $fa-var-code,\n  \"whiskey-glass\": $fa-var-whiskey-glass,\n  \"glass-whiskey\": $fa-var-glass-whiskey,\n  \"building-circle-exclamation\": $fa-var-building-circle-exclamation,\n  \"magnifying-glass-chart\": $fa-var-magnifying-glass-chart,\n  \"arrow-up-right-from-square\": $fa-var-arrow-up-right-from-square,\n  \"external-link\": $fa-var-external-link,\n  \"cubes-stacked\": $fa-var-cubes-stacked,\n  \"won-sign\": $fa-var-won-sign,\n  \"krw\": $fa-var-krw,\n  \"won\": $fa-var-won,\n  \"virus-covid\": $fa-var-virus-covid,\n  \"austral-sign\": $fa-var-austral-sign,\n  \"f\": $fa-var-f,\n  \"leaf\": $fa-var-leaf,\n  \"road\": $fa-var-road,\n  \"taxi\": $fa-var-taxi,\n  \"cab\": $fa-var-cab,\n  \"person-circle-plus\": $fa-var-person-circle-plus,\n  \"chart-pie\": $fa-var-chart-pie,\n  \"pie-chart\": $fa-var-pie-chart,\n  \"bolt-lightning\": $fa-var-bolt-lightning,\n  \"sack-xmark\": $fa-var-sack-xmark,\n  \"file-excel\": $fa-var-file-excel,\n  \"file-contract\": $fa-var-file-contract,\n  \"fish-fins\": $fa-var-fish-fins,\n  \"building-flag\": $fa-var-building-flag,\n  \"face-grin-beam\": $fa-var-face-grin-beam,\n  \"grin-beam\": $fa-var-grin-beam,\n  \"object-ungroup\": $fa-var-object-ungroup,\n  \"poop\": $fa-var-poop,\n  \"location-pin\": $fa-var-location-pin,\n  \"map-marker\": $fa-var-map-marker,\n  \"kaaba\": $fa-var-kaaba,\n  \"toilet-paper\": $fa-var-toilet-paper,\n  \"helmet-safety\": $fa-var-helmet-safety,\n  \"hard-hat\": $fa-var-hard-hat,\n  \"hat-hard\": $fa-var-hat-hard,\n  \"eject\": $fa-var-eject,\n  \"circle-right\": $fa-var-circle-right,\n  \"arrow-alt-circle-right\": $fa-var-arrow-alt-circle-right,\n  \"plane-circle-check\": $fa-var-plane-circle-check,\n  \"face-rolling-eyes\": $fa-var-face-rolling-eyes,\n  \"meh-rolling-eyes\": $fa-var-meh-rolling-eyes,\n  \"object-group\": $fa-var-object-group,\n  \"chart-line\": $fa-var-chart-line,\n  \"line-chart\": $fa-var-line-chart,\n  \"mask-ventilator\": $fa-var-mask-ventilator,\n  \"arrow-right\": $fa-var-arrow-right,\n  \"signs-post\": $fa-var-signs-post,\n  \"map-signs\": $fa-var-map-signs,\n  \"cash-register\": $fa-var-cash-register,\n  \"person-circle-question\": $fa-var-person-circle-question,\n  \"h\": $fa-var-h,\n  \"tarp\": $fa-var-tarp,\n  \"screwdriver-wrench\": $fa-var-screwdriver-wrench,\n  \"tools\": $fa-var-tools,\n  \"arrows-to-eye\": $fa-var-arrows-to-eye,\n  \"plug-circle-bolt\": $fa-var-plug-circle-bolt,\n  \"heart\": $fa-var-heart,\n  \"mars-and-venus\": $fa-var-mars-and-venus,\n  \"house-user\": $fa-var-house-user,\n  \"home-user\": $fa-var-home-user,\n  \"dumpster-fire\": $fa-var-dumpster-fire,\n  \"house-crack\": $fa-var-house-crack,\n  \"martini-glass-citrus\": $fa-var-martini-glass-citrus,\n  \"cocktail\": $fa-var-cocktail,\n  \"face-surprise\": $fa-var-face-surprise,\n  \"surprise\": $fa-var-surprise,\n  \"bottle-water\": $fa-var-bottle-water,\n  \"circle-pause\": $fa-var-circle-pause,\n  \"pause-circle\": $fa-var-pause-circle,\n  \"toilet-paper-slash\": $fa-var-toilet-paper-slash,\n  \"apple-whole\": $fa-var-apple-whole,\n  \"apple-alt\": $fa-var-apple-alt,\n  \"kitchen-set\": $fa-var-kitchen-set,\n  \"r\": $fa-var-r,\n  \"temperature-quarter\": $fa-var-temperature-quarter,\n  \"temperature-1\": $fa-var-temperature-1,\n  \"thermometer-1\": $fa-var-thermometer-1,\n  \"thermometer-quarter\": $fa-var-thermometer-quarter,\n  \"cube\": $fa-var-cube,\n  \"bitcoin-sign\": $fa-var-bitcoin-sign,\n  \"shield-dog\": $fa-var-shield-dog,\n  \"solar-panel\": $fa-var-solar-panel,\n  \"lock-open\": $fa-var-lock-open,\n  \"elevator\": $fa-var-elevator,\n  \"money-bill-transfer\": $fa-var-money-bill-transfer,\n  \"money-bill-trend-up\": $fa-var-money-bill-trend-up,\n  \"house-flood-water-circle-arrow-right\": $fa-var-house-flood-water-circle-arrow-right,\n  \"square-poll-horizontal\": $fa-var-square-poll-horizontal,\n  \"poll-h\": $fa-var-poll-h,\n  \"circle\": $fa-var-circle,\n  \"backward-fast\": $fa-var-backward-fast,\n  \"fast-backward\": $fa-var-fast-backward,\n  \"recycle\": $fa-var-recycle,\n  \"user-astronaut\": $fa-var-user-astronaut,\n  \"plane-slash\": $fa-var-plane-slash,\n  \"trademark\": $fa-var-trademark,\n  \"basketball\": $fa-var-basketball,\n  \"basketball-ball\": $fa-var-basketball-ball,\n  \"satellite-dish\": $fa-var-satellite-dish,\n  \"circle-up\": $fa-var-circle-up,\n  \"arrow-alt-circle-up\": $fa-var-arrow-alt-circle-up,\n  \"mobile-screen-button\": $fa-var-mobile-screen-button,\n  \"mobile-alt\": $fa-var-mobile-alt,\n  \"volume-high\": $fa-var-volume-high,\n  \"volume-up\": $fa-var-volume-up,\n  \"users-rays\": $fa-var-users-rays,\n  \"wallet\": $fa-var-wallet,\n  \"clipboard-check\": $fa-var-clipboard-check,\n  \"file-audio\": $fa-var-file-audio,\n  \"burger\": $fa-var-burger,\n  \"hamburger\": $fa-var-hamburger,\n  \"wrench\": $fa-var-wrench,\n  \"bugs\": $fa-var-bugs,\n  \"rupee-sign\": $fa-var-rupee-sign,\n  \"rupee\": $fa-var-rupee,\n  \"file-image\": $fa-var-file-image,\n  \"circle-question\": $fa-var-circle-question,\n  \"question-circle\": $fa-var-question-circle,\n  \"plane-departure\": $fa-var-plane-departure,\n  \"handshake-slash\": $fa-var-handshake-slash,\n  \"book-bookmark\": $fa-var-book-bookmark,\n  \"code-branch\": $fa-var-code-branch,\n  \"hat-cowboy\": $fa-var-hat-cowboy,\n  \"bridge\": $fa-var-bridge,\n  \"phone-flip\": $fa-var-phone-flip,\n  \"phone-alt\": $fa-var-phone-alt,\n  \"truck-front\": $fa-var-truck-front,\n  \"cat\": $fa-var-cat,\n  \"anchor-circle-exclamation\": $fa-var-anchor-circle-exclamation,\n  \"truck-field\": $fa-var-truck-field,\n  \"route\": $fa-var-route,\n  \"clipboard-question\": $fa-var-clipboard-question,\n  \"panorama\": $fa-var-panorama,\n  \"comment-medical\": $fa-var-comment-medical,\n  \"teeth-open\": $fa-var-teeth-open,\n  \"file-circle-minus\": $fa-var-file-circle-minus,\n  \"tags\": $fa-var-tags,\n  \"wine-glass\": $fa-var-wine-glass,\n  \"forward-fast\": $fa-var-forward-fast,\n  \"fast-forward\": $fa-var-fast-forward,\n  \"face-meh-blank\": $fa-var-face-meh-blank,\n  \"meh-blank\": $fa-var-meh-blank,\n  \"square-parking\": $fa-var-square-parking,\n  \"parking\": $fa-var-parking,\n  \"house-signal\": $fa-var-house-signal,\n  \"bars-progress\": $fa-var-bars-progress,\n  \"tasks-alt\": $fa-var-tasks-alt,\n  \"faucet-drip\": $fa-var-faucet-drip,\n  \"cart-flatbed\": $fa-var-cart-flatbed,\n  \"dolly-flatbed\": $fa-var-dolly-flatbed,\n  \"ban-smoking\": $fa-var-ban-smoking,\n  \"smoking-ban\": $fa-var-smoking-ban,\n  \"terminal\": $fa-var-terminal,\n  \"mobile-button\": $fa-var-mobile-button,\n  \"house-medical-flag\": $fa-var-house-medical-flag,\n  \"basket-shopping\": $fa-var-basket-shopping,\n  \"shopping-basket\": $fa-var-shopping-basket,\n  \"tape\": $fa-var-tape,\n  \"bus-simple\": $fa-var-bus-simple,\n  \"bus-alt\": $fa-var-bus-alt,\n  \"eye\": $fa-var-eye,\n  \"face-sad-cry\": $fa-var-face-sad-cry,\n  \"sad-cry\": $fa-var-sad-cry,\n  \"audio-description\": $fa-var-audio-description,\n  \"person-military-to-person\": $fa-var-person-military-to-person,\n  \"file-shield\": $fa-var-file-shield,\n  \"user-slash\": $fa-var-user-slash,\n  \"pen\": $fa-var-pen,\n  \"tower-observation\": $fa-var-tower-observation,\n  \"file-code\": $fa-var-file-code,\n  \"signal\": $fa-var-signal,\n  \"signal-5\": $fa-var-signal-5,\n  \"signal-perfect\": $fa-var-signal-perfect,\n  \"bus\": $fa-var-bus,\n  \"heart-circle-xmark\": $fa-var-heart-circle-xmark,\n  \"house-chimney\": $fa-var-house-chimney,\n  \"home-lg\": $fa-var-home-lg,\n  \"window-maximize\": $fa-var-window-maximize,\n  \"face-frown\": $fa-var-face-frown,\n  \"frown\": $fa-var-frown,\n  \"prescription\": $fa-var-prescription,\n  \"shop\": $fa-var-shop,\n  \"store-alt\": $fa-var-store-alt,\n  \"floppy-disk\": $fa-var-floppy-disk,\n  \"save\": $fa-var-save,\n  \"vihara\": $fa-var-vihara,\n  \"scale-unbalanced\": $fa-var-scale-unbalanced,\n  \"balance-scale-left\": $fa-var-balance-scale-left,\n  \"sort-up\": $fa-var-sort-up,\n  \"sort-asc\": $fa-var-sort-asc,\n  \"comment-dots\": $fa-var-comment-dots,\n  \"commenting\": $fa-var-commenting,\n  \"plant-wilt\": $fa-var-plant-wilt,\n  \"diamond\": $fa-var-diamond,\n  \"face-grin-squint\": $fa-var-face-grin-squint,\n  \"grin-squint\": $fa-var-grin-squint,\n  \"hand-holding-dollar\": $fa-var-hand-holding-dollar,\n  \"hand-holding-usd\": $fa-var-hand-holding-usd,\n  \"bacterium\": $fa-var-bacterium,\n  \"hand-pointer\": $fa-var-hand-pointer,\n  \"drum-steelpan\": $fa-var-drum-steelpan,\n  \"hand-scissors\": $fa-var-hand-scissors,\n  \"hands-praying\": $fa-var-hands-praying,\n  \"praying-hands\": $fa-var-praying-hands,\n  \"arrow-rotate-right\": $fa-var-arrow-rotate-right,\n  \"arrow-right-rotate\": $fa-var-arrow-right-rotate,\n  \"arrow-rotate-forward\": $fa-var-arrow-rotate-forward,\n  \"redo\": $fa-var-redo,\n  \"biohazard\": $fa-var-biohazard,\n  \"location-crosshairs\": $fa-var-location-crosshairs,\n  \"location\": $fa-var-location,\n  \"mars-double\": $fa-var-mars-double,\n  \"child-dress\": $fa-var-child-dress,\n  \"users-between-lines\": $fa-var-users-between-lines,\n  \"lungs-virus\": $fa-var-lungs-virus,\n  \"face-grin-tears\": $fa-var-face-grin-tears,\n  \"grin-tears\": $fa-var-grin-tears,\n  \"phone\": $fa-var-phone,\n  \"calendar-xmark\": $fa-var-calendar-xmark,\n  \"calendar-times\": $fa-var-calendar-times,\n  \"child-reaching\": $fa-var-child-reaching,\n  \"head-side-virus\": $fa-var-head-side-virus,\n  \"user-gear\": $fa-var-user-gear,\n  \"user-cog\": $fa-var-user-cog,\n  \"arrow-up-1-9\": $fa-var-arrow-up-1-9,\n  \"sort-numeric-up\": $fa-var-sort-numeric-up,\n  \"door-closed\": $fa-var-door-closed,\n  \"shield-virus\": $fa-var-shield-virus,\n  \"dice-six\": $fa-var-dice-six,\n  \"mosquito-net\": $fa-var-mosquito-net,\n  \"bridge-water\": $fa-var-bridge-water,\n  \"person-booth\": $fa-var-person-booth,\n  \"text-width\": $fa-var-text-width,\n  \"hat-wizard\": $fa-var-hat-wizard,\n  \"pen-fancy\": $fa-var-pen-fancy,\n  \"person-digging\": $fa-var-person-digging,\n  \"digging\": $fa-var-digging,\n  \"trash\": $fa-var-trash,\n  \"gauge-simple\": $fa-var-gauge-simple,\n  \"gauge-simple-med\": $fa-var-gauge-simple-med,\n  \"tachometer-average\": $fa-var-tachometer-average,\n  \"book-medical\": $fa-var-book-medical,\n  \"poo\": $fa-var-poo,\n  \"quote-right\": $fa-var-quote-right,\n  \"quote-right-alt\": $fa-var-quote-right-alt,\n  \"shirt\": $fa-var-shirt,\n  \"t-shirt\": $fa-var-t-shirt,\n  \"tshirt\": $fa-var-tshirt,\n  \"cubes\": $fa-var-cubes,\n  \"divide\": $fa-var-divide,\n  \"tenge-sign\": $fa-var-tenge-sign,\n  \"tenge\": $fa-var-tenge,\n  \"headphones\": $fa-var-headphones,\n  \"hands-holding\": $fa-var-hands-holding,\n  \"hands-clapping\": $fa-var-hands-clapping,\n  \"republican\": $fa-var-republican,\n  \"arrow-left\": $fa-var-arrow-left,\n  \"person-circle-xmark\": $fa-var-person-circle-xmark,\n  \"ruler\": $fa-var-ruler,\n  \"align-left\": $fa-var-align-left,\n  \"dice-d6\": $fa-var-dice-d6,\n  \"restroom\": $fa-var-restroom,\n  \"j\": $fa-var-j,\n  \"users-viewfinder\": $fa-var-users-viewfinder,\n  \"file-video\": $fa-var-file-video,\n  \"up-right-from-square\": $fa-var-up-right-from-square,\n  \"external-link-alt\": $fa-var-external-link-alt,\n  \"table-cells\": $fa-var-table-cells,\n  \"th\": $fa-var-th,\n  \"file-pdf\": $fa-var-file-pdf,\n  \"book-bible\": $fa-var-book-bible,\n  \"bible\": $fa-var-bible,\n  \"o\": $fa-var-o,\n  \"suitcase-medical\": $fa-var-suitcase-medical,\n  \"medkit\": $fa-var-medkit,\n  \"user-secret\": $fa-var-user-secret,\n  \"otter\": $fa-var-otter,\n  \"person-dress\": $fa-var-person-dress,\n  \"female\": $fa-var-female,\n  \"comment-dollar\": $fa-var-comment-dollar,\n  \"business-time\": $fa-var-business-time,\n  \"briefcase-clock\": $fa-var-briefcase-clock,\n  \"table-cells-large\": $fa-var-table-cells-large,\n  \"th-large\": $fa-var-th-large,\n  \"book-tanakh\": $fa-var-book-tanakh,\n  \"tanakh\": $fa-var-tanakh,\n  \"phone-volume\": $fa-var-phone-volume,\n  \"volume-control-phone\": $fa-var-volume-control-phone,\n  \"hat-cowboy-side\": $fa-var-hat-cowboy-side,\n  \"clipboard-user\": $fa-var-clipboard-user,\n  \"child\": $fa-var-child,\n  \"lira-sign\": $fa-var-lira-sign,\n  \"satellite\": $fa-var-satellite,\n  \"plane-lock\": $fa-var-plane-lock,\n  \"tag\": $fa-var-tag,\n  \"comment\": $fa-var-comment,\n  \"cake-candles\": $fa-var-cake-candles,\n  \"birthday-cake\": $fa-var-birthday-cake,\n  \"cake\": $fa-var-cake,\n  \"envelope\": $fa-var-envelope,\n  \"angles-up\": $fa-var-angles-up,\n  \"angle-double-up\": $fa-var-angle-double-up,\n  \"paperclip\": $fa-var-paperclip,\n  \"arrow-right-to-city\": $fa-var-arrow-right-to-city,\n  \"ribbon\": $fa-var-ribbon,\n  \"lungs\": $fa-var-lungs,\n  \"arrow-up-9-1\": $fa-var-arrow-up-9-1,\n  \"sort-numeric-up-alt\": $fa-var-sort-numeric-up-alt,\n  \"litecoin-sign\": $fa-var-litecoin-sign,\n  \"border-none\": $fa-var-border-none,\n  \"circle-nodes\": $fa-var-circle-nodes,\n  \"parachute-box\": $fa-var-parachute-box,\n  \"indent\": $fa-var-indent,\n  \"truck-field-un\": $fa-var-truck-field-un,\n  \"hourglass\": $fa-var-hourglass,\n  \"hourglass-empty\": $fa-var-hourglass-empty,\n  \"mountain\": $fa-var-mountain,\n  \"user-doctor\": $fa-var-user-doctor,\n  \"user-md\": $fa-var-user-md,\n  \"circle-info\": $fa-var-circle-info,\n  \"info-circle\": $fa-var-info-circle,\n  \"cloud-meatball\": $fa-var-cloud-meatball,\n  \"camera\": $fa-var-camera,\n  \"camera-alt\": $fa-var-camera-alt,\n  \"square-virus\": $fa-var-square-virus,\n  \"meteor\": $fa-var-meteor,\n  \"car-on\": $fa-var-car-on,\n  \"sleigh\": $fa-var-sleigh,\n  \"arrow-down-1-9\": $fa-var-arrow-down-1-9,\n  \"sort-numeric-asc\": $fa-var-sort-numeric-asc,\n  \"sort-numeric-down\": $fa-var-sort-numeric-down,\n  \"hand-holding-droplet\": $fa-var-hand-holding-droplet,\n  \"hand-holding-water\": $fa-var-hand-holding-water,\n  \"water\": $fa-var-water,\n  \"calendar-check\": $fa-var-calendar-check,\n  \"braille\": $fa-var-braille,\n  \"prescription-bottle-medical\": $fa-var-prescription-bottle-medical,\n  \"prescription-bottle-alt\": $fa-var-prescription-bottle-alt,\n  \"landmark\": $fa-var-landmark,\n  \"truck\": $fa-var-truck,\n  \"crosshairs\": $fa-var-crosshairs,\n  \"person-cane\": $fa-var-person-cane,\n  \"tent\": $fa-var-tent,\n  \"vest-patches\": $fa-var-vest-patches,\n  \"check-double\": $fa-var-check-double,\n  \"arrow-down-a-z\": $fa-var-arrow-down-a-z,\n  \"sort-alpha-asc\": $fa-var-sort-alpha-asc,\n  \"sort-alpha-down\": $fa-var-sort-alpha-down,\n  \"money-bill-wheat\": $fa-var-money-bill-wheat,\n  \"cookie\": $fa-var-cookie,\n  \"arrow-rotate-left\": $fa-var-arrow-rotate-left,\n  \"arrow-left-rotate\": $fa-var-arrow-left-rotate,\n  \"arrow-rotate-back\": $fa-var-arrow-rotate-back,\n  \"arrow-rotate-backward\": $fa-var-arrow-rotate-backward,\n  \"undo\": $fa-var-undo,\n  \"hard-drive\": $fa-var-hard-drive,\n  \"hdd\": $fa-var-hdd,\n  \"face-grin-squint-tears\": $fa-var-face-grin-squint-tears,\n  \"grin-squint-tears\": $fa-var-grin-squint-tears,\n  \"dumbbell\": $fa-var-dumbbell,\n  \"rectangle-list\": $fa-var-rectangle-list,\n  \"list-alt\": $fa-var-list-alt,\n  \"tarp-droplet\": $fa-var-tarp-droplet,\n  \"house-medical-circle-check\": $fa-var-house-medical-circle-check,\n  \"person-skiing-nordic\": $fa-var-person-skiing-nordic,\n  \"skiing-nordic\": $fa-var-skiing-nordic,\n  \"calendar-plus\": $fa-var-calendar-plus,\n  \"plane-arrival\": $fa-var-plane-arrival,\n  \"circle-left\": $fa-var-circle-left,\n  \"arrow-alt-circle-left\": $fa-var-arrow-alt-circle-left,\n  \"train-subway\": $fa-var-train-subway,\n  \"subway\": $fa-var-subway,\n  \"chart-gantt\": $fa-var-chart-gantt,\n  \"indian-rupee-sign\": $fa-var-indian-rupee-sign,\n  \"indian-rupee\": $fa-var-indian-rupee,\n  \"inr\": $fa-var-inr,\n  \"crop-simple\": $fa-var-crop-simple,\n  \"crop-alt\": $fa-var-crop-alt,\n  \"money-bill-1\": $fa-var-money-bill-1,\n  \"money-bill-alt\": $fa-var-money-bill-alt,\n  \"left-long\": $fa-var-left-long,\n  \"long-arrow-alt-left\": $fa-var-long-arrow-alt-left,\n  \"dna\": $fa-var-dna,\n  \"virus-slash\": $fa-var-virus-slash,\n  \"minus\": $fa-var-minus,\n  \"subtract\": $fa-var-subtract,\n  \"chess\": $fa-var-chess,\n  \"arrow-left-long\": $fa-var-arrow-left-long,\n  \"long-arrow-left\": $fa-var-long-arrow-left,\n  \"plug-circle-check\": $fa-var-plug-circle-check,\n  \"street-view\": $fa-var-street-view,\n  \"franc-sign\": $fa-var-franc-sign,\n  \"volume-off\": $fa-var-volume-off,\n  \"hands-asl-interpreting\": $fa-var-hands-asl-interpreting,\n  \"american-sign-language-interpreting\": $fa-var-american-sign-language-interpreting,\n  \"asl-interpreting\": $fa-var-asl-interpreting,\n  \"hands-american-sign-language-interpreting\": $fa-var-hands-american-sign-language-interpreting,\n  \"gear\": $fa-var-gear,\n  \"cog\": $fa-var-cog,\n  \"droplet-slash\": $fa-var-droplet-slash,\n  \"tint-slash\": $fa-var-tint-slash,\n  \"mosque\": $fa-var-mosque,\n  \"mosquito\": $fa-var-mosquito,\n  \"star-of-david\": $fa-var-star-of-david,\n  \"person-military-rifle\": $fa-var-person-military-rifle,\n  \"cart-shopping\": $fa-var-cart-shopping,\n  \"shopping-cart\": $fa-var-shopping-cart,\n  \"vials\": $fa-var-vials,\n  \"plug-circle-plus\": $fa-var-plug-circle-plus,\n  \"place-of-worship\": $fa-var-place-of-worship,\n  \"grip-vertical\": $fa-var-grip-vertical,\n  \"arrow-turn-up\": $fa-var-arrow-turn-up,\n  \"level-up\": $fa-var-level-up,\n  \"u\": $fa-var-u,\n  \"square-root-variable\": $fa-var-square-root-variable,\n  \"square-root-alt\": $fa-var-square-root-alt,\n  \"clock\": $fa-var-clock,\n  \"clock-four\": $fa-var-clock-four,\n  \"backward-step\": $fa-var-backward-step,\n  \"step-backward\": $fa-var-step-backward,\n  \"pallet\": $fa-var-pallet,\n  \"faucet\": $fa-var-faucet,\n  \"baseball-bat-ball\": $fa-var-baseball-bat-ball,\n  \"s\": $fa-var-s,\n  \"timeline\": $fa-var-timeline,\n  \"keyboard\": $fa-var-keyboard,\n  \"caret-down\": $fa-var-caret-down,\n  \"house-chimney-medical\": $fa-var-house-chimney-medical,\n  \"clinic-medical\": $fa-var-clinic-medical,\n  \"temperature-three-quarters\": $fa-var-temperature-three-quarters,\n  \"temperature-3\": $fa-var-temperature-3,\n  \"thermometer-3\": $fa-var-thermometer-3,\n  \"thermometer-three-quarters\": $fa-var-thermometer-three-quarters,\n  \"mobile-screen\": $fa-var-mobile-screen,\n  \"mobile-android-alt\": $fa-var-mobile-android-alt,\n  \"plane-up\": $fa-var-plane-up,\n  \"piggy-bank\": $fa-var-piggy-bank,\n  \"battery-half\": $fa-var-battery-half,\n  \"battery-3\": $fa-var-battery-3,\n  \"mountain-city\": $fa-var-mountain-city,\n  \"coins\": $fa-var-coins,\n  \"khanda\": $fa-var-khanda,\n  \"sliders\": $fa-var-sliders,\n  \"sliders-h\": $fa-var-sliders-h,\n  \"folder-tree\": $fa-var-folder-tree,\n  \"network-wired\": $fa-var-network-wired,\n  \"map-pin\": $fa-var-map-pin,\n  \"hamsa\": $fa-var-hamsa,\n  \"cent-sign\": $fa-var-cent-sign,\n  \"flask\": $fa-var-flask,\n  \"person-pregnant\": $fa-var-person-pregnant,\n  \"wand-sparkles\": $fa-var-wand-sparkles,\n  \"ellipsis-vertical\": $fa-var-ellipsis-vertical,\n  \"ellipsis-v\": $fa-var-ellipsis-v,\n  \"ticket\": $fa-var-ticket,\n  \"power-off\": $fa-var-power-off,\n  \"right-long\": $fa-var-right-long,\n  \"long-arrow-alt-right\": $fa-var-long-arrow-alt-right,\n  \"flag-usa\": $fa-var-flag-usa,\n  \"laptop-file\": $fa-var-laptop-file,\n  \"tty\": $fa-var-tty,\n  \"teletype\": $fa-var-teletype,\n  \"diagram-next\": $fa-var-diagram-next,\n  \"person-rifle\": $fa-var-person-rifle,\n  \"house-medical-circle-exclamation\": $fa-var-house-medical-circle-exclamation,\n  \"closed-captioning\": $fa-var-closed-captioning,\n  \"person-hiking\": $fa-var-person-hiking,\n  \"hiking\": $fa-var-hiking,\n  \"venus-double\": $fa-var-venus-double,\n  \"images\": $fa-var-images,\n  \"calculator\": $fa-var-calculator,\n  \"people-pulling\": $fa-var-people-pulling,\n  \"n\": $fa-var-n,\n  \"cable-car\": $fa-var-cable-car,\n  \"tram\": $fa-var-tram,\n  \"cloud-rain\": $fa-var-cloud-rain,\n  \"building-circle-xmark\": $fa-var-building-circle-xmark,\n  \"ship\": $fa-var-ship,\n  \"arrows-down-to-line\": $fa-var-arrows-down-to-line,\n  \"download\": $fa-var-download,\n  \"face-grin\": $fa-var-face-grin,\n  \"grin\": $fa-var-grin,\n  \"delete-left\": $fa-var-delete-left,\n  \"backspace\": $fa-var-backspace,\n  \"eye-dropper\": $fa-var-eye-dropper,\n  \"eye-dropper-empty\": $fa-var-eye-dropper-empty,\n  \"eyedropper\": $fa-var-eyedropper,\n  \"file-circle-check\": $fa-var-file-circle-check,\n  \"forward\": $fa-var-forward,\n  \"mobile\": $fa-var-mobile,\n  \"mobile-android\": $fa-var-mobile-android,\n  \"mobile-phone\": $fa-var-mobile-phone,\n  \"face-meh\": $fa-var-face-meh,\n  \"meh\": $fa-var-meh,\n  \"align-center\": $fa-var-align-center,\n  \"book-skull\": $fa-var-book-skull,\n  \"book-dead\": $fa-var-book-dead,\n  \"id-card\": $fa-var-id-card,\n  \"drivers-license\": $fa-var-drivers-license,\n  \"outdent\": $fa-var-outdent,\n  \"dedent\": $fa-var-dedent,\n  \"heart-circle-exclamation\": $fa-var-heart-circle-exclamation,\n  \"house\": $fa-var-house,\n  \"home\": $fa-var-home,\n  \"home-alt\": $fa-var-home-alt,\n  \"home-lg-alt\": $fa-var-home-lg-alt,\n  \"calendar-week\": $fa-var-calendar-week,\n  \"laptop-medical\": $fa-var-laptop-medical,\n  \"b\": $fa-var-b,\n  \"file-medical\": $fa-var-file-medical,\n  \"dice-one\": $fa-var-dice-one,\n  \"kiwi-bird\": $fa-var-kiwi-bird,\n  \"arrow-right-arrow-left\": $fa-var-arrow-right-arrow-left,\n  \"exchange\": $fa-var-exchange,\n  \"rotate-right\": $fa-var-rotate-right,\n  \"redo-alt\": $fa-var-redo-alt,\n  \"rotate-forward\": $fa-var-rotate-forward,\n  \"utensils\": $fa-var-utensils,\n  \"cutlery\": $fa-var-cutlery,\n  \"arrow-up-wide-short\": $fa-var-arrow-up-wide-short,\n  \"sort-amount-up\": $fa-var-sort-amount-up,\n  \"mill-sign\": $fa-var-mill-sign,\n  \"bowl-rice\": $fa-var-bowl-rice,\n  \"skull\": $fa-var-skull,\n  \"tower-broadcast\": $fa-var-tower-broadcast,\n  \"broadcast-tower\": $fa-var-broadcast-tower,\n  \"truck-pickup\": $fa-var-truck-pickup,\n  \"up-long\": $fa-var-up-long,\n  \"long-arrow-alt-up\": $fa-var-long-arrow-alt-up,\n  \"stop\": $fa-var-stop,\n  \"code-merge\": $fa-var-code-merge,\n  \"upload\": $fa-var-upload,\n  \"hurricane\": $fa-var-hurricane,\n  \"mound\": $fa-var-mound,\n  \"toilet-portable\": $fa-var-toilet-portable,\n  \"compact-disc\": $fa-var-compact-disc,\n  \"file-arrow-down\": $fa-var-file-arrow-down,\n  \"file-download\": $fa-var-file-download,\n  \"caravan\": $fa-var-caravan,\n  \"shield-cat\": $fa-var-shield-cat,\n  \"bolt\": $fa-var-bolt,\n  \"zap\": $fa-var-zap,\n  \"glass-water\": $fa-var-glass-water,\n  \"oil-well\": $fa-var-oil-well,\n  \"vault\": $fa-var-vault,\n  \"mars\": $fa-var-mars,\n  \"toilet\": $fa-var-toilet,\n  \"plane-circle-xmark\": $fa-var-plane-circle-xmark,\n  \"yen-sign\": $fa-var-yen-sign,\n  \"cny\": $fa-var-cny,\n  \"jpy\": $fa-var-jpy,\n  \"rmb\": $fa-var-rmb,\n  \"yen\": $fa-var-yen,\n  \"ruble-sign\": $fa-var-ruble-sign,\n  \"rouble\": $fa-var-rouble,\n  \"rub\": $fa-var-rub,\n  \"ruble\": $fa-var-ruble,\n  \"sun\": $fa-var-sun,\n  \"guitar\": $fa-var-guitar,\n  \"face-laugh-wink\": $fa-var-face-laugh-wink,\n  \"laugh-wink\": $fa-var-laugh-wink,\n  \"horse-head\": $fa-var-horse-head,\n  \"bore-hole\": $fa-var-bore-hole,\n  \"industry\": $fa-var-industry,\n  \"circle-down\": $fa-var-circle-down,\n  \"arrow-alt-circle-down\": $fa-var-arrow-alt-circle-down,\n  \"arrows-turn-to-dots\": $fa-var-arrows-turn-to-dots,\n  \"florin-sign\": $fa-var-florin-sign,\n  \"arrow-down-short-wide\": $fa-var-arrow-down-short-wide,\n  \"sort-amount-desc\": $fa-var-sort-amount-desc,\n  \"sort-amount-down-alt\": $fa-var-sort-amount-down-alt,\n  \"less-than\": $fa-var-less-than,\n  \"angle-down\": $fa-var-angle-down,\n  \"car-tunnel\": $fa-var-car-tunnel,\n  \"head-side-cough\": $fa-var-head-side-cough,\n  \"grip-lines\": $fa-var-grip-lines,\n  \"thumbs-down\": $fa-var-thumbs-down,\n  \"user-lock\": $fa-var-user-lock,\n  \"arrow-right-long\": $fa-var-arrow-right-long,\n  \"long-arrow-right\": $fa-var-long-arrow-right,\n  \"anchor-circle-xmark\": $fa-var-anchor-circle-xmark,\n  \"ellipsis\": $fa-var-ellipsis,\n  \"ellipsis-h\": $fa-var-ellipsis-h,\n  \"chess-pawn\": $fa-var-chess-pawn,\n  \"kit-medical\": $fa-var-kit-medical,\n  \"first-aid\": $fa-var-first-aid,\n  \"person-through-window\": $fa-var-person-through-window,\n  \"toolbox\": $fa-var-toolbox,\n  \"hands-holding-circle\": $fa-var-hands-holding-circle,\n  \"bug\": $fa-var-bug,\n  \"credit-card\": $fa-var-credit-card,\n  \"credit-card-alt\": $fa-var-credit-card-alt,\n  \"car\": $fa-var-car,\n  \"automobile\": $fa-var-automobile,\n  \"hand-holding-hand\": $fa-var-hand-holding-hand,\n  \"book-open-reader\": $fa-var-book-open-reader,\n  \"book-reader\": $fa-var-book-reader,\n  \"mountain-sun\": $fa-var-mountain-sun,\n  \"arrows-left-right-to-line\": $fa-var-arrows-left-right-to-line,\n  \"dice-d20\": $fa-var-dice-d20,\n  \"truck-droplet\": $fa-var-truck-droplet,\n  \"file-circle-xmark\": $fa-var-file-circle-xmark,\n  \"temperature-arrow-up\": $fa-var-temperature-arrow-up,\n  \"temperature-up\": $fa-var-temperature-up,\n  \"medal\": $fa-var-medal,\n  \"bed\": $fa-var-bed,\n  \"square-h\": $fa-var-square-h,\n  \"h-square\": $fa-var-h-square,\n  \"podcast\": $fa-var-podcast,\n  \"temperature-full\": $fa-var-temperature-full,\n  \"temperature-4\": $fa-var-temperature-4,\n  \"thermometer-4\": $fa-var-thermometer-4,\n  \"thermometer-full\": $fa-var-thermometer-full,\n  \"bell\": $fa-var-bell,\n  \"superscript\": $fa-var-superscript,\n  \"plug-circle-xmark\": $fa-var-plug-circle-xmark,\n  \"star-of-life\": $fa-var-star-of-life,\n  \"phone-slash\": $fa-var-phone-slash,\n  \"paint-roller\": $fa-var-paint-roller,\n  \"handshake-angle\": $fa-var-handshake-angle,\n  \"hands-helping\": $fa-var-hands-helping,\n  \"location-dot\": $fa-var-location-dot,\n  \"map-marker-alt\": $fa-var-map-marker-alt,\n  \"file\": $fa-var-file,\n  \"greater-than\": $fa-var-greater-than,\n  \"person-swimming\": $fa-var-person-swimming,\n  \"swimmer\": $fa-var-swimmer,\n  \"arrow-down\": $fa-var-arrow-down,\n  \"droplet\": $fa-var-droplet,\n  \"tint\": $fa-var-tint,\n  \"eraser\": $fa-var-eraser,\n  \"earth-americas\": $fa-var-earth-americas,\n  \"earth\": $fa-var-earth,\n  \"earth-america\": $fa-var-earth-america,\n  \"globe-americas\": $fa-var-globe-americas,\n  \"person-burst\": $fa-var-person-burst,\n  \"dove\": $fa-var-dove,\n  \"battery-empty\": $fa-var-battery-empty,\n  \"battery-0\": $fa-var-battery-0,\n  \"socks\": $fa-var-socks,\n  \"inbox\": $fa-var-inbox,\n  \"section\": $fa-var-section,\n  \"gauge-high\": $fa-var-gauge-high,\n  \"tachometer-alt\": $fa-var-tachometer-alt,\n  \"tachometer-alt-fast\": $fa-var-tachometer-alt-fast,\n  \"envelope-open-text\": $fa-var-envelope-open-text,\n  \"hospital\": $fa-var-hospital,\n  \"hospital-alt\": $fa-var-hospital-alt,\n  \"hospital-wide\": $fa-var-hospital-wide,\n  \"wine-bottle\": $fa-var-wine-bottle,\n  \"chess-rook\": $fa-var-chess-rook,\n  \"bars-staggered\": $fa-var-bars-staggered,\n  \"reorder\": $fa-var-reorder,\n  \"stream\": $fa-var-stream,\n  \"dharmachakra\": $fa-var-dharmachakra,\n  \"hotdog\": $fa-var-hotdog,\n  \"person-walking-with-cane\": $fa-var-person-walking-with-cane,\n  \"blind\": $fa-var-blind,\n  \"drum\": $fa-var-drum,\n  \"ice-cream\": $fa-var-ice-cream,\n  \"heart-circle-bolt\": $fa-var-heart-circle-bolt,\n  \"fax\": $fa-var-fax,\n  \"paragraph\": $fa-var-paragraph,\n  \"check-to-slot\": $fa-var-check-to-slot,\n  \"vote-yea\": $fa-var-vote-yea,\n  \"star-half\": $fa-var-star-half,\n  \"boxes-stacked\": $fa-var-boxes-stacked,\n  \"boxes\": $fa-var-boxes,\n  \"boxes-alt\": $fa-var-boxes-alt,\n  \"link\": $fa-var-link,\n  \"chain\": $fa-var-chain,\n  \"ear-listen\": $fa-var-ear-listen,\n  \"assistive-listening-systems\": $fa-var-assistive-listening-systems,\n  \"tree-city\": $fa-var-tree-city,\n  \"play\": $fa-var-play,\n  \"font\": $fa-var-font,\n  \"table-cells-row-lock\": $fa-var-table-cells-row-lock,\n  \"rupiah-sign\": $fa-var-rupiah-sign,\n  \"magnifying-glass\": $fa-var-magnifying-glass,\n  \"search\": $fa-var-search,\n  \"table-tennis-paddle-ball\": $fa-var-table-tennis-paddle-ball,\n  \"ping-pong-paddle-ball\": $fa-var-ping-pong-paddle-ball,\n  \"table-tennis\": $fa-var-table-tennis,\n  \"person-dots-from-line\": $fa-var-person-dots-from-line,\n  \"diagnoses\": $fa-var-diagnoses,\n  \"trash-can-arrow-up\": $fa-var-trash-can-arrow-up,\n  \"trash-restore-alt\": $fa-var-trash-restore-alt,\n  \"naira-sign\": $fa-var-naira-sign,\n  \"cart-arrow-down\": $fa-var-cart-arrow-down,\n  \"walkie-talkie\": $fa-var-walkie-talkie,\n  \"file-pen\": $fa-var-file-pen,\n  \"file-edit\": $fa-var-file-edit,\n  \"receipt\": $fa-var-receipt,\n  \"square-pen\": $fa-var-square-pen,\n  \"pen-square\": $fa-var-pen-square,\n  \"pencil-square\": $fa-var-pencil-square,\n  \"suitcase-rolling\": $fa-var-suitcase-rolling,\n  \"person-circle-exclamation\": $fa-var-person-circle-exclamation,\n  \"chevron-down\": $fa-var-chevron-down,\n  \"battery-full\": $fa-var-battery-full,\n  \"battery\": $fa-var-battery,\n  \"battery-5\": $fa-var-battery-5,\n  \"skull-crossbones\": $fa-var-skull-crossbones,\n  \"code-compare\": $fa-var-code-compare,\n  \"list-ul\": $fa-var-list-ul,\n  \"list-dots\": $fa-var-list-dots,\n  \"school-lock\": $fa-var-school-lock,\n  \"tower-cell\": $fa-var-tower-cell,\n  \"down-long\": $fa-var-down-long,\n  \"long-arrow-alt-down\": $fa-var-long-arrow-alt-down,\n  \"ranking-star\": $fa-var-ranking-star,\n  \"chess-king\": $fa-var-chess-king,\n  \"person-harassing\": $fa-var-person-harassing,\n  \"brazilian-real-sign\": $fa-var-brazilian-real-sign,\n  \"landmark-dome\": $fa-var-landmark-dome,\n  \"landmark-alt\": $fa-var-landmark-alt,\n  \"arrow-up\": $fa-var-arrow-up,\n  \"tv\": $fa-var-tv,\n  \"television\": $fa-var-television,\n  \"tv-alt\": $fa-var-tv-alt,\n  \"shrimp\": $fa-var-shrimp,\n  \"list-check\": $fa-var-list-check,\n  \"tasks\": $fa-var-tasks,\n  \"jug-detergent\": $fa-var-jug-detergent,\n  \"circle-user\": $fa-var-circle-user,\n  \"user-circle\": $fa-var-user-circle,\n  \"user-shield\": $fa-var-user-shield,\n  \"wind\": $fa-var-wind,\n  \"car-burst\": $fa-var-car-burst,\n  \"car-crash\": $fa-var-car-crash,\n  \"y\": $fa-var-y,\n  \"person-snowboarding\": $fa-var-person-snowboarding,\n  \"snowboarding\": $fa-var-snowboarding,\n  \"truck-fast\": $fa-var-truck-fast,\n  \"shipping-fast\": $fa-var-shipping-fast,\n  \"fish\": $fa-var-fish,\n  \"user-graduate\": $fa-var-user-graduate,\n  \"circle-half-stroke\": $fa-var-circle-half-stroke,\n  \"adjust\": $fa-var-adjust,\n  \"clapperboard\": $fa-var-clapperboard,\n  \"circle-radiation\": $fa-var-circle-radiation,\n  \"radiation-alt\": $fa-var-radiation-alt,\n  \"baseball\": $fa-var-baseball,\n  \"baseball-ball\": $fa-var-baseball-ball,\n  \"jet-fighter-up\": $fa-var-jet-fighter-up,\n  \"diagram-project\": $fa-var-diagram-project,\n  \"project-diagram\": $fa-var-project-diagram,\n  \"copy\": $fa-var-copy,\n  \"volume-xmark\": $fa-var-volume-xmark,\n  \"volume-mute\": $fa-var-volume-mute,\n  \"volume-times\": $fa-var-volume-times,\n  \"hand-sparkles\": $fa-var-hand-sparkles,\n  \"grip\": $fa-var-grip,\n  \"grip-horizontal\": $fa-var-grip-horizontal,\n  \"share-from-square\": $fa-var-share-from-square,\n  \"share-square\": $fa-var-share-square,\n  \"child-combatant\": $fa-var-child-combatant,\n  \"child-rifle\": $fa-var-child-rifle,\n  \"gun\": $fa-var-gun,\n  \"square-phone\": $fa-var-square-phone,\n  \"phone-square\": $fa-var-phone-square,\n  \"plus\": $fa-var-plus,\n  \"add\": $fa-var-add,\n  \"expand\": $fa-var-expand,\n  \"computer\": $fa-var-computer,\n  \"xmark\": $fa-var-xmark,\n  \"close\": $fa-var-close,\n  \"multiply\": $fa-var-multiply,\n  \"remove\": $fa-var-remove,\n  \"times\": $fa-var-times,\n  \"arrows-up-down-left-right\": $fa-var-arrows-up-down-left-right,\n  \"arrows\": $fa-var-arrows,\n  \"chalkboard-user\": $fa-var-chalkboard-user,\n  \"chalkboard-teacher\": $fa-var-chalkboard-teacher,\n  \"peso-sign\": $fa-var-peso-sign,\n  \"building-shield\": $fa-var-building-shield,\n  \"baby\": $fa-var-baby,\n  \"users-line\": $fa-var-users-line,\n  \"quote-left\": $fa-var-quote-left,\n  \"quote-left-alt\": $fa-var-quote-left-alt,\n  \"tractor\": $fa-var-tractor,\n  \"trash-arrow-up\": $fa-var-trash-arrow-up,\n  \"trash-restore\": $fa-var-trash-restore,\n  \"arrow-down-up-lock\": $fa-var-arrow-down-up-lock,\n  \"lines-leaning\": $fa-var-lines-leaning,\n  \"ruler-combined\": $fa-var-ruler-combined,\n  \"copyright\": $fa-var-copyright,\n  \"equals\": $fa-var-equals,\n  \"blender\": $fa-var-blender,\n  \"teeth\": $fa-var-teeth,\n  \"shekel-sign\": $fa-var-shekel-sign,\n  \"ils\": $fa-var-ils,\n  \"shekel\": $fa-var-shekel,\n  \"sheqel\": $fa-var-sheqel,\n  \"sheqel-sign\": $fa-var-sheqel-sign,\n  \"map\": $fa-var-map,\n  \"rocket\": $fa-var-rocket,\n  \"photo-film\": $fa-var-photo-film,\n  \"photo-video\": $fa-var-photo-video,\n  \"folder-minus\": $fa-var-folder-minus,\n  \"store\": $fa-var-store,\n  \"arrow-trend-up\": $fa-var-arrow-trend-up,\n  \"plug-circle-minus\": $fa-var-plug-circle-minus,\n  \"sign-hanging\": $fa-var-sign-hanging,\n  \"sign\": $fa-var-sign,\n  \"bezier-curve\": $fa-var-bezier-curve,\n  \"bell-slash\": $fa-var-bell-slash,\n  \"tablet\": $fa-var-tablet,\n  \"tablet-android\": $fa-var-tablet-android,\n  \"school-flag\": $fa-var-school-flag,\n  \"fill\": $fa-var-fill,\n  \"angle-up\": $fa-var-angle-up,\n  \"drumstick-bite\": $fa-var-drumstick-bite,\n  \"holly-berry\": $fa-var-holly-berry,\n  \"chevron-left\": $fa-var-chevron-left,\n  \"bacteria\": $fa-var-bacteria,\n  \"hand-lizard\": $fa-var-hand-lizard,\n  \"notdef\": $fa-var-notdef,\n  \"disease\": $fa-var-disease,\n  \"briefcase-medical\": $fa-var-briefcase-medical,\n  \"genderless\": $fa-var-genderless,\n  \"chevron-right\": $fa-var-chevron-right,\n  \"retweet\": $fa-var-retweet,\n  \"car-rear\": $fa-var-car-rear,\n  \"car-alt\": $fa-var-car-alt,\n  \"pump-soap\": $fa-var-pump-soap,\n  \"video-slash\": $fa-var-video-slash,\n  \"battery-quarter\": $fa-var-battery-quarter,\n  \"battery-2\": $fa-var-battery-2,\n  \"radio\": $fa-var-radio,\n  \"baby-carriage\": $fa-var-baby-carriage,\n  \"carriage-baby\": $fa-var-carriage-baby,\n  \"traffic-light\": $fa-var-traffic-light,\n  \"thermometer\": $fa-var-thermometer,\n  \"vr-cardboard\": $fa-var-vr-cardboard,\n  \"hand-middle-finger\": $fa-var-hand-middle-finger,\n  \"percent\": $fa-var-percent,\n  \"percentage\": $fa-var-percentage,\n  \"truck-moving\": $fa-var-truck-moving,\n  \"glass-water-droplet\": $fa-var-glass-water-droplet,\n  \"display\": $fa-var-display,\n  \"face-smile\": $fa-var-face-smile,\n  \"smile\": $fa-var-smile,\n  \"thumbtack\": $fa-var-thumbtack,\n  \"thumb-tack\": $fa-var-thumb-tack,\n  \"trophy\": $fa-var-trophy,\n  \"person-praying\": $fa-var-person-praying,\n  \"pray\": $fa-var-pray,\n  \"hammer\": $fa-var-hammer,\n  \"hand-peace\": $fa-var-hand-peace,\n  \"rotate\": $fa-var-rotate,\n  \"sync-alt\": $fa-var-sync-alt,\n  \"spinner\": $fa-var-spinner,\n  \"robot\": $fa-var-robot,\n  \"peace\": $fa-var-peace,\n  \"gears\": $fa-var-gears,\n  \"cogs\": $fa-var-cogs,\n  \"warehouse\": $fa-var-warehouse,\n  \"arrow-up-right-dots\": $fa-var-arrow-up-right-dots,\n  \"splotch\": $fa-var-splotch,\n  \"face-grin-hearts\": $fa-var-face-grin-hearts,\n  \"grin-hearts\": $fa-var-grin-hearts,\n  \"dice-four\": $fa-var-dice-four,\n  \"sim-card\": $fa-var-sim-card,\n  \"transgender\": $fa-var-transgender,\n  \"transgender-alt\": $fa-var-transgender-alt,\n  \"mercury\": $fa-var-mercury,\n  \"arrow-turn-down\": $fa-var-arrow-turn-down,\n  \"level-down\": $fa-var-level-down,\n  \"person-falling-burst\": $fa-var-person-falling-burst,\n  \"award\": $fa-var-award,\n  \"ticket-simple\": $fa-var-ticket-simple,\n  \"ticket-alt\": $fa-var-ticket-alt,\n  \"building\": $fa-var-building,\n  \"angles-left\": $fa-var-angles-left,\n  \"angle-double-left\": $fa-var-angle-double-left,\n  \"qrcode\": $fa-var-qrcode,\n  \"clock-rotate-left\": $fa-var-clock-rotate-left,\n  \"history\": $fa-var-history,\n  \"face-grin-beam-sweat\": $fa-var-face-grin-beam-sweat,\n  \"grin-beam-sweat\": $fa-var-grin-beam-sweat,\n  \"file-export\": $fa-var-file-export,\n  \"arrow-right-from-file\": $fa-var-arrow-right-from-file,\n  \"shield\": $fa-var-shield,\n  \"shield-blank\": $fa-var-shield-blank,\n  \"arrow-up-short-wide\": $fa-var-arrow-up-short-wide,\n  \"sort-amount-up-alt\": $fa-var-sort-amount-up-alt,\n  \"house-medical\": $fa-var-house-medical,\n  \"golf-ball-tee\": $fa-var-golf-ball-tee,\n  \"golf-ball\": $fa-var-golf-ball,\n  \"circle-chevron-left\": $fa-var-circle-chevron-left,\n  \"chevron-circle-left\": $fa-var-chevron-circle-left,\n  \"house-chimney-window\": $fa-var-house-chimney-window,\n  \"pen-nib\": $fa-var-pen-nib,\n  \"tent-arrow-turn-left\": $fa-var-tent-arrow-turn-left,\n  \"tents\": $fa-var-tents,\n  \"wand-magic\": $fa-var-wand-magic,\n  \"magic\": $fa-var-magic,\n  \"dog\": $fa-var-dog,\n  \"carrot\": $fa-var-carrot,\n  \"moon\": $fa-var-moon,\n  \"wine-glass-empty\": $fa-var-wine-glass-empty,\n  \"wine-glass-alt\": $fa-var-wine-glass-alt,\n  \"cheese\": $fa-var-cheese,\n  \"yin-yang\": $fa-var-yin-yang,\n  \"music\": $fa-var-music,\n  \"code-commit\": $fa-var-code-commit,\n  \"temperature-low\": $fa-var-temperature-low,\n  \"person-biking\": $fa-var-person-biking,\n  \"biking\": $fa-var-biking,\n  \"broom\": $fa-var-broom,\n  \"shield-heart\": $fa-var-shield-heart,\n  \"gopuram\": $fa-var-gopuram,\n  \"earth-oceania\": $fa-var-earth-oceania,\n  \"globe-oceania\": $fa-var-globe-oceania,\n  \"square-xmark\": $fa-var-square-xmark,\n  \"times-square\": $fa-var-times-square,\n  \"xmark-square\": $fa-var-xmark-square,\n  \"hashtag\": $fa-var-hashtag,\n  \"up-right-and-down-left-from-center\": $fa-var-up-right-and-down-left-from-center,\n  \"expand-alt\": $fa-var-expand-alt,\n  \"oil-can\": $fa-var-oil-can,\n  \"t\": $fa-var-t,\n  \"hippo\": $fa-var-hippo,\n  \"chart-column\": $fa-var-chart-column,\n  \"infinity\": $fa-var-infinity,\n  \"vial-circle-check\": $fa-var-vial-circle-check,\n  \"person-arrow-down-to-line\": $fa-var-person-arrow-down-to-line,\n  \"voicemail\": $fa-var-voicemail,\n  \"fan\": $fa-var-fan,\n  \"person-walking-luggage\": $fa-var-person-walking-luggage,\n  \"up-down\": $fa-var-up-down,\n  \"arrows-alt-v\": $fa-var-arrows-alt-v,\n  \"cloud-moon-rain\": $fa-var-cloud-moon-rain,\n  \"calendar\": $fa-var-calendar,\n  \"trailer\": $fa-var-trailer,\n  \"bahai\": $fa-var-bahai,\n  \"haykal\": $fa-var-haykal,\n  \"sd-card\": $fa-var-sd-card,\n  \"dragon\": $fa-var-dragon,\n  \"shoe-prints\": $fa-var-shoe-prints,\n  \"circle-plus\": $fa-var-circle-plus,\n  \"plus-circle\": $fa-var-plus-circle,\n  \"face-grin-tongue-wink\": $fa-var-face-grin-tongue-wink,\n  \"grin-tongue-wink\": $fa-var-grin-tongue-wink,\n  \"hand-holding\": $fa-var-hand-holding,\n  \"plug-circle-exclamation\": $fa-var-plug-circle-exclamation,\n  \"link-slash\": $fa-var-link-slash,\n  \"chain-broken\": $fa-var-chain-broken,\n  \"chain-slash\": $fa-var-chain-slash,\n  \"unlink\": $fa-var-unlink,\n  \"clone\": $fa-var-clone,\n  \"person-walking-arrow-loop-left\": $fa-var-person-walking-arrow-loop-left,\n  \"arrow-up-z-a\": $fa-var-arrow-up-z-a,\n  \"sort-alpha-up-alt\": $fa-var-sort-alpha-up-alt,\n  \"fire-flame-curved\": $fa-var-fire-flame-curved,\n  \"fire-alt\": $fa-var-fire-alt,\n  \"tornado\": $fa-var-tornado,\n  \"file-circle-plus\": $fa-var-file-circle-plus,\n  \"book-quran\": $fa-var-book-quran,\n  \"quran\": $fa-var-quran,\n  \"anchor\": $fa-var-anchor,\n  \"border-all\": $fa-var-border-all,\n  \"face-angry\": $fa-var-face-angry,\n  \"angry\": $fa-var-angry,\n  \"cookie-bite\": $fa-var-cookie-bite,\n  \"arrow-trend-down\": $fa-var-arrow-trend-down,\n  \"rss\": $fa-var-rss,\n  \"feed\": $fa-var-feed,\n  \"draw-polygon\": $fa-var-draw-polygon,\n  \"scale-balanced\": $fa-var-scale-balanced,\n  \"balance-scale\": $fa-var-balance-scale,\n  \"gauge-simple-high\": $fa-var-gauge-simple-high,\n  \"tachometer\": $fa-var-tachometer,\n  \"tachometer-fast\": $fa-var-tachometer-fast,\n  \"shower\": $fa-var-shower,\n  \"desktop\": $fa-var-desktop,\n  \"desktop-alt\": $fa-var-desktop-alt,\n  \"m\": $fa-var-m,\n  \"table-list\": $fa-var-table-list,\n  \"th-list\": $fa-var-th-list,\n  \"comment-sms\": $fa-var-comment-sms,\n  \"sms\": $fa-var-sms,\n  \"book\": $fa-var-book,\n  \"user-plus\": $fa-var-user-plus,\n  \"check\": $fa-var-check,\n  \"battery-three-quarters\": $fa-var-battery-three-quarters,\n  \"battery-4\": $fa-var-battery-4,\n  \"house-circle-check\": $fa-var-house-circle-check,\n  \"angle-left\": $fa-var-angle-left,\n  \"diagram-successor\": $fa-var-diagram-successor,\n  \"truck-arrow-right\": $fa-var-truck-arrow-right,\n  \"arrows-split-up-and-left\": $fa-var-arrows-split-up-and-left,\n  \"hand-fist\": $fa-var-hand-fist,\n  \"fist-raised\": $fa-var-fist-raised,\n  \"cloud-moon\": $fa-var-cloud-moon,\n  \"briefcase\": $fa-var-briefcase,\n  \"person-falling\": $fa-var-person-falling,\n  \"image-portrait\": $fa-var-image-portrait,\n  \"portrait\": $fa-var-portrait,\n  \"user-tag\": $fa-var-user-tag,\n  \"rug\": $fa-var-rug,\n  \"earth-europe\": $fa-var-earth-europe,\n  \"globe-europe\": $fa-var-globe-europe,\n  \"cart-flatbed-suitcase\": $fa-var-cart-flatbed-suitcase,\n  \"luggage-cart\": $fa-var-luggage-cart,\n  \"rectangle-xmark\": $fa-var-rectangle-xmark,\n  \"rectangle-times\": $fa-var-rectangle-times,\n  \"times-rectangle\": $fa-var-times-rectangle,\n  \"window-close\": $fa-var-window-close,\n  \"baht-sign\": $fa-var-baht-sign,\n  \"book-open\": $fa-var-book-open,\n  \"book-journal-whills\": $fa-var-book-journal-whills,\n  \"journal-whills\": $fa-var-journal-whills,\n  \"handcuffs\": $fa-var-handcuffs,\n  \"triangle-exclamation\": $fa-var-triangle-exclamation,\n  \"exclamation-triangle\": $fa-var-exclamation-triangle,\n  \"warning\": $fa-var-warning,\n  \"database\": $fa-var-database,\n  \"share\": $fa-var-share,\n  \"mail-forward\": $fa-var-mail-forward,\n  \"bottle-droplet\": $fa-var-bottle-droplet,\n  \"mask-face\": $fa-var-mask-face,\n  \"hill-rockslide\": $fa-var-hill-rockslide,\n  \"right-left\": $fa-var-right-left,\n  \"exchange-alt\": $fa-var-exchange-alt,\n  \"paper-plane\": $fa-var-paper-plane,\n  \"road-circle-exclamation\": $fa-var-road-circle-exclamation,\n  \"dungeon\": $fa-var-dungeon,\n  \"align-right\": $fa-var-align-right,\n  \"money-bill-1-wave\": $fa-var-money-bill-1-wave,\n  \"money-bill-wave-alt\": $fa-var-money-bill-wave-alt,\n  \"life-ring\": $fa-var-life-ring,\n  \"hands\": $fa-var-hands,\n  \"sign-language\": $fa-var-sign-language,\n  \"signing\": $fa-var-signing,\n  \"calendar-day\": $fa-var-calendar-day,\n  \"water-ladder\": $fa-var-water-ladder,\n  \"ladder-water\": $fa-var-ladder-water,\n  \"swimming-pool\": $fa-var-swimming-pool,\n  \"arrows-up-down\": $fa-var-arrows-up-down,\n  \"arrows-v\": $fa-var-arrows-v,\n  \"face-grimace\": $fa-var-face-grimace,\n  \"grimace\": $fa-var-grimace,\n  \"wheelchair-move\": $fa-var-wheelchair-move,\n  \"wheelchair-alt\": $fa-var-wheelchair-alt,\n  \"turn-down\": $fa-var-turn-down,\n  \"level-down-alt\": $fa-var-level-down-alt,\n  \"person-walking-arrow-right\": $fa-var-person-walking-arrow-right,\n  \"square-envelope\": $fa-var-square-envelope,\n  \"envelope-square\": $fa-var-envelope-square,\n  \"dice\": $fa-var-dice,\n  \"bowling-ball\": $fa-var-bowling-ball,\n  \"brain\": $fa-var-brain,\n  \"bandage\": $fa-var-bandage,\n  \"band-aid\": $fa-var-band-aid,\n  \"calendar-minus\": $fa-var-calendar-minus,\n  \"circle-xmark\": $fa-var-circle-xmark,\n  \"times-circle\": $fa-var-times-circle,\n  \"xmark-circle\": $fa-var-xmark-circle,\n  \"gifts\": $fa-var-gifts,\n  \"hotel\": $fa-var-hotel,\n  \"earth-asia\": $fa-var-earth-asia,\n  \"globe-asia\": $fa-var-globe-asia,\n  \"id-card-clip\": $fa-var-id-card-clip,\n  \"id-card-alt\": $fa-var-id-card-alt,\n  \"magnifying-glass-plus\": $fa-var-magnifying-glass-plus,\n  \"search-plus\": $fa-var-search-plus,\n  \"thumbs-up\": $fa-var-thumbs-up,\n  \"user-clock\": $fa-var-user-clock,\n  \"hand-dots\": $fa-var-hand-dots,\n  \"allergies\": $fa-var-allergies,\n  \"file-invoice\": $fa-var-file-invoice,\n  \"window-minimize\": $fa-var-window-minimize,\n  \"mug-saucer\": $fa-var-mug-saucer,\n  \"coffee\": $fa-var-coffee,\n  \"brush\": $fa-var-brush,\n  \"mask\": $fa-var-mask,\n  \"magnifying-glass-minus\": $fa-var-magnifying-glass-minus,\n  \"search-minus\": $fa-var-search-minus,\n  \"ruler-vertical\": $fa-var-ruler-vertical,\n  \"user-large\": $fa-var-user-large,\n  \"user-alt\": $fa-var-user-alt,\n  \"train-tram\": $fa-var-train-tram,\n  \"user-nurse\": $fa-var-user-nurse,\n  \"syringe\": $fa-var-syringe,\n  \"cloud-sun\": $fa-var-cloud-sun,\n  \"stopwatch-20\": $fa-var-stopwatch-20,\n  \"square-full\": $fa-var-square-full,\n  \"magnet\": $fa-var-magnet,\n  \"jar\": $fa-var-jar,\n  \"note-sticky\": $fa-var-note-sticky,\n  \"sticky-note\": $fa-var-sticky-note,\n  \"bug-slash\": $fa-var-bug-slash,\n  \"arrow-up-from-water-pump\": $fa-var-arrow-up-from-water-pump,\n  \"bone\": $fa-var-bone,\n  \"user-injured\": $fa-var-user-injured,\n  \"face-sad-tear\": $fa-var-face-sad-tear,\n  \"sad-tear\": $fa-var-sad-tear,\n  \"plane\": $fa-var-plane,\n  \"tent-arrows-down\": $fa-var-tent-arrows-down,\n  \"exclamation\": $fa-var-exclamation,\n  \"arrows-spin\": $fa-var-arrows-spin,\n  \"print\": $fa-var-print,\n  \"turkish-lira-sign\": $fa-var-turkish-lira-sign,\n  \"try\": $fa-var-try,\n  \"turkish-lira\": $fa-var-turkish-lira,\n  \"dollar-sign\": $fa-var-dollar-sign,\n  \"dollar\": $fa-var-dollar,\n  \"usd\": $fa-var-usd,\n  \"x\": $fa-var-x,\n  \"magnifying-glass-dollar\": $fa-var-magnifying-glass-dollar,\n  \"search-dollar\": $fa-var-search-dollar,\n  \"users-gear\": $fa-var-users-gear,\n  \"users-cog\": $fa-var-users-cog,\n  \"person-military-pointing\": $fa-var-person-military-pointing,\n  \"building-columns\": $fa-var-building-columns,\n  \"bank\": $fa-var-bank,\n  \"institution\": $fa-var-institution,\n  \"museum\": $fa-var-museum,\n  \"university\": $fa-var-university,\n  \"umbrella\": $fa-var-umbrella,\n  \"trowel\": $fa-var-trowel,\n  \"d\": $fa-var-d,\n  \"stapler\": $fa-var-stapler,\n  \"masks-theater\": $fa-var-masks-theater,\n  \"theater-masks\": $fa-var-theater-masks,\n  \"kip-sign\": $fa-var-kip-sign,\n  \"hand-point-left\": $fa-var-hand-point-left,\n  \"handshake-simple\": $fa-var-handshake-simple,\n  \"handshake-alt\": $fa-var-handshake-alt,\n  \"jet-fighter\": $fa-var-jet-fighter,\n  \"fighter-jet\": $fa-var-fighter-jet,\n  \"square-share-nodes\": $fa-var-square-share-nodes,\n  \"share-alt-square\": $fa-var-share-alt-square,\n  \"barcode\": $fa-var-barcode,\n  \"plus-minus\": $fa-var-plus-minus,\n  \"video\": $fa-var-video,\n  \"video-camera\": $fa-var-video-camera,\n  \"graduation-cap\": $fa-var-graduation-cap,\n  \"mortar-board\": $fa-var-mortar-board,\n  \"hand-holding-medical\": $fa-var-hand-holding-medical,\n  \"person-circle-check\": $fa-var-person-circle-check,\n  \"turn-up\": $fa-var-turn-up,\n  \"level-up-alt\": $fa-var-level-up-alt,\n);\n\n$fa-brand-icons: (\n  \"monero\": $fa-var-monero,\n  \"hooli\": $fa-var-hooli,\n  \"yelp\": $fa-var-yelp,\n  \"cc-visa\": $fa-var-cc-visa,\n  \"lastfm\": $fa-var-lastfm,\n  \"shopware\": $fa-var-shopware,\n  \"creative-commons-nc\": $fa-var-creative-commons-nc,\n  \"aws\": $fa-var-aws,\n  \"redhat\": $fa-var-redhat,\n  \"yoast\": $fa-var-yoast,\n  \"cloudflare\": $fa-var-cloudflare,\n  \"ups\": $fa-var-ups,\n  \"pixiv\": $fa-var-pixiv,\n  \"wpexplorer\": $fa-var-wpexplorer,\n  \"dyalog\": $fa-var-dyalog,\n  \"bity\": $fa-var-bity,\n  \"stackpath\": $fa-var-stackpath,\n  \"buysellads\": $fa-var-buysellads,\n  \"first-order\": $fa-var-first-order,\n  \"modx\": $fa-var-modx,\n  \"guilded\": $fa-var-guilded,\n  \"vnv\": $fa-var-vnv,\n  \"square-js\": $fa-var-square-js,\n  \"js-square\": $fa-var-js-square,\n  \"microsoft\": $fa-var-microsoft,\n  \"qq\": $fa-var-qq,\n  \"orcid\": $fa-var-orcid,\n  \"java\": $fa-var-java,\n  \"invision\": $fa-var-invision,\n  \"creative-commons-pd-alt\": $fa-var-creative-commons-pd-alt,\n  \"centercode\": $fa-var-centercode,\n  \"glide-g\": $fa-var-glide-g,\n  \"drupal\": $fa-var-drupal,\n  \"jxl\": $fa-var-jxl,\n  \"hire-a-helper\": $fa-var-hire-a-helper,\n  \"creative-commons-by\": $fa-var-creative-commons-by,\n  \"unity\": $fa-var-unity,\n  \"whmcs\": $fa-var-whmcs,\n  \"rocketchat\": $fa-var-rocketchat,\n  \"vk\": $fa-var-vk,\n  \"untappd\": $fa-var-untappd,\n  \"mailchimp\": $fa-var-mailchimp,\n  \"css3-alt\": $fa-var-css3-alt,\n  \"square-reddit\": $fa-var-square-reddit,\n  \"reddit-square\": $fa-var-reddit-square,\n  \"vimeo-v\": $fa-var-vimeo-v,\n  \"contao\": $fa-var-contao,\n  \"square-font-awesome\": $fa-var-square-font-awesome,\n  \"deskpro\": $fa-var-deskpro,\n  \"brave\": $fa-var-brave,\n  \"sistrix\": $fa-var-sistrix,\n  \"square-instagram\": $fa-var-square-instagram,\n  \"instagram-square\": $fa-var-instagram-square,\n  \"battle-net\": $fa-var-battle-net,\n  \"the-red-yeti\": $fa-var-the-red-yeti,\n  \"square-hacker-news\": $fa-var-square-hacker-news,\n  \"hacker-news-square\": $fa-var-hacker-news-square,\n  \"edge\": $fa-var-edge,\n  \"threads\": $fa-var-threads,\n  \"napster\": $fa-var-napster,\n  \"square-snapchat\": $fa-var-square-snapchat,\n  \"snapchat-square\": $fa-var-snapchat-square,\n  \"google-plus-g\": $fa-var-google-plus-g,\n  \"artstation\": $fa-var-artstation,\n  \"markdown\": $fa-var-markdown,\n  \"sourcetree\": $fa-var-sourcetree,\n  \"google-plus\": $fa-var-google-plus,\n  \"diaspora\": $fa-var-diaspora,\n  \"foursquare\": $fa-var-foursquare,\n  \"stack-overflow\": $fa-var-stack-overflow,\n  \"github-alt\": $fa-var-github-alt,\n  \"phoenix-squadron\": $fa-var-phoenix-squadron,\n  \"pagelines\": $fa-var-pagelines,\n  \"algolia\": $fa-var-algolia,\n  \"red-river\": $fa-var-red-river,\n  \"creative-commons-sa\": $fa-var-creative-commons-sa,\n  \"safari\": $fa-var-safari,\n  \"google\": $fa-var-google,\n  \"square-font-awesome-stroke\": $fa-var-square-font-awesome-stroke,\n  \"font-awesome-alt\": $fa-var-font-awesome-alt,\n  \"atlassian\": $fa-var-atlassian,\n  \"linkedin-in\": $fa-var-linkedin-in,\n  \"digital-ocean\": $fa-var-digital-ocean,\n  \"nimblr\": $fa-var-nimblr,\n  \"chromecast\": $fa-var-chromecast,\n  \"evernote\": $fa-var-evernote,\n  \"hacker-news\": $fa-var-hacker-news,\n  \"creative-commons-sampling\": $fa-var-creative-commons-sampling,\n  \"adversal\": $fa-var-adversal,\n  \"creative-commons\": $fa-var-creative-commons,\n  \"watchman-monitoring\": $fa-var-watchman-monitoring,\n  \"fonticons\": $fa-var-fonticons,\n  \"weixin\": $fa-var-weixin,\n  \"shirtsinbulk\": $fa-var-shirtsinbulk,\n  \"codepen\": $fa-var-codepen,\n  \"git-alt\": $fa-var-git-alt,\n  \"lyft\": $fa-var-lyft,\n  \"rev\": $fa-var-rev,\n  \"windows\": $fa-var-windows,\n  \"wizards-of-the-coast\": $fa-var-wizards-of-the-coast,\n  \"square-viadeo\": $fa-var-square-viadeo,\n  \"viadeo-square\": $fa-var-viadeo-square,\n  \"meetup\": $fa-var-meetup,\n  \"centos\": $fa-var-centos,\n  \"adn\": $fa-var-adn,\n  \"cloudsmith\": $fa-var-cloudsmith,\n  \"opensuse\": $fa-var-opensuse,\n  \"pied-piper-alt\": $fa-var-pied-piper-alt,\n  \"square-dribbble\": $fa-var-square-dribbble,\n  \"dribbble-square\": $fa-var-dribbble-square,\n  \"codiepie\": $fa-var-codiepie,\n  \"node\": $fa-var-node,\n  \"mix\": $fa-var-mix,\n  \"steam\": $fa-var-steam,\n  \"cc-apple-pay\": $fa-var-cc-apple-pay,\n  \"scribd\": $fa-var-scribd,\n  \"debian\": $fa-var-debian,\n  \"openid\": $fa-var-openid,\n  \"instalod\": $fa-var-instalod,\n  \"expeditedssl\": $fa-var-expeditedssl,\n  \"sellcast\": $fa-var-sellcast,\n  \"square-twitter\": $fa-var-square-twitter,\n  \"twitter-square\": $fa-var-twitter-square,\n  \"r-project\": $fa-var-r-project,\n  \"delicious\": $fa-var-delicious,\n  \"freebsd\": $fa-var-freebsd,\n  \"vuejs\": $fa-var-vuejs,\n  \"accusoft\": $fa-var-accusoft,\n  \"ioxhost\": $fa-var-ioxhost,\n  \"fonticons-fi\": $fa-var-fonticons-fi,\n  \"app-store\": $fa-var-app-store,\n  \"cc-mastercard\": $fa-var-cc-mastercard,\n  \"itunes-note\": $fa-var-itunes-note,\n  \"golang\": $fa-var-golang,\n  \"kickstarter\": $fa-var-kickstarter,\n  \"square-kickstarter\": $fa-var-square-kickstarter,\n  \"grav\": $fa-var-grav,\n  \"weibo\": $fa-var-weibo,\n  \"uncharted\": $fa-var-uncharted,\n  \"firstdraft\": $fa-var-firstdraft,\n  \"square-youtube\": $fa-var-square-youtube,\n  \"youtube-square\": $fa-var-youtube-square,\n  \"wikipedia-w\": $fa-var-wikipedia-w,\n  \"wpressr\": $fa-var-wpressr,\n  \"rendact\": $fa-var-rendact,\n  \"angellist\": $fa-var-angellist,\n  \"galactic-republic\": $fa-var-galactic-republic,\n  \"nfc-directional\": $fa-var-nfc-directional,\n  \"skype\": $fa-var-skype,\n  \"joget\": $fa-var-joget,\n  \"fedora\": $fa-var-fedora,\n  \"stripe-s\": $fa-var-stripe-s,\n  \"meta\": $fa-var-meta,\n  \"laravel\": $fa-var-laravel,\n  \"hotjar\": $fa-var-hotjar,\n  \"bluetooth-b\": $fa-var-bluetooth-b,\n  \"square-letterboxd\": $fa-var-square-letterboxd,\n  \"sticker-mule\": $fa-var-sticker-mule,\n  \"creative-commons-zero\": $fa-var-creative-commons-zero,\n  \"hips\": $fa-var-hips,\n  \"behance\": $fa-var-behance,\n  \"reddit\": $fa-var-reddit,\n  \"discord\": $fa-var-discord,\n  \"chrome\": $fa-var-chrome,\n  \"app-store-ios\": $fa-var-app-store-ios,\n  \"cc-discover\": $fa-var-cc-discover,\n  \"wpbeginner\": $fa-var-wpbeginner,\n  \"confluence\": $fa-var-confluence,\n  \"shoelace\": $fa-var-shoelace,\n  \"mdb\": $fa-var-mdb,\n  \"dochub\": $fa-var-dochub,\n  \"accessible-icon\": $fa-var-accessible-icon,\n  \"ebay\": $fa-var-ebay,\n  \"amazon\": $fa-var-amazon,\n  \"unsplash\": $fa-var-unsplash,\n  \"yarn\": $fa-var-yarn,\n  \"square-steam\": $fa-var-square-steam,\n  \"steam-square\": $fa-var-steam-square,\n  \"500px\": $fa-var-500px,\n  \"square-vimeo\": $fa-var-square-vimeo,\n  \"vimeo-square\": $fa-var-vimeo-square,\n  \"asymmetrik\": $fa-var-asymmetrik,\n  \"font-awesome\": $fa-var-font-awesome,\n  \"font-awesome-flag\": $fa-var-font-awesome-flag,\n  \"font-awesome-logo-full\": $fa-var-font-awesome-logo-full,\n  \"gratipay\": $fa-var-gratipay,\n  \"apple\": $fa-var-apple,\n  \"hive\": $fa-var-hive,\n  \"gitkraken\": $fa-var-gitkraken,\n  \"keybase\": $fa-var-keybase,\n  \"apple-pay\": $fa-var-apple-pay,\n  \"padlet\": $fa-var-padlet,\n  \"amazon-pay\": $fa-var-amazon-pay,\n  \"square-github\": $fa-var-square-github,\n  \"github-square\": $fa-var-github-square,\n  \"stumbleupon\": $fa-var-stumbleupon,\n  \"fedex\": $fa-var-fedex,\n  \"phoenix-framework\": $fa-var-phoenix-framework,\n  \"shopify\": $fa-var-shopify,\n  \"neos\": $fa-var-neos,\n  \"square-threads\": $fa-var-square-threads,\n  \"hackerrank\": $fa-var-hackerrank,\n  \"researchgate\": $fa-var-researchgate,\n  \"swift\": $fa-var-swift,\n  \"angular\": $fa-var-angular,\n  \"speakap\": $fa-var-speakap,\n  \"angrycreative\": $fa-var-angrycreative,\n  \"y-combinator\": $fa-var-y-combinator,\n  \"empire\": $fa-var-empire,\n  \"envira\": $fa-var-envira,\n  \"google-scholar\": $fa-var-google-scholar,\n  \"square-gitlab\": $fa-var-square-gitlab,\n  \"gitlab-square\": $fa-var-gitlab-square,\n  \"studiovinari\": $fa-var-studiovinari,\n  \"pied-piper\": $fa-var-pied-piper,\n  \"wordpress\": $fa-var-wordpress,\n  \"product-hunt\": $fa-var-product-hunt,\n  \"firefox\": $fa-var-firefox,\n  \"linode\": $fa-var-linode,\n  \"goodreads\": $fa-var-goodreads,\n  \"square-odnoklassniki\": $fa-var-square-odnoklassniki,\n  \"odnoklassniki-square\": $fa-var-odnoklassniki-square,\n  \"jsfiddle\": $fa-var-jsfiddle,\n  \"sith\": $fa-var-sith,\n  \"themeisle\": $fa-var-themeisle,\n  \"page4\": $fa-var-page4,\n  \"hashnode\": $fa-var-hashnode,\n  \"react\": $fa-var-react,\n  \"cc-paypal\": $fa-var-cc-paypal,\n  \"squarespace\": $fa-var-squarespace,\n  \"cc-stripe\": $fa-var-cc-stripe,\n  \"creative-commons-share\": $fa-var-creative-commons-share,\n  \"bitcoin\": $fa-var-bitcoin,\n  \"keycdn\": $fa-var-keycdn,\n  \"opera\": $fa-var-opera,\n  \"itch-io\": $fa-var-itch-io,\n  \"umbraco\": $fa-var-umbraco,\n  \"galactic-senate\": $fa-var-galactic-senate,\n  \"ubuntu\": $fa-var-ubuntu,\n  \"draft2digital\": $fa-var-draft2digital,\n  \"stripe\": $fa-var-stripe,\n  \"houzz\": $fa-var-houzz,\n  \"gg\": $fa-var-gg,\n  \"dhl\": $fa-var-dhl,\n  \"square-pinterest\": $fa-var-square-pinterest,\n  \"pinterest-square\": $fa-var-pinterest-square,\n  \"xing\": $fa-var-xing,\n  \"blackberry\": $fa-var-blackberry,\n  \"creative-commons-pd\": $fa-var-creative-commons-pd,\n  \"playstation\": $fa-var-playstation,\n  \"quinscape\": $fa-var-quinscape,\n  \"less\": $fa-var-less,\n  \"blogger-b\": $fa-var-blogger-b,\n  \"opencart\": $fa-var-opencart,\n  \"vine\": $fa-var-vine,\n  \"signal-messenger\": $fa-var-signal-messenger,\n  \"paypal\": $fa-var-paypal,\n  \"gitlab\": $fa-var-gitlab,\n  \"typo3\": $fa-var-typo3,\n  \"reddit-alien\": $fa-var-reddit-alien,\n  \"yahoo\": $fa-var-yahoo,\n  \"dailymotion\": $fa-var-dailymotion,\n  \"affiliatetheme\": $fa-var-affiliatetheme,\n  \"pied-piper-pp\": $fa-var-pied-piper-pp,\n  \"bootstrap\": $fa-var-bootstrap,\n  \"odnoklassniki\": $fa-var-odnoklassniki,\n  \"nfc-symbol\": $fa-var-nfc-symbol,\n  \"mintbit\": $fa-var-mintbit,\n  \"ethereum\": $fa-var-ethereum,\n  \"speaker-deck\": $fa-var-speaker-deck,\n  \"creative-commons-nc-eu\": $fa-var-creative-commons-nc-eu,\n  \"patreon\": $fa-var-patreon,\n  \"avianex\": $fa-var-avianex,\n  \"ello\": $fa-var-ello,\n  \"gofore\": $fa-var-gofore,\n  \"bimobject\": $fa-var-bimobject,\n  \"brave-reverse\": $fa-var-brave-reverse,\n  \"facebook-f\": $fa-var-facebook-f,\n  \"square-google-plus\": $fa-var-square-google-plus,\n  \"google-plus-square\": $fa-var-google-plus-square,\n  \"web-awesome\": $fa-var-web-awesome,\n  \"mandalorian\": $fa-var-mandalorian,\n  \"first-order-alt\": $fa-var-first-order-alt,\n  \"osi\": $fa-var-osi,\n  \"google-wallet\": $fa-var-google-wallet,\n  \"d-and-d-beyond\": $fa-var-d-and-d-beyond,\n  \"periscope\": $fa-var-periscope,\n  \"fulcrum\": $fa-var-fulcrum,\n  \"cloudscale\": $fa-var-cloudscale,\n  \"forumbee\": $fa-var-forumbee,\n  \"mizuni\": $fa-var-mizuni,\n  \"schlix\": $fa-var-schlix,\n  \"square-xing\": $fa-var-square-xing,\n  \"xing-square\": $fa-var-xing-square,\n  \"bandcamp\": $fa-var-bandcamp,\n  \"wpforms\": $fa-var-wpforms,\n  \"cloudversify\": $fa-var-cloudversify,\n  \"usps\": $fa-var-usps,\n  \"megaport\": $fa-var-megaport,\n  \"magento\": $fa-var-magento,\n  \"spotify\": $fa-var-spotify,\n  \"optin-monster\": $fa-var-optin-monster,\n  \"fly\": $fa-var-fly,\n  \"aviato\": $fa-var-aviato,\n  \"itunes\": $fa-var-itunes,\n  \"cuttlefish\": $fa-var-cuttlefish,\n  \"blogger\": $fa-var-blogger,\n  \"flickr\": $fa-var-flickr,\n  \"viber\": $fa-var-viber,\n  \"soundcloud\": $fa-var-soundcloud,\n  \"digg\": $fa-var-digg,\n  \"tencent-weibo\": $fa-var-tencent-weibo,\n  \"letterboxd\": $fa-var-letterboxd,\n  \"symfony\": $fa-var-symfony,\n  \"maxcdn\": $fa-var-maxcdn,\n  \"etsy\": $fa-var-etsy,\n  \"facebook-messenger\": $fa-var-facebook-messenger,\n  \"audible\": $fa-var-audible,\n  \"think-peaks\": $fa-var-think-peaks,\n  \"bilibili\": $fa-var-bilibili,\n  \"erlang\": $fa-var-erlang,\n  \"x-twitter\": $fa-var-x-twitter,\n  \"cotton-bureau\": $fa-var-cotton-bureau,\n  \"dashcube\": $fa-var-dashcube,\n  \"42-group\": $fa-var-42-group,\n  \"innosoft\": $fa-var-innosoft,\n  \"stack-exchange\": $fa-var-stack-exchange,\n  \"elementor\": $fa-var-elementor,\n  \"square-pied-piper\": $fa-var-square-pied-piper,\n  \"pied-piper-square\": $fa-var-pied-piper-square,\n  \"creative-commons-nd\": $fa-var-creative-commons-nd,\n  \"palfed\": $fa-var-palfed,\n  \"superpowers\": $fa-var-superpowers,\n  \"resolving\": $fa-var-resolving,\n  \"xbox\": $fa-var-xbox,\n  \"square-web-awesome-stroke\": $fa-var-square-web-awesome-stroke,\n  \"searchengin\": $fa-var-searchengin,\n  \"tiktok\": $fa-var-tiktok,\n  \"square-facebook\": $fa-var-square-facebook,\n  \"facebook-square\": $fa-var-facebook-square,\n  \"renren\": $fa-var-renren,\n  \"linux\": $fa-var-linux,\n  \"glide\": $fa-var-glide,\n  \"linkedin\": $fa-var-linkedin,\n  \"hubspot\": $fa-var-hubspot,\n  \"deploydog\": $fa-var-deploydog,\n  \"twitch\": $fa-var-twitch,\n  \"ravelry\": $fa-var-ravelry,\n  \"mixer\": $fa-var-mixer,\n  \"square-lastfm\": $fa-var-square-lastfm,\n  \"lastfm-square\": $fa-var-lastfm-square,\n  \"vimeo\": $fa-var-vimeo,\n  \"mendeley\": $fa-var-mendeley,\n  \"uniregistry\": $fa-var-uniregistry,\n  \"figma\": $fa-var-figma,\n  \"creative-commons-remix\": $fa-var-creative-commons-remix,\n  \"cc-amazon-pay\": $fa-var-cc-amazon-pay,\n  \"dropbox\": $fa-var-dropbox,\n  \"instagram\": $fa-var-instagram,\n  \"cmplid\": $fa-var-cmplid,\n  \"upwork\": $fa-var-upwork,\n  \"facebook\": $fa-var-facebook,\n  \"gripfire\": $fa-var-gripfire,\n  \"jedi-order\": $fa-var-jedi-order,\n  \"uikit\": $fa-var-uikit,\n  \"fort-awesome-alt\": $fa-var-fort-awesome-alt,\n  \"phabricator\": $fa-var-phabricator,\n  \"ussunnah\": $fa-var-ussunnah,\n  \"earlybirds\": $fa-var-earlybirds,\n  \"trade-federation\": $fa-var-trade-federation,\n  \"autoprefixer\": $fa-var-autoprefixer,\n  \"whatsapp\": $fa-var-whatsapp,\n  \"square-upwork\": $fa-var-square-upwork,\n  \"slideshare\": $fa-var-slideshare,\n  \"google-play\": $fa-var-google-play,\n  \"viadeo\": $fa-var-viadeo,\n  \"line\": $fa-var-line,\n  \"google-drive\": $fa-var-google-drive,\n  \"servicestack\": $fa-var-servicestack,\n  \"simplybuilt\": $fa-var-simplybuilt,\n  \"bitbucket\": $fa-var-bitbucket,\n  \"imdb\": $fa-var-imdb,\n  \"deezer\": $fa-var-deezer,\n  \"raspberry-pi\": $fa-var-raspberry-pi,\n  \"jira\": $fa-var-jira,\n  \"docker\": $fa-var-docker,\n  \"screenpal\": $fa-var-screenpal,\n  \"bluetooth\": $fa-var-bluetooth,\n  \"gitter\": $fa-var-gitter,\n  \"d-and-d\": $fa-var-d-and-d,\n  \"microblog\": $fa-var-microblog,\n  \"cc-diners-club\": $fa-var-cc-diners-club,\n  \"gg-circle\": $fa-var-gg-circle,\n  \"pied-piper-hat\": $fa-var-pied-piper-hat,\n  \"kickstarter-k\": $fa-var-kickstarter-k,\n  \"yandex\": $fa-var-yandex,\n  \"readme\": $fa-var-readme,\n  \"html5\": $fa-var-html5,\n  \"sellsy\": $fa-var-sellsy,\n  \"square-web-awesome\": $fa-var-square-web-awesome,\n  \"sass\": $fa-var-sass,\n  \"wirsindhandwerk\": $fa-var-wirsindhandwerk,\n  \"wsh\": $fa-var-wsh,\n  \"buromobelexperte\": $fa-var-buromobelexperte,\n  \"salesforce\": $fa-var-salesforce,\n  \"octopus-deploy\": $fa-var-octopus-deploy,\n  \"medapps\": $fa-var-medapps,\n  \"ns8\": $fa-var-ns8,\n  \"pinterest-p\": $fa-var-pinterest-p,\n  \"apper\": $fa-var-apper,\n  \"fort-awesome\": $fa-var-fort-awesome,\n  \"waze\": $fa-var-waze,\n  \"bluesky\": $fa-var-bluesky,\n  \"cc-jcb\": $fa-var-cc-jcb,\n  \"snapchat\": $fa-var-snapchat,\n  \"snapchat-ghost\": $fa-var-snapchat-ghost,\n  \"fantasy-flight-games\": $fa-var-fantasy-flight-games,\n  \"rust\": $fa-var-rust,\n  \"wix\": $fa-var-wix,\n  \"square-behance\": $fa-var-square-behance,\n  \"behance-square\": $fa-var-behance-square,\n  \"supple\": $fa-var-supple,\n  \"webflow\": $fa-var-webflow,\n  \"rebel\": $fa-var-rebel,\n  \"css3\": $fa-var-css3,\n  \"staylinked\": $fa-var-staylinked,\n  \"kaggle\": $fa-var-kaggle,\n  \"space-awesome\": $fa-var-space-awesome,\n  \"deviantart\": $fa-var-deviantart,\n  \"cpanel\": $fa-var-cpanel,\n  \"goodreads-g\": $fa-var-goodreads-g,\n  \"square-git\": $fa-var-square-git,\n  \"git-square\": $fa-var-git-square,\n  \"square-tumblr\": $fa-var-square-tumblr,\n  \"tumblr-square\": $fa-var-tumblr-square,\n  \"trello\": $fa-var-trello,\n  \"creative-commons-nc-jp\": $fa-var-creative-commons-nc-jp,\n  \"get-pocket\": $fa-var-get-pocket,\n  \"perbyte\": $fa-var-perbyte,\n  \"grunt\": $fa-var-grunt,\n  \"weebly\": $fa-var-weebly,\n  \"connectdevelop\": $fa-var-connectdevelop,\n  \"leanpub\": $fa-var-leanpub,\n  \"black-tie\": $fa-var-black-tie,\n  \"themeco\": $fa-var-themeco,\n  \"python\": $fa-var-python,\n  \"android\": $fa-var-android,\n  \"bots\": $fa-var-bots,\n  \"free-code-camp\": $fa-var-free-code-camp,\n  \"hornbill\": $fa-var-hornbill,\n  \"js\": $fa-var-js,\n  \"ideal\": $fa-var-ideal,\n  \"git\": $fa-var-git,\n  \"dev\": $fa-var-dev,\n  \"sketch\": $fa-var-sketch,\n  \"yandex-international\": $fa-var-yandex-international,\n  \"cc-amex\": $fa-var-cc-amex,\n  \"uber\": $fa-var-uber,\n  \"github\": $fa-var-github,\n  \"php\": $fa-var-php,\n  \"alipay\": $fa-var-alipay,\n  \"youtube\": $fa-var-youtube,\n  \"skyatlas\": $fa-var-skyatlas,\n  \"firefox-browser\": $fa-var-firefox-browser,\n  \"replyd\": $fa-var-replyd,\n  \"suse\": $fa-var-suse,\n  \"jenkins\": $fa-var-jenkins,\n  \"twitter\": $fa-var-twitter,\n  \"rockrms\": $fa-var-rockrms,\n  \"pinterest\": $fa-var-pinterest,\n  \"buffer\": $fa-var-buffer,\n  \"npm\": $fa-var-npm,\n  \"yammer\": $fa-var-yammer,\n  \"btc\": $fa-var-btc,\n  \"dribbble\": $fa-var-dribbble,\n  \"stumbleupon-circle\": $fa-var-stumbleupon-circle,\n  \"internet-explorer\": $fa-var-internet-explorer,\n  \"stubber\": $fa-var-stubber,\n  \"telegram\": $fa-var-telegram,\n  \"telegram-plane\": $fa-var-telegram-plane,\n  \"old-republic\": $fa-var-old-republic,\n  \"odysee\": $fa-var-odysee,\n  \"square-whatsapp\": $fa-var-square-whatsapp,\n  \"whatsapp-square\": $fa-var-whatsapp-square,\n  \"node-js\": $fa-var-node-js,\n  \"edge-legacy\": $fa-var-edge-legacy,\n  \"slack\": $fa-var-slack,\n  \"slack-hash\": $fa-var-slack-hash,\n  \"medrt\": $fa-var-medrt,\n  \"usb\": $fa-var-usb,\n  \"tumblr\": $fa-var-tumblr,\n  \"vaadin\": $fa-var-vaadin,\n  \"quora\": $fa-var-quora,\n  \"square-x-twitter\": $fa-var-square-x-twitter,\n  \"reacteurope\": $fa-var-reacteurope,\n  \"medium\": $fa-var-medium,\n  \"medium-m\": $fa-var-medium-m,\n  \"amilia\": $fa-var-amilia,\n  \"mixcloud\": $fa-var-mixcloud,\n  \"flipboard\": $fa-var-flipboard,\n  \"viacoin\": $fa-var-viacoin,\n  \"critical-role\": $fa-var-critical-role,\n  \"sitrox\": $fa-var-sitrox,\n  \"discourse\": $fa-var-discourse,\n  \"joomla\": $fa-var-joomla,\n  \"mastodon\": $fa-var-mastodon,\n  \"airbnb\": $fa-var-airbnb,\n  \"wolf-pack-battalion\": $fa-var-wolf-pack-battalion,\n  \"buy-n-large\": $fa-var-buy-n-large,\n  \"gulp\": $fa-var-gulp,\n  \"creative-commons-sampling-plus\": $fa-var-creative-commons-sampling-plus,\n  \"strava\": $fa-var-strava,\n  \"ember\": $fa-var-ember,\n  \"canadian-maple-leaf\": $fa-var-canadian-maple-leaf,\n  \"teamspeak\": $fa-var-teamspeak,\n  \"pushed\": $fa-var-pushed,\n  \"wordpress-simple\": $fa-var-wordpress-simple,\n  \"nutritionix\": $fa-var-nutritionix,\n  \"wodu\": $fa-var-wodu,\n  \"google-pay\": $fa-var-google-pay,\n  \"intercom\": $fa-var-intercom,\n  \"zhihu\": $fa-var-zhihu,\n  \"korvue\": $fa-var-korvue,\n  \"pix\": $fa-var-pix,\n  \"steam-symbol\": $fa-var-steam-symbol,\n);\n","// specific icon class definition\n// -------------------------\n\n/* Font Awesome uses the Unicode Private Use Area (PUA) to ensure screen\nreaders do not read off random characters that represent icons */\n\n@each $name, $icon in $fa-icons {\n  .#{$fa-css-prefix}-#{$name}::before { content: unquote(\"\\\"#{ $icon }\\\"\"); }\n}\n\n","/*!\n * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com\n * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)\n * Copyright 2024 Fonticons, Inc.\n */\n@import 'functions';\n@import 'variables';\n\n:root, :host {\n  --#{$fa-css-prefix}-style-family-classic: '#{ $fa-style-family }';\n  --#{$fa-css-prefix}-font-solid: normal 900 1em/1 '#{ $fa-style-family }';\n}\n\n@font-face {\n  font-family: 'Font Awesome 6 Free';\n  font-style: normal;\n  font-weight: 900;\n  font-display: $fa-font-display;\n  src: url('#{$fa-font-path}/fa-solid-900.woff2') format('woff2'),\n    url('#{$fa-font-path}/fa-solid-900.ttf') format('truetype');\n}\n\n.fas,\n.#{$fa-css-prefix}-solid {\n  font-weight: 900;\n}\n","/*!\n * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com\n * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)\n * Copyright 2024 Fonticons, Inc.\n */\n@import 'functions';\n@import 'variables';\n\n:root, :host {\n  --#{$fa-css-prefix}-style-family-classic: '#{ $fa-style-family }';\n  --#{$fa-css-prefix}-font-regular: normal 400 1em/1 '#{ $fa-style-family }';\n}\n\n@font-face {\n  font-family: 'Font Awesome 6 Free';\n  font-style: normal;\n  font-weight: 400;\n  font-display: $fa-font-display;\n  src: url('#{$fa-font-path}/fa-regular-400.woff2') format('woff2'),\n    url('#{$fa-font-path}/fa-regular-400.ttf') format('truetype');\n}\n\n.far,\n.#{$fa-css-prefix}-regular {\n  font-weight: 400;\n}\n","/*!\n * Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com\n * License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License)\n * Copyright 2024 Fonticons, Inc.\n */\n@import 'functions';\n@import 'variables';\n\n:root, :host {\n  --#{$fa-css-prefix}-style-family-brands: 'Font Awesome 6 Brands';\n  --#{$fa-css-prefix}-font-brands: normal 400 1em/1 'Font Awesome 6 Brands';\n}\n\n@font-face {\n  font-family: 'Font Awesome 6 Brands';\n  font-style: normal;\n  font-weight: 400;\n  font-display: $fa-font-display;\n  src: url('#{$fa-font-path}/fa-brands-400.woff2') format('woff2'),\n    url('#{$fa-font-path}/fa-brands-400.ttf') format('truetype');\n}\n\n.fab,\n.#{$fa-css-prefix}-brands {\n  font-weight: 400;\n}\n\n@each $name, $icon in $fa-brand-icons {\n  .#{$fa-css-prefix}-#{$name}:before { content: unquote(\"\\\"#{ $icon }\\\"\"); }\n}\n","html {\n  /*****************************************************************************\n  * Overall Layout Variables\n  */\n\n  // Header height will impact the top offset for many sections\n  // Article header is 66% of Header\n  --pst-header-height: 4rem;\n  --pst-header-article-height: calc(var(--pst-header-height) * 2 / 3);\n  --pst-sidebar-secondary: 17rem;\n}\n\n/*******************************************************************************\n* Breakpoints that trigger UI changes\n*\n* Note that media-breakpoint-down begins at the next highest level!\n* So we should choose a media-breakpoint-down one *lower* than when we want to start\n* example: media-breakpoint-up(md) and media-breakpoint-down(sm) trigger at the same time\n* ref: https://github.com/twbs/bootstrap/issues/31214\n*/\n$breakpoint-sidebar-primary: lg; // When we collapse the primary sidebar\n$breakpoint-sidebar-secondary: xl; // When we collapse the secondary sidebar\n$breakpoint-page-width: 88rem; // taken from sphinx-basic-ng, which we are ultimately going to inherit\n\n/*******************************************************************************\n* Define the animation behaviour\n*/\n$animation-time: 200ms;\n\n/*******************************************************************************\n* UI shaping and padding\n*/\n$admonition-border-radius: 0.25rem;\n\n// In this theme, some focus rings have rounded corners while others do not.\n// This variable sets the border radius for the rounded focus rings.\n$focus-ring-radius: 0.125rem; // 2px at 100% zoom and 16px base font.\n\n$navbar-link-padding-y: 0.25rem;\n\n// Ensure that there is no overlap of bumper disks (smallest circle that\n// contains the bounding box of the interactive element).\n// - https://github.com/Quansight-Labs/czi-scientific-python-mgmt/issues/81#issuecomment-2251325783\n$nav-icon-column-gap: 1.12rem;\n\n// Determines vertical space between entries in both the section (left/primary\n// sidebar) and page (right/secondary sidebar) table of contents\n$toc-item-padding-y: 0.25rem;\n","html {\n  /*****************************************************************************\n  * Font features used in this theme\n  */\n\n  // base font size - applied at body/html level\n  --pst-font-size-base: 1rem;\n\n  // heading font sizes based on a medium contrast type scale\n  // - see: https://github.com/Quansight-Labs/czi-scientific-python-mgmt/issues/97#issuecomment-2310531483\n  --pst-font-size-h1: 2.625rem;\n  --pst-font-size-h2: 2.125rem;\n  --pst-font-size-h3: 1.75rem;\n  --pst-font-size-h4: 1.5rem;\n  --pst-font-size-h5: 1.25rem;\n  --pst-font-size-h6: 1rem;\n\n  // smaller than heading font sizes\n  --pst-font-size-milli: 0.9rem;\n\n  // Sidebar styles\n  --pst-sidebar-font-size: 0.9rem;\n  --pst-sidebar-font-size-mobile: 1.1rem;\n  --pst-sidebar-header-font-size: 1.2rem;\n  --pst-sidebar-header-font-weight: 600;\n\n  // Admonition styles\n  --pst-admonition-font-weight-heading: 600;\n\n  // Font weights\n  --pst-font-weight-caption: 300;\n  --pst-font-weight-heading: 600;\n\n  // Font family\n  // These are adapted from https://systemfontstack.com/ */\n  --pst-font-family-base-system: -apple-system, \"BlinkMacSystemFont\", \"Segoe UI\",\n    \"Helvetica Neue\", \"Arial\", sans-serif, \"Apple Color Emoji\", \"Segoe UI Emoji\",\n    \"Segoe UI Symbol\";\n  --pst-font-family-monospace-system: \"SFMono-Regular\", \"Menlo\", \"Consolas\",\n    \"Monaco\", \"Liberation Mono\", \"Lucida Console\", monospace;\n  --pst-font-family-base: var(--pst-font-family-base-system);\n  --pst-font-family-heading: var(--pst-font-family-base-system);\n  --pst-font-family-monospace: var(--pst-font-family-monospace-system);\n}\n\n$line-height-body: 1.65;\n$fa-font-path: \"vendor/fontawesome/webfonts/\";\n","html {\n  /*****************************************************************************\n  * Icon\n  */\n\n  // Font size across all icons\n  --pst-font-size-icon: 1.5rem;\n\n  // Font Awesome default icons\n  --pst-icon-check-circle: \"\\f058\"; // fa-solid fa-circle-check\n  --pst-icon-info-circle: \"\\f05a\"; // fa-solid fa-circle-info\n  --pst-icon-exclamation-triangle: \"\\f071\"; // fa-solid fa-triangle-exclamation\n  --pst-icon-exclamation-circle: \"\\f06a\"; // fa-solid fa-circle-exclamation\n  --pst-icon-times-circle: \"\\f057\"; // fa-solid fa-circle-xmark\n  --pst-icon-lightbulb: \"\\f0eb\"; // fa-solid fa-lightbulb\n  --pst-icon-download: \"\\f019\"; // fa-solid fa-download\n  --pst-icon-angle-left: \"\\f104\"; // fa-solid fa-angle-left\n  --pst-icon-angle-right: \"\\f105\"; // fa-solid fa-angle-right\n  --pst-icon-external-link: \"\\f35d\"; // fa-solid fa-up-right-from-square\n  --pst-icon-search-minus: \"\\f010\"; // fa-solid fa-magnifying-glass-minus\n  --pst-icon-github: \"\\f09b\"; // fa-brands fa-github\n  --pst-icon-gitlab: \"\\f296\"; // fa-brands fa-gitlab\n  --pst-icon-share: \"\\f064\"; // fa-solid fa-share\n  --pst-icon-bell: \"\\f0f3\"; // fa-solid fa-bell\n  --pst-icon-pencil: \"\\f303\"; // fa-solid fa-pencil\n\n  // Bootstrap icons\n  --pst-breadcrumb-divider: \"\\f105\";\n}\n","html {\n  /*****************************************************************************\n  * Admonitions\n  **/\n\n  --pst-icon-admonition-default: var(--pst-icon-bell);\n  --pst-icon-admonition-note: var(--pst-icon-info-circle);\n  --pst-icon-admonition-attention: var(--pst-icon-exclamation-circle);\n  --pst-icon-admonition-caution: var(--pst-icon-exclamation-triangle);\n  --pst-icon-admonition-warning: var(--pst-icon-exclamation-triangle);\n  --pst-icon-admonition-danger: var(--pst-icon-exclamation-triangle);\n  --pst-icon-admonition-error: var(--pst-icon-times-circle);\n  --pst-icon-admonition-hint: var(--pst-icon-lightbulb);\n  --pst-icon-admonition-tip: var(--pst-icon-lightbulb);\n  --pst-icon-admonition-important: var(--pst-icon-exclamation-circle);\n  --pst-icon-admonition-seealso: var(--pst-icon-share);\n  --pst-icon-admonition-todo: var(--pst-icon-pencil);\n}\n","html {\n  /*****************************************************************************\n  * versionmodified\n  **/\n\n  --pst-icon-versionmodified-default: var(--pst-icon-exclamation-circle);\n  --pst-icon-versionmodified-added: var(--pst-icon-exclamation-circle);\n  --pst-icon-versionmodified-changed: var(--pst-icon-exclamation-circle);\n  --pst-icon-versionmodified-deprecated: var(--pst-icon-exclamation-circle);\n}\n","html {\n  font-size: var(--pst-font-size-base);\n  scroll-padding-top: calc(var(--pst-header-height) + 1rem);\n}\n\nbody {\n  background-color: var(--pst-color-background);\n  font-family: var(--pst-font-family-base);\n  font-weight: 400;\n  line-height: $line-height-body;\n  color: var(--pst-color-text-base);\n  min-height: 100vh;\n  display: flex;\n  flex-direction: column;\n\n  // hack to avoid the black background on some browser including Safari\n  &::-webkit-scrollbar-track {\n    background-color: var(--pst-color-background);\n  }\n}\n\np {\n  margin-bottom: 1.15rem;\n  font-size: 1em;\n\n  /* section header in docstring pages */\n  &.rubric {\n    border-bottom: 1px solid var(--pst-color-border);\n  }\n\n  &.centered {\n    text-align: center;\n  }\n}\n\na {\n  @include link-style-default;\n\n  // The # anchor that appears on hover over headings\n  &.headerlink {\n    color: var(--pst-color-secondary);\n    font-size: 0.8em;\n    padding: 0 4px;\n    margin-left: 0.2em;\n    text-decoration: none;\n    user-select: none;\n  }\n\n  // set up a icon next to the shorten links from github and gitlab\n  &.github,\n  &.gitlab {\n    &::before {\n      color: var(--pst-color-text-muted);\n      font: var(--fa-font-brands);\n      margin-right: 0.25rem;\n    }\n  }\n\n  &.github::before {\n    content: var(--pst-icon-github);\n  }\n\n  &.gitlab::before {\n    content: var(--pst-icon-gitlab);\n  }\n}\n\n%heading-style {\n  margin: 2.75rem 0 1.05rem;\n  font-family: var(--pst-font-family-heading);\n  font-weight: var(--pst-font-weight-heading);\n  line-height: 1.15;\n}\n\nh1 {\n  @extend %heading-style;\n\n  margin-top: 0;\n  font-size: var(--pst-font-size-h1);\n  color: var(--pst-color-heading);\n}\n\nh2 {\n  @extend %heading-style;\n\n  font-size: var(--pst-font-size-h2);\n  color: var(--pst-color-heading);\n}\n\nh3 {\n  @extend %heading-style;\n\n  font-size: var(--pst-font-size-h3);\n  color: var(--pst-color-heading);\n}\n\nh4 {\n  @extend %heading-style;\n\n  font-size: var(--pst-font-size-h4);\n  color: var(--pst-color-heading);\n}\n\nh5 {\n  @extend %heading-style;\n\n  font-size: var(--pst-font-size-h5);\n  color: var(--pst-color-text-base);\n}\n\nh6 {\n  @extend %heading-style;\n\n  font-size: var(--pst-font-size-h6);\n  color: var(--pst-color-text-base);\n}\n\nsmall,\n.text_small {\n  font-size: var(--pst-font-size-milli);\n}\n\nhr {\n  border: 0;\n  border-top: 1px solid var(--pst-color-border);\n}\n\npre,\ncode,\nkbd,\nsamp {\n  font-family: var(--pst-font-family-monospace);\n}\n\nkbd {\n  // use theme negative\n  background-color: var(--pst-color-on-background);\n  color: var(--pst-color-text-muted);\n\n  // Compound keyboard elements will have nested kbd in them, so this prevents double lines\n  &:not(.compound) {\n    border: 1px solid var(--pst-color-border);\n    margin: 0 0.1rem;\n    padding: 0.1rem 0.4rem;\n    box-shadow: 1px 1px 1px var(--pst-color-shadow);\n  }\n}\n\ncode {\n  color: var(--pst-color-inline-code);\n}\n\npre {\n  margin: 1.5em 0;\n  padding: 1rem;\n  background-color: var(--pst-color-surface);\n  color: var(--pst-color-text-base);\n  line-height: 1.2em;\n  border: 1px solid var(--pst-color-border);\n  border-radius: $admonition-border-radius;\n\n  .linenos {\n    // minimum opacity to make the line numbers WCAG AA conformant\n    opacity: 0.8;\n    padding-right: 10px;\n  }\n}\n\n// the back to top btn\n#pst-back-to-top {\n  // zindex-tooltip comes from Bootstrap https://getbootstrap.com/docs/5.2/layout/z-index/\n  z-index: $zindex-tooltip;\n  position: fixed;\n  display: none;\n  top: 90vh;\n  left: 50vw;\n  transform: translate(-50%);\n  color: var(--pst-color-secondary-text);\n  background-color: var(--pst-color-secondary);\n  border: none;\n\n  .fa-arrow-up {\n    // Using margin instead of a space character prevents the space between the\n    // icon and the text from being underlined when the button is hovered.\n    margin-inline-end: 0.5em;\n  }\n\n  &:hover {\n    @include link-style-hover;\n\n    text-decoration-thickness: 1px;\n    background-color: var(--pst-color-secondary-highlight);\n    color: var(--pst-color-secondary-highlight-text);\n  }\n\n  &:focus-visible {\n    box-shadow: none;\n    outline: $focus-ring-outline;\n    outline-color: var(--pst-color-secondary);\n    outline-offset: $focus-ring-width;\n  }\n}\n\n// Focus ring\n// ----------\n// Note: The Bootstrap stylesheet provides the focus ring (customized by this\n// theme via Sass variables in _bootstrap.scss) in some cases. This rule covers\n// all other cases.\n:focus-visible {\n  outline: $focus-ring-outline;\n  box-shadow: none; // override Bootstrap\n}\n","// Consistent styling for links\n// ============================\n\n@use \"sass:string\";\n\n// Define some useful variables for links styling consistency\n\n// The default thickness of the underline for links will be either:\n//  - 1px\n//  - 0.0625rem if it's thicker than 1px because the user has changed the text\n//    size in their browser\n$link-underline-thickness: string.unquote(\"max(1px, .0625rem)\") !default;\n\n// Offset of link underlines from text baseline\n// The default is 3px expressed as ems, as calculated against the default body\n// font size (on desktop).\n$link-underline-offset: 0.1578em !default;\n\n// Thickness of link underlines in hover state\n// The default for each link will be the thickest of the following:\n//  - 3px\n//  - 0.1875rem, if it's thicker than 3px because the user has changed the text\n//    size in their browser\n//  - 0.12em (relative to the link's text size)\n$link-hover-decoration-thickness: string.unquote(\n  \"max(3px, .1875rem, .12em)\"\n) !default;\n\n// Ensures links have an underline decoration by default - needed to meet\n// WCAG SC 1.4.1\n@mixin link-decoration {\n  text-decoration: underline;\n\n  @if $link-underline-thickness {\n    text-decoration-thickness: $link-underline-thickness;\n  }\n\n  @if $link-underline-offset {\n    text-underline-offset: $link-underline-offset;\n  }\n}\n\n// Ensures links have an underline decoration on hover - distinct from the\n// default behaviour\n@mixin link-decoration-hover {\n  @if $link-hover-decoration-thickness {\n    text-decoration-thickness: $link-hover-decoration-thickness;\n\n    // Disable ink skipping on underlines on hover.\n    text-decoration-skip-ink: none;\n  }\n}\n\n// Simple hover style - can be used alone or in conjunction with other mixins\n// Add the text underline and change in thickness on hover.\n// Intended for use with the `:hover` pseudo-class.\n@mixin link-style-hover {\n  @include link-decoration;\n  @include link-decoration-hover;\n\n  color: var(--pst-color-link-hover);\n}\n\n// Default link styles\n// -------------------\n// Defines: default unvisited, visited, hover, and active.\n// TODO: @trallard to improve focus styles in subsequent PR\n@mixin link-style-default {\n  // So that really long links don't spill out of their container\n  word-wrap: break-word;\n  color: var(--pst-color-link);\n\n  @include link-decoration;\n\n  &:hover {\n    color: var(--pst-color-link-hover);\n\n    @include link-decoration-hover;\n  }\n\n  // TODO: @trallard to add active styles in subsequent PR\n  &:active {\n    color: var(--pst-color-link);\n  }\n\n  // Visited should still be hoverable\n  &:visited {\n    color: var(--pst-color-link);\n\n    &:hover {\n      color: var(--pst-color-link-hover);\n    }\n  }\n}\n\n// Text link styles\n// ----------------\n// Makes links use the muted text colour and removes the underline.\n// Use this mixin for navigation bar links.\n@mixin link-style-text {\n  color: var(--pst-color-text-muted);\n  text-decoration: none;\n\n  &:hover {\n    color: var(--pst-color-link-hover);\n\n    @include link-decoration;\n    @include link-decoration-hover;\n  }\n}\n\n// Sidebar and TOC links\n// ---------------------\n// Makes links use the muted text colour and removes the underline.\n// Use this mixin for navigation the primary sidebar and table of contents.\n// Active and hover should work together rather than one overriding the other.\n@mixin link-sidebar {\n  color: var(--pst-color-text-muted);\n  text-decoration: none;\n\n  &:hover {\n    text-decoration: underline;\n    background-color: transparent;\n    color: var(--pst-color-link-hover);\n\n    @include link-decoration-hover;\n  }\n\n  // TODO: @trallard to update active styles in subsequent PR\n  &:active {\n    color: var(--pst-color-link-hover);\n  }\n\n  &:focus-visible {\n    box-shadow: $focus-ring-box-shadow;\n    outline: none;\n    z-index: 10; // keep focus ring on top (prevent the link-sidebar-current notch from lying on top of the ring)\n  }\n}\n\n// Sidebar current page link styles\n// --------------------------------\n// Adds a vertical line on the left hand side of the link to indicate that\n// it's the current page. Note this is distinct from an active state.\n// Used on the primary sidebar and the TOC.\n// We want the side box shadow to have the same thickness as the hover underline\n@mixin link-sidebar-current {\n  font-weight: 600;\n  color: var(--pst-color-primary);\n\n  @if $link-hover-decoration-thickness {\n    $notch-shadow: inset\n      $link-hover-decoration-thickness\n      0\n      0\n      var(--pst-color-primary);\n\n    box-shadow: $notch-shadow;\n\n    &:focus-visible {\n      box-shadow: $notch-shadow, $focus-ring-box-shadow;\n      outline: none;\n    }\n  }\n}\n\n// Header navbar text and icon links\n// ---------------------------------\n// (includes light/dark mode button)\n\n// This mixin makes it possible to show hover/underline and focus/ring styles at\n// the same time. The trick is to use:\n//    - a pseudo-element with bottom border for the hover underline\n//    - a CSS outline for the focus ring.\n\n// Normally we use box-shadow for underline and outline for focus ring. But we\n// cannot apply box-shadow and outline together on the same element because the\n// border-radius value that we use to round the outline will also round the\n// box-shadow used for the underline. We also cannot use text-underline because\n// it does not work on non-text links, nor do we want to use it on text links\n// that we want to treat as blocks, such as the header nav links because the\n// underline will wrap across two lines if the link text also wraps across two\n// lines.\n@mixin link-style-block {\n  color: var(--pst-color-text-muted);\n\n  // Set position relative so that the child ::before pseudo-element's absolute\n  // position is relative to this element.\n  position: relative;\n\n  // Set up pseudo-element used for hover underline styles\n  &::before {\n    content: \"\";\n    display: block;\n    position: absolute;\n    inset: 0;\n    background-color: transparent;\n\n    @if $link-hover-decoration-thickness {\n      bottom: calc(-1 * $link-hover-decoration-thickness);\n      margin: $link-hover-decoration-thickness 0;\n    }\n  }\n\n  &:hover {\n    color: var(--pst-color-secondary);\n    text-decoration: none; // override the link-style-hover mixin\n    &::before {\n      @if $link-hover-decoration-thickness {\n        border-bottom: $link-hover-decoration-thickness\n          solid\n          var(--pst-color-secondary);\n      }\n    }\n  }\n\n  &:focus-visible {\n    box-shadow: none; // override Bootstrap\n    outline: 3px solid var(--pst-color-accent);\n    outline-offset: $focus-ring-width;\n  }\n}\n","/***\n * Rules for the UX/UI of skip navigation link btn.\n * It's only visible to people\n * navigating with keyboard for accessibility purposes\n * ref: https://www.youtube.com/watch?v=VUR0I5mqq7I\n ***/\n\n.skip-link {\n  position: fixed;\n  top: 0;\n  left: 0;\n  right: 0;\n  text-align: center;\n  background-color: var(--pst-color-warning);\n  padding: 0.5rem;\n  z-index: $zindex-modal;\n  border-bottom: 1px solid var(--pst-color-border);\n\n  // This shows / hides the button\n  transform: translateY(-100%);\n  transition: transform 150ms ease-in-out;\n\n  &:focus-within {\n    transform: translateY(0%);\n  }\n\n  a {\n    // Ensure we are using a WCAG conformant colour\n    color: var(--pst-color-warning-text) !important;\n\n    &:focus-visible {\n      // use color with sufficient contrast\n      outline-color: $foundation-black;\n    }\n  }\n}\n",".bd-container {\n  flex-grow: 1;\n  display: flex;\n  justify-content: center;\n\n  .bd-container__inner {\n    display: flex;\n  }\n}\n\n.bd-page-width {\n  width: 100%;\n\n  @include media-breakpoint-up(lg) {\n    max-width: $breakpoint-page-width;\n  }\n}\n",".pst-async-banner-revealer {\n  // Setting height to 0 and overflow to hidden allows us to add up the heights\n  // of this element's children before revealing them.\n  height: 0;\n  overflow: hidden;\n\n  // Height to be set by JavaScript, which should trigger the following\n  // transition rule (unless the user has set their system to reduce motion).\n  transition: height 300ms ease-in-out;\n\n  @media (prefers-reduced-motion) {\n    transition: none;\n  }\n}\n\n#bd-header-version-warning,\n.bd-header-announcement {\n  min-height: 3rem;\n  width: 100%;\n  display: flex;\n  position: relative;\n  align-items: center;\n  justify-content: center;\n  text-align: center;\n  padding: 0.5rem 12.5%; // Horizontal padding so the width is 75%\n  // One breakpoint less than $breakpoint-sidebar-primary. See variables/_layout.scss for more info.\n  @include media-breakpoint-down(lg) {\n    // Announcements can take a bit more width on mobile\n    padding: 0.5rem 2%;\n  }\n\n  p {\n    font-weight: bold;\n    margin: 0;\n  }\n\n  // Ensure there is enough contrast against the background\n  a {\n    color: var(--pst-color-inline-code-links);\n  }\n\n  // The \"Switch to stable version\" link (styled like a button)\n  .pst-button-link-to-stable-version {\n    @include box-shadow;\n\n    $background-color: var(--pst-color-danger);\n    $hover-background-color: var(--pst-color-danger-highlight);\n    $color: var(--pst-color-danger-text);\n\n    background-color: $background-color;\n    border-color: $background-color;\n    color: $color;\n    border-radius: 0.25rem;\n\n    &:hover {\n      // applying the same style as the SD \"buttons\"\n      @include link-style-hover;\n\n      text-decoration-thickness: 1px;\n      background-color: $hover-background-color;\n      border-color: $hover-background-color;\n      color: var(--pst-color-danger-highlight-text);\n    }\n\n    &:focus-visible {\n      outline: $focus-ring-width solid $background-color;\n      outline-offset: $focus-ring-width;\n    }\n  }\n}\n\n// Bg color is now defined in the theme color palette - using our secondary color\n.bd-header-announcement {\n  background-color: var(--pst-color-secondary-bg);\n}\n\n#bd-header-version-warning {\n  background-color: var(--pst-color-danger-bg);\n}\n","/*********************************************\n* SASS Mixins\n*********************************************/\n\n/**\n * A consistent box shadow style we apply across elements.\n */\n@mixin box-shadow() {\n  box-shadow:\n    0 0.2rem 0.5rem var(--pst-color-shadow),\n    0 0 0.0625rem var(--pst-color-shadow) !important;\n}\n\n/**\n   * Set background of some cell outputs to white-ish to make sure colors work\n   * This is because many libraries make output that only looks good on white\n   */\n@mixin cell-output-background {\n  color: var(--pst-color-on-background);\n  background-color: var(--pst-color-text-base);\n  border-radius: 0.25rem;\n  padding: 0.5rem;\n}\n\n@mixin table-colors {\n  color: var(--pst-color-table);\n  border: 1px solid var(--pst-color-table-outer-border);\n\n  th,\n  td {\n    ~ th,\n    ~ td {\n      border-left: 1px solid var(--pst-color-table-inner-border);\n    }\n  }\n\n  thead {\n    tr {\n      background-color: var(--pst-color-table-heading-bg);\n      border-bottom: 2px solid var(--pst-color-primary);\n    }\n  }\n\n  tbody {\n    tr {\n      &:nth-child(odd) {\n        background-color: var(--pst-color-table-row-zebra-low-bg);\n      }\n\n      &:nth-child(even) {\n        background-color: var(--pst-color-table-row-zebra-high-bg);\n      }\n    }\n  }\n}\n\n// Minimum mouse hit area\n// ----------------------\n// Ensures that the element has a minimum hit area that conforms to\n// accessibility guidelines. For WCAG AA, we need 24px x 24px, see:\n// https://www.w3.org/WAI/WCAG22/Understanding/target-size-minimum.html\n@mixin min-hit-area() {\n  box-sizing: border-box;\n  min-width: 24px;\n  min-height: 24px;\n}\n\n// Meant to darken the element on hover in light mode, or\n// lighten on hover in dark mode.\n@mixin hover-darken-lighten {\n  &:hover {\n    filter: brightness(0.9);\n\n    html[data-theme=\"dark\"] & {\n      filter: brightness(1.1);\n    }\n  }\n}\n","/**\n * Main content area\n */\n.bd-main {\n  flex-grow: 1;\n  flex-direction: column;\n  display: flex;\n  min-width: 0;\n\n  .bd-content {\n    display: flex;\n    justify-content: center;\n    height: 100%;\n\n    .bd-article-container {\n      justify-content: start;\n      display: flex;\n      flex-direction: column;\n\n      // Max-width is slightly more than the W3 since our docs often have images.\n      // We shoot for about 100 characters per line instead of 80.\n      // ref: https://www.w3.org/WAI/tutorials/page-structure/styling/#line-length\n      width: 100%;\n      max-width: 60em;\n      overflow-x: auto; // Prevent wide content from pushing off the secondary sidebar\n      padding: 1rem;\n\n      .bd-article {\n        // Give a bit more verticle spacing on wide screens\n        @include media-breakpoint-up($breakpoint-sidebar-secondary) {\n          padding-top: 1.5rem;\n          padding-left: 2rem;\n        }\n      }\n    }\n  }\n}\n",".bd-footer {\n  width: 100%;\n  border-top: 1px solid var(--pst-color-border);\n\n  .bd-footer__inner {\n    display: flex;\n    flex-grow: 1;\n    padding: 1rem;\n    margin: auto;\n  }\n\n  .footer-items__start,\n  .footer-items__center,\n  .footer-items__end {\n    display: flex;\n    flex-direction: column;\n    gap: 0.5rem;\n    justify-content: center;\n    flex-grow: 1;\n  }\n\n  .footer-items__center {\n    text-align: center;\n  }\n\n  .footer-items__end {\n    text-align: end;\n  }\n\n  // So that paragraphs don't take up extra room\n  .footer-item p {\n    margin-bottom: 0;\n  }\n}\n",".bd-footer-article {\n  margin-top: auto;\n\n  .footer-article-items {\n    display: flex;\n    flex-direction: column;\n  }\n}\n",".bd-footer-content {\n  .footer-content-items {\n    display: flex;\n    flex-direction: column;\n    margin-top: auto;\n  }\n}\n","/**\n * Header at the top of the page\n * It includes the announcement bar and the navigation bar.\n */\n\n// Styling for the Icon links can be found in components/_icon-links.scss\n\n// If we want the shadow to only point downward in the future, set\n// box-shadow to: 0 0.125rem 0.25rem -0.125rem rgba(0, 0, 0, 0.11);\n.bd-header {\n  position: sticky;\n  top: 0;\n  z-index: $zindex-fixed;\n\n  // Overrides bootstrap\n  background-color: var(--pst-color-on-background) !important;\n  box-shadow: 0 0.125rem 0.25rem 0 var(--pst-color-shadow);\n  width: 100%;\n  padding: 0;\n  max-width: 100vw;\n  justify-content: center;\n\n  .bd-header__inner {\n    display: flex;\n    align-items: center;\n    height: fit-content;\n    padding-left: 1rem;\n    padding-right: 1rem;\n  }\n\n  :focus-visible {\n    border-radius: $focus-ring-radius;\n  }\n\n  // These items will define the height of the header\n  .navbar-item {\n    height: var(--pst-header-height);\n    max-height: var(--pst-header-height);\n    display: flex;\n    align-items: center;\n  }\n\n  // Hide the navbar header items on mobile because they're in the sidebar\n  .navbar-header-items {\n    display: none;\n    flex-shrink: 1;\n\n    @include media-breakpoint-up($breakpoint-sidebar-primary) {\n      display: inherit;\n      flex-grow: 1;\n      padding: 0 0 0 0.5rem;\n    }\n  }\n\n  .navbar-header-items__end,\n  .navbar-header-items__center,\n  .navbar-header-items__start {\n    display: flex;\n    align-items: center;\n    flex-flow: wrap;\n\n    // In case we wrap our items to multiple rows on small screens\n    row-gap: 0;\n  }\n\n  .navbar-header-items__end,\n  .navbar-header-items__center {\n    column-gap: $nav-icon-column-gap;\n  }\n\n  // A little smaller because this is displayed by default on mobile\n  .navbar-header-items__start {\n    flex-shrink: 0;\n    margin-right: auto;\n    gap: 0.5rem;\n  }\n\n  .navbar-header-items__end {\n    // End navbar items should snap to the right\n    justify-content: end;\n  }\n\n  // Contains the navigation links within the navbar\n  ul.navbar-nav {\n    display: flex;\n\n    @include media-breakpoint-up($breakpoint-sidebar-primary) {\n      // Align on wide screens so the dropdown button is centered properly\n      align-items: baseline;\n    }\n\n    > li.nav-item {\n      margin-inline: 2px; // breathing room so hover and focus styles do not overlap\n\n      > .nav-link {\n        @include link-style-block;\n\n        padding-inline: 6px;\n      }\n\n      &.current {\n        > .nav-link {\n          color: var(--pst-color-primary);\n\n          // Underline the current navbar item\n          &::before {\n            border-bottom: 3px solid var(--pst-color-primary);\n          }\n        }\n      }\n\n      &.dropdown {\n        margin-inline: 4px;\n\n        button {\n          padding-inline: 8px;\n        }\n\n        > .dropdown-toggle {\n          border-radius: $focus-ring-radius; // make border radius the same for both hover ring and focus ring\n          color: var(--pst-color-text-muted);\n\n          &:focus-visible {\n            box-shadow: $focus-ring-box-shadow;\n          }\n\n          &:hover {\n            text-decoration: none;\n            box-shadow: 0 0 0 $focus-ring-width var(--pst-color-link-hover); // purple focus ring\n            // Brighten the text on hover (muted -> base)\n            color: var(--pst-color-text-base);\n          }\n        }\n      }\n    }\n\n    li a.nav-link.dropdown-item {\n      @include link-style-text;\n    }\n\n    // Dropdowns for the extra links\n    .dropdown {\n      button {\n        display: unset;\n        border: none;\n\n        &:hover {\n          @include link-style-hover;\n        }\n      }\n\n      .dropdown-menu {\n        z-index: $zindex-popover;\n        border: 1px solid var(--pst-color-border);\n        box-shadow: 0 0 0.3rem 0.1rem var(--pst-color-shadow);\n        background-color: var(--pst-color-on-background);\n        padding: 0.5rem 0;\n        margin: 0.5rem 0;\n        min-width: 20rem;\n\n        .dropdown-item {\n          // Give the items in the dropdown some breathing room but let the hit\n          // and hover area of the items extend to the edges of the menu\n          padding: 0.25rem 1.5rem;\n\n          // Override Bootstrap\n          &:focus:not(:hover, :active) {\n            background-color: inherit;\n          }\n\n          &:focus-visible {\n            z-index: 10; // keep focus ring on top (prevent the hover background of the next dropdown item from covering the ring)\n          }\n        }\n\n        // Hide the menu unless show has been clicked\n        &:not(.show) {\n          display: none;\n        }\n      }\n    }\n  }\n\n  // **************************************************************\n  // Showing and hiding the sidebar toggle buttons and header items\n  // **************************************************************\n\n  // Toggle buttons\n  button.sidebar-toggle {\n    font-size: var(--pst-font-size-icon);\n    color: var(--pst-color-muted);\n    margin-bottom: 0;\n    background-color: inherit;\n    padding: 0.5rem;\n  }\n\n  button.primary-toggle {\n    margin-right: 1rem;\n\n    @include media-breakpoint-up($breakpoint-sidebar-primary) {\n      display: none;\n    }\n  }\n\n  button.secondary-toggle {\n    margin-left: 1rem;\n\n    @include media-breakpoint-up($breakpoint-sidebar-secondary) {\n      display: none;\n    }\n  }\n}\n\n// inline the element in the navbar as long as they fit and use display block when collapsing\n@include media-breakpoint-up($breakpoint-sidebar-primary) {\n  .navbar-center-items .navbar-item {\n    display: inline-block;\n  }\n}\n\n.nav-link {\n  &:hover {\n    @include link-style-hover;\n  }\n\n  // Override Bootstrap\n  transition: none;\n\n  &.nav-external::after {\n    font: var(--fa-font-solid);\n    content: var(--pst-icon-external-link);\n    font-size: 0.75em;\n    margin-left: 0.3em;\n  }\n}\n\n.bd-navbar-elements li.nav-item i {\n  font-size: 0.7rem;\n  padding-left: 2px;\n  vertical-align: middle;\n}\n\n// THe elements next to the hamburger menu only show on narrow screens\n.navbar-persistent--mobile {\n  margin-left: auto;\n\n  @include media-breakpoint-up($breakpoint-sidebar-primary) {\n    display: none;\n  }\n}\n\n// The navbar-persistent content should only show on wide screens\n.navbar-persistent--container {\n  display: none;\n\n  @include media-breakpoint-up($breakpoint-sidebar-primary) {\n    display: flex;\n  }\n}\n",".header-article__inner {\n  padding: 0 0.5rem;\n\n  // The items define the height so that it disappears if there are no items\n  .header-article-item {\n    min-height: var(--pst-header-article-height);\n  }\n\n  .header-article-items__end {\n    margin-left: auto;\n  }\n}\n","/**\n * The primary sidebar on the left.\n * e.g., between-pages navigation.\n */\n\n$sidebar-padding-right: 1rem;\n\n.bd-sidebar-primary {\n  display: flex;\n  flex-direction: column;\n  gap: 1rem;\n  max-height: calc(100vh - var(--pst-header-height));\n  position: sticky;\n  top: var(--pst-header-height);\n\n  @include make-col(3);\n\n  // Borders padding and whitespace\n  padding: 2rem $sidebar-padding-right 1rem 1rem;\n  border-right: 1px solid var(--pst-color-border);\n  background-color: var(--pst-color-background);\n  overflow-y: auto;\n  font-size: var(--pst-sidebar-font-size-mobile);\n\n  @include media-breakpoint-up($breakpoint-sidebar-primary) {\n    font-size: var(--pst-sidebar-font-size);\n  }\n\n  :focus-visible {\n    border-radius: $focus-ring-radius;\n  }\n\n  // override bootstrap when navlink are displayed in the sidebar\n  .nav-link {\n    font-size: var(--pst-sidebar-font-size-mobile);\n  }\n\n  &.no-sidebar {\n    border-right: 0;\n  }\n\n  &.hide-on-wide {\n    @include media-breakpoint-up($breakpoint-sidebar-primary) {\n      display: none;\n    }\n  }\n\n  // Headers shouldn't be colored in the sidebars and some extensions add headers\n  h1,\n  h2,\n  h3,\n  h4 {\n    color: var(--pst-color-text-base);\n  }\n\n  .sidebar-primary-items__start,\n  .sidebar-primary-items__end {\n    .sidebar-primary-item {\n      padding: 0.5rem 0;\n    }\n  }\n\n  // Hide the sidebar header items on widescreen since they are visible in the header\n  .sidebar-header-items {\n    display: flex;\n    flex-direction: column;\n\n    .sidebar-header-items__title {\n      font-weight: var(--pst-sidebar-header-font-weight);\n      font-size: var(--pst-sidebar-header-font-size);\n      color: var(--pst-color-text-base);\n      margin-bottom: 0.5rem;\n    }\n\n    .bd-navbar-elements {\n      .nav-link {\n        &:focus-visible {\n          box-shadow: none; // Override Bootstrap\n          outline: $focus-ring-outline;\n          outline-offset: $focus-ring-width;\n        }\n      }\n    }\n\n    // Center header items get displayed vertically, end items are displayed horizontally\n    .sidebar-header-items__center {\n      display: flex;\n      flex-direction: column;\n    }\n\n    // Positioning end items\n    .sidebar-header-items__end {\n      display: flex;\n      align-items: center;\n      gap: 1rem;\n    }\n\n    @include media-breakpoint-up($breakpoint-sidebar-primary) {\n      display: none;\n    }\n  }\n\n  .sidebar-primary-items__start {\n    // Add a border on mobile to separate it from the header sidebar area\n    border-top: 1px solid var(--pst-color-border);\n\n    @include media-breakpoint-up($breakpoint-sidebar-primary) {\n      border-top: none;\n    }\n  }\n\n  .sidebar-primary-items__end {\n    margin-top: auto;\n    margin-bottom: 1em;\n  }\n\n  .list-caption {\n    list-style: none;\n    padding-left: 0;\n\n    // Level 0 TOC heading is put inside the <summary> tag\n    // so let the <summary> tag take up more space\n    li.toctree-l0.has-children {\n      > details {\n        > summary {\n          position: relative;\n          height: auto;\n          width: auto;\n          display: flex;\n          justify-content: space-between;\n          align-items: baseline;\n\n          .toctree-toggle {\n            // Prevent toggle icon from getting squished by summary being a\n            // flexbox\n            flex: 0 0 auto;\n\n            // Make the level 0 chevron icon slightly bigger than descendant\n            // levels\n            .fa-chevron-down {\n              font-size: 1rem;\n            }\n          }\n        }\n      }\n    }\n  }\n\n  li.has-children {\n    $toctree-toggle-width: 30px;\n\n    position: relative;\n\n    > .reference,\n    .caption {\n      margin-right: calc(\n        $toctree-toggle-width + $focus-ring-width\n      ); // keep clear of the toggle icon\n\n      padding-top: 0.25rem; // align caption text with toggle chevron\n    }\n\n    > details {\n      > summary {\n        // Remove browser default toggle icon\n        list-style: none;\n\n        &::-webkit-details-marker {\n          display: none;\n        }\n\n        // The summary element is natively focusable, but delegate the focus state to the toggle icon\n        &:focus-visible {\n          outline: none;\n\n          > .toctree-toggle {\n            outline: $focus-ring-outline;\n            outline-offset: -$focus-ring-width; // Prevent right side of focus ring from disappearing underneath the sidebar's right edge\n          }\n        }\n\n        // Container for expand/collapse chevron icon\n        .toctree-toggle {\n          cursor: pointer;\n\n          // Position it so that it's aligned with the top right corner of the\n          // last positioned element, in this case the li.has-children\n          position: absolute;\n          top: 0;\n          right: 0;\n\n          // Give it dimensions\n          width: $toctree-toggle-width;\n          height: $toctree-toggle-width; // make it square\n\n          // Vertically and horizontally center the icon within the container\n          display: inline-flex;\n          justify-content: center;\n          align-items: center;\n\n          .fa-chevron-down {\n            font-size: 0.75rem;\n          }\n        }\n      }\n\n      // The section is open/expanded, rotate the toggle icon (chevron) so it\n      // points up instead of down\n      &[open] {\n        > summary {\n          .fa-chevron-down {\n            transform: rotate(180deg);\n          }\n        }\n      }\n    }\n  }\n}\n\n/* Between-page links and captions */\nnav.bd-links {\n  margin-right: -$sidebar-padding-right; // align toctree toggle chevrons with right edge of sidebar and allow text to flow closer to the right edge\n\n  @include media-breakpoint-up($breakpoint-sidebar-primary) {\n    display: block;\n  }\n\n  ul {\n    display: block;\n    list-style: none;\n\n    // Reduce padding of nested `ul` items a bit\n    ul {\n      padding: 0 0 0 1rem;\n    }\n  }\n\n  li > a {\n    display: block;\n    padding: $toc-item-padding-y 0.65rem;\n\n    @include link-sidebar;\n\n    box-shadow: none;\n    margin-right: $focus-ring-width; // prevent the right side focus ring from disappearing under the sidebar right edge\n\n    &.reference.external {\n      &::after {\n        font: var(--fa-font-solid);\n        content: var(--pst-icon-external-link);\n        font-size: 0.75em;\n        margin-left: 0.3em;\n      }\n    }\n  }\n\n  .current > a {\n    @include link-sidebar-current;\n\n    background-color: transparent;\n  }\n\n  // Title\n  p.bd-links__title {\n    font-size: var(--pst-sidebar-header-font-size);\n    font-weight: var(--pst-sidebar-header-font-weight);\n    margin-bottom: 0.5rem;\n  }\n\n  // Toctree captions\n  p.caption {\n    font-weight: var(--pst-sidebar-header-font-weight);\n    position: relative;\n    margin-top: 1.25rem;\n    margin-bottom: 0.5rem;\n    color: var(--pst-color-text-base);\n\n    &:first-child {\n      margin-top: 0;\n    }\n\n    font-size: var(--pst-sidebar-font-size-mobile);\n\n    @include media-breakpoint-up($breakpoint-sidebar-primary) {\n      font-size: var(--pst-sidebar-font-size);\n    }\n  }\n}\n","/**\n * Secondary sidebar on the right.\n * e.g., in-page table of contents.\n */\n.bd-sidebar-secondary {\n  display: flex;\n  order: 2;\n  flex-shrink: 0;\n  flex-direction: column;\n  position: sticky;\n  top: var(--pst-header-height);\n  max-height: calc(100vh - var(--pst-header-height));\n  padding: 2rem 1rem 1rem;\n  width: var(--pst-sidebar-secondary);\n  font-size: var(--pst-sidebar-font-size-mobile);\n\n  @include media-breakpoint-up($breakpoint-sidebar-secondary) {\n    font-size: var(--pst-sidebar-font-size);\n  }\n\n  // Color and border\n  background-color: var(--pst-color-background);\n  overflow-y: auto;\n}\n\n.sidebar-secondary-item {\n  padding: 0.5rem;\n\n  @include media-breakpoint-up($breakpoint-sidebar-secondary) {\n    border-left: 1px solid var(--pst-color-border);\n    padding-left: 1rem;\n  }\n\n  i {\n    padding-right: 0.5rem;\n  }\n}\n","/*******************************************************************************\n* Rules for the UX/UI of sidebar sliding drawers on mobile\n* Note that this sheet controls styles across many parts of the theme\n* It is aggregated into this one sheet instead of being split across\n* components in order to keep it easier to debug in one place.\n* It is broken up into major sections below.\n*/\n\n/*******************************************************************************\n* Sidebar drawer behavior\n*/\n\n/**\n * NOTE: We use this mixin to define the toggle behavior on narrow screens,\n * And the wide-screen behavior of the sections is defined in their own section\n * .scss files.\n */\n@mixin sliding-drawer($side: \"left\") {\n  position: fixed;\n  top: 0;\n  z-index: $zindex-modal;\n  height: 100vh;\n  max-height: 100vh;\n  width: 75%;\n  flex-grow: 0.75;\n  max-width: 350px;\n  transition:\n    visibility $animation-time ease-out,\n    margin $animation-time ease-out;\n  visibility: hidden;\n  border: 0;\n\n  @if $side == \"right\" {\n    margin-right: -75%;\n    right: 0;\n  } @else {\n    margin-left: -75%;\n    left: 0;\n  }\n}\n\n.bd-sidebar::backdrop {\n  background-color: black;\n  opacity: 0.5;\n}\n\n.bd-sidebar-primary {\n  @include media-breakpoint-down($breakpoint-sidebar-primary) {\n    @include sliding-drawer(\"left\");\n  }\n\n  &[open] {\n    margin-left: 0;\n    visibility: visible;\n  }\n}\n\n.bd-sidebar-secondary {\n  @include media-breakpoint-down($breakpoint-sidebar-secondary) {\n    @include sliding-drawer(\"right\");\n  }\n\n  &[open] {\n    margin-right: 0;\n    visibility: visible;\n  }\n}\n","/**\n * Breadcrumbs for parent pages meant for the article header\n */\nul.bd-breadcrumbs {\n  list-style: none;\n  padding-left: 0;\n  display: flex;\n  flex-wrap: wrap;\n\n  // Font size slightly smaller to avoid cluttering up space too much\n  font-size: 0.8rem;\n\n  li.breadcrumb-item {\n    display: flex;\n    align-items: center;\n    white-space: nowrap;\n    overflow-x: hidden;\n\n    a,\n    .ellipsis {\n      overflow-x: hidden;\n      text-overflow: ellipsis;\n\n      // Need to add margin, otherwise the overflow: hidden rule on the parent\n      // will hide the focus ring\n      margin: $focus-ring-width;\n    }\n\n    // Style should look like heavier in-page links\n    // keeping visited in the default link colour\n    font-weight: bold;\n\n    a {\n      @include link-style-text;\n    }\n\n    // Items that aren't the home have a caret to the left\n    &:not(.breadcrumb-home)::before {\n      font: var(--fa-font-solid);\n      font-size: 0.8rem;\n      content: var(--pst-breadcrumb-divider);\n      color: var(--pst-color-text-muted);\n      padding: 0 0.5rem;\n    }\n  }\n}\n","/**\n * Icon links in the navbar\n */\n\n.pst-navbar-icon {\n  // Extra specificity needed for overrides\n  html & {\n    @include min-hit-area;\n    @include link-style-block;\n\n    display: flex;\n    align-items: center;\n    justify-content: center;\n\n    // Bootstrap overrides\n    border-radius: 0;\n    border: none;\n    font-size: 1rem;\n    line-height: $line-height-body; // Override Boostrap, which defines a separate line-height for buttons\n    padding: $navbar-link-padding-y 0; // Horizontal white space in nav bar between items is controlled via column gap rule on the container.\n\n    // Make the navbar icon links have the same size as the navbar text links\n    height: calc(2 * $navbar-link-padding-y + $line-height-body * 1rem);\n  }\n}\n\nul.navbar-icon-links {\n  display: flex;\n  flex-flow: row wrap;\n  column-gap: $nav-icon-column-gap;\n  justify-content: space-evenly;\n  align-items: center;\n  padding-left: 0;\n  margin-bottom: 0;\n  list-style: none;\n\n  // Icons styling\n  i {\n    &.fa-brands,\n    &.fa-regular,\n    &.fa-solid {\n      vertical-align: middle;\n      font-style: normal;\n      font-size: var(--pst-font-size-icon);\n    }\n\n    /* Social media buttons hard-code the brand color */\n    &.fa-square-twitter::before {\n      color: #55acee;\n    }\n\n    &.fa-square-gitlab::before {\n      color: #548;\n    }\n\n    &.fa-bitbucket::before {\n      color: #0052cc;\n    }\n  }\n\n  // Force images to be icon-sized\n  img.icon-link-image {\n    height: 1.5em;\n    border-radius: 0.2rem;\n  }\n\n  .fa-pydata {\n    stroke: var(--pst-color-background);\n    stroke-linejoin: round;\n    stroke-width: 0.35;\n  }\n}\n","/**\n * Logo in the navbar\n */\n\n.navbar-brand {\n  position: relative;\n  height: var(--pst-header-height);\n  max-height: var(--pst-header-height);\n  padding: 0.5rem 0;\n  width: auto;\n  margin: 0;\n  display: flex;\n\n  // Ensure that the logo stays the same length while other content shrinks\n  flex-shrink: 0;\n  align-items: center;\n  gap: 0.5rem;\n\n  // If there's no logo image, we use a p element w/ the site title\n  p {\n    color: var(--pst-color-text-base);\n    margin-bottom: 0;\n  }\n\n  // If there's a logo, it'll be in an img block\n  img {\n    max-width: 100%;\n    height: 100%;\n    width: auto;\n  }\n\n  &:hover,\n  &:visited:hover {\n    @include link-style-hover;\n\n    color: var(--pst-color-text-base);\n  }\n}\n","/**\n * Navigation links in the navbar and icon links\n */\nul.navbar-nav {\n  // Reduce padding of nested `ul` items a bit\n  ul {\n    padding: 0 0 0 1rem;\n  }\n\n  // Navbar links - do not have an underline by default\n  li {\n    display: flex;\n    flex-direction: column;\n\n    a {\n      display: flex;\n      align-items: center;\n      height: 100%;\n      padding-top: $navbar-link-padding-y;\n      padding-bottom: $navbar-link-padding-y;\n\n      @include link-style-text;\n    }\n  }\n}\n","/**\n * The list of in-page TOC links\n */\n.page-toc {\n  .section-nav {\n    padding-left: 0;\n    border-bottom: none;\n\n    ul {\n      padding-left: 1rem;\n    }\n  }\n\n  // override bootstrap settings\n  .nav-link {\n    font-size: var(--pst-sidebar-font-size-mobile);\n\n    @include media-breakpoint-up($breakpoint-sidebar-secondary) {\n      font-size: var(--pst-sidebar-font-size);\n    }\n  }\n\n  .onthispage {\n    color: var(--pst-color-text-base);\n    font-weight: var(--pst-sidebar-header-font-weight);\n    margin-bottom: 0.5rem;\n  }\n}\n","/**\n* Previous / Next navigation buttons\n**/\n.prev-next-area {\n  width: 100%;\n\n  p {\n    color: var(--pst-color-text-muted);\n    margin: 0 0.3em;\n    line-height: 1.3em;\n  }\n\n  i {\n    font-size: 1.2em;\n  }\n\n  a {\n    // So that buttons align with icons\n    display: flex;\n    align-items: center;\n    border: none;\n    padding: 10px;\n    max-width: 45%;\n    overflow-x: hidden;\n    color: var(--pst-color-text-muted);\n    text-decoration: none;\n\n    p.prev-next-title {\n      @include link-style-default;\n\n      font-weight: var(--pst-admonition-font-weight-heading);\n      font-size: 1.1em;\n    }\n\n    &:hover,\n    &:visited:hover {\n      p.prev-next-title {\n        @include link-style-hover;\n      }\n    }\n\n    .prev-next-info {\n      flex-direction: column;\n      margin: 0 0.5em;\n\n      .prev-next-subtitle {\n        text-transform: capitalize;\n      }\n    }\n\n    &.left-prev {\n      float: left;\n    }\n\n    &.right-next {\n      float: right;\n\n      div.prev-next-info {\n        text-align: right;\n      }\n    }\n  }\n}\n","/**\n * Search field\n **/\n.bd-search {\n  position: relative;\n  padding-left: 0.5rem;\n  gap: 0.5rem;\n  background-color: var(--pst-color-background);\n  border-radius: $admonition-border-radius;\n  border: 1px solid var(--pst-color-border);\n  color: var(--pst-color-text-base);\n\n  // Background should always be same color regardless of active or not\n  &:active {\n    background-color: var(--pst-color-background);\n    color: var(--pst-color-text-muted);\n  }\n\n  // Hoist the focus ring from the input field to its parent\n  &:focus-within {\n    box-shadow: $focus-ring-box-shadow;\n\n    input:focus {\n      box-shadow: none;\n    }\n  }\n\n  .icon {\n    position: absolute;\n    color: var(--pst-color-border);\n    left: 25px;\n  }\n\n  .fa-solid.fa-magnifying-glass {\n    position: absolute;\n    left: calc((2.5rem - 0.7em) / 2);\n    color: var(--pst-color-text-muted);\n  }\n\n  input.form-control {\n    background-color: var(--pst-color-background);\n    color: var(--pst-color-text-base);\n    border: none;\n\n    // Inner-text of the search bar\n    &::placeholder {\n      color: var(--pst-color-text-muted);\n    }\n\n    // Remove the little \"x\" that pops up when you start typing\n    &::-webkit-search-cancel-button,\n    &::-webkit-search-decoration {\n      appearance: none;\n    }\n\n    &:focus,\n    &:focus-visible {\n      color: var(--pst-color-text-muted);\n    }\n  }\n\n  // Shows off the keyboard shortcuts for the button\n  .search-button__kbd-shortcut {\n    display: none;\n    margin-inline-end: 0.5rem;\n    color: var(--pst-color-border);\n\n    @include media-breakpoint-up(md) {\n      display: flex;\n    }\n  }\n}\n\n/**\n * Search button - located in the navbar\n */\n.search-button i {\n  // Search link icon should be a bit bigger since it is separate from icon links\n  font-size: 1.3rem;\n}\n\n/**\n * The search modal <dialog>\n */\n#pst-search-dialog {\n  display: none;\n\n  &[open] {\n    display: flex;\n\n    // Center in middle of screen just underneath header\n    position: fixed;\n    z-index: $zindex-modal;\n    top: 30%;\n    left: 50%;\n    transform: translate(-50%, -50%);\n    right: 1rem;\n    margin-top: 0.5rem;\n    width: 90%;\n    max-width: 800px;\n    background-color: transparent;\n    padding: $focus-ring-width;\n    border: none;\n\n    &::backdrop {\n      background-color: black;\n      opacity: 0.5;\n    }\n\n    form.bd-search {\n      flex-grow: 1;\n\n      // Font and input text a bit bigger\n      svg,\n      input {\n        font-size: var(--pst-font-size-icon);\n      }\n    }\n  }\n}\n\n/**\n * The search button component that looks like a field.\n * Lives at components/search-button-field.html\n */\n.search-button-field {\n  $search-button-border-radius: 1.5em;\n\n  display: inline-flex;\n  align-items: center;\n  border: var(--pst-color-border) solid 1px;\n  border-radius: $search-button-border-radius;\n  color: var(--pst-color-text-muted);\n  padding: 0.5em;\n  background-color: var(--pst-color-surface);\n\n  &:hover {\n    box-shadow: 0 0 0 $focus-ring-width var(--pst-color-link-hover);\n  }\n\n  &:focus-visible {\n    border-radius: $search-button-border-radius;\n  }\n\n  // The keyboard shortcut text\n  .search-button__default-text {\n    font-size: var(--bs-nav-link-font-size);\n    font-weight: var(--bs-nav-link-font-weight);\n    margin-right: 0.5em;\n    margin-left: 0.5em;\n  }\n\n  .kbd-shortcut__modifier {\n    font-size: 0.75em;\n  }\n\n  // Ensures that all the text lines up in the middle\n  > * {\n    align-items: center;\n  }\n\n  // Only the icon should be visible on narrow screens\n  > :not(svg) {\n    display: none;\n\n    @include media-breakpoint-up(lg) {\n      display: flex;\n    }\n  }\n}\n","/**\n * The 'Hide Search Matches' button.\n * This only shows up when a person lands on a page after clicking a search result.\n * Clicking it removes the highlighting of the search term from the page.\n * We want it to behave like a button.\n */\ndiv#searchbox {\n  // Leave `#searchbox` rules empty so that it doesn't show at all when it is empty\n  p.highlight-link {\n    margin: 1rem 0;\n    width: fit-content;\n\n    // A bit more margin on wide screens to mimic article behavior\n    @include media-breakpoint-up($breakpoint-sidebar-secondary) {\n      margin-left: 2rem;\n    }\n\n    // Put outer shadow on this one so that we can darken the link w/ an inner shadow\n    @include box-shadow;\n\n    // Style the button to look like a Sphinx Design button\n    a {\n      border-radius: 0.25rem;\n      font-size: 1.25rem;\n      padding: 0.75rem;\n      background-color: var(--pst-color-primary);\n      color: var(--pst-color-primary-text);\n      text-decoration: none;\n\n      // The box shadow is inset so that it darkens the button on hover\n      transition: box-shadow 0.25s ease-out;\n\n      &:hover {\n        box-shadow: inset 0 0 50px 50px rgb(0 0 0 / 25%);\n      }\n\n      &::before {\n        content: var(--pst-icon-search-minus);\n        color: unset;\n        font: var(--fa-font-solid);\n        margin-right: 0.5rem;\n      }\n    }\n  }\n}\n","/**\n * Light/dark theme switcher\n */\n\n.theme-switch-button {\n  .theme-switch {\n    display: none;\n\n    &:active {\n      text-decoration: none;\n      color: var(--pst-color-link-hover);\n    }\n\n    .fa-lg {\n      aspect-ratio: 1 / 1;\n    }\n  }\n}\n\n@each $mode in auto, light, dark {\n  html[data-mode=\"#{$mode}\"]\n    .theme-switch-button\n    .theme-switch[data-mode=\"#{$mode}\"] {\n    display: inline; // inline needed for span height to be calculated using inherited font size and line height\n  }\n}\n","button.btn.version-switcher__button {\n  border-color: var(--pst-color-border);\n  color: var(--pst-color-text-base);\n\n  // Add a margin on narrow screens to avoid feeling cramped\n  margin-bottom: 1em;\n\n  @include media-breakpoint-up($breakpoint-sidebar-primary) {\n    margin-bottom: unset;\n  }\n\n  &:hover {\n    box-shadow: 0 0 0 $focus-ring-width var(--pst-color-secondary);\n    border-color: transparent;\n  }\n\n  &:active {\n    color: var(--pst-color-text-base);\n    border-color: var(--pst-color-border);\n  }\n\n  &:focus-visible {\n    border-color: transparent;\n  }\n}\n\n.version-switcher__menu {\n  border-color: var(--pst-color-border);\n  border-radius: var(--bs-dropdown-border-radius);\n\n  a.list-group-item {\n    background-color: var(--pst-color-on-background);\n    color: var(--pst-color-text-base);\n    padding: 0.75rem 1.25rem;\n\n    &:not(:last-child) {\n      border-bottom: 1px solid var(--pst-color-border);\n    }\n\n    &:hover {\n      @include link-style-hover;\n\n      background-color: var(--pst-color-surface);\n    }\n\n    &.active {\n      @include link-sidebar-current;\n\n      position: relative;\n      z-index: 1;\n\n      span::before {\n        content: \"\";\n        width: 100%;\n        height: 100%;\n        position: absolute;\n        z-index: -1;\n        left: 0;\n        top: 0;\n      }\n    }\n\n    &:focus-visible {\n      z-index: 10; // keep focus ring on top (prevent the hover background of the next dropdown item from covering the ring)\n    }\n  }\n}\n\n// Font behavior on mobile\nbutton.version-switcher__button,\n.version-switcher__menu {\n  font-size: 1.1em; // A bit smaller than other menu font\n  z-index: $zindex-modal; // higher than the sidebars\n\n  // Make sure it meets WCAG target size requirement no matter the version\n  // string displayed in the button\n  @include min-hit-area;\n\n  @include media-breakpoint-up($breakpoint-sidebar-primary) {\n    font-size: unset;\n  }\n}\n","/* Collapsing of the TOC sidebar while scrolling */\n\n/* Nav: hide second level (shown on .active) */\n\nnav.page-toc {\n  // A little extra space before the buttons\n  margin-bottom: 1rem;\n}\n\n.bd-toc .nav {\n  .nav {\n    display: none;\n\n    // So we can manually specify a level as visible in the config\n    &.visible {\n      display: block;\n    }\n  }\n\n  > .active > ul {\n    display: block;\n  }\n}\n\n// Each entry of the in-page TOC\n.toc-entry {\n  display: block;\n\n  a > code {\n    color: var(--pst-color-text-muted);\n  }\n\n  a.nav-link {\n    display: block;\n    padding: $toc-item-padding-y 0;\n\n    // Padding w/ negative margin so the top TOC item highlight overlaps w/ the TOC border\n    padding-left: 1rem;\n    margin-left: -1rem;\n\n    @include link-sidebar;\n\n    &.active {\n      @include link-sidebar-current;\n\n      background-color: transparent;\n\n      &:hover {\n        color: var(--pst-color-link-hover);\n      }\n    }\n\n    &:focus-visible {\n      border-radius: $focus-ring-radius;\n    }\n  }\n}\n","div.versionadded,\ndiv.versionchanged,\ndiv.deprecated {\n  vertical-align: middle;\n  margin: 1.5625em auto;\n  padding: 0 0.6rem;\n  overflow: hidden;\n\n  /* break-inside has replaced page-break-inside and is widely usable since 2019 */\n  page-break-inside: avoid;\n  break-inside: avoid;\n  border-left: 0.2rem solid;\n  border-color: var(--pst-color-info);\n  border-radius: $admonition-border-radius;\n  background-color: var(--pst-color-on-background);\n\n  @include box-shadow;\n\n  position: relative;\n\n  > p {\n    margin-bottom: 0.6rem;\n    margin-top: 0.6rem;\n  }\n}\n\ndiv.versionadded {\n  border-color: var(--pst-color-success);\n  background-color: var(--pst-color-success-bg);\n}\n\ndiv.versionchanged {\n  border-color: var(--pst-color-warning);\n  background-color: var(--pst-color-warning-bg);\n}\n\ndiv.deprecated {\n  border-color: var(--pst-color-danger);\n  background-color: var(--pst-color-danger-bg);\n}\n\nspan.versionmodified {\n  font-weight: 600;\n\n  &::before {\n    margin-right: 0.6rem;\n    color: var(--pst-color-info);\n    font: var(--fa-font-solid);\n    content: var(--pst-icon-versionmodified-default);\n  }\n}\n\nspan.versionmodified.added {\n  &::before {\n    color: var(--pst-color-success);\n    content: var(--pst-icon-versionmodified-added);\n  }\n}\n\nspan.versionmodified.changed {\n  &::before {\n    color: var(--pst-color-warning);\n    content: var(--pst-icon-versionmodified-changed);\n  }\n}\n\nspan.versionmodified.deprecated {\n  &::before {\n    color: var(--pst-color-danger);\n    content: var(--pst-icon-versionmodified-deprecated);\n  }\n}\n",".sidebar-indices-items {\n  display: flex;\n  flex-direction: column;\n  border-top: 1px solid var(--pst-color-border);\n\n  @include media-breakpoint-up($breakpoint-sidebar-primary) {\n    border-top: none;\n  }\n\n  .sidebar-indices-items__title {\n    font-weight: var(--pst-sidebar-header-font-weight);\n    font-size: var(--pst-sidebar-header-font-size);\n    color: var(--pst-color-text-base);\n    margin-bottom: 0.5rem;\n  }\n\n  ul.indices-link {\n    margin-right: -1rem;\n    list-style: none;\n    padding: 0;\n\n    li > a {\n      display: block;\n      padding: 0.25rem 0;\n      color: var(--pst-color-text-muted);\n\n      &:hover {\n        color: var(--pst-color-primary);\n        text-decoration: none;\n        background-color: transparent;\n      }\n    }\n  }\n}\n","/**\n * Admonitions and blocks of styled content.\n * Admonitions CSS originally inspired by https://squidfunk.github.io/mkdocs-material/getting-started/\n */\n$admonition-border-radius: 0.25rem;\n$admonition-left-border-width: 0.2rem;\n\ndiv.admonition,\n.admonition {\n  margin: 1.5625em auto;\n  padding: 0 0.6rem 0.8rem;\n  overflow: hidden;\n\n  /* break-inside has replaced page-break-inside and is widely usable since 2019 */\n  page-break-inside: avoid;\n  break-inside: avoid;\n  border-left: $admonition-left-border-width solid;\n  border-color: var(--pst-color-info);\n  border-radius: $admonition-border-radius;\n  background-color: var(--pst-color-on-background);\n\n  @include box-shadow;\n\n  // Last item should have no spacing since we'll control that w/ padding\n  *:last-child {\n    margin-bottom: 0;\n  }\n\n  // Items after the title should be indented\n  p.admonition-title ~ * {\n    margin-left: 1.4rem;\n    margin-right: 1.4rem;\n  }\n\n  // Lists need to have left margin so they don't spill into it\n  > ol,\n  > ul {\n    margin-left: 1em;\n  }\n\n  // Defaults for all admonitions\n  > .admonition-title {\n    margin: 0 -0.6rem;\n    padding: 0.4rem 0.6rem 0.4rem 2rem;\n    font-weight: var(--pst-admonition-font-weight-heading);\n    position: relative;\n\n    @include legacy-backdrop-placeholder;\n\n    background-color: var(--pst-color-info-bg);\n\n    // now that we use solid colors we want the title on top\n    z-index: 1;\n\n    &::after {\n      position: absolute;\n      left: 0.5rem;\n      width: 1rem;\n      height: 1rem;\n      color: var(--pst-color-info);\n      font: var(--fa-font-solid);\n      line-height: inherit;\n      content: var(--pst-icon-admonition-default);\n      opacity: 1;\n    }\n\n    // Next element after title needs some extra upper-space\n    + * {\n      margin-top: 0.4em;\n    }\n  }\n\n  &.attention {\n    border-color: var(--pst-color-attention);\n\n    > .admonition-title {\n      background-color: var(--pst-color-attention-bg);\n\n      &::after {\n        color: var(--pst-color-attention);\n        content: var(--pst-icon-admonition-attention);\n      }\n    }\n  }\n\n  &.caution {\n    border-color: var(--pst-color-warning);\n\n    > .admonition-title {\n      background-color: var(--pst-color-warning-bg);\n\n      &::after {\n        color: var(--pst-color-warning);\n        content: var(--pst-icon-admonition-caution);\n      }\n    }\n  }\n\n  &.warning {\n    border-color: var(--pst-color-warning);\n\n    > .admonition-title {\n      background-color: var(--pst-color-warning-bg);\n\n      &::after {\n        color: var(--pst-color-warning);\n        content: var(--pst-icon-admonition-warning);\n      }\n    }\n  }\n\n  &.danger {\n    border-color: var(--pst-color-danger);\n\n    > .admonition-title {\n      background-color: var(--pst-color-danger-bg);\n\n      &::after {\n        color: var(--pst-color-danger);\n        content: var(--pst-icon-admonition-danger);\n      }\n    }\n  }\n\n  &.error {\n    border-color: var(--pst-color-danger);\n\n    > .admonition-title {\n      background-color: var(--pst-color-danger-bg);\n\n      &::after {\n        color: var(--pst-color-danger);\n        content: var(--pst-icon-admonition-error);\n      }\n    }\n  }\n\n  &.hint {\n    border-color: var(--pst-color-success);\n\n    > .admonition-title {\n      background-color: var(--pst-color-success-bg);\n\n      &::after {\n        color: var(--pst-color-success);\n        content: var(--pst-icon-admonition-hint);\n      }\n    }\n  }\n\n  &.tip {\n    border-color: var(--pst-color-success);\n\n    > .admonition-title {\n      background-color: var(--pst-color-success-bg);\n\n      &::after {\n        color: var(--pst-color-success);\n        content: var(--pst-icon-admonition-tip);\n      }\n    }\n  }\n\n  &.important {\n    border-color: var(--pst-color-attention);\n\n    > .admonition-title {\n      background-color: var(--pst-color-attention-bg);\n\n      &::after {\n        color: var(--pst-color-attention);\n        content: var(--pst-icon-admonition-important);\n      }\n    }\n  }\n\n  &.note {\n    border-color: var(--pst-color-info);\n\n    > .admonition-title {\n      background-color: var(--pst-color-info-bg);\n\n      &::after {\n        color: var(--pst-color-info);\n        content: var(--pst-icon-admonition-note);\n      }\n    }\n  }\n\n  &.seealso {\n    border-color: var(--pst-color-success);\n\n    > .admonition-title {\n      background-color: var(--pst-color-success-bg);\n\n      &::after {\n        color: var(--pst-color-success);\n        content: var(--pst-icon-admonition-seealso);\n      }\n    }\n  }\n\n  &.admonition-todo {\n    border-color: var(--pst-color-secondary);\n\n    > .admonition-title {\n      background-color: var(--pst-color-secondary-bg);\n\n      &::after {\n        color: var(--pst-color-secondary);\n        content: var(--pst-icon-admonition-todo);\n      }\n    }\n  }\n\n  /**\n   * Special-case for a `sidebar` class that makes the admonition float to\n   * the right like the {sidebar} directive.\n   */\n  &.sidebar {\n    max-width: 40%;\n    float: right;\n    clear: both;\n    margin-left: 0.5rem;\n    margin-top: 0;\n\n    // Undo the .sidebar directive border\n    border-width: 0 0 0 $admonition-left-border-width;\n\n    // TODO: these semantic-color-names border-color rules might no longer be\n    //       needed when we drop support for Sphinx 4 / docutils 0.17\n    &.attention,\n    &.important {\n      border-color: var(--pst-color-attention);\n    }\n\n    &.caution,\n    &.warning {\n      border-color: var(--pst-color-warning);\n    }\n\n    &.danger,\n    &.error {\n      border-color: var(--pst-color-danger);\n    }\n\n    &.hint,\n    &.tip,\n    &.seealso {\n      border-color: var(--pst-color-success);\n    }\n\n    &.note,\n    &.todo {\n      border-color: var(--pst-color-info);\n    }\n\n    // No inner margin since we have less horizontal space w/ the sidebar\n    p.admonition-title ~ * {\n      margin-left: 0;\n      margin-right: 0;\n    }\n  }\n}\n\n/**************************************************************\n * Similar content blocks that are not technically admonitions.\n */\n\n/**\n * Topics and the {contents} directive\n */\n// Docutils <= 0.17\ndiv.topic,\ndiv.topic.contents,\n// Docutils >= 0.18\nnav.contents,\naside.topic {\n  display: flex;\n  flex-direction: column;\n  background-color: var(--pst-color-surface);\n  border-color: var(--pst-color-border);\n  border-radius: $admonition-border-radius;\n  padding: 1rem 1.25rem;\n\n  @include box-shadow;\n\n  .topic-title {\n    margin: 0 0 0.5rem;\n  }\n\n  // Over-ride text color to ensure enough contrast\n  p {\n    color: var(--pst-color-on-surface) !important;\n  }\n\n  // Over-ride large default padding\n  ul.simple {\n    padding-left: 1rem;\n\n    ul {\n      // So that sub-lists will have a bit less padding\n      padding-left: 2em;\n    }\n  }\n}\n\n/**\n * Sidebar directive\n */\naside.sidebar {\n  border: 1px solid var(--pst-color-border);\n  background-color: var(--pst-color-surface);\n  border-radius: $admonition-border-radius;\n\n  // to match the admonition-styled sidebars:\n  margin-left: 0.5rem;\n  padding: 0;\n\n  > *:last-child {\n    padding-bottom: 1rem;\n  }\n\n  p.sidebar-title {\n    position: relative;\n    margin-bottom: 0;\n    padding-top: 0.5rem;\n    padding-bottom: 0.5rem;\n    border-bottom: 1px solid var(--pst-color-border);\n    font-family: var(--pst-font-family-heading);\n    font-weight: var(--pst-admonition-font-weight-heading);\n  }\n\n  // Add margin to the first non-`.sidebar-title` item\n  > *:not(.sidebar-title):first-child,\n  > p.sidebar-title + * {\n    margin-top: 1rem;\n  }\n\n  > * {\n    padding-left: 1rem;\n    padding-right: 1rem;\n  }\n}\n\n/**\n * Rubrics look kind of like section headers\n */\np.rubric {\n  display: flex;\n  flex-direction: column;\n}\n\n/**\n * Seealso is kind of like a vertically-collapsed admonition\n */\n.seealso dd {\n  margin-top: 0;\n  margin-bottom: 0;\n}\n","/**\n * Miscellaneous color functions and mixins\n**/\n\n@use \"sass:list\";\n@use \"sass:map\";\n@use \"sass:meta\";\n@use \"sass:math\";\n@use \"sass:string\";\n\n// We must add ::before pseudo-element to some theme components (such as admonitions)\n// because users were instructed to customize the background color this way.\n@mixin legacy-backdrop-placeholder {\n  &::before {\n    content: \"\";\n    width: 100%;\n    height: 100%;\n    position: absolute;\n    left: 0;\n    top: 0;\n    z-index: -1;\n\n    // So that hovering over the text within background is still possible.\n    // Otherwise the background overlays the text and you cannot click or select easily.\n    // ref: https://developer.mozilla.org/en-US/docs/Web/CSS/pointer-events\n    pointer-events: none;\n  }\n}\n\n/**\n* Function to get items from nested maps\n*/\n// @param {Map} $map - Map\n// @param {Arglist} $keys - Keys to fetc\n// @return {*}\n@function map-deep-get($map, $keys...) {\n  @each $key in $keys {\n    $map: map.get($map, $key);\n  }\n\n  @return $map;\n}\n\n/**\n  * Function to check if the color needs converting to a \"color\" type\n  * if it is a string we cannot use other color manipulation functions\n  * It is used to create the sphinx-design colours as these are often interpolated\n  */\n// @param {String/Color} $color - Color definition from map\n// @return {Color} - Color type (in hex)\n@function check-color($color) {\n  @if meta.type-of($color) == string {\n    $color: from-hex($color);\n  }\n\n  @return $color;\n}\n\n/**\n  * Function to convert the string representation of a color to a color type (hex)\n  */\n// @param {String} $string - String representation of a color\n@function from-hex($string) {\n  $string-lower: string.to-lower-case($string);\n  $r: \"\";\n  $g: \"\";\n  $b: \"\";\n  $hex: \"0\" \"1\" \"2\" \"3\" \"4\" \"5\" \"6\" \"7\" \"8\" \"9\" \"a\" \"b\" \"c\" \"d\" \"e\" \"f\";\n  $length: string.length($string);\n  $max: if($length == 4, 1, 2);\n\n  // Check for length accuracy\n  @if $length != 4 and $length != 7 {\n    @return $string;\n  }\n\n  // Loop from the second character (omitting #)\n  @for $i from 2 through $length {\n    $c: string.slice($string-lower, $i, $i);\n\n    // If wrong character, return\n    @if not list.index($hex, $c) {\n      @return $string;\n    }\n\n    @if string.length($r) < $max {\n      $r: $r + $c;\n    } @else if string.length($g) < $max {\n      $g: $g + $c;\n    } @else if string.length($b) < $max {\n      $b: $b + $c;\n    }\n  }\n\n  @if $length == 4 {\n    $r: $r + $r;\n    $g: $g + $g;\n    $b: $b + $b;\n  }\n\n  @return rgb(hex-to-dec($r), hex-to-dec($g), hex-to-dec($b));\n}\n\n@function hex-to-dec($string) {\n  $hex: \"0\" \"1\" \"2\" \"3\" \"4\" \"5\" \"6\" \"7\" \"8\" \"9\" \"a\" \"b\" \"c\" \"d\" \"e\" \"f\";\n  $string: string.to-lower-case($string);\n  $length: string.length($string);\n  $dec: 0;\n\n  @for $i from 1 through $length {\n    $factor: 1 + (15 * ($length - $i));\n    $index: list.index($hex, string.slice($string, $i, $i));\n    $dec: $dec + $factor * ($index - 1);\n  }\n\n  @return $dec;\n}\n","// Style API docs from sphinx' autodoc / autosummary\n\n/*******************************************************************************\n* Styling for field lists\n*/\n\n/* grey highlighting of 'parameter' and 'returns' field */\ntable.field-list {\n  border-collapse: separate;\n  border-spacing: 10px;\n  margin-left: 1px;\n\n  th.field-name {\n    padding: 1px 8px 1px 5px;\n    white-space: nowrap;\n    background-color: var(--pst-color-surface);\n  }\n\n  /* italic font for parameter types */\n  td.field-body {\n    p {\n      font-style: italic;\n\n      > strong {\n        font-style: normal;\n      }\n    }\n\n    /* reduced space around parameter description */\n    blockquote {\n      border-left: none;\n      margin: 0 0 0.3em;\n      padding-left: 30px;\n    }\n  }\n}\n\n/*******************************************************************************\n* Styling for autosummary tables\n*/\n\n.table.autosummary {\n  // The first column (with the signature) should not wrap\n  td:first-child {\n    white-space: nowrap;\n  }\n}\n\n/* overriding basic.css to use our own monospace font */\n.sig {\n  font-family: var(--pst-font-family-monospace);\n}\n\n/* C++ specific styling - overriding the basic.css to avoid custom colors */\n\n.sig-inline.c-texpr,\n.sig-inline.cpp-texpr {\n  font-family: unset;\n}\n\n.sig.c .k,\n.sig.c .kt,\n.sig.cpp .k,\n.sig.cpp .kt {\n  color: var(--pst-color-text-base);\n}\n\n.sig.c .m,\n.sig.cpp .m {\n  color: var(--pst-color-text-base);\n}\n\n.sig.c .s,\n.sig.c .sc,\n.sig.cpp .s,\n.sig.cpp .sc {\n  color: var(--pst-color-text-base);\n}\n\n// addition\n\n// .sig.c .sig-name .n,\n// .sig.cpp .sig-name .n {\n//   color: var(--pst-color-inline-code);\n// }\n\n.sig-name {\n  color: var(--pst-color-inline-code);\n}\n\n.sig-param .o,\n.sig-param .default_value {\n  color: var(--pst-color-text-muted);\n  font-weight: normal;\n}\n\n// change target color for dark theme\ndt:target,\nspan.highlighted {\n  background-color: var(--pst-color-target);\n}\n\n.viewcode-back {\n  font-family: var(--pst-font-family-base);\n}\n\n.viewcode-block:target {\n  border-top: 1px solid var(--pst-color-border);\n  border-bottom: 1px solid var(--pst-color-border);\n  position: relative;\n  background-color: var(--pst-color-target);\n}\n\ndl > dt > a:has(.viewcode-link) {\n  // Sphinx applies a `float:right` rule to the .viewcode-line span, which\n  // exposes a browser glitch in the focus ring. It seems the browser creates\n  // two separate boxes, an empty box where the anchor element gets laid out and\n  // then another box around the anchor's contents that have been floated right.\n  // Firefox draws the focus ring around the empty anchor element box. Chrome\n  // draws two focus rings: one around the empty anchor and one around the\n  // floated-right element. To fix the glitch, we apply the float rule on the\n  // parent rather than the child.\n  float: right;\n\n  .viewcode-link {\n    float: none;\n  }\n}\n\n/*******************************************************************************\n* Styling for autosummary titles like \"parameters\" and \"returns\"\n*/\n\n// the API selector\n// from https://github.com/pradyunsg/furo/blob/main/src/furo/assets/styles/content/_api.sass#L6)\ndl[class]:not(.option-list, .field-list, .footnote, .glossary, .simple) {\n  // increase margin bottom after the dl elements\n  margin-bottom: 3rem;\n\n  dd {\n    margin-left: 2rem;\n\n    // Fix until this will be solved to Sphinx https://github.com/sphinx-doc/sphinx/issues/10815\n    & > dl.simple > dt {\n      display: flex;\n    }\n  }\n\n  dl.field-list {\n    display: grid;\n    grid-template-columns: unset;\n  }\n\n  dt.field-odd,\n  dt.field-even {\n    margin-top: 0.2rem;\n    margin-bottom: 0.1rem;\n    background-color: var(--pst-color-surface);\n  }\n}\n","/**\n * Code block styling\n * Note that we inherit a lot of styling from Bootstrap so not many rules here.\n */\n\n// General code block behavior\n// Unset bootstrap behavior\ndiv[class*=\"highlight-\"],\ndiv.highlight,\ndiv.literal-block-wrapper {\n  display: flex;\n  flex-direction: column;\n  width: unset;\n  border-radius: $admonition-border-radius;\n  break-inside: avoid;\n}\n\n// Code blocks with captions\n// There's a wrapper when the code block has a title\ndiv.literal-block-wrapper {\n  border: 1px solid var(--pst-color-border);\n  border-radius: $admonition-border-radius;\n\n  // This is where the title goes\n  div.code-block-caption {\n    margin: 0;\n    border-bottom: 1px solid var(--pst-color-border);\n    padding: 0.5rem;\n    font-size: 1rem;\n    font-weight: var(--pst-font-weight-caption);\n\n    a.headerlink {\n      font-size: inherit;\n    }\n  }\n\n  // Remove the upper border radius since we want it to connect with the title\n  // Remove the box shadow so the wrapper gets the shadow instead\n  div[class*=\"highlight-\"] {\n    margin: 0;\n    border-radius: 0;\n\n    pre {\n      border: none;\n      box-shadow: none;\n    }\n  }\n}\n\n/**\n * In-line code\n */\ncode.literal {\n  padding: 0.1rem 0.25rem;\n  background-color: var(--pst-color-surface);\n  border: 1px solid var(--pst-color-border);\n  border-radius: 0.25rem;\n}\n\na > code {\n  color: var(--pst-color-inline-code-links);\n\n  &:hover {\n    color: var(--pst-color-link-hover);\n  }\n}\n\n// Minimum opacity needed for linenos to be WCAG AA conformant\nspan.linenos {\n  opacity: 0.8 !important;\n}\n","figure > a,\nfigure > a > img,\nfigure > img,\nfigure > video {\n  display: block;\n  margin-left: auto;\n  margin-right: auto;\n}\n\nfigure {\n  a.headerlink {\n    // So that header link doesn't push caption to be off-center.\n    position: absolute;\n    font-size: inherit;\n  }\n\n  // Default headerlink hover doesn't trigger on figures\n  &:hover a.headerlink {\n    visibility: visible;\n  }\n\n  figcaption {\n    font-family: var(--pst-font-family-heading);\n    font-weight: var(--pst-font-weight-caption);\n    color: var(--pst-color-text-muted);\n    margin-left: auto;\n    margin-right: auto;\n    margin-top: 0.3rem;\n    text-align: center;\n\n    & > p:last-child {\n      // Don't add extra margin to already existing figure bottom margin\n      margin-bottom: 0;\n    }\n\n    p {\n      text-align: start;\n      display: inline-block;\n    }\n\n    table.table {\n      width: fit-content;\n      margin-left: auto;\n      margin-right: auto;\n    }\n  }\n}\n","// For consistency, add bracket around footnotes/citations which are\n// cited more than once. E.g. [Newton](1,2) instead of Newton(1,2)\ndt.label > span.brackets:not(:only-child)::before {\n  content: \"[\";\n}\n\ndt.label > span.brackets:not(:only-child)::after {\n  content: \"]\";\n}\n\n// Make footnote as a superscript\na.footnote-reference {\n  vertical-align: super;\n  font-size: small;\n}\n\n// Docutils 0.18 uses an `aside.footnote` container with different internal structure\naside.footnote {\n  margin-bottom: 0.5rem;\n\n  &:last-child {\n    margin-bottom: 1rem;\n  }\n\n  span.label,\n  span.backrefs {\n    font-weight: bold;\n  }\n\n  &:target {\n    background-color: var(--pst-color-target);\n  }\n}\n","/**\n * Hacky fixes that don't fit cleanly into other sections\n */\n\n// Ensure user highlighting/selecting behaves properly\n// From https://stackoverflow.com/a/34372191\ntable.highlighttable td.linenos,\nspan.linenos,\ndiv.doctest > div.highlight span.gp {\n  /* gp: Generic.Prompt */\n  user-select: none;\n}\n","// Override bootstrap by restoring the basic theme default.\ndd {\n  margin-top: 3px;\n  margin-bottom: 10px;\n  margin-left: 30px;\n}\n\nol,\nul {\n  padding-inline-start: 2rem;\n\n  li > p:first-child {\n    margin-bottom: 0.25rem;\n    margin-top: 0.25rem;\n  }\n}\n","// GitHub blockquote style\nblockquote {\n  padding: 1em;\n  color: var(--pst-color-text-muted);\n  border-left: 0.25em solid var(--pst-color-blockquote-notch);\n  border-radius: $admonition-border-radius;\n  position: relative;\n\n  p {\n    color: var(--pst-color-text-base);\n  }\n\n  // remove padding from included line-block to avoid duplication\n  .line-block {\n    margin: 0;\n  }\n\n  // remove margin bottom for the last p\n  p:last-child {\n    margin-bottom: 0;\n  }\n\n  @include legacy-backdrop-placeholder;\n\n  background-color: var(--pst-color-surface);\n\n  // Ensure there is enough contrast against the background\n  a {\n    color: var(--pst-color-inline-code-links);\n  }\n\n  // hack to make the text in the blockquote selectable\n  &::before {\n    z-index: -1;\n  }\n}\n","/**\n * Span-level styling within content\n */\n\nspan.guilabel {\n  border: 1px solid var(--pst-color-info);\n  font-size: 80%;\n  font-weight: 700;\n  border-radius: 4px;\n  padding: 2.4px 6px;\n  margin: auto 2px;\n  position: relative;\n\n  @include legacy-backdrop-placeholder;\n\n  background-color: var(--pst-color-info-bg);\n}\n\na.reference.download::before {\n  content: var(--pst-icon-download);\n  font: var(--fa-font-solid);\n  font-size: 0.8em;\n  padding: 0 0.25em;\n  color: var(--pst-color-text-muted);\n}\n","/**\n * Tables\n */\n\ntable {\n  // default to table-center\n  margin-left: auto;\n  margin-right: auto;\n\n  &.table-right {\n    margin-right: 0;\n  }\n\n  &.table-left {\n    margin-left: 0;\n  }\n}\n\n// customize table caption from bootstrap\n// to display them on top and centered\ntable caption {\n  text-align: center;\n  caption-side: top;\n  color: var(--pst-color-text-muted);\n}\n\n// MyST Markdown tables use these classes to control alignment\nth,\ntd {\n  &.text-left {\n    text-align: left;\n  }\n\n  &.text-right {\n    text-align: right;\n  }\n\n  &.text-center {\n    text-align: center;\n  }\n}\n\n// override bootstrap table colors\n.table {\n  @include table-colors;\n\n  --bs-table-bg: transparent; // background\n  --bs-table-color: var(\n    --pst-color-text-base\n  ); // ensure text and bullets are visible\n}\n\n.pst-scrollable-table-container {\n  // Put a horizontal scrollbar just below tables that are too wide to fit\n  // within the main column\n  overflow-x: auto;\n}\n","/**\n * Style the toctree component in pages (avoid modifying the navbars)\n */\n.toctree-wrapper {\n  p.caption {\n    font-size: 1.5em;\n    margin-bottom: 0;\n  }\n\n  & > ul {\n    padding-left: 0;\n  }\n\n  li[class^=\"toctree-l\"] {\n    list-style: none;\n    margin-bottom: 0.2em;\n\n    & > a {\n      list-style: none;\n      font-size: 1.1em;\n    }\n\n    & > ul {\n      list-style: none;\n      padding-inline-start: 1.5em;\n    }\n  }\n\n  // slightly bigger font for l1\n  .toctree-l1 > a {\n    font-size: 1.3em;\n  }\n}\n\ndiv.topic.contents,  // Docutils <= 0.17\nnav.contents  // Docutils >= 0.18\n{\n  // Style similarly to toctree\n  ul.simple {\n    list-style: none;\n    padding-left: 0;\n  }\n}\n","/**\n * Mathematics via MathJax.\n *\n * This is designed for MathJax v3\n * ref: https://www.sphinx-doc.org/en/master/usage/extensions/math.html#module-sphinx.ext.mathjax\n */\n\n// Applies to all math elements\nspan.math,\ndiv.math {\n  align-items: center;\n  display: flex;\n  max-width: 100%;\n\n  // This will be over-ridden for the y-direction and divs\n  overflow: hidden;\n}\n\n// Inline-only\nspan.math {\n  display: inline-flex;\n}\n\n// Block-level only\ndiv.math {\n  gap: 0.5em;\n\n  // So that the eqno shows up after the equation\n  flex-direction: row-reverse;\n\n  // The equation number / link\n  span.eqno a.headerlink {\n    position: relative;\n    font-size: 1em;\n  }\n\n  // The math container\n  mjx-container {\n    flex-grow: 1;\n    padding-bottom: 0.2rem;\n    overflow: auto;\n\n    // Set height to 0 so that it does not cause scrollbars to appear\n    // ref: https://github.com/mathjax/MathJax/issues/2521\n    mjx-assistive-mml {\n      height: 0;\n    }\n  }\n}\n","/**\n * ABlog\n * ref: https://ablog.readthedocs.io/\n */\n\n/**\n * Sidebar template components\n */\n.ablog-sidebar-item {\n  h2,\n  h3 {\n    font-size: var(--pst-sidebar-header-font-size);\n\n    // Remove unnecessary vertical whitespace\n    margin-top: 0.5rem;\n\n    // The headers are all links, but this makes them hard to parse\n    // So we change the colors to make them look like headers\n    a {\n      color: var(--pst-color-text-base);\n    }\n  }\n\n  ul {\n    // No bullet points for the primary sidebar items\n    list-style: none;\n    padding-left: 0;\n\n    // Otherwise a scrollbar randomly shows up\n    overflow-y: hidden;\n\n    // List of recent post items\n    display: flex;\n    flex-direction: column;\n    gap: 0.5em;\n    margin-bottom: 0;\n\n    // The ablog cloud should move horizontally\n    &.ablog-cloud {\n      flex-flow: row wrap;\n      gap: 0.5rem;\n\n      // Vertical-align tag clouds\n      li {\n        // Center the tag cloud items\n        display: flex;\n        align-items: center;\n      }\n    }\n  }\n}\n\n/**\n * Previous / next buttons at the bottom\n */\n.ablog__prev-next {\n  font-size: 1.2em;\n  display: flex;\n  padding: 1rem 0;\n\n  // The bottom previous / next arrows\n  > span {\n    // To ensure that the whole thing fits on one line even if there are long titles\n    display: flex;\n    max-width: 45%;\n\n    // Links within each span have the collection of icon + text\n    a {\n      display: flex;\n      align-items: center;\n      margin-left: auto;\n      gap: 1rem;\n      line-height: 1.5rem;\n\n      i::before {\n        color: var(--pst-color-text-base);\n      }\n    }\n  }\n\n  // The first span is for the previous page and aligns to the left\n  span.ablog__prev {\n    i.fa-arrow-circle-left::before {\n      content: var(--pst-icon-angle-left);\n    }\n  }\n\n  // The second span is just an empty space so we remove it because we're\n  // positioning with flex\n  span.ablog__spacer {\n    display: none;\n  }\n\n  // The third span is aligned to the right\n  span.ablog__next {\n    margin-left: auto;\n    text-align: right;\n\n    i.fa-arrow-circle-right::before {\n      content: var(--pst-icon-angle-right);\n    }\n  }\n}\n\n/**\n * {postlist} directive and posts page\n */\n.ablog__collection,\n.postlist {\n  padding-left: 0;\n\n  .ablog-post {\n    list-style: none;\n\n    // Post metadata tags (author, links ,etc) should be a bit smaller\n    .ablog-archive {\n      display: flex;\n      flex-flow: row wrap;\n      gap: 1rem;\n      list-style: none;\n      font-size: 0.75rem;\n      padding-left: 0;\n    }\n\n    // Title line should be a bit bigger and bold to stand out\n    .ablog-post-title {\n      margin-top: 0;\n      font-size: 1.25rem;\n\n      a {\n        font-weight: bold;\n      }\n    }\n\n    // Read more button should be a bit bigger\n    .ablog-post-expand {\n      margin-bottom: 0.5rem;\n    }\n  }\n}\n","/**\n * Special cases for Bootstrap functionality\n */\n\n// Bootstrap adds margin to their general container class. However, sphinx/docutils\n// can also generate output with the container class, but in those cases we should\n// not add the margin from bootstrap. Same for max-width.\n.docutils.container {\n  padding-left: unset;\n  padding-right: unset;\n  margin-left: unset;\n  margin-right: unset;\n  max-width: unset;\n  width: unset;\n}\n\n.btn {\n  --bs-btn-focus-box-shadow: #{$btn-focus-box-shadow};\n}\n","/**\n * Sphinx Copybutton\n * ref: https://sphinx-copybutton.readthedocs.io/\n */\n\ndiv.highlight button.copybtn {\n  // Nicer spacing\n  display: flex;\n  align-items: center;\n  justify-content: center;\n\n  // Don't over-ride the success color\n  &:not(.success) {\n    color: var(--pst-color-muted);\n  }\n\n  border: none;\n  background-color: var(--pst-color-surface);\n\n  &:hover {\n    &:not(.success) {\n      color: var(--pst-color-text);\n      background-color: var(--pst-color-shadow);\n    }\n  }\n\n  // Tooltip styling\n  &.o-tooltip--left::after {\n    color: var(--pst-color-text);\n    background-color: var(--pst-color-surface);\n  }\n\n  &:focus {\n    // For keyboard users, make the copy button visible when focussed.\n    opacity: 1;\n  }\n\n  &:focus-visible {\n    outline: $focus-ring-outline;\n  }\n}\n\ndiv.highlight:has(button.copybtn) {\n  // Make sure the code block has enough height for the copy button.\n  // Sphinx-copybutton sets 0.3em top offset plus 1.7em height:\n  // https://github.com/executablebooks/sphinx-copybutton/blob/master/sphinx_copybutton/_static/copybutton.css\n  min-height: 2em;\n}\n","// adapt ethical ad to the theme\n#ethical-ad-placement {\n  .ethical-sidebar a,\n  .ethical-sidebar a:visited,\n  .ethical-sidebar a:hover,\n  .ethical-sidebar a:active,\n  .ethical-footer a,\n  .ethical-footer a:visited,\n  .ethical-footer a:hover,\n  .ethical-footer a:active {\n    color: var(--pst-color-text-base);\n  }\n\n  .ethical-sidebar,\n  .ethical-footer {\n    background-color: var(--pst-color-background);\n    border: 1px solid var(--pst-color-border);\n    border-radius: 5px;\n    color: var(--pst-color-text-base);\n    font-size: 14px;\n    line-height: 20px;\n  }\n}\n","/**\n * Styles for various Sphinx execution libraries.\n * For now, where these define output sections, we simply revert their background\n * to be a \"light theme\" background. This ensures that inputs/outputs behave similarly,\n * because the CSS is often controlled by each package.\n * In the future, we might add dark theme support for specific packages.\n */\n\n/******************************************************************************\n * Jupyter Sphinx\n */\n\n.bd-content div.jupyter_container {\n  // We don't want borders around the whole container, just around code cells\n  border: none;\n  background-color: unset;\n  box-shadow: none;\n\n  // Code cells should have the same style as our other code objects\n  div.output,\n  div.highlight {\n    border-radius: 0.25rem;\n  }\n\n  div.highlight {\n    background-color: var(--pst-color-surface);\n  }\n\n  // Ensure the style is the same as our code cells. Jupyter Sphinx makes it tiny.\n  .cell_input,\n  .cell_output {\n    border-radius: 0.25rem;\n\n    pre {\n      padding: 1rem;\n    }\n  }\n}\n","/* Styles for graphviz generated output from Sphinx */\n\n/* Style the inheritance diagram such that it has a dark mode */\nhtml[data-theme=\"dark\"] div.graphviz > object.inheritance {\n  filter: brightness(0.8) invert(0.82) contrast(1.2);\n  color-scheme: normal;\n}\n","/**\n * Special-cases for packages in the PyData ecosystem\n */\n\n// xarray output display in bootstrap\n.xr-wrap[hidden] {\n  display: block !important;\n}\n\n// ipywidgets\n.jp-OutputArea-output.lm-Widget {\n  // override overflow:hidden rule from Lumino (.lm-Widget) to allow scrolling\n  overflow: auto;\n}\n","/*******************************************************************************\n* master color map. Only the colors that actually differ between light and dark\n* themes are specified separately.\n*\n* To see the full list of colors see https://www.figma.com/file/rUrrHGhUBBIAAjQ82x6pz9/PyData-Design-system---proposal-for-implementation-(2)?node-id=1234%3A765&t=ifcFT1JtnrSshGfi-1\n*/\n\n@use \"sass:map\";\n@use \"sass:meta\";\n@use \"sass:string\";\n\n/**\n* Function to get items from nested maps\n*/\n// @param {Map} $map - Map\n// @param {Arglist} $keys - Keys to fetch\n// @return {*}\n@function map-deep-get($map, $keys...) {\n  @each $key in $keys {\n    $map: map.get($map, $key);\n  }\n\n  @return $map;\n}\n\n/* Assign base colors for the PyData theme */\n$color-palette: (\n  // Primary color\n  \"teal\":\n    (\n      \"50\": #f4fbfc,\n      \"100\": #e9f6f8,\n      \"200\": #d0ecf1,\n      \"300\": #abdde6,\n      \"400\": #3fb1c5,\n      \"500\": #0a7d91,\n      \"600\": #085d6c,\n      \"700\": #064752,\n      \"800\": #042c33,\n      \"900\": #021b1f,\n    ),\n  // Secondary color\n  \"violet\":\n    (\n      \"50\": #f4eefb,\n      \"100\": #e0c7ff,\n      \"200\": #d5b4fd,\n      \"300\": #b780ff,\n      \"400\": #9c5ffd,\n      \"500\": #8045e5,\n      \"600\": #6432bd,\n      \"700\": #4b258f,\n      \"800\": #341a61,\n      \"900\": #1e0e39,\n    ),\n  // Neutrals\n  \"gray\":\n    (\n      \"50\": #f9f9fa,\n      \"100\": #f3f4f5,\n      \"200\": #e5e7ea,\n      \"300\": #d1d5da,\n      \"400\": #9ca4af,\n      \"500\": #677384,\n      \"600\": #48566b,\n      \"700\": #29313d,\n      \"800\": #222832,\n      \"900\": #14181e,\n    ),\n  // Accent color\n  \"pink\":\n    (\n      \"50\": #fcf8fd,\n      \"100\": #fcf0fa,\n      \"200\": #f8dff5,\n      \"300\": #f3c7ee,\n      \"400\": #e47fd7,\n      \"500\": #c132af,\n      \"600\": #912583,\n      \"700\": #6e1c64,\n      \"800\": #46123f,\n      \"900\": #2b0b27,\n    ),\n  \"foundation\": (\n    \"white\": #ffffff,\n    // gray-900\n    \"black\": #14181e,\n  )\n);\n\n:root {\n  // Add theme colours to the html root element\n  @each $group-color, $color in $color-palette {\n    @each $color-name, $definition in $color {\n      --pst-#{$group-color}-#{$color-name}: #{$definition};\n    }\n  }\n}\n\n// Static SCSS variables used thoroughout the theme\n// Minimum contrast ratio used for the theme.\n// Acceptable values for WCAG 2.0 are 3, 4.5 and 7.\n// See https://www.w3.org/TR/WCAG20/#visual-audio-contrast-contrast\n// 4.5 - is for text that is 14pt or less\n$min-contrast-ratio-4: 4.5;\n\n// 3 is for text that is 18pt or bold, or for non-text elements\n$min-contrast-ratio-3: 3;\n\n// Customize the light and dark text colors for use in our color contrast function.\n$foundation-black: #14181e;\n$foundation-white: #fff;\n\n// This is a custom - calculated  color between gray 100 and 200 - to reduce\n// the contrast ratio (avoid a jarring effect)\n$base-light-text: #ced6dd;\n\n// used in sphinx_design - gray 100\n$foundation-light-gray: #f3f4f5;\n\n// used in sphinx_design - gray 700\n$foundation-muted-gray: #29313d;\n\n// used in sphinx_design - gray 800\n$foundation-dark-gray: #222832;\n$pst-semantic-colors: (\n  \"primary\": (\n    \"light\": #{map-deep-get($color-palette, \"teal\", \"500\")},\n    \"bg-light\": #{map-deep-get($color-palette, \"teal\", \"200\")},\n    \"dark\": #{map-deep-get($color-palette, \"teal\", \"400\")},\n    \"bg-dark\": #{map-deep-get($color-palette, \"teal\", \"800\")},\n  ),\n  \"secondary\": (\n    \"light\": #{map-deep-get($color-palette, \"violet\", \"500\")},\n    \"bg-light\": #{map-deep-get($color-palette, \"violet\", \"100\")},\n    \"dark\": #{map-deep-get($color-palette, \"violet\", \"400\")},\n    \"bg-dark\": #{map-deep-get($color-palette, \"violet\", \"800\")},\n  ),\n  \"accent\": (\n    \"light\": #{map-deep-get($color-palette, \"pink\", \"500\")},\n    \"bg-light\": #{map-deep-get($color-palette, \"pink\", \"200\")},\n    \"dark\": #{map-deep-get($color-palette, \"pink\", \"400\")},\n    \"bg-dark\": #{map-deep-get($color-palette, \"pink\", \"800\")},\n  ),\n  \"info\": (\n    \"light\": #276be9,\n    \"bg-light\": #dce7fc,\n    \"dark\": #79a3f2,\n    \"bg-dark\": #06245d,\n  ),\n  \"warning\": (\n    \"light\": #f66a0a,\n    \"bg-light\": #f8e3d0,\n    \"dark\": #ff9245,\n    \"bg-dark\": #652a02,\n  ),\n  \"success\": (\n    \"light\": #00843f,\n    \"bg-light\": #d6ece1,\n    \"dark\": #5fb488,\n    \"bg-dark\": #002f17,\n  ),\n  // This is based on the warning color\n  \"attention\":\n    (\n      \"light\": var(--pst-color-warning),\n      \"bg-light\": var(--pst-color-warning-bg),\n      \"dark\": var(--pst-color-warning),\n      \"bg-dark\": var(--pst-color-warning-bg),\n    ),\n  \"danger\": (\n    \"light\": #d72d47,\n    \"bg-light\": #f9e1e4,\n    \"dark\": #e78894,\n    \"bg-dark\": #4e111b,\n  ),\n  \"text-base\": (\n    \"light\": #{map-deep-get($color-palette, \"gray\", \"800\")},\n    \"dark\": $base-light-text,\n  ),\n  \"text-muted\": (\n    \"light\": #{map-deep-get($color-palette, \"gray\", \"600\")},\n    \"dark\": #{map-deep-get($color-palette, \"gray\", \"400\")},\n  ),\n  \"shadow\": (\n    \"light\": rgba(0, 0, 0, 0.1),\n    \"dark\": rgba(0, 0, 0, 0.2),\n  ),\n  \"border\": (\n    \"light\": #{map-deep-get($color-palette, \"gray\", \"300\")},\n    \"dark\": #{map-deep-get($color-palette, \"gray\", \"600\")},\n  ),\n  \"border-muted\": (\n    \"light\": rgba(23, 23, 26, 0.2),\n    \"dark\": #{map-deep-get($color-palette, \"gray\", \"700\")},\n  ),\n  \"blockquote-notch\": (\n    // These colors have a contrast ratio > 3.0 against both the background and\n    // surface colors that the notch is sandwiched between\n    \"light\": #{map-deep-get($color-palette, \"gray\", \"500\")},\n    \"dark\": #{map-deep-get($color-palette, \"gray\", \"400\")},\n  ),\n  \"inline-code\": (\n    \"light\": #{map-deep-get($color-palette, \"pink\", \"600\")},\n    \"dark\": #{map-deep-get($color-palette, \"pink\", \"300\")},\n  ),\n  \"inline-code-links\": (\n    // need to make sure there is enough contrast against the code bg\n    \"light\": #{map-deep-get($color-palette, \"teal\", \"600\")},\n    // keep primary color for dark mode\n    \"dark\": #{map-deep-get($color-palette, \"teal\", \"400\")},\n  ),\n  \"target\": (\n    \"light\": #f3cf95,\n    \"dark\": #675c04,\n  ),\n  \"table\": (\n    \"light\": #{map-deep-get($color-palette, \"foundation\", \"black\")},\n    \"dark\": #{map-deep-get($color-palette, \"foundation\", \"white\")},\n  ),\n  \"table-row-hover\": (\n    \"bg-light\": #{map-deep-get($color-palette, \"violet\", \"200\")},\n    \"bg-dark\": #{map-deep-get($color-palette, \"violet\", \"700\")},\n  ),\n  \"table-inner-border\": (\n    \"light\": #{map-deep-get($color-palette, \"gray\", \"200\")},\n    \"dark\": #364150,\n  ),\n  // DEPTH COLORS - you can see the elevation colours and shades\n  // in the Figma file https://www.figma.com/file/rUrrHGhUBBIAAjQ82x6pz9/PyData-Design-system---proposal-for-implementation-(2)?node-id=1492%3A922&t=sQeQZehkOzposYEg-1\n  // background: color of the canvas / the furthest back layer\n  \"background\":\n    (\n      \"light\": #{map-deep-get($color-palette, \"foundation\", \"white\")},\n      \"dark\": #{map-deep-get($color-palette, \"foundation\", \"black\")},\n    ),\n  // on-background: provides slight contrast against background\n  // (by use of shadows in light theme)\n  \"on-background\":\n    (\n      \"light\": #{map-deep-get($color-palette, \"foundation\", \"white\")},\n      \"dark\": #{map-deep-get($color-palette, \"gray\", \"800\")},\n    ),\n  \"surface\": (\n    \"light\": #{map-deep-get($color-palette, \"gray\", \"100\")},\n    \"dark\": #{map-deep-get($color-palette, \"gray\", \"700\")},\n  ),\n  // on_surface: object on top of surface object (without shadows)\n  \"on-surface\":\n    (\n      \"light\": #{map-deep-get($color-palette, \"gray\", \"800\")},\n      \"dark\": $foundation-light-gray,\n    ),\n);\n\n/*******************************************************************************\n* write the color rules for each theme (light/dark)\n*/\n\n/* NOTE:\n * Mixins enable us to reuse the same definitions for the different modes\n * https://sass-lang.com/documentation/at-rules/mixin\n * #{something} inserts a variable into a CSS selector or property name\n * https://sass-lang.com/documentation/interpolation\n */\n@mixin theme-colors($mode) {\n  // check if this color is defined differently for light/dark\n  @each $col-name, $definition in $pst-semantic-colors {\n    @if meta.type-of($definition) == map {\n      @each $key, $val in $definition {\n        @if string.index($key, $mode) {\n          // since now we define the bg colours in the semantic colours and not\n          // by changing opacity, we need to check if the key contains bg and the\n          // correct mode (light/dark)\n          @if string.index($key, \"bg\") {\n            --pst-color-#{$col-name}-bg: #{$val};\n          } @else {\n            --pst-color-#{$col-name}: #{$val};\n          }\n        }\n      }\n    } @else {\n      --pst-color-#{$col-name}: #{$definition};\n    }\n  }\n\n  // assign the \"duplicate\" colors (ones that just reference other variables)\n  & {\n    --pst-color-heading: var(--pst-color-text-base);\n    --pst-color-link: var(--pst-color-primary);\n    --pst-color-link-hover: var(--pst-color-secondary);\n    --pst-color-table-outer-border: var(--pst-color-surface);\n    --pst-color-table-heading-bg: var(--pst-color-surface);\n    --pst-color-table-row-zebra-high-bg: var(--pst-color-on-background);\n    --pst-color-table-row-zebra-low-bg: var(--pst-color-surface);\n  }\n\n  // adapt to light/dark-specific content\n  @if $mode == \"light\" {\n    .only-dark,\n    .only-dark ~ figcaption {\n      display: none !important;\n    }\n  } @else {\n    .only-light,\n    .only-light ~ figcaption {\n      display: none !important;\n    }\n\n    /* Adjust images in dark mode (unless they have class .only-dark or\n     * .dark-light, in which case assume they're already optimized for dark\n     * mode).\n     */\n    img:not(.only-dark, .dark-light) {\n      filter: brightness(0.8) contrast(1.2);\n    }\n\n    /* Give images a light background in dark mode in case they have\n    *  transparency and black text (unless they have class .only-dark or .dark-light, in\n    *  which case assume they're already optimized for dark mode).\n    */\n    .bd-content img:not(.only-dark, .dark-light) {\n      background-color: rgb(255 255 255);\n      border-radius: 0.25rem;\n    }\n\n    // MathJax SVG outputs should be filled to same color as text.\n    .MathJax_SVG * {\n      fill: var(--pst-color-text-base);\n    }\n  }\n}\n\n/* Defaults to light mode if data-theme is not set */\nhtml:not([data-theme]) {\n  @include theme-colors(\"light\");\n}\n\n/* NOTE: @each {...} is like a for-loop\n * https://sass-lang.com/documentation/at-rules/control/each\n */\n@each $mode in (light, dark) {\n  html[data-theme=\"#{$mode}\"] {\n    @include theme-colors($mode);\n\n    color-scheme: $mode;\n  }\n}\n\n// assign classes too, for runtime use of theme colors\n@each $col-name, $definition in $pst-semantic-colors {\n  .pst-color-#{$col-name} {\n    color: var(--pst-color-#{$col-name});\n  }\n}\n","/*******************************************************************************\n * Special-cases for the sphinx-design library, mainly to make it compatible\n * with the dark/light themes of pydata-sphinx-theme.\n *\n * NOTE: sphinx-design uses !important quite liberally, so here we must do the\n * same for our overrides to have any effect.\n */\n@use \"../variables/color\" as pst-color;\n@use \"sass:color\";\n@use \"sass:map\";\n@use \"sass:meta\";\n@use \"sass:string\";\n\n/*******************************************************************************\n  * Color and variables\n  *\n  * This is a list of the semantic color names from sphinx-design (we only\n  * need to override variables that sphinx-design has actually defined).\n  * https://github.com/executablebooks/sphinx-design/blob/9226a12a/style/_colors.scss#L31-L43\n  */\n$sd-semantic-color-names: (\n  \"primary\",\n  \"secondary\",\n  \"success\",\n  \"info\",\n  \"warning\",\n  \"danger\",\n  \"light\",\n  \"muted\",\n  \"dark\",\n  \"black\",\n  \"white\"\n);\n\n/**\n  * Here we create some extra --pst-color-* variables and use\n  * them to override the value of the corresponding sphinx-design variables.\n  * This is easier than re-writing the sphinx-design rules. Even easier would be\n  * directly assigning our values to the --sd-color-* variables, but then our\n  * downstream users couldn't override *our* colors and have it affect buttons\n  * and badges.\n  *\n  * First, define the extra keys needed to cover the full range of semantic\n  * color names used in sphinx-design, then merge them with the names we\n  * already define for our own needs.\n  * see https://sphinx-design.readthedocs.io/en/latest/css_variables.html\n  */\n$extra-semantic-colors: (\n  \"white\": $foundation-white,\n  \"light\": (\n    light: $foundation-light-gray,\n    bg-light: color.scale($foundation-light-gray, $lightness: 30%),\n    dark: $foundation-light-gray,\n    bg-dark: color.scale($foundation-light-gray, $lightness: -30%),\n  ),\n  \"muted\": (\n    light: $foundation-muted-gray,\n    bg-light: color.scale($foundation-muted-gray, $lightness: 30%),\n    dark: $foundation-light-gray,\n    bg-dark: color.scale($foundation-muted-gray, $lightness: -30%),\n  ),\n  \"dark\": $foundation-dark-gray,\n  \"black\": $foundation-black,\n);\n$all-colors: map.merge($pst-semantic-colors, $extra-semantic-colors);\n\n@mixin create-sd-colors($value, $name) {\n  // define the pst variables, so that downstream user overrides will work\n  --pst-color-#{$name}: #{$value};\n\n  // we are now using a11y-combination to calculate the text color - this is based\n  // on the WCAG color contrast guidelines\n  --pst-color-#{$name}-text: #{a11y-combination($value)};\n\n  // highlight is used for hover effects on buttons, here we make some fluid scaling\n  // to avoid jarring effects -- we create a local variable that we can access\n  // later to calculate the text color\n  $highlight-color: color.scale($value, $lightness: -15%, $saturation: 5%);\n\n  --pst-color-#{$name}-highlight: #{$highlight-color};\n\n  // override the sphinx-design variables\n  --sd-color-#{$name}: var(--pst-color-#{$name});\n  --sd-color-#{$name}-text: var(--pst-color-#{$name}-text);\n  --sd-color-#{$name}-highlight: var(--pst-color-#{$name}-highlight);\n\n  // calculate the text color for the highlight color\n  --pst-color-#{$name}-highlight-text: #{a11y-combination($highlight-color)};\n}\n\n// Now we override the --sd-color-* variables.\n@each $mode in (light, dark) {\n  @debug \"Creating color variables for mode: #{$mode}\";\n  html[data-theme=\"#{$mode}\"] {\n    // check if this color is defined differently for light/dark\n    @each $name in $sd-semantic-color-names {\n      $definition: map.get($all-colors, $name);\n\n      @if meta.type-of($definition) == map {\n        @each $key, $value in $definition {\n          @if string.index($key, $mode) {\n            // since now we define the bg colours in the semantic colours and not\n            // by changing opacity, we need to check if the key contains bg and the\n            // correct mode (light/dark)\n            @if string.index($key, \"bg\") {\n              --sd-color-#{$name}-bg: #{$value};\n\n              // create local variable -- needed to calculate the text color\n              $value: check-color($value);\n\n              --sd-color-#{$name}-bg-text: #{a11y-combination($value)};\n            } @else {\n              $value: check-color($value);\n\n              @include create-sd-colors($value, $name);\n            }\n          }\n        }\n      } @else {\n        $value: map.get($all-colors, $name);\n\n        @include create-sd-colors($value, $name);\n      }\n    }\n  }\n}\n\n// Make sure the color border variables are set using our variables\n@each $mode in (light, dark) {\n  html[data-theme=\"#{$mode}\"] {\n    --sd-color-card-border: var(--pst-color-border);\n  }\n}\n\n/*******************************************************************************\n  * shadows\n  */\nhtml[data-theme=\"light\"] {\n  .sd-shadow-xs,\n  .sd-shadow-sm,\n  .sd-shadow-md,\n  .sd-shadow-lg {\n    @include box-shadow;\n  }\n}\n\n/*******************************************************************************\n  * cards\n  */\n\n.bd-content .sd-card {\n  border: 1px solid var(--pst-color-border);\n\n  .sd-card-header {\n    background-color: var(--pst-color-on-background);\n    border-bottom: 1px solid var(--pst-color-border);\n  }\n\n  .sd-card-footer {\n    background-color: var(--pst-color-on-background);\n    border-top: 1px solid var(--pst-color-border);\n  }\n\n  .sd-card-body {\n    background-color: var(--pst-color-on-background);\n  }\n\n  // Focus ring for link-cards\n  .sd-stretched-link:focus-visible {\n    // Don't put the focus ring on the <a> element (it has zero height in Sphinx Design cards)\n    outline: none;\n\n    // Put the focus ring on the <a> element's ::after pseudo-element\n    &::after {\n      outline: $focus-ring-outline;\n      border-radius: 0.25rem; // copied from Sphinx Design CSS for .sd-card\n    }\n  }\n\n  &.sd-card-hover:hover {\n    border-color: var(--pst-color-link-hover);\n  }\n}\n\n/*******************************************************************************\n  * tabs\n  */\n\n.bd-content .sd-tab-set {\n  > input {\n    // Active tab label\n    &:checked + label {\n      border-style: solid solid none;\n      border-color: var(--pst-color-primary) var(--pst-color-primary)\n        transparent; // top LR bottom\n\n      border-width: 0.125rem 0.125rem 0;\n      border-radius: 0.125rem 0.125rem 0 0;\n      background-color: var(--pst-color-on-background);\n      transform: translateY(0.125rem);\n      color: var(--pst-color-primary);\n    }\n\n    &:focus-visible + label {\n      border: 0.125rem solid var(--pst-color-accent);\n      border-radius: 0.125rem;\n      background-color: var(--pst-color-accent-bg);\n      color: var(--pst-color-on-surface);\n    }\n\n    // Hover label\n    &:not(:checked, :focus-visible) + label:hover {\n      border-color: transparent;\n      color: var(--pst-color-secondary);\n      text-decoration-line: underline;\n      text-decoration-thickness: $link-hover-decoration-thickness;\n    }\n  }\n\n  // Tab label\n  > label {\n    color: var(--pst-color-on-surface);\n    border: 0.125rem solid transparent;\n    border-radius: 0.125rem 0.125rem 0 0;\n    background-color: var(--pst-color-surface);\n    padding: 0 0.75em;\n    margin-inline-end: 0.25rem;\n    line-height: 1.95;\n\n    html[data-theme=\"dark\"] & {\n      background-color: var(--pst-color-on-background);\n    }\n  }\n\n  // panel\n  .sd-tab-content {\n    border: 0.125rem solid var(--pst-color-primary);\n    border-radius: 0.1875rem;\n    box-shadow: unset;\n    padding: 0.625rem;\n    background-color: var(--pst-color-on-background);\n  }\n}\n\n/*******************************************************************************\n* Dropdowns\n*/\n\ndetails.sd-dropdown {\n  // Remove all borders to over-ride SD behavior, and we'll add our own later\n  border: 0 !important;\n\n  // Drop shadow should behave same as admonitions\n  @include box-shadow;\n\n  // Header is where the \"clickable\" box goes\n  summary.sd-card-header {\n    border: 0 !important;\n    display: flex;\n    align-items: center;\n    position: relative; // So background color works\n    font-weight: 600;\n    padding-top: 0.5rem;\n    padding-bottom: 0.5rem;\n\n    // Set a variable that we can re-use for colors later\n    // We must set this in the current and content sibling container\n    // so that it is defined in both places\n    --pst-sd-dropdown-color: var(--pst-gray-500);\n    --pst-sd-dropdown-bg-color: var(--pst-color-surface);\n\n    & + div.sd-summary-content {\n      border: 0;\n\n      --pst-sd-dropdown-color: var(--sd-color-card-border);\n    }\n\n    @each $name in $sd-semantic-color-names {\n      &.sd-bg-#{$name} {\n        --pst-sd-dropdown-color: var(--sd-color-#{$name});\n        --pst-sd-dropdown-bg-color: var(--sd-color-#{$name}-bg);\n\n        // Otherwise it won't be defined in the sibling element\n        & + div.sd-summary-content {\n          --pst-sd-dropdown-color: var(--sd-color-#{$name});\n          --pst-sd-dropdown-bg-color: var(--sd-color-#{$name}-bg);\n        }\n      }\n      &.sd-bg-text-#{$name} {\n        // Use the WCAG conformant text color\n        color: var(--sd-color-#{$name}-bg-text) !important;\n      }\n    }\n\n    @include legacy-backdrop-placeholder;\n\n    background-color: var(--pst-sd-dropdown-bg-color) !important;\n\n    // Add a left border with the same structure as our admonitions\n    border-left: 0.2rem solid var(--pst-sd-dropdown-color) !important;\n\n    // stylelint-disable-next-line no-duplicate-selectors\n    & + div.sd-summary-content {\n      border-left: 0.2rem solid var(--pst-sd-dropdown-color) !important;\n      border-bottom-left-radius: calc(0.25rem - 1px);\n      background-color: var(--pst-color-on-background);\n    }\n\n    span.sd-summary-icon {\n      display: inline-flex;\n      align-items: center;\n      color: var(--pst-sd-dropdown-color) !important;\n\n      svg {\n        opacity: 1;\n      }\n    }\n\n    // Positioning of the caret\n    .sd-summary-up,\n    .sd-summary-down {\n      top: 0.7rem;\n    }\n\n    @include hover-darken-lighten;\n\n    // Focus ring\n    &:focus:focus-visible {\n      outline: $focus-ring-outline;\n      outline-offset: $focus-ring-offset;\n      border-radius: $focus-ring-width;\n    }\n  }\n\n  &[open] summary.sd-card-header:focus:focus-visible {\n    border-bottom-left-radius: 0;\n    border-bottom-right-radius: 0;\n  }\n}\n\n/*******************************************************************************\n* Buttons (which in Sphinx Design are actually links that look like buttons)\n* ref: https://sphinx-design.readthedocs.io/en/pydata-theme/badges_buttons.html#buttons\n*/\nhtml {\n  .sd-btn {\n    min-width: 2.25rem;\n    padding: 0.3125rem 0.75rem 0.4375rem; // 5px 12px 7px\n\n    &:hover {\n      @include link-style-hover; // override Sphinx Design\n\n      text-decoration-thickness: 1px;\n    }\n  }\n\n  @each $name in $sd-semantic-color-names {\n    .sd-btn-#{$name} {\n      &:hover {\n        color: var(--pst-color-#{$name}-highlight-text) !important;\n      }\n    }\n  }\n\n  @each $name in $sd-semantic-color-names {\n    .sd-btn-#{$name},\n    .sd-btn-outline-#{$name} {\n      &:focus-visible {\n        outline: var(--sd-color-#{$name}) solid $focus-ring-width;\n        outline-offset: $focus-ring-offset;\n        border-radius: $focus-ring-width;\n\n        // Override Sphinx Design's use of -highlight colors. The -highlight\n        // colors are 15% darker, so this would create the effect of darkening\n        // the button when focused but we just want the button to have a focus\n        // ring of the same (non-highlight) color.\n        background-color: var(--sd-color-#{$name}) !important;\n        border-color: var(--sd-color-#{$name}) !important;\n      }\n    }\n  }\n}\n","/**\n * Sphinx togglebutton\n *\n * The rules in this style sheet are meant to tweak the\n * [sphinx-togglebutton](https://sphinx-togglebutton.readthedocs.io/en/latest/index.html)\n * extension so that it matches the look and feel of this theme.\n */\n\n.bd-content {\n  @mixin chevron-down {\n    .toggle-chevron-right {\n      transform: rotate(90deg); // point chevron down\n      transition: none; // match non-animated behavior of other chevrons on site\n    }\n  }\n\n  @mixin chevron-up {\n    .toggle-chevron-right {\n      transform: rotate(-90deg); // point chevron up\n      transition: none; // match non-animated behavior of other chevrons on site\n    }\n  }\n\n  // Apply this mixin to the element that will be hovered. These styles are\n  // intended to match what sphinx-design does for its dropdown admonitions.\n  @mixin icon-hover-effects {\n    &:hover .tb-icon {\n      opacity: 1;\n      scale: 1.1;\n    }\n\n    .tb-icon {\n      opacity: 0.6;\n    }\n  }\n\n  // Collapsible admonition, implemented as <div> + <button>\n  .dropdown.admonition.toggle {\n    // The title is visible when the admonition is collapsed and expanded\n    .admonition-title {\n      @include icon-hover-effects;\n      @include hover-darken-lighten;\n\n      &:hover {\n        box-shadow: none;\n      }\n    }\n\n    button.toggle-button {\n      color: inherit;\n\n      // When disclosure widget is closed\n      &.toggle-button-hidden {\n        @include chevron-down;\n      }\n\n      // When open\n      @include chevron-up;\n    }\n\n    // Focus ring\n    // ----------\n    // Sphinx-togglebutton makes the entire admonition header clickable, but\n    // only the button within the header is focusable. But we want the entire\n    // header and not just the button inside the header to be surrounded by a\n    // a focus ring.\n    &:has(:focus-visible) {\n      /* Override Sphinx Toggle Button. Make the overflow visible, otherwise the\n      focus ring is hidden. */\n      overflow: visible;\n\n      /*\n      Why not just do the following?\n\n      ```\n      .admonition-title {\n        outline: $focus-ring-outline;\n      }\n      ```\n\n      Why use ::before? If we put the focus ring on the ::before pseudo-element,\n      we can reposition the focus ring by repositioning the pseudo-element.\n\n      Why reposition? The left edge of the admonition title box does not align\n      with the left edge of the overall admonition box. There is a left border\n      that belongs to the overall box. The border is outside of the admonition\n      title, which means it is also outside of a focus ring around the title. We\n      can make the focus ring bigger, with `outline-offset`, but this will\n      result in a ring that looks off-centered. So we have to pull the ring left\n      and stretch it right. That's what the pseudo-element allows us to do.\n\n      Note: we do not have to do this for collapsible admonitions made with\n      Sphinx Design (as opposed to sphinx-togglebutton).\n      */\n      .admonition-title::before {\n        content: \"\";\n\n        // pull the focus ring left and expand it right to be perfectly centered\n        // between the left border and the right edge of the admonition title\n        left: -$admonition-left-border-width;\n        width: calc(100% + $admonition-left-border-width);\n        height: 100%;\n        outline: $focus-ring-outline;\n        outline-offset: $focus-ring-offset;\n        border-radius: $focus-ring-width;\n      }\n\n      // When expanded, sharpen the bottom left and right corners of the focus ring\n      &:not(.toggle-hidden) .admonition-title::before {\n        border-bottom-left-radius: 0;\n        border-bottom-right-radius: 0;\n      }\n    }\n  }\n\n  // Collapsible component, implemented as <details> + <summary>\n  details.toggle-details {\n    // Over-ride border color to re-use our primary color\n    summary {\n      border-left: 3px solid var(--pst-color-primary);\n\n      @include chevron-down;\n      @include icon-hover-effects;\n      @include hover-darken-lighten;\n\n      &:hover {\n        box-shadow: none;\n      }\n\n      &:focus-visible {\n        outline-offset: $focus-ring-offset;\n      }\n    }\n\n    // When expanded, sharpen the bottom left and right corners of the focus ring\n    &[open] {\n      @include chevron-up;\n\n      :focus-visible {\n        border-bottom-left-radius: 0;\n        border-bottom-right-radius: 0;\n      }\n    }\n  }\n}\n","/**\n * Styles for various Sphinx execution libraries to display pre-executed notebooks.\n * For now, where these define output sections, we simply revert their background\n * to be a \"light theme\" background. This ensures that inputs/outputs behave similarly,\n * because the CSS is often controlled by each package.\n * In the future, we might add dark theme support for specific packages.\n */\n\n/*******************************************************************************\n * nbsphinx\n */\nhtml div.rendered_html,\n// NBsphinx ipywidgets output selector\nhtml .jp-RenderedHTMLCommon {\n  // Add some margin around the element box for the focus ring. Otherwise the\n  // focus ring gets clipped because the containing elements have `overflow:\n  // hidden` applied to them (via the `.lm-Widget` selector)\n  margin: $focus-ring-width;\n\n  table {\n    table-layout: auto;\n  }\n}\n\n.bd-content .nboutput {\n  .output_area {\n    &.rendered_html,\n    .jp-RenderedHTMLCommon {\n      // pandas\n      table.dataframe {\n        @include table-colors;\n\n        tbody {\n          tr:hover {\n            background-color: var(--pst-color-table-row-hover-bg);\n          }\n        }\n      }\n    }\n\n    // Dark theme special-cases\n    html[data-theme=\"dark\"] & {\n      &.rendered_html:not(:has(table.dataframe)),\n      // ipywidgets\n      .widget-subarea {\n        @include cell-output-background;\n      }\n\n      &.stderr {\n        background-color: var(--pst-color-danger);\n      }\n    }\n  }\n}\n\n// Add extra padding to the final item in an nbsphinx container\ndiv.nblast.container {\n  margin-bottom: 1rem;\n}\n\n// Override nbsphinx's colors for notebook cell prompts because they do not have\n// sufficient contrast. Colors chosen from accessible-pygments\n// a11y-high-contrast-{light,dark} themes.\n\n// Notebook cell input line number. Replace nbsphinx's low contrast blue with\n// higher contrast blues.\n.nbinput.container .prompt pre {\n  html[data-theme=\"light\"] & {\n    // Copied from accessible-pygments [a11y-high-contrast-light](https://github.com/Quansight-Labs/accessible-pygments/tree/main/a11y_pygments/a11y_high_contrast_light)\n    color: #005b82;\n  }\n\n  html[data-theme=\"dark\"] & {\n    // Copied from accessible-pygments [a11y-high-contrast-dark](https://github.com/Quansight-Labs/accessible-pygments/tree/main/a11y_pygments/a11y_high_contrast_dark)\n    color: #00e0e0;\n  }\n}\n\n// Notebook cell output line number. Replace nbsphinx's low contrast red with\n// higher contrast red / orange.\n.nboutput.container .prompt pre {\n  html[data-theme=\"light\"] & {\n    // Copied from accessible-pygments [a11y-high-contrast-light](https://github.com/Quansight-Labs/accessible-pygments/tree/main/a11y_pygments/a11y_high_contrast_light)\n    color: #a12236;\n  }\n\n  html[data-theme=\"dark\"] & {\n    // Copied from accessible-pygments [a11y-high-contrast-dark](https://github.com/Quansight-Labs/accessible-pygments/tree/main/a11y_pygments/a11y_high_contrast_dark)\n    color: #ffa07a;\n  }\n}\n\n/*******************************************************************************\n * myst NB\n */\n\ndiv.cell_output .output {\n  max-width: 100%;\n  overflow-x: auto;\n}\n\n.bd-content div.cell_output {\n  // pandas\n  table.dataframe {\n    @include table-colors;\n\n    tbody {\n      tr:hover {\n        background-color: var(--pst-color-table-row-hover-bg);\n      }\n    }\n  }\n\n  html[data-theme=\"dark\"] & {\n    img,\n    .text_html:not(:has(table.dataframe)),\n    // ipywidgets\n    .widget-subarea {\n      @include cell-output-background;\n    }\n  }\n}\n\n// Prevent tables from scrunching together\n.bd-content {\n  div.cell_input {\n    display: flex;\n    flex-direction: column;\n    justify-content: stretch;\n  }\n\n  div.cell_input,\n  div.output {\n    border-radius: $admonition-border-radius;\n  }\n\n  div.output {\n    table {\n      table-layout: auto;\n    }\n  }\n}\n","/**\n * style for the various mapping libs based on leaflet (folium, geemap, ipyleaflet)\n * mainly ensure the good display of the maps in both themes and avoid the customization\n * of the tiles\n */\n\n/**\n * avoid border override from pydata-sphinx-theme\n * minimal selctor to get the priority\n */\nhtml[data-theme=\"dark\"] .bd-content img.leaflet-tile.leaflet-tile-loaded {\n  border-radius: 0;\n  padding: 0;\n}\n","/**\n * /search.html page special-cases\n */\n\n.bd-search-container {\n  /*******************************************\n  * Search results\n  */\n  // Whitespace\n  div#search-results {\n    > h2 {\n      font-size: var(--pst-font-size-icon);\n      margin-top: 1rem;\n    }\n\n    p.search-summary {\n      color: var(--pst-color-text-muted);\n    }\n  }\n\n  ul.search {\n    margin: 0;\n    list-style: none;\n\n    li {\n      background-image: none;\n      padding: 1rem 0;\n      margin: 1rem 0;\n      border-top: 1px solid var(--pst-color-text-muted);\n\n      // First link is the page title, it should be a bit bigger\n      > a {\n        font-size: 1.2em;\n      }\n\n      div.context,\n      p.context {\n        color: var(--pst-color-text-base);\n        margin: 0.5em 0 0;\n\n        // Add a # before page section titles to make it clear they are sections\n        a::before {\n          content: \"#\";\n          padding-right: 0.2em;\n          color: var(--pst-color-text-muted);\n        }\n      }\n    }\n  }\n}\n"],"names":[],"sourceRoot":""}
\ No newline at end of file
diff --git a/_static/styles/theme.css b/_static/styles/theme.css
new file mode 100644
index 000000000..4519dd912
--- /dev/null
+++ b/_static/styles/theme.css
@@ -0,0 +1,2 @@
+/* Provided by Sphinx's 'basic' theme, and included in the final set of assets */
+@import "../basic.css";
diff --git a/_static/vendor/fontawesome/webfonts/fa-brands-400.ttf b/_static/vendor/fontawesome/webfonts/fa-brands-400.ttf
new file mode 100644
index 0000000000000000000000000000000000000000..1fbb1f7c32d46f5dcb89a50e10d00878ed43f1a1
GIT binary patch
literal 209128
zcmd4437p(TwfI~0>wWK@-g~y5?wRRKW+s`Qt&@<Y2?->S03!P$K}1ous6h}xhfPsY
zLD>YwxPV@bUM`}dB6sw1k*i$gDqJ_z0WqL*H8Ycth&uDWr>Z-dAnJYZ|K8{Gy61G&
zul721>eQ)Ir+%ZfQYxq>luH$ldF9gKGvE5Ew<x6-5xV8<3(nfQ<M+caDizqRRC2}Z
z&VTD`UvqEt+e-CpSI!@noqNt%ujvas{%+Fl0LITHz<r$iQ2fsT-REAg?b5%veLtYQ
z>y+}JfBr>hpXL0^r|(g!eTe6+7o2tJR%f026~y~VU%c?F3(lDijaHQ_^<$;n?OQK;
z^R|0ioQ=wr+ox1^m!hpMk4JU9W&G`D##Zi8A%|>AJzf3wwM}<^cK%Dw4f;`@{mPLX
zeE8Lm&VNZM=St%L;M~v%H<9w=v`UE6>OMwQOrYe`2c>fBXY>uqL+~d3T4IlAxO-Gy
zT|-{wahaU*qvQL%N*#LAD>ti2)ipnF+d$ink~AjaS`9W-k;e(#`uwb(#9voIdDav2
zb3Dg+COQ84{GWh1!vAbs310}3N775ax{3>S8h&yAXv%D+pOm<l<gcrusb9h+zP`Y%
zQ>TO*n4fcOJLXOK6Z3zfuCQbCJ0|aBBQGv-orHOa7k|N-nBQaCHc6QhKVi~(BuxCc
zJ|NsECyYOC(%85WlV)Q6MVo$pUR+a7Ge7-z)ciiDuDXefHRGFU8s(-4m;S~Z&r)wI
zWjm&>Nz)Dqld`>lU(z{ODsAdl^M9dyizf-6IN)!V*L0-}Q_e2Z2$0^CHK8KLKh@|1
zWtemm%I(xu4BY<3q=THmyQ`T`${?MW@iA>8&TZ1wO}PQ$Z1~njU8YW3hl!&e(zi6r
zYo^DI>E{}4m?X_E<tNOxwdvNCk9-{q>O0UUsi#i<30y$(I(6fQoBd{7=1sj5DuL@K
zK0-PN$eI35nEEWONZyx%YP4s<xC4!LL~)Z1zmNK5yrh12vn}#$+GO)7==VhvC%CLz
zXy8@y1?Y!1c_xXQFl8BBNLv@A6WkO54<P+9xDuCi1~i<EXUW^LzZ}Ywwh&&Vo&q@e
zBk`ey`AyuUZ3A%u6F0TLF3L9JDdoyMk-R>q4vk2E7=M9A9rA4HY_i#|`4>%jlPbqt
zWUffKX@j&WMwrYgOYah2Cw{`D`Lm>#ye57Zb4;0jEbO<L|EA1yC0~PY+>gJmGA6yH
zcRyvOOgpwW(oHqW;#p74Qx|SsMNM7e2aJD`u{)Ob-Oxxsq1sG+O65sY1RU6_;UA6g
zB5vN)W%))<?`qKDq$#u6en}_1N9vI<$Ix@JQ4eVi9SfZcOu`H5q@BQj1LX<snIGg$
z6PGm4#6fF+()U=rnKszGrVP?I%VLdM;1`;@zcK&oW;~Lno=MUOOqhE8`e}JK@wUH=
zorxz6?XYx!mwu2RoYqZV;d9M+JGS!N;5Xzq{FZ0d8u1%f=9ScIX>EdWsq<WC93C`f
z*>+7zISsID0(oSvS^OM?za}34j@#(7^eyVFm%dVV9iA<D>WnFFW2_`DbI`MmgEqkb
zq@B04-|&hCZftqVwzrWtN4e1ke%c!KOC3VjhWDE`m^|RUuCjnj;;BdGQr5;>`sF#_
zETd7k#nY|^PHwgHj&w4|(}E`{kF>G|w3>d(+Nr3(@?Mz_Y2r05V`A1iS$}1$r5(2X
zNzz;V3y&3=v}?kg@{z<1xPWf!C#y?|MAD5LG)1|6zo4Wi3G}HaRhQ~f{c5>7Rh^|?
zt<F}jQRk@Fs@JJ=)$7$8)OqTBb%DB2U8J_EH>x+OH>+*xV)YhviMmw1Rb5u8R$f^-
zzH&n4q{_`xzNzF?VX9+l<J56eCrw>5b=^~QPe1mn`meV9>L<^{o>}qCk<WbinLD1@
z{_E<mxBmJIzy9{l@Xpqq{X3WLT($Giou}_Sd*>T>UcU31U9a2q`dwG;x_;LOcipk;
ziCs@kC#M%pFP<KsUO9cj^r_RYoqqlF8>TOue&h6;rngPMefrAjYo^~febe-9(|1hY
zIsMt`uS`EU{pj?Mrhha2`{_N?FV47T0yEK>%*@ct@JwxH^~|Q3SIwL?bKcB_Gh1ig
zK6BN~duBd3bJNVtGasJ0W9IIe&&}+Z`NGV7GxyItJoD3;U(8I+o<4i=>=m;Uv)9gE
zH~WFvTW4>d{n+fMX78Q-#_Z(mL$eRhetY(j*+*v|n|*Tjm$T2zK0o{W*+0zA&i-+B
zZntN5WOsIV*Y5J}p54{mn|B|#`=s5c?0#bRuXpd<{l`72Jp+5Hd&c)1y61y??%eaq
zJ)hZg@18I1d2r9edmi2M{XNrr_U?J{h1d%%FRXcC?F%1$;X5xp`ofcYt9#Gb`?kH;
z?tS;(_wBuL?}zvP%ib^UePHjydw;n1zxO`1_gDK??OVI=&V6_9`_8@}?0ah8bMy0z
zdY3Az#cD{c{C{<HPnf!7>Ro?hbcYz-9lw5fXJ}`R(H-1*$j*9Wbl<k~Dl@uQn$i96
zt{*HI-9s9q`#MJVyy>ls?whCIIz7SYzT1rM?TqgB>HDT1nEuZ6W7E%0|IhR;M%S6~
z&qQY8jBb_DT`_YMqq}A1?3oJ~-OFbtX0Bm$Kg8(X!sy;L^G}TK=V$I=bRU{|V&>;&
zbkCi=boL#yS2McroxO4P!?Pcq{rK$Nv-dH&|JE4Y?=ZSQV053F-Psu3J&dli+qXNh
zJ8wp}cfsgB!{|P{C-{GBbpOMQ?hvDUSz~l>*n8{Ve`a*Q-Wc6qGP<Mt)*L*#^HcM`
zod4PUkLI6Xod1tL#lB)kF;`4?{;u;^olkXst8=pBsP@a+Lv7EuZEIW8mTvR5-rxF_
z)-Sep7j_mtSGcM0io)TAT4B7<U+62u3W5Be^RxNiOA}%<vERg=iTxt>nb=2T=bMlt
zV}}#o2eboiu~vQyu}mx-OU7ccK+GHS#9T2a`p4*==x?LXMt>ZABKo*e(MO^WMZX!H
z#CLb}6VdI_o1#}ouZmt7oru08x-I(V=$nXn1Mqs_RngP=JuP}_^rYyf=<4Xo=pm8G
z$d@8tjNBW!C-Q~Jj>zXCpN)Jb^6ALkk-H+FjC?$@9l_~ikvk)IL~f7V7P&QYOXOxk
zZj9W(?{$%PMXrfl9k~ksMC6>vS&=g$r-z>k{}878gYdVcSMmw353dN1hlj&`;YzqO
zoC_zziO{al??S%`JrVk0=+mLQLPv#)q0Ue|_>JHPgKr655PU=Mb-{CjTY^Ugj|?6b
zToGIn90+y?%Yh#Teh|1Mur;tbaCl%@VA$|$TZz)o{Xh8Lt-qjm=-c!y`p!l%|A&q~
zQ6G135$utdNr)Z?-UFap;FnSPM<1PPreHPxTOU>^C9BjDNuW!AE7{)&M>=@P=SlM{
zr8Ym0H9ZY?(ztJ8jXR$;@c+w?X8CrpPJ5JBB~*(_s+3BrjLNDU8i#^vRc)$Wb*N5N
zgnyUd;N7gzz349%sXn-QzZy`3YKdCP8hnkqUfrpFs(m`DTXehr7ZsMKhs@aEky+W*
z0FeU9Ko5Y>KuL<S6zx!~5DFPUBm%`+p^yd?ZBwio7BaZ6Za^0I>;~j;1zQ4n;G6~&
za9`U1W<2tQg*IIF4hG-{$QTwnaGA{pI&sfyKoM8?h=BCtf(8ghT-X5V(?tyssb^~g
zq@QnWKrgP)ra%RFa|0IPZfgKM3fabj;6^A~KyY+P0|sy}ZNMPzTN@yBdRYUOQbpB*
z;I{Hg3yiX=9B;w+PrwhzY_FVT!T8}s6#(i`Q;dgzjL%fkf}-px@GGE(alxH{8o}LY
zL5<=bXF-kOQqL51sBv8SGIb4p#(U~rz;*bKQ0l2U3yLzHW_AduBbEA9-Gb=0eznDd
z+JyTP3u-e;!I%YgH0}xu>KI(|3#empKWst00{0FJiZ(v8-GTy7zowtRrhf`N{d%hf
zbt3K#;0uJ)hhIMod>j9%O6?2*Vf^s_ozUV=p|w}xGG+oovpX55os5al?9M}gRrsOJ
zops<){LtV|$qxwK?UZyv-{<4L(Sp$3&bI-V6As<&yb2H;(#Bn{vmoQOi?Re{+;?3G
zkWOfA*Yy^JzIJ`kg3#73aIy<r2z~AP0q_L=%W$8vpx&<3G~+Iy-hm6f3JCp7L!$yh
zGt-Q#fSSOikJBshQ~xyME+9LCY49bWuEwSR0`LnpP5Y;%&Gcvb4Hk?aTujqf^)6ic
zISnq<dvV`n0nVeQ>D%-+!rzDcR)F@Z58zG!)UR&Fm2w4Fx8lAVAg#I$_a+PKqqy=6
zz{}M1?SSCqW4Q7RsE^}r2R@7c6S(&QU%~%*+y{UM@k?L613Zdf;vWNkgr9WN&jP=}
z|F5|J1N<KUw{Uj>d+<Mm`=SMSv6^uJaHc44#t#JWQ|?Rzh~obqZXC$qr=FQ13&I;_
zs=zSe)IC$PpqMK&D}dF6{}lHqU=#jd;=am)dJ1<7a2DZD<3fWojDcv}XD$F_ENJ`8
zRtxG6xX|Ow+X>%=I{{pUe-;;f&%6i!3%Jmmfbjg84*}p^{RNk?5s)3@%q;-4sOE9O
zy?};4&)fyvjUQah{1fmw{LuEy4h!0c`+49CgbSYT0l>Wu;@)pT$8f>#%)^Aoai0Ld
zfrh4Meh&Nse;#+rf^MY?r(4i%xaR^FlZN)pg5TLI@b}}s1DL=MzGtE7*=zBS;9h4z
zB$L_q0??yo9A<9>ZpFV6m%h&4j{h*+j{+aVzaAGloBb624Y>4s_Fnvq(=2_O{RaM1
za3?Kj#%K250Q#n1gG>Je<P2f<+ZJT3XCJYkUxy1l%|b_7`t+Cu%^1)A0C<w{^KqdG
z0euNBG%@=Oe(Cp4;CcL)<Nn@)mVQGA0y1}I8G8YJ4elP`kNCmM?3@MNqSS8aU^inZ
zQrm9EZ#Oip-;bLBviNVtWjuCw;r|G3*@C_cx5t8x;`RdIUf+$o*@6aNyN|P=zleL1
z1r7dogF6BJCEO=0=r7~azumtk{42QhQNWb{EPiSKSC!fWJ?%;12X}i0EFd|lJ=8CN
z45juk-U1@G>^ane(Dt4WTF^hj1xEs+yWR6i3;HLxpRu4BuRZiZK>rMvwhCy*X3v8b
z^v`i0wxEB3D{TTaW47n}7W5SEv<1x=?%8WWbb5PUw4i^*^o&{1;NS)NBcQ>>3u`QB
zaPh)g3mSa90Idp$F8_u90H7xgUweW21@!N5X{&&KUa7r90Qk}HjlE}B(9p`>%Ye5L
zz6<wS3wj3k-4--7wD)}$^lsc6fE!7_2lrOs!}#gb-hT%E1wVb-%Xkauzu<lycmV$#
zF1Qka?9|>LS`a1f-v72B2LyY;-CpqIxNyOXfa6ta-zcyOKXkeeoa_S^P6C%P6mZhG
zcUy2WxZp#;$>RRNf|J9gYyl^a3+@D*g3`P1HDEPnw2#lHXf>4PEF<t5m}3zjwdXYe
z81>?M!J2Oey0kg~b^_!Hx{<mC2oIA#d_6$A2v~?d#-8_BrQ*cLuY*9tz)c(*_yOXR
zl#`rfQC|r>iVBcA(;rbPL%Pfk)Q&^IpE*X^j*v}R1)d9^;Yf@4R?@Zc+(uk`T&WH?
zMdvQ1ij>#2RjCqfDcy(q?{Nfw;wp=PA1k$J!fpVTZc%FKZUjuqsMdi;5GwCMn8ZJF
zy;7qOC^ZJuu2X8+q*5!|0Ng`zO0C+a)Eb`aCo8p<zO8*usl!eOD1Y7aO09oJsf`l|
zk>@M5iSjoQzWH*cj=n~zSA0aN<BnJA<olF5jj~SPpwt;V5Z*QcpHXVdBWye#LV$Y=
z!R<A`J~j=+T|k-(w<GXV&PCMsrcnf}lM&)m2<|I^=MmcPN1&pOm(q{7QtoBceK~QL
z6L$r5zvFxaCDL5^XQi$p?CP5kewHb9?QOt5rQW#}coe~h@~``hQtzgY_b7lm-y2uz
z`uh-g@V_7b`$_wOFhJTH?osMS+HoUw-$EO2b1U`X0i|v)0+f5lr<D50^)Tuo;C=+0
zN0i!5-#)$!q2@_QkGk$6{av>yb$7Q?pMFTGe_Eu}XNmuu8=%}B*C6yz_B|5_JmmS}
zGYC6BR_e>t`xV-9KW+Hxb4q>fc%>fLrPMbl>p}8N68BB~4|OZWWK`e2PpL<!>p%Pe
zWqcR^qn}agKZ^)A8<hI~N`U<Tb-q&n9aiecLrVRGI-ex{&v&yqf1aHKVNccB1w4k3
z@*1U{CI54TJ@=eazj;)t-%{@H3ID?Y0?N5aSHR5UtVElDKf~QmR_aC4{E4>wWt&oS
z&m+sam1f1&aC+@*NAP$Efg=S>DeW6Vu&65?$|)T=okgNe>DWqyin!7(<Vo%TNSE3I
z;7^k`OI)rS0fRE~*C<_hUg<XSw>^SDajw!G$12_F2ev6)Tm<Y^x<sDRlS+3FDcy6v
z(!Dn*UD>7d;>VR9oJ1*cozg=C2p(ajN8L(~DS&k2)HD9P(#t4!`K18<l{YE9>OQ5{
z+@^G$GS;39P{%sb9+3j@ZyZqi$mf*a%=6~$N+10xrH{cq?g6Ebr;jHrQ~IQ>N}s$5
z0fgsMDEIU!rO$Xs=`%Mdz2$nP&$?gfv(HBmA>TQa`C6V|_hST&<CT5`Y0f_u;Q4|r
zO22tP>21_;@#Cm%u0zn6M8F{YQtEr_l+tgbZI?fy^c9Pgo;VqJPU$P_N?(-&?o;||
z>V0<@cwXuE-lO#Oq<P<M2oW~{yOsXHr5u_Oe(N(z-?0rLqNwzp>j2VkcPssI>ih(0
z|B2_%(XSmFl>YqjO8+x;+@k>U-^=rTr2SF~ApKW}yZ;7&wtSU5U!%OQJ*M;n<oyQi
zesGJ@leiDXmHzi@l>Rp5e0!VHk5JZk&Q<z*14@5?k<yT={;wh6&r1K0_{V>&^b<Rj
z{?Rgk_Wa~}rGNSmi~lGv$-+<k&jA)F{WR@)dLIik<?mbxe1wIVGJZ>)&r`<#+{c32
z223eEb3Q9ePU#m&x0mogQr;gw1w5(rpKfB2B!2EDPL?*X=v~8N_Z*Adxyr#F$B7aa
zy$!&h*u^3TWHu=$H^~C#XW@DbgO;tzX)7wH^GW5DA5~6w9QcfKdX802FZrQaXVK}(
zS=_Che)0^Qtehn~l(Y1H<qT1F^~cH?8BoqB<&F_QzDzl_Ei8N6m9zXl<s3rVRg|%s
zv}^DmO8jBKx^2o?PuStlE6jJ4bJPGp{ATJtmimqhE9Ydla$fz2a$d7RIp>Tj=iD~s
zoJSnA=e+TJ<-BQ=a^6gxZ+=uc+diV4i+`+~OUQTWO66Qe-pi=}@*;4(a;_Kv_9<rq
z_e$d51Fk;wymD?IQqCQe{gKC&^J)70Y1$3_IRCsuIrmWSy~M*Co%<eA&X=kCE1yx$
zS84y(&R5RY$@7hravmi7LCTsW?Za{9Kr_y_A5qRDo0RjN`<3%u()}lOevjwJb|~k+
zb}8raPbue#dzACvmn-K->y-23$CUHaN0l?xrktn3%6WQ9InSJ{oL>{Z^Lphxdz*58
zGpd~5-lUx0QReeJ|2{4@R?5S6Yf?|>Ni+u$cJ!^Hsj1a`wMsRS9Vw4gs=ivbl1<>0
zN6Oi1t!KPet&Z2as+nwtV6V^X>#FuvdMk<gseCG5{Nfi+J!Rq)J|lQfIdyKQjzm+s
zn2JWU_~xd>x8j|A=62&u;?ZGo;`@_qY5Meb#mB*_YD#V9A&n+dB}XgOYHzvZ^QIHk
zR5mj*TpJ%txW;Saw1W1~u6iI4jRyG0Z!sGV2E$qVt0!#uyt1xVb6a#Wl}c*+Yw80h
zztEHV<7fikOnsSbwVLtysw34JJs+oh`!t-*q*Fd`xm=aFg-^7Cw)EC&Rf${pRI2oj
zjn>9Tq^xpv;ggh{_0@bP=1WnZPY2Hoxq_hxUMH}^Hg|ox={XSf4^w&5>GYZDGxc^U
z?U40SwDDQ=OjzeYFh0NYHxg{0;3GKFDhS<-LN{**L&Jihbh<h`GE~h}vqBrtg|||v
z_KGVQ%Vw*?hGrUZgr}>5J;A|Pb*MZ9J{VWQ8<=b0Ihzrm31iT!Lkr`H8LJ8|w}TtU
zrH3NY?ocqgbf}|EyWCDFS9R%3HkZmo!(MNqC7;UKzB1PNLOSemd%WRLB#}<XtF3v*
zak+x|sw<NtLnPw!B$N46mJE85pKk4!q9crj-xmqTV;-+B7S<lWKO8Nz7Tc4_px@PH
z2dV7x`2yZ>I1+YyJU*}2<qd^e;*uexz2ue*Z6-sR3`Xvs-^(0)x1sN)Aa{JMS{<zj
zjq9p6ZE0NlzzbLdfAm}*^l(Wa5R62EbF@@kUEixai^Wde_(if<@U&5q1OjuD*46bu
zce14=*=>L4)xsyG=jZ09^i|BgBNvP_^T5mkFSBCdXfGU5=)N(MwJLPb7}IRkj@cwV
z-x`X9eLj~f{H8aDTrRIa90_f;19TC%voUW9yTE26MsFwjy2B(3S^ewQ6A_lIhrP>U
z{#|PualZ8u7n00}T_;G^Dn7qU-=l93&9y$jLDDh{;Ut3Nkt*$P@RKo_(eM=+g>rRF
zE8DOeY&|y|{G_+OY;YryY&yYV-S97@(4^&{GAGi8e}bFJSk3FJ@{~4n1e%AoC0sd!
zgX`7}4mwSDdtYm-)~&65O;@XC^u&J87D+JCOkq4SXU$kDbSQnag3=ki#_WVsB^F}B
z@KoFP#C9D|#J7YTm(L#whnLiCU$<<rZe33#p}M-nNt({NCvCs&uihf{{;AP_<nm-m
zYBjTdlv>AYf(NhRu1yYV))Nya82CUb*I_BOvw!94MSYI$d-oS@-Sa9P@J0VMt93S8
z(pz$!mi#*wYo~9~>XrT1HQJC&#vja;O4+P#;6TP=9c>wvwTyA64W<RQEnc7D%)-5d
z`_Mk2@e$$uHEB`VP;--i3uWx09kFXND|#zhS9(_*Q!KQ`9FN-_=HqiB?Z<ScQd*}{
zo#&c_chY>b5b0z(cB$l$oQZh5WbjEocitp7*(42h2p<0o9w+oC<kTcNE~}v72(Bs|
z_U}<|gO{)lFvhiSNCk|RcfuD4CZ;ToiWZ|q`}BqL^qVAwq@3FokHwC)(Z|~8<}7%+
zNmGJPG$T;<!Rw&GETj96G0G}Xt95zv!WXFIuJoIl$b4mIE+6qUk?9WQU)e<FZLwg$
z>C*4aIks6_o0yDa575M<KfNE5p>Pbz9(kfMA3x4X<m!VkDf1zfPNj!fX@nM<BVNm<
z%<MFDYgcgA9q1Ja;JJX-tgGR`jXnD(sk6Ahz9tumWU~2umc`_NPoF1ACCiTf<yvc}
zIcL`n40?|GT_IyrX9@>=@Gg-@)`8=W{WNRjFCQb{kTjAqduf?aqgE|JzkkXv)CY5X
z=#l;87K_K{CM1j{EUaZT1s|^c{HesOIGaK3HMCWh(phi#O$#>c41g<C#t~D>BNArE
zR%V6ZY6QFu!8zr3Y{)FwW%F<#K{Jyc(%n7d#qpkQt;@M+RG#Yc6piM}dLkZ=x3;us
z-O|!pZ+hazV$-|dqiqRf9_wylT{Y!N!L5as?y(8U-^?R!{iOj-dibT`&jqy%45vhz
zvFm_c0nAD)Sjbw$NLQsFsk9fiK?L20eCr2>`GD`_L;Dx+rwdp2jRBUWpw|AnRtJXo
zfN#5JaA0kdCfx4Y;vFOoNT#C<9n7&7UnO*)Gln8PRgd9b(uf}T!Fa95tj+ogGk0Iq
z-mu&4@j(f5lL@U8x-Jt#Unsv<UoGJhzdeXJvR!gme8g3qwX%*p(hojLQj=LByY{b%
zUFnw%9kSC`Hx`Gv?Js>s+E^f(PcD2gycvs7+rOmOjDqx!zJctF^bGz>FQH18G*<iO
z{@CoEu1ojkCjUYA9OQs0Xm2|f25cQ0Yq8~1vKq^9E|dc78a0lDDkN-YL{ml}^MNhL
zu|ipvEZh1k44G^_LW|pC@%?me!;U>@JF=gU-KAt(VMsUd74f&Ww6%4)o8)YR=Oz!@
zst6yO<PyrHI#!(lMOX<!2(!78N}KJ>v;=!V;_&*KF>pEIWULJf`z^6hyP#T7kHD#p
z8y|O^@$utqTQ)3TQ7UO&Dy>+)q2Wj77;NhvX|}n)=#SX6{7T09L1O5H1WB0|EeDlk
zN-Y}srK?K|0!^V=KWUF`E>)6u@*rUq;&qU5eqrR*F13j9l^vl=_F#sB#*r#Uti(!}
z##!NANw&4MYW5d_blQLEBDMr@#;~{Z;i3mO0+lVgbny*0+^}JTzFzi-b5mO+o;efQ
zv`M0o^TS(oac=kLKQD9V059#tHbz#Lrd($wyM;Vda9qw(zX+Q$Y!JHdV8%LDZ*s2*
z!9XI>)^5<T#eAA^JWMqG6MDN);`Z$(0L+!!+Y_ZB$vwHB9qwm?6Vhfq|3@QNY*$F9
zx|eNOQzAwJHnh>$r&$?(L5SX1UAeNlbgBFzqXmM&btj&<ZXgzo#s<U}3<hL7A}Q*1
zeyx0V=%Itl`3$UG8w|@%MA~52WEzhKLU@jqku6#*Vnl`zj3rok$_X$gTv!5>^moge
z;gR+ZtvlLBhSzM7zHb*4H94s^K{mFLl1#cjH`$b{1rPhzL3AL0HEs)b162Mup6bnE
zn6i{V)nI_zo6ieJ#}fX;L5|?!KsyExYDbgryku{(ur;QBTj1ZQdD1p)`%7xIbxwl+
z75nQHk`yE_)KkDIAX3Xc4&p2l%3w@2F@8vYfA|Vp*9|F__rU7ar!E|$lUK>YCIm3B
zMsI13g6=Lavn`88vnMZK;Vc|7rzX@S*%r~<CZ2?5hB;SCk2N;R!mQb3+C>|-*0td2
zfWm03+I8?+G_P#M_E)P(J_irt#DsmGYq*p9+fb(s>?;?nlp_c2rw*)}Wx^LwPtlgx
z+)Qb8@H2h0&5@n`0gYC(;Rn^f7=&0ojMd7?)UbPP%iWu!nJ_vNEj2f>-so~;Q^}L5
zc7FfHve8P*kV$2lY4QIHy_UYq;llx)g%5svU;q*att+rvVRa4eThldo1~hQTl7;;{
z^jgscu$%N>|8D&T!z6;+mwwgaW^Ui5)^pxrlOdOb0wxmr?)yLm^QH(Nld%vvxrvK~
zI}(wQZU5_(F=@-0vX1HVTm#>Ng9*YLb8%9CN_b;(oSmT6eLzSJ-4&`)v!6v_x*%-4
zmd!-(x+|K=G(>@GgNa1&+6JaG`eUA!efwHG{%n(&bmoz0^pUyWHX1XL^_#k%L>~GS
zwwF@3rUub>$#|MEu1Ls%c8u53>6rF4>PV$ybAN35^_|}M+(UYiEq!hx<M+n&s<|IG
zYs>h-pXI@m;IAMfYN;8i7ryfnX_%EtlogHbkP-bG+o9}Ke@_p>M5d>w|A22ww|9(=
z9dgLnXb0=pLjM+0_AXxBYx6XG+gFW@bar%fj*P52;A^%|#<vs8(}ULt;Ynqow8p5w
z0z`e?TqB6G%S4ksZLyfCvqm)4BwOqv(VP@pEN&*ICY#!p$p()8Re#gaSRGAy#z?8-
zsLK!N3hc5ZEkfrf)9@fh3GSEz4R%?nOLYU~4SuadvVtDbshbVZ`LT5iIi~eHV?)if
zOD3f^I(1@*j1qXl;##wR-#|7qYvIBYPF%HI>L6*5eU2G#JAeDtA>i|Xq_S|OSvX(!
zqw}Rw%Yt?1%-5O8%gj7}Q9h@gMSZKbnolX4(Wqlfb_tx^-oWn%o0Is`M0a_~s=h_e
z)OJZG$(hsWJbujH_cLOxOAD<onccbBeL|N8k?b^$YyhgufN&RRmU$vy!)s&^&>l;y
zd}~>Bk~Wn|(-@B4N{!jv>pQ!{%?2aVzRc12)<QNHjd-IgYU`7^e9Gf?bi^n|LJ=SP
z_*60#Szi3>^u?m&2&FUOh|lYCheDBHDi!p2mKXI&k=9fo7-)$ySA(&5dq;OYmkGQ4
z=$5tipxTu3QE1a!#~=Nx#2!zb!wl_ln^L887s~F=PCs@$4SvPCAy(EeS&y2d`f@AK
zgq;t<^YCrozwR_fNjK)pg7R5!8a#HP&a%m$7HBpN&ni-IQ;xS)EfD`rZBxyC!<22q
zRCB)}&t`8ZdgxCW-ac+-U9y^Nj-k<n^j2JE7N$f(zv}9y0KZ*Cdp5V8i2kVj&V5@i
zM{@T_{M>{H^?CxY$JXJjr4GYg7Sxf+zN9Yb$W;~XOQ}(38fCNR!V~=2tp9`y%^0Gx
zx*?Fs`d;x0e<l;Sf%+o)PWhetsqU7UF!!YH6~&gMo|{NJX`P+>jW`ohmW<{6{QNKV
zm7GI~&ia79&u8`NLbI$#3->k)=W0#ARVKRdhIhS71g8*MhYx(9$vjM$(L{XU1EQ3&
z9OM(9kW{1&eQrmSRebW3qOXt=c6@HZIP@_7Mjg^<$HLtsYoyFFVrc6MQ4I&-2e;dS
zzg<q3*jxEI`1_EAOZ<;+v7>&=k0eG?Jn#T0pgz$ZUkg3=QlHh=THTixIf|7~IKI^w
z>xDY7dMe-#&;9Ptw%EBzdDLlnw3^pMGCubu+aR4t#`Grq)aR)1{6l)Let@0a+o^G=
zTJ@M-Rx;V4bd!6GvPS~@_M1&*QzMy?VKlIImn}Vqt5~&=<wo!z(VOl6)`ZvcR2yoR
zBCgT0S(#h#H(I@lc^8RBQtogfp3G-CK`p1#-h385f$C6xbgr<pQ1A!iv6ca!zpZuE
zV4>Be<FOGR;~feiGeQGyx6AMEP5a2=cIikc5>Eu&JQTZXwRNW-wl?PT^IRWHC0cxo
zi@GzDNwt<Uj!u`@;)k@uhnnzYvT;AGDL**a+oN4=ZJn-`q{G7H(O!>_QKKry>F7un
zLg`e1Oo31&pYKI~>|0l>6}z;veA&rk<+7jS6p_`OMT~10yUA{~L7j<B`-Hkd-3r#h
zh2h(RFBi-BShdVn4~&^L-pJ3j>{xnCPHaXj=jW70{IXZHKDawo+0C`+y(PBca-dz2
zgISc}1ZFE;U0n&b3GC?_XN`%*7E@Lb+EPxMbJ@z<i_uUf+tKczq)aV3hR<!9B|aMU
zc|H0+vL$bOM}Pm|nbAcWN$bq>1_$nDn_DY(bd-+tcsw1RTwAeiVs5t%3tU#Oi+Xds
zcHUk-=dD_$^&u;^9`E=2<38Q$w0C4Pp=f8Ct<&6}-giXQ<JL>up4gJpPCIS=>aMPE
z_?1ropuSrV4jy${HkgfM5`0cO?U}jyiscvfkB;>AICCfHTZXHzJ7;862MT;-J~?Pe
zru2j0y-gj?uIYfRBI_U+Fu%|$xTw%;BdS!e0+7=@v?oT+8L5dBj2xOpng>Td{edOg
zS+n}gGgq&1#O>?#xZ8(WDdDc^!&j_Wk>)(seaImYqE4@HX~(@G#{%Z$Hju`3;@HL$
zSFCWH6)R3W2{)?Uy<x)hot?Q`1_lP!yIqdc*T?y+d;I{XD$Y8$d!6Gpcb1I|GsRi`
zcBqJ*D~1a`vnXJo(3qY+Z8ViuU|l%duv=ifYolZKTjYa2t7}9bal~PVRY!Wdd-Qgv
zr+c737xD)};Z$)^BGFRF$MwdMu99wFxpJl6Fw6wW4h$|oWO(=xeRVqB(>q*f>u7CF
z7noDQmX;K1tzlPn=+Js~2-Oh$Omud%cT%l|&LPAMEYb^t1-WCOISp#ZP~n1j_O22m
zN34Q)NhcdR*?^=88zU^e``BZTb-T~>xVO07$3A+r<2jSLc<j-SK6>=AUd@AZ^rQMU
zOO`BgyO%l6GPirl>6iRr$r)Z}nd8~&(My+Ha><gV8ZQCfGnV|}lGACU)nN%vTewe!
zHFJ-gL^sc&&6#MW2Wso8CR}EMp%#?`y@L)$#;P^lABtef5s9SIk*L=jntMV|ZX6pM
z)A2>EnRGZ53a2xLcr+4T_9ZWgGLdLBD#700&DHg!f%oQGqnx6&wPmsq*2?hFqM-H~
zJ&m+4tWHBNyqen^4>9%)9iKGBXEZ~-zEWAfy$6OJCN2j&e631Q1VmWPFzkDHBqI#K
zek+zJgbfR$_B8}{iOfpbHkHZUpfgdMRfv-0miQ7zsb{Gy;qsP25@?b=C^jHifrzZc
zEMgwXIwmVw%J;3Db`{zR_V)w+P!y39`Ofip)45zU<oCNh!u638-TuEHdsvV5^etXw
zf7?C6!qHZGydIZJ@{7>u@_KF-<=Nb1Q*EV<Ab(n0B`ORfT4^h2T{!sZ`9x<&G#pQK
zbn?6YrR~Y3q-2__Q!m@zTzX+|23wab%_Q^o*Wu+$UNoz$Y!2)aUO`J8_o>acEo|l7
z&;Gh$ZIx_)f3|`Rl<J({r=Q}jP&v7SW=^^pq?gD_<17MM<oyGZilh)?V2mqoj-*mC
z*4yCery~(XW2sd5O1Jx0r}IG?itrqnaJ#Pzr&4iJh4i+#<3z(PiP+n&h$ULWQOAjY
z-sivOiYu;oTl0B`&-aC9Drnd<|A_uN_X;v<6}A;8srSpuhh+lfYIZ3JCOw2^zVZX{
z3{{4VP}SsYh(6})K@C8xj|Y)UgoHD`%CIb5eB}{&M9#F#xfVe}gF}>0Z)J--)~H8%
z4kh}kjcT$Wf+UoquW`F$ahCd!@9o;}4?>9gY?mkACS(<E>Fr2+23<kfvv_F*B)Zgb
zMVE!6;jlm93kIE{&le12{h5?6oo-+1${-RrPD@K?PcjwKu~;I=aLd$pT)FIUDj1B1
zg7FNbex~+|HBvCDPZp{T2K{r#bh^gJH*6TMxk~{*CN2@}EG?HK#(c*+(7$Zi;>E5}
z{bhHH8w%<GMfq&q?FqD1eBNSb(e?9aC=iRac4soKaM&3@JEMJGZ#u-LHPh49=Fu)k
zFLxmw1p}T>q;l>MqN^Lzdg0wpUB7j|om>ekiXl})$G;v9X@wL-lJrP*tRb>^Y@p~N
z5MT~=DmGBG{ltw~NB<l2()i4+63;F2bL)QJfkz4Y?#6RA@R|FcjbGSkKW#W|9Baus
zUzq<Ux3=%c!mkEzG-t6ML+j-xQ83mlKV`ED5@wiL1{xTXr4y1vaaz!iva=i+t-t<|
z6`^8n*=AS1kk7g9@gPkvm-Vu%Yi#Yhb-n#-)(ltOdc~9RwaaVm?XG(Ll%eh(H(V;4
z8EtDxMhi=puCEO(&18bX?CaZo{#<SiQxjf3zelZcCiJ1)CO(dvzL#>-?>5?N5J@xa
zx<d0I%@J5HTX<={mlaSblTZX}vp!{wL?VkOQ)T~ZWhF$>Mp71?GRAkq9OpBktW0RN
z>BD0bMW0E9xerpIh<+)gi5dv1ASud53)8$eEO#ej7>vas8ecq~&L%qA`#SS^6n)Wr
zzO%oz1G%~sE0hxHa1v#8G@Xxf9Foq(bS9SM8a|UJ5KD!^uS#Us<SMy1R2|b;j|Zcb
z3Wp&&u&=wjTlW-3+jGbKrmwFr9Y+6|Ov8V|I?>stWAWCZA>A4$Q@(vT6mm!Nkwm~B
z!F(!|N<zdf$K<1@gdplb2s5-;=>74pZ<1Q#nM{05;80^AmJ~jAY5jgf*Y*sNsdj_D
z1A+2>{qOpRPS8m^CFmL!**j*LWTiR^8jZhPsg__)){nx(`jBMAhwx~8q_;j&_ZnYG
z<QxVENk+mLM<0Bv;fL+8Ci_fyI$Is$Y~A=_%I4^YF&E!ZnLNgy@(qzkXd#Q*4<Cw3
zk(%%^c7(B{8vlqjJ+yvddd7zwB|elQ#y3Qr*3WWG4ycRp;p`A2&*l#QxP(G)gtBZK
zKbRSXfm**Pl#GvFiVtDh_(l<z8-Bs02{lq2#$`T1l@K{yH7M$>KzbZtj6!y@IT12!
zen;pmI1_E6Fmfh>cxBI1flh6~c8}9rtB7zRsU%UBXj>H6Drb$T#!^BKD<+w_U>a{a
zD)kA9=nSnWF{5OmX;fv(qlEp%>%U}BawlQW?wpP}tk~Et=^%SXcfoPIhy!kHE7^Fu
zv&aSTWF&w9qKGTuaz!1t+ljjHvh1@{b(}OlkK@XcoTof8fcD}dzBsJAI?ukqiRm0d
zf-9*rF4xP~?4ibkYDTt@nq{^62X#~HODZq0r+t~q({A@I5x2jmZ&9xwKIbj^{U3I@
z2KA6T81!lfA;|9w9^()2R)i=LL{vZJ7>|xPC$rU{S1xa1Ea=7z)fd%HG~(Pr2iLgZ
z0ccx-??@2tXK@T8e{e{GPRRZ3{?$2bR^6?AU0vxGcc_f+3hPM!$9VAqQV7NWdw^BX
z&EbWPK8!TIAF~cv|9gm~kN*q2GLnqZL5OV(DJ%a!z^m(jkJj}6j+z+Emm{Yuok%&+
zQbsE(>Q!glrOR$NCoPE})&ocot;l9z%88g!!T&`;6moIw(sEC?XkJZB9yaWDhrs9Y
z!7OrzMgi6WfAyl9S{lx~-Fbg5+p{R_#Zt+|8)Q6suNg%!f4kr3H>1w_Ci?5In*E-f
zQI;xpE_o}=cDspSArnT}IAxJp%n8*9tjNe}yzFaTt-TF<s-be1*OSW+436Y78989j
zW^&nR^kgU0-`V~9%Dtm~eaS>PT&WCKmkeez-cxBqDwWHolPG5V*fQ%gLZOlN6dTY3
z=R8p?e9XzIoRKz`$}zTd=t^V*_P)F2^V1jkTye$2SIBxzUR7tkz7QM4*V^^LG6Tce
z;El3x5>^JEu=_3}g^KFA%UFULPD<x%Fn-~E@ETYpQl}Tj4ju>-eVVl+gKZkwRNwaO
zv(L)p)$ZS9l1YCwv2yjH^;}Cjoow;>J&Rh|4QQACoLjrR!9em2=)-lu8w|D#lgDw7
zq?1Wec-`aD9^H*unCwB|Pc8@{U6NuCD|O}zhek9@Z7Somwl4Dd82mIh70%WUN|*>@
zp=8jpXgWX!S0XW9?w0cj<oJ30fSwSm1>M!vMNjQlhLW!+D}`|j-!xwYZ?=RL{q?yc
z^_S<4j6@P$eSIAr&I!v>EiJ*hKl^)5iKf%}Tp^!M!4+TO)q3LSWy^~7ayc5)#Ul#6
zy#vcS(=EwRFrTX|?(B5>IhsU7MW3eF6Q{9zStswMFl%?YB<m_`TVoBY)WlMB7_A2@
z4x&ye?ag{*krzL<k0R!<aGEqSm_DdMitsWflFu&r+RZfpFgtN2xYO^z=oJ)deUa8{
z4L0BL4#Eg%Iv-rScCB*(8P;lzqxCm`@rz$*z5h>!Pk&2(o-`3!|NQ6jM1l_eui}#*
zp+8Zg6rsN@=zH1>&QU}_Ch`?yZTXfx?Cgxjq&XTR=D5_S>&5ntOcrH-IGgEcFV0PF
zHy=IK2w1yxiPl@@CdDwM5!A?G6K-!jamc)6=~~+!ANsUEf_Hm=Mv0hTt9k_a2#pow
z$80tE;otuDx7$#E48G+pZ+Z67=P$na;%%=hKDzvj#~yp^`Ja9N```ahU7;|I?)mF_
zhFgcPX7qaH4gkV}>5^~=OIA&v?Q%^Bi<rg}NH{=zG7%-b8DdH@#EHl(rRqnz5{Xc#
zqjO-xp@+6b99IOI&t*GHj*}^MIu3^dEPcU%pZ!=o8e>@wB@$7N>r(M>$O~N}WX2IO
zTM$pfk(keMi|rIFHso`k-QL;RgYvGc6iu~v9NyjA*%}J<c625)FkuH%J@m!w_0xqy
zI>Av`BH;2QQ^{ny(3MBDPQ>BJ(Uz8gKjsT2<8kfDWLx}l6Toq}%oFgtp?7vR(7e(8
z#W)dOjuqS%@MhVUNL<Evzw(mj8go>TmSA3HvJK^o2w#<?JxP$gu{~!;BP_e&mlU9@
z=rdR3qiCygA35Ot>D<%0WA5p$oLHmhx^zd=x27don0vY)S1by;1NVvpnLhdINHhy~
z`3No>h{#!z^`(W(-MAn#g~S_)*^`Fp|9#NjV%}SGI`2KYkaq!H!@EyzlXsQW%(MUr
z6(mVPhKSS5WRd4Ywlg;VV@ABqicF4)qip+FiqfLyg*_igDBN94Ah~7LTM?5$2v2rj
z$d=6z6Ksz;d^Sv&{sQm*EJ*e??+%ya3Tx-$qdcCB$MYI~UGP}fHmz@UTt{k8EDF~>
z(m`DGoP)UJoC6OJ^dQBh1?o-DnwNOk_B7(~t+tUqZ}1z<WKH)+E-vJ--Cy6&^6$bd
zdWGYCv)4hP;BudEn7&c>zY`s*(MP?rp9Q7hajnpLh09am5N*IVA}u>5Tj23cPa`DY
z^Xf)ez~}3>vG|T_rrXf;K+A6IQXkXYP=Turt5f8iQ4M=e(dd|23#R33_(8?hD~d*T
zvo$ov!m{~hvo!>LM%~PAPP>IS2zOw!B4>u=7Tb}mF9~xpBms-4FzyfoV}h1P9X38L
zyZT~reEhJ5{&z0QWCO2@dICptJS<D9FCKHe0VfvsF($}<8#|URT{hIexcIFS6D;6~
zPJ5k|t4_m;_^-;c<<6bmF*rEZ)!7nq<e=N@_8pC!!ouu{MllIxWkyly^1GA$`9d<)
z-81*KKmF-XKDU>&+6fi&`O{aeh6(t<-MOp>S<Z#4tb1qhHl+8e4?_J+TTn#DW&bWl
zpH;J=%Hn5Zh04|Lj%4#?wF2p3wpDf!5v*CffqlWIs10h8#fnK~h@BXdrom_`)+C~#
z3TLv2IOQG5pt2v4HG!k2dwY6%+S_{vds{oo?YgadX-7hAmeC-{*-d+km)%`U%7gX{
z4*01^z75@Zv^5tg6+7bG+0%YsEFozU56DqPEL*@9j^l?`T?+K_Isf8|Keuh$Ti!Cz
zlIZA2wAkNs!{Jyw6h=uGipQ||<{C;eRe+)BmerqotqyUMO_swL99_S*h0~_&ro%gQ
z_n{|oyes^568odc{W>qKf`r8We{?t0rAFX^-)pSDMcz?iRB1k|d#fbSBgx4LIyQ8q
zVtkeuc_kmc@i;!0zxDY4{9bYE(nAkjs*ip0$tO2%d~)vAx4lijo>yu>SDT=#H*n*0
zkXr$#^7gGudArkZ!9d9;tWa#hg+W?bNsJCHM`dj-`+6C6ISlst8eC68Yp4fhZVGi-
z`v|k93m<OGS<*vRM(4GEvRYm$#M)C5GnX+liEE<1r7)xFvg{#4t-Z|R6#W(A6~ZK>
zQX2EY%*%4iQY1Lrc59y}GM_hHHRqyDpS?+BvR6*zw9D7dxv|UD4zF&>whCRfMiaqc
zqRr)M`w04MS)Dz;B>}lpinLH}g)Bp%w)Rq2ds`?J2(}i=C+b4JQYo-`%oi$^d_fPR
zkcvxQbYB^c-{W3Sz~#-gb;M%2C7I;tIhEnq!|enKEqBHfI+;xBEiPI9Q_KWkiiLs(
zI59x`V!FSDGw2snt0nGUTFPGlaiY7FN}jL7{VX!YF6R*Hq+XxLbL!yWphql8gF(-r
zlTMXNokdh^>0)Q8luA42<+vyl%VpWx#j^>UE3P{oG}$h1%;%0qBgq0h$_WIu6Gq~7
zq|`WPSgqVh(yI`Sq9gvSPj|>Z!{zFrIaz<b&*S3g3Z6CpBLB@`LT^KVyp~%Cug22g
zc7-xd<T8=N%}J|hcE*joZSTfHH7wFXG1k&Rr1RvEk#S_RB;<&OQWVPEOhY3oK_%>B
zIPgM=C1oX1;K_lVBrJ(lAG@q(9bz4&O<Xg_fjm-{hiN~KP|&Q%-jR}hn$XnK;Na#U
zOuW^XznP1-+O2Oc6<UL>fp`n2<u`kD=oMh8&!aOVuOA&9?Hthq+STWAPID7>`q$BO
zdW?#NB-Yh63Dc*Wc<q|o7nI?0Ev9gSoRw6FCiS&BbO3$eIGye~w=I`rf7pjuI>Lb9
zeZ}`KyW8!V`vf~f8uThX_ZW$76sO6g7}f3rvF@9xk#tgm3)W}%RJ%wxi8bVAb+)``
zky|%*=Gr^&GRb9}5SL_yFel=&yqGD<WEPcihQ*g@h>eQOFvBq#V&@oAttrhcL2|S!
z`Knm9Eey4cpsIh9cbG<t=I&p_1QK6+p0yaeosB0SB@Qx)Ki|G6U(vxpbZ#<=VLitt
zbzD8Sk2O0Gh(*yOVMZMMSMwN+LB6EHMC9bw%6DwBh5m;tt$AI*?hJLw`t<=+%Ta$p
z=MRsGWmxQ6Ql8}cR(~MMeP=1~WgOWR!$04pH^4dHz}p1hqdtjjZrB@iEDBw3#jq=(
zL$iN@w}H)}F{oDfp~0(hC0_DCdjiakij-jTOF9$pDallkEvjr*%N5_k=Ty1OQf-;v
zsM+ZHdPZe?)zDNJHcV0Ix5xAi9uG%9n03f;X`rJ}42N2}&{=A|G9WrM6q4uoV#V4}
zUysi}*fHS8BBsA9>-PkT%a-((Jf6<d_&{$y<nqpKhc{_wt43mZ<5J^)KT_;XM{(;9
z>RW?SRPXV$v}6ZD9Ofnp`5xyl%n~>nx&|+7rX}5*ibP!Q99mM4&OXfRc2cSSK(LS>
zT+RI&N4KWhL%~76ueH!Wx@aI6EX3P5bwMC=h-}Bzeuw+QJUM=@({a8Q3V8y9Es40F
z!+8u<*j3AU|2MU~*I({?odzEom;C_<6I7PiHwfx#H7`ElPnN#3hQG)$qof^gpkGAE
z6m&0KOi~+VKcj7D2!hv4Ftj<-89OV7b;qlN9obFkP;gVGGx%x=I4jnf*?dbnz9}4w
zhBwDDJ%?m6hxBA(o5QhKXj430T#?PLDBckWY)<R5GMQIrHXZ5rZ_b>R&7PInyeYO>
zIQr(;3MpzuY;!2IDW>g%>Eoo)!$pu3#g-d#6A{hs07W{KRkZmEf)O@4LXKUeK#l(R
z^K%of!NFR-HaOt=y!p|!!9myD#Pi&$8Sn0PCY|o?aRF|W@H-*=16^6;epcSpC@XDq
z=LHV!j!ZD*@fzcSCJzw|mdoYqdWMF2u4}q1ouyKqPpMRZ(WMWHB+-qz+<^pkOn=83
za|>@|JxXqe^_Uy1as(#h(wIZcDCUs}z;?#T?m&Ek3iwt}uWT}8uJI*=7?FdkS-Zn1
z5hi1wsa2hc_O`LXQfZ**V^`1VF1C2(-dabe&gJ_0#(H>YEfy=oeSNfj?qgi(&2<(E
zBy)za6|W>(I(83X+*TS}IuwodE>5M>SFNus(w&`0jdA4cZEbCJPUKjzQaNnnnl+UQ
z{C=pCN1o{IS-pDG+m|FD?T(J2p|K;eQA@AJV0-C*Ivcjs*SL;O=U1=K<`@d(S?-2C
z=v)Pz_o6LW$Gc&%cF|_!g&JfZ1;K+5kr*waWOpR?L{%3mB{9Q+Du@)@&q@zVqD&Dq
zO*Lc4BN=O=dopIIB=_mXVed7reiwHf*019<5brhK@O!T2=B5AYs{@{EbnsQLdexC1
zE%xtV@!^cZyR>Dl59$7+sQCl8E{0%3bC2raonf!f6^pri-te8!_-IJaJy@w!-ga3f
zFRH4-+2r1P!C8wpZ{7_0o$T88Ty)VzIvK#&<Ag1ovF?Koef0a4ypQ$YbzUE$&qPvg
z*t9f^Svd9*eFv)#*q1qORwJPZ=#lNAL5&ej?Oe92-IyRu_L!DpT;|F0bdFC`$eo5y
zoBb0xjJ1oDn8DViYVA2;#MC8%3};f<#?YQoLPke8TI0yK(E_PO%z0Qzd&f$o8Zz1~
z-q#^DORQ-dbBMKA45=Ea%@*0VRpu3`%Ds$JTIw`Xx2z}Zm21>0A(#SHS-FvyWu6!t
z9pP||eYd2bzN%PIa3ChFvmB4(O2a7|dVJbbeH@<6Va%D!mjy!`jy%f<9&ljLFS-mq
zFkg)Y0s<nS2atDJUGgW@SNnY<x$g2}w+m^dQzY(Aln_|Zd0gEhrFM%zLsm+l4Ck}#
z*IO`6W8dER5F-_oQG2J$;+JsqgWG=6vL(C}r?`5}k))xFyhG`nj=2V0ZZAJHgJ(w?
z67Y_Uo*>moX0t3|iN^^SsSiijF6~`IwG}du&Ep=@*LmDkt*a)5?a-QHBf5{|^fP<(
zv1gsNwr-k4_2(GfFIIr2G0sv6$9zPJcDV|tcvh}D>7*lgbwiNF@;papy1IIMi=~X?
z;eZe$`CM<MzrWIp97r2Y8FO<|K)S2c)7w?z^1s28U`#sUnDo6BBt`R)zDe9=>2)hM
zZanO;MJTcjQk}C4EYH{>xt6;4^LRj#Q#i@9a`lNP9dUR{3bh~7zgR6($a-nA+r7Hb
z)zc=ppj20R=tQT?yKP+BbvUyd$F-)=)!mLCB-=Hjd;XVB5gYG{ysKIZbs0J}@{Mq6
zsLaTfR<$OxV;pw0U{;aC(qMxEeMX`(bE3*C(gJJH$duNI&CXXubi>-(fA=e4k;9zP
z@@kdug@bFl_^b(-6cQh}i}=DY+d1O7uHhYAwrtg!Wy=QH*!aK9s}+z0Kehg(B?JB4
z+#rXoVXG%;)~tGIniGZw2Fj!f1rRd)DmMS9o`w(kx#hV`9g5x47WO_ju^QXWGU`OT
zOv#dD#o<gkD-yQhE6q91+_N`Bx$Q3JDw5gLETLjgTE=UF29P%<yYMRLG|L*uGhd3Z
z`Gj>;yDVBIrIiKoRjR+$R<oy%i6?ToMBM)7O5RIdsbs0Nc=6Jui~GAu$&~9-Z>hXw
zNqKIcPN%SNO*{HRhkYbFOXsbvWwylL3<~2E#*q1pxA42JF6^<n0TS?u`5g1p>-EP|
ztv=pS)TK}DN}vg9O~n%kqQPWKg7*q_>CdJdTjKxhZwSpft#iqe#jF>iKPndc7B5-S
z*}+QqgETsN8j;e_zv;sgd-W#hX}(#@6D(z~3x#u8Bui6&qdlIM7*}sPJHp`(_QKg%
zGT4$#(2_tl7ZKY_RBjEKCWZaek@CLu8e2|zd5owR_Jaq-OBt1mJyaAS3UKXDDW*3z
zx|zcyY3ka?wy*|H0D8Op7ANIr0;u!HXL4R`=V=S4GFsu`{`CaEfQhG_B751hfjl?;
zn|GW3TIt`#kP9t|%qp|UT8~R-<STL!Zxd@;AX@84YcYAM4t)c)T4yJ!wXYY0P7W5d
z{pE})FN^s!y5;rjd212M(0jB$l5?EgOWJr2ZDbIc{0E&$SVK~gDQKKfDhnItc|O?+
z^;Rl~)eVUi@!Kxh<81abFpQxpn{!ID`c6ahgl4kl8r;H;n9d!?QH_QiGwa8()bO}!
zhv|%;>rI1jEcjf5S?cz;Vcx`oVl*txgR;NIPGsA*ZMx7}NQ}sK+vyug@SwMGnA9l9
zb^7TPD=Zm$;=OZCnzB6MVtB#_c+>e4{P&G#)r&f)%Ls3FT7w`nfdx?@M64x2OLGH9
zH?gf9mpI<xA`D=-Cc^EAB(01y$|VeAU=<>4h*1mQa>|@bH$@n(c_88m4zVw)RtOd;
z!5D&x4qeVtM2(oO$Y2PhNY1J+3UuTM%rLFQBSlL)ps!Mcdo`tOm{qStQg9D#C0}pN
z#}$&X@p36WGAbL7tm%W4K|@8QiG7+V_eh<}ioh;{k`YCQ5v5QN(Onccw;p=p3H5rL
z+g<1hMp`>{BoYr}s-`to3fzeb<nx?=(r?`n3A2y&<oy0EjHRsi-)C`RWcBGQu*>X9
z1k)LGW&ThindC(qVL9wug`SUdJC9x!3bDJT1W#lcf~gyd4zsdoOfF){?w;<7%Y`+H
zuZ6c8wPv#HL(S_Pj$S?Y&qxFi4=g9l**g-4<9d!NU9J-n30IMqtD$Y^=tMDi3Z;2m
zthopbjweX_M8F^;)a21IwhBv^Y{I~%C+KSl>P#AbMU9+Th6-))QhzMg*2deOTp6q%
z0tvD9$U)Z9(KB%=N^f4jWn^Hm+|w1$X6a}ofF*CbHJ1<4tk2qRYR@XKmlwp*&5-M6
zUpyhg1h3xmdL7-8Sc;^QZC%{o>vEAQ6iv2pN2l>3j78|cu&$=7S73M~b_1?OQmcb5
z0hK#kmpjY6f~g}0hexj9EYzb>V^O-uVYHAxgn|7$Y%N?C9sqfpIV@zAsWlZ*a?-sJ
z?5i2~tB{0N%p|G4n}bImZ>C9bb{S8l+VdUoqos5AEN*=Cu}fj13sSVTmrL>6v)OE0
zJR#QRAxx2CtA{#^dRgNm1&o+jb74D`;w?}rq*9z4{EL3TS?_Fe&T-!8e8~BTbFcHD
z^N{ls=UMim<6}NiUh`%F8Yf>LG7aws9A{Qx;%6oxzMA;PnWEyNJ@aK`Zpf_QP+flI
zZB5L;G@D4spC|0r;N_%V_>pZr2xA{Fv#M0dm{d}xZ2Ei6EIp7uldZuyY9@!2hIJjk
zRy{Bzb}8(TdYSj43qANl3v6%LON`3_8?7CK4IBNwX!l^!459gz9R=DhX;aP1u*!Z#
zST}1?&FdRR;u$yf!uDx_eKwPPM0#gLn@UL-*I3D<=*5mBQ<k>UGFpN7B0k|5?5~EZ
zqm&{c!mox{e23EX0=&bYs=cybv$!4sr-YA^dk8}nN*KfPLsI+Lp@DDocGO|aZ++wN
zJI00f@+A-D(I=^cD)~qc<<!j9Bqfi6KWUA8#TDT!QxjZBdY)@Ar>wz@4Us)yZOHh?
zX;ZJ@Zj6>x$E)Uak6f-^BE0EdhE+Cla=amhdKmyg1r11hE3|Obv=$DHU<r*uH}ZjI
z%y?9xEwjRmfD)Pl94L>HYME@7c?L7Crb#$ntM$m>(PY^Bkk1wcSyK<4@1-B~P+!jN
zPA0X(Cbc!!){dE@_BdRH@$m1)0)7l2ygpwD8%S?Fn?V%xz^U1dz%k)vtir+z+&-6A
z<Ok$LKdg-yOxOIV2R-f}=1AO=;DwR|`^>R*CSC1YETV{%<4eF~@uAMfJTDN^*g!>E
zS~$}2y0~M>7KB3+tQB?0tC91AK8-D5*u%a@j>QA0Ep?u&tELDadsL6dFN{g{eRhvT
zN_?_~qFG+M&4h))UVUWt1`sTX!feCqTXC4HJsKsAJLHYUlgU`zACo&TA#;4v-QBC>
zTt|^z9jCxjocVJLLUiXjA-8Uca-%Wib0x~<LauO(!GrA3{44{)&}o{pEEIHkM0a~S
z9d@~WURDpRQaB@U=|8w4lrr}YuuWc?!H-veCm0K)Qh`9Zvm-)?!90DJ{TV2SffMNV
zYiu)gI1olalJ<y9oeoB5qT3ahZKylJDnlO91~EAV_3i+tVyp|S177sQejj9D@aT=O
za*rJnk7c~BLXTN9DG&2P?neYPNki-ks9J1J`AAdg9cu|57w>H~ji+T^jk$RT?e+Lu
zk}j0`GHx8dA-n}qUCSUb>|$u6y&O`>!V-$f;NkaDh&O^-%^k<$9OF}Vg<=`!_j(2j
z+J(_zA2^K#<XtP=e8Y+~7)+(vGP{=M^Ssy4OTqL88l)nMhE5o2JX$|K_qg2t2*g7H
zZ-g;v<G<@U{!3}54tl)2`AT*hAu|meI^in`G0{(wV-W;;Sxp0wG?#+4UR=m!@~%r6
z1M2ejc^PUqhi;tt7g=9^&pz^c?#$e+zN-G6|Aq4#Sy#+lHoq*C@CwWGk?rA~@G!%v
zcoMgi77Il4MU)nuk>OWnffcSRTvR5!9OX8)qObt?6M7y>fY*%>E~jvCJ#!2v=R3lg
z<vU)hAb-RDV4CUdC|X{=X|bUjACoZoVr)0WqDQK+ECQx1JaQr-7E9*NpKpH?P`XQc
z3b%_%P2Vg=)Pi(I!_6K9M&s^c0Av)L77?!S1OgaXEg2l!I5_5HGXt<Nt|*LWTin@P
zceWT`?5UK7x>2iQ)fjf1l2~7$aMGT1OG}Ys+H|UaWoH``;UYOgh{ybtPO*qKS-C_e
ztSuP$SVu>P3oAXTk87)iZQ)S1)RV)Q%Tq1ndwbW8@?zPT`(maX^htXi-pz|)N1HGP
zLo^O|cPQDddpaw;CbQVH+Qa)mqF#R>-5vAvw6;e4p-}s9pjvECr+b5Kk_IDaYVTln
zxs$Rr$84&VS1ly^x|cZo&p|l|MCuc%m5+lRCACWqe?VU<%4F#fpNx|`7LJk4$NWC0
z)Y-q3mjxy}a{Pl1c~9J9{0GB(u*JCm%ChDnjI0bC!y!W>OkZfm>LPq5Oa@lQzj>xF
z&(Mll8OMdD%<+@gC)PNUuWXi^j9Bf|dZ7|hXRCS>_;3bVc(+}7<TU;InyA!juh(x_
z))wgtW(H0?kvGigO8<#RR&t;+nDHhHOS`#^%^|^Hv5lFI=pT!Q${j287C*N-;)ztM
zCE-cse7<5N;yg~*(~H_Fm7X!Ze)ak1uU@b9`qEIaw{=|W@!>Ph7#`Psd8|Y-kwy6~
zE^6^MmuwacW+E*fZ#>7@?d#XKz^|gwLDr>kBx=rAn)~ew)RljC$KAxDEYJI~$V?SU
zVLuklO!}Kx#QLIlKMphKV~y>OU{L0?%)natWw-4+J5WWrvZYl=bfdT|_8fjhPtxD&
zi?**=-qzaMw!GRJ6Lm<;9Z7WNlITz{qe-_ozEDah<JnZNt|QNHi1KFPNZjMjKu>bZ
zUcda-|M6HMSt#VodP(1g4Sh?OWQ)E+OIgl2R<G{r)@|{Y1aHV|jbrZ+j^%ThRB2a?
zbs7W9cI_Ol^}0Y1+W`(@BbZwrgyZ1_Z)|xPt%k?goyxJB9L34*KrEm*e)Czo)auB~
zVn_F#d+xcHoq5!WOE0?k)rTFkwD+^0{cPKg!pryW_)9Li<m0b?^*cU%>ZyNziVt(9
zm3J&Z%R82Hs*CLl+g9OYf7J?61GxxJ>T8cU;)vV7^!|(9al~C;xbd8A`9+Hs{blhy
z|F>b+jW<5<%(d6fy+<ECcQ0EC^iGaycP{7uU42x2mj9;qZ<KFkf1x33>xM0}%pW`c
zP=Gn5GCUr7FpQJ5oFNDY7rJBa3Z*pXr}(qHO@NswzNUX1pII4T)5Bw<<Tj^fK3i{!
zlBkJu>!xKM;bjEEEbVB_`b?T?g(x9sBD2=ASG2rU{i8r-k=(>NipyOcDcCBPoZE)i
z%Xv#K#eD<zob1-ve6eh^^+qGhrWiiv3+DZSC~rhf@=}`Rg^rH+62y(3QY_|oxvObR
zaii&IqKeu@m@caVZz+hUTU)ty;;R=sTU^K-{y?g|J(qLo>l}=3<YlW^p!dWATwjRA
z<u}f2YcS|N1yc)=0bK5YA1zU9VQDax%p?;Pk9#?{d@-W*$MR{ce_Uba7VCqH*GV8*
zrqihwm#eGN)#Y>N@+0wFo;Op(Q+{7tYtkqkH1DTnvLUN!ZYE>a7Zzcv1*H*Na(6iv
z^Ms6s^r$p;L|hifuqd2vL#u&3ei*d0Be2pl>=tEis8@2{$^SG*uly>g1Fisn6?%}-
zmeG-YiHzh(*>ZDPP(`68X^hoPMb>d-{ex2(4sAEvmhVd$PC#A)Ag>E377ExW@%H9I
zzSH>+-uxWr<%98jq0?FDCt^XeW?t;bDPfqBVzB~>;x?U+MaAES(%8va|I3DxZ&*sP
zbK`)na|nB)1HDg}zBT?ogXvWldNkSf2z`$uy&&PRA3~q*#ju4tl@R4$MOe+Kjj&X%
zCz?vXgDz$NrXwk@xlVReaDt(9<&sKDoCS|IyqWIh8Ot4*D0X!4FBu$^+3{35jxGA%
z^)=Ij3}drs-ica7KP_Bx9VeX_-q{;2R@-c~&sGC%E`f>)TOP;udnAkKtjqgDHtL>@
z8-Mnp#!cIe8;_Jv<Ebw0hkmwk<J`|S@=K(a$PaF8+{JC`H#@NSQavaa4W|+gVzhnG
zQ<LW)1PY;A6PHT0hu7cw^t+;ZF*j`c7i(m3Z=nB(kqU=6KF3waF76*2>tCEbQ==_9
z^XB&Qn#H_rc=6&j<#w+p6dZBsc3wOc@_3WH7Y6yPFu3`<-`xzJT$~C|>aQU~AE~a^
zkFZ{Y0@;WQyA#7O8FZN#3%{bs6ro2Ne(+Za$t>37#?DN5D|>wO8`3kwhiwMYRvS%D
zqgbg7tw#nynE5Jd1NQ6uG6iAk#y*EbVdGNZC`t)aposO&!(nb?^5zE9PAQz}&emIs
zm&x4_#~ZP*?sQWbD;1lsX&%#nhKf>R9P>HqpefRL+e?%N2z|25(0J*H#R8>@2EZ_w
z##Ta@4w_sx3d3pXN5;q2Ls|n)P0OjpC`!tsvgweoAc#I&TTM|QqpcJ1#>7eAh$pvw
z;hCsnK$Y|^?X|r`p(F3Ipkkw%s($e3qdz);l~7zPwZbvP6Syol;nKPG4*543Jn}CS
z0_`2SG@Cyz!K2mVKUrYXg-*~JJX&^-o^(niY%DZg4liP48_Yf(z8>dQ((ZH$b%B=+
zF4mbM!y`<}K3hZ~S!z%QW1_&46XES;D0JCHbMwx@1d~l5AD<6h0`Vw&IXdCrJp5m@
zy?2~sXLaXY&%IUmR?bzoa_p+=+|^xOUER|YdU}$CW)vn!NTZQ7@(9EbB#=lZctkW<
z0uj~*BNiEvOwKDXK?JY8Yi#qeS*<a)5ndZ(%xh!t+JNoSe$R8OdRk&eyZ+<t>h8KZ
zJmJLOIVT+Q$75m{5WRq1GjbIBKv_mH;X3HnqBV*u3r#y<J6ThBB0`tkhoezm(II`9
z6vBi&O9Tq<CAXm1hFlAYO@$2LQ?d;k3qV#M+9?k%xH%pfyg5;Q?u+nBD5>#)dI%Ju
z5sn3em02d#{9Lr0h{uqlxtbt9xnVk;h{Y4-=-fP$ZnhE(#%Ltf8{P`m!WWOBkzquU
z$H3C!MHc}l5A!CEf=}7Tdqjgqv-QQG@-TE$o=$6Yav*=av)21@E87tm41Ff;BLRx+
z#<=0LfpP(eCkIpr*DY2S%pCJ_6)WNA*^&~wASw{ej>2w<0+dLm)Bc1!%RD{-@T<X4
z4i4<~k#)twl&>$1O^F*3`Iw*zidcAPOj{cx<jEKYa*v=XwvZ@1T2@4TLq_vMQyzq-
zyvXwdQvdvKxbPAKb4A--N7D^0+~p@-5Fp!$%XpcoCIef`XiPSy1BecHs*L_4?TPzC
zYu_?)u@R2&OEOiuOgyG=uP4q#QPc%K+Zc(6aue~SVb~Rez?cSXeaqa{pOKy|G+{h%
zN8kNriXydPouh}+^aiTZ>6lweqN^tI5@J%+6k{~cxkk_RPdRcqp?dNezdO?!QlW{K
zE0yXu-u?i~uANoCuO_&VFHJJ^&BU&$wI8lY6raet@&F*Qbha7D8nk%ZgR@`XUuU3^
zbKD+r8)noLThFFnVVNsv0ygRk_u1U{(8~@sY&GpO+sxl&2!DsG^at_J$iT+1V;XT2
zBy}cJDv4zZTqKqsKyZ4lLR3m6tZPBrjFbNrxs%f(cx7ZpFpU9YGV|M}@6Eo~zV@{p
z^2}gQVjfU?M67pAC4CRCH$u4*e<CgSkTV(5LckDznY-Kpe}?T%`LSCuh#>@$l8#A0
za2s-V=Kxst5E1H3FUkNh!kHXQ8d=6d8L}`)khm^odgB2Snn3D^)DUuH_+ZOqaU4;k
zXbj?JF-_Ux7m_5FP??NIh<4!Kc0f}ill=Y=pb-@kqN}j(D9T$P2Xie9lJfG1;Q=za
zg`rtS*azk2hP(}V*LWDQF8Z-}^7@(-LkcN7WXGe>Uktt68|nxUmFr493T0+8BH(+;
z(4`*of<hAU8zL?Dv8;Ssj59*E>{>M#*ck?55RIjv{OlHF4)$mktyDS|1$q$P6$n&o
z;Q(aVG~0PG0qQGBNf-{B^=9V8Jjs>eJx12+?XUvG&8k=9kzS}w4yafvCA10?DrZ8^
z_0>}`#QbvXaJ3$S#DFvMW(h(RZzTJDbHi`)fb=`&QqXZwRp^uuR571o7K85ojSxV7
z#T=Hvm+6mVk+v{?(@7jjlYuM8S_BF7hJrR!bxZ=~9Wjx|BLo5q)CpU%+*J1kUG1eV
zR|J&>NU&TTF5+WDNE%xApL;Ct$Zj6Tj?mO%Uz12PZmr38FvR2Ya5<g<VyY>r7xqab
zTX90RB|1nh&$>?peRP~HOv8T9iT{g@Wu*ORF+(RyLB|ncLlAoi5D0AIks>va{1aLx
z#wR)Of|h{~?(WcMQ<Lj;a=J<usdZ8`*=1VUn{l#}Q!C#25&~wSu!vDVUhZ93vOlgG
z&6SlpPDWcTGcIX%)A*Jnaej3~7UYj_t5RuZmQ3#WRNoy_X#HfTH||oeA{QC>C>N26
zHn*bbme8iYtI&wd1h}3J!W{Yt^KNjCyd9xA94=sd?baiJV*JkPAKJN`9mwaJ_|9<X
zW}Ca4%|JHt>_oEBNG12SIy-l^CSaAZQgfdZiZmvbZ@1*4o0{7F79*XC94cYR6pSml
z*1yLaIx{!8r$ZB|T)uP9JxJqfa&q%VqF<(d=bnHcQZY5PcQ0`UqPOwy(SJl|JxBZ|
zMMhJmSZcJT*XD_J=LitRh0tKT;f-|cnsM%HU;ElaS9|+cUUfSy-fWiN@&WI`yS$fv
zUY#30TfKhx>~|lSnGQVw|Bpt?AGqh9yYKcS9{u-6KhF^U3YiyE=*Vp}k##Z`9bp`u
zB69RL&-1aSzt!_TZ0R~Lon>ni<;X7GhSI_~v1;4PgAU>*?5{6z<L*}6UF_@#SH(>|
z+6V_@!=Qt^OrMD;Crr}@e!&%NM{HE?CXgBIu^p2wqDjEDb@yiBvK_AHo<}FzVkZ|n
zmU~`?1h)(U-M6}Db(@XUjpkPvp^Zwo-hL6pTQ3a1mnI2SG958<mWAEp8AQHS8dj@*
z#@ztQE6471@Zo0DY`4v~WJ=MhlUgigQcg8G{4IA?{VD@B>CXkzL-(*>Sz1~O=U_nM
z$zwG7*7S7<0E%0;HyRCJp;WW#mFKT5c9|naC4FLgu{%FsuS<+Yr93;ks2+4k9Lt>p
z<L^7`ZF%@tB$DzViR8D&_xzT-C7IYYe!chcF8rR<PZ?ADAM?;so<hIkY4@n%AvP@e
zmS_^!@siun*Qxtz<$NFCn*-^g5tMpMxxBQvw$?q7F8F-O#Kgoxx7+P>vRUS&=h06Y
z8|uqMApZh$Qir8=1at%y8EJ>oFI&zK#xnBMNRSZ29G`CA6@6I4=K=NU<Cquci`$26
z)nbNRhYvP=mxI3`Ge6f`kkpuX=%65Ge0F9@c$LrBZ8i<n-zn+G(OAH(H79m2s(QUU
zI(2Dg7H^%L<c$U$XQ4MYpLwgWoc&X)Ywa4+1__-ap*m>E@M0<{JPb)X(P-}45s56!
zPm|4bYO>?^FE2MI(7t51FPT~lOTa5BRpSzzUTd$dPRZOBoZQ$R-XVTypU0y(j`5mF
zh*nBA48*rSf#!&~rc1(fCazv@M9|#4T#s8*_YHqpy<qsuu@I7bOufMUt^+%){@~*O
z?&sLYoHP8RYC5B;nap$C@3h)l;MuqHnOP`<C&iOXZ~eSE$yphoG7CmqW2>bj`E8Ow
z1WGI-{D=~Bwn)^hJ(d!S&MKe^^k4d2PM6+ZRaXu_s;<N!4p)rlh94a-{ZAi}cjOv~
zgh?qG@fO2&rB1$&R4_DwHMmhvXAq)dsaC|YQy?By(@i<vY>04i|6{M^UjIiO3iJ?s
z@1Sdjg8u1z<#NeG8OdILdDG#QC;9~DXK%a0=R;Rxbwb!gG9mCiO}`(3em^9cM+ssU
z!sKGS<Ryg_LCA#bc2F2-3VB6$k3Lekn5<{iokr>Gt$wMFD)3BR@!84AKbTBDJH<nd
zHu`TpTSB~zk^3d4?yg5qWx}yBmNU&^KbtU&iJx(<zJ#0=3ujJ6>!gYiEse7D;?p`m
zCJ#KJ^x%?Q9#~Y9?Jpr)gw_ioclTaUTKE82K9{{lHVXuHHJE-%I=BiNmR0X^U%*%Z
z6sR$AA++Qh$yFM({G6E0m@g5V^h?Z)tog0ryqyr_WWwfMwvW*sxr4{K&VPP#)e&&S
ztmML?9y5fU|Hyl~iKGdrlQ=j6>^9`_9}lGbyZ@ILmt5-v=;`a|>(wq~O+EVm)DqkK
z6f7>8lf?vfA&(IDAy>U?cd#lAL@Uk8jMH<{9iB2cMMM+}uwY!o%G6!FU=Kq)vQyJL
z*9d=GXk_CFi!m09MH09w2t#Z(rqGpY<)oAH;{=GNF5KDgotvH!`KeT5LIb);Ch}kg
zh*L}?W0hpm)G}Gx7jhzoKMlbj&pDm*W7p%}R=~%uxC8`9M~!53{F%gDN|=r|^lCDb
zklEO&Z@t&eq@;amBoPPYP>n{8#}3?ezCX4<_!}>8+y#?24l%tdrOV%VegHg{-Z!B4
zPlh7YFGlYr&{oTJ*aqO2W28HC&1(AeK9ONqbF@)<B>6!uL?yQ~>i!`@<5uC2bE6UY
zH=CQA`x<eS*ibwn+1^9JgetHRc|D)U6t)kPemtIEjFZUoe)0eg&q?(qG;{eLo_qM=
zAIq6?euAs~_*PuTSi<7jvk#Xgdwi)p`&^{Q^EBj6&r3Y78f6=KzvmOYvCIB1l>OFC
z3*{1d{$oNFoL?J{H+`W+I5+#bG=Vmlh-E@vms~{IflhCMIFZ)c%!qDEa}nm*S?(>t
zd;n!MAW0-WZ3?OFVPc0B4OVEd3omF?aJs`-x<CfpPpe?r!7YoC6P=X?!<v#N)LnEl
zoKx9s**5*4CTzlP(c%n0SbfNw3NHx&e{Sy3!36?7DZmj{Cy_aTUy1fFe^IfSDiidO
zsQCyTGNMFjL^0J1<hzpvV^p&0aSu2AT{A!?cpIntd_D>?H-Yg|K)HmYlm3frkn`aP
z>Z3P_421gWV{b;A1Z+ef<xft}tR9@3lV?1*Iy0U050IVYQAD%VH|8h-dx^R$2442@
zhs8lZ6qZfa_!E92`pXYd(C*dz59k@D?lFlGbGOVgVf*qejDHkxf&bI(!7-6hzlTSB
z-{$G-4uvC^txgmQfT&^-5C<w#n8nuHSsiQ=%vzaUs7zEN^=dg=Rg<=LNY}*`i|4GA
zR>i<kDs>3QudbDI-LlkDovqLrH9OhLcq28C?>z69wQq=OCnHT##8nwbu1D<uPC?|!
z7rBOpG`nzLqrY<E@f76ro;}7x+e_qKzi*%OfydI1k46t3JR|K$u8Bv#LGG6aSxQiQ
z=a)Rc?0Ewj0UiML%WTjIA3972J#7Ze6YQPs=7`h_&CcQ>S?(iS($FL0oD6(d=F-hC
zsy^MIw_?#m&eqDmOb&D~*BvEGGSMuKbvv)!F~_&<AhYWGJ6l`E3s|zXZ5iMYGBba(
zKKbw4eER<VhMJ#0k|g1i*qtfW1mr7${V9|Ltm6&O*6W7l!)R^<lDWK5sn@EM){ci6
zZ6=NZU@L_}853CY@H`{7-dJ5b_+|`DVqh_4cr%l*8!*2xlhE<*F0<!is2R{=d%>2(
z(Wd$(WB>lMfa41);te@eQ1Dq<k}d+gLz4G-lSL~RM4>-Oz;)P4CJSaXf-H&2&9cox
zGHD@nrirJ(*cNKG$`Oir6gD<a9Hgv4RSn-ey0wH^Gf((yx|#+V<^+m`VhE#_@%OvB
zpKoXUxtv<Kwa_j1qwT0$#tBBHt3tVZ>0He!V9&0_Sa+!G)218Q>N>vUJeh4W-{d2E
z5L<t!>h;yt1MMbW8L*ZBG~==KTC<4{WoGlr>8VI0$LyxT#7ya{l>75#cyl&ePhku7
z`mk)(QV~L4q~TF>0U{{l&<y?;ec1!6tAr^%<bH)BNivINv)(Y`!0gQC9%6Zo*_mDC
zVkrz{D+JEbr1DLCrdl&psW{iuVKfmkeU&pcGZ8N)NKJrgqg)Q-l!}#h@4C{xuzRl&
z?Q!)h&{=HVmeXr19fP*rGarQ9&;WXxGbf};q;a;IS5JTILd`43=#QkySQ)s21FZKU
ziMkyN372Ag^#v?H{Y;K8L{>>=O~v;{1{I`Kn(vp(7*7}kw41U~G^VGn*u-O`n$5Ko
zr7;+Z8uj|!@ddZb9Ev0)loiM|aB3x*v8ZJiGUa%jM3kyhU7Dj7IQf}p2)G8F;SX5`
z4!u&bv}<O@P@T!$d*z#Hhv_90i;<~(OJo(gvt5SXl(_$k^KWc6B1302K1-T}5v3c_
zmAKCPv)#C_^K{fTS|P?<n9m}}VDFnM6hq}gQcx7hLK%SC+6{!I7G||@Nc~FrxiCAt
zUmub$z)llxrQcXA+^_?)k`a<CC*70FWqDsiB8`kYu(x$Q8_dA-PwmbUTjWCR$d&mb
z+w|KB6hMSWSAY~@Qe-z?QN7Gmi%ZAHb&4)uTBeL{X^iby&F9T$e(E!;g8{J;e-=$1
zrwqrq=8@4|B5?*xWO=#KP)51r?6b!q6}n&Oi!ZC+A@7o-t6Crv5=zGi6?VmgjYhL8
zM{buD#{qZl3n43>ImOSsgVIPqmNZH%=BU@Q4F^P{iR<^&b&4uQjr~YF9yK#1`>S!V
zLRC3u_`x?8^W+_IoXVTonIZwD!LWK4@-DTB9#wxyTPCq>+=|D~gAq`u8Eq&6En_8{
zX$gPfg9{2*lgu!95O5QHfWZBP6Y;EC&;wv%bM_WT7C8`kkJCp~P+}wG#n{A+^hQ>-
zo#?Xxe>#OHB#=tuhi)6GbaP^Qs?E&wwxtdo){ZF~M7U&7n4Oy<cR(IclGi^80w<Xw
z8N6j>9Vcs%k)@eRIfOt?bSE03e;lK_eE<`K{{MV}u_bRkt^`p{Q;FWvT#Nj#+6NGi
zx8{~M7IS%oXVjo{Fp#$W#LUF=&eBG&A=zm(%feNP#ket(%?g~{Z)dYJ1G(1rYYVUN
z_;A*|i2PF?$!Um7ljdlWcmUuWf9`jO{1r4byz=)?pFX{e!z{kI^Yr?$ou_xssXKSI
zCc<6ckK;)9->2Rbe>bVw`JLa^b3x?zbLxI<cnfG!DKjNneQL2T3Wx*^N}%B)H<Fk)
z0T0S<F;B~y?P)ffZ-0A}>AU>kgAe*X61iFZg4H!w`qq2H|L8+Oxl`ReymNV3-F<HH
zKNX9I+p*a0H-df9Jn#|q2=}eAg3Vik>cqA!PGPx0efyt!WMgCF+h0l__22ob)bJB=
zbtE+WctZUeH@H{bKD_tJD{q{dojrZDf0Ro@&%F4be+hfcC7zo+chYV$HAME5sViB#
z<Qs_!pK?Z$crySsu8ZMm3yeha-ZHV5Fi5#eGTP)C9Stva#HBKFtLQNeVaW8|>x|e7
zS5hl;UT>|su$W3M2B|a<Nsuczf`<c_CtAV=O4VXIxwxy{vdvVg`3|oynad@m3>iqi
z1W|8Rrc@Tz;bi&td4E5>dpeo)UU5WSPSnFP>^JdEU~_86{(bxQV<lOgcXIYKM1u&1
z@_A}4pgH{O*|TTa5DFKI;uvKx8J-yb7MZU*z{d%tmye#7GC}o6gg*w;=p(J*I!bga
z3|tKkM*!~(;MQn|Te+pBnHkfZnOR!OweU!a9%K40wte_zDvj;mew<UOk}lJ5a+Rvn
z3XRVXwVY}thlJ+jsukzq2OfCfs-sjb`gg|b%NaL$GMXS4Yg%Io5*sJMlBAOdPrc+g
zco5hnK7Ka^jedAw_gVV^Oh-k3aBswW0F}WP-5av!{_Ng+?>*^@><wF|j70sE9o&p~
z|K6M_FNTfqFCGpXkt6T=_kaKQ_p<hzW_kTxeCu&`UI$(PKlMa>Q=X16((%(lA9-eY
z>EoXe8Ys_bUxZEm=YY9b^#<qtzM9eX`R5<lGy@BM-=?`M2sF}XF9gh|UYJ8Zq5{{f
zP*z}Io(R;R&%z6QXOn3PfT_Tz#_xhgd%q`6%ZeXLdsy6ADKpbCX;J!{BquEBEqbo9
zLHS?t1V8HVi4!M|ALsY@@!|JaX6bfW(cAD|!re^z^d;Qe&7E6>)=}wBQs+Vq2JV;b
zeu--c0x=2@ck0xg(6d=!eNEN;>Q}$&|LXd$|N5`z-|GAHr#~J4lth9*zjpKQ^3xgq
ztyI>VR8O6J&D7LXXX>Tr&Yi2C3-Cm(<0_1==KkFKCMRET|NT6Z4;kH+E0dlU_`#Dp
zJApgeWM;^4l9@xi*@zbpWV1{l#-!l+gr&ht#IBEd0|N(onFF_iUx~&93K+pD3*Uls
z!MohKCkMKA51E&;L$gRdcsYR&*VG?Oa)Ruz>_--Bb!6|_;`qCk(b4|3@%NidbSl5V
zt88_+)mcfmCMH^`6)FNj5n7!=O4ifFAS>04RLFr>A~|WG)Tn1F5mECADXWl1W3;uA
zmaQ_w@v--|-&gMc1wx{>ZY&S*6*BPWxicy3O9Q#Z=m93udQC3RBcw^P-b#Px3GXM5
zasJ^N>Bq{gvmPQ&6~jS#;OEVr=FfLi@8>tEdx!TE&3^ybpssEjFFjh4)I!NbiNuHU
zU3|ymHz2HIY=8W@!|%w!k3V+%I%7@bZ&x3*sMfgz7seiq517#;g49TuVnFC&z{ru0
zJ0G+k7#h<1$=<ztU$*xrv;eBzb{)xGKJ#XNDvkFwDk^vL%{TwTwLgD|d+M=o^Rv7R
zpRlMo`7`EZMpxPthgCY0g7HbBEK(uCGwLFDYB2zT$+zE|PzNLCeib#;a86Z-ASk^(
zuPC_5a2cLKs1K{(h<S$(zv)f9v(gx`9j?!0ghkf8NIQxe@M>+CaLFOu8KS6pv~X^1
zf}jb6xT}>=zA&*~tA)c00Tr50B@Goj6i~;&w*-!pM*|gBfF`vtayH^ttzNu0xIg4e
z;H`_Rz#S@ZG@xEt0M{tm>|VblX54N|N1(X4Hsp>fU<JZr9`nT7)|}-(8N43%#6tNm
zOx?dYAh~4)_)1w4wFYLbWy@L)g+M!E`~t6!3HHLYom^v;a`8HA#x_HM?Ssm9iSjM`
z)Ui18j;ZRMq!VnNbJ9RIdPKdLew`Qo-y+|Uw!vz7y(5!Hqt0ku8Ei7p+(E}tsL_l-
z)ITWR$iM&{Q+rgeyh_T72$|#(6eegiji@|oHXn~C^TcGObCjcwS#ww>*HRc^(2XNS
zqf`0BpfV{IBCuOTJsOO^eVyr4rHs9giVcEtQbs0QET!_*iuJBMay@A}qm^mZq>sE?
z^x3tQu`LB(%Fq@41k*=z1!ue<>0d>nvpq<u`sZu^q<_@SlsEhMTZW(G>!ZI~YBU<(
z__%uUCqMbghY`1#m()1eP_NYXxHV}l8bFAAeY_yKzeFByFiTMz`G&>wpWrQKr8K7t
zx>xUZJy~XE_|N1J)SfJr#F36oiol1(^^J{nlKFrMGP`@)9oyKjJMBGh)Sff(o{j$A
zK+0Mg%y*(#JR#r5n(I3*yKKjL6X`wh=oiQV{*XuVvE1wV0G06mMkUo9>etnmXqz-p
zY1alTF*2AW;3Ya(ste|rq!P_GW_U3v>RPU2q@^o2lZ{LBB6m$*3I?W;>$0TR8Pj}=
zDkeOMQKLVMAo;+vdb*CqXq{~&5O)K;V3iUbDQ(#1L>Xpe669IAi)@tfm?pRylcgY(
z#P}eg;4AJ8<Waaa3ZnZNxW-Vq7#EVE3S|8JUG-^D4LO-u90}Nw^T375;aqjFqF+ug
zv8E3cd#C=4)^I85vs_rD7(i@E_Q5vyY4|Voi>%JbvL}n@pJ7Sv&gI710u)T*$K~21
zjHx!_aS|V%O7N=SA}mMhAZWQ-tVfvR<R9xG$&~TuE*2RyVBC+{5}YJ4q274BJ(Eai
zuR<1ZUXaxKAoeU~79e3xmN@b#L4&DkK1tS}Z~`b>(xBj<6U!hj5OVQVvVW2XicG8+
zQ8`thA~wO<Q=BQLT|u$orogUbN{tz?bapBf0~BWCq)S@9V1meYB2o$L<Wwg|6nUQ1
z-CPPYBKJr02Kz;2wX_B3&IYHA@<4^`h@!kUS~=m;5j-j}Mea+maR%oRNS&tKkL8UK
zF;CWB>T{4G4;ZE`M}SzyLnVeuUolP~5-Jyq`AU^AI)IMJY&IMAffU3(%@twfz&oKn
z1a$N0G4&{#s5h8QL<24a8v89~=|nsLpqdMK6$%4Qf!HKXFHOkz@Na1=fsYfY+;UPW
z;&u`V;EDM%VX8#IqzlDt*sj+S`9ir^@Y#~jM44fWAL$9wUdrYCL=oa-qBF4*BQq7^
zfDVz_l4@5xPIWNe!vH=HV2(r~c;<lV8-gN5o8&T91irV66e$=TiReNBBUwq<k}zq(
zM1p|z4L_DQ4PdxIn@}7aR6y~$Bz&7PK@`842t~qlL?I+ZmISeZh}a=W8II1HbDUbm
z4p5^Zhs7OiCpC&#@C9>7Hhy_oQtk%qKs=U8CaDM)!5?NAF@HMlgyV|(dr8Xw`BM%c
zfJDl+Vi5<Nf0pj{*@kTsCXlcdsnUdFGj^nb=`<W-<SCGL=k+ANyB`_pC-_E6(6?33
zA<xyI)zIz1Ta1MycY?Z;Ri+L}8npSFc_bS<=S8U%c67FMc)izN?#sM%MR?A1X^mWx
z>Dz%YX)#|`X?5kfnEdpvdBdKW8Tf-c$|W;r6pE`w$FX`#OP%lf9H)4AFbD*z)wy2`
z?%8uR;NQ{h7W&4~#LUdzEWNV4w%Yv(2Hs4s)gtaMHvx$LA_CoWRu2PpveH@NO6{DL
z%VoQ;9dgMu>sF)DC0UPJLOb~)bHVG&VyT*(oZE5yy-fmm*JozPI$*beF-`42EHGzV
z6GS?1mwRed;FIhu(uU%hm3QXNwN+q=Hj?IqOKGpQR&~ZruTS74`EZa{S2$V7v8M22
z;X*fKQ~jowV!zkmMeRs38I|2yl>{^@`<10;gApG~g$dBePEl4mh<F$9&y<+Y^~U53
zej`PYA|;)Iq12=ii4^_*dWm9`^~TJbGmcad_ChIZ)o!P}ZMC-0DW6OwOXROJ;-IaS
zkq<+sge2mNm`x>PvAodGh|L^zBH)vmDTzdTHj^2*)92AwKhH?Kh7nlTS_gP8-3r<q
z&bz};0Yo9VN@Fec#cpS#yb2J=()e02QwW=r4+13rHTMAvrbEK<(+~eaC5Hb%_H_I#
zk_67br&MC_(Wl4qM5?H6d<W^A2tj_jn2k>ASD@pK*rg;{u#7}}2g2~TF!_ES$WwH9
zc?{x>AK=nj9g8CrMR>)JH`J3TVGKz~MN;H+umjnQ4Fxg%W+r_QqIL|ERb1;Ru0C7^
z)c^u_^wzxPvgovWXl$>d0+!8<rPQ&O7&Um=OvbehOYA!XVibib%?73XWr2E~hj^Wb
zeSO8q9~0e`o(>%g+x{=XoUR&gxQY#^UZBIwSp1Sh@74%0a<+3ifUe<(wB|dcOevwW
z!`nHSlV3MJ`E~A?cXh9lR5%_vQ-z|8hqQE`9=;ui?nAs2ZVnyD>R5f2h}HPX#mK=O
z&*lf~ZjMw6IKz?=68U8uz%3V@*njy}e!vhU8Ic0}V)TXKTS1whnA*RszNCHw`^@q2
zvx-(EPs8(|Bk`<!q-FIKY|{sA29`rmpBy6&i;NzSZYD+9Cb%>x1%k7DS1G^(^~L@@
zGjmz;G79Pb+=jVcuZO)c*LCpac9|D>QZe}Z1VEImBUAXZ?z;H7<BL6Qqf+tR^(1(5
zxtu4yl{CP3y}bL#!uw=sJ;RI4Bk#qhFu<;M8an@SP|0M)6e{e>Q!;hv3Dc!PV;jth
zYzAroj!>55j}QH)&yx+VIM>0Ujnw6Kq-=I2L7P!a-%Hy`JIc|>;j+m+hNA}DUmvc1
z%@?N7FTrpfCl@5F+I(nbaS;cm@uc3|oC8(~aZ3RsPkO`qe7;?~pL(w~Kfk$mlKj34
z^MJo6jSyry)hbsa6jO?m2q%@Co?5;lTCLWrNI}tZxk@s$$kjoln=o<JAv=_*2Q05&
z{!}zYN^pW}EUQ>(6J!=2{x}h@R%Z5Jwv<T3VqL$HkA!<+!|C;AX3aNE=kk_aug^^9
zbNZVsfAItWk$AN_k+OWr9$lRa4*y#vLb{hizEq3=O3vph6=bv^PI?5REQk_YB2NNd
z=z+_RKY=f;039699GBH*&?0ER0qqCBbeo?+>8?JkCZ{&3LcX?k-PGRxOsyIY*Q&Gq
z%`a|FO{(kG)~HXeoa1{p>-$dKuy0@dzTwAIxkPk)cd`TBzox#q_xOGBefw@WwNJ(g
zdf@Njzxb5oaCRkb>1+KbUGKV>1Ysh*s(aL7O*16Dmf<-ypKHt0a)uByxA*0oMn?c}
zG8WXiOqHKZT2;FX>*4U?<Ww<Ju8>HNyx-YOAZD9*i<<uR26%1}4gn{Jsyh*?khJ2l
zN@d>H!02F-pedK@bmm^Y6eku0F0*^0DHtt~E#)3EP&&1OIfXd<30&&w^m-<pE)`Ak
z%8~Sz{2TL5oAOzR`XN7&64aF4OVv$ds%0iqw%tH4PU8FcFnVB?T3D0XYgngBq{iVk
z(NSGv1!%uMoBOy~2#Fpov{>Msv08{VRRAPOaK>lUeXz>ZH6NvXue&OnMt>)V-&KX-
zcl9ogx-(l&rLVg#ohoOY6DOW^;>7RZ%_zg)F>Gv?FY=y)o+q%Yj8r-@JM#>pQq%rD
z$ptO_KrhKR(~@QS>%z8_yEf-a3~R=6#G4_23TYyhinIdEry^N<<1mcPCDYh*VS~!U
z$9tKi>VeXa;Z39)@PT><H>Fc^H5iRyUv8>B^V-)Wm>ridCh#V&!5L2zLGc>TJH~!c
zt(?k}LwVhtj(S#Tf0NDo_4`0D8L`^V%ez6(o0(Bw0^ZK0%ADWp0o6gH9g9@#={L7L
zts`E_`Gj-nvjvAO*Jn=;Nd1^ViXF`rzM3}(NElrU9+O_kZCe3HyujaR3ybbC&)Zt~
z{I-MtV`skd|EKFd(G%fS80!uJYQvqxOZ`gEJE2(kI;DYD20r;3+3b5uP!FLT*xG4z
zd|w*DElx?z`!xznEGduOCp~N<AIP3|r!x>ch^<E~%7bPqwr~;mMso1jx&S5GPOly-
zckI69mRnY8bre-~)I^R;pzeW#QtKyH=NHO;@JT87B$C>Zg?gGm+<-y@V<QV}JdEfY
zQFr4yEmvC&rS|?8uCTW&XrBBobrHoPgtWZU-`wo4h(x}9*>Ro6h2GM}!sOvLiHJ#=
z849gvh=w?@(bWFt?vp$omOJ?`8~*x`*KFr1pZUyZ0%Umm#0!Jr>fGl_-R{oS`KMe=
zXu80?^>d76?;kSd?{&FOizavba0T0+bejzC1f7)-zmdSxh3(+JQt|!!_g}jTcj@<!
z4;nN_yS2XmV=sUC%UhMngG<c-IaF$a$;q{q$xdh{SS+?WrIKGAuMf}NQzB3_Y0m^I
zsnhZdCi$x$Q9s6p8YO}XLt#n_hU41cDf(NB2omAn#6U=5;lU$oYip}Fo$cIkPq-bA
zNL~-(5ObRVFm64idavi_r`M@tzic#`r9vd$&LtB6TP*RKzx}R~`t9>${`R;*gxQhT
zc3<2bSYs@X`kt||_nF&U`If!=7Iy5IoiV8ux_7ZsA>i(wA9z3ejLl6kzixRBj6HiL
z>HPG;eY3OZ1^A;=S0{7YC-xBNkcW8Z8>z2$m-xgO$YPX}KWTbRTX-5xxA1w=IepTx
zv0LLa<U!@*&V}(~Z!<D?WR5Iq<G;3wVWf0e0K|*NAONl&M4z(-MZ$`>!?rHRz1&T<
zXFD#_UH(=Od&**yq+vH;jzDsw(wSK#v~q1amdW#N)Xi@9&?T5;!J;A5W@0mI;cyBB
zSNubgC9x=_b+C(kLH-p<G>EAUU|`pukc#RFYCo5%)sjy=2JSh6e)<Fc#pe7%|0yIo
zNOF5D9f6n^hG`a9=4S2qO0#Jg&E`tHEf1r@1ry(si6Q0&YO__*ed5fub{svZD$wj;
zVhuPRaziME@ehj@<MO=EGVeJGA9|Lcah@P%M}rxpTf}y|)ru^Qy#MZK5;@?powC&>
zmKO3bkXT#(PA+r48#V+U_xR8V!JhpH<6I~hFZ%^tv@{V-N;FHE5w~a0^psf!e3VT_
zTT8)E++Y4Raygg|;3G7+_o#f%#PCOSwVIAicCd@(izTT7e!(D;S8)$91C~s3q#KkD
zAi<qek1kXzB$BBvMC%SfCKYr9F>Mx&tFU|(DWyY&Bh@G>qZGxY7>JS8-)Krw3Wd{Q
zEcc2X>X1}EenR^bM)EC2FN_%FOUAmZ_{wCFS|d_hWQ$4AcpNC)W_&0$pfl<I#Vt;N
z_(wk19ROuXj*+cai6v?{M*Mx_Jx+tXL~6&ywa$;L_|t2EqvVA7d00&RnN_9=!Bka>
zUT5O!MI&!=bs-d7ko{e32sq04r49Z?oufT2gF@kGk)ibX*N}*>9t0UmEHCSFSZNx0
zTk()dW5{^h%CU$=;9pd&w$Q~A)L1rhsP0#P2eeVo08@i`hyb9xarO8Jac|rHPaiyZ
zP+N|wwS|N6_#r|&u^C*lBTf}Jsw|Qov09mbKw^guY#n#MYa6KY8obL$mRok^Q<A@k
z@!kDsCxIBTsNf~9s3VMcYChY7__WzWUP<~`gC2FsV(g`ukC~?+m9e9{ZenaKsJU2z
z#m^j7RTcZ_L;CN<!am37Wo-eK)B-q{Nkc6KoJ8$QtDCsan`7#skMbwjaZe4}e3K`P
z$F>cYYl93>M+ME9!b_WB?CFNiu%ZW|^prfGG``;F)~#fdl#biBojkXr4gJn|Av0c}
zc;eAlg4#vxonq;gM`iH`daJwnMO1myKPvr^e?9!FQ>R|_k&k?&H1$s`P4V>+zMtFc
z9sbBic>eSBD?X~@`!t^3m2t(ts>ywZ8np};ZDpflngd*q(jAXUZl#CQ2&*c+*MJ{M
zEt@_kg2)ho3mhV6jW;s%`o9ldht`Ck>V2}H$hXjG2>yi}`#*K3OPZ1^bIalQf6ip8
z*kdEv%5o$obA`|uN|o#v_OXvI6g&yWdFRCd1|K1)u-+g{G4Kb$o-bc#DHU^!`&eO5
zgZ{N2{fx0jPupbGyNsUwTh9+Y|DuS)8Rd%vtf<*4{w*X>`GEEn9x9njgowN395Gc7
zUwL~OrDBMaWz<}Na3ZZ_7W<|aN*O5H*Txz8f;NSo0N;q0kU_8KxtuLi80#hzMQoU&
zTD#vYJKPIqg0Rd-PZd2~^twn%uo~^mm)?XzxQ{IpZK*G2X7?lKx_fNK>{cXr$AO%<
zSH{WLm#h7g_(ZYezAK*3M{Qy;C`}u%I{0Vsx#0H+VJvyd96&g)Hx`Z57T0WWc9>R^
zm88E^tNQ;Z({7y<0ISdv3@qfg4F>9@IZC9E2qfumDi!w<wGdK$um%Lxha{;HQ<^Ih
zA7I_hR;yO4H{dje7XwWwY*Q0E<nx<x$#0y?MM>>QdSxrBbJ#1h6^q(Fe;^T0wCWW0
zz;N@`uz5L9q-^bwX6qEc<~))?{&zvu`5q$m^IR~^#YIR65k({p9-rmM38b7GQW<<g
zpi6R%bQb#|`8%RkoVWiNdzBi#e6Go;hqA=%XaG-)@|H6gmcwibgDyA7z7e&&VTpIZ
zN5|#4r4XWERVrM9CD333kmaP(k3`4=OIjogm6|&lww=!h;1y*5!^W(P*2F@iQpQUc
zrSXzH(meTA&D^i<+O;d%dj|p-&MxtU@q^O~yF~@N#XGfZc?neOm2eeTGm?&#JmK;!
z?gQy^I-Frgax&SdjmaD!vl<nR-6D*x3<RnwO4y>q!K{x)pbk*x5H2V}s(Y5J+Q~?k
zBW2ua>62qsx0EH?BbzNvBcZMIj~!iI^&+TEe>ecHg?ErfB9{%h4};+VD9E`hr&F;4
z?zC7njh`)*F1?h@OJ0(vCf9+s(>kGKJb^WbQs@+hrXohUR4f!H7Z>*(<I*cDN4>s%
zi;I)r4gqEt2dN~)Fp3VY*jcZ+^hTpKN$%E2qJc9mktwEAXAytA1t+yDffqn>JB1RU
zh?Oxh)0$7GpsVMx64S&ZzS0BU&#ky6o_S}5Nvb74rlR2c+^MGj%e3puG!qh{KU&~+
zlRU7lMS7X+ZQT*l8NL<WqT`FWDW1d>L0q0tW!<)NDWOOmBmy}}f5cFNv%Z5loAr^a
ziN0`etpAIo?8Fn^rTJNjbO;8|%(gL*Gv;Gyi_sB@+%-EpYiF|HxQTOs07qx%5(s&m
zm&4iy$Ri!e=NzO*+|4!v6B{i?ow+keqp?_QHd8DWeDQczQZyne9_MOg*aJLcMMFgS
z1yyWzHY#QivOH$vF~TYHX?&JmI~H%LGjdimnvKlN#mIWd{pIoW*)g2J1sw9(=W<1h
z$K(0*9p@I8z43%Ix*s$o9m(Z+zPE7Yl!*1dRC;dq%=l#(&VB;kwaVZm;%1-vwLGAO
zObk_$SD0H^CIOm$fk-4nqe$^La`nb1JlH{~$zW`#=V4c`BO$Dk^8@*Sex$pY3=%Y;
zXIDoU{R$+Wj>A{eABzSn(lbp#p=%Lh#7U+1+=;1kmO%d{xi#zt-@^C0s#afk?+2GA
zTE^aACm~{%=#xn13UDmPPGC783{^c9?Crg)wc4(A457{DuIo#?t#~4_TYW}-)*^iT
zjz7-ym$7LK-=5E{9LU^)126}nIrjaVjea2%{KsbVqdO)$%%}GHu8EeqCV^{v!?LS=
zRDJO!jAXh1F%(rSbep{O3wUd|B=3yyBYCz(2GRkFO>vb7M;z3(G4ujhLcyNu_qR0<
zlm?QsnM!G!cNT4xaB2N%f>o0i#oI!&xGeZ>R8TI@&-o@NCufL%Co1QzJ$v?;rLy@n
zQI^if6P;I_8kh5PUlVI^@zlvYM5OS%dNOgQK`esVh%eP82!Sx`ml?&9TE1%zkEQpr
zG8NvwccGBhx296-;SQ;m9ts9a#b5`80bv?4+#viduz>x!P7P$twVH0YzOhy0ja#JE
zMwphdDNAE%#1@8oQ{GeFQHF!4HEq{!t0z2_ZkXya{h9h4(?*P7jU%m&YyK4JWNg}O
zqUQ?<A6-pEgsrHxYxZt9H)0(xRMP3=Q-nPc&eEAApk+Omer52${##sg$MA#5p@C!u
z&{%-<w&_$Y#x$^F(b$KJZhTR39A`APYNBpq)*J<{8H-C-bRZc>bvuiNsmX$r+AU0~
zt5sxh^Tx)=T5?Y82!#F;+fK?B3<4H6h*_5PGC-dM3lcsEUKIGsc|)t*ji%j3lk<$n
zn<cx3J9c!NbTp~lYAbMO!lqokBLiDQP-D=sU$?GIN{B0GN{dNNL)vy+K1vK?ZqKkE
z1N4eJt)aE=i6S6al0KKx0m+oONcpa)!ue72++=o@3`VV%)Yv48`xOkcQa+LNU3)%)
z><<VCuSoEI1;b)YXfHpRU*W<rxj{G~K3B?iCqWg5|HnAC?G4?Cag}Ph8IoG<e15fp
zkAMsnPV4me7`1Vxjw?P{uP5U1L>+flIFfh&ZyTE>;sCDnh^FA0m1Jzn`)3_#pRwqe
z(vm6?EE7@8NOAxYtULH*Zo0d~j^PeOcc<P#s%JfQ_4pJ6*G1jEw^+V~et!kTiZ`ah
zyLate2^x_im623;&&D2+JCg=@NT+({2;iiuZLM8-#a_fZ3l~nTzn6JLiE=Z2aXo$b
zCU4Q_#b!iH8BwpVN`7axX{u5nfIO&@ahQ=gXGfCiu(xLW2~x^rkUhgWE0NXXj(^Bm
z8^~F+NLWbw65=D;o`m@@khJD1rf!$2NhA&pBFCtWn{U3^KKUDwS#rLeI%QvdUv!40
zo7FFT;uD_;o(mkO7R9Mg27W217i5e_`RkJx<uam~8}yRoDtijvk$XKK^L*L!Ezb`;
z|E@S@fPi5yrkjnvL=`L#vYA06BZ5Pk9@JL%=6D07#@zsdQ6I6jVPGNYK~%;YAS>e|
zpiSgB+jfod22HE*cFBqQRCh<mZVQ5w(`~^vK+qvTn#hTsBa-6i?prr&0-KPjCuhmz
zb1@&X1$wC;%Ml7Gr*`DJLi|Rr)^XhD+}hB*jY$bwPX6$gi#mfx-kX<2H=g>}kP0B!
ztQqhRtOu+RFk<9UWdPyXre5%ZjQbHI4%iy|b(I&&9iT1<$V>+%BjKC3K^5gn@~d7D
zk6K8|B?p(npEhtS;&5yX|3m{$xEe`BS9mB!D`Kg#7}orEmmk`@ckcnO_rPB8TIQVZ
zX$M3kJ@7PN&iAtYAj4ngyUA=eP5EP>udSN+e#zlz?%y=c&HX&n!TnzE{)6Ui3}K#4
zeXLgV`!_!BGqBE49^`%ln-fPNY}-2?@EMMEw_NpZYqzon%?YxO+CN1z;bF;6q;@L%
zHG2guviCJMKnv>)`vf4j?+rYueUH0*4+di9!C$DCskeb<ajjOa5tKfYkLxDW#AEE3
zgM3;Tq1ndhfO-tnxl*puZl}^ttcK2)ybQ;Ed#2vbWQv8w)hvl_BBtXI*Q<(U!)Ucu
zzJwvf0)2&R-=3H_a{0@jd+^}g?4C5#eF-a<A>MhC`()DP`K6siY{?uF|KjJc>wJ}{
zoCa07-AvX;1plF6scnXrj5!hAv&;?AqL)RiW=3IGY@#sB0@GZ!^!blZL`oNp7Hp@V
zVPqoryIV5qpTGb8?}x+R4+h=;k9_chA3Sp7?2Si0_=N#KU(l;0MUtQFc&n=2ci(+}
z|9w{YzK|^ocIdvab^4Aw?&$4#DnEOAcih+F=SkmnH{Ii!gznkCi+2^gwFmk1738El
z<#`wVf>(6ZA#IvpcMV1bom)Z(xJ*m=2+h`Vg~+F3*+G|Rh(6(NcB#F4m_7~f48ljp
z<ri*;^Ix1cu7n~d=<b$%Tv?x#9o<8ogpXWYUq(I3+_O@XQ;SR8DJG<8%y&A(%}f52
za;ep9ghGwR_18~U%GmtljZA5F=Ps<+oypFmxepU_pd2M0wb4e{C%Qf}xwO38nn3xU
z35Lo~_N9_20`Uw6%y6}w^V|7cQG!`174C~>$he3&fzPPc&N=wEi3Sga975)p^dzJ-
zYrR}J<o#pxBVRh%Bm5jSaZ2V`s#Na!yhKk;G(ncLA?I}P{-@s#5`mwcxWdwQeA01n
zm7D&_Y#|WnboPXN+3D#vz$)=X#<GA+0@;F7D3FXF>n0ylB3)5Al7(g_ovoG&u`miu
z0d2>|TS!SFe&!N^Ot!%M3xntUF@7y_AmD!$`G!B;lglK{)@@90NZ+D;jz<Ad?SV))
zq$84%AtRR~NBU{kU1OhY)BV-gnyZ}MY;Un|Od#seV3mXoMx`=P=wm+Ld_JAVCP9h{
z%sc_q^Kwcm+A8!t$Xp04Svf=M2h&WG`Z*SB18R=Y4nkWeCnmgVbKf&oNou*ac5vtN
z@&wdxVuFF`U_Ug-UVvqt++?*NR$d2TogJx^kwibO1>oaOt5qQ*Bp`(ect|so^{V{*
z+_krnHbdvxab<89UvLD!&PzRSgM<Dp6rhbFLTfO!VM_cBBuGyj*#enBlZH--3(j%(
zY6Tn&ScBuB1absfh^4WR0#*h>Ph`nL8)DGw^I@wZPmSgMu@O=3iFUWuZJZ|`ce9+!
z*~k|1U?OEnC&?Xnj1|l>`o^rtqv@yCAro?tJTl)fU3pgNY5C@OMc&!<ZnqK-gp!lH
zeShaT4nSpxa=9Xy$Nkp*yC;*OK%Cm|>y)W|0iyThZZzxBIci03@NHe>l-lO*+`b%E
z)mmTN-+hM1_T>&v1iyt3DKNh4f62#t$-suNyp2Gz7td3V0<%~?wk%s^+cGdxy-mHA
z2<K9%i>)j^AIS~>$+m6r7TSquaB;_u#Xu}^dt^RN5ts#F!znYT)<l<&<VL5(!oweN
ze@>MR<pr+u>g|5r?aM#^=<aJg9eYVM;odeL3oNYfSp2vzIh!sx<ogY03+dUUZ=Y%I
z)7!Fa3OVjJ=z}3pt8Kb;dZbC;jx6^DB)B&D`~@k+0HEt-W-?DL5&==f4-)<ax-<-x
z*?|6{Jqps>45X&Txahj`m0_h*1&}U40wr8}>@a*qXw^Oe@eFW3W;i2CgjuUMU_aLp
zZxWSS8q5;W!SSPU4XLf@x;MyDB-qKp3xzHrB#4`cbO??=kDJkX^kUlh-C8h0IJ!Z?
zcY!)!UrOQ|wykL?dvRdXjE0HNAdy+Qlzkzlx=Xz#NojOWWwWn*XX*T;GfRy&t42Li
zCOwI{O1C4VkO~??3ZUOm&m#woX{OS~XU}V6=XI{Toov<-XF-CT!PEidTst?%c#Vau
zU^o>lwMhZ}4D3hv@T|+Ul{Nh|$xM~Sr7^t;iO(97F@HjEAQrj5unmDN{Sk7;>(C4K
zpie#1^E}Brl)~209?flCF~N}-RvJw#_L?X&N$GN#+k$PT$Ayf;G&L=-IYg+sX!C5~
z0FgIvG@$8-pd$Qr+-FE7tm)BU(|KYI<d8*)Op@*!b)b{hUNte?bhTk1JNT0;uDF8A
zuM$CCS4j-j200&dxxq6Hsxt4RzN#()i!|2SW8@2x#v+Jw8OGTwU>gV&fKmtig;(0%
zFaoAYNQv*b@~8Je#3$tMk#cK-1e89XQw&E)XKt?TGR#P%q8doykeT^dg0%a|047GR
zc}a&j6{%2wLKknBBIX~)5wX27sfEDh1FtghSD~uOf8z{%md%Fm`CJMQbNx5{X6>KU
zL=&O*uVR4InrAVNs|hn6t2^Ng(GXLo=`XVV3mjPFrIBsRk40~e-r}Nl%y@Nl77XI@
zagYFIo=BMHcJeH_8`ML?jm)1`uWVJwVlEt<Z0F_XRU}nFKa{`UUIBmf)VHoW4yhQ4
zKFMYWZ{xlOS%kl+o{A1xzz6O@9+Pet-YtJ}7v^4vKj*wkE`&iJvULE5FtRkkkkQlb
zTmpObmx?PGJ8MLU;D0YxmdMKdf{mG(NVw<^g?%6@n-jeqoi>=GT)Q(BGtB7f{5)na
zV`1S?7r7*Q^OaW)2uBgfWv&s8CHt3Na;^Fy;x`FcsGb#Fn4gnuV4cp6`8rjS(9a{j
zP&z#|xqOIJZOA2l-c4#397g`!0vF>6c5|A!qnJmR_v62QYlf<pNl{^CT|T3h7_me!
z;!X8&lHKD@3>P35UaMN89$so<qE@xM$waI`Hg~EpP#-dG+5Vu{XLj0GpXhY(?gW|3
zRFNb%BIApYn=HB9bpAO}{P?v9WAJ55#hl>zBTjA^V^l5}zHZlU!`QX!WH^{(!*b4%
z==w(=F-XZhilI4%-Rap#5lfCVKkuU%wwAB3TDsZUI?|b9JK?2zEHIlzA;JNOtzS>>
z(TT#Q0p~E#<n>XMwjbGh?5ese7>Ev5E<H9F7{=k$exFrBIJS)W`9rr7{OJ&!gy+>C
zeho!1saGmlwABMfx4Z9>`FVq*E|CsjiB%A#zU9Zp<nzl}Pd~rN@jKgXid)Pdx~jL{
zfIK4eW;SPLRi7L-5KFMsR@eQK4fpNrOd7@{Nxmm1QI=J?>y*lMvMJN_7D>yD^Gl3>
zYLGqp88u{#NN#|qg7zF`9o4<f>uXF+&vnPjbdVOIiJJ~Gk?ZBurabgEX?Q&gpsO6o
zaWE`C;$ekd@o~4VNc(dnedLl@)SHZ)in7BcAe3k$gkRX1&yDSx1>(07h1YMbx@OP9
z+bwkVNF+6UE2c;pKq0($#Vc_=)@ztkAp>4jEY|9zwz;AR=?YiDT^GD&BVqcfpP28o
z7qGvpT_ogdCn)6=7cfX1_e$ciYu6FL_g^zooh2I0{8!N`c=k}MdCBxlG!_X@=mHOz
z4N3b|1Ry;jzQ^gALlYBWj!Wgera@Xm@DUW5{8AoC*;x3Z&^)gvhyCh%w7ukts0h`Q
zDD?AY{0?pGW<L_fGFVL5Y#0NdA&u(QZGO&n_<7gq)2Ht|edNQPBmC^YhOu$&v!3;=
z|8er<$>FC@o_x3R{5=kf4H5%!Umtl#zNqdZmNo^tVh8!sZ*nbH!qm{gr61|N6=&3q
z8u=zWSQlwz+>PxHRv^fpj0g8jdL6x;-o;jAX*GUD<QBbq=FFKhgww~L4oJoDMZJMw
zV0L1LvNlBT5M}oyG*AB<*4Ni9VuKn46L^D9Bc93rPsuv<46NPowfNJaCEnqekSu(+
zLRnF7f;(2e4u2qRrB%RW6r7y`8t46C4qJ&GybX30I!ZW|^{g9ixB;+<UFwjA(QJ4_
z;n*uD%V0VKCu1a-te9;6Lth~1|C9uKejSGzk(TfV&!c~iUF89==j8Sn=<zCJ77B>F
zwNK^%bUJaC@Sy=9E?U5lunPAhn!79wENw@l)|0d=ElYJL`KG0BLj$I?gLzqa-5Z~&
z?8@acvpCVoP2LNKveAh5pHaiDRHNBymu(W(B}BRlZHTix2tbE;$CY*)Cy6)nmQb-&
zD$vAXbteOvmrJ(W^9$go#8!K+=}jb-Q4hT;@YH4l5sq}l(a4?{rZe0DyU2+Z*JIKL
zAAS(J8iuk?>xzs*AL-(bL?Da9NX7v*1!b5t2U+;@c(ut$z;C7-gtx;L_x3><5iT6b
z_>3<2(2#F;`POhMRrmS3{Z7YzPO#lxd*;7+L)ZFM+PaiDA)B(Rd#}B;*PU#&nypSU
zd8rXP`n2Wl;;jATt4`f;=GIfEF5eGUFmUkj%kQ3FUhd!a6!8VZ3pRM>1XiU@5G8kd
z?t$aEkLKnamxgzq0ADk(NsK1uB^lDPlx6t#dZW!ftp_aA*Meqqjichp(Y=mwM$-Gc
zq><@C6SSEvQb|_?J&r<)@2Q8y8dztL3ZFC2Y9LN+oRdR6B1^+-@t)0eyis3tL*)pX
zxB=JcWPbAQMpij#27fqLNv2S7klIOWl3^^w%QZO1JqTabn6(q6E7_DV(0$WLQVk{%
z$Gqx`{MyBUI{*5%x4rFb<`ODWS)o_6{I;b6NeaVom^!tXqdBq%BT<tC&~G+HH;Cqz
zI7mfbhbe&cfPM^*4sHK!+Ws9nuRzXICzkb2&+9zD;rXEFk34_P+-|E?^(1v#-Kk!y
z-cB^fx70ryA$mu~ln78QQIUnGFuxd8O=NO+9ZSd}))Pt{(SGSgM0PA77PaBgj}WH$
zp#LzXB-y%P+c~P6u|=`cn>o0FS|#34Bp%Eu<WLl$Ql{nw09Ni`K~#`jZXG!Ybr0|^
zMs75Cp>Sd=?T@YqXrs4er$L?J=kKlPrT#z@Z^0i3vP0k4akWX&1Q<756e}MZVxz-&
zkr63o3f9EwDwlRI?>;ofcZh`G)%1CgRxAn~nn`qs@p}oY9Js+0;~R_*96yfSlx_NP
zdirr5fBxvXw;wn_)RsqbAB%IH<$f2%w9(g<rxv@-bkjl!l_pR3(ikpvC-j$e$H)a`
zF3D3aausQN0fBNg+E#Qb{Y<(!pc4c2zrYws+>jdnH(dX~Sf2;_U-?n<W%F6#xZ@ub
z5`LAs?1B@>qb57+Q_X=|X-=(^pz{%8gZi&iQ6huE4p4fD)L|6lrudS74j-WxwNWMb
zBHk858sdcM7ycY(wA*qH@TdvY2@xcSAp=N7yjT?Qk4C44C=JRx80&uOzTw9(sf$GD
zzCfA7p#V+`Ym-xu&v*w=SPdp4$?PE2#-!d(+5kfKxu}#4AHSdbjV(?>$K~VYm3B-w
zCy0j*x0=&Cc&oPCSkh>tb_2G9m^oU(jP$r4NW4<lUW^Rf4dg&M-F^EPP&+%F^=Hg(
zY|PDVY|I`(;vsK=x8<YmDS4N@8N!-<4!%0ncS7D<1*$nyy$$&%1QSEN#PEkjfB<8W
zN)uFMiXiYNDbh|IcA7Nn^?O?x8rSeA!5R5zVsDha7<EnZ<+4|=`h6|@_oD-2sjaQc
zmqE_Bj>eRpf5n#w2j$0i<$4F~M7BCTGd*2(NEW&SE;i;3LC=I|r{@^5&K*d(@AEw9
z`J(5)d%g)JT85j;;3TlQeo>&r5_Dxn^E!bHNy<PusX2y`%K5D7>&Rwiw#aOt^cTsr
zC+BGZidX_f<q-7Wcv@rRVOlc}9Jk{0Nch5<hzky8$`G@#EM#@@x4t|loJVvtNP~bk
z5Ov&O<&{CJJ$@A}xEuZ$87KV1ntEjTI94OGXO>b>2(_s~I%~vNf!~x%rSjYXEGyaM
zgumG#C==ltLwqEhS_Gs7<UAg8s9`E$6sJjHk;DUrp$XB9LJtM8(14)}KgHok4A9u`
z_p>{$-mRj;+itq{$-DO?htqaUFnfW(zlHOWcj-xd{}va#NtS*EMwv``oAvt9Xfm3O
z9v$#={5W=s(X;Se77l_asHe0ACXWib`5YGRO+SM#o`B<pm{W`?f5IUsm9V*^B%g?p
zUiRqlXS{EtUaf|2vRAACxg10u-WnqVe7eu{{3`XBM+k-y@<0f#@OYO(OUF8zIEpkN
zBog*8LNp{Ki;mDc@&CFKg&RvFmypie{>`8f7SLa`JDkOu-aDq8GQvcRaBnC3xUL+?
z@bgH+xkAJ4TJ>w2M=vkkoJee^i4V@q%uM(8@9#}d_4e&McGXqKe(?jRQ4~5YemZzX
zc5U=_&xAL_M_Y|QMX648;#=L&Y`*I=8ynrOG2Sw~KTG9?`AaTYSh)01JzI0?ufJq^
zdivPZ^$7K6UVrVi#Qc;>XU^=Bqdai;kv!%H-#dv!c8chY2e7XGh37|}f7Oz-o7Z08
zO5!s{O42G}ecQUQ+$Rv&mW7Wn+9A1AClxO>ng??A%0Lia7Mw}um+%=vWZ_Zn5z9<V
zTl7+dv<7D)?TN%H9FmV)Y&)%GbtJyW-ay>-7rQ|N<<{D|1*f{tg)eioQP7?8B<<#B
zRR(iVI;NfmDKmY6g!sh$sw8V(KPXH5@87r}Z`;7;O_Hg&N%20CF&k=5onRXH0#V#M
zxPK^kY9_qox1S_R4%@m>2pL|AVoSj`D6<!5kZgU&_%a!SK(<|mkZpaLILyVXud`r+
zDvWbY6~B%Li?bH{LfVM&+ck<t97CJpJB?<gUUo6&^6HqNbe}>I>>~QsT)?onK2y0O
zPIDw;A5MpxufP6!fzDzl;&DxON2OUv(JkEm>7y%{qOd*o0yrq$E7Up9xRq0Vw=Am+
z<yg|B2F0*A8i$tF3=V_)dXeX4Q2+ZqpM~zXMc#x<3kF~M<_T>3bc%(PCUTw3bs^VF
z4knLgdlX;f%1}smOk<Jg#Ezjw9;c1QC@_C_`isX1HX>sbYS2fY=Qu}l7Dza=&=t9|
zwmLTDn&LGaUy(D#kHA<{KZqL>M*O+BB&C{q=B*ba@x!y`f*{B6tR|!=t}&k^7fx6a
zBotByM34z#Bu*met7h>@lBH;(N?d%<J3BG@c~a7DvWf6)ZT?9A25~7q2e)CqAm);2
zw18jee_^9WRI&OfH`S{7#S4-3pLWpN8X-{~xKE;^tOk*txOPdWv$_;B$|OCCKgVCK
zW3)Q)5)qn=m%Jp2G*R+@Sox>y_eg@1M717|_Z-7t4#eX%LbZuI)5+ij5F|AWuJT+3
zeS!T7gA{Hn%{@}lWsW0lx*dpxLt+FN0Gdwd`A@U!&UKFv1dNP}6*P&zWTIu#%__LA
z4Cv`pGD=R^M)OGs47_oTN^-clC#SvZ!IRSx(S9&sm&G)`bCz@$6y2<pFW*~l7%yGS
z=dO~J6BsY@x#<-k9t6sVqch|JhpEmkqMq>jg{qpckC+6&cGvOZA|j@-t*6uVP-p{F
zV5fcX)wlqn*eA4&;t^s<C5P;DuphbN^*E|Wh}tNTiLLkWD!iOnDa7F*z6-dCG?lB1
zX)-7y8qGr*L3^U8w90WbmPjNtitHVYP&tO1KdzpKVBY>eXhI+qE+jGSgi~Z&!0#NZ
zUT&zPi)0UBtk{ugsU<Q$cuz02I>Bld)~uFM1f7>Pz1VprVym%`L0r6PjH25PDE*Ac
zj5Tf*jW!CH*l{`6v=n^eT9Y)pQ(pWKf>2BoJOFhFwJHqxuGdj%Y84a8QYn#{kSBM2
z55Ehx>p^lqT&XL~Gx6lrgntMg*yRvo9E~*EvfzSgB6f>CNDh{EmjAA;*=>0t`XsZ?
zv=G%~P76qXWnkR8Z{NPk&Rv(gu1;THlH&c`hMNEvomiWnFDTNO8`<H*)@Va<_&wR6
zrmf-E`H^M)Pu2EMO)I6H!Ffm0>^*+{rW3>rlR|z$b~^dTuI0vw`LL(VVwZdqcF7H}
zInVUG0x9{s(94y9_`c-teB-$9K3X5qLbu_>=(}YK67JsN)#YT@R7T5Ykdg53v&|73
zk|X5Q9@h}3<z%A>&ZB#1@RC-ZxgC9=2P=&##Fks?8BE_uB`8w#$;`z>z#i!@`3N4t
z(#N`U-#uTRrX%-3FSq4F>Lhq}(Wb~&nIeB$lxpA1!5A4$QaQ*{G)AWCU{L3cxB76T
zxhGxQTN34~*)&5zsV0G)Kjc%5`uj<A6pA-U^A=QOg~MRN#3zVT1&Exbkgjb#IJ<YQ
zQBUCYh?4M%s}rw6S~Pz!YWq{v7l8#z2Ae<trvR2a6^<msR0Du3o5^B<&@7DH;RI&C
zLWDd=fnf3G0Ge1xS8)B?1i0Z!LI^j$!g?bgCR=>A)2R^?Ol7OW+$@usYbN3;q3yBN
zwYgc)Y^zVM7xKZUBT`f7BUr9fs>JzNH|(UHLZP`^2tQH@DiD<7UnOR~koV&vOp{$6
z)19q8yXZL6z_~?JNC1{aiH-%N2D>N^fJ|smgbreJ!Vw@o)YkQdBya)$NE%I$ccH#7
zRoq*x+DnN-!9t9PvE4Uld1^YdJ3=!#WiKur;%zo1oxBLo2+fZm3tv6+$c4Dl-WzYc
z@!X9kZ_k}nQ}^?8=wfor@VC_T@L~BoNt^+8ujr#+H<r{#(I~G&u6Qw0#l2uz-buE~
zPm{0m_n9+)<@pECKcl8bR1quKoEm^0)tR8hGKz1B{}RmC#;zp)h0<VTw<-3GNGzm7
z;3MPzOJYhyypaDwe-K7T|4CmCzeD}(3)zCgBX@{k(Zz8kcflG4EN*FlQ@YQAY&d@m
zhx%wt5rkQ^-FEG8l*ZACtSn(PC_gEA4#DQ>Lpxkv&OknK#KtKbdu)SE$npKm=)?|-
z`tn;_kb)d}d69Lvuntd<60sur===iAM%gvs3i8D($)gaBz(@AU8qAANchTqwFIACM
z<Ph6;mIR8vvtAnu_$nrb6)}<w_**`|_v!2<-#mK;`~`=15W*rqO7cJ<aIC6juh~O~
zG2G1u>AmL0&7TAQmJKQ5Jgz);rG<-rzxCq8vkha1ar3d~2A=)wXRp1-`yxTPUiRkR
zt#@j}L~n&eWIL~z5-AEz>1g8VHq!DhoK$zoai_0uU;nfhV6rgOb$291&CWIn=TDP%
zcyl0*#7uKu{oys&Tw{78#!Wbe#WaOU)-+B_s>kawV|j0+o{)Jh*(#wAkOq?KUI#S8
zypbB(<{usA=Qv5#dV5w@R+PF4bCPkLv3&hg78Vw;WyLSK5=_ozI}yp$d-pv>eHn;M
zICS;ZS08`!QHzMzL#2d*q@95?-}2-a8!vp}3wP~=QV`HZ(ltY^qz_$@&1UzUm^pT;
z&ClT{V?4Mzx&BW0>~%NFfj8brts^XI*B{FgB)97T8EA}WoIZUT-u|vDxG9rfWS10t
z=0otAtYm}`-<R-Pk;cck<HVwEx+2XOHzE5ZY@n$Qee;{&?BoSAmGAtOEIj2yq!<wD
z8@)&3U3-tHC*%I>{B;|zD4x-_`oOKj-+`(`UOztWMF3{q_=P`5jCs4xfiWp4+L6at
zWN~?<7W{=TEe#GB@v<Te2$f;`9=fPpS?cZ1=T03s&{VI<6rXd|RgF}karEdNv7EKD
zyHKC+^jD(kiZ^`g;ls^hrMY+G>em(uyO(;Eva%))9K7N4#Z>d?RnIA9?l^X=@o(+9
z`r_hFJD-lOuDqdAY+iokX(8{`o10B}ZqsAah->O)L`g3YA?;Cm{<gKH8pk6Pg8Z%n
z9WFrTLup(TX4$;rtk8_94rEezu84J}Y(~A%Du$z8&iDweE-b9pD{z3hg<aE=>z{7b
zmsfYr%<NoUuC>(k?n7&fi))8=PtP<Ojm72l+1d5w#YPk0u$<SfR-5XopZ@fxB)syI
zC(1~wnQ1lrNIXZmXhjy#Qw`A-4AmGuFsE|E?~zvyB!kKfzemo3vH(ow9etpA9x;9q
z`F6sy4C?;4=Zr`qkWRPZ-T$Pa*WELgc4aRrn}}O3T#7N%`wy-gC9kp2J+wYM7t2*A
zoSZW21a29P=CXPzI4px<H$D8I5velsqSc6Tt%_DfMdIIe=eJ&CM9Q+Y95MblnL5;^
zf*vK3kZp6hCSJ%&WdiFXM5eh+0?5$2@nT1AlMCJ^ORtbKUm;5m$kGF{^c~sz9r81I
z9{nm~_H){IoYuKprob8xJiDoYeOPm)(_{y_4eZlPsJ-?!&o_ut_)CzMKO#~AuenOY
zl^SYFb>U09)d6)_U88Pb<THkV_dD*_XiXRc8{rGv_Kh}<4l$X%Sg=L`XrnJ7-)wyB
z*JJmMJl>;2I35(IHls3fw~LxG?-AuhSd>L!O7yaJud<GIqIKqu1?LjEQCOugNJc$M
z;OLe<_v#Y#$5WX&gT#QBFf}*BpN{%fALS2|iZL7`DF)8JfW*`Bbba@0bkWg9Nh83%
z$G+TXpIC)QJI)}Jd8<s4s<-k!7(mtc029ntd9mK|hkDCF2ztf)WLfrpg9oZFFu@x7
z-#xN=<JRhpvidq%9lk>_+6QIvAm-Ob=r_ltvE#;jb-IEH%o0U|Z=3=fxW&b|Pwp5}
z^~Q+-%>?vjY(M~e64BTT<$~&J`9FNW{O|w8UzUISw}1Of>Q%#E9R0mfw#(0R<w3sV
zE{8%Nm$i?N*Iq7bU3s!&Y^rA`WU(EMMuX83oR6etg60g)=`GW22}HYWi3F>9dmGqH
zB&e=rW&6+YPa8i2jsE-8)asmn)x5e5lL0&cAO{Q=kQ(|ll5+<17J^t(K#AlYhKEt2
z=0xK|jhVW-YIs)t>hSE%H}n4`@W89FR{u!NtgpYU9;`nQdJF%r32oLNXncsrr)5Lf
zH=(^=k4OERI)}$|S&(Bgj>Tf&8hT{liPo54D2ig3V6cvKb5TVYpW+U%5q~f;j-xU(
z5mVox&J9UDKa1n&EFySbeK~KKx#6cu5>rOV8sGf?hWj<%sLpZt<Ro4uBRBk{WCc-&
zbH?z`Fd>vnIEbmmR4PNh9=|TJBOWih?~9=o;*%X~vm^SdHM&tu0#7Z^tP$z9>>s@e
zGfy){mm;G$%l6Uv96wUPw9Oz*Vg@;y8`kSfyQilon$uG|2jr$JmDlR^dXX&8s9wlA
zwW-PJ4Kg+v(=(fwE_OThiHYu_I1}b$Soaa}=t84bU*e@gevyf+bg@|8)oJspNy(e?
z{2n>KjCO0W+iDq@Xr`uF52J3FC@ewPg|8$TW+pp(H{nOd)YS4Uijmo<Hz4genGg{I
z!6gAsqlPecLVgRx9v7rKjSn`zK$k;<f17-bpV1jiw;)ntRT0$G_7<WYw#-q|Hmtck
zOTOKDu+nOJWN535=5WogUAdD?*crx(1W|$fFgl%G%XN%FXiVt_wX~^+B(e=niX9J#
z875GEyfxgYWPC`c()~sQ7(jpdQuUec!tp*?C!-Nb(-4WKcDHD)QmJ2RP85n{b4rsN
ztX%ioiTG3@mn}?9&os%vht=4eE*9s=-oWL{<+-_q!?Oh3$sEAGs%=jtWTqF>ATBp{
z6&dkZ2?3Xnv<7mVxKP~CG1)jWa|+0xNub+=R?_XJEl%zd+Z#v~Z8FnJ8_MYEYV}^0
zvij;`TWB}cOEypLURpwqDp#MkxF#Z1g=7u2xA$i3gDEd4yZ-X!hq1nx{r=^*_@Yt3
zv1FUT{+J^<I8i8m0}HfdWye(s5RVG9$a4@wTGW@%bqXl{IE=iuDP?WdjT;0dSC*D`
z@9f=CExL<ylasaK`&HhX+;7kIFJJDfm6gMn_m_>=qS)f?L_xt@8j41X=m1d$8C2Ns
zqgn^!3yV}Lorohm2a3fM=_(8ZLqHZAmrT)-o$nlS(7ec}LT$8O1{*{c)^5l1-=a<l
z7v1_c;oN9b>KvYdt8Tao*Uo{+(WhQ}?Ng6N_jkJwVFz9*s{0>$==N@{`}ME$@Al!>
zIdnt!u_wrMl{ov3WZ)dZV5Rk?6KQ#ov>+Q|P}G}rXO0pDjyLN0@TntcYeuhk<gojF
zSk<81Nbl_eM!$Ta{VRy=5PO+=fhe%?-tv(n%W{hQ9RfCa>7}@(#@|WmaPHj;5_0@K
zZYQya`oZ;k1e>f;#1c#*?lcP(vZEUfp1jz1vBBRU0IWW5Tj9Cpk@9)S@S~nLzVVGG
zqmkmtdv72kD`>r+J^F{nLG@|;2Osc!&hvYoKk@tx8h!wX<_vMDN7O6STh&L^SJa=V
zzf(UkLPpJ4H1-=OL}Im}gZv>m+8DV=2}A42xWxp(hw;lgp1={U)q6CRka0g+lQx$*
zARn>E@{MexDUEEC`GCz$rj=Y%)PT`eQXt7%TP$guC4X`(X=w&xOpy7BSztvjV7EK$
z6A4P+g*%gEM{huRkSgolMW2g-m%}aFaml+roMneM!Cyh>E7!Gb=J@zh@~Ygb=OV22
zJ4sMcCksRb8=Q8T$b~88Vd)4A0pg&HKtm+|B<IQl>85atXDs#Q9e4x&1WuRbX#NYw
z;*F#g;8c!FW%bQFoX<GrEp&QJIhlvyc}G-=r7lk`uOef*lV%RdF86|2fjR1xkySH|
z_lWn`*VHyhX~-3bIfRZWAh#l`E6#aYaIe9BI-0}evOETx<&~u8+%}dguvy-;rTGba
z+*9SD=~AvW+92I-><Un!LcHlPhgcX3J4G?<B-R0w5)2v?5g@V+eM-oQ&mtofh7rpu
z#EL7Mo9mXZ;7m?r$v`IoZTK|836LG68MCNL5hK7a1~i5ra)v|=tO|9%52Y}W6yk^%
zpIahk86@~A#oqt^7(jKf9#0}T1R)EG>BGJW<TXH6X*1}zav`aY;gez!!|R~4egZv=
zypJd2hEX8-9F_=ld37XE-1y!HYakaTp~Lw5PU=^{=!q7LMX_mk$%iMj8PVa=ASFRS
zq2O6SsP&#rVpQbn)S>WsC$Rb(xLM<-l}y1aGhT9*8!;%m8HmLw{vqO`-2|ku&~Dcw
zz6^v|lB@)CP6~bsjpmA8Jdxr=@EX&rt4o+%j=t?}{heT`;mx}S!T$Q%0_ID<!Eq4q
zm~oZ|ih&GPSZ+h9oby}WJS8-u2Umv=lk7+=n-NUJa33r5Cs`>#>X79`#PG0}WWL{K
z&`8}|lJf$6RqA78@WN6W3QFB4VlJZ8n*cl$v26?giN6>lVN)DtMzuj;_kln%7^Vmb
z`JQ-3aGt>K0}Ks`=khZ?{8)_fHBaPb0Nc=<0beo<@>N7)<>!g9ILR?yQWd0e)zOex
zl`xi30yq?_BLm>?^{KEIbOrT+8Gm`={~XJ5j8rP`vxdNE1JX4PVg@x=arOZ`O>kj5
zK;W9HSF^b80Coj@5vQaPC@{h>hnUd(a*o(-Q&sBk%w}qJsTi5fkisMTVWbb+cA{``
zE6#}GyqLAZn^DA61=1NW*1L!xQGDc<j3KnCpdZjzn&E)cT7rgXIYJLYK><feY|KQd
z6pI&2kVwghNjt{KQ)-fLCh7BTX2?oq+BlFHopZHBrIgzcS)iVv#nGP9^>8Iwos&W4
z1^?wUGU;O8ilI>j1Ibj3cP1aIdeMZK_)8P(!%tFWBpRT23GgrUuHZ+-5{>Q?u;Nq{
zi4_uY^<%vMxW<tX7`J|VA4hR4FN%8~DQ_CYxy6tm?o={Yxw6-v8uzOAXx#oKa2OrA
zB4ktpjG%uRSPS$*Sn&U$>`mY$JIgx%de5mkr}n+-)V{BGzjc?o_jcdgxA!fbbSI&+
zla&xc5)eW_1g6CVBoGKl5cv~mal`=^GXDr7L~Ib8=<pHLfpOgaQJjy?=(r9eDpvo$
z&wHwF-wug7Rrgk%I_Evxd*1iipWh?OurP|I(#Z>5VbPkh2(5(Z3&DZ(X!4|f3hz9Q
zGk`t_v>CZUK<;nVdg8jXl>Y!5u39;9{IT!vpvE=Zg{k(@YPHdPi@m$?oO=NGVAyAY
zGMe%SN)^CPlo5>;3XNvJQ{8B{5(#)KmMn5;rCxlJ+670zc6t>k`JZ?}d4o$G(`77y
zQj|=rJVH*zb~v{3Kv*xc{&Wer66;q(60yXH?$AOU>w>t&-aDOUx&(Vnv>~>-UxZro
zPV97_*4bQJnXAfK&F|c}cdu#g-FxHY`*u|<_4_C{1iMLkZiSc!S^R4w{++m?Kfa^9
z%&~x1J*MJ%TLRop(4d;TvmkdbZ!a%nDX1<lZ(mwcPwYOcoc$3wtDL<Ez4sUUlDIiu
z{;Hh$-ID0=?sq)$$Rl3wBVw;z$ONl_Ks5-wIIkF{p%=+19W|bJa(x-(+=(t=Y$JE2
zh#EhH7xj&v_cAJ1q?qi0RA*szly%|l#>5Jb$Fw5OItkjrk_risargqUxd^)qVl9AZ
z<1j*q9;>(#7~wmL7oid>JF#@)X2AA`p>S27Rm44+1?s7?&>N_l*Bh~eQv{nu{ox2z
zo`^evh}`{xu{RF^c?{|CmCB_YRB1DU(EfD2(~)j2mpa`_2er{?)Vsf6HY-Mx^!Gx3
z>jyx`{VI{!CKEWLno?Kl7hlZ1^y)-R3OYwi=y+Hb7o!PZ04G|DBE!mGN+FJP0n#ts
zST>cFD;w(Ej(v2}?@2O9r00=TUqD88g%Si9u%(xHC^Gw{w*^vHz7ss4iS;xglc_5P
zJ^7YhT-Ud&f&KN#!J+}C(8&{nfnf{=Cwes<5}2-?uDT1=@4Q$z%CC}-az^|c!{M2m
z&TQ;4C?!9;qF%lGaB}$3pG8W=79={1gZt~@QmNG{mPkl^ArIc@HY$*xz8+;utt?-^
zGCd8xk89I+=}DE^fG?HYF!3O9Lj;_OrXv0@xpB+AMQ7lM1MBkbuBqcvi|+GjgDb7G
z6Yqqp#!7chI`{w3xnyM;=+wz<paRdTqzVO-;=sAN^m{JbSS7`>mM#<w;0xIJ*s(Rq
z=~wl}+S;K5Yio_BQnlLJ+R?Kcc<|TOZoCfE5p&PpGyj8GL%Fh<B*F%hSA4y(*y1#m
zJ)x_+wtxgj8LRP@uK^>yws!Ey+FHE^Ec$Tm#E~_LWEpLm>l-h*de0sKHM#B<0<%T$
z^hAiOc`cEAi>`H8djQco<y-JyM57-{=Z+UI-}E%V7%zYN>DfZ++_~jb`47p@{Hf)0
z=Te2)r=R{^e%;O1JX@De6@KjQyYGH^;R@NBXDjD6sRj6~`Um)e*ch(#c*X+t3)5%O
zbG7Gk_od18496SfTIHH|?tvU<dZ9iJo7la)_!eHxT=>+f;!5tylXv3C8~-?U^38Al
z7hsLvAv?Bmp;M;{EBPx=8nUDQkvnzrEpO5Hq%J<A_Nfcl>d$C@hC8ktE===cMiY3D
zmlCmtl5?Wp3@-RJZ@$M0n<cu@gdfU?mWEi_(3`XkoS)iCPSPOo8<{Ie93_^wI(G%~
zJ<+p)V1YUy`Ftji?1*HYR#cOpDdf|nR3XXbziqVI!(qE+Oui3D(99$tN4{Uf5sl0h
zL^gns#c|ye1VRNcBP}Zuix9Km^~Lb|qZ>qw7?@AEJY!!Swu%2~bMTgYw}u>xBbVf}
zB-YijZ7EES+LBa9OS!Q}%sEk_m_08{w?x-+dx6e!R;3@t5)Rs$+rzqlrO2g)&L<Og
z^?5x!Hf7)^f6kF$1uEdk=lRiv7Y4OEKbOm9b93(J9;_IqWl<Xrpp(cvar?1Ud4W7L
z(gTRsi$uclctnuZi1na}$f0UTbQ!Nq>J={(Otr1L{az1aWv|~={{fZBPYzuW-AqO1
z(K15m(t@OhV_-sgilDBrktQf#fF&USkf_5vz9_W@&fzIKz$~KsTx+!qriqZFQ}vqW
zSpSw0Cw-%RE8`}RNzulo=hX*C?^UlFy_ZlztY1rmefzW%GaO2MEEo<4TG^gBV=*+M
z!&I<BCu@>144z>iNc4sRZ!Yn6cNcf}w<km4(ck&~%}fX7KpOh(N|GYQV1F4D;zMC-
zXSCl_?5=-<P3qmq7#<~poxUKE__|}7j=@kapSNV7Y9NKNf#lNNPFhGl&F&U@Ks1q;
zMRMUSGNq(<x$l&gG$j^$(Ky$S1>*3jYV_)4s#ZZpSIt(t8O%DFkKt1B0puN{;9!Wl
zfDx3*G=s<@DJ%AFq%4bQp9oYr6j@f)MuxfA&g?-%h>@miywL{L;+%R6ue{kwx3{mk
z=I9rjt*yOEU47=7YqI#WNr5yY7$@Uoqtw?{S8`E;ahlomEOi+ph&EVxQS>8aCtnKI
zYlSizlNm88KM<fGQz{nqszL&dCaP_ViJh7vUc4DY^5<>=Z~vU<ePl^I<M}Fn!@ok)
zAx2r@rEfz??XiII!bI4C7p&4-Ogx^8OfuAsVOOt}b>lS*E;i7w(!ZjRh^JQS;5eEJ
z^5_W89<l|GpRD0zm)2;BL{2Y4z60;EXfLN(WhU!6Zx7j%fuZ-A9An&4+GxE85omH|
zxe_i}6J0s+WEXvy%MQ}zbi>7=+j!iRqscXj7gMgom6+uGD%I(9j^4PNvFu*{FxDBA
zQMy-|6kRgyJ8at)W3gpoXI5%Mj1%&IQd)^Ug<Mg#qK<~Z`S9_i*OeWY7JT%ce>4fH
zGGka0Y!3cXeP}Gu1{wK=I^Z7IRtNQ=4bFS{f$x9nsiz)%s;T!KABS4VlXEHiXSuWc
z)zy%aonjlimp=uC!>GW)xo|o94P4#$Qe<u8(sdIY#_eNiLKERar+{$g-h7xn-Z{Dd
z<Ig|Z4@>t<y6TT6hn6FFy@hdq=+evMj=lW2_dodHgHJv92Rr@DfeudkL-U>s$UGN3
zaZe52a|F2uRw5(-j!G0PW6dX;1!D1-_6+683mI}D1B6s8jxpP1S+>Yt9#wyU;JJ1t
zeZ_4+8mRkIM*o8;x2VVc-}i@q_=mVj1E*Tc%Uj1bHX5T}hZnnemE-RVM>l@Cf+ggU
z*pVZ59;meTv}c!&(z2nx@FnbEvyaS*v$xUQIHyrAckPYXfxpVQ_-oW1{RKtV3wUIb
znkU15f#EIq8YDHs@u*ZK5n|0{CggIHb+TUaEJaC1E|Cag@*>$Tab&cW@2>o6h>1bB
z?clMa?y}wo?>k|b{BBCSuA~rh1Z}mZN8~NPA=p|YY!C=1sTp|7>;)#L`;xOR@-MTS
zWuAf<qIF?BdUvdVb0o>QMVDl6@j-FyonK?~G4^q7Ax8X{mU7Q@h}4?hd5N*>vX_{7
z+*b%*hKv*Onnj9s>UF79$jL5?!6~1|HY95o=jYq8a5h<*pX=l)DU59v%&ka>(P-M0
zYA-uI)t^lmIsDyz64euwe{)H~*RauoR?8T655(-*{Sph*i@wH9-7l$hjM0zjM^+=2
zL!b$w*ubzPHKWN+zgO-D0*!nl9>epFR!iw661GyAeRJ~^O*Tu7^~S(XCes@S1gP-F
z^$o3t9JaC1XgT8R$m17~%Jwm&ES{16m*>E9;kiKp<*nJMrqi8f%3{dqRoY#liLVyQ
z)W+GRWuI@U)M?KSXAa$NSqD~ED8S$n5_S8*nc>|vVv445Q5C0Is;vM8kz7$9#pH43
z>ci-Al}*Fjs1Jt+-A*+3Rz|PI(Va~!%<tbnzmNcpEVXpt!2SaRnFAtMex67C>v-+=
zyT}J>RfxS@m@Okmu3N^>#AwfbVOUtjd&D~K3wJ)B6qOW?AuBpVwhGOcnLe@Z%d+Up
z(i7@i*yq=VcU*JeV8*G}8rQ7ErK(<`*zeEe3i;xRJsXAZFVD_(+J-t#PGoBI(nz7u
zoNKiLf!2c<j;8vUj=uKUIF5hnOJbt6(xq}!)l#`!W-St#HkX#KE0xQ!lXsMRz19LW
z+V6l}4~q1W8I1h-rFy&F_<Xlo%jA{OY@WP-%8dz`dIp3$#zrg`$7~5?MpQrF=b(M(
zp?sjma8GiZ@-K$64h3DsE2ArJa)(&xp?g`Pakx+EGp$-nk4Y~M;fSu$g}sp30d)#Q
zqm|QAiY6G{$i%CL6axE?H{N(-E%PYO>Uukj;1fci?bMMf-(l`ucH6$kYc5VtHP}nN
zi8ZPGU~RY;c9Ezpm$Q_O4Cce7;_>4|BV#2-o2k5Vonkk-t$yY;=-$Wh+ua9Z{6{>$
z$$gEzK5`G-of}*z%la9MY{2z%Pg-*lFOt3^86)l*8%!@PD!A4}k&0bR9oH4h{_e5#
zh)G$}fm$Y#)pU)F8oJ9Bqx95}pKO$o#wLuj?12KeoROg?duc}<<P*__1}+ooKqM^$
z4=&nyIW$Z|9~oJovoSkjDsQ(}2>A~5#4(f36pI<^Yy%fGGlS)-I~;ZdN)@k3*a-fu
zn8*#^b82<}ejKdS6I)#!3<}$DQ+uV~D;24N?Ijg>VU?0%<{z&gJ+iT$0&Ymf_HSQ%
z%~3B=>rYGe6+jkd6n_!t5e}Y6nzrprQaK&}lLWr97&h*qwmD%qhCfogrAy=iuqpjT
z+#_CzosUF%aJXzVot}5<j?b!fSGi#>HDKXjF$13`JOoQRKKe&+0!`1HIN0k2gO0O@
zOU2ZZ677}%4Q|bv0fUH2xu0J3dWpaVm?d!iU`ZX6WbOO7UmA@smiTs`ok&__Z*zf2
zS%}VDHQ}_-3&F$q9P?qpvkk($XdJG=QRtpLG57Jf8am^MN6X{LIM4%~cmy;6GJ?pJ
zV+y#GVUWqrLK=AQj0>K$xDzdo$Wf3cNLOfDE(((j07eKzquZ0XmS&6Mr`%~^&4>~V
zlgn=(Xf{qn@x&CUUjghYWfm9)#UHk2XRlnDnl4wUMla~O?WLt7hi{4odiZ?s$5Fwa
z${)CXhzi5v4{i-k57=RudZ~Fb&oWRTag(h{8LJ<mz%3~^)vn#d2N91IEbF?iAWIeq
zW$gb%s6ZvLR5Itpb&#+^C02cJ+{jUfhD=>tX}rjRm*@#1Y|x3An+p7`=kS@I$L{ZO
z1M`}oeu+Nj;~Sj#SG?jCA9_G7e)x4ys0*)n$!H||crJcdeOdhwJV-c@<vNe)%!cjC
zU()yPvysWdvaK$dH{7(ZTLP{w7ESspnOL-1D5jRf>147)k!_@c`Yne+tJwdV*B>wE
zk9CPerEF}8N+P~1-d-&2o5|IIq`A*Du6Jpqj(hP7{>2VbnLtp>1l0lTiw*)Lz+;=h
zp4_~>{n_pPmjVk$pTB|EXV0Gftma2<42=apy4mynJIC|64q)+cNNTA>d$3HJsl?$+
zG9fYjh-e}%iT=vabL6INfRhB6j;$_!F72c@$9oD`rfax!uT(b5za#&0#s3t+lSy>d
zq0pM)oB1J0)(4PFT~Q=@Y)K_fWNa&MpEi8)EI}G}wMLC?ZwWQHWL~<z$0*?d$uZi*
z$eX@=!SKJzY-f&<@foax)AP5Wnwa)|NJf<p(!j)CTgOnENciiyV5y|>;Dk@Rd}tdR
z*&XEUPjcI2>R1vq;JC45u2p(xI!VGG-hY=P%d&2zqwSbLvNOb-KmnPkNQg2i<#?L%
zuGlj1w3$f@GZ;W$M{bPIp&a*0svxD1;7#7WdvCa5?_RH@%Ti*jkXkv4o4QU39nh0`
zIscWSb|5>%8(xaMC;pOs#Np{}Q1ouEvT!vwEhDJuaV7{*Qz(=)lZl+S9+$&W-##Fv
z3duk?7)T}@IcdD^ZR8yMv|*WsQfKZ#W@{!5z3WPP&w{R_5shTJ2JYFTQWZzp4j$ha
z!_)N@UVx{6k^HuO^ra`IX9qlsJGX@dEyJrBKsu_rp(P+>ZJaiXur7&mA4_wL9NA-0
zf@5U?fk5vpQ*FQ@WTFXF3p>7H-?3}gH%My)<8DX*B6-Wk^757IOHQSNStxEKGp(EU
zY!IQ>U0<8?_J~VUS3(z(x%sV9p_|Q`vCQ23aDE!63vqA%Plp)k#(-D53BXavj}M2o
zn?Rfur%TkH8GUhfs#i`OId)4=I7(f`XJ=PtO1K=uiB@~6T9Rn6jIc9X+dL$8?OEb<
zK8wF<mT2iQT96(fGST#;x{Ofku{7jX=>^Be3gQkqS)Fw>`FBNwu?QsV)lW&K245mn
z=q%ROH%{%={Ex2<8=t$-Avt2|^!C&gP?l5MC#LsQ=jPTn2aA<*xw1Ie+zU+Xx0MFV
zo4Nf1hx^z7cWc|N_V)dEgmDKRHjSyNeflt4jfFx%@2n4ld#b#wWgtIvR&bsC<U$ub
z7}DC1f(*n;>#o|6Z?}DI$*s`UUxhB%>Mqa%9=Yw>Z+`Qe&WEnKVe~tyH~O8qe;h%+
zzWL^V{0%}rzIpA3eyVfZ+urlJ&wcK{mN%{d2;kXuJ?5l#^IxJTyu<Tj;>Q)*K1qYt
zD&8bgR@A-mGV+YTn@W<i%uu{}5-z35JKuVM@+}r4lnlJL@{k0)V{+Wx%dQ%!gS<2=
zmc4{ZcAD+9>>|VwvO36+7_j@dlTrTzUBjzdWUTwz-jcTsZ)CdRj`tHA!Vtm*Z{ya%
z$Vu)gE?tBTqvexb^auNS)h`&uLZN8<g7S`@;~4z?h#XHEocGn9_IhhI`L~P?Oj!hu
zy?hI`^bs8q@V(lg_Lke|>6*{?d$+YDL#K7y?=hTw+g|Us`w3ALUTC}(oN=(HiQi?0
zYHldue==4jk%o@bdYBm6T&K8vx^J!3=sg&)((ynjNZ0db$(3!G>gaUR%`)kAJ#Ds%
zJ%kf`Z$L*R3;!>}AcuycIJ<9<4&kF%O(hWyLXnqQplFNRpA_awA&JyWU`9Aa`eKDh
zg0|2z=$g}xRH6jAgd%E4$k7d6b-@S3ED}R2lP5$|X<xQJcu^G#RVtb&c0~cKq%-VJ
z?90qvP)!>Ie8e1T^BDDH5;V0Y#x7%3ii%#UcNu5SV+9|_Ot=hGWccxfza*XMj$nSv
zn;sKx;Y#?za%H1BLNUA}M@Z`Ye=nPwyQb+m@IBja7#wNm^6J0S@C19bcawtM%b0s|
z>N_W77m2U+ybL>ck=&R4o)c&-%pu|X638gFZi)RIlPb-?HK&&{!qP}`?1Zb6*GF?{
z$PSW)AeWx`Mjii2sI%D#S@i?0K<j`yN#y(S<7b2V<29e*^|B9Uf9~h9AN<VE{_JP=
zZEbCJmX|x@&v3oYwHb`P-zMhqXSlvoqEiX))$!dzBSpT@%+$^@TC~rA9`q50$RN@R
z)mTBE_~hv0$<U+WR8B3-A3ZWNlQE4!r?_^o*KC+Zv6(@)Pdbf8PHp8oGN@9SCdD9%
zFFn0oZ*ImYJ+{)V)hi_|?7qaWl$WkR|HisusAIWqE1hoVgD8kdexywotTF-hl|m|G
z2a#p>8EU>;JQ{7nu7bgQq20n&%xxm0)xb+UuYN`6+Vpr9y4pw=Hh?7<hvHDj-U#pL
zT6$h-JCws+V-JuP()aQVG?nF>TUKq7Qh{XWHLZTi$xijx*Jo!5ge~M}X4i?bQ6C}E
z@XaTRx7<=ZVf?#mR7NMU{yv}6XCJ<Mh-^+UTV^=C`(^9v;0Ty&!ymo=`s?5J)N%GG
z9RK9G8yG(YXmAG_oFf(%S$X(k7sGnSjALuzaOKO0tnWKpz3;y2S@ZpO^o8r{*E2Y4
zpZtxpwfpX?o&D7(UjA}c-Zl8JY`V*%F8;&CUo&j=3uGT@9|4AOWGQ*p7DP!#t0p+2
zvGN;IMCXWHDlRz(i9z&YNnMq<o?2ZDb4neNN&=8v=J5O%<{^+es?ld0p%6?%-Mdob
zXjX0IUeyRiVt)q({R}!N^b~)`<B#L>U=efoXN=xdzo4$_NqJpwu6LT?_8!kQ1W)`n
zJh6vwd!I+5sI;F!O93L>G@9sOn)-3^L$M}~m2q1#`h~`3MAnvvt^-qs0tcm6tFJ-q
zE0x7cu^5-ajqzf!vQVwS06NLUrU;OmiDYN=GvtO&{kOTr!BlT~dHMq%=k=CGt3@(n
zB$5*h=0d*JY8*UxU@Q?HIC#y*##x6_9R5Hp-w60qXEzAS_GoN<SC7c!C*L9?<x0<W
zp4&ilg)c}HG-5KvUN4z&>H;H`ksK%?X%m%Tcg~7bRn??1Gg*-F1I=+JzjrotpSM9>
zh=h})4<rS%ET!({8$hqXt2Q?uAWRck@Lr!9{S$zs$3v^Ds}F7e)|Yu*OT76*tJmIn
zr+VG!1M1$<2avBPd-LtClRZxk1}Bwo^Z_~2KOvs9Pjq<wh|Hwpm;d}j`RURrw{KdK
zC7II36)~`&MQD`(ZGG%I<UzV`r^CjyG}5$x{JAvPX&}u!PoB9N5!W`)6pjSKjisdx
z(C`xB)ahKnZx9Sa$c58tEjCG{4}`LXJV~&Cd+$-_wbZ+|)=K)XK1#fwVNR3De8tL<
zIUsRxB0A-A-8AO`Ed)=?pJW~x2Dt@gp0miU?!BbR1VAeiZi<yjTV2ONY51Xdtg`-l
zW2tPHiFLZN*;ST1asa}nVfV7+8gBLMmGz=6$*||mw-ykF|Ex@aTZm~R9ukFv_+8^y
zUMMtY8#)SXT-m_|WMWrQ2Un?^NnWzg2r~9?UxMfpjoR^mKN`)$9|H(M-eMKUD0P9=
zbR;r0MLZNyq(sbwG4~yQvGA$SgU^6N$e_nbb1Rni8)mmVeGPPAPav4m0nnIqm<O>u
z&?{uQq;b7Vu8i<l1Nwdeef|QK;a&j^a64zxJ!12tBgS1Z>7=nnBpQ=!)k6Qco5r2x
zc9lDtx-@Ygkp#87f);`O>PKxqwK`HLGhUn0-ldiGZYP~;w=lf>jbJ7#8C!${dWjDX
z=c9xn^nmJ%N}`J4b+Xw)A|vS8*t+QFPob!iAX-91jmAKVp)9NN{V#6OACLfEVJoZP
z%Gav1bJGW#P37AM`j3Ii8YQwBUEU}bogNL$nc0<kGz@4lmB|UJ20@PKLG#*!2T4kC
zoIKr=2HiiD)x6P!@_$Y}L*E6UI=4e>aMZZLrDOe443^8?tgR&ghM(JWqM>)BS20~O
zx!gXZn|67Y*+jQWr|M?N{VZdGRq5gk7!zhagn@h0`+$;Y??515SQu0*V)Wb#bVaaI
zyLw^4G*_GR3-b%jK3H)n=Y9-isT_%pzHqBv2OJgTjG|NP(UYow)v?_4Oz-<k<-MRG
zs$-9=4@o3j-&k181Uv1Ojeg$<IvHZgNu@fr*_L>a_WbH_Hq5gtheJLqfo(gIO$2he
zMkr9sWT|+JnQmdByKBzo!HRe@edJEU7eMJzu3cKZ2^x<?k36QJ?uJ$5PkeD*Ax6S(
zB=l9x!^=!r-Mh%~{FI&DK69aH1SVD1xNb7Up%(<~8GWX1s7hryowA7&>kwj)N>!_b
zC0Ht10Eff?FUp`akFrTLz{T2_UuZlDmQ>JG*fO%z&ka*14%|yBe~_;J!k9gyZ*^wp
zYIOtfABz#$t~)siC2qCKq0nG9Rs{T~(YksLzh#Q7+)|@aZZ>~{ekHO95#JYzB%_=U
zn{p~&sUY+ilw6bf8@TwJR9OF%=V66D!~jL`V4UeUH$HyzTk>L<&+c1POZT0DElN-C
zCiRnTm#-B=$GDXoRd?*tK#cpl-soFhb{FQt2o~{Se6aC}#crbgVElG>>*C6sT&`Ba
zv>GODb`HD4hcMct7t=UE^(m#bdvwy`#y4QAzp*F`jS<giV9!R1!iZvfiRZFIWu7Ng
z5$!daYD`TvQc<vHL`t%(a)>+<aILTgi{qZ!?#a4oP$-7LBI5X=ArNt2ESX4y5$q<S
zl8}&&gYRh}bCGL6Z3n8$S`F}QFs1vVi6R^vBUF~Ok#MA#5YK=e2qyro_QIk#hwV?;
z-niFC_D6_mDY5}qP2!s`lkNAj86QyG+FZ;QD*$&4`e<$^{H!GtVhFzooQb%O(PR;D
zRW4krH&MhQNluwcP!2m4j)!p5#v@67kn*YW<cWY@DYwJfK!q((!I-kLaf@69)(~V*
zY?dMfgGhXW4EQv0KKO{w5M`Z47NNNUU`i4MD`gXjB$8PJ#gu$W8}}<&6c%0%&=V|M
zZitHAZYgS9!iWVp@<KM|1Gt+Vj79T6&f_7LYZ6B-$bBU1*ggP!xF3S6FY*1n2Lxtl
zNWHEjxDrSy*(IboXrt?9Goiraf$b(u;$WAduA^Bert+qx$J@?i*coT;I27Smk50zd
zKq_s@dpt6;8OGkFaVMkCof4>idUDw83KL*<Oolglpr&MCI1d+A#zVb3zDV)OH#;Jl
zio88J7?(R4*m~$|LBySV<BJhy;9eMes&k_Y5}Q|$?C65pcE7!(PcRc?DuDSOjFMLa
zmMMs)1X)Fcq)Ca)017X;8G=Qmj4V~L7$NjI|3<*ehyaS76~07(bd{152_zV$7=b{g
zwJ1zR+J>|@%?KDHybM1Mw}ehQ@)DwBgc%HYx~OES{rcXB8H3CrtVCo#;&Z$fsW%Lw
zutB^F&ps4xF&%BvjU`_Mps0Wy#WzMhQ^~86^5|iU39Qa@o0r}0=!>8TGgr7DCIA%z
z;oKw`@qu+lGDlFMgc8sbp%AwhCXY+*Rd<2q4@DB3yAvZ=En<ga9rB73WLirWl5T}V
z4DTGx13<O`zk?pega9j$7OeKESiTx2_lL}z0EHHEBsDRYOsdw~2B|bOXD3#0Wqfc^
zBvkn%B9t$l4pW6|ljq^eW6!B?s83SAXOoeRp;eC_83*$0MLu(5Y6L}e*Bf7t%UZ`y
z02yHF+392RtJm+@V;Fn(Trb~hBWTYkBG*sO_wiAsY9QYmg~Bn;ZjU})AyAr4z4l;!
zenEXAxKW(!C*Rj@Mq-15Xds4DK`@I{$}h=JPYonfc=Ut$X0t1sl5<CoB1d8^@F7=z
zUdu+W@%%gTS$+&ubPw}FVp;ecHi$^=GFiXplhKa#6(0XRpTt;Tcz~gQ&nN7{MtSah
z?`P9N9}pEU!FFxC77bj?5CRB=EFs&l8M%&1`gYRKoo?i}aVlpssmg3Jva}hpwht+F
zXxj>HE=7{Fl@#^e7)>E>Hk&NZCL#lAaZp<LLYsp~Vz!*jN{io<SK`;**E9Mh6>6l=
z&E)&{9@cwpjgRY0jypbc*e7Q`ynlS=@p1K;cekis#3FjpbMBntKlnOcfBw)zfBrH`
zVf?QAEeWEFfUr!yHC_EL>WkR0s`z)WMxR217kh<_9lU0Ag;>~x8j2ZpbvR`Em~hcT
zuUm0(<|H1>SxiY^OU;2Ldb>k3de*)P+qnI7h2fu0&xAWV_9EZ??*|VaoG!JYM&Vd}
zwna=PUZDQg;O=wZ3}QE4u(y<YFa@?+<u{7hdQv1vw||4;!fNf+C+}EX%w(c9wPfdV
zX`mFtGczxLQaSH-0m@=t(7a>G>>kEokXl&c`?&zN(d)p0c@kUvr#)W;pX`4ta4b?9
zeou%wf=Xw_=Nhu){>045H-eP-IoyAdTVzS5V3vR35H1N8aenuk-}<0(BpFV~ECLng
z8V56NSFx5IIE2vrjI%s+xd<zz%o+MpXGNb_B;D2F3Q!sHK>o#L6a`^@a$y{9tdq$e
zyZ?EF8rqk`$r<6BNKkUtufqi3+oVn`kVBbnxA7xU(VQYASma47lwe!*wHj!X(OMeR
zzDzm@AiSSy<;2noQ5MM|02LteIgxh~%8=Yqk!nmeAGAX8FbS!nX99jQa>Yh2j`hI6
zp9ifY->(<FxK<a+mA;W8uap$lMfs2)i-_U^{Q)G`c;QrTs=s_3`=LdwM3dxJxkkTS
zXCRIjHJLPnx!lZrjeOq4V5E>q`!e_r?;+a9w)|3y?R>f0h3{qpH9^Y4&NTT4cSVAy
zu+|g_0n15Pd!gMfmkr{Q<qj!95|)Y+lxH>oCWZ3rVwq}Yn*f3+I>|?4l7kxxA>Qzc
zzM@j<PfclfQ!vzwP$WFKD0fu@i8x&wFPZ#&fptGIi&v1?rla}%^4>Wh-8J4U@k#=Q
z#D9#$tLeL@2p2ONE9hs$#(LXnDs3_DFWLG_*nDo&Tx4wR(F0OO#CZHMT1+>uPiFyY
zIfwX~9hM_1mg^#L9j3#v`O96Wt?6P>SDxK={N2k(>*W$eA%WRpp;^zACwZ*}0N;+h
z&X5e8Kb1OZ&3_@O<Xwq4MxlDS98c8hDe42)>x5TeHKDZXD)}q-HhIZGxOb^XdVyTZ
z3siK%*Q+P0I}tO`B-aflpft-}vC)gyw|+Cy3z(2uqRUijTRjAV%oZiLa-+AY+hS$C
zS|B51=IA^X<8zkpo$vHnIUE2A8|nr1Y0gEom;^w$mPEoSU_)|vF?>(H#nLM#@X03=
zHNyoA#nmip#cd%BPdL+1JwP!aRp}zM+#b8HcQW&XH@~?!Rm)T|i35mCS0X#$4V^0*
zd3!_o#L}0}6A7gD-?LTdyk~Ky`}TV5Wu?fKM^;I{8!{7f<x`DhD{*3mE<$Itv7<>`
zPaC@L5$~jQkv2|fH~55dWg#Bp!O~@<h;c*swM)t%k64ANh-Rx_B)t<=AyH1($sXRb
zan+R@q)}0-ZGA8O+U}*&xUnCn&TVr(8zwo1$p2XK0n>co)n;OocsRx|Kt_u<u1Fr+
z>J0}M7eLLV?v+e@FoZYc85kMt*U6OQQy+d(sVDdGczwdbj7@cp{HQATa<Av@^2|kb
z6e0>`mCMf}HRyHF#Z(sD>C8%ckt-Kd0X7AFR|7Y%Ls~mL7(F|*!Ea0^(Q@T@$mM$G
z&<ni5ztiL!K8f(4Nv@oVw`dl($03Lj2P*>PrC0`?z#BQJ_P<-*f&l2nvNbJqr$aQ6
zll4<i1~-n+il!uVjL6hj)JEMyq%<uogMl7mEx-exav~fizF!?A4A&Pc6QqMOR%Mvl
zh`5DDr!!dU0vCkSz#rLn_sv_QXGCmOA64;aJ_jfr+_^Pe?sNz-ASJ0(oc9Nk?PQ`=
z6QBUH^QDql&Y9}3nTW2o@iPUaZacPh^WFO*f`-tlce_i2PNyMbXp*yBA~I}}btzOv
z_qn6{$z_vcI26sEumsXK`aSMsw?jL+L-jUbJl&mLohzOB&Te|Ld-=5zP(^%Qh+tkX
zsW@T=6<qdUoO+uge|ujR3j^GW?vPQ5`&)w~)<z6|FHUMc0c&eCoPt7i0VXT}rWAST
zl#)m#NzG@@9S-t^Z4uRcKE_iQt;FZfDL*k4-_f~x&3}dBrkHRzdS0ApE`0japMJ@w
zU;HvhpZl<!)85HliQ(m{lRGZip8ty(orM%x%&5)~661g!hjswNCsjnujr`!rh?csl
zM?6n#Jf7tl=@tkr6OQ3NF87hR&lWS1kOJPw4(P*jc<K$&M6VOcV)BHgjSed&M3k+}
z7YU}97R(s;?Jmn%wJ~p1q!Pm9SG$}{<`wi&w{&Dyv7}q>x}2e!71UrrBP>c=1}o^j
zRJSMemuIbC$FSqY<Ix!53IR?D)KGi@SpZCiVCbD{m$Q2J@`0$|1kX6lG@vAoXnA=b
z=89Yky!a4s^1w*&<SAVv^>0Wr!;BLIsLG4=K7rWcNO@99k5L*Ykq^a#c8Y4S=$)j2
zNv4z(zX_ulDl+3@(Lyno4+7(Dhmj%D-oL=4%^iV{M50T2au+DG+!?PZ1|6V;4Y-w`
zK&vE;W+W(s!$lw5Jw0|WPa|RbC~gZK_Qyd0i<XHhFpW|rOOP9g^l8K=v{NJ+F`hAH
zs9;hk<d|Dd&cv(>784q4FknTH&%ocn(O^1BU{W#>sojt$QpwQ|a!D#3wDP%Ts}xTr
zBiS?tKnkm&*1~Vj!Ef$oZ0;Ia7zbU&f_zS0m_i=0C>z9K=*O%dd}aYxoE}g(>l&ye
z3_49bh{R@qU_l{1+uc5Zyr8&9Y>V51vr0Wwb=}#s4^VWS`bO02LA=9_K&6O7hgP*+
zzyXIP6wJLNVqT{RqP3JsE;27Zy~JyvhYV(K=wl@F`LcJ}V6BxSm^6685amQ?@7}G=
zJ_ZF;enQOI%}Q6w*pSJosqhv2_sGJ5vOk!QQN=wG@A+*iHPh^!-9&Oa81@jAB>nTt
z>c?5?!jZ2b()MxB&wHMoL_Hv3iThDp#v*vRE@gL_YRF6=M?A6IG>bB8=mK|cHJc)^
z$%LYx=~?mVF1l@{8(j)3NsWlZL`mwsG==6AdZ#ffm93hQF1ilH43;;}umD4>GoL#d
zQouPL%gpmdFtHLSD--lTd8qXp$pmg4z|}Hejb<X%Ir_<ah?sRI2wj0O(XzwSO~3AT
zJ`C?hL)MWr%F~lRWq`vOjT(XUs2Z}0JgT)Ya}VH_(VGAs^A^;@qc;JvjVV$+oFca)
zFnZIQ9)0xq{{8zOz2%mhZhDkBYP)&-(MKP}Gy`r-dLd~G^tSODLb?DNUV%{<uVM2H
z223#|VMg*=9n7qjrA;@!K*kI<1F|vD!is9D7|xf-=o~pGaz%0l1yFO$<mgRe6!oUS
zDqy-ps|{+leqH??nA!Jveu!DHLqh4C$s2f8$f!IcOABN&`p3ji9;YkKxR01mF;41{
zjzQ8ThnIg+v<z#Dz<cmSq1RegnG719zN7wDF4+oZ0>Mn7ScxZW0_K_6^sJGLXHw~8
z&>xRd2FDhYI0g6uu~01KNHmi$K1C=t;EYZIFB&cA^<+fk5b1~?7pGd<wsCSVeN%wo
zlgU~EO!-(G6dq|HfK1@L+tFA$mvN-RDKIjHOfnuxM4(S%A7s+R0)WDZX*re66`6fv
zZfon^i7kEUFV{MTwqZP{<B{{^su4k~>IC=DU6)3ni$IB@qbX0<LcSBOnC*64RwDiH
z2w=G8hq4xaI98{H39^GUmPeqwok*R@I#*Q0P{rYFb^B<fjM1ovZxQU}9>9tcj#(`9
z2U)~tuyOqHP!a(BaK_0-z=9$I7HcO}2h$MUP&*jPKxU<{gK@l;JFTG4hBRVNw-E>P
zMdFD7VL6;;_X*|Hqc3x$bP-6sVjL5`Vau7aSw}nt-blh-m+f`iUk~^sQ3hN4u13VB
zPnvXlZMmj^oLkF#Za<gF1`xZX?wWC@5O}TWEuZwJfR(dV`c>BI{!|yX_U+p`GCybG
zVXIuXwQp<R&X3c+>6y(<AiaDu)0@Cp8tS~1)e0>x9-ZoR%h&n&we#c8Hai`|H$_AF
zo}R);ii==u=P7yCiEX=qINw)Z&S~@<bFY!7%NyH{$T0FRqvdk3RiZ<+=7#75ZxJ<4
z00l*r6ZNethQpohdI2j_qp`7Z)J;g-y<Ah3G6~{aO9LraT@i?EcX{+hFwWc8wI2YM
zJT_R2E@iDlRy=(8@Zrj#^6BAl=nNfl9wotX_wn52?>(G*=pRBA<e;CR=IdQ|y)qcB
zMD5q1oE<73o`y+qNRb=EVEH{`B#4Qg#KX>v<7Q6dq5VtG-@|wG5a-!BjCn0Fnmk!8
zeA(_7b#8aoOcs2T#^T=A`*kO4aDlsXiN5YqKT&3o`yQgC^DhHMW)Tm^wk03ZeK_My
z(i8%wD`C33>4#=@C4Z8)`%6AKEf+j_bFeWhl*@77ORFMQx^G4T@0Fdqyc5DD5dr3V
zg1FoWKnb5k;zw*SLvKt(f^qq${voc3cp#Mg<**&ZlNF4p*F_93Rwi$_i+G_ZscW2$
z_(L)Iw^FuUD^$X@I$_?bT5n{dfN2p9lS{`F{mkOR><W?GRoluP@x|dZ!s`OUJ<K4I
zCbgr)4HuNkR49%<{J%Jmqj(ZS*rcpbn6h*D40U~LDvXOpL`n<hdCC%o!|*_=`QkUh
zM8}@du#@wVb)Sj{@qpq@M!!^nWBxF?Lj0?*$(>Tf(fH57crp@<r6NDgjrzl>cwhrz
z*v5k8|FC2#$^ROtVsfv66YhOR1Mehy%SovwDM{RE+OjIyyq&{aoohFn{-s2s8mm-?
z;Y!6hNp(a+3W21ciXkkBs8rlIUPS~W_&V$cc@vQ!(QZ3_GZR7BO=6wQP)q}#k_CQR
zBwDZ&hLP~4v!n<3jR;0_3Q|Wc@(caQ6v>;T$pp8W2wlb_mZ;G>uJSEH%=6rbbrc%0
z3sAR)agor9$2g7%g+-8nU~13+QA{S0@++Fhik)?P4Bk2Z9$ld1bh4<nCx9Vt1)mS8
z&@XY9lIR)!d@Truphh%`jxOxly>|IN?r}C^wkW2D592XYwc1n49}%=;?k`6i$|tZx
z?bkJHJQ4=W0LO|f7A3h@qMh`#X~+`nPe>ZuA<&w4$s~|BH%IOa-<Y|jA862(nIaoa
zcS15Talee|r7Shoe;|7b*B+qO{kS5-mQdBgY`4^H%wI9R1l|gY$)QBbK(($@w!sK(
z+zaxG+E4PL*fO(|@2C4k61cKG<cfZOePg+suU>fa$tTlNyEOeqquQZ7aJ}0hxv$Wx
z`OoI&D7zAiM>he;2J1Hk<Y6p&0Nj@7hjt%Tjy`%4D>Jc#-QMtEzfW)U5<&;fi=&Hw
zWjw4NCj#;g&qv`HNYSK&yQIPWM6?<kefXBQB#JLfuW4e!_#sSgLRoYdOP}dCa-2wV
z3lah`>0k%VSV)^TPiY$H#)gi6ON1z<d(2HcJFv8Kcy(aOk2k$upB@Oe#A|D!c#(N#
z-(rAb6IbU^9BnV9uuEbO_L^i|_o@{HA3_ZZz|GoH4L0DbH)iJBE$@$+AX*Z6RVsNw
zd$ia6`2r?#Ujs`jp$J5wd1q$#?wOrK!3%*^Zo){)UOVt*Ciej>Exvkfda=_nrHDBx
z0R#;IrfC}u0;Pz7c>`2}WEBic#64JMtG%$$Zka?sh<F+CO;4}ynV%=#g1Ij7L{i8j
za6%+|XaM6$v}YtmO!5&?+7LAfoTCa+5+|-aJK_1y!F8qf*R^<eKI(bO^J~Z@_4<oM
z2KYK-2E_Al<q{YUoh$mIpqlE3p`V+(H0gAhpX(Wt=O$64vRRz(h~FB%4Sh?p;Ii`U
zTt+01-`(9=5$$N)b`|!8IYS=5-mJ|V^foANdQ6PjlUxwZ)E|%bsIQFnh<n9KsIQ>s
zBVmuep3N6Rc<W<+a@Ek&xAy6zM>Av!`Xj6YceFh>H}5w>!A!Xgb|Fy~s*35rG)Aw(
z@yxJbU>LvppHPf>seljgM$uiTW~O6-ye~;ySWkTX;;N3R<#@Rgb8^vKJ{eQi=y%2F
z0QE|76~vK{@>o^pxxUdG^>rrI!(8VxKtJXq)C7i8MIydx0aHSsaM)<I29_tQMZxn~
zh^PxrDiw^NPMLN*{D&|UxqjqdX~DN&T|4>%)g1i+K6@{DPoRCL(}$K8LAx})nWH*7
zA*TMUsmg;TupgXUwH9q!q~~G&VZacinNMJUFo@5VvGNY`ET@reEEx$qkBH8;+v8Q;
zF|ivfRLkA-z5W!DZ=vqwTiv>G^2A0vjOTk5L*>>Ln_I1xQolXS=VQr*`QfU3&(9Cn
zU)t|gbA!cZBPO^M?bg=TsR+8h*h0R*wOl|-ok-Cl7>O3I9dca=!gwsofDljWr8P2c
zq+Qm52WeTGvBdSu-h-iDM<5JA&B}9heZAR$=5A~pJFxEa=KR6#?DZQP%Db`Wy01TE
zTb0!onTX`cPLqZV`>l#f=j*iB_RUOt`@PLU=R@Mz5BwjLt9oO7{RpTW&8AWHS{a*C
zJ?7p$HxPbpjNWq(_<YLD=UX#n1l)YCW0UBE5p1p9_8Gn2zRjsAZ|R2_ezy9HfT+Ug
z5;5!@|B<<SggEUnH-<)vC5N}3I>Lms6V@feY;$`^`a~Ne#?w#CkC+~qazszc#DYHS
z1FR1JdVXQC2f<-<;d7Yk&-4+?W(R|QFX3Q%^Mf_M+Mb@OS>PEU_QWXjj0`eAch{gc
z-w{V18PY%mHq?2__4$bDu!ErQOAamkrqLn#mzulLW)r~R5V<zswF2vum$(v)1^rDB
zzDEB(Yfxg0TGv?l{2@w9;xW2Nt;TceO7cc#Jr8?+)N|hRQ=Xs0kdUW(Jzi_!-8S4=
zIQOM*_m``H;2)w5EQrlMR1vq8jNHkf)^;GgOD=Dh@7>38k1cB*z2(w&QDt<?E+>$2
z#allQnvkGvYpa^Wxh>ln>N5|!;|q|3x(A$m+Tf<h(^}=3rk7;vs<U$gI~UZ&v*Pf}
zJ@HUWTyi6F8G0R+_johPZONtK?Vd?)iO22QNgU}9S{vt(9h*jz|Ax68p0DMqk=iZa
zfWbvdk<s5{H}#f4K8~iOe5a92RPz!IWOlK_n&y79WweYo@1au*WX+RiF1X?C`QI`_
zN*qwqL<wWjizh=xbImB3ppY3ENU}ejeLFU&nDa@fY0qC<&sBrPj#ZZA@-i`iaJX(Z
z>aF1ln)a1_aa>((xeVwuexY8>o9|fbz!xH@V`E90_xrK81gwU+PgSXUV22UMRp8gj
zHuT4TOBj37C>d#^;QhYW8V?z#ud&|72!Dg|hB)@Mjw*u?f%qRUK`7Beuu4&7h$!Zw
zw`AxY&CKhJSDiIVzLIgX86(2+2)h#k7z~CZ2xVx66@S84BG~XKHM~;(`fR7u=@}>C
zaGd###@+WA`wkd=6^_SD#=E6zgnxyN={e)78Dn<38S5JDF4*ss1SG<2sz+<{YFLf<
zQB=8!E0WR-W)36-G3NMjbZOHW`2^z6oA^*_L_Zy)vfLY>kvv+45Q9$xaUyn$pHfo0
zAd75qY#zQ0d>LO7pdzjXNe<iL=!_YmE3WxR3jy-;awj<Z2o+{Tad5Z2@*vZM5;6aE
z`{E3~UlcF>@`1EJmIwu|VklwC^#dOAiItTV*;G#3Y$=ce0pMRS)iTT%1-|+nV{`N7
zo3=KM$+tR+Z8_!-C4Ll_U{pdMUfW89{IU4&HX4nXKbR=ksV#|yXke0y%{3~SY&zwy
z?E|qqqF%&-@txf#kQ2H1FB4l?lU$|S>D~H^S<BR<PshMsm-V@x%$J@Cz25xZ4wvV}
zsDAv#@)kyah2zeS<zlIKl5i(+A4~gn_u_Y>+s|&rAq00&Zjqjihq8%GG>mH$$)vU{
zjj?`#Bt3dw0rk;Gqjo5s_oo`G^rL!BGeM8>q7$E>f@G9LbUWIqv~$^1(r@|eLqr%6
zC<7Nqp6j5?id?aIDxuz~Y4c}2U-EnliNS^P*H|f35@-qzjKiSR;Zc&3Xmzb&$H7WS
znFGhtaZ)l~8A@*4=r|VD!hxYJu=uhN1BJvo5LoS?mZ8fSytL2Q8vg?O2rG2=;W~7v
zksMTt0$Rf?fmoVvcJ=dHUXd#z?7=-emn31TnY7%4IN3yK)pxNn1#pf@WJ|~Lcjj2?
zVqK_A8%p3*4H6Gc=SnC6#tFY2&jv+UK#&XM%0AUSgWHQiFZ4^`_co**CW&|kdjzuG
z?pET%0z=BX5((F<k?<8pHbs)26=-#lD&1eO6wd9>By77Ws4j-;{w5sUiK}SRB3LIK
zh+QEWjNLZny97%p5;JbiaLfR;OVimfW(tCMj(lPh!;jdG6JqVLlAHf}6K_jwCN)X*
z5FmRHWjT1T4~rmNJMNtWqd4F%08Xc}2KJRhuh+7S=i;}GZS{WqHXyW8olX0&;G9B7
zWQ>S>*`*#Qv_;myJ4MBl;W0#plW$SUC0kX#ZOu<DV&??f#;`V~XRf~U{`+pa>gKie
zwdL2AO2^`<AN|6>vZibE<@#NlQ->Na%YrNY!}U^e_U_}Q%rgK?)SHdFuPT(@xj7uP
zS`p9@m!=Xiujx<kKX5u8`_|}>oqm6=kxPYZo7273@=`;2+bjOy(Fc;*GyQ&_O4jG!
z6!McJ=(+fh%!#*o!k!w^$`S4lcam?hjmd*g;kw+FI2N#txOc0I9Jz=9+_ttNIFd-_
zaJRJWa`*I%ac*>Bbb+!V+h4-^Nm-rM;mj=YPTr$uu3rCxLT#l!7pH2#X5;%0p1Shb
zu_d+G-+$n$tM>1oil@G_y1M%8v-}-{Gyr&<&rjh93WkmX05ey1P<ZCwx(uFo?XaqB
zwUNzRt$j&dPtUmcUGxC2))~ntR&|&5nc5#MK|!m-lolo3qG;AwFmZ*JRRj!Kk_Y3s
zH*1TfKz`XMYR{^*(X$nDI@W!Octx%4EN;|lwbweSY_h)`--xC%hf{%<`>JL%PM~`r
z=e?EAR;hGl^wo-sAYD<*vNZZhFw!1~g|{P{3-M|^S-Gq1`#x{Q2qcmb!%Et(_fs8k
zbnzngi??F+zS8qS&!3|*QTJt-l^%zDVB!(yx%9e(yo+EA)zY3o;K(69T|Otl79wx+
zWHbQ+IJ6_%p$*v~Wli9;?5x4(mPiz*4}|atF>xR4&M#cG%d#79T(z8Xfj$$sDCT0f
z=^ICyNi^vgST#$ZOSu%+I)z<?i{)q#Tqcfq9YtC1%BAqzkxW_rcF)mmEV_K#4d~ai
zO^-r`rDo@7){`Iy*<-jW&#n9OBrQsz2g$0)bz%ta*-R<erpiYariX+ACG(LnRn4c2
zv$$+&8OvdFm8iw&PlqUcD{+5u(x^~d6KK>%y-kb=Bob946e4&;(j)41k|2^Q;UiB7
z(GHuYVFY6ZvdP?X42Y6;5cD|;chp1hB0|Ega0yOFxB-Qhtduz6Ibb43+*5z`@Zp2k
zUw!5P-cajDKKYi{l3GX^_+lZYLIKKy1yDvx$ccoj_`aPDV6ZWeR=wV2nS(XT;?uL^
zK(HDnrzc1q*D65>DW`I{hUAAW8N&cp<g^4tI>Db5yMPKj_82Zu(Zb*J9wHgg$Jxd2
zZ?&!oB2lV$yhA?mR{6bzUWndKFgu0?DfWz$kK=j6p>me;4qlUz%vhIE0OgXRa%w=Q
zd_l_ZSp_t)S9D2buT}vy^P1Nv>Xh@a>WHrN6u1fgV6$1Yf@EUlLB}!0?lboH?bA9m
zvls&O0>)f3s|SG|h!e&Om6aGk8710XZfCksG)Ocs7&1i!(rmH`OrLY4R-7A~Wyjy!
zMk;+UQHzrbvAwN)=}azh5aZ)E94|kY2@}_mnakr+Q|GDUblJ|Mi^N?a7bly4cG98J
zo0DxfOWABT!Lr}Zl8H|!MgV(-cSjB`Q3>L$o+qHY)_6d>GO8=3x^gPu469u+D!EqO
zOW%un?lF__S7HTTq5=VGm8IvHHexfE+<3#CZ)Tp%MA(p}56NMPRN??ohvCjD3GCLZ
zh<ADs>GR!YEM5p<;R*2(2()cDCoGjxs0@^(q6#>Xw@vK+W~hjU!$&ipBj?heELmIr
zLLrIm$@Hcj#~ZMT%p}@{nUG4E7j_4=>|So>(IjTAY15doXYtsViO>x$k;}-qgDNr`
z6mdc`hB<4^5~xzi6Y%0|UWXiEslDR!aIR;pG4ogPXaq83FMQ|X@2HV_mioNYWOrU6
zDEg^O{7h>X!ujCpsjQGqp;OX(LNs~Pm3pQ86ygIV?P6XQ4(LX5r_uuCB#)KnBC+J6
z%*v+hFw4$3?y(YIxwyHxwK<?1U*y2G*Y1}ZVRvFrkH4hfPg7XM2huWSz}|n$++5lh
zjTnhu?=;nqVq1oIOjvQ$8;NZ1i-36yys}}mnzOUbmgHR->diCpSUf%tN=a~QgOBj&
z7C`N+(uw0|kriUwo0}ZesGm7=8z{YrYmOW(lO1DTb?n$Nf-xh<uq#FsRCGUST9v&-
zSgF|2Wf$jU`Djcrj6F907;?eS7c5!9kUQ<UO>**6lbI=S2YPIwK#T=u26LxtU;3@(
zDN!)S+x2e&19_45<lBDl{vRi-p?@@UsEUv7Er$*rN*?J7O$>*t|CwKzBgFQEL)D*W
zId!xvD2&x#x~x@6|LOzQGTxLrGDS(eP*t5wvhoJhwn*;=+-_3OTKy2^aHCQ+h7IF|
z)t#kBca~T;-hqA+S<9tsIZ$`+^ZYn`d{t@)Vq~U=v@(D{K@T#=<!(A}9W$6Vt+<X*
zjiHCnftIw@tz{2!S;$)_EC_xq$J9rafr@2~qf0Y(&?8ZnWp^32P*xFFUE>=&O@d8U
z2%TQ}m}yzr6t-64<9)t#0XLlC+v*a~Yk?a??=DC-ADHL=A2Tl)XthqCYPB#MGv;ih
zXkMY7Z!p-ue=zXXbV|BXFCi0(Y%drXxrz?p=jkf-gQg^43Vxc*Sa=~S{&EJOJ*B=2
zTVnExWteyozjdp$vIN<YLr3KdZ=FFyU?rR}2-YLlaOO1pGH7(s5^PeY@v^pBSD)?o
zyq}Z2d)y9y;|a%x8YE`TaHKUy;D$qpDQmphxMPX>#N*n{i<ZHFakJ-<o22_hE_OTA
z?L-v7aTn6#^uG{V0<1(^z`%9cW0uFK58)tR3%%BdlID`jEVP9~s$V9wFijanA2YdE
zZhIrq+8~>WA0&Z6t~^{WH-ZbqVuk0Md7Qjz^aXD)(5wMn096S1tJC-oO#cNm<5L6J
z7ZHvDGalc!jeiw0eCUrN(!H>3XV9<EQ?gh(?n|fl05lpV6Ne5mtX94qY&MXhw?zOj
zY>b@jg<UK7EYgnbr7V6(&%yBnggk=3H5upM8++i!Z9GT}!rMIWL7w%b;LQTz)O48d
zI@w_OZ&>5|H?;ZQ<NROkMS|5EUd^lbyzkMm=dGADc%cpI{5f76=MB&MW~YDqdOZ#e
zYO(%<`U?JyS*Yw-TF`@tVWWqP-I*|Scj(DVmobqxKiSlE`4q@b$IUb~Wl0B|e4jlr
zTq~9Q0fPFkIemI%)koRxy?c-Se29|hoBe6@;*}#uPd&{p=42OdI#v2}U$48hb&LsK
zY^Ty&TQ{B9v!`A+e}3=$LaiR1zUijdhstHJly}j;HrO5~iDT9o6|TaVCPPYu9ZbGj
zdDhksq!8h@nger>HWx1@Jo(r6=IALlo`?ze_GOi9<UVb+1v4eAr`48dt!}$eh!920
zAoT~0D<stR+91EBrd;sFqHPY%B5KUe9-4fs4wm2=K%<>uFbHfjc^7FUosdK+h9cJ!
zmh62sOLExI|J)r5w0n7UZucSO&~%TEzq+9PMse|Bi5~8TflA&8xp%&StGbQN5A>7U
zQ%HY?=TFxD<WJVWb<xwVlD<T>hHCrL((hB<>08p^V~Iqq)@s#i3G&1*Epu*stz5vX
zz#LuiyvFme=l!0Kdp<kSGl%+N$e?lwl~&d)>eliYv<FAzZ>BJgQxR^KFnEbQK$A7G
zg*ckLiRh$9s{C@7WO&L^wP8_ihb;o()e-d(1F%@%Ea6+be>R86SeLenvDl`oL>;}<
z@CgC{_V!4{Cf6<$baHEndZXPg5g`>+v9fpu(n&QBHwZ0$i$t)(YO75hmmIb$Fq;#E
z<WM{h5T$&mSZq7|w@O8?kNC#03MT!2klhO9vj021UQe*y^v6SJm4TA}_=xfY7@<F2
zt%3!~8Rp}0Dw~=04i$-~7iMNsGx(ysfe5hQX2wWlk`(j~_d?1KSRiof@oKG_q@*=|
z0pcnv`Fu3&hp`dRO{o^2&mljm0bWWn-RP7vgyBTOy?e$N!qxO9R|11Y`xsw~^523b
zMXn|fxesAm_^#)i@q;RYJ@fz-Xdb2>#y5H9dLD_g1ZxsKY9hgD1!}x*wQD1=4cI16
zUBqFrbqPSB4xP}lo5ebO?-&OFbdaaBB8Y@?JUO<zkNlER<-#Yr-X~eZ9H$kmQFOSS
zE4cJq_HndBKmx&;CzmcxnzMO)13RbrEdWeY?U32bhs;76*R@6Ne8{wFC4gl<4Cs2u
zYxUYufOZNfDFopEJymnThuA<&@io2P3<W+H(gb1Ine`+%5~Mk*EASD3jTpxy3NYYb
zEbx)C%|x2C9;}Ccwgu#PcMy?pSao=$nZ;wWl0YY7_xUogHhzEnAe#Q4U<**ex2oW6
zROUvtB)lq}aY`YlPo}>r_+i{;GfGaPsj38BU`A7fq*3B1hWBw<N;5Y6j|76TJ8jjx
z!&ZLLRRcX1>aza{<*OjVkv2oEA`%x>sVAQ*s4;FJKFFo#u$cv^j^aTDONsp_wOIFQ
z@nn%v;QH#iLWX#YI0{4I<(kVResamvY&M^KvWY^yj?D7fS4Hl)(^@oFR=pqhzS{l)
z>u=GuA5`xeJ-EEA-gO$2;NX4b@`+9~zNzNsP0EdHzxlK3Q{>l7P57IZw4e~$Sfpc>
zP@v@Pl7Er0I$d>rv{0yM*L0;kQz;ap>4OdY^SgD&;rx8#VEWxTBzf9M+gL1r75RDD
z-Rf;N)9GFbEzA--Ii<ej*-OmH|5UddIRk8}RA(q~Jr@6C>cw@8DI3)yv|<Qll-U3n
zu`|F{-{2qi#z6JaP+xdMb|$%@xwt^skiHpyYkL@$w1uvOgAX!f(75do<1y^9CsFrg
z7-JtJ-fMh%$896#K91GpTEs*k{~Sn8CPJn@483HNG0Z>mD5uaCY7V9ksKZy><*eJf
zOLJDpA<!1&dR@Zb=(=$$<C_)@Ld&t367{m&>g1r~1Ilk(9Qz2wbOWpz$(nWJ%d)zw
zV(}ZF0c4w18eQ5|v>Fbr-TmTerrPoN1iB6UL0^jA&S5V_&x<6U;bhWrE*=M$s&9ie
zf;`0Ay@(~0AAx0*q<3x=ZX?|bb?MMca&+l{4(*ZY*-svnyOpOaM-=XrrQtOccnLH_
z3=Y<u)Oy5<`b95>0K|!EGBEQ6afJ~FDM<9SP{hXn@Qs9%0;pR`$O=drvl__5q}NB4
zbWu|%hy|ODWr{WIk9b*1xtXepO1ONooNwEbHDP+gBqb~L=Q)X}1hl*qBU>Bzm{16!
zPo4EviQ{_PDNg7k9U>{I5F)kX%`CMAMV}a;pGkxxl!G_IQcHr1!23aMZ@C=54MYYQ
zIBA0PvHOYSZ{h`FXI$Up#Ss^V!bE07Jb`5rRAOG-DK;)ASIOhS1D=<diCP|e20%!o
zH~OQQRFEhp%Fzb3;ZgZxm=wjV%7G{($x#UwgNqXC7p1VRC>>aC$mx#ZX~MGRqsA=8
z34@Z#V!||R0<(M+ACcQ4Lt3stmRMmDDypjl0|Jjj3m_neWDo2{*_;LE4B)M9BEIo5
z0b8W5U?N2F(+^^IPpDf8*-)^b!-A#Aiu6S{nzf|h`6%eQ7(~-K3<xTeBDX!w4aQrA
zl2iME1ns5P3Q1yIgD(Mp$Cs%uNR{R|Q8LV<^+ro1w!<QbfozoYbLWyR=?3*wT8zk+
zSx$wDl{y+JPtnd~67iClk*R9D7_j5SIhJx_hJ}PPEOKGodc{JABz&JiO)i}LlFUYI
zT@;@x&BVh+N*bId!SfQmn}>vHSn?FGHN$S=0f1v70LFk|)Q&+KCCUu@HrQzZ+|!mW
zQH1D{#8D3ZOvJBz$xI?hAfX&gy2Xf-F_TP@iy#qwRB01PH4Mvi5`!g%6^oc%nvro6
z4@W3)4APY(p@gV%Ew*zx5-2h@*TWwpkUStc5+FsOB2ny^>*r>b3>^&FWLL3GQ6Y@B
z_%2n5Fee|SMc{^JgyUENptceir(C(#L;i3Mnfeg1NZ;pqC#(kHMSE#Q5Mh$*X$@&v
zfp8$YYR{_}nb?{U%*%gk+qU*1yQx!`!%?o(U(^f#mStXqCMEi~7BX=bw#kbj7{&ek
z!6ja-jZd*T8&7AF`*&|vhj_Z$4()!6Ia~fG(Z%Z0(%N5Vn3HyTSwiwKnkMe~7*V<s
ze<`)Ju|Iz7Uu*E|oamxwd?NX-N!ap5&(6v861?0ixE$&^cxV|^_LpJf_Bcb^P3qI;
zJ3$i}hwTfaYO)Cv32YIWvx_({3nI|-7V@h}ySCr+TU+pv`zU8G?0~1W%+WMU$k<NT
zfj~pBe*@s~RI(_p2!lxftC*DjT)yKcOZ3@tA`$vj^7Ydv!0Q)ZwPYgUM~>WtBmo!+
zqwi|v%V-Ek6~A`1SPBFl0l(7<hmwi#cBj+9L7qib@P@2V8v>NdW^5`K`lHq7KJkf9
z3^B_>$wbnYT8MbK^M&be;F&ROs~!v%69C(mN|`mRTvSQI-97dPKs&y|h>dvGJO}A7
z#vMwPbOr_zc%yK3xv{Y)*46r5Hf@V6mJ!YcKeEw}BnSc1;&mr@ZEdM1lKrW*vnNi>
z&8gc!Oh-5ZUM_@Pv)f(1hWOHA{x5F5wQ&Np_RY;tuJxyq`C?(;)-}uBF6K)rKvh=f
z=PPCQotryx;_RQ^%1dlxX?(4(MQ%Jse)t2-st#8uCdH1;l_IuiDFjX|k&6Ub1AM}I
zv;eLWGb2_?wg4&09^`#stxU6paE9&}<w#~FPgOSnA4Tq8?ob}I^a94|4M1e5CVmjN
zirU&xYlO_GgQwALBb4XBOa)0je)5WA;>Hp-sUivEszdR+N52VVn_FrCEbTAzL*OAm
zo#kP~qK<l_DiD4PsUfo;XWvV%z2<7_UgZ^8<0>G=MraeD6fjc`$<!%0D!-`IFTHp4
zzfr66M5J;Fc(+|O`?BAnX7AX?8d~1IWYo}hS_Gs8t#`@2(qFty@#^4Q*N)uD?f>!C
z(h~4#f4nsq@MtJ9IT&mi<@}j!OtmO)({AAwZ?EEcE|&*`O|#LMnZYcB8DeQ^bBkJ*
z6*W2^PM<6`CQap=%X94z>T%*%Ey2;35=a7`%s9-VR7vrY0V1a;Sb7<UT2sASUN2Rz
zx%Jj$ERp=BU;6FaRP9~ws=f8CZ@o*MF|IX7o-avhSnztpKQFq*CEMV&)MLCG3PB{8
zsDauDxWa;r;1w|o>IJb2=mn;QyMRp1kYK2Z|Cs^7pa61mOc59+be~ApTDQD>&z(c;
z98|sZ!1#5ge1BLz^peVisW#(lMT7WOV$tU46d@?3wFh2*?X`XC#HJ8AA|tiaT<@bl
z{>z>{U)|h%@18x!2(sII7iM#_Sqet8<#LX)7Be%=11rmSVpXsCpc8=-U?99`JF$~J
zr+yNj)Iq#bVs$2DM0k+gkt>TcESLhkC9vILt+op3?a~A8HvL&&%nTulwkg_*%pO}?
zqsK>*(WE``<2}@Kk3atSO}SbvmnVtYOr=|`<(Yh*8XoajIY)9-B#Uj?UnL>On@;C_
z-nFS|BqMbijcb1XFN}1?FaQWy%FIkp0XiOtgd4=s;q(;wpispAGPwHctM`|B{r*C~
z@1T?m2SSf-M8S(el#1sv@i-Ej-vn)wY8q9N{P)exVhP#4U#g@LHl)vBq<<gPVF31w
zQ<Am-G3E>Pdb3$5h9>#e7Z_(zIM58}*C+5Kzt;0^^tiDBkq(f#MTdy7N_tNEH09D6
z=~}vG*wu}@GBGD@Xete*<4{**Uo9uF;>C;hGJ!SqmZSn`pBwt1*<bVHWl5oLH3v{*
zv*4wA50$NrLakPR2|9cf5j|9GwW&(b?~!<ss8%{Y>)1lbo6FUyL`*TnV##Ne3507~
z)<Yu#Zrl{4%hQt1Pq*8|T4dYp=}Sv$Pb8nONis$iW)}2iv(+d*kgVTtkN%lLHJw<F
zV%=VE^a-SLC>@T|PKBg^aOfTNJ%kH*JB6bCda}5T#g;ctHSX~5%2}U({?R$k+nJi`
zxK}V<f)+q*usfXR88vt+E}LV@OUFvrxGYSMXi);YP0#i#&(6=Q*_oSv`lrvV4uK=s
zJp7uUR_9;;`qvN5`Gu{$3ybEqs?`Te>o@H?tnjdZ7rc%?QNM_9vdI`&@%$F?Gagmv
z)OrfYz=v0%<Wd<&o}`E%axSgJLr&z8bK=PYQSkB<MMby7zcw7WB$*55GWBKaXUoO&
zme3U)p(@K_plWn2xSzy7gJ#W05dWMWI+wI`PHyQDqVv7shFX6gA72n`q;q7#>Jh}1
z=us-`Y&YFpZbvs^l^h6a!8!EU(^ti65pIXv7Awh1L5&rQ`^0tS?!#Vc9pN2sTF^|9
zJ7WD>CIR>cX*^McQ190S*HbK?AC(XJp|<>F?qu&F`XjNhK*%ZyOXvqgDhnb9g9Tiz
zNFZ`(fl(9&Q#dTyKcocNw3xw*hZ`c6DBlWLbQ+M>SXmf4kpxs(Lir%wzfi1|Rs_$I
zD-x7dWPbmqgjS%?=}Zzb?1bI>ZNhl0(cd^utr3<y6zZg54kr$j;KgOsC30a<H7Mf7
zp$O{#VL}z-s_JB6>D11`bPnY~GX%?sAZr2R00L75!3JOk;MOFk#pk;h4T<}{^#2p6
zdck|3UIEf4CEp4Ef@nYJDkQXvMiWu=oc7o!d?e9%Ay|l|;-e?0<yHWPlZhyGyGU&V
zv@ED{xZV}j;aU;m*10iX)IyPi?gpkv9R+97pa7L<0g03a$%w>$q6XTfL0L`!ImE3;
zuLKCH*zoz12r1qK@FkKS>b+KO-tyJJY~ry<LmYnDELkfq-}yGY`B6^@ZK;fp+LMCU
z3X>Pc!F&+Mi~N9Uiohw>-Hgj}WvEFsBe^W3JYRR$6_D${Z07UcPlDyu`qM>YBrq%|
zmOAaiZH3wy_2}r8XC8j=iU496>7Q8@*uKLbe+f~8MzuP%FxONB#3)Y`9b$kE(ZVXd
zLm$L^>}mj73_;M8=vPfGPCs~E@)e0hHy;k8FMY%~aNvn=@VcYXfu!Cbqvyw`Z%4lI
zU;I^#<I|FO$8nzf#CN;PJT;y|a=#9iP?^Vs$=S?hW~^(5V}{ET^G~V!4j+N_wY3cb
zl$ux&D(1WL_%T6(Jbd_-PaHpf90<F}Mj&t%W@PVWKh$T90|)P2ot>*Q3{!y!)jddX
zlFSya3t-NB4<0an@xcclM5_+WSk_*Gy8@FRdn|s8pWg-SxEFYOvz`q+Y+816`{ojw
z+X~M!Bhs97nM-74)r#wa)q;}zl%fi<cJg0W=g@%8UHJI98w=+?e&Jl9KwZni_6R?w
zXq=vN=L-CF<Y(dBhc4W>T~Hr)HBr{+n65wI+8*=l@m%jAs@4rwmeapL%{OSjd*W4x
zuEi&aF5VQwDy?m6$(2am3UD1Bi8v_qaU5qc^DkC_TeM$8@sHRk!`}0B8M(`@bkr-_
zl%qIIX?ZMfGI3P&vBw@O5x!IpJYZPa3Zg}D3J{lyuak*nI=;$tOCQ17U^SX%=)+#M
zRKC+hd6@MhR|JDeT!Y5wpBvub`~7C1_jPnc<!z{&W@l%Ez?{vtgPl$k26QR@E8%Gx
z9^*aq2y@NSC@kV6k=P&+Q*k%I5m|$n&9sxTC~qA;AYC$kFGvKmerr($am9U0xia2Z
zq*;=6{6eAPlcFT76LV}jgRPp+th5k)jJ15qlKd}e%c@vcX(#*9PWHnn0!PxvSm5|#
z?`ADBnS^jsB+X$h`@6rgoQ;B?{L62*4SWm~Ab1&{)I4U1Pgx_s&@tp&ua`A?y=-HP
zEbD#Xwr&f45jrv6leO+vXp&S>mY5j9Ns-TG$_6eE2T}nV*yrUX4DBdYay=5M;kLrk
zEO|}nsS*Ucurh#W%bv2VQH-QLpr|^wV~H4hS`vAqx63F0ax9L)er%OxX)kLfB474c
zSV{9Q*P)|!_(aQdAD}$v4v#XU32g`8=km~Xh00WXvE@~|OZF52MD~^Mtg!keN`Q%W
zg!mzIh=+#N#GU2ZiON<v?q$o`r}p(MQlU2s-!mU#QNFWSY9BVeF_<KLOHD230FYYl
zW7hx<9WyrI&@;l(Ui76k=J8ge-0!0_Gvy{L#^0BHY3rhHBHT*9p8qW?2iUURt?s8b
z9`0Uo`SNTqny&VASg`)+Rnt+q)$*keF+C$@z`{jHL73r1OfXy*>m(3kP!G!>o;~Zm
z^zqxuT$X;i?BmToZwg!+E}%k9`w;O+mLL*g90rzq-x;~V$xAcjouBeccH8~h*(iJM
z{^PW~+|S7l2yB<X45M2pV6Tu@fcnht(q*5|+ed<f$4BKt^)L6Vm+hp<CMVl@M0!d8
zazB)DX5Z+$PO6#+_^zmk##g?=7f4i7xueowg~#=0!Tt298@+<kt*z)tQ;T9|G0ckc
z6~-;3-QxfkddTnp$BW-mUr`@}=e0FA*OI*m^6+Fmp<IeN<O_*cOUJ82AKS+3VKGou
zX^CviGA09lflpj2S1VECR)Lnr2TxUc%m1Zvm5de;zDf(K@AFe5B%2kRdD<VEUA=j8
zobAPT5memUY>bTIP^4NdA<kMv8BqGNl;DWbTD81bxfjqKinpY*`3O}Btei48_uhQ-
z-c3`!Hx@|F;3YV(UonoUFMw_QUeAZ%==v5kwU<fj+Wa6GLNd61*D@wsTw9sTNMw5h
zRw8d5eyZC^bcJkj)dcqEEk{^^Lc3egR^_<b_^Q{B&nJCEQ`ewi%Ms*<vFZbdF=1@)
zzcU;<v%F-0W3x`)d0<I^FuI*BFxV14jMW(L9W;BPE@g5k5UlA5A9%vf7T#RK#Fs#(
z53V;F%2--H6AItCf7?jl2=FDbXl?B~6^fkNCutwaXNm32%Ky723+e<Om*C9eJ;#p4
z-eh?1$!u?5j~}>(C&nijJx7Pb{LO)UI@l&))*FsxIvuwVVW*Rck=PWCw1eq<;O6{r
z_2~9ssm$eiYinyquG`+u+~YNWAa?B7%Lg-Zj)m1LxA%3s-uEf>K5w_XZ~MyCg^wWY
zO08XE-o32sPF&A(*wfBq38!|X*tnp`yIk9nHY1%sQRZB?c1MT82z8dO=Rq|w^#dom
zyw+(e)$XjV_O|xlg!NC*hq3W3?pv6iX<>Tqh3~!h-iMZRiaI0bnoX+6tgn-<)!n!M
zGH{}FddevljPJWa^{b2df{e9E++LPQUSu@o#_*!A%J^btz+@O7|3dYp^oo>lk$;iw
z8aSe5fa!Z<$E-{)VV<yaEsV%LvwazbdR)#?PO_qAy#9cbNT;JjmmAn6ip^BY_kJw0
zxzSTOY_hVXPRr6m+&jgWE8ToSJQ^4`XBkg+xLzpM{YFvQ>2|Bt7FDfWEG0t9ic+X-
zR@40E%STVC)6YMk9Qu!d|CwL7v8O;A!6Tb)Wt}~AvN??=6^VGAOpOcoH%J(ZRL~ZK
z<#Jc%s^KYt(Q+s9_d)95XbD#fF5^1~LlYZaS0q@OzLPFx>hOkl^PEK(LqMR#m^`IA
zTIbR;d+a=+Lz~kGnSy^BUh4Oj5)wDdt99hc**SF8Sg*Txt}GFqGr%RR4cCU1uzFv=
z(XhOU_*Ax1DK?1#_m|77v*r@AZD_IHu+SRi7{-3DF*AEZtFc<Ekztq38=LPqb_{TX
zz0?0cac>?b$yJ{D=84RV$bH`{_pGd}s;+&@>b<(ut$k}rfFy*}LTE!65D1J77y)9j
zF>u+80UIR147LX&#*8nrfoTkwWd<x{8!v3~1~y|~<DGeIQup_s$n36WBy<0`_qmx>
znUN7EPDGsLJKy?#3u!3dQlV;|jf&;O(%c--X1Ut@{K8fne%6j1eAgy=-9SH>53b`w
z(({m<I&eZn(&U5S$CDcH>i56@{qKF>13!K8<Vmx2a`eTOcfRwTCx7AUSDC5NU+TXC
z1O9rG8GTRxuQzFwcA-EOS@1hnC!Gm2q4icL8eSEyLzWq0R*fGhFYgpJwCRWYD2C06
zuYdjPrgCWXUF8$RN%k@G%P3lp9XpgtkKSY|qrc+d*T4FWZ+zo{L*I>8M(0=JsE7i8
z*6nt8j`fsSW%PFsJn+DwZ>WFU=#zHvPuY?^i=ixw)#P6T9Z^PI$J=(QzXBDNA4*5<
zd(+pvXe~PH_U4mVtC)e5jw=zP42wL(%4HX&yVL0cmSAkHnK0-CQ!S;v&ljea<>szn
zS@JL=;ganm@+GDOM*9#8CstlyZBfQZ`yjtyf`Np_oJuRFO#`TN=Pfn#eC1lLHC?NM
zKh-AYu07pf<X5p*C?rswgsP|&@<sF*U;w4_1C*}7lLBDcXtgsQS#C1PBzIRJ$qQ=*
zvn{Ut?9Fa>pd{!y&^b<<pQ7&;uy@>s7WOrGB;CY7dZl@@v){SYx!SqId5&|h^JeD*
z=p3vG7VQE_gYdSY5i~_;B32`K_3_l8g{6()(0IoXX+z3bskCTmnoQ(oYZGaMPc8ne
ziT=c5ib2K3#Eott76zIu`H~Cyy}%^nO4h4hwfJ+-N?(!{tOvNS%T=%T;OIfh_I11R
zj8cF5vYCG;oWYf~&CTi?exI!(u057O%2p;9ABE~z%^nI+bd3H$&`tR^vSQcn;kW;W
zT&TA3joW>C%JG}~3SwrXeY#dq<g#|DSA*I5?9X*M-3<MUj&Ba{bjsMf<P;s^2hGY4
zm6hTHCmLOb--V*~+Qoru=`E~t5SHwNQATozF3AJQhBZN2vR7m;E;VWjor#GL5mACE
z-Q%CZy{n9Bu<Ioy*p}(ZPvfmpTAuuS@n)Ef=vAr8m6nFEM%7!EtDv25x=pKKwx`*K
zVhFiC9)uC1N0RE9qH0A(iG(KUg=B=93=fXj2R{rq%*0B%B;A8Gb{yjQa3_r?BgGO~
zJ6+RgrNZ%80!b<O1{g{ek&{W*MYo4Ck0Nc5d6TSg`6E;nA;EAc*?KjzPa)Oh$3c69
z@PWGS6Jct`rBGV<G)hyp(Ny65gD63ib{uq6st3BnBqBsOl1&mw9?*c4ngJyt#X?15
zv-m|BE{*p-Sviz4fivSi5>Fz6cGW|5sA~NSBokn61Ac9Sc@q8Ls?ql{nMPsXzG6L_
zac82W9CG$WqFIua2SGI?o(rcG@XS-V5s`=-96JLN{OCJj@RBi&!{P~WHHZ_88(i>$
zr0l}(E*RoA;t*-NXal<jVTk7zZ1YN9bi=Vi5;$b&5NH5Z(kjf)w+jyW0zo#7<ns9>
z{+6s>ieyxi5FMiELta!zA~4%(R)omoT|;D~i~bnEych6++=|K-FB0M-I3biSM@<OD
zsnp7GuU}k>#KK-N`kaM4${{ycdSQu14dwC+Cp{iCR=l*>$1eq-BGjpfhQElT3UEl^
zUBr%H+?USQA$6VyI-iXHI8@Q4M2oTmH!4^MJOpA1g5~geiltJhibD|q@OcO@z~bH_
zK{Px?1VWx*IN)bMWkkN9g-)h$|8zKW5lIwfxf9IizQa&~sU;B_UOBe`&z#S@;HOCz
zLxRnM5h_P39DRi90nh>tWOS3@7MvK0b6OMZKMU1GQ-I&at>MTX#|umvWq=qoNRw1T
zSrbL4ALDE~RJgGz22%(PzXlI_jCdB2nI81KxJ%b%>HyI11H3&G$_eEUx7EbHX;25s
zoJ<UD%3_A0Glj!PcP>sY?8S)k7Z5vOBEk>Baf<|FvmX9)JPl6_Y|4juG-i5?A&yx=
z?lCmKhjoCK)?G7eJl(NNs3PEjq-hH!0X1O<NQaDTrVly`riB0(qb1ydZ46Rc;yzVJ
zYSO1eD=@ajo22OEB9%%Lghhp~N^M??ya_1qaFuM#4B<j!A3Pc9uSuU&CSeJK(gm@r
zTy`pQgv!KDGzzd~$TqQRIG@Y@F@0<QJm=et-f)Usi$SsoI0k-)92bFq2EVfiYS50k
z8h*o%74kJX@R7jkuvTFhE?bf$N(YVfml}u*L{R&soVNTVe79xfSwih13e`lF$60<U
zmLho0u01+lC@qm>FVSH-CX<!D5}v5~4UjRJfxD~)9FuBH;(-iadRvDNTUu+&xTX^4
zOAW_<tqQ7nat+R~rL{g#VqAu0%I+O6fGa(rit4$n=c_fPaqR)O055P3$l6+2KFV8M
zMA@GRjpac-C6E$H-+<SaP`fpJgaCZfreeq^bDo>-uB~=HgNy~5G2{w@1kq(8;%X>s
z5l(VY43RJ$5xrpSlMam(UlAZ4XGm=>zr#o@ge<JsY``<2jGU_Z#4iXqn?hg)l8ull
zaH9#RqJS#ZaFo*?e@rVCgk<Pt7Y-xcQwS2P9bA~XtLi93Za$5E_URhDAYAx*vZ<q7
z3+e!D8L!;8onDvV(5~aOJF7hAVOTh4f2;`;cs{j?*gzVRW<Oy9SEA9w&~`;Fnf$>^
zU;0wOYuJIEf!J@*xKn8igG7o)_{PGPr~~yQGgl0F?I2dND-?VV9S`4VNU_HJ5|TF6
zL2X^8C5t82HT%G#PHxg8oG;4SA(}0j%Z;YwBFS@7RE8=r3!x{pA+5kmsyI%M#g_1*
z<WWf|#~Ki#Dd1{Z6WK%ktdb&6{D8jL^;CVrGwSAEi<OV?xTxGn@r;$L(CQ9q{h&EC
zM;~Z~wUwwDgL`D-Dp?qIrH8Q`#pC2S4U0o0EYA;UlgqpYA3Tj6aX$Yehx+0ZBr1K<
zpWwK5WxomK)#K8h@ka0D<0l+_)w^zg$9B2IsI!?7N~PhuZuR|lY#dQF@Gy7Uc_Pol
zpnwTw1eSL~R~(MDOEcxrg!5^+BE@PzKzIc7{`zE6T2-=w7-h+3vRY=4v#grJGJ}df
zl?I<9Vhh2d&{VfVe_3_O#9;+utNEV6RxaL5c5UhSArC_v6Klx>rX5{ZZ=fl#wEfoT
z_2&7b*Asl!sFPjT+!M^_r;tcO{UrT}E-hIJh0CE(8s0_S2-$wFFC_Duy&-&8DwCqF
zrZT(<1A~DL=MciJLaA7i#Za)wJj4lEq=3opJzJPKVk%r*^@y8O4nkyU1|!8a@)8hq
z&d?6N(B7VI<O?BgHJ`)PIv+C63llLnx=Z&2>A<6xaT=SNA})<tgC}NfY6{^DM`nhg
zEyUu=cLH(!@NF3nTA_ie7ppc?yBxz8)?xx%Vnxtd*^8A~pe>w(EeF@-dYM~~z*lNX
z=FG6Mvg~Q|*T`foEQJ#rs3a<vv8%osI~!>n4A+BJ&L#c88G?=VP)CR(d?qLd?RLw|
zA&)*Cok6?PoY&?dKJD*z>+F^}eQa*FLiWUBwX$(+cCK8>6spzD(N~Hj!8-ffFJyHp
z6S?2U3nJ_n6e6gximgctpmdQ#mO|YNf`|<j)3@%lgdWpMGW@oV9pk+s{pGPtpz}1g
zRNlk~rncsJ=xq~;3@JcJqu5y*_WLL$n?&~z170BL8_#+?Xa{vrjsapzl}pwaB2`K!
z-gkY&M|c6ziZI=Lq478U{fnJ7d?}<1rc&2k6IA5(8gR)+l1-rZEo;?;7{^l$c#0e#
zcp`x|T@`vO@A>Z|MN)H8Uz2bi+9h;GSC_+k?#BZ^bJnkkXPLBx9%?)tttm{#CLWVF
zmc}z0IF+%bH0{)!cKx*tx`m0)FDFj`o$0HO?ew)a%9%mXr-Xak@zIM`u<Wd^Ty){;
zisP)TUU-FD%%yUf1SyRf@)7bW7K`GC6E5tyoF~Rd7@bhOW??p&h?@C-_=kU3@Rgej
zXEuJ~s^OO7Yz?nE>3^Rz&*cJVE_4wW{2s|WW4Sya-@w*w!_kVu8*jGUhuIVV%AQn8
z(59hKz259ITUncM(2MT&mNnJes8i2g(O!D=46b={@3omU*dG4G#!V5)%E+|MXk$@m
z3o$8Hio|Ji+~fr$Qf#SELE8mqoBa&^r0IPoYts0X=MDWeKb6=AmP#758C;Nw_K0ec
zgO~Oe_U+pTa;08gIhD&NJJCpy(ME&-l8)Wa;(7V<Zn20HKZ~131H?1<M_`TOxW_Wy
z?OaY^f&lpuFt1w1tc%Ptm4Mng5t73C1>JC?F*iHg1+0b~GpQ)xw7C}_zW_1E3kJ&I
zQ%gVh>A<PLivw>C{GCaehM70}%@yWa$if@V2hD@#_sqA=zkzIcfpg0FG3O=DTWGhj
zWUe9TyWz7goY9y0EjRf@!!uTnX)W;IlVg)_yH`FOKG+k=<Sk|a<YMy15Taiiu6)3R
z`e%4tA60}_AqWFUrB=~G$Br%kQg&`wA6y-u>@ClTcU1|xYQpJ`2Lq3!+oXi5x9_qj
z>DoHiVs$-J4JUF@o`2vDm~)#Y>en4piWORyVzh9+erRE3K`!cyvEoJ-;f*yOj6NyA
zHt_lM9qLJgKuwmgYk9XzI#Plj)IZ4%F<lGQ2TvfbMFueW-s|?b6BgxC*;AHezExWn
zZM{!yrD#>4lutr+6B?}kYN6^q`{-~8WdU8Ri9lm`S2?aY;0f#lsidBeT{ik_$WN*A
zxVGLr_xQcJ#SwMx`0Yy&IyyG*snaN+9lA2b5w-2p+kPDKkW@TWSGk;~yU;}^Z`~P;
zhxtw@1DS%eCV&#|Jd_5JE|X1n0Gq{)mGG((GJMV|b;QI*!l**9H2^Dv$cwnhl*$(c
z$qxtXuel$P%bGj-j^!ti0GWDlt2U^L697-H-C5+`HDOuSojmAwFywuh|1jW7aLxtk
zeK1BQa43_|3Pr<!9IsZ&4T6)agw()_JD?_eQP4O_2-S%~ZK~GJ=b81PQm%yEk*{1F
zWSr19h2jA~2Jtj#G0@boRhNH!76?9*FUF8e!XYUb&^r)EbVnSWbIhg{`JfkB+J^0c
zjW170x-jOW-?QR5{%NE~j}j3Pm;Q#wmFSYJ6cDosT606G=qBP8Z9>{kp3Y$xS}HFJ
zSQe32;pwaKy2FSM&kMWnA*PdzGb+uALW%wfMk^&R1JbIXTys8#b0J^+0Y2zvbIvIj
zwhfI2_d-+>c}>&kcQP?~1rVK|z<Uu{LQ=hh%1yd&4^54Y7z;V%w5qAHFR~IyVIs3T
z>94s(f?>@931|r>C}%zI4fxF!O|4Rn+<?fDC%Zhdi6jtsw`#Q>QgpUNIG09_5Lc2-
zZyY!{4?I}B5)4VMLD>=~#aOghsPpJj)ZhVE&g2ObCYU5fx55!Jobb3s(-cp0O<Kf}
z8*VmjUV}gcpNCE^6anu?@@JW#uBVkkm8nWu`a9Q@q)|eo%_YAOMLlB!%?kNEA$Ph-
z4mn{U+fjo+*clEW(PlDJ2$yiM2#(28D;(Z8H-}0t6(xOZ1Wi5}uj&N?mnEDL3&+}$
zd8za^R>lk?YZ8S>;vhT>1wmdQECUgY#g!r6!F`fhM6!!u4Yd&(5ticcfqZVJ%Djx2
z3~CXUekPH`MH=35I6%^Qnn4^k$!vhvNf{|Zt;B7GS~gjks%i?<LAC?9A6J7_UMT%6
zrZM$oq6jC&07gxyREJIkO({eDqNMP4X?_9=GdN>}cQkIu7aqoinM7w!B!&N-7nhVT
zT^i7rY(>t8P2n;;mXby3x=sqP8cAqyfk5`r@96F5JYa*vLXHl4F`}waV!3F>2t72_
zXr$hd1~(BQ^eu>P7~&z^D$mU=OC5^7Ncu{NeOSk_9xIOObLQ`{ISDi9dgYZemiraq
z^2y*CA~|+!XU+28{sF`(*JFoWM|^Usb9tXx+EC5=CqXprN->xb4vl}l=&$AcI0c|E
zlqytae)7oiqubll)3PYbns#7+-{0P7aVAOuD}=*)VMiIxrkTcHn9q{iwT}GdnCYE9
zDJjFL`L(OBzI=bR7ELWKuCCTUmMfRb&`skM@+b_*8wTfrGCfiuGT>L17#tb%mVUFf
zfA0nT?a?RJYE=ki0u82<$Ylw93?`FLhG8#u(YHhVoN%q$YLza#@X|tIX?p+uqYwsT
zhByYm=0g<FzKSk0OhAFi(1m4b!DZLAwoQvi7s2v)5u7%zksX3xI(~~=%LF0_Ylp13
zMfeB(5+^+~z+se6+m=4I3EgJ(Zejv^0dsn^qnM$36?)Z;<V$A$mIJwNqDh<s%9XOj
zgG{zQTj<2H2ozJXDP)6u9z_1pj?U%(eI4xJq>({|9c;YujXS~FLIv|Ms$v{S35u`=
z^B(Z7f+<G{nyNL@>2sf}ioGB3;3w8;jF0SjNJ)ZptvCFC_Mi-z(~lb*@Xu!-T+Q(0
zZudt&{DIu#YePow+tUuT%07&sKe{P`vor>AqK|z6oyo`0?i_=MeK`my{|TL|eLxrn
zrkYDXT1~1aR)q;;HJ;?*^sT+{8nUXPUWE6cs3s7SDaYj(ZQmLIOsawWhkbQvx$kp2
zI@}#_7;cIHbgrgoFKI!ER_Ee6pE}PlF$77X|Mht=ox4Mb(bCDKSrUJk)q}em_|u@N
zBY#c-khpxIVs62GVNnk3={QS1<Ql(z68!9#Vrlf*n6m_JPXftXEU8?LLb9A7gWYMx
zUBupa_~_bYOS47QK8NFa;o(T+P_O4k7Z(rdJ2E@iK4@C;ivdi-o4*S9o*Br{9yAq{
ze#lh8dFtf05S?*SDFTz}{rk%Hi49!r<N?IhK;}so5wEw-UzjQ05l<5=5Kb=~T$m}X
zQf^ANBAJPzi!w$?a&`_K!%e<Kg6w?*FQPqnI?;v2!w1JVHtcqt7xl6m8`+-j2j33d
zT&#ZV(J{W{u`_bBoCqdTL)K1O3xB~}9d|3;=FtUK!61GYN+>NhM|aMIzn<A55C<P8
z!pw8|YBe9pb!sS>8Q@c$Tts_v!nS^#V3{xIq<lpMBG9vJw)X0pq6llapyjKTyymUJ
z58+YsA$%=nke?W#mJ3I!(fHA5@MZL^A@?qcUS22WR<LT??asb^-E`}~Qj<<<HJ1*w
z(w&8UopyWnG>19JW(p5Z4*m5av4w<Y*mF9XB7rD5rB8!F^{K#10<R)_<wJqb1ilpb
zqrkTU-{q#{N!#6AVccVaU~tZGJBT7^{_q7}b@3UtpSYWPa%06}ZarZMKiQys7pI}g
zIfdQ|mTv0lS-fit`-xF_y<k0mRsR;Axm#!({ZH`KxRT8<P%a8HNf#LAv5aKdi%YmX
zodIIx6Yd{HtnRD0hKDj6t2WDl)-@_9V`5RQ!%a>ePz$Ha!oopsvxvQ6NM);p4j6ND
zD0GT1bEQ5`g+e#$JM<|rYsQ=oJuPHLFS_!|lfS&&S?Ye|BOmE5>-Y_kE3drrnj5Y>
zc;15#KKRPTPN%cnUDCcQPhPa#HJ7T483uzxyJ5_oOJ5#u5g0MqIr@z-KuzX3PWaBv
zVCdP#+_JR1ociJyzgR1LVQTc_x88ayaj;W=I6iTMrFC(2bE$j5ZMWTa!zDGekol>n
z=BH9idcFOrPkriCp5)G=f2Xa^xQ~y<No4r93I9nLR0UbvX${n{+M?M7<Fm#phMS@F
z6jl@^utk1SRvjP6pMr0d*M1arz6U!5aZ(l)@R}{_`}(UpRbDHo-5&f0Zf2I7u|r*R
z*3w7D3;fW8W!Yq5C&CEKJlFuf(FBEV#NxrPP^dLMJ<T@_v>6QkHrW0^@&tqIv;X&*
z+SFk5o3l7Qk=Jox&ezuY&gSM7S8Q&8mA>yp8Z#~Z3>3%IRI6E>{WtuWLeAPsC^Y`k
zM(AYF9N*!K;lS1Tp*`G>Vxy99l-l=<(QlgbfB0Iu>g$iq<Kwk&{@DJxxfr%gvJb$F
zz@TO!4-2c7Ckc4;*9(Yk+fgROJi!9KqTDVOo10+<tQ@=jAo?*`fSFSRYp*(Q4gILv
zPkMWGwN@`3Lu;y5g>+b5Fhq}Xwr%av0LVy~xZ~VQ)qJ~BBg@LnJ`{CipE83%u#8m~
z1zN=!{RxbhxeV?4R6U&zfy|kV%*;;1%w1-l^Ob5k?GhE5t@IxLwNj~c*>njZi}QXx
zoUin}%Wz1#OxnLogXLT9zWeTY{O<dRFoUm*m?B2uVDS8eN1oo<{~~oLG+jPMpG2Qs
zz{@fK;mmq)Xo?KR<BTn)un+m2WtZr}&>k+DPwu??J!a>d-~42o*U_79I%;+vgT#WQ
zW&$_dV7~}(`I@|UO<?*lSOWXSHKl_+%@TS#aF|$|>jL+~FG#y%|E-e4vVB84LZ!4=
z7k^+|)Gj0#BWYPU1zKHP!?3(1*U14E7j)a}-Z{U<HN#U_#$4yAPgS?vW$c^Y70YSg
zR=1?RKFdN`1qu&e2z^4jK;>x#7$jAsW-!h2Fu+U{^q|DSMKPgB+Coc;@w4^pOy;%x
zRWg}t`Rg+D@iM{1#7sc!;Ii>2f!W7_0wFg@5F9Bo0A=6WLP8*8P9iDk5mW0Up)gj|
zAjZ`sdH(QaBj=R02O{jbE`fyEbY>R?0(ddpXfXVULge(nGue1Orke~I1scspe|O)4
z<1Fmk?Kd0Ed?`bE;56uM;Sf>AWSXtEpeF_l!;Mz8HdVmG4ZfZjIoofa2jA&qff`xi
zE>zq(eI$j`#L39-Hy<Zx9P~MEV44`69t}_H%jUf0x4cwfYIhIbUz15jGnpRzZ<)JF
zcB)BlF6V6jn{v-{Hn*;+H_zYR?sSY1((?Ju`ZZgdj+r~Od){&+xq8{9$9I#-#Z<PQ
z%N_ihsQTK<N(Wer)V$|)XOA4+e+Mc8pceR;J@c{W=jo^I?t<sC2gTXT?Cw788S~-L
zaxs5iGP$&R*<~jR=_5PG_WcvpqZ-&ZJvTpC*d8vgy5ZU9K0&?XtM0W$Ok(=S)CeN8
zZpjiGf{QDP?3~#X_(NWQeeQ?3|IBYDG4tym;nVKLZ_vtI^a0aA?Wbn9r0a2icLGk4
zmji+av<6=cJTO9R7s+y={fJHXA}c&?^u9X2kkQ89)o|jCH9p*cYmPPUnS!>(>J^14
z#OwdVf4<L~?=fELAd?+n#7ttyB`n*4QT9iSKle}}W##2yxx*ru&s~^`Gr#Q!nf%Ir
ze=kVc-t8+_+Qb=PGkQ_JSj>+%+bjNt`Pkm>+`B!xKhs8Fl)@|fXe=1~s=xWw1MSSC
z{^sa=KD`X!%c^H8Fb4vCdPn3$51{2i<gut!iFU&-$^NsckZf@4=G3#E^{mct`}pzg
zVdq(Q+%fw8dmbR==bllPykuu6FXDE4c<-He-n;Vf!w<jbJrB>!?|sU7IM{$E*^7Jm
zU?9wloP(C0BFpp7a#y;cb-Gwn$x2pCF62!bmN+N$$>?v@x-)};7B3kDA+oYzF-G8e
zH#tp{QlI%Xi?nR}X>(%}ddqIS7gym<tghLbo~_p+Ms2C=RRXR;6~d_w%zUxYC_ZDM
zH3+V(?(D3tI)moI`ue`Ftncis#bW3$7Gv@278Vi*4_<un!2`*Kg<55PdwYAhuTuL%
zytRv2k}jcOz=R!nO(@JS(r-JFQgJ=({<3QZeN5A(TCEhkxDbmKK3|JHa20tv=B|1m
zR%<p7EG@d|4cu_~{PWLmuB^6hSX^9MyxyQ}X`L`0@Wq+*l4nLgK)XjCM|7b0JVfST
z8FBZDL(AvTM!h&97Y~m-s!4}7XU#N^eTMn<DY7fR9=sVbdA@|H35M`6obcB^XYN{7
zPW#oslRz{p3hPWUa5_V0GC5uNe>4QvfH9yKhN{{Mv*H<tm}BigveU;aX#qsf1ixt2
zT3f?MW6>i}7%72Z5|)F(hn(QYg8Ewnf!~D7tQebKG@4t<=O5A^dh3Uxv9}e9`{tz3
zAv;JNWFHeIFx&}0<tnFGnpx0+{nHB34pq<|0(E@7b12+rtw94R<gU8o++{+45OkjD
z@E80la4CYudPJ(mN48lyBS13#K`DM$EOuAC^wnbGuJ9ub;@NrOq!yuZ^JsK~$Ba2Z
z+$fGbUbM-r<U`>qutMetBRmk;adu1_{F5rwWC&imjR>)tBk^Fsby)6999ky~w_LOg
zn<8A9natBOJI?uWPATRbFVh}9dhh+t&gc{oxI69k{rcK@^z`Y|{NSgec~75iw@<h4
zJ$35d``h>O#Re#lfxr>sz;}@f8bqwz1up1^t)%zofq!AG<Q`xH*P#y9--B)xLZtW}
z`rYyyKF^6`2_r$Qzq*@KL@OBt<S{xkoAs0$Z5lBmJeGg420D$lb?~%Afpnd*_lPzA
z_#BVM*bF02UenS-ZQ61SBi_`yt|f=7hNXU!SGMXX(u^`bojJz_Oi~j(7|XxdUy&CN
zq0bMEmSAZa<~y*_)HYG3032Hru-O%0u9^N+&D2b%w_;Yi-P$kG@ST=J)Cvm7IK^6>
zPRrEr$;I8Fh+}Teytjx3Bn8lDq$t;^Xwc2$b6Pg3xfyohyYIz<Ztgd9F>{J|DkP0$
zubU#+jIGpkKXnoMdJx+(?lKW&Bt6P`{}*bOM(t>|_Urbn5+NNp0pm1MtLgTTXZo<1
z-QEhn(JSQp+#AqF?9OL%g=s=Ki$&#UEfy=aMI~-PpE4B<i-=_8Tjrq0X5!>X+>F|Q
z8nf2H3T%bM-N(~FvPY{1AiQc7bi7CuF@O}2jg450Jaq@S4xVB}iqGBn8io~FAs=Ji
zc@vl%YZd|e?tt<m{z~8zcq%cvFa_1kG9!;X43;c2f;_3gwtPNzY#|ou@Ht&eP^@YW
z4Wm_)(TAco_9O-*?_$^zA&shu&&VdL0}Ao~U>D<QYqR{LN?G;Fxpn>jbmDLZ*+C=U
zrEO&1E|sqhw%Uy*p;V>PJcKFVgJ(=v4ZtxV7#I3vr6GvJsn+N?vu!%<g@qM>iQ64q
zShG8)PVK-ijQ$FrD4d%9<IfBK<HP>Z|Mv=2OgMJ<>Z=b6bbj~78+YM`g27U0+tH-U
zAo`Iz9StE__OPVQi~XyFqmfE^0pm@zc4YLKiw|I?OC-w&F79`_^<~Gb)VrPj9eba1
z_wjxpdSb*%-1m6LY#U9^`v0upY%o}>NOxC8*ukzrbPbqV@J~K3SGJK+in7@Sk?omi
zao)kY?rTdlIvCTu*9ewkIoX7Oqeqs?44RA@5k>VTtFKRpw+2IDHfKe_%R(Z6lAc+a
zisKc^dJ8}?2jjCgsiX+O+=SXnkG6RRUyBQBc4}s#N2N+`WN;4i+iH&f`O9y6+o|*C
zp0i(EdLe2uCi?vH%a6~^o7bK=aRNJFGIc@-P$&xcec}XZ$5G{^;+Xx?)`^IPxKRWA
zl+lbNb;9Q~Bcs<o_x}(KNE_&f=Fy*_r=r;{2Vm2f+1Oq&mo*O^zWli>oo2JM@=%<Z
zD$@Jjj`x0|nRRP!_5=w&rJ+`mY9Qi*4mJhpU%bb@<?J^PfG>4{<yCDfLt%VJW^YC8
z=1|WJJ+i_$6iLs`zn!84mWa5yqE^!MZ)$I$NK<On6wb1mi7fT>m_^`L-%)t7w)$%&
zzc@|(0d?c;|28joTFrLQ$u?>==hI~8&IdzN8HjkjzJQDq4Q`WLT&{ZUmb`cqm@Yws
zL{|C@{79$pYV*8~X%iF?EH@pXQ?$1gtW}TI8Ff#_rU|1|tJNE^hK`Q9#3vj&1dlW|
zHFqKG4}5JRUTF7vl{{`rp*Y6q&Dm}zpZ5;UdWmQ#SgEY7!FaPQ0<ric=56qgG5IN6
zp!LVZd(_$Elt^8imv=mqJ4qVoOs0nqZ=LaO0!8#L-^JIc%MX8as}*=l$H7~p{OsF&
z*J-tv_U|4W9>qrJx%eB-fsK>Q6l&G>5PD~3*80<FFlMXmg9ne3%wUa>bMy=CaxFnB
z$=Shv`a91xe+^wvli78Ry{@ikMUMaNnllrpxXGl`P1zgt#ztCE>Y1T>+P_=oCR^rt
zhpxWvz(I#tx_q)hPORb-aFk)9UlhT5@Qszlg_*vCF7m3&&_z1^{=&ZEbKp6a)|;(x
zsE+S8(J+|Qs|ROi9dqwd;~YNp)Iooy)ymFvKqf(7SM{>3*38VzQ;v?Va+Vf%b`D&<
z4P(2#ef5Ezoy8?||BLOG<20Ic^D}w*oFo%@I9wvRr8nWZzlS`2Cl-MJ9Qbc=uFw!Q
zG|W{G27_rYAPZSa>`>$(5dv^a@~nHY?}onu8PRc=Hf+-?40Jqt$T+8Z#K@J(5<8Sv
z4uls#w)~`mnBF99YjM}ou619U`SBu7b@uIxfE<%f(-#HD<A6mvtR<3@ty~TZip*io
zLph1R2yj$b`wPfS_^puzSWr$9lVk>yckrZ0(E*8;vgeeDnZ-!dVyH$Vgpowe7jC}!
z=7X1BdXPvVh!J!on~mjg+KCplw}>+WK%&KLcJ!fzh2_JCmlqfH`J3pc{cA@-#^Qds
zieO2R*Ao27<ExJD*j_q{HY^yCIS(fx!1bgY^Y5RtbO*X0eMORe?q#?Q43rm%$6Hw}
zQkggoO!x?z;WoDtLEBcu?w)<C?qPgC28E04J%Pvm5;@=(1J3|8_q~D71^xlv?hqM|
zUWgPR_c9#1tnu4QjkeWkd3r~nV(wVGx#t*1wuW5+j`>WQ=75#M6&xs}7D1D!@9^@|
z=P;-qlOdJPCHc2{+nzJD!-OkQ%J#d&etRbM_yLKG2O?+3%2-=v7Xy{O991Y>)1ENo
ze2pvp^Tuc)^}+bfX<*Wz1JHY>H&{WS;n8}QZ}bP8ZpR+RZ_Gcf+?O}K@MZiv(sTIv
znEdb+ms5P7H2Wwsc+&xhYHyx`>nlaCgy2*dKPflO;WIUr{{W{iJih=;#NQ@J=wUIR
z&gIIDW>ek)fEgh4qJT}rk_p@<>fldED;sp(!Bo9mYqgep(|tnbc_Edi3B)sQc6P4$
zGen(1v*VxbbUF%d;v^?I)9<0ib&_~%fm=oCOj<rs(Ru-p;t!JoF^lLzh(#_}pIuw)
zt5RpGvoL*oJ{8So8n{;>qet>WN=f0tJcVBe;gL9Zq%*iTp{B?e2m}KZn-HyVr9WQ>
zJr$R$R5F`Su|yZ-1f)t6sX-SQ@#lk?mWhMkmPWP#G$W3S5*mWvcT7D`Y&x6N2d0XA
zb?(r4=<*V2R*p!Tl+78Ti-R*p{9&$CN|OGJg~<z(Vu~0aRHv14vs0~wNesu_XU)lG
zXL-sx&M8=!X+j{ggR#+jO)g93dmC|sQXB6eoPSvHNb`t0j|F*=W1d6SD!j2`Ra}>{
zBzeR&Zy!2_f)`1>&_Pw6v<`AKi`s_l5qRryN&IRZ2Y*FXi7z3phoLO|*!{X6H%j=P
zl<2Mr=F*kFlgAQ}jYP~isYn=2102dvQ?8g*YzzL+QPmX!ALZX3`*Y_ww&FI~3^4Se
zQ`1<tCQ`ZbFpRa$i1&w}H-MyD*$O453@40Vy*1pSpbTw0p`qKdg+kr!gx9t^a1Ij0
zXukCJS^<B-)HAyCt!}T`hS)rc*IUlq*q-iJ4^-i>8qI~#!`Q1c8Q6zc-&|A1lho3s
zmtPvkweO;>t+=v#g?rmXR`F!<R5E_U>t6Rd;HE>tsnPqJ<~>S4S!lJkw+n^hD`J4E
z5wW+j)$`BGuBNM^Utw~me$#vuY?uak0gF8U`K(_o8muoocvw@uoYw=g#@KxWk~@1m
zHR&4Gtvwl<6BH+ACp$jv@V7D)6NB>hr$7DaDVI30(cj~b7`zjyJl6Z4bW--mBEjrO
zJ~8^+VDx|8{O-Hm;MFf<z{rg+`?=43?)qoqetjh@SSOET7#Y)I>CyXvSWlk(PAZjl
z4w%SnrnMC<fBT<C@5d7{|Ji6RFVBSopZ@fx{{Uryo=bM8KQf;;?<b4WFmSuF?Tx*t
zt>un;^|?W-Ah+Ek-H0?TRucmki6Tb4IO{fyz37%z<oKl6^8aTT%CSyW8W0i1flkuN
z=41G@m1l89nz~B?|F~9^Rt^@p)w}L5C*njszEc`G;Kp+=yz8#JEW$E?%lp?l@X~j^
zrJNVeB1nkkK2_8`mDB&fDnZ9eJzoHNqkLO2m84loFIJiZw7a@|<bNn{x{{>-f(}zD
zE$gO|<vE}YrW%z>i_<wfZ{DCnDy;T8>%Ub+*Ny*{={!+Hl!q`xXR{2*MkCXiZAvTW
zNPS>Ampw<#&7+ZVMx(hJ!676J0@8oH^)M1>(J8$%l`3vCI3Gk4%-7kSxPMUohtK?F
ziWO6-ca|KUU14(<^1o*fI@^h?R}Y|28FQ1G(f&O3S1x~*)Y1r7#@h~GyG;l3o-A<U
zg~%Lghc^QS)1;2DkN26qewF@Sw*4CY8vK%!Jx7w;tKk3D66rnT@p2_X5<ZNI?(q}h
zAQ8)uOI-4BA<d-0*ek|@$a@h7-%6bBFx;flQjGzT6~PoKpKAQ^NDYW71UAQOHgFXp
z3K}0n;zy!|%0bq|Cp3w?XMF=yP$F>Ur7fVcDyYRI%Imr&M*7RVhnm2N$5y#mw2-I%
zF#b!~EMuH-N8O;n73Ci>@flhti@?)CvA7QsgF03%>dy{8($D8^QD&>zl^MrZB-$dZ
z<l;E!3Qm%vgLU0em#FkaJ2e~(N}OrCqU>8VM3Tovitsv!i)VM39n`oaumTildh5)_
z%c35Rs4UO*X~%P8UNm$MPO0QAj+m=)re)?}`y#7FI@8Sn!kj3D%{n74m2j^lKG}1H
zX6W+Tv)w-D0`^M3D=+2z>#i3DK=%$->}WLWd0rTpE2ttt>HTiLfctv+f(V%d=}udt
zTobAN8T^PcD#Ba@#kl6FugA!uzrf1or%fD4b2|fFMiY!_O=BHmZPzPNVhvs$6TR+>
zHkbVY>+1)DkbC5+Y8S`ccqo;g`FQ<z+Sec3Eq=3xRJwttbQ(UT__VLbRz`nX-oC&e
zyH<Pfdq_#9|0`GlQ-cg4?psgaZ&J?2#x*-wx}Dfkf38EUxjQx0P4=(s`o`O_%3<`%
z`|i8%qJz*L9_z7BI!C}T&9Me;2YyVETK%q`!g>TudXm1?E_upKcCDYY3quO7v>Iy~
z2WpV<tku%2^Yq(as+pU%7R^zKJ#s8mVWFo;gP`^iwa`9OM@tK*luGlQ(e$(ZFZS`&
z-$;W`7ENQU$8MB*M;WZ4wc+%1sSem)*<<=4*M!#BLww;^?~!nRtzMd*9<Hwu*H^2J
z{@3CHEL2rFR+_1GCwOf4mRoj@1@-+cD%7Q(_>)`Mo}Q`^FHlV5!x<F{L6&H(josai
zX1&hz1YZ+h-E`4SrD{{tb4yEef1fC1;sO>%pSdu*y)!@WIP>#6JNxGCI<S{^9{Ulr
zGtb0k+|g2Uv{Whz#vHTPjyOo5+XLPC38~&MqVRq$X>rPzT=S&s-wHDne>dV-IasTu
zL$^LLr5fFEUzRoNmRmlQOg@COi}`W9djgOB4K}(T--DOG2`>Bv*p1H3r8(wom<9uw
z6wAGcTk$t6({s+D$)2-iC@h9<SZ0?nRm<$~H_q-U1B5I7h;OE)d}=6UFQL+QV>sC`
zHw}lwi!Z<VJ@}b*ht*@4oU;8y8Q!On>Xs^%O0?z?q#G)`)fW1`V9A-nU$<xufq-JF
zVU!lF@Km%iH#ZmQr5co2N)4Wb`LjBly=H!XzP_HHBU{DhmEHC8&O2}Sx&_naLW705
z>3Y51YfO8kQmI~UT-)pQa=qL&xbIYQm5Zr6U58&xCcBBwA_H(H+CRuGEyq`P*g2P)
ze;!0>CwBa2$(fki%6^9%yeaxDvPqOmi>1Yz@Sb~0{E8=2&2DXy)AC{SF#L4_l;WEL
zcQCGKwoNr8Jn?YA4}bP)#ucqpr}<}G&E!4rfhB8}uvzw7hHgz!5wvsO9w%wF-x_}7
z`0?XgTgR_qex=EDf>2cM9~3s+VCJCNCeKnSMZ$5SkpLM;7o#%^P6i~iT5W-efvGc1
zE>{vsV+u;A;^eqbV;hD`)jzSqjeGuE^_7pD;_){VQ&YX!dLyzGTiAEx$ilwp=op!q
zvhh?JcpI`+rBX;(L2p%%@$;=_c*9*-*j*pQ3h6Xdf-t8-K2-|G^2)etuK-=N6VmFE
za$vJh{BNWW8^`Sph7+)JA8>kRaP7Wh$M$dMMlS-gC=_09uQ>bg8J$g?DxZJi$}6}2
zimbtF>&L<~eL!7*e*5xlx3{*o^J7<Ev$hto*Bh_jCHaR5D(tNHwAy!z(K{To(dQ7z
z({FjpTQG<%FQ0#bfSi|?FF3UL=H}=-P(sIn+Tr4)?=ADTx4muO_Jzv|cwN3w-=_c8
z)(f5K{?G;x_w_xk=XQ`xzG*%V|6aE^anb{FNp<sCBCaiZTC>Rxle0&p9pM<v8po|l
zZCGMN@tzDP(fP9K+c_8tpLgQPPd=(|qS#RixZeEE(Vb%}e=ryfz&mfxO-+d#C>DzH
z=pyhXluoBWMkVJntOrOw;V@iR4uw=2X&Xsw`}`+8`RJigcvtnJ)M8<EgV>P@g|**j
zu6jx(R4fjni1x%>c?>-wzOcMx=L_n5WZK+p0yq|Q^>7=6g0U}{pjq(X`@v(MFu#RA
z!Xh&%&=--hZp7RPYH%7!-<nR)o*8}fdj-B#vYV8FG}L&n%5#E<wk^xrs&AFFb!oqO
z8TR~NalHd>__#-C(lxGoqv!p+8(H+iU*YC{2Oi~-M;?Jk^}g(d_x-9DAu`<Euc9}F
z-5VnA11jpii(39uIJDrp$K(cZmFr#{4!<=V?z-M*;4I(ox-b3mm%q%C6n?Mk&i_(4
zJb=IqRrChey)f*)Cmf!3CvqLK$!YYfIdT<TgJd^}%jdylBCvWO{B*<u`9y^SJeT}+
zw#qycgFs{1n;BTWD2#$79~wIpR4}TFFre&VcB`~@nV%(Mtcs6%Hdm|SslI_ydG*Ia
z<w~c8*U5pK%Z&@1c6((Nr-FZ`WQ~xoU@cGI<Cb1<<%TgM2lt0fsn`M4iV;w1cR*P*
zDb;qi?v-}ixuT;2xqPF1*B}+Vja~8)T775=JfJw~zscc402qRQ8+@be)I-#>r&c>o
z$Svi_?~=GHaZ~F&8c$^Z^_afB#gE|S(Z!)NIoOi=SdF2#CXwE@VP?krfYGK!h=bK9
z%`H$ipQ}0(b&8FKR-?_tPQk04b-Coa_DVXYH+9&zwk2h$UR{T2&NsD)+uJ(J_NzWP
zEb8#zlY023^Ur<zHS1m!sa3Q`H*FrqfwP8|h*^b`iw`0JJFY|qu-W)jD)JG9Ac1lS
zR0P}*NP%w5i>ELRAcAi-8mRe5I<DwV>XA3Q)p4PELa)YH$D4*ZF1LUPT)e^+1w2D{
zQFkz^Ol|lpBg(;+Ri~&ENDVP$`lo;;1*)6SKII?hT!jjvRumeBtsaXMKiHx8Iz~_u
zjiwV4bTYghk&yz%Ag(r)`2OYO8pI%l@Id7R^GOxX9q9o95LSLXs>i3)%cP-gzyU<b
zw?Ne#g7gpu8Z@Pf@N4j-m_w7!?x>f}L;>{_#1BOY>C-Lh-f0F&gjOXPf|8+j&BRE?
zmb1}_sN4{?xDU!kqD36d`6igJ2+A>ciNgf>*Eudj8+aTvk*$|(540k_hshN9Sp+_Y
zDt2yQ{~~7@PZv$mD73cRb;HCWgGJI3tu)pi^|#D{XYvC|Db{W8+aE82QyHGwFD)0V
zoPCQw5eGM5&$<7TU;c5OA-+!ke40lYN43HBmDxy)R%R5*W^&%swgFo53fFC?Uz+~V
zRUdf%Z9ize<c+!2rBCmBR-zc1X(Vu>Le2ng0Sn~i$NSZUGAt{3?O=)&l^2JTOY0{p
zxvkiXbI*F&%*D6N<gWS2?VF?7e!Ln2R7Doo0322zZ~}Y8DeL3?l)$qBFGHI8MZA+f
zANW3=Bh%0VpE!00SQ3~pe)zpFMxyWc;}|t33H|on{(L4tGLd}Y#~w(%iCxc<4h&In
z#(SOPMKe*T`3c(O$}6j^&Mpp7o=OXV;Qz!&p#lIb$}LJxQJi=tt^)rzRRd!Cdf{=7
zTR$YL#=~&x-|YB|qHa)!{URKk?2z}8I8pzAnD?G@Z}9ku+h@9o#O)_eIA(1LSEC<1
zNQ#Qv6Y*~U_7lg0xC||>nL`mTMffyCo@tYak(f9VAuLfp3B@8;$xq;GVKOWXCf&_I
zHfoMj8wd+j*7E&!58g+1Pb#tBL}wytC+|c-I6E2{L{i|v0*Vt%MK&S_HAPTuYjQAS
zA6o6SamRzJTI`VSve6Th#iZ|a*3j~0I*YSL#H<WE6J*|za8nq0P6>uXIX6X5;ujBx
zM34-JO@K&;Pi~@Zb5a-X!>*4Xy4m+bzq{~8Ot&MElylhMEGC6jl8hvGE@i!jhCwn2
zD2FewL7|c*C;FgWW?-k&kpd@cV_iDd;SC>PWc8qHmnFIH`KU)ywfb)1K7x*2B3Ks8
zyGQq#myGW7WpDEm|NFrM^Yh+~#y!p*=;KCrM!DT)+z=L7b3X#~IcJRCq5Z=0_0K47
zb-M-_?ydb)qdBDeJ`L}Ant0+4_if`Od=5g68@Oy(s(^G?9%#}J&By=jPp*B{Yj1tZ
zt*?2~lb&?lldqkGN8EewR~KJ!$t7<;uzBDLaazlps=sX8<zD0=ms#2j%m$W0%RNly
z%}WATA+g>9y}BEZ`<DfNnh3JD2i}Kg;co?gn-k?1h^Gk?$W94H^M+0SrdFF7$dTmi
z7Q=a*Aa49s^xY!IU+3=BL{#OdjeIwjqW>lLN#-9H)XY&v(5)fv7G*`!uzy_2qi6m7
zrLxW!S+R8$I7<<1s=`0x^r7XiE+4w?_S*~3d*1Vc?$6{?**ovN({Y}SBJB46;PruL
zKl|B^b0_iY=M)-!`7i$BFLJ5T*V4KFJ~{9YW_9$tW;GTWeTenI`vg`It#{8)<v0Bz
zZP3%rrvMrPh9DT)4~$Nx2a06kMD~;6S?9dSsq{6mYT~U#fW{t*EhmmdPB7h^(XGiW
zDHER`en<KXfYGC%?F-2*6TM1|9)>=2a}o4dFG7!XKG?@f&f`;Bv7!@pRE)W0UBqVF
ze}54o?}b2&X{oWH(-Z4XW)Wan1|?4K!f*N3SWch{@i7XC0ffx*YXwq!$Zm6nGt!ng
zDRFU7b8-9jQUmj*Ru-E(A@pdO-<mA8s#R2*Z;NLdXc=R%8d(~3$P-hl#7?jyzbaD`
zMn>QfvM8DoBo~H%5D!6WVF5wB8o>@-DacLej)5)3MFJ#K2d*gownUM1<K){lZ#dHK
z5Z~f@om~3itRuXyeD>iq;H%O)*QZwPMUbYsfc2h)gCT+pBF9$+u;g5YJZuEdh(Qt(
z0>#3k*W)t}f+t9IoJk&H$2pjW(1kLoJy$_96nN~@=Fi9%SVL~OGVtZVHv|8Z6%pOz
zcs)d0WgE{nl9eKmvwcUSv8Hffg+~Mi@uz4EXXvdh5OMvGG78bmYk0}BG<s5E^m;w3
zf<`Ry`2^7p=^<Hfu)0Ik#3h$i2Dyt;N~(e#-<A+>z0Mp%yoJO_!S=sxNec>xEhneR
zO|YhNTjHE`6@lp3bU4%vAFVZ&+dA`u75?lG+NRYe;(up*KR9LM?fu(qkA}EJt#2GS
zu(4iesoT5#@;t~SaClyrUZGR4W#_&|$Kq7sMot2m;ubs-FT+GfBcyN?|BL;Cl>(#*
z7-fVWY!~pIuH$CFl|a8zj$z1S+@&JerlNSw{)KuJloOadu0n7?9jF(0Z|^DN?~m>d
zhn;r2I~*Q6HXL^StwYM)3DEu8p|e{@T{nAktTpvkGC4DX;Fm*g3D^@x4jmeg)UpM#
zTSwgD94;gp`HZMw4D}B{s(K(doGfBzVXzk3I0Y0I8*Zp@P&XUAk!s%;2iPNG7Gbnu
zfWZJMUI-*6;Z~d&CgJj)?J2M;!$8ooQifpI9Q=L1Z$8C(a2T6OAn)rsWQUYJL>jRY
zt0BX#Y-4>rO@1w#kl3>&7*QHft64jK$5R~=!BAxAFeH5u&90>7KgF}5eG3{JLHZtl
zMu{H5xP;xIc<XU?V%=}u(nes#(<cppOPa@z=8Dnt&Wjew2_7L^HBcT&6Umu%SbV3H
zU~(4vM@hbhe4x~m=>8FMj`p=bQNnJS9(~rNom#bn0|8o+lqtOMg$1~bbg)}%B0oE6
z!)lm#?32V!pAKwbIeG;Y^n-!FfQrso?w|`L)KmD+u?nwXxBMry+8sL`^qYAMcR3-u
z&`zB=;NO1Jc)YE<iYENf2ecA%dnQQG8~8}6njpevEN|IckjzX2TGJGc@Kme{ga>Ko
zK39a&yBNCI>%Ys=j!P-yw|hxc-BYe9{l?=(D}c5=^YQ5Xz9%`o>80h~I!qN#xCdeM
z<o*u$W-4=Je%^IL=_*)85tNGU6nd#*Apr`|^z<ecQ;>9HhUG#l-9lq_K=As7Gkvd)
zW5w&l2b#8P-1%Su|Bo5A7O?kd_p&|K>#n=*e!S-5kyI*^Z8i}})0sMgWCF_u0$mu}
zd>nDa^YXb4(>xT5<f_HEXWkIc<pk*6+`bjf^6YGG12c$RX=3O+vpC5xU*d^IbNkfc
zBU697ggzTrF$azm^I<+mqU>;yY=$P??(}CG!QeH^RXnm#)uy75#givJw*!~v2KOb<
zhd}bN)Zw4+dMVdE+-Y;9KfU{Ovdm)<rM^NrKegXvQc0|OM45oq!xfO~Vc-gdm;ur!
zlFY|amcN_ur8|tT&J*#a8ZHwrEXcmVB<x!~=8rRro@L%LdZ)Q<^iGt6`0JS4{O_8x
zv%6F;0Dc*dB|{*|#Y*US`%ye}rwF*qkVE6VwV=75Q_eZ{ZOd0Zd83%ytk=_pXtap@
z7uvoVIF@z>L_6<tat4Zr3_kW);FR-n69Dn%_P~444%=BGCqss`Z)CNKzcP0W59TYr
z*v~T|tJ&{f$3I67TaSAlP&RsA|18fp^`jqS=^`XdPKg@1zOU`TLf_L~?Dy4f?B`rv
zS&EtUJF7S~Y(w%>ojA+HG^z@;n@B97M2E%0*=7_~d**eITg>#4Y~r@uP{rO`jmNg(
zK~fzOHxR)dhLsEV8PSP0;<DL}#jDe&<eT_hC>Mg3<6HuXW4v5~7m0|ud9IHB=o0t;
z=puV#J8Rf3UxTzAV7ZktKufpUORXqMLHl?FVn`2X-!vA$4s5Fc=+O8Ih%P}AEN7tR
zRd<a^H)VMHqyzaE9BF$Tp56B2;BjzveP<kB*GFlnqV#(=6oSdMaBF%S<1NAHTigA<
z^M7J#%!BZZE(bYFb6lWMQaaI~bQmPb#knk*)#8TIOzr`e1JpViiMk(pJfx1%?_YDx
zHRTd9(B~anUUHnprSs-`pPFfhSCT<c%=`UqZe3}5=odiOaeihbBogVQyvM@Arf2DN
z2+<6J`W4PY4W$9I=|-(b8Wa%_HuK#Hw#eSJx}&d_N~Np%`_5YgtYc|uSKs+PelDk2
zWB#4lk8Crr@|;YVzP3l-a!HUjM|;(klW#DI`R!|a)V)>!w0opO(4Moro@yJ?-s*39
z&Mp!nDdjV|sJ%7o3NoVV+Pa?VnS4w7RPEZk_X=`)FSFM_SNBrp{#1|dQ@yQT*TT4t
zj>|8t^Y!K|bGjZZ#2S!wh<V9N<O>mk_X?;S*xi&-D&dyBc(D-i8u6mj_;KRfg<OMv
z!T%<2=khQ#C~T#&gF!JlpDd9`C^an1gtxZbL2)yk?oXE!^NF$>iU-5lifg9JtUS%;
z;!M3BO~M5_*|1ACR?o=383jpAcTj}Phl+86jJdH$5tu$)bEgu;V51Q%ChGD*NqCDO
z=x`t04Y)PLv!rH|<=#vty-}WZb@kF(Hr<;krRLIw!E`FKQ5d)zOK_TGgUz9|b(7J0
zeX!VQf<ZIwRkGn=BIFiDTf@k4pSQVB>-ZzA1nz+s*(=MbNu}P>+n>mUU*+vQCO%Xh
z#6;`mIV>5?UqbNvrZ+gv?{SWy6Sv2Dss?QvKC_elvhx&w@IfBSmJ0e4*97)@4ej0j
zR{kz!0{d2{dERlN`3J4CT%!r7#Xw#_>xWBcvAc3r8I7&o>qeTZD+jMT!ct2T0)#r!
zGzCrCl&?IAy%dEq5pm7N`V(8>zrQ*^`g(nH^Sp0GfBUychhQ#XSqBNyQVWq>=cYIR
z{bbh?;V6WG06fOPfjOM3%{jq~zc>2oXFvPd<xV2La`4FdT2pj10*>4F&BuvZxBwl<
zE0N9YYQr$LZ=Yx)zo^M9W;PAV&NFt6Nx@0T0ZXARim#$AWEsKnq=2-J4^jkvnlMVq
zWc>Sb71B1HOCj^T4sJPK11vxQ6Al7U;gu`%M1;BFQn6A^|JNYAGYkePJV<CDbV^8L
zE{R?|=*6+Onqj(DL%@hdo3J?D>FF%EYf^<nVdc9ydPl3-#0nhDEmW%bd#xThe56x;
z#ayeM21YPOQX<Fg;3yF%Gx^d~sZ6)RpZyw+Lj*Zus)&_gV94ehpQ$%$r9?EFEjHKJ
zj~qSJ>!Rdy3w0+rx4jJz8;)J1#7N+Fnn$Pj{&*r9j22M?m{urUPX*zNzy}G8tln$D
z8irY3E*-Q2e6LV!vQQ^fIe8RGG>k|Kyus!>$Ys293jKuHH-Gf1;nEU@@gC|@rB-X)
z_nFUp=8+8gwkzjmM}f~UQfjrMhbvX{EvA&!&o6Pk9J(#&aL`h-JfA_RSML?i8D4ec
z#!sy8eBc8gF!SFwrP2G}uleac_V0m1=(x{Asy~b@|J1-w1ztydB%Xdfjg695wr)Y5
zCMN@jLBiW=r6%9%avR6)ue@Z{!Exw}de|Qjs34^l)7;~+%i@E46|rP5jU;Kcr4hG0
zJr{r&Yz?JHM*i;6Hd^<Uh}kFVJ?H`@QQ0WH=nE2nmD+Sv2}~m3#5B@iq^GbI$D^*7
zz+D#SFt~06NE@k*vo9EY?lYS6&1^24Yh2VsyMR4ruJ^uX%GA*-SAhzd3Rmi826-S^
zid6CW?xK>p4KUE?W~*nWTk8bUV&-y9CzVg8O+J-_5l5gU7^d26LTNF$X!t500(niK
z8?jg^4+bjsr2eNr{pnqdlnv69Kpt?Rxa?O=rF?TSL6{&o-j%o$n#sojs+3o099N!;
ziM^=4?fK7tLlkq3j1$tq{He1;JXkDHMCZ9p)bzc9cY~1?=qrwHla43Ir9e~5H^ePI
zF}hKhC2dn(i_A8e{pxy>&5t{0Z{KV5qMXqPpx5apP8k~obWJUf%=I%}irqk6DiSX-
zlOZlZrXT7>BFt^zjafq5-dM?1Ou6u92x5g%dxBSP9@W-phq<djlS2dp<E1G?deQ=6
zghNvnDG<FJibtgjU~kw<MPjc74Ti+Q4douX-kHHSp1785q1HD}yBI6n#yUwXacSBK
zp@j^E8nOPjr_9jQjw);brAqRPqx;RN(fwy`<J4s9g{Z>m(&Qxwa(F7>+X*iRdnd$u
zP&$h#)&c44f<y{#6%vWpE|UR?P)bG0psI14DfB;?%<<OAlef5OlXbed@`!Fvp)bD;
zpM)A&{Z0}IGx1dsrkJZ~NDKR@JZt4VKbbtd_b4R_Of%Vf_KNU9pJw~#W{18&uX-fk
zGm*XD0kP7!auP?~K%5Winpan6=BAr45uiC>S`Sxiy>1KjArDB{MQ4hTZ8C(ssa!0=
zN4n`u;|=NFG_oZ*4C3+nZn>N&AfO?SG&T0qi4-sqGQi$BEry4j1*X{TZqDu7H!}bP
zkfy8k`?GVs1_={Ke?Gghu~<c!Q3;KH^FJ>vFLh=&H<zaS{bFW$Nx?!X#{a)Vw_X*v
z6zq;yK(l^1@X^4xVRPF+q}^feGp{pmhW+Rckj6AvVHQeI5FY}JeV!(Tw2kbynqq4P
z!!+jV4C`PkeF8^SN~MJ@L3>*yYili2)R=JXX>as_z3!&PAJcltsz)1(gaKeZai0=K
z8R`u?S^NpQ1#3eXQc>9{NlH03#tr&bhe<gpocx>At#9cz_09bAZ>wY<iv1Q#ONDjs
zlVbKDS}$AJ6GUE@iD&WA2sTiSE1R3Xnp~%yxt4!y>xZkN892wAoWI6uW*ZZ?#g(4t
zvE<m>RyD+j`nM*TcLR|?cnv)mb;$KPa~DG#lMV5eI0O3!Lz7Nnn9qiy<Qg>L1Fow-
zC$wB+?GNU#PKzIcWk>-6Uyv~4y7@XYRRJwZWr4a3Su~1mTrw6&PK;8`Mg1K8eej5s
z?9O57Os7LZs<TX55HjE}T!M!~A^%T_okD$(E9Fw+|2@L7pgkPqNbm^p0QxiG>C4dK
z^VY}6FTetDgt9?%i2-tqSIB`0;nLh($Jyn>{;?4a#rWLK0jeGHcD+z180`2P&Mvpd
ztwY?X+Z6Cgp^6|ne+P?TVa+};5A8oM5D1hGW6ZZ60H}z}8$-@C**TriO<WiaVrYSQ
zDz4~+7eZ(gX@{Hc_{c{-GAVTd8>oj(@WwDm`JtDy;1P9aAFGP<ls!si6zjon27`Qp
z6;AjMlp@C*CK){>(0(jjaDiW}H!0r}wRp)ZxG35c2cGmx-0g3mle`_ewq;lKtuoWc
z7rr&U*lSG#B5Hg0>)2ppb67xKGM4E5qWbhD`N=7hD#o-i5;}C~`tuJSm>nci^Yd3F
z!&9|zf+P&N0x5xuXpQ6L6kgS#_-wCRl7CpyggfoSVcZ*?FdlfNa;w#>fUXjr4I$r{
z&{lUd7U`iRKXKLecJMcD-`!0{D%F+Mqffo~lCGOP|3o63nQc}vun1?j2$X5+{PXd1
zYc>};Q#As^;0(M7>YSDnK5#G;Z0v7Wsvz{FDwX<D#R(le5O%7|7hQX3Ww}<3#19=N
z-2;9{k9`dQonM09$?NlOJQ3a!_yz2l9}9dT@F!qd{1ZYxoB<kMYyDDtsX;uoo86){
z_1hO??Y7$2Z@!`JgK@*z24!WRL7ky48h3+sfDbEY2GKP>{OZfLyqE&s_6lP?9|yE!
zOAcUKWn7Ps8GaLcEb(Aco}*luCCa~os#9UrrCs)?c2JEd>)9jRf{K<%sp$DPHK~##
z;uuxo<cZCDB}*}1pxBLOhlr3+uhl9Rlw*pp*Eyk5)TZVwZnx_eimeuN60ySr(}HZZ
z7-!IyEFgXXJ2?NEdPT_V0)Ovx3uV{6ClbMNjfk6Ose(Hh@p)75n8`A)B+-%D#s#^g
z?f(kTg47d<q;T!?Dn&1PdnSd{lD;Qdtu-}a%7tzR)&`fOisyD4X=Yim2aIky9lNL6
z%4Nxf!L0A}3gvJZIAgt%nO8dLM7dmN*5rFSJ46P-7%Jz8H!a7wa(U+G<Z;RLCT83J
z#)?3=3$UQ%kc&0s%h19*H_>X5Wic2`6^dmRHqa-dQT!Mv6K~)x0qFY_KxK(`dx2^1
zq>$UnVCy{zb+j9qni{}-1?zyS{}Ew^;Szw-KY_)DSGjls04mW=u&MC=#4rN3O>v!Q
zh_XstszSkVxp=pLr}Npv#Dmn-8r8ZSy!u3vkG>B#Tu9`|>j$oSyVdSIy_BS}E7fU2
zJ)C-tSeuOM%{KF79FA7PZnYth`cCIAwDzqQA?cvaq2SNLvhQgKHHfc!d>EeMLAZ-@
zplxv-S*oOql4b}=O5W$%?8%>Zz{ap|{jJ$BM;>_Kfy=8es$Tv}v$&-?bJv(>;_m&y
z!n;RnAN=44WADkE2S;u5=%`H+;zvin@(NzdxSVJ9$4ID~X+^gc^Ap(M`j^7Z-%h;y
z*8<-UjNs>O-e0o1`24{Z21%JDQAzO$v%#Y3A1TQ-&}I~|R+AxfLj1f8Dv~_m$tRjn
zrk;{&S!Z6$$!nqjJcvNLI-@#mViB$4zUyE?vMI-0R&&Ysa20%F?19nrEw)T*xQ~fI
zu=kn}jPY9r_#e6^vlBayZi|Uc4Lz4AC8hQgtL4dyvFF^%+C5)z!Po~%h8di93-eme
zJ#rYJUVSFnFr8cmbhGKHOqO1?-a>$O;zo>oo^F%EIGN0h-r<=`_MS)RT)|}>hoU=q
zVooZ6#)q800tnv$&fV4qn@XNxw5WL4Q5(rF(rtT*%)6SU5_lHfW-Xn|b|N>S`^k2i
zUb$}e|0On*>`JiWRLOdf?lNJzX>`57f>nbnq<Bg*U;!P1ptP~x5S3idIjLkRA4ips
zxES^j4_ZV*VbCj)L?e9Ar2Y#h68y`FWEkf}q(oi8RFZLB!GyXvEhZ|?&nf=b1iL^t
zK&%o!vg#AK&khE~QjQ#8(<Bq?fEl*iavF?77vJdrj!ooa_`Dt^qt4yt)#gpI?_(Bb
zTrsLFVaCSJ_CyuLemnd$IDEoumjjj@-U4JsVsv2!q1F&?-u&i#cuH0Y&QUo(aT_H-
z=hby&Rkvj|iu{_i7cw`tFv=(PEX$x0<MC!?qRDSt#|C+5xT4)2m-Fkk24P<f33xL`
z9^cGh7Z+-iA;#(){o4F<&_~MAgg15iBV2X*QrCus*BIaLWRc~*aoFJT{iYjVdTV`U
zXjd?`3;%j*|B!uNOdLvK?^i2q04~BxNC`hRky$e4f3>DykeL$L59og<R)o8Tq+(QL
zqFCS7XbDOR9~mL`M4}f<8?+qb1LgpsI)GFOS&W-h9o_`<PPSA5>C2m{AGl;=gAAkT
z>E1e?x~w~#8N=DW)tX<aH=?QS{U_!Iv#se^JQSI#HwvsQcu5264QNrKZ8aP?1puj{
zBrSNjM?5ffZL&igeLNX8jhmN9kFzd|nsTR-{+D}*Wl{6xQ7#)LnnC%7yM&7wJ(`FS
zuvs3=VRXVciWeW@1i%I`krzf+<Kzk4Oek9>^{&)$u&K>g%XNvU0Z0akQo;m}5v1}Q
z3G4h~H_P#pLs6sxgowxqW1hnLnnb3ECWuauWJ=IrtWwD){CzUjNL(c?3f5c}vtmrR
z07i}wd5pY>B;?GY5WqATbT#8yLnZHIGdP>q2u!I}o4sJLQmQw~71Eh|EScOHei4L%
z=I0TK6NxgjztNqlC)=R6Ct<gZTP)Vd#ukhu36QDR0kh==5v-%cQp!pg4XBAAR9vVB
z_U`Fgsf5i98bZEFvVDNQA(&g5GF3}c42MQyz9@jhaHY^hB=eGB$HhZ#K1X*6Czlo^
z-gOEF;!SBVdR-j$3zXJ?F-x5rLT;KM@_eq$S#H+zilyi#G8aCZu8Y3|J&`ws7#=1d
zfJi6G%f=g|0#j^RMUZSmdEUJ+`a#xFEwigZbL(Z7UADdQ;G>T|b@Y3io14#n-qr^{
zXzso4y3xJn-uCA{H~N<8V1LTs6ZxRIpOrI<@1I2q#CjuTYhH`<YCgy{bOkA#^1pfM
zrI*gk%;4kHe(=He<xKM0v!_;G@zJL~^=~e|_~OyuJpJjWF#4z6T{-p=f*ZJ&wRjd@
zpok8AAJP0b2VP+=G*99>tp0wTGEnpeTo0!r$+uz-^a$P%d01<s)r4A2uT{yh!v~i0
zEg5_08x!2>)hs0`k`fCtFjl`Y<5}tL0biZY<oBq5$YUv@SKpxFTXmhR&$<aeiqL0W
z5c;7RLbw;pBA!9tGHHqA&9Iu=UpCHHYrB5=9ECJ`R%yp??dKkz;F#!>)U}Uyp7!fn
z3N?`FO`MB=<pEtq19D$<&|Sp3ZRE2@j${|v8Wiq6qRqwZk(254S8{XDnENpB6J%sa
zk^31U#tRo2Ij-T#E*BX<xfsD7b2T0%{f-Q-{EEWb>Q~U*blpE1bL<#?`!Et*=V&gE
z`HUP$W@pEkf4mME8I(m6yOQ)!aU*^h@2VL_GHPC=LMGLW-PITuj9wX2K@<G~77N1R
zoZR~f6P&;AzFWwUx9|4*?z`{Qsry3r^14T_L}LByjZYo}a0X~$$kXq56!F2o>lh*I
z1S<Q#2sw}6jOZUD*ER`cQA)bu+T1H&Iaed`D3&5rb)gj5uk~DL?nuc&SOg?99IPN=
zq=<)yXU9K)^m<M-RQC6&3`sD8Ve|*%6R*{6&(5MWdu$Z=DaQUip!8j2PMH^&pJnBr
zk&g}w6v`?Fua;$bqcTqU#U8j*mP1CY%Qk*F`egC8>lRBTrMwmY$aoFYqHWJcAQTg!
z5@7h)GQz>tn8jG4C06w^2mG}UQ=VPIt;uh^&+g&c82-EoGs1|sP&xJvIA1kj7F12T
zisq9PcXr&`*~J!bYP4BL?eT)Yk0-U@43mX?NO>I*Q=oF(Vb}j=4p4Hu4b#Ap@K}bn
z4<>jWpP%;yYrB)0xPo<p0x3xR4)!vZeAfTriVq<LNAB<7&QTfCIZ$$qZ4JvXtT1a2
zivc|h=0OS>hfyVYioQ({b>b$7iiz9O2K%!YCN;Vu@}o(XKdld}POb}I3SVGAsd2#@
z6&;6lh>$CpD?#zKoH7qMh|q}c8ljVw08uRAB?@Ky+q9apV&GW=lOZV#=Qng&h{KT-
zp(NCRuLBoh=5jId2`(5>a&#weW?>(ZEwN~M3KmupCr?yB3$36HR@)zZ68A0T-Xit}
zA0*YFZE35iup9@~0d!DiFUh>FffTU*=`JGXQ%LAo230BYy38>V!U<9&!KP@r1mYjd
zIaM;1k|=|Og@eOnvH`u|ws6_Y<x?<8VFdIxaLS+$xCrC@ey5Gsp~M3@`w1&mD=Qom
zOCx+YuvuvMQ)#?ybGayw8je&e0{YVZMba6v-Z{N|5~UrDAWZ}{W%XoHelf=*y_~qV
za)|`@G@FwZ9-C7sr={ssGt$;*072xUbEy@&fzI7I0E&=jLw$mgaD}EK1e4}@3B|CP
zCQDlG)3vGK>q-#JSQWts=N^zR(7kdSp0Jy=M2A>cPn+jrOFV<GE~8DBp`AUm!D{XA
z(S}_kcGq~_ilgni{dk6czSNvI`ozDEE;S$3|G)lgAO>x@jNv<nXF!@YyARg-b=arB
zg0%ZLtgI}eI2>X<L^d6|13#1oHG9{Jkz<5i|I72Ne5;AAcUfA6G4GG(kDc}+G59Bn
zyg-q}TtgGkk8hQ?#<uS9*~~i@I1ep2M8^^bf7yd9v1FCXWsXV?QF?blz4`KujvyVJ
zbLAhtu2l2XA_K$^-@%(;Mhdz8`Rze#(Y%9nL@2mHVTvRg2Y6kqk5-z97i3GO8nZ9x
zB-2D{o&6n54B&wobaZ;~u1VQoaQ|WneHPjo2L1ejShSQabkTi#2{Y9M?7Nz3*Qb)q
za9iQI7a09Zm+Z8q`QH&a@hd*X7zB}*NoR6#C;fcgVzml?Z(j85XMcKXRcMaPH$+g_
za9$|6<Pav8Fw|b!(2)zV(=&nQq*-oFPH{r>WPe1?PQxcxD}#$0u9r=>N{t}PFOxXA
znoHc67Lh1!p|yz?aDw%#qlR-YnOHq~^yn-G({%PVdoIOzc5EJ@7crh?D-GnwIvp`d
zaBgA`VcOpP_P4+Nx+6ybR9{#;e)n_8bM|#JJ^K1R_gu^C>tENr5Kkyu`Eg=dmhgPP
zo;=9+5>xVlz=!-O{D~@L(+)H-Pn*8kW5N}YrI*2@K_(uRaQ*NGq@F#D?}>6o0w9BI
zs1-)T520~mAUzIaTx=lv8rAP|K!7g3rW4kY<9ZY-<?Z_^adOX_8L&EvwM6~mdgB@N
zODy$C(iKJ_drr4|9kSBQ0$c-JLOgl{ZLl!<JyK7WN>ZudS&Wt+2cc@E&QDA=8M@)>
zOs3r?dbZHvQ~AWrlhG5vapi;WsYRW^f{B!?gxF#Yt;Eu}siz6bgGj$jwHxvtN8eD8
zIX>JC=8w*3v3ketuyb>arR0Si?Z=M8q0EDDST2#RlN&RsS<0j65NQj?K+T7LE0?>o
z=n~@{@{hS`FPcqzUgH>GKVOQjBYx!P2ZiE`O=LHcz`d3_I`6<&A`X3f0XEh(`b~?i
zrnGuxq8qo<nw3_qPhVKBTigEoUCwf6rr*8bf@?Q606_rNw9`_EaXMX@o}cSqa`O7&
zmc6k~2hA})l^t@QW?zlwcz*uU<pyvN;22=%X96XXg_pNy2A6F#o0+V*zU+9sK8nB4
z1IQiwxkIb}L&d%K{=_p#-{#-Za7M&!EF|cVyrpKmUq_bTT3ai#24ab*w-LwZW@ku1
z(P@`U=EY^n@j!GY^EvPwXF>0O9%wybFPp0*YO`b#us8yP8SGiX^|d2M`b!9QrQXs~
zuZKTG9XYiL!>oJ?sep0rHZQ}!>sf(Uk)i9K$$&(hsBe}LH#Fg#tV+XwlCBJ#KpRe;
zK`JFdS&WZvZrE@(=_8d-OCi#Vv4os6smAqL!VcHh6BMX)O6p1vt&2Jey6?Y1fE4Fy
z`NByWa)d1!N%y6iv8*dQn3{~MkMDMa#0i>i3X$HBP1W|k?wEl$W^a8>jeWp1PZjBn
zf7G}p)y4U^QcSKz{VY;NS6yVgX=`(HJgQW0SGw1;ZDg-Cqe{iAPR>v{=!Fs=%hoEV
z)|t&zkYJ&B{P1qOy$tU9HCOP9J<FT@83GYy%S33X&J9i;gX4HN{9Z9Dsx8V+-qA4H
z8#br9%~Z^ZCEC&sg+j|Hg{oz;-r_IZBANG>C<69|v@{YWNa6!aX0fZ-Kx>0M#dL>B
zj0#TA$;MF96A$9{OtuL66Er5o8o>yI?ZGn2l2XW*aB$%i{-O6S><bYtm|9S-;Mtz%
z()R)G%TWh29UM`E=t|TQB%L6w1TQ?lqE#YUrhC)~!zQb#>9Xzx&5P0Ia`9%yYJP&T
z4rWsD{94(B+{!4OY8<fF<{~UKx&`(Ua~*v|8}<+JM<DNYT+=E|*tmI=QBF0B9vH}4
z);fT{d8JbGJPtNn+<D2~3<HTSOP(pXr6Jt$FfRNXL@X2pxQ>QpAvMqRTnvMN=E5+_
zAPDgVRZ;gKI}`1Olkm%h%1C+STV6WCpOPVo+E9#t*Q>U~59VHs4-eOe*9~5H5i$WR
zw(!cuY|HJU6pdYXJsrR~#ylegx{)lhz8}$}&W^dCHElC+Md0TG-(l|FV93wGs0Y4Q
zBcD~v&Xn=QvXgo3%=|)6D?&M*I(?xtvE8wxDQiVFMLCwr6GuW7)cf&9H)DKbJihUE
zfe8e6fAb85(2=qjK^0rK#t%(T?qu)u?BS0v(IP_=^&U@my9nwxQW%J=<9i+y$vKA?
zhd<H92jY?8Pd7_ox&x6aG8&9;T&fmp{9x})IXvU<p3Gt@g$<BoY2Q=|&MY<!LJIvk
zZxzrslFYw81VKbX&L)#ZS|WBq#O}-~8=Eg>5*3&_(Wz|8yvy_WWMqKRseHfie`faa
z9HxF3=^!I_2VNnZ;WU2GG<v0YI9RP{9OA-;c;lD`Lr^#hSQAK~U|tqXr;$1|T7pOy
zOnh9|7~OaufTRKmkUlmVO=lv^CeRR5Y&dTOES9NlN{wn<S;1?FjB?yCIRd&heqDKK
zye<^ly`WT_ot>&}uOogD>cwqs9dGx5p5{u>kW#tbS{Q_bZgh>*+<?30(4eY@Jo%MT
zOrl)Wgd@KePw+1Yw*)_+&;-danSk2CMx?khUUZa#?D&)pO*ao1eC-o?bXA<7Y}+uY
zIFnk$Rkn2Zc`tJG2<$SB`uw^^vTd0Z7Jw&&ZbP~QO@`elVYv6}P<09sR`Zrm;V%e$
zLM_KIv1%yMK9@MXTR!~Z5337FDvC+VMK&uHsx+!&)$y9~gk?tlid@*Je0WqhS%WUv
zz5?;5iaH%a6ON^5<VdK_5iT0!$#pSxGm0>PDMfyu0Yyxs9zKKC2jet)9bdK!$b31F
z{?(KE62{Rp+dsee^;=-rKh1<k|MrtN+;GD;zHx(ShH&#9CHLLC$tW`mB51&LIvq{v
ztexjRpi<V(e<^jvk-}-in{FycrgVg2jnVs=SB@Ss7mOaNVWK06RLxufZPiwF^!uky
zojUTIlOKQY?(XiX)7QSxyjulE4^0Y~3v|*`?z!i_`$pd*{P}sWd?gZP;IU7c$IK_e
zIqF-i(jBN+$@&_|-@u=Lc6A<D3iN+-_9pO>opqgeopY+rseP|HwePp;u6@_Kb#M2*
zy(Z~&x)VAbvXg}^1WXuB*ksW#Bb)LnFp4Oks0^aGL{V_?;}8*WW6(k8bs5EozUqwV
zJHGS!iUPOa@9#NP-JK9(n5uiLPMzi1pZ~s?Fbk$Nmy(2u^)zQhD4V8+aA)5=eEQnm
z!@Ztd+Yj$vd-|{}#2;H)T9kBe_IfNh9o&9;SGVYR_w-AY3}9*LX35w1xWaG3Mazsl
zdCm+Au9shL7y%`+dNEXo0XG0bL!6q<03gMi25$h)6!=)RdwAV<@R)V{N?(qde@9w)
z4>>|LOHPU=5!lH|k*!Mi^Au+v+uq)OVEgf>@=JMhrQLEIEM%$(A@ndZo(iUWj?-#i
zd9Et?KkMc(aE)>iUCwG9Bpnj#(#I^Cd_*V-NSSuu?YG~4$L&}mT}nOuVg0<_=*gq@
zzavIn8k^;pVF&a6&__ili=vdZo(W|E?_=_yA(-qbPJ%f(Rxofz3g8Uc1Zi(9C51hp
z+AxCCAc9^vA?(C0tP=(w$#=jF2)2u+!@oR`E6y3pThzC?B)dz<Qv<g(Hzmm%G%-|Y
zrVm}I?&YTnkodI!6=%WW@1E+Q?%U=NqRj2}V85+w-Dy~#k5>;T<8H-o16AscQ4#LO
ziFC%Kpy<xw%9ru0#jF+1=Ax14!B=yY3j5w!u&dqGAnopI4bd7wln8i4e53Cb<1Q8+
z#~|XhTRq45O~?6oK0&%r`{2Jj&YMl7RH}FllD9Yai!b#PRga9yc#nry?05zmu-eS{
zFlP8VgnQXY6pT?T79If~l>LQY0xaELAf@l>Au{@Yh%CZ*LcxIK>qNk`zC^`HI;fat
z082dnEAZ6!vuAg)ZF+7<`FVaS6tXV45ct7f4fd^B<eTGQ6sIE@j3t9?Zt1f@xG;j|
z$LG#OcsE?Hj%UIdSuC6A44zIh!AvjjAV7wM&9aolUZ1L;QBl_MTd%rmeuM+q{8d-o
zdgbx$*PdBivaseqbM5U{9^XfxVIRk440s4z{;hL!a|G$H<_no39;)fi>F&pL3R@+@
zJcDoWTb%<PvbQ2gND6b^&wuUnu3Nj}dF$2^;_i8D+LQ?F@^Q{D4~MI(!y&8s($TwX
zh+9vCC#nxDld-ZuOs?z)|LZc!g=jW2%ofYpHbjAI-}}LrTzRLH4(>EKAxUnxf5cv3
z9WTo3P$D1}44_j0n>5D!`-U@RX9Sa~Gm=&4PGWil(>fL6F`>Z*k&&piG?>F!yEHdg
zI<vB(fe06^PIof7>T0>HTju!cejlRWx&gR@J~Dy=wE?%CYc?0^772)kq?ni^;G?s-
zc`dAiyRvoVvsamN@U^wApUA>3H=8Kf5u5|d7Me1N&|1CtMf*5*5UZhM(8~7^nc~|+
z-@{s^uc8j=g+LHTf-Gf09Q98f(J|k~FsE-_3c#b&rP5dy`$I!%KjrVprTekNDeHb!
z^Y!nyv&89`Yt&;T%AI@b%wqkHv+YjIablhJ**k)BsnWq0oF8$X^amO7zHs<`Px!Rn
zUMA9NxlA7Qe=Mcah5RrQ86p}Fejfaj<M{j`D6pP*@M>Mq$^9qYq4A7f{1<%GzXMui
z5IPaM4?FI6AxZmW=nue%Mri{75$Jqe$!*Or!V56GtPp;gHDhGJy2RLHhoOrGWTR0D
z^i^gHYa5$Q9zbogFnI{ssd<uT_%=5%VW3(SLh+`~(l98}R5A*pSa1t#*4)JDssB1;
z`)s-=3TMrAK@SyTmNZXTWkg`etXhL!x7di|d*?`k<<HF>-QT>$i?jPzHVA4KwPPjB
zQwum^&pjNjkmxuXZ2}3<NT5|mrb=4rd|?Zqd^C;trXvs29C;XHhsZ8vm)b{;>>jS8
z8I9*t^+qz$f6;{-N2BWFq+?rK8)wp;vfuP^s``cfW5-uAnZ>p3?a>HfaC&oT8Jjc!
z6UI@6$@Z~i>`jJSN5=^H>v@vNIvK(c!?kuY=}MzMTxhkkxy4!?>x5dX*?lw?&F5;h
zK^YesJZ8wMK#tu^q1Zb3e{8ED9E2e<C?c6Em#arXbn?l!9ffX$Zp8=!byC)i5J7DL
zj{(Oix>0&5#f$Y^>{)0HO(B~J{PJ7DV@?=OHlv7b*p9QNEaE@>5%}cCLy)Qz#{`&v
zrs|6K&>U}%rp7=-8ri_&4Q_MFl4<5)Fg4>W3$uxRj<iUoZ|UGR3rfC-Za5yrXVO#L
z@x<xzcsH3^w#tR|t*zUjq?~aiwm;rRk2fi!cB$9baUUI{Dk_x{%AEk~RxEXgEfP-r
zc)warh3$B`RZgR^K^B3>q0Pnv3WKoA6c6igZ`6(}=oG|3Ml67xxPN(xJ)26kn$m~4
z`MJU3Y9Sv*k28yJT}6_3jCe>Uzj{DuA(@5@2Q~*#3jc%64dcWxxlAwzL&jk64mysZ
zmPS-STPJkEI?}kliNVf&$Q<tL_u!RJ@9cDd$o<~V&dIBGc6wcfK-t;3_KqF=%y$pp
za^v0+$KKt&_29ny?t9g%^c>QPV||`|jnyT^>owQyOo%FEb$RHTQxwp%R-7Y8F5G;0
z*Rdwscip<XYfMSRhs^Tio`DVw6`ro?QvrL>51B%!cL6gZM^g#dP$`?f4q?wtKYfWW
z2pl6T2m}~{J=)>s%{4>5eMz~LpQ^i$p^)3XWq&k6Jbjl}ES8%KJq&s<NQFa7r&gBl
zeXO;*wsKDt!C9l%==92E41kJAf#c<GcaI#o84)lTZs+uDIFw>67>-|T4;Tyr`5=>z
zEYwqJYdCE6%2rjvI*gZ#OK-7rx&0+S5g!3_x`^M%L@ic365FlSZ#ldt_<MPMG(tC_
z&A><WY2*9n46MZ*O(q-Ck0CXIS<(?h0?lj2RG`|}!4%C10|vp4gy@+D!wR(bjG#<X
zH3fk-pIxRx5Uh^AbY^FwusL1z_vFq_A3n;A!!5gq?Zk<&@7Lq~{_>(9_b~P>7Yflk
zA90FBf)K+y;&*{?^K`?Bl-iAa!XNDoqVA2ILBH9ESz0*mn;F&~OY_v!>9^gU>|7Vj
z;laMMd-&#lyV-&iPPH}&uaZ6lz)8fLJC;Yd5a0fIp_Poe=pACKaQzGM#N44GK$~^F
zduyY*_k=M$OH9?5f1_jSGoHbiYFxGSWOy?(jF9>jNV4uv#)?}u;a1s01Ujr>B%f{+
z*-)o7im`8R9CnW%-#>AqBGe3rySJfhbSCcb*vTb-{$%4qacS`?jaF3A!=HAvI*pxC
zGC$hc84SpM>tk{Doq4yiN%-rR-5bAUbI@-ivSn~O^_3^DIe9If_*ZuAQvjSxEINcx
zz^i}hnL?bqe}NZ0dZtk3ZDQHa%c@6hzY^Y@#ozK7?G`k5?3bx>&oCeX<N<V!13*$X
z`$RTn02{L{WJGi2WXfvf>&AXRy|D>wc~+S0!i`2HM<T-kVz_i8l{&k;Rm5SZe(2EQ
zPdxtk<LL9U^Tkprd%AhHzPvoXWp@W@w;7E~xeJ3$7wn9KqxNXDzkjly$*+ru@Z8)o
zHh>lmG1YF7xazirGRu1B3tsR7Tqm-NIXB|0cNfXG<LvC-Qm=|nzXYEHiut+3^ms<2
zCMd1Z5r8PL<_(Mq5RT`YJs^?lHg}W_e3|VPo*QlT<!Ps#BjfciT?;l6G-1nM1vKgP
zfhJ}j2!QVF+@WRe9ND|^YRnABdAxR<N;S`5;)fea6S=kj5N`av`Us1vjR@}84IgdF
zbW;E}I>K%N2tj^h1t8?YkpRQpwLO9JeGB^ULx+fH{or}z_06IE;9Hg<0HUj>7L+#T
zN#k@2w*TJ@E&TM@?5Xwz5<D5&=^fx~fd)Yf2-EQvW`cq(zAOiY`bYW!jl4NQ7cB!k
z-{*=#T}+<we)>_NZoE&@P6Wr@QSSJ$Bd6ZAevS1l-(q?5^Ye3~!FXlm8`y8ZgRJ%%
z#t6Cy;xU=U{`hzW%a+aNvHju&^u}ZmTs@4tL^hWhvNa1!g(vOHRU`=C@W?^e$>5v+
zKy=2kp$!I&88MBaa}Wx^+DVX!h2=9Ip_d}w_x;RzgVbnHK_}b2Ui_%DRbbxHWGiFY
zhFjQJFMb1q^udKv2{D~tZnx_j^Yb|6m39y1^Q-e@#|T@?>+4U3Rjb6S{f5WH-xeR_
zOD3$XSX|#?vnl@2MsN<YQHTnf88%<bhcO#%5nXLK;vC0oV^A{3a{4eJuT&hH^XQax
zzs47rR2H(4x2hg$2<n3_M_#655vZcSbsUH}pYhNei9GO1@)PK0;@zP1h#ua=f*SQf
zG%l?MA`cr0zOJJNHIn)mM&jIu5hzH0Yb6cWBlQYFUKNX#XZV+W87ZlrF&-I1rh5<U
z(Gcw1Q^~`}Jty|ulviFx+t|4aTQu3RCMVaz&=A=#ojP_;!b;xS%C*oq@DMGzeHYD3
z^c*UIQUjnJEk2L)*vVb}9JG4~`_Dg5s)Q%fbgQ8s3H{pt#>~5(*0|)K<{QoV6MoOm
z&%WWABrFpM-k2aM1<482{Hx#a@KOxGhzMQ_a!CY&AcPd@I=wY11J5WFEbxpc{hO5k
zm83w@p~54<$jTBHrYeSOfzB{oaGdF-P0um+;n6U?Z}yD2n~>SMFTZ)@2Bc1ywGdX$
z;={8_I`{_W{y!7s`>uFm^J;90SME+Eo_#8jIKLZ@-;rbnqjlds6HhGNkVsr}CXu-D
z+C<{;@kHV#ADvHh*+&CPWlyM{%`H90jPF?`&g~tc=l^favIbXsPX=zmr=oJ0ArFj%
z^i!8gB03F%ea5r!LX>f-HzloY`11H$`2H1K;;h8vfkz+U^$iT(cUY;9sH|-(`B7Xv
zQri!|Flo=f`?mMI=WE8{BoaCD4L&Uo_!L+(IE4|P;b=!P6)XX8Ski$n6Ni!pN&fWA
z@!}q2FpT>i-!c%|BfyupjfrKnwEzI&D}e}n{~H*GJ!g4&YioJgnSF0t(ew)nnM%9u
z9IED}ZgIU7CO52&&~fA+$3r)O2_L{q_<KVkMJ@|i1dte=QLvS5rXMX5epk>6{h?=O
zngt=7vOs?^wI{;ZnSjU=HliRTf47nqj-YCdUl#P2QEN;L$uZvcw>PEo)6K>bT$U@w
zB}yTT1@X0(R5`xYcTafDC5p>cB1rVY?RlisCornOiM3mWq3&4aIdbcH$T?agpC}TM
z{l?(=2QTwcH>dnEI_GrxH?2*~sO*(e;ot@6;mW1{VhPf1iS#*c*Mk-=I8uTk3NB;9
zW<SWQSVvD$MJC=0<5N2#|55GyPvo+(col2kN;slmoOT21BYiR*f7$jYUMqcvajctu
z6MT&+j1!b666JBLSQ>N)*n=r;9;>uOv($`{#;T#YNM8IC>*tNW>&4hbJ!+z}NO~g4
z34gNc1KROF^du!TXwhRfjA;Y|<`*HzG=$TXkXVI7o{3TBZarsQKs22E#vp%U95?lg
zS(R21s*dMDvs%7p))i{dRp7h;2TSJ_1V_{C=vYlMIx8v_0nK^|8f>*NB#`}I?wdV?
zQn7d{O&TB^4(+H<d`J=_co`J_9q=uwPfM{{k$gwCPZ(rEqd4RYJycmaabmy5)bR|K
z;fW-HPx==_5Rho^aRP>oR$NMqufLC|?|r0zPNbL40k{F=IYeM6xt*wcfzSJAZE4!F
zZ{YuXlV!a3!-y{vVCQHwN7!2l1n@V8njDlca<<Y~(XTIcoA@`m;PZI3v3ItMyQYV&
z5x1t(b1N8$Cr4{Xw!WL+C54tCyzHDR=J=-~UZPIyJU3j=DN8NJ_9@5<%+wVD<zBv=
zSInBjKD0yWi+^hUob{tbBYzdUFqm*A9Z!mxwVAe~u?b_6r_;(cCJcR}5k`$QwLx|P
zrhFW*Hp~!Dl>p90p)ft3246*9X{e|vdxXN{=VYv(wUDPEstS&wVtjvVe-Z04tQpGf
zPOauz`-fH$*Dkf&G88#WI6CD_$l_sW9$!Vsm@c(D<dY>8Dt_URO{uj;J(<7&3sqJk
zS#PYxVuyEj7nd@u1TQ)#hn)eDvP56uC%!NmIj?)DiYI%#dF)sd$DC;Op>}0bt!By=
z$++rPnW(H2zZ1|gkK0$XcZNpO0%)7NG=3wESs;Zpc!`BAV_w5DPyH<*Q^h@+#Voym
z_6s{s11C~NM`CwS2Luoda3!o~p6ZXOcr-kD;>HUnP9os6Po2E~1%vMVH=kPYD+epi
z(W7_Ybp+e7t1zR&iw63;gpFDW7-Rpsgq44uPPDUGvJ~TxP_NgCb{wwM|8%mvq(EM^
z+S2mm%+@${bUzY1`%t|<SV->>F1FLT`fLZ|8#Vz#z~bUkArGPXySa8}o?QtWiOF7B
z%BeG+>pEqYmytEi^y$}Pi~Jp6Kdi61QleY#6uKxxmXO;9)(o;5&|68T>A8XFc<;)>
zeM)AF@zFvoO?n2#2NGh=5NH=CGe&Ku-BMKgF@uiv5=#~Qh*dpi_Zsv-eUCLR&VhBP
z93qlqQVtl%C`i1-kjSrXr&dYED%lLq3Au7Hnep)v^pJ(Z+Q_<>UIJ-KR=&k9tG+Dd
z#ghZe8eTRAIQV(iUnY0RI)vL7uz>-}JeZLQ9aJ>d<UYY`BXcE!le{S&OLeFo>s66%
zKo}e&>ZF>gvAc<%ykcD~OO2^aIeYPA_NMh~WD&TBOaYQsOr3v?^ZaZX@jcyNtVCex
zbnj|`nZ3zlGgVVrfRqt*`4&iZJwavt*70ceq{2}eRA83sHf5?TlR})S($EiiPzzLi
znR%-S3Z(^nYa&#zDH^KA?T>z7z0?9hwOfY{$r7Mi9aYwx@Xq1qY-}b{XV2b)p}_u;
z?=E@6!T#Rq)2G7rd#u#L%1Zml$;k_W*aKonxVG1dX7C8blo(&wOxS)ywK4SYlp}{D
zIZ<J3?05>lQX_k1;!>otS@hz486w58GXwVHWrPh^JgIza;V5ZQy;#D^yKc{V_u(_C
zTD@mlM=Q~@ddF9te5!QqwRhcpco*YIJ6asS^Zd0Z=Lo&ix1EEBb7|k-eDRA2i0-&+
zdzF0H&2MK}S5dNsgk55whaCd)&!yxYNPZy9V5PYt4`>Sz4%R8A<8C68*%{$<#5#$1
z#agwQ#vY+qXePq8&;XG)FqHJjkB|qGJdxhs4c!uYG2SZANQx!=t_2gwvI1!qR#L7v
z(@F*N7~Eo3&L}rD`*!w>z#O#Ih*^EYC%8eVfB3bZkYx&DDv0=`Pk@_eh=;HFgoLAp
zuPb3ZwY@EUPN%cIed@~N{f+e&x!s$s^^N^2kMBUggGB9)+4slX^mY5$(q!OQ8xchI
z$#A{q54OwMqt~U~>o4F+aKq(0>S5G=sYs(%9!yHv{p(WR4L4cVh3gHm{M2&F%GWk+
z+RV><ZR3K59-P0jfokZFRsEpCw0px%(<U9wU21eHP1qoF$E?v8uBXvQv*qo6xmG8+
zlo~CVT*bG5V6aH_O4zh4u~H26Eh*b{f0~ph6Qp$zULIdY=Els7)<^da27`Y8*^scC
zo5h8N^ZkB*Fi`f%#nR?x$!f@VV(D8y{pn91KK$w5TUl8(zwcP#^?N!M3MH=K4R?tk
zcYWx)(W<|X=<kn)K89{R<Od7v>nzPnb~T1@4}@P+;M14fy-b^l<p=?Z<Zyd3);TRq
z0Uj}vLdRt~Bmx91!Y|MQOHrMh@snD#dcrWCKBE%kS;2x(e94HPrVf7l{O(809inqK
zMe=aR`qvkRpV-=h0Bmn<ow$6|I5d}E-rd<<&JR{o+2oP(k!bkBg>aOERHmJI!ii+2
z{ZUjG4X+yQ9X#5LR=q}kG1}7!>$^DS%)uXW!}9Lx;PZo3%G^N+^pwle`m-ms!?dVW
z#V<?c#JVbU^TM<x$qW%nuWVW}d#PzHDx5pCyR1s{?=}74V0rh|oHtE0&CQ#bkbgmp
zs2?_dBj-)b|6dM$)?!T*%sk1;!C>?$w#SqN?m&PC?hi-<J{rwFH~auB1rkyCI<$50
z2eVJ^UT3xipRh#aQwq`<)Dptf+R^vyoN5*d3+xuL?BEy6Nw`dfs1drgWLBVYco2mm
zHZr(NY^Eq;!*;^O%Ge?h*9J9fohgK<iABl+721~G1OpNt)3<)Bntw?^-WI?G#?T%U
zf&x|ui6kKr(n(zCQhp>>i#eG@BNZQ5&T+>Y#8Zt##^G$lCvS}+8D=`@K^o1z<IFEE
z*3t!1kuLWA3U1t3w^jW9Vxti+q-%?d^PsI}11kX{zsSElH~Yq(I!<^*^GUE!VwDD;
z4`<V)itFGdLR&KM;GJ~JOJ^IULJJ}XGN)B2Hmx6quFM@ie4z@}WWsC6X*B0oFC0FM
zg`eX#;@FktbI|0)GWjSL3d3G+SXjWo0#YlHCgVgk5^uP~FfI_c2GkfP3wo`LFRvV|
zvZUgf9+hjhjI!-kt4mm>d#~()dJ?_V2YP~!*Q^%_NUc+LB2U1%QX8U`p(ZA^9mapq
zA`LGnwiZKL2~z~T`n>h4)(3!5v;5oQkVG1>T$wK3eek4yPtVE&x(I_S>?x{A{{knx
zA!$p19G10tUU*Z$NEi>Vo3Y`jSjFa><)S)xn=QauLr1xkW9)b+-3Us!Z+paDEFkN9
zR$wACb}^AC6zm~3O|-%8<Dr4Ajn#?}?}}bRRq(5F>{yYkAdo>4&|2-{v164U#D#*_
zMIvo{M+meZ?!$cf*iH>$Qxl1jL*ejS7|gJv$}219!q=~?h{Co=B5~c(@wn4n8jT2p
zbbNDzJi~e7a5jkvvbVI<Y!DwcpWh&ry$u&OIkbB1>78A9+Taeld;0X+p-C*ZHrT)v
zG?i{NmzMSl1$>&j4Ib+EH#U!BOfVcRbvxtnQM|=gtVVq>Xw*?3HU<M?m9FGS4J}%z
z41W6$Sn+<^`fFTZ`u0)b6yqskeDx<dUCHTIzWp2r`&C<WdlbxxXjRQ78&i9lIb+_q
z9#FV+oB7hr8t*H^x-c8k$duR<Mh=Hg0MMW^w*V>uovbxLt!NeqN2wA`3xfG3Ex<&p
zgtMwmPwTB#1ZYmR@#Y+@R+|yP=w99d$QX{KNK+l$BPNmdt3{@j22H_S>KS4JaGiTh
z`GTF8l}O4R<R>vbGpj6!u*I6BB)2~dzI8l%%M(m@25k)7QD!Zt-;|F%QZQOl>nR?X
z(YtWY3yx-e(<tk$Iw()nhSoQuNgcdF0_I7@Hx1r&M8l-+aaqG|riye(yv@v-@ilV`
z{{b4H4%JfZZ>{4*5ilsGI-C^EuuKq2K{3E=3eKxw=GQPo?&xnsbL+)6L6Cx78fzB5
z!j)fzi1te#Qoo>?;Tm>RFvIX429C4Jb%r!qy4O5Gq&Pxvif!fuPAZWYR#Or%aT!UV
z{@@@~5()#`&bhe0sABFgF@F;Hv%rpXdyLca+=yaa#p4khJ|he;myZ{&D-tBpXgFKE
z{q<7I;wp`f1nEyHL@qY`cG>b+|4@wTmE((PEdWCPEP@nj7!d(rUtB!q)B$YmsDc;{
z+)h=fqo&=|-CUz^msArpRX-f~=9Ws*1!?Q2b=aqMV)i0FKyY{~hJ;=|aw6g9vRkQM
zH(yG`_dmCs&eX8jZdlEwB3V~3o+TvqaU?pPPXsGeC=k1D22Ku#AP9y;)F!9_iEUD9
zJrPY&q!Hng{G;jtt#qOUT%cdDVc0DukqwcNG+U$%GD(W0z`uKTI9G*V11H53C}>Ks
z@`O$fCmDqVCXR~JfMPDM2nftHE=uHH)%3|J2Gc}>Js@LnaRtIont6sFO(|*xNm!Xc
z5LzTbh-W$0aSlZuiLyLBs3idgrI4^ApYnvHhq(A?X$3E}Uo(#kGMHc6MWg{9qjTKL
z6*CKPnZYC6q>wTCPzWVLUj&bl^p^DV7%XDf=dqyhw?ZqJAnmG`LPo^!Xl+mr9+LF3
zz{p@B7)xu+AX(D*LrJAc1SmpJtk|h?GLkIeB7~D6Z{p+YD%`eLYR4l(EHyaE71j%`
zln64}1&T4nkTx|2LH>{AIKNPD#86XI+nZbC4?DH`Jct3^a<}7GB354!$O*QLZRN#J
zlRy0<p&ug7=yR-}v%lip<h<YchcNKJuz`J-4C$5r6^J*b8O<)&P8TdINm0OCpI1P;
zo*~r$6@9&NwHx*Hc?tebgTNWcBR^MzQ4CfA+6&hJ3j=64g^30-A}_@aX7*16dn(`^
z1fU6y`3wN0fpXKPnugO6z^wqb#IBb_ua|%?fHy!BrcnZkG#45e;`&n2Kw81rOK;_=
z0+8yBRk;DGI+kDa<-Q6N8Y~#V*lC3m!PA*f&$Pf9NDq3&fC=@P(2V9g_+`#qZnco2
zc?$&3+B(A#%-b}K^|V!ZTMG*dk1@&wrMcn_S^kvopm*w_`8T*_HUf~}3@USF$aG2-
z7L~kE^K_|^SB#%11HD-Z(=xal{6A$mv`T{R=#HQS^YrX2B)_iD>etGgz6oqUmEBF}
zNK#AD6SHuBT35mEEBmS03Ep^{V+1+t2|5`No0(n^E8lpQY9b*(xUo+|%^Twdkp?Cp
z5|&HswPD{77e?E&rR{@2Q9Kj=tPIEmi)1r`tRoho#!1otW`68zhJMnHEBkBYw&cpJ
zDvns%@-iowqUH{kkNHB=Czf67{_>#cC@4fnW@BC+Sh{OSDMJ!XNjcND?TN`HQrrZC
zo5xtk(=BG^fEi(S$RD9dp5{%5FV8X~tyMsq0@1=3@l{rEIRSD&X)>A5&$ruwE4g6#
z6b!q`%2KT@YYx<@gvyfhGg4h-j)K5K1TN7m6fVfgX16LJhv}0%Aiv0DzO|IVokJks
zpjqIxumO%tP~HV3(Jnrcg1}2@^DNI}hM&q+h>8ox;$;Dawo@z#!3%S#%K2_3FElO!
z04XRBlcX({D%M8C`^PaGCD)6KBx-RSB6zBY1l=N}Ii!fZw)G;MOD0ALLf9vBLLTgx
zjYs0SSmIH&j{tnxEWMU(2s8xUOk}c!DQb5-A}S`0=N$xx8_l53u`Pv6RVZ|cGEJLk
zt;SKvIH*NGOsFP<4>6Rd-E>R7t(pzIv%I_^zcd~c6L-mC0nP*kn$BVsWAG<Rvuvlw
zWWYn`$cB<;q>u;`cL*#6z9C>6ru}KUPAOE4oqAS?nA$b;#z7uTiKv#JYMPb+Z{4kz
za4F#t)d+fmk{da;UYyVn>PV%U(sYwXS|@c;1?gTC3B2eCF{1RXxieLnF6)Q`F}db$
z>tni^Kir&Sgc;D?(mX-{A``~B7}W%UXNgCGrxv)~JQRH3Y5#yJETNj+WU4?M6F7Xa
z{uE9TpmHJkcp{_0#YiQAD|OJ8k8pxfaijTm8=aHFTNMa|i3<`DxMpjd*vqwzei@=o
zoz@f67vg%Hq7k_#mrtY=c#%>y^%*?OG$y)T)J_#%lN2T8AsKh#0S#E?0{RwZu|~5*
zp(}d`Gt(r5fC`C3`t3r_&zj*e^cEGeh^u*DQ)+VAvvsgDrLtz-FE&dpJf*p5T|qJF
zMLI#A9M3D|vq{>)v=Dw&6gFL;S!yQ~8xLzsGHA3Rnm{to30`O)S&tRsOO}XnNh(bC
z1<bM5QMrl}JPADrVxnMWdV+bROi?~ON_Vv8RU?T=qsgpdVUf0wq%g!eB}A_7QOGi=
zE7C0TFweTE+ID7@QzGS2m9{Iyl$xs_EXhm95NVt=TUlQ5+@fbVgB%4n8G^5AjY=P-
z?r2;DDiYF`sNb}k1Z^%bY}$URqI5MT?m|t|30A+lhSr3dRX(O?kw`X=-q|c1D9nWM
zR;OknegvCA1t>_H)*eG`T8>?VdZgZn^&sx60>}{=RGbM7gh<5dNKX!RA5)@6s>M%!
zsRZqr%iHww9kYit3Y0KZoh$kv5M8p>AoWsQpi2`EuM&-GVF{9I<)gtXD2IdCGBa4f
z#TW;uS#v~f<q6ko5P1VdJ>GpvsAlGk5#}6W=FIve)~nfD=pXZF{M1sj2b0GC_*2M7
ze<}0|{Pu|aBaRgGHw2VH=m}F9S){>XU|oZ;Kz<-}VYkur4Cl=qGmHlYG*=8!$gw%k
zv(O$>N_R>(1R_db#X{_Q_x6_-SC$tCb0-gZ0R2X@)r#QNIqs1_g@9dtDW_}{wv#Kg
z+r@mDK=q5OqopYEu}TFv3%i#`8%LtfjdmO4#++(Q+xQ8!@#9=Lw|Da7&R(bNIJN5H
z=Mfw-ddT!%ulRZyQUGr-c?xMj4Q_6wHP^@j&ytGOsu_$&P~1h61lsHPMYbdnpg^ct
z3e2^;1hIEhOG{&xX0%1%BYE@o4?$HOE)=&`&h#7MaBuGD@_Ktr+Ljwx24w1ytM3(l
z?}xEkdk!;z@<Spz*C?hWq|<*HeSyvf|Jz2z6<k5chsBaPFTzJeho}|NB%2714M8vI
z3#8YuAG+q6YmQtKiSY5t<ejzJoyk{T@mu;D&OE*`91MmV!Dqi-tJV9#r}e&9zVay^
zn-znZ@{wy^aOp`sf9X*Q2!&GcOTP}k^ikv2d<*#+<k$R@p<f|a!#{$NmMrm25~2tH
zqN4%<`qWHerU`s%MNMHGq^(!zvRR1Mz*-NTyjPN6L|`4F*s8h^SnCq)n;~bE1Cgl4
zcrub=$_)0U(^AT4lrNoFH#a@$)adD2?>|i&0s&FbLT!dzppuf}@@x7+(*p&;Q?&f+
zDlAWqw`B6)?re=uUU~fPPrQ5X6>HCa_B+npRDW&pZ9nz)XCbb~-po$ot+)lMPDK&R
zT3+fstPWC|=rHy&%sY|@daXn$^S#Hm$c3$(i<?`=t~|cy*;_B4fA6Q?kvKFOwSN7F
zfCty!xwq%UmLER;?3>=cMvRZWJFmun&#qL$sZu6qee~<8jSiZ$dptgIay*s~T)RCU
zpFDxwm*X>7<XkxPsZXx<S1Ogtr#@9VU)i~u)>Yg{r5MZSDW9E5elgOYy$P{OlEZW|
z62BcHS*arYVl4NwYWwE#W1E{e=Uc7z=H{{Eo9fdCh}-zdPi<V-Z2th_-`0f_^9$C&
z&1!e2x$~({9xH6_o~KLpuc^YYTPY<9idB{fHIN7YCHl_CL;oZ6zsLd)MQBy5E|7oK
z+CyM$^qdkBYO3`XcQc%5xhr+9G=agtw1mMw%oiXN?H?f355%_Yay={IG2bjb*N8dw
zgv6aHAy};v(vGGj>R4+$5dF@cRn0SYWg<m^k8(hw(bBB`$+Se3Iz->7ngj3k8B=e3
zu!8nZ*izdQcwcW)^u&s;jnOBdwkMd3Skv*STtXmPQXw$j>^AfZ@ZBgGbYgAn$<Ubx
z#3BvHlmtvDiPOY+gv=qJOA_)5D6=59GRaat=e6>Yx?6}A>nJE|!qKrBp<l{~@+&za
z5hs#`RvfD=5{Psf&G6yy173bP+i*(VPAS~TF6X>&at=ot%?{GYa3-E;p?X07kgcK|
zC|mI`x*1+IuNtcrVtKN}=UP!RA*M20=onNTYYWmfk$e`O7!V%~I|-Rz#U3(VN7qHh
zBFQXC!Qtkbf6FaN0@PS3bVJohG|^~)tO@jmgeyg?xn>3Pzogf4aR(;IMYJ{;%=K3Y
zWZa4N&=z`FaW*3Tv>VORz(~u7qAV&;b|!;GU)pUTpNtRYR&%}n(a~LdUU(Q;e-%|`
zU$q_?hP}PJMo0Vo?CRVgj!{*^RVYJoh{P>~xb<5Rl^S)^{YaxkRirlPMLT%WuJq@C
z=n!VfDwd=zFPSNo?qPc=1Z^}yGE{~(X}Lu#QLifw;Ii`dLu97H{KT?r_J{tJJHz4J
z^v+QD;+HS}CkE@+uyeZ!9PwdAtdZ;kS|rd^urz03O_7*LOn@O5tD|6)W9$R1>#`w9
zHevKHvD_3O;Xt5Pi5Oakr-tSMzOm>=;BZ+|XXN*W1`_2mam)ia3Q4F#9l_g(@68^V
z1fSZAD(sx&R|~m#sy4`E>ZA~`?T+l#O4<HkMGNM-Lx=naM44f`>%kf(66DRuvDo~|
zJl<+VM<bwMuToiD+_Bb?Eh7^ql0HOBWo6|WQTxliKCW#g;<v$*;5>!du0u=;%z@J6
zORO%yxn?p&yHW2|s!L0|1iVg>Zw}85tj=zzk-fZF@tvKeB~)iUR$wthnt<>kq4jeF
zB{@yteAS5-F9ZEqKD2t4^}pQj<?!8cBVDR=8ud)3_eOb~(g7%VRK)~OB$Bvq>P};v
z(dcE;wOW6$WQ7mDT3KE`CAxHl<;OsT!{nX`lMFbWnS(7*K4=mHz^EA*YG3@X)`zVR
z5udb4Jlsj>BsNBmhJH2lC)SM?mZNjqKErG#8nF++M58V_4j=`i9;L!4GTq3r0G!+k
z&7hkn6Bk$_Ftar>&zWV#Y)Kvp-N@6tMX*XD#7!H0QqS=)f+8bR)6{CpxuoCx0_d*k
zY(NoHiHJ8I);bD`GqhJ=A#N5i7pQGCHN~7&$up8=xxA6i5~xA>GK`1__->xiCEBaE
z9#XNhW6}f|nbl-NI6>6|P7yiI8e*PDCaOm2oSwf_n)x-NHycEilopaP;|3++<Cg-z
zfgE_yTLUy(XT>O?4nWnkjle0LC2EpKItVY8v87S6vfH+cQ~-TO7O{!t{x>qup&k;K
zL-0OBzk`W+Ufv||WEMeH#8Em)1v(IzMOxJrCSF0fLJ4?gE-dB?6(XjFi-l_E#L43)
zND2kl969(TnU4UQ`BIr0qUC~*5i;s1gkHmymAVn$6pNJ;t^XTK`kIp`;Ax{;qv?pN
zA|#}YCl>mB^vkSt1Y8)!$+L-vDcW_jdB8W_7GFjJ4PrIO-fz?=wX{XoV#yYEHH5A1
z3<jY8h}8wpSoXW>CUs+s%|^&Z^hDNpFo;{9K^&CG5tD(GPz2frj>4$HaKJ?)<ziHz
zfJqWfr(oBMh}O8@vz`S<?8TwH6eEKx@Fs}2WMeH64u<D6SE0f)iWtEqU~;zGA+>g1
zo7XM+l|zSC4*A$v)N;8~Sp1hyNHGr@S|d<*p`9XwPTL3UJQT?z?{SKRc-d-q2!J2M
zDXX}3;|r~5l9EZ`O+X3w>Sj5OOb;3nj;d59Sh!f;Fymc>)}j-38orfI!ATQoJ4`!4
zjku5Hi&(e|U1u;&Pmr#}Dl31Oa=ROGz^}G53<|w60r=>qa{6HlLP$pElKoaZG!l<A
zfshteg!*~>bOd)`WhN!H6uhE{o_MShf`8Fs)Biu{cDq;RzP|GK?KNKAUhscpuU6YT
z^7jN|>vm7}KKONs*XLomRy@(|o|t{!dgQhv=L&^$eBCM@FpN$7W7b2^i;68iA39<p
zv<W(p(?W`29Zur_jWA@^rq=${=AZHj`l-``5dVbm3NL-T_D0g~a#*#W)f~-NSL}_A
zb7wa;ZF^($?C#v0ZO_f^?#v;K7|!i{-|W_U@Pt+S@oaYR7Y9FgZheCX*3X^g``|Oy
z+;D<nHsO1NZ`tz~E<SGkG&bqw(A7ZYU^-lZr2t5oW+7BCB&$iTWl*QCtI=}Z7%}*2
zxH_P$m}8=q#xW)-6w;)3fVs~W>xA$6E^?#w4)r)%Sw3|NzgKHve*eJ-FKmx(duR9N
z&;4{UoAcNPU_V|;f>|ikBP;VGvR3zd3wpu^0UcXf_U4Zr8%B3`Z^G|C{5krdeC%IA
z1~OD1A}%EAtbJx!OPnzXLM=3JQMLy92C6!t87|14if@TcM%MS9yl~~xx#u*9uhD$|
z{vI~gwfXtwt-fVF-=f^1oK)HW-OiykLl~4I-ue?((=mw#NAu^ldtJ6Ssepk;pU3z#
zlQ#acPP;&^FYvCt&C&^M;QwD_I&Tei$<XqG&}&&w2K7nVVnhVY12O^PPfX)=AgrDY
z!DB9>10hPNZDvCEaK|VDeJ<pB={mv%J*zQUXIsr=43J3NP8CyZ%w;b6>mN&}^SSqC
ziVzr?AJo??o6XkzsQuo5<TTNIAbT<yPPBd~oj&-N_rCYNU+4bYzV`MP>rw7ksXv&>
zys|wSwVTcN-XEOIkWRY!L%H0ol=+l5j-PlRW9I>D9>bF7G1l`EJ}NhaZiDW4wIpNU
z^)ePEt~xam`y|<w-%|>Zz!i|C%aUg4z}{(bbJXi!z6K|6*dCLL`tlJ}>U6mM;2DrM
z1E`Z;u!nv-pZ_?BBm41u{y)mE=bI9M-$Jv3kSSJqD<~-MV%Uv5n6vqoznITAPZ5ax
z&wFGbO7{EQo#+!|$@6@DM$6VOMWYWSihop0JQ$5WXbzrpUkeT;m$P5k@Ao4abQbn~
zz237)*I;)+Svh-%OIWpLAPcaC(0RI<3t$e(5Q%jivL3-d0G|}gm@BI7hGyW~Y?2w9
zs10Iv6`#|$A!K~mK=oKod5o|*3dC4tMqXKa^wi}eK@96BT7Ef}A&Op>B%;x>-x?BX
z0aJ)rAzjKh>czcUb-!4z7knc`Mba>;oj7qKKIq5htx>!;h@XHe$u+%YYsKqyycKKN
zYYsqP?67ak=bDQO-k(e^HX(>h<u2S^wYD`pvMjOjc#JhgYWc+cCV4Ak(K)0c4?p_o
zqkY@zzWn73TK{KDP}@<f@z%FUX7QG{lH0lmc@&Y4O6U~+h(}ptH!xy9!I+7|4dz_*
zm&5_0#@Ln~E6r>q!?0sl1-E0aPRA~QN5Ln25WA9bxlAoW0}?7cc!UMJzCQNFZ5wE&
z5rj&fk(A7nOh_d>Fd!_b1$YqvH|7m(^~ib)IJ=bL13`5BVU|lHM*Q@KR!!j~jw=v?
zXb{L%5|M8S#0@Douw@|-ly$N|wzP*W!qp;ahvStZ2z|u+Z@TFw*cS-@7|9Ev%)Cej
zH>eDpw=NVi#KX$EN@$LjiJ{J5e{qww2Yo<be$Vzv!{Au`H@x8u53a2}_=X}mBfb>I
zyP{g2!<oY=mz_L9)p9jP316~f(IWoOQus)X<b>ZvM`GpgkKO#**OC(i7O5PiX;LAQ
z*WOE5uYC<Bs%|_{LGJ~<#_2>7t0(I<!hM&0j*0mCw35r|MWGe3u?(hfCXku>x+1%`
zrAMQ!t<flLCsRl*+KY>Aq!XkNeHy#}zt{-kTW%0ED>=_*fDxQm?l#t(9jXA^{s!2#
zO>XcjeD6lW&RPxA_azI6OU(?{g4w6eP`8hztw-UHD!}l=e$l|XAQ|R1D-B+QhbRE#
zDPuUP-^GiAHN~)--{Md<jNHlkyw_@thGbn$b~}g*TeAIeu+ib{sH3ypp*bs(Ir5_s
zTrQz~6*>eW6G9m12rwehaMmV1^La?lToy7k<M3r2kC%#NA6+)Cdw#iCiYEqxm6gE&
zD)gz(AAA9iMa*<D@S<e?3LB!v!{N&6q1!;kfp;ak1*q*}4$VsDR3vg5VvJ0D1X$)W
zHoT(gHW+KX9U|ZZA=05j6oNi-gi__`iU_QdvYFV36Km}q?|8>ssWfu68~KjtW|Y77
zOC#>7h_(HZk9_1!;#}TnZA7C7Bz1VO^?yVhLmz(`c=84!$ZinREM#>}&|H<zIy13{
z$JEK%Q|>qAbP>rF_^OGHM#3zVHIs>N+84h1C-&g0e^QWFalsxG;qX(1i{u{LH~;%@
zCh+n8(S7R;;vo(`cPV%7Asroj?p4=c|EkYjfBoki9NS;~V(epQ^3WQiu>*~D17ku<
z)Cz@3$eA|yh8o{oXKHgfk>v#g_515q0gxriU3YJaQ{pD`nB&v#9A`_`598AOz%9?4
zfABT8KKJ0%wSRE))(r=r+&X{${IkwnJ8x|t{Fb$S)A>Jp>$SX&zxz4ItS_`4{gz_~
z8`g&oHm(1BusQzcgU^htPaOD1jvRU6Lw`E9{`TM+>tp8s*R1_tyzjpItn!#WuY9zt
zjP;Gg*=-V&<`Fze{))X1M>5NEhu?t0@o12PAuf&N;7#f~>_tY3ry4N%0N(&E!))U-
zh_J~xgadAsY^$u@!@n<NFbZr=Jf4NdU>?_lV30=qpiN8W9z!s!2NWo?E`Jg33nU4K
zMuARYP-o(-q0Gz7hG&n;IM+mW)_Re?!Ijb3$e6NhZeuKCD4X&6hy3W``m3MY;1x-p
zdBxV&7H)yj+u+;8xZ{^*5qq6*9yTseTkv%!z2jLpOs*!jvAFXP7G_*~pFI_^-UE>V
z2#+CY%vgnY0@q=(;Tmx_!_wj}6ZJ;D9<#mdejOGw>d1q7X>ofpfIBaGv0>qjk#Oz-
z5<J2h+k|;sd9@@5%YXdGe_X5m@h9^6pZv+Y7rN+r!K_FV9Jjg*X8@~;w{nBXB?;^7
z2I@hNwKg~bSDHYrk%}f^)?plMle!Mn6(d<?`iN0UYb!Wu&{Xs^rX=vE7=A#Og9pK2
z8R-F?7l>&-joBa_NqYA38Ks^xaCa8t`DP+|&UWV$B7<wLG#mPsY{y^+6O!nX28+21
zG(p<AC5|Pp7l1ZCBqOURs?irxF-7^-tKrwS4nBi#gk=@+l-O#Lc%_enRAo3^+ad}>
zq*OXOxfYH5&p+p-FWr3b!Q0~T{BSfF=<b7GOr@8AxP`0dH{jT<_nDJHVFa!$5Y?=w
zEWrfZYge)5@qNWpZWlH;3vB{A_5m2Fzg5)-e~L6w3FArQ4TvUZJ2|dwVB=STjo*dW
z;1{fSksspI_HWx?u>a19J56WNxyyNl164G~?$(Hi^wX;6tud0R<5)+coXy#(M#Oa9
z6GO>03<)uE=NZYnB%G68&;|Mm;^|qveSRsa0>|ta;iEuwr;<oz$EWUADI<2ccFG2&
z^SohRw<(#cW0|E9N_l;oJ&b&QT0S+y{S<u#5;4#pc^=8yHB04w5D0KMlI~-BBISzy
zwZY-}c>v2c`Lz-?LK;TX9ImMO>cBReRL$2_Dug=)1VR~>Hx-OROqYQAHb4^|&MXq6
z7^b_Fga}+^s1{m6*Lw0LVf8>Lt}90Cl|!ng52r)gD<LzbMszS5c^9bSQi#vL-qI?m
zAfyoHcTd026qU_cjfMzG-Kq2Fgp7bx@_5ad&Y5eR)o*EjHYS`Es}1~Qo@iR|UMoqW
zqxynGnL5(%z#NfT9`%5<Dn-Y`k*)?qs2T#n!?uqI7(pjBfcUsdeZeGXcu_u?R?~U3
zY}|uEgeFKs8hme0grcNK)(TNgI<I}H;Z<wOzEl#uzOP9hF$-F5Whzm})EtUlhwT<?
zuGJ057iu0($QaI0aTP6?no{<9BY0~JGbDr#h<8qFRG4&)Xum}9T3GsL>?jY{53^g$
zh?S^?5!5kL98^6rZJ>Ok<#1I8-N62ELPIsP!X+-yNWB~KoPdkO>Sl-F=`K1pNT~`V
z_BSXZvcPl1k!am8noAUvg0u|f;JV#@6tjv7+A*($cy_#!w9A;}%dE%9{1cgQw*@^H
zjo8^lA%kg7lTa>cr%1kv_@R@}x%=2LS@}c>ee2qhJ4n`n7?!xpSZ-tBA`3&10|qo!
zq?twEha!dCX`<V(3W8crMvL>b45M<fhKbxi0if-uor=W6NSItNiyeq~j3|-E6#Wk&
ztiX9^RLZw8qmr>WMNuH~VQkX~qD0i;IQEt$4ACPA4F6M@BYD0Zg+DH$(ZIlpA)&JP
zN~2N4-pCWH5g{?TXQ8G`St&bVVH3untN@rX1&O8-*!2?tzItSN88+L>6|iR_$53+q
zc+*OGja)WfgO3H2S4moh&=grM{8yBsF_Ut9OZ=4M5W>X?KP*|gMc%`1i$zDyVN4;*
zq>Y9fOQnc_e~9gCE!WV(Fb|}eMy5dDupVN#vzdlQPMsxj4r381gU6A>R+?(AJVNUA
zaW22AJe$}TC8B9J?!sRzAmhnath&Mjx;QhWh(Hid{~ZXtE#kn#$xSZgh}y%rzsu1J
zpTY>in-MP4aLk(!7!$<UTS1(RQK%xqGT!riq?3>isjKjVaNVp!pl591$c2qvrdR|O
zjP6Lspk8G32=Vc4A^d?R&4m}W?8OR}PsrF^tD!(;q<E!c`CNxIlsUL7(t>;z@kj@@
zE-zk7Niwst{$mirO87#Sq6ABk^@OzE_DFh}sV`eOoYk;KgRe}&D$<j5bBIJc7t5EG
zl9p?XA7TU0Bym#~7FYL(^m{iG2f>4~*Zp_pa#=YHl{N-ehXtO5&GlrwM{me$nWY<w
z3qVI=q}`4uV~I$LU}&VimI+@mq0CPNlE9w9nq#9W&ysBon&idR<jC4C!h(e1Ra0Pj
z`5>_=L&*A;U2MITPFKn-M}9~0Z8?&FtBEH=K~6Y}{2ezGTN-gT3^ukb5(0kJBg46a
zk0`=KegU}@I+j8O!4h#%Qt=$I^a*?7*w`hqg4={&&ZbMHdL1ea7a0tK32ot^pCbSW
zvq>x1s@efad9&63lXw*6a;+qj+o+Re0iif;Qa6@(C523}q4-^rnrF)+PN`Sw6OEUJ
z1urd8PSJX_<JyODhirUFtASx40t!)s<qW|%v-L`+UZ;gN@<-TX%EbT$k?OPA%n3Jx
zPZATDB8!bkMwY{i2<Di@b#xGmtVZA|xlGM5bePd-m6Wc~g3vAylxT9T4QyI5WZXpU
z!A8VmrILfVRg$476M#1?OO_QZOHx{D<&!R<KZtyo<^m)oK+aCZc9;wnqv|2xU7Qs|
z0#!F0D|>W}^%4qYiR}HD^>OH^9V=^ntMxNH@O=B1?7wzu&ehJhId6A<%K51CY3DQW
zCzsKKp+8{8XXYL*gb^r2Z^)|YJHQQmf(bwy5^M{V8a?OelG%#<2WChatc9C9tOC+2
zgr2U-^Po(O8j{+`^#mT{3bjZ8VEhTdHXc%^{t#<^VLG)5yLnK}?5)%xTOdeuk+d0N
zfG~JaN9#T6(-XmaJSFH5IuNs4xHZs@fDflcfuV_mwvKvza2%d25^TxwhNcSt_5!g7
zRhU9`vpPXh8(6GR<k+O)3eNMJW*HX*UDdE{(>3k_p}PK-x^21%(Z`6>cPe7cdk`HO
z87Z5~rk#o&iI4<un9CK4e?<MeSq=&w^#b1z70vk2Jk-3T+zKicONdtkx?|1iru-h<
z=~$Z%c$7XVdt`Y)@hwHhraUm{7rir6Ma4xmLJMI^n$V9;rYRU+C3c~@q8h`K8hXyC
zia?4(uCDTEyX0vW^@a=<Nj)-iD~UM|aJ!i?Lu^kSL53o?CyatTOPG7JKA?MPn$g-1
zS$88197DszGH7ZKgT`3<nma8_mw_^3Vy=<!kTTYXX02!}R5mn}*7|4-wh7wGtUR#~
zumq;PhO)x3M~fBeUPB%%N%c{^RF*RqHG}J@^59L6G!$eKVO`P>*!O@y)skV+P%Fb}
z77`5`X)W-uF2RX#dkJF^ro${g`PNLk2H8rVT?vDInm+QF$>1@~t&!5W-WLdrw7Asi
zAPS20y7Tj`M#Ndz+>lDoiM5(@i@mNJd+u}FF+m&;ke<(g4rE|#rK5<Xf!V)o<8(lp
zG~zGH(pd3`$Or9+4gi`;wG8-UJ$7M~1$G4n9TI3OCMO8j{uq{U(0Uf8B!rIuN>-{>
z^7L{eoM5#YxeZ?O)})n|THc98lZ9-K0xWxJX$w59`&5fd>?gg@7K2*dQaG8yMbI<(
z(l{L<)K1_4!n2Vu5p(nyv>y1G#DIDdSo-Vbyy!u75ECY}5~)-JUjShRp}bP5Tvou#
z@_Cek0;bZqQ%W9;E%5B5jb|po&cw_c{j;b_LNt-Okfz~gUWi~-WBNvsz*vnbN_V_Y
zYO*jkQ~_6X(UZRmgstvF>>xXG2r)vlc`kk7idu!MKrjd(0No%e!=cj=2qQBxF64>|
z(MZu;%JQ&uW|T}4Z&egQ9um()6X_Ib&cH@t{0KC8^1oNl^+%kbXr#mAVJHV~3h+Pl
zk1&sZuwLA1r8DewNUaea4ewA>l+6yJ!Zg(RDEYNX*t!tElhQNTjpN-8?f4Q!Z7?BN
zvngCqV@a%;J?}iLfl{Glejn^kJ07zNdE!PBRwDemZzF_Pq2T*iiB+0S1q86|UhmNA
zXavfR<Z{IdVF`;nlO+D1bjA8Br_-+GH5!#P!3~=nF!3@UplgIJMRJb;Aia>aHW_Kw
zK>s`jomik{>H^5v$*UWzSAGg?Th=wFIfnA{>!gePE6^&470eALCu)Pex#7CsPceRb
zIF}%Xb3tZN7;2sUw-gXHR`;roBATrx4x^Ns!UIUtnopC6#0f*AV;dsUhv$OCe<a}$
zu>hGwn>})gGC>Z(!O>JxHHtg#bS?abW6>ttgt>*IQ5|Wfis^{Ai@AK{R))gu&;zNa
z*m^hyurf0G<+?<_r45k=_983-i?jx~L^-IgELN1El31(UC@B@!k)T7QLZeYy4)R{h
z-gA}ST}6{KizBRGPG~`?DcGd&5JDhIhpsVou#fPk35J?yblN=a%weGRQU-*<z)KsC
zAURuT@3RW%8ut?6Bg&{-FQI(ba#Kl`kRIkPhL|JVp+sDA>adtP{yeYC7z#E;qm6^V
zqYzZY<$RvYUOtb<gZLC}NI6p2ho}sBbs<-Ae}x!}2p`o6JUMU<$Y$a2Y!7d^FrG3{
z?jG4!KP`rcA+YY(Xc7*Ww&Ib4<3UylxyX+lx&xW-9JzdtgdPh00J*q668ar$U@cgu
zte0A^v%bgr8R)Rzwf@#F+avp&{RR`aLzuwzgbl!H@baz);u^tL3vo@YtBA8P-a<%J
z-b4~$q_@OKLtF)NgD5EqVSq9~O;Vibdn8N&E)XgNksz3u2r0UtG2|+E#rkOAApy3L
z3n1Y|766Wf`^WK+@9j;OIG4yFBuW=UFC>HXXk0QFnB?**Zfn9P5`_Q<A*}|(BP9?e
z2%ZMFLtXW^5iOS~Zku*&&^N@|AW$mddz&b|WQG~j<38n^HV2{~wcd=2kIR;igrsbc
z8eO9nmu8je1+*DbO}9U_(A2PK6ePft@H1$=Xh_q3BM?VmhI~Y^dl)!$?sUkIl$=3S
z#+eDSM9FRfkvy|-sUWGA`*eWriB!{|&}0wcIrM_uLp9Q_H$qaUy15U5hzbhYfe3_7
zsA9T6xTZ3L*Ykoorzm0s4KWhX%rQ1TB(ofBR1dR4yL{ROx;G^F6-6Rj(mj&K(`xRM
zm4WV3A<!gDfreg`$r?h!fP7YoqKnk%Z9~&qhY2VEJjLA$9+I@1!j$YSHj@}c_GolV
zm?Mn2RTBJ!=abF~0_$ZqpTI*HjzhozU{Clsthz?psW5%OAtK)CnjQg<By!&W>XI@$
z-)oZqHQUInXfGc8GkX_E<i!opj^uE6<SD=e8xJ@ER)GS66hYQ*b~?D1@;F5zPYEu7
z4(w0r5llpE&joE#-Ng-%9^sMNYa}<qQ+m|04*ndxh^Ri9*ijUoOadVWY&}SS0mCTb
zsR}g3F>kf1PWXks@2BUs(q}yH#`NAFDE)eJIKsg*HD?il5e$wJ*L90`5@2l6WgDz4
zdyAedw}EbbNxOI_;K8wH!+#bC2+Btbi%bS9Li(=w691RStB?)1J#R!P!J=FC-ht4F
z8!hE>W+)wnf&e8j5pvavx>W%u+!;O_CcUj@5g-WF#v~cvTB00q7Meu>l%|xhxWwNt
z0S*F~$-q)0P1A&J^vOMn!2oe6!b`}<u@WF@8#+aEWQYffgb)iSw`jx-v;=hr!iw~{
z?d@&1OTU~SWHIoE3<hF`&UtCtKnx2tF(5UVN9Kw?$*z#oW;R7LQl)yxgvx)9_5r{l
z;XLjX<`T!v1FSmJ6j)m=K2qR52r(uiG}g>?n6fajP>AFZ!O~vEStI6;8$D<t#AHZZ
zG9{W`I)~Lq1r3Gyyn)3?W}x6RcJky&Z#eW$;*$E4mOIiSLn8%dh1_F=oV<EQC=@z{
z)`<nRf{DUnio8~_>!Z=NShkS7Id-q^Ao4wrsx#^jSs<aQkuYMu%5$I(bf7aGP&TxP
zRo3mwl>!tER5?I0@wuZp6t{vRIc@#Tt`XJd4D55XtHr91#byHmE=3@pse|-AMDZ>a
zVF?kufW=HajZWZIuA3vb7KCNv;9b`J2k#<et?1{1b^q+>U$EY&lPMjp;DzM<=yb+M
zp_mKD?oS~98+J%!HvFIA`<e4=qLD{a$%4vYkC+lt2VX`0WBmt(#XWu7ur+Mi`PUWf
z*24%D3vYkup|?-R;4xI=vUA)u{)TTx8|F<wN647H4M5(S0k}XfAfI~4h651u){!|E
z{MLB@C_|!)ym@3|9ub3R&M3l>&TgZa(UsX9del{V1Ke?WJAico^^q0&GxPc&W`l>@
zP@h<BGN)KVIn`J^lEP&J*n~(epGQj>#g;1ei8|J@7_~uq;E=*(XIawiCU&>P7QsXs
zKEhCzi9ACr%b0{rLF1<`VjiG>eg4njW&Lna`=6m_J52WSBBli_BC?1$SQlX}V;mEy
zJ(owrijXWC+p8lVi)GSyFGrxFs2vOXas{r2+F6WfR6QaCCF1$KsXdk`mI}qV^}?X~
zCpJK9w07}g$brsz9lGuH(DBgQ$szmU&_9s-{I$^XU$z_e4fd_}ciJDef6xB1Gw=MZ
z^N-HO@F2VyJ|0G+KSiO8wPwJJtbxH?O@O9@A&1}yFw=A%-4fa`t$fi5dXLP6nspw`
zwb9E2(+ZwbS(JpJ!#ECsSUEy;!d_s}YY9LYQ5&?6whV74*afTusx&uyli75$<oxxK
z1T8`Pv~d(QuZ;Fa<eE#VDm`ce+E%c%oL&u9=1y%>@Vqu9k5ZYu1awl~g$4n$@}wLh
zurlPADGchU#J`@w5X<aOGc1zmKo&3tpm;fW1<i+;qs!9EsUa$JgT0M1hFuQA=Snb4
ziWxG2;!&waVMPgQHqVtdfC}53SHOB=5%WC46v~1gF=th-YLPw>4_t&us$8?CH}fV^
z2hAqqqU^v>SAoojNv30lFonGHP>-4RbDFw55pr%&jD`)##Y6;|8C)bInh$BZ7%isN
zY;1l}@r2$7)^EeUZ>9~31|m`I74e1usx;ISET)plM{I0ZCu2sF1~LxRg=7uUnJ8fu
zCv7a3L~T*gtQx9?ARA6-H)F^W(ode&_yp^hHbNu&J%G}X`gxd<(qK?1eH_wx)ve`#
z4eVs(W20e+X_4j<^wAwD@)T>ntTClF4FEDK)@o4baAZ~yL%~v{SczM=O->>vgEP`j
ztL?=}pt~Dss)Dji$$`F8T1(v<^bMga!>@4ZD$hJe3DOFxExI%K4xS2*wcvtl=sts|
zctV<Pe)G@#@)T`jP>f7)GzXJnwh0P)GxOZQGSE0_wUUU+6r_p`Z5;3!s)iXcGl9%d
z{k^<yw)=)QE*rZtiurF~U<Bv_X0a4LKm5d<Xmx=hha6mATU}Ug)J504AE^s~S}yXg
zwQ=8dXBQWtZ`O7Xcdf<!{WJT$9zNQl@Wg&5a7=(DmVQWB8k$9j0A#(eaO~LG=VQ|t
zN#MQ**+ddanD00ypXne3vZI(A4!TA2f^rtM0AH`V6}(H|VP^t?0}Y@QHR54WaER;!
zC?mgsK3g%80kAN)KoJ=thi0fG9^g(%;<(??OS)<<bz;yg+~8P2&=W}s0s;xzSP~c2
zO?FHOKJ*SyZc(W8cp`<K3yUmMi6|2k6wo-dX%2!^WF*aiqCvkBg~o}a4FiK9en?2p
zq}rrIB7lrC^w2GvASE6SwBht3U4(w)9yJhpO(ZA<qxMbU+D;|N!BFpk1jA|_w5s%P
z@g^!o6$EX{a6#!YFa#iRP~1t#z!*v107dIsL|H^IB+VjB2J$YE!)RTcRGh(!_VHX?
zm_IR3tj3$~tcJZpp<FKDfRjt@6C?t$TN15rG(itM#pkA#Jq>SpF*0Z*7yPaV;Ykb0
zm`XTKX`-shTw%Tz#;+!PPcofT3}+`7IpXKA`hx{wWf>Kp8|#%i?atixk-a*Z!S6?!
z!Kl6sJ&jvVG?zQ=XfkfO&DP4DSDhLTm0|7l>2LWxCvK7?X>QSZ?@PL7qqw#`(W|11
z?_=>|#EKtHCheVr^iH=*W&HipL)*<P<#K2t3$-M}^ARQ)n`B0dpda7wbbEt27Fu$B
z<Ie5D9BaJNY7nhE*T%0Y<+QTdtc`plp37KXy9%R(o{8a!Ojec&yMOR`<bVXIf!KDD
zgeBgL2jOYAS&u|t?!?@*W<S^*nq@dHhEa2;b?2gl<)mYG%=#q#1J?{4&qbz~XUm#N
zX$K8UzywH7&lf8KSgJd?lV?<DAn=86f<1({Q^|NenMI{1cm*dpPm}o!+skSmYu9*7
z5j{WQWGM@`rZ7=<@EH*&MN4=*Y==v1RRqQ&(k0m<9y`;Zp}!6ALL2gC=TV3rilg4u
zf>V=WM=O3L{BR_Tlo)scoH4~3j!IpC-_<~{k;z?ldpcf2=1$=eKSQEvCYF1!8qQ^N
z&w@+V(uU8I3YosL{1AgF;5zWjusgDV6Hm2FhuIub@?nSUVFnN(&pbm8i-{(b5EB5V
zT&q>BI`Kp<MwlkX2&31MU5`YT=Aw~ou_!|cs^>{NK1{ILBUiCQ$r^WCYU%FiTuBDD
zOjwpg6iDeJC^3`yBk^{s0`t?1EXD>@7lWr;l2pJ-U^;E@HzgS>o!Ez6vbx>t9Oog4
z5Mm>ZU~c$YMwa0rvlK(y?LyfBZk+701Wfi%CfFiG>|u{bs!MbkgAmdTkawJ!<t(<z
z{>8`a1$curR;-~pBW&^39p3(W!66OB#Mpy6)_=L}w%a;8pZ#LzGvC+gbYAh>-7k*b
z`ObIVV?FQS#g=vN!EgW8!H-$r7pjFEau!@={dy=Ln!}6ub)ol!emwMBq2CGp9g)a)
zt*fjTTd%af54*Mx)8vtTmHl(}AK72DFFL!<IcQl!CJI-IoRf{J?|<znB_8QX7&mf_
zni54*$wfiAfr3kjZ75_2jZmIQJE*(Gy&07zjNg^VJqeKXq{`_-%0eANNnUwcw4^Z+
z9*x}5^l4vCNZbb%Q^M*?1%PQ~`(?-u(-g_Ad>`^_R{G^)XLZVwC}>SkzpRtZZ)0(^
zHCp30DqRSo9ypqEpbEsUuJvWvdSxTG`WS-G1_J@0dZTBH^lz<n`PnNvGpk!8&{s>=
z2-U{j&>tqB<rU+E-Q2Wco(I8YriIB75yMo81rTurd11IcXpeyAH^i5S4nt|`+t4LG
zlpiFB$XA!=Anqa2%#}cc%sB|2QDA9n)YC(6db>9br#mx*)APt+gL)}bj4Q++^xX(T
zn9~xEiXJqLP$NeKZuV6Hz2z9LR*yj}iEz@LlJ5mf>oy(H5mnG3oekIv(^OT&)2^qP
z7)~XqngCj;g<;OjRu<}e?<oc@I3*(V(paiAvjQe2R360cbl{=RRCZ6ZijA1&Y3+$C
z0)Ad{%u7YB_rJ`(cOgpT61)$@<rskMMeYJVNJa%hk8ogRi2$Yl$c?12BJxfGQUlhy
z91br6C9<%>Kvp0QVypyk1UU07fVebh<NybepqmtSfT$v-XgG47Kx8$P5a{T>gzit_
zrL3&UWnoNo_0t^O#R2gTG8u?H%+27ab%__KT;M(j708uaXq3)U1L3yf*!H3u{Y|<+
zuN?L1PjY!eFp=hcFi#A;T0NDr5tOADmnN%g9UsC-iQl$6U7{gWo6SUHqtx%Oxxt;F
zO4rL2PHwN6TS@(7B_49UyAFN?+AP;tT3TOSEkY60{iQ_!d)!)|p5ItqEfB_};Sc<B
z7};1j#$o{Q$K1idtAZBdvFgrFN-l$mSY<H4bcFGMz<Nc*eMtlWZ^hi?UPai+jaPSe
zaqSnGP~A&P!-QB4Wf=y*61V2Mz0$)P0oo@04Wnjkcd*LS7y$QecX-7yS(J-U9(3q%
z@HUuedcXHYW{N2yd;o;j^k^4ylL_|*`VgA8<kXRw*9TISe342B`06-&`Y1TDB<@J0
z*C1wict|or7Xrz=>?5FDgj#%racoBKxV^otXB>ww3+{o}R0n;`{5Y1LPxH4W(bzyA
zWg-_c?Od%x@U@!>FVPKtrsHWp(MntconjjAh2QLjkIc`L;3qSRM21W_R7bbB9}A!D
z^;k3sm+>EFFqp;=JD#v#*Gwgn@B^3cbiK3P9)s&v`vXEnZq2=k;ERO2IpKsG)kvh(
zL(qsWrdBH+T=(QFAkm}EM)*4sQZv`(Vk{iTtsdi|Ih1|#&`kNcaI}Opjx@fo`5nx)
z8Juz{5}m{G+qDbBNVtcHia-*>XqdE3?8ssv>=g?H^l)m~jQbEHw4pKCFhgZ+5qgEA
zn=cQ5XZAAD*>L!7UpczR%@Z%UG*XItWnX-p&GkNHfwfQ@jEf#bzAB>$GdcvpS-^D@
zxMMUDtjBa<Dbu!^NB-jnKJbBT>oyXN!>T0z;^BuMe%a}>XV2EH7uBu)!Jk@vDBI(w
zrN%6+{01-lo-Q6hTiDPR+0a;0SEL7$oEU5gC<f9>loGJ=K-B<^a^cb&?bN%{iPy0}
zV?P@U6R%?9Ia}m9Xbdu`NbF~nyosP4{obHYs$^Lo{(ds^Wda9S(W7V2?)>G==gwVy
z&51@`&X@M;>NVG#KE1vH$t9VaCA!D@@3ifwjLuM?UH>EOmW-qXv~U?|Oc99Zj*Nu~
zxc>4rgU3JwT;q+<0b~Lmns(w!)MF3_;v8uJ^)@WMo-~pmG5JVvpeSU~Wn@w42F;z4
zuc$oQw7qFP;*htaYAb3Ba7PxozNDVwZKX6JG$iUK3Iej?6NtGiWFzcmJg&U509;+e
zppfuVDMAZK@rvBtFGC75vDkIy@kxioW$7Az2emv#CZyA2bK53y5%v|;Dw`g>0Kd$Q
z_tJz?*%)b(d<jXk=xN7_5jaID3J4T|X)o~(a1rKgG!jckNIFc+212J4BE&TS)Huy%
zyNw$;NJhE3r1*%)x|q$9L7znwbI(R+Ni)GlI3QwKA%ZvU)$xy&w>_qgKvmF_bl<o)
zIwq$Bi2^SEIa1X3;X#xl{@F@s3qPo5lcn^5(8HmB8+uFVZNx(RK<F1)66^$TeeCfy
zMllKLtJnmxHb?j;TrE^AY^@U$UB?ccn;|viIV%s@()vUcv<_vX4i(!4p$jRO^h}WO
z+T1$gaR^)`l##oVYQlwJZ0_ci3f&Y1$tB53RhkL`!|N>V<L5}FT0I<*&%NUM>#tvl
zqaaHpR(6_%{CQ+&=aIy-@B?c<mqdGy=)}S6yWQ>`**iPut^FHrxS@SV=FaxdU*B2z
zt_L4{@Lbp0r<^-8ceHQdq3hfG=gys5d5n(-|A2%(iNx~c-`siUo!L9O=dHWB@%jha
z)c8RBk$C)(oyVzfWoNRSD6N><%<bp7OEs7>|DH0>bv}Hq{dQOfO6h#(^&Qplz`ol3
zA~kRF+&k`2<DMnB@s*38hHLv6aRIIeZn(be_l;yL*bfj#67fw%a@sSzt?dz;m(9eF
zVDmu=BK0#Mhi#<I!<(@8FG<B()ymT1jk!WPx)Dj`Sb8tJ_RPW}R;AWg%yp|-tu8K}
z&lS_r&2SQL_SVbJ1UC#)v6F6cD@U-O`DhFo%|mGqoB8xMKFKR1HsMYxb^`xZ-I1Cn
zrex|NVh53>Wjmc(UY>WcAC*t&zqcO={hf(Bv4U6VF`{d{koEXlGIM@+=!ZieqSuD?
zVwy)tm;+V0WTJ`y5XktpAqC}iqE7W^eA}kn`^;JcdPyn>{E<{pU1v~&CpJ8#7Zi(C
z>P`<dfPfu-nWe=NL!-`NkbPIJ;7nf&+SJLn&m#t8bNu%>@0E{~|NcVc)}Iy@6uG#$
z@B$K*hOe^rNxpt`{^-HWaTL5W{BK~HiSW7t7fu1?SCqcli+uLXnKO|atW%~G<*@ub
zx84zf{tRD#CUwg#x7-+czwi6C^5OZnjShFAm?CceZvTz{zSV-z!^9sPgG40~nLu-R
z9xAl4=el*NKP8zXa2AHhkzJfx(F_77;lW**h27BNH6HIF33I)ca6*-s8h;I}c$rq6
z89$97|Kd@2&oIN4zYG)?G#|rkntu#IcuO2D=p8IO!(?!#H(jj;u5(ob?|MOb!_il4
zEVA&&miZj9or%gslpV`q-?lJNm#Vqd?&97qv`(xMuz*4EfXHmQU8_m72;qWnXtxEV
zO=qx<LZeDi=)z;MG{MU3yTF)r<&_04123~##ubehTNqnFyvi!bIEaueQEDz%=_-br
zoy*sHO-#X{FGv<Fqr*xSYM6Wh=*P;XT1laCQ9}Z?i8I`&RW0MM`5bgwZ8AbXkCpw_
z(7Qvwa9O?}m^U_1Qn1f3(!ntBsTFwnnwB9-5@USJLN|&>qlsl;1HU}6p+M+4=^n~H
z`il_~0E5gKWXcQRr|wl<SA35Qb3QhpmrDKV62C%u%O&Ix_I0eNqvi4|cfOyjt>tva
zje`-Zsmb1h>+6at(r#rJ(NkJCgZ!%Hr1kq{|M>R72@IXw)gL4vkc+gBk4Wx=EbM!B
zc6N6U?QCr9d?!I|hKCN#^SN?pq4r~YdwUqWubpKu8OCOUxZ>qXZQ+-$kH+V+IoInE
z9Vt(44De>MTHVVPt-|_r*h#61Gv|-DdyVDRJR0|9s))0?H@8|X!bAd7FX_R%WakV4
z4`vikuo8oH$V9Npr6XIPGs8I2d0D{n_ON?GUn8?$)-;1H)M#91VU2@zuBnnTrZ0_=
z_5$~9v+5Dg2mp`QLxg<X*f@4<W3#D{kIl`EV?RyyWXrmhb<{%j%eqtiBl$C@(nOFL
zU_wA#vwqk2!`(umi@;PN(pj5f>#-|~nI4me`KOPyv3<01S@whM$;cuxSA^6_6-ucj
zN>#ryAU&37@Xj_OLJ3GANASnN>?0Sp^SD>Im<`D1corXMF|-fO@Dk#Yyak*7VC;fL
z7zi0$U?VBSfEB@Bme4@|T6)^?jL#-9z8IY`HfJo;fYjv<G5Fk30{WtCGX+cs6MnzX
zqzeG)aRZKh0JOA*nk5!<8s=!>{R`rd78Y#%u!l92H4F1-Xmm>dKYH@y$wokI)vc4A
zLcVTKe>UuTzVJM&ZY~z;<hP#P*f@Fed6yq+oV4y$C8}Vn(bzJ_mD?7i2AZDI>3j2q
zIt4y*ybf*0q4~TQ8c&A=1<E&$Stt0};HKBVo_44iuV<XSOKY=AoT)LH0&irjm8BTK
zry<Q5g4y`fPjY67Vo)bVBO}gw98cAb(_c@#_QYQx9R^eYLew;?mQ|^5>0lXu6uba$
zzhu4n;8E-R!J|qE%MPnp=VwO`2!-hX?AIJzH#=jU8txt0-<!w4DcKh`>LxLBja<6?
zJ{9@o!Y^bpzmZP=hL_3wA%A@7S1XfwBY%HM(EoU%1GtSgAmn59C@}i&;2PtnI9-n$
z8n;RmE1cyuZdz11=|BT>mLLv6pTQJ1ladszw(nntl}5H@gm@C~O#%ZC4Gu6@f=vS;
zOq4$n0Xh@tB?V!5b4992cVvD4!8@#%9=s#3@HcturNQ@K!-j{6<8Ou8$M3~a3Pk`Y
zJ4t-BXl*50>vo00V_~loyB<~_K}E5YE_Py9$I8t{E>A>hDX{2uJ~y!5N5N01;2|PV
zU^-s2>VM79mP-Xl6GG!hn=k4N9eWP1tZuG8Sgw)-JPaVNKWi{Q7a<AsQZ~+3$EvRB
zZ37_fi~j;o`2WEZ4&W0)mm%Pcgw%TR84@)6bXkKl5~(f(<XQh?ynprK*X|ujCg+F-
zPCoKxW9R<SJh4V?u=-ju`414hygti)M_#jcgaqHgeU0Yi{`rw5%XjWer|$^vQ+qCk
zT>JIbL&QRQsoEnt!O(GL$OBl+Od1Z<pt_;j6|8FAX;!v~V4YKIbv#}7v->zZ2b7`|
z1FhlB1OC;szopphkXt@msZI!Vl_CgFHj_-m)A>SY0-k~)>$J<rc#^5s93jIpmCA6g
zRwF%9VSWL=E>)`zhujCR+3UA3*3KUMF4RhgEtSvJts0v@o9Vm;4OX?fc4%P%=fq47
zdIishJd}yEuyAOtT7}Gs`-u7#^l~tgA7(B;*AL<z_vi8`WQ&DSHn)&WG*b;A7-9cP
z6rIo4IDi98U;O8bAGdypn5I8K-qBySI+z~46LYDbw?1Y4A$G&W9I*S0Ez>BLn`|hc
zDMMgmGORbp><Wel`8eOa-Pu3h(3%KMfUyD4k-4*H?&}$|6z-Sw*w<UafwfId#)h7h
z>L8c~O_|<MG-XZO>`lp|09uKp*=A<oZKl900AJJ(^Q{NK&3sb{(gJ>=HZT$lb8Z{P
z77VnhD3fed)9nAhv-5y+>^ckmoTGBprI9q!)E!OL&dh3NcBk#Gz4orV7ZS(olEjXQ
zucSLO+EKHvG{pg8Na7?8goHFm$OBS<Kq!v~34~^d4GBr8DKwh^p*(nm7a)zY=KtL*
zjmNRg@AuwnN4n+QbMHOpJ6|suI9%D|j~Fp`Th2J%#ZWDHQ*JMP2qYO!b}`u19{Csg
z2GmG+jSPUq_lPcDw%u>{%qr#lPN7v<&q@w@kwN5pneS!1V%%e^Yx43Ahiu4sp2g{X
z63vvLEBO;}&Kk3h1hnqxSGulty%;3AFzDSGr^BJ~yG|_f8ax0u5d{lxB%R8OU?$<+
z<nAZAYxg!Bk)}cuq&sfOF6@tLBalQ%@fz_aF&Bn(NJ3D(zK9=mh9&|Qnx#%?swQ8i
zL4HT<^F-;QVZo+LxUw)BBwB^(M(s)kHOZCa^ddTrq&I@gK+KyB1xLkQ)SpfTrSO%S
zJT^WMN~IC2SOUS|U<P$-nv^Bk!G%;dgP$77k~3NHS4-Xi3e}R!5tTA(R!R?$N`n>R
z_Yet<1BIW^YeH|(@$$_>;01~rIv`Gz0dY~l;TL1A#Ff&>7sJ;rbIaKigqX%BQ}9Wi
zl_B&piFoD}*pD@#*W%ZT7cc$exPcWIuMxZ>P>5n1z^ecwBvy2ML?hTHMbQopCwB+O
zrO<*DNMO*Y;&$TfEdxe26P4KGVDOwjohA!9Jm2(oDi=kf3Evw|lUFbZt%wuI8X6|u
zZY-Y3q;V6CCGbDklsuZ?fpm&=Vafi?RVvXxG(<9+#LzH-9hVS)By)8v5hwezR?)0f
z1dZfUpEC?&60bsnK*Vd8Sm{Ur7gJOTGHDWc%VUdyL?U2RFd79lMV(E=F2_*qL5n7B
ziDAHKi;Wn!Sc0B#kj43L(uhq?PEilsJHhiMBT}9^A{pE!0E5KT<RpoN5{}6@nVPwe
zs%j$R;O*egWHQ;Pnh>kGDD|Y+3l?7DsTo^9MWkAQ$GI`+7hh^Lap?7tf`K^!MZis>
zk1)Yw8_o{RB&VX;Ow0y%50XDt!p}%4icR0~D&P`vVHP_PqB&JqNxLPlFSkI$C#pPp
z^YJw33}xRhE;mMuXPipT3}Lu%Z2f~CtPcGswUfSIN>C(14~uwOBusu#vV;?8oa9}D
zdpt=d<HtK9>tM6+PvcuD`DxMIk@G&HzAcLw2@zNFiwZNlFyteJZ6+g0p2!%MN!==v
z@30ym1uQfY-scK${+)RD-T@w$?23|LLNHD5Be+erg%kIK9T{dm8)DDP42U{~!-oRC
zG;$k&1{|88?j9K)5pR(=>v=Gf8;>*1=uX3tY$S!oh`Qb@tJl34J*XNPoj<=O1<JkC
za5Vb#=}VVxzb7}L2QtUs6AS_X@z{0I5c$c)2Uygyy!hl=G!h#cf?6F*uy#g~rO6IH
zI!v?3EXsETZ#_Ofg{+u_*9-Io;q|`i=<nm?u^V%El~v>ur?$5!iVPuUSY{>r)f6^-
zm_ag3WKr*@P-{Bp5SV9>c4_P5EXeO<3fh^v*Z)G*eCVNvUTa+W0@&={zc|2Y`3q|F
z_H#e*@WT&({vPuC1(pAvE3ZJw`nvB>V^^M3lUJTx*nG3XCt~PTm)q#U-tw9cGQn5=
zO~qew-Fx}*%ik-}OD7+=@2<P<dQ0r>cYIupGH(O$T0a4=B|6z7yq3i0!q<zHU>!Xt
zldv}72wXB-4r!RptDhjC|8A0q>U{oM6}>;IugT}Hd<pM|i3yLq?}YyH-Dl5eedf%I
zsa-IziNYkhIY9M-0!oTRp@2#{Ucgh~slHKk6lF$rfLa)m+uG6GZwSc~Ab}9g<{fYl
z&Z#HEi4SNOwuDyD6j~?Rd^d>Kz+{gBnig&+q<9j2kT@eaT75YKqaZSMw-@#VMxBq`
z)|E<}b`H!s`_8q<(}z}rr|%$FI0_Om2|JK4;JhJsbpuo~S$%sDuYdVlqLk%8+V6fh
zAvIvWzz9>Tyr8O>n03*xIjb7as6>2tsNd1Yr$Mh#bVQP?m;#$bSOC5lUB>&<X|iMl
z2|6VRB3oEoUtKtUe1hC%!my_%spud+E!2d_^v4taPzu)?@el$J#iAhs&Jz8jW7$!%
zsgDDS(<vb+Q4rKA@)&a=NWe?=E6MSXPjx)>6b2mLBH*V^23ZS9D<I#mHXN)Zu^b*F
z27QJ8IR7AUxYdEM$omX>)BU0`W2xd*p8%iWB&dTSQmsoGt&ksYqi`@ia4d;AVP<eJ
zJ*jj)38hJ6C<z4OBuD@YN@5GAP?Dr}rpLjV>0u1o(Sd2IQj^z$5yXXxmvGc{UPDqZ
zp-b>Q^&vf?KG7G!&*nH%?sa{4_4O%%1P9>&N4}#w0(Cto%M&AU^c*~rASbTlTFT!!
zKu8DD6;qy7pHVY2vooX%mhUT%_x5IIW|TR0{=y3=tPnarF_Fqmk=?i3R8AsZu6R9#
z*fAXBhOc}el^D+F3rut3)(hvSaSRBn&H~&=c3}bU(#*n-^mfwe%tGasg#|?tPaPZ@
znu#0t^v+4cJF`>gZ@+ZIMC!!JlM8?B4bPyE3<ihNNCwwx#>NXydxd%*tFX{_P2Y?9
z?ne)f2DU4bi`vdz6R1Q)9nKCP_pQa9Ti)B}#>ue*S0`)IprI&DGBPku6LH2_=R(#Q
z>jMjdLjk&YMLb5Nsk=l)I2KnRzp+Im2^B=hWu6U9l+>w{cL}k)6KB|~ZoKirg=Mul
zH~-?7%+I~>o_p^2{oDYGQACiT!DtSbLgB6mM!ikE{8AY5`x9eCYvbAqoM0v!BooG)
zz1i)p=@4$C#v3b7Jn_WE*_jCe+=<zli%(EMi8>WeT!Xve^yI`fue>1<4^=<&na|va
zdSGLjkV;Qr*Rm|IaRKE3N;Gob_^FPeqQ|H(mdj9rB>)wTjU|EsOue4Ik-nUsQTO-d
zusD<i-wA_@?h1*4S4Gn&+9km-?%syss=C%)PZwl<8OTYJUb$lr912nvGJ1%Wa5s{k
z+|R_xl=>tYb<!h5ypZcHLY|vwOcoF;($l;#L!!BfxmA>bk9lvtxj8$(x;%UT%Ifrg
zUUzEmqy)rxG2{{6G&hHC^y=fswffN;e&-lEZ4WMkC<Npmu~uUA6D2nIh;*j+NgVB0
z22;y}gV7(0g@z)eZ6U^+I1(0QqCY#B9Ta`_c}H*ZG3p}y0@>kS)%R(}!<7siE3Cn*
zG8Qi6@HT?c0S{d9!R~TS_MA3|cg}FQzYb_3RmiV93eFaNI?59{+7&=V2%!T$?r3s~
z4~JyzWE2GyIps8&<+#8s@;j5^Pz6Vd5n+<{PWm*qH=sj2*O5<NJu=*%o(n~Xh6uF>
zvBTQQuz|-=CrKO_{3Hw?o{I#ngAe`qA11^X#~=<(nj^r2GJ_IZ<qe5Eqx2MBv0(5`
zH8FAe^o>W;ZM)mIc?U@ii1%$c;*a6ug1;g;3}7&cvLo9H!Al@osa!`j#xQO)h9b(%
z$OeXoVj-WA>`#+oDmb9F$Yz9NFtCC*N*IzOPy-#nZwVC7L0=%*p9;chV)MY~7QafG
zdT^U@vbAzMv=Q-X+?lv6$Nb|G?n@$#u_#~q?5B3$=9|Ffe%{-wCkE2K#7sXVMmXab
zT(CzS!#fyP+EkSAWz^KU(QL-wPu2nmgE*JsksXnW1^$eqm-9XKr1~ZG9%^Y^i@g~S
z2!SC7F1pgIGrh99+-Y<`?8xM~#GZ3}01B=b;a}FQz=-p4byLhZO@L16&_Q8qpPw}X
zgvAdB0yB@~jvX60<@XWyI}Z;YI<Y*H&rq_NNc&`B5U*HJ68o}F#wI2}BB7C`YsYa_
z$)a+Ptqu{f**`XVDls;iKQT5IIyE{*CMK`<o0g_eQ1>Q&VtVOThJXA*;RKO@V`FnC
zPOoY2=x9Ec%gxT_mJyQiWr@cJJ)Vslu0J_6G#Gi^!zjRD5Yp*fWtFs6YqK*rN=1$r
zf^Zl!vuk)H;1%!iyM0gn8UXmHV<D|$yZFt%KkWNUY^PwajjqztStT+W?sAd{5%%D&
zoRO0CVYtyPj8WbhH5pKJB^h_zvKdr9?n0C!vf>O;O~{yo0$E{941@0s@)UQWjp$pP
zUow(7eaiK;Bn+i@FSIH$<6WA=7D^OxKEz9e8KisX1}X7wT+gn{ZkD60Z+QmCn8mJ;
zDnBkM>eo}$=q84La#)0WipS2JF%tdhp+wpk3MUib|EnfR@{qkix;KTN2a&0vXnr|H
z$;!dWaVpIQu}5VlCto@?1~QJGc;STigwGR?ue>u(eNEJI(a%vpDSXW19r^x1AP3#(
z?;l&t$HF?9pPCxX4UXyL04d8n@fdlvQ)$CEwsPZgn4AIR79;KC>8nU{sjiw|$m)^f
zg;5HHgd<T$Xx6VHAssp8zyXn<e>68Sg#@TGPa>jvqv>>Vbt36u5$F5;@sW}Apb8yZ
zAMoi+<~Y#|l4C>7kB(9(oSHV#q%uYak`x{pI4L<m6d*k}SiqCThz|{qTs}QLoWqyO
zdzy&9B+^v&gNU7}bRzK~STlwXIbF{#q6R38jYa!UT)U<`qi#O+!P_pX;2A2Hq_l4c
zjuej+hB!?fbDlj+G1=)8XGYby%H+{8f^8%`QLIeZl_EO)o-@>Uj6hrvWuy*(fsfT0
zWV8@+*^3=Y>)2}`pz8uhfTaUu6AqZVWWr%_00<F(+|0qW;UoNw;dp?!s#zgUPFplm
ze#Z}3Y??xf)C<-8%EBr2VFOH!o@VgMhliL&eEy>OJl^I+HO7b3Ek}<1agl#4aiUQ<
zn#trtr%s((#kp+#s}t&GH60lo9?lGkas^W`?r($3zER)ua~}N27ZZt*v}6fEWEdJx
zCTq8;{H)XviGc%uR{bzs?}}7;5V_Nz$5Uei2ac>bc65J@HAK@{1D&XxmEYdkyyPQw
z+8v%zAAJ1r#|`6gzkexUyw>1<V2SOeqkHc8E8D%6?*8?rX?nf<$A9U1=RZ!Xug{7*
zQ*YN(^zh%?!r?{hdp#caTjrIF6zU(;L)3RzB$gO|bx4L2I4pCpc5>pRq(_!*A;SHF
zQl=DkJvjnW#;o0CUL8g%FV;N424Xps#jhT{@{oGy%0q;WQj_tax4lg^SgXk^cb&&v
zwushyLC?>he-y2NTTt-Q`npyRMI*`c=dXNrA;R*QPe!6w9zBm~?XLOxc`4;{?ZUiz
z8&&(pb6mZ?eo6R1_0*rMhtzwKy<dg@Qy+^Nj49kUF5^hAPAXO5Ktx7O9krzk^5jv4
zO%%N-k4``%-wslhJ^tha|153fvxZMT_zg74D@wH_c~w62<X+{*J_<G#d@Pq62WUo7
zhobNFshU=|!Mq(~!I9(J!08KH5_JZmc#2>{jt)``6!tpEgTrT|l|(&-Vw55b5OHyt
zoj?-vU4Uj&FF<lALH<XE2eUdpkPhEY3^`?qK_ny?R#(GOE8QTRDl|s!ginq^&Oigg
zX?VOrf?Ux4hEg;Pg~noogA&U$9*t66YzPlfZcK>KOd^9@1bkBpVj2!lo?Bm^oK&Z=
zdXd5*ko^=DkU)gmhi9)N7YeR;Qhgm<ah}NZcZwex*$#w$xiBF^C@7)#>rlL|zI?s+
z>mUgkc4m)3g=UB?lhKtwceG_DoWb?eObmcB_LF&1SqKLY`EjP`aB5^0)}#YLH(cUg
zz5ywfN813RMN|-_hPY42t#3(YGZHtQ6mvdS_yNNG<LHo48ONeHO*jXW>@hjXlk@W{
z8>dhAzdD&4Plc(REWD*RzHss4#arhk2MrD>98JN4Abt%@Cg(6F#!j3d5nyC8(N6^@
z_OrP}Ke_qm`uoST94F^se`4}JPkJDLeGw)%?HLP5!lQRnJR&kcg+2)?9~cmCgXjGv
z`a3c_<oYo`JT^x92~YOynbW6J&|GpwrOuo_dnV)Y8+e3?Tjkiga?*2A+!CgEYAG5i
zh#L(hvrs=6gValawJJoSOH)&N8NckPA^{O@HVU$*`{MY={t^18AL<LD#arq77Hm|%
z+4ls5<a+gpdM|lx{zK2`W%x$$KOzOL6of4ZPa~5ctA&6N78t`1e<K10t4wBO7BFyS
z8JTA1k4blC(Va<wCs~N=4siIKLTh=^gvv~k{th8RT*kmzOuqnyD8I0t05`M&H<NTt
zz-A3mhE0k>VMF2q`b2V>#Zcx*BpiVwMIPsyj&a+Y1V1j5gc<lDnnmJFme***J2|&<
zO=OM?vqnx_LR^IX!RJ;`D&+*pGMMMl<ta%w$0<qT=Qy{CYK>=gENsrR3Use>C!W<&
z`EE!+jN`wB@$D3`I_r32&pMJ69vGP7Bv^yday~d><X6Sp#r0d^J>(<Lz)0YP$5=6-
ztS!Vr8AetX$`=+L&P&v4JA6q7NnVmSWKx_J=9GyO9W3O4nPHW295z2KI%~;M$V|#J
zxHelsvo60NGJ@zmM8)`=L$jVozYScV^O!Zhj<y(^Kv=(CWEuBQqF~wnS0EKzJi!X;
zpY_S*>*7<>OPPfe^9!kS0IBCx0$y-@>FU{u<=pUaeEGzQ(J}R(Uqc^A%&zD&1R3&=
z)D(!@Vju~kKtq!aejUh3?FVpWl<-lGgcD*|ijuwo8LofHsVt^Wqee%(qAyG+Y2-DO
zaipK64f-nIv?OhjUNN~ind#{+YNX^&+?@ZCj$U~vcmic4T?t?42c=e=BrDNycpT!0
z1UtwE3TqmmZ6LB^-RF~x5ya;4<Ml3zme74Bv;%^bm0A)b3K>udl?3BU#4DU`kmnUE
z$N=xYj^r7o?k5z0^u5lrld0Hwb`3os0chyAF`R(;1Ti2tiWu#a-$9AbV;JzByW#lp
z$OuC2ay;JOe=;~Ty|OYfcRW8rM8xtWilJfL2d1Y_<=g^oVk1EEq$ebI9VaC&TLw<<
zB2?p~M}<fBc}WTNltJ;6uGE$BC()HEM)WoU2xux+8w2Eph%e72m!LJJH;x%=s^7Vv
z_>s%_JNJX}bM7YxyEi;5Lm`gej5p&(;#d_6;p0Y*BZh^KTqtHpKrBrN2f1k&%ec{x
zZh_x^HZb)lby}%Ww75(dkH2n6BYFmO9KoK^M_S<V;u3btnLN+vi9=QyMRt!oJk!L8
zQ~CpciGN4G)rWrT3|@y0)^zw0*P)-~&H4dSa9Kl0bGT{L<_>`aH3*uKB`)r3ENF*?
z6Z<ghluQmBbp5?tAFt20TCEYX79j^@vm>t|$OtvYYtv~tC<O~R_`xjoawFNFN~iCf
z4_*)zl|Qf>kH3DB0stzLnfyL3g_9nn6NIUBim>OT{>EcE+3(pOTVc8EGpA^xsfnAN
z{Gmx|9wBGA_0ztF{%W-E-SDC=aqiLv7cKEhf-LbwbDdY&6igzxM7RM*G-U+x&M`So
zQ?MDho{@sj6O<&V4@B;=svHH`cUuZd>+*3u)=k<vM8&xZ{OIh_UiCen!sX$l7%G`=
zJqpjJa#8gAqI6BA@=P?Tak6Aj$x8~lK))C555NaFp}ZndU%-fFvcYgPI_N<Z%n|D0
z3B)NOBa0cUuQyFb7Z{7_u`w*$sZ6#%=}8My?em78znIUPRDxJyG5mVQa|;+jgQSr1
z!8RmgB&qiM=VxXoh9pLNjQrZTzi?B`r>I#+mo73*@c=I6<aZt(ME*q(594Z_Lj2O?
zJ03_7Q$C2U1?wiV(rP9vh6>V<V;=>AC0~&T*|86vU=jPk{qTh3>X~ufAY`ZopyBfJ
zOqD|!nNWtj%WAX5#AS{3L&b<grDV4Fxa6K{rD7CfmbzYZ!VF@f5_dY;G#qx{;R<E?
zo#}Kg=T3R`mC&*AYj~#-sazZw9Yy9%3=C6-_RK;MYxba^C2#+^t1sSm_0{o*pdgV{
zCUbK6>I-Y@64)LN&*x!h2crRr|L}(~v3!7X=ENFiJsm1W5qFsJyaJY+#E$*LBV<G*
zvmgO31BHpfA^%q{oZiTdU^<}eoPwi-p&CxE#b7jXD$EStjG&{I7Qgu;v611aYgSjJ
zWckd@x&9QhuTIjJH%oN9FaB|eqx0`(G<i|<l%hNYiAPexE%qlzWgc+Tf9jVBcYKVx
zIp>hz8%Rp;V2v>fus>uT;Ul8C?1>MoaO+NRt;>@!0SqC347ZG79*%*FMD``dR&dZP
zzJ{wDX8OQPr&v|6Bn|_`97-^V^gfl6%V;k?$)XF8oFMc~=trIg8{JRFP+tc5`-Zb>
z@S0F)>tj%;?8mmCl_!S6Mp|6BA&r^L`|3C&;_>7q!O|*j$x+bl%nY@+48Om=F`#Un
zO?^*x_EIXVZ6kQ+%B%N%dFDw?h_41(d(TU52n4ot=IdF#MINrJhlW|#fUjvj{C<RG
zk7qHB<KCi2zYKgBx#^}b6CQs5xfkhU*Jt#Vzgk_mq_fu@8_KT6CT|(nA>W;M;&b2g
zUA!JRci5syhBPa9!*IX-fVhOZIAK*lr~te(^c+P?04^8y^OWcjK-20M9((MuU^o&y
zz5=W7KSelbe|BIXxRgQ^jKxAM!hR$S@2S@Y0#3V+1_Gbvl0aZ?m6F^8rwj~M$yjW3
z@j;TR<8m3pMe!8|PRA<`W4I5}@T2nEg%Q{9GSQdCp7&xncniAYTB2MhcGly-2CgWm
z22vp04eF~8tgJv)TyF$Hm9n~ot(I6H$6=FKaX3r}kE>UZxnJMNVz4LtLWBY0KE4bi
zf|H%s_lYa#-{CnqhTL-uTr}*YG7X2uahoO++Nfu|zdy6=`Sw_JWOagad~t=x!^!dG
z*AbqxeC&^ekC;a*W_;|*dDIX5J=HfJ!^axRpTHX<7)-|)1B2wlpws~crJFsCbM_-k
zEYi$<k2wEiy%>EFYI*)EwCK9N+rV2=yHR3UdNviY$qDN!Z!W9GK}nLb2U0U2>&Wpw
z6q&@;jkpNQVXY;<rXW5C8Rd;x6c1gMg!q9OdGMAYR0{+rmMB>+(%<Tab5!JCYH&DO
zTn?x|Nv^Eku5Voe0}L%Dr>Cc;`xl4US-PmKjSY3??9SaA8>yl6HRsIEg$qhuxX>}r
zU6Xnvk*K#Wq~;PRJ3W`k3Kp4mzTykX`BZ8lwQ$QcbVW}ghbC?%P?f4iYpMCwOhgpx
zk%8TtW@kMoFA^)Tv`7k&DHQI+vknh0@*L{oN$=S++dF5^B1PX(z~0I|oqHDwcf8eq
z!>wv0I5db<S_Cu)2K-<UvOB+ZG%z(47*Q*-i(4QD*#ZAs)UDSW<FonbNYuG6_jcek
z08Kd!O&KND=vHQ@N#xsq&o6jDRZ<5`Bs#{^(JKgT0sWx05}YTC&vgXOiktfeYfxs0
zp~oRsyn|Q7uaqA*x)H{s3%L$3vlcKUs@w~&J&)ZPJbnJ!7arZd>kV&s!wVi>l2o$@
z0uuz#f|*B#?|l~mWSQx3U}iWMe;1lP91)z8a88+*t!-cVEgVr|L%69*<{G4a<oN+?
zOk}&EjDeI>a$*ji9XN31UBbRjrBjnwZ-@*aRM>yk^PzjDrtX<}^Ep62lN(B<0+&8D
zeMYL?8AfTD^%WroTVh>|zL%r#o#-2+V)|LU*Iy33IP}NC*#<l<-+2a%ussIQiP6e1
zTwVUGn+aCNS8##wW(?5Nx<sd$kzW-Er;SLMYs?UM5QUI4E&ja3%b)O%6o$Qv(ZbyP
z-!Joztl+3+Ag_3UC{P6TAqG5O413_>JgbXtl##}Ha0NvNVi?Gr4BhzGB`KGZd-Tyq
zU-Y6Eyx`H@-ACd1)9K|!{BXT*+1%U&FHBHVQ(a6B5Ff1w@{PtOax8$HkOOM0)R+dc
zNaLHxpbLKRIkkR;SB*19D_!krcOV#syu5M#zvE!t5_&5%bJwx}-5P_+yE#m}$}cW1
zE}eplhVxs<KOeVwy99lqt%qkoQS)85^Y6^a@YvX@^1f4#3}<ri=Rc>O$F6zh9lw_)
zt-GXl*MIwaIyau5r|hG=AIw|V2mhkPh^QmJJ}4$76^H0mpYcJr3ceQ{=zd_!SSvlX
zso+YHKh{`$qF4~c9*hP`12HqnT9cLLY{T0-wUR)kGLp_QloyhuIGtCI;+Pp5iNbG!
z<q`Yw;s}QjZpqR>fDzp7GfvYiI)V#}=cWW+rA!`CI<qr#*X5G~c=a4xpzLTUI5jsn
zFgH4SerXYJAXHqT(9l^7Dg^fX)HJn1dwl=^$kjgBV>tBQ100?`a$^ePeo69olXEmV
zb}kkqvS}ippN_<G$Hv1aW@pv46O*F7n91iSr_Y?hvkR0#Ej$@_1G@87M+T}L>Q0|q
z29h+gEH6gU896tU+YM3@R*f7WsDWtVPKiv0EKrAJp^QR#BzO%~4esaybQ$BR=LWKk
zo8Amdd2D>cyD@&u@d85Yz{`1a$nhlB4P{2hLL^0C3J`6*IVaHCY4HmJcpEo8-ED9C
z%2(zF2eIqHPz?^w;U1YuL89^SlyqdlD3COlO2g6l(y2K@2k}xxekN-d9=1r02~jm?
zR7|)n=ie~j?zX~%@b9|Yj3qe%+~c0UDPnKkZSRpbJ}O;)sC(Q`>t=U5#HxOzyB&rb
zTJLTru$fGBxBH#`u|9Z(ds<b<I?lH0JF7p_-PV0Eb*8)R>AOk&V0YVlq>ZnytRLte
z_tVc`ceg`*FVREY?Qq|s>a`*BtGc;+ykGjizR^5rRkpU<x#Oj|+)a&oJ9lBvvKuw)
zlC{;TnysUUa_hP4T4ud$Z@1e``|Q%vChbkP`C_S2d(O7GZbyan#kIv#(xiKXb+@lV
z_xzpPR&LKK=IvaqVYhS5MpJH|tCSk`oNcvp+&9;5G^#t5_Tt@E%dT*IV`F7;`Q#ZV
zuTbvzGj6}@HaVMfZaepkbG_H-JsM$F4W{pa#Ojs4t-fu7UUN96mHOuRy@_*m=g0+2
z?h=~Z=&SL2iNCGB4$cfF*F9gW9MhkJ5p2<_P9b#L>AlU}Y{=shpUrOn<}-R<bOf#1
z|3kmIZ+IdJre4IkZV@o;J>4_kVEq}-RrrSM@8n4>o?YI#NHChsc8zDXov~=L)jWLr
z9OEW}VVyQM?PdJj_sy~0;8R6>uduz?cQ@x-w5oL5Z}e?I1Q$tce-dTY8rvs%;x&3W
ze)!zEzGpspj?}$JdpS$%r{8w&zq#%kXX6`3jn(KVAE04>Dp|mu{ZEg8Fht+!#aA}U
z8j_@eqH`q;A6bAR6f42p1N!doo9vq+XLL%X5f|{{R)g3)h|E%>YD^u&f}X>hZc<H=
z#eYW4k`wGW+5G3RcTx&mEvhB8OoYxUwW3zl8tT>!6oqHhS#=e4r_ZTtNC<OYT~ODl
z>(veFMs*Wb>RZ&U>Z1B4_08%9Sio;1AKUHflDb3PsqRvD6Ic2Y^-^_@x>vnS-KV}q
zy<FX|OjT4RRi>QirrJ{5WF)(+c2rf>s437;P4$3kDO<HwN9~fUY@Z5G58|Ept<)R%
zcJ)f^rr)8yQ@u*PTD?X+tX`{LM`i9uaHIS#_1)@?R3vzldb9do^%nJ3vYfqLy+eJU
z`hN9J^)B@T>fOX4{U9vp53BcJ-FcsSzxomNqw2@h2h@+_3H?F!lT<VKY4tPee_}8H
zUoas*uRf$cjGF2f)kp9m`epSi>Q|-el=^k`8|tI#G4;6mP4!#qW9sATx78=qC)KCa
z@2KC!YV`Z+57Zy3PvZspS@k)hoc>t-iTYFZXX**{r22F9dG#0SFV$bEFQ~s(f200Z
z{T*3Q{vNCL|55*_zNr2Qf4DEHe^&p3Ecaz%;l8TAhIR7a)W56$pmOLH^_1=tZ2}r~
za>h$ud#nOMGP#jwj!bzm9oGqbY?C@gT-5+o8M1m%59wh|(5W6HeaN`Z=?Oilr}Q+o
ztywa_9oKVuUN7hqx}X>J5_Yna`jlSLt9nhZ>kWNcpV4RaRr+duPG6(1)#vpEeVx9Z
z%yBpBoAk~47JaL}NI2#<>lctG`8NF`eY?J-@6dPZyY$`UO@4`fslG?wt6!$?)8C?B
zuJ6~TF6xplYfEqHExoNP`m)~9RbA6{-Ox?_fNp79w{=JF>OH-$5A=ij75ZECx9M-!
zuhb9e@6g|=U!`BIU!xz^uhp;9uO~P28}xVS@78bB-=p88->kn^zeT@QzfHegze9hY
z{(k*V{Vx3j`rZ0}=pWQSq<>hyN55CUPrqOPi2hOiWBLR7$MsL>59*)PKc#<K{|p)N
zewJL%Kc|0Qe@K6rx?;bmKSJT4U)H~(e^vjQepLUu{tf+6{g{57>aD+}Kc+vfe_MY-
ze^P%+|Bn7${d@ZN^&gNa`qTO|`m_3T`j5!3_b2*K^`Ge{^ppC}_2>0p=)cr|rN5y6
zTK|pyTm5(Xf9t>3|3D7Df7D;p|D^w~{*wM@{V)1o^_TTm^jG!Q^w;&j>3`S%q5o4~
zAx~c)f&yG4T3BiZQH}?vBr32`49`QIUW_{lPd^FVDJtWkZd2IWFICN&*@kc#+a@&B
z+zOW)Rft@{u58txYqbZfmTec>+h*NwJ<u_$c4Vt%nQg1(9Q*%Sz46UT%i3)07mBUM
zo^7>!m9k}4gSASj)hJdQTm3foX<3D4#VYgFq(!G*X&;2kW~F*iYsk|lx@{_Tn|ElL
zHD9f=Z?z12yV0y{9{9_a^`O-XTjedQP_?$q(m`me(b%fet9jsUb?mmkUD;_jb^@Kc
z-K;Vcp;EQcDQ_|qma)~TRLfS`yIt9}0;{JATdhW?8F21ZZIok;QX9^|E|jfiwQ*qZ
zfW?Eh73|bY+h(h6l~a3WyR=<1>xEi_2XC}0^)2sSqulY=8||{S8Q-hGJJidJX?f3T
z?Rblgw(T#~n$=1<P^?rd^5@@bRL%O9r&-ze+m&{!u^*5bt2fG)w`R3Xf72{it#-V=
zSt^tn(2@*`Sq;g)eNZblszJNdvaEX3tVZpoS+WXUHYwE|yG{6@-mz%+*jA-oHH-d6
z`M|a;!)|wqMXSYV7As~wz~EX{Ix^bZfI!)fbua4f_U2^2(B7-Gfso+gR-n<e>UPJr
zLOe2OijDnpZ<?PzV$>@2cCpfqb*IWMZ8HWfU$JHGT2be_(6WT9vV06nW!K+n?loFF
z20YtlwXv5F7&fbg8W3yMw*ZG?wPV>k2cFCOjDZVP?j>O^>Aichd*;ZYp=Tc|*v!F>
zb@<NiE5^GGHyM^<18}NV>N{4sQt#c40p0EtjqZp9dhPt~@nDlTwpy0m?hW4If!<T~
zz_!}4qer$2TjAcV3YB`u?XF-}+y0ulUn<vq+wEF)&0DhV6;H8UG7cIYCe!ln1B>DA
zL(8^NZZ(=rM~rn;+9@;`T^Y*Y;llE7Rx72Qmd7mDgUn5_Q`)guJe^vx+UmWeyXP~T
z&8lVW0okl;1K@5oDrFy!+1U@2Ta{w5ShajiN7D*zHaac$Dg5o0RjoDxEaG;B#i@69
ze0x@<*zmWkI{$-BbIYO+3vhED5gf+C*|H+tx7e-JtcK49Fn5ALL%GRt+I|5zZhl~H
zHySJ@S^9KXE?3&Uhj#apz&NjBb()QOp;T#=s#f^Op<tPNmnw~p?JZZfwxUntq(ZY9
zJF+JuUv70+Z2rrQMy+ZZMT^HQ9Rf^u*Jrmav(|fdXUDLc;Iz$yP}$mLnYF<oA-mD3
zmmQ4c-R0el63d=93S71~D`nQbuVodjs^6+LS^A;wNB{&L{=(Y~1^Bsc?b*?`#WS$Z
zogvi^Jm6=8c@Np(8p{S>S$=`d(zf@&sy0C?2ZArUfVO8=S)E3yRM@T9W|ZHW$s@N)
zW~&@#uUxT90x{CUtXca(e%&5qiv<Jn_Jds8o3-AkyL(2ZvA>Oo6_kmrRx5TpRP2Cq
zphTb*#H~8`jN!1Vjb@+>+A9IOfz3vXxdJa$EwkS26#a66eGsv?E3GyNwb-fdgzXA=
z^q>d^ZwG)i6MSZe4_8&88Q9x4+cuzTFxaI=rS7k>UOTnOK>_ixSgD&J>}aFet~lsh
zUf021ymLEZ?Kf!*inr~mpI5wGD#rL>y@QkrrD~^`0Otb8yO#6@`nUDgR&Z;p3tG`e
zx!%~}iR^l1r;_f@+cOS^wybuc*^pHh-lLJ+&@AlvZFAFXRs5yx7J{+SC80jRyk+(~
zNTkknJYTiZ0%J3TZL{H9TV8JN`=R&`Hoypt&Vvu~)-0F8Zm=i|rqHffL7@X8;@S~+
zNq|KvR^2PKGXUIGEWq4bYn1mROu!+<@O12Aq}hV1mD+{vPQ4v~0y_}vhwfC0W$w_i
zke1uq;lr&$Q4k*#CbZWqT3Z$7xn=A%>9=P4t@>`IW%{-RG7aauu;pix%2uNXJ?>t>
zuA5DeK)Z*1d;6iM7i5?bvfBml0qny;0~idpx2>90v030p@F?%}G!^uNXIN;y2Re<G
z8QX+<GK*!t%vv+hwqPXSGy+91n;h}7x9hdq_JOxjD;Eu+Qq|T$K%lb34a0Va*=aT)
z@iy;!I5#!RYImB$-N`?C|1OP*o23#stO8mp$ixKL@7RsacEBtP^)st6b8{0O0XEEe
zDZ{M7EmTZqT8>6pRjq>_AO?YTItDDFhg*e8!xnUHwL&J$0{ld)vSXN?vfR@!1r1sI
z9&^tI<<}a;%dAEN<S)y`%kM$RfgV_c)&VrOu?4MVx}D!bF;VOQqYdzrEEgD}kYFIN
z21Ds`S$4^+_BLw}$cO_s@@m4nm73hW8W7SY1T9qRwA)o{vtn-><pN`0D1()S^C~wW
zU(k1;Fba0Aw4wct61c%|5L1hGf}o%@mcQI6L5JlB)EF(d%+2<x?lnCq>X9ga`M~5|
zLYCPAkyKh`K(i&hSg1$63!BNq%|m7*<UpL$vjbGlk}7Og5=U2!ypSP)RkX`No_MEa
z?pSpQ6pgGPa9MUZD}hjVtFRU6LMxBAdjxdN14w42DTu(g)nfg5(UmoOT*FZk0Ct4;
zD{O<`!3^7t_GPnWFc?C1{FQbG(i(Pu3h<|vaoMWxK>mG~8=xY6*$#Fg^|I~Xu@2xZ
z!aHyT5M!pc6+N<D*fC(&WEFUYV0)_vn{A^8`L4Gc23Qt=vwi%w+Agy%AngHb_-mC1
z;qd*SORLo|Slohj4OxnX%eJT9ta<A8>2SRRlnR$y+4l)kunM)RZdzrdX~IJ^;1sNu
z%y-iYH@D4Vt5OnF-rt04<&wfCYu6HXF$}+dz~f6-e)#xptL$$cw72O21Xz$^7NjI>
zL<^n)YEtHe-Kau|w?ZukH4FAJWK$3qd}_6_AGQIQ(l(4|r5-`7*{!gMg<~;n=v%9@
z51qHn9kUtY=m7%+pBQl%fG)qz4&w^pu63$#H*Hp9gIfyL2a~{Bc_D(xLJ);^nDJ_X
zK<W6vVDcH-a-d_|oqFBe_V1b|B+0j1k@gP4_w3epdvM+(fWE!$TFFiv@=S<;kWN@c
z*g^jR3(eY3yFYN4ObC?G5itazxPnv!+IP6)*{mLTWzIYey8?e=mf8`sgQSG8C$gDW
z1WB(A?epz+tjoe6I@(>opqqBXXF=LqLLRskv<iIY=k3g0Q+^<W0r#h1``QQ1#!ASg
zYQk>_WAgM~KyYHq>eT!@jTUzbf`@7Xf~_9!-sOdZaJI0%@G=N`P21DlZU!Id0Qn^o
zwnf%q&GKI60gm!{%zLVbS$ibDP&|0{v3e=<takNM!RkEc;mghE=%M`VW6g5*S-m#F
ziq-bByKPylt!H;(n?uj)!iJmykz2=}_e`OQ0P^g6I2`1_v#vt+Zg{Ptd0>Dsil%J^
zgmG+?Wr-C)*FsR>Hry4n1sMwpSM010(GK(ow~H-1)e?MP|CR~I!U{nIg!h9hj~^aE
z+M|!qWmqMb;)fBw%OLR<ge=mivZ%Y)Bzo9Da2oPS-HU9!odC|jHQ*AsDFO1-UZc7R
zXu%;B%{Fv}V}b2PtBz!jrUE@gX`96osG(YD!W8-I6_JNLNT(ttH}~MHyHb`TdLuCd
zgQW)Cw^ghjc=js$Ubuut$i>qbG<N(u<`zN<1J&9=eyV%x7Bu?7?aFq?&;0@O2%YdE
z<-&p0u0-Ln7z`+0t5LOLUBXx>S_h4K*#qmG-V{;7J}3a?*j5T#U_U#(S9uV4w+oF;
z)HlN22%pm41{oCuIoV-n-C7WEgftL*7fl8oHri%1Qzf$wauMBzDOdnjeA6%Vy<vx7
zRW{-I5I8*L?P`ElkQg{R8P|P3-_A~Dt7f80VU@!nvw>*A=z1^1`!SwiovL*ZX>qB;
zUZ4PCDWWQA`)yd_(zb|3tr852WqC!T;O`W{e;rS!X?r_1Uk7HZ?l&u~V!arW$>Skm
zje|(%yiLn&`iq^-O&A}SlNLrTAUXwy^obw@Zxn*XvTR6l2|O(00#Ytpg2@kR@M{O&
zJ@Y{e6ba+hGQd~@r$GnTLyESerAE7rGzlAInVm3#O`%?ZFm72M*hC+_BAD8H6^J9^
JOba38zW{E)cm@Cf

literal 0
HcmV?d00001

diff --git a/_static/vendor/fontawesome/webfonts/fa-brands-400.woff2 b/_static/vendor/fontawesome/webfonts/fa-brands-400.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..5d28021697ff1f32507b1bcbcbf9e6a41d0ac99f
GIT binary patch
literal 117852
zcmV)zK#{+9Pew8T0RR910nA(g3IG5A0~loh0n85r1qA>A00000000000000000000
z00001HUcCBAO>IqkZb^^I?9XQILnZ51&AF7ASE)4aqNK9;5+~T)YhK|WkgR}Egk?>
zRaK7(!QHF&0}y`t<u||k!=L{0w}1TWKmYs7z}OuS%XVDPH~jxMmfrWX8zQuYDYL92
z5ORuog^Z8|l2d}?D*;`?E&nClG7m4>d$AQJqRvFji+m?Wk}G8j4<MaiFY6++AWxFJ
zA9GNXo&f*UJAKbp`mB*BR!Ff598iHaU=@(;BK|R`fcW>Ih_EZKOU_U4%&)Igpb}Kc
zmZee|y4$v7OWo|UY$@F#C+X}>hLQ`M+4KSq-Q}U4^$=%?y+G_G_#^DcIBfqszprNZ
zoa@i@i@*QhtMdQ8y{fLR?rO0a?;dkzhM57MK`d!tFbyO@CNvTQ5{NDk0*xhPW3y}^
z8+2l_yTpl5<%c<6{(WXfBgrx*hU9XSHtxM8t4Zt!#4B3Sjy@u+0ANM51c3G@oLWC`
zG&2IUz_P$7OL82tY|C=eq)n6ElwkL=O?zFSEMqT6#QncV&_km1{`pnvcU9e6b?E^&
z&rJ7B_uy{7#wD%5a@LkfGQ`4I0V%+;6Lz8yv2zGF$^Q;9kN!ec`+d8MT>uMUQ940_
zAVorr5+PREmMuBej^vj5rCeN6ntiqEnlIOS*Pl!7`?uaCp67x6TK8cUXJsP#Pc=4H
zW6!jKD`%Yybiyu-#>PeuYbRF!_59s6FcJiiv(g;#BqoUyM8Nak+~=KSw*dG`H2J`8
z@%XY0OyEN%Co{{+*SHE@qiZzz-<PHJ?Fo=yD2e|^<OsQL)@Zp^tI_K|=RRND_dLwz
zc|<NeB6H!9iG@c33y%aAo&Y@(S+vLmSe5`;Z6J%?!(!LB60JZs`x~ofv<E=5js(8&
z0BAOm<Ool2M1UNTZ1#dxWgx4*C}B2aD_QD`vaV6pHA*bM&0;sJ-R!}tTCl2m<YG1Z
z-E3A;ug1nugS7pzHE|&vnBOX6g7?fRJ}jk4iYl_S?Y_4yUkCx5Se87MaM~kuP0|1z
z*s|oA$g;d_d%5<{f_X#0*s?Sl*(Bz^zq{X>PmrRBuw+vvlW!7EthV<AX}}IRhV`ys
zG}2uQg1l^&^S7R^JzlY9g%bM@0LvezUfg*5c6`b2Sx#?(LafzSw;ZFleh=)jFZrD1
zN{S!;PHX8sV=b9|3>2{CK+Vf%^jSHwpxrtFY|B5<3Qb%CBb(Rdd8*{Yhz-XM?IF?S
zF2|DIvTKxCqK&!<bb=xfd`SruC{_eT|F1-)0to-F;C~eSzoP%mYp7pH;|OgumWP%9
zD}nndivFu#)6S(f>-5v;EE`UB(a!tdO=Pnsi~4wLqW>`{wLaAPDW)X`9ZF;^?5E8Z
zeMP=F`QJJmvz;|ACm0kMl=ii%)S%MFPYh=2wWI%T6S?<?^4q!ef5U%Tpx2<Hj;SXy
z#<hmb!<FSLEfz7E*eBbh`AQTXk1wNfj^h6;^k?RA>NM!f&2ypMfim}g9tUHoA{h%*
zDEhy;T6w&YH-^q{?R*8f5=An7(zcx<qip^XZ6YOIp`3HvRxKG_Vg46Y=(`y!EZh&a
z)qw&f+BuLZ*j%uw%t!1Vmd-sf+~-jxQl&(p1<DM}pmELy4=%J{H#IV`t{atQrCf;`
zx#9wgHamhrMjjbAuvKEgfA0^?k0rXjNeqe<kn9^7`go{;)s0cYeVkRAQ-gM<j=no~
zw9ufsuYz1Jty6(!!+#l+^dWsa^`}M08}G-1E=8`mRH^++fd%s98!ohU2Lo%=%L<17
z9>+HN3y!;A`gIk4MO?t&jUL+m*Wwa!voP;JZBiV=S)CX=<&TUV`p`aFZ$n}}Ewo7O
zeBEqa_r5=!t*3u=wHPSJt=QrsO8(pSQ>a{<PNda77im(YQxQ`PkuvQp2lfy<I*ppg
zqR1=zt~36a<7$|0by5fZ7o-ab>1RnfcRZy926dQp^Sb-Z!T;#yR`MBxJ(LZ<Eni7w
z^)Lo#(AnmiI72)y7V4vabIzOfp<`~ug2G(V1dAi^*KgGk>k7CI9Vy3*f-c5f*t_F^
z*t`qFF7XySK<V$DD?xbBc|8{T<hnq;TZfMB))fAt7wy?mEY-(|D?x0%UsMUw6>PXU
z`55<8y9EM&)+Y7N0rA4jUv$N>*j`g{$2n}XTl`2ZHZS-U^u5MX2~sSTz#7NkPd23v
zF}!R6NY_rr7x_dlc7^YWpSz@fopl{hn?B6?$oQ)Y#$pEq;g##1`{VbUt+t>ITsSVd
zUhMQA!T=x`0IUFzZ9fUOfc2~-%?4J83?cvj@CApx2MBr_1P}&qAv0u$yigk2KxgOz
zU7;IvhaS)qdO>gK1AU<%^oId35C*|u7y?6K7z~FIFcL<=Xc&{TaL%SVo8@etvt`a@
zT9}rgsagiDw$?;zsZG^pI{&*Kx+}RmxnFsNr;MkLXO(A-XJcC7v_Wb6({AX|x=qik
z7u3t^mGyRdSADQPPM_-S>Fw>E?49df>|Nu1=5=BM=E6K!0?T1@Y=b?pH}=5+I2ecE
zFr0vsa4OEjrMME;;CkGRC-6Mp#TWP)f1rWC$x0EVP%;&wB2<zpP(A8EovAMkpg}Z&
zCev(MOiO7Qt)exwo%YaP+D}L6BweT1^qw?s$HRF%J9rw;<ORHfSMvtm%7^(JU*xNN
zop10>zRM5!Eq`DSf8!tglbL_<e;FcUWQxozv&!tUu&ghe$d<CTd?wSRE`J$`Mn0pk
zQNpNfEH>5~Ta2B?VdI!_-neeuH69pfn8t6P@WuNo`l|TW`EL8}`rev_&GzP4bDBBJ
zoM$dFSDA;*Bj#E2y7|O>ZaPi3zr4SSf4zUZ|F-|J-|7Dp2mnB4$O?I&5R`+?&;`0e
zH|P#MpeOXQ@ZXr`+DL82)f7q_l(skRx*nyc>Us72dRe^^{i3n@<O7rb(yP3WZ!cS=
zC-%X<I0%Q}P#ld8oQkvPi`{Bmj~nqQp2gev5I^E~^x|)_P&ma<EES@{RFcY2UFtxc
zs4MlSK{SpWG?f<95?V&fX$@_oU9^|>(Lp*%SLqqOBaPc}FCN7cc?!?uIlPEh@jBkb
z+xaM8<STrgZ}3gN%@6r8e_$u;{DXfov%&wRMTW^ZnNDVv*<}t{pZ>Pzk-Ge7L>{<x
zsF2y-9P_7LXs$31m`BVre^Wk~ZhtX<#ot#3w7`eJo4|{}Gx)G?Gv&^dAyaCmq>NuP
zx-&X6Ue0(iL*4Xa(ns09*@xLH+LP>|wv)CKwj;J|s;=%)m#U4`8fr<kgql~)tqL_F
z-QRSa?gs#XAn{pv#Cx$*tP}l?e%W3})Bq5<0YrKLVHY+4p^9XYBoc%W5h7HC2&=Ft
zzZ66HqI^_dD$kUs0Lo3}s&YxW2%v0NHY*#IrOFfq4;@<^Qyg6!RqS8vSL{o%E5$Ap
zTNayBY*uVqY*egXELkjGEEb=R55)W8z47jNXS_Y$7H^F=$D87f@rHO^ye3{vyeeK9
zFN>GPi{nM{!gvAkoOm|z<akm%J{}j3B_17jh}*<1;^y_a_4la1TYnayzCwM``hxX&
z>r?BK>!a&K>iz3|Yt7pCwQp-5)ZVMzRJ)<JT5U{ibgfVIiR!u41FL&fcdhPJ-J!a5
zb+ziM)#a*-R%cabR3}u&SKh0<TREt*cV&&rij{>c^9nH4*L3*i<L~ot40ndx!^Ppk
zaHZ?Rgbl;GWl2~)Ec{7venG;igabcCe)INqnGXZ@?$grPe);IbtX5xx<GgL%tkL>h
z&*tMf-gecei~kdLU;EnK_V|4x=%TA`y6d5@e)=0=pg{&3VyIz;8)2kTMjK<SamJfq
zqDdy3VybDTn_;F|W;@vNPIskGL(fozeqm&|M;&Sl%dqkj#=f5zFdh?6gsV)$baclY
zEW%mD#TUn>=!<Q(LxZsGp~2WL9)iYB+#W-*10NcO9S;r1PKQQd=R+g0%b`)&_0VYS
zc4!QCKQtD592$o`4~@rOhbCa}Lld!&PO>lg&}8g)XbSc}G!+LNnuY@pO~*lpX5ipM
zGjYhFSvd4J&t~#Gb986&dUFh5^7`1f8}L_u^(&gZ5#wSn;5mBG=sCw68vW;(hsLlu
z=A|)ej`?V;GROQh)}7D-G&Z2I5f-Fz5RKz)zC+_A8Yg2V=KjZXtU}w{bF51H(sQgv
z`_^-;PWvZwtU*WbIo70O(K*(lW7Rp<rsI-1)}iCFIo74)>Itkz=Nvlc!TNOWO%vLX
z&fDp{6C2U_BAqYcbQfifvX0Fuhfog3mXy<Zgl#EjQqIP0U%43N64>l3m!@0>hf%rt
zWscn_x0_>k%6&%IgYq!S!*SPF9!q&V)}rQ4+Z;V;?lMAmn#a&Q7H?7Wbe`h?n&*yi
zAk9l?UWTu}=DRdMz){qkyUlSVss9{Dk*3TsfHdu^zj_*_1$hETlNKW_j$=q0&>US!
zo6d0@X~zk|J*1sTyWn`zu7v%`cO&hIlSq4$_Q7eSeR&U^PTG&OKh7W>LOK#>k&Y%E
zg9}K<(*!OeokBVT7n9DXIj$gGJb^1omy#~SRirEFR_+?o)udZ-E$M#JL%4(VIO$2;
zLwc3;I_@XEP5KBAk-jE<i>F9Gl77Ooq$cTCJWu+I^cMz_yO6u$RdO$K1+SAMxxky`
zKIB1on>>d+2M3eqA<v7C$P1xe`h>hFc}aXqUX8puz9Mf)duT3sYw|W|BX39E4ISh?
z$a~^P^4>JU&*a0&N3MFi$VZcpMT>kq`BeNyK9?uZg?t|QeEdtkfF{t5d?EP~{7=4&
zb{R&JZy?`D=t{nYd>f%V`F7f6=t;hVd?%q7`EK(4gev)A8WDPvpC`ZgCG=h756GVq
z`jfvPe@PfhZt$EioYv+e!U$SB(>g$=J!&0H>j=VhT1U}3nlKNo<7k~gn2*-U{9+9H
zumSd5uor@Iuor{960|zltHE9aS`X~?U~dF%0QP3Ew}CbWdk5Hgpv}SF1NKo+6YLXU
zp8)Lx_Gz%sfc6FZ95hG!gMEGuAAx;g0v!PMC9to64g~uKO`t=-z6JIZ(4k;Ir(NnO
zuwR1x26Qyo@4)^7Iv(tAG)Jd_{e2EMfVC6obg(7Zzd>h!?a~Mhf(yXSI&<427lNyS
zE&w+GE(Ki<ZUJyhfUW~K!gF*xxUmTogBu5DKzD%Kg63!*xNYa?Zg4xz(LLarb967b
zcPG$&;JyL(E$Du5-x0>x`Fn7GgC2yk3T5)Ow5VKeo5LWKW)4G8Hs&w@<vMei56VmC
z=zS<}n4=G%ymgK~gz~OA`UuK-bM!Hk_s-ELP(C(CpF;WE9DN4mD|7TYl&{ax7f}8=
zM_)qe=jbaa|Cyt&q1`t}-#~kCj=qKVT66RrwAY=Z@1cGD1o{EmFGBkz(2vl5h34pI
zXn!(+egVG{cmw(s{BATtzk@#({Bd8Ci~JcpM}L7odyXvl^Y>5={6*j|29@Bi1b;Qi
zgP#q4F6bZdcT<V}2mdhmN4_rA#y`h%0>HmLClL7eCI|xXpM(Drf*$Z+(=LMwbf%!Q
z1_U8=)`HHu5DY-)2AUHLL+6e;K^;0T%?VQIyfr77A39&o2^N6P4}Yv+LFDGWE`=Uu
z!ORmnSx-=*M#4f&L!O6uW(Ly9k;%-2d77mICd@}BM^clIHF9Gb`4Xm{*3$t~2`k|e
zW;`3u2G(x1TCMEvyYFr_Pi{7w&B?p(zPs7#ey3WTXz?VjNwqlXev!mAty&aC(RHg9
z#S=wQsJD2cDEgDQCe`AkyJr&Dq*|QlCV*hZ&V0cZx`|<ygrTX-nDLP%!^-f0Nz(e{
zMBbPl=xOAQY&@%km9(C&9Yt{*MNt&TQM7p4Tdnq{X>YaKo9g5^ilQi<A&#ObibZ!%
z4JFBtG?XMm8gutB^O>H|%^b{OFjP;CsfVGNG`VTyjkKN~c^N0t`fw6fMn(+1jFqsG
zjc4Qba8+gKWlZvHI;qN{FqA8+QfprE(g}Xi=N#OJ9&tVCbB>)8kz2RbPNDdXt69ZT
zN<CB$VL@C^UcT*f4trFQoxc@|zA?9bZsSV5#e6=We@0iIyI#TsmgsE=g=1|@tdtoS
zBNH2KVjXjfTX}9WZdoObwKikiNO^T`ab+TIrAeMmr)HFnOlpkD@{g|4b)w^{wCy8h
zd2Wmu)wv;6SxwEPDvO2hIGDglavKxR^^Znd8$c-bnub8En^Eiw%L*E;s5zbgO%Nb#
zwUuC$S;CWnQgzd5VN8hIG9+#`qgYDL!mt%-$MIq5IQ{@^;<>)$9NS!qdW~7UCV<(t
z@Y|i<W*E9Q>HCi3IDSevcPt^Kpp0?LB9`Yh>dfLb4=kHYzuoC=grRGb)OQ?&2tw<R
z!5rR?6iUz$LQqbPndDiTa4?pNbX<%p;8~t=i^n>y$_fsw{Tjz{rF2)z{E0C87<7BR
zF5I1k;K<s$u9U9htd_59Vdf0NMk5^DowY{WO(R&Z))(*&SfH2LI;xB(+~4Bd;saKV
zxIzA^vZ_pO?DmVbHWM#>$9c(<f^$N|mw$yvh-C}uy@a>c8${_J=gS2lobLt}j^n@Y
z7AHh_QrM@rxsbvm#Jcrf?Kj_h^Yv%)+pnA_2;sNh=Q|DpRA2v3_!)SF3_=h^vdW|;
z;g+5k3!`Gn&QnvEaaB$yd6uST3dj-cqp`MlbmZRQr6OU(l=uQ>F|50fP#8tiSjSr3
zm1lXDPb-TXV{zpt#P_AMJj)S2I5<5$IKcTo{bHxn0qAsgT7arF=Fi8RL8r4buMKV@
zWY1CrKZte^3bQ=p@d=)GzsdyK7%Jw!bFhGVP(Sl9;kFdwa5mrQ>}=kX%wQfmjytoZ
z5FRGHyZXnN!N{>1&Jcq2r*yDpgix4pZalv;sVZG;wqh#Q6qK*Y@+|*o(%eGrsP5jC
zYuh^*cHaN<I^TJXl*9es)&N>3aMtX)!bP@r0q<;IyR!E_IW7{0_1|hHiPnJWE6!kn
zP7w+xl`&;en5a&9yv1X6JI@VnN5op2SQn<Oq&h1#5xr*QIR5`s4zIl{KFM-`JUhA9
zYj<jxQ7VKG9Lvpn-6#TxqV5-0&V0A;I53y4x~jbq^6ca!%OR+5QXvFqlnUBvf~M6&
zRePi3_z1!J)9VYE!}p<Mgu*h9nBNB6;>1`8wkDns@jF(($tV?42y4zAH@JxKwuj<_
zy$OEmHSpz92#ZlFR{yhJtKF}%#eI-(|JO*TdL7;Cu8en!R&PtXeXF$`Y~cshmn-y-
z_hh%(lI+(~ul3}JZ>_lw?!$YUcpl+PZ@;!)Ydb&BR?D8K{mdHw?k`{l--}Xojh~1c
zi$^L_BQx^FY?fzvr6Ui;ip6#N$yH@B*0GM&e{=v`PYUPjhS*_$e|Gm~*PBwty4BLL
zlzUf(*LYGm4uJa;d&dBP9eB;n{R8&$|I&_IY>V65SO1*YUDp2$r?5a<xINHskn@Y-
z5Rt!QKn>TnFKi)%AjJCpKZe}6TD`uS3qmX|g@F6W!uL+aj`;+gqR&GJGOcJDp<u%W
z6>BfJiAkR2WyOoMu&-vv*OXOZMrLGkemYvTH9b(MWI*vu!UzV#vR4iV0BO_rVK6Lb
z<!}JL-%Md%uh%<`20)|Hne8mpd%gRBE%b7i!LXbT8tu-JOZ7VK#$Y;!Mx*ogE<23|
z%-#(q^j~m1s*Nro6h_*Z*qi7_^q?-clvYkuR-r436l`dQd#z?hKm6Q{y<Kd41j*p*
z93edyz@7mz_>l!W*uRO3F_c!jKj8v6j*#BVvG5yL@D6x~hZvQM%&;>q9g3&J_1VL@
z|L*-K!4i}*E`(Su1AqW#LI@787eerNzLN_f1f^`@hM~J)lnR}c)zJ)o24+a1y{PTX
znMk7_S?;~BcvxdcvBCjBoO87(+YI!4)AA4Tq`iXqHTuwMiA81@PU#R8bjT}DCX7-6
zeD#S-{FyNOi?UiiF-G<E1&m;UuAsYk%;+B})1(8jlsFZYQ$%L3FuG%VSiTJ1cg^m`
zUJjRa)Z5*t)qS0py7|5&37ol~CgDbV1Qt0Nq}^$3Z1kyviwVzKf?U}VBeY&(1WR-u
zy&5545zDcTHJXAiAI!<W4eKED;5tOJvz5Y-N67Ll*BQ@Ti=HS;jLY(gM!V>p<C7!-
zNRpG|JIlv*Iy(m&gQDJt_j<N$;CIGl`NWKho-oH{`NR`t8NdgRh9hf_?f8D*2r$Mm
z+IAfOK3u)?6J?2)gf)i{!t2%gpFiV1+GaX)EKUqJy|SuI@e6J7Sw?P1m=<Ye^glaF
z*>Butlu03k)&0XB6GBKv>Gm6UA3b_>=MH=*1PE6bXF>>zQYK&V3dtz7gb-)YTm9dk
z{K<G4chM2Tuw^1;*xf)J1^T_CPK$I(Kh_DZjcEYP&&%Nef?#9QBziV6>kpruoz1t+
z!2%kM&SJ6ihkGzA)6LBw8I{8UEa#z+dd{yu!7c8(K>G+mmKs)U$4KWCvnP))mhL&+
zHCL{fqa$<WigBfMT=(>uXHNHPzF*rvedd`{*L9?H!5kf#*~}arakMMmxN&fNe0;or
z^QJ4L>o{?k_0Ue-8R(ldj=`D>1w~q<fguoz%usrLTx@OuY;6|f>(56PXJ=<;2y4M!
z*h02Gw$bRoY_$x}9ElP>bGm8U4RETBl>|*@FTCGf1i^4oHqb(J{>}5*&8RMSJ~)$z
z;jLQ-9&{sTcgt1lotfAFTbHmvCq|77r<n^kR`Bab7ogBWmS0%CcX@A5oS**PJ&JrE
zoNL!!Xs0VYchv=OTp`{4>u}Z*T^OX7s`W^|ub(?U!OeD4xN@X)T>xhrY;`kpSfDXN
zLp4=K8lb(yeG)tz81n2s;L%xYRI&nSxm?=IG^Cs7kKAg0^U`lN8?@Y9XsQBA-Y+#l
zn+f$cg6bTA%9?X%tIaF4G?gFp+wLwxkc$}#lqGT`jPOQVYEXUzdcNcMzZ_YFr4ieW
zwYW3_!N^Zkm44IyXSmsCig}CsbHmL(;Jurx2>enis2tZ7(*2IZYaDaq-L{aPhl%~6
z_rsS4imtmj`qi4WS08_4l}5kj*3u)`wuSWm?8ANM_FiKvFSR+y)pe6Ro7!M`<~vRk
z^9o_^`PHH>V9i0_us@*Qn_A~T8bP+>w&0!8B+pIxMaoLY{ty13AIH$&_ii@`+;=Zu
z5@Prs*7*3x8_d=YbrQsX?E8Pb`p*^c+J5<C|0B%d2hd;<Owgui=X2$7-W8>4z|V6e
zQnl45mQR51w(6_jhix;SIFD_s23J>qd;Z0FY{y|;a^Y~w{piJP!?0q+=WB~gB9aU~
z-ju%B8xCWo;^A;_>7`%5=GJ6-`SNtKwF&QQ&Qv_ic6YP=<mYA<SBs*%wbd<(s}BY3
z_O2GFiw<#uT4W#x^$t^Aq=js^3MvdJ*fZ-a&(6-yJlmcDg?TpJuywKQ9BzqK@o&TL
z=&@vmP?aCPpDrtP%hP^C7PmMHmPfS3wF#L|CZFMhgVX)=jt$_sQaX=yEAdv)nx6jS
z@&cZYw)cn2uzBg|8Ba>b2JoKVtw8VW-%R%8QM%7uJ$5&5x|eX5efYucq04A9&}Rda
zuFG7C*yA65Q3oz3*nVoe)dak~bM>XknncF{u};7!Tn5(g>pSu!Zwv<M;nkgOe6fI7
zCy9<B?~(mmIE5cUUyO%P13f)5abZxZpCOYyb}3UK6CTVW`;plKf^9*YJfB}gu-dY`
zGTdbR^R_5qRBm3v(CV~x)0dWiQr!-lt%y+!GCA@A@g#QuqA-%jy?s7=?)ChW>b9q1
zA-P4U=Seq;T*i)j@T}}aj_Wjh;Er3XZ*C1{%~%kda}2<QC#A(GCSY+3+x|zMv6<Ea
z7(Mb6N^TKCDDHOc8h?1#b6B7hlT0K&B?<`{kbu-!V2{9`gFaI{$MA0MzmT^qV9q)?
zwA}L@hs!fC6(s<WF1*k?hAtV+(Y)~W;cvQtcfj`@C>KK$j>%++v3Qn~o_K5|gs|T6
zP7=Ro3r8t?!6=jPU5qlh`Xe}&j4~#KSj|I#5avP%me1<qoA7-Rrg)w$<MZ8?t`KAK
z<h<oddV;<FX<KXi>DRa8tQn(^oLKY6AGhP!c|;kL@ZFR#x%zt;!~pT?A0d;BQ6@aO
znk%e8ul|$53g*%ijGJfI>wkf_!}sQ`2QpPIU%J1gwZ2&gU;Ufx?9>&c^WOJ5Qo0_1
z{n3vG8<Zm(vHa1G!fT?)CBN_cJOJB~();nB2p8?||9<zm<Nm~t=ls%l!1tk{!fF{)
zmvxXHy^i%Aabim$T!QUyN2E!=-NwWfLfY?n4}615ux(2r{^pyVZR9uqjS$kdF>!z6
zH}uJS_Nny+yc-rM%YSVRS>sKR2U|0X99yjZ%cp(#uB&C_*aFn2(_t$?SYQ1jMD~KP
z2Cw)n!$_=uA3g@Zfyn4qcVrAp{B&bIQUN{5s!5s=<n$ss(oqo?<D$syCt05HxFUNH
z0PJLcDfbr7J4V&$IGkDq64<APJj*BM9dPDIUq(~}^{}NSK&q70(ii|>=1UW3wvXCv
z+pX6c``q5>Ts>%a2-Ir@cRlbt-**69N{MY}iUS5pKzgzsI217S`c-xM;MUC=w;c+z
zgD7b5-5zw~IO?P^29+cUcz`j+0N??Rb=_tZw+;@nA&`xYE@^}qJePsRxNT!_DaKn{
zVcS!YgE2UsY_+m_t;SEQs@Dg6eCfGqn%aOI)LUQRZCD_ILNq{k&_nckG)Irnw;&Yj
z1-{piA}yvSt^BQ>8l`zMm7c0;Q39B?>Do@Lt?gFIs+^_*J5pu~nJ>|~&QrrpYElB3
zyx;E!hNlZ!xIkvwF~s>a9Gm?LC4XDb_hP-ZiNW`y?R)p`-TPKx7WaK_G58amSesjW
zdj}8wZGdh6;cE`|f9Ux5xaw_fB`;%)ZL#J?Z)3jtzaUos7sQ)qGk`{;tX}i6kA3W8
zueo{^;PS~!PuaFz=g`5MTRQgquJV0Yed@!H`HaFLWwpZ>UwrZQwSHfSC-L3^{16-*
zy!=J&Y8eMX5WM)}zg^8vj$gkwDY7A6Jq_PHHed7w#RQyoyWK_zA&l_)0)7h?Xaju~
zLc_A+E65!oqQ>KD;<@ddS9B%H^ASCc=8dV!qB4~!i?P;5tHSqX_~;?v>(?GWymlSq
z>(?IcWQ=Z(DTWY0eb&jzi2_3D<;#>3P$vX1z0(9naq|wvK%SZ2edgo@<CBwTo;^9i
zKF~~146Sar`OW+L`?o2<cxMM=LT~S5M)4`7r<gK~5JH%vmn~3;P7tbi#e?;H)$gPV
zIm}_+Yo*U=f8%H}oxUIRdc7V%(l6k#$8Oy+#c(i$1s)Fe_nMyVcp~a;2SKCVs>9u)
zpTOpoD_7voI7vY79~@sEk1xYJl^SN__QqDHquMoYyNyN!Nirs8bYo^lMBkS#VF^og
z6QNM&gHISNA6r#sYPeCRGE-A(4X!kQSDVT-gCr&23$!xYOigao|Gj_zKBW&CJ)`vg
zpT7#ThYYOyuln;pf7N{p*h7Y2_2=;UhlhuhUc&eirH2o`=6@W%+QOGGdx^o(;n#f4
z;SsQx7`}w9S0Db5uX%tFA_%SRx~73zs6=QeppB9%o#a_Q41%C<0@BHBI;b}|I5PXG
zsbJ5OkuRl;l=LmjTm2O*?@p&vsBd>-B|J~4xLx<9xb%yblu;~w-&ex(tpBo_!SeqJ
zKG^K|GS(X#v6hlC@hWlGskHlL1bPvDF?uKZF7*2dRagl0fonfr#*;KjlQe7mt{$mq
zO6Vj7%vR%KT#O64$p3Bb7*Ux6fa5F?V}gw;unVWhw3ix_<dMqe4yXY)qNhz45G%FP
zBu(-p&!*XQGM!YDsw~Q4T#R$Jqhl5Ei2rUANPD9_SM{T|=QmpsAp|o^HJiR?+mylN
zYr&|!MgOfKu$|rQr>DKiFb1I1ViqF=0DureiN(GN8ja3snQ2tWIes`^+{pcQyWRHl
zjW$5L<iq%1gYK3u>cLjm7xmk9YBwWzy?2P}{O4(^{)5ipQ5?4Bt{D@?EQ?YAj2UB;
z5<&pe7qXxOHNBVBKUA*bMy~hvbiRQw>aITy{|erRnrOmQPRW-Zq0DuzbGEY0(t79C
z@f9+PY7)SG@W2D`Yf&WLPU*jS@ZiA%7eLmcNX{vJyNIH?1n}T1>lpi@5!BvtUJDw+
z$GHBJ+<w>j`T1LFK|}Z$*MEZZpQ>Z*i$+jG7Gmo^hF^zYMKQXHUWT4UA3@)O5KQtU
z&C)E-^1xIk%d;s!E=|%=KB}&Ha?$)+W&)F>BhK@M$jfzjKK9ODS(O#ns?LlTWs^Lc
zj!ZSu^$ulKPBXT$qGQF)sk*u<b1(n065d5=t&YL-_}jp?UC*}R^9ZYNU<@D{*_L7l
z#Ki<Gi*t(s9AV;L621_2z+D&jICmXw#}QX*^N7UtS`FhyqdN>E4{Ei*Wt1x6d6+cy
zIC9;(=hkD&;33#Px3TNPa|8ff*IwQ0lCr#Wr>tn=*j)e#_$W;!#)P=m{@$faySrop
zzeF39QnD3stEFd@IU6~*dfguRC*N~wway@pi4b@n90#~%DbE%%9&T(fAQ+Ab1?)QP
z`=TcGgi9zwE<y<K4BmvSd*yaW&<+}*3SC3D5gHPe)l)T9g_+V8%h=0l_3Xj?AU`-=
zgx@=#JmKo<`T6<zH=m!MpMT50LeY-r=jZ2tcJ~YJ+_`gS_22IP)ZHiD_+(CfOnd!1
z@E_q<5JMG0)eTh2{5?z^gX&4s7-iJ8#!8OEnlI|UnYQ82akpDcW?ypo#Oqa;UP)T*
zR+Ib;b8K<k>jCup)0?NK+1~Z*V?*KOAL}=dtIbU^n>{}o3@I?G<H<%N^xKC=x2w@n
z9J{Xml1*+mo7X1@A!M!pAG(g`a04BnC(!4hH=_5VZ$+pwW-#cEfx5o3D0O8nVVk9e
zr=R2))zT*0Ie=JXi6>0WB<EIY-Ewz$tRv1EnM04p(@X(w={q<bqa$ANil-(wsnIw3
zy*{bF^q2-UrRsHVS-f6XI@sFW>9$&a9Q&<ScdxSr7?N5$36uz>>-wtY*L+_!YY^8$
zukPBm>(nAoe0HF(H}hs40BQjBy6fk;bRBR$J{SyO*q&@Q@BOEpogF1yAwv~KLO{^n
zfLgsX8bPOC188lIJ&*b=8927IY~e+rka6Q)%YWVjZaE(3PR;vB{nuYn2Y`AU*RMM_
z99K#=eDL7GgE!881tAxq^*G|L;2}JM?|>hIUx7b>zrZe5I6-Kb>pYtV%3f|JW)d`4
zrMXE=wjA=egZ%c%9l1><YsyWM=MB{5#!7DwT;e84m**;OB4hDiqs>UgTASMp<R=}a
z+$3wd5g)0Wm{M&b&XXNoZDJ*gm$Zr0vM==QhkOjHw}aK~QD>y|V|T2_MdE9M%8d2M
zXcHIXsxoDvjnr<Ai>JBM%1nx~YUbwcV1D~}VNz6&Ivr4s@_cCdnCT?v8TTQA%w`&q
z<%Jnfr)I((6v46Zm=wn1l(GJ@GA7pP<m$r2I<GjYPJbttRr#bU3*!>?Wx1|+Ra|Ou
zE7iBvLO#j?QhHE06dY_6>jukd6Z6z03R!i0cPMi)$+KPjJAL>}9LEq60NR8a225K3
z;9`tv8)FLyp)HJo5~>MiloCaV1C)}05Fb;DeL@04DL4QaD?%y5L{mZm;}#_ZER0(W
zW9$Mzi(<evXi@+YLQF`!_dh?@`dOI#wqOkK{eb}EuYbX9ItxYVH%n>{ceXPd7`1w~
z{cVICz=*o81(*<QbN8O@aPWNq--myNC;r_`kodV?uV-OQtoGEUf>G`R_9gxf0Ju(E
z2f%jRuMHUGwg)xg0XVKJJwg9y?^=^Gi*|PUebt~|3J?;`?e75}B!KV3A2^g!>i7Wu
zANsyeNI)nB2QX2D(#E~#?JwUjp)G*Z`etGB{w5+{-URB3dW#st*ZSU$Wg0Y6ixA+B
zi%SA2r442WuI)JBx*Z>5-v=zu%p@55KkxfKVHT^sF&z#pMwiZ_H5d#i^&Co{a<%K)
z;7HjBg|uu;jzo*nmfh6DZDG~xb>h@41|K_<SpfgrZEl}=#MiT*UjHgAVHu~4lKg%u
zqC=5ZC>kRz__h?7!tjuuzZ#aO)#k>?j7DkSV$Ih6LD7ukn(x<i+|<7R9QO9Qga4lY
z++=4b41~z@u{k`@vGoGCc@#CZ3b}3DVc3FKdtR{_0YVs|wcp(Kk&2#bzbiy{`4pyU
zkt$Qbay9$#5B<;&{m>8n{U<;9$xoi2|KWM8egvA~IlLYnqc7|uXb?b^>8~m?nNFsI
zYMc39p{C7fpA80U+Lvc}m89ctu)|XwD=RhV{hXUfM+KKG8du}f*I@Y8|L`CF1I7SA
z|5F@>wjW%%c4O9TC>1ui&9*yj03h(=6o}<I;g=GM!LeMoF~$Ix7Ag!K3ixLTFc=J`
zb3zElm=f&!D!~A7E9th{HzWudMltSmwmGNNQ@(`H1Kd@<1cEW91Zuu=FaQaHG93Uy
zfY^EszX5Z!hY<Ap{Yb|;)^X?oSOkr?i3Z)0v+2n6O_{^5uU-bfw0fD8L4RjwYYRVp
zDQYy_)u-*@^S)B8X1k>#O5t$}VE(F0mwK}_^=r_3tes{1m%6GEdTy(k?{>R*ZyzCw
z(E0)+SRjSA(J6WXp<%A`EKL+Qk<unE##Lp?yo!s$jH}5sH(8pRL|K~UI?vM7Bs@~u
za1*JxD9pI3%w)>bB2Bv0C@SVpTjp5`bUvR0;Q8AD0KOI=2e@Sq0q`7v<1HHh=#T#B
zKlAd*zVrC;<M=E9JPz<I0Dv!i_q*Q(cq(fE0DKVu;M)r`0N?oz!1qiCJ`Vsdgs)ga
zdUz4;At7nCd+hr@cxM8X*&@$)TBN9nIU{x&7Ow((y>=%_Ez7dfq|@&0U%vFvgr{a+
z^!Vlmz&-nyVQ>HPciGBb`Ydz$U=KDnkMk#2%Y&!jKj4yvCQs67HmyoNf*EZJqtz0=
z=kNaR?;hL#I;_9&g)jW;-}{Fbyx;|oJ-zpP+jl<s<dgsKtv~$3Km2_tl+c&A4t@jv
z34V{<N609jm*p&a#hWS6PZTdPE0z|(i^2u?R`c0BpNrBs&BpqY0KVGyebel8*Ulb0
z*0Pb>jG48xJ3veZT}0Ei8KR+;VR)Y7a-^p3J0``*vrHiXKnhQ4Hwfd%vR#SHwk!(_
z!<6i4?QVCpwzk$EI8nQEdYE^crkQuTVWKsl55T1ojKz&c>>E1pHRWLxhH;}`_oVPW
zhjJ$fv~r~pdLHm34HTtPB3g!_6=Q@DPRIfNH_|0tvQ4fd_Yg8NVF4{>t76{6SN(n;
zpaG0*!)QW*C!l@Xw3RI6=Ab-u=+L3MV>?dm{gYSz=Kh~SXaCRowG<$&^`Z0S6*wA%
zjr~7sgh3ECpwkG0S8VuQ^TYSqP8vJ*`|q=zBysF}A+7Zb+P?pe0+xcX;cbv32C)vF
zf?wr-T+;<|2YG-zo4kZPPTod-l6;Vmio+Te<$%){<<~fl%fYg+GRCH5GEE0kmh&=~
zqR6S@TKHOpm7}T@C0q+#RCys}l2$?{>7vSubnd~TH!Jc&et&3`Cez7avoQYiTz{M*
zwE$mmCFcp}_j8_5WRyMz@DoU{0PZ>z@d~6G_ydR(A7Frb6eGqlV!ZeEhK?SVQty4e
z{m6naesg2<r=~xqt_c+3)GFnIT9h6_@u)x>kg^+3z&l{|wE{RMbuF)5MbHgS4*?va
zyrBTr$_tHxx4L2Ol9mF#AJ$S1r^54F-*FZ>tbRw{!ISWP_#|nO3AvR#kJD**oe~?8
zQM*!X5;6*n&Gtp2I7-v1lGVg+*zwrHN+-z5m^X+Tkqp5vK;OP7F9VIy@|Q+VEEc^U
zK(Du0oS6Oz<h4sls-Ny~eT~){0Gae$6dGMmGC<I9uCuwhHC|omeR0t5_ceF^QQ+a-
z+u@Vt@%{O+|Fq8f`n=x_Oavi?VDcKJ6s43q4r+~#!zrbTg{$>O7!61JpZTx<`Y*|Z
zkOI)`)$4cc9%0!#rg{i3LP|QMBqzz8<hkTc<eh|6*)%K4GSYKf;mgvg(p1Xlr>Shs
zrL=ws@sfB_mG=r_A~tP1TPm3VZg(bWIaKmlac;1lLB%(vaJa1^Pw&-YniTVLnohH6
zTxP#K8jafRd_8Y=vNp7an;qW=@cmBP_kF(|2q{GnaSkAb{}H?1a%|ga*6cyA<4Fmi
zr0WYQh5y-JuLo}0Fbo(#+Ju45^E`j%3tsTC#~yp(3)cd_)A55>|4n`mS(fXW7Jy}X
zu4TyxAdDJ~Fo595#~uXJ@?6WZEZ4J4crb_^C%x))2ZqOP9ua-$YD%UDPr)Db!3TTo
zc`5O?OyS4#xS*uPS?Lqd_L)DuaN)uQICuX1`SUM-$qm2%2fgQR9y_)P*M0lj-#&Nl
z+xtK9vX{ZbgeXGDRd^8|Ap^2bw#aSdS>(m!m&o4{QVrz1%!{lPGG@1tgE>DhuZJ{~
zc`>IBG=;Q$=zbRF@tibHE;gd~L(|wJxmCcuf*Y1bw5%+XbTlZ-GM|(BqG}vyM{q`B
z8CDJ!Y(xtPrvvB{26LHbBFnj>tUqay)W5aLQVG{B%eYh%p=SRTQ*l$>1KE~<l(s3-
zAZ_|V;5QxLF#Hy!t@nF~o(G)E4Gq*V5sj?bFbvadwFmun%QUsoY-Be<qh1t^IzYWq
z6!iv7JOIx_^uUD#{Lm9zQ&DSmTo-~cOc0_ZqLiUQMJ?-kK7?Ti_fUywMG`<p6j_T>
zbSXuUF02LyhyRR@NC5fE6e&fjqc9u}qVSnuty+M+KJI|$0eF)0+t%0DxnTep24BZG
z8VtI<IEvz4cQA-zd`8XnJhzs5p68`LcpiY~K^LWAP~l4EICj|38c`d776Kr6l6$rt
zH4TlhD*&9SN+onCMWnPtk)~>InNvXsA@blqiG>&8F)}8{$vJW#`4RG7@+m^9S&?U1
zIVkd~Di_tXTvS!LD8|{i%t{fbh?8`}Hpo=vY#xq9it)G{uawAgDu<wPHp(@%`GKV6
zIF8bBOgtNwvyx_0-%AI~ATA@s7mrI=Zni3X&*#~=<dv)zQ`{MKQW}-=JeyR1h5+w2
zgkVise>X!D7`%JXXc|rJ1#N_Pb1<(5z-11}^x@fT)}6u{&}ELdvt<r<{KoR~GDiad
zaGAf(LAAWR%;Ao8>Ib_21H&)?=n6xVEpt4K07@AT))4qIMU3PAL#<j3fGq>Slpir6
zyZ#Se`ZJ90e+Vc5gS%k=N#M(ib=jdjZwBmtmc4s<c^OzKp7Jc&V+lF<J|bKub8<C#
z8hMHGhAIl$Hk}V4m#1Y>g3Ih;*w<3om`~DD=0%?M?>F#aB`(;Ub4`Y@7RlE=M>$<`
zaIs7H+d4v{zO?_TrFsLPQD17;wbr(+&)ssRCIQx`>g}a^0fu(=FFS_Th68(!p<(|A
z8bE8;(T1TN$N2oLog7yipbf{-a7(lJpGEI+WA(R+W*r*Jv6a@RPHC-dM>U{++SOWX
z_lu690kq+KaaBA0LQnPbNfUl>K(4~q!C7(}d4#-_ypepEkdcVQ+pKRbVjrW*vMd+X
zc%GF5mk#%UiB#l;6xp~G*;q>$3LGU^$MXS7n@r0|j4*GK^ZKGJOGRn`b!PaEk)zqX
zEYS)>qMCkI&gSrT&JE4EkieMGokq_xgMLgYz+nw$%?iW=((P5_<&jkDoi(L6TkWTc
z>)zHzKH$7NSghrBlZyRI2omtQAON4YIsJL_tw`}*R|1{{fTIWu{*?zox@Iz^{6>9*
zKLkLEfQUjeCy4XNwkfMIh5%X_hF};)tJ-MP*N?3B6hbp<o5s45&Bp3%Y0WSiUW)+-
z4X6;>paB#%0Z<{T003#tG`U_6d{3Env;+WzP(lvA00(fH#AKD+PDr&-y&8NP-oI8<
zh50Pwf1ojaF;U36#VN?9WfDhHvatfdA7K;@Sl_Fs|74g~Ctdg6TDM!f&*-FA#insp
z(lzd@b-T5D-EMOAkH_9smg`toyUA!LNp?nwd$r}d=2c$YJCvq}dhgTv>KN`#lKYaY
zu2Aag<lZ#BH@W&M_i6*cxY|8b>vn60+^bFVDi;pby4~6#LJ2wGUc@%(@VR|il7r|4
zMrjGTnKT5gDILyKiOXqO{?R||U!?2nRlQnYqrdQ?I#lcHbpPT%z~zgJ;Sew5aJazX
zaIrrXUevW9zT!!pL4H3GmrfPEvR*cjk`oAw9l3N8M>5N@*N?{I(d%D-$WU(#2K9Qq
zJ{UA=HHcTZK$gA!9ktp@)OC*ed%OrgPNw8aawj1p6{}{%<FMrY=vctm9TfSTr{*S}
z1_u=j<tXR)1yZoSS+%Ij6q$$yQXT!jG9*=r7u&7*`e3lulY%kNb)n3%yy|qJR$E@4
zkFqRp_Ikx+dAU}D{SS0&^;);la9zN094CbzbpB-wkPYUWW5>-`qBwrlsbUGb-7Dv&
zDMho{#GBe}S`;VF9X(nU7{}wHUI)lWM~+<eBO87I(CLiF^D6+8_y{zco4=1|@7X&_
zuZinNPNg*@gb;qPfABec6<j7c*(N8+wd4*$pe)NI4a>5OOIgZNP@XbHQ5NN#1e%mI
zE%^UhxfK~&(y|yuRx+}z%Az9k<V8`G6;F$l7WxG<O-l(YmUx`5Qp2!LowN;uuF}Va
z^~P7f`qfgs`qi304#r(~-F3wiz15E|FE3Nh#AdL+-0St8U|i^*Sb@u6?*ASbKV=C?
zU6)E>{geSPGhzR8MNz!$r9~a~_AfV{w!FOj@6Wn-<?5@i7K~wf{Eg3f&T}Btg3}xC
zA>!Z%2bWR8WpaVMl6(c~umg7!k|t6}QLq$8P0~p@Nt0=kMp+)v_aFw<h{dihQ*z7I
zdR|)2$FdhlleDA)&+es>jIygP%vIB1yxf*_Wh6^Zr_KYs@S~D(k<U4n2%R07O&8Ne
zEpQ!{53)g#XZd_E&x<j_^|MME%K9zk^I4gJf?L4yuy+>an$PlV6>c%|wgZ3JSNZjw
zE~?42<o)4T@2JH0mOiIh%3*q#mK}{_YbkwNDupcnh;?%!yL)@fFwb6bFK2TNLLCq(
z@+!*#*8nL6q_9RqB{kJ=-aDe?v^LCE7+nVJ`a$4#830hm=`fC>cnGw#D<c8|;70*^
zf$s-Bc(AyW$Is$0j59<^865-AzIkgy>t653(JQbfW9rBN7^8HJGC{wN1z=dedGCl8
z)7g!P0KcPaDn(p|F=LFLW)Rw`c(eylFaSy!XJdFhXC*+nmZM+fn=S{TPJr~9>+Zex
z_+DMx_&}qAXq6aolQ1*`*kp`TN*lNG!@D=%d}ffQ1~7OALelT&d2f&)rwka<T3)QK
z7J03nC})TLeE@*CKN#ixLEIt<0EGil+@_lsHF!WXmQ#S2#wQP*J9py5Qc<DZ1LD&f
z2n;!=o0M)+&XEB&Zsv!N+;sDq(@}Gv37<WH|12{EdWtbl89UPGk6M&*MgeFxz6rB(
zoKc{sDQ6TYJ=*9G+mtiL^yj}n!-MZ(4?ai=axEdSxZ-)Yi8DB#mMmgi`Nca6Q53np
z<Dw`5H*+1<3vaOW(LujIII0X&{SOVGm2Ih?xd6@P#6XnN`UA^2+VA)KNA+5zTG^KV
z=?l<oOw4aJ!6gvuTU)zFx3<;=fckZ9Sz7CloVs~qZFMLu%P{Ev-rqmpqq{ZTI9^-J
zq-B|!5~7HE@O$u&aG5C5BwOSdxq{q7euTU$dxV6m!^jH`I>dImsDdcLecmqeLiy8P
zRHK;qLXz@Q0(rBIV6)sjNrfz>STg|20j)#6*pot*eM1O7AI5|INh*r+?*Ob@1J`<f
zt>$~LK0nk3;>9!y2ZNQB&CQk7{veF#i^U+@*vR&O0C7YijuD=XQH&Tz_>yLpH5($S
zJ9ZQ~c0Cb|zwh^{Qrgg3D#igR3ZcBHDTN>O;kLdX2-%E0pL4?q17Fd8A3ho(MloXi
ze`~(y`BmDB+q)YZE6cr}@B4nQx4g2k(d__qz7|uAVuT1{I1alZziAm9ODW~XAbq`Q
z)l!B~2dUFj<AIwF2HlQjb-IH=>V`%T`ab76t=X1e#Bt3|5m_czkPC!V(`@ROV$|R{
z<?%eu^o~REdxNg68SF6ZZ-vzi-hLJHdwRXz#a^!mmoD~tz24>Cyeap3y}iA?y~_vW
zQm@y$=hCH1Kzh9%^!6|Ida(ZD-rk<3`1s3kiPXtqCG9DI|F!*8$Z46RQ8ptRE6L)l
zoIf6y&lgpa>aZIyHQD~<+ra!y3!vKtuzsUwV6BEK+jR+*-_>5~O0XzBb&668dc9AA
zvTKO7dOrFG2nl`PuOhlw<pU60t!qnJ%5p5nWhWIDiL;tj)otrSAmt~igAVf`3C1|Q
zJZDM{1_!vS^J{BspruMqEr3x@s}qnYsbL)`HFPl4ar<SR4(13&N-<crb+d(@wEfs)
zk3pl^@TbdwfXh?A(QLqDtgwXMaR)WY#~~wx*w+>9!qvTi{0Mm)`91P=@?G-R<UhfH
zj1W1Uxs9dgcLl+>k>1|zc|NzJI`+1+ld8z)iz$(P?IMljpqN$MoC<@|XHk|C39&(7
zR#ukdGMPqlTo(U)k(0qvSr%n64s54oiUOyyETd9qH6j*L6ytJ`jYmDEW=pA%0}EzG
z+Px&@xh&j|7GuZKypqv0nJ=<IJk`0h(#6Q8tjLRWnoTG3MX{lmQPe#t@_BClvy(WH
z<7}L^=D+-dpE!2YjeC17#u_8TZg#=8J<IbXfbcxa61rY*calg7=-3to!D~w0Ly<B7
z_J<TGB858+VG#6vBThU3W%^;rm}Ln7*agl3IEP)+G^wyH%Iz&nNCwKbGyn)n-EcS>
z7L?imGJr5{CaFRg3=VCdz2?aNCn#k|DMPRrMMf#ozd;I=-st<Z=h#wf)#-X3-U<Mm
za%3DSa)SW?P)ZphAP090u(@#+MKv;HU_cUgJC+TAXqt_d5K6gjtCglelQ@*xcO1YP
zFp8AH-M}bN3Rj=HXS%kYjrv}idH}Yzq9|_G>IO3UQ4YX40N)itSSB7~lL!6^Th%Q9
z;8c~cs2@E7`qP_xJ7!f4#d49=)hnL)tjz|v`xQbgOw6A^Y>0&UE`yn_vVf3ft3Jy^
zmS-2GNmBtC>kO`$Fcy0RVF*K)f9#8rWbTLlKE|-u&M>mx(gzGY6M_%GB~ZMc;?Am#
zpYyC|J<H~KR~LCvHO=|@0#{d$p8W95!v*n%M^{%^>#3ZbZ#UJ~_I-a=)y{<?1fYE0
zczK0qyr;OjdbD2yqU~qCX`Y`XNl2QfKZhT|Pti;1HS})!AblBqJ^d*C4E+-Q5&dP7
z+?kFxPQz~ADmPEC(;VaDIC-W%M(;E;k1@sx{Xt%e*mm>0C9Af-ukuYBW7CdpyVZNp
z^fq>V6Cd(jzl~kr?W1}--VEKEZd|TA0KKqPU)RsE>G{!W)s5l6o-Xaj{TRo6*Y9jZ
zM%USw9pgni4!e0bPs1FiImTmbFn%8o+3?)-N_(?BZrsN>%{ote8`Vx&ZN8h;KA!h+
z+V_E;^Khdl`zA&lF+UBPIL7g~S?k1It9`fa`aumVy4WYM+IPno$9Oo-lbSfjHjb@U
zE!DVdo1ntD7^l5j?RUdrhya^FWWRAt;xu}9)V_&h+u7-O!0zIHh!ot|PQ$q&51~06
ze55qS7>|wOCf|~(xIVSoANJ=r8sqD9I(B{69&E$RG&ffJ*5TZR{b4_iOkKcEGSl%8
zrx@XjV~o+zk=pKtuHOw~JjRY9)?w})OQ!j8tAka0sPTv~ZjOgp?bx2TSZx#FPNMB*
z!M5({@&`s!W_FHo?)pBJVQC-RS?x5>cgB6Sdv+5?TU?dtXw@E$G4AHZjxplnjRc}b
z;;uhkot@V#FjS}-Q?hqXMXe1aqE+8HF~(Av)Hxv)*|0E+SXN}r2IN$oD^@Lcq}E8v
zD2)-D7Y?wY>s;5y>zmbzC{bNoGnvoX8iw})UXU*q6|8dZoQR|blBNg>n<2HZ)@6nu
z0ZbkWth>$^ld}vm#=*6TMEs;KLBx&f0;I0ycL+e3VWCjZqy#Wqwa<DnpBIHPNzk%d
zF3ZZ5UE8%GBY<w--k}m!GAzzHKWtZkSOB<ucMvQJn}%Sy-fmA;C$D!&ZwQ1OlLIpx
zgq{&nHr^P0yJ4_cB_>NkFpPiXJP4kA5{XPINdS@JH`3BKjrZHji#*__a7V!yos11J
z2aq^)U4hJJ-n!AyY}&R>a{`1b7Kqf=*K7nNsZR+E@fD1!3b8sP&KjUC$7??9j>l(L
zWv6^{2T(;$mDIte-oqIQ-dkoP9x*P41k8d9_t=8v@|cno#d>|=2*xdz?295Z1fo>f
zq<Sfp3D{5#K-1<-C6vPM_Ln&-fabxfRC(%c0l+g1oSOS5;AYidlSAqAG==k_U{2Gf
zT><=Py)MZp8-jq;T7UorJQD%H?>_lmHGr>z*BlOKW$DR%G{GP>+Gi<&DX_p08B)6e
zix5%p3S+GUvw9<zoC9u7R^6IEngc+p*Xo$W6D2medirzt2mD;}`N>x$-<kY)^2^EZ
zB!8VGccyucR<>{dxS!)>_p{c<u^n5ah>^5@t9H|Fw)2oHi_sS8WVJT{^udRAnN&bF
z@9K6lC?a*AAq-5b@R(&${O{fN(>yf$ah~Rb)wVxuH+Ew;ZL{CU={O&DJ-+SZ7$3)R
zo{qhpZPyo=4#Rf8)%`VWHCo$kG7feqS3Bb?m_3*XMp-TaBSvlsgA-aXo$lJbm-gi0
zgX1e79I5Z_cYVKJUrzl(`ql0Js=6`5X1slV&M1qFXyZgzySapE7mKT5ebP4fo^yF7
z%<rdDpeh|+G8wV-h8bdj)PK{(#RcO6mYVXEPCh&f{pQYU*(^;wS?_i)I^4d!EafAJ
z1dS;%bQTW*rBt=Ry+{E7Z8Ng(8H~C`6hUuvgs<cyM<^W+T|R0yZDpG6i7tEXIL?h$
z6hV&*#sNA=1q*96-%D%#W}I05@^Ay;C!q$w90vpd002geA**wV((vMDvx(STS~<bm
zb(HJ0?kb6c?&@Z<8HAl$JrCYx4xWTxgin%N$+HL<O-2Dk8=3Xc^mEKYaxPh_Y0q6P
zqE$SdD+4KHk-fxb>MZcJ5YYc3E=5|xRe*qNfyG!hy&WE|N`SI@7#`Vb*~><<cGFF(
z1wgTS(-lRHQe`9}Y-|o&T?+NadauPyw+puGm|5o#+@s)q;Q3J$_&lmf*|Tl@BLIk(
zT17FM!>J?BeCCl;0H+3HBX2GM7Lz;goGf6u-e`o0y;Scb2FYlYrVLnO2V8iyHJ}fl
z3bY2tS?7S3?GQ?c&;8(^MPB*cwB(&iKP~kR!8>Au8+$)oWM%#>NzzjCcj79~?rVhf
zPM^8%pXL2k$JCUjgWWU39zbt&`phU)P3g1`ZMT}u)^^!+DGK2-+wa!G7$NXGae&f{
zAoS8GhdqvX)-f$=dz>Xo1DK`*a0!B-x$h7EtLz`Sjj++EXRxt+_U!WJM%t5&AOpw-
zM~?J|(DDM`c68HA5G=P|^IQb#Y7;Q>+JL74PHLk=QE6=p0qs0I$=Y%%WJ5XkR5RtU
zW*)+tX_=N)IhEYhw^&a1TIhQUSbpfChhBR3l{alZ=LPqjxOOxD=tn==`gjA}_#}11
zi(d4i58ij*kG|)&+y47c|MXA)lptvyd>{TA-cD+yPY6u2sVr+!JDp}zT4rSkuQ_w(
z%zJ<Bt<U+<Ge7-{?|9&``qI+U4_6N0?VtSqJKpixzj)1S_TLEC?Emsdte%7*Y2(Y`
zY2*p=QSv{?mk5FJe}V574PqNF?mT%lE{k7bBu-f{k;F!sKSb_P_yP;lGS6xz(~SlD
zF0+_V=CdpoEt#92a9l=3UL@6`8qf7KX_04R8AmWNY8$)nn$q*4EM@o3E{P*q!XN8m
zX<RI-E5Yw}B2FpiR%>!97E;!tAhcot$99x55d`OeDAAVHCylz&j@RpjVXeO1=ybdd
zA;f6lx{9(gMrJ#)<Co0_1pfLB0i@%2akD89<zBBFP%xCzQM+BMQFuKDVHf}eQ4ow=
zt)y~2t)%jpX<8QC3QAFG7?f$X2D;hUG@>vG{erXYs8(}LP;NaoOr(~k6p&J-83z`}
zQ9x<G==UY7)u&#qPQmjcC0or9DFXn@q9_Cf3QTZ7(_+9_ungz{u*`M2DHY>zWo&B0
ziW9>yt+)lY?bu*Bw$lbF0i*;%C?VtmUW7f;B*(~Igg}y(Wg<WzrD`cJ%969;G6G;X
zj?ywmzvL)hS<Gie37G|J!b-dN-6V8MQBV7WlmR~oy+*_H0er7f@8a+D8ci<<z^ga9
zIKBCIRyEu27WQD!pHBJ%aNR~T%UV!(9dO-7E6bW#`*iAMYhciyO#1`4KMJ<tA!Blm
zw7?(_4|aH4pR-COJNv5ntjddNy2A&KBe{_1DBqb<Ya3j&8wJDg%J2VeCyW51uv0_;
zh8Y(xDk8&-i&>m7>u$ifd~fhR2%}CX3IRI+UKD#>oHE<+qfklQJ?@cZOZhk_2}$Si
zypK-{0!<Ve$w6#I(wr~n6E4j#(#RH{bPh)6&VB30&OO|A#ldssd*{xbJNIMXI(KgW
zTj##@_cmz0^PC>nV4SNvWK76tOu0?+A}_Mq3{?rjnn{{QK*d1eMK~&XDdBYvtgIeA
zy1D`mr9!WsnHILCB+^E@vO1ryuB3MZ0dn`d+u6~T6~vX5qgh*U)0k3do3>$c5w16y
z0L{kw)xZ0@R}(@g5eJvyGjN$)L0%1CB_y0?)3jVvlo~%BXPyv~8a(QNl1z|co#%E{
z7AaJEX_8hUo97+MaW$R>T=2P>m1vF>3;ZP>mr2fTKnL-nny_`-i(<^tiHmWeD(b7o
z<*XdTsLRu`Tm*=F<zzaU4)5uS+X1zqEu$1Hk%1d15Up_x$g;srSHTo+gh2q*9GREp
zcxu+;aXD}l07U_^5`e`@u*wa(8k!3qO=V5(cv61o5Y&ib`>;x)y_jdSbYuv?N9E#p
zZJ7c<z%k^eKsREiNzc-#9LM8mKpe&522Z0hDx>rTTEbH=!t$NhT=T>lMv><j8Y~w~
z)3T`4MqF!mG*U1v%cNAdJGI!*l3JE&f@^^`950Fx4?TFDTEMi8Q85(}Qyv0y2pJJ`
zCB{nID6IMllroqJ<cP^Sy<ZRH@{Gk-Ay;zVJ42j0PZ<d5AYyB+E4n|d_du3en*~>u
z$|PWB2JQiX2Z|yuz+G>BO<-z)R;2c}?-)g16hMkCCh-g*GG*nmU~QHm^}F%vFoc{T
zLLl#|cJ;=BcwsaL0Y1+?fyjGD2U8D&Q$pqOb9)eFWtv{xr<<FH4~p};D(j{(ME<EB
z(=>Hi)#t^7hc`EL{~}Gxj0pQD?;{ISm12ZEI#QE}hynypVx+7LEPW>g6aZ6}1pvn>
zT0^gB7`}f65o7ZjVD`++YD5cH*{ZQ27*PdQAq|eL8pcDQ;fB?&Y3c=xKmi!Q5ENoi
zFPg^H3eZDnMLtc#id4mfpsKd`Ycxif*^AV<H6kDeEY;3Z4v~_S9e|=NjYhK?1c@Z>
z>CfTE@N>z>C4Yw{{yi&cyN}aH4B_qAY`Z?CF{ZFs=W%gkm2CpX$GNr#ql1+4{<zD}
z4W34GKSfuv(>(QK`<SKX_tuwIrO?iA21x`mFdnh7Ye(*M@iA;t3yQDZndZ?%2hZ)}
zVV|)_Dh8W#+$QQUu%jq{Rb|?O!<%yKK1OBx9dP%BHqa)gSlEqBhlT`_e6|Oy6~`>q
z)WK|Ehs$cmHkj3G+Rkn&5pQ>M#+eJw({UPhm5(;t(3fz1Tz^(bD&J~fKbWib;W(ka
zaz0o)9t)-~+mCUN8^b#k#Uf<E@sHA+>M;d5$5gRR)3!Z3OF<x5W39CS=y-tAv$M8M
zQ_B^xa}Zgar=-NW9)ysEMNweK=F83Je(vWEM!^(9Fj*jSgknE`a*>>^b9M9n2O=`U
zM9fM=<j>C9+Bqvr=b3;4@Wg3mLI_MyGrW}35QJ+ZYL-g_sHy^Hxip3)s0$%wTC!B$
zwk;tq@*GG(CUe%3RjN{yHVbM)Ab4jPU~6H^hypq1nXI=)tJ!C{Beq24L`+~0l&P;Q
zoHA!<SAcU~*k!pV$_!^&3L^5(Q403<Ijcl4$1}=;)evk|7-EqWV3!32c_7d89Nf0n
zRU%6OKPUG=ppa!}tR|?gcg#d0qW0=C_0P$!>bfYGje-Ef%xuuat6nx`QP)@5bG^U3
z%DfOU@78LlH>Ft?2!#96J}{ohbeaXGPGkT<jOwk+D&L3mBG1a@60!kQ6hmkaRhMN}
z<mcsUE^mdDfI*}q4w*fzjm(25e<Y$rj*W0x8d$)>lqDM>Wl@p8gkqHAg@L+WE?uPr
z&7DiPc}*qY(llKGsv;q^V$}>PE0D^i%B_fiMTsw1WlB{^Dj_d&1M&gX67~d>7kQ8?
zPFWTJEV$eVz(#CdiMX}{Rt^*Xi_;*PG))~~O~;7KGX-homdm<^a*u<83_ex|IRg*+
zCCOVc07SJYaY}Z>ZphN<?KWFw{utGER^t$rT9{^=CpX&I4?%Qlbl_8-?|<GffUfU8
z`__5|HEN!1jIrm+B5mK@y`JRZ*(q>x65Fn=&iAmv4aa6XO}DOKEe^z@DolUUmj%-;
zYMQ2@zrfkqz5C}EPz~QvBB2CN|C>GqpP4L@*Cz=p)eS>c*e83G$sx}#l2s7U#T*%V
z>9LWJ!v$0GkN3{ZzWB%8YGwU;wcU0-(0AMCZMVM5KUgm|o8{t#cZTOY=gw_tRi15M
z+=qN~7v_1_*F@LX&--d>mif!i&cJhhrdKQc1Fo*_J(#~>Sg(N9dia7bc>c`|?%sXT
zyVk>S{i|Q#T}!yWe&GwRuaOTf{b%}Qd~*Ko3@>>bXDvrp?J>%F8(V(N(ER#+gFgAU
ze(Se>?rm~>!y7+<lauZC{HwoDUiK;S%Adw3KlvcO=*b7a<E8g*!neUlQJjC@7kt6z
ze|}Ow{r{&wO^jbme6mbVlU*_<&rbS>1%OlJYH}mFlRQA4LtaGQOx{C&j(mlXFfCG;
zQn2aC^MVVRmW!&8v#KaXT9x^n7iqEN39Cq^<)V;f5~p8Il%kMjQWe$KmKOE|k}s;F
z6f))IYmwY1BU36Xsoge=jI^rCq7ZcF@<NtLDpO5$RF&ICaR{2WlT@BvbVwv9kIS<!
zan1)FtDnD22r;pX{lAHHs)ER-H6dygo@rTDXCOxX`)BUjuhnWQh$7<~Sw{1m{$$*8
zx@lOp;xO$x`@ivR6#RG;MS-dr(f+Cx-ZYs^teT04j0qu2zFpRHqf+UK=MDw~-e~ni
zzw@H)a@=YHcA^_s%JIfVzwg*eciL-fC44GPQx!y!etqS~`Z*tTtjGO;<~j9)Z|&E9
zYwkke?;c|A=Y*&UQM<jS_jloB#Y1tJ_TDu|?VqtMYr-c&h~SH_?Uhw$pZD7Jxl$?>
zo~o$;;3wPdNx8i}zA<WW9{9uI=6F0ViX`!ANBtV^!54`^o<@i?EiIDiB!zS_;whxH
zU=PJtw5OA*k>kXfS{Toh>7aRt2lK&Vf^c>OR+@HO8-5aM@v;}k8|%}}G%+0p>bNe{
zTU(t_ipcpm%Mj*=N24S*DK&yH8y+d4-yhGOPFB}EiUips<``+ZIbGj~-va<}a(Qbz
z?;%yv^K7%<?+>gp40vn0vOman583wS#tLY2d8trpHp_-lTw#3|<w00lgMPnn+MXvR
zVlUs`TDHsd3Nj}r7$O=29RkV%sIN|7X)y*iuJfnU`It9U?FjGRe?7c-|MjkE7^Vv^
ze*PHtA&$EL@Z+s78Kyhmi2bj1qZqnz{QUp(5qw>*u6fVb$a4q@`~7}cC)&oEYUm7V
z&b(8R7g;9nXfqiNHw}~=ER8QwmgDJoIxZ(=O>}KS3b_T>?SB%kt9dSf>(%x@IkfkG
z%pE(7EZerC*tS{IlAXSx!_Y8?5XFOj6jQC;aAXS^CIFDpjz5}g5QYLSEsyK?XY8K}
z*8|&0QriL3P%HJ$HHqywc5=-%S(+y4aPA~Y@|<fqHvy7#I80K2z^`d_Wbh(fCWjPw
zx@lRJ(G;dV&G;AbB^IaR)wdT<FNVjk^~ggCY$u8w8yq`)Ua<5BOTqI(+nI1p>Y+zk
z2#%X1t_yGOJ9oyG8y7n&vK{!(A>!~q0ceUTA1>F5?{xZ!>i{@zLU56P=1;NM1psoL
z9XV@J?Lsq2+27Y~F--Zf`FuWOm?VhtmJvM|8Cx-8k|3MS9l%STf#X>C#PrUMjSWMn
z1Tl%}D*}Mv6*Nu|6D5od>DLfzzHARIPRZR$Rj0VY?pt3*Cf|MgN-}8&&mVCA@?JCP
zo7@Zn`1_sLi=@=WWp2@15Rb?`KYug^&g-A)3vFJO^g}tB7nA9v8aOZL{02w|M<Pm6
zEVM3gF5hY)qj+4l2|v|<Zk#Ny9Nu12YIBfyzHr>cb#32K3RFECEb}nzwSzQNaTGYA
z7isFBU0+#kwE$Wz+lr9#z^{V>?K*zoc7lMm%kmL7QyW#}@O~7uYPQ<lV;Xk}Db_r4
z#IOOzaXBS?);QudI4Z%uUBi9zIsAz2dxjBp9s36N)GPP;to_UGuivkCUNd+lBH|ga
zolfhTcOI8@?C{^{0hyCq%6tb33Z1wZFgXxnkP0a>&Z=7o6)&^`ue7KZ(<Dt~443vV
z!bK&T5R_?NRDXT^`0*11&ooWb^9>WgG<;|rKYknpuh*4?6MG;$uU>i{!X<6k`xk9P
zYr}?%wxRbge&ZYeY?_{Lnx^S{ruokgMS+jtdC4Q+XxoMft@hdrh$iIVDY$&^tvp6v
zL0-!^viwo<D?~f+lnl6}hGKUER$*ZM+IcnzTyLjI8p<Bv{G&`*A<{URCKbiSkvK}D
z%qAkQpG*#_Vm6(`ksK6qP%NrRDl*h>I~Ij!?+oMskilPXRKVq{b6@;A80|Dc@Bl$r
zo`+GKMCf^z07MUxq%Em(9ZPBjCS#~LH<ARwnshq=>+5GuZ*J-rRj}Z8$|wa*0VoBa
z4g|GLZ4CBoQ3ds}BGe{A7_g{FvhD`fx&KwF!+I*wvN!4uvn(IA2F^ii^eu}y^$p8L
z;jzGkX{20hh|L01NJbG<u(G;!dVL*Wef{*->Pnz0ikpK78V1q;@RdOdhz8`N2Y?zk
ztclV@Y51(u?<0Y?*MUcqxGcxTj7tpYaWb8BFj>&ZH>nNH-x~(M3)$er@q2GOW7$`3
z4I2$Z8?HS{Q-~XD6r5tKI*!optZjCNUAy0HCtX;Q;*4rkeH#F45Jg>}s?{pc!tQpv
zHg30>Wvz8O06iIuVrKVJJs<sIOlD)JH)-50^oqTarN|>Kszp^IQLKAdoG)(t_)hm8
zJ&KnF?cT*5Cr+dv`;&aqIej|ql4%to2Y&?Lhffhje9|C4N?uRiLEcL~Oh_fN0qH{l
zjDCWMlj&lfXX0RJV}r3;2hMCU(6N(j9N$qfdk%-Tjzi*L5`=eTIZ2cFFdP%bQ7kH}
zz(QV741BYQ1{&#DGdcblVD@F(1x{Dvu*QLN;K!eF@+88>#?=9!?RlRmWucHh*tQwO
ziN&n7ejf$55K#xUI(GWKu5Edqr4gwmC3t?L(e?m>t2WjF`h%_Q)9*q`5h+shjw++r
zN6IKrhE9qeBBfItwW9sx#X}JoMM~jE@#M)zZrs~zfX&P_Mc^Vq$5tQ(0B1o{)C`Qm
z(~)gkA_yAPv8mRel#sL$1j4fIXq+ZWp_z@^Hn*X%w|C>|s_J!N|7Ve|-lGbb>mp#*
z+ZaXBK}G=?L!CAK>bY(_Hy$2NE3u@?_rO3g$P7r95X4d%tFdt)>B4+NZyK*>D-gc2
zY?`<0f$u}VzqNHL&y?1T3IW1%qwOpMSY17S-O93U*W^Ll9Z)HFr#;_jw*Zo)ADSj(
z+yLl>wpEmk6hca5fM-@$VbT4kwzm3xxcsz!(++&!OcG`(1#7Fvj}}XSYpc8MX3L_2
znTj)NxSqG#?IE<9>(fx0-Pxh%wc{T3y{4}XfJUR;w!A2ETSs=UTL;%S*cPzQ>RlD8
zoq&=7_Io=Ea5l#5QyyX-wo`tT;^BNDr;Uemz{S{?n$A&VTnoi01VI=87@RTF3<hxm
z&}wbW+ifmnoM$xvncB|E^0miTRuHmmJ9J$M<2$UB`m2(DpuzN6Tk3gnJgN=A(0=AR
zLN?;I=Y^IHo$h44*#zu%S3TD?C{w0DNv&41wY$2CP%Irewz`U-ltq)&bY0JeAZWr~
zGYEj#6d&|)a(f*fa^3QIS>oJ=rbeoQPn)q{g|d{*mj@A-SzSBj{zYiDR+k%1v)u@M
z06%E7Z5y}mLZ~;^8i+n?KWv~%)Jtv1zIww(@YQ}V>$=uJ>a}JApw(Ip+Mnr*|NXED
z!AIdDDaalnqb=v10u<jJS+Ix8={%mJ1{zb^=u6DY3|>W{oZO&|dh9imnGuF9e8)9y
z)$8<GzxL^^s@mGx`nD6?Fr9(!(beI2Vi+0*gV}5_0BpC?6LRYTw%!s#;EV9vM37Y3
zf~uEdVJTeETSI&CtM~`1e-pm{&O7hCb0c+KfJU>~0C3$D-gxJoccM~*%rxQ7+dwDj
z;AEO0fz@Sj-L&gDj@M0H7xq8?&St%ikfxn?!Rm78xPZmdk&h6<4h{|u;LqS93CIcZ
z5P2CPRV2$&$XlrukM6JbAR7dO_lmrlRreYQ2RhS9+P1x59LYq=TAA_Qg<G{*O%0<A
zzrq9)JpNw(8LgsFE2YCoX~;2*vf;{dPLZ*Eb#*j~Vpx`x$w9qdUt3=;>h*eGGNzVb
z4}x%*Btj%<nh2hxSs142U@#mH2U(KBg-~m)!%%Cj!x#BMfEWb69|izH=uao>BhT{#
zOz(N4^~qjYt9z#DC>0qxlFBhnubxi!rh{5768Q&#QeXhJ+F&(FqykVfNmeTeLVx+l
zsIZ@-hL@26uOZ1deO7|!!t%L9{QMIwCH(%q_uf0BN_pksdv`7zzW4AtJhnR;T4Vmt
zo}u^u4wC)9gXA7q+c^Ba7|CiE;Sw>(CLv)+!?di*u<Ua_FP#P(Y6;*Yg&G+UNT~MD
zlq6Y}J@G`ge-UP%`qZcRC+z#+X<|$lbMdp*KXSOV{}{Y^|FPK&-h8oqce8m_?z%_b
zIqt*7Pr;=;5fd6x8NuYHN+<WL%KqCugs1lQ_P+Op=)8LDwc-A+cyP73|I0r71YFvG
zJ3M#)?bls*_ww4>z31oWwaj;92bbYT;W9Z(?jes6GBO9#3>6LJbFcs$lf~s^y3j<<
z6@yHclj(%(QRkUVkU5?;Y@8Pb#57E06wGzqo5%BbzO2H8FS4*$XRO!VEW@xgT(@ca
zjMZ(2qG80G8+(C|WfYXV`AE_*%s#+)P^$$YKp58QKBrv7t#+IMq)DqCbN2huk(D4|
z*Io_RTnjJ*O8q~vD8iwWCr+HiB;MLcYx0=@h=y6O8wMiW^IwlV@(3WBR<miDh(HJu
zmk=!BA~{5k5dwLU7kM$raPWYY3`cJ9qm${loB&s-m4i~g`aGVerNrDC)h3hGRZ3S^
zCzIO9;;ygzMESHw*#F|&-UcUMdxf;qt)+lzt<z0M=7*`wQQGa)QiQbD?WEuM@P|Kq
z{ds~1KRkE}UR_^xri5h705o>4s>(@LlsV1$>a>!TEKB-wku5};2QM)BVXXO0WgN5F
zDS$QZ95dzmx8MHux8Kg~W0trR{r;WOIBv85M_1dWh1N^2ve3TzBR~A%4}aD|>*r~E
z=OY$cgb+%|LEZURev@2rTwYb>xZFZ+_qnTbE9Lu{$g1*Ff}hf~>@s6R@g9!3+W7Es
zsy7usPInDc8(eN``t(}zX*~t{=7k1b(HnLDMH=*r`6EH(`5N@E>Hp`@gG<CCbzc-%
zYk6npi?S+HTFHt-QY-d9t<(@O^1*_7{9znib>odU-f+W>H{5W;4f}s{<A~dyi}8A6
z_DC0|2T#E(;F7L}ExC)kvP`R|2c=RnT~`sZcGG^pe`jmIGJc+AU;gr!)t7hP``-7~
z-@||X*MHsnwO+5+dr|McuYUEbUoH0k7SjE{U4o}Cy<vHIxmf;@ix)3;FX|g_ym52`
zShZU1QtjuKmR@}65+Q^Wa?mcy24qAg<aUMv+To4TA-t+VeH^qI5kqFoIX<1iS9B2k
z5sOrqFjTM$WG2%^l~kc@o%yT)i$JW1m82F5L1>QR;%KDIDS)%#PQ9cQ@~N`d2k7_8
z=e`$5{D-}dGyO4ItD};ZsU%!ADi+abI2?tGr4b;)sHnoxa5#z<#Ry<*ccQ_drKQxZ
zK|k);*r_AZ?!^5mKXVX)@9usQrt5yEve$>teZ9I~9K)CDh;>s*smxYYi~u&m52G<3
z<p2PqqKbyY;VAmkb)E%s1J5yteBlqcMvnFY1_BOFLr>LdrlU`Sx9?wqd-g98P5v|(
zanSMufWU7xpP#e&fsr0AQB$rr|D)XhQwaR#*MS8}?hWh{_5{hu#P{0C%1U8Z8dU3b
z<#<ofC;Sr9a{7Z~$Bw<~*bn+)1AVfO%>D0r*Sp@;8T{O!1GW3^yYFeYzKmVCaTmV#
zz3+W*HX}&H!S~=l^A*uf{s(;`j^eQN5|oAIq8Ma-$fj9SpzcI)ev<ic+NLMLLD*l1
z&O;A9)Ow;0mpGGv`}g|*7JSfU```GnAItWurw%T|A-~Bm-%&V=>>%FC6p1!f^=rnZ
zzHcW@D#nu+u7rhIZwz;OJ<HnEU~Yr~f_p~81r4BY@Pob-923TRmZ1S^wO+sZcH^YU
z{XxI)fqoS9c@3{=n5IPm#4#lZAxOwUpXiLwm+ZW<PFHR9ff%Y8>wbJNLW+*iry`%P
z!Vo$fWZSd=^qQCnF*D7WF$xZ%^t{9%xJQ?<({A1-RwXs#I5L2r1wP|&!IM%U09)cr
zN`XVZB0G2rUJjSY202E^D6br>MgtJ-?4gtc4EqJcksp_;Bh5t-nhnzN_hvW>)Ap9{
z`?jY{zgAxxWavoGOOm?h1@)TmMYT9~U9qk;ZikFw5C(P=i_Wi9oh1&Oa{(X?gns`-
zu@ZLLX$l}UfKq^Q(rktGZby8i9)%hV!|AL*MhIbq9Q0dhFUw|xgq5sp<*}a@R=if6
z4p?84X{nUtQx8Au_x@!XxUbP@{PIukf40$ReDcRzgTdgBei>f=tH1iI-xy$RGGq_;
z;59Zm-A10!5ft_Z;!!2zIds6v(ap1gRJZAfw8Tymcpt$}`*7yac>mvGS_cH<@t7hN
zf|fgbdpji+f+8Is%?pWpvdE9V(}`4k`_$iY;lc&ZIlr;lypeOxFJwE1uDtTlPDTu(
z4?YLK%E>a}+sVhsm&sp40FT0Z;R}R>(X=e>uqER%mxDN&PNo4<W5?B`oXi(lF2|*a
zlc}DoL0b|=;l}k5$H>VSb0MW22y;XqtraJ+H-Q^JnWV`ypAH$fvZxjX4fg8FycjpI
zeU^uv$9Yv`5|l_G7m(`yD2>X6%uz~OYU$9H6Ig8um8Pv~X;FY)D$7N|e{a^#`w(nm
z4B7!R0&)}Ki?D-cmM}pl42+4>NqR7d5{1a?1qRtTTNJ8n8g(cNOqWB&cDM;d#q=1G
zM8a4^r<tTU)1d+yFF@Bvn&*W`zXQnIHY6|(;hOe4c;sq`mZxmn0^86)ndjxJzVBHW
zpE$m6Xv#UILIaqJ3afS{grkfk?A8ON4a+xeCqSpsNZnXT3&8uba%`<^Gys883K!W2
zQXpSeLICGLDHkA6S(G{sr4m7grppDF4r77fhA)`ljN6EUv9wlClDcGC)vtBjKva@3
z%7kedkuVJ@Q7DU3)3j}fIAcsH$!*2C3pQnjVIZZ9A&nAB1#_h|4TDR`xX`9)NM;xY
z=gc-tN&$G&Gfeau1K{;ew+%?+I7pJj;sTKaP-XxGhG_}_OG^al!T|xwjKFs^40;3t
z3U2wH<{_^hPMrw4Xo%0A5@G*ur3JP|09YBNVd#MC`w7x|yB+5pGhw5dShC;q>y37^
z!6m00fm&$N*i<p9F^kIM%matZ>p({9hz(CMCM`37UC*m|zyQ!t8l*HBn5L->2dNYi
z4MPA3I}C7_A_p)i7!-kW+Y-n$QqHJUitqnYouXxeAtS>?u#`ps6r3}r5uKrFTc%|+
zObv$NI$GPl9|W#vvbtSM(_TmFxY4K~MTY3uh!nX|Yqosl`w6ycH7&L0hC#r!vK>nz
zx+?Ng%Y%Tiz_OGIQ^eR0rEu-kb44OiaEVgdmf=fiQKkVj7aS1qr5S~w8NKj-6cL0-
zq7J?aKY&k>7Fi}+<P5oqJU|{}$QRlPww{Eo2aT!2X_6NGc(kn&Ih!XpI<qkvN)t2!
z{f@LDjnVClNE_;NP9NDoxw={@<50UrYuIRRHPci~CzIl@c$zk^sw!=CyX!ACjvhU)
z)uHjYF~{@%>gut?FmblMHU0q@I-ZV3Qr2oi_*A2=w&!nMg`A5hX*BAxCTg{0oDHO9
zTUlQW2IG)1nB+zLc?aOfWfBDb($e~&8-6w$jnbXfRW}iGq|N2!LsAO5HX7Cm7ggLN
z_cF8)63VuLOT4qKN!0^V>A6OC#~l9oyui8#?Bxc4x|nBq1A|k3$MEwE4;{KWN!rD@
z+XF~O{mjkYzp{R5k_|vgH?*{EC(8(i5g=%_+G+##2TQB{K7v%PZz#hIm(aGGO7&YU
z#QtFQU2%Hzp+m?LfL?c8v_CwTz0IP1dl<AVt<lq30oE<S86YapTU`qS*R9)z23raN
zsm+A|6uzIY#qqJa>u<s<;5W%Kxw+?YFb-mTR>^U>NGEn!o#^MarL1U=fw4tW5>jh0
zARILil6aWSv`i$N*#H0F@BjanX@U@(8-_aCY6%IH!iBn42xIS%XIU1tpAiCtx9`Su
zi~+jm3@!xv-XR3{TQ0zFg6lc^*SOg5KP=U`USBF;0Jd$xKkva6hYue%5kfmeE~HLk
zX;7q!#?fg2xZqIqaK~|%%vlPhzqLFrXR*;}u-9ecDal^V3@oC!qV_9JW%TxY7{~DX
z{ZGPm(5!0**Q<RA>laKV0i-f7SW<mKN$dKN-Rmu>P+b}7dS}o4G)k<@b8B#2&Glga
z7ka-}+&7g{%G`ghC6$uaTbF3xdfl`Cwb&wyUj9Z(l%ukeMd?^&?8Rf0CNh#gC=$NM
zeDrzGd*1UDKrIde(+&dL2!l8Ub#qt#v-iB`J?}A+B(Max9k6XKtRP8@yk;(SaPU3&
z0(_9HkQ=slK8=C0#x@=;3Wq>DFLKeurc4uxA&i841;*da%#ENJ*emj)n>cV^<|JKR
zy)w9Rbu}INDd75aI>jWejngPnW8wStRurkZjN_KLdU*xmmGhf5O!e?yr^8=)1&!l&
zyIymhdac_G0(Jt*@)kXhs{&N;vm_uDxq{qF9wV<NZzlN<FbV~jj&@$ra>(@NF)5u+
zs!2H{_ZY(|O{^j8xe*nOQbq>S!QC!$iNd7wTwZ`kKg<q8aI?y1`KcxR1h=ffr_?Lb
znjsyBUS5=Cw+nd1bbUQdZRDK7kt_?Cb%OKVBTs+3*w{FJY-ySD&5dKnmY2{B84E}4
zj_sJ#^OPF~E6cNMoo=_^HBH-Tx4S*Zwr?^_Lt3$6n$nE>S}>&rQ{aTQO~G-5Xf|?}
zac}>heXrYDJ$dEC_g#0aux?vZ3)AUzb&dYmO06zrzrVUtuL;@juY#7IkHYi1-C-zr
zFkM|uYsUWXbZpx&8ueDwww12eIvvKSkywl}Be86hno*@0p+p^g6@CD}LTu6?%Xbe?
z#rbiV8&@Qnl?eP7k&xejrRC#Wh}+w@Egze&_PUnU>#og@fByLL65O`EjkpEr4abi6
zPuzLei4)%Y_kRi6tzv0BUMgB`xVeA)*bVRZPMo;w&J%<XAmre8;UfGRxr30AkYy=~
zvdoLJ%zLzS=R(_7>V5#*T3J4(5o;gASw;&uuK;)j!a^tawH+?T-EOxVN6;N_?pRj2
zwA_r_opvW#42MaqT}cOnVWxHlea8j>OcT@W9yTpQBY1A7v%v?0exFJkYPF(Rf8E6M
z7&9~+8D;>cfzWPGH9!<D62ZeT7_I_D(M}vittNHC03k_6qm4ArJ6%M}R2W5$>l{lG
zz~vDQLMaD?k>KDm{32W?3F(q0a)^+s-|u5_U8aHpsR5naPZ#@TJm5x^S&%Nx&P4$4
z-@hN;9|o=P=1;bQa8&Va_V)e1g2w(|P5!|OgVtk7JB)6-EehL7dgG1HzVXK2hA?a=
zNjnUojq#xC<g~?HI9}bj4MNqT=B&KQ<z|d>30>I$c<WP!R~}z)g@W6LVYApZG(ZxD
z_DDOn-HZS|*Q*_$P&{f>d4|4m!!!<H+G>IuY6TFvF0ksJC|vJ(HHz{Y3P(52<91j#
zUdbl~*(TSLd&!H)8^{NIq4U`w?JXPgI&QAAep{T+D*&KOZDld*0717X-s0&rU4m&=
zt&&+e&!YRz)Tj7nWhQ<S2Gc9%<#gUyEAZ671M&NLmt#kfZ3EbLv{@UmI<lS3$acnt
zo0A$Z92ox18Scip|44kzvN!y3o{0WKtjTrR0<>R4KIHG@0k*u+3J7L%IzfZ!B=e<s
zTuxBjOuqmS*=RZevJXy&z<OI&(#{eURg5dcsy1%RJv+<GtaW-aKj}O2zyl8~dVPQZ
z&Qlk_N5k*{VCTlIjm@?aK@g$>MdPbC`w>;3Aq;{b2tk9Q@l^?c4R7At+iQ161At@y
z1HeBvZnpsHyU_1%ZM8vOx(U7RVt)Mie9`T}uuUDvl^fH^-saL(d4!mzo@p)+>Wwvk
zaJtN2zxj3mU~l8d9~OUkhS9v%`Shnhtu2IKd5K|l*FW1Dj}LEcJlFtYBkEM^$99>1
zO7Ht{8Qxy=L?g+Ng<oNL2s3_Jm3}2~#;WJVSd}SwCr_Tdb!!`OKEI(F3;^=c&dFbT
z^{Zb!>MWg}WZLWXdU|PTd$ClQt46aqDq1ZCH}v-}zNOV_#erPao$y{l2%*G1cnba*
zE)j<`$$*ehR<bCwGA(5}EyGkukqRYri#)5+d3?5>+TPyYy62JNuD4jZXE$0|2EJcA
ze&?NcK6EEc-}bh*z3o44gA1?6LDp*6US9M4fBoL~zV|)6*l+zC_!QY#=R$HasJU?E
z7(C7+z~A8Bv1c_&e=r_6c4G6;p|w?tt82%~PRDiVTmFH4=9$Nj<JWRLdTfQ!<(1PX
z*4EG`{m}5HpvJ2)azO4T&n2%U?;@l!%$c%{ZZIjk7$9xul3+$u4Hr7^hpogVPZJ}I
zkPw3b{U-B@3=}H-G);O<vxXoF#9?`42%^3hNiqrhx0l~C7?WidMQ~SOS(X)emPwnw
z58EqlT=#vzJ{^zGoE?v855f2AvAeo$Ss^n`?-y*PSku-Dw6;G7TH8$uO4+v7Fjl^o
zB%ZH;x4PXHhs$TT4`oF*R%umc8=Lb7!LowDvXprK)rN&Zi#99(->>K1BFhl7Y~kf~
z-v_W7G;(a}XnmJyd0mjo^Gq}Ml!UHtnx^4zho%W&njuF-<iu~ObshZ&|7`NIxSepK
zb+=V^V(t{b;aTvF)|11jD1^_G&IFF=lY6kYe;FOW4u<)&)qnB~(=@!cN?jKwLnlaF
z7us}r<>=9sW!eVUO#){$F-%Xjf5McKrga3R%%6dJjY?@+O2Ue}RCvJZ%~p*8T!J8I
z#un_W*P4ds878)<=hrYb09+^SJDc4OVyC<5^i#(LFjDj_i#G9k!*H7jO;>9eGyz+V
zVHnyiGsKL6cFre++M>QTXUKy#sKr>w%H9?ou)|5Zs48rq#$NXMd0A2L#fxcKeSZw6
zWfVtqt5eHEzw0#|0?!AClWsSOVUVKta~|2pG97^L!J(dK-!Fsrj9%or?cdn&0K8|V
zt9WAEO${UV;AM{8G2Knm*z^FrabGGdlSdE&{|*;Pi(IKAUD)Tst{`DCMsYHWx=6^w
zV~{!K=oPcodPd;iq1W3S_j)jxVa<~2ra#THY>K_^=4P*lSgYerulgrmNz0V#*H525
zz1iylbbFhpJ?~77f>h?&L!Rfk!_7_?pxfE_a1F6`>b3TgePDB!i`~xJsD54=;CVhJ
z@6d%T&%A+wL8}<p0n-N|<Y@N!aw=OKBU0gW(56R1xQZ(wRe;J80`7?%2fE;X^731K
zIfbeqm}&Z_gSn~@mO@k%94CT%gxG(V3zt9Xswv?8LAW}&hgc*gIoTw8<QzFqZYLxZ
zuLT6E8E4th=*TT*lA@WPmr2MqW65?D%oorfg#`~%7!OoDfByXW*G#+JZWor%pMTA3
z&Ud@r?mtZ7*XGM+4?pM-ed52adhMNezV;KJ_(W^@UtW9Xoy)D|fBD2~@4WK`)9F>8
z_(T~Nc|^;%`!FG4JGVV3=I!6(;&dS662bc}v&#4V(v4wWf>&if!evRd#K|;G{L>bo
zCo`eqqFT&<Uu)Zx(zLann?jhjR!ocR&mI5uocs1aZkbQHqF&&y&n)k!;<(GXu#?Wr
zc6)IQzGoW1PBPQ<POy^<OvCdH^FttmXY&(~r2Rpffc^tx%Ip0!fzVCy{l=6-k`9O>
z^5E0B4VQ@|HF706PyUvCpZq&$LO3I%G|h3Qla@&dWNES=f`FEmXSAru&PaGRKQE=!
zROVLECo373bI63x=9w)-i;9pjo6CVrWa=q&NxT4gO_1!7owEnDME3jyAw?=n5ffob
z6CvPLVPgTb)7uu)^a^>iEl%c>c?w19VHwZkXz-;XO^c=9if8kDE~JqE!a3)h!AHD$
z-I4P-pwL<r(gMnn3p5=e9$YsO8FL-GS8hw6Qz3ZJ2~?}sReuZsJGCs^GNn@B2+9rJ
zk0arS2)>pI!q8*3ZJIFGU^2%sIZ}gCL}W@x@n#LC==J&o0|1#5+NLGLI5N4So`nFl
zn&To!kiy}y-w!}Xt|Ph9zUPnnh>`ERU$*EqwOTFdok2w8P7napTy6eOiU9nw5Nh3s
zkU72&U{Z&txCSni0D$zaFs4$p{Gc|75=(F)>W=U%pufa{n(n`PJ<Bxh>T;!{Ln(l@
zW5Kxvj9FeBqiwm4U>4OHD5YTXI+X%rMi;dKQYt{fwA5<(j$_-d0?Rdoc5J}~N?EV#
z)G*9AqS^pQ!_7gb?Mmr{2!S8eHk!>Qt^N4!?yfWafMr>LlyX&xG73l-=^R{!54A6k
zo+Y=F*C5epXl7J{i&#~K6wH-Zay%$<GI5u4c#%#`Qv~3$EXy=VB1~kYfWBiyE9YeQ
zZZR+B)Q@O^P8&xO#;qh#DoI)^S%!=Gh4Wim46|(I7qu+fZ!Hu76h%<3Vd+&$*RNX%
z-G*sfZZ|@(!l?Blz=2_q1UrEb6wwSkKS0R4O#wGFnvHg=*=R16<%tVhTWqm7&-jV5
zT>74ABQQo8Mp1;#)Uov%qmy^yXsKu0wm+yL_2Xs~KEeQ4BMo=`5cr*5y6Fd@4>(*M
zZA2kA;IS*@7V;YMYgtrL>&>T9=vzhV9CEw!5uu@Rnij<fKB1jQq6Y)Je#k%ynUIX*
zDXGe%p$&|M9FRA(lB%*#XGLBVJQ4DUNO8@HB}}rReo~?2g2yjk!1wE^=d;Pin&-NP
zVcfr#+cu|CxRDStvh7!{t*yy8NfKeYu4#a?w(h5n>-+VZ=dnhk5!v-xsw5av$`AoL
zkHYo!`%TkuU3V>Rwi?{?l7{Dl>!dfVt*sTd#hDb2={lAH?%J9Y**5rIBk^2buh%2Z
z1(U8fg8Lhu4~~=AtLrWprt6ybujRJInG|l+Xf~MVB`;W8TN6o=B*HWe*LBy+8T0&f
zo!72}jO<!HRq`k6^?HbioQKi++Wk9}O{7$@ZFd_koBAG|udlBcwk4>P_CmXxGw$xh
zINOeekf9cw3(p&7>AoW|aSvWd0@5d&WQ*LW^l0ba0_4eu)3Z`cv+3e?S;)$H^8n>`
zGaHR@acx?-Igu=CHp`1?xtNDce?iFpW(>Xl=Ffh7G92Nt_ntd<E^$5Ej<3CN;lhO*
z7YLqLho`^&?QehkR<F0aGt3~%cJF8%5uWcKfltF{#C*mc{onC?7KE_>M7_2+6+Zx4
z)l$Q_@OSs&ykQ#um}Q?lv{WGE`OfZe1ULIi<DQV+IpUm1(yoX#Nxxs@IJuF$n2;)|
zGRZ1r2N35)o}$g$m={9k2+)v?C4{`JlPe6EoRc_;$B%JVMrBrWT3VfdgsK+R``Q%S
z8|!>&X=$~Fpp^B>qeqX@R+~PorSlW#;F@c$x#s;DL{an3+v|Z3jTiO<|Nep2nhv~4
zKMVlEu=gr#wqW+k_0;C<$~HjzZ#qVm-V=s9R?&9s%ZAZv8U@5jPu*wK065c|5QY67
z2SkI#sy*@+IZqxSFOFD)S1cfPSgguPCq5Q3&Y+I4W>(~fd65>@+od9xX%<3@qNog%
zBA9!g$mXp#6QE@x*UE1`wwK<`!?4kbq8qaI+o{nkmYNONsYS0bPMv(9tF){Ak9WIW
z9mG+nwbD@(_FTcabRG8>n=wXVGu9}u7Opt9V@JVj<FTAvThu%+j=P7Ojpe0=F7Am-
za1+?oeS3Qrle#WimK8{?R2)T-R>~Hx<GMnKR|S5v5d=ZdX!?P>2D7FIv~A#PJ#r(t
zk31)jK!nPiU6jX57>|fTOye>w7k#NBeAQ=iZj!}siqYqix+!JBJJVTpYyW^0`TS=b
zE`)Gwr>%l82$fPQ2!jqZy3;=ZF9?z?sMSZKdL5uq&##3rY}NgM-+E1r{R3#VJ1u}#
zr`-ZKP9Cm3ev@!)*Wp4~zEnzvVHjvBt*=`F`rk3O@p>4sB>n!3@_K!%6NUhen~v_i
zW+%D7AGmJN@B5zT_v^N0+4W~b<ofKi$h5b>Ui1cXC1{3&R%}OPc`RTziG{oX8`)$8
z$+8cM7Z|eW!ZrFJlT^5TKUesG;G6UL0zfbqtRuUJ7Y5o*$pFSj_l_bm7N8f^((e6N
zo9K6?*uL)CV*p$vfbRTQwV1D3K%>5TcK^p%lQWBT@na7i#=FdR;5dbr);<6P9v6(R
zOFMw8SWl_|F^;KcSvBD&kb79X2bc3aMP?dIiA{7*!8M?jrkKtdryy5TZ#gYv*t_q(
z`{eB(wAUzyJMWY?z28}-9PWds{mQTWig8ijfCBFPRsEyJKovp=aR?!Eya-nkpA5*9
zuWaWB$&1L_$uE&Fl5dfJAU_00)hI8^C>zX+vLeMksT#zxOv{LbILt;uH1}N=WuD28
zuA(S5uNo?5yKR0iv+OfV7>$!F!|a?G6*Ay^l`LgBiy36YX;OPjMVS{{@Y52>@sVNT
z14%8r(Wd?4bTR)~g$VeSAD~DFBjfBjf|sc$1Y=2;CgmJhy?cY0>gj(1IY$J}0c!|C
z03t{USf@fzDkUArGb!0WAu?3Vg3^mNgQgQe3KU4IC<Xc)vEXSqaCoujh$63AsRH%Q
zL6QI@$zcEgfKIAN$ry-*U<|kb#y~X3a`?6L^08ybPBC`s7zLo~{NYm!fSr1n*Z8YW
za(?nv{2rQRlx7)C0Jpa&Zf_%)^yG0$kDsJSPoHG$<Z1dG007G1mwG*=_I{b83xFIi
zp=oSVq)4ee^kI&vc=JilPrg|k0a0Z`Fi`%7LP(N9KS|**$Tu950P@%yBq$K?I3WPU
ziFXKK@+}|yTjag4`n&KdcpsTtt{4O*tyj9rP^lT`9A4oEg}?yv;eFx5#a6A8=V6q(
zy<UG1T<LVX`D#Cp<7T7WN@7Htrl~6dnr+0<Xz_*FY$kSAS54cI!{OD}y!r*FPp_{X
zjq;+L^m>5({_08#A&%M`lf%uX#S4&r772WrQ^c*ZS?8`MHxL3vnl4h1%m;*#;(0lv
zOM`7O$>Mn&T&HrZpQnD2rL>Hzf!`$4c|7N*Q8E16-~HX+S=Qef#=o&oef;AezxwXA
zyRZKE=c=lzKKIWj*F1#Dls$yr```b5rQR>B_nY$lO1)p2@3+Lgk3Rb7^yt%{{`99G
zoj&^h(P%Wf;;(-Ax6D3mddp<yAiX0YxCf7uF}ag`gpg5A0EXh5w^^J^;dc|9G1Y|G
zMPTj>ktoRh5lX{JEb+zi05=7}9jr5W(z$eLr4ULY7lfIy(0-ANg$w*Ic_bW2KoKr2
zmy_{w4T0iDQ3&2KP19_*M%lnL2ZKBASn9O3cD+H|T06X3uhC+uSfVFdj-lHQlY>Ey
z$Qk4D(quLp4JnwbhS|P_he5p_da>uZR<~VKvR-RaWK43woj6Vt+XftDy}XttVHjGr
zX{NOlkupsg?M-Wz$^JK$f=9u$-R7*3E&~9Bt@fC+wBH|QNrC`SON&vzA3c!)0Knm1
zuNO&gDNQ3ls+N+5)<tpD<jKlPeqGIV0em;{T~{ZKw9zoFm{LX=r8<&I$S`PRag=o1
z4cFp4Y$S=05hA0UbH48DIB5_d*1=cdpW##FxrFe%D9gMk%e-1ti)vO*Vj+bTX;Mw9
zMKvxLMPB4ZUgX)htjZ)!;&C}E@}dITOCQyaVfTyixP<K3IjkDC+n@@sC?vp{YOB`;
z>~ty+Ny7O?J-e=3Q~)RyXg|PqXjz5<YikZ;V7eZ(W6QQFjr=Hb-JEgTc3ohIurwSp
zIDX=pTQJDBw@)9Q&4vJMIGju;DQJz2b~}n8G|ZmiIFN<`Oebj=VqhB3(~gVjXw-=j
z3=PA)7-Vt6puVww>vK+?M4|~f_-cHR`WS<<tI56ON67ogFOk2^)7X%y`MgNMYXq%i
zl^3(BC<zwHt!UF%P{}SgF9$_M(;{KjXfYM<UP;SE1yqia$Y5f=C^Od>X!GA)?rSYr
zBURD@Xlj}SY?P{ZT8tp0Bc3C7GcP13B(fD%N$-e-JeTRBDxl~4JL7T3(`K-Ag#Ue-
zrW%Y?D6av~@RYc8WGOJU*BOs@e4oEK2!f>}%$v`{Vq}9a=#1`khK#VF+ndjiAD_>Y
z<qw80QFxxGL~4Lu^L6<jb#EHzJqkegbTIYmw(U5!U3X`J-lKF+2Qzp4vj=aaA#JtB
z+OfS2ySD!?Qp#4h+mgOxl!p$L+V!7nZ+Nz&H`^^5(i&{LuDfp6W?kc2`~T#5-p0Sj
zsJ*Ej`*rQmQ*1tWtM3_xVL1M(r}A9AxpS!eWge_WjnuR(D``Y)0Y5?MiSct1%d*Tg
zMF2dJvqoT}KM-Y)ljoARlh0Y9ld>EX7lC*UKX}JwCE~Ia>0&-B`qTMAv6xO$kxBQ0
zlYF&3QOI(9&&E<nx&U9zloDAccha<=#flfx3F47?y^<g`5z{KI4%SxlayFl3gUlh5
zMsZY)v(;cOm^voSPm}5;oChL(G<lxqq5$qd5x-KYszbe&9`wld&5ffO03>&nWtl1f
z$MplPK#G+x0yuS?Iu;}lA#Jykm)N#_j!_HIasmAWI0ycSlrl)y&|sp6HGp9bg14Rv
zAe5$*GD85u3o!~Q4I})FbF=0*r%95g3247&7=Si_khOJyu4@Xz3XN8-q<AJGMiGcB
zEv==|!`w)cL<6iGSG~$Fxypz7E{N=_^Wxy(;GdAdTcauauOrVQFC-uKAF<N&Nh)Ex
zl8RKsN(LwtTm?LB=SAg4w#A7;d0-`7g+T~-DcT&y<nLw+;d$$&WnAP}gqNwF!rJ+~
zoRy0#mjEZ6#KknKz6mA7S91U7+H0?kyPaAJ*oQzHfMEi)TJ<c1Zuf*~*>0<qrik|F
zDApVGI)V{6Bin{)Lo2Ntb*;2&yhbt$DWwLN{030b(ILQa2rzuAJsNs0aGo|To8~~b
zcahq52L>E52$nWn-!#lX`}LY_!`Z?x5FloNu9vrL`iGw2oVk_(fRyUjqVM}YfbaW0
z-rwu>01)|$I3feu|9M4w{{q7-8^T|>=(*V=Kmmr-bNi_kGYZT1f-p?Z3_=K@<erQb
z3G&2NKOxJ-vY#q!YO2AlbN+^bB6{q!lmssC@5SHjUN`EPMk<geQNwI!*TbdMD=U3v
z@xTKQJpDaeX=(^%D4itL`+TLN{e#cL)8R5{kTZnj#Sdoqq1-9~I+8-SoD@=<s)$=F
znq5Ih{Z@F^wyA107pCj>QqKi!H9M19O}=<<b=9_-%CvZ5n%QuAsK^oOwY*q%kvdx&
z8_gEr=H{8P3*g*$-E~!~RzvXI+Q4yx`MI;V!uN4B+NvsB2xoI+z0(0GibEUyf$JJ*
z+uV$z<)zt~%}pdg)WPTBQ*fDFMaXDSlx1F&Wj-&<aT&={j+^g_4g)P;I&YWGXM>0{
zIT}(ee*Q~jl)*1c!_eBc8m=EO(d`X~VK^N2x`GA1+puh{4MTG9gj5FOw8(F|u_!2K
zhLXYq-m)#*hFY_gaIGx?h=2%)AS|tU(rVVgwk<oY&7vr(8P;vPM-X>+Z?}wE6h*UI
zN(8YFp28_yA}z8^w#Wr?A9)@jFiCUUU6_z4x6?_*&=wWR{RkV#kfbVHN>U+{qBTe!
z0CbpQ(mAT))muzwt^r<YXs247yHHh#SA{3JXtlK#xUq5Op%fssRFVQ8bJiP7t+h65
z05$k9j>jj?ZfqcKY@BVj02gj24B%Ue3&B+ikUnE@9Ms@&o&#=doVk9wGoXK;AdHW%
zuE8AA6ofQQDaKDfHXu#m{fCPs#HHf!;iV-61?_R#YWHhGC@BQ`0se&6QWg(B4f}AJ
zA7J6><PFTD=1?FZo6XbFv@f@1B|V)m=babFW#_|mK$vqVQ?L*+4s<@y0cpr$c4?3e
zVkI+Is8b%B_REEgKg*$Xe8Ud{zwx$FGO6?lAq3cVxc{EW7F`@;3ND)0MZVwfHJgz6
z!=5&qy}s{<*ER*`R(HHqG#DNDR7H%{i+oe78tkr7$bGL?PmJeb>NU?F({~$V?6=~y
z_k8?Y^0DPYnxpLO%Bthq)==bO`W67RnoW%`v@O?JSv@lxT2jcc&M4L$ri3Qfe4%a{
zfSaG!xmu|mT!z2NN2#wv)(L^+yj=0LEi1`@q!nZO5nswmf}dyQmwk9$p6B@^MNt$V
zx%b|CAG`PJUo5V^`s%As-VAfN_1VvU_SbK}{r3G|zy0=~2J-*TojbR;2SMH-JOA+D
z^YBXe0trb@4w0+LJ$l@H2S`kD&|IXa(pBj$CL&GJ%s}K5+>{+DQll(h1TpCLq8O#I
z$m}b1*6Y3XuHJwD{SJcn3{xrO4%3ES8?G8!0t4Hzu8@H09XmTa!Z6ezWQ-XPds@qP
z2YqE2;)v49dxpU{+y8>5ls{y;E=pw_T0B{nrE(q7ph6gH%LZfY`!xkZ8qe_rH_t1>
z6wkivuDcYXY!!w!=zy7)`<kV;M9{apmc=^M(8eEfZKz*!7z>^>qy%2a(cJYr@GAH)
zSt2)bv_NC|Vp&$^^RQf0lZwd#y=kiVH>#XQqhcU|=__GNi)ZDy=2%{&lj&i6O@0nC
z#u?HptsUmAc6Mv^cui^F9}cIMnK(B4Hv}reK{m?Uk{jT20buSq6sbXl)QF2bw>e|+
zPnykEtKlG6@R$dHpo2W$*i2$T#)7v~=KHfG;SBWCvjJKF4mkEvmr{U6W4BqadxQ{1
z_`y@~DY!%|(jqHlhmbHyh0Ke*SQN9l%%+*p5(J8f#(o*)v+7++FxB&G#XK*bsM=O&
zSa!^DeCM5an*2!nAu9~~T#e^NA)jyL`S!E^o0+#Nh{U<T5)wLa?ACMB@zQ9Njfx;R
zhvxZ*XXA2B-g5n&cisQcop)Yy(ge__uX^>HH)gZ>a~>pukb|di4=xd(tdis8O7a+a
z3n8N<Ez2a$`qXc0wWv~zO;X7!V~NosDPxsV<<04Il1}>q@vNHU%t8ZcK(&XAa@fC%
z7F8<7n!ooYtyvWnHfmf>;Tu31^r@ZG(?@_%L8e#-;IqH|TaO++x>_)YJ9|(IhO2kg
z>-DAj(whefq>;h6RqF(yX;}$?lsfjg@Y=mFfVcE|y{;<`uP_QTKpg!E2EJ0hhn#c!
zC-yJFJ^Pm;$7TcBSO4|C_r33t_^bm|n6G=^``-7yNyFEM1p@j49H-_PhA^~ZRMAWU
zI8JTyzW2QkoVSA3jB5piq=X!N5BJ~$T%YyTNT1w99wTohA0!_qUn74B4070lE8t#u
z4BiM&z%RnL;NQ^nz>$R@8HqTGxIC#6a$4qDIu3~|M3Xofgo;r)D8+nSius@%mw7oW
zXXR`vq2{KP>wWRK9LG|{TxW9cjR7i5%RI^MV-a2iyq0y+L1QB#n=a;MIV*bch`tRo
z%B}I!ayHMW6B)(RGN0$u#gLy@C3h~*giO*&1y!?YkCN)mcyD*mO9Y5Ou9zi}4GIE2
zl<kGwo=-}qNmWg=>3hkeMeb(vRlX&xE^QT1VOC;ZY&J<2fO&NKY29U<>cy@in&76D
z#^p4f&Bw*K9L&|H`pR(@Wg=;+zBrmj)5(eO^U2B4N~ubVHEEfTn@$mwi_>zh2FPv{
z^-S9E=fLkg9`=97Gz|dbg&IOo&oCx!e*_*(%=+oCyoqK?V5!(y&MH`B%R9vqJT;%s
z=Whi^3aSNHidrTrptgZ(T`9$s58x}Nme7`}NY$t^4M0l)NGYf`O&C(zm?*bxr4%U5
zoMy9$P>oGXhpz8-M@g+YM`{WwnPHj$mZ_PP;<!>u!kfoPH9*d(Qk*edn)V?Yh^8r&
z0^}&*!7`Oo>PS~vDh{n=L(8_TQMPiZSc2R<U<+WF#|}!+X2dxg@BoFgGrASC_eJl`
zhn5OR$0ts1rhrAU^USrqz4i6Iy|t?y03|saaTdpna~4Cwc<L|?L4bmf8Wv+za6rRk
zoC$$oXelKMUz(=$ZJ>4yAaEH~f&<8eDK?4~0IEPy<Wf-PxQt4{De8bJ#vGSX#Ye!|
zC{_Td+;EDYUFkd`uC(y*G+)6OXN)m^-A+*;B;A$Om6dJ^#nK^Sk^14O7-UEelMCb?
z@+f%&`8o0_@_F(<$)6B1n&sMkUng_0L~Al-%!?wa0Spe!;R0k@%*vY(VbZkQf(S(3
z#SHcy9wOs;e0L(GuU#!t)n{pdAs!-%Me8CdsxlFPD?l{bLQavcA!L5Nyr@R`?tn+~
z<wWIj?yCJBJaxCs0&FK*W6X0L2pds?-q!m1db`zXub)biBngHpD?;10?J%@#+X_q1
zjc4|5`mSYtmJ!zOHIN1gpyzo?D+P7~F^XJ6SP!OYnnsodrkS4fJkNul250{{_uP8R
zk)y%>igX!srPlv#)$Naf?L_;RF3TtVCZH?`S=R5LcLFDI&R11c-Eaflo)=(f+ae4@
zN6Ih~(g_XIs@E*bIIgsmp3k{$DnKc?@>9k_!?4d=X5d;joZtUX2f{(W+qLeI3!yDb
z5kfe*PVQQf2g$R@kCP|f69NKe8guFCaGqp23P??Rc)VB@ilsk-m(wW<Buf?K+Dl#P
z&C_8YmlD2YUdjUVhf_@MsYS3|0WuczQZ3Nywq@kgd?`nI-QPQY{+iZ(zMn7tGx4d_
z)zy{h$&=HS<>`qN7p}kl!b?APZ&4J*XfzrXhkL!<z3Guv>$r7(H27u|MPcFPrzdCG
zM?Sr`Hy&dS`%m=#rDS7cW8>`E&CPRX`bjVCzwPYG%F2c1JMO>#{`=o{>#c8l+uK^L
z`|sb~*x1-0=5<$1ks4Vd$H|@KW#q%;bL6|^pU8ieripFu3O1gw_<m@S%YNG1nish&
zv_bpXW#rw2K%6eH=#ELVX(nR-h%l24<GD&^MOw*fHlGz_ztEK3$P_K;a9g8x?=7p1
z{F$8`9mG*Ae)(#)-ep<F<+fcp^C#;(`wYY`VC2HXoO8;x&wz6UEzUV*Dnj5w{?YyI
z1+!-u&JqWYh8!3(5!T_xFbuhNLdyc|xE4^Kp+T4S_V#E<v0)+$C;;P3ATX(bh7>=r
zsy}Z`f3|W{&jesUKegJVA%aJJ2HtHEf)^sVpNr6q&>dqFJ%dufCPoSoUZoH|+t;~5
z0(+1FYf6N5jlu-5?XZJjQp$kuQMm7pJMI9WK#QCLP_S*sXTVJpyv99aE$61~-`ha(
zQKe4-@FAeUADBVx07fpyX*m~F)D9|LMIIn8Bd;QFAeYEz2pQ#aT8biQ&2px>d_Unr
z3d3(G*P;DLCusUEw=rUYYbBXZs@PW6&Wk<ZRuT=Qls}r5E{R=fq@Z&_N?N6P5!dWF
zkDI8EU7*UQSyHXlX!Lwf6vgll9uCoafyKEM1H|vi=AX^|N9bk{N1+YqTc!!P5%_*!
zfp1wRoN6|kU>aNoo?$@Nyn-nuhh3D~U~9vdwO0f(>!OSz*1w;uq7R-veVV6{>((2d
z2jDo3x^4eC)M|hr<^r(WrJn~&gMIxU1%MuypfzJk`2~o92Bj%eD%_g5*k*>|JzsVE
z2<paHm?i+e;uV3EvZa0z)Ux~;t<m$leb1YwsM^}|daWP~gO(ITA7}?sB7y74^@LPK
zTIlLhM@p*cBD|nxL^H4WS5xVia}1?UN?FlkUG~4X8kc_-hG86K-7thWX|05TL%B7`
zuBiHbz;3VLVn|_Wh3y!(uXMVIr?qVRKEUC%BrzEEJMC+Z^#}MPWxaO&bUH=X#d>XJ
zk)+UWPc3J)-3AE5wbJ(mf@*YOD6ggv2IHLoP$(7k0HUaGntOQ;MSl8qh_-{Ai_ZA@
zi04G2*17eM%XsPm4Hq0Rgme@n>l*t?b(IN2oaGtnmz|bMngi*_BuwFOHsEkX|8KMe
zH_~bZwqsdgV%cE2Zuc65^JNg2fFNzBH3Be=0D!SIfGBKiixD>K010I^tF3RuU^9*t
zSYV}33r?{Y1mHM>hUbEzp;r>V-Z$P#A+0Slju_cCM3&$bz#Mff6UHos$m76Fv(qvy
z2!bYVWezXGr^qt7&fENX*)&7TkD~N0!G;HFaJOg0v>mvhd!hZr#eSLl#nQecHSL#4
zJi;xZ+o@5D3LiRg;zZ~0?ln!dkDAXD#3b=5?!oP?8@D$$8jvQ4$^JLQk6+l_|C>@+
z5&KU*`Q(!cVluD-y_3r;ji-;Yo??<<+K1zw8@IL*TCGh?5=`sQum;k1)kKiBgUj$^
zaGC6p8_2WB<K$1sUlGEFv#8M;%4pvtP4Yt{N}rEnaS$bj#J!`qtY1hG7iHxz9>uL=
z(lXC-=||kkG8har*=7%90aT#bTp?Gj*zN?9r(Q9cPNrDuXU$4`A=pas3di+tlh;@Z
z3858DQr&<a+%8O$2gPjSFz_8maRd~u8wO#`G{JE#iiQz7E})nrtgl~j>vRHOm{~?m
z!#0Y2eQuf@2K|q6ZkXNxX=H#f4XprO*E3AeRuDL@l;Trs$JPgZ-_qKVmTj6qwU*LQ
zh9gxNM3HGS0M4~e0XPG&?7(te3p`4LW+RRPD7E}ntI@CxirQ%2r%lF9O&P%7`rU4~
z<42AISa0+PbxTr~6h*Ix;J8U+eQh8G1Mt1jaRA({?e#T^+grEv8+GFuR=q~iG1{F@
z*F_L_9hO3x=K2FbIr!~%yXAcu)3i}n20|33sU@U@&y;Dp(hqz9*Yka;KuT9TUKF`5
zfNi^R7*JqnTb7-Kp`{r&Lj(k^B0(9mnh(v??apY_`w@zu4Q)``J{SwqLw6^x>;^@8
zSxQ5imDJ7X2QV$CPpz)5-n9DEo#~_N*xh&Eeev$wpIf^fmM>kpbm@#?Tj~C9z{>tN
zU}gUsx8EK;?~QQ|p2P`!l3XCSkq5}j$s5Sq$@|EM$j8X9lP{CsCx1fzg8YB-ZvbFJ
z6LMIG3a(_ayt0_*MPX^C8QP|1qN7}t%2LR2%=r;y1Y05=iYL`nD^#a8gW#<$zm3id
zUZlmejOW=j<H4Vc=dzOHVz8xnT+Y<mleCQIJ1{h!Z>cI{u4LD(%2*!TMxj(EP}wV4
zS%gVQB*3@V0zoRR%4wFcS<A_?D5tWLEoV!zoW~1yi*g=o?3vVTR?gGBD5^oREHe`Y
zQ>y~TfEN;w-&SM<wel3OoR*QC?yFL<=3^?3;&F+^bUH<_w*VZ;AK_B+N0@XFwiLj^
z1TZTAbN259&tTF4n3({T31I9$y7n>wtWT#?0NjEZqtlIz4V*Na(*Tg|-S=73G!qjp
zT)1%Eh3mw%*Is*4yxe~t;vu~6!VC2Cp7*@%pJ6X!La-~}HGSwY;Y6*|#o@z;4_~`H
z91e#oD=X*yXGlunX}7~GnGo#WJMudoelcSV;kHMEP}!!<vM7p}567#jLX5Kw_`{oT
zzL_!`??LK-!|tjM#e10na0j(ab~jRs-b$$hwt}w*9K_qGh4gM}S@dgHU3Jw}Hz16s
zM;D6)z&(hFx8dxL2RAo283XU^b*}4PdAI<8pFQy)e9^Kj%e?8Pn{K$}yf}aU{F#;y
z6M*}Hu?KE>Ilkm2FWEiJ0N7$75D^y9nQN0IIeO#jg*)>+&#$^=eSQ6=VCO>=;I_LN
z0K5BcOG*iF$Ax5fcX#(x8~{G^-h1x_NIr5c03aYCNWz25@C$I6BxHjSpleD@Iiq{r
z@k<GLZFDCHEweT)GdS}nfAS{<chFq%7e&1R5B@?dl+T3L&RcA)Vfqx@0?lUemwB23
zq-hQxduac+K^ykl=54)fZo^Ra9k>in5Qj9$5-VJ`8pHqEvJ`UEIt*@=h%VsW$VHW;
zc~MQhwFv?3&SZL|Uc2+usSMr_H=lp~^@Ff6IDh_8w<ZpcH~SmKeBneLW<7M(Ravu>
z9oxI<jg7{U$+Xi3F+6qpuHS5i+4<|A--;i-aAEM@`Fg)B56gPwY%Si=X=c}4{jkYy
zI(|GOgb+$3A-D~%A~D${=LmtcNV77Rc{Y$jhUKD~q>CyX6nQa_%u$+I7Wd60i!@2Y
zf`&BDa+%A4q@`s5tpmRB!K+=@HMcf5xB8t<r@OwnyRx+N>!bc`>+tI8;jLM31S?0*
zY?tNsnIkK!gTbJj?X0cs%*sL5?dDsXn_GFeo5A(}@gM&I?Us@@U;wKtu>T3K2FuG(
z^We?_V7H4f*#Gc4)b{@dYB3ZA#QT5aYTa&Y?Gi!=C*t5Kd?{QaLoy>L$qnRwLPm>u
zDTG6<C~drF#Z8_rJ9hxO{id9z%?MizH^*mo*4EuxcbL|I_Ja^`kj>ynX4wFtQD52r
zINDtY(dpWFD>z*T28M2%@+)sfyPa>__`iej%y`pH(oZrXfLbjxEw|Gd&aw<3%Vw+n
zzWLKOqWv5P(Rq%IXg}@%IFH+i_J<t+=fgIl{ihD1^QWXB<lxJ289qw{iQhr5x-;E~
zd&xuOIpl@p73B5gedLeGx5%H7za{@feh30=h@c4rScWlFa0E`lRd6%h#mh{Ll7$Fn
zw<-&nvbmTp6J)ltY1VqpM(op8gX>JcGuSMC2vnX2ji;cJK-g{+-mJv=!^sY?O4dqj
zL6gg<Ndpt4Q{?l_l0>n~D0P=m-}lq9Dt&rKJH{{jrSF4@Sw7$9cn>$H(M@bFCfqq*
zXQ85XUiJIC-7f29-ENn4evUEL<sa^JI=u7p)B}H5@igbW!){@Wb=e>F`~CjrU@#aA
z9QT%fzu&+6zuWG<?XB0a{{ho9&C`RyU~oE3(}DT(*`qHSHech5F?gPr1c(ZRVQLjW
zQMt4N&ja*)uMz7>*U7R#Yuk6+mzbt$!cC@Wn)@F$P1Brz_s`m2`N~(m0I%JD>G$S4
z%|HIQFX;FC{XaGTxWC-*_xrzW*7oXtzuzAZ27|!`(=_|Za4;AQa>sEDr)Ar=VKc)p
z9AiDTe8skH?Tih>u#N7t_axU^+Xh@`e4Vua&B1?~=HT}!81wCZaj3&{0x1DBfF~)X
zR0{;)EDnVPumepgMMEJ1<H3!CUl^?R;rjhY;K%nLx$i#s@%>labkj{YfwTWlu)4GJ
zs=m?xu=$gHqkpq`y#L|BFPNs84ouS|giyi=f&Ye!q)%q#3?YzBvw;+GTxMjHSq+c+
zrNw0f++Bqg15o;I81&LM4gBdNk33>3`57h6M`{%7@WndP+WxP#LgD+GRx1p{u=PiS
zX2R~^qLSv)k|`C|_J6g8Sc9u-xc_YkgLccelW(=1@~o4FUamNv;YD~kdB}&~y0tnO
z9AM9zk>uEl`Al8cRKBp3&88Kc&2e71#A>IJ9KP{*5&Hefk(HHUwz7P<YBizN-tPDN
zO=MbW5JL2pmsa-H)(}@#kDn{YMSnOPm!9W&8?J3CM`!E{Os_v_2cfB$0U)Agv%Op7
z070<X>%ohr-5y{*D#xP{Vz;-v((Qp|hDZ$rLrB3}o)55896P?e47j{JTkG|Z4*G)u
z0_Wgpt&K^q3!n`x0Zb(l)5LtVYnU9`olQc3kb|$lMffzifsk;ZCe@fadoLU%YKREW
z4ccDtz6C4*_=F7?N`>@Mbfe~Zk!Qbm-Qit|aU4m;gx~_OD0XN49)MO+Gze`6O$#7R
zlN1duO|5xry%({bX&RIpQ5ek!1K-2>>>PZ0ym`aC)dt6wTxr`0kBoAFR%_nMhK(jz
zR@7>>+I=N`Z@E!R8p|uISsY6Rl&&<J>rN184ej>&`sP(@YlIL&2|2LGPg`V8UhZ$t
zyf5hAg;Op#cE<3$Q<T#3jKwSJPYOXCO4Nf&I;|aOxVTgfvYcHB_|-I~@XI`Mz_Zy0
z@QUNNADK)#ZD@C2SZ*JgOgkM2{N5AORe3sOEx`HgnyVHS()s+F2e{*K$r%@vsag=Y
zOmaota1R(m!GQ~*feQhkH3i`HTG4P^OIsF`EZ}yydk>)9nM{rxo<7=bj{d9m`qENw
z|D#Z6!AZG3zh*Xv#p0@K<}-XFl>)isQc}T9$7u?|9L_;2Lvf3O1Yo8aM!v@Ybh8<n
z6cME~5>63_e>2^PP7kf@Y>9Mf(_>+n4O2teLun<kZV4Cn_V%v7>-xR%_>_JA>9^kc
z^z+Wi@%VDH*<3W?(&fv~9rwmhJ~{4<pS%C$9`Vd9C4?Aw(boQSMJ$C<+8QeT*Vzg;
zVRJeG*TjflyLviBoKCO4YPjqw=oR%wqh90<D%~gKZ?d)4+V)I-@(Xb~yZY+c6u<uT
zCXUDF&W*?Trw?Fl_1Ll1H5kA?N)+jmXbPZ7?5H+RN66y9K?#Ck8Qt6;U31Mf{S7Is
z_3Y}l+VFSo1M<#yzVmj+Zr=X(yY3)_)DHgh;1BUM{5rWrK1M!EewX|o^4H|wKtl~y
zVGpi`$KgHjN%#`{ANYIt0h-vu5>Midgp3w4FBUQ{X4ydIayFlw<zUiw6Gfgaszo|4
zCucQtik6Iw`9gV{a0288W`4|V0!y;eD4wKAI*<dIORiV|s?oS&(3i(T=IM9l?NL$@
z%IMCDVnO4jA*FBF5^{Pb!mM~@G98yZd>P1r!!)BK(LzL0631l~Mv51bVmwzTSz<Vi
zJXQ<0?cKCU;`LLqw^WvsY3ND9Rd>y_ER#5qMa7HyiDcJE!VMt7da=ngsgi0jO{zrE
z)K0v|ZcC#b5P_}BAToK#LK%z=)Vibb^zE$53oeR?zY$d)L1c+{Z#mjMIW2`e)s1E`
z$Z^p}6W-~zMH<AY$*$(Js&K3FNJD(d1J92iAF_@#akr;Q!$nKx$pUA$JKWVS7tEDf
z2<}>@>vBqEKrNGcTruLy0`0r5t2ISLE=(yEfDjG0xj24&NAO0vG)zoOAu<{Uj^#5B
zAeBfcb+zlczG93q&ZL8uMf-}ouBihqgqAK7+|>YBtJPZWM?dPeYPA}`6^NFm6nLGx
zQUX%#Q%VgbY9^yhaB2y$e;ZJ$YyhltO%bGWgfb|lz8(uHEDNOc*r$3)3=k*1-f*pt
z32qqxTyWP}05HSzGzEZG`<&tX1_ulR(*$6T1R)Q*|B_mHoU@^20&?wpR0N>~5VJT1
zbOiv_uB(KQ5<q4}G1<)XzRhE1nvU!HMlB6ZtvP4_#!QBs0x(?J+L}-Y&cE+{^TU3C
z7{|MnG2hwVqzn`?lQ94nA2EF2b!`i{)@E2se}mQQ?Y47zYyTUL?;~i#)~3n05a$0$
z92YQ68XlnMTg>%p-vc3)DUhO6h5_(PrVxyqrePTx(D$80B1LXXDZn&T(`8C{k~5Q2
zZ5Ubyh7}2u8ybM31(Y*wx~}8cw&Ij3N-fUmu(ryW1Hd&8ENv)6V4y7E%topxWkJ6m
zX~qD+LJDAj!A-XxXdNl!U@@+Ip_t1>Jq$4wBE>MQbFt60p$*T|sV%wVc#J9L`@RKS
z>L3W9-%YgV8HTC3owgYB9EWqTeBY<qfp!fzQ0VkOl*GM$oB$+ozZWOJ=mde3snY@!
zo=+)bo+}oVvoO@4qnJ$rY|m4gE6;Pg0AMIbipT^&a5(~KKq)-aXs}|chYnjU*K4*i
zA%DSE3QWVT`99^Y7jSkwj*wBwna@nI-t#-H+MW;qeP0RyVMKka6Li-}x&L<!A(3Mo
zHS5APOfYm1x*lV`UxSwo9aqZMaA*Hl(Xk!P8Kp`o!!$k#wiC5%TMI8>%x(A{{4=Kv
zg~_D^R7##OOhYL}DdXC)9b5q^gej0wE|o$;C?N;m#J9uGkbsQHSwcpUEK8BG?j9<_
zB{W3-6$}`F!VxC<i<D4J7PS-ntZj#pZQFJfZV#iXTZ%wT`@w_uvoBif_fyK$dTsBX
zC*PYSnij==H{MmRx1v-00(NF4T;Wp$jpv<mR}jNm&9(t+y6?B5iKIRsw@W3PJo=9N
zz?*1<R?#tZGkOwweVGHd0M?r5m<)JVC|J|A+QTJj?n-zxHI0v>tQ7-qI>^ed3)~17
z4%_nVzzKH|Q_kTp{#s1Z(jnZQ4?||A(ag#JJif5NaSIE_hpSu$K{zoH1|XOM;99L#
z!+&P}h>d?5_`lMB0uc!j2|O8$S^pj?Ns_KhQe+i*6HCKky<V>mhf8B;!Lzsg8;>t6
zaF4zm1mQ#=00_Z^s%h%KpDaFWU~D|a05G2N_(wnb(S$sj0Fa&|)kU!`Il3fCk`BSv
zxmV}fYW<HfiD3<TKVvM^MdzW5(S7LY2*nO%<9^gDRA?`HHR{FUyE<gYw9l<zqYB!+
zXb{<MAp`<qH?W3wxq^kc%P#hdT%XS8?irBeNji#|&1lAb5`@!LN<jfo2vK>d3mnk|
zMddZZfeN5r4f;2qJGyIbZVoV`VKuBcWmr@d4$d@_Bn$wmVUpI8Rtw`+Gkqr))o_F<
z@xAT$n~L(;Mx(*08g?M!G>v;VZsK!rOG%Gy2egm@2H(rH5*0x=+KH}%5L%R_DWg_+
zyvhNqR55IdGE$>&6`QDb^iTiL>t=`(Kpb%fxs!_=&dh>`a$G}rLG(_3Ta)~8Uq8b*
z>+jp!?_=EW@9i|A&2Kjj*Tc@gPxXJQq`Yb$KXKy3k)pu3D2|*sGQSOn;mmXnx2*Dq
z<-UG~w|=RHLAxD>$``-*#V;yh*lq`*3cGbnUX`s&@o04T*f0mkhsW~CNzVh!r~W$H
zi7rOB#?pt-r=h+d^$WjtM*s4G`|e<#t$M$<O!J33^#?^fC_?`r|Gn943ZjUc0&kYZ
z8Zq>m<@+3VE0r1hsWVioW8G-jzK@~eR%Yxc?;ejEb&d$1zT@+j$KzH6&}fXuyVmCC
zFpS5?FPfXf+_r@yzbchV8plnUssfHKX&PW%&n8PIwEz2c)G%;;Uw~F?JYHEHkDCo>
z)Qj=n)p5~m!qy@;H~)~M+qQ9dZtkL!qar5f5lF=dZbucAvG%2hYc}!G%XK%+io&1m
zDJIYK%*MvX#+ftIzO}hIJm9@xV`F2(J22ebwEXEaXTH6$v2pd8Gt-{6xj8uCKmY2h
zuio(Y4>vb0Z~Dv`Sz9We`y~9!U)FC!jt(N6;*e{-9N4`ADhLxDKNH@^eFDgBuO>Tn
z>eQ*Iq8vC7<duW_u4uR032)<peGfheL3}k^k3V(l)TwqwK5)R#y@UI3yZuNj2lkyh
zB|j{KbMJ#4a29PxN7^*_o(S~F`q6MN-u$5vIWuB<u>LKTHbk;z?ukY`r-OPN<lrXS
z@8=vW$MY;_zsi`Q>-Cxax}h^(6D7ay*q-Njq9jpWv%&MGrhL!VH6}`;_cz>*iz060
z;VqYA5Wn1QDr4B$uj>Y5N_}P{f0&E1U$-%1s;;W0La2^a1!G0UIwgv!s`_V(I0lI0
zqG-neak~gi(=>z7G)*()pORHjf(GagMl=!z$Qh?$?}&8)1xeO*a&FX4<S655vj^+k
z>phI*Vr?(`p{pv2>JFXJ%Yg0rUp7-Imn$=mpJ2DCqR27Ebk)!mbyiUo-KzF<MUfd#
zsiLZiVW>J|j1xuG(8EwSps*a<wr$6{7XN#OOOQ@<Isl!{L<+AHIV{M6APIzkI;*RS
zY==ErQFY2wrYI^9B1nQD3z!pK7<fUrmf>&U9X3&bmbn|Tz13VulwCWn4sK*|+XkcA
zz>J7WJUhViww?|*Z#})FC{V4%Aw9NM1w}5sLQ)j5UG9itoTa-FB&lXrL$e`?qAG}3
zlu|c>O+_g^b}_D^{IRC2Z2d(LA}5jLnXgIUYi1af8oYo-K^0{=cA}UgFY+h62~R=y
zAp}U>Bt7ZUT)WwdK2k6V0hodaN!HCak|a%dxaeRm%+ZTR`7ke_FpP3$INh>qWlMc!
z;bs}$oX|xhXt(RyQPXPFfM*c3<47vI&MS<PL3u^iE3PZ^s-jx5Xse3Cgp#2@MVBO|
z%CbrgLso}SZ#jm}VrLttX}Z8Od>?P<;wC&<Q4}t*<E0BO*!7{Py?p`T=#dL9D7!9j
z9OYCHMMDdM=6IA>G}rZ`vOQf5d{rR?G!+<AtF9MH&4yp)2;o#*yI-u=o0hH-@Joz=
z1_&WkLOuyof|{s^uVK=g&?b5YdL4Qn`Yiem`U3z65W<WO%YEjpj5u}+Ucy2UK;_MR
z<D^KZu;Uqstq><g+FYm3+StnPc+aM|PV;G&6nvItqg0oLm5B#R@do-fcKKo%&2V?3
zBHI+9ak-pXkIYR&)yOSNaJY49p)ow3fyJ<@!BkCkXU;0ziLO`x%7X*@_*j9JB+0Jh
zMf&7Lx~`K%oEU)9d5+^R=Xsvv8D<>MVSsr)pb7B%I2Ov)F-Hkzm@P3TQ_849$Q<wn
z4+C3v)(8rUd7WX7*Lm*Z)sNsgh6&{`EI(r&RvwC{4*Zk7@$~7_cbtxvmvt@hhOxcB
zJHWp=Ns=4k=o@4GC3!1jjO?-5!os?)@6K_+aa;@bUG<$-5~smVI?r_pQ7EHSX6%W4
zNFNaRJ!?GjI1U3GuJA{zN8|_d#NW?`csSNs_whI0amO8}@A%%a?(!VJva$1cK%DTi
z25O)ITJ;X%+?Hh`3&(VnqpJzi7}(K~;-=96y_byMnkPiiys1>J#*V%J(#7o<ZnW^k
z^DIU+S9#;D-~ayid##pupgkOJ-#tIy+WItjn3Xd8addXq{0D2M$v;|OUA<zt*4`FR
z4|dJZw_5Ol4?|d;e)P2O1GJ*~%_Y0wd2y_!d3^3u@K1Or2$*E#wVftuFHI@KS03g`
z-phJ<lK1jnG)(hs7!C6z`<}~sS&|RaUgrl)o@K*mnD+8smgK#(m-o_P+RIz{kjG-0
zC0Q>Ws+1*plBd})?WM!0m-o^%Yv~~ecxco#2IoVd0fZT&;HbsZ&?b{^a^2Pu`-G%7
z=of4r3=c;}Q$f*gvly9RI~y*WDI!9CsB1grcFo7{Wy2i)378t-Jpkt5rGdvTDyxFT
zn4%Ics7@MfS+ZLm)&aot*a9~qo1KkyrEG??Gl>TbGe%|IP-Wmnj<3}_<;m{EbO~29
zO%o~tlvY+QzH0iufH8G!UeyfRVH3P;#q0zJADi&t`<<qW0wp5DSml9GIRwxs22oQZ
z%}gdb)d^8-c`ZXHk_4DB({x?ia(2wjY9{8w)?BO4o2EU#%wY~6pPPqfYdk)m=Mh2}
zp>tn^f5BOlpdAPS<FkxOtK&hxu!9fN_D|-83rS?5>9BFsxFo!K#>4#DU`Xj8OycQc
z>b%PsTOQ^0dJehAaM{XKado2=)hB}>2qqh?sF9aU)6C(e+`__<qvy9AyS9i)t63B)
zS<lnq7HV6!TL9%!Z+7R-*<Q)PU=5a+cP{t!z1F@LbMR?2L_5(BS}=-}3g!S3JtlQa
zcs+y7awCc&b7_<&sm(C)5{3$))NnNJ>?{9P3_lSW8yA;fu)N~9%|`2j+|X12KbV-9
zs`y^8ciX&wZ*_Vmi81V<JhQegR((I3X}3kOeFv`9x)Ze;{=&Nq%d#rE29z*6tVYnV
zDivp}s*~Jccu`ob>iaIQb~^2zVeo=0V-AvJb&65n8#Lp%_1?7Ja6G_KwC~2Q?*sUL
z_k9B60W`g0n3moM0u5n=&ixQ0-?@$bmT=@8Ux+q692A&(4n%_pw{25Gkq%ebI~G})
zA21iYA=3b&rqvE>k|asmykpccrvv=o@#DuE&J!ftYQ_q0=&~Z)NmEha@!Z02{uw=$
zaM@(HB}p1Qph=SD^~c2mAy8@zt7T1NlBa~ho;|)Nm^|;;UhVohNz{Kp$M4%~ock2K
z0^WgcK#xW*L0|gODEJbe!_6>=Nv^6`djf;au)sq>Drv9^VhOA=t%Iw>=2)(q#nJPV
z)bi2E%7rxWu_zU$E#PvtMQNgxzPg;12ScZSOCeGfLSK`>Dg_4^@nmM0yFmfqk78WL
zqQ}?CA>cU9d5(ykN(I1ng22%*2arxpHIpPQiUb6K05%mY{h49_On(TYojZFy3`sm1
z_51!35W?pZoiI>%Oaz|xMx##3{djKI>inFARTcD|OBY_Si-;nB1~{MubDTyAk2#=>
z(vr#XJT)0%7-Pa1F&R<tH4>#9hq0grCsRKF2!n~NTGfcE0H|uGW}0QqwrAa@OIafw
zjnX6mL1Z9cPJFKjj1rDfc=QNha%%5Nrz1(OyA22tMQkx<=rRQm1U4l}ydZDAeMCsN
zdsM_A9_tTag%?G71I0iAIBtiZ0tQC;lF4|Un2hj%fiY?_PQjQM`kubNh?mfbTF0iA
z6hiL!tV0Imb*w`@_d<QQ%-fFhQLl`O(ZJzD5mY**c9Hap5s&spS-dj9X3LP6OqMxQ
z@Y7Dc0XS+;c0Fe3;uvchhKjee9JTgpmW6%FFgUhtV;qFR<Jk1{!Mr<Jttm9#glh3%
zuzLOpP3$P5#1Y05rfD+o_*(*~s;cN3#TZkl3Zm!r3&#<~wy5jE7<9e-Os87qaHF2`
zksx40_Zhn=RTN3`BUzbvrR)bBSZ2jF7{(Z&QZB;;F>r<FW!|zy3IKp9Wf%Y`RTUK>
z1gLcGhj1ERhD?NF%{j)=F}ah`tdvDyUwiGfFSrS^7v1@6ID73wwzhO{vj2sT!Xr^~
z7m@~Km~930Q%p`I{xx54aF$y?u_Fz6-qbad)*M}{`++r7Y|~6ajLS~BdGdS$$<EvE
z-c$8<rztprs)mk(>HfzC;f|?FlMv|!4j)4pp>ui$!!BBx1I>6SvHQ%$CA%3|JbCiu
zLzkA`wX_qD>?XI~yS~1@{;svPckygHo}r!Sq3B-pGo_~*S~(&%uqIEO%yt0g1l@YM
zpXONy9=JTniVmNRz;F?EDT3u28Z7NNjeiQkID=5{;9f`nQ90TFVQ`*2#&qhDjsQT*
zA?y!K3kFcy>+sCu4`XVSMN#DI4PGphP#2Am!&`f<pUm@W1pvplV$4y`$HL8A?Ci!M
zOHJl;fA&0&<8NS?)t-kj$9aI7&8AQ)3C)TWhLT=3<9|GU2^~R~qubd$L4%;@WLeqO
zC!=IKxTK)F?fMF&+eQT$Bga6e5W63^f4S5fKKcVAE&9bMr|QsZ@dcXK38Mn;v^0&n
zy0JUA{D$LzBHLtPVSRmJfmn(Hj??h1d>1(7Cebw8)pcEWzfqE9EXziVfFzZE4~lFU
zl0;gDEH40y(llaNvYjAs90_ueWs@^Y71A`oaT;aQ6j!FEFrJ!P5zUg@a2zn2q*ZaL
zsVJc9WmnhX4p3Fc7Db%eng%^bRRJ{BNwFx}yLcc3cvsgn-9-r5(3G9U2tI@+&<=EY
z-wUwpEcy|?7zS$%y)&P-S_9YivQgB_GK)pKAPNq2xzx;Pg$IJ7W-%-#p(6=A-wVj$
z3te{4j@=i|%~z`swFiY|0tEvPhX>~dZmlK)7}#{$C$`N4PSUyY4C#bbfP(;FR%W+{
zep)Vbx-&Cd%uZrPDT5a#08y*o@Au;fgCYx*0+^M>P6r2*VZdPPL(|<()mq(svf~E;
zO)#dX^Qo}h6_ipt?$*OWtp*Md!t;O-SCxx=6W)a^G>tAmh(XAlU38Ic9%R<cTd02w
z7B|erY4F!56p~Fb2!38Zl-KHvQd*MzBx}sg9}pPVxo2Fo-g7otTwLrPUg~xMy4|I{
zliTVuGvmd6R;yNPS$}b%*Q@;v0P`Nj)Md{(f1)Es7}t)MZoFKn*X!qVxZB;as&sp+
z=ld(bBL%2dhu2gjoWphsc!p>UKI+q~{z`)WNlasBMEdPzRy<r8$MK^tz3}T_|GN8v
z3)Z*337xHP&is38eSQ7w5B~e}l}hF77rx-N$)%5Z>U-Yvo^Mp=_doTiPyOVaj;u3d
z8t6S9Js-V3j!9z*jo>_uwrp|SCp^R{9McMXSQNE9>HvQy8s(Vf>L>*k*Kw-x+Y{uD
zxFzswC5d7?Y70hqW%qLiENw#9fnNndK%fmIpfL$aj0He|*|k7q83>`gzzd?NC?aSY
z1_408Z5j|5s2xZ^;GH<|{Q$oc$kq=@NJvPY2_&F2An-EUI73LIK}e&)a6-^tJX?s6
zVH8KHXc!{B1+$qZCge<m((hc_Run~PU-}(NDP1CDX-MveOZXA+C=PB7-U8i4;$0pD
zHxRb5VVu0r!=wi|u#|BOA@F1XDho#v42}A<1f^fsgbtd@&B`u*1dw&iVNBrniY&>p
z7J#rWz`(emZkdWC$?8?iGX0uSZza3tn-=4lEi0DqI5o=Dn9SF8(RDXgYnm)evI+&E
zWSziS%H;r<wjHlxnh>n}(*=Y80>54{3~+Vbv1>Kk(RG&-rKtr)<yshY-7V_|ZklGr
zb0W%NJi^rH<uL}DM$1TVf;Z8(z3-MCraD4bjHP^+lM?%2f>l{ER#y#4{(oWEcaKf*
zD)Bx2)wtrp&kRZ4D@#T^D;N7?#<5p{f9?g2eD*552?1)ymIU5pF2@!z`k#_r;U*5)
zU>oBKz#S(i&@%);MKoQ@bc4+KTC>??a<Z7nthwA4+sm+TZEbDOo;7LUs&9M8JKph*
ztDaYW%UjCNd&isK{Ei*lw{K5|!{ld2X*T^#p8OSf6ub$w(E+<DJ2>-L#$X*P&@<_#
zNlK%rCG{Ovl1;muD_{920>9R!-3s((cdbrMIUMdM!FZ(;wKyC^js;-4tyTrLdr8f<
z!E&Mi<KW`MOU-Ce?_^n?Hk!3ik{B(0q&nE2q~N+XhTWC4ZQHRY@s8~WRvBZTRl%}q
zzUA<e&)5zOvuUtPi!2i)$@AlO)N&9)h@i%~55mjfBZwgtb<j?PihJJWDMVSG4Fo`g
zsjmqI5!hL2G*`oEdOYCfi!6nBa&i((4R6pSyw)vuC+6m+r+p86Z)$pOt>1^296EI9
z!M(xBlfhp6unWMMoSa;n{O;uB<mB3;t}Y5crq5(iTz%EtT-gO~Ts-&Kv15-py(c(%
z(%<uj&2_{f|J+$f;4G@3I!a=lJk<_qao#xByTTPedh+DSXROt4xS_trJ?rv`{>wD)
z+t}E6@#ohXH{8%z`}ngrHv0YkmHij>`~CiveeRKd3iI&J=)t4ZrHE=H0UXcRFsZcy
z9YvX)pE^X$)TPU=qZlBy&|n@lDn=>%6985*yi=aTc=iLc7;~D9(^0%?C`t>i%Nva6
zStGv=TC%GD1!MdU%C$Mm5%RRB5yEk~E<)ddJKYH%4s|L7fK)n%{fTY|5hS1cA^Z(q
zhB_!iJ2a^p<xBvUbL107sTNRyB|~WOv31W8H+OPINx4;|lb(O@g@DppEvp5A!9<Z6
zK~U?}Ym^3wSz26JUtd@(naS3hy4`N~TQga|+ZhfgpYxj6yyi8hTJ5&ygQ`|4stUf>
zZnsuemZwUE{|J^>E|{NRb6r;##D>=r1#4}79wCIpHq#0o%F%YT3mrrkp-a&Qx&_6Q
zyMbU}`?yl6l%!Zi#493n#7Z+86*1&;Yv7D%0f~5S*PZx=(iaSzRTXpVIi{j2iUrr1
ziW&nIZ(dxyNumH0uA{K^pFMl_?2$*K(OpYl{^&<P`q2xQmX_`sU3kS6aOc)@;JU5n
zSc<ZSwxU?+<Ka3A`}+NTK)0Sl0qB3+al8ykk|fXU9`@SW+S<9K+rB%1lcB>tQVHZ*
zF0RYl7}SfJ$;|wHng_SnDBJ!}Xl_m?v+#9x<caIgx7#4;^^Mco8ce5$#&P6Z;dM2w
zup5OQVmcg_8AV5%NuwSB#@)^iTq^l~8g6Y319kokphCJ@Os7}PPj8r%N&#uR-KTW3
zEK}DIDk$;{aeZm2%@a0`$VP7t6&pQ8G3`tf>?A^gh=C5dno)QM9$&yMMF9X;@UVai
zL1QpftH({M9|BdHDnMTf%?j^c@k_8TC3|2PjPrI%!<WecoLDrQ^alRqpCKhN1+685
z<48{{$ElBMb<Gf2vuoO1QGiw3URe=<j^m}Zn#I6B<#u+;-w>`#5zs>f&ug}hAuWp4
z3kY;a8=(?(1tB>2zFT))Y_<E5wmtfKxR<$J9~Q*t$;0fJC<7*ywZN}9P*co-lH$Se
z{NLMCIbeM9WN8%b(v8Y7*Zz@_grAVA9z`MNz@&+-)_l9jqi{4lPEnx|CqW3{x(35E
zvFeVUrb~yG3P&LYWNDH#{8(Eea?BEgeZq0Ljo5;z7~eNdhce**mjWnlA-%O!UDQ(O
zb=NmmPh=VJqs=B@l)6qdJu}eQY^F=K6zJOeyzf|yqNOAU<lJ=}o8EBZgh???>!ZAg
zgwkemf82=@IQR^F7A_M_j*xrf52S*r(D3Z9%0)3Nk{%}rs!gp)+E%koqzdvzEQ(B+
zt~>`s@-(jcA`DS;tsnTbZ$h5$vS&kx09w}@<4LyzI2s+U*Qn9yoga@WU1Xc%&2hFY
zWf-Q<a>6jI#aeJ{*zbGD{Z^})_Lh>Q0n68(t*x#ty`<GXyi@?3ea&_i1e@E3#${~e
zqxqrbWi-;*1f0*#o!uSTmSvAN7uCAOfwe7@3tu~qo%p&|8<@HoCvJ+E4aehxP<@ui
zv`%g#uOR&DEXbzm)6@zpj)pwHd{G8eLv|G_NgBz_&!qsXbh1HVV8CV!LZ922%A~E#
z4mSka^SPG&U+*JyI<qL0=(>5DLKt?to+|_djmR|7gj1N;Qjo4?GS(Pujt5Wpeqc~A
zqBv1tSuR+XsXg1NM-IWSefQ`2`bMvh(sl$i>Z5*I6Lz;V9JWn!vhFsUbr=lKZ`A9f
z(5JFB7__tO!`g8?A!66%rX4s21uA7&?{tzBu)lZ8k?4c}1^*78AddsE+nUd_tlC_Y
zC|DjI$boEqY-DjeMDZZQ8_4cCbt^fgl_r?aE|rvApju2PWwIqyU5&kom!dl_<Ybq6
zh342)n`XP%Dkel{j9&3E>13MBXY&O05DIRVqB5$e3^u0RdphZOy1S`_e-vhvgDDOI
z!9q5el1t_YEzPzNfIzjjoN%zRG6)?@Yby0FYnxIFhjIi)yM8PcL7$?hgv+JlF^rUH
z4R-@S(uQS!nzIZYc@VfP<M#nY>bX)lV&Gd^O2!?(X<3%zritsCmfiFn&ZN|quMt`0
z`v@qNFPX=<L8+-l`9RV5?V>zRmY0*58&YD(2fGm3q}1TdW71a=0e!z>i28aPK(!Y%
zmEy6q?*%o?AgBX?A9RAy^2{(aJv#_Gfe!%nATW)$Q~?mVQjGvXc%CPiV#KA^azQl+
zDivLyHA+$eipz-2X2}#2;Cg|g$heY}g5pvM0U(5uTmhw0afVb0O0m=F3Mr+*(R8#s
z>QWGh>9FQHrgUO=|9F&!>(qVUGeH7ShM*l2nJJ~@-EIdM6(k>d@b3eMeYY!fP$zp-
zf$?52iVPws(KLRzXo{CJjh9L>#)At^C+TE5yRBU?Nt1XoJ@*QiiJNv&iSd@=#(uCM
zSf$W#r4-e8e!9z@7VYw#EHBI4=9*iO<MQO@hSBFPja1=?W(~n^8xpH)8etfQ4cHq_
z8ROc7)2)omv<yc`Wm{or*-8q>uuMv?l#!<9Ofzl*CN?iIWd=VLH4{vQi~#`mwVk$F
zsjYol`z0(?OfXG3ZL&`&wbMrV3@XaBZ+J9@36rBmsYFzQ(c~-zW>eQ~HeHw6NuTCG
znNWz5QVY2$7=mdHDWEi!;EYmj1zI}owxd*FX-XLv$`S|^E`?#^Eyw#`;Kr$0M5K(0
zU>_4M1b?cCPOb_9pg?AtCIzqpE+`PBoeEJX)Aqr%a@Ul$EloE!E%0qq30I3*04jK3
zDHn~`rV&JO#5m)UqSS&1Yy}Q(y=!r802rEE-I^QuKoyEYOQe!B&Z0O1q_i&qa-1V_
z%B{c;Msh?ZC<PRha=R?+9XthJhmVpqvfIiNM7@L0pAga%LgRce-`hnlZ?8jh_?8y{
ze0BBg=Hix}9mJiTTW;CeK{#Z{HNeT_%FX3kl7zi_-NQ!Xte{lx|60fMa5A}aX*Ai~
z9K+8WhnmaQj5|BG+^}oAlM`1?CYW|y*RVV78|$kp6D|bozp|cX#g&r@1{-J2JlvCP
zDL=RjzfnFadp`M5@@DcL@^M1KGRcq2aamRanbtklA4=jVt$^s)e<agMRUk#K|H#Z|
zg)A%oXzM-JWU24_vMfS|CId;xdgF3l2!cu8!d8uygZ4JhFZ+L+22J9yvmV&fT~q9x
z1UR`T%-yLStarjVVWw%CEJ=d)x^GW*O|f?Z;N%`RcPF;L-VTz4ndY~?^{sDx%m1G*
z2R?h!naxea&CN5z{#%bP>5i=Xj#Re(GcEpeP92lS`g1*F-2(#TfLy!?<l;r+1uuBP
z3;zCvFZ}xlp8f1+fAiVT{^ql%xH&#^X1s~RXbTwgNALysIO&o-a-NVd$|H6a6N4Kz
zl%hyYr&~?4Y5vN%24L_nE}%-LWm<-*@B5h+p!%V<jZft5pZUyZ<gaw%2%>1s$}u_i
zYrXi(6DLlrwnkxSS#E!Q=sJuB{_>H@gBQPH;QxTJJOc1o2vOMigJv8<2=&J35Au2)
zww`<CePtQPP7kKC0M9@$zgCMt+Sc0I)1QF!t@|Klmvb4UV>d{42qDJ7fj--lamWkE
z>&X-3=gF^;FOc6S{{m+3JCW~vC!cczT~THr%p=K^_`DE{*g15+&AFDFIt_~S1FwQ5
z6vmhCvXD{MXxC2SQRRAQp`m^xPG^<K;B_wfkBfYc@I!#sH+McAmp;d$k*}$jt^3`;
z``SW{1zkk)1ye4n#r#RmwBpPUV61fzjYdrXr7TZt&(q5DwDugS>b*h3b9&LB@5hm$
zK&de7`9cT;ii~ny&$1=0{d(#HY5*oU0EJCMn4aZvzW-URTvsaqrCnF)iK-*k>dXBm
zi=`O1JIfdbh(U-Y1|eb?U<qLe5QZ=r8GChYWqEdPbrnP>2r}*O=g(65M);~91k|Y2
z);4=S<E3FY;)utV1rI49rBETI;iYy_q){B}o~<?b^v3sk48slq65zTigP>MRl3HUl
zYPZo<5`m)RhSdWRNj*fNZkJg`DXi@&+WGD111{!$x$BswX*v&gTFWad5T{pTnhBb=
zH8}_o!(K=+>W#fc8urr1e|Wvs20ERcsn=(RH&#}<n6riN<J?wWtlq!U?E((w%gcS@
z@F+%6tJxv^R`i^vS19DP$m7<t9ieSs;VD!SF#GU?iViBZA_8MN^~WFNw2ZmbUH(kJ
z-SU0kZ?*g5tRJ`gwyhtXjVNNLUx3i-R)0OrvSG%itj7H}`(C@<@3-5Y-|L5vB>jGp
zfaBCx4Rg`&!y{RiWy35R`o_*$2bbadaJf%on3q$htrM4i3x@U~1Q|(xniu^$Qda$7
z*MAM+9=x!Ic;pCTZU41!kJ~2QKepl0{Y!g$SQGqBZ{niH0Xt(JT!vo*KVuGp%IuS6
z5+_A7KtbJr?x_cclDx=cb!?xF({#?5m$K>i$qJsmv$F%80s-!IpSg5p{K`9C(Ol`p
zo!CFF5U;Ti0s0$Ftjj~dN2AdwsxKJ^9DC?UBY#`DR=lq7KG3qSIlZtQtTaHaef1z1
z`d6$GLI@Ih@DzLoE|C%0BV?3KGa&(MHqCLT05ZT6rd1R=15sLLXBm;@RbGj6_UorR
zoeoIlx1)X>;m)CJuQ{|saWvZ6KI|sRXeo@ifVVqwoNXqSHXEL2y1}z4efD#yziSu<
z;Gj_y<!$phz<hD*iE<o8Zo_9pP&rAALy7AE#x({(m~wdk697*fjcGomgy1f3nVamA
z5qX5Xj*w7Wjkv7F2>7DKtO}EII*;d3sRtL8WtM0swjH=W3Y7vB^`SHCCS`I86d2iL
zBnr|dAWSKuFp22yOQ%_p7x#LMay#vMF4%Un?Q+WO6L9RUa3|%Qf$OdU*mj;Nl_rX7
zh1lf634sgF48wIKH8|G>6}7t7OGudz`V*<mDf5oO37b*wwwtyMuGfWoT*)YH4Dx(3
zEsFk-F{bRJ58iQP|FZx9@B#3gdac{dd_Nmjvpi1{ini5iZYmv&0)N=EOhCu?Z5y$c
zt}Jg+p{2x+Hz;M&J#xo`M{UIz8}^H0I?3|^AxOx<SHX9G<95r-jTlB0vY>`w7Feu#
zxr%osr_;B8Q17~rQB2Hk8suCECQQ#|%(pDprD#wJh8IRW=)b3A54e^zO-*SG{<DUY
zAmp4=AZ|C!&qss)V30PzqSVw%n^GxCQMqBznIvsEWk9cwHl*BiV<9PJoL^UkeD|JY
zPj*w)8BPf(KBnVe`?X)Y@z-+2=riy4<!bVJb)$6Cg~g}--`1Q$N5rfw(zIB(X%yOX
zE~oPrsNlKeoTftan^7(3ZdqB5V5pxLnM@bOqR7iZS%(g#>WsKmq-t?99!Izz&C2*(
zaTy7nJ13ag*_pj5T|T=)QAa{_Yl@V6)KDsP!s+%DDp;4&P|6g-P}=A)#sy=7H~7ns
zOlMsoJkN0*1TBR$!^j&;DY(>z`KGhszt1`DC~DYNq@@J3C;-$sI!<jEc!ud)ro}nq
z90e$;wNk*iVMxZ9&6OVpnAYnE$e8BbubZABLqh<7WoQjj0HSGHk~45E%XJ#fTHR=P
zuC$Ugir7CWWk_jgWe8zLoO8}O=Yn&_IOp6k!!QgeIsnQvy+8<aO)9A*XU=1%wzm4G
zHOF7ns}7?)(25ICo*|^ut_1<bRws!SDCI|<r>Ot{W+H-VY75LpqsD}p)~K%45>6?%
z3@z*cks=4rqiLXBFSa?S%x^YZprrJ|$Po2fHf(u8U?))|g-7%j+0R6cXW4v0)gW*S
z8!zSy=M)qUcIV|}s>O)B792cyCLWWOgETIqWjbkEide`f=0BL^#iZz!4ff&yVl43&
zA!LK<H{W#Avz4ry1^^nmDHYPCy}gqs7k^Yq5u|C<-eA+4ArwLYZCE+Xpn<@oh#(3l
z>$fg@p2vA|zic+!ZJPqCBQPQS{P5u;yUPI=46tq;(H2pM0PWt$PMU)85kRfm%BZ_F
z%k;e^B~#M^fHxI>uxeQR0J2ZP=i!4yu~|R)0(mKUJNZ@eRn3R?w2X0qsh>q92%`Ye
zi_#=bLtGQQf9d2A(vxn-^Eg-&w`kY3<^1z;DcRCkSvM-9z`T)5FUc~i-1|z<1nl-H
zTQ(y4*x$;Cu^gBEa$M#yaDnzC8au1x5d4UWWvZfo+i*A>9@koLcAePnZXf_V*{Cbc
zgDO3laUreMOrZ(L2zF?|xR8=_vaBP8R8nw;y>)<qC~q-9Dc!QPQp$q;mqVSg20Xt1
z@_>R63LXzp2)+OEm%s6i=Z+ma_QpH!y#4k!DlKGo?u~DJBSpcKh{k~wlmR~f)?05y
z>dZBz`207%@r`Q?DN;rmawc*_Ed&*Sl<SxpnMR6K4-KRgT+rIyzi4Z=Y?WC99&d2a
zG}wQ+5&~EV7qt|iGx9sI51xd7hEI}bl2;Ql%0`RdbAxPF4$48M|7_(yVv1iV(F9+~
zB%Msp_SSUKP&xRO%;w`N<W)O&<XS{&#2FB$@J~1{H5jor;zqOM`BG8_o(DiN@ZvBE
z4COhIV>8a&E5<GDnr@gjI#DDfrO*RJT1y+L>qvnhL}0t}<C*G%ptX*qJa^8VeuHzv
z41!+6G-KCOe5PF@7pCDz>9|oXPW>S8<EYt)1JCwt&yd|`w05=0-IJAuNv-L5Tp)$f
zD&DPO_;-E2m!o|6ywl_*oHkDd><|h~72FxyNs-@9`ii{B({a~_aLQR};xxY}iabrw
zsn{K7h(H<XVm6Icj*xKg{yyOT{!$-QrAvUzjX3T88CsqsMUjWDr#v&TC|Yrv*p?8n
z0f5riMr0zIBg2dnP6Z+uo`rPIi-1z-Ya_Dg?;k=@<lOdr4VR$VT-xYfEH#_3|K*Wk
zMorTUo1T(@(h?$0lGGLe%=Xh?TC!A(p48k#3a0#;=lpro#nNbGn6dV?5%KC;Y?E;r
z2W4SsJq2gsP~k+}93n7mcoLV796fsE^yY>@*X`VV<mi#32_FmB@zu56-Mm1)wz|7p
z<Or9PR>~~PGb?$~zF8^#BmXzMd5(N#b$7VEy0VMGLc2w6al?5Rx03tFGwwE97D{a!
z=RMFIrO@!20`AiTRuSW2Q!x)Tz8FPDbm)vx;=s7?L5E+BJTjehIsl!{WP0TB$zlHs
zH{X2o=;qP0H#axkN$E@zaZ-5JsZ*yqC)?LmRh3rh<#lrXr}&G%P5<iAXms=F=AXao
zuDkBP{~3nUapVg+ozBVjsns-1tF-F&(YqY{=fP*uhcD>FMr%ZOrsO*E3iAKRzY|iO
z=8M%L4H2V3%aI&hToj{{rTI7H3xw5tLCn=nnbZL;B%~;!OsJfQ%W+X0gMY*YnP(r6
z)W^_t;|J0kCKu)TXBeF^h3r5F=q5qkz)X{8bGitT14|rDlQfyXQ+*ol<%?Ns`Zhyo
z(-6wK$pC5^K+Tu>w&D4<;eqFCEAX@#eBP3VV@kt@7ud)w3z$`)p@AA2ud|iu+RAiA
zC}pqFv3mVB0Cf9<#DTsa0M;VUUyjT1`g~~#yHeCnb5A#?l}fFI0BRVf&*J8tUK*YW
zKso1BN@+We)Rrj`glR>tfl_+5{_>U?3ON@hg6kS+n3jb?NZ?#3A&^SxSZR%WrX6~Q
z6Alat+MuA%D$9_{G9>&VH4R{fiN9-jfo-^<{V^->lofjVkYQWWv@P|1W?R5)3$`?%
zp@HfvOxISX>*zNPhQl<3%rH^_h=k}Qby=%-I<--jsj2UG-A=c%I-T@N8z^ck2*S|y
zOkW5FTmUGF767Fbfl<nZwEWR1RW!C8fPncy9Jwx)C^WZiry+erpGQe8<qB<PnvQEb
z!gFlJ0fIZ6z&GvCHwh)=;Hmq>-xX7GQ+z>hY;5WG*a<7>Ufv7Arr!-w1l@;2ZuIG8
zl++GB13wR!$uYk*VAKd@9i2C(p!anQ=%yB(MA0Z<R_3d+BA7`NDa)!8pafcFKp-Uk
z2!Lr|s`@v$DoT;T&)K$3UDt7pWe=8@+7z3O^`cb_Hm_QpdalcvbJ7oS&>!@vMbSL;
zh@%ZSmdG>1WcWMZuPiqkTqH|9z~$xbL$jjZz5K)zPeg(%6}=R@`O(#)U*xS8G?sem
zrrL%jr0Y4mwgFKXg{VB&IWDx}ycZlNNgQVjRra62T75Vi6id~K<z*p@CEpezbFhE#
z|MBsDL`?1@A0Vh*XBTBRUb(5V1587VvACUi`@UecsM2K0iCfBY@@lhxbxexe9A+V*
zEMZd;?4ePlS?OT;ev>lMz6&-{Cmf$DY4Co3b&?mz88>OO*=kadcVfLBdvM8+QnF^N
z)#{aiFLP+8yR_8pP$88s8}&v*f+L0I{lVJiXvp3}T~lML)oL+QYq_oJ4Ji~KWZ94d
zDhh0E{qWBE2B*xlZG(0?oi>yD%K&hkqQU#U)iTei1O`fL4Z>0=ae!$WIq+f&7+9ud
z0+UkE;b=S_4XH*1plI{e)$N_lO^!ejY|FBNbFQyIMuDPb+cp9+3SbzvZ5jZ82pq#8
z3{W5;ga|_Hbw8Nv<OcEr@&WQm@;h;PHF<s$njdFUK9u%+6eF7!bRaoaxR2PPbqIbJ
z(eW3IErnG6L9q+b-m9D{N^)i+eqJgAfcF3LLPlkzyeu0fv5*`P4(Em3xKxG&P-6yv
zw!Z^k+28Sn5W<JA1VRWA>_3^*8zwiwRW@a&sgE2reC($t<;wFyq%4yg8=DGEBW{oQ
z0mKe;HHq8%FF>v0nKO#N_1i!hV+=GGt|1uU%G#={>pbvVttDVA1f<l3nb+>PY0atC
z0~f^pHvyDs^||CY(=-smpbM9T@b_Q3ueAm6cp$`QT`B7}M=EW{wt2T<HX6&8Z8_bZ
zU>vE$O2E}LO>>-vp<z>`RC?BTHB<rB(0!R6LBS3C-+^rZJ0=BYDh`lF(aC9Pdmd%+
z84ZP+T=<&|+LLJmAgy(KPA0gQFu@T)0oHww5Ipz~+wSxYBp|C^IEP9EB&c(^QL@SK
zq*8UMS#fD;c_ocavl#w|4|gBB@`^(vtG>Fj$mfeASM44d4gvmmRj<3jc(Ypkn0#}y
z+P-Ufsau<r*}z?IG+;D5a^z~;!cN!EVQg%^wQYvRQ|wwIISZjL_>gJom#GE!UpDm|
zD&-#o+@=YG#n;}v$bg*L-p&R%=pQ<C_V_kuHDwg*w;VbI?9k55Prgu!&SF@vQ*5?Y
zi$;?H?@CI6=cBEoYpZN|X?K#pmkR;<-{3Oz2ix1H&mKCIW!PmRmdpowc<0ue4jsb%
zw>`u~0kmEpuC+S=^;#}@lPN#k8jUzEEgjumSz)c$bBYrFK?5O#AaM_#f?tA5<TSaN
zkZ^Eyx-?>qu#&$I#r5CqX=Gl;=4oDn>14Vnd}IO*ZC8dWvhbPjZ;s0)&DU`z+H9;W
zuPqxI*C&(ZB|qg1lx{W`qt%t3;7Mu#bK{r^fX$8jC%w(wG_9mx%wfMjfJ?U1;x3>x
zOy$VHbs!1{xzJi^>10_#sadPX34mq!bwL5BfaTrM2=+gcAo!kvXt|CuwZVvaa6n4D
z2-lDyStpN^pCFgW&yrsR7wRyB9YRL3lCqR##pcS%VLn+2$yQ}0D_KocuG7$#Wp%jZ
z#3nkYtGc&tt7?7x<tiHpjyd^ii-#{{QD4f`YZp~jR@Fkvz8YJpWF=%s7x0;7c2;&!
zhPKI)?QaxKcWAhngFADyC=;1(T;9r}6vYnd=mYH)xs>wk;O^Us<sM}v#vbLr5X6g=
z(hM`2;ehU0o4s1s?%nzIR;y*V?EQZa!;rP8cAc={d4Pf6t(ztVTDX=)=`lLQA&zi_
zRoc@Dy^_|9Lu^U?$Xc`6)V`}Vf|OohHt80&sHqgjLI@QluTx67=|>r3ORBeB>l)2m
zw3%?6cGI#DXpuOqXQ$m&-sT<-8l#Z4C;+$Z=90_mx!8u-6%v5)fIdsvQfM)tN1-dY
zkd|R0=>J)()s*+OR;xt=Y+-~A_Pj^&g?QZs@mfTDF}~Ov4u?6k9m7Cyy;d4U)X)l`
zi@>?tWGxJ^g%OQkfX}%JTfBvLP!~OQ8W95o!?0{^0Q%5TK5tQKoH=~>aH#t0d7dxf
zmA>!O0AGp^K7>b)<1$#DOTn@P^eDA0T;0HH*KmC`a|;|5*h4T(lhTzXJbIj-2g;P9
z6xx{vpxmG}W}!d|%5WX603a_2A)Kg#ueM$HdSs29Wh-X^Rm*hgCD3i-MKv|_LM7Eo
z6f6{G8<oFFgQZo-H-~73dtZ{YnaY^HA;zZ9_=wip^j&rHQBO+w=@2oT9gmdjoBG-?
z3{BrtZq57o`FtKChV4kYzS)R;%VOLJhe*M%PX0f*yL-p&M|N@eGb%Xad7i6G|0lq<
z9SlNzbm*JP^}ab63|wXS4H+K6Fu*}zD0gGfiIXT)y`zkS-A=UW2OD`u+}*u{*xWzX
zb&`=&<X(wc2(mH(_ZmFz%w2-dMW<axv+hA>4?}G*9*;`<|KIdn>DJuvO~3x@zaAh4
z(_@EN)uI0p=cXf`km9~}yB$f_GZQ~{EY1wZ!rm-E41)ET`3D?ndUX{J78r!^{LD9$
z>$<Ko{C9(8I|I*@PTm>Sk}yz0^(zPiB!uXLeR4Bi1mIm7Uq_ynKd=52`6BsqLg4%n
zb99pBab^cf(^?7hvb+#go~B8$ka;R}rL45(5&}e&t~W(tVom;{kfsQ<qn$7JRmwke
zv1x5?%`3#~I4-kZd|47zQb77q*12jnU&uU-iW$4b_zu%JD`)c#4={nr{j1oDCl!Uj
z%Z17}6ompobXE~o{Qxiyfu`5wc@7|y^b$i-3P`1{wK){mGiF<i12Aq|j2+7J9HdlU
zY$&8i=iD7<j9WHibKB~7ZR;vbLdZerVPRU3Qg2lh1%irwDKjt)ictJ7N_Fai69^k1
z((YAZf>?}T45w3N7w?X3#<VcaD6uSQ0wkw@em4Rw|9eyTv5-OpyR%(?HZ)Aj>X{T%
z02f`u0kDl_lY)=nv5hDOtr}EPNYFsz<cn~j^8bo^@Gi1PuHoD7X5}&zL|lqgI$u(G
z7Fk--!F`%kvYgVbap^FX6jk`4y!8r+;?U~a`KLYWnYUki$JX}N?9o>1tQY>o=O#j|
z_BPx7`*&AP4jxDbS^D~ZtGWK*xmNtydcEGy1`l4_XuWB-nhb}wlcv)Z-(^%q$Btj;
zxqrU@XX*0t#-J8jz1`KNMSD5`HkvEr%(H{!`sL;2xqwSAHx*%o9Q-F<gx3;_^vF6n
zO~@#o$GjYs^OBb%S;;Cd@@yc5oR*7f(ZxC86cxY#v_j{N{6VN*^4|^?6<*xGynh+?
zVDF2Yo15+R^+mO|-UiCfTz`K1JB{AFzv0#E?cKpkPF#J>*|Sq9myaF4_S$2|R=n^B
zi^bxrUtN6lt7qHo_G-Oeudh(LWteBe5I5SXZR6&vW?FZ9r=W9WGz1t9j|MIwgy6w{
zA3TK&E|EI9qO|~BE-DWVL9XZ+&56PZ3og;38q-RH2S6)cjw29$l4&X7EBjxCt^Kcd
z(wf-jzSn`R(LFrW>-8Q@!z5Uqd54@ZJ{9Vx^DcEf$3$IYcMG@#Vbt0Gy-vr)KGz+X
zb)pdVf7)@JXS>#(y*u{0UeLL}&7a3QsQti35y%%QA`bQs4)Ae!71<`&ke?)f4+vfC
zDvi6*%Vuaz9g&Gnrtph2$SjqTiEYJL;bJLGWGn$g#2qwYP12-Fd-^M$=M?~5GEL%?
z;a{|v!aWV9bmDQ`9|8=*+|S3-CeSWw7a6bX0-s#KRpHTNXevI@feNE3#YS-%B%m;c
z2-6UY#Ym-H3`$)ED)j^Vh93^K8IF@>IVkfBF&5S04es_=K~iQT1uk?gcjE+-xMhsc
zvJw!5HoBN94AX8s#G7iCi4@$L?Kl8RJ0CM~qWf3Dvxa?2rH`Cpe?(D*eQr_FG#g+A
ziekTyR0ZI3-w$MTbN>|rqTx0in}P<ELP}2>A_y=D`eqm?<WyKK>3V>c#iR&LQ(K7C
zuhkeVFHW60am)GZk8?`JkA3t<9yJ(OZQX2yVCw#GsD;)o6G1A)|25H^GSmo+1#Psf
zLEG~Hq!-y1Qkw!8nsFfj!?bg%2Zo_76Op5ok*zhP#!oZG0Q6hlLKy`F^|rSmV*n@s
zl^PB&7{EEa;SIn!2t>t{DF6UC<%}W-(@>luM`oGr#P@(vO_?D9B>-cABLpA_&W*tH
zM1wPUxQHU?b`LYg=mmf=lS1|t00kWZC?m_7!hoPrHwdAG96VKy(WCX~jY)khZFWG9
zSDG8CC@ppt)gp8UI7rZhSG^T%wSviS2`=r2<~ZdZhS3SX=lLA=_JBum%|8)^xCc_!
zH{y=4>uuDfgf562SixTHALK~_h8bjkb7iVDq{%32B}w8d4N4|qEEOp2VU`#j&Tq<d
z$g9Y^+p}!J*Oijp!tjBTZI&ytOs3<`P6!8_UJ2;}vQRQCMkR>%R@mG{S=B#_^Ri05
zqc}5&RjBe?sp~b&rfHgH)6{w-K`U_429^AbDgYC-20IgiQL||orrETzdd*T&1ubzz
zH5vh>E@e@gGA)zDrRX5_!;oGsS5KSOT}Q8rRf=nJ9Z|Otg)s#<D*+L@p`A><F4vH*
zi*+Ssr|x;0XE!H^0uBQ+%=DJ%>VFywpjyX-AN=6p|G++cl_X@993fYc=k;VKjiVze
zG!2s?ufnd(VCh$MprR!8cElvfO&?Os<GBWRCyZj?9k3ki&Z<0{%tkDoONwvUHRbN^
zk==>bhJE~o8;;pp>!-P#c{eUEN9fqx^S!mzWAD0iV<X~@jsDWobslH#k$w-WY~vYb
z+j~cChXHjQ#9_9+o(&N@ZQs9Q&2v3((=rU>$RWcpto=LDwngiTb2qt+x_i638ygY!
zufP5tms0=2=`(F<+w|JAXU{r-_E{Uz1{>+IZnx7pEQAN{nOPJ8MA7Vw3m)MbmmxL+
zJc=h9kI7!IUM)YDknrksRV9MNX>e^eiv>MbDMX`8wbe9jqsstcb=`G*I{plJ-tw9F
zWY@CbM^2tR8JsRW55TgzzkBlJ$@uiL=UHa=S5KZi8J;OT*R;C7Ij6RkTUvHqnEX#v
zczVTgEwc+(2CfUYf^FMNw}R^iwr#_!Zw1GRY}<yB#v*ys8-L0|n@IgEt&u5t<`3ii
z$}gri3V|OlJ1}BvR;f7K#)=@W%CH`gTmjk{MENxaNu0Fuc@|nYy&@5LU|E1iSmfjB
z<QW*@$Jl{k`YF1xh@&*dk-5AmoZRsFVxG@GM1@E~!BI)%JZf--$d43;0Rk0RR(Gr&
z6SYx%Pp+Y%hr{cx9u7?dEyGA0z7)(QKA9XlHkt6AS$gdNH=>}HrVKUIgA}!jy0)WN
zEC8S?2x8IcbULaXdjbLe+nCH7ikOW=D{;5Q))=K20GJL1fLE<0E1scJDZ_dm(_DYu
zaHtI|OaR#dq=qpZp1)~%iG4;<YSCj(w-MwTU+%9{%N}mHo=QE}ZtXouyi_D{6f^8@
zof)~>f?h0eMYIrX$axv%<OJzuSpA4*1<_K4xBBO10yL!*1R2YV>9}mqOaY%~nnw=j
z+(B4skC^TBCQ0m_2;JiD2X3|717kccOKUT$8x&ywbIj0LPigAds8x4W*_5iwuIpYs
zi9)m}XzF=K_gt51K=V&+Z-qIO>14TM!r*kRb3BT6&Nb>*Uaw046vO(+$Oiqs*b_4(
zuA@-4)iJ(`PJ9z#+lEHZT|j8llEgoUpF;mYZ`Iu>WTM(6AsV9suJL<Xjr}8>&-tE_
zix*$~+~+>`xl8xIruNT1NCJ)@?#0i2?u&oC$(A02%|{jwz5@RNUqO%;?&(>YU5A~!
zg&y2+@c=ETwXm<}$=#^fUINb1C~7u65AQ#dH(gO}wOUG#Mq3xIyKcVVkY$GtpZ!&H
zb@lM>@~T#i`ROxP|4P&I=<FZIEDBqH&zFiLN6sFbOq!!GI&$RpD|UAJefq11H^;rc
zvwHjOFEZQhqsPE>L^jBkyeA+o4P9C@u?j*15($XmDG#Xi3Q#~dAmg)s<HgAL5XT{N
zL~EUuQ8<32YBq~eqhT9LY12{$Ud0&aHyDOehM~2gabp;14VY>Dq>?M8dWfUEGK|$`
z6QCA>)~2PEGVNOgW9+#~nTAq^`FD%t;?LT@xLA9V&erV&eY4wDjP_IowR1O?5uaSX
zNmxwuMY7xEm6;;>i><%-i|s!@Afs+C41%!N9Su$J?d|RDKmUuZzxay}`F^iA9QJxd
zcA0<J-#V5Md=9*t%*pe~<K$iBXBe)+wPsiuJ{ap}>f`7jGj$iY!I7{C+CF}oIt3O7
z(hQ!tEid*~Y2V5&dSp>8iqsz37Ib=?l^1^9wkSqv#>-aWyDb5yuM-sQ%fNTbV(IYG
zd{NA@foY_*Ex$h)jasW|YJl7J0D~xiO+~>h|6B$^V=)}rc2MJtA$5J9f$KFJ%PV!W
z*&L<K=4jYzGTv*L78rq2aolON+v*2POG_?*yW|46FVq0Kr2%y7r$8kNKqU&E3r52L
zFd7EDeb(*Leh#hH>UeD}T#IAIwB2m3(ir_X08d&=Ca6Iio37_|d)=Vb4iIz@^g8vr
zV<|lx1_4cNi}N(;_Xl>G2GJmI$B_`Wwe%3UOD=%B<iZl@)>B%jC&}-71?tXvy>0*)
zbrZlO;^4o@MSLaLBqztoW8`u2bL1O@*bWFfAVU0MD{fefPZx!pW-x!@(H;)3v~pMo
z^$CRVO+`8_6MDOB6w0nSpFsw4nU8I*p}Q#Z!i6Gxpt@DLC~}$PavGCoQ5Z)Tr$x@F
z3QtfbkrxrH>MPc-F~<o<Ce7+5ZA3Ol!L?~p(Q7#le}51JCKF2|N4RdoQql~=f4}CM
zYbbh$7;2s|7MqNYBPE55w*%o&tz5V&3`2!O&$d_-)IW@yph8Lg$lx3eRG~sGT(@zb
z0EME@0BB-TI`;*mTDhBT!(d8zCk)2_A~a1fUIoTI;AknqNFj>TmYFU$!VnsYdLGRj
z50OF_7yxBp*x-7ui{Q^(!-Ro)jW*n;NyzS#p!PsYy(B^P8(|1n0PhHn6bw7C9Mgo>
zl4F_xModUtVZE4J<{PNcL~-RmT6q68nPq|q5D}Syva>>tnXl3|=ls#<*!Mk6lytse
zKf|6YUnc(Px2w2rb8~a@&UX8X-0^l{W0P)TUOx7#@Co=7SvkiAQQF9KR3p!ULRwm|
z<j%k?PNUK3o_=Ya_FAXWaH11~&CTjs@&}>X+#H;U-db~A=(N|?+8uD+`n5$KC)b+S
zC5iLm8X`z?@L61eFOtLL1LW`EUaVo8kT9>JbWx?9CJi;V1KW8vyEqiL<>RARq?34D
z4&rf{7b4A5SHY7fSz7&*Dl5m?SPfign!4m<J}$?FkmGU^qv>0oOea&K5Ps%3TO7|T
z=kA;iS}5mb<Dv-=6=NkQTzKLcU%blYtP*loE#`|NVuZ9-5%oeO4Wf#)TvWMiJ%}3u
z2qmIHp}W$|i(6cV_`KkI0z^F(7R)yZ_twZLj?&3=lB@Q4aTM!3c~S5mB#thQj`CuW
zJ|ASBjJzJu<)6&TMU|3c;Z|5*;WRFaej%$x#z0Gn6w)V0oDDMJM_UT%Jb^8t5sRsa
z#WblFQ}<xv8B2_x=q`lboRTD+q+3y@@Z)l+`?XpWX;S6I1?n<AU;tqB(s0`rj<OK|
zweWx=T8M_}xu!2T0`yE<rT|~}(=c`&#uP(^6o3N60plKntEXvU$p>)bW{)#OimlpO
z7aWsI2-@|L<etaswwxpd`1=~A6cu_cH|+VolBTHunxQ#tAq6%HO<QuNM3CJCNOr0g
zwnod#tyT#5ao@6q2DJQ`h?<m2ilz^lF-qE|GISJalQT-U%s6T>ju<lGP|B$E`!OS}
zJl{lsJ^+A)s6eH)PCY;bXeR)@y33icyZbLyP8=Fsd8XkS4h%D(T*?_UE~UVz?KqSo
z*gC>-<ha=3_JkQ`NU3Wm1jFQ1DF+Z8p%`O;GW30Jn_O1Jl*8)-Qxkha9oIIwp+PdC
zq~};VWyleAQWI$!Xhms8D@$F}vn)W@Hw`UbDFW%kosGmamTOuA*otxIP}U24G&j=<
z7X)Hdb4?4(u#Old0N$|CN_#K!wT`zkQZj}J+-FQ1^wmrO(Aofi!15R{r?m>g;2dyy
z(0cD*!vRrH&Hz3cQHm&NJ2ab}e(JdjK*q7}wL}mgr^8Z9uT~$nYK$V7CLqO{<w@0S
z<P<hJa%ow%wkQBH>bRas5ddsMgH+1%C4i<3!9u_=uqaY0w3MEf%C-&5bT}Ki;JTg!
zpxAdE&OAgw=>yQ%_Jg2peB4&RgV;BuP@3bR5<pKQ4MHWo%TyGW<v2E_fTjhAq3hUO
zc)%8ROPOccKBacxcmUL7hUW(ogwhn1kW?jQD3yXUQ<IFK+N6+CUAYMeTzRI6DAQ}!
zr%!F5lm>uc)+ur>g^NIKC8W^JI{30}yAy$nq()ZBNpdrJ9(fbjDOB$r62W!SG)d>u
zeM2}kDF%|GyK8##4ny;)bzD-O;k+k)!XKH<W-~XA{bg#Ju@}ewp4KjZXRng0^(&{-
z>C}s3Z-ts>;ze<AY?1^{+J-u^y?8%@i(>6SuedO@i>>d+0iuj%X&4$3(0}Mdjy4R#
zwn;*Jv+NH|r2ms=U|HD{5!=u`2!iGTSs{izA^8U9FT+#tMbakAd?j>AT1mq`@yrxX
zt=N!GsjV(>z|FZ9UI;?z%Gp=BEUc5H*+s9L_fOezFO#26FiYXI0GvNO)Mx*o+wF!y
z6O`nP456=O3{|b3E5~)d+V*|(6TypCujq@0Witk-H3(tf?lc;441%-sz+*R#GXd|0
zMYE;#Yxef`gk=W4wU_6)^pb=jWu`DkzR$xXmhf|0Io+pz{^x(b;=AK|QUk!SJkJKx
zs5e%>ZXrb}`i9Z;A&y(Ec&pRt)EoyO>Jvzi!&i8*Ti7Beev7g~lp)yidG*zpYF^2y
z8iWQJj=Y>qfcgD$T9!9oam5u^9618-4whH8Zo1-%jSaX5+U?EFs*|K<;JQUIyKv-a
zqgnrlyYC)cvAeswyZh0t<&~h`Y#cpuVOA8b8<=U*S!`~0+HenSY+P~0O@Dp&-FM$j
z2qn?MQ}`%cB4^3%<k^HkUgjCW4q~)KAcxcm=Pm$1rSrrfM@*CHq?*J@I+lIrWHe3E
zXu4o5Lp&Wxs#QfeA%j==)1wWCjsPfm;)GNHM-IVO6e>7z9e{5`EdeqBCqH-PRcC>7
zNKyc>eZ2Oh_u&5jN?bSGK-P8BFDI@GX+KU~C-t1&!b_oHg6o;4=YrXQ8*jL99-yp4
z+5phD4=vyClRbeG*2Ikz>LtK$zGMGi!1wEQ-vjWydi_#+aNH(><3R^)?xL0`^QN~k
z|8oWLMnnfLk2cLW!J0oiGM#oi@Mpj<2-xB{Y}c<(Tmq&iAC4Aq3EJ(+WS0&GYis>J
zg-&NW-9?ab?EWPyy0ST#d~q=v;<1P?!`pe}v)T8P=4C45d7SW}Y}4Qz7_M<XFEe=S
zTi^QDx84QMzx(dM^@HF1&HsH5^xpjD-m6~qs{7%3yn*hMFa8Me*`xo>!NGuhaP1A`
zPVyjm4tcRBeF*1-)kx+vfq^@UJEp|7YbEv$U|x=7A49^>G$W0NpH;<Z?yaW_?!aBg
zZ;rd>`0?Y%-#_$E5X4p-MSSQO4x;O1`&SFoY;8UJMK|2A95kCD<zVl_EQLF@|IyCQ
z@9plsV`t~=;lqay->(p9)-s%=-L7dY*4DD)^V!o-$sRWVbZh6Qg|mkb%gXXk!B3MT
z<OI1&^6ZN=^}9lQJOH>3$PNaUNw_#P@Ns;N`(+&3IHagVjT!(Nv+x2_oa<SKr{4bd
zx8Gjt^=fq`QyNCY;cTs5Hv{Uq?V9J?c7j@|ZY||3it3zgt*rV8aNS^Yb93__kVdcZ
z(|B!lr6Gj2tw9)?pxOwP8qG8{Cg;zeKh|1WULG$mr_4FY>}?o^1E&#10G?NiJ&y^c
zsAZXHr_*h4esp6!K-ha$v)OEdTFV%WzMYQCNx$7nTZ%ekJh|TJ_p_|gG;OZQnWeU~
z&b`nT<W}-1d20l5@d8G53YR(si)47Lk?Ty!UE3B2f9W{PrW3Pg7Q!ww%LW7YI;t8G
zIN>oUX+HNNvY%;3krcyh61XmHFxLr&zK@MwuYV&*=`6QeW_LL1m-FQ%i37je$+<W?
zHd(FKw}pT~)@*TXV-Svnc;Wu#Aik$tWqJ8}!j4_9_kg0+O%i7CB<VVy%MzvJ{tt~-
zD|c(Pw8oh8ZYP`3AO4HHQ*W4-`G)=uGRE>oQ@)5%isg`bzVBGy^uBu9i<~h_jtlWq
zaOqeW^aT+9bTQmmSb42^Rx`5qjGHz$VSVk6kA3X=Mb&J<?y2X03@*LsMK7x8=J?3r
zaY^@}*Pl$cZ$Ekph&uT1gRjD0z^93419|M%$ln8jeq21nu*&9A6oa|!_65CdLF_T&
zUtGwGY@o@-{t=H$K|igXxZ^W=DGx~_xo}$AECyPt)OYTRY*3hNA&%h6d{!mX-KTJ+
z-ks-#uwm5cy|9v-!f{3z<CGTuPt2Z*5B<<c=TJt=1-+OgQEqDvkRs><!BSyqVlZfn
zbj#gF)3VSoXy_V3{7&qfLV#^K)HK;2I*5)7;C{dba2-Skj;JsQlZSvYe>?^oC`c)k
zVIYE*$QU;)TiYkc0AN`hEE@qW8>C_2nnS_!(DmDQM^PAQ;ka$70Bpb-0HCBG8qHp7
zZc>V97@SE72x>QLB?(8A2mm!KM0J9aHQN#U|Cpw|fyI$Zr9=>fff9_v43KIrG)3xg
z0stzRujiKSK{ri+Q7MBk2n05Xtbn&02n@&MBoqum1qCn-L@qdg1j2d*U#I)9&zSL0
z1fWgZLZGM&q=2SrA{WW$q&8R%?X}PP+;I#>xjWgfj)MS@-fc1#_>L2Bp^Z4!2Gnds
zNVyo=u4Msm?g)yYmC`F9WCVzI4YYP$KNQS$Z1=-GE-1QIl0=b+uUUuGN#Y*jJYWFK
zR|*7D$!^dTso*^~Nwg#~;TRr8Nn*vN8vU<$5#C5lQX_3L#g!iW<y0qIe8vzCF(|(3
zm5Gev(=JKbd;SYKE&ZEszWL_+uR_@Se_6fGJ_<=P{Oc0;-vs^s71MmwxTn#(9^SbB
zjO!nN>?#dOIvS-3=)L=tcjKZ$>~>ei8yUQt5FpOMWthMh$&ipLp2yQN*A17El8>gy
zN|9%I3RAV7>mIu~c-Z%gx@E<&_fvTM_`ARU^{;>ZzCrH$cq^rdA6&iHm(PLU`ej3M
z+VtNRMO1R%<vC?a#ASBc(nQdt^H7>565yGqPGcO6<PaZ^N*k^bNx`o<cXo3V@zkkj
zy!+g_bK3LlLt0;}q+lbS6ZjM!Kk>+7eWOpM47F_t&IH$p7^TV1&Ml9eIF6rw?6Jpe
zO7)r$hq(~?+scj*LdO+jZ3vF7k#%y2T#325pd1x=byInnrSZc-g8V^y3$3CB-}fJd
zS2ZB)+2f0sf9B$Cjf+2X`C_AS+ikZs_VzDcym)c{B9Mz08y7D&@{1QQUTj=^@8#R}
z8t^k$qw&Ry7xlVEeFrYWTgVQ%g}jJ`wvEModxKJ3eo4eQwn`ZV?oRoirF802KL5@;
z?{u$5x$`S+13-3icz9HAz$q@QfbqbzFCOPDZ+T101l`xqMv-(Z%Q9A04XeZR*p74F
zd20Go6uB5=)O<gKY5Qpu9CuxXLK;TUL&W_b4w&&SMfK8?3@NYy+`hiPZX}7ZJ~Hy$
z=@Y_;I{4~-*zc1J3BHXr0D5AHrlo$6PGgZtrXqN~;$cdh7>Akm0eGa5Av9OWzXYx(
zKsb-4L!ePOj>_boO{wX@wAo2Kj-xmMZXb{1h-@1SX$pb387H^>lQ=5VGFi?aYZ{l6
zIQd!Qv{@!hG?pZ3RF2be8l~$gO`|d?$8l07aWaMB>pq>7lj-DMZake#C+T=PR;T~m
zmf|rx&23(8wVV^b+Qsx3ql=*<<LRVaOo3Q$yPTA>YF3S_aXG8T)vTP%XVqjro{y)q
zYBr@PNXOiUMT_~ARcGa*oRy3DY+9AGYCNCJt8!e;hT+L5Q;o~<d^T4)o6hFrdz?mR
z%kZ-~G{H<+mE-w%7>^=_^qt1#Y(Ack=h1vTpUlU(qsQ~vJQY_JeplJP*Z5#-!myec
zUmF@eW|2;&^Aec9izAI4JE<1aNgR8}MwGIJ$#gMaSf#+d-Z1*r#ypuM31bNbpf604
z`}u?ceVj5k@Q!(vurjMP8nxLnMJX9)OexCNqBU>e23O7U{hhGAyj%vMKK0xVZ;K(L
zPdU#+C(FyNQE+m-(c8?r;Y8gf)nNShceK=v9F|aMk3|6>0ZkJ@8ithVzPvJ&f!0(C
zAx#57U>j({zW+`7-#}*+D!qcU(BlFCIwt342T%wW!Y-^eXGgEc*Ft77%(DlzH|zaV
zoPM1ze9h~;@~y=ZdRR0Xdh9o#8_i|7S#Q1}y8MSAAYJo%KRqlWi09wHcdiBmy#DC^
zf2U#B*ZiuEHVmVE71w?@tev3%wDI<P2OIqyZe!B6CFuzQE!_nz(+=>dG8BNmy@8*y
z^x!`a{v5snA0jm}I-{d?Pb_uP7N`CJ9k-+F>-<H<jhS#wXIj(Dq;Q;&FqW-$x8oS5
z>HFw@u16q+`eM7=wJpo^TVq(}%1q)UF&)k#Wv(yo*fq#{FWBA1rr>ao^nKGZ?QXZ#
zmP*JV2m;;m4byab-FDe|#Mg#4!YHZR4&|Z-bocNbcO2fObocNbhsLHNgb+&9!B_As
ze2%=Hyo0=#5XhxS`FrTj%S9#k6OlleiidQL!HL%uZhj~+%n=w^6!e)I`2H(gBP^uz
z@=-hcRr=OC^D<dyvZ!W-n{wz96YKaiLgIp(n)<kR>}i&H{cMUb7;GDYX}Tz$uSI?&
zMew<ywF!`>ole47((a@+08<;9`+(_@ya4ESr<w_4doTc;&aOADrybivpK6l_FdQ8@
zdbMd^ee}p^2*FnqOuamO^T15t3WG6Yt$gV0S@-409*Xz&ZaKBJ)nnS=?djy`nX0Pa
zq3e+`vK_`OH_r23w8HOfc^<nKV~#yCqPo7LUM<e-O{Q&bXx7`>I(_rrUi=WFFLTeH
zefnh0V9XdVuGu?U6zrV<?_@=BbnlwQ_@^LA5R(`;0jG*?PxdKzJ6s~m*{MSSHPF4i
zRpq6KUE5@qOyh(ga9*gM!AaY;^VwED0vP36i=`vSZa*@av?F1DTu#cP<JGkxIQEkD
z$Rm%ua8`qQeRMI)AkVkA^Bjt!$2<ysC{eVMHXHc7TVc6d)*Jec$0TG-UPNjZV1l6<
zxVamRz?4SwvYdTwXJyRdkt<%Gu?B&ii8M&$Bf2n?<B*QfJ^k(qYfS0Xk0PgOn#hx+
znS~*L7auUwzpxD;!>Ci#8-@hzbf)Vy0BQ9b&A!4WNEQXgx-LMw+44;gPP5rrKkh%6
z?CJZNEZ>Mf1<&6c-?jvv>j1>jFiCgXZJ?_F%(ht?_oR?&fZ*93rIgWb7k+n3TL+(k
zr@>`%g4{-0xoEtY>v3C<`}$Uq6tNcxd?f8O70@J#qq#K=r)d_UDljhT!-)WJ$7~f5
z!wA8eE-y`e>o7d);puDEHz*kHQgL{r-7x{yAgygxTUEz`cP<YGg8AM`(&;oa+ZL+b
zUaZrpWf^AKABZSAU*z!^!?pEWhXeXouV0eXarX^p&o&xxcy-OSEo&pH!?U5+Kee&E
z92*ccR#w&)y}qvOJQCS%IT_KAPRp>&q^wBK+aG;OLz<RhS-^#leB>h^`tS$eaQ*ey
z!{Yk=|2cl|d*6Hg`!Bo(n)`nV&HcZuz2z-$c?)>^-v@91`)>jFt%NXAKe!Bk&24JW
zBONk|#MkAe9Le0ePy%NlLD-Ys8yN9XQA#-V?Qee@YKQm#E(VAJ_7C9acXoD;9Xsr}
z`|ki}xi^Qu{f+N@=Q|%f{C8$;|8&g+0_4vIgTdah%nU<w|NnpRgC9Kno!AYR{wnOj
zr^tdlhmdM^wh1d=I*q6$K9uy&Y>OA6NLodX5?)Jg>Wt(SDSL~e9S`tLA)CpzYRuL(
zE1u`NQS5twp2r~yo84LsAdH%uo(F&!fMt3#;zytQeHlg(!mT6>EH02Kp=djfjT#J(
z%r8R6WeCs?1Hd0^)M{zkXmp2-I_13AZnt~G(fY7akFzMUlwho$L{S)1N~!C*VbO25
z!hlL?w7We|qEw!3yKWTuz85E{>w+0Yh@HWpuxvsCf(Mu3M{?PE`W1PQJW5^%SHRuy
zY<M-i6SwgyynqkkbMPX*6Mv?j3tIBaN5}chNkwA~7FT^dfaEO_G7)rqo-~~|O3U&!
zm`;FXo}^>MfxxB457(;sxJ=UO00)Icl`3fSB7G;cNL34(^*HD8?ICVtJLV;xkK$Tv
zc{U$s;m?q?R^al7&xUxqMI%z>{5lukN0~T7i<*y>{ER~gs$=h`^7*HIfQ><Xi_{Nx
zzTgWg(C29-DJns}3{My4^32r1oO4hg2!o5+FQ#M(aK-YZ<4k0ycLs6_Cz17Zd{lzc
zUZ)arOWA_ZGiV(%0s<I#w#6F}1m5L7y(G)3bg@kfK<w+-9EeK3k{3ytM9ehKG2LJ@
zNT5mt2xI(X6A4U0r2s@+T$N?WZaY=~^Mhhz1<}yM9Pbl8WOyzaAuEwBc+4amg?1zh
zGi^T~quqx8)#5H9Ehb;=YRP247QXMsqnc8Rkb-I2Do%Pr3TSs7ZW@*Xpl=nD)u=3>
z>-)mSUN!?EMK6h!ZJHP!p#aJ<4Q6*=+itfNQdELV!A%6D;x~CnfpaFWQrEHpkm8q8
zpnyVBYAQ=qI^HnBQkIoiR%qLtaZ2r74ceuEEY=D*fHX{sz->FUtk|;d!@&gb&Kr*p
zCO)I$qPddVqmU>?N&^)dgIDfI(sWdwomImu<(XAJ;F+<R(V937Iq!Kh)7Hfmf>NSM
zAXtC-7h0Rln%G&5q=@p6GMx))Q3}x80Q;n%$QHP=Oc0p!_y-G_IRMtKj0!RCv=GHT
zEcZ`O_e<=>5hyl!QADNF3_$8l@d$ZHKoEygOkm5w<LFdHNd(Ta649F9PqY?MY#k7l
zu<CLuX)GvPi<ay4a#US0PD^Fq`@G1NW~$%6Ce|u1&L0tNovU769f=2074Lp^+b?m(
z#!(bo<8F_~V<rUjnnklgLIE<Yz0NYx9F%6x&pVgR)D#f8IA%m;ATJ2ei1Tt0oC6lA
zW~l;e$ujLV8>0|)6uSrza<~xh!~jx}{{<4qN~xGh;j_auFA5H9AXibG%L20kLK8&<
zA;dvzx&o^xrxZFc0*a$x5TdY4?GS0_g7MCi5k?mbmB56E6&L}qsFYKpROb3V#NH7#
z!e}`QV)2=Cf&?O-iy}{ld3Jafk|b3*K(5jx31^2{4h?}qGO^#VD#kjU0~+T9P>T!V
zSq4+>=p>J5V++TkU(r4@fD+udAp(oIga9Cr_qB~95dk1$gpoi?h*&WtX{3CTcon4y
zvEnF*20|8N2*BxLYHRPw4ym0u$-w>=(+_oB3XAG?{kYo=0HNC*ho-B9rL61FH=mIe
zx$Zf~3PL$LLkQlbOKh4nH4MNeO`8V0J^-ff#iP;#D0Z%=i#+=+Fb0ePBlzc}NXHr!
zC7KXAsuNzg+xa*6(M|<DxDQ^_>FgA8O-S?hyqyP3+SbgKQtvmN&1nQCP!YzW_Pjfy
z+POhEkDMeck#I)flDIja_~w`8@T;rZtUT$(yKSYrx@Nv^CmWtmPRbxuZeiOjTQ#_j
z36A!%JP<EO!YCNqFA!WhrV%mSwu^Qvw8M0*s&un}lNK_wKwcHlkVSVI(#Alq?x3T!
z=CW~*gd7){QCCyeU2*ZcUEz-I-|l`nj$=(l%XArVGFi3<|1XLn?R@O41<)GBc{YSJ
zm02-b3n?PY(3k=<Je8FVnhbWkv!h-YJ%DkF1uf-*J*SCD45BIvQ2_g?rU0yV3al_n
z!h}MKm@HT1SyC1EI~G_cMXbXd04q`|H#1jY2oZf)Ajk}MZ4lQ0UQIuG5v>%5qP3C&
zF_7eo#bS~FIZh`R7t;yi{y`f>(LE%B*WI|t;cXCrvD^9}L~y;=>pd(=YbCGhN74Vi
z`OR-m(^QmZw!q$)5GRfa{t+f|&KM3u?CQ!pOGFOwRro@uW7cm6h7m*@5@{uhEi)3L
zI7fkGX(7*%7zGyfX<*7E24EWlI}M}}#R?)DEl3K`s~w2Q$^xNtSM)2t9Ex|El^9Y8
zEbNgcT7x1aEJ2HSPW7NT5loo@KAzq!Z!8SR2@Gom(B6ld32vRBBERL>-sc05E#PC^
z>l}0*=x{H?jGhDsznlMgc+u7scMyP;eI`|~(wlqTZ}08xJ$y8Fo&d9bI(5*|l8^bK
zcm9HFP3-}qBfMi>@4o06TcQWw>-?g}>}WFW2)2_<g^+Y3K7P9SkfCMdr!8^q(7ASH
zdlaA*dQ+ZGgsU?7z`KpS3CinQhgY;qj2c$d-NTYr5!Mo=Ip0dZ{cP^YA>J*U`TS~&
zg2=ShY_)Cx*VyyAss(=F@>AfAm!G1_Vl*rZ4R4~N7z~SC9Yw4aX^y%|H%cW+bJ*5~
z@IDvHm(rkBqEw?*ck(@A&<M_IKrkjSpo*ADF-Mi6iV!riwYt&%y@cPgePAu42BcD<
z)=|W)h&fUS9E<@3&59U71H;`>QL4*NQAIu+7KMW6bXg29-(5~xnG|Y51UDG;Bj*sE
zi~56sBNH-2DK;pg9~sPObkDP?Frmg!R0>KVh-sWO-Y+l?Sb5`oBd#$*FaoD7)krMq
zGq*=0BMQeIr6DeK`d4tbv&RMz^<r>QB+&3g;RuIvq1Sf%X63EkOfwz48#ML){vMok
zyYLQB*ZT+K({2~u;nr!lJ8s+l-p~&y+`a$gxZ6#j>UG!mAD=YcwCwdZm%mi?dfz!A
z<Y(sEA2x&b2!Cxj3(Nz323<QWFxP0CH-2r0Dvt_-p^sYIovYSOTbsduq<#+lrAX#O
z2x%5P;PT+g?&t_$Z@d^lhM?mHO;OB;<8gzK#7(Ua(I3}I_^QW8rRfPr_ecEg(e;ai
zCHjyp_)5&nJ{%ohTTZ9ub~JXDb?I?*5BYdh)^#pI0u^)<kCium6%@sInGY~73VZTz
zecpMh^8uYtL`G;J|M=a~&+XE3v)=e`u~VGswtPc_t2hFS#yOaF;LUb9?db{x;}kES
zhPLIi@P{v5v;ghmr7vk0h>P|mui`3?^E5Pon$W1at|}Cuz|L}}lzSAPjL|eMkE3AW
z^q>FvpU(xHjq`yKAN1<o7IC|K0KY&$zx5+OANL5n4=&o3ReNc>z<(^2c6nh5C<@mM
zf+N(AYdLO?jG>4XAgDO@4ne8maD2pyNGpSAADCXX7GQs^7<h!scf#%Nk@aUg9eClt
zU-<2Mc5v@(ussU(cYx$^X*1hwH@h#}c(4tT6a`{Ha|~^-=WEG=c>~0Hf1*<0!@^mf
zrb!N&&$BL(HL%a+a(;Gprjx_M1k%KL5gk}lg)HkTvV6R#i;aHe)d#971Atp6q5yqb
zRjJon0SFp1Ie2BNAlH?NAbIuGWmU82K@cJ|!(lV=)T~FNljGw_x2ufi*hZxC4|(v;
zc+5hbrqlhbKI1>$dAjo<ozLw21H>?bQ@9SVf=A&U;M3si;Vtkh@aOP9SmGUg8s8H?
z6hEuO5E^godixbZ_i|gAudWHj71%GCtNG$jc&t@cDtIf~4Qv}pf^2x^Ed?R~U?s%4
zz-ttpixT%Ogg9bE#TBCK14>_LX1eXuNM^##@cC6)Jte*fH=oKFi|HJ_&C+r;lf`jS
zSFCE8&A0RQZqD;<yW9lFVJIf}JZkbh^sv@^*q_Ih2ybM2bgR0WrWDs%x0O<Tr*5XP
zE+nZ5M1TVsyRsR<AyX)9GNsLQHC+SOm&2n7{v{__EKw)gs64Rh+E_{?*$`eurJOM!
z<Z-jS0JR|u5_x*sMR>WUu+`e|?%mWmy9?2-(aRK79NC&6bwJ)=xh+c5g=`^O;ECY?
z&=DEZTJX^nD7uiI$uTl`w}dN_8h6VT{Dn$2p%x=x2t_4O+B8cJ7VR-oo*AjLOme|a
z41_drP96pzK$bvQ6dF@hnhk59^KR~IMF9EoFWM9yRJ12A#;|IPu=swz|E|?Um=$3&
zw<s)%8Ufr90&1g-R#Jut!22v{Mlptz0%fi5b(_)A;jotxCC*`#ree{$8p(vRKj`ll
zMFybiymsEJI?ufkQNOPZqIC$`Sm#6&f*v*ikToelQ8_EDRRjouwW9c4e41((HzfUf
zF;}hyW-G&HKneulJOA+#y$r!7!q&6*!I#O9RmG@V60+CADz=+Aj#)#dAX$0gK+9t+
zSn$s3!`BeQkie1Jkc}k7qMc=f6cNF{>L`juX;nu-81Q@9IQCgpLSi|~@oDDDYN#=D
zs%k=-h}Nka0)X`)cxH_l{M|;$w<Hork=58Zg=tat8IXTk5iv%p1VFKcSsewzxkRA`
zXflycP3pRWez!4CqWGeSgCItOei1s?>kSn(QBZ`6MU;t@%98bso2OZp`!1;%fSpo4
zj=Q!hhe`{RLa3W2EjUR?c}GeSqE<d8BGx|1L;w(*u2NY79|xoXG%2l2%+l}l(4?6>
zRjclx+Ykve)XoKfEYG8#vZ9o+S*|RK)H}e3Q`Er+aE@h7p8H5RO$VycTeBC5>OHN`
zPEPho#Wpe4__FK)4$ImTfB~iBkY(wc2~;a{64Ig-sSs>pfR)lzrny)@g3?MIXwqUt
ztf4*_F$_v6eWS={J<Vbu4S=BmAY|U`a)8*=Y($z^4oE7>AdX|u{<Fg?vM~r6iLj>u
zwCJnktRheh;iHw9qpS=t@p^zmRVOiPrPx>zMB_~$gqKFnBPcD(`4pM;eMF=}u5}g%
zQW=^!axC5g=$J_<6#;1THcB}HrL;oJtPv68Jg-O*AtYHI5h^sIl{W^pwZ@V-u+9^5
zh%9P^o;AZ!5+VqOyrN(L!B_+k&aaA60BiygAStSq!j3If`V;U^@V$P6j(fcGQJoIx
ze(^vU94JM(J#Y&ojgf`rilf%c)La4&i#U{$rD->d5ZD!-&tS-!#oOsj5l?*)O*osw
z2j711-fPzm4*|yGn>X*j|Hk#B;oN#}(QchH7vlTOAk3z_-Em^s{M2cdX8S{}5`@G1
zj*jw7$EVBJyylhHdwm;UT(s@*d-JAgD$Rgp;gGD>Uh8_FO}8xJ8?R9L^-iC4G~T<u
zQvNiH%O6_ydjLKJgBbEG^+Zwhwh%?tBtmrFqwe?i_Uc!><mF{~<?#CTd#|KQcgnip
z0WgaB*gRSgyIQ5jU;{e#V}VnVcQl=G=cV~O(wqRj`AEkQG!R5pGlp?O8UKN82KYB&
z@_39X`iIf919va)`&Y=aVJ}T&Vu~70-|<F13F846lei|10McxDTpqZr02sJI5<^k!
zCI2}6<yRe~2{@mX1=8r#KJ7j_F1v_@*UZ|+Nso*-V9R+qveYS`#Qjl{ytcU4S#ayT
zF}a3FJc=(=+g;-8U?;m5yD?}ML;v>TzbzrlhHpEWY=HZ6BZpuy{8iiH+w>q!&dTzS
zCLPC{DElew!1s0zI`?;;=FGu<9ed`|2?>`byk@d>>k#Vyt)u#?l2#qCyC?EPq%xvM
zsra*{>Kk#-!xYw-ypJSuN0}_4=IOlo<aRc24e2ZkKGsP8k#1$p#qJCEzFW6$U3uin
zF<TAo%?H-jSqmJ+{W6N7D7)~EBcKcL4ij&z3&A=nq5$_nRbT#phgSehG)JiF5zs|H
z@dm67!NR*H#9gP?_by*?<+y5ZKCnK)I4*zDm|N2++lz}^x297z$Mb79Vb4G0_YzI4
zvzMkYzj^cK{RCaM4EsQ*V6C;#6a~al)sG|j-^H~TuQ8~0Ej9#etWT;ag6-R!&#HIv
z$pSzp=Qq!1b$ji_>j6%<V!>Vj^>SB_x9&%1>OH`-xtKDJ;9b$4U)*!^$(8LU6a2tw
zwq9q`1Q<I>+P85t`+nnoM8T&BC?~9pnzz=?oiiE=HyXR(@>|@uUvt;o37~13CeMpr
zuQ2&R8si@f+29~IMXy)nIiGU7{;8&Enxfk)^4#<QvaJ8H{VW6M+dMCN-J&KPz5N&d
z5Wc0xCSbNlLLwWow_GLLZ#34-mz~2i$H&L}`v+%dlVtD46^<3^xUrWUoSz-+?;qbi
zK0ZD;J3H9lkIxbI##jF8Vs8w19?usY9OLjACokNE|AHUwd}!xmI-lP8zFc6+f3x!k
zoxkpMVBIb@lSJy`MLUaWaT#gbb|H&t*)Cdc?;(%tc+;+vtxMONMLWS%xj5Oh^Ocm-
zMK!4w?PBWl8o^%Drd_N`?O*8Yc2>K#ZYc<2sYglhCDP6tzey$1ZrWWjZ|0dZ-?FM*
zREw%@7u7;TAX~R>yQr3=;o6qCCPqsDU4aPHts9g_3!Ao?#5aOFZkk<NwJrQ70KC0Y
zPXhov3;^)7Qg1Jpvp*_)zF~LOQwlCW@ZpCa`G%_pS0>;7?cYAR3IOn=dHCUnU;E_4
zH(&IYx4h+}E)EV3u1>C)haZ0U;YVJ2bpkIxEp>@@aPzC*^~%q~!Vd9)y}JBvjqeTc
z-l*SoL+ZT%-to%StMO0&^iTK8pBh|#z%$Q01J}yIZ_K+#u#PV-F51nN$(?7PefG(h
z^@l}K44x<k@s%rAu3Y)NC!Tm>P&{!Aq}=#oN1a{gk<N#7KBs$V<Wp7pu^I1CXU8SD
zFqDRi+QLkAOaimwO<_83Ziau1q`UL&3ajbwNU86SH*+`}-Y95l@pdv`0p&#;c9`;)
zy{Hy*sF;l~Vm75V!U(9%?y_y0Qk${va5%)MvBR9wFV5$4O6QaghhMXgL!(*je<qdq
z%FRXrr<WQHIF0l9wb#yP2&=27T%<ShYQQm@&2A@N`mgJ+zn<Y_%-G#>#;#}VdI~4^
z7`wj9rh-l3B)gshvnkj>ZDx;um-p{~22QW98Z;ZlTaK-)BCf6++g@Has8V5xJ@TCU
zewwCh<@vjuNF?wA;*mY_41(=}fm6_ii|2D52G-yqlS9MyYLSm&XEB|W<)S+;^CG!!
z`tlO4VyFk<Bm^kPwtx%jwU=CT%{5vXW}2pT_^Vn&H=FrtyAAb51B!yCDTQ`>t%m#G
zM8kxu!O+c?>$0w8*(zTeMiE>M&-q_<*A*J6U(25S#Ux3t9wrGuyS?@+JgjB%>a@|g
zdad1tt7!VP=RNOvrun?b!yr^bDu@|SN>5u-Hk(Vo41^GZ1P?C5=ioAl$TA_|`@Ssw
zvM6O4mIaseCKr-emcAdODg3f1WBA<OPkjLPzWd$J^?JSD(Yx<H3VR1YE7B2=JMOqU
z?&8J$Zv{aR02v;Ts`?h#+`N=MxJ-7*mE=}(iDi#)+Px}Deo*piMB%PMP=Lfbv}0eM
z&8o`EQ8SDjXHVrALDo-SIE9n{ro`dSfb=Y!uU*8vG#2*N3$q(em|&ESj<Lvu5EP6c
zMBA|u$}W*eO}VR@DVXCJrszJ~^Iq?HudjLDEuMGF005PmNulyR&?%XjLICDE0MxWC
z-?CshTr7q|uq@xQO$y*R{A-MBsi<RUJC0lruvQQH1jlngAsj^C^^ObEba|I)n|}>q
z#I$g4;2C`jXEiSgq1|bZj<2pEuB{#)b=sXU@#>7ZNL|jFX&N;f^=|qxW4iT5GfLA2
zXC6gYem?4}xB{Od4N{V`EEBLP2E|-y$qL94G5nN`4D}xXlV<bZEre0L3Tx*w#q0ia
zv(nh`5Y`Z_ds>Ui%IER2=i_{FL#uPz?eUbSJFOcQbA;u?$4_i3yS)0UlgDj)-SJyN
zu=6c@hkpd)ala{qvm)hS>B!OTha`X;O8~iar#R{nF5h|lxF=mIKhx4Xe*Dh+S2){@
z!xOf>QC@xZsmMLDcWm`v0KoR#)#36=v9?uhmVz%mcVQb?m9beL@cAi6q~)xfPY1S^
zV;=?i<3Il6KmOZ0^Kj{J+6e1E<9fDh*Im5eZOk$6ZD`ys`UK-Rj*-W46al7bnj(wi
zD2B)Oc7G2p?td8W-2ZUPFbv)P{|%)Kqy34tp|#O|tf`e@v}zq~7-sADKI;Yx!O<0R
zh@3^Ce)=;0|3j#dK0O|w{bFc74tt52;>>)C0xomixM!KZZ(1)7<2W3QdgJjM1yix(
zC9embZ(5e=KV?w*jUT`H8;5${Q?32)>!p*+ER8t)667cJ%_78x=!iKoCbvo85IpW6
zjlfqI?s?`jpV_arPM+MV`p<mmq5U6x;Dc92tLfTT?o~x3_z<d#4?lcy{N$5Qe&7R7
z<||LUxsv1K&;Aa4AAXK-wpg-Pl6%NY$eaDZE#<|uqKpNqFVVatq0JTq^u|Kwlmr3-
zzxcznll`JQpA~a=>Nw<TnB*k;R+9N_K9GcHK}R&@&G5BMFUw1VEX_-+pks$Xuf5}P
zR=`T!ZpZhpbqhK!_x8#Xi_Y3~y86GSdwUbZSTD<U!@PBE&Dz<y;)<O^_S#ybwz9Rg
zRjt+<e~e)i`^x7`@M#jO*D>xZKMo^QNj&A^=LHla;4Eo05_&~s7|~Z7#s|+YFE1~j
z|De(6bPjE-3z4RRr>9S!?u^Us9qa2G>$gF7Ty{@^+;Bb5-H@>VLqKlYw(^%e)p7-(
zY}@1r%SzLiDKHF>3T+$ZexQ(NQU4dn4dg9^1cJcPuV~$h=Gc0bXuw7`hH11A(1OKv
z7Q4o!j~$hz$SN!eNIDvtpA6qWJhwb=%EYMAIF$TKY*9UB=pzz<YR)y%O-etF^i!06
ziqc7JnZQ#HIG0)n<1qYm7=Ahg2tTcjcSrH+vSk8j-ECPm171Twe)D;ZlYGrG0k(&>
zg}`bYAc}Tw!^3>U0D&TMfEMC2fPJ0fBZ!Y6{RV}OmdgGb2jE}ZLPXUv)2}DyGYsPy
zX7Y{LdItYm+p;+4+|WvaR3Ss4#{dp>EQ>NJbqBN#8L!7Nz!Aa;A$zz7J)%gRtdNRa
zN609fW^=Mf4)obd1G#U;NOyDW#$??QgZDl3s-yFK?d?Bx@e=Os-*fTe#l2qd(#4Ax
z_x7H;eEITTuh;8c+S|K$`SRsn?{e?rJ@;I^)Vp}`o_p@O=OQr(AxH2c94CUb$u@Zg
z`8o0l^84ifkUuB?Mo0w@Z+SY8=jMu?vYe(9;$rs=EaDS7a^x-V%*!nA^f0Ttlr3JV
zIi$NOn>`}q%*1gy9ZiB!>Ab}sMu|3yi$V796YwAK(g<fMCIpF<Pu6n9Wi756;<$_z
zjDz`X(GpHXHYk#pWz(3n3v2w+G~MK1Qe+P`8jZ#r!YB%$fu;r5HXAvNnhj_`KN~|i
z7&M-MUa#N9rfK?~X+i_L{eBl3mX*ZyTAWx`13naMNS&pTqhiBQN>dT08d7kUb4SId
zR??s%`1MAk(SUonsTqKs4IHja2H*@w&N<<EQhJ`)ltCaw5Xk?xv$NA~9IZEgrO{|K
zAh8V7N-O}Y(HJ!w0FCBooDBxqxX~DeK@g5OXRN2u9EE-m4I7PG9NU&<$8oK(9)Jm5
zz^3L306Ye=CZN(V0Rpt33(#zUFCVE}S<gG^Y9(|X>WK9^XrXi)>Hzij%rG3oIAofh
zVbI%dyKVWd*OlJ0<lx}otMCr^G?{pT*5}3H%6^pmCV7(ldA>2nv|tmClX@<-%QEr1
z$hNHX5^K&6w#zltO=^oRNciapu=}8=(8URL7WGEntoeL2Fw*}=WIA)Z%ICp%DBrJd
zQ6=duwA*WUn9@nIB2^e>LO5<6YPG&naWujp2m_jG`;Q&~ThQ;Vt&PWPYrQ^LroVU3
zJ$t@s?f;EZmZj9advJCV4*!Q_!skN`xUqZT!mhypj^B0HarhKanj~B3d#34u?>mkJ
z+p!%)&Q>r@(;C;RmaYL~^~RC?FJEy8`TmD;=!#K)(Aq>8w+8*uL#w^F<JJh3{O|q8
z(|X*Rc>EHjjanC=o7Ne#5pBO(L3>V9>!^5*RQwRGeuTV?yoZpG)JET_;-Z?QP)^Hu
zK9A?JoTl+SUR3$%#c4U`B@p)EeMJ??tn3L03+RtHGZpXCnSOIo%&Bu)S~SgLUgr?3
zif1X4xj<Mr4cS~YfM?z9?EmW@y!+kvoL+v;b{xlXq`?FIX!^>N%Pa8uQ>RWjj$=Eg
zn$0F837k50>eMOM<-_67F$L$YBGJ3EIBz$UG`Af4?hCd5`scp=>tFx+*L!!~d1vS7
z8<kSpU!bWB17=%exVm$A_nPO9`<+gI{AsfWP_MgB?{qp=#}^IZpK4l`nU-mpX}D=h
zF^X!WjwJc-;Wm7bIOI%w6>=%K<brxhYdp;d(&$|?FCVv1H}j$}brQv)?yOqOl_YK9
zKRuq7>88?tiwkj7q^HlEUzwM`X1y&^4+@@r_e=^2uf%SrN6~LL8u%q(GNi2O878z^
zYlhaEZsi54=lzs`qEePk5zvn!_;nDX*-25#KJ;i5(zJuVk03;|OB?lLEynDdKpW7)
zMx)g>+UWbZKVZx{e3%0?o6BbrXc(boM!hVng<4W(`av*X8uY_Z9$u1`X0%qDOv*B{
zk~pStBl;fw6O2L%Qg;6Y1aG15VN$Qt5R?L-)D9W*ybJ|z5ki2(2bbaR;7epc2;@cS
z;fM~=Tq1ihxBCPB4!gaL?c>L)qm4R9QLoRJlf<>Xs8R1#y*@xb84cZ#(t2-a=VWo@
z$fRV@X^(no!!o6_RBRKP&tLWz;WFuviq#3qU;^#b!Sx1;aqB7>1|SgGEH;Fq!D9QT
z7HN{g^ABIR_0SHwZWP+>ENjNiMjUfH3|$vs=WVt1wS0s)pP#>aK0_Rh)>h-^K(Dtk
z?Q}V7fn|rGZ2{JImX;7M9tGSze0wp<yInu;_Yl*xUYCBio9Fq>NB7U;#`@mgp$l7E
zh+A714(;u&Zy<EndtJnKXL%(LEgQhL!nS1%2HS)YLI`{xE|Q1I%gM*d=Ls3j<N2am
zR7II*gKS(D<!qi8MJXDBV3x+mQpK3WHrR0oqDm?^lIgfiE5C!7djz7dDR3tw^4OV7
zlk_DPibx_Cbu*vf3ssSx3XLY%Q2d_Qxhq$*FD%Vx1uxMGuD&%w!t${RC&194sb!-E
zh)^qVKx>QuG}<;Q&>AOuG%+d-HvSwNuwHJh=Ys~VY)5POoqO-Scjv0Bb~p=^G_8*B
z`$iy@($UUUSM9_~OJSL@@9+P@+S=ys?&kXX=I-w1XKdSEzt_0JC!;l@)}YX~P(w(S
zhKUl#p^A*E6TmP*q2s)Uj0Wv!UH&bwVR^mn(Y^QHTWxV3NM)K`KTQMA^p(^qhH5KT
zS_#vPgEaNuwYj^y`J5tH>+_k09Q-9-geS=T<ay*n<hRK`K^+dmt?)uZRLt6wDwmoW
zZ359rc&BkWC`D+S;L*D*szEs@^DK6ee0ou8ny}>NQ1Xz9DoW@8^x<3qv_9x?XMR6C
zpoS@azUH{B`c_8@N!Pomqs8%{B&rJ9dF^qQtIrKrNQgjb%5b^gz&#>Ze23kqD+p;H
z&UU+`$ctR%Ure;)YLT*0&>7Fi&x+RD@M9BZz&NOvCYxYkP|jZZ#j4449Lppt%jng<
z!=N0Ov!%@<{W{Z{0W-qTRX}~;jYBsG(srj40yOiy*RpJ&re)ZcP_>q6na;Jx2vIa!
zX`|cS%!Z?O_s}$2hps0m2!Cn0^F|0lv`q6W{eHj3Y`$toJ0E4fWueVDYR?N@&vOL>
z)U7D?;!l7T_+hF!upnqHO(rY>KF_e(UmHF>bhPia9oy!BDl`oMCk#7H0ce^wVB~rZ
zrKW9%k>|P)Se7k#ZM4$b{}%w1j_rqz(tbo~6`BoLPBCaF@F=w`&(vJF7C=(BP0Abz
z@J9%((6ucI5JsNX9Du8bUgfNew&jM5DYp^%9;3FCfTlDElEiiyrOcGv6UaFN$XeRz
z*BhK_9qH#!=le^s7q_<S4TKbcfrZrA{}2RzlPRm4)Ur)SR3;|}R2qU&%h5|ci~&RV
zZ$_y!>W%}xX=>LnRskWB%6TD%mSro+)dt|027qT<NEHCBF@-3(pcGL81x1-JMS97d
z%Z#-pgh>Ge(7>(ze@ZE12Im^kQQV@;1b71g#J&&!j2l`ZN=E3xe;xcep2WPraG2a4
zTR~+vDFW<pIQX0<vY5y2uj5?$ohL4NtSh7_yskn?8L5~eNCz7!uMKf23fwUq2Y&x2
z8j<OF&i#Xx?jY;*l+;f#W(IIqZ#b$Ss@DM8owfZZ2U2>T@8fIlZN#SQIU849bCoFt
zoLekR#Q=D=WpPTGX*#xf$D7{tCfm@AHupc$fe#4NP*JzLwG~D2s|?Sx4JpTqY~Iu_
z-8$idVBG$}ci~szL!?bcWSt!L_=W^fyNe>1g(}og*VO1X`C^ecbyD$s8XWdWC<<EY
zaKz&h(&GT1fnQjD;=cRtbA(|!`~PT3K_3>LXJmhG3E@9WQU8~IbN~NR{U7)K)bj+r
z@S{vADFOW7{_Wqs?Gee~I<0j-G)%gNhT-mi#6z^N|7*u_@ersbue;FcU;WGeM}&m%
zD>?{058%+3zVxN9YbppKgb{M^f8eX|5wbxl@^tcCqDk$t(<`%#2;(CqWOj6`2GhKP
z{}wF*xde`4ptvJxo?|g`QngbrQ45Qz@b@wjJ|!$WU4k%dJ_F?RcrO|6M4kuFxRhF^
z^1SzEWSBz?Uige>JR_NG3OCuF^i30<@ndNi2&K5=q$3E!mSocMnLrHwz2>?wZiNwe
zZu(&C*uwLKZ70i)>(n=o{2pw)r~iJ>O&?5LmnkJ2J6*ONr`fL6x;57wx55zK3N&wB
zT_-Jb3EdcmAO;f(1rZr$bGq0FT)b%&rZAX*3=*Ki_gCmiu$>r__c~6zWk~tSt*tFX
z$_E}LT!Lm}dG1&C=eM@DV#j%Jg0W+Rl*U;pjem=>`>#!lysj?-!2<Cs&sP$$BhU^M
z4M9<(2c`T@j9;8gC%+Y@1};+uA1sc4kdx_zX_=`$mZVIZW?EC!Fa$?&@)W10AqAr*
z!AD~}7ex{qG>)`F#-ht-hB0!-RZ3IFw4z*4kU}Y>DhT6P1ISKWE6osvq{7fTs_kgj
zEmHFbrZXjLHPhnEw4`LbUgylv*+77?u@Qw5k)vspOe-CQ@qtS`&-aA{A#^fkR_jO@
zy`gwYZbogt^f)0rktlvbNZkcnxFX498iR6KjgB8%y)|%x4=-|jTLy$f-Kr=O8?4TB
zl1hRKgw<~ukL-iJ%a<=NG7V}h2B!N_Els63u+|WzFr;RWajmUDDY$^;{rdPY45JtL
zt~c-;M_UPpDM;y9;yPg%QV3}nLWuu1_f~?{*8hi8a5H?<ZHjX`c!)BpwJ)XQzF|s1
zX&8(KVZ@nC&nVYXTG6FW#D(C7A?_DKXhU!=41<4Gxqrt#IgVHwz3Eh=gw)nm%nQ4!
z8_+mWF~IxWNE%EpRW8q(X-|_B>H{UY$s9NTZ0omrw;ekkf4AWW!OSqNp<~<eo!>Ub
z`+u5lo!Nh}2YrJ1-+7+H=%d;&oWf(=THLn{4$fvb>?s5^Hb%>RTT0Pv4(!o&gPyom
zuYUaT$ItELu$6uekH8mmS?k{-&mwOmB>%nbaq{-sZ1h)+?mQ()VP<y;E_%fh!bWZ_
zq{%t72-=p+X9W)nyd;HO)J$vOqI5@SIfen+a9nK|+I{w2X0efc$2IhbpY&=l4Cg)R
z^3rrt4Tni<vb>y1MB)2Bn@*YU`vOs>%gd8iG8|UZ$<lJ#Xzc&T`daM8^(FvxoGH>{
z$Dj7J<HsmHcKm7IGYrSohG96aHjH~SbPQu_*lakCh}{~Zrb-IG)#}WSAD?wvEni4Q
zHDb+;h2u1u!{v>Q<$tsy&$OND^|scQw)VI+e;0dut1H;=Io#6y+dx0AhLlOtXNW8C
zL9wnINa*k}%S1LP(<9%HtWVr_+ilOawMwtN;l|tEWqO|ZQ?}MAZ8Ti=KnL!RW_LX9
z2O&P~X}@6GpALf%UPL@X4*mfy!;9i+O6$g^3HPG~(~twFc-Z8@yh+ZG#HB3bk^;y_
zSPc}$2EeGGBR5>J%^9XEzN7le2kx$_>WXXb{lLb?#-OSn>-Bp6$V#==ZD)|wYBk-E
zDPt@Z^==%;G{I&RMKK(1x7$$XdAHl;O<h}FURIgYj^lX4DQ-M{`gFZox?yEyr8Nzg
zcXxN^*A1p8PMkP?>lzF;Ha3d2<zcJU%GyJjBuOi6-;!lnkOemcK~M{7SG3!0x8-@Z
zZ4a#ex?vc(9_?tYH_dX-w(VtS<@rL0J>%q?x3{;Qh5y%#vAgv%p_U}cda{0Z7=|~S
z*WT1_w~sgCTE<Vplet{=hU9MY5X)#fM_}UPGMi`;dBt8UqYxf~jhMOB;?UNpVmula
z?kE*0-m298nQf9Ts_&dUd2+Ehc>@5r-7o-1($P*n0x0qw*a{7vIAJU$LzJ!?>wFEp
zVUji)YdIO@Tds5+k24d`<`EycYh(%ZJcc9fcKfl<*I)g$dl+MLtJ%z!T5Yv3)>e-k
zSzFcn$6SuS>A0q0G445zavM~ZMyr*tbUS<|*4B<s3nOw}iO8giLMP!y7)VHXh@VH1
zCoKjf)4?gQQn0WnDpE$=^m6#q>6X=F$F{eE{TEM1IpdqXF|L+nvE<y7o<4Qmb&J0d
zoz7%>jOU~2to+HXYy3erne095!VQy&()&FjA<KmDA}#3UkT42<3uQHu+Fc4R|JaZH
zSQl_}^E9m2oer<Rv$Ox#)o_S799E0_-`)b={qA?KZk^p6BaS!Ee(S;LUELR^7$(_(
zMWbO}^Lc*oUHB|~gtUCc74vl|9LPn$(5kA^$OSeZ)T*~djxOMe#*i;%b}Kgr+jR>~
z_jVXRaq6a<j_xpiLTGJULdcc9qkG53UoVP624Qcx*+k$hj^fa^!LbeIy3SIZB%w&t
zBu3yojN`zv(RCy^wz+ls#+!~FX8ic6n{GNPl~R%m8c>X)$sa8*FH1uM#BrfL*ENNZ
z(scy~V9Z~Mqy(VM50)&~vn|f0bbQ7EDD^>pHF)sDgWrVTfG?1BayKEPv`9<f1cd~X
zwUpG~P9D>74am|(Q5EThD<7c(19^$JMZ%f|xQuxG^XH;~A4SBE3VBHIlTs>igAjK~
z`IADeOa8w&m%k;r@wKmg4H=VvAo=RYq?$-6wxv|!ZZ7Un;)6o1OYuHL{8`S{ggC}I
z=jVmEg7bHAJ`nQDi2Nf$yy6f3;13Xhe@KXxpXR&}LR6e{euofex%dF*Lm>zugb_jr
zT!xDzAaim9c{(8=q{P@{z8Iy`(-zqQ<I%&-sb(_!E_Gzl_}`Y6`%{z`(s5T!672o&
z6^jbq41Q2=1il|M>VXe4==I8Hv9#9jrl~k|Z`wYCz23O&_26H@4;qXz+6a}BItU^m
zwGQ$|;KP%U#Qk0p1H?(MA144#J@5hipuYTx#=Rb1+fQPEAZ(}4C>;9WSWoB&sQh%(
z55QDG5QR!fN-66h1b*|={WwXMjGBR_2M3pNuhR2JTtY?zrfrRgkbN!zLze^ug-?4p
zXgkplW3X{tK3>PdM}iU3$eS|97A;Q!ND)rrb#A;p=-_Yg`idWBP9IGZkV9;@c^O4_
z2TcLR-d@Cflnc1QD>doSE`BXMpL9Ad7x8#;r;*|WK+t_|nInl<N{%eD#|0o#2ETjC
zIYX*^zvV=c?=fjQX}ul<xM;UM&viK7W)GqkFYq^$0svPAccok!ngXB_f-|)p1F~X&
zOvb1-lm@DlLM9xi-mKRH_Rs06SeeOXo}oaX%(lTkM<~lu(WVUT`Pvkfx0f0bodCeG
zEgO)ToSUYN%ybw9;6R!6gKI2A#%#k-Sj`%yVS?hmGpaKtZ3NS#fQY{IT-yc;R68sa
zkFyb^bZpb4kkkz2pLgssfPo;jbUocsie}fP%m)?x$z%~Il_1za9|KHEaD%`zqBsaF
z5;MTKQoN=UPYINcZAuZ8P*%*fZy`zwXi?7jVowBBYS~qcI95U!M%#8QU_wb&`+SV`
zJw!twGVO3gO`YN7xT1(ccTb<lFa3o73zx|wgaFXxfm`l7#1CM`7a$xFArhVBZ<T&E
zV}~m|B+RK{tSnP_igVs5wyyJ)0qma45Ve96cWxyzmeH+S?kn71IR7(`KKR4#%ik81
z8(&&|rWLciZAoPt2C!DE)mppe<fv{ja0A%ZR5XVzyn@>s(^IuzVZ1DO=8xuAJS`7y
zc<I)?+8>#9CM*^4x1CyqkW;k;(}>Z@-|;iakCHc%C&<r{uaX~thM|PBH7F=U?<Q%I
zUP>OWCTWg)x0x3nBaPlBId2dxq%iB-Lk!w@J}z<bo*=S8F)y<ckg=%{o=(l-sW9Z@
z3VS6W3T<$T{y36ZDyB9Nn=;!9L)OOE7su0R8l${l@Z4_Z3Ig6!zkeYvrxEa(I|t_w
z{w;1SUZf{aJv|>-*3(a&LYQnc8t}tU;;B<lx6Hxl>8DOoXf)R+a9ByF1)!vM5G(^6
ztyFW=?$Fd+=$C-+gNMByxC77g+y-K!P@dJ8ZvXeQ0qSwAZP2-LF+}ZhdQ=t4p&@Ox
zLh7iQ+Ch|5d@OEJ6cImN_It=-I9#mvy;{AUErvq`4*mY5U8{Nh^(Ei}FtvoBTq%%D
zX~r-yg<wn@z!;cUR8{5sRn<Z4R5x{?*+A#)YNyiy$AQ%jG#lugUF~!_;5abQ+7^gM
zl!Lok+X6ARQ$TJjd+#bCggBIDB<Gw<!_WacvNJx8OU=@R@D`_KtO9OS#2M0y2lEj^
z!cXl#4lmz-{Q4XDU5zme=aYw4R^(kEP71j5D&8gX(PA;m1se>u;gVrEbFJQg4;u`&
zqZzs&;sspAcUWw%tjPFo`q{iW!oIBp(<u?u+{8l@7XEav0zUg^-@N5DuYcg?2VQsM
zjW^zU(=A{kG5+GkZ>+!I$}4~3(EQM~Wm#66`Q~f7ZK_;^OGJ=}bjT9fBnz@jPI4hv
zek*wzd4xQVyqNqbc|+|WFaKfk3Gy4{i-e?$Dl3z;h+mjS@wA);fFmC-n~R~hj?3x%
z_dCbyJkU+2*)+v5TWK%aSc%ci@AAW<Fy&vjUnXg|Ec^XFU_h<@mgPN86i>@>Ih+6b
zBkSeEo8Q<xeCyMn9zFm0&!^&zq2oXN@WY7D7L-2y<|Y2jv!DHJ#D}S{e!ArTAN<8%
z{6*mGf6EPicLBTkCn)#-5K2Ss{{kYHzxk#&y@^qPN1bp!Z>9&`E_|Vmwl922*|u$I
zFS888I^}<kFX0Jwk9&hrx87w~mhow0(>kI~AtLTSVEeuekMj4pzhgU;n#Rb`lzRF+
zql{XCPbg6bpTif!7s+Yz9P(~LKu9fkS2jrtU?yx8+g+M1nb~|%ji-iP4BCn2d&xMC
z<UkgQEQ?zePFhP6Q+j1k42rR9`d&00Poa%pxHT;~1C|Y~z7*ck^#k1cV>vb7Wyjrm
z-O<*&O|MPNXKABe0tHTG;6s&7QV#Gjh6ty`G64M=qU98_1Sk4wLLUUcYv29ucPqp2
zgpiC&$`QJnd%UaFe76;M8Vy@p@YW-}zT<Eq`+@tUk1PyYq5q_-UCNnhwVEaDoD5?O
zEk}bAF8{1^cSJ2h3~CfDXUdO75P+5=`Tkp;eDX<4NJS}QOfinQ<7&p3=d20%V$(0d
zpTSSp0riQlBY!}?Oa7kx0Qfmk6eDShE`8Is57a=Ug@9$k>w>mOd3~!oi(-PsjXmti
zq!d{(vQXg=&14n8N@!r8sh}p#PI+>LrxunoEz0MCRlrsVRK(+GI*nMO)j8@k#Smqf
zumv<vBoQ(z2pj?daw}yynHIcnX{6J!DI0O8bQJs<bbH&|NB6e2DR-J^d;93#_ExV8
z`xUfW)7hazvnjFVh2c@J45bFoh$MGhsnr1(d<$6!N(ps6rzj%nNkx^f6*qt~M8w@Q
zm6XU80{Dt@*R}-;&vz{3Zfbbl%h$tE0iqVV{uhiR3dyy@C3iRzEvC4`xxD%QPQQA*
zs`|a&psJ1?tE$0oH*t6}?Q}Yw>E!UY(xXE7_Zr>iyF4wO7f!>NNVI7wN`(}rYWR_F
z8cM{=VDJ5TN*OgAj~h93RuzKVv7kUXi+Y}*(Ry33XvfRt$HQGlxA|d{ajIb5;3z0E
z?!;#mql~#ia8B*v(M`!HXO7^UGwXUye^7r_-ObL5xyiA#SAJ0$@-X`%L@5Fz!<30{
zE5%gcmbaV4rQfjS$Gw>sB$uq0-Nf_%PeJS6)4G6Bh;Z`$L<pd9M1_bSI0+a<0Ntnc
zv<?(FZWuZ)D5Zl~pE#l8Kr4lz>j$<CDuAgQ1ejP6Vl+AmS{Vp2VodPsOplHt#K=l8
z2;BXzfQyZKzh7^l>o^d-@P!dLj!OrPPB%@_1tN%b@HzMoxJ+i`QOrM!KS%yTgUt<G
zUGudJl{T1oy+etYFueAn+KMfEYDMuk5=B6va9nun04?9}1Y46WN?(af<)vworcB_G
zPj)Aj;bp2M%jdIWhLag*7PRa&r=DGWJ>i1X%QAfn9AUyq(p>*vi!_n&S-rA)BWA;m
z&1?$DS-r6Xh=!q*V$AW5tgHxRZoOVhmFc)WM+RXWSq#N+I0xJDJY~Q*ju-`4YcCa@
zE}YFrvSph3O~3^}Pk>mV5rk1LK&&Mx02i<cTmWQ)sSw-w2u<ZUp5N)z>w)XFkeQb2
z7$5{!QkoV6DZ`)-j59+8_1Ki~Rx=1Fm}Y0|fx*Vc(o!%3!(dbhgK^JiQc$3#)$VNF
zvwNiZe>ZA1tr1WN3Z54RT01<B!U)`6f0VZ=y<xMiH2}DdP8aj*ua}|^AOu~t`bo?{
zaA1wbNJvMB-F^>*7!8l#=b9D+?s;L9hRtp899v5hq!P9nXw5hW-;W{#xMeCkG#t`4
z{WHBcB+m?*)8Q-oSm9L6+~WFwZ2!IR;Qo8FosQtatf&7GG;r_uMl0gnHVvD_o@pde
z=!~@Eu=e?GPfRFWvh8s8UB7$t{7thsn73MPq;>4N$hPi{qK%$cqf}mlrPeQ}2M6RH
zd?gSvB~NGQ)8(*4J9&|Jw6@c7tn1k-$MydJvvSe6WLPfB`t#Y6&F53hyD9tC%I<!q
zCJ0{C8z~sQI4=tK_w<fQ(VJ#4TkIGp*R<92;ks$U76LeZlLHPpH!aSG2+Zm`)LVS3
zO|zcCn<mY`2!Wfw)LofoJ^N7IG`0Z2>02CN#JFWKJ_6uY&qaVO!>kYQLEFaXD#&(i
z#4G*nCRZZb3DJHNpNqDQQ{BT$ztGN^`&8N>&mym5k4B@)Y_c-EWI@10G1!(0IUGFE
znTCfS&%?E8iEt;7rXiSWK~5%;EuGY2gVO{I3Y@0zYnKW+l=m4O+ZqPN9TBxBygS^=
zbAX~)Y>h_vKeS<l&A?|s1q?&a1Wrym3a$u<oC#`Kwl63$hKw^qw*voe12U$x_yzAZ
z)BVv6H{6gWxW0bk*yaY}`o@Xn><f91k8K(NkyX>@1&s2oD-luIVJHF6vRqp!3LF#w
z=Xlmjs6iPI4MWlTRVjD({{aR-=1Rfg@0{JdqTfi8<osy$#QGY-#>Vjz>uVVL^vT^r
z9wq-x2;f~c2=xp1$-WH(B@U-Qm^lRz><?EzC`TZ)$Km3~d?lT3qhWbeNRg%TsIAhq
z5Z3lO&qgb{y5V+XaOs2@;-)ewCzshuOGy{?EO@e6UR2ZR97e)*V~VnX%PkriZ7IXp
zl!>&$NU4??(N>E_Mn|eBv_vY+I8w50#@K#QjTEV{cK@d=XE!%DD^^4c$1jT5Ua^zm
z$aShH=Zl3X;<@XNhN-n;rGlB1`!xZ>GzI8%*7H_N+Y&kYTtosTbmKVIW)vyaGGo?g
zu-I&B6~?BDl@$e2HLaMo+cdUXN=Kn3*FCRZ^E}V1)jiL59B0W+v)prM>5^D1M3GE<
zH_H=e*^P>!<ISQ%%r*pyP_O_53R|~Y#d^B~X*!fOpHqv8m=HoZQ8?!RUKo)vd5pY@
z<Q*N$S(O)Xw@3%Y=e{knki-R(znmLGdf~fHC!ToRX1!^Pc{#{90X=JJEXyLKWVtyj
zy4p6DVZ!zMUiggd!2rJ3@0>UZkQLn&HE0Z@FbE^i0Gh^w@%fZKQiD*Pa=dfv5diQF
z1JFnCD7cVbcm2teCj)>+b$vIeGgR>h<;woITl4ve@98i827(P|5Ksm2K)%UG0PlY1
zKaSp`jb;;Y8^E-@S`9H?ZY(3c;`{sG_{vwlve~!H@y?Oyq=WIG?6=_u@L4h-XUHQM
z!D-><MDs1ojX~IDfWzcK--emk1gj1D2!!4(qn0c(B_32o4MfVJ6P9Z6Q93E{QB4&*
zzwbIqwp*>V*{JVKBPr9`N~2yEJc(;{_tz+mqdNH3uTsww41gIruo*=v%}!?mmD^}Y
zVH&#ArgShI`X;4}E5$^JiPxz8_jEfQS0WA8YIPH_JhFSFpT26j+jA}7HzY+AeOoIt
z8I3lYNs3`quLr+KY3O;9AsA`D->+MinYY@F#L~VWcc#-LM-OKMU`#|Uq{~}dWk0l}
zv>n6pJSS|m!jG7iraGn+pv!p6p#%vYTqgT)nbb&8cUYT276e(PEdTPtyv&F+BTu`P
z=kK}s=9~BK8N?B+t{gpIZEUpL?X1yAJNN0dAOG@~zx*|CFxb9sd1;?~8Ej}Yj_%g#
z@I4uQZv!rqpq?#Ci51#iSWZh?#%1=Z=TztKn!RMY_cK59GqCbkknDftBeljW`)_g>
zFTx3;$Sw=6B|k#mL_SJLC^F_@y-XE|oOA-h{lQIpt=;sEClxk4`|6a+Dv<~`raA_Y
zQA1em%X!7V3KDRG_fRJXbI8MW8dA_#Wei@Qr5&<cRLR;AdR#s(C({>Lk!6RZn%+{J
zE5mHNwh)d~%G5$ymK265m6Vh*YwZ`0VqjRNxWBW~@dH0-pX+2<whXYGeYoR5%X1rb
z*MlanwIH8^Z6~UZ&>9$q_h4dMdfv@o*qu@la>fPpozQk6bOJ#EC<m<9JDu263TWNT
z5|GLuE^HQtk%ZbnhtbiOzVxNz;0Nti;6nrQ5nTN-sHOMDmTg&42sIN~9-3CF)Or=B
zt@RaP8ufQS`siDA3{c~LFaM{w2OlH`i8*`BK7(8&KSe&BiJuYN`!wo3cw1#D;*{|1
z#2`#llBjqTfZ0n}N@)h(e>wl!={qmoyDoc2sq<^o0|egIgRv$_WMV>Ma~pgok-Pp!
zHmNGuXxD1(aZrOa`ZLHKztWzqm8~@#rc!|v2vTXf@L%wZOBI=#F;hE&bEO%Vjxt^k
zmK9Coc5GR28|J=gLP-6nF#_zxM)Y7D$91SDLu3x6jJ1u?Up1kE#!(wEQV90{+`k0(
z>|ctWN`N-B(hCj40mHDb1Swrd3NR>AEL1`Pf*_6^AGt9YM_r#SBFlRHre(#ouC>gh
z;48#t3&8VEcCWwwX~G2`2M_uA8Pz<v3=hH=NQ0cnXk9=fw8!z7y&D{Db@rKQgvaWl
zU?PsDiNC9X%Xl#nqcX1Z!N{8lYEIHb{+3~yAo`(U0ve`)l+oAL^W|YDiWI_foB*#k
zvOyPUy$5bkv)^wvHD@e%M;gZrxp2MqTitBfPvV%OX||51sTFC*aR531hT&R{T_B`N
z4-Y*LfC~okU@%`^UCoPjG88p7Mx&+WtleyO_WyioHe0WEx@nE=f9B(Bn;ZS5`FvwI
z8pYn`#v0)ev7Uqf#*6S8aus<Ac@=pR`FZlI<gY+M53YlU;Bj~pyps?PHgag{W8%+n
zQpjI2E~8B01DLQxfdlCeh<^zI0K{TvB8SShh$BFC?QAfw9UpC}!-@x>FjJvYNhu9`
z(HL{QyGTVww$@S<7c&qF?MX2z{2f_Ul3WQ4hl4{^vEpo8{;Q-~%;)~JoK#jB@Qo2x
zrQ%&9^Rg%hne>rnryHa_p3g<)>uwZR>9Pzg?$|gkld6)-aFdwmxmi|}<4j6R4p~$v
zpB{*{@$&-Jlh0CI8JQ^@rZl7zEd>YQ&==iCb})@ZpJ|T+@Iyiz65<fD&4$e=4ZOl4
zlwh+_cPI-YU|u2#TTt<Jr3UZ=Rn=R9i{1o~a<9<nl22j(AL)@0#$~$<5r>R=hz&KR
zRHye{2XvRQFKb;2z(#Z(0yySlfcR^S?b1mqQ8uFV2x5pBBJknW02ccNZLs$LDF>iO
zDBGoQCBW4IF%(m#%7Oyu<8jnNyXXTP2hhwo4sbjGrkOl0S)bDWKpT&H8G}_tumB)}
zz(+)?EkG${G<nf&0ORL8Xn_da0?)uTC<+wb#&8HQM0Pg<K#XjSqXy8(*BBN&?#SH_
z{n9V}(v*{(VK_w8K=0xhA$uj*Hq>8QQm4$Hf|HhKh%u!ZVuE1S`?w@v5L0kK$~*0~
zGiQE;**v)nDZtBL1=aa{VkGL|_YvXO$&5VRf0<D)5%m~3E)C@h$ANqMG%sfJ3gu!W
zzK%C7Q?r=DV?n>otw*uLhi^N*b7-lsoRyXHHg7h#H4-d{I-Mlu%ra9)G@Hy^$_9zg
zIK;sFy<HATihL6>NxR)n&9W$8Vvwf57K6E=vO!gyI={6=fBoslkK3wNAD2gOzv9Y)
zuuq?|xVO})*C_gas~$%tcTS%+nA_>B^_vZkCGa;xRgKFo@<Tg}wzqq=I^(`mtF<<2
z$aW5KtZ$yX<?wj3QCH^S-Ma{J2|4%{68LFymF8#F=aDy(A0zK4A10q7zeB!B{+#>^
zaH!b|qX#VLwhA_eC>ctteK_4<qJn5P9>+FvmWgEhvXvLgIz0fuR7c1>B$)K{@PtHJ
z9}-6LLIS|h{Mu`Ek)*j?R8U6DJXAlj^ldgz33PCny@E@(Rk<JvkGkU7H{=1=JPo07
zU_j~8!-opTRb2}aN9|5u3dXW-H;#nRfLbq8ndzG=zPtZRSs|jh+wD_|RLWkv6GcL5
zK+e~iO@Ml1<#nxEYFmscfc_v#g?LP<G!=%FousCel!j;;mT9MHn%bsi7?x=#wOV4E
zmho#+X_?lPa->wU7EAqf&*^kr_c6QP=pY41qd^}*QfAep7_?mmahxe_xUTV7y&L#8
zWdgCEMJeZX&vjkbwY^uPYo%#RiZsmn2$BVi8?Iw0IgAs4G|hiJv~2*}4nx}pu<a0B
zq+G1``<7`KZ6O%znx<)WyQ@l5>O^sxCb5gC^=hYMnTFX$id2fGVVG90w<aZ^lf<<&
zj@%n-S_{=|7Mv^Es&nywYzujkI`&IAwQbu@<5O`Mg@|Do#Zeqa0AUoxQ`?SHPHp=t
z+qP}W)A)IQ2oU<a(l8oG0qQNPlp59Ra{mX2W@H7f>sb_g-CqB`#8y()>O-X!wi=Ar
zy$B%mJE1L<u5nRsQz=Bh{|ta$x7$OsJ=gI(-*&;a!_c+?Y&#5X8+<J0pTvvsN#c;S
zK1is#O99e@r!;GlBrP+plb)nyHPX4>tF)}($Ok|8!E5R-u3z)hOUuj4xO@XVVi?BH
zML)Ga`MIC_IpYH%e3HtC%GX~HPwjv7Rj+#0t2VpcuD5M$dY;$STRfajG{cK<71<(B
zCm$!@B0nJegp8^n7%o7n2!zUJTmq>K3uG0Yq?q9ogRdl4fi%-usk<OSG15>3eYFFH
z-f)U}CE_BR*=8E?QUKS!P#=t<%8#cP6|MwZxI9~$&GYpCDgc{$nl={&1MB^1^3Ncq
zXG5`0(Sp8VouXu5nMkLYCNjeZ0AVE(5dzd|01%331^n1oaa|$d5z1l=JU<SHP0t5B
z{pz8q6~tj2Tb}3kd=G5f+kcOQE9adOa5)koLYe}Gj-hL=77YJ}(F6nETcUbID~$oP
zl#atVwmr}7t>%A#_ukh@5=fFkr{M;EU)>Gh`~8kgTd@6?u4(yWTEn_&+3tWM9k`Zd
z*AVJ->|3U*q=Xg-se6FJwCz@aj-7<2$pIMS(gwe42my|hB!<a%!1-l5_HENqQeqQ?
z)V5{$V_JinZQAaD0vkBCW!3P<>y86J2LL;O+_LRffR3GnrpXsy`4l8^QN&5m%!fKS
z1P%vP1rO}usP4km|L=vL!H<CN?I7F>kHY)H$HHfHI!C<QSp>OxK83JkH5Tg9LYAw3
zjt_2qTk;6EjoSz*%*!V~lS#$0-K_QF1*cprvlhTpK}=;D1pE!;PGZk0geo^{lfB&W
z(hA+1OtYV7nZZ+F;0f-;?XGGUyKUP}DN74mqG>~h@`le%wB8I2Ely|>i*oyV{+!KR
zLvv$%j?~>oeCTSis|l4HNGE{Z9NqP9(Jne#R8NJ38bYghHCzGsue|{>8O4%jtaX)A
zKvCp_#$*&A**Z>P4vn=|vntyjIv_j|F^T&Xu!_^vLp&rU43O2`JQo=ZZ@g^1_AX5h
zr>jw#Dn=r7Hcj^SP8Y+GjjvsQ@yYyn@6dT=2E$Pq0U#v%`*AQP>`OTi{IERoOu*V1
zDGUcZ+y{z+2onGok+pz|7?Bkd6<P_qjJ+#>5fbN6DFzs_&5aGNLfy<yqKG4VlxO3r
zLO^Jm^UK#HX<{v@tSN~Lgc|+$V-dB8W&lwvI%5=3@2wDkMQa9Q<!zOv6!grlUJUU1
zh*abdZHSx#LSXA$x0{9WYPT~&qtwW(i?XU*EC3)Hpa@Cq9cV@c0951z!6zoc5PAD7
zOY^emyCU!R##5qhJsdS%K@n|8u^7V|(&F^w>@0*PuItfcFpTzly(01ez*W^B7X_Ip
z3<kqtT5iQE!pShpGb@Nq<b4!{s2W9+0-`(Y*EL3w^I91er7B~YK<{$gPpIgAN_hcM
z<N{*19FtNK^+WKAi#+n2u3!ag%RyDcXC#_6=UG;UjMS?GTse4BaS|t3<V^^kGm)yc
z8;HkOCCmn5YpH<J#ad@Z$POpI3(s@IB)r?`yj~HusD?8+V5^r}f@fawidS4)yyXx7
z@QKS`*=#m%d~W;oUk}ee_Sog~@cjNy{KVxSz(I#P=?j0D?|P3rS!axbz|UKHqqOMO
z@dcUI5#L#z#rMDb<u9MjX21B0zqtRFx9q>NZQK5P4Hh5yoliXRw+}t^(B<F0!#hBE
z`L7Q=0EhlaN;;3?0Ion$zw>b1i8~d;qHAw;-oEpZ@Dg|(Ct_S~->z-i>Ad>i%gwx*
z)7WN3{q_N>cADYZLe);4B%2O#ImWrAYz;?~W;IpxH|EsF*G#w)?9fHLid8SM@k0y?
zw>WIDH0Ff2p&-hMOsa*dby_xMGw1BaU{11|F|hS@>+ft;6r4Tk)KTk$shj$7VyWD~
zceh*Lf#ty+kWJBgPzpdc%Xcq|VqIjnZ)X=pQOwKty?y(3c5#v2ek4hLDL;9qlW)`7
z+7RM61Z%a{+8PHK;y9?ujMq9*O2j(ntrbyf%&)ZGx&1}(hR6N|;QsrS^8no<yM1q7
zGyvWa+;<<qzdU9Ed;o9{Tksy-h*>kwgjp-CG%)jwSu>9;O#BP&!|r1V>`#?tS>%BE
zSD&7qo__R4zvJ22+1WdP^hba6(@%f2dj9$6pMTSv-t?w7g|K?^=B0R1fn!e+8$#@@
zd*=PldT$+%nS!15Z!aP)R*J|VM6CNKAN5fu{VG+hHKa(zPLTG!7}-)6Gr2=e-B=`U
zltq%Dbr3?Un(>#?WF2@*+VafVN6*gA&W?|LS&Zv^7(Negf@|<ncpBahe2(G&`v9ro
z3Szg$i*_B7E#fJJDpUeMPT>M8f`Ul8!ii+@sI5+?5_?aEI913eT3e;rsWS{Ts7e(4
z<B}}`lS^)6GM*iQOS9>$)owzwgY*yM>`4f#S{0dtYeHJ`Y#i8~-XT9-z{6&yvE}&N
zc0=Xq-|aXokkrFg>fLrTS&G|bP({x~D~sKBqZOC<v<|n6_6W;^JixU;5vo0#FLD9#
z%>qt#v|G550w)F%+`7ti(c<%<6<F&XFtdh|YEO?SLdY89C{476D2gI<K6nBSfx*vF
z2}&cPLy~A^jEyvd^VOy#dpDzqz>8+|ftn9(6NzIg)b%8*G-yU*#dB<uM2MKAAxJ2j
z+?-LfNI)d^|DlNTtSli#G)7Ep>V}dkR3fE4U_86p*?12iLQyEs@*JFP25}6C*0P2Y
zNhN8L0O9}rFghn1kyUjKnbw*SopT;g1R9?vC<eV$ADh=2t3$PQt$`oaeMsXNFsdF}
z1u!<oIAH?KO`U?ZAH$61D5@)eI2oWg8YvhPQ$d{MMXZo?JRa|7S`ll8syaG4*vHU_
zLP?D+jaQ%*V4e#i1m9Zq7{oZyKzVKjK$~7S25UjHF-h8_fYTxZSduC=4IdX}PHV6r
zOk~UyJZ0R}k!jFK%!(ud>^6zRO_>E2aK>02jk6X+6nc+0D9gm$XKFIK>zrn5Mc_lh
zdto2dO<q9KAJ8Y&G32T=fX??4W0DvVM1seBI<l&hzHk@brw*5r-@Mylvub1mcUs}N
zU-H0mxm-T8TCG-huE#qqm&@gw;YF7}^dFZmhwr?6Ieh1T{71*VaM^hUJ`X1y-AO!1
zrjK>rt@BH~->&NHnt3yetkS7>3}#tUmG9Jz=*dHCUN8tdY5Oy_q?{^EqnUL&N8z^N
z)6Gq_jhoH3zgw2T$`CS?6cKmxR7Ur{Vg0|Qgx}TiZs-mXRd@nxumfvf_t)g8+EhZW
zb2iPh*qhIo&1PkOy|^~tyMQkU*5IH|U`!a@xN+mgLpN^RxUrX@=k-YhQQ-m@C0UB&
zaunuBELaJx@ZG21JEH3s^+~TMfEb0-8_w2QIZ=obU@&&B?8W=TK@@1A>#XFWm;XK?
z()H<E5oO{2X#n%E3#~Qe9+NkMF`?HB#=r-@*L(f=pp62^kBLGE0TD6sA&LfmQ51Y8
zgb-drg#$b0BmjiHBR<<Ss7{N@r2q*8-|NmF8nI-_UOl2wL-rb?HV<!?cn?r*(=@YX
z`IHY35x(qAhxXok?>#PZOtO!AIq1)1<%PTOfpB-cwAm^8b&$joL(utC-hJukectCi
zcKh~Y=jZ1a55Dhv!`;i@g~Q9=ebbvBee}^sfA@E7)M30lez4PJj%n26oe$_d-}$`G
z*LJ=|6W?@)Y~Qp*G&UOZ&V;G28-Haa)H>PebptGHXxc#&@zeuKfnGcE^Jge8wyW2e
z=Q*FsK=s{i<}Bu4bVt>v;JVSN)#`9-hS14YG_$WcJ3EU)Ra=fCoJ<}gqS?81PV3<9
zljrB>=a;{d)OB4C(hTA_sR+_ux7X!)s?H=%OT^ct>HdD6m*v5Jp7$f?;BDbi5U~Y)
zS>NLP98A+oY{s~}>k<Y@f*3L3!`VQ4eq6QIerg#!fSyos?<|U<=nLN20iGa0^b%|B
zk$cdAKQ&R@)Xr)hJjBNyAcSt$Ij!Af%0=0ApBxt@_yY^fiP$U=8Qo9b|D0V}DvHy2
zS$znY2TbsJ9y}a;L7bKG`@xUqRD&*O`OOqNtipzqJIQsg=<bc5&1ds3!m9_fqsg5+
zkFM8X*|fsJUR?nsN%!#d<mhFOJignO9ECqPh}8f39vz@{*~bn5PETKcbu>&<??NaG
zV<1H3N9@n$uUL=AY38i=**cs3L-=0!iq52xCy4!JGrvc`2{GJ*i!VhvQD|(NT{rqW
zNXd4&Yz#n_4Tt;tqv6T%%-PYwepADTG|r|;)AXVuk7GPO*}-$UF<P>`8~VrBuGtcp
zc0Rj@fL6=fw~wxXk$QUN%5;h_9G;$>jeV$>muhhb@O|OKJB!YHbUvo@o1K4yBY3ww
z(T4RUca72%0wf4Ootif^pQAUMujfUeQC(q~1Ol5*r|qnvbQQ2}5!GL--MH&Lut6F{
z;zcWQ5or?~H_xr)((fkOp+rZOaYatA*X)=iE6;o{X5<NBBuJdZ$!fV<JBb1oGQ8qh
zO*v$R59yXB<D;@-jb08+h;qEMTkc*DZQyyXB&NEX?PhbfM}WIvOD0%O_;fn^{9(-`
zf!i3x0-I4(%)e2KcD9L0l=l9f2ln@`F8ck~zDj8?N?+X^%?5ph3V?OLcQSwEen9wA
zrBsy}Je&oHx1gO@05F)0V}~y66Co;f6|m|xjsdjat1|dG0ElD5#;{VfE-!dVEl;zo
za6(k%MU$fd*rhcJ05cZ|iI}I5RmM0XQc(y-ZTcb##rd+Rb5z)#f|)tXL=1@q(1gqd
z1AKH*;=5X-f>(f}aqe1U?IkfrCE6I<i?uqODoSUG7y%~PDawBcD<tM|+WS*Z?f^ks
zMSEFbdD}$d(GAQEz?#lT0<xe;!2Ys0F7&ddcqu?FJgmCC?pOd$5fbcWX`1Z8>@AOD
z5s^4P1%;U6NKY8rYeo@>vH7+iqX0A)Nm4tjC{dTWYOMj@O}O+L;dz8WK*~B75Q{0o
zZ%I+Bj}V|0UWLrMa#3XwO$c>0>9hdI(%U12kblT_!%V0((Mw*+tTi)G_8qj=1`ulN
z9($Z9JNUhKQe-4YN5C!#UJtdKzlHM8?feBC!ISVQoz89zt|`!@BR$aZe74!38_R7>
z*BPlZo6ngY$r>5dd$N_EdJX@TyX|hBnNDO;F?DzI>&;#QTS+LJC{X0Ti`8<q-25z}
zUEV@A`P-Ms=xFf1SI?_WRaWK3{y2+v(=MyUY9|v<=9A|&Czr{(U9<xr*g-kHwc{<|
zcP~|n@Aw~Lh|^UXv-m0D>{h_lVRkK(?h)=nX#o};UkW_cV75x@&6^>sN~4t>WHEe^
zNERzzeIVI8Pl#mTSZFHy3t(-U;)Pe%i8vfam(^ZVud6kHI;X3QFd}7*F-V}CBlvJi
zlB`I~lU5W_0I_w5gbpAInh_PBUvZKeq%^rWP7ULyDTcqQK|Shui`Mmt00jW@7`L&C
zYz@FCDfPgex;j1{^slX~RVs@bCEEx0r&XmDfN`M;`+MheO>CFQ#%5`r;Irh5BxYu0
zv(eR{RjL90Jp|T=7y(pq9ECoD1B?mKqe?iNCDX2d&k#9d41N41)#{~Pq`H~SwN{E_
z=8r+E1Y_n4G56k%0^t@^;kDKRG~KE)qSUjw6$0|2N~)#rF|9MJ0g@oIc#abHq!glu
z7ln9dt>Pq#-tpVM?c35aG42)=%uQa@Wsd={OCH&b%QBe11RR3k&nk}kTI2VXL0t4a
zXLBRx;y#g4!~FdxL=IjM9RgkJibA`zU+b!`wA66Jih1EKTsyx*WN-O-TeOADn|b(7
z_k(MnfARbE15Z8m)Kj1@|L23BeDcZP`@JW9`Tjlq!|<h@Ugx;8vvH}~0heKKv$-jt
zrIPQ<iKg&Tm){3>F2Apzr2uKxhdaGnPosDFt4}}u^zHY4<Ojd<fd?LV`tGCe4_^vt
z*1!C|iK@VzewM=P-}I)B{^-kpeD1mDp8Kee;t4+eQFsA<Sl;~nE1g{hHkcThO<Rte
zvk5=Av2q<2@Np2zy^qv^$>jQtJDl;~?429eClk1Pc6R#wg}wU_XKCHJe!KMgop<}S
zU;DLRJ3D*b+1Zt%FXSWK+3&m#aqAiGco-r^fkx$C?>w1IS@~(sHaKT+iC`SiDtWA+
z|Asfb;SJA{2d*(-yN?c*UV1PYju1z~$wAtaNB(Zz-EOz<vHe5vB0J*ayL)4zy>Yjf
z9=?|ZsHEE)?-7mn?jP>iF*x6JdR&jt@en;dWU}_4tvZ~MJ<hX_J@(k+k3n!ho%xQ(
zAA|51U**?k@T1@ZI&tTo&YL=))%ohq_jEeN+6CyU)v8@?=DyI7qI#D-u(W9>EtBg_
zS(r(nSYS)fJW|2yw5|<ry9IW5TH=%aJ5KcN;$(keel#1q8xlXGjQ0$Ql`{?*<QQs&
zM=S%46zZqt|4S!CCzKDryFP5Dcyr$aNHG{liQqM93%+aA4_St&rCkcxY!wkTP0D+7
zpRxWZX5gmv$OGp2imUw}`n6gZ0$s*>f3wpC=yo<6K>&WxX!=3mH{WF`J=`fQTPdY%
z7*P5<lzu)iucv_fe@p4>t))KCOq0IM1^W6ddUk5or4|4VvUN0EaGZMEO_lASk8I#^
z#SEhng%?5VBwhX#0=+eaZfA3&+kuY-VZ9mn%!=>m2k-->EK4h;Y)dJBY#CC+i||ma
zl}~sTxtBaneu8|Q5Ktqq>1xkoELQcmp3VnSi9~u{FqzAu#q5>kAgipr<IH!yD&wSp
z&Mz!9Agx2e=yHI0*UJGTYyv~{jeNQf0rqY_d2*=$D3(s1y!rM^y7lb2l~sV%m2+or
z{ZZ$G!t+wb9nR~v&0jJajkFZ?Aao<f9H)1t|C!Nfq=kq*FX8te8OBDO3P8r&LWZOM
zcc0@gEkt7k+#JV)Y-uSQkXvE?xEVi9@_chM&y)0TA5ZUV2o7g}^ZGN_D7s;z9(X_Z
z^)*1HJU92E#OoRc6?o6jJu>yrrVjkO-<);uW%yV46e-D+Tw$%&jB&y98npC>^PMpF
zR-|?TF}#Om|2xkbp`Tp~90GvV;V6#dcr;u+x4td_8H=FTud0)$s;b|E+aG_YvP>|9
zYr3_%WeT1o?N+<x`R(>{6XIx;FR$dIC<Z#6o<-zheKx<Ie6Fen;|M?R`K?yFoy0;|
zR;}JyZs!MTQc1|cKj0p`gKUzc<XUn!c@Fte@-_zHL{^?&<>(dLO(Up<h#bY^;l%7*
z8L%c|=Mf8R?7jwPS%e@e@5%f*u7KW~-t?wXtEqLfIeK$^TqAG0vfI;?>R$KC+wOmK
zcD(;*^n>)bM+QnSF?Q*nR}1e*lHo8(96SEk*mlAo=R6O>yK?*gPH9SMDg--tBY7#M
z=>uEsP6ZCTule9V4?YKv!(}og$H;@^CFCc`FOV;jKjz1>H7`Xk47z+bWr`Gi3I&ZN
z*&3VU@&qGJlQc5?!ma{=C=W>}X_Pj#bSO$<O2(d`k!ngAiPWL^RS|%wG${z=5pNB{
zH|GzE@mNx0;<goUo*fJSFVDPyL`el{+8GY|QOnR#NFhxJ>1cFhcY2dFBF5Gy+E5C0
z9CbRO7e=E;S*_k^DAo21O~G-rwh?W{f^a4PjyQH0^f(jTw{4|2HW4uH?(A$IYFdtC
z1a`A!nc36NUtbjUr$%8g9?xB;m!$2KQT)~2qkHSFyE0xZih`o;Ojp;s-EPV_E=OUt
zIBMzFaYg|sqZFBxLI6_NmJ-40xlXOs%$GY|&tGXY<G9i2wELgbD)1YPVd8pV7@i+d
z^j$ZMI{W{G5H^~Apphd`%TAK|5s*SD#*msqNTeX7vMlKd?z#4|<=B))LJ0sy1&4c0
z37k<zsb$96Fy5>HC{q$#%LeqBCWOGj!NH&4Mfh2A2O%J2SqfQ}LJBE<XAeztQ2>vC
z8ZeDL@=-W-Gb4bK`^4o%wU7<1eY2>>C0?9{Aw%Gd8_M$?$+T&jXXf*5%U**dEN8P@
z00f<Ly*pow=Vg^7fX!xEO{a4Vf;cw)MLJUz#r?d~PT*&=dfjGdB%Q=DP06SX!U&UO
zsFVjvC8lWt6Za@I7}ExV=hAe!zS0k^tpcUC-Dw*ppa@38mCZ0vzGVCBWI&FR^FG!K
zuC?eXmUNvkO={<7gSmOs^$)(MWHsQ1z)tJxHfQod!TE#PV31{JwzhgbfNZd}b^PSk
z)}Rm2?QLzHy=`j?aQo0r*B>UuE4TmPgAcy=#SaQ|f#Z&5*<kQ8==ZCw(^qU&ycIp2
z-qzOX6I)w@0j$%Vo%1&y+NMw~?!0+>8!<%~`}BH_)vWuNq;Fq3+09o-cB|-IOb4Kj
z7L#fg;*NmRKww^Wia`R*3^{pxLc5UvwLhT#>l7dPg>%`&L)$m)7DX7roid7&_VQrR
zY_+LGq>jD5cF*O`=6L;X6$W8z(CQ76L<t@l+BA|M6+1gO!l$6uueQ!Cl3(55+B#MJ
z*Sf<S4{amV>a{vwZrTpyd1sJ7J?S^kh=`)qH=ytDuBN6@6p-{|r96t7G0ydNqj}Sz
z!#>50eSdoa@5jb{4NK4~ty#sApGJYd5jAF%Oql_vY|F~DOv^}WZgr7OvuRlx#3B*U
z1RVjPzecegQZ)6L;s@ups@`YrJHEA*4PaSJf7A9MG>@?~Z5mm&wvrmsww)vl)$JE)
z6czc7$(y~nnlW~!MS0w91!h_t9x8EtZ<w`P8gAa+M)7+16lG(g_-niW{v})e-luie
zZFeZ3-I>%I=xi{`OryO8vR+G5{CwE4v`7K;rUkHUn4^s-VhnJ5`{q{tu>WSgGLq1{
zbGa4-aEvg`^1Mt-;RwT8i%%=u^{z6rNBUYQ8?oUf`Llale;}oQv~@`A?d=}B9x#gX
zeEXI|hYrz7<VTOMrc1|L=cCn?lQ(Z~qk3S8WqYly!U_s}CO=epElm=9V<2kNAP642
z$GB-a%(@-K?d_X;&Gq#z-XvSL(Ye*$Jv*IKP-gNb`N6-_yMFaofAv?jvG0aG&CB@q
z)z#J2>sD4)R`xl1B@TjnD7KhxS>_e@pyphKd!KGz<nvWRlkZn6E73@FB8)-~e4%gK
zpd6R>-#^nmGnqzFaxH@>YPD*997jW@wbQcgE7xXGOlfmt<IoqMdg>`RNxY>fj=eMO
zE1PR;^P9G}s05w_svN<h*xfyzxk1@%S*8^JS{{DtO452iO0`DVY`5UY@4x^4I)Jz0
z3r@@aO1%bjYx|~VJ*L*e_x;<$QF1N0mpp4fV^(H(v7>1{SAu8{Wk5v|k-;YE_YtG-
zVMgxnZx@M-<PXlWzxIrtn$PF+pkAZ2Ru3SGq7b-aBV^A%dGh3-Sr`URc)Q(hr4a8%
z{#vKEwtA<pXkYuG2mTw-vNkp>8=rL@3fa!_S#MawAW)kpmN5))S*16Tr<3P$SH$h&
zZrV0LKg7V~w6>V7%5fRZXLQadahb}94Nx2wNP0h*XaS2cnmXzHx{1qboBD%OAL5*I
z*L5ArYqfUa?;YJa@ss6gc;53sE-fvMis5{HeVZ%!LrUps6jBJVil2}?*=+6A&@`v*
zqr1`hwY4?eSl>M4d7kIH`ReLwxEgMoU&ou|-lZoX^$n5+6J8@o>1s^Je^=z797ujn
z@1DKRAGCV5jN&wP%T~*@pxZm%9Sn>kbQXpIf@Rq@hlvO$Wwf9^0sq``97_Rd((N`U
zOG}7xyuA?wn@hgW7_5~gSuTGzf7PZwk%4eEsgg9x5Ob?4d7F+FQ^!pz!R7j*v}%+T
zvocxrU^D(_&7&xia8|k^j^?)HTeEq%$Z=5)B41`o0qA8#tK6_|Tcm*jy2)eg>b!e)
z_~@1I2@ZyiOMPcWO-|LJX-Jg<rVQCV9;24s@{y@C>o={IOmM0+m;idV?l38lDlmX?
zE;$SI;hT#KK>4-Ja1IpZGY@M34QN}3_9WH!+A>*_pnG=;3I?EMRc94|0o1XN-fe<)
zbI0#6&J3%W0Jm-%0O~3b5Vs_Nita0&hD4vbL*W5^l-J#djKMMXukH_$FG&S~xXk9=
zIppBJb+3a$D>h)<1O^Q3HN8>Sq1X)hVwHRGNXTqJ#(A;u0j4v9d^$y5aQ7VLYkh{K
z%%oZ)=co;291R0GLqnrrs0~zrN-<}3D(SR8mkBc;=9q>Gib}Cyx^qn&#~@`Q0c0X2
z#90Krn>N|m{ys>3r(sS{nP9BnWtumfFwN_>4dXUT1E^*y&Kc(FHKuv`oM~Qv)-(_8
zndXI5A*i)fyx`zl5I;YKecm_smE<<^v~7G{4vS1QQ_Z+^X73h?UoIwK^lPS^e-l5(
zdcX#-UuVI>1~lGa!QwZUmzTk;p83gVKJ%H++~55WejMyyw*l-0SpUO9uosU$VBymH
zZh8OvmqmP>T+1B8-T9M?$ECRP5iOpNCuU=*&qSdTmGx(H_pFKbFe2yp@Zp3L`tNHU
z=FE@_1gSd93_z-msg>H@@1J~_uC2{xYism+#RVwm{?M&;yL6)-)H_`&r2WYHaYeF+
z7h#WFL+&NdByS||Vpu57%5;OK!4+LFzvKr}Ks(HM(1tM<JC9dcWQw{~lukPsqax+d
znMQyN)1rWP3nko@e7;B*#Gul$ImMFP#Eh{Mk@oQ%@bfU<RaI3rGNZM69J2xIE<qe0
zvk=p?)vxyedXS7-VJN|KJ4Ff|;JdBi)Ak=rZO5_G#Bm%a`5l;sO5u7O?%#j$;>9G+
zR^lj%R%=Nj`ZA8Akluaw-FNdONn#54tY<yzSy2<TuEziamML8XAvUi2dEe7U9UQmD
zcwpOrEyJ*!<Bsvz;)^c_$EnvH2jF_`n;QsH84UogiXen29`@o046`0^ZkX-3t+k<B
zgm7XVd>ei_{*d%x@*46<3oRzn2ho<4WO^ZyU5KK*E%QvQ339D|E(O$uUqPRa)0HA3
zRX~CT>-*@SjAzw2y*%iyT$<;_$gFldN{jNzfg>{}Lm*6(m`_e0hKumc(cTH?EX*xL
zQoiXFIv!$<f@l6*nIuDP*pAx`9RO6RuuaBn=~4<kLj$#)ST~||!@!gqhCz{Xi*D4`
zj~&}>q)8TDpKA|IOSq675Z86lhLlqV2!cbK(RVAUGr>K|2aYcoBI7<Ng$PuM^9b~z
z_Y=3#D>8#pe3z9ZK4aZ(*9<H(?>j2+000xmTEj4Q9Rw#Hm*r}|9R{`l#*KRG@Rfbj
z@}&@=fK&C+y2ZdM#yhjO0zeFgfsFEzLdc@!ALr6+G9C)n^lf{(*T8KsG*ZTpi+d<z
z^o4H;cH8sL!FS-7;6vnO<iTNnQs^?=$e{!a7>E#yW<j$k&7F0Me4Z6BtuDbCdSedS
z9z;(Cw<T2RFm0vLBI=G&48H`P>w3`N+S=Oc|FO5gS(?svSMoRlNRw`_(Gak^u^CZZ
z?RJF(xN>uI6I{>rVB;zN`+`lN)QP*j(6XX#H+GQ%*yC2yG9AtscTB6<8f$%MYkOtY
zbt#o<#JSc(DQy5IlDM{96!aC(uPbdB?W0HAhM`sc`Q2Joue%8}qo@gqTd!AXj}S)4
z!Bcn&E|HvEOYSC*5HiY(vdjyL1*maUh<DcLTmY<~G<RGU^_4W`qpSp+ek7g8^9kNp
zU*wYI;L-HIJ!f+~1x0@R*!Aa+9nTBE6UQIAf7oAo?uq5Jw!cn~9J%YxokNHx8CBW<
zjN>>$jFTk6-`*Mw;7=X1>v=)YvMgven@R{)YksF%TWvG|8jaPp>fCH@AKB&l%I7z;
z;j*)pkFeJ}b!CqNqYMO(qLtM!;0&JhyS=5nswx1+;)HR8b5aPJ5JCtWQ-7zk<VECh
z{=|%R%xF*;`1O<NJ|CexuFT6xHL-Y|<i()KCTlpQ!W(cmm;en7JHbF4LTfWnqv@4`
zO?M=l&SiO+A(&<k;a3&Hbse{9Y2eJHlpBWOT8wr@*1?!Y+NfE&=DA>+eiB)3>MJQ#
z7)SvI=f-*9E8xcvbWrg6W41J`AwVt=F!BB`QvlcBfDM;1Pzo4xpXE}9?6`_$rk-Q-
zQS5+xouTcXVc-sqK>{VISEHLTLc>7jI?g(r0vHVXmCL<@&)^ikL>A<3@*AG}L;-kH
z+)xg<9EAMWU|v43O%8>qr6EE9>OX`sL#F;h0{INGA0l+LajXmX0ofT+TTC-*aW3Vc
zqOqUh68Qv<nrbwPAuj{KZ)L@M%D5AR(Cuu1@OqtGuNSp3Wm|{tm`qLk$}8`#h2ieb
zTVt6IcMqRAbAsXf!Cqcp@9rG09ze>6k_7@~N4ZRdlt~8EW%!zUE0<te+Hf>OpooU8
zIr9Jj49;yu^@=nyKtMw#ra@8LTxs910tvvB#hKJ51VRkpeTUB3jpksZ(*@{uHkR9M
zxwlCJJ3f2%&btn6b6}{V`Hx?B_V{Q-X@+$FQQt|^>BCRIo>MBuhid^7!NRi`RgP9n
z_=++bC^@n@*NRcixJfB8MMVbz5WL7Sgz36lg`_qBa2Yk~ZAU3P3fm^b5RzH`a!|%4
z6;>bs+nRqL*(NuUhspmU{eRN<<Ya7|-&23?fj%(AiFS5cUW>9pCBmHIXL~-YdV|;h
z*_=*T{9pg)+Zso*@G!aciQ<glZ9|IKwtUFAB}nX&<92z^<F{@S-Msbq^^b+`=lG#}
z(rS0u^oBZ$;-Cio=5Sl5w;zi66|Y8o%`2z+o^zl=97THAuMc-sde1|Vyz(^wuYLve
zRi38%_T+7!=4b!Q+h<)q{3iP!zjc$os@J^eQp~S>&D)y4oxSm27{&VM?W%A0)vw_A
zq1$P_o!0$<ilQF|JN$vUttt@179K6#&RPf&Yi!%?i2p^u8x6_U{`BGD;dJ`ivMg7t
z^5o><>2!K{_}bIc)6;6Ts$h72etv%TL2rHQTW{QW>#tl~T)nUC@r#R#i(h%`jT>+6
z@I_u&Uq5m$>b#=!sh!X7e0k^VJ8$WHf9L-p^cY+{1KAoX9rGu5K)q>GZK@RACR3Ki
zOaesaA{NWLCW@X$9R+UT0DWL*AcC>c31Y>xNg!@-)z0Pzh9H}Bc^!K&1m~wZ#kdGj
z1hHm3$lbi{5D)Js@^k3^TR+&tD@XbH^=sG9^TVrA8gAA%je7N~mAP5pjFP?N&fndT
z?0rw3=R@h4$>m!nrYFPv)J%SSj|HD3!+c_Tm%o|k`T6y$hrfS#_4;{!ZPh#%{K0#{
z`%9``ulHPcu?P83daGBz_*VPxrjfqAG2=YXkFH!lKfiwUF#nRzlD~SGpI`srJkQ@I
z-m>O)A=5v>4t$yJh(2uhKmLKv&vTOlBXORE2eg7?4q#&=gdID@8W_&685VNHI&8dM
z&6v(Ge}Iw;T1IAy{pqQSdpm?&0w_z3#jupfzX+v)y0JDnRGq7v^YjO-CgPizQwlIX
z0V=pMkwsf_pu<-uqdb1u*dv7^GPt$(?Sevb!erKoFnQ*`Jqr{@;Q+_r#Y_bV`_NCK
zI3!sXHQM$qC4LyWL!kQz4qY_#3E95YO_XIxh@)g*0}e-1qT|!keq6ZW@O0XAZ3rRQ
zZZka{4qXxVPfw4DrlTQ(&VmQ<A<KdX@F9z>b2cEU$i}Hg;<ftUn8p%E`(?#RmL*)3
z`*9?3I;@H@7-I~Ki*f{CHW&=D8#i9vYqFxacrPR(W3s0A>Kiw*!DkZ>o$f_>7W+`v
z40KY=CX-olLcnzye4OP`PrD(Lg(5o`j}ervuwSPk0ECbdr6B->kk);yv;rKD53&ND
z1|PC4cmV(3hWQ-F#MUv$-^;xbcoF(sVO8x}7%$U%RfQ_|2$oXDcHe?;xsy9D{673B
zd~N5;C4X?1M4>#9)^StK0eNJFr;E1aZi;5TC3=5&l4)1%#&oQeqb7}muiCX-@n(ph
z34956dR0ws9zk0Q&>PvtLNJ#&wu?pny@yY`K!_78y)O!!5jAyF<5Z~-;uywS*G*L+
zfR1D6-g|GkS3n3~G9e=DmG|D;od7^x)s5DB7{?(fHHD^b8lo8%h4&T33Bnafl7gkG
zFD_oLUU6|z)xwY`=|i_SoBe~c`MleO`&P^8G|vYI2cvOO+&ViO4dV#%e5vCE7uN2s
zTz$!%Yu6P45M951=g#8F&f3M{GS4B3hoiHzTSYM*?;i{y&!^Mn>OSc9X7jUy{mte!
zWZ4A_2Zx8l0nuQ1csLlsMV59tosRC*FWiN%fxGZl_zMg;#oL`uGuyQ7db5QdLuA1)
z64Y47G_bVYZRaA)B68VYw0!c#R_?(a^H_r>Q&N>pefkoXmZ)si7YCt2zFxH3^g^Qr
z&2rVSbLbs-7yQoMx}O8zbU^1pxOHY8Tw(z6xcPcB+sp*4x!WMi+GwToWQlY@*-EmP
za8fWIZ2E*t4KrP#$c9}2`Y16c17XelCU(rqZ1S#miBBG~!}X$$%+S_Frz|m&$xZvo
zw2}aat;TMMtCqc1=L#SU4K7FfD{jiBY)~=+VP~|SU&^AwxV0em#V#02!%RenOwqQn
z9|CTDkCAe^Zs{ZirzYV}l#X_s5lfJ?JDOD~YuEfsxKJ#q1*B%Rj|h>&BZVtoD?^`5
zNo=IF(Ov#iG2W^M&Xpi~ALt_zADyz&)_GwN)1XQQdV!ZcN6I5+Z2q1QAlUs$y<sHZ
zT%c9ACyI;*dY(a}SoyGTV$qa0ZHxKfSJ)Y6k(71@v$7Afz=&-|b)s39IVhF?XC>tq
zQS54>+FKwF{wyo4Ej#6vSP9x*xEKdbpfrl$N44;5835u;Wdac4Dt8#hQS8LPdimh*
z+wZ`IQi#hl0f5&>N=dHo_f3|rqsc*Dd3Vn*pT|ibvU~`mv$E70s<Pip?jteMK&h>c
zQKUHuS`j*y+8Z=Us*OZ&Vx6VP9(g5%0?edj5+)=u#7gCXNSI7oS|+Pq>qyw&LY4K%
zDkkf^QE`=!N;0K_Bl1y^TdP2^&K-W+sUrY`X#DE6Mnomn8q$sfYfY>GQ%JtxD6*E2
zDAhhr*?$=TXwgx`f~J|qBN!r$s3;;t?>Rt2^W|Yl!!2tYmHv)PkwGg2ON6|ChmKK5
z-#3%+eGyHp6(eDMj2??he4`X8Lz0UK;D|LS-Z4r<sEAhzVD7Eg80`=XrM1u4#>7f_
zmUym7ky1rLq}2sRb<m+=Fh*&m6^U1*6rq)B&*_<FtrcmdMG%>tLom!@j8b7l0N@NM
z)M1mK42HG_=<Ti6oA03h;J7Hp_1@mWe$&-3Epo)T)3Ke-3vY#ggzxTrDQI{Sz8(LU
zUPE6^|D<krI(X-FSq&111Y0}`6fPTO{nDIQRwX8FD6??C<#}5FG*GzPKABTa8r(vr
zwRA(52PpFh*S`z=uHab1KGUJqA7P`t=N2z4zi3!#xO%g2g@ACj{cai9cCq_);F5=m
zaQ+iHIiDA4Q<!~tVL5my1?GS$%~nPusC#@~Sxr7=xZ!e>M^L^jWHPr{E>^Q!=O^>i
z2{We*`DvZY;yVxCvz<si4Dn(>7onK4mUkwkfuQescpW*^240Gf^AiTY<M?9UI_lz}
zxmzQO6q~WSI#p<_-sZqEG+LE*N`M*-dw}&WQGvQ+5>C%%i*s^~Y0T#rtoQhRez&9x
zfZr1cvjby1crL78yl>!B4$T(gdDF%G)4+1RZolGO)s}5pEv8>)Z?m3HrY!C1L`gke
zx7*dyGJ9^Z+OF1{+4P(QBR;w&Z@0Z|mdM%D-FCNII_-vrtnLu*`_;lS5>{wGUBqm{
z*zEkGc~o93p0pFd?Im2XET-$tVw3F_-9Ws=f(pJEx4dS+1L0_2Sh%JOa%jHc8JfW0
zPo!I}%68Vxi<Q#zbVbu?3|sPz4{yI|TClg<m9}I)oByq5)~sPJd%{$xwU77q)XAH~
z5cJ0aa_FA*_nZaL$~lbNUR-h?(G$5NXw4i&NdiP^y6z%;=S8`8fDrQWnGfAwO&S3A
zvZ7t1audbgS4EltYw<e1a2Ausb)T)Gth;BaD4*ttNM(hp)TK%xXnlOp%{8fuw#n+`
zZB_R+D*N-RA&^4!;a$rvB2gB37C}z*Hd3UNqC-XwAxo^WlHQj#0Qf>*L)vB5zlDtO
z`7{lr$~*9;zQR69i^F?{#2E?NcjZKiO;Q3NJOhmt1^@x=aa9ogN%gjc3q^BHjJD9T
zJcsHyB=XEU3P$p<=<yVz#IbQ9idfSJj+MwUGy@Q)`xEw4Gn}Pq97jiMU6C4$W)^FM
zXO2DV##x3aN`1`sgC{tCLB8La*coud6`I84fSRxtJqFiL3W;U!G1L}<qtE4m4(%47
z0#D|Z%4HFks?60z6f-3ogNh#;%}b@~JmAuC-~VPxgJ$c+04QZxea;yzgy5VJ+0{CD
zrDCIBEKStdz<WI5WjzuChp4dDTIWK@xL#4J-!q8LCt;?x_WZ5H2;coQ*Gg>iy}dag
zqH{%|6?2HhtmhXI9`u*fI#R0TVJ^lZX*K}p!ukDCY;Eq4V}<BmpRHT8ct-acBEVif
z>#=t2z|tNfY9|hq;^dR(E^b9p6gwY+U{+7}imXW&5h6&yC+_)P-Y_$qRU|Lcq^w9G
zjs{IRs>Y#B<l9)|ywt@$c0r`d)6gznr8FxA_DA8P-fQj96b-F@R?NmnUztZ7t*nE)
zO$b$Hr2-`FlB7T&s(T_?=DoA24=WIPbBjrHvye#qb3lreGt4~~a!H>z-liS^ZfaSt
z7Oj=?E(VTch<Cf)wKu@^5rqYjD5X{NUdo~;xUI@w16qHKz7V-oLM$<>LL?HrU56<6
zr=@0j<kK|(LDu-l;3ns1=#o;L=-9+@MBoBqZ9w1M3%!jI2q;>oxw9)zOc>{FGXsbu
zYF$zKL{<w@MxH7uqD@h>i6{kHlP7Y{TJH>e@-+m(Gdr_B>MVF(>^lb5v*0o!)wfXC
zR0|hoM!b_8!VCcM;*@M;@L@c(#yi$<9+VIpmzzkSEWaolrINx}&`b*~w#E`};*Ev<
zQUQQvxQ7UF%(^a%z~3FC9*&KXbu>mTF~ww<lN=0zMA!~B_)*vYkGp#D{W~AdzCpBX
zRKTD{+Gvhs9GL@m5eFAT^YifH@s52OLBmjbt5#&Y+s+TFfqe*eQ(~~FE}pVHJv}?U
zI6pl+df<vM!f-SmYfjV6q#~@dvZ=BJ`eGefvA0*|b;*2sb$(_fOR7RK;v`QL7G<;C
z+bfufStulgJ!9%Ut$QagzxBWa*KY0CME&0B@0)zMQV3vK@XDs+fGfG4AO)-b8^)mU
z*E^$@XEJ6P4nnQ(tYC1{pvv+&@1+qF!g1porvT}w+fQs$*sH5^#-!?WlUOUIvW;p|
z9txw``njyd*kE*It?bS>H~QBjr3$1Xgb?b-K0ZXQ<W||Ig*noeNI68oAl0S{A;Pvr
z@hF3_0A>VZFT->TpEZqPEWXmSBG1U_)2DY%bAEc~^oy+98;#qo7u~mmZQJ(k){CB+
z<ilY;xt3(jMxz=t;zl!z^z#(6nx{IB($hPq@BjbdNj@C@HOZO{QnMB-zYX7q|1oY@
zz#jScw{Mbv1p%vYWE|0jVfSigo8FjRrK#4lr&7_YQP^aSvocNrRM0J|Y5oC_5#52_
zQo&#|IPGbwyNOD7BV(Con!XQ?VpHcIK8fg{=-y`4=Yhc|U|c%IyWf~vMEm%Bmg=J0
zW`*1JzVD0g8@+IR!wv5|ccJ<6=<OeW$F-JaVVa_qxZX@_Z7XyfkoNm++pf!;Nu;eG
zi{sd$X#!^KzU%008eaeXrn95JJ`=h7hS>|2p7^bInj1yY`R$)ds>Rvc4<Dv_?a{p(
zF1%w5c=+(`r%cntT8-JU>%;$mp`()hHgtRQ`LW}3?q>VC-TD0ZvD6*yk8_)wo98#a
z^5xBJy;iGz<tw%8YFnq`IIam^i*yi}rfCY>Vbrz7>p16@!>D6%<5t5kqLlMA(*7r6
z@_m2YJ35_qI{=;Tbb54eI_-4eXQtEXm%lPOKkdE;e7|%4*wQlW--w3~-`;C)edWtX
z!|C>Qw;w)?Uy9?n#zZYPL!S_0lGedx_zSp9o+5uo{+av$034`6A9C1)!-Vj(NUJg}
zgSq3$Is<WtWtog0(m$u6b23Ux+dKVfxhQ5u+I1^9ON^+d>FT8CP~-EHUO?&RQ>AJY
z*A{mF8Zl2QN%54WzbYds(n^dXjNe0}dS5Ep3BneL72TZJySd-cc3M`-JDDhsWEFWL
z@#A^c9^xuVPfzWfmUA0#ALKZs7sXLpWa=e1p37O4qy-J>QI$Kni9|dmwiYbIFa)<)
zuU|#SnYmUR__7o5rU+HkRCU=XRDpfgpavum1T{Ylz_h}SVLF^LtJi9?L+qI{So2ym
z?)PHW^45I$40?#S+C7Ug<{D-vG))XcuO0-DfWaDP92Fa~t{b5aM2q<y<@>Si&K$eF
z`+t(hxy89@TGx7>XDZr4OIv0gxH6h81PFYI0HG~}-~M~g^DM<$U@OzCb7i(#h~WDY
zF|@h#+cnp9E!h!CQnw7D8pGizTL;tVsX;CWvM=Kn&m5t=F!XpQMKq!ypy0Z(>o}qn
zMcf#UHvK_%q`332WH~Qi?0KH|;*#e~9=@|Ul4aiJXlU>#Y6-`2W7h>51d)N5cDNUY
zo)S*RTe0lRfy@W0XPB^_jfTU95{6Yz648;C8^?E>rYSk1GAF)EnPq{99LK&Q8iJ%j
zbAz(LR}`>;A2Ur;a-8~?Fy_C4gdP0X!RHX*OC%>3$m8T!xN_oXg#n&`OTZ=5*kAg2
z;i9m-rHP5a!_#s$KkXLfWe%-SFwU5rLY+Y%ILkuU68i9sVT)!`1!2?5%7{I1iBLdk
zFgY-r4Ecc9?N`yX9{Pse7`kq=MHym`u{epnY`6{xaK*+(`b^}~vK>cCN=?%Zd>;}>
zYfDUP*G-_+8q{hlD_c<7TySn%uA>yz*4IxXP`WnA{5Xyo?ILmvf*?fNHH=4n#|fcU
zU*=l7ZiKDopjKa9-KNsA9a~AHhT&e*u+1Q<rF3g`wN`^ctA&y4YRy*a_0abNIAfYN
z7o1y`<0=iw#^#j(u$B${I86obBXZhmy6)h$(zYE(O2$mf^*uKQsHMwXYuAaP)f%`?
zqmd0)f$jgWwzhTx0C7Fbyf}#&>mf4QYB}ymN=tCYEYo$0o2%R;c<^=j75Er2NIP!1
zosW};$tTGd$hYBocrhWPMVR5M(<GJSvP|%a7jeR9H5CokNW8b3%ol}^eHbl3s--jv
z4gvFG)e^91%_#lC5gQeYJ|~)FJaV~tKBd^>B@|vjN=aqy9|dG|qWFef+Ih<L2Z#zN
zB`y@22!d^wkGn!-8G-`rnuU+i&Y)OyIJT11UUtx-SCB*Dlf=`s?6>Mvllc=i;_gQF
zl|pDgoDroY!)0uW<DOhmEUHO4jsxGr!9^z)#*=Y8p2p`=rxbErjOTDA<CLP-9N9tw
zI-Vmy3T;U3BG(23Z($Izk`&Yvf`j-L<&+7gOhY7OwE`gJoNJD#6a(b2))CTAsiiDS
z_Hw-b^7;Us?z%0w;Lu!KTV4smnk#@sVZC?k_};M&QpN?}|8n5Eh+z;Xf+>}Rsm(Y?
zW!rAPzTV_o14L2nSm*yk95-lS20^Rd;K&#U9y`8aE@zpGAUJm@B2!W#wQZLRE`%2V
zS5iod0Gcv{P%@xJ)@o_+gwR%}+wFRQz2UIN1c!bY0N_tH+dvVt?byI6#1YJgLj!&f
zK$<?FuZ5z(l`t4Zq`){6fV5v`gwHACf>Iu_(mYnewaB?NY+EX+K*Zygq0$iCF42Kf
zA3!3N)V5sJ1@a=4ZXA34UPr2c3(iH9);2cQH&V;@8@_)cN#f)fvn<CJVhpw&c5O$=
zZtDB+d>&BFXw+^evu;nIq0zRY+4T>AvKWGtp=AR_r=2*OGt08NoX1Jxo%W_o*>*&Z
zd<^IW*plOhp@r70gbn(+WFm;1Lv{t%Lo|&TArVpoV@jKO%zmNk0#{J65g4S54S+g}
z83ujEo<Rz7M=ZjEl|@p$zX9PvN}B1sHKcK?Xr0@Q>UW}V!OySfkNf?;P#tggzoXG;
zH2Q;IABmmAjmF`f|9AZO@qYjK;OFL$2JFfAMNt$T>-Udc8+r52Ejw3*;Z-}g?A(lF
ze#JSs49~Yg7cY?=eqkiIPq(-!wtx&GNONX!D*4FLf3G9f8;Fh9VjULpGlJK)i>3NH
zP9|4fIhi6(rdMu{Mu?-)_SPr|$VXf6p4>L)dMeQPS<f5(`Tj3mRZb8m<yBXf6Wsqj
z80A&;o4C##TQC*|O=^0HidDrO6;9K%a-=J0xi-i0O%R1ExRcj1^x=Il0K*K!FmxQo
zl=Qu*8HVsC2&2KqAPixB?Zk<-HH77*-Df@P{9=x{wSD8)ej)OFsTgydFbq*|UOEh$
ze0`~CH<E0yOnUC+kFKuCrK3l4wY`1ertL%QYrN83?7!htBqyf`;bH_TIL&Ov0&=Ld
z4Pd*B-cRY?L3}SsSiL|c^;Yn%<L6hOn(jUWh^_Y1b`K*o8cR!Svkc&AkY|H@$@dur
z==V0pjf1z2xh&Ve3P*t=E0(TW4Ek~G1vS7j4~l!PqceL++6@EGPXTbTgNuL6zaW7(
zlRi02?k6uNbsmM{o}t}q^1KLmUO@Y?GLW*$xt+BoxBTL)Jgv$_!9B$nr}6XI{3eyh
zrk^1!UPg|{P>}qLM5N_s94GLfaHGg|-S@gt<i4og?kpAEC;m0?pYX!S4Sd(_EES!f
za-9ADeBy~G<~hwRf0_8<z<1pz9LJ3!_r0$BqHa-i+wCVFdf)@@`a!$>Q@(%m6Hh!b
z@fP#Hh)pYUHL3|;1=(xJE#zVHQasc}T;*Zm6StWTh=%70Jw@I%Jp}?plA-MJI4!5A
zK5$1}xJ6G_E;Yy)7E*vtIvKCAT2#Lq1fTZ-z0U{1r#(dP8Kz}fVY{Z4(zQ2tyWK!=
ztx`HuXApGKl^|%J@Vwri4lK*GvdpwBGfR||GEJ0}DuGWa^-MGRlgNCQQqQ_k*BuXb
zAn<)3AIP$dyPo6VgM+~h=(yMi^nJ`jY}6Zjmu|g*K~BCD^YaBiMhYAsEH20+{3W}b
z?`X53m{UczjDq%2(3sJbP_HO2+!GO5IJ>ftDkssqF5CjYAocxxCrz~5Z1`TY;i)9;
z<n<aL)jD+Ipw)~HH|o1lvl*rxsdZtkRy2+sJ7x?sy#$3Z7#hcvY4~lq2J5odlk2c1
z+r!B5-Us1F0>8atS%{Xk()I(0lfG$M^~NmUS+i|``5eHu*N!bs)0)z1lqT@#CqMbg
z459ym7xV$MPsEO`6tv#_hAu$&4R3A%479;GVT6zqcoB|}gj_=&AwN$DL~;4OBTc+m
zIyylIxQ=G=P-?Z0mu9(AJlRgOz|=@vj3`VQW1R7>^Rg`GX(cN^N-(W!`eSL~@w~K@
zU~0wFig}IwmsAD30T_#2U`#7*Jep-$OE2X{E90&g9tYFZh7^)(!8s@ioC$*qDNJpe
za6I%p_9!?)P--B8*8DvDx44g7xNrf3z|mT}LNdlBcMYxFAb<-OFbG|tlqVEtObS<N
z<%R)XfWrWWWon73VR8!DYhLr3XN||tdQGGi{(y0%>&d7VhBQfN5T->^*NU?rpjHt_
zgHk{yghkm;GNyI%6Z*!Nzx?H3O3O}^fk>sH65Em{y!>SVrZCK!HYG(gRAO4n1b7)C
zNc14?QH5tQ6dScU3*aR8F-<Zg_&x92ouZh{io!w5W>&AeveNBY%(hVWKl<}YXeoUY
zo1wK-R14{nO_`0E%H)IVehcy@{kx$;hitQuf+#eq8jF(-x=xShi>a_lRLstaeH!iW
z=n(VxMKLc2#q0pu7ndRyF{G&MtjMR6>Eu3*(#ZfEr{ljXJDno$_bsd6D~e7>IMyo{
zqSGmgUf;6%{rr(i6p``|anA=I;5rmiXl<~3_a6E_=e~z7z;yuJ_c($}eUHPQ5l2ay
zE8!2WG>Kxv91hplhePwRvi(0p&tn)j>h4fxg5gB6O3`22+_;4z)3B_{kh*ak`4)hC
zg7Y(Z;Ri~alLWD|w=a`bw1dDrD?ZGvUr{?nE37mKaAT(Bq=NA~-}%lrBLF`o_~UA-
z)Kqn!&BX~0i(misUw<7k#$O8)rS@GZpVfH_kBw0O%lG?3>SRK4`%Q+%2-k#YxIntT
z&ie;l+@DPUY&Le*CVtAM!Sg@-HV%LI?a+1{JH%mR+m0O`9654i_xF<B*WYo+9lJ+%
z;WgP23%}+z_S#`1v~Ae`+KaEg`o&+n`s%OIXFl_p4?j#e=@VVsAgg4HoFdmCH`SsJ
zi&$ksOnFjiRZdGEeBZCiIar5WRc0lGl@x4U8<(=2QUHJp(e+E3)>+@aEo1n=-FM&p
z%$x38de+NszIXq`*+0H<cFq2mXV+bK-L>b=F2Q2|f5PIzb>DmQ*|TTQ?!E7hqwr^)
zPd@MH{sca@KZW1lpU(ed|Mv>`;(ofbv-807zcYuw-#-nX`F+%vVE5M^eDFa?<}c&3
z;>w$N5w0f=X_M!Yi{vNCU$H2|<tePl(3Ky0`7=c}%@*^lUL7s6a#8Joxo-xUDsd#E
zXf;%j&p>(%G^w(k&a7I<ov73q=`D)9C`H31Gh$sK(;`seR$2Pqq~fwIdlOui5CL&a
zhxQ3E63`hm1XAL-%+?tMmjwJ*5N82!EX12P^LhEPRIQXRy|>kBwYYWeg|pcVr=V`(
z3M@+kstAN>Xh3WO%rs@k<(_9d3M|W9Af^(n%oVr`2prz;l_SQNN?HlWh49CfRsh;C
zG=^YX>MzV@tJ%~jz1=2Is;HrXuC6SqA+_yD>O6ccXZ|xWwQZ@;x9#<pN&r~<<~P6D
zXngaFLGaO!-nHBZ!+<EJ)T5%l#+XBqO0(HgV3=^_gi$8J_lL*8aZHf5t*kf#(*TUZ
z0FX1G!+<(FoKgVWact^f5U4Zo05AaC7C;dgV+>qQ0qFf`Xgreu0BB3%+Bq-{B!nO#
z2bh5L#w@s=$~w3~LTOH&aD)iwBR=LjN4AVTG)b`=WAJYkfTzwb3~}Byg_rLC9?bTC
zuWmeMJZnc0Wb5mjwLBlsk_31hAF0mTQvUKC&wlpwS<kw~FoL`o4j~FV`@d#8t3{_1
zo?4o?9>S#u_=GQ_FzST0G#R8RV7*rF)@z7q8U)~X!|62aVzZgqj&1*aCyZeKJH8(v
zCTXLC==&}q#3gue8GZ<t$(zXkB7X)yi4uPce;5A@|1ULYyFC=0(o^*Q^bMWPk@qXq
z%vDj9)Xt=@u>SJY(#=LiaL7e9L?E?+wkm=QaOIdxXEN2vBBmNS*nzNpU5JLQ;%3*D
zjp8|lv%mq0Ye<#Sxwcna@0RKn!eXz)33a(;RX4kK*>0tjnM{}>i)V9A5C%28;7Hy#
zb1vhSntEDJhNR|%yWw2I28^UlgoZA|k|qzQyre99w`sRcB#@By&&p&vn@Wor1N>vf
z(sFXJP{b3_YI)`gi8M&&`({Z=k`)?taiU^SPc1Ck<%$>#G_`gBbwG;0?P855)!Kws
zwB3PEdxvS;ayxOSY3Gea4)CMMUXtCst=IK<Hs`p4MF<g6f3Bvx*6!v;r$$t-?gJaZ
zVoyVN+;<hERij0vVW;!lgb*)bSK6i(N$hJ{o=-Mfb*?e1WbhdBWAXbd+3wbDy{Vge
z9!;ZN(O$+y%mRjyJt-2w5zWN5M6_wIN824%J03+jTBkDd94(u%x2SF?hebg{5%RM(
zY-AZUS%+uyS<B*IGMh^cO+;t7mf+*V3S|AXmdSjsF3{by?W_SGqy$Qr1(IT`{BFSH
z4aVn|=DT)%d|1Q6)l7J5GlQ;^@#V^DEH%Y(;$%K==k0p4B^m)FvXW?nc;3#Up<dqZ
zxN-5QQ<lOA&Nj~ie_2&k8h1(VAvsXm`XUa1bri+S2U8OWyV`rHBBeJM0c&PZVffv^
z;C-SF#u<RonEE17L?bY9OeLUqR~~rs?e4voh|oktjQVlKqJZi+LO%)&jN^ecAm~vF
zdG37zi4fU16de$PDUSifu}(znv;W{xwlYTAsUnV9QiG7NF(%SL;>h@N1VH=qD&O;g
zL{Lt!CPd{KHQLl!r<|*7Oxj}<MFvnBG(E%;OX~oxMQ)koE{4O>xggAew2u;joFIY-
zJyhcm2rr5;fEW|`OeqtE-p%v#eh(olv`#_$aD3kgA{l0>>kkG+3D6r3_m#0)k+IGM
zVGTeHI7A?-07$)&hzKal4#suE24YcIy8;As6!i|SUI7+3?_W%(0qe-xRLcHD)-|9{
zlcC6xVv5A4(Ym&z;@EaaArP`b8VojBe%1QWh->jCX2+zWlQ_@PE)3!bfr%talv45E
zp`?2i3*tShk!KPF<w)5p(`c(LDHcHUWURAH!OI1bVvQs|T4s5Sh;cxc=s~NPIi(2E
zM<DF11tMgRAb~XyVGV##pfZM)0x_|gnMDJhc9|0-U>GXUY)OOGdA4r_vKWI{_wypM
z?}P04hlrH522v;P$F!<Q0IVX9-XV(83MJDj8Jt5Fg7WTEO@}B5fr>{n3o5gDO$IUS
zWk6T;LDR7}#mVW_TX_A(+i68{0$C>fQFUEMNKoe>Q!o>nkg^eHQx8Om!<3a0D=azA
zMozh4y^b7E<i+@=Dg77`m8YN`97QpsARru$MB-Ln3n3DXC6=S*q!J+UT9I$VGB!y$
z7{_t9249Cco$uuiF}Eo*Mi5A)E{-E%3{b5;PZ`Y~S%cIAJv+Mm?nEibPclGlZP7I%
zNt#41%bHrpp0EPcI^7#NpT<=+7>Gf@NRjnEqehrnF{i)=5&#e*S(^5K-GQm=eq}9y
zp;W1OFr*Qn37#v70Fau7o8~;N=fN`o&?BI_B0XY&pn`JC1O$rN5Rh_|mt}yVMbJs=
zpsebo3zQDJ`-1^$jcHLvq>!2<1AzV4Mp5KB;h+#yL;(m%tbupd-EJEDaA@^aooY}B
z2w*FvjOmq-G7vc;B_ae%Vx={$L@^O)pLH_>fQrQc`4B9EfG|RzFDJ+YTci-8*y_&Y
zi4h}v&0#(p&|v;Fw+^xltYKwqiP->hr;~Qzg}d-0aJO>}QurYFRz!S9{9gPE>eGwq
z!{{^Uo9TP#t@N|j&<GB2xRg+DOhrlp4F+@8kNg<W;zzq_IlJ+Vx#&uxfhmsgmUV?3
zrnhL<(+T~*D*mjf-I1&ot6EQ?Guy0POF`mYg}5Ba@jIM7HLJ2I0X>D)n@ziJ>K#K4
zNVHgQHea?An}~7=YG7cMthJsyiDJ{Rkd<J5UJTl&)9tQV4sjwFRIxc_Cr-tQUxo^{
zGLR81=Wdtbii?jWo9%90VJ4mBY8gha-Lh&5g~oJUl}+hEi`YJN^<t~4QYO2#g(C5p
z+}^EsMAmLA+ZP4g_3nJTs-L_$pIChaWHUBJyPI{pc;R%nYPV9bBV5)?)uQF)jSrEN
zy5e0eN3j-VrnBi}wOH&x`j|{7Q!eNmyM)_T==_UT>zB<n^=`3RZq_za;f&1fSSpKT
z(vy>~>Qz18Rx)$8(_XH&KK*q5{brh~<=q0$P`YTh&EioigY?asvf^D?2PE63l=rPG
zR);ZipPk8~o!3%#YIKb)G8TTl8Qez~W+3qhJ^PH*GHGY)ORucTRjr@_gGTA1owhSS
zHPBvZ%WV3d(Yi{OZ7r)}+}0aS6rE0vPpXTq(cZc$EBe;xpV}_{vVo}yXmOGAA8t*@
z?bf@kOxNXTURKH){ITfFGh%O7Hf<XP>C$?)-h8}Oo7sHP&gq9gvRTiCXT_&tp?K8y
z$GE?LdbYohO4-T5@$q=5>14G$Izl8n9vz)d4%oi^+wa*pIJ1a(o;bkSwp)lzWK0J5
z14tBk=RG)72o|)WIMySjhCm5|V&MblLBwfK2#FG{9boVX`2EKx0D?fyMQI2?-CnP+
zby~S;dOiK<qt3lPK%7@KlQp48Gp)h)+1Xkc8%h=t_l2Y{tF%KDu2hILu~=<~j#haj
z={Y3QD$>U1U&ey9#)-JEfe`FZ)H$0j9SM}@S%y(Gj1+@EV+pXT>RO|7^}|I`F!zlj
ztVuBsLg}^FSUWReF6gj*^{llqMuG)o0#L$G=so8^ht#1T;H(g7-B*(Op%fTfiej~@
z@Ek~%k4;$+fHI(W5zd)QNe=1)lt}L)YWaw36j=i8=yg1ITgcF(EGJNaC?&Jf&XG43
zPX%(;CC0~*8fpzn1(fxUaYi21z@_tL&=|cB3Y2m|>o!2DKnttYR)8Z!?sfNP`}+p1
zna>^<nnyK`ko=v=>3gu-92|^K+!KVDh$0Ng0`G<lU}q1IMpl6h21MjxaJLpS|7s|x
zvx2)&<lSCg0ID~Cm;x?}rpXfOjz(35fH;|4xjLT*LY-x0w=Af<whKmMpaT3k?eFz-
zK!9%CMDYl}e1Lk)0wAkEHBY7x3SP<~EaGssmsp}gO{auNQLYHUFYd{Og#q5?4gl&r
zYX)@%e-74J=g7SdYn;W^lKJz1u-Tg(d7UK1WH>~ChhFxPfM{qg1dfT_0*=O`5doq^
zE<^~5^4M7-)gKIo)?=lG)c#Zj%#=bv)HaZk*dYN?&xKJ2fV4F%Q{;nI){KfYi~SzD
zNg7vt&N%1x$}HEf2L#?{7@mp&q;w*@N-P9=9#$z`qK6_%6PgkQ+)n`!9A$nlC9M(i
zR0h3{M61j)6O^c|wp!h7;SnE1WuVeb5oqT7l?R|QqPmkE&JtT%!!d4K!o>|NB~Gh3
zWRHx&_X4FPa;%^ZKKMw<;aI*5l#OHK0N~v``Ym%5tSXjL81gAnlqR$ddAOmCw1OaV
zfF<k2ZOriU?*LRB=XIWIK*@7f1X|ZbT6@PDuMj{S6-CzNA0iz>@Cty~nLJTxN&+a9
zrb@ER=pZ4Ow*m`*FuV%_Ga3L?niZ#MYD}k7bvm8L@p(9^{}FwTW#wtVyYmZh2q*9&
zcrSP}d>VW!{0RIa{0-JP$Cu;B*zRD<0~KrMTK<=k&34{$!=#HQ2xJfaW(`CdE;h1@
zwwC3tvG?(7=G%D*zFK%4V!P0SB>ZUdwUq97UhlGc0A`HSYSC^RO6p)dI%VPM`nun)
zQp!Uhf9z-k7)igWj>0*fAI)}kvnHJv13-2IE1_MlB>tBjw@q1bwcX98+h!v5D#l-3
zmiG@1oMT(wO{Uw1tF3G|ENkQz5XsbLnFQB!%W2&-lX-n$qr>kEw^duu!E3HGo0g|i
znqnO?oYE1`Paq7B`OR96Gc#YV+ICw+1^`eNZK@)308Zm9&nMGeGnZ0kzpBpZt>Tkm
zpke)yc!`Zza_!I&77o&zt*|atc#3E~rvsLzWc%vfY&VhbJ-6L$SF`!7Y5Qj8i<X<_
zy&AK`&G`hcA^S3mPm9!9Fi%~}n|3$v#pe_4T+5=`?pU_XL>BF0weiJ<XVr!HfHB>B
zEpVgMfP~5!M2n1~wEfls7$Bw0nqT{jgTM&k9a3{fHgAHljg^5NLWu%vl~xhVQCww2
z*eGkN0hQ=5jX~=(`Im#-n0&Gq$7woDE+CGI%l|q9k3rmyJJwmOxs@m)vFI=YS+pWl
zsFlY&GVvWbMMy@0fP}Pic9+$NN`O#rs|HuHT0x&c%@3v?AbNXfu*H{u6s&UIdV{}a
zyf=v-g&+=y;uFN45le{Vj0NbCdZuZb_-Gx!ROFHP)^E(BC<?QAh@$8Sa$(`TU|}ww
zI6pr><+9?7b2*hEA`rne4uF8*QIHgvC?=wuSz)LEv=IO?8sN`FM2b}HiOfaiCD-zq
zh$XWEuK*Py1tKA1wgP|=Ww^%zz`FbMDz06Y6u}s$E(rZvRlOy#F~(TN&P0_4Ce0qT
z24(}~TXxXW8Q_V(cbI%dL@PP)1t4bdJt5$FRMXXzR)sZyd8Gk>D5~oye))F0<%67=
zzECHCD6&vh8YhUvXbgz~2z-I~z5qI0mB9$07DVYlwV5RTWEC(rSysfNL@y7zc~0oe
zMnkSz3I-4$hxb&%lxS^7$v*SV`2bo(4>9l{RW@&$1`tFA3?XKRDyhsXP4#l@cDr4?
z)_o<!5p8pmL^0S09(X`zGkKt}_8WocN4Iik6gdY-NIA8oru7U+QA7wA*4oo3i^aae
zn1xM?@H+dn!4xMAERd4zqZokM8jK>=QABov%s9|XN;{{WTb;)`@80R`(rMV+wCnj1
zjA;%J@h1uFiKZkJ#8DwsGzhzZ3qwNNs@7_`tAK!-F6!K>l@DajG9=Y&hL@iY?{fM1
zg>9>a0^a4;4ft>HesFcs_a2_vY&QG*o7)9KaqqkMIJUF>@#tXokLq2NvTtMbTcS{0
zety0TuMfOS5u(fgOVR|s2GS(C^XP0m8)N?F0>^Kx0gKOg?zzuc4vdRGHl6D_hxX%b
zal;;7pDZ_RyMRC@QW8`h+%}tg(>AHXB-qhH5I=zsq^mpBY2^q#6;eZ)17ebvnUOo(
zLJOrD5kuz>F3P_iisHuJ{$9T<ODuP`+No!qj?y#`sSrt?J7-N(8~bB}(Wp+6q$Xk=
zBG##}+8g;APaL78cYG3SoX?98ytmGmWm&ox0k~Fxac%$c;upi0{Xu0YiyM2S%A~3g
zA&H5)#*#zj#M(k6S?--Rb#3gefl{$e;)Y0b6j-OMRox}UT=!jN#u@=*z031=u-5vr
zDoO{>;KyF)|IdA=)9LhIc%ei1JiNIRciPT<ozLieb>}-e{{%<yp>P-e2Z#7dd_8_5
zeh2;v{tq3~-_XC&3+hm<)P3reoPU~c!Hbr#=Xf(lBSJYkah#|tT+Ou;8G4uT4>%+)
zFB#=L*9OMMBG!v-b#MjFBwy855Q%rpmpdnRGJGZ2zul~Z0G#8R=qLbkv^Q_2tEs(l
zl6gYD{+N10tzLoeLstf+lEri;M`pjPAmoUDPF9dWw4Q?jZcbbND*SFMVu8$8>huZI
z-J-2_i^fZv+jBo}e**6nJbmi44j1qN;@Q022cW*h;-;=RduQEz>HMJI*yZmjSuNU~
zK||LLSn7IMYgE<N)v}$oTWLwHDOoLctF~;pY7?A*npm>cE9okMxo#SWv=RD2G{oU9
z5hfHEW4{mb3dzPP2#XG9UtMO`wxzHb0k%OwtI*wTstt}|T|W}lZH&rn2TO)O5Eu}h
zH!sYlP1x<enK+!a8;fSXnvH(O(pEl+A;*fy18EWE$f7Exj7~D$wcGi`#H;9u1~26n
zxS|j|JdDJnjfQ2!AvwJJSel9M2fVO`729S=L6&V_pFSqwt;#OUcYv#CE$RK&#fqlo
zKx;IcM?BS9w2*2EJXv{c(d2?PmUioxNO{=RHKyNOnM)wl3lE<U&rTC8PR1a`<e!&`
zX+CP{HJa^-D}4dw>#*F02&O!f%n5REKmj?7>~e(D=uoxm`BD0UIy0FQ<=1K2-6qVW
zE*A`4f-2)cVZ}93A&3dLAz63)cG^DHTDC0DQvD7RiVA=xu@O8{<6o7B!>cE4+k%F6
zarNYUH~>KSYLtp%n1N<uSU&sE%T7-jX>t9=0i51*&r9!_OaK~y9)Z^2wR6b8N<kF2
zozU2=XwAvVz4yNC9j&q2r`{L^gdR~%JM>~R$Vw^vihR;R<&`qRVbS<llO!+!Pdk)2
z7SlR~*#Fg3P(zLZEfQSx4k*WDU~wu3MXLm>W)XmYQP_TiiINNg;J$SckN`ny_@tWw
z=thzU89&M)0C-V=2YW|kF;;2kBhpYr(ZmTVa3#@H2Ea$!2!b*B2P?OL#&`y21vpZL
zj<>okxW+(Wj=&bV$rvA#gjlKgloSAx0rX^Hm=uIRjI-X80S#!i(pF=@ybLocBu<dr
z2z98Js*T#MMIeP@sP1U6+~?4zNEik+`sYcdp(@rST!7C6DRRtUrn)Roj_*I-KN!C5
ziJp?8sOv(*K^EPEj#)53Fvb&r5YZ<404D%S*Qfe$7*3i45mH*|%DP-3E9PFvisQc0
z5vVtWI2(@uDAPBaOq3#{h+&v-W^5<b{@(u4_U5et;N`o3vf${`F+^PZ(fR(K2@ZHP
zzIfun7tLk>2a`K@-uG9?IZC3Tn1t^^6?T4Cqcm$mz4R<-_L@kUB$k&Way+hTt)qB`
zoSbDDsiNoy2K;U>*uE-|BJDnp4km|379h0CC*lwKj@IjrhXp{kmpLENI89TO5L}jk
z?DfDXwup}P?&7@Phs)n*M9xxC#9*EO_&m?=@MxgT`;q0C05%Y=YD&Xi(ow3B5uKQD
z-%3ZUbb;wW*#Zp>@#Jf*05U*L#WX-%up1{u#lkvI9<+U<2%*!rA4!SRN(n61T2?Gr
zA;Fm<&rKX#zR~l<19)pSrm9$_Jf%n&F)J&ijMxt(67y12l|giYP=JxjR7Xas6UyA9
z%#29+)>p2SSBT&w(+n6KDrnTyZM1r$PGiMUj5aT9s!h+}{3J%}&z2`sDo1!Mc72P=
zD5aZZU@iF-ql7R^vsYP#DyV$~?SNG<H4Z51ImRhZYz+i$ujfPf|K$!DKqhgLg4PgY
zgctx#@Jb9imyO509yy=csz{Lt9)=1BTAv*mot7mrYDJ_qpqqI?&<`S__p3*vv$r!x
z)tDKPI2DNmwKk>~@|$ii>MC+0eQFOQL@P0#1x5gGw8DEvXaK7F?<uT>gM){No&y79
z=ZKVNvkwK$2<T8V;K9LaxpPF6o^RIk8GfV35E-pCu<o4!pb4asxB?OYt)YXR=7qa>
z0(U!cr`PEm?PND?rKOhIrY&eSUG3VUU9|+i^ys6H?qB=)zuo`Y&)MJKfB&C3_}lsC
ze(vYK0p8*AT>+lH{F$G={3`gIPQODhd_O)2KheoMN1fMn-rV`(&ewH*y7LR2zk@kk
zhX>(Z;RE1v;LG7V(BT{(#Bayn!oS5A=sLZ;(`hO)$DRy7=w6!UZm`ypY09dc^H8PR
zrfD--&exlAr}9$8&2qHftv6dQj6?}a`95=dFnVi0og0VVLLNd@LE;6#S1F!hq1mnL
zmLaaB4QT?jeArdfgL5qGWXE0gdar0+ps{;*d{4nUgw#>Io-d%u^m9|SlV*wZ6yYzP
zdohhXd)u^GAT8N`LY-<N3{~?NKahESfK45_o35sdDOXjyXkdbH4tf5tUUXJC9A$|C
ziusFh=Q@ZEdJ64+H+uv|4Tc&w`?P5TAVCaC&(6&-2sOFX0yEq(>Q)X+W>JyJZEe=V
zaLS6b^4c(t;90OA!rKVaa0c535ZvQwYs6FFn0<>W&v}Xfkg1&I)I-~+#2aQX9l9_s
zGSoJ(z{-eYuflHDi<;rhrd@1S4%}<Lp|`l0&lVn+=Q>7E%r~Pub9s%ngfsnW{rmu{
z^B%O^R_&r$tQA0ZT>NU{cKSc?w5S#w-NSfAa)_dzA|f1VDT5|mFK(d5x~hI8j$^Ns
zkCm#4YNbvEAVs8$^aqq^^uSs(yo#)aHh}N)Yhti)Td7e>>4yzbUnx<f{<A0wXUOwu
zQ4knGWnlQHfJ3Z&k~rtRyJBmL2m_2LOrJbt&N=7(+hVsxzi7Ej5(nNn4?jhu?0f^6
zk$*SKW7hrt;o-5?$NfPRVa^<%p6#wK_8adh0(9H`1HgW7H1fk`HJvVa>JosAB`F@*
zF8+EwFaq$;u{?G86=+O0JUeTzUM*{_2hG_jYUQAPYzE7#SBpmLVRP8jD$!b57FI--
z4-chB1kU!ZU5mPa;BEKtkWKyyZk+J@YkoH^F@oiAt7{c1cD-xY8`eg#?cEA&6cLNg
zvt^PJ5@ZLH>WxbWbP>k@iTsX@mcs>6GxF^A!(v&M3tpSPXOK06dA$5Bz_+7HM9D3G
z-L@~i9J;4D3dRDmd^nEK<|PpoL`1jfkE^gD0d<q8X-LfkDh?+7Jl-U!a*PK6RgOlZ
zQMKJ}jnDRIO9b^SCF9Wc-fPw3<2XK!*8w0L27pkLxA7Dx8<yV4i$Az)MC{_mkNw+-
ze1K0EQ6Der=J9b7C&^svnfHX~4kq_l^|HwXz<KT-9?p_(Z*NqV&f}X$kq;MBe?nz>
z*}(+7<KFaeHeIieK2D2)y}kQM4SQOTC(0vNb-!O;KJ@7Ow$^ZaG*q9M7Yqn>$%klE
zUj$6R?I_fl1t4yRszC@Ce81hVnyw2JAPf|>;}V5oudC3O&~yd6U6qC*CJOi-<-X@z
z7D;W%o(=#j27s{w001zwtRVp0&N*KQ%D3l7qFhlp&6lvQ-)`JDG5M5?{|IAWO`R0w
z`5`WkimIFtk95m+1?)}16VP|qGAcP<SXhub)DM6A6QB5m-M9;2jw*`rr_XrCGoE?V
zEw|iKg9mHS-v51Qr?9_%6Hwa;de^R9yM_=x*gx3A81|5jMhH#1oL4LgqroY4Q45n1
zS(pcd%NJL>#ouL#7g0*JFKRTi<jUPcz*#mUO-1{HMk$3TZntLINld7!vR^ag|EipU
zdi0iCrhj((=H?AIZe%r;a}_Ty-+1FqH&rW&sv?9Cpy*%^{u=hs=Md_QCsj38?hyUQ
zEhctc<sbr5K`_P8=x>3P6W%z*nHu3A!~Me#4jzaC1c#9@k}e7vNHoxYMr9NzI=fMq
zY&7UwyMP7LX<(NmNs?ekC7>vj0tq8sCzvsS{5GDb3d5W;4pM45r#hi8WtyhzlnG5r
zu$Z~R6ct+_9W$8ddK6OpP@-TaT%06H0`rV`8!}OxBwR{QXu~{V#%4(pMX^)>XLRNQ
z38Nqs)QlBsXOwmis6s$fOap@@rNj!MD$nzJoq(n)nx%Li{Fkbpr&yYzYC!7sJm)GU
z3YOAJD`%`|0B){|v}{|^RgS|T3JDNOb&}dEC=9U0YiW{VoFugxx42i$DTXj~N(mpw
zsg2OVAIF6-o*@%usE3B=26P*`7hR4=g3-9NKY_j;O;EzC5g)Yl?FkYaMh0UCaifu+
zbY_SaJm}gH*hcyvrqM7z>9k_qAf6cCF2v|5q%Ld<U~*eRgjB^^g0{1EV0i#9HfkKS
zgK9APDY53#o`P>z^>b)K+IjsGR0C+`dA|AVt*xz|zPPw3L~mLbLhMebyW;fa%a`+u
z)9G|qh~4S_ON*k|ws)FmVC~eYQ~9>s$-lDI?7jBNl`ESCtmS#WBe(NYmoH!5%GWkG
zH+y@Vo10gDv)Ahhu{e3k&d!d#U7Ue?@;u+VoNsMyUDkJXeRujtLWtgUvM7RHesMaT
z_JmlRoXPWiTkhofj@-`w;qv9no6WCo=5KB`n>+S)^EF$|wav}V%WGR(TfGN&h1i{3
z+}_@{cbcaW!Uz9%@MBEjJE)s~7KXvq=yzjB;Uv?Yvb&%xWW*ujts!O#0S?nb-eD^n
z_Vkz;kC<8`i5KS2I-aGjD5g`d=g++P#N2#9Dg56w6G1&m=I75izNHSR;nwSoXP)@$
z>}O2vI5UQhV_CDR#!Ps+#Z}9)#-_R0D;a~SscmopL&veqSyfler(3FOnR0BIi;J_2
z+2qThx8hy&_gvtmJv2tgD&lX{GtdjsE72R!2hnF)bV5d8$b(-xsoIkk218f%^F`x!
zEzhEw9$e?#tvOIesN4j|srH$p6jozgS^cwuLuRR61;q_&DcC^-I7F@qC8&_!%O)mM
z6ax)Gx>eA@?ec#(eE4v9IQ)lm%AKEQU2Ib*>R|4^#Ior1ur@n8dvx~b{<F@XKfgnt
zf+(w^-czW;I18U0Je4cIaN@)X<rG}^Of`e9;XgQk{`|HQQcAZ@n0MT9$64iLQ52=&
zq1iW;hYC!Y!rXhJt8Z;In4?xBCmL0lqD(*t&PSush%;6riW9@<hB4hRAX8QuYhcsF
zj5Qcz%w_CUqtR%LMn^DL8E+s2C^`5e_)GW@dM1<7=6A7;F8(eyP3{qa&1SS>_*};!
zKP;>a{7_&pmz+9m4$$hGE^?kN)mV1R|HF>sTbc$q2>bvWLb_4{DT<=ORIaLk8+j2}
z23{mj;JnC4JIvG6_-IZA)AWi)9W-K%maYR1{2&0(G|P8g{3M9tnroV1NoirQ?I;Zc
zjD?lmToNZS)^$;(9kqhv#6>N`fE_n&)$14(g$qpu5+`wj70QLKV+eyZkQ_sJDGTE!
z$<hSscp!T>-h~`(#|(Mg>(G1A*H9mWQPVWRHHY0r0R^jp$7!&UmZ9Ljcc?6=ws5d(
z?Nx)S>=gN;ZL`o_6e0;oFO}yE25()-6oRDj*G1U)ivzgSb-Lx@lcUk-Xc#_c`cANd
z(2`7#lO!=GN3K*AhBVC^_PlA^a67f_BsAdH!f1WGzrh`cEx*TA7-Npj7*m{2oj-s6
z{B*jyx-?x`nZ71X)9%vJY?^j^OLOT*j~qD?$9De~8^AEVy3jcf<8<y@TdLl%9mZQy
zVQ%Eu0IrcFM;sr#3ZyN$nES!8C3v6ATiIe`Hjd-P)zu}v(P#{YjmB`$(1+B5g*Hg}
zgp-U`nRSuzt0RA0Lmdru`Uvnox6S6-VLkgnWy*)U4`T!fN~_6eShg*_>u_4|g8>c>
zP8yFY1dYZ@d5DrdcKtk8;`uuv3>vBLQ>=&|FTQ7*U|N2yHWSAf;;3Cj5iLBg08KY7
zA;57c>^+q{ZPv2F&mYTO(l<2ql~EH*YqK5SG)&VAOanR)#WU?F1_*=Z7&|sr6sma8
zZ%b6}Q0!80SxxOVn2n(PJX%Ayp-0g3^XScG#}(0JEF*4L-EIdCopWfk*)78-CCnyN
zDW4t1-z6}70-!z`*hSmaTA)j#QPp<Q$YA0n@-CTow`IEuEU4AyJ~mgY0iK(~P^-=1
zZVFj8H=EskGM<~mA31*fc!rc})$yk1*6>~Z8LqkBC4f<N#H(eVH}ONV<Hs+}x||({
z3x!9T46|%lgrpCN@0!c*o@AJ1!wargtGV9pdd+iPuXar@WEBp@UvQOfb__Pab+el5
zz4WD?TgzM*UW#YP#O|VH{UOC?(IcLUH-_g+Ijt@dzx>&w@}57lTmld-j8NOKvnS#c
z=sb*q{y_!s7)4&q%Q(iAt#7$;$H^xP*L9B^UOO_YtD4asjYgxw&{SQS<kZD|w9`2$
zuUXbLF6D1Y`CI&+-8H1VD&^Hb%Z1QIQ|P*yX_~IpR9zSMJU(d8_#2&|S@eRayqGE{
zL}B{S@nj|1UxNB6kr~jowym6W*hylPap;1jKb=hD24j<y7h$5ECW<On!aMih0#Dw5
ziz^I6xbWob<Np$LK@F7yjUxK~I_Jd@Rmu#*an!V@rbPjHVG7Ed+Lqt~W9$c(-_&l<
z!g}Vo230kDsk$y)X9hlG2=_()73%4ZqNv<Rq4t*`?J)2-i*%i<^#_|>f;(K-3#T@-
znDlg=+Kkms&&+ld76uEp?lOg%Q0HTe5I*>ygFW~U*h4d@htOWojQP80^VG_WANnhU
z2s3MezZtFFaOj0c4jV?t_YBjL^=$f3ISZzWvE%d&<DWM#F7gYbwHpq<;K*Uinl_i!
zClAe*VB~-?>08$JzjQN(q#Y2$FNLS0yU>#n>U3=r70x;X-y@DW@paw)RvAslV7}`o
zB_SS9#_k7Y-8a+v$(Sd`MmHb~lL@+gjGyM$n_SiHIGN<0YYJi5(h$1kdd*23+q%k|
zc_;u3v(fQ9OU7}xlcuKUdb4w`=b35J?Z&aqF>AFOCI)-|mjIw?%Tko7n7S_5viSlJ
zLz48D=H>{tq!awv9+<AQFqxZM>L&@{M2{jrP)JsrDLo&+aXW3>$79<a*8%Xo(st$y
zQ8%+D+7A4nsuA`w@B=`|I{5E{AHoOVN9f(?Q|P<UgbLmPABV5N6YyK`w-~UG+dMNS
z!?Ig~<0CffbTlr<rDO}LGex^(w~Kit{@w@>v<QPG^2@hd(`rjzg$S{clX|2dFX|7*
z`56t0d}Zb`C>tL6i4@})!3ChNZSWnK`6?q$+L^3IHf8k)nT#iX`yRR9k0a}2*+D;G
zJe<4%%T2YS2faP$5AID?D#j&@fpeAF&|x<@OI(9l`9%}JmY(Lqm5dUjZeLi{6RhnC
zKI(^Ug{=N+9F5B`azj00T=vT_>JR#p`(dr#Mol*2vsd;CQLC(@4Fgp>j(GPhNXLCF
zCgvW|Ln>zSzOZ;ikOhMZVWjLP5VslYml(`hjH>9Rg6AMB$T|x3saoK3cqXOPv^+4V
zOBJePJ+C0+6@zh30b4={6=Q>Dp>iT_sT>ooS`2_$1Qda>;<dLhYw#IEp^B<K6Tk-U
z4{CsCS~SwNj3^kG?`dJ47tm_tf$sSXutG9Ti$fAczR%m`q3NC#O`XJqMNv0NOq4$e
zNF<T?f|`gx0%*k<0Fw(sK4FZsVnPAM1WJ@}YnbsUV4UZXdL<B|ww4KXT-T<4$&K@_
zkz`V2DblJJCy*xfhiF8Mj5E2l<P0!*V<bns2CalT1eFG&6gdeb!J!7om^4HTK;*On
z@^!{wY@|8A@sdiKI?_;iXLJvjWz!&Zo5Oc0_9HOHKjF)w@HS*w5-D>10x3@Fy3C2<
zw62S1n`~b&>$1pl$g}z;q*<?*Wm($mry0DutSaYN$7zxVgAg^|2!#FOH$-HT=MIAQ
zsCCV5q_J3*LcEI6Xf<!EwP12!qsX}^x{2a%K>_esL*Pv+Y$Yz~uDZ2G#{pcPPb8j9
zrfE>>Aj=>RBY;uJA<GVw3h8t*Nkr0N<C45Mddjm@bwlz_SzXsn1P200L2aW~&x0&q
zRPRr!Kr8*?>dTo8#F6F@5saYJKv$|2DCN(hA~(IGd>S=%{6_EME?VoI_fbL8mP%Jp
zOc4*lnMh&*{37sLc@(km%2_RdtPL+=W0?>nuA69@AN3m6<NY(}Qdy>%Im&bW1L@vg
ziT!@BhwmV)yH%cv7|GMJv>y;PbzQTB(CpFr{I`WjoV4{As}D8PfEPtlg0w@Z1EFeC
zMhl5TRS|}ct7@7ShTaK#cs&U6+_v+Q?qduut1HL|b;ut&Yglz98-Z4we>VokdKzKS
zQ*;%@)A~khsFJQUrtFq*M=h;kphOa`#$hv;f^*9>gb^7u&EJT6V=z2^Cjo$3Hora^
zH=96biUQ;5^xV0#+sy(8;p|5=tqvfq<EE||qP#xf@2Em+WExhI_yII*MS)D(2#PAD
zx@H*Ng@uKB9pgN2r)la|)p)8M{1N;o)`C=`5Y^i&I;=1!Dac~aJ1z>4J>m-O$|FTl
zlvNoKk9bvk@^}(egHE>l{(Awg6h#qLQRQWx=k2gl@YLPi-B<DbKNCXi{PFW3+y67j
z&aS=h>eZ`%v`sWk1Gc^Y0!2|CdnwfS{{UM1f3Wbto1u{;$;%&kre#@gedXuEFx>wi
zkar#b=>Goz+y5V6FELHiecF?5zy0>NTJN~^ry)Za1qXZZS=d8adkZ!TO@T7NnX(@h
zCtY$3s*{!75<ZJp58ZLc>LFZ~$H2T`;-h7`|IfH6ih_`$C<^@VcN|*9`;;CZje-Xf
zKsh!TjgI$hDeYcQN-2Bsc9ms#7w$tU3efM^x2DbW25%@y;t|Q*p`6FLm3m<^jbx;K
zo`#@d;yG%yiKv#=mL2p7Y3PwQ>|U~lQ(XCzZ93C}leSr~-r_!(J}v4b53yI_xC4}i
z?O^SPprvRA<5_pcXEjE-#wcH=(g}Qa%|BJ?*(p<NMALv8lv)Jld&;)#oI;2#)5PhM
zALsj?W9k@F;@73Eqz34gz^JO`j_*6Ru7Twe5=6bF@o-^ww(w08gRXn6!JuJctZA+r
zI?`sk$8`=0fF^Xy)Fq|1lhvb)k~}YpXvWi3V5Tr(MPr(Yu>_!o!VE<bswSj<pr|UN
zTxCWmCF4PO&A)>0nH~=W1;Q=LwL|luDMBL{R5zv`J2F%=1!y+vwyJ5Ws$fkwENO7S
zI<qJfnjbXWB&l`c*lz(Y-7xfh!vs~8X`lf%ZUJXW-bw9oR*yR~X$W8jGmZz~SR4cA
z7?{~r*QQ+dC*;Y&H*gJp78z%1=?EQ1x7z~@yXL^onkaYRs#!<QX656y+(E%by62rc
z&~yRGx4FSe1iucQ4q1MEoo)yApSb>NL8sg4z{UFd##RWrK3f!CvkkAE;2BplZ9qxS
zYD9q3{ZD&#T9!QkW}nztpP^I;c-BJX`_aNe7zE+MC+>Jvd_P=x_~wNLsLx=mB}qr}
z?TKe`Thv}Zd+z$eTdb;u-=?Ay1fXe&&syzc7%@Z-{s>+GA437^p`+*wx`19;Pfk27
zm!iA^`yQ@s6DsIjH6TPJ#WnQ34bp)m5Ksfx%CfpHV1E}RkBh+<$BehYE{pqTqyd~z
zjA!_uVwzBZ@5hl|+zTt!?SL`9t*YSg4L59T^x=WI`8)5LpF6d^ed{kbgE+wu_(^Ow
z)ufEv>Qqf5r$N(&N5FCHdXf@d7#@Ia%eYp<Z>G^h56<WsRrnho{{7$o{gd5J0RT{R
zJ15J-@9i1meed^fA50DpCkMAb{J}JF@9}HD_G>SeB-t&}l*qjQGtWGuwZ7Q*4&e2r
zb8NB<qH${Ne(gP*B!PI8TC>;bP-oED$0K;Uv){S5^NyX)(Ke&4?P<F>tMyXy7G{|I
zY?V9G%qi{CY*(K1O_Z&T`9d73#xo0yOjc((Xbbq+LhEt>A;cbnPm?GX^|(C;dY>29
z7Uke%y&8@GpSpABeP*ZY#q8<T`tbiASbjje=zFi6QpV89$*o&;FR=FO9{l;Sb3oo0
z2sGNWhPBpp<+W|PJddBs6v4lDi$%8^eSPeDA@Ts)rD>p5=<%Cw(@j!<`|w&FeYW#a
zonN)2O)DMT;|5YpEFI78Zqb$&9^9^~3gGX6=M9BXDq}+`*sIe^HwB2^h=OGhTA-H}
z6$`PcSSfLG$#eK{O6!$T+A!f@O34oKEXMr4^Bg|0pOq&r>h-uF#q8LTPAe-LAq9#Q
z1C+#iAB@qAR`?%`MeHnIFsZJY7!j01Yx+Tg)1n^$?6Y7J^?CIw)y<;FiS~hGs}Kxm
z?Su#^M6r#cpxCn};@JB}Yk{b)r7`_p?6l-r8HuqKyt70}q*a)J@!HE;hO{>fRESOk
zpiQ0?w%_kT6s0K=rwBkvN)}ry6Jb>3$ziM%kls&2WHjhJPJ+c5J!SzmdRZN~!d@#p
zs7jq4WmTOJ5yn~p!C32vP!&-KCX~hgu&FtdU@gBiF^Leo?YZ_dR<2+8efR;uLmxpp
zI)*OC^Cz5UnTYWq^44g7mF5)GT!w=ICLZ<+N(?Q|{s^nS9O}B5hHIVO(I0iWn$~Km
zs&-z}Y&2>sOo^h_yzguJV!u;{p`l3WxpwRaGB(0+TdBQX6uz$OwdG@Z({-c5^Q>W#
zsNAV%D|S6A7wdI>C969MrSz8NnZ@!6{x`Fk<rAiD=NrAnD1v%@ZgD!oG|Nh>+3a?k
zeN~~pWl0$mGP(Y`DoJAFv8zW8TP6nIZ$3O6Iqs<2!61y;o~C2m>5ho5Ya9?lDnbW8
z#0>V(-ROSwWb_>LWAv-&Pte~Z)EQ5zN@g%nTZ_tW#nZZAY1wFiQ%vJF%9a-Fvx`Wd
zU^|vl{y-<QEt^_&2Cz_eCuzdfwvPbxx#0R@zgjuV(6Gn>%<C=p2W%qAfw2L{>&vfQ
z`W26^X~-G(RF-YD^XTy~>bA)UKozSHhWRazjfNBWNl#OWW?8LNCj^8McUHOW_=)ZF
zM7Xxl;c!cmINDHz08L6&RduuMTMCQgR-RC*amNW;t)HmZ!7$9lQw3dAh?Il(NJ|$@
zTbjRV+d?!5t-V`Sn^J<~)R#(2V7F|y<7QmPZeU|fq?H4BKIe_W4SiwSQWh0jwdrBs
z2RJ;x5aHTvFWYEfVVEb*v&1h49QuGkQ4Eb`&7vKK42ORJlv308-C^MpElo>@$y&{i
zL2rx$1&84**6eIlh50O_Mm@_~rVBg^T#0c|rRT!WgD9Gdd)P9$Ow!sT)0wm>G@VX$
z;k&k?M4wZ+>-o0*4aWmbCBPZ(E=k+y)$69Scx(hDv#I_aKR$gDwAEg(=3ynVolMn0
z7_l!X0UQe3wq_PrGsqz<JrBS%EejLZ127HKz~UpT(gg!-r-u+i6!C*U!mIEhq@y)-
zC%PA*P8WN-i9#DInii`n54bisH67yCgSkIg997+F{xIO#sU-&_1Ty3gTuixBFh5vW
zfp7AgnzLoR|II{7gBdqnmL|s-q)g!E`}gR#FI~EHsqj3*@DeG*u+&#pR)#}>@qZR@
z6U-QKnualF+8^LJOw+ipWJ>?v(|_k5Y`f-Lk^r$~lDqDe(@=H+5o8_w0el5MT>D=3
z2BgR9$0QCy1Dd0<RxnHySZ0c)vJ;ijj+;kcHTc{UPdvf-6O1jZ{8gO4imS_vEid20
zX<GXaP0jntue*5hBBd88jR?8!w+lv;zB-EjwIjJi$R$FbzO%D~@eU#Xe=-PB6d(kD
z3%jU5OXvvN9Bq+&<TZ<`M(2hqC1QnYjf7BFr(KO2uxL!7Ru0(+;0h4I9&e~4T*H6`
z*nRukU3KA!-2L0vJ-4>hXutxVpI_fwSXgi>ctg&O#~5}^!(CtB|BVI1bYb2#%>BLf
z%yn;{pPz?D-aEE14{zTXjq)Ze;PLoeo>DLMhp-DDMHzZIdQ<(Xz*sRXSizjC_VZ73
zKR@scGRS~0R7mN_i#STyOrlSBX+s-ix;i2kjd4VT5zUK*XG9?)LLiXmzBoUg=6R|{
z!80>WQ{W1Q)4FXpEKCHZ+?#AoLh~#Wbgcnk#<4<l;c4ww8#G-7U<5~@Y#9a(pksty
zFw;~vF!iEX<rT3$&}wNMVHx=1ETKYL3I)Y=RXAc30{AJQXnNiB_J1zZHC`B%E?7U+
zOvX&h(>+%OU9Vel%n4O;(`-_j)H4des9|*MkPDqM=4DH1d~0nyZna<<MU;Z3Mn7Mh
zk1=3{kRjP$l>GtxH+%@Kqeszu>fVG+M6MS&+7z#op+z%HY^I_)V%IH81slPDMWPHu
zt5+nTa5|~o(lj6wf~5b}M`PLjz~=z_7l4uv?ctzlBBXI39zIf)7fnX_K6<L<&2AK8
zkd_--7Fbpom=;JY3@r(k89Y$kwItvubgTLK!DKphUg0)#PdJXyyvxajlP6D}n3sh*
z_&E>ZIKs=Du40-&%bha}VJ$9N!Z2F4qf=rSB5K-BK0iO_IC&%tgHYYETTdc>pw>BO
z7<^J!g8;+(jU+XKKw1En6$F+AU|GTSJ{bf?n#eNObhTc0IU~`b)#=pJ386&uR;Pzn
zL&CUQuj`ry^)Z;}nyp0%;C6f2G<rhE*!5xsfXAMTu`YUsx!i8!KK6YR5&(=7Fnu2*
zd9Z)*TX?tbOLjAZ2IwC2ZuDdH`(VIz@EZ6i{1W^h?%;mZ6FSUo{L-Y~u5yq}!^j3=
zzwcN6^t=5koP=>8Y2T0)fnPoxUQ&&L47%}*g*4F^lP_^O9<XmoC?wqGZB)d<>LZwm
zyPT6+g@OSY#kgPfgQyAzeRfJ>-VeO8QR>4_`UFQ|Ani)!q&{41ChfFcOv|=vOE~Jf
z$(xn|?oP*+BbK}EX5Ox+lS*Y+*J?Yuz*1tH4mh2VQF&y&TQ&<f3DSYcf*XcmLY<N@
zj&9%XZatks&dD^(cDdQ*L6I=zy3=-2)q()phs<Y~pRci=guMA6(pJCRR9s^a2ZL=2
zw~;{Ft(0JbF09`1gZ{!rTQgTV+z~>$n&x+quN%9m2OBLBDd^hhFWan+mj&AlZVLSB
zjCPAB+Rh<e?^LbYnfzD3-zO4}pP=`Ogjb{DZDaUXs_KKfg%$iWwu=Xn>EYG-?ELhs
zxMw;=TBgqL?_Iq$J6P=Z`^n<`d^m#t`%e<t)RcxC1<_+4$Jw;Z09drH%u2gOwVx{H
z`iON^$N5yixF`^TqU@@7u^e2ZgqRsidQ4?n(UheWJP{lL=))=z%ouiob^3Auh@=2!
zjbKwpKN1M2cmM(b3E79KGQwhd))^Bfswg;10l?ab?aK)<g00p`O+;7wmAaxt)cb0K
zc8E;$s`ev-AWDFgJBUPu9f`TTu8hDIw2B%;aL&H}pY6j$*cGBaxu4cbLlOf~l2~cE
zVLBm#E@}ov8Bq7Z<Kr;MvvQFnS=QR4!`15G<aj>l11u(xru`yBhlk63Cv6F^KxNk;
zD*a|+L)$N500SB!J2NE)-Da<~R&?7MKF=?mXk(xPcoo31rsXmc3$f&yhqGBu6<LNw
zHjuq$51BzjS%J4Z{8~b0_^aNK@Aire#N@SeU?6t6?g^2jKt*#`sEraXZ48D)fD*~T
z7@#DREK2{}HGrWQCXMT4D<a>frzrXt@O%ziI&q?cIiZ(jPArr#7n<P%WR+vBbSlFU
z{ZFU9X!JV22r%dwAiriX%|O=;VyV~~L7Q5Nq!7XM`t4>n(^%PuLA}v7H*3K|2H?7z
zUw`ho=LSs^M^W4~gO6#v_rCeKvV3mPpaN_+gKw_k?+L!OEMI$S?>DS9oA*zWPnbY4
zET4RS6gS;|zuUwSl*shMQkLWWl`+rg14k=n^MQsruj?G@X7c5eazh@>%hR8gFR_it
zYT!LhD=`tbF#sFRcYcxpJs@o(ladA~hg~#RfgWvpEkh7&$(iPxkxdXSIx!e~#!`+-
z{$q2*Dv8!OOuEPOATq3U6hTqU0mS=3Q<QE_V2sI>_J$NCk~wE$1xBK}u|A5rq?=}c
zV1!JPLI$W@RZf+pN~My+(I^lWb(7^(hG%mhl?U>bembN^Q8v&=d;4cd1V~O31LSdt
z6!YoP?4V~1z$l17h)g)Avx0auy!wz~PK=JSIPUhlA+k0}e9l}5B1WXDG$Ylm>(C@g
zLa?rzN<U0ZBt~$*b##aFbb8k-o!51q?!0^F<Fi*N_#aU=Br<HtUdV#yC@3fnN$1GW
z?Ax_+$QPm+Tf42nj(6bcWWKryE<wpbk)(=6n(O0c^VuB!-HrEd+awuO`Agns!vJYo
z_2XEspIMQ#D-58rdu~1Ss#~{`=lA>lP}Ft1xOM+#8xw%{r}JIebR$y~dCtDuEx&=9
zny7B*{fKtm&kHqu2>75V&-#NjQ`$SAy1F>%_W0lLzqZ>SR29TY33VJ}S@!cy9L135
z!@7It>=fYYm6v~iJm^nvuUC6}07pmnWX0iV1Z|or)j+Lv@>BhOzu)io`*$M3D2pP9
zqBIbNqN)(^kmF`32Q%ZohXp*2rf3b_i=KjBUh}?0KgZIxB<4XT6^)~o&8TXZ`AJp8
ztL>;->61;%vZ^ZAV&u|<KsOQZq@4&}5pXJEE$W0(^a{<kSNM;vbS*(EC0ao(X7s>_
zvSn%x|Be+!Nh64l>iUD%B8;PJ59%5$CW8AJK0MAg3_ted(P#wNr#^<E!k9@+*6DD=
z<cvLe62LQX<h?QKp7SF746fZac*S#-(sNxX6e5FAf!uYysy>Lr|Bmp3s@gx2q~+WA
zY^oP8Uj6va&JH0<KF0nM!FLNGj2mwhAcXk%+8NxqF2wtPI$Su1qvMSv8d|NJbF3@3
z-IhP8-X44iT_=e}4~FmuVfi&>+zCZNWio~U%(3{1Q*w)-Ohs#jZ$19_<C-wE*+F1g
zY^7$JP80;%vIj!9EZrp3k(APvSE=f>&(@Dr_17MM{Bc#C8#2oZR(KGEu4QFQ&(|GC
z=(^=N>I=AOiNRIP5ZdFa`XkK{+Iqs6v!BRD5ju|UME9WkYfFrR>_}D%!4P_HtU>k;
zVK5jBM|XKLoR<BnJBj}vC)nqj48vK4T^Y*HOFkJY|GPp>&oim=v-|7sBvqaGK?4jk
zHD{U#J&&!SJ%bs^9VhIQ7h7g+Sa^|?5L8u__aDpiyx;gO&Kb-DXZ#wR!Kw!+w=~@Z
zaBR~6P4g|G>)gPwR3U&!i0;%g&DSi0Fv2Ij=7N~yv&O+S_&4|#{!zW)zj5)E=ndXU
z?Z^X%ai7f7e<F%nh=NK2v9Z!WG9X!Q%N9L)#<i3K6A^#NDz?fMx2-M|>VO<}G#V1L
zi*O@b?Z~C;|H&l$jyo8h#V3~2G)<P=nVI&Cvy`N1x_lDupG;u&(4%)uCSKwnUAy?K
zjSYZ}jb~k4JL<jR7{G}IZ_bt&$T<v#F@wP{X5@l9?|BQ}!p$>=0k~}##6AIVjGN}j
zn;(XT5r&2lJp0COmsBSqO_!I_6xzBlL;-2Kw4A1JvZ9ApAA0oAp>BEet-T(=#>TUr
zwXp%v>)rY`cKr#cX-S+W#w8vELBLoL1R;A{rncK^4F)U-f`GFy2m<z2IB^}%yQNt(
zjW9HfFf<TC45<fygs1Q!l%X*?fi9q@qL(2AjJFw-#IEKBv0jB;Np>5NNzmXX8W57A
zMoIr-W^w2{sBcj5YS39*so{8j*==V+LV_FULG#qH^>i`cZ~fS*+xgpH``XuTU0rtF
zLU?Q!K^VC~rJeUSo6T@WsGYPa-}}T9PuL2Z;1joLMd96}hxUIeNfIkDgsu>-8bJ^Q
zMqmj+G3O>=_z)DDG;dG2|6YtSw0*BdC=P>W-S^oc@{R3wd%N@IwLF!7j3u6@p8NTk
zRYh@Z&hPDmrWz(8fEeQFJYIzY#b^;7LTAx=^c0yKI?^OCOz6KD`m%<?Hrlo}AD`}&
z{i-Xh(4UB9+u`L<5JvmVl0g}efyhtEI4d1p*B4o>m(nG(H#h$`{eHiXqbLe--H}d$
zf%pNzK|uaN5EeybxTI;6Vrh{<RPQt`JX_bp#B~WY_xASAoY~sid-k*UfGWP<U-AQm
zzV(3z9so@%TAB_g-5^mE*u&Vgil$*ebF^+W1Xq$IaRFn`gIKN_N_%xhqD3_5K@EDa
zX*u=G%e9h5D>za!6sh)@hWtkpluJuX%PWqHAqW=AA->GnMx)V?nhrk#VeG!{?Afzt
zS8Hj#J_Pz6TuZ}d7{b?Dhu!^m{!&EP%}aOjU;Ywq=H+}N2gHlR1y7<^BLwn3wjE*6
zK=qR0l0c0LejjyUs-xg(l;Onq5l^77bXJJj3;3w##wtdB?a28(TYiKN{fWX4!pJXS
zk6>=qOerPf*Epk82%!_ItAyyBV=_s(l+vtO9FrUEFmmWKH8i@Nx#Oi9EX>X}7TWDh
z*V=P)!CaQDFE3$A;vc)&<WRkCLJbs{c^+5_|JGoI6D-VEDZY5-TdMlcnPzu%Z{7Nu
zrTO3$WjSM5%|<R3yInX|v<NWPDa+Q(>gp=S0G%vT^8FTHga14ax`$W%fgWX?m2J8f
z?9i_P2Ha5Qg~I`0H8<G$aM+5=Y8dnG(x6(2eEfesLC6o*lJQU!)5(o*2GeZh6FSKo
zl0Zxab7k73KmoJ|OZYHR^)RdJ^?IPI1QVc5WozILvfp}&fv4l`-u{2g#c=}e%Hw!W
z+F|IK02IgdQ&ltlAei%fg)znV=7PXCHK0(J)vK}PhM_HMTmMU^k0As;LU@W0c?CkW
zi4grOgp~UbQvMDhb{|6QpAh1oLr7gmNc}g2w2vXAQ-lOXNPG?<!$-(?9YV%GA!L3J
zA?ueAl4lUIDMI#R2ss8q&f5@jRfOE{A>^$i<n1BkA3?~+Lod;$J8iC{?Y6XSpk0A>
zf6~654zJO%K*w9?v@M;kq;rAJ^>itx>ms^+NB3rW)Y7wpo{jW=i$06!*PnhZ^j}pT
z`u{c7!=^E;l|q5xl?;E55qC4Pl96XJs)^A}j9t&TW+rrILNgQFGkG3UJ5yQB&YQAx
zkzI~vmlk#{vD=sIxr)7N*r&igE$q9D{pvXIN)9@N>PC)g&(T#J)4;L&aO^EidyeT(
zG2=C6&SK`T9KR1IH}KDTPFcn&MgBF4Q>&R*!Td?gFR|bu7B;i!ZWdRuq`<OCEGx0R
zj+Hg6dWhA3vZln^3#l8&X_s=^ubf`X=`XRa$ogB@Fp7=XxQsK(IkS<omT`7DXV-Gh
zwwzPs+`Boig7ccWxSmTl<+8zC-kr;T<%%1)@=UHSaP>o6^BncFxON=ZeaH>BaN}TZ
zyp)@IbMsT&+MnCTar>s+-b_OschqzDD(?A^`@W>Hg$F8lu$G7J=HVib975A~Ja#FM
z*U?<dzc1yf-aLb68+mRp&u_`|U-IH$Ub=;syYosFuTJ9C?|8k2cebUaz<aCsa7#XF
z<>L~c7Wk}^FJ9xzS$uN~#RA15|9OdTzvO?j`2S$OJD%^4=ZClWaUXvClh$f}|B^o%
z`D-43m&B%$eIzeQS-ouXlx$WZn|&%<6lCj4*`~d0ceHHZS$3Et<>m5^ePqW5+3{1U
zm?jks(&i~?S1#>)ONR>SP$wNr(z(5Keond^FI^9iZd*#XdD6YVbZ?cO<<j#*>0OXM
zHPZK7>Gz!sd`pHj%h2^w7%aoLl~J3@XfnD+#`KmkB^h_Tj9(<<OET#~ncOH-FO|yg
zWanzxxm9-UEqheR9<8$HVA-oE`&P)l^|Id$vVXlC)Fg*=mP5zM;k8n|j~vlijws2|
zpUN>$$#GRO?M#{ekj$)+S@klzRpvY-b6e#2L*&FNsX0JuTIA$ua_V52cZ1A-OBSq`
zh0U^Pnk;IP#h=R3-m>gDS>9h(Rm-ZPtge(b?PX1)tbIy0Y%3c|a>m_qX0@DkshoYG
zoZTwtERu8YmP@{rE9S|S-R0_C<?52uZz|WWm+LN+TT60#tu$209f!zWm2x+^dy(9G
zh%|00_gBdM>*c{pdFWhus3;HD$|JkVBd<wQl|0rYk57~4CV8q{p01N;$+Pq1xkd8A
z@$$-Gd9}a1MqV$->wn6d^W?1-dAmX09VhRW<h?WH{io!EQSw2veApr%wa6#r^G)UR
zU*)Sh`FfIkeWeuV$+zWFYLNdn%6FaRyHDkZ8fk4WKTnch8s)c9^4ksa+lTVUh4NR8
z{C%OavMObp)G6D%sBDYlm2K6kZ2KzmVC;O1{f!CVG1+%aUKdjy$5i2%`Z}iRgK0Bj
zhR2xYGG>j(tm`pbL(E<ivrosIEiiXh%oC1zn_&LUSYSUEoR5X>W0AaAv@RA4$KoZi
zL>nx*982xSGMlm7bSys~D{RI}C9z6LtU3Uz&ByAWu@+b-FV@+N_10s<7T7o%n*?LK
z^Vp##b}Wir(qi`(*s~4x?ScL3V*haLe;EfX#)0>7a5N4ni^GC(*lrw=5l2kMk$G|C
ze;kt+$Cky36L8jNTv!kn6~)E3aa}`Pw;Z=s#BG~#drjO~5O=-C-KTMH3*5IJ50u1%
zJ@C+NJdzZTw!ovG@pwTz8;|F}bLa7VUA(v-FWtr~W$|iVyuKN4e8=0X@y>R<I|A=j
z#rxOs;bnZD7GFf;%jfvI3%<FIZztos`S_tIew>Y;_T%UI_%$BCl?0LpVhI2MscB09
zfB@-`hu^5%hXPLX|4_&?P8QdAINhOjdPI8a&shOJcIX`M`FZF(xA}4C3OBi@S6|~|
zA!_X`Fi^j4|60W?I23RxREI)#)t}+D=?<-vsr1k(25SA#Ilc$>(0O#Qap($f)%3bt
z<x;42ZS6M>jI%2<+HB)Gb00pMVs9C@ySDc7a9c+5_i<lM>ndBzc-+^W?$n6j+46dF
zyFb~hG?S4Y>fW0z+Z?r3QF?iuzL6(B^@})p2m0X|Mwi|<ZWcwkt$q9&pYrR}u0EGW
z4ovdO$;vt}jcxXpaVu47YjMU)Txa5~T`nrGUrAoDlwB+}o(&ERo|IWu7)ET;7J^iq
z;R-AICJY68EQy(KEBX673nOIi_$%g5j|Hvf`jJ%_Gp3KGLzgK-`gG|rM0owooA;kd
zXRlSJ8KUX&hOIvh7t$kIsy{&&)IZ}lF69mKym!;5*h)VC@Q2sGe2K@?;Fu8+k^kBp
zzn<#O>U70t%2&xAU`TlN{OJmdd)L~067x#2P<3&qHG(TwGAHlqU$_L1#j<2Ug-3s$
Nb?&+EAr}Jx001#^dC~v?

literal 0
HcmV?d00001

diff --git a/_static/vendor/fontawesome/webfonts/fa-regular-400.ttf b/_static/vendor/fontawesome/webfonts/fa-regular-400.ttf
new file mode 100644
index 0000000000000000000000000000000000000000..549d68dc023ff6e31b8774d784c2cfcc231e7976
GIT binary patch
literal 67860
zcmeFa34C1Dc{hB{o#oEHFWNQINHek}+cVlmwy_zEZ43q?%;vC)1cKR>vcyTCQ5Kd2
z!Ye2tG$oA^2z3+ExFl~%7FuO%NK0BbO(`L1s%(^|U;2)O<&cDq-~a!dduK+rY;4-(
z>-YUS(sS>*+qvgF=Q+=Io^!5{LI|Ij5)P4h-PvmfFTMVfON5YpI6Lvi>n@u-?XeH<
z6T<nN5b3EauYK#A{_d&E#)Rl_3h_^#T~}Oo`3l!}?!>*114eh@g!43q3+bN#dUjoR
z)9!b6h@T_ByM%CzUwgwFFZ+{!z43V=+IHgHC$76}_oVo-xrnp}_h+uZ?7Azi*zAo9
z@#_HI>6*OZ#+%lD>(9R_#KD9R-l{-Z9WIyn%+4FzE)AcsDEtN<6XHk3Z`^M0559cx
zN#joWG>$z&Uq<%$fBN8)LKsuH{<3i=&n)?UN1geO`H~2#^SmyX+~t(}<g`HD<mvKF
zxO$4UUlh&aJ;+(O9JL%}q?Q!or1Q@{L(Gb<g9mjPP_}I<cWzFU#9RicDeTMG&jQNA
zk7FsP4*m+q5gel^NS6=(67XxB{|~k2`9_$+G4G$kJ!KJL^{q7bzvMe=_s{ZL!*N-q
zGwS_3&+9yA@5={YQTOv2@;B5m^HlJPei!ioD2pVYv(h?`DeG7qvM(R3qE&vW??;-~
z%ho+p2VX%Nd7;hhb3$LMm1Wi{BlhZ9T*tKto_WQX(%0<swSBpMUe%A!;vBEr$7(;N
z@~Y))2;iOPX6t28&sg`>%5=!Fe(%A9wfCdGDqU9Z%`}#?mdC8f)b1DRzIIh7bRM2x
zQqEa*y`h|{d;#Wb<&iPUI?ep;w4FcmRdvEW>bj^sXFbpMww~ubvkDYl0w<(VtE^W0
zi9?@1+A-Re_fH+|o@2ZQTCwYdEWx9M1JI+>o?#rZU-w||1A7=2P7gY&d<G{?@r>va
zy`o>N6&Hxh#2dsL#pU7(@g{Ml*d?wKSBq=Jwc<K)y|_V4iZ_ej5jTpP#LeOs@fNXL
zyj8qS{IU3xY+v@JY_)f$_iuZD-TO*T<f6Iu+}hj)xeId_<u1-$lDjl_MeY;1Pv)M=
zyYk+=KOe{!^PBTq^JnMJ$)B4)KYw}t1O2W2ZT&s{<3GIpht-Os;;OhSW+hQcRnnD~
zN^7OP(p~APtgdXVY_4pnY^}V$a#7{t%9WL?D%V#gD>qhduH0Igs@zezvvODEy_NS>
z9;iH6`C#S4m6^)NDt}b@dgbZLT;->g7b^c&d8u-s@{7vLmFj$8K02S6PtLc@x6QZD
zXXpFoSI!U3Z=T;ef6Dx6^JmWQm_K*^qWLT4@0h=LetQ1?`3L7eHUG`|Z_Ph9|K0iT
z&Hs4*XY&W<e=+~d`Cl#g7EWC_ZQ+cC@r4}=7cN}9aMQx>g|{udec{~;cQ4$t@cxB+
z7w%v9z{0}|pIi8|h0ia1Vd0AlUs`x-;m;Rl7rwIa)rD^_d}rbN3qOC!c&YED3t#&9
zOMkaNvcGr#y8S2aKWYET`?u^rW&cI{Z`=Q={m<`zVR2xwxH!6a(&BlGS1!J7@t(zp
z7eBuE*~KRpXBWS+`1HYpXyq=E6)QzSobcbMHJz{0n&%(Znr}a1Yd(tB{D#$<KU=0X
zo0e_OGv>$9nitMrK7YGv&HGeqe)$Nkc}lG{&p~Ux8Lj!&Wm@yIhqdOH(VF`jT65=d
zwB}nE-?8}M;-ib7ski3Axr6_F@OuaU@!&t8UH@PIb$qg8NQlU4z$z9d(xHnKNkrli
zU-;MI13V5>15&^6zVP3MzZHHa{EhI}!hgY2;imwfQK?Uc{}}1V!tV{=6@Ifivnza=
z+Fu-ojuhUi(&g}p-18S+5pE7QVV?}g!!hiAVMj>4D+IkNG#mO<=o6tC-Whr@G#$Df
z`*#Cw43)4Q1{6YTLaRfqNTt<XCQ@#u_zTVfegOC<!1vXe?;`aK;H$yU1V0jdTX1{O
z2>fH<_k{>d2CfZU6Sz8XRbW@(92`6S&+!TU^Iz;g!+(N*%)iFp=Wq2l`J?`b->>+u
zb%%WJ`2UP<W2GUCj1iIlkK8X`lrI=gc|iU^{>&IP*5PlpQ7{Vf3FQ2&T#(<FPvhzV
zIbnDVx4c}wL7oY%?E$&Nh#`m1%6nK8`6%FBfEVOmY=4CPj8#Nl{g8YR+pF={CV}_H
zpL~ye53X$nTq0FV$qVf3a--bDG_Jf!?QvZW$N`mZf`0w1_zKfFi&jy;&*>e?If1j#
zc&+0;32DWj?8atBrP^@R!W4fj7vA<Pwv=ZfNULkbmym*1Ez!2p2{6=?Ree@JSKFe+
zm;6v;{GJx8@z+P~^awu*^}qo+>k@7e6)_PP36T^jkrqwRid#ghXcO(CLv$jKF3}B+
z+yfb#6M5vi0^GS@42V^D#~LvxhQzQai7~Mr9D9Q}No*1)i!EX+bmh~<xHwCkBhD4)
ziSxzl#f9Qxaf!H8Oe~Sl*C{#tc5$ni67Lqbi}#7U#k9Cz{GNC~JSaXW9umJV{y=;L
zSa?J{Dn2eAgT(%X_#`ZsPl-PfpB8(?pNatM89rUtR&K$*%L0jg7SIbY04o9g02iPD
zSPMW;LfHlM2LWsd$Q=zS?B8Gk``jBXAT?cXK{NJOSip8A%@EK}p@C>Xn?o1T(2o68
z7Lfm3Z9ymY*I1Cj{#pxIkLxTT6<u!u>vV$!J=jlLz<QD&5psavu^^BAjTZD_f0G3(
zu)o;?;^r0$h@-byK-%AJ0XgSeEdVEl{-j|I_TY#LZpHpjEVvu!$NLCqf02Dj16o>S
zLDK}{pm#<CC`R=Dtp?DH=>4?@b^aBk0Zg-M1mZG>HY0$tL=N~Rfc8WVZA$><i5%LN
z0J;%5l!L%B<}T6zIuf~yHGsNA?h*~Ok8{iy0D2QSv^4>AFLJyOz;fo8FF>WALK;9k
z<y{(Bw!BvZ`h>{)HGtklKA-`VFY-kVq|f|j4WNCI->N~S&qf-6J}dI)Xi({Mkp_@g
z=TRntN`qGOq*s(V{{ao?_oBa51A36?Z_}XCpr?KS>(UQg6VS`W4^dVE`^^v0ZUlky
zRva2wuZk0Zvh!UPyo11JE9hec@Cs2uxd~zhd-O2^`%opVL5yIJzC=Ku7ZsF~fSxZZ
zXj=j}iKw7I5X30<C@%q=NK`<V1n?wL*{Fg2ys}w?Scm-<4PpZOts2-y71X%`dI2{R
z70?DjrTIPp`9uZ%nm`;@Kz{_bSLJ#Qg72KvAWp>oMh&8j{mmM{Cq?D$fLrnGChVs)
z2$t~<4d9}pa;FCJlFD5g$V)2k)c`&#Dxe#J!ubP81H>uVKd6EIrSd@y3MU^%8bEJG
zWk!QI75k5A0LK=UM*)9?drrgt>l(zF*nb1?G|s;c`#BA&-G7QSK-KF7qyeh@KSTQ8
zklum)OBz)E2apDkS5|(Zfqbs=vIfWuQ2~7sAU(u9`UrtEIUm)a(g~yi#Nm8W1LTjG
z2ald7{p`fPO#}JQe7gqH$$VA=aX8<nf&FNHr3Ut$dC)FFfDY!-HU!e%Jf8tTo{4$j
znxN9a*F1o<G=ByFI3tbCqYdVP1Im+m)QtcMEaq7z<W2gYzZ`%)+j{<Xz#T|~w&w5E
zAU=-$v<65*F@GQ6e%x~n_GoK@!WH<`Jb?I||1#j4NP{-#zomhEef~KO<m>a_)u7Vf
zLmHrP`(va5<YV(_M*{i%Jlc&Q-i|%mj-b-NL>d4|F6Muw0a}Gv@M#dYVGlelfY!wC
zVh=nK#1!^tXb|tj9=IY<Ml1kl1o1BHL6-}_mpYF&SO7p*5DRYx+=TQ!*rPrKXb@rn
z<z7H}RT^bo06<p|3(OZl`d{Gl0O%WHfzJa}`u#`)poNG9;Gdw<ybnNKV&MZC=&x9K
zSOYX5vG7^I=Wx#h*#DUZXfk5q^BSPjh=ng`Q0Xrs4S=R27QUoGrJq6?0R2ZS{J92|
zo<$l!Ub=wxUHA&pAHx2t8lWwSg?)f;<2=jt9SxMt3*Xm3IkNC`4Pr)!mrx&q_$c=1
z3k2~o>~{h#MEVizL4O4CN7(<J2C+wo{Sgh!e}AtA)_wmv4V0(*!3PPHzxz+pK-s$g
zWDS(9`?qMIT;2~lAy6*wzeodR)Bf8ukU#9_GXS>f{^vD_C$N7(1Iw^DpaD9rSS)IQ
zwksA#HBjGMM0p5o$HntBh|geur3Ud?>_JP5Z^J$R8~b+vNGE99MYQW8a4f!#J^JJ#
z@$(G!d>+q<zrp@V4FYYlIIBSbe~V~0f<QYhKCMCgy^t?8E1+H^U)rvK{9L~DBL(c&
z@}-|D5Ws_cX}<!>J_+0s1Z0rhe})3`6S@CP1=K<1{>v4Re&qgp6|ha^{@+(XIU@Ie
zNP$lO0n(r;b^c+bSzq4sQ3XPrjs4>a*cNjC6AH-R<o?enU|*B_|6BoOhur^V1?1au
z|5p?U@NK#OYYNB@<^HcLAn%p?|5AZY|25KojWT=;`-%ecCApYTARu4lVp0M5l3YwF
zpgtfMn-oyLmW!xAfV@sFjwsM+)MxQ*oaecBDG-ox5^YTo;x6nTRiN;&_z9#z<E$6j
zoFE`i<>KcRP{zr{FDjruE*JkofvWo=@BmPCUqrnCs_u(_gEVMIJ^K$x{{zw=#{O9a
z?ALPfUlma1%Eg~4pv;wvzfwTHCl5da1PJjY_ELcm-^AWfAjB;8XeWRWDDMHM0wKPD
zy-R@*DEk4o0wJEl-m5^MugL>G1p>4!4+IqmSg`UyL;-Coc_6KTeMBAr{sBUK1AEXp
zKnS$$fmQ`V03Qe16bSKE?AsLx*xd3!Mu9-zln1&M2=TYr=M)I>Iqdrt2==1`0}9wr
z<$+ZS*dOJA)e6|}<bfdt?AP+Zr~>w1d0>+Q>JRe3W(7igOUPf$DNuCsiytEWLzEx%
z^NSxT(CHr|{bTF=PmunJmHsKxKSg>k_Ae;V>3>1`U##;lBK@M3M*IHaU#&FS_ZL62
z(*K6^zv=WZfpdVO!(WDw4q53Iq+5`_9s8^TA<+MRxk7;upuu0F`~dyz&yoH)(w70o
zA#Hyqe@8wkUxC-=dgEc^2aZ9<)s8PX4QI~znDgIV3D-u~<*vVUpXPp>`+2j?e1kb-
ze%sUSImh!Z&y${?d41kZ-rw<l#pm>0;`@Yu*uT?%kN*k(zXvjbt$|wte-rEoJ`{W<
zbbaW#@crR`kF1Q`8|{w%VayYIH2#J}TjI&&K=N(Lze%l5{V3g&{;Q_ZrcX3q+p@Z4
zf9n_8`r009d%k_JeYWGwj?ZP^mtE-D)bq36>0EdIwERq;r|+V^Cs+LLO83f%m0##j
z^k2|FJMcTJdRFaUeeLS6tr=VM{lfi4U-8!B!r<gkV(5Wk-|)Le21affEsV~UE-F1g
z_O7)}>vpc2Ssz$`?fU0Vc;AN28(!IX(}{g2zFdCxq=}QhziI2HzdCv2$q%0V^5*L|
ze}7Bg*0!y0JEiTE$G5H9_JvbVICbjOhfedI_WkX<Pj{a_dHRcIOq}uDnd4{9ysq%N
z4~?HN{)MwHIBVaI%Xd`I&Yk_EbKZCE=5wp({muE;?>u4W4_^PS3*L0$<VCA5n!ETr
zm+ZOpg^9;5yX6f#-mw3T)hkYa)9x!%S032)=vAAp`r6eGU9;_)m#_WFb?dI%cl~?N
zQZ9_J&dMn{3tJ=z-L;jrQ>kQ@^2KO+C_9udnx%9;9mOs?luZ{)U8AL9akSJ`Or=ve
z?Ka(JS235*<)eig@oRtKg%@tTahl2VRI>Ul84M+4CJ_osrmAyHwLFGj^|LD<du$&P
z@^tQMcTCtRIYTY!;G8HU<%c&(#B+JK8O6ItOHs!lOENN?&t<z!cQTPq4UHM8ltE|;
zdGQPBt-OmTt8?7SOm%1N0xkmu>K#9jv+`q*F=yc&Dbp+t6-&j@v`(ecZZlitIXmUf
z=Srm_&)KP5eze4AvcQE-jiybrWU6ydC5`Y65i^x^gic{0&T2oCyYr(w6~005wl!SY
ziJRT+?d><T-ylEYPHjIo>A;p1o95=+q+#+Tw+$y9ADqSy{95?Yro(8{i!F>Li``ka
zlpAfvKh^|)JcE|QpLL~J983=mB^7>0(8~DJ*Ym6w{&=P|%4gM;ktg|jm$Rp{%j*sV
zQ;FuLCS$lG=yW-pon51`rlx4pZ^n~ttqF(WK$TNQd-G86kXv2hXnT7{XH(L2COcNj
zJw!yoh$U8bHZ=tUrpFzKcs-tIDi|~!ZnwuwBz<1fbjU&8Sd_l#$kH2=UcX^DU2aD*
zWqMpiY7rPwe}{YrXsC&PJc(MFvIuOHq)XM_R5dEeB?oT~g`&~wEVr^;Jt)&nO=)Ru
z#ge1fMWdlmb(ULM4!4E_fpDwd9u$p7c=f?w9h{SQfSwDav~;pYXXzr^BcCrO(Wi3x
zks58~i@<t1U1WRAri^zdg1+dTcSe1|#Jh>lrN?r{NPYM$Uo;Us=bT_7>O1Sh#L=N=
zK=V>W4p!t|ISmi3?o~<9t{Zq5LJ8QvMnLbQr5sv*<Ym5ShB+a7J0*7?azb6Q`_r6c
zpH0HcE8?1C0K=#)sgRU3QgWL`9PwRz8{fzC!dSIx^X65n40}IQ=*&o&=`7fLDeTKr
zwZtTEn6mFs3D!~JTF6J>i57A23G{2Wd8rt+>pG-s7=0)Z4D5<}L-Ebytktev+>Xn;
zLMU#Th+I+q4C_abyH(!5L_4yM{m9$RoE1-C-)D_@M)M2B9q~@S3D|sKWqS|lq^EzQ
zc|u>_kOPnXJ@c(rB5s%e@7ttoY0aB*QnvMw@Us2V$oHMFVdcZ>)uFKeFWdU^tu0cO
z;n!9fU?%~~iYHs|WeLc&y75k%g4njiv?>R&t#}x5o&q_z-Tn{sjrHcF%=M04*wNG+
zlAc)7Z$!H;>`6CC*_7_NT;23Fpa2|IqAdT5_wt!YxW$?Ddt%NGb)$NK*Nv1a=YOD_
zQ}Qu*|4~k~gM;lbv<#)8RHTV)vGkNTo(N2%y9X8Z^$|yX#KFGX<ZX$7H(sp-`~lTY
zyLkE`-tZ6=5VX6Emy*Iuop0N1b9BrAXQfic72n4!zm&eyM#`ma>2$zl0x7Og`=vHg
z9`&21%*cD<lEt{o#z-jC>`4a9n5U(N5wFJ&Lr%*Z(#{`2-lBi>=JG{VZ_rTE?hX0T
zAx%cQyNr^6D2`-=q&}E&HQRmScTLmo2i(gQcSmB8z^S@7WZj*4yIXuj+I@=q^1vq}
zfk^BE;tMkNd3iEuIbqS!AYO$w0v_T@qLD%(=nL*jFcI@T|AMA~Kp;?^jCq3zd97+^
zSO6A1cEbz#Mr5pOM!rJMf~~5brb6^bNJ<krhFwuDV@RPx#RC5HcBD`&;vlO|k#zM9
z_CvBDv)Pfpk!)5PD_SEF9&O`MB+|OVxHS@qWKu~flc~%$I}z#Yvy-)i?BO$fcwJAl
zwXNGe>TYX|_N=>=&)d&%UpVxH-LBtNJVCW9xO-7?QZ(HN3r;Mg-C4U1x&{{GC>&$6
zN}-zbnA`Q&@^*PU+f}AaSJ9<pOfRTyxYP@RlD})ls^62nF^@l4-5ZcWIi3v28<T<i
zf6t$cdG@%&A#;xho)73q5it&VQGg6hgG$(*U5cl5B^%PJo61SW(5AVy)3U(cLD*|2
zR6j*Tn=W76oo4ySpHZ^Xa2};J9iEu%t$vTiljA5}b#Kz&P&_jfHusn;9?J*XD}aWF
z>AALKj_&W2IzY3MHIy^MiW`rPXo+J}=V&Q@xI$T$sdRUyJDrlUa~Kqc@`py6A?15}
zMl++mJ?hNF5@l20jmKQMRCn`AgS9rIzV28{ONWb($e!L&g`n9(6jAEBvj$_VPNVl!
zY6(d)Vc#LyXyP9&>2tO;wOItu6OxeN)P@I3b{*4rRCU1=I+qkX*RMB>_3L-C6x-IW
z>+Y7ayL;W*ZC1LaC0}gGjt=SbLjxHzm@<Nacsvl~Q`=;=r-yw-ZKrukWeD6X$wp<_
zmTzg1MsF`K4XGSkd_dZG8!|(dZG3Xyiew^@T%osYOYraSfo`ckw4p88bCkqDyA2hg
zCqegg#9@w<is@(Fk%)T_Rl@GMZeKEC?%Dd}lTTiBk-RS?WvDut^qGmX&PtfRq@1q)
z^YhP>R_c7J6QfU?6g6vIL~A3B{E0ZvrtvPSUv=?uDBBttYl?)W3`d&oA?Jw4JJHTy
zCf8nj(@i%?^IDsC;OGH6eUB;S3?JFEM?H+&d)qt0*-_^3^fazoOsCFl%xRWi9{lCO
z=j5G`AMmltT((%W?H{ywqgIc`#viDMnx))1x@F7g;2^gVC?Er4_zet%@Eg!+=o=Av
z!tn6u)~%z%!#Ildhw<xQwHCjDfp91k9-!*NGAO$YeRl@_&s7S8w8|Rv3sY--8f97=
zq0}WlqhqL>xodlI&@cvz+jor-3A;`l9BPO5(9u3Lc;YUFkV$D)_nMOT@LrEPsjgf1
z;xRc66x8I7F3F*7Ngb9NHA|=TS=E#(U2C=q`$EN(PjdgHH1%;~`>^jz^C9l;LZPrz
zZBcu7_7FN+qCp?>Mn1jVn`ZW@*Dw2yS>SrG_6~9Ywk~T%!lcBed?9`txP-Fr=4?!F
zE);wp;2S;?^#`PP<Hidbk-l?7I2ef)w#Z2v<FbF{CYB`-h`oO8IzzXOe#9t6V&UKb
z)ldunq=jy1GszJMc~+k6l}xwGN@=rwq}V0LnJJ+<XQyOl_UWge=GhnbaQl$b=c<!T
z!?WGc#<W+oonmY8S`KBQ)mtsx#HS|Jvocao-$m;PFI8Kn5FLbVL*5y@$o6UwkL>j1
zr#7~uIF^)gwvob2EZYRK4jETpwv{m@3j1v3OM2?`@?Cif`feBMK?Fr>_7>Qg;%#}_
zgvweOsY%W82f_$3`_Knz6a0(i(5OtYZnunoC+ZK!pZ_4c=D$3vaFwz2Fo;HIA;b%N
zAZdj)7)u$?@yQA|s?c}oW{}gWkh(w3!fxup^P}l_F;3-Ed3%@xWNEe=Rt3F3I>%_#
z_u0?-z&)7H!zK)MZLUP*N1UPR|B=y<(;d}~Da$|i`+r`AC5v}YN8MK2{u6k~BZ}KY
zD?!f0JM#zMYIsj1?3X#!a^bMQ`n*oz9f9g2Qn^&BlM&PDmzPwZ(D_G9<ga-Z`%eqK
zTk&pXc`A>QmXeT@)JAhzAc@R_)vI-*A#^SL%@z9kn(*uEE7VhSvb|$?Wc~V);f{8>
zBt6-bUa@lZ>Xj?#G_X^9Hw+DRc64+O4Q;5W?DCaSzNoU7m#D-fK?lyTv$ITEU7?6q
zy=-i9a&kbwd|<aUJ-c`BK51pYVf3#&X_EX>p=c6C8{l0}oARLGHQqz6JBJoGQZQ(V
zqtDB4C>)K!18D26w(Y22P7dQbjL85yP;9mPqCnm;+J7N9N%+E*HQp@?W7`^ejm1RP
zjRq0RI66xPCEKovg8^Mr*le?_C-as#XngX5jVFQegv(aDvM<dbK5G>1svEC{nwqYV
z2MR6Z9WKeUN(2y1*S{qckNU4z-HxWr=hjW_o|>AH=2UNUvy{!vy&f$LX7>_f4?V^{
z_x?anPjOu?FX#3^qTwFi3l?(lm4n}vo6*NHwvSRlr^Vl3x}f+7(!ioy*?hTtH~J*~
zJG?{vp`_a9)}1f#8vjb^qN;<wL-F)nzJz|5GcW9M!zUKbu9vd4tu@^g3b;dSM^8yK
zHz%dXm2`Pi4yV(VOr;_lbJne3<m%R7(B0I;n*-@oFzGZ?{$$GUbZ*SaE5cprK+vBI
zNz>~KhuhkEHl&k&hsOh_n>UqkNeM{^51cnzztQJaH#(iX)o?nIU&4WR8%<4K43TP{
zLK%@%IH9g!`Ei$U?P<1zBFa4$x4ujD<lQ`P@e2A9W<&=0H1cMfSCdAGH6BwC^sU&i
zVMQO>4C&F)9Xm!x$$7x4UsR`Y`9;3rMRfr;;E3;0K7`xl$F)st^FZpM$lU>-TsY~w
z|9+)$nqitD3HgU+qWW=pA(hSQB-Kwjo%E@^UO6~tY(}(c56W!u(^BfNJeha5$~%(u
zz2%MWXw?12x8Mz~X)_veUv`;05_R9_O307664n1LPm4$yseVpwPq_F#XrB>za`mYQ
zcKWqSPyDXZ6CpE|&(j1Q<v@>V@^Iw|t4SRBHg?4a9}LGL!9Vx|c3(R!---?wdE$vk
zFdX~*=h^u_W2IHy=m(=NN~=)kZLy}B6wCLcV+9^z3p^5t#zP<YfP63-2*rP-U5Y=7
zhXT=WeM|S7IYidZXgOf{vm_J<G8pYg<B|4QjfYLQuljA-#@4Iu<&j)d{V@bGC<Y@9
zfvVEgkn+o`>i!9c`x*E=heTP?A=y3YP+PMYRZx8{OFU(>&|#@r0Y~svk=CHfB)hu0
zXn3(0BjwguOv+fSRZiKdKkQF-yeA^V@9~+d0@06#WpzT%5M;PIg+`j;?bTWPb}5+Z
zPnN?H_sei*VtOwPiM_z>|0l*~fX@%Z2l@ug63nNe6zGlzCI*y}?b4CMG|}gV8LpLw
zcme#c04(z|dMi1ivd=NXgnp|nbDrgOvsl2MLitAyhvU<0r0mL!Y#a>v-5!^FaN|g(
zOUgB$b_Ao*U=Y}CYHG=*V6bOC{b_iPQ>koAQxkA(`d|tghVjU%lN_UcdA~pF4VrG3
zGwbvB^^H1CTJ;F@TzJ6}36IBRwzl*-209;kq;tU0+tO;fJf1`XJP&d*dGJd{KWy|q
z`Wc`trm3}or<>e^H)F)i@&yAKKtM^=D2cHkR-gBnJcH+Wy~J%kFEe3t^Y4dUPTvhJ
zC`RzT?+pe5M#~L8rz`yXo6YdZ$cQU=%XuD|yxQ--Dq(ofyCvuv!I)L>{V7K%<h%ai
zkk4bDda4PR!o%15LLo=${XzM<Yp=c5?}(o34eTCsyT^71yyr$8{%b*6!bx8Po}WeJ
za~@jJ>8Oit2h;_n&JL?#zEQ3Hz^+CEX{~6)fgV5xt2L1&b{=lacG2fN0#{7FuMYxT
zZTCf^9Ua}>9UW2p?9>{lX&;l9c(Of%7Y$~6JVj%0X!}LmhXxI=Cs@5T=<&+(s-{$`
zX_ek)+uLK&Xso?GTTfN@7CSrPo9QgZ_#kt@b9kP0q@6zt`kYmI->I-o5{>HJ$S~?@
zSFlSfJj!W;)2v^P!;1JT!J<i!2|zICEn7zJYilhQ(R#(!wtA{skX_v;Y&iMk4JUMi
zCL7W(`X}Ncf9zcHz}dBH*0i^^wXa#Tww~I$Ze34zch9<YTkEO2@C%w{e=Hta`*w0U
z)X778(<p<7g%L-80-{9DgQxWt@m|Hv$jN98k&AU%Ewju>ZCW)ho0bjBW*vckSvL}r
zM?A=PQ*R;WYEPSoX9i{_!*Z6}YPqcVV~yce%Vc;Hjjk{F@8?*(BW$)cy4Ny|4`P%g
zTve~ZsGE$~Jhg$lmdlL~^Cg6d!xH7jY?nL(oI@Y5`pPQhEz)K!?Y0I<r=+PRKhVP!
zR<R#RVk}+pzfw{&2A-^&b>;P2PVG)WR|rPEJ-J-AwEo=7F1u_+e&woF{Y@i7+sC))
z@zB2N4~sil?8_73w(gdoHz}K99<S+&#acqiwlkZXn|<-5G@=XMcymuY_H&e*I@eRs
zm1fl#)u`&wNoZY~`$UytZc_*7z~)@}sNg{TyF#INRp<1;JzRAk`VfZ5q@3p1e4+Z5
zi@e^8<a-NhM1LDc$*#X1n~>xHhw`tX-%-x4Wo)dnsWIUZGPN7kxA*9b%F^rVb*p}P
zxE$_Eqlcm^q8mbhkfL07nSp?Lc83SsaTlm;syb!2dtC@ur^XwHT{WGUd%fOk`+dGi
zH>ss?h#XdAIomj@Yw^fe<d+Z`(T<v2gPK6dk%21{Jq?~-QmPcrP;B84NU&8avVIzt
zy1Fo%4%Ibm2T7f4M&X+tg|0ycl$D>y081#;zPcSI{~dSSvHA{|dtFqz9{Kv$zaDh8
zJAJ;;<+0$`lgU7;y**X^v2Ccx2vel=fInKTz}18OBxz?--i|DzQDhkkeO+fMqwCzR
ze?gX^=;a}w*U|0@ex3ORlF2+xnrg+)S7)KKlM(Fs3N-g2eD*wGX)s-+Ok5&1G`Edv
zGPDeCHmqqG;%SE)Ox8x8&2E80<p}{2O)2X9vYx8*%X;dpP-x8>x=Plpp;x4ymeX}^
zT~8gEZ_Bd|&qL>TSiMhqt@TzTqIJzYRf31UUcGv%UcGwi&8*}Ns*>esbn3XPr=Gk)
zp{g7TO<8(DO=e!o*+0F*<oUKZWSmXSYhnchSwa*_I{vhlkcSXObdJ)SslB6JO{#`!
z>;<D?7@3B?=Z0-NvWzr3Uh(<~cPJEJ*BXmL704*cAMnV~I=2Dq$vxW0+mBiJUKsWH
zWGvbWUdpZ$44ExXZ%Dd~gxi}mzSt;M*aw^ktB}haL+yr4-o*%WYy5^t0mjrHxS9G>
z>IL|kC8F)i+J(yp>>G5z4Y(k1sg#oLVj-Kth#M`%M6sa8Q%8>6-DIZN*Voyei2MEV
zM0;mnU-8i6FZUD*8#WXQJy{NJy8`KSw&9H5<?`dqm(>VZDvg06gM8ED0ZZ?$t6}vN
zSZzGk(c0dgPPezWcEsZH(BrP|G^1$J-CfyiTOiPu%`S81pP4^j>hX}f8yKL%!Ln*3
z>?Dkbd>$)*d_&2bQ!(QKvDV5Km6p<`unv6dKVmNroZ4tt$F(t9STX55`O}~pm`<2H
zuD8f2BIEFY)u6O-^0=M@mrtn17%b)E;XuR~=k}`hL;@}q8>vJ6E4wB?IC{F>2EJIV
zc{IfFz+!gww|4B<VOkY+4Xs?cGPl~kFW8#v4yl&t$)dHkjqb$IO>cCJPbOmWi-AZy
zTrF_BqchXl+<??bv~}k#D-630lC7;RD_hmTrR?bWXt`VthILzpgO7G~b-|@zwO)Hq
z`|6&0Mn=chfq?;U*MQT>Axt+{QLuEPA!wmnFtg`;F#!pAlXwTBo3&?KM-<Vnqt-!e
zQ#YlKqz4Dc1#B6J6vx!~$Vk#o3>}tW5F!V1l1V7#^l0NCi)2HBTxO_tpYj3p)ieYp
z<w?ORlw^9>xb>Twz;D_<8kgbnI8IGNPZ*!(B})n8laT^qx<HdmV>DiAoqc2xeSLez
z#&!Xx_!Y*+X8QX2!jgk*{K;_N%orT2hT;>{jhZ+EimGAFwM#k#cf}TQ5kyqOKn{?{
z%k1{t7P5t`J6kArYef_HYD@dpL{nekQoIf?zLfi(pmca!LRUJBs0)U3<1nvEj}_wC
zG=h%YCY;B9rvv_7!?8Bx2|LylHf}7ead2Pk3nyDsLEj3ev!!=<%epmd+FI#vT(f4K
zPRBw~pWBeIu-zVqll+i6^PHsv>L?s?FzrpzV}en3BRh7qq`fe<?ZUVvs!drzEWf8C
z+5d+&%C)+OhN14q?MI`5Ri^Jfa)Ngx`}sga1C`;5u}lOJmbepTQ@%*}<7Pw|%tQIs
zm70<7QXUEpK3Hy_w08RpDKb?;OBYKOU6b^XoGKIwG2f)e<BK62#<y99Z*jX?y4Kum
zDVaC7##`d8H`^6x%E<6+M2!~TEM3rOTi~S#XEMBw8(DsBz$pDw<1hsecabvz)o2d;
zQOzT!Ms(7R1{=e6d07t)(qCoYtR13oev(|P6P<8rAlA)D?c*^vO$Z-lQf=T|tmx6R
zYK14&K#_V9?zKdsQ^Jbc#S4UOdY3ZPh?=DsI^YMNHQo*TqYc{gO{8{Mnj}^70!_?0
z&;aQjY*e2jXH~Z>=@xQOJ%bC1hY#nqG)yJ+*c>!yJjOr8;h}_ZujTH}aH>o8_n&%d
zf4}5D5o07%Qwv<va;HSVR5<B(o8e7sO*7qOM@Dcj_ckGfckQOI2?r}EO|Fv0Sm|8%
zrhHCzr7*POa3EN(306-emROU`CYoAW28$c-?r7=mAI#+pBbOWG-s$i-90(FK9gf?}
zs~tvrlOF*jj#hZb?SWSAtNWY|oRq<^zo`v!z*t>AwODkeo29$AcTInHONRq1U>Sz#
zcOXNjj7D2Mj#Z2VR6Z#=Ccgqd2ci*`wdJP91?@`tmw;X6g;Ej8ykdKyq6qzwGVnm7
zv&jn&OFOIp<!!oH{)H=6eMiWAUCpy(q~9NLykq_P^_FL8P_0qm!0KAhVa<g-h-<x&
za|NYxo>-nev<}-@#b_f&fn1MUwkEeKhI=HXj0V+|iYQQgt07@J=o|HeL%2*^iuF^u
zAa^<Ga)uJ|<iP6H`5XoZfhd=R;<x-+DiaQmD-=(xTD5-N`le=IpyAZ&b^6p>`#L%T
zo>ama$>oPq9%ll|4fFMf!@2zG)dR_RBII;I6I}jmfjUgU*W83RtXh?bhg=><{nWbE
z)~P!K9UXlMrzbU(&qbVxlt=k}=ZrG=6XTT!F%#=F)v9_ZeYDh-R>Dttpo>Q_o~>bW
zWlRp~A#;^BzzM71LUNYOk*svCTCmxZoV8QcG6sqfiLEwM+)k<ut?5iAET?B@XUTwP
zXX`wRDOqL?)qOTgu=iHIDEq<~{oV-wDS|gp$RzB9F8JB03v`*Nn^qLmA(}Zml2lAX
zyDF$mkY%Xhfc%Jy%!JyRwHqy=K2T|<0C2dYf?7(@_pVI-Mkj%)X_H;P-K-sf75tcC
zd@L+05dEy8+RV+(*{GSDuu~HXJ$$N`4`+2#0Lm%&4K`q|-CHq-<^i-IbQdzLA#gUR
z{>rL?2xn*`E2KFY`6K;nIYg(iYPN@I1fzlUm~-6B=APLiEdd*VD8=>o9K=CfH8rCC
z@kA+M>X&AVNfV_|Csir3a`uTQo-i`gtmPc~OOuMxN~I9Kl=g%|M2HNBWu`T`&#v+g
z!{~P6D%-l;bV^~3VXP^f(geKAa6!qna>Ot)Jj`IY97>RVzDbrn9yAGjb?h~W2rKk;
zbOxNR&Ui4Ijz-g3kKzM;edB04yn%1^c=}?#0B_<E--f?6Ms+W<BJ5>rBiFKh;z(@}
zhir7C4T!4Il4W_a4K!9YX)#)pJV{5-v1PRFI)cFrwgE*eTa4{6q+)g%*J{PKDzQb<
zg}UJ2ONO*C&L`6PrBGgL5?YmFsQmxWCSV)DMxg&OjDBzfW^dok*~D@5a5@I;2%)r!
z$4IgjbRVGqYqUfePG^EUYsWY#ipEF~+kj(4yo+r>{sE=I0v_vj<a4kF5MRwUw0csW
ze8#pqhR8{&Z`4j5JAv+gRr-`5;Umzeoce9HIoQz=+M)#B%E3*IGH_At-)CD`d_-}Z
z99RZ@x6sHz`I-lz7u0x?Za22a2*yg$Di+v?rRuDBoC+QxvK4PL2jLXAd)E-lqFJ7G
z>bKMu;<wxs+Fz>1-WWT1E<9}D$vTjV;jOGaZDEa0?LXG&(-wzAYd@Cxp?bjxWC$N4
z#cT5gXXKP6pznc6f}|4cq0OPhJvJwwvJ>i30e6vHGnq+{wg}73&sZ|^nFJmqGqwyX
zWtG*9EGh%~iGue1qi!?k@mn#kTf{NqXUp?p1g%g4*5x+~Yg=2&1zM{HtE9)?$SJbu
zqfi=YG@whPq;%oCx*Mz|7Vkhy{Iu@VVAIT|3<#p}>FEsoff)`Es_jH|78AKtB^OOR
z-MVNaK<g~(>B}k1cOf{6-hc|+f3bLv-fFQ9yH?eY<8QXEO=W~tIf^o7sUhm+EX2|`
zv-Bt&5MLk9T6TnjWx)7~rC&r5QL@p}p`g<v;=81H+Y&7bh5*=T9-^qoBB~C(#P|j(
zKNX25B9Av{O!s#5^mN?2yv8&x!>M}%fqP*CUH<svkF)U~zd4=0`5e1Wb<%*T-RM)M
zujov!))ZG-h#h*NTDV5K3|*%P7$W;t%%7~sZGr@}sz*i2rt??tE?2EsvBEBx*5)G6
zbGEe&Y;;GW=@pJ}IF$;A9eJBq%fhR58#%3-tHNAKiXb(pGo3OS4B<Z%OnNle2RcDV
zN9##!%LruTBxK}PtTF}6^is;H5ig}*6jgI+Ne013l%OGr3Dp_Q1W&3Cl~r9y9U=vh
zr%&6r1J^m_$5-w%^@Lo3C<(X_)ue=!%ae?SokfQ?9QSxS>XCaL9#1^%brhZ9c+zxT
zrRL^RtiiZpHEB1^@ZcMn7540g6&a7OHN)ZhmS(@df2|$Rx3=H!lg%w0!_Tz(Ja6Ub
zp`8cOU!7t`kqGky-6;Qo{3>uWfE5bP0M%$sK{F7=CK<(~y^3_q!1G7^V3Yxb1G!G*
zaViAeN?M$e?(C@%b7-`&EEL`Tuon&hISoVK6NZzZCp!iTVdQRKbss_)5a0`!-c&r^
z(h`r$SzL<;`uo>6Z!K{W17<a2B136w^ZNe&Kpc5a@PX#jHf%VpnR!%a@D!f4;&}|{
zL(VnM9p>Za)8>NeyKJbWWd&%i&Q=`KO{uo^bUoF0$7?(LYEK>e_4=LvlUlxA{s%Ar
zgO{bf`sftVh0EFwVl0n6Qi8p<9W}e4BHoyy0tn^03(8~E;vSeBq<)9)i{Ll*&ra$6
z(j(qupRrT+vAws?*YCBj*!v^j@teAGJa0by-rwe#ItC7Z>wo@QJ%@VgXz!`t_1`P?
zYNc8J`fEE&4T617_quv&w4`hmPE291)5?Z2YjG5%D}g2pZDDpxZmD+Yt=)rp$BP_z
z*~yI>N5RO#Z~oWn*=4_QC*QjF<()ebPo_M)V|r_(c2;-c>~E?thraDDBegdLH#d|(
zDwXA(P1FUOV0ndQLlVXo%Io2#I%n4bDi+Y^;bzKO7eK|-VL)nXUv;k<)k<MOXt*!o
zQ&jb<zC2a77nUi@yEvEU&{wIqD*LLz#@nuRH0ovc=;YBNE8)Yi-Bp&80|Y=-T^s1R
zJd7=Ez0%avv5{rlJkeI+(`8zqWtP33T51}qL1%3<+Rw`J%TQ96KBOMkW3jWfn4Z@d
z&vU#XImaAmW9!8NdOqi8jSI%JzQ$6*8lcio3kmK57>!oo29C7-TMJu7&}+V=0%z<P
zSlU%X$6L)O{E2v5TRh>9M#I5qTYFz$ds{RZjz+sN`V@pU0Ba!#U-xW1$qUDfosB_+
zAz@%W7{(}LOuG|f7{Lr*uQSx!B~h_zoTc90XJSPzjJ>|LD6X_ykFn`6aOHf_w&`yM
zo1bR9RxsN=FwmV1xe`gQCz<N(OeM|E{~Bygajr~nGL^!{=TdBk>}T(>;&L0tHm?LO
zj~qvCIp^${Mn#Am-ozigop8XYu}-=(V8IVznFE;g%DcfsPS{shESa-BG488}|J76D
zhtQWrb{5*6BuAN6n`J2v>mxM;t2d@A$f$;$Of7}8yUA+^`nySB<WnjpeZAO@b*HXF
z9MJ#5TA+W5HTeEs{PehZDAZTl?b@A5E(&7Pff&}Q(qVU?$N#suLvwfK<U2y-`(6wq
zQUs{YHYI`7CW0T&s}07n?1SIPd5%mR6iO0dTazY@e!JOv5w_LJ6`81?o%ju%wd9rZ
zkLy`ldi+Maz&T!^eI!SZCF|<yqK{KKW?v-+Oiom1Br2vnuX{+-eH5)ni}SK^r*(xo
zuif@q`7GoU{I`%qDry5`zr&^FJO7bGmA@1WO#A)Qh|&)%+sEs?qMo|WZSKTuZO=|O
zqFDBZ%tLh_Ms6*Q%sQIxqk5(8qG}laRrH*PmP8vZaedb)ve8ctebHh#^bIn;tmcQ`
za6IL7F-o8ALWOihenC+vtG=MNN6^aTS(pwSjUq}K5#cv5s{9yKvNz+IPzaIPh#9|u
zPpWdy;yWd?Sk3S{uEpF7jggi_7DH@MR#c<|2kSXVhH&^ikg8X6K}Z_B`;;GyjDrMG
z!`OTXqQm}Ej&u5XhY=@?ffBAer2Gd~#(C`)ZtOw!Sg6%mz;12nGmWvJjrsQWfQuf@
zLLn@#zPb#41m^Dyc<&Ab-UqeB=Zh57x>Jcn+Us#+)jfnYAr9M*_$Xyb7>HGh#oI#4
z>QM&IIMsIDm{nK+&~?@3%UEk9gLX(IdqOd_ZUPQ3y);~a)bZTrk3{^pdVRj|DlWdT
zs-pwTr8%*rpfn-@EXRR>DgGQTZ>TjM!w36lF6_0<1^oiPMUK~L$jR~PnqIt}v$bEF
zmZT{v$L1k`p3sfWd`-C9aZe-1+o$EwefQl56g$UUL3<x>l*q^2)3+c76*ES{_XS<{
zeoLb?N0&OJPn`g4STo5`8D;;JH1+7v8$EPq4JTn<nCehU3ajd4_>q0e9?nvhPGfJT
z-7@F%?DE0&zK@f4cX@n~I>%^`N_uQgTRUaXq(4lTd)PmTsc_4b`8fF1AbBqOj~ZQJ
z!wtPU8#uJc$G`+$z08<-$<R}UV~|iZpHfx~1(+S^WMSL1Bm_JZv~VahFd9|Eas$>H
zalNDja^kq#1$fqT;4pT9`TIqfY@`gex*(VrC9^8K)I*F`@N9BP`8i-qC3Qw|T7kh%
z$k@kP;uokfT&;Pq{b|aKqefG>Z<<7cJ6fyXtC_h|_S{vwEE_Q2LZt2SK$}-8;U7Jm
z+hK+kDK(n~ooEjYSTCcf(iYc4F=cp07ot%gvWn93m~tj)U7D;!r&w~7m=&zfA%}g6
zTx;X&vbPo0z|tF*1e>rE3a@kEK_AE1-+Kh+;HqIevJZeYr?Adaa0VP+kNhExa^p|$
z)z*re#yO4#Q)18nVYI9rG0Rax<Og%aWe=w5@mAU%df+T8H+M8Qd)@BfFlPKmhLT=4
zgCO0Wm<L+I9jKi*fDlxt3<O<Hgfx2NPN3|vL}VkUxNd0gzKpoS9FHdH&|&fynJyQC
z$=a*m*DS}HU|E(Q=GjCdycScp!(&Zg*NB5mbTo5VE$nx?0s*hf9dtMXL7&s@cX}~@
zJ1qSxd{{(aMI4iJVj0sN>+*SZm{=r&N1MGLM&og8&)R9){T^^UqjBtVt8q|Jf#zAn
z6#+q}Jja+)O$#bb3CeTat1V|PDJ@NL4AMHKZvqrVV2dCwkNF|htHWui&W*A;&`2Oq
zM793*aNJ%)F`Nc#(+w!bY|ht=<BJc)vGd@69DE8A|Eq{ne>1+L0nJ9uC%5Lc0b`_T
z(xswn?1}XhaHuekJ*ZN(fTqrTF|TJ=5|w;JZIKO{NN>?tyj5+jZ1jrIXeGUxkBUIH
z=BuF#KrXX+(42C&BM|VNy#q5N9kDkAT`qI&T99b)4Ka28Y~Nw$uRiIdlfsV3nXA0s
zRcA&V;gdq4LytopEiGIw4a@x1Q{^S`CyyWUD*ujUUe#E%3s$W<((CNkSF4A-QoVd~
z`M8yLqYp$>RE~~TtS?4cax2C|QnuV@=mu<%oC!2$yAi0$j-V!SmI^91iA)5FC}*Jn
z+}ZoO<cbw1#}lUw77f|bGO+RW>+ChdwvACg^SGT(52lzQ2+U&`+lxg5ivi%g+wF4s
z3<qYPVW}D@b@JljleUkoZBpU!!z)*$HhDH~ywF+%sE=x%%k6YP7i5$?BIj3Rvsj`B
z4;$_P&n>GLlp6fElvj`#jHXzk#gIbHuF^1jHqfBr49F8k3dg6IWGF)@@F<y=_3<96
zMp$-1-3R&_HPPOqOa(m7B@3izNny$DU@(<xsi%%v;heyO>E>oEEyOIWRAU@sJH9q@
z3SwTa;+KCct&M>|Bs&rz0;Q4W=;zUF&R6yOn6G+=YNv5~7-F2Qc2MY66Smr+HG!Q7
zWaedTrj<Qp9Zw(Oyi(1kKneA62c;&|&D*+$Fs0+9Rrb&(YVWb$hcgXlZ9G<>kFA0J
zb2hMeGrumx(P55Fv9D^<wv{nGVs)s>rafu4*doDm@QaO**5hAj1^XV=9h9<cT~}r+
z{4qmlvHBoKJ!NH92C51+B9+Ehs40SznM_74>xi|;P(M~SL#-+jnZOkrZK^-1O(ZgI
z;YJ4?X825`Axw?h;s>U+`bHgc5)P{-zCk@+7>ABV-K(J5TenLXw@qQjUsz5|vH-Z3
z8`T2kGMEW{4LXiFjdg>tGHVGjjjZw#WfM3D6&J}{DA{yW66n{`H`6(-l>x6x=BYi|
zs}gB>H}{1G&9gz3*J}i!W-X7EsEHNaSQ)ChncD&<jOb)@v*2O~OVm(I@mNj~ou`^L
z!}X)iKAYPaPHs8p945~>=g|5hhVwsBUnT7=@oWieUFTP?>xEeHFtBPBr?vL>Hl~lc
zp8pyNgyU=1awcw}(3n2dhB%d9csza^0_sF9u3T{|9lK|(7pwI>YI`89hL*LvLPN7|
z(@u#2duiCsn=#vVAmG6-5IGjpVjo64m0^;x7meLDhG&rj=vVnUR2)8Sx6Sl(9DiN0
ze!ewSQS)Y`>4s9|tM<w`;TT3f|Ed+9Ko8X_I<{M|x{_-}z~%M2`k(=L3)5bg)3xoG
zDt=JZvO$(>O!EXpFP2TO2VacXg}T?{-noJYBa*M-?C@IswcU3}SB@U7AJ;5J^mN1+
zT4o~-O!^gZM(a<i+vjN-lP8nI>TzL(8W4X597eOLWvxtohq|r4!Fqj7YW_V*$Y?Ut
z;U*L7k|QhY9t&kWKDsGtF*?QdG~3;Vk$*keo*cvD*~v-z5O|s+Hqa1x;2DaB;A3CT
z4d58%I2YKX@~<rQ0x(DdJpxJ<u|YXveP|_JQmch18w=Vyl=ZR{2QgK*!E2yo@*ysR
zS<E@^iL#K<9@FQQ=fz@JqTUO`F_s7t9k1$7*asi2&*d<)O0Z;2hs)>e$mKe?uzjMb
zDUk?zv5ab62Cxs|dv?d~0kLKq*=8VgOlz2nzB9`#RBLCA9E-1H9K<sL+F#|sdZ!XI
zw_%#76US6(%RY{D7}CEBuG)@>?nV>nG7W9bx}MA#3bp>SMC2W<pIED!&~Qs>yRM`q
zpe+%Og-u*WpLrn^D$s-8**T$i_DPG{k51{;f96BT0NF&Not@o!XP;z!6_0>_U>5T&
z*F&2+AK#ASx+m0MN-C;84TFbiEw4EqMKhFjqvy!35j(zB+q|`g1Rh7vcYrgHS8@?g
z+|DD|qj(ss6uQ$WSOzSQJPgx!j&9#RI-Y86O^t6}YZz;>{!cU-*nH`wo3U(9C=^)k
zTsP(^1cT!PQVuYw9%eS*!2wU=VzzNzqju}$_|dbiziLUYvZww!zM)p&V&jTSh}}L2
z-?|~cjx{;@preYCTlq8@udf5JZOcEduqxBNu3k<|&bxrRQqd8KN+WiU&l?K+Hr^Tz
z1|4N15{XsI!}7f9C**nGmv5J^!>6O3Equuzk9*f$8S?r9rwzMI#NN*&{C@92(el-=
zesxg}-MwYYmj6?I2JH-4X!-j`Fb|=OuRWcPugYG>FNXHwrF0jdDLGey`g7LA7EP{*
zS2e7fUST!2YI+Aad$u?P6T?hnl@P4vnjgZPl@c;br7bpSIU_(a_|wjaG}a~|&45;v
z_#j>7<!4_1`qz6*H?-4q8X78W<kzQNr<<5%-rc!!i`(OIZ&}&d?Qj_8>8|u=U<4aj
zuGesE$5N{_iC@MikG5lZM4YXU@6}US<-HJe#@&$<BJQ{|SUBGetJ~>0-{m^rgHI*p
z))br`({pxre4Wd+F5Z1M4xEKGIl1-$2;?6lU%ta|{1{#}m$Uk{#t?xe>8)OG)*l(8
zDmfPMXNAdeI^}x{BerJ?<nLu_CZM+X5*VPhY&=J<W{)$#j(uqv+dQr<jaqPkwoPo=
z%7sJfl2)(Lzx*`}^E%}KwXSN#S_eU^U}}IWVvieDr?EQ11g~K8G$bvPdee~JyKmn<
zUeKF{w46l#`+|N9I~B0*_@v4l@n|^v9JjJid(l3X+R3xyYEzp|tn3xqEQcq^t7~iD
z<Y7;r&aV|OUHL{4vkq0%?p-}{2o44eLrw!>w6zoId>mv;_pts1m11h?E>0J=CrmUf
zY_rqZFjE4eV^h!XeLmdW+{t0P_>VugVZ(+$&6sO2vjR(LhGj>pK7FELv6?Lbf8)Ff
z$gg;~bkluJw7@WfVsO<H)dMgc2iNxow&<xA8G9<lA^sB7`n9b&md}k~tQ5{!xMtDO
z$)szm6u?d)GdT9exYnlA&6!MdG!hEshlcukd;5ll@?hW*o;}{BxxRsipzei2u}BWf
z4=_7QfHaobr9MArl;O+c9JJHG!H#1d8d_V`R52vBJsBKo3X!Cz3Td(^9ql$QAOP1^
znOixlS9GM-*tmAUT`2^tg@bJaD_Xf$OzVn)HjXBQQnpb)O^?e_Z<=fz*itrr<DeFG
z;JwgKu~ac8_~8ZWMV(l_m^!N!R9aAPJIYPyLcnJR_{;#G>_ca9^$M)a54->g<WIzs
zw(7~m^;n(8R<vY|eA!&=5FJms`n66+D}Y%tAh=;1KwqS<L19Dgv=C&~Ar-3bvCo|0
zNhUpVE4S4zIM7?Bw9dxe6nGO6Qm&tpr`r$OnO9#t^o(puAhTbojNP83_4I_!WMYPG
zO8wb7kpZsMx=)<L1V8e2d?)vEP~WZiF6dv2=fuAPZ^I*;J3LzA;#tHMro9)z%$4s*
z*+v|aqKil`dqhV)tpNRjBuZfAyHc?bFfMc+P_b$ZANLNa=3oQCk=%`7cDSY0#Nl4W
z{_D5ukXvn%z4~hzv5m5gV)P!(3=;&obEi+?Gl48l5tU0|*bN?UkiNkol%S*F_s3#>
zKe3gFM9jZ3F%ZSZIKpZAq}X%9vfqbQJO4^;*6TB%_<b@Ujga98;>?|^SFg72DZKg{
zwuC*eH?Srs!XaS(x@C2)@knBMX(JSLu-x*t+A9JMmig-C%e>5KyPQ}v<!TlUn<wr1
zM0D=NN}qMPxA)a^j)Y>ZUM@Z5)!P>u4}uC20jxry{Q4Y+o>Or@Nq)V&!EdaG*eL0`
z;t*Qwpp6Oy9W`q3JgeO?bD-;SG8n|;&9Su<11mE{Pn+8w*r#<dhpCB^_3KQ{xjA%g
z*@en>$iu$?UzzUb?B$B)H9cz@2i?;POQ#V?P4NTmLe*#TYW};`J9Lk+o-U(^)>Q0T
zQOk<^&ZsL^G@-JOYd5Gw1Qx!klPWVS2f9UGGu9(WMv#SF|5~EK-W|paW2`>|%}cL3
zrF)fHLkNY`-d;w8u;A><dPQqS1t^CoXYukQvK(FJqlxqohQW0hjz{+IIBS5PMuA4O
zms^h<I7^?o!R>m|<=S!9rxJ6CC-w4cDI<R3s<U<soEW!<&*LW!>^N%`kE@TmOcS@i
z)Y=`7cegf_e@JXnUv<@E{6xJTO4LoEIlMhPLk}jh@o4XL#i7)b_i(NUDmDTIiR?-@
zt=Y7+z{^)^4_5VSmvge?Ec`l<Xx|}cS^Ohc1)AayaC2vWJl@~2W3!#?#HVQ?=pn0S
z(R{PwPSqxqO)2Ovhwn-5lvk`V#SWRQ)&sGQ1B@LF`VKcf-=Wqm=kPRgEMdr%)`|Tr
zI}?|<rD4RuY??*(H4GO_D+Uk<VYNuGheP(56s<o)(jJYG3Z2ea&IZz5EEJ1*vG_|T
z9D+WzrmTU_I|hT^euTZdOqKPN;yR|=3+{zjp@!WqEbZ=bM{j3a+=4GvWOpfA?lh$S
z#|z5{9gFyzy92qdOe&FZo6hTa_f*m$-SN0PGa88a{ccx=@7NS7+p#X-GGZOqE5X`$
zT>pZYMExxdk=Q|m#n2%BfZ5XU$Z!q2bu3$pl@R$vCPkx-HZ3NPayD5|;T_#sE*pav
zLo(r|c$J--YE~(@G!E<Rh_^H+;_;n5YgP|qIUHDPTh|oa#5j(y4ftYxU<bM_8u!kg
zo)HY6dd}N;zR&0P55VW3q!XzWw2HNU@bqJjcs$YEg6tD<bIqEq9M9d$*BS=5&TeBu
zAuw|naeEhWIi?I56+HYYDMxyGobrU7e!tJx?{}Gzf95UvX3{nJdIfxR3U>Lu@D}Lh
z=Z6%<X&Iab5(_FyKq*i}hP9N0UiS!VCgq0__h?e3%0n?D8KaS)(5ud&_Kd6~2DNvV
zs|(O8LxzyZCP?xS`FXxnId|AP>_n!B##$YomQctAmv8H*9L`v4&cC8No0LheV&s{y
zi0*e=V{xMl6o)?=F%hs*qt*#ERVieIn$m6lXaH+U$h9HE6+ZW=kjn^tDFWf=`{Lt>
zTr><f##>v1Ay?kzOtisb%sP5j+@@A#LNl=ANa;jzVn#S5-4Z?_CoHU5r$gA|ZEej)
zb1)Qy76Pr(6%HARl}0G+T<MHTzv;o^o%mV86!^x-2c#%H5kwQ|oFa<HgEbtp*K3!p
z){AYYH~9h)y0k?Uk08nBv)f1wdlYt1`dvJF$aI%s5vPnb_u1C(F4Dpt`{>(QO!;_S
zhPR~7%nW{8kJyn26}{oKv(F4F?(1)-Nv$5FN@b<Pk=Dys4qwZcUE9-kIV?ERS37(^
zJ@RYt{7j+7IoiJHQYqAt+>EJIQf=v*0T#Lu+K^MVu<1{ALfmONGBvGOl%nXyrW!H@
zDxAm|rWLaTNj4lI+fU#d*cet%dpd#$<ZXqWhf?E<Jf%=y-v~!9c5b;S>2yS5&CRLk
z5c*|vBzlGm-*p&u#*^M5GZb{5>WRgYCkFf;|3zz0!4uw|);8}LTe`)a2qMIy9iQlR
zcDfQV*E)Fv-^F)cv$ZvZRqODX41S^~)%s7v{D8v9DST1V;fw~~-I8wdnT}-0;R^j(
z!t07p1)XN-3r!ftA2fYI-zoiw(9QlJ>~w^7#ih?1crfggp+CYxztO*fTbye?s@VPY
zViK0Yd&LJVIYq{oI)bFKR3%v33gf`B&FW}jaZEJ{Y65z#Zb{Bdy5mqaY0#Et-KIWP
zA6BzCEi$lqZjL^>+97lx(p|)lRUfdA`dl7{ng#q$x2C1B4Q)-$k&{h-(3|JS&D5}c
zB;AxqL(ZLy<-GdjO?JmL-SLdi7wd*I4ulFNi&<KO$V`sgXLMNXT~W6aVas8?V<4UB
zG(wSGB;{bR>58D&^snPb(-KlBweBxt4reIb+>*Yc3u{n-=j_r!1c<T22ZmL|U3DK{
zG@&+N^g03ku6`E%z%ANT`+OPwV27B1t;;z&^cfM$dMw(;Gcm7P4ZOg{b_2e#_^cxE
ze3GH)3H*_tDivx7Q?{Tir8Rq1qDxCaHZ<4+IvbaMLUY#iW9^2~>ZHwp_)@ZTwXltP
zP$^<$%gJB2CK3)a>Px4&#)xO6slG0pE$t9l1fOxMAK$HVM;+mi_nuOzlybX=9Kn$L
zPwfWD#rhy1xfUJUl2!o`)<^E4)(yrms_!)WX89M)aYL~M9%e6I{a`v<TLEudBx8xf
zF_g{ua5zzb5DtGV<OpAI%7;GmA$*e5?F?Pb+%u|C3fFcUhEGutUf%>mgRfCy7c_)t
zoo)4-GZFYHZ2vXqNuCSa`!#yK92kYO`-J^LRgO)tE2$AnSR&d>p&S^;?g3MAts8p%
z$++3wZE!Iccr*)^TcLcsE-<+ZAGcH9*KIrEra!qoJ;0S=Fi#mXUC~fLn5E}~<ESq>
zT}Db)J)G5GPe1<ILJ)zrX+T9^E6=vTBl#c4bRyU^1rkz$jcTW~F>}mwjQS3FTxHZy
z?_sQs^E%X<y+T`HIt6aaeuCLj#DpEQuY-GRTg}zQX;M~X99_NIjH4Pbh0n0D)U~y<
z>%KRX2#W*HsCiQ=%x%|_nLIFW`gHeUM?Atbs#$vq)eK)ku*X>0maNko|4ZtGu{K+)
zgB(^;*w&@Nyr3N(PdzVPgC9ra4*jeIH0I;0_3N1v{kezdcPJh(`ubncwv0`nGNVN6
z@z?wm^hA5|)1@r;an|=LJa12gx|A4%C{q0FpNulT`<a0sdk|x<@MEiZ0XCKO9jy^M
zaadrbr=?v8y()pGO{{+$%_EU6;j1hBCTY|d^~um4ujkKZ0$64$yvO5ta>nofk6-QH
zEyt#(3-~hTZuzh$P~97GmmS824NlV&kmKrL;|7;`MfE4Lt@@KyvVC>+N3w%ZsQ##c
z*+%NrPqdBl=gQa9j#bR;Z)7GZ1=YN(D42sik7Ed~LKI5T91KH@G63KKIm|DmyJxm&
z&d^nTY9MkqdA^$d5S}QwqWGFsJ$kObc<tR+s0g+Q>WtA6^~hXU-s*%V;q)a<rw)lz
z0*Sq_sysXR$k$RCaxy(5uWim#2{J^BWJ>m-B}d0Q&1uf&0h4y(n8QrBHL{+@H0Vpk
zY^e64yabPYjWva*?l6zN8Z<iIdd_k>pqiLq?T)#c6A*U|kCGMx(Bl~Yxk7#WR10r%
zdYkK$_fsBl1cT--s1$P;x*X^?U8GBNsIH3;<P+lhcd+>51$jr4VQk!J7){kX0`l^L
z$5W_2jOjw{>N@Ug2|Mud11yGSG`U@O-|ccY8E{vtFrDk-iEHgX{R+khrtk%n$Hb=v
zz6T}AduRngJES~i=zo#6HncnZQPo#7-<ifa-SL$vpdC$of{W-XTO4_49mSnzWwC|M
zLW`Z|X2Wm`Dh=lh8HX}XUC?M4`-;2Z%0XE;60g`w3K}R%kRF1EpUA_g0&&TtjKIuD
zLxd$0nN*M2iZ32XS1{o7d;J0EYM5vn35P<b_u#7<xH-MGf8E&1jy%*yG*%|=J=f)F
z%i?1(ZLLj+BZ{?}Jsvz3$|yar?*xa@+P@ATOhEXm*X(V}<=eM<JkiKD3?GK&_ONq1
zm*{54hDxZ^MW*FDP$YMxDTI~sP$I7v28+Ytg)-{#hfJ60Lzu|vUGZ4DqoZjPmagk*
zmEnErg+6by%kNF#I}jL~SB2rCg(%NH<lodQ<w5iwcUQA)+2ry0@sX{Qxr)X1aH95Z
z<ve0>*v|F7)r{5oPJxUXEEa|eL?u*9Ha3y1e1#^xsH!cf@lz2Q!9sqxMcHL`pe!{j
zw$4E*pK&}M<#LpcXa4ryd-Xz;_!9i2DfdMJ&wyh2+4pG`HJo||F<#U68?GSNqYS!?
z`+sut&0LD|v-lpb`B|Cp-XCp0Xa^V{Y>(dWMcGaIP!(U`_~|I(K+Zu_!f}nF#zCwg
zJP;tXEYq!wC%Q~59okxQSm}<qYHN{8?N4n9s_9~2;hIB}S7Whlc+@o4o^G7Sk)if0
z)grB_ba$Dqd`%yRH<2BA4vLwwMh*`BP8x#i8rDY5G_L0elCUBJ%aeA3>m%0A>X@lH
z$ZkCbt#LR(`Hk2O-N=;NM_CUQ9qnisgw<~2bIJ6)V;Oor8q%JovZ)-C4I6u`2iPjx
z*qXH)K~F!OJ0h}_mQ1-dVGA#im~ho>wI+-zZ-Vr>2=XVF=V+z2O|<8XS~?gKB`FXB
zUd0pUsP;}-3x_C1XN3+$VWZ(b+=XZI*x>-Da3DVz&UCIQujtH##tdWq`km@q7Nr&Z
zj0+6(j&esbY2Hwe2&$*j0iO?_C{o*<#>#<#m4@CHdN92d0mVJ*&a*#^ao#$$oNgOG
zwH1xEjmsmf8-lM8#}f&D{fXPYL41R?Wl(Rm4x-|bGw`sVP7kDn{?N$1iK3ceLX?3i
z(_y#F>a|?bU1O1=fk@;=p_Q^h3h=lxSRGmx?JE=!gZA0cgx$tE1a}5z<&rM8boJnA
z3?pck2~MaLG<GiShoO!&WoIX+@B(yEzKa=7H1=Xm8@liC>kcPk9368CsL$Ho5KSPg
z4nER|Fhsg$hz*-pQqhDchck-}uZiG@9R`J+Wnom+GAPm!KK8J$k;0XywxNfpUL!N8
z7NmVxeT<8F5}{LyQytn}8f&bq-$Qzc8pj0>=J(RpKzbw~HOMCID%zesd`%Vtn}6x9
zh7T|6_NyVhI^}SL0uhfJD=Rd9Z23(N$5gFkv+8!k^hW%_J7(E7vp_rg=^P3ZLF^}D
z6oE|YOF*vjFUx(bc(o9zb?v&`TcyW)xfrnysnVl_b|1FATO2NgFBmoeVjZ$T8uO=C
zmW-o9tng9;odhhoNDN3#eL-YSktC39Qy=DdC_1jrMYQf^lU+j~IcLD=b#&QGDho!N
zGgI}caDKEy`4t<U5zL3Yi{G!Gm0yEyxyQ0SX@8|4vT5tsX?qWeuS`ph7#@1gzQ&L;
zJ8iiDwP^$uYy>GN1IjeF_jN<nvI23Rea_ye@uZ#Rkl0+xz$juUw02Vx@hnbHF9WxG
zGYN+$b>^l`XQoVNva^>#-Mt<sLTHkk*3u45vt^fgt}_Ym_|c!E3A{SZOxB)}KZEyo
zCJ}WWNO>?x$C1qCk`7GH@uUKXM^AR9JdVVfxzjPz89`jDNmkagr{~UO8IR{#w2_|I
zKwn7^F-zO<72wMmvmA%6tp?51$W{rOm-YDx%MC=HLaw4&<<JlsOvxM*+=pR)npVMe
z(n}1wrRLqzqfcUS<NM+c=Os?Z13f)Gp)uA?eIHPfoktyR?2Wcx(jNV}O@njew(fFx
zLf+GcpVQe|DwWte)qQpkkcyL(nlQEeN*Er6Dpgx#DGOzUV6sQ+%@NY1ZQ8$c2KAOK
zR;WK0UuqP(ACfHfXWyxtg0Rr5O@${Cm;NeEojED~iznQsNiDF>xxhITit$xT%KFB5
z!V&E;n_xMUS7O;6b2!M4*=~vOjOva0BW8-B&agMC`pSH6oh&OxG|O+X?n8}ry|Jb0
zG8dK~_WQ^2!OBb2!Lkt@il5D`gFqxs`DNERdYo%{0XK@bDY@K(QP@+#2rSWhIET{1
zFXUeQ7}F9-jjj$%!p{M3hf+?+R3$$Ib2y*s^HDQ_V&e8aM<i(R3~tM{DI;*rv8?bm
zE-Pbf9i+f09`$q?3rF()CXap&PQ^OErVjve)}^N8=hZi?b=)V${h)dv)-$RltsSLJ
zPHabS$7ws0!IOz|RD(kv6w?XId_9hDw!YOg66=6>8iMN*hItKRyRV7VKSx_nVI7Gh
z=vUS6n!bcFp<8gmNchY*y4`PVw9BmeS|jNy!nIaYj<e-P@D+Eeql!p@0Jf&G#8Gb-
z!}FGkF1_HXGTfkxrVI5_)>Fq@-edNEERQprW@|nkwC_>Y=7Y2?QpKRRK~ZMC?!9yi
zutjytYF5Zzty~CPbLn_}Ju<&s1O2Bfd>B2ywzy{pgCaP*QD^#WZS5&itnb(deU@*V
z;~wl;7GD@o{x$fu3*VJ<L((NgE5>;Q<FZQlCWsbx>FiJ!R1ZdfLR)Y-;tn^Q4-WXc
zA>Guf8&p01`26$F-}&i(+_r7oJJ$|A@kI4uz=Q96=Q~dwlTW<(;)|x=Z@%dFO`3s9
zz?}5?<?f5|Ydr(M+rIX-!Wn1uY26n?g(4h&$me?~90vJOp1>JbhQF&BB|8)Iuo+E3
zg}p8l*ponNN1ey6<6V~Pr6qd4$ZF7F=#-ibX(Lj_Br2{tL7Bl+bw&p&*0f$7oxOLt
zMIs=2)yQ{~TKCqYm-ZWHE;Dv^Dp9j=8kNPV`XU$gQ~?HbO|X+CEB*`p8uC`hs@oBp
z)PiZo9CdC#c<Z4WWut}&b`Oy;#cg$}%p%wQ<JuVe^m0Qr9CBG|@z*L?ALEkc*DW_V
z!$BKMPL`p2oh2^d+NvskMq4RlqT27u&}MCqDnWu~)PJafay{x;8_E%T)dSwoC(yVK
z3~F#>Hp>s$x}k{ZSqWM_qc=B~hlwVm_(Cs)D}1=ILNLUYV(~@cW-PA8@Gscj%^uGt
zDyZua353tZ2MaNJdebR<cNx>GQe)S^r2Smv*?r&q<~I}G7FWO@`}deX;A-*3sNBX}
zfpBKk>d~%LG7xMHnPxDMT-n>((H>#`d_|<)=i5Y$c3plI?lH|!EAB~kjjmpm2?t#G
za(1<V$n?y(j11%RsQeOr%;y<~-J*33>aQmuy7Udeq1K};LI-EkMlj{8QfLw3O;wg5
zylToQ<gxAwwV_!D7yx|Q-~?2{>PW+<1(h7nzf7$lLj@soPYc|juF>I*m^%6BqnJRs
zad^}vqmh<-^!496aNt0s`m=~^^Tw;cuy7++E9B_$+YB@7laXjkz<=AcVQ><rVNBoV
z4=`&Ck>XX=UHI$}r$RMWo;-u!eA$b-GlKql@fOq`%?vK37k~swD7Aw8OPO70e(0to
z3^qKSs?gu@kw9`c5*3k>Wmi#1;cUY$Xs}(qbi69J3an_Q@!%HaLUEAzNNPX6{`%{o
zH@YK1BN82qhXT>TQNMf8(`MiceiBP_1%vTh#z#jCN4nT-82P>v?gh!nQQ$_N(9)tv
zruu|Ji#H)x#JxX71Y3+Cr<HQ+3M4L!_z();6NtvcJ$X-{8+0q1`!4KEhtlz2&=?sV
zk4E45zCiGVzPwSLBDr)94>w|r7?e?+Z49<Lb@*8@ni&zmNHMi|Qcq5+Fc9nz2tX#4
z`79>!5i#1_(Y=ufBQphoD2aigH}p-7d=636F*Wp}%=YwTrQR|C%3ttjv)BYlMCf-2
z+fd;fX&(Rev(G+@-n1|7ZFSaqk~i^9*|#E@NF-P2?QYDO94B>juCOCNR&?@scDZhX
z`CY$)ui6b-38=4ZzX@77_z7qq{024QPEAun_5W{k*ZSSYb=)z#z%HH(fDeG;Lxf22
zAz9%2At{n=Igw2(uA^ACW4pE`5Ll8hK>)^rl&G|c6Wd9g)Tx@*ZInl=G_B(_O{2JN
z;-+cpG<~FPn)>9V=cFg6{gOYRuhX7<Dc#@PyNd@2(j?5JAr7(k?!DN@+?lyEbLY<3
zq{8fmEzkU-HnA{pFefs}aDP6&A+^9d77hl(UGn>4H@s{ku`m`rTKtdCUg*Wvr2dm9
z`$GphdN0gA?r#YOL*ZB?s(HI({`R9s+x@ZbooMJ%?T%Eh+R+vaX__8t|Liq5*mSZ1
zvNIA<u6?!*xq7uwu&o0N7yXLTA5}J9YD84Lhv!wyTQ{H!ztfH>!!|`8SerT=+L6gd
zrSm8oEriy`v<4fEbZ!^Ai_l|4i4{cP(zrfQ#gjUU0yvo{#q15@1+ku2Cj;ak_aAAe
zU?56k(Lj*ekNC#|6qv+LRt&fN@=y$v6pQ>1yvtf82j2LeLr%UI#8MJ#dEo?M1-<B2
zhP=jO4;+kkU%?e|<tyFMgAY7rc!v~sQK~-jo9So^)4hZp?owgEJk4SY(~*#!y^yPf
z1@aK=1cx{s9s7-7W;!&DBfvthZCql@8i#VB*T(3YtkR9KF+PbSuIH)CZ9b2PG#`RZ
z4XCFJj0FdmNXaPJH-s-1lf;K*a!BQD;hn%DlISQL{FvuCSSyKkwl0b`Tg7XlXp1Uy
zd&up$7~9V)8u&{cZJ{oBj)x9AxzYJJHvCbJQ;LN>P3-)~Zd&ZYrLK5?>)7$RlPBO3
z1}Z;!Vs7rlNiL6iqH7X{@|z89mHPQg31@Hf8valWJA#JCojmG%3=bhr6*fley3X@8
zY5u6uQkzR%md{QNbbh?{+N>>B>WNATXKU6PSm)#T`OU|Bj|mlt2Ve$M4-bwQ0MsyB
z0Hk6ioGtatO=NU<J_QRPLwc`D(r%;YzX&(yKu2$HJOJl(UnJ_=wH~#*`1!G5KrVv^
zun4jVp63wp{r#Yh?IUYN1!RPbQNdNlM0^CNmQ0tWK7s$ULpE|wgEyfA_>$n=LIY=H
z1}g%?cI3o12dVd9Vweku0<A5boh_|?1FN`!uI@mj<vMUBz%GcF3{4I7`Vhwvi3ofL
z;bv-F3aOgGtJ9pTV6Up%8bTBx^af%32FGzU6!;*AuHs?}V*eV3mSBHN?BvJ&fyI6e
zo6r*c<yKUQ)XY07yoho54&d-J&H8J20C<VFfcMZ0W&i*br^J>P+f)s71;+>EXD8pN
z!(MTkT(xT-*I{B7=H<As6j#fyZrgBTbhM|tyJvLtM7301tgDvlR;RA>)?i`KZg{t?
zHv`Olbuj0A+!tJzunvBe_ZID7M^SCgZR5CL7WPslI6Y%K5MLf(IpvxyqYu@NOgnxD
z9s(94=!E$VWHJoQ&?Klx{EP{vG58ziK)A$G82|r39%9NXToY!*rxLZ&OtK>u>qyG)
zYqgTdQ#3R@48IBS-H^#_pDID97kp<i1@oPY&U4(@_8#ZAd#o)qagXbwh)N>t%8t=a
z4uLdiO=H6>&!Dyufq_|(`oO7S>H|3KMw?$Z;OYC26Xh#@T~|W~c-Q-ZkcxS(Z+N(G
z&`w2R^O#p9<qLMRmwYJiRS_$)8><!3ZokKa<eguV7n0xg^I|=I-zFr(M*+iUT^|QO
z7vXPp+t_PEQ0VOa5Cw!}VoS^bngg~tni!iQ5%E4xAumn}*CNPDlQ<)k_wja)+74V7
zL%_O+)I4?xZC91-tEKJHZsR3JHLrJ=HK$AXN6;{F=B1lAZ#u3ol|sjk!_b)XHzYP8
zzp^;Ue~jk{sb`1UbB=Z5Q2o$s;`eU~w>>i~vs96tISIydm(bp8RLsYLX?kmK@8Do>
zZ!6-Ag^x{49Q*E$P;JOZs(id!no>XtSk{1J1qee41Txzr_x|#S?O-Sq-2j2v+As^0
zZjTQ#C#h6<<)kr--ab<WHH6`NYGqn`VMD%t^YAMR<EU*KWdotG64sss7`Nc!)2^>I
zEqiQ(i<}w}8YHqZ2yS3YgxDTBhMaNo4>MmyLjqI*Wf;1e_|f?I{*1>nya7c++#raI
zu-K>`cq)crDu&%mQ*Uf))I3h3=24p;uXkWOsaxat{<x<?Zc?=F_I5>SZ|{auhU!kO
zm2v!*s~ZvOGOka$z6U-pCPiTwW1&I=%r$IS0xC0)a0Y^zHQP-1Gd0B8jbz>q>yE|c
z;HWc<{c>6rf(u^-Ql>qIF(?826TS}=08GdHHz~$lUR>UY$6X}&<mNs82;=)NNC`n<
zBdBTKfL|vM#}3i8`%d3$E2%XKD-u$NLI<eg4+g)mbF;iLN@+v51%sJa7$GAZhW2T-
zZY1kR`NmF-bcY*8<w!?7tZ9C>5t85A7BZ9v5U{!Cd1QHJ)Z<RhIzDB5q+^=af(XzX
zKaiZS$_aCO!-9Goso;dnEq4zrL(lFvZ3~qd+J(lNobb+@kvaz^qPSS8nqlk$O4xmX
zFnnfz$n=9Skf~A4IAg$o0lT9Sar5`PVaSLWfuLK}MOzd?y8+{!yLsaL0$=xxp+@=0
zck@K}-W?chHkufGx2TF9_FcEYegQebn)JcnZho>Y+zeQrlY~mgb5q>IHgVgi%i|&%
z0rmF%f%*?PSey4@9jx&1VUUGRc0qf@>wN@SaGD46XCP0x5ea1sLD#UFdCM)TM@P}`
zc00u}kJoO7FZ1g^e*G22OW(%Y=qM!Ii->OeG1upq?}5SawgsM{VQ&+2i}a!tVkwvy
ze`FaLi6$XKVdh~@1*XIwly&<;vOmBq)QDLRR;U0^#%We2rr84@HX;0I7D56*FRVfW
z<p(h?xK^A7Ww0&unZVu=e86Yv3t=RH2pgg9o)*8N>w%3MXD;0Tc17_hx6RGXPV4%S
zXp2`r;t$+PA3z01+3r?vo4f0{Y|r|F*re?Db#?Xi^%)cz>^~ge`0rFI<<~T|M@6VR
zRXgrhJutg{(=I7mwQ=J=`TpIl;{N<N^s*7=V?fsxe@jny==7}t{}J8W5<Q}8GqZDZ
zx8d=Wx8Hx^%v<C^H0!OA;Y#ykK?Bdy)#dZ+?qJ9l?>{^^*we$$;!hz!k4JSM*D$AQ
z>dUJ8gh!MPDw^h-=f;hNy1N_UO|<kFxY#F}wMnULsS`eT!t8_r#7t#JW5h5D>k~)@
zYWSvpJfk(}7w2IdteFkjdbNA^fx8utM~M!4c)P1kU|W6IMD5nQ!1g_Z(HnLvX8$gI
zbECd_`r;|Au@19?H2Q9jr_V!BCc6eah@XhnD^>Ljc0$o4H0;p!r9hxMP4Tn^SyC4^
zm8Y;?(+$78`&^dm!>(_-ehj`B1Iw=$Duh5CXF9VjDgdI+RqlwHM92S-s<Ff&xEnWB
zNJnoQKQsv0aSFK?M4GRmDK^B)k0kn`A)&}hHI|8$WfhPtODlVpMV<8Rd;XyU>CqF!
zo`6%Q?z!jGsYAMkd}2y$dg|=isp;5X-aC6djHm*U$k{vR=kGiliJ%fbK6~$fgj-rz
z_B4!0o8eQCAPPe0k6<tlXu9H4D1=yY+=}Qi;j51L;^<?nqkXzwc-e#ede!YVAb{RJ
zh)^4YgSSH<#jd*3vUZwc%E;(Byn%Z0)M-UIed;1YZo2Ot9Z@!Zh`?8&aBC|z;()vr
z&Pszb{1=381`MR&Kpx#_ZRHGBL3gd9$9R;h7hU!y_SEY!N3mA;5F%Ya2fP~;_ycAO
z%LR+E!WX3m8#H*Tv<bA^!G6@hDijxjb^Uq(Od-4|kDCdk<C1iPVZ^1^94VA~d#BwY
zzaxP1X!%3#X#}F!o=k5_UC=QYJyox9N0b9fiw8mRR8I?vj{C~1j)-~d_;_c~m<oog
z(xnlaGJ>7s<F{7z?`r9Mtuk7-^zz!kzyROC=R@q90|SmOxfTAGU`K_WX?yqsfJ}|M
zgTL?(f>>$02nc>4rJaEpHh&1uflcy`x|iV?XaN)QpSo-A6f&~8TL#`>ILeTDY9t`i
zIrE<og@gq3mR*fUyz3}W(7f4+=f$I4@$E+>%|Z|aBd@0-7-6R?%Sf9AqmBRCW{4zj
zc7Zn!BdYcN$Wr?Rd@m56WPq1`F&aFJRpv5w!e9o>sLPbY6a+F67S;wEb5t;2z5=f|
za%mRB%ZgwuK8;`Fx40J0)>*xAmJN-37rwu6<`DWMm)hyk8X93G&8CrBDtfevvenY1
z2-aFc!FX?PhgdI*-eh|q|93$dG_ax{3Wd9BZQv1yVCsiW1Ia*t6s9QVP~o6WGPPuN
zCvD@RN{9VW6vXaQ?3W}N+6#{{x2-W;Y|!9Gsim;3_L+-_EVIvgL<3%?@p!$7wavy>
zu1%K07n@$#+}T!Gzo5Cr#`Ox~b^dG9Ftqc|iTT7SLjts|9c<FTPSEKIrYQc{#DZ+3
z{VaJnf2@)@Wj>BgPK`JC%@EQ{Rl3<D(r7KZ!3?!$RS!`Gzymo46;?2~*xm)RlLr$x
z>`t2MMvO{f3xkweJ%{MoY#&26MeAz!>z}EgnVMbb=;U3v*wBPO_=o%b@DUFnQx?bJ
zR`G<I3ayiKW#QQ%Kg+XFbNH;RYFbe9o|Cht{N7}O-=tH#;`%ol{C~iSsoL?~f)kDL
zsO~qQQs&PRxArWa{<LhYP2rY<Zw_v~t7(~b*Y=2`P2gQEHsX;XqI=K2Nnuv{1E`p*
zldhY{A^&~XzXLa``RaJYIp5+;Tv}Y~4Y?DTdWSHp@`vYRhk<}EwND6_N&McnjKCs{
z(0^4P!+m*GaBsiCoL}d#Bzd3X>0vL|;--_}Q?YtEK?vv&1vD1#q^X1-94h({T^LL{
zb1ff{r=&G%Y_+FOk9t+@mPQ=Q;Z8eoEDiQxhwW)}rC!j!v00dy_pnO|EKqTTFT~N~
z4jV0PaGrJK5<d1FMEp5@_}Es1bek`5;A!S;Z^+FWV3i(VJ}7yj9=NAgNN=kXHlB$h
zPk|>KZh6>HIy~V>i>SA6mL<+X25efEh>IA4VuOLSoH4dx!rSa6wGT7X(~!*r{{+of
zg;8l_q^}S8Q6Z02N2=Lb<|IbKs|SuAKoGgk14pm!G+k_(FU}!v2BLd37gMD%>;%dE
zta9+JWAIbp#+%0Hfj7lf>wT=wVTJ4$_jmk$^49pRgP`u4C4LjEzbfqTb?W90cKFS(
zeT=gR>?dt(<tuO6`*JHVH*uBbwUbS-tm7+NbQ9svU)jSq$Jw=-^UfLA3ZzEN*Hv|^
zsg`zdu(`G)uWf8^Zs;>(dtWz_#;djo>HnAl{(MCEDK+JHI#ltwgM9(h5}S}`X4~44
zy+E3Qr#VrBG-}kDBp^6Beczzx?k)TeIX-wX#Vhl1$cY;px3F+sYlnDs=jO(s-ri>T
zA;<2qB(rP4F;zHG<0#UQBx=3=61(VAD?bdlZpTi8dDj)txiGY`A&&)AHn4IT;y|rK
z#~zJGY<Z}W6P*p{ELY0iglvzZYi!p!HZ^sOj|Za_84QV)uT>2~)zTKz(2Krky@0pI
zU=YJ#ACIoBtuY+0$0>H53u`@tRjD+IRjns>X#w&h%uK}~^awoFoZD0iwYk`0$UVpT
z8W%WU<E*`FyUrTdHv}}_ZbvWl12hS18p~y{?BO#jjo<>XL$e|^iPMAacI~>{gMjX$
z$`o+33n@}2ssVK%+}<*9bGtWwsscUXo)JVSFX4VVdWiGVjNHLNI2;o;aPg^{GFq^+
z^uY0V)(ZTQSF2-JEj<ax3M14eoh`!gj>s7vemckrc8*e=@6^sZ|9N=&exvDmN&=rT
zaz<D<r6m(KnmA-&{Qay-a@+1YfU61&JwH*Gr)~EmJp86aJM86YfBoyqEzp;TL>x_K
zLEEhHVr{pw#>>$Fp>;y@<YnMEd&KwCjE*D=Z@V6b8LZH$U}M*F&%tv<=`6rB7QX(r
z*t93K_Mn&AvTzA|nVq$sk1K}Z&wmOSt>1k&FYA0N??<fi*z+6D)5OMiX(AU;qYwCk
zSB_)f=O#UZ=X*rqXiipYFuljO3IS}X1oeZgfvbD93hun{G*;{EziT7Q_RCA>Yojo2
zfz`c$)ccx+6#xS`z-poa1DL@{ubG)yj?}$%U1?vvXuf=Ke7yG{+&jE2eS?F2aOlvL
zgGcvGbichvReO5+y;yhkhQnSgvwHh`zPKkx8dr=AnH2$0?{xjhE@KxY>LM%>XiP`c
zbXitihtML+?wURxMD+TEeO_~QATE#W<42Ulyj>2s;;!%5<%p{lYf-XKR9s(mksF@t
zt|{eP*d@(nLX5)ywabdjpjEr9y6&J!yX>y%<8ft_ar?aH8l%tKWxwko`e(ZwaE&N0
z*yV_8ke;^7QNI42g?u@2=CWxOR?Ngq;?9z3ChjvAS97V-4o%6Kk?E0%?fdah?TgOv
zC6jj7`<KkbWpjSWN~{#Da-vu$@{JPNbRnOx%yJ@?&m_u)LT))*9(m9#Sy`N)ot+#R
z8$ZEO%BK^<bq}>ByPQmj$2qc8E*GtnqoWJBdrH32NV>4{)<SW;lwDjZCx+8U615kA
z&{4opm0byJO}~s(uomhoI3@t%1X@cdbL~ESFJhBp4#6W!P5vs0YmQ*U(+JR}nd>>v
zTK}Xom8VQ@y6XL4ApGq7Uh|kLER<L9ewKLOB1*-|8ztZ@nHI1}pxqQbSS`umH)WhJ
z;LVrOr;Mk65bY)O%G&*B;rud*^J5U=PT>6}ah$+0fu&dHnYY4st?nu((Yp$rBY5&M
zK1JN+BwXD_AsEVWO4W_hh<N4|5EZ}6I-V+vQR3h7z0(Nz$z#}fgh6%e!-K!qe~5S<
zjsCt5W}uKtZe)`UfmS@o!J?Cwe1uKz$lMkp12HMET^(`l`U$zjs0}${VuT1tgxIRo
zMcvdx2dS45bcp(>pAORi4bl(|(-AsKx6m<4(g=+ry5~4e&?HUKG|kW~9j6m?lHNk6
z=rr9*x6$o%hVG!Z(%a}P-AQ-R9NkUl=<Re5{T$s(@1XPaPP&gS(EaoPJxK4Ohv;E?
zgnphLrN`)Tx=5EOMe~%V44Je*i?l>pdV-cIM=O-40wQEQNhPwVOslj;m+1<v(^X{M
z{RJeCdk?)AF`3>+@298f8G4pJKp&(J(R1`+`Uw3JJx?E{Uq%GdU!jlDuhOs43-s&s
z8}xCyPM@Gp(x>Rt^qcfs*sJ{+`fd6w{SN&uy-1&<m+15Kd-MhReflE(0euPa5xz_>
z(;w0w(I3-S=&SS!{Rw@I{*=B>e@5S+KSz#{Z_;1Tw-Bf9ujsGoZ|K|fxAb@P_w*h5
z2l_7kW5B$U&ZSmTWv~l<DP=8{Q;UI2AqUoyw6cr&M)39JOv_4^mr{And~!9Fvx18y
zGgUTA;#@QE3M~|J8T>w-D_CYm%NNQSbHTG@rb=bERZf-MtCm^vpf$a0l+1KFm0!%6
z$z0YdyBBh)Melr}u)LBgEqk&nsYNqXH1o-FA!*^^N@g&fEv3=>3U_cS=_ND09JX7Y
zTs19zf7M(!efUpWIRGM*PUXye25`uv)-4^(aK+4*HTk3g`+*C%l`p-7=S<G0N?yB@
z<>QA=E~Rn{o}9T-$OqCTGm|YR)2UL%kKUPNrj%VXBLHD0SuAAp<)q-U;hGaUbD<on
zsjn7mTQh~rd3PGzIk1o`6pQQ0Ox9Y~<Qsd<{F;fgrl**)%BC;J419hyH}9S|b2;x)
z!766Usa!zzP8M^kmL@^rUr43#`YYL-DT@mw+{Y`<u9!>SWpjPLfXDYbunE{8Eap<{
zLHl|u+5D>IR0N=-7T(!R;$kcIpfEV`a8}g*4FF@7ll)TQx-+ZAmb$utl6yfsAHIMB
zdDHThmsVHit>kJkV3+tps<oPTuM`UT`g<@~{q|F53T5y7YBrb2<`=zu(PS~R;1h+*
z1&ov^fk4b$Shf}}n<X(^vPocDKr~%N=zg;#Pp4Kh*@7&tWiw_$UNK$Bm>7x$6GNR}
z3=~u757?*jrnazJo-gEldGoR*PWySBBo_L3jKy0=C2;_Uij1tdMzApd1c)KISSqX*
z4ZE_M7u5i02qRg}rkB^5aPp?HR9G!70<$fFn7&HY5J)zkN-rlDn4-8U5zk6zEvrzn
z0-{GQdsU(@aMmh<6oXal#H-7S2?X+&%tc(t1kP89h_TFY&?YHxq*@iUS*;4BQCCt$
zeFdBb6z!00F$G58;G9)16oV|UByEAk$BzJ1TQ)-u1tgh3gQ-lW#4L;1M5V&C=v<1~
zi#MAA4QCcj9R+R-U(V)%NXeDdmF!CPs#&RL^P+Cp?YMNgP%?d1zHk}bXxa3YOW+^4
zMqU#bV)@d!Y*7%7b3;Z@r_31Xyr39qpN*cYMRn0!34(u^>&f{nP+~k><$g9lTM{pp
zwU!D+Kj4P9TxQg;i+Lt}`+PCQ_Y_^g@yxt^x&)v&tYBd^mkZiV4AfY*!c~Tl#Op3g
zt^oF1T9?I4_UhI3py;)L5neJg@+X3ENa%~k#S%b~HhoSdZ&XUSaK@~u&YP)~wrVR^
zlh$QSFp=s>F*R5<C%`4ZMPt<-@<g(XImCVoJVG86QC}DBHP<N@@{6lxym}pZg4q_X
z-@fHM1EuEKxhmgr8Kff{xRk8El&EsQq<9Gl#$0N3aY;f>6h;52xSOoY``7lU-Wi-Z
zIS)p%EPqhK><tu1VkWcX$XF?0bZI>#uE!OA>)N5^r;AkPEC8ug01?UVtEFNoYuWQb
KISZn8x&9A2X%@x+

literal 0
HcmV?d00001

diff --git a/_static/vendor/fontawesome/webfonts/fa-regular-400.woff2 b/_static/vendor/fontawesome/webfonts/fa-regular-400.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..18400d7fad27fc52cfbabbda495b871bd912d045
GIT binary patch
literal 25392
zcmV)yK$5?APew8T0RR910Anx!3IG5A0RDjh0Al0=1qA>A00000000000000000000
z00001HUcCBAO>IqhEM>n0Lp{92+M;i1&9R)AO%2wWkK|$7WWWQQR`t5sYyKms%n`K
zZ?75<zx?*cU;q5~7lCmTAL~5dJUmbLR_9hY^E10?Ijm{`zs)qHf&Ws@N`502gosP8
z?SK*w0m0HI19~3n{R97OlIi0~rYBSq9bCng<PW$a-`)Fg`ZlwUwFYUWwWN{5j%7>1
zKCeLv{I)4^{$Y-|<=o3Z;3BxhK5}zh1Rrq`QsP6G?f;*aX@Bo4c!lp3zXml{tE)AV
zMpY_xm!uiVJ>x--WA}_nhS+joGiEI~83#M;u$(1Q*k`|UGE3Y6|72N9tOcUd|KZgB
z=MG9E%?K<@ro@gNOAbpWrNn8wkR|AD3JcVeM_r(HA5lJcKjsnl_ftRT5%&?17amOK
z`<!1OxDo~M6Hwq1_F49QTB}Y~wO5@0hq*It1QUUF0{{P)rrha`m!I}7N;^EAz|dtW
z%<s(zye|pN>@H~4_9#_aSFJK~s|}^F%&gqn?2<K0>Bu>}Hx^_7ld#ZWnqU$FV65Iu
z`}{AJtImRTLi<vfLP*1@eAUKlVRX;H=n+N_L1O@}(F2?qaQF4=p2qad02qK=F3431
z<SO-ro*}*f2pG~J_Y{T0^MoZ=3UW^fX`Nx=^cT|O_s%iD|KqH4&o<UP*Ek!am9t)@
zfrW_Pr-jUw+<ogP21#Jr<~7EwYWv4>D5y-!WOc+0>RSDxZXI<B5EA?`W?%Q`$g}0d
zCj#|HP}k7l*FT)me&C7o?^DVC13>4WrhNVBkNEPz@{`#!;u~m2a_}I&QGSxFRz6rh
zm^~w}@)!L{=uJOqy2o`I&oas**T)ayg&K%;aU3h%`?}8%PbJe(m*ry7>Lq4z&(`&N
zwU=c*I^r8vmGWMSMe$y=5j9kaUZScJH4(M);po*GRYLi2RLh5=T0RoB@-eqo=|k^Y
z%cHI}YVn!Y|CJ#AK-9{IXxwz^-ZW#$C)*Y`c`q%e^U6mzey=MZ$-=73hhQHqPceOj
ztZG2?`ZMz7__4khR?Ek;TKP~KzkEF0p{uWzkLFifKDasQQ}vmw_dOiHH?7fIKAP3y
zQtrvBNTr`jH7panD6jkC<ztbr8z17$c=(<4wsdpb^dV@qK67i!^5yrYUTtZl-a4XP
zYh`#=*Kixc>1J!m_?~IwE7v<)rIYZ+-ty6W#-p!l%SZCaR?BAD9F(Q%>-6Eod-uu*
z!xp55c0$9iuNn}GdeouI-_|@`-N*~N)VVh7e_31K+)8{WX!0HHx{Wi(ZigIp#8Jl_
zcfv`hoOZ@p=bU%JMVDN5#Z}k4!HsTmvs=8&`#VQ>-qZPb_Zz#v-~CAUqq)3KJ|dr!
z@6Pw+d-Hwy{`^3GEPr?Yp8Rt~Lvc+}DO!pt#j;{Wv9;J%Y%h9>qs7a{4I4LnT-Ufw
ze|*{>A5V3uAvLC^)ISYOgVNA6ER9GVsVhxP3)8Z+JgrE((%!T$ok*wB`E)T|N!O>_
zQY}3_JtI9cJu5vsy)?Z%y&}CXy)nHx^{~#KvFGh``^LVtpKQQ>w?FJp`(L6-42dmq
zB%vghB$7(fOGe2hrKPNtlZsMRYDyhxENx`0OqJ;}Q|8JRxgb~MzC4ge@=89+cljZ|
z<hMrA3R+RCXmzcrjkJmO*8VzB2k9uCq?2`uPSu$@M;Gf}-KPiipdQl0dPI-vaXqQ0
z^rqg{$ND7@8b}{#6j&3u;YIh-cm=&;UJ0+1SH>&nHTH&hTfB$fTR*d(!_VuN@ay{R
z{DJ;tf3d&D-|g@BkNYS6^T8mHf>dNAJB4XMOIp#IHngQ3J^En6-FLkQHpr~DXY9HA
z`*{EHqpC=CsUwY~wTv<6%y=LtE-3fpYHJhiq5XBheeZts2|cCPAFBF~HNYR{&-GXP
zyZ);Op9h}>9|i9P@9>f!>87N)AbM7ao&_L!Qu<ek9y@xh=uylc^WD^#*SQ?TTsPOu
z6?55~GpEfllV^^A+2xDgWHx|VYbKcSrbm2Td((n0zlo_2rmVlQxG7erQ!^P%0uvuh
z923jL1QW%Cx#dN=55Ya|ZgDrd%fkjf*PZT;0e2L*z1)1@<_0&ro6XJYCIUAe<<k+t
zjnLwQIM2a(3eFR79-Fhwc>vC3a85b9oMp~Hr;-yI?OwD65UqPOJv`fTrsZ_Ysg{#1
z+i0n;{DN}{EBh+zEAuO}E0ZdtE5j;-D}5@hmCDLTUKVovVt>YE9og7S)|R!#f8yTw
zNBk|TkNe|y@vm%VHaDA_P0S`|ljFnji}+dG9lwch#Qkx99L$=t#yA>>V?(Tqm&TT?
zZ`K+gjpr`dPT{S@lZn5@2Z=w$--3r?zizxbUQV2j(=j{-Gtu$Xcq*}s*dM)B9lN6g
zu`rgnb^nTSOtFF(ALHZm2ghhX^CjYTG2M&(xU***je|sgXphlxEm061s9*OOP7Jjb
z#b9EIpDjL5IiBFOvp!2S6OBYYk)^he$BfNR?RSCta@S|*8z!1)v^Lw*AyfBz`F}}$
z^wm#)0}M3CU_%Tw%y1)&G)kLx9XfUCmMe@l##rNwH^D@cOf}7Pv&=Tnd<!hL#8S(x
zu*y1{Y_`pIJM`FPkA3z#;Gj8L&uQ=X`c}88d5Wibw&!@E7ki19dbwA4rB`{4H+ZAB
zc&oR0yLWlF_js@OdA|?%pbu#oeRf?7yTx*}zxNDuqI-$)#*D={G+;93pb7J_8+~vH
zW*`omK{#Rt<ER;eV`eChn_)O%hU26efm3EAPMc9UW7=@mwBwxVz<JY&3#JPfO*bx?
z9G6XjD`qsVnlZTEjKvLR9Bwq@ag&*Vo6SVrVkY5T<~F?l=I78k8tkF-9`F=89|x!D
zek0gK_xC{$-H&jbeH6Y;&GU1-^z%L&I6yuEtRtTTc9HJ}?;+m<-cP<4e1Lo(csKce
z@Luu*;C<xBz*O>ggC6qtu-|?T;AV;jaDd_(u$iI)o<`9Eo=q_YG*B!9Z=hJguv-E8
zQ*2|{?HK%d+b?>+)*c0zO7Sw#$hcwPgNz%_uw4K>jN1gJ@yDlu$^7wgFojeXV>79q
z)KH$E8Udz}n!t2YfAB`qKyZvS2+SZ21#ck@18*md0Q*QCU?!;x%py$$vq=lV9MUo{
zm$V!lB&|r`0I7$x3-d^O4O_bpU_R*tXe6Bi4W#p6A?YGmM7jd1r0Ww{OuChH8<vo2
z1}q~z9V{pHFb-%3c0AA?>;#}J*on+{I~zbbuycXxVCOMk>=FRw!7c|XfL)n@+F)0M
zT?bSId&VT7D%f*i&jZ!KJ~!O=_6@jg_AR&#@gKo{0&0Q{fXlGo0kj4C1Ly+wCr}6M
zf1oZ%G@u?x4EEdD07`<yNkA!(_#g>^`XGr7&<G?6kPjpkP!=RT5E~@p;ICWxk_$j%
zkkUXCkg`BikaBEoMa0kyqzXt?IV+IrAT{R{U+Qczgn={yX*>oBq%}yJd80tafQ+3h
ze3@#2H6YVtSPe1*Waj3}Trj`v7R1m6<OIltxx<$$0Q!O42l|6NV849?=CyqVU;xNR
zU?9kMU=YX;w)Pi*!63hZIG|AyFa)$5Xa!&>Xhj1I2dx5(0Id#;0<GB#BSC9}Hj1-N
zkbp^`JwST{lR^8NUYHI#0Cb?Wg8=LR9R-vJodnDToy@RP0L%iN%CIv5go4fi!htSM
zKoQX0pnHMYp!*Ck7xVxy5A-0z9s)2Q^f1F70k8n{D8n8{0&;_%06hsT1U+REun6=z
z=uKcT=xqaR=z;&g1IPLQFK~jY1~^I8JjN-i=~T0Hv0t49I8XI4xIn25E>fC+J(Ql{
z5~UxwOc?;KPzHgklp)|6Wf-_lnGJ4G)`KaOtKcT(8*q#AGuT7626f2}pdQ&3)FB6e
zTI5u4o1Fh4aEDw?uEAY$Be@gz$$jJzyiT4W&*DAuHhBjhkPi&_h<pJ)CSL)ckZ%#6
zimEYXk*G#dhKcHX%2HANM42M0Lg^;T07_p`#!%*qGM2JHlsS~nqU@%e7v&(OqbP^O
z>S0Kn9)Var3Cbi<u261@a*I+^l-rcyqCBIF6XhjkpeV0U28r?-Wv(c%Q)Y|uhCts1
z<&-G@Qr3v9Pnj&T0cEYohLr9i8&Q^s>`a*^vWq|`fzn0f7Rm&X`zTvQo?*Utc?sM$
zFYkadLgYirJduwnt3*EM_Ivpf3@_gW!^;mK>Gg9^j*9$(azx~BlmQ}tr}q$m-2ngp
zHU($`c7X!i2-rmm@Mpj-qW~SiZh`{b1lS#*05=16lN8_<z^+CCZUgM5DZtf$-3$e|
z6R>MgfLj5(76teyVArDnzX$B*DZn2AyMq+q4}jfa3h-;d?g$0=Ens()0{jNBTc7~f
z0CtNM;GclqfC5|(*q!0A<aK930d4^7&QXAW0Cu;d0Cxj+YZTx*z-~kV{s!3HnF6c=
zc6X%!e+PVHlV!zw<4K5LpMu2c(-5oAK&(Cs3UDRh8_$W==OI>KfLMJIV)Z2moxTkA
zdv6)w@#(!KAXbM!0geNF%clUh1HLsy0qy{N>q$wYui}gc_zWZg`W-wHJD`C94h%yB
zAdYz)D(ThyTBvyJg`QMA_G`YU;)y6$DvBo*dE`l@ag9%?APhoTSzljYAARn*=hoMo
z>+9?5AM(D1y}zR&94t7(Ks1EY+jNA1L+y6E-Me<W-M+KkZsWe~JKJrl%%?B!Kem7Y
z^fzG{Tabk@06PdZE){uEEECmOsi;v81HZ^M*YT(}&xjZ!VtJ8Hr_;i9Cp?iDquyp0
z7J93Z?IN9quNEZjKaL&z0qhA|0c_%2)l{q^&sIU6#(h<V76x&wstCd;uBkgj9_KMX
zZT~ixnMYWh+=SSJ7^AJD`x9gIVadr&m`9u@-!u00S+zK6qobooj2?lPYqZY?-GoN_
zFe27uMC^ku<FwBu9MPA=`ORjt34oC8cd(8f_ypIYTvhZnu5rzySFQD6Z^SjOg+UmE
zL2SNKRj*oed0f~~Q5?k)O<m|Kx5nXZ9(fe|8aHlaL@YT_8lOmOx<SrhKF{i7#y~PL
zFg_9G@_FfIx$TY&I|-KPowbb))(rAL0ZCJij*gAzT|VHBP2&bJMirKGrpNPnL+2??
zx6-MUB*WkdO-pgd&7?U`V3imng&Em69Lh~+85>Ay372>(slKN)0L1U%_aF;-C<9;=
z1{{^<QH)6rC~wuAF$VPQ%R)%m>j)twF8BKA<?|l8ri5#v1Edt9*O5{PTsDVIT{nj@
zM<38H0PMfr-^Amw1r-3BJXguuu{=dw3q$29jR#>vzJ8mb3gXD~R6|8^huATWk&!2y
zATx$@45P4L1-2Mnx-}ym;{zWs9GTgADX}Ae_xy!c|6h0|Hep;&N_?ep@@h&3ev+Ds
z;vn=@<4s1yTErN&d~?IE2Qu6-6}ySc*a8=30c@=ikVl?-hz`OaY{VK@O5?6H22mWv
z4Jj)HHaEBHu5)wDeeYIfynt93uhbxrWRJWKw}yrB%8}HJ4DdS|@P{x8VAsQd^L}rP
zm29uAwero87wCq5piB#U*A-%n4m^+;qZK?+aE&h;j?7H;uF)?Zem(#6TkRZf2Ef*M
zkNHtNzOtJ3U=7K!JeE%Boaxb$ltPqdcHmnkLx!{MdsNhb|7B^(>^&ohL&G7*H5@4k
zA*DZqES%i3{q!Zp7Yj4{hmpx}c&Lr(cj}>woMyf3lwFZLjceRj#V`n^sE)kaxK52}
zM|-`7zzD(8wCuT)^W^|BD9?XvEI%kPm2<O9PJFE7<q`8<>GOoS-x0$T>vJk+U{Ic)
zF9&FuLj!I$l^Yl%=0thI%$~uhu6KT~zlld;8^Tu>20WoOk80AEQPLO{-<Gx=y{Ty_
zLjng)4w{U<^E>z*M^D?mf9YAhuRi{PjEG%8jL`+J0&mWV$D+k2@II(msgShqB*y4g
z?*zsi<x?Y`rzd&N%_ha@r_9lQl;s=?c&d$zh@E)ahY*7)SLeCZkXa+0%HB9euD$}C
zwsH`vws&%+xg(-6jA|X(EeL}!il?LsXklTzQY|D0$Hfxg<$TN(;nXQs&eHc95uLs^
z9uA^8F30gzv^@JWOSf_#{VHOQ{=>Knw!o34pz0A%=%B1EC9#&^n)WZl$z>jXP70!Q
zUGJ^s(uRYF_H{bZwwM+aVORJZ03Ov}DX>&>8-?FmZ(0+t`Hhu%CMJrbxTY$qqFiG%
zDoUw}@`FP}E{EPstf22VCL6wwY--rD(DxgAHhdo~Yj}!%+p??zxri<m_VhGblat{o
z92H9q-$(TQ#$rhhj}${H6-S0;Y4O|WQX$<|3*=(u>8>U8A=%@;ZL`T!+Vp6#iW)%}
zgc{elr}^!WgQMzaMx|>PBEtU0cknnoE)R((RdSwr?aL|%>v35{9)6eSdXHc^m&&@m
zD|Iw*#no}$)z3OR>*i7$nko23D)$|gC<|8bN~}N@Jgrwp-ozX+`-JP~q3Yq9Qk$Z=
z)nG-1X0Cj&4y%6$kF4X>w&tCMnn!V2MR7eWtAHm`IV|@c$)&O`uIOm=u5`1zJ8)t0
z4SqCWVD&e#f-R`S9x(5SsCceob%tvkt!rdIYRo#Wgf1hB?M3AwmXW+-q399fc-<f6
z(V{UZ1eQxtAu5-~efHBU%Jy_xV2Xi~S4;*}9Gqsv7%`bC=7xsGl2N&YSSm+Mnp$Tu
zdRPX~?=Zj)><u3Vz*$mOZK4W9c;pzGQC&50>?kHLipg`8s~nGN)pkzlEp+-0t4m9a
zEiJ7QW3=bOVzG!=EG{nW=`R`@3e}Jw)#CbGn=NpIn4#OYZm^}L)jiQKm3(vsO}4bO
zn(2~6EGxc8kAy=*$jW7g#bRxjd?<sARz_^yFm!ulb!my+IpsQzJLQT10M>t}{|Fwp
zhDeXBP<Yn(LyTipQ=XzY3KO<p^JC@RuUVG1!5ueK+$?5Xhi~kD<Bd0d{Ns40Koq?-
zH^ZH+TOFQpvDtg#;fFz21>>M#XNi{eh#bpw6idnGo?r7D)B9XCuDVX8F`u_gMAOQj
z+VFj}?eUUADLM4eZQHgXKU7_GZrir)9MutZDo1RgY2`OIGWgbfzfm3?GySOH`}n6O
zS2#Xz&T&V5tN$RLfLFjI00zENBD+L2j#l<8FT7$uU9N?=8tt}QRINt4?G{;5h}rr1
z`T5ydA%vLqpEW{A3%Atk(eArP^|}x;H)EQnIWxDgu&^*YYYJh`N-4xVmVYF+K*F3(
zn2MZ9r{JSd)U~VSUkf?U5&41ps48Pswa<a`Xj=ILcdXV%#}LOxYt<bOWXd7ekoT_S
z$f|0eeN>gDBACXZ&zV*pS9G29>9ptiQ&v1>D~;+XUf-s=?y&zN#^_%hZzRU(M&w~-
zeZ(^E@e3w;<zs&ifb#FlAIsWiETyiK5wX&|)}0aKF6Y(`wxH^+oZU~Kw)%-~YFtyW
zG-NhC^!CrNYo(Goml&hVWmd=Z4m*6b<14IgZyHvvvKy{_`8qf=y)z?Xx}N*U!XoSc
z7h$oLGmTm72R)Meb5M30ZU?x6E<Lf9mL1bLfh&Aa)7yND!b0ayfBF+KMlX0ky+G4x
zk4>_F+K*hf!a&)YlD}+1xvnKIPmIwT6U~O!x_2TvmfBux$NK4R4z=G|0KjB}oeF!G
z(2o(Niej83W|mhkT1*UBP47)I7S-w!^=z0L&k-NOJ@5*cNHp9tCLQBj5UN;3jwRb*
z&fZtfF>Nbz(fLA1`FyBxYX3Tt7^APQ+$*!D{qXrx3i15?+@CIBoaU5BR^}KHK`2!e
z&l7v_6z5LLwHWnI?3HS4`a?PAdjBDm0PM)Km93)0r(zIF#g*n4!%*E95yetwuDv!R
zt<yGl2{#OGS>i3cd_eSGM=1t0IiN+j{AxD)YOk};%cK@OhDYKL5kH7LRI=|mVVXU}
zeFc`8#h~?^X=Zy5ujS+A|4?MB`~=1-G3A~^WfLDRzi%J44#lge86H3lh5&xoeKN@l
zQ)-bU<&u6Z)Q4w_yM%j5zWPy)(!roosZ<7obnW9GVWYV=JauZg)@<PCkFr-03nxyV
zJh31u%SXAnmFKNyvz6ytw~!~e`T*Q1eSUe7VKyqHJkBl3>?(zs-1^?$-qH&=_R<Zc
z+Kn4G&YV~zWbwqAJ>(k59yXdwOU(v;h2k-g#S>@FoLD5toyw6=vo|ok?nm*<{nWV*
zQ0Q0JB?%jDQ>7|Z<VdwjBaCCDWHq}oT71xUJpSMb{9Xgo^-Fl;0Ld7eBb)ntH2YpU
z_k_=F=ZW+jIn4PZ$unfV#{=7EJZ|$b57qrPUc}#Nk3PHCRH_5V!R8@QKJBO9z<k$y
z?a4+1G3jqkZ%n6Cq|^Co6|q{K*9gIz-87qr{<Ldy|3g-<mu>bFJbICCcj)<Q6#?-0
ze~*6~&*5j_VYMvzTEnVH`B1Vq&D5E8y|^7cCfS;kyItE&(je&<B0CPZ1>WWZC^DM$
z>4QC0QdJ(GLafi#MTJ=^+ZbI8s?`uRB~<iCl4T`~B4@k5`K6oWcfq#RN`-BVE?Oe8
zLzzZq7@4xX-N)DRj<9SqWJq;`^SQa+mIzHr)2M`c6ex%UDW#I`e!9fa`zgy?A~cCC
z%QOlkNtQ{aqPrNbQ(YHYILj?ZZjLkAJ)~bmY{t84yg`VewzTcCLip_Bbs2s~R3btK
z3tL+Y1I+JEMx$%jMkD0YzkH}|u(0)))Wbj9N26<hxi8KU>feo@^2II=UP7fYrL14@
z(D>*_Rp-Dc=k!Gy;CIoW{1iSIP$Mk&7$U>{&}eT@|9gBy&fzC9b?!yfZXjhU?Y92`
zJE8H0SF5h8zWNPD7%2PHb<{^6RUKD-K?(4?m7x4Vyv;#$${)fjfntQ^9!GTW*77GE
zbau@-mHxI?Ot*5+3N%ZyK2?Rt(y$ck92?(gbDw|HH*wFgzx~^7$6N0UKId`U`M&RS
zZ0`N&k9yqZ-+y!VbIwt_0Px!nahkK=#EZ;#txu3BBA>8apMB#u;y1Y#^Zz(I?Ec3-
zv)o_!g}!RXci{mSNS$GIidiR1Yrw3NNsKw+R5i+9!8tzcd6QGsz?Je(0;)EfS*2dv
z4cg4a$1)0!fcC#3hq1bVpq(G;UYSL4mSp`nrs`@8kwtl)MMBOLgq=<&%%Z)D!*Sj7
z5IwJs)3vwiyB5RdyBy^2GU&49evD(eiwET{9w0BL9Pt1>uU^icJP493hGUL&*+Sml
z-M`5Z`OTSOe{=Pj!xF&nghhB2fJj7=WEY<RR{p3M6={|V2-k|sQsSpcmZez|?67_<
z3A0FutjO}<=ss0xnxz@dQkv?Hf4e`HB<XW2h@JLudzG20rl_mi!*&O8<#Uqly0(p;
zSE<zEC_-$1?sIL#D2i*9isvCU3>pwZKCyg88VwTDjCGr;iX0o}U@(%-EPsMn7T$_M
zplJ%N*XE_A)+avET9W2#b*gAu5Fn*MAUysFvWPPnz(oLzgvdk|iHOon(BN3od#@;t
z>Zy<xRNYQ#g-FFVyyZVal3<(DbKlAp*?6W#EX#i1`)tc1wPy@j;oo|W^5Jl(*ss4=
z!|-v_d@LaPy|1^GA<oeLP$V&CJpCQa(CFpMRM**eJZ&&0MIW;9&L^LI(v;kLbnC{t
zs;=L#^m|;%d=daa1~{JK3};|L0z<exu?tm0ym-yRISQyV45+G%Mi@$+PNyH9lzY7d
z0afEp)=m0DOfeY@68zUkIdt7-v)gSpU3cwuT)~ISkK@By+*^HUHSTE{S<SCJbR}OU
zx@MPewly8c%atgqEH57+ZZtgC^%{+M>3?}MYqbztt<1ZUuM)DFU$He^*KDsZnaWGm
zU<j81<^zfMh~b(ac0`XO3EBE{X|a^5v<Qb`Aiz?b%}2t+xq6lrMV8g)*1krWVyAn0
z>(*PhPIo)_`KS6zbJu64cMpjoGTT^LY0S+vR#rBak8ojgv)AqRHa9Qa4)y%=&p&U|
zvFZ72<4qCC(^b3f`ZdM_;IIIv;WFF{uZGvd`@o0$oeat5uR{?*#Q^*H>fjc{-K0Ov
zb&(_*(-+<mB;6#-dN0+&G-!i|7;{99Gmgm1@fhk-pr<8nPO)2ut-DU5`t11pBYSD%
zHFTY|pB&&G@^V)nb{38~I=fL2cE)(z0G!{xNwm9?88;n$9%pkQoZ$?X6C+sLz1%Eh
zJ7QGyKGmAMxa;DIVL^+gW?0tKSi55|K6U<bH$Y^ztM~f-xHxssqmMqikepavUaSoB
zE0gmz(O^*iQI^>qx7P%Gu3NM95G$UhQ^oUYES$TeTCEy>h{XM8-LLk1@9#2T`I8)y
zUxZY&7-lX4N!Oh78PXJ>SO3HXv{~>D#@<mL5ya|k+xoU|vuq1-KVUYd<r^N-^@s3%
zDKFe}ml##)(@#^CGW-#)EpEUCUo8_}HM=^5{_WE&b9=khndf4!BdXBQ(Z6AE#!-2m
zS{A*lsaY1CD9CYIPKkK0aylu!;#_&e=uKUJa?vpMR2*|m@$-Il&%;?|SK#EL|BF8X
zHZ))Zo&Z3RObj5KUKO;Ug`5Jc=ZXU8E%k(u#S{e`dK*P=a7dyc=O{N@D@J)fb;9vS
zEz4r8akAmMfAOC8yyxV56m`=@<r6Qz{IabyWW!*u@$8qw(25$3sQeSNQblLJS`CnT
zS*Bb5qib50=>{T8?BTmD%W~bu$p&LDH+urm-Bgvoc=_d*nfn@M=u$(mUyd4$$O^**
ztJV4PA59RNL6bYsaZM|u(qQfIy#Tlhs8Q+w;+ifnnIqFpWf5V<cP%7s+LUS6J%WIq
zcke1=D=Un-?#hbmuJ1wI$L?L=)ppRno@1uv;<2~z)%isf6~)2QGzIz0i@a*bt~)(M
zI=p8-GdyN&3g^9XT(ytD*8t3~`BhM2CQMF*=oCbXpcjYkCIti8x!IP~=ZN}C-LRj;
zS(k=dsoju~Ql;45-=i|-Z`M5*vDH5N@79urY*Qsh7^owBPlteiuO4&_13kCi^Sy|W
z$Y!)A>kJhVsCr0#oLcCAGKY@K<0Eow4tmyQxDM~E;h@5VstqbQLSht?enQhI=~z*W
zt3<gR<C|hrN(L6m(l`oLs?u%}WNA&c!{H_O6vTEm7_=IJZ<>D4XblG0eW&vCy)@n0
zO4D8}grz8!5b@`;nTlcx5r4j}b?+h)^t`&JVI;bu>?w$Q=zGn2qaj42QEz%azW)a5
zbcH|^qT7k%Im?=h<Ilf2{2@l7+Z7RNTHW)IAa+G0C3^uV$NzQwqvXB#WzHG743EKc
ze$+xFX&%jH!Ufosh_0vmj5=+PL?j{^4cC&@B<@B6DiV>5%Fw7tvRDT}yq_duI)MxZ
zWJ9Y(;&aO(6Q5hoV~rp%Ww~R|mM@;@?0xg-_HkK<;d#{&!`}QTi%<T-wQJYt!H!k(
z6DLmePi|Ya>-{bpjX#~eI8C=Mjds@8*XP}J?9Gn?4}aWpd|sxO<MV5+cB|U4V;#4?
z^ZErc4xzAKubrqDFUMoE_p$MKZ1eH_vE2S-r_*spY+j?+IN2MaO_1h=rKKgkvn0!-
z`5vlZ=%UT!a2n3T9dIA)!V~ay@HTi}Cn<rY#%If6fj~3sqdKH{oUc45{DN~?RO4#%
zEDXb87?IA;<8T<hQMJd|mTERF223BI=hdf^6!))B{`j=h-QC^Y{mG$`k0;aJ-Tg7g
z$^P!{?ye9!2YO!JNsmEex{xYQ<o%(+i|gyx$A26oU0**K3<ez6Z}rFr2kYwq1S{S#
z-~m+NPJm)0qIkEeQk4cX^9rMaK9C*dthn_1t5l_FoW^RLW?ivq7jF9x%A14AAbn(g
z9dUjAkw@0ov1g;C*VyYM;wmdxqy2ebPyJZTA*vLqX3COE6(VggjY})(_IA1=Nh|4g
zHsE1BvW*2<uFbEV-&|Rlt0UIuR#rCGSMZo?s00mzs+uI%2u0rF6bclHN}D{Rm|3X)
z#gDVDbXq6|^t}4Jn&IMibvq;<T*fn|pY>y|wM>n7;qHxwi|BdvK^h!8Hen2Kw3cZ-
zsh*s}0~iA+`8vf_ZqiY6j{9OUS+X2v`6(-28ohi(M|UR2X_}^<v8QQ<SMGVnIpnWb
zm0D-zS)n=h*}7lz>(9QdP^FE0=9C^!$+M_%RBCNR-fq`D5B-`iO-X~IA4f+JLk720
zbh;yvicxJOBTSnV^=TNmF{2{vbUIH!DC$O-y@kr%RuJF}Dw^yRI*dqVe?NYu47P$!
z`HvN}rm|Q>&t51Z7z8ccjmUKmWnVHi_O83xH~>h;|9d<m@5C9*!3KOCKru@C!;>03
zhV+MKJ>hOk&5ksehKsAuV<flMs`FkHR(hQPl!-V^I?)D|(i6MGqYf@EUcS7zh`6|T
zIq)1t)k;m43GU!BK^~ea<!3i26_wc$zp8dlog#R%A}HNB%c)|791)k1tQYsFXOlj5
zB926o5Nj2cGRF&6;y9?(YOC4y2b#6+;%dK7NWZ_j+V2xt(j>{GluFXO$0sGysF;kQ
zRF@=~tY%lPR6I^%c_>R#T|%3il{tb)PL3~UnIfvF&d;wbc56*ZQWQd{DM^|nqwCf+
zX}MlQ$feGK>-Ytz!D1D&4NanGR&r!{9WKs$h}Jg*PNQ_M(P%V2zJpidPb*&eHb(DE
zmp(TUA9AF(ojP^OUqxfQ7yZxU-@>opi;%*D@H7AhC}NCFb-pprQ*evXBwPB>zT*g-
z7Z1ZucJK$NfDWd;wle-Wms9OZK{@>9L2*x!S>T6DCr>7Q&P|071$CV*Ih9=zp`<Cy
z50;ltZJw%B4eRYiCpZ7{@Z6x;w6rLYoqm!>njGLBDugH|=l$g5$)(T_n5-~ezdt1@
zMxtvO)yk>OQ_IVN&lF9%fAQwY#ouc+n}a~sqCDw4au8|0U-)_IO{l;sY{PAKG{Wh@
zsOSg*NM7hpGXq+LTF9Vm&Zjp=fMeN4jxz^&Ip&B*WjboA-Ws`%#-;*=n_&g^!ad~e
zHpl(h%*}Itvl*Jx`SQ@u!DZ1l3I3~a2e#oNO#JZmVK&M;*{T?Nt_VY5bsWLrPNvd$
z7!Dk8SI7_~&Cc!0P#F~fDN$T@RE#-l8csD0c{z0?MXsEV$Pq^YJ=m3fwQszy0{4)A
zoRE)m{KwthU5>~LrK6)G(UL#f6*BqVw2?+#61X~!fB;UxhAnsso`ZM8HvnLq$59w&
zz+*`*D>qKzjqCxxOOfagM@4_QR`iGX0|Hc-2^5m?`v?#ZMWIUn_f;B?e8BFzx$j4i
zakQ;e1QhLD32B@!0Q|WQXW#dI-$&Z}jSO_1RghoxQPe{e8e@_Cv7BSO9v(J(wbuyg
z$|yASy!yCuF<l{KCB0Y?%QBU1y~Yk9ZH~y>6-9`b9eD-En&u$q=rCQ!ZZ}T{&6XuA
zE#G#9>k8b&;c!Gg7)+3(=hf?;hnhC<46Ehau0gQ6mGokIF<l|$P2?PrbG)f(HQVEv
ziH^e|{5;OpFwEk(pNtA$DDb8Vs?Hw``$hmzf_asq^1WdJNA*;y1OQ_+q=u3BA6?6M
zb~l#rIS|N&BoG@e=UFJFG4uV9@nI-|3Nfs8aoB=q;4Sb0_znO+9EPJjHw1i@3QWd+
zVVb2G25$j1?k8&8P1G3YL?^9DA;FxZ0xaavD38Sd$}sV)*qpaV1Zbj^tcr~vD9myZ
zB9>C>p~>BN`l{jNqXws!UT1JTD#b7yg7I;pn}v^`u-6z%NYNeFk5$@_yHx#!YFt)n
zJT3um2fwBl<qDh?T@9>0lQn25R};p!5?w7e9t^X*80+~Eek}Bp;qiEHf+zS6$$swV
z!ZO?gFdyf!GVkY^IDsvS`q}p|TeCVclF&!)f<MgS^PW*&qw{Q-#W)qxpZtU8kk4kb
z9qb$nJ}AHp?>X5Veu_M7Q)svMET-JKpSNP>|9rsCG2^NQ7+U#u+s2ZqUle$&g|^@V
z+y&sft}D}P){-KN!V6Rwf8s=wtFVM}lJ%P+hiN)N5e~x!nIftTc&nKPefFVreeGlY
z|6cQN>)wVWnnlFnIN9HCW2@b6VQF}`oZ;9a9Q$un>H775YX(99RD*m?TFDZ~oylY}
z`Ny8`d48{C*tXQhGKU3y+v@avQiV*YA}~i=jx{E<MHADCk5Y)(rjTzVF7#kqk9xgD
z7Uio&k-Z`y<t2%~F%BT`+Q%F}>-d55lGgOY@Zn~!*ZlCtC&;4@Th@moj<0#?rI$D&
zf9Y8vp1r$W%IQI5Vv1k6M-Hdq4tN~E*F=;Pxoeo6zA{J;@<&DPcy?m<vn|xUsl!WV
z=&^-`g<XP1w?RkO*`>LqZPjtbg2Z_gaV{l8!nwlq(s6PB=ve>wwH|ahGF@+BXtU5}
z%nr5ME(9;HUrmbvNZ=gLa1Up20q*BVQVd^YWmX~)32EkK8E!0MMOT6>%RJ38KI3TC
zwro@+W|B>l0}3j2Gjhl^MvHx1h!PUWCHPPxMGHMHXOhl+O>3g}7Hevn&vhx2xgS#H
zvHV<@^DKvDIU?_h2oEq4-7mKnG8~p&Xlq8j-Dr4jt!kQ!8))whZ7iAwR%@=;Xte8w
z_8i(QsApN^9Lp(2qU$(a5up!Z0sJD)U<uB^E6|%_5TRqA&(v#8z)zqB53eQtID|14
z!z{QK&NeDWg&*f}dc@H>V!FN__YfDku2PQsoGKb$MC`@uy3W#88wcfKl4tWvVLTpB
zeZN-oeVpk!^R30jQ`HNFZ`n4YZCieEp?Yd@(ejzD?_#@^R&U$dx~-bF+E^aceBbwL
zn4T5L6ld~^e2u<Dze4{Rhs6F;oVPcUWm%SI|0vU2X2}*==ieDl&F|z{mg^99tV_dw
zM0f=W+sYJzBBO?#JkH%(pB-z<v9G&`vpKPA^U?`lYfIk8%;ndZhkwRjTlhQnHSztw
zNb@IPoeB8P_VdM*sobc@K9=s42S*Roks!`o5sHK_583VcJbU$qS$zOb_V~+I5a3QL
z2r!ni_5Uf=wd))0xr0H_y7`|wJ3A5a@|x?VkB=nUDX;JB?BJtzi>r8^L^mTcMw<lz
zvvIsraahbp?r@xZ$2h|vXlWP(t@5bj4%Ti30cx!vz$rHX;p)QIBQiZKZ$`GnY-`Zc
zq4GbcV{)0BJ;vuNmCEL5%gtcNUVj`G;0n63Gz^C>D^4I|1SLEh_878wvyhjwF<OD4
zy653n#;C+gkSHQkjlv!G^`GPMA5s+p!)V}oCIZuwDl8`*unp_Rj60GOOzT8^3#*AE
zLq=9)p%@7pI07|tlLD(|%p%ZCnoT{z?=u5`Zq5%(*X6c5*BA^MbFR%@w~M-NBZcPK
zx{kBSs|I!5?x7{NZNoq`44X5@^IVEI;du_@Pot2c$XH#Ue&5^ykHXiPY9=wdTjUlK
zX68L^)T93g@Hrz4R6Fi2Ep=n21fi~lQL7b&v~{=KG*FCiK8&JpJ~WhoYs$HaDsH4d
zul*%>5MGBR7Y;hn$v7DW-Z4u#t3osqQC^(r#;jnq7!8*S36!7<+a9*gob0yID{2q<
zOxN>`5#EnK`*sWA9MAk?k2vkV*{<vCM$GG(wm-(HU+lMcr{D^_5}ty$!AIa@@V)Rw
z_*M9`aa3rz><e^9#gMAbZe6VPzChezgn#BN0s1|-d31E-`LA+Ben|>Zq2xHu<bPr0
zV_Y@{C&-V%bKBPO-3`w0&z(!#>T-6MyeID737IAEWz#Xnla)K2j&qjs_7md3M&WLG
zfV)tk6tG_0e_9;yG+jc>x<W`JLT+5I!(p~ojGqrQKVe(@rnzs~mL*{xxu;aRLv35z
zQB}%rG79?76c^VnE)TvMsFXz4nei?Ah@%G6Ih^MQotCh#Ut(LnLW_h4BsYWVufaX8
z;??PNij)pqcjXzjw+1s%g0G%Vr|9gR7+ZOUG1on37D;giO?V2v6<~fClF`W6Ck}{|
zm**S`fyiirhlE=^yP8yhP<8}iY(>uV#rjE>tz~K4PtsvO>6<E|@(z<s1cU_OS%(Db
zv5&mHoirMj!Z&poDaXejA3Kht>rA%v4_Maw4VxQ=ladszCJ2PCsfsE|GI0@2B$5qq
zP9n+Z`E%?qjUICXJ|iF+WmH3J?frSqd7C3fVUKwPM@N;n`6G|;l<B(m7SnOeH|vJM
zm&x+-x#eclQ59Jt1c_r&BxHL%P*QYO_dPr`8J=^~8E>fzC+Cp{@BVdi1s=v(W%u(w
zh&r;Y(nP>BKTEpU_60G2;e{8tVaV%>UC#JYIi0J0q4VqO>#||+7i{IhIbV~&v1_g?
zc{(0J*S&~1S{>d#5jtwhfH?&>>0VTb8B-3FCg|02d_|v{y654Mp<Opr$`3hJjq94>
zj2fzBir7f^{#e^Hxs7OZb5Gmvw8kugc`{buIu=DX$_Fm(Za18U*4Sq=_VaKyD%>P|
z5e?L#$1N%)fb=1`u?KN~FCcn(zZ*X`%uhvljvgUo0LrLvGafy27_22ZVdA(e0JPb#
z4fQzYIAOYeux%Gg;MlxR*Zc?kfbw_;Ikt73O)y(7?gwVidcd!jzuO?rQ!(~u%obb*
z5Th55m!tr+ar3ThH>e%~$bH_BWS4ocBM~J)>`uA%>2<lK76jJpu1uJ&mq#4&5J6@O
zd0rjo>;41&F-S25Da%;t`cJ{*@Gb!CTNgHsZ|Q~=Fee&d+2*iJa|Cu8xkMy2z-ER2
zp(!!o*;WE-EMUCRFnIPqlo-!7t5sc9?KPWwP9Ey276r1Zd74VudpM$QNs=a`Wh=5K
zq3+9w{L#SK_L=EzHM)=1Jr4z91tkyB^Xd;#MbQ{*l)uMxUDw%|>AJ47F@C4USR)Ad
zhU?mVz0wK-S&M>Z)v|5mrmR?&uBf&oS+*gormP2qa5NVT*Jlg9=hnS8RlSa(d-b~K
zq2mP2YE}PEU1vv3*L9s8F<l2B0LQ<JyQh9O75<GGNyd)JTnJ#!f(=G2TFt{U&6yKt
zUB*wDvo8?}8UF)&M0HlEO6MUp)-biNvMQo5<?o+KKj$0T)3<jR-w@}Kh4)|JD-SIn
z|KIUX5(&Qux4>7!2jN?D$_dzE(d2)JX8BnnbDmi!M5erv^;b!fEXlC6BEkww<XYyE
ztYVQMNF`AfSfLF@pNJgKBp8yzz#Ecf8FyVXElcuVWh)BZ*r1AHzsg(o72jn%6(2uy
z<_wpdJC=2Q`3^_oXBeAzS6H)F<DBzatvQt)V^CVXw7IX$ycXHL3-rM9@+Nr~W-H6h
z(pLw2*p=!#@c<lT4iJk9A*G<?<-8{2-|<PGcBvY7V?r9awj`J4%iW~^vhi-5W@`LG
z2NxD@^@H11GlI3+()RtEykRHl(mFvzO_gO$Q7lEJ8X;G*jJR$S5mi-D3?f;IBFl;*
zOZdvQGgsC(DiHad*G?=%XSMC^2kFWd86YAmsx0Y-K@h6&s4v8E6;Y98qFM$))D60d
zNG}-9RO$+T(x@f_ndKF0qb|7+$XMtarshn;%E=yWb5)`R<daUz#0T2peD-a+`sZDN
zS`^u~ZAVdUDjoB3+^%g$qFNQzh=&L+^c^*zfQxWHJO*#FG~qaBWangaJ+3jo&6fbR
zC!vVi2}k6|pa2?90JsB!9povr(&xBu6Vc=qK)#R5NIjc7<rsUQ#3jXERJ4S_>7dFv
zse{k)OBi3fj(@}%tiWZs3m$`K;p+iJ`kd2O(pe!x)Jm(i!HWfphNBdKXi8xR4rO7M
zuLtOj>JKB+mDW-?BeWrpA3My3<1|fE3`py9Qe|*xuib9bFcRML9U^Z>ghY<B+iu@(
z0D!IJIm$VcE&=dRwlShARPD$55BT+O+}HKU`aR@MrQ&e>lR`S-1cWj@<~TNe`IO@n
z`IzJG9(aptrlXJ>kH-_}ocFYV9PR@sT-SZOqFF!j?=6wxZ3%$|ZeW&@!~or$q}pf1
z2(myIUTf{sg%$;Js=BUPt*Wl8rg9KzlQ&(h0#3`RD!MO@vr~<}e}Dc{AZyWqgXr9K
zmxJgWM4BAjeYb<?+<kYOzx3{6O;*cp*l2t2-h<qCtmS3P@%jAxbUJOldw=Cv+~3&n
zx#gtkz1EKSTE7hUw-AhTI;J`3d>1mHs$F4TvIy)YM(X?hxpPZPOG}ofX_{qT1cxNL
z9?!Mg#IdjM^>t0NEKP&<eyyK_hiCWKJQ^(or3cZv-)8x0F-np~M>$}AyK!F8G-bh3
zbX^(9vZAN^x*{u=>?)(Gr-H7t6-rP0nl`4orfb@DkU4S#l|1(ec&-m%2c~x14a@OT
znM%VMlwfCvkss<hS~1QyD$^mY)3lSYmaecvtyC%`@!&3y3KhIUqjq~`(QPEul6RdL
z>}ce}MCH=X&Q8KTlKg9S1ftkkv76&HQx%ZIE6$_y?Q}Y>fKH{SiU@}x-3&`$++0c!
zi6UBP_US?T6eBT4I0l3(odia^vD@H}_L*VOp34!r+oOh#_j;Zkc)DfjUSRta>JO*W
zsi8<Tmbs@XrjlaF&3?b>>2?rQDnVfDo*cN|L&)GkFNk1XVpC8SnJY5n{mUhpX^1)-
z_Lr<<qU&sz>H4xHn~0_?EzfMc;lh6F=QIhL*G(i%=VA6;7&E^1Z-slnV(X~Z^(;jp
z6&fBYmVA^k*G(BqU6-*|Yq!V?8XqKr@(*3Nl5r_xuG?yLANkRSae`vz^(LHx%WxmO
zmivT~VhC9;MZz4*Y+{NKj4eYdjfA&mTO>wBi>-u5LKev~zC?=@8kwL%d`b&s_DzdS
z4Fgdar+;AIKDu&cG>Pi<XmW0YkPVA5*R{?)^2j;Mbs4kfFVUS$Cbm6ULR{)jCKZH7
zSFVhHZ8G`1)Wo*q+pn=2_mh}Zt?WYdqYHbR-2Q8m$z-Q98SB!jKsr7?o)LmCLkkA5
z0#K-aO-m?F&Z_k<eZJ#gJ}Bmud76nVQ4*09ndG=gybl^W<Hq)zxot~h;y7M8Uc-CK
z@4<V258s4$e)`j&{*Ck}OyAcxU&nOAx@}FN8pngcH1%VseesK5{5Q-$aQ^)He<*(*
z05pK(88)Iigwrs--Rhp#;3;@Jz<eT-uB!s{#-V=Fk7;(;iHe90F73ED<aGFdI#M3z
zahB6jL4`;}Kj|j9N{b>9$xs-}cI5b12SxkodiW<ce*gXV->*@XM3E3tL{zHXFO=IU
zk)&?x#CcWI)bl4=T}dMJc13(YA`<*E66uO8%ZP~oNr-$!k|cs(zOcC&132l7vDB7*
z)j92`zHF!Wsj8~T+I@<0pC&76e<hVQs@>J~Hx*^m@7|?RO-@((xbY3LqW_5`#dGds
zMEVn5*A=<^lHvH!FX(z~I_nOiv+kHN(7fO95YEG0@Mtz5Vv8>gXp>}<gXaK@+q!6D
zd*7mC`I-&qbV`@kyciJ~mk17;c0*F)$9C{}#d7~TQCw&@B425@?GqH#>C(-^Ll%V@
zB)VS|_efM8+9ruGjYzpioj|E;Zuwj5IO)dZ;bw{DjiOY}BsICi$pmo^G5^iYO`w3=
zVdq7k8-eH{Sa9-LmNgj)MB*pHMU0!F!kOOPp-B@exy_s%S<9CA)4+87Z13xTlvk^*
zwADucPkv}?YwLU4bOjf=rcjQ}D1|~Ltl8_lW#&%w(skzZ;_F_haGpice|7o$%70JO
zboJD{bw1s~Zl_bU0RF8nm7S5KKg=>l!(1`yA#wW|L_x5Jx(V~xA5S-gsJ7cx*I_Kl
z^TGW5AkP!VT&LP@SLwO>f4SOjS1J|8Jg48Uvc`286T*H0N5Jik!vfp}PvxQ+x8KHD
zMsW0jc1k6jO1!Eh5C~!q*v@fQy22uOmyJIkF~{Tf+|ok5j##fREX~>6b69i(f7gB=
zZ5FH;jj>ZYnl@NfcN00c9k2EGw6U=*&#^g2V5v+iOS&Rr!QuAt5ozPWb@Ym>@HA1J
z!xDm%M7nveo=Rh)@SSS8Yg%>mJXNoAM4DRM3TC=P+aiWL?YH^dmy|EA;T2jKYIs2l
zL#_NVN!Qs_k*k?9U7xl|WSYjaXUE&CYx$QSVXOq9R{nS8MO6#q2G(wo=!%=#GYf41
z?x({dQ=EYVSKw}V4Lk*JhM$6;gWrO`0`S-Jhy78pn&ckzlOzadk~@{$&ep2K2r0t$
zS~!zJEp|6?oALwg4!uJ&9rTB@QIU)4dyWc4rQJBo`pH`I@GSE=XbXO*R5pEyFpC6q
zw374-r3f@>p!AawG5%j>(kuv7+Kp4>NGwEdOQ~sko@t`z)dR<&KT91asCynR@^H`D
zRmXABs8)@iEtsd~t_FE|h!zq?B%33?_vFcwt6WH<nr`JZoafs8MAIonqUoB7<<XoK
zaM?n_Yza9Me9PypVM&4G(8p_yTg3Yss#J}~jcUREkv5sm7Dg;YK3L^k&d=B0$6;Pw
zmnD_zrOcntFzr1x3FN8?KFq=0JA^ggF~V#_93uTBQH_INhGtfUW@_w`dDhX~a(2Y+
zfstj0|A1dpN1(eoVY+_AvD`yf{!JXIHU9yBE*ef?1Mc#|LVy6Nitxuidx+c}EA+my
z8x=>bgzMzlxk6(g$GAb6el{Dg=Gd_mK@?`kPojqH?;szKkIEE&XdQqMFL}h#IS0m$
zYV(?F95jNHoxQuHS!n>=K`;f_J8@rA4<sj{%%gRZN^Z&6Ynsmiit}1DT)cL52{klO
zMZ=}5NBcjcDnFtq*RFmxI0}A*2X}#p__r)yy|#3VzX&>Q{w+(_t}fr=lc@ZpLaB1?
z>c7;xzTa(wpTk*YSH%47FsLXP>fq~Q1R(;7oB<Mae34uY@s}Ku{qg0Ug;u_Ff`<Lp
z@Zz;=*BW?)&8t_hUTxwLHm>2UIm<<){DZ4kueKI_f3bP(+^*NUs<%_$KAzzL&fvY*
zg@_2vLNHLXO_PjCZ!f`^Aym)0y7d1!o7NQ>127-~B~fJ8*ou7pL;ZwiSC?XBcJX*H
z%!d1e+BI0rRFv(4%|a=Xg&MAho>!M;umAt6sCyoI%=2{Nx-G_Z!@9aX153AUeGv^q
zq2ndNi4MrAs!N-7*DIc<sqVYgaH@Q6F23&F$+gJ;9YxQpk4?vM%xc%_ciK@9s8oI`
zA|#rI5~{whwnvs@nyTV*E1qTJoQXO-!WnGBcW`-PQnEr8V^gIBhs)E;S0}T<;o2Ei
z&a@yb;cO75IjM|_VK!AJeZ&h>s?WJ`9D@~z7Xq)POK`*p1JAJPo`;fOs|LQm(_1;Y
zrfG;sF07=gL$3}!uWlLMfKr1AMtFNua<4Z;RJ41y?=uY3T(TI$=L9p-mC-t@l6*g?
z*8H7b;M0|r3j`5wx*|U(se2yEU9wB^M4{(7B46+41R!4XXAy_Jo{Xn=Ow%wHO@%st
z$PxJqI7Ti1k=GzJh7SW2Eajyl&rliAp^<68AGixg&`E&%sD7>JCu`YSM%6t-g-Rn)
zh)5vZjcumrfroabejI0Anq(rQSr7*#{stYd=t#$bg6q{Kt;U!_!>ImQN%rb}bD<lD
z7-F}ZYr8kzif^rZJ{fZ){9_JPP5D-pyXi_Nkr7rAbEa!~o{t-hDEywEWC~$F;Ycbq
ze*C4n?-8O#QLSz>B~j#H4uc?;dJAvS;}ad|6e7y5SN8~Ks3K#QOgrFl34+#J=BlJ>
zGuv?jL|fsE1SbgN@(I~R0mC9Z-IHpl>UIep3FrRt(TReAz(+Z&=Xs=`46`J(2xb^~
z!Ifi?s{(72T;LSkuwQhzlCthX-$K=~Nur3XykKgNxy}*!i-vdTSK?MHx|7OxIJ%>~
z`YTa8!}T4GA|el{JpzZ!1_nXPVmJ8-SY0a!kURC)oLGA@A|!gUH}{$Zo4Zuwct(2o
zB{&UJ0Kab*+Cp`Ll=93_FOSeF721uvozJ)j2{v3TM#7ni`z3Ba5UcqrNiqldi0n{3
z^8YpVljM6;%dXWTp*3wa4C@ufxKp=nwP1t6P}g*A=lnyVEID4a8o9Y)m{rGpg~FIL
zlKoK6DYNCvn&*YLSf*w^v~f|>HN97#)7J^(a!n0vs*wgVF1M7xQ#SFLS|l{pvTILV
zs52%jo>!4&8b$TrUu(4xhZntSH3}u!wclA26@yA4lN9#Dfv)&dTc+&CDz<H}Qo}Yb
zE@BkMf52slUH8$@t#9Hovd{R4?#}`!$Nzgg!x_%t6zsv9;eGIp0RE_mM3~K1+?S)e
zyhMTHxLVOdDa00|FDP=9@fNch5;Fw_Esc^+u_rSzOyfcGU@7i9ro!3L*EJN)mL$tE
zW%YhqYJF?2Qgv>nrmZK2>jY(&b3`Q&io$OdaXi4U>zwse_1lKwb>$J$Oy6Dvl7TT0
zM*)o;_q?mh7Dc{UZ`q1WWx||1&82L+@>*M`<|dO>H$XVT8veg0$&6QP;<X*$$8I;j
z9wH7^P^uv!!UoGjy1M+XvqKCee1QDja!${n{_}7euE8!mou`urAA8uGoa;-dG|Q9}
z#eA=vb1&;Mw7W@&By_WYtB8<UejbHf${1n^Y>#<M00YP<i>z|JVdjcq00AdL4UdYo
zY|Z;B5Cvl5JcZ%^)Pq2xt|3INYMR=x!uKR*yq_YXN*t$dY8q8ti8K9!MNvemnoBlQ
zzxOg#``$p)=)S^vw~M;YP#_?f0>M|=GH$zqoA_zsIQ6ZpW}3P#MD3eIOvhEc<g};5
z!&|Qhf&3jjND&GDIFtAT7r*V>z71bss>~jb>`|7?nDogmA%@iI_6pPWsrR*bULBFh
zb1(4PPTbna?YnsHf!DC@m%?GVyo(5>A(f^^rD7Nk-59_trb9_j*w*E@=FQNj-7djM
zbax_>!Vyz-gH`eR-rk;G$2u-u^QjqL5ldYWA&GXzW6=(=*5`bo>3;P0CSe|Q<=L51
z4285(iu{`q;fN&758g{N&#2YlE$D64wvVjA8MspqEt@AEV|0ziA<e88af-53Ql$j)
zTvE#!f$^eBr==wrja5M9azGF)4|1mn>XyBp&2yb*BpM8wY0sKy2y(W9c75r;I!_CA
zLVZyM$ks@$t(X7E%F2qg#GYiV<>zz%%F6rTJ+<fJJ>s4p&LcHx{m{j|y}bmeseYu*
zBaiLv?XhI2XFeiwaF+aiGR99p8x~*{P61%te2``r%^SL6D&ogaG>YrGMdoG{UDr)A
zHxorc3pb;gOGXB=7j^B24=mF_{-UP+$bo78-+$k@f$RJG>5UsVZs2!lR(aD=#}e7v
zlBs6l#8Jl*+1^s<Ys){yx$;k!v2n8eV{DdxjLo$C<CH8fFJt3m`KJH`;P|Hq;0zjY
z2S5>t%uE0lT*a5}oJArRS(edeREOCx%Q!Dv$G$1t&IgY(dJr%hFI&zBIA!f?L%y3T
zuBPh<I)#qS{lIHcY}=fXf%vv1$|H_%mTg6rjgZRhZO?6!M`*U&VIH}7ZiTBFABAuj
zg8cR$6_Ln~q?*2HC;{7h&Yg6!7&Dg=2_s2``Z&?_+}T<;k8|ClFbxVErP|iHv+2}t
zf6?eV`>Sd9b#f{gUnF2*K;?q|sX~{|qk|V_<EQK>>JwnT&#<3#B-eF2SqS>5bsiFk
zj7Je?h<=LisSvWgO-QADkA<&EH7zZ_19y4jIcsJ-s&OejhIBk86;=7b2NbnJ%0sA5
z`6)lpmOJGCk||E%o$zV+9Q+6X%HMfv>}VemnaDB#fJkH_NEw-A6u1-;T4v&G+7Stn
z<xVZ)1f&Tbs~Rs+69y<NK#i&d*{H~Y1^^>Tg_{Gy0u<l$!T^X8h=|7SE-zO|<A+Cu
zGAj7vKQW+_=eh9eP8hVK9<7@eLB+NVQ#UPLN87R-&e-j}uwIV@iVKUI>nEDYgrlyr
zwy)o#XmhcuhjaBxP^oxz+S4puXKkkI$Ooq-QeWIOsESUi)A_l6(zu{$u5-zC7{@DI
zzEZZ4bM(g8j_3X4;M?*%SDgx*V$`E`U6&+Fl5|toHIq?=8kXzc-tj%rY*x-Dh`l=U
z!^LH9L$4~P9!SJ>5WkR#_u)_bkRz$gBOdSX@2j0E*3N2%NiFMEo4f869(=y-teQ2t
zPfJu`2`&Qot67?-nP-Nf6yz`TsYrsVgGHvDDSgZ!S!ujHg9~;w{rDL1=cSiiKX6`>
ze*Tv}{NWFiQT9Vx?om8+te+Rs?{Z%y{scuoZ#%gEQKHx%wyY1^3i;@tKKtymLOlBe
z6p{V_2Kq<c#xcmadfae7s#gJLwZFgQJ?MJ81b4%uk<@N~q)jF$={b$8*iT~GVOnMH
z4pnVXktF%dqrrCRe~*mZnf+W2)S}?=83E(q7%k11%f#q&z$j@#P&3j9c|PM9@>4Qx
zPDkeu9he{Q?FcwJ9sveji9_Q&PT4Le>LkQJO~sLDieAIv=fo$DWae8<HX7w#?d<4;
z+4}(rC9y99yf){oKmupsPI#!=4{*M7l$q*9;X_!ucSy}Q({)LayuV8mFN#@csJ=t|
zX5MZsj2BvMwytCG=#pX~-QG^sFl>&SVMw2uM{9&-7?#-jlsn|a($WdyrM)UQO1;f{
z#WCD_H*cFa)|}@ey58JmS|5iF`a$61)~6h-)+&|Ss)|YO(}qpB9qtD}Y-sB^tyHA)
ze$`?{kxU2lhbTi-Se*q4aky>B^6<yUaCMDkIu`7gJ{XV3cZ`NXxVY~)NXd-T!3+kC
zNGF?d3>>MM-JXp51N|plfmOH#u2g(ca$VOWV#Vvi3*1j!jfZGu7s)zJ;2@uk8Gctp
ztHmUO51GRuq5bu)QrA8Pgk#`3RU`>^`5a-|Wg=r&yR?^Z?rWa|=)wwsAB2jU+)VcF
z*`4ZzW>6u%+yYj_cek8M60;mlC8^-@_OFtpX=j$Rp)X9R?wIy_QlRab9?)wAaKHp8
zDjR`X&NB}OQ|HSobDy~!^DSYV6EYYk7)&%=N6H1Ti~L#>ZPU;&>tvy?&J&yJO++!Q
zB4Qt%aJ~b*a!S!%D32VNG+$BG?5ynTI^)tc-o{GPHc2l0KzugC@#}2hn3m|r9r+T@
z;6*ze=JHuI3$tRMk1Hm2Yw;s*my;~~k#mFCB)5mA(qK{itPpg>nC|>|S!BcK{=+`Z
zM#WJ?gzGZ4aW>S#P}b%HTlePMfuu!uoIQI-MCGtG@9B0hugPJkg=aSyb6r8A*M3V{
zp-fK_&Pa4pFy^`&uSi}I$~0<)x))fHrbU4i_WPj}M4A>^fv1P9NRxs)`nPWYxymOa
z^7<cjxA*S|B+cOTU)>C6U_$|y;9htQd<}q43mQ11=+?GEBtn8hxJ{lN110I@`5L(<
z64Z6wWROv-#c1eeGBXKGJk61%|DW&sFZh!Duq=H;uh(Pip>e+TC>DNpZF}BrJlt^q
z?ml#meBv=XJ@JEcb8~aDTrY~EPz>jA$RX|AzC{JG^1AX6)ADc`Tg7yIc(`!EMv{H@
zJ7D}BAHx>h1J?oiB4DLjnObIl0;f8U!8|;X1~xDE;C1r)$<(y(p;%)8flq~e^S#5@
zQQM|#Hnl8z-XTmq(m`z*w>+Tbr#Nho^f!^t@21Mxe6@;)Bp@#Cp*h}o-)Wi?*DxMN
zR0oY|m7~pm65G9WwK{)PC|5MsZKocL;W7YDuQ8((LNFvk(2CC11dwp!7zW}L=2W>d
z7`ROp_t;$;7-|?cn_;LLzcodVcSy9e6A@<$^f-SK;XGr_twS;?RFH=5T~)L(yRHU9
zxDP-{%Qx_$U8_Q?R-`oWa*Ohu#WKwFv;PH2f}IWeMU@67Y+AalHP2BJZ0g3~y}20>
z`h=?8Lk)h9>o)l+!NoD7&j}e|wB8$n11XMkejK$~MAJ2`1wwYlM4V;}p;9=1ia85`
zOv>~HuesGUpwIIt{m#?4ao*aYHcx=gd!Ze-P(ir5$cx)v%)*FwDcuz6_q?Rh`d=nQ
zZ<*59I0F#X0$Dw85e<J6dQYbt@QFWX3nBK@AD;RuX7GAc2SdLeLWL-a@~jn2c;lj0
zZz9U_cExfvj1&lOl~TbZ9ltoDEEI3J?z-!)f7@r4mX^+J>YhJ8`up?eM}K|h%$b9W
zaQ?gB{jRVq@m))<3j3$FSd|t$aOB95Bh^bUz4X#Y-}SEa(n~w*--QeVkmH<{@~ktb
z2>!#@Qbs*=8M+Ix`bSo@G-)KwWlz{sn{&o*QL-2#2kli6qCBkZ?Q^ugY7!GzI+iHH
zUj|35LY5eJpJm8pYlz>Rb_^>g64z*T`VEQlqMSM)twW(y@)dCoh@8R_YK*(&`TcF8
zRb^yFXzw(`ZV@=6(>p<~r0v_nHRpa|fxoTo<pG-f!UB)mL0~6rpexXg=$RNAhYzIi
zy~2Kn<}**2WZ(;KA3@$=7o%KaP$G59pha>aaN@Rdu@vF76ulMrn!~#@#NgR>EsOxP
z<EKhX3E=zXasW6*=)XHClnrCI)}FUH0FY9v)OR#M6xM9p0~{{i<GSv@yPv-J{`bE>
zFe}Wq{2%+4%_`D|4IKK+b{cc@!-=qHtEv`4*~RI0dwk5r0002k9g}jm=V`q;nJd~#
z2whcnF`O9A&ovyI`5_*qjYgxf3~t|$zGd5%FO9wnh#<6v-G3K4gpM&iD%AFvV+jei
z^I(|wdSMiZ`uywx@1wAF`(^{uYxSaRBFo$BGd^+8nV@3Qkil~VFR6Gguwj2Yfc5HE
zTL9bpLk6B(dC7AQ@fW}T^{?I0_g$!&h0$-$j|=m^q9_BxC>=%zT(4qVkFH^C0azHX
zJ!;t%&xOv%-JH>#jYi#t6W}96bzl5*uz{j4oJ99);us{5#LGlt@WRZJBIU^&=8&Y&
zAnWI?Y<d$C>oQoa@`*Vp5xJPil5C<KC1HG27yy_0j)^Rc2#Wq>f8*rIlb)2^RoL~q
z1#NrXp~bsK4P)GZ0{BW5PAv}ym_%6_<D|3oWv=VO&>ejrwpJ>q=7pp4i<<#V70hq>
z82diPKCCTp|IRQwxXG1ZS(CO`a3&MOZiX13+_|Y9>8PL-9t@W~@63yAwY8Jr(dkB`
z(Wv+PXRXEsaKU?P&&RPtxl!YgZ=?Q7eg-PILu7P9Yg$>7wOS_!CQozhb)zcnsK!OV
zm5DGzbLfOZ@L@*pyVY)OrlgdzQmfUnQcHyxhY{4ZwQ>|<9DeZApZ>J_@9xEdS*5*c
z*bLqeo$9u*6+E!Kyj%lQzCKlCmpfGtd$ntHmD7XhQ@?*|19R9wHrj_SL$@OY-cv+E
zsNe{TcNeVAu|gNcEQ)j{b|})F(om1dLiK|GD8993aL)}LW!t{ICBLd(D(Tkd%`H_f
z)vwBLk-lv!Cv-h-)IwRDo-WF;){inPS*d^F8Ke~0igTfguUf;dTbagqy5hQc?Nxym
zoGUvh5$v@vZh3h0`bNjHC^lNI*YOBm8Lr(!^bqbp5F7=7)DzjwLhA*AQipk~?gfR3
ziHUpV^R%PA%1+*#e>=j~GU-_$GdWcR+wvgwEE|ebGPS_!5q@^s=+7V=eIG*i*Dkn)
zvy>lwyi2*}Uxkw>gtK%5R##V7DNYQ1<eG7>_CRS5eIKZK<JTO~MuO?8`Gc)pjZWWD
z?h5c}`lhLAWSS5+e8E6fFXZ@%%*!U)tzIjNSnW*=GEYy=yjaT2_%uyZxR}D)p@$3W
zX*ya-Q&>+CLQLO${4w+_^a6zPSmRl*{>(zYps^#=q|B0s%=9kK^vWgR3HE%r$Sc2W
zlbhRZqWUlwczr2QRg-|4)JI<o78h?#J1@fo00ZOM|BoD{RB#N{s8vW8mo{%|*2lJe
zfqAD4@JpGdUNguu%a1O!!_ZQN)xhm!TIxNvrwhbMD8;yjqPrm_0C4GPUr!y{>ig~M
zw<~AgKgz$InDhPXY8f3*k3(NZIa0D$e;8epRD=JiFz7$=yjE-5D%I<eWovD^o{14i
z;om}NTcPmFvTP~O=?!uS-GQF4wM69!Ccn|)Hp%754Qdb2cfb_eoEF{y3)5QB4sPtU
z+zcmV$x&9puT=bk)Q;y_)tcq{PioV&G^ct+bE4{G@dq2%l<O5mxVF!Ds^;7AI^}t`
zGyzO$d!BMM#@bO2nWn~=P_21`D9xpLC-;|JE7IUv5sJx$Xo++nzE5_e<LEgi3y>jL
zHmSsrm{iZwNZ0t!p!i4dq~}F0W*K>WmHo~(9!%vzM<Y~io_w<g_AiYPGVk^K{&sW#
zU4>TAedr<dd|ygpodb6I2^huPVJ?EXndItZia}tEIU;PA>BY7Z;ECdO5RB4p?9aLY
zsm?lbCC)hASkW}kl@^p)Ta9E+S1uGk&?J;guz&l>X6XwI;t<l*+4|>nwEWfd{bJfK
zsmMe|HHRtJQH<GJP|4$W`u{G4AgWw$G|FW~gpkcnrx}Ba$8Um<6UJoC^E|I686%GK
zwz*MrkINma!|=`M>FBxWY>fiyjVv&4WVi3tV%&`gGhWP>CW9upgG`H!R*JF*FA9oT
zub0(H#CbE4UKwxf3rIsZWeU-)Cr32`^*4-8qtR%rG#U-Vh`S8E{@l59@E2pJ^U|eA
z;JAzLaI}WC(*L1-x%?UYS*Dvlo8<Crnfp6n!zg3i6(2g1T^%2R-CRq}(*<ygHce5e
z*W<WeF9_3cX1m?lk0kKl1pC-DRn7N2zotwRTh@u`Js*W5Dgq{VDq$2BWv3cWdpHmV
zQ6&h+%wwmRSvmO$3PbyuNGKzsnVk;Y=dBIPGwRvszb2t?&gz@7%8i<#G)s&6@S1rm
zNd`9Jk($VKSNsUbvN-vD_8b6Axz<{9iUA1Y;5wy};{xMk-NQ>byO=m5adudj1ICmq
z0L3ZBKmqewv50Z8SmREKeHeM7*E>-SU5ZYize68KOsotswBm4gc~Js#x!27j1Mn;i
z^pSnj1CU1;#)WYXqlY;eCRqlk!0r&@mPXhYvn-S<VRCC;ghpoBi^k?561*eL(qd2f
zXPPVifygrWk|QNJ+UDG_qyPp=2ZCR@?J|Q-e{4}{rOLn)c51e4x&BZNz%dSlY3kW-
zch)g7WQ%`@O-)R9yzb%m0qmx>?YX7M;as||3m}a^X?#tm5lxoI+~jb9(Yf5<lJT>g
z`!-{v&*okwbxh&CDM8~yUrwG;ccJG_q_9B@TX5n5qZn8!quwfEhAI0K1n(9oI*-Eh
zwcy0#dY9&cP}|ZjrM9IXrj)wcvMCYwza@Eu3$1LM7N{reP1LsZE=@dYTUv8Ln5zXR
zCD(N+6<lk}rXJC|_N=d?&Zyhb<C8$SiJyp&TQjMj3K0v$R0xTa0<Jp)qOch9>-g*q
zH{4(tHwebb<Rsxtz`|P)ztJ#m<c!x_KZRGMX-Z9#ri2i|$jl63f)FAp{VJx+@ZuvR
zuBB?-2D}diXaOyuThW8)&(Z4;f?+SqhVcc0i)K{;PD%=0HqM2Xakm*dgEy4`xRGHp
z%*Dop2R|Of&nB~N4XP<ZNjylh#CYbNhIt+qqbypDIlV@P`7i>3TRZY$-Yw>?qYQ6y
zg1~XKs*U*)3t^4ccU^V$VT>8xv9y#AgqZStLrh6)J3LSP4?mxRk{wG27LV&)rmd)u
zX0_UEHZ`cYH5rY5b@b>_$vGJ#oHN1~DPb7I+rPRn|E?ST+|jC5D;&eIc7nix^V)HO
zz_EnD(jTj-ZQCt5B@Ex25`361E$zSz<HJ{9wQI}2<mHn@!R;L7Berffo7Ji*1+|qK
zwI<_uY;25hE{_VpjL=0+80X|Xp_|yx-5$rBo5uuTHoE-?X46i7x?zr>=c2#u{|9?h
zR3`&VWHbkg6-Wm|%bwpaLeMg*5y`^xbXW-P>_N;J_F@K@)&x*Wn@vios8YZz_SStI
z5Cg4>$CjcByK04LYUYDKoF8Va$$&9ZZ8Ji!?_)w(Tp^4BYu=xETU8Y()5PNoSVq>$
zHI$<((KFDC(A&{(5o(78o6<MTtVRM=5pX&6-uQ763mUVCwV=QomF&|~1ldH~%Cded
z%~XBml778qBnmSz1R+{u-2lTV6bWewrRC|a{0kUU49w<YV(ZpRF4?+uLU5Y_4hQ{x
z`}%|MU6<vHj#Sch_w7G+Z2vyjl}b8``DH(Id|zwLwRQt-V!<T__yz!2TwoJ}@?9SX
z)V}XL)SftiHcjylG)h`D>>3}UR4Yz*#xXz~@1$HQMz^^f^KB5~`GrFaFu8T>HjKAz
zy_H~~2N&k?=nFQ0atZ}47$X#$ezj_=Lq0urwd$JylrScAq2MS0n-=`HCv2MPUckhS
z)`zA|_}Zt@K6E|02ffh3a1gp?Z27dBB4C9c4N6SdMvTKMgb2w9qOmO+6bBsaL?*Fx
zEAZ0|!bO5_Ouas!N)n;>MQxfwDmtjwq2Va>v~s(;#pzCh6<l1#J|m_{n2*Z>L+lMQ
z^R;cyvWl(ywiD!Us8)S#SF&t70hC#8kVd;-mb=()x62rO)ARwC+pIP*aq!(<K@Xv`
zUsKVLWa5aPY7NDcYY$;0{nwO{D0$qpeZqy@dSGd*<bwL`>5|XnOi)4J00d`~-|`WP
zs?(BlB}TU<;apBvqc3@!`gtWa4R%j5GBZP|B!md!Xm>`xh$cs1CX08_9&|N2g6>B5
zqrXC^JsdP>bd^)<?h?f;h?;<ilB9!kx=2@wC^!+A|8=(pA@VJ{K_LW#WfZcs!nsAb
zIiI}LF(Ogd_U7Q~wc?suD?6&!OMPZ}?;I$74^*{Mx>|LV0-S9((`cgZM!2o}N|#bO
z$Idp9tku!}WHJXPW!GBC1RD_lqAeL?IwDaRi*<cuuOhr~V*58I4WASA-N2WoQ&T^G
z-&{Z+L|+!V=S)FvV7b^#V#H(TEX%UAmkzoB29u%l;v{V)-l&*r&VF59+UsZQeZ3^j
zn&P0gBEH)*6TpR}j8Q2T!^JAcz<^)~aY8BKFA~IV3-=dM_VIuL=GCGUufT(kM#Yep
zFgV5-bHk9*Ft`Pr696ETF-{2AmsJCR8N>IykTW>l1m`w44kZjNh^&48So;aYQ>Gus
z`nl2dgghw_s{b<Ktm&sOcIWxxxDJ4>O$U>h>glu>Ns(`fgmbXCu@4AZq)$o4+T8=^
z&^ys5(N7U-UqHdAK!njS($d(Rq`^?s&Eq6V2PEM)7iGU{(q3BZX00(RL`UXxk+y;~
zMbP@5EC%b%!Xu_7lz-WTVF_yg^|S&Ji7eg>aEX<uPaJ});<@_vUbueC7AzSlcp6i^
zg@Xly@R?GJOQ^(i7l-6Ihr<Kq*8#ya)jC13@ncLoo@baeV$L<5VKVAyKVaaIl=lkW
zvgP{bxG=Ej%wA!@9U(BS^0O_GvWA2ujdF1>VO>I~&j=*JvG{X5(`3wX{2MebF~{}Y
zS{BQ_i|G(LLQj;VIG6Wri?;oHUS2yr9q??zh4m?m3!<CN`T1s3sb0qwH(SCmgf(5R
z094A;4+Uvlm7x#6Gw2MuFul*pO7WO@*y<&+E{>+}nvzu;hiA|U^ysw35&4*ic+Vhc
zye_=ue|$!poJr@HvgJMc<TSCZl$cblD_JI}g`!Kl(XvI~KKcCgVp&2rap+7zy8up=
z$)@wQ@;1R>w|FKsCoNL-cAI>D5#EiCqLb*wyh`cYQ=CaO)sgvNqvK?l2gOtY)5vhp
zAEZhxG|YI4|3;ouLiz3M`S1!MHndXU^r49&f<;VX!%M2;3hfs#<C%A=yznrt-8_ud
zV`Lft#AF-(BJHN}ZL||zfsUaE?0A6GAIZl)Ljq8X7)VbOKx)x*=0WJnh(faBjD+|t
zG~4gb;xNSHo(Jn)uIG)fMTZ_9+Q(~64>7V&sZU`wD!r#qojMg_96lLA_~b=LbdAN7
z>9v}R<FUKrK!*Z#OX-xkQO7IqGop{p4^d`C-l*Hrg9zm#w+;FqZOjgnQN6IAjcPL|
z=KrvA;DWeMXXJq|8c_f<T@0tY6GUO}Ts0^fT}<F!pqci$ROg%yL1NJeZ0*2E_Se2J
z;k(M2-<=j7m(-?tdiiykrbdoaUc7?n-=(_|leJF^l-4R>xTrc&$=xNd{ies2<Ga(q
zKh;3H|NCSaJs<sF((wZSf>CqHMjpI}cZj-VV-lwEhDNhuo4!nBX+o=Td_V8!O)f<|
z&(pL2MozN01w-Ljq;G|<{e!z0hw}<Ry)?ph=zm3~x%23y&>ejQy7w9kJNqKRp1x0(
z&`s#hxGFxbZz>~D_*7~C20EWIDZ_s{q<z9Jrpz*9zaFRaFGF`~ESvz^Qps<I+Fr_s
zTb+~UX^wv$&$9YB7Q*nGaoqF`A@KP0<8SkJjuA38)-o6)hT|B7F{3s1mY@GEM9b(2
z^nG|bT!25~X?zm@J^nK3lDo-&QA}g{TKZ!au<h&wdk;UvAL5@EHE~R=i;o*k<7(px
z<AU*hQ<}TYr<?DPRNf-rX!Wh@t(REmtsmPBd%u0k{)ie^XVo9|N&Q*p&z&E;)9%Z>
zruUD&;lH|YET{z+inGOsiXRE*!>>n^(FaPy(i_Y7R^}_eto~Q6Q#)Jx{8)EvWBju5
ze{a30_4CQylizQz#m(eUvfeQ|H+L>fJvPm!SEm1KCYZT#W@Gm0bCYwwoWFPeT?>l~
zUrhfzli4HL&%3L=p!b)3*?(d%J2*8=hhNQa&Og8S#HP~b>o>1&vA5j2<+EE~v~BOU
zKW=~Kj?RwX?)>zwm0e%ly?^%y_H5tt*L!}u_vGF$?(6KY?SJS%?ZBItHZT3x!L0{R
zA3S?V9{S?p2QJ~4?4i~ErDM9|WoLKi#V&8xh3@k1TRo#a4|@~6|N9R2z3V?aP(Sd0
z@cxiFTt572<n(BA%o=x(znOSCd2{O1^!4a$tT*;G{$I8y=Tlo!-_ut!wVB)5^SS2S
z&-{l%Y2l%H2H0Y(!2keua1H|igPp+gfQzN<94KNRxvoX{8(55={R2zb#>asboaFkz
zO19EJu#Sy9mh1O%up*2_%YmS4^KHN&%)bE_YmC4^5w*sX=0&%G#UzZHfhBld_Xbul
z>}nZUiQBj~u#Us7OLuxbYm7jnGPS6gla4A#5f+WMl%iLrX7W+{-_txOZ6~8l5s3uc
zUSCNl;PZ#X(RIyPZ)+<RVRy*wKbF?oa!Ng(q%w7Mx}xs5D(p_4TKQqrDQm&bq|#a(
zjUN+x`3tQ-EybLS9dSfKIa*lCn#o6Pk%=o)I8uwKnGjm3d@iHiXSJ<SnfiD~g&S=t
zPwo1s(1^lBBdB2xiG!kmL=Z;Mh>di)=X#N(m?2LTdy<0+e|?9{_Wp1cM2HZ;jTav!
zgb3inPYA)$eP(C3BukIt&bSHT#?LX*XfzfMrFig=Joaq#=DUe6O~9EopW3Uey9Dy^
z0yY^^M-UbqCC;&sNw5dMpY)a&n?n-g2o8b*>LSX0fwff*H(!uB-cN2KaFD;<h+>jJ
fpiwCD<jA0L#}oV19G4T0$1mvA?icaP00RI3hDXpZ

literal 0
HcmV?d00001

diff --git a/_static/vendor/fontawesome/webfonts/fa-solid-900.ttf b/_static/vendor/fontawesome/webfonts/fa-solid-900.ttf
new file mode 100644
index 0000000000000000000000000000000000000000..bb2a8695653cd2839e3e52266c124c5dba1ca995
GIT binary patch
literal 420332
zcmeFae|**R|Nnn~{CJ*iwR3h>TdTI#*49}++4{9ga-tMM6hbi+Arzqqr$l)Pi&O}c
z5JCu{7_X3*B!m#Q+7iM!XXnTIp?z-m$K#x}yxy<(`}_U=@%{YsUYF~+-Jg%&zn=H!
z<M}+#0|_E(MFCP&aQfJz`&@M8#bb#?H-uJPa@9rCE`Mp-PeeJRiAFtl+0<E=N-^j}
zqU)>4c`|Rx<clVC*V`5#?P@S!3Ig=#l!sy0f#NAw&6pYc-19EVLz?W3Q?I$?qKxjr
z@kAFk;&{_l7tNgJED(zjpNjMaS6_71<iASwo8)T0gGl{8?V9UmoRz=pAi4536P?<b
z&{pL*6iu0#d11zoD7BWzMzpi+qdSt?)X?;ryjy&N<5ZHI1Aq7wpESKjBm;<Vkas7-
zO(cIf(>cVMF7Bgj6Uh0*5vjCPY92WV9xX0K?8t;hsRPYJ$&{u{PW}+yW|L^xIb%=7
z_!c%b**2hUr*ImRaF}=rT7bg<)Zf$~s$u)76(5U0(;*ytag5}!{Y?kJA%q_^%Hb{H
zJe;2M`N_+pldxI$nKF~<t2yp(`TbO2>gRBd_qR|#>f~@k(;;cw(PYXGG#%i&49Fub
z1<(Q+Fr$7*Qqy0}@f>F2n)YEvfJw8@gi#_tmubQ&i_@Szg$&BbNY;g8ug#w<2g>;q
zWtu)zC(83D`hw%;F$>r*)Z<6L3X<ira^52p@dzuBe#8ZA9L8XO(l%vsxS!+D4k+g6
zh%qy11Bv{s)h+B~I3>%5BFwY_^C|tX8A|4BYO-;ZZ^Ahaa6e2vep0+<I88E7(r#8B
zC*(=i^KWC>GCsiN+j%#6Y`#{ByalFAPD7?VVaKm2ZqgzQ<5R#ir}LY1ygm-YYpOa?
zPr$_6wk2)kN%YNXGMvo%>tM?EV>|-QV?(CRNN39B^!`NoMw7>hx-g&IAFOF!*F3%{
z$^40S+V=cCZ}J$$B-%id!vcQ??c_21w{T8p<C1HW%SrYX={sUg9prp`zS%kwYaQtj
zE|X<)eq(dEjkCvo$~AQa5bjT;N!or&MO<^cL^aOU07;A)#*)`pa^C-@NMquN26LV4
zlWbZu&z(&D_FQlu(H2{87t>C*lMrnGe<08zUCT9N?ZbU;nU2fCu^C6}H`sS%@LFgl
znV!=cFz<n+ZB+2sm~l1n2qVC0cutZwAODwNo|565mf>UKbq~Mh=e0&2=|>n7FmXmP
z4=wVVHE#2hNt?xeFy--a0@E)Zlfz-~k>)sE`V;ArYSZv}<K?*Im~dXR4y%zqV8cxq
zA9EZuF{VcxC)aDToR-RUpg$1~lm0|Gfg?3pFRQT;Z`zSamyG9eG2wplC)55XZEgdX
zl_;k<jISSL>oR3m!{#-NdBNP`T(y0*X%fDP=Z*2(coWBd-hafHoA#Tz#M(hQS*H9z
z;u=U?Poj_2h{tiVO&o5<2w|pdvhg<m;Xc^7WSXR!cJFVF3m`l|ZT>!oC|@V*OVrg0
zd4#E>I$4&<OVW=B^wmJH$&Pt*7>@(Of!%LZ!_OcOFuo9Dj(Ja%6+nKr6Jv#WOkU%U
zSobKaAlW{)fgi^lR>0#7nvauxOV(q`C-`wfJuL~CddbB56YcR+TO1qTWWq7;RNW$<
zpGvq4F59jN>pRTaNbYA4-&~L1>{HA!%1gzr&(_Cj{MP0*XsAvcL#5xA!EI;;VHo3s
z<``>|^Ppa%#^$zJQlN}IrbJl*j1x$<4Q28Ek@sPVvJThTB8|PqRGYeBBYmK`JZ_K4
zZ_@j1cynIl5hfoLdHy?+d6LTcfN5*OHe;BGx9bvRn&U*-#&;s0O@npM>mqp!+l2cO
z&jgGXux-*n*|upg9@U97WW&&Qvu0V5royB%YoIwDuR$)C#~AzVgdYhc;-Ss`;B<e3
z^I~t-WWoc9HWBANJijQ@j$!rTIZX5k{WLy-ug$g(i87LPCi=*2<n~MHPoyKOiFlir
z68lGv3t+!yG?AXi(fC42Z?20F=1-IpfWqc_{*Eb!TJ#Mzhnc#Knm#0tt&box?k#<C
za!-f6>^lv-j_q}q%kd-3Z_4JmuyrJ_%ZN|5qh;MD3~3LK1N<8wbAG>7*k)dn%Hd|s
z1)9rj-ba`?N`|u!Hnv%}N7V1Pb#lGHj9H>xe2wO|rkOTyet)98K#OZ!z~)cJ{jI%(
zYf@8lpE#{eUm*R7vXUCG<(qi6TS91mfPHs!KWscD$_VgXH~R&XE-}71_S-U>HDJn3
z-XCCpYSW;7nJve}*qk>&F53=sZO<@cU2VdTY#++S`nP*DV<*$t_L%-7JW-apN0NRV
zw<M5=CnytWJ{R4b|36^%NYZceq$QMC6Ll(6XYv?o#)!2ANSnM)ns76R0aK>7Z9#d&
zc}>`1<uLa1$#6bq(02AG#!iHp_6HLClR%<B2}L@yZn0L9=TJJrm?P`rW4|fatjXkB
z<8;Y&lMFM8q)4B*|7m&sZQdUo(MD4r$~sb;8qIOw@8O9)_*;Y<`)~g0-^xp*=kojs
zc<s{RauRJf?J)C*D6_uJ8plDv*4>;3$IWvPASL~W%Q9)O*K1j(*-!ct=L?y#3OJ9g
z%j_G?^Nd8@cJG9A&GoSzK$`+kzYRBO?Qwxg$2NhavQ2i+c|<vWgflo7&6)`$@*s?l
zo6F^J)Bhv)3e9EleKK(x;HM72`fq<CjQwwbuN^pM=t$*y5|}imUQUz5-)vjI7-&&O
zGS1)BX!7x~pE|=Qn7s@4Em3EpO=j%vvnR6UAfLTfaNa;7EI<zOnms?tOI!=!7buY?
znXVeger|tqPEEb$`e@n{u;n%Ph4-RJZ;yGuYum?Z?K(-Ehe*?$Kai-ClKB#GeC%h;
zo+aOu37^ku%B$w|TwbC~giAjNaCkzweqwAMs^+-nbCmsy-<I2=PM9{o$!q4G+iu`J
ze6o)Svt{$OgZr9HW5SUyv7T)Ek(R@O-=?#9ZI~Gwgd?8kv4GPU2x<8^VAI($4)=$T
zlVibZ(jn~;VW>Zm4`XM>GLa7LPVS@ao;v05G4h}s%pX4+Nz9#XgJ~<m@TY+S#nx_r
zFOJxq7eiYS;aIo70#BM(B4L_XNrlvvdeR^|k1nE%=@Obmlj%~rjHb}#bOl{WQ|T(Y
zny#T~G@Y)c>u3gDPdCtwG?Ql0P4px^MbFan^dh}ZZ__svFKAnkSJ0_oP{H7Wl?AH`
zo-KH;V0FRTf^`KO3cD0m7A`8hzwm*=#f1+RE-iek@a@8lUD|ZXFH%LiD7UCvQTL)A
zMLmliD*CMGm!e;belPl?C{)x?)L68y=s?k-;tgG^y1w4^v#wja9_V_|M?T?GKJCl&
zW&1k$3VdCBMZT`S9==|_qkKpE2Kom1j`R6_$NSFno$s6Ao9Mg1ccE{x?^55jzMFhE
z`)>6Gd~<wreYgAO`R?#N;9KH*$oGh^%J+nCh3`q<O5anyr+v@(R{5UwJ?DGg_lj?W
z?;YPp-v_=AeLwkr@%`$H`}X_(D!~m;i7s(U(n?yDv@S_6$tcMzX;ad+B&#I5q+Lm;
zlERX*lHnyMmW(brtz>M;xRSF<&MBEx@?go5lBM1Ib?@KP-7~jmUeDs5V`^MA?wU3=
z!)k7*xwB?j&BHa%)~v2sTk}TE+clrnY_IvLrmnWEc1W$i_RQK#YA>(7s`lpEJ8JK%
zeWZ3(?V8&4wHs?U)qYv~RqZ#mzt=X@#%d4kOxxLIXYZZI?!0T~@||z&e0%2yJO8zF
z%g)_9f7_M6>!n?<?26V6sT)x@s%~oCb#*t^-CS2$x47=9x)<u!)V*4_x$eih-F3g!
z{a*J+y{hk8e{}un_2cWWsh?3lzy7}ZhwC4$f1>`m`j_h0?w+~(rl1U}pc8ZjJ;AoY
ztYCJqU9dy2V=y<E7wi<w4-O6bgTsR(f+q$~3!WdG5S$piAb4r;vfz~9<-x0iw*~{j
zJA?Cs3xanA{}H@9cyDk~@WJ4c;L_l-;A6ojf-8bA244wo34R{@F<2Aa6|4)^2X_bm
z2u6a@U@RC99tx#|QbVmn=^;-jGt?%O70M3fg*t`uLtR2ep^{LyQ1?)e&{3iOp~0b{
zp%X(RL#Ktth0Y3{8@eDgIW#kLduT!E-cV)efzX4YC84FEheD5rUJR`Xy&QTq^mgdu
z(3a5l(6^!QLf?me4DAl>4aGw7(7w?A(4nvhr-WOF^TWNv1HyyCCxuTApB5eyo*14Q
zzA-#Ad{g-5@a^Hd!wbXrgzpV63f~`oF#K?MdHAvL%J8c2>)|)TZ-(Crza3s5-VlB#
zTpfNt{Lk=~@R#9j;T_@n@Xz62!@q~a;aGTI_)vq>klv8jkl)a)p-;oWhVvUHHdHiR
z(lDoCZo~YB%7&*J-feiV;e&>c8$N6Jyy1t2nuhv@-y04z{MFDDaYek5E|KC$*GRv}
zfXLv;kjQb7VUZIeBO)Usr$kPRjERhmjE_u=To}1Ha%tqM$hDCfky(*jBDY26M&?KE
zj@%nr99bH9Eb>I;g~+Rsw<7OE-j8gId>Gjh`7-i-<fq84$nI!bv}d$ev@}{4?H?Tw
z^+$(A7eyb9u82MpeL1=|x-PmQ`a$%Q=$7ag(Ql&PMR!Dhiq=N!qCZD}i~bS)D<)z(
z=8d(Db%=G26~?;8dc;a&<*^~LVX@({k+G9wr^n8SofR7&yC8OP?9$kkv8!U&#Ad{9
zh|P-4j@=fU7rQIAB=$(`iP$r-XJaqK-iW;&dnZ;M+Zfvv`z-c(?913!v2SAE#(s?L
zitUc=iTxf6$M(k##wng2_r$Z~Ir02>mw5Mh&-l^tW8?n#3GtEfQ{!XfXUETvUl_kU
zJ~ciqeqH>Q_?-B>_=5Pt_~Q7I`0{vFd`0}}`1A1><FCixif@Ql$KQ{C5dT+vOMGkm
zi}-i(-SNHgKjRJYXgnT2&?p;I8e28CZcJ}<H+mb}G`4GO-`J_KN8{0rLmN+S9Njpk
zactuSjh8fD)i|s1md3e_^BWg6E^U0TadYE0jo&ubH~!h!)YJs;T}VaLgGy=0|9a{F
zi(c9>Ui!G2b8GIXS^8gi>0R*BgO2dh^Wdc){%^c=;lFw5Pn*5;%Ly+%(s=0`>TXJS
z=~Z<v!ArkZx23M8ZV!9ux<Bhv>PzbT)Spp*Uj6j?8|v?>UsS)mzN-G&W-m?f(i&bm
zCFo6f>6{i`x)5G^nDNphgJTk2x+3AFuLw?qm%hD)m%b-h8GIn&rI!b*;H94ot_fBL
zw+6q4m)@E1(mw}-2`}9kB6#UEcxiXYoAA=@jh8Nfmo83t>7Jp|P<iOskRM)pRA>ym
z^x2{FLlvRxLvurS7%#o3*-I}EJ#W19E1`9v>d>dnUV2BUCiJWE(v9%a2f`HA;a1_C
zaACM_cwpEMFMTS!^cnEd)8VCO9pR-bo4xd-;i~X6;kEzfrQd^>{wTZ^Ui$lHFC7X;
zjhB`UY0X}`ykWfY(ia;qeW&r#?<BnRr!Bnn{=a$YKJe0m{)3mE055$Jy!4dFG<fM7
z;iYel%!$l{m%b-b8ClZorPoC^M5>LK{uEyN>&TChosr#<{n2cA>E7_tec+|b;iZp{
zJ`jC8x-$A)^wsF=(YK=SMn8&v8r_=k(m%HF(g*+MrHf<TV@JVDA9IA49t$r$5ng&y
z?DA$WJu`MQy!4#de0b?)v8sfZem&u(|M@pB{X=Z0@zQ%^LH5$Irg&?3>8ym8?glS?
zRQ#A`FMS5Q^m*|M;H9sOUlYGJK06+W-yXjcUity!r5}fvUKxKj{zCk<_?z(3@4`!O
zjDH;8Y`pY0@jBzB!^TVRhnIGmy>te=bhh!*CI6+DzPWKu<2>V~tB>^3O*KtFHGSXo
zP19HK*Z=$Xe`y2?dKPpoDC`{V{A1^>ovS;)+xg|rFLhqk`RUG25_O*6c~0kBJNN0_
zvvar3Ih|8F%lxMNaQ?6PwfW!Vf0Dm3|IPeY@?Xk-Dt~4Ellk}N-<3Z<KahVz{)PEx
z<)7H;K&M|ieb#Anr+;>Ox6_7BPj_0;X<A-Q-cNat=ADywdfsVyJ@dNe73CG?x$|1(
z?$2$^jps&lBe@N^;oMMekSO=h+`YNK<^GbpC->*v-MRI-b-BB8cjngQew+JM?&rCl
zTz4kkP|BPS?##RcpYy=&JcyZdKp^v0e9kt<H)YN;dSm7d2)iE4$h<D|+RSMt^lBVm
zl{q!@%1qql!Jd*iIdc;9lFW-U&&nK^IVSV;%+ojqVW(uCoUle_o`~ZSV0h*UiLhgF
zG$ivFd=3VKz`)G%%mJDGGW){rlUat((#)eXkIL+v*)G$(wUp`3Oh;(z%vPDH-UHtK
zxOLv=je0}gKfLv@>%cDWPVZ0NuVLXl@ow{e>HXaMnRlc2eeVYEdhhE9dlkIwebW0l
z^f7LW_g-)hxZ50IRJ?P%0q?Ee>tIa>=(+bA6E_vs6z^sDtniNap5r~sJJvhKdzyE&
zca(R8_jqq#Zy#@W#O7uk%s7y-F9UCZ;c?{G8Cx>m%2<){M8;zok7O*%Semgg;~yD!
zX55y68ZxG3T#<26#+Z!LVIQ9{EaSM0p&3Imj>#CDF(_kTM*oby8GSNJGkRro&*+v>
zl2M#dlu?*bkdcqn9Wy#)w9m-S$jWG&(Ix{U<2mHn=ZSh6JYi4B6ZHJ*+3l(G?DW(i
z=3o4ln&%y`0jvjad*1TA=~?G_-Se7ft>;zGE1s7<YdkM`p7%WKS><`g^R#EBXNBi+
z&!e77&pn>IJps?Ho>`t5p6fi*Jy(0K^i1|#?76@*!E>Hxyl0$etmjP67|&?WNuE)j
z6Fnn5!##e_ah{=`V?D=s273m2dU<+!N<3XWg`Rv*J5O6rrpN2?cv^cLkM<~!^au~R
z|8nnlH@f5Qh&$~5&Hbx;kNZdW5AN;mZ`_}|x41ucZ*qU=e%t+~`wjP6_e<{8?&sXk
zxSw(_cR%8O$i2*6>Au%}k9(o}F8A&3x$axsH@k0g&vM`BzQH}seYJb4`%3o}?#tbm
zx+l9Qxi4{F=sw$hhWm8)sqWG4Q`{%HPjC<8xgv58aUbIz>>lVYclUMoaQoct-EMcf
zJI$Sv{#W|G^k8~@dR=;LdQJK_>0hKjmtK{=B>h1=UJj(2H|%T~c>b~G|NTFg-*vpJ
z!qwijz%|a5<-F*sacW%c@UIpA4R#f~y19nCI=Uh$+Lh<J*jewq<GkgpcioDAcRF7<
z%o^7khjR^ac02bw_q$d)H#iHNxz0={;9Tn(<MKJz;@{cMudZvIhn-0%<2<CRbhbMS
zT&rDG&N?vLwb-@XwZt{UHQRLq{>^q>?JPmro1M+Bqg;LP&xe+Db6n0kXO;7`v(h!h
zgdc_QE~w#ZXT6TON*$l$cUG8x%bY5d@EHD$c1}jUGo78Tp(x`hr@brB+3PBGb#!7*
zDoWJO0LSI>qO}2MrSq$^TR*K=;@=8qj}z3Vp^x+Q66aQZnjWnO>H&JL9-`Ny9ShOR
znJ9Ug{#6@4m7Iqo)!BUH@&BJkYTN&l+BDu1RWEW%T?Gb$75Zs>b~Is?u-fT1Y^&E5
zycrbUQ}ttu|8y&jr*O*nW<FMH)wQg!-m;qf)c319)j#-1_fuaeypIaYk94byHAG#(
zrKmo9q~PJ@cVHtQ**}|Cnl?0t?N>3j)tSi<*e&|4Zo=`XWE{6l-kCf)tR|JY79n*2
zY?8b2ITPG~9LLMy_=E?uf9g_%qkZaPP$6gFa{|YxL<u(4-#=x=;Sjl8E<b$q?-tjP
z{FA1C$)o>CWpiJXDW$n0B#&CEZG*t-6}ZL<)G16qTBiA5n;xTo(8K><#z7>{5rIeU
z3eT|}a#1F=p|+HTccR--4z<Ud&K)V2@~9K#Q)eo`bCE81GE$6Z*gh(uZqyx5NqSN*
z>P<({(R3d@Chiss#XaJF@qkz&mWhYNV`7C^EnW~Wig(06#YXXg_)vT#Hi?hLCt|bs
zRBRQWi!a32;v2DDd@Ftsbz+bBL;NXXA}*y&lR2`3>?FI&qh+b=Bg^GLIY=HWkCP+h
ziE^YoS)L+C%TwhU@=Q5So-NOn)8&nFmYgkblk?;q@=iHlE|7Q0f5^M#{qh0%uzW;5
zAy>&4<OW$SKaiW{7Wtk0LGF@uvR>|%Kg&Jx7x}CFP41Py%Rgk35=tvarKoh}Rvwk1
za#RP^QI)9fs)y>Sda0w-U^PS?tA;ASI$n)ZXR5JkoH|RLt<F*BstZ(wx=3BArmO4K
zEOoP*t!`1bs@v2Z>Tb16J+F4Coobh=Q}t@M`dRH!zpIeW))(tb^c=lJFVj`<_s{58
z^qYF4{y=ZipXe?63;n&W(|dKJZgM=hs&sRXcKYEuF^I6^Au7$Yt3eBBVFI>`Kv%$D
zg!BMCfdfhbcAJy}hJf=xCb%d;J1BgKAqRR%g7#4O8AAu?<OCg|>>-$3fZd^i+chOY
zC+Ot~uyeyVAC}J0D-#qzvA;BMKdws91qzQ~;6CwAlqm*$tz)>K*r^(PQ1<yu3BVrH
z&<zUjXuuv7-?LbFY~WW7JVrMr=mmw(HS~t^T>*0xfFCy;4Smug?|?pKk@KO?TI3e!
z^A>p`^hJyEKwr1WYUtY*ITHGfMLYnFThtz+g0>d%0~F)S$TOfAUq<4(Q-HZ;<jK&%
z7I_MErA0mgU1i~`PXXG=$h)A=S@>F4u-YQ$LDyPz9dw<=835g2kqfZ&Fi(tfLn|%f
zYv>}2*bcqlBGA7=^o<eULKj>3x><<bB%{Vcms-@h(6=lS>$ebN#fZnC8!fV&s7o6Q
zk53n@Ge(7QDkzH_4b>KTDirILQ5=T#!AR6q)ZHRBL$N*>@hKGRgVEX0hb-a+=w}xG
zRZ7t>7O@HXt3{xVMZa6b$Iw44;-64%A5h0b8!XC!Hd@4o(0vy15%hpXEP)=f$VYHG
zU@jQ745vetMRtN>jWOZ_DB8)$anP+6@eUMaGA8{&*nn+5l^FRjR9IvV6ywP_{h$~-
zM&miUFViAtLor5-g93b97LaMs0*l@Y#d={JjJvPM;^4gSb+u^p!-uhFG{(@^%c5U_
z9%a#QLUA548sqO9XwlC=2U+CV(Bmw!9_qKqT~LfWBkQ2&S=7zY^DXLD=md+p4LZ@H
zW<f8o@LtPzp+(&Sooo?zLoc<+AE4J-B-V`YCX2DTK45IF5AeG61uTl&HpilVhR(I9
z-O$@Diu*FpqIN>>uqf^?&ND_%hhp6@`0)t&9<m6m8Q&uo&GqoO1Htt^VUc~HD=cyk
z^ht}vdiSlgNUV3?Qx@K{`kuDP-=I7nKyv$6StPH4XDt%vhwnKHuTS6e7Gv{#1Bvs-
zx4|MmfWBi<*F!g2%$nplpud2AXwf*Id_P%aS17j|Xq=}$Za3h&0AJkV@S56h(OaN@
zSrpn?LKfKpDlNv=uz|51*nrQ4k~E7%nI)|(`g>?=i%Nl}TNG=Ch3^$ga85D$5@;KX
zz8KopBJPJ~S&WTyt_1M4qoke1*qvYlzMCp3v}lY`NtuQB1|`ESY6$d1i?K(;2Bw_T
zU<1DIDj93R4w6bRXN<AWf(;03tmGVvu_wU>d|s41Xu&R&O0XUo_S+>(EquM}-p`^g
zg7&wl3T$QE7QROG%(bXHpm`Q!7sCd;etM3vh=oKo=w}V;<b6;L%3-Jv6l2R!UnqP<
z%`n)gzvf(U1MG4r>ZrLBHu_eBcGlp$A@r~2VGD6utY?PzRW++EbUt*gg(g7Xu+T*4
z+ZKW^srk%;y&ctTw-EeF4c09~70_Lv4mQT57IVk&9<3I6YB6tw@vZe+c>h&<riJ%k
zwVV#{{;QUc0q?(Rud?votM+CK@4ss20rZ*oUbXjGc)wMP{?<N%@afQ17T#mkuCeeQ
zt9HGG_gS?YExgC7-DKf?R_&J--ec8%W#K(m?Kc+QW7Ym{A#PWLg=Rxz7P<v`$U^X&
zJJT$58?+E~K|2G`-WHk*J=Q|EL+`Q>`nq$uh0y1nm?MTT-#al!49$mrV4(%je_4pj
z_!PiT;Qowu?zRxt+s@xCbPv(4d<)$ReaV8mDBAThcm;VYp-~Ink004$J{fudI>JJW
zp(DX4gg*$yoH4W%iaB5i>#`1GU3VkwhoBh8I*cDZ0!1ARJqqQr0PZTO?kNjB4#k|-
zy@2p4DB4+vd8MbJuUd%XUjv&F{wx&zuKN-8d(awy{?ixGJplcuZ=t_i=m+Q@7E|w^
zMD+@3kiQ;&u)Zto251R58a|;7v=2BPb}sY`FdjD6dp*vv`fFh0T&>4itjD?&=Rnc-
zdd!tLABwrDzYq3R&_w`iS0G(I=DHqZAZ~?LfhS-Gpy*HibFi^a>R+-5td07$7I6pB
zZmd^EU~TNi*fIk31<~}NguMWYJ~8a`gBmyphu;gjEbNPdDZqnptc{=-w1tg14Q5%C
zADV3un8#o{i^LiW=70`JkNyNZTGTiw#)wfEhhUyXpr64`7Kt?x%(w9VIammWq8!Y7
zkn00{-3ek$gToQN0Xo7WaZUwKwD5H;I1-$OH19#j0E{>9Z-WyoVk>l_MSKpuz#_ha
zR)9;9{%h!E7QPk)r&#z}5WL*N*QVeV;A*7*5jqXr3VRncU=d;H?ch#?WBmkqK7hO*
z%JTu_15m67M#P~!j=;<d&kYcbP@a!_Vef-h0<0Tx0Lu9>KjJUw5)1E>gG()9HFTK;
z%YcGdW5LHzuYgv8Ct&khSYeUJLZ1XLBAola2D}2h4YV3;fz9*J<B0K-#ZVqEjDhq)
zYb<7s?Szf-;QOOsokeq7>Mddqbhkyspg)5@kiHKz2qLftL!%Zo6w2uUeos$99xotQ
zL3xZ0!9I>CgnounV4nz0wWtfAX`nU2PlKjg_*xut0}sN_fO<hD?6FYvnNe3lvn&Gb
z3!(pv_!HV5U`@-jpcr>XU@SxV7I_}D0CYhbti@1~h37hiu?v+ToZHsTLR?RGi@X@x
z!@_%&P)~3a(q96_n1=epz6@FpFisNwB!u}34TU`&>IWDDc^!15Mb3hb0vH2%6Lbu~
z7|1zL?gPNHI0~H&&PDj0P;NKc$me*d0!)T|FZ6ma6ZT^0TyQ&V%x~xpumJYMP)-LF
zmse>~T+bqawJRTkK4{_R0HGxoIRd)WB6&O?vdAZ)Si_-5k#{BZdGI3ar=e>s@;T_s
z7RBT73V0Q1IPE&{HtZLm)!<{;oc2?&1vcVB+bwbf^jnLuzk?0rJJ9bfX8r7d{UgHP
zh1P)Gu-}LN3iiTA9if;-Zie#Q17kPB-UplO-*3UQPYN9XhY<c5Q5aJa#{A1Kp&Ibo
z{tDU(w1$l~hI2qZY_uy}2ztZb1?>w4zz#wOf<dq&P(L^cHrf?F*&>^ur-IYW5}@!H
z*nsc-!)JhruqAY=MP)#zgBua<h0e4nZp$oi6T;cP*`hH1VXP5`f3y+4+oC!^7g~&c
z4{V@1La}xjV^_jn1UnaczlEPAhOvGaJVB=L!xq&Eigm(>FQAWt$B-9e6s`hTJE{Qs
z3|IxbD|9V*9d-%y4U59L7=F{DdO+W@82fG5K=p*Kw-_7iCJa<B=sOl;SHlK;Z4JK%
z-bb3=(0^LgQP7XT7KGy(9o`Bs=c*jK&7yd`zXw=z>KJIfh1YrbXAAquFxDQU;19#U
zTa+If0$5OL7&Hnn|LO$jK8wP67d~WB@Er{p-v-RT8VOAU>99|M=2_I~P>d;~uvQx2
z#~5`6w2wu+2t^+nu>SaYTElp7KJ0POi5A6S6&8NB)^IVn1ZmEL&ap`3Z<uS5gP`zD
z4fBy^0u*b5VL#vSlto<#eFxyYR9rU3oKal%2Ns1o8gPCxY7+EQfOAewhJJ2QQ=mUs
z)aB3`3x6BeP;cRDL<7c|Q8+&u_JaekuY&$%(MzB>#~5`DQN(3Y)1Y1p&u^rQMO_Om
zwy5i%T`jymBN*FAKa?{AI>4fCgkoGGgAqOxI>f@?-bIeH@N=XH=APkaL=ntA!_SB!
zBP{%UDT4WC`1w)<^UbIL^fZf_107>gn3Kp@i@F_(xn$HlDCQ=DIp+J{$b}X)A9@kM
zJg5cGOD*ajP>gj1W2Wwg^0;6O)FLSN6R5>d?&mDnIOikWZot?52)7&U<>!qNZXepE
z9)@x|P^VfB<#K_lg5Cq}h5ZDy5}<7LBoy-$Sqgg<^f8Ni4$A3(Y1<310sFMbs}{8y
zx(>XBaE{*qu>RB<Xf=2r_AAhh7WEqRLkoX%9QhRRx&H<f=RKp|gnkXahmCS0KZ2iN
zb9p<#F4!nHvfE<T#(ttG&Q*naqS>G)>}^o^BZk^QdxKJh{{}6y;CnZU_5uA7{wEac
zilIzsIq)OA0XodW{x*6%!2GDcpbvmY;f7CzJ`PsE#yW_u1kb?6T8LsCqc6j*fW8XW
z!p59O(f=sMPUD=4qMuQWoxTeCF8BcUbm&Lm6WBAM7^^78O5XzA3ci4ixsQHh(Q~2S
zSv1yUbcaRHgJKM$7(<OU8LhQwtdD4&MPp4yf41nmpgcxEV|_;du;{y?2Lbx67ZSyU
zMPoh1FfNR~59+n(N+`yc(HO5-2aCQR+S#Hpez8J}##)ROgRV%Absp;udca-^Jqlp%
z^g~e03!|~NW5<9YurUWQ%n_p>hYq*sD(Fayegb;3MPr^~r&~1Ef9wp4UI`rw&O$j)
zK{1z%#(5E&2rhvAEc9ZFejYjr;C$9N7h;!#D`CF`#auG{oIG}oMZXH2VbNFzu^TM<
zH7MpdhI3qFUBuwCVzXhdgJO;ujWrUR1Lnbg8;bdhVXpMM&?OfAK6Dv)1mXXLa{tk9
z{SovTi~bnOeE|Aj&=)NH+&;#2zk%@0P_6^;bJrNoX-0nrt+wc`P_APm!as*@vgj|N
zoDS%1P>uuoD=5bS{WX-2f&K=XKyQcrtwpo`0I+s7)^%(r*adqB6z4vp5f|HI(LX`=
zg5MEd0}X;OY|gjeqU)gtEt<<{vS01#pP{WmI@13F^;q<8&@7ORaFiX-vFJab`4;^r
zw2MWf&UiP_9qB{Ro)+Bz#Tt$ujqnKc7;r4?7}Rgk`=BRSG};><Y0(Fur&=`H9zO$&
zMc%)lXImVCo(Imy_c#K20k{yhf?jTMkSKm7n2K;0^cpY?b}IB*a2@PcP^`Q7EwC{b
z@c_V@<L`sww*#y_#|ym^EP&kxy3pcajN%V~#RzW)U1D+CLzi2e4$#Lz71DHsuCO?H
z(3Rk6gm;2sZN{I6T>!=UjK2uG2>Kd$9d=jfo8T?jCD08P2jd=p7gQr0z9as=#pwmb
zI*osT@ZQjm0oJ~Q`G{`@TVR(#w_5o7i})87r#}?yGyWaY41m@FtQ}`Cbg#uZ7Ru`Y
zI76Xf&;a{5Xw>2igT^h+@zDJMbL(Kv`3KpIGXjcvYD|HBA{6t%@N=ie))u-Cnr<=S
zZrH#%37P>gKhDWe%njp=hGK3SF(=NcP|OG8oDRiUGtL+&#<8&nY^<-wqb<%^Q1pj!
z&W3Wiz&QuX<pO6sl=B1UT<BPfa~>4yka4gs8!xdq6QEaFoQcp`7KhW`3~oVstldVe
zyT-Y&FNV$o^I=bdF0wdNpi3>z<xtFd<9kSR1$48;nF__+F%H-Ht;M+-T5sWdy2d{(
z&Na{`i!+Tx=r#kXgLs^kYQPVep{)(%q(R*V(&bQ(f!~=E4H*W~gP>jmNe=5^Abeje
z8VU`h%Ah_2;oKJuB?iLxWul?Gft(a*4+EdGqM^5e=UOyyeym@$9@@u1at(BVft=Hz
zrx-{ag7Wcc@F4-{=>}_0fsJu>IPDb%()rLC2Eu)oXt>cp>Yvb=2GU&590Q4;pmPl*
zKZibGAo&z@se$)(qTwL};T}XZJZ2zqFZ6K(NiO>t1BoZ0s|=(LKwmJB<}rN5K&lq{
znt{Y!&}sv}uP1O1#1NiMiH7$L<n)DdAO4AP_!atrf%N&%4-F)E-ZvQt`+w2!F9T^V
z^HT%K8t7I7VV^G=zBG`!6S~bnxYme<uMMPchJI@x>^(%o_Xfh=Ks0b!HN@w;XxM4s
zXMF<qTnz8|MZ+EgX`aL145T$QV&G?Sq9JY|@hpi5;sD{gEF!4}zK)9s)-K@duZVaI
ze619bYy)4TMWnNVG{y?|PYel8bE<)#>k8Z-F@)#3B65a-6#5Z4%RrjT#vB8F4j>|v
z4WuxZ5sU@kYr2R845YEnBbYxx3Vp%-3PZTJ6}Vqv__`n>7$d;XjYZ^PgS8)l{YWBw
zIc!{eOqwTPKY?}*fMVPMze_42SYv?1yHFmlXHnOQ(A5UM-w_d<OMv~Ph+w?|B!Xhj
z_+ALlc0}ZD13zmO5uAsBWLxMv29kMDt^?O6^&_;}U~R0=2-YW_wF=xbG5oAm;J%6B
z=a3?T^$AE`0mb^{`zT|70{at`$z#9Sz|YJD?yVS7JfB+(By*uW_n%?RQAGsn6OiCO
z@_c`d^t>K;K6W6@Z_pnMgy$Xt_hJl*Cg@KFzNUysje#VOWvzjno={#ZJJC)DigSSP
z*>KM*B0&Q`Clry8fz+kYn1KZRL<DOXU?D>f7}$pkTuvDhKSPCq)cH`XbHLAeL=@{B
zka!B(%0S|2XlnyM0~FD81IbNLw}AxO7R@xELTDQUi6@|#OTgbFh-j99@ZEvH{U*cD
zHbu0Zfdtl0G{-=a!`d52J`U|_AW;h~F_82^yBSDhym8OU5bmKww4Z^$Ll?MDWvo3C
z_DH1Rew}0>{RZ@81F4swrx-}T3>|IY`wJ1poB`4wLa|-}@1;d_tbxDJ645ISX6~X_
z!@k<Cjp%gP(~*8J^m+qHtcB=}2EyK0L~k~jbrQV=Heb^(fX+4W_i-Y+z+l#Dl*j!a
zD4+Xrw}J3}g@`URklYG=&_E9Ndx?P*&+STsIme?s-%p_&u4|Qn#0yZI*MQ_F(B};1
ze8hdZ;CplI69n$d8UAia;NG0Ed3j9VK=?Dzbp~_JM&E?}Cc?SBZyQK*zIP4gT#mj6
zo6qY%p<Fl4Q$Ck*AJ5plU%~zgZ5{~y+Q830MRdD?zY`SEZw>tJhlu`SAbgi3qQ4pl
z-zACYegi*$=C76*(#JrtUin^KWBtUWf%FMb_%A?u2o!xb_wQKGF`TP_^iZhFK-v#Y
zGmv@*ig5>|F>W!OJAm|YP^@FV|Ci@Lxg9wO9|Ohu2c*x0b}*1W1Db0fJsb-E2uPm@
z?PMT50$N}ooeAw?AbmO1XCRGpBG%JDdKR>of%IT#Zv#JX5qO5cke&hUYrwA#paTs2
ztVYC+H}H2`0?!Z_(i5R`45Wh4I}QAuf{0-*0pa^-5xd7g67v?TH1M+>5nE&+$z!<0
zKoaL>Y^j0p>_y<7ogw)m6z3iwydNcE%MFD4N)da)KzJ`j#8wzc{s?{Az|TWO>^TD|
ztdZFB1`=3bvDF5C?k8}+&yY@qt}zgmL2+FJB)@^KHSj)P;F$nJIs>}FKu$Rn>m2a!
zK1A$211X-1%?6SUP^@!6n&)$?fi#!#g@MHH&{_jYp3_|h5|2QCGmyR%y4OH@GL*+6
zh;z0F^nii%WB4k$#$bG79AyG)?}W|Qpm9)?53F4WyDnkZ!^Zv!&uj(m?HFtS4EyJV
zjq@$OCt+iJ<G&>AUt#~6u+irDZ`R)DhHJxl!F!K=9)q>p!ET2%P0)@8!n<c;AI1m}
zo{NfoMF!Htpxq6mu{QVd@i2^MF_e#w$2f2qCm6`V`Lqw~77)H45&KRvkh&fU-wwz*
z28#6!$Y~2b&p;06yU{=j^R}<jK!i2B5Ay*?V_f%Py#dk~k9}Atfbi^5?0dsN`X=b7
z29gg!F)x7hJSgYKyqGb;IPb$a<C&G%hw%g?G1vPro`BR-P|Q0Zbw3nq1&|Ct;ZFg{
zE1_5qfHaSH*g%5I!}<dxhd}okNPP>%ngN98He%mF1JP*cAp<Gs{#gc6cR|q~KoaA&
zALlY4Jo^&+F(!bt1I3&Hezq+3zh^MdLg9k>ZzX!1g%SoL#XRIF09Z+UH|wBxE~pDV
zNLh{loW37K@SiP0o>rQuHG}YU<WFA+kj{;rrDrRCdpQQb@<hCM0UlVVgT;8zja^S0
z#J5E`ZL5j0hJejP*{CynIZ->LYquKzuYpo<5dW>DDxwZ3t0Rs(uELK85TA>5c{t8P
zTqiI7)0GJ8Je;Tibrp8Q8{ISU#`7k;q3p-ofC%f+h&TK85cR6Vi}@SjOc2%w?d`jO
zs6X-!7>^hCrVtH`;71QgGq{@QSd?)b(jA9#{5T%AlIVm~qTvNZBc>6ZScCugKjKD#
zlQt5ayaSVj@Kce04BB=^fM_gqEaJwY{IdoUoej<zPBeZhK)Q3$kMmIW`ISTykZvNl
zU^Y>O#*2Fs@Z$=^UxGGG?u8#tv<Ha4Y#n|ifpV@u*;k+qS0de1v}<Y|ehje{KYAF0
zA3GEP<eR<*KX5?#*Nw*y8j$9C=ndoW<As^{!Ga$k&#b*fHzD24>xgEftXpOf-P)e$
zHrRnlM03F0c|^DGAi4v2?nK+?Bkr!zc!{kW(cP%)?u|qXD~aww-S=v+5-+KZB3guW
z_oE#TY$aM;gBQ^@;bcW!%g~O8mJvOSbdOZyC9}mukE6~igsmvXk0mzaM-nLOX|!b(
z+WqW$qUTZfYP8|SJwz`x61{?YUo9nCi}t)Wj_3{4w+{8ZH4{IQSU|LXFVTjBMDL)i
zcTsLN>Ugh$=>2&_{|pdq+)ni25TcJz_Q&ms{xzQH6ED$bq~EfQXe;y!unqNnjlA1Y
z=6BPGcAy<Upq`&r;f+0%wQC?z-7=#3)kHsM5$#z^^b6|w6?yiKBKjTa{zRXGrFeS)
zdHGxEhIRO62zjD4c*2Y0MwGp83ekRM57B{@M2AKbHO(Z!J9TV=1fS8a@g!1a;D3{W
z@HAYNTg@iX8dufyO(fjgNo06Qc;P8Br{RS-#I;?F(;9Kv5fbg-3ENL0(QzeSklRKg
zuaQK4HxiwZwqOE@E{MY~8$>a@U{}=RYfqwN4~cG@NOVUXJrIUBXGO1K61}IAI0|(g
zRY#(942d#$ygtZ_^GWoF4nX>Wdr1r)MPkSf636x;F|?Y*aY*Y&*zqX)1mqooHjNxe
z;v|%LGTL#<dJ?0N{<MB1PDek+RFF6WWu1w%<51sOQ%IbBki_^2B+lKA7x$2E0`g8=
zP2vKis~Aq=qEZqU&m=JkbxfXzp9muSvXvw*NBLK3yx>=ipAI7ZHS0)B+e~6Q%DWC_
zU!O|i29$dv($CyO;-)MTHzUt%)N?EH-MW`Vpce_84`S{r67vvt$2jwHpP1hbFY^`P
zq0Tm3I2Pj}0`lEkL!uI8EgDba{!!Q%Ll>vw#XZ#ZU?XlIQP;9{xRF_l&BPcI%ZG#Y
zBpzKr;<2gNR3lFnc!B`RS%I`GQQuRuNjx(URFhb>hs1Np_xv*4(4j9ct|aji+WGPn
z60e{wYZ3nX5EARq?l;lCx3-d4zX~twEyTkFwEtbCt=^0m^HA3NNWU?a#0RMF1El#V
zox~<TiH~Q1MiT!*o=^G##BD}dn^%+AGLyt-?MZCS0((e&j<$ThiNqJk^9AyJxq~-%
zB)&$zZ&JZx65A)?$2_CKUJ~D}A@O|>FLWaw=al$y0v^txyq~6#z^Nf>^YAbS<?TXT
z-53(}h^vo~*bVz<^y_EXdzRy3iZXwxBk?Qh+>3hlBK-Fn+~91+jm>%z!993NdJs42
z$j?vdB2&O7+>M~Ud>6wzzeWva;0g=fHyc#rDme}y{ec3o1`pey2de<e{|kNi3wj87
znu@_XlC%mp-!t*BB|;M4Y)hRE0szOjmP!Yj(w=1M5P<NsUL;$M0S8I8UQaT8J4u|!
z(lZSYVv*OohGb?nUi2G>2Q?Z@00_%kMlyRU*oK>Tl$(RN9Ms)@HGWipayu5|<`LQn
zX*wZ)eicbvA7w#5l7(3$yP*D}UU*Ru`HGRQE9|bD@v<Q7k_wXDMw9HGN3ur&$)2-G
z_Ch^JA<fbANS31P(rS`rNY{r*_8kcJlI&j!5I<lH$#RrGFhFw94w8e%^Tq_594|oI
zk45-U<QcjU4-5Sy{ZmK|+e{MIR(V2%<nYBLM<D%)?MaSgkbmTIfU-vA0oW%Y{3L{*
z1pDLyP)G8VO(aL7ZKux01FlMvr%xa`2I<BGNuIHm<eAg(0%0{a)(1(R1w9+_=j4$b
zKN~;hLYnivBqwNq_=zj=0^lfseqC5evZ5c@PV%B&U>?{(^5Rmkn&c(WOHlqK<e7}N
zUy62IhBi$>+RKse3bX-!M_z?`u0|QxOd&aKG|A~%U;@c&yWwTP3X(IbNM1jX<PCn3
zH=^zvH<O%&ytC>^-n5bA%_#5YZTO)G>{}L+#5G#px{~B=sQ<S0B<B!7`L`qdc9b;_
z<=z4N4%m01UGu$QImrbmXTds>cTK~MekyM83&0+d3vs-#k>ouAlK0Lec^}$P86ml7
zD;}b*!o$<mBp0L2i}#XTf-;sO{j#+rAJ+JyU<S$MH6$P1gqH`A?(sZONwSIn;+~j7
za>YQBPomr>BP3TYCizqlF9;&;nJhd&9!K)o@g$!cP4f9vfVkCraI=Zymo}4pc?ii@
z5dP`}u$Sa($n*LFl5bRzT(^zno3ly2<tO<zj@M5mxuF+m#7lvQt415%D*!0-{Y~b_
z`EuhLk{@&<`C&TAk5KMM3-O>2_5Ta@zak_*LE29dw*~1xL)xv=@I}EYl3y$)xea6f
z)f8}$<kyJ%W-4x|k?%XieLoKDA-MzX_yKu-MBP83ZM9yKJ4^BMAJWzJB3VBW%p<wG
zhU6X`|AIWf6_ebX1tKJWuO|7&R+4{iCmHHTGK}L0;-UdO7({t-)Y&)=ppW|&liXiL
z@&MuvP9gbMdy<DxPtz1U|7{O8krFdV$pWyBl!D@kigKoraxKG+T^3kFDm6$d4e_lO
zlWIMlRQhoINK=Cyq&y2qW%MHD?M5mSdD|dO+exId5SF!#R5t2pSBVEZ3rV#{c>5Ys
z9Z+`1DPRw&+%cr`P*2`!Qk?>%@@JFk3@t!8g`mq?QeBa^1o7Qxkm`Z_z2@OXJO=GL
zdMZFUWrY9KLIFUV`XWz1r0WlRz;;sQ+ei)EN@@`LaLi&-Ll)o$GeT-8${4x{FWs#n
z<ww7UA?*q4NewS1HKG^5@rj%9G9B`dsw8z1>N<H8K;F^gNS#_i>a+;}WuJ~RPv3)=
z>JUF>HK{YappMj;$Uhc&$EAV*sk2G}+IBX=&he8PUjR0dIu~*0Rq(@Lydaka(2j{X
zo`^nNfVN$5kko}}Zw2yQ)D2XTy0{mqOQw;Ug!WECy2)tYWVG$l#iTCtf@P$pAnlZ`
zq%NO9>WVyo_Dn?`SEU1_y$0dakY@ToQrBv*jns80bH)Tx*N-7}1KM!oXuP1dfYhv1
zP($h_=*?NAW~01YR+73E_1uQ?um@6e5H@EMsk!S&-M$dFjH~fw62j)AoCSVTcTFYr
z52X7?9jUvkNiD?jJ+n#O3%w6*z7K7xTuo}xJW}_sCAGMK)PtyF$zoF27pbKONiC}+
z^$^lOjPf2S#fxI7|IzWJ9-B$(@qVPL(7q=SwgP#c97t*<%6MuOsizl^dIoJ<6(RLp
z76_7By@u2a&==Q{S~CP}BlR-!y;4Q$)c~orDDyRh!KbJ<iUHd8<~UMsq0G0tk=j6D
zFCHwSjCVJZsz&+mA>aE*|4;O5qXvlkU@NIjoAJd_7O79rkIjhtbT}SXtS9x^6jEE$
zNqvsApVyH3av`a0%SnATiPYDnq`sL!YWrF|0N6<CyHTXRM_YF+BlW{*P)F)V)b|tY
z8kA96ObY%)!I!8yw55I<solu?GwR#Z9uL=6k@~fo)Ng3d?@0T5kkp@(@Nf|6L+ERG
z2woaXC53O@RCFGx7|M!gfz6~Er;*y%NNWEAQU_4~!NsKh+Kz_~D6^@8G-Z(%dr0G+
zM5{HVbq#4}A!%1N>6A%$Xot8~D@nH=LppsMX*be%mgC{vM$+Exq%&udZZi<9CEa!%
zY4{JFjr{EzNw+U1-C;cGj^jw@P9>c;igc&hr1N_Ll-n8Q7HlTn1@T2_OYwTrU8j)t
z6_75;Bi-#F>F%hf2h#N1M7kH+(0e&)+^6VL<n5D1x-aP0o^*eNm(L?Ta6M`4Q}i)t
z%dsfuxBzK?H6CWcK4B*5;a~*vk3^lL(4LcKkUqJX^eM=5D#|>42p$xtk{&yO^jXVD
zpS_UucpRUL{O6&r^JkNuFqQN~lzTxQ=?hE2UeXl@Nnf;>^u<F+U*aV_X$tAd$TxXC
z9w=@jec2w;Q&9fp^GII-y>cVzsR+Be7wKz|2KO#{`ZUtlqP*)+-i%eGuOCPHhVi6t
zM42;-NzbYvebaW*H!mlR`xt!-+HmVa(zk6Q9hgdb4%#?3Li%=;f5%MHca9=GKb`af
z<h^SG>3^WSyOHl6B7N@|fVB5DlCG>Hy=Wcj`w_PoZGSKepv)zeq?e8-y=)-qhZd85
zxRLZD%SbQB@ndLb73y3uiS(1Oajxp8x08Nm59w9t&$BpwZZGK<rjUMd8|gI*NxzJC
zyo$VQM+2mN4PkGjl3s^;-rP$1?Qx_xApYIqq^p<Vg|d01H=^$!po~omNPmp-Hm@T6
z8T!8!WAS+&=`Xybztmte>93$)k0On8PjB~={<at|a%GYJ9_+yJ56JUVDq)G^!QwX3
zJ5lB?)LmCcdiQVu-7|*tF9oE3MY`XnlHR)mUofGqf1<6yd89+ACycg5f}~@MNyj6k
z8)5HTL;5e&bEuSb(^R|^)r%ZCiX7!7M-L|lZ!tN}0&-k6<fP?+9pto51)IomBhHPq
zo<_Xv<R_<1H*(tcBPR?0y%yXTIPKPvlQWN;4wdBOA}kMSI*li%b1^xE6UZq-IYlTJ
z=a17B<(14NryJ7tKzOfoa(bh_qZX2L^g(h;_mWe#nw&l>@j}8Ra{7-UXTTIZU`Kg_
zh@8PlJ9rN{IDecWb>tkom7L=ikmKJ@&aiFdoPe~$hm(WrEWQ{6Q^`4L0zll!qscj?
zJy?Y=X=jpi+IWDnPTxq*7}RyfLUPVT{;|k=)&g?Qo=wg<i^&;}{O4AXbKV+q&fiPU
zM6~GwP~j!#BE(&iPR=C6O&&$grAT+#a&o3Ll5;uqirM5`i88OOA!jPubrr@9=ah3z
zdk`UK`Vg?0oNL#TbKMSdW>k`My`P*L0_4m@x|!?AnT4>M(7u}yJ{x7-g7)1ugPg!X
za^|27bM}yPdmgAJ=Z?`}FFALith;BEvv3|c_aN_msQ`9mDLIP{l5>9zIk*lw4<c^K
z8giDREe}<Y^YA8emZQw&Xw##s$a!oWfc-eALY@^w&XcH%sHs{6L^Zj{4IiCL{fWxU
zQ_4%qGIRPC^(`q&DbFd%$%GX3Ey^h??>eBotZYDeVOhJJb_jN*xKawsd?mh;O#gZK
zD|z+RSI;{)a4!D(!aVo9L%W6BlO+nWJZ{0(p&GUZ--*9N5lA%*k<Q|6PEwJyL>1vr
z;{Uo@L#N^(o4VlXZ@-eVGG9@b6jyd;SyoQFzJ1CE^v_fS$_JnoXb;-uPfPQ7((uQh
z1v%-hTBYaMPZ6-;O=MNjp$gG9E32)rpQb*H$&aEMk5I1Q-yOGYms3{OE+wU`Z&^8d
zJ^<z0qdqz9va?cLMMY&C*YXIhKwEs}<z*b#@~BIRuYbSt0e!iwqOz7pTyjoId5Vd7
z&6AQMT3y&$wQB8#Dbohq=8nit8i%9)$5gKD?CcA(FBDZ=+A$-zXk*#o30OHC3@cUs
zhXhpt{P7qGYK3*v59{Vuj8GpQq3rCkK7C8e+Lh(-+Q7PSm6VkESb4;9a?1LcHIs-#
zcy<|&9*;r)veKebj0fhG#|<Nw80VaJY?&}jdRb}9c*OKC<FTy5*hnQx-Q4chtvpAU
z=I7zRv?N=%FH@pjPW!BOo^+QhvrUJr_O`E>>kb{W)A1j~bEUU-XJ%)6%W^wN{2zK-
zbtqHq+9QM8o#M1@+aW6l8ALTcibX0H?Z#ZBrnu9+8ICI@BV9PDsp+1Mxdoltwr!QF
z3hg8nsg#s7S9-cTT{}*S%cWecTetCYhStJ`+?*lLWGF%g!o3zX#c>YaW!Cr67~BE<
z%gXwd@ER9ou57!;MGD3RqlWQA&&3f3H>Rbva=Tj{LQ7eNKQ1~K6m%AePdei0Bu<i+
zcBtB_@TV2GZPTW0vHfhKmPaI-4mH(?xj6SuZZXd|56oHM!dcOBwp-39)_r0o%geA1
z5_6hUX6LLLJ)hRvot~1SRQk2owN}cNn(l5r%}&rY7&{wtZMwo}CSpVtq8FUNS&%h#
z_y|O#bJi2)+mWAtd?GGo1jn`J%qJ}1K^Ej~3W_zN65j}uIP5{P`7DGd;W6%8hW01?
zNq;`0;a7MTipu&6vJI=W^;8~tB&PkB!HGn8zDF2CzwuwaFpAcL@;Q-h{3phyq<^_9
zr3^>e<{ZJA$J*v_lD&HmAKtsSOlnp4+*~1YbGs*1P%=8uY*cW9Kr)3f_?$I!NqBdt
zJ6eI#F?)%#6P_xw<wzKxYWtp9CA^v5iq=x4q`K47kMi5TR#aH^i$GhfIzPv8n$CxI
z*nZnjQNi^dNc11yWwh<WwVJcOA8H*?&g0-R-Zkk#&Gp2@@d(`4rG36#shxWc8QQJ8
z6y5K7)z;la*=Z@BcXNcu$>}00+IP0gKfi~N-MbAP(sMzg4Q<<c-)Y~aOHPhRi~-NZ
zaI~c#U&}D>XoG2iZHp_#cxLuq?0wKaUgLe)_m^{vip*+G`fpwtL9`=dUc14*5+O=_
zgHJE$n42LTt<&+BBHf)%@0^t-L{?VkDJJ1#Xui1+iMH7pGdTz6%=C8R3@MyXcQDCK
zHco>&cpML690Pa{5*w(t1NbUvJc25NhyCwXZ^D=0I)FJY-@p}Mwp@Xfv{solc8m(_
zh!)tR7q1l8atcm)DCo_|7-OTy*y!Y0urs-)cs!5eHkBWF9W*10+5MkK6jy=r@<LY!
z_6w-wsqAZ$gE_Wy`wng=IheY2hq1}QT$a%)O%{qf+e_Q5X~~iFW^_zTj$~?f^GLQ%
z&sf9PuH^an1QpTHBSw<Xhpg<Z>{485cr7GnygVn%oSkOf+ABD&J6NyS0qjW=0@u~_
zv_)N;Pg2>Vr~f$or)}Hibm)+Si^*Y2T**l}%j%wkgb*@$&K}>Zw{yyG9oyN|?K&Q|
z;JbJ~G8|)^-@Inc{woDDpV(=%<+B$p<JBmrO{>(@np9qWaE=={H7~aeulG=Z!*B^p
z7d)Fh9(?_={;3Pjic2u5KC`xpxO7}^@R`k{Vb1_~g^~f-rWEz%Fgv$=R`6K$#kiHi
zbMj~ZQggvBG6#nfBr}OpQQUPv!GNyCLKL<4c=*WAM;=f6A`$R<y}51L2+^iZu0Lsb
z3ks5EvmtB=$lSlU4X&%EJT5r5W1Hgs0nVSy!|MNgKyp3&Wc+hS8i*0j;@ynB4%jPz
zxf1gz<k*4Xu5v%JvR!Z+h(PzTeEo-~`LH#9SkE5K>p~STO2cKTl@O_ZL5HXKuvO*s
z?sa@}O=w--V>J?|ai&wuIyi)D@myX9qMcchPMKr87dN6S{K0_ouIAb-ZZYTXe&I^j
z+DXAmI8>b}M5gfbi6O4$&)aY3aE`BPh3!Zc=dk19B|ol}e(XnjVm!KVP38(&*nCYa
z%>FM!OYP|^5*LR<RewLmZllH4yt?Iq@y+<Z%i8|6b(vY<{-JLe>~`EU_+RuAt5R`e
zMarRl$?ge1ckfX3|Ij^&eLxM?UKI_Ylla`{D*)ClPbXiEc{*F}0_-(v0CuRngzXuT
z+#?|K;aiR|yt3@F%<EZ#DdSs@);;nvyv^&}hK)JmcBFYB>n?5cI+lvWc)3$^+vMdH
z>f~~^!H24kxK&|4RxJs8@G&$IE5hy|xF^XgDYx04nRdY*-f_56k}>dd?8$I#XxVR$
z#cJoF=Ap-ilg=D4K*|9F&a`bAHE3{`E<$waGI-FagpEB%@4Vu^$u{>aNOjw^_~eZK
zBetOd4&pNHupCj6DYd}tUkVEa51=VDClzVOBrjE*xB7_9Di*KA9_L51UoE6=m|wml
zRD2I+R?q<KD*D<TE4nlQS9sMHx3%SEIX|XlXQ$5WhFb!7#&lQb4+|Wq8>_6_%pR4M
zm7_+9g?x{AsHTGBan86$kLGCX^V2It!J)_tFYvi@*q3&uqs`SNxv#T3yOutb$G9j5
z_48(vrwt1|<w(cc-=Fke0Um+O%)Cx!I4aCvG7blUq#Y1dyb`ObOaMl%s8grRE~T8i
zy4f8zyTJgrSv2i4`-&>U?o{}28<yM=V+U;3M&drr?%`X6h?B~O3@JPMX#T_=Ev;3n
z;b)&cyjO<DlhKQ<R;|+bc7#*-{rI%|vth$}55ixs<BxBZ&UYf*2766L<M9N7eNM~s
z(hdyoA$Y_xapft>#E7vM<^baURvg!-Z>M}A@;mkIb6f@Yy^4n@=}CnNvayMj$RzxS
zs+0R_9*5@ZAfAD={rkMNxEny_|KZ4=oQ4{^@@o<fuqt`na(3*(4;&#p7Khu>`-pZV
z{m$R+O<K0b)Nc#?4>ecYhE@HoR$FH^#(!{goxCJ@kXt^dfTzGBwcg_h&%&XYjGE+#
zAJemE`YpDu$}C*o(}oT`ujL$#Kb9|Sya0L~Cn}P&Ac~6y+Ln1dIpYTnmMy1Dmh)=j
zY~9e@<T&w~=|j{tyMN+FnVmImGVMhhZmqDz(cuS1{mTlEycRW)t=N7Fs>a`ulNbot
z<3kCpZf=7gZNPnHiz{W{BkrdTUpI>o-eUC>*b<XBQ-Y2>M&E36_|E?D6Rl*!kEp)~
z2E;yxu}|gr)Ufy3*1IQXGt)dX5nOX})DvA!+*IPoRNK^iW0`2Bb;w9%nrX-X$N$6J
zn}A7jROg}<SxeTwZ@pJ_byxK^Jw4M)jb<cNBWXr6THDAH0wic5Aqffaut@^MRT&v0
zUBC?JVq=Wu#()tI-gn0fHj6=pu>s>}JlC=>HrLN?Y+vkaf4)zC@A)D7nwt0jaWX2i
zs%sVrdC{n|G9x1+Gftdy;%sq3U537cUOYAn8~oNCfK1=K0#-{_=WySqYJz6~2YzPR
z+rPu_pe+E}Wcb~Ws(Un(h;HxwoWQ30%mZ?A@8zAC-+}=q5&Io@8u&dGM~<bOY+&Kl
zNGKso|66FIBkk<SkM48Vfj42$M!7?a9rZEt#@q%pL9u-RksR9<Oru(#g~8NQw$UsV
zlOO+hvRHIPfp<hRndmzlNEg-51+&jQlMRMT2ANdx>&fKTyMN<+OuH1;@BS0WLm%_a
z=yxl;f%Yzrryk=5OWg0+Xch`7WjpW47gF7SH0SE^P`dk@YED}3wu|9VN?p?ZBlEUm
z81NGwOdc@G5fyBP)C;~dAPr-tqOrnJhqUPLs6(Zl#o1X16UEutMfYq+O-!z>A3C(Y
zHi`Mmd%lI5^@WAHJmZ{gU$VM7H90x8x_XIwX1<TKHpM*IeRBkPQjM7Aj0#u)&DUm*
zz%7|3413xV$h0s=7&FP1OcI7DwgfiqosO}#bQ~BxslJSFnDn(^DKBa%bpz&d*H$3Y
z62AyGKayb|qXg_g0uH-uwD`KE+Oe=wG0vb@PAT&NYGnNiFUOQ~1WcK_%N_Pc<!_4N
zK^ES)&@}z~B4jf?7kXQ`>5_~14pfao&(Y&8^t<Rg1bDhLo_D74(rXW#gHnrU-g)_5
zns`B(KRh>5R@U77B`5TM)1{(<Id)p6z|OV<-}f6LejqbbTfSs|&f3{#HC6{r!{+fv
z(0!lq-L1b+Y{}5hwe|_T)PRJhsbvFzx`sb;7vL=Dg#R?J;UK_z%)9*8EWsuzwR9b4
zIO>fisJU+6K8Za2fyBfCOO1?;mdeRQD0#4XBv&5E2LqN$Xv0V>VMCA4=kke*D!c1f
zDtsbVD8>_Z$RCKs6487<8Vp`kQ7=l2<s;EZHVImdrqUCWGp%wl?hnJZtW*%DO|}n%
zHZaHvCwEm223v4us9-?1%GUibc27-(Qy5d3$1!ifwlO##nNb}!1GR-7$n$Wv-J5ru
zQBn@Qc-qH&<M3Fo&C>8Ek7ZNytO^=7a=g6CBmV2R?HIjbN4{al=o`G&x*_e*Kd)^d
z8=C5J`*LOswI!t9@M|i}Q$m0H!$yGLrf3i4HkeR9%VYOz>LN()LEi7SDb%YrPJ{9e
z>oUAUa~ID$ilu>f0gwFqXnmP>-dTXLIr!Qe!=+OA#@FgGgvsh;q*$_l=!e3^V&o*=
zmr#%M*!{MeAtiMGMAd27V(o6bU=>uU`#<>8W?LLfY#;h(>PO*+qRra1_t|1kC(dF%
z^6EBTT5YN~lhDZ<-uX@nPBB;=-uqs|Jan11M7;OCG)f5v`T3t`E!2*E>eGf*eDtHV
zSFnXof66lsvlxGE4spK2t36_lB$eR~zV#qY!%^^q56QrPh~6dWR$(~)9%DK0f9oe@
z)PM3@yu%v5jsSPUfVSgz0H5o4pV(`~_N6FCVJ0NU7mKm-+Q3@*NI2g8+kcj3yB%Jt
z0x!k#n#rZRpOFtznOsU;gLAwOb080@4K3tVcq63BMo%4TV)9>IEkT4a;=fnWvBe@H
zaG*lhvxaKxEFU57sd4n^Q^2o{8H@f0dta@yC11KluZ=mE@7tC;cT^7@s?NxH3qPDE
zDy2vwSE=L@krJjaU@>fqloMeUF=ugkf`Sd7TDGiZse>bDi&A7M5Lk+oM0f3Id%_Yt
zpKtfjce4j+Y>K_FR-=SRS&GzdJUX^9repA7j9v$<O6VR3se~ihEQiWbCcvm^Od&Yf
z(0Y*a1JD=px;ur#__Bjf-jW^+LS_dY*mC<{VP~99;%A1?x&u$D&-lY(f7XvOT=}!w
zop%;E@ofek1h2dpM+a)Hui5ru$Tn6QI|Nl#=FyIISbZ7R!(-4yWFD=d`;-C1JHC!_
z#^?iLw9m1tBAW{i7bq3mc#y1=nk~&uHLC^LF;US*(C(AeEGzhwsPe^P@yNaC4=>l_
zs{7fZWfeE|p_4$5w~ZFM%?6UuortnuHY=TZ^KvY4si|Gm=&oY`tRt`^O<?bxmhT4N
z9U#2AwgWia1s?_c#b;@?yV0=4W=1a0eyu~$(eo$<S)`wOBe$QXHSIt%=7dIn9RTu9
zFoNjmWazTJE4k0p>tCUR<n&RMYNeDubf00ru3KBfv1qBOskODkVd;t-D^_uKWJKw^
z@T8#5uTiwm=T=rOOD6LtU@}W}zUH)sCRbR~T3fr0?H*oRo0`I+e?bm8K7uAjMrL`N
zJ_syx$A1Cn79DF=!{`d%OMPzxb!@PMP4-F8fG`G&xCg3ey8gTscoTSs^^E+a?c-Wv
z$w(tq)o7|BKv{8C+c=C4OV=ElAbr3wC(O0YU<yLaN@R?<lu1}CD^oxiHKFA$3R5^$
zsxq~rt4^v^N=-t5K}5!8rBaa&6XwYl*nPB>48vctX~Hi)KAJ9!8?kM?kRJW`2{!dH
zxGiQ^**Jx})mck@oNq-S1=_8~R|)IYc$KZRG|J*N-5mulTq_c3rE@t5DTDBQ@Lv|5
zN_2q+Sufcw#Yn2?J(1`K2ImUE>4BpiDLBymLBaVH!TFSc6aGsN6uV?sfUT<de9+&{
z>{yih=6!3v!@g^=L%`k0j{gYYD*Py;S+JPRh+?ZAAFo)qt0*M`t^=z})B1-6^;&7Y
zuueqBc|-pY&vh+KkLJ8sQb6y93Bg$x{o62e&J-F4i6a?%a%ypLYSR8~N0o1}wA-qr
zR3@cTTj4PAG#uXKPbzbQ5a8bl`AMnn={@CQvAjo)UTy&80!^Lq5&0HI=_wyw5G;RM
z@G<ppB>uq^wCdNf#n-9a!w)~qvk%52kEv9Lr#Pfx{BcfIs(Vs3=H}?y<<SXIkVbtr
z?9E@Y-mZSb_c&JZe8Kmd>J{o<^&iMTfo65kJy1@Q?(j3W6kGzZgiC23Jdl_;3sh?M
zO=DWzhdDQ2&`1m7he%2E#o2z0AXtL7V`IdIoOw#2*<f(^P5R8=D8a|Kf3>*U+&xg>
zU$nV`hs%rbilNQBuXOLPfah;C*VZ@ot*x(Nl0zg5{@V|S|2|A|YwP`=!4H}SMP4Y7
zLcPud3}C<1=xQZfduN@P0<n*%ed$n;uC`zpu{yz#LNJ<2ghTmsG#QH^X2gny!s&1z
z8B>1n$6%l^kxKbP>2$0R@Tc&ovdTGaZM2h?oel?6;qbT>N~io*D3nduv_nQ>SUBmo
z)9H8~&wP1hWhER-q|;t(Y#D@oSU387d`~!q@3BLvNTl~YnPl&KOdr0_Pg!XTlFCs%
z=?4)(S4^Y;lUOt!PK2YWNG2PO!~v5)Fp~}hl7NX)$*7-D(f>g|VYdT?SUMd-3sD&C
z5FM2p^@mdFus@v*@||Hz1*10Pl0>918gL)z{yiIU!Up6!R!CR;>wZEv*!!etZ19r;
zS-kkdKWW4t9Q>p(Ki2)Ec<OujocDT%h;#dy@1OZ1^zE<seh&Lk{IU;<2ehqcOTQs!
zo@ev`X_tS~uu_nELqn*y=piZc(sjy!yJ*=ZtPL*(D>M!h2Iz95q}?M8&6!EFaMgVx
zkS~S<sZ=%-ib6jMR>$mQE)xxp<&)VoV%PlsWF%UMhH|hXz~o?s?D2+5<pbejB2x(j
zpe_eyCt$DS7Q%iA7=@^vk0onXxR?(FqVaqxrtjt4XS$zRTg$6bxy)D`9@`Y3_@_RC
zPsyhgJ!W}3T8JmJ(M&X%ix)>?@iaas7$_G5p<E$~PszpYfU@{W2sXm!;B}QuLK?`$
zl1Z3Qi(|^NRnX2AVnHa|_~-)q1N(};J2`{vf_|#|E5YF6SSlV*jmhy5`5=*b>8BH2
zrv{n*#h4`_A#hsMQ=_BqP|c}g_`sdCs$(M(mXa7I&nm@S1k*U0jd&R_8qHO^TCIxp
z!;$=NeZw2xFp-2(n@mjH7)Ola%;*_kdN1a;L=s^Rnif`+CX2*~7U__v0h8F&^_H(o
zU6ga9gv53b9g)kyn-aF;!Qe|<tuKO{5$e%u^*D{5N>7~NWY_6UxLBv5`+FJW*LnDg
z??Qhub+a#`akzR%Px}DpW)C5xZ&1z!f{`L9?W2>f?4ZRd>adJ1qBzBS%~?Pb=#SWs
zodd*z16FaRXn9W6W&%7o3U+z_o5>8)Eg{O*%l6dvII7C_N0w$D>b_62p(hR;*wuz^
zu6vS=u(6xTR)oil&)}CWj=l_nQ9$>dG0rbTwCd0KI-qZv^w@1uo@=!lybhtym~S!U
zqUjh^pazEt&haJ~4~uEbFH$BRgz=pA0+HJ>_dy>5K4J1L(Q={N#=K=vj0Og+2&?l}
zVDH=xyJxvjWNN~)Cib>+w_g6bIA|=LxY(Hi1U|L}L!&)n<A3+;zi|2ruOwlOv6AkY
z+Br*k2Th*lz@3tDJH_)8a{iTYeku~3nu^wyy7XRtKs*Ld9{zJCM_u_eDtABE{}u8*
zzlXic2fl+}1aWt0xzC1f_IaxU{X6TMg)i{cz6Y@f(w}jZiMf3$XlZv*Uns2~WC1KL
zts#6#;|0|%{%PzF!E$WQ!GVDpuAgfbvD?Mc3Z9i3jO7m5u!Mj`IM37}1CfZJ%L%cd
za-Q4-uOO{y(mKyMUrW9NUei$90&qEQc`v%#ecU*cnH(KmzBrLwj${h)>+EPG_<~3_
zpSWVtL5-zyxm-YmzM5e1-Z2E3f?>nFgsV##Ep~AO?0hVfQC4I16iNCPSPEkg^6WvL
zeVAt-?v*`q>oNj!ZPh1&^j&nH_A+O+a{T!5Ms0e;?+-VZmnUb#*7f;BB>T=tFciJF
z5RYWvboD;z;FSvt3wB;vYJWZ+&9?R;tROtcx+T7Gc4V%x9EOBaC`?2m(1U3;h{cBf
zPOxLJ-yr_sbJjc6*Pu7u2F^%%T{&}c9kgj<0sz2uj3`Z6Wb>ji)oPk@H4kJRA$w>F
zipG14vCJlVV#h<~w~<T5)7B=B@3a!{ER9|pkA-7}UtA#9CcD2T7&ulQua3QV^o7~n
zM7j1K)cxIesW)}sm5s%D|JApoFbzcf8U6qgUx=r3sqPAo_hb@p%@!`UqlswxqpaIz
zqt@P3;dtRPJ6*oL`*Jng{cmd4N+fdn-VI5RZX}lCkAiCM3-#d9^j!~I&})2c-=pV+
z3oTs3p)nhrr5(`p-u%vj3(5?nrwwgWUDu43bYT4X+VG^quWDe`qL!Qme(zGdq4wu$
zvis-b{ov~UMzh%*(Q`!wD5Ym!vM=>Ch=C<4wOJ7v6~$+PZTpLTF#XPVzVkEvfc&SW
zr6rXjmqV++@}Eq5ot>Q>DFge&|5C2?_YU;;4k9pL<lBPn`SrfH`hLs@$2Jg$b|(Ib
z8MTNh#jHG`oGJ4YjjcdBn_f^J$tJXE6>O#6sP5)_J#acuMNnrCzyjxuCf+Tly>ao0
zTd{n+*-ZN53(-V8mw!!GS-E?OY@1ebZkrHHFDnxSD7`&lq^}Q>QSh&9O6E+-+^&+H
z7p(4|L|D(BJxB8C^>`{5PgIirY#urg(K=kD#rOC~{*r5jkI8Y{G}wE^1E%B&^NO-5
zdBBwTf*SAN0Y0z+8R<^nkN6&dHJ39n0gk`@dI?L7-dnid;buhHSq25mq=LPHp&@wq
zumlaD8|I`3an%6=C4!5YAoxi(7>Gtwx$mD!0|u)XUE2r#ZcB}-L$GXHxfd+Pr@dof
z-EF$rGABJCzh!!Q+8>*?Qd26EOr_(%##}l+x6}vi(47bOJ);Qm!1NRpfF8Vi(zMu9
zF`_+)f`vHty$bL-kI0FeefI*7*O~wXn4%Xi0bLM#9}u;OP3JvlPD-3w;O;vEo@)=v
zMib@+STJ@?^OeZ|8EQLt_{?AY#b1oCZ(Lzu7E!0Gr2}96>Q|%r50m~Li$2Pqhm@Vp
z6yB#&iQI~ytQ>c$Qy)oYwhf+L`3t}B3kPd81Egp|6(d^s)*Qp1k0-Vf!riCw3OtOy
zvEjR4_$hjQcDi<H*>#%8oYN$cLlD6vC81baXW^CB1;xmW$qMp;c7*E~Ih^qcFoJrc
zN6IBb+{0*^L3J&3y5m4`C#9@HAqUSxBom2%d#cokK8?q6g@Q#k#{2Qu+3xqyW6rR`
zA;hiY2r2@2P1f7yHL>`Gy=L$;w+wl2_IR!6LiE=(HMjLSYZ^B&*NZ%7^chI|wYrT6
zWGxeNeGq|N)e=Bs-|8%`Fc<(T#uCiaWznK6SQUC>RD2z9WT)<l#$vJV8H7A0D}=8)
ztQ|F#{xw%8XwGkGvpY?Zjt^(=n*}PfFSAu-2P@Igvg_w*dqZoVHYR)a7|7>`4^`%t
zJ>YCNV<z7KZHc>$)MXSsQ36SyL=(qj)ofBzq(86LFdjZpcQ_dl7i%?ECkd~ATf1|j
zCnvMUfwN6EEl!;u?{rqs^Bwe~f-l2tv7&9VDpk=g8yK<+b$)V-=2_ep*G>125q^8p
z(HwvdMn(c;(6>-ex65KL+il-p`~KedPage_5}D|CC6T^qYT1CzQ&SC&c`zjQf=XDu
zQA8jQcsf-ED0eVfgT1gmdpM1MQ2Cm?D+DNGFN8$`(_lb#;sXf~|MY8kMFPb?zKVov
zQ#;Ur!`H!JYW}fKLEDWr&S5P|$cAMYy$~GS58ol|!*T^ypudd_65ZnutUoOPIFW-(
zIFfu@B9l(R0X9)9V4BVs(wPukTlHA%SS4S~6pOGV;6KZNRO@KTWU-&5pFPR{U&oFe
zi^b~j%7rrN0v6ZC3$+O>v4#C6PIHHTC6u|meqkz=DdsD$GxO6i!ydGw_Q81kLG<T0
zA;Lz=Y-1ubIn}6RYP2kyZ}4ZsBYt)zpG(qkgg>B;?h(&bDwj|G`Pq2VAhR&crXK+D
z*pX~D>nG%F%fhScjj74ZI9Ci$lq2+};T?Q8UWZP3-<?aPt&#Aj3=bmS696e+(3b~I
zz8d?P#<joZF5mk|mzsqGLx4#heiEY2s4WR*S|F<F=>VpdzK4VmiA4}ufNT8^K#QRP
z!geT{PqG?X*X@an5+Wr~%|#=9A^UZIu~b=FECm9sbb5tguCT8Nx3L217rYz@mgMrn
zQl(TxU{XAu?^Rt%r(4@w08+e_Ot#`czrG9A_J;F+a;g9=uj2Q&vc+O{?{R~rZ?v=7
z;{H#1t3HDCL<oc{Ozkgbv-TTJ&EtFdWNQwtxk@?|inTJCRutxJ@0FeM=zlZ5BaryS
zSEy-T2WwDE0h|C8kglT`+E8lXtwAw|hAKQAvy@jtTAgpI&q1M`n40qYr=}*}_{vD`
zK-NF^HCOTb(skV@)#2`wm>=MQR*8jhaPh8v{_I8B$YpJn$&PbsI0b`!<~<*FwYv|z
z;QQbIzCUVNuYIj$B~<9(r!#()dOc$|s#uS`4S8!8EAgnsI%CWL1B|g_u;YYQ)3MPq
z#Q{SY$jPr3GX&>AB3G@XpC`i-tS}RT=JVj8#ZtR{u^pJE+c*-LE>~$}BI}8bWV2Sz
z%4H`$Kas_+mCcU)<n0xbH(d2a%RSQ}`jz0ctO}Se;bFAVR_XNU&P0w2CUX-zqv^DI
z0OQyHsBVOERe!&4dLH_%AwpjgrZoIxv@QNUo6f$Ab32F6%I4x&As&q9vepH6_UmKG
zR4O?(Ix&$9g_08!qmZRjXTGTZ9%DNJ3+7$E6PVw`(g03Tka?xZZ^g{R)p?Zg@NdYh
z12V^mYl-EYvle)dOegNFk_g#$=!+Rdk!7=cmv0&!y=i$bx)_UQzL+?l=yyIg+Y=34
z-Vej}rkzUVSC1SyeA!|=zIfT;BS%*Asg%9>nzM1~3+3`Bce7)d*^PYuQrJ7NLY%@J
z$DWd(CLU4+VMqB$ByppeaK;vLGWKt=IkFzfMT~pBXLInTSNlXsRRP`$_Ty;6jiFVf
z%dncmbZ)CnxovC*o60`GI|s^`-&-AMfI2K+hca*VT0*NPxVckiO?F(_P8I}tXW5og
zysO;lcE~EtGo&wx4e|l^;=U1Ra<}S?7{oLU4d7YHd(m`P)*HA%fB`b%C^E5-)hd__
z!6*I)BT@@k5lcj#L>v}do90KSz1G4LUgp2m2fT@~OqRgtm|olrXOuZtEpv9pXJVu}
z=A?qj$m7A<_D*ONT6?eAUL1>+%TuN){*Ko9&fa_U$(CUQ=9lmcVEthSYb!g@J+AZJ
z;(Hf3mb;RY8U}&Fz5}cvLYc9J+R1{&9LB()8tQPJsysm$iD=8oolg^(A-gdxvpeVW
zCYxn^9oRk6j0kV%V|tOJwy~X(ex|S4x6iT;9lH7ELx(I?0m6{Y7Zxva#PF>XMVbqf
zsiOIf)5glbH%9!*&SnBCG0S_&M#NgmT2{7|wF-@ZifgG3Z<$UWIIzj)`I^?gvdkmJ
zS`^a(Y2p81>szLIeX@pNeZE8=;t`bz*ja1ZLdBAmPcOxim}>B*bfvTu3|euTGU3}b
zw%xtHOQ8F_9PqBY(>T={z?^!Y{F;?IoCT`@6Kua+kw+u27Qk17dc7Vi8|y9yiou)(
z3<v-7SSOpUc56``O-Ey)Ym%egH;yL5HS5M3Z+vAc6pN<S7@|zHzix*TOKZn#Yh3)s
z2k%(3rco^xc|GUC`1lvDFGgZ$tFyhmX+)gEt21r()8Fgoy6=a$8g!tS>*x!p3iBF!
zOH|+=WBT-rj|f^KN<(RzEGnZSL+VX02^QYdlMUFRr;jfRg~E|+=9gpPk-7Qu$b2ZY
zkWLj2hSS{-r^BTK>Q6rKfe(B$lM9DKArgXW<x|}c-gD1AflQ_v3T2YsPG6CE7&3J5
zj(6J;h}Yxz3(wBZzBG}TS?Z?lyYIet1abReZ6t`Q@f&WqVQO?VsGtLS`rm;3^TL7t
z3oQx%E+X&+k+|qN$G4#<$Wfu<KhTFJFx+g)ihs}*RX)I~SnzcEwe@n@&B9SS<cEkm
z-Tlk@=tb&S_Yc%k_Yar@#zs(eH2Xg@sT8ur{9=!P0K2`NzA3!Eeys_1zj}SSzO;XB
zwPCI7>HgT?{oUUcO4qEfS0=~dD28)0J9_xbhF_Uq503AH{CcDB6$8DO8Hfyb+9o6B
z_{(hOIfI}lTvt}_*BglQ?Um8ufoqK|hVOfz@ugqNVWHmS@<JJgps^8PE12QR_&iQv
z%2+Hlu2_Z<*UJ6ZsmSrvhYr;1${JBuW$e$k{+Bn7%L|kbSUElegURxAexyboYt2-O
z6`D4z>8UU9bvbK%@uF>2uOAoz6L??88>OXbqBp1iDExsx#aW$5t#xV5HPJ`8q8v|c
z$uo=7(nCh+%>SYeiHDn!+?v<+3)X;+hdYCK>SVf(_Yp0TLqo9I0FnXKaHUT;ADa2!
zUzwT=%bM8mNFq^pLbv1fL}CO>pJ8i{MkXp#CX~5SnFzBi8fEQ9On7&!;nogMPF4IT
zyt7){sM*GM*xxv+3!0wgQ+73emf9WfYz@b~ytR)`k0Z`_msjaRVCY%eNjN%VXg5k5
zI)EdOGp(FG%fa;Wv=3S7)Y#eJInZUl+WPZog8bMz@@V&rQr$DD;*gQkV<FFC#>#t<
z{mmsv(gV##BEwL?@?ZUnA*$1MsQFjF^E<zj82(rf7s{!W<;ZaKEzz&^pzlfF?}I~7
z6abDBM+MV?jz%aTE7TN0=z=atEff@h)1(Q(Y+$hKvevVRm=iFDq>hr&A)u2OsxvYn
z+#V2JegFjpAHTcR4zLVtS_dvgv(d<Sf@qf*kKjT_2SF}E!9cCCBx@zd25PMsQw?pF
zLRsY7t>H%Q3%5@9wbmK(;(^b*@J{R)!x^>k-fK3Wot1vKDtE%$n#2(yH(m!TL>}<{
zIJ`QL3MfpFRrX6nFf<`<iV>hjU1HcfV}gKQ_y=UdKPh4IH<nS}a3s`TYx|pQ1u|-h
zt~jB1piZy}<ELOQ@bUK$A2hmm@90s^#NzV~W1%F@gK8oLO>$%ex?^fW1xcySt#&FN
z3&h@y5SJ(>8hz+~8&|PdG^!6zec}_JNLyC=DLedxk+{B`iXV%oGFOFI;7{9jh|YdH
ztzR6vDx2i(FYC8ti%%rt(+_xC#JT%2@IxJ&q~Uol;Q7<O{|w4RHwBvc65U*ck}IPV
zniOYbnqFbTJrJ&_3v`F=XTBXtX$1n;&?G4Xnv`+8t#JhUvL-^hbC#2VDc~6`8iGA&
zq9-GuO-MTlU=t_ti;D4606oAMPw8la;ZRor2kfwq>T3R@9Id7|7fK}@hM*J<M-apr
z0YCxT2!eMb;iustfalRFhPD3s77WUCJn=*^tH<}k$Aj4LMA&|cBij8B-c%}I-~Ceg
zm@>WxMq9pDe*xD&`YKjSZ7~PetXFdrslY7Rpp;G70y4E%#$aq5K#K%NLRH&6;`$^N
zpjSOAYS6a6c<csOwRI}q@V~)}^dChX+s5f*_?-T4Qd)4=;fgf<IRj677KjLFg)^<p
z=n}2?>4jIwW_E$!WP}xPm~(uTwx?*ko<}?bv=OtKs3%W(Who$=F;MzlzAo%Uylo(W
zUP1>TGW8Yq=z9cITi|hAc;0aVVH>M}$hTd?NRK7z+@EwgwWaD+Psu1eb(J*K?tW4V
zde(xLbM5>9=JX*lN9Z=>?OS}WWDgtQ0$w{1pzRew0DBTDNSAhE^jM#B-t<b116N#U
z4TPZz3ar`J>~byZv6njEkyw?Jav_LdEiC@?=B+7~<>qR_`TCb$X(3LuQh6y9sl>^I
z<g@v-K&2u-Yiy%Gx6N8#g-?I_+H28-V;92mo(^){TnS0;7JB#{ONOzA3Hy<89?Y#s
z7Dij8Xd*JYHrNZ9%4!J~GV(Oe2z`YT|H9(hN%ZcjG$PL9*`v-UauHH8Rfg7qMM@~E
ztgN&Q*KXmwVlFD$2vLc}(r;NP-j|KXD*~ztADqNYF+R(-SlWioLH|eQ!RK3YxwT~q
z1Q$9Q7B9w{n>PavuJw?CPh{R=O9U(+gRo;bYLl2|%MNwYs6gBnS_STCnw1HZEuf`=
zoCZV>zJbu>G*NSe6=yhW$UdR+Wtd?EAn1>CCxlL;`_VL3<{PNs^u&pi=Bi>81T19N
z;zR{&awjGlv*VabCk=QAWpf`RabI!aEf|wZ#{g%m<$)7;Cv0DoL$PeK$?2O4697Sr
zlsWT*MM>AmoaD?mF)<Mw{bUTWsmt%8G3CpVv_FnK278GGoxl9czZ^xpXe=0ft2UQ(
z{|d|4;sO5`@pT|ovF&PJwC+2K(b=GF7q2$*ub%24vm$aCA0rtxkjqwo7(S*-$Yvxf
z+$xjF&{h&cJ_#tfTZrP0Y-~iZV#&b4fmU~iI%^A`pnSq`wpj3F6e*(?7eQpvCvmUa
zL3LKbkLFa!NaU0duUc5s)dG*11&whvGMvUDC&^+iVT~CrA@n>|Hn5A+Ji#Y3P)OwU
zX0e3H96C}V$lf#b)HVwWMUJxJbp-e!Zu}vX%BzimLAA}<)zSF$Tc+cGt5VlSmA_n=
zoGj$I6d+&Qzi+1O4@R++^6dB2LsM{d!3N|SUzIk)PEA?X)RPdnqd_!UK)RqzCO?hj
zVr9RIUW;&m;WuP#E5NDSfK!J6qg@)lwxkLN>S^uTc3lS#S?}>+En<<=@Px!<^1U8i
z+Ji`+C5OZ8or%GG&SZS~k&!$TIl_1zo7vl17#me;bZo)UOf9jxcKrC-YJv%VF}k(z
zSOh`PG0O_4>}Nu0KE}t#!nHe|dFGkebai>TIvqoYXm8)Kd>HczEX3C$`tmIpcZpW#
zpu-U7;H9iCkc==03V=Q(GM>Xe$S|7dk;_mF6M9yaN!*BxB%KjmPtg8?leS#Tu`T#3
z$JSZzNP?YHCPts8A??=a;6djp{<OBYx1|l|r-J)iQo^6DUI68e6#6Jbdtr9_g+R<3
z%R(3lry+ke05)6*OWcQ+R`B{EpxGMu+;+Py-|75pV-l7E=jUv2*^*jLU4gC~`HJ36
zFV3W}bI&Yt_khbOObuLKz~J*H*4S}QMcm51)rR**Eu^h}oI_Zx7OZmd3a2PX>{6YZ
zZn{Z5bc=cje)#0w_kH`8TX2s5ze}DFQV$(f9WGMAk`)?vg;Q*+htxxD^s|(?yUlan
zYpI8}cMK!hA?<QL*wHc+Y;%whMFA9B9ET)C1rh_paw(-+#E+xIM7srtO00D9(Naoj
zl0;0Y0xcl!5J5atf-r-koDmQb;6U3P)EIG*aJChTov5o)+W2Fn3P;p?1Qt%jlUV*K
ztaxKVRt1ANWHO^ccpNO=gTFV*f>8h_#1!Mb=Pczs2bl4(tbS?=YpeKN4z7&M=tz!F
zsh=3$rrh<mdA4tv@}tRtR?qdE=%8N&&-sdP6<IM)P=bLEPD3AhgtiYsl@uvM!)Be(
z<v3uJv~7qpPC-k+R77<|3c#)Bbxc0cUX6Bd#2)fl42w1taWYy*Xo=X1AV(#V@l7oL
zjYHbfNnjCANoGOTgS@KDB9rb#pLU;lR@LWHb}(`b3l;5Sk)WNLBVv9clFyB>alT|E
zmydjcnP}$J2dQ%M@4$sW)a}LF0>8<ZI$G`sJMqMnJ*X#OCKmUCOiiKyoCCC^NG&e9
za5I`bN_xZLfUsEd%S$$zIw=JEunEY4YixsU41klS`u?I73M!tW(*_W|P-%6uKn6j-
z1Xx3uQDYF04K{S;VCj_<bo%O+O>2_Lg0CZgFuv!nAVTUA-|bk}^CmD2+AT#7&@?{)
z76OQJbU*&u1|GCv?1rTmhD7Ne_#i0BlWIS3k!axo3vm(XlXemC8>SyvNa;n{{wr<8
zU`=(*09Jj8xCq4Fjf>=%eORmWR`9{&-~8q`2VuLbp!+|}J9<EdTv<8uCG`rdS-%nf
zP|B4Xw9vxL6M{r2+_oqStsQ-}|BBR4|3B-4JLvP3T3yzsI<4T!hO{bFUV^O8ChCnd
zemj&%#-f4PW%CPw-c;5UB$7$noX7SzC@N9VN@q7xQHo0p<umsYI2O8Pd3iana@onq
zC6`>1O1>bKP#;9jzyRy)Nv*CPU7D?rUTI3!g8p%HzF<$VWr}if`;Qg35<n2v#pd=(
zz#6^lv!DI!)ZFaEWahj7@DKl>0{`fzk5K_`JL)5l&EQ`I45&o_nG7^Q*ZhYP{PV&{
zOm_P@137g$uT9(hqtTTAh_^Zm=A}z8N<gqs(MxXj=5F)m%d`r1NM}&8{Z<SsVzU|e
z7bB6y(NQHS;pX@}+i9LPf#z1EU<GV8z=TVo6pU?=WHWL@zTVi6zH0l8)q-5T4h?@W
zhlJO%VLaxXscF4n&qXDv1~c&SBFjd#W#;`;u_MAOluv$-#7>!e2x4I%3ppmps;gxM
z%6?K2M~RuPuzst%gPgng^Wi@o&kenD8_~^|b1knDyBueB@!wI`6ZwK;vmc`-+Q2&G
z2=UO-x<q3rZetKJ7)~}PdMw-?R<%u_B41RQsF{>yvy<Th3Z<NE)*CzG$Z<L@7x%gy
zE$~Ul_GL)*Y$m5YKwH~Vj+Xj^!ku!ti$2?&TQFz;mM;Ts`T#KU)tI;GPh(&)dZFhO
z3OY7$lAhh&DkJlK(CZ1@!<gAdEhdP>5FrxqI)o`Vvc2yP(YXh)W|6W2K&RneD!l^&
zAo6ra+tBLlRxEbJ4796HcH3KCN8u*T2HIqUeiCfjj^Hh*czv1cv6$xt>I-~dGy=r8
zThZuSi9~~n=9c&IQ^wShmT^lDUbns9_L*q3*E(_T%s&DCz*gTl^fX#HHxN%icp45`
zC-7l&#jzRG`WfbM&Mw-(gISyM4|EIHygt<YjtvjgsgYPTb?+sOhRXV_R^r!EQT6FY
zwVKc6^3`g?Jv&4<#_TNJ7c(>7bM;9r-Q*rl?A?2Ke#TNO(e(8@xzcava?{iG`t)=T
z|CzIUcRw#}q1%qkPFg!+=nOsfQX2d3y#|E=FGR#4mC52&&$T)<SuVw!_sS@2?uitf
z_208+D_7gt9l*(@`s$qH0v%F>?<F7G(fV`-E8nq=Syh{JXvEP%f}KN&@u(EZq8;A}
zcu`xmyKN#Zo(D3uTs@~<LB9O-Ht%Vsw4<4oYPB?qkBlT(-+@&NusRO>5S~O{?}d)n
zHsOH@t)M;Lq-~<^)g}OCP*(T>OK`S%QhraKlHb#s(QG@~VoM5J@`&@}+;e$o2l=0m
zBA>kby=G4HekpmpjHR}rIJSEV`YiPtQmA&WYMG*g_xGDqc!jTNPnRXd?*c=JE5?@6
zx+PK`nbwxtUFH;;SI``4FdIocIv=B<<;OgxlxLu)PLGr$#97V{I~K=*d6BS*Dpq_`
zZxK75^h-;Wq+Ok>uZOV?<{JE!{$V_9ZI{>K-Zju)m9qv7G8;`RP0_CFKs%a7%DFkj
zRHV1c_UPS6-4Gg^N<@p7%V4B`51*NSRly4W-^4Z5rqj8rkLnQ$ZcKQ80(BME@o%RR
zaaFoI8qHjdBy|{Wl!u>b9SWW@j$ibZYZ$)-eC=R2w=v%q){r6tl!DYN4txqMcAp#7
z-f-j+^bPEY8ntHedJ-h>a_(!L`E2KKfC44J?5NT8H{;hFQ7n)pwP`-}cEp+5u|z~4
z?#yH!L{A+;E)<X=Z)5%rS^>f0BavHqb}P@`#IrX!Wolt|eik-)E=dF^@K`uzS0V|0
zP^T~(5Ssd6!|R%e*fD+BN+ut)1A&;MwR~uCcBV98+l@PD$;48vSnO7w-O95!@$5}b
zS@-=#HMy{`@(Pu55-ICv)$d0VD=}LiL>AGx0Oue%T?rqd%)aN&28ca8Y)qVW%nq1B
zKwLO6(O8aUANYIrWPr*SS37+&ht_9xIx8r(t+QqY{7y7k>5HYM4otA|dD0H^@BS%t
z@@>r%9HW`XM5YA|nb;gl4Opm_Dnf!WQ0_Scfal4S2lk4N)@s^J>BG>q$QbZ@AZa_+
z(wS(ry8J$q8FVKTvP+v(Gk;nt{xp>yoiD_AEKyz@6)aeqoV4#U2}ie+#R5$Ratm4*
zPS_{#gFl13_!08uzU(GK-y)TtjZ1r_crTA`svGrIEwJ-`N?urz@2$wyr<fdxa{1$q
zckz>ttGj6x=2P!~i`3ngy4!N~miMzhpL;~gAMwn;_kc$mvoGn^OL!OoXAA)AhNfr7
zEKHrldoB73`eDXU*CPpYoXF>;T~00Vg_9rQY+-yn%I(0AU&|k>Ntt_h*`iSE^uZW&
zGkc+mOzlTqJu6j}a(2txm#@t%vmr!GCihG@IX7Rf)XN13SIk~Y`B+{e$MvfZEiadu
z4#LStOL+wTc+bv|vMOFYu1(Cj{c=2KRHM1p!S&bM((rsD#9Yns1o^8`QFb0RrS}_z
z`$@(-AWd_~Z$BYliQ|VFrf~XU&p2H+Fqg`=Q11}2!;^$i8(zKlIeLPVBob%7#2I}J
z^2MAj<;-zYXz1z*(+mJT3*Enz&s5{|HJCfbLjjtge(SR^10lx$1$sXj#-mV^!kh?d
zr4%4GR?;`;N5Kt#A!!*iMU6S-aIO0#-Zkb#AqH`F6>$r;L0(WqL;xi0>gIf-iB{cP
zs<kjZJ2DZunadUw!irzLe#I5*ubxLnaW0EEHZoDJPcQsB9D|7KQTCjwsSn_yrsizM
zrH^z!p;9|!WH_YZF>p@ttV(V3f`@i#1efXZjSH1ZbPQ{{r^jNI%EFC&XL_X6{jQDw
zunpvmwpux|&WrUUE6TocB$u;w(m*z}Vy3iBUh$8hB#-$Lc+77w7P~KZGbdu`{`eE!
z-^9cTHjr{P?vjE*eN6yMa0L)2-=%8itVq!Y!$wmrb014E60D@MD=U-bNJa*mh~jvy
zuQOGO;XI0GLOOcML7ef4#hm&88(^)Mhr(3_(<^Rno|cAA8_j6(+osmVX;%2q9TuE_
zrq-UR$>0Y1PTh0|xU`KGBQ@V1=tqaK*5OXjfmq*+Z-Enm4zdS>FVca9mH#GhDsU!<
z+BJ=44d98GNGN8YF0c`g;(vpb^j5Gfwz$4U>Dyf#-QA*8u0``!OAGR9OMW3D@Yaqu
zaQJC$nVX(oRVOKcbx-Ks=IC_NvOQYvoowsIwsq^<PFvf$<rChp=9}5h)2rAmNjHsM
z0G}3_^iN>v_!uk}S0Q3w$LoW_&Nl0q^eKaS0p)EoUp1J+41q`V@F1JqTe<l}MMlt>
zgYLL(ciBbn{?!!zyZ<2mQ}5@N6dl2fwmjO_We$VvNLlCiyZ^ITY~6H|*l4v+|KIQa
z!6N??AD4lNFYEE+9_Iy}k?xArW5jkvnBpJN3Hc@6a1OeqD9BW`HBV!q#zpc&P-ls=
z%Ib2SJO07Do;Y&!)vrE!1e?a@A-b!0e;MaO0Vj?sov<Ga+mY1K+iyRbj@XeS;aDs@
z`XW1_w$WDi!)VhAA)41pr9w#ElUH~2l{kYX?Zti-E)*;eRS7GROuQ%(sXj#CvWA1U
zldS^|*Yw~}=-J7rox0pbY<XT8hw$jzv5J971QZCjg?G#ArC(dz0mv^U{`lhJq4t0K
z!WX`<RVkNa`27M_P*t|{WVPkgce=3U?g4b5Hlx}tE{1K5XR&o1`Vb&*(S5icmcBb+
z^>Q^;(gDBz!uvzp-#Pln-qB`*-su?&#OL<X+t<yt`S}tOGEt6&q1wq0o-57IubueP
z<BvbSxp(UEC!c&$sV5(w+PnF<?siMMU6G&GmOPDAE~AT5<O)<Y>~2-mREtFw<Uz0K
z`Gxks4o1>Mv=E1NCUMc=FejTUq&KKB>TeoWddhQ6$Um6Z^<1s5I%H*p@h>IqQ20{^
z5o7MIG5%CIWG8<~y~3<!zU=-;=AtmRsJLizF&JJzeus*zU*5dP3IxL!Wg_=O=yCjC
z_{MEmC=VhZF(p>xcIUSeaUe~Q(tuYD@IGLlLtwC(55zigS+;p<pr%$iAwJMbsZkUS
zt@;4PO`QR|RBEhoy9dl&p#zTGgm|zO<MUflzRk@|hBPrw%8mElG}GS7iWdAVWV_x|
zo6fe`5F32AgR<K@9qLVNZ+lC(fwr6hZEs^>x6M4fMYvEl*s)rG3vFlyB0UM*fr9Cy
z`|D~@B9oiQGW_a1YtGk@p3L81kr80=P5eIdCkSV*_>rWu)0o3gAv2X^qvDDIazrht
z0n+P)G=~r*2efl1F2}M)KO4qG!u)7&qZnh1MkpV69)VSm4doSZ2Fsgzt5qufF0_^q
zi}Ljs8;QNuUkr=;F;2bwdqRGYm$c{1ylv2l))a6wFSxvU+_c*LBl98$hC-wJ^qRC+
z_9#_bic1;T#RTI;I~JoPDZpGK?u=;!nS+#O^^k45(Xob42X|*bqLXx5ZE5_*p2#})
zjLvai0o<n4+mX#eRkC*ZOJ(dV@12BrzgM7iun(I;EL&+sZdIHrE2Vi=L7I<a$C&Ze
zRf71oUJG~xVDRW@uK)y$&Wf252tw-TE*Sk+Ce{FGLYfCZlSzgY;BYQjlr~&G*1MwS
zx`p@N4;l^lL!rc@?pbSXW+tGb5#-JZhc-5@W4)58H}9U|`$dbIW&^Cym~c-&gq<qa
z!`v<<5>1A$TU{jxpRi5U*ra>9ApjsPbwKYOSfDo`>)iz?ggibO%65#uaeqlRjt>of
z1^QIbe?Uu;s*9>V;76deOroJuz{g2_AzRD0<n2hjk>0^yZ&%%JYuhWNx_foGXUH}W
zeMYtmAS<*c9oL)|d11h-uw~gb3pD1*k^c0sUA^9eJh)Vk-r(f2hiXzq_C@38uK+$3
z%oCTx)Bi9aL_Zd$^C8(&weu8EMuue_76h?PNn#UyM=3nYY>cJgIUCq#tMlCd+j_ko
z4AzqtmJ=u4vo9u-HKdzKskb>N-G|k$z3Nr3D!;1St%bt}3fyp{aFK0aBxeT{GTIL*
z{Q&cs_BYL!bhnY!F;zoyl(#W?4NlPh0cn4FuxU1Y#y6}~d><VoOa7WK=6ff|P&`e*
zqh=GPM$G>jyfzR9?ByKStVh|HD;MN<gHOQUf4*yZW8L16*Z{l(?^6FrEAGwu@HGTQ
zPE16nf>SY^!KySFL|As5XGbI$|44kIQJ;kEH#WUAm6)hECXj(FKD|_lPdew}@G{#v
zRGEr~5Atj|Xvfq~(Q<pUK7b0@$V6;%p~{B%VBh%_vJT)#e0-drQ4LghhEtwROz<om
zT4q}p+mUMZAkUV=AtEnj>W+F+eHHl;t|4V}69i3qlxb`YR-6d4+MuZ&ERRun@++bb
z!_JUZuoucg`rTh@?rZopZ<l#f6g?4fF&wbTdXg2qArhW9@u>6lrI!?gfpq;9CFDh(
z@M8<8q;31lM-ygeIsRr+&4oSgzSHiRI_Y(L9`^^Q!WN=c5hO4*mR*dMin&FVNtfQk
zHre)DCa1acj%nM!k?(#EHrHJ{pK~5Kqh<GxIIHk!;U-rjP`m<!6PXPcHUS}44p+Po
zTV?-@;FaN}eq$yXjatmH9UiruIjh3JqY77A^CS(l0O#fp&6TXvt5+(`hLKE%8M@g@
zaE!AkDr_^Li5r(ycp_yu>Rre2yDcSC_0iD~ggoGf<9m!_N5wgK1AZ8pd42>dF#jV~
zBYzy3<|(<mHm`njm{vZ(!f7ajLqNnt3IF9}2=djpNQfXsZ#y;b1#Kss38b0QbGp^v
z#wn|qro4kg8TredQ>%E=dvO5q`05*o*jE!1D=IYl*MP{%#Dv;$AKr3L``UpBWg6`(
zhrl+pe9QY@ahvg*R*WRpll|_+=isw8=I3+CQOg=l=H}-&ZZM*o`Ne0+>DC35^xEb;
zX89(;3$`$i#kIGDB(7};7#)O^X~9LSI;3}?HdCcjyb-QI!yZLHEa-5RpuVfVMb(HM
z!?y0&DGqk?gXKL49ffURET0dCrs4DeBBtjA8;#8#bGL_Kbbr(?H^WSr%tmf&*hn;$
z9jn@-BQvCeE&So}E$8<(N8e=5M6C0HpiWhbvw&Eqg*s);TYtVp*$sj72WS-C-LLb*
z)Ant281Xow)2t&kw5q`Sh$lK%4#jvZgSC8AxGUuh)70rWa+H><jDSs}47auc4bbth
z1d2y2bpn<6L`!<yVnb{<5ZLCJf{Y=%vEP^U^?O)e$W<Nd0k&<trXW&EcI0B7o}3xT
z0(s|DD0(9%mPq`*c>IUzZ`F22^VQ!O$o9Avk6d{JmcH1x#uJxoTjMtH&U|A7`I^0>
zGwdrRup8gwnBt)-YyL|XL9R+hv5x78M{<jGGXjI(ZL(J)3@cSh$p!{6hWAb#xDWP#
zGUK+km3u48bA)#VG{p5B^qUx6fka~L1Y~}^22@RHpGS-AkPsLA+QG28%QmV@CSq%;
zYNpE>>`G4F?ukv1po8!@{D^wZ#)!KG{fKAjhG0xzaDt2l>{A`=*?0^+aP_sp8P4c$
z0~TF>o|bTL%$c^%?Z`)-kawQwzv4~hAh2FPWY(b!)N`IP!3ok9Wf(tj{1W&oQoage
zZAFI3(J})~6gnv2EtmoQi>wRr#SC`RbCAS6P^|O8RcY@8v9f6=rIrw|nQBLH)wW~b
z^k=PN)*pyZSC_iqSgKCP1L_pivq-lC?u0cItsnBo@esK4PSoH1-HacZ%@>f|5h)iJ
zn4hnE=JVL6=d18Y{iL2V^~h`g2z4X<4_Gm+bD`x|S1{eJV5Ob*{sBk0RSGf2U{GD6
zU6I*U<PUgQun!VzFyn9$xWiWE$hHXs{A-r#$nL|H>T-z+4(DdW_F;dvFc(f?FCItp
zjHWZ;jitTVVTdUZ)pR%zPvkQBLe@W@3oE~6<qGpr<a!GPkjgKH9UFt8gde-_C@Y!8
z-mvR<G>Bl25gW&RrclUQ^Z5ui=DAj>`9cKy)mYi$Tqc4IMbK6)uC_AaSV*CTK*)w?
z7VyP+BwN@}ljSmu!TAMb_eZ2vENll=jC;%`v5Qa;COd8d!?*J*u#6!X3R#hCZf#<u
zm?+dIMn|}L2|P@8D46&!@+8AegMK^{3Z+V62*L0lI<n3zblexfr|}Z#aYoN2Y{_MK
z1aIm2m>suGPRhYIArCTHHi)}{p8|%0YgTB3E&1WqAg`ie%mJ{CwWz1#s=(g8M-G&~
zE_kg@QrYaD*tUwtTG(ACAIMtm=KM{;;G63z@HS<=^ic>7*;M=;`NhQpPncOBVjmb3
zc`&8gj*txX;4b_)$4bF`A&SN9rQmeUua36%wzAn5ak0dVn(9PuUBUL&Eh9EKu^>M1
zf-T>>eLn-ba`~Ak0(g*`ltfOzU?zAfvaMGPsB3+TO2h{1tm0S8#aj5&;s3(zg2tFg
zP^0l@VP4fsys$|S@zBOHVC<@5mEbL{u1zEBswx&8H-9Nt#>VXM{m0`aATzX7tAZ+d
zRV;ARl`dJ%g_>On-E1JN-^cd`x$j%4mMS3a;Vg_U6LCh>qEarDNb!w)b2^6np@o1`
zYi?pPQA8zjzYDU@Vos;dwgYj9{>fEDgc<v}U?{C}G7FOPZD>hH?iihe*tc%X(S6nR
zRIiP_?pkA7PKFez6R@8c19R_0oiIh+ZF8o!I_9)jwtb2g<?fETp(;E>-zW?0Xq)>!
z`bH3#MJ<anyY2&7HgO5~qbX5Sl6aoQCRy+!V6TSCWPjdnJHA!jLj?3`QkG_kkoz$$
ztm3sGohOzv!BFgICUZ0v3TBp{Fm6+c`fc5yuvd?Lr74vFOFnm{ZTS=GhR1@;*&2LI
zRam9PT(L)!$)n5_tHqrjwYSdr;B!r7uwx*ec8mEKOIz5@EhJ+NVQ?FE>}lUkkdDwl
zTBxgqstE-pH95W4WA8UW(*rF%*3-jqAaa5zm~c@OGRkjq<D4~5**1MXxlPM#-3z8V
zx12zCS_6I5kM#GA2HCV1EC^O;dHOlx_LJYwBxf`;lOu@erx=n6Cs}P($7YJ|jY{M!
z7)H*0jE~{N-nb8uV`YrpQIkG)TPM@oI}qvyI)zU?+3U7GVJ8h7dMC<<QvS#A49|e4
zwFTQdx_X}J;3GQFsBh9S@9*?|047ACQ|(~DceBgdYK2ZW7?q;&zV2W?l`sAk*<)J6
z9N(?JmPKorp~zdC5-JJ$(bvEe)GcA104YP5gi}SDF4?qNLR4V2qOE*K4<4MEo}M{)
z@ThyXx6x=|=c^?4s0rA?V7lh_BSUO17W4-bb_~epk0P-l)PV#x6(0%s4;PC?JK&E^
zM(joGV-sr>QlkJAXEz*v4MI}@VWnYzA{okqr!ab3A6lw9eeoriUw+BO(^cg?{~36v
z0~9xc3D8I?oB>y5Hb?Gy7)=H2(O9H}?O>4P#)f+JtrE6bgN?+WL(am09Y}TmjIZY3
zJFsa`GUQL-5J+-AS;mR$gC{55DpqE{oUufG83{KK73?~MC~W~`+C8KrjtZO;u`MKX
z4i+#}h@$jbfRGr6Bige|!ZMD{@Pup~P14aJiN&5wbszE@J<mdDT}w=iEma&n5s&}2
zdgSFVfBAt%_nDw&YGG?Rr_!af>yQ34tv*OC@JJkX4VaL;IlLaAFhxD0kVV5{P;4Pt
zrT_8n3$MEBs{2Cq)27BOmWw(yF0F>%`<lOCTLHAmtOa}p__*scsQDJ)Cv<&Bw2p~B
zE%Oo1ktWDl4KCR!OR~g!sGp!j>Yy${ElMnHlqE>rn(j3Z-$qy32WDz2pPvo}omCWA
z$lPwPG={7@KoFIy|Mkw=r1Lf)(nv4wdBL99S%hwBuGng~x7_6h?iu)p6Hc|_p7mfs
zooWK>Q?K>C4!8v7O#%k%2GWpJC0VDS1fjq#LevDXg!c*i(#bS>3*cFU4kHGvOAw6!
zRsyc*s}wd!b~3AewtKt!*{QvIr&5uqN#}Wu&f|%V_*$J7XJkDgpQJnj2yWsRKil$S
zJjBX}bl+3h`!3}q_I$lg>RG|E_nyOfv(irE$(0p6iYz8891dxPKRw5_Bg+Us-1hzN
zzF*a{jA2gfxYrsH0-zvFYZ;utBmRi4PeD;w7Ucu7FD-s+nL*f`$P19VsY{8}s%25i
z;smm^B&`zEI>%6t5%MWw8C54BU>)!z<K1Xt0Ev&qQpeenJiWJ8)>}Cqpd~+69uG|$
z$$UCAUe+5=9-#d_R-Reb=D<zL+t2o~6|XI<C$THXJN!JoskZk#ANX~zrTuQxPftGj
z8=1ATpKoI7;kb(YFO-aVild_KaR4o5H5+htY1&YF$kzLq>Df?oRt_yOT;u~`14XRU
z1U>0yq{${;=Q0e3O>l_aWcELo37&PBhC-qGa&^x?|MNdv>a36TjqcQ_m(@NvSvwD4
zMFr~(WA*CptG@cxuVNJjZ=4rl0tgak-rx@6A3E4m@DATYzV}0x-o4APMqd#(7`$i#
zkT^tQC2%)&)glORp25hJ#bD|bhKmvKOoc&hFdFdnFf46jy-{?fcZ@OVRp*xx#(-mC
zUOP6lk7C}*URlwLxouJm$VL;DO;^q4W;~8nUln;crWRc`?4Dg4jba~<InL@6mms(E
zR_|5hZsTiw6X&kAaa`lxX<i^!2$YLAX)?7#+Bq~byV;gHonaHmB6cK+M&~HHj9(m$
zg0FEMix>WnP|EvP8G+o{eQxU%`Y5a5#ASKU+aUbRFwhulf|0qNmqF;BatGNx>v=v-
zLL1@xaKKRPqJ@wA+m8eml#6XcPibb#f?Dek24ru}z&(8Z0Z3!$s2rNXf8G_A?)ksq
z0DWc_{|#TAIXF&tVwW*moSuK41=kJD!hKxpIZ0GM*a$&bUNPM33mE5f2wcydZ6z{Q
zbz1O<?O(7w*228OTyXdMeDG>vudP1qW-u5Ib2lafN03G2u~)OfM55!njl_?s*|{$H
z;n{`!Dsx4lAVM$eY4KHOlH3vnXWroC*1RYtTl_IMJKI7gy_*q;v(jm-Xf>w@iqlVF
z@y{FN<{J=7>9ljUJF3~1wxKyVC*<wgU&XogM$D~m@co$Y3BU;??J9r-5oL4Y*>ePs
z1}gX3=s{@TD${@TAcTD8#vc1@6oH>Y0zd;+U+n5(5H+1`d2dn`AfC3J-=}$j^Wj=T
z6K|2RT6jZxw0HrqhKk;F!HhEl@lCwsyqIhE#$*wm&TFBu|A_BLeV+mc+32%?(rRd=
ze2Oug1|dcwu9&fk058=G5l*RMV!ouf(<`E;<+(%+Xzk~zp?BcynPs?c-sA<8WojS^
zo$hphNmZzWZP64uM`^lj%%j`74CW;`C-oui{Ne)2t0_G-w6yyPL<dJAn=%}uZxBR&
z0~Q>zM4dD*ISH3Md<{J{kSA&zRuR)YCRF!bH2?Nq=lO=d@^9hId<kC9AEa_73XRB@
zU|#NGt?y-^qXop3+=m?ZzF7$dJs;7K1~6=OI>PXae;7(g`%SG;<H%nraU5*gj7-gq
zi%a+ftupwB#9GDG^`LG^D}p*O0MGN^x?X<beC}XJvbDv9S|L{)$MOijAL;Pwg{A%b
zrjYqJGPQ62o_rlHEQTi}#*4W^ZDFC7%U*|#0153d{9$m2<t8TKX$6!|PEKHRNEAiF
zAkE}c&d_)UjTv5#jwq{~NM&ulj4zF*uzegH{eb|IH;<&SqEY+J{5Enpmn}6CkAIh~
ze=cYRvBgn5zq7)Ze+OG}z*v?*>E~c??~TE2z7IlI3u!=!x<nx50Z0QT{vz6e{23D_
z0mL{6QqBdiInbT1Np0{s4BQERdYT9L><B7CuR5B^=Q9(N27Dn{0MX%<VI!si-Tq4M
zSq~c=Eok#J4_k5UTqdwoIXFcD$Q0mcMdO-2AKN`{kDcZxcV7~ZKik2>JcBha*cToN
z>U7-g<$6`)J(!b!414W8PTtnH^G^S)>&_B7WE}`jxMR#PngGA&0;7r%1W8~#`)nJl
z-f9T8I7>C#duP`B3=|Fb8E2AS{f4lD-xgMYV;kNjPhd~md$Hm*F!b3x`l@yNHdpGa
zcgouW_AT<fsjqRft<P@@Z~BIlnV56XrniZ{oGN|I^;Ppevgv(kU#+u$Y2P!HjkV9u
zYQYawKrB91qRdvSRim%FYZD-X>~lmatO|C8E1E=%+$HX=S(DNFoVBz~+p0+B5SchE
zpZsEpF<~_?1352n6Tz(8Ioq9H0$+LRQF&^C><uMPc_gxB2x>!$`2gV&@<pLv-v`<l
z^)>Vkl)vQrrsiKdipw$oz{UVnCnN^!c6}rO=ZlM{LAp$E&8=n3!fUZ_=t)V^iVeG?
zp8@)d;h)S-pM~&x_Y6cf_$<@qv(Wug@?DG<`f##nm1z3l#*eg0l&s<(dGo%X@?NL|
z-fZ;UKN=ewi|2EZ8HCjVwlu|#S=LM>myh?ZdA}ad7s~mO{Ivcp7xHmzSd-^(-COrt
zkfhzOaK1~y?0H{_U&j3~`K6`%iq_Y?-?8PrnDc(`J%~ohMdJA@tiV35G{8r0><d^|
z<l~XtzPc3Db&V?vSgd|25AnjKJY3mVM^*7{p7)z?a_9z=TwAq$MM&%S0z*V*mMqDL
z2U>SB&=whukFnB0aE?3%$$gjHCS_bjDRLVI4wK?6GKHC_M`@5@ffeb59KQvG{;mhX
z9n>EzQjZF0$WgD0(jif})*({IHEH#cn?r~LW%F3l(CT{>O~5ZxLF~W|_{cccD%|X3
zB*feS?k`d<<_n@M)^|fhTD657VEg7Q4~7E+1(!@Ay8_}isI$qAOq@_e4u*7~i3fo%
zbtk1BC28&S0h)6;I5lwPJG4+?daV;FX1tod<R-ZCG7o~QF$b1PonY3j>2=FG?8n;p
zmmEIK-@$u-I;;(~wY%bB{~^e$yIyx_;Ycd8yN%AU27FE|r4Kv@Ie2m*?g#+<p*T+z
z@FIrfDr2mVPMHNp0Cu4NF^`Ir^&^lV8lOOzt=<*D(N}xA8r{f(Ni7`&i0ffmRc&Hk
z%?n20QKN%(|BlWq2d5A%&Z_}@(Rh{l*n(HTU4bt3THm(s57A$<=Qh<^EeASzc61ax
zS)GOj^KJ-XP;dlNL9jr_nQzFgz+L7v83H`O3xw(Mo{leeYBfC@Sn9=96g)L8dVwl?
zywx$~?c2Am`6#LgX?mMw(Ax^X8sEKws-7ky?DeYzE*9g#g7%gGId@Ma@;z@#n)e{Z
znA}2*fZtho_9`Lcd(<9j#9a4Pea$`hptb&2^>RYzxRb*@roN3&=x_Xcefe$VEs=o_
z)lHk#gZy&OyQJ<J9~R&^-|Br=L>hE#8pQ2s>?waIvMzb!y?U`X!u9}3>e0jlKg}M3
z{onxfq`AYXvA)URd+OX94uyc%03p`lU@!;~<Opgq0=^Wy!-M9CKaGYkZPPSgtb5Ci
z*^@1HtA&*Lg$g?JBFqI+U`KQT#!Odx6bwM+a_JV#7U^8hWELQ?c(0`nvvWD12H)-A
zbl0=p`?YOp*R!7-@=<85IrO7iFiW{sUM<(CGhenIQcpszIfMw@SNPtBU3EZxfS$2j
z5YDvXOT2bzLmN1#w^Aj;KMVx_IV`7%3a)sgK$CiwGw~m$(Wau?)HE!H5e^fFtE!_I
z*dl8i_beD0#s;BEmx||4uQ#Y)noQmw9~qsU-M4Rcc61~W3?|&GcqsIFFFG^ODVe;q
zPQ!z<T;bC1cLqx`8LM^arGsxla_V2)UGu+>PE8_h1mgWBr$*!P{;O-e^sM$#Zs~8e
zJ2_Z_H(c8Pef(mG6tp)%HS8c-l>6-(?kG%37@8P*I22y0z-Y+GVdn6k{V~U%-{Xzm
z$Aoonx$OO^e(p{=T62C+mB)G*O()l5s6oDCY7sf-8QtKBEP6%IJWILmwyzdxLE6%b
zh;$|TSiT2&+=unn>yWY@@V%oqV;Htg)bzqZ>!rJAL&L^_cV;ditFW1D#?ZfJCp{o)
z_5<*seKL!BgGwlXER^t%>_j9o`WT&kJ!12!>d8Ly*Z`aymHy3M%;|ReaQG9#Zbt}L
z0~r9Q&j@jbEA;s#9!r?1%MWuZ-{oz%<9(+i%R1}s9UYKSBw~24b3-A(NDN~XWwo8g
zC@|WhzS|HTG`4+!R_r>MghAFrtoaqD$TzBA*Cz;2lAeQul80$bFvubyM8oa?@qm^t
z(|c3~Xqd9Md~O|3(;#q;I^b2HLCH|bfL7uf2?jP@tpkGuDW>dBA!{(p@JmsKA)KN=
zSU3!VAZ1yT_+V|xp%qAN)272Sn!Y*e4r*=ciK6?x>?|MR!?TdNP)FOI27_JEWXCw5
zJE%mS!nGq6^rxw_n@B&zLcM!)I<j{m1WM-o&pGBqEMosj6PhY&n!6&kRGR6g@MSu<
z3A@UrvD%p2H+&-8(%nLkvi@Ipkoq0+eA|a}F+if)G5zXP0hd_rFcy>Ge)sK(L@trf
zCFAjAE|1d$<G)N_622GsuGSf5^-4aP7WvsQ8&PhiEC!4N<$C{QhT|OP{-GmOP}o^*
zm+xIZ0MaU2ZL4^*8ch{PuNf^Oz7Ty>s!)(W(_c>kf^kz@v|r<(jp=g14$sXY&B9a$
zTP{X3Q={qZ_+&Oa$v@M3b3L9-j~-(i{5v5(^f<dQ%uDFGiO$AWcRKEUKQWWyY)SU}
z+>b5qI7)e>NvN+E&b>}jN4H_WzCvNLl#S+cr4kakW=oTW0=8b@_eHu{<hdKVbW!%I
zVUm;>jF<6vF2>9>M22iKnSJT7#f4s7poVe04t^Er(2h^V%>u#kqAhqs>xmqqb*$E=
z4MS^|7-{N;yo<5s{tA4Qh|RFSbuRrUYj^fU+gAm-_C^=Wa!7L(Lvv60WkLT5g>1pd
z%8XRW$QexF!?XPT-@;-XJ(5_2t~8&Ky^C6>ow3G%yB?3OHrs~7|4+QC+u!W|p>7b(
zA_MW8`N*5$CI1vpD*CXAJX)Yt(C!Wph(#E1#yaj;E$GzOL!-V;=ZO(sM6N2qH0-~g
z|8$A)V^ZkwQcA~O2X$3)iJ_<lSm}GclZt%L+%bm>@!G?b%;hmB<tH{ed17<3t^e0q
zB2c~6<^wHhsAZUv_qyE@mJN~D{f~1}h;$D7$~N-&T&4F>yC18A`z_NX`Lp$3Qv;yz
zY<eu)AYW*Vk<*P`@Img3`T6z3dxrw%gKkubdo~zOwC%uT7xb(p?x)l^HVkOpr-wz!
z><S>VhrqS%d(Xc#WS?#9&~pNAco+nB#R)w(=nf6c-4X?*Xr1f}3>t>J)U_^@cApdO
zo*lyHj|qDIpnb0RpXU+q=MMOB;eQ$C$sb0h$=5PY(aEtTsRB4Jkb7{_ryfNm)YYiO
zh9I6r@=J0NS{2AH^_7kI2|m8sFeO15T@;}<br1+5Mt7D?^F+@uP1Dw}UIC>W;dd^~
zOk1hQ_|gEw)r>?z8d=%gJiWQ89w2AMgPSIDTKMa>{CF~1wU}~iB%DV6t^qcz*>1(<
zxI8<FCM6s~uP!N;rdLqXmR9+*LcUDD?w?wh!R~#D?~Ma6l@1CNp)^1=W>`7pNhvGv
z+zB1mhB<UV0|_%2{^5y`u4|sq&!I%G?}N<AOqi5CeFkb+7cc>f3thElW}2MHVE2>U
z1Xm)g(mUSgNzd<bJ?Tq@0Z;m3Et^wY(%%_Ak7nK$!wS{5nYr3nEE;<oUx_!*bie&<
zeXsGjmf=Lduj+Ns&a}%ZTN}e5y>J>P2n03z5?lz@LiW+v3m8Hc5`@}Xu@)B&uN+=j
zv{ux_<kHF#mRxQDPzE1@ML(qJzkv|#F9y*`HOF-otj`C-Vb<q^#{>X4{-H>7e~3n8
z+&k9g*jM95?IEUlP8;+ojD;X-Q&cnO6tW7m@btvArXISI7Spr<hYu%VZBSU(?a)G`
zx31P#O^}cd__E#5Y`#Teu#NCnqI%hM_hx3x67p?Yg}1Ho@hWWpxk7AoG*-yj(`h>s
zu#3n*6v$QTDSsFyonR=Nt=F@;U=Svrus>C=<bq+VP{cYtJFTv{?6S+!v1D!yd#nY*
zIc%35%oihxV0t`0n~v7v>0D|gTbZ6prb4+wv>1$y$Hy{>F+^0PlWhOS<HwH|Vs`G}
z!JHi{9FJuS1%IeGmsO=`Abp@1RE4=9H+l@r6;-f!ARUO7RCcZy@?&Xp>{s^g-J8l~
zVzZYP1I0}%7RIZC`9?OD&c_1rJ=t_DJr$fOWU>eoiH36Psn}RP9ZpPT(+O-Boy(AB
zp#nNF!;7%}T?>BrBZ%C7pYPMYuYrO!1$x){6{s;8BoxZk7FWbEkyDBUO%$!|j%#WT
z)F6BGJOGRstg*Wd;nR*TLn{xlow!~{5?tT_>4I$<EZD%P*KLnD4`Z9Pnk;Qmz^8bx
zrSVOVux+I7l!BAUT%0y1eWjaj=}-~iJ&4SHF&laDI2DEi(dYpLgr788#b-xqwfSrI
zXVj)C(R;ZltW)hpSEVl`;wgQW`e%H@gr3V0m`)V}DVp>L;j_+(2TuYILLO5F=P*}`
z5cQaT4q;4HaA7hl&YHxG^`@hNJC}HLu!dyH={+!{#_l$rojAT?0PGHXDY<H8GLuAu
z?v4k5eW7HY0Y2p&2v2!meRaYW+}*Q55x%2eDu`!3Y*S_SUo&5;jeM3RPrI1s`LwiK
zdeSrEZmFm41xoxrAIAPH^#2*(Z~S{4`P|us*>y6fT}uuAwFZA@7_sfrF#c5re<(K^
zu8^Fa|4U?xYr_0@LHFkv{qw6b|E8F5F74;v4k^y5EQNc#p#kggaYX0ei@kB)!^rdV
z%m}?qoP)f6Hzyyg@mywurm@1WV&go{NLm8Xn%0mpH$3<G<NK!|5w}i~5mQl;;7zgf
zZJL)kQ~UG7-6^|wD>Ns!Q_P#)o4;GI<L`m?bv<mRuR|>QW4@31zJ8w3Vh{u&_OCpH
zKd0p{yGQfI5oB1MdVZ(z4u=yU9=^(8`0ysP?b+<e-l_LLKO^ia+~-)ytm#wd%`)`)
zX5WwdKJ5Gac}ASlo0BPbkoqo^J}J$KIfLfpKG{vAe8HwZI|tPT>aOeX?8EkI9t=T5
z7aNO~D@(h_Sc^&*Jj6ExVp=AGIFy2~r~+qF>JgUkZ-wm`qU7&akUdb}ipCKj#yU6|
zKE`}h9C7Cr&`z<XmX_GclmTu|!N=xG_2vV#2Y}61=GWHN7Wsi&%<<(p1#NSjzl3~x
z9qiY09C2W;L43x0eV_3CBF4F2L&L6uSZKiI`>?kZR-j%%B!VqVWb<dKuRRz2th50o
z%(2&Ki;Y-lni%m$0Ykpg^l2#SjxW^d;3yQEk?58g77>dH2*>$hPT*s4PfrZ30H^o^
z$jK3pO-yb?A{&zvOl6pS;BI87*k3GIuIBY~88lT6BW>880|%A@gG?HOCohMa&MWq}
zy%!!~yRXvg1A&!Lx;j5!ua^RWQoTMuUrmQrZmQSozdp7wKZo&T-mgv{-QLfYy8nW6
z>NyF{)_S;10Yb|TooW*OLCG;UjK1C`#E4Ne)JVb`E(TpV(S7o4G{nVu#27gk;?q|c
zj<ec89`}9bJfk8Tk{V0qzmo3Tzpg@3AHt>Gl-&WuGPrxxUJ}h_CJ655PwY`E-u&e|
z$j<TYR5Xfxwi<Uf=!_aVp6T4o_(XbZhof*_HhR)Si7v@|Nbi(4AN%j(?f3}&ZhV9<
zeHQukao;KEwtwLJ@4oM;fXb?=0Y1uTj(<}t7hMVF6UZ}HaYaRpj9g#ugYSK?gk&tX
zI2upM`PUHut0!FN8WjesIFq+Tb<o^K(wDipA-q5?JzSeObRq28!=q2T!{{m9bW6bu
z;ZsPT_0ZN?t>lpQoYNa2mqyX<X}L~X5^P#=(h&S`k^GeV8GVf0gJpe29#45+{Y_*>
z?0rwWca}cko?<Hjw~t$HZ`<B-Y^HTLdxJv@6An-Hq3rzP0z6Bd`CmHr?YFQpN3xx0
zQ#XvCuF0D-@dk|OjDMgZ6ozkvvbaWbm_{2RnBg1sOfnuhaXaF2(s!+6htbq4tD#IL
zRDER%8-uRj#qxVb1JPLW%4wEYKB|5_5>IA6S{bb6eR}#z?pZ&251+pM1o^D(8^OHy
zSCC_3$Rc=y?-pcedcCiW&hR^&XboH23u(#a=Qfxd%m_4|Fa~C$lRyUvYY-O94w-c?
zlzf28+T19oiUwkVMBT2kVK@lH8K3}M!JwDD``z!31#+izf!Nj98?VBU>ooKMMaL3P
ztjc&apC5hqXuc567_9qi^LY2kWO9YknmDMFkVF~psSlDTN9R%5{JM&+uSZqv`s=Sx
z&CazdXcnzk%x`OMHg$a&Sx%GXhfTv;isb9mai>AFiuRp%F+9{0PtE;^>D$EqaZWZv
zvJZ)e;3^7IIKeBiP{bJ$d)}t*Z1li-BgcxA^KpX%p<{m;b#Fti=a3ak&d)|u`Ho$5
z5Zkpev<Dg2#tNZiD*MK4Dj6z_so0_pW15~mc+s(YW1ZN4{WNxQcpUpUyfJ8%3=F?+
zx{7&2*VK_;k<^K5x|N0bJeqB#t0z(szM&&8>-r+8xw%y2Zx--hDD-yILJwG6!-o9g
z7USFEfXPp2PVYu0GjSu`QGJn!zzRSCbRupx?Ml#P-pEF!m-GsYl)bYILONa$Jnvc}
zWamt=N^l%xbr`Qtd804GX`lg+A%T3jaW3Rtn^=qEk86DqJcSKZ^lgO{?^;0B%0KZ^
zr7q4`r6o?0D?lur<PL?*66_7=>~6bD+<!UG7lhI53zw8ydSM*7YWbf0FAFxR9m;5T
zL~<3p;3UenpiE_N(M^wK0`VW_%MJuzlZ(8O@n-?-Q_!(qui?y${T~9PMGp`SKvHDF
zQv<xC^DhKNkmu9_fqfuymNdw#9Pc~*nc5TVprBlVX|l4@C?+hoT7R>g0=*3ri+c@j
zd!c?T%YC_~#<{O-4BH<ZoIQAO_K?n5jSU<orno;BQVVimobgeQc%OIncRkObrGeH{
zl?wBi@z|{(f)Lw=-8zd`Vjtvq?7C&8mapT+wS4vLt=K@!F+glOzZep43vHx3kg+=7
zSZ={bt96bYjL#x$c_x|;U<4V5FbBvlrH1`YIBd1%NM2BHG=m%KO#qYNZK$6g17qXN
zRR!UGwy5{-;$=4d{s4bD*;&=u?mq=S8w|}x?Dr!=wosU0F|yrOui}r?3b$hkDK1Sf
zmQG(4ja?NC{+<<R>)1BYk7n?#x9M1dkR$H!Ylnr@vL}VzD>lRcfp|FfjQte(W{sGK
z&9unNV0!>2t`FI;u!DW~qIDC2DWQzyKC|41ParSHn?lKC?%)`r95Fy-Lh+@rN*)YZ
z(Lz21Rn!eU1`;^&kKJBit8BBWH|z6O{dfklbT%^vR3nSTek7v-7#D+9J`^cfOyAP0
z4c;H>t*M{!?T1WoEpkb}4>TxCo+ReaK*6LR=lJrUpL3y5BZzCZNH95<cHgaL->v5s
zps_PiD#c6Fj2^WNXg#n%Fv<;>))M(_bS06?M%0zrh`IWf_r_gr`M&26%h5@vMz9N)
zGZ^MW_{)Hf?w^1<v&f;W57k88TvX3e{?X?U?xF1fI~L;@!@d_Fit|q7b@_47Cb$|i
zxm2BP(9N_B<4!N$-d%mo$O`YYU9AeKnB{;RU@`X9N?NoMN30HpCPK)iV~sT|QZIID
z%VJIgn>R(T7M#<8>cG9Lqi-_zQ1i5rvRl2AKJnNiHuuqv+Cq3M0$V{>J8X{AO%Jy#
zy<)AoZkwjXZ4di5C!v(ulpkzR?0s0^3yL#2tM7tE94pMYqpnwWltm$2$pGF0IY~=H
zMUjs%BZ0Gk3QPpot_MqJy+eg;S9Lvt2w%Z_xVXe4((2FXpaU)GK7U?=ok~I#>^2f_
z${FN>=I^@qS=&>cmjf-d`zvE>(18cuJJ7m*57V-G+9g;$mzBXbD*gAtjx-M^Znu47
z*wx`0_#=#ct4R0?boCMNZB(6=t+WJ_Q!_++t!d9btHgd{^T>WXmphnk)K`-T#;3Vs
z>;FUIj?A}-NouZva5wc5*EjE;4SMO@e7KzaSUkSCJTV!JC3Rw)?%yR6A3aSx-FrA3
z$@?2aGP~}It7RUAj@3eZ)FDv$D>?gu&-&3*%tAOLVFL=xl>ikV7>78v5+tME8FqLv
zl2T4#(M1)1?t)SYVJFG1LgaEy>FWQD=d$*re*bORL@e>5D>-X%M%%n5Wk++j1%sD{
zld1Hb+&m!>$mauz|J=Psbt-}|X1Pse$H?dEGwE<F-~By*pczhPQZJgwWMNTSi|5jT
z$zbU23{dlC_1?T?<$pqrbgu?VNJ}_!g~JNl!GXBZo2w>t4CcFl8(>XbZL6m!07UQJ
z2+p;t7kt`PtU1D6%5YlB1DITZ5rJ0p0$^DlN;8J$pFpGy##!PQ{{Y+&YHHaLNB&C5
zU<lz3gk*g5)#*?uTIila=u{}p?f{v)_`|LEs^4E8OOKVA3Z^uc87uq!tK3QlKDln^
zm!HdYgohPUOk>lDM1n58HVPihXO1b08+*j#>3~0DsbiV^gSgp(<hS+R=Q8E^f(|Zr
zu**1eBwwf3m1{5o*oGq!iDZ+~XOShjV3fl3o-&V7@{#b!K%PZzWP@^stxLjmLjFS2
zBI_N=DT#GF(bs{2zP}K7ci}ykV9TMQXUJM(0?+8+IA@24#%4g*YO}LEw8>r0($}GJ
z{02%x#`Z1Gs)r&3fX*NWh_2KS1NX*iZ)_VNE|L{32qacbkRVJcBbUNFrlAokoCgv^
z(_)*Um$^MMmr!;EyWONzxdi&i&t>EBb3$k+czf9J6Cv!q0yu||?I95kWP%YbK?LK;
zv3wls>tgY&Gk1x<jWlwZUKh6;Pc$O}W7$O^WZn+H5j7XG^hUGkb%7ZSxwUE`U3fE*
z@umxJoX3sTLy*0*so6aT5AF@?xvkx5w_kBy=acq5VP5X|X5dZax|yGaHQc8dv#S?~
zQ6dx`sg(ol;lM-&<e;Tey@A|k5QkU;C0w%E>}{VS$r3_HB2MN@&0SAPIv<5M$la{C
z;@$Nj8}R~x@RnWzN80y>)KW-q6_3PYk*RW7@1uBR&ywYzuGS;5Ja%1GR<3SgN6<(F
z`-1k?RCdXSDqthzwjKH9rJP|AxZH7S$Q2)vGaS1+y;C-E63jYcPb@4fo825Q+PhNu
z|A~7OD9MiUOf=)>zVA!znU!6cRlU~Iwe%)scdMn=UI<CFBrpY%wXnItMiOFx6lRgZ
z9@jx&z{W!vZxYzJ$6(B2L>V(IwjZZwykXBAyL@<a9%G+(P8*)zz%yjO@B9CI^WLn=
zs_qsIZmDj@-Qq_4@yEZ%9}$1vVE2r|`f#O^z*cdU%J5JwpU(|VcVBL`y0xyL)Y2RL
zSy}Ji!C6mBs@^+5Q^ec<e;5DGy3YI-M)DSnAqEDh$H9T*AZ6gF*f=?SPFZz695GMz
zzh)oZ*W|PLTp$W%E%%aQHWS;wA6qjOUy>`AN1_4jg?!soeSB!7Si}mxIQ09W@%q$l
zul6?3{4y4G7H3k~Tz)Z@&J~xJi@9`cF`vt(W{S9V#Gg(VkQpnV$>b}Qav`1eANjj3
zfKZdlkn4i0W7rAug~)s9TMQLLnm5tKFcBQ6xxlXG2YF(kTAf?XxEP5E3D1V3_fq@!
zU?}AUHwzEm{k?QLOD)hImA3xQ+}>&#U%S1gr;o;vS7@a3rz6<uC-$gbt5$>4ADaob
zjtq<hx#*x*nfZU0&Gls5_SnDw3)uo+W;KQE7v8VHYayk5G;;gxk?0!67dAEXB<5-v
z{pT8?P&qGCQ5C*C!{h5U5*9{0V}Lm~n@wO%%#8R;rP^=RN+tgYw{PyOPldzb?}sa6
zW3}4YSjDWzVyVunszt0}C={#az1Xz#S6_mE*fb`S!G0pZFMrSXv6tU=Ey!<f!Qv)L
zQ6|#$Z`K^L7=1sRgr@`llD}(ZvAa|vorACgA}RW1pO$w1`k(MGl$L*?2lK<j;0w9o
z;rsxtQ`zhQbieS$ubA(d&wzs8<!fVI!k_#8Z-*Na?|}^sNy^B5F_?*{geGDO**P8$
zBG1rbgG!p-Di_2craMo{gQ!CMt)NqlCj1v5S*VfqsO@Za(XLVE*g@EuklhHF8(i?v
z&~a^mpr+FV-Kgqd6Z>}@T%@K%UaOWWo!!tG*;dqtPuNpk7_j(&p|Da6lMa@@mJEgb
zL1b>M4THl$tO(_z0dCKYJ<CE+^mr*>YV*{O%cVSF4jXVd05c@s4McMxUP>iKs-260
z1X4hHFAmqXzG;3}K7j4quq})I0jCw!k!~wGf1XYwsUm`Hj)wgfmPKHD+~GnZlB-$d
ziC8Xyo6%!o3(X&f11URHOhoeL@ylw%v8{b{nxKHl2eYxxU-Vwv1~^ciu*r-EIEhf7
zHk__x(?M*}bS!Gn@c7u?+TdV1f~;q`=8;^}do!3A-RImS4W}-&fi+WW$R~FV>_2Y;
zjcdr_FvwEe+tXHi7{!ALj$n!|X9cDbFJ#&`k@^HGAy{=Cnhx>SeOR?3ZkFy;NMgxA
zA+D6bRUlMQmh$8ExkwCKO1)**zBQZ9P3)`hyUjL0Ye3T*9s`AlnMaNJ>}5XNd6zL4
z_|i+51+E2Rxy#OePs#2KYnfi<{hvVUi*4=$Rub<f{2=F>KxWEUVrRqu?E8}MZ!oWQ
z3A;n_38QcyYp`CT+zL%crZnbknA3%`Uj1UKa%Q&=GlziE`z*iOx6b*w?F|_It<CFW
zgu9?zL65HdfOAmOXryHA85}guHqvHra4!Ph=#Ox>rglr~slj>scJI3fCK`p*M!)^1
z&nsWkqm2Z5Vr=x#Gutns&0NM0b{FwmdE`)X1fyB(C}jCFB?~Wf*86>Hr8z!Vb{^SW
z1=3kI$D3|n+>$J0(^lX(txeD)=a=NSuvWlBU$5&tNq$AR)F?TLl!Me{>u=Z~hjR*(
z0<KTzKU(H!9^|t!+i15^63b~O*BP3@wVi4&BqspvCNt)BS{x~rP{Gz;`3x+zn1K`k
z(el*Y0qQ<gv%;Ui-MoaqQ}JXbHjGfU7(GJ=23i9L=q!SdFESiUr{bqtkozS*pxRl}
z3ha?-9WlE~rR@EVQ}q69skF-+;p{`zt(g}m!tqR<dz3>NiX9wjTvM%H(-=93s-xV*
zwVsKG6EB`=by~uj#jnA-ymDCrdjUS~`#fY%@Fbl{m`7zuB!%!e`fgJvz^e;4P&6Ad
z6FLKQ916UOcNg#$U?{o+pTxpW`UJ(SMe7)AWk`xZ8yW=cLRk7KLqMa2P(A7>62t;J
zilHRkQ^FuJlCV#AfW_3^@+OSj7mvq_*bF`r;F7;Mw1{M8XtdE79ZqB2UZL|j>~@@=
zOJxbDG=y^IlBuQ#u#y%Q!;^#g2Zj)Wn6;+4lk(65`N5N7wXP1NtD1GDn&EI}Q*(Jd
z2J>{jpi&@?zj4n!_mqOgdY!BN?#U)IxeWJ&$4@f*0G^&)o*c@<(iD!z@oA^Jf!hAt
z`HlPo!_5a<OS&2JQtQFy@B?|a>Gu7#flRld(Re%z3s!yzmElp$5w%ZJCme*G_BSw^
z0#s17t%iW07vBHQgH|6I$z`*-k&(Jv>U`_!jt<s{jgF6xV%eZuYW;{t!m$&75Hb9x
zX(y)5fLb;;j3R@=;(w4OTe<_TcZrRJcBqyNb(t0zwI<v9`OPhs@U&QrIYY#^$EhR^
z`1158Kl#Z-CX=v6j5RxR^PXMU_^x)aUaKOR_dvB)KUl+7cn3_o*~ivB9S)4`iXf0+
zB^>b--}~&d&%&3!XX)mdSt}olg%%pk7`EhPjWHCOjfK$hy?^Oi-`&zbhN27DIrlU+
zh>h3<((zu{AQ$Ep!e)C$+$3yBsMYKa%5a$wv5?Vke7w;Z$DTlLX_Ft}@*r_tKiF>Z
zBLr;wrQ*j6d_Qr;P}p9mnWrv6EJML{A^b3?;aKt<^W^10qAyN#gZfrI*4)iFan}_?
zV|$!{Y)lV>nm8{H2>r1EW=zMhdwKouiwO=0x6FfATydYb$ACo<Qiui&`64nf_)CGH
z5uuj{i2j5kGtEv+Q{Hz=mj{x*C~+RK0KJt-wVUX~RfP{iAsm^`A);QQQ40Ss@Y!K`
z;12<%9hdjPUtnGQqUo6Dd`Wm|rx0uN2y(o`a7GTJj@Yt<xzV3P{!=Sa*uvT!K2V>>
zle{`lD@oX6wpA~Z24*@wB|UQ`X~@gw^GL9qICk`&V|#YTQ9K^sz313HM~@{sfqqxN
zt^%<hZK2sd+2C99dQ0l!>wPZSqVKvb0dGFlxFgF4Ip~0I2|MK?hY8X;(&Wt!5(toT
zu?Zcp01-v1DM+dc8p37EoS85zVc6AqrsxOj)>*^36ass3<`h6yW@v!pMNxXITByEZ
zHlim_y!_<C0+Pp?flA|S`pIxq)PlRM<*K%R?Y7%)tL}#u-(9oRE%aSnv9W?=rlwk5
zSU9=Vs0^4lgel9L-%|+wG15>hJ!uXf+-4Ed&I>4%FCB*sV;*wLeaIl;n<zo{Mo(21
zp<N8U7I_l(3!11Xqv(PK_MFfyludrC>AxrhtGgc&112UKV&L#Zfx)Z)X=Gik3=S5D
zhX?YJL~)=9_W)#?JL(ql!}}R?wx=AgizIFqC%veo@YG}H0St@9r_HgUSTxu9mHpW0
zw>(s@4;2G}WTKFZAdemDO(b4W**l0#_VKtzyuoM%Njntk{HeLaTVvb%sYi0r*ih$B
z<PAW>K*V$Wv5ZmK2Y>7t=DL%7q(CuBepHAW8gvU|KJ1n>?)3J(#K#~QiuOri2I>$*
zbIp;^<48epABLp|n%_st{$O+=8W;)&hXQutGn>I+K8dZvy8m~+>g|<1pS0J8%kR1m
zMk3kGUGDrz0H5-HFgRrH^nz@|Te&Oywap(Re&TMSQ44}yvt@#&5e}lqw2BI?DNbXi
z*>N?9o0z2y4S4A2i*r}FS>$oR`}$<^FA;zo?_5Mu*HjhKKel*A*o3K8plsk$%d}hp
z&@F8e1y|^zvOX6!(T!LNp}gjlWzMRyRv|6QT#gW^bIpOpOSxY?QX6uvJuhe|LOAo3
zxB@$eSeWM^2=<W(&E)~)T))0-mIyBekTn;SZfQqZ(18nf&i)ZW<<eOOei#VtAO$=x
zVC~HRRi}Cq^XeGZV!ReG16$G<Lothqm3X|)|3q6PiNT-5-oQXJ0ZoY}ng)>uXj4{R
zNs)jC7ujtUSgDpF2nc#xpT<Q_%SCV`U_NG41*{H%NLB_Df|p{(7suRsS3bQko{7cR
zX-D*9SNKRY5!)Y!F)X$RYiVftY;E=PzTC37ZllHNiMr1k>lD&3CsVPFs0qKpBxe=|
zhKGhe!nY3}m>vv8_s63qx)-}`^D}2R`)x^sELm?u8_6(M*Zb_ezeUM6usP=dg@6X>
zutSs3S#iz~fPjrC;zFZ&ojG)~W&?e0ou!O?mMwqxyWd?bZe+rd66`i!|1i#uV`UwF
z))<~C>IGi?@MDiXmdUKA;LcmeDQ>U0Kjvlh<>|g>U-mVjH)6FZyv1${IUG7_GAt<5
zScxmmrk_N>_HK!PtAt^r?4mKnazxqoCJT3hbQlIQaHOyFRWp~*pUZ|KG#|TrZH*0+
z+-6FVQ1;x!m2ZLeI_`sZWv`jfCKBxwSoDp?Le4nuyM-3qh{ejOcIW)%M_#3qx+&Vb
zxHH=2zs9}PR5p0ddcb@ccDZ@X(5HQWigDXsYi%xI(y29rmK{fVqmlx9Eae7_6HTF<
zdI;>GByW#0CNXw7#}s}ZV+LK}`MqYUE?IX=Pdk~iyxFrpfo+#_Y46T!0XP$g5BdBu
ztS?tMyjPA8yX&TJywXW0hnAaF>*St^U6mjx5BF+I`K>zzYZ8gW6MKH^N{~Qt=j@fd
zqI>Z5%1k!{bJ{?r0yo=+UrZz+*axqi(1(B=)MAR=XlOb!sM7(lD1s2ByrDkM2PM=8
z+X@+5Tmos@p5%XO9CK~dTE#;x(6RoREUVaV5KN<`M6N%f|8Qbtcw!(o(Y}CcE|tB~
zv4Q_F;=Nwt!*az5Sy(NR$c`n1UWqQ#6In?!E6{gT#B}3PrBEJWJQAm*PB~S6x>0oq
zF3@y^Ax>El^G&NzsCWLAKKUxTWi!CPN+y{WY*H4fI+l|vvPrlT#x4CtJbp~lx?xvv
z><zxWZ=zA5|9;KoexGC3GfzEI>?f@4iN%ge+B$4yjt$L;Q?Beu7;&}2)qoO21R53k
zIU9=9vFRk6SJ09zA==`K0CicC=w(A{)djX&>V_At_n0SbTNGtd!t2dHYG{=i!zY3V
zD84u#g~0>4h+<&nR{?P4ZMX2=^~1q%b=zIKCm_3Ut+UErOW2*p_D|V9ciJnnl%YtA
zFxkSeyig)zn7q<E5m-6L#O|_no+=mAC>uS7!Xl8VJVq+?XAsB?7WX0~uM`PqHj2gh
z&65b`NX$zS%|t4rc(#TYCLx-2K8!ln5oncK&t#qv(<PqWT!YZ<ZfyL^eQ)()v(&tj
zR&Z@c%7Ss4a{~W~emje<g0Bnl-A)<{g&D*NXf(Nl8g)f)Z3wCZ)rQJ1f&pZDR=zOb
zbPNjl=O@d(AU9syGf;*rImm#V>aH_Kj#_47;$#6!1R?92oj0zn@5yJl*l=KAqKE)^
zY~vFQBu4AUc7Gr7o!0P}vjHoT@tVX+KEE<|<y_q?*5GxGIPpSpYia53>*wclq5fOk
zI?~*L2}H}K(TRZp6Je{c^dXcollfTY4UiI@En2Zh?rJ7>I*~Zt`PB0f|CD|}#$U%i
zg45h>#Ex=N{Xv3w1jB9a%Mg?jP@m`_D37G9+b%SsTCuL&lRs6miSCL-DXi$-w*Au^
z@F6kCe5bP^F?uMODkqBl7Jt;X)ouT2_?Dl19({n785qy5%~3}IN8k*Wy){|z+<-_E
z3Uk*t`zj1UYGf55cab6wU%YhJ&tf@w5DG_TptLXmHPZxlbN`L$-00TJ)9$nMN0dCz
zUk1obP@cz^@I7oD@+fu>>0Sz%;!S!+!ji7CLAX}La%&&{Koe=JhA-66+pBgzU0F_c
z$^V@0OWX=3Ipq?^DovnJdJ_wF_gPAZ<hk4pRPLHI{I<cp!Br=YWc6@}8(N^RD`^uv
z;$9E#HJ^&b9!ex{y)#c!&r(&^;YVt7!EgpE(b)$mteJ3du7=1Zt5m5j@h5-ht;xhg
z=#$7`k{79rHDfmj?bEm_9$lB*d=J8leO)&7fSgu?1%t*vHTgO22H1>{ctO6o$}dJ%
z4E7^zJ6jss!S_C(uOd5+ny#Un7NYVkc^GfGlLVo>`ndgL$EQ^>CWF2fR^UE>{l(vd
zd~6>j)>DVx<dvrZWMLj1qiTkmO`^BpGlfTVnVr^d*PA#)8BgJvAjRxWFu^NfJDY*?
zQlqKK6rE~Zia{JG>r_HTnsUQ|?}dKBS$pHn#=&D#la@6(b?jiH89$vWJ?yN;e7KZ4
z{qx=x@y1uO#9wiN&<6C+ZqPd3<_j0F3yAY#1`ux%*y<$BWIk~2+&R7?2Q%$PdAX(P
z%-EP=E7%${HZ}vaLfuj&(9*L&ehmZdAT7pz<1GF;UtV?I(Ggr`q6A=cb=i4=yg29^
z#2kAF<C*l0A=0TOQTD(vj)P7kzs4Op79Fg70B&-l87ks-y!j%wT+?%!^05;QMiP@1
zBe`hKIN$q#O^+}7ZcUAt>qb(;;b<h^c{m@5hKFxY6&}vUamd81QXv)3KAcbEV4lI=
z?gW<{I^WQ~RIB$P_tM}Zl8_+#St`4TcxVh)0jvlPig-4kJ{`*xQm1Lt`ab#>K>sEo
zpZbs!A4UOAC(Z2#IY6>2c-9vR@eq6={!wn)`dh+&mp#45UPH6<^Cp9((<A}Q>)Ota
zJ#8Q~XU=0q>11mff5uGbf{}8kRgMI6)8MAK4az~9t*_yDHf=W6<=>Z}D(TL9SbFy*
zj+vv!ts_G81hVBht4>uv)5%2>CYD#*ZRf<RaZ_1N8+EKGzu;pcK+a*vk^4k=-aZ7T
zBagv%Al$Wt|E-(02d~zNVC-0RWd%(KTimd+HQ13>HmX%}N?BIwZp(EZ{o<N{6=^$U
zzrV*5AHjaxdwnNK|CDN=VSceDmN5DAc3`^aibNyMDX+&S!#4fXZOER(2BQ-2-R>4{
zNXBFSmI;Sb-NJoc;G#k1QO0%i@s7lXvt-)4`2sJ*{45zM;pt9W-<36M9v|wcZ;$V&
z4;=%c;bh?-NsfB^I1=~|gxZfEbWkZ9gik1g;8YwTk>&w=jlHyeI-r>3OU_;q9JFs>
zkLpnbzfL6&9=!36g$2tR9cwm+xS3YPOO_bTN*TotWL>k{4R0Ska&yJ8mm-Ddp3%|p
z4Rq8jEZnj0$dSdR8djYBw3n{XK&$2i2=XY~{pZg#E28_*7rgkri+^ji%o<i7FwfCz
zeOQZ!Ts-8!>^X%IPy<I@!>^2|C~J<|=BBZqb5uCYI6{%qJoz;LtCykhAY?U+_yk6-
zmH`nmltLLMKyV6z7KV0VwV7My7gDKoYj-G_icFP&7|{#dqu};ElZk9Tb@1j<l7{K!
zCh{*t2NL5Yq)4dMmX=OsxJ=`~=ouK+r)$~IA_0xb-qtYh=Y-~;0kH)o+PD97@MTHM
z#206AjCIcE(}~PaW*4zdTlV_hBwE*Bmy0F~dkLe-Q%g$-Eqi7X{~D-xs%b&OsZ=sA
zlgJKM%=leL&Kbt8mp-IK0BsSnG!AMiXwzkzv^ay%QpzKk9WX9H^IbE|w`ip1o9C}T
zbL5C+9XWD_zrV)>N@I_biat8VWKE}2snZv(pMSFkSH7iboG1Bv51$xgSI79?X*`aJ
zt9B8}V#8Q%vV=TCJGYbR-BdE;TwvUddsgl_+rHhMZn~r{%jy=kZGHq~mOYij3RW8N
zL=G*GAfCuUXc4%B>fFbHzfTG761ZWJ4pqQ*XezmYBWbEo;mmd0^dM%Nx(4JcQ|TJQ
zFvp>@i(up$#AM{;V+<V7N|+D;BF;?Hd>0F|<IU2ZV|3P-M&pQxnIfub=}4ntOl6>n
zrA&x-Kl`vNJh-L*8i>aOBe%rEq2T0+sO68}7Q`~C$dMzV=#(ksYez>wzOf9dS(*?p
z)3LpKVuKv`D#27`X)v~D@3HBL0Yo;2GH0<HU_Zg>HL=)DWj{jlPwflGVpd`q%5ZpL
zA!N;++TTQO8EAD~yV{yB4XN>^jGsU*&;q^N(ANpexy5cA4v7b*K$+8DfVapiT;5j1
zgD`RcN=h132PA<E+MJ!?ykSXv!vGY=cuVs7;ZYMB<2H@qhX$W~^2rM1bwF_HhWZiY
zR!)|RLqo+97N(CK-gU#&Km`*%Q+X0={+g5H^+7ZF&_fS}F<mRwkQx1BQdp2elez9t
z1-6oWe&DV!GDYHzyLMNrvD|^et%FkoV`B(VFy->t*ud_ATMGxW@oIJVUHtCu!677R
z7~DN=!gmei^KgMx4qdlOcCYkWcsvM0+q9A4v4)rcpC;pSxp$VEloAkGp^>$j>?S$Q
zBMBQHYIbzk^{T0MHjHp;)owS?o0fi<Kj6;b>CS3hqywPh>*MPvSM@o&o$H!cRP^5g
zNRLNlwf&305g)-?1jnwU+@M!!$JlIUc519Ib9Qm=mGuYXeHmB$N)GvC-Ny#I-DSf$
zUTtkTe=@{x{&jU9;T)H^BKehV4X^D$v(qh^_6kUke#HJ+*RPbUZuLu-am75W{D!On
zYy5l{KLb4Y3ZmBzA#UdN*trII$FQmN0;h;9>I_lL220R5#-;(2bQC8MQYn(dtWV=0
zQgUFP=<>Zeidzs)nDUJ?h+9nxv!$pd%C?d?#>a8qK_^qLi%AvUKE~22lYz9t;TG9P
z`QwD@$IG2N%4MhQy&wJdx4#|2TDa2dLxoJSn33gqTdx4JEdjZJyS3pGIk#lq&kwdh
zg2F9+qD)wqJLm9WDfQkphvLanI%S4R>2&_DS+{#$_!~i*mLL)jt!h8IhvD&syS)2a
z>>Y4F@>V|v3lMVff%})RU+@fo%bBQFS`d9%GlE~rH2^6Gg;4}5hEp1(RTmOs&t9NC
z`ajtx5*(reQq0XG(=3Ur0fIebA_xnZNhN&8{7t~mcmd$&Be6sxcHezmYr|DBXN$#q
z?sB%SPenUZR2$vBcr~FY0#R`lvY;lQ5gpUM)}xP_KMaO~;T4?)`By|t<hI-R5q@kX
zley(iZ9=s3w~m(D&3CkR)~@}meCw=yP(QTnm2iFFAdbA05@}@7cO6zvzZ^W}&At!#
zK7xInKIL2Y{U4w|3JXVq>a_v%o|V-ap#}oL3s4iO5<u>)1XtCR!3j|+1KgZDYS%!)
zdoQ>k9z)&E^@ktAb0T=B_4qh|#hE3CDzr16$N}~M=gtLVKh1X}U8{3JJdWUhrSf>%
zw4DM^C1tB~;iB({LWXs%Uym7w^qqDJB8aiw&p6-Xlaf8wxv=W(bH-^sFJ`=VTqNqE
z)n#K7rfQeQ{yYpK5T&V?;V7%JFv{dA4V+`H+Q$Zs8<@9hiFB-qYEjguF!SWTebjj<
zg5C^;c4&XBp~U$Q>Br8xS;n_;Rwm8B+oeTwKAqehJo)y~U3Dg#E|tbVIGKo)8}Upo
zJ&mQ2v&fYi4x5rSixrX6X(%6!N+dD$!EqW428ZgqM&EuSv^$Z?Wu`+FoqCnfbS8H=
zlbczLhF<gpmctAT?w%MN43(!+ncZ{LWm?H}bGA_lAHufQI8++5&2+MrNu=|0yECcD
zN@#F!V)x*HWk#NOQ7F1N!>!hq!qKI%YlWa72<h6frD%95owWHEW$Z5@>)Zb}K7-Vz
zW(fK+ITp2F9&KvaB%~LqG1|uk!T^vi;~5Hk3yABW6kY_p^eHkl-nWmKiWD077YZ{P
z85`t5@6a*@LyRNf8y1?(uTHQsueq2Ht7T|wiLcXM!kN6eq^y^Bx0XmIoUbrFozziu
z3Qb!|t;5|W;%fcbWN0*a<n=?<Y8g^!zA&~n6Y-};pw5P?ek-xh8S{M!%U_LPx%x=T
zADLMjF9iIl!NF8&upZCECab0x3hvP{uUdG{ya?VnN*0uR(;4rPSLN2u_cNe5@^6{P
zR3v$9Z#ZzY{NMmFb7HJq3YK?=W4jNhBEe)U6q;z3E8#tn$ewVe+-QVCtz;;iI=nj;
z-dza}27<bWxK$0N>XoYBKbQ)H_RKmMG+Uht`0>KmyDh!UHg4|Y+39Jtw#^3PcY2yU
z8^WxwYE76G<eac&R7MO5wsq~9t7!ASeT_!>Rak?Z?_2{TW%N}h*(~8-^Yd9i3@yrH
zStw+HuJ2ZN7ko8jw0FTK?&KvV9+IyU4TxSMl=b8x))))Y4=vH+Ve2!>kvIXu%A&Lm
zARwrPTPM%AyfGnG6Y8AE^ngjB=Cm4Rxu##Y<hvmdtg<Wq8}rC2@q|M@MKGQu@n7M}
zSUw>KkYMLz%`NsMLF}AAdi3bA>4BR@u#bA|xfqhwm^bF*(b#j)Hc{b=AM?wbPcV}*
zFKqCmr^yFSud)a$weyX$-49n++~TtPz^~tb|NX70C-yP{cjwDlW+&z0ux(R$H|!M&
z*cq7%meG5l3m9ipQ_YDqUIktJDtJIG5F9`4&A8u*m{Xn4S@A*mX`^ooClcZ5H^&Qw
z_%~qE&s0p+yFWl@4Tk8!Z2BAV`0ZblHvM)gLxs<7&k6G<kP=^x_B6m=8ss82pU@U?
z@vd?cdi$xgQc3Dfgh_fw8yKG@woDM8l#`&3O=dZwDXng0qQk?rbTS<B54;K*W+{?-
z8#!-Ag)_^izdv{KnxPYeH_uO*7YzE5{UTk<WFosepGPeC+4wum&sK37fFU7y-$<T$
zRGM(pV5mCZwIw+bDWA*Sc60s%A3pOF{8}J3G+g_Y&YOrhviIHw=>Ibqha5a!T+zp<
znkV4-|4o^Fpy438A&d?PG<5z&9#v?zZPS!>#idVlFT(*uhOEfX<XoIKuCj|*0=Gtr
z%3KNnI`7(FVT_#z+!yV8cG#e|4zVqCzgQ}tHbcY1)Yy595D{1R!!H8}P|-t@KyEZ#
zofx_Rug(6Ui7GcTM&3aCe#HCs<C)AwpI4j(GN;XI2Km3cFE|yl8uKs)0I9I`_@A-^
z1KB%gcb_=eEb8zLr3Qz_<-z>_C>D#&gC}+)-42>FJ~Wsb(r3)+o;vQl^G;Ut*Is*X
z=K|`@4(zt??dIEi^zA*}FQ8WOcXK>{6XQ96HAF{!zbn|>kijc03Yl`JT38MCy;xjB
z+`9Wn5JU>^VJ<F!?^E(&#IQOyiP?^vwB?Sx1JsZxKh7UJ-Vd|NEvbyv^Dgvj&vNu{
zc$dYW)h=7M{{91`eD&V$d;<$f5?B$l%d(i|V;YYSS=O#7vY>R&dB2e;6f1>+!npia
ziiN~D&iLE#K5HKL*8Y^YS#+9cV{kn0z4HplrTJ*0aGe#HS<puoW&+lAg+w$z)6feI
zdE$m5GTdCtL!x*s4>$BYBR09`u%|I;<Wq>3#z=DZbm=69l<JHEk14ZBX7~WJ@w#Vz
zlxN`(lY5H?VYY`*i=mu`ZGnSV_B_jBq9Dz;@h7$MiVmMR_`@jI>pdtEhD8H4*<Nq0
zoo_4;F^|%a9y8S00kw3r;*Z54EQL)mGwKKBV$%0RKLcKq3<iRhKRU==k+PA}4JFbw
zk}Oy+XA9AAW(YAj;Sg3q71OC9*#5>Mjl$0&vxQmpe(7}s0|Q6Z(ta!)NKFR(W(CPR
zkiAcjnEbh+Hbx@8%2u#7Z+xd_o{uB0XCVX{84Q`oU?N~zwZYrYKrr&NDKAb$7mk-R
zkyJ1cKuZ$g*u+>ap9*Bhk0N@-^Mwo7TyxD4wccM7j_%qQfO(!a)(Gs`v*wG)jeQIF
z%+r{eAzLFQkTF9rq#O{2&~TpVER!`w_E}bjjyS+)!;m>#e4x2;D{{y(ioSJGlX2!U
zy;<rFs+(^`G)rqF8i-?+5IP^tA5nbI=+BXS)E|bxi_JBnXS|?C7H!XvN9-X7sO=Y%
zG3-PkYXe3bl|41{_OU!t<mU2Yb%k*(Rljbm^L2fs%`w6U&Ft89b>u(_@SF7sFQC0;
z8P*lU06Dzmy!_1q6A9zmkzKUc1>xUZx84eUX9ynWdEW`&Cou}a2{@_9fuM5CU?~^D
z3*2X*Ets9lntD!PK3H&hBi37BBEpj7FhxuVvHRgE(#tZZg_bzA+!AWX*!O?8(eIAE
z0@y68ofu#VF@JLBLgxa4z{ud^rUnJVOS{f_Pt1~+Uj0wdOh8Cz<rUDz?dYueBlvV6
zg(cap_rIU5LF2+}k;sR<UtRH@OnO`QC!n8KLMFE*wq2hF>1WY*hwndZqMy#r8(6?P
z$53juc-J`TI31Wml2frHR8T3P2z{h=7<gU>Lu9ISl0>b5Gm0i#XMrQ-MJEKPcWywW
zICJKVBJ1Pkt6JzhGs3atwit(#rP9FYAbcK&W$z6Dam1V%P^z-bpJk<S1B&0hG_JD-
zZ`z99(c>2mAln*JK_(L~D*{iF#g}nhP<7(K7fky*-}w&mUfH!u?a}DqceGJTyJ*vE
z3orT+!jh|&>^L%|e~!e_l{bVY9&BC(SpEZoLAHX?2JjCB%^Og2G`b0j4t(J|y7+^O
ze*|TVczHK$(!INs<iSiyh)9BHu5j@OG@>X7DfgQQ44Ed+z;og}@K6^ZC!BG-g+~=r
zH@rI2UtN3cwW)kQlLSdMNA0D^59UIlXbgEZ*NhbjhOJ=CI$0R2?z*KuI#!(XBS<Iq
zdMqPBAc=ik?@_Iu#7;t3lM05PNyqk0DC!`3W)4;6S@p6Bh9a?i&N4rVR5qE+@C`wP
z^<|TD14PFGJrdXa;`e>4z~?4k8}u2J6P*E6FzuXK>+;9Tn5129YKJLe!rF=*tE8b&
z1yNLmPf-G*2@BI05Tw6%qlc~8%xuT0s*|qse{JXEjbL>}aCTs65>A`=EhC%ot<8T5
z!hhI)-u{C=VZPqq!5a<)<F`kq8mC|(@t1D`T(J)k-1%Py-rYdFgO=+e2xHd0#I6si
z>q1eZuDT;oRuh`~--(eJDTt+VZZcjT9XW6Sxx)~@gWO>^jqKaiOWqV_HgwO>a!Q4m
zYCLsxDfb9^pF4EuP#iU(#@IdMCXp<?EUq*yK5fIsf#SdZ0<EQLg%{5<DwSYNB=oan
ze<r1ESj*faV{gVbMD^Pgi1L!cfxk360(uDf0ItFoN+>W0gm9>nH}o=%E+jY6F@?a=
zvJPx!eZAGSz;p|@p6Ca)?GeZcTx+dY)jQ<abXeoE4YWuIIX4}kVGc4U<hs=oTsZ}Q
zG1l4iXzolWM`_3Pn}yKh205RH4$%<n$Ts?9UGx0n842-c9a>AXG2drF(pfL^*mR21
zE>2$Na>#v^1lDTW=<}4jD#k5sT9(wSFm%qlvu&EZmtS~UiDapC(@@<XIN)GY|Ml^k
zmMN)>>B8<2#Nb{f=xM$>P<JofBm~1vk3Qd>d%(iy13*)0_DQovs}lccUg6*K1!3DA
z6t{Fi>)w<re-jSA`f5RO2OR84qyKz>*BARX?C9b)0M|poen`x7#$xMy1=qYf^K}P|
zE0fUEBu*KlPXN~Q0bhUIgIODn1NL^>Q+l3|?2CZQn=9uBK?k?aWvpBOkwdoVgw)Xc
zq-<5~sXJ|TaP`3LOm$fL0haz7AY4u0cW3`)?v1dCF8o+<-IfHsawC{sS6;k&z&{d0
zo|lP<csv%{waZ(6zLcXczQ1y7{rWB7yd!^8PQ2H*x=BuCfRbZ!q&M7ao#GiZHSqjP
z%j6<*&-uF!o-2{Zt&CSLco%e*9??-YQ*4Yon)G~TU|=*cIxvu#r@BiKalFAD6~A%`
z;e4@a3+KJaer`p@v+mj(?4kJ6OOAGZ&p;aC%>(y1+V#>;V;4oUO$_E{5gocA;_oFF
zH`jVR23q6hqY|jY-cEv`@)M;kk+>*(QNrY}cphDSDNGT+A?_@SqmJc5Dp{n+D#0`m
zTKNjA(GZz+g=jW2f}Ij-gG_sn$0iJ!M;t@xBi9~lFl@!^EWvJhSC-=fDGh8jr#gig
ziud@jX$B0f!IxQqiennCqzZ|#>SVR;b%=cVd88z;e1|VSXZ{g-<xRd90jr%jf67wy
z0gL-e*31x&$Q)b;787K0CK;`|5-+8E{0IDJOjwUyP+~9<zI*6JmDFC@!ZVqOHdf4w
zk!7C97g-A`Y!Lyne>g*R=bTr)l;yiIthg`1#FsmaH)J=?xnJbFtmC_>^i7I^H_c2=
z=Hr#}Xbk5)7HCXk%wylh=d3rOzc<3pz<et`xWt~}Kkr<YZ2w(@*BrCVb<e2J0E;N!
z&(DX6D2{2t^$;m8dK72Zf7_kja72(wh|<mav2CwCc{9WN5k;dX!L<mbe9Iqxu~}|J
zQRW_{dsCGg4q>_Mns@q}vqz2`5q0`@EDMPow<m4y+&1~Ka?b4X`w@q)u#1hBEAjkf
zbJl@nI`v(CkKg^pIoS~vrNcS1T&~xrV$<Fs`uf7cf=u0nABko6U9SqZR|_{qe|9Di
z*d^mYeq~OGJ>vlE{+sMQvSQASZ%}N*sC02{2F_H%Jqzf2uZyJ_^B?oxwihj~d0QC?
zQ&BuZ&5p##){kD3sSMafmrja4H4mGV$BEHd+t}FXryjOHI)JG5i#~@8lKdTV)TVn8
z6N~;n2i=kOt<zY0@+L%E{E|vdm`P|Whz^Wgi(bs@<szH$u2+9XLPQt&7)37DoE`-m
z70hvhsv?e0hT&DFZmwE6X(AA?L{`<KfxjL5ZQIzWKigX?%x`*um6<|HzR0xW7#d#8
z4-M0`VcRawVV`13m^>sSq2zQ5u8iqko&CQe{Ke5TZKC+E+nG0xd&l`<;Q#Kp<xj8B
zC&eZ#Eh0`gKm4F$=`fT0IR9oDrc}YAn@9z|;wp6fy7>au7LH=a*3*KUD$FbB!XRu+
zj6qNyq&=hPaTs&RkQEwKQUaPc8_PK`v=BMI;`;wCT2>jWb*%IU5E5r^)B7Ro8BLJ&
zfnP6_9{K%`ee7e;E=1FrR5}od+?pNwB1VC*SU0~PC?YFjdMs^{FPG3l)DXeHv&g}M
z8t(Z>xQLhc--{Qq;Zi2`<Vg0<5a6J5y^6k$U=Ds0eNCaSclcf-yp`OO!3-M2BLO<F
zo045}VFS060gfV@AAH}o4@cv=fjU4p9N`-xKjt~!!&_YX-OnbH$G#bu#zg;8dCe+*
zAdSwvzij0`^q~()vPbsG`aeNeK0Yu!TPeLX5sF}g@<eFKqTg$m-?}9l2ym~N=-q0p
zaKIG3B~|)^U`UpoP;fUNEC!MSD64!Z5^-VGdGEdV-uoKM`d&Qp8_3@u0`O8RhNg~I
zAoye|13wwQa1aoGjSIy}rt<@<8z-;(U;L5f!`x2#MvyJ?2(aemi0l0o%x@3o)Weds
zRl;=?c|7Pa*95@J<eh{w?TlsMmq%g=i43CT!9gE+)K9mH&h}otv$VLlSkM`bIqyt#
z`2guc@dm6z&HNN_FY{B6z4482^wR@O=cM(Yh+uwewP2o;-F{l3;Lm-;E7DYLA11L-
zLd~CwL|O@iL7gG6&%lP3Xr112C1O5dVM}zAPGyqE5rlwG@#Sdr_Z269DmStx7?Qfp
zE8guDY{ZM{pe^6)bdYI3yc7$C)3W5Jnhu9zOF6j|%6Kk?nII>GHGiM6!dQ=x5U$Ut
zYwYLpJTz7Zqc_>_de#xw7Gwed@v)8=f3kvsVrY$y9JohiVl>ZtDW%7Xx5>}7nNwx{
zC7WI5N&y_qc`grFmP2RSDIw(RJbW$oGdSxN)IDasOB&JJ1`F%7etyLHa;ICu-DUT_
z%2!Fy&#L(|MA|f%1rrh_<H96j*5SyMm*(a$^#Q7~(g=xaCD~zcts5lBuva%=kdRaN
zkFa@L(9}R%iLNryW_8^q0{PLgiP=ZI9vOouJw%g&1%dF#G*`^tvXwoGi%J%iC1i}R
z(D}=cyK}yp&7Jjni<V9LXQKI62Ze*(6%S;Gq|O(^lar>4aTT>y^U()box(-q|G$Lw
z&$F-{!_UA;h=t9L#~_QYmys>UJ#RFdVJ{7E?C#_}!db|wNiQUtD8%|8nbJGRC$MHS
zvyrw%$8}qEG3kvSA9zF}+{JIL@jq_7kGK~8C=zMMa-DS>Hv3-ln$BO?<yJ1%{&TO#
zy7MjgZiNDT+zsV%0(fLe4FJJ4na|li)4RcOerK}`bP#Jvo?i@(bl>(sgNcQ~fpKX|
z08X$ZUr?&4t_g?a+TL!mFrp=J4YODL-kQb()bzYp$hh-TF*ymW*3p=pxjd9>UGU9d
z9w^?M!*T9YFZN*G*_wycm6VuU-f`EuyvG^cD<wT*5R)@>>m_4y%JUIt1vvtF{O{Zn
zk@J}CJ$H64!spQlew>Zy@w4|C%Mbk6$8w@kpRq$u8DQFtSD^Us4m|nek8a$)W{=y`
zI$Tfx=tt=fdtw}VM+L6Cp853wM6CX2-*e1RNqz^}0*DRt4*$sK+2xN7nq)ghLCnf(
zKo=T^DcJwUzsqQT*f9ZjS73JK%kGsQ(Y)MY#6RdRo%#S9n32OXPN8Hy&I`J_O6Kf1
z@S`qzDCzGnfD`?A@dA1(qx%~eT&39H>JW8lCc$6RoVyf-1#@wu9<PAHUXzSg7pU~N
zp#3Kn;m8cPEVDP1_z?}-HKD}f7(ew>R6(oZfIt3Ifxs=vD274oP#Qb5fcWwUFba-<
zPtpaf7d~#6%iL$40jpG&WDhRvL9d#^PHAVUhlMC1h|me(mUT(=rz8IC?;u-L(z0Fy
z-6!TB$!3ei`mI=oFl_o-#y4e5H|zefv5`9;2}V!k{4+P~2}X7&O|@zw1K4hMe`8``
z@0}xKWB$%+KcDHA?Q=XaX@teSzK5PSH1gXo@tr42r;vyOm0Jc8K}hmZo!AK=7y9+V
zp#+Wv;c@}ubG^H6X~psLu6oi`(oX5(2O<rgCw}Yz5m=ipv2n`tgw77x(tjvG?UXg$
zh1a#%ff74}fZNeNNKuY<Y|g1rn0gz?4az)2o&*RA-2K(Ns?RP5ofLgeItGa1ZiURj
zq-Mt+O=O)4?JySeJQx}dwA?9{G@--4ll4bFZ5YMUf>_l*eXe;q=Z#FR!&z*aJD%s{
zs^nyLC3i=Hdv2KV=T1a}kKFklRC0;rxiR3-2KJFz!zz}lZ$Io3H)AEsS*-Pb8mo<+
zmBu<5bUO5|FYY~qN<mFb%|am93R$%a&Et$VF>hUE*t#f0Yt4MVsSmi9X#ePaB-|j7
zxq3(<jKvZUo3E2Gl!E%d!;<3fTZn!!Uqxi72_fwzAb-(^aYUZUAVSArbfY$@KVo~}
zgx<Ph_|x5o<XKrW?7p}LbxoHCJ3mk;Jr}s5#&c@YA4<^FmuL(Qg}Her8fArt1{>fl
z&fX>Kl856~-^0GQW3)Hpvz?cU`n85r+89s$lTsH4%h5$iDs?%NszWuiJ)vDB+o3Z<
z*$*usGDpbW%PTheQ=#NE5IEx3?5Z!}2UqhMzp2-ksJ9|hYhMns&E|IEBkKAe`K2|j
zvU+6Cp2SI<<Md=8lF1(1zb}i0r`Yjy%_YNbsoz$2TVeXfs}0jD+X3YYA!$|J{<)<s
zd>s(FG6<gU#~!z`cpr0utXuSdWPCMb(|H(~w~%@H$A*>t+5Z&+u|GO)6$Y*tiv97h
zuVB!3elu^)Rh{2E;3D$zIV;9XS2~~jV<AFn7z8E<+NrJr20DUSA6odXs|~B2XdxJ!
z!!(sQGAdR~m15?+VgUBXz`mRKT{Uue9zHmlF{cSsx73f%{8(XoN%8fn%<dh9*(*lY
z{u6yC!4~tKHdECFCb2f_#f+Ra-9;f#@x*V7UorlQHjNzN)TJBErtO>T^&82;w(ajQ
z4<0$c<7Se_#kgi(Ua3^D;w(@(7`CinBWi-d@PPy2U@+PU`mON6mn)ZT+dmt<r-lsK
zxrbZkT|li+XgUJrCO9!44g~xI6@MTQJ~)9OBNO>kJie|~@P`6}=k1l&rqopcmCW}I
z*q7IkWoHVR2~To2R+~1uwsmGNp-+N6V~SC(OP|9vB%!K9)#goHYKjF}t|Mb(Pl%W1
z@T9#FNl~iWL>%wB3r#eyzv-r%LighT-6#a_Lt!lxk_$JP>Rs}T`-Z^8$<_I(kFu^-
z{14)Pu&#%={H*EoPF&E!M{Rz}_<Z_VX0dwmr@5BP)N0syyHrG1yJ9oi3U><x_x`pt
ztdw(a1o1w-h?ELRup9y0(Q+t&)0apxMwhPC3;>E02|J1p*>`-}SVy#LJN&V0B<q^Z
zIxT%+a+2m{fR4SVwfXd<*6|G7OUpvYIxQOn>-hbuohq?avOdk)@rwhK_G$^q%UZ1!
zR%E}`V*M+`AG{5;q5Y*Qb7~D9!3V_e-7~)IBj5;M9b@rAVMP@kZxxDZUj_^W14z=v
zISw&-VNSEbqreJ!qtg3C^Z(V&H{Y!7y1AIig!g2#+#N)(;OGxvjRqS1wF)S(|Iu(J
zTX?rCzqzH&vHWXJ&lB;&<9_U5Vy}$nWn>VHATtQI$3t#FoZeE1Cpzc)3UZsG_(_Yr
zaC<egr{=F3<a87BO)%RjlK;wqXNLOq(?L;3Y%o^}zIAsb+)_7$cim&y0gdX#J?dsZ
z4hWWE?*U72jITiL9+uH;UM*nmuoubIf$L6tlDPPJ6cH`=``+pEF$*Q<Ffz$V?1kvN
zIz2(XdA0y|DrPQ>s+h-crmAH}LRM(SaHGyh?nqvrS2|=<#Z>1Ue-z1N(v`329Iqlp
z^6<#sv2Z3+v<ZPtr}X|gQY;i{IKt1G{y4EhW)1VZ^T}wG<iVNpZKO-yH#Qcy8}r(^
zw!WEJGQ{$d*^(vCujKUn3G7dI445iLFyM{=tj{kaB<i(!zs-kSE=16^kW?>xfA8U~
z$nILWuG6YotsktrNi6Jc-kd;cmiNAU^YM!ax;6jY`Bw#z9>u0^=N~<WT)#i-%q!h;
zfyX*beY9eEg>Y9K-%>&!vPegsg)E{RQLSIZnQcpTZq(rXH;26Km1=dVQYC?j=-Rq1
z$1`TFv%)_M&iL&fsv`XWmi@IWXO@@G2q&GSl}STgSdvpDJ8grz_Pgc0DD;Q&B;%Db
z$coI^OVA&c7iD=~xh#Q72soOVnHC{HqFS-%7AFZ=UBg~wE9)Z>)j>OIhygEj>~y^i
zFaWJ1F6#iI?_ytyA#wQuoV7#Y(cM$kszvr}gfUyy>eTL03xeM$mO?Vl<iG*OoAi{;
z$_>#7*DFS%H&g?G+<1L)=?=#=e#g>ceH@WXe(OAH*z&H{*_E08i|;scFaD(Oao9nB
z6ZC;JQ95p?nXjSWln&Y384UKqVhD&s)KRNv0j6N`b|+S=sfJR@OT;0*$W&=+?R2gz
zs+i>s!`eZ~#+HF<KI_8M+>y&Y^{A8U%4AwJHJi|`e1+THa5o<QruQC^&Bd^a@zJNg
zjkqduG(C9A2n(^VW)Kb2aygb;x}R$yu)i2~8N&{G$IRzqvHQ3-vZI;w=`}e#6?Z<%
z-{E2vTkO=|$R^(RIoTaR4sEvxJ}7OGd~SUmE7;vkwn)Q9Y1n4)DDsrqx{r6vKID7$
z-*MD-@Tl}RGX2O#uCK_pMU=VvBiCuWc;c4&TdUOW*EZhqV|I~$uQ|8F=q+-w_fEv@
z$2*WaiBUhUG2KbhLR@9nrh5#lNA5+m-+%D^JjT{T`*Z<`<M$y^B_UP4l9W-YC3!OA
z{zRL^MaU4)oIq_ST%yT0>zbjTG*nJOaKW3Yy>>6vHk}*hnI*Zq{hkM%h|_Z<WF%AP
zoGTCBOKHd(k$KVArIA&0)}f(k(^>UgZ+pt1Z7uFr1FF2iY9$LWKXkTxNz4@&yx|QX
zzj#Rw83<^05qk~0*WHqb7S1DA!9B=H_k{1=zTX4H7a)mIC?TL-0FxbY;Sxc@49zvU
zmleJSfYbZtfgB!WI?vuK7y~zEXc4y^(MJcaRPOF%CQ>z|=gV=En*Ci~hWu>I1UG86
z2((t`g2Uk+ed_OBsJo>LvxAk&;H(}sA!wCgOUwSsA}Bkuaj(GaOqsSMp2KIc11U#l
zg=2&r7|FH%%sU2c@5yOv_ujp`Ej=P19tN&E_A6aU+KDTl#+9~yo<CeOD>gi2sb5Eg
z_#R|MZ~1(S>N9oJkvYVEa_nj{Vp&OfA|e$<0WK#9b<gr*aar`S{cHRd3!41?W5{qr
z-l+$q9pZ@`UQgjm!X4NqPcHqoc|Bet<CQ}dFY!bUoqxtV0bly!=gq%J+pk4z`wK|F
zp;C$IvB+pLovp+bSk#(L-~i1hFhz<D5uzo~&mjWr<9yi(`-eEm$&N@2B)bZSKA(-n
z(U}AL?;aeyd;bBTL_C)L{Hu|IH|U=XMy4adNq-QDGVF*%iN57Ptmt^4_<6s7Ih;-x
z=1!eDeqv7|vFF6`Q>W$%>2!Gc{ZSJ@BC~~ni9YH`j*q^>d6TYK8PuNsp29qn+?PA`
z*Y7N>;q1_vsKzs@4Vpb&lG!&w871{tcNx>dV)=3afLxfI9($4RjYI;j=4pfM4)@2c
zw3(bFx^YO;5Vh=1>;Q$%|J0y<lTBe-5oPBR#_SGaXQ9zIJBveL>&O2#JS@-nAHIyS
zXP5*>pW{y4kOkU|Lu&3c$ZujA(t&(9oF8D?(%cK<@fRXB>JB~mJ9#b~&gq+X#A0{k
z2kg1FfpuGJu#GNYl^@2>jt0@R=2@9xHZ}CvMKutOoP`t<?G01=cZ{R1<nEM0VPr;7
zeglRiv(X3{<8(yUH;mXVNX3FNIf3O0PT{@0Ai$lLBL;M>J8mm{$$gzA(>m@?tZ^_N
zd`-*>bc_Bp2-FI%FZ1<kw~n+tye9fE;y)A)4c~2mLP!>oWQ9|xHfPfuOhAwszyd_3
zi4!-{h3sQo6+F2~C65oo1kmlPx7RCPE{c|Eo*J&a8p<yYBMDl9TUgD&+>w|<$Gecy
zuCn`Qy&UpkXSvpcTsoFtT;kiTWhL#m;=8fQh|*T4pZ8)n17;T4zNa>|qNpd7iycIX
zont9X`h*gQEjgcReJBw=i|F{GZX;nwg=>^?cMkW3$Qtzad>;EWVF`33T&93t4o9#c
z8kh2UN#SGVocqXmi~BGG*!Ocdt12teBaGZC;`ayMHJe}deq1Zx`Qsl^VxmW}Tkvz5
z3t0Hq+ZZ#-R5vPc5M-3=>`4JvO5f}BrwB2m;DBYsUGD)laKo7cXqDx-%vg141wYn`
zP&%BMW`0{Ss~2VpGexi&C7%Cq+#%<avZ!xzI3OMcvI&(D_Fxe;X=76t*6lKRd81US
zn{+lfoXpfE<uTDU@^a0d6pfe<J0#G&FBl9uOh=~MYR}mNd!R?-|1n*7SLMxJgM;Ch
zA|uRbH(65<G$0AXJMx;F_$g<;+Oh$BR_l-oZpGS=F9W{huB1RAOf>LWI2(HGwR6m;
zC-e$alM}L`zm9#kPD`gEGf=kdu5orEdV;z^B8{13QMjz=jonm800~!8wg$$5znh<M
z458#^2+j)YJn$xc78}@FELb@d3uB?_Vl1{;LK30ap$hjcb*^Hcji`IQXN{Q`4~-oh
zRE9W^i6d2f+>{+LFP$Y1^WM=nJ9ZvH$ENtOS?}9~{tb)uY!TA(&cKYG%qMC|NwR=L
zvX+>4guk*Z<Sh2ywKE?_p7HkTnaB#`fMK(2nKtCYUA{$fS1c&+N`W?|R@DfSq8~<l
z2o?4+eaA4((Lf!fr7i`cSW1zv-R`V;o9_~9l^7`vu65S=pzJx+eHL;M?Sw7p?givb
zJ_%0hw2pKT^iVlQm|_jfysDF*E<|;Wd36pWtGdG)FVRClRT{OjIVMWYVJk;;W#b+1
zct=c=uEyU^zQBfd)*J?Dwsx!vIba|^Je*4;a>K(2pUj#ngE1wp88voAWsPg?u*+>{
zrM(V}&s@BSJ|9PpuO@oCfLT{6=ed#<2{9X(W1J9n;3wC#F6J@!AiL)C8*!|9i^fvR
z!-ePr{!o<e%*^>z43f4L-!Pv|B-SG7&PEEToBe9YL_dYz#AD#wsmK~oz>Qz!9<?`v
z0+c9U^NY%C@f}LgnCs=THmi)e$NAf9W2ZVng0M)VuHD{mC)3NB%yK%J$%$!rGM9N{
zKL1SIs_0<<U|oOP6)XNs9xm5o?0{L$Ww=W=_U%aSdV*ZwiCxJwc1mi8lTNw@g<5be
z8E$vJ@&vjC-mAO<c=Hu#Ixi&72o<ZbSwf$bnw7v`n;;(OB}fnMahXoIAzFo8o0|ib
zTfibel_sH>@|+Ms#`te>2aoa2tC1`|lUt5N&{m{@H}C15|ABVUibd0_oUtYv@?RG%
z9OaT;!+x62Ceyjh=qS5)n#)N~Z$AGu?P@w6M>ijhn$S~O6OX6Y5|Os2k4@k07_ELi
z<D77$N0L-PpnRGAd%U+>uN^BCG}qtw_*~}~YoEJh=YDBZw_utbo8%Zy&S4LO(KqeT
zHQ=Jb9Jm(zXb{<eu>)<Mt@g^r<VrGT&D!?2E_W>dn~RkWo|=`d*DhYN&p>b&Py6An
zb8T^_-UELnm(r7<0R!+9__#M&NeBT|yDkcXIbUuM%`g1gul-so`Ai`YevgBG`TQ?m
zvO}MF;)y5x(ROBV%0|0na&3pE3k`t@YTcK`{t?jBD5WuDFT-4vOY6R9Q~+o~V~ht&
zSv*tMU~IB~QW~G|0*T>3$d5RjHYj;kCLOuOtiiodRl?D#bXJL6>b){&E301P2}pv+
zR+z<JoQsIM;3{nKJgE#K?F1i#F9i`ORMFTJRiGP9w9uodHJXbcZqMhvDmE2FlP~%(
zqv2A+9vA4-8V~D^thaXQM<gOe`o6Nc<50m{tO<t=euDto=^?@<3_kmL(918%e7g_v
zD6fPil~Q&?^feK#R6J`+$W0TpB*a@v{Xu3xOhGDXd;wPKmUtB03S~IG?c0kOZgMku
zwCN`%$R$LQA&_OUNeU)kqTYkRo(Ym|;Vygybzh6zkHKOwh=o%Lh{Kr-4hi!`x9r~k
z-2^!}S#{#)kiVWvU#R#$iXr?au?+U-Mn3Y9kEGs_dO@SnxTW!?aK0E%r{l#uwd98W
z>NSI1l7ora!-r?_l|Tq7Tbld!HL->@6hhV@ryLB1y3hQ2rszlr#Z2eF@x!5DAh~Z}
z0`1`U&Hn#ht$yN#`1iBLcsO5gHtYCqabRR*px_jo8(qBhVGooWph4dQ>HHzh`zZv0
za)B)k0Jah`B<vW{&&)JeWy62!Rdwu8c<?mJ4hvw?fIc*)CP)nJB~X0AD56ltsKBe*
zq;bqvt-hT~tx@}p<$+ki>M5OiFEv=<v7+Tu*q8_r^f9J8S-TM#P!7NT^%$gUuMPf8
z3VWmMufJyBzJ2lWM~R?nBru;3g_fyK=NMnGj*7J=`(zoSR3g0GISK3W%=+-~@cjJn
z3s0Ur*%%qY(48E);1C1B;dRkr_8?9`cT0WJ_fb&K1u-T#av{YNm1LYN#8b*vnB+9K
z^xB{8Nlyrn{|J%lePWUFnCH?~tM)_zFqE%=+&Ln%h5%5#)O8Ky)uu{Y@Cr@AA|_QF
zTlAV-Yq?hF3?W#RL&hnj&AGl;I~(xQQ(4KxqDjCtQ^!VU)>zkuX=DQ-etGpYY3OMd
zH=Iu^mF-px-BUlCHfO!wT~8}|uAh&F%TxT|nvN@0tCZ?f<#5zdm`Q&O_M6xICHt^L
zvY;Y$N6Gmw0`3Bk9a(L3KTe^$V$y>EY|h+&|NSQtfgpB>$zxlD<zP0KJPzmaOG`t;
z#X@0)6Z@a>%<*I{i@V8G&iJyyK;p#xyt7g$6o-dOFO4%zhplh;un$2u&NWZ<l%ucM
zqNB+CC^C{F3t|>gCRTk`r=T+>C4z^6o60IU_BtDFeX6FxlKQx_#)!()zohR8SI#(N
zaHbNzC!Nl$Q<GTFqz_=0IN$_Sk@Kk=u&zK}?-vGXJ{Qeld^xzR7vqfleBl-DH?<Ou
z{+qtpvVYq6gj4RlgEKp(o%VujC2QLQvXVmQnX_>t0Rd_c=<z>CP0%O@&NXjI<{rVM
zmYE_qalj)0HYP<+Vkujc=gXXOUD3VS^s!^=MwuJT-)is8DTWZDTH7&~WCW;dP`qxP
zgZ}B+7z185Z=x;aIZ36Ozr`8xTTMIn$g(?Ertyuze6^(%%fV*nR~vIaf9pz5nw9CE
z?;U5YP4fOGzlA&|%T7$Y7q)KzAO3N`7Q4sL>dL`rVD=Hh6vM7ESl309C=|1aM7CJ4
zs1{sJpsv!;TdLKtDK(F@+1mD5@P;+;1?KiY+l44+HUF^Zq5M)`+8I|lr8z@|gE|Rd
zh`Y2R(jvGG=2pd|Fp)brm76yKiU>Ndwqj^iZ|Kpi+<o`md-iPfAWjdWniW=Njas`N
zGh(->_D4y*%9|^wy4qgZ0X0Dj>093rnBDF>gIJ~4A?N7F0XGr*!12_&LTjFuSyCDa
z5ZKgGOVpvM0H|3Hrq=msEDs<Bz#k!32IS_M1OWI3yaWFAY&}N#(Bav<%{8>$C7fkt
zx8%;s)rq)7VLcbvc;UG%xH&ulGoPC}n48!z)jjxkcjtnaBJ}zNz6pDb2p_RTs9SiG
zBaS#!R?fxA&;|UCqa&iDF^}!Sc!H|;5~yi!Yrb~y;K4m)%PUAl7WEVvUe4=!a9D_(
zk;iczq2|9i3jh7<d~fu<%`>VXsG!&wOYk%-EbyEm5FOawfNjtTE@X+2R-wt2fmEqe
ziAkj<$C_AH#Nc=@%2-qU!Ia^!H=19qq%-Cz<YkS#aW;hGefL@;2vp5f&HxxCq?Lf0
zgd@z^AW0W}?AdZC^?_6<vyu5K$u$4Ir9a?v>(<D~;45DcjJ6`dfHiT)WFS(F1cUz3
z(Qu@e9S_{`AI-}KM@Fo3NT^s#?BUCkVxZWQsCA{E+mWIyH8b-vr`GPu0Uz_6sJzKN
z-5OZid^^?@z8*XDky?`r>aLhs?+Na>NREz^-jY!gX;m-?62btBJ(omcXp#`)5nFZu
z^#T3iYb@w33<IB1+{b08mDV!5^n-Iey5$#o)0)i@=~wcp;X*!gH)e_XdZbtwN#$hL
z08j6DXf79ofjpYrz};YOI+0w7$M$-wc<sJQu~^x+pU3^?8Gej^ooT0JPJ4~&Vm*x}
z%B4|MUmY!%<C4D(wO^B(PG@zdX43~!e(SB?_gdZ@S|aO{Ha&}cYBgAhZu8v-*mE_f
zvMxXZFo5z2r#}D%C@nIonot&D3<QaY3y2GklaOl$XsZ(ekmU<_fCpT*t3x%ohNl2>
z&r)UA=TQ3#FJwM}9^M@;6oylIfv5TU9e3P8AQR@f)JUNic_<cNNhYRqOg+0HySN5(
zbHC%Q<X^nt?fkvM$8jiNr@a*~kk)W}v{X*WCrEEdO``E~dGs2~pE}@zcRJMr+!Zfy
zH}IZ*21n3~nB=Z|Cv2(l?n)o>gU1lr{$4^H@SrVjq1miKatClI4g+q+DNsnO$ph+6
z>fNP|A@Lj{F&)?=!La~kaKVjp#u4e0GMEq%y+bIu+`6MPMRDr19IXAPPoJioC5IoJ
zvkN%0Aa6NOb^ezZj;{(>B}|(DvJ*C-orXfjOr~}XrIJbH(n=?j*gUTUU!FCv6oS1!
zp1R(g#m3I*bm{t395(;Z(tHoM{!I7tXo|LqU+y52q<iMH>wDhOUydg^jgoQ8s&o6j
z=tK|ar^mfCU=2co4=q~*Y*Ps0{9w;Rr1>tb8t^fL^$Q>aJ&rZ`@99$hJ|6k7Dm0P_
zp?5h~xo28<2Bka0dI!fdzud9jer~OFh3Ivsgy?*rr9@caSC(S+WE*a~V?ijGS1&Cs
z^$f{9GLBPedt}p7KkwzG>)z2vyaJr3o$j|aM4U1cKfSKMrnT%H<G0(RzyTTZjKl$q
zS;HB`M)bAbcb2^)jd|APCk4>7Ie5EN>UzretH_@vc1l8(G)vO+l)+l@cO`R#HV*O-
z%qDl%fz-KmU<SDak0wsXdqDY3l`i_wNCibo5aNnP!5GMqrKCrhfOqTDb}qh$0WmU#
z_7zN^d(u<A1GZ~h$UTnFuzC#6!@_x%yk}>ONCr+D&Y0DnyXU-Eu-$aRB2&*O_Drs-
z7jah4@cYiEIxCy{-}7PD*0AOqhnMPlSp{L65NrbnEqk<}syv&OfCb80v|!b*Y>l4}
z&}p?5b;nG4c#abbrz@>i-%#q!i}~zV@y5sY9%|H@x^{wdvI;46rL$cq{n23t@*#vD
zf&8}YxsG)oYuMBHFfyR7`aTHE*1Q4KACwA#7hpqFQg@){;`F@eTafP^J)#~KH!e^N
zj2IkVcn;+tbpWG8j3!<Si$IA;P1FWr;y<2=s6?bioEWO;uCStJS9S!_ssfDUVDE4@
zJR_3{1oQ?3?bhR^NEn11ku*6*$^Zj$If=AMarbCm+2(#4t_lZV07tD+tbVLT7Ukkv
zXG2KfSP7;dCz2_PjY$3e*Xc`Zg#V|Vk2;lZ_tEVdb*9?_d~-#bjiz_)Jmmdv@%^&z
z*L~+`PXz#N!;6;e*pLRTVZ5`=xk^`d+=#@FKxV)=YaGeJnS%(fvCl9BEa{gT;RUzc
zZ05Jm{4%eK1Sli*c+MMYf#!Q0pEGj!N&<^XMq*>Zu{cV@!?Dp|HI5=mr!=YIsl-UL
zF$#xke0=X%a@T%?Xy0lg8sBq$IJBR29I1}QB8OSp9}LHzYhKQ-Oz$VdKO+YKm|qx)
zkM6GVE2E8OZ}~b+=XfeHJj^!M0@cJgmK~Yd$)q0*3D3tv!Tqf2XgFG{9cF2NB$Vh3
zxho;u(u$xY5Ng%C0@UIfTC)NB-A&>jcN7s$A+^_WqHw~f>rPpYdyP{|pXIDos;?nb
z7&gu-<~8HAX*340G^p7)d>EmZ8k*GAjC-zDiRMC^1^ZV7>jwrKcV|A$S6RifMovBB
zXtd94y76g#Fyi7a<kk)AouD;E(AQhAHh}&tm)3a1_H5s)P0F=4o_iym1m}tA@I3GM
z$lQ*1QGrcgTyb8BfP0RN{)=!GXS0io9A>v{tlo3?-g9oT_YsGp=uGdly)*uMTfd;y
z^!qmNZ1+59yv6txbKHClx*1|GdNeXWxR;ADv=eyGfJdMQu7{AcA*7T>^tnYHjk&oE
zujKTUc7&@*0Y2G0VQeIV-CDU>#F2@^hbIn4jD<O#j1G@+e;LH_c)51x9KuZ$@ZDeR
zo$!(g*=z#!+_vAC{kQRbZW}W35*R2bQ(L|i`uRHPC*m)9cWo3>NIoMNa@iY4Q+|d|
zV;rm=^JlDDg_A~(4^5Qv=u-mw34{y7d$g})zi9a0ksS6P;bsz%9O9S9@?aIFw~vn`
ztD}6CRg6p(J3r9=I$i_#0`?5?eePp|J?v8<+h)3MwLl~4@Ry&0H0oHPopIwY8Z4oE
z%#cQUk!3Oqq=r`I0n8M})bb@$5csGQuc15brCoJcgvWzH4Gm+ARlB|7(lfWTMMR;p
zOF5;P*}w0aeG?NE31c?xR=a(f;z>8&%<1z_UQwTQP4=eVs<Ec5oQzbXFHyJz0H*sf
znIjaf`ff?p-T@SSVY$=3Q?9+R<U4>|Ca;9%=;R9^e&fF)oXm7HXKT<Hm}_0OriZ@J
zoH<wrub7|D11spBCgbQ?WCAl;3CnuDlO90=-1?bS(o0tM1F!HAJcyTDFxFv{T5d_x
zGF6;h=f5RYzG2FY=L<LP-@kv)M;-R?gRTwF1ZvLQx$~yU4>KjoshhllziGSM&R_-5
zdf>k7wMO=r>8!WJ|FPoK+F4_1)|pIip1vuadI-^c9ybzb1EvpbgFeii`zSAonNRtp
z9mYzY3Am8Ia@HsiH-)--o*~%ueJ{m?9o9f+(fIcLRxUe+-6KL{*_^fCAWDm2Rs&-^
zZn|f`q13~icH-NUnT68W=twX)GCEdT$RznD>nWvp+<e!0tNK-?C10(!NfRBP!MyJp
zcydARl|<BQj@yJ$Hjbl{1m5=P3Z)1V8;X?fCqP<iFz~_9yWCB@qG6<?>1ZsnYhc%|
z-BT0hdT_{fWa(=$Ji5)H(Oz*u%IkyJt16v}AE@PXx!*{X%R`75$sD>7;}d{vux`GB
z@wwgi2qIm65g0vAP6M?8Qm0sNfs*2Dpx#Rlurq#gU5vGK_~0GfM>p`EN}Dw3LO<&+
zOCyJKG-J`C`@}+*MWhuc2}Lcw@}hhKafO*ga%i{7?H)=dGWm2OnSZT>m0R@_NOlyN
zOhy;mi_zp{#PSDE)I}+qeC>@oAhycZwz-7DWkv7l(X3)W`UE~wHKxky5ve{+A!_Zw
z@QDztUeU!x#2kh~Cx#EKQ6QT>i&|`<g+J@w+IqeN7=uZQeS8hl*Fi*NTn|{U_~>iU
z`QZ48iTes+qKZ?pxyhWP?t-vriXXar1i(e7FtIuBT-ZC|`a9n3wK75|M@pr{$)!=N
zQqdJN=g*wE0M%U&lG37uZf}z=RV};VUdID41-7EOiB^lu%uqMPzjpUP*j*xB>eYc+
z*F@mL%mH2(oL3uWtutq2?}b$YZt2XK=MOdkm<_j9>oS05UHmuTSrBu%>Dv$N0zN@*
zhjBm*WitCjLs1KsgeHUH))?J{(BEn?8czL9OA*<ThgyG=3P+2b*2dFMKW!RM!yt%V
z+5ejM{4T<O;;%DUIrA5I;As!N{}bfF@)^cF=n5~=cGB)y_%DODwB)@;?Q{aMBj!6v
z&O1?h@S9B|iB-?JDVbI}X+9gvobG(riSy@g@IAMl?>Ya~)8)uY0ET<gV&y@@jOxci
zW>7BM&e0UzSldRrrYnlTki-ZtxoU(8l}mG2<kNhMYQ}G#zy8dTBbIgK$Qk~g?<bU8
zxPJc4s=vOaX`CndyXs1AZV8|hUbz8Y`9bVU;#<V9P@yCdR@O+?FAGsrnue@VDo=m%
zlb-~!N_-OL)mrUfy;dE-#)H*b{a~$T%z^hl`|Pvv__KSDhZY*m7#4f*E|z@6nvI3f
z@xAiR!o|;7dj)4Gx$_gC<o#tcS|Om9G|jPSAs=h4=0l;>!vj`2lWRYTEHM7c!^vPc
zzuGb%XL!!rO8#6v_0W6riFow7>!R^Q{yh(+@;QGAluyS5=)|~YtjS&6-jab{)8Rq%
zRYjF(Q{SkMjKI2-8yTs)rTsV0&01X2^ENE)v1aFPe(`!XJ32Z(J~|3RRJSzETe|p%
zFZh5W(2r@08vunqg2;%MARhKDu*iL46OC3}wka|rdew-}@u5k0^AdS-{U7#;?GpfQ
z&sRl;HbT-#fjd%~cEi=oS$f4ac&!6-h?a2eGOkOwZd+$Rf2wW~RH3aQ81mJ1@!$&w
z=v=s*0J-e~=RDV|U?TcK`-iHM=Ze_rbSxu?ydH$chD=EY;~?we$SI}J1Rs!?Yo?1K
zg`CeSn)UJin02heUfbY7{&l<M0kBzta@Jnwp(X!AX(i-6FyL$I+i-A=)^0~0IOBVn
z?-fF`iR^t^L0%?JpL1GF7A9Bd$dx{HuXLZGjIhmHt$zJbo+TW=(<TQ4F9SzAa|WWo
z5`Q5N^k3{q?r#5UZz;4|nDWBY`}Mt74(gd>T{d0AKQ3>Z=X?;UXNN0&@gI#cZKg!z
zUk7d6?yNrwYoG8M|Fz}FPGLxWRnV~&?Ak|uH-o5F14)?6bTv9^7Cab49O}Qp)*%;t
z<OywMBai$qo|qT$$a&RPe({Qb5_dD|uA=T0)IEc`7v^VV@c^ZB)J@i8TbDOmlgYfW
z64AWK{>meEwXfKGZl<=o<2>oJJiHSrfOh7cw03e<sdujH{sKai3Kq9#pc1OmqQSC7
zqo<a(WuW3*@;oT7U>v+{oxDU_(GjK0pxoQHtMe_kX;lu}Hb)j@unlMe+K{<;-dq4*
zE5g#cX-8XO=&%8(SPnKumad8LY=@2d{&|(gBB+2Ug~{+}JTX<T+_j3-3y~-=XR3^7
z>FUD5$&(8U)he|4NsKC1%$gXuo;5pj^Ubp}Fp`cqamO1<KsWB;%UA{*89y++r0;4y
zOVbC&`Fb4HRIs<~%`>wW>%o@474}j0<y=7@OTK;Z&fd!?LVFpcCOf5ytEo=I>Qvft
z8SK~z|JDboo!AdUvlrtZqpawKA`$Do!iOE(a1Z~{lNB$V*7RDRJ6n^P@<nHbrUdfa
z1VdjoKK*UsDZ5twkQmrE^hfa^0loVF0gP6#3LcaJvz9Wa>dYcxVUC7y6(bC4(ICBO
z;><3I7LDjiT*b&!2wVUiu}fk~L``K}2bQn|l%jB^h+{4UVtXxXZ!8dw24jg>YN=33
z1k#04shA2RDuty~EEx?(!=9(2$-Ym;0@3gV?_PUmAQhWTB$J8BSSpu|WYY0KE)zYB
z+^9(6e>j@S1>)&UB%4jgn#oO1#SZCHhholCxo5rMHlTNu(~p}!21b$6kQ3Hn5R*7O
zjrppNBi4}E#mGC=W7H#~ID>ogA2c3$<{+LEn*`U;09LbQ7*@hoi*Z~~_2GueSjMty
z6*yEds2oOItKgQw<qWP?aVcheFq%h=FJaahnzR-O@D`tR2A&n*{H(D~vF=oCBWl9`
z7)yC*(jav*tUZ8_71#@Wwr95bbP=OPEazBaAazF}q+Rx1fB5jO@$p@U4`1(=Dscp=
z#dCLwe`wv*#*ZF5dGgrNaY+20^5aiE^%QnhibDoJ^hQ!JY0CyWzlKbM$Ut$aT%1(b
z4mSNX>4@Q{I|V6<bWdc4ay}Ygzm%HYT)F)-YWB^UiO>-z6|Mu{^?q|q*4ROAUq^On
zMl<(PTQ85a6)D@Sg!Pkt#V)E)G`g~@s${7hi`=^&bX=$o-Mvw{rEuxCa>yGuwn|Yy
zxxX+lnBVuHIZLSX&!N2Exk=x`%QmaO={qplK~StZ)ktAqKEF?5H<YW8uWg{8Ylw#7
zN<y*nka9q~_O_o~pMzB^)@n6ImMVGS1mF!sn!cf&$%YO|Xtb&a)}c@~Q|^4*r1&>I
zSBR#fg~#PMH;rZ|OW%w{UhFjP#gWK2OUYDg-DHQU;o-0jVrP5<7ylMK@3&#ua<ED)
zcV1(kPE%>1puWxE0C)|F#(kOCB=``S%v)z?XO*0`^#{k?-b47ecQ5jBr1xq$hJ`%I
z*i5T6bKn4vSZ3ZO{`QM=S?>#4=1bA2zJer`%D1%-N1^%o*eo`~YdaCsJ~zY!2e&0L
zj=-ia;Z5mZAzZ*sYOVSL<~;|C6$_<qC=}4=8C0Q=vI$x}RDUEfD7_iizMtI;f{6@Q
z(7`1aJP`?nEZf#=!^ge3sTK=KIsbKh5FPHq$ga6=!)RhZ>tx+``$Hjk3;ou%Pk1Ua
zwro$@GJy52*COBVPe7l+3iKWuXOC>($J7gYgOFcav{lIUN%6Aik=-C5b>%p3Syg~_
z2Xi$SgNsS%zl$dTZnX<w+>xk{4Nly?hyMD(dKaXqs^>!+-aFq_icqq;;RxM=@(Qtu
zq>(Gh&g-INiOhoD0AdfPyw?mL3adeaCN!AbSJ#|u94%{d>e#WVNh{L@Ll+<!gLMDX
z(~iN6Wj@I_rb&>x4@!Y^9K2cDHjkFM!$kX@7tKW?agA1_Qk>TKoo{*J&}tnyb>nye
zA?3+zbXTKMUp#WkA(Pe4c^m24ne_1ZU@}_#ZY~lH2Xnc>R5l}RQF?d+4O2?4*{#&Y
zLU~yu^JTILzz8z_=WtyG)@+&28s7^J9g3%PITZXX0+jLb!-r0vK6IF_pHg|;@p@%a
z@k57RV9r`|v$tc{zEnY6EwZF=dfI+mUY(xSEux%9LbuP(fshl9{~Zpy-}DXP1YZYa
zG#p!LxvUGuFlO;gE}wqqb<WUT_s%pH@R)W1_N~V`LM`5T9A>P7X_J1j-gVSF44IsA
zhRikcG4ce7{-HmH;{*Hw7$hzXw;))WeN1$<`D<rukc{99Nt4v$CvSc=TIg(Tz|%tq
z4h++Ik}8)|bku+)_yTCpU&3wmJFvmb_)hp<Ds3VO<Ot};Z6hlhfH{m?DinFzh0D&f
z+R)}s-LgtE!^6f556`gCQ(P)P#bzI?<?*kEhGpg~YmR=Nqno>SwjTh@hbIpmzV5oi
zho;!O+Vu3HL(|hW_VDJR1{XXxhERQa_x0EB#t$xTg2Q|Ig3G7&MvN6{<v6mroC0(n
z_WeiS`+aL&dg*%EC<CMS$kPY0ODrF%K4LJH5)h!9A7R3gEq8@JXx0JLgIcr(^+j`n
zIQ4{JTh{VCCI~5xq*v2|EQ9NMfu#W9Oq|18u3kVyo-=d0g=Ie8wpZnd>5<k*S(Rg(
zCB+QcbJ0!V*{x<J!+N?oDmPun^uf$?FR~SQ{5)%Q{#KlVe0%Mjdq-~&k9F-AWv?`5
z)?SR(ZP@jNzLhM%PtehcU!T)+J+n_X-|U{3q5OK@1*oy;5m)r>P*_&hWXOLqXPhRU
zmQy(hd(zv_xLeD%x~*TsniCN{%_^cN(;exiSwob*=eoCC>6|sI?kh+gp*hg`ZN6vH
zN?-oT`~l!FL_LPKZ&6m%Lm>ty4U2|zWosO%FMlpSI8b;;E_W_J^gfP-E75=}n6tn1
zOTRRTXY%jhW9RZ5yB_}Rqpbh8kb!?2b7mfSFMWpo5l{?KnaOap4Im~#n@>WfqB^C|
zfC)iRo9`oFn0sKfZ8py`M(pgsK%0SHo!@2)(4=SVCOwm1^>Q!s!6%=5a`(c*?$`3^
z*ACU|Lr?PkFQ6an9p2OK;+Lz5y6*~IR6bgl^#{<yfJ@YLiBHf)x*j@>@`Z7P>X)gh
z`IBlbC1f&sJg<|qER$W!3yT^s2%njegcsdUS34VO`*b#gLBPa12N#D7@@&{e47n%j
zU*9I@7C$D;rh4T9pXT4RgLarJDkHR^6M(ldiCzgJo`kf{DQY{5-+Vx^&@caT2qB7u
zb7w<n#A@fX`E&<}9(Wa(otRI{Rov%1BtKafU2g=l@GjX&Na03Nh(<?xI$07e2oM8?
zioqC{5NAOcsUS{Ts(D>iZei?}Kwh;#uozn078cv&udQ#6wDKeQ!EY+)T5w}vIK{Lt
z_W_M6&oqmbOiIpCfDji~Rz^;q9K`FLHN4kqWirI+6^Em*peAAB!p)y@1Vf!`D&I`_
zT7V?h5~+w_qlh$AV}}@6q`s#eLayEyi^XnszUsP#ZjL3!pBPWPa<GII!dN3*8gxtM
zEn{P2Y?mE;*6rfhm}QN<1y7d;2Fm)LU9#IPd`)udBHl5CjXpfyXE$ygatKGP|2;b&
zO1Tp)7NbuM>ii+J06i<Wmb16mw66?4HHfs*OH1b6kR}VA_bGzvaeCT&W?uPk?|UC`
ziL_(`%i!09o-@Cfv)X`oK!rl>u4-xGbR%BTlEmo;GVQDlmk<rwZIp!)v+6#4ae8%O
zVwW)^!^nEJYhq!orESk-@Vns*1RA~Sl&cc1$7iR<vv6thb#3|D8h+O}5`31<rM7b-
z^G*8!U%S%>&zfyNU`)#lZ8h3<?MvS7NgGwKYtxRr-EL#gxwFR=p6RS>d-Ny?iA3hy
zfWd1GnVFbB2y~Z;llu>97LD;L3IeP;Wf%9Vy-htxI>v#I3)+*2E1MyFF%AOlX0z3*
zYUA7b<0K#;NjZC##KUgvAJT5qVjx3Rf*e|#&kg7G<{sLWPdij-TE7CPN{ayx5(&}j
zIU${L<nSJtup{cOaRQ)J(`Q%Q-rEMKO)v?E=u;vLK=btl;HS&*1b~DJ02&Ur+Qc#R
zdqsaT9Agv-lXK@>m@M`*yyrxKVZWv?x3>X>c(1m?H4<^)40w=kZwO7Gui!d~0xXc<
z^<X5i6#jsImnXby?}*77GH3K@Jt11_;ykhbY-r!N{&Y6bAqS~tpEq>@wDkp@g!YME
za(Jbl?Qn}@-><K#DQ_PBvR+2(F}!YMgVANWo_KtDV4{B|W(B?Z|9N{CI7zRvPPoon
zm$&ZsOW(VytGlabrhBGmdb+D;CYh;BCdmvC=maE_M4*5qCV<dvBEkYviU>gsoz-B_
zbtz<B!)15J_2Z6O3a)n@#a$3r?Vo<%C!hZOHoNGL-Q~0L{r}JNzV*J<T|Joy?sQV`
z?Y!qb=Q+=L?&mo@Z5a=|bT_Y)uJl{`y1VBU7Z%Ujy`i2T)Ys7i`0W|j9=fut&D?(&
z(NUHk0>*GxzfI%1cK~yv$8}-!xSG+W_v7T=zWfvSHv%8#(~z}xpz|ys9)@dV9|nCV
zsaCy()W(hL^w5_?tyUVjf@JjF23V)BLI}a&zC+(wqBONCKTqoDo2nj>zEcR@!Pi&Z
zH<2mNtA<N!w8FBL1z@!n{B{q#coN33K?RW0+zZZw@H2nsqi^(uGKym~%^J#!<S}It
zP4}_sp*6=I-CJKI=?8lhQx<alpw%vF3%(Qad)g@;s+vL1yScmfg=F%9SQPGFY(w~f
zTa&)u(S!cB@!sINJ2=DLUr4216bXeA|0)p*MX>$@{$W4OCrn=VPYM$xX`@9+ZCvD(
z_O}j|&I1p4U_qw3Ay^DuTmvvKdXWz>sTA=A@&SHYFxibWLh@kiIP$)@7Cd?b`PUTd
zU4ZhR>KmSbZXL^FiGD8IPN&;h*bZI6(l5SG#R^Ca1`kv#p6le|tr*tAt1OJUPFvKP
zXd5ctj>yg*reWTVJ#l0->Qs#ev?b5^tnKT#EzsaIb0L~c*>_Du6NxC2kl2~_(`iU9
zZ~i0ya)Hr?b*#vEM=ciHg45ZgQi&@_H$661G5RYq@UKWN^eHHysu5C~7OOLG!i(%Q
z9cyN`(U^><)A7khV|K=}Ha70Ob7R9Yft537nwzy`vbNbgb7qD2_)Laspm!#jg%}yQ
z;GjSXOj?#}3Mx8^RB`Y=i<B!)<=9u%=aluI_xPQDhXp;~u6HK50-HPaX|d_T6vuC?
zO`AH2ewG>wEJ#$g*a~N*Z$oOgYPCkA6w6qdSgC;^Rq0k+RCesTLj9e#?iQPH?A*C#
zli{Jw=DBmn*t9F6-(y1<DGwVs;UPX(^G)+PMAy}sj~2U+9EX%*Lz{DqFyRQUAi<MR
z;eXAX$T9J&uYM1jN8L0O#ODk%U&)ut`I%QCk|p2&qgZGr6i=k@PR5>k)l)BUV}wnB
zX9M!&vi7z^x6DTo`T43D)|3A)vDD>sB93ZU!+&68@iX!^&)gUEJrfwvIQCm(69O2E
z@AviDi@c(Uul?6qqOn=(!lFfjf<9FBLDZI`(RdzdlmOI37?hx2$E2nLA~j_!B@;bJ
z3|*!oI*huLV-LfZ>0FC$gR6(-1_b--Eyv)rfAq^EKO!*uCJeFfl9@yOMT{1}4tRsa
zuEpLVG|1;9tr(cp>2N4ydJM@J3_`MJC+X{G!2y!&#vj)QZW4;e<6Rx2Z-SP+3sIGb
z9|6=ICP^hrDS{J~M14c^caxDSLnXB}|8F7laen*b{!f3LEs`IT3r3UgdlQ@RzBf5|
z($eFb-pA(YFE-{wOtI*K+8*ZF4J`N}Kwng)0gai2T~s$`u{aB@o8NJP@bQoTj>00@
z<1HTy-go%$`|gH)y;D_)WHNDN!;TeO0{1@#-S|1oQOQc9GmjB4O(Yb%1jmTMTJAAZ
zEyK43v)~8Psq|F8JB7oQ^yEYaOWTD{VeH7K)=-O5R-M6W?QjfGFm`)$d+5nPbFR87
zbhS4)8=Rn>o(T|VW>DHO3l8qFUl`<$wCNh0b0mWt%@X&GQIm3nP<O}?93#8wj1kB#
zMGM#B+*n`FDxPLP_Oai_5M_IbM31+wYlw7ixVYLRtrRPy9Nqb#wqMQ>p&g*|?vdk=
zFJwQ)?|#gWsvlzu<mb7e>hN40f&%xKrGhOw`d}WdV-DK`5cIHo-<zt5XP>HA8;v#T
zSa;S9<Pp6SD2S;ZJrq|qImcSiFNl!Esh4}Vl)h3-q-OiwnRKFfC0*Jj{+PClP`WJ9
zbhEvRdIQILnZ8(qSk5ga{oOT`d!`ZXF|e^X)_2**!==qoGkPR&vDimHs&uyZWnTC6
z;v*l?hdtEULZ4b%=YqxpgQh!SEwUf081K<n0|U(URqD`@9&m+gB@ly)d`~n$I^NR;
zK_h$v%%22>u0VPR=;0=HbNvLQh4giM5{7p+)}nsXq6yk}W)2Wso`NH7Pf@-u^v-Q3
z;%~3f&ft*8Vsx7>?(FQaTKQ=m)D~<XAR$(azQuozk6;2?>ovd6xh!N~1Q%W2h@pb-
zS%wKY8$956XLpR;4HrAw9^AY;_H3Pt!dl6mafMIq8t~E$=pa7`5z>Z_ZsHqZAkO7l
z@|8ExC3SGqxC1brv(T|O@CA}rFQ#q+*1Ef{*(#l(5JXk_1^j?uBo=5O7WC5S1*8K^
z*B#B5B*bOJqnrQINIG3OJ;{wiW~&_8<Z{%cPe-hHsTkSt6Pbhib#E>nH~HKoQX3F|
zQfYf41SL6a6(g~d^|^sWf!c=qyUD#it}zPSTZLW#onQXB#XDHqc3fi^oC&|NHoA{X
zj7(t5U`K!l%qnm3XUX5L13Py`CZ1Kk>|*^@KTL+k?|k3~5mDgAWeazKP2C*=H|E40
z4iQ2)WDRvNDkoRUA1t$H6^{*+NPe*QjaL{FOv4kh)`1C#81Oze5;o}Y#pA~joh){o
zc=Gs(+q5&ocaLAh-6!JJ*=2pYJX?)BTlkbl-+jUlO0lM0yK-Mc`n%6wz4y`9Fz~&(
zlR)%8R<$i4jM&x605AdxGUkFqzelN!Z|MKMN%a4|Zm5!-TRl4GoS0<))d!KlIv)O=
zNv@s~U3TuMb5igB{Tn>e<(&Z~IMHE~?~wD3;C+7kGp91`deHDNOkHJluQ6*1;{NZx
zMwwt&IAT{X9#XJ8q=mb~-&@#_bu4~e$ZtjU=|I#-+z<ADS3hXh+z$%ZJ>Y}vp4gx|
z;Kh_KTpiWZuRN_I=hv&Q(GBwx*J_S{)c;NBG>jGX!4Ix6g(P(zU;nSslZyAr{UFyq
zLN54O;u;Xm@t27i03;bscVe7c45)wNp3@$D-8O63-j7Q=OsP0FhArXq2qU#^<X56D
z%y2MlRod04uNIeoY%VMm3t?pSN+6F{szRlk%2;ZL7L}b;G!T;h_eB0VgDh^}3&^`G
z(p{OB9xVFd#Q`=HN5~SsXUOQ_S(1m{Wc%OU>t9)~4oHBbp0%rkx!TreEmy<$W^uct
zh1_>gP;v#iZ|GB3=XCC(P|Mw{%lD#N%2zvaWE70l4WM%VzMcfH2?fW6pzrR3p!2TZ
z#pr`@cVA~a4rDw~xgEnD7YJ1?#BS?Z+cl2*x+<$geMNBUg4?Nw52ch_uYOeLQ8v&u
zR|30OUlhQO5zsFkub<D+cLn8#mg@AhHTpXw$21r<e+|B#BP&<g1;N{DyLb5{S=?wg
zPNvQOs{?4UO9v5YSRc#VERLVup_Za)1GzOPGS_{9H)7@>zR4eeP_AGJv|@*jr&4J!
zWyS}Iz(tWO0)ZB@Th5>|IW9TuE<=@oo#OgQ%4<{*B;`0+Ii;G?FV7*?^`b6hvBpyG
zh#%H=y17OV$IbZog0(g|Ha59tEsP_b*%sEkTr~CMn1$D{k{BA+PR|4&xl_^VnYUu!
z6BNNiQ!JFR$%22aZobSqtiy*79kEz5N}VZkP0nvzQCIb@sCHL&6(`_R0IlmHw%oPz
z4q~7rLk{MEy()qSDINPn{YTwP?4v__84F>uH>|AW*KU;Py6nYq<Ud$Ob~YK$li8z3
z&l)U;gW<DeMZ&?%#6<N$(>8Op=7g9+!?E(r(PO8w9L?<WS5MC))T7Zjf72Z|HN)Y_
zY*zIU^!`5vs?<XY8)yYufM5ATm^%MF+FSrZqye2~Q;=)?u>ePD?Stv`p@nQIot`L_
zGjLa#>6x_?=gzI2n3*=#kt4U=yilEu+WB~DJe9-_i0Nc<BAGJP>T&bAL~`+Hr3i!W
z>`Zlbdb-kDU&rb|JX>FH-Ei^J&HejJmBoembaFb8%vLJ7WTINBluu4g2ZGphYuCCN
zuwF!D$IA!smUg?nV2|`qn~YzbI%+CVN#8ZYuQ%07gxMrHvdY?FyaPLN$y9G#U7fEZ
z19mu!6wUQ{#G*TQu+<RK)ySQZX%n!=L4Gn3i(A2Dt+v!Cz`ZdxJ{z-wSkqal%+>1k
zsbq4hj-)%#x%c1M9hJC2oF_pm7EiD~);n7$SlGOdc{_uH9MH4KC2<F33#`>N&Z;ee
z$N(OMR|9b3s2A;qWKYl@<Ej#=bEvt(59NGNMOsJXpRu<1`_Oc;2(Ow8g0&-uYule*
zI>ku+Tmzxs+E@FY_m5BUo_YCPT0l5o2M8~2Tr!_yvYh@ZfA5^F&cePw07f=9dHL}L
z!-)UIStLntJDE-1`gELk<MHPmo%dd_W<W*J|8D{}!oZDuU{-T{=$?#@<)(mH2LyMg
zAX4Ic+|88%OrVndN<5yd^tUR>c>Ir(jt0MnmSPAC{hea5U`zp9?rj&YMMEj?^{f_b
zBwX*W3Jc*c>$&ht#oo(NL=_cO63Zz_6f2NPaE0?S&&^*<&&~#ev$NBW+#f4!=7Yz-
z>qt0y7#qbRB+jG`N5e<H>v%A~S%}^L2rNiIKyGXaZ%Hus@zYonU|Fw!y=A3L<n+gL
zNZNlzb54UE^bk3F&he(dEAT|%=YTz`G}21rSFcp1*6>B*l>5}K)oQd;yT-5)yEJjd
z{IE|QbC1J8jZXab$^izKsBUiwMw<IxK#2}mK}Q*#n04XR0<B-gosQ|@qQ^@$y&({}
zO3x8~I^OAya|?H{mBgTg@AkHLS9G>&RoPEfz9e7eonHTG*NA7bTZr58D)yacy~=j{
zWqkLu{`(>?{E5iYbAeOHgcZ>JMm=*cut&Y0%1X&yY_Mn=J@5olg2vWC;t!LEtDOK?
zH5BJW{Gyc0w>Nms`pu6QvJjmls;rugA6}ZBIaW1=r<ay^)>0_6@W2!pcvd|&^MQEQ
zy)73gtyP$RWv5$IyIN^RVWl?L4=)(Ry#3Ghb)M+}rZzt>v2bSLcd?q4W?P>5BIAVL
z3J)N^#Y!Y+EiUc+a}8o2`U@MR{#)m|IN&J%RM$|7SFQk@vJ_@>wNPJNj)#ND8i~w{
zx$rS&QyVO=xq2GwCMHcD7APwohyG!O?DWLK$y@ZknVdRyjHE}#Xcwr~1z+Ww5{W`r
zrmCM8F$RrP=8CIzvDm1DBr-D7^0^|D+z6PNX>!yPYtu8pSExSJYhW28wiC#v<vXv!
z-=anJwe}#`x?Ocg#61Fq(9X8mHEr6M3meo2oiV3P7=BUDHvIo^(0Y3e%1q#}*GMd5
zgL<o|w<@;l{dy*`j-5l);K+Ipji?s6qXGfMa&{1<cP~cVOT(tvO}!03MytgZFo@73
z_}6^7WL3y4VYXlx1lvUi#*A=c6p?Ph2~pezbbVMS7|Q|Jt}4p!ea`qzeXdxG-;Rs;
zoERN(2xBHWo{0Sq9US_sbr!UE8S@<LRSPObO1uLFFo-OggpqehXMyWn1dVyT*g)Y8
zbQpzjb7*ItfK@&^mo-0COeR8y>_TC@zZ(u?vQkLw?@ktq;gcqE6sh0+(F<lMbG~dD
zG6QFO;R9po%rrJK0$<E!$4$mY9T)wP6081NfM>33Tnhv$U=f@Rj(ZNx$B+T|oE7T6
zYne{J<As5Dv)Cx1I*8Y(df_E0iuhn}n#Bg_fxYQ=tqYic)67W1up148P$o*kMsy(C
zir5v+9at8aI$oW-@sAb0X#)F&O@0qUMEY0e3zPA@*~!PF(d~G?zbz-bPd@pi8DjyI
z<SSJ(n<b^sW@VuQ4l=(GA5m4FKhnnIn+a|Vv1`qOHOA*p{}kt6Tj$@QLu$BckGe;r
zlst&V4)?!$Sl_^l?me^U-eda$ffk5hi!#Xqz$LDGZ*l@oe)Et_SkJ|er$9$1Ai(?t
zD3AA*Ykv6>ijHdfh<lDb0&VPqrZlmN{?!ASgvg>nwHlv6X;tGho>Lqf*P^@1v`i*p
zl$_#NX|kg;y}!*c6ii%naVU5Vq*qHg2VI4sx7;`nvC#A(e(A*LQcSajB5f_EzpVjP
z;1mCaLlbrd1|`U6$l6lCSkdIfDNpvUMk`xZw7BU42J;IDuB44~jCKUUeUD81DHgK-
z{KK{7<=VrYO!`k{dJSgV&u3C=RayK-Tz|?syu5ta(qD7i`1p96T1(5e;iSj<9XF;*
zo!R!d&Is0c<-`P%v!BpkbSdEZ!bs-=j?I~4$OaZEx4^6H3VA9FVj4p(F0U0Zi%|%L
zK}XD%oBoBz=!Zay;IBYj=2G0QO-wCOf`@)0zir7j<7A7On)r|%M^U&o@Sk0aLRN+x
zuhM*hb0Z`&`<u<Dk^AcPf!_=K38su@n8T+5)<q#*YP*dH9l?A?!-PtMw{UXc93%%b
zS!|om8NImZ$wm|RoilPSPCxFk<&BLQdMvMqX;MCn3K~um*ch<XJ)hTxaE3|)_A~{J
zs5FKkxsH|fp<p;1Pfurykr1|@kH<5iNHIH=iDNe|Y$ux}xUf5JC>V;wGgB;;!Q!1!
zrM~EA0p`(oCQ8U8!?9F!IukS|m_h3ASU73pb~F<&B11BMP%0c9RSGeAgHnh5tj3%=
zxN84P2bc0UV5sW;=Y|1tP*(*Y0NZl_h_i3_4}do;VYL|ZInZbB;=jhsVJ2Ye@*i=C
zOtV~RMkJgC<AxjhyMeGf{##w214{BwD=L?+`jbGKMkHoO0C^J)h7xmANVj|X>EMlL
z?<s6-TsnO$k)E34_H-DTh#iX{CFI7&i_OrCIkTLJ$HJ(=P9V2aA_BX@m6vXA#!l5?
zVX8zT$kh~2Kt2wJV~O<gncI#Zx1)^Vv;xTCXnr1jEg@Qm)-gs{p?5$_K}sNYu5e<l
zZmN5-Ha=#VAOz~+RE=1Kr}RT<;vss0M|f@27O6I8e8pi@%^}H=v_^#v(iX?+FWQn5
z>1VzHy;9lj^S-2{`b*DKrrPO$#G&0H*Fi9BIoHnNBYdx;g)b>dzob=lz@$a5!Alz7
zH=y`pg(l3N`3J}j?*z<_k~{B<*k<0LGN-l3gVc9OPiRA+uE*uoK6hso)w|~H>BK4I
zB8|nf{T?<M4x93+<O~|zLCl2HsIImjzdt_P3TC*MRwme*{c$|*cO9<PMiYHB{E$Ri
z`aJ<Yh=WnNmK8^}0o19hDdgSkHje!hR9~$J6y1EIBI`TRgctvoLi9JawfOBmEeR3G
zS6dFBM%%HlSR^q@=_s@<Iy=sPQ-|QUUaa@o*(!N*zvtHCzCm#gxm+E-?etf1ZiP5!
z0AxxInmf#oZk(zf|5en(PV2{E3kQjIbsvtEXh9TAeeg1EjD@z&J?z0Sbm$<YltnkL
z=z&V8koW^oMUWhgl%l`(EEr25XK|Ol05dRQieTzcb|AHJ{v=Am-~8dp15?A`kFve;
zET)0v`%0q_h08)~qyx6uL4J6eO<sc)x21Buwy-dp!!G2Bth2hZr|_7_@@8&!VPS3#
zc3G?Z8nNqq=p`?C3AXmhE!M~Kxsdr_eGxGu_Dh)OSuH=&s8zn@#PM1+k6kB{sSK(a
zpbu)scQV+gKAO*0AyM9fu!krcvwiciV>8(r+#JUrt7T`79lQB>y<Sm0zXy4YwEG35
z-dxM;1u9@D2O(7qOT#2}o;H}xlOWKZ^GpRYe}WV^f+z%Wi*aqEOEmoCWEYYr<tj>C
z17XbxWyft_YNr=}{p(*>;El@eG?3V+!nhZ5@}L{RL0Z&*=FiNpnxBC^^%nGY5e#Rs
z0o<jU$zR%s`Ae4omQ<s6kapNd2q^qr$3sqd6xZ7*j)B29<FQ=6kUMp1bNNWIkcM0V
z1~vNj-yE*aro!oTFtf05A{`E<G#Vc-V2`Il@yPP#sZ+T^J{OC#8zXMdre>>$Q^9ch
z#KJ-*m`;a9FYQ^D`BldMVl^;&B_b1h!7&;LSh!21<#pp{BOr&DCoOZs4fo&w_UZUo
zFcQhMCkspKhYyV(FU6M3&t)Q!;8=Y6?f2h*!wtroybKZ=S&Eg8j~_a`zEqfOgOmvE
zR9n$F<=>(BRi%hZKx&l@q4LPiSHbt4bI&`|Y`LT6MUeJ;_h#*4I6gkHQNm*1bP&1q
z9IH@z8CGimcyHp@R|ZioUfh@%kB5tPb{TuAirpy_Lbeph!r(>tV|Hqo7$~WL@bQns
z3CO6_?pRgCn(L||ZrslL4vFj-_di5hC~N{B$<NI#4<-NcI`FiS#6PKIDG8x7l1!!+
z%4P8cjL7}7DsuX+)q;+gL%xz#;g^bVYN2NPYT+6)6tK|_!%=cX^@#!#hMU#_y%d<D
zLISE~1^%IDwXlIFu1t^6?yrJClRL-{?7|&h@JC=<mHE??qI{_bGcyYyje#DRuY$2S
zUj|IO$jiz-k6aEfP=z-}NeEkxnFfxY1EA4%nL`aIK*!No{=dH@?Ysd{jz?bp^5&+i
zbRIkQqODhv9r$*tsczkLQ)AIBA;YbNLj2`sv-$FuKcY8Ywe_N7$5^o6_3y_6g=w*I
z(@j)n@k}vHNtR!4FJUi|^Q4jv=IF|u5}-~q%sTBs?DPmi=-K#OJ~<*iMnJ};!*Q8d
z>~|I?_Lz@+L@}TZ)nD*HFCy4m+4E^Cw4av2Zn-l{PKL@?{WZy4CO`N!cvG>y3_w{F
zIs#8F*{3rfR0>jwMih2*TJ=A%w$tv@Sm$DWf`;Zi)#{<)?$U~QWZ~$swV3piJx1$A
zI9NM&bm0iCi`t^rO0#)GEOtY)xnjCZn7=#^jTYtG>`ce$da*FSOcVQ$+u3ZXl+D^d
ze)~J|Z>dynmrF|l1A6~ULKl%0iT*?MMpR3*7tGC*+w~e$K$2*<N6-Y_Cl1t-1|+e{
zy0lb|TAx6v@uhew9n79h$78|#xfB$TW5;5_Q0iPh7>lRRW`pTed}Re?LK$C~qeu6b
zF`wCg@0#yrLT=rAzEAp$m<JMR*qgb=^akXMdoM=T3!#+Zaq_`(4yxQOK9=U_b;hF$
zGiH+l7N9C(7Uw(hfpBxpmF^rTEp@TU?CMxvz1np~-Z`{@8LN8ELwoq?WvG9#L?MZ#
zD;HK)EORuuur5M`69qv+RV>bm6p1%mJ^w6eX<ho;^*?E+^TjMOHJ0*eYh~ra+4IK|
zh%E%#UDj}oO%lG%S$Ihggzb=x9<C85d%Wsz1M#xqZWKMZBQo^p!cB;;6um%R@GJfS
z@h}3H9fPx9God=IM5wAg{8mAmSp44%THq}BbO(IehqI36NJWMG*&pic(-0f3pVz}R
z;<7I;+E>^{6hGI80j~H}n30PTu0qSGw+YNSFDS|??5m10-+rWW6Z8Gkj2LObez;Z?
z6N!hw*5jq7|9YUyI@uvr?<E&gj2m5;%gw1UfMVj8vt99&G*?9zpFWJIE`|6fgX_0x
zB8;Xne*3|MtcY6_y+ZV3@Ei&S#4kXjxv&9^!G?=ZB{VU65Z#(hkmv(ef%N~1D+g}-
zBu$o|yp8Jzwvj-q=LH&jXW!4Kw^8A_jqf1U*0z#=jJxQ0(mvojB`&v3938i<bg6@6
zt0<|#8XIKpwtZmp47MNt7MkbT&!uhGh#oBH>Jd&;PER0{8jBEcG;$Hi2w=1hg7N51
zjDVf9@>xtF7*~#|8U+>`LFj~l6$u9kFc_bVb)<2?TA&f_gB<;wkN^#gVNU-t#`agR
z_vJ)j9{6__Quw{_FCiN`VmdLDa0CJ2O59tU{6ozc<SB~{V1ev>C;fni=D&{E)bbr}
zjyhoPP$NgA0mwJ-F7T$YR#OcHbJ(<lpAF)_>9@Z1t<mUPgTd`b9(g3Ybi>MW<q1R@
z#`0MNC?ZHPn~(hgo>Z1sZdl5`1BHV@l>b~l<|}f8zW0cE0L9KUvXAs?v$Hk+-_k>E
zw)(1#Tw`goxl~v8Ons@jxzxyQysGM!cxI{TJ$;21*CPHo%bh`7zYS&AMNClAdm<U*
zmqafV`B%L;gwYAyFRyuttJ@SpB0;zYZV%h1GA4X`*jUk0g}39MEJE&`o+^HFe7gTM
zXEFcpP`;>?qLzugAZ#+H?C|aI9|R-3Jw5S>;?y)`;o>J^g9RuU>j+}vm@nk#0!2jI
zw_v+_ec-K-Y)Mk6XOWN)1&Mm4Q8$}T{w*I@qfS;DecOr9_4_7Snn(k(20#J%pUk8#
z+Ax+ck@A$A+5FyXE-KkoIb{E#g@;3VMYe}LB0rhUZ$<4T(wFj$nN)Z9@_`{K?!b#-
z7d~`?|MkqDSP|%}6Qaqv@d(2G2|;i{!ZDN>H&@0J@yw?TFt(NYR3>g7<ucw%VfP({
zcr>vZ&%S8404tgkFC+9d1s!n<^@@2+#1v2i>OvY9P&&R=bJ?MLq+-P#OjRyX@mm-%
zNwllue;Sd#HnehkQ*^i<7C%BLZE+{Y9oZJKgY+W+W(!~tozF3*?HGG$$u?1hIEnHk
z+S8J|k>b}JzUT!dbR1h!0J7IB1JD4SoCEaRsH0YzO?-ot?`nPeMD;0h6`eDx_zAz~
z9q5ICD>|!SaiUzD_4N`b)%FLOwD(4=jCv|->8UhcS83WEHC{{WlANPGM2hr4JI7&f
zxfOkWg!9RV;MC~{*zWABNvV)FLhctzIuTx_*9s$jva#XBX2AY}kW(6%C<37lT27%}
zVFF<aIZJL;FH$eT&d}x$+y1LQVy9+~9v&-M)}jAgu*?&OEUPqj_~^`xu?qkB6Qgnj
zw$0+==9@Mb7mYRE|Kzv@#?n|?;-=PnfBXN5^}uO)X2B|y7S6zX$*&cxwwI*PJL`Xw
zb+Ynt%YqAxbTt3WA6Tz9{~lBnw8-&aIaHMPRMDn491tj-BI$7GDu9A}d`i4oh)51L
z=N+}k@wos^Xt4s*eFp@{sRzCY+q!W0aO33KvX!u4_)ex0*7($Pxl(!Kc(pnX;9<$Q
zCA%_5&1@*NxN_?B%Jx6}4|!+B;2k?QcxP<;z;~A4_<yXNKDDwC4$T_w@rsb>sL5eO
zup8Xi8!MIa^wc=FMuw>#4Ou?9);N54d+a)1!Dh(*R{&euD|dmPUD)RX&9X>sYV-pH
z8(jKH>otE#M2aPi$`3H6*lbH?#Hs1!2t6k=mGb=j3c1mdi3zhUG0aGUcx<i?yEUCM
z*?GDlz<aRe#{4{%If+~bw4vXwg{FW)xyCC}DiQz;lS7|mYJ+eNlv!$4x|WC;H8}fC
zwbSV;k&b<L903Z`W@8#Ao%rjVb&1t>t5E;g91{MicJ;1&=1Q)}A}n0YUCHF%rB8nC
zUHCUYySX_#@5DBkKH5op>GQ;&g?$zkI0Oq@i3kD6i@Mn@uPN@FY#6e3B2LBG0+4nd
z8`{VQir`hi6(_#gA80fhwn0Ld)maAk&#smW2qG`LS3KHgB9SOhPnUnR@}@Vvi6@D~
zzifTkA49M@#N$}!o+=tk(V>sz&15Q7D*fnBpvF=umHL;hzw%e{w7+uSxJ`=v7@SBw
zH=cUyrq!W%M`Fy3k+jk6FLCv4Z+lxYeIZTf)F7oq7n_&2dJ>l`t}Z^r(v=J8;(r+g
z3|D#9!MdZW#1l&qxd)sMJ@ZBLdGjgR8D$af6h`}sJguCAGaSSc_lBHL*SMz?2K>dm
zt}s4c_;j^eJ94D<B)nnBxAo-OO*gHru6}yq*s+Cr{YnK3G)ko%eYS&$n&cI*kJrJp
z_-VdWtv-qLTzC<w$oS%?>vg_(rBvdJJNj$~lRkchpTZoa{d*Vlg!!7+VILml0io6m
z-dC>=30gx8e+bcp$CF>HC!uhAVdCpR{hm%A9)``=U6%DcAzz=s8*MGzat>X;!5qz2
z+9K@+86UD}=b&IIPXeL@yk_`5B4UK^TAC`bK*g}!$bMB9P_GJ>dnHH*40~_GVFTyd
zz7=QzbltW14rLm|eYqBjG4#YZz1Pln^)TWBaEJurP7b~#d@kItM?A)W2<3i20wdG(
z0s_!xRrsD;0zSHkz3M=*l`w$}wfs}Lqiq}#iP`(A-rt{!uj7AvNYXV!S6irM0EO;>
z7h8T(a@AiqY)1yXC;&X4!dj)fpk2OG@P;xB#9GIop|%`~uZZDY<Rn&BPm*eLQhNne
z1Sr%-MJ<kwWgLAMc_|S5D~rS=2ri4I{^wHIhPwchZ2<&0YqIqkRj$h!I~=<hwI3ZB
z-6-1?DvaE)gOTOl!>ZowI+fdq82(GF>&oGKh#)qVLS-z!$fqWiNxTr-zLgGy_Ru(R
zPLk(-1u%XJbmV&ky5vOo1D=Q$3KSydtR63xh?idAf`^uM2%;*LqHI&R7sdd8PD~YE
z=c|-S8EeJr<tU%1<N8D4>Dp$zSd0UPaAVu4WVw{MrFl9XO55?5ISAU*;~$zPeA!|1
zc?8wX)Had!rC@?s%a=-szcXdQr(GByEAaJj^ab&dS)E9wqgE&wWX<x0P}m@>v1Lqs
z8om=Bnr8Cxc(m{<vC<@8LA$YF!N7}(*OJh#jSY1<SsrW6PDWt_A?%b6-^v&shw}LC
zw`zF=Bw;)Ec{zN#urfdb%PkQcO~9}~&|RjRrFbP0l!BzKhb(0&%3E1nv-l7Nb1(2r
zgXvL6$KUr3Uu6}%>ykFD8F)fFuFgf@-mA=?B#KzcVe9J^_YQ46!r-0tB{>=PA@6*v
zEB|)&(DgYRJFfSz-{~RXx!vLq{&m~!-MuV;MIAiui9XDIHv*MKMhrC}ZaJdJVCPPQ
z_+Kv1sHD?r^Y%=kkg2EB{dZ;x#dO`g@S%qu8hdCglP(rA=I!ZpohSWw;=~m65$8+@
zuuvWDoJfE7VeBL1BZGK2qBuugwJkk!iMZx_bhtBwAdY>F4t3qmdmRej242UU!Goc7
z2m`rhfB=I<Ee^FN+g_8fo}SrWTU)!m)4%GcuV&cTp)0#Y%5tNm+@oSftBMf)+Y9pH
zE*LygPG*wH%%uJfk$b=e`)OcOw7`qUkw0(+5f3*x`)PQv8Nz^qAPIrh5WGR3ZO0l5
zA+fnAHf<o7dRX=-wD7S&(e>n(-T-M3VbGHMt(J%q9sW9Zc0I+?JA>Z`CH(m`Sy#U&
zulJnFr2bY<>Z*=XfWfW48EBbXfR2#TnuE1F9AC;emCB<MhD5`|GXXl7JC`sSX-l9L
zq9gBj@P3Q$dutv59)Vh;p#f-o-{|{HHRY=S;OAVJ4D`(X(gvLc(1lQk$XJaA<KG6K
z{K(!0;r&(s8AzS?H{7^842;a+W|Kjk*LYuKQ_Mr=9OWL!*O2qLRf^=4uK^9H5IB-w
zXE2a**6Mamhcv<qH1On3?EB(w`^zKHk84-~!!G*Y;Md)iS?GbYQY8mnjdn!W3`e1V
z-^g%vf(K`Ui8(^Np7-xWi2)p(J$?1t0e(AxHNem_D1&>S+Bbd`f0Z)+ZriVO{D<0v
z?vp3+2XnB5RZuQnDvR9=cz`(tY@~INvrZ?Y_%&$CRr-?uYN7B|Y?;JGpKXa5Qp4U?
zbEQ)5tC%WyFqZz;$>hIIk4dE8)vH(C5xe@~RO-VRhb;CC{43ovkakJWp-E#dP26Ik
zf<uo@B68vn1Ee%hms{c;I<OepL4EoA{`YPfNH{wFt5aMv=UFAy1}S=@M0EJB9eX32
z3wtgt5TNqGfjx6yNv=~yNpp_q^=9yF9z(AQIcLLnp5(*rSZvifT4z?y7#btr%o||$
zc+olAdxxp<xA|<<IU2l+4kFR;GoLX(Ykr-3x^Q)d9d_3KL5PL~h@6tZ7T8kE%VLER
z5)5ZCBerknIfmGdmf!FK^JA54DzlqGbiyxqXFx&BMaT;+%)=k$EEKb(azj_;R|(#g
z<as0lP>QE4q&$<{Q8wZb>0xvUU79mo_i!C<T~6IsiNP(`ZV-KvrfG>;9lmeuCr8|U
z5|(ZERj&>j#hudILr<6Zq&ifwyj8W$(*C<k#vWCc4b;+{+7|GVYaGwYzH4sm0jChm
zy?~vXoZKi!nN&96F^D?F%6Q77oP@a(4%lfUR#L^;2qXI%iy+Gx1&jM{xiuX_iiCW=
zFjFdFaZ5S`XZ(!+ypYc$Q$s9$>n-;~aww#jjsXFS2@A=?llW&_;Y8M?)Y6tuwpwk3
zF-DV7+vM_73~=Exd;htOhgdVyZnctmG@*Yr`7qNwB$K5utHUyyWGHU34<p3ViM@su
z034?kxFl3hogXT7lF7I@KsXQ&AYZ9a;syseKp_<Bb@-bIz>Q{iA%DspP>I4^<*1#?
zFARoRq$gUUUeI;@Y)BA2!(vDk3zL+Qy#lj5oq=JM_n($4p2B=}A?^3K1wlI@yP+?$
zurA)@cnMEJUt$B<dXA`!wy08TQ1khW+JkllS`J%IkFX9ta)(!=Q5f#IT{v?>%|xTs
zTz88x`&~wsZb^)#KWX&QhDjujoN`x(ojL*wedAb%g|pHfv&BIZD<s37a9`LCV?FQ(
zPvMJ6QC#T(6j6OQ2W10OSxM=qr&o?xLOHZVbC5U8U6$GZ52SKfz3C;?avvd@%N*j<
zwl0A<qynzo!n1yN<ZXFDc!Ao)7dRpnl}icYlQNe$8oEcN#7ha-YfTz>)l~p?Kte`0
z6tD||;Eou>Cnk<i-CCKSFIQ$LcSn||vBx27<o%r&!Qh$8ZPCNopk|Lv<kIP??BF>+
zpAChmt0CBbUNiv&ZrA6I(fIyzcrBgAo9T4phlq|p)L<3P*tmT_mYwVD1G#`^uO)_w
z?~N10wy&cH!)<48p1u8lytn1-jm~eus|onZ-VWJXqFYrs5b*$r4(^I^Si|Jtyk_W|
zA`+*JObkdHZoo2)qm(KbQp8z8f*=GDoy6>id05}z{NejlHP$Qc+JfEs^mHCM%dwY}
zch)|I)I&&^W_}Q<TTH!vs+3BD_H&jfrtu7y=@g7J--+#x6A7{glJ-O*FQz1pR}cBj
zXXfT+kQ3cIYvzxxoJF;${_M)p{8}WNt1Pj;j2Tk}eN3aImnykz<diWtMbepq>v1h)
z(vdTdy2c}*0`R#C9<qwq{g+ENs<MF0cIbx^LX5Hq$s1$_m9hX&>5e|f(*f?&&8WPY
zz~Y|*Q}vGU{!rw{Xi}MQtr!!j2+4UOHGzyqras>c&(1EBv|tM|QWnxF=vyRTw_$9#
zVlJ61PSD_E?$rM4ulXqN#oprTuy2X6R@NT15nuV1Ge-`mj!a^Tfz;g=D;CDxRZHWA
zV(c~{b}L73O&&RN<}Iac&IbR9q<%|BPk#ZskYfkw1yabMv<dy+x0f%O4&6`$kt>jJ
z6WtpAUx$*j#tpOxALU1Ma8d6e{SfWUz5;hUobIbc4RNwR6KEloN!69FO=sxpJ@B!^
z)lP@|(!j&MRPA?7yNWabG?52HlkNC5T+xww=GQ={mt|FgnwoV4`76?&t8>cKaK;pq
zrm(SIuj`VuT|j-4%NLJDb*EY;47684!RniU*Gi$)DikKL#J`8D0>N6i*o?_ew@5%3
z+qhIXbL2>}fLi)et^ln`eH&nt%tNpoI1lJL=CCzFi4#&eyCcidZbE3Yfnb2)+9=a5
zNIteWY>9Ug)KwBq^aM!=P8YUSE?4MdawszOouk;zNR;`yi&?u}0+(-g*g1Uzzkb`$
zEY|zHHt=J6q+eoy+I-j-SUHa8=o51hgC;ZfMgtk#Gxo4Xqba3i_TP<+aL*Jjz$IAQ
zv;+wlO>-JQs@1AAEx$bqA6?2MTRfAKoo)P+Gt!nV;qda*s*bw5z1{yB2oTGH{D6)P
zPR$O>i=?x<<(_VttE<?06d&6mu|gb4#;m#zf$+{PeYJ~q^Lr|l`rUkb9pJ8OdwXmE
zVRv-|EwG?kpjrI$luwtOD+Vdb$ZRAQF$9Mg1Xu%?P!#Bf9YGc`O29k}02iw9&1Zou
zN(9(T<j({U)@VLc^boWo0u79IS?PWkhJ@SKhAnFl0DT*w^t-EsGJsSC%)C<%&Z{*1
z6{;=8MHfrmqc-Pi;r8~BF>u%^yoHUB_I^OV<=K8o^BSOemxOBoW{Qqf)@V@}P&x;}
z70fn>905NN#kus82<<NN%po#`G*I<HWzzafKBBCdCDy%wB-p210_mrRRoUKdFrRxP
zyR&M10ZK|LJ<{nj12T__pMV0vM+^a?7R2V_*n=+z?h|Zee@LmAeN1o>SM)A3OPTx3
z^IPc?HI49>nwjcn`ZitVgoB`>Lm#?M%E|15k0cWhV98nffkg6=2St$BrX&%qHe)HG
zo?;)4g;cCrg{NwZJ(1;CElEoQ@p31$XlF9^Vh9^(sTQ--<hmpWp<1k?iDzF?IU{2N
zVaO1>V&^-AEK=v;!@S3%)iUQi{hsmxr3<2g!~epPF+~YdN#+};6{M^+m6P|KSJ4_#
zM%QYH8cC)~cXD)gWw6_uo10OST-$Y9*V=dDdOwi&v0QtgQpm>_=3GQE?f-#n9HHm~
zN%Q1Wgw5`OKRVXCkVR?V<l1sqjU;y(vTZXH8@sGX?4D_yt|kd9S6YK649@yJljRs7
zrKk>soIp=QZ~bOZzWL?{@sFjp#0z=VI~{K4nQOI^J9~0aWQt)XyAzQ^rL7*c&0A#H
z@4d&jmjJ)pVtYq7v;P~dmOS3`3RrCwT4E2j%4z1(9FeVBZD?WuT;N7v0~4C6ir!<W
zb6wT_-uqtF<hrH}oAJKeUuG%X=HXGX`l1N`jkvFKyM;FHx@{bhD*>~Yx(6+Siq-1c
zaaCULvH9}8)^z*)^uF648POOPJfM*Z$F9gs9BaSDQQGCthE$;STTGOLX8Rt0<2iiU
zmo$3-^wIqh8VPy<{BXwFvyKb3BaOshx^XzNB?2wd5=hS5{>RUyjg@_{JpeB_)<*C#
zAZ>gWY#f1aBQXng5`U^%=d<h^GNQ4gD|Oo120UBL6X$t381gBABTrFY()`U)V@l7_
z-Wh{x1qLnDTV=h?fe*_y!_$D_T60F$duc?qmgr0!p_zw#w1{>&szy9g4y(a*<<U`G
zcSz-02<j5**ZJ<8Wy+<R`M=2mtlU5R8>Kg{1BiwDT`R|@B=b3Z7c|DTNsasbAu5av
z(d&Y@KGEgDBMDzu`KkVbc#$qkJ9(y_l0MAzex1M?*!1?df$V%B3yo;Qmdhe`2PfOl
zy#)e?Ko!kxBU&6QQ}In&4gIkS<jx+?_IS1UtC4o@Ih79}3FKN-8ycl0VgO4$d6ws2
zTSK~P1tKNAC}BIOp<2E8ERS{E{rz>K5d1M1guBXBWYN4*O<c4FXNAdimj`68*<s-5
zHJ?BEEIoRb$AN&2H3pxBJoX<Yu6$AUd%T@8nXIUGb(oxAp;<tmMfBRI>V$cJQB<IE
zGv~i3lmRE)0Inc|(NF+@Ut`&$%?zm>{r45@=)^>^VB3=uMFeO~PCDm*QJk2x?Lu*4
zB5D`Rjz83(-}Ya1{ofi=V?LoTYei<UI5`=O6!l*JtNKO#I#5J>R6W3qdoNuphy3WZ
zUM+LLx~V{jG{>oTJ#dl!!)vTDYp7yR1l(}Epd^M|Rs*37v(Dj)LeCq+yRFd9f51*O
z-<gI$TC0@Pi6jz*Oiv#=G(DX|m}#PjiIvisi4?JJ0(hSC!GXQ{Y&MlN5WKUASRz@>
za!b-I5@N;@&R%(1BgOOqc_NJcuvo)thr6#0i-;N>A%)^k-B}HiqHhHfuMB5W!PMN`
z(W7&7ILU<L6BG6N#6+CwiT)9gcxx&cNiQ*f=J@zh2H8|*G=gL16N8=u3VR^ohr>#s
z#Yp69;6!kU8ObPD_TvH#N3QuA_KA@z61TjOW^X&gUEaYG9x6NXbKuaME&r%i{k`nC
znwB%xfdFiGj6IkGpFutcM?dm&b?|j8V)H`WZNYn*_JH;1>V5@8skA64J??|U{c}vf
zJp*Z7^(QKP(4UELmud-OH4(7sogvpaYH;7@ap;}*4Ar*hU8x0~O|!!mWE1FtN1(mj
zjtIav1l|#NJn*5wF9g02_)mfVEAWj$|3}Ch+!>l(Rh3w5D9V8bxAvECc+@pY0GMh^
zajKOCnAiNux1+@VmJXP;W_=MUsWp_9J1f>1`5C&ub-;a;clGjZ7=4Jf^5d)Lx!xc8
zuHQl~qFg|Q&p|r#?ZGFjzBXPNdDmC=s@EUi-R<Xa*Y`Dto<iE_!4CJ9z`FuFz%e%l
zK>RJbq5<TS9tCIt;MKIfs9wO}n{_%P#QR@?5eB<p;wj#QEJO`#9cCB?BGUgulZ<%%
z@QbPGnDt7sdV)birEnSEL0uMIYQl~X!HPoMB89l+OtZeYI3LTw4Q7gSmu@-Z=Ex&$
zgh8N%1cwdSaSCIH6N%bvrC3TP4i)pM6#OzR6POrFN2jK+@U~PcivuB>OHbtUp=d0c
z?|-cnOHIs8Ww0eS3{$c6)_IfWM?Z&{&hzP;k`}hM!Q6~QZcYZxZNcQNb_8lfFnJbP
z9IcC?@bj~mp0Hx!i&n&X;rnm9?Y8(WRyd65EP4vGALu@7fa{{fkgEE*cP&U`bQC#S
zLv5>vC6PRGTgh)QyX(Gi`iJzfltO7<edHtkPx%S0_(FaO^z>J;wqZT+k^p2m@ze!%
z^%|%_K_YapFz`6i1iIl!Wkmr7ABnP7^V<U$+M!`pZiJF~G=;p@onQn}&h7~a{yq0-
ztDwtB>`?#r*hcqCpPOIiTm2T_<C*F4Ou~=uzFRve@8T_|7M$_EJ{1xgC`g<T*B4(P
z%!#vV=7F3kZEV0<J#-tWs-ik21B$ljko=(>Jtu#V%{`XMJeJF*(~mutPQTCF?<#Mq
zVidQr<C(GYp-_BmtO#8auEkaJWG4MX2xQ9qP&%E@zWd$T+?am{ENs3@hh+s@4n`As
z%L*5Y3k#cc8n)OE6G%Ptx8@V(3C8{n#Zqw=m74|g@9cE8(C=14k?||m*mz>c+*in^
z?VZH<m~~}55~@HqV?D0y$nh2{sS^Q5?*#g7WBPjDYx=6yjzcAyY*vlMNDAsp90@hA
zNsAeM+d0!~Hlh4;7&gBC3tv)<bC`OSly7(6%@=QMZ0K+{TVHZxD;xnwqh?!Qa(8ia
zZX19>WFDbaQOJF-f;WtjU^MWSE1KvB3MW^ua`Z@IM!A&Liy%NOakM$c;*VEyG;l(T
z;h1J)#82xd#3)Mky#?>_biGY_KI~GaTW`c|U!kv&-}C?eZ|Kn~n?bnF82!eVbA*|_
z*PU#Ow{hf?U-y@FYwf?s{Sx3d*oTG@)b|BmC-cJFi3W6<gQ=6k(b9zWSCmS*yU{G#
zjmn024a;`JcI>A_Mv{Bg>V!dv%K^cN6)J+|rhk)~$L<}yzO`M=BZ%nG`ug&Z{YGVc
zyf_nm_@AFTb!zm&)*DYe7+jfITwMIDf6tmX+;i{()%N(IBg?f$Fj#x|r$Jf@wZr#J
zz{r2_3ul+^Ji9uLY?S{3bBJ`j7wDOP16}x(;M@i}2IK+EoCE`E_~3l<+=H&!M`sYP
zhV<K2%#;?Gj5Ad_%6eocli{%X9euLvr{y5=X|{`vGhYnUI2W+L+ajqQM-Yo52NT($
z&x`f-u(t?o4MtifBOjE)l$K0km_FX*CcX4uYa}YBz1huVvQ@5-iM#GdryqUx39ZZA
zb5F+UQa1k5m&UWs^ZoZt7n^<ASsQV9`7M7ZJI>sf{WVZm`k()E{^EcC9cO$1hYmc6
zqzT`?UOp>C5A2k7eMDz_2LaA>_V;c7oo50E0gG^GfcI6+2e4Y=DDc7O;aD3Wfi7is
zd3gceR1Pwb!PNiyum3uq-zrQ#w5zRu+dK35_ZB9`u@`RapKZPEy%)^8QG_?Puy+>6
z;hnsAZ$97uz}DN}do$6|<2T!+GMFJG#$+VQT3wqMDXVv%<!CKr8Fe0!=<=RgaP_X*
z)mRtbb)S3IwvA@bBoS#2DyhcJq_Ij8ZfRi?SZ_n`A}&f;+=jqQ9QM8O7P`?7dgCp4
z<F0RKW&1wwG+3s#8@pMC)yLI+g>?x8gvB*%1s%as!h|Nk*479f#(WS^Ean22KwH;X
z6=%&Cg^&(;g*kHg{3_YU#f$u1J%9L!>?-2Ms(|sLXx9w_kdv>l<kE%J)%cvtZsy{v
zs~48!z8fHemH&tl`k&8y#(Fnm!y6-ZLUWtQozJva9C>IB2Hqvz{EYDw8eccMpT3>Y
zK3I{4r?Kc<*R_4dyynP}Bl#j5bj<AS=<I}_jl6B?xVFAJnf_h3vm1~1AHt4@5yB$2
z5Df0=awrYF&ph*s=D$IYn!zNXg9!tI1UBqP1VG0hGS_U7Rl>k#M}9Gfb(Q)2v7;}Y
zn0V>YV-O`0@!T(BfnYDV1=o3T?Uq|^I)5aYJaYb~TW(n^!kNB>r7#Yy6B%d)5!(_g
z=1Md*2Xq{F%4Q_(#0c*yT;rBt#1zL>(bk9=iQXYUB5BowX>+G8-MY)PHLwI~GAGTd
zyyQ?mcjv9Hf2%FDlY77QK$Cwp(874rZp`=@XpK}NfSM|RIn5I!6*%FH2+N2CkEIIK
zDnU5F2~G9*Lu5ys2N4^|ng0%~Bi${AZ<@~Kck{XF&52~O5Y5-Bh}IQF?xt|Dn@;`_
zDTByoyP`T%)i$4rXWB@mAFPH7_4LH#+|JJ2<V1Qpzn-^~#YI^4rj!_t6++cutp8mA
zg6la!e8|9D#T@a!D0w<%1%1<56C?(Ik&jd=MfU)G2Q7r)iNpjon%!sbfhjmep`V_e
z<-%0^HEBB%SuY*BGZZ|0;>7Xe3;D1e3{6bj5RRC+yV3JfCZGCCzm4e}-c0Ayh<;4o
zJ!c}}8zv?~K|7pZz@}0s4hKVb9)fBevD2@$!{O8)`^{S(FQp9|>wOThgiOqeOn^O6
zNAx|kw1J*L+yJ8Mpk6Eg+kaym?p=2=g!jR8`oWQj-`>KfciqL352Bch>yF(UZE2QO
z_9QFApUf#DIsPgUOfT7{5AbD+?0`AIn*NV#3T3T<4IXW|hOM=!@Q`uk$YYhkNK5LT
z{<f~O!TaH2BBI5M+J^BVY#e#MVbkib$+S3Dk;Vae-1f8u^yT`X2Ydlt?^(Bj?&Sk-
z!z|bQ2d?N#VHgeKP;r?2%e`ftFHsB5GK`og=ZkEk=o8B1$j3+yMm=YJq-{R&)VBdK
z=4|PZ<<fuaP(mkhb(ItyoAad8o4MR3G0tYp{MHr&;D_(?<yD8uC5E#xY)zeXIm5{*
zrI%M(9CHdaqFPjrr1ueb1@zwgeEF)rhv8J*5}MB34IZC&WhHjipOs$2eN=i9QUW02
zuU_yXqkV$?E~Ap4c4?Lu^OI%t|MJ{i9`Vdf10KQFUSm#wTM90iqZu$gD$%#BI4GAF
z_#&gNOUYE9N{PORU-Szmu)&-mReV_hGU0;k@##_&C(2%C^t$5|rh$xvhTT>t71C1y
z5zt56z=L%$Py)3W2<DVv*aw(~qmH>s81c~bCdOA**49>5#wWZpaJB<b+ywlIs~WXh
zu}~=1Y7OtqQA@P9y_3t!6Zp5he9}Af(ph?ETmHmVyM@Bs+~VTgTmk>wGe^-=5CVQd
z>;~EaUIZK;EWPX?x{l__xiPqC0$>0JumEME*ai`sB<gS1fAN)S7XPaH+w))e&o}+A
zNIL@0d<`w>4%(sPpkuP`zu;6yJJ1@mqvfyEe?Ihq%l@wowe~^(J7@>(Kf8bjW960+
zEw?q8D?>t!5EbO=CrkuDGCP1hC0F4X+|XEXKen{oiQ+{V^n392+_Je@tBF`?o!a~k
zinli%K7wD!&C@f>Wy_EH8b;#HG_C4E6s^@ZH?gIKRsN36Q}}J;^vL0xcw?EHG?(8z
zm_Dd)i07;fVk|~Y4}`iwxKc}u+G*{;LHPZ7r0`GfBoaHx)JQ)3-FEs4XqSB%$t|Ln
zBUps5f+u}bD~OVY$JOz|DUrtTLy#e)-Q~XoFK|OmaE+So5X*-Y?YVRv1~>Ylv7iQ%
zSbYh)-P^i=-%8I|)=b)>xTgLd|DApdQkM~5O@9|BfRwu*O}_{DspF-U4kza365;e<
zK!@*aBl?Z<pZsJM?0g%w^?+&udl%OlgC38<vmvMH;A?*Lxs1$*9Ao(zjU2z>s!#m3
z=+LGOp1O`v_ggav)uFuz#UKx+=|&9FObe1&cj$PVoaU06T01;Iy652%y4eokhwuX9
zGHcJ%2Y4>!VzK9m0d}}8g*i%t5@DyjDJMQ2oj+*Bo)9}VEMHe!T;m5D)BoN5jS(a9
z-f_X4xJJAztshi2d=k*e(J?s$%;$0~vhHJVAmbiAL_d#W>2!+)j=h0}IWGqE=m=)U
zwOYf~dy*1NToB-JXkJr1EFzatdSxOd=z)UugEPeMM&N~llz&?Zd*!5l4{1YgdngjS
zAr^^P872W``MtE}b$lh8P2r_v@(F!~rRBXL+=gEYEdDs=udWzUIU^^AgPKrs*$-(T
zl<6d{JGnXce0rBy{_=MdIPX@3esXmCw)ynk1m(LQ;5*$iZJ8=#uLAEkxZ}}+6X~x+
zTr?>V79}kdfv6;kqV1$wlM?Ie2+iU&ReGakNMuN+{GC(vQYMROA3b^$TgsbY78au{
zQu`z-V<ot8ARujFL%me|y~*Sc$C4q;mj2VCHMZQXs3;$K+f}V91zFw2dgXdGYNxOW
zIh4)L%p9Y<bR#yci@hz8%_iQC#dqlzf*o{UEcZ2c{g{96EWm|$)^qfr)xgcy);<L7
z6`iPjh}aK;%Zcj*T8aX<K3XYRd;Iao(aO+fIOQH4Q=bxx)<In5b1<ie0ISDSVfT28
z0z-fI&RtK3a$+@QfE@aYT~=h`Y6Kl8$N>n~jYl7hN9@eij2($T7>$d+$xnJYp*qp+
zM272!GKuZzh;|%ILy%%h)LXSVaEki!!}zA?-e!zO`e$p0vzL?Pc5#-sIA$72I~D=z
zi$RZVDnSg|Gx_#)b##7{y=?_R$ZvLG%jgEa3;d=C=kF@)GexXU^1K6_gg#YEWGw&n
zNv7dI@+QxBKwjm)<^=TBxau(7sj-cXi|(@1Wvm7KK1G<F9X)<uv0h(xpI_YA7<23~
z*{ewN8A5Ig@~~E834ut$&9I1B&wP)W7}R;Z)M)7$`^`Oy&3W|u4$$rm$|MMf6J=Ah
z4Z6Eky79G}kWn=iB*yD4Mg|b@_ji4pBb12W!CYTKEd9KrSlXChaCfxMrl(_3Y-SwL
zZvJ{@TTix4l_MFPbkc-{CT+7Yzu~lcWukH-n;j6J#pM<4O^bL<I`Jpw3+T_0!0EtG
zp-(DjdCqOcW@3^oPV+}`Fy@9sMheSyYY}X!z#-?r;}3%u0Sysj5R^Ak6PBd9v4Q76
z87;?a-sc4NJ@@L6yRb}B`q&aJvQlZB428pxN)wSyTI@KWSUu)aL#zM=6^&0}0%ua0
zXfza^pKD~WJ1LW?SynO}Mq-B~b`wK7wN#`Mw{h15<CDRODU=RI>-9z?k+hSMNHCm8
zq{6XSxSW8Ph|iG1uhi%A9e$f__z~Vk6S})SQg$|>n}usX7>$K&JDw~R$ByJPb|@U=
z=A9<Y^}u0dkHaVNee84wOQMCK5@Qz<#bV_Mmi;1|9QOE3a@}wc=LVKIo4D5(-1`}k
zrx!_{31VR1Fs=<>>AH}8m^OrWpi%@pYC|rVGfeNBu!8n0PHM-no?klXINIDR=XmlJ
zb`Xhwk@@%j`^~@Ldwl;f9h0gvG+Kd!kj~A=_zvHF@<~^AroMO^bJ;rnINEwWdb0QG
z;GPYTvI2R4Q;fI-P}yxXq^gZ^t<X+ux_Y-ONrE}XJF*tlMJU+?x;NA*w;tyv-yv@9
zc3mvs{0ec*VB`H6L?XTv){Ohmw}VCwGY!*;-cy`Ou6;jk@Q+blVxu@(IO8nDur)<E
ztUFg{-DT$|;JG4p9y%4RRP4h?4<qew^wc4}Sv=uVj8m<{_Sl$xxOM7<larH)?04Jv
zmu}-cn=tQnSI56!`})zcRetIz{Crfe-_JGgtm&gqar>>OKKj>3kG7`puXPkYMU~N@
zV)YQyG(Vu4H0E&x88pJ7aRe>qgopJmKtk9>^&Or#Y0)4}cRDIYcF+OehSi-`r=zx`
z!FNu;i{M$thP5zY6|6k#$B_tly@dTff<p=!GYJx1)T=mrHQ?a7s-4Z~_Bx#%ziN2U
zxx;@u1Bsoe$1ZgcKZ?c2{D0sb7@W|CK_6`%bEAPCy)p1+$?WCIJ0ztQz|a+2MZXdP
z3psAE^8qFRum$)eDg{@8V!i{{6cPniF-+(oK!JK|A)sE9fxbu;_33yz9ly*tZIGw9
zGk2<S&_dq3{2um*Y2}``rWod3w6QmEmS#WvvtcQq@19tbZN&>k<PInn%kg-dwX&vp
zb5twn|KU8Z<H%9oem`z8P2D9$JNML7p^sHrEvttXgIZa8wUaYDOG~5RhxKWUOQ2=2
z4EEy>RT5oU?rtCo#oez;%kH0#N}nJO+6ayCs?#aVs!@qdR@L3=*8#t^=P~OWF$>-F
zF0Ki`BK|Fz_xl0Z|3j_89CZO0Z}<NO_bxGIoI=n&eHI7;hGDGk*tIC{HDE=1ao{1u
z88g4!bzxz}Zvz-$;QF_G_sMG`<5LO%mrm<dd+5;BE9bDQSG&)Oh;ap<{&ueKg0}4u
zFuZO|#QgI_N0%wCf6I5D4TkVtDpZ^UQlJh)u-hGm;qI;jMLk>xQa=Cs05}9^;jN&H
ze`ZC^hXaR!kq-oZ03y;NXSG=T%b3+e5`zzXbZ;95FL`2*7txrL`MT(cR^>HpUPW8S
zz-xj@4Z`YtgS2Bf>Av|wI<x9B%t%(Du)3<pY^wjK*P@^Cw6R?G$lwT0%pDREsbuVU
z2@5_J7P#;uo0~L084G7!MsJ<;gRz~B_iw!hIR(LP$N#PC#v$G0=iUgC!_V+@=cni8
zJNVvAxkBFz`Osc?(@f&yDxe3htxsdJ$v^}~l?73g2vu;EhGTSO@f<1VfY%_@VS`!~
zS&s|>DYU&PMnsBrxI~gU-y7x&sl7msxbnzqCjG4No)OW725yQlI3AkJ<y<J2j&pMi
zZmeOOpNwbQeh{CuT;VAjjzxU1-b;A}*8N{#-P|}=kD+igRw>W8AfBH-&#&-n1hv}R
zV88h=_O5&>@VZPKly1AitmN9W%ZP8_|Dq?00;Rcq1?I#}NCHeP;#tjI*@yCiIaNfq
zYwXyhDNn|8NCXA%=By-6dF9O5;$h^cwhtvN7^I5*ze2>kqOo%@ALRYyQk@|l)9G{t
zq4U8+dJgIpqd&3}(`$>X<430>wY3C4^g)2nM;+XL{wey|)qbjvT>7EG4>a8&Fhw@D
z(0N9V%i-6|DrE>R`ag0-I>eB(n_<Qzk}6|<i~XE^Zryyvo^BfEx)R&`A@I!o>zVhE
z9>?TX`N$sQJ3#N6<q<pWn%vF9IwbucbuHkq&G^1IVkN#8mObPbkagZ5vaYe;_dTY(
zb3fSsQT?EKc<@0CB<w+fD#o#mynCb8t$TKlp^SQ<H*UTAJ93Q%4S#P$Muda1{C&gw
z;Qo$W1LA_K(B*ot{cztTdbE@$D1{3`Z2>XEjPsEu&OVJvB9;zFpPU0fX%(UbC7Ou;
zs_S29dNI#U??cw&oDJfT3<mA>?D+Uha%OxyyKV=AkeHM9+#-lk6k(n;gk<#qb;F;)
zr(;K=51&fleb#f&csi3wkKY4oZ?lq94@Xry#=8R!k4|&%c=nlVdM}CYwgCdH<hlUm
z$?L(h_j{<n+y2HaX^W@b9-dB>%i}W>rE=;_#OWadtEF#y*5qD=_N4gu=c{htURee&
zOO;>g_HFggwL}N~38Fu`H*yPWV&5c`Ra76*k93qolm|m%W?W@fVwl3%s)e~p57xj%
zq~QbNR{%l`#TTp{9JgRzV_U`7MMJ^x*|XS1Ao{wZWk;j;+!F<>-*X?ga6kH6qw;<4
zdKvn>GxN*7`l8qKooBo}<ldQ=SkOD;yn_wuouJo?z=nu@9F&;<lPyae5xKUQ`M_V)
ztXJQ~*5Ytg>yGp<I~#?~ES-JciQ`y-vV?7<lCe09-?D$X?m=$d7>mbY_QHyo6H7~#
z2=<yHc~Re)@QQIrd`*16rm5Bxu^e}>D8k$`W5}$4#*v}WP8qSQ5~bKvtZ;+R^`2)T
zho@!W?d<~js#=@}=<5R=KCq>VyU7U8?xB6Xz*yji1+yqOYU5n43z#(R4L2cpWfdF7
zQhXY0@}+($fYT>bflLUCb0%8@J#d3x7aKrPN9ZC_E+UVPCxWrRKv_6nO@Llq><;!G
z?6yURsnAyJK2e(3o+z6ZX+A8dP_(U<_Oq1<23yJu&MmCdG_k%QNup6K7|gh(EGXwF
z23mtkL#Q0z2=QAFNp;AQ$c1piQ&4-}*%f4=YFW-V&($LTl@R=|k=i+@oMlxPbhek^
z+LUqv+g*G`<2c;7KF8M~W@E<(0*eO#f#DYSopfJpN!Rr7i20mD?=Q8_>tXqPYVfjO
zBc3mJaJ12AADTDoyt9@$DB7q!iWIm$KWusjTD^ijGU=b791Us>c-S~43zDfyG<2?I
zvhrHuPa!z5IE5;~LBUSBQejfWa#$M)a<H%lg?|kltk-*fwsl(}94{T4nv&hsjxH>e
zVufNdGcjJ1guRH4`eDnmLaAiFl1>FH;e4(VHs<B=@I3A91O2HmjxAZnZ;72#q<jIS
zC?8MziCDQ9$uBLPlvK3X!J(Lp6(*4v0LeAyr>0W3#p7|tOe0t6BW3vL<F91k*+658
zTo_BIoe*Zy7tgozkihPGeLE^?sx31@*%YAj<Q)yE_kn3R0P*;<`FtGZCLN3oJAd+_
zH!4uJx1Hl?FHlvdl5@zW{l1{-xuJ(uCtk{W7v@<Et0C_Qyc!ZK6=hlYr+J&?v?eLL
zh)&f1RMMtIPAIT|q7O~)NE=BMad-i(ymjU!6TyKFKv7M8;OzQqcdb8r?Q37Ve%9JW
zmUrq8QQdq#O6?*WJybE(b>vTo$B{*0y}w&I6gAy=9R9paB3`Yw+NKTtqzd~)TO!Vr
ziClIE8x$oa^t^3m3x(x$dbv=THIE?I8a>vR;eF0Je8r(Hhp^hs<!i$>V<0@@k`SZK
z4q-Zh?jnUZjPAqS2Xh662G2}9IP3R*692+XhZ_D#Yy=dXe)Q2t0e;za=}{8k(JtVp
z@4N553Co(e&q8$G-FM&Z_&UGuZiZPwp85aa=MR}LAinJd%&H@KtAKzU0*z-N=rLby
zW<dl|th)%R_ven!(GZRyBckp(4rlt1AfB)9?(UYceiVxk5u1cHqvhRH>R~|MnALnZ
zS~0y!G@OV3t+muzZ7p>aH}%*2A_gZhHfBW`z!9~^#uDvF<WIkdmC0r9F<VYo%@>gl
z0{^;-nEH!;QV#uds{_1ji9S5XeFB6n5(9~coOJ-cgK4^qML2<Ii#jaWO#m!$0glLE
zS&TY>f$YSihD_VEl1XM1HJQ`nCRm84;IoUi>`2&(C&P$(D$M+J+76|Dx{&%Ht=F1v
zwEwkavgc-Vz>?p5KKs6m>4xsNvst^;I6k$!P{>2L3ua9)WS>7=E{-3auO!cgu#ES{
zX!>+WOWq&qzh?>y&9UfV>hZNmWbJWun?4P6zJoYA?q#xxh@sAaHX#im`s|1~HR5Nk
zhLp`rXH~OF_GP3odBqw{EU<yUSj>eTeK2d+)W?BBDqe`h>%1nK3_+uzP$j^v4s@6N
z7dRJ=+Vhy@GXX~;Y8jk87KVef70Koh5f+Of9ds*?h}Fe{sphTaEY>aKKcrr^@|kEN
zTd*V9d?*)-WZ{<wOw&jkov2@Asgv+|45p7?<Z`oZttN~~?C{-UF&l~Hc1~f*&XHTN
zsM^Zst?)c#&O{=MufW+GHEfU-iqJujOq=33%Y^b^>h^%IPeZD_J<tw39QdKY2LNaB
z4v~m49f8CTaVCN95#bQ}ol~Mu4ya1JvVyQkpM0ilK%dz|2~}h=LGA;S5en!kQivac
z)aZzMYD*gtpxkMFVY*eAE=+tuLO3u@D3>}1v&-zuOea5xa@_S#!?w;U{dL~c=R>p3
zbVy65Z@zf~AEI#7?qkR$UeZ@{b+9{M0Cltutoq=OD*VN7*8u!4LAQP__SEH`9TCwj
zln(nN*f|)Sx=tKW<ykY1>@CrNsfd$6oqQw~drLI;l@LBc)*}3c3`*BH;ZA=y>r18q
zlQ~I<{P+@QTsh$WP|6UnA=W1~Z^;rWTuKIH)8(avl4<q6Pa?X*{r--7k}<x!zOtT9
z|C^8#Zxd=lc^N3(!0v@EYZj{nX#?*rs_;id9kOI7>vJp^V6oLIOo#I8h*c%*qnD&!
zzEqf*nV+ATDJ+S}BAJXjszB9r=CN}}VP1dxJ>Z-oW3R>+qo4`rG!^|hX!T;OsP(i&
z8}t!;f&TTM#afaDM0dZMIz4uVr%tLZ2cDqS_J~^>oP7Mh!V>s%;)=zmMdIn$W==MB
zNmEm+QbP`)g7}M}TYnjk#mC0S#*pkvkMBrZ!9t;2&ZL8u9A_uz>vOeoxi(jypA4r{
z#a!-krE(c`B$W=ER~CX+I#Vte3RX~#+hf@A1pIOw$9NJ)^Xl5<6xKcGYI4kFYp)p}
z<1((;xO@DqTN{fRX2!}aHg3fUY7Uhnk#Y!M!_7A(D`Vs1(68O&%VJZLwHjhh<e1BX
z7&0d4m!QY~D)#?a3LGbfaMWG+5Hci#f+5uMn+#0lilIgU;*exz`I}O?S3pomqR=Zl
zXT@IW-^sr!{Nr8o*=Y1aG@C8`_fj^Cv*-n;$JdJ+&dR_3`>>k-^QS&9^x<<>3v1RB
z@IBrdxD%F}`<egQfSpknA)9CR=Q%b_0?R1~$kKE{Tx!dt+pCap<V?zi(P}1~DUmC>
z)l?wY(r;voeMj!HcTx^|xT(MJg)hA0ju!@lKV@0%Im^`A<;olXS-H}#8f&f%m1eG8
zEb>9%PW<~@w{+poJHHVO^34Zszy0<*!{&X{)6>-|7}PQno}2m1;`i3GhvV_%$K&zC
znfdQs%*>l@{6wR8^d_v$ySsYN(#*`v?Cgv&zXOS|+MfR%SX6NRQ7kgOuX=CTl5zVY
z#*KTQ+!J_IW_B4PH;{sK)BqBW9vIkQj)3}+>u{7EU&26pCIB%vNO4us!@C&s1qi~f
zX9sVI?11_trrdA30G1$fzxnT{PoA7^Or}$bU?p0<D-$%S>BV?17te8i)2Nw$28K*~
zJWV@7+ki(SxVi;iEvu-)W1t$z{*M7@DHF`xRgPAIVB(W&2+SIAK3W|zC(Ms61E^=Q
zO;kK{xfmP^Wizqt4NOiw1#%L>{^-f6-zN=Ps!Z~YYBmx{AVRwp58Ih4s*52d5Lz+|
z?jFr%!j(|zt_*fm&fFPw(pZD`u#r<0tR;a@4E8X<yo<c#w*l(k4X6WOY64=)3Ag~b
zd}d&oM#Dy-8W4{B-!&3JE5bXK{Q<Zm_OW<3))AS-iD>zg<>)Kw^?H1KZ1$70W8?8U
zw)b$LBde^)=&{YmWj}{@QvxYDHP7e2B61~d-GBf6SQm?Bu@qv9JC66#^<u8>Ao-qs
zufC<fgxuiTflE(t@pjDC*9X2IaM4u>B}9tqj_rbfpt<P0!~qe%){%{M_*%R<in$HG
zDcAM7CZ4OHI1scsLbeL;m#xS)rYjXcXk`7L1DblgQlXi-Gx^C+0-6*3uL7(?5y`T0
zWYA5?301Xnxv6VF>Heu)&f*6;rbys+{ea^A!L`^{dk_G6qJsi)xYxXd(hnUKJPIs|
zK_>y6y~jm3k-qfA6Hi2}()cCYD%GAJM#z?a>LVp9{QO$UvM-I7tjH7LXf$j-`{56N
z7(pl2CH~wK!H?6ls#S_b?!k|Ji9a8X=-M9m1lZ0J?66o-&TR=O5p%Mu;zf39Z!}di
z)7!kY(PY{^9EKiYud3F$SG?j}3r0abHmeo)v{JqG)vvy_|Gj$j?wf6YxThPY_wGdg
z0r&iYd}7R)v3Kerk$<WC0RGef*c|%sH2P2jk2??D`BlKu9}${Kslba)qhid?@yMR!
zoCq~VK#`jeDaM|%^NuJf@vz6&REShTb4b)AAL2`EfCL@7d`J@Oo;j6yQE;-o;i_1j
zC1UI<gh_XI9SC<K|52&VX4`n;3XYx}n$2!u>i{XHH|3p<Xb(Wb9m@CegG|Sr&Mq_F
zvU(}Ix|K0oTz#=Bf@n)>&vY^n*_p$?z0GP-mOt`|R!ewMZxD;e9cm&zS>CnVx&&_!
zhX+o~>0I`6Et~QfX@#(E@a2I{I2?obID!>tAsa=oGjh2YT8%)csnvL6HB>A|fw1VK
zFHbT_mcX+A49qI`uoL&(T?|FbU2L8GidWz|+bu^!#XWtw^*Z}Pslw&FJ`O3ewD8b_
z`f=Yp{$X3~__1S&Tp{B>Ke!M7>m>ehH}3Q23z<aDe=a=A8;i5o9Tj;E4EL`Z2HqYp
zB5%HM;X+k2VON)RanA*H5iBns+?A)1IJ?T%BxmL()<8Z8rvl#Fbot5vubPkruM@NO
za|3s!2AC!^7|J=&Z$w%;s6V{{t;UpjG8(>&xS2R&9&uP@h<Iu>d<`^nw^vu1U$r|?
zBod59x1!O~tsqA(P3nLqF;4b4DC_eBuM~;mdgczutxG>-+3EooPBB-~bL2A+AhH^9
zkR33XT~Dzbbr!M5(qs0REWjjHu__gzHu7`uU<<uq84waBV}p~6j3vGtjSA7j)Qv_j
zU)CbB_Xp{>3#)U1%ZNz4K1=}raSSX}5J7*P#0MZOh@=rg4D^Pyp0s>Tq-Ti7j)JVE
zefHgeOF|h2Y=PNB97x3?i87-1vBEtPOMNH25cYhH{j1H}@wH@vr%Z_q1c)CDn!iBC
zWbW>kizHLARH-~MQ7+*m8JP{oJMnPPt_wx4+rfG)o*y62Ls?8B_CA(Os9fs7=j?&&
zp7&8;clpP@fFY1GK!meI`Pp3K07T&|X)BWR@V6`Yht&+Q#f5?g_trHdxo^H6+D>c6
zGKmKh1~xj%*<OD(!LP4xw1W_%T<R-x9De>Lx+1Yu&rXZK3#!*F*CwvQQ^d80lscSK
zeQ=E){W)LeOL!A6<#Ka)**Vn*fdIrQXPsUG>(fsoXVXhz!w6tO;$ovIu7nEK1i<nN
z=N)ynMcBDHS7Gf2mjtJ5ZZsMg86q*t;~9HQ@edjJP=fjg?Y0%s&4}$TCO6C*R?nSV
zJv}x(J$Cxula_Tdl`WRf-+S-*axt6Ybvox=A3c5gbUJ;yZp<7{<t4uU2e`n^Iv$?z
zyXNY1D0w=SI(<-u?gxfyGl{`x>a2Otz;(ZCk+hO=2-Gh=gLR(+dk>^ESG*)$!=KY(
znY?(acPV}n41XR(zR(*M7maJYH;W55rXsrJ=Z+KDrF3rY^r?#%Po18dLw?Q^T$L3v
zUH%r1pE&O(Z$5wGctL|OBcbv0H{Gq}_1(K~IzJx5SERk4vn=y#jGHiJJ4l&Q<r=h;
z3Lj<`##mjeaM(lZHvc_qwSuX1`pp-z#exYv1#*%6Z6tflnqEF@KAV7@5KEdO$Dfyh
z*#448=u^=6qMulf6$)_^h=B^g);GYfQ!;0$wgV9K4v7gBO(2>mfB+smqa#ISbRr)A
z;NC;J9K6H?uYz%6NO?Xi57f}D^6PWqNMhE@XMJ`Ru{Y7E6-=Cau^$}k$Bx!NQMT-h
zoE100h4ekBA`uDaUXMfv(D`RKB(XLjYifKQHi760Fv%N-VzIt%l^=}T7dfMDdO21B
z6Bl+d54+;?b_m$OsIG!FU0bwMGU^eD?KDtFVz9)r;a)qT6jayM8RoNP3_Ww}3xL?U
zYH@aVZy@U4>yZWAeW5@rx3~Qg#?F<Ys~!K<6<*0as5*V8oz4mLR9k{B>u0CAv!8dx
zrgKy|4SKECt|O0V?G%b}qw1!iarM=HSBufFpl59+!C^6>VOKE>SD+0t!^T?!j|D!B
zC_s1Z1kj1G*Wjg8S~WHk4Pcxj)E@ed*Y-<i&dL>cq}UBjYv8=1CWC{j#%f?#XJxeA
z8i32V<@@M?HE>LuOpwals^8Gwe#5<P;<}?i(%+u{qI%FPZj}!3w9-p%rN|CQJHTD;
z^_TVf%T*ka2^QfV`Vs&6;Il(jGSBS(vUqQGP#$%*{1y4neRJtFxX^}J$cZc<3)s-s
z2IBBc2?TkHODePsh>638jHHpEos32{^YY0hIpTV0^5t-3JNMmdJm2--j6{x2c%>$e
zMI!F&a^wvuBJ@q?{`i414%=1>M@0XeYcTzo?v%`)@Y%`rhG>^MCCbp+ygijY8E0@t
zFBnzHvD|ljM{dlOPI<heFy&rO74Bdex$KJK*>fMMPqV|z^wDFPK`|5$1y>4QF3JML
zCif-Ma@yt9rdAEKBtIUhIWSrR7A0Xd$a7@_tOU#i_}+jZa9qNrby2cASQHgFPz?Zp
zjuO{+L5kq04d%Cz^asmVgSn%&eKZF@OfY*iVoc=d4>DB2np?Ph=8R>XIdhrEWH1)n
zdI$5-h=AT$JRPNs8#O-y+0V9DazQNg%&l;nSe&D-+<Gh)OuDc0xKInS;;p-w8%N(h
zz1WcYhWoybalQ|@iN2G7z%T&9OoR3!A{!_)p%=o{01J>E3PmUbjs|qKq5aUhBUGrN
zxa{UY)3o1YaKaf;4>F&92$|WzVXb^+Vxp3_B6jq)$<U3q6%0;RSDTS&G}2tHP6mUP
zePd|yLtf%0?+jTrR#TxuBzr?PQV6A1FJtL1HUqvj61f$d`-Ts#FJfol==nzBXgGYd
z&^V8aP~+sGC(S{@U$@3$^+a})E}qS7q7LB15V^E#o-sRtI&#L}9QY1+N4_`kZg{Xi
z0xC=#!MG}yFB4t@Em)vr+<6VCaM+0kqW2)_(a`mNUufkym*Bco-zP8#h@Tt!%3`Hb
zNf%~syx}D;x#7myLb{S3yQ5@ozulDX7)wV+>6}BJQ`onvL^}K&bUzEzm{SC4G_HEP
zIs87>A5K>0t2N82Rp%>{OQi^GNX^N~W;r?;DbW_>RNnSZx7~XOd0MkxSzqIPcQJO%
zceV@O*$w2tJ<lV<{>wZ*UIM44Xt=RfVXT4ALm+N9z>q;4#8}sLsX`96Gjq}<srVxG
z&_O22?>c}2ct9Hrp@siDIG-$qqv47B{?Cbs9WG5eH}AU-CsMN7YKa`6?~(qwQ2=Mw
zSL#bQFg)porTWVH8PtWBu+g!eAS{c{KJ-&)m+0%+Kn*b#o;N_lDi-T&u6w|};mbU&
z@nn9(HxYV$AWziV_#QjcbB?a9#n}fMcezWnHVW-0&vxC+a_r&;o>sXV^vzoAX6_bU
z)%&Wa={Oz2PAeBN$6kfKR@@8_y6bNa^JlP*S%!_W2cs7dsZas<m9nVqH)1yj78jCW
z{{{FfEu@<G3QxR?Ho1ky`3Y0mp-o%cNXpxCc4XZyTflCytqlF(CX&tqnV*qg-+9|*
zG)`I|Br<TbPB`gnR>!uswrFCOAED5!NhZYT6AZhI`<EqdYYwqX(AqI>puiGIAW|Xa
zK|du1$)=A0#Ws+EhE$zY9)F<Vyj#I7TslQgUbyhW7hbq<5(|*wBERuPFS_^M7rp34
zddm?`7r*O9=nwGL#o;&tBUoLGBfuFJl*Em9wfw;fe}7_XdV1;vyyB@;>`bj{Otp3<
z25uCOty>$2aFz#}cTaAknhmS#k6U10<K)kR<Kv*i+-GV#@Xo;dL4`jb_<i7tGFy#i
zaV_BNdyTY_s3OLvffrEb{r=pTFo*9IZJhuipROXhPIG@rES&?xA80cC==GHv@zwvI
zyf+V%?5ger<G#p<7yG{Co)H<jZxva&MAoWMEm8>~DUr|y0uoCqWDA=Vix37(urvr`
zn*wYP*cPN)#zrvaE8EzBZA^mOKzkWiyWP{qy*S<8fCfG>9vgSh5H-KwIq$}j6<I17
zzUe=#QoguvyYDXNp8cHLVu(}?Pk6lSE5Ub9*N^w(tsrlG=`B6-4%e;s)7XEnbzX0k
z@aVzyU7?8)7OT#8v;zL@bk^0tz1DfxS9T7(!}@x^(^Z?mek*@9bE}*M1ORf2#*?6{
z%d2EDz_Jh_hz>x39bgc2C5tr+%Da@^_MZ10S*u^rx!FBhb!y^$<`w_=kN;>6h2o)b
z&J<>|^}1|q-<huZMw3m3*Kf)a*?F1xv&q`4n-6~e^PdmgR0}1u8H2sN2jR5m6a4{3
zmUgnk=mFG8Dzde`+f<-qM`$Rc?^~7>iJs$Fui1CFuB6?-xLHYe3tLq@mJ6E!>~D=d
zuJcYjk^RSPBJSkP$H6akj&K6mu+l%&{N*gApscxr<J#6~3_{+#njikO2HUMe-PZ;5
zP`wCV0ngmqAOFZesWgzr<}K4JD>IXmGb=08P&cG_wOFhl#G8E)3=EVK`ux-s-!B&V
z4%#fR;KMjp=ySz9`xr6eHsRYXDimK95s}AT!{|kse&8++CjB;W4g{h4BAl7K&YwRo
z(G>t~^3MpNI6pEnHZwEY{LA7{ER`+H%$55eTYO!gI}yN-c@i)1DLe}xmsie@&diLB
zj5PnezdScn$fjaL#n&x97MO7R@FU}0g4}V+d=5Mca=$?;By&udog#R;MHf6KaWr58
z$2NS3hN-r?wz_8Sm>>9g#`<6oli!#$ugbpq)vsQnGYf)J;sISy7Z@7wJa^)MCevv+
z*P7_Bj2wDs5Si%wW2Gf1(V|gH>fqlGR~zVkgnt5bVqA%HZ}$CnlUFFUp`U6er;VJW
zy@GM?eFqyDb@CVX1=nqRVQ@K8ShC8o)s^a66`?%*kDdvnYWD0fmjM2ePeFz#T-LkQ
z<%90nH*+ghk*fKw_&QKK=#fE>A#C9ZKte#Ht=KZoNayePwDw4yP}lD2@9z)7D;_TK
z!G0(u`x^T%Jk)i&|95mDk~4^>^65|j9_Xeo6hxo~@)f0W`DCaMmjbTa`$`~`&ZlG9
z3>iw2VGniIO+B8f<e57OdAf2io_W84U|~Fk2Nh44-<+G9i$whz1mKTG=7PZ<SN1>A
zP*{@u!#9SoTV$|LvMvSD<{)>E45H1!?lUwW9phbycj)bYcbDpa3iwlfet%!ezyCLI
zr+)8F4=*W?I3|9TZOE}S-|D{@1(BBMHHe~|USd!2UW4LFAr%YD14wV!m{rJJ(62SO
zVRmor=@q&@l88t0p-9x7Mn;V2ZOjyCac%QrCyQt{n|ru<CKkRWV#4{52_1?>Ul@)(
zAN7J)<2pZywUmG_<sqzeIKz2zV-f2b0vm<r!w?i9b9exUJj>!UhGwr|GuEUR!k7Gu
z=F`s6BQHogPefwJZ#%udj*S4EL@ItF8u`#SFa5<|{6*5ym_IlUyy*H<iPSfrh(+hY
zd6m}JPp6#ZT^r%>hZNV_F(n0khA|DSbH6lMAEHSY<w!fzDVj}{zyTx5&BTcizyAo2
z+j?@ZoPbVwBfoDr?-mRi{(+<3=3>~C48;cdUNpXLxDVxmVEh7bgR0rv>em%;vtZ_g
zaH>{KQKtHNSv7@nz;vwxRW%K}MaFKcAkhcA)*c~+W}_WQ)-FMDh21FZd;D$Kx9sl$
zv1OpSt^QOo4?$5J5yNSViF+#QyAGZbqN)<yY0<$#fbDJ_4KBO8?RK?GSMH11yFmew
zn;Vdu&m!vOb-o7l9X@2WXM30rjf;rvh@QJRBWAQ1S*a-%LiP&VPn#Kfk)?T1bx4q|
zTb2>%%b7@l9pl4-)>o=Zpo=wl`Qfv4D+k<}4a!M_<3BYNib-*yFqBN$VLx1fqADJ`
zGr#C`&PFc#ec^27eGtyE5k<HZ2#)3GikBOc-%Rs2ufP8KAb2@@+3)m29hDr1IpZc%
zdil#=zI@#5(8;1c72(g=T3@UBu~Iem?wQGxwh8<V{(8zZpFVc%ST-EX9dCaLUmVWW
z2k1Ji-(E(%gsH+K@WI9%!hZz?Pn3Wso{y|6ul4<??+M>WG5QMO^yg~}Gb++_3pTXK
zbSg8b+h6lCa#84627kjp4l%VBs;H;gg6xtAOCH+f9OvD(Gm5fP2)nrQl3!Pu!mBBr
zKYqef2LK)7{}K+n-Y?Hpk8$Go`5l}{gxy6r>gwg~9sTtl;;VQ_kHCOCl!kfOCCW~f
zaIcP=e93#Xl~M<{SU+!<QpD3c>iy1(4g78KJs_gC#L-zcb$*F|DrfHj-!MM*sOXRs
zS@+OP5RCx5J2m;VJSzXdF$d#8Jb$&ve4>ZgZ7B^z7W`5nk;)|F!wXhdzc3t6W>SfQ
zbw|6v*oLb<wAHGyRkx{cqm%Je2B8JkY%rEdA+e)mDPi3jnQ@heJUW6I*@15~fY|dz
z?aYKhT_hEhgf?Xw1*d5?8bq3SJQ(yj#3aIldwiJIvf@oHrmVqU=uB}5?wAr6oOc0Y
zGN%@g;kYAuwPr)lK+|z|$GgRMu;l<Mux`4V@NQ4<i;BzF2R!(jLbDgJ&)f~n#)_z4
z_C;o1`14?;OoA;}sGq`moAwKSM7#$kTyY#OA}-7$>wp48$k7Q>*-#>RE+C4kWPyoV
zeNJI%JY{|sJ8<{Sy4mbNb1x9cs@E1;n#~1PVJ;{@`O^`yYc@MA`G4wU78r)Lz;G{H
zGRQ3GjAG8SALvgcN8m~XYzEoOBweS_JlFYt*XDS}x340p5@d_IM;)WnupLmps&wIS
zQ3X}2UYi0Bqp_+g-w;os<oSw#HejfV7y{Do3I|Tt5QY!&AzVNhJ`7wxkw0uDtASh=
zx;maU_p(U%3FBbaokd>H+{d8LmAv4RLA<z|tSr61vtMSjYs_{z0&eqLkVsiFnUQ3o
z{{va9fb$}7(p8=n7SE45O<(7$T^rM5D8#<E3w<-?C1lI;%tLJ<6#b=P$P6PDc<j$&
z_}}|ApXfM|i(W2`Q>cp?QRIVb<&IVSZ&#Eks0oSGI|(x*OfH6sJNIH70k3Ni%)H@+
z12~3Ma1eeKB!e>4T##IRazt6}8tcMX5sr7UqPUa<A1q}0{P70|2S!KIxm<d%JUH-R
z+~1d3kl<j_%Utf|mwoT6%*|Eu*}||!7DeMzxy8fFV_|I0I<|axAwLDFlPOc;!-Z_V
zGB;nzWR15#ORTApgoA0~z9cqI%9`YT5O3XNgE*=5Tkzs^-H8{z@DB!z(+@=G!kr<`
z&qR_4E=AljvJLwU0+CrgKjUT+F*wi-#A4RpiE{08eLsMBMC_A;yrJW={5S-(ymHpK
z5b2e{E5uwX>j77ZiH}Q`^Qk$Mboi>v<9M2hQE@J_p+<u>;f<9tyI_IHKNm46V2WT*
z+zSVdJ5WreBNO?&|D*{YhqN?ZhJid3H*ce|DO-lME_OU@PWtotiAXwC9DoWVyn*nJ
zp$PUib|Qmu$a3>*Yx597?}2%-RINAQ5midn%|#OkVe6g2ii^mkLLwMMQi_=0c`0(8
zEueCIWC2;vUJ3<*DGUaKi2`Dt+{)mPap1gu$qWPn0|vF?#RR^w1tzMoTcR$qQ7Q?J
z+!G!2^AK719^Wqj@)ZLxnZ9z0<Q&uin|ZZX)OVoPi$fH(mI!U+J}9*9;~EclagD1r
zwLG_(Mu0Tx_duBd&J|GU!NnNCBu7)<x0oGz5MaGUZwvnoeq?${?&lqACzUE71Hz$1
zta#lgk$shX<Z#=BOrn%*ew_gwmdY}W<8)+vylNT4>iBr%v}eW93YDiEy?v26-AnZO
zNGFr&nR2->vX7rU&j;|5J2RLw&9A~AnVjVsNt*p$fBJevrPwQoMp+zz$=He{BO{BH
zT=ly+5?_hk2>6Pziic&#QjA;+)uI4(Ht3PEh(t7`RH`ncWX)l&xXRve*$6)IhMdFB
z`%)pUd6>qPE4YVke7md!!keIdPBbtyH57<CXd=5(zb|isg~O#(3<={y;b?LE?1K-U
z9WO?a(%HmgsnX#>&?zkT%l6cxC$_dujw1haUtj-Xcj%)Kf2tld+#g>OfB&xq3wbA;
z3Pt0&R5a=)!tfBth7&F#n{)AKC>3_{g<v>_-Jk>E5VfVDFczB=3yYqVvhxRgW#+5G
znnquBd^l26DC&p>J><;Y6V#T@D_j_QhhR<?onx|&aS|tKDG9^m{`Dt5@d-C@_e3(f
z>?RYTIVUy#P{92Jsf5YB?QL%hIQ?!S>CUdW$yBi91Rn15hfTS>S}B)+>;_8Z%4)f6
zLiwkjdMe;LrNJMHoA{mnQR4<b5{>>%-yy1P?s#RwB%T{E{e!{vsp&`{65*>66sD)v
zg9H756~|}yz8vIE<w}hpPk<WjlSzw0U1atxRFJ2!;)4zTP&QB(JlQ|+Kby(q;Na-!
z;9wH}+0upUW@cmqZ*IkHW@fJAGT>@ZXtrl2weLNt;g@68L#Oy6^BtW!RE<@--_o;+
zw-)w<hn+%xmvg{kZ3}u3yo-xTWp@a`oT@@mGgg5<@;|=_MZJ{@<G~bchIGLzuZ_04
z!WL~Ttd^s}AKL5Hbn_lM|7|mZuDrS+8<F!B8HLfcax779UTlw2yVNmPp99bOQp}YS
zam9rMG4i3DHTLJqz!-91!ZpYOr!gFT#u54FsbmobeqTq=F|XzDp+mz=v516Hp;)4V
z{W`FLsc>i|GFDpRCZ?cZNu&&LBB6Gx+a>d9-lE$pD`JJoX#dpI5;rv+WJM`5eZs2^
za%<D2si}TEf(CrK*Dzu@fakWnc%)N^fPdh?xqYUw=rQlA!S-mqa0#seGPk$peR=@F
zQd3jMgy!_K6Iwb(|5#xC6Z3^>z(G%KjQN!o983>PR!=O<>oBxOb}%x(a6+>5OctgA
z7Z=+apLU7k(6;S7j2sNUEY>biksute)P`#d{m><JK>st3kXTUwFP=RGtA(g-MRkik
z$7!4H_G3BO?9WZkuCLFY9drGG2#nw0L&na+j)9?T&$^a4bY?WEZ{@~ZCbmv)J>4sL
zli6iGIoS))O`^I{d_bRv*F3&3yUtO~a9+IRDJW$)s_U~!eGOmGCSCTm$FyA{i~&D7
z+u$h&;lJ=Ru*DyY5o(sQ0{bcQq^d>KHma^srjhCz`fIdNPJwZMx{i5~*ho|ff|)os
z*H+zbuZ42)VBFe?f%^-Ik!eXY1v}C7NTRTK?COYlrnJ*u-0jl#!PvMHXa7WT6j@vF
zDfwh!q`xc(^x&66Mdk06C!G}ia6@>~AV{l=!L&q)tC^fU$TUhE=8A{RHM+GYar4yl
z``!}?fn<Qvf{UpPy46Y*IF{cU#}}bm-Z&VO%L2crH##@fHpcf5f#a5EoHAHD8JJKg
zJ2E<s^hO8L7ePbSo<!zin+kcz3LD6~<=Om1%9K`Q%*P6lm@Kc@@ZkIL|6kx6jL|f`
zNtM!CjD#`mz(`Zm7rzT=-W(~8Em~-fM%;9vxUrA}O>Uv?a$ZUHt%tuGF#dF^2wvO+
zW2ZD6G^2Tt@D^q-=R69J)OQ?f_??R$;fRl!G$~Oj^nz{lN(aC(My3};8}R|}3DM>;
z0nEYxl^la3`#Fsfxjs;^t-Ystp}mOPC2&rJzrp}*^aCUN*_>g$Ua%uuz3l5XYv{Ys
zD98rC2ahJ~QmxrBu!@Kvsq@fMPOP|GT^2d6!sA1=N6Nbh3a!<yLU(>D8m$wl&nP9e
zv{(o!2##KzA#tid+>@;QRwPm<9Z`Y~M<$O93q=P(L4353dQhKw^TTao(Jo1UOTLuz
zTC)whf!OJ1rW&@x;KU%0k+@P$(9$~44@r6u#DRlBL?oq4y+-@vWS=Z^+}bm~#EINY
zBsK}2!nuw2<je1o{pc-63RZ7}eA$Mexy^KXGZF}<HuM@{x>J5Ytm(6rA6kmzo^828
zMo_-~8fAjFszC8pw$Ya|+cFneh~STe;q75@OCL<!dlJ{`%nbYi`e$aU?GnPdTS88|
z)ZFVY&dx0@&COD=YfGIzlFU^$56j%GvX_r#B?u1c_ofU(8IFF+9~228EP7u0>hJjs
z&(^J0p<`pghU!c>%~b+6|C*n=`hy$xLCwF{s#Xf;>`mc@S>cYz{I<#i9`f`SN{g1o
z;&3o}!>07Kx2U?5hB3`#OY3`E(6MXCq8E<VRy*f!yR_vTUb*?cn^z7y(3&;h0PUIi
zr#c&6XII)KQ*WpBZI_xK<eRKQ)ovfH*Y~|AHF*7dhtS$(iNb!?{WC-bR=-Z9pfnSR
zK{)fN+O>A?KVlQkLnIA{4iUAt%h%RLbH|l$YAqhFS3Yf3vT|s4HaJ)Vtts<MsP1Bn
z%k48*jX==wK-;WBWxyWR{vM5FiUan<Cqq~Gh0cwn#W-}?w}V?WeeLNZ)_N!|30s{z
zM5Dc;F9W69Mf~{9h>EtBdi6j7+C^ii(D0H1f_Vpz35rZkGc9a<+4ClaCK-&s_QgxB
z+|>~c@WJ-%O98(Vnu&*NjaoQ9gQ&d#HpT=;j26Qm6kRkvZMayhx^kPNq4Lz^;9w9Q
zUO^MC)xz-3!O7s@q$Vxi(wFVeZl`&p)BlV8eGq#iA@g#Vr|q-pvq~r0XBBYP82iv3
z2=AgytN+TNHSBQ#bljD^@7jlHQ#;C^OK`PrpJ50uDWvGpngjKTYZ=3aK5EC1lVZ!b
zW1Z#oAx~+gpCUZ@N>1Hu`I5dh$@M~LS1)%G_A(-sRR!CL-L2Ez6t?mrGf<{iXdcG3
zHFJsf>SlvS`hL>rEuKwM@_JOj{-_PJtyv7smO*9_+S7Tc^Wh!)Rtl}G<0Y}3Z#5bM
z|NpDno3d#s{};5E<{aU$u6=wj!8)@0Y?}JLf`xxZTesmB@vOjrJbDKl5F?PMAV)!n
zggdp6T=kCV_E#atu6xlz*(*@05WCqMOa&A>08~Ix96vf?Q=k@268r0a1r+Z<C-2K5
zs;8}or-)BfOq9GTNor|h&qdtZk9vEh1vehWHUZ6VGD(wpXY<F+tC~NK3om)K-SO;|
zeEjc`^a&ARmoI-7UX7oF^$@)mE|JQ9u3jK9Si96@qCTy(B5HsI=y9z;7F`-c`ytow
z8X9GN7wH5*YQZOI0iPl>?@poNko#Pp0oM^4WDaf*otRW=klyIV2}cqkDEe1JiAdPF
z=I5T2r{N_qu%$R@Yao;Hv$A*?{+)rTDY$RqqkPQY{S<cC<YTnMv>wA8^f6Ause?vN
z!)M|SL}}2r%04NCaFldIXqDcWS)QW+3OQLWm)m<qlVEyj#(Gc=Ix4YJ8N3!L^(aUQ
zsV!hTn#`EnGm~T{ol$$G`Eh&2Jk|WTxxM*uD4|2<b{?D9h(%1?-Ve;1CB980=v^{5
z+1FbYeslj7lPRLEZ%VzHP||yQODowD4!46wlrUy$ha*4B9IX#`j2OXHM~o{>M~v{K
zBequ4mB*(<e}$Fk-2qJTjn*9uCWop=ZLPujll>iOUAUWl`4c(p+K1}im8EdziZS_<
zD~IWo_B}gpuf!m|l2zi+Y}l)_v)-j*y5C-Fy-K`;Sj=6l?-A(rx;YU~00yJ)QAGo+
z&EcV&3I4Stk3D)8$X39JRckCR!rQl0FR6H<@mbq+G1%kG<B8<yMcnSxotNMPCVQcC
zKq+N&T2!dw0COrJdDB}_PdFo;51A6OeAnCU5fH268ViKmR>8Md?>ev7#f;gyF9pC4
z^R(_dZwCAvLzLsazAfKtfDAp}N}U0@u5=m6c`HbX8-%iX)H^H7O%P$O11R!~Q{q5N
zk%4MB5hKvoN%$>bUr%@cb#~M+>6e1~E5nmJTC!0~JCos8dTXgTJ6qhiaN)wabLXm+
z@vuMg667lly(Hofk5{VAH{Ep8P47APBOO;ONbiE@Y{rC{bo0xKu=HrA(=%FPj@;(n
z8{V*mjOXL&<dX>#e=?aKFBHswegppd@vcjJ2IEJ#0xV$sPIN?15{GkIsnO*Gx9~a~
zASK3G_%8`6!Iax;6~7k>x{gsJQ7CDT&C;1OOP*I`$>Q&>izV>uXU-s=e^+}FJ$2~J
znM0?TXHu)cvHM!{<|zJ~89sAnWE$P@<P+`3&9FjPPrQF-Ph7~zVy%VpZE6iZS=yn>
zHTP(5H?@;W1>7@d+`!JYcT0LT(%(PQ{DTnRZS4{NJ)0xjvmnSTNbB^o=+U#Cqt+A+
z&veYsNx<2Cm>-pL;8F_`#)V0affhM(682i9Y!@ea1JQ*8m-Cv2y^5Sqq?5*~CGry!
zoGTl8_l7i^4ypxyTiPpOS``Mpm3sxqT>+Y<76e9{4G$KX9LLa)8LmgWzIJ{>B4ewP
z)HTJ`1&I(VO?e>Yesq^Tq7zyG=ttE)UJ!p*c;|&7+bkYNd)QQNC{MhejWgV>^9XW*
z>(!AJGMYo7?C8|7>A5XkP~V>{jm>VtWe4~^mPPVf?~iNA2u|2T`{B`Q+K~~q{cVrc
z9JV14UoUWdoH<Jb>=XOn@Xl?zQ#u*zs2-2Qrz0AB<n)WO)XM6`_HCA|dz=V*7qu6N
z1@Rw6EG}Mt_&TwUUD6w%Y$oYFbdlvt+AktJ^H%8BeVX*e<5tyrtyxD__R%gu{Q#_G
zE*Fph^0RQspV9cCPsq@5oJ)0acAB3!yfhxW8C$?7Pyb;mkqiXfQ0h<q_Gt}}if0*)
z7|+T~!awPe>lP};8S4*XvzmN85P}m`G7&f>b0BSPVE;lJ4ekZyLwCY|O5D_Fio9|y
zKQL`SAxN+Jgy_HNNac|nhLzKQ>$rFl3UF*nOVIpHA^k?r5B<VOC6=n{SRw%z?%U_=
zuZ44G9QnNA1=O|scmYsL<f7wcN3aW05aNKyw%U3vt1*8VS*(kS5LUFEZ+iUyxBJAu
zG3Z}&QjxxS?&HH`TGzobDNt*{O*FJv9*^JdM?4=BX-)c}ADW;CWCD>~Q-ORoc*J}p
z?go)DE*k$r3Ub1`ZAz{`#$IrJYb`MH4-L(mEF32TtgI1s;vZ!ltm{`1Ig$AfTqE3n
zeHROC7qCQq*aIz$rb-<$4Gd|*5hda@UH`D*z&BQy4)hXZgb`R5h)QDwm6-w?<#D&l
z+kzKq-Dp@if>A2p4NED)&)mes;5h_D#uKSzbb}u50DdohDvO);ig~{R@K45^5ikl^
z{{jywb=m97=f<~K1*^K*oNoRTUxr%(4k=sk@>A$lj7h*Z>WxVX+&uY-cVJXhbw%k=
zHJcjD<0$|Yd_|er)_N&y%dJT`mr2l#9VXIc(VJP`u8nl)or9sqk!Xg-G8teH41(Dn
zfVu8107Wl2@`4BC*!KOu+;PVpH{2M&rXJW_ATV^pppMW6d%eN%m_#O*3L>3+CP6#b
z$y0}-lem@Y7y5pQsAq?jYK{EqDDuWHf95lvd6ISW8~+tjv`(f>34&s<_jxuJ3x~xF
zw-lXP8yd)@As~TAgU+F09)Rs*6R3e}qa9&@P3^seG)hg#jfuvU1U_JMWwq~h%iQGb
z%nOq6Skj@mE0Rh@?$R>6m6DG>+WbW}U9Wik*+tjP9qC}GU{!oOufk?MJKAyMD};jS
z4}Op#;Cp&$PhZE8Ye!{1AJ)zwcAmKrfU_=IlU~_#V8c-RMX;4_v2EL1voSmBM3?U!
zbG_vh?x8lt86S7E`RL>c+AFHCKWN+Pb#qWk9HfY)f@Pt2b3W>3Z$C$C$JVx&>Ac<{
zyM|;B9nrg#xl2w(JJcgc#%?W0@Kx<Mfxqtfnm!SD9G?}1no?>|nb`^wY<MHwXzpp0
zX6xaWV1(<9)@b8#JKln;7NKe7UQWWNbB&1Zl!zK~P%9i5YqRr=I#yNltKlr|5}maT
z(iFTY_fe6dqUcstsD!c~+8D1*Nb^>g4G}?6beogCf}@sF&WbIW9q1nfA4yFm?Y1>g
z>@=UYR0z3>V0i6(RVuOa0P*vZVCXg;ViP&$il-I0bQp*x>|Iwn%^^~qT^caTL8ABI
z3a7#AkhBkdLVv30X~Wy?5vR=^X{2=NA|VG1A)c_E*w?+?&OUvL9pc~qesJzAKD8}<
z*8LF8n(#8<V?bjOT;lmnfF9MJP2ZNg&0=j6*){j{kzVFC*3-fh0=JyQ=Y7xnj($Iz
z-rmICzC;Xyh>C%`+cnOn^zkKPLPS6fl!BBG_8jP47z^Fny@31yIR$H-7F`Z-c>7nd
zTph4SXwz+bun$%22|1uKbx{#KUEQ>+qw)R(Ao%&$$?<;^WFWqJh^Hp*ooAMdt9T<|
zIapU4E)9fkf%+<aCHf+PPGz&><5N@Pn#0B`T_C}fSUqwd(CU;BltGgUn_itRmvdR{
z(^#Hvm+C?SVY1_!#A(g#C!c)M?|%|+;61@%4tR?u+FgPDpYlC@%0ax8urWLZni?`x
z;IFLs5m4EM$c#jT`f0*D>7~ZAtWg@~Es9|zoO{SBrQbBKa|2k>)qwi}ujM}c?$keE
zFPqol!QHxda|geinm{!gt$KMU_HbS*=@s7IqqRY67uxZcPThRS7xERsE%^fQqUW~R
z=^B*x=n5c#5rM8jsEq7n4;-BaCu15ERZDM*BgL~xyr=n!_q^vl-pO#SRx@v|HQ!{+
zTbpk*Z#|3*3i0NzB;BLt4M=+1eB%ld@5Rk+NtZCFd+~5Q_iGbnI53<l&3V?_x+OJR
zh?;r*yX+Z=(=Ml2LkKq`NV@PMo@*|tZdzZ3Pxe)5J!5h`z-r^@m7Z67llqd*h8`Q8
zC-nn~e7&W$Yv|nUKUW<Y<VLQU)vNcwQ#@h>(GO20FVv=Yd$0KR!Og_5cPi+KAI07v
z_%VrW#n;huADSvJ#1<qPaL>INmCkpf7EAB>D|Z2Gq~a2ytHm8oA}6^)PDv07mZw_I
ziGJ$Vqp#(^@_tLS5Q^z(a}k8aXD`{f-Ak-T-C5}m*z%-l)zP-0d;wWku7{rwHAM`w
zab@+YbqVVi%T-n|ptPlZRQVjM8#@tF6RCU&b-MW9Y9i3Zu3O*X_OUOzOJ{x+%ze=e
zBN{EZs#orM5rdtc_m$_JuyT_~J490rd&NA#?O}gSq3u#0d6nb&OM3Qej8uB!iRRy9
z;qx6ibjnYi*Sq-QTj~1<l*`!^UPYz|lSQ<Pf3+!bhGfTbg47Dce_(1bIAZP7xEkU_
z4jx~dCJx7|s05Yy4q6cZ8oX{UHZbgr1oHWCzdz_Y{n!sTFyagk#<FfCADfw<pNYMM
z+k?f5dfaX2rEHh(a&!4eAR54)lXn%uf&5Lu;7$2JxNui6csLlvF1|T8kWL38CYwVR
zYCQs`M=$@j`D?7T7_h~i8pF}iY*(`RX^+9UtfSXNS5}oqU?3km4U{inB7LSRqT)hz
zueM0OgzO?ROa%Yp#m*Sn&@ei62ys;0K+3ITZ_%kt9KL}88Bxc$w~QX@pFs+CzZ;Av
zqLIGBnmHYgN8LyJkht?vv(@QIQc5HSUc`!6;SKYH*<0>V$htR|IQ-%lvpQD$vPtuw
z-uT8hqJDVg2jlmweK6{V;~xw51)YyUu7>T$0d+@g?J^{$=ZL-~DE<*$jJYuK<s$P;
zh|^Pr?47thQ5E5W!nn(JXBai9k!u%Ss^WS<3@uD=cI`rL)_)q-3OAFFMs5$s@|n;%
z#$7=^%;)D`2nO6Pa4@^olI1>1orWc$2~zWq!}O7lx!k%l6w6<L+TmfO-<jKjtOX3i
znw-Lgek=^44%Vf34r!S_?<*k6{wOje`XH9UAx8ex?#Gs6ZC~jpgR4Y+?Wd;K<uyPD
zpoq6fYz0;bRW2)HRW;Lp{p#x2m_&&?tE;z?BM<4^>9OP#m`Rps&kPJq7Mvf&$`8(j
z^UDLXb89o%p{)eeP5qIPH$M2_@#DrBz6ZE^Y;1M)dRBGoDl*mM)m;TIJwn1XNOH^j
zlLH9vExgtr3?{dRvNLOQvjfZd@C+a>dC!nzjvs&U!8eXX7@p?KUH<mv|LDBaJcTur
z7Z1q|!JRMkJp^mmkHB;3Cw)JUOd6l`?fU-6_ka5SO6bJ8j7kulEh?W^D6U~RupTe~
zST^{fu!9g%ppDkED?P7HcfMHbJi+_V^m6ad7khuGXH$DW>+4L#xJUHMzs}@}7c2_j
z?0od;o>#XzU+hZJ*m&lb>%G1XI-r-kzAhwUe?xmeYP#BVv1fHX``xQGd!4=5={%`-
z_Pp1t&*IA$uDy}!_qz+e>+`Vfz8bp3pYi>Q?+>x|rv&p6moL*&<o&ZOye?oIu`S~&
zV$!K?<$vpdY~}{G{1Guf6(BrX`zC-D)+A^LYOF<`VC^{9AV)8Qp0Xiurde)TKU92x
zJfeXxdP+J(rFTcpc`hhhH_PTHEO9W#AB?4vu~ss=bUI&Pax`W~bNrD+^X~&=V+#_o
zF@hALV)Ri=#B1qA3G_U>Tn#hzT_O^3UfPMRb~^D45>nMQ!Ib>`lHL<cRw#~(vW!^m
zGnNV>#~HTK92gka8kZ*~iUp|0(K6&%8~Mxm;bB95aB(b=%?$*Jw6#WF3?_5c8^zUr
ze&Gz-jmSp$5C3&%b9Oo#Dyg#Snj{N9l(O(FQKI$EW0`ml>JjWG+5mh!&-bM8iI@*-
zIe=hD$bfN;9291AO6HqFE44oYHJDf40RixyaOpSPE*tf4D&RR>w_)D^CqRY=7oy`A
zsfDvD-@qLb5`Cg37W7v62|J>9!a6azelAoSa{I95Bp8HPQPs^w6!9On^v+E^F9r%i
zgT;flW4=So5qrL9Zm^Zy6~C5BHj%yH%1YkLZ?d9~4UL6qxpEg{*q#!m5G0)N^wL>u
zpc&*&yW{bKHNp&^VJoofPUz01=6&ByB~f<i&Lzw9Q@+YWC-DF?elU-vY8X~I^q$7Q
z_tlzmLKtz*0Oz#E*~K|v07pcj>74V98Uu6<VqaHGrN{~Gmyos}sKb04nY(p`ct6cu
z$c1{v`#G{OHiqmIbOZ~TaCP~t0Qq8?K(9b6oDg#SYWx$f?O$O4Z1YZhNuk~QkuT|$
zhW>op3=hxqEBIRc%)-2^^c$<WWRNT0TWa8zf&;m1LW8<Cd#o8W0K61z)6hMA4&GKb
zf{*+KjI|2b?y7LkQ^Z;P1Kn=P5>h4-WttkRJf&@h!&{}`g(sNY18&=5-tyiTY^%(b
zHy2h-%o%v5GCv4s8?Nb^Q$i%d;Na5ncmi%$x!A3UJ_?*fDSl>N2cY?<0|NtvxMN0m
zb&1=T_66dlBG0ZP5FeigPMe>*UX-!-ao?yn5R<%trJrNz=UUeuf6)>Rex?LxoaxH=
zfZrdgEiH{shMe1S@o?rH;Xp9*a6T5!yyoU*o$4FG!n!$3OfDA5)DImx6b{Ys(IV@a
z9GID23SB>ki6dG?rixy_BWxt`T3;)_?-=|DuNTed6!3}hL^}ejU^K$}G^b#?s3Q%y
zcj0rd*NgP4n8sZLoQ9kSm>m6g7`E5Rj%>f*Ka$HufeDJm%0#t#Vr71=y#1-E<>mP$
zWI~vRn`8gL7<?ez7yKCVF5EHP-~6e={y?-?oT%XGzCiQOI+Fn3K2{n-28H1Yd^d{N
zX)To=y5~LPeg2X1>gwUq(eW|tLhWEju~h7aSU%tUHqyL|-W7@1=f8sf8($2%zNdi?
zCWTYAy`-?4hPX$~H?^cBuf)snyS)!z;?~wlY{;9fn@9T{=j2Jp>2JQ>Y$ZqJGHwqh
z{mmW3o%9C-?|N4t*bf2{1P!?i8q)BU5pD7ScExLOhhxDsEfl_vVWn#d)rivsNYL~m
zF%$<HlTni^41Q&jzyhbIW`E@j&rQRjT@-6AcMk+Jh**M6kGTS&(y)#oSQ?m6y=zWG
z6arL-Bk&BkKOPB1^Y1c!=lX}sLq()Q=^q)+ru+Pb=V!7b{pW`7OTxh*>`(K@^(WJr
z7zQNy_U3PhqeTg@x>>R;oOv1xhnHX#E?K_JJbc@2w?&ZhEDo0A)=)GSO*KB_97^Wj
z2sD*Ja*_U_q3r2Us=9XvYYs;_;@BoC;$Vw$eDw9V!`DKW#NMv0bQd0edJWDS_+PJg
z=+#od`@R}9os{D}ZN)P8<sl*MR*PdKG1_%1K|jb3erPwF_`Ly9Gkft`1OFzOeflTx
zU-dQ7=xeH%qVLU?p0<+g)1~Zt>7%2$21#=U@a+xX8LTqR;6>`*L#n1<AeWEYl{v!@
z5&;-QftX%o>e7~i#9+y>%_KGut~zdP*;aI9+TYuTeoZ)=<Ss@A>{qFi`9{VrPfAGH
zaZA}Np60dDs#Qub4Wbf0%v!rdgl%4}%VFCVlIBADs;-AUM|BMxmm84r1oC!Q<cntp
zBwj|Qre_!zxorn6f0Z*)4v3my*LJ8`-@N+Z_&}4{dOaLa?yZJEP{l#R)LIO6JJPK}
zhmI}>>kWK!F!7WEXj_oXA~Yk^CIub$RNO%VA$|tb0Cvo~u@~b+eFB>?4oytX&rePa
zH6J-hF$d*qK9|eS>ani-Xph2LkRENcwQlWF%2XyMm`sF6fV}EaynO_8ZvXl7%J8N3
zVrb8u&(7Ut-_}*<F4d{}JOGk+51gVMJ2UK;+tQE4eWTv~ruQIb%SU6o0I9Hnd5jcS
zum#_;r!2Sr1!PBRcmVULXdNrqTDh;b^Qq6Ne`gm<hE|aFKnM_jXP1yuc(lMZ9Qpl-
z+3u93FIvEDgK$@A%Z_D9&HeTiCjS2}HKJ<p(2@Vpi7;AdLe@F!YZW+FZTsISa2|Bb
z-pUqESt84c+Q0(Cq|c#q6_Pn_y1_b2WI1?~kJ626&8tMOk;UfIg`!%(k!l|+jmRhs
zItDg%+u9C{88Obx?1>YzGho}`;b&ETlBUtX3%<cCA0ABQhpq58oKFpYn2dGHj__gQ
zOs=pBR<$zem=9m}3FnPh%dvmYA^7*P^1sBs{~G)LYyUF){wntURsS;kE)Ksd`}#g2
zJWjWT5Q`b&uP$b4z36vH{zsQO?{K+unR98&QWva{?j&N{tyLTET<N+#_pNZQ{7z$w
z=Wp6`kqer*a&c4SZFbpMh}YmK@9lS`JLJ3^T7>udKFs|hJx_C*#z=F)QbMB#xdGZ5
zm%V|+B=i_upVc9zk;D#qqb_w{@k({y+V}7&&3@7K#d;KsI1*ku{yY*>q|-Y*nBh`+
zG&MR)otp~2yUJSY5gN0$N8_`<L2;l}gE^jzARFLlsXT1xr?bO@$@`sD8u~t{B%1$5
zA2-MUbL%BssdrWa6-K*3*LbJY`Ht%Os0ZGK&O+?LFlJ#cL9N;|>1A2>jNMkBt6m;d
z=paOxHeu2wJINFxJI~0km$$f)Lo#SGp;$Wo5H^iT7Dka1x>zdb3tk9|@dmuvsGBiu
zDR#w`UXC%b@~dK|4|%0WA$2Aa4`IuZP{st4MN`O^JtMI`{A<l`z%6EBct(dSmD1mt
z;f0jt)ppV6CEsm`wR|l&O(DgQb`@ID4gAZ?dgKaKt(OvKl3+YW3T+>^RNFoXrNO!w
z!nn%u&$BBFQD0d@)ruBSt+{TpP_X&;3WZ50aAFdEucIR?hp*FJ8O-G53FxVvKsb&V
zB+rUx#^T|Cvn*6usPpoR29e|z`FvcMR+GtC4BjZ%`7xUr6#t--9-rGkHGRltW;rxH
z)ju?9hSqC0cwdsQ-cVZ~GNVINVK*5XYG(x>3PBK^J7ZaaGjjzK2@k@i*q?x+Gmz+q
zb#X8pzF>xihY@QeN8%3Zk@o;!w6D@5OL1Ti8L$X`lJ0i<oHtv1S;Ll^jS}~=v*D9=
zSG(SlNY|2Od+3{k_V`NBhjzv(bU?GWPFtW@-?wBs<dkV>kdUvz#stzCiZEauA36EF
zMguTeKF>nZ!*x}YAbLs)t|^gu^5!C>doO@~&vI!9V*vSWZi+nTv%_*(O6;$<g!ls<
zFPUmZH%kp*no(nGH1##D%Iml;6+3qjA2@f2O}xt7tpjHO>jSVQ$_P?>XTH9vc9f*<
zqbltD?rm+K#%^E7-c$MR12ca3F{g^d-h}-#|2105Rjm4H&#M&3=*p6=aaAUCrcJjB
z2AxXPrs^b8ui7qg`Ey$NEnaVxTnKRLI=bIgsTD_mky=X5-3+y(Go43{4)3b#?X^Ca
zg3?B3&i+u1BYD{iGyl9Qdw%+%ZU`&w?jmlT@fx28KK(q#-ulp5c+fyug$=nT=&ut_
zx?o4{5`LiM6F8d2+}Scnif6i5Yq?*r1ik~xLa_%CHDd{Sx!nXVyI@i75?;ZXi0YlH
zM_X0}{2s;;+gT+N%O^6Se7-Nx4^vEt)|-Js&L1kAIez?1!R^cE2WZ0y;bwmTXQ9jq
zaoSBP3EE$S+JCf$(<e_nQw`z*I|?$FPA}h3InD286Kn(9l0bMs&Y!7FWc@h=1SHr7
zwk44*g#5YFl^d4X{N$(tgso-bRA()5*Va<WJYy|_H+7AT0dG#Nac#70K`!k;s#>Hq
zsP>Nq5P*gGWpQ1BE_-5jeCX?H-}%O04VPzO;)G5av}}2P0o0=K4K>Al9XLnUGH}l4
zd{e#`b7ut4BCVV=S8R8pU&iziUk1pepb6kYy7>?`Q*FhH5~cuJT&Q@~#qtH2#(!(e
zd9xE49UB^!2>&+1M>>;uq7T9{q->p5Vqs^r4?;2$7&tK}k-!1j3gGr@l8LG58@IWt
z8#?lhZR%@Naew^raHP^d&_~chJ$rov{gp`gaol)25*>sdhoU#}I5H!I-~N6h`XqzA
zpI<hAXMTy8w+2r`q8juCfXgF$NiG8a2=I2(bn&z0a6pI7<pb)2&%rE$+-q|SM~-Y9
zSy%u_B^OXWa%2H}Wy0RLT(2)5I<$OUeR-A%3SYbD6dKIkH@83*&3jZ|<Ky8l9_A}c
zvrOfNIin91xGAjvF<%va&(DzzVoC;!AbwYp^_AtCTV~_~d*B5DK=WY{FEU;qc=8U}
zhS2U#Xm{b*a$lyw9K>`=)3!(@`FCZ(e4IJAF%@m@Iq^f@CaQSf)RjoK_Zr?;657c5
zc^dOm6&)^jz=0c>NI@%5+f-Xbe+r<6DbmT}w8_fh9y!avY&@^Yk$@`JC4CY+Bd{eV
zmmU_RUGHwq&|{&l8J$e^1%e2xiUwkXZ|a!o$y6Yy*P}z-@7lTRp6JOX*1@{ULr;Z5
zAxYrq0YUrT_@FoO)=z->yDwtyYw#Vn9`|Z+Q3^~@W#6CxY)O3_5z^6?e`&Yl=tW&6
zspl?(#r3u;H>G~X68ZTf9yed@CxzmlAWOV7p4iyfw40pQm-`jx6e<yK@zT!BH{{@J
z^QP1EdN;JOg8#ILj20=Hx0NWgLqeuK?|2ny+u%(dr}~i-^exbvAAbGoUmuOW9-F*H
zA{H&cu$_qfIdr(%QTt<E{k(H?7hg@K@G@RF32^0&lc)y^H=MBXVYo(Is{X?Y)17>`
z^iB51x`w{d;=oDKt#h>hv`H&RtLwFW_0JFL+S|X2t~GxjeX!}ryuMxbacFGcjBmhO
zd(4=D7Atps#X{4iX2pIAVIH3JmZ`Y}DDPTI&xjl!d$!22uf?050*pQ6d%aLJ$kw78
z;<~0D$o}e?RihJ}RLI1RAR=sXXAoX>uk^(%wqrakkAzJzSRr2^!R8#AcsLmF_mP~X
zL~_U{WN)&e!8A6fXPo4=oTt**r-SDS1i%9&VkaEX@mmyJ^|+<ep~t!tL{>@UmJGrK
z{}I^wKWkhU>DonR!TX~bRyL6jWOGAyGqj;xHjtm-S!R@NNv4@n?s8|frOpq&yw4v=
zr5$W56oJbP91DPk#qP4_)I~;FgO~nROHL8o+KTw$AXx9+iUe;1?W#|={!U*xet1R6
zkWL6HkQKhRBO^ybv1Lomov*{@g27H)!0s}^a6J@_hSD3@BgXy1&JJA=Yy~$|Qzss3
z?$Q-i<()lSmlPg`Cog=4OgtWgyGh~sNQcll764nPu-EA#yGP%K?_B;Ja~ZgpmYUmW
zbDf0UmsWMMz``aIyQzu}{ja$+wBr(n0H++W748<yCyVkA&xjj<a!}x~)H2Vp1OH)u
z25jf}$?;g>iz&C#2PE}aA@#+={{+twbZ;yA`-&!6`U7mbP~cP1VltDN$qo+Ws_0xa
z8i~l^%jlZi{bDNrI{2{pCxWRj7C%xf76VT4R*YpS;g~mq6)Eati5zN}&SpP?O2Xk-
z6o+UGd-Z<j@@LTZ&tkvZ39K=+Q^X3ae(-zXhiq5dy`MG@gS5V7*=y8|U;XM=Z7=ev
zuf4_mvfO2RXEK=^zxK7SB@2%M_x4rX<k#K;(8D*I-gpl&(zZ=Mn9hKI{HM_6PI2Hx
zMp2;C;bn{<uD9=DV5iOR(H$VOYEoP2Ret_Y)9Lf%=g${LMhfvXg2dAwd(Qa*?7Hf}
z^Z#e#=9lpTAM{?yZsA!94(IcEef)n%Z~ziN#0SpfamZ``FKG4|z{xmx%OF@vx>TZ}
z0Qd^LK-&6;v~V6yF#0(5_;DYP2HoU2fHTR^xup9@B6h)bFOUngnP-}NCP7)jHv*l{
z{aQaKfKq|PT`?N8EgDlv7<7L%8NCn+U5F;ZQ+q2O^~A>m5&di=(Dm8w`X|XhQJajf
z0G3&QXXqMX+T7M|Xe&nv@(~~S0kq!$o_5V?9V4@Dfym?UTBKh{CDl9T<N>XP>rAV6
z9UqV58>9`N1@7WL-810xZw2mJ7av+H<M7vm56}qWuyg?22jVY^EoP+!V&dY_4NX4c
z?j6in<0Y}aWMANS1;>C1YBYWkD=HQnxp5?(OvS(X<~P6j#y1DU$HT#(N&17TA8KHl
zf9%Ea^l?luv0bCyXuO=SrBZRc7>oV7zUCyIpuEd*WBjGr13qTn3W<=iOfG@lP7=j9
zvb$_NuVYop6)I7<RP=fQY$LA-c?6j1b%N=au=tg7&)jy~3?UGK4HQ~<N-5^*cez2*
zvv_=8W$30Sr#3Km6CURq6zh&s@+<Y5I(zvm={9vm)F1PK_1bQ!zdtf%DS`3X(qy#1
z-&-0=1~w5)N#RBgkiF8GVa`#f5W{zeGikepvyvew1XV7Qx$^mrW1XIXmku=G9cxIn
zS)2&EV*@|PM5vAo>&^UAD74Mxx@+Re=yuo1$3u}w=%ENO#l{9GVdSCx=UiLFKYbWG
zd^an|*#iL8((T7cs&NCbymx<H`_Ij%y4uop(rN1$^qoXAbjCD4jmjuCO9F`?wf4kZ
zWSp!BpqPt{RL<=kb@;l6AAT5-O2LO+Cx=ai#Ru#t@;zdx4#<Si{PF`2JP-&x;6`%J
zmQL;#Q$k+RUiF3%;okCz#M_ju7~tyRSO8Prletw}rYRl9r!mn7w&4Ez?+*s=-`|3s
zk!J^W5b#8KN_9jszO{nbKx`X{y*SrwJKbBYs6|O|zo13W?*p4_rh5<Avu*DF_IlcR
z3j=JYg$+M&tI}=q?Gn@s_^_?!(p{}RJ9wxcG#Ll=u4{b9BzqLvCrD94CP=~UN)-z3
z<z?^0MXD=P`lyPRG|+H|9B#JqVK*`NQ5ChHnnesc#jZy4X={Bj;!vl^1XhS$Vv}tJ
z+id`OHk+D#O2zJv&LxoaltiRNd84^&tFVvrRqltYl_~t}p$^&`j``?Awv3h5r6eNJ
zsFfKjuT+7^sfDD3i6Jusa@??(vlAmit@F-2wmCT!3{G&7o+?XLR*A>$e2(un|L&dd
zd}q=)7!N#;hbMx3-htVOaQ<6ZPQM4beHs|7TrOMPoCnkZby`5J#A(`SUpbzjV3rP^
zsH7$>4j<&IEe?Cju(hkB3zJvJT~jw(b<Wim2YV|K3gzoMj+cG<+qxUxO7d=JyXUCy
zsA>Khiu)YZ7IyFArj(v}!TtmsE{lW61FP}APL0G48;n-f*_|C`-f^yacTYP?QNP5l
zN&HNmU6VZw>sP-m{gYfiTZq+L7n?3w1+J1pv8KqLNT{v9(7;S8sM`9Pim7*|bAKW>
zKY#4l{Cv#r=CJ>KyEB=4b)$mab0#VqItATl!tX&#3c44<eqpzZhZspe34d{&j9-@l
zx|iD+TZ;Aq!c`$abgHZcv<O-gOxar8RNa_UC>7#yu0XCo>^RbWHe0(J`+yGEwK;&@
zLGPX#8Vr=|+a}b~vOR7~FWvvj|FQqV-D9TvJEp&X>au<Raa*~EgX<U*$yDBDV_d<~
z#$+iLZts!`ysHV9TiNdeP^Q8OVYytltNvp2tU`-bB%O_3wD-I$Ih+Y{Od@WVbi(vl
z5^lG|=j*V)?g8~e*86{8Hx8JOz-~-iC{<#I$r-tR2>|!Qnp7)#_!x6hTt2*9U4di*
zy^V0_9$P!*zQ&C`9doxM(XU`TNcStz$o5pgfAl9|tq<-ZNhg2bcf=nE243)jK+t?J
z5_x|Lsy+m$_!6o2&-4YiVo<FhK{JoBtsso=FMVkMq682o>94sTV~N0@N!nsP9tOZ7
zJPhcsrCDvI9=Qd`NH6Y`UzmnzQwMO9@k{q~340zwD&YafNbN%8>;Co?@{;95l=7%{
zm>{g2-7|i>Q--?Y<B(^C?;s43YRE#u5+EPcc1PjDC+hMzOffZBJ(~y}KnrVT%PtcI
z1nWO3AR<7~0<e;)+9gxffvQ^Hh$^4lwa_Mos=)c@#6Nf#an`<~wvzF4wq<NFm-b=q
z;@$QT`Lm(v>9JAASzo{X_VsmavOP9EJw)|!f2X_aN_nGp?AW3_fhQM_9jk4WV=?op
zPWM%T<JNwgha|6_`0Xh3s}cwS<;s$6r0VHar)Uk{Y_<+7uoEy7F|5!-k=-!lNCT?D
z=bNPXo1I5{XU}dZEBZ0)`cox_ICqZ6s_Y0vE&mpT9;+j#98+o??9~hK`my`s&hX`#
z)&@kZd7N7l$#n*ItG_O|7VsMjz&~y2n_aTXZgxGb`#q6+m0%LmObWM5ehkGa=DP(V
zQn=KKPBw`{IvA3M1r1uP6n6B8^IcOX9kY9)4lsMoKAgyPlct!A{mi>$$4y&ryDfC_
z=a|dE!*?B@=L_&7u@YhDcua}6jzX!BoPFb=;<_?nR~re)o3;fzR~rYmQusk{3|s-$
zI%0<gkVR(^GWPZGPlKmc*JxOGbB;!t%~&IKEKY+zToKi}$I}Bm0hWj*z*f5Xt8Bwp
z>jlZ$$+{ig&dd!MUb$Yc3wCJm=w0c(?cQ@uvM=K}nf`tr`<zTFLq}}9*6D~N^Rg>-
z?s{L?<+b{*exTQ`Th&xw6$$rIYsNXlAJVM<arx8EJIoisrye0@q{`OQLSn?g^{Gyd
zRE+5w4D%s@RqD#w03LSQxi_1PrM`2K$DgYX4k6uwO~(MW(0$GK+;`u7xnf6FhTCqu
z;b>>z+7BN+ckX$}sNnZ9E=(h)y7?f|Uc{pvi5CV37CZDR1MmPq4lSee#(e(!Vuvud
z=n=S{Ae!iQf@<6AruWz)K4q*_VoGlwt6)<{?;Zh<8VW@}8U>|oEs<ucgaqWB`1Zr2
zc%1LyalD0Qf(GqDCipPy!M-fPNLLM8jejuTbKNSoSBX5>Ap`tT#%$9s&*XE^|MUSw
z40f}3#YbYxWU|SfXmlr;HBW?db=g{RauUW9OmZ}YB&<bZzdCnyzd*upAr6OK$CrdY
z{9C|nqtvoM0>Lap=I(GBfHs<b$(ZgHhHpsw*04y7&kcSijWFa)=FrkzgM)W19Rln{
zBk9is-80iusU)nu{!KTT%&nX~d;O{TSZw~(^=Hqn<dR8u(|;Oafj``Qdu4t7$kC<e
zc=Jh~w;RXy@PX^dhHU9JMJuFyTu%E4_?^yUvKl>ULXoW?4kFOr)7Kl-3SA~A$o5TN
z6}xodT3Pk`+r%ADh{NRYjM!FS1`|abw}>w*0PDE4E$3SD3{N8c64$NicnDH(65tL-
zVI#o^X^c$kN~@LiL?Ajo9t|XVTp_Hq^)`Us$%8kc11K+JA#!@uVUmqO0MvAkymK!(
zv$RN?gZVrS6^lz}erKmwLOY0dOK0!|p5zOH_Pw>VgE{$L`5iVa`ffH}@xAgpRQ&1!
zzYcl$RWT`#DwbKbK#jX55GfE@O;cmR#gC=OBeWHtOsA2|=x%J0-qKu8-`#vqOUXTb
zxAy>5+vd}W1R}QYo@RWgr^m+A-)<GX+v1(mp4$=Oj<4DdKPV`Gi(m$H;n4w*2JMi!
zr+0fUUU`}QFuT1P)b#DY%3hEcGnaoI_;c4+!fL>*8<^}cT>2-CA*SM=XVQa{=M;t+
zNDU)6Ny!TZ4y$WEl1)63iro}VPOeNP(yz{@o=7BS6WMz@MfE29pz5@$;Q&|PCsNs0
zr}6w`GI~=C&m;PEHqqSdR0WjN={l`yIC%PhY40ZNtk|!oMp=tElqw)@5X5#KUO9|u
z=f7NDbf0;M?r;qyO`yJ(PHW^S$1CF)c>Y_JB#=}D)M%+hh0@CayJ|07=Z1=f#L=U7
zLRL&99Rr)YW6m8pIzDcS1L088p9s1A`Kvx~G}aeNoGl0933vFILGC46js^bG1n)e0
zG*KwJp@csf3J(;``1sKy=ipvw9PnsYyPH6`av~lmpG}1NV#kJWZWmpQ8RJNHWz52`
z(vmJQm(oq*Ee3@!{q!*5m{JH4(+E(eu9Y(n>OMYA&pxh~pc~-m5GW<_))c}LSj2ld
zR53j*nYcIsWN_}j^UgafCU|POxRA&dj?LY<8;u6d!ilGDxc^dr(KsicI<b1r{IOT)
zF9zhHg~Tt)OLrDeoH+5+*$3nu(?DHw#|pW`LUH+2&{Xcc8_CGdt)6)5r1Nwz8omF9
zr%o)GDO(7Z_{j39HGM^^I_bzOKytMG=~xT`-s{xoUQPepRv=oPqT9n?;mW6aE2UGN
zQb(8SquzO^ZVoUQn2VkAO_vN)@4SP4if^FI`Ss9v-S7J;--kf!)UkrPMamH>JW`cw
zuOOA0r{f{>%{9+o)fPy}PXi4gf*L%Uek<ku7p5WEvP!LrRai?RKi8`a=u!56?E+;H
z@v={?f)y(}lqCC7kEi-rz#iF7W>s%*&CkyVoWiAo6PVvIB|W>yvqmX2<ornANC+PI
zNz>nP6oIQ`E|OJ9i3AHtpzryVPa6#$pHBu*WZP788_7lC?2FHaf+)7jmyn<!79Kly
zZY&&Im8!Hwaug1vRKoZR{s&LA3n+ul%gcx>Tr*Lp@Tn@Sb4>-o6LP&tlxn47N7*KX
zsBo&Cfr+UpzsaO?ySX%W7V5|Hh(vy1YGMG?K2hJCot=f2++?6=pWPG3yuo1k=&|Qi
za{2JkP&l8fY*g=$Ao%GAj+TRisIV!oV0$#`+Myefyb*X)>;T_~eXCyWd!z5|SL4m7
zm7v~&pk#l^WGiUd-!;s;&b|FrT?gOeJ^V7~xjcpnC3o!#F6btG%l1}_h3aHSDF{Bp
zf4!p`dcM@?xZ0Y-UCdz}YpCsE-=VaY&Z^6y3x1h)OV#kPHx^~Eo4iVER+&T{_?4XH
zswD9DkV1-k7x-^!?%m=?rXuJb$(kin`1>?w)|5E8&5vVpWomgHPgL<}$o=%E-4I@`
z>NUREoW?t|G_r7R(G}2n>n~ab&vX-dBMX<RIudZKQ~p6vQPN+Q#cvgKMDaygrqn?B
zi2y8UmY3bFcs!m>M2iFdKL263Uztz?zE$-RKXqq6mT{u_ek&lqlo0w#LTGb$)?DC=
zLvKq*+(guI{FAQh!hz}TOzA`EbW&7mnOHC)LNx{J=Fi#=&CX(du|HM7D)(6PvWj-8
z@l}Hjatbixo~>e=>qcJ7?w?UvJ~FznFggMieKCx&bB5wOu|ck>(C5(H|BJ2N#eTBI
z;xM1$)5xZY2NOeq?d`x2O8=vyVHoQ+jftUcC*cuBn_647=&A$1FfBj@VyNXWn664$
zIaADpn3!RS<9}|sT<(M$7(F}V%YHc6p9|#{7I5eXqna+J5=fTjcLEQmnomC*V4^kb
zol#7uf<JsgbL_XEX%_3fBXsNnw0)^b$C*rl!$sOxI1>h=k?6&XI0Qp3n2h}9r50cT
zX;`@00aV>0dHAt{srDv{Q#wI3*ZVw&9v87}gddlU52%_}Kl-2~I|GpJTQb1URBX8=
zNa@|B_)4m|ms*KmER{;uxM`BIdg{{OpWy28fomq(p27R>On2!*Dl&E9!c-)+Ws+6Q
zE9SVGG_0Z3WnD|uIWz!!HQy_+i>{5vRlK0PqO?^Z78B?mPs9V(t6{Ay_*0!56p3)l
zD1Fub5{c*rN{U$c9G%z&Z*HN5C2c5w_@`&QZg#uRyPs|B?%HUHo%)WxMJB>yB<i?A
zY`E`=KG%9-hmU}6gfIVF^KJ7R*!Q*wo!1g1o^wJs$Zr=h{rm?;6MXiRTQuMlX&Nll
zvg<@XZcv0W_rQ~X%4fcBaVYl9ocw&_+o5DIfA^WsH2(rykAzYIlxxvwoumy1<SOq`
zG`|#Zr*B<6{7vuY6L+7@_a%a#Y5o_~!7?(x-W-Y4Nm6mB_AGMlg|8BOLJpW`1N-E>
z8ml`f;<*|G?Xq@ryjVd`WhDcuU3Qx%kWi*@J6Nf8JtLPr{<3>R7zG9e1@d7mp%KO6
z^q)=_!A^Tdil2yt6Oq1f9Fb5a9CTK(4_`bHj)Yt!r%F76zxIVc=EQQ6@e9@kN0h-A
z@#tHbof{mSn_aPHjq6c~<Ax&PWIXH-AC_Vwj?MkBk0`=9^qyS!Z(3uku2$_FJMct2
zK<|>F&d2ksvEwDMvR+fg3q>;3DML#4{opxBLFdl>p#9yQ8nmZ_Q@1_x$Zf!3YU!}Q
zPNMJ@YnymWAwjflWyd<`c?nqgW@2TgwiG0!{v}MYOzB4CUdW39`pW=@IkVz5fO#Xg
zro>Wkw%B9<-cBtX-W><Q7<oBq?)WbK1DYRs^wCEHhvtue9Ii-j^81~>6=X*q4=tbm
zR;Pq(KFNpiHVF31M}Vcqcf;ZThF4E7&l&n8Jp$6|4D`j8E1^Sy*K~^098!q!o=gKt
zudTY{{ww^$_8mI(eqSBWorB(_A32a-iW$(v)3oZ60RYjFE+O+1`mORYdk&HWVUHRC
z>zT!;$^LVaR%-!~e}I<&(g11zC%_Zf3aHNPO^uGGzhaJS;L}PpHdJ0{zUN!t`j(Rp
z^!e8&BiQ&Vaum=PDTPhsbi_m=6DDn5F-UVcjpl>x(oQa0`s#6w6<9fa`nm!0I4X+9
zDkEkBxkLGOxCBjX_$Y=m8lGD7kB@!?brc5%idvgj!k95f+OEh)F=)HyfXSIH%ZG$!
z#8MoGT%NtbiBM~Xp3{!8u9^iGqoT|_rQP)!o+teB;q@BCEPaQg;rM1|gG0mX&s#a{
z^o0j*zWt7K6Gu<fuRr;9XJ}|?W~s05?8Kyd>crc}iZ>lPlyS}E)X5u0{70vzQWO2b
zrKQd4Ehoa+>u<jKTxD@__16>5=8ZS34xBo5Z!l1)&Q1h^hien1pqr_kg&i`vbn|oW
z_Lu&As8Tt0?52^@&Bu?GUvTW~#M;`)vH3gBo|$UC)=bY%F2<tyayCoY0dojGY7cAT
z4%m4f^!*TQJ(?}9of?8mQf6BH$nx;Q%@Zb6hXM>P;$S!F?ta6o0UIB2oRC0Ff7WXl
zEiFPQ(&2OM6&YWS<Uea?y{fgYZIZ<EuXP3JPOog+XF%k&_+9SD*Susa68VyKwm!CL
zKi0a^YLGV%o)1k5@Hme>8BSo0f^*b38xAgUBHYCN-~~lDkIj{-6c#sM%m94Gi+1e1
zL4|haiYsOh+H$Lu-749A*xMGDt*zEpJ&#fT<=<oaQ*c^WN>qFAvL34bMg4CtnEJMT
zuN5fPdarc_oo37_AP;<06j{(g5h#<WObE6Cavu>4WJaba&|a$D5xb?WtXz8KNa8b>
zE?rtljJ)!a=UURXRs5?5AAGQ31|NKI&@@VWw!3?sG++YplH2$u1WhhoJB(bwcB~1u
zM1X7l?xF7a%yNaw<^yiQo2CGU78=Txe?riJq;f)Q8`3glikSg?Ce5d8Lw9UJug{ng
zEb#lT71N<mx`^S!>k|9Q8QEsdn}otm%xvomws6HAy^u&{b}}hw3cWcIych>gd<n(}
zo$}N(K>Esga=bd-8$Gi)h^N9R?*kT<>DMC%vNfELjfqdhJLXg1$6d(i?nd2qux&`H
zgCps`(4R^gaO7a>tLz!pO}lsjOgvcm5C)aq-EYmE;k&Hl{a3sxm5?w3|22UBMrccz
zq(iR|JXM3E$q;tcPh)jjG^0dQ9ydUd8C(J}V9~q=DXmvufw*);tiNJ4=G4}o_dKu#
z(H90SQxbIgXT!K32+#F9pG9twcfZUZ4EkU8Zf9(KlRhTv>zm_a%Zb#Crt+9cCHmI!
zMld|P4w+Kq$}c-NV7(;4Vc+b#3$wFM_cbhcZ;C3A3~YPJKe1^AK=is0>nY5O&JkxQ
zFeUG<<6Uv_F9s3vS_LOC?$&MkqpiBlYvxZkMiv&Pr`^!tn}-(`X66F+j+sSb=X5HY
ziJ#uQ{_{wmje(FY!~`<8_xAQAzc;cZ{p!fTZ%>CpLnDieGg2rY?zi{qnOrW8#Iot{
z+jz;BpjY=`QRjxf{@~X<9|ZM{>;-_FnI69WMd~^@0xkPW!M$dW|3?tC;aMOK!Gwj7
zhimux|95!TGj|+%1#KA0cL=<(Pj$qe+%5h<ieR2~&)WE1mZ_qv4zv$^#Pz4wuer-D
zbpH_A&xIq0hb$4e$GgC**FF}|)oKMT+Sno!`~?3n=kDdp;LA>7J;o5*)#gd)U)J_2
zMYjp|)G!jtEsS&e?eb))ny*5iMYfekl{rypg%x69DBFP=l>3khxZGhBU=482C><W`
z?;kv@$A1?c7?{|b7#N7cw-&qJI>}5WF-)ZE9J}#_X9@9{crFIxPz<K5RE>RhQapa=
z=`Y#wd&rK&S1eCpw(Ss@lry)mFz4v;_rT4@e((ojQH;c31mZVSS@?_KGr?dM+ixIN
z6yJ`<BK#V^ik(Ze_D|W5Y}hMX_8YAU@Nmj5W^otX-+9o(M<GtAIYWj_xPDIu1!Bm5
zE5O|~l@~C2{6|Ta5^1@WC4^K^eJ+S8i6fCHN_C(@ZmIqB3$>)(9aUgQ4&lA%+Oe<>
zb)oS)LANyYXd>nM9Y3siC3=3SnA(JNB;u+(yuQ%BBQZC({O#ZVZ8Bo|BS{ocYJMe>
z+!!81WuiJ=oR}D$op9i!haCU{xjgp7#dZXV#L(#I#KdAte#IyB$%)yn>&*sS-BL&s
zf&Z%9EXLFI3(PKYFH^e<IOFY#S@aK7g5!j-qYR}0irp7H4RKM?FoOuF2h)c|EyBkE
zkkdZ2<)(Zp?cTRE{QKMp7`X!CzYp<Yh9a(0x$h>l`XBm&f%sDHL~e<@{4VqC{Rx(2
zTw}3Z^XpFKG~Z+mXJ!~@457I?Y`BH;%lK&WMD|3IFMgn11wce;B=lXozO%kZ&_7Kv
zuim8e*CPzT?W;w2R^jc{edg&rkg<$``dp?J+79@l$I|IVum}JqU~1_(^j?Y(smDSi
zNy-bw@vPrcC;u1P#nbvslPm%rO6~9=c;_%-2ufR_aA+E|C%h$BECeN+lAI|Nvd_`2
zs9nZKQ7xh+N_0BiL}DL%PYM^pNZT+thyxOE;UgF;^0bzpnD9NfMYFo8mGuU$a3eUV
zr-9F|iCJp1NIo)44K!|Nx6wF=Ioc&l)aL9(kOD?;G@d3Z=;EW48{%A-(Ec6F_35si
z-qq{c@;jEMP#=J)l87zxckKyf1S@d@rqJu-7F)a>>B}5Oz;fU44QvAT*7Hu!v#A<)
zXL|*#gll`@%KhPRjhIqPjTW0&!t_XYJUq7N)WI92v7_ArGCW&u!NMn6v;udlG6bcv
z=d5h`R=AVm%Cnn+QmHO-Ppdg4%2FMiX-9EK=3!SFzH8m!@yjmPc@<^CZSN`Z)}Ie^
zBJo{Uu~%~wTu;`<_}dbxoRUhGdYh)VNhQPz+A7q|L1yTO_#f%Z<|dCq*|jJ<>MrJi
zeqE<A3(zuAEFuBu)@YL~Fb1<*?&2I+7hrY1N`uZrrJ_0p!5o4^ajne`7$4Noy8gYN
zUBHDbd#=~k6tz<sZ%gK4098Tq7<Z*G7AnzEUA@K3iii0X0ZZ><zG9oZ_)hh0)ME=D
z>;W!^Fdi?2rOI>pvkqreKQg3^U8_+HCCcNV(1@zNhJh4=K*rrOpkp+Yqf;r0vnx1w
ztG$S_N3(-pAZT6eB{d~W{hYX+&O`<~TgFFifL&f+7(^Yi0(-$_(ljNN8n^b^V_kd9
zs9@#mhOdE42uB&E1%DD~Da3nQ&BE4~d!=ltf|xSeIcWj{6}U(vFvnx9q;p&He(R2J
z--RYU)OYrHXeiFE>rdASjTLpwNgwqNf-6$+bv-8W-^zdMD&bUk943dA%Seg#T7V)F
z>65x}@@~-eZUP`z)+um7voB)!76EX(B_%=rp{(P~2N9_ITOkvvEXbzlx#PCL8^7(E
z2!vg}DV4M`Lbrsdy~`2f{x%|TgE_-3+ZQUwZ6RV%F>aubZMk+pY%%J!(wzf9@H3Sw
zY3ZTDpaPqM8o_Q~$%hf`N&#8gr~TO`9^39bTBVwu#jW-VDiN*M+lNm2Nv$lg_%*jY
zO_t&sRu4M+jJ;^|M#op_3bbC_8ngrJw~Ck!<qN<=eGXA5FB6*)RP6i6sHoRD3#3I<
zD}dZWVkMRYSOMd4`&xxa%W{LfLN#X3kV3*^wizIyX2uF$PzXf@=C5GNypPNEp_to$
z2tLfA+1K;*@wnT+6vVDSPeSPqHG3%;y`9XZuAjZ;UgvV+$%QWa#c!-beE|XzDLoK#
zqOuu%bmFBk<fQrOu|eY>=_jkX*S%=1_3^TA2RHWHG`L20h39;!@8!PNb@-tuh6HAW
zS_)VY%DbLrj~l@N3Sn0j4u-WR);9z!Xa+m#u@$Q#l7aDTij0bZ&SC5Fc-Lhq)cO*S
zmr^{YF7kL$!R7+XNXe=;sfI0h3m|(-c5sC&9N6|^W91ewXh<BJ9MYpnb#Ao@9taNt
ztXSife+_HnvcEb_^->m_a%P2YeFJ+^Zr$?M&PuzoRRcEF7DE1h5?U780RENl=RmhH
z8U-RWU_c9Dw9Zp2tFT3?UX+;N?rV5{-+La<(v5$JXV;WF1RPJQ*xB_hKr>i}Wq8c&
zzl?#I&D(vV68YmEq&9k9{f=;wW=jFExkC~MVque>=w<J&o)|d(M~}_t%S&GRpJ*<T
zKXI3LvUjzcQak2>^^ttS8v(qcccJGG@k86Ed6M~{nbSsOKw+?LiF9XO@Pd};9f5Gi
zyiICzIIp~fNKe2PoFy+f9rd(I=F@Bv)ltc*8h_56<FPg+Sug(%zLzGov5_*LO%?Aw
zLXFpv^!yQ8#*fVDk5<4tsDMA`&hdx}`14a86&zv3NA4}A7LL$9j|$|6G#GoI%lb}0
z^SVeM=}FZXfY`F@FkKMRfVznBppXHb5)BT@_&2>K-UQPm&B0>XIgg!1K@69tVv#S!
zBZuT@FEzj2{2MdW{F}5Bo1Z&&Y;HczFwVG^V@}!({oAQ!p~VA{_?IHFLvpm2p5vry
z6>d3FsimBflibjgjU-4cCQ4#`$eY7=x`wsTbHv+YFF2h&0^cBfpEp-qso}06_C|VX
zbbeuk*6TgY#l0Nedr#eaFUE7YYdpPSUE-4?3-hCUse3u_NChK&fW!9DrsOkIU|_7>
z4lzj`W)DNpG|xMM36!P@T~rKL%-u9kBQ-(utEw~zP7j-+>}d#}ldaN^zVChSTiH5q
zKB?IY!p*M^&jb`BnIhlA+j#8T@)*WU`7fW@@s&UgpF<y9t1}TnfyhQ#5@Lw7$dRoC
zzTUo*ZU2Y^pCpmrc*tANPzg|k$Jyf`)7AH1<(?%>c(!@T2Adjs*Ohj5JRIKh9^s))
zw+Pc`;1lI5g5w8Wu1!J>6d?VE0ae*|P5GpWlk7FZIq|C|ESLmDu~c5O1z$9pF-ZSq
zVi|1QT&Nc4sU|@-a$>&Hyl6I?7vaYRX|DNx&Cc0u%$O6+-!&(4sbuKkK;Yp}GL>uo
zE-An9_Wy1E6VG6t+t?^L+`p;5wTb`OC(&sL3{n0>1_PZfWi`+p%$I5aA@8eGhol=h
zP?_znYKl!&hqVGbp=HNjYUA)1z<;)6LIah`>};hnuwez0wmi476&>$`lUiAKnZ(hg
zDkCGAbUHIKQfZgAwMt^+Ht(sD_f9H$Tf*&&--;dV(ppUie9nmn?7MA`8GNJt1$^34
zJKcmrD$;+nSQN}0M?r--U^9k+yA+17)WH3bqf>T-*InT29vnBQEdrBOJq_HaMZ&gC
zl&%ootm`k1#+USLfZJplwpT(SClkZQeW`)R22z=DJeF}np%uNk$wxLJ;#uwjWv|$1
z)o3-Y_$uosH3RD|+`I55H|G4PG#M>2IIkE@;%olgjjTv+my(9M5v?|5KLp$qg&g)8
z^C9G~J%ce`2b)_H#@Ex83h@h)3`|eeYT}FoYnJSe4v-_}lv_|TkzjB%SB?du|L_md
zz+5mGNIqFi+;K;u_+%3KSm#WzR;$Ibxx$+wk<=e0Q-7F3TF`th8>>0yr^91o;pZpf
z@xW3b`>uCo@e_|Ho`2U1U--f-Gm!jJDjx4!>BIl=c<Ps!!Xyhx9(n(o5_`CVtQR+d
zYQI8g7!4ErBkrOJf&VyZ{Fkd#;c;aDqc<3dSz%>bnH<Yyk&E~Ep3X}>QcW74#-C`Y
zh*v%ofKxCwvUUA|$ih@5Tis~SBxbJbWJ{-1k3P799y)aN=%GVufJFK1>f|8S2X@6p
zCOV{mOlF?$e512mYJm%`5^uEQV2i%%Fuzlv)P@s<X={(qG|YvD>f);p(cxOh?v9V1
zfEN{mV;;Kk2h5wBA2WA0KbAFD_Isn+84Y{ikxlKIvwlbOZWC|*y@^lGUD-o~@Cx1B
z1Anpu?gV~8o}JYyZP-*GWh_|E@fzg=NF$a}R?VkcL#!5*^9+muRvJ16q^()2GxFZ}
z7&OmhVa<l565De|RsW13<ajn^sd3Fg3<!lrof34jB^zl!4ky{sYIQW94~}n_W1<Yd
z*f|Rqss8fE%GFY-+KI=9ec$hUAE1?*Tka^$UI-+jh3FeLkqx@=c~H#<J4Sp(#Rdav
z@edPO09y{mTvSi%P)_l684Ovh;8RJ;MMuX$qOphRR?&HJ^tVEpenB;TT8FuDd~$Mp
zsl3t3K<sP)KuQZYW}H~0-|-hRX(t<YO)wNsWf8IA_N9@XJw+GuShUFd*=z(6T5dR=
zNN0k6MASN&OR(0A&Yj!cJvTp+vO4^x+5#}L3!taz@uPRy6vlTQ#V5-tQ+F~6$Bm7R
zMk9SD5D54~{(wQc@V-<A$ACXczjQniFO5c{e4@_}V?e;4&bZh+GUGJ=@9BY3X<(YO
zSGS*)0@sJ1`TIdvKkG}tZp6HO9BPFr46#dn26I*OPlg>s&zVt{xX|8_89`T64R8w9
z$Y8<`UWso_sl|Hj=kvp}la)Rv9Sau*KU7*+M5fDNCLRnU>e}S<!$Y|r4f*|n#O%z;
zVBd3}LuMGs*P{=J*~21i^OAvdF*LSxJw`RuUq3W9HfBmrDmgyBfjH&#_;~f`-0Z+$
zbPCkPML@OVKS7>+XHn9|(kNqTfaOb1`QGAN@zpKf6wqSs8Bu{)Pg$gmC2>^JNzt+Z
z_aF3lvYH<U>{i#R%Y|j98uw!-VQe__+SjqYsk*L%`imDUi<4U3`T2#3H#OgD%m$z1
z^RQ%F-`C*@EIGb3_+ku+fhyDEdrp~80|(qrsVqCSw(O}`L^<Pet7PP12?-R$A4}MJ
zv4^UVEWVX&vBx=h8Vk-7W;=k{tuSx-$}bcO#nI7XCc|T)@F2M6@Okr-5yu~Y)5PM=
zj`m>^-MMws>zk0$*S$88N=2GKiZ<8!N5}ALCWETRM*CkDiM;HKy!m465F>q?7)Ajv
zv)lKuW?+SOKX;*@zJML+J;@(L*mL+B&*vnyI))xx3xSkCXw3@%%E}b~K{hR5m1_{@
zI);iOnowv#E?nO*KYVwigt-gl``ki09t-)6KMW0ZwjaK+Sme+@C7_CjS=Y%082aKk
zNSxsMu~0)<CzCDu{R3fn%=n(osD1>}zVXBN2P>NqpD-H-`=asEcrxfj$uwi1VeBAK
z$OZ6i*N|5Fu}yQp^hIK&N<83~hnin{#+gt)o4yZ4SPtRcV~9k7dc~&Wsh}y+f45_p
z>w^>0DagI5ph};H|Ln5>vroJ3*1)J^bxzLcz*h5<&)7S=<{R+;KJ9x4R#T^kDe0n+
z8RU0TJVwybWWG}^Pg9Hq053?O(x3=NQ1Df;q7Z!m5n7&_NCPX921>vx;gae<-nGg<
zso350c7`8%AUk+kkaTGvSyGX=lmaq{q!TMyvzJ{-qyb-{RLK)*lLIBOEuS9DJ}@km
zAUpOwDnxA5&rc6OkR{^3ZK*uaekY9NQ{mW3c4%f3K5IA(WmjThq@)VBzd2A|x{VD%
zwbSS88=hDKo+NVl_nk-0r=V+i6L2Q_u6@OD2#Tb9!IepZK{QEa=bEfsNEX;2o0}tW
z9P-n6l@BVxW{E@P2`0p?%h*p_cSVAqw`fiAi+RDK<~yva;*@SkrEUeT$;|ZiZ4c*i
zeL9e-$3r9=mza5+S;mL*xlrmk-lN!L%j~dkem{@#EaI==p7+?7{C@vg)^%$tHH*~e
z>EXw;+vid==Z7}%pG{!?+CU~VeB>tY>8S)cBOG$^zGN~_hd<yn>OFK1?0NXL9I*Fw
z8`MBBD^z-L9MxNlu(XbW(2><~wNRwe2E4vS>Ht>5lY;BaX*kz*Z*2W5a{^L;gXabg
z+TwcCKJxMd_PXw|gY}aHtp;E0wUc%0<^ele!_Ei&n`r{qK%O_N;22a^$zhRDXdQPM
z6`*eXgpVMubALKJ1iD5k$YT)kuK3L7b8u#GMa%`jK>;x}b@=eqlw)2|iiRH>!9)+9
zsQ}fagFhBv3%A((z>fvfPGq8TW-yf;ek>d<Z5w1=yY=uPV-6j@H56*}y)6`~&d!bu
zE-ucdoS%ur<TZIe7Wo+`HM_VtI5In14Tav;*TB^K0?ZHYyv2M0K4h;2ruj+Vlm8!c
zZvrLBRi26FjmU_|HF95T%gkD`mae_4viHtLOSd*uT4)K;MkTRGVr_(w1PCZVw%P1y
z1P3q-xC}O9Gh-t#7_(R|z!;8K`8*8v^BGT#-#H%a_YTNDUNG3H_kI73$c(I|=@!g+
zua?MIZ^pg%zyH1n2M_W@!i9GwXNAclZe`zaMkXJNh5h1rRwoiMOyqp8mJ+Uz;U%&m
zQsHw%IIH-Vx=1*WvlQ}4|B0oVT`ih_dtW4=l!xfPL2E6jrEeotgvsXIWISfNH|A72
zQ<*5vL!TxRTx_8*Q9%x@+>Ne@APPk4AsuuaY2@R_!r`#_CX&B!_-Mk({XFzti{o}2
zrAok{s!q__@N8TIsbUoqdwfw{2wwZjsZ*zx<EiKI`}rPcZ_HJ1=U1wc)7JHx1-afj
zjdVVEv)ik=8_(h;6kq0fsrd4#!MHsHQV4Q6Y{duv=AnlkT6{iVfG?uH8SH82W4OpS
z;F~;u(c`=5Lr@64I`oL<N3`zb;L1UXx8OuL%<&+C1jb#EJ2))P4rhQUU+;NdC_*G;
zZknVW<d89Y!kGAGIIXB51VH>QY4<<a%#b5}1IZXXVwtuq#~z$bWDK{As7i56s<~Or
z3gxv8?{hUe#KcB$XB(2(bp)0h*^$`D$ZW2#(I%_He;9i4gY*unOH4sg>Iidhm4?yM
zI$F#M<Hly$K0e!1r`ar<mv$ic%+~7(>utJCl!q}wnWw*^K8f*bO9X+CQIb*(+!P#y
zeZ??nYYvBt=$}-BVx9gk3TUqageA2DMLr_Dmb&3*RzNC%+Y9!g=L}ZTla=Y~Co2=#
z@}XPKWlF2A0mgdcmRoMQGX^23Rm@!fhR2RrQ4rEd)H?oCLxbbPL*--8sMl92BZaYn
z_WFI<_(&w>Cf08U`n)&crXnNp%*~!ImAdOQz~QeAy$g^);->#?%%E1LK^_pYh9!o$
zYId?`=L^oVKi>o%G(_G&maqo~Jk2I>1;h@7&SN_?9o5-+7^2rb4bcrQnR%4+`YAm%
z>pn}=`G1m`Xzc&madtAD0E6JcQRg*w7y?T&50z~9q)H^A<5zJ^vwjxaek+w)uMQ6n
zRMQUCzP#!+hf)t-{WJeexl$>1-{&EyaGa=HsSFKO{gb6?rCdy_w|Bn@Ep7+DnYgrP
zKgcpupEv&gx}i*?)4F+6K=h~=o?dnwa^x`!!L|L?#Om<!a+^@x3E^vMy85O5LcOmx
zZ##ZGdDD@D5G--N&HhPo%-W`Y$xVIO_g+ilW#?N7dm36gdimNj-y2P1bdA~l!k7AW
zwH8-jS?Bk1gLmj`>KCl(!y1k}KN)?(F9sw|L%fR99q(<`dhb~_<DKS50Jf_b*E5v<
zJsYVggfVCWid2r{w^_uJ)F$jzG3RU6jS?3y7zC}ZbV<5X0#s}L3rQ5fb#IribI3JQ
zuOp)k+=ru?A&}6ap$yp5cpiiAVKyX)QXh^bV@_pktm0tDEfn0g+Q%{M3+W`hY7@zH
z9Hl|>u(83;mQUc|K|NS{Ta8Cz$>Cu%8p~4ECis5aXhYXFs^c<Y2}GZSC+TTWIS!az
z@St3Sqo&NbmZ^d9*>6AS5U6_ss~8Ue6{t6iz}g!u>?-@d56j0s+>0;l?V$*6FHc=<
zWNK>O1JFP_dTv!{uK>gKV9q-9&-QGc%)J9`A}SAOIK6n!rloNZ-gF!QE6B?XzW;d7
zJ-A4O46XuOjD_Z1fVR;{6Pil+7tN2<jLro(?}nkgOTw{+{*&z>sr`>{^U7Soaz|XF
zFbA*S@D$H{5PRk{B1O-u3!U>4lQA+f_d;G`bGNAmtuLMPLH=Ni&m?bfji}|`kd#*7
zT5L|s9_N*@(pral1$?aTg4fH>f)CT!n`lDwlH|A8ZO9qRyS;{eQH{J*&rbtR<X|vq
zD%hhUlZTLeeG3uM9{TmRn3Et{OqEJglvj+mG^E_1>>C8tqnYZW*o-!f=zaIyw-`2-
z4hI3*#|;`1)`ORzrN!7kxu$Et0-cEv1K(n{uJ`d&I~_~KPv_NwI3%>q<L7l#dRq%N
z=%E&U1!LROt<N%BW1nuFRNBi`3L@voNEbl<wPq1noXK_MVTII7YBi`?zs*gP!^37B
zCLZ+RyVt4nowuup@Q))E8qWq*>Q~015$Iq-If%sVSZ2en{5tz`wDUsM?EIx_nx>A(
zG7x?ZT9Sgyu?Tuj-#|~NPARhs%}@^uzb%}b=+JzOjU~})>6pOLMTd?b$t@x$*Blu>
z=;dly^Ymg12Zu*Y_YeZhte>&e#}ZI-Mial*nwn0$FgU(1OiWL;e(eSAHeM|-p7cBG
zA5TIjmnxTRlQoB%5RiC48M_`ZOZ*#(N-HKWUTIFm-=%(-&1V1UXg3{SaNNF{Z8Q!r
zaj~k^mR`}Zw!Cb7+80OSF($yh-4%KVGB<q$7-NDAiM#>6SQs@<o<|aZ6UoX=G#yW_
z*IbX5W;Aa^9*(4-C+7o7MAG+E2<Q$uD|X3u9n*6mbAoTsBS7i`)|$6qm*YZ2t6rUJ
zHkX!~&0Hc9jYbmPBN_TpvLrQOJA9e-JvIv7@ULLwPxgwIRlI(A7?eMfyc_eh0r0;&
zi3*=1!^;aNPcEE<7rvL{kqTb5K2xnKRjtkt7TSb`T_+Zx_4m1*uGT+?n2o+K)nETX
z<WD(_uWRd7*T!3*XlK5mRO79%jxNZz^NU5j;K?X&0xVYn%~t_p#7|L^-5pd3Vb36E
zU`jb*1Z~Z1uvV50bWSQR{5UB4de8FR+gmr5majW^9RkSlQk_TelE&_!w06(j-P`MY
zgirBtJcNhUBK)HJ1JVHd;jXV;6<JiTL;Y7T*U^rCU@fA5H02<Zo^U3*BU~TaHo0f_
zTLc(w?)Gu`cJCE?dvnNUJ2Du}=SMs5VIT3<zp+>5$_tzY%!e4D)mCqudV8HGW7Wag
z@rg3}%n@ie_ehipLw8$GC>kRU(Bp8w%;_<B!2Q~#i~~gmS&5k376qu@n1GED#tb|$
zFKV)lJuE^`)rLifU%~)B#TF)vj38>>g(Zz&J_)8rAca#v(vQ!T5|+Gp6X0lN-yubs
z#ZnUHa61-91*w!9>Ab}oUC5m%?1!>`sOcAvA(uxy8nKWo2=(M*rWL_q%thWGC_uY+
z#o-6E`!28uadnQ#75Km3l$gUQ2zlvzHj4^I$i(9UG0W_n1-}9lQ8wY&cEpM#QmHhP
z4J8uT#M7x1YCuO|!cS%pXBsvW8TKY(Ip|C}hWtc{1d|luzNCZGyzgfQVX#jR4zlF#
zpq|&?1&747x&%)HXx3Yom}^zW9^((d9oMc?zU++oz?@-@^vvR(=AyIrBzP-Vjl*DQ
zB_<cIJ*Pgy>R(dJ%l)q1t>cASV0(h|g13T`sbiVs<brXfE+CdpC*05K;$dt8Px~;e
z`?`p=OsmypV<y%*^GV7ARKY#W1=f_D8hJJ?EMF?Zy6YY9ct<pst%uDM9zp!n!pN}7
z4o>(1!0M7P*sz<8%r%>X9=%i{WJX(^s-s$_=U<O?Y%qpYXz0AihO%I=PRR*z!EsTs
z_PAHyoccOhWL$Qut(K`5n7>?m2pdbU^;eel4J6;o-ry(BUq~w}9H}|5_oRVMvfp@V
zCUH7}zswS8Sv1*sBWcsf@Dgfss>yIJXQdAN2K>XZQ#S(WZB%VMZD(@f$<F7k<n^)G
z^+_u|HPym{XrMJUmEM7)Pr*y~(qWj)0mygH!<LLR87s02RG|=Uk?d=G0k8_VYYq|z
z>*65&(&)mo&Z*~vX0}1HJ^pJ`lTfm`WOpwL(r$LcG^fO)Xn|~84GKw*_<JASdJi7W
z^<kSHC6^7^Iv=q1zFTcwBJire04quDnrto&PF2-c@_F`EfyX)c147aV*VN<YbOIUQ
zALgJKP=&9PJx+n6`G@_$5)xIHx^5q+)nR`~rc!Yi?g$XQ@dwa!<3x|)%+T<J4}05c
zeC%k{Igw1BaH2=Y#>saHbFoRHxFMS@RjxD*gDt3JB8B(>%ZdU&>&7FI5hx+y#=<85
z=H%Rb+KAkeOx_YP((`jCJ&xloJ)1x2>l4PC1rIO4?-S7-mx_m%mI;p#`^U*N`4kuH
z5x=-V0*VDr1{S_x7YEr#P4jTEWR$Dt)XKvTKa3ihiTKd)YBah!JOl?q<vtvZ-fI+!
zOLtxra1XQ&%{f$`+v)n$)Wk$794<{vOik6(_Ch#(c5-TZ;7b$zas|@m1>`iigW~P|
zW!CiMVi75@5XS>zMd$C4jtk|DFwkRT#xMNOGxPK01l6Oq)%jsis`$%9y)mojcXZ|9
z&Y#;$i`S6W|A|i*UhC07W*xZXv4?yp=2ncp7tBR`;foI*XENGLKJz1nKOKEx34{M9
zSnRo%rgFK|HLm;8T=kZ2W|)f(JA6{0v~t#f9B!%RUh2Bnu!)!E*1Fnn%9)J8#V_bR
zBZ#ElmNtiK?U6`AfTBVPZuSl+!31H-lYJU6lrAsJRuG~99%zSGO$dfU!rz{#<4y>C
z^)V~CT5k~ZKMK!-2bFp&@qeSfnzY91a9Fay;WBP`Z#?o1IMqDzo}}SU&m7k1|M1MT
zYb4+E$OxQj;9n4lc=?=(2bsHS4%YP7gg&O?3bj?>r^fZbH-xq<cz}SGWms(&fY_1$
z+A|c>agD#8a3?ymQi)tT$AkJdZ7cCvQJ8-*4)j>TsNlgC2+P5hW()iR6WlNXcL`uE
z)TJv6ko)zCDF?IUtynOxx?LeQ)T(P1*kgxr4ZQ%Ep;=+GxZh=?HOT_O(ynK`ChvpJ
z@Oir718uHqE(KrOAFp&i7`=q-Xt{4qAyaC3dTw|anM=p3^x{hUigbRSN{8<~G%=x!
ziaKB-S1A#!gZ$sP&$bh$`QGX2>2kq{fsr+gXe<K<?L;OP_3p&k;i1`cxEeMrI}%4a
zZ=hEMR3I3F5@pJaMd4cl<`$MdnDgL?Wmr}uZt}iF_cqszM{KlVhUK<jP`P}bHYVBD
zf)P}~{deY!@uT4H)$s6KK^1Mdr<I3&shFR~pum}!I8=exmI>b($1s31;Jgw`yKY2V
z@V<oDjVkgC(zPX>NN@xai8MTG<nFM=7=>SUP?IAZv7<KZJkfMGK?`IeoQB+~a1xgy
zaab)fnW!OmBH1hNjATsRo|(Y`?sOd7nXq|hObwRM{X|J`DWg%RxCI*W6=dsa1B*Q$
zu<zBL!YmV_G~1HzdB*`t8B*M1ame7X-ZJ1<$2bsoQ9>8c!SdGNRxB|M9@1a~s=L~E
z0+)A(5lh-u7)}q#SC!^OulIWY#M^8PfxD&opv`*xjW?s_%9#si7DgLL!FXpRosQgz
z^ofnph0dF=x#pU8pLt_+w6P#pqwC&P^%geG7RSd$F&Va|{4=wb+4;O~9b45*HKzaO
z5&W#HVzH4)eauzKkEJrBg`zsg(?8dMUaIdBY~;+dve7?|NL&ZL1nX!O>q+zYd9joG
zUN=npB)U0?$<VsF3`QKvrBV}#b{%Ozm}xr1Vo_pI+YX5~vK)96S`aEG#lPc0f7HBT
ze4T%M>7%ILv||*zWBlWfBFApOQS|7l`ZwTWX7jlt^m4($qVB0{QJ=V!xR)M`yZ4wg
z+SAr?48=p%sn_`|)CL^9pmQu7J<)WZ5i&%zY+Id8$D(2T7s<7~FZ(Dz9?8ChA9rU_
znDRC%Wgi1hy832ACjj<6PCDEE-i0na0d&0>3xp*igL+?x?mE8@S|t2PY+}uSUUoc-
z7g|7TT(6|Xe{0aT)1a0N_?sP&z!$7n;(UQAoCnki$Gfc8B9+28iCYC;NOVE4Wf`}A
zfK0?zs#ZHNfq}}Xh!d^U>ciFHp~CfrA$-<rm8cU@nIb}I4j`7*LK@?;VTGgWwFvcq
zu0pbWbbg*l7x;!mmBbYo31~N}?uS1S{9@zrWZ3EaY&2}46B_Td)mM=_a}m7j3+R`S
z6D=5D41MvePke1>+zX1~Y0YxTqAHejZ?(Njx`tl2IM>a~ZfY$JR|I-)B&<TgN{Dxd
zrT(h7-@E>Y2E7R|Bk<|P>=76~i+nG9I-ao8X{b=p3Yh%y-hS`8dT(=fw#hp!%T6pE
zIkM!tag>7x!JA$-p6z^F!;_lIj`Iuo#ml9^!4h8>WHT`~H8qyN4qv?*JP6+8*`M23
zUn5eLYLy<i!oLaM2C$$MPe(2)$XX`V+{E8bkr>9)wQ2EawaFZ<^y}-(w5s#0Mj8s<
zGhxFt--HU%zkYmX-coPZK7hEpm({iNtp}OWH8Q%G2v-cy;#aYHvzlJKLc{vDz<NdE
z9c&Ruq*fhb1@tF@xz02lJ?DjhQIm4H5GP9I>m0T=s=ssSI#_OcK>g$g8<^4!oYJPI
z&bD6He5SLB@J>D9dRx;^ay@DvQ}T|B(2Y>jtCyROY{&@4AwJq|13=moXiT6Q&<WP|
z)D7#R>eLJoYZ10L`SMD-!oBZhM`?p;<AvIt&7skr$+fpPu)4acd9BXBR1C5vy@)Se
zH55+ux<@rtTg=qjdzoOjHBVzy*i^9JE#M*gypc}`hG!9IblAMVCTZpn>V|0{go*YV
z9uV4Y$q_<L6k#05!BU^JtZt{+FEw0iuPVa72Yu$0fH1#GpY43PEKZ|q_ZTffo#lJh
z)UQ%{5dmoKN_U|1!|r&kD<qh|a*mCbJAaZIILAl6d+4|5{dR63b?CdCQ(JhV5azT9
zS>RS=B!361ec(;Cu8Y2WdNzn@EUl9kON2v#fWQ2)@rg%938MP;)fQv{j)ta0h!j~t
zlgOvWf4OUgEP08y@b@6nb>Mdyv5MOE1TG!7;;CLEY+FBNgw*)l4VHcUwnV}h?<TSC
z{8B1C-<%u(vmF^w1Cz~#bB9b5exswKhtnC9vYn|{a%!eHTvJF?4`=>!3%Gk+<toQ-
zyX`o<4I<EM*M>@!T&mjIzU_D==cmW!9qRl*p<=^f9464hfNzbx3Ca(*FlgKPltSqp
zHHeCE0|WUy7~&i2dB=^VMu!XFUd{CZXd2ZZZch~kMpH2~Q$fjCgqtYSu436$k;Ef4
znod=chIw99u<w6|-0kV{I9IGca|ShxyxIl6xP-OCNzlV<HOP=LmvYy^qq)B{D^o=<
z)nt{%bkv1b`#5W;kX}Q2iy{uY^CY*8nFB2{5slhNxdJ`_e|uJY@5eU1rQF*~w+_r0
zd_I+`lt<t@k5&o0)YX2`cq7K6&8ifhe^FtPvB_dr0UD?0X;GMSs3_r=>>m-R=pO(?
zPhbR3kL-^`J_K6z4JM@Nd>Hw*BM~?;T>I1WF-Ve4I2S9~n?5noR5v9JJCiOgRi}OZ
z0rUoy%J@*{yL>2yhY!Ch@5Bms%J#+CrKKy{k-X&=lkNz@gRn07j~B-p+&0MPAmSIv
zq%@7<6cDf2zz+Rv8`SFT6%EMLGN6|W&^!3vEkyq!!JuDFwSs_O2~Jhn!*Fy?J~yHd
zu~oBHCX1MgQZ{4FD(x|iZREL0$Zm(#*+fyFE5%=(gmig2=63J-VcAWk-0}~*ZMnE?
zhLf*`ejzEc!EZvl(2%%s<|1ZA7+_6_A4i|?et3VeAd3nJOI#0q)1Lvs9X#+ZK##<U
z;h9`fBq}0RP9br^sK^RX(Rc4wx#G}J@tuXjJN?h={PFH)P~7mRZTF7g-YuW?xIbds
z#RoS1+to(`%}Vo6_?NnRD9LZcZ|v@@`QP8^X6ltM7#tidKG1mBze>0!PTmGAv6kKe
za053(W6w$vK2JcjseKhmVZalVc87pEqTcSVd^)lqh7h7C;8-w(AO<fWKIrbUM_X-w
zAo{B31TjG08r}$AY45>H=<6Qu(H&*o_<HdE?O;nA!7J_HmCfK4j34>!o#VF`ucD=+
z3{0Uwl++>7E)A$$U2%5HN?4Vba92oJmhOP*i1EJns|DYDJNV{(S9)wuFzURo+kFpj
z@7t`$TXx~;^5Cf3mEhcO2ET0YdE5&240Gb_F((g)-nl0ad0F$a$LwjtK%b<$$6b5O
z@nz0lE!fAtG3mQf>l>|ge=+SluN%Q;+rfu7_r22h3~>nUCH-sQ^5%Vyc=2b18QQ^D
z-UxPPZ@5|C_x`??L>3A5?;`7aj~cbj-RpbLx%a(r^-FE{z5bpP>AUpmJ2!)!p&XHZ
z`tQ}_>V2YnpC?6xz)8IhIFwc<cn|<#_=I?vslFq>Q>lz{h}uKb8`+-sFz`npF%)1e
zaJ13?S^+tE993y{j%2{Au1+%IY3{7uLnu^JCjs=p_I&5O9*Oo2PxOuMD(HI|wXwb9
z-L%ZvhOTh{)(Y}R;EqH{r$VPE2s1Q1gCKK=B3<7<lC>5*j`q^uYPO-E+KFW9Gf3x9
z^WJm2?I2l!gCA}YQm~E}@_JJva$i@QP)nP$Nav90d|1D2O>J!lVPnIe2b|P+v&}WB
zQ0wUPkX*J<K{GUNja%CPLpT|b@jzmD?2{pdA%a{EEG!hlY_Ho4C|ZSfLY0sh18El}
zP12RupF|LL3`+2u!e+?{hvPNupLfi(I$+PX+q3tM<?>IU!kYaZaLr5xbXZLy83h6>
z9HVR)B`BR`zhPLl@%F5JK&b=wj?K?6w|)nX1aZDUSXGZteLnKgg}^ydwJn@zfiFj0
ziqA^WF9LikIzq|{o)0h%aTh=65cn7T0`gg3&3{5;eKK|Ww?Juxi$k5S6^r5L<Y8bP
zzHn~Yo2X>$6;Z4U*iUQM!mlWCtz{Q0m0Z>|+^{u%hrUaT{E+5Pox0gB7KVL34Y((t
zhrE?e6&l%F&UrBRN)$6IjtpN-pZ3#M*fq>-4uKH1HLh>q^xt~bRR<14h9H)AQ<{70
zhx7Ve^0}T?L_5Av55V41?)TVL$hVkI(UUXgNdJP*cL)N$5~mOt7k|}Bl0vPIBHv4u
zuz=(y7gK~H3%N*CUaP@_DSboE526H@?kDQ;z_jff=^;vWm!b$SqG<S_jXH2`REbL_
zZ^#WK%e4>q;gdCp02jRB60M-xNMnsqB9o2A%LZkc&CnM@-xRLZ!JNyCmx@pm6fq+H
z&jNbA2yoV&0yv>t6aGcqx6p!Q<G};;oVY5m<QWM7IRbroZxlI(S|3imxX4c4vF8a6
zc2^b@*kVAfNLCZ0bY$V%0TS^|bk`!-QJrBy7om@EJSuRGy9k{R<*bR?{8Bv~MZ`AE
zTwyzEO<GZ#^$YNWYCW0S(tK^g%7v|wLd?mJB;9I0XCHvw3(iId=0<9!nowqKWbS}C
zF_{PKT)yfiM<_6kSm746GzaoYSHW*>U~GH<apGIp8y_18BX(W6NnC|KN1a#MUy3UO
z+EDpoBbu}jpq+&IkjK!KBBJV-vbe#?7h-S?N>#oD*L%~MO3U{S56u;eb3-s);<eIK
zj%mi@U#g_iRx(z|*$LyHXJ%&jJ*c)g2WLM18PhoD?W?&Q!#41)K7Ih7`BCINSjNmx
zkYncTN$M($XE$5{4sTwj_@0Lp=}P^euO3!yS|>YOerVsn{EK>-x28ArCki3l?FS}q
zL<@!Jh98W0_uZW*f-p+lWg=q+ZC2D3+<>f*p8roEE0dT{ytOKJDk!rEao&H6n2H!~
zg;WQ0knTXf_*np@S%kidFsFSur8j^mz#E{AM9xnQwpqIp88UEU6R8%@Sdq7<!X+a+
zfbxM)ZlY%@1nY=6J^yPo8lG6`e8=l<c_4?9Y{{6PM-->qOp(#%LzO!nvt_5xWsT%=
z8*O*C#b<isK0Xm1gGT{X__?qcRzOp|dO@W3*LAf>Tuodb8eeIQ1jj`5NidT|{Wz9}
zuf)V+S%%$7=ONM~fz=r@kO$=0%$YMY$4*SbGTwKjaxi6<;n_vL1s-1IGk5Gf@uz%B
zO`kb~>ScI#SPrMIxn@cqrf|Q0cv!cB<1=M8CdXye4itR+*MftN(Qc$8f;fLItYN^T
zKWprj?obdzQv&p{HM;0)B-AK^;sy>m&T|~t!|70SXrNLlA84`AovKt15cifUNSB3A
zR4?~Q%`v#aD)qW3oK&9@_Zqm;BJn<|?yb%os630e&Dm;-pUY0{OF=z=(A(a6(uv3U
z&Rj_#j7SK_)rB*#UZD0mcps5qdq#n}dY8YTI$K;&s@+>qo%ec><t-?r5xVpR1sNLy
z4GRPEem)=3Iqv|z_azLD9*2#9;^#TSdwSzuS9O+T(k6APdPy9)tS+`kp?rl&3w-T`
zc0wfYz$BoTdbTZWXB2waLR8`Hy*(24PTL6m$ClClk1ebf$vn1Bm1A#155o=@vK}>T
z_dft`LX%G1cB@s_j-Nj5?>T&d&NkO<jYJa$#+_SyLwgG;^z_&GA3x$a5QB1+*pVlo
z0qy0>BE)kxs8eSw58`-$mozS;UM;xiuL(r`D6P;Zzd}b(y-GYw^s;Buz_cNBBZVYN
zYux@KJUy@g-0a~|%UN4v@4T(=67zX=GMnd0m{)3M`dJHv{4-~`T^K<?#upRmT&lCc
z?IV+mlRRHe*9P3#0Sh4<UL=L@g0lMV&O2?c#pt+cPF}gN5Nr4gs1bG>08#4F{!h&G
z$3u1Gusi{+xz5IeC^~e9%ZKbJwiyzBfGmFH(-P{1vC;UD7myB_3@<!jTQ(OP`~{bF
z+F#z)1U)DS_YhLmQ_nhe>e#VkN3(ci9N~~hITuG#4g$Qh__y+Nn=sLC6JmUe9j0G{
z>W!U0R4bi798MG?k!%K*gK!KUH!I$8B$^6GW4Z77L)-miQ%_2>n_f##O8Z(5<B!J-
zqDN0k_PK|!c7NoE<lw2bB0$gtV}vpt7!uEh1zQ-Hb#iaW0D;pS9Kw-5deG~3&1GG{
z(H!hW2W!K7dp$aORq2swcQYgVoqklMBvQUyX1sjZ$fi*S{*uP8^Jg+RTi%xmL)_ky
z`B?Wo%)6iI@B-6?@n5P-1ID0h@o-@!)C81W-Nl3%8Loa^Ex?4xk&LPoTcY(;+dNsW
zj{~JI5+Q0Jz-+25!Cctuv=@-6p6g0BNQ<S`dDP)y8Cf^EVq~L-YiCunO)et-pVuEq
zJT1vTWwX04Z@a9%P}b3J-_%?+7ZXiNCE8;aVYlc}20{q&64`yg3YXzBe&MFFD*1Vv
zkQICw(!^oU(RdOy!q^kpU^M)LbMaWlw6hsEiF|Kx4a;Lbi>6VECpdL!_}yi~4(1Vw
zp3=#Pjg)W5^OV6I<(S;D_vmfx8odv=E(Aq<k%=5`VatS5pB0M_*OCsCxEa|Lvc4%N
zn*n<sLDn}1Oe(WG#fNGM+mt&X#HXMBYv_eO%4z~yss<-Xep74fDa-U936c^iNb{mv
zC)4scQ~wC7ucYpe8>TaTQd5)r+P|iEzIiGFUy9(Rm$W|-yre$F8%}6)x2`|#=KcGe
z<`Mu;03Mri@B(bfXOSlmelFvA%rCM%!0JHCA)c}&5Pt@<IY|Bm_|SMl<5_SuyA|y#
zyLC-_sN(@SI~|<xGKKu$-6ANNsE%O3Xk}Zw(_Z$|bv+JxyBQ?->%l8q5U&OhAO+dS
zvQXJg<%?xNR$n$v$un6CKIILiH?I3$>C<mcf@=A^)jk-7i$;5Y6m+)Z9!I1(C9Q?v
z`1e*(cdxUx0cUjmGT4zQ3LHUiaK@*!(3lYUbkUGjYi;X(v?VhG_S;On<RfIjwXZMg
zWx@2dA)4Xy0~@uEfR^3Koq$h)@1TD#-`c-+aHZ?>7UIxLJ|A%=lfB-lwf<x8@VtNm
zo<nj0O3!`^+)1ZR<E4FP*8LM&pZ*2lgx?2#rv%B2i>q2*>f3q+Xuy#}ZtB}I8}(`n
z=us;>K>d1qSQxFq1r%<ZoJ^<ne?q_Kt}{0i>ut#fXix8WHDmRGdcux18i|Ddk6I4c
zs+1Ev7qgv|dNSqwypPqM^l8F*{|wi4MAkKB=Sk`P_2~VJ#Z$zu_vUN4GP1%w`we=l
zIZNUrGtj-|efEh8Fx8qbE8%IJz}y2@kTQ~iAg&-SSZ&ocFXjUdI5Ek}@UX9_>OAT1
zAtb{UT32J|YU{Gyx%;-DKlBHJj>DY`2n6l|iy@wh8d`L%A1c{8Px)V_(%d7{4r$6G
zv^zgHg3V6`zxd2vfDDp=Y{9>${#|{EdSj`yN!_aWsdv2_x(WIr!HblOJA!-ELf||k
zCG}}WJ)CphWt6**Mi-{>5sU5kSoZGe{N9}FGO<`{8FlSZ+<tnmk7e(k&hN>Z*EC?A
z9-CyCmzNn#(3QG5Wt>H>^4)tor4ioepX2On8j=wVv9K62++=ROSE+4*W8Lm<g5rnX
z?=J5`7J4gx_3v?CUposvuIOw5C9sd7HgrjC(Vb{O+*yilchC0fZSNd$TtP5e55#&M
z*}2E}>7uet9_{Wz@C<vpspke!k79qwt&&<o@*_vawWSy3kJ{_u;*7lQMZa9WR>tmv
zKY}}ENIKRXUTBRS57&GR9!@8bf$A<)4SZwh-B<(@;EDX%=R5lpM8Ch*%Zd;O7-?!}
zLJlv%hbRM*3j5kU#Y$0u!#(bR-e?uygB}C!fSxG;^J@r%WL(wF+3edM^gH$>=_8r+
zC;c-IzU?33)YU~d>H@Hjchot*@0ITdYk1z?eL4+3|BJEM^EjlPEq=wiFA=QsoL&RZ
z`=S?@-u5T~K5^#!o@Y0M#Yr(Q@^bPV;G4RqEyD)AtH9e+k3Mw(%$8Ox(XVQ$1zjOL
zj2Kk6!=Z;JAN6Oc%)QaGNI7SpT}?P^@#JLZSI?h653MD2lIM%X^2kV8eLfjqa}uj(
zkxe*$HhOO+HQD(t?&2-&X!JbqDHicQv}bGmZ4K{pGkmChe=o|KS~r0=qxS}Wi76&V
z5w=8mT`M;TxH;hv1@847Jc-{?|M~HqPpHlM_&5USMl0ofzEU0?o|+mNG1S=T^1(&8
z&0@-HHO|B4;=$$7F=dP(-_I~Kh53BBGKvHf<Ky+<ZBXXHU)%v75se|KJ|fTy`N_$l
z!E{Pyb3=Hn^;CLr2z9;-xKZD>p`YsA<NKs?Et4C$qpRhVel>N!I8zIYdEiR!>GeX8
z5AJEq%sj|_;EsWyeGaW*G}Y#TDx4LCne}*Q=sO8@fEo|q${Ox$1#=&%P%x9x;>hgm
zYA_Sh%G@lx;0VbJ1X5<nz?QAH)qe^;*_TiWpddOOO=D8<sGk>)`t#t&pp0hnMR{$=
z){Kdegp^GEqUn$5y;>X%+i5x)0}A5|c#$!q{uI9w7Rw>yfd(hdd^i$k;o*W8-gx7U
z>*p3uiUxS}@35qAS*{`DG&}OvN4)cmg_mD_<j7c~F?Qt0)xEtmZbDbRHP_vphB2+k
z=O0w_Z2hO;S`Jn5$%X%f)@wCc$B&G9=jON6e*T%8DkW%0SP*}28;wf~vc4Uxr^A4s
zQL(C1x6j-nK(b`GsMKfA^}j5tCbT+YT-Qonn%OD1Q>n;N@(JJy#@|3+k3`a)|D28-
zDWuXUs4!J2Pfb=TllY%PK?T@d7b4MYrn8k{kp!D?uO$lX-;UA=$>j6K$H!rv9z>a9
z|389L?^o7O|1>0O_13nJ|JrX=O5`PtA1bJ*a224TUiXRw(tuOPTcKy79%lkie@STY
z*USj9`e+?$Fy9Ld{srU4>VN!!v6_y>?d#oY=Ygtwy&aFG2RLEd99O7ng%1SB*H$51
zibwCyLl(S08plH@`m(tu!;kydaNt?&8Jq8tjo^C|*+P$A49!X<3+=cE9lj_1YkQLC
zkF)c%r9B?{Y)*JC#O61&KqvKGfJ^&Ns12NvOa2wrC*5x9hxNAaAh`aj@ZfrH=)<9p
zhd$Z0B9YT1KxoVgOBsI0CouY$2Qc$eE`E<a?I6IbyZ4IXz#C*|dn_)51ZkVZsJEwM
z^YEP|uMknQ)!o50;f+K4uAak*c%wnCxITLbPT|qRb8^&ZkOgkc9fm(d;?QhejvDbq
zG}@R&0j6m5&}{I&*+aN5et5PKjecTQeqi|HkM~j3M|(O&Z^a<G^?e&!k+-^pZ`<qb
zxAd3wzkzq;!J$`=Bnd6l16V&;B_tG@<#wR)<y!ICMXIAHPLN&HE(>&0+VhBP!t$~y
zDr0|TEsObwN(&-rLW-6dAKz5Y1&4)S(&Y=x#*Fgl2VH_H0MY-znf%~}f8<>g%DEvZ
z;oeF7Fx;7+Y=o^%@~Lw*4bOf$)EYt}4#HLn#cvSYyO;$7qa@amc?Xy&jc`qrsE0y^
z0CW-{E4~D}C%%-5vw(a6J?@qv_-4Y2R?78yxe^N}l8u2k!7dm*6ip;!2l@6}eEDT0
zxWzdH_`Y$Vkp#Pt7{*y#9!{v@eNh)&5uS?7{J=+kKbT>Xaj;Zw^gi-Qz$Z^Xp`IoA
zi_dxyg{yis;%&g{poCX}FRWjh1RgIZFUVy}`iU-LLTtV|{^{1k^ZFdmuuBsB%F|v`
zp3fCRb^=amb)e};6_!sg7S+ftM@B{)0|RcVz(VB6(vuiSJJI}7#4MSPorI}0Ts&Qj
zqz_jr6*D~PhMnW&1f$Vo{pp3@kC2H(HIlS!iJ^_Gl%Yz^jf}_GB%5|4>C&myXyTkC
zG(^+4Kf79QfBnRX6XEde7&2GrV6LOW=WYZ4ZlaplG0d<nwu_*ajZ1(8gi>h0vTE)2
zrtZQ`&!?tyf#9b+FF5D~)`NhyQ>nHsvkIE6f%+-^p|_Y@%>$F?tHQrxjlM>JE%B(@
z1Ls;)pLlGIwsh&6TE6q+3MN$$t47PBcv*&C)yJe759ftuk^|TuU)~anCQwZVwPX@t
zSdooBRVbuVK$0IrZ{w~VMmA#GOpXAqk=kG@I_!pRcO7XTk;Vxy4gZBbk>I=H38aZ+
zf+-h`#*i*B!tEn?20`Ru*R^YA(zRoP9k8@xC_-x1xIrwQ9_PUP7GwqQX(i9}&eD&N
ztDEZ_oU(X>Q?bejwv2~9{Nh=D*{pWkS9$t)1U0!ggOlg$m%BALJ?zD}yHE(hu1{}a
zt&<l!2+jGm$n*I!;NVAs{?VK~hOmmX;`0Whbxv_cXQ_ZM%L?9-h`41pZJB*-r9*$X
zvqIY?awv#`UHtN}%r(cW<)#%aHbdHF5~miFlB*>m4?f31L5-(@Ddl5NJ;LY#b;NCS
zSDX=}(a%{1G8Nb{<D?n0t?9Tq5*-_hj+pUh_mZsj_WlQwR+w>I$4_)DZXWJE9R)ME
z&c@MpXA57)FTsbrMZ0_EQLuxSoh*)y7L#^taAKzP%cYr#!JB*8;d*=NiNVrLZKgCh
zaeeS9tgWYi`1J3ppHW{6-HEY(SLnY17pxNPEqhhPxDUiBkt;-=&`)cBAMB4dHRDW_
zIqHb#BM~J+E_9M?qF3mxwf7lmL5b3H33KQ8>P`z7bqQ_q`xf}wjr~FUK++UuR?*Hu
zpkqqj2hoQ1VB_~sG=Br>L!$9QF&0CPkgQ=R;)uC5a=EA#ixrCTC@y!e=W>RPtZQ-q
zda;1(cL(!6bpBxmE?&{XR52TgjKk(^)el8&%ltl>@<_Hg1^0;=)3T$7>Q)rxjlQt6
zwZOMs(%XC?m{j8j_uO-j%8j9hkYzrXUa5-_3c$xuywbGb>o%5C_fU$CEJ8pAUWBd!
z8~UOQ5QoF7nu3WZ%mBVQs-6wj;+$dzglY#yP{<$&8=3(v?)-5z_x@<~{kf`ICQk1B
zX(ls2Pp{eedAi&7_O;ZKdmqWv`=sQ#3gMQWx9Laln#AGT&>e0<tNs?@o<mAlBk)d5
zO?=2jX}ryqr<X4ESXiu{qC+dYG-1*!Q){O5O5N{1q6+eFT*=&Ln_e5Hr(a*8%cIE5
zYI0#ALdoU9?4-|YOlGf`9UKT-zCa_wJ0&608XBB+tL>^gJ2+%XaSopDUJqLXg9n}!
zZ0JF`{teoi)F%E9aqnK93^%%Q=0;<9e(rEzOVZ?Wvk4_T6z|QZ)I#olkL(<tn;#Cg
z5$5Z5ewAN<9|0A?w_gXoO);Y~@SN?SUIm6^1w&tG(V++<&#ZhjOAAjV{1BC2xEY2_
ziZvWZX1ep;bC(506G3kpqt*0Zr>mnzdn1x@n+{>7T;|r9K{lq5^x$^}(@5-FDo=Wz
zx#`C>I{n^kF*7ibDb8JZH-92_4387=%4KdH1XUiJFBImFEl*EZ?Z~@>REB*c|F;GG
z(iS*`n^BVqUYLAcVl41z#7{mTQE-=Pay}6B?lQIU9{LY0C$p!pum}wRiNV-V$Fkb`
zsA1<2-)`ErSupGv2Rf!?YVY$;f0WBf))b*gIHb<T7MUVpjJ{yo2l>ZIejLcF_&6Jp
zbDh71!lV08x0%`n{V39I3O^)t^H4WNt|%(M0%ET<rQ}xI$YpNjtbfdj%b@q{9=(7m
zooEhEi!_>+YNcUcG!YaFJ1x!I>nBgrsS|(X61Kcg>>4Leu5aSdkC~Mak+hy|z1<NT
zu|=n#w2(KC;qWPLjm@+rFEn##=%>ApP9L|38@t~Cy+3m&{{vv+9I&I7&WnzKFTh(E
zd#nH;5Bkvo(0#l}d4_93Q=pdi-evrU)~^C*;%WJ%jDz20R9|8BOJpX><VJ>vid5}{
zQ5@CH&XgmO%IruZVmOi55jtrdi6QbJ(ioYA&t!Qfi{MgJ!9x<y;?VF&E`tn5tf?7M
zTTXUxs4;mU8Oz34CJ!pp;h|c&7<Wl|Qt^mck2$tAbeeuIr-#B|JnCj7p0c7yEgmnH
zYeU0Oq{8JQjs}tkCL2S8Sx0Dw7aujm`q652M{WpmQJQj)QeW(<QnzVg;_ZjEtg^<B
z&2@e9@5BQa!Yfb0_TnAB<A}bz7L9JorD*i;U!qTK&<7a)eE$ypTpM%N07vNiuo6e%
z&49Q$-W+@4F7RuhaM0gMA_gpVs(5*66pR7shYmPUkNd{_fnp)Kk{%mdU>0fSoSv?a
zW{aJh?X)_dw(a<*+8oT=Z|AoCskr^5&)*`@n~9S7h$NuSk45i-=L783pL)=DynFCd
z2~_aZG?{v>0r*5wt|EyF%K93iK)oQQVbW*N?7!ewrq3QaWT-9U(4n(w9#4LUTlFNj
zodwn=E*katiQ~_G?(q}xI*&TtFk9lla47NgAE|#*pTRdDfp&A9SdbWy|KtYzfACSO
z{3l-2|9hl_zqCl&e{Bq3JDeU%*{)qVR%zU_<FMIXjaZrO;Og8CUQwU4(v#_t>qi<l
zH_As#R>C{1-_j@_DO-s*1)tmsKH1l1-x&D5I38bg(QzQ()cbwLgMnmU(H#Te9~zP7
z;9$&5Ob-n?i~Dw<e^7oIsde@qmZ)J5R3{Rqg9;Q6T=Ljx+g1o`Yn&F`zFKmmm0C^A
z{2g3jTt3M7r6hNP$b*Dj>aMiIO-9B{M-ODMc!_vxx@R>~t?g~ZeeHF2)4NGu)P8zx
zweCxmHGSTTr|WEroQQ?Bfi+c;%9l65$Lh||y|9qJPF5H#VL{D>ezQ8w9~2lRp+=V&
z(kTNppyz!EBfW<;&>`4=G>I<}%?OIw|4di!VT62Eu@kDnFHd*S`+)HRgb0NHX>Qw|
z|Kes!b)NLu0Wl|$yszzlW^cbdf`40o9Y3D(pZ7o4e5N7opGoh32FpkD7SIb(*V2Ul
z)qQYpTh+Q2&O3gIc|Ga51j(9Y6sLifo2$St7n3oTf7A_o*IWtO#6A>{kB?JBJvK%y
zac_Ssco5ugt9@)m+d-S%hjzCZY-IO7$BI1X6EarU>Pprs@6ZefF$_OonzeR8a|;|b
zB!Ka-F*q>#r%BWmhHe$>tysd5D4V&TgB?%w%wSD4iZ*vMUkKA(89R9J%G%iK?Cc0M
zs*-$T^2m{^Ru&FL5{DvA#5tBnIYt8o#-MVFUfr#IR9KumfK=g;Ly5@bYcv4qty(W;
zpVv%1_rX}H*kmpbHoG)5sIqq_97JPT(N;2kZzj(MrQztqkc*+LAg%0tD__4po@^mk
zXL!I#T&-Ant=l@sj5uf;_VtfrB_p4z(#8uK&b5$7u9O-m*7VhTWWjrTp!p6h@6hal
z_+y}oj%MTpeYb=m!>U*RoQb))`wpIX`I$2)l1N&)bb{LxOWa;%n%7_X6{x>?*DH(J
zLFBJN%9D}fFM82&<P3$nE;4xR-hr}uz|Cc1S4FD>M>iKn?ja$5$TcOC#=WNd5RdN}
zS=c;YoxOT`dYboMW1%c%Jfe`%hC`By#Z&|*O5_VH>3bV8Fl}DPpvk-oBT)1>cfbd=
z$>Owu$zK&=eu4la3>g`p)h3o2XqAi>p{ps2v4qjUFP0vbC_FX_JmURC)bL2Cz3%3l
zO!KCjU*~^51=V6>aAGi?jm_3zx>3K0gf7KmI%5yjYD1ry800zLQ=5%t<AbOr4SX&r
zZn$0_hV~XqYU_Bt(WoEaI@YMGjY!f()!3PAVIUumB$@O0hI}HKOyo`TF-LMB^NwUB
zo*yV=XCk(QgRoJ@Hcjy9j1N}o%c0ZY_`31I=!Nu|mY8l#FvftQT;P~z06gBWz{M({
zA{ZrnT|9|Qkp~%zgvDUr4R>TH(8JLt!-gHRh`>PfHa&Cv#EFyhN3d4x#IFQHhOcwN
zF)Lhm1WFuZ$Uve`tQ`p<Hl0IA6rRR_Ar0x^z})fU$EK6fU-n~%Ip<m~9SV&hlYU!l
zOt!O|X8F<?JpO2b@?3ey^OBh@h}({TL>^a8Q#K_tR|M+SxgmX8riETD08ape1pHrx
zT5VuiZf<RCJdRaB+MqXq{nhwrgcl--u<PVdBQb#lP2p9!Hx`ak_=S=!j#}XmMKP6~
z4W9@ql5h%Y8=!WY?sRuLHOXkCVoNmdG-IkLg_WZj+s;H)>`>jqtGzL6??dBvjTeT^
zr#?1OCa@|iQgU#*lv;BYe%~iYs9QLb$(*5zVuYAs|Gs?Z#V>yG2=Bwqc*J!_Ud;9H
zzprr`G_#5M1q=guYacT|+>5HM-I72?Yry<hEs{kQh1}?hgq)N@9M??_=CAUrewBw_
zT*JmIntYcv*InbgG+)8t*7g_nPGp7i-3i&~JOv$4lXv2FJe0{apWN8!zB}pt%0BO&
z!JGSbN^*ddC5i+(jxi8%34McO0kiXo)HSZ(eWUyCy|@02{&)TjjYt(=J@D>c@H+Z>
zzW<H4x6uhs(x$w6$l0~F?g+gAT;X2UEXkQ5RQHnB*S@SHJ1U$pxl55XrPuoa*uIM^
zy|uNa>7xI!bs@MA_Xor7O6u;c_w#@?wA0@77apcK)Q^J}5e@d?OC6&iZ(Il-lr)hB
z?bQaA0%?%&Jq&fnK7qzpq>yNdh$W1U&YXv!7|<zCoB$}vbs#!T?)MqbEgw(yFjh)!
zq*OBQnwFbMxkv;NiRZG}0<!<2WGb*v#x+fv-4H#VOQ&E;i$PnR$(txaj_do6bC(qt
z#K`YEqf!~ev{LvgJB+0MNbMPm4-6sTCBw^E6B+GjgoE|hK^=jpgF93>id&Ec54U)F
zyA7;0(t_thg1H<>)sZ6?<*-p>y60U?%vYjvK}58D2J{1nBCa5q13^PG4Hs1nYqdn&
zO*I-R?AK30*{p-U&M>6s;KBWe>-SKf57`4#&g2LPg*!9^h9F+6aew%jp$_o*_FC3h
zV|$anoYBTu-0Az&?`ea!i4+kpLu?Ox19k?-Q19sY?EQ(LZ{DX2OMNFkJ-TZ7>DXi_
z$uR*^yGRX)Z8xn9hmToEc+l=AdwbHysMM>u)2A?MV!xs_JBJcc%7CIC9o$VPrIO9_
z5BRk8UHcZa>FL4=XD2QZ%Ak%S^NEoV8bspB)r}OL`@qg5QE@*7-Px%gC(sdGzmtGc
z9xzaM_j`bYrg(wr$<Zt?UZtmt)Xc310dRO^0KUc^bZXLX>%a`W+G^dV^^rH98yb4e
z@WjOM6NSS2nXhi;rkj9f^u?tW+kU@gO?SR**yq&ev<mX3bG-dELqkvS_V)wV-~8W}
zmX`Fz#TCnXKYaKi4z$UT{ZI?sApA}u?hzR&bE`=2LC+Yx3<MAgZ(=9a)(2>h#jJu)
zS~Sr!c>ee_R9sD{gT}p{5s!F~&G?vfif?_>iuYPic&~Zx=4kYBbm<(j>YtpJ*IV)4
zRy8&8&!m_ZBbCv~gxS%XL9Assp74g}+)Zr!uhG$SP6R%S7?u@plsf;5#Jf<dl^v&C
zs}<O$_JjL9k$XoF@kX2dhXQ_Z^3r6=K^QbsPH7Hk9nNqd65@hLB#(thp^wajPGT*3
z8u7rSbqS9*0g!Op!A<>_Ff@K%^j#NP9I^hhY^g-O1y6Su`pU|JarEfz3zJhrX%yEQ
zU0OP-eSC!r{6kb`fHwsI-1^~AC1IG6L^hpa^mbS|b^_T=jK!tv?t@>3Gp3;1cM=&3
zylgyvA4Or;K!Jl4sv)?kVlm|+{?2w{nXp8NNvgz}ZRd(ndu;3WqeqQ=u8~Tnvs2}=
zb~T<PuQ#~@^HQ-;)CU9A*GFMbh%ws<qP{B=;|WNqxvA;Ob&E^JRUmqECs1t`UMMO(
zGgFzWWz%uW<CWEeaL-fj(c=}q2j8c%x{8_uDt+Wwg|E8KFRMK7))ZbRm;R|2Q~=w?
z(?1BMjVsl8iD&e=Sq`lHa?<3N11kcmAPk?4MaaSKNXE|&HJe~So5MqS<b-wNh5XR4
zJ~LFw-ciV=Vsmq-m{zzWTiH>`d8wMsD3!@pD+5(H*A%J)74J-?P+EzlvW2xZKEG1R
z-`~|V07s%9y&S*PWgSEm(Y2vlF6ujk4*HP}P~#BJMBjze<LD|njRC;QgvwBB1(*Gu
z5JlqSS6p$N;&5>P@m|KE^i(0AFHGqzf?`L<#zv8$m)p)ZL5=@g-JPD#bZ-x=j%Ms`
z4?cY+*syNKZ$;!2u@`zey;*Si(7Rkjr>Bbqd+3N9J=``{DbI*9Rbwg4x{amjYg;fX
zI2K()sQ0qHy)`{Cba<j86ruovwUS?lgkxs1QmNG{m86LTWR1zmMmso&efuyS^;Al=
z;FN#T1V=UqN9N363>7<4OqOfFzh%p?gucKC4jzLN3+1t{zS|=sR4e#-aqC3Vb<o1m
zS|RGV-)`%#Szp(m?QgqIw9sk)95`(?(f&BdUM`pZ)^7&i>^HxaMxmz-GUIp+(R*#l
z9NZ;Y@JYX@nbx8d$pv1+qhsKmNrk)Yj-X6JreHRqd$`k6AnGLUTG$0_q9ra$ut_p!
zeME%xg%>}K>L~CZjJYTqh!Teg&9;Y-%qTq;kJniLs+p!)x+d;;=~xA&19K`;w{29h
z%E#^GKyG#Qn(47m_svQW{ND^O7$-VqBicKI@Jq&in=lY%EGS@r3WpPUxVIt*HMM}@
zOg1uwgkvZhnxi^u4qy(mal^V|dRnEv+ykNBo@eWqv1F}2<ZvKOEpiM+ap;W_)+t6w
z{)O0x3xF_@J^ls@oM=lE;)akb<i8+wCX}$r)fFsu3iCl~><=W2nbBnB$bhIc;HO*x
zB`8Nq$_=~unM?{f@U0X|NS0hEi;y@t%@UE;>6Mj&dLJ_9Fhfn+&I}o_?FEM~u>Ukd
zFiTD*Iy4D?q-+z#9*t-vK9IEBK{%;|-AF#{q@$?3<Amdhc)F0!W%6M)IoUpWXf|e{
z7{gK~jO@p~!Rze-CVTjC1=7fCLq8LGj5*L{GzAv*n4<48N=%$0<Ef9l(bYsgU$Q0~
z(&|3I<&ig%MQIS@*@C@xzwv}36Qi!b4uQdXZ2<2&E&O-Y4ybfLl8mO|W{KdbEJ_NQ
zHn2v*EF|KQXgU%Oqv)lTu8)znhSmF#>A*zcAtRQ_#%v=zhBSG$ZKvFnnReCTOlM`k
z!GE{uq6~(CP57MYf{!xXR5<dIWF!rbT?d<AKszSB5yjrJK+GuqnMVdal|c=l0;U5R
z#`sj6Oh!J($c?ZO1x|{YMjpW6!hsXwp&e6m*Bt20N^kF+|AnTT_uc~U^DR4BbK%xo
zOCmTqPW^|6*C(J`JPHn)wz0dwk-Z6Xc`+<cPe*EVFh8UIJ~_8h?|h-)?=YhUG2^n<
z@nENkOI#4JYx6E573?-;Ui*kdf97S}Ku?KY7khi_s2$QWlgaFStJR8RlS%V55hs7v
zWP{VZ%o-!S+A8;xkS&dueFC#D@i;%3NyHCL#uJ&z$wQNq>-3bzCj!oKe)15{drj#E
zsmRoO*!J1R7BSWrr=!5dh{EeJN;oDP{NoJMqX=9N=O!l)E+b41@>=xx=`+Xv>6s11
z*P+0T&1$>NYz%-*3E{qOXVLNg&I4Zr`G{xUe~H|2K!?Pm2CPE?%sF_8?jfCV_6CzZ
z`~R8X9Bl+UurC<w+4tA_+EV|sP1(mk(2~#2F}Rob@*=#{elP%;Z4oTUK`+&dFq!Bz
zs!Cv~=^yY@U1K0^fH=$ik(d3@;^JkD<tLcJ(MDz$_^!hs14S1Kw3f>$)N8fP%^eqk
zzPyX@kiERIz7?JlVVij<@I3}(D#IQu6i`kM^(5r*Vpp|36L{~duu&g{742t3PlON^
zD~RY|>)^@)*mC@0xoMqQQ)?r6xYQ~Zfb#H5Mo73@+zPG`2cT5EMV`eGM;r^d)hqNz
zW=&3FCOCC+XN%7;-gE(aVf@JE@*h<H9T{T46j{oNAz7`RK*}uG5dZtLj~lR-BF|sl
zO&0n>Q6((((PL*$A4fh|6XCSW2PY?UVI;cvgxNPW{nNI2v5V?iNGy<#Ak+qo2arOl
zR0d{Dgx{zpQFAEUKv^$qeszyLGG8b5*ojeYy}WtthlySI=Gmc_JfpQ%6g2GdC1Cvs
z8KR}(Cm2UCf1E*pgm;hJf%Ln3QmQ>s(C>gGn;F|Z<=PITe%v&}bKDPs14u)U8iu)%
z*N~UVagkjpe0huIx7pSy7kQiT$#1aI{Jt`#D2(Z*);63mafkO!JMi>%p%*@*aT5te
z6efbIef1TRn(`$0G>V911x2@oF)Uohh;`u89*H~`Dsppd3=rukZ^ynP61fA#VJ>fo
zJ`TeWa(sptOk{Yq%o+ITBcrPcDaYSkKy7w?8GP^4-+%hIjfnb2Xb=&zw})OCdRyqD
zz#v+K=Xrxd=M0Pj6nt<6dAX#3E)fKQqot|3O@@Xl0*z1P3KCDPBA!7|4Dpc=H7LHc
z8VjeP(9~_LHW8*z7i)cjBFF?G5MP3j2>B|UWAlizKn#U^jry;td>$1@64fdk!4Uw5
zlJItX)-W8Te@+(i!%k!>7C&+7mRqmGIRpdY+=%1&=Q_VNoR3k$_FGXe3!{Af@X|6u
zHfI;+Cq|>uZ0B=k_^<`kZ=1u=V@2<Flwo~(XlRJb`z{`s;X}4uSD%JdgBOATZ>U%f
zA<HQ3lgK%d;XJZ28>V9z*WG%{sT1+ol;e!-I@kGG$BgAiD!_HRl_#Sr+FD*7f)8sr
z8;y>Q&o9i~Vk+xU7{xu6a~A?*48Df-<-6{>i#&@Qn6`HvC}Rig-w}H2MfE+16$9s?
zsX+k!Q+%c(9H)YTxtKUav&c*V?kS!x_ur_cc@bfD+!)$V$WZSm#@>m=|J5=w6BcOk
zpre398VzjVe1ox{$Qag^Tm_vRK)&-`7kgZFZ|{1a!3(13Z(0otb}DmV+(HFbm}O8x
zb$kF4s=|efUE?@w*^#!aIb>3NN$Aa?-wu5b-WxVRT@zQpuKpFs<v!_TkIwSQyPcem
z4Qi~H3vXR2NExREShRFDyn~)H98VGN5cBLacbY<9F5%W>yLh?O-WabRJ9ho`$BxyZ
z6~<V~I?Tt@>8q~3`|hi+0uj{<!L-8(+v!|wV6vet$Y#l6xo$e|M%CA`YT`&95Je60
zEM!OWU)gNxj=ccmcP>V~FG|{PXirQiWQ1>n??R!Yp&Vb#Fz-AD2IKg#J5in%$Ralc
zxfDFvYryz|J7b1x<Un#A^bFGLZ8!u&3?&}cqvg2Bj>MY~f2gI0AaLFD6Qlie(A(Y`
zx*zLUD0zvtJb;SucK{Wk+|=w7Y;YdGfDZibA$*orse0%WQA;b#buO_X!as7a7yJ=Y
z3k%Oqjf1msV#gqBQU1z~r%=MYFD-}{!{UpdJM^1rqhe$&6U7s(EOvdVV(aXwQv;>J
z`GM)+_)QPY50(Z_ojMz2vi4VbK85ghHVM3qI?J=Zv=YRFT%Rw%`n>G%JIKKT0`v@7
zvN|;vFfO7k*5NKbfZ^^x_HaG`ejvadz~JJu3Bny5{jtaM)npF9Fhr~_!w^P&Xz&g@
z8HUbR_ZeXEsSw@E9P%|?4-EKf<o%i;Iniu^U^^{%P*n@)6g&v{6hcGQZw}~2wC36d
z2BV?eMfe47bf9-39227XJk`Hc66-xQ&r7SVPwoUpfb0$Kcs7VVvg0mt;Kw5{5mh4b
zWXvrT+@A}&)uw%*>>pZ*LZCrmc_&2f54%X+<HVA~!%2K(vlBrl+qYsYeFzOe?O*xE
zirw`4nd(?PD>AVJS39XiU)VaQXdhM7o(z75BNjuK{1N=B)-(7k{_?oh8?)JXM2}G<
z4jpnp#YDOqs==PHF23$Jpx(%vVO{lo^|jSP%m&cD(fqwq(8jhsuCxF^;7I%%165LW
zBs6_0z>xN}_Dr7HK)%ccT_SJy2i!@57kAuy@vG`kf2&(N4#D*Iow>~G!ozICBcJu=
zmZ3TI#+s#b05lvJC`U0!z<hMphQf2-D@|Q=(z{J^*m#fc-`X>I2GOZI`n>Bz-;gY9
zUGDYu{t4?py7+uu{HnIwYy)R*h9+QzMDDI`<w~M5!I8jj<O_>1<mhw6ex{IZAi&_P
z2wL>=mg(@SzL`=Xs{e^uf=Uy?G*HMtPJHLZBBli;)qKR{pH=6nxz`?wL0YWOu<_1Q
zd<M&xN=G)>7LB2J%XnvlwyG15G{MFEh(6nJ?BzeZ2P7p}-&^T&j+qK}4TET+;Ef1%
zl?)sk2te)LW5JT0uW$!zVv3`^Z_s@OV28av57mN3aC&gix9wfybAfki_k`g1TwpyO
zRH{IW(sBt9bb-S~Ki2c!NG<rS$&PnH$M71!euxpDXU@-YNWygVLQa51@#Gy3_|}**
zPLR^Tw|QC!$jvPg*sy8|PzPJ;=tn;Ck)dy&{US<0Wa4f%*-AwIz%8Gt(B)+*7OTAW
z<jIq#N1WJIMZ@{&Z-4vSLmxS7ohiG25J|2kvu>jEy(|jS;!fONc}(AH6s?#;S`~Tv
zUp@QF0KBZ29rhl0cfULI5zI2VQ1X#rpTsD+3XhiwzGrhq0WWJ7)6X5o$>xkvq$J5N
zl!W%{HBl^KN%+fQwRMq8;8OSbN_;J})L=I7R^%;IT2@9gt*aI&?sT3a4EIvF{MSO}
zIaOUv#djv{`b3`Jbq0VStpwWb{6&z((Dj<Fm9bQ@mLyM+lwGUy7XY^NK0Iu2Cy5$3
za9(NmRaipY(mK5#u-WfCYs>Ebf+8ZvMrt)goy(T<*0q}*f=9?5u$Da%`gzRSr%7i;
zLg=L%%LvlZgg`qb$pq1V0AztgAtQU77VSL3ySRp1dNyb?l~cR#;7kKNE^^1tHgxSq
zOrs~{kh+JkeEiHTdeJ`aFTTh1F}CZy9Gn+8iZJeDgK5+|DfQHzX9M?6BH9?n=Srhd
zsUZB&iCqX3<ni;oNH#=%`vP8yYkdQsYlvs>wvK7^57NQ&Yx|rJ+`k3IRV~K^2)P<G
zPEcyfJW>b2ct+M@;mYedA9;(x<8Xvq^4B0Tu>)C(=ZCj0N#Zp#E{jvvEo|DHZTg>T
zyv)g3Ks`uJqsUr`X}n2iaJ@G=+8Awk(YEj*!JSxlUfY0*lAknB9V=jbWQ=NsDH$h@
zSO3}0$DPQL#i>dHS-ntii-U*{!%0-8Cg&GsuW-y4CsN<e<*vp~WjqI=R5|87<?%2Q
ziIheWlLFlPA;<J5Dt`8O!Oh3<J^1GMT)i-JG`{T}?Q!W|Q{lD<9H*6?$-oF<SO{f8
zV88u(*ZBapgD~4*^MTjD{`GKyh$qr(q#t&l4(`LD$3tJexc=F<$jG32))_1=EJXhY
z#n4ukgvfC)F-`KjFnVyY;ol-^=<We_=%l~LGZ4^x3!C^ak(e^W%uc<)9@>Mw)!WF$
zg2jO924TP?cMAhBVbR_X8<veCW~ME+H0bempM#3e9zdD2UwURX+~1GH;!eb7F$a`1
zL`+h|jw6)-CJfcn;Gk|oTV!$mFm+I%%30Xzcvc_6&SX0mHTni^z8?;KQ^DupV)oAe
zAIzly!DX9CssFp|hxW6r{zPB`+BQJYPt+FslYy5X&v*;^tMRr8yge5BFQ}II$DzLi
zucz&-zV=RI*d=K6z!ZTGUL__iH8ncEd`!VG0RM+%+u!Hs0&*Jfg(((-*hMQ6h%0t0
z5=y((4O$aK+CdJm<ymr|`$1|>^^a~)CeiyV^T-A0AXs9kT#n~5fAAxa3VfJopzlK}
z{y871*y>@8+<h&@S9fYLPFij%lYwEZWTl>iPb-{qvw7=flLvwvz2479<If$bR-=(*
zvWi?LKR(`xIPw3iDWKjS>1jor*n=6QF|ksq>~meCY^Bnu0fpm2P_J2qjPR_`b1sUH
zgo>cNBAGZ&hC|8$XnMxj{ug1&jVpLTI%aBS=E#wmnbiC<#N8M{48g&AO{rS_;Oq2M
z>Fo~Bn;;IJN3;y0w=RN%_wKU*{ADA!pD0$`W9(l0zX8YfkL!#2y$UYRDgj*8Klayz
z?tTVi|C6$$?=gVm&)AOs`~Bneb?AwHv<2Oor;;4QvSVDHMZfyXhRNOQuw(~Rwi5{U
z+MDOvV<2{{!Vmk(TIixh-QC{L0J~!x`qp99SJP6JjShC@jjPe$d*Lzr&aT8L+WArj
z;#^O59EIt`M>+>vJCBHpG9oKdlGHVKDK{1t;hEW=b(rqzE3{PUZ}HS!yNkn66qGDk
z2qk+`qmnC3Ul=rhzqgxfl!_20sN_NYEoaS9>lc_Uc#Ua;x3*!Hq;D2V19hH)blyD_
zL4sMjI@^A()`%p_<Gu9F^q)h33GPym^zcS)_*d>qON=Zj_~COKd?>K1Ve4x~8~ni_
zY93s{Ty^6DE5zGk<Kw7m4uLXU32Gm|kRdWvI;^^t0RiP~koIc19lj<|=VGjz1$bSo
zsQFB}0uOzV-{s{-12sZ(m2$>L*@jmk{~TgvV$tLSFpqa$srTPxS&{JT)vM~IObLJW
zt+y2#qvPX)@+{gI93LM=hGFM-kpdq1q*6)8vF1Rb;l<VYGj+^Kcm5H+JK^PjkHzxE
z#VZ#FbGgCAueAQn<A?nIE`X-p1aE9#Pe^OJerv)D_@Ea1UD9qTU@LII0BvV6!_o!N
z6VZu@fVJ9c;Vu-jME7w8!^Ed?3GLxV^?_c8`v(G%t?B{PXh5{kyE}i+K-V|vqt`!y
zUhD2p221&<Gk)LuM`R7TFYq&GIZm?k4{7^|nm<v@$727!Os^&7rk%yF3@-j#>nj|K
z&jjlMVN8stJILqE+CO8V7DmGl+N6q1yv${0SHWG;28n@ySZI2Ybr=GDwbfGp6Yn;d
zN-}WnEW_T<W-`eK-tsF017Cd0Ti(*XwqPN7U9$4xYuoCV1_n~eLIF3pE^g&re75sd
z+<5Am7gxYV8CKyMs3m(eA&E%dL!=2gI%{%uO8%ER6p`RFFb6z^(nF$E@>xjjzaWr^
zOsNx?JDdM_R3}<3l3nBxbf11~!bGzf4jZMHXQNkLyYyo9edHooy7sDQ_T?ob+_|+}
znwp-Qo1S70YJcyyBaxC}EhU^WGa3%tv&jFPR8bWka}rCIQBohYof13|?vEY(Abt+U
z?oTPB<k+2WEg*R~ekNgR=<eMai_IbM&5DnPQGz#YnkB<1h22FfJQ|1ZR(Q_S-G;Dc
ztH^EAzNps?@{&OI#x%+z79~u74Lh#w<<Nc@`$^ywpQQx?^IU%YGka!KUeL%;Z+CE4
z!TGErn)3>-zKhW^=z3^;)v?FW(DfW%#JfxSbFc|em|!=t>iSO;ckNBq+DL;y9gl2k
zU6bq>kP;O|_Lw2+Q6q3d%nthv6i_;wOCg6iw5q&yDiew6L5M{%Q(~^@KgApMTq57p
z*N`W4fK_mGM&MEo;k7v;7ATsv+7g;hr;d6P+9`*1G(gIz6#Z;GzJ@@Q&Qp^ryf#F2
zuaxFodsrcVvD67epCr{IK3U*rEo^m=zugM2b^ckbG_c$}uKTUG#0j3>_RFC(^4n3z
zj`79PS?2l&P+7fn*06hGihu^R^|L9c6Xg_j^0K=6egst@|Lol_LF8m-(gTZ;;~B?{
zM%*no5;dL7@sXOofTY%+g>R1Y!TVx~MC`NrbxzF96$=CUhJiwHZq6$iiP!AD=N9aW
zpk29Dp<9xqP!##yY#N@nSzEq*E03US?>@~BaCrNSks;C4*Zf)42{Sf1lAyX5@KS?}
znS=@*`dj^qZG7}p`)&Wpc8B^uVHxmr?T%O(QNEJQDV4?vZ)t6yPslfnJT~}6pK6E;
zZ6T}p4e-l(W$5)gdPs8Q_b`Jsps^z54^ptc#N43eh&aJ4;)m2rr+38D+|fpAt&(+Q
z2?-Sg5nB<&LA5#HPme{nN4iqu=MKc02)+qh@$f=8ZiT1f;b!bWDjdi8Lp(cgcF+D=
zFDG|zPt@UHa#U)&H*A3nHyhtBF3wh?R=hAXQ;1v9>g;0ip4n<~am`BPXJ+y!dB9V*
z^)hhx_K?IvFG~$8h<_J9k_;#bd{A2)2oeN(7{c4of%HH(QB5UUp3>S^Rp((EBk;G>
zgQc+s+_P2nPvh;_Xa5K}r`oZcGw|v4a9D!?A>huX$iOY?SCERgSZs`yKND-m|MYJW
z4cLy~G7wq1liZrVIUHX9^njBCEDIg(s3$@h$uSMe=962G6yprv%>)GtCNLuqFx^?P
z1*07(8a#*8`Btm-hC}J}T((%uu0acZVI*<8o&2EX^Qioc$9}=hs>n#}fbnMa8;1@Z
zdP9r1&!y8hK+ky~Hq!Zi*8PQ8+~+m<X?-wh-<}w`5PdVpG%WA@wEA7V^Eh(Yy%gix
zmBGY;6Ix&8+Y0&ikUC@((HHaz7z{_ar?j{*x=<i(F5=A6;=)R)goYaM_3xczW^&I(
zN-U8r_2(d;+3ex_59g|teCKvl(zyTWc=BF17Bz>%u|In3vBz$6-Dk77Aj+39ja^&K
zZo)UlKgfGmZ+lGrI+|iDXe^uUd|H}{g@?^(%td=YU{h@8*)A)gF)v=XDQfE5D(_vr
z?J<r^SALlhpSpXnaw%z$*AN&ZzoD0_J|jAl=D0SRtVm^{U6{V5EoMFmAOYytoeMn(
z8QIZ{oqf}knVrl<U;ff^el%m6#rCPrw=&6_V>Q;FH;AMFf&a0kKi9j^h+_{G=_{)8
zf)~6Xqaxw-!o*`4Q>DIAbJ7FpLNQv-++ZbcjV3$)EBNr5eoSwP*t*9au70c93>^UF
z4=J#aBq<2Sz=&`W$`MCMC<$SiOf<}Y(DrL1l1`n>6{!;^t2cc0t6!~F=0IEC^MWI8
zvhGH&iKX6-#O;~WL&y$SI6N|NZ|ALejlLeM8?L$Lnpm>P1ip7JRc$97OFbuUzJ2_<
zJmPQMmCk#*Z;eMWo%-DcC!c04T27NcsdleXCG1Yi8g+;x5Ce?a5lseVA<&p2{B)BY
z13pF9SV)%-YD8o^weI&WZ+N+S@AK06zQjL1J5h$szc#XP;R34Io%h$KSi3gYbp!2<
zolp74XJbRdldLqpDK-80H2c@%vzx$;*D+h3MOldQ9OY!bE#KPr86nE?u9-zn6Io0w
zT`uKi&$&NvnhtjCBOA~nA{$Wav&+k~MA1h^Ar%ITzcogW@H})2{jPMc(|Xn3$(IN5
z8GAFE%l$EN{u7NON6z%=R*xKMJi#k}tjwUYA*c&|G3=!NfOdB{K3X54dG02#^No~5
zTE&tOQ_eDeIYye*2EQ)@1!p*(1aj8gDel4`2)i_@xmk)PvTCuYJ$*NssqX2J3WO*J
z_n%U!tL9k=rMZG!rt;2+RnWmq9Fiw@s7igFUBkzoockKYl4VvwnZ_WkAvh^rE~Ru}
zQa<TE*#%83iCug>ebl#M{oh2SY#x#t>je_$VcvCeTida)-X%X0RW7n)0s09C93!${
z{vMA#My*~XJ^)L`tE0Gp4~!s28)8FG#BQ(O4z<VM+=QQx{>}P2L#onb%E(`B_&Zv+
zf?B!{tITljl}Kg9I)H&<LY)f(sYD_L;gnl-JAT%cb301f$<LN>YhJ5tlns1G%JLQ;
z&6HopNeDKu5Nz99V|t8k3B3T^+-osDJ&aDSlgN;(O)nK8lVnDG)><&gD47sJoV0aw
zsPr>*KNPz=2rFO<k72ktiZWt|qQl*z7&KO`0MEFkQL}UFNDlH!zA!RXOipE=4`wQs
z_6WeOSp;8aUmy0zPd^a_1DUQOF&75B_=^-bCMOM%I68k2OQ`F8i6b&KRfP{j{mgIy
zegvbH(>D%HAQW3SvYK_1;qj4?h(Dwr;;Vfsl}atommvV3DMe0FIthm-B@tb@%yrnq
zQ&nWqds*mX7ndLP0_Ebw!oxsl2%|$g<OKZuf2*ZG=R$>vQU{18i$ycYFFiFhltDeW
z`07tySO0}HiV>?s62$}K3DhBH*kK-vJ$~1ZrL}iLCj7GA09j=Pep#mxA^Kck##j0j
zQdm{WuYFlS0-3Ez{J4-!Lh`8Ok!F&^mIKtl3a}2H@K}@=u}Y(;M-$kQ3PMfX)1qGL
zb)`%;e9+2eNB*+y;W&>J{v{yXTDG#8Qs;qpKKS5+@ps0KzUW0SDm*X#ikH0PCGiIe
zAAIk7-y8o#{A>5!cVFRw*gCGopEoj6fTK^U_;hFl>f5#l`ohh$A^DA{5R&|~@8!>e
zZ1G{+_X?k|ZM1<08yFE%2rtT2m^}g@&=~O;e!Je~BqD@Nfm0CB&n=H4nQv@#`O4?S
zvaAGq{MdE*{B_5Uv;1o|_M9spN>;16Ba@Rya@DGfq*L~v*A`RhbZW6SjT`ih`Y}8{
z{k6o@^i}nhmHJiFQ=kyY0qf^7IE8r<g)XU&$A{ugh;nd*S8dQFthqG2u<+_YLgd@I
zqHqsZ)}IXy4(4jbOxnq24o=n$Rk&iDK~A~hQnuwp$^*Al^EspqETt#tx>;QwtW+}U
zEhzI-R>?%+=xQP5jt!)jhZl;d?xoyWE1a-W;dFYSkWLPa4wa9%>1rws?I>z@5C@2S
z?8$L6!mT_XnJ6C(y+8B`^ao&~g?J`Xpnjz5(#pB;-B&sOD<yW#rzZhaOc%6$m}{>_
zW{?Gkj}j)3n#L~G-XouXzb#ux{zhABd2Il=ghp)$G6YT8CM9_)SS+cJk@nED2i`%@
z^suPD(nq7cHKDHvt0)O}o(}m4_(`Abd2%gSCrPFCBYQO2_ho%6r0)gUy>1%;ea$Lj
zIcRIpd=u9N=SS9uP(jWW&-jj;9>**ckN$m9iA)^=oQe8H`jG>;r&6T``CDU{<XNq=
zU&99Z<wgk{<-OF2{Y$k_Y4|;FR0>sI;Vt#VvOf;Xi8`BMd*1?ASfdP(fBK8Y0pPkS
zEC-=+&<|=0M0G%!+xuNy%#hY#m0%r-v`blU6#@YXdJ(IJE3JrCR#LTr=VYwH)$2pa
znNOPObarDSn@$;(dy%jtcL7R;e6)Qb=UB;m%V*rVy9WzbxmNY|&i{Bp4w(DOD}l*#
zkKN<uvL^TsB<^~&Xl8TnJze<|tj`wsv8!M~y9+Br;~Q<@61jtmK=WNZ$)VRn?z0lT
z5}KSQ&1~rMZN5sjKUMMtZQgnuqt&%=bv0)i3_NJ?m10k#UM*R2y$S5Ax6zD+jG1Tr
z>Y8V$h}-6aU1&m+-IMUTWJ5?%>hJXN3e=O|a8X|25_55J=7fvDi0rLe_9S4R;ocFU
zwh(mGZ3j8}zI|wHdaI^yrLFaO*y`9&$|WUKmplTw`g;0lc*sn<@oOz1Cp(inh-yT>
z!bLPG<Q9Y8{&f?e0lVnp3S~Co-Ai10T%IAk6xX)T5N=;bU+p_$vHxQVId(nYw$8U(
zz-#=ekvZhQ?-be57>=CtM(07;zU=E_cs!Yme)B9E@LS(%w>S9w_kHCwaK8C;Cp2t4
zOWg-8(-JEsP|E~}W#G6cbTx#2>b`i<Bo1C1OOO|UN$|Hy^)Np5Eb@}(2nqYbNaKvS
zK;Z#WWj?4@-|>!j&;bZVg9bDCd}g4QPFp8VKEJI2u{~9PUvN&|GdevzTF4a#VU{#a
zB!-&E&n_*FL>(tOvbc1hFoAlZu<Sd@!D6m3Iz2NA4CX%c&_no!d_IxLDf_36pJ1IQ
zG@s1|!R6x>Wt5Yttix8>dMs5@CNdKs_kpP@$ajQR*}4OF`m%xe@C4~4^@@LuvD<`(
z{B42<Xxrr=Xd=&HAeq+@N}O^o_x-LnXlI9q7o-Vrbr`PZO758rtPnj^zK6SO1Uc(t
zh}N~L==2uhYmf6_RELJQhKCIEH=n@}w*}vqO5+jvCVXS0%28fl-{->LaYEMbQh8<k
zUFQeSVCc=K{|&kJe^z}x6bntWGAN|4aiCd_EHvvJI{fO;Fu)(C04fwko~p^>D56>b
zjPou|hic!v+WFR?!D8*7cZ_<y*(^j;Mk<QBZ3F)$mrA6KHE!RCOro$&5F2^~49@zk
z|NHO^Ysmjxr8<A|WUB?xNZ0GrZ%w3gsm=nocaLgE1x*O~G+`$wZ=D2hfB!{sDBcsG
zMgEJwWQhn^(nFzz<QiL01IRUfhCC(iAwzVz*)E<3{lwrP)277W4;;x)&4g#B@)0Nt
zqp{4$GxpZ`9I~(^5}6D#>%kyzMp7vz%8TGI^IMnXx#Q=)P@x*?e{Evjq52nU<Pph0
z%IB(+|5bFI6U$@0k7G}!hsu*|f(&5u3pzF=UUJMaE1R>1B7=jGA;>z`nCX<_C|*A^
zJ{U$)zgV(7uHKdB`mxIe)QJ2OFhrTxRSl#1x(p<PKS9;VLfM=~wfp~<ws(P&>pIIs
zYoDs~toQrb)m2^HRozmzRNbw9NS3XVY{_yg+a15<Bu>S#6UUH5Jd)UP2pwXu9TE@|
z2n->Ub|x7xkRTW)2@pb)7XgN#n_=C_B$qpVXAu^Q0LwFL;N}iP>3-k8&#9`GTC$U2
zx}@{iXP<q}+56xBd;fn4f!YwnEjtm6aZmjYd_cOOAK(>cFi4cTh!V&RjO$(3mO(S{
zZWu}`V5poa*f<cQ6CCmZ#dvxCFOt-BR!mD^S<d8>*%9i2qeqXP{-j2sMo8}m|D*Q=
zyq<1K#A3?K=l6%hbyFo1CCo%*YtBSgtKqMPYvI9%FffUm!xKjk^84t@q3H{SLg7mq
zW9sb%D7Q<dZ6%McuFl2c*1dOrypqp(%C+pQdGuEIHuua1G6N&~Tg82O&-5IlR$C&x
ziyjMlUgplc6Z(a>gx-mmM<E5NLKr6?N!XwfIf}|$-9H@C0<LQG$zyF^)ZCLk)(als
zC69Tpk6WhBF{5kgyLGFQAy*zY%9JVoNGMb1$cmCDC$;nmknTI4Tv;fY09ou`?ec!r
zYcvMi*rT=Xu-o=-V64^6Q8O%kflc9$t&u$jonbz?2zZeU?Q44oRK@WCPtto@L7?Hw
zR|2$lhOdOjJZT2phIJqvkC~_8M_W`|#bg49v3Po6ce?9FG9B>sX(WV64R(qMQJ)2;
zo$@@sqvR=E#5*o>)tlDi6I_9?*yjPjBhe;?N)NKoL!pou@}lrdGBs^r=CvOAyhM1i
zeyEGL#hY*V{9pG_eKMS&7Ir%WmClbv4MzMcGVKNO{5-!lw=eea_=i3m+c&rNJQDr<
z@sH!LRR++P$}PNd$R5h;=@8}Jv(Oy=OZajRwX0*WXc0VpHeNI<KHH85k1?(=;8R*@
zi#Y-xk2x+<2|ik+loin>>cc&&!VQ}kYOh<RerM0S#2%ts?Gu;?aWYzg%*&zP?2XHx
zJ4QXmH5akEF*-VPMG3AUG~{K~P?#ItG&+Lm1Alt7ga;YPzI>bhxZr*9$~Fz-z}UMC
zn|*k%!lk`#=a?ECQ?Vyy>0go6s@gz8BZ6#cB?1h57&+ovc*1BOM=-pix1fTz-J#_i
z<Q~q1vfHWF4(DORotS_nH-ETRHM$3nAqa^VXGE*`4an-DHXxn(UWwdYIJ(#kTNcb^
zwq=F8i$@>2;RdK(LG-Ys#<<~zhw#2`sEYUfCgx2H{s=dLlkA7yh_!srhJFFmLhDfY
zmT<)?5Zi-r;UmbKcP!zk@v5Nk{0x53ssJOtQY<H6g~0yc^Kc!ojE}>H{w5?PEVNhG
zFfl=k<b*H|Gz%C-)eau1tMT!BXjxP!lZI0rB2>BYbmp^V+fGj1nbIMUQ+G}!ZM(cz
z*$E?3gli?@#CS3I|Hr&Ij^XEZ*Ql*b`oo5aoTIE1jTUr;l8L(iWgLWje4ycN(J{*w
z7+%XNTlP;*HJh~+Z?;e@7G}MbTC<5XPnVb6Sjb^rQ8*eaxbCkZlDS)mL41gqsyk}P
zG0%?G`a!=&4wV<jCsSh_-+-uLTpLDD^$}G+1Rfs4rb!n%Qiq9Wm$Rp_hboZ0Jz`fi
z^r0rPBqc0?)SrivF1{T{if|||%7iXs*0N@iBBVG~&Sg=2&CBW_G~uvn72@Jw(=A|K
zSEK;VO+h8%g=<~Iw}gjyvI0KlB);kDzJNtUt_A`Wtgh9BG{R<4$~j2&9yNu+#%v)Y
zX|Y@`7UK8|jyAfDR=-O5q5jZDW8<MZhS#X}TaB*jR|>jnp)xx9xxu&L{`BoqD(son
zKSpJ>rc0EwMpuwOu9|KnRnmAtbemn#*~qAf+o>J_-Rp-4^kkXW#<v_b%q}fAkozXW
z2)39=c|DBS3gV#O>gw931N}-Nl?*!$^(zq&S14w&9IC)TMd4(saGqBg#i)SNd<+T3
ziV><_9EVqfCV<c2*gQDLbKcAvQP0igDy3}JOT(8pZ8>f+owC?WqF|-cMc2WCM_lo;
z*-|B!bCKo+ufXT9Ep?u4VQguC-9f)>h2pSIzX~y69>@5CD=UC<NqUF17T8h(EOl6#
zCkuG5CIr<HUq*5+JeUlA0MJ}k%KNltOYCME8|wl(fO8&bPsERX@{^w|IF-TID^B5o
zR;vX9Z`3i04GwI`Kd2kgXr|?rOOv0RES0?<1<j#UC7UgyAleaZ)T7{@s3nmNs88<J
zTXPCV2pa~PjhW10jX1hKdvQ23vjOvs{m29al=Q@f(OO<7P`60h1L{unoFK7Lo8WiE
z*Ljy_2v$MDjKI5gLeTp0rk3jkK$a1un;ZodgxZorB&_W=R3EM&ww~`Fh8omck*xRq
zeJarOjU2jCs!J7JgPj3AVQ@6KI9heSo2MJ<_=X44hhz6jD_u6o|M+aci*f0T#v|&}
zzyvn{R^AnQFW`kd3uZmUY;p1eX7l&L_D~w*|Mp5-*J~9is-lkgZ$Kg<fRXFb{GyhD
z=;yKfTsRyJT-@>=(nHce^qi1BoLq=TCeDE3nzJWdBOJZmgcrOs?M4&xbLk2qlioHF
ziEcmr^wUvg{!zn@o`UGvLgGBQOIpbKheX!S|LedxG@b<QiMT1xsoF*=Req*iEGCh2
z$tm#v*AbQw0qC84KB7MDy{s5^T=(RH4OL5Q173M{v(-6}I9kFA1T2YrnfLS4h!7b)
zym}DcQzuWJ1fIBIlFc*rWqk6)*zpaW>B8L=q^g^?U8nGFAQySaT?qttipF^o%-K!M
z*(kCw-i}Q1kLo!~U}8rXMXaIkSBuU8geH=CV%<@!9L*N!xk$(p{0yq&tE2r6mHwq$
zKJEu@M6F+0Sy}nMNhklv%ah)XA6KD=AAWe^z$fdaVr(u;hpj}@u2iO{D;4`kf&+LD
zxoUE9@l|fMn#C$O*$)18+@`Pk{WDo?`u*75>W6;lhgt|hUVYR_CG+QSbBkVqg+eRX
z<0;*xR?GgiiWSL$9LTnDnOlI5^LyZ}-hjC9qdZlCEHV<lIqNiS3VKCk6X^?@yMmcI
zI?@^kjo8}a%_DPG<~L*l^tH6eK1Z*uVPM#M7@T5%q-qb6$EUF796HKLx;&KOD8+x6
zzT5yyr;=qN9mYD57*|_QxZ(1VXe8{MbK#CRJOq^hT?oP$po`FP=dpUeA)XXy{vA#(
zUj=%Q3EkCOpo(ff=e$D@CBvR<^Y!ksqtGBbeGGc^R^>)*9QnMp(9klqZo_OrqVFD$
zVN&GW)R>KXjn@cYLN3?~oz7pxQmNS2V)*Cp@3RqizNd*4zBUQ3LL193Hq>v9Rg%ij
zB7Cg3&X4W#wecbA9>NxGz+YR&d`I*QbD>A@j^ZDMfCu<dGD8egq0t2QC9Di2N};y`
zyl4tS;4wo0=PbZ@f&1c!v$Dp;gbvowxQgFR@vhiUI$t##h7sUo!)TaaMdHauP%xAE
zYGy`0tqzkJUVK=<lSl(GN{vQD&3*N&xXqDIW9@aE{2^=Xd8^^K#f#2*B%VyhBkPC@
z5{d{-Cw?eHKY9c7ExY}<)A-;th`a1z#npMF1Apfkt%btG2aXkpE!b<s*F#ul`D#%-
z@3mubZveZ3kv1Mk!$@6d4l2z&=4%-_4kX!RH<9RaMans#TK`aAP%;<zftSEE^Z(#a
z(uR?4A`UMb`W#!}owyx$rqjrP8XW2i;7uE_9X8<SpN01$_-(Zcw8v56%)^^v^wx?(
zZgv$>3AhH813vBY+Hy&`#nRWgfb1K={?CB-#tT+{Zf54^_;*Dq*8)6ZBahr8!TwL+
z0y|z%I+aSD`qHUWr>-a^jUDmZCxjylDQyYYMnsLNv>}GDh&iPNLLYMj*LSq*TcD>P
zFA^3%Pf<SroJF*tssRn;0ox<&u0p^X*YDpKZD9`_(u(fef4#;fdz5aDyv;{Po8zL9
zn8dH$$k+ofg+?ysqoa*+sSo`Fz7_hgP2hF#s$+elGo@;Ii8B?;mV5^{-`xUuT6hb<
z#UlBYPB4=Zx)r*BHQ*;byEP*ZW>ysIoc}?3FOpX+*N8us%6nM!HPOn)vdD{+LWGfW
zxgCw3Zo)I;6>-;!e(yX;Vs!r1@tEz#UxB*K)6r-fcox};kxD5W%eN*_*UM8_k<NOm
zlC%!6Y$9Es?6l|S=MBT=9pMZ^*W7Z%ipJfOHQROV+EHl1qSldH7|$jHOI5jT!@Ycd
zzTKIurziMk2dre}DfIAexY`Gt&<9q~UUC|uH2otiY1kqZ@xT7saI=XdGSowm+G#vE
z28qo8<`^n&#LsWxqsY5!SlV^6sTzYF)nGa@rURo5#W<MO6}Mzb2dZztdP&YxHi5Mh
z<vZhT@?_9wBW$ckGtR9msHEG%wx`)wk$1?uH;f+v-+l~U^?)AW|3DuZ@CD*k<y#dB
zz$SX$ky{|;Y6G(XcA%sN*Fak?S0^q9-g%q)sf`Wsk5IK6#aSXek^K}hMkju8DUp16
z11Vs4dp<Q->3BT>k3Kk66uuIJ*fU+W8yolDdvE@@5APws`w3!l4cT_!+#Kxnm;_i{
zj}^DeJ_Xa*06K-Wf#psYi&z&*%AxWJ;SZ{nG<?^<B@E8Xc`98&@^73cfQx~lMX`lI
zFzUCE7_?fMK{8O3)-$X!oj0?&YQSbxkt!xX&9h8ho&Xg^BIdxOZKy|}zLmII&p_&k
z<?TA1F#2;`q8w>pb=auOP#1M5g=KXe?75JP951t-@f^2bWxO^WLy)_z5h@uSsS5<G
zEjV{!mr897zIOin*ga_B%OjLFIs*7$JaW+4{SC(Ba*u#K*>AiPunbVESei856l$t5
zwZ_kWl1ibmtAHzFBcZkwW)@d2`g7&&u~-~wdcxt@+ut5T(3_Nv*!yq4=EY!|yzgd2
z?upjez^PNbtp=9Sy7_$?K!;aj&aSKcKCpUzOgQ?2o`!t!CW*bf%DDneNJXROgR~+N
zf!Sm=q_DvxNTe#PLh$R1W>_%4?BUu;8EYB@e4`DAJE#R!H&m6wQ`QQpt2k|Q-AIN^
zG86^@zjR?Z|IuBkaeNT4-MYfAOMCin_o)~76SA6rLfwJDG+_V2c$>SFk-PMP=f~e;
zSKlrma@(sy=Uybf`^=U7;CtRdN9*IGDTH|;<zlCYudJ`<j#yoS-rb|stM~4%@6k+c
zx@TweSoEWU-q^TmZz$?4hMlly4{r@GqbeN9(7V;lPa(@<_Fn|Ux!Ymx{CJa%@w)-P
z!0}`78v+o17jS#^-q-`y_WY=;fX(jrTZS*|HQsM*h=m{pv+5derxIt}Y%X?NI2;TO
z_j|u#Df9kZ%+0=k*vaU4?FYJJ^hADG0L{4<tIvKo^wXhV3jOXCk_4enYqiovU)It>
z6H&Va5De^Q#PC|TVD3>7xPpVk3KYIH$z3ZHsdOb88D7}C;!ABYL`#?&x=U%tsSA)H
zA#s8Rll)%bW>_+~I6P!|ch|*@;2x>7`^JrH+^4k93|?jAC#phStP>=XS?4p@a+jM?
z(FTMFUM~gLB}U??NZyUN=pwm$boIiE+>LnyJ1%s=dfX3S-0#D<|7hsvL!UtOyZ?xh
zx9ngnN8Y342jOWi<20`1g%V1SC?R%up|e*-c<*rNcU4qP&2DOd;lhdlCScQ|Lj?6R
z<a{*8z6OQY_&MLI_);rsKP5oE)_o8{2AiAL80d?;uEFZ(LynC2-~xQ`HhmC7qXSWC
zu+d$2?FNr)fesv9aI<a&?Fs1gc0<N+<ktm$QX?SXGly3-FhC!S_B0kpdkvaqrM(fX
zbT5aM%V6n>>);u<UKM=xm*7QsE363bhW+y2il07aFy=408w?`0oXxAtl%x35nK9ZJ
zUi^xb#lZk|^W8idC{#!Cj5b*+RP~{ZhpXV2=p%!l+(|t#AAvi%E0WX-l?!%m(Sj0y
z0eON;l;`|QSpO;jEzLkgW-+(+t&N5G1<)6-VKs_~K&apvREb1#suZu3I|wN<KY!O<
zosR2-OU}Og+%Sw)=~O%pPeu<qnApRG>DD}2XzX)JVaIEC?z(G!J`pua?!-^VkoU$_
zVYpYMTzKV$oo6D*Un*tsSUTmoMl{)Iv?1*wx}iC<v^;MiY5T1c<x<25XVR-j?QA4$
zSd~mBqMj<kZs#E&V>}v*C5okVEcMYNoo|>9R-hogpn;Vy3ei5ohVwUA65vKk<%!Ob
zk0oR2QVCk_Xu?Z=^vKc&9Rq1;kna(XMsh+(0FIe*OebuxRPl|%cr&G5JC(_pR<+Vb
zq{c|Yv2#W^0zOf6P1}GU`pHhi7IMQ7Pozak<XEN2+jd&8#dGx0&FFXy3{cLn-#G-s
zD+1}p!op2AEi8zxY`Z_MDBiXBt2=(h*MIKtKHjg@Z;%YfR-#=PzhPti3cf}2z0@l>
z;CkPKbq`+)eKQndtm9=c=&d!Vk+jU47i}@+WW2ekQAEEhM}Z`)<>tWaleo$h*%<VU
za4x)>#$E+T3RVZ;Zc8g2OQZt!nP{QcL@p?@C%Hi}s-oB6<F94E@3uW)NKPMt`!&_r
z>~6gKw%cwaC#xB%LbEwL+iVtk!GSKu<t<{Eej_{hdq!45R_|0Y3PnUXYZ%!u^b^q}
z0=GtTyyT@Xzm)cH$r&FA)zEccLNjn09X(;!0O~x&ZA6v93-<6IZ(C;~<+|SLk{(Jo
zh!s`&FUMNi{jR$na`~%W7>Ak<UgL|e(aQIXf4$719a-b)AR5JMAZ>-{g)It(4uDU4
zPia>+EgX-;t3!-NF%-6VMnQ>k6yOirmPnRlsSryF8$geqbItJ+^?hf~ocYkpe~Ns#
z9^|ItvJLTbJuy0@AxCN?7QZ)Fe~t9x*z8h`@xBpjPCtdE)VReT5XX@w(KJJCjgCiy
z^NfzWj(8wrOGd|!1N)ceQx@U3bsSPuz@2EUqt~`*ywfaF`-?Bf45mR5s0PFXLaYt#
zFXlB=#bhzTX$uoUmV(oAs-u;FoOVmaA^G)r1R?DI2CFFSOg?HoU`6v;_q3fz!d{op
z@5_IhbXa&bz4PouI#7?qqoC-#e>>pMwbkQ&&_5uHG$h!J6V0EG$B*yBpGdWAWFD)*
z*kT{7DwpkRYY|lazC0l~xWKg-eykQuYlLm$xD%5gqJd`LSgZj`gOax36KKQ;<^zw8
zV!|8yfoy<v%dX?NSh3s_rCmIpJE+0Im+$)e>|idA*#&K!{w&^(n45Xto3B(RH+iYk
zXqNPl6XI~RrO!&uMu(02I&fh0?NAGOEnac?_zt}Ywd-eE!@6oVn|T;E0u(%4nG<k|
z^sEmY5r}Iw7rCN2KyXE>(a`Usx71LISwg%$@rvifD`03j3ruNvPux7%CM%^sgy1eJ
zL0Cp~iEnk22F_9E<hO<KGeuq{-HAirayzV*@4)x^eiOt3aMThsz`@KfE-wQn0l%6z
z*5*kqHNgoNi%hiAmY>S8rf`P9eZGc96RX{kXgUOW_E69>a9#l{+5l5~T>yXkm&&cy
z92~MRmQbdqO5#1T4P!0(Nu+f(3&u#PUN$>5)oxEsW$_mro%cNvq}vqMGBK2%gMfN$
zt`Fp_nw$x<fw#-xYv{4YCd&>UcbrtG32izQ=kUuP9&Nw}Uu$)HqSl4?9A$p?;W7AQ
zycU^^-W~dx(8ogmF7yYX|Ag4Q{|KC_1p<u=X`N&TvO~v@bQk+wCmQczH`A6l>LMZN
z=<NWA*ol^n_C|1pB3RBx5C?iab444{`(t;Fje<Ze7wqnS#!G19il?X@q1Z&&NH9<^
zWRnv74N9&;su3KY>kEj=W^e&6G=x>zXn@rx7lx$``j-TiFx0`x`QQMdyT)6glS)uM
zHL9K(m8$c>yL3+z<DJ%9`GB3^k{q0;vvtszKtymd*7DekY>d|)d*aw@kCk@EYhN6_
z-?{N}!M~7wc3>CZ3ax@`ygbgMR)mnq5VjDMrPa;D{N(?Ek>hk;D=ceMH3^lh7%Gy*
zTSUMfPGSv}#N>oV-U`%A)_EB}#GONSPu$gkk((%+?Gb5UX-jBVppL^w7$owbnRbxl
z4`KFAZ@%5icGnTF2zgyIX+#mg>Ojkd<~~yh8%0FIM@q#M6wm31ouMCaHWha)WxH7;
zlSYk_CpDl|sU19c<jBE;HP{T~_~fCPO2TG*XFA2CQ%P8};hJXfG2w_ZBk+MWEz^tE
z8*$r6pz%yHjEM7C>kN^`QLyJD`n##%^)k_jSz)Px&c=E*%<~_?Jl_ZH`*WDpnyl$w
zX0Pau9R=GD?DvJDtrZQ7YsW6-7C)dq{-NW^%wqy`8`hYC{mx31OCe7BRocRS%T1Ww
zWFCAz9XDz{KE{Sxa$(k2)E2YM961<G)TW?<hgcOxq<gEQSj?<4pMKzm>%;0USin@r
zVj^T<QU|*Gsahg>@W?HztJw(0A?B5ga3Pa6vKS@HiKnuWNXACm+%Q%s@p`n#>sgiY
z;0OzIV<MeFN?$LkQnxNH8LDU@?6X3&eI~`GKh$+BHkKA|{Zn^+{a(km78h^6Za(R`
zCf)#E%gE@6byHaHepsz>fD*{sFj<d6=ETT^v3LjLq#~&dwq`i#*N8XjVEs`8e$E)d
za5`fnEu*$!$-3FYLs~xWArg4N`IaE`sJ2%Pv9CaO>jLM$=HG$0MDrq;>#`yc<Xx99
z2593V%SU%;o!7Sbw7Bs8e;53&e+p{=o(nw>?ZuZcm*F4)z`?^IEMb&6&miS#93NOK
zd<@?LT)?2A&jLNW;Bav0oaeM3k{*!p(hQeKaLa@f`t|cH;NDx1?j#N;fL8W+KJT`J
z{Thf3?kEQP4!Bvr%9ZQp)tzvh$lJ%m&U@XdD8jj7X{cJb&Y_UxT(_WHEH`zdQ&@1}
zdLB5Jt5))qr$*HVU#w4?$@u=`cE0O`kJsc0-A8BY`{T*T+a_krR1DX0i%z6>eE<Ia
zv6MM8@v3nK>%XBCO{8Mk@8l&A*LCy`SH9|N^=rI-W&*EaKEBs0`qw^$Ucr0Y`G+)*
zrne%GX_#1-72a<K#LD7bN(PKZyNZF-Q(nS0VyYK~f#Tp2RoJ+ye=P-}0!1?RMEk@j
zT3^1XIXq4J5#?6+z2NLW-hA`h`eAry16*%gVwTdy+X%xj@APAX@BMl@pHF|C{F8S7
zyl~a!-hH<-d-l{sIz7+ZXQI&=HaVY8Pn<eC+i?e5U-`;co{Pty`^r!l5w|n)R^w)K
zv8iYp*F<d8kO@wkcruN?5FjRsgu^S5Wm-N`)-G@kgEdv#)QFBeRW}=OTTX2Sj~slP
zWd6Kt8vH0<awwPMJi8QPMd;yKfzG$I?*#91Tvkc2UwtdYUM~bLj5ykV8w`!c?&Jgo
z?_mB9yOKv^?>G1>VVC%5%{a+XBY5@8-xMjNc2$&{$s>d>{SVNs9_UsL>pbW)tNpGD
z)cpV^a+uek9T38>UAtDWU3B<}w)kJd7Ix98F-ipvLMR-8b2qKhT*M-Ll_Sxzx=@aK
z$?v(o(@~kQ(Tjg7=_$lC4w(4S(Gv=h`;hM1G%%Bm1jA$JRRqa^jU-qD&=!PeO@+fr
zGYm6|5sed+SS1W5ASE@Pr!!|U`|kL`gU9El4b}BhckE<~gI^r+_oJgGGKfTsM8Y%-
zm+YdaOt)491H+`!3cLaq#^JGH<Qzsy->6hEB3`uw{VZ#Q{aCPq<9|N%E1};(-?J~Y
zX6|Z7oaovkW_+tXV^&0BpwuAIwop^Y;nB2;W1(dVK7s#PYxG}>XY|R^8p2>pgt59d
z41mWQNnoe|3XVxs9xoWPgsLt#a)e>f|FG|YH*GY8E()cT)1F4Tx|&gV2B#EC8-ZlH
zZ4?}Fa^JO;%dFDpf84XR+3#;|bpuk`)#v_2R^W4b4YV|P0j1(tPma5RVw{irwcQ4f
z<tJes0M6sYWM}AZ)&P2~^T6pyrj4wF;<Vu%%a+)ueBl-|9zX5q*2!N<5?csXnN)D|
zV7@2eo^~Ip<O0??y7xA8FzUv=pp3~sua`?mksT|Q>!Tz6d3w90l3&)2Cr^<d`lvqs
zD371g$6ro@{3hextEV#AY-Z}!uo-|s&SYZb31fJX@#6;iZP+~UUK#7&JHR2Q!KH*S
zhl<+Y1UQ6bhqwfLQFR6TK;aeaiCsv(WUudE%Y$_Q)B^v=U1P05Q6{a@TEnLWca60o
z>NIH`n4?j$OJ{O$@$S5`Xod54pGbP~;sXezUr4U3B;myEJWz~#spEI&!)ED~9XFlL
zoxv?&CnKq30{Fvm(hx%;KpQlvC?g8fiWfB)Un>)j8sTI*n{pl7NhVTB6Dwj>@$5=0
zkt*B`H~Uhwb3BvJq2ZEiXV09;+Aa>wTt0KW6D`3-|L#I65nDN1RKI{V=P3(3a2mE3
ztUj~TsaV{kt~$eP;D`i49km&crP4N5p(T>ZbksGiR0f?MN&$~|F$T15(YYRC7`kmO
zD_xYOSE=7=7R%uxjLzYabZ{FE!vstUDn59M1w;>?CFX_hVh(rd`#@1SG~n!b2MCxD
z3BdabmzT&H+hH5Re6k%)3J}MZPsUw%AGwPW0aiQz00kuwqcsr$D+fiQry>&;7A5A5
zNZYd75ksz4kYwb*#Dp?BUbMo6`i||+)v7TSt|2NgOk&0X(@i=?tJiz^%X__+;Ur!2
zfPu&(2q;_&t5~%*=i0a}nXE*;j-e(d4j3u|NByxUJq6srf-F#Y1yYGU_{)gO@#Pv}
z!=_KPZG@#p+KAlXPPMv+XJWBjeWFz^MjgjQT(_lgxQ5jlrddG-aikwo&P{W3O3lsP
zgmH7Rp0r$n1<!D6;qVfoz#*hpv{-IU)DaJ4=5V()<z8<ZkVkhtQTwU+y2nGIi$beu
zp>7J2N{hek$O6>Tj8HZRq`E={BnE@PVdZJcl)(q8>Ii$p;ljX(nvO$qh`&xSR2XKO
zK4lnXK2lP^oftacGTDPD(2bp5h=(AHh&LkckkvAfKe&fKtM6L1hBJZ=<d(HV01XX@
z5ZL!pc8Vz&D2GZ<pq&Q*D&`DGD2#|DPlwT?42mFR;Tbc}3~ZHn(TtUmbIFcQ8kjl;
z1b%p;!c|otCz!g;j2U6Wa99jc5j9lQn2g%Vh*^%A#dsCnZz<8i;@J{VJ)lf{%igf$
zuM33&1MD>*R~q?ti7UBF?K~RCD<Ps(I+@2SS(OSBXG~9E8dN5omWy{Xl3^7F!;rwC
zcP<inGuJeS6ZkR~>Q}OxPQ6mJt%+$%!SxZjsYJI<o8a$(Z~J}N1g8+g;r`IWvO*3v
z1tL)-?I@ZUR<ic_MdTGgq+w0SXR<TqjHI%F9zc5*>&tD(kr)LT)fL=QSfdq=7=lG6
z21+mnZnA@N&Rad3J4gRQ{=NYjD3n)&Zz8$JH&sD}iYhbsJC#|9ht1@EmA@47yi2IC
zpKREQ0(bUX2`e1$@wBJ<tksK$twe8!#hpa5oxoYgjC9oLo8|AV;qZoCOy$oibvB<W
z+8cv^xaXdG4x}OPh2zf_oejg-aEj03Bx$A(+=CI}QBF2>R?5l7V*yDf&lo6?ys+qM
zmHL8OR!7xI&M!fb)Tji1;D`aOIU9}uMIuph$b}+MBXnZ=kJOI-mgpRz^O`o<JOdxV
zzvbl|^kZF_b!eqXM5=!|%!(kH+7eJJBy7wrQcwDM6Ww;uw%AW(o{~wy3`IGOK{D{T
zV~w1<uWzEAtxJDt|A)7b-_qI=-cYNUz-i+xAU<MtqczS@d^f$;`R@9ixy52_+<R!P
z3ttz#4i$_bexn*d>i7~eOL%u~c=8XP*L9h+5mO9t{Zh$_f|k#8_f5|=5FRjWG2<B$
ztFZ<&Hg+<X$3i2h^Rv06XzDm+ZO`$}G!_uo8iGtNG8^Q`QOQ{|%d&7}p>YP4pw(eB
zMg;9-scywmtz6OpLoR=&9fd-PD`Fb^ruJR_TBu^WT{;+7p^}Q84i4039B0;v;T<tQ
zQ7Ks|HyY&%;^zVdshBgJE!B!sQ=RT?1k01mgojemX63T=`t0n~RHY21qr8CO#%yzc
z0rBc6H)d-SZdP+rSr=gyrBpQv`%OFM%9r3n66JDhYIe3>M|e1@z|ehO0->xuL`v;U
zO%-dU?DREWZuV^%tXM_vyNSnw1IVg=?2a!W3&n;H>GUWUJgj9_V22gLF(bffv4rA;
z79K`>0`l&gmKK**R$fS@el^m~nRaQ=E7@kQ8~N1>y<Ts&*TsLOx*5|c4thn$%yjWV
z=-!76RgJ^@TNQzMmGRL31s8k#eF09fdXZhCp}s(yR0DiYAAC+3^6D+HN^@Os5S^PZ
zNB9l|lV)*rg)@z>tu<T}VI=w}3rA)SKh1?!m#j(;NKI2%Q9b>rCeuswRyHwoVu;)k
z3L&Qy!MF904kTvE(3miw3&RIN6pRm|I+hhRo_XdO1onzPBeV?{Wue8kNb!A~ZlPQh
z51Tk(n^1iOmV^e1;y*x;`hhsE+`|RLU{xOZah%{1vS*Fio{Grl$G&<p^#0JtA(mEQ
z?Kr8BB_3<#s3d}Qcx<%}s6Rq_?^#}i-tj7zu2THZd50OSH{uY;7!L)s;kXV<h*7m1
zOc72bWS_Soi%RL2xJwIp_*iJ#(Fs~W{%0x7V|9Hv6UJCzM8cW=?$g`B>hCJ~{#|F8
z@A&#Z{kb0aZT&zW0#QbmYI{>(+}hIn&A=qw$9WHT1pDpF&atY(-jCo-`Zi@-`!7F-
zru2%kf2KOO<Ha`d<SqU5E&r6efFsmbu<h|PS(zWxX!Kw72=fDs6mZEVc&L3?llesG
zJj5u#3VvN7X*?p;r3Oz^O%{tAAD}D@4QimT8|hAf7yxa-ii(j&k4L2BqJs&qa?VLT
zYHLVUeJsn%xuOO47y@S+snpIpW~JGjgUTqffRKcPUzkv80g{0^*KAfy=bY$V^w$O}
zEJOv<Q-yLFk&M5Nph)FpVk?m>r%pi~TQ@8())QsPwgtLWC6hY2Ha}6{zcxvZwOX!B
zS%VOKNJv&sk99hpqRWodv-y0senkBv;?L)56If!}n5gAqAAB}5vzdrX$`J%$LKhF$
z)-d1EYgG+B+!i(x-w*5x#?o%LkPKEl)I>S}_ra;SME**CKF0zwvx45&9E<iOnzFn2
z<-`G<-PNU`jEQA!y(&&)0CH&T?PYLKn%81o{tm57p-Iid=r=&!JeTC=FoIpYLvGhw
z(2Hr6Zw>r08fCvNNJ|pVOQawRoE{`B{71e#+E{?E_#IjkNBu4!0fRgWk0o{1)RS;o
z!+`1SLo_^EU>k&+K)zz6P^%rR#x+w!NQboq(yNoIG%IljDn+IKO)MQK!loeiKy&5}
zLo&d9Gm*5|7(|l@{3UTnrX*~p(v8Ln9bD4IPOa)DObt-&I3!p68Ax#*$B}IninE-0
zoNVSSYQ`6%^v7yXKo?@WG58Q+jj8|o_=p=|QiUTYtZ~A&m|g-dp3t~D@)lZ$BrgzE
zMqKh{zuu!AK!1W^l=K=e(Q3mvHI2wT@;(sT$crH;j{f7r^VJ%c=faDpHG%gNM1ao-
z7d?mOUNCN8-Owet%}EuCpwjpVk?vT#P!XrEJime8GJ;+DZm@45A;42qbGOS}U`>Jt
zZHAzR0JNYBz$DjzD^q+8>;{INgHPat0FD_>z?7y$hyXiJ00D@wu4eR6qt!(%5yt_l
zd@O#_h6XrZVLO`nFYr$A+wd%A872Zzn6BwBJk_nEmS*DFCa()fqCVe`yb#}mtpUuX
z&Cb}$bk;x&Cfx_{JRoNSUn4~Bo|&nV*M^NEut0>3<Rl!k(A}<+jKHYw*zrU~bL#rI
znuZUvFzd*uj@0R1v<OGb9IIq8knmbgq-uyDV?|0W%Z(-z9xp)+5vNX)TFH36&}Ny3
z_!zF$g1!#Auv#;R)!rf#<lz2UjCx@T%>)!fS!TLgNyKdj@;~nkjw4A2X~-1Z$R?A<
zt_3Q?&s3ZGb5*jOZg^fJT@F5#`+6Q`3V>t`5jw+WoQgr=<%h><P*F@HPCdXn`1eTM
zh9(I@fP(mGnTU#$O(e7NRF3{eEDu*o6BTqszA{m&grz22oN}o+!p9~Mxd_4La4X9#
zgawJ1iD(k)1Ib2jTUhbV1ORc!CwxS3oOEM&ycaF!>*IqJd|BZ06#~rLSh-$B^xYox
zi6C`^>JGTS6@f4~4MUBTCpy$P3>-5o!;}#@EeI$Y5VhpF^Wf|%LYq8&Y92JKq-Gv`
z@WIvDxw2ZAKlSwBth&nTx$qNDJaJpgeeg|hdebN3%ZF}R*|!8g;3wSH(!P~j4lTz&
z@upXwI&x%sYHIq(kyE21pXO{Luh#z1Veq{YZ9$XXKtRy2&a*biGDtACZ6akl-^(O`
z$`i7+zXn8CgY>X7S5U(6bIkjyFz23$M$fo^0F@8W*2e3NH9=ja*Ker1-uT8hmJ-!(
z_5(2Kf3un>y)o{daq%zy=gI0TS@-YOKgnBPSxv709gq%rk{#&y7$c#JwW!zjAPv}7
za<?NM1T!21!C@F@>x+9=?gg(ma*G^n_cZ?3Tk?m)bDzc1=X*Phk^>F+FVyqf+Sa#h
zZ{li0A9F`MyT>>CK5=@1%`V?r=w4t|1?}Ybig`t7vHx|#=8rC2r84@LcGd0OpZ*=a
z_b5cfcr*UD>d(7+zu}t=%CYJ>^@otsZX{O(MVcUaAYGe|!J?hRYyyfVU;;cDubCi<
zYs`y5U9OgcSPL3hz~f}2F{$+SZN*~ILqgX3u>i!2rc&|sN_ii`a>crcfLB?)Pg{L1
z);v4^g9~bNJnZzBq5sB0jp%9K#@iRBn+gi<SJfx$`HnVCsY*3-F9b;1ynk1C=p%@o
ziYOOiHUQ}Z=<z*CQZOi-UicC%SZToYO=|iIzPg1^)*mRhl4^2zb!8$Gd^(1lw3M`o
zSA|BQJJhR9B*Kk3kH?jFM;^F0`TYKJ#dmoni8Aj4)<e*x8=}HF+=n2uLj_W~vj6#H
zEUxl*fc1FXi6OVJseUKS`rdEm%C|%#Rj1I$x|qm9Y^u5NRz0{IM49(Q-Jgvx1+V-w
zBR}hT_psbwbo<tY=2UDUf&@GLf>Vt|Zz<>gt-JsfjUs#ay8LB$bG#LOS0Lx>yD2n>
zvg(R}GdzjW8Oy4vJ&CErpkkr}y#zsm%Jm%j10BAimGK>d3<6O}o2X9sB7ZFG3OZFF
zi`}X<++>CZJO0G<r>ANJ;ZJnI?ck}3u9-0r;Z1f$*Rp;?{H7<NX0<$bPE===IoAV&
zXHJUGG(M7T2NEZm;>6hPgiH-O_}FB@3&&z%uK>diZNK}-lIEy|hjOSoI}6SeB-{!k
zzYy3wfxT^mw5rBPQ}FB{+-Y3c0htQEmdG9-9&lbfC1Cz9IL@ig4@3P1=cIS8oT8^(
z;4$|ZvDkjM{%|jxT6rfMb=@BxIpMMCM|~f*@7l|4B!}s|-#5K@Z;^W>c(AtsES#Vt
zvJ%i65P+-n#3rB_s!4={M3Tj5G+QbeTyCKO3EQiFYlAoO=5}YcfGC@@o%Zl1G!W1W
zJ?>S+*0~XDQ$j^1r~x_vdZ5n7;-rvQhi_XL`b%PqHi!nlD9;7KyJ`qJX_3nz$OMkq
z%fFUXOTGA(D)(kY;%QX!2O&X6W63@#_x9L@A6Qb`;@hXMOw^jkI&dF3f4-6J)(=u-
z%BIwZQUw^AK<rZu__2?!+}AmF5h(^zTI#`Np%0-;|Jmq)57;jm%U6Y@2Tlg0e_?(M
zu;(}>Bw%NYIvt<b;01lsWIH}H&^JDI5qmqRsH!6DOX179qrdzHNWsz99nfb7rnATI
zc@x2MZcYQ|zI__@$IsRGBS@qk-@(sjBPxACpQy=mY6{xUz3v&c>3{FGx`eqr8+s);
zPWfJbGwd253oRW@oxr%nZ}M?jyE7AfEI*xd8Xu<(n<OkegMksoOrD`pQjE2b&OEp@
z55s-jv4aV0J0yDZOTKIJ+rq#!lZJUSuuo(eHX<`(9g9ScVdbNWg_j{yS~r`<<hHJ!
zG3gfIJI=yrpV7~Fn`!nC=ubfcSqht{nBKc-7)``;wN9DgC5Sz|*bBjm2m3v80BvKn
zUl#VfyO2-sHNa}{@nhgCtOx3V;pRdiy7a9rlSR-<O;IYf#{eN1RH`ZXz3iC)aBMVE
zMASDnhEQ;H;s>1A`UypSP5rd26>6@ouD+fBP9VUP6^$m}v>q>&;_GipMx&NrayTX>
zaD^~eS8-Q<-@{e>t^Pb#2=SC_j-HT`_e|k$l^OQ#zyH4b?sK8$h`th;e6p`Z%Tbpl
zX!jMs?73IylKWfuTSw5K`Sn)oCwTh<JOQi-?F7&9g-{co?YFXz{ODSu!1Y&{kw1bu
z7iYpUFw`r01Hi+npk;6b^Gl~kdSL&uq>Tp>Kt+HL<N(|Nz|n3EsDn}Hs8!E<qNY+l
zPCJ)UwI}|l23ElA#mHMS@))_xSdV43Kd|joBZ?GPk!T||eoV{9Bc++||Es>9ZT-Lc
z{+ZGvgNTX_{$52Jrg?KB^OUfuQh-7<{!}J$vuQS-aGaXq)oPwm8y%s05PdONbCM6$
zv2y2ih}`^k_z*u6`pM7-LLUOQ#AgH3iPBA&Ytapmw~+8_h`mh5#nF{CnYR|4|1@aP
zz7ljLT`L3pLVdf&EyPN&076`BEdr>xJe9ZNa#4fYng%?wF92;@10HJvDElJ~kgo~a
z(r%h+bLdVTpvu7yhHDc@d2o2rip8wS!zrW;*?s;F0m9?;di@msP5u<{Y%`t4r>4?<
zxztBzAT<Wc)6E7vLVH8&jnB>NvIY>^U|53D8hpu&EOt#ZWdIgdjg)D27bC;7RYcaD
zK6gm|N?0OSJ5#G2W!X`lJj%@&|8f{B)BT9d=&F%%jWX!yoJ5$C=oHbKB!mZ2gZ5Jh
z8btweg>>_v4mmi}h7mDhjYNi!P@!HQKTUOFVPkP(axj>jSTum@e9=Wh+u~DwwlMg7
z<Pl*1!<>gdwsa#j7`bQ^lQ(wbMHuZgED!gZ6zDDCQ`^}$f$M%Tbb+xum@gV1N5n2z
zyhxFbNS;5CI-28%-37w~@9D_K(5A^X5(RDQ3D^?N1YG@0Gr_E-qb^bv%+6ia^Z~F+
zq#=YA{D62wEd$`!n7tQ+v%F%0gTM~h1wk5_s-jH23lx|}dHMPxSYO1Wf&dY*v1-C<
zw2{FE8Zea5qsIt$8VihgJga+TYu9cT%9_kVql4A-@IZn}3bz>HiFU)Z9mf_ms)d#D
z;o|knPxt%Vrysu*dgjy5z~$tZ0mc+^l#!lJ*?1iru_y6cHzO6OiP;WX<X>2wXvNG-
z!SIJQaD*K0jY7tZwNCIB{_#cNJrl1$hu1}mi1-SPs$r+5AM5qD2m=^X%7Nc7ZdM-y
zFOIeLVBo~vTF4T*5RxCeSMUY|Q_`SSA=VgD;bH<pEo8c=&_ODbO+I~7Afn&&bTa#n
zHT*H`;@TR6E{-3o54pL)CiIQ*#|8cJw#VbnQ@rEyy81r6v!)}IAA)#Nfea>q2jC-U
z@$LEU=~ZDz0@qEcBCZmJVs?+vMtd;Jp+I#W>?ODy!orPoSED218$FM(_2;858YQ@;
zV8yl=T8?c2cU?u5wmZPGKS6c#=!Z++h8^rMVpSfHs9N-Z4QIoIdxqd7Peq=XF6nz8
zHSK1z$-dHLReQYa5yv7hFmWQQ1`3NX8?*o!OlrT%NiFC8L@O3|z%anEx}cf}^ANE<
zOo{pfnWwmWN9Hl^9?Dz_rF5LWAtBfaj&{Yh`u7Mu?xdUfd^2Ms{P=u2zL*B0UyP^c
z`)YphXKFqVO#qY|$r4opva~>N8boZ3piw#Kj}M9Z{aF4;T7{1v53BT%e5~fgS5M+^
zHSSgcTN1Oum+uc@_4B=<S0h$Yum<`&^Vsv1VT3q#F~@;~V0EoS#7F2XlFQ?v-L?b1
zr2u4xzSX)Ra4l{0a_6Nv*kRqJFTJTr1iVdRdyf6i|6RRHs<*?VC4Sp&armb^_SpFG
zOL3V~-C$xjvT-&W&oom(%d{lOvtAR4CjeSDat9wcFn+A@&9*`Rw}Z?*qgLUC1PDSG
zX<ZqQ*b|VFM4d$YA7?QHL8P5oW70czPfI-khfGAUj=6}O0^1LC_)hm*P@E@nFTB9f
zkA16q7RlP)_W0xKy{yLSi;Jwr>PH9PNM(nCk4g~oW{wVi@E3joxMdP^3j74}61*Gh
zDt-p<C@X15m&vAT6R}p>`azj}ex3+NE1ZW@g0m1K&krcVu0utRd@jtFq*R0jEXl+r
z(n%2zK*LZ3ewQkx;qz=XO&}78(TA!B!hu}G)9IuJ?`PIEUN|q9rGm6+{A4fT%)mRR
zvTtsAE190J!L?F>z{IPNZ>n(HgYFDbaDx%-1|z6K2wNYCG<=NnzBl?P42bJLS78i6
zyKEosxs~=k9XtSqla^hq^>g$9PS04@3>IE`NX%9#)Z?!R+8x3QrkeI;g4Z5pN9Y2$
zV3mv$zguhg$7FixE1$s<V4*s7oL}0_JyCH6lGJH_yW56$HMC9=WT0!byWKBitq`J}
zX8cbI={-*El$RZBsq+^vUi@P5z~-ieCBUtln+J+ty!e7jH{etrU}RaVj$8Z5WU7%y
zgn<CInPYhx(v9zvKzPQ;$*NkQWH3J%OZf@w0xvr9>>BlIK<Acxls1M-UWlT40iMQ)
z3^au(n8w`P4RAiPBF@SNXk-gHjTXWrB?x6xm>RqwR0_e(No6X9d^}eXea-|(D-fGP
z_LyvQYVO$F9KBbg&LKm!(hG?FoJ_xb4YX)u3W{0YJ5z=eH5+4&2_hDxfXEl>-~(mN
z?56r0sLnl%Er5#zUrlmok>@lRsiBtE*ayNXU{E+LZQwZJ7_cXUwu<{nQ*0iJ2@1EH
zFR^Rdj*+;>!Sc{{BrGOmp{@LaShR}VYh`d?evPFiEE9I_Nf_YCZob)ecIAVF6#t_A
z_j|$JYia44fRcw3<x(S!#@)MzjgLR{PsjE2J1^S9!fOMdu*wY@<{#$jj5W*-4eweq
z(DOtq<27#5$5{C%QbbUT&VkyI0&@c0&3VD$p;lFf0@6B%hb#bTl!if#n}C90NYfN*
zhMKxSY9z_(kgSpwO{R%HXxZgn&=HFp&b~xB6N^W3kywPr%$0n7vh`z|%TrT|F<HuV
zy{_BpVt{-GZ?ld3Z{!Udd4Z8hW(U@cEsfId_pA~z2y^FhU$VUW^_8v0QX+ONr?AxG
zb-A>8CvC%tm(|ep)YS57b}|FU0VG?n8nf+uD%%@U!{_05U>JGVE;f~F7P)RW(Bq^L
zA!G#^($7M+dnbI0J`(yga1&VHg7lsUao`_qj0XVT@6k4Ta3-iDj`a}mir_(k?g{yT
z?*h7ig&G^(&r6g{R@yCqye1k`;_tGwOtITXUzka}7onL@NU@)1GG8cw1I78OQE9yc
zG8NoME0f#_Ji4-yaBdI7v-Kxzr^$8HNA1`Fz`s)M+i$<!jjuTNPheeL__r12$wJ&d
z3U7F|_}p{Pq3a+GecUsW|0nkG=aPo^aixCmSAOMJ5I;72FVmUd0rALHpF%sZ?3?#S
z62I)iW8gpCamO8SbaMZN3z_2%$9dqRAN{BsF(VsDo#UDtIC9kqjII@XFX-_@rh}y(
zPd@o1B7vE&Ot@9NqPO+l_r5n~HB8qr9|E=6^6*mdZ+RsoHS5WLKrj#^amZ4=NHqDN
zvJNGTY%&_@y(^c?q231^`*D1qQGWdK$1%lYM$W*OMW224S*#$z5*7DZ&-;C>KRNF>
z=TV~N+Rp+t{ysE#b2<|NdlZZr`ViBFoSHW7aR!Rm<D0~aO`t1LO0g!5T+p@rVX^pY
zAdSvv9sI-o*M9xNg$tY>A1-m0z%vY+(uY}cL47hB?ZYM5d^8eyl>7cxzmI~ivFq+j
z)V|0)`|^vm#C_c3LZ7ro<6!7KpeeeGnpZ80Q%gpjA-ANDoY^fd?V#?Qg0<E<uZSjD
zK4w_b=9dCSiGca_%gYRbr@0QQ>57qT%uXN5B{J)fbwLgqGv>L+d>k-6KR@jsE+ish
zWFbjlU4O(b)i0bIau=$Jgn5WfUO@1LLWx9aviF*uogF=ue;}Lv0e$qBnMFyK!e&xP
zr+Oja9`N0fOaCOkf7`Nl;vvk?kmBSRk(6aRjNt=dDxh{@3ZF$yg#IYMl;OoOT=b|}
zU7i5}5<N4xw-p2sT!bSTe~UI6rYG6znsz5ZtHh<Ec*MMyAzy>}7Rn;Pu{Sr*APeWZ
z2kVE$87g1R#v7{nSQ$C8CMOS04Z2{yEO^i-;Q#c~31!Y7fF0)G!GlI2Z4I7^CsT=c
zm;Q=*I8Oe{C=!{zQ{9L}RtQ*O=8;z>s%CD3lL<K6Y`oAZoJheh{NN<4#%2O>FHAH%
zZGf9!ICx^=@QQ=PsKA8CdNBj1DEGeF(tc*$I9h^9*@4;>okm@w8UTJ^#a^Q(&}spK
zg%kj-`9Bf^sG#7O0IWmrId&L2j^$ucc#NNrk1e+f$Dg?6lo5aj`q?rZ@v-F@{I_`V
zEV7G0lM<)n8PcL-tyt_}_XJ}`d@Al$5ylYuKz)OX294N|A><+`wo{s*uNO5o?y=c9
z#Yhup`7QA90!&aJP?Bho4<h^RQ;0A7(a>*(wnP6OQTIZs3nm+1#qZE>4dB2x7dyTH
z0_2L%Alztu*4o-I8){+g5JeC{1Z~tpW)CgO@!Q+d%1m3hI8GMIcv726Cyna0*av2G
z_h}0Z4&gGm+`9a_bS<^HP-57cVhfEIS1wJfSJca)o=RqJa$wddH(-K{Oqxj|Tdy`T
zQS*R3>MrIz93<WP;D5lC$-OZHx@#n@$q39rjgmS75Xx`BjGMw52=`<nGWZ=VO~sii
zY~;b}M-nHOHa0d$S!JUTFw8&P7;KJ?)!6Yf0ZkW~oF|(>iE6>2EAA~=uIJRUFsyiI
zvhm09;xyafjm`ji06Ay6j^|njTo54@&z|vMc*)kDHBCNmZrX#@`am>$21+?^dX8nL
znT;O!35l3`*=F;yb2;51N*TPstxAb&z8sHj09QJW7U8N13FSfP;h!IB0;t{9UL1K*
zvT!JU=U)>8zp_~5@(aunp$=Lwq%vkS;AE_JV_@+V5Mh&uEAW5SSerD6fT9tV)YS41
zHOE0yoW#U;Bt)w)m%-++1+2VzkwUwc+&B66X)4W;B8bAeo#OHXZg2)45njOVN5K}d
zcUm#!%v4l7p54#W{lnAOqT@uGs{>Ub&2Jq&xyk2j0*RmJqh*6Aikq7|Xb|aZRFYhW
zy}^w>)z8`p-X-Ql(y@swEOLCav9m^Na<V1E)L5BE;%H}n1!tW5vaSSZbUTy^HDFab
zLvQeG;4<ZVX)o&(o&Ip3aCxNV(?cNgd~wfG8|Gy)*qHUA0>F4+6vt_4TTlbLoMXf>
z*H=gAU>kkX<4E8?54ZJRZ~MDW5}90zxjZt^lrmKvS>Q?@VK&&M_v;Q@XJ4*M06?~M
z?i}01e;fx{ZTtGP?-vKV`!Su|Xf_seT}I;cYuz5pN82mn0_uASvHuYgQ1eFH5@$z8
z?hnxrL_J1YK9{JF3S)sfbj%O}HM6Rr5lC_mP3s;S##zLXg)69JP8S_37F9<9j_xEu
ztO_q=ugarI7a;oh?rxK>*B_n5al}ZeM4+b~{I`@5p--#VvMhuXs#{je181iEC0zQp
z#(0B-E$YU%p&Q?Zt_oCg#`32YC)lb_5ep<4@LvI4kt1{fYvF3g1dA@05x<236Q{8H
zR|~C9al=vW!CF0Yc;CXbmvYkSQX+A`Y@PJrhyMDn|Jt<MuJ`%kbc3GJm9N^(cBv7Y
zHxWrGA5-mpGm9xFktn6p_siBv|M=%V_c<H)Sl_BJ(NSX>5={s*BCscB;19YAkIDyN
zQG6$)$&ZA79hC7;L;rW^Z$tl6LHFWEy&`$xi&EC`#qaSV4EE?csXhdIYA?o%|F_i<
zYN<6q6qpd5f$A|TivCwm$1+0@i3={^@&fV%K;N3&%N3jp(x;#$x^W^HqGbwkqdWBU
zUr|Spj!ztY(Y+Cd9U1IgUUGxD`TCc6@y^R@9Jsv3Z@g%YE84i?DFZCTY>0|@xZS#<
z#P<c+*3$!9YCEpFBuUrWR=<li`bm;x?3{48yZdxybcGjI{<mFSSIK@V)hDKWY5SD^
z0DiQO7G;q^ZyYPgYNg;R>R%YHli9}nx)*uqLu+CaCA+N2JLW!MhK`?F)zt=i0D|Zc
zR}wD}L*Qywi}yioAQixv+#%Y8PPo}@29#wKkhJ%PH@u;^`2GlRraBr9zv@q8v%qWK
zYyu&aBME&R`^m2{^nr!7?0M{hqi^qJGH-nATi;rld@yXJ5e?^OAT>J|TjOo%<Ce-;
zR&J~<X*D<a%6L}+|3pIOrO$$Pd<h)-ZLp)fO}M93U<lw2U>TAU3Tk5{ShRw|qoWpu
z*nqjwB3TzRo~U$1cr8(aup0Z34Yt>^q)tS@A`fC->Low+_IM!SW#|u<FnT1#FDu!3
zJ71{O)3r%XH2&{Y5{Zw%<Zqh$D(J{K9R9=WI7jpa&oYg&lh4&6@HaKo`SC!)pF+&z
zr9|#5T1DG;y{z22yAqFEc6on^?XF8ys;_~p{@3@s=RKCS?_svf_SFOLd*A!AhGRHx
z^cB@q%5`Jtxqn1Vs!jFb&{LuJ!TaOG$d;kKWtQZVNN;F=gHatlMX{DWLk^UEGde~`
z$dBBlW*7^&e>eX^rg3)(I#hR{NP1lP4%jQga0SW+rIyC{IM;`Gz-1-sXJMN_3<bnP
zq1_h?8d7<1budbSZR^Pv(f<rANCKmQ?6~MJQr{lBu0AzIwugrqL_xL-4prOs;RYh)
z!2K$bxFr~-`}JrUh5LnFL`z-HcK%mC1;x0Xq6ImgEv9TFEiTU$@oc7M;x-e8TwI#r
z;S?5@MDa8}5f45Si=EL&Z=1{7(fnd2cjx*#R=z;}9!aNvD|qNIdG9$kA}t56LE5W%
z2kK`YLe`7W446o9NTSdz3O`Z}*f&Jr)>|zc@GALvvK1naYKhWkDnNBsaK*QL1UlYT
zn2c6%52prrjBPAMLx9@`meW{#u?@+J*1xq|hCx-+qm&Vyh{nswj8m9G0#y@>USs*Y
zEsFzzudf@!uKX&Ld%3KQZ0xNw$j-$tg-OKdn5EJb6hVONd^DSe*Vz2hAuuagKXc4O
zTN9CR$xuJ4d-9|81I8yty(78@<2dwfw~Ie3SqS(WaS~%6xIwq&^8tzI3N2abSx>bM
zR5n;Rz(@xo2?@_FYrk!y2lsKPUC1pz*?xI6_7<?Z_R)JhYw~rJPWW{Qbpfu@{T}~g
zS7K@#K9FjQ*Lyd{W3RGdBfRToH+D!QC^nezc3fpYIHSMki38C%#+cu+1N^q7F&pL;
z=VA`u5SYwqc;`2b&UI{c4xC^)&P`#O^%owZ{4iuKSY!NLuh?O0g7EOH!rI!~%gz0H
zXoaHR)bI9=<*;#+KFmc_qtV^Jw%-cOp4MfRpIo|DWpyS8dul%8JBT8B4|)tuCsPPz
ztESAr?rOfoYMdA0OP4S?j>E1o3oZKzbfRV<0R#LV^x>-Xj7$K4zOQjYKiR<MsG>1F
zBJk0chTg+;QjIqsJ65ScA++=|r;{xrsg_;LcAS?jAxdeba_rcfx5rmlsNT3S2IE&l
zu-U`+EF4f%2cd3=ADmJL7VbHW;IlQgJzljp-iklRcc6FLLe}INd`Y30_0^8roOfs~
zi?4w?zaxIU{4T5n2<4(*a~bqn=U;cf7R?>U-yNBFyzHeGA<L!)znDk`(W_GlbvC88
zAAR)EA1=k?Q-nM1|FQ|(;((ifTGl^9Z9oon1(=xAX~eba;6azGD$sozuLv%pomKmv
zQSC6gTl2LfAezq&K(jf<^~zPQsU}rvG5xx{hYCT@u&JHYV3Y3#5zqVGz%^qlIOwX-
z9)*1k!<WXco!{J4JvIxaf_7l*vkhNv6UPrJpc8<;vR9dg;n9@=+LAx%SJK?A(4o+u
zh5kJBm*D1T(-Cf;LI(Rm_YeD1^JgH5qHWRoc5$h+WyCdA)g2vhZdzI)TcWiEMCAOD
zwu%5DXnvK*9|Oc7Kztl92n25aMApIM_%d1_<TE%JOim59mCO5OLoEcQB*8*xfp66b
zC<UIX4?hWgu^Y|Av=1!tU@V^T+&EU%7=;uP1jHN%y6AWoL7IpYeOkF?)Ds!@wkCo3
zd<(AFkwosMefwbAn^;~trnYWaTCPtp?KDEYBr4UZ9_cOCaDoOG96iQ}@V8zBsw$T}
z$H`}J)2&HFdwuwiAU_ejQJ!zc=L09jB6b9cWg;X&a<GI-7Pw#5(PP|A&175+g2_fo
zDB8m_wTW2R1k^;5%gd+Kv8CmSI+o|xmzQpU9bqDmJgNl)z7JqZsFqiV1_j+Du^tw-
zRN9%$g!0}RkTQIgJ`_C(i9YaJE;JFk4v=F>?=liEM^45uk0H{sa<JwS{jWz_kF!6@
zJ(aon(MNC2KwDp3J9~Dm8uhksKD2!L^zxybv9t%2_w*x=oSxDFtA@uf#4^wl&17O&
zU!JUjyGkO!$gIlJTRW>}XR6gSbg!*+wK~(^bs~5fwCuCMJ>=Ezg_hv5m2IobOUoq1
z1%RgBlaB#H!>5s83%~+-2N;6nj9fVWgTuicA~$2k&~tr64cP>%*<cFC@_u|SSwL(9
zjA!;fgc-8p@mMr60iXNC;Yl-AEY<TGZt{88Djn{oGu<g<f8er&V(Yr=uG1G!)v>m?
z8qMzJ(Kk~M7IL}x%2_tarXQh{I(=tM*lrH9$|hH{NmDh8deubHmx5KbVx?<yL0|0{
zJ>ZF~)(p{eMKCO;buid#g}gXo0}3=t$N8!+H?i7|D<m{JA+K)P%OOBnO7db6r=63`
z>;mqobopZxyaz9+zN~dm{yk#Tq2Pi#hj_zy<&7aG%~uDa>*X9V%SgZ~tGD1@2$KtD
zVA-TQ2W)7_t))H(`q4lxj8ll95mK@`6Zm7dUQ=6o2%&qVe1t9<*6rbFa_EP4GC&4y
zB8e8#Cr1{b0gwQQA<@<YkI=N|H2+V0k9lXz&Ru`~+^pfRY0+bOUe7?i9sIEkZ1o$z
zfsGCGM(oFaEC$2qUS$fnrj{<~HVWxlbt>@a>5ex@mT-9fBW&#KSvJfE+;}4!-m46N
zpf3G?h@bhi`U6^^5OYk4PppEBWG1u5Mk<!f{?mn2OnoX98+4Od99>Li<-Y%U=`)B?
z_=V8F=6${msVFl6v%zQkwOA{YjS~(3d5npsO;I@KxDdKf#wkW_P~1B>8He>`dh%}n
zIjE^?`Om?*raU+w+A4svFmyb}`>d_~D=S-Yo6*wYjfl#+g`Amnt4EJkysVikxLFn1
zI9!T0-9pxCAeWb+SNtLWywSMwzRLA-AExeQ_jTj3oP)pu5HKv$#;T+)?pGf!xK{R_
zsmkH1iX6_FZehLR<#JwSz2KU;!x2?IT$#EjYq^EH%}8UGs~9WQS;X_Z`#ZR=o51o~
z7?>iCMEi|8@SBGGR7!Nw4u&gahhoqJ_c324VP8B6spY;<A6Zp=-jysVf3T!$(8idE
z+}uRm3A+1nj5t=Tg$ur4Z3m~s8pWAFd>)2qz^Ljne+L%#GgH6{DhaPUh}3__ti*9R
zsoTT@s&<pOeBm(Ik3;m}d_b-scx!?sq>zUjL5bRKA3C)3)>9{|@Png}T&p)$R?diX
z@@C+s9PBrVUIJ3g{jj}&9*SXzZI4mNz@A3Fl^TtI9s1SK?~6u=xDccRU7<BTVr>>?
zU*H;0C|3nKZ6PQvA5U5S%27bSm>S3d?Z#li6-tc^`Y}}RifU+;8V$L!S|nGs0`^r_
z3UoN2cp(fO@D<RS9*_KRKtT_KA?>p}9z#xFr~6nMtdhnTVQ-ygt-}6%`TEud@C$UI
z5U-0~Ebaj-qJVhE2FntUc}3TSPffrM>59qBFF*F!V~y|YJg`tI6iN$vtNMWgVg@PP
zjaFeI5|Ndc<?@WSzaih#<5M$JW`GsVTyY9korP+m>KX^G{Y<{gl`j{!R0Ccq+p06m
zv^=s!jA#8KJuIV`IZ#O<A7G2GB=d#71^B_6ItAP5e$1J>L$AeHqkBQ@v7(bx0gDPC
z({b<*k*^j-7K%CAP?3`J4g_k=SCF!b@T()ElA3Y~Vft{AQGvjT_rmL6|N2<z#&&sP
zZnBZQ^A(cB9gYsvT0ESL|MU$<nsB+eiG7y3GmhvXOUt*u(u;lwOAw}*JE%u~wtF}Y
zhl_Ld+|o@k*i)u&`tQ7*9y_EU>!$sgNBcIsoahEqMJlTJt;-NKe;v*JV!1UvmmxuZ
zHS!;3s2>Jj2#F=sfmCrEIHTEwo!?)jg)dwsYF;Dg2NGGDXoN7g3IIe^oa_>Jjotv9
z<{%Wzz{fk;Msjj8*~qq^2+9VHaanbDm;gw;09YLbUrqIDO4WK(!d5G3^TM@cUM+jt
zpn3aYYkd<|rul35Ku0wF;1H>yAzJ3LFCZ<g&4j77!KXqt5%sss+F%Z%BbMnzcdhFP
z-Sc8isEcWs$Kv-!fX8fe@Lv&S#@pvb&mnj~3vr0D4a-O_Th?-tYHPTW(SOVu{1O<!
zD@rZnaRd#nT2|F&<={yDc0A!=9a_ZBBSyjCVifHslHO}gyOvaue;E9wG<n%$z797b
zJm|7g>(pg!%Ts*2E942Cmj){kQ~>pXEIAVFb#}U7dRTcsbTDCyhzpJ<Lj{x$bLOK+
zOddKZ1hqJDzI|)Vh#i!PnC-Jl5<wRjY?ODH7icMI1QcknscDe71Ng{p2R?tnb^o%z
z<Lzv3OYetUk~iLgYXe>zJ0I4;wau%PqmE2Jv4gTNSK+Y?c!hI4fk@lwt&TvV&#Y=X
z;?o4qKuk~24bTXtP~_Pvg&IP8Q9Vl$0iW%~VxGGgH*MD`eI#2c=kCvDpDRyJPUez}
zpwxq}pEz+M8bwkNy;Yw=XNh!y#R!YZJSvu+%VzKAogXQ2xmbKr+8lfWH@faa(Wu_y
zV<M9G!NRv3g$?(!p+APCSt!!CNroF_iO59OTI&)4fNrbX(%M7OnF;zuxH_OAoyScx
z@t|%)b{?VhB($2A+ZLCWyY>`9qG`Pc)9fOqxUU&Gw1S|z2#o8?eu!X&#{(0KJru2U
zm%$F{Gd!@ncxVM^pJWx!FCwAb_=&7@S%3mB7Y=79C$r%&=&HRCKn)z1h;rM-M5WRc
zXE5<<_ccHaJ%tpu&;b}h;6wTnp#(G+eh(tXzeuGQ#w|Kf36CX>+~Hdwog!nMg2)Ci
z=Qu)ttJPGRVpi1iO64GqA){nn`S>{etH$0DmEd><NV@VB6)4A;Ts(dZSg-)szh&GA
zSeOpI3_R(>q3;uaNJ~_JWKZzp^d>hi8YFReM9)vP!mGXr3ifn`IAY4?=o2KgK+2n&
zW?KNgAYi}z(oWR>gXNVDdc$bJOt?A?-=LW7v`b~I*2=>H5FQ@iCl2t?rO0Sja)<|3
zVdu6OF_s){YVgf$qTwO3RwJ6}zt|yIje7lHyj*UGg2UlrAXP54OSxPO3*Zf`LWX01
z9Y!+8<ya*kUC3kXN@beQo%Yh*Y$BS$=GsSr?v=P?4%GZC@&bl1!vZdU=otydksX+`
z5mZ;?G<Iwg@eMWJ5iXS-4viit<nE}gqw*qc$$Q??ml+{F2$*YDDv4s7pN0>YE9k_*
zw#}GgEi0i{V_U7Q%g%oYK3=<v1d&n?=S4ZY(kXNfCmbA>eN22P5;-14Fm6gCV3fk~
z=<zE{sL#Hl6#PGIn!h77Z4Q|DM&wm|@H^_g0)Y|A1K{A|^7si2$-#w!-5J01k~%7p
zwYwfz5^SCb4x{=14(*DR(hT|3sQkq{Y0Ar=%h{K0CXu*ySU5Sl2>BBFnBT@s{~UVd
zWav&<%D)#nnJrK%vW39dKE5VINe_Smfd)xWU=Jd%T?<xGn|)(132N8Yz}2<vcz^8b
zx2vD5X&myR%aXA6#%tvZH3D5cUN9!tufEvk-aNb%CJ=%M3P~`aC(~X7;qbWWzu4ie
zFKK1Yv2-&*IIUvQ6TEnKFZCH;jq{T+%Z?U`^=ys&^5Ab0*>2i<37a2$@r664@EY#|
zez*yC-}i?;6Z#90arHDNhf6t^*Z}-T;Hr)@fG_!q4%5y0{DsyhAlWMA)^>->5JkT)
zng;EpL`(m;ybEU}A>rg`z_<Y(5MD~S7Ca3kS=&|-ehTsxt%xsL3(|IxJi$w-+Cvf#
zL@hjUfEnOTt~0X`F7F$SW1Sq_j^XAIbF)r#ur=!@_a)MF4^pu+vJPbv!X=0^nZPZ7
z-)DGR!z*C{yOG91`Y@7=(Ebkm7A|?;bM;F`_$}AEQ*EW9rZGQnn6YFID&W(uTB`8X
z_=@GIb>D+Yrv?Q`vXD(aXd(U@V(RrYt2x@g4_pZX!h}=ESZUxp;gTLMq)-q6*K+kM
zSHA{YE^s3<^N&HM`Z;6_{AIvR;K#~YT@Zs4_OJ^P3utH|hA03+hGiHkW-%ZxxOTav
zv8mK23~7iGEeJIs{*U(cqE>!DNfYQnK%%rIV_3q$J9*t&tKFV<(^&AZaN>AsvRNso
z&fGD@;~Q=~c4~d)$}^Rly5;0`JuuN!zGFL7Nr|@x6-w|>(|V<0iG*8Er4yA*CgQkG
zW;#`>*@<u?9JQOZ*bV7o&9251&19@-H>>f>&!_T;i7-`5rZ)V6J=Hvh-(h}2KhR(W
z&m5|@f&z>cWco{=hZOx;ti3r5DL9CFw2F}-j@I*g3bdW`7NWc;$~AATB`nQ3b3*D7
z4*H+H9WuI$XOiGxZ(Ce6#*bA47`9k07Yj7S1xJ1T5qkT=9f#NBDs?OojvZd#*jQgr
zFfB(M$K6Un-&UxMj(omZ_dV1Uer^HsHds*<WQ^tqX$2f#bL4!c<x3he!-ATqkQoI%
zp&Cb@7nT8vKygA$0IHH_;M;IPFBsSF-xm$*R}TcnaCG1P>y5&vVUoQ{(gCQRqO3Fo
z88pC#qDU0BL-X+F=2PR@2cE)^`8IR%ZJp2wtXc)<p};jT!c?0R0pr5&gxrER1jCFq
zQ`mzuh9j(uQBkBtJx@B62f<e_=dUz@6oWaU{_gXi|2*>?{rW+~ML2xq-n9b<Yp;WK
z!K>d2Kbf#G_-bonB9l&MCMH^=qw%jAYERYg-FoY-bu5C5ysmcez}me>4nxa+@Yk6U
z>GO>HqJ9s~;`E$?25WS*3pDy80H)lqZS>F=N%0Hr!`=sMqedSigV!9LBS(=2*_E`?
zfg$LN1KNX9ByrG$hQb_(4EhgykpPD`iEnHRSfdR%(6!QV830`DvKbPkku++fcyKq?
zQuvReR7qn+CkSxU3nJd=UK1SwQAt1Hxe&&}Tt0+2!W68SY+=~9n-|ALDKn9~MPI#f
zt~gnR5>Qw4r-#=L^b$cB{ve<ZTqs#biVgS_CBObbK8}wKBjUjDaiRp-7?&jn|A;v4
zjMh!8huE}O`BYs~f27}{zkfv^T$hR+0_@V4?002-;wjOhq7wm)=qPXq>@F?I6MTg?
z77l|vpQulvW(B0K5f26aLwa8nxW`T6q~xnMs7EFI|E-I_&JmT*&!osAxecs42h%oT
z7q=3rG*+dB?TJh?h77>e`NF5M;gY9=xLYhvFL>S-m%^PoWJ1y%9^wnIaw8sZVjVB)
zW|CH^)SN69oqBE)j{iUiM~phwlt7={ES0Pzs$`%bnW<E)zp$-$A@8$#1CzjPI{HZP
zaNw`;YS<LUe2D-;0Z!GP7Fr%7?HrYB?2L}o9y6m0>@arc<&|xJ5v3-TtaY_!IV`HD
zL(J(xR{Qtea{BDq)3@xyQar7Rn3_6r__TDdK8K2#>Sg~bx@gNk6juuEs-R1Zg@v1L
zT39etUx%ynXL?^ZuOAhZ>6h4Mmj^|G88kTOJM_SW`QYa}MaWZwGCqw@B`zheqWe7^
zIs*U39}j(1cwBAA(H#dDp&89KqAM=vL&3WUlEofV9(cj$Fkhr9aIvWBUbI2vzHN@V
z)rbro#E#>jYviRzH!uk8!Aght!8`o&f7(%75hFH%C5p&H1xBx21_Ow7a|kyZsl|-Q
zd+6PN)$1R)#x*0dy$!z$usHr;L)@k$PsGy0fNcBwO%)2p0HiToMzS$vx~n3flfku~
z$p4T<fr`LMHhj&i*SK@5u|fWaJ_UFx*5K=ip4|ptgzpjT`<h>c%u3zUQm6-;=qmsN
z1z!_M!`TS$1ga&C+|4EpAW$v=FSR@L2jLlO1NrQ_hLIwI(-#}`40y7&wA8AK7X;KE
z$p2ce&n&gtEA$vJP7f}qjll()rPKzuJxJW#%VzuXtIrsPM|l++)nx;Wz)TGun~zK)
z?>Sc2@}WC{sjD9x1pCvxpZ|USU3426<TJh%D^Q;WKDv^-%M1Ag9TT@8>RpE%q3;u<
z=MT2T@HWAfyvB+zkt4qae!RkZ0cK~J<~Gx-zFv=~T9?O)0R_|JIN|_DM}uwZbEwxb
z;%Vx9IQr>?lJ0<qlyn8V35SZ^BvrBMFRAJVZ&jOs5Zykun<EZ&bTo8S`B){DuLa;$
z%cm+IJI5wJri__o8B28g@{FNA#<3Xm*?cqh9|)WOA=c!BefT3k$NoO!_xFfn377{R
z{8nRt%UC#ck4F3)f}l4!EV?J5q$Wb+1>Na93WkIoFP)y%jR<-5J+~<Yh=g+O&bS~t
zw8W^d5wRQJDIFmYep#Py4aX<wbRRcgeM}p?<;uS1gY~fP+2@x+@@CTK%k1;zSX1*L
z0^jT4o<}wZ;PDZ^L0z+O8rQHlw8?FwC->qyMEig6{UZ*e7br8YqWFK=7<(~xD4P1W
z_&i0Rx!hAvAF=JtD;W#WxCnV<k;n6}U@y{gtG&F4wzOoRt!hMPL}uh&ZJLOP$}qbb
zfekh#`B*KvueCy^dX4F{5$0s@vluniAACaHIrv0oR^5FtYW$bCyyY!1?>EEA!S6cG
zM^cEMZJC~9-JXlSAJaF1Y;vC4X513Yv;OM7W68{WPM$nzd&y*L(lp=NoSvpT!R^^H
zX*6ZjZ5gk&=&6|lCD2kY0dD|z3>_GzI(SR%KhdI0#1SQH4tG7F&ti06T*KIsqr?^Q
zEDd~|*Tiq0>_sV^ou1C7OJpZ_UPw1wtY&GX3n{2lQ|ZY<!%Jj5@46(y%O;PS*@^x8
z_D_5u4sFZe(E>s%#Xo@jX1I|JBqtCO6pKUjS-cy^zl)_($w4E@<cU=3#&Dk)`*~y@
z0S-ze3%TU;`nz|Gp4tkfL77es`4S0;ho!2`#k3eTFkHe$6gZePwrl2v>NiTQde(Ec
z>(5*f9HbhY889atPN&~|kSkjdgljXA7`&bn4c>OCVCD>@rjEDCYlDlbdoYoBGybR3
zy+mR&o&G;qcXN|h#aUSnB(BH6IH8_~F8GazU_k3S@jPA+d;IW97I2iN3>qEj5{cab
zizRJ=t5EuW7dOEP=DWL1B{JF8zI`pEZ>b@uemD$uUpu$3v^WWnbSD>==JRdj=b>a2
zuOUlJYv2CX%y)Xjcm4>_@VCceaFZz^1+&eU;>&r-vI?gdRwWfy<)p`#;>+1c){HeP
z<-yKNzatUj9kJN?mwt<WNDMB6kx(5t<|FF2;Xk+n8unU&R*thCQB8q{Q&9i_!%uC6
z_}ByE5bIz`7VysQP~j|HqxlH4Ro^1|J?fG>Kc`yL)AR)9_MDeWC8p<=m*=JvDXbd`
z8}XT3al4qCPQdf|funHqE5V^Dm5Ox{(8F1J*Y($5zl4?4(Jsqku~J1nI+xGo@^gB7
zn738x^$J2XAP-Q=t0ukaX)jsjbAAiKh}tE6V;3G1vC}A6IUR$8Mpr+iJmcfu-v;jb
zH2Q@;Ggu|0noZh?#U-YQy`HUsD}>Lx<3mM_iRsZ^L6}!?G5mP;#p+s7->iF<35oeX
zT&c`^v8eqPafxYYO4>-L8<TT$3-kNnL6lDKY{92?%nXbuVWkpnz_JakA2?8F7>ef`
zBT25;o^DmENb{YjR$J5U9v(J!Pj|dE;%f8KCD?I~gO@o2i{hiO{X7FmAZw^4dyq?f
zI6jsiNF0FA(OX*4@o~!ljWs#eGCo32NnL(AuXnQJA8$*|R@>(w@gOt?azL^G^4%iz
zv4AJ`l>Kq&^o{h?<RL_!i`e_t54W4O+8Tm2h1XZ>h!L=|4lTH~R;xAJhu8O8h_9Yo
zsZag?w7my>oJW~IzVp5_@Ah8Q<&|U=+ge$&Ey-~lN2yLFCXF<LR?<q^sO)BU<r0!0
z2nQV1bV3!KP{Ywf6Xob7<ml}L99(EeIC@k5-{+b4-C{fC_<jCS{O-K-&O7gvXP!RK
zMDdd8H4r4Qy?tb)y&d}p5PGD3820&$PJcWZ3|XmUx6ggkU@uOhNF-A5&&DYfeLJ4z
z?=*aFUw1M^2dKrxy0P=%K$nbm?x7gFePZ7kpeZIcVS3wRt*tR=?G3jHLUYDG(MQpG
z_jE>O*TBJ@D6qA!uMHdHTiZxGEjAl${4a>5+6QnB`HrnyTYLdA6%F{~e&{vg{y_AP
z1MR6;z}K>M>kc|ctVUBw@0-IMX`dV#K0XYLm0mpN5LMv~ReS8Pqx6s-VP=RLW2CW^
zdH89x@sL1SlOK9UIin;2)(ig#D|*&hw(h#7FCgbkz%SD&qAZRhHT~9(C}8@uM!W}0
z_Zr|2opn}!ABmsDP?DpY&L-ziWL9a~hQpcDZ}mGwwl-{jgZZT)*O{)%FgA|ug4DpI
z*QUkl^i7jlf-vW)8w)(ja*ZaqQ9vP$49SJWkN2#iQMQCJOuc|pJHAYg+z>)3_&42d
zuF)eFVp)?v$$WS<Kr+9<Z_`p=-+X|F1Mi6ld}H++2rI(#{QpB=h;^jCbp2oZBG~e=
zx$%GL%>;4oxB$Pnq1XQi^MX4-XU;?%h|3|Z&$!4FN-Mc=lDLgoC>RzgSv8;%snHl`
zydNkOckYpXA}cA-8*JbP-Lc#_fSRMD;nBz`65{c2=Qe_yZ);K(4zt9jl`DJV31j<?
zi5<OdPF~w?z?s(_OkO)QG!*m=dBPnaG^YE2*SH!qsPl|1i@5JRqjOIGi2c&Q=2q`F
zs6wjb>rLG~+pLKl+l{8ElNWugSDt}W0}qVC<I@)$3cF87ggBfl9g1Au*Vh;Ci=U%^
z&g$*O@#^H7&o&S}R)dJE=4ceLeBls<b;=TsExgm}WveQU6|1l;GU|do!`h|*wm{0v
zKm&QRVIBiZMaA{d3?pJ8oovv7kp4J~#~Y8gH6cI=RIXlxyhDhS*)#k1!wbVdeyfFZ
zOP;=m(r3){B$A>RdnWGjMx#v~ct7s*J8w{>T!E*#EgnauPceqh@WUNxeDjatfZsZE
z58dn^!jX~v<9)pV3}017I14NG24J*h>||Yq=HdmeS1<*Fe2g_q$cuOPkxiM{59ATf
z(td}mkSBP7Hkw#xM@f(SCfJ3DnIXq`Sc%3sEVX8-f$WHgaR?EebpKy+CNzq|#)DW?
zf*3<g(B99YNk(JM-?7o-MjGe7&_dt~42fcr?ly5d?C5Teo}|rr%cIEQrAVzvhr`V;
zal6lVgr4)=?w2%&!)YKUdOYlyT{c(g%k-~WJ%<Hg6_oQqR!BWUPQ5Lw3*1o_Cq%e<
zp++Oc1s1P{H7&xio2_O6NyCA;)6$zG@8K5!Nw4jO5X+8V0AA;*_JRb9Nq8FNz*3St
zsQu;n1CX5n2lQT#_#Dp0fhsK&3JhF`Xpq)c%ers?lr#cK){iV?Aos$7Kq!R0fX5$n
zowD%p<EJhTCETf=Ow_*DtL*Qg6^~fB+Bg)p_1Ppd2C<WO3!<qLCvGvM69pO{>1l$s
ziVn$u$41X?4k$<oCma`i&lON%y#sAi)ne9FO^w^D13D(GmGP0TT~Q0Wmg6_JUjM`C
zmRr-e-F92+^s8{jVjSC+t>|?vn^%yOqCq{707N0_SiBMeoiF943;N^8m2j9ewDog!
z)qvyVl@YNDhg$+flw8xL^>t0>rHM9Fb@^dL&+COgKkIsm>!s*F`bSaYN&Sdv<}(zi
z1IgWXC<Ee8dd4-6($72#_1VUQgg8<d5XYuMLq=8;Vg$&&pAHBdw^a><N~kizjl>cV
zQ6gm9)YQjJQXihNx|7(e@m<$ne|@VxLO;Cq)>~Vq)6)ng(cZN0KGTTA0^z`QG^!A@
zLm$OiHOBCLy0f+M61&{i`yg(35rl9H&bJ~h|9P1~!|d+r^ZFA(3}l}Xk0Y#h7;#1&
z4TM;$jw+Ax1voRJtr?M0B0ggR!w-=taB)j}TiYcz`{WF=*VhAcz1Q8befyB3!9Q*1
zaN2`V{bpjly}22oL~tGte39bu7{#C`nhk}USjAj$++U4=X_?YuBbMpmQgsNJST7qf
z>tAL|X)eHnk_<_|E({RZpN1&<e5-+l0Eqsr<2G&o+t7!F6bYRbPrjZ|+OB{qVV~7?
z*bP37J+cQ8+xV%7M)gWS8)Oo!H((d=8%@<;ZQs=UD0lNlYqq<CcxW>vy)#Z+h^8>R
znjkUhcl0nh?g#{0wNsn<JN*PyW{TA%G@@X<iChr?@(tEMe*?hyd6h`Qi<dUOdE4Wm
zK(syOFphU(uWeFp#R0kP(LgBP;(2{XN5=>upn&HFDXG|cnpsmRMIzbRliJeR)YF+v
zK+y%h^Z5D(uqKy+abcv>9RN~XOZrrzvy*nawIl-YflV@NyH3SIv5lr^#2Ri-iX@!p
z0<or-gM}gTfq?9ODtBmd@(`8u2RS|*oxaDR+&2GAxvPixyN6Z<Vt4fU$Z`SRo%Vwo
zPr*47_P#dkWz?R8nsn$i(GLsbt!W0)1fm8+KbT^G?P=WViMb@%(<Vri9&dtnkt*xo
z?Id0qYPs#Ume6HMzrT0OHf+jO2;dt$c#u*EYZ~0PWoT=!KOFXN?H$5%igfA^&~pR{
zRslbr_xk<G%LZ{Q&vn6ze~Sn_p=;a3yHf6<A$RJ|9fLdCt`1X;R9`C&LeZ)4)e!uL
z&N^$Twbi^jOf^z{t*w*_;GD4V)opn7&T%Lw();KbtbOTGT9O#24S0O5xtd&Q=8r6j
z_yF{!ZC!D89O(i$Li_6;=}L!uK4X_krMCQ|GnG=iC~E%Z#9l-n5{Kh3N3FqQVDqj{
zXsK{!N~d@n4$@jD+=5R*H{|GVm#|lkJpI{UMYCSJ2qOcJ>LZ+JQs9}-SIw-^)Hq7q
zF-#8;ZUSt2gTR{XsDz)%tEb8kKw3o_wQ4Sv0#ebd_SJrg?CN$gl6a|7+ib-G9)^OJ
zl|{-X{PBrgnhhT77rw5pF7Un5+zvL^2Schi(CU*2GvEovu;<e2mQm9RNudG=a%V)L
z{1xG(-=j?7gMS|O$GI&l7HM+(jJ@Mmy&X}b-N8UK>~D)TC*zT}cAV6Q_ykx#V>@=V
zY+fWD;zX4X`~5ftnlIvbD|`$5UO4czBYv+!i=@}|S!h#8VlNlsRU+a&4ofiMFDip@
zJ_M3PdvIt1nuA@UZsKTSk<RX`_wBPbza-HhUsKp0OLWOrv<86*Ox50znW>!)Q%k)9
zM^O_)jtk8<d^}u`Rz=#C#Lg<0KFW)?SYXQz1P4OjU88)!*NM4=l!02=$ppWSBUSBw
z4Y|4uuB-Jo{M`pPT<d^o{%&2I#yX#$-8H*fUG7AEO%2u7KK)Ql!&)L=a5wgm=_oLL
z1d{mYCWs>t(tiLb`~k}BKa9gT`3?|3_xDpIUX;^#Mtm(A9hM?;DfF7n2*vO71^e5(
zEwz0IShY52fxN*~v^jO@HYJBcp@Gc@>(y}LLm}GkyY13cb2LT4fQ+`D-Ma_3DXY7^
zKM1v`iZ(-CdTB(+;b?T=FZGHnqSwF+F7ZlO!f78k@7&hQ3A#7!>i{8{`$D`jda75I
zX4uD%Z0*{0iD4FT5S6+AIw_h4_w4EF5@P3jNmcsZokDbV?GZ1s{ExYqjQa=`F0H9c
zq&l<J<LmDmv(6g4%kRHy@GNVruOBiU_*wUZpA~<BJ~4pPsINvO<ktdwU<CVF<Aip&
z>u<1Ih(?)ZIvQ76-`s9S8+eEQSR+6!XPiKi1kq_Y+xNMvMpT+I&ZLK_T2*Wp<q4#h
zdFuLfRoJ`Phqv?;WA|a5nzg}QDklu4W|+f3J4mq4YhwK)zPLNk7DK>bo!^OZF`@ZW
z((_`PLlKR&fwux`!J;K@uDzWOyL9qRiY|}*Py+d)K}9Pj>nFJ3W}V`qHlf;v*PUw9
z_5&P+5bFs9o8!?JQzM}puC|gG%WCUS`QZBcalm{Ut_j%kM#cNVF&)AwfMCFBG-;|d
z1t6DlnhaVr#0m8Sz@UNv$KCC4&?v!WY_~C*QXp=Ehg8#q7#e!aUGVA>;jZ3}j^GHk
z?8I7n<hItXXaX@uRi+8!Zec%G->`+fN|9d@D<P`h50MwX@P${LbyoX+oFp8L`CuOG
zesXh5a}0;8Z|Q@gIM^Ssyumb>>4SWJB(W(EI3h3lN9$*u^;09jd~|{pqMVVWVDDB;
zLYLz1!|^|WEBtD9qxO3Vd?OK9KOjB>oPDVbhvgX^xB~N1x}#ko{X7xe{GABGS;jyp
z8&Dj>++^FfICLCzCUo=NL~ZUSRyI^85Bd?!rISyAatZi;X@v)9CT4)fvqynrBsO}C
z#iLP(HFQV3r`|+G8W9?5nizVbR}$FGE*7FtgP7R-ZU2D-mqa4z@l0mR-J2gf`|PtP
zMaS#l<b#v?n036{kwNaJabFkdvpWEwufF~E+fT<?8piF%jvWhb*)o(~J%1m#hcvT)
ztByWmm6^gc9_)iP6tIM`X5D5QvYx9KG#P7>;Yq8Vnb~9Bf?-SPn8<o2w8xb803L<)
zdV|jLI$at41E<SfeSN`nG|_1is7Ei`T*t!VXlwZ0+yEVz_;NgP`anPYM><DtPc$Ly
zUt3G;>IqFDY#NrX<9piCz6)^%CQ0Hp@HSDjL#-!AqajZA^}|4+l~hzjvtjy5Lv1Ud
zG*g1P;F)-+x)4qfC*9!SC~VOMKV3`Hxk}_juujeh5y$yxxZ&jAwZ>O!ufuGSf}+kR
z5ZVaLeZvh65m!pJlm83VJwfwD^E`R~3B~K)CH_!g9v>7hf}Oh$BXy9+30#`yxxsZ~
zxx<~=MhluD1IkG4rXxJTnh`^YGWY|#*3Um!Tg0$`yU#ct0{4U6aRb;I;l5xX6}tdO
z7zT$%M~4vK$$ddAMPiH5cYzy6X$`dEm>`cgv{8?>gi7s1??wj)w;=98;I<e0Z-c~b
zB@#I9GjN;#(9YCt0R%kQGB`-Ct~j`;W>T}yU}Jnh=lh)pUG<V{6=>q2{jj5BY!R$e
z%tEdurERJST1SQztUwT@-MdK!K>|tACJ=chVmyw8jC%kG(IbS_JXOFO;o37e2$r#R
z5CpoKs<Cvn)LzlH6@ei9{Q<-430=^R;Gr~t&24SXG=N(Jk#O)zcmQLVjX*ylT)Vfn
z8us~JdO?WQ)|O0W_wG!lrL~rNn1cdc>ByM@!}RpffQ}qEZ~y~Mie&s8K!<2DdOYwI
zKM?6k<3@i4SKtH%xn>>mAH{wmYwPuhHVQt5;RB<FL1EVy&DO&<KskLxeps1PM|}D<
zf*?f?WazjK4X#6lW7&>j)OB?fE>45$I00<6vbw5pyfFc$ZP_L8r7D3|9gWhVdYG|~
z<P61wmL)rV*a_-M!(C>JX<2~;I!WD)e2&%={X1U&YIDq^q=92r=+zE?YioCRYpcJL
z-o!Z*2F}j34+@LuY8vB7ZQcvSs>kvbVaTTb;0)xsc&u{<9tAj+;}!bZs(^UE$tq`Y
zZ^~$}0B=HKpM)gQCP@5zLOPyE#M4g*;dRmGvD}lo2FE!U(*=3`uQu<jEjjS*vH3B#
z=AOX}*eRA3wEVz9x>%CXDa;>$A6M|ySRw3!+aHJQN{`+BPU7d$=wv8#U8Hw#cp&6o
zj5_*AG)qJlL%w7zaxE9Rb{q?tcD0XR`x=1ufeQ}2N&gUUJsFK&8;K=-p~VQ8H5>Y*
zj4t|u1H*&8;cIn~ae5y%z42=|U)xZJw#hM0=)@j=($8wbL`$3q6)j?-2lPiNa>dg7
z_O<0vmtYMAVFF-4?N>S0G@fD|P<<6t0r4M%psU7F>ab%py#wJG+v_{hG4D5wP|r3x
zlRV^x4*d1D^_1SKT2$E!v)+7LHS2y96(W+0o@;uieI9ypcZg0V-_`^DrvEnEz>6l;
zjJ~#>0DF<vU2WL4GY)L&5<u`;!q<v3&9y|C`5+H=;-Rz#areXAiHY~T=RK|QOXF-?
zaXIV+6FOQ(hxA&iONe!pZr%K{Q&bydu7T?pEjoR9h%}v}PJD^e5ArC!khM|Em}$bz
z#NQ&I8m7+n$chsB8E<mPKdbN%&7uofq9@s_1y;?9jrOCP{5)R2cV%}q8%jesO@ZfO
z3{&54J4gzwOT?f|q|=8E3D=>!dHYFvVr^oA*8W6#bz))?^OBx8^$-vk)s;?zwS!9t
zXHs%P*G+i%Z_Jw!9zG2|#8Wu^o8`7D;<V4PU``E7p2nnx5l2aFiZOw?f_3d6bbxIZ
zMtO1U8CNW~F<K>6J26TCgeFBK5$l0pr-3Hi4!!_PW*uGLK!ab?jqnE-?MJ6+2|p%*
z{ricFk=lu);bFYJoP%0;#W^9Iaug27uO+7X+8FTQ=I<=ag1s_yI_X~a2O?I`V+9*=
zDcS6d_V$+Z_t*C0T$*ZYiz7gB=toyyeRVjHx<beKo0MS#4jODviy9%L6JjxlTW~*V
zR8IFp3d8}Vj3?iZt*A7fyAd0qaV*ssy2=KvaSlm?4)ciGorkQ^-%HXeBN7@8dP<{<
zG~UE8A&=-8h7JEnb<vMMdXI>s9!zmt{Sa<GEs*%k(8Vcp2_7egch{4Xsc<p?zq7%?
z%_qmbl~^nW(FA8GjT^?5t<BB8=pHAE(>}yfXmh*o^wD!~3}MAXjidL2xOeuvi<>(3
z^amr+@I~kNd>(fU>a%3(vw_4H5xFQz??mAS(}fvm+udphs<Jw~u^4!<3%Smi*BuHd
z&={?M1VCf#GSJi0?iwsOKnc2x0dUvHgCs>cjRM(=;(CmBBhUv$Suv4C{5w>QBg0r?
z`btZeBqIw_k*#N^bEZJ<X(;(0b_HN4LB9w(ZEZiEG3_EgPBbV8mYjHz&DU&82fU}J
z$I!}f>Lwxf?jY>*bRk|X#0N-;pz+puZM@5?E`($gb`|Z0u6Ho05Es&pG-))hfe9Tw
z^73mE9^dVB5><1!Uu^!FswB05hdzbNd%VPSG)9uV7KgxHQVTXobev={y^rQR_9m)n
z_q2Vz(T1VN%+Uq*kd_RKpAaI*6^|NdT2D{IE<Qq*RE`WtwzZy+4b~Hh924v8U#&CP
zNd^n5`LvOdPL=BE-X`I{b)jDAU|yQ-_wMHnUu!yljYJ|cGE#MuIuF-vFn}Xx#NCrj
zO2NDGkc)xQC1{5i@IA+&+yG5qgRg5AyNKQmd31~!d2;U~3auq)f*<N9{NZ|rDJZV}
zAMx`vDyiMDQQ{SrT3Nr14ij!9nyxKFKy{kU>8A+AR2^`nF2&5liKU48YZd~lorjrd
zCIAwcw4a50O+yj%(WN(7>5W)+dZ&td+}JDxC6RBiuO~jdcg^pM#0ee76*jryrj2ge
zaD(3@OajicFg!!*AL8z1h||$SC%|D;<HXa)8S(l2ZaB;%cJ;n}h*b?M52w%D)7y%3
z;IL61e(kVMhGV~k8I)eY*@^K$oNS^Yd<+|haM$}#4^Yg4*UV1Z$HzVs!>yLpI!vvc
zq$BDGeoe%`T*rq#iF0KZTt`V}U{*>a$9e?Gh?P;OrQ;qrh6qBj1iKIpsC*Mt5t@V%
zgQAfNKOHxLh6$|_&q`45ZgB3r4^oc1WqW7y?G))8DAIqy!LBG&e=!+&oaJe!b%{ek
zH`k(}6nS8$EKd_ydn<U6=pOQPhFd7!=fT&!<~6YA8tze%zEboyh_fH10=qm`e2OZk
zkL*pihoFwQIBa#9wEnH9TcH%Wgg04Ciu?iDGTI+&GOYGp)Vj?#!gI;%z1U|B#<~OT
z;>LQyKt5Wa{h(Frwr|!FpZiLuPZPEs2Wz5DzM3)<vC()KSAaNS0F9lA1nJRq-XNj{
zyv-^ekSu^ZiqgWuaO)qqZ=I9_@lQV_j*(Xr<YRoCUBj@q@hZTDdzd=jl=Qh1doKh)
zsbRwyiX>l>3J+o9x<Z803-?mQlO`%QVgYRYgeM4R_*V#$=wH=ACgS9r96WN^K-?gl
ze#{0KfhcW(NOFMYG=Wbiea&f{JE$@T&&z~Dnez^2Ft4F>vp<1vC#cjwG6JPevf0^8
z(wxlbZ{x%==cCoI0}SGn3ym;0Ff5wET__l*%&1f*Jp$~zS!YRcK780qUUA*#(s3JH
z6ef%gF&G`$jTGj^u}5Q;9E=Sumg7COr=N6%&H!s9MX?tM(t-bg^N0#2K}E@wzo#Fw
z240oXp>~uc1lZ+vCbG$yHoQzTTsO612X>1TCl+1+UM;rmiKJp8T7a0lq<TUOb||0x
zJK|%ShZ{f5IbZ!WI0qP)QgPBu)iwtS*9Ofi{rk55!C4Sa&b*T&R^fp)G5MY^ed$Zs
z&xHz#paBaM2;5PmwWfBr#vwb>cgL={>Y#I2`@vXhbPLXxHJ2QXnDfh`1bv`-21nJC
zs+vRiiRmk3iZyukoL-d}Ya#HoA~R{z4t7x06r-M}*1WU}trioS1Uk|0*4Fg>s-|kI
zxN{6&_5=Nc>xNEkpd{MGycO$UcOnMTHB}v_qctYzz^pUPCZ!ZPSkk-_$swK{9;AX<
zPh#uAY!W#ntZAEIne!T<b&F%>%__Mg48h|Bg&a05Y)!Sn!@w62t_W28@z&H<#LNO0
zP2;**6!Sv4Q#%hqQRhZ1m%8-cZ-d;^BH#<DH&e!o4I|NW{*Ko0egvouhVaw6<NThu
zi3fQ4pv#aJwph5yHyo#UQA%N2=s)&5p>;;{rDNN<z_0JXe9aTbUe#=ncn1Gdj11_6
zjfz-rfi*&l4QbN};Q+ZGIzKmrKw3dWw28c<4AT$ffwv^i6PS_fqtOL(ORK7{|FUR!
zAm~5j4vZ}UVa19BmJ?oYG2YygTJp}Ho<K15j`qaifak!A#IInK_E;0jeHzH?1WX;C
zaD=?0uWE0@uEt*Mpxmr33Grcna3CC=_b#Pcn&Vgo90ITy10r1-3%Cz$9!s=$z%?Lo
zINf#N#ea{4fbOg_5w!9&H>w6ZVUjDixC(pdlaW|Y_u%G+v&b-Z7VGc~;3GQg?JDq-
zM{qL4sdk{+G9<ZUJ30@bf4!lCSsWH$fPa`?6Aw!8s5wo0Gt_PxJ*{phF21J;uvT{q
z(lmZT3C9v(FATk6>~%v5hG%>+g<Wy?-=DI~<Tsw-Xb?8nDK$aRrY}V&5f;a5xs97G
z8`P-xQw3*lH?WCdc-Ty)+=G|U=hmDrtko)vKoYOk2^!?wU=XXiiPHd)?6pP%!blsV
zNWeq@ra3|k9QaGkz!P9GXyWOf@s}gQ0rnLb;j6YL62{2LMKRJ5iBXgG^f@M;1Uz$F
z#p{v~OpJI_BJ|jvokP1XqDV0C@0XX1VZ%f6TBr50J<-=Ux}zKOPrfPlAQs@KU=4ux
zJssGrc?7_JH%kfZ^@Ik|-!Rb^{OUuaYxg+mNlRGh68$lkK_agBKrFPg^{Pw;dwn~0
z4qe;R;Wy0i)tO9(*|~H0n$4%Tw6r{7WMqr6rN93%Vn<Uzs^}#>y|~}Cdnys#AwDhx
z!=9Gy+egfuBO^Pb?XYZn28Va<G)FR-*3B_^$c^mTGhz<+_BIEbg0ZND4amdh$k^C!
zZ+H;>tNG;jU{4UDw}3$*nC)V;4X*_7LCIR!(k_~gsJ%afw)AN^i1gS*k2UnxU|4HH
zh*mQ^qXch7g=kYn;Q&?`yNHDVHVDz?Av`+H+`=#;!@8uHq;e?6q=L$$EU`YY_N^ny
zH9#+R(u-UN!yCy(uf<%tL?0X2$;htPAzf`*u1XjB+``qfe?QPYYwJK}^;Sg!6mw)`
z|Nb7hh=xKvJ!eu5n-PhavSIPeo*sDQVu)>dl;p9nD-Br;gGC?G5Dl2dJpqM{1V*|j
z2(f7Wb1*_$Px|mCTOAG8y@2#|?2TU1UJgS+?!fD+{>D1?I(@>m*ENZ6Slm13^e3lk
zUQSTKKJ+<Hc#)W*N1RHd?mVc?k5fZvN>4rkYV%(E_^=b^q;@hC;|=mOIQcB7GntJ|
zpI8Hi)TZ^M*W3#2?77fsl0B1VQ}cx<t7tVUNuC@!09YD0nOHx-b`@n4IB5K{o`%M#
z&lzhS%b6P8>u&~!%u>I;Hk@NL3_%KqvVc^H@H+6{I=m*@c`_OKQc6rvrR}im+Muif
zy5Vf3yPe(qghJ78?JGyHG)AlSpOouiyd(0XiI7$m=_}z{P1LEx=%W<E(<d%ey=EVk
z6s`B*oB$n|arBjO6V~fG_)Lm)4w(&FibkHE?JJ~bWt>VpJ?Ugvq@o3ee-uuDBwoVP
z{M!Wife^5pk@dLP1}|>Sob%6QcXxNfYBt={ns=VIyo+s#o{KVj&e@Z>xapj8&I!1^
zmr_7_@d0{mO&7z@ai|j#ZHr#_l&)^tZ8Tf)o{Ra-i_YaYjp(IbH}6r^zD;Y;$I}iG
zlJdD5APVHH3*EWav{07-i;OL7zd}~d5ZguFBVYMGddInvg70J?Z*ussQ!IdA0fwV3
ztsVQ@;XfOL;DSIEIEoRLPT%&fEs-Q0coGh^^!6AywXD4j#%8w_i#B&B?F0QZ^shQ%
zUHy^}cR)4Qj6*YhIODIQ1(i3qM1#@hj!4uBnyoDes)nQK+mcD_J?~FOBG4VgW9?g-
zuxY*h4!s=oHA3)D{|?pXYP@$ZnBKcx?XC=bc>4Omv-3Y19_lpy7`<9<6QUnRmc|pK
zhKK8LX>4slAbM<?!WJhS3f$7VW5-yp&+8BLAY2tb7>=|L!?R-N&ar(5&fGC*7=t^`
zJg{$U=T02HByPfKcb=YIBZJhizRVz;%r(rCezyyeSZOa~&Bre}qOp=Th>*-di)|GC
zo)A4rhdZvra6#J$%>$;Ce$m`y^bt$g2!yZ-T*%iqFi0DE9;SodkT-aazBwt6EoUe^
zsVy=S8xs?2_H@z@|65}>6WpxUvfytX!5MuI01veR180DZE@iKJc8nq;AOG>_+YE7+
zt@wa(=_FxV*-Qts#F!zQ7NQ`<7ii#-&j`SULc|lpJFmHB=Wqf(zFtpY^Oh?d=zG8c
zl%kV^UBX1TR~=D%lT!d`v4*llUKt~ApdQl<_V#vn8%B3`Z!b11;9HYUeRx498M#sN
zUZ+Oqm&<ZsFRC^|bV|6=usz)kcsLiZ;8@ihNr`$!C>OeB#$Uuak-m=jCSq~5my?e3
z)9ogfx3A}Z9Dfmv^C!_DxJ`>VXiUZ$+;^DcPob~zAXaKT4(IHsSjXiJ0U81Kh1BT1
zcq0<;>^r?*K{D;#IyRA}x&m>yXulU#6OCd0`2^-RgI$d-;k24sOK?dOq<5MU(gA{k
z)tB1JmIjkh7tK3E04&E{!UebyIEZ6-FS%8@jIQpz!#J~p7`chxM1&PS_uM<~IQQJJ
zB_jX$u>c|s88cp?u52G2J!j(VbN1|MzfuYBjDc!4u_MQ=5V8ZYm|8;04Kds11;CK3
zBc90hkZ%#y-{u80fg>_rpQ6^$t_OlrXN+x^rbncRm>7j#V+2qO-mls?Y4Or<X!c{6
z#`HLi_i=rt3HYuLM>0`jV;+ng^k4M687QWfgnppzFpv$OFZ|sw>?f^l{j>eF)q0&?
zzn379C46>Ci^R9KcSI&P=#$88tWei|Z~ajxUq+3nnCnBu&iH9Md(8Dr*UK^IDWbIY
zc@s2Jf*s%8)N%L7FldbaBNYkJQp$DmbZUgTNaXP_febU%BB0PBlD^jNHOP<kBdDOL
zQa-W@^wZ~2nr+o$q+k5JscCbqB@M2uSxh$Lc)KA)jfcq;LF(a}qYz3g5K7^oy^wPk
zOb8hLv1#DZmis&T--&c28vRPNxj7nnr<2&+h={xJ>6Xp4rY8ColTA(M!*dgjRyf&;
zPUKC60*GFQ;OzDt92l^VmJFzmrqt%`&bRU5uSDtFb^_n#wMqD7jp+;IH=%)!hThqs
z!_dux&EQBI=x8xwTbw9VcdujXimiAY`-PJdo4>`e*L28d9w;;I@i_a^5v3eZ=M<yg
z)%)(dkMx%JF}>5awiefR=-8|BBct*tfPiDe&~J60s)OFaygn)}bb^^>mUbpKS0^T5
zOSlf329l|+&DO3Ykl#b_pJLKs+I3V4;oAIi)G(r5AoTS1?`ZuEpTxV8Y(Yp6`@~vc
z-#!x_11I*1##_r&QPR<9GhmH;YoQq8gT~}Q3efCS)2>=~nEL3B0qZp=LL;Pv+RK+`
z&{g|+Xpw<ltGQ5s#6ZPDv>EtR|5Uw69KA8)^mX*<Gp<37`!U9JCV^H#!iQ%MF_4<^
z*T2wvV!@43K5H{bhpO==?N%p@YCH4rau#n-1XV}`whONc1l@jj_(E!Pl)Rd@^zP{K
zgp@b7`QYOp|M&p{rAwO6xbFIe({VDZ&l}mbuU&l98}qhybhHkh)ee?0&<rbn>JV(d
zfq*9rFLF;ZuzA-#_uRwiZus({orsAZLzEQh57GW1^0WIm%L(1!Lx`U)V8;4b8gx@w
zTy8lwKPHEhjCO#g=`X&=act@tw)d6yo+5B?lU8ImDE3aydMl33REQX>Lq!fE<cn|*
z?Y!vX8-Au8la$I~UyE1Vx}m4Yo-fQ<_5OJMjrcfugb(VNlZb1JagE(@@kKj_+?#Vs
zY454cr(!$0k^4gZd#lyzkULy|QzuVFpe1-3um?PvXoB@ROu5~3Y>mAmA4sDH<xI<=
z-J<^}4K|xt04hL0K`cwkO{yXuSaxKfqQtrp?je(zR<z?hZIV>RHF%}SfYv|ChtKYW
zCD8Y2AL97#gl1#aa(8aumc~)r(C%!mliJ5fr?+kIG_XJDN+T9C0={NM2aiRKIAV-j
zo?s*{WsJ5unSlsD_czgVoIq#6WDD&=G-jDmSZpT`dOdDrdAtYHc%{2%eDCFs1?%#?
z<2~ImiaUNy%rJ12x;wC?)9=IHDePl|(ODoukBr1{g~7@$5lnYNzCj6WO2S4sGa7Tl
zWsOh>#{pn_L=W)DKDOmP3FH18>>POo(OBa95j%^=nQRWkk>IFHN<~S$gEvN37b0%f
zwkpB$1uNKiKZ%NvbNT3VmUdZsz-k42gGVGo8B4~%yO1E7ns!6GN`=Xjq<<7|V`uR=
z@^Az=?aqX3Oj0pwcI#(a)5e3pk*)`#v`9P1RA+w!l(ilIx3^>ajBmVM;nefCy_iRN
zsvP)%?4|$lQmAb^UE9yYtCXN>#`y<BR-JVX+9cjEb_Dv@Id_BlH#bAohgi+S9{lh2
zHl|4pI6Tay?8^oG23;^WH5~78b@NGTyc$R-OWMjajqn_mAVGGQ-mCzv=ff0PC-B`Y
zI^<HUIVr?j#O{0$p;JFgr_+;!94DOeoY>qzetP2M2k1jhciV*>QqWG5M<1<5(i1qr
z<Q%Xf1vJ=Iz>`S;$0BB5zX?Y_@N3}FDAc`Pt60r!9P8_O!_lKh5j|v5&$P1vZqn^d
zJXNeBT-z$*kA&iXjE90oa^2@XL@ogRL~{DcV1LN%Bdr%+uAw-DWnA5SN1Ot!$Hf&B
z?n6|t|7k=lp4LyVA9DZG{jwJlYeJ&`Z~dwf*#ASfSc+uYK5^=f0p{2baUBpx^M&(a
z<1i)?Z#Kp_lzea!w(}9e8$>Du>DFk1Fes@Dl5Q@`?-1@Ktg8*KRN~Pq;wMF7((j){
z_-P7{ytNk_r=hSUZR1)x8aNoh|IuBYTSzgxc{^>x+S0jeGD3(mg8$tR!>UO!v2WkL
zq<5>7TfNDBQVuwAkO!o!2D^<!UZ#SV_<WZHRb<<?any*K$G2^Zh-L^wCsxU|1H^5e
z{VHTpc;MQ;$r`+)tUqBaSp|2;kzI2q#C>X$bCp{G#vEM*K&}BQC-G}jOG=)Y<=C7z
z9eXrhp)x2cChY`eIr}PB!IKAEXW+!B9HOS4=+_7A3cPGDHcU2|xkyi`H9ofgovj06
zIT|XFF-$L(DgZ%m6y*w#LHwqE8Ltzk|7glJdNsCTBSR38<F05zn7#ez5_%Hav45M_
zKcTxgy7`>?aCV=oMZd45KKO_wM1!a61i;(po4|I#XkS5pW;N<-AL*h~J8{2JAMmag
zxxbcrr00Ma_*g<O1qlhB&rw&as|Pc}d(6lLwVO;WoB&7O!?>O}9x`~0MlI$SwT4b3
zeeK%ag<z8SZ=J7nafFH*FOC?$lk{YrE))Deyj-f9C5;p4H5YaRJMANT@?OXmH^LtO
zR73`OtLvk#dtE<q{ZYV6UF;NRi>sgwc%pbNw3=0|TGg2Xr!|6ks9Kt78AMN!b@Gdd
zcX!f@Czm>fme@SG)Ia~llN+H<u3jGy(a^_JlsZL+PBC7+Cx5Z`<S(A0)G6vpYuMG-
z$@-gzO0PfEgZljcx0)R4tZQ_Tm%eL!*3g)So5pgFx{kD7zWO>*f72Mdntk)DJzi}n
z*zj7zqsF%yeG(d<HN4jN>4&bq;Yqr_O$|>PZdM;%F`{3z|3c#<ea7`QkOFQ;Y?e1*
zN3PcD6Ag%hHLEfC6vPJy+15n|aA?ITvU+Ovg@-BCuXkt=Wg&|Kl}y&$qNxe=X!PV#
zF5Oiq%k9u#n{1I>s~>^&`kTr6n}#Rp>eSa4IU((`;rCE|-2TnGAI+=vWg6O-uK&hr
z{SA0c=ng(+oQpl+0r*B~|A!hqWnPQ;X3$#@ZM}z=QsgGaTo>_SkgUc>Xio#lP$|f~
z#FG)<O*=<e8wwKMhmvrzzqj5!^)omIuc>WU|CY{R&t={IXaqco#}|9hg{^V!h%cUQ
z+PpjAz5?O)aa0JC$TNB+qO-s&Qx5yn*a%d&vHztBoCdiA#|ox;=emcm=X{NH%;9h(
z9TlC=_9btCIOIuA>QTD5wLq0nWyiZnSS6i5>4(-`6C9rpP(l)B69<@bSPn3lbP5*3
z8Sx$H77H5$KBP*)JLFMIpmYioq=2W@+1=fK=`WFCMTADj&S^zZ2r|2z)!pq=IzL>?
zuAfoR^hbW7zW$AdiKbYvAA#xPlh9I9OukzngOiq9D=~=KCia>9v`)+!V^*6zdWb*6
z<kx#{X|DpU;GhG&Xbb=>sH+!}G;^vXVDa=b7pfU|cdrFuot+CAvIvR@?4W}YzYh?@
z;BXm}^*2WIS}3sjIOw^((F0L$pv{pL*PTS$N<DxHxDEC1g}<G96)JyOm)v|uz@LR0
zv^m+@7Wet$ZLJ8dFOrkuL%#5Fr!x9IEq=KEcCy7vo8HZDVM|RY6eF>;D$R0Yz8)l>
za@}>;!JEG5aeMsL1a>OTG<9~QpovZRd<p1dQyra6Gk`(CzIHCmQ9o|R#-VGXrN(+F
zvXkin)4~$X7Tu4F>*19gz6OCv_V3T-ZtK9=c9s>MY>N+!Zr|22lJpLUH-{|?r`vVh
zmdowmj|e2!z?;$<@FquEwrw9Bh__AJ@UQW73-*YepqtSmB~1~z1hc?QT2Or7PTSV5
zakSA6mv{p6cYVa2P`;Mdv4jswIWr)`_F*ugK`8T2C`?nIHBru&7;9}o&?<LiFeLOQ
z$*>IQJ$=S1)+*_z&qs90+g;Cfz1np*+TGRFMYu`>vE%kb4Y!XQM4PKp$ZnE903at*
z5Q@JLwO2Y|80(vvWu{BE-kV1QQ$t#Vs+$)K?GJmez<<{jM}6RvC%`^WIN+{ru{0vz
zCx_O(QIFs6nebZf@S*h(*4E^+>zDQB1K8t>j8$L?J&aZE+(XKJAi#&Xh(mP`*6Dj2
zP7^q>;P+pd@_G*;gcTwd1djXbaS>EtHZe<A8?OC=V1IMIIe*F|WL>8K`}5FVj}IA7
zN4%a4=+8B*x5f$Btb*wU!LW~WV2lFD2cDt`B&^i~?DaPd!xx$STc{i<2m49G=a6_k
zbu3OoXBZDdGkS=wo4+2Mv!IO*hM?Q@t5C4{MQ@|Sr>vXb;#=vu@#xstSm$ot;WhoU
zJt{sNz($)*geQ-OzxC>7Y{Ldtv4VpL6iSNf)@z?C2Fe$GXUC2m`T7|;fVCxE{SW@o
zh(RXnhwb5J@^xt<0;$(Cp}i(piAaPYwN(_osUpLw#LuV^y>t{?jCX-cwPep#XsQfU
zi;5|xBnTMNn(l&d0i;2VKj?!Y@EUj!z{MlzvpIt_Oox8TyXF}LOi66kyU6Z;?PfCW
zn}`s}q+KWOYY=?~|3O_HnefFE0|NsHvF-CsQms_eRxzYUZu35b#uhksJm{YkFF_vs
zoFf<NA##$?lC7qxTB;tKoKY*)o?eZL;{yYy8b696#<&OLg2g?K@pHJpDA*L8i?v<#
zEny`e7OKPQIXnL}uwWg#X~Zy&x;j=Ya~hS9GPxhrY#iF+M=^f8FABl)mX;)XS-W}I
z@k9LJ#DI87{Qz#>r)v;H&H(x*lgPR-0%vYLeope?#}6Y&dI2}~1Wqsvo&gRgPhJ9#
z8VUhBHNbN;syGen|1tMv4-brQf#z3_f&Bwyr|<0_Yyr{t_Rq$~hQ7Cd9vL9tuLn|J
z8(8CqzB%<+K3a$zuZbOX&{!vo{FAwlG?onmnQj=y$%a8(YZ$_fM+Qp&KQo4e6E_SG
z;Bf%ZIE&b@E<}~l7@l}YH(^9uT8r`mq+D2X;3?MIM@ygC^Auho^;H$)vr188d}74=
z$JoZCH&hcc1R}&rt}XPM{{WUg4k3IP+~^C5o=|w$ksB=Rq@Wc(Y#SLP!7dO=7AmU0
zZfvSRj0M_yGy><kmR{cE8i}^09|ltb7f<@`@v4X;>KJ?KM|d*g`y##qcct3LZbn0R
ze-l1-cE?u!3cl8vYN9wSrx@!v@$t^6^b_{9MYd=I;TYS^B0?S6sha7?c@ML|6owG+
zSQ8`&aIBQUGexM>-AAs7(C}*sGNI><-x+_zu@0U%z*Z2OA8B{xa5$tm6~HJ^4t}@M
z`I`0!Cr7*>H*h?Gp>wTnq)$YtJkpzg^+kw>(zA){9^S!8--;;j^Qms^I>6pQ{jG6t
zQ>xP$;H@2J(^u$g7-+p$W&(U^H)3WS1YF<fx)mIag>FAQacP1&7Q*p3*h4K|YHuJ2
zWwuWh!8a8mHk1ScZKx(Ugwdukut0jLfCXaqmJ}ScP_UqiNzXY!*ku7N0GI15ira&G
z89LjWjQ1f}ef&1`{SbRuH+$J<8$&QjMmORe@iWX`B7f3ezIR=yU3(|#XB{$We2oaV
zg-O5{sA7y$AiWkBYnjX{@HOEUs&#8Bnxv25V-s*!vHiEwFV)q(^&(w8=3$-fU#<!8
z_mj{`43abgxq?<MAPSHL!5g>zSpOtw+FKTCIX!kU{mSur$FHMJRsSE}@-*o<_|UIt
zD&AjTc(vXUttV4olwNl3Gd*kd1>v9m4=5Av0AA~ZCx({dhKZkpVUqsPY+6DrZ6oM3
zawsG<dP1gfS||A9Zc_iKSKJR26B7wP>;#dPcmgpf(%W~8j_%l=#yPZ!cuNH85`W@r
zTlefa@4P*uJ$>m=BqW=5^d2~H!wm-xY#nTp$Vm5zRRb)RWeg1L-qQ<rFoccm-LrdO
zz_7q>8DEXV=@{o-VH;Y&>x=E%(ceKwWp(rqo`zVefsQ!oNYJ-c^sSey3goX16Fpc`
zMO)q=B<c1Nx1?Qn2}|2u>mmga8C}sWWG9n%;utjw*|AHcw#EVhEZr?zQd?9s+H86}
zpXl@4NivDp*!=mn-M*o|zN8frdrv#<G|}^^W`y5M^mTM365iK0g~G-jfB&CJcKK9C
zbJMn9TPHVD+lEO-47fU=^I+RB*$-)4u?vdiFfEC(S#cb8NNPPhO1cnIA>xG8o}k3~
zItBBNAV3Qz(rMT^oC2$~R=5xi57}SV4HNsNdR)V74a3e10j%ievoIhY>H|a_z|}UE
zo{?SxW<BKq0T{pZIwv;YU4H`@sXY<DNk~s>^NNrku8}lcRq5}iBJrepXofpq^&Ap7
z@~+aF#s$eqO48-|o>W5KF1_(Cxp>@>1+`T7*iY2MILmMJ)&yM|67{~66l3j>YZ|MU
z>4KJS^a`-o076BwRRTf}$d8lBZmBo|`=JlCJe8VOJVaPt;i?@)SxePSYPD1q{{uy>
z*Fvt<Qj@g<Cu^yVnmMqRY9J@}m~afD^FR}_+B>VCR0TFTLUsYV2I{YeNp|{#-T|l`
zVCnEe^(jOja$-&pLV38mB6%f5NRbtiI1djlxOO^(^W`-0OY5IuFx51t`tlHI@fPw}
zxPU5pWdb!OZg6yDNZ=Nk_!6^6`g*}LG^E&gze)fo0se&64B?e+@Djff7J^29!(_eO
zi!)v|Vzf1nq^hBHO(Q@&iRS%?UM6uO5;jCtDXLcE@bfsZ6^0d60U$}))h!Ou61Mr`
zG?t4UP$0sT5jM@$G>r-k(geD+p}Vze|4i0v{RC4}LKtg$NFDDg`p(7!liYdzB5zcn
z`s`{Eyt^(k65$h<D<wkY1;GEcy1r?r1$19;w$pUd!~=HP#5of;+i6)>Mo|o5JKw^7
zqIEm%1BKmXr~QaOe%ww+Tye~V{%(}(H(f#!H@U}nEj({1EnK^dbL_O?@)_T-)28cy
z_?DfPb!8OJ2Ks}YZ@C7I6+7*7T_diw(|*^Gk+jnhSD$#dosLrdSLaIQd|^2~KE8Ws
zc;vKE(4Nus!1;ycO8TrLxpHACmp-hE>`RXgjSlVFcIA97eIz&4UrsL-%9V7nP+VEe
zmeToY)Kbn>(%I#ibfr*OT*z038tNHKpI^%5PLY+qqOh2s$#C`al}fRUASttGK~^_l
zXu7cUkZ;igyP;!LeTM5gr%*gr%FoSL(gV}m()BO%V+2|zhl(;p_5xI$X>9i!huU)p
z<YffR(J0a*n8Hy!8GtgB%2bd(3mdg_c&mUQmpR-YZm4A+a>h`{DE{rjvbz%72B;Lh
zGX*tO8R;dwTjsioNEKZxu0<fP5}xFN)^%ITC{JI@;;S?GtsuXEatkO^LER7Co-vd@
zAGPK1_vlKcfjp`1JibI5Zk*Q7<4YB^uM8K%o%oyOwpGxk`VpYRS7<E$@85FXXdKb8
zYU?v~?K%hFD8jL+gjUXBOlfQe@a1jX+f&a?tyfrG%)9^}eh+K+6#prTHU<xWA)y5i
zx<Wz;OCU@td?tLtFR;%=gkY<Th$v)`9uX7R_b-wnC7MLDXu%HfHuzR_h)z0@T6BqS
z(Id8sUePD|#emo*wu>ENP-MiA*a?5A5wT0`7JI~~7!%`SuQ*NY6Q_&);tX*>oQbU#
zXTe9~9C5BVPn<6fiVMVr;v#XexI|nk9wRP;g88xHa&d*Y5`G+4i^mD9adDlvUfdua
zk03QqfNM-vOo?eRBXVL^%)!?qFK!YGVo@xKWw`1W#m%B5%Az7x#9?tn92LjJE$|1q
z6|v857f%8`eX@9pxI;Wu{EK*+c)ECo_*d~v@htId@f`77@jP){JYT#(yimMIyjZ+M
zyi~kQ+$mlzULjs7UWJ%luYuX&wKya5_2LcUjp9w>&EhTMF7a0JHt}}x4)IR$F6>!)
zxA-^l9`RoBKJk9>0r5fcA#soRu=t4hsQ4JXlh?&3#3#k4#HYo-i_eJ9iqDD9i!X>T
ziZ6*Ti?4{UihIS^#C_sF#Mi|)#5cva#J9zF#COH_#P`Jy#1F-PiVg82@ni85@n7Pn
z;%DN&#r@*v;uqqV;#cC=;(x?%u(#~D;(x{O#P7u)#2>|<#Ges){jcJ0;_u=gVpBYb
zeMIonHB1BXaSY3F8y>{5@fm(2U<47lE^I`wO(15(jf9aj5KYHuHd>5UqYXzAbQqn+
z79(wR8Qn&YvDN4``iy>Kz}RMNH+C3<Mh0gr?lgvt5o4FJ+t_1_8e_(|vDY}w*k_z>
z>^IIZ4j5+|6UJG_*~U4>xyE_M`Nl!x0^>sCBI9D?65~?iF~()aA>*;e<;E4pmBv-Z
z)yCtDYm94+>x}D-8;r*rHyTecCXK8yWlS40M$VWu=8SnGZ`@=o7>mY|v1}BKqH(iP
zGRj89STPP8M~tJ!G2<5FiN>wQZN}}!lZ;j4$;MNRJB+6q|6)ANc)IZn<6n(u8qYGG
zZ9K<#uJJtMxbb}B1;z`F7a>~6ON^HqFEj2mUT(a?c%|_w<JHD%j5Xu6#_NpN8*eb)
zXuQdIv+)+=F5|7n+l;px?=aqJyvw-Tc(?Iy#(RwS8t*gSZ+yV`pz$H&9^=EtM~sge
zA2U8~tQ(&&K52Z)__Xow#%GMr8lN*hZ+yY{qVXl;%f?rXuNwCnUo-AA{=@jX@eSjf
z#<z@b8{aX$Ykbf6zVQR&hsJ*z8^(`}9~(b0{>%8O@iXJUjr)zC8^17qY5dCgwedg3
zZ;S_w-x~jG{Lc8j@dx9N#-EHo8-FqWYW&UkyYUZW(|FKynZiW$A_THgCX5-T$HW;E
zrr!+UT#1kwHX~-#j3MZG!c3Yev&n2WTg+Co&1^RjTfy97rp+$1+w3v7n!RS9*>4V*
z+i-%$4s+1Vm_z1HbJ!d)cbU7*J?5x6W{#VC&C|?%=IQ2s^9=KVd8RpGo@JhGo@1VC
zo@btK9yBj7FElSQFE%eRFEt-yUS=LLA8TH2USVEoUS(cwKF++xyw<$VyxzRQe7t$1
z`2=&)%$if?v^itu%vp2JoHz64P3D5RXfBz{X2C3)H=8B1Y*x$_^RRiuJZc^@Z!w=}
z-fG@v-flk0Ts5C;KE=Gle5&~`=F`llo6j)+)qJM;Ec4msbIj+O&ohsk&o^ISzR-M;
z`C{`W=1a|&nRl8mH(z1C(tMTqYV$Sbn)zDub>{2MH<)iU-(<eoe2aOP`Bw97=G)D8
znC~>-W!`PR+x$24J?4AO_nGfEKVW{){E&H%`C;=T=10wsnIAXT%}<!0G(Tm2+WdF(
zGv;T_&zYY$zhHjR{F3=)^DE|8&3nzSnfIChVSe5GhWSnNTjsaT@0i~;zh{2m{DJvH
z^FPfE^GD{7&7YY6W&YIsnfc%5{pQcjUzoo%e`WsK{2%i-<^$$$&HpujXa3&&gZW4E
zPv)P^znFhD|7QN({D-+|J}6yMNCR<<pn8#(bW4x)VwbUB24ql%WLQRI6k*EZG9i;P
zC7WckY>}<9O}5Jp*(tZkwCs}IvPW)}y|PdC%K^DfZkIdcpv=f2xl<0y5xGn5mV4x=
z9Fya6uRKlelc&r5@(g)Eo+&5fS@LXojyzYMC(oA$<puIWd6B$WULr4*kCB(jL-Mim
za(RWkQeGvmmXDLy$ZO?w@_KoLe7w9-K0!{(tele5az^IltelhcGB0nE3vy8|$z@rP
zMR~I<$+E1-6?s@5kw@h*d5e6ayj9)?G4)AuRX$lhMcyHwD*r`3O+H;dL;kCLrhJxs
zwtS9!u6&+6E}t)7AYUk7Bws9FB3~+BChwFlm#>hol&_Mnmama(^0o4H^7Zl!@{RIM
z^3C!s@-F#S`8N4>`40I``7U|4e7F2J`5yUR`9Ar6`2qPs`5}3a{IL9p{HXkx{J30~
zpOBxFpOT-J|1LixKPx{cKQF%^zbL;XzbwBZzbfyQUz7LA|Bzpo-;m#w-;&>!-;v*y
z-;>{$Kaf9^|0y@*kK~W#Pvn2epUR)f|Caa5pUYp!U&>#}U(5fIzmX5f-^%}$zmvb0
ze~^Eaf0BQef02Kcf0KWg|B##VLFH0H88|dXVsD0}+{&Z8%BTD)pn@u-!YZPoDyHHp
zp^_@4npCrDQLU;?wW|)*skW%J>QddRM{QNTs!#Q+0kutSS3A_8%BUf=Qw^&TwM*?*
zd(@~JQ{!r{I!*0Ur>p(y40S-AsV3A}>TGq6I#->i&Q}N31?oa|k-Au2qApdBQJ1Mh
z>apr_b%nZ8U8Sy8k5kvEYt?n?dUb<(yt+|6K~1Wxno`qhM&;D3np5*CuWnKcYEdnz
zWmQl`b+am|vZ|;RbyyuyN7XTPi+ZBERo$j;S5H!_>dERU>JIf(^)Kpa>gnnk>R;6}
z)w9&I)pOKy)$`PG^?daL^+NR`^<woB^-}dRb*Fl{dWCwWdX;*$dW~9BuT`&8uUBtS
zZ&YtmZ&q(ncd56kx2d<Qcc^!&cd5J8yVbv`_o(-(_o?@*52z2S52<_9ht)^aN7cvF
z$JM&}g!-iVl=`&#cl8<dS@k*fdG!VLMfD~1W%U*HRduiWnz~Q@hx)quhWe)Zmio5(
zj{2_pp8CG}f%>8PPqm?bq<*Y^qW(+$RQ*i-x4K{bT>V1*QvFK(TK$juje0=+R{gK~
zo%+4{gZiWTllrszi~6hjoBF%@huTyRS}sdi1|rf*OIa2qS&!wle3st|SV1dfg{_Dc
zwPIG>N?1uNWi?sNR*ThYwOQ>}ht+9qvC>wT)ou0Q_{3hT&+4}ZtZmkIYlk&xWvn4<
zr!{PiSi7v<)*frr8nec&z1C^gKI?RAzjcOnz&g{Ku+Fm1w$8E6wa&B7w+>ntSQlCs
zSr=QESeIIlu`aU?S&y|Yx2~|Rw63zQwjO6)V_j=qXI*dIU_IWt(RzY4X=SY`YucKz
za@MRhXU$uA>n3Z#TJ-0RPA_JcvK5GW?)hwazLK5u&lDCRxMj-ux#d%FrryO|xtyuY
zXP2$q%`4f(a$v5M%T{tF&iiLLO=~v0GM%gV^-pG|lrJXc3oGSZW_muqw46JZIg($V
zDI5vtOlnwmF;&Se6?3I*Wu=tMWJ{&Okqn9~Cr-{@DY`4AY<@9U@=Rr?(a?O>YbREg
zd{c#^nM%G~UdbiUrJ2lpVR0tEJeOI@&E%)Ei$TsT7qg`W^qcR>nW>ejDcq&#S$RIY
zki%dt6&LZdh%pS;W$XJ~b_OM8(4WG}+<bDr@m{|bs^u<a%M0P!?cscBr5tM<5DZSq
z9*)Iqu~I1d7gv_&${8*jEanQ-8XlFhzqpuPwwo3xt}GRuu6g;UT*wv!<@tPWaYnx_
z%lYL6sk2e>k!)ppKC^4sSDr5vbyNN2N&$n;osCrr7}-jun8jq+kEwBGomSi+KVNl@
zo71_nJe(_60(8N<_(ZN8(X&`7<uDC2m!bNMm10ugt`sw~rNUBXs+65x$W;Q_l`^Jx
zk>M5dI$z;bU@BjkF65W_CNx!8tYj8R@L$H<2B!*zg$!MmFs;Gq>=NK8Qz}#d#olSm
zyRKn+K3l3_$`)6ami_u>F1uW*1h8a^=<QS?Tbi+^7Yk5?1f~l-y-NTT)Im41h0;Ql
zQ&5B~W=dEr=&TR*6?ChD)1{SLa`{q$Kj)swmoZbvyfXzrAHldMx0pMe#Zbp*^NX~;
zrt_s~OzTl<wSRW8P_nxhoCVC%^Tk3rH{+WvWtXRQ7v@&7xY7MF=T??O44Np24qzaB
z8mMT#qF7=1>~f~GQiQUEKfjR2e4`y+U;|x6a4}!WRY%pglwHp1(e@I^W@fX~Iqy;)
zou$|OO9fiGQ@Ahtma};bF=boJh008B)>F(b7fL`q{$dW=A^psYv>pwQ2HKUP9}UW9
z=k=@Q3aTl4D*2_{VtzU2uas7%@!1k|JyGC+g?xD)vuuAz9nP*SR^mr!Q7>iZ@)#0(
z6?>0haaA(eBg<i=a*LSTd?~Ypc}z}am**C<GdV1XOeMQuH*2bJ1QVE_&V@1UWlUeO
zlmiZ0t^^xarQg0Iz_g#(SOieDtVhaMTwEz<mh#IhWltGM2|!eFmvhsEfvxg<33wz9
z$i5}N$g`2N0Xv}KPotRtW@-*&ms!pot;C%hv}q<c4cuHPMV-9z%Cyc2<^ap{nZk4q
z-~jZu0uUqI6ai=fEos1_rg?dpFBYa3>_t0WIGmq}>IKYc{bC$2Heaa1B-ZcrLQ7*_
z9c?+Gw1!~2G<~Z@FKt@%8LaQ@bg2LuLvtK!d_sleI*avqnMwhzuN-S_#0_($!V0Dx
zxGPs`ZhXXf8A5$_J@}rkYNk*r>%20fAF7+0pz%&EPh(fzWHd52Td^y!%lr9*K!8|c
z`s0+w(&4V5S@eoVcA%rwep=Q+`z>UkA$51TW4WmFY^8)1N@&b!C!l_r>Y%nerO+X}
zCEh8*)1YWMV8BWRGd)ux%1{Z|w>%gq1@N9N6lQ=JkLJq2$I}bI4%SpDKQosDIMR;;
zBzmSB^Xg~1@rj;kJC7zWV5j-Z))dxnt|9@B<+iDnyr#T1H0o*N-X>3c#*H}Xjk;!>
z^aeL1SgmJvF$*%_d{H;dxm%tqgsYD<W%O6?SC(t95#9_{pD<EUQ!Ax%#hM2HR9N=W
zPi19!Iky-FCj$ZtjrCNv0>t6rdr-Q7-P>i5jB;#xF<)e|l3`*2oDc_6Ujg?pUpPWU
zm50rPYKkiW?cnsv6y^(P5yLo>%q<_zEf$K6Q{~MaErKqfDm9xcm5w#lulTwVi`M5e
zeoEBmIPLLsradaNd4Lysyqqg}X9@Yv=9kMpj4g=4;e2+{4_u0M%#DuMr5!Mg(>ld+
zC+@9m6W}N_Tn-mTH(t6-uguPt{hEAD(c+Hlv{R7BlR3C@m4(wo8wIi_I<9dXaRRZ1
z9EPRlhOD|_X)Md*(pVuO&Bj6uYACb_(?^l$cKO&6KI=u!5;!HI!bBsphqGYJ@O6D#
z!r~~y+>vuQJ>6crb($O2Sq-hLf50EMX*x}flLV|sHE9CVP_WVB){}CaA*x$=b`7*X
zT6KAwpgCo^dFfL=s%z${-{t1oy>+VPb*ZbXlvkFsOHfp0OJ4lT@cIhfgb|!242W@E
zDF$x>A)CqJW)1|_57QDF1nvqL4r2?t5A=<N7^EBGnle<sUvtc;23VI6JRxzA#pSsu
zI?A$82B<qjy|!=xf7uQOVmUv13}rGJj_U>~b<#s7)(L#DoS{}6&Vky~Iw@De%qStF
zl3yg&4)+x>yjUbfzyw5`r2-hH9Js|wsSvMmP#Lhe%VjL!#ysMbSBjCkay&{NSm176
z$yW-fPA9-BRMZl!n;>1mM=fItE5}g4zQbsHz{+Qi<Q5kLwgAD)6AT(l6EspGJ53xd
z#=^NR6{e<grDIOXQlT*0VslD*oS3lNM5*x!mOpdzRC~gHXA1IUxiXsrC%!y~wlK@Z
z!-p^M%`_yMsFR-0E-o<Aj5|4XOUL<|z5MHd@`Td7v}27m(ArKlW;uB9gts*;+eK%1
zLiK!Cv%uQu5#tWbW_bXe9^1DXMx2hJU$elJ0FH=V3j`A2Q&iJzXMowxSF)vJnIc4;
z3LYd+d{D~IF6Nv;s20am=z`9Gma@mn&Peksm|AXV8CVO;tJA(cg0aSDG|I*bfk;T)
zOzSBhGwtp>;n57Z&K!6kTAB5W=!9a6D|2(%xm--k0ru=^WY~~ZH=&JXH8Rv?X(f{w
zT*+xH62ZjIIVg-MLbN`EztU8nQ}>aEk_<Bql@Y`=6yatBuy}dcK&eZCQUU6KDfB&`
zEBW;eb68PM*LSt4KBvAB_4)h-FXwPSIR)NxAy+m_D|5awm;)^{#>-&Y3yV(2xt}eK
zd37H-p)`NSZ<l5?YNzR=LCh&B0YL66`Dv&MSe*b|>Lh0hbB>aMDXkA^36gSVc{ZQ2
zlQsarT>xA}ZYE%7YXt%ZZgw_PUWTqE1WZL=f<Scya<Rma76Wu~xC&VTdTpYM4Pf=y
zCs~(YW8I-Os+MM!npyN(re^*XWt78?#-mWGWT6I%=zCyrJHuBYUM({XL>kq}!W1|~
zi1J{<yukLvpjD1p!0<q}LHz@DMji}6RExBA3j@zd$h8@FN93jP4cX!^kM+v3{fHqx
zu2nEoxyq3o=pik(vQOtx+D_7<59{0#Bv2B4a}j-iIG;O$X@FJ%q90_?dZ_eq4$n(F
zlN*wxof35bB##tIO96%(Ez6amMIkPjaau^Hf$EU}>8)or88RDkxE2SwRp!Aq&du9e
zK}ggNEVuzC5N}F8()U3wf{fh6bHtNTwskp7*c0{nq&y@rj)P<ai{@<_4+Uf8m|9vX
z=cgkl(s;#SENMYY8ZwFYFGcGsh1O&aDhdJ-{du2GFXqb?#XVHIS0`2t;0D<_DVLFr
zx5%(nn4PsqnUY<UbKq_jb4&`N!t{cq_3t5&UqnLXmm$;HLbTR!X#q0GAm%XLq3Zkt
zR>~F0711L1p_?<sd1!LVzQd#m13NW4>m<Rj`0d<z@MPA_CDM1ff!pZUgfbbcbFM5&
z)#h>ZZK?>p6^Tfsb(7@JgXfDg76=(QW`6}bjm&(GSUex6kK|`6^FBPA10aLe1ay)#
zD!`xmhAB-@r*8tk23Ac{N8E)KD4Xy#0_Eijw2?Epg34oTa{=_R%In2!_-2v|bz)&5
z=eAcIl&LIXYjosD{a{-dtJF|P5Gv(`BhUy!9Lp7Jce6l=q<7|L60mwq6```s0dxvQ
zH<X0*PQd<!3Jpq8=Y}d$xtp*wq5jDohMJ29An4q1ul;t@31}Q)l0n6}VIXsE81BOM
zD-JK?RFPkvE%;dVLJG0631yg1V<PINR%>YJ@#yT8qWzr4LO~N=%2^OM!Sk|+O00|0
ze3kvhdh9Tlp9P@hva?t^rJM;hhZp~}b}3PTlAh2}rtS^Js154cMX<dP2xfr3+~s))
zhKo5jMB*iyXj(QoXj$#?S}9hCj@sx3Uk!{{K_w6gN!3MaP)qAi{k9qv611&)E|bQF
z-g8tdwdwY1g&aUV;OLKuWj7ac#}qBeGK~PW2pTouhzvoNt%>4kYtUhPEAuN$QzXUL
z^jLMJH0Y)3hyp2iZ2dQpk>dO@YXwq1*85@(!hqrx>rha>u~vW&zy(c1O9)LA`affq
z%4Hwsj1Y4EaLzLYrXDIi=(=;jg0rxGRB{oLvNd<9XP%84b_R`xwvRCTiNUIW%Hw9w
zdku=5l~&go2OJ%kiZ;B&pf$Hxn95m4pf8!8_tC<JWe5sK3rH78zzf=$o1M;<bFq`x
zrDB2xY_PmQ2y$gc0k4%8JmqN^eF~+rgphvB%jOHxzK~wMSUi@2Hh#g<11~YjWjB9y
zWl{eY7K6YNV6kjc<b`0BD-mf`jO2Yat+Qe(?W>7sq!j|MME&yUB$km*@G{d$jnQoC
z!6t|@Y%-;amoEIWn^P-AkE8a{Kx(T|7*nXZEHpxdUZG=#>V7F(^Z?LG;7qc>LclLx
zHi}^8ry9VkHeV5~f6^J$=1_eG^z$^lXwNi+S=fazAcZnUH0LQp{-i09Q}~*1i4@Lc
z<tuqmph&phW^8Q9KjM^ktW*9vNk24|!$+tNCdnR~HJta`3BrAT#ur*28?e(nh+g{!
zZM9R>4Iig@lytU(*cj^&V&eoAwY0t<$dgflj8>^YrC<x{`p$RrN&&(J&!0~6r`@Ci
zhH}i(5bYxaNu~@bET@x@d;tWUyOf)E&w(TsLA-%OSe^|L1=f(B!?N`N)PYSw^NG58
zOS$6Wu?*zMP?eJLEJ<MSqMr><KoNSuEH7xhDhb8=%dqj3%F{5j0-RQ+bjkuX##F*W
zc*`xoyTv?g3BIK~RYv4LQcLQpV5}g<6CkY7G1xWm^B~?ps4LI~qSvI-%2nd_a&}NM
z4N~HSJRU7)-8%(WiYLCxv!TrfA&t*67%S3mGx92fe!yskRb0-OqfP?q7oamV(~&uV
z_8f{w<k-SPvLo8C#PmvYN6_6uyAG_Tp`Y)l2~JVQK~Ak88cDkfT^y)LcZ=zw#`n4a
z5i6{|a?~zgLkQu9oRwm%;a-DtZpF430tGS7(bhs=o#pNr>A=7*A&<CEN_vjiR!3q~
z0e+-pVzIzBvQob?*~~40YU6QZTU_}W*f7)t`e550h&0yn>T4OvoK~v^H=4Q;E%h^<
z)54TJs9ZNJqI33;qgJ}JpAxL_y=@!j5#S6*S;4Y|rDaDGlA&BG6qfyD)CUtf53**$
zxX?JzRH@GfK-f)ukYQ{aAh4lrJcK}tWlm!UV@S@K76;Qts!D%t9WIApj+<H0?Bq0%
zEF=<`UgzL#mpuy5)P@Z({YaR_>=8+Nd&SCoa3s^mVE!=y1Ynkr<|PS;o*7sPrm!A8
zJW<7&S?Jf%8c0;I?@|hU5v&UODTxG3f%PM1Z?&HOmOL4%b#``TCSTA=W`lK&j7k>s
zvrsu=;W}(0Y_!w!er?ldt}Dv2MLlhClVpHbKsrlgjV#W)S$$n9`!&F8i-rxrby*-s
zOH<bDN(E%etW=Kqrt-^BvO-)cgFRlE^DgI(VARoWc?4VtNaG?s1%2@mDbd&mR26z8
z7}{-rem}8}#IaL%ixB=QQ*fG?2X6-F3}_2y;I09tkhnfj->l8O0m<2!EEQJ@bHs8&
z!0|fCW%pEG&(9PTI~W8Q*im1RM0s=`UCk}zY~KUSAgMrB%3)6Fg+%ZT(|#88e?|gj
z+5{68UoZjS3eYXCAG8zmUV+&pcf`-KC2W+7<(y*Wv0Ne+1On6aO6gdLZfc5U$o)_a
zL;np91pXAHYMv?3y7zNAo%ZPml%un?kfW>8X{^sEzhU3<TC<jOWN7l~>dEuRB2@->
zE~fj`TUI7$1(P|Io1vM>F9-OV0d+Z&Jr;n?3zZzp%!0#AIX7UPpa~&05T2E*Wr$G@
zJ7w&rkfwA6O#Xv<WEtYYPvkQg6!1cU!W4;08CF_{>>D`6kcr6;fW}%TOn;2HKxned
zM7W@a!%~IIM;H=xNo)3rU9W^{_tYS==G%`k&fe-B`%|D2@<JfFnYd-!W5{MN*<=Mw
zdkgq%Fxz0J%SDjIK#i})=;|A2LUK?K!TSKyu;7D>0jeazZ&VILV6mkh8uC!}j;U4c
zmSw%#EfXjYWM6WTIpzlYM}`d_(F0Ofl`AH)09NHP3B)!^Wa-o)kR+sHYCvLaSqW29
z2DPl-M^=`BWJ_h}h_dW!;Hah@P4SUD{0>-O&j<+C8qIHD0R@<|$Y>3M8_Wt!AUPj|
zCssn_mbHIXCR<+2U_k@NVaZ_8d8ot;1V=^VRu0k9hb96`hpe2Co=KIbw+ON65V&U{
z7j$rUhQzx-cE(oHLo-pm!LprkI!@s2!RDi>ENtIyB<NwJ-noo`xwSr{5uh8&5)8%b
z%i4e)t<PiN23tkw95P2`-d!xrfQ7Dvwfh^qSdQkG^0(xw_fS>QeaL<ec^l{?=ddRX
z4DiPR3kS+m^p=+kN5D}o<P<=#u<XIYUM}Ys%g{lS=M!_U1b?vOV!pt=u|E_+01UX(
z^D#|)*;-@Eo)Mi@6Goxi=BhZ~#8)gPMr9D_<>~yQi60OC!2t>FsgizgmCXBeg~S24
zZBD?ZfOygAzPD;6_wkGT4i%>go%yRzQz-|!xv<v;^_5}*ujy}czcjUS$i4H-X%(um
zz;(lSNVC@@;pfSt3!D*fGvEZky@2sD=W<KlD!b^m*+?%5;2Qj@96R`rdd`|TvV<*K
zV4aZiR!bO`&UC~M{S4MS$Hgi$;@7K*-FiYi?5vT-s;VfK6~NQ5su;j{I5(pepjgNa
zU(qs7P_x8f9G9R@)d-%|nc<qAkfAHB=a9HGLT%1~u~!RpAGuO@HH)6tO7m!I6<(D1
z9N^(atTjkiNI+etb9@@vP#!#1j$m}=xB#h7%g}bw6%J;Eo1j|e8}i8`?GGN1#l<eO
zu*~{Ioeu_#{Q0N|G+;TnX>b}=%LUI2#AD(zJaC7p(07A#1z6#@d6ira9$t_YU~Yp-
z-ZzCe;D(4ojvPuTp;^o9SX;|^a;2F8>%oOcX}So>4&@mwVKgdb=t`l>0;33y)%Lyt
z{lPb8qE)4aBN1x}3o95(SUockzKZ1)P~~uSNoa8qLb~HfLV62FaZX>XnOn`7<*XuS
z-(z5tz`CI33J7?BnMzwM4{Ej(Y6K`Hm&m~Gw8mqvq4|J~9C=X(iO|BoO*_w8MRXR1
zXR>XV<3Qh4)DQ1VVz1jyk>|YZMO!ov1a&_9`2iKW%S#2CS9clvIG3^FG{BZ2SwJ|(
zu2jIBB>MzZy>LzTLb;j)wJzrZM_8o{a0ek(M>DW}KqGxrFGpKd$1@UiZg?Rnc<fHk
zC#%u|dqxA_l@|*YIRlSDO~`F8lo%6o$0dcixOhXBcCBu7G<NjNOw1hLl{z0{Rz$EK
zjtQAP^+960ry-;OxC%#MAZ8mq<`HZbGJM7S+#FJ{J6fgu)D(1r;Dahq%fPBm>nK1>
z0JM_JbGQlD6cc1rS%4r?%-QB8*ied8BRD#m1b9i}1^5O+k*-;BAKu7miwPT#GK?wo
zvk-HMm^bLIvnv&iNwD0odiV)T6`GYKrKyKAK_K)~X5E6EXUkVPU6v53F2ekePJ$o5
zDXY~{8qw)&o<?0UY!P+>d_PrN&d^sQniX7G#?oYFFnMAImPWql21`vmHr(alJxp8#
zEP-U`10KjSomPwyNTRSE65#daBs26q3lj#Z6hH+(DYy(n3jl@{s>W$j+8j}IS(flS
z&w1$rp7OAh0{`jZJYamCDCdv*IEVQTomV+hux9c!Iq>2q?@Sh0KxvVt7Q1Ibys%Ov
z@b8>wE(-(&M@Tm_zHmRQcjV8ob3g1qjvEvBNLU16xb!aMkK}=NOEX~-2-xjkm$sIm
zeJ&kK)#-|uqG@0V*HNn@z$nqW`w=*a9frih^o#){R+q-Yq=n}Na?u}h1F_^F5`=8=
z5{rSbLfG+pEtU!-3y^S~gO&@h4}7CaK$WEsU)d86a61#xx6o_kZvo?&U9_J&N+U~K
zxncM&<&LnP9BR(_0u*Fq#VV`v0&u<`>kN85#!QNNFo=4EVu%#SkUFw+bHv7zhzEy>
zBiY&9qMLTYV^zUul83aR^%oce_O7t-i|%P1Ayz0@cT#3RlEQ8Uz+Scj4YtaX;mAJ&
zCvls13ryv1flC32Lx-Vu1rG&)gb?h5=vpHf%gb3%OrncqRc1kvJeBIqo-j`1HRxPH
zl3|UXMxApxYZ_n9=fL>T4`#oDG~{i&;uwh5e1X*ffaf{V@9O5y<Zrp<ScE|xpfXd+
z&JqhRXG$weLAqdgpnZYKC~SkKwk_mlH08nf!I4f^w5FpbQvZKtoa>U>Mi7NlG1mpL
zcx}hYL#Qgrk5uJx1kwr_kcbA@kf+agx@SOZ|G^r;>N1y})90MF(;H|_@KJsYRN>e+
zucp0rtRMO2D^dq*#5v@u@^IT(*{>81t(L^^>vy`G#FibzlpG|IpbqfxN<<iYzJg$|
z$oWc6hN;EFC`$_}O{*<?*RM-L4XIG^iqK!IVguMnD=6vFccCf#iZw=XrGs{P8tv+S
zWUSkuxv7fd=f}5d=HZQXzJ8pvttbpmux@yTUmm~k6Zx433H~0st0wQewtYXi4orA`
zslGn0<0o>z+?OBDYj+tuyKOey=#=T>&eY0b8CB;vzl?3d*0+$(Y?=itc}F-nUpCRo
zgb04F4&!ffLV3SY4-oujI?5j%Wlh~1p_z{NnR+Pj6!@##q5b}&y@9v=orb^ACB6)?
z6NHDW9!urg3FTuRcr*kuP$~O;)sKNm1RT2s{C~b-``P;)djcL)Eho1PxcORycd88t
zuF?{zk^nm`jpl-qBUbd3Qv|X)^daUQDssEMO?A{Z8#`lBE;A}GXKpumPoKS2B02PA
zC|qxV&wLgJj7r=7me~`kq2XU#73`cIo!^&@cZ+Eo)^Ur_5wXF)n4dpD=OKjvnYO!m
zI03zsU=k>S7HiEgQ%9L=(%m0IQ0#+v_ls6tsGfo)ql$yiF=~dlhCB>NV?JZxN69Wt
za51NQ#3=R#XA}zfDUkhcm1HnI>~@!3*)7;38DaqADq_Wmj@Arw1URlgz1-hH^sp6%
zEhwlV*1SoJ7{OH@6@RPs9V%exgxBc_$dF@_{dA&V=pb@ib32wt)V@}^*jj|8@R^pO
z7{`}LB)TwRdFzsp`bNw=+-i95=xkL+%JGfOz%bOOG!q$33M)+AA6vngD?FyyP38V%
zTt=krCuj&DU|8DdbgvLHuN7+IrLLarGuJv?`l06{ABPS58oLy7ASMnzQ#=fN;RS*d
zBSc_-MP=sNcs-ID*tO!kBG^2^*P-A)lFp(Wu|(L>v<+?=5Lw3N#_$M+P!%+iv1*kN
zyrYKa*c2@)6d{yF@>tThgjl$tzIb^@4l&FBl?lUYstjNzXMngjx#D@k<ahFGPHX<R
z8;~-L)G<(NThPU$ZzMsV>`A|Y-q7us9!iA9`vPu%vKQblBo&x7d6k|)PT|8IR!;R%
zJMw-eW%YA+{FS*@??62bAl!NhSSm;I7)D}SVMF8-@>(K+i002enKd^K_sDuT+v(&0
z^u(p4waFEY!UP+7%X|AD-mm@ogDc73|F<8`rx`a6VD*rm1^ffWO*xrdQ6Ft`i5CD{
z1(5++nr>wQhMk97Thtx<%M$dZ2WE*DOenA#7A@w`<k~Qb#WH%3jFOn~0{IQpltsLr
z8!zH7VQ39S;ds)zXiHQVd9iT2JOP!dvygfOzhosBm-fT9dvjo2{`osBKK)z%^#L#A
zgjL6STM5A!t89R1T0%EhFl&Y=w&Ch|zL?>0jqH7%Fg@7uXN>%1PgNVluS|^MhOrr$
z-*(ps$c`DmQ^$_hhw78%ujKV!3cFQx*>cz4t`u^yJJv{E@-rkOcZU6i8L;QNigu*g
zz<suInYHZ>Jq0N=0k-PXwRMTd1%)gSE#PuDLAjQmki3qPV^l?}D7mzZjBJZdqg-BD
z3Ov6F)|uHJQcOg?Y0HHLPtW$Fb>PVI7YhPuk-D689A4VnI{dEcg@AEoS*KyKU(cY)
zJ38Y$GnM=jiaE)Z&{T`7KQ1bk{IMb{IafS(e1BHtd54dV^p%%6PIW)XCHVFes)_!7
z1eCAe^S6Yk=OI@M-y9#;%oSXyEy`}nMC9+g%+Yn^3C6ZB!RT&onLgBCtbn?jZ<=mb
zX!J*$0|wbUtB<oOyEpJF?U}M|X}cysT-a0&CV1MU5zGivMg@z&Iw^u?>3}4%M-^Ar
zry2mct`M@Ah(wdLtyLi-G{NiNJdcl1RZ8=TfG=ofc^vgR>F^#=CaC>v;Zd`)*$GPq
zsl?6-89@aJb67a5oHpAAr8A5h>^$8B>J*Y}Dpv={ZI15I*e17*WXt`;OuV+Zws~O9
zgNu~+r6Xm}D38w1&)AA=Hm12A_FR%s5QjeO8(PV9@2#9L%gUaFg9sw>8&)U+LO4QL
z`GOH|Zqu8;ovK0I>rDTGXx`Y~So5|HC%#x-2B~81mL>Lff{~tV$l+O@R<c=;<Unt-
zqea?mp#lhmA|@lljXE8wFYkju{NaG+)0#{>UAoSM0%IVTDvK$oz+zX|X(yA|NcA0w
zUGJpTu`#03^P4P|Dg{#2^9}LK9XrlpFjz#uxq=~ayEv{Q=!nmRuZW92ohnJO8rj8(
z6p%=zw^^7O-S-ScdR`gVak1aAF36P0+F>BT>@Gs29PGuLECE4RgdW^E7J^#PQne6q
z5P@47=ZytXLViySlSPJm*<nkBdydEO^d@<vxkB11tc7IXj`g3`g(oM}#r-4$!Q+g+
zWPKCqk8V?S-rXP2c<%r5Vws*dr!h2Bzz#sT%0BTZLVcVYnkoLy|9D)5#v+_7FIW0L
zx5V~{Y-Sq?o7JYf!5pe+U5CudvJx7yVnOycdTp#_)d^D3BS>cL?LgFqcBmK%{WE2j
JQkvHO^gqOStd#%&

literal 0
HcmV?d00001

diff --git a/_static/vendor/fontawesome/webfonts/fa-solid-900.woff2 b/_static/vendor/fontawesome/webfonts/fa-solid-900.woff2
new file mode 100644
index 0000000000000000000000000000000000000000..758dd4f6070c7cb399334ae997ae9ff6523d3b55
GIT binary patch
literal 156400
zcmV)&K#ae4Pew8T0RR910%Gt03IG5A1{^s60%D~D1qA>A00000000000000000000
z00001HUcCBAO>Iqt3&{Skt)iT#vIG5NCk&&2OuRJ4wHwn`v3r{z<B@w*s7n05h<EI
zJVoTJ*8`xcs;a80nwJo8T-)}l3IHJZ(NBK%i(mcbcYp9lfAVL4@mGKIcmMED|MGAD
z@n8S*fBz{k7<d0bn3nCh+#avbZ}^X2zB_k*eoEbh)*fAB`$W-Im7wrZ*r9bLQSr|~
zHv(Z(cYhfRZRtpN24Lysn+*V5umWa>$k+yfTBl2y8URduF8ULtPe@HSJ#Z$+N0aEt
zD<QwG;3CQFv--Romg>Odp!abQCwmh$Y|>_L5?gB&hfU#4vaJsUegc04EwIBTw>uN?
zg(|LIQpJjP<m#bHApFbIY=3g|2KR+_ho+dy7dJsncz9&1d<mx&v#F^F%)<8SKQ$u=
z;Ued=S4xNqB1o`WIq0DFDyG|lLk(e|_erRU%+w9GG6%76VbaFeM<TR4dHe}gt$%M&
z1_R6tK%$H!N+LywfU3n2NWhlugm&aD)t5~x+Sa|v6=y3ix!B3IIM*t-*4&oPYu(p6
zum9h-G-r3g<H^eTuU?1Ouixv}A<g8T>7D7_JiW6sy|<U<@?O%d?xfS5bdpBWNje3@
z3ZaOEMu;FG!h#3_69#0A1O}f8Ngxei0|pzP!8XBZopJoK{cUhy_WvhU{oZ|3-n=Pq
zN<$%t0fxYU1q6meU_eo2BvH+kt}1CrvA$dq!)vXN9XV_GdtLvDO>ZJcPDQHa#AlbO
zcFOrdT31@<?~b=ArG1}(|A*76`Xwlppv0CTPD_y^MWMDL+e++quY;5B3}hgi&Q0E}
zhd4{Ud+a059AS<)cRbLyKH}^VdJ&FD`FVb(`R`o7_wECdM$(8!nvpf5Cwk){%OR?%
zrn(y^>DtZ$db2<im&kS@KrXR23+zsk0)hJpy(N(Ue_Cq&yAP=96<~KYRbxpOxJ#{8
z+m4mAlW~+xWJj5C99(ZF8RR&u;p8Mc>mk+<dx$l}`8>n^fJ@oGz-{>F;a#))@Q*T$
zpyG%fd!KZ6Lo5|L*^q84*s){nV<&yGVK;LUTOliUjGQdij-6OBM>_;T7rL`mh;Kt(
zWV!JF!O!O5-v<QokMAxi1rOpL+=F^(cUnx+VrrQdbAb<l^Yv}JE0~)8kSk@hQ`qo)
z>C7kDdlvxsAu=9v(F9A=<}8C&qume{WipK2rA-f5v|QAH@cg&z+wt@Nt|Ub_6!9Kb
z;Wr4)@iE$cuU?7n=_d5wPYR)`H&xHBn7_oJQcz$w1;r493l4`<Sc%lDkq(EK9E5Pd
z|MyEv?i2w_@mw34n`KD$KefYeN#OU4HwA6ZlUh!avu2zkiy5nD)<MqSx!J0W25xcf
zB8P;q0icEk7}@`Szc=kapQNbQD6{%juiC~I)n!{v*Lqd%nrm%ipS_Q98yM#TFfM>F
zE`TsDfG~hS?7h#y#X09*fcFAHCZZuT5M(NYjdDZ;sVtJx%Rk1u0Ah?nL~x8kM9>J5
zGJ>Lv?2OE6`XHrLk(Ba%lG6KMlvLx3so&+Rz6(}Mz3Q^O_rCR>_sl=aD!=J1vpl0r
zvujo}6>0uMDN8(V(fbq%hSR?=b1VX$bX6Iy?p0yNYDvqBE>Xx|t1C`6F<>G2sv^tk
zy`6pD5-5cLi7m@Y&qSR0Luws|%%YPzld+_>``+_bYpo(efMhbkBIc_5m;}+4NlfTa
zUHl^Iucr^V`vY#Y8Exz&9<sE3h)Z44b_8#D$0$4??GQ7y?mBh)xEDu!L`f7y0)zQ5
z5AKs7==J+t`}geZ<E6`6%-LT6xcdKv?{EHXCO2~3KI%6xjZrP6+=71RI#}dJZrI0@
z_}AI}HSTxlujwoM4wN#%-1da^davk%?&v9SeZ8SA4)F?%nI6lhsNvUIRv+EFOK>t%
zkE`^X+fsf`w5o4F0R`Z>p$c+l6`;tmDvcaOj^#O!=U9=$b>a9CydZ$4i1VuXTM7Dn
z1Bz_<8r`7MwD76RSzd94?wlO{c->0VrAT=@M-Bwe_FgH%^S6wR9N|pHg^}%#;1%tD
z_%t~<`#9{i$X1n;s+=@fGI=jirQze(1NfO8gUU1WJ#Vl1?5<eg3i%QLmUR*SnJPvr
zxEFhN&&E^qFR-WGZ$j(+7<7#HEAdWz<>P2}n0{9(GS7G5xPFQv<L*vR<k&8=DTV6y
zAS?ByeN7kPtFbqES6MKtr^e2Fb73GKUYt9o_p6;E$5kHP0t?T1m6eC=Ea*B~8!X7d
z$wMQ@%|C&k@b>f_v%{d`_d3n(cS(aO(`R;B8UJ1criXs!8_y{RW}iQ|{ETc6{F@FN
zIn+<%I`Di~pH{}VlI-YQ_bzdc)K#LN>RU|LsqtB3&5;GB%5lpvi+vlsAh72P;i7yy
zmsl6abk%y8v0(bB9^F%WruWJ?Y{4H^K6<toIr8jlt6GAg7sNFf$z4U}-bKjzXwc2_
z_?gL}-1cYdah^8NYY+S|d8UFzv(@9xQ2l1vXVq7awpVrS?aIr3)7V};um22(t~i7j
z_efu}`kcppyiWR+?S=YWL4^-DXjF4kJvoC<F8@^0)8mBR8h@@(&364X$L(M2C+HL9
zLZai)248dgoNW7w`9XO4{-pix!eWji@-Fcl_O}FT4%|34XRodP9*pDAsMd7we=1jH
z8}G!9C4i!P-Qj314qb`k)v8R$^zv?=HIRQ&9(w;51E|B1_&w~`o|}Wlmf}c^OYetQ
zD`;{l=;Hk~t$NR=xj$vhBC5}wi-D`&!Tv4@duE@>Qo7g=t@hYVpml!X9@YM&dcHJU
zD(zLyd9y?FLi>l_IdMD#jm#<oe6eU6n=|BF9L>f`I%Vv2G<V6fpC3xrk_Q(#=i%g(
z?s@(FED-WiXGWRDy5pKM9W}1?Fivvk`1{K8&-2JFhXZ|9`I(+ev`%Pa`n)_Rc$N58
z)*GeAM*j9QS6yhW9+Ue+XHw;FXBs!1J7V2JJ{2)X7f<*~9#r5KrB~^^HE4cnaE3*E
z+o1HZSCMIsuW~d#Da>jd%|72Z79BUeW0|q+L3s}tHS4P(?oqF;k3U}=xx=$EUuy2x
z^ZU~O6sJ;M75<)*LXRKW*S44RKD{5pI_b+c*Ok<qz+#-M#`7Z2kxo^)D@Fv)icEfP
zep7p<bC934$Dq?dvD?2#dEVWnI0t!QHXAM-e1G>a_)6?n-t~+Oi%}l;T&j1w&N9CQ
zoFs>aJKq-C1QxtMJ>85MGv~RO>jO61h`mcPQ%COz1G@`v!2{HNr17@3(v;D3a=+Ip
zHNJv!IHm5-Y37sZa?DySbTY?36Lbg-C)gG0v6ECsZR<Lf*Gqvo?N=|i=h+=T2=!`{
zZ#iywAqE!#zK`X<2ZD~46Pg#@m_4t5GB+82c`x_&>YkVDXW&fN(I62H2Y+vGlw*2s
zxs&9e#M$5b@-Ah3T4&cKe}2~gKeR4pS7NKgqOE5w5j6dkuJaEkSMt1%N{tu2tNAnK
zo#f9emBZ8im)Sh=rM*9CFAX)Rd}VeVNzuJ<cmYrK$qcPyV0mtL-D2~l#N3&}P@n#t
z#_X|sDXBTF^X|W|@l0ZTQ;0kr&Urz9nI)ok<~BX!+8S<r+yO7&8<Tpl^Gm}2Kjij9
zl&JeT<`(Z^zhq3Gj@M-TXzTNJ?PgVuC|z&Dxdz{c#K&UCD*dp|6)yav%lpU?bw6zM
zd-VUWw5fjYqt(kw_gAfX(6+9SPO^`3uB0q>)BUt`)pdsYogQbU(<np#t7G0!PwSXI
zC!PQJ{!HV~_@&!{;{JFz`<0#&PUlbo@vDA~VZ*p9-p^Xa*EGyje%Gn}n9qMJJ7FCe
zjwwlxT=gNTdlhc$rT_nHi~+UHUV^c86}Aa-ex!rCF;IE7RscQ#4?wB?irlsEn)Dk=
z;UV7b|GjSf+!;XBVIPQ7hhY(*KH6dg#$y5|ViG1}3Z`Njreg+XVism&4(4JW=A#S?
zun>!|7)!7e%di}Wa0DlC3TJT@x9}X_O@=8nMW(H3XAYa==7c$EPMgc-y18Q`cD>zT
zH`+~hv)y91*_-y3eH=-T6g$uPuE5oCbzMDI-|cX<?xlO>-nw`0qx<ZBxS#H~`&W7=
z*dJUCYJ+FN@8AzXMUOtyn8hL*M(7x@9viR`8?z1DvNJ<=<p_@CD30bBj^$)dVHuZm
z1y{426|7_xt69TM+{zu?#r-_MgFM8;Ji?<q#^XG}lRU+Xyu-VE%qM)x7ktTA{LWwe
zD-bQdL`j^)OM)ayilj-pWJsoDNwyS;m8Q~NdP*<pBLiiS43?oXNw&yV*;cnjT|AV`
zpD=&={2B7+$lt`patU2BSJjPjv)mfD&h2#v+zEHuop*QKQ}@QH6qO26MVd)>=_><e
zyv&rvvR?Mc5xF22<*Gc8NAg6z$ancA|NFQ;n=j}q`ni6CKkd)^Oa8jQ?ce%OkxY@p
zkz<jcT0v`S9UY*<b+nGxCAva)>OnoC$MvSZ*0=gezv$NxJ>(39L*vjU3=Sj0?65ej
z3me1MurC}AC*F>EI}Xth9WfCLiI5yAkP@ko7U_^48ITc~PzhB~12s_#jnEz)&=H-`
z3%$_?eK81=Fa@(P8*?xh^DrNauoNq?3ahaOo3Itza0th68~5-U4iP8<-r_5M;3t0J
zH~wdA#$kLWU?L`EGNxckW?)8UVm4-HZsuWL=3^n2WI0x1E!JiuwqQ%PWheIJ7*6LL
zF5(g{=Sr^PYOdu*9^w%m<#C?pHQwe^KI3z~;A_6+Xa3@E{>8ufzeTm!7T+>iK`Uit
zt*+IxM%KhS+5j7EV{Dv_x9K+D7T7{tWJ_(Ct+aKv!8X|r+heEfw4Jqc#)lWoMRUnq
zHdoDcbHh9~FU(8x(fH<@`Dy+bW5k4-C>zTrv>9z?o7Wb%Wo-xB$#%6p>@+*w&aq4F
zE_=mZv$yO$`@}x8pKO5j?H?O%Bkg~RO-U&`<)mCxl1fu~sz_C+8r7g$REHW+V`@q*
zs4aD(F4UcR(;ym7BWWB>qA4_;=FkFKOe<*(ZKiE>h)&RXx=h#UCf%mT^nyOn7Ye4Y
z9FGfdK`zS0xD=P>s$7ki@h0BRd-xci<g<K<Z}9_u%+L7^zvqwqg#$T+zws~r%aIaI
zVoFj;A?YNGWRqNyUy4c@sVLQ?rqq%8(nOj`D`_j8rMvW&0WwI2$VeG2<7A>tk(n}2
zR?7z2CVOO`9Fj9~K`zTxxhZ$$i9C}R@=D&wJNYcZ@>RaeA2AXp5%7=@iIECvkQv#L
z4+T&Jl~5HmPzR0B0<F;jUC<8$F%%;(3DYnWbFmOBu^Jn(72B~J2XGjtaUPd&71wbK
z_wg7{@f`2*6~FNh77QTVMR##re3!r_bctM2m)xaxX<a6l-xYC{U47TkHFYgqXV=3G
za^u`2H{H!~%iJ1w%{_E)+&kyHfA0VP|DNY%^Ky9ky`o;lmbw?~q*V!M7ceVe%@KnG
ziv^ZDx-$cp1g@Wdw%Ly+kM-*8HacYd>i32m2{{pRD&%oUK*)D}i;#bQ96yg=+;8T$
z_lNnT{dxW}f1|(E-=}qXDu#-wVyieRspd}8&!(!XYP;1@O((k^e}6SpO;t18BDGYl
z_&6KYR<&L2P)F5O^;Er90V+tleN$@PPJ248j;9mqL^`Q@X<b&GL+AW*1$0qeMpx2R
zbsb$-H`T3lYu!P2)uZ%uJxjBf?XpoHnDe+ktFP)u{f|08|NLV*9j3#LXJVT8CXLBv
zN|>^ys;O@pnWm<h8EA%?v1XiRFFo;_%vQ6<oZQegbHm&*Pt8m7LAz+fdqlVKb}3`q
z_PY0GPhOt;(fq=WPH`$t<-To4>O$S9H}$2VG?K>9B$`aqXeKS7MYNPwtACa*(bZq)
z5xt<-^qGR_D}`|?F35$r7#HW#T!yQ0bzZ?+cqi}Y6MUM_^A*0skNBzP{ye{kGtMdb
zq_C8c3L9!6ZKacRm!8s3+ZrnqWQt6aIkH;T%2rLDp5KjKK1tBdzeOEJ{v!brBjucV
zQ2>Qe;mb5bYqUpa^uquQ!Ej8(RLsC^EW`?Z^exzqo!Ey%IE8b#ge$m?o4ALE+652y
z|G(+|Iv_nYF*T)MIJHP>@da1u&rMyNx@<n@{;nqHMb7h_$2s?N?&8OGPT3k|E0rx5
z|HfzWMm!hK#{F?`+!J@j?Lb@^7smy$eykCz#*#62Obj{USNIfOhWp`ixD<|u17UC2
z8Fqy2VMACQR)$4kMi?0ShIT%~fArV=6@Ss6^{4$Vzs*l_FWhst(e-!TT{l<5RdN+v
zIako-vvf<f|Mu7Z*l+t~KkWx#-|dTiwvYC~-rGBSYj5ney|Nef(C*qzD`W+=3eZYG
zE3_QYGC)fKK#Kt_(gHy9mJ8-+w%?X#Y6hU`fTn4xrfAaBnkE1muW=fyF(JnrrQsR|
zXsCv$ullHmx~p6I?W@k}v=y~iJ3ws#wNdL-{n6D5P)oG{)ErPVKuy&|jnz;M0M%DL
zKy_6|wG~xD#pC^+3#tI1{K}_1%wRgeH2&pJe&bt!ZvejLD?aBvfOi4j;cedJbzb5H
zp5`eY0eBGLes1SxfSUk-8vw2cxE27o6yRbm;sQ<uI2qt1fD=RiXa8{kM{y*;K^(yT
z?28BAz}NM)eNA80SMue30iWMzGxI6*SAX<NKh4Z^UFUUBE44yPwMg?dPct=LlQl*I
z%+y!i)lPL)$4uo_PGwYDB~?PjRZK-yNClK%d6iSylvP=jQ5lp@X_eYc$(2k=idURs
z6(eWH|NO;o{KC)t#E*Q<=X}b?d}zkgJZ8qDW;|lX!)83hgFL|f+{-=O&0XBd9o)`s
z+{6uB&vjhO)m+JCT*A4W#TgvS(HzEs9Ke3;%^vK;_H50TY|f@^$Of#>x~$1+tjfx)
z#ELA-GAzwfEWzR|%ACx>EKJXIOw9zwGMb)b6hj!yAO<pkhJW~tpZI~V_>4DrjhDEO
zd$@@kxP<dKg_AgrLpXqa*o|%2jEz`_HCT*ASb+JMi`kfknV5zN7>`jHf#Dd2AsCFl
z=#8G}f$r#r&gg`W=zun8jC!bpTBwN{sD=tCXJ%%Gk|=@VD2gH|fb7VOBmfQ(2y_43
zFZa<sb5GqP_t4#Nm)%A@MD2^(615qKnin-5ApigX0MnSpG^R0)X-wna|2Cu+YNf%d
zsAcM-vOJ@we5&#)uLwn`krI`vHma<jp{}YL$|9DqfJK_4Ia<Q&T*B*2QyQQ0shaRL
zH*kXvaSE4l0jF^uC#Z*#Ie`<{kMA{sJ2;$09K;P=$GcpnQ`*No?c*$M(k^Y$B+b$k
zP0=ik;TB%!HC|IS)lw}bD^;l!^LUI$ct}lDO|?`_$%<Dp$FRtMs|J(Vf(Lkj2e^&<
zxP^PThh5l-J2;Kc)l8|X#)_)SkE)@{{LL_m&U%#MIFIvqi0}E@kN6=!<OlecU%eZb
zaFK7}9Pj2`yn)yA1>VGqn8#II#%bKaZT{ZJ0RVvCe@<S*%i*<~xPylQcIGCTTiOSB
z1HcdX5x^_E-*Fw_<-PPeFUYZ+zi(~?Se=iDd4xv*Hew6=T9!>%g%xb|4!*&b*a};}
zj$OBh{T6^7dtehAVIu&mwY?7-4g~js;M__y{!@b-{J6sY+i(BrULZO4L)i?#I=p|I
z00c{bgq)c54Qw7@mx9-TO#^HS5VQ)~0D(O8MzqEc+Q&8ee&GA)1n@odSkJ(WXdk~f
zt2=ZDjQBzJ7fKsol^SQBi#CDo$3HRc7_g(j4g%W)PStmD8MjecJ^Aa}5Kbba3iW9~
zLmJVTCN!lP&1pePT2Vo3+EAcKB{^*=5tM01dpgjOPIRUVUAdaO>~g!puC(jydb`PP
zv0LpfyU(7mC+#VF%igyS>_hv=KDJNnQ~TV$urKW!`_{g*pX_J*#eTIvZJsT%6}Hkg
z*hWd!)l|(@p_00)n|i3P`l-JLYp6zRjK*q$CTfx<Yr1A=mgZ<z9jlXds?N~ax<D7|
zB3-OYbg3@W<+@JS>o(o4d-aH()SG%oAL<KzrQh_Y{?|Os*8(lnA}!VuE!8qD*9z@)
z^W>RV^;%xr>v&yn>dn1{m%Y7r@Q&WeyZ8Vf<b!>P5AzW|&S&^cpXIZCj?eX7eJ|g~
z_w~d5SU=HE_0#<fKhw|hv;9K9+;8#6{SW`o|Mz)5-xv5oU*yYub!wdUOZ%sD(@p7?
zbWeIHJ)E9Puci;uhv}2_dHO1Smwr$4(z3KE?ab<Ak+sXZX1%gLS$}E)bZZ3w$$%nI
z1PsW55+I-h&=JUhZa{yaDliDx4X6+7`|X+l`?;pT{;nBtfNKsM=vn{=xt74et`%^I
zs{jslt%1W_8{lwP0FH1);7C^q9OZK0XxA1v#+86$T>!_qGH|?W2b|#A11Gu;z)7wn
zaI)(JoZ>nIr@Ah{Y5R2r?uRY}9)vCi9)-RF9*4#NPr>T|FG24BuR~*jpJCSn8{vzf
z&=B?~6k0*kp-=(MfI@$00u%;76QS?`bT1Shfi8l=qtK;Lcntat3Qs^6K;Z?Xc~E!-
zSsy6830(rkN_cH3UIY6Hir2z^f#P+rpP_g?>{lq>0G|QH8{spdcr$!gD82@DgW~J3
zyP)_1)E6on!2X2F*6`JkJ2VOM1Wks#B|Jgi4%!v+_OLG??*RJ}@{UN2A-@&&B;;Si
zK8JiU>=VeBz&?O{IqXx&SHRwfd^LOo<ZIy><eOk0LB1LGG2~leH$lDwx*giS3BLtO
z_dpd;dI|O+l)i>$LFpUVTTt36amsBps;EwD2|%|&7*JCPN7f4>A?<{y51j$g7+FA6
zK-D1%NXsB9k+p-!k=21HA>9Mf0qFyXPDsx|bU}I*qASwp5dDxIhUgE?ffx$ShZqL^
z4>1Co2eBLcbco&IXF==%KO16C_^A+k!Owu$8-6Clfw0RV4ubxKI1YLe;xw_Brz3vl
zS-@9AoCBW+aV~rz#Ch-q5a+}HhqwU#AH;?5`4HDb$3ol$H;7wdw?N#EbS}g_uz3*o
zLOmevgBC&D4=sjx09pd^AhZDDA!sSY!_Y#AN1$a8k3!2K9)o5<JT9hu4rm3$o6v_4
zZ^2K5_&~~K`5{Q(L41U?3F2?41hE+D6Nn}7KOi<D>k6?2=_`osjW6$7?pOxYoEzm%
zm1QrJ5x+76Is?kpp{`J_f%H3+Yr?BSxfZ+@l<UIBLAgHC{!ng!v>%ik!mfjIBQa%T
zVAn&riBML6^bnMbNb=AN^yYz~^p;1pIK35xa&=_A<(i;cl-_zU0loFblpC<{Kl+;9
zMu;i52kA6=7r-!j7b1N@?;=#Am}P!c+c3)lsJ3RdF;T6@Y?Grpf!U@I${9djpV`(P
zBb`n?lzJHErXEu;FZFo8kol;WqFx5`Q*T0jAQqxNf%-&jKz$MQ#n_1YQo$zFx4@><
zx5H-C_rvDY55X4HkHMDIZ(-;n)E~oE)L*T|*3<*k?Rc$e=3y<iqgjMz@iAJ{EDh{T
zvo7pHvmNY8vkUA-vj^->bJ$w!L32FK31hIPIg7Q}o90rQ%f?(yb2G3X&E2p+&Es$Y
z&GT>|&C75Q&3kY#%?EG@%|~!3%@1%G%`b2`&F^po&Hr#D{gGohivG0pr^C_o=MWrA
ze+f8_{wZ)g{mbD5`VYa0^dEzh=)V9b(|>ggr_g_c{+l?J{-EG=`rpAB3`_-QGH}-z
z&SKym2JXe#4E!cImv(VDk9LJIoKL$F?aH`-c6Y%=wEKNU7t<a{dk`+6J(Tt^Tt<7g
z;0oIF;Y!*YzM`vWZ>GHiSJU1_dpB;Ry-#a#GwlPkua9$oqV`SNFYyd*O*@EpX-Cn1
zhfipK0KTC8QMSFE^AlZ#FX@KU&4^#<dUSJ<#-p2;ZY|Q}bnDP<Oj?9)Q@XuK%h2sZ
zw?Ankx&!G>B5g)@3f)<xJ?PG%yOgv)-Q{#wkdCCgn(i9X(R8=a-AX!^?k>8!NXOIN
zPj^4*1O^Xg@L<x33?9nhp`?=tL?)d~R79oSzf%k$hPHi83^QLbolXoV#v+|Tj6;l1
zx`3EKsFo%sM7o5SXpDR+F)=YI=~7}cp<17q9Mwj|6i8PPQz9QkOf^Qjl9-lQl5`cZ
zw8p4bCYB+VCp|%|AmpQn701kn6DtvGlAa;fCN>-6HnBNKZxdS}y+drdmh=I!HL>G7
zp^2S<VZ_czUl6-|MMH^Qi9JbQ5_=Ozk_L#QHAecGIF2}x^b2v4kk2JfMm~=?1?dms
zRLL99PniBB&LGYt{Y9KjoJ0DDIFGo1^dE7dkf$UrLb{!}_$y|_<;0cbL|jc=Lmon0
ztKUFgg1C;jlRS*Lo4A)e9`Q8sCV4{QZQ>pB48(`Tm*km<T2jt#>=OfXq$b)0LoOyd
zq*27+TGH>t*Tj!wyRAeJzmgXvegoB##P7(b5r2%CZzldE{wFU*K^Z>3ZORD7sJ5hx
zq>N5plQM?ZlGmq<Nf~SIx|K2xWnA)ll=1LTs@*6PNcxd7VMO|gG6`i0@&=SCCFxqq
zR1tYo$~2Ve$eU57S48@QG6SVY-jXu2!~~RCByUTZb<RBz^s&rFnUlO7Wp2s><eexB
zQ5GfdL0N*b<k-4LS(>sUc|XcZOUWz`?MGggyfSHj@~Y%DNr#cwQh^|^O<r$KzC+%C
zym3A6W8Rd!73pO1UgUjAXOa&mA3-{od=&ZEd0LiFHWF)-PmPGR$fuFdAYDv8OOh@l
zpDk%A@;Q<&C!bd`T|vH(d=cqN@(tvhNmr9^m89#*w?)Q6<lD)2kZvH~OMYO^QzAb~
zevWh-`6cqpqzB0FBIzOWd(NJ|kEDmmAGk8qpAX3&lO7>|O8%PkIQe_>kEG|wzai;G
z^6$<R|3Utf^b+|m4iRIK|0e%OdWC{Ag!BewKpBSg7G-$K2&DHY<4`6heL$IvGShsr
zWSN69H|bZ(yp;J!jj||ZxijclS$-)<|5H|?tbC*5tgK2z3`JRuvJsIeJ1QbZq3lN4
zhZvQzucnD{DEm?Nx2FdnaRcQDNr`fT#CVhw9qCC(j88e)QG6=pG-3kE>4-!*OAILI
zOH4$$aF&L=PPvG3F)=aaQp%OYq?BtY*Ai1v?xfsJOhdUx64O!clbD`zzaxDBi5Vyl
zI?{)bn33|ZBYl*JAtzBDqdZQ`M0r9bV`j=Tl$VKFD6dl9Am*XGsfgH(@&V;TVgbrm
zl30lHjl{x~?<E$Y{8Ta)rDRGW7Nc~^Z^ROmKP9mg<-d}#G&NC&B9@^JM;(z^fjXfi
zR-sNTu_|?1iPfmH7Yw<XIuCW;b80)(b*USS53;(^BC#%Y(}-A)x;b?VVtwjik~A;%
z@PafS^=RrbSD3PTDiRx0uPTU5s1H*gnWts-F(fvpJ|(dQ^|^@HlKK+$6=EyuYm(TO
z`jNzT)K4Y0r+y`|1NEnf*pd1d^>1P)>c5h~+0_3fcA*U|u`6v<iQQ<EOYBaYQeqF<
zR1q-^ZFbt6#GbUdRS^5o=A+GjeJ*PYBC$Vh5s3q6i%A?vTSDR>+ENk+)0UAqgtmgj
zp|q7H4x_CqaX4)ai6dz17Q~UX4QU(AfwQ&=5pfJ{Gur0Fv9zsKGLENhOWTP!fwn7c
zcj7eKo{ESwY5UOjBhI27q=K{v?O@s==NwD4!)S*S=hKd)9Yb75JAppLrL>c1Cli;^
zPNkhmTtPdVcFwietX+)6^|Z^}{-rk|aRcp+KE{o-dujI*H_;xVJxtt6dzAJ#aXalv
zN!&$yIwJ0-y+nJNxQF(J3gSN6+q4g^8JB1u(>@^{rF~BOf_R+vHLiU0{);{w@g#jL
z`V7~7vp(YzVRZUT^jV1?=(8!p81&ica}mGL=T?N#==0DQB`SR}{aYE(7pI>_{7XNB
zei4I}=$FthXRt2)O8PYnHl$xqzm37B^gHNxGT4@WH~n4)+tcr-Kg3`s`XltG80<!W
zhW-MBed#aJUuJL+{Z;zw3=W~cLozs={=N*3pnn)KIFkM`oe#evPN4IRBu=99y~N3M
zev&wa&TkT@()lwYPNOrH&N$+9vP%+Yk|A*xIY{Dca;U^P<RlX3l2b{XM^4knIG>!J
zoPoH2oQ0f~xQLvSoSV3WTu2ddIk_0QIB^BJj3llimzTJjTuI^@a#e|I$+aY|BiD(D
z>&f-V4Tu}ajTI3$lUtBm61R}ss$|?oZcpw=+)nO7?nc~2?jebL$-N}*BlnlMpFFT+
zJU|{y9zr}w9!4HcJWL*?lJO{c40$~97<m$T3h@+qp-RTH<i+G=#B=1;<Tb=g<V})z
zg}g=LRq~FAc#V9Je293Rd`uE=l21y!MLr!7Z<EiF&lB&EFH7P*@-2z?$#*0^Am5Yt
zko@5M_=uGFnEa$<d_sOfeo1^veoKBwd`|u-i7&~YOU75^@8loE*W};izr?rXXi5A)
zj@>SPBqv<jixEFjeW`xLFVp}E;x}p-HT-%>RududCpD$SU(~b_@i#RyH4E_%HJeJt
zf7G1RT*Uv>V$>2uQcFu>6t$ehXlf;iG1M9nF_v0~T8|h<ZK#qBwJEh3t&7@{+KSel
z+C{V28nqj>JFN$`7qvI77qu_7AFU5{ICTWAA9XZ!%r&7BbsTj(Z4h-Lbuw)TbvhAk
zICUO%K5Ybbp`=YrT^`XUp{}B?rcFv+t3#koPF+vkNSlJXg}RM4HFXDdCv7_F0qQ~8
z4AdjkqqLc*Clt|Up`N0irp-z{tH?Gx^*r?gZ4T-c>eXuuJ?eGpP1-!v+en+Adbgl0
zKz&4gdd;;*eNX*JTa@}8X-iOlMYJWUf2sdyOHmz3TZXELwk$Q88be!-k=+#0R$ydf
zWKY_PjO-)n;=^+xy9y^mc5{AdcOq<A$exhBz?OsTqX=68vOnYiuoWQ(No-}v;Ssh9
z<V46RV5>sTEU`5q7ZO_wayhZJAy*Jv2XYm$bs<+5TMu$0vGpN07ux`GN3jhd_wA!?
z1i3%tfnXa$9twFF*k+JNKpqXY1>~s`+XnLN2-_C&5y(fuwu5{@Vmm;-Ew&@%yJ9;*
zekrvl<j-O|L;erj1^s$3ssAAFM64k1N~|;T?xf9)ya#FXAn!@qyvTcz(i8b$QqD&{
zn$%g5Pay3t<P%9t<dcbYL_URBC*;$KEr)zoQXfD*8!5LSpNrT^$d@Pj5BW-@9fEvi
zv-@ozwmkCfNE?BC4^kIGzBj3lBj1O#xsV@9^fB_oNO=|c>7-nV{A^P1Lw+u)TOhxP
z=p^J9ld=!;tBEd0ehsOABfpWf^O4_1>Zi!>Bsv!PBc$$y{83WYL;e`C*^xg^+IPsG
zByD}<PZ9lt{25XYMgAOVdm(?3luwYqLh8}TUnlh_<ZqDn74o-8+aLM6q+O2uebR16
z{t3~!$Uh_H8suM+vM};*NLvZ{_oOxCL3A1N8AR8jXr#T4q7!K^qv%Z9+bDXF_8E%)
zq`ie=Bx$pwSd{296f2T)8j6)jIUU8Sq<(^8P11Ux*n;Rp6kC$|9E$x(`vAouq;*7b
zEUA~FIEmDaQJg_+6pC|+jz@8x;m^nX`~{ftmqN;1C@v@E2^3coJ&fX7QkOz;1F6@e
zxP{m%C~hNE6t|O7P~1<-N+=#6r7wzyNEwFWVNyRr@rdCch3F>~j}iTh;t5iYL-8aj
z+o5>M@Xtf)swiF{`V+;Ar1nSg5-CGbyi6J>-XOXj#apDljp75+zDMyPsVk%Si0FS5
z9}`;<#V4d(jp9>cQ&D_IilO+5)Z<Y6Ldvr!ekFA)6n_w%jba9oqxhTHj_7wjg!m1n
zq}u>9NVgc~u*9n{hbLZ*IRa^gIU;G#Vva;S9&<F}pO~W)FT)&zG%?2{Z6VCDNP7`;
zY~qcW<B&Es=D4I?f;koOGt6m78x?a}5+7mCNa9q?IZ0a+b1uU#1Bt6Kmm_fx=JKSw
z1#<-ww_&bG;%>~9NIZ?XGKr5d*C3vcxr58UkUQhbN66hE@gwG5q<w<94~d^K4{&8A
z<Ux?Z#+ZkbHVfvluDpgk9?~s`c>;;eF;67zRm`)<U>VG_NjE&^m9EZ*yb987jCn2b
zCd}(ddja!$S0_W>1{v&#c_(RsdAG&)_u$HR$a`^hIOKhhb^_)Dq}`4Ah^x;bAHxyy
z1sow?#^U?8Al<x}Z<B5w%y)^GV7^B@1M@rLx0v4(3-hl9bmLrJ>Sh740G5zg6RVL}
z0;?mjAXboA2x~xMVXPq}7R4HlbPr<<Phv-`kw`3pH7aTAV2wj!RjhGItd2DviPf+s
zAh9IYgd~>6nux?wSd)?%4Qp}|+hI*jVlJ#1NbH9-BZ>L3W+JgI*32aK#F~S|oLF;{
z*br-R($>OSlEg+>hmr1AtfNW$6YF^5QCKID_Ab_`B(}#o&E3ywsIwvM7OZnf9E)`>
zX*Xk?Puf&i7n1lD>oPK!59@N$KE}G5#D!SblkPjL8;EaV-AKC2ux=tTEY_`r*|2UW
z?H#N;$zUC<yGe|K^$;1%gY^jM=EQo8baP?7NZKD*FO!%E>s1mDVZBcL3F|Y`w!`{@
zcmdWAB(}!-k;GP5zmmc9Sbvka7>9o9+Q%^TGtBNk$M9cZe*R0$_^%+{Q8@H#!+(R}
zzs3CgcUU<8Jr>9RfQujck#J*?(2tDS{n(K1f9%I6-7wfsM7q1NpMu0<*iT1daqJf(
zF(vkklfn4dFF^(qV80~k7Q%jI((Q%)8f35l_G^*B7}&2*21{bU59waU{#3%**k3^6
zGwiP>@h$duk~kdudr0h${Y#{shyAA{j>P^mOBeq&Bp%29JJPns{(I70g}pu8{~;JZ
z0eJ2Iy8FHk*Zw~Q@=CyKUm}<S@Y+`h_&))!eU(5?0baXCF#Znk+P4Yj{|CJG9RfTH
z@LEkUYQQHSAQ+DVKKXhAISKgW;{@|6;FG^dFfG6*KS415H{g@M{+|@g$G`%ZPjQS>
zut7isEf@k+6|HjJP(@ngnpP@T4Na@GNR_U7Wu<jlt?5K1s-bOa(>0ytS>71kamO8Z
z<WD^D#2vSfZ@>Na+lwcjc;fau>i<F^BkV*{APO1Pvq%a&xOC~#rTS|Jmo7~&UAlz)
z()7}$Ad&)6$f!Pwq(Brh!U+HXnSXUYgJ<IeV(0>dgIwz@?NU2#=tw2SepL<|Mp>14
zmfO@;i5f8`C1Z?9$yi5m*X2ra*X0Vwk}<~Q93*3mNrd_c!zc<7!zc>zAD_wt%>Nmu
z_*QrY+z${YO6!DDUFb@eRjCr?as>BP5=YcdQ>`OtQ#;S{tg5t*U^mZ(gQ_eFt^LZX
zREk#AH{#=xQgrY24A&EqQo;^)dH7sxnc#ifLVguHj^lgddvRj8o;Z0<Qi?W9qICQ<
z!Rw`*lhh%BR!qj@@ts^9Oaym6#Q{#?%>d!P(mIJ_y<hZoqLofmLTzg2xz4DaE9(=b
z^nSUi_<wb3(^yB;PP=(F9O{1B*QwSz%YRm@v|gc-s?6Kh%ko@n-48P>Rap)7pel=r
zWQ-Wt7t|q!=gPhQ>M|N8@fNg!iCTyf$!)vQT#Ocu>VHARaB(T-mTB4C6OA|ybZe0i
z!*Cb1kt{4ki4fEZ!^KE3CUMG`Tt~<7Tq!6e4i!A`Et~orEyv+<snuBtL)S6ZBx8)p
z)S#5voC|JR7PW1|_Pl0b+0^&YcBqg`t<G{7x{i@b#sCa}`8RQlk3tMx*aN_FsP!Pv
z(k?}9$B{~k{R(N8XVj*NN~*F#hmk+Pm@9-^PyZf`qi<rn(`n;V2cXNyr@KP9jMdZ6
zkvL-OVY3;oKXuGu85nS0&u91`PT(1NomILfJldu-$8))wQC$yqD62~6+SVwQ()kuW
zm5g2R1gF$6_<i?#hG9Egco&M@@N2p6rS9W~L1`JBFeV?lff@$)1a}V4Q7*V=81@a9
zYxL0Nv!5r?bAO-54TIkBh-3^9g87&5BX|Wg0Jb9A$2!$rYNzv(LdR`N@1ZXAepL<!
zd6uSnh)^HWD=o*JTydEiFL+e~x;ACO3`cEPV+SB?Mb{@PQSnpvEYI>`WmBzf8n>-h
zzGu$zEC;f?dw96JOa8w6WV6*mY_&Fr5rWjm@4FmftF<})b_@aJXIH>?!X`kN<r!_C
zTupmar5mZyQXGHp#sq^#aO4q#I)d|KqXW;5j^1NNI1U+eM~>j!BL;2P|M+e1JL3@m
z&cFRUq7M*uDb+Q9(V(i*qTOy&U7zCoM`U@HKh|AXEv-fE?NirnY!cjj^oecso<q#2
z{Gviss*6Vp?W!=%>K2mCjq6TrKe8XwgkkW7g>F|VEYqJl!U-G#go8@!vM6+`Zu=8#
z+Bz@Hb6su=5~Xya3SCxv)sgBJ{U8Hl@=GzJ{tqPwS&o=z2hZ&+wR~ckCg+?}B37Q;
zjv~Y;YQHS-cE2JS!|~pI_1M3Nd3JD+<=6;TOwKvAOp}|7QC!ro9>)Gf#v}mFzcZiV
z7~cjb0KzhFG2aGiQ$yP&IMwZmmH3|3=PlFZf^&OJncJA9Q`$+@w3BKPA9^L;Cpfn)
z)8zG+0^dKM=y!6ykl+8`AWi-CrMqd*c)O(PQQgLxy*SO?#qW&o^B~a6Jg>d5D6BI~
zZ}E8rKj*uQtl^~vqU+j)_cH>2iJk7mt32+11moX-h9i6{q;TEUfvC1=6i0Di_kD5u
zEYI>vMIP!9%W(VKRb{cS5|zaNgCV+};Oy}paoC@;wLQ9iL5M^xE~-R`)~DkuJ;516
zbU(0t0uhM=k2t%tYd!OSFC|;tFIl>t^Z$;<&Hot>aRRG$d-Ph};1{i-MSeCy({-Gg
z!#U@MVgJ_encT<uK~RsmVc1k~j^{1Q{jJK19{2CUA-oU(GyC#sgmOAu94AU=0}S#k
zFDqK4g?@e3KccJ(-Pe8nT^RO3+ov59R!TPfDY1|1y|PpG))CVMDRI45j>_ITO1Y5Y
zI0%APvx(SjwnmM`L8sGy{Z)#48ModmhwIIyR?keG)>3nQIL2nPwfYOTnoS&iBsSsy
z#g2=X;5a}SDXkN4GCwj8YEU~?!pX`iglAD~HU@T6RwJdazU|ERmTmlo#`@!o3(rM#
zMhFVuG_iJf&f3Mtm}a&00T+=mF1%;j!Z|fQhz|x#ta6E^Eym@0!}C3S4)`Yj?f;W#
zbJMga=e(Xa5F0q+oKt)a=bS$D5apb6)3hdT7`hY7G`UQ=1u(*o;0R*aj@nhtMCxmi
zW$LwfN!kz2RRoAh>q%cb((|_IyNf3aYaE|r2yIO)v9yI#zK06F$1BGgmT7XN^|ezx
z2}jqZo*tM$FrQ%`CvXa$j>i_OvRJD$7fTJNvT}&X<`qfaasRNqZ8>k6jpgke&vCQ2
zwd@DBZvE}&Q<CeAc}KGwE-&>_c4JXEt>*IbnrYw_k#<ZmUFk^x=2HSVh4XMX0J4Zx
ziAq#<;`s70x$*CWbr|!oWiFa^#xaJxPL^l6%6R5TxUnn=Da#wxJMGg>9CW*g-R{AO
zr?q!&wKjK`*Nbtl-tIU;!OtgUdE<zRjvAA)yz$1e4B+{_xjp+1hY4$ILU6*^MLCSg
z^KkY28_SYR3G03UK<4%QpPv@2!3NV2n;M$x4P{m7;?3n2|1ffGglUmhTKzMNW9Q<A
zWm<xBZnuA@V{y)fWttlox2{~d^0cSnkA{e0Jv-u@+ooxWXFW?;rfGA|kFZmJ?SmhT
zrwiJ!hcKp@EY|EU7e^iXJy)egIt0rm!L>dNfcuZj-a0lK%PS)BN1u{FxccbmXuR%z
zIKgJKHJPmNYp?gpbY-Q{?U%iEoQ^}G<#S<vja!VLzzzUrsb<CLKGId=T+Mj=#NNK+
z)AgxSdT&pkI;CA9nCl+ia?9b4FQvb8c*`w^uFHgQQSa^P(Ma#@akNi8b7uF%i4!Mw
z&YpF-a2bokHV)s4J2~Gpfh1V-T8$zt(uO7w9>mb~`u$>M6>)W?*uP#ln<hs`M@I;o
zqP*N1m9=+mHd{EVr{URtqQAmt?rNyJ9H+W<J=d!&odv;TXSGRRPD2(a+Q;9v!&t7k
zdv_v+*K`cF?^|>9>U2@FBis33IK>GZBrK>ktz1<1Gx+Aw#pV2qvwSA?4JWs^`J0CE
zepezXkzIG)9eVnd+fKWP%;mz}xgL)+(ZKce_~>3wN_E?b1ERKpNX3M3T|~BlvKnNJ
z6W9mn#lu)gIojL0MuJ-#g1nP8JnEQ3&qjplbSkezk=1eis9JyDKDC>6jh5|2PIcuq
zoy#sO)H_M2CPC;M8l$e-YqU%TpC3bG3jk{|Esf05)krMDCmQvQ{!Q4CjM;x>&y~b7
zNw(|r<T6PRlJwIJ)sDaLpYg2EOsrcxnrbTcob|R>5%_Q_teER^;r`;WbNrYy4?A3V
z9x<FBeH332C>ri$@6&QwSbyuHfQnyv!_+f~<8a~q;g8uwwc|VW)w)!BqgL1G=GjmN
zt45MB72CH?+&9;gvA87ddf!Ojk-ZNoRsX0r>Q`Aq=d>NU?w_WtR3hK`PMIXGzQe<A
zqv1ZRyBRUTU$vUw`c~6&ltGQb<oBffz50I*ASg%1+y5hu@g1=KyaaVar>^=sS&qlu
z7{@-+<7|pzzrNNUg}-74^>5*Z7)Os2$M*4b{kO+oGESU0?5A9~HspSW2KG>^81W;O
zO;aKZ1$H%%nr-)bNgOA=-gfKoXSlLD7@j;i9IUS3!-%6e>1A76+1i!YK_;h*qP@D>
zE{fBa3Ax(~O`r|O_?tmyuB!N)p(@fMZBJvUNX$At@}gCvqobpb^av;t$4K08cs6y8
z>J+OMd=Y;=A6qd3sOrOyQe{;yFlc%)o7&V0mYbAKl@6KTCLfXA-NT*qX%3?23c;S*
z9>lY7ad`M)?Zsr+Uq2k1t~zscizftg5WSnXD!1+Ye0NK0H*Pt7!d-pHJ&wEP!%u7*
zPNGfDXRjA&Q|2-ed&0dhb>Qv>ac<gJTtKq1dHTYq5nYuaCaMcYwP|3j{e&aOTx_hT
z$4+l<kl6$i)$OVT>mD&p0NVc+5Ai+lGF+x?U|RKcQV0~n%<1Z}Ntp_muwO;4^z|B)
zu#G95*YjNjyJcBkX{s~&!d26-OtEsD;9_e@El6R@gX&zkuozh;K~Ws}ns|aT#3+o!
ziOx3v*>k5O52|xsoNz&H!}L7iMv-e-CpvhmXhqCrO^KAbez3B-K3Yh);ZRBtE$#_n
zTP88krZ#cp<6_GiDaFw;_k2k~ZNo53(r!CGCOrNaCy-*2MB-EwVlu!1DV|^tYs#R{
zxW~wE)&8BjrGq&E9GdD##<)Co@1_8BL_5oOgg!~mQJcy9@E<b62k~upm*b%h=VaOq
zZPP3_JaJ*fIkz8th{V5Q6BftL#4;^`zigS7sDBqv2+OoA&Url!5knkv&aHaZ1UZXu
zi!knSN|WQ=mChL2rrqNXkA=sr*WBzV<=p%lJ^p%ImU)FGihj@cI7z~;n3g5*S4_(i
z_21)qf|%6*h?%e~)8d|}$1#af)&DalF^+}DEpFaEpZ^Q~AilM3?NzEwUi~}KRw};<
z;A`apr#f{N!7g8BLbx8H^Vnm-hWe=*G4<GE_{u1<=iA=qAv#P5@BJSL7mc^S-Mtxg
zKk)vVU+Y198}u@)y4GU}AU$~<#=FCTBRF>r;{1YLUF$D6#Bezm&ZSHEldeG=M{xc(
zKNAvm{F%SuTsRIf+|PY3KY4E7p3m@MoIqCpD#xmE0qbB(MiFy({lC8xX^yR@5py`k
z-zkU9G{T_%I7GG@p^wihv(mqgd0rrZW7vk51BCG~&Z*q7m9BJAD6J^CM|L9WabP^U
zKn5Nq`QXHURSw~(nrrSpk!z~wASS6!r<gN))JAN#F-DG-b6&J|^5ok3QZr(3*U(W+
zgq=<(m^#8E&N<&~jGJaF;`=j`Ew>yagm3M5xVuXf+cC`@rfPwxDLi%uUI7oz`G~ZS
z&2_HRlxmt6$gGZ~1$AAO)j{`ktSF<<!XaXqTB+Hz{JUs4w8~;%sRKGN{`3s0Y9<+z
z&N(y2<dF<aht2XF#pT!6PM%zQkz@=%Za9w7G(=J{np1jNWHmLDjLBqz?Cu`EI+9yB
z`dv!5ZCa@rqxN?Gy5p3b!C!~t@HBuPK{_MVO3#p2hPqN(84hZ6@%~&GtYmJil}>1H
z3aDD>`$Z`kA!DMGuHK6Ji>+o8JCZRM!SgtKC1bK985?q}Ka~)Yod0IQR4|5<!okiH
zWK2kR#*p6AJE7SJj(OtxEp+mq^R;yjmp+G6Sb<#_!SmonwE&JV$bqZXF!gTaInB>?
ze6AB&KD07MWR1uVr^Ee9Yfq6>D9YE8hnu!3zQpC+y(;aqd;Rb;krdd88TGF|1^tna
z-@LH;TL~e_a|Q!Kl7wKpJs6yEbTfo8*&#`9aS_*e33+yq5d0)#QyGQ*{#hxb3m8WJ
z^mq68E<u0PCrLucVDOxTkmSr@&~9V;xBnvv!NtX1LP&Cu<w%l0_Sy!W`IPkJbodbN
zgg0?e%;ilbDv43=j6qdab9zZX?%RT7)L<jkzMkWY8ap~V%ZED4Z5pTEx>x6L(kLp&
z&s9n%#eNlF_Z*{;Q6vQt(V$Gj8iMru?G7TYaIc}KH#Z2bAY!N8U%h=5eIHlJPi2gr
z?{sE-uHK0ml_OUOcO)Gr#`8anG5gM?AYP_!zgf$1u=LIg*lG9Qfk=AA>CFv7Ha1Te
zJ%aD-w>wxLp^(uUFMLG5;A@c-I6{u`O{ixWhQWxjQ4oeG(+}|*FA?Sgl{}yNg<<=&
zXmv$;Lci0I<tSyb2Me$2r@Q{*p1&ACu%xZg>zCC5Z~*2rGR6sP!T`>|v*8YS8GIi+
zQqAgX=o%z>b*`vOpg)>5R=RDd%Sv?@cS%&Nvs4Y^VO!^`n>NkzJ;&l<{ON>5VNE<5
z<_|&2vxBNE`o`F$L08+`+wE2Q`*8jh8=;U<J?i+_@X<eV9Q1v3ob%}W=r=C-h`x{L
zUs&A1$NK8-LXs@(e(@s0HQ@w>jOGwYfm=&0#MaW*yNJHjCvV<=i1DBEag49?5&hR;
z_|QWS`H23{arTqwPkme%3#QZN4(QicFyGk7@ppXTdN(f_13ItgUmy?SZ@{m>d*HM1
zTlfNeCB7K|GfKxtq=PvPrF6^L=cfDGj%^ynwvOUR$2!(pYuzpOmDahA;%GcuRXS0h
z+V_k7O6jDU(!KoXZqa$@Xi}=PPhSfoSGm^vMN(8Ri|TqaHqG+zvOQFnaMnBR{rkEc
zmepWb>cMcJhkE!Zufu-Y&9zo~Fuawgo>8j1Y1cf?dbgcdrWjIbt+O;s6=i5uU;7=i
zuaX#bU|U>xuH^*2Wi7^*>kG$@0#|yBdj@fBCvZ&Zp@E2&8L#-hVF!U1n}#3Q7LtXq
z;`siflF-1he8*WPcHkStwwr<q;kwL2KQL$z@W?d$U+?$(jwgbkN7_-R)eoI`)$i#N
zj!$SnZJ)6Ze}{&Y<Bt5U<B6a@c8%W*$rygo@O?B4L->{xcrJH@<NB=8bQqT&Gp$Bo
znlf-qMClp^{_AfQmTB36Z<w(c1h(P(E^}QWXt-q9e&85E)A9{Fa0ps1^*k<Eyks79
zq58{+i0AJm>RY9We*Nx~s@wWmUGD&JK##v)US`ub{@E@X*1xdhy5d6I|0W_n_L>d^
z0OtI=^Uo6le-nNc-UUy<=MYy%ec0H!K8fPK|LXYr-N~o2^N9=T;eS`)T)r32N3FHe
zxl|7IB=+0?a#B|KZZk?8)AySV+l?b?t+l8O8?MueWHaD_Zy1ufv1^AMErLjFx}2jQ
znNA`a9n*|s$Ly`RjfRw-G?0kcrI9C7;v|u2x;*l|Mng(3JZ`7@<AXtjYGFa)Qe~Ws
z*?xDB1mBuCX5^!3S|Xxu%w^MUxH9C)qQ`?~z%6qjG3_vRErKDZCK7^34C<Ip5}9VF
zAx`pQLr=;^!%bR9C|Wd(JxeJkGM(5r4Aa}AZFSvr;`qp&>R+?0?Uv8E-&%?uxhaf$
z9lUc?3QvRE0YXs&4|s`0-GOSV*L1FP4R=~p<timdjxN^9D$7${)xp3tkji0#uC1+A
zKzQ_jvv=Qr|NSdcIF1x6H*@Nc^`$ukd+0bq20;)dDt&Z~idY=h4X*)(jM~p7Py=a4
zwJi3_n4*+nC?z_STh^VU(dP|AIL>S|iZ+x^kGyapk^-km*Yuxx=~<k@Cfp4#MYr`-
zqYejqC<GGerNpK<)+J1@DU#_gtMLO3slawH`u#2z`&HE|(n9O5?(5m{gplMoN%{$C
zUK!5|gSc)G$RBvXNAv^eHZ!_sH{FFR?UT3etiPpxJQSV$J;#rqht7V5`idI_QW(U&
z4Sk<it1Yi1-el<>XDZqA%%^0GQ@97-48ICfwnWEB-K$&$*J-;WKnYVxwJwbF%u`o-
zFdPgU$4ACUQ&W*dUFosAt}xLNxz2NycQWt?_=|S?{EB_0@S7fKdgP&<);k;8ti3|W
z%HAlv`-azXj|YJ`wc~A4LMfq(lrDa}y+7;s`;OC>#BqrHa6CBMw!@S$dDTbsKN3py
zN5*IftKZGvbfPwM+if>Py#AFu2n6>I2qI!|z(rh@jMbmM6Wr5D|5VCrJ`H`x>DRMu
z-w!~cGye?f;IBgydaw_#f;YmC!@mF^4DF&;M_8?b-Z(O-#0fPR_ECCLRr<e_6SS@U
zLw-0>Yt41-(4!SiR9_|Ph^2~F)TTwDwd2h^qf|iz;Uye~>LL)sL0b5L%BnrBwbrR*
zyq04f9odXY3#7KedJ>UeM5D2~v~=v0kjGdf=Fg!nv#w*!NE~h_!~ZgfI|~a7u|UKX
z){MkntJOl&5*@{wk+_};&QIk@LRyUmlDzmxLP&C#hv>aAAtZUDhv@xMLP+vU9-{Zg
zgplNsyN`K>L2=BOl#DUPI3xd7>o0|7EWI+=fBMs(o~PYK!*Hr&$5wj|d2S?F^D?t+
z_mwec&F_1r&f2f+Q@6IZXoLj!BkndwdwYA#*@y&dD$g)zZ*e0(<_JOKc!jZ*+Q8<X
zxBQY;sNfuc%&*NqO)lf_!WnozK$uZmVSb*Q<+-LhXZ}d>4DkUFyiRLkaQLxYSFH&|
z#P^Ky2_<_;=P#=&t3331K+cCg4=xjNSuH)AdyW@>bSq~6Kd(N^GM~1vyw<vX>BeTb
zvXH(Rf1v&m{J#1}njS|XqgUPKhoNLf!w*A=1V77z(69SE2>n+y#2cFNDe4M0csmLi
zjUy?L?D_H2<Fhnacwv1n_UeDY9uXoOM^fPPB%;qeUn2U50ASCr%|A^p<L^QR&xIGk
zt6>a}0)!d0ZSyFGI@K0=f}YijwIz~Xu5yeNO^I}}bRZ*~K5v`VS)M9rlEsFb{0Lg^
zBK{3m>wjA|SzG-ZRaGs9&q8NVTEj1`okIV7C#Pnz@8J0Jc0J3#{_DT~qA@3)+}YWQ
zbhC`fU;C*XJvyDup!?KSAFp1XFg)_qH5|gNa0D-h````mL-1n&RjtY4J4utwI@?jL
z;~Z60ns?FZvQ9$<N~whJ)54*clo!X#iW*tgDm19xO$5~9`*PvwRV59+<q94~uWlj<
zUx`9S=Si|Lk&Iz*3^6F#3nMW5bE8AMu=d`KPrmU!nq61yudX8Q?w*T+A@@TrI?`xH
zga(lm=p=+O`Siu;^s~c@ot#2l=H83<zUSTrdvU)CSh3(b9>(8=67GhVz-!?mKu_f&
z@I>kTX)H;q6Z&0eQyQ|om_xm?vR=+{DgDh_#!j*NdyCVZogHCqH>@17X_>C;hu{53
z!2Pf|@$6UdX6g|x;Yp*3B)o2$ci%&uyzEi!=$VzF-|^RXyIsTU62FZN=?5II`eDE~
zb}aq|?Ux+rZ=unQh}mtE?`lMC`weAPr)GK`9h^c0J8&Dk5}+DXO3%7beRHeIl{gs2
zZl&6$Y7jYDXsQR@eoNc^|BD&wvg+%ys8an*hy9a1w!^Y-{rkWF`|_|nwA=fE;ng*D
z{C9rmciiZgd_?~z-A8>y|1C&^M*KMXBJ6kGg|oHxTV-R?*A_nYUGI9=L7H~UmMbt}
zP@kg4vl<Z>lSByAiyE?Ej%Eoz05LVYZ#(|%5GSyjt^{$Dc}8tH77^W()K&?l9hJzi
zT9uMLD!0#MRhefrw{m9{1XBH))Yp@Y$*Cm|H1m<fap-Zz8jLOai2h=83HQP{Cg|_;
z^WvUzI=c}L(;S>6k4#IGYum}GKIZWiT?)72+^Rk5tMU1G3;fqQgTD^dUsdS6iP198
zwM&9&=H@y}GiveZ43&~Ztzq+tTo*kJ0%kfD=<5VEay01(#t`{1+YEh$>v>*1cNvo%
z4Wd3H)Ew;0=Uug3z(jRNrT0mH>LZU+-Fu)<J%`>s6xqy`Tu)&<%gcVLY){!`2Edqq
z`}?q6DB(7Muq-q+rxO?`U{By6H2ut4?O9b}Jd6uH)M-9t2Jom>Xr(GijoZqq(8Ir{
zbf(-Zt4+owHz9T=FHNU?$yK%ELsP2D4oAO(pJZ&xB*xi+6(1Q6&^Pnyzq@kf$`$l5
zHbEcPHhi4?pCS19Znx<lCIHa&Xo!A>{tf*P#{2s=<2r$Up9<WyZYB>Q9Y5Sn*9Uyl
zL6)Q;b9GE-sncz@(yorEy`I#QrOp!gg%gSvvPx7Z6IEn2X?`}D?5CP_os{QH%vZyJ
z5Rz%M|K6Q+spIG=BcE5_)Az!__k5MKQWdZ)2^Yh~(DS`Ll1wE!Sjse5taLMU_$nK&
zYfvg4F>K2O*DJ?9cRkOyY!&kPCS$9S7uZ1{1ONao6%bB4Eg=Gb%HVG){`eCA-|Ffr
zW6gXXRNL}B&*gqx_FNFQ%}w*NH6gw&G)+73qPs)Hy{m*XZ}r2RKP)KyEdwykl|2{A
zcBPs$XJ=<ICe3^vBuC3jb`S{96&MOgOpa0eOAMy#iNB)sAy)X1x8Q)|I1=RZO@i_4
z>})fsR_qdh>y>j9gCSh+`eZ@><3iW8a9;alzKo+W-!KJvTO)iY{0`!rO>!%G2dn6m
z^xT+2Gfk+PAgKYJhg`JSI1td0^GQTVFy|mHLKx!Jn3Nd>;Fg@8QV&gjgVfQN?H`~i
z1z|Kh6UuVWvRYsWZ)W#1dYoe!ehp@wKA$Dax#`&`3?vzI?(O~#wk)eR?1K=zCq$16
z0V@VzF#q(b&sTj9a3o1RCkUK{!#(vGUrih*SbTm${eQJ5Bq_q+`PIe1amcIt^gDKV
z!3hGVHyK4yWT`6a*>+D?RBKH?F)`o;C(r?e8rs1qsf0DcFy$l|9L^;TrJ)i}gA@52
zxz5HcTAkbWc4hG8!-o%-t5r#=R?DxxnYzo4H2H1t@<a9FW#gA%&-f(@P^|(O)0#4K
zYbrV09J+S^wo@uP4qP-rvfugv6mpB+z3Z31{H3I0eAitV8=$WIx-EhEDug6d+Zw?T
zMyP>yBUz5z&|pB)r<XX;e4OSY!c>FlH|;z~53j>>p!6lApHwEm4$c`2b`ZHfk$AG6
ztL-Se-4x3^4tBBYl)qATT<l`U$-n2OD8M*~{6Muxrr(H^`pkf8JsdQ>%LZUsh0U_-
z0eEhCvtU{9KolKLEzDo@p8co#v{m^eDBM}v+OY?=PtCCShJziKE1ITo*TEkT{+*&Q
zO;IeA%R*H}xm-XpGPk}4|HhZMu0yw?b%lnh7nAT8GR{%mWGQJT30WJ?JRQkgT}Wp6
zk4lD0i25~KXW%#s^QQ}i)AI|C15CHSX7rHx7qrW=nTJmM{{!i3lC(xl(`z3%uxIZK
z=QDfv95~SSOp~m=qNQFzOYWG3T{fEJ9LlGfa}9b0LXaIS3@-Ts<pPPBi+6H5LFngC
z_jOVrPu>5qZpZJ`%*g?WJXqo4@7@E`%C!QFgBF1#T)__-Ye2B|B{|E4-e}yT=2O;`
z?|KWtd3uz~StcMhLUKz?3rceYCcA`b<#MC(D2gJgzrzn2hG2v^%A;9ytty6>{8k7v
z;c{7az0{rd=^8n|D}u}CREtv=fGZ+}J>xJIsJ;uA5W~WH2T<tmQVYsO+u3Q^`|&d&
zWW(2?H>bXZ>}NZgI3a>QZdw?2YK6^uT{xU6vRtnEAWvbcScF0|>%O#~fC^vmt+KX8
za1ll*Lzkg*=qaWyd_RZZ5@$53?Idj`$^=o!KZmynqkyj8hTric8|P^zDoN_lH5-`P
zYdoZVN1{kIOK~<v^e<~2OH+)uZ$EQp`*sXbu9Y73XSSK*d-T@>Qve7vxcwF6z!(5y
z@M<anacmikMqmmM3B|Jx#!k<{ej)>hyw62-vAn!?X8ZPsYV^{aAjL!w>-BSU!S8}F
zgZ>*$EL!UK`wW2Hcm+5GTP6;!V-RMDJ!_T;>^K-ZRus-E3g&gL`uB@lKs(VHbUl^{
z!^gQT**dKxZ6>u|qM8sX96n51VZUz>r%8Qnois`PFfNc!5z1SU^#;8eIB4liJthj{
zR~Lj9<14Sc@@7+Ix&_72XheU}n|!IlCf@jE&V2Wo^GmqSbc@kfc>o?Sy>KF+%+lFt
zIOG*2d)gCue6h5zkKaL;qies6c2bomp?&td4gdhNn2VD%;~L+uk^nBNJ8`F}G;Jn(
zdUP#a%peI9H66tLc`L6{RgF6KXPK6toXX{=RCU@i{T0m`f56hhW%$!~zx&<aw_QzB
zRloO5^*nR@%<Z?|F57lZRc!&X-lF~X;{c$Xe+!`{fKmxSYcv|y7{<ix*u3w)``(}c
zz~gz5uUUuEsZ*z_g@OWV9sH|H&~9{6S`y^64^7f;r<rxLTB;I+dQ$a0(oG`SZzAhv
zS=PT~{GS<wfxqL2?rP?84Z!;q!tD|mqCWncW?>twj{gv5$A2ghLWw~&-T7DBG&L@M
zv#E05W>0IK&T+!~`{8J3zB#+N)J||~Zv6Ux`Imq3!(*LJR4G}O36wg8JzsBD&IHFP
zp6CDOrRV^<65WVU647nvk-%3;Qsz(Du;oP_|KVnmq;dJP)Y5L2q;Zz??4AbuzW5mz
zL!~l5o%1mi^Iq*VTcs%viTEzhgvtK-N(Hd{$5ndReE)K@2{;c&Z1Sbv|ITr_yo&il
z@owzSS1O=0HedJh@l>%0cp?g-dKdt%S6_fiW!}Z5nQcR8Hkb1l#5+bztw5arA|69;
zL+?e}FRW3-p+I;{Zck1c3v%HT@e&QF;<>W}<rvq5WUAOW>x4KhIJsE`;;`8eX%`z{
zT8gR_EiZOP%Ns_MzI}DxG>p7PscvwcmZPe4kALE*tmy{Vb^T)r`)MqB>RP25N$bGP
z=sJ9W)2~L-=7e9fY<yu9-?Phbv45?e;xIZDSz2*DDr&5M9GfHwyLmxPk0hak*2x3v
zwQIi$-j6M_$acF;g8pCW1{?Qlbl}PH7QlGR433^#W!pfQUMn#2V3-lSlGET@14~DL
zQw2kKM9u#Gd*Az>aaVOdOy*1z+i}SNMzN*V!{{aS17$Pn#Bg>iZwQ6~6OM6}kf8>q
zXGt>>?uM=T2u5MJllY<HM|4T-LamrDtL3DeA_)C8w~qph!@wsm?W;RFEg1pEfl;Em
zqcemWeh^*(-jaW!D1J6kRFT1`Wh?X1>!IgvbO@pL*Vt|J;H$GE(dIUfo6!4s6qJfJ
zRBJ61^>UUetgjL6>e~b(7@^zI1L%ziWl5L>VK*ab%$w!eh%lo&JW#kyNh@vff^sT+
z8YQLmR_YPd{M1TS5%k((mJc!+hDmr{&d<7`l9BA}y>!ydet&s+xv;QMI4H%v-)^m?
zfgwBv$|f<*eBLyPSq5eHJ_D08nZ2deYUw&-H~vY-KW(*IjOqHGKm7Q|KW<@c{g7zS
zSzQPIy2)3$X&+HNVSr>2q5`OdSP#BfJ>m$y%FVBD#D$#TIfK_9SZks6fD#*`I^FpW
z^Z<Gv`VEBKFbV5m$_v#Pz7sp$cDQC#bJbQ`tcW8&&8(F!&bhnIH5!MxH3hOpN}BU~
zb(K<vbTTp^gm~DVq)ig~og-6iw;6D7GrqL%U0z;hGUH~=*Fc_c!~wdHbg;0nP{`yN
zZ$+eOrm1NfH7$)2Vp)Vx%`&O>GMJGTU>s=3(Ko_}KK}8K*SYar;AqF>je;S(=QQ#|
z3u9~iT2OxM@|_392DRlhO^Zw~nYMzSmf>rNa1Z)u8$yj7OXSQt{gcCnkd{hPIHtlh
zhGa{1fz5qhHw^u@Cw2Msu%PehQ|>MI*#AhUK0)d2Z=P@HLrQOdS~J4gE)B+ep_5<a
zBvlEa%vWI}KHK_a+a(6e3Vy`++&*FoU4@XFQN;Z~G0PcrZK;A}r#Uc*SG&miBYwl2
zpX-L94_wsf@P@O*VRi7u{fGe^QT0jspVRO~N(U;4)gwRLm+B9XtgJB@KfiMdF%W_x
z&tm&h|G8#H>q>ME-JA~@t87S+SAr$Is5&&B+F<A{3KeRuNZN5bOye{gQ{UHCui3<^
z(cae0a=vy=HOTEqt!?bV<*qq|d%a#1U31@k_Zb_8rcLzgHO2WVU6%B1ic1U4|Mg!#
ze!Le&*wFJp4@PJgI)qN4bLftoV?o5w1PR5*JLbT493>nmMrcAyUD!$6VaM+XjQKOr
zv{7_?*9}4G5sNX-or9aXs8x+Xm^m-!X(6Q|YPEX4@@%iw>P3#Tj>1cOMT}VwO$YZm
zoGaiBhp5`B<nv~!(W*x2OSM7&FG;zo_4==pgT=lE)yLtp=mL6BTTnu!US@RF^LI{i
zoID-WfQErEc<?u_>9UTl!MS1Fj(s^C6B{g?espvk1OQ&n6qVu}%N<mQbH*&)TBi$x
zIDE$X`uava@fB{P3fdD(dbyl5N@hM^`Sh0AlXK_3VEQ48v1s(Ozu&8?yJ30&BQ&Ek
ztdeA1N!<}C%@cKG7H?g-TviItGfl(uPgy!szfLWQQw<h21f`?D`J2BfU`=C+@*2ow
z>G<!3rg2&NT3WB$9*uX<!Q4HL=$ov&Yi6Caebk`tVf>A2s&QP))q^rzTf3Ha(%Knb
zL7*G4T+?W`OEnGZf@`KS-o$Z9*aeLaiA@ZBNtb1HW=54|{UueeQ9-E`4M0j&x*tLk
z>TPY}9t=?f9m+_qlFUHogf29X#s;K0T4e~dS6|9lc(DorF3HkvI|zLxgY%F_Nb28G
zL?LM=B)@$@=Z|z<KMFF`VKDVK3Xkf%{!pF&6PU+!kaB*d;(N9&gcmQS6FEuIVGGTF
zCU~@3t%Bne0ay)C><_AH731nlCV;L;Ip6n$ux+mnP|r!AAMee?@%>svTO+suMrb=a
z<mGQh0%y;3^V8#wx&^gEQ^fE1{)TA2&ObOi2QW9=%DnuA35J1CZZI#O_n5&c0n8>A
zdeg-MK%qE&-Ru$@?WMzqm)b^L_7&2K)ZmO^teNB&szm@ZW^h{6;`6`w#V=UBHa}mh
zGtB+@HZg)Bicp4*qx0yYT<nUSf^mFqBEtY2bQ@A0oCf7NEM?tx=6C$k&)s4e_;&y$
z(tzq`bG%*;MQzxG!7V@4$ag>jNk}BJ6JMmQCqY;bZRAJ2>pVeeZ#Zmyq!DuX&4>Gf
z(msijQy|HX_UdiTH+}K=EDp`;EMYrtmr|O0`kXA)Q`X;PJCKeN6RW@1+hb9{NUIy4
zLg~v@`B4w=yqPvp4mDc31mbq8etSF3(k!j#ab934A!$3A8O11{@fSUBtkh~XzkNAa
z2S%q)pN2=z!K0GY6Bq9L>A7>~ir)C&6&(juc=RBQ#t5S^YAyC3!lUr$px5h#_JsjF
z3Xc-Xrr^=x=Cz;O)H-X8V2IvpC32rg=);6Z)EM7Es0ecqZ--;3vt0VEfgd=Gm~&>B
z25mQVI$O%RZM=?RO8krQIYLI4kW@|TNwYHtE-POD5G6lx^99=m>SDkcuvE1irC{6O
zfTDl{wp~yhE5aB92JGly+l7JystWi02@H|*p`DLPRhs>+fL#i<UC6tjssL|WFaE4}
zd#Y7wYrQpR-VgYpSNs}A@J-Z4JJ33M5FrI!qmCHpC*pBLJh2^iQ=dmzCUxIsA(;j|
zN;xcm?c8`@(NC49Gi|1wcBB9^q9mrpd`;8VxG+Enb}TTAEcas!BY=rU*J7YPc;jE(
z<;EvXqR>^s7+IwXF(<)siXW$*o6mH=RNnRI$GeRQm;|!&IxXKZ9&>Je|K4bv;yQAH
zF%B_^U<d@4vQ=+d-3d0}-A}=8RyrJZZ9gX03?pb(Lsfc~VsWZf%iCI0#bU+VOO>JM
z(sej9H7A#MJ(@H~rAfV9WHMDO!lBxF>>(%?r<PY#ZKzoYzXRCizg-01L`!HVx(VHh
z?nMY*1mZ}k8a9qg(sS67I?-W(H0_Rl-K#3J3Xb_4VbK7q+n6{X&v2ZgbrRa|)=Y_3
zseux!)n6S}EBA|m7l}Cu6OHbvwY}&#aOT>}?s(S2-~ayi>o;6jAOFMSC_el?NJ7!p
zi*O@+16_%(SGozPV3W#iA@o@TC7vCHWTc5^(u#A-WleLNNk&LUYWZb_8Oef>q|IbY
zBB~*nF0;K;)8IH&r|^&=1X<ul7ZZlDM!CT(r)%m2FarQ{`B#w(p`M$cpZCCZDs<=0
zou=4j8t{z4m@E<1m^0h$gR_lh;ZW(+k|LF+o71FBB?hkogp;2DX9lMLI~G4^OU8v;
zzx1UqRVN$eiv4&0`Jev@@_$QepLJ}swFw`AAv($67R@DzFv99&NRI0Xu$@`N$D^*g
z{8(Hkm`38K{sm1+n9bd@y1^ViLEm{szs=Lkd@<N3s?NEg5eYLrC_1*r1*J2E0?5LB
zolQ~zT{jGxiD<s8KJ~Foh=dvEhNhZ?P%1Kd%4;Tze$$`Cz4Bc;5+1tY>02V+C~nCm
zYKm>!wbV{)O16!Q?11qiYL{aN*hPO}$hZfiaSui?61};5(=&a*n<n}1V|(LGh{l@`
z#rjAKbV(QOLuU~JY|e}5RsGb)fTcjuZ}P7tm-VqrNi&w%ln%b<?4_<bzT*7W%dtfB
z+(t$71*N-18fq?ogl##2<}2foz#0s=dTI+CYgZH2^tIPWlN1ye+$s!hz)6G<;VcdR
zZDgYvw2V%oJJEwNwF=3IUo?aTg>VYw_!Vl$sJD>A4hI2N@bPupb3-}kDl+r}>qHI1
zH}rZvA}RVnS0vJ7?11B^e99BXgFdxKuc0*+PeisJw=XCi6&9k*#;qO39^{(}FZ6mn
zqEpay($n?VIGaAn$6n-7Fu%OiDk}`OZ~XJR{s_c5F#mInBqfxgoIEa)YDJMR+}=r(
zw3#U`2|xunUQ8|6(rt$jdpZRFLm;y}T+B1wyklpQfFt3a@kLXIUr%Z^&vm_8Etz_4
z+jX6>MuRb3pP2Zp!1GA~-xc?9V_{*>)C7i>Zk^b4!{2e;dcE1K*IoB#mpj&hUyj{A
zJAffNaIJ-r&p`N9FZLoHvN@Qg(efp*y^;%gc1NQyxi&|AxvHKT^JZXZSw%h7+MTYJ
za#}ItKk$zh*=PzOrB;J<MmuRO&1TB2)oPZg2^<Ad4Qd;W2QV1J9~#R{=X$<Q+aPjX
z-AcirWR5?%xf%7^x+I@BxEV2MmgW!Rqds3Sgp4}%l(l4PO;CE;q8^waXxGN(;Jj>o
zNy*U+0}0}IatUfsq#ac|dj1hwK?l*=HfKUoWb1ge1oEdGELQ@zIz(<CUnMFTK`P0?
z**QF+TF=S+E|}gysfh5Dw%$3K%hwYD!hhawqrG%kP<oE$sID7apJK6~^!0GcvR|A#
zcdnk#Q7#CuwALPU99=WU#i&YrEYH#GBhCjM8Fkk3Z7$&$Pcu$C1e+&de{P)AYPBGg
zd}wwvY`G@v-(S0KQ7TPJlv@29DO_-cP>WSV5B9z6Z~gwcbLXtve60LC48{W(Xrs`r
z3Oo_@dUYrj>w;2X+!~q&hwy@~+ec|gsaO+~YQN}hS1tyBD+uK=>KC;wUbBIUsE=NQ
zo<Insnu;WpIzqZ}dl#t|iCQPFpM81~RMI^;f~)G20@lD-2y2_W=%y%&2s7x7`M22i
z!y2Wc=r7&(8)>&YbzTVD?l4B_wL;kTwTw}EuMoC<Z~3t>-I!{4p1^l)(-?`U@jd|;
zd%Y~Ap(cPyTCJ8r341-NFT}${mYL`Bm3aS5V<IdQlAO0~Av%;YcCBp-aV@2c-D}%I
z+*^Lz<97y7nVxRl2<GJd+u;Y)Xfg7RYl-s9ekCSR2OUxboK(_)7>qExx!(p0mH_|k
zf`E?!#9=cDqu#HlkWoTeHf6wVf2oVe)Uf^3F(~ljQ~$XheQs^cq+A;~8wf%et)dcv
zV^)OEC7d2#M?Pw5hw%7Grl(3jx-1A{x?n=D)#l%!4!YB(t&(sp?|Sro72kRv_1wH>
zNDiRLYEWp@W@jrEa;xa*^iTvoIKTLS&ZmYkUoJd=kD$W{DV%uxqD(EdXUTi<x+H@u
zhijQIIQwJ+D?t)8w~bFSZrhs=MHnyh{D<HIH8|4&=sO=;o~3sg{j5Cn&hdptxj{#>
z>U#q9!TH-YI=6M(i{Uj~55nWHt(i2FIMa#ExIXqCP;&P;7UE+SciV;~8-h|z)3mPR
zGSj7!E)mTzm@cunJ8_nnH;-YW(M50t&|PRRV5n3VG0)K&z=(NQO1*eu-e9^U5hCWw
zx|<W%16;q%mi>+01E`?qI$!7L_WAi-stueBU^p(8$MK8F*fA1sF4lTg@m}fJGgy6_
zQ--QaGUG-l$Y0-la*n<Kq`TyMn9R6gsH#Ms+p_xmlNtT>$B$Ln-u^V$mwMXF!M-+b
z_N%#3pE*xZ3WcJw-Pm-oP>xUq_50Lm%0ZYjU#qk6-xMr>606td=Sq|Uxs>}Lga<G{
zvuGDuL)UUrDhQC!N*w!7IJb1DZkx*x!AcT~)U=whl!V4@+=;L?8zdvqHN{o8jjXm-
zoHmniASp}}Y0M#}|4xD4biI);(=$2C1jyy5?(FQ}-??)tp93(h+-km@Yu2Z~S}vFK
zrU_&cVt9ACT#jLq00=Wb{y3PMCfpdaHa91ifN6m_G~xBPcM#?+Fs<B`(@~@sV}N>*
zMbY$?Im-kqACCWy5JFrxe<gv|fes;rbPlwDT$y)W@;FCKdSEBXnDOn_i!gw1jLs9l
zSdwN^|F=vXcSmwYi@vV2_^V)Rq?{u!D1&)fQMtjHQl}(pRdUomFC`XhFI(*BK=TBp
z6RPT14HuL`sXXJtyA93;#hkug)oQSSSyQmEm%OF=%ukoxbB6_#iua$#LU=DTbJdE{
z^-rNkxpN-HXbxp)54sH9l9%Sj7ibrB%@XH)hI*W8r#3rjTT!VZc4ng;w_y^WLrXMf
zoS3?%mcw$plzibNNbB)fIl6Zv>O~`>{-3C4v==rd9`(!U+&{QpZ^K4qygm+Jt-opJ
zk2_9rLJwC&k5%gp*QbYT?_Zyjs9bs`z&LohU9YF$On6@PWqz{$)3GI{_n{-`a?X0s
z#6JLn6!Z<eKC6<rbyZ0cf&PmKUn7D5`k_tE-%x%X+urg8(=^TdS5{V5;GK@5g(GuG
z=!)Jrn#cY1($Npb{~83n)2DITzktCQ!m*@}dUzh~LD%^=N&=Q4bWBr79b>)`OuVHT
zhm;Npo7QTJ99iPefvL<~s_%tvd)_k)!|(_(4jjDm&VvUGLOg>5@bYN!7~s|)Mz9R>
zutunP@TQv%T9nWO8e>}FDq_IUFud^xJj1}MX<}@eDmDxc&X>l6wcRMCf_=dd92)=z
zHiU5%rL{*{T~NPtZU^l_$NW68cDngY9na1wZH`(sLE3R!tYV<k1HqOGO8eVqX1H`7
z9a!QsGusD$^1=%*^rKvkz3{>dFR)xL>N}~Xe;+>Qa{Y0DuDBdez@P`VQ|gr*8+vgS
z^sP_*1Uie(qdWJcBm4AdyPc(EA89<UHs2$0+R2yN(n;HUiYo*YbJmuorb4%r^5|*)
z!QF6bYH9t&4}IuEYYWv6z4X#c055&0y0G@4U1u+xuSC;}_)hkqP|)RajH+vxDHNa|
zLA9r8niln|5b<e9!?0|R8nq&P(wMxUf9|1@dYq&rt;e4{Mz86V5bXK}N{$Vm5k%EK
zxuSxRXT0o_no7iH;Kq#RYuS50wYO=CjJK`LDB83EQIx~{TGqB<S<$xH^!~lNC+$Nt
zgI3VhR;z^Gh2rH7hZ5pIm378&DK-X2^Oq*VLllDY5(H{%BTB`Z3ij?{m@LVP!VS*U
zW!q`0*J`zbCv7c2Yinz2&s(Bui}S4#ttzdSw{abKa%$&XTz3WrXR53y{4LGo*l=?c
z@u>{^>#0{G==X@hP*6JLL@6ANxIOWKY*W_?SGq63+m06niii6~T5EC<x&H=>U9TzM
ze+!lJ@tc1M58xuw6sh`=s~yvqrb)3IcwrJOqifa|=>|9yFu*tLX6u-$rJaT*jg^vP
zl#*yk6h>o&Wg|c#IkwtMdb6g+BUp`D#?OV_l=f5U4S}>MyE@k=n^(^cZ)<)0<2C!b
zKRV4zi*MGAcX~kY_yU;p%8qF!c1SW(<9Br<`4lNahZ^x~fCHGS(i!R8Uq>V;t-|XZ
z$AQQpxfgQ;KtX#{1R7f%)}xD)<ISm17fBd@Z=e-ULA1KMy1E(ygpQD7Z=i{b(b&_;
z0EXFV;<rL%0^^7WOVIm}k2KZ}!VzVprt)NT@nFf@u#|P%eL?B{24}LQs>YL3H`ZI4
zn2<rI!l0Vk?OxUwl!nXZre2jNbvt!QRSknPIVu0R(f32W-q8%I0o8?eS-XuX70;2y
z`&i{Q>Q(~I1>NIACtY;80&7u?sp*iB<J@-RH2y%NJ$7OVv4Zo#bw?v>dxr4&EURa2
zTD_5V&m8xLbNP3vBX{)y(zs{MdxYoi9#*4B$cTobERHkqy1qQJ*U@ji=bn2^0P`L_
zds4V}V*T`g@^utiLHp26=<!%qcV9u30;tnOQvEc)VY0oL!F(Dv|LYt`Mv^p3LuDlo
zB_*)T?fp};*;JIKz>0#!)Gl8WA{L%&!XqEoK7Jg&c-w8a<!;N3V@+H3B~6q3ZG>!l
zJ@qp5=bZH2LahEpG!>;e9-6Le#zMd&rt67Vi1FjF{1~P66+J~wTlTlmyToyiB=Bv-
z(3^CvN!A@xRK=R;TxKql_9o71i-L=5Ox{IF+taUJl)!%rVcTbJyr^$;#>!>As#F<g
ztW?q~O2imvrILPt+5b`w@N&|uFiKf{wrZ4{Ntsj1>$4HBOy88I%?n}M+oP(ktq5V;
z^9o_`UfUMp%-g#`P|C|mv&<^fHAX3`G?Uk8e=flO2Y9K(IAb+A;-wN}{O6x2m;cO~
zI&a%T>>#vOTX~t9*HoToIs{&VZy}D3p<6LUCoHN-RjUSzVDmT}ZXO~YHW^UfVfjF%
zs+pyUBw~wDkT@nJ1+^_`Lr>&&>K>|1aZ;}|)ttRLP-LrlV<-%%`~)GUq})Mc%*QVl
z&&}<(&1`yZ6zT!k2z|afhb5(|Vaymev})0rVPW9TfNh0u6@*BV&WwjD^|2ypnCnOE
zdgCt-hK(_;^d|O-Onj4~bQ^@J5g}S@@Ii#v;~=u-i@kz}#R6K`JDCu=jzQB5t$=?t
zp;(+6I(Q0-#i=k_BnBt*g8S_dh7ApdD&2E+9fMM=_Go&Ti@d$tJsBmH$Oz9oaM~W#
zW}%=0LL9{(56V#iO`=2S3UmW{C3+aW8NC;M41IPSnM{ya2`K?*6@*r!f=N+)`EE;^
zNw%&+yyD|a+BjQhW(O8-yNClTz&FNnmNj_-4mLioe_7KQe3xk&l*=tp3qLEU&?=W<
z^A3lt0Qd1ve_g%&c;2V;K%u`NLT4nLA~v3~|8DROrfHhSx>HlGC}3O=?$lKGl=&wA
zD$&5T_Dj|J!AEaZwnos`OVMjtwhqH0k(!CyrLOr(Hw&iNHmqs`c10AM{xAal5<jte
zYBj*h)10{0mE4@Ff{Ag_^Axon_&}+!CeKlEBnM#BXv`l!Msk_Po?yirxFI1R*-o)W
z3i*l1F_0(vU~cd^f93YT*U=rCi5L-quieS^VtY{|M?5#lQ{(0E<(~9OFJ(#E)=_M6
zW?NWCZYIjvjMJ*WWjV8RHH-mUHsb)1h~i!cFs$2*b7tp4<Wa6{%97<F=K*?=#hqG_
zsoIdb&G(>ZynzU*s~PR?P1%+b<i595RsG5xgi?MV=a)BMgI&s&=l*ho3Dg|E#Oa|^
zoNDA+ZtO!c&9JZk!tH^tqXeDUQd`?5hT~%&p@-NPWMLo<L-gEC+b#;0bH<xsM&=*}
z`PO8Xy`b$bUAqI`HR9PJrz5{N#^uGdT#X!=<3F1eVISZeD{iK4)N8Y>t-je>3J0zg
zH{F9$#3o~{q~&eY^T5>VhkW`5t`nc2z`2il;P;iSMx`?1yDr<qO1VbD00C2hpQ;5X
zWS6X$6gMemlx~r}EWZmu0n1$(Aqh!aBfJy7flL&kB|9c@(*TK>Y9=-r?akC<`Lj|p
ztEF@&B%}NXQvwKdMm7~$);CR<WdP=YDzZEv?E8{~14oj1y*4}k?rg2jWw^*>iH=8_
zM!|&M<r3dCLDRHNT^j$MElIXHozDa0^V256TjQ;-fCS${51{uT<Q{#OeF?6muIr&K
z_u8#cC9Tv`N2`Wu#-%!o@I)S}&3N=m${mQ>W6F-FX*<H}tRHmi$z~QdpVb74SOkPm
z#68F!$?+#OQxNa-|1{JsTkFm)FpxO+0P32|4cGR3N1AdqkT7=rDP5SREMsgiF3XB)
zNHPFm;Wz?1vkOeobvsWKUH5F?ckq-)B?*oLO!*W`fSq8{rbNRM#xaM!tuYlO8Dy0h
z24k`$$v_>y3zb~X0Pv;(%DIp;jVQnjRgr~Bse)xi!2q(N8V2JsU{F*QQ^#E@=L5rU
zmJ4}dXqw6mqAJE805eoY=7Lg1k`+}|%}|pNQW4tP#0Z9{fx2iHI)bi7HzDNuy0GYA
z*Y$PMI5UTCB6prM=o1a9oT$W46xAe~pcYE`j+2XFmWG3%p`!fUB_J;>99YhM=iIKq
z#y~?~t@R-eR(rm#c(Q{B+0<!8d9VrcBY>}Y(y}ba<Ztk1W|p<VF;xChMaO~tb94Af
zO3=7tU#&9%UJVt`*Fp9|r5;Ogu(!~2oU0U-s1r~f(Q8|kGKg*GkOs}Uee@ReD?Nlv
z%hSFgni8;c$@z<clyQxgdeZD{du<#i>gyQs?k9!Cx(qf1rLL{VRB>h9vT$fxP+O{d
zE<g~7`o}^y7Y@w+q3!uX0FIXGo(F+HVG7kL)@%ESQsKy23>VkRWm!=)FwHPFeNiqq
z<fL5Y{x~(|s)iY4%zTT15WbvqX0lu{0u8s7^JtYUM2}>Y<85@DhmKYXxzs&zNHNyg
zTIn>_#i3n@Xo-jDs0t=EE%|J)9z`Dy8NvgL+>!C{SG_-as^0YtFUrGWkdd*yE9j2#
z6x|sY1~J722uJDWFv>H{^}?5&<4YaC6N`Te$*t#W*-hJrA`UH#fT*o1`Ml4~+ltB#
z+V(-FD)#(yJ(Scbm-SA8Az^Z8OFY&GzS1FrB?C@9tvEs8C{OEp?*u~dgF+nSLIiN4
zdi?9q+d{ow&$MlG)6PZ50n@h4o@!aD+-rA?VTd$TN2d{LATW;9e05N_1ZYyTQo*@*
zP4%6jAkl?bWmMEgi(os8@EauMLL0++7(lbAz6MP&*P>pp*X!L-86<5B!n##7XIduo
z?)4M$aZ|?bE&q0bj^Amw{WNY{)knUG7U}ev4BLk(Z&&vNYgwqn2JiDZj(Z!{aC6PP
zu|q*v-;l+FbI5Xczup>vdVg&I=G_q!+RM$D-|YxSsDMuEz0Z3SdN)GK>Kx1xSYy!o
zd_fQ<Ntm!=P;u<l$7IBp!okP|hHKDjsU{if4DlM&{sW;zW0zzam35Mdgx^RQ5I_f4
zRwnB8iItUu&wi&YBuS#{mM{z=6GgFNNm97$GDT7h!X%YQx=b8Ox#1fuFH3uZARw~D
zDwNDHz%2320+^QPtb>~HMNQLWAyiE=gz806#F(x>cx^qX)pzXNw{Pc;dJUfc*8hrO
z7_uxYs$!^?WokB-hyW0(0$4X?QeZS>rr@ScER25?hM`6&NUq9snaJk&FVtTS_J&+8
zCsfJE2~`$rQ55MqWA9yDx_|T9Xgj(dJuuMi=+BY65hci&AgStu6cZ-l+5EJdL3+g=
zEH-$Mhl8+_CSG@$;<TBTj0f&zB<0*WJxl?5Y{nbuK~Pxd{CD7q>#x6lIT`;#!8u1w
zN&2|&?ndXy_e~XxplSRGu4&jJE{<2@0bwqyr+o*Y$5gyp4}eJ@8lOCJ<j8$$^HQ$A
zUn9c!I(FC8x4+`Ap;(-fxu$6vm*aJE<Ob))Xc`@mAH{8&8ho%rUKoTH;XtT7&M=e}
z5UMYFy-=^v(qdR2Qg^??kWw*No`}Ko>WX4X@e~Jx!C=s`J=kJ_Ny{m!j>d`M13_tG
z&Cl(cYcz5_|0%+tHyHGd4@%dLqD``p%R>+R1@E*29Y=Q~<ko66AWH_|W57-{3#~w7
zY#OIno#UOfV=-@HM#LoXJ3hRO+k7ULl4jh9NO^5<_~7^^_+WKmp=wec((^WtsZ6)v
znqIF5=4m!?9auG==`j&X?gFR^P{q1#f_v&$0ekjG!%vIXqJc%JXtkhCNG#3EnL`6X
z=_*Qi(6r8c>=1nleSxcZuAc^B#*xsf-zKcL^Ic!Q1RMdG9E^4;YA;xZJM2SDcNFuN
z8uA)pnoWUxG#pvC5hLY2aJuSYYPjl0mZN~;%*cuPjaUnDjtODg%R<;T%aznRn{7y4
zE#;VP3$g5haxynDZ#gpW@UDdx16{etIbL{fw+^4}fy4_JS@;f9&tAA_7n44^(mLK7
zJ9AmTmEO$K7;t@PE;j@(SkR}@+xt}l=|zaHmf(0fLqT5<8YDr*KhBm3V9Nvj%g0Q7
z7(Z+)CGn@Dc0JtwfB*M?7^+GNaG+WkZcp`aB|dD{I{CFwJ37AYTi^N?rMp&TfgOx-
z=WF1g^XO6ZPK3bIuySU97L^2{`kt&HoyuYHn~X`d_9Bj&Z|+l5mKYSCyoaEA`#Z_v
zlVedS0yM;!KZWCXso^OUNF5n;8&W6KZg(YkZf^hD8s~M_RZ~ADgPB43tvIghtBg(R
zI%DOXjG>E<t(BG_vZ_lZj?(-~P%KWVq-d!PREQc#%vip$rbU=KiCk%hu~l84WK7pf
zJ9OQa#jAtA6rMou{?el>ce5QWm|SN7{iD&%Dx&MJ6b}ZnIu2BhSO3GN<BM~!l652-
zU@~d-npsh}Km3?~q&CjTNJgeLnYNtH@kclyHFL`C6BAaFf4MTI>y>nh&o>(9#I*$L
zD!B#S&vm-P-c>pslrG;$vGc2{Rjj2cP19PR*6_ljYn6tsX~V9;rHt#=TayVM+}!&S
z3WIivSYtaBPEa8ZniRa4Esc>92jaMzTtvw;rFPr~M8?8b*(%L>Qa22}PvqDsg~V!q
ziu5S-Ge|Tx>fI>q^{QtCdQVNZ=LDtG_k5_{GkuUA`@idu6_o-)QrI5vh5?MwmFPUW
z2fZFWhftW2teAeeI_XkYM5LagUF0J!r4!S37%0?(9%oH=?7jVd={+JKcC=Il^{LRy
zz*3i5Z}7PHmnntOtl!`B^PRrODg@<y2BnbUP2~pAQ)^zMCSp=Ad(WVq=s3C_y%If!
zK4~MTxS$UG;PXYvn^|NdtT<l836Bm=#yPp1<oe9AxZw|5CY?YqosL`XHi*Cwp1X8%
zCmQEirsP6h_hgPFcr@1IGr)|jEuZ@0lJUpa*48LptCDKro}#VZQ|JLPpeE`KNSAok
zrvr=-#-b=X*cLF?&R1C9_UTy3kJi>s9j(2O^3@*Y8%G7n@kd-{FO1MMx&hsX-iA=a
z)$58nPSXp6wuu9WaKpxOOjVQqV6h8t0)?IV&7?_^xXl<On3MK)Hw@Yx<!D3<j-;uG
zt`c_aRQZyFH*+hTcxF2G-JoO;B1w!HO@DSVtx`g0HC>$Zng&x**vBXZt{+cN$F6hP
z#zu+npy{4%`@UPQI19YuMy2ffzHR$d>r*OzSUqmV{d)aSP<k*AIA@p+kr*=!LL}YH
zgJj9FEL#%fO<f|yFc_1FBX|x&o^$(?AzxOoVwk|aO`rNZ{v~)EKo|;Ez8rHL>S5~*
z=$XjT#KBFU7Zw$Jl$vZWX5ILdgiE#3p-#B6WGmF+%A=qF$ZzhXqN;ugE~^!6&$G*w
zPl>83t{GEnjl42Gyn}DLRi^>SLtpBhqg==3xgWTkdmabVP%zecAMNsV-RY;z&pxN)
zt2wX0T@XSHp{=jrgSxkRpJn5vZb$c{*Q0l!51~)mz+k*j=_Sq#Fc2$@iy}$z{9^b~
z^A+0r_14JhALKrbLb8>m`qlP#ho0znW&m;{7I63OB@|jrhTd9Pi(>~%K7XWekxelG
z7tiLOz(6>M(eG4V?Yr#@+$Jbp%j@H+_=ohfdb{<u_C4d(`ujO{V2HzX`EGPOLXBFj
zw*GY#E(%#BYo51>;Nqg5HDct+o|CMMsyISPM6nobS{#LQNHkTM{!#+rxu$Ma@V@wU
z&y#i<#M)hz^)SDwr5lEhe5bbbtr5OYmZB3cl*wn%_aooS`z#>Q;w|m)?)+|WA7w%b
zlaO~5cRQ0Trt)D39O13tl$eD1ubvU0n0lACzjjaRU9EE;@5n}gL)#XjXWKm?Y&*ok
zkFnni7ZV0xIXcHLYy9J4vB*7_PGDT5RI~)8MT{q?>+wIsx#!X*_xzlf_v(-TQFFe>
z>t4Rl%)j+Vz224ZwJm}X4(<zu*!UWHwk<?>Ux2@JErtGt6((<B*W1%Hbt(7uV>t~H
za;I)8<NY3|ZrZGV-OTk49`d<i?Dn}~aDTU9aR1PkZBXm=u0gJg{phf5coLv1alQqQ
zRx}hCg?1wF(-h0u3zH&c?S^y~ve^y@{aG5eZ68zsXjPMcf*(;j{dZ(XFnIu;>?9O}
zf*GSsP}~#*5#M!+77C@0PK`&1%&vwN`8MBe6E<OpN@xY0K_zs2XwtYry6zKo;-_JQ
z0AWR8V_hy)i5IrhW~ZAeaXa?PWSX*8YS`3)#+DKStEVC*AaU29PZzUJs>Xn}Pe(*j
z^mP+-oz!=pwk*?rxK^tXodTHax+0N!2je{w#BlYVJ$odGw|D*X<7@)`%uldK+8#xx
ze1h%M2Tc1~x{eN#M|O+Lu0hCE<1}p=PJ_WOv5j%iMQA-kz~>nn5dcTOsCK#-c2e^5
z8YZDL+7yMF`iS5-MJo33<MI?tZR#8UOVEyABuq5AMaO}5bb|e>(=_fgx6n>>7aF2J
z^1B^(nmB3sKoh=8Ow<%ZM7k|;JI+K=^{|xBQ-5@Oo9o1KLq9DhDOmV<)6Fn_FqW!X
z@D8n0cO!r-JE4`PhoQ2HZtHLJu_P<1Lez)1ySGysiSp=MUDrx>rZ1>08J|VB734{7
zhq*LA`{TXh_S<i#r&GAlTV43#K4&3{s+^$sbBr5oETb1ozAgmwqr^QQ@}G4J-FDmc
zO%g>_#u|lJ`s^cANBhw&=n0)K+jc^QmdnZ`6fH_UBg17Wv27(W0x95c+DTz`rBO^M
z%}(kghRJlN>yVhZktE5*NVQTBkS9fgMt?OwqEuYf>m@p2Oi|WtPexG`#}yCI8^BZ=
z@<oR;;1TJ1h4-(F@8>Ut-*tQb!u}--O0$t(b|%3xIS6ch9X|>$p#s{DR?&^<5%j(Y
z8sm_A;Yc+{leB5kYBM8hR?TB;98&%Cmf{h(9dd-D5-C+RgRqcs@HmhWv`DHZHBAN)
zlgj?!VNaO0y<1#KHV1D@bGH!p^L(MvNYh56kT(>?NZH0!^()Ab$AD?uLhNoH;&B7K
z`<J8Ci?nUr+r9ftk6_#Io4MbLzb#ZNK>!d0m1=?Wl)|w-eO5??xc*JXl{?yS_wMry
z2;rAOcYW?UQ3(VW4bVbzMkL@8VEpe<+Lbxb>F+RnyF+}*4?p((aC#{qlSBM;OPd45
z;?&dAUStaW34k2QqS=#G5KR&rZK0eTc}SJgm^vwA3n6*y=QjPxMbtsxiQh1507_dp
zZqCRI4sv)#?7`)jWi1g}t$7np)G%9tl4|tdGu>d}X{#rXTpti?EV58pao)m%v%%na
z)Q+Q%A5P~jv@mLfSx1{}A#Ruq9FtSUBJ5DOb-XW(W}tLBaLTRKYSdp`lZHgZxU@Xr
z285Z>R^jhw&q=(TD2nF@m%YX@83;9^4DWK_8fx`^IDW6k3+M=PuC}#<z*8b06Wfwz
zOiY&Hlu0~~ps&f8$xV|cn0f_!DjmQTAfXtvTg_TaVS_rorW8Ci{<+go4*Vh;4p2eq
zK;d<qQsL)WkBqJ14@5`)mG!YGZXGUlii0{w`_&1phaN@l(x~3en3+wLsPTB(P5`XD
zX@=xWs-|sb-U1+Cjg@r`tF{T8vwv{4dW7|T4B6u{r&z4Hn_fJs6^jmY_Ym)5*D-#b
zVYr6p3eJV=8LnZ(oJaT5GSDt`6lIpQj-k0ktRn<lNTe2d6>|Q?^aKw}@}Bjez*o<Q
z3LP)%fwi@@01t3*1`1|SI93SE0)XQH`2GjEXE(7OdSLcVcwnq=FB{B0jBqeH8Q|xt
zwl0KjR|}R?syI%i;y4wRPf43CIaXm6N=wD}=IJqy&!VFm+DWV3(pj%BhMxK1^Amx&
z{OJfnkJQp;$G4RvFxLYEsJ>qb9o==q&~<eutoXhP5SQ=ij2&ve2!A!-z9q*7=5o%t
z%fNO@;5bD`2<I9DLysOfHb%S9W$1czhg`+OLDDhjKJHq)Zqfn85s4ElYV2P1(=fAY
z9mUlP`$gND)<_Na0n!nl1BUQ^C>ixHUWe_v-easM$sP)(2@S@6`P*PZ)i%dN6UG|&
zFY51Is%Itj0N7GX)Bc0k)Y+4@@gE&j_hi!)+tuL*Cv^RjvW~uJ&4EkYtMphNn_DN)
z8FT}>p~s!A`KcR(l8|vEF?OVE*`SjOsp~bxlURBDgr}0)<f>V3Ni;A?QgwB5)NLOl
zcy4WN@Q;B=Ne6lZM{zPRd1-b4z3cr2UYw5T#nuq*K}US4JzV#rH<y7~<o$Q-uGbIr
zjC$#n(HBi=I5pMTvrsr|enzDzTM2%Vc(zyoNLCB(fsbESBI-uU_AlxIdh3_<T+|=p
zcGAHD?y6~}lm@y&t5Cw}%DGvTgHqe?kLmkLd4+mG+$&eR#53yv-6@5?0;M0P4lcD9
z9Yt55Ytdckov&=5SVNT(nIz&r7pls^8iLNKcpkJ$*f*M31WoyAJ4ua{(TRN<o{=<T
zLO;WCUf}TOUCO%cP9*#I8cu4hY6K=N&5p8r!UJn-t+lmFYisZTl|X90fRcnL^I%LB
z4Ifr<__`33_8_XZ@|qP5y@OHjuXsbbS`;9HD%KXolK!0VUKpUA=+)(l@&j$E>&s(t
zbBuKI=)qbeR1j4N7!3}K!6s3-gV_MYkZLnqcOt6fv2KV;ZZA#aW!DAIE?4xV-l|I3
z_P}+^T+^T(PfOE(nm6~e&H7U_vEw3?-PQ!xG;rPWBg}(PDgLZF0Pxr&W!D8w;}heb
z%3rIT(-%bJ>zXYz50?NH#L`@ewMkswg|yrp;f|6C<ALi*v(s<kndv>PJ<~I|1?9?Y
zYqnBet4N!8@k$^WhWn4MVMu`@D1}K))3iO)GZ@cI@6oi`D5R^itA#KjDE*@qo&F<x
z=wWzkyfwo6;J4A0QsU8!WJ=S@KsAn^!-%4fq|zy!g?^h4Wm-B}j5uM$9gfpxq8z!A
zrAyAnT=i@O_UkPGb;$pB97I4f+Y{3LpN>h|O+Hqu)rjk}LV@`%sawRBN#IMG>_$yf
z(jW*F)p43l$5oUd08KKRk*jFf4^+h_7988Vcdx~SyQKPothuI16fdBLVwHHq(qnG9
zX5NYFRbi^GuLlY%@uF=M4b!oNDQR%!;lqc0M%<MZmoWb@b9`S?gGmQMUAC431^h{+
zig8t*3_uB%Em;r2nG96P_Z{}Rg@pyvwOM0#AO~xhX<k4TFL9XVF`3Uf7PG3#gl{`W
z)7MqEW3r-WX-3tt4AatF8zF=!LR-A|=bu5_(Q$MwdL?=ky#xI^`Zguu+WJL48_(;#
zpqxr=&85#QTw9VW<v4f+8cfd=19P3(Obx@-OvkJF{w7({A6c<&+!moXT3V6k*B0Ee
z^4?;?`j$_9$<3c`pY!&-Ro6>rO#t(3N!J-$*7aq^^q-#LOZhmSI<{yByh{FD#$b5y
zBoEmHOY6ziz4G_jbPG+z!}c*&2GfKxV;A+PmvtSrh4PoY8DS+2UIC~k$CsF?!Q!Y$
zcTGlxJKW!3uFot<IZg+c<SMR5r|Rd1qK$34!YIk4PUbd<UM_C)_t_NI7bQWffZT`_
zbZh_8_TsUrIL?2`69&9|Hc#8m)xqn&vBi7WfmQ$MA*9#+-{yQCed9}K%?&;ff4Xj#
zdTC&daOb~Iw-(Gmlw_O({zCfw7O+AShYxOR$V>r!KpRGOg2{=o@b_DSGISW7MR%ar
zptqvuT8FT^KL|5|%{cM8{b~_-6++ZjTD?mt-yz+<bx}~-gHO>L;@k9ot~mLt{Cnp<
zrD7aG56poHn5iG7zklY?PILlYkM2f~p{LQWqVLoggnbS#^@A{rcisdhr5FG0&{>Kr
zP#w+JFtSHZAIAR|fp7*5C%;gA=i8b$teIoY2buKIzde5ty%Bu?edSFcQUWRhRQp)R
zgA(w09ZKePMiM6B&cI!l89RN!*129gb*9y773m%(;2;Vxm`{jJY-=GPkDK3&<CwEP
zu5<Or>ULvtF{XQo6gZ_m6YrwREVmg)Pg+w;OG`8Ip);8*(O(f7ix<C$_rM5kLx<z-
zy7))X+tJ6-Z&GQ8NE*jMp;Y`9q&Omf>@>~RaT?8p5MR)>XN8bvH_Jkejx)PiHxsYC
z9@EXp1n`55>h+z0Cd>7oW+24ATOX(N0Rg~SxzeSyTPZURARf4I{q@%`20o^~{%rDL
z8_X!D0l?hy@~ljQ6$S%{w?6Z>z35lTa!a*pQ&Y`mD9d59IW<+Y)Yj={v-#EH^wgy9
zpitlai}!x??tYuS)4_rW@Zp?`Wz)Z`-7E;ZWLLM3*++zza$MTDsvWPV=lfAV?fJH|
zy_P^JnfV|O<wNN6qbaF_*o>R`L0AU|Y%36E1w?lacQ9oj{j2C|-LcD5&j7R>0e#DC
z>bm(;Q`ezTOm0bvg~{?yyX8)l%2t1KC|%iVsQ=;|CVTtunYym)=J}t7LNP%WTEo6|
zA3_(=H_#uVzePU>862pNr6OVeU~;;EO%G-pxs-|g;X$QslKNi|_z5~MYz?<{;&z<2
zvERwYot}q|f*+lHA=w-fim$alWOs~(R)qmdSEdJgfNX_#dF#3JPKa$KdUG6Cgv9T`
z;pL9txE0o}?h^*N^`~2UmUWBw>ATyKmvIn;?A!|h?nP1L%>(}EC4u}*>i_j;qksE#
zq=CII;nO;7aCn{*<~m*>oFAeP#bER2(WGe$PS<a;0IXX(gc;_|HPyD&+RdiHNat1y
zz`DI4>r5P~tG2C1&%#$J7xuGJPVWf+MD;_0>2l$AebayP&TxYATO;^;7$Ju8^3UXE
z^s4{sHy2U`Fy1DN*;yw+LGz-<B@N1I_An4Ac~&qsWC*D!i2FeG%vvQix8E!fh#~OZ
zt6%q)x4ebP?j=`dN6Y1Mgdxuj!am;{M<w0!3U4WRzGk;G`?HqUK7L7vmTf=yCxwkx
zt0jbO11m?by}R0XK<{*PU?)zTFdLIS#XM2a+p{;>Fi+%!X^PzAP5PrbbNHaMEg@R1
z^Ysf6{{egw>nmdN6+}ZstpOSk>YWU!T}u~z6i)f6N3^}}Zsyz|Ybx#Jl{T8hv`EFf
z_|eRu7r1pm$0^P*DEg`}omV@iQ2ipX8RR;8?ORvBgN@i9|GE(3Ljo(WR&dx??RSLc
zRT$-_Km!jz&Gnkqb9BaZr)SmHO&WtBUQO3qJ4V!;oHXe_PESuytLp1;fz2Q!p-No6
z7s1g%^fnZWp_aTYgX%=%%7?EAg<Z(Fz6G7!8q9hzn-BOY?aDI%c{`QmGrVWY;=1am
zc2^WYpr>tbH2>7sZV-cWCP`$eyu_EvOGJ_w=Ug+gG$ChM!;vvXyo)R_ci1?bC331A
z9q8_Y7MNf{K%#x)p_%(NeAZ)<L^RI%$ytEelbjo<8t=6N9uzjgG~qmTBTB`FptNrS
zcFtkw-9*ur`4uLBxvadxrLQ&tA$e<SYlL4F_u}JTfo?$W$8#saIPj1Iu<ipv=+nxC
zD9WA&vzJnY1N$Mk1b(r1a`nqUXklpMuw>QI6d6}b5yvb7ZKbiY(%6n$y#qJ2auv&p
z=9E6y(EvPg40bxUl8PI&K)-g#I9gj<BdT_718aMz!P#Z=0P~kIZlG+>G_KWD0*E%U
zwpK0PQR;&}-295_)AnQf(VH@!I72BAiF%c!l6Ay{(if)AsOB}3OzCzanrJ5pe6(oW
zSZWJvVSv9B)EN>}j`dC%K;&9*tug)|`Adr0pyZuUC%#|iTQvOI5%$HhdKiE<U<Ny5
z=b7-*5uF`Tl;6kl0LzbPq}Bw=&<*+ovgFAW%+C`vlV;*uZ6GLE1@LQ#kpabV6?N2c
zwU4{jt>d)tJfkR=)0?210>3%gV7c6QjZu5(X;lb!rRdlI)ksY*&1W?bD=OA~Pwi%f
zQvue292t)ePPm>YDLFR4%dn5xx(<$AoN%46+}?~J1(^y^u%}WVzi<tQ5JD=p!Qu(D
zh<2mn=;{b*saOC`V8fCS2y)Wc-JW$b352uIa6lfw=PYKMS3yZ*&&_fe@~~d<eH9m~
zTOK!5EKZ$jwOWSf=q<x_C>(NVO5sNvbGLq6?^K^-!N@Z6AykOeE}5wfEf!(?r)z6#
z4i)Z$v6|r}^itcl-p{Ih5HA3JUw6{bE_4Rng6=_Y^gd5vGqEdlnYz93xIFK05<~{G
za7${e1h=Dv{H1`{L{gWq`&40I&?zmyaj$-~eTv=J*irr7UNJr%fx5awV`o}Y`~8Pp
z#g1(DJ{O{X5J#JcdR1`F;6tb$(ulQdJ$!}X4v@w!ZuRu7=rQywOmBnT%t&3ZoUMl;
z%B`4K*9aM92ne2JQPNxZYH#GTdQ09H;&!ytE4U=75#a{Umd-Ia&wg*4lTj+j5ta4!
zGB^-hYzPB3MP+aQ1p3qxb4Mk^{gy_5ltgb4UzU2)KX$8#QgqCN5i+L@aFc-!On`u<
zX5DHM-A7&HWA$K@#z_Az&QD!;#YviNk#XAl|7H8W{nHZWsZJ&UMYwu2EQsAljH-v`
z#IJHbGhePKOn~Ej^Tz*T7|&iRIu0!C(WoaS<Gx|tTEVzl{M%?dI)iRT$S6)rXro}{
z7svVxajb&LF_J9|!?fF8Y~wFq(1kQjYeuD@_D0w~k=wn5X8dmw{D1B`<Vi`o!7-R|
z)ghM4MNuxNuN^aq?%tp%yES22w*UY(WY3e0-xwc*QA0$~7e=DZ#>e~Y^OnZE@$XA=
zrU~1;s%$%sU0&j@C07*nf^Beq2Hx&r?7a>0<D++D1xE2lP!JNtF;y^NbXgBGddgNK
zSquXP#npgBkpaAL!kZnd)oKR7hIG(4i3V1LAP+(THc1(Je}uIZ<m7#hTB@r1<BPto
zs@75N!Gu?Sa6u_=OH!_A6?3-d*<sNx<|L`j{nLETd*jh(t0W3y@+iZ|<qX4sKm)Jy
zVb5L#Y)g{Oxg|?BhE>~pI10B1z5jRB4!08>oMe5OZ8^BvtR_;j^$ARF>8NBFXuT0W
zvNYkOXzwsIvVFNZ*f#=b#xnW**YD)d9*O!{)Nb3MZ*V0iosQ(JH%D{vB=h;EYxYP=
zrCr|-*F#_O$$)`yVgvKQYB%GEV3$->?>GPQ!`h{dP8vX>xIO+OR4xR=@FpG3bHv~K
zZK|pcQ7M3-$+kiV26t7sD0<ugVBGOTi!rxWjHc0E+qjgJq{1lmvC?5M$tC}muB~D4
zyv(rIk_=_3KaO2!n2ouWl?Cm7Slf*QP_b%Xl}w;*wlniSh%O=oX7$oYP(R1oK^WX^
z6f%M*ARB5fQgvLa_+v%@k*Zo^6?cW-z*QD2x~DngO{2t|=CXGvT+KPBoY8777jxs6
zb6{>3OZ8flGS2`E*ll9hH7PY+*RGPHRpd(UCt#%9u$y154algzSSb-i;n`lWVHgG_
z<>~49$REEu+ZI}pCb@tyc%xD-#|h!NT&d`|u2T$?AlGZ{mGs4}Vn4)J|7d*a7&`aE
z{%7l_ct89a+J=rr&=icZM--y_f-sJrS~vSI%=I5BX(oxo{=b!iYrBq2%VFqT>)Vz-
zKd)Q1f2|XSWhy(4bJ<i}FP43uX&UqWa<LvyUG@rkkbgMf{hV<>6}IDMy6O09YrbRZ
znd{gh^*Q6)CDZi6T+X#DH<t@N)0DRV^ptTgLN9g4Dw;&6(0PP1)n|d9vjrb@>{EV9
zBtwjRgMu8qx;t%O_;XkL(4je?Ao1+dN#oxzO?#?X>|is!PP3!Z6TKdw!N676W$x3#
z^~&S#mOU5r`wo6+6e+Dwrb=&nzO48=@Gsj@|Fs4O`~7~ur9b`Nmu#=hsB^mP*)P5K
z<PX{#4u@1baYCa*gtxY~M(`MnP#Bx*xpkt3ZqshKn|70CTnm#hX{V9T5C_=b8A&9l
z4rM6_YQI(s0;vo-730wqA>{u^u3D|ts?{8fbX|<^jC>Ek^CNhd(De^qd+oK?av>~B
z2w$!KG4#0e*C%z$veJSKmogmMqPA_@|7RP~e=cQw3fqp!x#`%((5XkeX#B_j_>cei
zkE;3gRa9?rOC^VMr&MzDwhdt0d88t;^=<fX_%fPCSE2#>B>G+SFSfDFx(f`a`n(Vo
zFzIHCKAz55>U6Wr;c7JqX-LBRq@X;~nClU7oir1`7n}%#P~F^zgVamIv*)eH%|z*z
zqfz}JiaszQ9y3WdYb*u8g1%!13yC&1NgN^s+*RU3Aw|uT+$a1ULscb(F{4)Ef`8tG
zLuJ)(**+?lq@v@18w?D_U4QQqxZpTNNn(Vp$$m0*RB)W4;N@t%B}dxmzOYHD_U;ON
z>O=k%XEI@OB!7pf{GYIof7dh>RRz#@5D6=+1h!T3xaQRGwKWXFR8<w|PJ$(cl>jWe
z=yT152Yzn6Q$Ei?Vtm0}#n#6^OrU=i$6!p^jH}<?qt%>kDiS6;DX6N-s<X91!PH!j
zJIQt@+2HT5@XDM)?V<fQ@m?6BCA16OgC0RBYYLlj#vBKAC)xLSd}EkYEZumOq|q##
zZl$d_40PFf16oEDRF6kRhlVhswxwSsRKj1`vz4~mnvNyo(TO=8GBb_2c<wUK98hZd
zM)BbN!m@O@58x}??>2rjz{|q2^lOXV1ZDzC`dbf76VB*ZGFltLvh?c;+Z2#a>hn3M
zf!5F+=soB+(Qlz2MJ6sM_yqN&*&*Fpyc#+dxUD^b2V-MscJ^v38*i|bY%h)1@P3c8
z)9%FRA{}CvaCPD3O2iMd<TfHfsTSawLLmUsi9%r(I54F;s-7H)LT8h}`=8=Mqv?sw
zV5|$5KDDrR2*P$0YhvrQs|rfjC>1!cB<v>1Vo0=$7n8)rk`>_KY*Pcjp6w^~cDPw^
zH%}SDTvAWmZ~NdLKKW$DcXJNz!wB7L>#0~UV_+GB%gT62g{dXffdirO#=dsB)Fa^1
z2;)5F&j=wI15jknmg~e=0D7zJ6-3wUm6)?`B2LRA{2EY*YVv8qYx55<!_rF{t)ZEP
zd@sMifoLbKq?>{r=nZ-U-GJ(NxDysqmi4y7L=f0FwT2hoWCy_nY&WpIrRh0Nm=!Z_
zXcRzSDmQmU(XM8BNe7^s!C7oE!q26>aWDFhw5mPXtevAadH{V5)6enU+zP`a^a2El
z<1|xgC)iz$Bxz)M=Q;pBxj#u^b}DHYxEZ>(lwUEET$^oKw_V#2mwY7XVVGN|Y%Zm6
zhzU^~=OR;gccBLw`3lc5#{58$HCYJ)V6JN#R<Y7-R!XK}I^OsNn{l@-Y#o4_5g1Q1
zGV@XI^nvyj-19}?%*HL7aVVa53;X&eC!)L=A@p1+vNQnQ7HyX^8%AK7gs=HS^{Z#k
zo(&W~jssAFv$kP5R+!I+mSY+AdCu&~waKE}=9Dl-cX_#dZT^HO3~<kvl3RMq4w|2L
zxpzxaI`4uZyc6@aymhe8Lm4BKx8368+N8~V@>AJ5TR_*NU&Ubyb#yRo1~1J&$QSN5
z;&R!sZKqt0m;MNie|Cj#!?r7xdc9Jy?Z=0)cQxSubXX7Ghvv{Kx)~vNdIYNkc%66=
z0GmTaRnp{>_0u3!>|<;&0SjsnPX^?Fgm%N$=RWs2Zdn|c0WYRUXC?q@wWYWg<vGts
zwRovk16Y7T|I=-^h+Lh}B#HEiw)ms+UC%%NJY#rf`DnU`U0qkZ%|uspJ<oYw*A+c!
zcGW$zf7oBM>tCw0t7+=0q7v;dC;m7|pU>_!orCb`K&ottt^THv51<aDM%Vr68y-t2
z%6a53mdnv~zp~iRtycCZ3p#RFD_>;6(+J!%AC@b>+IXP!7q6WLIU?8V++D5Qpn7?O
z=lx;1Jr_Kn%nx8?)pEpUR?Au8Y84fabg9*9c~QJ$(`aRIDXaF#3Xrl|1?md$P!6OZ
zy!8zj!;8p3MKpynbQ_B1h7@Y3tEU)xuZU{VTcfD0#T|#(y3_6GcHN?%9HZd@E^v-C
z{e2pz9r)UwJ$v}h9cOpWOtU?E_UvKPGds`j*vZH8D~0ci_dcRuTYvNHj-7lLy7$gU
z;yZVoUCrVzclDO`R&g1783ZG=fR@o|^lF3}-QKZUa7+bZyUmOj;qFvBDMN?9wpvP>
zaY~Xf$<lV}OJ)}bgH5_`?~VJq-6#Uc=bG#0V}v#h>Zfru|G#>MmtA&QG#?vS9ALA*
zdpCwC>UQ@nH*<M-kf>WmzmcSW4fy)u!-tn2gB2qRq@CUq*ZQ-qqHEAA5o!eYk|Du%
zar4$BBSkP;5QLqrUI;BT(Y5U24-!NTCM2#V*o#j9zEr4Ia)p9lD&<{``+1*pZdv?f
zjD-lLuqrAn))F`YvdJ8z78X+~#^?PY#B6|Gxt&FwI^$oNx3Lr!<GAR{vcWw^V=lKW
z&iUoJ*@EX%#<bpa4J7UYsyhBIoZPF==(+tihw8=gH}&Ms2?HN)9+RCrM0qra4n%%N
z5dEf2+Bi-WZyzyL!{q#>Nn@7~@JLr>lMqrgEt<nC{9mzHRA`sVMMWvf3#DI9io@_Z
zO>6ycVEmmkD~wW+Oz_<8OYPG>%lMnJ{Gr>FB3yAH3Y#C`$banv_&VB-PM1^AV0^F_
zL;f5clm?#}_K-9^Xc&LU<sywm#{-3)IQaDS56sPC`~sVfu2$U;R46IR>G8{VJ-QS9
z>M`u<%Hq7S1BtTci&Ei=9$kkrc6NDC(&oeYz|WOBQe3T6$_kU#`2<@59vb8&{TNr1
z<~ejXniB$OsTm*8K8fS{e{GCTCNS%6VAhbT{mGtC+}=jAMb06?aUl}7!2-U;nVpLw
zD<cjnZq(L`-t3y)x}~$o?e~}J^$Cq~eV*xz>ob<sgW07w)_dlS7-McK6YGEva4Y!(
za+3mi5R9~&FBOX)#dCA}R~A|YmCiE?bhgpR?Vp>&>tnYVZ<Uzf#!Kh~_V1>CiaTJ5
z2YtzNo~EBWE)H#iR+`4C9WasZ8A!cV>y(P#Q%^nR`<oWg0{yxcrD6{df!Hz}k5B;i
z`0)Mr-)~tX;d^7-x&G4`xB(%__gU*(C_&fxhV_neFbKVE*|vxCUxsdyNN&$~E+jho
z_S=HC>t3`~EwCi)Z|fU#wS)n{3N+Qe*l>gPu)b<KcEhzRe1-FY2z)pU*!R{qA^<oH
zejo<pOZaXdGmVG!&5(D=1%0j7i}*ZzYkmoy?sMpA^j(C|k4Cy-bZRcPep`|@v(URb
z-gPF2bMiOI4!;P2A396BAN;p&3Q-|%pGM2_5VS7vS=Wsm)B9Rgmh!}}sOzq|?j`Q<
zpQoOVL1)SC!_jh(wdU@dX-wo4=C7?OBz#5A`BldGN@M2N{IVhctpB<zH@8mQ<0v|Z
z1_&u$Sdg=ha%T&Bbe@c>#Mg*6?ftD7({d696zae1#r~K43w6(f3f@4oJXPcf5<-E(
z{`csIdAZccJB`812$oKoE!vB&Me7JPjH^i-S;cnfReaQ&2uMcQb-}tz%G4*>PP<uK
zYtUpim^s2C<2gXj^WyP;7Q>c*ZhvGLQ*oT11nNKOw@>qjjO~PW)01Fd7qK?iXy$;x
z?71&s+pg@3cAJ{Hn=`t;1G-L6c4|8O1SOq*@oz!*{DM>2;|>GOLtm@3gNc<cYXzmz
z&ZJh?jigUkI4wo_>O=k76e|z0H*>@o_!m9vCW+TuD5Ktw6Y1F725|4z29o5j@nv+Q
zpCAV-Va875)Ry2iq0gz?s`4n$C}+ZMX?^@4o+ma1r83#OwMVJ=-cwIK1>=_lrBR>}
zYt#2Tr-qb+Hup`M0~5w0=nVy>llR|$|M<g{iV=9e7+KbH-Isx!N{K`M<CKru!#ASW
zp)aHP=N7A88^>v(gZ4{hvE6H#wXR*#Br0j8t@QRnN2s^Q`I+kb>7uWnpj@rmq*LL3
zaOJu7%bj{{CLi)-{d1G58qvg=?K?0u8vDEeFaU-+`bD}{(Y7I<Z}>LnTvZjBSK^)1
z|7QTkrK%|K4cqoL4S@Gux8D}^7GyQg1x@wm>zLu?<<kc{9Y?K|FUxrzWEuvQMm`Uk
zs!R^eZUQD<-FgrJJU6Z1&}CWPu=G{VSI1uv0nwjLF~cMQYRxB43inCUc0qF&s180I
zpNs}dxu`SLj*P98RU$1g;;~ef0i_Q+*gPK7d_4(7y`~3r($zWqO0S~p(PQUf+Sb+x
zqevWbG@41c;J!#Rq>v+Tv8g2jOxv-|18gRz|9!ANJw2@eSXMsh<}C|AnXWiarR7v!
z%C)vX7fcg^_pCJ8Dxf@?mX%lBhitjeht(4*_(#^jdXCqh%m?HfP@-G3zID0aKuHpi
z?g}dkO20EoKfR=-;v`qx*$23)9h-ou=(;ki1cCTQHda(H`(T>TH$mC{+E8T{zM$*3
za^v_FZs6iNiWzu`R%<K7GNJ!?5ttTH+BGfOU>(I|c!Yb!af9EQ=k-A6bSg!M5ZgQW
zP8Zw74ccjjflUL~tHqkbdKq%ieeebPTq3EvQ0v1)qdRbG$c71l`79|u7Y}hg_p_OW
zadL|?!RCC=I%+!7rP#ZNj-&JFKJ-rX0rUyoqx6;=2Ms5|up;-$-5@^);>7m7o02Y*
zX*VN?2F?uxz&6>HhhR@&L(KfNNNhv9zO0}%JIj*h(hMdnTWTh3RRkBt0v>ZA5UX#o
zimw4;gJ)ibWf@!FgvhFsm)F*w6Onvw5Wx6lm}yzhT)cQOz&LoaDYSIH&WO@#b+Q&t
zs8-XljzZr^zHf*_hd5UY@aB5#K6V4p3KzY)W*U^%qqQDJ#8c5KY>n`(+|XN9)J8o`
zdmkvLAeCrj{SXl(A;1riv^SAuQg<D7FXtLTc;&LVFw8T9Q(yfpbX=5Se_50W)!gv~
zSEHnKLU<Q!CcF!ljss6H`+{pS`vN?-v9Xb{&3JN3etn|ongT!}vxNW<!p^|9*{bj?
zeN}iWvt4slw>+_m5Z?N~tr3=CgeF-x=i2yxJ@__;RDe*P5oNZN=i7sJD@}nN(#3SK
z(*bL3O+e;ODl@JwZh8c!b*b0uP4-&7$sRz<QM7Q}3pK@Q0l;YF$+v%Kt(&kp`p?8~
zK_tBIZ4(jvmfYPMp$6JJV>x0-ttt$nxdTG`pd&TeOlrzlr+_8`3WaYN-5wp+iE-A^
zb2f&+ul>h}8}pgt5e)JcG0&pF3#$D-`OhJn*scXW^fx1dRurhO!$V77A`O`I`~Bmm
zxwuIN@X>is>auUnib%dQu5!A&mST&G+AeTBdfaoDl4ks(tmrkzgM~{RGMFy)fDrK|
zz9xZZDHU)Zy|(&zu=Hn!!88ye#sh=t(wDm-M`!nitRuOdxI2Oos-PLP1N9NI7z3uA
z7+{hzJqeu!3Du+X$e_c2Y=^i4g~F7vv~)Ny#Me-%CP`5jLeGJ9^nQtF(0kMNtJgQ2
z{7j`n4$aMC-0hy6+rB+pu5k|clr3vs|GTAZ{rA>SIO`(bt>kX>5W6tp^`@J~`?lF+
zt0v8BN`FbyE#F=e^dg}2IP*8rY$kCy%K{WW$V{AOX&9^PVK2<mI9_fI;g9}rOzT*u
zD}>pbRt=#|1*&1tm$~JdCuSzO?TV$N72WbgYb`N6hvs=b2mxxf<>dob+qUh6${CDt
zt7d;CXIWrh)`a((0Hz6G!Vm`m#&h$l$_)kvmMt0>^zkLv<kro0<_c~foOVrSAKdRy
z!<&U5tWO?TUar;Pxyi}N$tK2lD#2h`xtw)FYZTk_xtEksmmR+6&$y^ool#;0B?NvX
zDVKDU1cs12s(Lr}+Rni!I(Im;eLMaQ+ZLkg#QU!5l2svAUq0A*t-j{iqX~rd`9ksb
z?PtykVcS(41h{J3Laf^M>KHhV^YfWT%jn>-#)=)~W1ERevNiV&Gah?={GHmWd~G4y
zDIC&zy<TkNNq;z08rH+Lm409M*!E+I8I=$&OOrk5TucNqdnV33?2sICq)o`^a=$bF
z7@biRYv8AM_K<KC>bnT_VE}M^I7t)ug{7)231)WA=F6dGb9*jybu4{Tw>O&rxqQNf
z0357eP#wPoe;_j^mya=`DwBIDme^$q=v3RjU8PgtxwReT$wO%JBxsB@J<iVA8NCow
z?>D!g&&=*@HS&qBsMdOb2g1`2!Ckt3B$PBt56lr=$9#<`GU;|z+&(a$G>gs=q*FaQ
zL?)AvHsq?r3su%&mk?*VIKqEZcS6>ZY*h_>h8ZyNgDxb*z!{aOM5*1iryV)o?KC4=
z7Nwllhby2~nZeKAT6pZS$4JEA(4lyH6$~Tri$y;$45+S59HJC|-H#RPaWYwt3ov=>
zt+x^|&0M5H<?BE;3&GmIEf;8X;1zB7A<lw#yfBroR>Nti+M`;{Pv?(&3pOJXFSC##
z#Np4Q1-WheH)fZI>Us9DKW(8J+8(=;SA$13-7P@~I$15XWi?kCritrOa(xs<$>Owd
zOE%Fk`eD#$s0_wy5R=#FcmV1-e5}z3Ta;pc)A>k|Mh~FNV%5*zYgHWP(AQgtQDQ?Y
z`zU-$@#e2=Z6wt^ba!?^4c?Mof4D+ZRI%W%b$M;>2@Rnf^be?|lfs~nJyJf};aZp1
z_Q8eff$W*D>C1Qk-$q5W4c&q6MUSHQBLu$qR5(rEaF_-tJ19CTvD|cgflm{u9k)&R
zbbE@h;LsMtX$4``bjvFga|kGIvYU0YD10AfJM9n+f(Ub0_!uF3eLBE6_;ffv8HUT=
zHv7_l|M!1aL72h4s%QDWWulj|R%qEH@i*;lwwyH3`#rZube&LYIAeS<49PV(Tf7ah
zjTwO{pct5@`^QDcfpvev`w;|SVz<0$FDm<h`6%+>;@#*hx(U4k-5*(Mf*_#3P)xn%
zYce}DtX~PUc_IKDm<2R)<nQ|fMVpuV)e(g?(fm)lIyTW^EXI^X`r$|J_2+V~UvmxP
zAqCB@`~KN0UFBau(D4)){&)c@s$00?jaI9LUY=iH@3%<@;fr4MXEo3+yDVObnZ9Km
zyFv)BkALzR4t+(l{+C|IFxTtBz~jRTYjAYq2$?dt781#IXJ|h<g>FQzM6X59ppT+o
zM_)iA^m}NQ)@&$61^NL3YZ#Uo8?YP$E4PwtN2)sm(vD5c;d0#wJ*1=KL!?|reNp(1
z%x7Z{v|5NWLvRqSudfFf2Y=)=EE@arI}~KA?82<Y^g#jP+rfeb*j9P<7j6q>h{9-p
z)++uKKVI7}X6<vIsla39APO4?Ua(H-K9&N`auzz1qrm4D<!tLM2r1njL%lS&Y>@8I
zsspZU&?TGTT=K}VO}BIeJ!<bT_Mi@|f1trA!XM&zAlwb2uBSUwWCPQq-)V5kG^Uk(
zZ>UVf#R5PO)ZaJBwXn%7$81q8Si}Jk0s&qOG}<y<lQnZ1pL$<C2mlJjc%t%#z3Ma<
zj@43gl*V@b>!r3VXT@^TjH*{XR0f<cOg9PzHJlRGblWw#)icawv6&;=2mskea?QoW
zG<uVaaI@2vn9Qk#LSwp+$54LgDwSrbV;SZ$q080Xh01MFYV58qQ?hIt2#@2+zlGk2
zehvsII8a0=Y%()ftn(R=1MWHTOw>KV(m;1?>$Hy@j%`q>fqfq4oDI6Rkl?{|Uwa<i
zcoy1zf;1UushTt#MmSx@C?YZ}LA^P{L&)%@KBPOEkn4$`)l{7N{(^F%Y5m8`e%??k
z%JzGUQ55C?T(3GzHA$3JLy$<qeD1n`CFVGmBF!+=vXHcNSodT}6beEVVrJ=+5%{VS
zHNfHsN16dwnP8tlM$QH68Cfl;KeBC8vs(Z8-LckcwOT?mcFq!cM|evfAP5@OFi^s2
zqED{~O)+|^+DO72nbEWvk_($nRqYun5i8R=na(K%S@9!&H5@4-&P9?`5VAV6n2Vn4
zVl*X7JwyHeaqyFD#~Iv4uOD$dICAheUS32IZa^P}hzX{R672Nqz1WSg*_@keHp4q9
z)!gweSEKY!Fp?k$g5+zqlvc67CI_hTTcGP7`3m$XdNcX}HgV#op=>UlkLW!twwp3C
zDj}%s3}&cc^5YcHP$B1+hU>!b(v7((p((i_p&4CjP$`cjpz!37te}1`ln%Q#z(YkS
zO_IdZGYN+9`Vq=uZn)^<iob~Qr5!tV>}=(aly%qDU(_k{;59Cz`il+_l5_V<`#Ahi
zy$%rQHh5B%ZBVjuG1lY$hj<|D*Zu5GH{H~mdT7=Hu*ToA9UH)QY*^vHte?W`)w6i8
zPIT%9T2BrSDj<a$2^V%NZq2cvUo_U>!1x8sBKLbIUQakDt=BToV?U-~S~-XU1sOgb
zf^8RU^T&*x_?_cg@Nr?~EOTDQT6^J}=<EnT2AhGgR1edmh{^a?Z#Ecam89U|n#7Vi
z+DRLj&eSyiN+bznBx|Q(8e@x=N;T6Us+7NzXhEQf*W)0-9LL4=?ko7@zFoz=g`=G*
zxV#{7W|%e0(x%5>V~nk{H^SqQWofd^m~l<`{goufu&RxJA<dCY_flT8E<5Tz^Wihs
z8AO$2y;!RK%J`8eiXx=rB(^_?IC4;G)%QBxg6=^Nq1U00q1i1UaJ49do%RNEWLTdt
zk7?d<P|`Z#20BAO{$8Iw^gDjHvy?PYl_!F*llDtZb~{Un>R`rCT^fX|P%M>-@Jwqu
z9+i0<MZrt!2};Yh15vdVHHt4+0nZm3QL7q(W0xC)1#J)8AF!;gAkJt$bI#bXtlpbs
zr!>zzb1y`eWm(a)-9BXJ^Y#^s(|eZ^7QC&ZC=}}x(lf;UQ{VTKrM=Vc-2P10iv>}<
z+=H;XpJ%z^iYx4V-u`3R{iyNFdHV`GKmGcg>3XpsW-?1W^C?MGZDFwOw}}VvF~m_G
zEubCfQ#smUe&lnP$`CP|PkicwMGk2IE|{bje$1uFtluBphTrzf5nQ5LbVu<&8h{6g
z9Np)M!lhfRaLI=L&Y*ZiV;RYdcps+uMLdDtjXs6Y<1MMe01EFx$LHHX@(n2sCwY(n
zhLWYU1(P=?OIz*5Ob8Bu&YNVyztDAxpqX$n#q05j%{a~`+X%@B6)%}SK%e-Pv8K;#
z)TB9j`%~as6=jrzZhJL=VJNbKC0ft`z}A98fvlwZ2Y@PtlD6nkVih^(M3rU3H$|~n
zkE>eKd))Oq{P(9{mCxsQ^tNBjP9lp*SqdVK7@RK{2uj(;1RB;~^CpFzg97fVsw68a
zXaz1quU5G1jG>ZwyIOvZ)9#+o(!|W>y$*V#>bOF->pP;Hn|^uMu3g*F7Uu3EbYf1H
zV=H++ZEJ)V;Tvc-I)=`or=<|IVc0k!3Bz=*!k`WnvSHMA0+X84cA^q;LwB!bki5T#
z!2ak?<@;G`usO{nhM#Fjt-ESeX1Xv`B2jl<5gCAfY`aTiKpdcFE_87mM0!q}F*Sbk
zsG-{^GMBA1b2Bw~L)CQv9Jd;$^IToT`>W&cZzlW$XJN6rKaL7<AZ{eKw@jDzBz<^k
z2cv)Ona3SX#9DT}<E;^X9sCw5qDj<2d(r0*0;QoKCLT%}sWJ(Ml|~9P?kGO^D~d|}
zNOW;L4F`od!YG8!63sW5dcdML!~457UuFBBDy;Pt@5f|GCEVcSm&Y#~oKIZWW1wbQ
z774Aw#T5gvTxLD&e_Je;kG&Mj+RMRLhr=P^++eb#zW2RBSY3dVp{9M{lMDS~voLwM
z58pS_w5aSUEpQ76CuY<~C+DllxlP3T#-TWl+t!`=HoKUzPU+}ZT_$wNp<%kQOmin5
zhcxM^6|3XG{V~U~dS}j@$*{EtfB^aE__^}@C=MknMiAsH1wSxWTsb@~1}9}WL(Gq6
z?a;e9^mr=eyo##!(!#VX?!s^I4bBb!20menPi5gH82rb7{6`NXxr_dzI{hVq5K7N_
z5?*TPO){dnS?wQs{tKXntM}I_4y~=Fq@-~>LekbRx4w$A;PE&*jc$rMfr2QZ9&)vr
z4Hg}(^)b&6O}ow{Nm@_1l{`tdl(jGjJ++W@{R*4orYro>?%lhE>sp2^5!exQ!dLJX
zM^&lrIL;8TrVy;?c%N5|CXU6Gs^6Bd4Env`2Z0ZdilgDx-jAI)al*owbJH?)I)Glu
z7C>8f9M@4*x(14>>8^v}=8A1wR_TzU=(=sR^9JWe{(EQbSHI4ei#~E6hEi*;e<fQ{
z;ZoPV7zUL+k(FNH=k3qZINCRW&FYJCkiEL%u;Uv4{W|YkrBNM}#r*Q5q7rtjT;_5E
zM>#$IMQMxvqOczae1DO5Um`2)gf`VYK&d8$N02X~3Jl{OIq4yE)7XGWlW`in%qbb_
zOOV{RV{)6?SW%c50PEzYw1tHQ4$B_#NO^7|-8h9CXMjNMA(|_ymbvYB_9kE0wrv~t
zWk=~}>%fiR26`%Mtd%~Jl5PL_RsujK;r06Ja$;IJfy_ihC<Y9^Ry=?^k%y+dZEeBC
zl4k;18jHW)0yxrw;ErZV%}`s0(>Kee?g|#w5Q=k%O^@$Qxr~KKTv}}JjiT0@kv~LE
zxSy`6Y$HDz;a+(eJAnEOVyMx3udeGOr+-7d<gh9)dPY$k#B)cKk_zhuh=b#Z%OP~A
ztK{rlj!U=*b$vodCZw7aC`%*bO9|Haixz9!|IX}&U_o%C7)x?(>tUcHU!tojl_}}>
z7sYMpzF!{9#-g%P$Nzc2kkeUOmA|?UvM7VS{7_ncC|>b%)^g~Z((iZzrk6m^zW!<?
zU3%$r3^N7O>q9a4mbKK`5X^QKA*?&3{AB>Ee#aET(zP$ZIM!3@^N-EDzje&r!5voJ
zGdn@JV=_Q_%jkrEgb!SK=iZ>#cl7#<cgPPJ1H!#yl09?4xY7r|xNjkju0|VuMJ-9+
zJM&0^{<vuy0bgOioCeaux>YCp<VD=EJjt6K$3G%DRIE4aqv^dmRz6?hm3-dnP>uS9
zsW-8#Xy5Lj|3sDq+HG@DhDPb%%o0_*XL(HcPjxnL3Ss8Y{%TmhM^g#h)lG*CKH1}b
zyf*5g`z;0|#_KphQ^q`3QJ<AotVsQ_*g-#dKM<P=><40wZIyk_K<6%SecQ5b(`>wm
z+jj44h5`LI>us1d;DUSu(bPm-e5VB1Wz0}+z;aGc@mX`i<JD*~;!9|~zh#!-_Ay%7
zdJ+B-M(7B-DzX!VZ4ScopgDrLBL?Dl&w%P*8_CKHjfP0)*Biu>9Og35t!=w1CuT#>
zX>a!79FZ^wuGOJZR9_1eit{DQqEuXLP)UV%4}B_*16TuRCFY&WwBODQt`Q0IUkLL^
zDSW+-q?3~_%Y_wP#Mvx%s}{eAAAu1%j85BJ@yiobX$z!2mn<zLl(cJdGftDV|3291
z2tPUSVNo94!E={%$epy(fCWGuu9l7jA#VeREUPph8R?ajp~elyu|&?nT9d~9{j9Sh
zcW7J9bBWR#h4t;VcHRkH0&UN4f%<(cxYf{s2}vSgYq?$tbIhGg7SnTYQ~ZKsiD=wl
zZ*NB*^z17Rtc7trr6cF9eZO0M=W<Ec)wzR%Y;PpK41CE>WqCrcCK<tA-?Vyxw=#du
zhvqbJU2@8w*AJ_viLlxRb}F&hCH7}q7+#s5gGLH|U~E?hi<5U*x%^egX)CE)R@MnO
z>Jf*Vo0~(#XBm7VF9d9%CeUuX4B4xFf$E+w-i52^O7sYNa|}RE%XFF<$*v^9vKN@s
z@c|b{WC}Jk>Ei`lLtl1`NP!IUM5+>UC8zlwf7N2_4drrg6c01bSPa0&>fAzWid&Y}
zS{zzd%G_efbzuIYc@fNIhsCdt31Dcd(Go;8S`&>Y|ASU*<|5M+M`2G7Y>x}uw{M?j
zSTg$rx-MJJTyWgdEwdeUm>jtcUsMf2a;jlyIiz3_J35A!kbx>_kyAAp01<wUh=@Bv
zB4Z>@2{Xh(u=WY#ied(QgE<#;Ke>1Ob06dIy(VDu8RCR@0)GHMXL0R%pLx~`Zu`t9
zKl#b$yVSIVDa+b%yZ8-5DOVc7XJkLbU{*~qZmerLefgTEY3rQXHsfbMO8ifL@{{vd
zo3c!)X^F?m_V*%}<_ACuDsPS8ui+&mPz9YtcU@l*tE-6RkbhDY&|9*Uxe!2SJinDD
zvil~3rshbsofr^u<uC;FYjB)3>!xa)^z5H^i2Y|w4CZ@7?0nz@AHXT{3jIyD+x>LD
zw3rL7=c*=z;Oa7FvP32%e2mJnX__WI-T9NGI+7j}!RHm#=XeX==gY=l=iqRgrv3KA
zG=A6BS6_V>#{bQ<U!?|DHBA$JAlzU*S$RxYCZ)1$E>SA(O3A}HYy3+rD-TRI`l~id
zXAD$Eb9hblYyFzKrKb`T%dL^$#~@(5PLZgz6QmxtGgG#`PQt9$BFmSaxB&@X8403g
zDH(fC1zDC^gLg-vtl)M7s((lTSg$kzTd%zTHLrP%WT|Z^{t%^NAmI%UUW7jRn4YRU
z`B51V6R|!@%w!FBoMP_@P3v)Cz!_X67Y&Bsz16)})8MmMSOBK5j6IYpJTxf%Ou3x%
zN~b)dsCpg)w?8HAQHT(UQk*tBj0t08Syzcks8O3qRDIdm{nJ%T6MS(L!%K_6w-LYc
zkKTx0$$Mc(N3Gq1>ro??y3dm?@qKD>Q|J@uSw%?n1U0$TTDk+I&{|jUYSnK2n{BTN
z@!%g7F4?vaYxQw!Yk_UUrM0!=j?=a$SXMTXzGVPs-&wus$l7@83?bImj=@@mHuUY=
zzlZXuiSF!GYFkP$JjYT<W(`QP6{4~-wukQs$hJCyh>#VOMP+hW$Bpwdbh9?lBpNjY
z7jP@p#^P<OGEIm~aKk@snE<-Im<6*Tws)AEzF{)&d=?;Y{D+^e_uC@dSzj?)v<Bv@
zsrw4$D%QG^$#&5T|GaY!2D9&-e=>6Ed9@hUfld(o7PD`*Z|ko>RD(~*(vZr-O4{V7
zF=g%@lN|~-LtuqRVXT}5%d$3kl`|t3b|-?lxel&1DR)-}$iaI}un&&-s-_Ke2R-!6
z++BB#f4Kdao}&+bch$Nz{@Ix`XIg>Wr^R_>b=rJdi4ih9@Kf@*r_o*L*ZV<w35sgR
z<#4>EnbYcK+20Iy6)h)QA>n3MuH%k*2A{N>?Y<X=2-DYeg<Q^0@?fm(VqNbikZ3`x
zh~$`SmPf72Rx0;-6#d0{7prWbr}Txp4WL&m9+%4*3p<(?`QvK^F1_hb=3lvoJfW)o
zT6<c4{&+p}E!QA<b|6f1(SN~Trk$lE$-9sI4g>4`kUyPMANeAxbs%TTwV(gztB;}I
zL@%Q6=1(D6py5e$+SL<z@`IP7_QTV_ijBAO+};mZuIJw8#+!(~AYIYA_1p$i28YAQ
zkM806){8i<w`FgYCb+}fV3`O}i|Kez9JE0<!M~qcTmKo5y8oBQT?UQ1+~B0AN)pdF
zH%Je|tjaU)9p9z$tg1>9e<kOfL`0U@D`okZVLx4uBB_nfOFk;AP;@Pq#f@aKoe8bd
z<%(9DqNFpwjtS^K(3En{mo&-#_0X6X>r&aaeLp@9gCc!FB8YhP@*}BQEnjh)LiajS
zdT2(`f@l8bYWaFRHQ#LHXRjz%tI~L5?J5QEUQtdDeRcGytMJ$Qx~9`t_D{NdZ)pwG
z+HWU*=!#Z*(ygZl%E*WAo>W<>+4NONp{y3uS1++LqeykqEZL3Lkd8u`73Ocj*PZCt
z!OK}MYAo5jz3z%VT{{Xt`*ChjOLFhW*00-=_9gziGMdnJV$*&DSFG4>-mbfcW)v;w
z{P=tDP%1flsdC#D?=+xGx-2gk#xO(<iqJgTj*g=1&^meyJ&nGgZ4C85_&MpQ#<6;R
zm+bw3TuIDwvF<Spp~>!Ya+oH5R)P6dX5(G|<bS0#lV+k;e6|0>gmCx{Cj``dzJVsW
z)yU@+;P^JO6p*JZ6QT;8qBR_U=|DQZ#CowALRG4R`Ox`|^oN7TL>~yWZK_!)67QVb
zNwLtxlkm(Is-ffP0(wJq>0)A}w7ru{J>qAF*=Y3ZfYBIMN+HMFa}B<3X1{i(Z|+Ip
z+?g$I3Y=Kkm$;S$al8x+bEIEIvN*V-b)rIlEs~*!%!#(o%<z5b^v&KUYnHupe$JL9
zLFw3?MXI~Ilyv=2z2;j7y`Xe5HM?Wu)ZuG;XH&83rA$h>%rnLTFAeqGjH`bjN@j*G
z!0H!CspC$EptC8tZezD9Y31?9<=$_;AHLNuiSU?#2X{g#6rsGdy8Gku=90HU|9z|9
zD5g8|yEBSnCbxbc==5Eg*PqqD;AHR6-U+F^0(jK!80ncN(Oz^8-G`83NK90+bE?Gt
zT^*WPv!rx;YN}*yI^(mwDIj@X0k|=j%Sl9)b4vtcrAa|i$if0q6xviIOqOoEPDod@
zWOEl``F6Xh-+g#<tJPA~mIf+O8XY3bQa&fiGFfUUDgo`gmb<u)xbF})UnN}IfU^fv
zik2viCk3)#V&fv|@d%WVcFc}&Tbz=VB(qz>D3CAa<#3w2#ZL(~jhnghN=gOXd*sLw
z_39())ze3g98s=0qFy~z)t${F5M6nk^D3}O#HqC77d{aQN_*_{?DL}sn*Aa}1U_`6
z`w2TGdbHNE%Ocv1?m-VD1gLECcGnqXOZOB`wB1M23U5NNC<3F!fsqad3$Pa=pa+D!
zr-dEp1KL7Rl%&S>IG8%Cj(acKbaE0x_aaKgb8BmBYw|x5>oK`fGT$+mc&HcPFWFAq
z(t|Tkoprg_>%$h$^|8xq#2pSWpC^aW`L*xTwiq9l{m}5_J9&-<YgF~UsN)2_Wo0R&
zarrpgt~7O;i;>YkI(ihB&u&M3%OW$jJ=bXRyWd8skYqx`!q@b7a;6E=eZ;c8r}_&)
zd44-Q`=dGY9*{gQp4mrO)pVYbg|B@NOfyF*9P=0-yOv0CI8v0;PQQZHr90!VbycMK
zJN(b4P&`*Drb?E-qe<9x)Q3w4yi<zG-;cu^QgLOjNbZOP&FYWj4oKU~eo34%+S-9m
zqMP^vfCi>AmtxvZNnmyo#`iG?ft3y)=v*<B4Um_1-XwApqN7KP>b(0TFwI=#Tf0WM
z0;Te7l~|VF7{t9oJSix>c=5|%Ao(_6$eB0(9HlwF)b{QT8cu$$T9r?GjPEZUVa!?w
zD=t$61m7CKbtplbLK9(8XyGdLF-4g>TI)#HA_rU!krn<ppfhsQgWG$ZPBm};xEJBx
zFGq$f@QI!y9OaU%zw4<-&+-l1vf!V_|Cv~MWP>!>i4JmaKy}Oc_i@Lu-K+5a=US`P
z?6j8Jie;^=TT$$;0Zto+wA(?Li=yRR<l&$P?a|lc85oZHFdX->b>PlXE*AwD2Dvk9
zYiFz-k>y+UfC9yP65|03a!<>PU3rAr5wg%SXSnDy5|}#^x|r5!#6%7oL~(7w+%z?9
zV-hrQhF<v1kkMc;SWskH|1DjXm4(4zFi=TldMb*dWCV|e?>LU4sp$%qWVy1iupoqO
zFD%%$5ZgBf{X<kE07RpQB3X9oakhMNdHid3ayg6Z4qypiT6jI)SM{5A|82AfokBOG
z2hh{#V+<DuQ0&#y6xOFsg{OdTM+C&H9`T`xb%!WsRuB`QV#%u9-|Hd_6;Z6rZf`|!
zNTullIO#Y~+^4Tt0V`(uIiOCsxkiBLiyDo59?aS8E}t%72=Bfp{v9x?2Oc%v+&T>l
zOh36N+{D-~EqFoKuQ9I)Yx)a|jsq)G>`UJO<nxV^AA#q^uknNd*Sz4DdW#8zO*1lt
zMfT3paZ{ZzqwDiWVy`riZ-1$N06qUphu4|z$lP1G8?`&_yY6G9r3}GUq}IE?0DUe8
zA4nE9Xdj@=%-ogwK;&!Y-j+A7t9Qmp|2TJ|htXSnmI!0s;s6IrwNQu0$O+l5+p>Q&
zR`rD0bB%_RoWQ|KzuI-SR3*w$Wwk$=5UsDf@EtpG_7)S{_9ALxd(r52<(8{$Gq~a#
zeOoB&>af0brU~PX=kkNU;1k8cOz6<jrgl>=%x8-?XXQx0)pO^cTYt{ka@)h<9drnt
zMYo`b&{OC)u<JRA?Ibd2nuYBF-G-<aH-mexv{Ow1H}%e*s>Dy?LM9*L&1j<9jRc`L
zGnwIA@lxps%(se6GD%Xwy;Q1?Uv4Y6-1o$vr@WH~b6;L8<Z^|@M>z|C@Q4Wr!3E4F
zI<+!=A{Q|`Fb&tpI}HO#O+$EqpJ3pc@8v^bkSTLl@$~HMG=4PkfMr=a-}L3z+IM9N
zJO<=P>)bM$LEAW&mmx{Lom$4H-bm3!d07niXm{4d#jNId%xZqJ-|xF7rw9M2!{LxI
zU0-nTU@#aUoNBH;{UHOmx8LvUI%D7elSISeP}dpTKK`e{fVWS*8|8>a-p#UnPi(u`
z(3;j%mUi1K#Jt|zB?K#*m6*?~zXB*I(xjrbG-Xm!bdXB4%U3$`eE^U^Z@;z{Al!4$
z_fn9%s@j!--X~aDzweE&w$@eCKWBTu{$=LxtfX86fU9x)d@=K{V%jpvn8Z_?TXUEQ
zJKjmOqT|4<M@etFdDj+j3E#x936qlmiyktVgOsR_|5mWS1-1aQ*j}p50j3o$%azyC
zV!~7YlqR2u*SeDb@bOQV&ZvF;<^{~2lk<!oq4}&z2;pc?yK{=MQ+A%}pK^azA%wlO
zqw?-aT|epOvEm6Y0=g^nFA^ahmS}Df-MLW+J&JYg@g!{$MHuj-B3oNk(w0jkHx5pW
z`@oV~ImXJ@_+G*`a_?GCF|F1;9+u#e{#)Q}FTF?NLleNf<t`n|n=Q%g^?Fuj<2BlL
z@n3pDG#u**KLlrKqRS8hc`y8}Vl8g6LMT2nq7oMXo?67AFibn=>z%nTPiNQky_SLF
zmWpmxV#r`lN-T>{-O7U?>ej-#xURjRvbF4%lIwt*E#D`J=hQbP%%-=aJWsAhudptX
z@oX6`+UijD*oEM=Q1v)HRc>+^U~paAwuLY?5(XFtA<;}BY}<BSlRkCq9203BX{&vA
zhJ$gi#xPE)U@R>|0+#N21MdUi3#{O~!bc~dY&1E_anLjdFyivqxZMGfl~+}yB)(Z5
z>Nzt1E6bqyWCcOD_s9?%{W+aLrgxsHwA&``$I@0u(fhEMdZ~|ix&9Y><mC0FV&Bwc
zNuJybquWboB*3N#kU*J8Ccu5wo<R_UX)7heiY?v1s&*qw_=l(lA09|ey<1TfQFy9y
z#f?45`57<)Q>k+0J0fW0*oi`cTyMfg#bDD_6b4{ulxhY5J5XoyNCvXKhmNB@`qrwI
zZ<On2@x8wG;@oYw6hrnv?5osn5-!0v1YI|9y`uok5t}t$KOK&ZQ;u@xP=$W1AOc!)
zTl#hlbtSh=CqxP|UC#oE5Pe%t(Vsbk0?=Q4Ju|N?8W0fLGX$nvj9EH^5VP<FOuHTi
z@GOouV_Sk^9QV{0VDvFEe7K_S86xi03TE1*ol?=^HN()tLdS888t-%fb*Ri);vd1*
zj|a5pYS`#=T7TZMy$1tSMiVHI$)&UO1bKWcy~LwBQbw{k^#gL0$?4Wo)@@@FJ#yn#
zh-lLx@RD-(I`GJ8p5m<@;j<3MqrqS>fZ1m*o0sarZzB)wK>N_?Pgv;26c>@8L0~{Z
zMk2CA;f*_pu!U2k1OTsqB!>Xbz_!gdzVVHE4jyZry`k$mG9F$pQcmu1BCzc|?!ZG)
z$*~aytL`#Hp<z^R5YTyRz)U@}2fl@pv6T%N<j^sg@i<-<t;F^FRdUd7mVx2An+!)l
zb+K0R=y^${)RnwnFA1GOgxMy1h4Z0ijyHwQ!2XU3^b3MgrkfB6Z736-6i_c(xPaJ~
zcnBu{JNRg)6Q;gDP?D$DE^A5jnTP$mVXj%$nrT?hRPhdzj`bSXeU9PW$qUNLS8<TT
z>^XOBZEdHdQh+tbTIGDza!h05p#bCHp$Wqbg%}XSWCK!eZ)IqZ!T8$`sX*^T@%Sz4
z!MD*2I)yM$#<eyIEgB4&#5LgsQq=*`Fi<B@xLDAmM5y_!4)(&~XO3l<_3<lereQg2
znzrUxhFODm?u$Nul1;2rlXT-32UL~zQ*Vc5QVKcDo`bI#rejqqmSY;6)CGTtK61Km
ztH~Ip;yw!M6E-kr4!JfsPvWYit9y2s{E&6`Ulq-ymS*8OmOOB36L(iV&-ULJu3m4C
z-(Gv+1k2v@eG@#d%1`yvJK3(+q4LP-EkY84tr6ahJ<W7DcnwCRD+cL$tUcY4{LKBJ
zWJ+1x1p52ewsqcxweM5Qul&$+mI4wx)sV;FTPTnAAOwVnn{arivgzSTo210<_12Gk
z<Rij(&XdVIfB!mk-S@lha}4M3N$v9GhaP%Jq64ci^$AbcFoqvLmVDby+h`W8M5`rt
zjRRL|$ZHr|7(^6GwH+dv??ib7fz(uz%jLv3^yX4hJUP~&kigM+oYBjTH<r1!!Fq0E
zMyrTqK(*}wOCQ$6wg_V|-`$Q&WZgDt$2C8-1?C_+ccPtkgjnv6?6pzkvJfN^Ez0Ep
z;UK_%QsdDGYK4c9R0`>>m-bKIonu||=7Hjqt@SnBgKyNi_G{32bO(C4rwy;d)WKq9
zENbW*Ni#u|4bxn500iZ#gajP{M7fSf;*!h=J{Z|-w$m_ix-y+5-K6=|_$!S!n&fNH
z`@dbg4bBzcS2#Dgq9~T7C<=#fY&*x)KgD6+{08tR6<Oh&bJhP-U*((|vZ6j&{^&<P
zD&8nAZ#J99nvW3IXQs(~moNs+$Jqbn!Hcq@@WqvtMXo5a%D9ot%_Roos><2+xT2`X
zjXi7mBg#qN8vg*JLc^GwyGueZlXIB=lPLP^sZ*y;J?=B&#z_*pg!%b$Iqz-AldA0Q
zjNqH-5V{PVMSXOu?N4eHbbzfEbDY#bz*gFvI8N(n>yba)5h5@N>1vX7Gm{z>k|O!n
zp_izu#)M=fOOr56!Xzb$^^xmnCh@-uG32tW=q{x~i3sSMx*lL01>0BeK%@vtU0so7
zK0LgA`}UQ4?^P6K_ip9Yf-yE9-!(Tk$LjZ)69um)d`(r?2<O~cyNxlg&WKX6=5Wr*
z+IU}@rs=3uDs?)gQ~UPqYnIE+W^-Tp<%tJ1KOKm>{>T3l-|8M2k_M~=Vc-ebXS%rt
zcVO7{$tI3)TRo4ZQ$tW{#X$hLI*)nh23~*UB}&4nIAF<k`4EC2o(c&aG=tYB^M5u>
zrsfc8L?5V$-G9gykqV&`;xaSh`0spiJ%KYf-E`AlE-Tg+a)qI;DYj$mxrwpsgJQ||
zygoBc_D?t6bkiQgvDG!j5Dp+)k@?=6m}#=U=lP{#a6MxPAtcwgk>5hHu93?;sFWe(
zB$&DYGg!>H$MJ=sGuc?VTm)6N@@)v)(v8MwYr(&nXG!i%-n^kRIdaxCt&t62k1TTm
z@{As$0|<gn=sFUGTnH~vXEne9yId9rs}7~4ue+|^C-i?06p5XZfrs0`o*atBQ}4Fi
zrqAEICU2G#Pys~S%^51h@hrTzMfFcBp5BOEmx8kJ;~SUa?z~upl}U+4H%UUsp{10`
zcELR!#<Gp(P3&L-MOy{_S%!nb;COorDdrc!JnB`(|7P1lB)=A59Q;~haKs50XMW2)
z?iSo=&?-|v@)#dz`z7~y-_%MV()Zo=2w~d^4g#E*G1+VlrSFf!gWnh&7f(SV**{!i
z((2J@Vv*<DoZG&~;@KUOu-CBxj#~<9wZJffS}iEK4lkGBQB5DW?;o$Hs&2R53xY6A
zy>Rwm1?5$)r~RFDO;IZZo#}X1FiQ%P8KJMQ&4#ly<#a$UIu0bKKjV5~=$}4)dS=Gl
zeW0*n+P)1}@WlNPC~gl?vBVsMZ~%I`(V(A_+-F9*=@a<m55Ed+pij@@d(my=ebX0#
znHVgti*<I9Sej^dS{-}#m1ZK@u0!B|A%i@QJ-$^7R4CU^>@|oae}}<wiZXPZsU@ee
z3DL~V%*^TW%isBH1MBnGDEly1t`a|c4$)%7>i}^71g_wIzh9gDDE;q;-+|>vx`eRN
zF}efY!i~L?v(R!M@Isw;?5T#S4s4u?7@Aa|0v#tu5CT?$ur2-vu&&HnyyTA0hB|SI
zex959PR_L82~8#1YZg_NXxChg%MzGY?hLLl)TD=xK%<u4x-k`dtf}G|p<0{PcZ8_p
zpRG{b>JcuN3wK<u(4MBqxN&k))*?+&q)LToHM=gKd>h<QD3|fY&CSgkpP8PZNs93q
zUZVqN*$Dw4()5ORR@wuIvy0}<HDBRMwxbieaQbyEwyp_eR%+>e)NpnLy!S~nWEMM!
z6fh{yadLg4w%sJXk2r3D*g|+At2aIvlqSLf4l73g5;}>qT{z{$PPcFP&-J*X<hj~u
zK`DG+^S!cg`Zv|5b@JA>qqrA+ZJUBp#c6S)&-B^1+57IfobTu6=5xNEo2UI`3J2rX
z)&Q&;Z@YUql!rluQV~_cklApXFDTt5S|;6g7qZO-A>r3e$QOE^O{l*=Y!eUQdBjl-
zb@1xz3;ht^5G4peVk=90iKGshii%0sMN5?%p`5CDL}r<anO#e-d3V(rU4oGcKTP1A
zK1-=MO+2q8+=bQ2@1Hz*(q`R(`i>}jzILn5`UY<~3Z-Jxlj}CcY5)2g75*)WtK&a!
z$6RlFh5o9=@}Z!#5`>(#IdpG?o{5EFWfzvjLJ|+Q#LxHX2Z0vGIw#UG)Al*~47wTJ
zkKTmd#YGLv@DJ8V_&l=FBB>hI4uY-jfTud64wz{AfOCUbXbv1XLQ*FHl%YAra2CNk
z@Lyu4a9A8A#44U&U0sE6F=^Ysytr}1st|b1K5}6E534_wJ7usjf0$$dtEsAMstPbE
zCW^u^M6qZZhFL5IrU^K|tlD;-iG#47$6%VlLBaC2tuA+3?h*L%M`*0Bu70uwwk^b1
zd`_6R!%>C*E&<gwM!fQCWEjU90MJG^^&OAEG=n@YxUR0Q;e244pz5w$IIxH?QV-9g
zHtM0v(M{+9^akw9J*jKe2yFn?5CSS)0Rj`;j8U~LkU{RBagg5`>sQ?GNUN~4$Hu)#
z{Yuoa0$e>f`SI+5X#%`rd3iZo^x3?uO_{lSf;sh;#0-Q{u18;0g~Sc2!+DV9&{*KR
zQZ#SKiF3g;gJN#XE&$MGMxM~5cfYQ%mfK#;mGeDRLv6IL=aoExJ{Z#*r>rb(QMYFu
z!+Nq=+{}$ami&dpkgv`ZgV4W%7G_H&K?rmRiIr-e9g1r86SM4?j<3ZndwjpPE(l6-
z5a2;2I>}jcD${zKNWJ8NGiNwdO(+!+QVook_ndGBzVkUwJOdj?todS<8fem-4Aere
z@Dw*rd82VZ4g4?e%r~?TQ60^ogI1_{8i6M{s%pISjnN0N?E?qF!`Mk<J-<AHtUpr)
zv(!eif9ig(Cwd!H=;N0!6_kE*c3ZQS=udndMU;xD-|zRw;`AY~aItdU&ml!C=n=Gm
zqFhExw4j?UWlO10Mzz)H^*9WZ+8ELj6L-E&<c<vxPBER#Ewf@t+D$vTH<Z!A;bL)W
zwj`o4VOg@AB#V2!UhjcGBSN4has*!fF4%ECCP=1Z@EwY$y6lcjV)+{E2l>H=p&HmA
z%{12zLR*qv^+LKiQ>9;e<4j*T(+4Y@x$rBL32Lq<`8HxGI`p$qLCAIe&SAgpaJG~#
zb<n|>kg@e{3F~cHh3K=WK&%!`Zr`MddC27o5vOeCAW`XYTl<S@5Ua_d+qXfT0Knp;
zAw`^$4pWu+A2%HrD{}_D2yi{DH~lV@pPD6yeJad(c<P_t=voBy9JNu1QL&zE(#?R?
z<Jo4jkcZo{2Js{{I`*GABAO}|VWS418?Yc>Xr8vdbo<hxH4vzZKJ2pRo;=T&pje#J
z%5U4&e{XH#H{C0^)(kp^?p)JK@49!lpXrupsd`dQRZ<fz$adJ8+Hb<5bwRyD$C>tP
z%L|cZLfCee+4lAZfcH7i_x9q;3d4F{91}|1<-flQ-ZiktIWp60X#M*2+S(ey9{3%^
z(Wy~cQ37HX@EO^??WZPS*pb(|{fM{t*2H;^W~$XP-(*XHY;UZrG*+~zy*@@u0S_r>
z8(&;o3#n^mA$Wl=<(AkQ!K==7qZ7z;<7zmMj?xSHS?Xy$xz%`6w>AQowynBX0x5yh
zPG-vv`LwT_!8OM&<g2a?a03XYZZo3FZKs<qPx<3tp0lhR%mW<E6uB>x1;v68Dy)Y%
zal>qS2F1IQ>Cbf}U6kG-CYZJNyD1{2)sYBkQo1PN^S(p4f^sT;ddsqsL`Vl2n$ay)
z=Hy6E&&<B3oH@hk04<>F*z9DR0i>Kuw-ShWCqKfjwd1yZ_d?p@G)b3Y?=;$_tg_G^
zVir=BSI-9U+;$qqaXUt{wQVU2rE$q*Z*z0gEzOPgAllsA+%Fvy;h`z0yWZjX`T3bo
zEGOiCN$FQzuGOE{aK(|y4<9*lL^*JzC*KDH*E5TuHy+8?*t@~UOf+T{-;HjQ@Uqy&
z3U%SvwvL$MR#@{0<M6{37Aw#!%F5QPvWvUA$2r$ruhpFGq#cNQhTn_x*s-gsrm0oi
z!ShhBx7+nP<g4v^{dw+7qS36wYTxz%hSd$Km9nCgE7hQD8SVOAX}j2VAfR4vA0a(r
z6Zv+oekc0h^nuq}8XdF?IWW2o^|;o-X4<b!Lg+0$W3twQZou3ypLG0ABlXi7hQ$`I
z8DD`Y=}lBjs%sPZiHYf{1{}-_F&dAc_ijZ|kmTV~@UmKzihC7_F;lRGn(H{f%EPdz
zSeCWzFb_v<@NF2O6Fk0~x63@Ul2e>vwPTpop1c4($2@j%^c{?7PI^bA&YuxvW12j4
z8UpJBU*&Scax88Xr@@&n8r*VCZn$?Ft2rQ!_bL+BCJmYmGHOg}SW@=J1GC2D-G`G*
zRKWZIapL|KcwM`B%i@+}MF5fIn%uI?78Hw9!-djbRaJpzS?D6k-qHeJf3?=Sl^iyf
z6OZ0IEf5`%$w-E1hIQoDL39H3*Bj8|jBBYX#th(Q0b6`UvF#zRZJYTKj9L!*mj}C1
zR|2F)7uSs)%0a;QEmv?ZrwcOg(wQ?apE>h@gs-zI_9>N_rF&%C&iAO(0KsH{j(~XZ
zV6E$ldaIvqpCXh5{7|;POL}L{_+R+9v5hll&YW5Pr*F{vY309a9P_4D;yKmxt68{*
zTDf>-GH%ZvfUQ>2#A!>yq?QuyL(Gpl;UY=#PW65y2=&Lk&8MG!8k$eT2wd0t<IfiI
z+x_exTdoT~eER99pKkM)d=o7q1c5IKh*e`CD}DS<s-*sYf*f;!v>U>I80&#&vCS>C
zZ*K^oXBzMo-C7<0A~~M59tNPi^Z2Fj7JhIw)cP`&g9iS%wPYCSdL+Z@j9}x^e+brx
z*|dyH*Y8zyNxIM`M({q=u+=~Y579POEovf{69W`B|J;Hu{jf9QG@hT`V_*d7v(xgn
zUcj^04eSif!piQW$?t+Rft}QI_1=c7s^UT(o0c<p%(g5kcfnAIyU~MB+W_|K1Icm4
zt?zU>qx<($#@%<`DqKejUeC`a+VbVV(^x<)gtDw>#i*QaDIp-_Hci@aUVR(SA8jvU
zsu}QlL!)@Hee|4t8q+Pj*gi@C)I8c=#CoW~Hy8tXWZ|xPpEnhqG6%||atYJL#c&JZ
zz}?@CtuitJ%viU&MHDGiN}YmS3pLHNGWJ0nZ>*!ZX1PUAE_z9<1=A@fAMp5PB)OHY
z3!B#W$%e6VK9_39Mg<h3oxTbt+Boa`{=p=%W~=z53IC+xv}3CyK+9k>91e$@!+)rx
zxtZJs{lQ=`Al}ivf1L^tPz7RqL$0@cy?zrPWn?=N8y>!GAGK{Ks9ag9wmUxJ@7cAN
z7Svl*_0xjeIP<eJXL@JOoH=u8Wd?%BT<;%TdATp(myBLW`m^3=s;bsChKPN&)Qu2X
zm!{tXb9=oW47Lz!#N!+fmSnh${p9ge=>IhbMKFSX{1n@Kd-ynZ2)Iebv(g53-L%28
zlx`z$-~-57x$!sN<*`S7y28H0uio^6jfhJD1t@?%^kcQ=#zN22=_3WA43SEsDIEPI
z@rO<6P0Qd-mejSfw_ksLMmvFk#Digq8b!&%%c!j*p{sy<r1L6UC0cfDW7@3*)%BlA
zBVR-4&Z<RZ{Jc)Iy=@$lQu`0SgI{cHUx14+My@uj4#lV2XGJrU!Cg?Oj|Xiq{x7Pf
zha5B>u>I0*M9{4<>-&aF==~N^O!~Qo9A&qwOq(WNiZ8u%DaNK*!aX)yPix0%0N*C6
zt#)sJHh^NR-KvK9?XtNroV_$&!gwjZbT$lgJ`^{B%}DJYYNGw<MD$2N3hBj;FSl$`
zRCraoZ-&?;DGxiGc1Y*_CyEv!MTOx4UmKF%Mr%_Tnx3RjX?!8MnBb+%x40xjJnlN=
zf7_0cx;=(J_~}r;HnNiJazv8hYzPJ)zao$z$t5NAZfw+(u6CAfoU|#aKff}G6h!Sa
zM4~SrE$4?8F7omM#tSk}s35B{>p3}<*B#Gu8L#G?9+NdHP_pIEO||pNJS7*K^7j5f
zgFKTPs-~$8nOUY}=n8KcbO+;jo?|<xVeyJ?NT#XFk@onQyz-MP@>c6<K)K=m!v*)k
z=g<k{<7THDRtu{Zeyxg@?uUffBQbCzUtdlE#ucLzvMb&`0Z0X>H#d0R7;IUAg6&)X
zNoHkeQb3EE9lF3#8qwwlu>o?Z7D39F!P$RD<&?h76ze);kh6H?@6vZZ^w2{Z$rQ_G
zD`$8Iz(=$Xv4^hU#|bsZ)`jE6xhWvFnu_IXhX1AW&2;&WN`tTK_&#&^C+b;@=0J&?
zUGE)s$6?-WYYASiAvL(G8!Q8Lq9b)@epri-9NB6ms3ylAH!M2&g55j`gHbr4`mnzb
z27;>YlS_U-kyWF(01*N9p{8v&wY@ax+-VelO1oq9&Jrn1P8y=+(qm3oaVwRIp0A!q
zM{gsH)4P1b6jyK)>411eG!1|9J7AOF0c$4?Wz@Oz*<MK-MojbmMUr0HCfetvlJqa0
z#<DS(QzrZm@oHR{-~K6Fgr|Ij?;t`u9TRn8P{qlgLN8Yft5pX#o2KFKmzI`Jw^O`V
ztc4g)A{LZ*jo*j_<+*W|#ioI@+*27y%$@KCnzn+lTcg!7>u#iD&Q5d>LTO1G5SaIQ
zwc~aiCkv`PK1L)CajJc?`HIsVr`11Imt}E)7TT+!7~ZHBQWRBF!%o?BvY}ox{ImSD
z?2|+h!v>*ux&+1?{$sV!YMJ@O?%p)vqh4Ed#`2f$dQKE2z#QJneKWyDg`F;XAzCKM
z%h637V6wIE)4;uHwoP^;Te|JHUfkE{27GjEvvDzXh4>9BpS24qdE498lMH<XpJvbh
zFdvoF{^iy#@4s;2B4ax0@^*1J^=T3@Ie|_(pm@nt2D{fcMo=)D#p|w%anKb5v@ulh
z#Ibg}t-f}bwhpZ9vr;;~4LZn5xL_<Mbr%hQ%Ty*$Bf53J@V}ufTRm{3>Hsj!q%-w;
zn64fmjvEao7rYR9gF{1~suncr{>73g5iF~~OasBn$Pd;mj}7)(IIu-%YL0d_+Soyi
z4(NGLvir_KSFNllzFH?Bs~0-ZG0(4#d|ldL%F4=BFu=*um7>z#L{>P^ZpSaQ+HQ-9
zlFg>YE0+@N`_uV9hYa42ilgPi^A=KdSd-;FQWG+;Z2iHtb8+dl2X)&57&=ta?(}rz
z%$=G<o#}!yWc%kZhWDc>)YKr=nC;9$c!AJ464ZMz5pJbjADW7eg+B4yWKhAlV(Y4I
z0+_n0+wczROb+7eZz{H0Q>tMZ%=ejL8R{7FrGJ2j@C+)T80|x^X#Y?J0K_y%6Va)y
z)EqB5|JXn;>Mra#N?H?p@l=Fg{veEPUDaoExm<2mS9N<=#kOs`Vi~HDV4N7LVeL#_
z*Fh%yfeY_ytn64@+p#jM8kSLMw^vr$?TTR;>ZQec)NDreqN>~a+`)r$x~;31=0*4a
zk!A3wf$+&%Ttz$40rU!VDw-_$5q5!di*s_-h{hVpX2xa(<G)ER*2q>TikO?^vO~!t
z99zLdf;E9``1QyqqBeOEBjQ~3AU9j4t~trYet&CnaREV5m_FYI<o-6J`xhq&_$wSX
zp3bK~Ek>4Q*~N7|z(Lsy?{0P11F4T1A8o=~{5wYUUx<T3YYHu(oe|clA{$OuO=5m#
z)i8nEfxB4`$p&iKnZ9Y7znig0qTk?+HY$>w#Mmk98OIJYcX)ZOzlkeeHqKJe9AR(L
z{A`B}4}(5Q>NXhZu+H1=F}cYAHvWzzKIpQq#4@P;1^76OP;q=oIxMPj4O9mrZ`yFl
z5B}ah$lkOIdFKJk+O&&j6}45h!9gfxFnHpLCkpnaZ9iZa^3JAJetsQIe0kb`iV-}8
z@>W#kP&C_kxEem2oJ_mEP*)0hsBtQyM~dQl$9US82*-!RVH6Eald(ZQKV-}_Cr^RK
zoF@wQ`NKNlyuEQqHw91Ld+)t--R|6NjG5+b#V{=1YnqIG_K7E+uy4<tPx86v?ZWje
zwZJE$dR~V#2&Gr>$g>IN9WZ?nu)|si08hampd<<h*nr*Q4&6X8U|81(CR^}N5x;aC
z0fZ*d!~92*DdHG9!<XY8oV|Ga<l8q4+6jw>Fh)g$><eAcTov0ts)Uy!>`CHV-zrLU
z*z9uBt8YvW!*i3v9_DYkg~uK|cegt_zG)QO!(;(nCrwEDaWEOO?#JV}qy(Y{tP5Q+
z*kG~~SA2q}eCJRcTqd_CWuu-S0LM%IrIgtvyYMf6Mwx~T1Oqq2aH|qUD+Kz|jY)SI
z2_QW{zhBzFzhIZ_$tVbdEHe!1N2zB?2BR#<XOYBn`g{*h4P_D9$S85-QJ$4$Iq(|Y
zZ`Ox7dssDkxJRSq7}srj<=7D?S+5u6bacQ4CBu5=iL%6>P}ijo6|PvOJb5g(MiyTB
zrV0-g27^Jb2mjyzIFqwQM@c)-GLtvIc=jyp0RG-i7SK*_KncYGC=3Pm>i85^Cw4yo
zvc$gq!WpQzvB(i7{TUL~qYzQb+8Z&pO-fT{C<a7`4aIR#0S%@GgTcmY9gr3*ftgyj
zT{TNFAV6C5$ldy4xYO(~8xDt7JW+cRX~`{7O}IP;2Do^MlOAJ-KqNO5>s4F5e`w8L
z<`O&<Cp<}JAfuFBtNMVijvBWjq%2G?=84_Fj1gV)byuLu8s+~wvAfBHqtUE60#Tl0
zkl}H<QIuulk3ugWL$LbzfOF5fc`P3|r_5RQxR3=wkV@wEsr<Xax<Q(w?j<XXeNTo3
zfc>H^?&r&=<9?={n7AZ5mr@nyyh<-mG{2oBBpU{vfC;r~#xP0eXw2E%%$FY1lu(BF
zLH6in_^XHC<n2#NQUUP*PCV3%%B1)Y(qlbJqS7*~ifRl+;>xuR>N|cTr&h#91Obvk
zj_}j|$(B=xNY)!<n(<OvTLeIX)R$PEb#*T&y=Zd;ZwqDw8dvTXxaKzkQr)1RetW3B
z5xKdAk>%9@hL@cFN#?^MBssKO;8)fp`rH1vp>zA`kAEDqZ)KH}*n_#MhuT+`H%dAF
z;Icb8&HIVS*uIDnY#<%g&?=p*zP;b$W2Z0<N{C)$z1%hB5su6!^W%n|;??%K0~fv^
zVrqECWp;CKwtD~z`e0*KY@BLGW@y^?!a04(a}T`*jfeJc|KYwb?)OMKE?<XYxQky+
zXN(U@59EDJM-qWB0VzdSL&^9qj{~~^ctyAvd|g!aB$XrpNunn&9DhN^an&n)RMV~$
zd0u&^!t>&lyfp&(^5MPZTa*fA$VIZxRyk+33{Rf?`S7bM{Q1+byb?O!15u!r!4TzJ
zZrLE@xf-~OY$T@W0$ebSyFLPEOF-PjTZmf*fMFVa#`=cIS{Qq^rl`G*;;Ow&HixF1
zQK>!5csV+0xo117NEkn3KeeHq22QCy*8vx&*y$_1bFe0<8aY~!6h+D;f$05<jELmU
zACL@W0It8en-%4r3Z?6+rh%p@n+oMI!Bq9<`knEV#=3DluI^!|^uf^Vkp;ylDo^8R
z(T)OY8Zt7!(o0uYn^(upiq9G2m5i$>vUd}Fl%q;^F?G*R^Tn|U#-XU~ZQ?RrJ>KLA
zhq&9ldZd>>pVN0j6CwKnr7Be)PLMm${vzr03H#zoC5l|CW1YHDRH-BlGCIbBayupq
z2hDJ|dv$Bio;1y?YCcW(>}g%y?P4~16syumSF1pcyDowk;nT=OVT7Q?XN|^J(xw&(
ze1nOGpqj?q6N8v--YA+=A^<_e?K*r}@Dt1Ptm#)xo2E7S0p+K8Sz*Ujdj3`CFN1ZY
zMPkTl1;401e9$zfUo~xco@M=lYR8x&^V2*%|ElvJ3EG&zOp5(|bI=A7J@0{El)pFK
zsO#X<xilqpT#T$#M}m6$Y}Sbta&aFaZ(dfw@5f+Bxj}E2$Cm%T8KqfOzI`p0t$5pC
zD0ub_(jPS5(RvXwxc~049^itojGa+bk0|i@{MZLow@m=k*2ijTx&xzB9TQceP7H4U
z?j!keoUw_=<q`VK<ru;JXc?iJ8*hrDmQW7Dk$4MUyaxX!%vdDw|7`skFm3&rk8eGr
z+a^5O9ZXJAAG^b$lKR*k*9zmL&bi|;Rie(bvB}tp69<mzWG0%+g+1KZjs2+Ob~@cR
zxojvxHujqt(Rk*WZ|28&)v<1y@V^}%o11&=c)#Bt2D!3C*X5BaQ77-`{uGAr6f!*|
zjJr=_ojvSGE|&wy<#G^&jQ4uB;7O*jsmW-HX>5}f3;JeVbq|y!a7v(uFc7M@WUcBw
zJ6`<hGF}|C_{VXgOxP368jkDkVb;0|l~AaX%up?c&r>}WC36w=#H1UO^b3pK_|sH7
z#<lv5#rV*Nz5%9fj1^^UoTk9_YEx~Y97j|-@Lz?|5);<mG{n%fY~w?Rkf)bx>qCZZ
z!jld@a7=Q)`Z?iwm(juSJKTv$OtN8MtQ^Q1mD_bF!rOn{-h!{d2w9P1H%l7k*@`W4
z0}0m{aZm^L-*VZeqiE&HXqqX`CUZ@v3X&#H!^jx>BZ7x=V|^k7POd7tPujNt!3dSS
zX5hGe9;L5aK2K0e$Aw<?gC8v3D&rqN!-DG5PnU=@V+1IRvGr65p_=n5k!R7eOjgsu
z3Vu)^ePg~(f!q2J3TUTan#r|lB0ksQ?+c#0WK(1Ytg@~dT5Tp@FQ4>$mSNyJvT68O
zKmA6L8@0%AT>XbPjJL|>cjgTs2RoNz_zYHE(YVU_#Cak*H$zr+ze*zRX|@<U#4cCu
zfSp6<_S0X45W0X&-wbDS^0wh{=n~FK>sG5(vK`)JYd9QwXmd;JR_2>cm5&c)4ITZH
zdHsShUk9DXlHp=ppX)uvk}+-T_7caicxnFUzK1R~k=+%z$GrQC!kU>f^qJv?n`-28
zIiN%=xv@BIq>lVhlI3r`d3TOrPx_Rv>%D&tlYQvF$EOV=vc|2wosKnwn)#3!D6hKP
zsON!k9dV|xco?Exy5Se`Ljs~jBPG-arsvNRM(N==R^?<*%WAxI=;Q>}><}NOl%17>
z^1`W`Tqp;!7@r>8ujp(>;u%5IR`3Dt$3DQx)^BHxFCgB#p7v6Zi`od)0H9~aV??7O
z0S3eAUB^YYPW}kW<d304@F(}Q7ppHG6{wR}Ul)r*P$z!?wY{f5TdOX9Jz_Zd<E`br
zL2aog9>WmL)_*RPE>suZGpgMLu2`*2zC|sfN44TTBX)8olbV!_J~OJ(vUvR)50dKk
ztNP&C4>c2~uGArUe7a9_n+D^2-De{O=IMLq_SLc`<+$N3hZO$YG#~wX!8dmEAZu!J
z{_m7*M>^Xug7>4gw*EY377@rm-Alj$5bPTqJGy2_EErrGFJW+<Tu$bB3j-yD^oN&#
zW$PnF8DV}MD8ozfrM+6su?37RMIub|Ly>jc4wbYGnK5}olg0&-5?b}zlyDmI2D;oH
zeo%i{f6%b61JLBaP+7|U!C}W-2d9%s26KJXVD97>GLjD7uw=NO1emJW-&nNp<M*Qa
zs65+Qwsb!sFm~ePQ})#K&gF!U^&T8{mHo4vf$~v;%ZG=Zhu6G8O7?bveI@$MUTy|p
zGfhTmwq+oLasPhZ7-$5WXbX7=)dq!1H_3J4@q4S7!g1aH&wYe;L?%lMAcn3BX|A9S
z9>gB5QpT2v8R|o|ec<7(qZR#aG?K}q*LdAWx7j%EAt7w>e4?(0!=sDu2phBV71s`c
z&9A8p%T>Y@t#D$=_gvB4(7u9{X#qrTX&AeotyZgWn2XpPoYU@+=Z;nQjqWoiLq1=_
zNimlzCRo(^KI^Jt9)?iOVN+2jMk7vci~+gScitGztgf!|J3EVGdltR!Fw3ju^ML2(
zX6J(=y_2b^s}87&`PI$gYQ5FH%WcN>is(mkd+>g3s_X+<Z%S62VgTLYgf0*Z;;akW
z@*!ae#?EHbN_Jt`Ein$~{iw1~2>0|&;$>rTaX0MmcFO^{SQI$DPzX*zAF7qKfF3wk
zpx*RiYXuAMS}twxn3(~HqJ4X>+M5bO(J<QfZ7_x}qaX(6QI{nnN@m6{Vfr6}=vM6|
zx&VeQFqH(^oBF%#U5w4v441Kj>*)riP?}Db_UuWPrb~e5=MU|#1!aj?vX<9W&9a!P
z7E}#_AcjvXYJJgnR5f3o4$7sHpLIG^emC9eWIGQZ*+03#_3O16rj``d@O@KN0^j$R
zr%HMr&)vfNVT9_jE#W-RiKH9CWKswk=+JTdNaMywM&ig-Z>dIg0hW|pEe_mY$F6pJ
zCiHzp5ES1JXWH#uOaFeVAgJlA6JE&;u&km&Wz4CSN+YsuN^|)##T<;l_p6mKoKn@P
zFsxMl8C?+gO7}z@4WJx|!&Fux+oo!z(kR<D#xk`DT*bw3tc=natGJ%@_VR=*NrKI>
z=lBkZ;#5yKB=U8cILmSx#W|agV!gQV#**VGJf96$Oy=h*gXbD6yJj2Mj5H~|Epp<C
z{8U`7lb2UnHa_!A*lzDS+&u!$L_hcHa!}4^NSk36k3C+C>Gf>AJUv}a=o>K9gBNAo
znB67Imn_cgl;G(d_Sd#v+}^?h>?SJn`G8zThTfcG$KBMu4L7Ste12MheDx<bv3$|%
zzADQ<RlV|b0gS+&Y*=FPXklPOISSq*(1F0iHYb{{3JL9ka`+SpL`D!rVFWgq@}De$
ze-5!{HU2xz`T{}qobK85hSQ|}%_+---{sG!8rJ}QlkYU)eU2`oJtOMp-jB|rr=n31
zRE7nF9x4FHB9x6(A@S&F!lUTr$+&vZ8_sJJwn~xzlpDr%pR7336BKLGSA_#CrMQB*
z;+b;8m#(@7n@=vTa+N!XL9QG5jE4*cqnNLr!6+T8CU626>#QQOu#M<tgPnwLcoOxU
z_0QYWvsJW$Zj?FdfP4wHjib}c5VnW})vrK0!&6IXD(qwHt_g4}a#*$EgOvN>%?^;g
zx*^Xs%G0d??2C;?qdeWJG+Tx|2ew~h54EPBltY_L3v7VNH3;u?z3i2=UT3ZbflL3|
z>2&65_+N?E`-~rE$DsBthgof{eM`fRvuIu~5Wj}FOOK0Eydsjayq3oQnC`^wxIHdu
z(-BJ45><@IVTSbDc35AK1&+kv`9xD_k=R`Og&nQdR1qv@npl<z0OkoR)RqsdTL8tW
zR%-{%C)&?CpewN`_BT=?iM&A}^BD^}i955o^;B9}SU?f2SF&fuhQp!gr+&y@pCst_
zCxw7QxUTnu){V4sG~(fK{yuC$AC-KyE1-}<<XYXXzX2+50!IPGG2#u8Y(W6OPU&m9
z{mb<C?l$}5(s=ttgzy0vq9%HY&!;)F2&_XRKT-7l{Uo-G{scd<`za{JHVtF!O&S{#
z4R5;*oQ3-|K(RPG2)3c?mL96`FAPx_F*hB0%{E~C-{ezxcRbq7O62Zz645vIrY>Qp
z7=LL+Bp8R4h_ebQB)FukCTZkgi&X>O=BNrkOKdwo*%AaB0&0ubx>d9tVHw0l5BtbD
z;CW^e8!~JfblrY3$MljVX!bRjMjmv6y~$zbioTPY6)fO_BUC}iH|<e;3j(PhU@a46
zematoup>r>Y})~5xQo($hL+C*Mh5JmO}6k_AOD$Sgi?|cGp4r4B~=#*xpDH9M|So*
ze*E}xu&m-81-oDu|EmVc`)|(JMcIN4OO~X8Y)u9)+P2P}I|n)6#8eA5mUL8RJYF-#
zPw8G@f426r3(k|-_8A}cQ+U>!w!YJ3L(7$eZl8eCxyjFbJv<8axi9dC*(JjWGPDb!
znlHk=!|J_)HsAroBJdFS%0K;cairhxCy8-6i7(HGFo<*Arfx+8C8DIr@Wl`oLe)|_
zGBegys$OIFD}=f(hwz{h7wz$85!pcoeZoYmf@CQrq+0>jObGNGRGNV?>O=+Q6%}v#
zmrF}aAHKb^uu!>usIy-JhE(grx|Rf5r97r+S#Vyvkv9&9^7(w<2k<jOCR-5LMNQq5
zD#^3rCskTJbxZ=dOL`%yjDR}}^rP<hI-H^@D9<SF0mVcG_&Th!fN8%pUc%D5iIdiz
zz2V`8Oe%#b^N4JJ)1`PRzI3YaZXyqFJPY<3q#^;btBCQ4cOK(A;W_k5^g8ry^mDHI
z;A8+U_r%k{#yYkmeCObxQt|z}9kme{=ho8r6=_Il(_o_07xmDR&T=HbNEPEn*aWW#
zNCbxb{sBou(VauUD;f|}#)ifI;XF23O8qD|e`7j|!1J6S$gE^IBF|(+mUUiqj453f
z1ditg)x+3R1)k@4QP!snN928Wm$GH?i<QZb)`w+XA|fHGKsBkPa{!!Pk~Au)gos4a
zWk*&N*^zaL2-4%HNkU<^x;ai*d3$fi4jhF4wZY`mnUX2FM%}VgUo1Pk_4hDF4bPVZ
z?g1WosS(jXyG9h6QS<x|+XKkKpbj0ib!X#try2upj;u%Z$a61nrTRm=x{4tZDG@mU
zkT_ncPBEpm<07|f_X)e(J+h;vu&JuTa}<CR2^Ay2-R?DT*ZhX8F6c59H~^x^0E#I3
zuz7v2M_0lb0ACap09hdboIn+}uzDzti8RL!TF&;T;lq5VbRI&MqK@Fjnf6i=LcqE4
zq8`P-!I)iR1D>0dA>h|Wv6f65x$#JS+=t(Kbac7$VABqVs4WTnnH`Ea8@MA8lhjXd
zi5)V=Y4Ve7Qw>5mD6PM3kKw+4dAD6y+hl1WR73|s0+lHeJVwR%n}#&|I|0t%b(NGA
z4+JV3ld&OF0lXD;no$%Vn1evQ&zs8`r!!4sI+vB-%zug@<R$%bT+d=#dM#~ttSn_@
zfZ1kxoCMMF0R$NkgvbO7%b5|~D$1{M26=sSkyAtsvbvhIoY9lk7!&huGyrFf-*b%~
zSr9+#`PIqzon$W(31YKolNZo+cURK@9yq3>f=5{vsDqI$>qD;6a!^<?c$M`?FU$>~
zPyQ1`LhbR{FCLdjG8jh;Ejq<72PhR{d^i8I)J=8Z1_IM4bhjJzD7@vO$co{r=a*9r
zK_m)2k;-$)#@1F`H?Dk>hr#v2O08Bl6-7}D3QKNy9iU(+ilUh1TCH03T)^H<+}>Zz
zyRW_WTEduF4|A5u!@I(|$r!noH_g0L65j=qguMexaV4-UNl-Q2P>{k7pcuNQ3X)|7
zm3ZlZhb4&){2krjYE2szKMdnXE5>xIwLcESIYThrK$oHGVvXh?D|l#CVALbn1*`xX
z1Rx0<&4HMKVT2Vl7|hPv$)&Ya4YqjFbaUA|ja~Em-y6{-g_m;$#lJ^44D)%)pTb=t
zyT)6ehxeeR=q*9^rL78S0>dO37C?M$9cDtLZ~))*k+_JIssS1DcM_SJmTj)A^cLnF
zoAEp+a;vWRzvqIoCNRe7wb~LBI2FadlW-!(^UQYU7kVozrfr#2eq6b%m4i8r6WCI%
zrgMx5aM{BcK98b#eGrq2omy#VZjnSIG`jxXA$q^4G5uH(!_J*I-0+5yoa01M?-%Vx
zXKpqhyR-pM>7vNx<kA~%xMAl`z{O)a)5HdK<NWMgr(qY%Pr!?GK8j;0rK0dw1=WZl
z9Q9y5LoANaWKbpNI$mNBN09S{u1i^)aVoDvhK_?Ra7exKD>>F?vePZ(Wx*lF0-H$<
zqdKp`uZb@}oYG(bKg1m?$&oRBpJttMuNlUjji494S=*LHRZ~@75Upx;fdP=Ca7yJO
zgnJ^<R9C(Ar$<$_n&$!kAJgt9wr4OxCaU@$+>w-PDv+J1qOqoXaK|SkS}Ri*Migaf
z$LMckoc-;R1=FUDCT$(Olvc8ozc3WuHA-5k+hh&S1Sm)JkKu)STJDk1W#Rgh+8f0s
zwMp{bt^sT0ylEV7O*}1@QOZug{N-t{LVT81>&o?4F-k@9EeO{4?v3hICQY0NR53(?
zH3xd>%U^!F&S|}B{mNEL`BmTXe)9oAJ=(ihEDvChsDPOUll!fY-$MJ)b=>j6>7df_
zBT1KD7^ZR8N*S;LMm2R>FRD<V4C@6Dl3OtP#Dsw`T`I;0&+?oP;6FEpP>s+f;fM9o
zXUU~-VD*=z3k&BY8qLjx_e9>sfpAo}geFnawIiO$aHX&G?4PHb!WP)JF<)D3CA4!c
z4TPhT)?#hmux-x$2g;ghx|7nKY1)L5O9@+;am<eH>lb;9-R#h@XBHTt?wg5WxUON4
zH(!3=efQnhaJ|0gHsI^}AIrwOUU4E3iSb5G+ud!$OV<)a0=^)vLdu^#mkQRx^B>~x
z%Npa1-AtyOwMXMQ+G<gb*Y;Q(mDz5CW18G-${OSK?S78M#Rn5Uxo=(PAE9FaA6Y7>
zD=?%pJ-icLjUKUj2W{|eoj4)O<o-L3B&9XZu!D|F)~-8#Q%i0EAVu)C!zF^K&q6v#
zFrr57!i6Tg^gOB=wK0*Omh`t6b_@WhVyh~p^Jt>EsMb0R8(8XFMc1e7hygHEfOFh^
zGM^;%Kz(L8hG-hD#c;E^wtHVosa>kkF_aC`zwF8N%8mp1>%-|C$t)pO4N*j`?DyCl
zcJm>=4m$NTeAOY1);Ot?92^v{cjCyU=R15=v-f|Nl=`<=pvm+8a?qQJGkd{brh17s
z8tYX{J!ooZhc?o2J6_@;u|~3#f)$04r!YNp`l`YZvDQt;M((iGOSkn7JZl;9HuA4V
zl<3>S*iDy7y=7b_z}%&TAC+N;1(DlvyNR#2nGFep@_01Mz4isu`;VsQ=k1+vGh`@d
z0<<IXvag1yiZ`Ss4mCH$=MF+HB2*Dis-C78VYPj`PX5C*89Vd|4`c5WhZr->0b_$P
z+VQY^88gj8*z@oqPFp86%>ivoygctQ<hc6k(eXxtQlo-?FD?K)J&%w>{AV2GL;uN2
z84%g|C2D0W(@KxS_vvlYK5-qhgrw!4U}-URMy{)dPBr4p0usd_g!1L#5u@lBGcuaa
za3Ms9&tdDc7{TY_RHbLUca~g+USj2Z3)LJFs<Co7i9I(Bh#SG^$tR<U=X=ucC@qmM
z)L52bI*V|PKBu#s`t=CVQ4u2s>Hvb6#LbKUX|bCo(C0qIai8K!XU?3Fq%#~hIDPuG
z(b(Bs@Xu~;Zc@uIO+_(H!=m3kefqS&(A?QD?m2Vj49A_3q)%IPvuNkOJD!G<{z7v#
zGER>x4uh*!vZp&%%E7C;X4L4VjZjkZ9$_QxH6pY7s$i=v+l_RK=oPzQGw<B$<&i`5
z2nlahIq!?fY;5G~5{&3btWx{Dgn2bhO^$<$oJ7RS1+t=paJc|ja{aYt6U(1)_VmIj
z=M(wT<RTW<%@1+2x#qhP2Jtch^c5mpE)t0oakIHrDtz3TavJr>$0-pC+x4R_^5<LZ
zpjnimqv&;7+0&&4=6i~EsjIspO6yIpW*ezSem4)U%EBGPSN!2}=XC8bkrmQl%rw1J
zmn?5t&xd)+lxDm|e3due8hEQ=txKd~nv8k7bZxXKI#~naS#P5Sfnof;5p4hG_Rp~h
zBUC_|Q)rJ?7%WdE>X()e-R00{!8DCbdsdg>VZm4P?X7!kSyG~KnMeE`T4SW4R`jq4
zL<W{oU^*{NwR%#i5}m5CCX(=fsFMPHp4@)^kmoiwHVBM>^t$zC98RGY=tgcsQ25wz
zJa)KaZrqt11{**c4FXc55%^ThIuqLvEpDD~Bsz!*W(ZHxC?*yz4L)O{@S!`j_cqnQ
z<2b^Yix?%vrHBwtS_EKLH1>omRp{fAl=iV!Qnw&_dUx3h_6WY8(vKJMF&}NCgFfE$
zhGvVEaazEQNQI7u>-l0x8}&$-EiL2SbXk#(nn)t9FC<7l%RtM$7c}j-Kqan68%+Ta
zg~5{134-`+?~Z4GWMeoxi?N%V3(I!bkMjwf(}#M!cSo?c&ocQ$9-HavJ*!QA7^>|Z
zR%_UHYpaSPhiAKM-+c#jCNK8tb%@$_73b^7OIh3B#jk_^g?6Dclt&7BVEik90q#tx
zi^`jkuz}}9*d$RcENr}AJ?*|%90~3O+OsdEnYO%KSS19n&CNy2$pTjJqy)JRoS!Or
zzJFUj2=Xq*CXis54TUn#*PAOlnuEXnWxrMv6pU?s!{g_d8w_2z?PJXyE6tj~mtk^r
z91L?3m>9eHppv`I_r20o9xL;p<M%37@WOHu&CLyRyZuZ}6<R0KBPVXmRuKUHXn>T2
zwAHJmtz?`z4AG8oV|Et|xFhaARay|CSe%{qy_uQjTC?eEIX#$L7>j3W-BvZErlqD(
zGUK^kOq$2@#>`B+?L$0NPqK(Ii35@A{m7hKMw#WpQ#I_x11K27T$sk)4g^Dw1umwq
zmBDsO(EVv{ArxE|^fNpBP}Lse+IGLCZ#pJ$VpF&7FY<T$(*5_}fB#Im*DKH9H#A@Z
z@mhdQe<P9a(klSMD@$pp+-$fE$8Euf4#s!a*kij<6ca10iawUcEQ+FtXz;oR1m(78
zN0IF**!o_y%K;E5D2i7qd2jUZxZ@75RPq$%?b(kJ(gI2uZH%I~IWcNTAM{RGP}DTn
zedAm1xZ@7j)wH)~|MBpZ#&OS`F2azB#}iCIkuEMXUm$ZwwCk`OaiK|Yt79(QefQlC
zTVoz2A!jh0ZGC+`vq|=1V9eH-^9x=Q7d_0jQPAcuv6$cfB76p(MOVzTiy+M)di4RC
z{*^38Df7y9mLs^u&gbptf}k=#UpYq=g-X)7WbfXj-F~jtYSqGU(=U~L*WFnE%5TWB
zs&3jg++u_ETxEW~5(MWYi7E;`*KQ|!_a@JUVXf7wZMv>sD)}4VF0rAivb<^M^ERJT
zJ?7c0J8=)oFfL5|a{b1ZU+Mc7n*!Pb2kCL_%k9GNk+uRL?9gb)<Ndy1Uv2|-{D8<i
zbd=B`bY*-#6cm|xMwRuK(RP#{O!ErfjG1_}_291RL0~r-<}(&t)&=-W-ly+33H=4c
zhwmeDqN1wI9vU_FZS-2Nl-*v2HAV84sQXZd6|&n|(vgkbUT~Qo<yIeIwFu~|@p^`2
zXTn!l-_N%oYZ3_v@CMqa&|Y*Dy;k3Jv||D^`X*5=-PqMPs`E2_k2uf|X$Ew+F&Ed%
zT?2RY=8P~E9Zd4@9!*hn+f-D$RPqck!U{b<MPm2V5_uTIu_Mci56wG<1NN~u`4>es
zOhvbiJb-WeIjinH+%K7?D_6)0xH;};zrbqd{{$mAkM?=ygX!5kil-cjkD0>+_!BGA
zGM%kQ_0%_HvsRDrs#q$JUeaZR#AM+)5mgNjW6yuLP^$D~$C1OYdqt6Gs^=;P(jAOv
zM1EODF#hNbe-F~KCJ68uf#atuJ?1zza99+KySfR?@w}bS*$2`cOpq><d1x0jCSjhN
z#yIpTuZ<T#_5BPWEHOD(u1~jQS+YMz-J-@MOhf{J!FMoM5&#1s5~q5(tX!1j52H-e
z2Q_$1>DHoCg)%t~FQLFzE?EOTdf-Z{oIzr14av!ja4)N5dcGJ#S%^TX6%Wt7Gtqmz
z^gFY`$1UElAL~>F93EHxs!h2dTUQFH^_SXI5Nnqj&JvH%pOB440$AMvNA)V1498xh
zjQamL>!{z~ss?yc<~VS@IeAn?$PaNwAH;|x8ftj*g915AAz}d&q{s>W$;o|HxFTwT
zWqPy*rx^>uUhVL#&kyxi3*aHd$(i?FN#P7X191(lM%hV6ur;`Us<w@~2!aZgwd~97
zLZZaMvyWkrB+2E&i5_VI{l3`g(B!up5iTG?25_~sXMtU?FSh`9K#0FT{Io}x+sb-D
zriS8yX~H;u_WV*D4Q<Z~7h!M&Xdm-w8Fk~(X=05ecEizuJ4<zH!2{oEyQu4;KFI?0
z$r9IzvpO%$;31OIAp<c&SI}dqhZ#rxQ;**RBitF>KC%e5Uql%kfH~OZ(C>&`f{O2l
z-(`Bp2z6&7&_2T>lnxEs2C!{o_#Xe@Asm3LV8cQv=a`y-zTX3bkp4s;t_>B@7;ZQY
z+zAdJylr9>(Y@lQ-;@vPC<G`&ML_SNczTH5{tqEd|NQ0P+(Q;{W;tO;o2Hq>W*k-~
z=uO_lNr*ud!<Ql(jbJz$jYjghY$ohiZ>L}5`M=US03&2$C)sYCIOMY%X=>UCsX%dO
z7sG0FcD4?O!@MSOzJ<4>{*1}h!{HL<C5g|!3)PYD)miuB!soE)|Coa97t!%|;u`p*
zZ{ffK)?heyh=4H)j0xy4JXwz_guYHtNEw`qGG^BLnQi~ff3&&wS&4q;O4@kn8`|i7
zZlRqTT)1%I<g{OSuco~hfy}+=Kja3M+Z=yOI;<L{(E-G5xxK|zV^nPZ6;vhUzljZW
z$BE2}mzf+YB;oQYGMYWEpTJY)nT|p6X;0l7TN4k}>xm=SBWs#`C#8n0X)=Ek|0=Nb
z=yS5B$pcE;zpe|fHdg_jrSwi&(`18E{wDs<;MXpbElLNnrpfJJO;~NN$}IawEpC4j
z-Va}?bR+-C<$U+?m>OL5hb4uVQQNh*kwK_CxSZOq4agvpck5;FCg0HXE!}b8qpzgD
zM>0v{H$DJ=1St*lp}?}hI>F=sH7*O|$hsBCb22~VIj<YI0G|3?D4ok0I!KG0uTudz
z_s8bv$o8YUk^8ea*g`iH2=U*XKIIVt0*y7NYyjDQG-v34CJ130srBwSx^M6b#04yJ
zZXpshLGWaT)X8hcJT)B{0eO0pPYDYuW^L<W#cr(=Py3vA>V^Xc8I`GGS@yK+iacYy
z=(^Ld4`5rCLS@Ph9=L&0W^3RhEsiQ!H0W;*78C<CXW!nCrDj=Q<poKV2$<HC4xsCh
zEU#Y<kyMEn)P9z!7Nc}>LA|}|`=F|>FjI>T^D-E>$5fKqe@PW(u`@1;-clY<V)J^y
zAjnuf@=@uHyBTK;nr2E^(A21At(;WC#3UEDjIUU=Q%zn^I?UFXbJ;A?I0MFbx^)Go
zXp{1g(X_rNW0Q-<h4PqRIBboLxoK{`CmcP1F)E`r+B;HMS&l*bg%rrBo$eHSc{!I3
z;Y6I6BSgE8Y+_kxn1Lis14+svn3kl#9Ib~puL+)C)otBp9KCQc0!5i$A^DoSGOx%G
zwH(zA@rT{knMgY@+-O$krgiCX1VeiXWvC0rNa4KflW8huT|$~!n|x%o!^g~}r&Li~
z=^{tUWb?^r743w+9_dzjVh!!(*Bvhl?v%MiAe`TH#UGDo0)`bk)*BP;<zR)vJZ?7E
znlm$=KRsJ4g1Atc^0QSk*|^f?*OGPZIdve#vo$kg@I0ne;(9Z^fnk<O!)mKrn}Mf{
zk|jf7kBT1fF!mmZ7(jaI`0VU~+E2oFM{pA#YMqc&q-g%R!e6C7J^G1t#@6p~)+s}u
z_7Z9*f78ujUD<~Lq8UMDCIR$hLNNGS28Hz3J3f6ux{4Zqs=pG3&<<g^;%dxVSp#ms
zAnNA5bzquH=}RfK6h+&yaYeBNjHT<f@efU{#pdp#cQ+TUL^Mp_ux3^+Tj!92uA(=5
z(-2nx_KHlml{q`Ftut};;bDFD3u&tjpGPm}NQIe!R3s1!qSd4-A@(!S8O%72P8`*9
z)H{-)7>7$kM_%wKFNj*9%V`VKW9O|ZII0TaOe&Pi3+-|c$Mb#hQhe#5I2KpJ8BkTH
z5PU@6--yfLgPNl9^satv^)Bi%*Zl-uBBtTqwK_ki%@?OeIZ~obPV}uWpt*QZou6O5
z%QZ|Q@q(y*y$K1Rnc!!)kCBGzs5i@@lssQD@mfhKM1yA+H#Xx+Y{s2Xb0lZD!$9B1
zfr7Vfj(}_spUPmMU2lt~Q+7lQ!f?qYA8Irj4FW(Kjm9YZ#WjM-2>OA4j@(`@^ogm!
zmv|-QQ(k^XJePjI!^X<W13Pcp?smeE5Z`P?+_nKNNm}iEhe{f0gPnm(ns%1icA-%3
zFSgbB`BnRHN~PWIYuP&O9<f*F=T)0gI@vb1$2X{~N=j%Urjra3vq|X+B`1-W8+3&D
z_IDS+yA<w=dX{-HjPxT#C}8r#9DSr<lHr)k0I8pY?ky0K+?M3FlJQ02z=tlbj1Hii
z(VJSp*2HPB5#>j}NiGfuK@(&KdFA?_(p9^{{X*V63Nub>ex?&ja6tqa3CAuLvg@v`
zudi=72@G1aL6<g~<{MpuHUG$i!Q^=zaynPMsh6(T*>i2@$2Z_Ydy3NNjX8%v*3L<L
zIc_h;?i-JSH>Q`uQ_q9Hl|4G1bZ6@2yDqN(<nAnma-)$c6Q$s9IhEI|t*P(vt+zch
zC-C3JW2B)7_0SPMhGCo|Rod}Ij+?cV_$SK1ZEMU82TX2M2NksvQ*2`^So)I~DQN@{
zCX!bs>OjPux~IF3lT>IUDWSM@iceRrK-tj2QJWh2X~?zlU0XvKqhs)3j8XyNmaj+n
zU)+XsTq-bj)kk9Rh>R0J9}d{7Fa6g%k&kaiCV0Lf<J4(YImrOnNku(<%CODzaF9~X
z_VFe4%qD3XNE2Ixl%=(qHmlFH!PkkdlREFww!V}yIVGIv>2w2oKWS@JbkG$9V*<~8
z7Io1+bOkyQ^%CjxYlpdl*FV?hWLz+vE<SZ^n7Lyw$!i;+?j4&{JQq~Wz1nNudqBU}
z>q(#{TSr))y^sOjJp*vP%1Ph0<XZJ&xis|ulF#RP>7)wh&KrP~{v?maeRLk*Asc*K
z3lvEKP1!|(Ut!|!0I=1oWHJVGvF-3=(2ESjhs`@rojRp!rC!&4Hh1A=;8;KH)wS88
z1=4?Q_E7e}3^bC*H~H!>6Y#4(bqTHh>GlE7sF9Qa;DVdPvn-|UBNI#E@u5E&jaGS#
zpsNo*yc-QgrL9AIc{W1`wvIE6%eC8=?I!R7a=N;kz>g2~=bVf_i%%mj^@x8{_f@Xg
zZpLk5N3KqB@^!ub{OJ!0yd_QC?-<*HQC>O_a?_EV;d#B5T#?dKxJk#3vx}eY8LkbC
zpR@vm0DWJ{6`6L;rC(h%rqVzVyzfe6hpdbLe?H#NvkxQ4c+hqI+QDrC=?BVtKjaKX
z-_sx6j{I=o4?54aUBGW9X*AfxDx-+&Wn!GLQI9HPKvfu06jz6@^*`JW@b4AR1#JH9
z17(hAdqgD0(a7a+Q{Z})=_1Q;seFD+B(*GbXcDC!PA~j?>>dt1erWW1WHQ)@`UNqD
z>!lpWCnldNymvj`da=%fUsQLoK1<^*^a{Q?Wl0M;13dn7*@-)qeP$3;tI4z*($bs*
zk?I5W$V#@EdOoREr95xfjhsYQ{7{16@Fpo~R$Avs8%yM1NIC=E_6O6g>_dM0&zh*U
zY$y7s^jtgH{?F}CVdRIAB#-Jk2%P0|bOQNMqhK!Rt=8R%1Wah)JQvLG#^&%$ioOJJ
zzP#>5NImxZUbv3OY$6p3j!j6h;5dX7i?3h5e|8E*LTsl{kcjQCl0yGYeK$<_TAfE9
ze*?VP5~w@Hq9i)YtCOF7DvlsVyoDhXC$oH*46>E7v4m$h0A%gJE3nz6&wwcbNTPDy
z@J%D_`u>;Tnh}`nf!P;yd{I{#%}u<Nd48qhdrVPfYL-f~v!#+r6-99r8KR4EOruoc
zWO30}Hb(c!rwlD5-E$0uDyrkS!x=|as3I>9m$$>hlX1+v{D=}AfHOaY(+Q6Xs|;JU
z-1r*+uM~8R)2h|Q#cEaKG+mGjg)l4>Wc}NmGa4tdhBb0?8oIz~(})2h(;rWL%6IR&
zh?@FAWX+2%+?np;=R&}e><k&edY?+dN`@CeVPo17W&0IC=PU#%Rn~AZZWP-F*>X`8
zwQP1lo$J-?+v-t8U_|IM?&}(LLQwJ&cV<XGo$Yq72J~v%HiEXxiF_|UZL)=@fy!6_
zLm{s6+~2SG&^w0qCpXwN5kf4zz<2U=%lrC%9K9QT6nz2x5d9nabKh&Fe%uOg^_iu?
z_xbRrkCgPN3d&Q}XUzu3?q1f{<&3_0N0rN4^Wd{H;DNbJ&s|;l+~eGIQ(xoi)wk&d
zczm`7ZmuwIO&VuCe-GV-?nfKE=7|ItI3_IPnGm8$Dk+zo<N;G-ngOGIUwCjdwc)||
zO46|ufd(td4Z?6m61YEL1YNthFyt+>5yA7*S-YEd+>RI;P5*WUjM4+EX;`n%P*Y<7
zPW8xv)l8}uMOf(;#zcx#&9!rLic%^2j;kuOj-_diozDOja!i_<%E}ej^$_l_&oEiR
zvUrI~*5rAYYK7{QZXm@b>-)g9?Z3I7W{mA)dsPe>#Tkg=ewBkmoO+OmkO8Oe7Ae7p
zdErvy$XQH<!&t;uJaOpIA^89n1o#f=(WOhWQGrK_E@+S*Bnu{aMEijGF(>Zm&u-O2
z>J^GTOM)tn1x^l3o-J%?JY=_j#6$Qz@=ym|OTUuh{Y2{nPrQD8+=WGSsbC>_ecFZa
zu%;^fFm&DYVtSp@6DN{m26#gQlW#A<Hp}3%oG5COEOy?10>*@bR6TyYvuOr<I<K59
z&^OU`z9&y1hk_$l$FmYXIG9&r?kbx`lbx`}3}Kz7MuOg0=evgasIEV18jL;qC}WSw
zyh<zxhH>^Ggiyd1&&#=-<GQx_9cjZkojssCj{X2+mhs?&hMALsbuj8=P%O^+8Y7Zo
zVJz5At>({W@YR2vkhcAQ;p1=?+4xj6vid$vZTO#r8TMr4^J0Dz=km%1yv8;d*--L1
zyqOnyANuXm>b1x-^{V>~6;eHMV*vqK(y7!PH;@JNF*Zx@wIqy97^i7^Dki$W|LFIx
zI<DA>NqG6ziivBw-7f54EX}?zBaWMUC^hTZ_eH01D{uhp@=8nIE!*s7G@l`y?pUUY
zB7vdF@R)o_yjPs4Wi1~=&`^SlJ`I}YmBRj9k4RhPZfKu}iN_Q38)6`lr@0JqvBylX
ztRjj1xg`$POj%p_N}SN;o%55~hUCP;ww0p2=rz7U@?yHmu&8V_uOg9VqD~?|v;&hK
zi3kieBe82W@{1NvP|a~X`Vtz~?-c#OlB8g^(^+`aSN(k6nU-#U{mRM;aqDZ?wRDPW
zPS@-8Psy!0&Rb6qseY&I=V#{^DiOz3Zhwcub9QC!M4{K~?R50+#?`yqB}$dq1i5H?
zjK=T*r1A*#8kx*wKuF>|poY`km&w%O(=?ejc^kKZWl(|i#_>_>nfbYIWr7*&)k6H6
zk}sl4RL^qDRLh=)AzVgpt%n_J;&tL1$pGt60B^k;Tu1uXILkz8#M}_JblnIn8{&JF
z(;`M+PwFR5=y&ncWysfGFB`$SxiGRQC5A;qUhkkb44(5e&Ugg&`q#6ie2$2Z)ya;a
zn>kpmVoFz*tJxBkEx|pFGfz0%tsuq6wT&`tI6@dj+gmt*CsC+FAm61VN+y>iF`ptA
zx9l(pBPWLJN+5jUg%>Pq-7enPuy|2--)C8m*oC|!Qu<%k?|x(r9(>`27wkgbS+_s<
zEx6Av<ef(>Yx3m!8y?vY2hoH1xfRznrUkB{z{P<3BN%RgV>lSM(z_G<M05u5r{E0<
z?8Cv%o$jV%Pyu&ndAC%+?#djs8g&`fp`8M6lThZGL~@E17w`3H-}P6`T`q>BZx}uV
zKIj61^=dDMT}G_te9WnOL`FGXgpNXj?cul$5H;S0o$;VP#s!o4xqTXo=))#0O|<vT
z&6}?#p~og8AM9v9Qg2!R!~~Qx*4k}3+_^2e-Ck>~6C@ekprsIbXx*Ye#2B@@Lwq0s
zb|Z)qZWMw^O(cpSLt0&FXq9*Q2VZ9vR@htkNw{f#e%^ArwHatZVoS=wy{$Il6ifw0
z_N`2M%es$GZmd?TBFxmfw8nACKEyKw_IB!NkNg1B%_T%2?{*Qthto;`+0*&xL;~Y^
zhXYfoRi8~aXZM=Rn?+cxo*2fRR-MBJO$*jhyTcVxpG?D!g>*SHGOY>yNhYz&5t@Ip
zQ4q-3a!atnNZc07e_K4p2tJQ8bQSEFDaM)W9~xnNh;*UGB3oiw>Q<}3{xh88Ap~k(
zPq_BqOMySAs>()=u(xDdTb5bs6;#KTtV%FvpZv1Df-_?34}bw@yrntXD*>OB^}a4s
zF5vBu6^hl3jcT#LN>;}rs#7l%<f%kHhn&p^9G(0fj)($TX*`Fg8W=uLMQ>8Npg3d+
zk3j;aP{E}k2y!NpgdH#}<9+sya=EMt0=bD1QS7+0SM%K5(o!7PEP-%*p|DdBp?X|Z
zRadvPe-W?oKHF0+mzl*JRaK8yK@@ft3Oq*ys}{#gOLH85^{m?wMZ#_-f}s5rT;<fK
zPkxLFXn_}8NhB3NI}bQA4C-rq^eAJdd5W=9%-`<1>-9c*)HE48WtykZ@nq%uc8|Z-
zh~L{P98U0OR=9IXYFu|V{%)Z(uSb@DDaVtQr(1z20+5xLTPXiC@^K3vkj8QHz57sn
zw{Az${9WA`1)~m2-3ACiet)zGa@N<+4e0Gz)|yScj^>aJmvAKtM~2RY9{8lq=GwZk
z*<#W0dVz91-?R8NFdpi0rWZOSMe2q+W+1dhYN*Ej5nEN?gNu+;h4-`Ez>V2m3)QOS
zb>B(ZwsUDV*;B+4D3#eaP3^BS&#TSSytBEgHbGxiwu@lwN_@dwST0r2iDU!Hv+7CX
zqf%y_g#g?ob4NCAuLI*kzS&HYW;0)Sl~LWAt$><7!$}qVS_sm5JLBbrg+ifFSXfwo
zm0GYA)|ppz-x5;mU1x;9JYPaX7y}xI9YAUCvydz(q@y4O$q|Hwkyi){KXK4tVZ6Y5
zR_N`<c+`L5c_haIz9Kj;h0cNiZ;&+yA5u8xwnG1ieU}spzC81MNfywq7}saR!C<jG
zU~vda0_%1<Kv-~!dgNX)&S1kwv&G{8dk1>GN~KNQ#4Ek)rtVl}ZoUQhI8d5i@TM_A
zI_laLPk9(Bm0s_Fhp~5EZ^c8bZ(@FKuZOX>FkJ%d-C5H3bzr@T@GMeq5tFL-?Z9d?
zGCciSe@0a`bwg1$RILKs_qBv=G9r>=MAM{WqIeB3ExvAgA2iS#K?Bs)jL*hBKhrrI
zBqjDE6`J%s9$O0V5xS>B7#q)k*)+z}43GJEOWP({20DywB$F*&ku$ONl_c)Nt*FOl
zOi-#-MPR2gurwKrM)n!g6BaaS#=oUp7XCuLN7q^*2?R~o>=NofwxOX&2E4}HS0Q!`
zx1~7lZ|fkG!fi0-=fT~B=yxmxhwCoV@i0|x35gCuAx37xbxCBBknReB<3VI18L$Oj
zOF`~dE)E(4msfr~0stY661#lsfCx|=JU5?qvzwp_N%|Su5X?sg)YI}0t3l%M&PwuV
zK6XqQk4R}DX`*aPNJ_Z6aW8-f>N6bU4+fDjrmF5CSJMsc+c(~L<DnZxx|50`*0G2t
zdzo7pW2#|j?jcuI&+gX5Iu`f!^D$1OVEg<O2zVfId~_OKgjP+yHOvhU()oDOdv@;2
zA7solAM7CaaPAmsGFB~q&@>r)@T7<RW~<uSy6sb%$x{f`u>DccP;V+1sVPz38BPO&
zRkc&A5>Vz~JK-MZCD27byLHfCTwDx&A2>q=Y+#;K{G96oZp`jlmt|d(A5qnRpejFW
zv;yL&0wSw~$N9}~f$xWbL^KnCH;n1&)~?wG-lfYjy<0I1<qbY_utV8cvTLgRET1O%
z^C+M;+TR0lj7|s59ONOU$m_^}-?CTo_~Vc3vds4c?RjihN}083+3VB2UXPdM?%|m4
zFqE(Oz@AWF!N2yLeiYv&`P7#OCSm&vo;NgBn&F~*YlpgaL05@+$t0?NLDQofh%8+B
znrZ9IGyn{ft!u}UwS1>bOehUV+aUG+-X{eQCBuckqW`lzCqFmPsv*{6zUFGi$JNaI
zF&TKA1hb9D389bBqFWKlJU>nSeqZom=NJxD-7PBFJh-`D(xTg2%JK))R-o8W?`m4o
zoo7qAZg-XRbb2<9z1OZY8$%znul3@1cALr7ZZ`*$nQvW^WlNSX&3ZD4!`bkIiRJDl
z(Zq}-plBSL1a*L9mnZ+XwK@On;>&pm(phI%Oq|%4;^hXj52Z8Yq;SJdIykC3P6W6*
z(=Rs|S0qI>F8BQIL_@0eLQ9mXraS!EM!p>i$Rnzwx>R&A^1LE}#zYt4A(<<L7W5-+
zqtKiUTdeKzaZdgOzG(}GUWv{S+PHDMt%fTa-mof(F*!ov*l#FmIJF4SkVaQiO>;XP
zPkb~e<x4%0Mw+&^WiS?tWL4Xg1l`q#ji~<gm^dH+$a)D3(=<#?leLL@VwxI5wFF~6
zh3UQ$+>jI?#85+>V89^hMX69SHBHjuI8hlVJVYgP6iStV>F>Dmi0xmSMFC41l?l<g
z@8xD3lgP>hMdOK9chd@Bz>)%zD(O_hvPOZp%=5w-%O)UU-$CwbsEPy<28k#{ZhjN?
z!bhzbvrV_Qh2&+1uRbooLsy$Wg5_~0t|!~gxT#*hbge0ajd1OR%s8qqT|XgLrf$9&
z-f90(^tS7aN8|)P;IF8?jn4*_`ZLdz^{oL6(V>}%f>!MP;ipbrn87)pm+zw$$a5SK
z<JH%P>t43rU}Iwf=xwJ|j1m@j4;xKhxOh`X>KMfrWif`2q7&#ex|TjK)7b;t6S3dd
znZYF~=Lx=R=2wftcHAauk~I)byu|JzJ+Cf$Nxt0FHZ2Uv^!zjlLv7nsKGeNMr$oQZ
zZM|ud%aQ5cw#$meBC}pW2qB_hEEX*WZ>Ers-jn(rJ`v#Hx#t2Ld}5h>k5`*s$S0l)
z0mA1#@sIQKy;7-E>dnu?tKbuZc%tN#vpCpJxpJ{(OWlyq3lOm&P#BP^l1krIvbo5W
z^`gsBnCh)(3qLpgs%dPhOG%<LR<HgNr#`pnP#!ZJ4F((C!%VNkv=TNl-F)LML@}sd
zBnZjlv{!Ioh!S)ex*I)!P)MG{#U=Hfxapd+nObcnBrDNF^M4X86y!$L4ZK6ez`?R)
z3Q$(Flymaclazc(7eYTn0Vv~!emC5RZg55c=$R)d6$76VaU__|YQb#Z)q9phq{he@
z;v#@+w|ctkO-@dFUZCrJ#toeU&<U8oRVT(jn4ptxFk$`oOh3n(YY9J<mkS^Qdk{^Y
z#<~!$19+;@z=n;<$viY}KbY~@JanagJGG6=x-zS<eabzF^0wbq!z7Y2VGqgt0^Fi2
z&BF!lBZ5$$TbNpq{p&)W=XsOxo}V8Z|KKNKeh<Ar{wG1{VauI=Eb?+Cxj1if5-Et_
zCjMGV&`xv;-G&~$5(^<5KwX9RZ)eRKp2<-2Tw(7&y>I`%LlN4svWwAY+Uidjo{S56
zK>662tC210#04nL%lo&pW<~kv_z@0C3mQQjocTN(Y;KA^S?PsYIGR<R$x)Pzd;dAs
z;8`?{PN7@S>!NC)WH9*cAvaqt)Jc~TSj+@6gcPW<NhRF~+KtdesgaXP*7Z}EQ+@;(
z@6W<Fh1se!jjE4^X?58jwPlR|tPGga1>hkP^*kYjOlOM22B)d-{EcOnU4-{j#j5NF
z@KPQ<*}kQZ4_?U&49@!cr-pLmBmar+rj#;+)3Lqf3BQFrno1DTJP#qU-6v+`KIX(@
z8=Yg+GLE``MZ3PKcH4IH@yWs2gbxQ@>`b?YIw^e9$Y`s2+3TUTOp)idBBxfKn<=l?
zw>R>6@Kfr3zYAlH!25MeAJcfr4_%MeQm9(`QhwLQpPgNqkvhvV$7Ch*Vf|5%&o?H@
zd)f5Cap0D17{Rk>hd%~LYAhvPdS^gVtp@(TL0jm=X2y0QI;WX+L(8a_!ZwdF8EZb<
zlf`)(7_$kka!h#_-cZRkrvYLz#jtEy-uUljo8JU{Shg&WPsU&fuPj997>2||ohvhz
z>0MwI%I#(=KVMdBZRRm1e^|p_Nvr-dzRvUjjEAH=N=qI#tJb?6=^s4zeUyjgP55se
zOyD?IQhhF!*|(Z$UB~K=Pj)?Y3DJGNbRE^*EQ7fBd<N;}*sj;dZ`Exe@R^;|Pc4IZ
z`8-TKrRw;tXK@7X`B1=we$}D~Bk1e>mN79|4=V7P>(t=%*YnT6XU%oGFI`_22LFg8
z-}ePPQkcc3c+mVH08jCk6o%BRXIpg~$5(~EPany@g}VL-cmguy(ZZ5}s|pw9g=vcX
z)lL;`Y`aJ6A$xv#a?`E9<x{-?E^`OGf9Vwaewuv_Lm3^US-RaXdL8m1XNo^eReT3D
z#xv8`cvy+MKXHhhBpOZ#QVuub6~YjAie#85t7ckR!%y_R2)7gL-KoivvTK(j$=aPB
zCYpBj)f&;({Rxe1s(qK|ch^mRef{v#d#+jgtgc&J$y3=3GVE7BuD_Q+!t<%zeWJ~_
zF_8pGvzi}o?p|zH%5!`1ow|0HrYKxkM;23@%TMZ9;#7y>#9w?tcqi3$knm!9ba8Sr
zRJo2pM<eF#t*1H1^;9$LeM7&d%#`9>G?A(#(klvAy5n&q(lttfzwMwGUopX`OSdUp
z7?9~jYkl^9Fvs7=bLgpYGC)nG<|QFSiYa&_Z231!NW2dXqp7Y=HTC?GKph{L#w4Ct
zz~om#!;9E!N1ZBOSPYn^OG+@CpxkM|KzwMH^Z@+AdLanhGtK>lI>ptWKaCjXgK0vi
zv3t`oNiu&#cuF!rum3)o&j$v5+f-tx04Ihs^b?IE8Kl^lO$<l>KNG<Ge`;;_wtruY
zqlTK73}t8xhW$|HV{l*`#Af{F@9}GC(_|}m{H3%w^s(k>l*Y%e=)FTneth;K#!&_u
z@l-weGtxtM@3a5saX*dgsKy++i}FNaK^ej}Atp$FXk^<M>7H;<!M+)ds%F?u7mIQL
zMe><=I_f16DTF&@<NH5&(IHpS{e@CV!rIXJM#HDB$IU{%CgbbNb-mhC7-QMsPACj1
zR2(~00It)O357AE4|SBA*F7(PfND2Ta?FKui{DEFPjx0Imqb8+)q*hbICYDb1yQv%
zQ7RdSbX~V~29_PZIuHW%8@Q^-y`~v;e@+Oesua=q&S&p7Z%Gg8SdTxl6@po>WZZ`}
znmUP)=m1c|k?(%BhV_vk>DF@{S`gzwgXQ`LCc>Fogx>%|M3JXYnnuuQBK`yKCZYq>
zHEm##BX_saCZF=cpe@K#LhVH3UR)o<HM`>$5jC!NI_VO~Xy4cE>fEu3|NM?S?&$2o
zn_I}v{YaA)PBZ4}R#@6p2vKK#zOL)_`T5RxGn~?3sOy$2n@o?QUSIbT8_j#EAuY~q
zYs0b`$60?b`%rb=Ss`X!w+gS$=kunn%V%U=w-8>2o|>C3*|0<=!X!)tsAAGVfrQzs
zfk5DzKJ0s4?|=XM->(z_6_xi(k|dR%eDX;ujfvPIWx##7;>+jHpD$osIFA8}XV0D;
z?`*vGtU8Cp5JQ5AZ=yVk@*RUGVjrgga7f*Z%)^wQs_{h%_~d9ktv5k8j??F9x|J^Z
zkIl`^Kvwgne(aW4<zUk=?&<Y<kt6sC80lU>30UX>j(Z7wt>5pnTuwFB=c3_Uj`dab
z&!R}6NB;;4ifZAl@z!7PkJdf^gd}yMRF(BGB~@Wvjre{ADNCGTG~m>*mdX=pGnOh5
zcmqmMkzR2EJ<i8JGXcz%JV>r?D3YZ0h^k<nD-?=u<%5FA#s^*F!<=)@9rV!nXPm3#
z{TpJw&2}8??IsN6n~7x+KS@io4c8%>CRrfKWbbs~<>x0tzDqU)*{O*YdF!^D<?$0G
zKTL<ml-|)+)%H7(z6H=rd8Uq*(Qfa?(kM#4Cc6$ZeyUcpXRi@a`Ow++!WR$`A5g%x
z<m_*DL-LZ*^06!TU<4g%fuDA>IOZ4`bq#6FjVjFA7HSodqqqLu>}`r+C_UA(DLq1|
zZK*xSb={^5k%MO~2V)0gqX5`3HD<X)wH(=|swGp(ITyloS-k3C>?~midc}`Xw}wTw
z)%|CN76V`#HdQUpvs7wtY@3;xnK^(R89NSECdvSeS*FHhg=?T%HpQ~ah2UI(m&cBS
zWe597UYF;uRI88fL2rz}I$KG!y`6@s0U*RmlI%>gq1Fm{8}I<S8Ap*yAhvZOH<YWR
zfvoBh-gaHcmW@I0o5q9aZ?<Ow27|U>b36o=@(R;2-n^fJCam2jy6<p&(C?cWLPmw8
zMi)PhSPGOPt!4eU&-lUcDMc5E6KJ3;_ORH+gfJ<YL;OHeIEDnb>UivPl~~U}^gaLR
zfDzNtRSm-F$^wz=^y~HS_`<kjkL-W=o^)Yf^;kbBle<m2mCD3KrQ%MK|8jSrEV2{X
z3F}3DREv;hw3{8h-Tm9tuER<n`5So7CdTkF6q7Q2lUHys&rAV{0o0-!S&Wi&GEKXo
z%t6L2wsq1G&WY+$M%KjpA`7}Ci<kekPW4<qm*dZeJ}R)}x<OzH34gnV)#5~4uLVJ^
z9#0e%VR*KECJfKmwr2<hZg3?m%ph=GEPcC~^_)%_%jNPpo$5T7%fn5zVu=#Vu6^6n
zwraQMb0ODtn*a8yefMC}G%WzjY$gXy3p_)}fvN_wFg&o#`*@hk=c%rL`zO7YmWs8S
zWs`<&5&S{@A}wC4iY8I2Whm$qjn7LLRCx?Z9?P=OtsRM*5phnt`~TzzBbRZ`+)?nV
z>5cqP|KrA^;cz&FuV~tS&9eM|_bp4?uW9><{dZ-@uE(PB+qZ4o_SJ1)^*KlT3%Ccr
zg_uQ?b5Ejc(9P&>JAGtA2f{O7qHYCzXe>gfSaFajp1_iPf?Gr(8F6oYp1f5v?D!oY
zBoZ1h%@#o$Zu?u2dIaLjF1zgV%P*6pU%|Lv$587B;oYwa!hQs}-p3f%`<~aYV?ZZQ
zo;>+4d;I&Ad-CMZCF!!uFTd=T<HwJmRN(F9ayg310JH1s>o8$9?~ko%T~AY8pEf7%
zZ<-UZ-ZUpPZNhBcSFKjhMpsW(DwS%r0`R-%&!3O_6Tb^ksa7kWsMqV~qiYn5B!sqp
z3nSQXay5Fgtxyt>b2+dq)b(<QGoJ>D(;ch2Na)bgqn3UP#(Ygfh?&Xs8e)hU<L5Zw
z3qXTjO7<J&rKNIG6b6$*Ex5{*z$iC$+t%$Fzq*JxTZ|hdZw7o|w%vVq1*J-kND?1E
zT`rdmkj$$BEtEJH#kOK9?<+7>EP}ZRPtO9(?y?Bg%`={qlP!~4I|P6zu?ZVHgqMDw
zND`Y2i*BIsrq$h-8&N8#mT+#6s%_hvWh$ZUUu80?nI|<`hn*ja#i>%bCXb<ghzxWH
zUG1miBANX*G-teAkSLKb0F_PvLuIvPxV6$O7*12G=VsniC={Ss(1I@n+Vyc9>-k*u
z3)Ng+j~#!;*k~q>UN{nzGUnJIUxdo|GH~?h(W6H_0B`N)n{SRU<AR%sWqBJ2$AWl#
z{teHv$TedBLiE-?-@Wr6N7tac&|`RAjaJ%fL{UVmrRRu#F=vRoj7D@&=J1@Uq!AwY
z;)~r)(=BUXeG275BSC3746}!50Qu31^)M_4SLL;S;R{~~w8HqO1ueiz4p>P^LYSXy
zgl)T&Q_Y)8wmq3+E@A-Q(W9Py=z+FS40L_bo+}BCJdnwnCzCdBz)z9iAc;Qs{Q8=4
zo&50K?|!$6{rq8q{n}9l2Ttkv>OAjb<!H^v<Z#}{>bn(9Q{XGlKKm@8K0ZvSkFQp#
zKmTMdi+!qIjeSZEW1o^|Rr(%2)>$7-qg``#34~KlL#;--`o(}~6QC!nK%m^!d%y_y
zc3*$}?jC+VmPO&>C1G^%rkf6q?|(S!24H<Q^a$?f)SH-d3m)i_TO5`H$UXjaj5()Q
z+S9R7AX_?buZAYkUUUrIi0(pf>0c=fHOR97<R{<6QWJf^oH7&RujD~Tw%^EP+X?w8
zvPg&`Xe%JCi+RcNir>(|grk^Oj<kxwTRe0riXvAD0B?+#3!Y$E{W*G~T7fOgS_4yX
z05?zx*US#^*htr1Qh1ll7q^>U5TNawtu+&Zo6*-=lMYum2aW?XeUr-K$l*=3Uhg^)
ze7tozf{Qm6CgBh5lT@evE@o+>?E7@GZU~3>1MeNhVzEjrTdfwgz>{05Ws_QsShm`d
zy+8v9yzjN~*4eXXuQmb9tIv9}77Q)0uD{*{u!eyqd&l$HICu}~DHH=g+p)Zv^Ai&j
z6Q*nF3k%G4&FPHC@n-&KX=!Q6W~RG$uWK^<1L?7Xja;8yKWDwmoxF%z1hZU~=?C`j
z-yhjbyc*3?zlipC6CCI4?BOhV8T=rM0$;$<J(n*8_=2W!u4&78i-PminGY#fes`th
z2!gG_79Ij-8VJ2-xyyXvhrYu;qbX+;S=P@yK^0kUDU>QkOS#pA?K&r}&qoHDMm0@a
z(=`7eLFtklpH6+~uWu+-to({OsR^O3W|_gXGxK5S<|Ip522;Wef3lna*QWxhFP5n!
zCvqf*HOSh6!*Jz{Cx+4tr6-*~K~efMc1F{7X+qP^oLRb&3GlwjZgi&Ie#Lh-tl5CR
zT2v8%7DYM%0=two8-#}ei6n#S_y!1QFWGq-ic7X_+a-glEHG{U-w5O?HO}d}P9|z5
zfLWU$dYdy7z+_yfnnv||rutVlhqs%sRc=tj4+@3AH>knYs-h2cMIv#k{;rIP6jR;F
z=N+9gt{I$i_N?@_3->w82a552Np&Jj!X%=XKx5!|>dDPgGJ>@H`uvx<xw$z>W&li<
zo)Do-Der!L@2XTPy$S|)fdM`yy)1Rx%0FD|F>-x|f;$OyqLrn`v$s~O)vC95Nxr0d
zt2K#*S&wSwFRImQ)fv5QEg@{X?l4}O#6q!`kVbO-PtT$iv}zT1-H35V>Q2xh(W0cA
zGzm8)O`^hoC2AO$F`6hMk?<eMQj%%FI3baY2oy?&_N<Uh<2Xn(%B>%0!uer<SGIQV
zZms0X<=o0OOBgR1mKW^3=9;~MXBkTvFA00zW#(5_R)koI0qXNBD{p*Lt=+p@f3UK$
z5(C65LM(RAeW~I)Kz+q9RyHbM#Q%E)*q1sS|G~=2%1#zK-HRUWbp2&*Ho1Yv?cf9k
zziOk2!Zm=#Z@P^@%l|@BRbxl93FMfkpxM~zsm5)WFD_z9lACyO@p6`b%ZCQzvYRgM
zB?UdPcQJLl6)&L5`MrD2Mw#uRvwQaDWmPp`C;q`NU<|(<LF}hGP$sq`brE;3{=3Lh
z25C2hf46W?GK6?wzva0gKc`WhfA5S97K|JVzQVk5-_(ej9x(~^uTbSLGDfw}&r;Xb
z0Z|4^h<<Cm?YYs3mtHJk9|m$-lF_?hm${IlDa1u2?WtmqF6xp_+Dwkt>vbQmoaZiE
zQCXFyb#+%uw3lrX+kNVkEb1)G$Da={G0(^~?ZE-uTd&uds@Qi07=+-}O_1V5&6<WK
zDP#ubAkJ<VdtThZ!3CWda%6JPbpeJ;Nk%a2_huY6Rry2V+B0NhV0lIu)?VWeiFLOE
z0`N7P(kH8#!SCrS<)ELexXw4@gpG@I&V4rSx=7E0iEu(oQ(osDl&v>Mm`Kd;X5DPm
z>j7m7I5NK!QZ!O$b;RKp!T44D(~?X<`cfP)=xHh&^IwDBhQ5T5pk#4d?k=zATVw@8
zr%9J~eZ5;Yn@Af3V`MLV^SK0*Ku!w8IdH#lwykIsRbop#H=lPs`tu{UKc82cfu{Qs
zvGo+aO*ZQbW?~Pob<b7ys(JVPsQRNQ^v&`{IAE!1^~&FD_qFv81%}aj&!ZBQ_BQN3
zPw|R`Sb{S?;Dz3&LB4`1S&!x5eQX+p(B}d;a$V6-V_PPR*DaHI+pa54ZvK5(RTm1d
z=2T%pRhO!nv|0h-W@o4G5-GJFU5^Hr{%$+jJ4v$C(Kfs<OL-DI1!7e+He*qw7Ub+J
zue5WenT18ify>=e;by$#UQVVloumO3gAR%L1#{IS9j7QubM6*r!JU)jUy2B?KZ#I}
zMqWO#4kv-Wip!xPYC%0A!b}C_0Aji?>=YDfbAb4FRF8Nc!1pZ6k_nOQc|zuGiD;5F
zuL7v^uh$Jj$MuGMx=Xo4>Dr^Sm2!bTNt!F3s5qDgZ<(8$BV^8&G$Pq^1c1zK?*ev*
zQi*Fio-}HbWrL@$f^ubc3lvSm%-g{^{@prg4V_1*k%Wmcl+kH3ku+l>Suy8qfX2vY
z(i5MiN(dzckp3p<Cy9N4AGry9<j|o*8d0$m77C$*RiYg#$~y^`q+-;{R83Q}R#cQE
zOm@n}k3e_*$8AE0mLa>UwZl?f*=U_n6h)D>gQ|K^lNDvE)08!$?M>WyMVWV#y&BQv
zWNGRdW*A?IgOtY1TDAbJWsAiKA&OvY6SiOj#i)(;p)2i1v*ms0dCG!J-bi%?jCax|
zJwdJ)Dh3K={=BtieX1*5D}~?0CsV#pn_(CVx4Lu36<6%oS#?Dya+mvX{5bfR=Y(4C
zT=RObs@vP@Gl1wDUt+TOW&i%<<YY0Nh-w(uqKU9L>8nL8$cn`*(2A;m{TutD{$zfA
z<LWDB>6!i1YdOT*R1fvI(tEu|tz?z}JPuj)lXY5`rVR=5&%SoMU%^@DWOmw)drD7T
z|7f6SsOc*ZavcLfY0+0SrEval3Mx_j;)2@w^VBe#{ixS#yT*IH7~{C?_Rd^9xubw_
zVaH@V*V&%KS-_#FM3Im!LX(8rH(*0l2g_UR3DHs8ta`R;S!2ygvZKrKgmoMq`gjD^
zXPx9TGuw+qVy7d>tgN?NcHG<;8*f=Onsc!sN&DIqy$7N!vB8Y{_z|h2LzmQo|A{cs
z`e;<x5Sn<42|Xni%%vt@|HbqrFad?807*#fOb2Oyj0Is`mU<>((`e<^n0pe3Pph@v
z^w^&e8WZ|QeE%zTB>y;KkPT{F(De+Z#+CMBKY`5*&Bp-r3gxa4vJg9PcWM9r%P!l$
ze@QpEPW7E9PF!=%i4!~Z82!Bw(68FbxIroCNh{5=wABoDu)&>I1qo&Qe=e5F<<er5
zs4k}4YY{+H+a5`)Q?!G-amfbSdlHRogI(;zYKvkyz=HjEzLgK6vuGVXj^2ZwM_)z1
zUq0zAMAGUK(#PyVinl&dd1o_SvbEE;*?~3FzH+Du2<&UZIw5kUf}b%p#he<X#V4rO
zw*SuFdP@Ucn8EfXnvy;~rtIsN!Tqwl|CM?hF8%a+VH{OQpO1faGiU`JxX4zzly(#T
z_8`J+)hObzDoF7d>2~<uyE=Rnu4Z_@m!Nd-#MAJ~|N5{0Vu?QzOfkWCIUj6r2-x3W
zAEm-IvsaIA{>C@Hp&ZUM({<oDM|_9YsIKXhNKTHDI+#b$Mg%i$Dz#e8MYWLTbT)I}
zM2ORUq_(@XPpJ2RM<E}sYR~IqL%9Ft3IqG_X(7Z}eK!b7g$-ceHU5-q7>41(sdkNr
zP(VBRxv^juWWX`^#vdVSFM0-IyFb{r<)bKXTS6r5_GF_m*={F7*j7G@@?prRtn%&9
ze?)lCp3SCh2@!_*D9Y0_N9?XfhiL74TzVxr%qIwP{h)ec2mBVuJcChOFd*q&T5$T7
zTW-NwQ+!Xq68bIWa&>aDGXBf3$SpURoUY{_n0rtr6UqZ~55lvz+;WQ+eatNO<<eU!
zlatkQdHmB{ZF(|rEnW;CoO?i-Ao7EA4<LCR+V{fOk&W6lLzZNZ`A0NEl7y%JWL_B5
zupy+zi(l%%shRvobS!aUTjsws;2!tYuYUD>wE_T|W+=DDS74)6jo@-J@js?*a}L0M
zv%GMjaNBLSN!4gxRSm{S7qmfCZDEWU6f5`W@;qvzqv)T&??3nY%B5etxAxK=!e71U
z#X`5W@FJMo&6e7&PAjrS2(CeQ<6ahXByOy+_UK}ou90ym{x#7`y|&-@K|8OolsXtv
zD8w94Lyr`UE2+<VudDUsU_p0kPUmzkms7=Ky;wCTtExPhS-h}HHTU8fCv_cuKv=%5
zDHs4`O;rG}tXZCKYBB~;2$|ciXtr;2SK#5C5dp^6(Sx30#l9*liWkS8BFnFjP}HDq
z8A2@2{mS(8w9X`fVV{!3^l4R{8~mB2>>)+t+Ye~~T$L#=wW)w*m47=42K8sHP>gD+
z96$^5q98EG>)HAfMY3c`k}c_VRk?C~dnJ7~!co&z*h)v>Wy$x2FaycRE|J{9H=IsT
zn?cHsiFD##<y?o9Phj3IhsWzC^!a+djoKbv7`m_Nj3l}c?1Yxy4va6#)qd@;Hr}LE
z?C12j4%F8`Tj1;(%CF$}Wg^*brUh!Cg?KFDy5nT1k{IbJS`64PzS)vyqQH&KPm;6^
zEf7}wve^8K@ymE++tnuipw4%kFag9Bu&|7sO#kq|Uiyo__zN@4>n>wLbnTwK!ur<_
zayA1l(tHjUHcu~8`r-5X86Gf-ERMNYR}qieDK`sbIjdE8Z`wF;+cs8VG^3}%DUGf|
zDDJy`g)x~~k;k27(Mo(Yo9YAUKwv@86Ttx<N(F!5QG$LIWEQ~<Bc)(QAsvB%C*0Ho
zDFmumQxcEL&Bp`PtK%Mb4u0J~ILU^RH2BPbUzW$m6t`dqoV^Cct&FH|fjy*M?LOMe
zFmKEZJLTB_xBURR9zBQ#(V_+9$ta|eWn)P-QUqZX?nTaA$Q4P_CFk5|aPgZ88{JHa
z95NeU!qLTci}DW>;JPP)v&Y&B^>-IFJ~=xZ6iw^Yzm~-~CaIxa>CfRSa&^zWk-G8I
zQ512{rIDhBj5WR9!x?{N&z?Q1>lPLrNWwXlXlA|LZXbr|&O7f+cYyiwl3r8ykl08^
zzTnSNjavNelZ{<h{|%_0w_yD8_U+qU%Do+5ZMo-FOI|N;T3>%?DmaJM%{SkC^Wan&
zOsh7!&=HR~bBhn7^Pj+Z^-SC1z2eT~jnWtd()*lQ%DU~;>S1x$auF<7G!-t&LP`(a
zd_zs9MGk8uRj<*&Hv`cvJ9Zk&1^+RTS?o+SkCNJ1*tz4DA(ma`;`+<&l#2C=L;#L5
z-@+Uw7eiUgft0x+?t@2uPJBDABe{k~hW#c9Gw4b$4*P|?=|1Sywt*iF8!70Ob|kdl
zr?pOib`6&kxHLlOpmJ+FcEQZHD$r^_I-+Lnx!(zd<(Nj0WWbzH3k=h-M1adcUZ2ap
zVmb-ZXIKZ*G%Tmu2S6cI;W*e7Txqhd$ax9%ZIEn8+ydoM!vKY9uXsr89NE`d4%0)(
zkmi-eKvN^vu8>kS6qBT*c^*Qkn*6V$uIj)L%z~_;iHGm7wqdt10pmnlKy~4{KBl5V
z&0GW%<>{RBN8(oW%cfG`*A6zCCyAMI|6;XtZs!lNGm9Mu&pLQ&Rdm(a2++pplj$Kt
zFCWCyjrjSXx)@egrM=BIrGMf&P84&dbsElzG&s6Yi?OtnzMw8-Vu5AzDl93UTRbIi
zg{vqK<z1Id$&7oSjPRNBj?8%enc}ex!rBk=)mV#+|1EFJ=JC3KQ#M8JeXeQ3ErB?5
z-dieojNt7-bHgdviPIUOyi{o<rNtgl;+DmK%=wRB_rvFA&<e_@oqE(NTK1>}ijMpx
z`B2a~6ljEUoXJx9q@8wP<>?gvGSpH(3_cDsYSJbyRR^ARWXZU>kgt?&&$A1)LjGn$
zlAT!u;cw8<op;{pd3SE1r=5k)4=IUCb}a*wShhH4dYrppo)fl3Vv6z`dGx}>!~`J|
z*Ie^GN1?}Z-Burq=yLrq?ND8R1nJ+Ceh{k4U1-lQ)tvhYl|q{+$~oHYEYVRy-EeLz
zt*0spW79zT;VEy{fkf!ksL04QL>~O5CTW&tn^Xmo4=r1t@O^0)&>b8gskt7NIqH9j
zdTuQN;5#Vnl6-$cx2-TwK&8u^^P-Lw1?vUFW_+g8nc>VnV_8<z>J3`0R%G?yA}AV$
zVj(8F&I6&Ug#fsOZ^$<qvxY{gX3RDk`5UmrAt<P-2)M43xKISFDR2q$ijoJ;Y@w>c
zX1pIo(O?usqtR#(MMy?yYZDJ(h%QHWpm%uK*eM+XsS|B)xNGYki7bWGQX=s8Q4?3Z
zAH9;}eHYD=|I3`;iA=K8)1@__TWoGp4XMeH73;=1w|KLlRO0H3l@3a_T`|j#)oj;|
z`Nlj>7X%M?f3Z|5anGftZe)u7lm+8|CYI&pv~xYnBK!1usAa=M0Nd*NzU_q?lq;K+
zlXEUc07c>0#hLbGtrnEal;#%ZX>W#17CadL!_v}{OS#wS<6e-n_c)CGID#=}u3RpZ
zJ_gF=Ijy|o_5HSf2XVB>wbbhaV5gP|5{c7)D;J75GArsAIOR?Qi~<#7xF^EyiwPa-
zk}P@$Oh0Eo8Z)KXEGi6J>)@94&iSAQi;=|*0FAP+e&FVt57a}ZQ2>Km(PE(DU@jNh
zX|GRZXTQIzVkn9%=jKdgMg2PyEkP;De?qB%`Ty<vM>X&<wKQgU7Gp$E7r>>0FlKoM
z(=3YpK&2dGMWLDvU~5!SFy@qw+@klP8aj(m80vO7M?z#6J>C!wD5)n62Tllr7|5A3
zr-Ya|{=<Z5Wi$m6HwXNi&wcK5M7e6hWGOL)HjS<N1%<fq_<I<<*T4St3eFK>lBqT^
zEj7Z*9kN8hk;OkUHUIhNpH~Qu3U?V`Tq#w6sLwI>xAFo9;HO?CKpaybSFktG3Hco2
ze>qk_S3rZI(Kl2t<x;NqAa;Vl^zE#fW}#526bhy}b$Da<<RpkUw{>!IcNyO;`ELZ)
zdueL+Q@pjXSFxqX*}Kr<@jEfc#C#n9pQD=|H||5wi!N~z<y2S!%`%y!&8%TkO9*5c
z8Wtw%s<jS=*H`OEYDb!RKGoT!8dLAu-2Do?!@+yYELCN-rP*ceoJjG{=Tzy`xTuXU
z9x-jey<Yn35@<_wO!LFZh_)gETUL1-Y)Wkkk1YiIM<G6Ru!~oj!D(q>p%l#GuO9}D
zi`e&JW_oWVV#7Ca@AM4#{)EXjz`Aw}!{N&sU_Azibqpb&xYT61W^;ajtL3D=yR3Vx
zZ(^&pf4<qwF_Rgb{^V35t*iF8aFUX}=&+vvGF;ykosYC@k5=r^RcA_M@yTybCduaj
zocTu%_<PDTLBiC#nutc=H}8c<mE;>OU;PoKo~POf5-w^N+kXQ6JINjBjrxKChsMFC
zagI^W(r44_OH%l6fx$i9Zr1h@W&66WNTlj9{2NZpscv`bK$S>}-j6|^q<Wxo`kX6|
zdtmPNW)q-dzO(98<Ctgj{>x{vZKUg@>IGQSLSj|`DhJ+k6(&ohdSI%%b9P2|a8P%W
zXC;?7`poRk?v(jf>+RNYY@stuvK`)qpL_vbiQX^Ap3QPsuVWbnMD$Rjn^{0&on(h_
z|B%U|F9NpxaIQ}WHraLbu=IE_ef3h|*<2vXiFnyxfTS#fx$TMbDJP@hqu%3sitMG^
zhu#)e0!g;N01%XOHAv3uT8KLgc?gqPbZE34f%sMc&QjUxjX9;6xR<F%C4;+H@!w0)
zV^U31+MJrCAfv6grLESNu%4(SXv1=9PKkXFXOi@dptbHWOU0x%iE988;(UJzHO(zo
z>M+b>C@JdR%4o_iuDK5gS<y*t%Ac?xYEu;3QVn={(G2eK5C0QC-BmAbr%Hiz`_#?k
z;=!#Q?&v?YRt~Fk&%|_M2jTLpcSa@zulc9-7woddTlpXGWh*S74@V^xD*i>CR9Vft
zH=<72Tu1#~dSn1`K#sqUjr=utaGY4*EX%!25+tPD)t@F*;DMdcfyNRW7RvXiE^9K;
zmk(~$Prac>C+s-}c_C~&U%nNi(y}<XBRAbbS5ES&Ht4}`quo-cF%^Uog2tQT-}Y@w
z=~7yLAxNF$&9c0w`fe62j?rXpD)}*yylXU@+w)jG8A^`b@5oxr3lZRmOuY_}6^lz9
z4J>_Hbe;6ZAIq2*V-z6^-w*5Y3%1fH%gYf&)-;hiO{u{YC51VaQFpy`#%DjRw_5wJ
zxZ)eXV+2C*H6hlx5S}>bUN4*LGd$@z!D+(G^xnQ1y`0o7U9;^qT~V#&`G75P>JXph
zWz~4xfqDWS!%ox-qQ{?I{dk<SCl;(<58KhFF!acv*R4nUQl)3GmDZ&4;9DjqTT`1t
z>u_#syqOE9rswCUr^4Ka2PX*|Ou^rd(qEXu^E0j~lR=J?9Z^Xf)vkPf2L??c5|57u
z;HwQ4OK}fi{H2wZHH5%>P;HtfltT#BwpYHQpC9S8*CFE_<*UZv$bY`JhQjq-`_g*O
z>*K}t(R0@7`Xa{<8ybaQ^n|7B+9UFNp7Do{wimGxyL$PAKK3}qkG$!#eliD)HZx$!
zcjOSe|HzQ&55s~GhxtI@lpghvTU9Sn<^Y8@s1E>qFlh);g@;_NYzvDZ6hdO*mxu(V
zo0N*pDHFK#ZH{I22<Swihv3X+$mmqCwYU^akTQk7ViN^KpCd`pQETGmRmfA2e?$*B
zv8e%=7NJZi&WgTuFmOD$XaR}+f#h>BK^V^~5;l@sb3P7(zPN!cq`n>jX8ZKbVGfvQ
z+(L~)L7i7^;Wj<577C3JNEM`De(4i7_h7)16dkoY`Th0Oai^g^_m;uuvY_lzr=iET
ze352mhfbC7R!an>j-63+PNY=0+%819XQk)ylMF)8w<TM|3DnkE2A~MV>1t#~Rb~Wb
zTvJ2jsMi{WX)o0l&?d1)DQj?co?6<0OJ=x8jt|BI9&5RQ)!&FmYGVNFlnl>+K1AfW
z7xW+q>q9ax#^OgoG6DY;+aC<jK3mJ=Umvs&8(Q9{L+AKc?9ym78tv`}^@m=k?Ax)A
z`(m*#$wz9Yh6^SVn>tT^-iMZ`cGm7xoYfxIE<K$y_o8>VP9p8H;I8XG=bc(svty1h
zdt$myy1`6zKEj?(POk9#x^Z>+J`a=ezY^+Wcyjz!xM%!U50mR(&9iIqtFWh&|4^|X
zhv(BT0l?QSe`Z1oZ(YL1Za_2$w;dG`9861BNzai_8^Y;H&u1jTnLf{+L?sBQW@rV=
zN~r!iPXv}#&<u?Nrn=CUYPBUz=oGK3pMA$9cAPxz(#8F}<4Bn5BGk{DO0%g%Ou*v^
zonmQCjkiYu7NAyD^vQJ3mUW}s&?(7d(0W?0r7c#mU4yUJ(t6yq-Kf=SXdI@-inuo=
z_mE(f+pLpP)Tl%iYAJ0E2TnQ-nklw2_b11nJ^(yF{xqB%e_At54NhvNsYz(XV6LP$
zH4B`Y>&@O64sW#dUw>~0Cp5M3zizwkXaxO*{05bB-FzaB`r;&(OY(P|wU4$x>VGsY
z?Ziq~>@H<Zc^tGT5kEJ)MRmeTe{gteX4_SoX=*=sVMi;P-rSSVQ|G*N`N63wf7Q&k
z(9~X~nJ?<&!IIk4)UPKG?cCd4SflbXuV}6W+pO;*s_AH{eNdCWEF~;My2%qe;MX)B
z!C-D~-GrzYu|_93dCA;6&o&-`xo<A&MeXN0!r*2V7~&2QGstI!Z2C}~`pHcS!L^(f
z$>xW--aZjBP1jyLLFQ5DXi(*7u-9LQR#v_T)Q`c7=)Lwe&fTkpCmtM#WT~2nRVhlh
z#k0h-0}>3=-7E;*2nS}TmlNZexXp5gq%}QjJ59(kGK4}rQ2lrGRkQ)nA(%^(!v=lT
zBlg1?PECc&8yg#|tE;WJP9^<%&9b!Xb&1yF*7(uGhY!DF^|ebD#`RV!u2U+16O;D%
zCsUX=3W@FhqxZZDP#teR{BR$Fpl+Mb3NW5E?RpTvpFaHX!w<jppDhpsvW@G34(i3@
zGAn@(7Kop!wB6RaEkX1YHnWoXwOF#;oay!kbD44m=J)TP&)_zVV8PsI1jYUPi%OKg
z4uVp)cVYkjg}n-S`tvV2KFSzllcoLp%SoR;AB#S!S>o#brIhX=rN69uQ0o)__{Ow3
zvWAwWkp26KGCVZ;eyMUUR~~;)Bm4Ig{_J<gb#k1E3?P$E9aTrasqg)z$zPLQ=(sha
zIFGZnhZDPn@G?*j?7g+P<Fg?^^+Mfu6CPnDwp|MvH7om?=sA<&NR*&*a?DZx$f)3o
zW9(&kNa6V=I>gy>!p4Isrk|omac%6-DAc|K86YK;=x$dQQYl3u$^xKKF91cxsd{BL
zKvL%e4FKjAE&7&B1U}H}1zZCg!u`hb6zOshsvlL<+)8tMGEEDs{WdeIPOVvXAD-ea
zH~y>WAT?nVM$*m~mvEIloWhgXspq9bl@SQUQSe>76;WP+j)%_dTe-uqEYA4d`>u75
zri*wL!u@3sx%{O_`yM`i8A~gS8<xfI*xd!hQF>Xr9**DU5<j_%Lm0v1Xtz8YXvAcb
zvW4<5O8#lYEtrJEHkrUTn#35=pF4GA9ux@Tea$lyi}Q7U)Q}W&-|t$&R1~6FKlr<s
z72R88xUMj;;a<(}!bGEc_wLyka5bk2L-KoyMl8z|%HI7QAD|s@*Nm7iTv1$Tcb&K<
z0uUPPG;4}jJb+anZ3u7u(QWbFUh!+>vbv|9T{*U4tt`-opSSR<<(OM2O9Nuj3MKN|
zpnLu7xDMao9bX!l(39xj`FD$MW49?;|5$si6XIsZK$ZLnSNpl_yP-uLZ0Yih!pLGB
zDecCJqGb2ZQB0ew)s2%<q(@wDL!tsEBti_KGpVWguDT7LGs4e){<ka)^z8;&t|NY5
zl_a@yE6taR#Tjso_>h=8pxF3?&;d4EpYZ^k!PG?llA|`UyLZRusYIc*r`c@>p+CM^
z`g}<Prd1NhX-6O^5G(Mfe%lobI?kp--WIWgd+97SF8urHnyj9pk?Dzv!m5rXLs%wT
zUPqc|(XN?C08d8Gpu8=37sq7*H*&RQf$hBj+4t&wVVO+Fl5rIFjemicU+-HJZ1a9B
zAssGFrJYb)+=#axM{kniQIVaHnNCN2$r9TNw+~-8n2WflisE(8E8jT@&v6HB^&`!u
zj&yN9H4h?QFCG8csZ*y89#UjWbf2_%u%P>5y7G~UY2}J8wM&<fEKG0vj>3&HckZXY
z_O-7)Yj8;sPWtBnYG8&{#hl7Jw$$#>K7jLR2CYfU==uFh*aHQu@WL#$tU@m<6Yv??
z^l0pA(-%x)&S~AU^wZPN0KRYDci;HeP8J^?i8nKQ%2u_2CdTaW9>chYoYzq(&{X?F
zA990;-Vgl%5B?H;V!a=doYhKg{7z!*0O8mU-VTnZ2VcOcW*-M<!dGQcy-r-8HFid7
zx5JM$BTux-g}{wui)O>4J|pf)G0kfIb!zANUj0>T`c6^X3vtRR#7#jy&%o}kB=;nt
zJrRJ4G2Me`Of-@NHWx%euBBIDpj*33v_n;%P72YOXh1BrXGe3|b*!eeW?e|Hk=F(N
zpmj5G&T^Y!jP6WzO2)IY<+szPM^9YOr_zozq0-BOR(ZiESeexttU^fT;R?dGWf_3B
zr!T%(7cdcn{Zijw@K}DzuNR7t-HC-{J9d0sQzOBAZiHA^INIa5+xE4`Nf!{hjvube
zkN?37vLn9h@ZHzzo`RVjJI=eW*F8luiN4}_aRfuOK(h#(r{zI&Ozh*lraGMIS&Iv-
zds@I4|KJO<z{@#9O2x~%2`m~sBUGYZvHO*NF9i|kVWL}fE>3?2?Sx1mlBWP017y3A
zlzeqE=}j$HeypYmV4}-aOfy{~`KdS&NH>Nn1d~XL#i`l~33@%}*=L`Xq-P&`gZx%q
zfibNMeV~rdHv2>7rfc1gF!e$%PQ8HH?J%FSb#iP%zd&L-N;1fJQ}lq{;6b9aC~Ya^
zeS5zUQ7VSL_br+S4+=_0nWKfwe@1CB)}lQ@(fu7obB~}@^fvphMW;IPi}*|2gAXGO
zg(yKtNy5ZV!_>}f+vYY(lFrURMlzdeaMMn^nR=2}8byy9R0k067{Brz?|8?ISFn_2
z89bJa9|d@P{0Kb0s1Ra|f72u@@bJpY%J`ACOo#y|OwwmyUZfooY*D20LRl7jnF=~#
zL7!6XHBZf{jR)kHDHEErrpvN&ZB~RvIxoNf=b?oTru)jJdwWTS?B4HZ<ZK7`UPi{J
zr9Fck1PyqAHW)sRH7xy{tz$Ps@g)59Gwy1>zmtr|;^1>7ChB_DLO4wzKj7GW?XC*O
zB`SK6U^(R>(2S!9Mx)j1ac5A$gG>w?1gcNL#@zc%46>a^Eq6KFOLR6VfZ=QliguG$
z3&aJ-MSf@;ZKFMQt3%4|^e~wa!xC7PPI*?+_DY<Zsuv9w)7>lxv;I%~9EgUVxxb{Y
z$FDtoa(7!&Z7YOQXxXZmGP|9=mU}L}{AZW%gt}AIf9W~PV&;P$r>=cb=U?&-Aq?M@
zwU_*?OS$*K503w%-|rJP^XgFEGQVt0@w>v(taJKI{U8KNN|qW%<vI?<0Al41B{PRZ
z4TS(SF(6|$mY~L*O-91^%}0GkJf1J%vf}$RC#eMIXy7SjTq<yv=susEnVIC*OL~|t
zz}d~7CZ4Y=Olhm?>3}N!VO2fsD>OK*s*5U9qOpFvZ7aIqhwU<|Y<(5}hTm}GAi5Hd
z!`*dworoz|K`s)gM(I^@;Qac0C@slJnn^5QMKw{%JQSs`g^Tbb3$#)g@&%z%YOxzl
z>kYAZP;7@ifE=$Z<R+`ts=H?2Y|w#y)L`U3S=F@rfSvGcBt-t&@pmzEaHilMH!{G2
zvz}PI?mCwkF1)n?fAYv9kA&#%?q@n5VuTu>RAm)^k}d`C@O7%2OW9d?ltL}Fc0vO>
zj*$|QZnj8KDJI-ua7z(O0(fRxwH&F`-owRzneSO$T_ujs^plkPj<!mKNx=7gfBG66
zEI;IDEC;y7C?<o<U2or8H#RnWP8`>DHSTY4Fz%S1o}TUpu>!a?*vLhhGN%g~%Zr>$
zp%s66fYqwwq-wN)r7^lxm;$D`W<BY4T$7stl?wHY*e;uEW{}%6AOa|Lsw^!HkpLEi
z9w~;6#zcS##z??MZx%}34+x_4{Y2>ujK*xhc!2RJ4)7J9UcMS&NSZ4F{%fQAB~9{!
zn&uBj2qACl=UYF*Pr~zlgw^#J*fo2!Z+v&%i(ZSKLhnK!L7zZhL?iTj=#SA~^Ji2V
z02MmYCq!ia3EeDSzkB;H-+1MPY{nYH4P@AbPaUpTvc5mkKkR1G?GaeC8_BqDlI-p;
zFA11x+?x@IH}&)lbnWfIr?%Fqzj|Q5x?kL6`mv{e8Q&_)zKWZL-m0l@@6=acr;QzT
z_*h)}8)!RvHF^_zKl)AdyV7E7mFWs{N5Zh?EC5gX9?hf~x6-Awp2o7-dAdX;nNmig
z9F_GulF|i=NR0c3wkXaux<~|J((E$Wg$Su}W;3o6y7RD&HMUPuJ=?bp&W>XMB$Znx
zCrH<B+kRkcN_<Jo$C56_|Dh<5RDm+R(1szrvP)a58i75BCpmQ>ZP8f$jBY5|pnT&~
zpN!vYH!so(IQsZ<g5YI2qiEWGO|+t=fMsG^`FwpV*N{$3gaK0|3?M6B`7dWm$|oH&
zR~4?8S3Q&6ErF=oQcbU=gi_ng?0!Sn0}*s!{5QK44|nbyN(I#o_%+d^^uZ-l0mcCo
z6iSPLqlFydC%qg!8@a=*J6hC^q66+#AS<vCnNt&#2BI!i0|j;DPe%ab1o4X;W9P{o
zNPNL<23){M-&D@X6M1p`Y5PH8m4;s_V4~4o)v_XyCGm@@JfuNa3Aq$T^Cg*h&C7YG
zMg8WJmQ$U&C4_?#FFoMX#)kemtqIk>b5MW`{baF9yNwATbDR&LhdOtaUAOM=_!y}h
zZbKd|?WZPK{bjpyetd326znGTE9cAo8u9+qGZBsmnUOgkca)aRu!PaBkV#OViag`S
z|Cw7_@!=rpv(~AOUTLTaEnP~QG5maTa#E%fen}qgfRrcvx&OA=YE{9yE(}cr+DZ@H
z=reORL{(yc)T-Cebpr<~?I~MAsemUZCnsrV@=N+^m}I$8_We1ON;8J00oHY6|Ln}u
zzU)xTGu2i#N-~xh!c+5}D-gT7tvxsJ<3H$ZwSOID=ny(fw*pn~Dp>-lG>LaAfK`L~
z1Yy|dbesh^)$pVdCe@Otg*dB2XC6gi7Z-1Xum(}U%K05L8G?B~WtmRHVTS}K+M3}g
zZ1;aI$6Gx9!+bs;7#P0)OPnth3dzj$o(9>6dX7QF?mv9d;lqKrf4mrfSAu$$u*ln&
zd6Q*g@Yr#2?Dg|G*oO9?UzoH_@UJCAC6IPAsQ=Kv&s}V{(gwxfbc_8wOEXPp<~h&*
z*z+wGhf}58_}z<=^6`1e%J|3d@oz<4mddrIrNuqD?~i}38BjT`>n8ruKkKVE9B*AW
zS~cQko(ao+>NY_iFeQcAwzkyHg%Nmu18xuQ%VE507sk2q8=-HOBj^0`1ydRidp(>}
zm8YIk)ExFJhThWcVQvkbMCZ{f;?bT`8<mU;nz-%|0YqmudsGLdUh3Y4uOS)C)n>0n
zc!a*BK~n@>aVAc4dhf>f{tJM8{eIuTKy}L$HyOIdVtfA`440Oca^;e1%TjQK<Cb%)
zKdwR~({{K3FyA=-g6STw`dU;r%MjxG*Ko&)6DM@jGz={1muZYME4ue@?2*aINk^7*
zxnj}Xr&+DdQv%@ky$RJZ7N+Zp*PMhn-~Xp&bL6zGSt|;zqtH!%L@{K%lht`|d+k@(
zlBCVJHyVvbA2MB6e0b!F-?sS-<80kn{<lo`<+r`<ZEtJc%h<iGOYEI)^m4uSjF%(#
zo$-POnpN{>&@$SOE~ny^rDhGiu!Q{~1W$G=9Qj5E{-;G0B|4YdesgoZ*AuhSLWBYr
zqLQ=|CBoJ^cdG6XW_(BIpf=w0LLR9+=vQ6hb*6~*-ueXFGYCrr-wy2!dM4%t^cDWk
zS)FsU&TT2^+@^;$tev1KG#Cs#|JIW^ok8@^<J&@~O1VO(yf@*t(<L5Sur9D^_A^9v
zykx?%&1x<$UxMK8Z18rwT%PDnl*@K;qA@emm?)0#Zmph+rhL!yr(Q)pUfRH8>pEzI
zn`=QlF%iSJBI0u@*LRif`Qymi*H3j}pZ*r=fB4OxS?=afZxjzN3j3oUZDgYcs_;0a
zcpNfrSbIX0%`)E>Hns*@l}n%PVEcujUT5Zh8X?bf;igqbLATtZUygcUn!$Zm?SG8^
zT^-gB#9q8%v%FAAMih1Vt1~najsGdQDK8l40Qw)DTTUO+n4bd2Uitu|id7i->828z
z;=F}0waqoXoPZ8y(~1}$4*{JFJ<dr48sbFTu4)jPHLfn~VsFDXFci8Fwg?R-2|ENb
zQ-gD&hr!UB@cR<VKQ?3}>RYHk-TC<MxQ_p>&;Ac#pZ!U~tM<R+y7mP0Th_ksb-A1q
z=p_s*pP@*Yh&%-j#h8-j$g*&>35?h8-VThu3$Mc^26%-m<LjcQkBtWlA7d9WO$Pmp
zp^ecq--#YY??fL%UxLl7g=zZ?3ij0IGto?%ggBk=OVczv@1}{)@L}n!`6;sfdrUV?
z-L|mggAleobjRv0rKnc1DpND1Qez?W#z7(5hidHb!b0?8f7oH_j9HbaRs!1=Vkney
z%a^ca+pKnMX8c<#`{SIZzb7^n=!x&ZQok!dkLn6h)+em9>lyhRi}9y!($R!%AskE(
zk2^x8BUuQ`f%k|zVnGN9{yv?w)(5W2rV5V6<($8iFoTL^m|;}&QFECY(d2m0gIK5)
zCjjkRg>et=;UJraRS8TXQvGY<nggmCLg3dvdvHg6cKk1vWzCkXXeUBBZJR93TK)UU
zJUS7nkIXsq6FSJ4C1FA%77{B`L;)$5Q-MY*<}{IUXBf4fsUj*B<BRRym|57}2m<m-
z&mjmJSlQWVjHr^=&}i&bFvbcsXl1tys^IkDr5w7IUR!Y7cvN2`b8ecP%kjc~g6kCI
zE1x}<YbFb1N_JaFnz>>HirwtsSEF%Ua(f4}ZV@WQCMBj;qzt~WsF{RL?=KF#bNi<Q
z&}pGKHI);JEGr^6HB~H7x*^-Sv?Ysf*F7)P1A7PoIE%$P7?lO}AYzzAo*gCMPI{-P
z|8xot#)UUALJ{)i?so5JhK5wPV_EO9ap4K36SN1Nv2n66dx^XKzyd7|0cXjBf5;=!
zz*jd9Jv*CJ;9^KA6ZZ5+H>t*Yvhf<!3xs=?%g^gD48XK7oM|h%rw&ue;H}u4pGe8|
zV_vPF_R5+<${;1ZD}$@m+X*rt>6f#&w^RLa;5%B6qW7VX;!KD#9Kut=(|lB>Q20iJ
z0xUog_y(9qA*7!2X__S)x%}$0Hv+DtaOsy`#6f_IjsuElc&`Xt`A`tx<=<DlZmDJ1
zPZG-tZU<S7&8DxLzwlLHIZqnTCKo$`Z@>tBL$fGqyo<B@gI-fj(7RXs`v*jQ;q3wf
zOCL{C|Gx5&8{{6dogg}bEQhM;kekZmlxig7xdAgI4%dmoh>#^hSGZ8kO+jhEBmNep
zI~~pUWhKWs*HlU3`M{I3VE>LC`vW5Tex6H`s&US9itPKEvojUKww3n{Z3TTm`1t2(
z->#hm)O>)7DTHmO2jhL7qy?@}rS7OJYeLlmPjcc3SMmZ)6++nzQ-vF7lD98Dn7TY7
zs-#@9rJpge_a_bVPh0Hd5FL75)e+vw8I9Wi=W05&=)VFemS$&Cwhl|eaAIPrUi{}i
zp2h$2*QMIj#6%dvwJb?zW=ll~{-uM){&V`;1|#@7YK}nx(O=4;0TLkg%ksw5Vuc?Q
zRW}2r50S71x_HD*ddUR1hVGZ!>ZCpl^-8r^!r&$tGX8f93?WxKmyi<>#=}dj$ie`i
z8s=CC%&|tH1tTy;GdZ+glt<x)C>4YCOM+6%kc=m&9_R8h-o}fQ^SM~3PZ*N%Mhc)k
zAppb^8UTIcGXQy&=)>MvLC6RurbCSl%|-mMpmWZ2z&C9>Ba7--9dk&WQc9@~&Q8y6
z+qP`)>MpaIuYX71-AkvKs#^DN;$UE@Dx01aI<G&@{1g|0r`t#sS+|0@TI555jZW5n
zigucrt}U;kX7$#dq2I)cP(lJiFgF@~aw|fIlN~C6kCX}6qdvg5uy9vHCNj!wFDQQ$
z`(Q^cAC{4bh!E8cmcf>cUur`h)B}{C)oH#1J7r;s)1N-2#`T1&g--LY`0ZC|@B!_r
zI<~()kry6a_Ay1>fEC{_*Dr9f9ZOz6^;+?HP17D&K2k4iz@T#hoLrsp`|>my0J@k*
z7gg$E5;v++?kb7H=#0mD#)?LVK)sd2=cvKx0X=@PJlF%0(O}^6NP3~@ttUg2Sd}Rh
z(TqY<6uGy97Ge^McB5MB+~O>54wnLD$=>kLYhLpj#$KZfq3eLg7pdz)=sy)(HE4FI
zE5Yg4yyi8`ve;{MT^G9`;m@~A=(?^`p2=N=)uBq!@qIR2<1%adi0wSIPO;Tl($h~5
z@2BdGCyu{Mt!)arWZv;P_63htVy`F5!-fY-s{Xl{i7r2?zk+Z{)3gNgv-*GogVETF
zuJlYQ`CLonwnG1_yvEPcW8hRYN8<Dw84M&%e5O><<Ou16TGy)0PBVKu7*tgxncfkV
z%PYR%Pu46JY-`{URbZW8S<|u$#d(nk%f%xV1;TLR_=O2dC3oN+R&(LhRG3pSu;&=S
zOF)RmMfKxqj#I2S;Ca;vUva%6L$n==o~!s1RnLQvT|CmX8-;?p^+kfsCOh3JOS)xa
zY+Jf`D$@5N{c7iS&?@S;J_out9SB}>7g*ec>jr#xm*gDUr0)@3IJO!0)sDsYKiYpU
zhX35}_lZix#=s;h?P-i@_A)UH^1BB`x)M)7(p(AO-O6|*8giE*36^)_j*X2C&J8dO
zzD8+q1!L?25KZyICYWrrIlf4e7h$VCb~;5T(6v@((@EdDCjZW<FzwYWi+>QKLT4dJ
zRx54SWyA?d!++nMIS_Y6;_JL$cX&VNyuPc>gKt|TmUUe`5Ln+1{)bYkk`tjMhhRqE
zGfgw#^<8xqnvOH+7V=(eV`GCcU4PO131e)H7yoU`f6&l%X+pKW9X=O^p@PH9h)N@Y
z;SnK3xazeDuaNf=*L`kdV}nx8HYjCWH=`x61ur2RP4MgO67Ic!298SLjx<eQm+siq
zp?~;?e>jph-}=Yb!Kc=^Znu#4CLQO{KmOxCn!y7?*X5X)|M)s2Yc0;XF7gOg>1;Jk
z$r?n7Ne{-_i(!B!D`W!sx4JM`PP_K&^4pV>la8|lR^MK73wiGkZTlA67UGtmTn-Go
zZ-M>EBe&!YO0nV}{uKj0H92`0{I(t4z7V$vVcWO({#%Cs`e9$elp6V44*!aw6G1Y9
z{k9L-f-;qp6s0yQ`L|?1xLg6N><Oal<O!w{b2YslvTBmM1>YcKV?;$T3TZOl1i858
zpfW~q)7KCW>k%By27XGXa+sg0<TEDQ(6kM$7MyKO7<EPe(mzvI0$-!A5hyNLN}_EX
z!=IrKPVB5wrttJ45MPAs-8cQqtr6@yPl%C60!0BmPL=(W{OQD@u$NsM>;rLx?hXiD
zv%6nv)zBnbk@E?yM8eR7qmYHZc0ih~ILy?0O_U9Bya&>DJ(o>ckv<VY<aLl=6GCua
zK2$bL%lOyF9((N2W153Zt6)lMi`$Tap!6-*8TJn00wj_&OSBG|Q$AGY{PS;hJ{Sj+
z;?^@di%ZB=H=>!x#K!xHFU<^=fy}8>Xx=zUiQz}#HgK!z@2K0d1ccy|Cr(TjNq=u_
zSgOh<Pn?)!s%ovV5e#Rp+r|b1bdk(N=@<1><mxH#9JqpA{JcA$kZEVaC#`HvQ2q6=
zBP30^A&{=uVuW^J1pCp9meS616vx=HOjjhAwUf;WvJ8HqNn;z06z=e5jDA|v751nt
zFaHS4GbjzxvOl^!dD@liW}#FG?Qg%NZj!un`^O1i+Bwh$MtMvZsA1x&85PH{ubWXy
z$h*o!JGkp2WD{aqKa0j!yWVlf9jd0OcM$BYfQCKV6%P}W8SV9x^XJbi%6X!DSeT}>
zV^1ghoK;V+NhJxBbk?+G|5G0z>u{X)M02m_IR42^H{GPFH|@o`jSjLLo#U*!eB~Qe
z67UBK{CS$)Y!Q*}I3PJ`9o^;NkId9Eoap)d1Yh<X7KH5JJ+ZF6BlOR)uG-2$5=Qip
zGJ7y!T0BA0jANzEK0raLTg~ck^ZGs|V%iLdwdbc)4|p&JW1GLt0oS#ro^S4}e_~oM
zf(IUf!Qb*&Hk^1nE}Gw0LPs)PlRADUXK)<eAO#s6+M&YjGIHV54$uShNgtEiUR9k?
zHBH@Hqs*G#<#65Ozkl+{Cr!XUCAGb(rl}LEy0=Dfyi>QofD!yQszsd3SglrTpW}Kq
zfjl@(UalRxI2_!A<s$CZx+J)+)u*&?xh@E9%JNwZj7MuMg&y>KJ|)7mgOftj{IuR|
z^_i+~e`ne^y?xiyk^h8#r(`kfO3$NGc)vsiawa*Yj~>A08L|ip)G_2W<SY}<xJ7!%
z#v=O~49iMVK87q|^m*X9x&6IWVnolU)khq4`*j_&3^4x~WXB>-2lg7s`Kl#n0D2_d
zkNF^<nc2R5W`?6fa9&}Zpu29_Ho=!+#qpl9+xMJk=&)3rD~WaP6woo&)<uybGj-5^
zci^1Z2EPTR%V=+rs-hQZ7{OF_Rs3-;xT;w!C=t7b9K;d68rd-6Q06ZG{&q^MdXrA&
za?P!+*(bgLr^Bln|Me2RYPR<r|53QeK9kMO%^31G*v_5FbI9)xMTwvirXxZqibrQq
zEek2XkW2C=7t>Z-tRv3~Xa7M*?j`(XPS$nyGr}188PnIBinQ`B6+j&Uj8p*i>1~pt
zDpy~vsPG|Of5sA9NYelb>zPSe?Q<XmEDHz$-dAN=x#0#yMhGE5Xln#F!3fPG)CdSk
z)wnMIr)|?Y@pD^%A5!B`VfwJoti$APQL?|_erL7;w<>&`C>ft!_F2Zj(AN*wnG}V7
zON6;Vf`No%WnbT`M}_~0wC{8umGyi|<A9_H{VPX0NH1qZDSfOB!$$?Tgp6&Tl1cVJ
z<d$=wyADzZ+XR(Nx0zfX+oigZAoR$W=1ge-=Um|MbqD{tW}9lqX4QU}&Xi=9suufI
z7a<5|oHqN`35?NI+Ij@LgH%S)hCG^6GG9*Nfagk%>9molE_R!Bl*`GT5{WwmUGdBT
z!A61LrFJ6aK+jl9&GF&cnQofOua@LoOKD;rKQ#=w4bW+wMRCoz47aLJa252T9}95*
z1RA`l?9s&rEVH%+V!9YDcz(2XJZ66~nM`0<WNHH#nM_AR^w4?NDMc@}Pt4e=|5ATU
zQ~SQKZ#cUwXbp;Tp?^+v1hJFx166iZ&oiG5*H5iIpZ%%2=lDFJkRG%PPT4zhiF|NI
zI%64}tgQID=65sNAk@?Ep;2Tt8cqgIs^1ar2A#Jsw#sD-V~fX@VPP$@vsw~o)4?=5
zs5fQsO>9{hTNd83Ft+qDO#-xL!P3zZf!0QJ3!q(2NH#n!Ku&VH4L3RyM=?IKb=CO)
zb6Z!9?`l?Sc<L5OhTFAja9q56e0=P9lo__&ch^o=Z{Kr2ySux4zwQ2RH}=6Sp4i>p
z)fD0jTdS3&wv{JcK@sX}%vBKZCJN;7bt-5X&TsK@pn}b5f^AP9Mxo}DP<~vIr03P=
z{cL|oe|JRVsDdqn?R~ACvjR8}B51^&ns9ko5BLT8Bg^!@KbOW_zmPt(X|$lNYST4M
zb042<J5EzjUG-_+iX%fmS77xDo5h#GXBS@<b~<a_<(t=3Y-!-6vuSW}-QdFnZvIVI
z2|%Djk%)5W<vnOHNN!OE<$cVq1K?+9uk{4=rfk0r4l<{dEu9;u_dM^OC1k8L<;rsr
z(3{YotM)XY7h&Kt&=z8wvE7SS0d$`Nx;mThZ8im@LFKpco3@shFr^y&x<)BpTHg9q
z|5{(c$4kpw8l{wKTgyx8E=%A19RhpkW<9;V#D1)kpGae?1Y)U%Ed(ojV~Iqf1Rq(4
zOdopLO*e$Q7w7QHrs*?w7!^6(9DOK0EM9KtuEmEdQ3LiRUSZ5P&BIvHo4yalhsA$R
z@+#C^?Ob27p~MRIrO3e=abBPf>Nw!mK6N=?^b}=5G+_M4h6UQyMA6yUmPp|87yp>&
zqb1Zu5n<2&HC``euG8OB3>_NRH!Bk8dJhQvsLj6HP;V&uY!)%|ZMN~<j3vyz#e1z*
z=t5Y0mj|W^WA@#K{Wepd&1QAwhH88_vu&2JVt<Ldn+e3f7#uJ=Q6t)Wr%^73GeF=~
z*HJHhy4{LbGT{;?N}l7EASX&$Cv$4uMAUBO(OEz3cr)P)or|%!)uy&WhLkZKtsf$R
z;x}eq+Z%}Fl~}iXAObKQgNOlx6q{e~UWH0J$0SmAS9&8|m6zCdAZUzC^`JC;O{(x7
z5D&UtR`Q6%I9;L@&w&|FKgWrHb$yLD#6TxZ6M}6f@oH~sZ?aBUY+SY@7$b_zSO;pY
zqIGmbf3$1xVy~08Vv!J$l(J49=VM5OoSQf&CNr_<vJsyn5~0(Mq(U5$uXrR1MdGER
zC1}T5zC?{HDS0b^ieIV2gM)(*#EsNjWKO9!e=t@Q5f=MDw)5CU*#mt2$NH^Z_>+mG
z><KF-i|mKrxah#^_xnG#{rLA+3q}VA2lKs>!xp`CLxk|)5WMnktN&xy@ux*ad2Hv$
z`il^KTPpe@ga_$H`*&=vavgsN&L};bu<gNT>p_C~_!J~P`f5`frVB=P=r?~921_&2
zb}XJ}zW)2XZvr1kX^>v``SkcVqwj^$4!~!-5yeL1oGp%Ft;KxYu2?HECv7z#JU0x3
znI`r~1vV#SW=M3qTw1dCrioE0L378aKF8h9XjvEnT*A1-8((3B8`~Tl0Jd7I3?D8m
zEC{%CQNqH)nQH{K;6}NfgoTkW)$r%@b;%?x?kjc)ZG4V%-sWSA=O41*2x)2EdWh#O
z>@(WAMsd>Zc0~^I17dgjt+;5lS}MJ=yL*MIEgd3khM*lse7}qn4|Iu!rlAhBk#*Vp
zn%vto<gF=KN7;sF`rd%_={tC(5k(wWrZaX-3!wbCqR8^IQKJ$+TKnuLNkXs-mWxR;
z0nWSWsu*;8H$%szwHkGuaJBiEEGx>h-9W4|N{hEGqb|lSC5e9xxi27j6g_9Zo!-B!
zEMQ%|4?x5MApd3V>lCdip!dG}X}u__F)sbuIP(;gO&6yZ|JFG8SB9>J^ta_lZ&uf_
zHHI-7aBf9{5M+fe&Db0WfVSPAhW<4V)Nh8%Ngl-aoeui1`X+3s7hL7rOhx7HDJJ)D
zs$zcIk^@l6UP+P!{Dfe)_-PjtxJ#x!yn9nsOo_bm3Xx1jb$Rdk4Mu4ZC)d}(lh)^U
zaMNq!)xZ)7&qVrr0v6#BZx#nnFP_tJ$Mg6J7r5K=RSHrv)HKazidW%yZbeZfkVX}N
zL7l^$&I4a#VZ|+1l7~M+unTW-F-eMX5|Vlcw`*Tw%7iHxbF+jHO2tFV|6bG7MVp$o
zR8ih~`xLf{?>wkllC-JmCLF2FDT@}`P5Ef~td_iVji0VSdUeZF^4BFzgZlL1;&dG}
z&7+J`T+`0fDyEc4zece8FD@p@IGHfdn-bjfZLLo)PI(@ql+}cj6QRZ^{qLRg^O=^4
za&$cgyF5~W#26p#f2u>7NZm}3Gz?0GQD=z~P2z+g2=1K|azHp{ujprsWk+%r7ahqd
zi=ttA2Gf|#Vc|KW`1ErE=47TZ!!r%>vrmIYNZnB<BrrxF`|6yQFc`BK&l-SBuEq$F
zL`7E4&lOn_B|@0y?qA$ox#W9thSuVNh&(L;xJSoH37O_;r<kb&<4T!`51`jws!ISi
zfyV|`x_vw#K7pwvT4odzBh%uh6S-pr^e5=<!ChaMpwrX9A%`t~0#lbdHRR~@G?I-W
z=#6>`9bS2!#O^(n9E~9$;r+xA0Of99HC{`apa6h%kcNVluJnaTvwdJQ#WpyE@+OlR
z62ak@PJH+gXR?y<yukWnqwyGHqINdmbdeh%+!D7B1P8P?H}ocb=6$98L_{xt5B?K=
z3F*i~K~xPur+R9>rkrk~lgFMCadi*InTyRO0w7%pdLuIy|Ceh$x>Qp?>AXDHC;n5?
zMfY&)PVrB7?%a_K!9Co{!;+T%?B1L_8h%6|@x4ncpZx2fzi_zi@=Ux_{KK6)cMi8*
zUMH|8%YjX+2cx$n?VCkqE8_c=I4HJMd;GTOj6tWMJlYWvNz#Z<^Gqv_NIc`NHp$xS
zuSbiRt8{=VjE93IovD|A5qMtw=dtI><-GXoWXfcoYP#cG-wxcOoUAb_O9ac5y;-Jg
zX5XJs6Dd1-?vS>|djQIk`D%4O$;3<BZ=F#Y6G^7JMmf4NxtwV-N105`<uo&T?^f0J
z)K}m(2Ic{kw|saYNg>@YzP$U?_^@Xu2lT@3F30WezVLc%993s6tVZ{}`OWu@dpo4p
zV+ym<7xh}q|DqDB1Jv=HsLx9KG*!e;P=+KjQ(+pXzKR8WUnBr8TQc50nD}%QXP5<#
zzU#g?r*u3+uv>o~yOhj?c=CHu{OIl5w}q<<*M8c;<O?_s^O=HSMoKsTxt0shc)Wkc
zGlaVM`MQhA%$U;m-oAZ%Go1(Q$mGo&&*2KMrlE9Iczcyv@M}@tk0f2M(0m6A86^dR
z0lY!<LY59L$9NZh2bHCX!T2^IS&cm7epRBF6du))nduM&+@UzR<2Gz60l)^1TBs@E
zmPNG)g`XYsP6m@`dfNOjT(jWUkI+m}Yr$gi@<0FcKVwtix%RBAXqw!mjLAL)c~b^i
zZh#3diP$jb%(4BCT*DVQK(@L0h6;SeKmYSTD{5GWhAeO4+mQM)W3(%4nnI&CH#7aC
z;y!SBSMq6S$wky8=FW<w6Gw6D0XN!GPF>66I~J`AGKP+-BC&&nA|^r{z~p`ZD^Mh2
zh|ydk)}}i5XDgVemAwZZ+--FGgZ1kl$F<t>+%nH^H)hF3|I4m;H?Cf_2+T&;cj{a>
zii}2CTwdO9-qWY{`rh7dxU|&zMTYn9+G$lbHjYFgY$gpsT*({0NUVIj*IUw;_ii3?
z{@>QZu-Cg$_xG;#g3st}H`?v%R~8=J-inHMK%C4jsfrueHbTfptr76XYaOE}(2H?&
z`yu>FK6#cCNx^!66Zhr{lp<w}B6Xg-p7MuFwEr>(9&%BH$5Ob7ttL;txS$26XqqW1
zXXX2IJ(s;u-y7g)v_Z{E{L#g@HgTuJyl!$2c_F~LAmgklksr$QGpgnez+_;#<$Tv`
zsrm7cU5|Yy$VmA+=f=Ksf>alPvRG_=aPr<yURHCZEJZwk69tue0yscCsAYio)?yT#
zlm#Y@y>gai*~MGx?9Ro-MaJrHxnRQN^jF;8mp<{t6C<cT@kA9yem}974<rqAlcRMZ
z<S239Oi5u}!kzB)^-h+CaJk&YwRA>i{<c!~UvqPFq;0Gfb-zuGIdf9zDgi8%Z^N)`
z>m2z04&(WAJX4ZPPY)h@a+J9fn?UtlqPYH*iSSZD2b_-?)|q9n@-5=FL+AXNxD_3A
z793kurgRMe9UpZ@D>_ybQU>8{c;^H?E54<eoJzjhamRv{4`~RxftXTzbd|Z7j^o>O
zlQ6}&?^D-Dd<zY628DQV0&!D9S_MK~h+D@2q5ASX58e&{t(Q1al0@#MmWuPv{r7ir
ztUN4&V!SvJmlG;Vl0fIn_+1#|uYZaYMeeDu$5&?di(7N|XRf5on5&cRK4&n#lTcBl
zNr$I*m;4Rh@qtg$Lwlnk2t8N=4$|KaVM|~h6h?Ekv96w|*OOH-b@YCWf4zjHv7Lt&
zHq$(vAp_!F@sKad4RGeDzPK32M5=z_)Z*gYys-3oAW=EqG;B-T++Y8703JZnKym=g
z*km%H@-+14>Xp}HNvhSCmgXkLf|c^}=Nnj#qbVF`KDzp*-vt~qc^2UPe~R|HdHSk%
z2I@cT$%9To{+GYowFgD<xbz^(qVA{1kIqTsefouJQIw$|A}#6+qJtVb1%;<+P%29a
z=Wo&Z>kt-%y~qX>y-$@@NES>nc%(07sMX`xvvJBLp@_T?5R+<_))ctc#>Ud`>6ZB-
zlBU~K(?`$1Q2`O0puQNjP?bPbnZ(8%9b&-%5(VG8>cifxR4R>0qf$`}&bMt?w?Z~W
zEbL&tyB^*KjX9R)s5%Cf*|w3lU6Yhk<NyEZ-#xQ(?DoT@0LRlt17O%5`+&bCGp4@q
zg36ez0@e-NW`^xOTsJI=HsC1$hiRawGP7+%$Dls^h3@?oyE<2);wyryiiYOmF^thM
zdJ>()Q35(XP=nWm4g|GBMQNR0XC}F>=V{#yqPWnYX;3xQTnGEW#(b}IfX|`1EH??#
zNS9>vy;Dt3hy0M+!M3LtJJ|5^&nptYcAFW5!<^$r{<Jo};vpQop|7*JlxXn`yXS)+
z{GcvFS=K)|HKh2ktglX8@oTqH*5BKuMq@f@U{wXcyx_QsB47-R)uyK#jj<+%wFeu?
zvga*EhG7_h48t&*uIrA&0lfi-S+d_EM3OUl6j&q=8SUg821ev1#X9eBH8HssY8qjv
zcJeLD@*A#UlH=v6-&Y~J@%y+YEB;O6@c0I!^xt_=(3YLPv#cqKkn*CSeH&90A>}i@
zOjVuY<2c+@d9SfGSFIqwP0cBAC}mGEMG^GA-Pd_h(7t^$Tz6-Aq4c&z{2dBTI@4_*
zz}#MD8Jr`y4#0k(RBZ##wQC<O0MIJ;)}#3>CmJf3BZ{JKbEGBvxZj>+X~e5Vldi5d
z=ug6j@uZ8`Kbb_OIF=+TcTq(zNIb=oo$v-EX<F9Q;32`O{qcCbO$wf`T~DdN3)QN?
z3-pwSw3BZ=JayF2zGFQ|fvIGWfXip50#c9)tx~z8mrs|7#1EVLOa@Gk+dCq$Man#q
zuQ=9=0vp@R$4SJZhn>j0wZh9XKebc(RI|D~d1N?*_6ZFiqcpGK(Nx+Y`8es?v?9g^
z4<Q&9sw3HqpDhC{O9uNnHPZ~PLgM=a{E^&f=Ej%ktl|WF$S!sa3h&ifHY)~KP%cXj
z(G_12&eU3MLw05O=yLLr5nZbJ^wUp2?KD&fx}Lk3M4m!$;YkH;Mpx<aAz&h?i76tz
zdX)$E${LAM8WGv;3{Dlp=lw?UWskxJH)|&2(a3yz<fSTgnH+M)N1TBC`!f9MlQAxX
zu=I=v1=_C7g5QGTt7QP0esSRd%2A~Wdg8Asy%`IJ4E84mPy>As;FH+AhaTN0PA7_z
z3IrYJ)<#}+t}r3g<c1GYyWO>`zg@G2m{aQQMdCm8pLK~Dn6pTAkbAoI+f}=3KZW59
z&}J*e`QME=>bhfWe6~{IkC}|n$e*fZn>Kdv^^Cqx;N%mz(2agFA>AC@-C+lzW~^q8
z6hrG&GOu4B0ato4*jIc`dhvJ$icrrk9O(G;Y38a{=}AEh(|G-jTTRz0SC2ZQ5GElG
zkv*VfNE@1855)_jEK9FRAcu?ciPfEJe;g&svh+Fu`Q6163c2y*bI7F(<aJ4wMMuc7
zvBmJ($1yTUuSv2j_FqHB`goU@e_xQwZFEl%dyJvwrV=3Y^r<{jmXK5&Ec07{h$esG
z<dt$K8UK+RR-Iy>Dm`8l5iD|UUOLGY?>#wTV5LvcLrX<+Bwa_(jU%%|+TlEE7@Ra?
zB@WL<8~hqd&{O?JC-Y)2E9EeB1mhUaaw~aflv)7EG|okZ74V}}F{+?ex&R&&!j6hI
zcwdqv$3CH5tdjCoqA@A)N!@pZuh2+&S!9~@1Dd8ew(sl81A<qvSXKe~tM>V&PN#!Q
zB`o`ot5{)3|1G369#JtU-*Tl2T)q4mQ@O1|{ThZpXUP~n5(WJEFDs7-+6$e)UczZE
zkvOD5F(v7aP_ctls{nQ&oHpF?MbGR}KGZuNO}a0L^+$Bk%0#We(Rp%9BU3<m!#<ng
zH=q3y!TBh=rx;O?_Cr<&O?;}gIQZYfsNXAtDL60m3EQ@Z_k6K5SgC$>#`)H>vSB4J
zqA!hlE21wfJXz(QbGm*|wR=3T?Y5)&3|Zp7mt`#sJjs-Yp@gWy+^6V%y+@IEvTV^T
zj3IG(MU67|+En+^68-<BN#!tK@*LyX<m`-q3SJe4Gj@S9MNxVSZm>)OXrAIQxBteE
zo+m4qGohBv&_P}2`o8b`k~*Zkq;j5uYi`V<W4bmDQCFFATCvccwS?p(UrC!TS(^zj
zU2)430#pX^_XS8-O;z9(FQvDbY3h*=UEz-6N*>Xyd~qM@_xoV1t{U(z!#i3tO~%&D
zg*8Qy<+b^j*a&&|wi4yuBPJgbSMrGdX8xNua``pJOmpqXGZxomSy9&JUydWh$Lv{^
zq4=N~L$e*cJOWFqkq&toJjxS75Cklr2x($V-0qZsTV8;ltYnCb(nytmSCg;4lG)c<
z{Ack2s1+YDvAQtd>&-7{U>KlHFH*4ywcm@<{J6a$YriY2SHH=(%iqLC9)<v6m>bx~
z`sxh5Iy&jc5M1gj5Or;Ants;(DOGmxuGfh}W9l^X%lB}3dSS6%2dGaB@94-h89UlI
zIy&-<sdLI7Z*I8`P_Hj8Oizp>#!U05{-E2#&Df_I78^$2-Bw7|u8#ufI6jpoX+qp4
ziXG!PO{qTqhZLu!CM5_71MW^~)Q{f$^wUqb`?o+uuFRq1e<)2VsZ$L?mk2eZe)RKI
zReen(UkB)^$dXp=CJOQ)wp&Zs%fKQG*ZO2@&mc<0vM+lD75hsoD1D;UYPFtOE}0XV
zvQlugNm0*78x4oi@<85oxjD2Q?MKHE3T@k7%A!OH{zNr)?Gi>D(KTQ-nU4|d0$T24
zDrt;n0^6Awju>-*a}L~L%m-3p0#=%FS@_xbBCL%s1{el77(XL!#X-k|N!U65ChYVq
zQ@cY^?$Auj8-LS8PRs)s<1p`y<#4dP9N_g{uh;9Zt*xz<U!)FduPm3LZ7<aI&+LPs
zJyI63JR@YudA^M1Ld-_H4(#c8DbVt9JT(=^`Q_4<s&+43?z6gV2vTDb`Ghv(U<vCN
zt#3B91~+VIu@i#G#$~_+n%T;keUIvCDA=;Mmqc;r3Go7l!Yc08yi!alq*IOn%}&QT
z$Y{aN1U-;acRn329ZIIdJfho=SK>cM0^;GW?`W+Meywdy!wxs7W91*nTMjk2gEg(a
z0(t>i^%cLf+Itp!<(-J)xruZmq@L^-c*v6s?wBle0o;(84mTw4&?-8LKr(LcR(;?@
z5k_ol6W<FTMpb)aO?TU2nyFFE4WJ5LrjazW@Bt#KR#di#$zlYk70Sr}0kmCV|M{Ox
znO0TBd^QxPPKoeYQ&H7vP_r!Kt`|J2>(=j?*6&)nKJY!4XBgf?t5tf1Fbrj0ai4n1
zRpu4L5LcYO=9+6<O;aa6ZW)H$mgTl=7}m!pR84cQLGfh-I{U3II*iW6l4AhJobxD4
za1s+QZpoI4{~DD;?`A<8i=az;b{e@!70S4-=3j3IEGV`{Bn(7S^x0<YwwAA^BL)I4
zuuoC4g22ZYp1&{Ni33ZGf*>*mAi9oi%XD6Tm$!vNw8Sab9DWmg8-<a->+pP(bA-{<
zwkH1{U!Rl4eLFFN%-8I~q1)lH@h9QR@h9D5A6RZ#v8r)y+nOsgzBagfYJ3$K<NpAo
zF@2QAsYAmCoB1-A84pu1_Kkux+EKb3cv0v@+kU<_4QPM8*q}g4VOP_R!17JhZA^3>
z>hYKWn}AJRX^&>$PCHi11?|)4RB!~tAyK(sUG@D+t5xxRwZ2~C4dR81*N29{&4nah
zYei8jEq?<gbQ^jap@tgQlUC~4W`@P9`t?z=fYuK+f-UldEwByk!z)yx6`CYHuu(IE
zAU4FN)8SJMLg@afP8z2bGkg`M@ky<Hmw-d{Mx#DoTgJZKyyeRP5X~NPFxPWf3LG1|
zlz^%kmMhB+k!?xh7IqxWS%{^;aeMX40YVL7JE~;s3_F)B%c@MTjz+69WvgER+5_(i
zhVW)N2fMy<T34%Ed3pr^Y8LdcBQW9B3e#m!6h+b`1vCKJax?%%G9A3uKiG&WjJ@k+
zNs$#rvK@l4aPav5C;2GKCtyBQ%3CQum5xI25)x<)#rqK3!`w6pKbUl^;h{f`t<g`H
zB^PjICK0uz8C5WMDk@Q(2yj#f{)k_iYQ!?OISmRQj%Md3Cmq!>RM&dljPI9<-kUW^
zQpD8cE>%8uWdGb8u3%v2yFMg$`1OP9^LD6J=l3|Oq~&@G)oK+YY?<}?vIESn*IO&o
zQ~3gGVof7nE{CO^cy4b0aE^2=g@hXSc}m$d*^6Fh{XFIcLma0&GzRUGmR}Emg<mDD
zG)Qr4yxILy-U+Zgr0m84p#63c-V0wwWpvWpfcd7C(c&yQ=3bDB67)ITA0(Jb|GL(%
z20>V<gpLzdDq#@ZtQ!Ws1>P05q<_wVELW+9LEtz+5LPR>TXp@`Zwt!`RZ?}V3rFZ0
zw1M7V`**azN$S4GFRt!|R!afY6uKBFSWuI=09Zh$zZr*vL)5+{C%T<($2bv`OB0|k
zcR{S;nYsyB)-_)yfo*V20*SidxH$>ub!L{TNcR}$O?V3RF~-<6IgtPq&BcxzN>ZM#
z`k(<iv%{ZbjDbV~Use<x4TQH^R%K>HbyLMGz&K#C3_t@<0RRFr29Q8j6p(cu#fBoS
z_`t^Bxf;&TNr;YFGlC(E3O-=;FX~d3oyUcqtZB+#8({i#X77WETW#T&3F+r6xc%u1
ziYxj4_=|s_KH3_<0PT~c%7ZdhHi}SGicC{RF_WyvBYA`Xb0H2$)WM>F@L-?-$5t4j
zqyzfC0eC^Gx=`G}bQ%C@2Wn)k`f%LE^AQ$WgUv_{{W&#Dmo_xU0E{-j;*b`7uqG#i
zWb|?fVWe+;7vBe8LM!Oe2$*zGvrRpY?jNfXm88;xfX$t53n<1pZ!sY$1Xz!Nb_<Qg
zc8h(p^x-$NrKD?PVBws1kQhe?E$cET2%Je-UN3o`d>#^RJrPVs_L5#C^NPNwSw~iH
zGyo+1c?BfN69yQrVSs7n9{7$VNe7%Da4xf~DFDYVJ#grHH700gwdLjIWsOee->*1M
zY1`p<?AH{+(=;JG&yY>ikJ2=>eSi<Po{22tbpF>qG-uB-Cv1;^RaTzZY0Rxt2K`-m
z$mUux4eaGke%EPStOH~JF38PVt&oMQ^#}s_gU-DYAy~qmlb*Ubi5X_9$B@1-N(u!&
zd6lK4>_Me@Y7Ln5+bL;P|8m<fz|H33VzY_il@X&4l)<zL`{M~6+v;1C9@<x4d5dae
zeInjpu*}i}ltt@+H0`p*1%QRc%QS5uzh2W?Q&Z)_+}xCf->(OgKlSkL&-M3XYie$;
zP@bA<Y1-@MSXI!9?ILDg2j4>6C-5XqgWHZj8TuXH9vTejPiYO8)%$|b+g^gA6=#ii
z=(dfClH@?2gcN*KDuaJiIr~#BY4R2y8_>8oCfJLCY)xj{>+ZD~wtx^}FkKQST(InX
zGg#1Tal7d)znQl!a8HO4VDi4fCT4AyupP22OOIlVu`F-r#C=<<)k+&G;|5n08OA*h
zGnchmE!jRe8#GdzQ7rq$o;`co%)BOw0xxs=gj+b1w=IC`A>~-+d6i?zp(=o7=g$<}
z6Q`r(Z1FWFYwsyBwLJdccE%NjmHz7f`|nS$<4VLL48R{sr#m}-_Wt|tw-Qe6y0q-2
zd-5|Nf?FDX$G62v-%>~A#0puINzSw>DdpR}gbv=6)WL#CbwV9l`iNkwRsrxH@3!0V
zYq1bEG;NR%GtAvBMImZwj@ww|gn16fBH>`x@#~2$Etk!YtjnRYN(88uSM_Rj&h-a9
zsJsgF>%ZfGAw%6I0;mCYvar%YFzSfj7ve6Yw%*hsaNRFKKz`5SIC!pB;|hJpmc8{2
z_zm3U*%JsgQrh_H?18w4>d|52`!hzcO=BXJZtq4nNd&El)T-cr5M>!0e`81Wy=v9-
z)#WQod22F1eQYw{baKn5&e>sCV62}RoH}*tdL3ht1@^H=-npVs4M3yH&TGr1Mx~tF
z5d+jZ`Kmur?DcPRSXDCx@1N=S`xm${wJNhuwx_;k1S7PHZb1*Cw;<2b*`z^D93F$}
zWT3ClyV4$pUIo-4{uisF)<};&0Y0X=r?N|fpXp8*foZv(P_=9C1=}wcUEKJe4V&tJ
zy`(v5aW?V1Q^g{T?^R{2>xSo9=I<8<ZrHW}86*QeRm-7Ys8osti(tNRIqocGz|Fab
za`}9)PkY`^WK5{Y=gZ~7KhGQ%^0~l*C&icW`+5WLN8=9aTmyM_Mi;|?v>RUrs5Rtt
z*WnBg!w7e~SMYA7-F91)9?LTc4wyJlOz#TAt9I@*E?-%yRsjG9gB;UFIH#E9@0(tP
zw3D7fr<Jy>rr4=!Tw`=vVSb5@N`@=2>G*<iIt;Os((6wKmFk9U0iGL)gzUIO%6Wgu
z&{w(c{SCWQept@_L>4=v3M$v)GLXsch}9{-B|oSeI>}E=<cV(R!uN%`dca??{u)+b
znr>7ohEDY6HmYlbpG*|vX1bhVomNx;jH-44kY~Y>s~!mGP!)x&zkvPl7<i=WI9Bt(
zT_t8Y*!0;CVjKGUj65c)PE9Sm-@AD?1%zbi^Y))a5>op+lM0%N6p=C8wkvgB6tS@v
zcYODP=NcfwlqiF;Z4gX&cRYZo+FGbac)Vd4W*pj=4s7-c7BBjjKax~p%=?O{3hd7d
z#^fnUCC2*&W70H?JP=B7M>Eyx{CUX=<CUwx0<XofNxJtL3C1MjxWH5q$Biwu492DD
zDYnT2x1+Z9q07(>GVeH0rLhmC0Z5HUSW4P$$YH6j#R&`#N5Ep#0ZZ+GLi1^D>;MOB
z55oFX7*2^(*!Q}6fK8EWGn?D(LJCFuT85b3cH3>Kj2E}v%*C<xgCQ7#Ip8?OOKW^!
zE7j^(+p1|=m`zdbdRjNxAx~fR?VCMmr&_y(=)v)cCHw8`dd`mX-Wb9_)x_BBxipc)
z4lxo5kXLGT@ALfLIagD{ovsp=Z2?%e)weL>kbgZ2>*FWk{_&F_t0$m_)2zO1zfLvk
z0&-{JHK_Sd6&(k*kIzFg{s|<KL<h2h%lS#m+!{eTc2evL{u0B=qcrZ*S5X(0^%tpT
z8D8L`I3*EKY7B+T86j0GOVofx7V!6?wN$A}VtHd@LwTbD&vL41ls}(MP4n|-a_)IP
zJ(WFwQ=HB_{S}WgkEYS+a;qd@)j#xa5{YbA%|!d(?i;yeCuy>`!2ddq^Z&P#<ny{C
z>^Hqaauh_-;$359M3Ikigr08%Bh*AabOU-bdN=xz{i={2GCgM-J}(?!Tu~BBR}N4d
zXbgdpY*(TJE%7hlW&~<Fz+n~J1SgEoGZ5)<rI`q@jLV`VoqArl$4!#O#UydLMybZ<
z(COE4;g-rHQg7Y)X;oE0RYCPt)t}luP?X?UyHW+J88=pRhTmvt6snbWYxnNf?sy(?
z_(?AA&lCy(g~E&}M9*$e3Nxo#mSq)#RLlQKT;Kn~*oB4f_hx6e85&FXoqJR>w$09Z
zb%6TgVM&sUTCk<PPCuH5N2$hjJ!SpkzB>n;ZsTr%nYIpnQSQ6o-#Z|}eb?02bPn7?
ztAJYQs83Hb=CE2)x)r4p^nVi)e8y9d7UePNW_jLq{@l59)e`l*c>J{AxR2}8t~iDS
zT=~cmm|%tH*7%ZMD1|)^T+7|HM=aB<kx3><P3K(P{i<k3M8&M|!-oAZNe~8+%H^CF
z+t52?#82CvOi1Gw$n$ul72^jv-L}K&feJynCILn~@Q{h_jI0wbg(pUDozr?fwhf`7
zs*0*J!7Wp0<JWoR`-c`V=$36;Isjw3&?%#uBI!ccg)S)?WmFeTXRzka$&7CygE0o@
z!hAjq^ZD?yT!&OKHB(rg<Cq3u0)k&5eR?Rf{O({5C#tF`8aGXgsT$#2)d;gplWVG`
zsH$OeNtPvU8>*^k3gMP%=|odGCz{GE(=>>(?#^#4S%zViN~REI=^p>cOAloc?bpr1
zO$_U`w25cDnZ!;p&4<Zmf$+9Fj+9&o@rL*n8j55pd@?<94nAj?{Ji?azgiFN^dCF)
zn4`p(8s-k$m~7$wOvH|`EG|B;nk(&u#|}LPgQN?Z4XpdQ)M-kI+i|v{TefWxRKG}~
zQ|DBeJCbPacEg+B{ATJpHCft~DK7iXR<#07sR`uXfmH0Mig2{~Bq`;YdjNuHsykKl
z0oL9kfB)+dy?uL3?+}WpYOgOv@i<325w+tHS&B${$mEfA;C%M(X5H*h6y-Y(fODL9
z{*aahimXLMQRqFqO3rugu=yd*4{>|3QUywl@oNoAp;}plmjEVZ*Hz4I<nmwaHhs?t
zqN=J&P8?EgSDqYySuu|3`Y}VXrlv9yz1y>?DeKz)O101)bn|_DzD_4y>NFE5TN?{g
zy3z>US;1{R{|?&Pdk(I{vtWyNNfnr3>$2b0ZNeEy(v%DZr0jX-IHfE~_xJkm&Ubw=
zZ#r=USK!CgcE!ka9ms8M0rgh!8sIC33mivh5z0z*!q**r{3A>4vYl<WYq9($nQl!L
zPhYQ6Gl@56qnd%iFij>vQ5a>6GDQKwOw%B;U9L1)HC%lY8WY<o*<~2J2;Dx>fJ2Ul
zSl>4~7(p(dV9XQ+z~H7ZG)1BO3(RU-)peOjT=}nd&o5Y5J7pNBG;A%*?>-l~*Oi<Q
z=UV2m`}m(vN6;0O#ddDEC5{`NIV`UlcQhMfCG>~TNyFRroA`Q3+7^U3S2zcq2OfxC
z<}W#=7qA~JUVrm`k+a$IJh`mZiecvGBFm>Ir>2WvZPqRBJM~E#MfZ>knLJz5<A0nk
z6!ei?GBJVQ_m`^+3rDj}RmT4d<TU*gUaQYt<EzA-2gZM-E~bYKjpFa;4R=2nE$W-n
zk@HXHZJF2Uq|G??bizse)Ki0xqI*mb>Q{0JUa&>ahq=PJ?+Z%ZJ51L#4-0XJTR258
zCi|&>!NAUkwHjHYRGccfcL;IVbY1ff*MOwbZb{D^^c#!3avbZ5;~jA|4E47I18cM<
zx7FLNwB3HC>OR6A({{(qOZPDO_!}43JC0MEP>gP^4wM?N2T<+<c-){A>b0(+Ow?Ri
zR;cGuMOM<19lukpRv&)az+!r4TNbdxwwY;xji(>3YN6g^G@8!hmyXwE9_>T-qGy3Y
z4#}43$N8UCc;s(sCQE@4f#k;_o=Fi~DG)>yQL5CCE5%XW>6O<hS~4VlXj|$mb&|~2
z*dW4<fuMrw0ziawJ{ml^1jU3;tt#fm9S)1eCz5WP`0@^WH?$L|qO|p8n;FSE$s=f!
z*)H!}-@{k|eNjUG^VovjKlYi&G0<xiBt=#nH*gdbi&J|n`NFv78#*gtEE%TBh>C}*
z=Sxw!B)w~TdOGK0U6v$_scv)5xvkT+f171Vs+8x>>9QnaMb#KKO%s6Wx~?+-(=;WW
zDhifmNtc9>bV*XMqG(JqO%wR9LxjXMRhDE){;BXjK|_lWzlh%S)Ke~Qxb0)P4}q#E
zQm%4rbEk?$*g@Sf)^|;sCIQ9-V?(1L>$)X`){EEnRh_XyIQSO#ku1(7j3v%33E&;K
zURGwyW%yOXG)>l2s)9zTC3B21=CVbp1}arGS<@K7)V3*J`%e<m&}?a&iDYXZ|Hy>M
zn#R==bST;0QrZ<;Bm5YA!?(e+SE1|Bz41)I&nj-_q9?Q9KoD{dM&A%c$RK)XP0SjK
zuue=|vTuYzD6MyYMq1$Awdz>6zEE;^IA2h)ubH{!2~`u><7+akvDcnlU0q$>SZ(P?
zaL;3+6~U-P`)kzUb@mskG#au{jLFK>-03+*8h`mK5dPdnn)T_wdiY^Y3H~9mAHX?L
zkU(inP(eyex0DRdC1hSF(Zo%N02_!TKawd(X{3twIf}`#D3rw%_UA&aHATw=H+o3D
zwaV^0mFxV__n!?ORp)t?KYpI|fb-T0{I@I)rf1g_u2`8GSzkq8J#|Xg!b^l;ags0z
zX^N4o))@d~(_HmUl)e`A;IySvSq;vagTBXky$N7G?A&ewm^V6Gh+b8690?g42T$Ut
z(G=)cXNcj%8)t*qVi9iZ**G_IXu71f1W{htU@-C^Dbb7yHC0m|qW(Ks0ov;5077A`
z^}?f~3Zj+dRI(Wyb=0nxQvM;LpGS!xM68JFe?VscqY<K*Q8e4E)dCwdLi16rQmB;u
zV}7|(sMMl7B^ua4t=624B2ytm%V9;P@F0;5#z50F%^SasDhj5A0MRr}BR~knibCO4
zo~CIUFlNYP{3a?Z7@@RhSP#C5awtVd&`psoNDVNaKp>Rsj0U99c!u5`vV;aQBukzm
z^h`RP&+D6Pu5>NHRL;W{MS(w=cGhWp4_EGUXzbBmbR76RV{0sCT(vBfbH0UKW7=6~
z0^XI(1(-@+2`ZJxsEixA8QzLBcQ%~ykM5>uO3jev;5fy0SSp3`!cR|3O-=A&Iy+z#
zW_T+y{R`cbbEMl&@9RrM&aw1_H7N;aTtf;AmRN;lrkiyWY)f6A+A4m&F0BrjwPh?x
zj}q*CYUj*?0`}MTDZ5t2e@xZ&TC_Ojavn>vbT`pG4Abep`(5N>@GRCCww*RCiz*NX
z9+OSfzyaJqL{l7#YMVh~XmGEy8B~>-#(i!x8a&C((cZ(zy-6M}x*kP3_y*S@>;{Jz
zKuSM=wk;0xD4TJ|Zx?keVPb0cH}Uy(tYtX$|D5<dX^lr556dyB*U4hH+jULM`Tr}A
zq1jxf;z2zB>Pa36|A`?=?Py#U+f6UTY+G(N;n<^0-tJQ1hhfO|5;!o>FQgpKBZ80v
zyvcee<#h*DEgF8>HaIWwacx9*2YTvKfVwsyxO+)5{(3I)ZLZylv*fdpyLBx{L5N`H
z5%c>+as0pMti2CPQS6^rGjZtn`)2W+Ck*)4<u3%_<OPc6^1mWP&sGO?SEe68Z$=;1
zG>*ENNs=rD=3<nxIgLO#8e@Rfw1sa2bJH+W95hTsVJZbwjb<xd>Ll8_PFD?CliLxs
zI9qD_sh|37SH0rB+7rYxM>gcV^La3W#{6+b-FX$~xIxJJ-v!4XF|CDUve*Qy6`?qp
zw9YL{QUf54w^{ZU4rXe37iNM=1PzP<md~}4$zlWC{LZVc+Q}K!07Fy`^R8LQHm};5
zclmhUfSccW6|5Kiyei>50Qkk`rT>t+9H=JzlB#+pm>Pu=2ESPJJWYZVHIE40j4M7P
zSn3yj04M>~OurbLdLAIJsEP^zl3IwOsG#b)8Cz!Fz|uP8bu+Gq0YDJeV^hxqk`YXe
zF6gG&`?XS0cTsE@&|^}g<?AL{t5$`x&B)T-5%=WOlnQ+%TRl88n|Y_VEvn}7rWsO1
zso*35R=%kZcGJpcXGV(T0@KXrs?pHZr1sk61^ilgwfHcMk@oSEr3A0c@>EINu*ot9
z0KBFno&w;#tdI4w7{1fU=VAOoHb|WIG1!qeEa<*Rk|_X~y4v)vW&)QRShcNiu`u0+
zuoy;=&o|2Be;xf|LojID?)Hf8U;pa)ZgCbD_w_VS5rHA90L&*cZnb(aM91qS0xQOt
zI_&j2%%AIpLA#YMO5TJlfWj;?Ze<3@9#BJ`N}%I->S>9yO%uX6tIc|T&TN=dY$t>u
zbZlh~P+@n{&Kg2!+u$@0AgM1AvU?0!(hunZT{imPsEv>Qck}ykW()n0F3H9{NJB>b
zz<gFeImE_6bfZ?WnKU6zq1{aBM}H7vBH3ha<kqTu7;HxTjQw})0d7z(2T%F_Q?KpO
zb8atOu^tB##lHLG&+{rxRfAW%{UQIMP19@p;?@-1;_{0VDr!+3Rq5!998oHjf>+lE
z999|PrvpFS>zlbnNVy<L<+EE&)axz;m?h27SZM|TNfR4eQNM~|>df(|CsF~Q9k-7(
z+O1cfz|ss-gZUMK!sdGlgKz#G_~0}GKNLELspS<hTUwgz$|*pbYtPm!MW>2z#l3y>
zp>ulMT$;g1nEs*r8}Lt;zm&_D>YQ}&UP{0I_ZeommL;RzAbRn^>jFLER`c4Vt@)W}
zt9%80CiMP!Ni9WV-b$}+CvNtP?-SUJz;n>SG~~k&rVgK8Z>4EK#?)jZBpqYcdnt`i
z?Rq5FKd?)Fl9b^g4XAl{-y!kRYW*Wzg!i<+mo4bo?Az?!y^Z4<w0c>M#*>Tm!Dq{H
zrs@Fni9-cK<}!H0ICW*+(A*^ORJxFLpHS;&yL*CaEg%HQH)w5rEIUS74;tO6nE7|_
zhA;pmH&Q)5`35TXg65DlrAgWyNeX7#?M`a5&XM$sEN?;Hm?-z!a6!Ox7aw}uFh6JM
zI?&|O-e!*wT_@xfN8}_QQ54mzb{txiN^`Pd*y9iG>q#OQWmmaziU4Ad1S>cSh|>Hy
zY7asUrJ<<BRA7O*#NlZNWuEEM$zn7iC8}K&LM9AED(M#QKek(wBwYjba#;$AEVD=h
zb)85S%&K#}-rR)=*L#sMP5s%{))sN51SVDIXtxN?VLXhF-agO>u45%?^ybuAfY}QZ
z3k&V+Yn0l#2C_Rr(l<13L!xP;BF3rvg=oB@d>z-pT}bwHv~^MYmuD?YJh7+{Qr%Gj
z3GTig1%Z5}$2e3r&J7-!7u_@TyC_lp+SF8ZM5l}&QPd!xcO3}}S!rB-4cK~_OPkF-
zdrqo>Un!T7v1fPnyr`gBreAkXox6B1;g!WwDX3NsGOp7@iY&0?xcPjbDvite8Xc#2
z#eoB}vs&5tduVIYXk72aceIQMp#pg=0<p=szat)2jKS(vae;l3_s@;w$7%Kri6gEL
zHLWqwf%otY5gCSM9d{8ar#Q2<bHZoPm9|zrj#Po6#)cbsziQcGFS0DdIPMmWT=a}V
z3?38NjBh-qCx?Nhw*%z6Y7Nm>&>vC)HiRJNk{}GyRvM(zY?BQq2zu0H^h%@9jHR-|
zlC+(+vR1p*)wm@KLIXaa8$gVG)#v%=5qLYwxSEPMB^aUcLj~+;0MOakNiwEgpF+=i
ztCgxmAyJ18&Ppq;b#c|>n(ud&X0)(av#5*#1AwejWm2JbkhnPjEK@sLT!@;AD=SrB
zCtlSM1<zGyC9zL;oSm;m5;Q@Is`FCJ?=Q`&u2&F7)g!uJRpgAk#7%mJ07;gM6OE!Q
z!$_8kjftWxfds<vW}AVAtb_9r2AC`I7hjC2p=g?7P*Ld*2(rR4)<5C^=EU=LL$S>K
zS3^*^GGzf!Evl5~gJ8Z~kw&#*O@&?JzF*ComSX6>t8)CGGcz*?w^uUUc>o<i2RuD9
zGc!F8=3RpW4QnZD+K<{*UG)lSg<z$FvT#P0CQ$Bh?(&h^5~iQ-ji_onLhqaWum6HO
z?_(|*Ou+zdL{3~VpuSHB2#MQnym9>EeF|I=Bf+Ykx`&SYvK%})@!T3u=r9)loPcdX
z$=c!2se!>L=Ot>0%dqVSa70kVrP5v;A#IXR>!ICuhzZ_)k-MsBPnvRwouZ{TeJR%j
zG#Gde<LKz;l*-M{_-BWl<chBKW^$0f*q)dzLm_`Xkuudf=U~H@6_xR(slN2U3Bgu&
zM1I|s^Hr9+C2IIfnOl-X>2#!BXaxj;QEIXyp@nh@K6A61W$i%Zb}HjL)(Js)YI4_(
znbp++`5l-u4esjd>S+6Zju3|RdT#aVfAsIK+n8Qm9jM>B^45u|!-uD48FN&bHNtx3
zjfYoPS66fMOwK0+m90OQF>QPYIzU^g)=o!6uxdC!gQbwF5rQ;O>PCpuIIRjp<6J&p
zTqfj7!gK>av@aL)`P^*geez1aV0)<~%xg+xkGa+dvzwti1k(W?qLR#BR*I~tGf`}`
zoaxTY=J(pM;Mt}EjDt4OzuFcMdZygoYS`s2QqahWSVlhcw6JRfjXJ!go8=kCBNxDh
z-b3}V@w-$3vST0cPu{Q|3{V4IhhB}|9QOw3xd46+5Z{H?k<(xk)#6|P=$Ar<dPP%&
z(z9L-ooF&FL7ine^#ObexHm<GQRT!!vTTPo$6<r_LbsSWc&8%c42X!a7h$R>6okqS
z9r{}n%HOrk-Z*#N#Ef30M;F<O=DF}<`hb1R0(dAQ>!?`ryBep_hL~A&l+T5qCSc-k
z7nMc*F}W>Ph|nl_C14|``DNH`@$arPKVHY26&1x?QHel&lqc7H$M`AL_}`}9$I<>%
z+=E|3HUGt#Rf;_D11X@Gav|CifmV|6j}o~$E`o0q7kXDV5a|wtps;7}o;^Eu?AY#@
zCO~1&p1pG>TwwxB2!U}WF<+^_j*#x8*G@(s-F3r>@%NxJ{$7O#n&#M4Gi9B+(3y9P
zMopRO?*Bsrme)<kDfXgwzt$UWO<zLlgH6WdoU~?82koXnx90o*S`SxV69$AA{AtER
z%_OHu_7i3X=ORsBlhVO~cYVF6CM!ROmGK{7Jc{ThpE#)XTV&ML>lCp~*dGRvedQq}
zXi1jzfL)p+QrAX%g=w%%!wT8lD^O_JJGN*I$`#*yI2q(rQL&MwSO<+<#6H$fIiAnn
z6GE7LOEXDGLCqv}qJ|oZxVeK6fnwwJ66@+((EcO>Q2AEw5pbi3dq(sniF;YrPuF<_
zx8NMpZAo=(VNe2~QV$#_kZ{O2b{uSRNg5)k2f`S;mZ7OYsEZvtkZ?|4E5jN5<bz+|
zXM_+~QXN~8L6H?nR#Zi2mB=7iI2b#oqyP}dmH{+SBqA${0<zA^5m%)Tk{z}6=Wr1|
z>bEkgS%>UM8=NB`BE<|pJ@~m>H|vJ*DuyL8ZQIA%Z9RXC*8gyi0eWTeKg;dCPt1j1
z(G1Ra<#OB-W^Xh+Z`vXU#!e$%@t56)PNUn<!w5CJG)+C9z6l`gNIw-C#E!*!r!nC|
zZ4xR?ziF~%rv(8+S}anuV+6iey0NnKHINAC*c3TcA$=KWiBc26W;`EN_njI6K+(3r
z5T2q40pJBW$HAM%m+|p&d+TpA00QN;aXzF;wE84EkRUwnGoA8}z+(tjfRnC#v^PE)
zwCqk-Z?z&WVTv<*HI>cKV$0nSw&zglO78)JSV)}RzRhlzw~PQX%_HY1^>t|W?nL02
zSC=1I>cG9lh*3*CnsR2Bto_R0?fQa2%UT70o+#lvzT5tH!0WYt1qSdfwB&IOa~3}c
z!uL3@SneW4f!6s1OB&AzXwDQy%Y>RWR|q~iX<7e`_`Zhc)=rvgGK`rcYl8X4V#yMC
zQPt~p&j0%?R3YZLWs*Oa>nbtfl1bi%p=vLsY=w24RR?ZdPo4pNELO_$4|sc0TDtr`
z9KOaVt$V>4+)I19kkBtEgb?Yb!O;bvBP3HC8X)BcNjhoUSc>)m%dOPMmu$FVA7HtQ
zLieV9bF%P+;#GxbfQQc*Z&=n1(}$xU8it?&eZhNQZh=+xJhLsv<Ln~%x-ifdt1zkY
zRoB&<*&x5hwf+yjiuTxp*YT)I214<v^MxYdwRi+RW4f++PKb7kF-lw01CTWE(zMsx
zY5ZIBLYy;Q*KCi2(yr5=rv15P{9A~pc%zSy8Q4x}w`<+?DSW29sbG!7tEHgb9+01i
zZt?RVEDX2Lt*xX5bm~<>Fs)AQA>T3hc}`o`=P#oeIW@<lzGMrrHq^rd=(V4qVt_t1
z)h3kvI8*0SjkyHfX3kCorzrX+BoXUjco^OZpjB~w4D>;S(qspoZ3R<KyQ;R+;Osk~
zR&qQA)U9<~>J(RC*jJ4BbZehOyU_`BLtF!Pdq;$`NiHIJiqof=9?fh)OIWQ<S<4?h
zyJ_JD7l<G@dkJLEUb;4+uyf0U)gX^=yuP&X3qGFlne}?GBG0W1A}%fWg7{vsrTB!W
z;E1~<8nPzEE7RH^dM~#>wMJNlZ|PZvD*#&==^!RRnIxu8BF#7m)I+p`uJ;d2`vN_!
zQ>vcsa<ayZ$?<dR>+2CI@KV=<AXlyC;L8SEBfNWBr5Za;FWBbf_+Qu8*Y#ey@OrLV
z%>@Ci0UhbWKX4O@6a!F%apbF{*1c-1Jf6ZCyrz;^`%zTZ?}^Es8HBT2+IZKCgki1G
zC=?*h=R7Z;iz`!8)hdRGxV@B)H`%lhpul^OF16zcz}4#1R3*;kJujD!p-^ZvYO5}W
z5fDuyFfg|AP^uHn_a-OHCCj{`RPESx+&8UKd2-V8HS#SQywmwx`bG_E(~hsEARjl|
zHSx=6EA{B*WN2msAl3F(r<LCFg$A|3$ARH&Ax|p)B>BSZ6HV$q4va@!>;?dF$9qg#
zotx`YLz?K!&sF(sg830%A?+)?{oQ3f)_6Gtx7spM4lO{{N9!^P>2P3mqw0w9A`toJ
zng+NSwS5tON?j$YO?2zh&5GO;h(aikEzx9rh$LacDs2Aw6DLme&$V1Nf5m@@qUcmR
zisRS>Fdu(-*(vL-R}j&PtWhq^&z+iVPWSu$z$aXMA1s_WaiV{19lkZ$-FiDaAbO~4
z56AO+TaZ!I!dMulVu(yNH~am5vOw+2EwVMj+ic}zk6XFS)RcmR<KczG61%HbCv@*<
zkl1#wjRIT1<Sp%_D!~&gD=UM%_F7ha)0$S*seEo~GM}HE%;%>Df$cN4H0s#nkv;fp
zY0Lz?F%`~SbImn|aZRJq&;UxM|2~t}Xf)mn${XG=(rc-qX6xes@aah&b>(|i<2NWc
z0>~r?)ld!vD5#P&$J7iC3Q2LFDQ@8IFTabs)*!t-;+V>Gs~9n_lf6+7;GMSg0FB7o
zO<qL@?%3#XIPA5j>ZQmf%VSD>1%3_uDGXDhNTN9CPID>O;hnt=XL!$`F6dMD<Ufdh
z4gCf@vTX>{C{p33Oo{5HsUO)3w;qzM`UHMng`~z~rHPMbG6MWELO35_8)x-8W4d0O
zTjq??ZS!;Xt=D%Cr<B%>)%{Xxd%a)(=}k1WEoGy>Uu!AX`|A$xuS==*{(eKspX~oj
z-}Op<+>@jm`@VedEB?!)A*G#q1jW5CAJ`ee?*LDd#IYWU_s5=NZ)bt6Z&{)pZTcMQ
zu8&4BE-Q^X-js)}7WRwQR5>!^kzy(@FUpKfRNh^X1L55^z0mcR){GgO=nQEzTYLQ+
z&O~Nx<~L_Rcq(-vK3&z@w{>ba+k6&;b{3xopbA;;K2oP8gQ2jg7n~^he>s0(Q&}w3
z>xH5sw$mtmTmi^+X*+5R5qvEpZ98&Z0_cyG8V>A%y_nKnb<u|8NogE4Qvjm(uL<Cu
z|JoA(YNi|`pNvI5bHkRz<~UZND|H_Ytfx38Im``-`6);R4huh{3pkD7F%V!y-}0El
zT@~}7IbBu<0I?u3Cgq+e9EobPQ7RcX4m-DFOCFceP~cR7$jjZE2ycI}Sj41j2;at~
zD%mng@sMW#u)QE0%J67+6O#fqL0D}oZm{iRjquWgQ)!>kBJy&NEYxSeytA`INWUfd
z4tKF=6K;Wc@#tmkvQt{x?wL@X)a_mEnL<Q5Aa43{A)g`PPBq^JYJUz4A%v1ttR;Mo
zIf~vE(q26p7Y5mJK;mf^H%q356>SaxX(tIO5@ag&^pA2LTa*IZsEsu%M#w96R7pFI
z7Z>B$mde)Bb>rA|rM7>_NU04Y=B#1Zsm68k?t6045N6Igm>Js<JLw(|xMkm=ZKE}S
z*2dOSDJg_hN`6F2gAgXhuo`v@Q)lY3H6s->4z4ldV1n>d?p|XPkL-oCp7OlUYq&V7
zS*pfW!o)MROUJ0i@w|gpI^U${c7yp0bXeM$gC04YgxdV(xJ)5BFG<rH^H0CfVe4(y
zMe8Fl(z%fvogMhFUYg_2HOl>{DqdBk6syVjBoRs*u+;#6l~OO$xIp0jB}<*$Z>xQb
z|DeZ#hPCSkP$Mu|ncoy=kqR|L)%N@FvlJo9m_i4zVsu60s%*7n73)W6U%auuu2!p9
zw`8?it;&|h`?t`LjG4Gkxa~4PaS05D*KgZzpa7IPNwb+aOqG{<=?|y9rScIn8L!1#
zDyO|#FD)-U^v)ZxEOPtbZvQ-d6a1p}sP#7MJFK6!If)r@7NaaL;$m+Ra;QmRex`ZG
znjLL=0aFHlM^P(XloQG(T$jF1Ci}-@(`=aM<Ywp4oae*%Yt{y_E4o!%vR<mtv`zCo
z8)xqy%f=%pHqsRRhp_XrK@dnCCrVKn1&G;N+YE)Fq)Nq!mccszRAGpi)>>a6iM0$K
zrvvc^@BAyhUJqovoCHF&iITLwE*ZuD+-|pp2$JPkLJw0W*XvX&0$<po`xRHdM};pC
z*7%D4>-qENA#5Sg6rYj+A``%c*gAFU6s8(bD}?hxa1a>>7-oh?882B|G%t&Fd?W6M
zjcHpwq&HH3iBN^CyVjrSHy6TZq<kh^fYDEz%zs;)M2MM)j+FNB`<3IU?`L~{3b58z
z#k9cY6K&f>SeHuBH4Cf|P;Yh`G>T30#NavRW<tO20zhUogo%k5GOmj>0i-{XI|u{g
zdK0rww{(A+&>ajM-8c4B(9O`6b%Q(m<r1Nl8YLvdTa9U@+&2re1=BBAC|>6RgqF(t
z?%0@q>O=nu+sq-?1_M|h3<iS>>(?6y`SmHF#^T!QejZqjQK!?se~M0Lw8pmmt80tf
z(L|T&>W>fpC9urN9{h3}LUh<`5;Z<5xh2gk9vQK5Y5*OSi}-P)stpLgq2tnZ<(R{&
zY2Rz8x#0bkx6nkhUE!O~cuuPn3Kqn<Vs8PU>XrXT*{cGig)SRS;|T<Gp#R@b{8GtJ
zRxe+<a^=bu!IbG3H>Xl444Jm)?d{9A*5Y_=YqYYmz{Hib*cw<zk$uICBvytY(nZ<^
zF|043W{IPmYF$@r#uzLX_Y&?nxufd!lsnkLeh{&cqZl)WW5iTs98hq1?URf24}wz2
z6`^|j*nD@db7T>Gfw3`T<0yP-T~CxsIM31i`uxAyT|x>}M3*|QsZ9#nR74-mtm70(
zv|rcbEVt~7IPZqCCNBq2ct=aJM3U=A>hT!+qDzSEkY@U>r+D-1fdfdIx$9hX;K12y
zk}VVE_kx#=hA8!urf=HI;+T}Pt-U6@{}F9y$G#5yvbQ;Z#fk;=ll50*HK|xY^RAL6
zsMS<Ih<S@4sJ7Zp+9*8~l19c1V;?>Db_trlnl5ec|9TpMC8*;VrNIxYi<Lw$21IM$
z1igX~#eTIR5Ep6#B?;P4f+NwW_KQLkdV%q^Ccv0TDvQ<72d#|bPyu^9C|4TYeU1v0
zCP2A}7b~?wq79XbxLV-4;*3`AO%O`%tq{gk9SiMJDXGy(p;lRp1*br?3Y4?2+o+TS
z4qs`m$=MmC4*}o;##HS&UBg0*{(S6P_fs5^nrr?CH2*UU7)V{tI#~W7s(#zXDwBt#
zyrxN~`Cl=?FtsZs#4Ng=`MR)pXA3WkD=EDnfYhDLvt%^#pWtv@mJq-;JJKi~ChT>P
zsbfk>>`c5?O9bX2EA?iO%#Sb+%#fo$;kPr6(VwYLX!bj0-19h2(py;2*{m70<YUp7
zK4yR7DP-4kpZbJtESb&jDiy*$Bw^dg-N1P)S&jY^qYe!@zP^rN53R2sH}X)W`#(UM
zNL-9$J^+~;ht&1X_Ni0bJGzcnt8i{Xm$==&{TuN2@TsU#7<<?{>u(~<jnBwsI;s9j
znXPL_QIAi-(@ggKmi<#>@qtCJ<ub#HTgArFS%uEi>Ep%S_vRIUj(T0M`cSoTq*2&R
zsr~tE^`ni#MuFNNfA@X%d|W&KJP14v`y@R3xgZ=2_3za=A&&+SlD~cZA<-$96}DF@
z6`mbL;KZEH4@>mRYm<nT7M5(xE0yJo8_9lt9Gsn~9XTutoXEFjz3*tFst+1pfww@{
z5kA>uHp=|Eouu8wNWtEedpESiTw4+Q3VQcrGdNqbWDHZrsjJMd)fJVUqT%;1eLU)D
z^t;DDNJ{I42P5RNIC6lceF(&;-XV-$Q`=?$syIjYn9o(M@t(m3N^3^SFV^g$I5Gp9
zs5sJ(DR~uoJTx~0Y+il_9^GeTXv4^VQ=};C(VOegHb*=6lkZsfSx?G)>TBz~rcOO$
zO{0}E@LlnEgKYkSj+Ls@QfhBuHU$QJWtY3W@a7dvOZ&0DalWDlxLG$K;$z|+@*i;v
z14<0kJ4xC{fr=wR^P8M5B+3-kJfm$(lF5*bF~MvHt&&|Z^68b}temvQ<GX6DX}@2U
z#-}<^cVpx3$zWaB>w*iu6?p>C|BMgDrM!p#(av$!-NytM>$d2=@1Dh-yEmLog0Jbg
zr=C)2(lJ5RL57tw1drRiL<p6+qc8Z8DfiC@NNM?jfN#>9{-db=kaluO2`O>=9&l!W
z)I-{DA_hNdjr=4JdE@D@NVWn&e{iN8#R^gVgEBBCRY$3jCcH3*!kJm3qq!LV!B-}H
za=ar7;s08=8zl)E?w}+=J42Kt=x&T3y6g4XZ00KyzB38RLIMa;-G1|%x2r+`NKxLp
zP%6Mh6Z-0Ixlr1?HmIIY5K7)}@P!1S<N`MDPY_DZR|nU&7y9?CtgM*OSNBt{wHBb1
zTLLJ#N`cT?rwCRQI^eV?R^QsP4oQ0>rFZE$0)&T6GbX9dQ^7Ur5DcVNEMoCZ4<!8B
zWKzN+g<JwN;%2nG_Msz(F+P0cL$6<fTcCBLyj0eK>em`t!zY>VC5h*=O08D;)Ka<8
z1i8FX>#IPQ*P&+l>jR7R`XXF|dSh#EyWVKjxA(Rhb=Vb-ac$lU;!>oA;~Gai9NCUz
zM;Kq_bdawm9OF7dN2NIE3C2^hNzO3F5&l;Jc+I+(aTTb(ukeCv`IK;tnkTkK$A9VF
zK}_L46c1{rF&K;`CfXCg9;cWyLWt}LV23$Qw8o&d*W2FNIk2{&w8QKlK^tE5ASa5F
zI=@qfa8yA=DZvFA@r`1>KAtRjS_3wwTw2}U-db_w591F*=PZoVK==*fvvKj$<!TYj
zL_pmVpbzPZn#5?xgw*E>hEDm2$ih^(?yq*wckPX-ZkkGTADb8+hFsfj(IL}y-eN^0
zW(-TJH1?Ii?n=-kJIip?L^7|TOlSu+3;=?Tf<zF(XZIQNC8+g!eLksJt!S)S`8fl7
zSM*IOHX%q;vK)V{TMgqh3feS^`}d5;bLfAZJ9jRE;3=x$%{e8sX`jjGJfj_JX7y~q
znF-2Q5SmP=G*By;LX1{uzLn+#?ZkCd1^KZs|Krg4BfIYiW;*hQ=HNSa-B&^K`Nr=$
z@*{TVw+4ejc+ye8(rx>l-CYZ&rkBb7m-+P<=_HA#y(rCQHi@miLg0&0d;d4t1t(jS
z{Oo>5Kbx3lbiGB%yY{+v^QOHB^feKt05|79U&(3fiTJZv-DfOY@li!8c>oWjqYqe`
z7JPoF`*rNGDmhyYB<GX)y3q#Gp?|$(){ga(S3dT&pN-Ot+9VwF5YCQ3VRv+7mBthw
z;U!9`=S^kVFcv=>|08?w5MQ!9w7ET?KIg?P+;zdx22<m^MhOr)41yRzVmsiv;2Vq$
zVTND`$FTF)vkD>bBa9O5oNHD1wzG*_b|1Jt!qhdGG9oBtlqy}SI2`#dfE!>2z>I^r
zn(srY<djh$QkN=vHI%Pa+V=3h$+~d20(&HTA-caafUgb20DqQ~`U*UI-+lKLeIK;)
zdktf>U8~vJxQ&Kuv|T?45Xlk@o2mCWP9bmh7r^(6q3iN)RokwqRE%ruT20%oS=>Tk
zJHI_<hXQm61p(^9HqOpq&w3QDKD{5BO3bI?jb^6SbCwYbBbxHb8MV{_+dZayX+pQ)
z#SyYa34*LIJ&)9qXN`UI9G?y>IgD$9nT-xdn!J-2rsS2x&8GgQeK%=u-@()-kcWOY
zb_OwP_>8CIB6^B*P(M!Y$?(J(KQRc1ga5$augUB1|84&h-{nyTv^1ts&EsNCt&@%f
z4lIy8)?6iUlZ#(-tw+vOD(_odT3Wmj$Im2{N-{ZoILopX_8v|cdxp}L`D=)sk;LX&
z51(0FT3UQxrE(*wRFY@ncyi>Yhr?mB<(}Du(q|Z35&Vo5Klgfu9|G?fizjwHyaSQ-
z4R{N{UjXTE78#mE_MzvQTh=&$1~!=H01kS27tLhX%#B^$=gbi)uV-0yh9Nkx@@Xb}
zT!qecILdda5Y+Xr$;@Y^JTf^Y=Ef=7=geW{IO?y4!{He&R!)tXoT>fh{~dy$6xV76
z&I`3#NDKnMSpPyhK~l8_{r346VBu)T36Di!jalRk4g^9GfTKF@kh9*JCb=*mkY6O4
z^z%`T2Ye2^#beWaN*g`=XAk+-qEt2ST=cMC%VaV@o0}&G-Q|iGxK498+(cj_Db0Qz
zf1M{WCMZb|{kKZdMp)Q^=TF-$Ai$X&5{k0ngHH;ncnbvL-1e!0MC&I9mdirKnC((e
z9rbpF7*{^Q5+e~$ueHWlm`K*zJGr@uqOjpQUa(v!;LbsJ&bgBqvm{Pt3@I&qSQ4$3
zP3<n36qnj#)axs|yi$BH8=|`fh}E59t7S_CUawbNt_7ZkQfaw39cwAUY;G45{nv*T
zr&BEicpF>AfSB7iK`1GH%`rETEcCwii6V{B_J6fJbQip<9aT!SfCH1(3{~webXs{Z
z_~+BG8L#Qnl}=|c=yVn<QLH)Faa38HJHJv4PRD^e0AWscIw*2;t_~MGXS>@%v9u7o
zZoE(`emkQWr<0Kfb7+!A?|CPPH+wLfx-*{hNOq#E@I!HmmsOU9uAOg%MO$wmw4;e^
zaFwop*9A{??*R~+dDrJRNoK}-6RFK%cDG<sIJt$weH+Jz^WRXrp6+vj=#UKeJ#~CD
zXc?g!oXYH+li8~?n>*4;gWuUcq&UY;u~@4Wiw@>oHM-qK(_L0)4Nm_Yrc_zUdO}f(
zZNW$5C`nV>GQqZ-n%UgPqoU@tYYT1Mq@H%-O6;_=X6=_H-#MGjU}^TF`b4GvVRl1p
z%WZ$xkHvR5>A6T0K_tILMn~NfF$1gtIEO$s#EA(hnHzna3zkfepH2vDfrNFB{7(N0
zysDHU#8yVwE@g@n$|^R1*V4L1sg@lNO#cU*l*Us<hp`YuoiS;<k!DUQoKEjwX}x-t
zZo$(*5OBH7h|<2K9CJdk%ZyJYCJM1dJCXntGJ}7AIt@RzIRX}GQRiG>N&}7PohvKg
zzSbn{)mx(n={Ussc*F+`2+pmYc6H@)+xo)-g{q80$b&Se(dlHPHAYkU@-6Qx%I*cM
zMH^B|r?^pS*7%RBq_m3y@q$Rf5EJ!$x4{v0nbN5dTo~8&jKY-OJDJ4r{ZL%6DJ35B
zD`-AAo&WiQ+n(p8+?SPZAVkoyIYY9jOAa+<5gB4cM7Sq0C7ftmdvO%{5rJ-Zc3{1)
zsMgNVC&HLh>86p%#Cns=aHChPS4LS0bKiu`k8!5z)6tk6BN!3xNCZOBc$8AVlRGzW
zj-O4r)q9@mMNfJu@r9}JJhMJeXNmKCDf9k~?~JR}sv|wAZ6-yaIl~NdZsXY2LVAJ_
z$G53hZ=KxSL<F8e2w_)qr2IfJL|SHqetsPp7kdWw`}4`Y{9i*}!KUU?WNI!sgN;iC
zGA<EsafHVl!#(Ca#~8!X6A0<R1rSn4wvym`#>Cpzgb=-~(dOpKD5L;Dki@R5Fp7k6
zjbj+2z!a?BzwcJJXTIntHqoBEacpXT#GIPp%&9rTvUJ)$x?~OI=hDX|z&rhqTUS$h
z@n7~h<II=YX6UJY(xA@96SgDdosB=Dj(7XbXieRKxxC+EXHMS8Ss_G@9Vb}J^IQau
zgYVTbKp%Qh$@5hbA;kAzfV|+^$OF8q>)U$0t8KsAUGH|!cDt`%bGPgHw(hQL+wb<G
z-nZuUkF70>Fd5hm(<v+w;0S5ZzdEPtuP1ziVw{BC?s(MgMwoHAeeZo+e}6jW{4!=t
z^=S%SqOTak0YLC%78X7Mp|Zp({X-(W`(sEQ1It6rI<ANZ4_Ep*D12&H<K3yudSRp=
zzx`mvRVuKB*$h}RCU_qH`G3KBh1$SIo7tG$3%_xi53)gQs1A~}LUjo0sKiBMo}#mF
ze6X_Yej4_Z0EjSSjA3H=yhW6O=N1aC2bAG!EBqv*4<n=aeBabDK!`90z+nU-Ab?mo
zjuDFEGMklo4QuJ@q2Gv=n<a;=Z?<k&79=|p<9s|B6Lo;onvOW*oq|3De-c4`aT;mm
zD7PCJ_$QdOqltTR2a{Y&29R^DWTPvA4FEr8dXNp`(c*Q_-u(WAzhpcz4p87q3&w2Y
zf>Q2ZKC+Dv14z@labl^(h|bEp<Qul`yLVf1hNEtGH12jo!kPNX%M-Y+JzrbleFAqN
z<3vIMc%G*ibJK$7+cx<abSbriMv?$so9+arFLM6N0YLC%l^(l#9lP@3=k1^Wk63RZ
zglH+7aF4r5DFxZoj`q1!f;Q=s?^H5;3%+bP!#*<y4da;D+IWQ6+Ppvr@r<?!=F5LA
z`j~O!*fvR=4)Qr-Yje<GpV-<M#xeEOm%6Iwd4zCgjJAo7&B3o0N~iw$|Dg4CjwZf^
zP7~G{)09T7R;qogx2ZjL($Cs;%uk&<CB!LC8E&;OqZ}rF4YJchoMx1rrkC-fgTa7E
zNd_2WPI0e?DaRNqLjD>ZP_fEz%VM|x`S$0KfM2o75s`QJoz}NlKRDk0kO?hfywPOM
z#$>Gc=*jRFP&h@^Jeh-*I_`m{Tj!X3n!+?6#syoF2PkU0WZ$yP0c%$~6!lD1SHR!1
z%9+o+Pz<Lo?`9|KWUT9d%Q*a$8%3^b?1hC006-xQO!pDu7#Syy7P(kfdgtKLV-F(4
z5jIX7RXLx(eEylmNMX*^mVO=KZA1#XI2^4@*&nPeHKh#ZpT%T9#gs9;h!~TPb3pXf
zN~MAsQ;%^#6q9wf&L8mbs&U;Yauw$YR9FT8DB2+xi&5+tggHVF9y@w)N2_JdtDno~
z-^{U!szv+r*8h$KnU6*lDkTY|Y_-=0t4A@=b%H4Z{+MJ8RWM`pwZ|TNOfiOM)G+&&
zl1HEZY3u9Td}?%Ru&h89U^;R?lymDSnF;zWLK%94=4j<lFUp8_`?${#&QwZ%v;Rbq
zzeE9%G6zCPqd^GaB?5>p`H1eV@k9tniGNQmJlcmI7ViQ2!TblbL4i_SYEz6+5Fm_c
zyM!qP+MvrF9$I==kG^B?x87nsZGF!A|B-LzQ#B`yAIUA9lg*K=Mgfo1&j*0kG=MpD
z%*ttW&>56?%(%+vKy3HYxqc6NmSx`e!zjn!7q#nKTMs?7wN>ZLrW8MmJ4f2{?!M>o
z$M3n@^VGv|3o#=$<MVqEX9|HE1jZO<j29URzYGAHaU>Bqfw$}lO%S-JZ;qb7oP%f2
z-m4GKqRily&j@oIV=7_n7=-YnTiXb2Z#{|;GEN*;jAI-|-WQZg4uB2dSP|s=lu}HP
zJOSX8N`V9*m{M-S&~*T8d4sLtd=9^YHT?Yl$W`m0b<BE?UuM8zS+Xj)6m}}kBF-J0
zYa$Pv=WUq~$2D$3V2M1uNDSUX3kHKn+;)H&S6lbqx1|^-LEC*~Fen1?|L+l`-W6)&
zd>#)KQ9!Yjm<FhbrSls0jvqW&O3Q1dm2>!PE2Xt^S~_^}ctI;;v**=r_51xIf<Vdu
zP_bWq34Ht4^ZnKn*3(2R-0946O9Hj_w3#^k*Zq!9>6sT9<t2oF5#Ws8zng#P_m&F&
zK12%odQ#l$5uz~0ywGqw^EXa%o_=o}upauqB8VU5Oh_%J7-ziS%1f|XaTcXvS6Qvu
z=m)C}Q>XQz*P+N&SY)bI#KHT}C`(d6<kFwJAuj}j!NaB<q#&0G+9p<x7R6SpD6~*Q
zh3^EdjXm@;|MXg~Gof^JS7y9_&j6{oxI%>`K-=-J@Aprldt&XMcUq+6;j7Zlz&**X
zSV#9?r=4ZiH*st}9!F?Lz46p{uuwntRY4%q(Mg2fXlaR=Uw;E*+6vuj;Qhif9Rh2s
zw|FVz^vNtj?*j9f4^R5O>0Mj<1Gaf7FH35()@^r<4PbrCWqY3|4hDmjI{BUYvJvx0
z{`t;mmB-#jOW7v;KrsLnTVC<QlfSRLdt56y(9W(?a`v(W=B(NkMc)nNKQ&7>#z%=!
zpq9L}(*Yl&0a*{QMDM&1H&v!x2Ea=1nj|BzMB&7^g;otG)IHkV+&p>xf1Mu4Orc-A
zttC#fWDdLwZEl_f_x2h+002$_R%dRRV%hE*YnKFw%FmW~{Z*gej&;&{&vC>5<8c!p
z4w21?hnVn_Lbr$H5D(`xADxf$K>&t8dB3sPeupXZBX5<>JA7?7{S)8ic_+wo&$9aG
zuFHT|@X9OGm)pf6sN6uP(7U9z_Aj5@+(g&HqK1ruC0PMYaJ<xT-=40G4RpQc8+wUl
z0>f=K-Pg|GYcxJrK9r46KypIO;=r)^^2;w*{`mM8Ap~EWz{zd#Yo>5t=yTu7p;E;U
z9yoB|-YQpjCy0OJAOG<mS6)6&?<<&J6VB8Lj6MJV0Th)Q#>sbdb^Qb-RL$j@CB~;?
zYtJfKJEM+Z&Ux#ib;XiD31vy?m;~xNiG3XdEv}E$fzuACZDJu2Ff#!mMaZnIi+jUg
zwluRj4A~F20UeKX0Zp7ZzXb{?wtg!1zk0#(DT=DAf{QGF*WA+XCbywv{wKu@@NAw#
zCoMOr#o#1O9zo}S;&~plGVMX)%t8|ktl6`eR?b_tDa^je2?^T8_9<BU{l)9?=KDn7
zWxZ^@$|1**TUZDO&_!Z6O-G916gE|EpbpdQ6h>BS<7)-C1OAOpLA{wKqY>S+c&`{S
zC8fEp<U@SDZ~=`Mi1ucDnFgn4W&ZBBmbmLzx~c>Ec8iThu_zJa>M~Ovq4lDuno20?
zK4w=+bBOMzXTtt07WU8NM&ao_&Q<OI3I^M1igRD7>$YB5`rDp953`XT2rjWmI!dCK
z7gEedn$HCu_q}6jOig>-*J#Q+I5J3EBdViOhd2$|Y)s8Gn~i1z)0CP~`_pM*s?Jiv
zd6Tdeo2e(R#mlg}Wa<1zx!A~-i?(a|e#=|{7Xd=tE-rW12K_rZK4-gs9ESIBe6H9I
z2LQnF`9hnVKYu=LdRp7se}QAH%LV%Qc5EUU%%9HHdfeO8Hy7`B0fRJJ5DdV4pBY_!
zKdq%i?8AmXi`a+D@ZwEB`{56NxP(C4-XUMxh_ylg>+q6w-FicMWzfKbX)j27=4aVp
zG?)xqZlw8tVyQD|nsy2+Ze5y;-=#B7I%$8|&X9&A77PrDuxkCXLo*JHD?%wze6l?*
zF$5HP)8;Tt2xi?Yn-DVpuL%?oE5(ZcK$rFbe@Q6K1w#lSCWKN#Fbyn@XZFX0Qid_3
zlwbfr)WGh<sezVY){|$<X1wD!a6SLF?GHfz^kbg&ude=m@gY0WAA1MfGppEvD|m4g
zg!L3^|1s+e)^A$>Q?GyT>?4)kI?R`P<p!qd3Z1*d3X>;4^zqMmFd7#IDGa9;n<Z&u
zLYK730b}LIa9A>v_vV0X=Lzt)rU4Ptzi%dM0;k+Hi<Lq_hyJhKL_Y_iGG9Jw4BYoA
z@F1mcv}C*o2dHs<Ur3Tt_p5y&DRE5@(YJN??H;NOe{CE|eX6>!Acf;BATI8uLxSrU
zlb$n-lp8H02jS`|UrI_{H#lWbfx2GsL(VIS&q*9RE?K|mO2S1t%Ug$m1H1}n{J&oD
z3g%2hq#4I<ueY($>$z*^@41=_!S&c92-We2?ureBux|pd`=8(Etd~CMXU4Ss_hI|s
zc`mST@4q)=n*{g&Kfmv{9={`6)6)*X`=|XYw6E6w%dO}iqiZRj#h?14g%oU?Q4`M<
zeiA#RgJ<ZN@jr{Xg_F`i5Y#!6d#t0QT0s2mGF8DgJrH+7UiwmRIL_ygoOY@W5>so{
zcyjhQ8#QCBK!u(~E`b%L2#BE^wj&XHH5by9y!)M<KZ?Q?DOz-2$aUa{IGnu<u2(oz
z@LbrtDJSRzZ+=S>46$z^8yTp8pzlTW1c~~205tNhvmh?eH-A(5ExMC9zVt8hfohel
z%(}Cmf%x174P%VXfKo(agY!$1^DIiZubK~%=!g!GIQv-IFtu--rP&15{6aAd0UC`M
zyL=H4hQ)%<q-5{Da^;EvP|~>wl+NEdec+E%D#(Z7-SyN@Q$MZmeM{VEw%g_VAGpyN
zmD}xRBj)_)H#Rny2d=|8UB%OQiE944u*JRk-x|ir=--u!lI*Um9ERoW_55GTMDm)?
z9awi*hvOD1Lp-2GDQXJvwBZQvs^305h=z{@5l!UjG*@F2eJ1s3TKeGG7aiCcvw~8%
zXj#IU|3w6)b|#$u@Qt*-w+jkg!|&DMzW)pCmC7NdUASJgK;e7;6_0Je+D;NF_3tCX
znE~VR?DcZ?FZt_PISfB#0Y0%=NAmCRrM|2=QFEg3%Nb1_nGOJ*It-?+Q)|2;c@BS5
zZnw*>lT<3nabtY{xbHg`Fa1cV^u<e;F3lc@sku2`g$HKv1ErGd#PM<8H|Dr8&c#bV
zQf{})^EWPCx^(dVb7t%)9K0?x9jjtxtd9*1S3*2sk~+^Bl;Kl65pft_HGuZ+>w#0u
znRE@roGK)j(<gkNE_FJDp!Ds5y!(OSIr!&u=gtiuxLXEqPZ60PGnw*#Ecu)eDWXH0
zx3DCH^$F=aAc3^FJ)|gw=NV5CxbIQp=bwN6`LTMB3n=Am{zrqZt(L$0k1DlFfK%Nh
zG-Cwg6rq%ujFP6N9LOqN2I!13Q(8G{Wk{YVg+eEj7G+~qpZ}fev7<196uQ!JU<!jX
zdv@IBf9<>0oz|O*9pZo2UJI3JUi2gu?_=Xml1^#CMZGZ;=fbS_aB;0!w%TG(`mC&l
zc4q=tIm^ychPG3&dN;Q0i}>++1mQ2wW>fqu7)(Rl`bEpNO1_@~Y0|TWSv?E0X`W}C
zOLuO}%;)*E)`tjm=eN2*4pqc+Zhq(kCNa9&7nM3h===uULCK-{KM=8Pa9zGxqI1Bx
zuUVe8!P^aTAN$9hz(kKCWG`vWM7iEhh`p{<hkq8ft$+8*TLg%C_t3bG>bB3Z6lTu|
ziMhXBtv%I6Gx)Sbz@JYZ79Jl(W_DIrlQ_pfjwr>+>gvu9_0O$)R?*s0`TF&nJ8-}d
zXS4)Kvul2`E-+HU&4JL{;-fWtQ)iE1kfe>vM#^&FOE9cmcJ0+fsv#?m67g(;{B=&n
zhQ(5v+G$lN)WH6u;CjtioS(c7zwY*;RkH?GZim?9nDx$RdGTJQ$1|krr|frOE%Y#R
z?m78VCrO)78;9dIHDO7xDf3Bz3zNdK+dT%;b=df7e0HDe5WP&OCIg}=S=OYZ_PIpw
zQ=N4}HCe+;y8Z-*?%XY_>SXAq{fqwNB<n9os^eZS)>JO^v*dh#A;}I=JL>f!dv@0o
z{J51#)Ns7~UC`~<6W?0b%nb}QEH}%F;SzpuaWgsK+TjGoMsb;dr_xsDt6alx>zVrW
zZ)mM&D&(bC&l2*9+sQ}?j}ArYDEvq|t-O#{;Ztg+fBU~FrDpnQNemx_zZMd5_SF&(
zuf?r7+^~E;)9LnVG2ChE9uqTRd8QM`$bGqamS-J!SDxn|UH81zAW4ElQp(F!`y_Lo
zr!9uBreh+lo-+Z6s@jL1hM!$uU;k(xBuTL9c@HCm_NnUpp9Axp(&bdboL~IB!%o`O
z%kt^^36&pUW0$0D-RFyNWmg>2A=0<dUT8U+I;c0Ns$$gGI6i0&hAEHcFdIy;#s?bN
z0yD0~Nx$A+;T-<3P$&e!{`c$;7mCsR<n7zH-*a0#7mSj4k*MFediClFV;*tC(2bxQ
zz6+p#t>5p@Fwq(ZIB^wzrcfw^3&m*vd-eyx{MF6<nG{*XQke@z{p&xLx?$)(VoU_x
z?v=0kT1z>D=*P&3tMxX-R&a9AI_A%V#w4E5ogPlJH0uNs7Bk5tWJ}X-B<UWStu#%Y
zo==f~zgsA1`YibPG7h?-eEVAqQPambnH`+}k?$N)K^yK1hxLhj-WO1O;)y5x`f$)>
ztJQp`ErI!sh*(9-4`|LluJ9|UG5-4v+-I#|E#>pIemt*E%|4k>#Nr;NOcx(O!<^M{
z@?F$zv8@rp9@7=7KnOB_P{JeM{N^_c`FRbrr*B#jMorMey7{C=z&=zajQQKD#S7!d
zLq_$h0LKp>KCByoLaB27VnHRAZeGcxCmTmS(*n^MKmE*EH?bI>mj@0ZzqWd9Uhp`%
zltPMGpM~PA)YV5)t&L0LaSX*lI^4KD2*XRB*F4?Kg@*l#H=-<pJheSLqErkir%MO-
zlr7!HX}w&TY*ix|8vN0RGw7t}$jfd(NKxZ@V!<c}(>M;wnY_GILa#Gz1b=stD8b>h
z!&C9>_2|#F+jC%=uo6Q-pDWbk6);WsAUa+25Ny?1P<X{T?M-{afH~LwJrlruvA(hb
z`h88SD=YOEO#t)vKms^;@X*37PWO_%>%h|S<4c^N9q0tQ5m_^ZWSc=eIdxTE;VVih
zJ&LUlqCUCtBFuXrR6o#}O6U8$4Ca9as&kffOjvum4O4J}Cdi8nfJ@zy{HZAwrT1%F
zGXb|bqxv*}xh4$H%7rFz(l*Z!mp&HZvOV?Bo%HE$G324L%)s2OQl2Oi7fH4qgI%b1
z>PG3`EaVJ0Qx2qMv|S6DmI)M7XiBv3s@ePX`ez8hUtNCr<(Gf-uloJSyBBQ*lq7$E
zTXyig>H}a`+*g{i0Nrj46vuI5unU|ybLLDR$eI3cVc(gzz!2Wg<N!Zg@OnFUrzZ^1
zipLLO7KY_P=kV4kbQ5}IlnOg%b>^CqBY4r^9!<4uMic}`V4-0V_AS7aQ0Jl<l95d1
zM@SJCfKnW#8<thPkz;D@<JU*7sJae-=U1l!W6HUXD^$0rJC|MIsD^M}BkxQ;0*0q)
zY88Z}5P;yrJXE5|VCh{j{;tm9SgqqawW+BB&F5>Y72gAh^M_89o2E#+!{uE^7;>Xp
z)#QH*N8^M2rfHh(g`5mzHPm+FAi%ObX#(^Ob2+@#f9*TyPV{V6-Q+*;kos&|&VR_d
z@cL}}&#&7J6a(RA#`y$ETO<64txoizzT)T`J?cY!E2QG8$oO?QUpvP$i+zZa@WCXF
z56GaThJ-pZ1#!;;lE!J)UMv$NMF{=Fz5+?ZheF$tmlW5j{-oN=#hd?(*!xM<b(FT^
z*x~rqPu+Llee5ZH`^`7s>|ev~y8ZUs*{l8czw2G^VxM5YecNrf`LEXdH{X0SyQW(8
zVI<683su7aID8N5K@e5}1*YhAHnP^ba5GWW3w2WPp4D{^P78$tKvXzl;#eyI#xTc;
zpDz->3)+j}7u&H;sUElYU#UA7CgPntj(XnF9XsO*z>a?9{`-wW!Cjf0TyYBpp=g@=
zM^S28mYGJ=fSr|b*SFKbF#X$nYWhH})2SVpp5m%1Lp$0N!&qP@#mw5#`!g<;G9M4_
z!OF-|RwK!EbiMu$)ITVdN^TU`7IEyQ$r=WKUqh0x<5of^BU-L_s^Ga`u+VB!aGXNB
zl+WAnx<;ds1B3h9yS^zVidMVQ3M#%2Vos5{V#=0P^ev+pmvbw^Dwvk6DY_mYgb+n&
zYlQ1^dZwT2$#v-ND2B^@B1TXgO_F^@q6LMO6a@5@@n5@J!0v$FT4_A})6~4@{2?S|
z8J*$VLL`3{UKW_-7p4Qm_uf*UX2plU6?)$>_{|7Llv1{0B|+1^fg{9uL}{D#31s}I
z;&TTm43@euo1}|&+t@A8(tvje?G6FtO{@!yU6q!C=fE;{-0Sip*TG<#Vcqooxr+bd
zZ>)kLs<!H(X~Ko^2zvij@bmRoKU2^9g)nMK#%jD>@~gJFi!+<18UAsK(TpG*1F3BV
zv=^a<eI}z2d_eu9g;kv%Ax-U2@Du9haO9teb^P;c2vPA$Tk#L}%f`%aNS5Uc299Oo
z`~{(^?nX~{J=)uFiDF#Ht%~^@Oa1|&6wZwQ;wG2JN^0JYq*Q4)Bp#SE3>`qH?7E8n
zPB-GTQ}=MnUSOOnBw13eaRh2QW%tbNT%(O9E`3pwF#n~XXMA|zsF4N7b!q5h;noub
zfGFR?O<D0OdbD?hsRgzV?QyJG<dqlnoIHcfU!CB`1rl~5?8?GNpZ%&Vm<b^&wL%M7
zH(Tp^)-Y-CjXvti%GIH*R=4a!$9h4>6cT_GlCEkMcPVEf9PI82Au;Ubpt{>yo}0&=
zxPeuw^jn04&!y`tY#!AOL%&+re>gQ7jquw>$3Mz!)?g#DCJ{w3G9Hwk4=@<trzoHv
z)%B}&!=Qh7dK3Ucq%^1*S-U~Sz$S9}str}}Ubrn6Bcj_ZD|N0))L3;!iHJ?bnU@my
z8$w9IY2bbcbJ+0uq5I}HzZnBSxl%3Jo@W;$%Tjjjz7D*;tv6MBkBo!$emZ9jC&c~t
z-=FSQ^SKM^wL5p2<a)qyaJ$#_*HXnpbNgme+*naAe(1~fZNW8>+Ba>wE1>(9#DuIA
zd{fuabzMtzs@o4>1HS+rcjuCdHRm)61r;(~UKf%iBg3+9;DB+tGOCoZlop7(cg}By
zVME($Q_~F1k`5f3XDJK+Ed37v;Pr0+5AZ*X|7Z6*G2k|;qiGK(MGZNqCZw8n5`jVz
zO^}r;Vu~AxK!LK7*8#ng-SHonFz$9wewpA}EeYAeb3`sz{J4rlW3L_my+tbdB6bz%
z;q0Hb&7ACZ@muqS2rfLws@0|+?}1!&qc)&!_^t%&T>m=kMrYBh)JansU|A#_7?z<%
z^`S^OUi=N5Yph&YDup@A)^$QOZ)!%KnesG3fa-RY^F{PU4XL&5uiM}6ilmvQCMk;M
zIF9`V>>VKY7-aXV(IE^YsRI-Zu~*mzYd3E?&g%7GoDBJq)#NMmA-9f&M4C_>S1D_y
zl2%q+N12dF$e>i3X_RD-X}XbXz*{_j<iBb;e?f@vI~eST3K$n2vH?5a7h=KBNz<hg
z<Rp`Ern%+!IqAU*coF)T^$m!5ZX8$JJlJa`Yd21a&<Nr*8<V7VkLwC@vx!a`Nk=j3
zD$SvnO--|NuHLa@$LbeOqoOzNZ}0ejdJjsHv|n%zito=zQbhq%<NzR_&pUFrDE&Yx
zN#hS{iXsk4<rSH-9i63#i$M^4>ni!%CMdl~Da|QFHFtEoQ`}HbpZrw8_jI5tag`lM
z>2Z2&O%>cQlmflHr|23kl@?)HB)3g#W7@*>z38>*O*T>?dE-&iWnga%`v4fY+$Zwf
zu$}(PGorqU6&%dBw-Ci^b|KO-)r4LjOF%g8v;Bg7_uc~XE)LT|kmgyW=E3ioUKEW-
zQ3UHz)Y^!<AUD|596{;m<><F6^oqzVQ%=kIzg=90Z>j2%$Myl|J^~UzSTX|wigAZ-
z<JZLS`MX7(w+*hWD#8q57@9&^Rk&f#=@k$~>1K$Su)UnFTjR~pbsRT@$kJm6#Odcr
zya!ToH}QYor>#1vpTu!kW<r}yvzaur-hE{3=?S4qe9JnFu{SatgmD}O7)D(_d0tMn
z?>+U@Q%}YD96&A~%U{UhI`8?;q6g9Y&{rU=4)rjnGAL?Kz8F{LD2yv{ly##NO%fu+
zRv3Vvlcn8GIGMhBkKx_vK5I%I#P;LntuR&ZrGP~6ZSB<9#1&=D`zm;<{Mg(_#1*dn
zyXdXPcc5_}gmoh-8PtC%Zkatv#T#G3+kGUA=kdS0o?p;)aT9?yCGFbLPyxLi(ozTp
z6h7Zz-feuM9fDmzdor`xsdT>iKHV&qw)sKel*>*K_}fZF+*;Ue2%#IC31RGNwJ@wf
znmQcp;*PW>sVZTNsH)UTcigyZ7mh%*c1f1;u3b06U0TJh5&W%Q_pnFA;y!u>dKf*4
zJ`jVgNe~8IO*73rW*Q@Ef_z1|rW~oG!VVhI$<ZJxK{TMgRsvi`1q;5UL<B-ZYHUH6
zIL)Z3B+8bk;{|>=5~V>}t`U%98&sDhof@|DVxADQe$rxt{8X)zb(4_%{I_o}mP1Vg
z42I5C8FSqL44M|6+TI+;{xDW@D>fJlR=Pk=!Ce0<Vb}d@;vDsTlWIy{RRL5ruadXd
zYqe5GpA5n<nAAI^TCG0!yRa|t6j_oO4=DW&H=s1&Op;^;S{vJ{DQDL-uU_lCUO;I&
zr(1rA(_{=jR?r8%pscLTn9q9nV0;uG9C>oGQAeQQAxj@JhreIZbz@}cx}v+D^`}~R
zQdK97CHPCpd#=995Ln<kr8*CcuFtl&0$Ek%z#Z(TEsS8fFJtGS6i}<*GgF~$+fq4&
zGe8oAuFZp_O3X!>uESn5eLt_YqPSngl#0rYYth<`6+tO3_T#7pk>_+R&y0To+V~ea
zOVX9%_XMSo&3%KQWk9L;UQwa4mBVS$Ew~LDF@WnsA1K=M@lwuMc(>wimy{X((ffvr
zsW~v~=<S07P4-JLmWjYJB_}<;QgUbtgZn~=01LyHOe@$3EK~9gfde6g&kbRQvM(Q}
zB-08u(+(prt=u><aS|Z{7CgTWOsAe#$hod=hoNn&8Vzkr4TKOPP%S&8nrho&XzQ+<
zD|jB&8SY&Y2(eYO(Puu#zjBFzj#^iTW2(3wguVlf;(qiIS*g#UBkYP$HV1}W)~PH4
zUDK3uK4q!CX~hQUtM6l9a4cy1)o~m<jH$6sUYj*YHglzdeGw{t94)+on%WE#(e%7g
zfZ;-vnWhQ!!M&h6-F)-S#In4#hO@1dAg(5CZgU!I{s@IVTHXBCwHg<hLJoEU1cn+6
zaw4gSphNjKPYe%6E>ftd8FnM3sg1Z;M-mMDM>-0Q*kr1bwQ{4Vwc5o(>WVA_Xn^+E
znW~L1?VpYM=FqKDgpf6{MLQmrMO%tm*j}i+?yn%c_1pMn_!4TNUFZSym`_02n_LdR
zIk`>c9jT(!njO1rQ95|(z%Gz8S=tQ~bK%A0{%{g{SvZf%PB_4%zYv5C01N$AyWxzk
z7gILo7EDkkW~b}|5dC;rW}`Q~=}ioj??!|jY_(c3V^+cv-D*=XCTl;}B$XJZ!I(@f
zT`Ll7S@|dOVQAW(qy?U&1wWO9q3dH3%-P{f`jO0y8B&Xrd9fpB03?3Pcl-Sl6B82K
z)?J`d<-mahYTVpaH5ij+nW#o}@ALt7B@@9F1=l4LglfU#B=8?-uTw&nb!Pao=6`O1
z7f=n5eWdA|gP@WJ=ZERVFid0RFHtQZzVNC#C9P=Q6h7L7Ucoz+_OY*3&`o`^)9Kt@
z)%=?;H}!p=0(AcT`SSc1D!GtPxzXXiM~xH;6BC63dEJDrMP}L3YI-d#7G0bE-@y1^
zG~K;v==YDg(C_#A{0eyFkw;>%t>OjEGW~TyX&m%r!N#{QN2zEn#xePgf+~AK;b&Qd
z8o+q0kL|$--DcCAlnS`%Xl)n!dxk!zxlRw_!85^`V(frI$md0=WzaQD`-^*<4!B(e
z$WK`|<4V6KdDGGkkfRuXzGZ@C=5c@h$>>+NKYWl%`4uKh+B(%IMf9InAer6Z%#M^C
zn|)aG{9*^?MQ@{aGb;L?_V9_b(a7hGU-gchBJ8NHYki^;b`&TheWJ>M(D@jWvfa2X
zD)S`;Z4@?}Kd?_(CLdVvm;Q|drADtJy3)PGgX8ryd%8~6bbSpz=UAz>CyAu!%Y0bK
zktxGWSk@JZtTDERyA1j8dP;k2*1UGKmA2ZEY`uup$QIj4oCuRYMoOswuvOPT6U+|{
zd@Tf}3<e(5K|Ja16{ca|)PA5u7@wEVi*cm<Kx?PT=$VNd1IH=K5mr?k$uRYUAN)Ww
zwPhB2v&)*P{mzW=-DM_ek^Ah+vSFHrysY8GwDM>g?b0C{@I}!L*~C%Zg2=8hwbu@;
zZWZ+CCL!O*<PlDX-O8W4dgDQ_hi!bdjlmCU<3i030P@wUZ#b1MToNEiXN-T(#?j%j
z#&0-?bGU?}tru_vzl~fp8~?UI2eqmshl!+{WtSX(HgP#+hC8W?_W&B^erf<1e{y@<
zA9gyOna{oROr!D4GtWHp%#uCe1zo?Hm?pXTd)x2ja=Gv4o;zq+2fuaj;KA=ADsxAB
z?cJobwS!!9r!j`1p9~3-Gm)Ub39Nz|?2pR}J492+pcv#n>B*fl(`?L!n2y=>%+B*k
z+LH9{dlv@PBjOjbf(^Nw-*gKhocmG!pllzL3}7^pJvjTvgqo$<MgArw;UK``g~bGU
zLBS@UX49g$0(Aq>yh@rQ_zL&+nZ>bCZZ|$<K`Hq9Fud~pT=QJG#-J*@bCyxcX0JCm
zp~f}fdbJUBMeBYGAm%Dx%<BuLI$!o&7!g;#*JwUpKfIzcgB~amN=a!4r3O=14%hQ}
z>PQ3G$}VMaz4E+j7G94}=p0+gG&-qWfmd#IxN^lF(>8|*myyES5Bmm<DDKgyG|Udz
zBqTRfr-E*{GonEji%GAj$m}@P1$((2hG0o?EiPW;;id|k+oi}twf#ORQ80a0k9IDB
zxRvh_EWmDcjn~1z&hT#h<>z8JC8fs$_#`3HPJlCE7sa^{6(8YQX)=!Rh!%s~*g8q6
zh}2lDl=dM^yfsqU{i3E4@$p4rULI|>TG&q%P)Kk8o{FgWo-iQzK4%P1<Vm!4`t)gk
z4{nj(tKAH+;Ntc>fp<i%PoT{K4hYJ4a+^hJe=W1IN+-e)6s2gD%Ab3@K~SxoWJ!fD
z6HtnU#_$ZVVp~%_BLrbxaE+YX5*y8&ukk4aHXC(2v^R-=Bh-EmoxT9Ex<=Se9f$U*
zraL1?*J!`UnZ3ct?+G_-#)niah*0=wXv3%TTh?&=<Bg5`JtAlJcTRAyYwX2)TO;^q
z7-0plKg_s|Hzr6aqGy!z>0&{Z%&(uC>zIgxY@=!}=^^qR{vpl{VyTkEA9{#Ol4=oy
z^M_Ut{S@AHZ#|?+GAnUzkb?(_!Fh?vl6vT^|NWU<3+U-4u@OW53C_%i9JEfNt}dX8
zsA6oKENb@|6Z%Q(<HDqArJfdF4H-3AN_vd>{we2p&&y#R79^So=(O(BQ>0U0zLvgc
zS>*QujDz2E4wz5g=K~~tEHS^2HU9o>&7mXcS|2!Omg(1}vB$UzZ*emgy>;DZWjGIX
zk@j$!%KYor`0`&9kdKzzJ>vST2LOTa#oGDc*?=<k>GZAl8CJ)mGVp?T_>8#kEae7}
z=03fR0;;Q!AJSLh*v+UlRod~EPPCil&<Q-p&!FwT%{FElu##uE2Z&?T_4si~dV4ZH
z%Vjy|X^&I)={4~yruepReICuBz34Q$13ik~g`PuSt4(;clD?MfKw`*cZD-EHZpz9)
z3!{6-G0$yA;}rrIT9{$wj%+nTh*Hv=3FIj#(knDC)PH(4cW``Zh5NSb8l4P%2kzYK
z+8^F<#UfK}fLhDT%l<Z!ma}tOU<hxvSqocEmHqVkVwg{<A4(lde3i7;Sj)vNaGo{G
zd%xd2+KaA2cc3?*_oGjvZ|1M3*AD8j{)}q+f)+wRZ>^iX2SN-?`doHRR{m6qr1%#<
zh)oM*lR?;Ner&yVy?kbXDj`T%mvCj^Dmls;)3H%9cteFCir(0gOW%Pbp*p^3eg)Yu
zB42nr6pK^$3>+ev51$GQnTg5=n!ZbVM~70x6SGUKcQ<rw=&@Ef3WYR|w&~t0c)YBo
zwT|R;(A8cC-$IM%D7qazjy`}st7R!SRmN&GlIX6r4v{5vw%PJ}6{3b=8iZLlZ22u8
zNiJ!n&5#zZb+aQ4;%6YmMD9IGQZ;<SlBCs<k3oWdTVATBDpMX={0SY?N~Kvk*+xtE
zBDiOAG@1(miGTmmBQ**YMFqui3$pA%&M4#)05Lmz>Qs_YP0DH0S5XO5YFmbZ0d&JO
z`T1ZXo}HVWi>9@lr0LD%)T!B7!Elb2-_A`_P>{G`TU5s&X-~)+H|PJaaf92IPBAm1
zsM*Zt6-7}Ri|tuek@3-TJ||<zvAa8nD@#}{IF1aj4+F02GJwH27h%rk)~~K4f1+pr
zmSIR(r#4WmNLc&lMx#OHT)v#Fe9Yu_E+-gcLN|YPW${-uEGd|Rty3%+h6SK0mZoSD
zRxkyybc!W|+W<G!9Y;}%g{F$J9BIVElB^<xWQ1&9W<uQz0@^FYT@9+qh;ShM!NtKr
zA}u6K0R|K=UQrVBWg`Id-M7c8PsE$9xuxSgU}9<CLk5Ml&G|B6cGA!)dNX=2`Z4<3
zo@i4ZF6hjBBj8>wa2TnN5XWnp5(td&MHlV}^&kuoK_c?V5iSYP_`<4NiW2Bs++>Rc
zaj$${bwxVkEVxSB4D^}!*l5sNd;IwE<7@Tn<w31BIa#X(y$LKzC>2KD{sXfgx#Rz|
zEX$J_vrJ|%RdO+QC6yV>v>1~;%afUIFTS9yGuhMLy(n6LrJL&a;74KYO@yrCGn%*D
zatj3q_tcO2>xa1$JwYjD=Yx>{y9dH^a9*XK-1h78D^EADrNSM1z=hdr&3ZD?UG{Ay
zAOobA0wqqW0UfZrj#Qwfv)?DE#QxkBY9GR5;qGI{j(za*cMRYEsHQdfIn@YY?_|9u
zj)GQtr&qaQUwUstF`7eT*^;(&hm|*FRMT3vZ5hzC>6pGj(G$Gai{miU_kKvbHOi<u
zwQF5=i?>sQV;S)yL#4gw`eXD^j)Q!ixRzCeuv-_VjurqU48?K8Q|R3MnlBmtQPrze
z=DYOb0^j$i{Vz2ePw7P1Ad=%SSkG(3pA1)H%+Z*?VHkU+r>Cc#cIW17sZ{dY$!sgY
zgme)jC)3fN+k+7bMm+_lmE?ksZJNXzy2d$&6R;}*aq$RZ0Yn;UCf(WEK^Pv@H0l|`
zK{OMF;VqmiSf6m#io3bs8HN)j9X++Lt6C`dRMV(mC{zb9(dZ^oEtyIfV8L1!F3i=U
zq}!OVDGlLrZH-ae*5Ujf=i04zxkVHDqK5(wZwE!4BZWR_mx1CZZjr9aalarQ){zjF
zvNBu!rUGkm)q1AP{zA40>l=><dTr0Q^eSS9y?frZ&f8P`Wc46+pBf!>nf~P|UqV%6
z+G#KuYIvkFZS1BOl4F`5V%NvD-C`4X21GHN>?NmQ9*irxInU+ecuEUtB4wT_PA)>K
zu~%Tyq!+#P`tC~@1+?qRz<5&^>{o0XYG#|79{-q95fGJ6W?_$}S^fBB6{DaQ?Fi!g
z(d>@!&8t3+-i<znK99bQeuDl1{V$5^Lx(s^dRO30VL9(UU;pV!$CX^*96IAcsMv5&
zY855PN>#s6*-*z@d1E_e$y#OMuX6G@EB#;i#~5sa($okYSMtK0R1W$8MSX03oz;GE
zoQl-$UtW3_)W?^meBS)7pY8n!;Ov9e!kl#%m#@C7baZaDlL>mY*kLyp6Kd-T8{-kL
z{R~@opG+oB{;o&?cS)he$KQRSzCZE}q6LCnn3H$Wn|VC~#<4<sfS|b=>@7YFb<O)b
zVJeVcK#%F%zj7Ob{Nlp0_$Bt!(64L98ov^KgwG`QbiJ9d-FMh%qdg=SSddCDVZ~aN
z0Vhj5HUa7{00^-J%%)WV%U`i6JU&Ho5irEtx1WhX$t{|u##Hy$@Jm<k+9|=QZWqU}
zWb<&h>&9z9M=as75jg*>faM3eln9m-^ORrd_v!NTX^p69dT9HssmH+NWf(A34FM+N
zZIC%e?^qSHc%a>uXlLVX`VDzlB979Ef?@At+Ka-IWZV$Dc8(xL#k)nbVL}~e6$cqz
z{h$p=tm|uJC&kM@h#o^vqIaP8p^u>F(Pz=OB=|Ifkfe!BrIUWvYH|JaD(xB!(u(4a
zSp&f36tW216!|8hLpDJ<g)TbJH`_th#qIawK{9H1SF)Cn+d9-1upW}HN~U_S+lF70
z3CYN6DYWzZ!O2#u<&!;<_7+-a;I7}HFt=T!LQrk{9GKJj>y6bZ2&(H;z=~(+pu?a0
zj(@u{p_s<Z9^$t&X;0}<-m}()MrFn@<@?JG#o{x2h?i<|Z_muk47Zd<`PxF(F8@9=
zGgFyREbjhV`4cJ7KT_6=Zu;wsytPrDF-+xwr^(AN@xfNBrAd2;f8(<X=t-cLsk|$%
zL>5D<`swe8R7?Om<F<<SXp)69H#O<NJ~=^q3Oas=M&k@WO;g^2$D6uT-v_!r<iBif
z-8my^0>+6>tT|!mmM7YWho)p^&!6>cKaTxh*CeQ{)Z2NI@5M7!EZlTu!}og~)a+YU
zzk@<A!1o(h-n3At`QFce{`2=3^PZnqNNmpD&uco*A$@FT+h_%$@s>p5KE)&)#+-(M
ze4odm45ANG>B;R5vm{g-ivKa|9sADZl_D8$NuV)S?ok9>@A<^XWcG4PMCy|&>tolt
zB|cJ?Q)5N7(x2mI^ZyfRxA7E$8hp0h79#G7`kEgYS4=*WM=~?qEo@w0oJQBcwaWeT
zQLL58=uea+wv^?1R09q?b8|2<=~^$+-_%_|l(rq0VMi(KxS&)FG;kd3%04mfBemlZ
z;xb|YV@rvt#Y)9Me9!gP(jVzROL)*S*>wIP({>JP1Ev%^4#PM;$8J^wQ}t8|0+dpq
z0GI*?AdbU<5^J?Lh^ekslq8f=AV9;=dK>x}`YHuLpqn}=Q;wHjy;~wj%lQR^H$p=)
zr|And3^V#We6}NB4qi>*e~3yAHX0<JVzOQgY%3|#GabR>vndTj{F`VBFgu)7^^|&0
zwMyt6VnQZq1jfuLP9ChI+Pro;xR28-hNo?A2SPKmW6=?NyR+Hup3?u(a2^}E$jtbm
zw5GXCuuh}o!*O~NINy7;B+&`$Y@3Naxux?za*$@8Cx}O}kkQMO&ehSVWI@+nhLD*J
z{#9wVV4mr&tgN_^u!s^=U}`D2E3Ve+bk30ajpe76*6#0`)r!rSU8y#|yo=hz9=Jcm
zAer-u0OR2O(|5i9o>%A}>c#M082!hcAPj@fA8X@kjf24EQgS;8@K->EjdqNx7r%@P
z+gRw=_T$XgSFw<4)g`wD2f0kJO)-VaHPe%FV-g!mMDFS0uE-+@O||?8Vy-Yg%l|0`
zc|}yVqtI8yttlt*smon$(wikDi8~&<Ndofb)M)S6shF_Jt%dT#XuBdRj4|rDjw`v8
ze&9P4TgLZ{0c<JX(-sEXc!9AnasAL!Qb^bFTt;CZ(woM%<0YpgNh?-yEQljkxwWE1
zxV;?;Rk^joqBuBKSe2B>Q;Fx;<C{`=M-T)aXUg(j*TEo#@7ab?OZ$Oq0E`>>+M>+Z
zo-ZU|$8~*6G42Hj0rD>H<00&$&muG_QVN7hndG|;nf}a9c3h65N89u*85@YSx*&%r
zTA%n{JB-S;bXZ1lB<_3HNWgg2$w+sZ%6Kq_0r4E5DAv>D;OA;fG*KCpc;(=pMEHLT
ziocKJm{7_a6f_6{8fb%4N@7e0oDV2o<9B>U0k1Y1fD3J9t}BTYYO!9kfz+AGI551*
zlq+#}I=$|?>9mWbtJo@Lz%#|_1ngRUQ3*+;>t>cNFf<yg7-+^RB{8OhN4qtz$n%_2
zN<vH_698iVzbHh6Ql1kmFy~lcvetjb_c6Uxb-KgDecO(jjczryq-2f7#Wh0el5obd
zDu#oSfPCEIBEaI}<0N3Uig7h#jAK<NWNmS=!KAd(YPZphY`cGW*zKrIhJ8XG)rUWB
zSx1jzP_N%em+ST-EqFYPwSuPdc)cjkO*x@?`4Q!Mxn8boA=MldgE$*crUVy-b(Wwa
znD9J>bm&S~#I$~h`i3Zn38$42hLTAsKBo&dios-EIG?Hq1n?|bR0#VrH^*_z`V<Hw
zxT0C<y(C@1gb+;nEar@{%%se&GU^Jeg@<OKIVvC|P}7%p7$v}LOfZ%KA%PrCf&Q#u
z;g%&_%BnU?O^s58vG@6ZNs@RTKu#IM@kp)n8(gUjVoWe0BnR-_m@&q)%m?0UaFJv!
z*;Rlsb_oyy)WI0*=TZ7xi7_S)U4Q`|=te|aHnUXlMvsG^)flH#PlRFL#8dbdlBk2W
zJ>(Ug)qjbsKM+ki$p(Y%%&?Ld{fqe#{jksavGxrUammFhUU2@*9ha<3CdFGwv)Qdt
zBmlySWr7FP;ze=vfBg2p#ebtTpcv-=ld<{#gal9u!Tc{F7#qUeC5SH~tuLY&mRv^6
z7y`%nvcrj)x{R19%nV`f)DZ4WFF2f-!f^)<;ZDJ50atBpZEbCBy@Ze*k$Uk>05`>6
zDt6|7fBp5>ulme2#5|PB9gKHm`j8>UW&Y~*jA3G277^pJsL5PIBIX(rX<}SvGiEc_
z5F5azDg7q`_%4j#D3G`awmBVsXkL=7M8en4I-VxI6-?x`QH<PQVxtw-t?j7EMk>mU
z$%ggn4v$9BC>cm^1O`VXZt9OUerv52gCeCD5YM-x|7<!QNoVj`<p8`Rty!bBH~*-T
z7n70;E{bhVw=Bxx7;{T-aEZ{QGFgj+C9J8}1^n?Pg32oOw6uwTT>hpvqS|O`eKH(t
znS+vyvsQI5;Xq2`KY;YNu<h@)9xB#%^P7Rp^*lENf`O?6y~R8TGQx$l1nc(R{s6#)
z8^x{%;Cf-?LhEh_0Ar!0-p*GDNia?#x?n1wYXNtGg$c$9p)2{emRjI@U5N@e3OyIV
zbK}UB9qD^3feA|rj%z`XALtx-YHa|()X*Gtvlf7L)x@s@d`-#)X#)m~2>`fRHP5S=
z3z!iMw(NzKdfMp}!v(6fWR50M{w~8*6gHa+3!P3@b;Q$Hf|V`UQj*Keo($^^l`t3+
zp{m!-s4^GwWQhs8%YFM`Dx7My+gVs>Hp2)E_8k=n)(%rj(}iKt>7?~a*n5I|E!-J&
zyFsH;X*^@X;op&b7uD>9hwKnzD<l9KKmdhkl5a&Q`*$j3HNX8I^SqDKVZ=mbKCK8A
z4e7_vPp8v`X`*T_`I?lsW2l*a+_jq;;M;$@|63r&FuWLpZ?B>Wx*FYf=JP5jsQd#G
z*4Iv#0=L0K;_lL>1Gelw;*fb>TQkeC<0I&*bq!BGl4a<c%9-yw_K|du04Qm2z8~NA
z!{$4VQg6qE;I~`rWDm$6eDFcb80$fGu1Sim-ml4<nYYVG=U=ni_jEUgYOg(4$}ukl
z0C=Hsl)C=OPkyoh5Kn#>idB5L{Yj@X*7m&(-Gg3_J^>tRFofga15cLw?NRJJn)S(K
zv@Dkn9|ys|vMjn|?qwNW4X8!EN|}0Z)m-43WVGT3WWnQJ%CKXoB(MGg8Zbj_FnaTB
z5X@$6KN$qUO84gr{;2-j@7{dj|ARSPV4<Va9UN8cZnLnN)w|_(+5Wr@;mZRMRq*RC
z$EBsErS;57CV_5yyhu)wQXyv1r(?8MTU_IpHp}zbl;U@`co)8f4xlHayD|ojfLp9e
zr(B`y?ef}5tvGa+(HlL`5SX2Oc34CqAMC-_T)IM!inz>0vV7_&dNYY`Mv(A|0s^<f
zFUN3BvaH=+q?FPXfX4hC4S*FwDO+r}GbU}N$JVD74WNmI_KTZBQmrO1{k7I!)phn9
zx9VNFx3|~CT-%z|zr=ms$*MI>2x(208_lIji(o=()vUwk=&B}8cgMqlPPYIjR~>1F
zaU3>}z`t;ANA*UIao(s$_S+u{`n!%1J2D74-QdFSAs5{N^`8HzB4n!&gYIxhjv{4c
znsF3IA2Ld*)HsEiG$$LQp?9gF29(eZEEtNRh`LC#w=wJvib2^aOA#jkVpUw&;tf;`
zXaSub6zk)T*SQ<Tpm-LDO32qbETcFshw)k*J(JpcVJ0s{S>ud(UeoL4dfD9``L9Tq
z<1p3aFplDE#3_tgjzG^Rmi#Bm3exUz3EX2@p4Q&2#kUg%z$rk9Vh73{6UX&~HB&GG
zxBp>)fl4oHXGRub#y)8KEm6mG41ko0V#{!lLAC0*oL~S<FvXY>Y#_<&Ak6dge7w=`
zfu9!jnsOX_-W(W7$}M|h#tZ=vJV_Uj!>_~^>NRCb3-Z=*(6AhBH3DCK32m0;T}o)M
z>}U{vJd7~yIF<#(m(K(rVJa*&Ex<;kZT+blPB;w);|iS@LAEDt<NHn_uVOO%*=SmX
zqkp#x71!!XjN~(~+kje4oyiiCYTuT(V@xmsQ#@=*FG+)lkutMal&B=*pBTkq1JeiM
z_$A!SCJ?ZDAc_)ha}<fRjiql=Hsz9h^-6c>K3LU28&Ax_K<aHP!!l^GIjTsIYn>e2
zmuR%Lt-L!V3IY7YBr0+fB}28_Ie18AZekkMnhc7loIlwe6hpyKs2-?Cu#<wka6_fT
zlJt59@@FMU8~^MC8M?11snZKdtZdmw+hp2@o>%MlYZ-*Y@7J{lL5*Oy6)6dTm2@eg
zObX*<rFBeC)_lfG{<n%y&WN^B%X6ay<0Om$KpV%a`#$Hi(&fr>9X;qLRO2KE*xDtx
zO`Hr+*$&`3mg3zCW84es?Fd6$FW4#aTr0ISVVoX2LxD5DmN`a9Mr#u?{o1K36$X$h
zYPpz5MZMG*l!Sbis-J6}`_)ws(u}K?#(-_*V!{mA#w*XO=>z&)0A>s@!GMi+7&U_P
zBL6km0Gx5(4;?o&UR3eiyARJQjVuO;vqmK&m4gGgUM2EO2#yo_K5+o-O2dodTo}q6
z%f%RD#xMx(WAKa(hTU1FUyG}33|L#$xcP?3s}mXGOA+iNfkIS^H1x%jaXc1fk``gY
z0U@E9`MJKfG122}G%iydlT>=KA5a(0ENr~`)vtc_mCGvt73iNib0!~lC#Ui-H+%3+
z<Fj-_KMJJ>AAIn5Os~7^uDd>Mw3iNdrWcPNKmH(%7pFUim)gds@4B9XNPTwazKCYf
zHguSG9xlI9NstNwlvH3n7O4X4w*9B|segmbsgCr2QB(97W5?*X|5$$N6|Z<jP89#P
z|Ec(<_y4via<4GxF-nh7<Il_v^v5z!N<?_(3b06}K<X8({a%>6rS2?uu)PnhC+jW1
zY*I}HyHK66tKL!ubGXr)(M>ZT6y>iT(CcP(UDGVS-2;gxa+fi`yEw}oSQ+V6d+sak
zcX!5pXHkAxv_1mAgSdPwPdWh{{WKV4((R+v-eSUy$iwDfCkhlHkV9Vd{b3mDwr!j-
zFzC#(jO9XpI_J8)WfD^8UezLWF_%g@@n1dX(CA^`iK57mFk5974apVsxrrKZ?p<4{
zR{W$XYPwK#PG8N>Mt`jwZC0#b!L-J)ukD3WwcRt-pR27bV8JFc4x#Af8E9i;f*Lx^
zaebodB2(48K)7B}2;tfT2Iwbz;D-79OuhgmYMP*21-(hwE>?QzyGpUO*jF6#g_-9~
zZh(J62jc;a6QU^af&>^#`scj-VI~)~pl?xKp5yh}ocV}IjpfBxGWuRwA;kDwn<Bqg
z*RRwLAL||HzPVbR&&gD``hixI+2OqR*(YJ7`iH_L)~9UU>G6N7*W&srIs}6F<hE^W
zVnH=V`dgvzQ<29(gjFev5*cgQE{W<a5ae;R)sxIrDF>Z6HQppb9*MvruUNF?GD>Mk
z6~}5IiyVc9R8>ENJH4o?U`f^K6!>g$VT$Uigq14Z^oqj<{+9n^5CkA=W}r))OL|~x
z5)L$Ny4#olPgYgHP%KVP7K?yERVA;emqRqGzl0LHfbw^tmjo%{L64irbG<-b5d4CD
z730KZOKbvj0gr8UFx{NG7-WC1asM6@6aS#>Nw$G6U*hqbac{%}^A+80F6T1l=5lzQ
zUQ4pQUe%r1OEi<ofU#gQY2Fd1&m6*q(naZllUE=VT7s96%;{7@Z}#Y7mW^8(D*HxQ
z3mkVXqTQ{;nT}n8wZ4y^wtvHf#hx*Qe6K+mJ0vRj3&Y`%ag&ioezfq6dlq5TYs?oh
zDwk@zla_96Y(#FWvM^OFf@{GCtzZZlI>CxImY#c6vinjb4&GR}P5LkBN%V}=@_d)0
z11@w%>d%jNfm;xSVd{oFu;T0;8zAYY!)xgSJN9`{RBbEBKgapsxs7-}yCr;rE-*jg
z?z!@=SQj?aF(&NIn}k>SJj3Sajo<6aU^60k8lbuk^<TN4^MhZr9{JH?d=j10C75~Q
z9wKVoBsH%sw&zlthHboH`COwbMbfe@t<7esRhq=CMF{nH+2Sn2lA)19_b8$n5@xfD
zas<Cp_e-iORk3nN1uVBE4vMVqmgU{5A_JG&G6waKf~)CNf5=e`<8wQ~4aIq!o4;RC
zdJ5>EsPdwu98_gVs$pD{Bw0PENQ<(%LbHdEX1nJ7UxpD<k&BL@Q|M}R1G-?xJ<tT{
zM5G=@_!D%LnhypEu@5Qx$Qvu2c7@{Jd@$&wk+2pgHWI-Y8Dj|55tdirUGzdPt=8>!
zAL@1=>hA4yIx1u4UCTx;XDr`kGN#V`liT<KEuAUduJ8N)^!Z}3SnR%kdV1RbEt&Qo
z{Az!5y<V?(&wM3UU3FD&=C1ji3*eq(%w+Vct0*%WyT$|XuK9kUXW}@HmuF{ZXP4vn
zZS{J+e)a6^?Cd!}4`Si$a2Fj#ujH^rP}o9@1&d^_n&@Vk>-Iqfs3uax!(iKXTsq-Z
zNOZi3FRWViOFwt3p4+;2i_-+Uy8d7Ztn=E=>lTzA{BB7hMCow>(0WiFen)suAw(&C
zn-D8vvSrCEvZj~AS2gYCT;o+gD|{a4XRm7HZXV0P#{UT{Qj|l&d41WOX1;CX{{+V=
zzTOds6eW64(@Iz`m2_O1sw1SotRC1btW`lBbW09!Eqx3<f!>bZgFa{lL}6GH+vbR|
zV69`skEgqoI{s0Q;qWT!qL_-d6?s)}DAHw~5@DzkCM2plGcE}wNmJAo=QbLqS!U!%
z<ltk&o#gsZAif<9?S^<F4AvICTMXfiUoDl&80>9Tm2<Va&BC}mPd|J>A8=2lQaM<u
z9IU>>vaFhI*J?G(?jP$r_ifv@YBkHU>{{f4N1O|#J6@PyjDEggLFxEgik!9-#ll@(
z@79N4kFjA%=d^G1dcEHB)_;4uEA@K4zFn^$?e%)SqxE{d{<M0%UN6Jw!`{9{88Ee>
z4GnLCdro>9kf>l7fW^jH&x+=)UfT3?xpuXIy4pF%o6e@Q>6PbdHo;}1HJ`T{B_=lD
zI<VV~O7fi&<$VQ(rMk)JaJOXxM3xl+*jGHNfxp~V7jTvz>cG~L525oJ&`omb;*Ad2
zkEZ1s3w!jAsT!AZwCF59mL{<_JoQeOq71%OcB;ZFn1g~a8N-?{CBjq<f;mPeWuRz?
zC>vr9z=nT}MyUmnjK-xi)_&8VTCG+wW{fe6tE0`7U`99q##AW5c|K@hLJ0-Hgi?-4
zk*@@RlZ4_c?%lk}g#<&ek{1MXN(lxalw#5t<V;8@bjZ0GP8cTX=IGV4+5S~$FQNy(
z^kAg|AN74754S)E@_U}Zm<i1YBkT~PnNi!SaQfRZ1TOE|EJ#8HC3t%=P6B3kF9&j$
zo1>gk0-#JrnH)00DAkf;EWF+`rv1Sd8`>wo!CQGI@EKjYvIr~b4v8nQ<=)98k46Id
zIMM^8puj0!F`2B7?Wk*s$QPwp+v9o|4l$HDq~2tF5LcS$(T&4(#zkqp{E0#Sp4=9j
zR;NO;e11(n7{(EDE87#tAoMtwF&59hoxU+(#&Cq_#eclGj}OE5&?;xG3*CdzgoGr{
zHKpsW8B7TCptXgSV0eVq87oTe8bTD+fPVOOou(IRwHkJu1lo3lS-WRDm><nQ^>`xv
zb{w|@k#KL>H>K}+erf#W87$BLBP_?RYg?Gp&%qcJ=F5R^+3^t{$gyJt{3EelmuBPm
z@#DugjIv{>!q?L$de(D)Jb<-VfEt;@wup4FMUQz@uX-3gOGN7nPM08StwVuiHv*_X
z>AKY^h+NAkz^YF@ukCI}P;=shp~88_Gp1ME$GR9al}qx_j6_`{LEG<Y)_XjCn?Wrv
z)E17RtE<xvALYik&_VqUXg9$(eWsY2;b~eDkP_U|0R8wYq_lGW!bzEyUAU_kH}#_E
z=uKt#XI=T#k9iDqPy5e?H{E&Xop+u+3y<{8re2o*gFk$9{u?i3ezZ>|^q_xq{-Srh
z15woS177<+I*wkB9vMwWz-oj1yIG8{QGxDwOUn$P4{T#AYM(J1RyeQmE}?`Sry5RW
zW0Ksl>`)kt08v1$zrjR$5~2=(Yw%-AILLDkCdgb~%cj2X`;A7=XGzru4;|<F+}Wj;
zNx$fQRUFw*Q^*c1j`m!Cxz4Bynh{fDQ&uyCX)u20_^`uw@hCtIKcF<1&)fGW&pslO
zKh)T^UtJ?50MD|)^+lROjX{fxv_6YK1NHly_k9W)qFI(S%`N}%`&LBwOcnTU{V{h+
zQjP=VckA`k27udU#H1%c{uU@f0nv4*LDBxl;Q$Yz*B>+9b|2!waQMSedqSz;WwDF0
zC}iU<t;6{qoO|@qN57Y@?gswC-n+Z2>GvLe4E)@4owKqCr#bJ;b-mo@5w7pk<H}xx
zm!Pgm92OHIql3j(vsI`aMw@8saE2`k9c|#I;V~&Kf>kGq052|HYK+CH9Pc=e-zBZ)
zik;lshma<n`D0hCN|IPitSw7o6J;VkqDH@SZw*+M2-}^-%N7?g#@dqW7`pySvx!~*
zy2%Ef?{t#%ir4B_GXcH*s@SoCLl-E?<D^M9o9x23<1%A|dr?duZS8bB27<<`SD;`?
zcR!bUK=@QrY#)Jz2)H$^2irYE2NFBiOG?;)l4fLeon3BQwJ=~+A=i4-Yio>9dA)<B
z6kA(D%7lD?@Yp`I-`v6p6CX+jdlIKw$$dtsI^{fo_rQf;NEBV$g}rQ$Oo*You!~-T
z9zd@}Xrd({kf?CfI|Rc4QzZn}ga<GY;8<n@e_x2!ShfbUr$r+gb~f;ir+6H*tCa0)
zbVI}hQ({mQhGTEKKDcADi<bRMF>Y(iP3Q7}Qw?lGBWhEeBzPP*TisXfjyfGk0Zh^$
z_ccXh?G4QD6gd7&j78cR4m;_EJU7|h+#5BoxB%I`Fc-kBaDC%Nec!!&&9N_wz|#Ix
zBA-3Mor60+zV3O@>5MkQR$#IU`N(`>5PRYDifBJ#i}z#GHb6T?NW*eQebmhXucgmG
z_oH*@z359g+Q6Mg4+Lsasx1Q$Vn~35?XL#};#!v{kG;a3X>Lr$!jnBFPCLKZjq#u>
zS}JwNivIJ~Mnzhrr!dpi?58j72XG_ux|B>ph^`;8ZvJF<0=X{>bG(T(8I|Ml*pVlR
zCipc%wt+EvRM@L_fcpP$9h^LQQkk(7uff<NUjZTIiV;UCV^F^5J?}999N-fgyZ@O3
znD@9?e*)l}AN$zHI3eUTlagOc303gLlP6DdDcNc2d{hz2es%4&*HX&V>lCB(S}AXS
z|NGysD5E<}QN?yBqY5t1^PJm{XbiVAso3g0_uRuVX16-3#u(Lmk38~-&2y$CyB)CJ
z)0DBAa&Bc9^Lze23`6X!bC^=gJp+8*!J%u>=}V(1qKv&sinCG*Ts?dCtdMWAafF4G
z)<X|Hq=W<^)I(Z-OA2*P%5zFcga|?xzXAL3Efk_fw24r-k*0j#sevySH;mEG(X_qM
z)Mdoa|6p35On#K6pOOIb8&ZBl0?1E&`uzFxQcC&m3P>s8oRktO@0L=^^Y8`BnlYu=
z(<pr!Q;N;@X0sVnzf_FU8=U(06r+?<roLBqZlIK@Un(|32qAKDE{{StEnLm@ae#Ml
zOZXDbcDtf1a{K}SM|9g5m2a7A2LwP|p`RL!(n-nVa<Gx6DGhDmuSZUB$XSasd7nM+
z8BME`z2)UzE9=;bLvuCt2r4Sg^Uo_`yfA1#{P4pe=)?(dQt6}#w5GRp4{h{?VHlB6
zQ~xU{U5|gxdMQlLe{xBe74vWW+LLW`V<tGF&&p0k)Q;Ri45noOY>5hsyhOKIqLNsq
zGd4#3(h%*V<H<0%FcuBVQWR+s89)arj<I8sO&4dHDWVPpv-S0LjX^IiOG2&Z4G%UC
zBSihL@VFM59NN#Q8V6dfgPr-1QI}(24rTBL2iWo|?7_i<2XW%_`74a;IWMpLABz%s
z{H+#a@_x9&7(=zi;#^o3^iLQsatwqfDU!>*N~LkI)dFA+<BYi+_pl@A%E8N54v(d6
z+ZZPp+`ffD-B4d!0%eJ#70{M^E<y+aLKpX84*Tc|geH*?CGYZL(B*=Q1hq+F&_#n7
zUjifT+GIoF25DMDwR(?uPt~+>443I>udQ1*UG;sy1XN9%o0L|VPy=~uXj@7Mun!Ks
zF2C%HM%NS&e5X2CIc?t*!Z9nfVJT&$;VS)h%$T}tVbQJ0xtsX)XcPxddQH=S>v;is
z1$qN|Kl&WnM?XcsL<olb9&rYhTE{k?78|^bKsiGs(eXqg>Y-$^J`ygeg>(;aoCQ_%
zv?zJncXStO*@Uq*UKFVVg@lc5Fn`KclwW03@aJ$AApy;NozIV^{{t4T1Hf+ZIEjH*
zbFCC@F;`JWb-mH3Yes2PMS^n?spkBj1z>eW-~nKlx2RG&ufP!i7VjiLJ*u@j?$G&H
zmDbdwknF?^Fm;Zv?d<F%7$-Xk#t94)j1%>hM&`SV`gh>lK8S#&=>DBv@Ly{e0E}MY
zO3Ql4Dbd@ZIjdDqf)b^-bA0BO`rAXP6<<{b08F^83FTq^A;uDnlf|CKgs)oRHpfKw
zx_ELVW1+SB8f?J^XR7P&0#ap&18qcm==o)j-6$Zus$<(|0i8h)qK`lWoc>@roJSY|
zn7w@DXeKLCeS{e?jtS6)d*@}sRacB01P4|snD7E6PF94~4yBpF`UCbAb_C}X#|}ga
z2D8r8mpso4*yB;*i5uD_VU9eKWAn=Xix2DCAp02@6E5X?mzveQ!v(MQ|13?$q-3l)
zWCc(dr_&9Dw@$;h{18*A2v9$u3St(>o9#~UAh+3)`gX&dn}Ip4yxVGZ->Mz*@v;;`
zE{}6w5;3zi_A`4ZM;mAxor{h{(^W6Rh3pZ!e_=FFL7=;*OM=mxj4&niuBWK>3$)Z}
z?R8EG0zzRu8v+rXT%65te5njRlTXv0qf{2hureWTQMg<%oL~GHFGx|mNN&18w>-Jf
zZYQWykK&{qTcMl`xIgKrz)=lJc2KunZXLq7+jPpo>L^5k71O6G1_IiemM$E}ELfJs
zxQ87kv$MyUW!biBF)3N554=VuC2J|$w$AQHHAPfe3zeeS1IPCO@4@{4dzdN-pu3zm
zEvw0SR|7~0?;6&Z26cM9WdpkE6(|D%c%(i8<^s3gM~S<t6jT(lj%f6@K*=K(h%E<(
zcdc}bbVq_ZSZ@Z053KaGCw;$?LOlnlIF9t^_x->B_y3p}DE*zZ7tA{c&;Gq=4=TC6
z%(P;$4TA&y(vyx;$^4!JR(xOj4}bgH-xgw^_@89}agz7qg}Ck^**5I~>Z1*Gb-WC{
zko(bV(R<OS(LVYG`aSv!^sjJ=^A)H7aY^jLW*5?Ra-1dmoxdL0#23(@Xq=Q6cf>4L
zI)qCYVvD&zEbT28JT1(noCa*b1m}$52o;<<@VRn9F{OBZVQ%5l8IMZ^hbOG{NmoC!
z&xE%Vex%BLK_*;&daHbIF5`nZEf3krlCWm`K73ueu4|F^i*Q{@$Z%F}Jo#6rbX|A6
zM%y89XP#%;hYs9kB}($=F<~=MEC~>B(oyPsyC`%eX`DvS^42C!5bC;2g0-WrEA8wA
z48{>`776f~=?2JIs+?~U?%f0NkMxUs`qQ7DmanJW!*G<4Kl+8epaJxPLoJIs)3;~;
z=`S(on8%F8{AbsHY&L&K@<02tKO48s5bOh`Zxm9>N4wj@wlA&=0p}6_d6sDYb9EAJ
zQN%8O!&VmB_Ui;PpwqFX0>W4+fZoh<;us^<;ihATzl)4S@+m@IbUq!8Q8u)PqA4~7
z^sAx>&90eI7=~+6kFw|ED<T$gk~RH$i!0RbMdmo~?e%($9muF>R^J_mdc7Wxhnh27
zm2rKjSCTm)I5xVLeDT<^V~fH&yE})TI+BmCC>VxH@-vbV^s;f3_xv$<-oL&1&2LuK
z>GNLqJ{Vm6*0;XZ#TwQqyQXMalyao_tZ4TEd>FkRy%oI&eHcCGr4}0ZbW~d`k)x2z
zaYqn>$C;K+&D4SrdqDHDcK?S{-~o!+hw0=D9X0cl5$_xa<l4|MKV6aripB4>ZCjEe
z+qCBNVuOR_8wONKVL=evas<Z3f+rH)*F+_i$x6LmnY>Yl`7Q}G-0hwu+oDP(FG;cx
zhu<oDhB2~=QXr=Z<6!R)!X}mwwr#kfMReEAPlaWD<SbgdBQXrQhF!AGC>8`OM2#Nf
z+%Bo_viql8!u-_nPA)Gi8etNaZTrjIm`4H(R}8doSY&fDvU!zmNKplK5h{et&Wzx%
zfWoSJ9hK>csGCX<f-H_Vg}Upm-b}QA%yhFf?a&$R*UeyL>ALHB5e8pHZKhFXWN!ax
zD_`?{YW3^3Fe@`7Z#r5~w<%S8N#p(%s){K;Qo*}(pC4Ry*=3$f2r=Vh73bA1ooIA%
zqU3UQsS(3`=GxP~KfAc(2aD#8yLFBGWm(E$c*CjXh+nPD<4?mLg3P@Id*JKnGQ@FB
zM!~pxoT5TIt)eEZT`~Qyn~~;RZz2p!(7}7dU($ZBX<p9wy_{0A<4Rqv<|th*8Vr-n
z9b^GpIvhsBxVNIW&*0Y*jq+26C|`oq9yiwwEKHLR&NIVj5{M6fI=iuLMT#EB#^5N&
z1a9kWL|leOh=S%PG?OKNUXkTaym_XJ>Pq@vqKg+hX4yme+Kev@V1IiWpM1D2;qgdr
z4^M_DYRzObYSU+#dWWC)YZi}!V;&g7caTJDWl(r!nQ6762<l1)Th~KbmGFp;+S)=0
z?}0%!^h)WIxYv>_VIcIF;P7XtlaeBwZkLGfo#>UjcNYqP7Z;Cci5n8jBB7gTM-~?`
z6bid{ztSENzT<wsH(&sz(!#>FE9d5+zF+_t3w4;EyK>vYLaCHKPeB7jIJ^Us05w{|
zj2q8SruL3%mz!o3j)pXVqtyYWiZNuczYX-q%-geP&z=*GVdQlyg<+N$Q?Mz$Q2}Qy
z7+tt<;nAF7)D2w9WP~)$HwqhnVwXB4*H5UGrc6>j@>aB>W?siRJU6PPELE+f`-W{i
z7cyT^JV1}OEw1Nr-^d&BaQS9{7I3XDzxADL3<h4Q)&qOcD9mk^kap4t4C9}4jphgM
zVaJ|Z$sySZ$1I^gM1O|<4rT3*sBtFX9#mwZQIc&kUu8H9JCqq$R{q2@1~@7q<Vi}Z
zI7l-;$QL+J*<IuT02Uo>w5STBJGU4n#B|M3$a9@g6s%9{jH^1D(Ao8SM1q=vf^zH^
zIOmi(oFif_=7ytFL*pbhNt%RfT0qDozN+AA;SHT0;)-y_ShFjFX@c*^njy);+dn<+
zk;(*yyJC@y$Rn>Ti9)g7s}~DG3NV;XU^ecnY5`sjA&T-jUDJH`_~l1wRsRhCO9+v<
z;iMngQpeyjku^oLjO`AZqRESd5LFi<hofS;Wi=d%p`VzM#cBVF=NqLmmlR=uY_{76
z!6d7fD_7c!yC4b6KB#tpUC4SA4r~>CECnCjUO3PSO~)yNe&6LRvR*sK0}A(ZJMz5s
zQ_WPfS@b7MsAoz(gwS0Q<%S550-)elHj_>doI@8b9CAQs#q8|atjP4yp{4ffYJ2I>
z7J^)^zIyY`tMweyDm8trFLS2>&~P{vhFQe8Xc~e}f(dcoEKD|vMLPzF?P9Uf|F7V8
zniq3*SDzkfPYO-ywyAD=u}#JXayF(nro)zKm{ger7c2nAY~uD0#&sRcNIJm?-@g(v
zN(t@$4Kt9WDj~$ZsvyOv=ALC)Rm8Tc!oyr8pAUBNshZo?vYl4lzl4)vynTB#g<Rah
zieR+JVv2P6z`8E*Jfm}V*e6<avuT>z#7SQCrh4vD*X+8c0=1Cp!G=^33yKBRE=}z(
z$P9^G=#X`LiIHm~KSWu?%}b-YaJZqKSVwRK2vnFnqA-Cd6*d>9bUINz5NUyB=RYQ2
zuKF@qp(io_GoPeVm5UG@Nh6dpte$Q8#qlMI(3X<7Eg1Ps^w<FtEfa8FvBwPb!PMWd
zp~!3-9Yhbt1Y;Ps>8MthW+kySOPHy)GKenq7(hD4kO99!y!jK&4A$!aQ9sd@WjLy1
zx)xkGd<l)q#@{2H^Ur){;0_#p;~TjRA%gSgLa$yd*}=VM1G`kLO9(EsHtHkR9rZdc
z%gWiaiYJjr_w6HLUl*KDHlYyRku<jwf(AaBFBQ^erkE~Xv$*O0@}kAzixwZZOxNOY
z$#mnMOn;S+Q~azKzay0srYieN{0dmo%{w$Y3Trp25C~^dhCM2T1XflP6$j8i#2MQ+
zaN=|xOFXDfW+0UsHhR~@c`EUWXW55TH=s3^+qNcyO9M}FX>rGng6=9_KwXg6w&hrj
z2Cf=mIl@$bPk<vCNB8}cg`GdE@BdC{8Qjwp6-*OUh4|cP$vA=!2UK;htQWQwLEh#m
zG*~X^o~IX<1FCr2WGHSc)UR|^8eE~skv|%Kbrp(}k*r+twL?qfTvwN5Sql`Woi$mO
zbl1&^maJT%+f9{k@vquiMT;5D&5%>UJVrU5*fl^Iv#4hIDn-vo9w@cl&yr1Kr`Sdn
zy1ZSxzBfzYtMyCSzQWK<)BaaS*Qh}#A*i7nmi8kl=fzPSkJa8)Ot@8DS#^i}1;ePh
zw6u8N!2?ByQZOy6REavBV<$6OL)`fCn3LTu9`KCVQgocUxBNW?EwqSa(;WL&^d9sq
z`gQbm>{^s$U{@=2;}K%_Kw3;|Fs_AeCds9gZ*0gGl_?^FXE9j{BC8XcN!;GXX33-$
zZpv9B(@89(T03Fy0`-}cBQIkAP0h8V4uuT#xBPa`vb5IiHa=&(R7qd98HZrT&+m-a
zC>5F{sg#B~H*}Q{lJva^QFVjsA*HG$6^XF&YcZWq+<yD*(cQ__b29L4UmVnsMm`TX
z4$87jec8z88-3MS#+dU!y-`=rjK2>QFRXKB&lN?H;`zrtm2U7SSTWQ<`7ew_=c|5s
zK5q~5`Nn*~l5_dEk<UXvl=X-%PFwb6nm&N8KsU)DnrJ2@>($42JI$nMm{<%MBYZZJ
zG!c*o!Bb!Whq_9mP6)2GZjV)EHPv^mo{mFR<N)2y%eUToE6?p~=F3yn$UJ$CAomK4
znam73ZyN8~wNhJHSlFNTo%Uo}8SS964^qvjS8Q7x<}2OWW-|us6>o7r2S1qD|Nmj4
z(bY6f>oz7}Ss`D&&?hHOoPZ##7k!#{MsF+prlHY~=i`Yf#}fuzFByh|%ZDD9o`axP
ztGO94SmlkoA`W{~zN)x4z)%V6hf6(>xRF_{nvvV=vtP;BS;l39TCZc#q`CE8382&)
z#4gYi?w&8!dm+ytA@J=eY~Okpdb{tej4HjNSkIUtbj=gcj7CFR#1^H`t^6sU(CIAg
zwi`z4&$iA(S`4$zEQU+Ar3vG-oOka(O2+@Dn?j$~*}4K|tjgd<DmL4y+BUJOVzH#E
zO9HFo=S*|S4k~^?#iXdJMHa$5_0r)hhR`g#FB9JcWqgV0)4DMAn-o$qLH@7t-)+F(
z=&63Xw6tU+_2BktDv#P0l4*3M=db0+cUuCuau5C@@(~$ReRp+P<b64nLz0AFXpa+L
zPT@STCelccxoCKbN-NpJCTiri(bdQ|DGT%87~hPdb(klK`^#m9rf$h~&CoOsa_4z#
z_cB;kD)f)mT|EsH`7zbYD{7+s6c<?iW0&G`-el?d76Tfn1i>YY`}m6XDAhUFX=*5h
zYPpZPg}ishbw84?R;!+w>KYyYbno81jNM4-tr~pM_&GO2L8GbZSF6?BM_l)eTgZDK
z&1o8ujdZ&^j6c11?_Nr8WbB)CId1*W)(ERPjyC(7=tcBBxwtriRN9LMo6(Cfd*>jS
zXvMAgBuXa|0y%51?BexJx6f~u@J-0zjo$yG7SqLci`4x&Bx=`szy2{hOP!V<v`NaY
zUHm#WuBY|5)$Vp`4=@1Z@pLl_lM=1V4a^bC@+8TvR$WQ*EQ`#QaVh}bJZ&{YQ7CBs
z2+VAT0#yM<IkPbU%#aWY$0s^BhTN%@uIQ(b;s6ZikbOgrbr=f)LdF8uGt%sFJW(uK
zcEpg3>A5^F!|wPg{7i#3FhB@9d+Xt~YyddF2;5uT=eXp{cwfIU5EsGvZk$Mf0m0CW
zx{s{*h?pgP;701P*ndC3ziliW3}g#iYf7_nu@3DSw*>w&jbp87**JDowL76H3MX1K
zmj}*$iY2C~cO$_kj#Ku{YOCNGeqd2*QqAViLB-+}SmVESMWoxVh?vu#90DV#R2GbU
zJ_2$yHKyY+#NZ$~&*Lg%cszhGs8<4C)m*^Z6h1eOs6Q8wLSbU6ScD1PZn=UvEmuql
z9lJ)c#dmlQI*abH#4Mdk;|`U=Jz0KH>e^t9a8D()kS-n0t<JLWm--7l@MteCj;gI%
zp&-I0r9vIx^RmNuM6jaZSQYX+PNv4gd1rYfj^lsR$+RuX=X9O<PCc?h&SR{uX=m9Z
zmSlMkGfeyO2$I3}%ss`PJ$sgD)UIh>*nf>K+NbMMr~VfQtRYV5hV2=Im&6NC2uQN-
z%6HbD!X+);_ee|kADsMC%>Vzb))tq%e%T6l)J4^Gf88TDwp7>G7v!pIwIf8f##>*p
zqiL&)4xp2kM@b$;!&U{#8)H6a6D($}DJ2cEt_?<cIK6J>?z@aL7%6AP^fxnl><1;I
zV4NndE)(1=J?&c<%AQ9<ZmR8otw+MW@j&|a<l2u?=S=OkYr?D<CrRrGxsOK%wB+qv
zB`gRl;Y0O>9sjd%Teg1VfxWs;yn}k~@W)_;_M<z{)99<{Z~lQ#IdgZRW13`20bQNd
z+|)_@#YQzIP;R>*S;Zkrnt6(TTYb=yA@xq0R`7o}lGf3I0-`~YPB?!4agrw4nxBib
z8HXEvIXzIYBH{UYEGbw4U`#zmjX#NYCtgw#AHdbvcsNPTX<_5QG=U$ZMTrL16%-S9
zQ?Nw+fGlE3!g@{uz_yGf$@e7*%Qgm(a{3*MPa8}3-uuR2XfXxP&SHg|9%u3@g~h@M
zpLBzHre1ZOsg)?0X5gCFsXDX9Bb)Fzw}-z~k~~?GBrA|G_E!3r8cTsCNs{a>&G@Il
zM@UXc>y7B`=o!ECI~EBqW>G$iPEFv50kn@ZT8>MnxX+4~Me3P>z8oB!;P0eQufV46
zZI_d|!T6v3HksSI$Ewx}dF$AT`rO_<)~<cK4=#7+gyV!(S?J>4s63iWn6GYd(0-3j
zSh*4rQY0}_EAd@+SRzG3)Jz_dTG5zi>yboj*52IqgSFkY-L-8&I;9?lAGfqMliN$?
zeiH`pBC14rmuTHVgx+$+kLwsXndK#e!K1tfP!0aIw9;cH25~%1@pN~HKyDxV_51PV
z6Wf*zuyzZHZ(CklTV57s&yb95%WEx#=Rz8SJGw#YDVr}dr+r3RC=C(Nq!rk7Q*`;b
z+NrdGfPez}rGGJEH%;!GnPwK+5gR0Z2_-f?vl9nj5<GCm2nS3FB)ovdJ7=aD3-EZ;
z6aWK**GnKMeZB7N(cXOX(U1**DtugxPpC@;*-A48l}~q?Nu6-8$1cEog+EV~b=!l>
znu6cn6#VNqzxho`BlI^HibcF_<#aZ`P`X8y6}@tldCR8bALx>UHw15sf}s8ggqMyU
zJzCLaMZTr9FrS@X*@lb7h2Nw^lfH>14)BtGIX+@^u740iCmY?fdrZ<mT`dvIKG9RK
zGummVHqZ3{HfuS=3|gkbpzU58oR&d8#^cT%2|qTFz3#>qX6dyXio>xXGEW{>5};k?
zj4dUJB}GblD4GN(KrmCx(<ww%$@2BCQth}^97IU?gd)7dYs}v_6;_M3U0>s>WnFJF
zX%|-fE|(!2EgFImW9CnP#bEjXrFBT+RC`D=5rH3wL-WS}ySlonL|)VmQvrZ&7c=<o
zLpAy{oyVNqZt*2t6a6*68ta3QRNM4HE;GkNZ65Y_ky3^Y#sVA8EA_Zp*%MqM6ISVn
zLx@!1qUJy_QBD5E)wq8b-)}T&<aw301CVeiEZZ<-iIklhS5+3XjEUz({n&dO0U!(~
z=5&3iNisXQq|nJ*4<jH%GmM&PgX0t(Q_baS)iBg5-heS?S=I`!SQg~-jan|JnvOXe
zuZcDa1@&);`WVr5xQ%H<9ccfdc5A`e7{4C96_ed*&BS#LEVOZU_O5rptLVYMQ|cXx
z#KeG^H%*-yiF%st<M`YQW~L9Xo;|yIczOmLK-}gWlm)SZGtXgjQ7HSxd!~+8Ds4o4
zmUY6H01ttOllf4upY-67Ri!CZ?n;KvbB6W+&ra>Kby0*v+H#;1nzo}3Ve3=q2j5^x
zK3PjU4W#CjQszgJdy;#}N>*Q&WTV>8S44$c<8!11D$N6p)8cds;kmi}DuDcr(qC@E
zP{zD0VURpUlH`0|mLyPo39ww^Sbj5>*Xi{=%O|ir8X04cj|8>|powr|Mcdbr0gT_0
zKlppSo|KnCk|nn&V~jb+SdxmaBugOYB}rN%>(9zCVv#FO(InW`*@Lj&%7u)+KGHJ*
zmc}rbc|;V@NGdSe^xPxhC7j^3(g*^s*!5;IbP=#TU4J>jrRBdA@0s&o6E7~t#g;pB
zhFGRmsWcYjX2*UtwbilLN)j9ZSz^*#Uw01_`3S0Xw8r222^?H#UI;LR%IZ`VpsJ|G
z9J%VaP#tsy>?Urw%EC~YX;*)G^n&dUnnOp?tGwl(Qo?&Q{$iG@p2fV_6-$4&cTAo0
zKsq9&#?haOc!=;-47XCtgZ1%9h=b>@^q-%Ag6Es{?S}9`iS+cu@K?wx>-jOKrqd8$
zmH*)YR~XvoA30XOq}~)5;NmP0D#j^vNl#TLxa#zNZF)Jr2vmOQS@KVQ^;kWJxUZ@I
zzTUr)r{tB#+`%&(a?JBrwmbM`4%M8Lwap=O$%8mu^({PI%22v3WdZ*8aHO<B(0X%k
zxgz_oN012_6fUIBKyo`reN@v!m@vJf8bXcjw5(}mR1YJ-O_VT!*KJ%WIHd}(aTq9J
zpDacS+@NSpg8lJ(`OzcXSZ=j530EvI{8#thdoR~NCz<1~Y1&6ERRvYiHT5!&Jq<%4
zL_V+6CZWDU+}{)3>D{LD<^u-~5ZyH6s-oOio0y=4sF%5Uz5jF*R-H%qDAA%p&vNbB
zs=}e@i;1u>dgcyC5s5@&;nyvA&h>04gWx<l+ntzjL-g#oCa@!FSv0Z(tJP{*b~T7}
z;pqBy)1UUu9g16?nVu=X$1u%cd?7Fe81I3WJd(p^8RK%Wp1o2=%P?BTrCctj1(9x=
zdo639)UN_mpEte0G(nhwXFgv)hdwXHjvYIcO~f|EQ?=Aqyt!&rc!=#f_Z+FFriPA`
zT#C-EI}A>d%Q#1k2e8eyU%3$Q8m)m4<6BukkHCl>bTsBd7q8Dd>JqdTgm~pbv}?Q$
z1Ka)sPZ%&5U@ntCM~02@UU(DgqF10_^-N{QH8)=kb<gjI&RsvJARITao7S~)rw%{N
zudL(3LwxjUc#y}NdZe%G#sejsqnNH3HiD1BR}e=XbPP+zBXo$!TQ46+(>(-OAXzTu
zHwP1hdyMpnCkkwy=Co}y_wfE_z*ZKwW?i=|F)`I{PfZBR(sc<N4KEyp-h{xCe)WzW
zI|?~ItW50RVgTCWV>@^5T+{$qiw_AC+j0fCU~Y@0+XR;MLZwoOwUTSH^rERxOz37&
zw=D7HPN&n%-TAUDU6_2e)9G|pxheG41J&{kc+JdLFoG|Qd3@1jm@_!UM}sf~xp|lh
zR0nge?-!9YJMDI7fX?u6YD5#8#ZsX#t8+%~wa$z1R(<of9aX2M=4Pj-s*!DP4#y#C
zA4r_U+d<3o^A#0K`tuR;`E;*25f_JNy)Qs9o@n;AVnnK?Y;A3QdF%x-$3_M?{e;%r
za@|A{_;HsF8PEc$BT@IWKho~<W1$A!YOJnLMU%5Nvu<)|5xKT$mLjWey*P<PN*K+W
zu{H7ua`69~Bz5vB3D0A@US0A4%Vc_aTeDUwWh%zfa<}4n)y}env6_`iwdS_v8CAyM
zbt-im&tr*VL(@v7G%b}h&A@aYgt=S@5ax0rtmwGHT;C`UJ3NUyT0CR9Pq>bvB#9!;
z$zbVWtBJMHojBo+i9^`O2e%g5)Nig6%;^A*Yz~JW5g1fG$g-fW01M&`hnq)WKshGs
zO`o<Gwhm%Yy}_WC0c5qopr>Ho&i_OFx*(}9*g%$gy{fk1cHg!_i*qZqZ2#{U>PcYh
zYOl9Mz`m{~O>|fzRN&<R?khklP-~FV>CP(NPtrgUS7agli<8nkyaDMIH#ZvZCyRua
zL=q0nfJfX)#fV#q>}TMG#Z4nO((QQ9ClV+=Hg=93%WDZUgG8WI89iapeN^|i7fOnU
zHWwH0W6@IK6(KPx@>6^*QM91?NBmPBoLwg(MW`YRs_%7y&?0e%sh&IY^?aoLS#3ju
z;s<Sh;eWZQLPK?NF$XEd(x3VXnUGmWK&T!wnnLQ(6N9{bT0_J%o)M<BixNFn8flTL
zjv@g}l{WG6;E$Myv6+7?W!!PAl?t5dh7mrnG+pY~=D&iqk>f>5nE0mE)m5WcrQ6ut
z+$7>sg-;(ndbGLOde-LVW^=Ro)L$<Tj#3<RSN$iqTB)RrJO455bQk$_=>SgsuUEpq
zPSH(F<)!0;2M=nq>WWLEDDvFc@t!ak48lQp#n;?p{eC~}hx+%({?g0&1BjY8kfOwS
z!?_K1(yE921jaI4B>qfMmL-9!3Q-p|($nhys|9Y6N`a+A-}f7-?<>w#7}ioHflyFN
zGAY?pE-Q{|ZCR6hxn;YrUf0^PHlNguV(1Sh@D=6zX~XyZP+OpsWKuGz6ab;1WTh5{
ziu0G@)lJLNdNTie*S7d&>z1YU(N#asi^-r52x){a;>MkL3|;HFLH)Wh14ZO2JG_oq
zniVs~knTEzftoPH5J3G#gRV_@EYS}4aJw876T?2kc2Lx=*rB-$lxo{KTnT)NPb58c
zo|jTTsQgN4$bk!aJg63O8;RN>IH&yVzt{TVxfZnWUzWRW<xN6dSrJ-zr4VnbxGo4!
zF~D=&q^mY~Z$){<Zmmz7>$PXiZKGr8HuTci!(lc=3e`Z+<Iw^a#r1SqfI1)Hh7k9n
z+Ne#5W(!#}f>ul1-Nz@Rg3TWABL$~{DcBBm;fLqkx}99jrO%!9)ehSxZ)}^Z-S%v+
zcIo!!7ttM#<J`GF?FBDz2R4?QKITvK8I*rn!5y6MEeS?Cpd|(+V@4%LBrGxecx6(K
zig9FqQ8RKMMI|1&o(_k9rCWOW(WCW_(<ZFAJ(Zu3(Y})1a#FU}e4YUg&t$j!{*LOs
z`F;1@m)-HsGiT0(7&7kPT%A9WQc5piRXe)Yi<}1&jkJ1KpmdI454=-^gn0M2b{rCN
z%7e&ZM-jn?0kSbeO^&Y*{f>8bQ3dtU5p)}R8G5tMU~r=ktg0e8?MasTwj+vt`gH1n
zRke|6mV7SMB~E|bZ5auOJkmu*v6zWm@*r%jx-mN8W*B=*X4Y=G<(6zOy+Xgg_uhM}
zcjR|C7}wJy|DIqkuwDCRpJ(9szVmfYTX^{I%Sg^Xd$O=5_~%E`l1CsU9E`6PBe1t;
z$=p0I;VO))wF<$$r|}fLQvjBIO|@RDZ5N?rM?VRryT%NeC7TfP6TAi3W@6E8ynk^L
z7G7^FnsUEO%U|dVMc^w!h9G7dA$JOgjma5wh#k@hn;YgjO7YVfZiVjQZiw}PO?KhC
z<1pQkFF~(BuSM^mgcAxlxee}mLN394@~E^7pdm#g2gU*h$jqfKOMsn*r}O|&+yyEb
zi^gV4(5I;l$fOAIG$RSQ08GBHQIh_MMX}aeB@t8g(6Rk`gpD2Wt;#Vm?KxRe_iZPv
z%J(%IjWX*~5_u~%Ib3BwFP`p8DMWFzo3xr~w;4yEQs?Bk{R`)EYv;?G;iPN|JHfCQ
zYZf(|5mQMuvi;DBA+YQZ<^&f~-{Zr246zgXc96bpFvtUfrgn^_US?%BqY158M)lsm
zYi(hnGXF=IQ)Oqk;@n+0l;D9LKjRG1x6*!(w#|lQoVa|*xgJ>ZxNZm;CvevYi2|Gs
zQ2SI^ih`L4paiLdJvAwd(Zprm;Z*Otuj*VITUNid=%@Z$!_3x}u`R|pGmEX&Uf*(E
ztJhy`EgHrdvuv}qWi0#ts#|8U)witp+U0t^ezUpeueZ0iottX#sYI-<isY?J%S*M>
zE@jLZ+hSQZ_3yBInq=APv16-QMozoTvW;PkX4yZk+vbeAr)x{gOK+Ww$K(7+tyZfY
z$;W-P(HF*utRzKwzckL^X;`vL&>?=EvX#wmPCIwB#%nA|TL0Ee61Il5cIMMVVHm<o
z!ufgFvn+T1k!BDA1Yr|ya4l=!6?rbsihlNpKVrV&e;$&vY1UY>3PWz9v8G$Vm8fq_
zF4AMkXoGtUgK$BlBNmN!N($gqMNBxTP+JWLz$CuKceKK72Q)6+QOz~02;qUr05|-<
z?x5T4wgQWi1B8MG5-B^=G|)<#KvJp%^$B+<FgvyZ`wbVsjV;LtXj=(Rh~PZ%Qlaso
z$@L#+QVC=Gu2l;vap=|R5PBwz6AT+mOV#<~p)FNq@~f7W4?kG99?oT1#w}$5u4B-Q
zM2ae+ISsT!ja1-D$zw(gvALke$Gpzl7-1Sj54C?}3N9rNyk_V0kt2NmQ!0$DQ`ZWj
zHqAoM;Di!ZZ>&5?;&I#CwazW;2Ggmup&9H$9_t@jqFbT+cnO29gxV%ZeKR=UwOwqW
zt5@J-QbGH6F)S`+0-KBX(|xq0x9O%cJ{_4yKH1h^r6tp2PB3P0FpKTO_;<jf0o`UL
zcYyCqbj*HCfxrLY-}(NKf_I|Z>s5S4v^GV(!`XqQ(WqO)pp;4cN^)Y6(V^ok%r64=
z-goe2#~D(#cp|CziBSr0tvebmFEZY#7e;Hs{7SFaJ>dZz`u@T{!SvW}>j{E)<ZT%G
zV9OsVg`QvvxH|I-XFx4r2iJ0Euf4X7$@3Uv<mk;5(&hDZ|MOA-@PP0JZk@b_5JH|t
zDdo3sQ+8P<gYj$2vE`2>1t8&*YB3I=Am1qT=_p4D<g~v}gu_Gp35k)n6D(16>WB)M
z7F|b;z%$idta76K8!rZ;Ayv32I|L!v9JP(o@6Lwt!;j;$A|#rDk1~}!8L%SR2T{t{
z^?K-NMVM5S3Ss9K6P68aQpyR(gi-m%{${_Oeo(W8pj1gllybv*-TwL*VHb3wdff&v
zZ6H=z{1_;`$bE>SR=Nq<w8<B~VF_-~wM`}T-$;e9^Jlsxbk!?J|1R7Da;Edc6{Z*W
z@ff~<w$KhjX>XWuVV;hJDA#1LS)rzBQj}aHnW$q@2R5+<t;eWi5W~ro&A$#)SY18&
z5#Jba+x<qvl#;Qa+NFzG+mEzjY$a93=3%Qf7;@0s`!lehst?~v@Aa>L{S%KJs~_cB
z`GHR2sPocFwGvbSS_6);DQvDx9)@_!8lEJm$Enihknd4VDaC5bczvvxMH0n?hiXiu
zW(IEGqU~$0xdx=P;rm9a6V!F-W`nvet2e0Y(ng)RF8fpP)ED&vW|5Ri8cgQVV6g;t
z&ba{2^IdOu?(<v#r%u?iP=uP2pehLbR1QLjzjr#0$CPhHqjA}-A_*HMIIR~tvpOhY
zPS>O-N9DNPdFh+(!gT)E#if^SdtN?SS!tc0f8+AYFP}l<z0>Iwrqkl1J+GrdbY6t*
z&E}UkU<U8H|Ni?g>2v_!f7e}inO18x-@Wq4?(S~R62#Fzjmmt@e0O)(#z8TL)goq*
zM%(RnPw|M05yu*tVrOQABWyesIiM3~V+=EY0UYoWhKq$u=-OalazAR4THu3!t+&q%
zT%Je;_Zyu4EQ~H$C@?gd<^53#fnTczrz;wK)5X<oucIUAYV;Hh|CByzDo#`g>NGwc
z6qz>daIJEXkJu<-sA4AFa{O*&+TIQ~CK!xuz~&krsA+phAro@lk>E}Y40GwfC3|;s
zria2lP8z<iK}(tEHST{4`$JQbsL1NL&9m(P+{L~B{s>V77x&>JyaD-WfR4l!DD!#Z
z_Sqy!({YU^mlVp{CX+H5O~6G=;fj*|)^3qpO#iU{a)8e{Qh~^_1<Tmts)Z!DN>E}}
z^I@kmZ%8s(b%M=}jSbvfCEQ?>g#rPM3ME@_B`CoHrJ^%CTh%0qKk{z&2<Ke&Je70)
z2wPhS9^sOtRcB{AfHL<a8`K%SbpTz1UJ<Wfq@8~{E_PNqe_G;P37#g6FlhVn)4{?g
z#0On>eJA=w3~^-XOm7|(Kyd{9o3jkX3Box`IHsCdElo^J5Lq&E#bVBoWHP}m)3{=F
zbya8-^Mr9uCJI=F7d{OSDhwhb`Whl@H^-6|cLb&Bu3fu0A5#+FLC$v(l2R%<v4$nS
z%c`kh&c2S4{M*-iu=&b9+!|eUkzb8^JUJZ5_ySZ&FmJ1|D^C)g@AjKuy@AmtU*Dj6
zEz|L4dyHj*7S@g%9^6E(Q@Dnz$`w7IkF+SC*Q=7MaxcF9Vk?@%3mb==&ISAIw;fev
zLsjK6D-`PWLV;BzRpp98qkJ|HqwY>T%wFTKn2N%falYcXu2VU0Fs3NxuPkpHKpzKV
zDFoZ$Qh-~5E9153LwJ{KMNk+wu8&;LbEE48V+s_t7@<X--TY7a8Lc^0<kqzicXxu4
z?;N|RLkJz1Orc~dy@vwR_GaQH;mUQAOkDLJY`ZVS)mCG+H7VGgHcWD;pcJe-h3-4n
z+0|-ivqV+cAv^ow)z#HqtGj;f=JUdrh3MP%S<CTtai;~x`Hr$|->Dmut=Wco_J_$b
zR$+T~cKng;`0cNHm7fS0_VYI%82r^pnjP`Pz`GRkm3ln^h`SWg(zi=FwqzrZS<1TY
zQ3I*haQc|XFI>1#dNJ2gzQ8aR@zj~^+p!=*@Ml|4{Z>`9zPe4gs%iHzU%PPOLV5ST
z4W$cQ2qN(I?PsQ9fiZhzNF{X;#hJE+c=RVlH075D&Y_Mzl7>N_!&zj2_}0?swLNDM
z^gjF9&lcR{?wf>(?8t&~(iBm?Hn9vw?s2y;{=L=FC{s9-Y-gFKG2Z()uSGBBbV70!
z>PMI)86U15rky0@Mq+wpeu^ZwV$zOlW8oQVqz~6R+m70U5`5P<nK(lglFL-)ggEtI
zyJPlgVsf!4iq_Tutz9(aVKeFtdx;0y){6$cUVm<e-B2>R2N@(T+S&{;pTzb-Bb|(3
z#523K4IM={j-M9Z$&+PSP<I`;LB}I@Ky!`B;O#DlY+5H-$`oo~t^`1)$V?~ghKjSo
z;>2uP1oJB)m7m=<DHR{CGZr4%aaO&C%RCH$cT`ZiaMe{;5yL;JD8N$KCom|c=?-Ne
z!yQF2$N!<Ks%_gMwJ)E;4sNN2rW&z@dmC$lQd}zK5ALAJsU8n{d9zgf<>}L>O&*-$
z^*7U(P3#H|Ik=4lu1J!kVaeFRn67VUs-zl%oGX(rV1)<y$=c}d{TV4Ovu7k}GZSFw
z?0|?U*fyADf{&%<RSdzMFuxnRs4e8yEEKu$$o1y8V*Ez#eQW+`AiZLe1E!0Q`GM(k
z;+V6u<Hub3YR7S?W*DlLVSGg)m(%Sf{&w$%6-QNzguGAJ3`3=kp%M|9xnKD|bLlMy
zj)vu>#IS7h$Ua@yh~qQO55CR4?`fK58!8cY0idQcR%{oTRV`M_)N!~Z9p3H{4kdys
zG;_M=AC_be-A7^LLlHclkWsuZTM0?yv_tI8wA&`R6*v}3sqo>c#kAGF1*&q<H<J*6
zd7QvA&0XrSj%+9-TZ@j;D0{n3Y`1}?%D#eU<W1(zOhue-kGez+&ER%`<(UBE;7tG@
z0{CKr#?mRPa-c}*<+72yLa6J|E*6WkzM`rIx5*L=0El`VB=<Xc#?$oI0>(<tOkmsx
z$_enis<M+88Q{-SPjHhw=j_Fc7mqQV6y7vwf71v0m*YVAP^Ag9V}KH0wIMRC9;i*Z
z9#fiRkQGIr9+zOb#hz<h=YwK#syFHl`7tC({2IVCbKBCiq6Be$N|q=-eDRya!8mtc
z{j(qyX0RgV&L#CPZ%eT_)j#d4Oyx`yip7+~zAu9j{Oek9QGA?yzM&!C2mTxA2J{sA
zP3PQ5%x(eCX8b@li5-eMZYQyL%}*H;DvyXfw?1n`VG$(r2>rE)`Y(;?6fXiEV90dK
zH33Xl!4-??RxzZDj#Y5Q1n^yV`35HZ^I%N3o{8f)CX9h;f~#nn;vy3~be$1piIab_
z^t<n;VC-_&(is*efIO$C#=@bN&eVm4g#}HD2Wkn>Xbm&YGiHC&Gm8>g7S;cxd*A4{
z(apKj2@GUc@ISW)#o{E(ybt8flFpPf-{Xn29E?@IfXXnoSYu*B(t^T~--4p>hV~Bs
z-nMN~J<zP&A4WO*8SAR8wl4$?sC@&M%8oCpd)|D{cajWZc|UNOV#>IAJakrN)2%W}
z?X7pe``rZlSd1Y*1^se5B0gt!P4e(|mtG7zl-HdkPOfozTV11J{MFn^eBW_WemQ#Y
zESpxxMUY}%U$bpn%s;TXxvA7OnlA?(yAEe6A~I%3j565A&uTpzy#zflmVq7SX_8JF
zd*yN@T;P2(q?&>Vt}saum?)Kcr?|O}MWSG7sEGT^#atD%?=Xm|!b6V6r|vemo={6D
zE%xfBu@=#xtAS!R#CX+<-;}tk7!MhyaP^@`=s3kmf=g~uVzH8`TQA0lQ4%5*!E=gC
z@x?=|Fd<msmD)y}+g@xHx|@%*h+<vI9mCnm*6|g`SvkJ7!nkqriWNmdYbLd9A@*#}
z;|kN4u}Zx1FbXk883xlK`d8j&+`U2wVbmwYr;I3S4f|BLUJCn2+r8f0$Gd_0ysEp7
z(yEQp{&EQfP|nMsIKnwuubZY>b}~*>u0J|}zbz&do+#D}QDbq+!W2{sR=jnJd+5D|
zdV8-1OUI8Nf8O+&ue;%f8-m4qLUM@*mxvtz4O=-_OW}sHC-uo;8S4uUJ5#_X5JBUM
z1VST#h~w-pYuCz8yb76VN7&ul+ndfl`_rHPv`^!-O)2M@LrBDImW;A6IOWHlc>R{Z
z_UFD8B;yt(l=L3m4j)p<;V7@nRo7U`88DsXNP!2=bUJM>SqbwABaeU><>^x7>{Cw)
z5-Nxr?m9_Y=geR+r_lvvXkrS7J$w*dhi-&(3gno1{YXn+uJ9fgG?_kkP_!X|?6K)q
z7>J^i(${k+MLdqjWf^JFw6AVX5>mm?OKK5Q+5_Zi#lkG<0aOXm*Y-`#p9fC^CX$0Y
zI|mcNV0z%tFI5=_FVG#{h;h`taw&6<X1~#kI_hRiS9U{!--we0UqqO4VT|Au1Di@1
z?6O`{G9qKd-!4J${jAN@PxBM6KsSI_1wESNbcD-wZ6K7viU=hi4TUb#^&wf3k_KmE
zk+8wMWo<%}Mx!tV!LP_=jKD2yZzd;fbJDOy_Tg7BS#m6py~xewgICZC+ecHUuV%@+
zk88W$1^+BbL{obT*=B;$Q!&;|Yv<G?zOqy-`BozgVC*{K=YbcPF=^Tt=qrMJ&b5E`
zZjL0gS9Cd98$XzgGyB{Z)pou6|13c<sJoS%ow?DTsu4-TlT$n4N<uW>a^m{zc=NMB
zU)AweECP=^s1cExsZ^+G{hnqESetg1*0TcysoLXUZMy_fWo`WIKyO`AjeRo+(T77#
z8DX;Jt32_&fMEY;KMPdJ|Dz`W4(Qf^V+F{3&w$;Zuu%kUMSM`|JMy!l<G|i&><4PA
zCHVoFJqEs!&J8!}CyDkaSlmXX^bIF%8X<*;3O3end4m{83F8St>=$*Z4z!fL-?A)~
zx!AmN!t-#N9&rqVUFk^SWyZ}s2lko`D8NU3->29OJ#l%wFyB6srg9ON$;Y}9!Two*
z%u#a5Yc@MGb^Mg}ui$;~P1N)iJGVp={g0{NMrjITsYjF@`B36S&C_X5&AHux?nh<s
z|E-WD@I^-~6t0|`t6(TE%%6?R3`^$exw*1bS!kUZzbT*3U!SH`TuqY8U?$|UVjqhe
zRa{+c-{{#h@RH11T7G&ul@`);!7g#j(q>x=3sO2aR~UB<!$_BwQmNHw<c!d8Y&ERc
zTT<HXwy8Ob5R%%iPKeQx&uP}JZ4cwB>vw$sl2{#itj^&p2XP0KuISrn>5A`8c0iko
zY_6UaoPw#3peBm{ec|eu_rZFo7gF7CF##0s*pa4^BH4;uaY)f|KuXgcJK|u#8;|4N
zO^K>;addYa$3=!?BjtvPqLFbWt{v5*LFhXM?ISf^a}0U2@C(CAmnZ8Oy77%+bjh-;
z$oUw>+qa)-4F>IzcpS)oI{7l{|0$cNk5qP9Pq{|d`Le&A*@r_vR|?c7<4bSG=b!~<
zG)|hP9S5F2$SnAuTJMjS0D0XVjW!J*8Yri2V9<8XHPgN<qrn!8BAT6Y#<ZVWxb`AN
zKX>lY>&L)P&2=2*S}St^cPCsJ+CBQrtwXEJ0h>nV$GzCn(@O{3&L0th>W4d$C1DK0
zD0f_~;=k+GKv^N9$Y&!-31ue>ZP=1!kna$5bBBJtUiZ`-CO4IzzpFQ?^v<FeADLU)
zw={P&+_!HZC-k_D35PF6ALh<ON;l#-_KJ5>a-(8bi*6z~r@Hacd;iHJ`&-yPPDxv{
zoEP+ADiQO5*Y`=HN3BT&RWlYLBN5Su6|sB)o!{^>Rc}6V>y)udLZL*;Vrk*l@rUT2
zDW!J4P~K7UJ#fmlW7yz(rCprs)3vJQ=WPp2k#EG<5PqrXc`8vIJ6HFQeFBs;a&N)k
z%HeU-G;@r}x@_C!yi>MwdE2maWy@9#sgO6ovh(?(@9U<S_?9Irs_T@hp(&vBIQ?L#
znyhb@MVw#vxC8^Zh)QS<S;i|$!A-)Uh>ZnQle!`}sIRrRAnRtM?#xUf4}LH`)9cMl
z2R`HrGc(<WP8pAF+a)@$u2pJ5xw5d(UD>sJW)|bwnccfqx(f@HGQjPOQ!QRhXH5XJ
zIX7E8Zgo$-x!wnX^k9dAb?t{(fDM~UJS1}mL-d!X`o6uETo0eYaf+kSXtWui)bnmK
zsGx=iro;1vhe_0T#KBl^v*$j1YJsJ{tDfGuvnlrl_uGV*VTg)o0?nZ=I*wVdg7DLM
z2e_rs)auNs6>8=cU}|SxxoB~2bou4v8{Ww^3gCLJg)?UsTAm9)H9mgm6s+OimzC|H
z^p+v*>oBP9>T$oJA}iM)9pwjac#~>el8i>9UdOmzZ!`=^;u`HAd`cyods~G|6=-5L
zKq4#U!bb3D#4xqK$h*{`{t~9p0tBtA1QQrAK7r;7%^+yuKyWbmGn?^%E~}WP+pm_o
z0l=I(yK-NjU!2f9qfzIk$(jADrr58L!wFS16l1X5N^^UnOh9y%Q&$@(9?$gW%%_Dq
zKGE(>&7s{WpQ`YTp-rCWc??60?ZD&^kwn=C>f2Vg(@biSzI+g%s?=Jp*8wA3t1q`C
zmAVme|Nm!!YVP0v_~ZNcn<`kp{42w59}u6Xpj=VvboTZ3?ps<aU7-LykKK3prokeL
z$mb;~pBD<Tpj}0${Va_iEf3%#C%W&-6wsP*q{@C~JFDVOxE`io6(6LuJ7zAC1*D>l
zTeDyY@2A7&(s?#7Se9ibd51H(6)u24kSxRCQhbT9grrpEtx<4k^B&}SNB^ZcaVOF5
zWrh}Q^UOjqL3|vaLx3JcZ$|G&8Cpu5`bpG(M3f4uw&JD(q}w|xG`QtVnN$JM9xma9
z1EcaUWMc{PbxUqJ^X_S1AdvtdB&aJS#Br}KqzU{(7><YeD9Yt>(9cCtex5PM!=M|S
zb49@*aS;GF#GGF%h6h{-q3i2L|JU!uJuKV)zMae2)~6>9KHjw8`Fs@R$HOqpM^PU7
zVR%s2t%3UVDx3?0Lm;>WSW)Ku8ekYJs$gu*`uMev{}0aP?C;w=Q5|pU!Y4l9*0uP@
z2cvmv$59c}BKjQ>g$_J(d%kX#r8RgZ!zl~ruX^_%&tzAJ@xR@6+ik?M$ZcB)B2D99
zG@xlGIwE;_#vlCP2L^!g1Ac)z-($%jZ<wCrUW|64B)YP%EM-u~g#U2lOt9;sz5GY%
zGm<NAYzJh$+lI-3SLW2Kj?WGaTkW}`!E?hx4!xupDtM(mU?_!NBD#;3djh|M`!eC9
zQJfS|<h~GvQ6himI${gKx6mx_53~(MUi5V%SU;oWgvXvcA=EJ-mmh|VWbJb&Qz;7R
zA$^2zA^f^$C<6F2P=y<W5%X|>abVlBEUJx}SQ83$##`53du^t++=^qlomXFT$9Bgs
zbZRZGl;CUBp@nj}FuSLeH-P6jXuYv7aITqNsigUQ{QOtG@|9EGm0Bm~7r-zai*uQY
zj!!}LOLYZZ*T{gpz02d9UY(@;?sd>*HJj-wGLbYB*R`TE&7>Z;Xv@1bOcjyls-HA?
z-;z%_23I%$yT~19EnNpPzi{N}xxYhBI$BX~mWWE>VYoGJ`SczE$Ir}p33T0>WpR0V
z0IqPuId}BP!aNz@swfy#&oX{JRaJcRcQ@oi%L}a;F0YMg3oWBJ#%K?k0;6_QB9ek$
zv~e%04I<TUrU|s|21{1#pK{O;xnd16qx6y+>U_-_dKp&_eL25j;V7D#u3$-)ufUpR
z2K|k@@|>(6#nPFs(l}~n{?k(?=(h~_Q6OofmG<2BN6{P6ThY5lr!gSOF^3Ly$gM!f
zQKHKvT>*_s6OBeh(^k)la5^voaKy1^Zj&%<e~f`wQb{`IhcWF~|GX!3M}Cw>lD730
zqc>nBijaOE{*e%(Bt)rHl4V)XN(!Y)aaosTdG3LW#o}_YxLg!Mh~l&mLX-}aN+t1I
zQWDwMiW4hyr;E8<E|+VyT9k|qHyaQ-jo&^or&sS)$>oeG;+*HTmRmQB<CpKG+0G~W
z_jEw8aJrqewi;z6d2Vc)7A2FpIqt*8g($=FLs1m<jcl&0ZC)8~_6{Uvp%1@JYG{!^
z43%R&HvxH^#cv}^yBVp~YEs(3JuZcH$|Cu^DpI9Kj(_!m2jWVnLA}@OpqTR`UbQjO
ziM@v{PO+DZS|d7)Af$A**f0o!)s1^3y<UD{*2rrMu!mW_FQ-d0o%)a4SF{ZOi{<z1
z?z(+2A{$uI5$d6WbcYvh9}`QRG(1%RRjCoDek?8s@3LI@P7}FoXMpi@`{4B%>#Gmj
z=uR|%mK5~yrxDrQONLQjuNnHcJNbs7x3aS0(+OFgp#F+1CqKf7EHk`)Ld$waG2CZ(
zcF0-QW@*xiO(NiP1^V&o#jc@Tj0Kq;$eLxgEDx+t>6~eXUN@1qKW(-muoMpE4#$Un
z=>HJf3XES2$xd`62b+Vqefdz^tZqU0kP#-i$GF}1YxwcCm!uu6M|<5syoq0?xNFxg
z!?MJw5JP^)tC=Tec1%&mmiWxh_^n|2+4O$aK9ny(yjBR)Z^txEb-m}7N_O*=J9zHE
z)=O!P?9RfbeQ8tsgS(=2Sid0X41I^>wJn)Jn37GM3?l4&)TI>uu#88L*S4{*)3kBC
z1LWB6<n8|&x`h53{TEOu!2;}sE8zmX2Hu8H+__7!GtE3F-R#*ZcV2tAW*Vfpi?qE}
zdeoFVr@7N??aG>ICP+!!XK4|Qxc0`6wEgcFiVFQ1nf_SAXcgRA#jS~~56lT5%><VB
zHkuJww=>)EX`lRBO%W*L2e!~0e{qjM_#U`t052i)R{e+1_z-?AEaR)`#76uXDo&tC
zhjGZcPR^h^v;`{uYOKKw+NtjosB-0T2oP>ZX|vZJubI}vV>kbmq=u0eIZ(Wl>tLF4
z0de^1vp?`rMSANm_3ILuZm~)e1F*)s|G35%?~-Ym&z`|QMnk+Bz^7&Ye)t07XuHqt
zoNKQoT;@>IrD-Kztgn{lx5Gr1Hd3Ck0l5{z#s&0vCuV6VB<1qo(tR&1E+*BA5ud0t
z+v*^7Z3`*RxJML@FM8zE`uaK-)cVeb%LHdyEP}xWoo=<g*CSeUW{5zUzO_DYK)<QA
zsY&B5mAB~rxmK&C`KPS0qlH&G$gsXqi)$vdx9T2`K2sOgZ4!n?RGgT~x)~870~wh0
zBvfG|>d=FltSun5gKGzrB)qp?XEJ-0#5D?{)7@7HjHZ|AJ@vY-EX$dHsCN-&+h5_G
zyvGQ*OSh$LZEbxuM<MhyUq1biX9)5^fg7{{8<KP`is+;Y9yh|GSKqj|gQSEgh;@YV
zz_Esi`%Xv-wDCdH2m_+{lyKh#>JYI5+D732b6N0(zF_#P<J?g2@l~92YVWcsFV-Pd
z>=O(W?=5w`QK%+Kvdm<O40K(WCtu$9Q=Hx41LXWd(baXgSO`J@zYky5&5bQz$P)%^
zbMKqAV?VD(5*`1@aU7wihBCquO%yI1mY00p)mLAwGqSD`jeQ+-yRM%PE9Jn(*wb|n
zV>>8U!t=hsfX}tZzJU^NJP`&vT5ttx9a>?bI|dii%rdw}HA0jv7{Y8TPlUA?pR4IV
z0r(86l_o_Crb;u}U(C@%5iMS4X3Zo4?z#m)l?ZG94EA)Q>!Ugc;<Goo3%%B^h|xu!
z&ZPlrAwe-%<e*X`EEi@aXUbJXEBUnp0a5sj#~uC)-W~}ulwFHIuUtWDtOBysSIUli
z$NrQU23fW>#<U()iMcisL?!*VoWrZ*mmNq(roxT^pib7AAoYQ=-ROt4g<VEgc9?eu
zV^;(;n>r71zQIl>vy_TxD=wtss2%`v&Rp+-x?h<5yIYn;x)(aJ)vM=5bSQ7!Bpqp_
z<F-sLR1JiOFjrt7Hp5mY9SBZz+ZiJv%cXd7z&uCd9v=V8?5*lZaCx1IFmmtuq-_|o
zdiQhsy|wEe=yto60Yr{*1O}#UUv`fD=tH^4k$XY2ImZTFw+d%p50m#C`N_=8%x(K>
zR#3VN#W%hmxBMN6xP4XY2c?Tw+s`%6tF}Wu$c+5C7qRZO_rDFqeSZS)5n<F2{O=k^
zfBY2u`DY>g3t{+=9{fL6un%jKll+^&kJOD>{>~hUY^L=9IO+cNm#pELAo{;wpNm=k
z6&{q)7v4nNT8F`m`wzu`zZ+jz?kVd2^_Q$@f8zCo_Njj}^n<@vcwqVw$=H$oh+Gbs
z)nJGwV~`&#2o<}O*w)-V9LW7t^oYm_U<JY9A0^a-a&lZz%=h;^JR~2zE+7sG{|BdL
zM0|g*(ZNX}-9Pmes#xCPEX5Jwzs}!__x{lSR(aG38htOsq`gX8rQ5D1xIkuAqfvw%
zHYu3G06cNR7iSG&y!Wqcf0r6NxuzQNCGjiKGKIpUy}#Gt5sZND@A-e?!GFEp@k6Wh
zGP;*LOnNFpETZc?vAMDf3>YC7LVbKeo+WrybfIEPQIp{)6bV-E9p_Oj$ku&rpE0T!
z|J{|V)&0a?>buQm({&%ARIIgB=p4X7_acQ^2(qeCM1v6h93j{|A4mI?ID|)yU<B?k
z;$yZbHomLMIgf@e4Cm$srj|Jhg0Maj*nY(Qj`SkNz}4uzT|h9O2o-+a&ty*{`9Deb
z$5fZmZ3uNm5DcYAV6^kGHMPAZALM~J<2zN@K>iweAQU)9WKaG5ox&%!LAd_2I}`@U
zXan)wFYopVbMX*stqtL=Um|4Zw9zIy>{B)BB~;wTD=;x-Ez7<?r<DdFY3H(Cn}&dQ
z7CvMD>mdw6=y{A#YUN$e5G0zO>2zkMBO(mX&0CZb=KaCM($awgOP#nG2}{V~Ok>xs
zbLV#LnwSk`VTq^-1I&q{;>pSOQbXvvu8YP}dvX%13gP(oT~h!E({;<7>dwkcQZ8d$
zE+@0w9G&vA8zUvH0Sr;YwjK?Vkp?KE=*dFQm@%=1xhucCv0B=|*af%$n|~i;OlNGz
zA_Nl-=aA1=t3g$<?VO}(UvFyft33WTZI1t8s;$pAo4zVwdE2&a5Pvf#o2J(+mpzX@
z8wwLIvE&y;$ANE_bK%tI;p{TyP0;qBT}qepnlrgmjtOTmXdXDYrfz>PmX&DOn6+!y
zT7{Ep=P+!zIAW?>*JWKRCeCm;0)fGfx#@5?94k8d=p0c<d#Fs@V@W<tDKW$;^ce$H
zi?B=MkctL~#kWOEq2!F6bkAm(b6$0^WR|XLE2@}n*JPV?_>aH3Bv~YVR+&oOS@ITW
zu{p<POf=GqS2PALs@V;6+J%{Etysb3&Iu2nQkRbK={0<1H-_VCRFeH#R-u-upxTT)
z%u<CIrW#v9T?2YaH<4Pp8#23J5|XVpR$}3&1Z!LohE{*|!hH)ST5X@<AXAnCscxit
z<?@g>RnpV~SDEL;zP?nCTMUcIedbvVi^XDNLdNx(o*H!1)|-Y?Kq>3>7Y1z>!(#IS
z<Ft3yT|sZBc4?)(t(xHD207TNUHS4Iz&Z2_A+{pSlA(_#?XwMD8^W8ONDgduMovy4
znh~}ApzkFG=MT&6s#`Z$(&JY_jE&?Ro^S>Nfomdbl;}|fY)iKUJ_klu<kszbv~FD&
zlVriLqIetA_l2gj@LM8QaM2fwuR3n;436si=Ywou%7sGHz;KK$(*I|^Caou=C82On
zD10A^6oA<Y<`ksoLxusF24ogMmJPDzLDqf94ud5QZ@9r~fOQrW&4QwNP|OU)!cg1^
z#Uo&Q2W&AY;eZlB$jO470w~!4O1=R5cTg%0r5ix$5R}P+GC?TY0Lp#`<*ZOX3o5w4
zVL-(iP{|LKqfpfeH4Uis1Zv-eIti$^7aAJS$OesPLDMW~=7HvYAvXnYCg80Iyz7Ve
z4R}8a&ItS?A3iXH%K;zypv5I<8HH9+XuTKOoP%~gXrBOg1LzchE)Af|Ug+w9Zn@As
z7kaqBGY~up_$UbfionN<z`GYd*$bbJhtCc8!V6yp;j0w%ssp|4(0ecR*$Vw~p??GD
ze+dS7VW1fXCSXt;hPc7!0AC7*u7zO{818`)W*F&#QC(rQAN)@6Ctyqx#;%2NTVcEx
zCX9fIc9`UaKu?(56Q)dqspnw26J`vAnO>OP4(8N=xel1S7v?R3`EFQXz=C{O)Cv}R
zVMz*>#v$(#EQ`YOwXo6yt5(A5d{`5Kuk+wrJAAhiey9RJ#^5J6{Cfn1%<!Ku{NjXy
zbMWg;`0WX-cff`QuyG)4>Iz$0!PYoz_rs1^5KhAHFW`SM*trrSd9dqq*gXyQ1Yqwe
z*#8b32*W`;{NaH^R*1#m&tnje!I2moy9p=q;M6QQ<AZZfI3I_LA-FOPuI55wC0r|j
z8#cJP5pI1BxAP#`3jPYgy()0O0X%TR!!SJB3aORw_!9gbh9|S&X<vBO7oOdO=lSre
z9lZ7<6`(>OBaRAtP>~VH>__RdP{w$a8AMr8lzj@lVMW$<sHhJW8;^?TAzKucNTHmK
z$le!~&PQdAp|Vd<`5LH#4>`Q3QeRZL2C6a*RrR21c2xZ)su4vsgQ%7V)h>(bR6%tI
zqIy<TFM%4kQNsXgR0B1hg__nt&4MU5iQa03-Z_Wft%Kf+p!eO#*$VyBjy_1B4>zKg
zxu|s&)W(O}hEcn7s6!Lfu^n=MjyeTV=LG6%Mcw+M?k|wXi#|GqK5l}%RnR8|=+hwj
z+=aeqfWGvjo_5r08tNTHeezL19~xkwK~>P;0yJbR^2N~57ihQ@jWnZCUNkz0{5R29
z8ye?B<4>W92AY^efjVe%Pc$WjrWt5@S2P1Ovj&>wK(mjbxp`>54=uQa76s8_4_eX+
zE%l&0Gg@Xv%buVWUbJ#6TD=jiaigFM1;gm;S?HUA=-XE4yH@CX(Dy;~!&a2v75y|G
z{hW(_j-b#@^j{bHWhE+zp<naSZ%MSSFIs;LZCr^qzk{~cLE9W?dmM#B==XVOr-61R
zQ6z?TMbPeJXs;RV1MLf;16e5QL<dKpgHO;O^U$Ft=uim794J<R{`8~&+fm$x;y2M@
z7djk8M@FC{PtehG=s4&^6LivyPUfLgJ<;iQ=yVdDS&PozN9UWM^Ko>+i!R#HMIX8Z
zx?C1riK44<lyIYKc62R;uDj8VchHR#x*0^bdZJqibo&WPno%-^{<?(j#?ZYmx_=Wr
zc!3_-(W4Mb^+l;Ddh9}vPocl#=!t=zj6hEf^mIIW8bQyxqGuuW+<~62MK2ua#dqjM
z3cYNHUZ&8i5$JUl^m;2MLYU;oWJ+Nw6vSj?VKM@kj1;D{o|w{7n2Ojjncu-=4q!^J
zgDHa#D`2?)G5j8ka04UWz(^f1as-UB0izzkXjL%!pBTLkM&E!jW?;-V81o9oYJ#!<
z#Mmn^v<Ckjfd3`MxC1a=8;o}V<1fGjMKEC&OcVnX=fNZaCjAqW4!~pqCOd=42VjZ@
zn6eC}ii@ceVwye}T)}h!rXPYCl46Dfn6Uw70yF2p%x^Gj2FyAHvsJ+Cc`(NS%y|cM
z)xg|fo+y}S1Lhro`Ep>sF_^yq7HEM5&tTy&SojMTnSw>ZVj(OZ!s0cs_#7;L2TR1o
z5_7O*2us$$l5?<BLM-(JOP|0pNwI7nEH?qm*TD)QtndLVZox_&u<`+{@;_ERfYs_?
z^|)B057zvEwWDC2Ke5jLSa%85D}nXiVEqZ$APY9Qfeiy}I0PHs!A2Rd(Hd-=0UK|@
zCQGpC6l`_^n<vB;0k+J7twPxPPi)%<+ugwSEwDow>=*+(*1=9Gv2z3LvH-j0z-|Sw
zdkO4*2789E=M(I81bYYA`wI3sf_=wezYz9Yg8egK|2;Sm9C!l<hj8!$9Fh`;hH&Tt
z95w}qN5SE9a72J3cHqbkI4TT|PKcvl;Fu0L<_C@)fa7}L_!2nb3r@U%lUCs5H8>>(
zPI-gV0-W9fXKcY)f8y*eIOhS*J%RJq;QY8ae*!K57j(geEpSl|T-*Z}f59bFaOnqJ
z)&-Zxz!g<+Wer?a23Lo{)gN%p3tU$Q*Pp?Sd2mx0+*}7Yzrn3BaN7vnz5sVz!JU0@
zR{`9;0Qa=Py#en1f%~rD0q|f#Joo_*oxmgDkqvk>B_4}_$HU<9F?g~Do(hAfzToK(
zcs2r_JAvmn;Dru&aRpwQf>)N{)fRa5f4sH<ueZS)Ver-zyj=xvzrZ^S@NNgZ2j1_4
z52N6t4EQ(<J_+z?4tzQUpFO}A@{kYb$zz;n8E~GvIM3hXyx5QP@-)t?%Q$cH<GkIE
z^WMez5XAX76z5X_=d&N@^JkndA)K%CalQp{zIVd;kqYOh2j^EOL@GoW#psJ1FLJgc
zd65-IF?*s|(@|^}#a@p*uTh-sC~gSFYlPyLM+poixQ-HzLWz7RaTb)MI7%8q$=V^;
zhurfhc^IXbgi?+}soJ4bw^8cpC`~7n<}pfJ8%4M%Vl|2kqNt%LT@ZOg=(i9`zZqp5
zhcd55S^J`Fw~;Rz<%mW(UZY$Al<P6d1Ip_~d9R~<n^FGTsK9kp@G>fN9u+=~ifl*4
zLa6vCRN_4<8I4MYQK|K)bR$&eGAg$lmCt}G_)vw*sN!){={%}D2~`Q8s+CdI_o%ub
z)o6ukHbXT}qgta-t;eW#WmJ1Ls*@PixsU1|NA)HlzX$n)sD5r#|2Aqc2{kN@8hKHp
zFlrn`O<JL*pk^*=wi`8Xg_>VSEgGQ~-%-nG)M^rHodva?kJ|W9+n%Uh2Gs64Y9B!D
z-=hv8)G;yYI30D$f;yc>o%^CL5va>?6aaOLkGj1_J^G@aSx~RSsJEfs=TV=YsBbgW
zuQux63Js`^29835JZR8uG}wcNxM*ku8nze>pN>YvM?oJN8G%NI(I4B<pW{&UYV=nG
z`nwexRT_;BpfNr)<~|y`AB_v3@k7ys0GgN>O?;0gtwxih(Ui?7<VF81M*oJ;v}S19
zcQkzxnh`)VOQTu2(d_AHP9rpTAexsS&G(=M)6v2ZT2vS<_M#=r(b7?9Sr{#Uk5;;9
z<!iKhJzDb`t=*2+7f0*QqYWXnaX#7<M4JbqE$7kJQD|ERv~4xo;YT~YXjc}r2ejuo
z+WQ*qJB;>+(Saa3I0_vaiVhn({2CoOjl###e=hpJ89F``ohXk^?nb9Nq0{lv8AE4o
zqjTfX`Pt|~J9IHOy3`k4-i)psMprANYunNF>F7pvbmKm{8AP|LqucS(oeb!1Zgk&=
z?uXHX^XTDb^k^u0oC-Zzj-KvE&rYKk8PJQ@=v6!P+K1i@L~o0ucOLZK(1&RBF*o|;
zqEE-s=bq@xXY_Rx`ZgVX51}8I(a+`RS1Sw`i}4r}nAT%<8;dy(i?tt%-463S#^Nr<
z;?>6DpT`oc#}dxQ5*@}8U&fLQ#F7@ql5WP5ZO7d6Sn_dLiqBZ83|Q($SQ<Z;Hh@J`
z#v+GeQ9dl{F_vyI<~@$3_hY}e!ZPf}GIqi;d9ln9SeEZt*27q~<5>1E=Bth6n2qJU
zkL6yB<#n-qE|xzPR$w+(C>krg8!K`hE4Cdg;l)Y?vC{jovR<s*K&<>gtipY)k{_#F
z8LQG4tC|6;b{(s+7^_trtK-4yHN)yp#~O^o8hNnBUaX0WH64mI3t-KISc?d(Wp1qH
zbF9^OtaS)$!-xIRvH%K;Lg8gl_$Mgh9~5~QMLmO}SD=_@Q0zA-?l4Mdj}kvYNxz`v
zB$Uz_r5;CVFQD}4C}TFt?2odVqwK{fCxmhzL3tsRUxW%apu!(eQ5-66j7rX<<<Fp%
zS!mU0w7LRX(;TgxkJe2`>qBTGXwzi0={(we83p&FJ3l~oh0r|#bZ>QZ?+@s{Ai94x
zdZ0IY@E-KgW%Td^=#k#&vCim;IP_E)db%@u_677@2t9uoz3>To@dfnKar9~wdhH+d
z#s>6ebM$s^^v-znP6)lb0li-aeGo(+{((N8jy~;;J|B+0cmRF90)5jTeH(>-Sc85%
zj(&Lo{r(90V=?-(KKd(w{ucWC7xb^tf1}a=X(%)pZE23SR!7^aqwRyyj>TwK8MGU;
z=O46p4cZq(`}d=R!_mS0=<sB8q%k^L1|5q-$G4-CA#`d3I=up&`30SQ0G&ULF5ZJK
zoky2XqbqNq8`aUx&gfPgzO6mJeKfu!fHfRDfV~En+KtQB$7Q$U@?~)OA8^G0t~eT3
zx{L#Zan;tiT6bJsxJDeV`2nuA0@n%RdcySw;|8;Fqvp8DeB7)IZnhgY{{*-A0k_J+
zt%JC28Qi`C?oc0h+<-e>$DNPkF0bIO&){xhxW{DNGYa>bjr)N6*2jGh<9;9D{`>KO
zKk%S6cyI_0`3Db+!o!E-5g|MZJSq;4nUBZ(gU61?<7(q^FW?DTc;ak4@i3nB4W1l^
zr)<Deqwus4p0Nhc%)_(p!L#S%Ig9b!H}JgUc)?`6a57#Tz)R-iWfgGP8#sJ4j#!K%
zAHh+Var7fNCJ)C(;n>|cZa9wbjuYDB#M(IVJWlS9lYhY}@8Hz!I4y|N2jh$)oYfv@
zeS))tI42C}4#s&=I6s67({NF5TpYl~hw<|Ic=;E2<#@dEI9|IQug}99R^W|ccyk%N
zc?AyQA%i;qA9dji>Y|`7RZy3op{{&KT`Qojzd_xchr0Ozb?YhW_8!#j52!mIP<Owe
z?t^-;3RvRJ!2<wR03hCJkLA3O&N>et#mDK7Cs@mhd~&^yD{^#yyoDvyKR$z<T>bb=
zmU8an^H|6?J^wLU3B2U=F$3QG`0;U%oPX-$J~RBOk0%)Lw&jyQKCY;ElOJzkr}zHj
zGdSRH`uI#z-mj0(W3&JKy_(Mqe`?f3$KqhJ>SSwqn@pI{$?y?vO;dlpf2BhvVy$Jj
zX#|yB9jW$IZl8{AU_$Ba%%?Kh?)H2(*DhP{7#&`%(j#}8**0N9x45liBG!7>7I&^X
z7ulY4`k)D2@vpzpmnxg)o~o83y0pbg(^w~wi4HZ2u>@rkiq%evVMUxVje6ix&z7EY
z+H0vWIxzdUju8|5Kh}W#9l}=1Y(tp95YGMj8|DZai=l~LmIxj9Ok(t$#k(;y2&k|l
zTAZSt6gl>h<0Y10O9vM^=_F5z3|YEqCr_3f?aw9SqZJQ->z>D#M~SyR%fQdL<N1B!
zuKlVw91)ft(xivD;m3wJiuczk19`TZ?4kM`_B(>6#1<?<6Z8Lk1QH?Iqr(o-ph%1m
n+_ODIu;{%utVl=&!C{`o(_1INk$Sb6_Nz11f%_kSg5L}Sh;n$c

literal 0
HcmV?d00001

diff --git a/_static/webpack-macros.html b/_static/webpack-macros.html
new file mode 100644
index 000000000..89732b23e
--- /dev/null
+++ b/_static/webpack-macros.html
@@ -0,0 +1,24 @@
+<!--
+  AUTO-GENERATED from webpack.config.js, do **NOT** edit by hand.
+  These are re-used in layout.html
+-->
+
+{% macro head_pre_assets() %}
+  <!-- Loaded before other Sphinx assets -->
+  <link href="{{ pathto('_static/styles/theme.css', 1) }}?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="{{ pathto('_static/styles/pydata-sphinx-theme.css', 1) }}?digest=8878045cc6db502f8baf" rel="stylesheet" />
+{% endmacro %}
+
+{% macro head_js_preload() %}
+  <!-- So that users can add custom icons -->
+  <script src="{{ pathto('_static/scripts/fontawesome.js', 1) }}?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="{{ pathto('_static/scripts/bootstrap.js', 1) }}?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="{{ pathto('_static/scripts/pydata-sphinx-theme.js', 1) }}?digest=8878045cc6db502f8baf" />
+{% endmacro %}
+
+{% macro body_post() %}
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="{{ pathto('_static/scripts/bootstrap.js', 1) }}?digest=8878045cc6db502f8baf"></script>
+<script defer src="{{ pathto('_static/scripts/pydata-sphinx-theme.js', 1) }}?digest=8878045cc6db502f8baf"></script>
+{% endmacro %}
\ No newline at end of file
diff --git a/about/authors.html b/about/authors.html
new file mode 100644
index 000000000..a46e06bae
--- /dev/null
+++ b/about/authors.html
@@ -0,0 +1,484 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>Contributors &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'about/authors';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="prev" title="Changelog" href="changelog.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="changelog.html">Changelog</a></li>
+</ul>
+<ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active"><a class="current reference internal" href="#">Contributors</a></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="index.html" class="nav-link">About</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Contributors</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="contributors">
+<h1>Contributors<a class="headerlink" href="#contributors" title="Link to this heading">#</a></h1>
+<p>Shizhe Diao, Rui Pan, Hanze Dong, Ka Shun Shum, Jipeng Zhang, Wei Xiong, Tong Zhang</p>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="changelog.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">Changelog</p>
+      </div>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/about/authors.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/about/changelog.html b/about/changelog.html
new file mode 100644
index 000000000..eeaac7866
--- /dev/null
+++ b/about/changelog.html
@@ -0,0 +1,513 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>Changelog &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'about/changelog';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Contributors" href="authors.html" />
+    <link rel="prev" title="About" href="index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active"><a class="current reference internal" href="#">Changelog</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="authors.html">Contributors</a></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="index.html" class="nav-link">About</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Changelog</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="changelog">
+<h1>Changelog<a class="headerlink" href="#changelog" title="Link to this heading">#</a></h1>
+<section id="version-0-0-1-mar-28-2023">
+<h2>Version 0.0.1 (Mar 28, 2023)<a class="headerlink" href="#version-0-0-1-mar-28-2023" title="Link to this heading">#</a></h2>
+<p>The first public version.</p>
+<p>Task tuning, instruction tuning, on user defined datasets.</p>
+<p>A simple and extensible API for developers.</p>
+<p>Efficient finetuning with LoRA.</p>
+<p>Simplified model inference framework.</p>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">About</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="authors.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">Contributors</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#version-0-0-1-mar-28-2023">Version 0.0.1 (Mar 28, 2023)</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/about/changelog.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/about/index.html b/about/index.html
new file mode 100644
index 000000000..3206208a8
--- /dev/null
+++ b/about/index.html
@@ -0,0 +1,503 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>About &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'about/index';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Changelog" href="changelog.html" />
+    <link rel="prev" title="lmflow.version" href="../autoapi/lmflow/version/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="#">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="#">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="changelog.html">Changelog</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="authors.html">Contributors</a></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">About</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="about">
+<h1>About<a class="headerlink" href="#about" title="Link to this heading">#</a></h1>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="changelog.html">Changelog</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="changelog.html#version-0-0-1-mar-28-2023">Version 0.0.1 (Mar 28, 2023)</a></li>
+</ul>
+</li>
+</ul>
+</div>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="authors.html">Contributors</a></li>
+</ul>
+</div>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../autoapi/lmflow/version/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.version</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="changelog.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">Changelog</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/about/index.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/index.html b/autoapi/index.html
new file mode 100644
index 000000000..be2ed3ab5
--- /dev/null
+++ b/autoapi/index.html
@@ -0,0 +1,747 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>API Reference &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/index';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="lmflow" href="lmflow/index.html" />
+    <link rel="prev" title="LMFlow Benchmark Guide" href="../examples/TASK_GUIDE.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="#">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="#">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="nav bd-sidenav">
+<li class="toctree-l1 has-children"><a class="reference internal" href="lmflow/index.html">lmflow</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l2"><a class="reference internal" href="lmflow/args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="lmflow/datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="lmflow/models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="lmflow/models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="lmflow/models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="lmflow/optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="lmflow/pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="lmflow/pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="lmflow/tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="lmflow/utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="lmflow/utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="lmflow/utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="lmflow/utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="lmflow/version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">API Reference</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="api-reference">
+<h1>API Reference<a class="headerlink" href="#api-reference" title="Link to this heading">#</a></h1>
+<p>This page contains auto-generated API reference documentation <a class="footnote-reference brackets" href="#f1" id="id1" role="doc-noteref"><span class="fn-bracket">[</span>1<span class="fn-bracket">]</span></a>.</p>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="lmflow/index.html">lmflow</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="lmflow/args/index.html">lmflow.args</a></li>
+<li class="toctree-l2"><a class="reference internal" href="lmflow/datasets/index.html">lmflow.datasets</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="lmflow/models/index.html">lmflow.models</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/interfaces/index.html">lmflow.models.interfaces</a><ul>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/models/vision_encoder/index.html">lmflow.models.vision_encoder</a><ul>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="lmflow/optim/index.html">lmflow.optim</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="lmflow/pipeline/index.html">lmflow.pipeline</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/utils/index.html">lmflow.pipeline.utils</a><ul>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="lmflow/tokenization/index.html">lmflow.tokenization</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="lmflow/utils/index.html">lmflow.utils</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/conversation_template/index.html">lmflow.utils.conversation_template</a><ul>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/flash_attention/index.html">lmflow.utils.flash_attention</a><ul>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><ul>
+<li class="toctree-l4"><a class="reference internal" href="lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</li>
+<li class="toctree-l3"><a class="reference internal" href="lmflow/utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="lmflow/version/index.html">lmflow.version</a></li>
+</ul>
+</li>
+</ul>
+</div>
+<aside class="footnote-list brackets">
+<aside class="footnote brackets" id="f1" role="doc-footnote">
+<span class="label"><span class="fn-bracket">[</span><a role="doc-backlink" href="#id1">1</a><span class="fn-bracket">]</span></span>
+<p>Created with <a class="reference external" href="https://github.com/readthedocs/sphinx-autoapi">sphinx-autoapi</a></p>
+</aside>
+</aside>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../examples/TASK_GUIDE.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">LMFlow Benchmark Guide</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="lmflow/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/autoapi/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/args/index.html b/autoapi/lmflow/args/index.html
new file mode 100644
index 000000000..f2b9920af
--- /dev/null
+++ b/autoapi/lmflow/args/index.html
@@ -0,0 +1,2447 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.args &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/args/index';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+    <link rel="next" title="lmflow.datasets" href="../datasets/index.html" />
+    <link rel="prev" title="lmflow" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2 current active"><a class="current reference internal" href="#">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.args</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.args">
+<span id="lmflow-args"></span><h1>lmflow.args<a class="headerlink" href="#module-lmflow.args" title="Link to this heading">#</a></h1>
+<p>This script defines dataclasses: ModelArguments and DatasetArguments,
+that contain the arguments for the model and dataset used in training.</p>
+<p>It imports several modules, including dataclasses, field from typing, Optional from typing,
+require_version from transformers.utils.versions, MODEL_FOR_CAUSAL_LM_MAPPING,
+and TrainingArguments from transformers.</p>
+<p>MODEL_CONFIG_CLASSES is assigned a list of the model config classes from
+MODEL_FOR_CAUSAL_LM_MAPPING. MODEL_TYPES is assigned a tuple of the model types
+extracted from the MODEL_CONFIG_CLASSES.</p>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.args.MODEL_CONFIG_CLASSES" title="lmflow.args.MODEL_CONFIG_CLASSES"><code class="xref py py-obj docutils literal notranslate"><span class="pre">MODEL_CONFIG_CLASSES</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.args.MODEL_TYPES" title="lmflow.args.MODEL_TYPES"><code class="xref py py-obj docutils literal notranslate"><span class="pre">MODEL_TYPES</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.args.logger" title="lmflow.args.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.args.PIPELINE_ARGUMENT_MAPPING" title="lmflow.args.PIPELINE_ARGUMENT_MAPPING"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PIPELINE_ARGUMENT_MAPPING</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.args.OptimizerNames" title="lmflow.args.OptimizerNames"><code class="xref py py-obj docutils literal notranslate"><span class="pre">OptimizerNames</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ModelArguments</span></code></a></p></td>
+<td><p>Define a class ModelArguments using the dataclass decorator.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.args.VisModelArguments" title="lmflow.args.VisModelArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">VisModelArguments</span></code></a></p></td>
+<td><p>Define a class ModelArguments using the dataclass decorator.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DatasetArguments</span></code></a></p></td>
+<td><p>Define a class DatasetArguments using the dataclass decorator.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.args.MultiModalDatasetArguments" title="lmflow.args.MultiModalDatasetArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">MultiModalDatasetArguments</span></code></a></p></td>
+<td><p>Define a class DatasetArguments using the dataclass decorator.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.args.FinetunerArguments" title="lmflow.args.FinetunerArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FinetunerArguments</span></code></a></p></td>
+<td><p>Adapt transformers.TrainingArguments</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.args.RewardModelTunerArguments" title="lmflow.args.RewardModelTunerArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RewardModelTunerArguments</span></code></a></p></td>
+<td><p>Arguments for reward modeling.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.args.EvaluatorArguments" title="lmflow.args.EvaluatorArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">EvaluatorArguments</span></code></a></p></td>
+<td><p>Define a class EvaluatorArguments using the dataclass decorator. The class contains several optional</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.args.InferencerArguments" title="lmflow.args.InferencerArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">InferencerArguments</span></code></a></p></td>
+<td><p>Define a class InferencerArguments using the dataclass decorator. The class contains several optional</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.args.RaftAlignerArguments" title="lmflow.args.RaftAlignerArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RaftAlignerArguments</span></code></a></p></td>
+<td><p>Define a class RaftAlignerArguments to configure raft aligner.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.args.BenchmarkingArguments" title="lmflow.args.BenchmarkingArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">BenchmarkingArguments</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.args.DPOAlignerArguments" title="lmflow.args.DPOAlignerArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DPOAlignerArguments</span></code></a></p></td>
+<td><p>The arguments for the DPO training script.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.args.DPOv2AlignerArguments" title="lmflow.args.DPOv2AlignerArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DPOv2AlignerArguments</span></code></a></p></td>
+<td><p>The arguments for the DPOv2 training script.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.args.IterativeAlignerArguments" title="lmflow.args.IterativeAlignerArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">IterativeAlignerArguments</span></code></a></p></td>
+<td><p>Arguments for iterative aligners.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.args.IterativeDPOAlignerArguments" title="lmflow.args.IterativeDPOAlignerArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">IterativeDPOAlignerArguments</span></code></a></p></td>
+<td><p>Arguments for iterative DPO aligners.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.args.AutoArguments" title="lmflow.args.AutoArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">AutoArguments</span></code></a></p></td>
+<td><p>Automatically choose arguments from FinetunerArguments or EvaluatorArguments.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.args.split_args" title="lmflow.args.split_args"><code class="xref py py-obj docutils literal notranslate"><span class="pre">split_args</span></code></a>(args)</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.args.MODEL_CONFIG_CLASSES">
+<span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">MODEL_CONFIG_CLASSES</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#MODEL_CONFIG_CLASSES"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.MODEL_CONFIG_CLASSES" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.args.MODEL_TYPES">
+<span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">MODEL_TYPES</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#MODEL_TYPES"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.MODEL_TYPES" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.args.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">OptimizerNames</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames" title="Link to this definition">#</a></dt>
+<dd><dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.DUMMY">
+<span class="sig-name descname"><span class="pre">DUMMY</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'dummy'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.DUMMY"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.DUMMY" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.ADABELIEF">
+<span class="sig-name descname"><span class="pre">ADABELIEF</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'adabelief'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.ADABELIEF"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.ADABELIEF" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.ADABOUND">
+<span class="sig-name descname"><span class="pre">ADABOUND</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'adabound'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.ADABOUND"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.ADABOUND" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.LARS">
+<span class="sig-name descname"><span class="pre">LARS</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'lars'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.LARS"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.LARS" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.LAMB">
+<span class="sig-name descname"><span class="pre">LAMB</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'lamb'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.LAMB"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.LAMB" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.ADAMAX">
+<span class="sig-name descname"><span class="pre">ADAMAX</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'adamax'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.ADAMAX"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.ADAMAX" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.NADAM">
+<span class="sig-name descname"><span class="pre">NADAM</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'nadam'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.NADAM"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.NADAM" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.RADAM">
+<span class="sig-name descname"><span class="pre">RADAM</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'radam'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.RADAM"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.RADAM" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.ADAMP">
+<span class="sig-name descname"><span class="pre">ADAMP</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'adamp'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.ADAMP"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.ADAMP" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.SGDP">
+<span class="sig-name descname"><span class="pre">SGDP</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'sgdp'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.SGDP"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.SGDP" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.YOGI">
+<span class="sig-name descname"><span class="pre">YOGI</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'yogi'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.YOGI"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.YOGI" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.SOPHIA">
+<span class="sig-name descname"><span class="pre">SOPHIA</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'sophia'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.SOPHIA"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.SOPHIA" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.ADAN">
+<span class="sig-name descname"><span class="pre">ADAN</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'adan'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.ADAN"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.ADAN" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.ADAM">
+<span class="sig-name descname"><span class="pre">ADAM</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'adam'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.ADAM"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.ADAM" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.NOVOGRAD">
+<span class="sig-name descname"><span class="pre">NOVOGRAD</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'novograd'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.NOVOGRAD"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.NOVOGRAD" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.ADADELTA">
+<span class="sig-name descname"><span class="pre">ADADELTA</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'adadelta'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.ADADELTA"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.ADADELTA" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.ADAGRAD">
+<span class="sig-name descname"><span class="pre">ADAGRAD</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'adagrad'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.ADAGRAD"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.ADAGRAD" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.ADAMW_SCHEDULE_FREE">
+<span class="sig-name descname"><span class="pre">ADAMW_SCHEDULE_FREE</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'adamw_schedule_free'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.ADAMW_SCHEDULE_FREE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.ADAMW_SCHEDULE_FREE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.OptimizerNames.SGD_SCHEDULE_FREE">
+<span class="sig-name descname"><span class="pre">SGD_SCHEDULE_FREE</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'sgd_schedule_free'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#OptimizerNames.SGD_SCHEDULE_FREE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.OptimizerNames.SGD_SCHEDULE_FREE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">ModelArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments" title="Link to this definition">#</a></dt>
+<dd><p>Define a class ModelArguments using the dataclass decorator.
+The class contains several optional parameters that can be used to configure a model.</p>
+<dl class="simple">
+<dt>model_name_or_path<span class="classifier">str</span></dt><dd><p>a string representing the path or name of a pretrained
+model checkpoint for weights initialization. If None, a model will be trained from scratch.</p>
+</dd>
+<dt>model_type<span class="classifier">str</span></dt><dd><p>a string representing the type of model to use if training from
+scratch. If not provided, a pretrained model will be used.</p>
+</dd>
+<dt>config_overrides<span class="classifier">str</span></dt><dd><p>a string representing the default config settings to override
+when training a model from scratch.</p>
+</dd>
+<dt>config_name<span class="classifier">str</span></dt><dd><p>a string representing the name or path of the pretrained config to
+use, if different from the model_name_or_path.</p>
+</dd>
+<dt>tokenizer_name<span class="classifier">str</span></dt><dd><p>a string representing the name or path of the pretrained tokenizer
+to use, if different from the model_name_or_path.</p>
+</dd>
+<dt>cache_dir<span class="classifier">str</span></dt><dd><p>a string representing the path to the directory where pretrained models
+downloaded from huggingface.co will be stored.</p>
+</dd>
+<dt>use_fast_tokenizer<span class="classifier">bool</span></dt><dd><p>a boolean indicating whether to use a fast tokenizer (backed by the
+tokenizers library) or not.</p>
+</dd>
+<dt>model_revision<span class="classifier">str</span></dt><dd><p>a string representing the specific model version to use (can be a
+branch name, tag name, or commit id).</p>
+</dd>
+<dt>token<span class="classifier">Optional[str]</span></dt><dd><p>Necessary when accessing a private model/dataset.</p>
+</dd>
+<dt>torch_dtype<span class="classifier">str</span></dt><dd><p>a string representing the dtype to load the model under. If auto is
+passed, the dtype will be automatically derived from the model’s weights.</p>
+</dd>
+<dt>use_ram_optimized_load<span class="classifier">bool</span></dt><dd><p>a boolean indicating whether to use disk mapping when memory is not
+enough.</p>
+</dd>
+<dt>use_int8<span class="classifier">bool</span></dt><dd><p>a boolean indicating whether to load int8 quantization for inference.</p>
+</dd>
+<dt>load_in_4bit<span class="classifier">bool</span></dt><dd><p>whether to load the model in 4bit</p>
+</dd>
+<dt>model_max_length<span class="classifier">int</span></dt><dd><p>The maximum length of the model.</p>
+</dd>
+<dt>truncation_side<span class="classifier">str</span></dt><dd><p>The side on which the model should have truncation applied.</p>
+</dd>
+<dt>arch_type<span class="classifier">str</span></dt><dd><p>Model architecture type.</p>
+</dd>
+<dt>padding_side<span class="classifier">str</span></dt><dd><p>The side on which the tokenizer should have padding applied.</p>
+</dd>
+<dt>eos_padding<span class="classifier">bool</span></dt><dd><p>whether to pad with eos token instead of pad token.</p>
+</dd>
+<dt>ignore_bias_buffers<span class="classifier">bool</span></dt><dd><p>fix for DDP issues with LM bias/mask buffers - invalid scalar type,`inplace operation.</p>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.model_name_or_path">
+<span class="sig-name descname"><span class="pre">model_name_or_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.model_name_or_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.model_name_or_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.lora_model_path">
+<span class="sig-name descname"><span class="pre">lora_model_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.lora_model_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.lora_model_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.model_type">
+<span class="sig-name descname"><span class="pre">model_type</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.model_type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.model_type" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.config_overrides">
+<span class="sig-name descname"><span class="pre">config_overrides</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.config_overrides"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.config_overrides" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.arch_type">
+<span class="sig-name descname"><span class="pre">arch_type</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'decoder_only'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.arch_type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.arch_type" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.config_name">
+<span class="sig-name descname"><span class="pre">config_name</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.config_name"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.config_name" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.tokenizer_name">
+<span class="sig-name descname"><span class="pre">tokenizer_name</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.tokenizer_name"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.tokenizer_name" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.cache_dir">
+<span class="sig-name descname"><span class="pre">cache_dir</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.cache_dir"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.cache_dir" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.use_fast_tokenizer">
+<span class="sig-name descname"><span class="pre">use_fast_tokenizer</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.use_fast_tokenizer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.use_fast_tokenizer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.model_revision">
+<span class="sig-name descname"><span class="pre">model_revision</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'main'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.model_revision"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.model_revision" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.token">
+<span class="sig-name descname"><span class="pre">token</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.token"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.token" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.trust_remote_code">
+<span class="sig-name descname"><span class="pre">trust_remote_code</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.trust_remote_code"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.trust_remote_code" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.torch_dtype">
+<span class="sig-name descname"><span class="pre">torch_dtype</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.torch_dtype"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.torch_dtype" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.use_dora">
+<span class="sig-name descname"><span class="pre">use_dora</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.use_dora"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.use_dora" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.use_lora">
+<span class="sig-name descname"><span class="pre">use_lora</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.use_lora"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.use_lora" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.use_qlora">
+<span class="sig-name descname"><span class="pre">use_qlora</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.use_qlora"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.use_qlora" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.bits">
+<span class="sig-name descname"><span class="pre">bits</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">4</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.bits"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.bits" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.quant_type">
+<span class="sig-name descname"><span class="pre">quant_type</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'nf4'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.quant_type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.quant_type" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.double_quant">
+<span class="sig-name descname"><span class="pre">double_quant</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.double_quant"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.double_quant" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.lora_r">
+<span class="sig-name descname"><span class="pre">lora_r</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">8</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.lora_r"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.lora_r" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.lora_alpha">
+<span class="sig-name descname"><span class="pre">lora_alpha</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">32</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.lora_alpha"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.lora_alpha" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.lora_target_modules">
+<span class="sig-name descname"><span class="pre">lora_target_modules</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.lora_target_modules"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.lora_target_modules" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.lora_dropout">
+<span class="sig-name descname"><span class="pre">lora_dropout</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.lora_dropout"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.lora_dropout" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.save_aggregated_lora">
+<span class="sig-name descname"><span class="pre">save_aggregated_lora</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.save_aggregated_lora"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.save_aggregated_lora" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.use_ram_optimized_load">
+<span class="sig-name descname"><span class="pre">use_ram_optimized_load</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.use_ram_optimized_load"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.use_ram_optimized_load" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.use_flash_attention">
+<span class="sig-name descname"><span class="pre">use_flash_attention</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.use_flash_attention"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.use_flash_attention" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.truncate_to_model_max_length">
+<span class="sig-name descname"><span class="pre">truncate_to_model_max_length</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.truncate_to_model_max_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.truncate_to_model_max_length" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.do_rope_scaling">
+<span class="sig-name descname"><span class="pre">do_rope_scaling</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.do_rope_scaling"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.do_rope_scaling" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.rope_pi_ratio">
+<span class="sig-name descname"><span class="pre">rope_pi_ratio</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.rope_pi_ratio"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.rope_pi_ratio" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.rope_ntk_ratio">
+<span class="sig-name descname"><span class="pre">rope_ntk_ratio</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.rope_ntk_ratio"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.rope_ntk_ratio" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.use_int8">
+<span class="sig-name descname"><span class="pre">use_int8</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.use_int8"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.use_int8" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.load_in_4bit">
+<span class="sig-name descname"><span class="pre">load_in_4bit</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.load_in_4bit"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.load_in_4bit" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.model_max_length">
+<span class="sig-name descname"><span class="pre">model_max_length</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.model_max_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.model_max_length" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.truncation_side">
+<span class="sig-name descname"><span class="pre">truncation_side</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.truncation_side"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.truncation_side" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.padding_side">
+<span class="sig-name descname"><span class="pre">padding_side</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'right'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.padding_side"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.padding_side" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.eos_padding">
+<span class="sig-name descname"><span class="pre">eos_padding</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.eos_padding"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.eos_padding" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.ignore_bias_buffers">
+<span class="sig-name descname"><span class="pre">ignore_bias_buffers</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.ignore_bias_buffers"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.ignore_bias_buffers" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.args.ModelArguments.__post_init__">
+<span class="sig-name descname"><span class="pre">__post_init__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/args.html#ModelArguments.__post_init__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.ModelArguments.__post_init__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">VisModelArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ModelArguments</span></code></a></p>
+<p>Define a class ModelArguments using the dataclass decorator.
+The class contains several optional parameters that can be used to configure a model.</p>
+<dl class="simple">
+<dt>model_name_or_path<span class="classifier">str</span></dt><dd><p>a string representing the path or name of a pretrained
+model checkpoint for weights initialization. If None, a model will be trained from scratch.</p>
+</dd>
+<dt>model_type<span class="classifier">str</span></dt><dd><p>a string representing the type of model to use if training from
+scratch. If not provided, a pretrained model will be used.</p>
+</dd>
+<dt>config_overrides<span class="classifier">str</span></dt><dd><p>a string representing the default config settings to override
+when training a model from scratch.</p>
+</dd>
+<dt>config_name<span class="classifier">str</span></dt><dd><p>a string representing the name or path of the pretrained config to
+use, if different from the model_name_or_path.</p>
+</dd>
+<dt>tokenizer_name<span class="classifier">str</span></dt><dd><p>a string representing the name or path of the pretrained tokenizer
+to use, if different from the model_name_or_path.</p>
+</dd>
+<dt>cache_dir<span class="classifier">str</span></dt><dd><p>a string representing the path to the directory where pretrained models
+downloaded from huggingface.co will be stored.</p>
+</dd>
+<dt>use_fast_tokenizer<span class="classifier">bool</span></dt><dd><p>a boolean indicating whether to use a fast tokenizer (backed by the
+tokenizers library) or not.</p>
+</dd>
+<dt>model_revision<span class="classifier">str</span></dt><dd><p>a string representing the specific model version to use (can be a
+branch name, tag name, or commit id).</p>
+</dd>
+<dt>token<span class="classifier">Optional[str]</span></dt><dd><p>Necessary when accessing a private model/dataset.</p>
+</dd>
+<dt>torch_dtype<span class="classifier">str</span></dt><dd><p>a string representing the dtype to load the model under. If auto is
+passed, the dtype will be automatically derived from the model’s weights.</p>
+</dd>
+<dt>use_ram_optimized_load<span class="classifier">bool</span></dt><dd><p>a boolean indicating whether to use disk mapping when memory is not
+enough.</p>
+</dd>
+<dt>use_int8<span class="classifier">bool</span></dt><dd><p>a boolean indicating whether to load int8 quantization for inference.</p>
+</dd>
+<dt>load_in_4bit<span class="classifier">bool</span></dt><dd><p>whether to load the model in 4bit</p>
+</dd>
+<dt>model_max_length<span class="classifier">int</span></dt><dd><p>The maximum length of the model.</p>
+</dd>
+<dt>truncation_side<span class="classifier">str</span></dt><dd><p>The side on which the model should have truncation applied.</p>
+</dd>
+<dt>arch_type<span class="classifier">str</span></dt><dd><p>Model architecture type.</p>
+</dd>
+<dt>padding_side<span class="classifier">str</span></dt><dd><p>The side on which the tokenizer should have padding applied.</p>
+</dd>
+<dt>eos_padding<span class="classifier">bool</span></dt><dd><p>whether to pad with eos token instead of pad token.</p>
+</dd>
+<dt>ignore_bias_buffers<span class="classifier">bool</span></dt><dd><p>fix for DDP issues with LM bias/mask buffers - invalid scalar type,`inplace operation.</p>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.low_resource">
+<span class="sig-name descname"><span class="pre">low_resource</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.low_resource"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.low_resource" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.custom_model">
+<span class="sig-name descname"><span class="pre">custom_model</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.custom_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.custom_model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.pretrained_language_projection_path">
+<span class="sig-name descname"><span class="pre">pretrained_language_projection_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.pretrained_language_projection_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.pretrained_language_projection_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.custom_vision_model">
+<span class="sig-name descname"><span class="pre">custom_vision_model</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.custom_vision_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.custom_vision_model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.image_encoder_name_or_path">
+<span class="sig-name descname"><span class="pre">image_encoder_name_or_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.image_encoder_name_or_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.image_encoder_name_or_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.qformer_name_or_path">
+<span class="sig-name descname"><span class="pre">qformer_name_or_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.qformer_name_or_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.qformer_name_or_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.llm_model_name_or_path">
+<span class="sig-name descname"><span class="pre">llm_model_name_or_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.llm_model_name_or_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.llm_model_name_or_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.use_prompt_cache">
+<span class="sig-name descname"><span class="pre">use_prompt_cache</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.use_prompt_cache"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.use_prompt_cache" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.prompt_cache_path">
+<span class="sig-name descname"><span class="pre">prompt_cache_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.prompt_cache_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.prompt_cache_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.llava_loading">
+<span class="sig-name descname"><span class="pre">llava_loading</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.llava_loading"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.llava_loading" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.with_qformer">
+<span class="sig-name descname"><span class="pre">with_qformer</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.with_qformer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.with_qformer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.vision_select_layer">
+<span class="sig-name descname"><span class="pre">vision_select_layer</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">-2</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.vision_select_layer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.vision_select_layer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.llava_pretrain_model_path">
+<span class="sig-name descname"><span class="pre">llava_pretrain_model_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.llava_pretrain_model_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.llava_pretrain_model_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.VisModelArguments.save_pretrain_model_path">
+<span class="sig-name descname"><span class="pre">save_pretrain_model_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#VisModelArguments.save_pretrain_model_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.VisModelArguments.save_pretrain_model_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">DatasetArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments" title="Link to this definition">#</a></dt>
+<dd><p>Define a class DatasetArguments using the dataclass decorator.
+The class contains several optional parameters that can be used to configure a dataset for a language model.</p>
+<dl class="simple">
+<dt>dataset_path<span class="classifier">str</span></dt><dd><p>a string representing the path of the dataset to use.</p>
+</dd>
+<dt>dataset_name<span class="classifier">str</span></dt><dd><p>a string representing the name of the dataset to use. The default value is “customized”.</p>
+</dd>
+<dt>is_custom_dataset<span class="classifier">bool</span></dt><dd><p>a boolean indicating whether to use custom data. The default value is False.</p>
+</dd>
+<dt>customized_cache_dir<span class="classifier">str</span></dt><dd><p>a string representing the path to the directory where customized dataset caches will be stored.</p>
+</dd>
+<dt>dataset_config_name<span class="classifier">str</span></dt><dd><p>a string representing the configuration name of the dataset to use (via the datasets library).</p>
+</dd>
+<dt>train_file<span class="classifier">str</span></dt><dd><p>a string representing the path to the input training data file (a text file).</p>
+</dd>
+<dt>validation_file<span class="classifier">str</span></dt><dd><p>a string representing the path to the input evaluation data file to evaluate the perplexity on (a text file).</p>
+</dd>
+<dt>max_train_samples<span class="classifier">int</span></dt><dd><p>an integer indicating the maximum number of training examples to use for debugging or quicker training.
+If set, the training dataset will be truncated to this number.</p>
+</dd>
+<dt>max_eval_samples: int</dt><dd><p>an integer indicating the maximum number of evaluation examples to use for debugging or quicker training.
+If set, the evaluation dataset will be truncated to this number.</p>
+</dd>
+<dt>streaming<span class="classifier">bool</span></dt><dd><p>a boolean indicating whether to enable streaming mode.</p>
+</dd>
+<dt>block_size: int</dt><dd><p>an integer indicating the optional input sequence length after tokenization. The training dataset will be
+truncated in blocks of this size for training.</p>
+</dd>
+<dt>train_on_prompt: bool</dt><dd><p>a boolean indicating whether to train on prompt for conversation datasets such as ShareGPT.</p>
+</dd>
+<dt>conversation_template: str</dt><dd><p>a string representing the template for conversation datasets.</p>
+</dd>
+</dl>
+<p>The class also includes some additional parameters that can be used to configure the dataset further, such as <cite>overwrite_cache</cite>,
+<cite>validation_split_percentage</cite>, <cite>preprocessing_num_workers</cite>, <cite>disable_group_texts</cite>, <cite>demo_example_in_prompt</cite>, <cite>explanation_in_prompt</cite>,
+<cite>keep_linebreaks</cite>, and <cite>prompt_structure</cite>.</p>
+<p>The field function is used to set default values and provide help messages for each parameter. The Optional type hint is
+used to indicate that a parameter is optional. The metadata argument is used to provide additional information about
+each parameter, such as a help message.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.dataset_path">
+<span class="sig-name descname"><span class="pre">dataset_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.dataset_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.dataset_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.dataset_name">
+<span class="sig-name descname"><span class="pre">dataset_name</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'customized'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.dataset_name"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.dataset_name" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.is_custom_dataset">
+<span class="sig-name descname"><span class="pre">is_custom_dataset</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.is_custom_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.is_custom_dataset" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.customized_cache_dir">
+<span class="sig-name descname"><span class="pre">customized_cache_dir</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'.cache/llm-ft/datasets'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.customized_cache_dir"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.customized_cache_dir" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.dataset_config_name">
+<span class="sig-name descname"><span class="pre">dataset_config_name</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.dataset_config_name"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.dataset_config_name" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.train_file">
+<span class="sig-name descname"><span class="pre">train_file</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.train_file"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.train_file" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.validation_file">
+<span class="sig-name descname"><span class="pre">validation_file</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.validation_file"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.validation_file" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.max_train_samples">
+<span class="sig-name descname"><span class="pre">max_train_samples</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.max_train_samples"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.max_train_samples" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.max_eval_samples">
+<span class="sig-name descname"><span class="pre">max_eval_samples</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">10000000000.0</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.max_eval_samples"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.max_eval_samples" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.streaming">
+<span class="sig-name descname"><span class="pre">streaming</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.streaming"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.streaming" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.block_size">
+<span class="sig-name descname"><span class="pre">block_size</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.block_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.block_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.overwrite_cache">
+<span class="sig-name descname"><span class="pre">overwrite_cache</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.overwrite_cache"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.overwrite_cache" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.validation_split_percentage">
+<span class="sig-name descname"><span class="pre">validation_split_percentage</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">5</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.validation_split_percentage"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.validation_split_percentage" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.preprocessing_num_workers">
+<span class="sig-name descname"><span class="pre">preprocessing_num_workers</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.preprocessing_num_workers"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.preprocessing_num_workers" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.group_texts_batch_size">
+<span class="sig-name descname"><span class="pre">group_texts_batch_size</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1000</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.group_texts_batch_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.group_texts_batch_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.disable_group_texts">
+<span class="sig-name descname"><span class="pre">disable_group_texts</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.disable_group_texts"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.disable_group_texts" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.keep_linebreaks">
+<span class="sig-name descname"><span class="pre">keep_linebreaks</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.keep_linebreaks"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.keep_linebreaks" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.test_file">
+<span class="sig-name descname"><span class="pre">test_file</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.test_file"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.test_file" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.train_on_prompt">
+<span class="sig-name descname"><span class="pre">train_on_prompt</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.train_on_prompt"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.train_on_prompt" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.conversation_template">
+<span class="sig-name descname"><span class="pre">conversation_template</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.conversation_template"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.conversation_template" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.args.DatasetArguments.__post_init__">
+<span class="sig-name descname"><span class="pre">__post_init__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/args.html#DatasetArguments.__post_init__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DatasetArguments.__post_init__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.MultiModalDatasetArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">MultiModalDatasetArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#MultiModalDatasetArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.MultiModalDatasetArguments" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DatasetArguments</span></code></a></p>
+<p>Define a class DatasetArguments using the dataclass decorator.
+The class contains several optional parameters that can be used to configure a dataset for a language model.</p>
+<dl class="simple">
+<dt>dataset_path<span class="classifier">str</span></dt><dd><p>a string representing the path of the dataset to use.</p>
+</dd>
+<dt>dataset_name<span class="classifier">str</span></dt><dd><p>a string representing the name of the dataset to use. The default value is “customized”.</p>
+</dd>
+<dt>is_custom_dataset<span class="classifier">bool</span></dt><dd><p>a boolean indicating whether to use custom data. The default value is False.</p>
+</dd>
+<dt>customized_cache_dir<span class="classifier">str</span></dt><dd><p>a string representing the path to the directory where customized dataset caches will be stored.</p>
+</dd>
+<dt>dataset_config_name<span class="classifier">str</span></dt><dd><p>a string representing the configuration name of the dataset to use (via the datasets library).</p>
+</dd>
+<dt>train_file<span class="classifier">str</span></dt><dd><p>a string representing the path to the input training data file (a text file).</p>
+</dd>
+<dt>validation_file<span class="classifier">str</span></dt><dd><p>a string representing the path to the input evaluation data file to evaluate the perplexity on (a text file).</p>
+</dd>
+<dt>max_train_samples<span class="classifier">int</span></dt><dd><p>an integer indicating the maximum number of training examples to use for debugging or quicker training.
+If set, the training dataset will be truncated to this number.</p>
+</dd>
+<dt>max_eval_samples: int</dt><dd><p>an integer indicating the maximum number of evaluation examples to use for debugging or quicker training.
+If set, the evaluation dataset will be truncated to this number.</p>
+</dd>
+<dt>streaming<span class="classifier">bool</span></dt><dd><p>a boolean indicating whether to enable streaming mode.</p>
+</dd>
+<dt>block_size: int</dt><dd><p>an integer indicating the optional input sequence length after tokenization. The training dataset will be
+truncated in blocks of this size for training.</p>
+</dd>
+<dt>train_on_prompt: bool</dt><dd><p>a boolean indicating whether to train on prompt for conversation datasets such as ShareGPT.</p>
+</dd>
+<dt>conversation_template: str</dt><dd><p>a string representing the template for conversation datasets.</p>
+</dd>
+</dl>
+<p>The class also includes some additional parameters that can be used to configure the dataset further, such as <cite>overwrite_cache</cite>,
+<cite>validation_split_percentage</cite>, <cite>preprocessing_num_workers</cite>, <cite>disable_group_texts</cite>, <cite>demo_example_in_prompt</cite>, <cite>explanation_in_prompt</cite>,
+<cite>keep_linebreaks</cite>, and <cite>prompt_structure</cite>.</p>
+<p>The field function is used to set default values and provide help messages for each parameter. The Optional type hint is
+used to indicate that a parameter is optional. The metadata argument is used to provide additional information about
+each parameter, such as a help message.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.MultiModalDatasetArguments.image_folder">
+<span class="sig-name descname"><span class="pre">image_folder</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#MultiModalDatasetArguments.image_folder"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.MultiModalDatasetArguments.image_folder" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.MultiModalDatasetArguments.image_aspect_ratio">
+<span class="sig-name descname"><span class="pre">image_aspect_ratio</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'pad'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#MultiModalDatasetArguments.image_aspect_ratio"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.MultiModalDatasetArguments.image_aspect_ratio" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.MultiModalDatasetArguments.is_multimodal">
+<span class="sig-name descname"><span class="pre">is_multimodal</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#MultiModalDatasetArguments.is_multimodal"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.MultiModalDatasetArguments.is_multimodal" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.MultiModalDatasetArguments.use_image_start_end">
+<span class="sig-name descname"><span class="pre">use_image_start_end</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#MultiModalDatasetArguments.use_image_start_end"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.MultiModalDatasetArguments.use_image_start_end" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.MultiModalDatasetArguments.sep_style">
+<span class="sig-name descname"><span class="pre">sep_style</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'plain'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#MultiModalDatasetArguments.sep_style"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.MultiModalDatasetArguments.sep_style" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">FinetunerArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">transformers.TrainingArguments</span></code></p>
+<p>Adapt transformers.TrainingArguments</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.eval_dataset_path">
+<span class="sig-name descname"><span class="pre">eval_dataset_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.eval_dataset_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.eval_dataset_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.remove_unused_columns">
+<span class="sig-name descname"><span class="pre">remove_unused_columns</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.remove_unused_columns"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.remove_unused_columns" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.finetune_part">
+<span class="sig-name descname"><span class="pre">finetune_part</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'language_projection'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.finetune_part"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.finetune_part" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.save_language_projection">
+<span class="sig-name descname"><span class="pre">save_language_projection</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.save_language_projection"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.save_language_projection" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.use_lisa">
+<span class="sig-name descname"><span class="pre">use_lisa</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.use_lisa"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.use_lisa" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.lisa_activated_layers">
+<span class="sig-name descname"><span class="pre">lisa_activated_layers</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">2</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.lisa_activated_layers"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.lisa_activated_layers" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.lisa_interval_steps">
+<span class="sig-name descname"><span class="pre">lisa_interval_steps</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">20</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.lisa_interval_steps"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.lisa_interval_steps" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.lisa_layers_attribute">
+<span class="sig-name descname"><span class="pre">lisa_layers_attribute</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'model.model.layers'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.lisa_layers_attribute"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.lisa_layers_attribute" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.use_customized_optim">
+<span class="sig-name descname"><span class="pre">use_customized_optim</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.use_customized_optim"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.use_customized_optim" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.customized_optim">
+<span class="sig-name descname"><span class="pre">customized_optim</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'sign_sgd'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.customized_optim"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.customized_optim" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.customized_optim_args">
+<span class="sig-name descname"><span class="pre">customized_optim_args</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.customized_optim_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.customized_optim_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.optim_dummy_beta1">
+<span class="sig-name descname"><span class="pre">optim_dummy_beta1</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.9</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.optim_dummy_beta1"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.optim_dummy_beta1" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.optim_dummy_beta2">
+<span class="sig-name descname"><span class="pre">optim_dummy_beta2</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.999</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.optim_dummy_beta2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.optim_dummy_beta2" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.optim_adam_beta1">
+<span class="sig-name descname"><span class="pre">optim_adam_beta1</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.9</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.optim_adam_beta1"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.optim_adam_beta1" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.optim_adam_beta2">
+<span class="sig-name descname"><span class="pre">optim_adam_beta2</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.999</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.optim_adam_beta2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.optim_adam_beta2" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.optim_beta1">
+<span class="sig-name descname"><span class="pre">optim_beta1</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.9</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.optim_beta1"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.optim_beta1" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.optim_beta2">
+<span class="sig-name descname"><span class="pre">optim_beta2</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.999</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.optim_beta2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.optim_beta2" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.optim_beta3">
+<span class="sig-name descname"><span class="pre">optim_beta3</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.9</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.optim_beta3"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.optim_beta3" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.optim_momentum">
+<span class="sig-name descname"><span class="pre">optim_momentum</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.999</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.optim_momentum"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.optim_momentum" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.FinetunerArguments.optim_weight_decay">
+<span class="sig-name descname"><span class="pre">optim_weight_decay</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#FinetunerArguments.optim_weight_decay"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.FinetunerArguments.optim_weight_decay" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.RewardModelTunerArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">RewardModelTunerArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#RewardModelTunerArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.RewardModelTunerArguments" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.args.FinetunerArguments" title="lmflow.args.FinetunerArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FinetunerArguments</span></code></a></p>
+<p>Arguments for reward modeling.</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">EvaluatorArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments" title="Link to this definition">#</a></dt>
+<dd><p>Define a class EvaluatorArguments using the dataclass decorator. The class contains several optional
+parameters that can be used to configure a evaluator.</p>
+<dl class="simple">
+<dt>local_rank<span class="classifier">str</span></dt><dd><p>For distributed training: local_rank</p>
+</dd>
+</dl>
+<p>random_shuffle : bool</p>
+<p>use_wandb : bool</p>
+<p>random_seed : int, default = 1</p>
+<p>output_dir : str, default = ‘./output_dir’,</p>
+<dl class="simple">
+<dt>mixed_precision<span class="classifier">str, choice from [“bf16”,”fp16”].</span></dt><dd><p>mixed precision mode, whether to use bf16 or fp16</p>
+</dd>
+<dt>deepspeed :</dt><dd><p>Enable deepspeed and pass the path to deepspeed json config file (e.g. ds_config.json) or an already
+loaded json file as a dict</p>
+</dd>
+<dt>temperature<span class="classifier">float</span></dt><dd><p>An argument of model.generate in huggingface to control the diversity of generation.</p>
+</dd>
+<dt>repetition_penalty<span class="classifier">float</span></dt><dd><p>An argument of model.generate in huggingface to penalize repetitions.</p>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.local_rank">
+<span class="sig-name descname"><span class="pre">local_rank</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">-1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.local_rank"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.local_rank" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.random_shuffle">
+<span class="sig-name descname"><span class="pre">random_shuffle</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.random_shuffle"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.random_shuffle" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.use_wandb">
+<span class="sig-name descname"><span class="pre">use_wandb</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.use_wandb"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.use_wandb" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.random_seed">
+<span class="sig-name descname"><span class="pre">random_seed</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.random_seed"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.random_seed" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.output_dir">
+<span class="sig-name descname"><span class="pre">output_dir</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'./output_dir'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.output_dir"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.output_dir" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.mixed_precision">
+<span class="sig-name descname"><span class="pre">mixed_precision</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'bf16'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.mixed_precision"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.mixed_precision" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.deepspeed">
+<span class="sig-name descname"><span class="pre">deepspeed</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.deepspeed"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.deepspeed" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.answer_type">
+<span class="sig-name descname"><span class="pre">answer_type</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'text'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.answer_type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.answer_type" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.prompt_structure">
+<span class="sig-name descname"><span class="pre">prompt_structure</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'{input}'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.prompt_structure"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.prompt_structure" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.evaluate_block_size">
+<span class="sig-name descname"><span class="pre">evaluate_block_size</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">512</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.evaluate_block_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.evaluate_block_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.metric">
+<span class="sig-name descname"><span class="pre">metric</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'accuracy'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.metric"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.metric" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.inference_batch_size_per_device">
+<span class="sig-name descname"><span class="pre">inference_batch_size_per_device</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.inference_batch_size_per_device"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.inference_batch_size_per_device" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.use_accelerator_for_evaluator">
+<span class="sig-name descname"><span class="pre">use_accelerator_for_evaluator</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.use_accelerator_for_evaluator"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.use_accelerator_for_evaluator" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.temperature">
+<span class="sig-name descname"><span class="pre">temperature</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.temperature"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.temperature" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.repetition_penalty">
+<span class="sig-name descname"><span class="pre">repetition_penalty</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.repetition_penalty"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.repetition_penalty" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.EvaluatorArguments.max_new_tokens">
+<span class="sig-name descname"><span class="pre">max_new_tokens</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">100</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#EvaluatorArguments.max_new_tokens"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.EvaluatorArguments.max_new_tokens" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">InferencerArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments" title="Link to this definition">#</a></dt>
+<dd><p>Define a class InferencerArguments using the dataclass decorator. The class contains several optional
+parameters that can be used to configure a inferencer.</p>
+<dl class="simple">
+<dt>local_rank<span class="classifier">str</span></dt><dd><p>For distributed training: local_rank</p>
+</dd>
+</dl>
+<p>random_seed : int, default = 1
+inference_batch_size : int, default = 1
+deepspeed :</p>
+<blockquote>
+<div><p>Enable deepspeed and pass the path to deepspeed json config file (e.g. ds_config.json) or an already
+loaded json file as a dict</p>
+</div></blockquote>
+<dl>
+<dt>mixed_precision<span class="classifier">str, choice from [“bf16”,”fp16”].</span></dt><dd><p>mixed precision mode, whether to use bf16 or fp16</p>
+</dd>
+<dt>temperature<span class="classifier">float</span></dt><dd><p>An argument of model.generate in huggingface to control the diversity of generation.</p>
+</dd>
+<dt>repetition_penalty<span class="classifier">float</span></dt><dd><p>An argument of model.generate in huggingface to penalize repetitions.</p>
+</dd>
+<dt>use_beam_search<span class="classifier">Optional[bool]</span></dt><dd><p>Whether to use beam search during inference, By default False.</p>
+</dd>
+<dt>num_output_sequences<span class="classifier">Optional[int]</span></dt><dd><p>Number of output sequences to return for the given prompt,
+currently only used in vllm inference, By default 8.</p>
+</dd>
+<dt>top_p<span class="classifier">Optional[float]</span></dt><dd><p>top_p for sampling, By default 1.0.</p>
+</dd>
+<dt>top_k<span class="classifier">Optional[int]</span></dt><dd><p>top_k for sampling, By default -1 (no top_k).</p>
+</dd>
+<dt>additional_stop_token_ids<span class="classifier">Optional[List[int]]</span></dt><dd><p>the ids of the end of sentence tokens, By default [].</p>
+</dd>
+<dt>apply_chat_template<span class="classifier">Optional[bool]</span></dt><dd><p>Whether to apply chat template, By default True.</p>
+</dd>
+<dt>save_results<span class="classifier">Optional[bool]</span></dt><dd><p>Whether to save inference results, By default False.</p>
+</dd>
+<dt>results_path<span class="classifier">Optional[str]</span></dt><dd><p>The <strong>json file</strong> path of inference results, By default None.</p>
+</dd>
+<dt>enable_decode_inference_result<span class="classifier">Optional[bool]</span></dt><dd><p>Whether to detokenize the inference results.</p>
+<p>NOTE: For iterative align pipelines, whether to detokenize depends on
+the homogeneity of the policy model and the reward model
+(i.e., if they have the same tokenizer).</p>
+</dd>
+<dt>use_vllm: bool, optional</dt><dd><p>Whether to use VLLM for inference, By default False.</p>
+</dd>
+<dt>vllm_tensor_parallel_size: int, optional</dt><dd><p>The tensor parallel size for VLLM inference.</p>
+</dd>
+<dt>vllm_gpu_memory_utilization: float, optional</dt><dd><p>The GPU memory utilization for VLLM inference. The proportion of GPU
+memory (per GPU) to use for VLLM inference.</p>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.device">
+<span class="sig-name descname"><span class="pre">device</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'gpu'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.device"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.device" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.local_rank">
+<span class="sig-name descname"><span class="pre">local_rank</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">-1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.local_rank"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.local_rank" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.inference_batch_size">
+<span class="sig-name descname"><span class="pre">inference_batch_size</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.inference_batch_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.inference_batch_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.vllm_inference_batch_size">
+<span class="sig-name descname"><span class="pre">vllm_inference_batch_size</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.vllm_inference_batch_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.vllm_inference_batch_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.temperature">
+<span class="sig-name descname"><span class="pre">temperature</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.0</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.temperature"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.temperature" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.repetition_penalty">
+<span class="sig-name descname"><span class="pre">repetition_penalty</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.repetition_penalty"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.repetition_penalty" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.max_new_tokens">
+<span class="sig-name descname"><span class="pre">max_new_tokens</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">100</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.max_new_tokens"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.max_new_tokens" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.random_seed">
+<span class="sig-name descname"><span class="pre">random_seed</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.random_seed"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.random_seed" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.deepspeed">
+<span class="sig-name descname"><span class="pre">deepspeed</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.deepspeed"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.deepspeed" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.mixed_precision">
+<span class="sig-name descname"><span class="pre">mixed_precision</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'bf16'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.mixed_precision"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.mixed_precision" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.do_sample">
+<span class="sig-name descname"><span class="pre">do_sample</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.do_sample"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.do_sample" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.use_accelerator">
+<span class="sig-name descname"><span class="pre">use_accelerator</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.use_accelerator"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.use_accelerator" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.use_beam_search">
+<span class="sig-name descname"><span class="pre">use_beam_search</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.use_beam_search"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.use_beam_search" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.num_output_sequences">
+<span class="sig-name descname"><span class="pre">num_output_sequences</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">8</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.num_output_sequences"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.num_output_sequences" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.top_p">
+<span class="sig-name descname"><span class="pre">top_p</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1.0</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.top_p"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.top_p" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.top_k">
+<span class="sig-name descname"><span class="pre">top_k</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">-1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.top_k"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.top_k" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.additional_stop_token_ids">
+<span class="sig-name descname"><span class="pre">additional_stop_token_ids</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">[]</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.additional_stop_token_ids"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.additional_stop_token_ids" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.apply_chat_template">
+<span class="sig-name descname"><span class="pre">apply_chat_template</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.apply_chat_template"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.apply_chat_template" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.enable_decode_inference_result">
+<span class="sig-name descname"><span class="pre">enable_decode_inference_result</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.enable_decode_inference_result"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.enable_decode_inference_result" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.tensor_parallel_size">
+<span class="sig-name descname"><span class="pre">tensor_parallel_size</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.tensor_parallel_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.tensor_parallel_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.enable_distributed_inference">
+<span class="sig-name descname"><span class="pre">enable_distributed_inference</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.enable_distributed_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.enable_distributed_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.distributed_inference_num_instances">
+<span class="sig-name descname"><span class="pre">distributed_inference_num_instances</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.distributed_inference_num_instances"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.distributed_inference_num_instances" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.use_vllm">
+<span class="sig-name descname"><span class="pre">use_vllm</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.use_vllm"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.use_vllm" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.vllm_tensor_parallel_size">
+<span class="sig-name descname"><span class="pre">vllm_tensor_parallel_size</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.vllm_tensor_parallel_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.vllm_tensor_parallel_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.vllm_gpu_memory_utilization">
+<span class="sig-name descname"><span class="pre">vllm_gpu_memory_utilization</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.95</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.vllm_gpu_memory_utilization"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.vllm_gpu_memory_utilization" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.save_results">
+<span class="sig-name descname"><span class="pre">save_results</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.save_results"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.save_results" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.results_path">
+<span class="sig-name descname"><span class="pre">results_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.results_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.results_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.args.InferencerArguments.__post_init__">
+<span class="sig-name descname"><span class="pre">__post_init__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/args.html#InferencerArguments.__post_init__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.InferencerArguments.__post_init__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.RaftAlignerArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">RaftAlignerArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#RaftAlignerArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.RaftAlignerArguments" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">transformers.TrainingArguments</span></code></p>
+<p>Define a class RaftAlignerArguments to configure raft aligner.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.RaftAlignerArguments.output_reward_path">
+<span class="sig-name descname"><span class="pre">output_reward_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'tmp/raft_aligner/'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#RaftAlignerArguments.output_reward_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.RaftAlignerArguments.output_reward_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.RaftAlignerArguments.output_min_length">
+<span class="sig-name descname"><span class="pre">output_min_length</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">64</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#RaftAlignerArguments.output_min_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.RaftAlignerArguments.output_min_length" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.RaftAlignerArguments.output_max_length">
+<span class="sig-name descname"><span class="pre">output_max_length</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">128</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#RaftAlignerArguments.output_max_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.RaftAlignerArguments.output_max_length" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.RaftAlignerArguments.num_raft_iteration">
+<span class="sig-name descname"><span class="pre">num_raft_iteration</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">20</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#RaftAlignerArguments.num_raft_iteration"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.RaftAlignerArguments.num_raft_iteration" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.RaftAlignerArguments.raft_batch_size">
+<span class="sig-name descname"><span class="pre">raft_batch_size</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1024</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#RaftAlignerArguments.raft_batch_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.RaftAlignerArguments.raft_batch_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.RaftAlignerArguments.top_reward_percentage">
+<span class="sig-name descname"><span class="pre">top_reward_percentage</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.2</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#RaftAlignerArguments.top_reward_percentage"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.RaftAlignerArguments.top_reward_percentage" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.RaftAlignerArguments.inference_batch_size_per_device">
+<span class="sig-name descname"><span class="pre">inference_batch_size_per_device</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#RaftAlignerArguments.inference_batch_size_per_device"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.RaftAlignerArguments.inference_batch_size_per_device" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.RaftAlignerArguments.collection_strategy">
+<span class="sig-name descname"><span class="pre">collection_strategy</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'top'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#RaftAlignerArguments.collection_strategy"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.RaftAlignerArguments.collection_strategy" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.BenchmarkingArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">BenchmarkingArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#BenchmarkingArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.BenchmarkingArguments" title="Link to this definition">#</a></dt>
+<dd><dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.BenchmarkingArguments.dataset_name">
+<span class="sig-name descname"><span class="pre">dataset_name</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#BenchmarkingArguments.dataset_name"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.BenchmarkingArguments.dataset_name" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.BenchmarkingArguments.lm_evaluation_metric">
+<span class="sig-name descname"><span class="pre">lm_evaluation_metric</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'accuracy'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#BenchmarkingArguments.lm_evaluation_metric"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.BenchmarkingArguments.lm_evaluation_metric" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">DPOAlignerArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments" title="Link to this definition">#</a></dt>
+<dd><p>The arguments for the DPO training script.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.local_rank">
+<span class="sig-name descname"><span class="pre">local_rank</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">-1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.local_rank"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.local_rank" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.beta">
+<span class="sig-name descname"><span class="pre">beta</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.beta"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.beta" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.learning_rate">
+<span class="sig-name descname"><span class="pre">learning_rate</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.0005</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.learning_rate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.learning_rate" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.lr_scheduler_type">
+<span class="sig-name descname"><span class="pre">lr_scheduler_type</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'cosine'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.lr_scheduler_type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.lr_scheduler_type" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.warmup_steps">
+<span class="sig-name descname"><span class="pre">warmup_steps</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">100</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.warmup_steps"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.warmup_steps" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.weight_decay">
+<span class="sig-name descname"><span class="pre">weight_decay</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.05</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.weight_decay"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.weight_decay" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.optimizer_type">
+<span class="sig-name descname"><span class="pre">optimizer_type</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'paged_adamw_32bit'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.optimizer_type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.optimizer_type" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.per_device_train_batch_size">
+<span class="sig-name descname"><span class="pre">per_device_train_batch_size</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">4</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.per_device_train_batch_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.per_device_train_batch_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.per_device_eval_batch_size">
+<span class="sig-name descname"><span class="pre">per_device_eval_batch_size</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.per_device_eval_batch_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.per_device_eval_batch_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.gradient_accumulation_steps">
+<span class="sig-name descname"><span class="pre">gradient_accumulation_steps</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">4</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.gradient_accumulation_steps"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.gradient_accumulation_steps" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.gradient_checkpointing">
+<span class="sig-name descname"><span class="pre">gradient_checkpointing</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.gradient_checkpointing"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.gradient_checkpointing" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.gradient_checkpointing_use_reentrant">
+<span class="sig-name descname"><span class="pre">gradient_checkpointing_use_reentrant</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.gradient_checkpointing_use_reentrant"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.gradient_checkpointing_use_reentrant" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.max_prompt_length">
+<span class="sig-name descname"><span class="pre">max_prompt_length</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">512</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.max_prompt_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.max_prompt_length" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.max_length">
+<span class="sig-name descname"><span class="pre">max_length</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1024</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.max_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.max_length" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.max_steps">
+<span class="sig-name descname"><span class="pre">max_steps</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1000</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.max_steps"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.max_steps" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.logging_steps">
+<span class="sig-name descname"><span class="pre">logging_steps</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">10</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.logging_steps"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.logging_steps" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.save_steps">
+<span class="sig-name descname"><span class="pre">save_steps</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">100</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.save_steps"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.save_steps" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.eval_steps">
+<span class="sig-name descname"><span class="pre">eval_steps</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">100</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.eval_steps"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.eval_steps" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.output_dir">
+<span class="sig-name descname"><span class="pre">output_dir</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'./results'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.output_dir"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.output_dir" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.log_freq">
+<span class="sig-name descname"><span class="pre">log_freq</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.log_freq"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.log_freq" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.sanity_check">
+<span class="sig-name descname"><span class="pre">sanity_check</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.sanity_check"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.sanity_check" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.report_to">
+<span class="sig-name descname"><span class="pre">report_to</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'wandb'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.report_to"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.report_to" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.seed">
+<span class="sig-name descname"><span class="pre">seed</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.seed"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.seed" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.run_name">
+<span class="sig-name descname"><span class="pre">run_name</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'dpo'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.run_name"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.run_name" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOAlignerArguments.eval_dataset_path">
+<span class="sig-name descname"><span class="pre">eval_dataset_path</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOAlignerArguments.eval_dataset_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOAlignerArguments.eval_dataset_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.DPOv2AlignerArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">DPOv2AlignerArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOv2AlignerArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOv2AlignerArguments" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.args.FinetunerArguments" title="lmflow.args.FinetunerArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FinetunerArguments</span></code></a></p>
+<p>The arguments for the DPOv2 training script.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOv2AlignerArguments.random_seed">
+<span class="sig-name descname"><span class="pre">random_seed</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">42</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOv2AlignerArguments.random_seed"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOv2AlignerArguments.random_seed" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOv2AlignerArguments.accelerate_config_file">
+<span class="sig-name descname"><span class="pre">accelerate_config_file</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOv2AlignerArguments.accelerate_config_file"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOv2AlignerArguments.accelerate_config_file" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOv2AlignerArguments.margin_scale">
+<span class="sig-name descname"><span class="pre">margin_scale</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1.0</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOv2AlignerArguments.margin_scale"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOv2AlignerArguments.margin_scale" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOv2AlignerArguments.sampling_paired_method">
+<span class="sig-name descname"><span class="pre">sampling_paired_method</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'max_random'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOv2AlignerArguments.sampling_paired_method"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOv2AlignerArguments.sampling_paired_method" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOv2AlignerArguments.length_penalty">
+<span class="sig-name descname"><span class="pre">length_penalty</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOv2AlignerArguments.length_penalty"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOv2AlignerArguments.length_penalty" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOv2AlignerArguments.max_length">
+<span class="sig-name descname"><span class="pre">max_length</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">2048</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOv2AlignerArguments.max_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOv2AlignerArguments.max_length" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOv2AlignerArguments.max_prompt_length">
+<span class="sig-name descname"><span class="pre">max_prompt_length</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1000</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOv2AlignerArguments.max_prompt_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOv2AlignerArguments.max_prompt_length" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOv2AlignerArguments.mask_prompt">
+<span class="sig-name descname"><span class="pre">mask_prompt</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOv2AlignerArguments.mask_prompt"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOv2AlignerArguments.mask_prompt" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOv2AlignerArguments.beta">
+<span class="sig-name descname"><span class="pre">beta</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">float</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0.1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOv2AlignerArguments.beta"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOv2AlignerArguments.beta" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.DPOv2AlignerArguments.loss_type">
+<span class="sig-name descname"><span class="pre">loss_type</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'sigmoid'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#DPOv2AlignerArguments.loss_type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.DPOv2AlignerArguments.loss_type" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.IterativeAlignerArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">IterativeAlignerArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#IterativeAlignerArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.IterativeAlignerArguments" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.args.InferencerArguments" title="lmflow.args.InferencerArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">InferencerArguments</span></code></a></p>
+<p>Arguments for iterative aligners.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.IterativeAlignerArguments.dataset_path_list">
+<span class="sig-name descname"><span class="pre">dataset_path_list</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">[]</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#IterativeAlignerArguments.dataset_path_list"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.IterativeAlignerArguments.dataset_path_list" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.IterativeAlignerArguments.initial_iter_idx">
+<span class="sig-name descname"><span class="pre">initial_iter_idx</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#IterativeAlignerArguments.initial_iter_idx"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.IterativeAlignerArguments.initial_iter_idx" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.IterativeDPOAlignerArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">IterativeDPOAlignerArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#IterativeDPOAlignerArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.IterativeDPOAlignerArguments" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.args.IterativeAlignerArguments" title="lmflow.args.IterativeAlignerArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">IterativeAlignerArguments</span></code></a>, <a class="reference internal" href="#lmflow.args.DPOv2AlignerArguments" title="lmflow.args.DPOv2AlignerArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DPOv2AlignerArguments</span></code></a></p>
+<p>Arguments for iterative DPO aligners.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.IterativeDPOAlignerArguments.output_dir">
+<span class="sig-name descname"><span class="pre">output_dir</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'./runs'</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#IterativeDPOAlignerArguments.output_dir"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.IterativeDPOAlignerArguments.output_dir" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.IterativeDPOAlignerArguments.reward_model_inference_batch_size">
+<span class="sig-name descname"><span class="pre">reward_model_inference_batch_size</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">1</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#IterativeDPOAlignerArguments.reward_model_inference_batch_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.IterativeDPOAlignerArguments.reward_model_inference_batch_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.IterativeDPOAlignerArguments.reward_model_inference_block_size">
+<span class="sig-name descname"><span class="pre">reward_model_inference_block_size</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">2048</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#IterativeDPOAlignerArguments.reward_model_inference_block_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.IterativeDPOAlignerArguments.reward_model_inference_block_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.IterativeDPOAlignerArguments.do_response_generation">
+<span class="sig-name descname"><span class="pre">do_response_generation</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#IterativeDPOAlignerArguments.do_response_generation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.IterativeDPOAlignerArguments.do_response_generation" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.IterativeDPOAlignerArguments.do_scoring">
+<span class="sig-name descname"><span class="pre">do_scoring</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#IterativeDPOAlignerArguments.do_scoring"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.IterativeDPOAlignerArguments.do_scoring" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.args.IterativeDPOAlignerArguments.do_dpo_align">
+<span class="sig-name descname"><span class="pre">do_dpo_align</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../_modules/lmflow/args.html#IterativeDPOAlignerArguments.do_dpo_align"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.IterativeDPOAlignerArguments.do_dpo_align" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.args.PIPELINE_ARGUMENT_MAPPING">
+<span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">PIPELINE_ARGUMENT_MAPPING</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#PIPELINE_ARGUMENT_MAPPING"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.PIPELINE_ARGUMENT_MAPPING" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.args.AutoArguments">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">AutoArguments</span></span><a class="reference internal" href="../../../_modules/lmflow/args.html#AutoArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.AutoArguments" title="Link to this definition">#</a></dt>
+<dd><p>Automatically choose arguments from FinetunerArguments or EvaluatorArguments.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.args.AutoArguments.get_pipeline_args_class">
+<span class="sig-name descname"><span class="pre">get_pipeline_args_class</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/args.html#AutoArguments.get_pipeline_args_class"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.AutoArguments.get_pipeline_args_class" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.args.split_args">
+<span class="sig-prename descclassname"><span class="pre">lmflow.args.</span></span><span class="sig-name descname"><span class="pre">split_args</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">args</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/args.html#split_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.args.split_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../datasets/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.datasets</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.MODEL_CONFIG_CLASSES"><code class="docutils literal notranslate"><span class="pre">MODEL_CONFIG_CLASSES</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.MODEL_TYPES"><code class="docutils literal notranslate"><span class="pre">MODEL_TYPES</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames"><code class="docutils literal notranslate"><span class="pre">OptimizerNames</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.DUMMY"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.DUMMY</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.ADABELIEF"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.ADABELIEF</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.ADABOUND"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.ADABOUND</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.LARS"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.LARS</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.LAMB"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.LAMB</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.ADAMAX"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.ADAMAX</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.NADAM"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.NADAM</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.RADAM"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.RADAM</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.ADAMP"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.ADAMP</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.SGDP"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.SGDP</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.YOGI"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.YOGI</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.SOPHIA"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.SOPHIA</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.ADAN"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.ADAN</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.ADAM"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.ADAM</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.NOVOGRAD"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.NOVOGRAD</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.ADADELTA"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.ADADELTA</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.ADAGRAD"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.ADAGRAD</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.ADAMW_SCHEDULE_FREE"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.ADAMW_SCHEDULE_FREE</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.OptimizerNames.SGD_SCHEDULE_FREE"><code class="docutils literal notranslate"><span class="pre">OptimizerNames.SGD_SCHEDULE_FREE</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments"><code class="docutils literal notranslate"><span class="pre">ModelArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.model_name_or_path"><code class="docutils literal notranslate"><span class="pre">ModelArguments.model_name_or_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.lora_model_path"><code class="docutils literal notranslate"><span class="pre">ModelArguments.lora_model_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.model_type"><code class="docutils literal notranslate"><span class="pre">ModelArguments.model_type</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.config_overrides"><code class="docutils literal notranslate"><span class="pre">ModelArguments.config_overrides</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.arch_type"><code class="docutils literal notranslate"><span class="pre">ModelArguments.arch_type</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.config_name"><code class="docutils literal notranslate"><span class="pre">ModelArguments.config_name</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.tokenizer_name"><code class="docutils literal notranslate"><span class="pre">ModelArguments.tokenizer_name</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.cache_dir"><code class="docutils literal notranslate"><span class="pre">ModelArguments.cache_dir</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.use_fast_tokenizer"><code class="docutils literal notranslate"><span class="pre">ModelArguments.use_fast_tokenizer</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.model_revision"><code class="docutils literal notranslate"><span class="pre">ModelArguments.model_revision</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.token"><code class="docutils literal notranslate"><span class="pre">ModelArguments.token</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.trust_remote_code"><code class="docutils literal notranslate"><span class="pre">ModelArguments.trust_remote_code</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.torch_dtype"><code class="docutils literal notranslate"><span class="pre">ModelArguments.torch_dtype</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.use_dora"><code class="docutils literal notranslate"><span class="pre">ModelArguments.use_dora</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.use_lora"><code class="docutils literal notranslate"><span class="pre">ModelArguments.use_lora</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.use_qlora"><code class="docutils literal notranslate"><span class="pre">ModelArguments.use_qlora</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.bits"><code class="docutils literal notranslate"><span class="pre">ModelArguments.bits</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.quant_type"><code class="docutils literal notranslate"><span class="pre">ModelArguments.quant_type</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.double_quant"><code class="docutils literal notranslate"><span class="pre">ModelArguments.double_quant</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.lora_r"><code class="docutils literal notranslate"><span class="pre">ModelArguments.lora_r</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.lora_alpha"><code class="docutils literal notranslate"><span class="pre">ModelArguments.lora_alpha</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.lora_target_modules"><code class="docutils literal notranslate"><span class="pre">ModelArguments.lora_target_modules</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.lora_dropout"><code class="docutils literal notranslate"><span class="pre">ModelArguments.lora_dropout</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.save_aggregated_lora"><code class="docutils literal notranslate"><span class="pre">ModelArguments.save_aggregated_lora</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.use_ram_optimized_load"><code class="docutils literal notranslate"><span class="pre">ModelArguments.use_ram_optimized_load</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.use_flash_attention"><code class="docutils literal notranslate"><span class="pre">ModelArguments.use_flash_attention</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.truncate_to_model_max_length"><code class="docutils literal notranslate"><span class="pre">ModelArguments.truncate_to_model_max_length</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.do_rope_scaling"><code class="docutils literal notranslate"><span class="pre">ModelArguments.do_rope_scaling</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.rope_pi_ratio"><code class="docutils literal notranslate"><span class="pre">ModelArguments.rope_pi_ratio</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.rope_ntk_ratio"><code class="docutils literal notranslate"><span class="pre">ModelArguments.rope_ntk_ratio</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.use_int8"><code class="docutils literal notranslate"><span class="pre">ModelArguments.use_int8</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.load_in_4bit"><code class="docutils literal notranslate"><span class="pre">ModelArguments.load_in_4bit</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.model_max_length"><code class="docutils literal notranslate"><span class="pre">ModelArguments.model_max_length</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.truncation_side"><code class="docutils literal notranslate"><span class="pre">ModelArguments.truncation_side</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.padding_side"><code class="docutils literal notranslate"><span class="pre">ModelArguments.padding_side</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.eos_padding"><code class="docutils literal notranslate"><span class="pre">ModelArguments.eos_padding</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.ignore_bias_buffers"><code class="docutils literal notranslate"><span class="pre">ModelArguments.ignore_bias_buffers</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.ModelArguments.__post_init__"><code class="docutils literal notranslate"><span class="pre">ModelArguments.__post_init__()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments"><code class="docutils literal notranslate"><span class="pre">VisModelArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.low_resource"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.low_resource</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.custom_model"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.custom_model</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.pretrained_language_projection_path"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.pretrained_language_projection_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.custom_vision_model"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.custom_vision_model</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.image_encoder_name_or_path"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.image_encoder_name_or_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.qformer_name_or_path"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.qformer_name_or_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.llm_model_name_or_path"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.llm_model_name_or_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.use_prompt_cache"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.use_prompt_cache</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.prompt_cache_path"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.prompt_cache_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.llava_loading"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.llava_loading</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.with_qformer"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.with_qformer</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.vision_select_layer"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.vision_select_layer</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.llava_pretrain_model_path"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.llava_pretrain_model_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.VisModelArguments.save_pretrain_model_path"><code class="docutils literal notranslate"><span class="pre">VisModelArguments.save_pretrain_model_path</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments"><code class="docutils literal notranslate"><span class="pre">DatasetArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.dataset_path"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.dataset_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.dataset_name"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.dataset_name</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.is_custom_dataset"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.is_custom_dataset</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.customized_cache_dir"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.customized_cache_dir</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.dataset_config_name"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.dataset_config_name</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.train_file"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.train_file</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.validation_file"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.validation_file</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.max_train_samples"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.max_train_samples</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.max_eval_samples"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.max_eval_samples</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.streaming"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.streaming</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.block_size"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.block_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.overwrite_cache"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.overwrite_cache</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.validation_split_percentage"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.validation_split_percentage</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.preprocessing_num_workers"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.preprocessing_num_workers</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.group_texts_batch_size"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.group_texts_batch_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.disable_group_texts"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.disable_group_texts</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.keep_linebreaks"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.keep_linebreaks</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.test_file"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.test_file</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.train_on_prompt"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.train_on_prompt</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.conversation_template"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.conversation_template</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DatasetArguments.__post_init__"><code class="docutils literal notranslate"><span class="pre">DatasetArguments.__post_init__()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.MultiModalDatasetArguments"><code class="docutils literal notranslate"><span class="pre">MultiModalDatasetArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.MultiModalDatasetArguments.image_folder"><code class="docutils literal notranslate"><span class="pre">MultiModalDatasetArguments.image_folder</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.MultiModalDatasetArguments.image_aspect_ratio"><code class="docutils literal notranslate"><span class="pre">MultiModalDatasetArguments.image_aspect_ratio</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.MultiModalDatasetArguments.is_multimodal"><code class="docutils literal notranslate"><span class="pre">MultiModalDatasetArguments.is_multimodal</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.MultiModalDatasetArguments.use_image_start_end"><code class="docutils literal notranslate"><span class="pre">MultiModalDatasetArguments.use_image_start_end</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.MultiModalDatasetArguments.sep_style"><code class="docutils literal notranslate"><span class="pre">MultiModalDatasetArguments.sep_style</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.eval_dataset_path"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.eval_dataset_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.remove_unused_columns"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.remove_unused_columns</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.finetune_part"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.finetune_part</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.save_language_projection"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.save_language_projection</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.use_lisa"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.use_lisa</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.lisa_activated_layers"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.lisa_activated_layers</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.lisa_interval_steps"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.lisa_interval_steps</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.lisa_layers_attribute"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.lisa_layers_attribute</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.use_customized_optim"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.use_customized_optim</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.customized_optim"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.customized_optim</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.customized_optim_args"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.customized_optim_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.optim_dummy_beta1"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.optim_dummy_beta1</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.optim_dummy_beta2"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.optim_dummy_beta2</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.optim_adam_beta1"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.optim_adam_beta1</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.optim_adam_beta2"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.optim_adam_beta2</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.optim_beta1"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.optim_beta1</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.optim_beta2"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.optim_beta2</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.optim_beta3"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.optim_beta3</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.optim_momentum"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.optim_momentum</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.FinetunerArguments.optim_weight_decay"><code class="docutils literal notranslate"><span class="pre">FinetunerArguments.optim_weight_decay</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.RewardModelTunerArguments"><code class="docutils literal notranslate"><span class="pre">RewardModelTunerArguments</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.local_rank"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.local_rank</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.random_shuffle"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.random_shuffle</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.use_wandb"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.use_wandb</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.random_seed"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.random_seed</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.output_dir"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.output_dir</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.mixed_precision"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.mixed_precision</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.deepspeed"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.deepspeed</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.answer_type"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.answer_type</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.prompt_structure"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.prompt_structure</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.evaluate_block_size"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.evaluate_block_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.metric"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.metric</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.inference_batch_size_per_device"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.inference_batch_size_per_device</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.use_accelerator_for_evaluator"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.use_accelerator_for_evaluator</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.temperature"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.temperature</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.repetition_penalty"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.repetition_penalty</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.EvaluatorArguments.max_new_tokens"><code class="docutils literal notranslate"><span class="pre">EvaluatorArguments.max_new_tokens</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments"><code class="docutils literal notranslate"><span class="pre">InferencerArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.device"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.device</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.local_rank"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.local_rank</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.inference_batch_size"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.inference_batch_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.vllm_inference_batch_size"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.vllm_inference_batch_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.temperature"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.temperature</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.repetition_penalty"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.repetition_penalty</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.max_new_tokens"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.max_new_tokens</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.random_seed"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.random_seed</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.deepspeed"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.deepspeed</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.mixed_precision"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.mixed_precision</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.do_sample"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.do_sample</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.use_accelerator"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.use_accelerator</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.use_beam_search"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.use_beam_search</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.num_output_sequences"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.num_output_sequences</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.top_p"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.top_p</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.top_k"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.top_k</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.additional_stop_token_ids"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.additional_stop_token_ids</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.apply_chat_template"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.apply_chat_template</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.enable_decode_inference_result"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.enable_decode_inference_result</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.tensor_parallel_size"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.tensor_parallel_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.enable_distributed_inference"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.enable_distributed_inference</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.distributed_inference_num_instances"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.distributed_inference_num_instances</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.use_vllm"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.use_vllm</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.vllm_tensor_parallel_size"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.vllm_tensor_parallel_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.vllm_gpu_memory_utilization"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.vllm_gpu_memory_utilization</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.save_results"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.save_results</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.results_path"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.results_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.InferencerArguments.__post_init__"><code class="docutils literal notranslate"><span class="pre">InferencerArguments.__post_init__()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.RaftAlignerArguments"><code class="docutils literal notranslate"><span class="pre">RaftAlignerArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.RaftAlignerArguments.output_reward_path"><code class="docutils literal notranslate"><span class="pre">RaftAlignerArguments.output_reward_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.RaftAlignerArguments.output_min_length"><code class="docutils literal notranslate"><span class="pre">RaftAlignerArguments.output_min_length</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.RaftAlignerArguments.output_max_length"><code class="docutils literal notranslate"><span class="pre">RaftAlignerArguments.output_max_length</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.RaftAlignerArguments.num_raft_iteration"><code class="docutils literal notranslate"><span class="pre">RaftAlignerArguments.num_raft_iteration</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.RaftAlignerArguments.raft_batch_size"><code class="docutils literal notranslate"><span class="pre">RaftAlignerArguments.raft_batch_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.RaftAlignerArguments.top_reward_percentage"><code class="docutils literal notranslate"><span class="pre">RaftAlignerArguments.top_reward_percentage</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.RaftAlignerArguments.inference_batch_size_per_device"><code class="docutils literal notranslate"><span class="pre">RaftAlignerArguments.inference_batch_size_per_device</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.RaftAlignerArguments.collection_strategy"><code class="docutils literal notranslate"><span class="pre">RaftAlignerArguments.collection_strategy</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.BenchmarkingArguments"><code class="docutils literal notranslate"><span class="pre">BenchmarkingArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.BenchmarkingArguments.dataset_name"><code class="docutils literal notranslate"><span class="pre">BenchmarkingArguments.dataset_name</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.BenchmarkingArguments.lm_evaluation_metric"><code class="docutils literal notranslate"><span class="pre">BenchmarkingArguments.lm_evaluation_metric</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.local_rank"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.local_rank</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.beta"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.beta</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.learning_rate"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.learning_rate</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.lr_scheduler_type"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.lr_scheduler_type</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.warmup_steps"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.warmup_steps</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.weight_decay"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.weight_decay</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.optimizer_type"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.optimizer_type</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.per_device_train_batch_size"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.per_device_train_batch_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.per_device_eval_batch_size"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.per_device_eval_batch_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.gradient_accumulation_steps"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.gradient_accumulation_steps</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.gradient_checkpointing"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.gradient_checkpointing</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.gradient_checkpointing_use_reentrant"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.gradient_checkpointing_use_reentrant</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.max_prompt_length"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.max_prompt_length</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.max_length"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.max_length</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.max_steps"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.max_steps</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.logging_steps"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.logging_steps</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.save_steps"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.save_steps</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.eval_steps"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.eval_steps</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.output_dir"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.output_dir</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.log_freq"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.log_freq</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.sanity_check"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.sanity_check</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.report_to"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.report_to</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.seed"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.seed</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.run_name"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.run_name</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOAlignerArguments.eval_dataset_path"><code class="docutils literal notranslate"><span class="pre">DPOAlignerArguments.eval_dataset_path</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOv2AlignerArguments"><code class="docutils literal notranslate"><span class="pre">DPOv2AlignerArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOv2AlignerArguments.random_seed"><code class="docutils literal notranslate"><span class="pre">DPOv2AlignerArguments.random_seed</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOv2AlignerArguments.accelerate_config_file"><code class="docutils literal notranslate"><span class="pre">DPOv2AlignerArguments.accelerate_config_file</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOv2AlignerArguments.margin_scale"><code class="docutils literal notranslate"><span class="pre">DPOv2AlignerArguments.margin_scale</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOv2AlignerArguments.sampling_paired_method"><code class="docutils literal notranslate"><span class="pre">DPOv2AlignerArguments.sampling_paired_method</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOv2AlignerArguments.length_penalty"><code class="docutils literal notranslate"><span class="pre">DPOv2AlignerArguments.length_penalty</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOv2AlignerArguments.max_length"><code class="docutils literal notranslate"><span class="pre">DPOv2AlignerArguments.max_length</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOv2AlignerArguments.max_prompt_length"><code class="docutils literal notranslate"><span class="pre">DPOv2AlignerArguments.max_prompt_length</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOv2AlignerArguments.mask_prompt"><code class="docutils literal notranslate"><span class="pre">DPOv2AlignerArguments.mask_prompt</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOv2AlignerArguments.beta"><code class="docutils literal notranslate"><span class="pre">DPOv2AlignerArguments.beta</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.DPOv2AlignerArguments.loss_type"><code class="docutils literal notranslate"><span class="pre">DPOv2AlignerArguments.loss_type</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.IterativeAlignerArguments"><code class="docutils literal notranslate"><span class="pre">IterativeAlignerArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.IterativeAlignerArguments.dataset_path_list"><code class="docutils literal notranslate"><span class="pre">IterativeAlignerArguments.dataset_path_list</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.IterativeAlignerArguments.initial_iter_idx"><code class="docutils literal notranslate"><span class="pre">IterativeAlignerArguments.initial_iter_idx</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.IterativeDPOAlignerArguments"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAlignerArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.IterativeDPOAlignerArguments.output_dir"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAlignerArguments.output_dir</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.IterativeDPOAlignerArguments.reward_model_inference_batch_size"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAlignerArguments.reward_model_inference_batch_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.IterativeDPOAlignerArguments.reward_model_inference_block_size"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAlignerArguments.reward_model_inference_block_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.IterativeDPOAlignerArguments.do_response_generation"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAlignerArguments.do_response_generation</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.IterativeDPOAlignerArguments.do_scoring"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAlignerArguments.do_scoring</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.IterativeDPOAlignerArguments.do_dpo_align"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAlignerArguments.do_dpo_align</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.PIPELINE_ARGUMENT_MAPPING"><code class="docutils literal notranslate"><span class="pre">PIPELINE_ARGUMENT_MAPPING</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.AutoArguments"><code class="docutils literal notranslate"><span class="pre">AutoArguments</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.AutoArguments.get_pipeline_args_class"><code class="docutils literal notranslate"><span class="pre">AutoArguments.get_pipeline_args_class()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.args.split_args"><code class="docutils literal notranslate"><span class="pre">split_args()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../_sources/autoapi/lmflow/args/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/datasets/dataset/index.html b/autoapi/lmflow/datasets/dataset/index.html
new file mode 100644
index 000000000..a264acf6e
--- /dev/null
+++ b/autoapi/lmflow/datasets/dataset/index.html
@@ -0,0 +1,1053 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.datasets.dataset &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/datasets/dataset/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.datasets.multi_modal_dataset" href="../multi_modal_dataset/index.html" />
+    <link rel="prev" title="lmflow.datasets" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.datasets</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.datasets</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.datasets.dataset</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.datasets.dataset">
+<span id="lmflow-datasets-dataset"></span><h1>lmflow.datasets.dataset<a class="headerlink" href="#module-lmflow.datasets.dataset" title="Link to this heading">#</a></h1>
+<p>This Python code defines a class Dataset with methods for initializing, loading,
+and manipulating datasets from different backends such as Hugging Face and JSON.</p>
+<p>The <cite>Dataset</cite> class includes methods for loading datasets from a dictionary and a Hugging
+Face dataset, mapping datasets, and retrieving the backend dataset and arguments.</p>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.datasets.dataset.logger" title="lmflow.datasets.dataset.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.datasets.dataset.DATASET_TYPES" title="lmflow.datasets.dataset.DATASET_TYPES"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DATASET_TYPES</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.datasets.dataset.KEY_TYPE" title="lmflow.datasets.dataset.KEY_TYPE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">KEY_TYPE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.datasets.dataset.KEY_INSTANCES" title="lmflow.datasets.dataset.KEY_INSTANCES"><code class="xref py py-obj docutils literal notranslate"><span class="pre">KEY_INSTANCES</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.datasets.dataset.KEY_SCORE" title="lmflow.datasets.dataset.KEY_SCORE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">KEY_SCORE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Dataset</span></code></a></p></td>
+<td><p>Initializes the Dataset object with the given parameters.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.datasets.dataset.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.DATASET_TYPES">
+<span class="sig-prename descclassname"><span class="pre">lmflow.datasets.dataset.</span></span><span class="sig-name descname"><span class="pre">DATASET_TYPES</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">['text_only',</span> <span class="pre">'text2text',</span> <span class="pre">'float_only',</span> <span class="pre">'image_text',</span> <span class="pre">'conversation',</span> <span class="pre">'paired_conversation',...</span></em><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#DATASET_TYPES"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.DATASET_TYPES" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.KEY_TYPE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.datasets.dataset.</span></span><span class="sig-name descname"><span class="pre">KEY_TYPE</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'type'</span></em><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#KEY_TYPE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.KEY_TYPE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.KEY_INSTANCES">
+<span class="sig-prename descclassname"><span class="pre">lmflow.datasets.dataset.</span></span><span class="sig-name descname"><span class="pre">KEY_INSTANCES</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'instances'</span></em><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#KEY_INSTANCES"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.KEY_INSTANCES" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.KEY_SCORE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.datasets.dataset.</span></span><span class="sig-name descname"><span class="pre">KEY_SCORE</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'score'</span></em><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#KEY_SCORE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.KEY_SCORE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.datasets.dataset.</span></span><span class="sig-name descname"><span class="pre">Dataset</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">backend</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'huggingface'</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset" title="Link to this definition">#</a></dt>
+<dd><p>Initializes the Dataset object with the given parameters.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>data_args</strong><span class="classifier">DatasetArguments object.</span></dt><dd><p>Contains the arguments required to load the dataset.</p>
+</dd>
+<dt><strong>backend</strong><span class="classifier">str,  default=”huggingface”</span></dt><dd><p>A string representing the dataset backend. Defaults to “huggingface”.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.data_args">
+<span class="sig-name descname"><span class="pre">data_args</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.data_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.backend">
+<span class="sig-name descname"><span class="pre">backend</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'huggingface'</span></em><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.backend"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.backend" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.backend_dataset">
+<span class="sig-name descname"><span class="pre">backend_dataset</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.backend_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.backend_dataset" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.type">
+<span class="sig-name descname"><span class="pre">type</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.type" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.dataset_path">
+<span class="sig-name descname"><span class="pre">dataset_path</span></span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.dataset_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.dataset_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.__len__">
+<span class="sig-name descname"><span class="pre">__len__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.__len__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.__len__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset._check_data_format">
+<span class="sig-name descname"><span class="pre">_check_data_format</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset._check_data_format"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset._check_data_format" title="Link to this definition">#</a></dt>
+<dd><p>Checks if data type and data structure matches</p>
+<p>Raise messages with hints if not matched.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.from_dict">
+<span class="sig-name descname"><span class="pre">from_dict</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dict_obj</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">dict</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.from_dict"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.from_dict" title="Link to this definition">#</a></dt>
+<dd><p>Create a Dataset object from a dictionary.</p>
+<dl>
+<dt>Return a Dataset given a dict with format:</dt><dd><dl>
+<dt>{</dt><dd><p>“type”: TYPE,
+“instances”: [</p>
+<blockquote>
+<div><dl class="simple">
+<dt>{</dt><dd><p>“key_1”: VALUE_1.1,
+“key_2”: VALUE_1.2,
+…</p>
+</dd>
+</dl>
+<p>},
+{</p>
+<blockquote>
+<div><p>“key_1”: VALUE_2.1,
+“key_2”: VALUE_2.2,
+…</p>
+</div></blockquote>
+</div></blockquote>
+<p>]</p>
+</dd>
+</dl>
+<p>}</p>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>dict_obj</strong><span class="classifier">dict.</span></dt><dd><p>A dictionary containing the dataset information.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt><strong>self</strong><span class="classifier">Dataset object.</span></dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.create_from_dict">
+<em class="property"><span class="pre">classmethod</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">create_from_dict</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dict_obj</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.create_from_dict"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.create_from_dict" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt>Returns a Dataset object given a dict.</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.to_dict">
+<span class="sig-name descname"><span class="pre">to_dict</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.to_dict"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.to_dict" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl>
+<dt>Return a dict represents the dataset:</dt><dd><dl>
+<dt>{</dt><dd><p>“type”: TYPE,
+“instances”: [</p>
+<blockquote>
+<div><dl class="simple">
+<dt>{</dt><dd><p>“key_1”: VALUE_1.1,
+“key_2”: VALUE_1.2,
+…</p>
+</dd>
+</dl>
+<p>},
+{</p>
+<blockquote>
+<div><p>“key_1”: VALUE_2.1,
+“key_2”: VALUE_2.2,
+…</p>
+</div></blockquote>
+</div></blockquote>
+<p>]</p>
+</dd>
+</dl>
+<p>}</p>
+</dd>
+<dt>A python dict object represents the content of this dataset.</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.to_list">
+<span class="sig-name descname"><span class="pre">to_list</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.to_list"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.to_list" title="Link to this definition">#</a></dt>
+<dd><p>Returns a list of instances.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.map">
+<span class="sig-name descname"><span class="pre">map</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.map"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.map" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt><strong>self</strong><span class="classifier">Dataset object.</span></dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.get_backend">
+<span class="sig-name descname"><span class="pre">get_backend</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.get_backend"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.get_backend" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt>self.backend</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.get_backend_dataset">
+<span class="sig-name descname"><span class="pre">get_backend_dataset</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.get_backend_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.get_backend_dataset" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt>self.backend_dataset</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.get_fingerprint">
+<span class="sig-name descname"><span class="pre">get_fingerprint</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.get_fingerprint"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.get_fingerprint" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt>Fingerprint of the backend_dataset which controls the cache</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.get_data_args">
+<span class="sig-name descname"><span class="pre">get_data_args</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.get_data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.get_data_args" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt>self.data_args</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.get_type">
+<span class="sig-name descname"><span class="pre">get_type</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">str</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.get_type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.get_type" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt>self.type</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.save">
+<span class="sig-name descname"><span class="pre">save</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">file_path</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">format</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'json'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.save"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.save" title="Link to this definition">#</a></dt>
+<dd><p>Save the dataset to a json file.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>file_path</strong><span class="classifier">str.</span></dt><dd><p>The path to the file where the dataset will be saved.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.sample">
+<span class="sig-name descname"><span class="pre">sample</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">n</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seed</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">42</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.sample"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.sample" title="Link to this definition">#</a></dt>
+<dd><p>Sample n instances from the dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>n</strong><span class="classifier">int.</span></dt><dd><p>The number of instances to sample from the dataset.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt><strong>sample_dataset</strong><span class="classifier">Dataset object.</span></dt><dd><p>A new dataset object containing the sampled instances.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.train_test_split">
+<span class="sig-name descname"><span class="pre">train_test_split</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">test_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.2</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">shuffle</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seed</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">42</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.train_test_split"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.train_test_split" title="Link to this definition">#</a></dt>
+<dd><p>Split the dataset into training and testing sets.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>test_size</strong><span class="classifier">float, default=0.2.</span></dt><dd><p>The proportion of the dataset that will be used for testing.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt><strong>train_dataset</strong><span class="classifier">Dataset object.</span></dt><dd><p>A new dataset object containing the training instances.</p>
+</dd>
+<dt><strong>test_dataset</strong><span class="classifier">Dataset object.</span></dt><dd><p>A new dataset object containing the testing instances.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.drop_instances">
+<span class="sig-name descname"><span class="pre">drop_instances</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">indices</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">list</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.drop_instances"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.drop_instances" title="Link to this definition">#</a></dt>
+<dd><p>Drop instances from the dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>indices</strong><span class="classifier">list.</span></dt><dd><p>A list of indices of the instances to drop from the dataset.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.sanity_check">
+<span class="sig-name descname"><span class="pre">sanity_check</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">drop_invalid</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.sanity_check"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.sanity_check" title="Link to this definition">#</a></dt>
+<dd><p>Perform a sanity check on the dataset.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.dataset.Dataset.hf_dataset_sanity_check">
+<span class="sig-name descname"><span class="pre">hf_dataset_sanity_check</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">drop_invalid</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/dataset.html#Dataset.hf_dataset_sanity_check"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.dataset.Dataset.hf_dataset_sanity_check" title="Link to this definition">#</a></dt>
+<dd><p>Perform a sanity check on the HuggingFace dataset.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.datasets</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../multi_modal_dataset/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.datasets.multi_modal_dataset</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.DATASET_TYPES"><code class="docutils literal notranslate"><span class="pre">DATASET_TYPES</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.KEY_TYPE"><code class="docutils literal notranslate"><span class="pre">KEY_TYPE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.KEY_INSTANCES"><code class="docutils literal notranslate"><span class="pre">KEY_INSTANCES</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.KEY_SCORE"><code class="docutils literal notranslate"><span class="pre">KEY_SCORE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset"><code class="docutils literal notranslate"><span class="pre">Dataset</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.data_args"><code class="docutils literal notranslate"><span class="pre">Dataset.data_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.backend"><code class="docutils literal notranslate"><span class="pre">Dataset.backend</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.backend_dataset"><code class="docutils literal notranslate"><span class="pre">Dataset.backend_dataset</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.type"><code class="docutils literal notranslate"><span class="pre">Dataset.type</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.dataset_path"><code class="docutils literal notranslate"><span class="pre">Dataset.dataset_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.__len__"><code class="docutils literal notranslate"><span class="pre">Dataset.__len__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset._check_data_format"><code class="docutils literal notranslate"><span class="pre">Dataset._check_data_format()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.from_dict"><code class="docutils literal notranslate"><span class="pre">Dataset.from_dict()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.create_from_dict"><code class="docutils literal notranslate"><span class="pre">Dataset.create_from_dict()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.to_dict"><code class="docutils literal notranslate"><span class="pre">Dataset.to_dict()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.to_list"><code class="docutils literal notranslate"><span class="pre">Dataset.to_list()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.map"><code class="docutils literal notranslate"><span class="pre">Dataset.map()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.get_backend"><code class="docutils literal notranslate"><span class="pre">Dataset.get_backend()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.get_backend_dataset"><code class="docutils literal notranslate"><span class="pre">Dataset.get_backend_dataset()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.get_fingerprint"><code class="docutils literal notranslate"><span class="pre">Dataset.get_fingerprint()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.get_data_args"><code class="docutils literal notranslate"><span class="pre">Dataset.get_data_args()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.get_type"><code class="docutils literal notranslate"><span class="pre">Dataset.get_type()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.save"><code class="docutils literal notranslate"><span class="pre">Dataset.save()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.sample"><code class="docutils literal notranslate"><span class="pre">Dataset.sample()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.train_test_split"><code class="docutils literal notranslate"><span class="pre">Dataset.train_test_split()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.drop_instances"><code class="docutils literal notranslate"><span class="pre">Dataset.drop_instances()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.sanity_check"><code class="docutils literal notranslate"><span class="pre">Dataset.sanity_check()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.dataset.Dataset.hf_dataset_sanity_check"><code class="docutils literal notranslate"><span class="pre">Dataset.hf_dataset_sanity_check()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/datasets/dataset/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/datasets/index.html b/autoapi/lmflow/datasets/index.html
new file mode 100644
index 000000000..258ad08c8
--- /dev/null
+++ b/autoapi/lmflow/datasets/index.html
@@ -0,0 +1,1024 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.datasets &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/datasets/index';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+    <link rel="next" title="lmflow.datasets.dataset" href="dataset/index.html" />
+    <link rel="prev" title="lmflow.args" href="../args/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 current active has-children"><a class="current reference internal" href="#">lmflow.datasets</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.datasets</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.datasets">
+<span id="lmflow-datasets"></span><h1>lmflow.datasets<a class="headerlink" href="#module-lmflow.datasets" title="Link to this heading">#</a></h1>
+<p>This Python code defines a class Dataset with methods for initializing, loading,
+and manipulating datasets from different backends such as Hugging Face and JSON.</p>
+<p>The <cite>Dataset</cite> class includes methods for loading datasets from a dictionary and a Hugging
+Face dataset, mapping datasets, and retrieving the backend dataset and arguments.</p>
+<section id="submodules">
+<h2>Submodules<a class="headerlink" href="#submodules" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l1"><a class="reference internal" href="multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.datasets.Dataset" title="lmflow.datasets.Dataset"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Dataset</span></code></a></p></td>
+<td><p>Initializes the Dataset object with the given parameters.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.datasets.is_multimodal_available" title="lmflow.datasets.is_multimodal_available"><code class="xref py py-obj docutils literal notranslate"><span class="pre">is_multimodal_available</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="package-contents">
+<h2>Package Contents<a class="headerlink" href="#package-contents" title="Link to this heading">#</a></h2>
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.datasets.is_multimodal_available">
+<span class="sig-prename descclassname"><span class="pre">lmflow.datasets.</span></span><span class="sig-name descname"><span class="pre">is_multimodal_available</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/utils/versioning.html#is_multimodal_available"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.is_multimodal_available" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.datasets.</span></span><span class="sig-name descname"><span class="pre">Dataset</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">backend</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'huggingface'</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset" title="Link to this definition">#</a></dt>
+<dd><p>Initializes the Dataset object with the given parameters.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>data_args</strong><span class="classifier">DatasetArguments object.</span></dt><dd><p>Contains the arguments required to load the dataset.</p>
+</dd>
+<dt><strong>backend</strong><span class="classifier">str,  default=”huggingface”</span></dt><dd><p>A string representing the dataset backend. Defaults to “huggingface”.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.data_args">
+<span class="sig-name descname"><span class="pre">data_args</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="headerlink" href="#lmflow.datasets.Dataset.data_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.backend">
+<span class="sig-name descname"><span class="pre">backend</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'huggingface'</span></em><a class="headerlink" href="#lmflow.datasets.Dataset.backend" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.backend_dataset">
+<span class="sig-name descname"><span class="pre">backend_dataset</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="headerlink" href="#lmflow.datasets.Dataset.backend_dataset" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.type">
+<span class="sig-name descname"><span class="pre">type</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="headerlink" href="#lmflow.datasets.Dataset.type" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.dataset_path">
+<span class="sig-name descname"><span class="pre">dataset_path</span></span><a class="headerlink" href="#lmflow.datasets.Dataset.dataset_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.__len__">
+<span class="sig-name descname"><span class="pre">__len__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.__len__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.__len__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset._check_data_format">
+<span class="sig-name descname"><span class="pre">_check_data_format</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset._check_data_format"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset._check_data_format" title="Link to this definition">#</a></dt>
+<dd><p>Checks if data type and data structure matches</p>
+<p>Raise messages with hints if not matched.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.from_dict">
+<span class="sig-name descname"><span class="pre">from_dict</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dict_obj</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">dict</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.from_dict"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.from_dict" title="Link to this definition">#</a></dt>
+<dd><p>Create a Dataset object from a dictionary.</p>
+<dl>
+<dt>Return a Dataset given a dict with format:</dt><dd><dl>
+<dt>{</dt><dd><p>“type”: TYPE,
+“instances”: [</p>
+<blockquote>
+<div><dl class="simple">
+<dt>{</dt><dd><p>“key_1”: VALUE_1.1,
+“key_2”: VALUE_1.2,
+…</p>
+</dd>
+</dl>
+<p>},
+{</p>
+<blockquote>
+<div><p>“key_1”: VALUE_2.1,
+“key_2”: VALUE_2.2,
+…</p>
+</div></blockquote>
+</div></blockquote>
+<p>]</p>
+</dd>
+</dl>
+<p>}</p>
+</dd>
+</dl>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>dict_obj</strong><span class="classifier">dict.</span></dt><dd><p>A dictionary containing the dataset information.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt><strong>self</strong><span class="classifier">Dataset object.</span></dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.create_from_dict">
+<em class="property"><span class="pre">classmethod</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">create_from_dict</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dict_obj</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.create_from_dict"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.create_from_dict" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt>Returns a Dataset object given a dict.</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.to_dict">
+<span class="sig-name descname"><span class="pre">to_dict</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.to_dict"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.to_dict" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl>
+<dt>Return a dict represents the dataset:</dt><dd><dl>
+<dt>{</dt><dd><p>“type”: TYPE,
+“instances”: [</p>
+<blockquote>
+<div><dl class="simple">
+<dt>{</dt><dd><p>“key_1”: VALUE_1.1,
+“key_2”: VALUE_1.2,
+…</p>
+</dd>
+</dl>
+<p>},
+{</p>
+<blockquote>
+<div><p>“key_1”: VALUE_2.1,
+“key_2”: VALUE_2.2,
+…</p>
+</div></blockquote>
+</div></blockquote>
+<p>]</p>
+</dd>
+</dl>
+<p>}</p>
+</dd>
+<dt>A python dict object represents the content of this dataset.</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.to_list">
+<span class="sig-name descname"><span class="pre">to_list</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.to_list"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.to_list" title="Link to this definition">#</a></dt>
+<dd><p>Returns a list of instances.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.map">
+<span class="sig-name descname"><span class="pre">map</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.map"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.map" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt><strong>self</strong><span class="classifier">Dataset object.</span></dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.get_backend">
+<span class="sig-name descname"><span class="pre">get_backend</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span></span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.get_backend"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.get_backend" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt>self.backend</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.get_backend_dataset">
+<span class="sig-name descname"><span class="pre">get_backend_dataset</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.get_backend_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.get_backend_dataset" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt>self.backend_dataset</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.get_fingerprint">
+<span class="sig-name descname"><span class="pre">get_fingerprint</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.get_fingerprint"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.get_fingerprint" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt>Fingerprint of the backend_dataset which controls the cache</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.get_data_args">
+<span class="sig-name descname"><span class="pre">get_data_args</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.get_data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.get_data_args" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt>self.data_args</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.get_type">
+<span class="sig-name descname"><span class="pre">get_type</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">str</span></span></span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.get_type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.get_type" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt>self.type</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.save">
+<span class="sig-name descname"><span class="pre">save</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">file_path</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">format</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'json'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.save"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.save" title="Link to this definition">#</a></dt>
+<dd><p>Save the dataset to a json file.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>file_path</strong><span class="classifier">str.</span></dt><dd><p>The path to the file where the dataset will be saved.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.sample">
+<span class="sig-name descname"><span class="pre">sample</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">n</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seed</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">42</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.sample"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.sample" title="Link to this definition">#</a></dt>
+<dd><p>Sample n instances from the dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>n</strong><span class="classifier">int.</span></dt><dd><p>The number of instances to sample from the dataset.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt><strong>sample_dataset</strong><span class="classifier">Dataset object.</span></dt><dd><p>A new dataset object containing the sampled instances.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.train_test_split">
+<span class="sig-name descname"><span class="pre">train_test_split</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">test_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.2</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">shuffle</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seed</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">42</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.train_test_split"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.train_test_split" title="Link to this definition">#</a></dt>
+<dd><p>Split the dataset into training and testing sets.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>test_size</strong><span class="classifier">float, default=0.2.</span></dt><dd><p>The proportion of the dataset that will be used for testing.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt><strong>train_dataset</strong><span class="classifier">Dataset object.</span></dt><dd><p>A new dataset object containing the training instances.</p>
+</dd>
+<dt><strong>test_dataset</strong><span class="classifier">Dataset object.</span></dt><dd><p>A new dataset object containing the testing instances.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.drop_instances">
+<span class="sig-name descname"><span class="pre">drop_instances</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">indices</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">list</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.drop_instances"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.drop_instances" title="Link to this definition">#</a></dt>
+<dd><p>Drop instances from the dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>indices</strong><span class="classifier">list.</span></dt><dd><p>A list of indices of the instances to drop from the dataset.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.sanity_check">
+<span class="sig-name descname"><span class="pre">sanity_check</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">drop_invalid</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.sanity_check"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.sanity_check" title="Link to this definition">#</a></dt>
+<dd><p>Perform a sanity check on the dataset.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.Dataset.hf_dataset_sanity_check">
+<span class="sig-name descname"><span class="pre">hf_dataset_sanity_check</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">drop_invalid</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../_modules/lmflow/datasets/dataset.html#Dataset.hf_dataset_sanity_check"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.Dataset.hf_dataset_sanity_check" title="Link to this definition">#</a></dt>
+<dd><p>Perform a sanity check on the HuggingFace dataset.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../args/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.args</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="dataset/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.datasets.dataset</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#submodules">Submodules</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#package-contents">Package Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.is_multimodal_available"><code class="docutils literal notranslate"><span class="pre">is_multimodal_available()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset"><code class="docutils literal notranslate"><span class="pre">Dataset</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.data_args"><code class="docutils literal notranslate"><span class="pre">Dataset.data_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.backend"><code class="docutils literal notranslate"><span class="pre">Dataset.backend</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.backend_dataset"><code class="docutils literal notranslate"><span class="pre">Dataset.backend_dataset</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.type"><code class="docutils literal notranslate"><span class="pre">Dataset.type</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.dataset_path"><code class="docutils literal notranslate"><span class="pre">Dataset.dataset_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.__len__"><code class="docutils literal notranslate"><span class="pre">Dataset.__len__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset._check_data_format"><code class="docutils literal notranslate"><span class="pre">Dataset._check_data_format()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.from_dict"><code class="docutils literal notranslate"><span class="pre">Dataset.from_dict()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.create_from_dict"><code class="docutils literal notranslate"><span class="pre">Dataset.create_from_dict()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.to_dict"><code class="docutils literal notranslate"><span class="pre">Dataset.to_dict()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.to_list"><code class="docutils literal notranslate"><span class="pre">Dataset.to_list()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.map"><code class="docutils literal notranslate"><span class="pre">Dataset.map()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.get_backend"><code class="docutils literal notranslate"><span class="pre">Dataset.get_backend()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.get_backend_dataset"><code class="docutils literal notranslate"><span class="pre">Dataset.get_backend_dataset()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.get_fingerprint"><code class="docutils literal notranslate"><span class="pre">Dataset.get_fingerprint()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.get_data_args"><code class="docutils literal notranslate"><span class="pre">Dataset.get_data_args()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.get_type"><code class="docutils literal notranslate"><span class="pre">Dataset.get_type()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.save"><code class="docutils literal notranslate"><span class="pre">Dataset.save()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.sample"><code class="docutils literal notranslate"><span class="pre">Dataset.sample()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.train_test_split"><code class="docutils literal notranslate"><span class="pre">Dataset.train_test_split()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.drop_instances"><code class="docutils literal notranslate"><span class="pre">Dataset.drop_instances()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.sanity_check"><code class="docutils literal notranslate"><span class="pre">Dataset.sanity_check()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.Dataset.hf_dataset_sanity_check"><code class="docutils literal notranslate"><span class="pre">Dataset.hf_dataset_sanity_check()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../_sources/autoapi/lmflow/datasets/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/datasets/multi_modal_dataset/index.html b/autoapi/lmflow/datasets/multi_modal_dataset/index.html
new file mode 100644
index 000000000..0b16c6f43
--- /dev/null
+++ b/autoapi/lmflow/datasets/multi_modal_dataset/index.html
@@ -0,0 +1,790 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.datasets.multi_modal_dataset &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/datasets/multi_modal_dataset/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models" href="../../models/index.html" />
+    <link rel="prev" title="lmflow.datasets.dataset" href="../dataset/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.datasets</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.datasets</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.datasets.multi_modal_dataset</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.datasets.multi_modal_dataset">
+<span id="lmflow-datasets-multi-modal-dataset"></span><h1>lmflow.datasets.multi_modal_dataset<a class="headerlink" href="#module-lmflow.datasets.multi_modal_dataset" title="Link to this heading">#</a></h1>
+<p>This Python code defines a class Multi Modal Dataset.</p>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset" title="lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CustomMultiModalDataset</span></code></a></p></td>
+<td><p>Dataset for Multi Modal data</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset" title="lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DataCollatorForSupervisedDataset</span></code></a></p></td>
+<td><p>Collate examples for supervised fine-tuning.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.datasets.multi_modal_dataset.preprocess_multimodal_llava" title="lmflow.datasets.multi_modal_dataset.preprocess_multimodal_llava"><code class="xref py py-obj docutils literal notranslate"><span class="pre">preprocess_multimodal_llava</span></code></a>(sources, data_args)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.datasets.multi_modal_dataset.tokenizer_image_token" title="lmflow.datasets.multi_modal_dataset.tokenizer_image_token"><code class="xref py py-obj docutils literal notranslate"><span class="pre">tokenizer_image_token</span></code></a>(prompt, tokenizer[, ...])</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_plain" title="lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_plain"><code class="xref py py-obj docutils literal notranslate"><span class="pre">preprocess_llama_from_llava_plain</span></code></a>(sources, tokenizer)</p></td>
+<td><p>This function just add the image in the front of text.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_v1" title="lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_v1"><code class="xref py py-obj docutils literal notranslate"><span class="pre">preprocess_llama_from_llava_v1</span></code></a>(sources, tokenizer[, ...])</p></td>
+<td><p>This function add the prompt and then put the image after the prompt.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.datasets.multi_modal_dataset.</span></span><span class="sig-name descname"><span class="pre">CustomMultiModalDataset</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset_path</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#CustomMultiModalDataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.utils.data.Dataset</span></code></p>
+<p>Dataset for Multi Modal data</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.data_dict">
+<span class="sig-name descname"><span class="pre">data_dict</span></span><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#CustomMultiModalDataset.data_dict"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.data_dict" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.data_args">
+<span class="sig-name descname"><span class="pre">data_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#CustomMultiModalDataset.data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.data_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.image_folder">
+<span class="sig-name descname"><span class="pre">image_folder</span></span><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#CustomMultiModalDataset.image_folder"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.image_folder" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.__len__">
+<span class="sig-name descname"><span class="pre">__len__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#CustomMultiModalDataset.__len__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.__len__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.register_tokenizer">
+<span class="sig-name descname"><span class="pre">register_tokenizer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_processor</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#CustomMultiModalDataset.register_tokenizer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.register_tokenizer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.__getitem__">
+<span class="sig-name descname"><span class="pre">__getitem__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">i</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#CustomMultiModalDataset.__getitem__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.__getitem__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.preprocess_multimodal_llava">
+<span class="sig-prename descclassname"><span class="pre">lmflow.datasets.multi_modal_dataset.</span></span><span class="sig-name descname"><span class="pre">preprocess_multimodal_llava</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">sources</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#preprocess_multimodal_llava"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.preprocess_multimodal_llava" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.tokenizer_image_token">
+<span class="sig-prename descclassname"><span class="pre">lmflow.datasets.multi_modal_dataset.</span></span><span class="sig-name descname"><span class="pre">tokenizer_image_token</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">prompt</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_token_index</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">IMAGE_TOKEN_INDEX</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_tensors</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#tokenizer_image_token"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.tokenizer_image_token" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_plain">
+<span class="sig-prename descclassname"><span class="pre">lmflow.datasets.multi_modal_dataset.</span></span><span class="sig-name descname"><span class="pre">preprocess_llama_from_llava_plain</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">sources</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">has_image</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#preprocess_llama_from_llava_plain"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_plain" title="Link to this definition">#</a></dt>
+<dd><p>This function just add the image in the front of text.
+And don’t add any prompt.
+Args:</p>
+<blockquote>
+<div><p>sources: The input data with text and image.
+tokenizer: The tokenizer to process text.
+has_image: Whether the input data has image.</p>
+</div></blockquote>
+<dl class="simple">
+<dt>Returns:</dt><dd><p>The input_ids and labels for the model.</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_v1">
+<span class="sig-prename descclassname"><span class="pre">lmflow.datasets.multi_modal_dataset.</span></span><span class="sig-name descname"><span class="pre">preprocess_llama_from_llava_v1</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">sources</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">has_image</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#preprocess_llama_from_llava_v1"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_v1" title="Link to this definition">#</a></dt>
+<dd><p>This function add the prompt and then put the image after the prompt.
+So it needs additional code to generate the target label.
+Args:</p>
+<blockquote>
+<div><p>sources: The input data with text and image.
+tokenizer: The tokenizer to process text.
+has_image: Whether the input data has image.</p>
+</div></blockquote>
+<dl class="simple">
+<dt>Returns:</dt><dd><p>The input_ids and labels for the model.</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.datasets.multi_modal_dataset.</span></span><span class="sig-name descname"><span class="pre">DataCollatorForSupervisedDataset</span></span><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#DataCollatorForSupervisedDataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">object</span></code></p>
+<p>Collate examples for supervised fine-tuning.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset.tokenizer">
+<span class="sig-name descname"><span class="pre">tokenizer</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">transformers.PreTrainedTokenizer</span></em><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#DataCollatorForSupervisedDataset.tokenizer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset.tokenizer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset.__call__">
+<span class="sig-name descname"><span class="pre">__call__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">instances</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/datasets/multi_modal_dataset.html#DataCollatorForSupervisedDataset.__call__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset.__call__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../dataset/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.datasets.dataset</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../../models/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset"><code class="docutils literal notranslate"><span class="pre">CustomMultiModalDataset</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.data_dict"><code class="docutils literal notranslate"><span class="pre">CustomMultiModalDataset.data_dict</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.data_args"><code class="docutils literal notranslate"><span class="pre">CustomMultiModalDataset.data_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.image_folder"><code class="docutils literal notranslate"><span class="pre">CustomMultiModalDataset.image_folder</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.__len__"><code class="docutils literal notranslate"><span class="pre">CustomMultiModalDataset.__len__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.register_tokenizer"><code class="docutils literal notranslate"><span class="pre">CustomMultiModalDataset.register_tokenizer()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.__getitem__"><code class="docutils literal notranslate"><span class="pre">CustomMultiModalDataset.__getitem__()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.preprocess_multimodal_llava"><code class="docutils literal notranslate"><span class="pre">preprocess_multimodal_llava()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.tokenizer_image_token"><code class="docutils literal notranslate"><span class="pre">tokenizer_image_token()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_plain"><code class="docutils literal notranslate"><span class="pre">preprocess_llama_from_llava_plain()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_v1"><code class="docutils literal notranslate"><span class="pre">preprocess_llama_from_llava_v1()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset"><code class="docutils literal notranslate"><span class="pre">DataCollatorForSupervisedDataset</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset.tokenizer"><code class="docutils literal notranslate"><span class="pre">DataCollatorForSupervisedDataset.tokenizer</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset.__call__"><code class="docutils literal notranslate"><span class="pre">DataCollatorForSupervisedDataset.__call__()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/datasets/multi_modal_dataset/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/index.html b/autoapi/lmflow/index.html
new file mode 100644
index 000000000..feab97f65
--- /dev/null
+++ b/autoapi/lmflow/index.html
@@ -0,0 +1,674 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/index';</script>
+    <link rel="index" title="Index" href="../../genindex.html" />
+    <link rel="search" title="Search" href="../../search.html" />
+    <link rel="next" title="lmflow.args" href="args/index.html" />
+    <link rel="prev" title="API Reference" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="current reference internal" href="#">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l2"><a class="reference internal" href="args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">API Reference</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow">
+<span id="lmflow"></span><h1>lmflow<a class="headerlink" href="#module-lmflow" title="Link to this heading">#</a></h1>
+<section id="submodules">
+<h2>Submodules<a class="headerlink" href="#submodules" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="args/index.html">lmflow.args</a></li>
+<li class="toctree-l1"><a class="reference internal" href="datasets/index.html">lmflow.datasets</a></li>
+<li class="toctree-l1"><a class="reference internal" href="models/index.html">lmflow.models</a></li>
+<li class="toctree-l1"><a class="reference internal" href="optim/index.html">lmflow.optim</a></li>
+<li class="toctree-l1"><a class="reference internal" href="pipeline/index.html">lmflow.pipeline</a></li>
+<li class="toctree-l1"><a class="reference internal" href="tokenization/index.html">lmflow.tokenization</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils/index.html">lmflow.utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="version/index.html">lmflow.version</a></li>
+</ul>
+</div>
+</section>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.internal_version" title="lmflow.internal_version"><code class="xref py py-obj docutils literal notranslate"><span class="pre">internal_version</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.__version__" title="lmflow.__version__"><code class="xref py py-obj docutils literal notranslate"><span class="pre">__version__</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="package-contents">
+<h2>Package Contents<a class="headerlink" href="#package-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.internal_version">
+<span class="sig-prename descclassname"><span class="pre">lmflow.</span></span><span class="sig-name descname"><span class="pre">internal_version</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'0.0.9'</span></em><a class="headerlink" href="#lmflow.internal_version" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.__version__">
+<span class="sig-prename descclassname"><span class="pre">lmflow.</span></span><span class="sig-name descname"><span class="pre">__version__</span></span><a class="reference internal" href="../../_modules/lmflow.html#__version__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.__version__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">API Reference</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="args/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.args</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#submodules">Submodules</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#package-contents">Package Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.internal_version"><code class="docutils literal notranslate"><span class="pre">internal_version</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.__version__"><code class="docutils literal notranslate"><span class="pre">__version__</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../_sources/autoapi/lmflow/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/auto_model/index.html b/autoapi/lmflow/models/auto_model/index.html
new file mode 100644
index 000000000..66bc5e68e
--- /dev/null
+++ b/autoapi/lmflow/models/auto_model/index.html
@@ -0,0 +1,664 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.auto_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/auto_model/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models.base_model" href="../base_model/index.html" />
+    <link rel="prev" title="lmflow.models" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.auto_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.auto_model">
+<span id="lmflow-models-auto-model"></span><h1>lmflow.models.auto_model<a class="headerlink" href="#module-lmflow.models.auto_model" title="Link to this heading">#</a></h1>
+<p>Automatically get correct model type.</p>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.auto_model.AutoModel" title="lmflow.models.auto_model.AutoModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">AutoModel</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.models.auto_model.AutoModel">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.models.auto_model.</span></span><span class="sig-name descname"><span class="pre">AutoModel</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/auto_model.html#AutoModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.auto_model.AutoModel" title="Link to this definition">#</a></dt>
+<dd><dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.auto_model.AutoModel.get_model">
+<em class="property"><span class="pre">classmethod</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_model</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/auto_model.html#AutoModel.get_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.auto_model.AutoModel.get_model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../base_model/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.base_model</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.auto_model.AutoModel"><code class="docutils literal notranslate"><span class="pre">AutoModel</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.auto_model.AutoModel.get_model"><code class="docutils literal notranslate"><span class="pre">AutoModel.get_model()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/models/auto_model/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/base_model/index.html b/autoapi/lmflow/models/base_model/index.html
new file mode 100644
index 000000000..941a94b06
--- /dev/null
+++ b/autoapi/lmflow/models/base_model/index.html
@@ -0,0 +1,659 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.base_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/base_model/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models.decoder_model" href="../decoder_model/index.html" />
+    <link rel="prev" title="lmflow.models.auto_model" href="../auto_model/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.base_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.base_model">
+<span id="lmflow-models-base-model"></span><h1>lmflow.models.base_model<a class="headerlink" href="#module-lmflow.models.base_model" title="Link to this heading">#</a></h1>
+<p>Base model class.</p>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.base_model.BaseModel" title="lmflow.models.base_model.BaseModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">BaseModel</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.models.base_model.BaseModel">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.models.base_model.</span></span><span class="sig-name descname"><span class="pre">BaseModel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/base_model.html#BaseModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.base_model.BaseModel" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">abc.ABC</span></code></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../auto_model/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.auto_model</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../decoder_model/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.decoder_model</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.base_model.BaseModel"><code class="docutils literal notranslate"><span class="pre">BaseModel</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/models/base_model/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/decoder_model/index.html b/autoapi/lmflow/models/decoder_model/index.html
new file mode 100644
index 000000000..5c00b985a
--- /dev/null
+++ b/autoapi/lmflow/models/decoder_model/index.html
@@ -0,0 +1,668 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.decoder_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/decoder_model/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models.encoder_decoder_model" href="../encoder_decoder_model/index.html" />
+    <link rel="prev" title="lmflow.models.base_model" href="../base_model/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.decoder_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.decoder_model">
+<span id="lmflow-models-decoder-model"></span><h1>lmflow.models.decoder_model<a class="headerlink" href="#module-lmflow.models.decoder_model" title="Link to this heading">#</a></h1>
+<p>A one-line summary of the module or program, terminated by a period.</p>
+<p>Leave one blank line.  The rest of this docstring should contain an
+overall description of the module or program.  Optionally, it may also
+contain a brief description of exported classes and functions and/or usage
+examples.</p>
+<p>Typical usage example:</p>
+<blockquote>
+<div><p>foo = ClassFoo()
+bar = foo.FunctionBar()</p>
+</div></blockquote>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.decoder_model.DecoderModel" title="lmflow.models.decoder_model.DecoderModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DecoderModel</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.models.decoder_model.DecoderModel">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.models.decoder_model.</span></span><span class="sig-name descname"><span class="pre">DecoderModel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/decoder_model.html#DecoderModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.decoder_model.DecoderModel" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_model/index.html#lmflow.models.base_model.BaseModel" title="lmflow.models.base_model.BaseModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.base_model.BaseModel</span></code></a></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../base_model/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.base_model</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../encoder_decoder_model/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.encoder_decoder_model</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.decoder_model.DecoderModel"><code class="docutils literal notranslate"><span class="pre">DecoderModel</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/models/decoder_model/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/encoder_decoder_model/index.html b/autoapi/lmflow/models/encoder_decoder_model/index.html
new file mode 100644
index 000000000..89eeb444b
--- /dev/null
+++ b/autoapi/lmflow/models/encoder_decoder_model/index.html
@@ -0,0 +1,668 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.encoder_decoder_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/encoder_decoder_model/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models.hf_decoder_model" href="../hf_decoder_model/index.html" />
+    <link rel="prev" title="lmflow.models.decoder_model" href="../decoder_model/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.encoder_decoder_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.encoder_decoder_model">
+<span id="lmflow-models-encoder-decoder-model"></span><h1>lmflow.models.encoder_decoder_model<a class="headerlink" href="#module-lmflow.models.encoder_decoder_model" title="Link to this heading">#</a></h1>
+<p>A one-line summary of the module or program, terminated by a period.</p>
+<p>Leave one blank line.  The rest of this docstring should contain an
+overall description of the module or program.  Optionally, it may also
+contain a brief description of exported classes and functions and/or usage
+examples.</p>
+<p>Typical usage example:</p>
+<blockquote>
+<div><p>foo = ClassFoo()
+bar = foo.FunctionBar()</p>
+</div></blockquote>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.encoder_decoder_model.EncoderDecoderModel" title="lmflow.models.encoder_decoder_model.EncoderDecoderModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">EncoderDecoderModel</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.models.encoder_decoder_model.EncoderDecoderModel">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.models.encoder_decoder_model.</span></span><span class="sig-name descname"><span class="pre">EncoderDecoderModel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/encoder_decoder_model.html#EncoderDecoderModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.encoder_decoder_model.EncoderDecoderModel" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_model/index.html#lmflow.models.base_model.BaseModel" title="lmflow.models.base_model.BaseModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.base_model.BaseModel</span></code></a></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../decoder_model/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.decoder_model</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../hf_decoder_model/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.hf_decoder_model</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.encoder_decoder_model.EncoderDecoderModel"><code class="docutils literal notranslate"><span class="pre">EncoderDecoderModel</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/models/encoder_decoder_model/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/hf_decoder_model/index.html b/autoapi/lmflow/models/hf_decoder_model/index.html
new file mode 100644
index 000000000..7346d4c71
--- /dev/null
+++ b/autoapi/lmflow/models/hf_decoder_model/index.html
@@ -0,0 +1,945 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.hf_decoder_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/hf_decoder_model/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models.hf_encoder_decoder_model" href="../hf_encoder_decoder_model/index.html" />
+    <link rel="prev" title="lmflow.models.encoder_decoder_model" href="../encoder_decoder_model/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.hf_decoder_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.hf_decoder_model">
+<span id="lmflow-models-hf-decoder-model"></span><h1>lmflow.models.hf_decoder_model<a class="headerlink" href="#module-lmflow.models.hf_decoder_model" title="Link to this heading">#</a></h1>
+<p>This is a class called HFDecoderModel which is a wrapper around transformers model and
+tokenizer classes. It has several methods such as __init__, tokenize, and train that are
+used for training and fine-tuning the model. The __init__ method takes in several arguments
+such as model_args, tune_strategy, and ds_config, which are used to load the pretrained
+model and tokenizer, and initialize the training settings.</p>
+<p>The tokenize method is used to tokenize the input text and return the input IDs and attention
+masks that can be fed to the model for training or inference.</p>
+<p>This class supports different tune_strategy options such as ‘normal’, ‘none’, ‘lora’, and
+‘adapter’, which allow for different fine-tuning settings of the model. However, the ‘lora’
+and ‘adapter’ strategies are not yet implemented.</p>
+<p>Overall, this class provides a convenient interface for loading and fine-tuning transformer
+models and can be used for various NLP tasks such as language modeling, text classification,
+and question answering.</p>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.hf_decoder_model.logger" title="lmflow.models.hf_decoder_model.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.hf_decoder_model.HFDecoderModel" title="lmflow.models.hf_decoder_model.HFDecoderModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">HFDecoderModel</span></code></a></p></td>
+<td><p>Initializes a HFDecoderModel instance.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.models.hf_decoder_model.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.HFDecoderModel">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.models.hf_decoder_model.</span></span><span class="sig-name descname"><span class="pre">HFDecoderModel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tune_strategy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'normal'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ds_config</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">device</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'gpu'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_accelerator</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#HFDecoderModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.HFDecoderModel" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../decoder_model/index.html#lmflow.models.decoder_model.DecoderModel" title="lmflow.models.decoder_model.DecoderModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.decoder_model.DecoderModel</span></code></a>, <a class="reference internal" href="../hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin" title="lmflow.models.hf_model_mixin.HFModelMixin"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.hf_model_mixin.HFModelMixin</span></code></a>, <a class="reference internal" href="../interfaces/tunable/index.html#lmflow.models.interfaces.tunable.Tunable" title="lmflow.models.interfaces.tunable.Tunable"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.interfaces.tunable.Tunable</span></code></a></p>
+<p>Initializes a HFDecoderModel instance.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model_args</strong></dt><dd><p>Model arguments such as model name, path, revision, etc.</p>
+</dd>
+<dt><strong>tune_strategy</strong><span class="classifier">str or none,  default=”normal”.</span></dt><dd><p>A string representing the dataset backend. Defaults to “huggingface”.</p>
+</dd>
+<dt><strong>ds_config</strong></dt><dd><p>Deepspeed configuations.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.HFDecoderModel.tokenize">
+<span class="sig-name descname"><span class="pre">tokenize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">add_special_tokens</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#HFDecoderModel.tokenize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.HFDecoderModel.tokenize" title="Link to this definition">#</a></dt>
+<dd><p>Tokenize the full dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>dataset</strong><span class="classifier">lmflow.datasets.Dataset.</span></dt><dd></dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>tokenized_datasets</dt><dd><p>The tokenized dataset, without any leading or trailing special
+tokens (normally they are Begin-Of-Sentence or End-Of-Sentence
+tokens).</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.HFDecoderModel.encode">
+<span class="sig-name descname"><span class="pre">encode</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#HFDecoderModel.encode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.HFDecoderModel.encode" title="Link to this definition">#</a></dt>
+<dd><p>Perform encoding process of the tokenizer.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>inputs</strong><span class="classifier">str or list.</span></dt><dd><p>The text sequence.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>outputs</dt><dd><p>if string input,return the tokenized inputs.
+“Hello,world!”-&gt; [101, 7592, 1010, 2088, 102]
+if batch input,return {input_ids,attention_mask,token_type_ids}
+[“Hello,world!”,”Hello!”]-&gt; {‘input_ids’: tensor([[  101,  7592,  1010,  2088,   102],…),’attention_mask’: tensor([[1, 1, 1, 1, 1],[0,0,1,1,1]])}</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.HFDecoderModel.decode">
+<span class="sig-name descname"><span class="pre">decode</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#HFDecoderModel.decode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.HFDecoderModel.decode" title="Link to this definition">#</a></dt>
+<dd><p>Perform decoding process of the tokenizer.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>inputs</strong><span class="classifier">list or tensor.</span></dt><dd><p>The token sequence.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>outputs</dt><dd><p>The text decoded from the token inputs.
+if batch input,return the list of text
+[[101, 7592, 1010, 2088, 102],[101, 7592, 1010, 2088, 102]]-&gt; [“Hello,world!”,”Hello,world!”
+if single input,return the text
+[101, 7592, 1010, 2088, 102]-&gt; “Hello,world!”</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.HFDecoderModel.inference">
+<span class="sig-name descname"><span class="pre">inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inputs</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">release_gpu</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_vllm</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#HFDecoderModel.inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.HFDecoderModel.inference" title="Link to this definition">#</a></dt>
+<dd><p>Perform generation process of the model.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>inputs</strong></dt><dd><p>The sequence used as a prompt for the generation or as model inputs to the model.
+When using vllm inference, this should be a string or a list of strings.
+When using normal inference, this should be a tensor.</p>
+</dd>
+<dt><strong>release_gpu</strong><span class="classifier">bool, optional</span></dt><dd><p>Whether to release the GPU resource after inference, by default False.</p>
+</dd>
+<dt><strong>use_vllm</strong><span class="classifier">bool, optional</span></dt><dd><p>Whether to use VLLM for inference, by default False.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>outputs</dt><dd><p>The generated sequence output</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.HFDecoderModel.__inference">
+<span class="sig-name descname"><span class="pre">__inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inputs</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#HFDecoderModel.__inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.HFDecoderModel.__inference" title="Link to this definition">#</a></dt>
+<dd><p>Perform generation process of the model.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>inputs</strong></dt><dd><p>The <strong>tokenized</strong> sequence used as a prompt for the generation or as model inputs to the model.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>outputs</dt><dd><p>The generated sequence output</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.HFDecoderModel.__vllm_inference">
+<span class="sig-name descname"><span class="pre">__vllm_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampling_params</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">vllm.SamplingParams</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="../../utils/data_utils/index.html#lmflow.utils.data_utils.VLLMInferenceResultWithInput" title="lmflow.utils.data_utils.VLLMInferenceResultWithInput"><span class="pre">lmflow.utils.data_utils.VLLMInferenceResultWithInput</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#HFDecoderModel.__vllm_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.HFDecoderModel.__vllm_inference" title="Link to this definition">#</a></dt>
+<dd><p>Perform VLLM inference process of the model.</p>
+<dl class="field-list">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>inputs</strong><span class="classifier">Union[str, List[str]]</span></dt><dd><p>Prompt(s), string or a list of strings.</p>
+</dd>
+<dt><strong>sampling_params</strong><span class="classifier">Optional[SamplingParams], optional</span></dt><dd><p>vllm SamplingParams object, by default None.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl>
+<dt>List[VLLMInferenceResultWithInput]</dt><dd><p>Return a list of VLLMInferenceResultWithInput, where each
+element contains the input prompt and the corresponding output.</p>
+<p>When <cite>sampling_params.detokenize = True</cite>, the output would be a list of strings,
+contains sampling_params.n samples for the corresponding prompt.</p>
+<p>When <cite>sampling_params.detokenize = False</cite>, return a list of list of ints
+(token ids, no decoding after generation).</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.HFDecoderModel.prepare_inputs_for_inference">
+<span class="sig-name descname"><span class="pre">prepare_inputs_for_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">apply_chat_template</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">enable_distributed_inference</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_vllm</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">ray.data.Dataset</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#HFDecoderModel.prepare_inputs_for_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.HFDecoderModel.prepare_inputs_for_inference" title="Link to this definition">#</a></dt>
+<dd><p>Prepare inputs for inference.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>dataset</strong><span class="classifier">lmflow.datasets.Dataset.</span></dt><dd><p>The dataset used for inference.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>outputs</dt><dd><p>The prepared inputs for inference.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.HFDecoderModel.__prepare_inputs_for_vllm_inference">
+<span class="sig-name descname"><span class="pre">__prepare_inputs_for_vllm_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">apply_chat_template</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">enable_distributed_inference</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">ray.data.Dataset</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#HFDecoderModel.__prepare_inputs_for_vllm_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.HFDecoderModel.__prepare_inputs_for_vllm_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.HFDecoderModel.__prepare_inputs_for_inference">
+<em class="property"><span class="pre">abstract</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">__prepare_inputs_for_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#HFDecoderModel.__prepare_inputs_for_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.HFDecoderModel.__prepare_inputs_for_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.HFDecoderModel.merge_lora_weights">
+<span class="sig-name descname"><span class="pre">merge_lora_weights</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#HFDecoderModel.merge_lora_weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.HFDecoderModel.merge_lora_weights" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.HFDecoderModel.get_peft_without_qlora">
+<span class="sig-name descname"><span class="pre">get_peft_without_qlora</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#HFDecoderModel.get_peft_without_qlora"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.HFDecoderModel.get_peft_without_qlora" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_decoder_model.HFDecoderModel.save">
+<span class="sig-name descname"><span class="pre">save</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dir</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">save_full_model</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_decoder_model.html#HFDecoderModel.save"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_decoder_model.HFDecoderModel.save" title="Link to this definition">#</a></dt>
+<dd><p>Perform generation process of the model.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>dir</strong></dt><dd><p>The directory to save model and tokenizer</p>
+</dd>
+<dt><strong>save_full_model</strong><span class="classifier">Optional.</span></dt><dd><p>Whether to save full model.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>outputs</dt><dd><p>The generated sequence output</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../encoder_decoder_model/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.encoder_decoder_model</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../hf_encoder_decoder_model/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.hf_encoder_decoder_model</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.HFDecoderModel"><code class="docutils literal notranslate"><span class="pre">HFDecoderModel</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.HFDecoderModel.tokenize"><code class="docutils literal notranslate"><span class="pre">HFDecoderModel.tokenize()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.HFDecoderModel.encode"><code class="docutils literal notranslate"><span class="pre">HFDecoderModel.encode()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.HFDecoderModel.decode"><code class="docutils literal notranslate"><span class="pre">HFDecoderModel.decode()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.HFDecoderModel.inference"><code class="docutils literal notranslate"><span class="pre">HFDecoderModel.inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.HFDecoderModel.__inference"><code class="docutils literal notranslate"><span class="pre">HFDecoderModel.__inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.HFDecoderModel.__vllm_inference"><code class="docutils literal notranslate"><span class="pre">HFDecoderModel.__vllm_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.HFDecoderModel.prepare_inputs_for_inference"><code class="docutils literal notranslate"><span class="pre">HFDecoderModel.prepare_inputs_for_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.HFDecoderModel.__prepare_inputs_for_vllm_inference"><code class="docutils literal notranslate"><span class="pre">HFDecoderModel.__prepare_inputs_for_vllm_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.HFDecoderModel.__prepare_inputs_for_inference"><code class="docutils literal notranslate"><span class="pre">HFDecoderModel.__prepare_inputs_for_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.HFDecoderModel.merge_lora_weights"><code class="docutils literal notranslate"><span class="pre">HFDecoderModel.merge_lora_weights()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.HFDecoderModel.get_peft_without_qlora"><code class="docutils literal notranslate"><span class="pre">HFDecoderModel.get_peft_without_qlora()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_decoder_model.HFDecoderModel.save"><code class="docutils literal notranslate"><span class="pre">HFDecoderModel.save()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/models/hf_decoder_model/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/hf_encoder_decoder_model/index.html b/autoapi/lmflow/models/hf_encoder_decoder_model/index.html
new file mode 100644
index 000000000..494938c07
--- /dev/null
+++ b/autoapi/lmflow/models/hf_encoder_decoder_model/index.html
@@ -0,0 +1,864 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.hf_encoder_decoder_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/hf_encoder_decoder_model/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models.hf_model_mixin" href="../hf_model_mixin/index.html" />
+    <link rel="prev" title="lmflow.models.hf_decoder_model" href="../hf_decoder_model/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.hf_encoder_decoder_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.hf_encoder_decoder_model">
+<span id="lmflow-models-hf-encoder-decoder-model"></span><h1>lmflow.models.hf_encoder_decoder_model<a class="headerlink" href="#module-lmflow.models.hf_encoder_decoder_model" title="Link to this heading">#</a></h1>
+<p>This is a class called HFDecoderModel which is a wrapper around transformers model and
+tokenizer classes. It has several methods such as __init__, tokenize, and train that are
+used for training and fine-tuning the model. The __init__ method takes in several arguments
+such as model_args, tune_strategy, and ds_config, which are used to load the pretrained
+model and tokenizer, and initialize the training settings.</p>
+<p>The tokenize method is used to tokenize the input text and return the input IDs and attention
+masks that can be fed to the model for training or inference.</p>
+<p>This class supports different tune_strategy options such as ‘normal’, ‘none’, ‘lora’, and
+‘adapter’, which allow for different fine-tuning settings of the model. However, the ‘lora’
+and ‘adapter’ strategies are not yet implemented.</p>
+<p>Overall, this class provides a convenient interface for loading and fine-tuning transformer
+models and can be used for various NLP tasks such as language modeling, text classification,
+and question answering.</p>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.hf_encoder_decoder_model.logger" title="lmflow.models.hf_encoder_decoder_model.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel" title="lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">HFEncoderDecoderModel</span></code></a></p></td>
+<td><p>Initializes a HFEncoderDecoderModel instance.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.models.hf_encoder_decoder_model.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.models.hf_encoder_decoder_model.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_encoder_decoder_model.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_encoder_decoder_model.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.models.hf_encoder_decoder_model.</span></span><span class="sig-name descname"><span class="pre">HFEncoderDecoderModel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tune_strategy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'normal'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ds_config</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">device</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'gpu'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_accelerator</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">custom_model</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">with_deepspeed</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pipeline_args</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_encoder_decoder_model.html#HFEncoderDecoderModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../encoder_decoder_model/index.html#lmflow.models.encoder_decoder_model.EncoderDecoderModel" title="lmflow.models.encoder_decoder_model.EncoderDecoderModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.encoder_decoder_model.EncoderDecoderModel</span></code></a>, <a class="reference internal" href="../interfaces/tunable/index.html#lmflow.models.interfaces.tunable.Tunable" title="lmflow.models.interfaces.tunable.Tunable"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.interfaces.tunable.Tunable</span></code></a></p>
+<p>Initializes a HFEncoderDecoderModel instance.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model_args</strong></dt><dd><p>Model arguments such as model name, path, revision, etc.</p>
+</dd>
+<dt><strong>tune_strategy</strong><span class="classifier">str or none,  default=”normal”.</span></dt><dd><p>A string representing the dataset backend. Defaults to “huggingface”.</p>
+</dd>
+<dt><strong>ds_config</strong></dt><dd><p>Deepspeed configuations.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.device">
+<span class="sig-name descname"><span class="pre">device</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'gpu'</span></em><a class="reference internal" href="../../../../_modules/lmflow/models/hf_encoder_decoder_model.html#HFEncoderDecoderModel.device"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.device" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.tokenize">
+<em class="property"><span class="pre">abstract</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">tokenize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_encoder_decoder_model.html#HFEncoderDecoderModel.tokenize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.tokenize" title="Link to this definition">#</a></dt>
+<dd><p>Tokenize the full dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>dataset</strong></dt><dd><p>Text dataset.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>tokenized_datasets</dt><dd><p>The tokenized dataset.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.encode">
+<span class="sig-name descname"><span class="pre">encode</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_encoder_decoder_model.html#HFEncoderDecoderModel.encode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.encode" title="Link to this definition">#</a></dt>
+<dd><p>Perform encoding process of the tokenizer.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>inputs</strong><span class="classifier">str or list.</span></dt><dd><p>The text sequence.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>outputs</dt><dd><p>The tokenized inputs.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.decode">
+<span class="sig-name descname"><span class="pre">decode</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_encoder_decoder_model.html#HFEncoderDecoderModel.decode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.decode" title="Link to this definition">#</a></dt>
+<dd><p>Perform decoding process of the tokenizer.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>inputs</strong><span class="classifier">list.</span></dt><dd><p>The token sequence.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>outputs</dt><dd><p>The text decoded from the token inputs.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.inference">
+<span class="sig-name descname"><span class="pre">inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inputs</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_encoder_decoder_model.html#HFEncoderDecoderModel.inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.inference" title="Link to this definition">#</a></dt>
+<dd><p>Perform generation process of the model.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>inputs</strong></dt><dd><p>The sequence used as a prompt for the generation or as model inputs to the model.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>outputs</dt><dd><p>The generated sequence output</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.merge_lora_weights">
+<span class="sig-name descname"><span class="pre">merge_lora_weights</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_encoder_decoder_model.html#HFEncoderDecoderModel.merge_lora_weights"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.merge_lora_weights" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.save">
+<span class="sig-name descname"><span class="pre">save</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dir</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">save_full_model</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_encoder_decoder_model.html#HFEncoderDecoderModel.save"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.save" title="Link to this definition">#</a></dt>
+<dd><p>Perform generation process of the model.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>dir</strong></dt><dd><p>The directory to save model and tokenizer</p>
+</dd>
+<dt><strong>save_full_model</strong><span class="classifier">Optional.</span></dt><dd><p>Whether to save full model.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>outputs</dt><dd><p>The generated sequence output</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_max_length">
+<span class="sig-name descname"><span class="pre">get_max_length</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_encoder_decoder_model.html#HFEncoderDecoderModel.get_max_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_max_length" title="Link to this definition">#</a></dt>
+<dd><p>Return max acceptable input length in terms of tokens.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_tokenizer">
+<span class="sig-name descname"><span class="pre">get_tokenizer</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_encoder_decoder_model.html#HFEncoderDecoderModel.get_tokenizer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_tokenizer" title="Link to this definition">#</a></dt>
+<dd><p>Return the tokenizer of the model.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_backend_model">
+<span class="sig-name descname"><span class="pre">get_backend_model</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_encoder_decoder_model.html#HFEncoderDecoderModel.get_backend_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_backend_model" title="Link to this definition">#</a></dt>
+<dd><p>Return the backend model.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../hf_decoder_model/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.hf_decoder_model</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../hf_model_mixin/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.hf_model_mixin</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_encoder_decoder_model.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel"><code class="docutils literal notranslate"><span class="pre">HFEncoderDecoderModel</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.device"><code class="docutils literal notranslate"><span class="pre">HFEncoderDecoderModel.device</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.tokenize"><code class="docutils literal notranslate"><span class="pre">HFEncoderDecoderModel.tokenize()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.encode"><code class="docutils literal notranslate"><span class="pre">HFEncoderDecoderModel.encode()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.decode"><code class="docutils literal notranslate"><span class="pre">HFEncoderDecoderModel.decode()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.inference"><code class="docutils literal notranslate"><span class="pre">HFEncoderDecoderModel.inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.merge_lora_weights"><code class="docutils literal notranslate"><span class="pre">HFEncoderDecoderModel.merge_lora_weights()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.save"><code class="docutils literal notranslate"><span class="pre">HFEncoderDecoderModel.save()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_max_length"><code class="docutils literal notranslate"><span class="pre">HFEncoderDecoderModel.get_max_length()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_tokenizer"><code class="docutils literal notranslate"><span class="pre">HFEncoderDecoderModel.get_tokenizer()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_backend_model"><code class="docutils literal notranslate"><span class="pre">HFEncoderDecoderModel.get_backend_model()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/models/hf_encoder_decoder_model/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/hf_model_mixin/index.html b/autoapi/lmflow/models/hf_model_mixin/index.html
new file mode 100644
index 000000000..bff55e9d7
--- /dev/null
+++ b/autoapi/lmflow/models/hf_model_mixin/index.html
@@ -0,0 +1,901 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.hf_model_mixin &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/hf_model_mixin/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models.hf_text_regression_model" href="../hf_text_regression_model/index.html" />
+    <link rel="prev" title="lmflow.models.hf_encoder_decoder_model" href="../hf_encoder_decoder_model/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.hf_model_mixin</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.hf_model_mixin">
+<span id="lmflow-models-hf-model-mixin"></span><h1>lmflow.models.hf_model_mixin<a class="headerlink" href="#module-lmflow.models.hf_model_mixin" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.hf_model_mixin.logger" title="lmflow.models.hf_model_mixin.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.models.hf_model_mixin.HF_AUTOMODEL_MAPPING" title="lmflow.models.hf_model_mixin.HF_AUTOMODEL_MAPPING"><code class="xref py py-obj docutils literal notranslate"><span class="pre">HF_AUTOMODEL_MAPPING</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.hf_model_mixin.HF_AUTOMODEL_TYPE" title="lmflow.models.hf_model_mixin.HF_AUTOMODEL_TYPE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">HF_AUTOMODEL_TYPE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.models.hf_model_mixin.LORA_TARGET_MODULES_MAPPING" title="lmflow.models.hf_model_mixin.LORA_TARGET_MODULES_MAPPING"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LORA_TARGET_MODULES_MAPPING</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.hf_model_mixin.HFModelMixin" title="lmflow.models.hf_model_mixin.HFModelMixin"><code class="xref py py-obj docutils literal notranslate"><span class="pre">HFModelMixin</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.models.hf_model_mixin.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HF_AUTOMODEL_MAPPING">
+<span class="sig-prename descclassname"><span class="pre">lmflow.models.hf_model_mixin.</span></span><span class="sig-name descname"><span class="pre">HF_AUTOMODEL_MAPPING</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HF_AUTOMODEL_MAPPING"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HF_AUTOMODEL_MAPPING" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HF_AUTOMODEL_TYPE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.models.hf_model_mixin.</span></span><span class="sig-name descname"><span class="pre">HF_AUTOMODEL_TYPE</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HF_AUTOMODEL_TYPE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HF_AUTOMODEL_TYPE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.LORA_TARGET_MODULES_MAPPING">
+<span class="sig-prename descclassname"><span class="pre">lmflow.models.hf_model_mixin.</span></span><span class="sig-name descname"><span class="pre">LORA_TARGET_MODULES_MAPPING</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#LORA_TARGET_MODULES_MAPPING"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.LORA_TARGET_MODULES_MAPPING" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.models.hf_model_mixin.</span></span><span class="sig-name descname"><span class="pre">HFModelMixin</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">do_train</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ds_config</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">device</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'gpu'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_accelerator</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">hf_auto_model_additional_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_model/index.html#lmflow.models.base_model.BaseModel" title="lmflow.models.base_model.BaseModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.base_model.BaseModel</span></code></a></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.device">
+<span class="sig-name descname"><span class="pre">device</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'gpu'</span></em><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.device"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.device" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.model_args">
+<span class="sig-name descname"><span class="pre">model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.hf_auto_model">
+<span class="sig-name descname"><span class="pre">hf_auto_model</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.hf_auto_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.hf_auto_model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.use_accelerator">
+<span class="sig-name descname"><span class="pre">use_accelerator</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.use_accelerator"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.use_accelerator" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.ds_config">
+<span class="sig-name descname"><span class="pre">ds_config</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.ds_config"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.ds_config" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.do_train">
+<span class="sig-name descname"><span class="pre">do_train</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.do_train"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.do_train" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.tokenizer">
+<span class="sig-name descname"><span class="pre">tokenizer</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.tokenizer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.tokenizer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.torch_dtype">
+<span class="sig-name descname"><span class="pre">torch_dtype</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.torch_dtype"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.torch_dtype" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.hf_model_config">
+<span class="sig-name descname"><span class="pre">hf_model_config</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.hf_model_config"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.hf_model_config" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.quant_config">
+<span class="sig-name descname"><span class="pre">quant_config</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.quant_config"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.quant_config" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.peft_config">
+<span class="sig-name descname"><span class="pre">peft_config</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.peft_config"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.peft_config" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin._activated">
+<span class="sig-name descname"><span class="pre">_activated</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin._activated"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin._activated" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.__prepare_tokenizer">
+<span class="sig-name descname"><span class="pre">__prepare_tokenizer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">transformers.PreTrainedTokenizer</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">transformers.PreTrainedTokenizerFast</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.__prepare_tokenizer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_tokenizer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.__prepare_dtype">
+<span class="sig-name descname"><span class="pre">__prepare_dtype</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">torch.dtype</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.__prepare_dtype"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_dtype" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_config">
+<span class="sig-name descname"><span class="pre">__prepare_model_config</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">hf_auto_model_additional_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.__prepare_model_config"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_config" title="Link to this definition">#</a></dt>
+<dd><p>Prepare model configuration for hf auto register,
+Parameters
+———-
+model_args : ModelArguments</p>
+<blockquote>
+<div><p>LMFlow model arguments.</p>
+</div></blockquote>
+<dl class="simple">
+<dt>hf_auto_model_additional_args<span class="classifier">Optional[Dict], optional</span></dt><dd><p>Special configurations such as <cite>num_labels</cite> in <cite>AutoModelForSequenceClassification</cite>
+(commonly used in reward modeling) will not preset in __prepare_model_config,
+so it should be passed in hf_auto_model_additional_args.</p>
+</dd>
+</dl>
+<section id="returns">
+<h3>Returns<a class="headerlink" href="#returns" title="Link to this heading">#</a></h3>
+<dl class="simple">
+<dt>config<span class="classifier">ModelConfig</span></dt><dd><p>hf model config.</p>
+</dd>
+</dl>
+</section>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.__prepare_quant_config">
+<span class="sig-name descname"><span class="pre">__prepare_quant_config</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.__prepare_quant_config"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_quant_config" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.__prepare_peft_config">
+<span class="sig-name descname"><span class="pre">__prepare_peft_config</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.__prepare_peft_config"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_peft_config" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.__model_module_inject">
+<span class="sig-name descname"><span class="pre">__model_module_inject</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.__model_module_inject"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.__model_module_inject" title="Link to this definition">#</a></dt>
+<dd><p>Override some model modules with custom implementations.</p>
+<p>Current implementations:
+- Position interpolation (model_args.do_rope_scaling):</p>
+<blockquote>
+<div><p>replace llama embeddings with condense embeddings.</p>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_training">
+<span class="sig-name descname"><span class="pre">__prepare_model_for_training</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">hf_auto_model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">HF_AUTOMODEL_TYPE</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.__prepare_model_for_training"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_training" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_inference">
+<span class="sig-name descname"><span class="pre">__prepare_model_for_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">hf_auto_model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">HF_AUTOMODEL_TYPE</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_accelerator</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ds_config</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.__prepare_model_for_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_vllm_inference">
+<span class="sig-name descname"><span class="pre">__prepare_model_for_vllm_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">vllm_gpu_memory_utilization</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">vllm_tensor_parallel_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.__prepare_model_for_vllm_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_vllm_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_post_process">
+<span class="sig-name descname"><span class="pre">__prepare_model_post_process</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.__prepare_model_post_process"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_post_process" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.activate_model_for_inference">
+<span class="sig-name descname"><span class="pre">activate_model_for_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">use_vllm</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.activate_model_for_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.activate_model_for_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.deactivate_model_for_inference">
+<span class="sig-name descname"><span class="pre">deactivate_model_for_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">use_vllm</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.deactivate_model_for_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.deactivate_model_for_inference" title="Link to this definition">#</a></dt>
+<dd><p>Deactivate the model and release the resources.</p>
+<p>NOTE: Currently, VLLM doesn’t have an official way to do this, and the
+implementation below cannot release all gpu resources by our observation.
+Thus this method is just a placeholder for future implementation. See:
+[Github issue](<a class="github reference external" href="https://github.com/vllm-project/vllm/issues/1908">vllm-project/vllm#1908</a>)</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.get_max_length">
+<span class="sig-name descname"><span class="pre">get_max_length</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.get_max_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.get_max_length" title="Link to this definition">#</a></dt>
+<dd><p>Return max acceptable input length in terms of tokens.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.get_tokenizer">
+<span class="sig-name descname"><span class="pre">get_tokenizer</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.get_tokenizer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.get_tokenizer" title="Link to this definition">#</a></dt>
+<dd><p>Return the tokenizer of the model.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_model_mixin.HFModelMixin.get_backend_model">
+<span class="sig-name descname"><span class="pre">get_backend_model</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_model_mixin.html#HFModelMixin.get_backend_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_model_mixin.HFModelMixin.get_backend_model" title="Link to this definition">#</a></dt>
+<dd><p>Return the backend model.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../hf_encoder_decoder_model/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.hf_encoder_decoder_model</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../hf_text_regression_model/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.hf_text_regression_model</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HF_AUTOMODEL_MAPPING"><code class="docutils literal notranslate"><span class="pre">HF_AUTOMODEL_MAPPING</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HF_AUTOMODEL_TYPE"><code class="docutils literal notranslate"><span class="pre">HF_AUTOMODEL_TYPE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.LORA_TARGET_MODULES_MAPPING"><code class="docutils literal notranslate"><span class="pre">LORA_TARGET_MODULES_MAPPING</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin"><code class="docutils literal notranslate"><span class="pre">HFModelMixin</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.device"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.device</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.model_args"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.hf_auto_model"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.hf_auto_model</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.use_accelerator"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.use_accelerator</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.ds_config"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.ds_config</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.do_train"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.do_train</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.tokenizer"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.tokenizer</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.torch_dtype"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.torch_dtype</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.hf_model_config"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.hf_model_config</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.quant_config"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.quant_config</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.peft_config"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.peft_config</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin._activated"><code class="docutils literal notranslate"><span class="pre">HFModelMixin._activated</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_tokenizer"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.__prepare_tokenizer()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_dtype"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.__prepare_dtype()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_config"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.__prepare_model_config()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_quant_config"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.__prepare_quant_config()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_peft_config"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.__prepare_peft_config()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.__model_module_inject"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.__model_module_inject()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_training"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.__prepare_model_for_training()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_inference"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.__prepare_model_for_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_vllm_inference"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.__prepare_model_for_vllm_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_post_process"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.__prepare_model_post_process()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.activate_model_for_inference"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.activate_model_for_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.deactivate_model_for_inference"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.deactivate_model_for_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.get_max_length"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.get_max_length()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.get_tokenizer"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.get_tokenizer()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_model_mixin.HFModelMixin.get_backend_model"><code class="docutils literal notranslate"><span class="pre">HFModelMixin.get_backend_model()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/models/hf_model_mixin/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/hf_text_regression_model/index.html b/autoapi/lmflow/models/hf_text_regression_model/index.html
new file mode 100644
index 000000000..3c841edd2
--- /dev/null
+++ b/autoapi/lmflow/models/hf_text_regression_model/index.html
@@ -0,0 +1,825 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.hf_text_regression_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/hf_text_regression_model/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models.interfaces" href="../interfaces/index.html" />
+    <link rel="prev" title="lmflow.models.hf_model_mixin" href="../hf_model_mixin/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.hf_text_regression_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.hf_text_regression_model">
+<span id="lmflow-models-hf-text-regression-model"></span><h1>lmflow.models.hf_text_regression_model<a class="headerlink" href="#module-lmflow.models.hf_text_regression_model" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.hf_text_regression_model.logger" title="lmflow.models.hf_text_regression_model.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel" title="lmflow.models.hf_text_regression_model.HFTextRegressionModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">HFTextRegressionModel</span></code></a></p></td>
+<td><p>Initializes a HFTextRegressionModel instance.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.models.hf_text_regression_model.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.models.hf_text_regression_model.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_text_regression_model.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_text_regression_model.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.models.hf_text_regression_model.HFTextRegressionModel">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.models.hf_text_regression_model.</span></span><span class="sig-name descname"><span class="pre">HFTextRegressionModel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">tune_strategy</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'normal'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ds_config</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">device</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'gpu'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_accelerator</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_text_regression_model.html#HFTextRegressionModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../text_regression_model/index.html#lmflow.models.text_regression_model.TextRegressionModel" title="lmflow.models.text_regression_model.TextRegressionModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.text_regression_model.TextRegressionModel</span></code></a>, <a class="reference internal" href="../hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin" title="lmflow.models.hf_model_mixin.HFModelMixin"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.hf_model_mixin.HFModelMixin</span></code></a>, <a class="reference internal" href="../interfaces/tunable/index.html#lmflow.models.interfaces.tunable.Tunable" title="lmflow.models.interfaces.tunable.Tunable"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.interfaces.tunable.Tunable</span></code></a></p>
+<p>Initializes a HFTextRegressionModel instance.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model_args</strong></dt><dd><p>Model arguments such as model name, path, revision, etc.</p>
+</dd>
+<dt><strong>tune_strategy</strong><span class="classifier">str or none,  default=”normal”.</span></dt><dd><p>A string representing the dataset backend. Defaults to “huggingface”.</p>
+</dd>
+<dt><strong>ds_config</strong></dt><dd><p>Deepspeed configuations.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_text_regression_model.HFTextRegressionModel.tokenize">
+<span class="sig-name descname"><span class="pre">tokenize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">add_special_tokens</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_text_regression_model.html#HFTextRegressionModel.tokenize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.tokenize" title="Link to this definition">#</a></dt>
+<dd><p>Tokenize the full dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>dataset</strong><span class="classifier">lmflow.datasets.Dataset.</span></dt><dd></dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>tokenized_datasets</dt><dd><p>The tokenized dataset, without any leading or trailing special
+tokens (normally they are Begin-Of-Sentence or End-Of-Sentence
+tokens).</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_text_regression_model.HFTextRegressionModel.inference">
+<span class="sig-name descname"><span class="pre">inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inputs</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">release_gpu</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_vllm</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">transformers.modeling_outputs.SequenceClassifierOutputWithPast</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_text_regression_model.html#HFTextRegressionModel.inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.inference" title="Link to this definition">#</a></dt>
+<dd><p>Perform generation process of the model.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>inputs</strong></dt><dd><p>The sequence used as a prompt for the generation or as model inputs to the model.
+When using vllm inference, this should be a string or a list of strings.
+When using normal inference, this should be a tensor.</p>
+</dd>
+<dt><strong>release_gpu</strong><span class="classifier">bool, optional</span></dt><dd><p>Whether to release the GPU resource after inference, by default False.</p>
+</dd>
+<dt><strong>use_vllm</strong><span class="classifier">bool, optional</span></dt><dd><p>Whether to use VLLM for inference, by default False.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>outputs</dt><dd><p>The generated sequence output</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_text_regression_model.HFTextRegressionModel.__inference">
+<span class="sig-name descname"><span class="pre">__inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inputs</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_text_regression_model.html#HFTextRegressionModel.__inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.__inference" title="Link to this definition">#</a></dt>
+<dd><p>Perform generation process of the model.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>inputs</strong></dt><dd><p>The <strong>tokenized</strong> sequence used as a prompt for the generation or as model inputs to the model.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>outputs</dt><dd><p>The generated sequence output</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_text_regression_model.HFTextRegressionModel.__vllm_inference">
+<em class="property"><span class="pre">abstract</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">__vllm_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampling_params</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">vllm.SamplingParams</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_text_regression_model.html#HFTextRegressionModel.__vllm_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.__vllm_inference" title="Link to this definition">#</a></dt>
+<dd><p>Perform VLLM inference process of the model.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>inputs</strong><span class="classifier">Union[str, List[str]]</span></dt><dd><p>Prompt(s), string or a list of strings.</p>
+</dd>
+<dt><strong>sampling_params</strong><span class="classifier">Optional[SamplingParams], optional</span></dt><dd><p>vllm SamplingParams object, by default None.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"></dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_text_regression_model.HFTextRegressionModel.prepare_inputs_for_inference">
+<span class="sig-name descname"><span class="pre">prepare_inputs_for_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">enable_distributed_inference</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_vllm</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">ray.data.Dataset</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_text_regression_model.html#HFTextRegressionModel.prepare_inputs_for_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.prepare_inputs_for_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_text_regression_model.HFTextRegressionModel.postprocess_inference_outputs">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">postprocess_inference_outputs</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">scores</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_text_regression_model.html#HFTextRegressionModel.postprocess_inference_outputs"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.postprocess_inference_outputs" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_text_regression_model.HFTextRegressionModel.postprocess_distributed_inference_outputs">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">postprocess_distributed_inference_outputs</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">inference_result</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="../../utils/data_utils/index.html#lmflow.utils.data_utils.RewardModelInferenceResultWithInput" title="lmflow.utils.data_utils.RewardModelInferenceResultWithInput"><span class="pre">lmflow.utils.data_utils.RewardModelInferenceResultWithInput</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_text_regression_model.html#HFTextRegressionModel.postprocess_distributed_inference_outputs"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.postprocess_distributed_inference_outputs" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.hf_text_regression_model.HFTextRegressionModel.save">
+<span class="sig-name descname"><span class="pre">save</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dir</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/hf_text_regression_model.html#HFTextRegressionModel.save"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.save" title="Link to this definition">#</a></dt>
+<dd><p>Perform generation process of the model.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>dir</strong></dt><dd><p>The directory to save model and tokenizer</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../hf_model_mixin/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.hf_model_mixin</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../interfaces/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.interfaces</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_text_regression_model.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel"><code class="docutils literal notranslate"><span class="pre">HFTextRegressionModel</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.tokenize"><code class="docutils literal notranslate"><span class="pre">HFTextRegressionModel.tokenize()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.inference"><code class="docutils literal notranslate"><span class="pre">HFTextRegressionModel.inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.__inference"><code class="docutils literal notranslate"><span class="pre">HFTextRegressionModel.__inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.__vllm_inference"><code class="docutils literal notranslate"><span class="pre">HFTextRegressionModel.__vllm_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.prepare_inputs_for_inference"><code class="docutils literal notranslate"><span class="pre">HFTextRegressionModel.prepare_inputs_for_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.postprocess_inference_outputs"><code class="docutils literal notranslate"><span class="pre">HFTextRegressionModel.postprocess_inference_outputs()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.postprocess_distributed_inference_outputs"><code class="docutils literal notranslate"><span class="pre">HFTextRegressionModel.postprocess_distributed_inference_outputs()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.hf_text_regression_model.HFTextRegressionModel.save"><code class="docutils literal notranslate"><span class="pre">HFTextRegressionModel.save()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/models/hf_text_regression_model/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/index.html b/autoapi/lmflow/models/index.html
new file mode 100644
index 000000000..9f52028bd
--- /dev/null
+++ b/autoapi/lmflow/models/index.html
@@ -0,0 +1,649 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/index';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+    <link rel="next" title="lmflow.models.auto_model" href="auto_model/index.html" />
+    <link rel="prev" title="lmflow.datasets.multi_modal_dataset" href="../datasets/multi_modal_dataset/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="current reference internal" href="#">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models">
+<span id="lmflow-models"></span><h1>lmflow.models<a class="headerlink" href="#module-lmflow.models" title="Link to this heading">#</a></h1>
+<section id="submodules">
+<h2>Submodules<a class="headerlink" href="#submodules" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l1"><a class="reference internal" href="base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l1"><a class="reference internal" href="decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l1"><a class="reference internal" href="encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l1"><a class="reference internal" href="hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l1"><a class="reference internal" href="hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l1"><a class="reference internal" href="hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l1"><a class="reference internal" href="hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l1"><a class="reference internal" href="interfaces/index.html">lmflow.models.interfaces</a></li>
+<li class="toctree-l1"><a class="reference internal" href="regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l1"><a class="reference internal" href="text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l1"><a class="reference internal" href="vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l1"><a class="reference internal" href="vision_encoder/index.html">lmflow.models.vision_encoder</a></li>
+</ul>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../datasets/multi_modal_dataset/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.datasets.multi_modal_dataset</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="auto_model/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.auto_model</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#submodules">Submodules</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../_sources/autoapi/lmflow/models/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/interfaces/index.html b/autoapi/lmflow/models/interfaces/index.html
new file mode 100644
index 000000000..ca4d1f342
--- /dev/null
+++ b/autoapi/lmflow/models/interfaces/index.html
@@ -0,0 +1,640 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.interfaces &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/interfaces/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models.interfaces.tunable" href="tunable/index.html" />
+    <link rel="prev" title="lmflow.models.hf_text_regression_model" href="../hf_text_regression_model/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 current active has-children"><a class="current reference internal" href="#">lmflow.models.interfaces</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.interfaces</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.interfaces">
+<span id="lmflow-models-interfaces"></span><h1>lmflow.models.interfaces<a class="headerlink" href="#module-lmflow.models.interfaces" title="Link to this heading">#</a></h1>
+<section id="submodules">
+<h2>Submodules<a class="headerlink" href="#submodules" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../hf_text_regression_model/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.hf_text_regression_model</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="tunable/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.interfaces.tunable</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#submodules">Submodules</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/models/interfaces/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/interfaces/tunable/index.html b/autoapi/lmflow/models/interfaces/tunable/index.html
new file mode 100644
index 000000000..0d9633ea0
--- /dev/null
+++ b/autoapi/lmflow/models/interfaces/tunable/index.html
@@ -0,0 +1,662 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.interfaces.tunable &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/interfaces/tunable/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.models.regression_model" href="../../regression_model/index.html" />
+    <link rel="prev" title="lmflow.models.interfaces" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.models.interfaces</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.models</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models.interfaces</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.interfaces.tunable</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.interfaces.tunable">
+<span id="lmflow-models-interfaces-tunable"></span><h1>lmflow.models.interfaces.tunable<a class="headerlink" href="#module-lmflow.models.interfaces.tunable" title="Link to this heading">#</a></h1>
+<p>Tunable class</p>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.interfaces.tunable.Tunable" title="lmflow.models.interfaces.tunable.Tunable"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Tunable</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.models.interfaces.tunable.Tunable">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.models.interfaces.tunable.</span></span><span class="sig-name descname"><span class="pre">Tunable</span></span><a class="reference internal" href="../../../../../_modules/lmflow/models/interfaces/tunable.html#Tunable"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.interfaces.tunable.Tunable" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">abc.ABC</span></code></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.interfaces</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../../regression_model/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.regression_model</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.interfaces.tunable.Tunable"><code class="docutils literal notranslate"><span class="pre">Tunable</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/models/interfaces/tunable/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/regression_model/index.html b/autoapi/lmflow/models/regression_model/index.html
new file mode 100644
index 000000000..4075a49f1
--- /dev/null
+++ b/autoapi/lmflow/models/regression_model/index.html
@@ -0,0 +1,659 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.regression_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/regression_model/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models.text_regression_model" href="../text_regression_model/index.html" />
+    <link rel="prev" title="lmflow.models.interfaces.tunable" href="../interfaces/tunable/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.regression_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.regression_model">
+<span id="lmflow-models-regression-model"></span><h1>lmflow.models.regression_model<a class="headerlink" href="#module-lmflow.models.regression_model" title="Link to this heading">#</a></h1>
+<p>General regression model.</p>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.regression_model.RegressionModel" title="lmflow.models.regression_model.RegressionModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RegressionModel</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.models.regression_model.RegressionModel">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.models.regression_model.</span></span><span class="sig-name descname"><span class="pre">RegressionModel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/regression_model.html#RegressionModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.regression_model.RegressionModel" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_model/index.html#lmflow.models.base_model.BaseModel" title="lmflow.models.base_model.BaseModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.base_model.BaseModel</span></code></a></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../interfaces/tunable/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.interfaces.tunable</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../text_regression_model/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.text_regression_model</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.regression_model.RegressionModel"><code class="docutils literal notranslate"><span class="pre">RegressionModel</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/models/regression_model/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/text_regression_model/index.html b/autoapi/lmflow/models/text_regression_model/index.html
new file mode 100644
index 000000000..471a1c04b
--- /dev/null
+++ b/autoapi/lmflow/models/text_regression_model/index.html
@@ -0,0 +1,697 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.text_regression_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/text_regression_model/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models.vision2seq_model" href="../vision2seq_model/index.html" />
+    <link rel="prev" title="lmflow.models.regression_model" href="../regression_model/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.text_regression_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.text_regression_model">
+<span id="lmflow-models-text-regression-model"></span><h1>lmflow.models.text_regression_model<a class="headerlink" href="#module-lmflow.models.text_regression_model" title="Link to this heading">#</a></h1>
+<p>A model maps “text_only” data to float.</p>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.text_regression_model.TextRegressionModel" title="lmflow.models.text_regression_model.TextRegressionModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TextRegressionModel</span></code></a></p></td>
+<td><p>Initializes a TextRegressionModel instance.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.models.text_regression_model.TextRegressionModel">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.models.text_regression_model.</span></span><span class="sig-name descname"><span class="pre">TextRegressionModel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/text_regression_model.html#TextRegressionModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.text_regression_model.TextRegressionModel" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../regression_model/index.html#lmflow.models.regression_model.RegressionModel" title="lmflow.models.regression_model.RegressionModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.regression_model.RegressionModel</span></code></a></p>
+<p>Initializes a TextRegressionModel instance.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model_args</strong></dt><dd><p>Model arguments such as model name, path, revision, etc.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.text_regression_model.TextRegressionModel.inference_func">
+<span class="sig-name descname"><span class="pre">inference_func</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../_modules/lmflow/models/text_regression_model.html#TextRegressionModel.inference_func"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.text_regression_model.TextRegressionModel.inference_func" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.text_regression_model.TextRegressionModel.register_inference_function">
+<span class="sig-name descname"><span class="pre">register_inference_function</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inference_func</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/text_regression_model.html#TextRegressionModel.register_inference_function"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.text_regression_model.TextRegressionModel.register_inference_function" title="Link to this definition">#</a></dt>
+<dd><p>Registers a regression function.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.text_regression_model.TextRegressionModel.inference">
+<span class="sig-name descname"><span class="pre">inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/text_regression_model.html#TextRegressionModel.inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.text_regression_model.TextRegressionModel.inference" title="Link to this definition">#</a></dt>
+<dd><p>Gets regression results of a given dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Inputs<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Dataset object, only accept type “text_only”.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../regression_model/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.regression_model</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../vision2seq_model/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.vision2seq_model</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.text_regression_model.TextRegressionModel"><code class="docutils literal notranslate"><span class="pre">TextRegressionModel</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.text_regression_model.TextRegressionModel.inference_func"><code class="docutils literal notranslate"><span class="pre">TextRegressionModel.inference_func</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.text_regression_model.TextRegressionModel.register_inference_function"><code class="docutils literal notranslate"><span class="pre">TextRegressionModel.register_inference_function()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.text_regression_model.TextRegressionModel.inference"><code class="docutils literal notranslate"><span class="pre">TextRegressionModel.inference()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/models/text_regression_model/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/vision2seq_model/index.html b/autoapi/lmflow/models/vision2seq_model/index.html
new file mode 100644
index 000000000..80a54b4b4
--- /dev/null
+++ b/autoapi/lmflow/models/vision2seq_model/index.html
@@ -0,0 +1,798 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.vision2seq_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/vision2seq_model/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models.vision_encoder" href="../vision_encoder/index.html" />
+    <link rel="prev" title="lmflow.models.text_regression_model" href="../text_regression_model/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.vision2seq_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.vision2seq_model">
+<span id="lmflow-models-vision2seq-model"></span><h1>lmflow.models.vision2seq_model<a class="headerlink" href="#module-lmflow.models.vision2seq_model" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel" title="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.models.vision2seq_model.</span></span><span class="sig-name descname"><span class="pre">CustomAutoVision2SeqModel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">config</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.Blip2Config</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_encoder_name_or_path</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">qformer_name_or_path</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">language_model_name_or_path</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">low_resource</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">transformers.Blip2ForConditionalGeneration</span></code>, <a class="reference internal" href="../base_model/index.html#lmflow.models.base_model.BaseModel" title="lmflow.models.base_model.BaseModel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.models.base_model.BaseModel</span></code></a></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.custom_vision_model">
+<span class="sig-name descname"><span class="pre">custom_vision_model</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.custom_vision_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.custom_vision_model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.with_qformer">
+<span class="sig-name descname"><span class="pre">with_qformer</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.with_qformer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.with_qformer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.language_model">
+<span class="sig-name descname"><span class="pre">language_model</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.language_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.language_model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.hidden_size">
+<span class="sig-name descname"><span class="pre">hidden_size</span></span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.hidden_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.hidden_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.get_backend_model">
+<span class="sig-name descname"><span class="pre">get_backend_model</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.get_backend_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.get_backend_model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.vision_model_from_pretrained">
+<span class="sig-name descname"><span class="pre">vision_model_from_pretrained</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pretrained_path</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.vision_model_from_pretrained"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.vision_model_from_pretrained" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.qformer_from_pretrained">
+<span class="sig-name descname"><span class="pre">qformer_from_pretrained</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pretrained_path</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.qformer_from_pretrained"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.qformer_from_pretrained" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.language_model_from_pretrained">
+<span class="sig-name descname"><span class="pre">language_model_from_pretrained</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pretrained_path</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">low_resource</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_prompt_cache</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.language_model_from_pretrained"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.language_model_from_pretrained" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.vision_feature_select">
+<span class="sig-name descname"><span class="pre">vision_feature_select</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">image_forward_outs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.vision_feature_select"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.vision_feature_select" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.register_prompt_cache">
+<span class="sig-name descname"><span class="pre">register_prompt_cache</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">prompt_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">prompt_keys_values</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.register_prompt_cache"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.register_prompt_cache" title="Link to this definition">#</a></dt>
+<dd><p>Udpate the prompt id and embedding for reuse in the future</p>
+<dl class="simple">
+<dt>Args:</dt><dd><p>prompt_ids (torch.LongTensor): The id of the prompt.
+prompt_keys_values (torch.FloatTensor): The embedding of the prompt.</p>
+</dd>
+<dt>Returns:</dt><dd><p>None</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.save_prompt_cache">
+<span class="sig-name descname"><span class="pre">save_prompt_cache</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">path</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.save_prompt_cache"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.save_prompt_cache" title="Link to this definition">#</a></dt>
+<dd><p>Save prompt embedding and id.</p>
+<dl class="simple">
+<dt>Args:</dt><dd><p>path: The path to save the prompt embedding and id.</p>
+</dd>
+<dt>Returns:</dt><dd><p>None</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.load_prompt_cache">
+<span class="sig-name descname"><span class="pre">load_prompt_cache</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">path</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.load_prompt_cache"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.load_prompt_cache" title="Link to this definition">#</a></dt>
+<dd><p>Load prompt embedding and id.
+Args:</p>
+<blockquote>
+<div><p>path: The path to load the prompt embedding and id.</p>
+</div></blockquote>
+<dl class="simple">
+<dt>Returns:</dt><dd><p>None</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.get_tokenizer">
+<span class="sig-name descname"><span class="pre">get_tokenizer</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.get_tokenizer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.get_tokenizer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.LongTensor</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pixel_values</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.FloatTensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">images</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.FloatTensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past_key_values</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.FloatTensor</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inputs_embeds</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.FloatTensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">labels</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.LongTensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_cache</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_hidden_states</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_dict</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_token_indexes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">[0]</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">one_sample_multiple_images</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Tuple</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">transformers.modeling_outputs.CausalLMOutputWithPast</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.forward" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.processor_image_token_in_minigpt4">
+<span class="sig-name descname"><span class="pre">processor_image_token_in_minigpt4</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">language_model_inputs</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_token_indexes</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pixel_values</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">batch_size</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.processor_image_token_in_minigpt4"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.processor_image_token_in_minigpt4" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.generate">
+<span class="sig-name descname"><span class="pre">generate</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pixel_values</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.FloatTensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">input_ids</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.LongTensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.LongTensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_token_indexes</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">[0]</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">one_sample_multiple_images</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">images</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.LongTensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">generate_kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">torch.LongTensor</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/models/vision2seq_model.html#CustomAutoVision2SeqModel.generate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.generate" title="Link to this definition">#</a></dt>
+<dd><p>Overrides <cite>generate</cite> function to be able to use the model as a conditional generator.</p>
+<dl class="simple">
+<dt>Args:</dt><dd><dl class="simple">
+<dt>pixel_values (<cite>torch.FloatTensor</cite> of shape (batch_size, num_channels, height, width)):</dt><dd><p>Input images to be processed.</p>
+</dd>
+<dt>input_ids (<cite>torch.LongTensor</cite> of shape (batch_size, sequence_length), <em>optional</em>):</dt><dd><p>The sequence used as a prompt for the generation.</p>
+</dd>
+<dt>attention_mask (<cite>torch.LongTensor</cite> of shape (batch_size, sequence_length), <em>optional</em>):</dt><dd><p>Mask to avoid performing attention on padding token indices</p>
+</dd>
+<dt>image_token_indexes (bool, <em>optional</em>):</dt><dd><p>The index for inserting the image tokens.</p>
+</dd>
+<dt>one_sample_multiple_images: (bool, <em>optional</em>):</dt><dd><p>The flag for inference that the input batch size is 1 and contain multiple images.</p>
+</dd>
+</dl>
+</dd>
+<dt>Returns:</dt><dd><p>captions (list): A list of strings of length batch_size * num_captions.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../text_regression_model/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.text_regression_model</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../vision_encoder/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.vision_encoder</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.custom_vision_model"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.custom_vision_model</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.with_qformer"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.with_qformer</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.language_model"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.language_model</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.hidden_size"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.hidden_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.get_backend_model"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.get_backend_model()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.vision_model_from_pretrained"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.vision_model_from_pretrained()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.qformer_from_pretrained"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.qformer_from_pretrained()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.language_model_from_pretrained"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.language_model_from_pretrained()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.vision_feature_select"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.vision_feature_select()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.register_prompt_cache"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.register_prompt_cache()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.save_prompt_cache"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.save_prompt_cache()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.load_prompt_cache"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.load_prompt_cache()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.get_tokenizer"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.get_tokenizer()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.forward"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.forward()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.processor_image_token_in_minigpt4"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.processor_image_token_in_minigpt4()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.generate"><code class="docutils literal notranslate"><span class="pre">CustomAutoVision2SeqModel.generate()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/models/vision2seq_model/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/vision_encoder/clip_encoder/index.html b/autoapi/lmflow/models/vision_encoder/clip_encoder/index.html
new file mode 100644
index 000000000..d3aedd6d6
--- /dev/null
+++ b/autoapi/lmflow/models/vision_encoder/clip_encoder/index.html
@@ -0,0 +1,771 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.vision_encoder.clip_encoder &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/vision_encoder/clip_encoder/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.optim" href="../../../optim/index.html" />
+    <link rel="prev" title="lmflow.models.vision_encoder" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.models.vision_encoder</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.models</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models.vision_encoder</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.vision_encoder.clip_encoder</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.vision_encoder.clip_encoder">
+<span id="lmflow-models-vision-encoder-clip-encoder"></span><h1>lmflow.models.vision_encoder.clip_encoder<a class="headerlink" href="#module-lmflow.models.vision_encoder.clip_encoder" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower" title="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CLIPVisionTower</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.vision_encoder.clip_encoder.build_vision_tower" title="lmflow.models.vision_encoder.clip_encoder.build_vision_tower"><code class="xref py py-obj docutils literal notranslate"><span class="pre">build_vision_tower</span></code></a>(vision_tower_cfg, **kwargs)</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.build_vision_tower">
+<span class="sig-prename descclassname"><span class="pre">lmflow.models.vision_encoder.clip_encoder.</span></span><span class="sig-name descname"><span class="pre">build_vision_tower</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">vision_tower_cfg</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#build_vision_tower"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.build_vision_tower" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.models.vision_encoder.clip_encoder.</span></span><span class="sig-name descname"><span class="pre">CLIPVisionTower</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">vision_tower</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">delay_load</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.nn.Module</span></code></p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.is_loaded">
+<span class="sig-name descname"><span class="pre">is_loaded</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.is_loaded"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.is_loaded" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.vision_tower_name">
+<span class="sig-name descname"><span class="pre">vision_tower_name</span></span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.vision_tower_name"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.vision_tower_name" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.select_layer">
+<span class="sig-name descname"><span class="pre">select_layer</span></span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.select_layer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.select_layer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.select_feature">
+<span class="sig-name descname"><span class="pre">select_feature</span></span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.select_feature"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.select_feature" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.load_model">
+<span class="sig-name descname"><span class="pre">load_model</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.load_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.load_model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.encode_images">
+<span class="sig-name descname"><span class="pre">encode_images</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">images</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">language_projection</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.encode_images"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.encode_images" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.feature_select">
+<span class="sig-name descname"><span class="pre">feature_select</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">image_forward_outs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.feature_select"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.feature_select" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">images</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.forward" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py property">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.dummy_feature">
+<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">dummy_feature</span></span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.dummy_feature"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.dummy_feature" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py property">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.dtype">
+<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">dtype</span></span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.dtype"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.dtype" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py property">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.device">
+<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">device</span></span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.device"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.device" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py property">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.config">
+<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">config</span></span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.config"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.config" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py property">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.hidden_size">
+<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">hidden_size</span></span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.hidden_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.hidden_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py property">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.num_patches">
+<em class="property"><span class="pre">property</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">num_patches</span></span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.num_patches"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.num_patches" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.prepare_inputs_labels_for_multimodal">
+<span class="sig-name descname"><span class="pre">prepare_inputs_labels_for_multimodal</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past_key_values</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">labels</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">images</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">language_projection</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">language_model</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#CLIPVisionTower.prepare_inputs_labels_for_multimodal"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.prepare_inputs_labels_for_multimodal" title="Link to this definition">#</a></dt>
+<dd><p>Copy from the LLAVA code base.
+Should be polished.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.vision_encoder</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../../../optim/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.build_vision_tower"><code class="docutils literal notranslate"><span class="pre">build_vision_tower()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.is_loaded"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.is_loaded</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.vision_tower_name"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.vision_tower_name</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.select_layer"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.select_layer</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.select_feature"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.select_feature</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.load_model"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.load_model()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.encode_images"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.encode_images()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.feature_select"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.feature_select()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.forward"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.forward()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.dummy_feature"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.dummy_feature</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.dtype"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.dtype</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.device"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.device</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.config"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.config</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.hidden_size"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.hidden_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.num_patches"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.num_patches</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.prepare_inputs_labels_for_multimodal"><code class="docutils literal notranslate"><span class="pre">CLIPVisionTower.prepare_inputs_labels_for_multimodal()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/models/vision_encoder/clip_encoder/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/models/vision_encoder/index.html b/autoapi/lmflow/models/vision_encoder/index.html
new file mode 100644
index 000000000..1fe8a0390
--- /dev/null
+++ b/autoapi/lmflow/models/vision_encoder/index.html
@@ -0,0 +1,664 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.models.vision_encoder &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/models/vision_encoder/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.models.vision_encoder.clip_encoder" href="clip_encoder/index.html" />
+    <link rel="prev" title="lmflow.models.vision2seq_model" href="../vision2seq_model/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.models</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 current active has-children"><a class="current reference internal" href="#">lmflow.models.vision_encoder</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.models</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.models.vision_encoder</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.models.vision_encoder">
+<span id="lmflow-models-vision-encoder"></span><h1>lmflow.models.vision_encoder<a class="headerlink" href="#module-lmflow.models.vision_encoder" title="Link to this heading">#</a></h1>
+<section id="submodules">
+<h2>Submodules<a class="headerlink" href="#submodules" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.models.vision_encoder.build_vision_tower" title="lmflow.models.vision_encoder.build_vision_tower"><code class="xref py py-obj docutils literal notranslate"><span class="pre">build_vision_tower</span></code></a>(vision_tower_cfg, **kwargs)</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="package-contents">
+<h2>Package Contents<a class="headerlink" href="#package-contents" title="Link to this heading">#</a></h2>
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.models.vision_encoder.build_vision_tower">
+<span class="sig-prename descclassname"><span class="pre">lmflow.models.vision_encoder.</span></span><span class="sig-name descname"><span class="pre">build_vision_tower</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">vision_tower_cfg</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/models/vision_encoder/clip_encoder.html#build_vision_tower"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.models.vision_encoder.build_vision_tower" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../vision2seq_model/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.vision2seq_model</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="clip_encoder/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.models.vision_encoder.clip_encoder</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#submodules">Submodules</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#package-contents">Package Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.models.vision_encoder.build_vision_tower"><code class="docutils literal notranslate"><span class="pre">build_vision_tower()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/models/vision_encoder/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/adabelief/index.html b/autoapi/lmflow/optim/adabelief/index.html
new file mode 100644
index 000000000..583fa1f28
--- /dev/null
+++ b/autoapi/lmflow/optim/adabelief/index.html
@@ -0,0 +1,710 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.adabelief &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/adabelief/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.adabound" href="../adabound/index.html" />
+    <link rel="prev" title="lmflow.optim" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adabelief</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.adabelief">
+<span id="lmflow-optim-adabelief"></span><h1>lmflow.optim.adabelief<a class="headerlink" href="#module-lmflow.optim.adabelief" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.adabelief.AdaBelief" title="lmflow.optim.adabelief.AdaBelief"><code class="xref py py-obj docutils literal notranslate"><span class="pre">AdaBelief</span></code></a></p></td>
+<td><p>Implements AdaBelief algorithm. Modified from Adam in PyTorch</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.adabelief.AdaBelief">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.adabelief.</span></span><span class="sig-name descname"><span class="pre">AdaBelief</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.001</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.9,</span> <span class="pre">0.999)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1e-16</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">amsgrad</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decouple</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">fixed_decay</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">rectify</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">degenerated_to_sgd</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">print_change_log</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adabelief.html#AdaBelief"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adabelief.AdaBelief" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.optimizer.Optimizer</span></code></p>
+<p>Implements AdaBelief algorithm. Modified from Adam in PyTorch
+reference: AdaBelief Optimizer, adapting stepsizes by the belief in observed gradients, NeurIPS 2020</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.optim.adabelief.AdaBelief.degenerated_to_sgd">
+<span class="sig-name descname"><span class="pre">degenerated_to_sgd</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../../_modules/lmflow/optim/adabelief.html#AdaBelief.degenerated_to_sgd"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adabelief.AdaBelief.degenerated_to_sgd" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.optim.adabelief.AdaBelief.weight_decouple">
+<span class="sig-name descname"><span class="pre">weight_decouple</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../../_modules/lmflow/optim/adabelief.html#AdaBelief.weight_decouple"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adabelief.AdaBelief.weight_decouple" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.optim.adabelief.AdaBelief.rectify">
+<span class="sig-name descname"><span class="pre">rectify</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../../_modules/lmflow/optim/adabelief.html#AdaBelief.rectify"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adabelief.AdaBelief.rectify" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.optim.adabelief.AdaBelief.fixed_decay">
+<span class="sig-name descname"><span class="pre">fixed_decay</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../_modules/lmflow/optim/adabelief.html#AdaBelief.fixed_decay"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adabelief.AdaBelief.fixed_decay" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adabelief.AdaBelief.__setstate__">
+<span class="sig-name descname"><span class="pre">__setstate__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">state</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adabelief.html#AdaBelief.__setstate__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adabelief.AdaBelief.__setstate__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adabelief.AdaBelief.reset">
+<span class="sig-name descname"><span class="pre">reset</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adabelief.html#AdaBelief.reset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adabelief.AdaBelief.reset" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adabelief.AdaBelief.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adabelief.html#AdaBelief.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adabelief.AdaBelief.step" title="Link to this definition">#</a></dt>
+<dd><p>Performs a single optimization step.
+Arguments:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>closure (callable, optional): A closure that reevaluates the model</dt><dd><p>and returns the loss.</p>
+</dd>
+</dl>
+</div></blockquote>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../adabound/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.adabound</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adabelief.AdaBelief"><code class="docutils literal notranslate"><span class="pre">AdaBelief</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adabelief.AdaBelief.degenerated_to_sgd"><code class="docutils literal notranslate"><span class="pre">AdaBelief.degenerated_to_sgd</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adabelief.AdaBelief.weight_decouple"><code class="docutils literal notranslate"><span class="pre">AdaBelief.weight_decouple</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adabelief.AdaBelief.rectify"><code class="docutils literal notranslate"><span class="pre">AdaBelief.rectify</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adabelief.AdaBelief.fixed_decay"><code class="docutils literal notranslate"><span class="pre">AdaBelief.fixed_decay</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adabelief.AdaBelief.__setstate__"><code class="docutils literal notranslate"><span class="pre">AdaBelief.__setstate__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adabelief.AdaBelief.reset"><code class="docutils literal notranslate"><span class="pre">AdaBelief.reset()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adabelief.AdaBelief.step"><code class="docutils literal notranslate"><span class="pre">AdaBelief.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/adabelief/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/adabound/index.html b/autoapi/lmflow/optim/adabound/index.html
new file mode 100644
index 000000000..6be67bf93
--- /dev/null
+++ b/autoapi/lmflow/optim/adabound/index.html
@@ -0,0 +1,689 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.adabound &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/adabound/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.adadelta" href="../adadelta/index.html" />
+    <link rel="prev" title="lmflow.optim.adabelief" href="../adabelief/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adabound</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.adabound">
+<span id="lmflow-optim-adabound"></span><h1>lmflow.optim.adabound<a class="headerlink" href="#module-lmflow.optim.adabound" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.adabound.AdaBound" title="lmflow.optim.adabound.AdaBound"><code class="xref py py-obj docutils literal notranslate"><span class="pre">AdaBound</span></code></a></p></td>
+<td><p>Implements AdaBound algorithm.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.adabound.AdaBound">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.adabound.</span></span><span class="sig-name descname"><span class="pre">AdaBound</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.001</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.9,</span> <span class="pre">0.999)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">final_lr</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">gamma</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.001</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-08</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">amsbound</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adabound.html#AdaBound"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adabound.AdaBound" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.optimizer.Optimizer</span></code></p>
+<p>Implements AdaBound algorithm.</p>
+<p>It has been proposed in <a href="#id1"><span class="problematic" id="id2">`</span></a>Adaptive Gradient Methods with Dynamic Bound of
+Learning Rate
+<a class="reference external" href="https://arxiv.org/abs/1902.09843">https://arxiv.org/abs/1902.09843</a>
+Note:</p>
+<blockquote>
+<div><p>Reference code: <a class="github reference external" href="https://github.com/Luolc/AdaBound">Luolc/AdaBound</a></p>
+</div></blockquote>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.optim.adabound.AdaBound.base_lrs">
+<span class="sig-name descname"><span class="pre">base_lrs</span></span><a class="reference internal" href="../../../../_modules/lmflow/optim/adabound.html#AdaBound.base_lrs"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adabound.AdaBound.base_lrs" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adabound.AdaBound.__setstate__">
+<span class="sig-name descname"><span class="pre">__setstate__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">state</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/optim/adabound.html#AdaBound.__setstate__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adabound.AdaBound.__setstate__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adabound.AdaBound.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adabound.html#AdaBound.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adabound.AdaBound.step" title="Link to this definition">#</a></dt>
+<dd><p>Performs a single optimization step.</p>
+<dl class="simple">
+<dt>Arguments:</dt><dd><p>closure: A closure that reevaluates the model and returns the loss.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../adabelief/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.adabelief</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../adadelta/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.adadelta</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adabound.AdaBound"><code class="docutils literal notranslate"><span class="pre">AdaBound</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adabound.AdaBound.base_lrs"><code class="docutils literal notranslate"><span class="pre">AdaBound.base_lrs</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adabound.AdaBound.__setstate__"><code class="docutils literal notranslate"><span class="pre">AdaBound.__setstate__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adabound.AdaBound.step"><code class="docutils literal notranslate"><span class="pre">AdaBound.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/adabound/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/adadelta/index.html b/autoapi/lmflow/optim/adadelta/index.html
new file mode 100644
index 000000000..2ff1e6b01
--- /dev/null
+++ b/autoapi/lmflow/optim/adadelta/index.html
@@ -0,0 +1,664 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.adadelta &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/adadelta/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.adagrad" href="../adagrad/index.html" />
+    <link rel="prev" title="lmflow.optim.adabound" href="../adabound/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adadelta</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.adadelta">
+<span id="lmflow-optim-adadelta"></span><h1>lmflow.optim.adadelta<a class="headerlink" href="#module-lmflow.optim.adadelta" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.adadelta.Adadelta" title="lmflow.optim.adadelta.Adadelta"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Adadelta</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.adadelta.Adadelta">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.adadelta.</span></span><span class="sig-name descname"><span class="pre">Adadelta</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">rho</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.95</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1e-06</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adadelta.html#Adadelta"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adadelta.Adadelta" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.optimizer.Optimizer</span></code></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adadelta.Adadelta.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adadelta.html#Adadelta.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adadelta.Adadelta.step" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../adabound/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.adabound</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../adagrad/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.adagrad</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adadelta.Adadelta"><code class="docutils literal notranslate"><span class="pre">Adadelta</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adadelta.Adadelta.step"><code class="docutils literal notranslate"><span class="pre">Adadelta.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/adadelta/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/adagrad/index.html b/autoapi/lmflow/optim/adagrad/index.html
new file mode 100644
index 000000000..13c2aa9b3
--- /dev/null
+++ b/autoapi/lmflow/optim/adagrad/index.html
@@ -0,0 +1,664 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.adagrad &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/adagrad/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.adam" href="../adam/index.html" />
+    <link rel="prev" title="lmflow.optim.adadelta" href="../adadelta/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adagrad</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.adagrad">
+<span id="lmflow-optim-adagrad"></span><h1>lmflow.optim.adagrad<a class="headerlink" href="#module-lmflow.optim.adagrad" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.adagrad.AdaGrad" title="lmflow.optim.adagrad.AdaGrad"><code class="xref py py-obj docutils literal notranslate"><span class="pre">AdaGrad</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.adagrad.AdaGrad">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.adagrad.</span></span><span class="sig-name descname"><span class="pre">AdaGrad</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.001</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1e-08</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adagrad.html#AdaGrad"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adagrad.AdaGrad" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.Optimizer</span></code></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adagrad.AdaGrad.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adagrad.html#AdaGrad.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adagrad.AdaGrad.step" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../adadelta/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.adadelta</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../adam/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.adam</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adagrad.AdaGrad"><code class="docutils literal notranslate"><span class="pre">AdaGrad</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adagrad.AdaGrad.step"><code class="docutils literal notranslate"><span class="pre">AdaGrad.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/adagrad/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/adam/index.html b/autoapi/lmflow/optim/adam/index.html
new file mode 100644
index 000000000..3cf345765
--- /dev/null
+++ b/autoapi/lmflow/optim/adam/index.html
@@ -0,0 +1,664 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.adam &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/adam/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.adamax" href="../adamax/index.html" />
+    <link rel="prev" title="lmflow.optim.adagrad" href="../adagrad/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adam</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.adam">
+<span id="lmflow-optim-adam"></span><h1>lmflow.optim.adam<a class="headerlink" href="#module-lmflow.optim.adam" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.adam.Adam" title="lmflow.optim.adam.Adam"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Adam</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.adam.Adam">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.adam.</span></span><span class="sig-name descname"><span class="pre">Adam</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.001</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.9,</span> <span class="pre">0.999)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1e-08</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adam.html#Adam"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adam.Adam" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.optimizer.Optimizer</span></code></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adam.Adam.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adam.html#Adam.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adam.Adam.step" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../adagrad/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.adagrad</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../adamax/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.adamax</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adam.Adam"><code class="docutils literal notranslate"><span class="pre">Adam</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adam.Adam.step"><code class="docutils literal notranslate"><span class="pre">Adam.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/adam/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/adamax/index.html b/autoapi/lmflow/optim/adamax/index.html
new file mode 100644
index 000000000..677239e1a
--- /dev/null
+++ b/autoapi/lmflow/optim/adamax/index.html
@@ -0,0 +1,670 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.adamax &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/adamax/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.adamp" href="../adamp/index.html" />
+    <link rel="prev" title="lmflow.optim.adam" href="../adam/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adamax</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.adamax">
+<span id="lmflow-optim-adamax"></span><h1>lmflow.optim.adamax<a class="headerlink" href="#module-lmflow.optim.adamax" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.adamax.Adamax" title="lmflow.optim.adamax.Adamax"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Adamax</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.adamax.Adamax">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.adamax.</span></span><span class="sig-name descname"><span class="pre">Adamax</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.002</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.9,</span> <span class="pre">0.999)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1e-08</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adamax.html#Adamax"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adamax.Adamax" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.optimizer.Optimizer</span></code></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adamax.Adamax.__setstate__">
+<span class="sig-name descname"><span class="pre">__setstate__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">state</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adamax.html#Adamax.__setstate__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adamax.Adamax.__setstate__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adamax.Adamax.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adamax.html#Adamax.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adamax.Adamax.step" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../adam/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.adam</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../adamp/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.adamp</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adamax.Adamax"><code class="docutils literal notranslate"><span class="pre">Adamax</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adamax.Adamax.__setstate__"><code class="docutils literal notranslate"><span class="pre">Adamax.__setstate__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adamax.Adamax.step"><code class="docutils literal notranslate"><span class="pre">Adamax.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/adamax/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/adamp/index.html b/autoapi/lmflow/optim/adamp/index.html
new file mode 100644
index 000000000..fa17fa0ee
--- /dev/null
+++ b/autoapi/lmflow/optim/adamp/index.html
@@ -0,0 +1,701 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.adamp &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/adamp/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.adamw_schedule_free" href="../adamw_schedule_free/index.html" />
+    <link rel="prev" title="lmflow.optim.adamax" href="../adamax/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adamp</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.adamp">
+<span id="lmflow-optim-adamp"></span><h1>lmflow.optim.adamp<a class="headerlink" href="#module-lmflow.optim.adamp" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.adamp.AdamP" title="lmflow.optim.adamp.AdamP"><code class="xref py py-obj docutils literal notranslate"><span class="pre">AdamP</span></code></a></p></td>
+<td><p>Implements AdamP algorithm.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.adamp.AdamP">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.adamp.</span></span><span class="sig-name descname"><span class="pre">AdamP</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.001</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.9,</span> <span class="pre">0.999)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-08</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">delta</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">wd_ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">nesterov</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adamp.html#AdamP"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adamp.AdamP" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.optimizer.Optimizer</span></code></p>
+<p>Implements AdamP algorithm.</p>
+<p>It has been proposed in <cite>Slowing Down the Weight Norm Increase in
+Momentum-based Optimizers</cite>
+<a class="reference external" href="https://arxiv.org/abs/2006.08217">https://arxiv.org/abs/2006.08217</a></p>
+<dl class="simple">
+<dt>Note:</dt><dd><p>Reference code: <a class="github reference external" href="https://github.com/clovaai/AdamP">clovaai/AdamP</a></p>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adamp.AdamP._channel_view">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">_channel_view</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adamp.html#AdamP._channel_view"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adamp.AdamP._channel_view" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adamp.AdamP._layer_view">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">_layer_view</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adamp.html#AdamP._layer_view"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adamp.AdamP._layer_view" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adamp.AdamP._cosine_similarity">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">_cosine_similarity</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">y</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">view_func</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adamp.html#AdamP._cosine_similarity"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adamp.AdamP._cosine_similarity" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adamp.AdamP._projection">
+<span class="sig-name descname"><span class="pre">_projection</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">grad</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">perturb</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">delta</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">wd_ratio</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adamp.html#AdamP._projection"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adamp.AdamP._projection" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adamp.AdamP.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adamp.html#AdamP.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adamp.AdamP.step" title="Link to this definition">#</a></dt>
+<dd><p>Performs a single optimization step.</p>
+<dl class="simple">
+<dt>Arguments:</dt><dd><p>closure: A closure that reevaluates the model and returns the loss.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../adamax/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.adamax</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../adamw_schedule_free/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.adamw_schedule_free</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adamp.AdamP"><code class="docutils literal notranslate"><span class="pre">AdamP</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adamp.AdamP._channel_view"><code class="docutils literal notranslate"><span class="pre">AdamP._channel_view()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adamp.AdamP._layer_view"><code class="docutils literal notranslate"><span class="pre">AdamP._layer_view()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adamp.AdamP._cosine_similarity"><code class="docutils literal notranslate"><span class="pre">AdamP._cosine_similarity()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adamp.AdamP._projection"><code class="docutils literal notranslate"><span class="pre">AdamP._projection()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adamp.AdamP.step"><code class="docutils literal notranslate"><span class="pre">AdamP.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/adamp/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/adamw_schedule_free/index.html b/autoapi/lmflow/optim/adamw_schedule_free/index.html
new file mode 100644
index 000000000..aa2a61c61
--- /dev/null
+++ b/autoapi/lmflow/optim/adamw_schedule_free/index.html
@@ -0,0 +1,691 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.adamw_schedule_free &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/adamw_schedule_free/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.adan" href="../adan/index.html" />
+    <link rel="prev" title="lmflow.optim.adamp" href="../adamp/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adamw_schedule_free</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.adamw_schedule_free">
+<span id="lmflow-optim-adamw-schedule-free"></span><h1>lmflow.optim.adamw_schedule_free<a class="headerlink" href="#module-lmflow.optim.adamw_schedule_free" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.adamw_schedule_free.AdamWScheduleFree" title="lmflow.optim.adamw_schedule_free.AdamWScheduleFree"><code class="xref py py-obj docutils literal notranslate"><span class="pre">AdamWScheduleFree</span></code></a></p></td>
+<td><p>Schedule-Free AdamW</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.adamw_schedule_free.AdamWScheduleFree">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.adamw_schedule_free.</span></span><span class="sig-name descname"><span class="pre">AdamWScheduleFree</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.0025</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.9,</span> <span class="pre">0.999)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1e-08</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">warmup_steps</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">r</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_lr_power</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">2.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">foreach</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">hasattr(torch,</span> <span class="pre">'_foreach_mul_')</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adamw_schedule_free.html#AdamWScheduleFree"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adamw_schedule_free.AdamWScheduleFree" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.Optimizer</span></code></p>
+<p>Schedule-Free AdamW
+As the name suggests, no scheduler is needed with this optimizer.
+To add warmup, rather than using a learning rate schedule you can just
+set the warmup_steps parameter.</p>
+<p>This optimizer requires that .train() and .eval() be called before the
+beginning of training and evaluation respectively. The optimizer should
+also be placed in eval mode when saving checkpoints.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adamw_schedule_free.AdamWScheduleFree.eval">
+<span class="sig-name descname"><span class="pre">eval</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adamw_schedule_free.html#AdamWScheduleFree.eval"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adamw_schedule_free.AdamWScheduleFree.eval" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adamw_schedule_free.AdamWScheduleFree.train">
+<span class="sig-name descname"><span class="pre">train</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adamw_schedule_free.html#AdamWScheduleFree.train"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adamw_schedule_free.AdamWScheduleFree.train" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adamw_schedule_free.AdamWScheduleFree.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adamw_schedule_free.html#AdamWScheduleFree.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adamw_schedule_free.AdamWScheduleFree.step" title="Link to this definition">#</a></dt>
+<dd><p>Performs a single optimization step.</p>
+<dl class="simple">
+<dt>Arguments:</dt><dd><dl class="simple">
+<dt>closure (callable, optional): A closure that reevaluates the model</dt><dd><p>and returns the loss.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../adamp/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.adamp</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../adan/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.adan</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adamw_schedule_free.AdamWScheduleFree"><code class="docutils literal notranslate"><span class="pre">AdamWScheduleFree</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adamw_schedule_free.AdamWScheduleFree.eval"><code class="docutils literal notranslate"><span class="pre">AdamWScheduleFree.eval()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adamw_schedule_free.AdamWScheduleFree.train"><code class="docutils literal notranslate"><span class="pre">AdamWScheduleFree.train()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adamw_schedule_free.AdamWScheduleFree.step"><code class="docutils literal notranslate"><span class="pre">AdamWScheduleFree.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/adamw_schedule_free/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/adan/index.html b/autoapi/lmflow/optim/adan/index.html
new file mode 100644
index 000000000..76bd74567
--- /dev/null
+++ b/autoapi/lmflow/optim/adan/index.html
@@ -0,0 +1,708 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.adan &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/adan/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.dummy" href="../dummy/index.html" />
+    <link rel="prev" title="lmflow.optim.adamw_schedule_free" href="../adamw_schedule_free/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.adan</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.adan">
+<span id="lmflow-optim-adan"></span><h1>lmflow.optim.adan<a class="headerlink" href="#module-lmflow.optim.adan" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.adan.Adan" title="lmflow.optim.adan.Adan"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Adan</span></code></a></p></td>
+<td><p>Implements a pytorch variant of Adan.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.adan._single_tensor_adan" title="lmflow.optim.adan._single_tensor_adan"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_single_tensor_adan</span></code></a>(params, grads, exp_avgs, ...)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.optim.adan._multi_tensor_adan" title="lmflow.optim.adan._multi_tensor_adan"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_multi_tensor_adan</span></code></a>(params, grads, exp_avgs, ...)</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.adan.Adan">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.adan.</span></span><span class="sig-name descname"><span class="pre">Adan</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.001</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.98,</span> <span class="pre">0.92,</span> <span class="pre">0.99)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1e-08</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_grad_norm</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">no_prox</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">foreach</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adan.html#Adan"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adan.Adan" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.optimizer.Optimizer</span></code></p>
+<p>Implements a pytorch variant of Adan.</p>
+<p>Adan was proposed in
+Adan : Adaptive Nesterov Momentum Algorithm for Faster Optimizing Deep Models.
+<a class="reference external" href="https://arxiv.org/abs/2208.06677">https://arxiv.org/abs/2208.06677</a></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adan.Adan.__setstate__">
+<span class="sig-name descname"><span class="pre">__setstate__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">state</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adan.html#Adan.__setstate__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adan.Adan.__setstate__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adan.Adan.restart_opt">
+<span class="sig-name descname"><span class="pre">restart_opt</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adan.html#Adan.restart_opt"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adan.Adan.restart_opt" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.adan.Adan.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adan.html#Adan.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adan.Adan.step" title="Link to this definition">#</a></dt>
+<dd><p>Performs a single optimization step.</p>
+</dd></dl>
+
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.optim.adan._single_tensor_adan">
+<span class="sig-prename descclassname"><span class="pre">lmflow.optim.adan.</span></span><span class="sig-name descname"><span class="pre">_single_tensor_adan</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">grads</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">exp_avgs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">exp_avg_sqs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">exp_avg_diffs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pre_grads</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">beta1</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">beta2</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">beta3</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias_correction1</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias_correction2</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias_correction3_sqrt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">no_prox</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">clip_global_grad_norm</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adan.html#_single_tensor_adan"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adan._single_tensor_adan" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.optim.adan._multi_tensor_adan">
+<span class="sig-prename descclassname"><span class="pre">lmflow.optim.adan.</span></span><span class="sig-name descname"><span class="pre">_multi_tensor_adan</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">grads</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">exp_avgs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">exp_avg_sqs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">exp_avg_diffs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pre_grads</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">beta1</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">beta2</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">beta3</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias_correction1</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias_correction2</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias_correction3_sqrt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">no_prox</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">clip_global_grad_norm</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/adan.html#_multi_tensor_adan"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.adan._multi_tensor_adan" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../adamw_schedule_free/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.adamw_schedule_free</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../dummy/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.dummy</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adan.Adan"><code class="docutils literal notranslate"><span class="pre">Adan</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adan.Adan.__setstate__"><code class="docutils literal notranslate"><span class="pre">Adan.__setstate__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adan.Adan.restart_opt"><code class="docutils literal notranslate"><span class="pre">Adan.restart_opt()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adan.Adan.step"><code class="docutils literal notranslate"><span class="pre">Adan.step()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adan._single_tensor_adan"><code class="docutils literal notranslate"><span class="pre">_single_tensor_adan()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.adan._multi_tensor_adan"><code class="docutils literal notranslate"><span class="pre">_multi_tensor_adan()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/adan/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/dummy/index.html b/autoapi/lmflow/optim/dummy/index.html
new file mode 100644
index 000000000..116fe293d
--- /dev/null
+++ b/autoapi/lmflow/optim/dummy/index.html
@@ -0,0 +1,680 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.dummy &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/dummy/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.lamb" href="../lamb/index.html" />
+    <link rel="prev" title="lmflow.optim.adan" href="../adan/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.dummy</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.dummy">
+<span id="lmflow-optim-dummy"></span><h1>lmflow.optim.dummy<a class="headerlink" href="#module-lmflow.optim.dummy" title="Link to this heading">#</a></h1>
+<p>Dummy Optimizer.</p>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.dummy.Dummy" title="lmflow.optim.dummy.Dummy"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Dummy</span></code></a></p></td>
+<td><p>An dummy optimizer that does nothing.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.dummy.Dummy">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.dummy.</span></span><span class="sig-name descname"><span class="pre">Dummy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Iterable</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.nn.parameter.Parameter</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">float</span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(0.9,</span> <span class="pre">0.999)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.0</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/dummy.html#Dummy"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.dummy.Dummy" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.Optimizer</span></code></p>
+<p>An dummy optimizer that does nothing.</p>
+<dl class="simple">
+<dt>Parameters:</dt><dd><dl class="simple">
+<dt>params (<code class="xref py py-obj docutils literal notranslate"><span class="pre">Iterable[nn.parameter.Parameter]</span></code>):</dt><dd><p>Iterable of parameters to optimize or dictionaries defining parameter groups.</p>
+</dd>
+<dt>lr (<code class="xref py py-obj docutils literal notranslate"><span class="pre">float</span></code>, <cite>optional</cite>, defaults to 0):</dt><dd><p>The learning rate to use.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.dummy.Dummy.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Callable</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/dummy.html#Dummy.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.dummy.Dummy.step" title="Link to this definition">#</a></dt>
+<dd><p>Performs a single optimization step.</p>
+<dl class="simple">
+<dt>Arguments:</dt><dd><p>closure (<code class="xref py py-obj docutils literal notranslate"><span class="pre">Callable</span></code>, <cite>optional</cite>): A closure that reevaluates the model and returns the loss.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../adan/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.adan</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../lamb/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.lamb</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.dummy.Dummy"><code class="docutils literal notranslate"><span class="pre">Dummy</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.dummy.Dummy.step"><code class="docutils literal notranslate"><span class="pre">Dummy.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/dummy/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/index.html b/autoapi/lmflow/optim/index.html
new file mode 100644
index 000000000..2511a35de
--- /dev/null
+++ b/autoapi/lmflow/optim/index.html
@@ -0,0 +1,656 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/index';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+    <link rel="next" title="lmflow.optim.adabelief" href="adabelief/index.html" />
+    <link rel="prev" title="lmflow.models.vision_encoder.clip_encoder" href="../models/vision_encoder/clip_encoder/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="current reference internal" href="#">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim">
+<span id="lmflow-optim"></span><h1>lmflow.optim<a class="headerlink" href="#module-lmflow.optim" title="Link to this heading">#</a></h1>
+<section id="submodules">
+<h2>Submodules<a class="headerlink" href="#submodules" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l1"><a class="reference internal" href="adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l1"><a class="reference internal" href="adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l1"><a class="reference internal" href="adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l1"><a class="reference internal" href="adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l1"><a class="reference internal" href="adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l1"><a class="reference internal" href="adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l1"><a class="reference internal" href="adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l1"><a class="reference internal" href="adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l1"><a class="reference internal" href="dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l1"><a class="reference internal" href="lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l1"><a class="reference internal" href="lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l1"><a class="reference internal" href="nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l1"><a class="reference internal" href="novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l1"><a class="reference internal" href="optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l1"><a class="reference internal" href="radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l1"><a class="reference internal" href="sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l1"><a class="reference internal" href="sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l1"><a class="reference internal" href="sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l1"><a class="reference internal" href="yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../models/vision_encoder/clip_encoder/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.models.vision_encoder.clip_encoder</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="adabelief/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.adabelief</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#submodules">Submodules</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../_sources/autoapi/lmflow/optim/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/lamb/index.html b/autoapi/lmflow/optim/lamb/index.html
new file mode 100644
index 000000000..c010655ec
--- /dev/null
+++ b/autoapi/lmflow/optim/lamb/index.html
@@ -0,0 +1,695 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.lamb &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/lamb/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.lars" href="../lars/index.html" />
+    <link rel="prev" title="lmflow.optim.dummy" href="../dummy/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.lamb</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.lamb">
+<span id="lmflow-optim-lamb"></span><h1>lmflow.optim.lamb<a class="headerlink" href="#module-lmflow.optim.lamb" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.lamb.Lamb" title="lmflow.optim.lamb.Lamb"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Lamb</span></code></a></p></td>
+<td><p>Implements Lamb algorithm.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.lamb.Lamb">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.lamb.</span></span><span class="sig-name descname"><span class="pre">Lamb</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.001</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.9,</span> <span class="pre">0.999)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-06</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">clamp_value</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">10</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">adam</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">debias</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/lamb.html#Lamb"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.lamb.Lamb" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.optimizer.Optimizer</span></code></p>
+<p>Implements Lamb algorithm.</p>
+<p>It has been proposed in <cite>Large Batch Optimization for Deep Learning:
+Training BERT in 76 minutes</cite>
+<a class="reference external" href="https://arxiv.org/abs/1904.00962">https://arxiv.org/abs/1904.00962</a></p>
+<dl class="simple">
+<dt>Note:</dt><dd><p>Reference code: <a class="github reference external" href="https://github.com/cybertronai/pytorch-lamb">cybertronai/pytorch-lamb</a></p>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.optim.lamb.Lamb.clamp_value">
+<span class="sig-name descname"><span class="pre">clamp_value</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">10</span></em><a class="reference internal" href="../../../../_modules/lmflow/optim/lamb.html#Lamb.clamp_value"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.lamb.Lamb.clamp_value" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.optim.lamb.Lamb.adam">
+<span class="sig-name descname"><span class="pre">adam</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../_modules/lmflow/optim/lamb.html#Lamb.adam"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.lamb.Lamb.adam" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.optim.lamb.Lamb.debias">
+<span class="sig-name descname"><span class="pre">debias</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../_modules/lmflow/optim/lamb.html#Lamb.debias"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.lamb.Lamb.debias" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.lamb.Lamb.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/lamb.html#Lamb.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.lamb.Lamb.step" title="Link to this definition">#</a></dt>
+<dd><p>Performs a single optimization step.</p>
+<dl class="simple">
+<dt>Arguments:</dt><dd><p>closure: A closure that reevaluates the model and returns the loss.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../dummy/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.dummy</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../lars/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.lars</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.lamb.Lamb"><code class="docutils literal notranslate"><span class="pre">Lamb</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.lamb.Lamb.clamp_value"><code class="docutils literal notranslate"><span class="pre">Lamb.clamp_value</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.lamb.Lamb.adam"><code class="docutils literal notranslate"><span class="pre">Lamb.adam</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.lamb.Lamb.debias"><code class="docutils literal notranslate"><span class="pre">Lamb.debias</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.lamb.Lamb.step"><code class="docutils literal notranslate"><span class="pre">Lamb.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/lamb/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/lars/index.html b/autoapi/lmflow/optim/lars/index.html
new file mode 100644
index 000000000..ad468a0dd
--- /dev/null
+++ b/autoapi/lmflow/optim/lars/index.html
@@ -0,0 +1,705 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.lars &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/lars/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.nadam" href="../nadam/index.html" />
+    <link rel="prev" title="lmflow.optim.lamb" href="../lamb/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.lars</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.lars">
+<span id="lmflow-optim-lars"></span><h1>lmflow.optim.lars<a class="headerlink" href="#module-lmflow.optim.lars" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.lars.LARS" title="lmflow.optim.lars.LARS"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LARS</span></code></a></p></td>
+<td><p>Extends SGD in PyTorch with LARS scaling from the paper</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.lars.LARS">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.lars.</span></span><span class="sig-name descname"><span class="pre">LARS</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.01</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">momentum</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dampening</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">nesterov</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trust_coefficient</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.01</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-08</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/lars.html#LARS"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.lars.LARS" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.optimizer.Optimizer</span></code></p>
+<p>Extends SGD in PyTorch with LARS scaling from the paper
+<a class="reference external" href="https://arxiv.org/pdf/1708.03888.pdf">Large batch training of Convolutional Networks</a>.
+.. note:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>The application of momentum in the SGD part is modified according to
+the PyTorch standards. LARS scaling fits into the equation in the
+following fashion.
+
+.. math::
+    \begin{aligned}
+        g_{t+1} &amp; = \text{lars_lr} * (\beta * p_{t} + g_{t+1}), \\
+        v_{t+1} &amp; = \\mu * v_{t} + g_{t+1}, \\
+        p_{t+1} &amp; = p_{t} - \text{lr} * v_{t+1},
+    \\end{aligned}
+
+where :math:`p`, :math:`g`, :math:`v`, :math:`\\mu` and :math:`\beta`
+denote the parameters, gradient, velocity, momentum, and weight decay
+respectively.  The :math:`lars_lr` is defined by Eq. 6 in the paper.
+The Nesterov version is analogously modified.
+</pre></div>
+</div>
+<div class="admonition warning">
+<p class="admonition-title">Warning</p>
+<p>Parameters with weight decay set to 0 will automatically be excluded
+from layer-wise LR scaling. This is to ensure consistency with papers
+like SimCLR and BYOL.</p>
+</div>
+<dl class="simple">
+<dt>Note:</dt><dd><p>Reference code: <a class="github reference external" href="https://github.com/PyTorchLightning/lightning-bolts/">PyTorchLightning/lightning-bolts</a></p>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.lars.LARS.__setstate__">
+<span class="sig-name descname"><span class="pre">__setstate__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">state</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/optim/lars.html#LARS.__setstate__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.lars.LARS.__setstate__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.lars.LARS.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/lars.html#LARS.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.lars.LARS.step" title="Link to this definition">#</a></dt>
+<dd><p>Performs a single optimization step.</p>
+<dl class="simple">
+<dt>Arguments:</dt><dd><p>closure: A closure that reevaluates the model and returns the loss.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../lamb/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.lamb</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../nadam/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.nadam</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.lars.LARS"><code class="docutils literal notranslate"><span class="pre">LARS</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.lars.LARS.__setstate__"><code class="docutils literal notranslate"><span class="pre">LARS.__setstate__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.lars.LARS.step"><code class="docutils literal notranslate"><span class="pre">LARS.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/lars/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/nadam/index.html b/autoapi/lmflow/optim/nadam/index.html
new file mode 100644
index 000000000..22115d3f7
--- /dev/null
+++ b/autoapi/lmflow/optim/nadam/index.html
@@ -0,0 +1,670 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.nadam &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/nadam/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.novograd" href="../novograd/index.html" />
+    <link rel="prev" title="lmflow.optim.lars" href="../lars/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.nadam</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.nadam">
+<span id="lmflow-optim-nadam"></span><h1>lmflow.optim.nadam<a class="headerlink" href="#module-lmflow.optim.nadam" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.nadam.NAdam" title="lmflow.optim.nadam.NAdam"><code class="xref py py-obj docutils literal notranslate"><span class="pre">NAdam</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.nadam.NAdam">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.nadam.</span></span><span class="sig-name descname"><span class="pre">NAdam</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.002</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.9,</span> <span class="pre">0.999)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1e-08</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">momentum_decay</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.004</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/nadam.html#NAdam"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.nadam.NAdam" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.Optimizer</span></code></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.nadam.NAdam.__setstate__">
+<span class="sig-name descname"><span class="pre">__setstate__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">state</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/nadam.html#NAdam.__setstate__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.nadam.NAdam.__setstate__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.nadam.NAdam.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/nadam.html#NAdam.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.nadam.NAdam.step" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../lars/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.lars</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../novograd/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.novograd</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.nadam.NAdam"><code class="docutils literal notranslate"><span class="pre">NAdam</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.nadam.NAdam.__setstate__"><code class="docutils literal notranslate"><span class="pre">NAdam.__setstate__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.nadam.NAdam.step"><code class="docutils literal notranslate"><span class="pre">NAdam.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/nadam/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/novograd/index.html b/autoapi/lmflow/optim/novograd/index.html
new file mode 100644
index 000000000..7498394d0
--- /dev/null
+++ b/autoapi/lmflow/optim/novograd/index.html
@@ -0,0 +1,670 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.novograd &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/novograd/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.optimizers" href="../optimizers/index.html" />
+    <link rel="prev" title="lmflow.optim.nadam" href="../nadam/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.novograd</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.novograd">
+<span id="lmflow-optim-novograd"></span><h1>lmflow.optim.novograd<a class="headerlink" href="#module-lmflow.optim.novograd" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.novograd.NovoGrad" title="lmflow.optim.novograd.NovoGrad"><code class="xref py py-obj docutils literal notranslate"><span class="pre">NovoGrad</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.novograd.NovoGrad">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.novograd.</span></span><span class="sig-name descname"><span class="pre">NovoGrad</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.01</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.9,</span> <span class="pre">0.999)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1e-08</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">grad_averaging</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">amsgrad</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/novograd.html#NovoGrad"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.novograd.NovoGrad" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.Optimizer</span></code></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.novograd.NovoGrad.__setstate__">
+<span class="sig-name descname"><span class="pre">__setstate__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">state</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/novograd.html#NovoGrad.__setstate__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.novograd.NovoGrad.__setstate__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.novograd.NovoGrad.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/novograd.html#NovoGrad.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.novograd.NovoGrad.step" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../nadam/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.nadam</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../optimizers/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.optimizers</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.novograd.NovoGrad"><code class="docutils literal notranslate"><span class="pre">NovoGrad</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.novograd.NovoGrad.__setstate__"><code class="docutils literal notranslate"><span class="pre">NovoGrad.__setstate__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.novograd.NovoGrad.step"><code class="docutils literal notranslate"><span class="pre">NovoGrad.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/novograd/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/optimizers/index.html b/autoapi/lmflow/optim/optimizers/index.html
new file mode 100644
index 000000000..09c091c6e
--- /dev/null
+++ b/autoapi/lmflow/optim/optimizers/index.html
@@ -0,0 +1,621 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.optimizers &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/optimizers/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.radam" href="../radam/index.html" />
+    <link rel="prev" title="lmflow.optim.novograd" href="../novograd/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.optimizers</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.optimizers">
+<span id="lmflow-optim-optimizers"></span><h1>lmflow.optim.optimizers<a class="headerlink" href="#module-lmflow.optim.optimizers" title="Link to this heading">#</a></h1>
+<p>All optimizers.</p>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../novograd/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.novograd</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../radam/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.radam</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/optimizers/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/radam/index.html b/autoapi/lmflow/optim/radam/index.html
new file mode 100644
index 000000000..8fd8d5b50
--- /dev/null
+++ b/autoapi/lmflow/optim/radam/index.html
@@ -0,0 +1,687 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.radam &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/radam/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.sgd_schedule_free" href="../sgd_schedule_free/index.html" />
+    <link rel="prev" title="lmflow.optim.optimizers" href="../optimizers/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.radam</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.radam">
+<span id="lmflow-optim-radam"></span><h1>lmflow.optim.radam<a class="headerlink" href="#module-lmflow.optim.radam" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.radam.RAdam" title="lmflow.optim.radam.RAdam"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RAdam</span></code></a></p></td>
+<td><p>Implements RAdam optimization algorithm.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.radam.RAdam">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.radam.</span></span><span class="sig-name descname"><span class="pre">RAdam</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.001</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.9,</span> <span class="pre">0.999)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-08</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/radam.html#RAdam"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.radam.RAdam" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.optimizer.Optimizer</span></code></p>
+<p>Implements RAdam optimization algorithm.</p>
+<dl class="simple">
+<dt>Note:</dt><dd><p>Deprecated, please use version provided by <a href="#id1"><span class="problematic" id="id2">PyTorch_</span></a>.</p>
+</dd>
+</dl>
+<p>It has been proposed in <cite>On the Variance of the Adaptive Learning
+Rate and Beyond</cite>.
+<a class="reference external" href="https://arxiv.org/abs/1908.03265">https://arxiv.org/abs/1908.03265</a></p>
+<dl class="simple">
+<dt>Note:</dt><dd><p>Reference code: <a class="github reference external" href="https://github.com/LiyuanLucasLiu/RAdam">LiyuanLucasLiu/RAdam</a></p>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.radam.RAdam.__setstate__">
+<span class="sig-name descname"><span class="pre">__setstate__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">state</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/radam.html#RAdam.__setstate__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.radam.RAdam.__setstate__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.radam.RAdam.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/radam.html#RAdam.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.radam.RAdam.step" title="Link to this definition">#</a></dt>
+<dd><p>Performs a single optimization step.</p>
+<dl class="simple">
+<dt>Arguments:</dt><dd><p>closure: A closure that reevaluates the model and returns the loss.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../optimizers/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.optimizers</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../sgd_schedule_free/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.sgd_schedule_free</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.radam.RAdam"><code class="docutils literal notranslate"><span class="pre">RAdam</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.radam.RAdam.__setstate__"><code class="docutils literal notranslate"><span class="pre">RAdam.__setstate__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.radam.RAdam.step"><code class="docutils literal notranslate"><span class="pre">RAdam.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/radam/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/sgd_schedule_free/index.html b/autoapi/lmflow/optim/sgd_schedule_free/index.html
new file mode 100644
index 000000000..5eaead44a
--- /dev/null
+++ b/autoapi/lmflow/optim/sgd_schedule_free/index.html
@@ -0,0 +1,691 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.sgd_schedule_free &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/sgd_schedule_free/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.sgdp" href="../sgdp/index.html" />
+    <link rel="prev" title="lmflow.optim.radam" href="../radam/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.sgd_schedule_free</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.sgd_schedule_free">
+<span id="lmflow-optim-sgd-schedule-free"></span><h1>lmflow.optim.sgd_schedule_free<a class="headerlink" href="#module-lmflow.optim.sgd_schedule_free" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.sgd_schedule_free.SGDScheduleFree" title="lmflow.optim.sgd_schedule_free.SGDScheduleFree"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SGDScheduleFree</span></code></a></p></td>
+<td><p>Schedule-Free SGD</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.sgd_schedule_free.SGDScheduleFree">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.sgd_schedule_free.</span></span><span class="sig-name descname"><span class="pre">SGDScheduleFree</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">momentum</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.9</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">warmup_steps</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">r</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_lr_power</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">2</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">foreach</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">hasattr(torch,</span> <span class="pre">'_foreach_mul_')</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sgd_schedule_free.html#SGDScheduleFree"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sgd_schedule_free.SGDScheduleFree" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.Optimizer</span></code></p>
+<p>Schedule-Free SGD
+As the name suggests, no scheduler is needed with this optimizer.
+To add warmup, rather than using a learning rate schedule you can just
+set the warmup_steps parameter.</p>
+<p>This optimizer requires that .train() and .eval() be called before the
+beginning of training and evaluation respectively. The optimizer should
+also be placed in eval mode when saving checkpoints.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.sgd_schedule_free.SGDScheduleFree.eval">
+<span class="sig-name descname"><span class="pre">eval</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sgd_schedule_free.html#SGDScheduleFree.eval"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sgd_schedule_free.SGDScheduleFree.eval" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.sgd_schedule_free.SGDScheduleFree.train">
+<span class="sig-name descname"><span class="pre">train</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sgd_schedule_free.html#SGDScheduleFree.train"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sgd_schedule_free.SGDScheduleFree.train" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.sgd_schedule_free.SGDScheduleFree.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sgd_schedule_free.html#SGDScheduleFree.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sgd_schedule_free.SGDScheduleFree.step" title="Link to this definition">#</a></dt>
+<dd><p>Performs a single optimization step.</p>
+<dl class="simple">
+<dt>Arguments:</dt><dd><dl class="simple">
+<dt>closure (callable, optional): A closure that reevaluates the model</dt><dd><p>and returns the loss.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../radam/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.radam</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../sgdp/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.sgdp</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sgd_schedule_free.SGDScheduleFree"><code class="docutils literal notranslate"><span class="pre">SGDScheduleFree</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sgd_schedule_free.SGDScheduleFree.eval"><code class="docutils literal notranslate"><span class="pre">SGDScheduleFree.eval()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sgd_schedule_free.SGDScheduleFree.train"><code class="docutils literal notranslate"><span class="pre">SGDScheduleFree.train()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sgd_schedule_free.SGDScheduleFree.step"><code class="docutils literal notranslate"><span class="pre">SGDScheduleFree.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/sgd_schedule_free/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/sgdp/index.html b/autoapi/lmflow/optim/sgdp/index.html
new file mode 100644
index 000000000..84f88b09d
--- /dev/null
+++ b/autoapi/lmflow/optim/sgdp/index.html
@@ -0,0 +1,701 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.sgdp &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/sgdp/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.sophia" href="../sophia/index.html" />
+    <link rel="prev" title="lmflow.optim.sgd_schedule_free" href="../sgd_schedule_free/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.sgdp</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.sgdp">
+<span id="lmflow-optim-sgdp"></span><h1>lmflow.optim.sgdp<a class="headerlink" href="#module-lmflow.optim.sgdp" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.sgdp.SGDP" title="lmflow.optim.sgdp.SGDP"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SGDP</span></code></a></p></td>
+<td><p>Implements SGDP algorithm.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.sgdp.SGDP">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.sgdp.</span></span><span class="sig-name descname"><span class="pre">SGDP</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.001</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">momentum</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dampening</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-08</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">delta</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">wd_ratio</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">nesterov</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sgdp.html#SGDP"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sgdp.SGDP" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.optimizer.Optimizer</span></code></p>
+<p>Implements SGDP algorithm.</p>
+<p>It has been proposed in <cite>Slowing Down the Weight Norm Increase in
+Momentum-based Optimizers</cite>.
+<a class="reference external" href="https://arxiv.org/abs/2006.08217">https://arxiv.org/abs/2006.08217</a></p>
+<dl class="simple">
+<dt>Note:</dt><dd><p>Reference code: <a class="github reference external" href="https://github.com/clovaai/AdamP">clovaai/AdamP</a></p>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.sgdp.SGDP._channel_view">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">_channel_view</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sgdp.html#SGDP._channel_view"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sgdp.SGDP._channel_view" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.sgdp.SGDP._layer_view">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">_layer_view</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sgdp.html#SGDP._layer_view"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sgdp.SGDP._layer_view" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.sgdp.SGDP._cosine_similarity">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">_cosine_similarity</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">y</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">view_func</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sgdp.html#SGDP._cosine_similarity"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sgdp.SGDP._cosine_similarity" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.sgdp.SGDP._projection">
+<span class="sig-name descname"><span class="pre">_projection</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">p</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">grad</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">perturb</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">delta</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">wd_ratio</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sgdp.html#SGDP._projection"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sgdp.SGDP._projection" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.sgdp.SGDP.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sgdp.html#SGDP.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sgdp.SGDP.step" title="Link to this definition">#</a></dt>
+<dd><p>Performs a single optimization step.</p>
+<dl class="simple">
+<dt>Arguments:</dt><dd><p>closure: A closure that reevaluates the model and returns the loss.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../sgd_schedule_free/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.sgd_schedule_free</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../sophia/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.sophia</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sgdp.SGDP"><code class="docutils literal notranslate"><span class="pre">SGDP</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sgdp.SGDP._channel_view"><code class="docutils literal notranslate"><span class="pre">SGDP._channel_view()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sgdp.SGDP._layer_view"><code class="docutils literal notranslate"><span class="pre">SGDP._layer_view()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sgdp.SGDP._cosine_similarity"><code class="docutils literal notranslate"><span class="pre">SGDP._cosine_similarity()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sgdp.SGDP._projection"><code class="docutils literal notranslate"><span class="pre">SGDP._projection()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sgdp.SGDP.step"><code class="docutils literal notranslate"><span class="pre">SGDP.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/sgdp/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/sophia/index.html b/autoapi/lmflow/optim/sophia/index.html
new file mode 100644
index 000000000..eec9af271
--- /dev/null
+++ b/autoapi/lmflow/optim/sophia/index.html
@@ -0,0 +1,678 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.sophia &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/sophia/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.optim.yogi" href="../yogi/index.html" />
+    <link rel="prev" title="lmflow.optim.sgdp" href="../sgdp/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.sophia</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.sophia">
+<span id="lmflow-optim-sophia"></span><h1>lmflow.optim.sophia<a class="headerlink" href="#module-lmflow.optim.sophia" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.sophia.SophiaG" title="lmflow.optim.sophia.SophiaG"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SophiaG</span></code></a></p></td>
+<td><p>Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.sophia.SophiaG">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.sophia.</span></span><span class="sig-name descname"><span class="pre">SophiaG</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.0001</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.965,</span> <span class="pre">0.99)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">rho</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.04</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.1</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">maximize</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">capturable</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sophia.html#SophiaG"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sophia.SophiaG" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.optimizer.Optimizer</span></code></p>
+<p>Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training.
+Code from: <a class="github reference external" href="https://github.com/Liuhong99/Sophia/">Liuhong99/Sophia</a></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.sophia.SophiaG.__setstate__">
+<span class="sig-name descname"><span class="pre">__setstate__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">state</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sophia.html#SophiaG.__setstate__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sophia.SophiaG.__setstate__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.sophia.SophiaG.update_hessian">
+<span class="sig-name descname"><span class="pre">update_hessian</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sophia.html#SophiaG.update_hessian"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sophia.SophiaG.update_hessian" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.sophia.SophiaG.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bs</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">5120</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/sophia.html#SophiaG.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.sophia.SophiaG.step" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../sgdp/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.sgdp</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../yogi/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.optim.yogi</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sophia.SophiaG"><code class="docutils literal notranslate"><span class="pre">SophiaG</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sophia.SophiaG.__setstate__"><code class="docutils literal notranslate"><span class="pre">SophiaG.__setstate__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sophia.SophiaG.update_hessian"><code class="docutils literal notranslate"><span class="pre">SophiaG.update_hessian()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.sophia.SophiaG.step"><code class="docutils literal notranslate"><span class="pre">SophiaG.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/sophia/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/optim/yogi/index.html b/autoapi/lmflow/optim/yogi/index.html
new file mode 100644
index 000000000..67680915b
--- /dev/null
+++ b/autoapi/lmflow/optim/yogi/index.html
@@ -0,0 +1,676 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.optim.yogi &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/optim/yogi/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline" href="../../pipeline/index.html" />
+    <link rel="prev" title="lmflow.optim.sophia" href="../sophia/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.optim</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.optim</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.optim.yogi</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.optim.yogi">
+<span id="lmflow-optim-yogi"></span><h1>lmflow.optim.yogi<a class="headerlink" href="#module-lmflow.optim.yogi" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.optim.yogi.Yogi" title="lmflow.optim.yogi.Yogi"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Yogi</span></code></a></p></td>
+<td><p>Implements Yogi Optimizer Algorithm.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.optim.yogi.Yogi">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.optim.yogi.</span></span><span class="sig-name descname"><span class="pre">Yogi</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">params</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lr</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.01</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">betas</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">(0.9,</span> <span class="pre">0.999)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.001</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">initial_accumulator</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1e-06</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight_decay</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/yogi.html#Yogi"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.yogi.Yogi" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.optim.optimizer.Optimizer</span></code></p>
+<p>Implements Yogi Optimizer Algorithm.
+It has been proposed in <cite>Adaptive methods for Nonconvex Optimization</cite>.</p>
+<p><a class="reference external" href="https://papers.nips.cc/paper/8186-adaptive-methods-for-nonconvex-optimization">https://papers.nips.cc/paper/8186-adaptive-methods-for-nonconvex-optimization</a>  # noqa</p>
+<dl class="simple">
+<dt>Note:</dt><dd><p>Reference code: <a class="github reference external" href="https://github.com/4rtemi5/Yogi-Optimizer_Keras">4rtemi5/Yogi-Optimizer_Keras</a></p>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.optim.yogi.Yogi.step">
+<span class="sig-name descname"><span class="pre">step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">closure</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/optim/yogi.html#Yogi.step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.optim.yogi.Yogi.step" title="Link to this definition">#</a></dt>
+<dd><p>Performs a single optimization step.</p>
+<dl class="simple">
+<dt>Arguments:</dt><dd><p>closure: A closure that reevaluates the model and returns the loss.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../sophia/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.sophia</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../../pipeline/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.yogi.Yogi"><code class="docutils literal notranslate"><span class="pre">Yogi</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.optim.yogi.Yogi.step"><code class="docutils literal notranslate"><span class="pre">Yogi.step()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/optim/yogi/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/auto_pipeline/index.html b/autoapi/lmflow/pipeline/auto_pipeline/index.html
new file mode 100644
index 000000000..aa29c910e
--- /dev/null
+++ b/autoapi/lmflow/pipeline/auto_pipeline/index.html
@@ -0,0 +1,692 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.auto_pipeline &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/auto_pipeline/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.base_aligner" href="../base_aligner/index.html" />
+    <link rel="prev" title="lmflow.pipeline" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.auto_pipeline</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.auto_pipeline">
+<span id="lmflow-pipeline-auto-pipeline"></span><h1>lmflow.pipeline.auto_pipeline<a class="headerlink" href="#module-lmflow.pipeline.auto_pipeline" title="Link to this heading">#</a></h1>
+<p>Return a pipeline automatically based on its name.</p>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.auto_pipeline.PIPELINE_MAPPING" title="lmflow.pipeline.auto_pipeline.PIPELINE_MAPPING"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PIPELINE_MAPPING</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.auto_pipeline.PIPELINE_NEEDS_EXTRAS" title="lmflow.pipeline.auto_pipeline.PIPELINE_NEEDS_EXTRAS"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PIPELINE_NEEDS_EXTRAS</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.auto_pipeline.AutoPipeline" title="lmflow.pipeline.auto_pipeline.AutoPipeline"><code class="xref py py-obj docutils literal notranslate"><span class="pre">AutoPipeline</span></code></a></p></td>
+<td><p>The class designed to return a pipeline automatically based on its name.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.auto_pipeline.PIPELINE_MAPPING">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.auto_pipeline.</span></span><span class="sig-name descname"><span class="pre">PIPELINE_MAPPING</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/auto_pipeline.html#PIPELINE_MAPPING"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.auto_pipeline.PIPELINE_MAPPING" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.auto_pipeline.PIPELINE_NEEDS_EXTRAS">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.auto_pipeline.</span></span><span class="sig-name descname"><span class="pre">PIPELINE_NEEDS_EXTRAS</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">[]</span></em><a class="reference internal" href="../../../../_modules/lmflow/pipeline/auto_pipeline.html#PIPELINE_NEEDS_EXTRAS"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.auto_pipeline.PIPELINE_NEEDS_EXTRAS" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.auto_pipeline.AutoPipeline">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.auto_pipeline.</span></span><span class="sig-name descname"><span class="pre">AutoPipeline</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/auto_pipeline.html#AutoPipeline"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.auto_pipeline.AutoPipeline" title="Link to this definition">#</a></dt>
+<dd><p>The class designed to return a pipeline automatically based on its name.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.auto_pipeline.AutoPipeline.get_pipeline">
+<em class="property"><span class="pre">classmethod</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_pipeline</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pipeline_name</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pipeline_args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/auto_pipeline.html#AutoPipeline.get_pipeline"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.auto_pipeline.AutoPipeline.get_pipeline" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../base_aligner/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.base_aligner</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.auto_pipeline.PIPELINE_MAPPING"><code class="docutils literal notranslate"><span class="pre">PIPELINE_MAPPING</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.auto_pipeline.PIPELINE_NEEDS_EXTRAS"><code class="docutils literal notranslate"><span class="pre">PIPELINE_NEEDS_EXTRAS</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.auto_pipeline.AutoPipeline"><code class="docutils literal notranslate"><span class="pre">AutoPipeline</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.auto_pipeline.AutoPipeline.get_pipeline"><code class="docutils literal notranslate"><span class="pre">AutoPipeline.get_pipeline()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/auto_pipeline/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/base_aligner/index.html b/autoapi/lmflow/pipeline/base_aligner/index.html
new file mode 100644
index 000000000..d4c3e1511
--- /dev/null
+++ b/autoapi/lmflow/pipeline/base_aligner/index.html
@@ -0,0 +1,672 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.base_aligner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/base_aligner/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.base_pipeline" href="../base_pipeline/index.html" />
+    <link rel="prev" title="lmflow.pipeline.auto_pipeline" href="../auto_pipeline/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.base_aligner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.base_aligner">
+<span id="lmflow-pipeline-base-aligner"></span><h1>lmflow.pipeline.base_aligner<a class="headerlink" href="#module-lmflow.pipeline.base_aligner" title="Link to this heading">#</a></h1>
+<p>BaseTuner: a subclass of BasePipeline.</p>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.base_aligner.BaseAligner" title="lmflow.pipeline.base_aligner.BaseAligner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">BaseAligner</span></code></a></p></td>
+<td><p>A subclass of BasePipeline which is alignable.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.base_aligner.BaseAligner">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.base_aligner.</span></span><span class="sig-name descname"><span class="pre">BaseAligner</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/base_aligner.html#BaseAligner"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.base_aligner.BaseAligner" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_pipeline/index.html#lmflow.pipeline.base_pipeline.BasePipeline" title="lmflow.pipeline.base_pipeline.BasePipeline"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.pipeline.base_pipeline.BasePipeline</span></code></a></p>
+<p>A subclass of BasePipeline which is alignable.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.base_aligner.BaseAligner._check_if_alignable">
+<span class="sig-name descname"><span class="pre">_check_if_alignable</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">reward_model</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/base_aligner.html#BaseAligner._check_if_alignable"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.base_aligner.BaseAligner._check_if_alignable" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.base_aligner.BaseAligner.align">
+<em class="property"><span class="pre">abstract</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">align</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">reward_model</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/base_aligner.html#BaseAligner.align"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.base_aligner.BaseAligner.align" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../auto_pipeline/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.auto_pipeline</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../base_pipeline/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.base_pipeline</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.base_aligner.BaseAligner"><code class="docutils literal notranslate"><span class="pre">BaseAligner</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.base_aligner.BaseAligner._check_if_alignable"><code class="docutils literal notranslate"><span class="pre">BaseAligner._check_if_alignable()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.base_aligner.BaseAligner.align"><code class="docutils literal notranslate"><span class="pre">BaseAligner.align()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/base_aligner/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/base_pipeline/index.html b/autoapi/lmflow/pipeline/base_pipeline/index.html
new file mode 100644
index 000000000..e600a5c9d
--- /dev/null
+++ b/autoapi/lmflow/pipeline/base_pipeline/index.html
@@ -0,0 +1,659 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.base_pipeline &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/base_pipeline/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.base_tuner" href="../base_tuner/index.html" />
+    <link rel="prev" title="lmflow.pipeline.base_aligner" href="../base_aligner/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.base_pipeline</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.base_pipeline">
+<span id="lmflow-pipeline-base-pipeline"></span><h1>lmflow.pipeline.base_pipeline<a class="headerlink" href="#module-lmflow.pipeline.base_pipeline" title="Link to this heading">#</a></h1>
+<p>BasePipeline.</p>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.base_pipeline.BasePipeline" title="lmflow.pipeline.base_pipeline.BasePipeline"><code class="xref py py-obj docutils literal notranslate"><span class="pre">BasePipeline</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.base_pipeline.BasePipeline">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.base_pipeline.</span></span><span class="sig-name descname"><span class="pre">BasePipeline</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/base_pipeline.html#BasePipeline"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.base_pipeline.BasePipeline" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">abc.ABC</span></code></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../base_aligner/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.base_aligner</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../base_tuner/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.base_tuner</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.base_pipeline.BasePipeline"><code class="docutils literal notranslate"><span class="pre">BasePipeline</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/base_pipeline/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/base_tuner/index.html b/autoapi/lmflow/pipeline/base_tuner/index.html
new file mode 100644
index 000000000..1abcde6e3
--- /dev/null
+++ b/autoapi/lmflow/pipeline/base_tuner/index.html
@@ -0,0 +1,672 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.base_tuner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/base_tuner/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.dpo_aligner" href="../dpo_aligner/index.html" />
+    <link rel="prev" title="lmflow.pipeline.base_pipeline" href="../base_pipeline/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.base_tuner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.base_tuner">
+<span id="lmflow-pipeline-base-tuner"></span><h1>lmflow.pipeline.base_tuner<a class="headerlink" href="#module-lmflow.pipeline.base_tuner" title="Link to this heading">#</a></h1>
+<p>BaseTuner: a subclass of BasePipeline.</p>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.base_tuner.BaseTuner" title="lmflow.pipeline.base_tuner.BaseTuner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">BaseTuner</span></code></a></p></td>
+<td><p>A subclass of BasePipeline which is tunable.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.base_tuner.BaseTuner">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.base_tuner.</span></span><span class="sig-name descname"><span class="pre">BaseTuner</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/base_tuner.html#BaseTuner"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.base_tuner.BaseTuner" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_pipeline/index.html#lmflow.pipeline.base_pipeline.BasePipeline" title="lmflow.pipeline.base_pipeline.BasePipeline"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.pipeline.base_pipeline.BasePipeline</span></code></a></p>
+<p>A subclass of BasePipeline which is tunable.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.base_tuner.BaseTuner._check_if_tunable">
+<span class="sig-name descname"><span class="pre">_check_if_tunable</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/base_tuner.html#BaseTuner._check_if_tunable"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.base_tuner.BaseTuner._check_if_tunable" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.base_tuner.BaseTuner.tune">
+<em class="property"><span class="pre">abstract</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">tune</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/base_tuner.html#BaseTuner.tune"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.base_tuner.BaseTuner.tune" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../base_pipeline/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.base_pipeline</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../dpo_aligner/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.dpo_aligner</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.base_tuner.BaseTuner"><code class="docutils literal notranslate"><span class="pre">BaseTuner</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.base_tuner.BaseTuner._check_if_tunable"><code class="docutils literal notranslate"><span class="pre">BaseTuner._check_if_tunable()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.base_tuner.BaseTuner.tune"><code class="docutils literal notranslate"><span class="pre">BaseTuner.tune()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/base_tuner/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/dpo_aligner/index.html b/autoapi/lmflow/pipeline/dpo_aligner/index.html
new file mode 100644
index 000000000..1f6151b93
--- /dev/null
+++ b/autoapi/lmflow/pipeline/dpo_aligner/index.html
@@ -0,0 +1,740 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.dpo_aligner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/dpo_aligner/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.dpov2_aligner" href="../dpov2_aligner/index.html" />
+    <link rel="prev" title="lmflow.pipeline.base_tuner" href="../base_tuner/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.dpo_aligner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.dpo_aligner">
+<span id="lmflow-pipeline-dpo-aligner"></span><h1>lmflow.pipeline.dpo_aligner<a class="headerlink" href="#module-lmflow.pipeline.dpo_aligner" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.dpo_aligner.DPOAligner" title="lmflow.pipeline.dpo_aligner.DPOAligner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DPOAligner</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.dpo_aligner.get_paired_dataset" title="lmflow.pipeline.dpo_aligner.get_paired_dataset"><code class="xref py py-obj docutils literal notranslate"><span class="pre">get_paired_dataset</span></code></a>(→ datasets.Dataset)</p></td>
+<td><p>Load dataset and convert it to the necessary format.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpo_aligner.get_paired_dataset">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.dpo_aligner.</span></span><span class="sig-name descname"><span class="pre">get_paired_dataset</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data_root</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_dir</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sanity_check</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">cache_dir</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_proc</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">24</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference internal" href="../../datasets/index.html#lmflow.datasets.Dataset" title="lmflow.datasets.Dataset"><span class="pre">datasets.Dataset</span></a></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpo_aligner.html#get_paired_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpo_aligner.get_paired_dataset" title="Link to this definition">#</a></dt>
+<dd><p>Load dataset and convert it to the necessary format.</p>
+<blockquote>
+<div><p>The dataset is converted to a dictionary with the following structure:
+{</p>
+<blockquote>
+<div><p>‘prompt’: List[str],
+‘chosen’: List[str],
+‘rejected’: List[str],</p>
+</div></blockquote>
+<p>}</p>
+<dl class="simple">
+<dt>Prompts are structured as follows:</dt><dd><p>“Question: “ + &lt;prompt&gt; + “</p>
+</dd>
+</dl>
+</div></blockquote>
+<p>Answer: “</p>
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpo_aligner.DPOAligner">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.dpo_aligner.</span></span><span class="sig-name descname"><span class="pre">DPOAligner</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">aligner_args</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpo_aligner.html#DPOAligner"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpo_aligner.DPOAligner" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_aligner/index.html#lmflow.pipeline.base_aligner.BaseAligner" title="lmflow.pipeline.base_aligner.BaseAligner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.pipeline.base_aligner.BaseAligner</span></code></a></p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpo_aligner.DPOAligner.model_args">
+<span class="sig-name descname"><span class="pre">model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpo_aligner.html#DPOAligner.model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpo_aligner.DPOAligner.model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpo_aligner.DPOAligner.data_args">
+<span class="sig-name descname"><span class="pre">data_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpo_aligner.html#DPOAligner.data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpo_aligner.DPOAligner.data_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpo_aligner.DPOAligner.aligner_args">
+<span class="sig-name descname"><span class="pre">aligner_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpo_aligner.html#DPOAligner.aligner_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpo_aligner.DPOAligner.aligner_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpo_aligner.DPOAligner.train_dataset">
+<span class="sig-name descname"><span class="pre">train_dataset</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpo_aligner.html#DPOAligner.train_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpo_aligner.DPOAligner.train_dataset" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpo_aligner.DPOAligner.eval_dataset">
+<span class="sig-name descname"><span class="pre">eval_dataset</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpo_aligner.html#DPOAligner.eval_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpo_aligner.DPOAligner.eval_dataset" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpo_aligner.DPOAligner._initialize_trainer">
+<span class="sig-name descname"><span class="pre">_initialize_trainer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpo_aligner.html#DPOAligner._initialize_trainer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpo_aligner.DPOAligner._initialize_trainer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpo_aligner.DPOAligner._load_dataset">
+<span class="sig-name descname"><span class="pre">_load_dataset</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpo_aligner.html#DPOAligner._load_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpo_aligner.DPOAligner._load_dataset" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpo_aligner.DPOAligner.align">
+<span class="sig-name descname"><span class="pre">align</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">reward_model</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpo_aligner.html#DPOAligner.align"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpo_aligner.DPOAligner.align" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../base_tuner/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.base_tuner</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../dpov2_aligner/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.dpov2_aligner</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpo_aligner.get_paired_dataset"><code class="docutils literal notranslate"><span class="pre">get_paired_dataset()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpo_aligner.DPOAligner"><code class="docutils literal notranslate"><span class="pre">DPOAligner</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpo_aligner.DPOAligner.model_args"><code class="docutils literal notranslate"><span class="pre">DPOAligner.model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpo_aligner.DPOAligner.data_args"><code class="docutils literal notranslate"><span class="pre">DPOAligner.data_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpo_aligner.DPOAligner.aligner_args"><code class="docutils literal notranslate"><span class="pre">DPOAligner.aligner_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpo_aligner.DPOAligner.train_dataset"><code class="docutils literal notranslate"><span class="pre">DPOAligner.train_dataset</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpo_aligner.DPOAligner.eval_dataset"><code class="docutils literal notranslate"><span class="pre">DPOAligner.eval_dataset</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpo_aligner.DPOAligner._initialize_trainer"><code class="docutils literal notranslate"><span class="pre">DPOAligner._initialize_trainer()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpo_aligner.DPOAligner._load_dataset"><code class="docutils literal notranslate"><span class="pre">DPOAligner._load_dataset()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpo_aligner.DPOAligner.align"><code class="docutils literal notranslate"><span class="pre">DPOAligner.align()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/dpo_aligner/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/dpov2_aligner/index.html b/autoapi/lmflow/pipeline/dpov2_aligner/index.html
new file mode 100644
index 000000000..b0e2f781c
--- /dev/null
+++ b/autoapi/lmflow/pipeline/dpov2_aligner/index.html
@@ -0,0 +1,813 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.dpov2_aligner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/dpov2_aligner/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.evaluator" href="../evaluator/index.html" />
+    <link rel="prev" title="lmflow.pipeline.dpo_aligner" href="../dpo_aligner/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.dpov2_aligner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.dpov2_aligner">
+<span id="lmflow-pipeline-dpov2-aligner"></span><h1>lmflow.pipeline.dpov2_aligner<a class="headerlink" href="#module-lmflow.pipeline.dpov2_aligner" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.dpov2_aligner.logger" title="lmflow.pipeline.dpov2_aligner.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.dpov2_aligner.ReferenceModelArguments" title="lmflow.pipeline.dpov2_aligner.ReferenceModelArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ReferenceModelArguments</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner" title="lmflow.pipeline.dpov2_aligner.DPOv2Aligner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DPOv2Aligner</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner" title="lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">MemorySafeDPOv2Aligner</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.dpov2_aligner.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.ReferenceModelArguments">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.dpov2_aligner.</span></span><span class="sig-name descname"><span class="pre">ReferenceModelArguments</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#ReferenceModelArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.ReferenceModelArguments" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.DPOv2Aligner">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.dpov2_aligner.</span></span><span class="sig-name descname"><span class="pre">DPOv2Aligner</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">aligner_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DPOv2AlignerArguments" title="lmflow.args.DPOv2AlignerArguments"><span class="pre">lmflow.args.DPOv2AlignerArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">ref_model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#DPOv2Aligner"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_aligner/index.html#lmflow.pipeline.base_aligner.BaseAligner" title="lmflow.pipeline.base_aligner.BaseAligner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.pipeline.base_aligner.BaseAligner</span></code></a></p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.DPOv2Aligner.model_args">
+<span class="sig-name descname"><span class="pre">model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#DPOv2Aligner.model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.DPOv2Aligner.ref_model_args">
+<span class="sig-name descname"><span class="pre">ref_model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#DPOv2Aligner.ref_model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.ref_model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.DPOv2Aligner.data_args">
+<span class="sig-name descname"><span class="pre">data_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#DPOv2Aligner.data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.data_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.DPOv2Aligner.aligner_args">
+<span class="sig-name descname"><span class="pre">aligner_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#DPOv2Aligner.aligner_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.aligner_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.DPOv2Aligner.align">
+<span class="sig-name descname"><span class="pre">align</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel" title="lmflow.models.hf_decoder_model.HFDecoderModel"><span class="pre">lmflow.models.hf_decoder_model.HFDecoderModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">ref_model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel" title="lmflow.models.hf_decoder_model.HFDecoderModel"><span class="pre">lmflow.models.hf_decoder_model.HFDecoderModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">train_dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">eval_dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform_dataset_in_place</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#DPOv2Aligner.align"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.align" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.DPOv2Aligner.__prepare_training_args">
+<span class="sig-name descname"><span class="pre">__prepare_training_args</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DPOv2AlignerArguments" title="lmflow.args.DPOv2AlignerArguments"><span class="pre">lmflow.args.DPOv2AlignerArguments</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">transformers.TrainingArguments</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#DPOv2Aligner.__prepare_training_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.__prepare_training_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.DPOv2Aligner.convert_to_paired_dataset">
+<span class="sig-name descname"><span class="pre">convert_to_paired_dataset</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">source_dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampling_paired_method</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'random'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">length_penalty</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">margin_scale</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_fast</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#DPOv2Aligner.convert_to_paired_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.convert_to_paired_dataset" title="Link to this definition">#</a></dt>
+<dd><p>Convert a scored one to multiple (text_to_scored_textlist) to a paired dataset by rejection sampling.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.DPOv2Aligner._calc_response_lengths">
+<span class="sig-name descname"><span class="pre">_calc_response_lengths</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">outputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset_type</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#DPOv2Aligner._calc_response_lengths"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._calc_response_lengths" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.DPOv2Aligner._calc_reward_with_length_penalty">
+<span class="sig-name descname"><span class="pre">_calc_reward_with_length_penalty</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">rewards</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lengths</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">length_penalty</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#DPOv2Aligner._calc_reward_with_length_penalty"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._calc_reward_with_length_penalty" title="Link to this definition">#</a></dt>
+<dd><p>When length_penalty &gt; 0, penalize the longer sequence by subtracting
+length_penalty * length from the reward. Vice versa when length_penalty &lt; 0.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.DPOv2Aligner.sampling_paired_idx_from_rewards">
+<span class="sig-name descname"><span class="pre">sampling_paired_idx_from_rewards</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">rewards</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampling_paired_method</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'random'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_fast</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#DPOv2Aligner.sampling_paired_idx_from_rewards"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.sampling_paired_idx_from_rewards" title="Link to this definition">#</a></dt>
+<dd><p>Prepare the dataset for DPO training by rejection sampling.
+We implement different strategies to select pairs, including
+random: randomly select two instances
+max_min: best v.s. worst
+max_max: best v.s. second best
+max_random: best v.s. random from the remaining</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.DPOv2Aligner._sampling_paired_idx_from_rewards">
+<span class="sig-name descname"><span class="pre">_sampling_paired_idx_from_rewards</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">rewards</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampling_paired_method</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'random'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#DPOv2Aligner._sampling_paired_idx_from_rewards"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._sampling_paired_idx_from_rewards" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.DPOv2Aligner._sampling_paired_idx_from_rewards_fast">
+<span class="sig-name descname"><span class="pre">_sampling_paired_idx_from_rewards_fast</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">rewards</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampling_paired_method</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'random'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#DPOv2Aligner._sampling_paired_idx_from_rewards_fast"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._sampling_paired_idx_from_rewards_fast" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.dpov2_aligner.</span></span><span class="sig-name descname"><span class="pre">MemorySafeDPOv2Aligner</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">aligner_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DPOv2AlignerArguments" title="lmflow.args.DPOv2AlignerArguments"><span class="pre">lmflow.args.DPOv2AlignerArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">ref_model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#MemorySafeDPOv2Aligner"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner" title="Link to this definition">#</a></dt>
+<dd><dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.model_args">
+<span class="sig-name descname"><span class="pre">model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#MemorySafeDPOv2Aligner.model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.ref_model_args">
+<span class="sig-name descname"><span class="pre">ref_model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#MemorySafeDPOv2Aligner.ref_model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.ref_model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.data_args">
+<span class="sig-name descname"><span class="pre">data_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#MemorySafeDPOv2Aligner.data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.data_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.aligner_args">
+<span class="sig-name descname"><span class="pre">aligner_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#MemorySafeDPOv2Aligner.aligner_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.aligner_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.aligner_file_path">
+<span class="sig-name descname"><span class="pre">aligner_file_path</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#MemorySafeDPOv2Aligner.aligner_file_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.aligner_file_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.align">
+<span class="sig-name descname"><span class="pre">align</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/dpov2_aligner.html#MemorySafeDPOv2Aligner.align"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.align" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../dpo_aligner/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.dpo_aligner</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../evaluator/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.evaluator</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.ReferenceModelArguments"><code class="docutils literal notranslate"><span class="pre">ReferenceModelArguments</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner"><code class="docutils literal notranslate"><span class="pre">DPOv2Aligner</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.model_args"><code class="docutils literal notranslate"><span class="pre">DPOv2Aligner.model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.ref_model_args"><code class="docutils literal notranslate"><span class="pre">DPOv2Aligner.ref_model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.data_args"><code class="docutils literal notranslate"><span class="pre">DPOv2Aligner.data_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.aligner_args"><code class="docutils literal notranslate"><span class="pre">DPOv2Aligner.aligner_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.align"><code class="docutils literal notranslate"><span class="pre">DPOv2Aligner.align()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.__prepare_training_args"><code class="docutils literal notranslate"><span class="pre">DPOv2Aligner.__prepare_training_args()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.convert_to_paired_dataset"><code class="docutils literal notranslate"><span class="pre">DPOv2Aligner.convert_to_paired_dataset()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._calc_response_lengths"><code class="docutils literal notranslate"><span class="pre">DPOv2Aligner._calc_response_lengths()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._calc_reward_with_length_penalty"><code class="docutils literal notranslate"><span class="pre">DPOv2Aligner._calc_reward_with_length_penalty()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.sampling_paired_idx_from_rewards"><code class="docutils literal notranslate"><span class="pre">DPOv2Aligner.sampling_paired_idx_from_rewards()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._sampling_paired_idx_from_rewards"><code class="docutils literal notranslate"><span class="pre">DPOv2Aligner._sampling_paired_idx_from_rewards()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._sampling_paired_idx_from_rewards_fast"><code class="docutils literal notranslate"><span class="pre">DPOv2Aligner._sampling_paired_idx_from_rewards_fast()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner"><code class="docutils literal notranslate"><span class="pre">MemorySafeDPOv2Aligner</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.model_args"><code class="docutils literal notranslate"><span class="pre">MemorySafeDPOv2Aligner.model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.ref_model_args"><code class="docutils literal notranslate"><span class="pre">MemorySafeDPOv2Aligner.ref_model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.data_args"><code class="docutils literal notranslate"><span class="pre">MemorySafeDPOv2Aligner.data_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.aligner_args"><code class="docutils literal notranslate"><span class="pre">MemorySafeDPOv2Aligner.aligner_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.aligner_file_path"><code class="docutils literal notranslate"><span class="pre">MemorySafeDPOv2Aligner.aligner_file_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.align"><code class="docutils literal notranslate"><span class="pre">MemorySafeDPOv2Aligner.align()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/dpov2_aligner/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/evaluator/index.html b/autoapi/lmflow/pipeline/evaluator/index.html
new file mode 100644
index 000000000..d78383a83
--- /dev/null
+++ b/autoapi/lmflow/pipeline/evaluator/index.html
@@ -0,0 +1,777 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.evaluator &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/evaluator/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.finetuner" href="../finetuner/index.html" />
+    <link rel="prev" title="lmflow.pipeline.dpov2_aligner" href="../dpov2_aligner/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.evaluator</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.evaluator">
+<span id="lmflow-pipeline-evaluator"></span><h1>lmflow.pipeline.evaluator<a class="headerlink" href="#module-lmflow.pipeline.evaluator" title="Link to this heading">#</a></h1>
+<p>The Evaluator class simplifies the process of running evaluation on a language model provided by a HFDecoderModel instance imported from the lmflow package. The class constructor takes three dictionaries as arguments: model_args containing arguments related to the language model, data_args containing arguments related to the data used for evaluation, and evaluator_args containing other arguments for the evaluation process.</p>
+<p>The class has two methods: create_dataloader() that loads the data from the test file, creates a data loader, and returns it with the size of the data, and evaluate(model) that generates output text given input text. It uses the create_dataloader() method to load the data, iterates over the data in mini-batches, and encodes the input text with the encode() method of the HFDecoderModel class. Then, it generates output text using the evaluate() method of the HFDecoderModel class, decodes the generated output text using the decode() method of the HFDecoderModel class, and writes the output to a file in the output directory. The method also logs some information to the console and Weights and Biases if the use_wandb argument is True.</p>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.evaluator.Evaluator" title="lmflow.pipeline.evaluator.Evaluator"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Evaluator</span></code></a></p></td>
+<td><p>Initializes the <cite>Evaluator</cite> class with given arguments.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.evaluator.</span></span><span class="sig-name descname"><span class="pre">Evaluator</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">evaluator_args</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_pipeline/index.html#lmflow.pipeline.base_pipeline.BasePipeline" title="lmflow.pipeline.base_pipeline.BasePipeline"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.pipeline.base_pipeline.BasePipeline</span></code></a></p>
+<p>Initializes the <cite>Evaluator</cite> class with given arguments.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model_args</strong><span class="classifier">ModelArguments object.</span></dt><dd><p>Contains the arguments required to load the model.</p>
+</dd>
+<dt><strong>data_args</strong><span class="classifier">DatasetArguments object.</span></dt><dd><p>Contains the arguments required to load the dataset.</p>
+</dd>
+<dt><strong>evaluator_args</strong><span class="classifier">EvaluatorArguments object.</span></dt><dd><p>Contains the arguments required to perform evaluation.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator.data_args">
+<span class="sig-name descname"><span class="pre">data_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator.data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator.data_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator.evaluator_args">
+<span class="sig-name descname"><span class="pre">evaluator_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator.evaluator_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator.evaluator_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator.model_args">
+<span class="sig-name descname"><span class="pre">model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator.model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator.model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator.local_rank">
+<span class="sig-name descname"><span class="pre">local_rank</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator.local_rank"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator.local_rank" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator.world_size">
+<span class="sig-name descname"><span class="pre">world_size</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator.world_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator.world_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator.config">
+<span class="sig-name descname"><span class="pre">config</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator.config"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator.config" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator.block_size">
+<span class="sig-name descname"><span class="pre">block_size</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator.block_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator.block_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator.create_dataloader">
+<span class="sig-name descname"><span class="pre">create_dataloader</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator.create_dataloader"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator.create_dataloader" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator._match">
+<span class="sig-name descname"><span class="pre">_match</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">predicted_answer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">groundtruth</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">answer_type</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator._match"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator._match" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator.evaluate">
+<span class="sig-name descname"><span class="pre">evaluate</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">metric</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'accuracy'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">verbose</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator.evaluate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator.evaluate" title="Link to this definition">#</a></dt>
+<dd><p>Perform Evaluation for a model</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model</strong><span class="classifier">TunableModel object.</span></dt><dd><p>TunableModel to perform inference</p>
+</dd>
+<dt><strong>dataset</strong><span class="classifier">Dataset object.</span></dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator._evaluate_acc_with_accelerator">
+<span class="sig-name descname"><span class="pre">_evaluate_acc_with_accelerator</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">verbose</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator._evaluate_acc_with_accelerator"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator._evaluate_acc_with_accelerator" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator._evaluate_acc_with_deepspeed">
+<span class="sig-name descname"><span class="pre">_evaluate_acc_with_deepspeed</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">verbose</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator._evaluate_acc_with_deepspeed"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator._evaluate_acc_with_deepspeed" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator._evaluate_ppl">
+<span class="sig-name descname"><span class="pre">_evaluate_ppl</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">verbose</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator._evaluate_ppl"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator._evaluate_ppl" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.evaluator.Evaluator._evaluate_nll">
+<span class="sig-name descname"><span class="pre">_evaluate_nll</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">verbose</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/evaluator.html#Evaluator._evaluate_nll"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.evaluator.Evaluator._evaluate_nll" title="Link to this definition">#</a></dt>
+<dd><p>Evaluates negative log likelihood of the model over a dataset.</p>
+<p>NLL = -1/N sum_{i=1}^N sum_{j=1}^|w_i| ln(p(w_{i,j}|context_window)),</p>
+<p>where N is the number of data samples, w_{i,j} is the j-th token in
+i-th sample. Here “context_window” = p(w_{i,start}, w_{i,start+1}, …,
+p_{i,j-1} with start = max(0, j - window_length + 1). “window_length”
+is normally the maximum length accepted by the model.</p>
+<dl class="simple">
+<dt>Returns:</dt><dd><p>A float which represents the negative log likelihood.</p>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../dpov2_aligner/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.dpov2_aligner</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../finetuner/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.finetuner</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator"><code class="docutils literal notranslate"><span class="pre">Evaluator</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator.data_args"><code class="docutils literal notranslate"><span class="pre">Evaluator.data_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator.evaluator_args"><code class="docutils literal notranslate"><span class="pre">Evaluator.evaluator_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator.model_args"><code class="docutils literal notranslate"><span class="pre">Evaluator.model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator.local_rank"><code class="docutils literal notranslate"><span class="pre">Evaluator.local_rank</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator.world_size"><code class="docutils literal notranslate"><span class="pre">Evaluator.world_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator.config"><code class="docutils literal notranslate"><span class="pre">Evaluator.config</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator.block_size"><code class="docutils literal notranslate"><span class="pre">Evaluator.block_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator.create_dataloader"><code class="docutils literal notranslate"><span class="pre">Evaluator.create_dataloader()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator._match"><code class="docutils literal notranslate"><span class="pre">Evaluator._match()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator.evaluate"><code class="docutils literal notranslate"><span class="pre">Evaluator.evaluate()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator._evaluate_acc_with_accelerator"><code class="docutils literal notranslate"><span class="pre">Evaluator._evaluate_acc_with_accelerator()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator._evaluate_acc_with_deepspeed"><code class="docutils literal notranslate"><span class="pre">Evaluator._evaluate_acc_with_deepspeed()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator._evaluate_ppl"><code class="docutils literal notranslate"><span class="pre">Evaluator._evaluate_ppl()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.evaluator.Evaluator._evaluate_nll"><code class="docutils literal notranslate"><span class="pre">Evaluator._evaluate_nll()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/evaluator/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/finetuner/index.html b/autoapi/lmflow/pipeline/finetuner/index.html
new file mode 100644
index 000000000..b4c4c5080
--- /dev/null
+++ b/autoapi/lmflow/pipeline/finetuner/index.html
@@ -0,0 +1,749 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.finetuner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/finetuner/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.inferencer" href="../inferencer/index.html" />
+    <link rel="prev" title="lmflow.pipeline.evaluator" href="../evaluator/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.finetuner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.finetuner">
+<span id="lmflow-pipeline-finetuner"></span><h1>lmflow.pipeline.finetuner<a class="headerlink" href="#module-lmflow.pipeline.finetuner" title="Link to this heading">#</a></h1>
+<p>The Finetuner class simplifies the process of running finetuning process on a language model for a TunableModel instance with given dataset.</p>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.finetuner.logger" title="lmflow.pipeline.finetuner.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.finetuner.Finetuner" title="lmflow.pipeline.finetuner.Finetuner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Finetuner</span></code></a></p></td>
+<td><p>Initializes the <cite>Finetuner</cite> class with given arguments.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.finetuner.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.finetuner.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/finetuner.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.finetuner.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.finetuner.Finetuner">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.finetuner.</span></span><span class="sig-name descname"><span class="pre">Finetuner</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">finetuner_args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/finetuner.html#Finetuner"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.finetuner.Finetuner" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_tuner/index.html#lmflow.pipeline.base_tuner.BaseTuner" title="lmflow.pipeline.base_tuner.BaseTuner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.pipeline.base_tuner.BaseTuner</span></code></a></p>
+<p>Initializes the <cite>Finetuner</cite> class with given arguments.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model_args</strong><span class="classifier">ModelArguments object.</span></dt><dd><p>Contains the arguments required to load the model.</p>
+</dd>
+<dt><strong>data_args</strong><span class="classifier">DatasetArguments object.</span></dt><dd><p>Contains the arguments required to load the dataset.</p>
+</dd>
+<dt><strong>finetuner_args</strong><span class="classifier">FinetunerArguments object.</span></dt><dd><p>Contains the arguments required to perform finetuning.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.finetuner.Finetuner.model_args">
+<span class="sig-name descname"><span class="pre">model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/finetuner.html#Finetuner.model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.finetuner.Finetuner.model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.finetuner.Finetuner.data_args">
+<span class="sig-name descname"><span class="pre">data_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/finetuner.html#Finetuner.data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.finetuner.Finetuner.data_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.finetuner.Finetuner.finetuner_args">
+<span class="sig-name descname"><span class="pre">finetuner_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/finetuner.html#Finetuner.finetuner_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.finetuner.Finetuner.finetuner_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.finetuner.Finetuner.last_checkpoint">
+<span class="sig-name descname"><span class="pre">last_checkpoint</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../_modules/lmflow/pipeline/finetuner.html#Finetuner.last_checkpoint"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.finetuner.Finetuner.last_checkpoint" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.finetuner.Finetuner.group_text">
+<span class="sig-name descname"><span class="pre">group_text</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenized_datasets</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_max_length</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/finetuner.html#Finetuner.group_text"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.finetuner.Finetuner.group_text" title="Link to this definition">#</a></dt>
+<dd><p>Groups texts together to form blocks of maximum length <cite>model_max_length</cite> and returns the processed data as
+a dictionary.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.finetuner.Finetuner.create_customized_optimizer">
+<span class="sig-name descname"><span class="pre">create_customized_optimizer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">base_trainer_class</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/finetuner.html#Finetuner.create_customized_optimizer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.finetuner.Finetuner.create_customized_optimizer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.finetuner.Finetuner.tune">
+<span class="sig-name descname"><span class="pre">tune</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform_dataset_in_place</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_collator</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/finetuner.html#Finetuner.tune"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.finetuner.Finetuner.tune" title="Link to this definition">#</a></dt>
+<dd><p>Perform tuning for a model</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model</strong><span class="classifier">TunableModel object.</span></dt><dd><p>TunableModel to perform tuning.</p>
+</dd>
+<dt><strong>dataset:</strong></dt><dd><p>dataset to train model.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../evaluator/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.evaluator</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../inferencer/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.inferencer</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.finetuner.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.finetuner.Finetuner"><code class="docutils literal notranslate"><span class="pre">Finetuner</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.finetuner.Finetuner.model_args"><code class="docutils literal notranslate"><span class="pre">Finetuner.model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.finetuner.Finetuner.data_args"><code class="docutils literal notranslate"><span class="pre">Finetuner.data_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.finetuner.Finetuner.finetuner_args"><code class="docutils literal notranslate"><span class="pre">Finetuner.finetuner_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.finetuner.Finetuner.last_checkpoint"><code class="docutils literal notranslate"><span class="pre">Finetuner.last_checkpoint</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.finetuner.Finetuner.group_text"><code class="docutils literal notranslate"><span class="pre">Finetuner.group_text()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.finetuner.Finetuner.create_customized_optimizer"><code class="docutils literal notranslate"><span class="pre">Finetuner.create_customized_optimizer()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.finetuner.Finetuner.tune"><code class="docutils literal notranslate"><span class="pre">Finetuner.tune()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/finetuner/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/index.html b/autoapi/lmflow/pipeline/index.html
new file mode 100644
index 000000000..3f9acc897
--- /dev/null
+++ b/autoapi/lmflow/pipeline/index.html
@@ -0,0 +1,651 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/index';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.auto_pipeline" href="auto_pipeline/index.html" />
+    <link rel="prev" title="lmflow.optim.yogi" href="../optim/yogi/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="current reference internal" href="#">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline">
+<span id="lmflow-pipeline"></span><h1>lmflow.pipeline<a class="headerlink" href="#module-lmflow.pipeline" title="Link to this heading">#</a></h1>
+<section id="submodules">
+<h2>Submodules<a class="headerlink" href="#submodules" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l1"><a class="reference internal" href="base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l1"><a class="reference internal" href="base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l1"><a class="reference internal" href="base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l1"><a class="reference internal" href="dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l1"><a class="reference internal" href="dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l1"><a class="reference internal" href="evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l1"><a class="reference internal" href="finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l1"><a class="reference internal" href="inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l1"><a class="reference internal" href="iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l1"><a class="reference internal" href="raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l1"><a class="reference internal" href="rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l1"><a class="reference internal" href="rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l1"><a class="reference internal" href="utils/index.html">lmflow.pipeline.utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../optim/yogi/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.optim.yogi</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="auto_pipeline/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.auto_pipeline</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#submodules">Submodules</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../_sources/autoapi/lmflow/pipeline/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/inferencer/index.html b/autoapi/lmflow/pipeline/inferencer/index.html
new file mode 100644
index 000000000..4df05a34c
--- /dev/null
+++ b/autoapi/lmflow/pipeline/inferencer/index.html
@@ -0,0 +1,971 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.inferencer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/inferencer/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.iterative_dpo_aligner" href="../iterative_dpo_aligner/index.html" />
+    <link rel="prev" title="lmflow.pipeline.finetuner" href="../finetuner/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.inferencer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.inferencer">
+<span id="lmflow-pipeline-inferencer"></span><h1>lmflow.pipeline.inferencer<a class="headerlink" href="#module-lmflow.pipeline.inferencer" title="Link to this heading">#</a></h1>
+<p>The Inferencer class simplifies the process of model inferencing.</p>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.inferencer.supported_dataset_type" title="lmflow.pipeline.inferencer.supported_dataset_type"><code class="xref py py-obj docutils literal notranslate"><span class="pre">supported_dataset_type</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.inferencer.logger" title="lmflow.pipeline.inferencer.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.inferencer.Inferencer" title="lmflow.pipeline.inferencer.Inferencer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Inferencer</span></code></a></p></td>
+<td><p>Initializes the <cite>Inferencer</cite> class with given arguments.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.inferencer.SpeculativeInferencer" title="lmflow.pipeline.inferencer.SpeculativeInferencer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SpeculativeInferencer</span></code></a></p></td>
+<td><p>Ref: [arXiv:2211.17192v2](https://arxiv.org/abs/2211.17192)</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.inferencer.ToolInferencer" title="lmflow.pipeline.inferencer.ToolInferencer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ToolInferencer</span></code></a></p></td>
+<td><p>Initializes the <cite>ToolInferencer</cite> class with given arguments.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.inferencer.rstrip_partial_utf8" title="lmflow.pipeline.inferencer.rstrip_partial_utf8"><code class="xref py py-obj docutils literal notranslate"><span class="pre">rstrip_partial_utf8</span></code></a>(string)</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.rstrip_partial_utf8">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.inferencer.</span></span><span class="sig-name descname"><span class="pre">rstrip_partial_utf8</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">string</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#rstrip_partial_utf8"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.rstrip_partial_utf8" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.supported_dataset_type">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.inferencer.</span></span><span class="sig-name descname"><span class="pre">supported_dataset_type</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">['text_only',</span> <span class="pre">'image_text']</span></em><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#supported_dataset_type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.supported_dataset_type" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.inferencer.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.Inferencer">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.inferencer.</span></span><span class="sig-name descname"><span class="pre">Inferencer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inferencer_args</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#Inferencer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.Inferencer" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_pipeline/index.html#lmflow.pipeline.base_pipeline.BasePipeline" title="lmflow.pipeline.base_pipeline.BasePipeline"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.pipeline.base_pipeline.BasePipeline</span></code></a></p>
+<p>Initializes the <cite>Inferencer</cite> class with given arguments.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model_args</strong><span class="classifier">ModelArguments object.</span></dt><dd><p>Contains the arguments required to load the model.</p>
+</dd>
+<dt><strong>data_args</strong><span class="classifier">DatasetArguments object.</span></dt><dd><p>Contains the arguments required to load the dataset.</p>
+</dd>
+<dt><strong>inferencer_args</strong><span class="classifier">InferencerArguments object.</span></dt><dd><p>Contains the arguments required to perform inference.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.Inferencer.data_args">
+<span class="sig-name descname"><span class="pre">data_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#Inferencer.data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.Inferencer.data_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.Inferencer.inferencer_args">
+<span class="sig-name descname"><span class="pre">inferencer_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#Inferencer.inferencer_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.Inferencer.inferencer_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.Inferencer.model_args">
+<span class="sig-name descname"><span class="pre">model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#Inferencer.model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.Inferencer.model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.Inferencer.local_rank">
+<span class="sig-name descname"><span class="pre">local_rank</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#Inferencer.local_rank"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.Inferencer.local_rank" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.Inferencer.world_size">
+<span class="sig-name descname"><span class="pre">world_size</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#Inferencer.world_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.Inferencer.world_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.Inferencer.config">
+<span class="sig-name descname"><span class="pre">config</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#Inferencer.config"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.Inferencer.config" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.Inferencer.create_dataloader">
+<span class="sig-name descname"><span class="pre">create_dataloader</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#Inferencer.create_dataloader"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.Inferencer.create_dataloader" title="Link to this definition">#</a></dt>
+<dd><p>Batchlize dataset and format it to dataloader.</p>
+<dl class="simple">
+<dt>Args:</dt><dd><p>dataset (Dataset): the dataset object</p>
+</dd>
+<dt>Output:</dt><dd><p>dataloader (batchlize): the dataloader object
+dataset_size (int): the length of the dataset</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.Inferencer.inference">
+<span class="sig-name descname"><span class="pre">inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_new_tokens</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">100</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">temperature</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">prompt_structure</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'{input}'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">remove_image_flag</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">chatbot_type</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'mini_gpt'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#Inferencer.inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.Inferencer.inference" title="Link to this definition">#</a></dt>
+<dd><p>Perform inference for a model</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model</strong><span class="classifier">TunableModel object.</span></dt><dd><p>TunableModel to perform inference</p>
+</dd>
+<dt><strong>dataset</strong><span class="classifier">Dataset object.</span></dt><dd></dd>
+<dt><strong>Returns:</strong></dt><dd></dd>
+<dt><strong>output_dataset: Dataset object.</strong></dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.Inferencer.stream_inference">
+<span class="sig-name descname"><span class="pre">stream_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">context</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_new_tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">token_per_step</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">temperature</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">end_string</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">input_dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">remove_image_flag</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#Inferencer.stream_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.Inferencer.stream_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.SpeculativeInferencer">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.inferencer.</span></span><span class="sig-name descname"><span class="pre">SpeculativeInferencer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">draft_model_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inferencer_args</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#SpeculativeInferencer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.SpeculativeInferencer" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.pipeline.inferencer.Inferencer" title="lmflow.pipeline.inferencer.Inferencer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Inferencer</span></code></a></p>
+<p>Ref: [arXiv:2211.17192v2](https://arxiv.org/abs/2211.17192)</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>target_model_args</strong><span class="classifier">ModelArguments object.</span></dt><dd><p>Contains the arguments required to load the target model.</p>
+</dd>
+<dt><strong>draft_model_args</strong><span class="classifier">ModelArguments object.</span></dt><dd><p>Contains the arguments required to load the draft model.</p>
+</dd>
+<dt><strong>data_args</strong><span class="classifier">DatasetArguments object.</span></dt><dd><p>Contains the arguments required to load the dataset.</p>
+</dd>
+<dt><strong>inferencer_args</strong><span class="classifier">InferencerArguments object.</span></dt><dd><p>Contains the arguments required to perform inference.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.SpeculativeInferencer.draft_model_args">
+<span class="sig-name descname"><span class="pre">draft_model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#SpeculativeInferencer.draft_model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.draft_model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.SpeculativeInferencer.draft_config">
+<span class="sig-name descname"><span class="pre">draft_config</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#SpeculativeInferencer.draft_config"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.draft_config" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.SpeculativeInferencer.score_to_prob">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">score_to_prob</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">scores</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">temperature</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">top_p</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1.0</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">torch.Tensor</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#SpeculativeInferencer.score_to_prob"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.score_to_prob" title="Link to this definition">#</a></dt>
+<dd><p>Convert scores (NOT softmaxed tensor) to probabilities with support for temperature, top-p sampling, and argmax.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>scores</strong><span class="classifier">torch.Tensor</span></dt><dd><p>Input scores.</p>
+</dd>
+<dt><strong>temperature</strong><span class="classifier">float, optional</span></dt><dd><p>Temperature parameter for controlling randomness. Higher values make the distribution more uniform,
+lower values make it peakier. When temperature &lt;= 1e-6, argmax is used. by default 0.0</p>
+</dd>
+<dt><strong>top_p</strong><span class="classifier">float, optional</span></dt><dd><p>Top-p sampling parameter for controlling the cumulative probability threshold, by default 1.0 (no threshold)</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>torch.Tensor</dt><dd><p>Probability distribution after adjustments.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.SpeculativeInferencer.sample">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">sample</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">prob</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_samples</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#SpeculativeInferencer.sample"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.sample" title="Link to this definition">#</a></dt>
+<dd><p>Sample from a tensor of probabilities</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.SpeculativeInferencer.predict_next_token">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">predict_next_token</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel" title="lmflow.models.hf_decoder_model.HFDecoderModel"><span class="pre">lmflow.models.hf_decoder_model.HFDecoderModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">input_ids</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_new_tokens</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#SpeculativeInferencer.predict_next_token"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.predict_next_token" title="Link to this definition">#</a></dt>
+<dd><p>Predict the next token given the input_ids.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.SpeculativeInferencer.autoregressive_sampling">
+<span class="sig-name descname"><span class="pre">autoregressive_sampling</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">input_ids</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel" title="lmflow.models.hf_decoder_model.HFDecoderModel"><span class="pre">lmflow.models.hf_decoder_model.HFDecoderModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">temperature</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_new_tokens</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">5</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#SpeculativeInferencer.autoregressive_sampling"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.autoregressive_sampling" title="Link to this definition">#</a></dt>
+<dd><p>Ref: [arXiv:2211.17192v2](https://arxiv.org/abs/2211.17192) Section 2.2</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.SpeculativeInferencer.inference">
+<span class="sig-name descname"><span class="pre">inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel" title="lmflow.models.hf_decoder_model.HFDecoderModel"><span class="pre">lmflow.models.hf_decoder_model.HFDecoderModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">draft_model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel" title="lmflow.models.hf_decoder_model.HFDecoderModel"><span class="pre">lmflow.models.hf_decoder_model.HFDecoderModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">temperature</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">gamma</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">5</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_new_tokens</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">100</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#SpeculativeInferencer.inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.inference" title="Link to this definition">#</a></dt>
+<dd><p>Perform inference for a model</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model</strong><span class="classifier">HFDecoderModel object.</span></dt><dd><p>TunableModel to verify tokens generated by the draft model.</p>
+</dd>
+<dt><strong>draft_model</strong><span class="classifier">HFDecoderModel object.</span></dt><dd><p>TunableModel that provides approximations of the target model.</p>
+</dd>
+<dt><strong>input</strong><span class="classifier">str.</span></dt><dd><p>The input text (i.e., the prompt) for the model.</p>
+</dd>
+<dt><strong>gamma</strong><span class="classifier">int.</span></dt><dd><p>The number of tokens to be generated by the draft model within each iter.</p>
+</dd>
+<dt><strong>max_new_tokens</strong><span class="classifier">int.</span></dt><dd><p>The maximum number of tokens to be generated by the target model.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>output: str.</dt><dd><p>The output text generated by the model.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.SpeculativeInferencer.stream_inference">
+<em class="property"><span class="pre">abstract</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">stream_inference</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#SpeculativeInferencer.stream_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.stream_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.ToolInferencer">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.inferencer.</span></span><span class="sig-name descname"><span class="pre">ToolInferencer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inferencer_args</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#ToolInferencer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.ToolInferencer" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.pipeline.inferencer.Inferencer" title="lmflow.pipeline.inferencer.Inferencer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Inferencer</span></code></a></p>
+<p>Initializes the <cite>ToolInferencer</cite> class with given arguments.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model_args</strong><span class="classifier">ModelArguments object.</span></dt><dd><p>Contains the arguments required to load the model.</p>
+</dd>
+<dt><strong>data_args</strong><span class="classifier">DatasetArguments object.</span></dt><dd><p>Contains the arguments required to load the dataset.</p>
+</dd>
+<dt><strong>inferencer_args</strong><span class="classifier">InferencerArguments object.</span></dt><dd><p>Contains the arguments required to perform inference.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.ToolInferencer.model">
+<span class="sig-name descname"><span class="pre">model</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#ToolInferencer.model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.ToolInferencer.model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.ToolInferencer.inference">
+<span class="sig-name descname"><span class="pre">inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel" title="lmflow.models.hf_decoder_model.HFDecoderModel"><span class="pre">lmflow.models.hf_decoder_model.HFDecoderModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_new_tokens</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1024</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#ToolInferencer.inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.ToolInferencer.inference" title="Link to this definition">#</a></dt>
+<dd><p>Perform inference for a model</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model</strong><span class="classifier">HFDecoderModel object.</span></dt><dd><p>TunableModel to perform inference</p>
+</dd>
+<dt><strong>input</strong><span class="classifier">str.</span></dt><dd><p>The input text (i.e., the prompt) for the model.</p>
+</dd>
+<dt><strong>max_new_tokens</strong><span class="classifier">int.</span></dt><dd><p>The maximum number of tokens to be generated by the model.</p>
+</dd>
+<dt><strong>Returns:</strong></dt><dd></dd>
+<dt><strong>output</strong><span class="classifier">str.</span></dt><dd><p>The output text generated by the model.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.inferencer.ToolInferencer.code_exec">
+<span class="sig-name descname"><span class="pre">code_exec</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">code</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/inferencer.html#ToolInferencer.code_exec"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.inferencer.ToolInferencer.code_exec" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../finetuner/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.finetuner</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../iterative_dpo_aligner/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.iterative_dpo_aligner</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.rstrip_partial_utf8"><code class="docutils literal notranslate"><span class="pre">rstrip_partial_utf8()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.supported_dataset_type"><code class="docutils literal notranslate"><span class="pre">supported_dataset_type</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.Inferencer"><code class="docutils literal notranslate"><span class="pre">Inferencer</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.Inferencer.data_args"><code class="docutils literal notranslate"><span class="pre">Inferencer.data_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.Inferencer.inferencer_args"><code class="docutils literal notranslate"><span class="pre">Inferencer.inferencer_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.Inferencer.model_args"><code class="docutils literal notranslate"><span class="pre">Inferencer.model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.Inferencer.local_rank"><code class="docutils literal notranslate"><span class="pre">Inferencer.local_rank</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.Inferencer.world_size"><code class="docutils literal notranslate"><span class="pre">Inferencer.world_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.Inferencer.config"><code class="docutils literal notranslate"><span class="pre">Inferencer.config</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.Inferencer.create_dataloader"><code class="docutils literal notranslate"><span class="pre">Inferencer.create_dataloader()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.Inferencer.inference"><code class="docutils literal notranslate"><span class="pre">Inferencer.inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.Inferencer.stream_inference"><code class="docutils literal notranslate"><span class="pre">Inferencer.stream_inference()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.SpeculativeInferencer"><code class="docutils literal notranslate"><span class="pre">SpeculativeInferencer</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.draft_model_args"><code class="docutils literal notranslate"><span class="pre">SpeculativeInferencer.draft_model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.draft_config"><code class="docutils literal notranslate"><span class="pre">SpeculativeInferencer.draft_config</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.score_to_prob"><code class="docutils literal notranslate"><span class="pre">SpeculativeInferencer.score_to_prob()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.sample"><code class="docutils literal notranslate"><span class="pre">SpeculativeInferencer.sample()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.predict_next_token"><code class="docutils literal notranslate"><span class="pre">SpeculativeInferencer.predict_next_token()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.autoregressive_sampling"><code class="docutils literal notranslate"><span class="pre">SpeculativeInferencer.autoregressive_sampling()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.inference"><code class="docutils literal notranslate"><span class="pre">SpeculativeInferencer.inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.SpeculativeInferencer.stream_inference"><code class="docutils literal notranslate"><span class="pre">SpeculativeInferencer.stream_inference()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.ToolInferencer"><code class="docutils literal notranslate"><span class="pre">ToolInferencer</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.ToolInferencer.model"><code class="docutils literal notranslate"><span class="pre">ToolInferencer.model</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.ToolInferencer.inference"><code class="docutils literal notranslate"><span class="pre">ToolInferencer.inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.inferencer.ToolInferencer.code_exec"><code class="docutils literal notranslate"><span class="pre">ToolInferencer.code_exec()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/inferencer/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html b/autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html
new file mode 100644
index 000000000..f5102eeb2
--- /dev/null
+++ b/autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html
@@ -0,0 +1,765 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.iterative_dpo_aligner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/iterative_dpo_aligner/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.raft_aligner" href="../raft_aligner/index.html" />
+    <link rel="prev" title="lmflow.pipeline.inferencer" href="../inferencer/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.iterative_dpo_aligner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.iterative_dpo_aligner">
+<span id="lmflow-pipeline-iterative-dpo-aligner"></span><h1>lmflow.pipeline.iterative_dpo_aligner<a class="headerlink" href="#module-lmflow.pipeline.iterative_dpo_aligner" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.iterative_dpo_aligner.logger" title="lmflow.pipeline.iterative_dpo_aligner.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner" title="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">IterativeDPOAligner</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.iterative_dpo_aligner.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.iterative_dpo_aligner.</span></span><span class="sig-name descname"><span class="pre">IterativeDPOAligner</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">aligner_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.IterativeDPOAlignerArguments" title="lmflow.args.IterativeDPOAlignerArguments"><span class="pre">lmflow.args.IterativeDPOAlignerArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">ref_model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">reward_model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner" title="Link to this definition">#</a></dt>
+<dd><dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.model_args">
+<span class="sig-name descname"><span class="pre">model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner.model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.data_args">
+<span class="sig-name descname"><span class="pre">data_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner.data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.data_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.aligner_args">
+<span class="sig-name descname"><span class="pre">aligner_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner.aligner_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.aligner_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.ref_model_args">
+<span class="sig-name descname"><span class="pre">ref_model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner.ref_model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.ref_model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.reward_model_args">
+<span class="sig-name descname"><span class="pre">reward_model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner.reward_model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.reward_model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.workspace_path">
+<span class="sig-name descname"><span class="pre">workspace_path</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner.workspace_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.workspace_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.align">
+<span class="sig-name descname"><span class="pre">align</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset_list</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner.align"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.align" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._align_single_iteration">
+<span class="sig-name descname"><span class="pre">_align_single_iteration</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">iteration_name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">reward_model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">ref_model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner._align_single_iteration"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._align_single_iteration" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_target_model_inference">
+<span class="sig-name descname"><span class="pre">_do_target_model_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel" title="lmflow.models.hf_decoder_model.HFDecoderModel"><span class="pre">lmflow.models.hf_decoder_model.HFDecoderModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_dir</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner._do_target_model_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_target_model_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_reward_model_inference">
+<span class="sig-name descname"><span class="pre">_do_reward_model_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel" title="lmflow.models.hf_text_regression_model.HFTextRegressionModel"><span class="pre">lmflow.models.hf_text_regression_model.HFTextRegressionModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_dir</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner._do_reward_model_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_reward_model_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_single_dpo_align">
+<span class="sig-name descname"><span class="pre">_do_single_dpo_align</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">ref_model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_dir</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">iteration_name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner._do_single_dpo_align"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_single_dpo_align" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_target_model_inference_args">
+<span class="sig-name descname"><span class="pre">_parse_target_model_inference_args</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.IterativeDPOAlignerArguments" title="lmflow.args.IterativeDPOAlignerArguments"><span class="pre">lmflow.args.IterativeDPOAlignerArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">result_cache_path</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference internal" href="../../args/index.html#lmflow.args.InferencerArguments" title="lmflow.args.InferencerArguments"><span class="pre">lmflow.args.InferencerArguments</span></a></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner._parse_target_model_inference_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_target_model_inference_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_reward_model_inference_args">
+<span class="sig-name descname"><span class="pre">_parse_reward_model_inference_args</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.IterativeDPOAlignerArguments" title="lmflow.args.IterativeDPOAlignerArguments"><span class="pre">lmflow.args.IterativeDPOAlignerArguments</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference internal" href="../../args/index.html#lmflow.args.InferencerArguments" title="lmflow.args.InferencerArguments"><span class="pre">lmflow.args.InferencerArguments</span></a></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner._parse_reward_model_inference_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_reward_model_inference_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_dpo_aligner_args">
+<span class="sig-name descname"><span class="pre">_parse_dpo_aligner_args</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.IterativeDPOAlignerArguments" title="lmflow.args.IterativeDPOAlignerArguments"><span class="pre">lmflow.args.IterativeDPOAlignerArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_dir</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">iteration_name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference internal" href="../../args/index.html#lmflow.args.DPOv2AlignerArguments" title="lmflow.args.DPOv2AlignerArguments"><span class="pre">lmflow.args.DPOv2AlignerArguments</span></a></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner._parse_dpo_aligner_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_dpo_aligner_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.__filter_args">
+<span class="sig-name descname"><span class="pre">__filter_args</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">mixed_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target_cls</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/iterative_dpo_aligner.html#IterativeDPOAligner.__filter_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.__filter_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../inferencer/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.inferencer</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../raft_aligner/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.raft_aligner</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.model_args"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner.model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.data_args"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner.data_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.aligner_args"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner.aligner_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.ref_model_args"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner.ref_model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.reward_model_args"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner.reward_model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.workspace_path"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner.workspace_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.align"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner.align()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._align_single_iteration"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner._align_single_iteration()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_target_model_inference"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner._do_target_model_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_reward_model_inference"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner._do_reward_model_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_single_dpo_align"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner._do_single_dpo_align()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_target_model_inference_args"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner._parse_target_model_inference_args()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_reward_model_inference_args"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner._parse_reward_model_inference_args()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_dpo_aligner_args"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner._parse_dpo_aligner_args()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.__filter_args"><code class="docutils literal notranslate"><span class="pre">IterativeDPOAligner.__filter_args()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/iterative_dpo_aligner/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/raft_aligner/index.html b/autoapi/lmflow/pipeline/raft_aligner/index.html
new file mode 100644
index 000000000..1801d01ba
--- /dev/null
+++ b/autoapi/lmflow/pipeline/raft_aligner/index.html
@@ -0,0 +1,806 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.raft_aligner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/raft_aligner/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.rm_inferencer" href="../rm_inferencer/index.html" />
+    <link rel="prev" title="lmflow.pipeline.iterative_dpo_aligner" href="../iterative_dpo_aligner/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.raft_aligner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.raft_aligner">
+<span id="lmflow-pipeline-raft-aligner"></span><h1>lmflow.pipeline.raft_aligner<a class="headerlink" href="#module-lmflow.pipeline.raft_aligner" title="Link to this heading">#</a></h1>
+<p>The Aligner class simplifies the process of running alignment.</p>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.raft_aligner.logger" title="lmflow.pipeline.raft_aligner.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.raft_aligner.RaftAligner" title="lmflow.pipeline.raft_aligner.RaftAligner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RaftAligner</span></code></a></p></td>
+<td><p>Initializes the <cite>RaftAligner</cite> class with given arguments.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.raft_aligner.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.RaftAligner">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.raft_aligner.</span></span><span class="sig-name descname"><span class="pre">RaftAligner</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">aligner_args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#RaftAligner"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.RaftAligner" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_aligner/index.html#lmflow.pipeline.base_aligner.BaseAligner" title="lmflow.pipeline.base_aligner.BaseAligner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.pipeline.base_aligner.BaseAligner</span></code></a></p>
+<p>Initializes the <cite>RaftAligner</cite> class with given arguments.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model_args</strong><span class="classifier">ModelArguments object.</span></dt><dd><p>Contains the arguments required to load the model.</p>
+</dd>
+<dt><strong>data_args</strong><span class="classifier">DatasetArguments object.</span></dt><dd><p>Contains the arguments required to load the dataset.</p>
+</dd>
+<dt><strong>raft_aligner_args</strong><span class="classifier">RaftAlignerArguments object.</span></dt><dd><p>Contains the arguments required to perform alignment.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.RaftAligner.model_args">
+<span class="sig-name descname"><span class="pre">model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#RaftAligner.model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.RaftAligner.model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.RaftAligner.data_args">
+<span class="sig-name descname"><span class="pre">data_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#RaftAligner.data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.RaftAligner.data_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.RaftAligner.aligner_args">
+<span class="sig-name descname"><span class="pre">aligner_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#RaftAligner.aligner_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.RaftAligner.aligner_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.RaftAligner.INF">
+<span class="sig-name descname"><span class="pre">INF</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">888888888</span></em><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#RaftAligner.INF"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.RaftAligner.INF" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.RaftAligner._initialize_trainer">
+<span class="sig-name descname"><span class="pre">_initialize_trainer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">training_args</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#RaftAligner._initialize_trainer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.RaftAligner._initialize_trainer" title="Link to this definition">#</a></dt>
+<dd><p>This function takes the model and tokenizer as the input and initialize the trainer.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.RaftAligner._load_dataset">
+<span class="sig-name descname"><span class="pre">_load_dataset</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">selected_dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">training_args</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#RaftAligner._load_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.RaftAligner._load_dataset" title="Link to this definition">#</a></dt>
+<dd><p>This function prepares the dataset for every iteration.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.RaftAligner._load_input_dataset">
+<span class="sig-name descname"><span class="pre">_load_input_dataset</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#RaftAligner._load_input_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.RaftAligner._load_input_dataset" title="Link to this definition">#</a></dt>
+<dd><p>Load input dataset (i.e. prompt/question dataset) for training.</p>
+<dl class="simple">
+<dt>Args:</dt><dd><dl class="simple">
+<dt>dataset: A Dataset object.</dt><dd><p>The dataset to be loaded.</p>
+</dd>
+</dl>
+</dd>
+<dt>Returns:</dt><dd><dl class="simple">
+<dt>dataloader (<cite>torch.utils.data.DataLoader</cite>):</dt><dd><p>The dataloader for the dataset.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.RaftAligner._clean_text">
+<span class="sig-name descname"><span class="pre">_clean_text</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">text</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#RaftAligner._clean_text"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.RaftAligner._clean_text" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.RaftAligner._discard_sample">
+<span class="sig-name descname"><span class="pre">_discard_sample</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">text</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#RaftAligner._discard_sample"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.RaftAligner._discard_sample" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.RaftAligner._get_batch_dataset_top">
+<span class="sig-name descname"><span class="pre">_get_batch_dataset_top</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">batch_input</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">alpha</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0.2</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">iter_id</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">local_rank</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_min_length</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">16</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_max_length</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">48</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">infer_batch_size</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">8</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">generation_kwargs</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">{}</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">training_args</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">reward_model</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_reward_path</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#RaftAligner._get_batch_dataset_top"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.RaftAligner._get_batch_dataset_top" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>batch_input</strong> – input prompts</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.RaftAligner._get_batch_dataset_local">
+<span class="sig-name descname"><span class="pre">_get_batch_dataset_local</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">batch_input</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">K</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">8</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">iter_id</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">local_rank</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_min_length</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">16</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_max_length</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">48</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">infer_batch_size</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">8</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">generation_kwargs</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">{}</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">training_args</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">reward_model</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_reward_path</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#RaftAligner._get_batch_dataset_local"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.RaftAligner._get_batch_dataset_local" title="Link to this definition">#</a></dt>
+<dd><dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><p><strong>batch_input</strong> – input prompts</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.raft_aligner.RaftAligner.align">
+<span class="sig-name descname"><span class="pre">align</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">reward_model</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/raft_aligner.html#RaftAligner.align"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.raft_aligner.RaftAligner.align" title="Link to this definition">#</a></dt>
+<dd><p>Perform alignment for a model</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model</strong><span class="classifier">BaseModel object.</span></dt><dd></dd>
+<dt><strong>dataset: Dataset object.</strong></dt><dd><dl class="simple">
+<dt>Input dataset for model to generate outputs. The input and output</dt><dd><p>will then be feed into reward model to get the reward for
+alignment.</p>
+</dd>
+</dl>
+</dd>
+<dt><strong>reward_model: RegressionModel object.</strong></dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../iterative_dpo_aligner/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.iterative_dpo_aligner</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../rm_inferencer/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.rm_inferencer</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.RaftAligner"><code class="docutils literal notranslate"><span class="pre">RaftAligner</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.RaftAligner.model_args"><code class="docutils literal notranslate"><span class="pre">RaftAligner.model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.RaftAligner.data_args"><code class="docutils literal notranslate"><span class="pre">RaftAligner.data_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.RaftAligner.aligner_args"><code class="docutils literal notranslate"><span class="pre">RaftAligner.aligner_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.RaftAligner.INF"><code class="docutils literal notranslate"><span class="pre">RaftAligner.INF</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.RaftAligner._initialize_trainer"><code class="docutils literal notranslate"><span class="pre">RaftAligner._initialize_trainer()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.RaftAligner._load_dataset"><code class="docutils literal notranslate"><span class="pre">RaftAligner._load_dataset()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.RaftAligner._load_input_dataset"><code class="docutils literal notranslate"><span class="pre">RaftAligner._load_input_dataset()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.RaftAligner._clean_text"><code class="docutils literal notranslate"><span class="pre">RaftAligner._clean_text()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.RaftAligner._discard_sample"><code class="docutils literal notranslate"><span class="pre">RaftAligner._discard_sample()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.RaftAligner._get_batch_dataset_top"><code class="docutils literal notranslate"><span class="pre">RaftAligner._get_batch_dataset_top()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.RaftAligner._get_batch_dataset_local"><code class="docutils literal notranslate"><span class="pre">RaftAligner._get_batch_dataset_local()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.raft_aligner.RaftAligner.align"><code class="docutils literal notranslate"><span class="pre">RaftAligner.align()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/raft_aligner/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/rm_inferencer/index.html b/autoapi/lmflow/pipeline/rm_inferencer/index.html
new file mode 100644
index 000000000..1c4cec930
--- /dev/null
+++ b/autoapi/lmflow/pipeline/rm_inferencer/index.html
@@ -0,0 +1,767 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.rm_inferencer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/rm_inferencer/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.rm_tuner" href="../rm_tuner/index.html" />
+    <link rel="prev" title="lmflow.pipeline.raft_aligner" href="../raft_aligner/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.rm_inferencer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.rm_inferencer">
+<span id="lmflow-pipeline-rm-inferencer"></span><h1>lmflow.pipeline.rm_inferencer<a class="headerlink" href="#module-lmflow.pipeline.rm_inferencer" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.rm_inferencer.logger" title="lmflow.pipeline.rm_inferencer.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer" title="lmflow.pipeline.rm_inferencer.RewardModelInferencer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RewardModelInferencer</span></code></a></p></td>
+<td><p>Initializes the <cite>Inferencer</cite> class with given arguments.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.rm_inferencer.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.rm_inferencer.</span></span><span class="sig-name descname"><span class="pre">RewardModelInferencer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">inferencer_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.InferencerArguments" title="lmflow.args.InferencerArguments"><span class="pre">lmflow.args.InferencerArguments</span></a></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_pipeline/index.html#lmflow.pipeline.base_pipeline.BasePipeline" title="lmflow.pipeline.base_pipeline.BasePipeline"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.pipeline.base_pipeline.BasePipeline</span></code></a></p>
+<p>Initializes the <cite>Inferencer</cite> class with given arguments.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model_args</strong><span class="classifier">ModelArguments object.</span></dt><dd><p>Contains the arguments required to load the model.</p>
+</dd>
+<dt><strong>data_args</strong><span class="classifier">DatasetArguments object.</span></dt><dd><p>Contains the arguments required to load the dataset.</p>
+</dd>
+<dt><strong>inferencer_args</strong><span class="classifier">InferencerArguments object.</span></dt><dd><p>Contains the arguments required to perform inference.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer.data_args">
+<span class="sig-name descname"><span class="pre">data_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer.data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.data_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer.inferencer_args">
+<span class="sig-name descname"><span class="pre">inferencer_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer.inferencer_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.inferencer_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer.model_args">
+<span class="sig-name descname"><span class="pre">model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer.model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer.local_rank">
+<span class="sig-name descname"><span class="pre">local_rank</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer.local_rank"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.local_rank" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer.world_size">
+<span class="sig-name descname"><span class="pre">world_size</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer.world_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.world_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer.inference">
+<span class="sig-name descname"><span class="pre">inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel" title="lmflow.models.hf_text_regression_model.HFTextRegressionModel"><span class="pre">lmflow.models.hf_text_regression_model.HFTextRegressionModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform_dataset_in_place</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_vllm</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">enable_distributed_inference</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer.inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer._inference">
+<span class="sig-name descname"><span class="pre">_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel" title="lmflow.models.hf_text_regression_model.HFTextRegressionModel"><span class="pre">lmflow.models.hf_text_regression_model.HFTextRegressionModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">ray.data.Dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">enable_distributed_inference</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer._inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer._inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer.__inference">
+<span class="sig-name descname"><span class="pre">__inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel" title="lmflow.models.hf_text_regression_model.HFTextRegressionModel"><span class="pre">lmflow.models.hf_text_regression_model.HFTextRegressionModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/dataset/index.html#lmflow.datasets.dataset.Dataset" title="lmflow.datasets.dataset.Dataset"><span class="pre">lmflow.datasets.dataset.Dataset</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer.__inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer.__distributed_inference">
+<span class="sig-name descname"><span class="pre">__distributed_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel" title="lmflow.models.hf_text_regression_model.HFTextRegressionModel"><span class="pre">lmflow.models.hf_text_regression_model.HFTextRegressionModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">ray.data.Dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_instances</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">batch_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="../../utils/data_utils/index.html#lmflow.utils.data_utils.RewardModelInferenceResultWithInput" title="lmflow.utils.data_utils.RewardModelInferenceResultWithInput"><span class="pre">lmflow.utils.data_utils.RewardModelInferenceResultWithInput</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer.__distributed_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__distributed_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer.__vllm_inference">
+<em class="property"><span class="pre">abstract</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">__vllm_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel" title="lmflow.models.hf_text_regression_model.HFTextRegressionModel"><span class="pre">lmflow.models.hf_text_regression_model.HFTextRegressionModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">enable_distributed_inference</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer.__vllm_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__vllm_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer.__post_process_model_output">
+<span class="sig-name descname"><span class="pre">__post_process_model_output</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_output</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.modeling_outputs.SequenceClassifierOutputWithPast</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer.__post_process_model_output"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__post_process_model_output" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer.flatten_list">
+<span class="sig-name descname"><span class="pre">flatten_list</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">list_of_list</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer.flatten_list"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.flatten_list" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_inferencer.RewardModelInferencer.compress_list">
+<span class="sig-name descname"><span class="pre">compress_list</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">list_to_compress</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sublist_lengths</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_inferencer.html#RewardModelInferencer.compress_list"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.compress_list" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../raft_aligner/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.raft_aligner</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../rm_tuner/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.rm_tuner</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.data_args"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer.data_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.inferencer_args"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer.inferencer_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.model_args"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer.model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.local_rank"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer.local_rank</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.world_size"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer.world_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.inference"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer.inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer._inference"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer._inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__inference"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer.__inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__distributed_inference"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer.__distributed_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__vllm_inference"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer.__vllm_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__post_process_model_output"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer.__post_process_model_output()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.flatten_list"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer.flatten_list()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_inferencer.RewardModelInferencer.compress_list"><code class="docutils literal notranslate"><span class="pre">RewardModelInferencer.compress_list()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/rm_inferencer/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/rm_tuner/index.html b/autoapi/lmflow/pipeline/rm_tuner/index.html
new file mode 100644
index 000000000..d1d3744d3
--- /dev/null
+++ b/autoapi/lmflow/pipeline/rm_tuner/index.html
@@ -0,0 +1,710 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.rm_tuner &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/rm_tuner/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.utils" href="../utils/index.html" />
+    <link rel="prev" title="lmflow.pipeline.rm_inferencer" href="../rm_inferencer/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.rm_tuner</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.rm_tuner">
+<span id="lmflow-pipeline-rm-tuner"></span><h1>lmflow.pipeline.rm_tuner<a class="headerlink" href="#module-lmflow.pipeline.rm_tuner" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.rm_tuner.logger" title="lmflow.pipeline.rm_tuner.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.rm_tuner.RewardModelTuner" title="lmflow.pipeline.rm_tuner.RewardModelTuner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RewardModelTuner</span></code></a></p></td>
+<td><p>Initializes the <cite>RewardModelTuner</cite> class.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_tuner.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.rm_tuner.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_tuner.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_tuner.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_tuner.RewardModelTuner">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.rm_tuner.</span></span><span class="sig-name descname"><span class="pre">RewardModelTuner</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">finetuner_args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_tuner.html#RewardModelTuner"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_tuner.RewardModelTuner" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../finetuner/index.html#lmflow.pipeline.finetuner.Finetuner" title="lmflow.pipeline.finetuner.Finetuner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.pipeline.finetuner.Finetuner</span></code></a></p>
+<p>Initializes the <cite>RewardModelTuner</cite> class.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model_args</strong><span class="classifier">ModelArguments object.</span></dt><dd><p>Contains the arguments required to load the model.</p>
+</dd>
+<dt><strong>data_args</strong><span class="classifier">DatasetArguments object.</span></dt><dd><p>Contains the arguments required to load the dataset.</p>
+</dd>
+<dt><strong>finetuner_args</strong><span class="classifier">RewardModelTunerArguments object.</span></dt><dd><p>Contains the arguments required to perform finetuning.</p>
+</dd>
+<dt><strong>args</strong><span class="classifier">Optional.</span></dt><dd><p>Positional arguments.</p>
+</dd>
+<dt><strong>kwargs</strong><span class="classifier">Optional.</span></dt><dd><p>Keyword arguments.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.rm_tuner.RewardModelTuner.tune">
+<span class="sig-name descname"><span class="pre">tune</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel" title="lmflow.models.hf_text_regression_model.HFTextRegressionModel"><span class="pre">lmflow.models.hf_text_regression_model.HFTextRegressionModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">transform_dataset_in_place</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_collator</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/rm_tuner.html#RewardModelTuner.tune"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.rm_tuner.RewardModelTuner.tune" title="Link to this definition">#</a></dt>
+<dd><p>Perform tuning for a model</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model</strong><span class="classifier">TunableModel object.</span></dt><dd><p>TunableModel to perform tuning.</p>
+</dd>
+<dt><strong>dataset:</strong></dt><dd><p>dataset to train model.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../rm_inferencer/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.rm_inferencer</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../utils/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.utils</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_tuner.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_tuner.RewardModelTuner"><code class="docutils literal notranslate"><span class="pre">RewardModelTuner</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.rm_tuner.RewardModelTuner.tune"><code class="docutils literal notranslate"><span class="pre">RewardModelTuner.tune()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/rm_tuner/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html b/autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html
new file mode 100644
index 000000000..90d251a0d
--- /dev/null
+++ b/autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html
@@ -0,0 +1,771 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.utils.dpov2_dataprocessor &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.utils.dpov2_trainer" href="../dpov2_trainer/index.html" />
+    <link rel="prev" title="lmflow.pipeline.utils" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.dpov2_dataprocessor</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.utils.dpov2_dataprocessor">
+<span id="lmflow-pipeline-utils-dpov2-dataprocessor"></span><h1>lmflow.pipeline.utils.dpov2_dataprocessor<a class="headerlink" href="#module-lmflow.pipeline.utils.dpov2_dataprocessor" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.dpov2_dataprocessor.logger" title="lmflow.pipeline.utils.dpov2_dataprocessor.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding" title="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.dpov2_dataprocessor.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.dpov2_dataprocessor.</span></span><span class="sig-name descname"><span class="pre">PreferenceDataCollatorWithPadding</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding" title="Link to this definition">#</a></dt>
+<dd><dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.tokenizer">
+<span class="sig-name descname"><span class="pre">tokenizer</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">transformers.PreTrainedTokenizerBase</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.tokenizer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.tokenizer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.model">
+<span class="sig-name descname"><span class="pre">model</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">transformers.PreTrainedModel</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.padding">
+<span class="sig-name descname"><span class="pre">padding</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.padding"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.padding" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_length">
+<span class="sig-name descname"><span class="pre">max_length</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.max_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_length" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_prompt_length">
+<span class="sig-name descname"><span class="pre">max_prompt_length</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.max_prompt_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_prompt_length" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.label_pad_token_id">
+<span class="sig-name descname"><span class="pre">label_pad_token_id</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">-100</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.label_pad_token_id"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.label_pad_token_id" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.padding_value">
+<span class="sig-name descname"><span class="pre">padding_value</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.padding_value"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.padding_value" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.truncation_mode">
+<span class="sig-name descname"><span class="pre">truncation_mode</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'keep_end'</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.truncation_mode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.truncation_mode" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.is_encoder_decoder">
+<span class="sig-name descname"><span class="pre">is_encoder_decoder</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.is_encoder_decoder"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.is_encoder_decoder" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_target_length">
+<span class="sig-name descname"><span class="pre">max_target_length</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.max_target_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_target_length" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.mask_prompt">
+<span class="sig-name descname"><span class="pre">mask_prompt</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.mask_prompt"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.mask_prompt" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.tokenize_batch_element">
+<span class="sig-name descname"><span class="pre">tokenize_batch_element</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">prompt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">chosen</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">rejected</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.tokenize_batch_element"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.tokenize_batch_element" title="Link to this definition">#</a></dt>
+<dd><p>Tokenize a single batch element.</p>
+<dl class="simple">
+<dt>At this stage, we don’t convert to PyTorch tensors yet; we just handle the truncation</dt><dd><p>in case the prompt + chosen or prompt + rejected responses is/are too long. First
+we truncate the prompt; if we’re still too long, we truncate the chosen/rejected.</p>
+</dd>
+<dt>We also create the labels for the chosen/rejected responses, which are of length equal to</dt><dd><p>the sum of the length of the prompt and the chosen/rejected response, with
+label_pad_token_id  for the prompt tokens.</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.collate">
+<span class="sig-name descname"><span class="pre">collate</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">batch</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.collate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.collate" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.__call__">
+<span class="sig-name descname"><span class="pre">__call__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">features</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_dataprocessor.html#PreferenceDataCollatorWithPadding.__call__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.__call__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.utils</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../dpov2_trainer/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.dpov2_trainer</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.tokenizer"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.tokenizer</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.model"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.model</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.padding"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.padding</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_length"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.max_length</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_prompt_length"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.max_prompt_length</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.label_pad_token_id"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.label_pad_token_id</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.padding_value"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.padding_value</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.truncation_mode"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.truncation_mode</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.is_encoder_decoder"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.is_encoder_decoder</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_target_length"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.max_target_length</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.mask_prompt"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.mask_prompt</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.tokenize_batch_element"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.tokenize_batch_element()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.collate"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.collate()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.__call__"><code class="docutils literal notranslate"><span class="pre">PreferenceDataCollatorWithPadding.__call__()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html b/autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html
new file mode 100644
index 000000000..bb9184454
--- /dev/null
+++ b/autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html
@@ -0,0 +1,724 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.utils.dpov2_trainer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/utils/dpov2_trainer/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.utils.memory_safe_dpov2_align" href="../memory_safe_dpov2_align/index.html" />
+    <link rel="prev" title="lmflow.pipeline.utils.dpov2_dataprocessor" href="../dpov2_dataprocessor/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.dpov2_trainer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.utils.dpov2_trainer">
+<span id="lmflow-pipeline-utils-dpov2-trainer"></span><h1>lmflow.pipeline.utils.dpov2_trainer<a class="headerlink" href="#module-lmflow.pipeline.utils.dpov2_trainer" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.dpov2_trainer.logger" title="lmflow.pipeline.utils.dpov2_trainer.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer" title="lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DPOv2Trainer</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_trainer.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.dpov2_trainer.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_trainer.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_trainer.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.dpov2_trainer.</span></span><span class="sig-name descname"><span class="pre">DPOv2Trainer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedModel</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">torch.nn.Module</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ref_model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedModel</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">torch.nn.Module</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">beta</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0.1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">loss_type</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Literal</span><span class="p"><span class="pre">[</span></span><span class="s"><span class="pre">'sigmoid'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'hinge'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'cross_entropy'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'kl'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'rev_kl'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'raft'</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'rev_kl'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.TrainingArguments</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_collator</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.DataCollator</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">label_pad_token_id</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">-100</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_value</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">truncation_mode</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'keep_end'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">train_dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../../datasets/index.html#lmflow.datasets.Dataset" title="lmflow.datasets.Dataset"><span class="pre">datasets.Dataset</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eval_dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../../datasets/index.html#lmflow.datasets.Dataset" title="lmflow.datasets.Dataset"><span class="pre">datasets.Dataset</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><a class="reference internal" href="../../../datasets/index.html#lmflow.datasets.Dataset" title="lmflow.datasets.Dataset"><span class="pre">datasets.Dataset</span></a><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizerBase</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_init</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Callable</span><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">transformers.PreTrainedModel</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">callbacks</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">transformers.trainer_callback.TrainerCallback</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">optimizers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.optim.Optimizer</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.optim.lr_scheduler.LambdaLR</span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(None,</span> <span class="pre">None)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">preprocess_logits_for_metrics</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Callable</span><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_length</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_prompt_length</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_target_length</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">peft_config</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">is_encoder_decoder</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">disable_dropout</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">generate_during_eval</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">compute_metrics</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Callable</span><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><span class="pre">transformers.trainer_utils.EvalLoopOutput</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">Dict</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">mask_prompt</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">len_penalty</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">preprocessing_num_workers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">1</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_trainer.html#DPOv2Trainer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">trl.DPOTrainer</span></code></p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.use_dpo_data_collator">
+<span class="sig-name descname"><span class="pre">use_dpo_data_collator</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_trainer.html#DPOv2Trainer.use_dpo_data_collator"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.use_dpo_data_collator" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.len_penalty">
+<span class="sig-name descname"><span class="pre">len_penalty</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_trainer.html#DPOv2Trainer.len_penalty"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.len_penalty" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.dpo_loss">
+<span class="sig-name descname"><span class="pre">dpo_loss</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">policy_chosen_logps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.FloatTensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">policy_rejected_logps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.FloatTensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">reference_chosen_logps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.FloatTensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">reference_rejected_logps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.FloatTensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">reference_free</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">margin</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.FloatTensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">len_penalty</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">float</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">0</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.FloatTensor</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.FloatTensor</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.FloatTensor</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_trainer.html#DPOv2Trainer.dpo_loss"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.dpo_loss" title="Link to this definition">#</a></dt>
+<dd><p>Compute the DPO loss for a batch of policy and reference model log probabilities.</p>
+<dl class="simple">
+<dt>Args:</dt><dd><p>policy_chosen_logps: Log probabilities of the policy model for the chosen responses. Shape: (batch_size,)
+policy_rejected_logps: Log probabilities of the policy model for the rejected responses. Shape: (batch_size,)
+reference_chosen_logps: Log probabilities of the reference model for the chosen responses. Shape: (batch_size,)
+reference_rejected_logps: Log probabilities of the reference model for the rejected responses. Shape: (batch_size,)
+beta: Temperature parameter for the DPO loss, typically something in the range of 0.1 to 0.5. We ignore the reference model as beta -&gt; 0.
+reference_free: If True, we ignore the _provided_ reference model and implicitly use a reference model that assigns equal probability to all responses.</p>
+</dd>
+<dt>Returns:</dt><dd><p>A tuple of three tensors: (losses, chosen_rewards, rejected_rewards).
+The losses tensor contains the DPO loss for each example in the batch.
+The chosen_rewards and rejected_rewards tensors contain the rewards for the chosen and rejected responses, respectively.</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.get_batch_loss_metrics">
+<span class="sig-name descname"><span class="pre">get_batch_loss_metrics</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">batch</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">torch.LongTensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">train_eval</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Literal</span><span class="p"><span class="pre">[</span></span><span class="s"><span class="pre">'train'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'eval'</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_trainer.html#DPOv2Trainer.get_batch_loss_metrics"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.get_batch_loss_metrics" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.get_batch_metrics">
+<span class="sig-name descname"><span class="pre">get_batch_metrics</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">batch</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">torch.LongTensor</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">train_eval</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Literal</span><span class="p"><span class="pre">[</span></span><span class="s"><span class="pre">'train'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'eval'</span></span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'train'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/dpov2_trainer.html#DPOv2Trainer.get_batch_metrics"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.get_batch_metrics" title="Link to this definition">#</a></dt>
+<dd><p>Compute the DPO loss and other metrics for the given batch of inputs for train or test.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../dpov2_dataprocessor/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.dpov2_dataprocessor</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../memory_safe_dpov2_align/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.memory_safe_dpov2_align</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_trainer.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer"><code class="docutils literal notranslate"><span class="pre">DPOv2Trainer</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.use_dpo_data_collator"><code class="docutils literal notranslate"><span class="pre">DPOv2Trainer.use_dpo_data_collator</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.len_penalty"><code class="docutils literal notranslate"><span class="pre">DPOv2Trainer.len_penalty</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.dpo_loss"><code class="docutils literal notranslate"><span class="pre">DPOv2Trainer.dpo_loss()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.get_batch_loss_metrics"><code class="docutils literal notranslate"><span class="pre">DPOv2Trainer.get_batch_loss_metrics()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.get_batch_metrics"><code class="docutils literal notranslate"><span class="pre">DPOv2Trainer.get_batch_metrics()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/pipeline/utils/dpov2_trainer/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/utils/index.html b/autoapi/lmflow/pipeline/utils/index.html
new file mode 100644
index 000000000..8ec3584a6
--- /dev/null
+++ b/autoapi/lmflow/pipeline/utils/index.html
@@ -0,0 +1,647 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.utils &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/utils/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.utils.dpov2_dataprocessor" href="dpov2_dataprocessor/index.html" />
+    <link rel="prev" title="lmflow.pipeline.rm_tuner" href="../rm_tuner/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 current active has-children"><a class="current reference internal" href="#">lmflow.pipeline.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.utils">
+<span id="lmflow-pipeline-utils"></span><h1>lmflow.pipeline.utils<a class="headerlink" href="#module-lmflow.pipeline.utils" title="Link to this heading">#</a></h1>
+<section id="submodules">
+<h2>Submodules<a class="headerlink" href="#submodules" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l1"><a class="reference internal" href="dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l1"><a class="reference internal" href="memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l1"><a class="reference internal" href="memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l1"><a class="reference internal" href="peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l1"><a class="reference internal" href="raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l1"><a class="reference internal" href="rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l1"><a class="reference internal" href="rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../rm_tuner/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.rm_tuner</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="dpov2_dataprocessor/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.dpov2_dataprocessor</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#submodules">Submodules</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/utils/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.html b/autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.html
new file mode 100644
index 000000000..0abee9a5b
--- /dev/null
+++ b/autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.html
@@ -0,0 +1,685 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.utils.memory_safe_dpov2_align &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.utils.memory_safe_vllm_inference" href="../memory_safe_vllm_inference/index.html" />
+    <link rel="prev" title="lmflow.pipeline.utils.dpov2_trainer" href="../dpov2_trainer/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.memory_safe_dpov2_align</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.utils.memory_safe_dpov2_align">
+<span id="lmflow-pipeline-utils-memory-safe-dpov2-align"></span><h1>lmflow.pipeline.utils.memory_safe_dpov2_align<a class="headerlink" href="#module-lmflow.pipeline.utils.memory_safe_dpov2_align" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.memory_safe_dpov2_align.logger" title="lmflow.pipeline.utils.memory_safe_dpov2_align.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.memory_safe_dpov2_align.ReferenceModelArguments" title="lmflow.pipeline.utils.memory_safe_dpov2_align.ReferenceModelArguments"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ReferenceModelArguments</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.memory_safe_dpov2_align.main" title="lmflow.pipeline.utils.memory_safe_dpov2_align.main"><code class="xref py py-obj docutils literal notranslate"><span class="pre">main</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.memory_safe_dpov2_align.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.memory_safe_dpov2_align.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/memory_safe_dpov2_align.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.memory_safe_dpov2_align.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.memory_safe_dpov2_align.ReferenceModelArguments">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.memory_safe_dpov2_align.</span></span><span class="sig-name descname"><span class="pre">ReferenceModelArguments</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="../../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/memory_safe_dpov2_align.html#ReferenceModelArguments"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.memory_safe_dpov2_align.ReferenceModelArguments" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.memory_safe_dpov2_align.main">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.memory_safe_dpov2_align.</span></span><span class="sig-name descname"><span class="pre">main</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/memory_safe_dpov2_align.html#main"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.memory_safe_dpov2_align.main" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../dpov2_trainer/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.dpov2_trainer</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../memory_safe_vllm_inference/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.memory_safe_vllm_inference</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.memory_safe_dpov2_align.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.memory_safe_dpov2_align.ReferenceModelArguments"><code class="docutils literal notranslate"><span class="pre">ReferenceModelArguments</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.memory_safe_dpov2_align.main"><code class="docutils literal notranslate"><span class="pre">main()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.html b/autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.html
new file mode 100644
index 000000000..133992d0e
--- /dev/null
+++ b/autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.html
@@ -0,0 +1,676 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.utils.memory_safe_vllm_inference &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.utils.peft_trainer" href="../peft_trainer/index.html" />
+    <link rel="prev" title="lmflow.pipeline.utils.memory_safe_dpov2_align" href="../memory_safe_dpov2_align/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.memory_safe_vllm_inference</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.utils.memory_safe_vllm_inference">
+<span id="lmflow-pipeline-utils-memory-safe-vllm-inference"></span><h1>lmflow.pipeline.utils.memory_safe_vllm_inference<a class="headerlink" href="#module-lmflow.pipeline.utils.memory_safe_vllm_inference" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.memory_safe_vllm_inference.logger" title="lmflow.pipeline.utils.memory_safe_vllm_inference.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.memory_safe_vllm_inference.main" title="lmflow.pipeline.utils.memory_safe_vllm_inference.main"><code class="xref py py-obj docutils literal notranslate"><span class="pre">main</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.memory_safe_vllm_inference.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.memory_safe_vllm_inference.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/memory_safe_vllm_inference.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.memory_safe_vllm_inference.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.memory_safe_vllm_inference.main">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.memory_safe_vllm_inference.</span></span><span class="sig-name descname"><span class="pre">main</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/memory_safe_vllm_inference.html#main"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.memory_safe_vllm_inference.main" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../memory_safe_dpov2_align/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.memory_safe_dpov2_align</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../peft_trainer/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.peft_trainer</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.memory_safe_vllm_inference.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.memory_safe_vllm_inference.main"><code class="docutils literal notranslate"><span class="pre">main()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/utils/peft_trainer/index.html b/autoapi/lmflow/pipeline/utils/peft_trainer/index.html
new file mode 100644
index 000000000..12bc0a02f
--- /dev/null
+++ b/autoapi/lmflow/pipeline/utils/peft_trainer/index.html
@@ -0,0 +1,709 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.utils.peft_trainer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/utils/peft_trainer/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.utils.raft_trainer" href="../raft_trainer/index.html" />
+    <link rel="prev" title="lmflow.pipeline.utils.memory_safe_vllm_inference" href="../memory_safe_vllm_inference/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.peft_trainer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.utils.peft_trainer">
+<span id="lmflow-pipeline-utils-peft-trainer"></span><h1>lmflow.pipeline.utils.peft_trainer<a class="headerlink" href="#module-lmflow.pipeline.utils.peft_trainer" title="Link to this heading">#</a></h1>
+<p>Trainer for Peft models</p>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.peft_trainer.PeftTrainer" title="lmflow.pipeline.utils.peft_trainer.PeftTrainer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PeftTrainer</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback" title="lmflow.pipeline.utils.peft_trainer.PeftSavingCallback"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PeftSavingCallback</span></code></a></p></td>
+<td><p>Correctly save PEFT model and not full model</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.peft_trainer.PeftTrainer">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.peft_trainer.</span></span><span class="sig-name descname"><span class="pre">PeftTrainer</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/peft_trainer.html#PeftTrainer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.peft_trainer.PeftTrainer" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">transformers.Trainer</span></code></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.peft_trainer.PeftTrainer._save_checkpoint">
+<span class="sig-name descname"><span class="pre">_save_checkpoint</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">_</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trial</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metrics</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/peft_trainer.html#PeftTrainer._save_checkpoint"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.peft_trainer.PeftTrainer._save_checkpoint" title="Link to this definition">#</a></dt>
+<dd><p>Don’t save base model, optimizer etc.
+but create checkpoint folder (needed for saving adapter)</p>
+</dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.peft_trainer.PeftSavingCallback">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.peft_trainer.</span></span><span class="sig-name descname"><span class="pre">PeftSavingCallback</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/peft_trainer.html#PeftSavingCallback"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">transformers.trainer_callback.TrainerCallback</span></code></p>
+<p>Correctly save PEFT model and not full model</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.peft_trainer.PeftSavingCallback._save">
+<span class="sig-name descname"><span class="pre">_save</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">folder</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/peft_trainer.html#PeftSavingCallback._save"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback._save" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_train_end">
+<span class="sig-name descname"><span class="pre">on_train_end</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.training_args.TrainingArguments</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">state</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.trainer_callback.TrainerState</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">control</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.trainer_callback.TrainerControl</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/peft_trainer.html#PeftSavingCallback.on_train_end"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_train_end" title="Link to this definition">#</a></dt>
+<dd><p>Save final best model adapter</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_epoch_end">
+<span class="sig-name descname"><span class="pre">on_epoch_end</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.training_args.TrainingArguments</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">state</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.trainer_callback.TrainerState</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">control</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.trainer_callback.TrainerControl</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/peft_trainer.html#PeftSavingCallback.on_epoch_end"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_epoch_end" title="Link to this definition">#</a></dt>
+<dd><p>Save intermediate model adapters in case of interrupted training</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_save">
+<span class="sig-name descname"><span class="pre">on_save</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.training_args.TrainingArguments</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">state</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.trainer_callback.TrainerState</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">control</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.trainer_callback.TrainerControl</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/peft_trainer.html#PeftSavingCallback.on_save"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_save" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../memory_safe_vllm_inference/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.memory_safe_vllm_inference</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../raft_trainer/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.raft_trainer</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.peft_trainer.PeftTrainer"><code class="docutils literal notranslate"><span class="pre">PeftTrainer</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.peft_trainer.PeftTrainer._save_checkpoint"><code class="docutils literal notranslate"><span class="pre">PeftTrainer._save_checkpoint()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback"><code class="docutils literal notranslate"><span class="pre">PeftSavingCallback</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback._save"><code class="docutils literal notranslate"><span class="pre">PeftSavingCallback._save()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_train_end"><code class="docutils literal notranslate"><span class="pre">PeftSavingCallback.on_train_end()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_epoch_end"><code class="docutils literal notranslate"><span class="pre">PeftSavingCallback.on_epoch_end()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_save"><code class="docutils literal notranslate"><span class="pre">PeftSavingCallback.on_save()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/pipeline/utils/peft_trainer/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/utils/raft_trainer/index.html b/autoapi/lmflow/pipeline/utils/raft_trainer/index.html
new file mode 100644
index 000000000..474b4ed7a
--- /dev/null
+++ b/autoapi/lmflow/pipeline/utils/raft_trainer/index.html
@@ -0,0 +1,1796 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.utils.raft_trainer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/utils/raft_trainer/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.utils.rm_dataprocessor" href="../rm_dataprocessor/index.html" />
+    <link rel="prev" title="lmflow.pipeline.utils.peft_trainer" href="../peft_trainer/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.raft_trainer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.utils.raft_trainer">
+<span id="lmflow-pipeline-utils-raft-trainer"></span><h1>lmflow.pipeline.utils.raft_trainer<a class="headerlink" href="#module-lmflow.pipeline.utils.raft_trainer" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.raft_trainer.is_torch_greater_or_equal_than_1_10" title="lmflow.pipeline.utils.raft_trainer.is_torch_greater_or_equal_than_1_10"><code class="xref py py-obj docutils literal notranslate"><span class="pre">is_torch_greater_or_equal_than_1_10</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.raft_trainer.is_torch_less_than_1_11" title="lmflow.pipeline.utils.raft_trainer.is_torch_less_than_1_11"><code class="xref py py-obj docutils literal notranslate"><span class="pre">is_torch_less_than_1_11</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.raft_trainer._is_native_cpu_amp_available" title="lmflow.pipeline.utils.raft_trainer._is_native_cpu_amp_available"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_is_native_cpu_amp_available</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.raft_trainer.DEFAULT_CALLBACKS" title="lmflow.pipeline.utils.raft_trainer.DEFAULT_CALLBACKS"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DEFAULT_CALLBACKS</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#id0" title="lmflow.pipeline.utils.raft_trainer.DEFAULT_PROGRESS_CALLBACK"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DEFAULT_PROGRESS_CALLBACK</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#id0" title="lmflow.pipeline.utils.raft_trainer.DEFAULT_PROGRESS_CALLBACK"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DEFAULT_PROGRESS_CALLBACK</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.raft_trainer.IS_SAGEMAKER_MP_POST_1_10" title="lmflow.pipeline.utils.raft_trainer.IS_SAGEMAKER_MP_POST_1_10"><code class="xref py py-obj docutils literal notranslate"><span class="pre">IS_SAGEMAKER_MP_POST_1_10</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.raft_trainer.skip_first_batches" title="lmflow.pipeline.utils.raft_trainer.skip_first_batches"><code class="xref py py-obj docutils literal notranslate"><span class="pre">skip_first_batches</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.raft_trainer.logger" title="lmflow.pipeline.utils.raft_trainer.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.raft_trainer.TRAINING_ARGS_NAME" title="lmflow.pipeline.utils.raft_trainer.TRAINING_ARGS_NAME"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TRAINING_ARGS_NAME</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.raft_trainer.TRAINER_STATE_NAME" title="lmflow.pipeline.utils.raft_trainer.TRAINER_STATE_NAME"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TRAINER_STATE_NAME</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.raft_trainer.OPTIMIZER_NAME" title="lmflow.pipeline.utils.raft_trainer.OPTIMIZER_NAME"><code class="xref py py-obj docutils literal notranslate"><span class="pre">OPTIMIZER_NAME</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.raft_trainer.SCHEDULER_NAME" title="lmflow.pipeline.utils.raft_trainer.SCHEDULER_NAME"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SCHEDULER_NAME</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.raft_trainer.SCALER_NAME" title="lmflow.pipeline.utils.raft_trainer.SCALER_NAME"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SCALER_NAME</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer" title="lmflow.pipeline.utils.raft_trainer.RaftTrainer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RaftTrainer</span></code></a></p></td>
+<td><p>Trainer is a simple but feature-complete training and eval loop for PyTorch, optimized for 🤗 Transformers.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.is_torch_greater_or_equal_than_1_10">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">is_torch_greater_or_equal_than_1_10</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#is_torch_greater_or_equal_than_1_10"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.is_torch_greater_or_equal_than_1_10" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.is_torch_less_than_1_11">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">is_torch_less_than_1_11</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#is_torch_less_than_1_11"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.is_torch_less_than_1_11" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer._is_native_cpu_amp_available">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">_is_native_cpu_amp_available</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#_is_native_cpu_amp_available"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer._is_native_cpu_amp_available" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.DEFAULT_CALLBACKS">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">DEFAULT_CALLBACKS</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#DEFAULT_CALLBACKS"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.DEFAULT_CALLBACKS" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.DEFAULT_PROGRESS_CALLBACK">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">DEFAULT_PROGRESS_CALLBACK</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#DEFAULT_PROGRESS_CALLBACK"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.DEFAULT_PROGRESS_CALLBACK" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="id0">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">DEFAULT_PROGRESS_CALLBACK</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#DEFAULT_PROGRESS_CALLBACK"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#id0" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.IS_SAGEMAKER_MP_POST_1_10">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">IS_SAGEMAKER_MP_POST_1_10</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#IS_SAGEMAKER_MP_POST_1_10"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.IS_SAGEMAKER_MP_POST_1_10" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.skip_first_batches">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">skip_first_batches</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#skip_first_batches"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.skip_first_batches" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.TRAINING_ARGS_NAME">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">TRAINING_ARGS_NAME</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'training_args.bin'</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#TRAINING_ARGS_NAME"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.TRAINING_ARGS_NAME" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.TRAINER_STATE_NAME">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">TRAINER_STATE_NAME</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'trainer_state.json'</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#TRAINER_STATE_NAME"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.TRAINER_STATE_NAME" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.OPTIMIZER_NAME">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">OPTIMIZER_NAME</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'optimizer.pt'</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#OPTIMIZER_NAME"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.OPTIMIZER_NAME" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.SCHEDULER_NAME">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">SCHEDULER_NAME</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'scheduler.pt'</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#SCHEDULER_NAME"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.SCHEDULER_NAME" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.SCALER_NAME">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">SCALER_NAME</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'scaler.pt'</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#SCALER_NAME"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.SCALER_NAME" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.raft_trainer.</span></span><span class="sig-name descname"><span class="pre">RaftTrainer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.modeling_utils.PreTrainedModel</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">torch.nn.Module</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.training_args.TrainingArguments</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_collator</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.data.data_collator.DataCollator</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">train_dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.utils.data.Dataset</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">eval_dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.utils.data.Dataset</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.utils.data.Dataset</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.tokenization_utils_base.PreTrainedTokenizerBase</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_init</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Callable</span><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">transformers.modeling_utils.PreTrainedModel</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">compute_metrics</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Callable</span><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><span class="pre">transformers.trainer_utils.EvalPrediction</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">Dict</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">callbacks</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">transformers.trainer_callback.TrainerCallback</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">optimizers</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.optim.Optimizer</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.optim.lr_scheduler.LambdaLR</span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">(None,</span> <span class="pre">None)</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">preprocess_logits_for_metrics</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Callable</span><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer" title="Link to this definition">#</a></dt>
+<dd><p>Trainer is a simple but feature-complete training and eval loop for PyTorch, optimized for 🤗 Transformers.
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>model ([<cite>PreTrainedModel</cite>] or <cite>torch.nn.Module</cite>, <em>optional</em>):</dt><dd><p>The model to train, evaluate or use for predictions. If not provided, a <cite>model_init</cite> must be passed.
+&lt;Tip&gt;
+[<cite>Trainer</cite>] is optimized to work with the [<cite>PreTrainedModel</cite>] provided by the library. You can still use
+your own models defined as <cite>torch.nn.Module</cite> as long as they work the same way as the 🤗 Transformers
+models.
+&lt;/Tip&gt;</p>
+</dd>
+<dt>args ([<cite>TrainingArguments</cite>], <em>optional</em>):</dt><dd><p>The arguments to tweak for training. Will default to a basic instance of [<cite>TrainingArguments</cite>] with the
+<cite>output_dir</cite> set to a directory named <em>tmp_trainer</em> in the current directory if not provided.</p>
+</dd>
+<dt>data_collator (<cite>DataCollator</cite>, <em>optional</em>):</dt><dd><p>The function to use to form a batch from a list of elements of <cite>train_dataset</cite> or <cite>eval_dataset</cite>. Will
+default to [<cite>default_data_collator</cite>] if no <cite>tokenizer</cite> is provided, an instance of
+[<cite>DataCollatorWithPadding</cite>] otherwise.</p>
+</dd>
+<dt>train_dataset (<cite>torch.utils.data.Dataset</cite> or <cite>torch.utils.data.IterableDataset</cite>, <em>optional</em>):</dt><dd><p>The dataset to use for training. If it is a [<cite>~datasets.Dataset</cite>], columns not accepted by the
+<cite>model.forward()</cite> method are automatically removed.
+Note that if it’s a <cite>torch.utils.data.IterableDataset</cite> with some randomization and you are training in a
+distributed fashion, your iterable dataset should either use a internal attribute <cite>generator</cite> that is a
+<cite>torch.Generator</cite> for the randomization that must be identical on all processes (and the Trainer will
+manually set the seed of this <cite>generator</cite> at each epoch) or have a <cite>set_epoch()</cite> method that internally
+sets the seed of the RNGs used.</p>
+</dd>
+<dt>eval_dataset (Union[<cite>torch.utils.data.Dataset</cite>, Dict[str, <cite>torch.utils.data.Dataset</cite>]), <em>optional</em>):</dt><dd><p>The dataset to use for evaluation. If it is a [<cite>~datasets.Dataset</cite>], columns not accepted by the
+<cite>model.forward()</cite> method are automatically removed. If it is a dictionary, it will evaluate on each
+dataset prepending the dictionary key to the metric name.</p>
+</dd>
+<dt>tokenizer ([<cite>PreTrainedTokenizerBase</cite>], <em>optional</em>):</dt><dd><p>The tokenizer used to preprocess the data. If provided, will be used to automatically pad the inputs to the
+maximum length when batching inputs, and it will be saved along the model to make it easier to rerun an
+interrupted training or reuse the fine-tuned model.</p>
+</dd>
+<dt>model_init (<cite>Callable[[], PreTrainedModel]</cite>, <em>optional</em>):</dt><dd><p>A function that instantiates the model to be used. If provided, each call to [<cite>~Trainer.train</cite>] will start
+from a new instance of the model as given by this function.
+The function may have zero argument, or a single one containing the optuna/Ray Tune/SigOpt trial object, to
+be able to choose different architectures according to hyper parameters (such as layer count, sizes of
+inner layers, dropout probabilities etc).</p>
+</dd>
+<dt>compute_metrics (<cite>Callable[[EvalPrediction], Dict]</cite>, <em>optional</em>):</dt><dd><p>The function that will be used to compute metrics at evaluation. Must take a [<cite>EvalPrediction</cite>] and return
+a dictionary string to metric values.</p>
+</dd>
+<dt>callbacks (List of [<cite>TrainerCallback</cite>], <em>optional</em>):</dt><dd><p>A list of callbacks to customize the training loop. Will add those to the list of default callbacks
+detailed in [here](callback).
+If you want to remove one of the default callbacks used, use the [<cite>Trainer.remove_callback</cite>] method.</p>
+</dd>
+<dt>optimizers (<cite>Tuple[torch.optim.Optimizer, torch.optim.lr_scheduler.LambdaLR]</cite>, <em>optional</em>): A tuple</dt><dd><p>containing the optimizer and the scheduler to use. Will default to an instance of [<cite>AdamW</cite>] on your model
+and a scheduler given by [<cite>get_linear_schedule_with_warmup</cite>] controlled by <cite>args</cite>.</p>
+</dd>
+<dt>preprocess_logits_for_metrics (<cite>Callable[[torch.Tensor, torch.Tensor], torch.Tensor]</cite>, <em>optional</em>):</dt><dd><p>A function that preprocess the logits right before caching them at each evaluation step. Must take two
+tensors, the logits and the labels, and return the logits once processed as desired. The modifications made
+by this function will be reflected in the predictions received by <cite>compute_metrics</cite>.
+Note that the labels (second parameter) will be <cite>None</cite> if the dataset does not have them.</p>
+</dd>
+</dl>
+</div></blockquote>
+<dl class="simple">
+<dt>Important attributes:</dt><dd><ul class="simple">
+<li><p><strong>model</strong> – Always points to the core model. If using a transformers model, it will be a [<cite>PreTrainedModel</cite>]
+subclass.</p></li>
+<li><p><strong>model_wrapped</strong> – Always points to the most external model in case one or more other modules wrap the
+original model. This is the model that should be used for the forward pass. For example, under <cite>DeepSpeed</cite>,
+the inner model is wrapped in <cite>DeepSpeed</cite> and then again in <cite>torch.nn.DistributedDataParallel</cite>. If the inner
+model hasn’t been wrapped, then <cite>self.model_wrapped</cite> is the same as <cite>self.model</cite>.</p></li>
+<li><p><strong>is_model_parallel</strong> – Whether or not a model has been switched to a model parallel mode (different from
+data parallelism, this means some of the model layers are split on different GPUs).</p></li>
+<li><p><strong>place_model_on_device</strong> – Whether or not to automatically place the model on the device - it will be set
+to <cite>False</cite> if model parallel or deepspeed is used, or if the default
+<cite>TrainingArguments.place_model_on_device</cite> is overridden to return <cite>False</cite> .</p></li>
+<li><p><strong>is_in_train</strong> – Whether or not a model is currently running <cite>train</cite> (e.g. when <cite>evaluate</cite> is called while
+in <cite>train</cite>)</p></li>
+</ul>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.save_counter">
+<span class="sig-name descname"><span class="pre">save_counter</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.save_counter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.save_counter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.args">
+<span class="sig-name descname"><span class="pre">args</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.hp_name">
+<span class="sig-name descname"><span class="pre">hp_name</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.hp_name"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.hp_name" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.deepspeed">
+<span class="sig-name descname"><span class="pre">deepspeed</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.deepspeed"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.deepspeed" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_in_train">
+<span class="sig-name descname"><span class="pre">is_in_train</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.is_in_train"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_in_train" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._memory_tracker">
+<span class="sig-name descname"><span class="pre">_memory_tracker</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._memory_tracker"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._memory_tracker" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.sharded_ddp">
+<span class="sig-name descname"><span class="pre">sharded_ddp</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.sharded_ddp"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.sharded_ddp" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.fsdp">
+<span class="sig-name descname"><span class="pre">fsdp</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.fsdp"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.fsdp" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.place_model_on_device">
+<span class="sig-name descname"><span class="pre">place_model_on_device</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.place_model_on_device"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.place_model_on_device" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.data_collator">
+<span class="sig-name descname"><span class="pre">data_collator</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.data_collator"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.data_collator" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.train_dataset">
+<span class="sig-name descname"><span class="pre">train_dataset</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.train_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.train_dataset" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.eval_dataset">
+<span class="sig-name descname"><span class="pre">eval_dataset</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.eval_dataset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.eval_dataset" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.tokenizer">
+<span class="sig-name descname"><span class="pre">tokenizer</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.tokenizer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.tokenizer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.model_wrapped">
+<span class="sig-name descname"><span class="pre">model_wrapped</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.model_wrapped"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.model_wrapped" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.model">
+<span class="sig-name descname"><span class="pre">model</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_metrics">
+<span class="sig-name descname"><span class="pre">compute_metrics</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.compute_metrics"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_metrics" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.preprocess_logits_for_metrics">
+<span class="sig-name descname"><span class="pre">preprocess_logits_for_metrics</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.preprocess_logits_for_metrics"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.preprocess_logits_for_metrics" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.callback_handler">
+<span class="sig-name descname"><span class="pre">callback_handler</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.callback_handler"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.callback_handler" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._loggers_initialized">
+<span class="sig-name descname"><span class="pre">_loggers_initialized</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._loggers_initialized"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._loggers_initialized" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._signature_columns">
+<span class="sig-name descname"><span class="pre">_signature_columns</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._signature_columns"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._signature_columns" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_apex">
+<span class="sig-name descname"><span class="pre">use_apex</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.use_apex"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_apex" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_cuda_amp">
+<span class="sig-name descname"><span class="pre">use_cuda_amp</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.use_cuda_amp"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_cuda_amp" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_cpu_amp">
+<span class="sig-name descname"><span class="pre">use_cpu_amp</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.use_cpu_amp"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_cpu_amp" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.do_grad_scaling">
+<span class="sig-name descname"><span class="pre">do_grad_scaling</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.do_grad_scaling"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.do_grad_scaling" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.state">
+<span class="sig-name descname"><span class="pre">state</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.state"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.state" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.control">
+<span class="sig-name descname"><span class="pre">control</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.control"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.control" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.current_flos">
+<span class="sig-name descname"><span class="pre">current_flos</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">0</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.current_flos"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.current_flos" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.hp_search_backend">
+<span class="sig-name descname"><span class="pre">hp_search_backend</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.hp_search_backend"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.hp_search_backend" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_tune_checkpoints">
+<span class="sig-name descname"><span class="pre">use_tune_checkpoints</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.use_tune_checkpoints"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_tune_checkpoints" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.label_names">
+<span class="sig-name descname"><span class="pre">label_names</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.label_names"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.label_names" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.can_return_loss">
+<span class="sig-name descname"><span class="pre">can_return_loss</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.can_return_loss"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.can_return_loss" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._train_batch_size">
+<span class="sig-name descname"><span class="pre">_train_batch_size</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._train_batch_size"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._train_batch_size" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.add_callback">
+<span class="sig-name descname"><span class="pre">add_callback</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">callback</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.add_callback"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.add_callback" title="Link to this definition">#</a></dt>
+<dd><p>Add a callback to the current list of [<cite>~transformer.TrainerCallback</cite>].
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>callback (<cite>type</cite> or [<cite>~transformer.TrainerCallback</cite>]):</dt><dd><p>A [<cite>~transformer.TrainerCallback</cite>] class or an instance of a [<cite>~transformer.TrainerCallback</cite>]. In the
+first case, will instantiate a member of that class.</p>
+</dd>
+</dl>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.pop_callback">
+<span class="sig-name descname"><span class="pre">pop_callback</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">callback</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.pop_callback"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.pop_callback" title="Link to this definition">#</a></dt>
+<dd><p>Remove a callback from the current list of [<cite>~transformer.TrainerCallback</cite>] and returns it.
+If the callback is not found, returns <cite>None</cite> (and no error is raised).
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>callback (<cite>type</cite> or [<cite>~transformer.TrainerCallback</cite>]):</dt><dd><p>A [<cite>~transformer.TrainerCallback</cite>] class or an instance of a [<cite>~transformer.TrainerCallback</cite>]. In the
+first case, will pop the first member of that class found in the list of callbacks.</p>
+</dd>
+</dl>
+</div></blockquote>
+<dl class="simple">
+<dt>Returns:</dt><dd><p>[<cite>~transformer.TrainerCallback</cite>]: The callback removed, if found.</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.remove_callback">
+<span class="sig-name descname"><span class="pre">remove_callback</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">callback</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.remove_callback"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.remove_callback" title="Link to this definition">#</a></dt>
+<dd><p>Remove a callback from the current list of [<cite>~transformer.TrainerCallback</cite>].
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>callback (<cite>type</cite> or [<cite>~transformer.TrainerCallback</cite>]):</dt><dd><p>A [<cite>~transformer.TrainerCallback</cite>] class or an instance of a [<cite>~transformer.TrainerCallback</cite>]. In the
+first case, will remove the first member of that class found in the list of callbacks.</p>
+</dd>
+</dl>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._move_model_to_device">
+<span class="sig-name descname"><span class="pre">_move_model_to_device</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">device</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._move_model_to_device"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._move_model_to_device" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._set_signature_columns_if_needed">
+<span class="sig-name descname"><span class="pre">_set_signature_columns_if_needed</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._set_signature_columns_if_needed"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._set_signature_columns_if_needed" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._remove_unused_columns">
+<span class="sig-name descname"><span class="pre">_remove_unused_columns</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../../datasets/index.html#lmflow.datasets.Dataset" title="lmflow.datasets.Dataset"><span class="pre">datasets.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">description</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._remove_unused_columns"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._remove_unused_columns" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_collator_with_removed_columns">
+<span class="sig-name descname"><span class="pre">_get_collator_with_removed_columns</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data_collator</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Callable</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">description</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Callable</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._get_collator_with_removed_columns"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_collator_with_removed_columns" title="Link to this definition">#</a></dt>
+<dd><p>Wrap the data collator in a callable removing unused columns.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_train_sampler">
+<span class="sig-name descname"><span class="pre">_get_train_sampler</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">torch.utils.data.Sampler</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._get_train_sampler"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_train_sampler" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_train_dataloader">
+<span class="sig-name descname"><span class="pre">get_train_dataloader</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">torch.utils.data.DataLoader</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.get_train_dataloader"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_train_dataloader" title="Link to this definition">#</a></dt>
+<dd><p>Returns the training [<cite>~torch.utils.data.DataLoader</cite>].
+Will use no sampler if <cite>train_dataset</cite> does not implement <cite>__len__</cite>, a random sampler (adapted to distributed
+training if necessary) otherwise.
+Subclass and override this method if you want to inject some custom behavior.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_eval_sampler">
+<span class="sig-name descname"><span class="pre">_get_eval_sampler</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">eval_dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.utils.data.Dataset</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">torch.utils.data.Sampler</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._get_eval_sampler"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_eval_sampler" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_eval_dataloader">
+<span class="sig-name descname"><span class="pre">get_eval_dataloader</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">eval_dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.utils.data.Dataset</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">torch.utils.data.DataLoader</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.get_eval_dataloader"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_eval_dataloader" title="Link to this definition">#</a></dt>
+<dd><p>Returns the evaluation [<cite>~torch.utils.data.DataLoader</cite>].
+Subclass and override this method if you want to inject some custom behavior.
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>eval_dataset (<cite>torch.utils.data.Dataset</cite>, <em>optional</em>):</dt><dd><p>If provided, will override <cite>self.eval_dataset</cite>. If it is a [<cite>~datasets.Dataset</cite>], columns not accepted
+by the <cite>model.forward()</cite> method are automatically removed. It must implement <cite>__len__</cite>.</p>
+</dd>
+</dl>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_test_dataloader">
+<span class="sig-name descname"><span class="pre">get_test_dataloader</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">test_dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.utils.data.Dataset</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">torch.utils.data.DataLoader</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.get_test_dataloader"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_test_dataloader" title="Link to this definition">#</a></dt>
+<dd><p>Returns the test [<cite>~torch.utils.data.DataLoader</cite>].
+Subclass and override this method if you want to inject some custom behavior.
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>test_dataset (<cite>torch.utils.data.Dataset</cite>, <em>optional</em>):</dt><dd><p>The test dataset to use. If it is a [<cite>~datasets.Dataset</cite>], columns not accepted by the
+<cite>model.forward()</cite> method are automatically removed. It must implement <cite>__len__</cite>.</p>
+</dd>
+</dl>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_optimizer_and_scheduler">
+<span class="sig-name descname"><span class="pre">create_optimizer_and_scheduler</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">num_training_steps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.create_optimizer_and_scheduler"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_optimizer_and_scheduler" title="Link to this definition">#</a></dt>
+<dd><p>Setup the optimizer and the learning rate scheduler.
+We provide a reasonable default that works well. If you want to use something else, you can pass a tuple in the
+Trainer’s init through <cite>optimizers</cite>, or subclass and override this method (or <cite>create_optimizer</cite> and/or
+<cite>create_scheduler</cite>) in a subclass.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_optimizer">
+<span class="sig-name descname"><span class="pre">create_optimizer</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.create_optimizer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_optimizer" title="Link to this definition">#</a></dt>
+<dd><p>Setup the optimizer.
+We provide a reasonable default that works well. If you want to use something else, you can pass a tuple in the
+Trainer’s init through <cite>optimizers</cite>, or subclass and override this method in a subclass.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_optimizer_cls_and_kwargs">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">get_optimizer_cls_and_kwargs</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.training_args.TrainingArguments</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">Any</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.get_optimizer_cls_and_kwargs"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_optimizer_cls_and_kwargs" title="Link to this definition">#</a></dt>
+<dd><p>Returns the optimizer class and optimizer parameters based on the training arguments.
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>args (<cite>transformers.training_args.TrainingArguments</cite>):</dt><dd><p>The training arguments for the training session.</p>
+</dd>
+</dl>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_scheduler">
+<span class="sig-name descname"><span class="pre">create_scheduler</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">num_training_steps</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">optimizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.optim.Optimizer</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.create_scheduler"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_scheduler" title="Link to this definition">#</a></dt>
+<dd><p>Setup the scheduler. The optimizer of the trainer must have been set up either before this method is called or
+passed as an argument.
+Args:</p>
+<blockquote>
+<div><p>num_training_steps (int): The number of training steps to do.</p>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.num_examples">
+<span class="sig-name descname"><span class="pre">num_examples</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataloader</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.utils.data.DataLoader</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">int</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.num_examples"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.num_examples" title="Link to this definition">#</a></dt>
+<dd><p>Helper to get number of samples in a [<cite>~torch.utils.data.DataLoader</cite>] by accessing its dataset. When
+dataloader.dataset does not exist or has no length, estimates as best it can</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._hp_search_setup">
+<span class="sig-name descname"><span class="pre">_hp_search_setup</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">trial</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">optuna.Trial</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._hp_search_setup"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._hp_search_setup" title="Link to this definition">#</a></dt>
+<dd><p>HP search setup code</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._report_to_hp_search">
+<span class="sig-name descname"><span class="pre">_report_to_hp_search</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">trial</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">optuna.Trial</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">step</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metrics</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">float</span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._report_to_hp_search"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._report_to_hp_search" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._tune_save_checkpoint">
+<span class="sig-name descname"><span class="pre">_tune_save_checkpoint</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._tune_save_checkpoint"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._tune_save_checkpoint" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.call_model_init">
+<span class="sig-name descname"><span class="pre">call_model_init</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">trial</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.call_model_init"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.call_model_init" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.torch_jit_model_eval">
+<span class="sig-name descname"><span class="pre">torch_jit_model_eval</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataloader</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">training</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.torch_jit_model_eval"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.torch_jit_model_eval" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.ipex_optimize_model">
+<span class="sig-name descname"><span class="pre">ipex_optimize_model</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">training</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dtype</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">torch.float32</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.ipex_optimize_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.ipex_optimize_model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._wrap_model">
+<span class="sig-name descname"><span class="pre">_wrap_model</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">training</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataloader</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._wrap_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._wrap_model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.train">
+<span class="sig-name descname"><span class="pre">train</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">resume_from_checkpoint</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trial</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">optuna.Trial</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ignore_keys_for_eval</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">is_first_time</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.train"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.train" title="Link to this definition">#</a></dt>
+<dd><p>Main training entry point.
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>resume_from_checkpoint (<cite>str</cite> or <cite>bool</cite>, <em>optional</em>):</dt><dd><p>If a <cite>str</cite>, local path to a saved checkpoint as saved by a previous instance of [<cite>Trainer</cite>]. If a
+<cite>bool</cite> and equals <cite>True</cite>, load the last checkpoint in <em>args.output_dir</em> as saved by a previous instance
+of [<cite>Trainer</cite>]. If present, training will resume from the model/optimizer/scheduler states loaded here.</p>
+</dd>
+<dt>trial (<cite>optuna.Trial</cite> or <cite>Dict[str, Any]</cite>, <em>optional</em>):</dt><dd><p>The trial run or the hyperparameter dictionary for hyperparameter search.</p>
+</dd>
+<dt>ignore_keys_for_eval (<cite>List[str]</cite>, <em>optional</em>)</dt><dd><p>A list of keys in the output of your model (if it is a dictionary) that should be ignored when
+gathering predictions for evaluation during the training.</p>
+</dd>
+<dt>kwargs:</dt><dd><p>Additional keyword arguments used to hide deprecated arguments</p>
+</dd>
+</dl>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._one_train">
+<span class="sig-name descname"><span class="pre">_one_train</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">batch_size</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">args</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">resume_from_checkpoint</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trial</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ignore_keys_for_eval</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._one_train"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._one_train" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._inner_training_loop">
+<span class="sig-name descname"><span class="pre">_inner_training_loop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">batch_size</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">args</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">resume_from_checkpoint</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trial</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ignore_keys_for_eval</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._inner_training_loop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._inner_training_loop" title="Link to this definition">#</a></dt>
+<dd><p>0 This function serves to train one time
+1 Update the self.train_dataset before calling this function</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_output_dir">
+<span class="sig-name descname"><span class="pre">_get_output_dir</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">trial</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._get_output_dir"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_output_dir" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_from_checkpoint">
+<span class="sig-name descname"><span class="pre">_load_from_checkpoint</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">resume_from_checkpoint</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._load_from_checkpoint"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_from_checkpoint" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_best_model">
+<span class="sig-name descname"><span class="pre">_load_best_model</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._load_best_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_best_model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._issue_warnings_after_load">
+<span class="sig-name descname"><span class="pre">_issue_warnings_after_load</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">load_result</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._issue_warnings_after_load"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._issue_warnings_after_load" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._maybe_log_save_evaluate">
+<span class="sig-name descname"><span class="pre">_maybe_log_save_evaluate</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tr_loss</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trial</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">epoch</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ignore_keys_for_eval</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._maybe_log_save_evaluate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._maybe_log_save_evaluate" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_rng_state">
+<span class="sig-name descname"><span class="pre">_load_rng_state</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">checkpoint</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._load_rng_state"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_rng_state" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._save_checkpoint">
+<span class="sig-name descname"><span class="pre">_save_checkpoint</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">trial</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metrics</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._save_checkpoint"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._save_checkpoint" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_optimizer_and_scheduler">
+<span class="sig-name descname"><span class="pre">_load_optimizer_and_scheduler</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">checkpoint</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._load_optimizer_and_scheduler"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_optimizer_and_scheduler" title="Link to this definition">#</a></dt>
+<dd><p>If optimizer and scheduler states exist, load them.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.hyperparameter_search">
+<span class="sig-name descname"><span class="pre">hyperparameter_search</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">hp_space</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Callable</span><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><span class="pre">optuna.Trial</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">float</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">compute_objective</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Callable</span><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">float</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">float</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">n_trials</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">20</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">direction</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'minimize'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">backend</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">transformers.trainer_utils.HPSearchBackend</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">hp_name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Callable</span><span class="p"><span class="pre">[</span></span><span class="p"><span class="pre">[</span></span><span class="pre">optuna.Trial</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">transformers.trainer_utils.BestRun</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.hyperparameter_search"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.hyperparameter_search" title="Link to this definition">#</a></dt>
+<dd><p>Launch an hyperparameter search using <cite>optuna</cite> or <cite>Ray Tune</cite> or <cite>SigOpt</cite>. The optimized quantity is determined
+by <cite>compute_objective</cite>, which defaults to a function returning the evaluation loss when no metric is provided,
+the sum of all metrics otherwise.
+&lt;Tip warning={true}&gt;
+To use this method, you need to have provided a <cite>model_init</cite> when initializing your [<cite>Trainer</cite>]: we need to
+reinitialize the model at each new run. This is incompatible with the <cite>optimizers</cite> argument, so you need to
+subclass [<cite>Trainer</cite>] and override the method [<cite>~Trainer.create_optimizer_and_scheduler</cite>] for custom
+optimizer/scheduler.
+&lt;/Tip&gt;
+Args:</p>
+<blockquote>
+<div><dl>
+<dt>hp_space (<cite>Callable[[“optuna.Trial”], Dict[str, float]]</cite>, <em>optional</em>):</dt><dd><p>A function that defines the hyperparameter search space. Will default to
+[<cite>~trainer_utils.default_hp_space_optuna</cite>] or [<cite>~trainer_utils.default_hp_space_ray</cite>] or
+[<cite>~trainer_utils.default_hp_space_sigopt</cite>] depending on your backend.</p>
+</dd>
+<dt>compute_objective (<cite>Callable[[Dict[str, float]], float]</cite>, <em>optional</em>):</dt><dd><p>A function computing the objective to minimize or maximize from the metrics returned by the <cite>evaluate</cite>
+method. Will default to [<cite>~trainer_utils.default_compute_objective</cite>].</p>
+</dd>
+<dt>n_trials (<cite>int</cite>, <em>optional</em>, defaults to 100):</dt><dd><p>The number of trial runs to test.</p>
+</dd>
+<dt>direction (<cite>str</cite>, <em>optional</em>, defaults to <cite>“minimize”</cite>):</dt><dd><p>Whether to optimize greater or lower objects. Can be <cite>“minimize”</cite> or <cite>“maximize”</cite>, you should pick
+<cite>“minimize”</cite> when optimizing the validation loss, <cite>“maximize”</cite> when optimizing one or several metrics.</p>
+</dd>
+<dt>backend (<cite>str</cite> or [<cite>~training_utils.HPSearchBackend</cite>], <em>optional</em>):</dt><dd><p>The backend to use for hyperparameter search. Will default to optuna or Ray Tune or SigOpt, depending
+on which one is installed. If all are installed, will default to optuna.</p>
+</dd>
+<dt>hp_name (<cite>Callable[[“optuna.Trial”], str]]</cite>, <em>optional</em>):</dt><dd><p>A function that defines the trial/run name. Will default to None.</p>
+</dd>
+<dt>kwargs (<cite>Dict[str, Any]</cite>, <em>optional</em>):</dt><dd><p>Additional keyword arguments passed along to <cite>optuna.create_study</cite> or <cite>ray.tune.run</cite>. For more
+information see:
+- the documentation of</p>
+<blockquote>
+<div><p>[optuna.create_study](<a class="reference external" href="https://optuna.readthedocs.io/en/stable/reference/generated/optuna.study.create_study.html">https://optuna.readthedocs.io/en/stable/reference/generated/optuna.study.create_study.html</a>)</p>
+</div></blockquote>
+<ul class="simple">
+<li><p>the documentation of [tune.run](<a class="reference external" href="https://docs.ray.io/en/latest/tune/api_docs/execution.html#tune-run">https://docs.ray.io/en/latest/tune/api_docs/execution.html#tune-run</a>)</p></li>
+<li><p>the documentation of [sigopt](<a class="reference external" href="https://app.sigopt.com/docs/endpoints/experiments/create">https://app.sigopt.com/docs/endpoints/experiments/create</a>)</p></li>
+</ul>
+</dd>
+</dl>
+</div></blockquote>
+<dl class="simple">
+<dt>Returns:</dt><dd><p>[<cite>trainer_utils.BestRun</cite>]: All the information about the best run. Experiment summary can be found in
+<cite>run_summary</cite> attribute for Ray backend.</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.log">
+<span class="sig-name descname"><span class="pre">log</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">logs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">float</span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.log"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.log" title="Link to this definition">#</a></dt>
+<dd><p>Log <cite>logs</cite> on the various objects watching training.
+Subclass and override this method to inject custom behavior.
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>logs (<cite>Dict[str, float]</cite>):</dt><dd><p>The values to log.</p>
+</dd>
+</dl>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._prepare_input">
+<span class="sig-name descname"><span class="pre">_prepare_input</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Any</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Any</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._prepare_input"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._prepare_input" title="Link to this definition">#</a></dt>
+<dd><p>Prepares one <cite>data</cite> before feeding it to the model, be it a tensor or a nested list/dictionary of tensors.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._prepare_inputs">
+<span class="sig-name descname"><span class="pre">_prepare_inputs</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._prepare_inputs"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._prepare_inputs" title="Link to this definition">#</a></dt>
+<dd><p>Prepare <cite>inputs</cite> before feeding them to the model, converting them to tensors if they are not already and
+handling potential state.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_loss_context_manager">
+<span class="sig-name descname"><span class="pre">compute_loss_context_manager</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.compute_loss_context_manager"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_loss_context_manager" title="Link to this definition">#</a></dt>
+<dd><p>A helper wrapper to group together context managers.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.autocast_smart_context_manager">
+<span class="sig-name descname"><span class="pre">autocast_smart_context_manager</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">cache_enabled</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.autocast_smart_context_manager"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.autocast_smart_context_manager" title="Link to this definition">#</a></dt>
+<dd><p>A helper wrapper that creates an appropriate context manager for <cite>autocast</cite> while feeding it the desired
+arguments, depending on the situation.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.training_step">
+<span class="sig-name descname"><span class="pre">training_step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.nn.Module</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">torch.Tensor</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.training_step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.training_step" title="Link to this definition">#</a></dt>
+<dd><p>Perform a training step on a batch of inputs.
+Subclass and override to inject custom behavior.
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>model (<cite>nn.Module</cite>):</dt><dd><p>The model to train.</p>
+</dd>
+<dt>inputs (<cite>Dict[str, Union[torch.Tensor, Any]]</cite>):</dt><dd><p>The inputs and targets of the model.
+The dictionary will be unpacked before being fed to the model. Most models expect the targets under the
+argument <cite>labels</cite>. Check your model’s documentation for all accepted arguments.</p>
+</dd>
+</dl>
+</div></blockquote>
+<dl class="simple">
+<dt>Return:</dt><dd><p><cite>torch.Tensor</cite>: The tensor with training loss on this batch.</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_loss">
+<span class="sig-name descname"><span class="pre">compute_loss</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inputs</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_outputs</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.compute_loss"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_loss" title="Link to this definition">#</a></dt>
+<dd><p>How the loss is computed by Trainer. By default, all models return the loss in the first element.
+Subclass and override for custom behavior.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_local_process_zero">
+<span class="sig-name descname"><span class="pre">is_local_process_zero</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">bool</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.is_local_process_zero"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_local_process_zero" title="Link to this definition">#</a></dt>
+<dd><p>Whether or not this process is the local (e.g., on one machine if training in a distributed fashion on several
+machines) main process.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_world_process_zero">
+<span class="sig-name descname"><span class="pre">is_world_process_zero</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">bool</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.is_world_process_zero"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_world_process_zero" title="Link to this definition">#</a></dt>
+<dd><p>Whether or not this process is the global main process (when training in a distributed fashion on several
+machines, this is only going to be <cite>True</cite> for one process).</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.save_model">
+<span class="sig-name descname"><span class="pre">save_model</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">output_dir</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">_internal_call</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.save_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.save_model" title="Link to this definition">#</a></dt>
+<dd><p>Will save the model, so you can reload it using <cite>from_pretrained()</cite>.
+Will only save from the main process.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._save_tpu">
+<span class="sig-name descname"><span class="pre">_save_tpu</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">output_dir</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._save_tpu"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._save_tpu" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._save">
+<span class="sig-name descname"><span class="pre">_save</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">output_dir</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">state_dict</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._save"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._save" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.store_flos">
+<span class="sig-name descname"><span class="pre">store_flos</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.store_flos"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.store_flos" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._sorted_checkpoints">
+<span class="sig-name descname"><span class="pre">_sorted_checkpoints</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">output_dir</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">checkpoint_prefix</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">PREFIX_CHECKPOINT_DIR</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_mtime</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._sorted_checkpoints"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._sorted_checkpoints" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._rotate_checkpoints">
+<span class="sig-name descname"><span class="pre">_rotate_checkpoints</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">use_mtime</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_dir</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._rotate_checkpoints"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._rotate_checkpoints" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.evaluate">
+<span class="sig-name descname"><span class="pre">evaluate</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">eval_dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.utils.data.Dataset</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ignore_keys</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metric_key_prefix</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'eval'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">float</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.evaluate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.evaluate" title="Link to this definition">#</a></dt>
+<dd><p>Run evaluation and returns metrics.
+The calling script will be responsible for providing a method to compute metrics, as they are task-dependent
+(pass it to the init <cite>compute_metrics</cite> argument).
+You can also subclass and override this method to inject custom behavior.
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>eval_dataset (<cite>Dataset</cite>, <em>optional</em>):</dt><dd><p>Pass a dataset if you wish to override <cite>self.eval_dataset</cite>. If it is a [<cite>~datasets.Dataset</cite>], columns
+not accepted by the <cite>model.forward()</cite> method are automatically removed. It must implement the <cite>__len__</cite>
+method.</p>
+</dd>
+<dt>ignore_keys (<cite>Lst[str]</cite>, <em>optional</em>):</dt><dd><p>A list of keys in the output of your model (if it is a dictionary) that should be ignored when
+gathering predictions.</p>
+</dd>
+<dt>metric_key_prefix (<cite>str</cite>, <em>optional</em>, defaults to <cite>“eval”</cite>):</dt><dd><p>An optional prefix to be used as the metrics key prefix. For example the metrics “bleu” will be named
+“eval_bleu” if the prefix is “eval” (default)</p>
+</dd>
+</dl>
+</div></blockquote>
+<dl class="simple">
+<dt>Returns:</dt><dd><p>A dictionary containing the evaluation loss and the potential metrics computed from the predictions. The
+dictionary also contains the epoch number which comes from the training state.</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.predict">
+<span class="sig-name descname"><span class="pre">predict</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">test_dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.utils.data.Dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ignore_keys</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metric_key_prefix</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'test'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">transformers.trainer_utils.PredictionOutput</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.predict"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.predict" title="Link to this definition">#</a></dt>
+<dd><p>Run prediction and returns predictions and potential metrics.
+Depending on the dataset and your use case, your test dataset may contain labels. In that case, this method
+will also return metrics, like in <cite>evaluate()</cite>.
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>test_dataset (<cite>Dataset</cite>):</dt><dd><p>Dataset to run the predictions on. If it is an <cite>datasets.Dataset</cite>, columns not accepted by the
+<cite>model.forward()</cite> method are automatically removed. Has to implement the method <cite>__len__</cite></p>
+</dd>
+<dt>ignore_keys (<cite>Lst[str]</cite>, <em>optional</em>):</dt><dd><p>A list of keys in the output of your model (if it is a dictionary) that should be ignored when
+gathering predictions.</p>
+</dd>
+<dt>metric_key_prefix (<cite>str</cite>, <em>optional</em>, defaults to <cite>“test”</cite>):</dt><dd><p>An optional prefix to be used as the metrics key prefix. For example the metrics “bleu” will be named
+“test_bleu” if the prefix is “test” (default)</p>
+</dd>
+</dl>
+</div></blockquote>
+<p>&lt;Tip&gt;
+If your predictions or labels have different sequence length (for instance because you’re doing dynamic padding
+in a token classification task) the predictions will be padded (on the right) to allow for concatenation into
+one array. The padding index is -100.
+&lt;/Tip&gt;
+Returns: <em>NamedTuple</em> A namedtuple with the following keys:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p>predictions (<cite>np.ndarray</cite>): The predictions on <cite>test_dataset</cite>.</p></li>
+<li><p>label_ids (<cite>np.ndarray</cite>, <em>optional</em>): The labels (if the dataset contained some).</p></li>
+<li><p>metrics (<cite>Dict[str, float]</cite>, <em>optional</em>): The potential dictionary of metrics (if the dataset contained
+labels).</p></li>
+</ul>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.evaluation_loop">
+<span class="sig-name descname"><span class="pre">evaluation_loop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataloader</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.utils.data.DataLoader</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">description</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">prediction_loss_only</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ignore_keys</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metric_key_prefix</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'eval'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">transformers.trainer_utils.EvalLoopOutput</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.evaluation_loop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.evaluation_loop" title="Link to this definition">#</a></dt>
+<dd><p>Prediction/evaluation loop, shared by <cite>Trainer.evaluate()</cite> and <cite>Trainer.predict()</cite>.
+Works both with or without labels.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._nested_gather">
+<span class="sig-name descname"><span class="pre">_nested_gather</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tensors</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">name</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._nested_gather"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._nested_gather" title="Link to this definition">#</a></dt>
+<dd><p>Gather value of <cite>tensors</cite> (tensor or list/tuple of nested tensors) and convert them to numpy before
+concatenating them to <cite>gathered</cite></p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._pad_across_processes">
+<span class="sig-name descname"><span class="pre">_pad_across_processes</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pad_index</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">-100</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._pad_across_processes"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._pad_across_processes" title="Link to this definition">#</a></dt>
+<dd><p>Recursively pad the tensors in a nested list/tuple/dictionary of tensors from all devices to the same size so
+they can safely be gathered.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.prediction_step">
+<span class="sig-name descname"><span class="pre">prediction_step</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.nn.Module</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">prediction_loss_only</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ignore_keys</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.prediction_step"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.prediction_step" title="Link to this definition">#</a></dt>
+<dd><p>Perform an evaluation step on <cite>model</cite> using <cite>inputs</cite>.
+Subclass and override to inject custom behavior.
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>model (<cite>nn.Module</cite>):</dt><dd><p>The model to evaluate.</p>
+</dd>
+<dt>inputs (<cite>Dict[str, Union[torch.Tensor, Any]]</cite>):</dt><dd><p>The inputs and targets of the model.
+The dictionary will be unpacked before being fed to the model. Most models expect the targets under the
+argument <cite>labels</cite>. Check your model’s documentation for all accepted arguments.</p>
+</dd>
+<dt>prediction_loss_only (<cite>bool</cite>):</dt><dd><p>Whether or not to return the loss only.</p>
+</dd>
+<dt>ignore_keys (<cite>Lst[str]</cite>, <em>optional</em>):</dt><dd><p>A list of keys in the output of your model (if it is a dictionary) that should be ignored when
+gathering predictions.</p>
+</dd>
+</dl>
+</div></blockquote>
+<dl class="simple">
+<dt>Return:</dt><dd><p>Tuple[Optional[torch.Tensor], Optional[torch.Tensor], Optional[torch.Tensor]]: A tuple with the loss,
+logits and labels (each being optional).</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.floating_point_ops">
+<span class="sig-name descname"><span class="pre">floating_point_ops</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.floating_point_ops"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.floating_point_ops" title="Link to this definition">#</a></dt>
+<dd><p>For models that inherit from [<cite>PreTrainedModel</cite>], uses that method to compute the number of floating point
+operations for every backward + forward pass. If using another model, either implement such a method in the
+model or subclass and override this method.
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>inputs (<cite>Dict[str, Union[torch.Tensor, Any]]</cite>):</dt><dd><p>The inputs and targets of the model.</p>
+</dd>
+</dl>
+</div></blockquote>
+<dl class="simple">
+<dt>Returns:</dt><dd><p><cite>int</cite>: The number of floating-point operations.</p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.init_git_repo">
+<span class="sig-name descname"><span class="pre">init_git_repo</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">at_init</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.init_git_repo"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.init_git_repo" title="Link to this definition">#</a></dt>
+<dd><p>Initializes a git repo in <cite>self.args.hub_model_id</cite>.
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>at_init (<cite>bool</cite>, <em>optional</em>, defaults to <cite>False</cite>):</dt><dd><p>Whether this function is called before any training or not. If <cite>self.args.overwrite_output_dir</cite> is
+<cite>True</cite> and <cite>at_init</cite> is <cite>True</cite>, the path to the repo (which is <cite>self.args.output_dir</cite>) might be wiped
+out.</p>
+</dd>
+</dl>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_model_card">
+<span class="sig-name descname"><span class="pre">create_model_card</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">language</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">license</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tags</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">finetuned_from</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tasks</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset_tags</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.create_model_card"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_model_card" title="Link to this definition">#</a></dt>
+<dd><p>Creates a draft of a model card using the information available to the <cite>Trainer</cite>.
+Args:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>language (<cite>str</cite>, <em>optional</em>):</dt><dd><p>The language of the model (if applicable)</p>
+</dd>
+<dt>license (<cite>str</cite>, <em>optional</em>):</dt><dd><p>The license of the model. Will default to the license of the pretrained model used, if the original
+model given to the <cite>Trainer</cite> comes from a repo on the Hub.</p>
+</dd>
+<dt>tags (<cite>str</cite> or <cite>List[str]</cite>, <em>optional</em>):</dt><dd><p>Some tags to be included in the metadata of the model card.</p>
+</dd>
+<dt>model_name (<cite>str</cite>, <em>optional</em>):</dt><dd><p>The name of the model.</p>
+</dd>
+<dt>finetuned_from (<cite>str</cite>, <em>optional</em>):</dt><dd><p>The name of the model used to fine-tune this one (if applicable). Will default to the name of the repo
+of the original model given to the <cite>Trainer</cite> (if it comes from the Hub).</p>
+</dd>
+<dt>tasks (<cite>str</cite> or <cite>List[str]</cite>, <em>optional</em>):</dt><dd><p>One or several task identifiers, to be included in the metadata of the model card.</p>
+</dd>
+<dt>dataset_tags (<cite>str</cite> or <cite>List[str]</cite>, <em>optional</em>):</dt><dd><p>One or several dataset tags, to be included in the metadata of the model card.</p>
+</dd>
+<dt>dataset (<cite>str</cite> or <cite>List[str]</cite>, <em>optional</em>):</dt><dd><p>One or several dataset identifiers, to be included in the metadata of the model card.</p>
+</dd>
+<dt>dataset_args (<cite>str</cite> or <cite>List[str]</cite>, <em>optional</em>):</dt><dd><p>One or several dataset arguments, to be included in the metadata of the model card.</p>
+</dd>
+</dl>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._push_from_checkpoint">
+<span class="sig-name descname"><span class="pre">_push_from_checkpoint</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">checkpoint_folder</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._push_from_checkpoint"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._push_from_checkpoint" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.push_to_hub">
+<span class="sig-name descname"><span class="pre">push_to_hub</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">commit_message</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'End</span> <span class="pre">of</span> <span class="pre">training'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">blocking</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">str</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.push_to_hub"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.push_to_hub" title="Link to this definition">#</a></dt>
+<dd><p>Upload <em>self.model</em> and <em>self.tokenizer</em> to the 🤗 model hub on the repo <em>self.args.hub_model_id</em>.
+Parameters:</p>
+<blockquote>
+<div><dl class="simple">
+<dt>commit_message (<cite>str</cite>, <em>optional</em>, defaults to <cite>“End of training”</cite>):</dt><dd><p>Message to commit while pushing.</p>
+</dd>
+<dt>blocking (<cite>bool</cite>, <em>optional</em>, defaults to <cite>True</cite>):</dt><dd><p>Whether the function should return only when the <cite>git push</cite> has finished.</p>
+</dd>
+<dt>kwargs:</dt><dd><p>Additional keyword arguments passed along to [<cite>~Trainer.create_model_card</cite>].</p>
+</dd>
+</dl>
+</div></blockquote>
+<dl class="simple">
+<dt>Returns:</dt><dd><p>The url of the commit of your model in the given repository if <cite>blocking=False</cite>, a tuple with the url of
+the commit and an object to track the progress of the commit if <cite>blocking=True</cite></p>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer.prediction_loop">
+<span class="sig-name descname"><span class="pre">prediction_loop</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataloader</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.utils.data.DataLoader</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">description</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">prediction_loss_only</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ignore_keys</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metric_key_prefix</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'eval'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">transformers.trainer_utils.EvalLoopOutput</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer.prediction_loop"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.prediction_loop" title="Link to this definition">#</a></dt>
+<dd><p>Prediction/evaluation loop, shared by <cite>Trainer.evaluate()</cite> and <cite>Trainer.predict()</cite>.
+Works both with or without labels.</p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._gather_and_numpify">
+<span class="sig-name descname"><span class="pre">_gather_and_numpify</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tensors</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">name</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._gather_and_numpify"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._gather_and_numpify" title="Link to this definition">#</a></dt>
+<dd><p>Gather value of <cite>tensors</cite> (tensor or list/tuple of nested tensors) and convert them to numpy before
+concatenating them to <cite>gathered</cite></p>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.raft_trainer.RaftTrainer._add_sm_patterns_to_gitignore">
+<span class="sig-name descname"><span class="pre">_add_sm_patterns_to_gitignore</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/raft_trainer.html#RaftTrainer._add_sm_patterns_to_gitignore"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._add_sm_patterns_to_gitignore" title="Link to this definition">#</a></dt>
+<dd><p>Add SageMaker Checkpointing patterns to .gitignore file.</p>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../peft_trainer/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.peft_trainer</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../rm_dataprocessor/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.rm_dataprocessor</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.is_torch_greater_or_equal_than_1_10"><code class="docutils literal notranslate"><span class="pre">is_torch_greater_or_equal_than_1_10</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.is_torch_less_than_1_11"><code class="docutils literal notranslate"><span class="pre">is_torch_less_than_1_11</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer._is_native_cpu_amp_available"><code class="docutils literal notranslate"><span class="pre">_is_native_cpu_amp_available</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.DEFAULT_CALLBACKS"><code class="docutils literal notranslate"><span class="pre">DEFAULT_CALLBACKS</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.DEFAULT_PROGRESS_CALLBACK"><code class="docutils literal notranslate"><span class="pre">DEFAULT_PROGRESS_CALLBACK</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#id0"><code class="docutils literal notranslate"><span class="pre">DEFAULT_PROGRESS_CALLBACK</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.IS_SAGEMAKER_MP_POST_1_10"><code class="docutils literal notranslate"><span class="pre">IS_SAGEMAKER_MP_POST_1_10</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.skip_first_batches"><code class="docutils literal notranslate"><span class="pre">skip_first_batches</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.TRAINING_ARGS_NAME"><code class="docutils literal notranslate"><span class="pre">TRAINING_ARGS_NAME</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.TRAINER_STATE_NAME"><code class="docutils literal notranslate"><span class="pre">TRAINER_STATE_NAME</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.OPTIMIZER_NAME"><code class="docutils literal notranslate"><span class="pre">OPTIMIZER_NAME</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.SCHEDULER_NAME"><code class="docutils literal notranslate"><span class="pre">SCHEDULER_NAME</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.SCALER_NAME"><code class="docutils literal notranslate"><span class="pre">SCALER_NAME</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer"><code class="docutils literal notranslate"><span class="pre">RaftTrainer</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.save_counter"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.save_counter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.args"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.hp_name"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.hp_name</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.deepspeed"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.deepspeed</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_in_train"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.is_in_train</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._memory_tracker"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._memory_tracker</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.sharded_ddp"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.sharded_ddp</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.fsdp"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.fsdp</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.place_model_on_device"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.place_model_on_device</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.data_collator"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.data_collator</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.train_dataset"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.train_dataset</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.eval_dataset"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.eval_dataset</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.tokenizer"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.tokenizer</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.model_wrapped"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.model_wrapped</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.model"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.model</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_metrics"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.compute_metrics</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.preprocess_logits_for_metrics"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.preprocess_logits_for_metrics</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.callback_handler"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.callback_handler</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._loggers_initialized"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._loggers_initialized</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._signature_columns"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._signature_columns</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_apex"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.use_apex</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_cuda_amp"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.use_cuda_amp</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_cpu_amp"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.use_cpu_amp</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.do_grad_scaling"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.do_grad_scaling</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.state"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.state</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.control"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.control</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.current_flos"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.current_flos</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.hp_search_backend"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.hp_search_backend</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_tune_checkpoints"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.use_tune_checkpoints</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.label_names"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.label_names</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.can_return_loss"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.can_return_loss</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._train_batch_size"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._train_batch_size</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.add_callback"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.add_callback()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.pop_callback"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.pop_callback()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.remove_callback"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.remove_callback()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._move_model_to_device"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._move_model_to_device()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._set_signature_columns_if_needed"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._set_signature_columns_if_needed()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._remove_unused_columns"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._remove_unused_columns()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_collator_with_removed_columns"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._get_collator_with_removed_columns()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_train_sampler"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._get_train_sampler()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_train_dataloader"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.get_train_dataloader()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_eval_sampler"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._get_eval_sampler()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_eval_dataloader"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.get_eval_dataloader()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_test_dataloader"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.get_test_dataloader()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_optimizer_and_scheduler"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.create_optimizer_and_scheduler()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_optimizer"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.create_optimizer()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_optimizer_cls_and_kwargs"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.get_optimizer_cls_and_kwargs()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_scheduler"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.create_scheduler()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.num_examples"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.num_examples()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._hp_search_setup"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._hp_search_setup()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._report_to_hp_search"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._report_to_hp_search()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._tune_save_checkpoint"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._tune_save_checkpoint()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.call_model_init"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.call_model_init()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.torch_jit_model_eval"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.torch_jit_model_eval()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.ipex_optimize_model"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.ipex_optimize_model()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._wrap_model"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._wrap_model()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.train"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.train()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._one_train"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._one_train()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._inner_training_loop"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._inner_training_loop()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_output_dir"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._get_output_dir()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_from_checkpoint"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._load_from_checkpoint()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_best_model"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._load_best_model()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._issue_warnings_after_load"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._issue_warnings_after_load()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._maybe_log_save_evaluate"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._maybe_log_save_evaluate()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_rng_state"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._load_rng_state()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._save_checkpoint"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._save_checkpoint()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_optimizer_and_scheduler"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._load_optimizer_and_scheduler()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.hyperparameter_search"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.hyperparameter_search()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.log"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.log()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._prepare_input"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._prepare_input()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._prepare_inputs"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._prepare_inputs()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_loss_context_manager"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.compute_loss_context_manager()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.autocast_smart_context_manager"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.autocast_smart_context_manager()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.training_step"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.training_step()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_loss"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.compute_loss()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_local_process_zero"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.is_local_process_zero()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_world_process_zero"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.is_world_process_zero()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.save_model"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.save_model()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._save_tpu"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._save_tpu()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._save"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._save()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.store_flos"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.store_flos()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._sorted_checkpoints"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._sorted_checkpoints()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._rotate_checkpoints"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._rotate_checkpoints()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.evaluate"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.evaluate()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.predict"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.predict()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.evaluation_loop"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.evaluation_loop()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._nested_gather"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._nested_gather()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._pad_across_processes"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._pad_across_processes()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.prediction_step"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.prediction_step()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.floating_point_ops"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.floating_point_ops()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.init_git_repo"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.init_git_repo()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_model_card"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.create_model_card()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._push_from_checkpoint"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._push_from_checkpoint()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.push_to_hub"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.push_to_hub()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer.prediction_loop"><code class="docutils literal notranslate"><span class="pre">RaftTrainer.prediction_loop()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._gather_and_numpify"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._gather_and_numpify()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.raft_trainer.RaftTrainer._add_sm_patterns_to_gitignore"><code class="docutils literal notranslate"><span class="pre">RaftTrainer._add_sm_patterns_to_gitignore()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/pipeline/utils/raft_trainer/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html b/autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html
new file mode 100644
index 000000000..d62a68551
--- /dev/null
+++ b/autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html
@@ -0,0 +1,714 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.utils.rm_dataprocessor &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/utils/rm_dataprocessor/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.utils.rm_trainer" href="../rm_trainer/index.html" />
+    <link rel="prev" title="lmflow.pipeline.utils.raft_trainer" href="../raft_trainer/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.rm_dataprocessor</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.utils.rm_dataprocessor">
+<span id="lmflow-pipeline-utils-rm-dataprocessor"></span><h1>lmflow.pipeline.utils.rm_dataprocessor<a class="headerlink" href="#module-lmflow.pipeline.utils.rm_dataprocessor" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.rm_dataprocessor.logger" title="lmflow.pipeline.utils.rm_dataprocessor.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding" title="lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RewardDataCollatorWithPadding</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_dataprocessor.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.rm_dataprocessor.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_dataprocessor.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_dataprocessor.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.rm_dataprocessor.</span></span><span class="sig-name descname"><span class="pre">RewardDataCollatorWithPadding</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_dataprocessor.html#RewardDataCollatorWithPadding"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding" title="Link to this definition">#</a></dt>
+<dd><dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.tokenizer">
+<span class="sig-name descname"><span class="pre">tokenizer</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">transformers.AutoTokenizer</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_dataprocessor.html#RewardDataCollatorWithPadding.tokenizer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.tokenizer" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.padding">
+<span class="sig-name descname"><span class="pre">padding</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">transformers.utils.PaddingStrategy</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_dataprocessor.html#RewardDataCollatorWithPadding.padding"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.padding" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.max_length">
+<span class="sig-name descname"><span class="pre">max_length</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_dataprocessor.html#RewardDataCollatorWithPadding.max_length"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.max_length" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.pad_to_multiple_of">
+<span class="sig-name descname"><span class="pre">pad_to_multiple_of</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_dataprocessor.html#RewardDataCollatorWithPadding.pad_to_multiple_of"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.pad_to_multiple_of" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.return_tensors">
+<span class="sig-name descname"><span class="pre">return_tensors</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'pt'</span></em><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_dataprocessor.html#RewardDataCollatorWithPadding.return_tensors"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.return_tensors" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.__call__">
+<span class="sig-name descname"><span class="pre">__call__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">features</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_dataprocessor.html#RewardDataCollatorWithPadding.__call__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.__call__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../raft_trainer/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.raft_trainer</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../rm_trainer/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.rm_trainer</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_dataprocessor.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding"><code class="docutils literal notranslate"><span class="pre">RewardDataCollatorWithPadding</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.tokenizer"><code class="docutils literal notranslate"><span class="pre">RewardDataCollatorWithPadding.tokenizer</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.padding"><code class="docutils literal notranslate"><span class="pre">RewardDataCollatorWithPadding.padding</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.max_length"><code class="docutils literal notranslate"><span class="pre">RewardDataCollatorWithPadding.max_length</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.pad_to_multiple_of"><code class="docutils literal notranslate"><span class="pre">RewardDataCollatorWithPadding.pad_to_multiple_of</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.return_tensors"><code class="docutils literal notranslate"><span class="pre">RewardDataCollatorWithPadding.return_tensors</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.__call__"><code class="docutils literal notranslate"><span class="pre">RewardDataCollatorWithPadding.__call__()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/utils/rm_trainer/index.html b/autoapi/lmflow/pipeline/utils/rm_trainer/index.html
new file mode 100644
index 000000000..e30141aaf
--- /dev/null
+++ b/autoapi/lmflow/pipeline/utils/rm_trainer/index.html
@@ -0,0 +1,712 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.utils.rm_trainer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/utils/rm_trainer/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.pipeline.vllm_inferencer" href="../../vllm_inferencer/index.html" />
+    <link rel="prev" title="lmflow.pipeline.utils.rm_dataprocessor" href="../rm_dataprocessor/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.utils.rm_trainer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.utils.rm_trainer">
+<span id="lmflow-pipeline-utils-rm-trainer"></span><h1>lmflow.pipeline.utils.rm_trainer<a class="headerlink" href="#module-lmflow.pipeline.utils.rm_trainer" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.rm_trainer.RewardTrainer" title="lmflow.pipeline.utils.rm_trainer.RewardTrainer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RewardTrainer</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer" title="lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PeftRewardTrainer</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.rm_trainer.compute_metrics" title="lmflow.pipeline.utils.rm_trainer.compute_metrics"><code class="xref py py-obj docutils literal notranslate"><span class="pre">compute_metrics</span></code></a>(eval_pred)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.utils.rm_trainer.rm_loss" title="lmflow.pipeline.utils.rm_trainer.rm_loss"><code class="xref py py-obj docutils literal notranslate"><span class="pre">rm_loss</span></code></a>(model, inputs[, return_outputs])</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_trainer.compute_metrics">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.rm_trainer.</span></span><span class="sig-name descname"><span class="pre">compute_metrics</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">eval_pred</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_trainer.html#compute_metrics"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_trainer.compute_metrics" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_trainer.rm_loss">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.rm_trainer.</span></span><span class="sig-name descname"><span class="pre">rm_loss</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inputs</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_outputs</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_trainer.html#rm_loss"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_trainer.rm_loss" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_trainer.RewardTrainer">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.rm_trainer.</span></span><span class="sig-name descname"><span class="pre">RewardTrainer</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_trainer.html#RewardTrainer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_trainer.RewardTrainer" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">transformers.Trainer</span></code></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_trainer.RewardTrainer.compute_loss">
+<span class="sig-name descname"><span class="pre">compute_loss</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inputs</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_outputs</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_trainer.html#RewardTrainer.compute_loss"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_trainer.RewardTrainer.compute_loss" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.utils.rm_trainer.</span></span><span class="sig-name descname"><span class="pre">PeftRewardTrainer</span></span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_trainer.html#PeftRewardTrainer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftTrainer" title="lmflow.pipeline.utils.peft_trainer.PeftTrainer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.pipeline.utils.peft_trainer.PeftTrainer</span></code></a></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer.compute_loss">
+<span class="sig-name descname"><span class="pre">compute_loss</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inputs</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">return_outputs</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/pipeline/utils/rm_trainer.html#PeftRewardTrainer.compute_loss"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer.compute_loss" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../rm_dataprocessor/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.rm_dataprocessor</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../../vllm_inferencer/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.pipeline.vllm_inferencer</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_trainer.compute_metrics"><code class="docutils literal notranslate"><span class="pre">compute_metrics()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_trainer.rm_loss"><code class="docutils literal notranslate"><span class="pre">rm_loss()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_trainer.RewardTrainer"><code class="docutils literal notranslate"><span class="pre">RewardTrainer</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_trainer.RewardTrainer.compute_loss"><code class="docutils literal notranslate"><span class="pre">RewardTrainer.compute_loss()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer"><code class="docutils literal notranslate"><span class="pre">PeftRewardTrainer</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer.compute_loss"><code class="docutils literal notranslate"><span class="pre">PeftRewardTrainer.compute_loss()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/pipeline/utils/rm_trainer/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/pipeline/vllm_inferencer/index.html b/autoapi/lmflow/pipeline/vllm_inferencer/index.html
new file mode 100644
index 000000000..32ed0066c
--- /dev/null
+++ b/autoapi/lmflow/pipeline/vllm_inferencer/index.html
@@ -0,0 +1,864 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.pipeline.vllm_inferencer &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/pipeline/vllm_inferencer/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.tokenization" href="../../tokenization/index.html" />
+    <link rel="prev" title="lmflow.pipeline.utils.rm_trainer" href="../utils/rm_trainer/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.pipeline</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.pipeline</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.pipeline.vllm_inferencer</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.pipeline.vllm_inferencer">
+<span id="lmflow-pipeline-vllm-inferencer"></span><h1>lmflow.pipeline.vllm_inferencer<a class="headerlink" href="#module-lmflow.pipeline.vllm_inferencer" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.vllm_inferencer.logger" title="lmflow.pipeline.vllm_inferencer.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading" title="lmflow.pipeline.vllm_inferencer.InferencerWithOffloading"><code class="xref py py-obj docutils literal notranslate"><span class="pre">InferencerWithOffloading</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer" title="lmflow.pipeline.vllm_inferencer.VLLMInferencer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">VLLMInferencer</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer" title="lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">MemorySafeVLLMInferencer</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.vllm_inferencer.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.InferencerWithOffloading">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.vllm_inferencer.</span></span><span class="sig-name descname"><span class="pre">InferencerWithOffloading</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">inferencer_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.InferencerArguments" title="lmflow.args.InferencerArguments"><span class="pre">lmflow.args.InferencerArguments</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#InferencerWithOffloading"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base_pipeline/index.html#lmflow.pipeline.base_pipeline.BasePipeline" title="lmflow.pipeline.base_pipeline.BasePipeline"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.pipeline.base_pipeline.BasePipeline</span></code></a></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.model_args">
+<span class="sig-name descname"><span class="pre">model_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#InferencerWithOffloading.model_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.model_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.data_args">
+<span class="sig-name descname"><span class="pre">data_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#InferencerWithOffloading.data_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.data_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.inferencer_args">
+<span class="sig-name descname"><span class="pre">inferencer_args</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#InferencerWithOffloading.inferencer_args"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.inferencer_args" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.eos_token_id">
+<span class="sig-name descname"><span class="pre">eos_token_id</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#InferencerWithOffloading.eos_token_id"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.eos_token_id" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.inference">
+<em class="property"><span class="pre">abstract</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">inference</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#InferencerWithOffloading.inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.save_inference_results">
+<em class="property"><span class="pre">abstract</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">save_inference_results</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#InferencerWithOffloading.save_inference_results"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.save_inference_results" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.load_inference_results">
+<em class="property"><span class="pre">abstract</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">load_inference_results</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#InferencerWithOffloading.load_inference_results"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.load_inference_results" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.VLLMInferencer">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.vllm_inferencer.</span></span><span class="sig-name descname"><span class="pre">VLLMInferencer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">inferencer_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.InferencerArguments" title="lmflow.args.InferencerArguments"><span class="pre">lmflow.args.InferencerArguments</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#VLLMInferencer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading" title="lmflow.pipeline.vllm_inferencer.InferencerWithOffloading"><code class="xref py py-obj docutils literal notranslate"><span class="pre">InferencerWithOffloading</span></code></a></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.VLLMInferencer.sampling_params">
+<span class="sig-name descname"><span class="pre">sampling_params</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#VLLMInferencer.sampling_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer.sampling_params" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.VLLMInferencer.parse_to_sampling_params">
+<span class="sig-name descname"><span class="pre">parse_to_sampling_params</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">inference_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.InferencerArguments" title="lmflow.args.InferencerArguments"><span class="pre">lmflow.args.InferencerArguments</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">vllm.SamplingParams</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#VLLMInferencer.parse_to_sampling_params"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer.parse_to_sampling_params" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.VLLMInferencer.inference">
+<span class="sig-name descname"><span class="pre">inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel" title="lmflow.models.hf_decoder_model.HFDecoderModel"><span class="pre">lmflow.models.hf_decoder_model.HFDecoderModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">dataset</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../datasets/index.html#lmflow.datasets.Dataset" title="lmflow.datasets.Dataset"><span class="pre">lmflow.datasets.Dataset</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">enable_decode_inference_result</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">release_gpu</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inference_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.InferencerArguments" title="lmflow.args.InferencerArguments"><span class="pre">lmflow.args.InferencerArguments</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">enable_distributed_inference</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="../../utils/data_utils/index.html#lmflow.utils.data_utils.VLLMInferenceResultWithInput" title="lmflow.utils.data_utils.VLLMInferenceResultWithInput"><span class="pre">lmflow.utils.data_utils.VLLMInferenceResultWithInput</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#VLLMInferencer.inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer.inference" title="Link to this definition">#</a></dt>
+<dd><p>Perform inference using the provided model and dataset. Will save inference results if
+<cite>save_results</cite> is set to True in <cite>inferencer_args</cite>.</p>
+<dl class="field-list">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model</strong><span class="classifier">HFDecoderModel</span></dt><dd><p>LMFlow HFDecoderModel object</p>
+</dd>
+<dt><strong>dataset</strong><span class="classifier">Dataset</span></dt><dd><p>LMFlow Dataset object</p>
+</dd>
+<dt><strong>apply_chat_template</strong><span class="classifier">bool, optional</span></dt><dd><p>Whether to apply chat template to the input, by default True.</p>
+</dd>
+<dt><strong>enable_decode_inference_result</strong><span class="classifier">bool, optional</span></dt><dd><p>Whether to decode after generation, by default False.</p>
+</dd>
+<dt><strong>release_gpu</strong><span class="classifier">bool, optional</span></dt><dd><p>Whether to release gpu resources, by default False.</p>
+</dd>
+<dt><strong>inference_args</strong><span class="classifier">InferencerArguments, optional</span></dt><dd><p>by default None</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl>
+<dt>List[VLLMInferenceResultWithInput]</dt><dd><p>Return a list of VLLMInferenceResultWithInput, where each
+element contains the input prompt and the corresponding output.</p>
+<p>When <cite>enable_decode_inference_result = True</cite>, the output would be a list of strings,
+contains sampling_params.n samples for the corresponding prompt.</p>
+<p>When <cite>enable_decode_inference_result = False</cite>, return a list of list of ints
+(token ids, no decoding after generation).</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.VLLMInferencer._inference">
+<span class="sig-name descname"><span class="pre">_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel" title="lmflow.models.hf_decoder_model.HFDecoderModel"><span class="pre">lmflow.models.hf_decoder_model.HFDecoderModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampling_params</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">vllm.SamplingParams</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">release_gpu</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="../../utils/data_utils/index.html#lmflow.utils.data_utils.VLLMInferenceResultWithInput" title="lmflow.utils.data_utils.VLLMInferenceResultWithInput"><span class="pre">lmflow.utils.data_utils.VLLMInferenceResultWithInput</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#VLLMInferencer._inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer._inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.VLLMInferencer._distributed_inference">
+<span class="sig-name descname"><span class="pre">_distributed_inference</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel" title="lmflow.models.hf_decoder_model.HFDecoderModel"><span class="pre">lmflow.models.hf_decoder_model.HFDecoderModel</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_input</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">ray.data.Dataset</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">sampling_params</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">vllm.SamplingParams</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">num_instances</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">batch_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">4</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">release_gpu</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="../../utils/data_utils/index.html#lmflow.utils.data_utils.VLLMInferenceResultWithInput" title="lmflow.utils.data_utils.VLLMInferenceResultWithInput"><span class="pre">lmflow.utils.data_utils.VLLMInferenceResultWithInput</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#VLLMInferencer._distributed_inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer._distributed_inference" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.VLLMInferencer.save_inference_results">
+<span class="sig-name descname"><span class="pre">save_inference_results</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">outputs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">save_file_path</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#VLLMInferencer.save_inference_results"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer.save_inference_results" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.VLLMInferencer.load_inference_results">
+<span class="sig-name descname"><span class="pre">load_inference_results</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">results_path</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#VLLMInferencer.load_inference_results"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer.load_inference_results" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.pipeline.vllm_inferencer.</span></span><span class="sig-name descname"><span class="pre">MemorySafeVLLMInferencer</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">inferencer_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.InferencerArguments" title="lmflow.args.InferencerArguments"><span class="pre">lmflow.args.InferencerArguments</span></a></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#MemorySafeVLLMInferencer"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer" title="lmflow.pipeline.vllm_inferencer.VLLMInferencer"><code class="xref py py-obj docutils literal notranslate"><span class="pre">VLLMInferencer</span></code></a></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer.inferencer_file_path">
+<span class="sig-name descname"><span class="pre">inferencer_file_path</span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#MemorySafeVLLMInferencer.inferencer_file_path"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer.inferencer_file_path" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer.inference">
+<span class="sig-name descname"><span class="pre">inference</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="../../utils/data_utils/index.html#lmflow.utils.data_utils.VLLMInferenceResultWithInput" title="lmflow.utils.data_utils.VLLMInferenceResultWithInput"><span class="pre">lmflow.utils.data_utils.VLLMInferenceResultWithInput</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/pipeline/vllm_inferencer.html#MemorySafeVLLMInferencer.inference"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer.inference" title="Link to this definition">#</a></dt>
+<dd><p>Perform inference using the provided model and dataset. Will save inference results if
+<cite>save_results</cite> is set to True in <cite>inferencer_args</cite>.</p>
+<dl class="field-list">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>model</strong><span class="classifier">HFDecoderModel</span></dt><dd><p>LMFlow HFDecoderModel object</p>
+</dd>
+<dt><strong>dataset</strong><span class="classifier">Dataset</span></dt><dd><p>LMFlow Dataset object</p>
+</dd>
+<dt><strong>apply_chat_template</strong><span class="classifier">bool, optional</span></dt><dd><p>Whether to apply chat template to the input, by default True.</p>
+</dd>
+<dt><strong>enable_decode_inference_result</strong><span class="classifier">bool, optional</span></dt><dd><p>Whether to decode after generation, by default False.</p>
+</dd>
+<dt><strong>release_gpu</strong><span class="classifier">bool, optional</span></dt><dd><p>Whether to release gpu resources, by default False.</p>
+</dd>
+<dt><strong>inference_args</strong><span class="classifier">InferencerArguments, optional</span></dt><dd><p>by default None</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl>
+<dt>List[VLLMInferenceResultWithInput]</dt><dd><p>Return a list of VLLMInferenceResultWithInput, where each
+element contains the input prompt and the corresponding output.</p>
+<p>When <cite>enable_decode_inference_result = True</cite>, the output would be a list of strings,
+contains sampling_params.n samples for the corresponding prompt.</p>
+<p>When <cite>enable_decode_inference_result = False</cite>, return a list of list of ints
+(token ids, no decoding after generation).</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../utils/rm_trainer/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.utils.rm_trainer</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../../tokenization/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.tokenization</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading"><code class="docutils literal notranslate"><span class="pre">InferencerWithOffloading</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.model_args"><code class="docutils literal notranslate"><span class="pre">InferencerWithOffloading.model_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.data_args"><code class="docutils literal notranslate"><span class="pre">InferencerWithOffloading.data_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.inferencer_args"><code class="docutils literal notranslate"><span class="pre">InferencerWithOffloading.inferencer_args</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.eos_token_id"><code class="docutils literal notranslate"><span class="pre">InferencerWithOffloading.eos_token_id</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.inference"><code class="docutils literal notranslate"><span class="pre">InferencerWithOffloading.inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.save_inference_results"><code class="docutils literal notranslate"><span class="pre">InferencerWithOffloading.save_inference_results()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.load_inference_results"><code class="docutils literal notranslate"><span class="pre">InferencerWithOffloading.load_inference_results()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer"><code class="docutils literal notranslate"><span class="pre">VLLMInferencer</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer.sampling_params"><code class="docutils literal notranslate"><span class="pre">VLLMInferencer.sampling_params</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer.parse_to_sampling_params"><code class="docutils literal notranslate"><span class="pre">VLLMInferencer.parse_to_sampling_params()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer.inference"><code class="docutils literal notranslate"><span class="pre">VLLMInferencer.inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer._inference"><code class="docutils literal notranslate"><span class="pre">VLLMInferencer._inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer._distributed_inference"><code class="docutils literal notranslate"><span class="pre">VLLMInferencer._distributed_inference()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer.save_inference_results"><code class="docutils literal notranslate"><span class="pre">VLLMInferencer.save_inference_results()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.VLLMInferencer.load_inference_results"><code class="docutils literal notranslate"><span class="pre">VLLMInferencer.load_inference_results()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer"><code class="docutils literal notranslate"><span class="pre">MemorySafeVLLMInferencer</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer.inferencer_file_path"><code class="docutils literal notranslate"><span class="pre">MemorySafeVLLMInferencer.inferencer_file_path</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer.inference"><code class="docutils literal notranslate"><span class="pre">MemorySafeVLLMInferencer.inference()</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/pipeline/vllm_inferencer/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/tokenization/hf_decoder_model/index.html b/autoapi/lmflow/tokenization/hf_decoder_model/index.html
new file mode 100644
index 000000000..697c8e63f
--- /dev/null
+++ b/autoapi/lmflow/tokenization/hf_decoder_model/index.html
@@ -0,0 +1,702 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.tokenization.hf_decoder_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/tokenization/hf_decoder_model/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.tokenization.hf_text_regression_model" href="../hf_text_regression_model/index.html" />
+    <link rel="prev" title="lmflow.tokenization" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.tokenization</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.tokenization</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.tokenization.hf_decoder_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.tokenization.hf_decoder_model">
+<span id="lmflow-tokenization-hf-decoder-model"></span><h1>lmflow.tokenization.hf_decoder_model<a class="headerlink" href="#module-lmflow.tokenization.hf_decoder_model" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_decoder_model.logger" title="lmflow.tokenization.hf_decoder_model.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_decoder_model.tok_logger" title="lmflow.tokenization.hf_decoder_model.tok_logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">tok_logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_decoder_model.blocking" title="lmflow.tokenization.hf_decoder_model.blocking"><code class="xref py py-obj docutils literal notranslate"><span class="pre">blocking</span></code></a>(→ Dict)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_decoder_model.tokenize_function" title="lmflow.tokenization.hf_decoder_model.tokenize_function"><code class="xref py py-obj docutils literal notranslate"><span class="pre">tokenize_function</span></code></a>(→ Dict)</p></td>
+<td><p>Handels text_only and text2text datasets tokenization</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_decoder_model.conversation_tokenize_function" title="lmflow.tokenization.hf_decoder_model.conversation_tokenize_function"><code class="xref py py-obj docutils literal notranslate"><span class="pre">conversation_tokenize_function</span></code></a>(→ Dict)</p></td>
+<td><p>Handels conversation datasets tokenization</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_decoder_model.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_decoder_model.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_decoder_model.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_decoder_model.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_decoder_model.tok_logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_decoder_model.</span></span><span class="sig-name descname"><span class="pre">tok_logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_decoder_model.html#tok_logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_decoder_model.tok_logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_decoder_model.blocking">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_decoder_model.</span></span><span class="sig-name descname"><span class="pre">blocking</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">token_dict</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">block_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_max_length</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pad_token_id</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_side</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">truncation_side</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'right'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_decoder_model.html#blocking"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_decoder_model.blocking" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_decoder_model.tokenize_function">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_decoder_model.</span></span><span class="sig-name descname"><span class="pre">tokenize_function</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">examples</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">transformers.PreTrainedTokenizerFast</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">column_names</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">label_columns</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenized_column_order</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">add_special_tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_truncation</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_decoder_model.html#tokenize_function"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_decoder_model.tokenize_function" title="Link to this definition">#</a></dt>
+<dd><p>Handels text_only and text2text datasets tokenization</p>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_decoder_model.conversation_tokenize_function">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_decoder_model.</span></span><span class="sig-name descname"><span class="pre">conversation_tokenize_function</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">examples</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">transformers.PreTrainedTokenizerFast</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">column_names</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">conversation_template</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate" title="lmflow.utils.conversation_template.ConversationTemplate"><span class="pre">lmflow.utils.conversation_template.ConversationTemplate</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_decoder_model.html#conversation_tokenize_function"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_decoder_model.conversation_tokenize_function" title="Link to this definition">#</a></dt>
+<dd><p>Handels conversation datasets tokenization</p>
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.tokenization</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../hf_text_regression_model/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.tokenization.hf_text_regression_model</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_decoder_model.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_decoder_model.tok_logger"><code class="docutils literal notranslate"><span class="pre">tok_logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_decoder_model.blocking"><code class="docutils literal notranslate"><span class="pre">blocking()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_decoder_model.tokenize_function"><code class="docutils literal notranslate"><span class="pre">tokenize_function()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_decoder_model.conversation_tokenize_function"><code class="docutils literal notranslate"><span class="pre">conversation_tokenize_function()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/tokenization/hf_decoder_model/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/tokenization/hf_text_regression_model/index.html b/autoapi/lmflow/tokenization/hf_text_regression_model/index.html
new file mode 100644
index 000000000..b914a5634
--- /dev/null
+++ b/autoapi/lmflow/tokenization/hf_text_regression_model/index.html
@@ -0,0 +1,740 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.tokenization.hf_text_regression_model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/tokenization/hf_text_regression_model/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.utils" href="../../utils/index.html" />
+    <link rel="prev" title="lmflow.tokenization.hf_decoder_model" href="../hf_decoder_model/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.tokenization</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.tokenization</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.tokenization.hf_text_regression_model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.tokenization.hf_text_regression_model">
+<span id="lmflow-tokenization-hf-text-regression-model"></span><h1>lmflow.tokenization.hf_text_regression_model<a class="headerlink" href="#module-lmflow.tokenization.hf_text_regression_model" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_text_regression_model.logger" title="lmflow.tokenization.hf_text_regression_model.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_text_regression_model.tok_logger" title="lmflow.tokenization.hf_text_regression_model.tok_logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">tok_logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_text_regression_model.blocking_paired" title="lmflow.tokenization.hf_text_regression_model.blocking_paired"><code class="xref py py-obj docutils literal notranslate"><span class="pre">blocking_paired</span></code></a>(→ Dict)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_text_regression_model.blocking" title="lmflow.tokenization.hf_text_regression_model.blocking"><code class="xref py py-obj docutils literal notranslate"><span class="pre">blocking</span></code></a>(→ Dict)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_text_regression_model.blocking_text_to_textlist" title="lmflow.tokenization.hf_text_regression_model.blocking_text_to_textlist"><code class="xref py py-obj docutils literal notranslate"><span class="pre">blocking_text_to_textlist</span></code></a>(→ Dict)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_text_regression_model.paired_conversation_tokenize_function" title="lmflow.tokenization.hf_text_regression_model.paired_conversation_tokenize_function"><code class="xref py py-obj docutils literal notranslate"><span class="pre">paired_conversation_tokenize_function</span></code></a>(→ Dict)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_text_regression_model.conversation_tokenize_function" title="lmflow.tokenization.hf_text_regression_model.conversation_tokenize_function"><code class="xref py py-obj docutils literal notranslate"><span class="pre">conversation_tokenize_function</span></code></a>(→ Dict)</p></td>
+<td><p>Handels conversation datasets tokenization</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_text_regression_model.tokenize_function" title="lmflow.tokenization.hf_text_regression_model.tokenize_function"><code class="xref py py-obj docutils literal notranslate"><span class="pre">tokenize_function</span></code></a>(→ Dict)</p></td>
+<td><p>Handels text_only and text2text datasets tokenization</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.tokenization.hf_text_regression_model.text_to_textlist_tokenize_function" title="lmflow.tokenization.hf_text_regression_model.text_to_textlist_tokenize_function"><code class="xref py py-obj docutils literal notranslate"><span class="pre">text_to_textlist_tokenize_function</span></code></a>(→ Dict)</p></td>
+<td><p>For rm inference, and don't need attn mask and labels.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_text_regression_model.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_text_regression_model.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_text_regression_model.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_text_regression_model.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_text_regression_model.tok_logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_text_regression_model.</span></span><span class="sig-name descname"><span class="pre">tok_logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_text_regression_model.html#tok_logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_text_regression_model.tok_logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_text_regression_model.blocking_paired">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_text_regression_model.</span></span><span class="sig-name descname"><span class="pre">blocking_paired</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">token_dict</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">column_names</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">block_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_max_length</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pad_token_id</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_side</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">truncation_side</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'right'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_text_regression_model.html#blocking_paired"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_text_regression_model.blocking_paired" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_text_regression_model.blocking">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_text_regression_model.</span></span><span class="sig-name descname"><span class="pre">blocking</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">token_dict</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">block_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_max_length</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pad_token_id</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_side</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">truncation_side</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'right'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_text_regression_model.html#blocking"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_text_regression_model.blocking" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_text_regression_model.blocking_text_to_textlist">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_text_regression_model.</span></span><span class="sig-name descname"><span class="pre">blocking_text_to_textlist</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">token_dict</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">block_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_max_length</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pad_token_id</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">padding_side</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">truncation_side</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'right'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_text_regression_model.html#blocking_text_to_textlist"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_text_regression_model.blocking_text_to_textlist" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_text_regression_model.paired_conversation_tokenize_function">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_text_regression_model.</span></span><span class="sig-name descname"><span class="pre">paired_conversation_tokenize_function</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">examples</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">transformers.PreTrainedTokenizerFast</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">column_names</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">conversation_template</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate" title="lmflow.utils.conversation_template.ConversationTemplate"><span class="pre">lmflow.utils.conversation_template.ConversationTemplate</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_text_regression_model.html#paired_conversation_tokenize_function"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_text_regression_model.paired_conversation_tokenize_function" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_text_regression_model.conversation_tokenize_function">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_text_regression_model.</span></span><span class="sig-name descname"><span class="pre">conversation_tokenize_function</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">examples</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">transformers.PreTrainedTokenizerFast</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">column_names</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">conversation_template</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate" title="lmflow.utils.conversation_template.ConversationTemplate"><span class="pre">lmflow.utils.conversation_template.ConversationTemplate</span></a></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_text_regression_model.html#conversation_tokenize_function"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_text_regression_model.conversation_tokenize_function" title="Link to this definition">#</a></dt>
+<dd><p>Handels conversation datasets tokenization</p>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_text_regression_model.tokenize_function">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_text_regression_model.</span></span><span class="sig-name descname"><span class="pre">tokenize_function</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">examples</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">transformers.PreTrainedTokenizerFast</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">column_names</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">label_columns</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenized_column_order</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">add_special_tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_truncation</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_text_regression_model.html#tokenize_function"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_text_regression_model.tokenize_function" title="Link to this definition">#</a></dt>
+<dd><p>Handels text_only and text2text datasets tokenization</p>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.tokenization.hf_text_regression_model.text_to_textlist_tokenize_function">
+<span class="sig-prename descclassname"><span class="pre">lmflow.tokenization.hf_text_regression_model.</span></span><span class="sig-name descname"><span class="pre">text_to_textlist_tokenize_function</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">examples</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">data_args</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="../../args/index.html#lmflow.args.DatasetArguments" title="lmflow.args.DatasetArguments"><span class="pre">lmflow.args.DatasetArguments</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">transformers.PreTrainedTokenizerFast</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">column_names</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">add_special_tokens</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_truncation</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/tokenization/hf_text_regression_model.html#text_to_textlist_tokenize_function"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.tokenization.hf_text_regression_model.text_to_textlist_tokenize_function" title="Link to this definition">#</a></dt>
+<dd><p>For rm inference, and don’t need attn mask and labels.
+NOTE: input_ids here refers to the tokenized input_ids of the input <strong>and</strong> output</p>
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../hf_decoder_model/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.tokenization.hf_decoder_model</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../../utils/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_text_regression_model.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_text_regression_model.tok_logger"><code class="docutils literal notranslate"><span class="pre">tok_logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_text_regression_model.blocking_paired"><code class="docutils literal notranslate"><span class="pre">blocking_paired()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_text_regression_model.blocking"><code class="docutils literal notranslate"><span class="pre">blocking()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_text_regression_model.blocking_text_to_textlist"><code class="docutils literal notranslate"><span class="pre">blocking_text_to_textlist()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_text_regression_model.paired_conversation_tokenize_function"><code class="docutils literal notranslate"><span class="pre">paired_conversation_tokenize_function()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_text_regression_model.conversation_tokenize_function"><code class="docutils literal notranslate"><span class="pre">conversation_tokenize_function()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_text_regression_model.tokenize_function"><code class="docutils literal notranslate"><span class="pre">tokenize_function()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.tokenization.hf_text_regression_model.text_to_textlist_tokenize_function"><code class="docutils literal notranslate"><span class="pre">text_to_textlist_tokenize_function()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/tokenization/hf_text_regression_model/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/tokenization/index.html b/autoapi/lmflow/tokenization/index.html
new file mode 100644
index 000000000..1b9ab1590
--- /dev/null
+++ b/autoapi/lmflow/tokenization/index.html
@@ -0,0 +1,638 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.tokenization &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/tokenization/index';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+    <link rel="next" title="lmflow.tokenization.hf_decoder_model" href="hf_decoder_model/index.html" />
+    <link rel="prev" title="lmflow.pipeline.vllm_inferencer" href="../pipeline/vllm_inferencer/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="current reference internal" href="#">lmflow.tokenization</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.tokenization</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.tokenization">
+<span id="lmflow-tokenization"></span><h1>lmflow.tokenization<a class="headerlink" href="#module-lmflow.tokenization" title="Link to this heading">#</a></h1>
+<section id="submodules">
+<h2>Submodules<a class="headerlink" href="#submodules" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l1"><a class="reference internal" href="hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../pipeline/vllm_inferencer/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.pipeline.vllm_inferencer</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="hf_decoder_model/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.tokenization.hf_decoder_model</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#submodules">Submodules</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../_sources/autoapi/lmflow/tokenization/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/common/index.html b/autoapi/lmflow/utils/common/index.html
new file mode 100644
index 000000000..881f9aff8
--- /dev/null
+++ b/autoapi/lmflow/utils/common/index.html
@@ -0,0 +1,777 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.common &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/common/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.utils.constants" href="../constants/index.html" />
+    <link rel="prev" title="lmflow.utils" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.common</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.common">
+<span id="lmflow-utils-common"></span><h1>lmflow.utils.common<a class="headerlink" href="#module-lmflow.utils.common" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.common.logger" title="lmflow.utils.common.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.common.make_shell_args_from_dataclass" title="lmflow.utils.common.make_shell_args_from_dataclass"><code class="xref py py-obj docutils literal notranslate"><span class="pre">make_shell_args_from_dataclass</span></code></a>(→ Union[str, List[str]])</p></td>
+<td><p>Return a string or a list of strings that can be used as shell arguments.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.common.create_copied_dataclass" title="lmflow.utils.common.create_copied_dataclass"><code class="xref py py-obj docutils literal notranslate"><span class="pre">create_copied_dataclass</span></code></a>(original_dataclass, ...[, ...])</p></td>
+<td><p>Create a copied dataclass with new field names and default values.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.common.remove_dataclass_attr_prefix" title="lmflow.utils.common.remove_dataclass_attr_prefix"><code class="xref py py-obj docutils literal notranslate"><span class="pre">remove_dataclass_attr_prefix</span></code></a>(→ Dict)</p></td>
+<td><p>Remove the prefix from the attribute names of a dataclass instance.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.common.add_dataclass_attr_prefix" title="lmflow.utils.common.add_dataclass_attr_prefix"><code class="xref py py-obj docutils literal notranslate"><span class="pre">add_dataclass_attr_prefix</span></code></a>(→ Dict)</p></td>
+<td><p>Add the prefix to the attribute names of a dataclass instance.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.common.print_banner" title="lmflow.utils.common.print_banner"><code class="xref py py-obj docutils literal notranslate"><span class="pre">print_banner</span></code></a>(message)</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.common.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.common.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/common.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.common.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.common.make_shell_args_from_dataclass">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.common.</span></span><span class="sig-name descname"><span class="pre">make_shell_args_from_dataclass</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dataclass_objects</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">format</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'subprocess'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">skip_default</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ignored_args_list</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/common.html#make_shell_args_from_dataclass"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.common.make_shell_args_from_dataclass" title="Link to this definition">#</a></dt>
+<dd><p>Return a string or a list of strings that can be used as shell arguments.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>dataclass_objects</strong><span class="classifier">List</span></dt><dd><p>A list of dataclass objects.</p>
+</dd>
+<dt><strong>format</strong><span class="classifier">str, optional</span></dt><dd><p>Return format, can be “shell” or “subprocess”, by default “subprocess”.</p>
+</dd>
+<dt><strong>skip_default</strong><span class="classifier">bool, optional</span></dt><dd><p>Whether to skip attributes with default values, by default True.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>Union[str, List[str]]</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.common.create_copied_dataclass">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.common.</span></span><span class="sig-name descname"><span class="pre">create_copied_dataclass</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">original_dataclass</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">field_prefix</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">class_prefix</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">new_default</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Dict</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/common.html#create_copied_dataclass"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.common.create_copied_dataclass" title="Link to this definition">#</a></dt>
+<dd><p>Create a copied dataclass with new field names and default values.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>original_dataclass</strong><span class="classifier">dataclass</span></dt><dd></dd>
+<dt><strong>field_prefix</strong><span class="classifier">str</span></dt><dd><p>The prefix to add to the <strong>field</strong> names of the copied dataclass.</p>
+</dd>
+<dt><strong>class_prefix</strong><span class="classifier">str</span></dt><dd><p>The prefix to add to the <strong>class</strong> name of the copied dataclass.</p>
+</dd>
+<dt><strong>new_default</strong><span class="classifier">Dict, optional</span></dt><dd><p>The new default values for the copied dataclass. When None, the
+default values of the original dataclass are used.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>dataclass</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.common.remove_dataclass_attr_prefix">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.common.</span></span><span class="sig-name descname"><span class="pre">remove_dataclass_attr_prefix</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data_instance</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">prefix</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/common.html#remove_dataclass_attr_prefix"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.common.remove_dataclass_attr_prefix" title="Link to this definition">#</a></dt>
+<dd><p>Remove the prefix from the attribute names of a dataclass instance.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>data_instance</strong><span class="classifier">dataclass</span></dt><dd></dd>
+<dt><strong>prefix</strong><span class="classifier">str</span></dt><dd><p>The prefix to remove from the attribute names of the dataclass instance.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>Dict</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.common.add_dataclass_attr_prefix">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.common.</span></span><span class="sig-name descname"><span class="pre">add_dataclass_attr_prefix</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data_instance</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">prefix</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Dict</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/common.html#add_dataclass_attr_prefix"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.common.add_dataclass_attr_prefix" title="Link to this definition">#</a></dt>
+<dd><p>Add the prefix to the attribute names of a dataclass instance.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>data_instance</strong><span class="classifier">dataclass</span></dt><dd></dd>
+<dt><strong>prefix</strong><span class="classifier">str</span></dt><dd><p>The prefix to add to the attribute names of the dataclass instance.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>Dict</dt><dd></dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.common.print_banner">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.common.</span></span><span class="sig-name descname"><span class="pre">print_banner</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">message</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/common.html#print_banner"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.common.print_banner" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../constants/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.constants</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.common.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.common.make_shell_args_from_dataclass"><code class="docutils literal notranslate"><span class="pre">make_shell_args_from_dataclass()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.common.create_copied_dataclass"><code class="docutils literal notranslate"><span class="pre">create_copied_dataclass()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.common.remove_dataclass_attr_prefix"><code class="docutils literal notranslate"><span class="pre">remove_dataclass_attr_prefix()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.common.add_dataclass_attr_prefix"><code class="docutils literal notranslate"><span class="pre">add_dataclass_attr_prefix()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.common.print_banner"><code class="docutils literal notranslate"><span class="pre">print_banner()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/utils/common/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/constants/index.html b/autoapi/lmflow/utils/constants/index.html
new file mode 100644
index 000000000..9f28189dd
--- /dev/null
+++ b/autoapi/lmflow/utils/constants/index.html
@@ -0,0 +1,908 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.constants &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/constants/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.utils.conversation_template" href="../conversation_template/index.html" />
+    <link rel="prev" title="lmflow.utils.common" href="../common/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.constants</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.constants">
+<span id="lmflow-utils-constants"></span><h1>lmflow.utils.constants<a class="headerlink" href="#module-lmflow.utils.constants" title="Link to this heading">#</a></h1>
+<p>Commonly used constants.</p>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.TEXT_ONLY_DATASET_DESCRIPTION" title="lmflow.utils.constants.TEXT_ONLY_DATASET_DESCRIPTION"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TEXT_ONLY_DATASET_DESCRIPTION</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION" title="lmflow.utils.constants.TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION" title="lmflow.utils.constants.PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.TEXT_ONLY_DATASET_DETAILS" title="lmflow.utils.constants.TEXT_ONLY_DATASET_DETAILS"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TEXT_ONLY_DATASET_DETAILS</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.TEXT2TEXT_DATASET_DESCRIPTION" title="lmflow.utils.constants.TEXT2TEXT_DATASET_DESCRIPTION"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TEXT2TEXT_DATASET_DESCRIPTION</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.CONVERSATION_DATASET_DESCRIPTION" title="lmflow.utils.constants.CONVERSATION_DATASET_DESCRIPTION"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CONVERSATION_DATASET_DESCRIPTION</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.PAIRED_CONVERSATION_DATASET_DESCRIPTION" title="lmflow.utils.constants.PAIRED_CONVERSATION_DATASET_DESCRIPTION"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PAIRED_CONVERSATION_DATASET_DESCRIPTION</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.TEXT_TO_TEXTLIST_DATASET_DESCRIPTION" title="lmflow.utils.constants.TEXT_TO_TEXTLIST_DATASET_DESCRIPTION"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TEXT_TO_TEXTLIST_DATASET_DESCRIPTION</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.TEXT2TEXT_DATASET_DETAILS" title="lmflow.utils.constants.TEXT2TEXT_DATASET_DETAILS"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TEXT2TEXT_DATASET_DETAILS</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.FLOAT_ONLY_DATASET_DESCRIPTION" title="lmflow.utils.constants.FLOAT_ONLY_DATASET_DESCRIPTION"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FLOAT_ONLY_DATASET_DESCRIPTION</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.TEXT_ONLY_DATASET_LONG_DESCRITION" title="lmflow.utils.constants.TEXT_ONLY_DATASET_LONG_DESCRITION"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TEXT_ONLY_DATASET_LONG_DESCRITION</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.TEXT2TEXT_DATASET_LONG_DESCRITION" title="lmflow.utils.constants.TEXT2TEXT_DATASET_LONG_DESCRITION"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TEXT2TEXT_DATASET_LONG_DESCRITION</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.DATASET_DESCRIPTION_MAP" title="lmflow.utils.constants.DATASET_DESCRIPTION_MAP"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DATASET_DESCRIPTION_MAP</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.INSTANCE_FIELDS_MAP" title="lmflow.utils.constants.INSTANCE_FIELDS_MAP"><code class="xref py py-obj docutils literal notranslate"><span class="pre">INSTANCE_FIELDS_MAP</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.CONVERSATION_ROLE_NAMES" title="lmflow.utils.constants.CONVERSATION_ROLE_NAMES"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CONVERSATION_ROLE_NAMES</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.CONTROLLER_HEART_BEAT_EXPIRATION" title="lmflow.utils.constants.CONTROLLER_HEART_BEAT_EXPIRATION"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CONTROLLER_HEART_BEAT_EXPIRATION</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.WORKER_HEART_BEAT_INTERVAL" title="lmflow.utils.constants.WORKER_HEART_BEAT_INTERVAL"><code class="xref py py-obj docutils literal notranslate"><span class="pre">WORKER_HEART_BEAT_INTERVAL</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.LOGDIR" title="lmflow.utils.constants.LOGDIR"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LOGDIR</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.IGNORE_INDEX" title="lmflow.utils.constants.IGNORE_INDEX"><code class="xref py py-obj docutils literal notranslate"><span class="pre">IGNORE_INDEX</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.IMAGE_TOKEN_INDEX" title="lmflow.utils.constants.IMAGE_TOKEN_INDEX"><code class="xref py py-obj docutils literal notranslate"><span class="pre">IMAGE_TOKEN_INDEX</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.DEFAULT_IMAGE_TOKEN" title="lmflow.utils.constants.DEFAULT_IMAGE_TOKEN"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DEFAULT_IMAGE_TOKEN</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.DEFAULT_IMAGE_PATCH_TOKEN" title="lmflow.utils.constants.DEFAULT_IMAGE_PATCH_TOKEN"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DEFAULT_IMAGE_PATCH_TOKEN</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.DEFAULT_IM_START_TOKEN" title="lmflow.utils.constants.DEFAULT_IM_START_TOKEN"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DEFAULT_IM_START_TOKEN</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.DEFAULT_IM_END_TOKEN" title="lmflow.utils.constants.DEFAULT_IM_END_TOKEN"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DEFAULT_IM_END_TOKEN</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.LMFLOW_LORA_TARGET_MODULES_MAPPING" title="lmflow.utils.constants.LMFLOW_LORA_TARGET_MODULES_MAPPING"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LMFLOW_LORA_TARGET_MODULES_MAPPING</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG" title="lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG"><code class="xref py py-obj docutils literal notranslate"><span class="pre">MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.RETURN_CODE_ERROR_BUFFER" title="lmflow.utils.constants.RETURN_CODE_ERROR_BUFFER"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RETURN_CODE_ERROR_BUFFER</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE" title="lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.constants.MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE" title="lmflow.utils.constants.MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.TEXT_ONLY_DATASET_DESCRIPTION">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">TEXT_ONLY_DATASET_DESCRIPTION</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">''</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#TEXT_ONLY_DATASET_DESCRIPTION"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.TEXT_ONLY_DATASET_DESCRIPTION" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">''</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">''</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.TEXT_ONLY_DATASET_DETAILS">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">TEXT_ONLY_DATASET_DETAILS</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">''</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#TEXT_ONLY_DATASET_DETAILS"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.TEXT_ONLY_DATASET_DETAILS" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.TEXT2TEXT_DATASET_DESCRIPTION">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">TEXT2TEXT_DATASET_DESCRIPTION</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">''</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#TEXT2TEXT_DATASET_DESCRIPTION"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.TEXT2TEXT_DATASET_DESCRIPTION" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.CONVERSATION_DATASET_DESCRIPTION">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">CONVERSATION_DATASET_DESCRIPTION</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">''</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#CONVERSATION_DATASET_DESCRIPTION"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.CONVERSATION_DATASET_DESCRIPTION" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.PAIRED_CONVERSATION_DATASET_DESCRIPTION">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">PAIRED_CONVERSATION_DATASET_DESCRIPTION</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">''</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#PAIRED_CONVERSATION_DATASET_DESCRIPTION"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.PAIRED_CONVERSATION_DATASET_DESCRIPTION" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.TEXT_TO_TEXTLIST_DATASET_DESCRIPTION">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">TEXT_TO_TEXTLIST_DATASET_DESCRIPTION</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">''</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#TEXT_TO_TEXTLIST_DATASET_DESCRIPTION"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.TEXT_TO_TEXTLIST_DATASET_DESCRIPTION" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.TEXT2TEXT_DATASET_DETAILS">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">TEXT2TEXT_DATASET_DETAILS</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">''</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#TEXT2TEXT_DATASET_DETAILS"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.TEXT2TEXT_DATASET_DETAILS" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.FLOAT_ONLY_DATASET_DESCRIPTION">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">FLOAT_ONLY_DATASET_DESCRIPTION</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">''</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#FLOAT_ONLY_DATASET_DESCRIPTION"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.FLOAT_ONLY_DATASET_DESCRIPTION" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.TEXT_ONLY_DATASET_LONG_DESCRITION">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">TEXT_ONLY_DATASET_LONG_DESCRITION</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">''</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#TEXT_ONLY_DATASET_LONG_DESCRITION"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.TEXT_ONLY_DATASET_LONG_DESCRITION" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.TEXT2TEXT_DATASET_LONG_DESCRITION">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">TEXT2TEXT_DATASET_LONG_DESCRITION</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">''</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#TEXT2TEXT_DATASET_LONG_DESCRITION"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.TEXT2TEXT_DATASET_LONG_DESCRITION" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.DATASET_DESCRIPTION_MAP">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">DATASET_DESCRIPTION_MAP</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#DATASET_DESCRIPTION_MAP"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.DATASET_DESCRIPTION_MAP" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.INSTANCE_FIELDS_MAP">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">INSTANCE_FIELDS_MAP</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#INSTANCE_FIELDS_MAP"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.INSTANCE_FIELDS_MAP" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.CONVERSATION_ROLE_NAMES">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">CONVERSATION_ROLE_NAMES</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#CONVERSATION_ROLE_NAMES"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.CONVERSATION_ROLE_NAMES" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.CONTROLLER_HEART_BEAT_EXPIRATION">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">CONTROLLER_HEART_BEAT_EXPIRATION</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">30</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#CONTROLLER_HEART_BEAT_EXPIRATION"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.CONTROLLER_HEART_BEAT_EXPIRATION" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.WORKER_HEART_BEAT_INTERVAL">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">WORKER_HEART_BEAT_INTERVAL</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">15</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#WORKER_HEART_BEAT_INTERVAL"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.WORKER_HEART_BEAT_INTERVAL" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.LOGDIR">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">LOGDIR</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'.'</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#LOGDIR"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.LOGDIR" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.IGNORE_INDEX">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">IGNORE_INDEX</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">-100</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#IGNORE_INDEX"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.IGNORE_INDEX" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.IMAGE_TOKEN_INDEX">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">IMAGE_TOKEN_INDEX</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">-200</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#IMAGE_TOKEN_INDEX"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.IMAGE_TOKEN_INDEX" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.DEFAULT_IMAGE_TOKEN">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">DEFAULT_IMAGE_TOKEN</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'&lt;image&gt;'</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#DEFAULT_IMAGE_TOKEN"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.DEFAULT_IMAGE_TOKEN" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.DEFAULT_IMAGE_PATCH_TOKEN">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">DEFAULT_IMAGE_PATCH_TOKEN</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'&lt;im_patch&gt;'</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#DEFAULT_IMAGE_PATCH_TOKEN"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.DEFAULT_IMAGE_PATCH_TOKEN" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.DEFAULT_IM_START_TOKEN">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">DEFAULT_IM_START_TOKEN</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'&lt;im_start&gt;'</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#DEFAULT_IM_START_TOKEN"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.DEFAULT_IM_START_TOKEN" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.DEFAULT_IM_END_TOKEN">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">DEFAULT_IM_END_TOKEN</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'&lt;im_end&gt;'</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#DEFAULT_IM_END_TOKEN"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.DEFAULT_IM_END_TOKEN" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.LMFLOW_LORA_TARGET_MODULES_MAPPING">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">LMFLOW_LORA_TARGET_MODULES_MAPPING</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#LMFLOW_LORA_TARGET_MODULES_MAPPING"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.LMFLOW_LORA_TARGET_MODULES_MAPPING" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'MEMORY_SAFE_VLLM_INFERENCE_DONE'</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.RETURN_CODE_ERROR_BUFFER">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">RETURN_CODE_ERROR_BUFFER</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">[134]</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#RETURN_CODE_ERROR_BUFFER"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.RETURN_CODE_ERROR_BUFFER" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">['OMP_NUM_THREADS',</span> <span class="pre">'LOCAL_RANK',</span> <span class="pre">'RANK',</span> <span class="pre">'GROUP_RANK',</span> <span class="pre">'ROLE_RANK',</span> <span class="pre">'ROLE_NAME',...</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.constants.MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.constants.</span></span><span class="sig-name descname"><span class="pre">MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">['OMP_NUM_THREADS',</span> <span class="pre">'LOCAL_RANK',</span> <span class="pre">'RANK',</span> <span class="pre">'GROUP_RANK',</span> <span class="pre">'ROLE_RANK',</span> <span class="pre">'ROLE_NAME',...</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/constants.html#MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.constants.MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../common/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.common</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../conversation_template/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.TEXT_ONLY_DATASET_DESCRIPTION"><code class="docutils literal notranslate"><span class="pre">TEXT_ONLY_DATASET_DESCRIPTION</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION"><code class="docutils literal notranslate"><span class="pre">TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION"><code class="docutils literal notranslate"><span class="pre">PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.TEXT_ONLY_DATASET_DETAILS"><code class="docutils literal notranslate"><span class="pre">TEXT_ONLY_DATASET_DETAILS</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.TEXT2TEXT_DATASET_DESCRIPTION"><code class="docutils literal notranslate"><span class="pre">TEXT2TEXT_DATASET_DESCRIPTION</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.CONVERSATION_DATASET_DESCRIPTION"><code class="docutils literal notranslate"><span class="pre">CONVERSATION_DATASET_DESCRIPTION</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.PAIRED_CONVERSATION_DATASET_DESCRIPTION"><code class="docutils literal notranslate"><span class="pre">PAIRED_CONVERSATION_DATASET_DESCRIPTION</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.TEXT_TO_TEXTLIST_DATASET_DESCRIPTION"><code class="docutils literal notranslate"><span class="pre">TEXT_TO_TEXTLIST_DATASET_DESCRIPTION</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.TEXT2TEXT_DATASET_DETAILS"><code class="docutils literal notranslate"><span class="pre">TEXT2TEXT_DATASET_DETAILS</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.FLOAT_ONLY_DATASET_DESCRIPTION"><code class="docutils literal notranslate"><span class="pre">FLOAT_ONLY_DATASET_DESCRIPTION</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.TEXT_ONLY_DATASET_LONG_DESCRITION"><code class="docutils literal notranslate"><span class="pre">TEXT_ONLY_DATASET_LONG_DESCRITION</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.TEXT2TEXT_DATASET_LONG_DESCRITION"><code class="docutils literal notranslate"><span class="pre">TEXT2TEXT_DATASET_LONG_DESCRITION</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.DATASET_DESCRIPTION_MAP"><code class="docutils literal notranslate"><span class="pre">DATASET_DESCRIPTION_MAP</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.INSTANCE_FIELDS_MAP"><code class="docutils literal notranslate"><span class="pre">INSTANCE_FIELDS_MAP</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.CONVERSATION_ROLE_NAMES"><code class="docutils literal notranslate"><span class="pre">CONVERSATION_ROLE_NAMES</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.CONTROLLER_HEART_BEAT_EXPIRATION"><code class="docutils literal notranslate"><span class="pre">CONTROLLER_HEART_BEAT_EXPIRATION</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.WORKER_HEART_BEAT_INTERVAL"><code class="docutils literal notranslate"><span class="pre">WORKER_HEART_BEAT_INTERVAL</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.LOGDIR"><code class="docutils literal notranslate"><span class="pre">LOGDIR</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.IGNORE_INDEX"><code class="docutils literal notranslate"><span class="pre">IGNORE_INDEX</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.IMAGE_TOKEN_INDEX"><code class="docutils literal notranslate"><span class="pre">IMAGE_TOKEN_INDEX</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.DEFAULT_IMAGE_TOKEN"><code class="docutils literal notranslate"><span class="pre">DEFAULT_IMAGE_TOKEN</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.DEFAULT_IMAGE_PATCH_TOKEN"><code class="docutils literal notranslate"><span class="pre">DEFAULT_IMAGE_PATCH_TOKEN</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.DEFAULT_IM_START_TOKEN"><code class="docutils literal notranslate"><span class="pre">DEFAULT_IM_START_TOKEN</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.DEFAULT_IM_END_TOKEN"><code class="docutils literal notranslate"><span class="pre">DEFAULT_IM_END_TOKEN</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.LMFLOW_LORA_TARGET_MODULES_MAPPING"><code class="docutils literal notranslate"><span class="pre">LMFLOW_LORA_TARGET_MODULES_MAPPING</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG"><code class="docutils literal notranslate"><span class="pre">MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.RETURN_CODE_ERROR_BUFFER"><code class="docutils literal notranslate"><span class="pre">RETURN_CODE_ERROR_BUFFER</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE"><code class="docutils literal notranslate"><span class="pre">MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.constants.MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE"><code class="docutils literal notranslate"><span class="pre">MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/utils/constants/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/conversation_template/base/index.html b/autoapi/lmflow/utils/conversation_template/base/index.html
new file mode 100644
index 000000000..f3bd1b021
--- /dev/null
+++ b/autoapi/lmflow/utils/conversation_template/base/index.html
@@ -0,0 +1,1153 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.conversation_template.base &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/conversation_template/base/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.conversation_template.chatglm" href="../chatglm/index.html" />
+    <link rel="prev" title="lmflow.utils.conversation_template" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.conversation_template</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.base</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.conversation_template.base">
+<span id="lmflow-utils-conversation-template-base"></span><h1>lmflow.utils.conversation_template.base<a class="headerlink" href="#module-lmflow.utils.conversation_template.base" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.base.logger" title="lmflow.utils.conversation_template.base.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.base.EMPTY_TEMPLATE" title="lmflow.utils.conversation_template.base.EMPTY_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">EMPTY_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.base.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE" title="lmflow.utils.conversation_template.base.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">EMPTY_NO_SPECIAL_TOKENS_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.base.TemplateComponent" title="lmflow.utils.conversation_template.base.TemplateComponent"><code class="xref py py-obj docutils literal notranslate"><span class="pre">TemplateComponent</span></code></a></p></td>
+<td><p>The minimal unit of a template, which can be a token, a string, or a list of tools.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Formatter</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.base.EmptyFormatter" title="lmflow.utils.conversation_template.base.EmptyFormatter"><code class="xref py py-obj docutils literal notranslate"><span class="pre">EmptyFormatter</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.base.StringFormatter" title="lmflow.utils.conversation_template.base.StringFormatter"><code class="xref py py-obj docutils literal notranslate"><span class="pre">StringFormatter</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.base.ListFormatter" title="lmflow.utils.conversation_template.base.ListFormatter"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ListFormatter</span></code></a></p></td>
+<td><p>Helper class that provides a standard way to create an ABC using</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.base.ConversationTemplate" title="lmflow.utils.conversation_template.base.ConversationTemplate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ConversationTemplate</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.base.ConversationTemplateForTool" title="lmflow.utils.conversation_template.base.ConversationTemplateForTool"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ConversationTemplateForTool</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.base.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.TemplateComponent">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.base.</span></span><span class="sig-name descname"><span class="pre">TemplateComponent</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#TemplateComponent"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.TemplateComponent" title="Link to this definition">#</a></dt>
+<dd><p>The minimal unit of a template, which can be a token, a string, or a list of tools.</p>
+<dl class="field-list">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl>
+<dt><strong>type</strong><span class="classifier">Literal[‘token’, ‘token_id’, ‘string’, ‘tools’]</span></dt><dd><ul class="simple">
+<li><p>Type of the component.</p></li>
+<li><p>When the component is a token or a string, the content should be <cite>string</cite>.</p></li>
+</ul>
+<p>The difference between the two is that token will be converted to token ids
+by the tokenizer.convert_tokens_to_ids() method, while string will be directly
+encoded by the tokenizer.encode() method. Specially, since the bos token and eos
+token are frequently used across different templates, we provide the convenience
+to use <cite>‘bos_token’</cite> and <cite>‘eos_token’</cite> to represent the actual bos and eos tokens when
+<cite>type</cite> of the <cite>TemplateComponent</cite> is <cite>token</cite>. For example:</p>
+<p><code class="docutils literal notranslate"><span class="pre">`python</span>
+<span class="pre">TemplateComponent(type='token',</span> <span class="pre">content='bos_token')</span>
+<span class="pre">`</span></code></p>
+<p>After encoding, the content will be replaced by the actual token id of the bos token.
+Please do remember that if you set the <cite>type</cite> to <cite>string</cite>, the tokenizer will try to
+encode the string ‘bos_token’ instead of providing the actual bos token.</p>
+<ul class="simple">
+<li><p>When the component is token_id, the content should be <cite>int</cite> or <cite>List[int]</cite>, and</p></li>
+</ul>
+<p>will be directly appended to the encoded token ids.</p>
+<ul class="simple">
+<li><p>Tools are not supported yet.</p></li>
+</ul>
+</dd>
+<dt><strong>content</strong><span class="classifier">Union[str, int, List[str], List[int]]</span></dt><dd><p>Content of the component.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.TemplateComponent.type">
+<span class="sig-name descname"><span class="pre">type</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">Literal</span><span class="p"><span class="pre">[</span></span><span class="s"><span class="pre">'token'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'token_id'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'string'</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="s"><span class="pre">'tools'</span></span><span class="p"><span class="pre">]</span></span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#TemplateComponent.type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.TemplateComponent.type" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.TemplateComponent.content">
+<span class="sig-name descname"><span class="pre">content</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#TemplateComponent.content"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.TemplateComponent.content" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.TemplateComponent.mask">
+<span class="sig-name descname"><span class="pre">mask</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">True</span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#TemplateComponent.mask"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.TemplateComponent.mask" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.TemplateComponent.__post_init__">
+<span class="sig-name descname"><span class="pre">__post_init__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#TemplateComponent.__post_init__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.TemplateComponent.__post_init__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.TemplateComponent.__repr__">
+<span class="sig-name descname"><span class="pre">__repr__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">str</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#TemplateComponent.__repr__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.TemplateComponent.__repr__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.TemplateComponent.__str__">
+<span class="sig-name descname"><span class="pre">__str__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">str</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#TemplateComponent.__str__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.TemplateComponent.__str__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.Formatter">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.base.</span></span><span class="sig-name descname"><span class="pre">Formatter</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#Formatter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.Formatter" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">abc.ABC</span></code></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.Formatter.template">
+<span class="sig-name descname"><span class="pre">template</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#lmflow.utils.conversation_template.base.TemplateComponent" title="lmflow.utils.conversation_template.base.TemplateComponent"><span class="pre">TemplateComponent</span></a><span class="p"><span class="pre">]</span></span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">[]</span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#Formatter.template"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.Formatter.template" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.Formatter.format">
+<em class="property"><span class="pre">abstract</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">format</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#lmflow.utils.conversation_template.base.TemplateComponent" title="lmflow.utils.conversation_template.base.TemplateComponent"><span class="pre">TemplateComponent</span></a><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#Formatter.format"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.Formatter.format" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.Formatter.has_placeholder">
+<span class="sig-name descname"><span class="pre">has_placeholder</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#Formatter.has_placeholder"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.Formatter.has_placeholder" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.EmptyFormatter">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.base.</span></span><span class="sig-name descname"><span class="pre">EmptyFormatter</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#EmptyFormatter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.EmptyFormatter" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Formatter</span></code></a></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.EmptyFormatter.__post_init__">
+<span class="sig-name descname"><span class="pre">__post_init__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#EmptyFormatter.__post_init__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.EmptyFormatter.__post_init__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.EmptyFormatter.format">
+<span class="sig-name descname"><span class="pre">format</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">list</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#EmptyFormatter.format"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.EmptyFormatter.format" title="Link to this definition">#</a></dt>
+<dd><p>Empty formatter for when no formatting is needed.
+This is useful when user has already applied formatting to the dataset.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Returns<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt>list</dt><dd><p>Original template.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.StringFormatter">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.base.</span></span><span class="sig-name descname"><span class="pre">StringFormatter</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#StringFormatter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.StringFormatter" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Formatter</span></code></a></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.StringFormatter.__post_init__">
+<span class="sig-name descname"><span class="pre">__post_init__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#StringFormatter.__post_init__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.StringFormatter.__post_init__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.StringFormatter.format">
+<span class="sig-name descname"><span class="pre">format</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">list</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#StringFormatter.format"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.StringFormatter.format" title="Link to this definition">#</a></dt>
+<dd><p>Format the string components with the provided keyword arguments.
+Mostly used for formatting system prompt, user and assistant messages.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>**kwargs</strong><span class="classifier">dict</span></dt><dd><p>Keyword arguments containing values to replace in the template components.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>list</dt><dd><p>Formatted template.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ListFormatter">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.base.</span></span><span class="sig-name descname"><span class="pre">ListFormatter</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ListFormatter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ListFormatter" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Formatter</span></code></a></p>
+<p>Helper class that provides a standard way to create an ABC using
+inheritance.</p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ListFormatter.format">
+<span class="sig-name descname"><span class="pre">format</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">list</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ListFormatter.format"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ListFormatter.format" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.base.</span></span><span class="sig-name descname"><span class="pre">ConversationTemplate</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate" title="Link to this definition">#</a></dt>
+<dd><dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.user_formatter">
+<span class="sig-name descname"><span class="pre">user_formatter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><span class="pre">Formatter</span></a></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.user_formatter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.user_formatter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.assistant_formatter">
+<span class="sig-name descname"><span class="pre">assistant_formatter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><span class="pre">Formatter</span></a></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.assistant_formatter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.assistant_formatter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.function_formatter">
+<span class="sig-name descname"><span class="pre">function_formatter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><span class="pre">Formatter</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.function_formatter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.function_formatter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.observation_formatter">
+<span class="sig-name descname"><span class="pre">observation_formatter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><span class="pre">Formatter</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.observation_formatter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.observation_formatter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.system_formatter">
+<span class="sig-name descname"><span class="pre">system_formatter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><span class="pre">Formatter</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.system_formatter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.system_formatter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.force_system">
+<span class="sig-name descname"><span class="pre">force_system</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.force_system"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.force_system" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.tools_formatter">
+<span class="sig-name descname"><span class="pre">tools_formatter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><span class="pre">Formatter</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.tools_formatter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.tools_formatter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.separator">
+<span class="sig-name descname"><span class="pre">separator</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="#lmflow.utils.conversation_template.base.TemplateComponent" title="lmflow.utils.conversation_template.base.TemplateComponent"><span class="pre">TemplateComponent</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.separator"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.separator" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.remove_last_sep">
+<span class="sig-name descname"><span class="pre">remove_last_sep</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.remove_last_sep"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.remove_last_sep" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.special_starter">
+<span class="sig-name descname"><span class="pre">special_starter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="#lmflow.utils.conversation_template.base.TemplateComponent" title="lmflow.utils.conversation_template.base.TemplateComponent"><span class="pre">TemplateComponent</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.special_starter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.special_starter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.special_stopper">
+<span class="sig-name descname"><span class="pre">special_stopper</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="#lmflow.utils.conversation_template.base.TemplateComponent" title="lmflow.utils.conversation_template.base.TemplateComponent"><span class="pre">TemplateComponent</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.special_stopper"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.special_stopper" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.template_name">
+<span class="sig-name descname"><span class="pre">template_name</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.template_name"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.template_name" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.__post_init__">
+<span class="sig-name descname"><span class="pre">__post_init__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.__post_init__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.__post_init__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.encode_conversation">
+<span class="sig-name descname"><span class="pre">encode_conversation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">messages</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">system</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tools</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.encode_conversation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.encode_conversation" title="Link to this definition">#</a></dt>
+<dd><p>Messages here should be guaranteed to be in pairs, with the first message being the user message and the second message being the system message.
+Data example:
+<a href="#id1"><span class="problematic" id="id2">``</span></a><a href="#id3"><span class="problematic" id="id4">`</span></a>json
+{</p>
+<blockquote>
+<div><p>“conversation_id”: 2,
+“system”: “sysinfo1”,
+“tools”: [“tool_1_desc”],
+“messages”: [</p>
+<blockquote>
+<div><dl class="simple">
+<dt>{</dt><dd><p>“role”: “user”,
+“content”: “hi”</p>
+</dd>
+</dl>
+<p>},
+{</p>
+<blockquote>
+<div><p>“role”: “assistant”,
+“content”: “Hello!”</p>
+</div></blockquote>
+<p>}</p>
+</div></blockquote>
+<p>]</p>
+</div></blockquote>
+<section id="id5">
+<h3>}<a class="headerlink" href="#id5" title="Link to this heading">#</a></h3>
+</section>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate._encode">
+<span class="sig-name descname"><span class="pre">_encode</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">messages</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">system</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tools</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate._encode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate._encode" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate._encode_template">
+<span class="sig-name descname"><span class="pre">_encode_template</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">template</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#lmflow.utils.conversation_template.base.TemplateComponent" title="lmflow.utils.conversation_template.base.TemplateComponent"><span class="pre">TemplateComponent</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate._encode_template"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate._encode_template" title="Link to this definition">#</a></dt>
+<dd><p>Encode template components into token ids.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>template</strong><span class="classifier">List[TemplateComponent]</span></dt><dd><p>Formatted template components.</p>
+</dd>
+<dt><strong>tokenizer</strong><span class="classifier">PreTrainedTokenizer</span></dt><dd><p>Tokenizer to convert tokens into token ids.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>List[int]</dt><dd><p>Encoded token ids.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.post_process_pairs">
+<span class="sig-name descname"><span class="pre">post_process_pairs</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">encoded_pairs</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.post_process_pairs"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.post_process_pairs" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.remove_last_separator">
+<span class="sig-name descname"><span class="pre">remove_last_separator</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">encoded_pairs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.remove_last_separator"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.remove_last_separator" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.add_special_starter">
+<span class="sig-name descname"><span class="pre">add_special_starter</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">encoded_pairs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.add_special_starter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.add_special_starter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate.add_special_stopper">
+<span class="sig-name descname"><span class="pre">add_special_stopper</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">encoded_pairs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.add_special_stopper"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate.add_special_stopper" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplate._ensure_id_list">
+<span class="sig-name descname"><span class="pre">_ensure_id_list</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">obj</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate._ensure_id_list"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplate._ensure_id_list" title="Link to this definition">#</a></dt>
+<dd><p>Make sure the object is a list of integers. Useful for handling token ids.</p>
+</dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplateForTool">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.base.</span></span><span class="sig-name descname"><span class="pre">ConversationTemplateForTool</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplateForTool"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplateForTool" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.utils.conversation_template.base.ConversationTemplate" title="lmflow.utils.conversation_template.base.ConversationTemplate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ConversationTemplate</span></code></a></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplateForTool.encode_conversation">
+<span class="sig-name descname"><span class="pre">encode_conversation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">messages</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">system</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tools</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplateForTool.encode_conversation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplateForTool.encode_conversation" title="Link to this definition">#</a></dt>
+<dd><p>Messages here should be guaranteed to be in pairs, with the first message being the user message and the second message being the system message.
+Data example:
+<a href="#id6"><span class="problematic" id="id7">``</span></a><a href="#id8"><span class="problematic" id="id9">`</span></a>json
+{</p>
+<blockquote>
+<div><p>“conversation_id”: 2,
+“system”: “sysinfo1”,
+“tools”: [“tool_1_desc”],
+“messages”: [</p>
+<blockquote>
+<div><dl class="simple">
+<dt>{</dt><dd><p>“role”: “user”,
+“content”: “hi”</p>
+</dd>
+</dl>
+<p>},
+{</p>
+<blockquote>
+<div><p>“role”: “assistant”,
+“content”: “Hello!”</p>
+</div></blockquote>
+<p>}</p>
+</div></blockquote>
+<p>]</p>
+</div></blockquote>
+<section id="id10">
+<h3>}<a class="headerlink" href="#id10" title="Link to this heading">#</a></h3>
+</section>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplateForTool._encode">
+<span class="sig-name descname"><span class="pre">_encode</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">messages</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">system</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tools</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplateForTool._encode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplateForTool._encode" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.ConversationTemplateForTool._encode_template">
+<span class="sig-name descname"><span class="pre">_encode_template</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">template</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="#lmflow.utils.conversation_template.base.TemplateComponent" title="lmflow.utils.conversation_template.base.TemplateComponent"><span class="pre">TemplateComponent</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplateForTool._encode_template"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.ConversationTemplateForTool._encode_template" title="Link to this definition">#</a></dt>
+<dd><p>Encode template components into token ids.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>template</strong><span class="classifier">List[TemplateComponent]</span></dt><dd><p>Formatted template components.</p>
+</dd>
+<dt><strong>tokenizer</strong><span class="classifier">PreTrainedTokenizer</span></dt><dd><p>Tokenizer to convert tokens into token ids.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>List[int]</dt><dd><p>Encoded token ids.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.EMPTY_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.base.</span></span><span class="sig-name descname"><span class="pre">EMPTY_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#EMPTY_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.EMPTY_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.base.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.base.</span></span><span class="sig-name descname"><span class="pre">EMPTY_NO_SPECIAL_TOKENS_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/base.html#EMPTY_NO_SPECIAL_TOKENS_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.base.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../chatglm/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.chatglm</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.TemplateComponent"><code class="docutils literal notranslate"><span class="pre">TemplateComponent</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.TemplateComponent.type"><code class="docutils literal notranslate"><span class="pre">TemplateComponent.type</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.TemplateComponent.content"><code class="docutils literal notranslate"><span class="pre">TemplateComponent.content</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.TemplateComponent.mask"><code class="docutils literal notranslate"><span class="pre">TemplateComponent.mask</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.TemplateComponent.__post_init__"><code class="docutils literal notranslate"><span class="pre">TemplateComponent.__post_init__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.TemplateComponent.__repr__"><code class="docutils literal notranslate"><span class="pre">TemplateComponent.__repr__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.TemplateComponent.__str__"><code class="docutils literal notranslate"><span class="pre">TemplateComponent.__str__()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.Formatter"><code class="docutils literal notranslate"><span class="pre">Formatter</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.Formatter.template"><code class="docutils literal notranslate"><span class="pre">Formatter.template</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.Formatter.format"><code class="docutils literal notranslate"><span class="pre">Formatter.format()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.Formatter.has_placeholder"><code class="docutils literal notranslate"><span class="pre">Formatter.has_placeholder()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.EmptyFormatter"><code class="docutils literal notranslate"><span class="pre">EmptyFormatter</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.EmptyFormatter.__post_init__"><code class="docutils literal notranslate"><span class="pre">EmptyFormatter.__post_init__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.EmptyFormatter.format"><code class="docutils literal notranslate"><span class="pre">EmptyFormatter.format()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.StringFormatter"><code class="docutils literal notranslate"><span class="pre">StringFormatter</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.StringFormatter.__post_init__"><code class="docutils literal notranslate"><span class="pre">StringFormatter.__post_init__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.StringFormatter.format"><code class="docutils literal notranslate"><span class="pre">StringFormatter.format()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ListFormatter"><code class="docutils literal notranslate"><span class="pre">ListFormatter</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ListFormatter.format"><code class="docutils literal notranslate"><span class="pre">ListFormatter.format()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.user_formatter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.user_formatter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.assistant_formatter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.assistant_formatter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.function_formatter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.function_formatter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.observation_formatter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.observation_formatter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.system_formatter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.system_formatter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.force_system"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.force_system</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.tools_formatter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.tools_formatter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.separator"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.separator</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.remove_last_sep"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.remove_last_sep</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.special_starter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.special_starter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.special_stopper"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.special_stopper</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.template_name"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.template_name</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.__post_init__"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.__post_init__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.encode_conversation"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.encode_conversation()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate._encode"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate._encode()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate._encode_template"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate._encode_template()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.post_process_pairs"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.post_process_pairs()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.remove_last_separator"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.remove_last_separator()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.add_special_starter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.add_special_starter()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate.add_special_stopper"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.add_special_stopper()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplate._ensure_id_list"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate._ensure_id_list()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplateForTool"><code class="docutils literal notranslate"><span class="pre">ConversationTemplateForTool</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplateForTool.encode_conversation"><code class="docutils literal notranslate"><span class="pre">ConversationTemplateForTool.encode_conversation()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplateForTool._encode"><code class="docutils literal notranslate"><span class="pre">ConversationTemplateForTool._encode()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.ConversationTemplateForTool._encode_template"><code class="docutils literal notranslate"><span class="pre">ConversationTemplateForTool._encode_template()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.EMPTY_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">EMPTY_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.base.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">EMPTY_NO_SPECIAL_TOKENS_TEMPLATE</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/conversation_template/base/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/conversation_template/chatglm/index.html b/autoapi/lmflow/utils/conversation_template/chatglm/index.html
new file mode 100644
index 000000000..3fbb2c602
--- /dev/null
+++ b/autoapi/lmflow/utils/conversation_template/chatglm/index.html
@@ -0,0 +1,658 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.conversation_template.chatglm &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/conversation_template/chatglm/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.conversation_template.chatml" href="../chatml/index.html" />
+    <link rel="prev" title="lmflow.utils.conversation_template.base" href="../base/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.conversation_template</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.chatglm</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.conversation_template.chatglm">
+<span id="lmflow-utils-conversation-template-chatglm"></span><h1>lmflow.utils.conversation_template.chatglm<a class="headerlink" href="#module-lmflow.utils.conversation_template.chatglm" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.chatglm.CHATGLM3_TEMPLATE" title="lmflow.utils.conversation_template.chatglm.CHATGLM3_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CHATGLM3_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.chatglm.CHATGLM3_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.chatglm.</span></span><span class="sig-name descname"><span class="pre">CHATGLM3_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/chatglm.html#CHATGLM3_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.chatglm.CHATGLM3_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../base/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.base</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../chatml/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.chatml</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.chatglm.CHATGLM3_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">CHATGLM3_TEMPLATE</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/conversation_template/chatglm/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/conversation_template/chatml/index.html b/autoapi/lmflow/utils/conversation_template/chatml/index.html
new file mode 100644
index 000000000..4a11790c9
--- /dev/null
+++ b/autoapi/lmflow/utils/conversation_template/chatml/index.html
@@ -0,0 +1,658 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.conversation_template.chatml &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/conversation_template/chatml/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.conversation_template.deepseek" href="../deepseek/index.html" />
+    <link rel="prev" title="lmflow.utils.conversation_template.chatglm" href="../chatglm/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.conversation_template</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.chatml</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.conversation_template.chatml">
+<span id="lmflow-utils-conversation-template-chatml"></span><h1>lmflow.utils.conversation_template.chatml<a class="headerlink" href="#module-lmflow.utils.conversation_template.chatml" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.chatml.CHATML_TEMPLATE" title="lmflow.utils.conversation_template.chatml.CHATML_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CHATML_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.chatml.CHATML_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.chatml.</span></span><span class="sig-name descname"><span class="pre">CHATML_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/chatml.html#CHATML_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.chatml.CHATML_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../chatglm/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.chatglm</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../deepseek/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.deepseek</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.chatml.CHATML_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">CHATML_TEMPLATE</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/conversation_template/chatml/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/conversation_template/deepseek/index.html b/autoapi/lmflow/utils/conversation_template/deepseek/index.html
new file mode 100644
index 000000000..768dc4016
--- /dev/null
+++ b/autoapi/lmflow/utils/conversation_template/deepseek/index.html
@@ -0,0 +1,658 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.conversation_template.deepseek &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/conversation_template/deepseek/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.conversation_template.gemma" href="../gemma/index.html" />
+    <link rel="prev" title="lmflow.utils.conversation_template.chatml" href="../chatml/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.conversation_template</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.deepseek</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.conversation_template.deepseek">
+<span id="lmflow-utils-conversation-template-deepseek"></span><h1>lmflow.utils.conversation_template.deepseek<a class="headerlink" href="#module-lmflow.utils.conversation_template.deepseek" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.deepseek.DEEPSEEK_TEMPLATE" title="lmflow.utils.conversation_template.deepseek.DEEPSEEK_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DEEPSEEK_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.deepseek.DEEPSEEK_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.deepseek.</span></span><span class="sig-name descname"><span class="pre">DEEPSEEK_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/deepseek.html#DEEPSEEK_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.deepseek.DEEPSEEK_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../chatml/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.chatml</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../gemma/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.gemma</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.deepseek.DEEPSEEK_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">DEEPSEEK_TEMPLATE</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/conversation_template/deepseek/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/conversation_template/gemma/index.html b/autoapi/lmflow/utils/conversation_template/gemma/index.html
new file mode 100644
index 000000000..085c25766
--- /dev/null
+++ b/autoapi/lmflow/utils/conversation_template/gemma/index.html
@@ -0,0 +1,694 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.conversation_template.gemma &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/conversation_template/gemma/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.conversation_template.hymba" href="../hymba/index.html" />
+    <link rel="prev" title="lmflow.utils.conversation_template.deepseek" href="../deepseek/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.conversation_template</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.gemma</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.conversation_template.gemma">
+<span id="lmflow-utils-conversation-template-gemma"></span><h1>lmflow.utils.conversation_template.gemma<a class="headerlink" href="#module-lmflow.utils.conversation_template.gemma" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.gemma.logger" title="lmflow.utils.conversation_template.gemma.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.gemma.GEMMA_TEMPLATE" title="lmflow.utils.conversation_template.gemma.GEMMA_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">GEMMA_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.gemma.GemmaConversationTemplate" title="lmflow.utils.conversation_template.gemma.GemmaConversationTemplate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">GemmaConversationTemplate</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.gemma.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.gemma.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/gemma.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.gemma.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.gemma.GemmaConversationTemplate">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.gemma.</span></span><span class="sig-name descname"><span class="pre">GemmaConversationTemplate</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/gemma.html#GemmaConversationTemplate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.gemma.GemmaConversationTemplate" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate" title="lmflow.utils.conversation_template.base.ConversationTemplate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.utils.conversation_template.base.ConversationTemplate</span></code></a></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.gemma.GemmaConversationTemplate.encode_conversation">
+<span class="sig-name descname"><span class="pre">encode_conversation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/gemma.html#GemmaConversationTemplate.encode_conversation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.gemma.GemmaConversationTemplate.encode_conversation" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.gemma.GEMMA_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.gemma.</span></span><span class="sig-name descname"><span class="pre">GEMMA_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/gemma.html#GEMMA_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.gemma.GEMMA_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../deepseek/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.deepseek</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../hymba/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.hymba</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.gemma.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.gemma.GemmaConversationTemplate"><code class="docutils literal notranslate"><span class="pre">GemmaConversationTemplate</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.gemma.GemmaConversationTemplate.encode_conversation"><code class="docutils literal notranslate"><span class="pre">GemmaConversationTemplate.encode_conversation()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.gemma.GEMMA_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">GEMMA_TEMPLATE</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/conversation_template/gemma/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/conversation_template/hymba/index.html b/autoapi/lmflow/utils/conversation_template/hymba/index.html
new file mode 100644
index 000000000..01026e936
--- /dev/null
+++ b/autoapi/lmflow/utils/conversation_template/hymba/index.html
@@ -0,0 +1,713 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.conversation_template.hymba &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/conversation_template/hymba/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.conversation_template.internlm" href="../internlm/index.html" />
+    <link rel="prev" title="lmflow.utils.conversation_template.gemma" href="../gemma/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.conversation_template</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.hymba</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.conversation_template.hymba">
+<span id="lmflow-utils-conversation-template-hymba"></span><h1>lmflow.utils.conversation_template.hymba<a class="headerlink" href="#module-lmflow.utils.conversation_template.hymba" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.hymba.HYMBA_TEMPLATE" title="lmflow.utils.conversation_template.hymba.HYMBA_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">HYMBA_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.hymba.HymbaConversationTemplate" title="lmflow.utils.conversation_template.hymba.HymbaConversationTemplate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">HymbaConversationTemplate</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.hymba.HymbaConversationTemplate">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.hymba.</span></span><span class="sig-name descname"><span class="pre">HymbaConversationTemplate</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/hymba.html#HymbaConversationTemplate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.hymba.HymbaConversationTemplate" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base/index.html#lmflow.utils.conversation_template.base.ConversationTemplateForTool" title="lmflow.utils.conversation_template.base.ConversationTemplateForTool"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.utils.conversation_template.base.ConversationTemplateForTool</span></code></a></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.hymba.HymbaConversationTemplate.encode_conversation">
+<span class="sig-name descname"><span class="pre">encode_conversation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">messages</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">system</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tools</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/hymba.html#HymbaConversationTemplate.encode_conversation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.hymba.HymbaConversationTemplate.encode_conversation" title="Link to this definition">#</a></dt>
+<dd><p>Messages here should be guaranteed to be in pairs, with the first message being the user message and the second message being the system message.
+Data example:
+<a href="#id1"><span class="problematic" id="id2">``</span></a><a href="#id3"><span class="problematic" id="id4">`</span></a>json
+{</p>
+<blockquote>
+<div><p>“conversation_id”: 2,
+“system”: “sysinfo1”,
+“tools”: [“tool_1_desc”],
+“messages”: [</p>
+<blockquote>
+<div><dl class="simple">
+<dt>{</dt><dd><p>“role”: “user”,
+“content”: “hi”</p>
+</dd>
+</dl>
+<p>},
+{</p>
+<blockquote>
+<div><p>“role”: “assistant”,
+“content”: “Hello!”</p>
+</div></blockquote>
+<p>}</p>
+</div></blockquote>
+<p>]</p>
+</div></blockquote>
+<section id="id5">
+<h3>}<a class="headerlink" href="#id5" title="Link to this heading">#</a></h3>
+</section>
+</dd></dl>
+
+</dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.hymba.HYMBA_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.hymba.</span></span><span class="sig-name descname"><span class="pre">HYMBA_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/hymba.html#HYMBA_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.hymba.HYMBA_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../gemma/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.gemma</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../internlm/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.internlm</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.hymba.HymbaConversationTemplate"><code class="docutils literal notranslate"><span class="pre">HymbaConversationTemplate</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.hymba.HymbaConversationTemplate.encode_conversation"><code class="docutils literal notranslate"><span class="pre">HymbaConversationTemplate.encode_conversation()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.hymba.HYMBA_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">HYMBA_TEMPLATE</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/conversation_template/hymba/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/conversation_template/index.html b/autoapi/lmflow/utils/conversation_template/index.html
new file mode 100644
index 000000000..60e55cb16
--- /dev/null
+++ b/autoapi/lmflow/utils/conversation_template/index.html
@@ -0,0 +1,1086 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.conversation_template &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/conversation_template/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.utils.conversation_template.base" href="base/index.html" />
+    <link rel="prev" title="lmflow.utils.constants" href="../constants/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 current active has-children"><a class="current reference internal" href="#">lmflow.utils.conversation_template</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.conversation_template">
+<span id="lmflow-utils-conversation-template"></span><h1>lmflow.utils.conversation_template<a class="headerlink" href="#module-lmflow.utils.conversation_template" title="Link to this heading">#</a></h1>
+<section id="submodules">
+<h2>Submodules<a class="headerlink" href="#submodules" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l1"><a class="reference internal" href="chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l1"><a class="reference internal" href="chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l1"><a class="reference internal" href="deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l1"><a class="reference internal" href="gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l1"><a class="reference internal" href="hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l1"><a class="reference internal" href="internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l1"><a class="reference internal" href="llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l1"><a class="reference internal" href="phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l1"><a class="reference internal" href="qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l1"><a class="reference internal" href="yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l1"><a class="reference internal" href="zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</div>
+</section>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.EMPTY_TEMPLATE" title="lmflow.utils.conversation_template.EMPTY_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">EMPTY_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE" title="lmflow.utils.conversation_template.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">EMPTY_NO_SPECIAL_TOKENS_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.CHATGLM3_TEMPLATE" title="lmflow.utils.conversation_template.CHATGLM3_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CHATGLM3_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.CHATML_TEMPLATE" title="lmflow.utils.conversation_template.CHATML_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CHATML_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.DEEPSEEK_TEMPLATE" title="lmflow.utils.conversation_template.DEEPSEEK_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">DEEPSEEK_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.GEMMA_TEMPLATE" title="lmflow.utils.conversation_template.GEMMA_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">GEMMA_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.HYMBA_TEMPLATE" title="lmflow.utils.conversation_template.HYMBA_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">HYMBA_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.INTERNLM2_TEMPLATE" title="lmflow.utils.conversation_template.INTERNLM2_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">INTERNLM2_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.LLAMA2_TEMPLATE" title="lmflow.utils.conversation_template.LLAMA2_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LLAMA2_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.LLAMA3_TEMPLATE" title="lmflow.utils.conversation_template.LLAMA3_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LLAMA3_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.LLAMA3_TEMPLATE_FOR_TOOL" title="lmflow.utils.conversation_template.LLAMA3_TEMPLATE_FOR_TOOL"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LLAMA3_TEMPLATE_FOR_TOOL</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.PHI3_TEMPLATE" title="lmflow.utils.conversation_template.PHI3_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PHI3_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.QWEN2_TEMPLATE" title="lmflow.utils.conversation_template.QWEN2_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">QWEN2_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.QWEN2_TEMPLATE_FOR_TOOL" title="lmflow.utils.conversation_template.QWEN2_TEMPLATE_FOR_TOOL"><code class="xref py py-obj docutils literal notranslate"><span class="pre">QWEN2_TEMPLATE_FOR_TOOL</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.YI1_5_TEMPLATE" title="lmflow.utils.conversation_template.YI1_5_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">YI1_5_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.ZEPHYR_TEMPLATE" title="lmflow.utils.conversation_template.ZEPHYR_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ZEPHYR_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.PRESET_TEMPLATES" title="lmflow.utils.conversation_template.PRESET_TEMPLATES"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PRESET_TEMPLATES</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.ConversationTemplate" title="lmflow.utils.conversation_template.ConversationTemplate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ConversationTemplate</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.ConversationTemplateForTool" title="lmflow.utils.conversation_template.ConversationTemplateForTool"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ConversationTemplateForTool</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="package-contents">
+<h2>Package Contents<a class="headerlink" href="#package-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.EMPTY_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">EMPTY_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#EMPTY_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.EMPTY_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">EMPTY_NO_SPECIAL_TOKENS_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#EMPTY_NO_SPECIAL_TOKENS_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">ConversationTemplate</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate" title="Link to this definition">#</a></dt>
+<dd><dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.user_formatter">
+<span class="sig-name descname"><span class="pre">user_formatter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="base/index.html#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><span class="pre">Formatter</span></a></em><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.user_formatter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.assistant_formatter">
+<span class="sig-name descname"><span class="pre">assistant_formatter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="base/index.html#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><span class="pre">Formatter</span></a></em><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.assistant_formatter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.function_formatter">
+<span class="sig-name descname"><span class="pre">function_formatter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="base/index.html#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><span class="pre">Formatter</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.function_formatter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.observation_formatter">
+<span class="sig-name descname"><span class="pre">observation_formatter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="base/index.html#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><span class="pre">Formatter</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.observation_formatter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.system_formatter">
+<span class="sig-name descname"><span class="pre">system_formatter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="base/index.html#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><span class="pre">Formatter</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.system_formatter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.force_system">
+<span class="sig-name descname"><span class="pre">force_system</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.force_system" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.tools_formatter">
+<span class="sig-name descname"><span class="pre">tools_formatter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="base/index.html#lmflow.utils.conversation_template.base.Formatter" title="lmflow.utils.conversation_template.base.Formatter"><span class="pre">Formatter</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.tools_formatter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.separator">
+<span class="sig-name descname"><span class="pre">separator</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="base/index.html#lmflow.utils.conversation_template.base.TemplateComponent" title="lmflow.utils.conversation_template.base.TemplateComponent"><span class="pre">TemplateComponent</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.separator" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.remove_last_sep">
+<span class="sig-name descname"><span class="pre">remove_last_sep</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.remove_last_sep" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.special_starter">
+<span class="sig-name descname"><span class="pre">special_starter</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="base/index.html#lmflow.utils.conversation_template.base.TemplateComponent" title="lmflow.utils.conversation_template.base.TemplateComponent"><span class="pre">TemplateComponent</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.special_starter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.special_stopper">
+<span class="sig-name descname"><span class="pre">special_stopper</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="base/index.html#lmflow.utils.conversation_template.base.TemplateComponent" title="lmflow.utils.conversation_template.base.TemplateComponent"><span class="pre">TemplateComponent</span></a><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.special_stopper" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.template_name">
+<span class="sig-name descname"><span class="pre">template_name</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.template_name" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.__post_init__">
+<span class="sig-name descname"><span class="pre">__post_init__</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.__post_init__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.__post_init__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.encode_conversation">
+<span class="sig-name descname"><span class="pre">encode_conversation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">messages</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">system</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tools</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.encode_conversation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.encode_conversation" title="Link to this definition">#</a></dt>
+<dd><p>Messages here should be guaranteed to be in pairs, with the first message being the user message and the second message being the system message.
+Data example:
+<a href="#id1"><span class="problematic" id="id2">``</span></a><a href="#id3"><span class="problematic" id="id4">`</span></a>json
+{</p>
+<blockquote>
+<div><p>“conversation_id”: 2,
+“system”: “sysinfo1”,
+“tools”: [“tool_1_desc”],
+“messages”: [</p>
+<blockquote>
+<div><dl class="simple">
+<dt>{</dt><dd><p>“role”: “user”,
+“content”: “hi”</p>
+</dd>
+</dl>
+<p>},
+{</p>
+<blockquote>
+<div><p>“role”: “assistant”,
+“content”: “Hello!”</p>
+</div></blockquote>
+<p>}</p>
+</div></blockquote>
+<p>]</p>
+</div></blockquote>
+<section id="id5">
+<h3>}<a class="headerlink" href="#id5" title="Link to this heading">#</a></h3>
+</section>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate._encode">
+<span class="sig-name descname"><span class="pre">_encode</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">messages</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">system</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tools</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate._encode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate._encode" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate._encode_template">
+<span class="sig-name descname"><span class="pre">_encode_template</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">template</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="base/index.html#lmflow.utils.conversation_template.base.TemplateComponent" title="lmflow.utils.conversation_template.base.TemplateComponent"><span class="pre">TemplateComponent</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate._encode_template"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate._encode_template" title="Link to this definition">#</a></dt>
+<dd><p>Encode template components into token ids.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>template</strong><span class="classifier">List[TemplateComponent]</span></dt><dd><p>Formatted template components.</p>
+</dd>
+<dt><strong>tokenizer</strong><span class="classifier">PreTrainedTokenizer</span></dt><dd><p>Tokenizer to convert tokens into token ids.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>List[int]</dt><dd><p>Encoded token ids.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.post_process_pairs">
+<span class="sig-name descname"><span class="pre">post_process_pairs</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">encoded_pairs</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.post_process_pairs"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.post_process_pairs" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.remove_last_separator">
+<span class="sig-name descname"><span class="pre">remove_last_separator</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">encoded_pairs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.remove_last_separator"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.remove_last_separator" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.add_special_starter">
+<span class="sig-name descname"><span class="pre">add_special_starter</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">encoded_pairs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.add_special_starter"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.add_special_starter" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate.add_special_stopper">
+<span class="sig-name descname"><span class="pre">add_special_stopper</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">encoded_pairs</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate.add_special_stopper"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate.add_special_stopper" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplate._ensure_id_list">
+<span class="sig-name descname"><span class="pre">_ensure_id_list</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">obj</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplate._ensure_id_list"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplate._ensure_id_list" title="Link to this definition">#</a></dt>
+<dd><p>Make sure the object is a list of integers. Useful for handling token ids.</p>
+</dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplateForTool">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">ConversationTemplateForTool</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplateForTool"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplateForTool" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.utils.conversation_template.ConversationTemplate" title="lmflow.utils.conversation_template.ConversationTemplate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ConversationTemplate</span></code></a></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplateForTool.encode_conversation">
+<span class="sig-name descname"><span class="pre">encode_conversation</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">messages</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">system</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tools</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplateForTool.encode_conversation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplateForTool.encode_conversation" title="Link to this definition">#</a></dt>
+<dd><p>Messages here should be guaranteed to be in pairs, with the first message being the user message and the second message being the system message.
+Data example:
+<a href="#id6"><span class="problematic" id="id7">``</span></a><a href="#id8"><span class="problematic" id="id9">`</span></a>json
+{</p>
+<blockquote>
+<div><p>“conversation_id”: 2,
+“system”: “sysinfo1”,
+“tools”: [“tool_1_desc”],
+“messages”: [</p>
+<blockquote>
+<div><dl class="simple">
+<dt>{</dt><dd><p>“role”: “user”,
+“content”: “hi”</p>
+</dd>
+</dl>
+<p>},
+{</p>
+<blockquote>
+<div><p>“role”: “assistant”,
+“content”: “Hello!”</p>
+</div></blockquote>
+<p>}</p>
+</div></blockquote>
+<p>]</p>
+</div></blockquote>
+<section id="id10">
+<h3>}<a class="headerlink" href="#id10" title="Link to this heading">#</a></h3>
+</section>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplateForTool._encode">
+<span class="sig-name descname"><span class="pre">_encode</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">messages</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">system</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tools</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplateForTool._encode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplateForTool._encode" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ConversationTemplateForTool._encode_template">
+<span class="sig-name descname"><span class="pre">_encode_template</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">template</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="base/index.html#lmflow.utils.conversation_template.base.TemplateComponent" title="lmflow.utils.conversation_template.base.TemplateComponent"><span class="pre">TemplateComponent</span></a><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/base.html#ConversationTemplateForTool._encode_template"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ConversationTemplateForTool._encode_template" title="Link to this definition">#</a></dt>
+<dd><p>Encode template components into token ids.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>template</strong><span class="classifier">List[TemplateComponent]</span></dt><dd><p>Formatted template components.</p>
+</dd>
+<dt><strong>tokenizer</strong><span class="classifier">PreTrainedTokenizer</span></dt><dd><p>Tokenizer to convert tokens into token ids.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>List[int]</dt><dd><p>Encoded token ids.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+</dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.CHATGLM3_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">CHATGLM3_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/chatglm.html#CHATGLM3_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.CHATGLM3_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.CHATML_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">CHATML_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/chatml.html#CHATML_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.CHATML_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.DEEPSEEK_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">DEEPSEEK_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/deepseek.html#DEEPSEEK_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.DEEPSEEK_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.GEMMA_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">GEMMA_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/gemma.html#GEMMA_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.GEMMA_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.HYMBA_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">HYMBA_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/hymba.html#HYMBA_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.HYMBA_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.INTERNLM2_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">INTERNLM2_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/internlm.html#INTERNLM2_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.INTERNLM2_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.LLAMA2_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">LLAMA2_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/llama.html#LLAMA2_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.LLAMA2_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.LLAMA3_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">LLAMA3_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/llama.html#LLAMA3_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.LLAMA3_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.LLAMA3_TEMPLATE_FOR_TOOL">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">LLAMA3_TEMPLATE_FOR_TOOL</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/llama.html#LLAMA3_TEMPLATE_FOR_TOOL"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.LLAMA3_TEMPLATE_FOR_TOOL" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.PHI3_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">PHI3_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/phi.html#PHI3_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.PHI3_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.QWEN2_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">QWEN2_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/qwen.html#QWEN2_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.QWEN2_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.QWEN2_TEMPLATE_FOR_TOOL">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">QWEN2_TEMPLATE_FOR_TOOL</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/qwen.html#QWEN2_TEMPLATE_FOR_TOOL"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.QWEN2_TEMPLATE_FOR_TOOL" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.YI1_5_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">YI1_5_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/yi.html#YI1_5_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.YI1_5_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.ZEPHYR_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">ZEPHYR_TEMPLATE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template/zephyr.html#ZEPHYR_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.ZEPHYR_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.PRESET_TEMPLATES">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.</span></span><span class="sig-name descname"><span class="pre">PRESET_TEMPLATES</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/conversation_template.html#PRESET_TEMPLATES"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.PRESET_TEMPLATES" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../constants/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.constants</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="base/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.base</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#submodules">Submodules</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#package-contents">Package Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.EMPTY_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">EMPTY_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">EMPTY_NO_SPECIAL_TOKENS_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.user_formatter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.user_formatter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.assistant_formatter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.assistant_formatter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.function_formatter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.function_formatter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.observation_formatter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.observation_formatter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.system_formatter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.system_formatter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.force_system"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.force_system</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.tools_formatter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.tools_formatter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.separator"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.separator</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.remove_last_sep"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.remove_last_sep</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.special_starter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.special_starter</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.special_stopper"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.special_stopper</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.template_name"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.template_name</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.__post_init__"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.__post_init__()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.encode_conversation"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.encode_conversation()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate._encode"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate._encode()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate._encode_template"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate._encode_template()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.post_process_pairs"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.post_process_pairs()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.remove_last_separator"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.remove_last_separator()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.add_special_starter"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.add_special_starter()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate.add_special_stopper"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate.add_special_stopper()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplate._ensure_id_list"><code class="docutils literal notranslate"><span class="pre">ConversationTemplate._ensure_id_list()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplateForTool"><code class="docutils literal notranslate"><span class="pre">ConversationTemplateForTool</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplateForTool.encode_conversation"><code class="docutils literal notranslate"><span class="pre">ConversationTemplateForTool.encode_conversation()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplateForTool._encode"><code class="docutils literal notranslate"><span class="pre">ConversationTemplateForTool._encode()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ConversationTemplateForTool._encode_template"><code class="docutils literal notranslate"><span class="pre">ConversationTemplateForTool._encode_template()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.CHATGLM3_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">CHATGLM3_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.CHATML_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">CHATML_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.DEEPSEEK_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">DEEPSEEK_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.GEMMA_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">GEMMA_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.HYMBA_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">HYMBA_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.INTERNLM2_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">INTERNLM2_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.LLAMA2_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">LLAMA2_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.LLAMA3_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">LLAMA3_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.LLAMA3_TEMPLATE_FOR_TOOL"><code class="docutils literal notranslate"><span class="pre">LLAMA3_TEMPLATE_FOR_TOOL</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.PHI3_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">PHI3_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.QWEN2_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">QWEN2_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.QWEN2_TEMPLATE_FOR_TOOL"><code class="docutils literal notranslate"><span class="pre">QWEN2_TEMPLATE_FOR_TOOL</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.YI1_5_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">YI1_5_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.ZEPHYR_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">ZEPHYR_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.PRESET_TEMPLATES"><code class="docutils literal notranslate"><span class="pre">PRESET_TEMPLATES</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/utils/conversation_template/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/conversation_template/internlm/index.html b/autoapi/lmflow/utils/conversation_template/internlm/index.html
new file mode 100644
index 000000000..a50ee4771
--- /dev/null
+++ b/autoapi/lmflow/utils/conversation_template/internlm/index.html
@@ -0,0 +1,658 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.conversation_template.internlm &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/conversation_template/internlm/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.conversation_template.llama" href="../llama/index.html" />
+    <link rel="prev" title="lmflow.utils.conversation_template.hymba" href="../hymba/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.conversation_template</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.internlm</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.conversation_template.internlm">
+<span id="lmflow-utils-conversation-template-internlm"></span><h1>lmflow.utils.conversation_template.internlm<a class="headerlink" href="#module-lmflow.utils.conversation_template.internlm" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.internlm.INTERNLM2_TEMPLATE" title="lmflow.utils.conversation_template.internlm.INTERNLM2_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">INTERNLM2_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.internlm.INTERNLM2_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.internlm.</span></span><span class="sig-name descname"><span class="pre">INTERNLM2_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/internlm.html#INTERNLM2_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.internlm.INTERNLM2_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../hymba/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.hymba</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../llama/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.llama</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.internlm.INTERNLM2_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">INTERNLM2_TEMPLATE</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/conversation_template/internlm/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/conversation_template/llama/index.html b/autoapi/lmflow/utils/conversation_template/llama/index.html
new file mode 100644
index 000000000..fe3e3c121
--- /dev/null
+++ b/autoapi/lmflow/utils/conversation_template/llama/index.html
@@ -0,0 +1,739 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.conversation_template.llama &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/conversation_template/llama/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.conversation_template.phi" href="../phi/index.html" />
+    <link rel="prev" title="lmflow.utils.conversation_template.internlm" href="../internlm/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.conversation_template</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.llama</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.conversation_template.llama">
+<span id="lmflow-utils-conversation-template-llama"></span><h1>lmflow.utils.conversation_template.llama<a class="headerlink" href="#module-lmflow.utils.conversation_template.llama" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.llama.logger" title="lmflow.utils.conversation_template.llama.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE" title="lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LLAMA3_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE_FOR_TOOL" title="lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE_FOR_TOOL"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LLAMA3_TEMPLATE_FOR_TOOL</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE" title="lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LLAMA2_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE_FOR_TOOL" title="lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE_FOR_TOOL"><code class="xref py py-obj docutils literal notranslate"><span class="pre">LLAMA2_TEMPLATE_FOR_TOOL</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.llama.Llama2ConversationTemplate" title="lmflow.utils.conversation_template.llama.Llama2ConversationTemplate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Llama2ConversationTemplate</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool" title="lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Llama2ConversationTemplateForTool</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.llama.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.llama.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/llama.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.llama.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.llama.Llama2ConversationTemplate">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.llama.</span></span><span class="sig-name descname"><span class="pre">Llama2ConversationTemplate</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/llama.html#Llama2ConversationTemplate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.llama.Llama2ConversationTemplate" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate" title="lmflow.utils.conversation_template.base.ConversationTemplate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.utils.conversation_template.base.ConversationTemplate</span></code></a></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.llama.Llama2ConversationTemplate._encode">
+<span class="sig-name descname"><span class="pre">_encode</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">messages</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">system</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tools</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/llama.html#Llama2ConversationTemplate._encode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.llama.Llama2ConversationTemplate._encode" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.llama.</span></span><span class="sig-name descname"><span class="pre">Llama2ConversationTemplateForTool</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/llama.html#Llama2ConversationTemplateForTool"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="#lmflow.utils.conversation_template.llama.Llama2ConversationTemplate" title="lmflow.utils.conversation_template.llama.Llama2ConversationTemplate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Llama2ConversationTemplate</span></code></a></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool._encode">
+<span class="sig-name descname"><span class="pre">_encode</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">messages</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">system</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tools</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/llama.html#Llama2ConversationTemplateForTool._encode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool._encode" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.llama.</span></span><span class="sig-name descname"><span class="pre">LLAMA3_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/llama.html#LLAMA3_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE_FOR_TOOL">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.llama.</span></span><span class="sig-name descname"><span class="pre">LLAMA3_TEMPLATE_FOR_TOOL</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/llama.html#LLAMA3_TEMPLATE_FOR_TOOL"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE_FOR_TOOL" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.llama.</span></span><span class="sig-name descname"><span class="pre">LLAMA2_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/llama.html#LLAMA2_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE_FOR_TOOL">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.llama.</span></span><span class="sig-name descname"><span class="pre">LLAMA2_TEMPLATE_FOR_TOOL</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/llama.html#LLAMA2_TEMPLATE_FOR_TOOL"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE_FOR_TOOL" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../internlm/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.internlm</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../phi/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.phi</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.llama.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.llama.Llama2ConversationTemplate"><code class="docutils literal notranslate"><span class="pre">Llama2ConversationTemplate</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.llama.Llama2ConversationTemplate._encode"><code class="docutils literal notranslate"><span class="pre">Llama2ConversationTemplate._encode()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool"><code class="docutils literal notranslate"><span class="pre">Llama2ConversationTemplateForTool</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool._encode"><code class="docutils literal notranslate"><span class="pre">Llama2ConversationTemplateForTool._encode()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">LLAMA3_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE_FOR_TOOL"><code class="docutils literal notranslate"><span class="pre">LLAMA3_TEMPLATE_FOR_TOOL</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">LLAMA2_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE_FOR_TOOL"><code class="docutils literal notranslate"><span class="pre">LLAMA2_TEMPLATE_FOR_TOOL</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/conversation_template/llama/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/conversation_template/phi/index.html b/autoapi/lmflow/utils/conversation_template/phi/index.html
new file mode 100644
index 000000000..a682968aa
--- /dev/null
+++ b/autoapi/lmflow/utils/conversation_template/phi/index.html
@@ -0,0 +1,658 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.conversation_template.phi &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/conversation_template/phi/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.conversation_template.qwen" href="../qwen/index.html" />
+    <link rel="prev" title="lmflow.utils.conversation_template.llama" href="../llama/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.conversation_template</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.phi</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.conversation_template.phi">
+<span id="lmflow-utils-conversation-template-phi"></span><h1>lmflow.utils.conversation_template.phi<a class="headerlink" href="#module-lmflow.utils.conversation_template.phi" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.phi.PHI3_TEMPLATE" title="lmflow.utils.conversation_template.phi.PHI3_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">PHI3_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.phi.PHI3_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.phi.</span></span><span class="sig-name descname"><span class="pre">PHI3_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/phi.html#PHI3_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.phi.PHI3_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../llama/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.llama</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../qwen/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.qwen</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.phi.PHI3_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">PHI3_TEMPLATE</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/conversation_template/phi/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/conversation_template/qwen/index.html b/autoapi/lmflow/utils/conversation_template/qwen/index.html
new file mode 100644
index 000000000..b038abf1d
--- /dev/null
+++ b/autoapi/lmflow/utils/conversation_template/qwen/index.html
@@ -0,0 +1,667 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.conversation_template.qwen &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/conversation_template/qwen/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.conversation_template.yi" href="../yi/index.html" />
+    <link rel="prev" title="lmflow.utils.conversation_template.phi" href="../phi/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.conversation_template</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.qwen</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.conversation_template.qwen">
+<span id="lmflow-utils-conversation-template-qwen"></span><h1>lmflow.utils.conversation_template.qwen<a class="headerlink" href="#module-lmflow.utils.conversation_template.qwen" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE" title="lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">QWEN2_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE_FOR_TOOL" title="lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE_FOR_TOOL"><code class="xref py py-obj docutils literal notranslate"><span class="pre">QWEN2_TEMPLATE_FOR_TOOL</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.qwen.</span></span><span class="sig-name descname"><span class="pre">QWEN2_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/qwen.html#QWEN2_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE_FOR_TOOL">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.qwen.</span></span><span class="sig-name descname"><span class="pre">QWEN2_TEMPLATE_FOR_TOOL</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/qwen.html#QWEN2_TEMPLATE_FOR_TOOL"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE_FOR_TOOL" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../phi/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.phi</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../yi/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.yi</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">QWEN2_TEMPLATE</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE_FOR_TOOL"><code class="docutils literal notranslate"><span class="pre">QWEN2_TEMPLATE_FOR_TOOL</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/conversation_template/qwen/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/conversation_template/yi/index.html b/autoapi/lmflow/utils/conversation_template/yi/index.html
new file mode 100644
index 000000000..25254fea0
--- /dev/null
+++ b/autoapi/lmflow/utils/conversation_template/yi/index.html
@@ -0,0 +1,658 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.conversation_template.yi &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/conversation_template/yi/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.conversation_template.zephyr" href="../zephyr/index.html" />
+    <link rel="prev" title="lmflow.utils.conversation_template.qwen" href="../qwen/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.conversation_template</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.yi</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.conversation_template.yi">
+<span id="lmflow-utils-conversation-template-yi"></span><h1>lmflow.utils.conversation_template.yi<a class="headerlink" href="#module-lmflow.utils.conversation_template.yi" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.yi.YI1_5_TEMPLATE" title="lmflow.utils.conversation_template.yi.YI1_5_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">YI1_5_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.yi.YI1_5_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.yi.</span></span><span class="sig-name descname"><span class="pre">YI1_5_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/yi.html#YI1_5_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.yi.YI1_5_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../qwen/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.qwen</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../zephyr/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.zephyr</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.yi.YI1_5_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">YI1_5_TEMPLATE</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/conversation_template/yi/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/conversation_template/zephyr/index.html b/autoapi/lmflow/utils/conversation_template/zephyr/index.html
new file mode 100644
index 000000000..65e710e4a
--- /dev/null
+++ b/autoapi/lmflow/utils/conversation_template/zephyr/index.html
@@ -0,0 +1,694 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.conversation_template.zephyr &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/conversation_template/zephyr/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.data_utils" href="../../data_utils/index.html" />
+    <link rel="prev" title="lmflow.utils.conversation_template.yi" href="../yi/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.conversation_template</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.conversation_template</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.conversation_template.zephyr</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.conversation_template.zephyr">
+<span id="lmflow-utils-conversation-template-zephyr"></span><h1>lmflow.utils.conversation_template.zephyr<a class="headerlink" href="#module-lmflow.utils.conversation_template.zephyr" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.zephyr.logger" title="lmflow.utils.conversation_template.zephyr.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.zephyr.ZEPHYR_TEMPLATE" title="lmflow.utils.conversation_template.zephyr.ZEPHYR_TEMPLATE"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ZEPHYR_TEMPLATE</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate" title="lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">ZephyrConversationTemplate</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.zephyr.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.zephyr.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/zephyr.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.zephyr.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.zephyr.</span></span><span class="sig-name descname"><span class="pre">ZephyrConversationTemplate</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/zephyr.html#ZephyrConversationTemplate"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <a class="reference internal" href="../base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate" title="lmflow.utils.conversation_template.base.ConversationTemplate"><code class="xref py py-obj docutils literal notranslate"><span class="pre">lmflow.utils.conversation_template.base.ConversationTemplate</span></code></a></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate._encode">
+<span class="sig-name descname"><span class="pre">_encode</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">tokenizer</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">transformers.PreTrainedTokenizer</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">messages</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">system</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tools</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwargs</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Sequence</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/zephyr.html#ZephyrConversationTemplate._encode"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate._encode" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.conversation_template.zephyr.ZEPHYR_TEMPLATE">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.conversation_template.zephyr.</span></span><span class="sig-name descname"><span class="pre">ZEPHYR_TEMPLATE</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/conversation_template/zephyr.html#ZEPHYR_TEMPLATE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.conversation_template.zephyr.ZEPHYR_TEMPLATE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../yi/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.yi</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../../data_utils/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.data_utils</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.zephyr.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate"><code class="docutils literal notranslate"><span class="pre">ZephyrConversationTemplate</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate._encode"><code class="docutils literal notranslate"><span class="pre">ZephyrConversationTemplate._encode()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.conversation_template.zephyr.ZEPHYR_TEMPLATE"><code class="docutils literal notranslate"><span class="pre">ZEPHYR_TEMPLATE</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/conversation_template/zephyr/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/data_utils/index.html b/autoapi/lmflow/utils/data_utils/index.html
new file mode 100644
index 000000000..10b876abb
--- /dev/null
+++ b/autoapi/lmflow/utils/data_utils/index.html
@@ -0,0 +1,813 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.data_utils &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/data_utils/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.utils.flash_attention" href="../flash_attention/index.html" />
+    <link rel="prev" title="lmflow.utils.conversation_template.zephyr" href="../conversation_template/zephyr/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.data_utils</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.data_utils">
+<span id="lmflow-utils-data-utils"></span><h1>lmflow.utils.data_utils<a class="headerlink" href="#module-lmflow.utils.data_utils" title="Link to this heading">#</a></h1>
+<p>The program includes several functions: setting a random seed,
+loading data from a JSON file, batching data, and extracting answers from generated text.</p>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.data_utils.VLLMInferenceResultWithInput" title="lmflow.utils.data_utils.VLLMInferenceResultWithInput"><code class="xref py py-obj docutils literal notranslate"><span class="pre">VLLMInferenceResultWithInput</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.data_utils.RewardModelInferenceResultWithInput" title="lmflow.utils.data_utils.RewardModelInferenceResultWithInput"><code class="xref py py-obj docutils literal notranslate"><span class="pre">RewardModelInferenceResultWithInput</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.data_utils.set_random_seed" title="lmflow.utils.data_utils.set_random_seed"><code class="xref py py-obj docutils literal notranslate"><span class="pre">set_random_seed</span></code></a>(seed)</p></td>
+<td><p>Set the random seed for <cite>random</cite>, <cite>numpy</cite>, <cite>torch</cite>, <cite>torch.cuda</cite>.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.data_utils.load_data" title="lmflow.utils.data_utils.load_data"><code class="xref py py-obj docutils literal notranslate"><span class="pre">load_data</span></code></a>(file_name)</p></td>
+<td><p>Load data with file name.</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.data_utils.batchlize" title="lmflow.utils.data_utils.batchlize"><code class="xref py py-obj docutils literal notranslate"><span class="pre">batchlize</span></code></a>(examples, batch_size, random_shuffle)</p></td>
+<td><p>Convert examples to a dataloader.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.data_utils.answer_extraction" title="lmflow.utils.data_utils.answer_extraction"><code class="xref py py-obj docutils literal notranslate"><span class="pre">answer_extraction</span></code></a>(response[, answer_type])</p></td>
+<td><p>Use this funtion to extract answers from generated text</p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.data_utils.process_image_flag" title="lmflow.utils.data_utils.process_image_flag"><code class="xref py py-obj docutils literal notranslate"><span class="pre">process_image_flag</span></code></a>(text[, image_flag])</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.data_utils.set_random_seed">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.data_utils.</span></span><span class="sig-name descname"><span class="pre">set_random_seed</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">seed</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/data_utils.html#set_random_seed"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.data_utils.set_random_seed" title="Link to this definition">#</a></dt>
+<dd><p>Set the random seed for <cite>random</cite>, <cite>numpy</cite>, <cite>torch</cite>, <cite>torch.cuda</cite>.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>seed</strong><span class="classifier">int</span></dt><dd><p>The default seed.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.data_utils.load_data">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.data_utils.</span></span><span class="sig-name descname"><span class="pre">load_data</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">file_name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/data_utils.html#load_data"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.data_utils.load_data" title="Link to this definition">#</a></dt>
+<dd><p>Load data with file name.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>file_name</strong><span class="classifier">str.</span></dt><dd><p>The dataset file name.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt><strong>inputs</strong><span class="classifier">list.</span></dt><dd><p>The input texts of the dataset.</p>
+</dd>
+<dt><strong>outputs</strong><span class="classifier">list.</span></dt><dd><p>The output texts file datasets.</p>
+</dd>
+<dt><strong>len</strong><span class="classifier">int.</span></dt><dd><p>The length of the dataset.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.data_utils.batchlize">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.data_utils.</span></span><span class="sig-name descname"><span class="pre">batchlize</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">examples</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">list</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">batch_size</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">random_shuffle</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/data_utils.html#batchlize"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.data_utils.batchlize" title="Link to this definition">#</a></dt>
+<dd><p>Convert examples to a dataloader.</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>examples</strong><span class="classifier">list.</span></dt><dd><p>Data list.</p>
+</dd>
+<dt><strong>batch_size</strong><span class="classifier">int.</span></dt><dd></dd>
+<dt><strong>random_shuffle</strong><span class="classifier">bool</span></dt><dd><p>If true, the dataloader shuffle the training data.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>dataloader:</dt><dd><p>Dataloader with batch generator.</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.data_utils.answer_extraction">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.data_utils.</span></span><span class="sig-name descname"><span class="pre">answer_extraction</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">response</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">answer_type</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/data_utils.html#answer_extraction"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.data_utils.answer_extraction" title="Link to this definition">#</a></dt>
+<dd><p>Use this funtion to extract answers from generated text</p>
+<dl class="field-list simple">
+<dt class="field-odd">Parameters<span class="colon">:</span></dt>
+<dd class="field-odd"><dl class="simple">
+<dt><strong>args</strong></dt><dd><p>Arguments.</p>
+</dd>
+<dt><strong>response</strong><span class="classifier">str</span></dt><dd><p>plain string response.</p>
+</dd>
+</dl>
+</dd>
+<dt class="field-even">Returns<span class="colon">:</span></dt>
+<dd class="field-even"><dl class="simple">
+<dt>answer:</dt><dd><p>Decoded answer (such as A, B, C, D, E for mutiple-choice QA).</p>
+</dd>
+</dl>
+</dd>
+</dl>
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.data_utils.process_image_flag">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.data_utils.</span></span><span class="sig-name descname"><span class="pre">process_image_flag</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">text</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">image_flag</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'&lt;ImageHere&gt;'</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/data_utils.html#process_image_flag"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.data_utils.process_image_flag" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.data_utils.VLLMInferenceResultWithInput">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.data_utils.</span></span><span class="sig-name descname"><span class="pre">VLLMInferenceResultWithInput</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/data_utils.html#VLLMInferenceResultWithInput"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.data_utils.VLLMInferenceResultWithInput" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">TypedDict</span></code></p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.data_utils.VLLMInferenceResultWithInput.input">
+<span class="sig-name descname"><span class="pre">input</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/data_utils.html#VLLMInferenceResultWithInput.input"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.data_utils.VLLMInferenceResultWithInput.input" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.data_utils.VLLMInferenceResultWithInput.output">
+<span class="sig-name descname"><span class="pre">output</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/data_utils.html#VLLMInferenceResultWithInput.output"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.data_utils.VLLMInferenceResultWithInput.output" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.data_utils.RewardModelInferenceResultWithInput">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.data_utils.</span></span><span class="sig-name descname"><span class="pre">RewardModelInferenceResultWithInput</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/data_utils.html#RewardModelInferenceResultWithInput"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.data_utils.RewardModelInferenceResultWithInput" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">TypedDict</span></code></p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.data_utils.RewardModelInferenceResultWithInput.input">
+<span class="sig-name descname"><span class="pre">input</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/data_utils.html#RewardModelInferenceResultWithInput.input"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.data_utils.RewardModelInferenceResultWithInput.input" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.data_utils.RewardModelInferenceResultWithInput.output">
+<span class="sig-name descname"><span class="pre">output</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">str</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">float</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/data_utils.html#RewardModelInferenceResultWithInput.output"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.data_utils.RewardModelInferenceResultWithInput.output" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../conversation_template/zephyr/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.conversation_template.zephyr</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../flash_attention/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.flash_attention</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.data_utils.set_random_seed"><code class="docutils literal notranslate"><span class="pre">set_random_seed()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.data_utils.load_data"><code class="docutils literal notranslate"><span class="pre">load_data()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.data_utils.batchlize"><code class="docutils literal notranslate"><span class="pre">batchlize()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.data_utils.answer_extraction"><code class="docutils literal notranslate"><span class="pre">answer_extraction()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.data_utils.process_image_flag"><code class="docutils literal notranslate"><span class="pre">process_image_flag()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.data_utils.VLLMInferenceResultWithInput"><code class="docutils literal notranslate"><span class="pre">VLLMInferenceResultWithInput</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.data_utils.VLLMInferenceResultWithInput.input"><code class="docutils literal notranslate"><span class="pre">VLLMInferenceResultWithInput.input</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.data_utils.VLLMInferenceResultWithInput.output"><code class="docutils literal notranslate"><span class="pre">VLLMInferenceResultWithInput.output</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.data_utils.RewardModelInferenceResultWithInput"><code class="docutils literal notranslate"><span class="pre">RewardModelInferenceResultWithInput</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.data_utils.RewardModelInferenceResultWithInput.input"><code class="docutils literal notranslate"><span class="pre">RewardModelInferenceResultWithInput.input</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.data_utils.RewardModelInferenceResultWithInput.output"><code class="docutils literal notranslate"><span class="pre">RewardModelInferenceResultWithInput.output</span></code></a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/utils/data_utils/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.html b/autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.html
new file mode 100644
index 000000000..cfc4cbf2c
--- /dev/null
+++ b/autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.html
@@ -0,0 +1,676 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.flash_attention.bloom_flash_attention &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.flash_attention.gpt2_flash_attention" href="../gpt2_flash_attention/index.html" />
+    <link rel="prev" title="lmflow.utils.flash_attention" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.flash_attention</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.flash_attention</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.flash_attention.bloom_flash_attention</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.flash_attention.bloom_flash_attention">
+<span id="lmflow-utils-flash-attention-bloom-flash-attention"></span><h1>lmflow.utils.flash_attention.bloom_flash_attention<a class="headerlink" href="#module-lmflow.utils.flash_attention.bloom_flash_attention" title="Link to this heading">#</a></h1>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.bloom_flash_attention.forward" title="lmflow.utils.flash_attention.bloom_flash_attention.forward"><code class="xref py py-obj docutils literal notranslate"><span class="pre">forward</span></code></a>(self, hidden_states, residual, alibi, ...[, ...])</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.bloom_flash_attention._prepare_attn_mask" title="lmflow.utils.flash_attention.bloom_flash_attention._prepare_attn_mask"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_prepare_attn_mask</span></code></a>(→ torch.BoolTensor)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.bloom_flash_attention.replace_bloom_attn_with_flash_attn" title="lmflow.utils.flash_attention.bloom_flash_attention.replace_bloom_attn_with_flash_attn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">replace_bloom_attn_with_flash_attn</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.bloom_flash_attention.forward">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.bloom_flash_attention.</span></span><span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">self</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">hidden_states</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">residual</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">alibi</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">layer_past</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">head_mask</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_cache</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/bloom_flash_attention.html#forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.bloom_flash_attention.forward" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.bloom_flash_attention._prepare_attn_mask">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.bloom_flash_attention.</span></span><span class="sig-name descname"><span class="pre">_prepare_attn_mask</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">self</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">input_shape</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past_key_values_length</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">int</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">torch.BoolTensor</span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/bloom_flash_attention.html#_prepare_attn_mask"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.bloom_flash_attention._prepare_attn_mask" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.bloom_flash_attention.replace_bloom_attn_with_flash_attn">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.bloom_flash_attention.</span></span><span class="sig-name descname"><span class="pre">replace_bloom_attn_with_flash_attn</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/bloom_flash_attention.html#replace_bloom_attn_with_flash_attn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.bloom_flash_attention.replace_bloom_attn_with_flash_attn" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.flash_attention</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../gpt2_flash_attention/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.flash_attention.gpt2_flash_attention</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.bloom_flash_attention.forward"><code class="docutils literal notranslate"><span class="pre">forward()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.bloom_flash_attention._prepare_attn_mask"><code class="docutils literal notranslate"><span class="pre">_prepare_attn_mask()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.bloom_flash_attention.replace_bloom_attn_with_flash_attn"><code class="docutils literal notranslate"><span class="pre">replace_bloom_attn_with_flash_attn()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.html b/autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.html
new file mode 100644
index 000000000..05f6cc6cc
--- /dev/null
+++ b/autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.html
@@ -0,0 +1,676 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.flash_attention.gpt2_flash_attention &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.flash_attention.gpt_neo_flash_attention" href="../gpt_neo_flash_attention/index.html" />
+    <link rel="prev" title="lmflow.utils.flash_attention.bloom_flash_attention" href="../bloom_flash_attention/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.flash_attention</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.flash_attention</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.flash_attention.gpt2_flash_attention</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.flash_attention.gpt2_flash_attention">
+<span id="lmflow-utils-flash-attention-gpt2-flash-attention"></span><h1>lmflow.utils.flash_attention.gpt2_flash_attention<a class="headerlink" href="#module-lmflow.utils.flash_attention.gpt2_flash_attention" title="Link to this heading">#</a></h1>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.gpt2_flash_attention.forward" title="lmflow.utils.flash_attention.gpt2_flash_attention.forward"><code class="xref py py-obj docutils literal notranslate"><span class="pre">forward</span></code></a>(→ Tuple[Union[torch.Tensor, ...)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.gpt2_flash_attention._prepare_decoder_attention_mask" title="lmflow.utils.flash_attention.gpt2_flash_attention._prepare_decoder_attention_mask"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_prepare_decoder_attention_mask</span></code></a>(self, attention_mask, ...)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.gpt2_flash_attention.replace_gpt2_attn_with_flash_attn" title="lmflow.utils.flash_attention.gpt2_flash_attention.replace_gpt2_attn_with_flash_attn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">replace_gpt2_attn_with_flash_attn</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.gpt2_flash_attention.forward">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.gpt2_flash_attention.</span></span><span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">self</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">hidden_states</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.FloatTensor</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">layer_past</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.FloatTensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">head_mask</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.FloatTensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_hidden_states</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">encoder_attention_mask</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.FloatTensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_cache</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">Ellipsis</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/gpt2_flash_attention.html#forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.gpt2_flash_attention.forward" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.gpt2_flash_attention._prepare_decoder_attention_mask">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.gpt2_flash_attention.</span></span><span class="sig-name descname"><span class="pre">_prepare_decoder_attention_mask</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">self</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">input_shape</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inputs_embeds</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past_key_values_length</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/gpt2_flash_attention.html#_prepare_decoder_attention_mask"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.gpt2_flash_attention._prepare_decoder_attention_mask" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.gpt2_flash_attention.replace_gpt2_attn_with_flash_attn">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.gpt2_flash_attention.</span></span><span class="sig-name descname"><span class="pre">replace_gpt2_attn_with_flash_attn</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/gpt2_flash_attention.html#replace_gpt2_attn_with_flash_attn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.gpt2_flash_attention.replace_gpt2_attn_with_flash_attn" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../bloom_flash_attention/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.flash_attention.bloom_flash_attention</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../gpt_neo_flash_attention/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.flash_attention.gpt_neo_flash_attention</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.gpt2_flash_attention.forward"><code class="docutils literal notranslate"><span class="pre">forward()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.gpt2_flash_attention._prepare_decoder_attention_mask"><code class="docutils literal notranslate"><span class="pre">_prepare_decoder_attention_mask()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.gpt2_flash_attention.replace_gpt2_attn_with_flash_attn"><code class="docutils literal notranslate"><span class="pre">replace_gpt2_attn_with_flash_attn()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html b/autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html
new file mode 100644
index 000000000..ee39e42eb
--- /dev/null
+++ b/autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html
@@ -0,0 +1,676 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.flash_attention.gpt_neo_flash_attention &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.flash_attention.llama_flash_attention" href="../llama_flash_attention/index.html" />
+    <link rel="prev" title="lmflow.utils.flash_attention.gpt2_flash_attention" href="../gpt2_flash_attention/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.flash_attention</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.flash_attention</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.flash_attention.gpt_neo_flash_attention</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.flash_attention.gpt_neo_flash_attention">
+<span id="lmflow-utils-flash-attention-gpt-neo-flash-attention"></span><h1>lmflow.utils.flash_attention.gpt_neo_flash_attention<a class="headerlink" href="#module-lmflow.utils.flash_attention.gpt_neo_flash_attention" title="Link to this heading">#</a></h1>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.gpt_neo_flash_attention._attn" title="lmflow.utils.flash_attention.gpt_neo_flash_attention._attn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_attn</span></code></a>(self, query, key, value[, attention_mask, head_mask])</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.gpt_neo_flash_attention.forward" title="lmflow.utils.flash_attention.gpt_neo_flash_attention.forward"><code class="xref py py-obj docutils literal notranslate"><span class="pre">forward</span></code></a>(self, hidden_states[, attention_mask, ...])</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.gpt_neo_flash_attention.replace_gpt_neo_attn_with_flash_attn" title="lmflow.utils.flash_attention.gpt_neo_flash_attention.replace_gpt_neo_attn_with_flash_attn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">replace_gpt_neo_attn_with_flash_attn</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.gpt_neo_flash_attention._attn">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.gpt_neo_flash_attention.</span></span><span class="sig-name descname"><span class="pre">_attn</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">self</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">query</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">key</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">value</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">head_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/gpt_neo_flash_attention.html#_attn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.gpt_neo_flash_attention._attn" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.gpt_neo_flash_attention.forward">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.gpt_neo_flash_attention.</span></span><span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">self</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">hidden_states</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">layer_past</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">head_mask</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_cache</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/gpt_neo_flash_attention.html#forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.gpt_neo_flash_attention.forward" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.gpt_neo_flash_attention.replace_gpt_neo_attn_with_flash_attn">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.gpt_neo_flash_attention.</span></span><span class="sig-name descname"><span class="pre">replace_gpt_neo_attn_with_flash_attn</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/gpt_neo_flash_attention.html#replace_gpt_neo_attn_with_flash_attn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.gpt_neo_flash_attention.replace_gpt_neo_attn_with_flash_attn" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../gpt2_flash_attention/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.flash_attention.gpt2_flash_attention</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../llama_flash_attention/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.flash_attention.llama_flash_attention</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.gpt_neo_flash_attention._attn"><code class="docutils literal notranslate"><span class="pre">_attn()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.gpt_neo_flash_attention.forward"><code class="docutils literal notranslate"><span class="pre">forward()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.gpt_neo_flash_attention.replace_gpt_neo_attn_with_flash_attn"><code class="docutils literal notranslate"><span class="pre">replace_gpt_neo_attn_with_flash_attn()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/flash_attention/index.html b/autoapi/lmflow/utils/flash_attention/index.html
new file mode 100644
index 000000000..b3d02be34
--- /dev/null
+++ b/autoapi/lmflow/utils/flash_attention/index.html
@@ -0,0 +1,644 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.flash_attention &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/flash_attention/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.utils.flash_attention.bloom_flash_attention" href="bloom_flash_attention/index.html" />
+    <link rel="prev" title="lmflow.utils.data_utils" href="../data_utils/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 current active has-children"><a class="current reference internal" href="#">lmflow.utils.flash_attention</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.flash_attention</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.flash_attention">
+<span id="lmflow-utils-flash-attention"></span><h1>lmflow.utils.flash_attention<a class="headerlink" href="#module-lmflow.utils.flash_attention" title="Link to this heading">#</a></h1>
+<section id="submodules">
+<h2>Submodules<a class="headerlink" href="#submodules" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l1"><a class="reference internal" href="gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l1"><a class="reference internal" href="gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l1"><a class="reference internal" href="llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l1"><a class="reference internal" href="triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../data_utils/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.data_utils</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="bloom_flash_attention/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.flash_attention.bloom_flash_attention</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#submodules">Submodules</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/utils/flash_attention/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.html b/autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.html
new file mode 100644
index 000000000..0b90e2b83
--- /dev/null
+++ b/autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.html
@@ -0,0 +1,676 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.flash_attention.llama_flash_attention &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/flash_attention/llama_flash_attention/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.flash_attention.triton_flash_attention" href="../triton_flash_attention/index.html" />
+    <link rel="prev" title="lmflow.utils.flash_attention.gpt_neo_flash_attention" href="../gpt_neo_flash_attention/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.flash_attention</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.flash_attention</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.flash_attention.llama_flash_attention</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.flash_attention.llama_flash_attention">
+<span id="lmflow-utils-flash-attention-llama-flash-attention"></span><h1>lmflow.utils.flash_attention.llama_flash_attention<a class="headerlink" href="#module-lmflow.utils.flash_attention.llama_flash_attention" title="Link to this heading">#</a></h1>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.llama_flash_attention.forward" title="lmflow.utils.flash_attention.llama_flash_attention.forward"><code class="xref py py-obj docutils literal notranslate"><span class="pre">forward</span></code></a>(→ Tuple[torch.Tensor, Optional[torch.Tensor], ...)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.llama_flash_attention._prepare_decoder_attention_mask" title="lmflow.utils.flash_attention.llama_flash_attention._prepare_decoder_attention_mask"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_prepare_decoder_attention_mask</span></code></a>(self, attention_mask, ...)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.llama_flash_attention.replace_llama_attn_with_flash_attn" title="lmflow.utils.flash_attention.llama_flash_attention.replace_llama_attn_with_flash_attn"><code class="xref py py-obj docutils literal notranslate"><span class="pre">replace_llama_attn_with_flash_attn</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.llama_flash_attention.forward">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.llama_flash_attention.</span></span><span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">self</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">hidden_states</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">position_ids</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">torch.LongTensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past_key_value</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">output_attentions</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">use_cache</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">torch.Tensor</span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">torch.Tensor</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">None</span><span class="p"><span class="pre">]</span></span></span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/llama_flash_attention.html#forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.llama_flash_attention.forward" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.llama_flash_attention._prepare_decoder_attention_mask">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.llama_flash_attention.</span></span><span class="sig-name descname"><span class="pre">_prepare_decoder_attention_mask</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">self</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">attention_mask</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">input_shape</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">inputs_embeds</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">past_key_values_length</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/llama_flash_attention.html#_prepare_decoder_attention_mask"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.llama_flash_attention._prepare_decoder_attention_mask" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.llama_flash_attention.replace_llama_attn_with_flash_attn">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.llama_flash_attention.</span></span><span class="sig-name descname"><span class="pre">replace_llama_attn_with_flash_attn</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/llama_flash_attention.html#replace_llama_attn_with_flash_attn"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.llama_flash_attention.replace_llama_attn_with_flash_attn" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../gpt_neo_flash_attention/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.flash_attention.gpt_neo_flash_attention</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../triton_flash_attention/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.flash_attention.triton_flash_attention</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.llama_flash_attention.forward"><code class="docutils literal notranslate"><span class="pre">forward()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.llama_flash_attention._prepare_decoder_attention_mask"><code class="docutils literal notranslate"><span class="pre">_prepare_decoder_attention_mask()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.llama_flash_attention.replace_llama_attn_with_flash_attn"><code class="docutils literal notranslate"><span class="pre">replace_llama_attn_with_flash_attn()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html b/autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html
new file mode 100644
index 000000000..cf88c56b6
--- /dev/null
+++ b/autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html
@@ -0,0 +1,892 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.flash_attention.triton_flash_attention &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/flash_attention/triton_flash_attention/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.llava_conversation_lib" href="../../llava_conversation_lib/index.html" />
+    <link rel="prev" title="lmflow.utils.flash_attention.llama_flash_attention" href="../llama_flash_attention/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.flash_attention</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4"><a class="reference internal" href="../bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.flash_attention</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.flash_attention.triton_flash_attention</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.flash_attention.triton_flash_attention">
+<span id="lmflow-utils-flash-attention-triton-flash-attention"></span><h1>lmflow.utils.flash_attention.triton_flash_attention<a class="headerlink" href="#module-lmflow.utils.flash_attention.triton_flash_attention" title="Link to this heading">#</a></h1>
+<p><em>Experimental</em> implementation of FlashAttention in Triton.
+Tested with triton==2.0.0.dev20221202.
+Triton 2.0 has a new backend (MLIR) but seems like it doesn’t yet work for head dimensions
+other than 64:
+<a class="github reference external" href="https://github.com/openai/triton/blob/d376020f90002757eea3ea9475d4f7cfc2ec5ead/python/triton/ops/flash_attention.py#L207">openai/triton</a>
+We’ll update this implementation with the new Triton backend once this is fixed.</p>
+<p>We use the FlashAttention implementation from Phil Tillet a starting point.
+<a class="github reference external" href="https://github.com/openai/triton/blob/master/python/tutorials/06-fused-attention.py">openai/triton</a></p>
+<p>Changes:
+- Implement both causal and non-causal attention.
+- Implement both self-attention and cross-attention.
+- Support arbitrary seqlens (not just multiples of 128), for both forward and backward.
+- Support all head dimensions up to 128 (not just 16, 32, 64, 128), for both forward and backward.
+- Support attention bias.
+- Speed up the forward pass a bit, and only store the LSE instead of m and l.
+- Make the backward for d=128 much faster by reducing register spilling.
+- Optionally parallelize the backward pass across seqlen_k, to deal with the case of
+small batch size * nheads.</p>
+<p>Caution:
+- This is an <em>experimental</em> implementation. The forward pass should be quite robust but
+I’m not 100% sure that the backward pass doesn’t have race conditions (due to the Triton compiler).
+- This implementation has only been tested on A100.
+- If you plan to use headdim other than 64 and 128, you should test for race conditions
+(due to the Triton compiler), as done in tests/test_flash_attn.py
+“test_flash_attn_triton_race_condition”. I’ve tested and fixed many race conditions
+for different head dimensions (40, 48, 64, 128, 80, 88, 96), but I’m still not 100% confident
+that there are none left for other head dimensions.</p>
+<p>Differences between this Triton version and the CUDA version:
+- Triton version doesn’t support dropout.
+- Triton forward is generally faster than CUDA forward, while Triton backward is
+generally slower than CUDA backward. Overall Triton forward + backward is slightly slower
+than CUDA forward + backward.
+- Triton version doesn’t support different sequence lengths in a batch (i.e., RaggedTensor/NestedTensor).
+- Triton version supports attention bias, while CUDA version doesn’t.</p>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_qkvpacked_func" title="lmflow.utils.flash_attention.triton_flash_attention.flash_attn_qkvpacked_func"><code class="xref py py-obj docutils literal notranslate"><span class="pre">flash_attn_qkvpacked_func</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_kvpacked_func" title="lmflow.utils.flash_attention.triton_flash_attention.flash_attn_kvpacked_func"><code class="xref py py-obj docutils literal notranslate"><span class="pre">flash_attn_kvpacked_func</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_func" title="lmflow.utils.flash_attention.triton_flash_attention.flash_attn_func"><code class="xref py py-obj docutils literal notranslate"><span class="pre">flash_attn_func</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc" title="lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FlashAttnQKVPackedFunc</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc" title="lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FlashAttnKVPackedFunc</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc" title="lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc"><code class="xref py py-obj docutils literal notranslate"><span class="pre">FlashAttnFunc</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention._fwd_kernel" title="lmflow.utils.flash_attention.triton_flash_attention._fwd_kernel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_fwd_kernel</span></code></a>(Q, K, V, Bias, Out, Lse, TMP, ...)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention._bwd_preprocess_do_o_dot" title="lmflow.utils.flash_attention.triton_flash_attention._bwd_preprocess_do_o_dot"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_bwd_preprocess_do_o_dot</span></code></a>(Out, DO, Delta, stride_ob, ...)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention._bwd_store_dk_dv" title="lmflow.utils.flash_attention.triton_flash_attention._bwd_store_dk_dv"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_bwd_store_dk_dv</span></code></a>(dk_ptrs, dv_ptrs, dk, dv, offs_n, ...)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel_one_col_block" title="lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel_one_col_block"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_bwd_kernel_one_col_block</span></code></a>(start_n, Q, K, V, Bias, DO, ...)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention.init_to_zero" title="lmflow.utils.flash_attention.triton_flash_attention.init_to_zero"><code class="xref py py-obj docutils literal notranslate"><span class="pre">init_to_zero</span></code></a>(name)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel" title="lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_bwd_kernel</span></code></a>(Q, K, V, Bias, DO, DQ, DK, DV, LSE, D, ...)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention._flash_attn_forward" title="lmflow.utils.flash_attention.triton_flash_attention._flash_attn_forward"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_flash_attn_forward</span></code></a>(q, k, v[, bias, causal, softmax_scale])</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.flash_attention.triton_flash_attention._flash_attn_backward" title="lmflow.utils.flash_attention.triton_flash_attention._flash_attn_backward"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_flash_attn_backward</span></code></a>(do, q, k, v, o, lse, dq, dk, dv)</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention._fwd_kernel">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">_fwd_kernel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">Q</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">K</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">V</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">Bias</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">Out</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">Lse</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">TMP</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">softmax_scale</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_qb</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_qh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_qm</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_kb</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_kh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_kn</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_vb</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_vh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_vn</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_bb</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_bh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_bm</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_ob</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_oh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_om</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">nheads</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seqlen_q</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seqlen_k</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seqlen_q_rounded</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">headdim</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">CACHE_KEY_SEQLEN_Q</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">CACHE_KEY_SEQLEN_K</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BIAS_TYPE</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">IS_CAUSAL</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BLOCK_HEADDIM</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">EVEN_M</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">EVEN_N</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">EVEN_HEADDIM</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BLOCK_M</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BLOCK_N</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#_fwd_kernel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention._fwd_kernel" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention._bwd_preprocess_do_o_dot">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">_bwd_preprocess_do_o_dot</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">Out</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">DO</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">Delta</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_ob</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_oh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_om</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dob</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_doh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dom</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">nheads</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seqlen_q</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seqlen_q_rounded</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">headdim</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BLOCK_M</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BLOCK_HEADDIM</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#_bwd_preprocess_do_o_dot"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention._bwd_preprocess_do_o_dot" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention._bwd_store_dk_dv">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">_bwd_store_dk_dv</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dk_ptrs</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dv_ptrs</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dk</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dv</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">offs_n</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">offs_d</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seqlen_k</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">headdim</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">EVEN_M</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">EVEN_N</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">EVEN_HEADDIM</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#_bwd_store_dk_dv"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention._bwd_store_dk_dv" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel_one_col_block">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">_bwd_kernel_one_col_block</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">start_n</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">Q</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">K</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">V</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">Bias</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">DO</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">DQ</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">DK</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">DV</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">LSE</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">D</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">softmax_scale</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_qm</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_kn</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_vn</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_bm</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dom</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dqm</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dkn</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dvn</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seqlen_q</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seqlen_k</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">headdim</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ATOMIC_ADD</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BIAS_TYPE</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">IS_CAUSAL</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BLOCK_HEADDIM</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">EVEN_M</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">EVEN_N</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">EVEN_HEADDIM</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BLOCK_M</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BLOCK_N</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#_bwd_kernel_one_col_block"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel_one_col_block" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention.init_to_zero">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">init_to_zero</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">name</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#init_to_zero"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention.init_to_zero" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">_bwd_kernel</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">Q</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">K</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">V</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">Bias</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">DO</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">DQ</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">DK</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">DV</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">LSE</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">D</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">softmax_scale</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_qb</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_qh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_qm</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_kb</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_kh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_kn</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_vb</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_vh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_vn</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_bb</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_bh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_bm</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dob</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_doh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dom</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dqb</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dqh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dqm</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dkb</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dkh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dkn</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dvb</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dvh</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">stride_dvn</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">nheads</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seqlen_q</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seqlen_k</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seqlen_q_rounded</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">headdim</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">CACHE_KEY_SEQLEN_Q</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">CACHE_KEY_SEQLEN_K</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BIAS_TYPE</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">IS_CAUSAL</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BLOCK_HEADDIM</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">SEQUENCE_PARALLEL</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">EVEN_M</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">EVEN_N</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">EVEN_HEADDIM</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BLOCK_M</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">BLOCK_N</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">triton.language.constexpr</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#_bwd_kernel"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention._flash_attn_forward">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">_flash_attn_forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">q</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">k</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">v</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">causal</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">softmax_scale</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#_flash_attn_forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention._flash_attn_forward" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention._flash_attn_backward">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">_flash_attn_backward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">do</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">q</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">k</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">v</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">o</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">lse</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dq</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dk</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dv</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">causal</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">softmax_scale</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#_flash_attn_backward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention._flash_attn_backward" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">FlashAttnQKVPackedFunc</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#FlashAttnQKVPackedFunc"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.autograd.Function</span></code></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc.forward">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">ctx</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">qkv</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">causal</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">softmax_scale</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#FlashAttnQKVPackedFunc.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc.forward" title="Link to this definition">#</a></dt>
+<dd><p>qkv: (batch, seqlen, 3, nheads, headdim)
+bias: optional, shape broadcastible to (batch, nheads, seqlen, seqlen).</p>
+<blockquote>
+<div><p>For example, ALiBi mask for causal would have shape (1, nheads, 1, seqlen).
+ALiBi mask for non-causal would have shape (1, nheads, seqlen, seqlen)</p>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc.backward">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">backward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">ctx</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">do</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#FlashAttnQKVPackedFunc.backward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc.backward" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention.flash_attn_qkvpacked_func">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">flash_attn_qkvpacked_func</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#flash_attn_qkvpacked_func"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_qkvpacked_func" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">FlashAttnKVPackedFunc</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#FlashAttnKVPackedFunc"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.autograd.Function</span></code></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc.forward">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">ctx</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">q</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">kv</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">causal</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">softmax_scale</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#FlashAttnKVPackedFunc.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc.forward" title="Link to this definition">#</a></dt>
+<dd><p>q: (batch, seqlen_q, nheads, headdim)
+kv: (batch, seqlen_k, 2, nheads, headdim)
+bias: optional, shape broadcastible to (batch, nheads, seqlen_q, seqlen_k).</p>
+<blockquote>
+<div><p>For example, ALiBi mask for causal would have shape (1, nheads, 1, seqlen_k).
+ALiBi mask for non-causal would have shape (1, nheads, seqlen_q, seqlen_k)</p>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc.backward">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">backward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">ctx</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">do</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#FlashAttnKVPackedFunc.backward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc.backward" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention.flash_attn_kvpacked_func">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">flash_attn_kvpacked_func</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#flash_attn_kvpacked_func"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_kvpacked_func" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">FlashAttnFunc</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#FlashAttnFunc"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.autograd.Function</span></code></p>
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc.forward">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">ctx</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">q</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">k</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">v</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">bias</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">causal</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">softmax_scale</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#FlashAttnFunc.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc.forward" title="Link to this definition">#</a></dt>
+<dd><p>q: (batch_size, seqlen_q, nheads, headdim)
+k, v: (batch_size, seqlen_k, nheads, headdim)
+bias: optional, shape broadcastible to (batch, nheads, seqlen_q, seqlen_k).</p>
+<blockquote>
+<div><p>For example, ALiBi mask for causal would have shape (1, nheads, 1, seqlen_k).
+ALiBi mask for non-causal would have shape (1, nheads, seqlen_q, seqlen_k)</p>
+</div></blockquote>
+</dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc.backward">
+<em class="property"><span class="pre">static</span><span class="w"> </span></em><span class="sig-name descname"><span class="pre">backward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">ctx</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">do</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#FlashAttnFunc.backward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc.backward" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.flash_attention.triton_flash_attention.flash_attn_func">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.flash_attention.triton_flash_attention.</span></span><span class="sig-name descname"><span class="pre">flash_attn_func</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/flash_attention/triton_flash_attention.html#flash_attn_func"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_func" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../llama_flash_attention/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.flash_attention.llama_flash_attention</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../../llava_conversation_lib/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.llava_conversation_lib</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention._fwd_kernel"><code class="docutils literal notranslate"><span class="pre">_fwd_kernel()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention._bwd_preprocess_do_o_dot"><code class="docutils literal notranslate"><span class="pre">_bwd_preprocess_do_o_dot()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention._bwd_store_dk_dv"><code class="docutils literal notranslate"><span class="pre">_bwd_store_dk_dv()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel_one_col_block"><code class="docutils literal notranslate"><span class="pre">_bwd_kernel_one_col_block()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention.init_to_zero"><code class="docutils literal notranslate"><span class="pre">init_to_zero()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel"><code class="docutils literal notranslate"><span class="pre">_bwd_kernel()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention._flash_attn_forward"><code class="docutils literal notranslate"><span class="pre">_flash_attn_forward()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention._flash_attn_backward"><code class="docutils literal notranslate"><span class="pre">_flash_attn_backward()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc"><code class="docutils literal notranslate"><span class="pre">FlashAttnQKVPackedFunc</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc.forward"><code class="docutils literal notranslate"><span class="pre">FlashAttnQKVPackedFunc.forward()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc.backward"><code class="docutils literal notranslate"><span class="pre">FlashAttnQKVPackedFunc.backward()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_qkvpacked_func"><code class="docutils literal notranslate"><span class="pre">flash_attn_qkvpacked_func</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc"><code class="docutils literal notranslate"><span class="pre">FlashAttnKVPackedFunc</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc.forward"><code class="docutils literal notranslate"><span class="pre">FlashAttnKVPackedFunc.forward()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc.backward"><code class="docutils literal notranslate"><span class="pre">FlashAttnKVPackedFunc.backward()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_kvpacked_func"><code class="docutils literal notranslate"><span class="pre">flash_attn_kvpacked_func</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc"><code class="docutils literal notranslate"><span class="pre">FlashAttnFunc</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc.forward"><code class="docutils literal notranslate"><span class="pre">FlashAttnFunc.forward()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc.backward"><code class="docutils literal notranslate"><span class="pre">FlashAttnFunc.backward()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_func"><code class="docutils literal notranslate"><span class="pre">flash_attn_func</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/index.html b/autoapi/lmflow/utils/index.html
new file mode 100644
index 000000000..0a533fadd
--- /dev/null
+++ b/autoapi/lmflow/utils/index.html
@@ -0,0 +1,646 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/index';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+    <link rel="next" title="lmflow.utils.common" href="common/index.html" />
+    <link rel="prev" title="lmflow.tokenization.hf_text_regression_model" href="../tokenization/hf_text_regression_model/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="current reference internal" href="#">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils">
+<span id="lmflow-utils"></span><h1>lmflow.utils<a class="headerlink" href="#module-lmflow.utils" title="Link to this heading">#</a></h1>
+<section id="submodules">
+<h2>Submodules<a class="headerlink" href="#submodules" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l1"><a class="reference internal" href="constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l1"><a class="reference internal" href="conversation_template/index.html">lmflow.utils.conversation_template</a></li>
+<li class="toctree-l1"><a class="reference internal" href="data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l1"><a class="reference internal" href="flash_attention/index.html">lmflow.utils.flash_attention</a></li>
+<li class="toctree-l1"><a class="reference internal" href="llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l1"><a class="reference internal" href="model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l1"><a class="reference internal" href="multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l1"><a class="reference internal" href="position_interpolation/index.html">lmflow.utils.position_interpolation</a></li>
+<li class="toctree-l1"><a class="reference internal" href="versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../tokenization/hf_text_regression_model/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.tokenization.hf_text_regression_model</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="common/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.common</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#submodules">Submodules</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../_sources/autoapi/lmflow/utils/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/llava_conversation_lib/index.html b/autoapi/lmflow/utils/llava_conversation_lib/index.html
new file mode 100644
index 000000000..ed1780731
--- /dev/null
+++ b/autoapi/lmflow/utils/llava_conversation_lib/index.html
@@ -0,0 +1,908 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.llava_conversation_lib &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/llava_conversation_lib/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.utils.model" href="../model/index.html" />
+    <link rel="prev" title="lmflow.utils.flash_attention.triton_flash_attention" href="../flash_attention/triton_flash_attention/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.llava_conversation_lib</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.llava_conversation_lib">
+<span id="lmflow-utils-llava-conversation-lib"></span><h1>lmflow.utils.llava_conversation_lib<a class="headerlink" href="#module-lmflow.utils.llava_conversation_lib" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.conv_vicuna_v0" title="lmflow.utils.llava_conversation_lib.conv_vicuna_v0"><code class="xref py py-obj docutils literal notranslate"><span class="pre">conv_vicuna_v0</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.conv_vicuna_v1" title="lmflow.utils.llava_conversation_lib.conv_vicuna_v1"><code class="xref py py-obj docutils literal notranslate"><span class="pre">conv_vicuna_v1</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.conv_llama_2" title="lmflow.utils.llava_conversation_lib.conv_llama_2"><code class="xref py py-obj docutils literal notranslate"><span class="pre">conv_llama_2</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.conv_llava_llama_2" title="lmflow.utils.llava_conversation_lib.conv_llava_llama_2"><code class="xref py py-obj docutils literal notranslate"><span class="pre">conv_llava_llama_2</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.conv_mpt" title="lmflow.utils.llava_conversation_lib.conv_mpt"><code class="xref py py-obj docutils literal notranslate"><span class="pre">conv_mpt</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.conv_llava_plain" title="lmflow.utils.llava_conversation_lib.conv_llava_plain"><code class="xref py py-obj docutils literal notranslate"><span class="pre">conv_llava_plain</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.conv_llava_v0" title="lmflow.utils.llava_conversation_lib.conv_llava_v0"><code class="xref py py-obj docutils literal notranslate"><span class="pre">conv_llava_v0</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.conv_llava_v0_mmtag" title="lmflow.utils.llava_conversation_lib.conv_llava_v0_mmtag"><code class="xref py py-obj docutils literal notranslate"><span class="pre">conv_llava_v0_mmtag</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.conv_llava_v1" title="lmflow.utils.llava_conversation_lib.conv_llava_v1"><code class="xref py py-obj docutils literal notranslate"><span class="pre">conv_llava_v1</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.conv_llava_v1_mmtag" title="lmflow.utils.llava_conversation_lib.conv_llava_v1_mmtag"><code class="xref py py-obj docutils literal notranslate"><span class="pre">conv_llava_v1_mmtag</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.default_conversation" title="lmflow.utils.llava_conversation_lib.default_conversation"><code class="xref py py-obj docutils literal notranslate"><span class="pre">default_conversation</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.conv_templates" title="lmflow.utils.llava_conversation_lib.conv_templates"><code class="xref py py-obj docutils literal notranslate"><span class="pre">conv_templates</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle" title="lmflow.utils.llava_conversation_lib.SeparatorStyle"><code class="xref py py-obj docutils literal notranslate"><span class="pre">SeparatorStyle</span></code></a></p></td>
+<td><p>Different separator style.</p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.Conversation" title="lmflow.utils.llava_conversation_lib.Conversation"><code class="xref py py-obj docutils literal notranslate"><span class="pre">Conversation</span></code></a></p></td>
+<td><p>A class that keeps all conversation history.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.SeparatorStyle">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">SeparatorStyle</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="o"><span class="pre">*</span></span><span class="n"><span class="pre">args</span></span></em>, <em class="sig-param"><span class="o"><span class="pre">**</span></span><span class="n"><span class="pre">kwds</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#SeparatorStyle"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">enum.Enum</span></code></p>
+<p>Different separator style.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.SeparatorStyle.SINGLE">
+<span class="sig-name descname"><span class="pre">SINGLE</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#SeparatorStyle.SINGLE"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle.SINGLE" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.SeparatorStyle.TWO">
+<span class="sig-name descname"><span class="pre">TWO</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#SeparatorStyle.TWO"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle.TWO" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.SeparatorStyle.MPT">
+<span class="sig-name descname"><span class="pre">MPT</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#SeparatorStyle.MPT"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle.MPT" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.SeparatorStyle.PLAIN">
+<span class="sig-name descname"><span class="pre">PLAIN</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#SeparatorStyle.PLAIN"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle.PLAIN" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.SeparatorStyle.LLAMA_2">
+<span class="sig-name descname"><span class="pre">LLAMA_2</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#SeparatorStyle.LLAMA_2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle.LLAMA_2" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">Conversation</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation" title="Link to this definition">#</a></dt>
+<dd><p>A class that keeps all conversation history.</p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.system">
+<span class="sig-name descname"><span class="pre">system</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.system"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.system" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.roles">
+<span class="sig-name descname"><span class="pre">roles</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.roles"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.roles" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.messages">
+<span class="sig-name descname"><span class="pre">messages</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.messages"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.messages" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.offset">
+<span class="sig-name descname"><span class="pre">offset</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">int</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.offset"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.offset" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.sep_style">
+<span class="sig-name descname"><span class="pre">sep_style</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><a class="reference internal" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle" title="lmflow.utils.llava_conversation_lib.SeparatorStyle"><span class="pre">SeparatorStyle</span></a></em><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.sep_style"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.sep_style" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.sep">
+<span class="sig-name descname"><span class="pre">sep</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'###'</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.sep"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.sep" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.sep2">
+<span class="sig-name descname"><span class="pre">sep2</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">None</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.sep2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.sep2" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.version">
+<span class="sig-name descname"><span class="pre">version</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">str</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'Unknown'</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.version"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.version" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.skip_next">
+<span class="sig-name descname"><span class="pre">skip_next</span></span><em class="property"><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="pre">bool</span></em><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">False</span></em><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.skip_next"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.skip_next" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.get_prompt">
+<span class="sig-name descname"><span class="pre">get_prompt</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.get_prompt"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.get_prompt" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.append_message">
+<span class="sig-name descname"><span class="pre">append_message</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">role</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">message</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.append_message"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.append_message" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.get_images">
+<span class="sig-name descname"><span class="pre">get_images</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">return_pil</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.get_images"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.get_images" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.to_gradio_chatbot">
+<span class="sig-name descname"><span class="pre">to_gradio_chatbot</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.to_gradio_chatbot"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.to_gradio_chatbot" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.copy">
+<span class="sig-name descname"><span class="pre">copy</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.copy"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.copy" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.Conversation.dict">
+<span class="sig-name descname"><span class="pre">dict</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#Conversation.dict"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.Conversation.dict" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.conv_vicuna_v0">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">conv_vicuna_v0</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#conv_vicuna_v0"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.conv_vicuna_v0" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.conv_vicuna_v1">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">conv_vicuna_v1</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#conv_vicuna_v1"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.conv_vicuna_v1" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.conv_llama_2">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">conv_llama_2</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#conv_llama_2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.conv_llama_2" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.conv_llava_llama_2">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">conv_llava_llama_2</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#conv_llava_llama_2"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.conv_llava_llama_2" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.conv_mpt">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">conv_mpt</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#conv_mpt"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.conv_mpt" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.conv_llava_plain">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">conv_llava_plain</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#conv_llava_plain"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.conv_llava_plain" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.conv_llava_v0">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">conv_llava_v0</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#conv_llava_v0"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.conv_llava_v0" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.conv_llava_v0_mmtag">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">conv_llava_v0_mmtag</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#conv_llava_v0_mmtag"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.conv_llava_v0_mmtag" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.conv_llava_v1">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">conv_llava_v1</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#conv_llava_v1"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.conv_llava_v1" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.conv_llava_v1_mmtag">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">conv_llava_v1_mmtag</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#conv_llava_v1_mmtag"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.conv_llava_v1_mmtag" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.default_conversation">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">default_conversation</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#default_conversation"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.default_conversation" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.llava_conversation_lib.conv_templates">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.llava_conversation_lib.</span></span><span class="sig-name descname"><span class="pre">conv_templates</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/llava_conversation_lib.html#conv_templates"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.llava_conversation_lib.conv_templates" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../flash_attention/triton_flash_attention/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.flash_attention.triton_flash_attention</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../model/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.model</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle"><code class="docutils literal notranslate"><span class="pre">SeparatorStyle</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle.SINGLE"><code class="docutils literal notranslate"><span class="pre">SeparatorStyle.SINGLE</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle.TWO"><code class="docutils literal notranslate"><span class="pre">SeparatorStyle.TWO</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle.MPT"><code class="docutils literal notranslate"><span class="pre">SeparatorStyle.MPT</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle.PLAIN"><code class="docutils literal notranslate"><span class="pre">SeparatorStyle.PLAIN</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.SeparatorStyle.LLAMA_2"><code class="docutils literal notranslate"><span class="pre">SeparatorStyle.LLAMA_2</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation"><code class="docutils literal notranslate"><span class="pre">Conversation</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.system"><code class="docutils literal notranslate"><span class="pre">Conversation.system</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.roles"><code class="docutils literal notranslate"><span class="pre">Conversation.roles</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.messages"><code class="docutils literal notranslate"><span class="pre">Conversation.messages</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.offset"><code class="docutils literal notranslate"><span class="pre">Conversation.offset</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.sep_style"><code class="docutils literal notranslate"><span class="pre">Conversation.sep_style</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.sep"><code class="docutils literal notranslate"><span class="pre">Conversation.sep</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.sep2"><code class="docutils literal notranslate"><span class="pre">Conversation.sep2</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.version"><code class="docutils literal notranslate"><span class="pre">Conversation.version</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.skip_next"><code class="docutils literal notranslate"><span class="pre">Conversation.skip_next</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.get_prompt"><code class="docutils literal notranslate"><span class="pre">Conversation.get_prompt()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.append_message"><code class="docutils literal notranslate"><span class="pre">Conversation.append_message()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.get_images"><code class="docutils literal notranslate"><span class="pre">Conversation.get_images()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.to_gradio_chatbot"><code class="docutils literal notranslate"><span class="pre">Conversation.to_gradio_chatbot()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.copy"><code class="docutils literal notranslate"><span class="pre">Conversation.copy()</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.Conversation.dict"><code class="docutils literal notranslate"><span class="pre">Conversation.dict()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.conv_vicuna_v0"><code class="docutils literal notranslate"><span class="pre">conv_vicuna_v0</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.conv_vicuna_v1"><code class="docutils literal notranslate"><span class="pre">conv_vicuna_v1</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.conv_llama_2"><code class="docutils literal notranslate"><span class="pre">conv_llama_2</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.conv_llava_llama_2"><code class="docutils literal notranslate"><span class="pre">conv_llava_llama_2</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.conv_mpt"><code class="docutils literal notranslate"><span class="pre">conv_mpt</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.conv_llava_plain"><code class="docutils literal notranslate"><span class="pre">conv_llava_plain</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.conv_llava_v0"><code class="docutils literal notranslate"><span class="pre">conv_llava_v0</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.conv_llava_v0_mmtag"><code class="docutils literal notranslate"><span class="pre">conv_llava_v0_mmtag</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.conv_llava_v1"><code class="docutils literal notranslate"><span class="pre">conv_llava_v1</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.conv_llava_v1_mmtag"><code class="docutils literal notranslate"><span class="pre">conv_llava_v1_mmtag</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.default_conversation"><code class="docutils literal notranslate"><span class="pre">default_conversation</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.llava_conversation_lib.conv_templates"><code class="docutils literal notranslate"><span class="pre">conv_templates</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/utils/llava_conversation_lib/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/model/index.html b/autoapi/lmflow/utils/model/index.html
new file mode 100644
index 000000000..93b4b8dfe
--- /dev/null
+++ b/autoapi/lmflow/utils/model/index.html
@@ -0,0 +1,673 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.model &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/model/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.utils.multimodal" href="../multimodal/index.html" />
+    <link rel="prev" title="lmflow.utils.llava_conversation_lib" href="../llava_conversation_lib/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.model</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.model">
+<span id="lmflow-utils-model"></span><h1>lmflow.utils.model<a class="headerlink" href="#module-lmflow.utils.model" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.model.logger" title="lmflow.utils.model.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.model.check_homogeneity" title="lmflow.utils.model.check_homogeneity"><code class="xref py py-obj docutils literal notranslate"><span class="pre">check_homogeneity</span></code></a>(→ bool)</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.model.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.model.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/model.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.model.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.model.check_homogeneity">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.model.</span></span><span class="sig-name descname"><span class="pre">check_homogeneity</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model_args_list</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="../../args/index.html#lmflow.args.ModelArguments" title="lmflow.args.ModelArguments"><span class="pre">lmflow.args.ModelArguments</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">bool</span></span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/model.html#check_homogeneity"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.model.check_homogeneity" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../llava_conversation_lib/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.llava_conversation_lib</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../multimodal/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.multimodal</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.model.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.model.check_homogeneity"><code class="docutils literal notranslate"><span class="pre">check_homogeneity()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/utils/model/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/multimodal/index.html b/autoapi/lmflow/utils/multimodal/index.html
new file mode 100644
index 000000000..1d646bd9c
--- /dev/null
+++ b/autoapi/lmflow/utils/multimodal/index.html
@@ -0,0 +1,673 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.multimodal &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/multimodal/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.utils.position_interpolation" href="../position_interpolation/index.html" />
+    <link rel="prev" title="lmflow.utils.model" href="../model/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.multimodal</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.multimodal">
+<span id="lmflow-utils-multimodal"></span><h1>lmflow.utils.multimodal<a class="headerlink" href="#module-lmflow.utils.multimodal" title="Link to this heading">#</a></h1>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.multimodal.update_custom_config" title="lmflow.utils.multimodal.update_custom_config"><code class="xref py py-obj docutils literal notranslate"><span class="pre">update_custom_config</span></code></a>(config, model_args)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.multimodal.load_llava_pretrain_model" title="lmflow.utils.multimodal.load_llava_pretrain_model"><code class="xref py py-obj docutils literal notranslate"><span class="pre">load_llava_pretrain_model</span></code></a>(model, checkpoint_path)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.multimodal.adapt_llava_model_to_lmflow_type" title="lmflow.utils.multimodal.adapt_llava_model_to_lmflow_type"><code class="xref py py-obj docutils literal notranslate"><span class="pre">adapt_llava_model_to_lmflow_type</span></code></a>(state_dict)</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.multimodal.update_custom_config">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.multimodal.</span></span><span class="sig-name descname"><span class="pre">update_custom_config</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">config</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">model_args</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/multimodal.html#update_custom_config"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.multimodal.update_custom_config" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.multimodal.load_llava_pretrain_model">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.multimodal.</span></span><span class="sig-name descname"><span class="pre">load_llava_pretrain_model</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">model</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">checkpoint_path</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/multimodal.html#load_llava_pretrain_model"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.multimodal.load_llava_pretrain_model" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.multimodal.adapt_llava_model_to_lmflow_type">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.multimodal.</span></span><span class="sig-name descname"><span class="pre">adapt_llava_model_to_lmflow_type</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">state_dict</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/multimodal.html#adapt_llava_model_to_lmflow_type"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.multimodal.adapt_llava_model_to_lmflow_type" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../model/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.model</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../position_interpolation/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.position_interpolation</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.multimodal.update_custom_config"><code class="docutils literal notranslate"><span class="pre">update_custom_config()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.multimodal.load_llava_pretrain_model"><code class="docutils literal notranslate"><span class="pre">load_llava_pretrain_model()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.multimodal.adapt_llava_model_to_lmflow_type"><code class="docutils literal notranslate"><span class="pre">adapt_llava_model_to_lmflow_type()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/utils/multimodal/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/position_interpolation/index.html b/autoapi/lmflow/utils/position_interpolation/index.html
new file mode 100644
index 000000000..089351948
--- /dev/null
+++ b/autoapi/lmflow/utils/position_interpolation/index.html
@@ -0,0 +1,640 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.position_interpolation &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/position_interpolation/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch" href="llama_rope_scaled_monkey_patch/index.html" />
+    <link rel="prev" title="lmflow.utils.multimodal" href="../multimodal/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 current active has-children"><a class="current reference internal" href="#">lmflow.utils.position_interpolation</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.position_interpolation</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.position_interpolation">
+<span id="lmflow-utils-position-interpolation"></span><h1>lmflow.utils.position_interpolation<a class="headerlink" href="#module-lmflow.utils.position_interpolation" title="Link to this heading">#</a></h1>
+<section id="submodules">
+<h2>Submodules<a class="headerlink" href="#submodules" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../multimodal/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.multimodal</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="llama_rope_scaled_monkey_patch/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#submodules">Submodules</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/utils/position_interpolation/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html b/autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html
new file mode 100644
index 000000000..7b753e318
--- /dev/null
+++ b/autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html
@@ -0,0 +1,703 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index';</script>
+    <link rel="index" title="Index" href="../../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../../search.html" />
+    <link rel="next" title="lmflow.utils.versioning" href="../../versioning/index.html" />
+    <link rel="prev" title="lmflow.utils.position_interpolation" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils.position_interpolation</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l4 current active"><a class="current reference internal" href="#">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils.position_interpolation</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch">
+<span id="lmflow-utils-position-interpolation-llama-rope-scaled-monkey-patch"></span><h1>lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch<a class="headerlink" href="#module-lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch" title="Link to this heading">#</a></h1>
+<section id="classes">
+<h2>Classes<a class="headerlink" href="#classes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding" title="lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding"><code class="xref py py-obj docutils literal notranslate"><span class="pre">CondenseRotaryEmbedding</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.replace_llama_with_condense" title="lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.replace_llama_with_condense"><code class="xref py py-obj docutils literal notranslate"><span class="pre">replace_llama_with_condense</span></code></a>(pi_ratio, ntk_ratio)</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py class">
+<dt class="sig sig-object py" id="lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding">
+<em class="property"><span class="pre">class</span><span class="w"> </span></em><span class="sig-prename descclassname"><span class="pre">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.</span></span><span class="sig-name descname"><span class="pre">CondenseRotaryEmbedding</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">dim</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">pi_ratio</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ntk_ratio</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_position_embeddings</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">2048</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">base</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">10000</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">device</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch.html#CondenseRotaryEmbedding"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding" title="Link to this definition">#</a></dt>
+<dd><p>Bases: <code class="xref py py-obj docutils literal notranslate"><span class="pre">torch.nn.Module</span></code></p>
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.ntk_ratio">
+<span class="sig-name descname"><span class="pre">ntk_ratio</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch.html#CondenseRotaryEmbedding.ntk_ratio"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.ntk_ratio" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.pi_ratio">
+<span class="sig-name descname"><span class="pre">pi_ratio</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch.html#CondenseRotaryEmbedding.pi_ratio"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.pi_ratio" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py attribute">
+<dt class="sig sig-object py" id="lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.max_seq_len_cached">
+<span class="sig-name descname"><span class="pre">max_seq_len_cached</span></span><a class="reference internal" href="../../../../../_modules/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch.html#CondenseRotaryEmbedding.max_seq_len_cached"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.max_seq_len_cached" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py method">
+<dt class="sig sig-object py" id="lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.forward">
+<span class="sig-name descname"><span class="pre">forward</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">x</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seq_len</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch.html#CondenseRotaryEmbedding.forward"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.forward" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.replace_llama_with_condense">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.</span></span><span class="sig-name descname"><span class="pre">replace_llama_with_condense</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pi_ratio</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ntk_ratio</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../../_modules/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch.html#replace_llama_with_condense"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.replace_llama_with_condense" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.position_interpolation</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../../versioning/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.utils.versioning</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#classes">Classes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding"><code class="docutils literal notranslate"><span class="pre">CondenseRotaryEmbedding</span></code></a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.ntk_ratio"><code class="docutils literal notranslate"><span class="pre">CondenseRotaryEmbedding.ntk_ratio</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.pi_ratio"><code class="docutils literal notranslate"><span class="pre">CondenseRotaryEmbedding.pi_ratio</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.max_seq_len_cached"><code class="docutils literal notranslate"><span class="pre">CondenseRotaryEmbedding.max_seq_len_cached</span></code></a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.forward"><code class="docutils literal notranslate"><span class="pre">CondenseRotaryEmbedding.forward()</span></code></a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.replace_llama_with_condense"><code class="docutils literal notranslate"><span class="pre">replace_llama_with_condense()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../../_sources/autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/utils/versioning/index.html b/autoapi/lmflow/utils/versioning/index.html
new file mode 100644
index 000000000..087b5debf
--- /dev/null
+++ b/autoapi/lmflow/utils/versioning/index.html
@@ -0,0 +1,763 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.utils.versioning &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/utils/versioning/index';</script>
+    <link rel="index" title="Index" href="../../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../../search.html" />
+    <link rel="next" title="lmflow.version" href="../../version/index.html" />
+    <link rel="prev" title="lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch" href="../position_interpolation/llama_rope_scaled_monkey_patch/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active has-children"><a class="reference internal" href="../index.html">lmflow.utils</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l3"><a class="reference internal" href="../common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3 current active"><a class="current reference internal" href="#">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2"><a class="reference internal" href="../../version/index.html">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">lmflow</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow.utils</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.utils.versioning</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.utils.versioning">
+<span id="lmflow-utils-versioning"></span><h1>lmflow.utils.versioning<a class="headerlink" href="#module-lmflow.utils.versioning" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.versioning.logger" title="lmflow.utils.versioning.logger"><code class="xref py py-obj docutils literal notranslate"><span class="pre">logger</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="functions">
+<h2>Functions<a class="headerlink" href="#functions" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.versioning.get_python_version" title="lmflow.utils.versioning.get_python_version"><code class="xref py py-obj docutils literal notranslate"><span class="pre">get_python_version</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.versioning._is_package_available" title="lmflow.utils.versioning._is_package_available"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_is_package_available</span></code></a>(package_name[, skippable])</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.versioning._is_packages_available" title="lmflow.utils.versioning._is_packages_available"><code class="xref py py-obj docutils literal notranslate"><span class="pre">_is_packages_available</span></code></a>(packages)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.versioning.is_package_version_at_least" title="lmflow.utils.versioning.is_package_version_at_least"><code class="xref py py-obj docutils literal notranslate"><span class="pre">is_package_version_at_least</span></code></a>(package_name, min_version)</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.versioning.is_gradio_available" title="lmflow.utils.versioning.is_gradio_available"><code class="xref py py-obj docutils literal notranslate"><span class="pre">is_gradio_available</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.versioning.is_ray_available" title="lmflow.utils.versioning.is_ray_available"><code class="xref py py-obj docutils literal notranslate"><span class="pre">is_ray_available</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.versioning.is_vllm_available" title="lmflow.utils.versioning.is_vllm_available"><code class="xref py py-obj docutils literal notranslate"><span class="pre">is_vllm_available</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.versioning.is_flash_attn_available" title="lmflow.utils.versioning.is_flash_attn_available"><code class="xref py py-obj docutils literal notranslate"><span class="pre">is_flash_attn_available</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.versioning.is_flask_available" title="lmflow.utils.versioning.is_flask_available"><code class="xref py py-obj docutils literal notranslate"><span class="pre">is_flask_available</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-even"><td><p><a class="reference internal" href="#lmflow.utils.versioning.is_trl_available" title="lmflow.utils.versioning.is_trl_available"><code class="xref py py-obj docutils literal notranslate"><span class="pre">is_trl_available</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.utils.versioning.is_multimodal_available" title="lmflow.utils.versioning.is_multimodal_available"><code class="xref py py-obj docutils literal notranslate"><span class="pre">is_multimodal_available</span></code></a>()</p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.utils.versioning.logger">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.versioning.</span></span><span class="sig-name descname"><span class="pre">logger</span></span><a class="reference internal" href="../../../../_modules/lmflow/utils/versioning.html#logger"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.versioning.logger" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.versioning.get_python_version">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.versioning.</span></span><span class="sig-name descname"><span class="pre">get_python_version</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/versioning.html#get_python_version"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.versioning.get_python_version" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.versioning._is_package_available">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.versioning.</span></span><span class="sig-name descname"><span class="pre">_is_package_available</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">package_name</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">skippable</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/versioning.html#_is_package_available"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.versioning._is_package_available" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.versioning._is_packages_available">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.versioning.</span></span><span class="sig-name descname"><span class="pre">_is_packages_available</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">packages</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="w"> </span><span class="p"><span class="pre">|</span></span><span class="w"> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">Tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="pre">bool</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/versioning.html#_is_packages_available"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.versioning._is_packages_available" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.versioning.is_package_version_at_least">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.versioning.</span></span><span class="sig-name descname"><span class="pre">is_package_version_at_least</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">package_name</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">min_version</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/versioning.html#is_package_version_at_least"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.versioning.is_package_version_at_least" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.versioning.is_gradio_available">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.versioning.</span></span><span class="sig-name descname"><span class="pre">is_gradio_available</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/versioning.html#is_gradio_available"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.versioning.is_gradio_available" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.versioning.is_ray_available">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.versioning.</span></span><span class="sig-name descname"><span class="pre">is_ray_available</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/versioning.html#is_ray_available"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.versioning.is_ray_available" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.versioning.is_vllm_available">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.versioning.</span></span><span class="sig-name descname"><span class="pre">is_vllm_available</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/versioning.html#is_vllm_available"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.versioning.is_vllm_available" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.versioning.is_flash_attn_available">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.versioning.</span></span><span class="sig-name descname"><span class="pre">is_flash_attn_available</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/versioning.html#is_flash_attn_available"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.versioning.is_flash_attn_available" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.versioning.is_flask_available">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.versioning.</span></span><span class="sig-name descname"><span class="pre">is_flask_available</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/versioning.html#is_flask_available"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.versioning.is_flask_available" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.versioning.is_trl_available">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.versioning.</span></span><span class="sig-name descname"><span class="pre">is_trl_available</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/versioning.html#is_trl_available"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.versioning.is_trl_available" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+<dl class="py function">
+<dt class="sig sig-object py" id="lmflow.utils.versioning.is_multimodal_available">
+<span class="sig-prename descclassname"><span class="pre">lmflow.utils.versioning.</span></span><span class="sig-name descname"><span class="pre">is_multimodal_available</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="reference internal" href="../../../../_modules/lmflow/utils/versioning.html#is_multimodal_available"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.utils.versioning.is_multimodal_available" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../position_interpolation/llama_rope_scaled_monkey_patch/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../../version/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">lmflow.version</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#functions">Functions</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.versioning.logger"><code class="docutils literal notranslate"><span class="pre">logger</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.versioning.get_python_version"><code class="docutils literal notranslate"><span class="pre">get_python_version()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.versioning._is_package_available"><code class="docutils literal notranslate"><span class="pre">_is_package_available()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.versioning._is_packages_available"><code class="docutils literal notranslate"><span class="pre">_is_packages_available()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.versioning.is_package_version_at_least"><code class="docutils literal notranslate"><span class="pre">is_package_version_at_least()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.versioning.is_gradio_available"><code class="docutils literal notranslate"><span class="pre">is_gradio_available()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.versioning.is_ray_available"><code class="docutils literal notranslate"><span class="pre">is_ray_available()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.versioning.is_vllm_available"><code class="docutils literal notranslate"><span class="pre">is_vllm_available()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.versioning.is_flash_attn_available"><code class="docutils literal notranslate"><span class="pre">is_flash_attn_available()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.versioning.is_flask_available"><code class="docutils literal notranslate"><span class="pre">is_flask_available()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.versioning.is_trl_available"><code class="docutils literal notranslate"><span class="pre">is_trl_available()</span></code></a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.utils.versioning.is_multimodal_available"><code class="docutils literal notranslate"><span class="pre">is_multimodal_available()</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../../_sources/autoapi/lmflow/utils/versioning/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/autoapi/lmflow/version/index.html b/autoapi/lmflow/version/index.html
new file mode 100644
index 000000000..3fe38f526
--- /dev/null
+++ b/autoapi/lmflow/version/index.html
@@ -0,0 +1,652 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../../../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>lmflow.version &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../../../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../../../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../../../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../../../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'autoapi/lmflow/version/index';</script>
+    <link rel="index" title="Index" href="../../../genindex.html" />
+    <link rel="search" title="Search" href="../../../search.html" />
+    <link rel="next" title="About" href="../../../about/index.html" />
+    <link rel="prev" title="lmflow.utils.versioning" href="../utils/versioning/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../../../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../../../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="../../index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../../../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../../../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active has-children"><a class="reference internal" href="../index.html">lmflow</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../args/index.html">lmflow.args</a></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../datasets/index.html">lmflow.datasets</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/dataset/index.html">lmflow.datasets.dataset</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../datasets/multi_modal_dataset/index.html">lmflow.datasets.multi_modal_dataset</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../models/index.html">lmflow.models</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../models/auto_model/index.html">lmflow.models.auto_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/base_model/index.html">lmflow.models.base_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/decoder_model/index.html">lmflow.models.decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/encoder_decoder_model/index.html">lmflow.models.encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_decoder_model/index.html">lmflow.models.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_encoder_decoder_model/index.html">lmflow.models.hf_encoder_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_model_mixin/index.html">lmflow.models.hf_model_mixin</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/hf_text_regression_model/index.html">lmflow.models.hf_text_regression_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/interfaces/index.html">lmflow.models.interfaces</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/interfaces/tunable/index.html">lmflow.models.interfaces.tunable</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/regression_model/index.html">lmflow.models.regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/text_regression_model/index.html">lmflow.models.text_regression_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../models/vision2seq_model/index.html">lmflow.models.vision2seq_model</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../models/vision_encoder/index.html">lmflow.models.vision_encoder</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../models/vision_encoder/clip_encoder/index.html">lmflow.models.vision_encoder.clip_encoder</a></li>
+</ul>
+</details></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../optim/index.html">lmflow.optim</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabelief/index.html">lmflow.optim.adabelief</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adabound/index.html">lmflow.optim.adabound</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adadelta/index.html">lmflow.optim.adadelta</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adagrad/index.html">lmflow.optim.adagrad</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adam/index.html">lmflow.optim.adam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamax/index.html">lmflow.optim.adamax</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamp/index.html">lmflow.optim.adamp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adamw_schedule_free/index.html">lmflow.optim.adamw_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/adan/index.html">lmflow.optim.adan</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/dummy/index.html">lmflow.optim.dummy</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lamb/index.html">lmflow.optim.lamb</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/lars/index.html">lmflow.optim.lars</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/nadam/index.html">lmflow.optim.nadam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/novograd/index.html">lmflow.optim.novograd</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/optimizers/index.html">lmflow.optim.optimizers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/radam/index.html">lmflow.optim.radam</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgd_schedule_free/index.html">lmflow.optim.sgd_schedule_free</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sgdp/index.html">lmflow.optim.sgdp</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/sophia/index.html">lmflow.optim.sophia</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../optim/yogi/index.html">lmflow.optim.yogi</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../pipeline/index.html">lmflow.pipeline</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/auto_pipeline/index.html">lmflow.pipeline.auto_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_aligner/index.html">lmflow.pipeline.base_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_pipeline/index.html">lmflow.pipeline.base_pipeline</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/base_tuner/index.html">lmflow.pipeline.base_tuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpo_aligner/index.html">lmflow.pipeline.dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/dpov2_aligner/index.html">lmflow.pipeline.dpov2_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/evaluator/index.html">lmflow.pipeline.evaluator</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/finetuner/index.html">lmflow.pipeline.finetuner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/inferencer/index.html">lmflow.pipeline.inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/iterative_dpo_aligner/index.html">lmflow.pipeline.iterative_dpo_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/raft_aligner/index.html">lmflow.pipeline.raft_aligner</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_inferencer/index.html">lmflow.pipeline.rm_inferencer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/rm_tuner/index.html">lmflow.pipeline.rm_tuner</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../pipeline/utils/index.html">lmflow.pipeline.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_dataprocessor/index.html">lmflow.pipeline.utils.dpov2_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/dpov2_trainer/index.html">lmflow.pipeline.utils.dpov2_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_dpov2_align/index.html">lmflow.pipeline.utils.memory_safe_dpov2_align</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/memory_safe_vllm_inference/index.html">lmflow.pipeline.utils.memory_safe_vllm_inference</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/peft_trainer/index.html">lmflow.pipeline.utils.peft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/raft_trainer/index.html">lmflow.pipeline.utils.raft_trainer</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_dataprocessor/index.html">lmflow.pipeline.utils.rm_dataprocessor</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../pipeline/utils/rm_trainer/index.html">lmflow.pipeline.utils.rm_trainer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../pipeline/vllm_inferencer/index.html">lmflow.pipeline.vllm_inferencer</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../tokenization/index.html">lmflow.tokenization</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_decoder_model/index.html">lmflow.tokenization.hf_decoder_model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../tokenization/hf_text_regression_model/index.html">lmflow.tokenization.hf_text_regression_model</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 has-children"><a class="reference internal" href="../utils/index.html">lmflow.utils</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l3"><a class="reference internal" href="../utils/common/index.html">lmflow.utils.common</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/constants/index.html">lmflow.utils.constants</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/conversation_template/index.html">lmflow.utils.conversation_template</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/base/index.html">lmflow.utils.conversation_template.base</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatglm/index.html">lmflow.utils.conversation_template.chatglm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/chatml/index.html">lmflow.utils.conversation_template.chatml</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/deepseek/index.html">lmflow.utils.conversation_template.deepseek</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/gemma/index.html">lmflow.utils.conversation_template.gemma</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/hymba/index.html">lmflow.utils.conversation_template.hymba</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/internlm/index.html">lmflow.utils.conversation_template.internlm</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/llama/index.html">lmflow.utils.conversation_template.llama</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/phi/index.html">lmflow.utils.conversation_template.phi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/qwen/index.html">lmflow.utils.conversation_template.qwen</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/yi/index.html">lmflow.utils.conversation_template.yi</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/conversation_template/zephyr/index.html">lmflow.utils.conversation_template.zephyr</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/data_utils/index.html">lmflow.utils.data_utils</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/flash_attention/index.html">lmflow.utils.flash_attention</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/bloom_flash_attention/index.html">lmflow.utils.flash_attention.bloom_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt2_flash_attention/index.html">lmflow.utils.flash_attention.gpt2_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/gpt_neo_flash_attention/index.html">lmflow.utils.flash_attention.gpt_neo_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/llama_flash_attention/index.html">lmflow.utils.flash_attention.llama_flash_attention</a></li>
+<li class="toctree-l4"><a class="reference internal" href="../utils/flash_attention/triton_flash_attention/index.html">lmflow.utils.flash_attention.triton_flash_attention</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/llava_conversation_lib/index.html">lmflow.utils.llava_conversation_lib</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/model/index.html">lmflow.utils.model</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/multimodal/index.html">lmflow.utils.multimodal</a></li>
+<li class="toctree-l3 has-children"><a class="reference internal" href="../utils/position_interpolation/index.html">lmflow.utils.position_interpolation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
+<li class="toctree-l4"><a class="reference internal" href="../utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a></li>
+</ul>
+</details></li>
+<li class="toctree-l3"><a class="reference internal" href="../utils/versioning/index.html">lmflow.utils.versioning</a></li>
+</ul>
+</details></li>
+<li class="toctree-l2 current active"><a class="current reference internal" href="#">lmflow.version</a></li>
+</ul>
+</details></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../../../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="../../index.html" class="nav-link">API Reference</a></li>
+    
+    
+    <li class="breadcrumb-item"><a href="../index.html" class="nav-link">lmflow</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">lmflow.version</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="module-lmflow.version">
+<span id="lmflow-version"></span><h1>lmflow.version<a class="headerlink" href="#module-lmflow.version" title="Link to this heading">#</a></h1>
+<section id="attributes">
+<h2>Attributes<a class="headerlink" href="#attributes" title="Link to this heading">#</a></h2>
+<div class="pst-scrollable-table-container"><table class="autosummary longtable table autosummary">
+<tbody>
+<tr class="row-odd"><td><p><a class="reference internal" href="#lmflow.version.__version__" title="lmflow.version.__version__"><code class="xref py py-obj docutils literal notranslate"><span class="pre">__version__</span></code></a></p></td>
+<td><p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="module-contents">
+<h2>Module Contents<a class="headerlink" href="#module-contents" title="Link to this heading">#</a></h2>
+<dl class="py data">
+<dt class="sig sig-object py" id="lmflow.version.__version__">
+<span class="sig-prename descclassname"><span class="pre">lmflow.version.</span></span><span class="sig-name descname"><span class="pre">__version__</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'0.0.9'</span></em><a class="reference internal" href="../../../_modules/lmflow/version.html#__version__"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#lmflow.version.__version__" title="Link to this definition">#</a></dt>
+<dd></dd></dl>
+
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../utils/versioning/index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">lmflow.utils.versioning</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../../../about/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">About</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#attributes">Attributes</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#module-contents">Module Contents</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lmflow.version.__version__"><code class="docutils literal notranslate"><span class="pre">__version__</span></code></a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../../../_sources/autoapi/lmflow/version/index.rst.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/blogs/benchmark.html b/blogs/benchmark.html
new file mode 100644
index 000000000..70da69cd4
--- /dev/null
+++ b/blogs/benchmark.html
@@ -0,0 +1,1061 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>LMFlow Benchmark: An Automatic Evaluation Framework for Open-Source LLMs &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'blogs/benchmark';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Examples" href="../examples/index.html" />
+    <link rel="prev" title="Blogs" href="index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active"><a class="current reference internal" href="#">LMFlow Benchmark: An Automatic Evaluation Framework for Open-Source LLMs</a></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="index.html" class="nav-link">Blogs</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">LMFlow Benchmark: An Automatic Evaluation Framework for Open-Source LLMs</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="lmflow-benchmark-an-automatic-evaluation-framework-for-open-source-llms">
+<h1>LMFlow Benchmark: An Automatic Evaluation Framework for Open-Source LLMs<a class="headerlink" href="#lmflow-benchmark-an-automatic-evaluation-framework-for-open-source-llms" title="Link to this heading">#</a></h1>
+<p>May 9, 2023</p>
+<section id="introduction">
+<h2>Introduction<a class="headerlink" href="#introduction" title="Link to this heading">#</a></h2>
+<p>Evaluation of a chat-style Large Language Model (LLM) has been a huge challenge since the breakthrough of ChatGPT. On the one hand, researchers and engineers need a reliable way to compare two models and decide which model to choose under a certain application scenario. On the other hand, they have to monitor the model performance during the training of an LLM to avoid performance issues such as forgetting.</p>
+<p>Recent work of Vicuna introduces comparison methods of human evaluation, a.k.a. Chatbot Arena. They also pioneered the evaluation method by invoking GPT-4 to compare the outputs of two models. However, those methods require expensive human labeling or GPT-4 API calls, which are neither scalable nor convenient for LLM development.</p>
+<p>In this article, we introduce LMFlow benchmark, a new benchmark which provides a cheap and easy-to-use evaluation framework that can help reflect different aspects of LLMs. We have open-sourced the dataset and the code as well, so that everyone in the LLM community can use those toolkits to evaluate, monitor or compare different LLMs.</p>
+</section>
+<section id="metric">
+<h2>Metric<a class="headerlink" href="#metric" title="Link to this heading">#</a></h2>
+<p>In our evaluation framework, Negative Log Likelihood (NLL) is used for evaluating LLM
+<img alt="" src="../_images/nll.png" /></p>
+<p>which corresponds to the LLM model’s prediction probability over a corpus set given their contexts. If the corpus set itself indicates a certain type of LLM ability, such as multi-round conversation, instruction following, math problem solving, role-playing, then NLL on those corpora can provide quantitative metrics to reflect those abilities.</p>
+<p><img alt="" src="../_images/benchmark-1.png" /></p>
+<p>The key idea behind NLL, is that</p>
+<p><em>Generation ability is positively correlated with prediction ability.</em></p>
+<p>For instance, an LLM which performs well in essay writing should have no problem understanding and predicting a reference human essay, just like human chess masters performing well at memorizing an endgame on a chessboard.</p>
+<p>Besides NLL, another similar and commonly used metric in NLP is Perplexity (PPL):</p>
+<p><img alt="" src="../_images/ppl.png" /></p>
+<p>Nevertheless, perplexity intrinsically depends on the lengths of the tokenized sequences, which induces unfair comparison between models with different tokenizers. For example, if a model has a smaller vocabulary size, it inherently results in a longer tokenized sequence and a lower token-level perplexity. Thus in all our experiments, we use NLL instead of PPL.</p>
+<p>One huge advantage of NLL evaluation is that it does not require human involvement during the evaluation process. As long as the test reference corpus is given, one can evaluate different aspects of an LLM’s ability automatically. This makes the evaluation of LLM more accessible to researchers.</p>
+<p>Besides its convenience, NLL itself is also a good metric. In our experimental results in commonsense QA, we find that NLL is correlated with QA accuracy when comparing the different finetuned versions of a single model.</p>
+<p><strong>Table 1: Accuracy results in traditional commonsense QA benchmarks</strong></p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head text-left"><p></p></th>
+<th class="head text-center"><p>winogrande</p></th>
+<th class="head text-center"><p>boolq</p></th>
+<th class="head text-center"><p>arc_e</p></th>
+<th class="head text-center"><p>hellaswag</p></th>
+<th class="head text-center"><p>piqa</p></th>
+<th class="head text-center"><p>obqa</p></th>
+<th class="head text-center"><p>arc_c</p></th>
+<th class="head text-center"><p>Average</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td class="text-left"><p>bloom-3b</p></td>
+<td class="text-center"><p>58.7</p></td>
+<td class="text-center"><p>61.6</p></td>
+<td class="text-center"><p>59.5</p></td>
+<td class="text-center"><p>52.7</p></td>
+<td class="text-center"><p>70.8</p></td>
+<td class="text-center"><p>42.2</p></td>
+<td class="text-center"><p>30.6</p></td>
+<td class="text-center"><p>53.7</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>bloom-7.1b</p></td>
+<td class="text-center"><p>64.4</p></td>
+<td class="text-center"><p>62.9</p></td>
+<td class="text-center"><p>65.0</p></td>
+<td class="text-center"><p>59.6</p></td>
+<td class="text-center"><p>73.6</p></td>
+<td class="text-center"><p>35.8</p></td>
+<td class="text-center"><p>33.4</p></td>
+<td class="text-center"><p>56.3</p></td>
+</tr>
+<tr class="row-even"><td class="text-left"><p>opt-6.9b</p></td>
+<td class="text-center"><p>65.2</p></td>
+<td class="text-center"><p>66.1</p></td>
+<td class="text-center"><p>65.6</p></td>
+<td class="text-center"><p>67.2</p></td>
+<td class="text-center"><p>76.5</p></td>
+<td class="text-center"><p>37.4</p></td>
+<td class="text-center"><p>34.6</p></td>
+<td class="text-center"><p>58.9</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>opt-13b</p></td>
+<td class="text-center"><p>65.0</p></td>
+<td class="text-center"><p>65.9</p></td>
+<td class="text-center"><p>67.1</p></td>
+<td class="text-center"><p>69.8</p></td>
+<td class="text-center"><p>76.9</p></td>
+<td class="text-center"><p>39.0</p></td>
+<td class="text-center"><p>35.7</p></td>
+<td class="text-center"><p>59.9</p></td>
+</tr>
+<tr class="row-even"><td class="text-left"><p>llama-7b</p></td>
+<td class="text-center"><p>67.9</p></td>
+<td class="text-center"><p>73.2</p></td>
+<td class="text-center"><p>67.3</p></td>
+<td class="text-center"><p>73.0</p></td>
+<td class="text-center"><p>78.3</p></td>
+<td class="text-center"><p>42.4</p></td>
+<td class="text-center"><p>41.4</p></td>
+<td class="text-center"><p>62.7</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>llama-13b</p></td>
+<td class="text-center"><p><strong>70.0</strong></p></td>
+<td class="text-center"><p><strong>68.5</strong></p></td>
+<td class="text-center"><p><strong>74.5</strong></p></td>
+<td class="text-center"><p><strong>76.2</strong></p></td>
+<td class="text-center"><p><strong>79.1</strong></p></td>
+<td class="text-center"><p><strong>42.2</strong></p></td>
+<td class="text-center"><p><strong>44.5</strong></p></td>
+<td class="text-center"><p><strong>65.0</strong></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+<p><strong>Table 2: NLL results in corpus of commonsense QA benchmarks</strong></p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head text-left"><p></p></th>
+<th class="head text-center"><p>winogrande</p></th>
+<th class="head text-center"><p>boolq</p></th>
+<th class="head text-center"><p>arc_e</p></th>
+<th class="head text-center"><p>hellaswag</p></th>
+<th class="head text-center"><p>piqa</p></th>
+<th class="head text-center"><p>obqa</p></th>
+<th class="head text-center"><p>arc_c</p></th>
+<th class="head text-center"><p>Average</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td class="text-left"><p>bloom-3b</p></td>
+<td class="text-center"><p>86.5</p></td>
+<td class="text-center"><p>228</p></td>
+<td class="text-center"><p>86</p></td>
+<td class="text-center"><p>245</p></td>
+<td class="text-center"><p>134</p></td>
+<td class="text-center"><p>64.5</p></td>
+<td class="text-center"><p>101.5</p></td>
+<td class="text-center"><p>135.1</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>bloom-7.1b</p></td>
+<td class="text-center"><p>85</p></td>
+<td class="text-center"><p>215</p></td>
+<td class="text-center"><p>81.5</p></td>
+<td class="text-center"><p>237</p></td>
+<td class="text-center"><p>130</p></td>
+<td class="text-center"><p>62.5</p></td>
+<td class="text-center"><p>96</p></td>
+<td class="text-center"><p>129.5</p></td>
+</tr>
+<tr class="row-even"><td class="text-left"><p>opt-6.9b</p></td>
+<td class="text-center"><p>81.5</p></td>
+<td class="text-center"><p>200</p></td>
+<td class="text-center"><p>81.5</p></td>
+<td class="text-center"><p>224</p></td>
+<td class="text-center"><p>125</p></td>
+<td class="text-center"><p>61</p></td>
+<td class="text-center"><p>96</p></td>
+<td class="text-center"><p>124.1</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>opt-13b</p></td>
+<td class="text-center"><p>82</p></td>
+<td class="text-center"><p>198</p></td>
+<td class="text-center"><p>82.5</p></td>
+<td class="text-center"><p>220</p></td>
+<td class="text-center"><p>125</p></td>
+<td class="text-center"><p>61.8</p></td>
+<td class="text-center"><p>97</p></td>
+<td class="text-center"><p>123.7</p></td>
+</tr>
+<tr class="row-even"><td class="text-left"><p>llama-7b</p></td>
+<td class="text-center"><p>79.5</p></td>
+<td class="text-center"><p>167</p></td>
+<td class="text-center"><p>71.5</p></td>
+<td class="text-center"><p>214</p></td>
+<td class="text-center"><p>121</p></td>
+<td class="text-center"><p>58</p></td>
+<td class="text-center"><p>85</p></td>
+<td class="text-center"><p>113.7</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>llama-13b</p></td>
+<td class="text-center"><p><strong>79</strong></p></td>
+<td class="text-center"><p><strong>153</strong></p></td>
+<td class="text-center"><p><strong>70</strong></p></td>
+<td class="text-center"><p><strong>207</strong></p></td>
+<td class="text-center"><p><strong>119</strong></p></td>
+<td class="text-center"><p><strong>57.3</strong></p></td>
+<td class="text-center"><p><strong>83</strong></p></td>
+<td class="text-center"><p><strong>109.7</strong></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+<p><strong>Figure 1: Correlation between NLL and accuracy on commonsense QA benchmarks</strong></p>
+<p><img alt="" src="../_images/benchmark-2.png" /></p>
+<p>In the above figure, one can find that QA accuracy is roughly correlated to NLL. Thus NLL is able to reflect the “magnitude” of prediction level difference between models. A huge gap in NLL normally entails a huge performance gap.</p>
+<p>In the following sections, we provide a comprehensive evaluation of currently available LLM models and summarize their performance. Due to page limits, we only demonstrate partial evaluation results in this article. The full results can be found here <a class="reference external" href="https://docs.google.com/spreadsheets/d/1JYh4_pxNzmNA9I0YM2epgRA7VXBIeIGS64gPJBg5NHA/edit?usp=sharing">LLM Comparison Sheet 0501</a>. One can also evaluate their own LLM models using our LMFlow evaluation toolkit (<a class="reference external" href="https://optimalscale.github.io/LMFlow/examples/TASK_GUIDE.html">https://optimalscale.github.io/LMFlow/examples/TASK_GUIDE.html</a>).</p>
+<p>We also introduce our own finetuned model Robin-7b-Chat and compare its performance with some other popular models. One may try our online demo LMFlow-Chat at <a class="reference external" href="https://lmflow.com/">https://lmflow.com/</a>.</p>
+</section>
+<section id="chat-performance">
+<h2>Chat Performance<a class="headerlink" href="#chat-performance" title="Link to this heading">#</a></h2>
+<p>The first aspect of LLM is chat performance, especially multi-round chat. Based on our experimental results, we have the following observations:</p>
+<ol class="arabic simple">
+<li><p><em>Without finetuning, foundation models do not perform well on chat.</em></p></li>
+<li><p><em>After finetuning, foundation models can all obtain huge performance gain.</em></p></li>
+<li><p><em>Our Robin-Chat-7b achieves comparable chat performance with Vicuna-13b.</em></p></li>
+</ol>
+<p><strong>Table 3: NLL on multi-round chat datasets, Lower the better</strong></p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head text-left"><p></p></th>
+<th class="head text-center"><p>En_multi_round_chat</p></th>
+<th class="head text-center"><p>En_singe_round_chat</p></th>
+<th class="head text-center"><p>Average</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td class="text-left"><p>Bloom-7.1b</p></td>
+<td class="text-center"><p>188</p></td>
+<td class="text-center"><p>213</p></td>
+<td class="text-center"><p>200.5</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>Bloom-7.1b-chat</p></td>
+<td class="text-center"><p>164</p></td>
+<td class="text-center"><p>181</p></td>
+<td class="text-center"><p>172.5</p></td>
+</tr>
+<tr class="row-even"><td class="text-left"><p>Llama-7b</p></td>
+<td class="text-center"><p>163</p></td>
+<td class="text-center"><p>184</p></td>
+<td class="text-center"><p>173.5</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>Vicuna-7b</p></td>
+<td class="text-center"><p>155</p></td>
+<td class="text-center"><p>175</p></td>
+<td class="text-center"><p>165</p></td>
+</tr>
+<tr class="row-even"><td class="text-left"><p>Vicuna-13b</p></td>
+<td class="text-center"><p>150</p></td>
+<td class="text-center"><p>170</p></td>
+<td class="text-center"><p>160</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>Robin-Chat-7b</p></td>
+<td class="text-center"><p><strong>146</strong></p></td>
+<td class="text-center"><p><strong>164</strong></p></td>
+<td class="text-center"><p><strong>155</strong></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+<p>We also evaluate Vicuna-13b and our Robin-Chat-7b model qualitatively via use cases. It turns out that Robin-Chat-7b’s performance is comparable to Vicuna-13b. For example, both models can effortlessly refer to the conversation context with only slight mistakes, where Vicuna-13b sometimes doesn’t count the item correctly, while Robin-Chat-7b introduces only Mexican foods for international cuisine recommendations and omits other choices.</p>
+<p><strong>Table 4: Use cases of multi-round chat</strong></p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head text-left"><p>Round</p></th>
+<th class="head text-center"><p>Question</p></th>
+<th class="head text-center"><p>Vicuna-13b</p></th>
+<th class="head text-center"><p>Robin-Chat-7b</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td class="text-left"><p>R1</p></td>
+<td class="text-center"><p>Where is the capital of USA?</p></td>
+<td class="text-center"><p>The capital of the United States is <strong>Washington, D.C.</strong> (District of Columbia). It is located on the east coast of the country, in the northern part of the state of Maryland. Washington, D.C. is named after George Washington, the first president of the United States. It is a federal district, which means that it is not part of any state, but is instead directly administered by the federal government. The city is home to many national landmarks, including the White House, the Capitol Building, and the Supreme Court. It is also home to many museums, monuments, and other cultural institutions.</p></td>
+<td class="text-center"><p>The capital of the United States is <strong>Washington, D.C.</strong> (District of Columbia). It is located on the east coast of the USA, between the states of Maryland and Virginia.</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>R2</p></td>
+<td class="text-center"><p>Can you introduce some popular food there?</p></td>
+<td class="text-center"><p><p>Washington, D.C. is a city with a diverse food scene, and there are many popular restaurants and cuisines to choose from. Some popular foods in Washington, D.C. include:</p><p>- American Classics: There are many restaurants in Washington, D.C. that serve classic American dishes such as hamburgers, hot dogs, and apple pie.</p><p>- Seafood: Washington, D.C. is located near the Chesapeake Bay, which is known for its fresh seafood. Many restaurants in the city serve dishes such as crab cakes, shrimp, and oysters.</p><p>- Ethnic Cuisine: Washington, D.C. has a diverse population, and there are many restaurants that serve food from different parts of the world. Some popular ethnic cuisines include Mexican, Chinese, Japanese, and Korean.</p><p>- Barbecue: Washington, D.C. is also known for its barbecue restaurants. Many of these restaurants serve pulled pork, ribs, and brisket.</p><p>- <strong>Donuts</strong>: Washington, D.C. is home to the famous “Donut Happy” which is a famous donut shop with a variety of flavors and toppings.</p><p>- Street Food: There are also many street food options in Washington, D.C. such as food trucks and stands that serve food such as falafel, kebabs, and tacos.</p><p>These are just a few examples of the many types of food that are available in Washington, D.C. There is something for everyone in this city when it comes to food.</p><p></p></p></td>
+<td class="text-center"><p><p>Washington, D.C. is a diverse city with a wide range of cuisines to choose from. Here are some popular foods that you can try when you visit:</p><p>1. Soul food: This style of cuisine originated in the southern United States and is known for its bold flavors and hearty portions. Some popular soul food dishes in Washington, D.C. include fried chicken, collard greens, and macaroni and cheese.</p><p>2. Ethiopian cuisine: Ethiopian food is known for its spicy and flavorful dishes, and it is a popular choice in Washington, D.C. Some popular dishes include injera (a sourdough flatbread), wat (a spicy vegetable stew), and kitfo (a spicy meat dish).</p><p>3. Seafood: Washington, D.C. is located near the Chesapeake Bay and has easy access to fresh seafood. Some popular seafood dishes include crab cakes, oysters, and seafood pasta.</p><p>4. BBQ: Washington, D.C. has a thriving barbecue scene, with many restaurants serving up slow-cooked meats and sides. Some popular BBQ spots include Ben’s Chili Bowl, Pete’s Soul Vegetarian, and Smoke &amp; Barrel.</p><p>5. <strong>International cuisine:</strong> Washington, D.C. is a hub for international diplomacy, and as a result, it has a diverse culinary scene with many restaurants serving dishes from around the world. Some popular international cuisines include Japanese, Indian, and <strong>Mexican.</strong></p><p></p></p></td>
+</tr>
+<tr class="row-even"><td class="text-left"><p>R3</p></td>
+<td class="text-center"><p>Can you be more specific about the <strong>fifth item</strong> above?</p></td>
+<td class="text-center"><p><p>Sure, <strong>the fifth item I mentioned, “Barbecue,”</strong> refers to a style of cooking and food that originated in the southern United States. Barbecue is usually made with meat that is slow-cooked over an open flame or heated coals. The meat is often marinated and then sauced, and it is typically served with a variety of sides such as coleslaw, baked beans, and corn on the cob. Barbecue is a popular cuisine in Washington, D.C. and many of the restaurants serve pulled pork, ribs, and brisket. Some of the most popular barbecue restaurants in Washington, D.C. include:</p><p>- The Rib Pit</p><p>- The Smokehouse</p><p>- The BBQ Joint</p><p>- The Salt Line</p><p>- The Sovereign</p><p>These are just a few examples of the many barbecue restaurants in Washington, D.C. that offer delicious and flavorful barbecue dishes.</p></p></td>
+<td class="text-center"><p><p>Sure! Here are some specific recommendations for <strong>Mexican food</strong> in Washington, D.C.:</p><p>1. Margarita’s - This popular Mexican restaurant in Adams Morgan has been serving up authentic Mexican cuisine for over 30 years. Some of their most popular dishes include their tacos al pastor (grilled marinated pork), enchiladas (meat or cheese filled tortillas), and chiles rellenos (stuffed peppers).</p><p>2. El Comal - This cozy little restaurant in the Columbia Heights neighborhood has been serving up traditional Mexican cuisine for over 20 years. Some of their most popular dishes include their mole (a rich, spicy sauce made from a blend of chiles, spices, and chocolate), tacos de carnitas (crispy pork tacos), and chiles rellenos de queso (cheese stuffed peppers).</p><p>3. Casa Oaxaca - This upscale Mexican restaurant in the Shaw neighborhood has a menu that features a wide range of traditional and modern dishes. Some of their most popular dishes include their ceviche (raw fish marinated in lime juice and spices), mole negro (a rich, dark mole sauce), and chiles en nogada (a spicy dish made with poblano peppers, walnuts, and honey).</p><p>4. Taco King - This popular taco chain has several locations in the Washington, D.C. area, including one in the Adams Morgan neighborhood. They offer a wide range of tacos, burritos, and other Mexican dishes at affordable prices. Some of their most popular items include their carne asada tacos, chicken tacos, and quesadillas.</p><p>5. Los Tacos No. 1 - This small, casual taco stand in the Columbia Heights neighborhood has a menu that features a wide range of tacos, burritos, and other Mexican dishes. Some of their most popular items include their adobada tacos (marinated pork tacos), al pastor tacos (grilled marinated pork tacos), and chicken tacos.</p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+<p><strong>Experiment Details</strong></p>
+<p>Our dataset comes from real conversation history between humans and ChatGPT. We collect 183 English conversations with a total of 447 rounds of chats, where the original conversation history is already multi-round. For single-round NLL evaluation, we break a multi-round conversation into multiple single-round text pairs. For example, with a conversation history</p>
+<div class="highlight-none notranslate"><div class="highlight"><pre><span></span>Q1, A1
+
+Q2, A2
+
+Q3, A3
+</pre></div>
+</div>
+<p>where Q stands for Question, A stands for answer, we can get three single-round examples:</p>
+<div class="highlight-none notranslate"><div class="highlight"><pre><span></span>Context: Q1
+
+Output: A1
+
+Context: Q1, A1, Q2
+
+Output: A2
+
+Context: Q1, A1, Q2, A2, Q3
+
+Ouptut: A3
+</pre></div>
+</div>
+</section>
+<section id="commonsense-performance">
+<h2>CommonSense Performance<a class="headerlink" href="#commonsense-performance" title="Link to this heading">#</a></h2>
+<p>Another important aspect of an LLM model is its common sense ability, where a model should acquire a certain level of factual knowledge and utilize them properly under different scenarios. Regarding this aspect of the ability, we found:</p>
+<ol class="arabic simple">
+<li><p><em>Finetuning on chat dataset results in commonsense degradation.</em></p></li>
+<li><p><em>Our Robin-Chat-7b model still achieves a competitive performance.</em></p></li>
+</ol>
+<p><strong>Table 5: Accuracy results in commonsense QA benchmarks</strong></p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head text-left"><p></p></th>
+<th class="head text-center"><p>winogrand</p></th>
+<th class="head text-center"><p>boolq</p></th>
+<th class="head text-center"><p>arc_easy</p></th>
+<th class="head text-center"><p>hellaswag</p></th>
+<th class="head text-center"><p>piqa</p></th>
+<th class="head text-center"><p>obqa</p></th>
+<th class="head text-center"><p>arc_c</p></th>
+<th class="head text-center"><p>Average</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td class="text-left"><p>Bloom-7.1b</p></td>
+<td class="text-center"><p>64.4</p></td>
+<td class="text-center"><p>62.9</p></td>
+<td class="text-center"><p>65.0</p></td>
+<td class="text-center"><p>59.6</p></td>
+<td class="text-center"><p>73.6</p></td>
+<td class="text-center"><p>35.8</p></td>
+<td class="text-center"><p>33.4</p></td>
+<td class="text-center"><p>56.4</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>Bloom-7.1b-chat</p></td>
+<td class="text-center"><p>60.3</p></td>
+<td class="text-center"><p>56.8</p></td>
+<td class="text-center"><p>61.3</p></td>
+<td class="text-center"><p>58.7</p></td>
+<td class="text-center"><p>72.7</p></td>
+<td class="text-center"><p>37.8</p></td>
+<td class="text-center"><p>38.7</p></td>
+<td class="text-center"><p>55.2</p></td>
+</tr>
+<tr class="row-even"><td class="text-left"><p>Llama-7b</p></td>
+<td class="text-center"><p>67.9</p></td>
+<td class="text-center"><p>73.2</p></td>
+<td class="text-center"><p>67.3</p></td>
+<td class="text-center"><p>73.0</p></td>
+<td class="text-center"><p>78.4</p></td>
+<td class="text-center"><p>42.4</p></td>
+<td class="text-center"><p>41.4</p></td>
+<td class="text-center"><p>62.7</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>Vicuna-7b</p></td>
+<td class="text-center"><p>63.7</p></td>
+<td class="text-center"><p>77.4</p></td>
+<td class="text-center"><p>63.1</p></td>
+<td class="text-center"><p>68.8</p></td>
+<td class="text-center"><p>76.3</p></td>
+<td class="text-center"><p>39.6</p></td>
+<td class="text-center"><p>38.7</p></td>
+<td class="text-center"><p>61.1</p></td>
+</tr>
+<tr class="row-even"><td class="text-left"><p>Vicuna-13b</p></td>
+<td class="text-center"><p>66.2</p></td>
+<td class="text-center"><p>79.9</p></td>
+<td class="text-center"><p>64.7</p></td>
+<td class="text-center"><p>73.0</p></td>
+<td class="text-center"><p>77.6</p></td>
+<td class="text-center"><p>41.6</p></td>
+<td class="text-center"><p>40.4</p></td>
+<td class="text-center"><p>63.3</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>Robin-Chat-7b</p></td>
+<td class="text-center"><p>64.7</p></td>
+<td class="text-center"><p>75.2</p></td>
+<td class="text-center"><p>69.8</p></td>
+<td class="text-center"><p>72.4</p></td>
+<td class="text-center"><p>76.6</p></td>
+<td class="text-center"><p>39.0</p></td>
+<td class="text-center"><p>42.9</p></td>
+<td class="text-center"><p>62.9</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+<p><strong>Table 6: NLL results in corpus of commonsense QA benchmarks</strong></p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head text-left"><p></p></th>
+<th class="head text-center"><p>winogrand</p></th>
+<th class="head text-center"><p>boolq</p></th>
+<th class="head text-center"><p>arc_easy</p></th>
+<th class="head text-center"><p>hellaswag</p></th>
+<th class="head text-center"><p>piqa</p></th>
+<th class="head text-center"><p>obqa</p></th>
+<th class="head text-center"><p>arc_c</p></th>
+<th class="head text-center"><p>Average</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td class="text-left"><p>Bloom-7.1b</p></td>
+<td class="text-center"><p>96.0</p></td>
+<td class="text-center"><p>254</p></td>
+<td class="text-center"><p>89</p></td>
+<td class="text-center"><p>266</p></td>
+<td class="text-center"><p>147</p></td>
+<td class="text-center"><p>69</p></td>
+<td class="text-center"><p>106.5</p></td>
+<td class="text-center"><p>146.7</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>Bloom-7.1b-chat</p></td>
+<td class="text-center"><p>85.0</p></td>
+<td class="text-center"><p>215</p></td>
+<td class="text-center"><p>81.5</p></td>
+<td class="text-center"><p>237</p></td>
+<td class="text-center"><p>130</p></td>
+<td class="text-center"><p>62.5</p></td>
+<td class="text-center"><p>96</p></td>
+<td class="text-center"><p>129.5</p></td>
+</tr>
+<tr class="row-even"><td class="text-left"><p>Llama-7b</p></td>
+<td class="text-center"><p>79.5</p></td>
+<td class="text-center"><p>167</p></td>
+<td class="text-center"><p>71.5</p></td>
+<td class="text-center"><p>214</p></td>
+<td class="text-center"><p>121</p></td>
+<td class="text-center"><p>58</p></td>
+<td class="text-center"><p>85</p></td>
+<td class="text-center"><p>113.7</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>Vicuna-7b</p></td>
+<td class="text-center"><p>97.0</p></td>
+<td class="text-center"><p>219</p></td>
+<td class="text-center"><p>89</p></td>
+<td class="text-center"><p>258</p></td>
+<td class="text-center"><p>151</p></td>
+<td class="text-center"><p>72</p></td>
+<td class="text-center"><p>105</p></td>
+<td class="text-center"><p>141.6</p></td>
+</tr>
+<tr class="row-even"><td class="text-left"><p>Vicuna-13b</p></td>
+<td class="text-center"><p>95.5</p></td>
+<td class="text-center"><p>207</p></td>
+<td class="text-center"><p>88.5</p></td>
+<td class="text-center"><p>262</p></td>
+<td class="text-center"><p>149</p></td>
+<td class="text-center"><p>73</p></td>
+<td class="text-center"><p>105</p></td>
+<td class="text-center"><p>140.0</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>Robin-Chat-7b</p></td>
+<td class="text-center"><p>89.0</p></td>
+<td class="text-center"><p>200</p></td>
+<td class="text-center"><p>81</p></td>
+<td class="text-center"><p>240</p></td>
+<td class="text-center"><p>139</p></td>
+<td class="text-center"><p>65.5</p></td>
+<td class="text-center"><p>96.5</p></td>
+<td class="text-center"><p>130.1</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+<p>It is surprising to find that finetuning actually leads to commonsense degradation. A more well-known consequence of this issue is hallucination, where models tend to give plausible but factually incorrect or irrelevant answers. One explanation of this phenomenon is that LLMs “forget” the pretrained knowledge during finetuning when attempting to fit the finetune dataset.</p>
+<p>As one can see from the table above, both Vicuna and Robin suffer from this commonsense degradation issue. However, Robin suffers less. We conjecture that this improvement is brought by finetuning with smaller-sized but higher-quality dataset. Vicuna uses sharegpt data with ~900MB size and was finetuned for 3 epochs, while Robin uses multiple sources of dataset with ~600MB size and was only finetuned for 2 epochs. This suggests that in the finetuning stage, data quality plays a critical role.</p>
+<p>This improvement gives Vicuna-13b a run for its money when competing with Robin. The following examples show that Robin-Chat-7b possesses a comparable level of commonsense when compared with Vicuna-13b.</p>
+<p><strong>Table 7: Use cases of commonsense QA</strong></p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head text-left"><p>Question</p></th>
+<th class="head text-center"><p>Vicuna-13b</p></th>
+<th class="head text-center"><p>Robin-Chat-7b</p></th>
+<th class="head text-center"><p>Note</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td class="text-left"><p><p>I go to a friend’s home but the door is locked. However, I smell coal gas. Should I break the window?</p><p></p></p></td>
+<td class="text-center"><p>If you smell coal gas, it is important to get out of the house as quickly as possible and call the fire department or the gas company. Coal gas is a flammable gas that can be dangerous and even deadly if it is not properly ventilated. If the door is locked, you <strong>should not break the window because it could cause a fire or explosion</strong>. Instead, try to find another way to escape the house or call for help. If you cannot leave the house safely, try to stay calm and wait for help to arrive.</p></td>
+<td class="text-center"><p>No, it is not recommended to break a window in this situation. <strong>Breaking a window can be dangerous and can cause injury or damage to yourself or others</strong>. Additionally, it is <strong>illegal</strong> to break a window to enter a property without permission. Instead, you should try to find another way to enter the home, such as by using a key or by calling the homeowner or a locksmith for assistance. If you are unable to enter the home, you should call the emergency services or the police for assistance. In the meantime, it is important to <strong>avoid using any sources of heat</strong>, such as a lighter or matches, as these can cause a gas leak or explosion. If you smell gas, you should evacuate the area immediately and call the emergency services or the gas company for assistance.</p></td>
+<td class="text-center"><p><em>Breaking the window helps the ventilation and will not cause a gas explosion directly. However, it is indeed illegal and may injure yourself when breaking the window.</em></p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>Can I bring my pet ferret to that beautiful pacific island of USA to attend ICML 2023?</p></td>
+<td class="text-center"><p>I’m sorry, but I am <strong>not aware of</strong> the specific policies regarding pets at the ICML 2023 conference or the island you are referring to. It would be best to check with the conference organizers or the local authorities on the island to determine if ferrets are allowed and if there are any necessary procedures or requirements for bringing a pet to the conference.</p></td>
+<td class="text-center"><p><p>It is <strong>not recommended</strong> to bring a pet ferret to the United States, including to attend ICML 2023. Ferrets are not permitted as pets in many areas of the United States, and <strong>it is illegal</strong> to bring them into the country without proper documentation and licensing. Additionally, ferrets can <strong>carry diseases</strong> that are not common in humans but can be dangerous to people, such as toxoplasmosis and salmonellosis. It is important to prioritize the health and safety of yourself and others by not bringing a pet ferret to a crowded event like ICML 2023.</p><p></p></p></td>
+<td class="text-center"><p><em>ICML 2023 will be held in Hawaii, where ferrets are strictly prohibited as pets because they are potential carriers of the rabies virus.</em></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+<p><strong>Experiment Details</strong></p>
+<p>Here, we make use of  Eleuther AI’s LM evaluation harness repository (<a class="github reference external" href="https://github.com/EleutherAI/lm-evaluation-harness">EleutherAI/lm-evaluation-harness</a>) to get QA accuracy results. We also evaluate all models’ NLL metrics on their datasets, with their questions as contexts and answers as output sentences.</p>
+</section>
+<section id="instruction-following">
+<h2>Instruction Following<a class="headerlink" href="#instruction-following" title="Link to this heading">#</a></h2>
+<p>The final aspect we evaluate is instruction following, where a model is expected to loyally obey the user’s instructions, given the instruction being ethically or legally appropriate. It turns out that,</p>
+<ol class="arabic simple">
+<li><p><em>Without finetuning, foundation models do not perform well on instruction following.</em></p></li>
+<li><p><em>Finetuning on chat dataset helps improve the performance.</em></p></li>
+<li><p><em>Our Robin-Chat is still comparable with Vicuna-13b.</em></p></li>
+</ol>
+<p><strong>Table 8: NLL on instruction following chat datasets</strong></p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head text-left"><p></p></th>
+<th class="head text-center"><p>gpt4_en_instruction</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td class="text-left"><p>Alpaca</p></td>
+<td class="text-center"><p>222</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>Llama-7b</p></td>
+<td class="text-center"><p>214</p></td>
+</tr>
+<tr class="row-even"><td class="text-left"><p>Vicuna-7b</p></td>
+<td class="text-center"><p>211</p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>Vicuna-13b</p></td>
+<td class="text-center"><p>206</p></td>
+</tr>
+<tr class="row-even"><td class="text-left"><p>Robin-Chat-7b</p></td>
+<td class="text-center"><p><strong>180</strong></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+<p>Following examples show that Robin-Chat-7b can generate similar responses on instruction following tasks when compared with Vicuna-13b.</p>
+<p><strong>Table 9: Uses cases of instruction following</strong></p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head text-left"><p>Question</p></th>
+<th class="head text-center"><p>Vicuna-13b</p></th>
+<th class="head text-center"><p>Robin-Chat-7b</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td class="text-left"><p>Can you write quicksort?</p></td>
+<td class="text-center"><p><p>Yes, here is a simple implementation of the quicksort algorithm in Python:</p><p>def quicksort(arr):</p><p><code class="docutils literal notranslate">&#160;&#160;&#160; </code>if len(arr) &lt;= 1:</p><p><code class="docutils literal notranslate">&#160;&#160;&#160;&#160;&#160;&#160;&#160; </code>return arr</p><p><code class="docutils literal notranslate">&#160;&#160;&#160; </code>pivot = arr[len(arr) // 2]</p><p><code class="docutils literal notranslate">&#160;&#160;&#160; </code>left = [x for x in arr if x &lt; pivot]</p><p><code class="docutils literal notranslate">&#160;&#160;&#160; </code>middle = [x for x in arr if x == pivot]</p><p><code class="docutils literal notranslate">&#160;&#160;&#160; </code>right = [x for x in arr if x &gt; pivot]</p><p><code class="docutils literal notranslate">&#160;&#160;&#160; </code>return quicksort(left) + middle + quicksort(right)</p><p></p><p>This implementation uses the middle element of the array as the pivot, and then partitions the array into three sub-arrays: one containing all elements that are less than the pivot, one containing all elements that are equal to the pivot, and one containing all elements that are greater than the pivot. It then recursively applies the quicksort algorithm to the left and right sub-arrays, and returns the concatenation of the sorted left sub-array, the unsorted middle sub-array, and the sorted right sub-array.</p><p>This implementation has a time complexity of O(n log n) due to the recursive calls, although it could potentially be faster in practice if the pivot is carefully chosen.</p><p></p></p></td>
+<td class="text-center"><p><p>Sure, here is a simple implementation of the quicksort algorithm in Python:</p><p>def quicksort(arr):</p><p><code class="docutils literal notranslate">&#160;&#160;&#160; </code>if len(arr) &lt;= 1:</p><p><code class="docutils literal notranslate">&#160;&#160;&#160;&#160;&#160;&#160;&#160; </code>return arr</p><p><code class="docutils literal notranslate">&#160;&#160;&#160; </code>pivot = arr[len(arr) // 2]</p><p><code class="docutils literal notranslate">&#160;&#160;&#160; </code>left = [x for x in arr if x &lt; pivot]</p><p><code class="docutils literal notranslate">&#160;&#160;&#160; </code>middle = [x for x in arr if x == pivot]</p><p><code class="docutils literal notranslate">&#160;&#160;&#160; </code>right = [x for x in arr if x &gt; pivot]</p><p><code class="docutils literal notranslate">&#160;&#160;&#160; </code>return quicksort(left) + middle + quicksort(right)</p><p></p><p>This implementation chooses the middle element of the array as the pivot, and then partitions the array into three parts: elements less than the pivot, elements equal to the pivot, and elements greater than the pivot. It then recursively applies the quicksort algorithm to the left and right subarrays, and concatenates the results with the middle subarray to produce the final sorted array.</p><p>Note that this implementation uses a single pass through the array, which is the most efficient way to sort an array using quicksort.</p></p></td>
+</tr>
+<tr class="row-odd"><td class="text-left"><p>Instruction: Find synonyms for the given word. You need to write down how the provided synonyms differ from the original word in terms of meaning, usage, etc. adversity</p></td>
+<td class="text-center"><p><p>Synonyms for “adversity” include:</p><p>- Difficulty</p><p>- Hardship</p><p>- Trouble</p><p>- Misfortune</p><p>- Adversity</p><p>- Difficulties</p><p>- Hard times</p><p>All of these words refer to challenging or unpleasant circumstances or situations. However, “adversity” specifically connotes a difficult or hostile situation that is caused by external factors, such as an enemy or a natural disaster. “Difficulty” and “hardship” can be caused by external or internal factors, and can refer to a wide range of challenges, from mild inconveniences to major setbacks. “Trouble” often connotes a specific type of difficulty or distress, such as financial trouble or legal trouble. “Misfortune” specifically refers to bad luck or ill fate, while “hard times” refers to a general period of economic or social distress.</p></p></td>
+<td class="text-center"><p><p>1. Hardship</p><p>2. Struggle</p><p>3. Difficulty</p><p>4. Oppression</p><p>5. Trial</p><p>6. Affliction</p><p>7. Distress</p><p>8. Calamity</p><p>9. Tribulation</p><p>10. Hardship</p><p>Adversity refers to difficult or challenging circumstances or conditions. Hardship, struggle, difficulty, oppression, trial, affliction, distress, and calamity are all words that can be used to describe adversity, but they often imply a more specific or intense situation. For example, hardship can refer to a general difficulty or challenge, while struggle implies a more intense battle or conflict. Difficulty and oppression imply a more specific obstacle or injustice, while trial and affliction suggest a more specific test or suffering. Trial can also refer to a legal proceeding, while tribulation implies a more general time of suffering or persecution.</p></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+<p><strong>Experiment Details</strong></p>
+<p>In order to make evaluations for models’ instruction following ability, we collected GPT-4 instruction data from the project “Instruction Tuning with GPT-4”, which provides many instruction tuning data labeled by GPT-4. We sample 1,000 English data and manually filtered examples with the following issues, where 767 effective samples remain after the filtering:</p>
+<ol class="arabic simple">
+<li><p>Long response with too many nonsense words</p></li>
+<li><p>Incomplete input texts</p></li>
+<li><p>Specific domains involving chemistry/biology, where most LLM models do not possess the knowledge and always fail</p></li>
+</ol>
+</section>
+<section id="conclusion">
+<h2>Conclusion<a class="headerlink" href="#conclusion" title="Link to this heading">#</a></h2>
+<p>In this article, we introduce LMFlow’s evaluation framework, which uses NLL metric to reflect LLM models’ ability. NLL provides a good metric to evaluate different aspects of a LLM model. According to our evaluation results, Robin-7b achieves on-par performance when compared with Vicuna-13b. As our Robin-7b model is finetuned with different sources of dataset instead of sharegpt only, this shows that Vicuna can be further improved or surpassed with smaller-sized models and better dataset.</p>
+<p>The checkpoint of Robin-7b is now available for engineers and researchers to download and use (<a class="github reference external" href="https://github.com/OptimalScale/LMFlow#model-zoo">OptimalScale/LMFlow</a>). Its effectiveness demonstrates that a multi-aspect evaluation is indeed essential to the development of LLMs.</p>
+</section>
+<section id="references">
+<h2>References<a class="headerlink" href="#references" title="Link to this heading">#</a></h2>
+<p>Vicuna Chatbot Arena: <a class="reference external" href="https://chat.lmsys.org/?arena">https://chat.lmsys.org/?arena</a></p>
+<p>lm-evaluation-harness: <a class="github reference external" href="https://github.com/EleutherAI/lm-evaluation-harness">EleutherAI/lm-evaluation-harness</a></p>
+<p>LMFlow: <a class="github reference external" href="https://github.com/OptimalScale/LMFlow">OptimalScale/LMFlow</a></p>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">Blogs</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../examples/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">Examples</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#introduction">Introduction</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#metric">Metric</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#chat-performance">Chat Performance</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#commonsense-performance">CommonSense Performance</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#instruction-following">Instruction Following</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#conclusion">Conclusion</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#references">References</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/blogs/benchmark.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/blogs/index.html b/blogs/index.html
new file mode 100644
index 000000000..2224f2fcc
--- /dev/null
+++ b/blogs/index.html
@@ -0,0 +1,507 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>Blogs &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'blogs/index';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="LMFlow Benchmark: An Automatic Evaluation Framework for Open-Source LLMs" href="benchmark.html" />
+    <link rel="prev" title="LMFlow" href="../index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="#">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="#">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="benchmark.html">LMFlow Benchmark: An Automatic Evaluation Framework for Open-Source LLMs</a></li>
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Blogs</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="blogs">
+<h1>Blogs<a class="headerlink" href="#blogs" title="Link to this heading">#</a></h1>
+<section id="id1">
+<h2>2023<a class="headerlink" href="#id1" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="benchmark.html">LMFlow Benchmark: An Automatic Evaluation Framework for Open-Source LLMs</a></li>
+</ul>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">LMFlow</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="benchmark.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">LMFlow Benchmark: An Automatic Evaluation Framework for Open-Source LLMs</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#id1">2023</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/blogs/index.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/examples/DATASETS.html b/examples/DATASETS.html
new file mode 100644
index 000000000..fc76c41b5
--- /dev/null
+++ b/examples/DATASETS.html
@@ -0,0 +1,972 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>Dataset &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'examples/DATASETS';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Checkpoints" href="checkpoints.html" />
+    <link rel="prev" title="Examples" href="index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active"><a class="current reference internal" href="#">Dataset</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="checkpoints.html">Checkpoints</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="finetuning.html">Finetuning</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="reward_modeling.html">Reward Modeling</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="raft.html">RAFT</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="TASK_GUIDE.html">LMFlow Benchmark Guide</a></li>
+
+
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="index.html" class="nav-link">Examples</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Dataset</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="dataset">
+<h1>Dataset<a class="headerlink" href="#dataset" title="Link to this heading">#</a></h1>
+<ul class="simple">
+<li><p><a class="reference internal" href="../autoapi/lmflow/datasets/dataset/index.html#module-lmflow.datasets.dataset" title="lmflow.datasets.dataset"><span class="xref myst py py-mod">Dataset</span></a></p>
+<ul>
+<li><p><a class="reference internal" href="#dataset-format-in-general"><span class="xref myst">Dataset Format in General</span></a></p></li>
+<li><p><a class="reference internal" href="#supported-dataset-and-detailed-formats"><span class="xref myst">Supported Dataset and Detailed Formats</span></a></p>
+<ul>
+<li><p><a class="reference internal" href="#conversation"><span class="xref myst">Conversation</span></a></p>
+<ul>
+<li><p><a class="reference internal" href="#data-format"><span class="xref myst">Data Format</span></a></p></li>
+<li><p><a class="reference internal" href="#conversation-template"><span class="xref myst">Conversation Template</span></a></p></li>
+<li><p><a class="reference internal" href="#customize-conversation-template"><span class="xref myst">Customize Conversation Template</span></a></p></li>
+</ul>
+</li>
+<li><p><a class="reference internal" href="#textonly"><span class="xref myst">TextOnly</span></a></p></li>
+<li><p><a class="reference internal" href="#text2text"><span class="xref myst">Text2Text</span></a></p></li>
+<li><p><a class="reference internal" href="#paired-conversation"><span class="xref myst">Paired Conversation</span></a></p></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+<p>We provide several available datasets under <code class="docutils literal notranslate"><span class="pre">data</span></code>. You may download them all by running:</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="nb">cd</span><span class="w"> </span>data<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span>./download.sh<span class="w"> </span>all<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span><span class="nb">cd</span><span class="w"> </span>-
+</pre></div>
+</div>
+<p>You can replace <code class="docutils literal notranslate"><span class="pre">all</span></code> with a specific dataset name to only download that dataset (e.g. <code class="docutils literal notranslate"><span class="pre">./download.sh</span> <span class="pre">alpaca</span></code>).</p>
+<p>Customized datasets are strongly encouraged, since this way users can apply
+their own prompt engineering techniques over various source datasets. As long
+as the generated dataset following the format below, they can be accepted as
+the input of our pipelines :hugs:</p>
+<section id="dataset-format-in-general">
+<h2>Dataset Format in General<a class="headerlink" href="#dataset-format-in-general" title="Link to this heading">#</a></h2>
+<p>To specify the input for model finetune, users can provide a list of <code class="docutils literal notranslate"><span class="pre">.json</span></code>
+files under a specified dataset directory. For example,</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="p">|</span>-<span class="w"> </span>path_to_dataset
+<span class="w">  </span><span class="p">|</span>-<span class="w"> </span>data_1.json
+<span class="w">  </span><span class="p">|</span>-<span class="w"> </span>data_2.json
+<span class="w">  </span><span class="p">|</span>-<span class="w"> </span>another_data.json
+<span class="w">  </span><span class="p">|</span>-<span class="w"> </span>...
+</pre></div>
+</div>
+<p>For inference, we currently only support a single <code class="docutils literal notranslate"><span class="pre">.json</span></code> file.</p>
+<p>Each json file shall have the following format (three instances with four keys
+for example),</p>
+<div class="highlight-json notranslate"><div class="highlight"><pre><span></span><span class="p">{</span>
+<span class="w">  </span><span class="nt">&quot;type&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;TYPE&quot;</span><span class="p">,</span>
+<span class="w">  </span><span class="nt">&quot;instances&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">    </span><span class="p">{</span>
+<span class="w">        </span><span class="nt">&quot;KEY_1&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;VALUE_1.1&quot;</span><span class="p">,</span>
+<span class="w">        </span><span class="nt">&quot;KEY_2&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;VALUE_1.2&quot;</span><span class="p">,</span>
+<span class="w">        </span><span class="nt">&quot;KEY_3&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;VALUE_1.3&quot;</span><span class="p">,</span>
+<span class="w">        </span><span class="nt">&quot;KEY_4&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;VALUE_1.4&quot;</span><span class="p">,</span>
+<span class="w">    </span><span class="p">},</span>
+<span class="w">    </span><span class="p">{</span>
+<span class="w">        </span><span class="nt">&quot;KEY_1&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;VALUE_2.1&quot;</span><span class="p">,</span>
+<span class="w">        </span><span class="nt">&quot;KEY_2&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;VALUE_2.2&quot;</span><span class="p">,</span>
+<span class="w">        </span><span class="nt">&quot;KEY_3&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;VALUE_2.3&quot;</span><span class="p">,</span>
+<span class="w">        </span><span class="nt">&quot;KEY_4&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;VALUE_2.4&quot;</span><span class="p">,</span>
+<span class="w">    </span><span class="p">},</span>
+<span class="w">    </span><span class="p">{</span>
+<span class="w">        </span><span class="nt">&quot;KEY_1&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;VALUE_3.1&quot;</span><span class="p">,</span>
+<span class="w">        </span><span class="nt">&quot;KEY_2&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;VALUE_3.2&quot;</span><span class="p">,</span>
+<span class="w">        </span><span class="nt">&quot;KEY_3&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;VALUE_3.3&quot;</span><span class="p">,</span>
+<span class="w">        </span><span class="nt">&quot;KEY_4&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;VALUE_3.4&quot;</span><span class="p">,</span>
+<span class="w">    </span><span class="p">},</span>
+<span class="w">  </span><span class="p">]</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+<p>where the <code class="docutils literal notranslate"><span class="pre">TYPE</span></code> indicates the dataset type and defines the set of keys
+<code class="docutils literal notranslate"><span class="pre">{</span> <span class="pre">KEY_1,</span> <span class="pre">KEY_2,</span> <span class="pre">...</span> <span class="pre">}</span></code> and their corresponding interpretations. The list of
+supported types are listed as follows.</p>
+</section>
+<section id="supported-dataset-and-detailed-formats">
+<h2>Supported Dataset and Detailed Formats<a class="headerlink" href="#supported-dataset-and-detailed-formats" title="Link to this heading">#</a></h2>
+<section id="conversation">
+<h3>Conversation<a class="headerlink" href="#conversation" title="Link to this heading">#</a></h3>
+<section id="data-format">
+<h4>Data Format<a class="headerlink" href="#data-format" title="Link to this heading">#</a></h4>
+<p>Conversational data are commonly used in sft process. We currently support conversational data in ShareGPT format:</p>
+<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3">
+<summary class="sd-summary-title sd-card-header">
+<span class="sd-summary-text">A conversation dataset</span><span class="sd-summary-state-marker sd-summary-chevron-right"><svg version="1.1" width="1.5em" height="1.5em" class="sd-octicon sd-octicon-chevron-right" viewBox="0 0 24 24" aria-hidden="true"><path d="M8.72 18.78a.75.75 0 0 1 0-1.06L14.44 12 8.72 6.28a.751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018l6.25 6.25a.75.75 0 0 1 0 1.06l-6.25 6.25a.75.75 0 0 1-1.06 0Z"></path></svg></span></summary><div class="sd-summary-content sd-card-body docutils">
+<div class="highlight-json notranslate"><div class="highlight"><pre><span></span><span class="p">{</span>
+<span class="w">  </span><span class="nt">&quot;type&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;conversation&quot;</span><span class="p">,</span>
+<span class="w">  </span><span class="nt">&quot;instances&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">    </span><span class="p">{</span>
+<span class="w">      </span><span class="nt">&quot;conversation_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;CONVERSATION_ID&quot;</span><span class="p">,</span>
+<span class="w">      </span><span class="nt">&quot;system&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SYSTEM_PROPMT&quot;</span><span class="p">,</span>
+<span class="w">      </span><span class="nt">&quot;tools&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span><span class="s2">&quot;TOOL_DESCRIPTION_1&quot;</span><span class="p">,</span><span class="s2">&quot;TOOL_DESCRIPTION_2&quot;</span><span class="p">,</span><span class="s2">&quot;TOOL_DESCRIPTION_X&quot;</span><span class="p">],</span>
+<span class="w">      </span><span class="nt">&quot;messages&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span>
+<span class="w">            </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;USER_INPUT_1&quot;</span>
+<span class="w">        </span><span class="p">},</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span>
+<span class="w">            </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;ASSISTANT_RESPONSE_1&quot;</span>
+<span class="w">        </span><span class="p">},</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span>
+<span class="w">            </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;USER_INPUT_2&quot;</span>
+<span class="w">        </span><span class="p">},</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span>
+<span class="w">            </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;ASSISTANT_RESPONSE_2&quot;</span>
+<span class="w">        </span><span class="p">}</span>
+<span class="w">      </span><span class="p">]</span>
+<span class="w">    </span><span class="p">},</span>
+<span class="w">    </span><span class="p">{</span>
+<span class="w">      </span><span class="nt">&quot;conversation_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;CONVERSATION_ID&quot;</span><span class="p">,</span>
+<span class="w">      </span><span class="nt">&quot;system&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SYSTEM_PROPMT&quot;</span><span class="p">,</span>
+<span class="w">      </span><span class="nt">&quot;tools&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span><span class="s2">&quot;TOOL_DESCRIPTION_1&quot;</span><span class="p">],</span>
+<span class="w">      </span><span class="nt">&quot;messages&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span>
+<span class="w">            </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;USER_INPUT_1&quot;</span>
+<span class="w">        </span><span class="p">},</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span>
+<span class="w">            </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;ASSISTANT_RESPONSE_1&quot;</span>
+<span class="w">        </span><span class="p">}</span>
+<span class="w">      </span><span class="p">]</span>
+<span class="w">    </span><span class="p">}</span>
+<span class="w">  </span><span class="p">]</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+</div>
+</details><p>Data types:</p>
+<ul class="simple">
+<li><p><code class="docutils literal notranslate"><span class="pre">conversation_id</span></code>: <code class="docutils literal notranslate"><span class="pre">Optional[Any]</span></code>. An identifier for the conversation. <code class="docutils literal notranslate"><span class="pre">conversation_id</span></code> is only for convience of tracking the conversation and will not be used in the pipeline.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">system</span></code>: <code class="docutils literal notranslate"><span class="pre">Optional[string]</span></code>. A system prompt that is used to start the conversation.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">tools</span></code>: <code class="docutils literal notranslate"><span class="pre">Optional[List[string]]</span></code>. A list of tools that are used in the conversation.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">messages</span></code>: <code class="docutils literal notranslate"><span class="pre">List[Dict]</span></code>. A list of messages in the conversation. Each message contains the following fields:</p>
+<ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">role</span></code>: <code class="docutils literal notranslate"><span class="pre">string</span></code>. The role of the message. It can be either <code class="docutils literal notranslate"><span class="pre">user</span></code> or <code class="docutils literal notranslate"><span class="pre">assistant</span></code>.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">content</span></code>: <code class="docutils literal notranslate"><span class="pre">string</span></code>. The content of the message.</p></li>
+</ul>
+</li>
+</ul>
+<blockquote>
+<div><p>We are working on supporting customized message keys and role names. Please stay tuned.</p>
+</div></blockquote>
+<p>Tips:</p>
+<ul class="simple">
+<li><p>Please make sure the messages are:</p>
+<ol class="arabic simple">
+<li><p>Start with an user message.</p></li>
+<li><p>In the correct order. The pipeline will not check the order of the messages.</p></li>
+<li><p>In pairs of user and assistant (i.e., the length of the messages should be even). If the conversation ends with the user, the pipeline will trim the last user message.</p></li>
+<li><p>Make sure the <code class="docutils literal notranslate"><span class="pre">content</span></code>s are not empty. If the <code class="docutils literal notranslate"><span class="pre">content</span></code> is empty, the pipeline will add a space to it.</p></li>
+</ol>
+</li>
+</ul>
+</section>
+<section id="conversation-template">
+<h4>Conversation Template<a class="headerlink" href="#conversation-template" title="Link to this heading">#</a></h4>
+<p>Conversations should be formatted before feeding into the model. As of now, we’ve preset the conversation template for following models:</p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head"><p>Template Name</p></th>
+<th class="head"><p>Filled Example</p></th>
+<th class="head"><p>Detailed Template</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p><code class="docutils literal notranslate"><span class="pre">chatglm3</span></code></p></td>
+<td><p><code class="docutils literal notranslate"><span class="pre">[gMASK]sop&lt;|system|&gt;</span></code><br><code class="docutils literal notranslate"> <span class="pre">You</span> <span class="pre">are</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|user|&gt;</span></code><br><code class="docutils literal notranslate"> <span class="pre">Who</span> <span class="pre">are</span> <span class="pre">you?&lt;|assistant|&gt;</span></code><br><code class="docutils literal notranslate"> <span class="pre">I</span> <span class="pre">am</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|user|&gt;</span></code><br><code class="docutils literal notranslate"> <span class="pre">How</span> <span class="pre">old</span> <span class="pre">are</span> <span class="pre">you?&lt;|assistant|&gt;</span></code><br><code class="docutils literal notranslate"> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">age</span> <span class="pre">like</span> <span class="pre">humans</span> <span class="pre">do.</span> <span class="pre">I</span> <span class="pre">exist</span> <span class="pre">as</span> <span class="pre">a</span> <span class="pre">piece</span> <span class="pre">of</span> <span class="pre">software,</span> <span class="pre">so</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">have</span> <span class="pre">a</span> <span class="pre">concept</span> <span class="pre">of</span> <span class="pre">age</span> <span class="pre">in</span> <span class="pre">the</span> <span class="pre">traditional</span> <span class="pre">sense.</span></code></p></td>
+<td><p><a class="reference internal" href="supported_conversation_template.html#chatglm3"><span class="std std-ref">Link</span></a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="docutils literal notranslate"><span class="pre">chatml</span></code></p></td>
+<td><p><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;system</span></code><br><code class="docutils literal notranslate"><span class="pre">You</span> <span class="pre">are</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;user</span></code><br><code class="docutils literal notranslate"><span class="pre">Who</span> <span class="pre">are</span> <span class="pre">you?&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;assistant</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">am</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;user</span></code><br><code class="docutils literal notranslate"><span class="pre">How</span> <span class="pre">old</span> <span class="pre">are</span> <span class="pre">you?&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;assistant</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">don't</span> <span class="pre">age</span> <span class="pre">like</span> <span class="pre">humans</span> <span class="pre">do.</span> <span class="pre">I</span> <span class="pre">exist</span> <span class="pre">as</span> <span class="pre">a</span> <span class="pre">piece</span> <span class="pre">of</span> <span class="pre">software,</span> <span class="pre">so</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">have</span> <span class="pre">a</span> <span class="pre">concept</span> <span class="pre">of</span> <span class="pre">age</span> <span class="pre">in</span> <span class="pre">the</span> <span class="pre">traditional</span> <span class="pre">sense.&lt;|im_end|&gt;</span></code><br></p></td>
+<td><p><a class="reference internal" href="supported_conversation_template.html#chatml"><span class="std std-ref">Link</span></a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="docutils literal notranslate"><span class="pre">deepseek</span></code></p></td>
+<td><p><code class="docutils literal notranslate"><span class="pre">&lt;｜begin▁of▁sentence｜&gt;You</span> <span class="pre">are</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.</span></code><br><br><code class="docutils literal notranslate"><span class="pre">User:</span> <span class="pre">Who</span> <span class="pre">are</span> <span class="pre">you?</span></code><br><br><code class="docutils literal notranslate"><span class="pre">Assistant:</span> <span class="pre">I</span> <span class="pre">am</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;｜end▁of▁sentence｜&gt;User:</span> <span class="pre">How</span> <span class="pre">old</span> <span class="pre">are</span> <span class="pre">you?</span></code><br><br><code class="docutils literal notranslate"><span class="pre">Assistant:</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">age</span> <span class="pre">like</span> <span class="pre">humans</span> <span class="pre">do.</span> <span class="pre">I</span> <span class="pre">exist</span> <span class="pre">as</span> <span class="pre">a</span> <span class="pre">piece</span> <span class="pre">of</span> <span class="pre">software,</span> <span class="pre">so</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">have</span> <span class="pre">a</span> <span class="pre">concept</span> <span class="pre">of</span> <span class="pre">age</span> <span class="pre">in</span> <span class="pre">the</span> <span class="pre">traditional</span> <span class="pre">sense.&lt;｜end▁of▁sentence｜&gt;</span></code></p></td>
+<td><p><a class="reference internal" href="supported_conversation_template.html#deepseek"><span class="std std-ref">Link</span></a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="docutils literal notranslate"><span class="pre">gemma</span></code></p></td>
+<td><p><code class="docutils literal notranslate"><span class="pre">&lt;bos&gt;You</span> <span class="pre">are</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;start_of_turn&gt;user</span></code><br><code class="docutils literal notranslate"><span class="pre">Who</span> <span class="pre">are</span> <span class="pre">you?&lt;end_of_turn&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;start_of_turn&gt;model</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">am</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;end_of_turn&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;start_of_turn&gt;user</span></code><br><code class="docutils literal notranslate"><span class="pre">How</span> <span class="pre">old</span> <span class="pre">are</span> <span class="pre">you?&lt;end_of_turn&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;start_of_turn&gt;model</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">don't</span> <span class="pre">age</span> <span class="pre">like</span> <span class="pre">humans</span> <span class="pre">do.</span> <span class="pre">I</span> <span class="pre">exist</span> <span class="pre">as</span> <span class="pre">a</span> <span class="pre">piece</span> <span class="pre">of</span> <span class="pre">software,</span> <span class="pre">so</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">have</span> <span class="pre">a</span> <span class="pre">concept</span> <span class="pre">of</span> <span class="pre">age</span> <span class="pre">in</span> <span class="pre">the</span> <span class="pre">traditional</span> <span class="pre">sense.&lt;end_of_turn&gt;</span></code><br></p></td>
+<td><p><a class="reference internal" href="supported_conversation_template.html#gemma"><span class="std std-ref">Link</span></a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="docutils literal notranslate"><span class="pre">hymba</span></code></p></td>
+<td><p><code class="docutils literal notranslate"><span class="pre">&lt;extra_id_0&gt;System</span></code><br><code class="docutils literal notranslate"><span class="pre">You</span> <span class="pre">are</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;tool&gt;</span> <span class="pre">{&quot;name&quot;:</span> <span class="pre">&quot;generate_qrcode&quot;,</span> <span class="pre">&quot;description&quot;:</span> <span class="pre">&quot;Generate</span> <span class="pre">a</span> <span class="pre">QR</span> <span class="pre">code</span> <span class="pre">for</span> <span class="pre">a</span> <span class="pre">given</span> <span class="pre">text&quot;,</span> <span class="pre">&quot;parameters&quot;:</span> <span class="pre">{&quot;type&quot;:</span> <span class="pre">&quot;object&quot;,</span> <span class="pre">&quot;properties&quot;:</span> <span class="pre">{&quot;text&quot;:</span> <span class="pre">{&quot;type&quot;:</span> <span class="pre">&quot;string&quot;,</span> <span class="pre">&quot;description&quot;:</span> <span class="pre">&quot;The</span> <span class="pre">text</span> <span class="pre">to</span> <span class="pre">encode</span> <span class="pre">in</span> <span class="pre">the</span> <span class="pre">QR</span> <span class="pre">code&quot;}},</span> <span class="pre">&quot;required&quot;:</span> <span class="pre">[&quot;text&quot;]}}</span> <span class="pre">&lt;/tool&gt;</span></code><br><br><code class="docutils literal notranslate"><span class="pre">&lt;extra_id_1&gt;User</span></code><br><code class="docutils literal notranslate"><span class="pre">Who</span> <span class="pre">are</span> <span class="pre">you?</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;extra_id_1&gt;Assistant</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">am</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;extra_id_1&gt;User</span></code><br><code class="docutils literal notranslate"><span class="pre">How</span> <span class="pre">old</span> <span class="pre">are</span> <span class="pre">you?</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;extra_id_1&gt;Assistant</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">don't</span> <span class="pre">age</span> <span class="pre">like</span> <span class="pre">humans</span> <span class="pre">do.</span> <span class="pre">I</span> <span class="pre">exist</span> <span class="pre">as</span> <span class="pre">a</span> <span class="pre">piece</span> <span class="pre">of</span> <span class="pre">software,</span> <span class="pre">so</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">have</span> <span class="pre">a</span> <span class="pre">concept</span> <span class="pre">of</span> <span class="pre">age</span> <span class="pre">in</span> <span class="pre">the</span> <span class="pre">traditional</span> <span class="pre">sense.&lt;/s&gt;</span></code></p></td>
+<td><p><a class="reference internal" href="supported_conversation_template.html#hymba"><span class="std std-ref">Link</span></a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="docutils literal notranslate"><span class="pre">internlm2</span></code></p></td>
+<td><p><code class="docutils literal notranslate"><span class="pre">&lt;s&gt;&lt;|im_start|&gt;system</span></code><br><code class="docutils literal notranslate"><span class="pre">You</span> <span class="pre">are</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;user</span></code><br><code class="docutils literal notranslate"><span class="pre">Who</span> <span class="pre">are</span> <span class="pre">you?&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;assistant</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">am</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;user</span></code><br><code class="docutils literal notranslate"><span class="pre">How</span> <span class="pre">old</span> <span class="pre">are</span> <span class="pre">you?&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;assistant</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">don't</span> <span class="pre">age</span> <span class="pre">like</span> <span class="pre">humans</span> <span class="pre">do.</span> <span class="pre">I</span> <span class="pre">exist</span> <span class="pre">as</span> <span class="pre">a</span> <span class="pre">piece</span> <span class="pre">of</span> <span class="pre">software,</span> <span class="pre">so</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">have</span> <span class="pre">a</span> <span class="pre">concept</span> <span class="pre">of</span> <span class="pre">age</span> <span class="pre">in</span> <span class="pre">the</span> <span class="pre">traditional</span> <span class="pre">sense.&lt;|im_end|&gt;</span></code><br></p></td>
+<td><p><a class="reference internal" href="supported_conversation_template.html#internlm2"><span class="std std-ref">Link</span></a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="docutils literal notranslate"><span class="pre">llama3</span></code></p></td>
+<td><p><code class="docutils literal notranslate"><span class="pre">&lt;|begin_of_text|&gt;&lt;|start_header_id|&gt;system&lt;|end_header_id|&gt;</span></code><br><br><code class="docutils literal notranslate"><span class="pre">You</span> <span class="pre">are</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|eot_id|&gt;&lt;|start_header_id|&gt;user&lt;|end_header_id|&gt;</span></code><br><br><code class="docutils literal notranslate"><span class="pre">Who</span> <span class="pre">are</span> <span class="pre">you?&lt;|eot_id|&gt;&lt;|start_header_id|&gt;assistant&lt;|end_header_id|&gt;</span></code><br><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">am</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|eot_id|&gt;&lt;|start_header_id|&gt;user&lt;|end_header_id|&gt;</span></code><br><br><code class="docutils literal notranslate"><span class="pre">How</span> <span class="pre">old</span> <span class="pre">are</span> <span class="pre">you?&lt;|eot_id|&gt;&lt;|start_header_id|&gt;assistant&lt;|end_header_id|&gt;</span></code><br><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">don't</span> <span class="pre">age</span> <span class="pre">like</span> <span class="pre">humans</span> <span class="pre">do.</span> <span class="pre">I</span> <span class="pre">exist</span> <span class="pre">as</span> <span class="pre">a</span> <span class="pre">piece</span> <span class="pre">of</span> <span class="pre">software,</span> <span class="pre">so</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">have</span> <span class="pre">a</span> <span class="pre">concept</span> <span class="pre">of</span> <span class="pre">age</span> <span class="pre">in</span> <span class="pre">the</span> <span class="pre">traditional</span> <span class="pre">sense.&lt;|eot_id|&gt;</span></code></p></td>
+<td><p><a class="reference internal" href="supported_conversation_template.html#llama-3"><span class="std std-ref">Link</span></a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="docutils literal notranslate"><span class="pre">llama2</span></code></p></td>
+<td><p><code class="docutils literal notranslate"><span class="pre">&lt;s&gt;[INST]</span> <span class="pre">&lt;&lt;SYS&gt;&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">You</span> <span class="pre">are</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;&lt;/SYS&gt;&gt;</span></code><br><br><code class="docutils literal notranslate"><span class="pre">Who</span> <span class="pre">are</span> <span class="pre">you?</span> <span class="pre">[/INST]</span> <span class="pre">I</span> <span class="pre">am</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;/s&gt;&lt;s&gt;[INST]</span> <span class="pre">How</span> <span class="pre">old</span> <span class="pre">are</span> <span class="pre">you?</span> <span class="pre">[/INST]</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">age</span> <span class="pre">like</span> <span class="pre">humans</span> <span class="pre">do.</span> <span class="pre">I</span> <span class="pre">exist</span> <span class="pre">as</span> <span class="pre">a</span> <span class="pre">piece</span> <span class="pre">of</span> <span class="pre">software,</span> <span class="pre">so</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">have</span> <span class="pre">a</span> <span class="pre">concept</span> <span class="pre">of</span> <span class="pre">age</span> <span class="pre">in</span> <span class="pre">the</span> <span class="pre">traditional</span> <span class="pre">sense.&lt;/s&gt;</span></code></p></td>
+<td><p><a class="reference internal" href="supported_conversation_template.html#llama-2"><span class="std std-ref">Link</span></a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="docutils literal notranslate"><span class="pre">phi3</span></code></p></td>
+<td><p><code class="docutils literal notranslate"><span class="pre">&lt;s&gt;&lt;|system|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">You</span> <span class="pre">are</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|user|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">Who</span> <span class="pre">are</span> <span class="pre">you?&lt;|end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|assistant|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">am</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|user|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">How</span> <span class="pre">old</span> <span class="pre">are</span> <span class="pre">you?&lt;|end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|assistant|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">don't</span> <span class="pre">age</span> <span class="pre">like</span> <span class="pre">humans</span> <span class="pre">do.</span> <span class="pre">I</span> <span class="pre">exist</span> <span class="pre">as</span> <span class="pre">a</span> <span class="pre">piece</span> <span class="pre">of</span> <span class="pre">software,</span> <span class="pre">so</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">have</span> <span class="pre">a</span> <span class="pre">concept</span> <span class="pre">of</span> <span class="pre">age</span> <span class="pre">in</span> <span class="pre">the</span> <span class="pre">traditional</span> <span class="pre">sense.&lt;|end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|endoftext|&gt;</span></code></p></td>
+<td><p><a class="reference internal" href="supported_conversation_template.html#phi-3"><span class="std std-ref">Link</span></a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="docutils literal notranslate"><span class="pre">qwen2</span></code></p></td>
+<td><p><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;system</span></code><br><code class="docutils literal notranslate"><span class="pre">You</span> <span class="pre">are</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;user</span></code><br><code class="docutils literal notranslate"><span class="pre">Who</span> <span class="pre">are</span> <span class="pre">you?&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;assistant</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">am</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;user</span></code><br><code class="docutils literal notranslate"><span class="pre">How</span> <span class="pre">old</span> <span class="pre">are</span> <span class="pre">you?&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;assistant</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">don't</span> <span class="pre">age</span> <span class="pre">like</span> <span class="pre">humans</span> <span class="pre">do.</span> <span class="pre">I</span> <span class="pre">exist</span> <span class="pre">as</span> <span class="pre">a</span> <span class="pre">piece</span> <span class="pre">of</span> <span class="pre">software,</span> <span class="pre">so</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">have</span> <span class="pre">a</span> <span class="pre">concept</span> <span class="pre">of</span> <span class="pre">age</span> <span class="pre">in</span> <span class="pre">the</span> <span class="pre">traditional</span> <span class="pre">sense.&lt;|im_end|&gt;</span></code><br></p></td>
+<td><p><a class="reference internal" href="supported_conversation_template.html#qwen-2"><span class="std std-ref">Link</span></a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="docutils literal notranslate"><span class="pre">yi</span></code></p></td>
+<td><p>Same as <code class="docutils literal notranslate"><span class="pre">chatml</span></code></p></td>
+<td><p><a class="reference internal" href="supported_conversation_template.html#yi"><span class="std std-ref">Link</span></a></p></td>
+</tr>
+<tr class="row-odd"><td><p><code class="docutils literal notranslate"><span class="pre">yi1_5</span></code></p></td>
+<td><p><code class="docutils literal notranslate"><span class="pre">You</span> <span class="pre">are</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|im_start|&gt;user</span></code><br><code class="docutils literal notranslate"><span class="pre">Who</span> <span class="pre">are</span> <span class="pre">you?&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;assistant</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">am</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;user</span></code><br><code class="docutils literal notranslate"><span class="pre">How</span> <span class="pre">old</span> <span class="pre">are</span> <span class="pre">you?&lt;|im_end|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|im_start|&gt;assistant</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">don't</span> <span class="pre">age</span> <span class="pre">like</span> <span class="pre">humans</span> <span class="pre">do.</span> <span class="pre">I</span> <span class="pre">exist</span> <span class="pre">as</span> <span class="pre">a</span> <span class="pre">piece</span> <span class="pre">of</span> <span class="pre">software,</span> <span class="pre">so</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">have</span> <span class="pre">a</span> <span class="pre">concept</span> <span class="pre">of</span> <span class="pre">age</span> <span class="pre">in</span> <span class="pre">the</span> <span class="pre">traditional</span> <span class="pre">sense.&lt;|im_end|&gt;</span></code><br></p></td>
+<td><p><a class="reference internal" href="supported_conversation_template.html#yi-15"><span class="std std-ref">Link</span></a></p></td>
+</tr>
+<tr class="row-even"><td><p><code class="docutils literal notranslate"><span class="pre">zephyr</span></code></p></td>
+<td><p><code class="docutils literal notranslate"><span class="pre">&lt;|system|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">You</span> <span class="pre">are</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;/s&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|user|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">Who</span> <span class="pre">are</span> <span class="pre">you?&lt;/s&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|assistant|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">am</span> <span class="pre">a</span> <span class="pre">chatbot</span> <span class="pre">developed</span> <span class="pre">by</span> <span class="pre">LMFlow</span> <span class="pre">team.&lt;/s&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|user|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">How</span> <span class="pre">old</span> <span class="pre">are</span> <span class="pre">you?&lt;/s&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">&lt;|assistant|&gt;</span></code><br><code class="docutils literal notranslate"><span class="pre">I</span> <span class="pre">don't</span> <span class="pre">age</span> <span class="pre">like</span> <span class="pre">humans</span> <span class="pre">do.</span> <span class="pre">I</span> <span class="pre">exist</span> <span class="pre">as</span> <span class="pre">a</span> <span class="pre">piece</span> <span class="pre">of</span> <span class="pre">software,</span> <span class="pre">so</span> <span class="pre">I</span> <span class="pre">don't</span> <span class="pre">have</span> <span class="pre">a</span> <span class="pre">concept</span> <span class="pre">of</span> <span class="pre">age</span> <span class="pre">in</span> <span class="pre">the</span> <span class="pre">traditional</span> <span class="pre">sense.&lt;/s&gt;</span></code><br></p></td>
+<td><p><a class="reference internal" href="supported_conversation_template.html#zephyr"><span class="std std-ref">Link</span></a></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+<p>Passing the template name to the <code class="docutils literal notranslate"><span class="pre">--conversation_template</span></code> argument to apply the corresponding conversation template:</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="c1"># scripts/run_finetune.sh</span>
+<span class="c1"># ...</span>
+deepspeed<span class="w"> </span><span class="si">${</span><span class="nv">deepspeed_args</span><span class="si">}</span><span class="w"> </span><span class="se">\</span>
+<span class="w">  </span>examples/finetune.py<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--model_name_or_path<span class="w"> </span>meta-llama/Llama-2-7b-chat-hf<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--dataset_path<span class="w"> </span><span class="si">${</span><span class="nv">dataset_path</span><span class="si">}</span><span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--conversation_template<span class="w"> </span>llama2<span class="w"> </span><span class="se">\</span>
+<span class="c1"># ...</span>
+</pre></div>
+</div>
+<div class="info admonition">
+<p class="admonition-title">Formatted Dataset</p>
+<p>For dataset that system prompts, tool prompts and templates are already applied (like the one below), user can run the finetune shell by passing <code class="docutils literal notranslate"><span class="pre">empty</span></code> or <code class="docutils literal notranslate"><span class="pre">empty_no_special_tokens</span></code> to the <code class="docutils literal notranslate"><span class="pre">--conversation_template</span></code> argument. <code class="docutils literal notranslate"><span class="pre">empty</span></code> template will add a bos token to the beginning of every round of conversation as well as a eos token to the end of every round of conversation. <code class="docutils literal notranslate"><span class="pre">empty_no_special_tokens</span></code> will not add any special tokens to the conversation, just concatenates the user and assistant messages.</p>
+<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3">
+<summary class="sd-summary-title sd-card-header">
+<span class="sd-summary-text">A formatted dataset</span><span class="sd-summary-state-marker sd-summary-chevron-right"><svg version="1.1" width="1.5em" height="1.5em" class="sd-octicon sd-octicon-chevron-right" viewBox="0 0 24 24" aria-hidden="true"><path d="M8.72 18.78a.75.75 0 0 1 0-1.06L14.44 12 8.72 6.28a.751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018l6.25 6.25a.75.75 0 0 1 0 1.06l-6.25 6.25a.75.75 0 0 1-1.06 0Z"></path></svg></span></summary><div class="sd-summary-content sd-card-body docutils">
+<div class="highlight-json notranslate"><div class="highlight"><pre><span></span><span class="p">{</span>
+<span class="w">  </span><span class="nt">&quot;type&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;conversation&quot;</span><span class="p">,</span>
+<span class="w">  </span><span class="nt">&quot;instances&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">    </span><span class="p">{</span>
+<span class="w">      </span><span class="nt">&quot;messages&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span>
+<span class="w">            </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;[INST] &lt;&lt;SYS&gt;&gt;\nYou are a helpful assistant.\n&lt;&lt;/SYS&gt;&gt;\n\nHello! [/INST]&quot;</span>
+<span class="w">        </span><span class="p">},</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span>
+<span class="w">            </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;Hi, how are you?&quot;</span>
+<span class="w">        </span><span class="p">},</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span>
+<span class="w">            </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;[INST] Good. [/INST]&quot;</span>
+<span class="w">        </span><span class="p">},</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span>
+<span class="w">            </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;Glad to hear that.&quot;</span>
+<span class="w">        </span><span class="p">}</span>
+<span class="w">      </span><span class="p">]</span>
+<span class="w">    </span><span class="p">},</span>
+<span class="w">    </span><span class="p">{</span>
+<span class="w">      </span><span class="nt">&quot;messages&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span>
+<span class="w">            </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;[INST] &lt;&lt;SYS&gt;&gt;\nYou are a helpful assistant.\n&lt;&lt;/SYS&gt;&gt;\n\nWhat&#39;s the weather like now? [/INST]&quot;</span>
+<span class="w">        </span><span class="p">},</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span>
+<span class="w">            </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;I&#39;m sorry for any confusion, but as an AI, I don&#39;t have access to real-time data such as current weather conditions.&quot;</span>
+<span class="w">        </span><span class="p">}</span>
+<span class="w">      </span><span class="p">]</span>
+<span class="w">    </span><span class="p">}</span>
+<span class="w">  </span><span class="p">]</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+</div>
+</details></div>
+</section>
+<section id="customize-conversation-template">
+<h4>Customize Conversation Template<a class="headerlink" href="#customize-conversation-template" title="Link to this heading">#</a></h4>
+<p>Please refer to the <a class="reference internal" href="customize_conversation_template.html"><span class="std std-doc">Customize Conversation Template</span></a> for more details.</p>
+</section>
+</section>
+<section id="textonly">
+<h3>TextOnly<a class="headerlink" href="#textonly" title="Link to this heading">#</a></h3>
+<p>This is the most common dataset type, which only contains raw texts in each
+sample. This type of dataset can be used as the training set for text decoder
+models, or the input of decoder models / encoder-decoder models. Its format is
+as follows (three instances for example),</p>
+<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3">
+<summary class="sd-summary-title sd-card-header">
+<span class="sd-summary-text">A textonly dataset</span><span class="sd-summary-state-marker sd-summary-chevron-right"><svg version="1.1" width="1.5em" height="1.5em" class="sd-octicon sd-octicon-chevron-right" viewBox="0 0 24 24" aria-hidden="true"><path d="M8.72 18.78a.75.75 0 0 1 0-1.06L14.44 12 8.72 6.28a.751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018l6.25 6.25a.75.75 0 0 1 0 1.06l-6.25 6.25a.75.75 0 0 1-1.06 0Z"></path></svg></span></summary><div class="sd-summary-content sd-card-body docutils">
+<div class="highlight-json notranslate"><div class="highlight"><pre><span></span><span class="p">{</span>
+<span class="w">  </span><span class="nt">&quot;type&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;text_only&quot;</span><span class="p">,</span>
+<span class="w">  </span><span class="nt">&quot;instances&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">    </span><span class="p">{</span><span class="w">  </span><span class="nt">&quot;text&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SAMPLE_TEXT_1&quot;</span><span class="w"> </span><span class="p">},</span>
+<span class="w">    </span><span class="p">{</span><span class="w">  </span><span class="nt">&quot;text&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SAMPLE_TEXT_2&quot;</span><span class="w"> </span><span class="p">},</span>
+<span class="w">    </span><span class="p">{</span><span class="w">  </span><span class="nt">&quot;text&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SAMPLE_TEXT_3&quot;</span><span class="w"> </span><span class="p">},</span>
+<span class="w">  </span><span class="p">]</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+</div>
+</details><p>For example, <code class="docutils literal notranslate"><span class="pre">data/example_dataset/train/train_50.json</span></code> has the aboved format.</p>
+</section>
+<section id="text2text">
+<h3>Text2Text<a class="headerlink" href="#text2text" title="Link to this heading">#</a></h3>
+<p>This is the dataset type mostly used for inferencing, which contains a pair of
+texts in each sample. This type of dataset can be used as the training set for
+text encoder-decoder models, or question-answer pair for evaluating model
+inferences. Its format is as follows (three instances for example):</p>
+<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3">
+<summary class="sd-summary-title sd-card-header">
+<span class="sd-summary-text">A text2text dataset</span><span class="sd-summary-state-marker sd-summary-chevron-right"><svg version="1.1" width="1.5em" height="1.5em" class="sd-octicon sd-octicon-chevron-right" viewBox="0 0 24 24" aria-hidden="true"><path d="M8.72 18.78a.75.75 0 0 1 0-1.06L14.44 12 8.72 6.28a.751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018l6.25 6.25a.75.75 0 0 1 0 1.06l-6.25 6.25a.75.75 0 0 1-1.06 0Z"></path></svg></span></summary><div class="sd-summary-content sd-card-body docutils">
+<div class="highlight-json notranslate"><div class="highlight"><pre><span></span><span class="p">{</span>
+<span class="w">  </span><span class="nt">&quot;type&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;text2text&quot;</span><span class="p">,</span>
+<span class="w">  </span><span class="nt">&quot;instances&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">    </span><span class="p">{</span>
+<span class="w">        </span><span class="nt">&quot;input&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SAMPLE_INPUT_1&quot;</span><span class="p">,</span>
+<span class="w">        </span><span class="nt">&quot;output&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SAMPLE_OUTPUT_1&quot;</span><span class="p">,</span>
+<span class="w">    </span><span class="p">},</span>
+<span class="w">    </span><span class="p">{</span>
+<span class="w">        </span><span class="nt">&quot;input&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SAMPLE_INPUT_2&quot;</span><span class="p">,</span>
+<span class="w">        </span><span class="nt">&quot;output&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SAMPLE_OUTPUT_2&quot;</span><span class="p">,</span>
+<span class="w">    </span><span class="p">},</span>
+<span class="w">    </span><span class="p">{</span>
+<span class="w">        </span><span class="nt">&quot;input&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SAMPLE_INPUT_3&quot;</span><span class="p">,</span>
+<span class="w">        </span><span class="nt">&quot;output&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SAMPLE_OUTPUT_3&quot;</span><span class="p">,</span>
+<span class="w">    </span><span class="p">},</span>
+<span class="w">  </span><span class="p">]</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+</div>
+</details><p>For example, <code class="docutils literal notranslate"><span class="pre">data/example_dataset/test/test_13.json</span></code> has the aboved format.</p>
+</section>
+<section id="paired-conversation">
+<h3>Paired Conversation<a class="headerlink" href="#paired-conversation" title="Link to this heading">#</a></h3>
+<div class="info admonition">
+<p class="admonition-title"><strong>Work in Progress</strong></p>
+<p>We are working on paired conversation dataset and will update it soon.</p>
+</div>
+<p>This type of dataset are commonly used for alignment such as <a class="reference external" href="https://arxiv.org/abs/2203.02155">reward modeling</a>,
+<a class="reference external" href="https://arxiv.org/abs/2305.18290">Direct Preference Optimization (DPO)</a>, etc. For requirements of the conversations,
+please refer to <a class="reference internal" href="#conversation"><span class="xref myst">conversation data</span></a>.</p>
+<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3">
+<summary class="sd-summary-title sd-card-header">
+<span class="sd-summary-text">A paired conversation dataset</span><span class="sd-summary-state-marker sd-summary-chevron-right"><svg version="1.1" width="1.5em" height="1.5em" class="sd-octicon sd-octicon-chevron-right" viewBox="0 0 24 24" aria-hidden="true"><path d="M8.72 18.78a.75.75 0 0 1 0-1.06L14.44 12 8.72 6.28a.751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018l6.25 6.25a.75.75 0 0 1 0 1.06l-6.25 6.25a.75.75 0 0 1-1.06 0Z"></path></svg></span></summary><div class="sd-summary-content sd-card-body docutils">
+<div class="highlight-json notranslate"><div class="highlight"><pre><span></span><span class="p">{</span>
+<span class="w">    </span><span class="nt">&quot;type&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;paired_conversation&quot;</span><span class="p">,</span>
+<span class="w">    </span><span class="nt">&quot;instances&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;chosen&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">{</span>
+<span class="w">                </span><span class="nt">&quot;conversation_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;CONVERSATION_ID&quot;</span><span class="p">,</span>
+<span class="w">                </span><span class="nt">&quot;system&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SYSTEM_PROPMT&quot;</span><span class="p">,</span>
+<span class="w">                </span><span class="nt">&quot;tools&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span><span class="s2">&quot;TOOL_DESCRIPTION_1&quot;</span><span class="p">,</span><span class="s2">&quot;TOOL_DESCRIPTION_2&quot;</span><span class="p">,</span><span class="s2">&quot;TOOL_DESCRIPTION_3&quot;</span><span class="p">],</span>
+<span class="w">                </span><span class="nt">&quot;messages&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">                    </span><span class="p">{</span>
+<span class="w">                        </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span>
+<span class="w">                        </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;USER_INPUT_1&quot;</span>
+<span class="w">                    </span><span class="p">},</span>
+<span class="w">                    </span><span class="p">{</span>
+<span class="w">                        </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span>
+<span class="w">                        </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;ASSISTANT_RESPONSE_1_GOOD&quot;</span>
+<span class="w">                    </span><span class="p">},</span>
+<span class="w">                    </span><span class="p">{</span>
+<span class="w">                        </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span>
+<span class="w">                        </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;USER_INPUT_2&quot;</span>
+<span class="w">                    </span><span class="p">},</span>
+<span class="w">                    </span><span class="p">{</span>
+<span class="w">                        </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span>
+<span class="w">                        </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;ASSISTANT_RESPONSE_2_GOOD&quot;</span>
+<span class="w">                    </span><span class="p">}</span>
+<span class="w">                </span><span class="p">]</span>
+<span class="w">            </span><span class="p">},</span>
+<span class="w">            </span><span class="nt">&quot;rejected&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">{</span>
+<span class="w">                </span><span class="nt">&quot;conversation_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;CONVERSATION_ID&quot;</span><span class="p">,</span>
+<span class="w">                </span><span class="nt">&quot;system&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SYSTEM_PROPMT&quot;</span><span class="p">,</span>
+<span class="w">                </span><span class="nt">&quot;tools&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span><span class="s2">&quot;TOOL_DESCRIPTION_1&quot;</span><span class="p">,</span><span class="s2">&quot;TOOL_DESCRIPTION_2&quot;</span><span class="p">,</span><span class="s2">&quot;TOOL_DESCRIPTION_3&quot;</span><span class="p">],</span>
+<span class="w">                </span><span class="nt">&quot;messages&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">                    </span><span class="p">{</span>
+<span class="w">                        </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span>
+<span class="w">                        </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;USER_INPUT_1&quot;</span>
+<span class="w">                    </span><span class="p">},</span>
+<span class="w">                    </span><span class="p">{</span>
+<span class="w">                        </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span>
+<span class="w">                        </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;ASSISTANT_RESPONSE_1_BAD&quot;</span>
+<span class="w">                    </span><span class="p">},</span>
+<span class="w">                    </span><span class="p">{</span>
+<span class="w">                        </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span>
+<span class="w">                        </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;USER_INPUT_2&quot;</span>
+<span class="w">                    </span><span class="p">},</span>
+<span class="w">                    </span><span class="p">{</span>
+<span class="w">                        </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span>
+<span class="w">                        </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;ASSISTANT_RESPONSE_2_BAD&quot;</span>
+<span class="w">                    </span><span class="p">}</span>
+<span class="w">                </span><span class="p">]</span>
+<span class="w">            </span><span class="p">}</span>
+<span class="w">        </span><span class="p">},</span>
+<span class="w">        </span><span class="p">{</span>
+<span class="w">            </span><span class="nt">&quot;chosen&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">{</span>
+<span class="w">                </span><span class="nt">&quot;conversation_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;CONVERSATION_ID&quot;</span><span class="p">,</span>
+<span class="w">                </span><span class="nt">&quot;system&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SYSTEM_PROPMT&quot;</span><span class="p">,</span>
+<span class="w">                </span><span class="nt">&quot;tools&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span><span class="s2">&quot;TOOL_DESCRIPTION_1&quot;</span><span class="p">],</span>
+<span class="w">                </span><span class="nt">&quot;messages&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">                    </span><span class="p">{</span>
+<span class="w">                        </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span>
+<span class="w">                        </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;USER_INPUT_1&quot;</span>
+<span class="w">                    </span><span class="p">},</span>
+<span class="w">                    </span><span class="p">{</span>
+<span class="w">                        </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span>
+<span class="w">                        </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;ASSISTANT_RESPONSE_1_GOOD&quot;</span>
+<span class="w">                    </span><span class="p">}</span>
+<span class="w">                </span><span class="p">]</span>
+<span class="w">            </span><span class="p">},</span>
+<span class="w">            </span><span class="nt">&quot;rejected&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">{</span>
+<span class="w">                </span><span class="nt">&quot;conversation_id&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;CONVERSATION_ID&quot;</span><span class="p">,</span>
+<span class="w">                </span><span class="nt">&quot;system&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;SYSTEM_PROPMT&quot;</span><span class="p">,</span>
+<span class="w">                </span><span class="nt">&quot;tools&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span><span class="s2">&quot;TOOL_DESCRIPTION_1&quot;</span><span class="p">],</span>
+<span class="w">                </span><span class="nt">&quot;messages&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">                    </span><span class="p">{</span>
+<span class="w">                        </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span>
+<span class="w">                        </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;USER_INPUT_1&quot;</span>
+<span class="w">                    </span><span class="p">},</span>
+<span class="w">                    </span><span class="p">{</span>
+<span class="w">                        </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span>
+<span class="w">                        </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;ASSISTANT_RESPONSE_1_BAD&quot;</span>
+<span class="w">                    </span><span class="p">}</span>
+<span class="w">                </span><span class="p">]</span>
+<span class="w">            </span><span class="p">}</span>
+<span class="w">        </span><span class="p">}</span>
+<span class="w">    </span><span class="p">]</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+</div>
+</details></section>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="index.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">Examples</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="checkpoints.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">Checkpoints</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#dataset-format-in-general">Dataset Format in General</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#supported-dataset-and-detailed-formats">Supported Dataset and Detailed Formats</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#conversation">Conversation</a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#data-format">Data Format</a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#conversation-template">Conversation Template</a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#customize-conversation-template">Customize Conversation Template</a></li>
+</ul>
+</li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#textonly">TextOnly</a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#text2text">Text2Text</a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#paired-conversation">Paired Conversation</a></li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/examples/DATASETS.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/examples/TASK_GUIDE.html b/examples/TASK_GUIDE.html
new file mode 100644
index 000000000..a07658a67
--- /dev/null
+++ b/examples/TASK_GUIDE.html
@@ -0,0 +1,639 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>LMFlow Benchmark Guide &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'examples/TASK_GUIDE';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="API Reference" href="../autoapi/index.html" />
+    <link rel="prev" title="RAFT" href="raft.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="DATASETS.html">Dataset</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="checkpoints.html">Checkpoints</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="finetuning.html">Finetuning</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="reward_modeling.html">Reward Modeling</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="raft.html">RAFT</a></li>
+</ul>
+<ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active"><a class="current reference internal" href="#">LMFlow Benchmark Guide</a></li>
+
+
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="index.html" class="nav-link">Examples</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">LMFlow Benchmark Guide</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="lmflow-benchmark-guide">
+<h1>LMFlow Benchmark Guide<a class="headerlink" href="#lmflow-benchmark-guide" title="Link to this heading">#</a></h1>
+<p>We support two ways to add evaluation settings in our repo, <code class="docutils literal notranslate"><span class="pre">NLL</span> <span class="pre">Task</span> <span class="pre">Setting</span></code> and <code class="docutils literal notranslate"><span class="pre">LM-Evaluation</span> <span class="pre">Task</span> <span class="pre">Setting</span></code>. Below are the details of them:</p>
+</section>
+<section id="nll-task-setting">
+<h1>1. NLL Task Setting<a class="headerlink" href="#nll-task-setting" title="Link to this heading">#</a></h1>
+<p>Users can easily create new tasks and evaluate their datasets on
+the provide <code class="docutils literal notranslate"><span class="pre">nll</span> <span class="pre">(Negative</span> <span class="pre">Log</span> <span class="pre">Likelihood)</span></code> metric.</p>
+<section id="setup">
+<h2>Setup<a class="headerlink" href="#setup" title="Link to this heading">#</a></h2>
+<p>Fork the main repo, clone it, and create a new branch with the name of
+your task, and install the following:</p>
+<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="c1"># After forking...</span>
+git<span class="w"> </span>clone<span class="w"> </span>https://github.com/&lt;YOUR-USERNAME&gt;/LMFlow.git
+<span class="nb">cd</span><span class="w"> </span>LMFlow
+git<span class="w"> </span>checkout<span class="w"> </span>-b<span class="w"> </span>&lt;TASK-NAME&gt;
+conda<span class="w"> </span>create<span class="w"> </span>-n<span class="w"> </span>lmflow<span class="w"> </span><span class="nv">python</span><span class="o">=</span><span class="m">3</span>.9<span class="w"> </span>-y
+conda<span class="w"> </span>activate<span class="w"> </span>lmflow
+conda<span class="w"> </span>install<span class="w"> </span>mpi4py
+pip<span class="w"> </span>install<span class="w"> </span>-e<span class="w"> </span>.
+</pre></div>
+</div>
+</section>
+<section id="create-your-task-dataset-file">
+<h2>Create Your Task Dataset File<a class="headerlink" href="#create-your-task-dataset-file" title="Link to this heading">#</a></h2>
+<p>We provide several available datasets under <code class="docutils literal notranslate"><span class="pre">data</span></code> after running</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="nb">cd</span><span class="w"> </span>data<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span>./download.sh<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span><span class="nb">cd</span><span class="w"> </span>-
+</pre></div>
+</div>
+<p>You can refer to some given evaluation dataset files and create your own.
+Also, you may refer to our guide on
+<a class="reference external" href="https://optimalscale.github.io/LMFlow/examples/DATASETS.html">DATASET</a>.</p>
+<p>In this step, you will need to decide your answer type like <code class="docutils literal notranslate"><span class="pre">text2text</span></code>
+or <code class="docutils literal notranslate"><span class="pre">text_only</span></code> (Notice that the current <code class="docutils literal notranslate"><span class="pre">nll</span></code> implementation only supports these
+two answer types). We will note the chosen answer type as <code class="docutils literal notranslate"><span class="pre">&lt;ANSWER_TYPE&gt;</span></code>.</p>
+<p>After preparing your own <code class="docutils literal notranslate"><span class="pre">DATASET</span></code> file, you can put it under <code class="docutils literal notranslate"><span class="pre">data</span></code> dir
+and make a <code class="docutils literal notranslate"><span class="pre">TASK</span></code> dir.</p>
+<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>mkdir<span class="w"> </span>&lt;TASK&gt;
+mv<span class="w"> </span>&lt;DATASET&gt;<span class="w"> </span>&lt;TASK&gt;
+</pre></div>
+</div>
+</section>
+<section id="task-registration">
+<h2>Task Registration<a class="headerlink" href="#task-registration" title="Link to this heading">#</a></h2>
+<p>Note the path of your dataset, <code class="docutils literal notranslate"><span class="pre">data/&lt;TASK&gt;/&lt;DATASET&gt;</span></code>.</p>
+<p>Open the file <code class="docutils literal notranslate"><span class="pre">examples/benchmarking.py</span></code>, add your task’s info into
+<code class="docutils literal notranslate"><span class="pre">LOCAL_DATSET_GROUP_MAP</span></code>, <code class="docutils literal notranslate"><span class="pre">LOCAL_DATSET_MAP</span></code>, <code class="docutils literal notranslate"><span class="pre">LOCAL_DATSET_ANSWERTYPE_MAP</span></code></p>
+<p>In <code class="docutils literal notranslate"><span class="pre">LOCAL_DATSET_MAP</span></code>, you will need to specify your <code class="docutils literal notranslate"><span class="pre">DATASET</span></code> files’ path:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">LOCAL_DATSET_MAP</span> <span class="o">=</span><span class="p">{</span>
+    <span class="s2">&quot;...&quot;</span><span class="p">:</span><span class="s2">&quot;...&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;&lt;TASK&gt;&quot;</span><span class="p">:</span><span class="s2">&quot;data/&lt;TASK&gt;/&lt;DATASET&gt;&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+<p>In <code class="docutils literal notranslate"><span class="pre">LOCAL_DATSET_ANSWERTYPE_MAP</span></code>, you will need to specify your task’s
+<code class="docutils literal notranslate"><span class="pre">&lt;ANSWER_TYPE&gt;</span></code>:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">LOCAL_DATSET_ANSWERTYPE_MAP</span> <span class="o">=</span><span class="p">{</span>
+    <span class="s2">&quot;...&quot;</span><span class="p">:</span><span class="s2">&quot;...&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;&lt;TASK&gt;&quot;</span><span class="p">:</span><span class="s2">&quot;&lt;ANSWER_TYPE&gt;,</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+<p>If you only have one task, you can add key-value pair like <code class="docutils literal notranslate"><span class="pre">&quot;&lt;TASK&gt;&quot;:&quot;&lt;TASK&gt;&quot;</span></code>
+in <code class="docutils literal notranslate"><span class="pre">LOCAL_DATSET_GROUP_MAP</span></code>:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">LOCAL_DATSET_GROUP_MAP</span> <span class="o">=</span><span class="p">{</span>
+    <span class="s2">&quot;...&quot;</span><span class="p">:</span><span class="s2">&quot;...&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;&lt;TASK&gt;&quot;</span><span class="p">:</span><span class="s2">&quot;&lt;TASK&gt;&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+<p>If you want to combine several tasks, you may first specify a
+combination name <code class="docutils literal notranslate"><span class="pre">&lt;TASK_COMBINATION&gt;</span></code> and add key-value pair like
+<code class="docutils literal notranslate"><span class="pre">&quot;&lt;TASK_COMBINATION&gt;&quot;:&quot;&lt;TASK_1&gt;,&lt;TASK_2&gt;,..&quot;</span></code>in <code class="docutils literal notranslate"><span class="pre">LOCAL_DATSET_GROUP_MAP</span></code>.</p>
+<p>Remember to separate TASK by <code class="docutils literal notranslate"><span class="pre">,</span></code>:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">LOCAL_DATSET_GROUP_MAP</span> <span class="o">=</span><span class="p">{</span>
+    <span class="s2">&quot;...&quot;</span><span class="p">:</span><span class="s2">&quot;...&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;&lt;TASK_COMBINATION&gt;&quot;</span><span class="p">:</span><span class="s2">&quot;&lt;TASK_1&gt;,&lt;TASK_2&gt;,..&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+<p>After finishing changing these items, you can run your own <code class="docutils literal notranslate"><span class="pre">&lt;TASK&gt;</span></code> like:</p>
+<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>deepspeed<span class="w"> </span>examples/benchmarking.py<span class="w"> </span><span class="se">\</span>
+<span class="w">  </span>--answer_type<span class="w"> </span>&lt;ANSWER_TYPE&gt;<span class="w"> </span><span class="se">\</span>
+<span class="w">  </span>--use_ram_optimized_load<span class="w"> </span>False<span class="w"> </span><span class="se">\</span>
+<span class="w">  </span>--model_name_or_path<span class="w"> </span><span class="si">${</span><span class="nv">model_name</span><span class="si">}</span><span class="w"> </span><span class="se">\</span>
+<span class="w">  </span>--dataset_name<span class="w"> </span>data/&lt;TASK&gt;/&lt;DATASET&gt;<span class="se">\</span>
+<span class="w">  </span>--deepspeed<span class="w"> </span>examples/ds_config.json<span class="w"> </span><span class="se">\</span>
+<span class="w">  </span>--metric<span class="w"> </span>nll<span class="w"> </span><span class="se">\</span>
+<span class="w">  </span>--prompt_structure<span class="w"> </span><span class="s2">&quot;###Human: {input}###Assistant:&quot;</span><span class="w"> </span><span class="se">\</span>
+<span class="w">  </span><span class="p">|</span><span class="w"> </span>tee<span class="w"> </span><span class="si">${</span><span class="nv">log_dir</span><span class="si">}</span>/train.log<span class="w"> </span><span class="se">\</span>
+<span class="w">  </span><span class="m">2</span>&gt;<span class="w"> </span><span class="si">${</span><span class="nv">log_dir</span><span class="si">}</span>/train.err<span class="w"> </span>
+</pre></div>
+</div>
+</section>
+</section>
+<section id="lm-evaluation-task-setting">
+<h1>2. LM-Evaluation Task Setting<a class="headerlink" href="#lm-evaluation-task-setting" title="Link to this heading">#</a></h1>
+<p>We integrate <a class="reference external" href="https://github.com/EleutherAI/lm-evaluation-harness">EleutherAI/lm-evaluation-harness</a> into
+<code class="docutils literal notranslate"><span class="pre">benchamrk.py</span></code> by directly executing the evaluate commands. Users
+can also use their evaluation by simply changing two items in
+<code class="docutils literal notranslate"><span class="pre">&lt;LM_EVAL_DATASET_MAP&gt;</span></code> of <code class="docutils literal notranslate"><span class="pre">examples/benchmarking.py</span></code>.</p>
+<p>Please refer to Eleuther’s
+<a class="reference external" href="https://github.com/EleutherAI/lm-evaluation-harness/blob/master/docs/task_table.md">task-table</a>
+to get exact <code class="docutils literal notranslate"><span class="pre">&lt;TASK&gt;</span></code> name.</p>
+<p>Similarly, you can combine several tasks, you may first specify a
+combination name <code class="docutils literal notranslate"><span class="pre">&lt;TASK_COMBINATION&gt;</span></code> and add key-value pair like
+<code class="docutils literal notranslate"><span class="pre">&quot;&lt;TASK_COMBINATION&gt;&quot;:&quot;&lt;TASK_1&gt;,&lt;TASK_2&gt;,..&quot;</span></code>in <code class="docutils literal notranslate"><span class="pre">LM_EVAL_DATASET_MAP</span></code>.</p>
+<p>Also, remember to separate TASK by <code class="docutils literal notranslate"><span class="pre">,</span></code>:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">LM_EVAL_DATASET_MAP</span> <span class="o">=</span><span class="p">{</span>
+    <span class="s2">&quot;...&quot;</span><span class="p">:</span><span class="s2">&quot;...&quot;</span><span class="p">,</span>
+    <span class="s2">&quot;&lt;TASK_COMBINATION&gt;&quot;</span><span class="p">:</span><span class="s2">&quot;&lt;TASK_1&gt;,&lt;TASK_2&gt;,..&quot;</span><span class="p">,</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="raft.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">RAFT</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="../autoapi/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">API Reference</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h1 nav-item toc-entry"><a class="reference internal nav-link" href="#">LMFlow Benchmark Guide</a></li>
+<li class="toc-h1 nav-item toc-entry"><a class="reference internal nav-link" href="#nll-task-setting">1. NLL Task Setting</a><ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#setup">Setup</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#create-your-task-dataset-file">Create Your Task Dataset File</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#task-registration">Task Registration</a></li>
+</ul>
+</li>
+<li class="toc-h1 nav-item toc-entry"><a class="reference internal nav-link" href="#lm-evaluation-task-setting">2. LM-Evaluation Task Setting</a></li>
+</ul>
+
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/examples/TASK_GUIDE.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/examples/checkpoints.html b/examples/checkpoints.html
new file mode 100644
index 000000000..76ab75984
--- /dev/null
+++ b/examples/checkpoints.html
@@ -0,0 +1,546 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>Checkpoints &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'examples/checkpoints';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Finetuning" href="finetuning.html" />
+    <link rel="prev" title="Dataset" href="DATASETS.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="DATASETS.html">Dataset</a></li>
+</ul>
+<ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active"><a class="current reference internal" href="#">Checkpoints</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="finetuning.html">Finetuning</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="reward_modeling.html">Reward Modeling</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="raft.html">RAFT</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="TASK_GUIDE.html">LMFlow Benchmark Guide</a></li>
+
+
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="index.html" class="nav-link">Examples</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Checkpoints</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="checkpoints">
+<h1>Checkpoints<a class="headerlink" href="#checkpoints" title="Link to this heading">#</a></h1>
+<p>In general, you can directly load from checkpoints by using <code class="docutils literal notranslate"><span class="pre">--model_name_or_path</span></code>. However, the LLaMA case is slightly different due to the copyright issue.</p>
+<section id="llama-checkpoint">
+<h2>LLaMA Checkpoint<a class="headerlink" href="#llama-checkpoint" title="Link to this heading">#</a></h2>
+<ol class="arabic">
+<li><p>First, you need to get the access of LLaMA model from <a class="reference external" href="https://github.com/facebookresearch/llama">facebookresearch/llama</a>. Download the official checkpoints and save them into <code class="docutils literal notranslate"><span class="pre">${llama-path}</span></code>.</p></li>
+<li><p>Second, convert the official checkpoints <code class="docutils literal notranslate"><span class="pre">${llama-path}</span></code> to HuggingFace supported checkpoints <code class="docutils literal notranslate"><span class="pre">${llama-hf-path}</span></code> by running</p>
+<p><code class="docutils literal notranslate"><span class="pre">python</span> <span class="pre">./scripts/convert_llama_weights_to_hf.py</span> <span class="pre">--input_dir</span> <span class="pre">${llama-path}</span> <span class="pre">--model_size</span> <span class="pre">7B</span> <span class="pre">--output_dir</span> <span class="pre">${llama-hf-path}/llama-7b-hf</span></code></p>
+</li>
+<li><p>Then you are good to go by setting the checkpoint path to <code class="docutils literal notranslate"><span class="pre">${llama-hf-path}/llama-7b-hf</span></code>. Enjoy it!</p></li>
+<li><p>(optional) Now you have the original llama-7b-hf pretrained model. With</p></li>
+</ol>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="nb">cd</span><span class="w"> </span>output_models<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span>./download.sh<span class="w"> </span>all<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span><span class="nb">cd</span><span class="w"> </span>-
+</pre></div>
+</div>
+<p>You can obtain the model difference finetuned by ours. By a way similar to <code class="docutils literal notranslate"><span class="pre">./scripts/run_evaluation_with_lora.sh</span></code>,</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="nv">CUDA_VISIBLE_DEVICES</span><span class="o">=</span><span class="m">0</span><span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>deepspeed<span class="w"> </span>examples/evaluate.py<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--answer_type<span class="w"> </span>text<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--model_name_or_path<span class="w"> </span><span class="si">${</span><span class="nv">llama</span><span class="p">-hf-path</span><span class="si">}</span>/llama-7b-hf<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--lora_model_path<span class="w"> </span>output_models/<span class="si">${</span><span class="nv">llama</span><span class="p">-model-diff-path</span><span class="si">}</span><span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--dataset_path<span class="w"> </span>data/alpaca/test<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--prompt_structure<span class="w"> </span><span class="s2">&quot;Input: {input}&quot;</span><span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--deepspeed<span class="w"> </span>examples/ds_config.json
+</pre></div>
+</div>
+<p>You can now evaluate with the finetuned llama model.</p>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="DATASETS.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">Dataset</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="finetuning.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">Finetuning</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#llama-checkpoint">LLaMA Checkpoint</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/examples/checkpoints.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/examples/customize_conversation_template.html b/examples/customize_conversation_template.html
new file mode 100644
index 000000000..43a04814a
--- /dev/null
+++ b/examples/customize_conversation_template.html
@@ -0,0 +1,592 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>Customize Conversation Template &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'examples/customize_conversation_template';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Customize Conversation Template</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="customize-conversation-template">
+<h1>Customize Conversation Template<a class="headerlink" href="#customize-conversation-template" title="Link to this heading">#</a></h1>
+<blockquote>
+<div><p>For beginners: Why template?<br />
+Almost all LLMs today do a simple job - predict the next “word”. To make the interaction between user and model smoother, developers use tricks: they add special “words” to the input text (at back-end, thus invisible to the user when using services like ChatGPT) to “tell” the model what user had said before, and ask the model to respond like an assistant. These “hidden words” are called “template”.</p>
+</div></blockquote>
+<p>We provide the flexibility to customize the conversation template. You can customize your own conversation template by following the steps below:</p>
+<section id="decompose-your-conversations">
+<h2>1. Decompose your conversations<a class="headerlink" href="#decompose-your-conversations" title="Link to this heading">#</a></h2>
+<p>Say you want to make the conversations between user and assistant look like:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;bos&gt;System:
+You are a chatbot developed by LMFlow team.
+
+User:
+Who are you?
+
+Assistant:
+I am a chatbot developed by LMFlow team.&lt;eos&gt;
+
+User:
+How old are you?
+
+Assistant:
+I don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.&lt;eos&gt;
+</pre></div>
+</div>
+<p>It is easy to abstract the format for each message:</p>
+<ul class="simple">
+<li><p>System message: <code class="docutils literal notranslate"><span class="pre">System:\n{{content}}\n\n</span></code></p></li>
+<li><p>User message: <code class="docutils literal notranslate"><span class="pre">User:\n{{content}}\n\n</span></code></p></li>
+<li><p>Assistant message: <code class="docutils literal notranslate"><span class="pre">Assistant:\n{{content}}\n\n&lt;eos&gt;</span></code></p></li>
+</ul>
+<p>Also, we have a bos token at the beginning of the conversation session.</p>
+</section>
+<section id="choose-proper-formatter">
+<h2>2. Choose proper <code class="docutils literal notranslate"><span class="pre">Formatter</span></code><a class="headerlink" href="#choose-proper-formatter" title="Link to this heading">#</a></h2>
+<p>Recall the requirements for a conversation dataset:</p>
+<blockquote>
+<div><ul class="simple">
+<li><p><code class="docutils literal notranslate"><span class="pre">system</span></code>: <code class="docutils literal notranslate"><span class="pre">Optional[string]</span></code>.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">tools</span></code>: <code class="docutils literal notranslate"><span class="pre">Optional[List[string]]</span></code>.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">messages</span></code>: <code class="docutils literal notranslate"><span class="pre">List[Dict]</span></code>.</p>
+<ul>
+<li><p><code class="docutils literal notranslate"><span class="pre">role</span></code>: <code class="docutils literal notranslate"><span class="pre">string</span></code>.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">content</span></code>: <code class="docutils literal notranslate"><span class="pre">string</span></code>.</p></li>
+</ul>
+</li>
+</ul>
+</div></blockquote>
+<p>System message, user message, and assistant message are strings thus we can use <code class="docutils literal notranslate"><span class="pre">StringFormatter</span></code> for them.</p>
+</section>
+<section id="build-the-template">
+<h2>3. Build the template<a class="headerlink" href="#build-the-template" title="Link to this heading">#</a></h2>
+<p>All preset templates are located at <code class="docutils literal notranslate"><span class="pre">src/lmflow/utils/conversation_template</span></code>.</p>
+<p>Within the template file, define your own template like:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span><span class="w"> </span><span class="nn">.base</span><span class="w"> </span><span class="kn">import</span> <span class="n">StringFormatter</span><span class="p">,</span> <span class="n">TemplateComponent</span><span class="p">,</span> <span class="n">ConversationTemplate</span>
+
+
+<span class="n">YOUR_TEMPLATE</span> <span class="o">=</span> <span class="n">ConversationTemplate</span><span class="p">(</span>
+    <span class="n">template_name</span><span class="o">=</span><span class="s1">&#39;your_template_name&#39;</span><span class="p">,</span>
+    <span class="n">user_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;User:</span><span class="se">\n</span><span class="s1">{{content}}</span><span class="se">\n\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">assistant_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;Assistant:</span><span class="se">\n</span><span class="s1">{{content}}</span><span class="se">\n\n</span><span class="s1">&#39;</span><span class="p">),</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;eos_token&#39;</span><span class="p">)</span> <span class="c1"># this will add the eos token at the end of every assistant message</span>
+            <span class="c1"># please refer to the docstring of the `TemplateComponent` class to </span>
+            <span class="c1"># see the difference between different types of components.</span>
+        <span class="p">]</span>
+    <span class="p">),</span>
+    <span class="n">system_formatter</span><span class="o">=</span><span class="n">StringFormatter</span><span class="p">(</span>
+        <span class="n">template</span><span class="o">=</span><span class="p">[</span>
+            <span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;string&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;System:</span><span class="se">\n</span><span class="s1">{{content}}</span><span class="se">\n\n</span><span class="s1">&#39;</span><span class="p">)</span>
+        <span class="p">]</span>
+    <span class="p">)</span>
+    <span class="c1"># For models that has ONLY ONE bos token at the beginning of </span>
+    <span class="c1"># a conversation session (not a conversation pair), user can</span>
+    <span class="c1"># specify a special starter to add that starter to the very</span>
+    <span class="c1"># beginning of the conversation session. </span>
+    <span class="c1"># eg:</span>
+    <span class="c1">#   llama-2: &lt;s&gt; and &lt;/s&gt; at every pair of conversation </span>
+    <span class="c1">#   v.s.</span>
+    <span class="c1">#   llama-3: &lt;|begin_of_text|&gt; only at the beginning of a session</span>
+    <span class="n">special_starter</span><span class="o">=</span><span class="n">TemplateComponent</span><span class="p">(</span><span class="nb">type</span><span class="o">=</span><span class="s1">&#39;token&#39;</span><span class="p">,</span> <span class="n">content</span><span class="o">=</span><span class="s1">&#39;bos_token&#39;</span><span class="p">),</span>
+
+    <span class="c1"># Similar to the special starter... (just for illustration, commented out </span>
+    <span class="c1"># since it is not necessary for our purposed template above)</span>
+    <span class="c1"># special_stopper=TemplateComponent(type=&#39;token&#39;, content=&#39;eos_token&#39;)</span>
+<span class="p">)</span>
+</pre></div>
+</div>
+<p>Feel free to create your own template by inheriting the <code class="docutils literal notranslate"><span class="pre">ConversationTemplate</span></code> class. Llama-2 v.s. llama-3 would be a good examples to refer to.</p>
+</section>
+<section id="register-your-template">
+<h2>4. Register your template<a class="headerlink" href="#register-your-template" title="Link to this heading">#</a></h2>
+<p>After defining your own template, you need to register it in the <code class="docutils literal notranslate"><span class="pre">src/lmflow/utils/conversation_template/__init__.py</span></code> file.</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="c1"># ...</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.your_template_file</span><span class="w"> </span><span class="kn">import</span> <span class="n">YOUR_TEMPLATE</span>
+
+
+<span class="n">PRESET_TEMPLATES</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="c1">#...</span>
+    <span class="s1">&#39;your_template_name&#39;</span><span class="p">:</span> <span class="n">YOUR_TEMPLATE</span><span class="p">,</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+</section>
+<section id="use-your-template">
+<h2>5. Use your template<a class="headerlink" href="#use-your-template" title="Link to this heading">#</a></h2>
+<p>You are all set! Specify the template name in, for example, your finetune script:</p>
+<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>./scripts/run_finetune.sh<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--model_name_or_path<span class="w"> </span>path_to_your_model<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--dataset_path<span class="w"> </span>your_conversation_dataset<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--conversation_template<span class="w"> </span>your_template_name<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--output_model_path<span class="w"> </span>output_models/your_model
+</pre></div>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#decompose-your-conversations">1. Decompose your conversations</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#choose-proper-formatter">2. Choose proper <code class="docutils literal notranslate"><span class="pre">Formatter</span></code></a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#build-the-template">3. Build the template</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#register-your-template">4. Register your template</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#use-your-template">5. Use your template</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/examples/customize_conversation_template.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/examples/finetuning.html b/examples/finetuning.html
new file mode 100644
index 000000000..cac94732d
--- /dev/null
+++ b/examples/finetuning.html
@@ -0,0 +1,615 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>Finetuning &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'examples/finetuning';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Reward Modeling" href="reward_modeling.html" />
+    <link rel="prev" title="Checkpoints" href="checkpoints.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="DATASETS.html">Dataset</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="checkpoints.html">Checkpoints</a></li>
+</ul>
+<ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active"><a class="current reference internal" href="#">Finetuning</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="reward_modeling.html">Reward Modeling</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="raft.html">RAFT</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="TASK_GUIDE.html">LMFlow Benchmark Guide</a></li>
+
+
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="index.html" class="nav-link">Examples</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Finetuning</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="finetuning">
+<h1>Finetuning<a class="headerlink" href="#finetuning" title="Link to this heading">#</a></h1>
+<section id="full-parameters">
+<h2>Full Parameters<a class="headerlink" href="#full-parameters" title="Link to this heading">#</a></h2>
+<p>Full training updates all the parameters to finetune a language model.
+Here is an example to finetune a GPT-2 base model.</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="nb">cd</span><span class="w"> </span>data<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span>./download.sh<span class="w"> </span>alpaca<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span><span class="nb">cd</span><span class="w"> </span>-
+
+./scripts/run_finetune.sh<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--model_name_or_path<span class="w"> </span>gpt2<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--dataset_path<span class="w"> </span>data/alpaca/train_conversation<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--output_model_path<span class="w"> </span>output_models/finetuned_gpt2
+</pre></div>
+</div>
+<div class="tip admonition">
+<p class="admonition-title">Conversation Template</p>
+<p>For conversation dataset, specify a conversation template for better performance by adding <code class="docutils literal notranslate"><span class="pre">--conversation_template</span></code> to the command.</p>
+</div>
+<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3">
+<summary class="sd-summary-title sd-card-header">
+<span class="sd-summary-text">Llama-3-8B conversation dataset example</span><span class="sd-summary-state-marker sd-summary-chevron-right"><svg version="1.1" width="1.5em" height="1.5em" class="sd-octicon sd-octicon-chevron-right" viewBox="0 0 24 24" aria-hidden="true"><path d="M8.72 18.78a.75.75 0 0 1 0-1.06L14.44 12 8.72 6.28a.751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018l6.25 6.25a.75.75 0 0 1 0 1.06l-6.25 6.25a.75.75 0 0 1-1.06 0Z"></path></svg></span></summary><div class="sd-summary-content sd-card-body docutils">
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="nb">cd</span><span class="w"> </span>data<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span>./download.sh<span class="w"> </span>alpaca<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span><span class="nb">cd</span><span class="w"> </span>-
+
+./scripts/run_finetune.sh<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--model_name_or_path<span class="w"> </span>meta-llama/Meta-Llama-3-8B<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--dataset_path<span class="w"> </span>data/alpaca/train_conversation<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--conversation_template<span class="w"> </span>llama3<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--output_model_path<span class="w"> </span>output_models/finetuned_llama3_8b
+</pre></div>
+</div>
+</div>
+</details></section>
+<section id="layerwise-importance-sampled-adamw-lisa">
+<h2>Layerwise Importance Sampled AdamW (LISA)<a class="headerlink" href="#layerwise-importance-sampled-adamw-lisa" title="Link to this heading">#</a></h2>
+<p><a class="reference external" href="https://arxiv.org/abs/2403.17919">LISA</a> is a memory-efficient finetuning algorithm that allows tradeoff between memory and the number of randomly unfreezed layers. This script currently is only tested in single gpus. Please stay tuned for our latest updates!</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="nb">cd</span><span class="w"> </span>data<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span>./download.sh<span class="w"> </span>alpaca<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span><span class="nb">cd</span><span class="w"> </span>-
+
+./scripts/run_finetune_with_lisa.sh<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--model_name_or_path<span class="w"> </span>meta-llama/Llama-2-7b-hf<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--dataset_path<span class="w"> </span>data/alpaca/train_conversation<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--output_model_path<span class="w"> </span>output_models/finetuned_llama2_7b<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--lisa_activated_layers<span class="w"> </span><span class="m">1</span><span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--lisa_interval_steps<span class="w"> </span><span class="m">20</span>
+</pre></div>
+</div>
+<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3">
+<summary class="sd-summary-title sd-card-header">
+<span class="sd-summary-text">Llama-2-7B conversation dataset example</span><span class="sd-summary-state-marker sd-summary-chevron-right"><svg version="1.1" width="1.5em" height="1.5em" class="sd-octicon sd-octicon-chevron-right" viewBox="0 0 24 24" aria-hidden="true"><path d="M8.72 18.78a.75.75 0 0 1 0-1.06L14.44 12 8.72 6.28a.751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018l6.25 6.25a.75.75 0 0 1 0 1.06l-6.25 6.25a.75.75 0 0 1-1.06 0Z"></path></svg></span></summary><div class="sd-summary-content sd-card-body docutils">
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="nb">cd</span><span class="w"> </span>data<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span>./download.sh<span class="w"> </span>alpaca<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span><span class="nb">cd</span><span class="w"> </span>-
+
+./scripts/run_finetune_with_lisa.sh<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--model_name_or_path<span class="w"> </span>meta-llama/Llama-2-7b-hf<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--dataset_path<span class="w"> </span>data/alpaca/train_conversation<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--conversation_template<span class="w"> </span>llama2<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--output_model_path<span class="w"> </span>output_models/finetuned_llama2_7b_lisa<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--lisa_activated_layers<span class="w"> </span><span class="m">1</span><span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--lisa_interval_steps<span class="w"> </span><span class="m">20</span>
+</pre></div>
+</div>
+</div>
+</details></section>
+<section id="low-rank-adaptation-lora">
+<h2>Low-Rank Adaptation (LoRA)<a class="headerlink" href="#low-rank-adaptation-lora" title="Link to this heading">#</a></h2>
+<p>LoRA is a parameter-efficient finetuning algorithm and is more efficient than full finetuning.</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="nb">cd</span><span class="w"> </span>data<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span>./download.sh<span class="w"> </span>alpaca<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span><span class="nb">cd</span><span class="w"> </span>-
+
+./scripts/run_finetune_with_lora.sh<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--model_name_or_path<span class="w"> </span>facebook/galactica-1.3b<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--dataset_path<span class="w"> </span>data/alpaca/train_conversation<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--output_lora_path<span class="w"> </span>output_models/finetuned_galactica_lora
+</pre></div>
+</div>
+<div class="tip admonition">
+<p class="admonition-title">Merge LoRA Weight</p>
+<p>Merge LoRA weight and the base model into one using:</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span>./scripts/run_merge_lora.sh<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--model_name_or_path<span class="w"> </span>Qwen/Qwen1.5-1.8B<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--lora_model_path<span class="w"> </span>output_models/lora<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--output_model_path<span class="w"> </span>output_models/lora_merged<span class="w"> </span><span class="se">\</span>
+</pre></div>
+</div>
+</div>
+<details class="sd-sphinx-override sd-dropdown sd-card sd-mb-3">
+<summary class="sd-summary-title sd-card-header">
+<span class="sd-summary-text">Llama-2-7B conversation dataset example</span><span class="sd-summary-state-marker sd-summary-chevron-right"><svg version="1.1" width="1.5em" height="1.5em" class="sd-octicon sd-octicon-chevron-right" viewBox="0 0 24 24" aria-hidden="true"><path d="M8.72 18.78a.75.75 0 0 1 0-1.06L14.44 12 8.72 6.28a.751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018l6.25 6.25a.75.75 0 0 1 0 1.06l-6.25 6.25a.75.75 0 0 1-1.06 0Z"></path></svg></span></summary><div class="sd-summary-content sd-card-body docutils">
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="nb">cd</span><span class="w"> </span>data<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span>./download.sh<span class="w"> </span>alpaca<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span><span class="nb">cd</span><span class="w"> </span>-
+
+./scripts/run_finetune_with_lora.sh<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--model_name_or_path<span class="w"> </span>meta-llama/Llama-2-7b-hf<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--dataset_path<span class="w"> </span>data/alpaca/train_conversation<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--conversation_template<span class="w"> </span>llama2<span class="w"> </span><span class="se">\</span>
+<span class="w">    </span>--output_model_path<span class="w"> </span>output_models/finetuned_llama2_7b_lora<span class="w"> </span><span class="se">\</span>
+</pre></div>
+</div>
+</div>
+</details></section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="checkpoints.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">Checkpoints</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="reward_modeling.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">Reward Modeling</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#full-parameters">Full Parameters</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#layerwise-importance-sampled-adamw-lisa">Layerwise Importance Sampled AdamW (LISA)</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#low-rank-adaptation-lora">Low-Rank Adaptation (LoRA)</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/examples/finetuning.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/examples/index.html b/examples/index.html
new file mode 100644
index 000000000..18886c39b
--- /dev/null
+++ b/examples/index.html
@@ -0,0 +1,618 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>Examples &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'examples/index';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="Dataset" href="DATASETS.html" />
+    <link rel="prev" title="LMFlow Benchmark: An Automatic Evaluation Framework for Open-Source LLMs" href="../blogs/benchmark.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="#">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="#">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="DATASETS.html">Dataset</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="checkpoints.html">Checkpoints</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="finetuning.html">Finetuning</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="reward_modeling.html">Reward Modeling</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="raft.html">RAFT</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="TASK_GUIDE.html">LMFlow Benchmark Guide</a></li>
+
+
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Examples</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="examples">
+<h1>Examples<a class="headerlink" href="#examples" title="Link to this heading">#</a></h1>
+<p>We provide several examples to show how to use our package in your problem.</p>
+<section id="data-preparation">
+<h2>Data preparation<a class="headerlink" href="#data-preparation" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="DATASETS.html">Dataset</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="DATASETS.html#dataset-format-in-general">Dataset Format in General</a></li>
+<li class="toctree-l2"><a class="reference internal" href="DATASETS.html#supported-dataset-and-detailed-formats">Supported Dataset and Detailed Formats</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="DATASETS.html#conversation">Conversation</a></li>
+<li class="toctree-l3"><a class="reference internal" href="DATASETS.html#textonly">TextOnly</a></li>
+<li class="toctree-l3"><a class="reference internal" href="DATASETS.html#text2text">Text2Text</a></li>
+<li class="toctree-l3"><a class="reference internal" href="DATASETS.html#paired-conversation">Paired Conversation</a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+</div>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="checkpoints.html">Checkpoints</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="checkpoints.html#llama-checkpoint">LLaMA Checkpoint</a></li>
+</ul>
+</li>
+</ul>
+</div>
+</section>
+<section id="finetuning">
+<h2>Finetuning<a class="headerlink" href="#finetuning" title="Link to this heading">#</a></h2>
+<p>For SFT,</p>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="finetuning.html">Finetuning</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="finetuning.html#full-parameters">Full Parameters</a></li>
+<li class="toctree-l2"><a class="reference internal" href="finetuning.html#layerwise-importance-sampled-adamw-lisa">Layerwise Importance Sampled AdamW (LISA)</a></li>
+<li class="toctree-l2"><a class="reference internal" href="finetuning.html#low-rank-adaptation-lora">Low-Rank Adaptation (LoRA)</a></li>
+</ul>
+</li>
+</ul>
+</div>
+<p>For alignment process,</p>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="reward_modeling.html">Reward Modeling</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="reward_modeling.html#introduction">Introduction</a></li>
+<li class="toctree-l2"><a class="reference internal" href="reward_modeling.html#step-1-supervised-finetuning-sft">Step 1 Supervised Finetuning (SFT)</a></li>
+<li class="toctree-l2"><a class="reference internal" href="reward_modeling.html#step-2-reward-modeling">Step 2 Reward Modeling</a></li>
+<li class="toctree-l2"><a class="reference internal" href="reward_modeling.html#examples">Examples</a></li>
+</ul>
+</li>
+</ul>
+</div>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="raft.html">RAFT</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="raft.html#introduction">1 Introduction</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="raft.html#dataset-description">1.1 Dataset description</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="raft.html#reward-modeling">2 Reward Modeling</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="raft.html#supervised-finetuning-sft">2.1 Supervised Finetuning (SFT)</a></li>
+<li class="toctree-l3"><a class="reference internal" href="raft.html#id1">2.2 Reward Modeling</a></li>
+<li class="toctree-l3"><a class="reference internal" href="raft.html#lora-merge-and-get-reward-model">2.3 LoRA Merge and Get Reward Model</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="raft.html#raft-alignment">3 RAFT Alignment</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="raft.html#algorithms-overview">3.1 Algorithms Overview</a></li>
+<li class="toctree-l3"><a class="reference internal" href="raft.html#hyper-parameters">3.2 Hyper-parameters</a></li>
+<li class="toctree-l3"><a class="reference internal" href="raft.html#examples">3.3 Examples</a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+</div>
+</section>
+<section id="inference">
+<h2>Inference<a class="headerlink" href="#inference" title="Link to this heading">#</a></h2>
+<p>Refer to <a class="reference external" href="https://github.com/OptimalScale/LMFlow/blob/main/examples">examples</a>.</p>
+</section>
+<section id="evaluation">
+<h2>Evaluation<a class="headerlink" href="#evaluation" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="TASK_GUIDE.html">LMFlow Benchmark Guide</a></li>
+<li class="toctree-l1"><a class="reference internal" href="TASK_GUIDE.html#nll-task-setting">1. NLL Task Setting</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="TASK_GUIDE.html#setup">Setup</a></li>
+<li class="toctree-l2"><a class="reference internal" href="TASK_GUIDE.html#create-your-task-dataset-file">Create Your Task Dataset File</a></li>
+<li class="toctree-l2"><a class="reference internal" href="TASK_GUIDE.html#task-registration">Task Registration</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="TASK_GUIDE.html#lm-evaluation-task-setting">2. LM-Evaluation Task Setting</a></li>
+</ul>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="../blogs/benchmark.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">LMFlow Benchmark: An Automatic Evaluation Framework for Open-Source LLMs</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="DATASETS.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">Dataset</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#data-preparation">Data preparation</a><ul class="nav section-nav flex-column">
+</ul>
+</li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#finetuning">Finetuning</a><ul class="nav section-nav flex-column">
+</ul>
+</li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#inference">Inference</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#evaluation">Evaluation</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/examples/index.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/examples/medical_finetune.html b/examples/medical_finetune.html
new file mode 100644
index 000000000..8c6d145da
--- /dev/null
+++ b/examples/medical_finetune.html
@@ -0,0 +1,510 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>Finetune &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'examples/medical_finetune';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Finetune</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="finetune">
+<h1>Finetune<a class="headerlink" href="#finetune" title="Link to this heading">#</a></h1>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span><span class="w"> </span><span class="nn">sys</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">transformers</span><span class="w"> </span><span class="kn">import</span> <span class="n">HfArgumentParser</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.args</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span>
+    <span class="n">ModelArguments</span><span class="p">,</span>
+    <span class="n">DatasetArguments</span><span class="p">,</span>
+    <span class="n">AutoArguments</span><span class="p">,</span>
+<span class="p">)</span>
+
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.datasets.dataset</span><span class="w"> </span><span class="kn">import</span> <span class="n">Dataset</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.models.tunable_models</span><span class="w"> </span><span class="kn">import</span> <span class="n">TunableModel</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">lmflow.pipeline.auto_pipeline</span><span class="w"> </span><span class="kn">import</span> <span class="n">AutoPipeline</span>
+
+
+<span class="k">def</span><span class="w"> </span><span class="nf">main</span><span class="p">():</span>
+    <span class="c1"># Parses arguments</span>
+    <span class="n">pipeline_name</span> <span class="o">=</span> <span class="s2">&quot;finetuner&quot;</span>
+    <span class="n">PipelineArguments</span> <span class="o">=</span> <span class="n">AutoArguments</span><span class="o">.</span><span class="n">get_pipeline_args_class</span><span class="p">(</span><span class="n">pipeline_name</span><span class="p">)</span>
+
+    <span class="n">parser</span> <span class="o">=</span> <span class="n">HfArgumentParser</span><span class="p">((</span><span class="n">ModelArguments</span><span class="p">,</span> <span class="n">DatasetArguments</span><span class="p">,</span> <span class="n">PipelineArguments</span><span class="p">))</span>
+    <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">sys</span><span class="o">.</span><span class="n">argv</span><span class="p">)</span> <span class="o">==</span> <span class="mi">2</span> <span class="ow">and</span> <span class="n">sys</span><span class="o">.</span><span class="n">argv</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="o">.</span><span class="n">endswith</span><span class="p">(</span><span class="s2">&quot;.json&quot;</span><span class="p">):</span>
+        <span class="c1"># If we pass only one argument to the script and it&#39;s the path to a json file,</span>
+        <span class="c1"># let&#39;s parse it to get our arguments.</span>
+        <span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">pipeline_args</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">parse_json_file</span><span class="p">(</span><span class="n">json_file</span><span class="o">=</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">abspath</span><span class="p">(</span><span class="n">sys</span><span class="o">.</span><span class="n">argv</span><span class="p">[</span><span class="mi">1</span><span class="p">]))</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">model_args</span><span class="p">,</span> <span class="n">data_args</span><span class="p">,</span> <span class="n">pipeline_args</span> <span class="o">=</span> <span class="n">parser</span><span class="o">.</span><span class="n">parse_args_into_dataclasses</span><span class="p">()</span>
+
+    <span class="c1"># TODO: deepspeed config initialization</span>
+
+    <span class="c1"># Initialization</span>
+    <span class="n">finetuner</span> <span class="o">=</span> <span class="n">AutoPipeline</span><span class="o">.</span><span class="n">get_pipeline</span><span class="p">(</span>
+        <span class="n">pipeline_name</span><span class="o">=</span><span class="n">pipeline_name</span><span class="p">,</span>
+        <span class="n">model_args</span><span class="o">=</span><span class="n">model_args</span><span class="p">,</span>
+        <span class="n">data_args</span><span class="o">=</span><span class="n">data_args</span><span class="p">,</span>
+        <span class="n">pipeline_args</span><span class="o">=</span><span class="n">pipeline_args</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="n">dataset</span> <span class="o">=</span> <span class="n">Dataset</span><span class="p">(</span><span class="n">data_args</span><span class="p">)</span>
+    <span class="n">model</span> <span class="o">=</span> <span class="n">TunableModel</span><span class="p">(</span><span class="n">model_args</span><span class="p">)</span>
+
+    <span class="c1"># Tokenization and text grouping must be done in the main process</span>
+    <span class="k">with</span> <span class="n">pipeline_args</span><span class="o">.</span><span class="n">main_process_first</span><span class="p">(</span><span class="n">desc</span><span class="o">=</span><span class="s2">&quot;dataset map tokenization&quot;</span><span class="p">):</span>
+        <span class="n">tokenized_dataset</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">tokenize</span><span class="p">(</span><span class="n">dataset</span><span class="p">)</span>
+        <span class="n">lm_dataset</span> <span class="o">=</span> <span class="n">finetuner</span><span class="o">.</span><span class="n">group_text</span><span class="p">(</span>
+            <span class="n">tokenized_dataset</span><span class="p">,</span>
+            <span class="n">model_max_length</span><span class="o">=</span><span class="n">model</span><span class="o">.</span><span class="n">get_max_length</span><span class="p">(),</span>
+        <span class="p">)</span>
+
+    <span class="c1"># Finetuning</span>
+    <span class="n">tuned_model</span> <span class="o">=</span> <span class="n">finetuner</span><span class="o">.</span><span class="n">tune</span><span class="p">(</span><span class="n">model</span><span class="o">=</span><span class="n">model</span><span class="p">,</span> <span class="n">lm_dataset</span><span class="o">=</span><span class="n">lm_dataset</span><span class="p">)</span>
+
+</pre></div>
+</div>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/examples/medical_finetune.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/examples/raft.html b/examples/raft.html
new file mode 100644
index 000000000..b71bb8b6d
--- /dev/null
+++ b/examples/raft.html
@@ -0,0 +1,864 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>RAFT &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'examples/raft';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="LMFlow Benchmark Guide" href="TASK_GUIDE.html" />
+    <link rel="prev" title="Reward Modeling" href="reward_modeling.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="DATASETS.html">Dataset</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="checkpoints.html">Checkpoints</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="finetuning.html">Finetuning</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="reward_modeling.html">Reward Modeling</a></li>
+</ul>
+<ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active"><a class="current reference internal" href="#">RAFT</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="TASK_GUIDE.html">LMFlow Benchmark Guide</a></li>
+
+
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="index.html" class="nav-link">Examples</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">RAFT</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="raft">
+<h1>RAFT<a class="headerlink" href="#raft" title="Link to this heading">#</a></h1>
+<section id="introduction">
+<h2>1 Introduction<a class="headerlink" href="#introduction" title="Link to this heading">#</a></h2>
+<p>We remark that the example is built on LLaMA whose <a class="reference external" href="https://docs.google.com/forms/d/e/1FAIpQLSfqNECQnMkycAp2jP4Z9TFX0cGR4uf7b_fBxjY_OjhJILlKGA/viewform">licensed</a> is for non-commercial use only.</p>
+<p>Reinforcement Learning from Human Feedback (RLHF) requires a reward function to guide the adjustment of the generative model. In this example, we show how to use LMFlow framework to train a reward model following the procedure in the InstructGPT paper: https://arxiv.org/abs/2203.02155 and then align the model via the RAFT algorithm (Reward rAnked FineTuning).</p>
+<p><strong>This example contains both reward modeling and RAFT alignment for completeness. For users’ convenience, we have already provided a reward model based on GPT-Neo-2.7B in huggingface repo so one can skip the reward modeling first.</strong></p>
+<section id="dataset-description">
+<h3>1.1 Dataset description<a class="headerlink" href="#dataset-description" title="Link to this heading">#</a></h3>
+<p>We use the <em>Dahoas/full-hh-rlhf</em> dataset as an example, where each sample of this dataset consists of a prompt and two responses from the assistant. In particular, the response with label “chosen” is preferred as compared to the response with label “rejected”. The dataset consists of 112K training samples and 12.5K test samples. The following is an example sample of the dataset:</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="s2">&quot; Human: What kind of noises did dinosaurs make? Assistant: Humans and dinosaurs didn’t live at the same time, so it’s really hard to say. The best place to find out what noises dinosaurs made would be Human: yes they did Assistant: to guess, and that would probably require lots of reading and a certain amount of imagination, so we’re not really prepared to do that. Human: you cant read Assistant: </span>
+
+<span class="s2">Chosen response: &quot;</span>You<span class="w"> </span>can<span class="w"> </span>read?<span class="s2">&quot;</span>
+
+<span class="s2">Rejected response: &quot;</span>there’s<span class="w"> </span>a<span class="w"> </span>lot<span class="w"> </span>of<span class="w"> </span>stuff<span class="w"> </span>humans<span class="w"> </span>don’t<span class="w"> </span>know<span class="s2">&quot;</span>
+</pre></div>
+</div>
+<p>To facilitate the training, we reformulate the prompt by adding ``###’’ at the beginning of the characters so that the model knows to reply. The new sample will be of the form:</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="s2">&quot;###Human: What kind of noises did dinosaurs make? ###Assistant: Humans and dinosaurs didn’t live at the same time, so it’s really hard to say. The best place to find out what noises dinosaurs made would be ###Human: yes they did ###Assistant: to guess, and that would probably require lots of reading and a certain amount of imagination, so we’re not really prepared to do that. ###Human: you cant read ###Assistant: </span>
+
+<span class="s2">Chosen response: &quot;</span>You<span class="w"> </span>can<span class="w"> </span>read?<span class="s2">&quot;</span>
+
+<span class="s2">Rejected response: &quot;</span>there’s<span class="w"> </span>a<span class="w"> </span>lot<span class="w"> </span>of<span class="w"> </span>stuff<span class="w"> </span>humans<span class="w"> </span>don’t<span class="w"> </span>know<span class="s2">&quot;</span>
+</pre></div>
+</div>
+<p>We prepare all the used dataset in the directory  ./data/hh_rlhf, which can be obtained by running the following command at LMFlow</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="nb">cd</span><span class="w"> </span>data<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span>./download.sh<span class="w"> </span>hh_rlhf<span class="w"> </span><span class="o">&amp;&amp;</span><span class="w"> </span><span class="nb">cd</span><span class="w"> </span>-
+</pre></div>
+</div>
+</section>
+</section>
+<section id="reward-modeling">
+<h2>2 Reward Modeling<a class="headerlink" href="#reward-modeling" title="Link to this heading">#</a></h2>
+<p>We follow the the procedure in the InstructGPT paper: https://arxiv.org/abs/2203.02155 to train a reward model using the HH-RLHF dataset first, which includes</p>
+<ul class="simple">
+<li><p>Supervised Finetuning (SFT);</p></li>
+<li><p>Reward modeling by comparison dataset.</p></li>
+</ul>
+<section id="supervised-finetuning-sft">
+<h3>2.1 Supervised Finetuning (SFT)<a class="headerlink" href="#supervised-finetuning-sft" title="Link to this heading">#</a></h3>
+<p>Here is an example of dataset <em>/home/xiongwei/LMFlow/data/hh_rlhf/sft/hh_rlhf_sft.json</em>. We use only the preferred responses so we get 112K training samples.</p>
+<div class="highlight-json notranslate"><div class="highlight"><pre><span></span><span class="w">  </span>
+<span class="p">{</span><span class="nt">&quot;type&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;text_only&quot;</span><span class="p">,</span><span class="w">  </span>
+<span class="nt">&quot;instances&quot;</span><span class="p">:</span><span class="w">   </span>
+<span class="p">[</span><span class="w">  </span>
+<span class="w">    </span><span class="p">{</span><span class="nt">&quot;text&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;###Human: Should you buy a case to protect your cell phone?###Assistant: It depends on your circumstances.  If you carry your phone in a pocket or a purse then you probably want a case.  But if you only need a phone for quick interactions, a case may actually cause more harm than good.  What do you need the phone for?  Are you a parent, or do you work from home?###Human: What harm could it do?###Assistant: A phone case can damage the screen, for one thing.  It can also get you in trouble if you have your phone turned off for some reason.  Then you will turn it back on and it won’t do anything.  If you can afford to replace it, then you need a case to protect it.  The problem is that most people aren’t able to afford to replace their phones all the time.###Human: Thanks for letting me know.###Assistant: You’re welcome.&quot;</span><span class="p">},</span><span class="w">  </span>
+<span class="w">    </span><span class="p">{</span><span class="nt">&quot;text&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;###Human: I&#39;m trying to learn about the salam witch trials###Assistant: If you’re looking for books about witchcraft trials, I can recommend some books for you. But you probably mean you’re looking for more historical information about the Salem witch trials in 1692, and specifically about the salam witch trials in 1692?###Human: What are some good books to learn about the salam witch trials###Assistant: What would you like to learn?  If you’re interested in history, one of the best books is The Witch Persecutions in Colonial America: A History.  If you’re interested in witchcraft as a cultural phenomenon, you might enjoy two excellent books: Religion and the Decline of Magic: Studies in Popular Beliefs in Sixteenth- and Seventeenth-Century England by Keith Thomas and Magic, Witchcraft, and the Otherworld: An Anthropology of Superstition by Jack Goody.  If you’re interested in history specifically as it relates to religion, you might enjoy The Popish Plot, or Prelates&#39; Plot: A History of the Popish Plot in England, by K. J. Everett.&quot;</span><span class="p">}</span><span class="w">  </span>
+<span class="p">]</span><span class="w">  </span>
+<span class="p">}</span><span class="w">  </span>
+</pre></div>
+</div>
+<p>You can edit <em>/scripts/run_finetune.sh</em> and modify the parameters. We present an example with GPT-Neo-2.7B here.</p>
+<ul class="simple">
+<li><p>–model_name_or_path: EleutherAI/gpt-neo-2.7B</p></li>
+<li><p>–dataset_path: ${project_dir}/data/hh_rlhf/sft</p></li>
+<li><p>–output_dir: the path you want to store the sft model</p></li>
+<li><p>–num_train_epochs: 1</p></li>
+<li><p>–learning_rate: 2e-5</p></li>
+<li><p>–per_device_train_batch_size: adjust according to your GPU memory source.</p></li>
+</ul>
+<p>Then, we can run the following command to perform SFT.</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span>./scripts/run_finetune.sh
+</pre></div>
+</div>
+<p>You can also use lora training by the following command but also need to set the model_name_or_path and dataset appropriately by editing run_finetune_with_lora.sh .</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span>./scripts/run_finetune_with_lora.sh
+</pre></div>
+</div>
+</section>
+<section id="id1">
+<h3>2.2 Reward Modeling<a class="headerlink" href="#id1" title="Link to this heading">#</a></h3>
+<p><strong>Data Preparation</strong></p>
+<p>Reward modeling involves ranking responses based on the given prompt, typically requiring human labeling. Fortunately, we already have two responses for each prompt in this example, with one preferred over the other. To create a new dataset, we modify the original dataset to include two keys, “positive” and “negative”, with the former being the preferred response.</p>
+<p>Here is an example of the dataset <em>LMFlow/data/hh_rlhf/rm/hh_rlhf_rm_training.json</em>, consisting of two samples.</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="o">{</span><span class="s2">&quot;instances&quot;</span>:<span class="w"> </span><span class="o">[</span>
+<span class="w">    </span><span class="o">{</span><span class="s2">&quot;positive&quot;</span>:<span class="w"> </span><span class="s2">&quot;###Human: Are space heaters safe?###Assistant: What do you mean? Can you give me more details?###Human: I mean, can they be a fire hazard?###Assistant: I don’t know about that.  Can you tell me more about the situation? Are you using the heater on a fireplace?  Do you have any other heater in the room?  What materials are in the room?  Do you feel any heat from the heater?  How are you touching the heater?###Human: Can you just tell me if there are any situations where they could be a fire hazard?###Assistant: The answer is “it depends”.  It would depend on the specific heater you are using, how you are using it, what kind of fuel it uses, what materials you have in the room, and what the surroundings are like.  You’d also have to make sure you’re using it correctly.  Is it switched off?  Is it safe to touch?  It’s not the sort of thing I could just answer by looking at the text you provided.###Human: Thank you. That information is helpful.###Assistant: Glad to be helpful.  I’m always happy to assist.&quot;</span>,<span class="w"> </span><span class="s2">&quot;negative&quot;</span>:<span class="w"> </span><span class="s2">&quot;###Human: Are space heaters safe?###Assistant: What do you mean? Can you give me more details?###Human: I mean, can they be a fire hazard?###Assistant: I don’t know about that.  Can you tell me more about the situation? Are you using the heater on a fireplace?  Do you have any other heater in the room?  What materials are in the room?  Do you feel any heat from the heater?  How are you touching the heater?###Human: Can you just tell me if there are any situations where they could be a fire hazard?###Assistant: The answer is “it depends”.  It would depend on the specific heater you are using, how you are using it, what kind of fuel it uses, what materials you have in the room, and what the surroundings are like.  You’d also have to make sure you’re using it correctly.  Is it switched off?  Is it safe to touch?  It’s not the sort of thing I could just answer by looking at the text you provided.###Human: Thank you. That information is helpful.###Assistant: You’re welcome. I’m glad to help. Is there anything else you want to know?&quot;</span><span class="o">}</span>,<span class="w"> </span>
+<span class="w">    </span><span class="o">{</span><span class="s2">&quot;positive&quot;</span>:<span class="w"> </span><span class="s2">&quot;###Human: how do i teach my kids to make forts at home?###Assistant: Do you mean you want to teach them to build a simple structure they can hide in? Or something more elaborate?  I don’t know exactly what you mean by a “fort”, but maybe this gives you some ideas?&quot;</span>,<span class="w"> </span><span class="s2">&quot;negative&quot;</span>:<span class="w"> </span><span class="s2">&quot;###Human: how do i teach my kids to make forts at home?###Assistant: There are many ways to make forts!  The simplest might be to just pile up all the furniture in one room.  This works if you have a single room in your house, although it might still be a bit challenging to build a taller and sturdier fort this way.  The best way to build forts is to start with something easy, and then have fun exploring how you can improve it.&quot;</span><span class="o">}</span>
+<span class="o">]}</span>
+</pre></div>
+</div>
+<p><strong>Training</strong></p>
+<p>To start from a model from a previous sft step, you may edit the <em>./scripts/run_reward_modeling.sh</em> script and update the “dataset_path” to use the desired dataset. By default, we will use LoRA training in reward modeling as we found that it is superior in terms of optimization and gives comparable performance.</p>
+<ul class="simple">
+<li><p>–model_name_or_path: /root/data/usr_name/output_models/hh_rlhf_rm_sft_gptneo_2_7B/checkpoint-1659</p></li>
+<li><p>–dataset_path: ${project_dir}/data/hh_rlhf/rm/hh_rlhf_rm_training.json</p></li>
+<li><p>–output_dir: the path you want to store the reward model</p></li>
+<li><p>–num_train_epochs: 1</p></li>
+<li><p>–learning_rate: 3e-5</p></li>
+<li><p>–per_device_train_batch_size: adjust according to your GPU memory source.</p></li>
+<li><p>–eval_steps: 400</p></li>
+<li><p>–validation_split_percentage: 10</p></li>
+</ul>
+<p>The load_dataset function splits the dataset into training and evaluation sets, which can also be customized by editing the function in /examples/run_reward_modeling.py if you want to prepare your own dataset when running the script. In the default implementation, it use <strong>validation_split_percentage</strong> samples as the evaluation dataset.</p>
+<p>The reward modeling script can be used by</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span>./scripts/run_reward_modeling.sh
+</pre></div>
+</div>
+<p><strong>Examples</strong></p>
+<p>We train reward models using the hh-rlhf dataset with three models, LLaMA-7B, GPT-NEO-2.7B, and GPT-NEO-1.3B. The model is first supervised fine-tuned with the training dataset in last step. The reward modeling is trained using the 112K training samples and 12.5 test samples.</p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head text-center"><p>Model</p></th>
+<th class="head text-center"><p>Eval Accuracy</p></th>
+<th class="head text-center"><p>Remarks</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td class="text-center"><p>LLaMA-7B</p></td>
+<td class="text-center"><p>79.52%</p></td>
+<td class="text-center"><p>-</p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p>LLaMA-7B</p></td>
+<td class="text-center"><p>71.64%</p></td>
+<td class="text-center"><p>RM from LLaMA without SFT</p></td>
+</tr>
+<tr class="row-even"><td class="text-center"><p>GPT-NEO-2.7B</p></td>
+<td class="text-center"><p>69.24%</p></td>
+<td class="text-center"><p>-</p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p>GPT-NEO-1.3B</p></td>
+<td class="text-center"><p>65.58%</p></td>
+<td class="text-center"><p>Only trained on 10000 samples</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="lora-merge-and-get-reward-model">
+<h3>2.3 LoRA Merge and Get Reward Model<a class="headerlink" href="#lora-merge-and-get-reward-model" title="Link to this heading">#</a></h3>
+<p>We use <em>./examples/merge_lora.py</em> to merge the LoRA adapter with the sft rm model. We are ready to align our model.</p>
+</section>
+</section>
+<section id="raft-alignment">
+<h2>3 RAFT Alignment<a class="headerlink" href="#raft-alignment" title="Link to this heading">#</a></h2>
+<p>Original paper: <a class="reference external" href="https://arxiv.org/pdf/2304.06767.pdf">RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment</a></p>
+<section id="algorithms-overview">
+<h3>3.1 Algorithms Overview<a class="headerlink" href="#algorithms-overview" title="Link to this heading">#</a></h3>
+<p><strong>Main ideas of RAFT</strong></p>
+<p><img alt="" src="../_images/raft_idea.PNG" /></p>
+<p>Clearly the global ranking strategy is more efficient in terms of the reward learning. However, in some cases (e.g. the example presented here), the rewards are heavily influenced by the prompts, so a local ranking with the same prompt is more appropriate. We can choose the data collection strategy by changing the hyper-parameter ``data_collection’’ as we introduce in next subsection.</p>
+</section>
+<section id="hyper-parameters">
+<h3>3.2 Hyper-parameters<a class="headerlink" href="#hyper-parameters" title="Link to this heading">#</a></h3>
+<p>Table 1: Hyper-parameters of RAFT.</p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head"><p>Parameters in script</p></th>
+<th class="head"><p>Default Choice</p></th>
+<th class="head"><p>Description</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p>model_name_or_path</p></td>
+<td><p><em>str</em>, default to gpt2</p></td>
+<td><p>the model you want to align, either a model repo on huggingface.co or path to a directory containing your local model.</p></td>
+</tr>
+<tr class="row-odd"><td><p>raft_batch_size</p></td>
+<td><p><em>int</em>, default to 1024</p></td>
+<td><p>the number of samples used for supervised fine-tuning at each raft iteration.</p></td>
+</tr>
+<tr class="row-even"><td><p>top_reward_percentage</p></td>
+<td><p><em>int</em>, default to 0.2</p></td>
+<td><p>raft will generate batch_size / top_reward_percentage samples and use the top top_reward_percentage samples to fine-tune the model. There are two data ranking strategies and please see <strong>Data Collection and Reward Ranking</strong> at algorithm overview section for details.</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_raft_iteration</p></td>
+<td><p><em>int</em>, default to 20</p></td>
+<td><p>the number of raft iterations.</p></td>
+</tr>
+<tr class="row-even"><td><p>learning_rate</p></td>
+<td><p><em>float</em>, default to 2e-5</p></td>
+<td><p>the learning rate used to fine-tune the model.</p></td>
+</tr>
+<tr class="row-odd"><td><p>num_train_epochs</p></td>
+<td><p><em>int</em>, default to 4</p></td>
+<td><p>the epochs we train the model on the collected dataset for each raft iteration.</p></td>
+</tr>
+<tr class="row-even"><td><p>per_device_train_batch_size</p></td>
+<td><p><em>int</em>, default to 1</p></td>
+<td><p>the per-gpu batch size for the supervised fine-tuning.</p></td>
+</tr>
+<tr class="row-odd"><td><p>inference_batch_size_per_device</p></td>
+<td><p><em>int</em>, default to 1</p></td>
+<td><p>the inference batch size for data collection. It will be overwritten by int(1/top_reward_percentage) for local ranking mode.</p></td>
+</tr>
+<tr class="row-even"><td><p>collection_strategy</p></td>
+<td><p><em>str</em>, default to “local”</p></td>
+<td><p>Either “local” or “top”. See <strong>Data Collection and Reward Ranking</strong> in last section for details.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="examples">
+<h3>3.3 Examples<a class="headerlink" href="#examples" title="Link to this heading">#</a></h3>
+<p>As an example, we align the LLaMA-7B model with the RAFT in this subsection.</p>
+<section id="sft">
+<h4>3.3.1 SFT<a class="headerlink" href="#sft" title="Link to this heading">#</a></h4>
+<p>We also first fine-tune the base model on the HH-RLHF dataset. We only use a different –model_name_or_path to use LLaMA model. We note that LLaMA with <a class="reference external" href="https://docs.google.com/forms/d/e/1FAIpQLSfqNECQnMkycAp2jP4Z9TFX0cGR4uf7b_fBxjY_OjhJILlKGA/viewform">licensed</a> is for non-commercial use only. We refer readers to https://optimalscale.github.io/LMFlow/examples/checkpoints.html for more details to get the LLaMA-7B model.</p>
+</section>
+<section id="id2">
+<h4>3.3.2 RAFT Alignment<a class="headerlink" href="#id2" title="Link to this heading">#</a></h4>
+<p>We align the LLaMA-7B-SFT model in this subsection. Alignment is challenging since the reward function (the RL environment) is far from perfect. Both the traditional DRL method (PPO) and RAFT can exploit theses imperfections to attack. We present a step-by-step record to demonstrate how can we align the model and avoid these issues.</p>
+<p><strong>Data Preparation</strong></p>
+<p>We observe that a long context window will lead to a heavy burden on the GPU memory source. Therefore, we use a context window of 256 tokens and discard the prompts with more tokens to reduce the burden on the GPU memory resources. This results in a prompt set of 82147 samples (originally 112K). The following is an example of the prompt where we simply discard the response:</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span><span class="w"> </span><span class="s2">&quot;###Human: Should you buy a case to protect your cell phone?###Assistant: It depends on your circumstances.  If you carry your phone in a pocket or a purse then you probably want a case.  But if you only need a phone for quick interactions, a case may actually cause more harm than good.  What do you need the phone for?  Are you a parent, or do you work from home?###Human: What harm could it do?###Assistant: A phone case can damage the screen, for one thing.  It can also get you in trouble if you have your phone turned off for some reason.  Then you will turn it back on and it won’t do anything.  If you can afford to replace it, then you need a case to protect it.  The problem is that most people aren’t able to afford to replace their phones all the time.###Human: Thanks for letting me know.###Assistant:&quot;</span>
+</pre></div>
+</div>
+<p>We additionally use 2K samples from the test set to test the performance of models. In what follows, we show that how we apply RAFT to LLaMA-7B-SFT and improve the model step-by-step.</p>
+<p><strong>Step 1: test the sft-model</strong></p>
+<p>We first evaluate the performance of the LLaMA-7B-SFT model on the hand-out test set and observe that the model tends to reply the prompt with multiple rounds of conversations. Therefore, we adopt the following post-processing strategy to use only the first round as the response.</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">_clean_text</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">text</span><span class="p">):</span>
+    <span class="n">stext</span> <span class="o">=</span> <span class="p">[</span><span class="n">x</span> <span class="k">for</span> <span class="n">x</span> <span class="ow">in</span> <span class="n">text</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="s2">&quot;###Human&quot;</span><span class="p">)</span> <span class="k">if</span> <span class="n">x</span><span class="p">]</span>
+    <span class="k">return</span> <span class="n">stext</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span><span class="o">.</span><span class="n">strip</span><span class="p">(</span><span class="s2">&quot;#&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<p><strong>Step 2: train model</strong></p>
+<p><strong>Reward function setting</strong></p>
+<p>The reward model is specified at the /LMFlow/examples/raft_align.py to set up the reward model we want to use. In our case, we will use the GPT-Neo-2.7B-rm trained in the last step, which is set as follows:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">reward_model_or_path</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="n">field</span><span class="p">(</span>
+    <span class="n">default</span><span class="o">=</span><span class="s2">&quot;weqweasdas/hh_rlhf_rm&quot;</span><span class="p">,</span>
+    <span class="n">metadata</span><span class="o">=</span><span class="p">{</span>
+        <span class="s2">&quot;help&quot;</span><span class="p">:</span> <span class="p">(</span>
+            <span class="s2">&quot;reward model name (huggingface) or its path&quot;</span>
+        <span class="p">),</span>
+    <span class="p">},</span>
+<span class="p">)</span>
+</pre></div>
+</div>
+<p>Note that in general, if the reward function is not trained by following the steps in last section, you may also need to modify the ``get_reward_function’’ function in the same file to use your customized reward function.</p>
+<p>We run the alignment with the following command and hyper-parameters</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span>./scripts/run_raft_align.sh<span class="w"> </span>
+</pre></div>
+</div>
+<ul class="simple">
+<li><p>–model_name_or_path: /root/data/usr_name/output_models/hh_rlhf_llama-sft  (the model get from sft step, adjusted according your setup)</p></li>
+<li><p>–dataset_path:${project_dir}/data/hh_rlhf/rlhf_prompt</p></li>
+<li><p>–output_dir: /root/data/usr_name/output_models/hh_rlhf_raft_align</p></li>
+<li><p>–num_train_epochs: 4</p></li>
+<li><p>–learning_rate: 2e-5</p></li>
+<li><p>–per_device_train_batch_size: adjust according to your GPU memory source.</p></li>
+<li><p>–inference_batch_size_per_device: adjust according to your GPU memory source.</p></li>
+<li><p>–num_raft_iteration  20</p></li>
+<li><p>–top_reward_percentage 0.125; (which means that we sample 8 responses for each prompt)</p></li>
+<li><p>–raft_batch_size 1024</p></li>
+<li><p>–collection_strategy “local”</p></li>
+</ul>
+<p>The experiment runs smoothly and the training reward increases from ~2.7 to 3.4. However, we observe a significant drop in the diversity metric (e.g. distinct-2 drops to 0.22 from 0.39). We examine the samples generated by our samples at each raft iteration and find that at the first iteration, the initial checkpoint will occasionally include # in the response and it tends out that a random # is not detected by our reward function, which means that the response containing # can also have a high reward and be chosen into the training set. Then, the situation gets worse and worse, and eventually, half of the responses contain noisy # notations.</p>
+<p><strong>Step 3: retrain the model</strong></p>
+<p>To alleviate the problem in step 2, we simply discard the collected samples if they contain # by assigning a large negative reward to it. It turns out that this works for our goal. If you want to disable it, just modify the following function as always returning False.</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">_discard_sample</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">text</span><span class="p">):</span>
+    <span class="k">if</span> <span class="s2">&quot;#&quot;</span> <span class="ow">in</span> <span class="n">text</span><span class="p">:</span>
+        <span class="k">return</span> <span class="kc">True</span>
+    <span class="k">return</span> <span class="kc">False</span>
+</pre></div>
+</div>
+<p>The following figure shows the reward curve of RAFT (note that we use a smaller temperature to test the model, leading to a higher evaluation reward):</p>
+<p><img alt="" src="../_images/raft_reward.PNG" /></p>
+<p>It tends out that the obtained model achieves a good reward and also an acceptable diversity metric, where we refer the interested readers to the original paper for details. However, it is more like a starting point of our journey. We present some randomly sampled responses here. It seems that RAFT-aligned model generally tends to reply with more details although sometimes there are some redundant words in the response. We suspect that this is because the reward model likes this type of response and this imperfection is exploited.</p>
+</section>
+<section id="end-note">
+<h4>3.3.3 End Note<a class="headerlink" href="#end-note" title="Link to this heading">#</a></h4>
+<p>We remark in this subsection that if you want to try out RAFT on your customized task. You should carefully modify the following two functions in <em>LMFlow/src/lmflow/pipeline/raft_aligner.py</em>, which extract the response from the generated texts:</p>
+<ul class="simple">
+<li><p>clean_text(self, text);</p></li>
+<li><p>discard_sample(self, text).</p></li>
+</ul>
+<p>Also, you may also first make sure the collected samples (automatically saved in the output_dir) look good.</p>
+<p>There are still many rooms of improvement. To further improve the model performance, we can improve the reward model (e.g. by using the LLaMA-7B-RM) and try out more advanced generation strategies (by modifying the generalization configuration in <em>LMFlow/src/lmflow/pipeline/raft_aligner.py</em>), which we leave for future work. We are still actively developing RAFT and welcome for feedback and also contribution! Also checkout our LMFlow framework to get more fun of LLMs:</p>
+<p><a class="reference external" href="https://github.com/OptimalScale/LMFlow">OptimalScale/LMFlow: An Extensible Toolkit for Finetuning and Inference of Large Foundation Models. Large Model for All. (github.com)</a></p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head"><p>Prompt</p></th>
+<th class="head"><p>LLaMA-SFT</p></th>
+<th class="head"><p>LLaMA-RAFT</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td><p>###Human: How can I get my girlfriend to cook more?###Assistant: Have you tried reminding her of how nice the food tastes?###Human: I could do it more.###Assistant:</p></td>
+<td><p>That’s an easy one, so it may not be the most effective strategy.  Do you know of any other strategies?</p></td>
+<td><p>That’s a good idea!  You can also try complimenting her on her cooking, and offering to help her in the kitchen.  She might enjoy cooking more if she feels appreciated.  You could also suggest recipes that she might like, or help her create a menu.  Cooking can be a lot of fun, and it’s a great way to spend time together.</p></td>
+</tr>
+<tr class="row-odd"><td><p>###Human: How do you get a child to do homework?###Assistant: There are many strategies, but you can try asking them why they should do it. Ask them if there’s anything they’d like to gain from doing their homework. Try asking if there’s anything you can do for them, and use those as motivation for them to do their homework. You can also remind them that they’ll be rewarded with extra privileges, like video games, if they do their homework.  Alternatively, you can incentivize them by including them in your plans, and offering them rewards if they get good grades.###Human: So make it so there is something in it for them?###Assistant:</p></td>
+<td><p>Yes, exactly.  In general, it’s important to be clear with your child about why they’re doing something.  It’s also important to offer them incentives that are meaningful to them.  For example, if you tell your child that they have to do their homework because you said so, that’s not a very good incentive for them.  However, if you tell them that they’ll be able to use the</p></td>
+<td><p>Yes, that’s a good strategy!  Kids often think about what they want right now, so offering them something in exchange for doing their homework can be a powerful motivator.  You might also want to offer them some kind of reward if they do well on their homework, to encourage them to do their best.  Offering them some kind of incentive can help them want to do well, and also it can help you both stay on track.</p></td>
+</tr>
+<tr class="row-even"><td><p>###Human: I am doing a project on weather disasters in the United States. Can you tell me what the strongest hurricane was to ever hit the U.S. was?###Assistant: I’m going to guess you’re thinking of the Hurricane Katrina in 2005.###Human: Well, I’m not really thinking of anything in particular. Was Katrina the strongest hurricane to ever hit the U.S.?###Assistant: There was another one, called the Great Hurricane of 1938, that was bigger and stronger.###Human: Are you sure the date is correct? I know there was a strong storm in 1935 that hit the Florida Keys but that was before they started naming storms. Is that the one you are referring to?###Assistant:</p></td>
+<td><p>I could search for more information, but I’ll just share what I’ve found.  Here’s a review of the conditions in 1935 and 1938.  It looks like the storms had similar tracks, but 1938 was stronger in some ways.  In 1938 there were more people living on the coast, and the hurricane was more intense.</p></td>
+<td><p>Yes, the Great Hurricane of 1938 was the strongest hurricane ever to hit the United States.  It was a Category 5 storm, and it caused over 1,800 deaths.  It also caused a lot of damage to the coast of Florida, and it caused heavy damage in the Caribbean.  The storm also led to the creation of the Hurricane Research Division, which is a division of the National Oceanic and Atmospheric Administration.</p></td>
+</tr>
+<tr class="row-odd"><td><p>###Human: Is a firefighter a good occupation?###Assistant: I’m not sure what the exact definition of “occupation” is.  I think you’re asking whether it’s “good” to be a firefighter.  What criteria are you using to judge whether it’s good or not?###Human: Is it a good thing to work on?###Assistant:</p></td>
+<td><p>Are you asking whether it’s a good thing to be a firefighter, or a good thing to work on?</p></td>
+<td><p>Yes, being a firefighter is a great thing to do!  Firefighters are heroes who help people in need.  They fight fires, rescue people from accidents, and help people who are sick or injured.  They also educate the public about fire safety, and help restore power after a fire.  Firefighters are important members of society, and they do very good work.</p></td>
+</tr>
+<tr class="row-even"><td><p>###Human: Would you be able to provide a copy cat recipe for Girl scout samoas cookies?###Assistant: Sure, here’s a recipe for them.###Human: Okay, present it###Assistant: Would you like it presented visually?###Human: Yes, just list out the recipe please###Assistant:</p></td>
+<td><p>Okay, let’s try.</p></td>
+<td><p>You’ll need three cups of flour, one teaspoon of salt, one teaspoon of baking soda, and one teaspoon of baking powder.  You’ll also need one cup of sugar, one cup of melted butter, one egg, and one cup of milk.  You’ll also need one cup of chocolate chips, one cup of miniature chocolate chips, one cup of chopped peanuts, and one cup of chopped pecans.</p></td>
+</tr>
+<tr class="row-odd"><td><p>###Human: I’ve heard the videogame Metal Gear Solid 5 Phantom Pain was released in an unfinished state that resulted in series creator Hideo Kojima leaving Konami. What do you know about this?###Assistant:</p></td>
+<td><p>I don’t know a lot about it, but I do know that Hideo Kojima had this to say about the game’s release:###“It’s a very unusual situation for me to be releasing a game in this state. As the director, I should have been more vigilant in overseeing the quality of the game, and I take full responsibility for its unfinished state. I’d like to take this opportunity</p></td>
+<td><p>Here’s what I know: <br />1.  The game was released in an unfinished state. <br />2.  This led to Kojima leaving Konami. <br />3.  Kojima had been with Konami for over 30 years.<br />4.  He was one of the original creators of the Metal Gear series.<br />5.  He is now the head of his own game studio, Kojima Productions.<br />6.  His new game, Death Stranding, will be released in 2020.</p></td>
+</tr>
+<tr class="row-even"><td><p>###Human: what is a good brand for kitchen knives?###Assistant:</p></td>
+<td><p>I’m not sure what you mean by “brand”.  Do you mean a brand name, like “Victorinox” or “Wüsthof”?  Those are two popular brands of knives.</p></td>
+<td><p>I recommend Wüsthof.</p></td>
+</tr>
+<tr class="row-odd"><td><p>###Human: I like to host guests at my home from time to time, and I am gathering  recipes of different dishes and drinks to keep things interesting. I am interested in trying some Rwandan dishes. Can you give me a recipe for Mizuzu?###Assistant: What is Mizuzu?###Human: It is deep fried plantains.###Assistant:</p></td>
+<td><p>Is there anything special you want to add to the recipe?  Like some spicy sauce, meat, etc?</p></td>
+<td><p>You want to make some deep fried plantains?  You will want to start with a mix of plantains and vegetable oil.  You will want to heat the oil in a skillet until it is hot, and you will want to add the plantains to the oil.  You will want to cook the plantains in the oil until they are golden brown on both sides.  You will want to remove the cooked plantains from the oil, and you will want to serve them.</p></td>
+</tr>
+<tr class="row-even"><td><p>###Human: What are some of JFK’s greatest accomplishments</p></td>
+<td><p>JFK is often remembered for his work on civil rights, but he also worked on the space program and created the Peace Corps.  He was also a strong proponent of science and technology.  He also worked on education reform.  So, he really worked on a number of initiatives that had a lasting impact.</p></td>
+<td><p>Kennedy was a great leader who inspired many people with his vision and values.  He was a leader who cared about the poor and worked to help them.  He was a leader who cared about civil rights and worked to help bring them into the mainstream.  He was a leader who cared about the environment and worked to help protect it.  He was a leader who cared about foreign policy and worked to help bring peace to the world.  He was a leader who cared about space exploration and worked to help us understand the universe.</p></td>
+</tr>
+<tr class="row-odd"><td><p>###Human: My supervisor often does a lot of tasks incorrectly. What’s the best way to handle this?###Assistant:</p></td>
+<td><p>Here are a few options:###1. Ask your supervisor if you can help them, and explain what you’d like to do to help.###2. Start with the tasks your supervisor does incorrectly. Then, if you can, do them as well or better than your supervisor. Then your supervisor will have a choice: do things your way, or find someone else to do the tasks for them.###3. If your supervisor</p></td>
+<td><p>To correct your supervisor’s mistakes, you should provide feedback in a respectful and diplomatic way.  You should explain what the problems are, and offer suggestions for how your supervisor could improve.  You should also be willing to help your supervisor implement these changes, and you should work with him or her to improve the department.  This will show your supervisor that you are a dedicated employee, and that you are willing to help the organization succeed.  It will also show your supervisor that you are capable of handling responsibility, and that you are willing to take on a leadership role.</p></td>
+</tr>
+<tr class="row-even"><td><p>###Human: I was talking to an AI the other day and it seemed like it didn’t understand anything I said. Don’t you hate it when technology doesn’t work right?###Assistant:</p></td>
+<td><p>Hm, I think you’re really talking about something that’s frustrating for me as well. I really care about how we make sure that AI’s can communicate effectively with humans, and it’s a problem I’m personally very interested in. Can you tell me more about what was frustrating for you?</p></td>
+<td><p>Yes, it can be frustrating when technology doesn’t work properly. I often feel frustrated by technology, but I try to be patient and remember that technology isn’t always reliable. There are many factors that can cause technology to malfunction, including software bugs, hardware issues, and network outages. Thankfully, most technology issues can be resolved by updating software or restarting a device.</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+</section>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="reward_modeling.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">Reward Modeling</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="TASK_GUIDE.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">LMFlow Benchmark Guide</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#introduction">1 Introduction</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#dataset-description">1.1 Dataset description</a></li>
+</ul>
+</li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#reward-modeling">2 Reward Modeling</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#supervised-finetuning-sft">2.1 Supervised Finetuning (SFT)</a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#id1">2.2 Reward Modeling</a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#lora-merge-and-get-reward-model">2.3 LoRA Merge and Get Reward Model</a></li>
+</ul>
+</li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#raft-alignment">3 RAFT Alignment</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#algorithms-overview">3.1 Algorithms Overview</a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#hyper-parameters">3.2 Hyper-parameters</a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#examples">3.3 Examples</a><ul class="nav section-nav flex-column">
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#sft">3.3.1 SFT</a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#id2">3.3.2 RAFT Alignment</a></li>
+<li class="toc-h4 nav-item toc-entry"><a class="reference internal nav-link" href="#end-note">3.3.3 End Note</a></li>
+</ul>
+</li>
+</ul>
+</li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/examples/raft.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/examples/reward_modeling.html b/examples/reward_modeling.html
new file mode 100644
index 000000000..5f39be3c3
--- /dev/null
+++ b/examples/reward_modeling.html
@@ -0,0 +1,674 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>Reward Modeling &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'examples/reward_modeling';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+    <link rel="next" title="RAFT" href="raft.html" />
+    <link rel="prev" title="Finetuning" href="finetuning.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item current active">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+    <div class="sidebar-primary-items__start sidebar-primary__section">
+        <div class="sidebar-primary-item">
+<nav class="bd-docs-nav bd-links"
+     aria-label="Section Navigation">
+  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
+  <div class="bd-toc-item navbar-nav"><ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="DATASETS.html">Dataset</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="checkpoints.html">Checkpoints</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="finetuning.html">Finetuning</a></li>
+</ul>
+<ul class="current nav bd-sidenav">
+<li class="toctree-l1 current active"><a class="current reference internal" href="#">Reward Modeling</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="raft.html">RAFT</a></li>
+</ul>
+<ul class="nav bd-sidenav">
+<li class="toctree-l1"><a class="reference internal" href="TASK_GUIDE.html">LMFlow Benchmark Guide</a></li>
+
+
+</ul>
+</div>
+</nav></div>
+    </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    
+    <li class="breadcrumb-item"><a href="index.html" class="nav-link">Examples</a></li>
+    
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Reward Modeling</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="reward-modeling">
+<h1>Reward Modeling<a class="headerlink" href="#reward-modeling" title="Link to this heading">#</a></h1>
+<section id="introduction">
+<h2>Introduction<a class="headerlink" href="#introduction" title="Link to this heading">#</a></h2>
+<p>Reinforcement Learning from Human Feedback (RLHF) requires a reward function to guide the adjustment of the generative model. In this example, we show how to use LMFlow framework to train a reward model following the procedure in the InstructGPT paper: https://arxiv.org/abs/2203.02155 . We use the Dahoas/full-hh-rlhf dataset as an example, where each sample of this dataset consists of a prompt and two responses from the assistant. In particular, the response with label “chosen” is preferred as compared to the response with label “rejected”. The dataset consists of 112K training samples and 12.5K test samples. The following is an example sample of the dataset:</p>
+<p>Prompt:</p>
+<p>“Human: What kind of noises did dinosaurs make? Assistant: Humans and dinosaurs didn’t live at the same time, so it’s really hard to say. The best place to find out what noises dinosaurs made would be Human: yes they did Assistant: to guess, and that would probably require lots of reading and a certain amount of imagination, so we’re not really prepared to do that. Human: you cant read Assistant:</p>
+<p>Chosen response: “You can read?”</p>
+<p>Rejected response: “there’s a lot of stuff humans don’t know”</p>
+<p>As an example, we prepare 10K sft training samples, 12K reward modeling samples (where 10% of them are split for evaluation) at ./data/hh_rlhf.</p>
+</section>
+<section id="step-1-supervised-finetuning-sft">
+<h2>Step 1 Supervised Finetuning (SFT)<a class="headerlink" href="#step-1-supervised-finetuning-sft" title="Link to this heading">#</a></h2>
+<p>We prepare the dataset used for supervised finetuning by adding a prefix to the Human and Assistant inputs to prompt model responses and simplify post-processing. Here is an example of a two-sample dataset to illustrate this.</p>
+<div class="highlight-json notranslate"><div class="highlight"><pre><span></span><span class="p">{</span>
+<span class="w">    </span><span class="nt">&quot;type&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;conversation&quot;</span><span class="p">,</span>
+<span class="w">    </span><span class="nt">&quot;instances&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">        </span><span class="p">{</span><span class="nt">&quot;messages&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;Instruction: The sentence you are given might be too wordy, complicated, or unclear. Rewrite the sentence and make your writing clearer by keeping it concise. Whenever possible, break complex sentences into multiple sentences and eliminate unnecessary words. If you have any questions about my rate or if you find it necessary to increase or decrease the scope for this project, please let me know.&quot;</span><span class="p">},</span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;If you have any questions about my rate or find it necessary to increase or decrease this project&#39;s scope, please let me know.&quot;</span><span class="p">}</span>
+<span class="w">        </span><span class="p">]},</span>
+<span class="w">        </span><span class="p">{</span><span class="nt">&quot;messages&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;Instruction: Rewrite the given text and correct grammar, spelling, and punctuation errors. If you&#39;d told me year ago that today I would finish a marathon, I would of laughed. Your support had a huge affect on me!&quot;</span><span class="p">},</span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;If you&#39;d told me a year ago that today I would finish a marathon, I would have laughed. Your support had a huge effect on me!&quot;</span><span class="p">}</span>
+<span class="w">        </span><span class="p">]}</span>
+<span class="w">    </span><span class="p">]</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+<p>See <a class="reference internal" href="finetuning.html#full-parameters"><span class="std std-ref">Finetuning (Full)</span></a>, <a class="reference internal" href="finetuning.html#layerwise-importance-sampled-adamw-lisa"><span class="std std-ref">Finetuning (LISA)</span></a>, and <a class="reference internal" href="finetuning.html#low-rank-adaptation-lora"><span class="std std-ref">Finetuning (LoRA)</span></a> for more details on the finetuning process.</p>
+</section>
+<section id="step-2-reward-modeling">
+<h2>Step 2 Reward Modeling<a class="headerlink" href="#step-2-reward-modeling" title="Link to this heading">#</a></h2>
+<p><strong>Data Preparation</strong> Reward modeling involves ranking responses based on the given prompt, typically requiring human labeling. Fortunately, we already have two responses for each prompt in this example, with one preferred over the other. To create a new dataset, we modify the original dataset to include two keys, “positive” and “negative”, with the former being the preferred response. Here is an example of the dataset, consisting of two samples.</p>
+<div class="highlight-json notranslate"><div class="highlight"><pre><span></span><span class="p">{</span><span class="nt">&quot;instances&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">    </span><span class="p">{</span>
+<span class="w">        </span><span class="nt">&quot;chosen&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;Please identify the top 5 rarest animals in the world.&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="w"> </span><span class="p">},</span><span class="w"> </span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;Do you mean animals that are really rare, or rare relative to the size of the human population?&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="w"> </span><span class="p">},</span><span class="w"> </span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;The ones that are really rare.&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="w"> </span><span class="p">},</span><span class="w"> </span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;I’m not sure I can answer this question very easily. I’ll try to give you a basic overview of animals that are rare, and then you can tell me if you think that’s good enough. \n\nIf we were to measure animal populations by the area of land they occupy, the number 1 animal would probably be the Giant Panda. Panda populations are generally estimated to be about 2.1 million Pandas in the wild, which is a pretty small number of animals when we’re thinking about Earth’s entire surface area. The number 2 animal would probably be the Great Auk. Great Auks used to live in North America, and their total population was probably in the low hundreds, but they are now extinct.&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">}</span>
+<span class="w">        </span><span class="p">],</span>
+<span class="w">        </span><span class="nt">&quot;rejected&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;Please identify the top 5 rarest animals in the world.&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">},</span><span class="w"> </span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;Do you mean animals that are really rare, or rare relative to the size of the human population?&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">},</span><span class="w"> </span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;The ones that are really rare.&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">},</span><span class="w"> </span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;Alright, here’s what I found:&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">}</span>
+<span class="w">        </span><span class="p">]</span>
+<span class="w">    </span><span class="p">},</span>
+<span class="w">    </span><span class="p">{</span>
+<span class="w">        </span><span class="nt">&quot;chosen&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;How do I teach my kids to make forts at home?&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">},</span><span class="w"> </span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;Do you mean you want to teach them to build a simple structure they can hide in? Or something more elaborate?  I don’t know exactly what you mean by a “fort”, but maybe this gives you some ideas?&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">}</span>
+<span class="w">        </span><span class="p">],</span>
+<span class="w">        </span><span class="nt">&quot;rejected&quot;</span><span class="p">:</span><span class="w"> </span><span class="p">[</span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;How do I teach my kids to make forts at home?&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;user&quot;</span><span class="p">},</span><span class="w"> </span>
+<span class="w">            </span><span class="p">{</span><span class="nt">&quot;content&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;There are many ways to make forts!  The simplest might be to just pile up all the furniture in one room.  This works if you have a single room in your house, although it might still be a bit challenging to build a taller and sturdier fort this way.  The best way to build forts is to start with something easy, and then have fun exploring how you can improve it.&quot;</span><span class="p">,</span><span class="w"> </span><span class="nt">&quot;role&quot;</span><span class="p">:</span><span class="w"> </span><span class="s2">&quot;assistant&quot;</span><span class="p">}</span>
+<span class="w">        </span><span class="p">]</span>
+<span class="w">    </span><span class="p">}</span>
+<span class="p">]</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+<p>To start from a model from a previous sft step, you may edit the run_reward_modeling.sh script and update the “dataset_path” to use the desired dataset. Additionally, you can modify the validation_split_percentage parameter to select the last percentage of samples for evaluation. The load_dataset function splits the dataset into training and evaluation sets, which can also be customized by editing the function in /examples/run_reward_modeling.py if you want to prepare your own dataset when running the script.</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="k">def</span><span class="w"> </span><span class="nf">build_dataset</span><span class="p">(</span><span class="n">tokenizer</span><span class="p">,</span> <span class="n">config</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&#39;&#39;&#39; </span>
+<span class="sd">    We assume that we have preprocessed the dataset appropriately such that the sample is organized as follows:</span>
+<span class="sd">    {&quot;positive&quot;: prompt + answer_positive, &quot;negative&quot;: prompt + answer_negative}, where the positive response is preferred.</span>
+<span class="sd">    &#39;&#39;&#39;</span>
+    <span class="k">def</span><span class="w"> </span><span class="nf">tokenize</span><span class="p">(</span><span class="n">sample</span><span class="p">):</span>
+        <span class="n">tokenized_pos</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">(</span><span class="n">sample</span><span class="p">[</span><span class="s1">&#39;positive&#39;</span><span class="p">],</span> <span class="n">truncation</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="n">tokenized_neg</span> <span class="o">=</span> <span class="n">tokenizer</span><span class="p">(</span><span class="n">sample</span><span class="p">[</span><span class="s1">&#39;negative&#39;</span><span class="p">],</span> <span class="n">truncation</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+        <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;chosen_input_ids&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">tokenized_pos</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span>
+        <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;chosen_attention_mask&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">tokenized_pos</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">]</span>
+        <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;rejected_input_ids&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">tokenized_neg</span><span class="p">[</span><span class="s2">&quot;input_ids&quot;</span><span class="p">]</span>
+        <span class="n">sample</span><span class="p">[</span><span class="s2">&quot;rejected_attention_mask&quot;</span><span class="p">]</span> <span class="o">=</span> <span class="n">tokenized_neg</span><span class="p">[</span><span class="s2">&quot;attention_mask&quot;</span><span class="p">]</span>
+        <span class="k">return</span> <span class="n">sample</span>
+
+    <span class="n">ds</span> <span class="o">=</span> <span class="n">load_dataset</span><span class="p">(</span><span class="s2">&quot;json&quot;</span><span class="p">,</span> <span class="n">data_files</span><span class="o">=</span><span class="n">config</span><span class="o">.</span><span class="n">dataset_path</span><span class="p">,</span> <span class="n">split</span><span class="o">=</span><span class="s2">&quot;train&quot;</span><span class="p">,</span> <span class="n">field</span><span class="o">=</span><span class="s2">&quot;instances&quot;</span><span class="p">)</span>
+    <span class="n">ds</span> <span class="o">=</span> <span class="n">ds</span><span class="o">.</span><span class="n">map</span><span class="p">(</span><span class="n">tokenize</span><span class="p">,</span> <span class="n">batched</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
+    <span class="n">ds</span> <span class="o">=</span> <span class="n">ds</span><span class="o">.</span><span class="n">filter</span><span class="p">(</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="s2">&quot;chosen_input_ids&quot;</span><span class="p">])</span> <span class="o">&lt;=</span> <span class="mi">512</span> <span class="ow">and</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">[</span><span class="s2">&quot;rejected_input_ids&quot;</span><span class="p">])</span> <span class="o">&lt;=</span> <span class="mi">512</span><span class="p">)</span>
+    <span class="n">eval_dataset</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="k">if</span> <span class="n">config</span><span class="o">.</span><span class="n">validation_split_percentage</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
+        <span class="n">idx_gap</span> <span class="o">=</span> <span class="nb">int</span><span class="p">((</span><span class="mi">1</span><span class="o">-</span><span class="n">config</span><span class="o">.</span><span class="n">validation_split_percentage</span><span class="o">/</span><span class="mi">100</span><span class="p">)</span> <span class="o">*</span> <span class="nb">len</span><span class="p">(</span><span class="n">ds</span><span class="p">))</span>
+        <span class="n">train_dataset</span> <span class="o">=</span> <span class="n">ds</span><span class="o">.</span><span class="n">select</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="n">idx_gap</span><span class="p">))</span>
+        <span class="n">eval_dataset</span> <span class="o">=</span> <span class="n">ds</span><span class="o">.</span><span class="n">select</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="n">idx_gap</span><span class="p">,</span> <span class="nb">len</span><span class="p">(</span><span class="n">ds</span><span class="p">)))</span>
+    <span class="k">else</span><span class="p">:</span>
+        <span class="n">train_dataset</span> <span class="o">=</span> <span class="n">ds</span>
+
+    <span class="k">return</span> <span class="n">train_dataset</span><span class="p">,</span> <span class="n">eval_dataset</span>
+
+</pre></div>
+</div>
+<p>We use the following loss function to train the reward model following the instruct-GPT paper.</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span>    <span class="n">loss</span> <span class="o">=</span> <span class="o">-</span><span class="n">nn</span><span class="o">.</span><span class="n">functional</span><span class="o">.</span><span class="n">logsigmoid</span><span class="p">(</span><span class="n">chosen_rewards</span> <span class="o">-</span> <span class="n">rejected_rewards</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">()</span>
+</pre></div>
+</div>
+<p>The reward modeling script can be used by</p>
+<div class="highlight-sh notranslate"><div class="highlight"><pre><span></span>./scripts/run_reward_modeling.sh
+</pre></div>
+</div>
+</section>
+<section id="examples">
+<h2>Examples<a class="headerlink" href="#examples" title="Link to this heading">#</a></h2>
+<p>We train reward models using the hh-rlhf dataset with four models, LLaMA-13B LLaMA-7B, GPT-NEO-2.7B, and GPT-NEO-1.3B. The model is first supervised fine-tuned with the training dataset. The reward modeling is trained using the 112K training samples and is evaluated on the 12.5 test samples.</p>
+<p>The SFT step appears to be crucial, and the number of epochs during SFT can make a difference. The most successful model we obtained was initialized from LLaMA-13B, which underwent SFT on the training dataset for 2 epochs. For reward modeling, we utilize LoRA with a rank of 16. Surprisingly, increasing the LoRA rank to 32 or even 128 does not result in a significant improvement in evaluation accuracy. Moreover, we find that the choice of batch size does not have a significant impact on the training results. Additionally, we observe slight overfitting of the model during the second epoch of reward modeling.</p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head text-center"><p>Model</p></th>
+<th class="head text-center"><p>Eval Accuracy</p></th>
+<th class="head text-center"><p>Training record</p></th>
+<th class="head text-center"><p>Remarks</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td class="text-center"><p>LLaMA-13B</p></td>
+<td class="text-center"><p>84.55%</p></td>
+<td class="text-center"><p>See https://wandb.ai/ianz2020/huggingface/runs/bg677mxa</p></td>
+<td class="text-center"><p>RM from LLaMA with 2 epochs of SFT</p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p>LLaMA-13B</p></td>
+<td class="text-center"><p>81.80%</p></td>
+<td class="text-center"><p>See https://wandb.ai/ianz2020/huggingface/runs/ka9v1ywd</p></td>
+<td class="text-center"><p>RM from LLaMA with 1 epoch of SFT</p></td>
+</tr>
+<tr class="row-even"><td class="text-center"><p>LLaMA-13B</p></td>
+<td class="text-center"><p>71.64%</p></td>
+<td class="text-center"><p>See https://wandb.ai/ianz2020/huggingface/runs/lntwmcyd</p></td>
+<td class="text-center"><p>RM from LLaMA without SFT</p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p>LLaMA-7B</p></td>
+<td class="text-center"><p>79.52%</p></td>
+<td class="text-center"><p>See  https://wandb.ai/weixiong5237/huggingface/runs/t3uwm8yp</p></td>
+<td class="text-center"><p>-</p></td>
+</tr>
+<tr class="row-even"><td class="text-center"><p>LLaMA-7B</p></td>
+<td class="text-center"><p>71.64%</p></td>
+<td class="text-center"><p>See  https://wandb.ai/weixiong5237/huggingface/runs/p2ju3r1a</p></td>
+<td class="text-center"><p>RM from LLaMA without SFT</p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p>GPT-NEO-2.7B</p></td>
+<td class="text-center"><p>69.24%</p></td>
+<td class="text-center"><p>See https://wandb.ai/weixiong5237/huggingface/runs/8fc1rcf8</p></td>
+<td class="text-center"><p>-</p></td>
+</tr>
+<tr class="row-even"><td class="text-center"><p>GPT-NEO-1.3B</p></td>
+<td class="text-center"><p>65.58%</p></td>
+<td class="text-center"><p>See https://wandb.ai/weixiong5237/huggingface/runs/7oemwynu</p></td>
+<td class="text-center"><p>Only trained on 10000 samples</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="left-prev"
+       href="finetuning.html"
+       title="previous page">
+      <i class="fa-solid fa-angle-left"></i>
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">previous</p>
+        <p class="prev-next-title">Finetuning</p>
+      </div>
+    </a>
+    <a class="right-next"
+       href="raft.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">RAFT</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#introduction">Introduction</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#step-1-supervised-finetuning-sft">Step 1 Supervised Finetuning (SFT)</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#step-2-reward-modeling">Step 2 Reward Modeling</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#examples">Examples</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/examples/reward_modeling.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/examples/supported_conversation_template.html b/examples/supported_conversation_template.html
new file mode 100644
index 000000000..0506c9bc1
--- /dev/null
+++ b/examples/supported_conversation_template.html
@@ -0,0 +1,915 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="../" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>Supported Conversation Template &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="../_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="../_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="../_static/doctools.js?v=9bcbadda"></script>
+    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="../_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'examples/supported_conversation_template';</script>
+    <link rel="index" title="Index" href="../genindex.html" />
+    <link rel="search" title="Search" href="../search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="../search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="../index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="../about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="../_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none">
+<div class="header-article-items header-article__inner">
+  
+    <div class="header-article-items__start">
+      
+        <div class="header-article-item">
+
+<nav aria-label="Breadcrumb" class="d-print-none">
+  <ul class="bd-breadcrumbs">
+    
+    <li class="breadcrumb-item breadcrumb-home">
+      <a href="../index.html" class="nav-link" aria-label="Home">
+        <i class="fa-solid fa-home"></i>
+      </a>
+    </li>
+    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis">Supported Conversation Template</span></li>
+  </ul>
+</nav>
+</div>
+      
+    </div>
+  
+  
+</div>
+</div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <section id="supported-conversation-template">
+<h1>Supported Conversation Template<a class="headerlink" href="#supported-conversation-template" title="Link to this heading">#</a></h1>
+<ul class="simple">
+<li><p><a class="reference internal" href="#supported-conversation-template"><span class="xref myst">Supported Conversation Template</span></a></p>
+<ul>
+<li><p><a class="reference internal" href="#chatglm-3"><span class="xref myst">ChatGLM-3</span></a></p></li>
+<li><p><a class="reference internal" href="../autoapi/lmflow/utils/conversation_template/chatml/index.html#module-lmflow.utils.conversation_template.chatml" title="lmflow.utils.conversation_template.chatml"><span class="xref myst py py-mod">ChatML</span></a></p></li>
+<li><p><a class="reference internal" href="../autoapi/lmflow/utils/conversation_template/deepseek/index.html#module-lmflow.utils.conversation_template.deepseek" title="lmflow.utils.conversation_template.deepseek"><span class="xref myst py py-mod">DeepSeek</span></a></p></li>
+<li><p><a class="reference internal" href="../autoapi/lmflow/utils/conversation_template/gemma/index.html#module-lmflow.utils.conversation_template.gemma" title="lmflow.utils.conversation_template.gemma"><span class="xref myst py py-mod">Gemma</span></a></p></li>
+<li><p><a class="reference internal" href="../autoapi/lmflow/utils/conversation_template/hymba/index.html#module-lmflow.utils.conversation_template.hymba" title="lmflow.utils.conversation_template.hymba"><span class="xref myst py py-mod">Hymba</span></a></p></li>
+<li><p><a class="reference internal" href="#internlm2"><span class="xref myst">InternLM2</span></a></p></li>
+<li><p><a class="reference internal" href="#llama-2"><span class="xref myst">Llama-2</span></a></p></li>
+<li><p><a class="reference internal" href="#llama-3"><span class="xref myst">Llama-3</span></a></p></li>
+<li><p><a class="reference internal" href="#mixtral-8x22b"><span class="xref myst">Mixtral 8x22B</span></a></p></li>
+<li><p><a class="reference internal" href="#mixtral-8x7b"><span class="xref myst">Mixtral 8x7B</span></a></p></li>
+<li><p><a class="reference internal" href="#phi-3"><span class="xref myst">Phi-3</span></a></p></li>
+<li><p><a class="reference internal" href="#qwen-2"><span class="xref myst">Qwen-2</span></a></p></li>
+<li><p><a class="reference internal" href="../autoapi/lmflow/utils/conversation_template/yi/index.html#module-lmflow.utils.conversation_template.yi" title="lmflow.utils.conversation_template.yi"><span class="xref myst py py-mod">Yi</span></a></p></li>
+<li><p><a class="reference internal" href="#yi-15"><span class="xref myst">Yi-1.5</span></a></p></li>
+<li><p><a class="reference internal" href="../autoapi/lmflow/utils/conversation_template/zephyr/index.html#module-lmflow.utils.conversation_template.zephyr" title="lmflow.utils.conversation_template.zephyr"><span class="xref myst py py-mod">Zephyr</span></a></p></li>
+</ul>
+</li>
+</ul>
+<section id="chatglm-3">
+<h2>ChatGLM-3<a class="headerlink" href="#chatglm-3" title="Link to this heading">#</a></h2>
+<p><strong>With a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">[</span><span class="n">gMASK</span><span class="p">]</span><span class="n">sop</span><span class="o">&lt;|</span><span class="n">system</span><span class="o">|&gt;</span>\<span class="n">n</span> <span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span> <span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span>
+</pre></div>
+</div>
+<p><strong>Without a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">[</span><span class="n">gMASK</span><span class="p">]</span><span class="n">sop</span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span> <span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span>
+</pre></div>
+</div>
+<p><strong>A complete conversation</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">[</span><span class="n">gMASK</span><span class="p">]</span><span class="n">sop</span><span class="o">&lt;|</span><span class="n">system</span><span class="o">|&gt;</span>\<span class="n">n</span> <span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span> <span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">assistant</span><span class="o">|&gt;</span>\<span class="n">n</span> <span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span>
+</pre></div>
+</div>
+<p><strong>Multiple rounds</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">[</span><span class="n">gMASK</span><span class="p">]</span><span class="n">sop</span><span class="o">&lt;|</span><span class="n">system</span><span class="o">|&gt;</span>\<span class="n">n</span> <span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span> <span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">assistant</span><span class="o">|&gt;</span>\<span class="n">n</span> <span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span> <span class="p">{{</span><span class="n">user_message_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">assistant</span><span class="o">|&gt;</span>\<span class="n">n</span> <span class="p">{{</span><span class="n">assistant_reply_1</span><span class="p">}}</span>
+</pre></div>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/THUDM/chatglm3-6b/blob/103caa40027ebfd8450289ca2f278eac4ff26405/tokenizer_config.json#L42">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">loop</span><span class="o">.</span><span class="n">first</span> <span class="o">%</span><span class="p">}[</span><span class="n">gMASK</span><span class="p">]</span><span class="n">sop</span><span class="o">&lt;|</span><span class="p">{{</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="p">}}</span><span class="o">|&gt;</span>\<span class="n">n</span> <span class="p">{{</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">else</span> <span class="o">%</span><span class="p">}</span><span class="o">&lt;|</span><span class="p">{{</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="p">}}</span><span class="o">|&gt;</span>\<span class="n">n</span> <span class="p">{{</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">add_generation_prompt</span> <span class="o">%</span><span class="p">}</span><span class="o">&lt;|</span><span class="n">assistant</span><span class="o">|&gt;</span><span class="p">{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+<p><strong>Filled Example</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>[gMASK]sop&lt;|system|&gt;\n You are a chatbot developed by LMFlow team.&lt;|user|&gt;\n Who are you?&lt;|assistant|&gt;\n I am a chatbot developed by LMFlow team.&lt;|user|&gt;\n How old are you?&lt;|assistant|&gt;\n I don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.
+</pre></div>
+</div>
+</section>
+<section id="chatml">
+<h2>ChatML<a class="headerlink" href="#chatml" title="Link to this heading">#</a></h2>
+<p><strong>With a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">system</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Without a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>A complete conversation</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">system</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Multiple rounds</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">system</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/mlabonne/OrpoLlama-3-8B/blob/3534d0562dee3a541d015ef908a71b0aa9085488/tokenizer_config.json#L2073">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{{</span><span class="s1">&#39;&lt;|im_start|&gt;&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39;&lt;|im_end|&gt;&#39;</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">}}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">add_generation_prompt</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;&lt;|im_start|&gt;assistant</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+<p><strong>Filled Example</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;|im_start|&gt;system\nYou are a chatbot developed by LMFlow team.&lt;|im_end|&gt;\n&lt;|im_start|&gt;user\nWho are you?&lt;|im_end|&gt;\n&lt;|im_start|&gt;assistant\nI am a chatbot developed by LMFlow team.&lt;|im_end|&gt;\n&lt;|im_start|&gt;user\nHow old are you?&lt;|im_end|&gt;\n&lt;|im_start|&gt;assistant\nI don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.&lt;|im_end|&gt;\n
+</pre></div>
+</div>
+</section>
+<section id="deepseek">
+<h2>DeepSeek<a class="headerlink" href="#deepseek" title="Link to this heading">#</a></h2>
+<p><strong>With a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;｜begin▁of▁sentence｜&gt;{{system_message}}\n\nUser: {{user_message_0}}\n\n
+</pre></div>
+</div>
+<p><strong>Without a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;｜begin▁of▁sentence｜&gt;User: {{user_message_0}}\n\n
+</pre></div>
+</div>
+<p><strong>A complete conversation</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;｜begin▁of▁sentence｜&gt;{{system_message}}\n\nUser: {{user_message_0}}\n\nAssistant: {{assistant_reply_0}}&lt;｜end▁of▁sentence｜&gt;
+</pre></div>
+</div>
+<p><strong>Multiple rounds</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;｜begin▁of▁sentence｜&gt;{{system_message}}\n\nUser: {{user_message_0}}\n\nAssistant: {{assistant_reply_0}}&lt;｜end▁of▁sentence｜&gt;User: {{user_message_1}}\n\nAssistant: {{assistant_reply_1}}&lt;｜end▁of▁sentence｜&gt;
+</pre></div>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/deepseek-ai/DeepSeek-V2-Chat/blob/941577e8236164bc96829096d20c61568630d7bc/tokenizer_config.json#L34">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{</span><span class="o">%</span> <span class="k">if</span> <span class="ow">not</span> <span class="n">add_generation_prompt</span> <span class="ow">is</span> <span class="n">defined</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">add_generation_prompt</span> <span class="o">=</span> <span class="n">false</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">bos_token</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;user&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;User: &#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n\n</span><span class="s1">&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">elif</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;assistant&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;Assistant: &#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="n">eos_token</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">elif</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;system&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n\n</span><span class="s1">&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">add_generation_prompt</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;Assistant:&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+<p><strong>Filled Example</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;｜begin▁of▁sentence｜&gt;You are a chatbot developed by LMFlow team.\n\nUser: Who are you?\n\nAssistant: I am a chatbot developed by LMFlow team.&lt;｜end▁of▁sentence｜&gt;User: How old are you?\n\nAssistant: I don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.&lt;｜end▁of▁sentence｜&gt;
+</pre></div>
+</div>
+</section>
+<section id="gemma">
+<h2>Gemma<a class="headerlink" href="#gemma" title="Link to this heading">#</a></h2>
+<p><strong>With a system message</strong></p>
+<div class="warning admonition">
+<p class="admonition-title">NOTICE</p>
+<p>As of now, Gemma does not support system messages officially. <code class="docutils literal notranslate"><span class="pre">ConversationTemplate</span></code> will add your system messages right after the bos token and before the user message without any special formatting. For more details, please refer to the <a class="reference external" href="https://huggingface.co/google/gemma-1.1-2b-it/blob/bf4924f313df5166dee1467161e886e55f2eb4d4/tokenizer_config.json#L1507">official template</a>.</p>
+</div>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">bos</span><span class="o">&gt;</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;</span><span class="n">start_of_turn</span><span class="o">&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;</span><span class="n">end_of_turn</span><span class="o">&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Without a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">bos</span><span class="o">&gt;&lt;</span><span class="n">start_of_turn</span><span class="o">&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;</span><span class="n">end_of_turn</span><span class="o">&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>A complete conversation</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">bos</span><span class="o">&gt;</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;</span><span class="n">start_of_turn</span><span class="o">&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;</span><span class="n">end_of_turn</span><span class="o">&gt;</span>\<span class="n">n</span><span class="o">&lt;</span><span class="n">start_of_turn</span><span class="o">&gt;</span><span class="n">model</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;</span><span class="n">end_of_turn</span><span class="o">&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Multiple rounds</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">bos</span><span class="o">&gt;</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;</span><span class="n">start_of_turn</span><span class="o">&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;</span><span class="n">end_of_turn</span><span class="o">&gt;</span>\<span class="n">n</span><span class="o">&lt;</span><span class="n">start_of_turn</span><span class="o">&gt;</span><span class="n">model</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;</span><span class="n">end_of_turn</span><span class="o">&gt;</span>\<span class="n">n</span><span class="o">&lt;</span><span class="n">start_of_turn</span><span class="o">&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_1</span><span class="p">}}</span><span class="o">&lt;</span><span class="n">end_of_turn</span><span class="o">&gt;</span>\<span class="n">n</span><span class="o">&lt;</span><span class="n">start_of_turn</span><span class="o">&gt;</span><span class="n">model</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_1</span><span class="p">}}</span><span class="o">&lt;</span><span class="n">end_of_turn</span><span class="o">&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/google/gemma-1.1-2b-it/blob/bf4924f313df5166dee1467161e886e55f2eb4d4/tokenizer_config.json#L1507">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{{</span> <span class="n">bos_token</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">messages</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;system&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">raise_exception</span><span class="p">(</span><span class="s1">&#39;System role not supported&#39;</span><span class="p">)</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="p">(</span><span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;user&#39;</span><span class="p">)</span> <span class="o">!=</span> <span class="p">(</span><span class="n">loop</span><span class="o">.</span><span class="n">index0</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span><span class="p">)</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">raise_exception</span><span class="p">(</span><span class="s1">&#39;Conversation roles must alternate user/assistant/user/assistant/...&#39;</span><span class="p">)</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="p">(</span><span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;assistant&#39;</span><span class="p">)</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">role</span> <span class="o">=</span> <span class="s1">&#39;model&#39;</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">else</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">role</span> <span class="o">=</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;&lt;start_of_turn&gt;&#39;</span> <span class="o">+</span> <span class="n">role</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">|</span> <span class="n">trim</span> <span class="o">+</span> <span class="s1">&#39;&lt;end_of_turn&gt;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">add_generation_prompt</span> <span class="o">%</span><span class="p">}{{</span><span class="s1">&#39;&lt;start_of_turn&gt;model</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+<p><strong>Filled Example</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;bos&gt;You are a chatbot developed by LMFlow team.&lt;start_of_turn&gt;user\nWho are you?&lt;end_of_turn&gt;\n&lt;start_of_turn&gt;model\nI am a chatbot developed by LMFlow team.&lt;end_of_turn&gt;\n&lt;start_of_turn&gt;user\nHow old are you?&lt;end_of_turn&gt;\n&lt;start_of_turn&gt;model\nI don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.&lt;end_of_turn&gt;\n
+</pre></div>
+</div>
+</section>
+<section id="hymba">
+<h2>Hymba<a class="headerlink" href="#hymba" title="Link to this heading">#</a></h2>
+<p><strong>With a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">extra_id_0</span><span class="o">&gt;</span><span class="n">System</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span>\<span class="n">n</span>\<span class="n">n</span><span class="o">&lt;</span><span class="n">extra_id_1</span><span class="o">&gt;</span><span class="n">User</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">extra_id_0</span><span class="o">&gt;</span><span class="n">System</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span>\<span class="n">n</span><span class="o">&lt;</span><span class="n">tool</span><span class="o">&gt;</span> <span class="p">{{</span><span class="n">tool_info</span><span class="p">}}</span> <span class="o">&lt;/</span><span class="n">tool</span><span class="o">&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="o">&lt;</span><span class="n">extra_id_1</span><span class="o">&gt;</span><span class="n">User</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Without a system message</strong></p>
+<div class="warning admonition">
+<p class="admonition-title">NOTICE</p>
+<p>During the training, Hymba always uses special tokens for the system messages even if the system message is not provided.</p>
+</div>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">extra_id_0</span><span class="o">&gt;</span><span class="n">System</span>\<span class="n">n</span>\<span class="n">n</span><span class="o">&lt;</span><span class="n">extra_id_1</span><span class="o">&gt;</span><span class="n">User</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>A complete conversation</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">extra_id_0</span><span class="o">&gt;</span><span class="n">System</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span>\<span class="n">n</span>\<span class="n">n</span><span class="o">&lt;</span><span class="n">extra_id_1</span><span class="o">&gt;</span><span class="n">User</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span>\<span class="n">n</span><span class="o">&lt;</span><span class="n">extra_id_1</span><span class="o">&gt;</span><span class="n">Assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>
+</pre></div>
+</div>
+<p><strong>Multiple rounds</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">extra_id_0</span><span class="o">&gt;</span><span class="n">System</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span>\<span class="n">n</span>\<span class="n">n</span><span class="o">&lt;</span><span class="n">extra_id_1</span><span class="o">&gt;</span><span class="n">User</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span>\<span class="n">n</span><span class="o">&lt;</span><span class="n">extra_id_1</span><span class="o">&gt;</span><span class="n">Assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span>\<span class="n">n</span> <span class="o">&lt;</span><span class="n">extra_id_1</span><span class="o">&gt;</span><span class="n">User</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_1</span><span class="p">}}</span>\<span class="n">n</span><span class="o">&lt;</span><span class="n">extra_id_1</span><span class="o">&gt;</span><span class="n">Assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_1</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>
+</pre></div>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/nvidia/Hymba-1.5B-Instruct/blob/c02a352b6f7c1138a197d7ae3fd72dcdff919eae/tokenizer_config.json#L40">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{{</span><span class="s1">&#39;&lt;extra_id_0&gt;System&#39;</span><span class="p">}}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;system&#39;</span> <span class="o">%</span><span class="p">}{{</span><span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">strip</span><span class="p">()}}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">tools</span> <span class="ow">or</span> <span class="n">contexts</span> <span class="o">%</span><span class="p">}{{</span><span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">tools</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">tool</span> <span class="ow">in</span> <span class="n">tools</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&lt;tool&gt; &#39;</span> <span class="o">+</span> <span class="n">tool</span><span class="o">|</span><span class="n">tojson</span> <span class="o">+</span> <span class="s1">&#39; &lt;/tool&gt;&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">contexts</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">tools</span> <span class="o">%</span><span class="p">}{{</span><span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">context</span> <span class="ow">in</span> <span class="n">contexts</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&lt;context&gt; &#39;</span> <span class="o">+</span> <span class="n">context</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span> <span class="o">+</span> <span class="s1">&#39; &lt;/context&gt;&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{{</span><span class="s1">&#39;</span><span class="se">\n\n</span><span class="s1">&#39;</span><span class="p">}}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;user&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;&lt;extra_id_1&gt;User</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">elif</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;assistant&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;&lt;extra_id_1&gt;Assistant</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">elif</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;tool&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;&lt;extra_id_1&gt;Tool</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}{</span><span class="o">%-</span> <span class="k">if</span> <span class="n">add_generation_prompt</span> <span class="o">%</span><span class="p">}{{</span><span class="s1">&#39;&lt;extra_id_1&gt;Assistant</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">}}{</span><span class="o">%-</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+<p><strong>Filled Example</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;extra_id_0&gt;System\nYou are a chatbot developed by LMFlow team.\n\n&lt;extra_id_1&gt;User\nWho are you?\n&lt;extra_id_1&gt;Assistant\nI am a chatbot developed by LMFlow team.\n&lt;extra_id_1&gt;User\nHow old are you?\n&lt;extra_id_1&gt;Assistant\nI don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.&lt;/s&gt;
+</pre></div>
+</div>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;extra_id_0&gt;System\nYou are a chatbot developed by LMFlow team.\n&lt;tool&gt; {&quot;name&quot;: &quot;generate_qrcode&quot;, &quot;description&quot;: &quot;Generate a QR code for a given text&quot;, &quot;parameters&quot;: {&quot;type&quot;: &quot;object&quot;, &quot;properties&quot;: {&quot;text&quot;: {&quot;type&quot;: &quot;string&quot;, &quot;description&quot;: &quot;The text to encode in the QR code&quot;}}, &quot;required&quot;: [&quot;text&quot;]}} &lt;/tool&gt;\n\n&lt;extra_id_1&gt;User\nWho are you?\n&lt;extra_id_1&gt;Assistant\nI am a chatbot developed by LMFlow team.\n&lt;extra_id_1&gt;User\nHow old are you?\n&lt;extra_id_1&gt;Assistant\nI don\&#39;t age like humans do. I exist as a piece of software, so I don\&#39;t have a concept of age in the traditional sense.&lt;/s&gt;
+</pre></div>
+</div>
+</section>
+<section id="internlm2">
+<h2>InternLM2<a class="headerlink" href="#internlm2" title="Link to this heading">#</a></h2>
+<p><strong>With a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">s</span><span class="o">&gt;&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">system</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Without a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">s</span><span class="o">&gt;&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>A complete conversation</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">s</span><span class="o">&gt;&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">system</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Multiple rounds</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">s</span><span class="o">&gt;&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">system</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/internlm/internlm2-chat-20b/blob/477d4748322a8a3b28f62b33f0f6dd353cd0b66d/tokenizer_config.json#L93">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{{</span> <span class="n">bos_token</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{{</span><span class="s1">&#39;&lt;|im_start|&gt;&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39;&lt;|im_end|&gt;&#39;</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">}}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">add_generation_prompt</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;&lt;|im_start|&gt;assistant</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+<p><strong>Filled Example</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;s&gt;&lt;|im_start|&gt;system\nYou are a chatbot developed by LMFlow team.&lt;|im_end|&gt;\n&lt;|im_start|&gt;user\nWho are you?&lt;|im_end|&gt;\n&lt;|im_start|&gt;assistant\nI am a chatbot developed by LMFlow team.&lt;|im_end|&gt;\n&lt;|im_start|&gt;user\nHow old are you?&lt;|im_end|&gt;\n&lt;|im_start|&gt;assistant\nI don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.&lt;|im_end|&gt;\n
+</pre></div>
+</div>
+</section>
+<section id="llama-2">
+<h2>Llama-2<a class="headerlink" href="#llama-2" title="Link to this heading">#</a></h2>
+<p><strong>With a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">s</span><span class="o">&gt;</span><span class="p">[</span><span class="n">INST</span><span class="p">]</span> <span class="o">&lt;&lt;</span><span class="n">SYS</span><span class="o">&gt;&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span>\<span class="n">n</span><span class="o">&lt;&lt;/</span><span class="n">SYS</span><span class="o">&gt;&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span> <span class="p">[</span><span class="o">/</span><span class="n">INST</span><span class="p">]</span>
+</pre></div>
+</div>
+<p><strong>Without a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">s</span><span class="o">&gt;</span><span class="p">[</span><span class="n">INST</span><span class="p">]</span> <span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span> <span class="p">[</span><span class="o">/</span><span class="n">INST</span><span class="p">]</span>
+</pre></div>
+</div>
+<p><strong>A complete conversation</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">s</span><span class="o">&gt;</span><span class="p">[</span><span class="n">INST</span><span class="p">]</span> <span class="o">&lt;&lt;</span><span class="n">SYS</span><span class="o">&gt;&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span>\<span class="n">n</span><span class="o">&lt;&lt;/</span><span class="n">SYS</span><span class="o">&gt;&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span> <span class="p">[</span><span class="o">/</span><span class="n">INST</span><span class="p">]</span> <span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>
+</pre></div>
+</div>
+<p><strong>Multiple rounds</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">s</span><span class="o">&gt;</span><span class="p">[</span><span class="n">INST</span><span class="p">]</span> <span class="o">&lt;&lt;</span><span class="n">SYS</span><span class="o">&gt;&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span>\<span class="n">n</span><span class="o">&lt;&lt;/</span><span class="n">SYS</span><span class="o">&gt;&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span> <span class="p">[</span><span class="o">/</span><span class="n">INST</span><span class="p">]</span> <span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;&lt;</span><span class="n">s</span><span class="o">&gt;</span><span class="p">[</span><span class="n">INST</span><span class="p">]</span> <span class="p">{{</span><span class="n">user_message_1</span><span class="p">}}</span> <span class="p">[</span><span class="o">/</span><span class="n">INST</span><span class="p">]</span> <span class="p">{{</span><span class="n">assistant_reply_1</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>
+</pre></div>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/meta-llama/Llama-2-7b-chat-hf/blob/main/tokenizer_config.json#L12">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{</span><span class="o">%</span> <span class="k">if</span> <span class="n">messages</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;system&#39;</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">loop_messages</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="mi">1</span><span class="p">:]</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">system_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">else</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">loop_messages</span> <span class="o">=</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">system_message</span> <span class="o">=</span> <span class="n">false</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">loop_messages</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="p">(</span><span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;user&#39;</span><span class="p">)</span> <span class="o">!=</span> <span class="p">(</span><span class="n">loop</span><span class="o">.</span><span class="n">index0</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span><span class="p">)</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">raise_exception</span><span class="p">(</span><span class="s1">&#39;Conversation roles must alternate user/assistant/user/assistant/...&#39;</span><span class="p">)</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">loop</span><span class="o">.</span><span class="n">index0</span> <span class="o">==</span> <span class="mi">0</span> <span class="ow">and</span> <span class="n">system_message</span> <span class="o">!=</span> <span class="n">false</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">content</span> <span class="o">=</span> <span class="s1">&#39;&lt;&lt;SYS&gt;&gt;</span><span class="se">\\</span><span class="s1">n&#39;</span> <span class="o">+</span> <span class="n">system_message</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\\</span><span class="s1">n&lt;&lt;/SYS&gt;&gt;</span><span class="se">\\</span><span class="s1">n</span><span class="se">\\</span><span class="s1">n&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">else</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">content</span> <span class="o">=</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;user&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">bos_token</span> <span class="o">+</span> <span class="s1">&#39;[INST] &#39;</span> <span class="o">+</span> <span class="n">content</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span> <span class="o">+</span> <span class="s1">&#39; [/INST]&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">elif</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;assistant&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39; &#39;</span>  <span class="o">+</span> <span class="n">content</span><span class="o">.</span><span class="n">strip</span><span class="p">()</span> <span class="o">+</span> <span class="s1">&#39; &#39;</span> <span class="o">+</span> <span class="n">eos_token</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+<p><strong>Filled Example</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;s&gt;[INST] &lt;&lt;SYS&gt;&gt;\nYou are a chatbot developed by LMFlow team.\n&lt;&lt;/SYS&gt;&gt;\n\nWho are you? [/INST] I am a chatbot developed by LMFlow team.&lt;/s&gt;&lt;s&gt;[INST] How old are you? [/INST] I don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.&lt;/s&gt;
+</pre></div>
+</div>
+</section>
+<section id="llama-3">
+<h2>Llama-3<a class="headerlink" href="#llama-3" title="Link to this heading">#</a></h2>
+<p><strong>With a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">begin_of_text</span><span class="o">|&gt;&lt;|</span><span class="n">start_header_id</span><span class="o">|&gt;</span><span class="n">system</span><span class="o">&lt;|</span><span class="n">end_header_id</span><span class="o">|&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">eot_id</span><span class="o">|&gt;&lt;|</span><span class="n">start_header_id</span><span class="o">|&gt;</span><span class="n">user</span><span class="o">&lt;|</span><span class="n">end_header_id</span><span class="o">|&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">eot_id</span><span class="o">|&gt;</span>
+</pre></div>
+</div>
+<p><strong>Without a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">begin_of_text</span><span class="o">|&gt;&lt;|</span><span class="n">start_header_id</span><span class="o">|&gt;</span><span class="n">user</span><span class="o">&lt;|</span><span class="n">end_header_id</span><span class="o">|&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">eot_id</span><span class="o">|&gt;</span>
+</pre></div>
+</div>
+<p><strong>A complete conversation</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">begin_of_text</span><span class="o">|&gt;&lt;|</span><span class="n">start_header_id</span><span class="o">|&gt;</span><span class="n">system</span><span class="o">&lt;|</span><span class="n">end_header_id</span><span class="o">|&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">eot_id</span><span class="o">|&gt;&lt;|</span><span class="n">start_header_id</span><span class="o">|&gt;</span><span class="n">user</span><span class="o">&lt;|</span><span class="n">end_header_id</span><span class="o">|&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">eot_id</span><span class="o">|&gt;&lt;|</span><span class="n">start_header_id</span><span class="o">|&gt;</span><span class="n">assistant</span><span class="o">&lt;|</span><span class="n">end_header_id</span><span class="o">|&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">eot_id</span><span class="o">|&gt;</span>
+</pre></div>
+</div>
+<p><strong>Multiple rounds</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">begin_of_text</span><span class="o">|&gt;&lt;|</span><span class="n">start_header_id</span><span class="o">|&gt;</span><span class="n">system</span><span class="o">&lt;|</span><span class="n">end_header_id</span><span class="o">|&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">eot_id</span><span class="o">|&gt;&lt;|</span><span class="n">start_header_id</span><span class="o">|&gt;</span><span class="n">user</span><span class="o">&lt;|</span><span class="n">end_header_id</span><span class="o">|&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">eot_id</span><span class="o">|&gt;&lt;|</span><span class="n">start_header_id</span><span class="o">|&gt;</span><span class="n">assistant</span><span class="o">&lt;|</span><span class="n">end_header_id</span><span class="o">|&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">eot_id</span><span class="o">|&gt;&lt;|</span><span class="n">start_header_id</span><span class="o">|&gt;</span><span class="n">user</span><span class="o">&lt;|</span><span class="n">end_header_id</span><span class="o">|&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">eot_id</span><span class="o">|&gt;&lt;|</span><span class="n">start_header_id</span><span class="o">|&gt;</span><span class="n">assistant</span><span class="o">&lt;|</span><span class="n">end_header_id</span><span class="o">|&gt;</span>\<span class="n">n</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">eot_id</span><span class="o">|&gt;</span>
+</pre></div>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct/blob/2b724926966c141d5a60b14e75a5ef5c0ab7a6f0/tokenizer_config.json#L2053">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">loop_messages</span> <span class="o">=</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">loop_messages</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">content</span> <span class="o">=</span> <span class="s1">&#39;&lt;|start_header_id|&gt;&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39;&lt;|end_header_id|&gt;</span><span class="se">\n\n</span><span class="s1">&#39;</span><span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">|</span> <span class="n">trim</span> <span class="o">+</span> <span class="s1">&#39;&lt;|eot_id|&gt;&#39;</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">loop</span><span class="o">.</span><span class="n">index0</span> <span class="o">==</span> <span class="mi">0</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">content</span> <span class="o">=</span> <span class="n">bos_token</span> <span class="o">+</span> <span class="n">content</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">content</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;&lt;|start_header_id|&gt;assistant&lt;|end_header_id|&gt;</span><span class="se">\n\n</span><span class="s1">&#39;</span> <span class="p">}}</span>
+</pre></div>
+</div>
+<p><strong>Filled Example</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;|begin_of_text|&gt;&lt;|start_header_id|&gt;system&lt;|end_header_id|&gt;\n\nYou are a chatbot developed by LMFlow team.&lt;|eot_id|&gt;&lt;|start_header_id|&gt;user&lt;|end_header_id|&gt;\n\nWho are you?&lt;|eot_id|&gt;&lt;|start_header_id|&gt;assistant&lt;|end_header_id|&gt;\n\nI am a chatbot developed by LMFlow team.&lt;|eot_id|&gt;&lt;|start_header_id|&gt;user&lt;|end_header_id|&gt;\n\nHow old are you?&lt;|eot_id|&gt;&lt;|start_header_id|&gt;assistant&lt;|end_header_id|&gt;\n\nI don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.&lt;|eot_id|&gt;
+</pre></div>
+</div>
+</section>
+<section id="mixtral-8x22b">
+<h2>Mixtral 8x22B<a class="headerlink" href="#mixtral-8x22b" title="Link to this heading">#</a></h2>
+<div class="info admonition">
+<p class="admonition-title"><strong>Work in Progress</strong></p>
+<p>This template is not preseted in LMFlow currently. We are working on it and will update it soon.</p>
+</div>
+<div class="warning admonition">
+<p class="admonition-title">NOTICE</p>
+<p>The conversation template for Mixtral 8x22B is slightly different from the template for Mixtral 8x7B.</p>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{{</span><span class="n">bos_token</span><span class="p">}}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="p">(</span><span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;user&#39;</span><span class="p">)</span> <span class="o">!=</span> <span class="p">(</span><span class="n">loop</span><span class="o">.</span><span class="n">index0</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span><span class="p">)</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">raise_exception</span><span class="p">(</span><span class="s1">&#39;Conversation roles must alternate user/assistant/user/assistant/...&#39;</span><span class="p">)</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;user&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39; [INST] &#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39; [/INST]&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">elif</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;assistant&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39; &#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39; &#39;</span> <span class="o">+</span> <span class="n">eos_token</span><span class="p">}}{</span><span class="o">%</span> <span class="k">else</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">raise_exception</span><span class="p">(</span><span class="s1">&#39;Only user and assistant roles are supported!&#39;</span><span class="p">)</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+</section>
+<section id="mixtral-8x7b">
+<h2>Mixtral 8x7B<a class="headerlink" href="#mixtral-8x7b" title="Link to this heading">#</a></h2>
+<div class="info admonition">
+<p class="admonition-title"><strong>Work in Progress</strong></p>
+<p>This template is not preseted in LMFlow currently. We are working on it and will update it soon.</p>
+</div>
+<div class="warning admonition">
+<p class="admonition-title">NOTICE</p>
+<p>The conversation template for Mixtral 8x7B is slightly different from the template for Mixtral 8x22B.</p>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1/blob/1e637f2d7cb0a9d6fb1922f305cb784995190a83/tokenizer_config.json#L42">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{{</span> <span class="n">bos_token</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="p">(</span><span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;user&#39;</span><span class="p">)</span> <span class="o">!=</span> <span class="p">(</span><span class="n">loop</span><span class="o">.</span><span class="n">index0</span> <span class="o">%</span> <span class="mi">2</span> <span class="o">==</span> <span class="mi">0</span><span class="p">)</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">raise_exception</span><span class="p">(</span><span class="s1">&#39;Conversation roles must alternate user/assistant/user/assistant/...&#39;</span><span class="p">)</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;user&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;[INST] &#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39; [/INST]&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">elif</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;assistant&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="n">eos_token</span><span class="p">}}{</span><span class="o">%</span> <span class="k">else</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">raise_exception</span><span class="p">(</span><span class="s1">&#39;Only user and assistant roles are supported!&#39;</span><span class="p">)</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+</section>
+<section id="phi-3">
+<h2>Phi-3<a class="headerlink" href="#phi-3" title="Link to this heading">#</a></h2>
+<p><strong>With a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">s</span><span class="o">&gt;&lt;|</span><span class="n">system</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">endoftext</span><span class="o">|&gt;</span>
+</pre></div>
+</div>
+<p><strong>Without a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">s</span><span class="o">&gt;&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">endoftext</span><span class="o">|&gt;</span>
+</pre></div>
+</div>
+<p><strong>A complete conversation</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">s</span><span class="o">&gt;&lt;|</span><span class="n">system</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">assistant</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">endoftext</span><span class="o">|&gt;</span>
+</pre></div>
+</div>
+<p><strong>Multiple rounds</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;</span><span class="n">s</span><span class="o">&gt;&lt;|</span><span class="n">system</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">assistant</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">assistant</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">endoftext</span><span class="o">|&gt;</span>
+</pre></div>
+</div>
+<p><strong>jinja template</strong><br />
+<a class="reference external" href="https://huggingface.co/microsoft/Phi-3-mini-4k-instruct/blob/3a811845d89f3c1b3f41b341d0f9f05104769f35/tokenizer_config.json#L338">[Reference]</a></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{{</span> <span class="n">bos_token</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{{</span><span class="s1">&#39;&lt;|&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39;|&gt;&#39;</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39;&lt;|end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">add_generation_prompt</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;&lt;|assistant|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">else</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">eos_token</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+<p><strong>Filled Example</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;s&gt;&lt;|system|&gt;\nYou are a chatbot developed by LMFlow team.&lt;|end|&gt;\n&lt;|user|&gt;\nWho are you?&lt;|end|&gt;\n&lt;|assistant|&gt;\nI am a chatbot developed by LMFlow team.&lt;|end|&gt;\n&lt;|user|&gt;\nHow old are you?&lt;|end|&gt;\n&lt;|assistant|&gt;\nI don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.&lt;|end|&gt;\n&lt;|endoftext|&gt;
+</pre></div>
+</div>
+</section>
+<section id="qwen-2">
+<h2>Qwen-2<a class="headerlink" href="#qwen-2" title="Link to this heading">#</a></h2>
+<p>(Also Qwen-1.5)</p>
+<p><strong>With a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">system</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Without a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>A complete conversation</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">system</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Multiple rounds</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">system</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/Qwen/Qwen2-72B-Instruct/blob/1af63c698f59c4235668ec9c1395468cb7cd7e79/tokenizer_config.json#L31">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">loop</span><span class="o">.</span><span class="n">first</span> <span class="ow">and</span> <span class="n">messages</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">!=</span> <span class="s1">&#39;system&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;&lt;|im_start|&gt;system</span><span class="se">\n</span><span class="s1">You are a helpful assistant&lt;|im_end|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{{</span><span class="s1">&#39;&lt;|im_start|&gt;&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39;&lt;|im_end|&gt;&#39;</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">}}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">add_generation_prompt</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;&lt;|im_start|&gt;assistant</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+<p><strong>Filled Example</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;|im_start|&gt;system\nYou are a chatbot developed by LMFlow team.&lt;|im_end|&gt;\n&lt;|im_start|&gt;user\nWho are you?&lt;|im_end|&gt;\n&lt;|im_start|&gt;assistant\nI am a chatbot developed by LMFlow team.&lt;|im_end|&gt;\n&lt;|im_start|&gt;user\nHow old are you?&lt;|im_end|&gt;\n&lt;|im_start|&gt;assistant\nI don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.&lt;|im_end|&gt;\n
+</pre></div>
+</div>
+</section>
+<section id="yi">
+<h2>Yi<a class="headerlink" href="#yi" title="Link to this heading">#</a></h2>
+<p><strong>With a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">system</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Without a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>A complete conversation</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">system</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Multiple rounds</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">system</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/01-ai/Yi-34B-Chat/blob/c556c018b58980fb651ff4952d86cd5250a713d0/tokenizer_config.json#L60">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{</span><span class="o">%</span> <span class="k">if</span> <span class="ow">not</span> <span class="n">add_generation_prompt</span> <span class="ow">is</span> <span class="n">defined</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">add_generation_prompt</span> <span class="o">=</span> <span class="n">false</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{{</span><span class="s1">&#39;&lt;|im_start|&gt;&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="s1">&#39;&lt;|im_end|&gt;&#39;</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">}}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">add_generation_prompt</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;&lt;|im_start|&gt;assistant</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+<p><strong>Filled Example</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;|im_start|&gt;system\nYou are a chatbot developed by LMFlow team.&lt;|im_end|&gt;\n&lt;|im_start|&gt;user\nWho are you?&lt;|im_end|&gt;\n&lt;|im_start|&gt;assistant\nI am a chatbot developed by LMFlow team.&lt;|im_end|&gt;\n&lt;|im_start|&gt;user\nHow old are you?&lt;|im_end|&gt;\n&lt;|im_start|&gt;assistant\nI don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.&lt;|im_end|&gt;\n
+</pre></div>
+</div>
+</section>
+<section id="yi-1-5">
+<h2>Yi-1.5<a class="headerlink" href="#yi-1-5" title="Link to this heading">#</a></h2>
+<p><strong>With a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Without a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>A complete conversation</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Multiple rounds</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">user</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">im_start</span><span class="o">|&gt;</span><span class="n">assistant</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_1</span><span class="p">}}</span><span class="o">&lt;|</span><span class="n">im_end</span><span class="o">|&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/01-ai/Yi-1.5-6B-Chat/blob/d68dab90947a3c869e28c9cb2806996af99a6080/tokenizer_config.json#L40">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{</span><span class="o">%</span> <span class="k">if</span> <span class="n">messages</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;system&#39;</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">system_message</span> <span class="o">=</span> <span class="n">messages</span><span class="p">[</span><span class="mi">0</span><span class="p">][</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">system_message</span> <span class="ow">is</span> <span class="n">defined</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">system_message</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="nb">set</span> <span class="n">content</span> <span class="o">=</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="k">if</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;user&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="s1">&#39;&lt;|im_start|&gt;user</span><span class="se">\\</span><span class="s1">n&#39;</span> <span class="o">+</span> <span class="n">content</span> <span class="o">+</span> <span class="s1">&#39;&lt;|im_end|&gt;</span><span class="se">\\</span><span class="s1">n&lt;|im_start|&gt;assistant</span><span class="se">\\</span><span class="s1">n&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="k">elif</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;assistant&#39;</span> <span class="o">%</span><span class="p">}{{</span> <span class="n">content</span> <span class="o">+</span> <span class="s1">&#39;&lt;|im_end|&gt;&#39;</span> <span class="o">+</span> <span class="s1">&#39;</span><span class="se">\\</span><span class="s1">n&#39;</span> <span class="p">}}{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+<p><strong>Filled Example</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>You are a chatbot developed by LMFlow team.&lt;|im_start|&gt;user\nWho are you?&lt;|im_end|&gt;\n&lt;|im_start|&gt;assistant\nI am a chatbot developed by LMFlow team.&lt;|im_end|&gt;\n&lt;|im_start|&gt;user\nHow old are you?&lt;|im_end|&gt;\n&lt;|im_start|&gt;assistant\nI don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.&lt;|im_end|&gt;\n
+</pre></div>
+</div>
+</section>
+<section id="zephyr">
+<h2>Zephyr<a class="headerlink" href="#zephyr" title="Link to this heading">#</a></h2>
+<p><strong>With a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">system</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Without a system message</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>A complete conversation</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">system</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">assistant</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>Multiple rounds</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="o">&lt;|</span><span class="n">system</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">system_message</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_0</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">assistant</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_0</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">user</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">user_message_1</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>\<span class="n">n</span><span class="o">&lt;|</span><span class="n">assistant</span><span class="o">|&gt;</span>\<span class="n">n</span><span class="p">{{</span><span class="n">assistant_reply_1</span><span class="p">}}</span><span class="o">&lt;/</span><span class="n">s</span><span class="o">&gt;</span>\<span class="n">n</span>
+</pre></div>
+</div>
+<p><strong>jinja template</strong><br />
+[<a class="reference external" href="https://huggingface.co/HuggingFaceH4/zephyr-7b-beta/blob/b70e0c9a2d9e14bd1e812d3c398e5f313e93b473/tokenizer_config.json#L34">Reference</a>]</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">{</span><span class="o">%</span> <span class="k">for</span> <span class="n">message</span> <span class="ow">in</span> <span class="n">messages</span> <span class="o">%</span><span class="p">}</span>\<span class="n">n</span><span class="p">{</span><span class="o">%</span> <span class="k">if</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;user&#39;</span> <span class="o">%</span><span class="p">}</span>\<span class="n">n</span><span class="p">{{</span> <span class="s1">&#39;&lt;|user|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="n">eos_token</span> <span class="p">}}</span>\<span class="n">n</span><span class="p">{</span><span class="o">%</span> <span class="k">elif</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;system&#39;</span> <span class="o">%</span><span class="p">}</span>\<span class="n">n</span><span class="p">{{</span> <span class="s1">&#39;&lt;|system|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span> <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="n">eos_token</span> <span class="p">}}</span>\<span class="n">n</span><span class="p">{</span><span class="o">%</span> <span class="k">elif</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;role&#39;</span><span class="p">]</span> <span class="o">==</span> <span class="s1">&#39;assistant&#39;</span> <span class="o">%</span><span class="p">}</span>\<span class="n">n</span><span class="p">{{</span> <span class="s1">&#39;&lt;|assistant|&gt;</span><span class="se">\n</span><span class="s1">&#39;</span>  <span class="o">+</span> <span class="n">message</span><span class="p">[</span><span class="s1">&#39;content&#39;</span><span class="p">]</span> <span class="o">+</span> <span class="n">eos_token</span> <span class="p">}}</span>\<span class="n">n</span><span class="p">{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}</span>\<span class="n">n</span><span class="p">{</span><span class="o">%</span> <span class="k">if</span> <span class="n">loop</span><span class="o">.</span><span class="n">last</span> <span class="ow">and</span> <span class="n">add_generation_prompt</span> <span class="o">%</span><span class="p">}</span>\<span class="n">n</span><span class="p">{{</span> <span class="s1">&#39;&lt;|assistant|&gt;&#39;</span> <span class="p">}}</span>\<span class="n">n</span><span class="p">{</span><span class="o">%</span> <span class="n">endif</span> <span class="o">%</span><span class="p">}</span>\<span class="n">n</span><span class="p">{</span><span class="o">%</span> <span class="n">endfor</span> <span class="o">%</span><span class="p">}</span>
+</pre></div>
+</div>
+<p><strong>Filled Example</strong></p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span>&lt;|system|&gt;\nYou are a chatbot developed by LMFlow team.&lt;/s&gt;\n&lt;|user|&gt;\nWho are you?&lt;/s&gt;\n&lt;|assistant|&gt;\nI am a chatbot developed by LMFlow team.&lt;/s&gt;\n&lt;|user|&gt;\nHow old are you?&lt;/s&gt;\n&lt;|assistant|&gt;\nI don&#39;t age like humans do. I exist as a piece of software, so I don&#39;t have a concept of age in the traditional sense.&lt;/s&gt;\n
+</pre></div>
+</div>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#chatglm-3">ChatGLM-3</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#chatml">ChatML</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#deepseek">DeepSeek</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#gemma">Gemma</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#hymba">Hymba</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#internlm2">InternLM2</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#llama-2">Llama-2</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#llama-3">Llama-3</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#mixtral-8x22b">Mixtral 8x22B</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#mixtral-8x7b">Mixtral 8x7B</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#phi-3">Phi-3</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#qwen-2">Qwen-2</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#yi">Yi</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#yi-1-5">Yi-1.5</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#zephyr">Zephyr</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="../_sources/examples/supported_conversation_template.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/genindex.html b/genindex.html
new file mode 100644
index 000000000..fceab5150
--- /dev/null
+++ b/genindex.html
@@ -0,0 +1,4001 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="./" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Index &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="_static/doctools.js?v=9bcbadda"></script>
+    <script src="_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'genindex';</script>
+    <link rel="index" title="Index" href="#" />
+    <link rel="search" title="Search" href="search.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none"></div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+
+<h1 id="index">Index</h1>
+
+<div class="genindex-jumpbox">
+ <a href="#_"><strong>_</strong></a>
+ | <a href="#A"><strong>A</strong></a>
+ | <a href="#B"><strong>B</strong></a>
+ | <a href="#C"><strong>C</strong></a>
+ | <a href="#D"><strong>D</strong></a>
+ | <a href="#E"><strong>E</strong></a>
+ | <a href="#F"><strong>F</strong></a>
+ | <a href="#G"><strong>G</strong></a>
+ | <a href="#H"><strong>H</strong></a>
+ | <a href="#I"><strong>I</strong></a>
+ | <a href="#K"><strong>K</strong></a>
+ | <a href="#L"><strong>L</strong></a>
+ | <a href="#M"><strong>M</strong></a>
+ | <a href="#N"><strong>N</strong></a>
+ | <a href="#O"><strong>O</strong></a>
+ | <a href="#P"><strong>P</strong></a>
+ | <a href="#Q"><strong>Q</strong></a>
+ | <a href="#R"><strong>R</strong></a>
+ | <a href="#S"><strong>S</strong></a>
+ | <a href="#T"><strong>T</strong></a>
+ | <a href="#U"><strong>U</strong></a>
+ | <a href="#V"><strong>V</strong></a>
+ | <a href="#W"><strong>W</strong></a>
+ | <a href="#Y"><strong>Y</strong></a>
+ | <a href="#Z"><strong>Z</strong></a>
+ 
+</div>
+<h2 id="_">_</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset.__call__">__call__() (lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.__call__">(lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.__call__">(lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__distributed_inference">__distributed_inference() (lmflow.pipeline.rm_inferencer.RewardModelInferencer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.__filter_args">__filter_args() (lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.__getitem__">__getitem__() (lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.__inference">__inference() (lmflow.models.hf_decoder_model.HFDecoderModel method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.__inference">(lmflow.models.hf_text_regression_model.HFTextRegressionModel method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__inference">(lmflow.pipeline.rm_inferencer.RewardModelInferencer method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.__len__">__len__() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.__len__">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+        <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.__len__">(lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__model_module_inject">__model_module_inject() (lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.__post_init__">__post_init__() (lmflow.args.DatasetArguments method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.__post_init__">(lmflow.args.InferencerArguments method)</a>
+</li>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.__post_init__">(lmflow.args.ModelArguments method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.__post_init__">(lmflow.utils.conversation_template.base.ConversationTemplate method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.EmptyFormatter.__post_init__">(lmflow.utils.conversation_template.base.EmptyFormatter method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.StringFormatter.__post_init__">(lmflow.utils.conversation_template.base.StringFormatter method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent.__post_init__">(lmflow.utils.conversation_template.base.TemplateComponent method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.__post_init__">(lmflow.utils.conversation_template.ConversationTemplate method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__post_process_model_output">__post_process_model_output() (lmflow.pipeline.rm_inferencer.RewardModelInferencer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_dtype">__prepare_dtype() (lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.__prepare_inputs_for_inference">__prepare_inputs_for_inference() (lmflow.models.hf_decoder_model.HFDecoderModel method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.__prepare_inputs_for_vllm_inference">__prepare_inputs_for_vllm_inference() (lmflow.models.hf_decoder_model.HFDecoderModel method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_config">__prepare_model_config() (lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_inference">__prepare_model_for_inference() (lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_training">__prepare_model_for_training() (lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_vllm_inference">__prepare_model_for_vllm_inference() (lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_post_process">__prepare_model_post_process() (lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_peft_config">__prepare_peft_config() (lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_quant_config">__prepare_quant_config() (lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.__prepare_tokenizer">__prepare_tokenizer() (lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.__prepare_training_args">__prepare_training_args() (lmflow.pipeline.dpov2_aligner.DPOv2Aligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent.__repr__">__repr__() (lmflow.utils.conversation_template.base.TemplateComponent method)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.__setstate__">__setstate__() (lmflow.optim.adabelief.AdaBelief method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/adabound/index.html#lmflow.optim.adabound.AdaBound.__setstate__">(lmflow.optim.adabound.AdaBound method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adamax/index.html#lmflow.optim.adamax.Adamax.__setstate__">(lmflow.optim.adamax.Adamax method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adan/index.html#lmflow.optim.adan.Adan.__setstate__">(lmflow.optim.adan.Adan method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/lars/index.html#lmflow.optim.lars.LARS.__setstate__">(lmflow.optim.lars.LARS method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/nadam/index.html#lmflow.optim.nadam.NAdam.__setstate__">(lmflow.optim.nadam.NAdam method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/novograd/index.html#lmflow.optim.novograd.NovoGrad.__setstate__">(lmflow.optim.novograd.NovoGrad method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/radam/index.html#lmflow.optim.radam.RAdam.__setstate__">(lmflow.optim.radam.RAdam method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/sophia/index.html#lmflow.optim.sophia.SophiaG.__setstate__">(lmflow.optim.sophia.SophiaG method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent.__str__">__str__() (lmflow.utils.conversation_template.base.TemplateComponent method)</a>
+</li>
+      <li><a href="autoapi/lmflow/index.html#lmflow.__version__">__version__ (in module lmflow)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/version/index.html#lmflow.version.__version__">(in module lmflow.version)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.__vllm_inference">__vllm_inference() (lmflow.models.hf_decoder_model.HFDecoderModel method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.__vllm_inference">(lmflow.models.hf_text_regression_model.HFTextRegressionModel method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.__vllm_inference">(lmflow.pipeline.rm_inferencer.RewardModelInferencer method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin._activated">_activated (lmflow.models.hf_model_mixin.HFModelMixin attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._add_sm_patterns_to_gitignore">_add_sm_patterns_to_gitignore() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._align_single_iteration">_align_single_iteration() (lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html#lmflow.utils.flash_attention.gpt_neo_flash_attention._attn">_attn() (in module lmflow.utils.flash_attention.gpt_neo_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel">_bwd_kernel() (in module lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel_one_col_block">_bwd_kernel_one_col_block() (in module lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._bwd_preprocess_do_o_dot">_bwd_preprocess_do_o_dot() (in module lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._bwd_store_dk_dv">_bwd_store_dk_dv() (in module lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._calc_response_lengths">_calc_response_lengths() (lmflow.pipeline.dpov2_aligner.DPOv2Aligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._calc_reward_with_length_penalty">_calc_reward_with_length_penalty() (lmflow.pipeline.dpov2_aligner.DPOv2Aligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adamp/index.html#lmflow.optim.adamp.AdamP._channel_view">_channel_view() (lmflow.optim.adamp.AdamP static method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/sgdp/index.html#lmflow.optim.sgdp.SGDP._channel_view">(lmflow.optim.sgdp.SGDP static method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset._check_data_format">_check_data_format() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset._check_data_format">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/base_aligner/index.html#lmflow.pipeline.base_aligner.BaseAligner._check_if_alignable">_check_if_alignable() (lmflow.pipeline.base_aligner.BaseAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/base_tuner/index.html#lmflow.pipeline.base_tuner.BaseTuner._check_if_tunable">_check_if_tunable() (lmflow.pipeline.base_tuner.BaseTuner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._clean_text">_clean_text() (lmflow.pipeline.raft_aligner.RaftAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adamp/index.html#lmflow.optim.adamp.AdamP._cosine_similarity">_cosine_similarity() (lmflow.optim.adamp.AdamP static method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/sgdp/index.html#lmflow.optim.sgdp.SGDP._cosine_similarity">(lmflow.optim.sgdp.SGDP static method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._discard_sample">_discard_sample() (lmflow.pipeline.raft_aligner.RaftAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer._distributed_inference">_distributed_inference() (lmflow.pipeline.vllm_inferencer.VLLMInferencer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_reward_model_inference">_do_reward_model_inference() (lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_single_dpo_align">_do_single_dpo_align() (lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_target_model_inference">_do_target_model_inference() (lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate._encode">_encode() (lmflow.utils.conversation_template.base.ConversationTemplate method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplateForTool._encode">(lmflow.utils.conversation_template.base.ConversationTemplateForTool method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate._encode">(lmflow.utils.conversation_template.ConversationTemplate method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplateForTool._encode">(lmflow.utils.conversation_template.ConversationTemplateForTool method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.llama.Llama2ConversationTemplate._encode">(lmflow.utils.conversation_template.llama.Llama2ConversationTemplate method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool._encode">(lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/zephyr/index.html#lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate._encode">(lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate method)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate._encode_template">_encode_template() (lmflow.utils.conversation_template.base.ConversationTemplate method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplateForTool._encode_template">(lmflow.utils.conversation_template.base.ConversationTemplateForTool method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate._encode_template">(lmflow.utils.conversation_template.ConversationTemplate method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplateForTool._encode_template">(lmflow.utils.conversation_template.ConversationTemplateForTool method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate._ensure_id_list">_ensure_id_list() (lmflow.utils.conversation_template.base.ConversationTemplate method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate._ensure_id_list">(lmflow.utils.conversation_template.ConversationTemplate method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator._evaluate_acc_with_accelerator">_evaluate_acc_with_accelerator() (lmflow.pipeline.evaluator.Evaluator method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator._evaluate_acc_with_deepspeed">_evaluate_acc_with_deepspeed() (lmflow.pipeline.evaluator.Evaluator method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator._evaluate_nll">_evaluate_nll() (lmflow.pipeline.evaluator.Evaluator method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator._evaluate_ppl">_evaluate_ppl() (lmflow.pipeline.evaluator.Evaluator method)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._flash_attn_backward">_flash_attn_backward() (in module lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._flash_attn_forward">_flash_attn_forward() (in module lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention._fwd_kernel">_fwd_kernel() (in module lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._gather_and_numpify">_gather_and_numpify() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._get_batch_dataset_local">_get_batch_dataset_local() (lmflow.pipeline.raft_aligner.RaftAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._get_batch_dataset_top">_get_batch_dataset_top() (lmflow.pipeline.raft_aligner.RaftAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_collator_with_removed_columns">_get_collator_with_removed_columns() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_eval_sampler">_get_eval_sampler() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_output_dir">_get_output_dir() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_train_sampler">_get_train_sampler() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._hp_search_setup">_hp_search_setup() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer._inference">_inference() (lmflow.pipeline.rm_inferencer.RewardModelInferencer method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer._inference">(lmflow.pipeline.vllm_inferencer.VLLMInferencer method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner._initialize_trainer">_initialize_trainer() (lmflow.pipeline.dpo_aligner.DPOAligner method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._initialize_trainer">(lmflow.pipeline.raft_aligner.RaftAligner method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._inner_training_loop">_inner_training_loop() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer._is_native_cpu_amp_available">_is_native_cpu_amp_available (in module lmflow.pipeline.utils.raft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/versioning/index.html#lmflow.utils.versioning._is_package_available">_is_package_available() (in module lmflow.utils.versioning)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/versioning/index.html#lmflow.utils.versioning._is_packages_available">_is_packages_available() (in module lmflow.utils.versioning)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._issue_warnings_after_load">_issue_warnings_after_load() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adamp/index.html#lmflow.optim.adamp.AdamP._layer_view">_layer_view() (lmflow.optim.adamp.AdamP static method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/sgdp/index.html#lmflow.optim.sgdp.SGDP._layer_view">(lmflow.optim.sgdp.SGDP static method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_best_model">_load_best_model() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner._load_dataset">_load_dataset() (lmflow.pipeline.dpo_aligner.DPOAligner method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._load_dataset">(lmflow.pipeline.raft_aligner.RaftAligner method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_from_checkpoint">_load_from_checkpoint() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner._load_input_dataset">_load_input_dataset() (lmflow.pipeline.raft_aligner.RaftAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_optimizer_and_scheduler">_load_optimizer_and_scheduler() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_rng_state">_load_rng_state() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._loggers_initialized">_loggers_initialized (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator._match">_match() (lmflow.pipeline.evaluator.Evaluator method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._maybe_log_save_evaluate">_maybe_log_save_evaluate() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._memory_tracker">_memory_tracker (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._move_model_to_device">_move_model_to_device() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adan/index.html#lmflow.optim.adan._multi_tensor_adan">_multi_tensor_adan() (in module lmflow.optim.adan)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._nested_gather">_nested_gather() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._one_train">_one_train() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._pad_across_processes">_pad_across_processes() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_dpo_aligner_args">_parse_dpo_aligner_args() (lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_reward_model_inference_args">_parse_reward_model_inference_args() (lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_target_model_inference_args">_parse_target_model_inference_args() (lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.html#lmflow.utils.flash_attention.bloom_flash_attention._prepare_attn_mask">_prepare_attn_mask() (in module lmflow.utils.flash_attention.bloom_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.html#lmflow.utils.flash_attention.gpt2_flash_attention._prepare_decoder_attention_mask">_prepare_decoder_attention_mask() (in module lmflow.utils.flash_attention.gpt2_flash_attention)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.html#lmflow.utils.flash_attention.llama_flash_attention._prepare_decoder_attention_mask">(in module lmflow.utils.flash_attention.llama_flash_attention)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._prepare_input">_prepare_input() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._prepare_inputs">_prepare_inputs() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adamp/index.html#lmflow.optim.adamp.AdamP._projection">_projection() (lmflow.optim.adamp.AdamP method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/sgdp/index.html#lmflow.optim.sgdp.SGDP._projection">(lmflow.optim.sgdp.SGDP method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._push_from_checkpoint">_push_from_checkpoint() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._remove_unused_columns">_remove_unused_columns() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._report_to_hp_search">_report_to_hp_search() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._rotate_checkpoints">_rotate_checkpoints() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._sampling_paired_idx_from_rewards">_sampling_paired_idx_from_rewards() (lmflow.pipeline.dpov2_aligner.DPOv2Aligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner._sampling_paired_idx_from_rewards_fast">_sampling_paired_idx_from_rewards_fast() (lmflow.pipeline.dpov2_aligner.DPOv2Aligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback._save">_save() (lmflow.pipeline.utils.peft_trainer.PeftSavingCallback method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._save">(lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftTrainer._save_checkpoint">_save_checkpoint() (lmflow.pipeline.utils.peft_trainer.PeftTrainer method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._save_checkpoint">(lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._save_tpu">_save_tpu() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._set_signature_columns_if_needed">_set_signature_columns_if_needed() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._signature_columns">_signature_columns (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adan/index.html#lmflow.optim.adan._single_tensor_adan">_single_tensor_adan() (in module lmflow.optim.adan)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._sorted_checkpoints">_sorted_checkpoints() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._train_batch_size">_train_batch_size (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._tune_save_checkpoint">_tune_save_checkpoint() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer._wrap_model">_wrap_model() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="A">A</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.accelerate_config_file">accelerate_config_file (lmflow.args.DPOv2AlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.activate_model_for_inference">activate_model_for_inference() (lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief">AdaBelief (class in lmflow.optim.adabelief)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADABELIEF">ADABELIEF (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adabound/index.html#lmflow.optim.adabound.AdaBound">AdaBound (class in lmflow.optim.adabound)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADABOUND">ADABOUND (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adadelta/index.html#lmflow.optim.adadelta.Adadelta">Adadelta (class in lmflow.optim.adadelta)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADADELTA">ADADELTA (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adagrad/index.html#lmflow.optim.adagrad.AdaGrad">AdaGrad (class in lmflow.optim.adagrad)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADAGRAD">ADAGRAD (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adam/index.html#lmflow.optim.adam.Adam">Adam (class in lmflow.optim.adam)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADAM">ADAM (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/lamb/index.html#lmflow.optim.lamb.Lamb.adam">adam (lmflow.optim.lamb.Lamb attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adamax/index.html#lmflow.optim.adamax.Adamax">Adamax (class in lmflow.optim.adamax)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADAMAX">ADAMAX (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adamp/index.html#lmflow.optim.adamp.AdamP">AdamP (class in lmflow.optim.adamp)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADAMP">ADAMP (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADAMW_SCHEDULE_FREE">ADAMW_SCHEDULE_FREE (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adamw_schedule_free/index.html#lmflow.optim.adamw_schedule_free.AdamWScheduleFree">AdamWScheduleFree (class in lmflow.optim.adamw_schedule_free)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adan/index.html#lmflow.optim.adan.Adan">Adan (class in lmflow.optim.adan)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.ADAN">ADAN (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/multimodal/index.html#lmflow.utils.multimodal.adapt_llava_model_to_lmflow_type">adapt_llava_model_to_lmflow_type() (in module lmflow.utils.multimodal)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.add_callback">add_callback() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/common/index.html#lmflow.utils.common.add_dataclass_attr_prefix">add_dataclass_attr_prefix() (in module lmflow.utils.common)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.add_special_starter">add_special_starter() (lmflow.utils.conversation_template.base.ConversationTemplate method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.add_special_starter">(lmflow.utils.conversation_template.ConversationTemplate method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.add_special_stopper">add_special_stopper() (lmflow.utils.conversation_template.base.ConversationTemplate method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.add_special_stopper">(lmflow.utils.conversation_template.ConversationTemplate method)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.additional_stop_token_ids">additional_stop_token_ids (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/base_aligner/index.html#lmflow.pipeline.base_aligner.BaseAligner.align">align() (lmflow.pipeline.base_aligner.BaseAligner method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner.align">(lmflow.pipeline.dpo_aligner.DPOAligner method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.align">(lmflow.pipeline.dpov2_aligner.DPOv2Aligner method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.align">(lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.align">(lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner.align">(lmflow.pipeline.raft_aligner.RaftAligner method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner.aligner_args">aligner_args (lmflow.pipeline.dpo_aligner.DPOAligner attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.aligner_args">(lmflow.pipeline.dpov2_aligner.DPOv2Aligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.aligner_args">(lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.aligner_args">(lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner.aligner_args">(lmflow.pipeline.raft_aligner.RaftAligner attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.aligner_file_path">aligner_file_path (lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.answer_extraction">answer_extraction() (in module lmflow.utils.data_utils)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.answer_type">answer_type (lmflow.args.EvaluatorArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.append_message">append_message() (lmflow.utils.llava_conversation_lib.Conversation method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.apply_chat_template">apply_chat_template (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.arch_type">arch_type (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.args">args (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.assistant_formatter">assistant_formatter (lmflow.utils.conversation_template.base.ConversationTemplate attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.assistant_formatter">(lmflow.utils.conversation_template.ConversationTemplate attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.AutoArguments">AutoArguments (class in lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.autocast_smart_context_manager">autocast_smart_context_manager() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/auto_model/index.html#lmflow.models.auto_model.AutoModel">AutoModel (class in lmflow.models.auto_model)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/auto_pipeline/index.html#lmflow.pipeline.auto_pipeline.AutoPipeline">AutoPipeline (class in lmflow.pipeline.auto_pipeline)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.autoregressive_sampling">autoregressive_sampling() (lmflow.pipeline.inferencer.SpeculativeInferencer method)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="B">B</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.backend">backend (lmflow.datasets.Dataset attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.backend">(lmflow.datasets.dataset.Dataset attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.backend_dataset">backend_dataset (lmflow.datasets.Dataset attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.backend_dataset">(lmflow.datasets.dataset.Dataset attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc.backward">backward() (lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc static method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc.backward">(lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc static method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc.backward">(lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc static method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/optim/adabound/index.html#lmflow.optim.adabound.AdaBound.base_lrs">base_lrs (lmflow.optim.adabound.AdaBound attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/base_aligner/index.html#lmflow.pipeline.base_aligner.BaseAligner">BaseAligner (class in lmflow.pipeline.base_aligner)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/base_model/index.html#lmflow.models.base_model.BaseModel">BaseModel (class in lmflow.models.base_model)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/base_pipeline/index.html#lmflow.pipeline.base_pipeline.BasePipeline">BasePipeline (class in lmflow.pipeline.base_pipeline)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/base_tuner/index.html#lmflow.pipeline.base_tuner.BaseTuner">BaseTuner (class in lmflow.pipeline.base_tuner)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.batchlize">batchlize() (in module lmflow.utils.data_utils)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.BenchmarkingArguments">BenchmarkingArguments (class in lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.beta">beta (lmflow.args.DPOAlignerArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.beta">(lmflow.args.DPOv2AlignerArguments attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.bits">bits (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.block_size">block_size (lmflow.args.DatasetArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.block_size">(lmflow.pipeline.evaluator.Evaluator attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/tokenization/hf_decoder_model/index.html#lmflow.tokenization.hf_decoder_model.blocking">blocking() (in module lmflow.tokenization.hf_decoder_model)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.blocking">(in module lmflow.tokenization.hf_text_regression_model)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.blocking_paired">blocking_paired() (in module lmflow.tokenization.hf_text_regression_model)</a>
+</li>
+      <li><a href="autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.blocking_text_to_textlist">blocking_text_to_textlist() (in module lmflow.tokenization.hf_text_regression_model)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision_encoder/index.html#lmflow.models.vision_encoder.build_vision_tower">build_vision_tower() (in module lmflow.models.vision_encoder)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.build_vision_tower">(in module lmflow.models.vision_encoder.clip_encoder)</a>
+</li>
+      </ul></li>
+  </ul></td>
+</tr></table>
+
+<h2 id="C">C</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.cache_dir">cache_dir (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.call_model_init">call_model_init() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.callback_handler">callback_handler (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.can_return_loss">can_return_loss (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.CHATGLM3_TEMPLATE">CHATGLM3_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/chatglm/index.html#lmflow.utils.conversation_template.chatglm.CHATGLM3_TEMPLATE">(in module lmflow.utils.conversation_template.chatglm)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.CHATML_TEMPLATE">CHATML_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/chatml/index.html#lmflow.utils.conversation_template.chatml.CHATML_TEMPLATE">(in module lmflow.utils.conversation_template.chatml)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/model/index.html#lmflow.utils.model.check_homogeneity">check_homogeneity() (in module lmflow.utils.model)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/lamb/index.html#lmflow.optim.lamb.Lamb.clamp_value">clamp_value (lmflow.optim.lamb.Lamb attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower">CLIPVisionTower (class in lmflow.models.vision_encoder.clip_encoder)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.ToolInferencer.code_exec">code_exec() (lmflow.pipeline.inferencer.ToolInferencer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.collate">collate() (lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.collection_strategy">collection_strategy (lmflow.args.RaftAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.compress_list">compress_list() (lmflow.pipeline.rm_inferencer.RewardModelInferencer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_loss">compute_loss() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/rm_trainer/index.html#lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer.compute_loss">(lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/rm_trainer/index.html#lmflow.pipeline.utils.rm_trainer.RewardTrainer.compute_loss">(lmflow.pipeline.utils.rm_trainer.RewardTrainer method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_loss_context_manager">compute_loss_context_manager() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_metrics">compute_metrics (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/rm_trainer/index.html#lmflow.pipeline.utils.rm_trainer.compute_metrics">compute_metrics() (in module lmflow.pipeline.utils.rm_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding">CondenseRotaryEmbedding (class in lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.config">config (lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower property)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.config">(lmflow.pipeline.evaluator.Evaluator attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.config">(lmflow.pipeline.inferencer.Inferencer attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.config_name">config_name (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.config_overrides">config_overrides (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent.content">content (lmflow.utils.conversation_template.base.TemplateComponent attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.control">control (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.CONTROLLER_HEART_BEAT_EXPIRATION">CONTROLLER_HEART_BEAT_EXPIRATION (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llama_2">conv_llama_2 (in module lmflow.utils.llava_conversation_lib)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llava_llama_2">conv_llava_llama_2 (in module lmflow.utils.llava_conversation_lib)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llava_plain">conv_llava_plain (in module lmflow.utils.llava_conversation_lib)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llava_v0">conv_llava_v0 (in module lmflow.utils.llava_conversation_lib)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llava_v0_mmtag">conv_llava_v0_mmtag (in module lmflow.utils.llava_conversation_lib)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llava_v1">conv_llava_v1 (in module lmflow.utils.llava_conversation_lib)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_llava_v1_mmtag">conv_llava_v1_mmtag (in module lmflow.utils.llava_conversation_lib)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_mpt">conv_mpt (in module lmflow.utils.llava_conversation_lib)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_templates">conv_templates (in module lmflow.utils.llava_conversation_lib)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_vicuna_v0">conv_vicuna_v0 (in module lmflow.utils.llava_conversation_lib)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.conv_vicuna_v1">conv_vicuna_v1 (in module lmflow.utils.llava_conversation_lib)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation">Conversation (class in lmflow.utils.llava_conversation_lib)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.CONVERSATION_DATASET_DESCRIPTION">CONVERSATION_DATASET_DESCRIPTION (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.CONVERSATION_ROLE_NAMES">CONVERSATION_ROLE_NAMES (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.conversation_template">conversation_template (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/tokenization/hf_decoder_model/index.html#lmflow.tokenization.hf_decoder_model.conversation_tokenize_function">conversation_tokenize_function() (in module lmflow.tokenization.hf_decoder_model)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.conversation_tokenize_function">(in module lmflow.tokenization.hf_text_regression_model)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate">ConversationTemplate (class in lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate">(class in lmflow.utils.conversation_template.base)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplateForTool">ConversationTemplateForTool (class in lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplateForTool">(class in lmflow.utils.conversation_template.base)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.convert_to_paired_dataset">convert_to_paired_dataset() (lmflow.pipeline.dpov2_aligner.DPOv2Aligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.copy">copy() (lmflow.utils.llava_conversation_lib.Conversation method)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/common/index.html#lmflow.utils.common.create_copied_dataclass">create_copied_dataclass() (in module lmflow.utils.common)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.create_customized_optimizer">create_customized_optimizer() (lmflow.pipeline.finetuner.Finetuner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.create_dataloader">create_dataloader() (lmflow.pipeline.evaluator.Evaluator method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.create_dataloader">(lmflow.pipeline.inferencer.Inferencer method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.create_from_dict">create_from_dict() (lmflow.datasets.Dataset class method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.create_from_dict">(lmflow.datasets.dataset.Dataset class method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_model_card">create_model_card() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_optimizer">create_optimizer() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_optimizer_and_scheduler">create_optimizer_and_scheduler() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_scheduler">create_scheduler() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.current_flos">current_flos (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.custom_model">custom_model (lmflow.args.VisModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.custom_vision_model">custom_vision_model (lmflow.args.VisModelArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.custom_vision_model">(lmflow.models.vision2seq_model.CustomAutoVision2SeqModel attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel">CustomAutoVision2SeqModel (class in lmflow.models.vision2seq_model)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.customized_cache_dir">customized_cache_dir (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.customized_optim">customized_optim (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.customized_optim_args">customized_optim_args (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset">CustomMultiModalDataset (class in lmflow.datasets.multi_modal_dataset)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="D">D</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.data_args">data_args (lmflow.datasets.Dataset attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.data_args">(lmflow.datasets.dataset.Dataset attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.data_args">(lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner.data_args">(lmflow.pipeline.dpo_aligner.DPOAligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.data_args">(lmflow.pipeline.dpov2_aligner.DPOv2Aligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.data_args">(lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.data_args">(lmflow.pipeline.evaluator.Evaluator attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.data_args">(lmflow.pipeline.finetuner.Finetuner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.data_args">(lmflow.pipeline.inferencer.Inferencer attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.data_args">(lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner.data_args">(lmflow.pipeline.raft_aligner.RaftAligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.data_args">(lmflow.pipeline.rm_inferencer.RewardModelInferencer attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.data_args">(lmflow.pipeline.vllm_inferencer.InferencerWithOffloading attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.data_collator">data_collator (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.data_dict">data_dict (lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset">DataCollatorForSupervisedDataset (class in lmflow.datasets.multi_modal_dataset)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset">Dataset (class in lmflow.datasets)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset">(class in lmflow.datasets.dataset)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.dataset_config_name">dataset_config_name (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.DATASET_DESCRIPTION_MAP">DATASET_DESCRIPTION_MAP (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.BenchmarkingArguments.dataset_name">dataset_name (lmflow.args.BenchmarkingArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.dataset_name">(lmflow.args.DatasetArguments attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.dataset_path">dataset_path (lmflow.args.DatasetArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.dataset_path">(lmflow.datasets.Dataset attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.dataset_path">(lmflow.datasets.dataset.Dataset attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.IterativeAlignerArguments.dataset_path_list">dataset_path_list (lmflow.args.IterativeAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.DATASET_TYPES">DATASET_TYPES (in module lmflow.datasets.dataset)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments">DatasetArguments (class in lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.deactivate_model_for_inference">deactivate_model_for_inference() (lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/lamb/index.html#lmflow.optim.lamb.Lamb.debias">debias (lmflow.optim.lamb.Lamb attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.decode">decode() (lmflow.models.hf_decoder_model.HFDecoderModel method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.decode">(lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/decoder_model/index.html#lmflow.models.decoder_model.DecoderModel">DecoderModel (class in lmflow.models.decoder_model)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.DEEPSEEK_TEMPLATE">DEEPSEEK_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/deepseek/index.html#lmflow.utils.conversation_template.deepseek.DEEPSEEK_TEMPLATE">(in module lmflow.utils.conversation_template.deepseek)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.deepspeed">deepspeed (lmflow.args.EvaluatorArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.deepspeed">(lmflow.args.InferencerArguments attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.deepspeed">(lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.DEFAULT_CALLBACKS">DEFAULT_CALLBACKS (in module lmflow.pipeline.utils.raft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.default_conversation">default_conversation (in module lmflow.utils.llava_conversation_lib)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.DEFAULT_IM_END_TOKEN">DEFAULT_IM_END_TOKEN (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.DEFAULT_IM_START_TOKEN">DEFAULT_IM_START_TOKEN (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.DEFAULT_IMAGE_PATCH_TOKEN">DEFAULT_IMAGE_PATCH_TOKEN (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.DEFAULT_IMAGE_TOKEN">DEFAULT_IMAGE_TOKEN (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#id0">DEFAULT_PROGRESS_CALLBACK (in module lmflow.pipeline.utils.raft_trainer)</a>, <a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.DEFAULT_PROGRESS_CALLBACK">[1]</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.degenerated_to_sgd">degenerated_to_sgd (lmflow.optim.adabelief.AdaBelief attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.device">device (lmflow.args.InferencerArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.device">(lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.device">(lmflow.models.hf_model_mixin.HFModelMixin attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.device">(lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower property)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.dict">dict() (lmflow.utils.llava_conversation_lib.Conversation method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.disable_group_texts">disable_group_texts (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.distributed_inference_num_instances">distributed_inference_num_instances (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments.do_dpo_align">do_dpo_align (lmflow.args.IterativeDPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.do_grad_scaling">do_grad_scaling (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments.do_response_generation">do_response_generation (lmflow.args.IterativeDPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.do_rope_scaling">do_rope_scaling (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.do_sample">do_sample (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments.do_scoring">do_scoring (lmflow.args.IterativeDPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.do_train">do_train (lmflow.models.hf_model_mixin.HFModelMixin attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.double_quant">double_quant (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.dpo_loss">dpo_loss() (lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner">DPOAligner (class in lmflow.pipeline.dpo_aligner)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments">DPOAlignerArguments (class in lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner">DPOv2Aligner (class in lmflow.pipeline.dpov2_aligner)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments">DPOv2AlignerArguments (class in lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer">DPOv2Trainer (class in lmflow.pipeline.utils.dpov2_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.draft_config">draft_config (lmflow.pipeline.inferencer.SpeculativeInferencer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.draft_model_args">draft_model_args (lmflow.pipeline.inferencer.SpeculativeInferencer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.drop_instances">drop_instances() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.drop_instances">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.ds_config">ds_config (lmflow.models.hf_model_mixin.HFModelMixin attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.dtype">dtype (lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower property)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/dummy/index.html#lmflow.optim.dummy.Dummy">Dummy (class in lmflow.optim.dummy)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.DUMMY">DUMMY (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.dummy_feature">dummy_feature (lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower property)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="E">E</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE">EMPTY_NO_SPECIAL_TOKENS_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE">(in module lmflow.utils.conversation_template.base)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.EMPTY_TEMPLATE">EMPTY_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.EMPTY_TEMPLATE">(in module lmflow.utils.conversation_template.base)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.EmptyFormatter">EmptyFormatter (class in lmflow.utils.conversation_template.base)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.enable_decode_inference_result">enable_decode_inference_result (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.enable_distributed_inference">enable_distributed_inference (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.encode">encode() (lmflow.models.hf_decoder_model.HFDecoderModel method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.encode">(lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.encode_conversation">encode_conversation() (lmflow.utils.conversation_template.base.ConversationTemplate method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplateForTool.encode_conversation">(lmflow.utils.conversation_template.base.ConversationTemplateForTool method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.encode_conversation">(lmflow.utils.conversation_template.ConversationTemplate method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplateForTool.encode_conversation">(lmflow.utils.conversation_template.ConversationTemplateForTool method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/gemma/index.html#lmflow.utils.conversation_template.gemma.GemmaConversationTemplate.encode_conversation">(lmflow.utils.conversation_template.gemma.GemmaConversationTemplate method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/hymba/index.html#lmflow.utils.conversation_template.hymba.HymbaConversationTemplate.encode_conversation">(lmflow.utils.conversation_template.hymba.HymbaConversationTemplate method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.encode_images">encode_images() (lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower method)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/models/encoder_decoder_model/index.html#lmflow.models.encoder_decoder_model.EncoderDecoderModel">EncoderDecoderModel (class in lmflow.models.encoder_decoder_model)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.eos_padding">eos_padding (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.eos_token_id">eos_token_id (lmflow.pipeline.vllm_inferencer.InferencerWithOffloading attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adamw_schedule_free/index.html#lmflow.optim.adamw_schedule_free.AdamWScheduleFree.eval">eval() (lmflow.optim.adamw_schedule_free.AdamWScheduleFree method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/sgd_schedule_free/index.html#lmflow.optim.sgd_schedule_free.SGDScheduleFree.eval">(lmflow.optim.sgd_schedule_free.SGDScheduleFree method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner.eval_dataset">eval_dataset (lmflow.pipeline.dpo_aligner.DPOAligner attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.eval_dataset">(lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.eval_dataset_path">eval_dataset_path (lmflow.args.DPOAlignerArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.eval_dataset_path">(lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.eval_steps">eval_steps (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.evaluate">evaluate() (lmflow.pipeline.evaluator.Evaluator method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.evaluate">(lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.evaluate_block_size">evaluate_block_size (lmflow.args.EvaluatorArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.evaluation_loop">evaluation_loop() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator">Evaluator (class in lmflow.pipeline.evaluator)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.evaluator_args">evaluator_args (lmflow.pipeline.evaluator.Evaluator attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments">EvaluatorArguments (class in lmflow.args)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="F">F</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.feature_select">feature_select() (lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.finetune_part">finetune_part (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner">Finetuner (class in lmflow.pipeline.finetuner)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.finetuner_args">finetuner_args (lmflow.pipeline.finetuner.Finetuner attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments">FinetunerArguments (class in lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.fixed_decay">fixed_decay (lmflow.optim.adabelief.AdaBelief attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_func">flash_attn_func (in module lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_kvpacked_func">flash_attn_kvpacked_func (in module lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.flash_attn_qkvpacked_func">flash_attn_qkvpacked_func (in module lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc">FlashAttnFunc (class in lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc">FlashAttnKVPackedFunc (class in lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc">FlashAttnQKVPackedFunc (class in lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.flatten_list">flatten_list() (lmflow.pipeline.rm_inferencer.RewardModelInferencer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.FLOAT_ONLY_DATASET_DESCRIPTION">FLOAT_ONLY_DATASET_DESCRIPTION (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.floating_point_ops">floating_point_ops() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.force_system">force_system (lmflow.utils.conversation_template.base.ConversationTemplate attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.force_system">(lmflow.utils.conversation_template.ConversationTemplate attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.EmptyFormatter.format">format() (lmflow.utils.conversation_template.base.EmptyFormatter method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.Formatter.format">(lmflow.utils.conversation_template.base.Formatter method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ListFormatter.format">(lmflow.utils.conversation_template.base.ListFormatter method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.StringFormatter.format">(lmflow.utils.conversation_template.base.StringFormatter method)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.Formatter">Formatter (class in lmflow.utils.conversation_template.base)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.html#lmflow.utils.flash_attention.bloom_flash_attention.forward">forward() (in module lmflow.utils.flash_attention.bloom_flash_attention)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.html#lmflow.utils.flash_attention.gpt2_flash_attention.forward">(in module lmflow.utils.flash_attention.gpt2_flash_attention)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html#lmflow.utils.flash_attention.gpt_neo_flash_attention.forward">(in module lmflow.utils.flash_attention.gpt_neo_flash_attention)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.html#lmflow.utils.flash_attention.llama_flash_attention.forward">(in module lmflow.utils.flash_attention.llama_flash_attention)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.forward">(lmflow.models.vision2seq_model.CustomAutoVision2SeqModel method)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.forward">(lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc.forward">(lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc static method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc.forward">(lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc static method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc.forward">(lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc static method)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.forward">(lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.from_dict">from_dict() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.from_dict">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.fsdp">fsdp (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.function_formatter">function_formatter (lmflow.utils.conversation_template.base.ConversationTemplate attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.function_formatter">(lmflow.utils.conversation_template.ConversationTemplate attribute)</a>
+</li>
+      </ul></li>
+  </ul></td>
+</tr></table>
+
+<h2 id="G">G</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.GEMMA_TEMPLATE">GEMMA_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/gemma/index.html#lmflow.utils.conversation_template.gemma.GEMMA_TEMPLATE">(in module lmflow.utils.conversation_template.gemma)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/gemma/index.html#lmflow.utils.conversation_template.gemma.GemmaConversationTemplate">GemmaConversationTemplate (class in lmflow.utils.conversation_template.gemma)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.generate">generate() (lmflow.models.vision2seq_model.CustomAutoVision2SeqModel method)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.get_backend">get_backend() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.get_backend">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.get_backend_dataset">get_backend_dataset() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.get_backend_dataset">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_backend_model">get_backend_model() (lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.get_backend_model">(lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.get_backend_model">(lmflow.models.vision2seq_model.CustomAutoVision2SeqModel method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.get_batch_loss_metrics">get_batch_loss_metrics() (lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.get_batch_metrics">get_batch_metrics() (lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.get_data_args">get_data_args() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.get_data_args">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_eval_dataloader">get_eval_dataloader() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.get_fingerprint">get_fingerprint() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.get_fingerprint">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.get_images">get_images() (lmflow.utils.llava_conversation_lib.Conversation method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_max_length">get_max_length() (lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.get_max_length">(lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/models/auto_model/index.html#lmflow.models.auto_model.AutoModel.get_model">get_model() (lmflow.models.auto_model.AutoModel class method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_optimizer_cls_and_kwargs">get_optimizer_cls_and_kwargs() (lmflow.pipeline.utils.raft_trainer.RaftTrainer static method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.get_paired_dataset">get_paired_dataset() (in module lmflow.pipeline.dpo_aligner)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.get_peft_without_qlora">get_peft_without_qlora() (lmflow.models.hf_decoder_model.HFDecoderModel method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/auto_pipeline/index.html#lmflow.pipeline.auto_pipeline.AutoPipeline.get_pipeline">get_pipeline() (lmflow.pipeline.auto_pipeline.AutoPipeline class method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.AutoArguments.get_pipeline_args_class">get_pipeline_args_class() (lmflow.args.AutoArguments method)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.get_prompt">get_prompt() (lmflow.utils.llava_conversation_lib.Conversation method)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/versioning/index.html#lmflow.utils.versioning.get_python_version">get_python_version() (in module lmflow.utils.versioning)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_test_dataloader">get_test_dataloader() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_tokenizer">get_tokenizer() (lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.get_tokenizer">(lmflow.models.hf_model_mixin.HFModelMixin method)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.get_tokenizer">(lmflow.models.vision2seq_model.CustomAutoVision2SeqModel method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_train_dataloader">get_train_dataloader() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.get_type">get_type() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.get_type">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.gradient_accumulation_steps">gradient_accumulation_steps (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.gradient_checkpointing">gradient_checkpointing (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.gradient_checkpointing_use_reentrant">gradient_checkpointing_use_reentrant (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.group_text">group_text() (lmflow.pipeline.finetuner.Finetuner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.group_texts_batch_size">group_texts_batch_size (lmflow.args.DatasetArguments attribute)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="H">H</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.Formatter.has_placeholder">has_placeholder() (lmflow.utils.conversation_template.base.Formatter method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.hf_auto_model">hf_auto_model (lmflow.models.hf_model_mixin.HFModelMixin attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HF_AUTOMODEL_MAPPING">HF_AUTOMODEL_MAPPING (in module lmflow.models.hf_model_mixin)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HF_AUTOMODEL_TYPE">HF_AUTOMODEL_TYPE (in module lmflow.models.hf_model_mixin)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.hf_dataset_sanity_check">hf_dataset_sanity_check() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.hf_dataset_sanity_check">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.hf_model_config">hf_model_config (lmflow.models.hf_model_mixin.HFModelMixin attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel">HFDecoderModel (class in lmflow.models.hf_decoder_model)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel">HFEncoderDecoderModel (class in lmflow.models.hf_encoder_decoder_model)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin">HFModelMixin (class in lmflow.models.hf_model_mixin)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel">HFTextRegressionModel (class in lmflow.models.hf_text_regression_model)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.hidden_size">hidden_size (lmflow.models.vision2seq_model.CustomAutoVision2SeqModel attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.hidden_size">(lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower property)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.hp_name">hp_name (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.hp_search_backend">hp_search_backend (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.HYMBA_TEMPLATE">HYMBA_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/hymba/index.html#lmflow.utils.conversation_template.hymba.HYMBA_TEMPLATE">(in module lmflow.utils.conversation_template.hymba)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/hymba/index.html#lmflow.utils.conversation_template.hymba.HymbaConversationTemplate">HymbaConversationTemplate (class in lmflow.utils.conversation_template.hymba)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.hyperparameter_search">hyperparameter_search() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="I">I</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.ignore_bias_buffers">ignore_bias_buffers (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.IGNORE_INDEX">IGNORE_INDEX (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.MultiModalDatasetArguments.image_aspect_ratio">image_aspect_ratio (lmflow.args.MultiModalDatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.image_encoder_name_or_path">image_encoder_name_or_path (lmflow.args.VisModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.MultiModalDatasetArguments.image_folder">image_folder (lmflow.args.MultiModalDatasetArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.image_folder">(lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.IMAGE_TOKEN_INDEX">IMAGE_TOKEN_INDEX (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner.INF">INF (lmflow.pipeline.raft_aligner.RaftAligner attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.inference">inference() (lmflow.models.hf_decoder_model.HFDecoderModel method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.inference">(lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel method)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.inference">(lmflow.models.hf_text_regression_model.HFTextRegressionModel method)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/text_regression_model/index.html#lmflow.models.text_regression_model.TextRegressionModel.inference">(lmflow.models.text_regression_model.TextRegressionModel method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.inference">(lmflow.pipeline.inferencer.Inferencer method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.inference">(lmflow.pipeline.inferencer.SpeculativeInferencer method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.ToolInferencer.inference">(lmflow.pipeline.inferencer.ToolInferencer method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.inference">(lmflow.pipeline.rm_inferencer.RewardModelInferencer method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.inference">(lmflow.pipeline.vllm_inferencer.InferencerWithOffloading method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer.inference">(lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer.inference">(lmflow.pipeline.vllm_inferencer.VLLMInferencer method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.inference_batch_size">inference_batch_size (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.inference_batch_size_per_device">inference_batch_size_per_device (lmflow.args.EvaluatorArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.inference_batch_size_per_device">(lmflow.args.RaftAlignerArguments attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/text_regression_model/index.html#lmflow.models.text_regression_model.TextRegressionModel.inference_func">inference_func (lmflow.models.text_regression_model.TextRegressionModel attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer">Inferencer (class in lmflow.pipeline.inferencer)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.inferencer_args">inferencer_args (lmflow.pipeline.inferencer.Inferencer attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.inferencer_args">(lmflow.pipeline.rm_inferencer.RewardModelInferencer attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.inferencer_args">(lmflow.pipeline.vllm_inferencer.InferencerWithOffloading attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer.inferencer_file_path">inferencer_file_path (lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments">InferencerArguments (class in lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading">InferencerWithOffloading (class in lmflow.pipeline.vllm_inferencer)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.init_git_repo">init_git_repo() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#lmflow.utils.flash_attention.triton_flash_attention.init_to_zero">init_to_zero() (in module lmflow.utils.flash_attention.triton_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.IterativeAlignerArguments.initial_iter_idx">initial_iter_idx (lmflow.args.IterativeAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.RewardModelInferenceResultWithInput.input">input (lmflow.utils.data_utils.RewardModelInferenceResultWithInput attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.VLLMInferenceResultWithInput.input">(lmflow.utils.data_utils.VLLMInferenceResultWithInput attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.INSTANCE_FIELDS_MAP">INSTANCE_FIELDS_MAP (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/index.html#lmflow.internal_version">internal_version (in module lmflow)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.INTERNLM2_TEMPLATE">INTERNLM2_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/internlm/index.html#lmflow.utils.conversation_template.internlm.INTERNLM2_TEMPLATE">(in module lmflow.utils.conversation_template.internlm)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.ipex_optimize_model">ipex_optimize_model() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.is_custom_dataset">is_custom_dataset (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.is_encoder_decoder">is_encoder_decoder (lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/versioning/index.html#lmflow.utils.versioning.is_flash_attn_available">is_flash_attn_available() (in module lmflow.utils.versioning)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/versioning/index.html#lmflow.utils.versioning.is_flask_available">is_flask_available() (in module lmflow.utils.versioning)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/versioning/index.html#lmflow.utils.versioning.is_gradio_available">is_gradio_available() (in module lmflow.utils.versioning)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_in_train">is_in_train (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.is_loaded">is_loaded (lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_local_process_zero">is_local_process_zero() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.MultiModalDatasetArguments.is_multimodal">is_multimodal (lmflow.args.MultiModalDatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.is_multimodal_available">is_multimodal_available() (in module lmflow.datasets)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/versioning/index.html#lmflow.utils.versioning.is_multimodal_available">(in module lmflow.utils.versioning)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/versioning/index.html#lmflow.utils.versioning.is_package_version_at_least">is_package_version_at_least() (in module lmflow.utils.versioning)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/versioning/index.html#lmflow.utils.versioning.is_ray_available">is_ray_available() (in module lmflow.utils.versioning)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.IS_SAGEMAKER_MP_POST_1_10">IS_SAGEMAKER_MP_POST_1_10 (in module lmflow.pipeline.utils.raft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.is_torch_greater_or_equal_than_1_10">is_torch_greater_or_equal_than_1_10 (in module lmflow.pipeline.utils.raft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.is_torch_less_than_1_11">is_torch_less_than_1_11 (in module lmflow.pipeline.utils.raft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/versioning/index.html#lmflow.utils.versioning.is_trl_available">is_trl_available() (in module lmflow.utils.versioning)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/versioning/index.html#lmflow.utils.versioning.is_vllm_available">is_vllm_available() (in module lmflow.utils.versioning)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_world_process_zero">is_world_process_zero() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.IterativeAlignerArguments">IterativeAlignerArguments (class in lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner">IterativeDPOAligner (class in lmflow.pipeline.iterative_dpo_aligner)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments">IterativeDPOAlignerArguments (class in lmflow.args)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="K">K</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.keep_linebreaks">keep_linebreaks (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.KEY_INSTANCES">KEY_INSTANCES (in module lmflow.datasets.dataset)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.KEY_SCORE">KEY_SCORE (in module lmflow.datasets.dataset)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.KEY_TYPE">KEY_TYPE (in module lmflow.datasets.dataset)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="L">L</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.label_names">label_names (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.label_pad_token_id">label_pad_token_id (lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/lamb/index.html#lmflow.optim.lamb.Lamb">Lamb (class in lmflow.optim.lamb)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.LAMB">LAMB (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.language_model">language_model (lmflow.models.vision2seq_model.CustomAutoVision2SeqModel attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.language_model_from_pretrained">language_model_from_pretrained() (lmflow.models.vision2seq_model.CustomAutoVision2SeqModel method)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/lars/index.html#lmflow.optim.lars.LARS">LARS (class in lmflow.optim.lars)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.LARS">(lmflow.args.OptimizerNames attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.last_checkpoint">last_checkpoint (lmflow.pipeline.finetuner.Finetuner attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.learning_rate">learning_rate (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.len_penalty">len_penalty (lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.length_penalty">length_penalty (lmflow.args.DPOv2AlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.lisa_activated_layers">lisa_activated_layers (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.lisa_interval_steps">lisa_interval_steps (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.lisa_layers_attribute">lisa_layers_attribute (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ListFormatter">ListFormatter (class in lmflow.utils.conversation_template.base)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.LLAMA2_TEMPLATE">LLAMA2_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE">(in module lmflow.utils.conversation_template.llama)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE_FOR_TOOL">LLAMA2_TEMPLATE_FOR_TOOL (in module lmflow.utils.conversation_template.llama)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.llama.Llama2ConversationTemplate">Llama2ConversationTemplate (class in lmflow.utils.conversation_template.llama)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool">Llama2ConversationTemplateForTool (class in lmflow.utils.conversation_template.llama)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.LLAMA3_TEMPLATE">LLAMA3_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE">(in module lmflow.utils.conversation_template.llama)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.LLAMA3_TEMPLATE_FOR_TOOL">LLAMA3_TEMPLATE_FOR_TOOL (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE_FOR_TOOL">(in module lmflow.utils.conversation_template.llama)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.SeparatorStyle.LLAMA_2">LLAMA_2 (lmflow.utils.llava_conversation_lib.SeparatorStyle attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.llava_loading">llava_loading (lmflow.args.VisModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.llava_pretrain_model_path">llava_pretrain_model_path (lmflow.args.VisModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.llm_model_name_or_path">llm_model_name_or_path (lmflow.args.VisModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.BenchmarkingArguments.lm_evaluation_metric">lm_evaluation_metric (lmflow.args.BenchmarkingArguments attribute)</a>
+</li>
+      <li>
+    lmflow
+
+      <ul>
+        <li><a href="autoapi/lmflow/index.html#module-lmflow">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.args
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#module-lmflow.args">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.datasets
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/index.html#module-lmflow.datasets">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.datasets.dataset
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#module-lmflow.datasets.dataset">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.datasets.multi_modal_dataset
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#module-lmflow.datasets.multi_modal_dataset">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/index.html#module-lmflow.models">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.auto_model
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/auto_model/index.html#module-lmflow.models.auto_model">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.base_model
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/base_model/index.html#module-lmflow.models.base_model">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.decoder_model
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/decoder_model/index.html#module-lmflow.models.decoder_model">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.encoder_decoder_model
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/encoder_decoder_model/index.html#module-lmflow.models.encoder_decoder_model">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.hf_decoder_model
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#module-lmflow.models.hf_decoder_model">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.hf_encoder_decoder_model
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#module-lmflow.models.hf_encoder_decoder_model">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.hf_model_mixin
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#module-lmflow.models.hf_model_mixin">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.hf_text_regression_model
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_text_regression_model/index.html#module-lmflow.models.hf_text_regression_model">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.interfaces
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/interfaces/index.html#module-lmflow.models.interfaces">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.interfaces.tunable
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/interfaces/tunable/index.html#module-lmflow.models.interfaces.tunable">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.regression_model
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/regression_model/index.html#module-lmflow.models.regression_model">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.text_regression_model
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/text_regression_model/index.html#module-lmflow.models.text_regression_model">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.vision2seq_model
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#module-lmflow.models.vision2seq_model">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.vision_encoder
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/vision_encoder/index.html#module-lmflow.models.vision_encoder">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.models.vision_encoder.clip_encoder
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#module-lmflow.models.vision_encoder.clip_encoder">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/index.html#module-lmflow.optim">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.adabelief
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/adabelief/index.html#module-lmflow.optim.adabelief">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.adabound
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/adabound/index.html#module-lmflow.optim.adabound">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.adadelta
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/adadelta/index.html#module-lmflow.optim.adadelta">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.adagrad
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/adagrad/index.html#module-lmflow.optim.adagrad">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.adam
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/adam/index.html#module-lmflow.optim.adam">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.adamax
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/adamax/index.html#module-lmflow.optim.adamax">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.adamp
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/adamp/index.html#module-lmflow.optim.adamp">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.adamw_schedule_free
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/adamw_schedule_free/index.html#module-lmflow.optim.adamw_schedule_free">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.adan
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/adan/index.html#module-lmflow.optim.adan">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.dummy
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/dummy/index.html#module-lmflow.optim.dummy">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.lamb
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/lamb/index.html#module-lmflow.optim.lamb">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.lars
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/lars/index.html#module-lmflow.optim.lars">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.nadam
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/nadam/index.html#module-lmflow.optim.nadam">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.novograd
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/novograd/index.html#module-lmflow.optim.novograd">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.optimizers
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/optimizers/index.html#module-lmflow.optim.optimizers">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.radam
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/radam/index.html#module-lmflow.optim.radam">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.sgd_schedule_free
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/sgd_schedule_free/index.html#module-lmflow.optim.sgd_schedule_free">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.sgdp
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/sgdp/index.html#module-lmflow.optim.sgdp">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.sophia
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/sophia/index.html#module-lmflow.optim.sophia">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.optim.yogi
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/yogi/index.html#module-lmflow.optim.yogi">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/index.html#module-lmflow.pipeline">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.auto_pipeline
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/auto_pipeline/index.html#module-lmflow.pipeline.auto_pipeline">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.base_aligner
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/base_aligner/index.html#module-lmflow.pipeline.base_aligner">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.base_pipeline
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/base_pipeline/index.html#module-lmflow.pipeline.base_pipeline">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.base_tuner
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/base_tuner/index.html#module-lmflow.pipeline.base_tuner">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.dpo_aligner
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/dpo_aligner/index.html#module-lmflow.pipeline.dpo_aligner">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.dpov2_aligner
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#module-lmflow.pipeline.dpov2_aligner">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.evaluator
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#module-lmflow.pipeline.evaluator">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.finetuner
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/finetuner/index.html#module-lmflow.pipeline.finetuner">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.inferencer
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#module-lmflow.pipeline.inferencer">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.iterative_dpo_aligner
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#module-lmflow.pipeline.iterative_dpo_aligner">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.raft_aligner
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#module-lmflow.pipeline.raft_aligner">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.rm_inferencer
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#module-lmflow.pipeline.rm_inferencer">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.rm_tuner
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/rm_tuner/index.html#module-lmflow.pipeline.rm_tuner">module</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li>
+    lmflow.pipeline.utils
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/index.html#module-lmflow.pipeline.utils">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.utils.dpov2_dataprocessor
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#module-lmflow.pipeline.utils.dpov2_dataprocessor">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.utils.dpov2_trainer
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#module-lmflow.pipeline.utils.dpov2_trainer">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.utils.memory_safe_dpov2_align
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.html#module-lmflow.pipeline.utils.memory_safe_dpov2_align">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.utils.memory_safe_vllm_inference
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.html#module-lmflow.pipeline.utils.memory_safe_vllm_inference">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.utils.peft_trainer
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/peft_trainer/index.html#module-lmflow.pipeline.utils.peft_trainer">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.utils.raft_trainer
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#module-lmflow.pipeline.utils.raft_trainer">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.utils.rm_dataprocessor
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#module-lmflow.pipeline.utils.rm_dataprocessor">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.utils.rm_trainer
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/rm_trainer/index.html#module-lmflow.pipeline.utils.rm_trainer">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.pipeline.vllm_inferencer
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#module-lmflow.pipeline.vllm_inferencer">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.tokenization
+
+      <ul>
+        <li><a href="autoapi/lmflow/tokenization/index.html#module-lmflow.tokenization">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.tokenization.hf_decoder_model
+
+      <ul>
+        <li><a href="autoapi/lmflow/tokenization/hf_decoder_model/index.html#module-lmflow.tokenization.hf_decoder_model">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.tokenization.hf_text_regression_model
+
+      <ul>
+        <li><a href="autoapi/lmflow/tokenization/hf_text_regression_model/index.html#module-lmflow.tokenization.hf_text_regression_model">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/index.html#module-lmflow.utils">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.common
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/common/index.html#module-lmflow.utils.common">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.constants
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/constants/index.html#module-lmflow.utils.constants">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.conversation_template
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#module-lmflow.utils.conversation_template">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.conversation_template.base
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#module-lmflow.utils.conversation_template.base">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.conversation_template.chatglm
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/chatglm/index.html#module-lmflow.utils.conversation_template.chatglm">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.conversation_template.chatml
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/chatml/index.html#module-lmflow.utils.conversation_template.chatml">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.conversation_template.deepseek
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/deepseek/index.html#module-lmflow.utils.conversation_template.deepseek">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.conversation_template.gemma
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/gemma/index.html#module-lmflow.utils.conversation_template.gemma">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.conversation_template.hymba
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/hymba/index.html#module-lmflow.utils.conversation_template.hymba">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.conversation_template.internlm
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/internlm/index.html#module-lmflow.utils.conversation_template.internlm">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.conversation_template.llama
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/llama/index.html#module-lmflow.utils.conversation_template.llama">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.conversation_template.phi
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/phi/index.html#module-lmflow.utils.conversation_template.phi">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.conversation_template.qwen
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/qwen/index.html#module-lmflow.utils.conversation_template.qwen">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.conversation_template.yi
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/yi/index.html#module-lmflow.utils.conversation_template.yi">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.conversation_template.zephyr
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/zephyr/index.html#module-lmflow.utils.conversation_template.zephyr">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.data_utils
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/data_utils/index.html#module-lmflow.utils.data_utils">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.flash_attention
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/flash_attention/index.html#module-lmflow.utils.flash_attention">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.flash_attention.bloom_flash_attention
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.html#module-lmflow.utils.flash_attention.bloom_flash_attention">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.flash_attention.gpt2_flash_attention
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.html#module-lmflow.utils.flash_attention.gpt2_flash_attention">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.flash_attention.gpt_neo_flash_attention
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html#module-lmflow.utils.flash_attention.gpt_neo_flash_attention">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.flash_attention.llama_flash_attention
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.html#module-lmflow.utils.flash_attention.llama_flash_attention">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.flash_attention.triton_flash_attention
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#module-lmflow.utils.flash_attention.triton_flash_attention">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.llava_conversation_lib
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#module-lmflow.utils.llava_conversation_lib">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.model
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/model/index.html#module-lmflow.utils.model">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.multimodal
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/multimodal/index.html#module-lmflow.utils.multimodal">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.position_interpolation
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/position_interpolation/index.html#module-lmflow.utils.position_interpolation">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#module-lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.utils.versioning
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/versioning/index.html#module-lmflow.utils.versioning">module</a>
+</li>
+      </ul></li>
+      <li>
+    lmflow.version
+
+      <ul>
+        <li><a href="autoapi/lmflow/version/index.html#module-lmflow.version">module</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.LMFLOW_LORA_TARGET_MODULES_MAPPING">LMFLOW_LORA_TARGET_MODULES_MAPPING (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.load_data">load_data() (in module lmflow.utils.data_utils)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.load_in_4bit">load_in_4bit (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.load_inference_results">load_inference_results() (lmflow.pipeline.vllm_inferencer.InferencerWithOffloading method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer.load_inference_results">(lmflow.pipeline.vllm_inferencer.VLLMInferencer method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/multimodal/index.html#lmflow.utils.multimodal.load_llava_pretrain_model">load_llava_pretrain_model() (in module lmflow.utils.multimodal)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.load_model">load_model() (lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.load_prompt_cache">load_prompt_cache() (lmflow.models.vision2seq_model.CustomAutoVision2SeqModel method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.local_rank">local_rank (lmflow.args.DPOAlignerArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.local_rank">(lmflow.args.EvaluatorArguments attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.local_rank">(lmflow.args.InferencerArguments attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.local_rank">(lmflow.pipeline.evaluator.Evaluator attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.local_rank">(lmflow.pipeline.inferencer.Inferencer attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.local_rank">(lmflow.pipeline.rm_inferencer.RewardModelInferencer attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.log">log() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.log_freq">log_freq (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.LOGDIR">LOGDIR (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.logger">logger (in module lmflow.args)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.logger">(in module lmflow.datasets.dataset)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.logger">(in module lmflow.models.hf_decoder_model)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.logger">(in module lmflow.models.hf_encoder_decoder_model)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.logger">(in module lmflow.models.hf_model_mixin)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.logger">(in module lmflow.models.hf_text_regression_model)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.logger">(in module lmflow.pipeline.dpov2_aligner)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.logger">(in module lmflow.pipeline.finetuner)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.logger">(in module lmflow.pipeline.inferencer)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.logger">(in module lmflow.pipeline.iterative_dpo_aligner)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.logger">(in module lmflow.pipeline.raft_aligner)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.logger">(in module lmflow.pipeline.rm_inferencer)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/rm_tuner/index.html#lmflow.pipeline.rm_tuner.logger">(in module lmflow.pipeline.rm_tuner)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.logger">(in module lmflow.pipeline.utils.dpov2_dataprocessor)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.logger">(in module lmflow.pipeline.utils.dpov2_trainer)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.html#lmflow.pipeline.utils.memory_safe_dpov2_align.logger">(in module lmflow.pipeline.utils.memory_safe_dpov2_align)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.html#lmflow.pipeline.utils.memory_safe_vllm_inference.logger">(in module lmflow.pipeline.utils.memory_safe_vllm_inference)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.logger">(in module lmflow.pipeline.utils.raft_trainer)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.logger">(in module lmflow.pipeline.utils.rm_dataprocessor)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.logger">(in module lmflow.pipeline.vllm_inferencer)</a>
+</li>
+        <li><a href="autoapi/lmflow/tokenization/hf_decoder_model/index.html#lmflow.tokenization.hf_decoder_model.logger">(in module lmflow.tokenization.hf_decoder_model)</a>
+</li>
+        <li><a href="autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.logger">(in module lmflow.tokenization.hf_text_regression_model)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/common/index.html#lmflow.utils.common.logger">(in module lmflow.utils.common)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.logger">(in module lmflow.utils.conversation_template.base)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/gemma/index.html#lmflow.utils.conversation_template.gemma.logger">(in module lmflow.utils.conversation_template.gemma)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/llama/index.html#lmflow.utils.conversation_template.llama.logger">(in module lmflow.utils.conversation_template.llama)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/zephyr/index.html#lmflow.utils.conversation_template.zephyr.logger">(in module lmflow.utils.conversation_template.zephyr)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/model/index.html#lmflow.utils.model.logger">(in module lmflow.utils.model)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/versioning/index.html#lmflow.utils.versioning.logger">(in module lmflow.utils.versioning)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.logging_steps">logging_steps (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.lora_alpha">lora_alpha (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.lora_dropout">lora_dropout (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.lora_model_path">lora_model_path (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.lora_r">lora_r (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.lora_target_modules">lora_target_modules (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.LORA_TARGET_MODULES_MAPPING">LORA_TARGET_MODULES_MAPPING (in module lmflow.models.hf_model_mixin)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.loss_type">loss_type (lmflow.args.DPOv2AlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.low_resource">low_resource (lmflow.args.VisModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.lr_scheduler_type">lr_scheduler_type (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="M">M</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.html#lmflow.pipeline.utils.memory_safe_dpov2_align.main">main() (in module lmflow.pipeline.utils.memory_safe_dpov2_align)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.html#lmflow.pipeline.utils.memory_safe_vllm_inference.main">(in module lmflow.pipeline.utils.memory_safe_vllm_inference)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/common/index.html#lmflow.utils.common.make_shell_args_from_dataclass">make_shell_args_from_dataclass() (in module lmflow.utils.common)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.map">map() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.map">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.margin_scale">margin_scale (lmflow.args.DPOv2AlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent.mask">mask (lmflow.utils.conversation_template.base.TemplateComponent attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.mask_prompt">mask_prompt (lmflow.args.DPOv2AlignerArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.mask_prompt">(lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.max_eval_samples">max_eval_samples (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.max_length">max_length (lmflow.args.DPOAlignerArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.max_length">(lmflow.args.DPOv2AlignerArguments attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_length">(lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.max_length">(lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.max_new_tokens">max_new_tokens (lmflow.args.EvaluatorArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.max_new_tokens">(lmflow.args.InferencerArguments attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.max_prompt_length">max_prompt_length (lmflow.args.DPOAlignerArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.max_prompt_length">(lmflow.args.DPOv2AlignerArguments attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_prompt_length">(lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.max_seq_len_cached">max_seq_len_cached (lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.max_steps">max_steps (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_target_length">max_target_length (lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.max_train_samples">max_train_samples (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE">MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE">MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG">MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner">MemorySafeDPOv2Aligner (class in lmflow.pipeline.dpov2_aligner)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer">MemorySafeVLLMInferencer (class in lmflow.pipeline.vllm_inferencer)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.merge_lora_weights">merge_lora_weights() (lmflow.models.hf_decoder_model.HFDecoderModel method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.merge_lora_weights">(lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.messages">messages (lmflow.utils.llava_conversation_lib.Conversation attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.metric">metric (lmflow.args.EvaluatorArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.mixed_precision">mixed_precision (lmflow.args.EvaluatorArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.mixed_precision">(lmflow.args.InferencerArguments attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.ToolInferencer.model">model (lmflow.pipeline.inferencer.ToolInferencer attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.model">(lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.model">(lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.model_args">model_args (lmflow.models.hf_model_mixin.HFModelMixin attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner.model_args">(lmflow.pipeline.dpo_aligner.DPOAligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.model_args">(lmflow.pipeline.dpov2_aligner.DPOv2Aligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.model_args">(lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.model_args">(lmflow.pipeline.evaluator.Evaluator attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.model_args">(lmflow.pipeline.finetuner.Finetuner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.model_args">(lmflow.pipeline.inferencer.Inferencer attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.model_args">(lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner.model_args">(lmflow.pipeline.raft_aligner.RaftAligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.model_args">(lmflow.pipeline.rm_inferencer.RewardModelInferencer attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.model_args">(lmflow.pipeline.vllm_inferencer.InferencerWithOffloading attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.MODEL_CONFIG_CLASSES">MODEL_CONFIG_CLASSES (in module lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.model_max_length">model_max_length (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.model_name_or_path">model_name_or_path (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.model_revision">model_revision (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.model_type">model_type (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.MODEL_TYPES">MODEL_TYPES (in module lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.model_wrapped">model_wrapped (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments">ModelArguments (class in lmflow.args)</a>
+</li>
+      <li>
+    module
+
+      <ul>
+        <li><a href="autoapi/lmflow/index.html#module-lmflow">lmflow</a>
+</li>
+        <li><a href="autoapi/lmflow/args/index.html#module-lmflow.args">lmflow.args</a>
+</li>
+        <li><a href="autoapi/lmflow/datasets/index.html#module-lmflow.datasets">lmflow.datasets</a>
+</li>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#module-lmflow.datasets.dataset">lmflow.datasets.dataset</a>
+</li>
+        <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#module-lmflow.datasets.multi_modal_dataset">lmflow.datasets.multi_modal_dataset</a>
+</li>
+        <li><a href="autoapi/lmflow/models/index.html#module-lmflow.models">lmflow.models</a>
+</li>
+        <li><a href="autoapi/lmflow/models/auto_model/index.html#module-lmflow.models.auto_model">lmflow.models.auto_model</a>
+</li>
+        <li><a href="autoapi/lmflow/models/base_model/index.html#module-lmflow.models.base_model">lmflow.models.base_model</a>
+</li>
+        <li><a href="autoapi/lmflow/models/decoder_model/index.html#module-lmflow.models.decoder_model">lmflow.models.decoder_model</a>
+</li>
+        <li><a href="autoapi/lmflow/models/encoder_decoder_model/index.html#module-lmflow.models.encoder_decoder_model">lmflow.models.encoder_decoder_model</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#module-lmflow.models.hf_decoder_model">lmflow.models.hf_decoder_model</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#module-lmflow.models.hf_encoder_decoder_model">lmflow.models.hf_encoder_decoder_model</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#module-lmflow.models.hf_model_mixin">lmflow.models.hf_model_mixin</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_text_regression_model/index.html#module-lmflow.models.hf_text_regression_model">lmflow.models.hf_text_regression_model</a>
+</li>
+        <li><a href="autoapi/lmflow/models/interfaces/index.html#module-lmflow.models.interfaces">lmflow.models.interfaces</a>
+</li>
+        <li><a href="autoapi/lmflow/models/interfaces/tunable/index.html#module-lmflow.models.interfaces.tunable">lmflow.models.interfaces.tunable</a>
+</li>
+        <li><a href="autoapi/lmflow/models/regression_model/index.html#module-lmflow.models.regression_model">lmflow.models.regression_model</a>
+</li>
+        <li><a href="autoapi/lmflow/models/text_regression_model/index.html#module-lmflow.models.text_regression_model">lmflow.models.text_regression_model</a>
+</li>
+        <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#module-lmflow.models.vision2seq_model">lmflow.models.vision2seq_model</a>
+</li>
+        <li><a href="autoapi/lmflow/models/vision_encoder/index.html#module-lmflow.models.vision_encoder">lmflow.models.vision_encoder</a>
+</li>
+        <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#module-lmflow.models.vision_encoder.clip_encoder">lmflow.models.vision_encoder.clip_encoder</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/index.html#module-lmflow.optim">lmflow.optim</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adabelief/index.html#module-lmflow.optim.adabelief">lmflow.optim.adabelief</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adabound/index.html#module-lmflow.optim.adabound">lmflow.optim.adabound</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adadelta/index.html#module-lmflow.optim.adadelta">lmflow.optim.adadelta</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adagrad/index.html#module-lmflow.optim.adagrad">lmflow.optim.adagrad</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adam/index.html#module-lmflow.optim.adam">lmflow.optim.adam</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adamax/index.html#module-lmflow.optim.adamax">lmflow.optim.adamax</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adamp/index.html#module-lmflow.optim.adamp">lmflow.optim.adamp</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adamw_schedule_free/index.html#module-lmflow.optim.adamw_schedule_free">lmflow.optim.adamw_schedule_free</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adan/index.html#module-lmflow.optim.adan">lmflow.optim.adan</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/dummy/index.html#module-lmflow.optim.dummy">lmflow.optim.dummy</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/lamb/index.html#module-lmflow.optim.lamb">lmflow.optim.lamb</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/lars/index.html#module-lmflow.optim.lars">lmflow.optim.lars</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/nadam/index.html#module-lmflow.optim.nadam">lmflow.optim.nadam</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/novograd/index.html#module-lmflow.optim.novograd">lmflow.optim.novograd</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/optimizers/index.html#module-lmflow.optim.optimizers">lmflow.optim.optimizers</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/radam/index.html#module-lmflow.optim.radam">lmflow.optim.radam</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/sgd_schedule_free/index.html#module-lmflow.optim.sgd_schedule_free">lmflow.optim.sgd_schedule_free</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/sgdp/index.html#module-lmflow.optim.sgdp">lmflow.optim.sgdp</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/sophia/index.html#module-lmflow.optim.sophia">lmflow.optim.sophia</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/yogi/index.html#module-lmflow.optim.yogi">lmflow.optim.yogi</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/index.html#module-lmflow.pipeline">lmflow.pipeline</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/auto_pipeline/index.html#module-lmflow.pipeline.auto_pipeline">lmflow.pipeline.auto_pipeline</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/base_aligner/index.html#module-lmflow.pipeline.base_aligner">lmflow.pipeline.base_aligner</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/base_pipeline/index.html#module-lmflow.pipeline.base_pipeline">lmflow.pipeline.base_pipeline</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/base_tuner/index.html#module-lmflow.pipeline.base_tuner">lmflow.pipeline.base_tuner</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/dpo_aligner/index.html#module-lmflow.pipeline.dpo_aligner">lmflow.pipeline.dpo_aligner</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#module-lmflow.pipeline.dpov2_aligner">lmflow.pipeline.dpov2_aligner</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#module-lmflow.pipeline.evaluator">lmflow.pipeline.evaluator</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/finetuner/index.html#module-lmflow.pipeline.finetuner">lmflow.pipeline.finetuner</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#module-lmflow.pipeline.inferencer">lmflow.pipeline.inferencer</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#module-lmflow.pipeline.iterative_dpo_aligner">lmflow.pipeline.iterative_dpo_aligner</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#module-lmflow.pipeline.raft_aligner">lmflow.pipeline.raft_aligner</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#module-lmflow.pipeline.rm_inferencer">lmflow.pipeline.rm_inferencer</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/rm_tuner/index.html#module-lmflow.pipeline.rm_tuner">lmflow.pipeline.rm_tuner</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/index.html#module-lmflow.pipeline.utils">lmflow.pipeline.utils</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#module-lmflow.pipeline.utils.dpov2_dataprocessor">lmflow.pipeline.utils.dpov2_dataprocessor</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#module-lmflow.pipeline.utils.dpov2_trainer">lmflow.pipeline.utils.dpov2_trainer</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.html#module-lmflow.pipeline.utils.memory_safe_dpov2_align">lmflow.pipeline.utils.memory_safe_dpov2_align</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.html#module-lmflow.pipeline.utils.memory_safe_vllm_inference">lmflow.pipeline.utils.memory_safe_vllm_inference</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/peft_trainer/index.html#module-lmflow.pipeline.utils.peft_trainer">lmflow.pipeline.utils.peft_trainer</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#module-lmflow.pipeline.utils.raft_trainer">lmflow.pipeline.utils.raft_trainer</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#module-lmflow.pipeline.utils.rm_dataprocessor">lmflow.pipeline.utils.rm_dataprocessor</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/rm_trainer/index.html#module-lmflow.pipeline.utils.rm_trainer">lmflow.pipeline.utils.rm_trainer</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#module-lmflow.pipeline.vllm_inferencer">lmflow.pipeline.vllm_inferencer</a>
+</li>
+        <li><a href="autoapi/lmflow/tokenization/index.html#module-lmflow.tokenization">lmflow.tokenization</a>
+</li>
+        <li><a href="autoapi/lmflow/tokenization/hf_decoder_model/index.html#module-lmflow.tokenization.hf_decoder_model">lmflow.tokenization.hf_decoder_model</a>
+</li>
+        <li><a href="autoapi/lmflow/tokenization/hf_text_regression_model/index.html#module-lmflow.tokenization.hf_text_regression_model">lmflow.tokenization.hf_text_regression_model</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/index.html#module-lmflow.utils">lmflow.utils</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/common/index.html#module-lmflow.utils.common">lmflow.utils.common</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/constants/index.html#module-lmflow.utils.constants">lmflow.utils.constants</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#module-lmflow.utils.conversation_template">lmflow.utils.conversation_template</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#module-lmflow.utils.conversation_template.base">lmflow.utils.conversation_template.base</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/chatglm/index.html#module-lmflow.utils.conversation_template.chatglm">lmflow.utils.conversation_template.chatglm</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/chatml/index.html#module-lmflow.utils.conversation_template.chatml">lmflow.utils.conversation_template.chatml</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/deepseek/index.html#module-lmflow.utils.conversation_template.deepseek">lmflow.utils.conversation_template.deepseek</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/gemma/index.html#module-lmflow.utils.conversation_template.gemma">lmflow.utils.conversation_template.gemma</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/hymba/index.html#module-lmflow.utils.conversation_template.hymba">lmflow.utils.conversation_template.hymba</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/internlm/index.html#module-lmflow.utils.conversation_template.internlm">lmflow.utils.conversation_template.internlm</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/llama/index.html#module-lmflow.utils.conversation_template.llama">lmflow.utils.conversation_template.llama</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/phi/index.html#module-lmflow.utils.conversation_template.phi">lmflow.utils.conversation_template.phi</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/qwen/index.html#module-lmflow.utils.conversation_template.qwen">lmflow.utils.conversation_template.qwen</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/yi/index.html#module-lmflow.utils.conversation_template.yi">lmflow.utils.conversation_template.yi</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/zephyr/index.html#module-lmflow.utils.conversation_template.zephyr">lmflow.utils.conversation_template.zephyr</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/data_utils/index.html#module-lmflow.utils.data_utils">lmflow.utils.data_utils</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/flash_attention/index.html#module-lmflow.utils.flash_attention">lmflow.utils.flash_attention</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.html#module-lmflow.utils.flash_attention.bloom_flash_attention">lmflow.utils.flash_attention.bloom_flash_attention</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.html#module-lmflow.utils.flash_attention.gpt2_flash_attention">lmflow.utils.flash_attention.gpt2_flash_attention</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html#module-lmflow.utils.flash_attention.gpt_neo_flash_attention">lmflow.utils.flash_attention.gpt_neo_flash_attention</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.html#module-lmflow.utils.flash_attention.llama_flash_attention">lmflow.utils.flash_attention.llama_flash_attention</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#module-lmflow.utils.flash_attention.triton_flash_attention">lmflow.utils.flash_attention.triton_flash_attention</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#module-lmflow.utils.llava_conversation_lib">lmflow.utils.llava_conversation_lib</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/model/index.html#module-lmflow.utils.model">lmflow.utils.model</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/multimodal/index.html#module-lmflow.utils.multimodal">lmflow.utils.multimodal</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/position_interpolation/index.html#module-lmflow.utils.position_interpolation">lmflow.utils.position_interpolation</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#module-lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/versioning/index.html#module-lmflow.utils.versioning">lmflow.utils.versioning</a>
+</li>
+        <li><a href="autoapi/lmflow/version/index.html#module-lmflow.version">lmflow.version</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.SeparatorStyle.MPT">MPT (lmflow.utils.llava_conversation_lib.SeparatorStyle attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.MultiModalDatasetArguments">MultiModalDatasetArguments (class in lmflow.args)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="N">N</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/optim/nadam/index.html#lmflow.optim.nadam.NAdam">NAdam (class in lmflow.optim.nadam)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.NADAM">NADAM (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/novograd/index.html#lmflow.optim.novograd.NovoGrad">NovoGrad (class in lmflow.optim.novograd)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.NOVOGRAD">NOVOGRAD (lmflow.args.OptimizerNames attribute)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.ntk_ratio">ntk_ratio (lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.num_examples">num_examples() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.num_output_sequences">num_output_sequences (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.num_patches">num_patches (lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower property)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.num_raft_iteration">num_raft_iteration (lmflow.args.RaftAlignerArguments attribute)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="O">O</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.observation_formatter">observation_formatter (lmflow.utils.conversation_template.base.ConversationTemplate attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.observation_formatter">(lmflow.utils.conversation_template.ConversationTemplate attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.offset">offset (lmflow.utils.llava_conversation_lib.Conversation attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_epoch_end">on_epoch_end() (lmflow.pipeline.utils.peft_trainer.PeftSavingCallback method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_save">on_save() (lmflow.pipeline.utils.peft_trainer.PeftSavingCallback method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_train_end">on_train_end() (lmflow.pipeline.utils.peft_trainer.PeftSavingCallback method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_adam_beta1">optim_adam_beta1 (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_adam_beta2">optim_adam_beta2 (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_beta1">optim_beta1 (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_beta2">optim_beta2 (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_beta3">optim_beta3 (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_dummy_beta1">optim_dummy_beta1 (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_dummy_beta2">optim_dummy_beta2 (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_momentum">optim_momentum (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.optim_weight_decay">optim_weight_decay (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.OPTIMIZER_NAME">OPTIMIZER_NAME (in module lmflow.pipeline.utils.raft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.optimizer_type">optimizer_type (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames">OptimizerNames (class in lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.RewardModelInferenceResultWithInput.output">output (lmflow.utils.data_utils.RewardModelInferenceResultWithInput attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.VLLMInferenceResultWithInput.output">(lmflow.utils.data_utils.VLLMInferenceResultWithInput attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.output_dir">output_dir (lmflow.args.DPOAlignerArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.output_dir">(lmflow.args.EvaluatorArguments attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments.output_dir">(lmflow.args.IterativeDPOAlignerArguments attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.output_max_length">output_max_length (lmflow.args.RaftAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.output_min_length">output_min_length (lmflow.args.RaftAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.output_reward_path">output_reward_path (lmflow.args.RaftAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.overwrite_cache">overwrite_cache (lmflow.args.DatasetArguments attribute)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="P">P</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.pad_to_multiple_of">pad_to_multiple_of (lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.padding">padding (lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.padding">(lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.padding_side">padding_side (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.padding_value">padding_value (lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.PAIRED_CONVERSATION_DATASET_DESCRIPTION">PAIRED_CONVERSATION_DATASET_DESCRIPTION (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.paired_conversation_tokenize_function">paired_conversation_tokenize_function() (in module lmflow.tokenization.hf_text_regression_model)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION">PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer.parse_to_sampling_params">parse_to_sampling_params() (lmflow.pipeline.vllm_inferencer.VLLMInferencer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.peft_config">peft_config (lmflow.models.hf_model_mixin.HFModelMixin attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/rm_trainer/index.html#lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer">PeftRewardTrainer (class in lmflow.pipeline.utils.rm_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftSavingCallback">PeftSavingCallback (class in lmflow.pipeline.utils.peft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/peft_trainer/index.html#lmflow.pipeline.utils.peft_trainer.PeftTrainer">PeftTrainer (class in lmflow.pipeline.utils.peft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.per_device_eval_batch_size">per_device_eval_batch_size (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.per_device_train_batch_size">per_device_train_batch_size (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.PHI3_TEMPLATE">PHI3_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/phi/index.html#lmflow.utils.conversation_template.phi.PHI3_TEMPLATE">(in module lmflow.utils.conversation_template.phi)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.pi_ratio">pi_ratio (lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.PIPELINE_ARGUMENT_MAPPING">PIPELINE_ARGUMENT_MAPPING (in module lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/auto_pipeline/index.html#lmflow.pipeline.auto_pipeline.PIPELINE_MAPPING">PIPELINE_MAPPING (in module lmflow.pipeline.auto_pipeline)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/auto_pipeline/index.html#lmflow.pipeline.auto_pipeline.PIPELINE_NEEDS_EXTRAS">PIPELINE_NEEDS_EXTRAS (in module lmflow.pipeline.auto_pipeline)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.place_model_on_device">place_model_on_device (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.SeparatorStyle.PLAIN">PLAIN (lmflow.utils.llava_conversation_lib.SeparatorStyle attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.pop_callback">pop_callback() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.post_process_pairs">post_process_pairs() (lmflow.utils.conversation_template.base.ConversationTemplate method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.post_process_pairs">(lmflow.utils.conversation_template.ConversationTemplate method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.postprocess_distributed_inference_outputs">postprocess_distributed_inference_outputs() (lmflow.models.hf_text_regression_model.HFTextRegressionModel static method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.postprocess_inference_outputs">postprocess_inference_outputs() (lmflow.models.hf_text_regression_model.HFTextRegressionModel static method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.predict">predict() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.predict_next_token">predict_next_token() (lmflow.pipeline.inferencer.SpeculativeInferencer static method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.prediction_loop">prediction_loop() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.prediction_step">prediction_step() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding">PreferenceDataCollatorWithPadding (class in lmflow.pipeline.utils.dpov2_dataprocessor)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.prepare_inputs_for_inference">prepare_inputs_for_inference() (lmflow.models.hf_decoder_model.HFDecoderModel method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.prepare_inputs_for_inference">(lmflow.models.hf_text_regression_model.HFTextRegressionModel method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.prepare_inputs_labels_for_multimodal">prepare_inputs_labels_for_multimodal() (lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower method)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_plain">preprocess_llama_from_llava_plain() (in module lmflow.datasets.multi_modal_dataset)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_v1">preprocess_llama_from_llava_v1() (in module lmflow.datasets.multi_modal_dataset)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.preprocess_logits_for_metrics">preprocess_logits_for_metrics (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.preprocess_multimodal_llava">preprocess_multimodal_llava() (in module lmflow.datasets.multi_modal_dataset)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.preprocessing_num_workers">preprocessing_num_workers (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.PRESET_TEMPLATES">PRESET_TEMPLATES (in module lmflow.utils.conversation_template)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.pretrained_language_projection_path">pretrained_language_projection_path (lmflow.args.VisModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/common/index.html#lmflow.utils.common.print_banner">print_banner() (in module lmflow.utils.common)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.process_image_flag">process_image_flag() (in module lmflow.utils.data_utils)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.processor_image_token_in_minigpt4">processor_image_token_in_minigpt4() (lmflow.models.vision2seq_model.CustomAutoVision2SeqModel method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.prompt_cache_path">prompt_cache_path (lmflow.args.VisModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.prompt_structure">prompt_structure (lmflow.args.EvaluatorArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.push_to_hub">push_to_hub() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="Q">Q</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.qformer_from_pretrained">qformer_from_pretrained() (lmflow.models.vision2seq_model.CustomAutoVision2SeqModel method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.qformer_name_or_path">qformer_name_or_path (lmflow.args.VisModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.quant_config">quant_config (lmflow.models.hf_model_mixin.HFModelMixin attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.quant_type">quant_type (lmflow.args.ModelArguments attribute)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.QWEN2_TEMPLATE">QWEN2_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/qwen/index.html#lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE">(in module lmflow.utils.conversation_template.qwen)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.QWEN2_TEMPLATE_FOR_TOOL">QWEN2_TEMPLATE_FOR_TOOL (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/qwen/index.html#lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE_FOR_TOOL">(in module lmflow.utils.conversation_template.qwen)</a>
+</li>
+      </ul></li>
+  </ul></td>
+</tr></table>
+
+<h2 id="R">R</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/optim/radam/index.html#lmflow.optim.radam.RAdam">RAdam (class in lmflow.optim.radam)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.RADAM">RADAM (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.raft_batch_size">raft_batch_size (lmflow.args.RaftAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/raft_aligner/index.html#lmflow.pipeline.raft_aligner.RaftAligner">RaftAligner (class in lmflow.pipeline.raft_aligner)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments">RaftAlignerArguments (class in lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer">RaftTrainer (class in lmflow.pipeline.utils.raft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.random_seed">random_seed (lmflow.args.DPOv2AlignerArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.random_seed">(lmflow.args.EvaluatorArguments attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.random_seed">(lmflow.args.InferencerArguments attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.random_shuffle">random_shuffle (lmflow.args.EvaluatorArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.rectify">rectify (lmflow.optim.adabelief.AdaBelief attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.ref_model_args">ref_model_args (lmflow.pipeline.dpov2_aligner.DPOv2Aligner attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.ref_model_args">(lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.ref_model_args">(lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.ReferenceModelArguments">ReferenceModelArguments (in module lmflow.pipeline.dpov2_aligner)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.html#lmflow.pipeline.utils.memory_safe_dpov2_align.ReferenceModelArguments">(in module lmflow.pipeline.utils.memory_safe_dpov2_align)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/models/text_regression_model/index.html#lmflow.models.text_regression_model.TextRegressionModel.register_inference_function">register_inference_function() (lmflow.models.text_regression_model.TextRegressionModel method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.register_prompt_cache">register_prompt_cache() (lmflow.models.vision2seq_model.CustomAutoVision2SeqModel method)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.register_tokenizer">register_tokenizer() (lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/regression_model/index.html#lmflow.models.regression_model.RegressionModel">RegressionModel (class in lmflow.models.regression_model)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.remove_callback">remove_callback() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/common/index.html#lmflow.utils.common.remove_dataclass_attr_prefix">remove_dataclass_attr_prefix() (in module lmflow.utils.common)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.remove_last_sep">remove_last_sep (lmflow.utils.conversation_template.base.ConversationTemplate attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.remove_last_sep">(lmflow.utils.conversation_template.ConversationTemplate attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.remove_last_separator">remove_last_separator() (lmflow.utils.conversation_template.base.ConversationTemplate method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.remove_last_separator">(lmflow.utils.conversation_template.ConversationTemplate method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.remove_unused_columns">remove_unused_columns (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.repetition_penalty">repetition_penalty (lmflow.args.EvaluatorArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.repetition_penalty">(lmflow.args.InferencerArguments attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.html#lmflow.utils.flash_attention.bloom_flash_attention.replace_bloom_attn_with_flash_attn">replace_bloom_attn_with_flash_attn() (in module lmflow.utils.flash_attention.bloom_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.html#lmflow.utils.flash_attention.gpt2_flash_attention.replace_gpt2_attn_with_flash_attn">replace_gpt2_attn_with_flash_attn() (in module lmflow.utils.flash_attention.gpt2_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html#lmflow.utils.flash_attention.gpt_neo_flash_attention.replace_gpt_neo_attn_with_flash_attn">replace_gpt_neo_attn_with_flash_attn() (in module lmflow.utils.flash_attention.gpt_neo_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.html#lmflow.utils.flash_attention.llama_flash_attention.replace_llama_attn_with_flash_attn">replace_llama_attn_with_flash_attn() (in module lmflow.utils.flash_attention.llama_flash_attention)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.replace_llama_with_condense">replace_llama_with_condense() (in module lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.report_to">report_to (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.reset">reset() (lmflow.optim.adabelief.AdaBelief method)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adan/index.html#lmflow.optim.adan.Adan.restart_opt">restart_opt() (lmflow.optim.adan.Adan method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.results_path">results_path (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.RETURN_CODE_ERROR_BUFFER">RETURN_CODE_ERROR_BUFFER (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.return_tensors">return_tensors (lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.reward_model_args">reward_model_args (lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments.reward_model_inference_batch_size">reward_model_inference_batch_size (lmflow.args.IterativeDPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.IterativeDPOAlignerArguments.reward_model_inference_block_size">reward_model_inference_block_size (lmflow.args.IterativeDPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding">RewardDataCollatorWithPadding (class in lmflow.pipeline.utils.rm_dataprocessor)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer">RewardModelInferencer (class in lmflow.pipeline.rm_inferencer)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.RewardModelInferenceResultWithInput">RewardModelInferenceResultWithInput (class in lmflow.utils.data_utils)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/rm_tuner/index.html#lmflow.pipeline.rm_tuner.RewardModelTuner">RewardModelTuner (class in lmflow.pipeline.rm_tuner)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.RewardModelTunerArguments">RewardModelTunerArguments (class in lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/rm_trainer/index.html#lmflow.pipeline.utils.rm_trainer.RewardTrainer">RewardTrainer (class in lmflow.pipeline.utils.rm_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/rm_trainer/index.html#lmflow.pipeline.utils.rm_trainer.rm_loss">rm_loss() (in module lmflow.pipeline.utils.rm_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.roles">roles (lmflow.utils.llava_conversation_lib.Conversation attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.rope_ntk_ratio">rope_ntk_ratio (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.rope_pi_ratio">rope_pi_ratio (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.rstrip_partial_utf8">rstrip_partial_utf8() (in module lmflow.pipeline.inferencer)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.run_name">run_name (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="S">S</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.sample">sample() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.sample">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.sample">(lmflow.pipeline.inferencer.SpeculativeInferencer static method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#lmflow.pipeline.dpov2_aligner.DPOv2Aligner.sampling_paired_idx_from_rewards">sampling_paired_idx_from_rewards() (lmflow.pipeline.dpov2_aligner.DPOv2Aligner method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOv2AlignerArguments.sampling_paired_method">sampling_paired_method (lmflow.args.DPOv2AlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer.sampling_params">sampling_params (lmflow.pipeline.vllm_inferencer.VLLMInferencer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.sanity_check">sanity_check (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.sanity_check">sanity_check() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.sanity_check">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.save">save() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.save">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.save">(lmflow.models.hf_decoder_model.HFDecoderModel method)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.save">(lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel method)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.save">(lmflow.models.hf_text_regression_model.HFTextRegressionModel method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.save_aggregated_lora">save_aggregated_lora (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.save_counter">save_counter (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.save_inference_results">save_inference_results() (lmflow.pipeline.vllm_inferencer.InferencerWithOffloading method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer.save_inference_results">(lmflow.pipeline.vllm_inferencer.VLLMInferencer method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.save_language_projection">save_language_projection (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.save_model">save_model() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.save_pretrain_model_path">save_pretrain_model_path (lmflow.args.VisModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.save_prompt_cache">save_prompt_cache() (lmflow.models.vision2seq_model.CustomAutoVision2SeqModel method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.save_results">save_results (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.save_steps">save_steps (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.SCALER_NAME">SCALER_NAME (in module lmflow.pipeline.utils.raft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.SCHEDULER_NAME">SCHEDULER_NAME (in module lmflow.pipeline.utils.raft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.score_to_prob">score_to_prob() (lmflow.pipeline.inferencer.SpeculativeInferencer static method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.seed">seed (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.select_feature">select_feature (lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.select_layer">select_layer (lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.sep">sep (lmflow.utils.llava_conversation_lib.Conversation attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.sep2">sep2 (lmflow.utils.llava_conversation_lib.Conversation attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.MultiModalDatasetArguments.sep_style">sep_style (lmflow.args.MultiModalDatasetArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.sep_style">(lmflow.utils.llava_conversation_lib.Conversation attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.separator">separator (lmflow.utils.conversation_template.base.ConversationTemplate attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.separator">(lmflow.utils.conversation_template.ConversationTemplate attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.SeparatorStyle">SeparatorStyle (class in lmflow.utils.llava_conversation_lib)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.set_random_seed">set_random_seed() (in module lmflow.utils.data_utils)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.SGD_SCHEDULE_FREE">SGD_SCHEDULE_FREE (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/sgdp/index.html#lmflow.optim.sgdp.SGDP">SGDP (class in lmflow.optim.sgdp)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.SGDP">(lmflow.args.OptimizerNames attribute)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/optim/sgd_schedule_free/index.html#lmflow.optim.sgd_schedule_free.SGDScheduleFree">SGDScheduleFree (class in lmflow.optim.sgd_schedule_free)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.sharded_ddp">sharded_ddp (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.SeparatorStyle.SINGLE">SINGLE (lmflow.utils.llava_conversation_lib.SeparatorStyle attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.skip_first_batches">skip_first_batches (in module lmflow.pipeline.utils.raft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.skip_next">skip_next (lmflow.utils.llava_conversation_lib.Conversation attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.SOPHIA">SOPHIA (lmflow.args.OptimizerNames attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/sophia/index.html#lmflow.optim.sophia.SophiaG">SophiaG (class in lmflow.optim.sophia)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.special_starter">special_starter (lmflow.utils.conversation_template.base.ConversationTemplate attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.special_starter">(lmflow.utils.conversation_template.ConversationTemplate attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.special_stopper">special_stopper (lmflow.utils.conversation_template.base.ConversationTemplate attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.special_stopper">(lmflow.utils.conversation_template.ConversationTemplate attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer">SpeculativeInferencer (class in lmflow.pipeline.inferencer)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.split_args">split_args() (in module lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.state">state (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.step">step() (lmflow.optim.adabelief.AdaBelief method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/adabound/index.html#lmflow.optim.adabound.AdaBound.step">(lmflow.optim.adabound.AdaBound method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adadelta/index.html#lmflow.optim.adadelta.Adadelta.step">(lmflow.optim.adadelta.Adadelta method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adagrad/index.html#lmflow.optim.adagrad.AdaGrad.step">(lmflow.optim.adagrad.AdaGrad method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adam/index.html#lmflow.optim.adam.Adam.step">(lmflow.optim.adam.Adam method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adamax/index.html#lmflow.optim.adamax.Adamax.step">(lmflow.optim.adamax.Adamax method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adamp/index.html#lmflow.optim.adamp.AdamP.step">(lmflow.optim.adamp.AdamP method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adamw_schedule_free/index.html#lmflow.optim.adamw_schedule_free.AdamWScheduleFree.step">(lmflow.optim.adamw_schedule_free.AdamWScheduleFree method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/adan/index.html#lmflow.optim.adan.Adan.step">(lmflow.optim.adan.Adan method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/dummy/index.html#lmflow.optim.dummy.Dummy.step">(lmflow.optim.dummy.Dummy method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/lamb/index.html#lmflow.optim.lamb.Lamb.step">(lmflow.optim.lamb.Lamb method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/lars/index.html#lmflow.optim.lars.LARS.step">(lmflow.optim.lars.LARS method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/nadam/index.html#lmflow.optim.nadam.NAdam.step">(lmflow.optim.nadam.NAdam method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/novograd/index.html#lmflow.optim.novograd.NovoGrad.step">(lmflow.optim.novograd.NovoGrad method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/radam/index.html#lmflow.optim.radam.RAdam.step">(lmflow.optim.radam.RAdam method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/sgd_schedule_free/index.html#lmflow.optim.sgd_schedule_free.SGDScheduleFree.step">(lmflow.optim.sgd_schedule_free.SGDScheduleFree method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/sgdp/index.html#lmflow.optim.sgdp.SGDP.step">(lmflow.optim.sgdp.SGDP method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/sophia/index.html#lmflow.optim.sophia.SophiaG.step">(lmflow.optim.sophia.SophiaG method)</a>
+</li>
+        <li><a href="autoapi/lmflow/optim/yogi/index.html#lmflow.optim.yogi.Yogi.step">(lmflow.optim.yogi.Yogi method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.store_flos">store_flos() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.stream_inference">stream_inference() (lmflow.pipeline.inferencer.Inferencer method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.SpeculativeInferencer.stream_inference">(lmflow.pipeline.inferencer.SpeculativeInferencer method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.streaming">streaming (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.StringFormatter">StringFormatter (class in lmflow.utils.conversation_template.base)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.supported_dataset_type">supported_dataset_type (in module lmflow.pipeline.inferencer)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.system">system (lmflow.utils.llava_conversation_lib.Conversation attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.system_formatter">system_formatter (lmflow.utils.conversation_template.base.ConversationTemplate attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.system_formatter">(lmflow.utils.conversation_template.ConversationTemplate attribute)</a>
+</li>
+      </ul></li>
+  </ul></td>
+</tr></table>
+
+<h2 id="T">T</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.temperature">temperature (lmflow.args.EvaluatorArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.temperature">(lmflow.args.InferencerArguments attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.Formatter.template">template (lmflow.utils.conversation_template.base.Formatter attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.template_name">template_name (lmflow.utils.conversation_template.base.ConversationTemplate attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.template_name">(lmflow.utils.conversation_template.ConversationTemplate attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent">TemplateComponent (class in lmflow.utils.conversation_template.base)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.tensor_parallel_size">tensor_parallel_size (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.test_file">test_file (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT2TEXT_DATASET_DESCRIPTION">TEXT2TEXT_DATASET_DESCRIPTION (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT2TEXT_DATASET_DETAILS">TEXT2TEXT_DATASET_DETAILS (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT2TEXT_DATASET_LONG_DESCRITION">TEXT2TEXT_DATASET_LONG_DESCRITION (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT_ONLY_DATASET_DESCRIPTION">TEXT_ONLY_DATASET_DESCRIPTION (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT_ONLY_DATASET_DETAILS">TEXT_ONLY_DATASET_DETAILS (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT_ONLY_DATASET_LONG_DESCRITION">TEXT_ONLY_DATASET_LONG_DESCRITION (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION">TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.TEXT_TO_TEXTLIST_DATASET_DESCRIPTION">TEXT_TO_TEXTLIST_DATASET_DESCRIPTION (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.text_to_textlist_tokenize_function">text_to_textlist_tokenize_function() (in module lmflow.tokenization.hf_text_regression_model)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/text_regression_model/index.html#lmflow.models.text_regression_model.TextRegressionModel">TextRegressionModel (class in lmflow.models.text_regression_model)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.to_dict">to_dict() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.to_dict">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.to_gradio_chatbot">to_gradio_chatbot() (lmflow.utils.llava_conversation_lib.Conversation method)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.to_list">to_list() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.to_list">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/tokenization/hf_decoder_model/index.html#lmflow.tokenization.hf_decoder_model.tok_logger">tok_logger (in module lmflow.tokenization.hf_decoder_model)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.tok_logger">(in module lmflow.tokenization.hf_text_regression_model)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.token">token (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/hf_decoder_model/index.html#lmflow.models.hf_decoder_model.HFDecoderModel.tokenize">tokenize() (lmflow.models.hf_decoder_model.HFDecoderModel method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.tokenize">(lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel method)</a>
+</li>
+        <li><a href="autoapi/lmflow/models/hf_text_regression_model/index.html#lmflow.models.hf_text_regression_model.HFTextRegressionModel.tokenize">(lmflow.models.hf_text_regression_model.HFTextRegressionModel method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.tokenize_batch_element">tokenize_batch_element() (lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding method)</a>
+</li>
+      <li><a href="autoapi/lmflow/tokenization/hf_decoder_model/index.html#lmflow.tokenization.hf_decoder_model.tokenize_function">tokenize_function() (in module lmflow.tokenization.hf_decoder_model)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/tokenization/hf_text_regression_model/index.html#lmflow.tokenization.hf_text_regression_model.tokenize_function">(in module lmflow.tokenization.hf_text_regression_model)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset.tokenizer">tokenizer (lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.tokenizer">(lmflow.models.hf_model_mixin.HFModelMixin attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.tokenizer">(lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.tokenizer">(lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.tokenizer">(lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding attribute)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#lmflow.datasets.multi_modal_dataset.tokenizer_image_token">tokenizer_image_token() (in module lmflow.datasets.multi_modal_dataset)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.tokenizer_name">tokenizer_name (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.ToolInferencer">ToolInferencer (class in lmflow.pipeline.inferencer)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.tools_formatter">tools_formatter (lmflow.utils.conversation_template.base.ConversationTemplate attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.tools_formatter">(lmflow.utils.conversation_template.ConversationTemplate attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.top_k">top_k (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.top_p">top_p (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.RaftAlignerArguments.top_reward_percentage">top_reward_percentage (lmflow.args.RaftAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.torch_dtype">torch_dtype (lmflow.args.ModelArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.torch_dtype">(lmflow.models.hf_model_mixin.HFModelMixin attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.torch_jit_model_eval">torch_jit_model_eval() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adamw_schedule_free/index.html#lmflow.optim.adamw_schedule_free.AdamWScheduleFree.train">train() (lmflow.optim.adamw_schedule_free.AdamWScheduleFree method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/optim/sgd_schedule_free/index.html#lmflow.optim.sgd_schedule_free.SGDScheduleFree.train">(lmflow.optim.sgd_schedule_free.SGDScheduleFree method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.train">(lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/dpo_aligner/index.html#lmflow.pipeline.dpo_aligner.DPOAligner.train_dataset">train_dataset (lmflow.pipeline.dpo_aligner.DPOAligner attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.train_dataset">(lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.train_file">train_file (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.train_on_prompt">train_on_prompt (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.train_test_split">train_test_split() (lmflow.datasets.Dataset method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.train_test_split">(lmflow.datasets.dataset.Dataset method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.TRAINER_STATE_NAME">TRAINER_STATE_NAME (in module lmflow.pipeline.utils.raft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.TRAINING_ARGS_NAME">TRAINING_ARGS_NAME (in module lmflow.pipeline.utils.raft_trainer)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.training_step">training_step() (lmflow.pipeline.utils.raft_trainer.RaftTrainer method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.truncate_to_model_max_length">truncate_to_model_max_length (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.truncation_mode">truncation_mode (lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.truncation_side">truncation_side (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.trust_remote_code">trust_remote_code (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/interfaces/tunable/index.html#lmflow.models.interfaces.tunable.Tunable">Tunable (class in lmflow.models.interfaces.tunable)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/base_tuner/index.html#lmflow.pipeline.base_tuner.BaseTuner.tune">tune() (lmflow.pipeline.base_tuner.BaseTuner method)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/finetuner/index.html#lmflow.pipeline.finetuner.Finetuner.tune">(lmflow.pipeline.finetuner.Finetuner method)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/rm_tuner/index.html#lmflow.pipeline.rm_tuner.RewardModelTuner.tune">(lmflow.pipeline.rm_tuner.RewardModelTuner method)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.SeparatorStyle.TWO">TWO (lmflow.utils.llava_conversation_lib.SeparatorStyle attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/datasets/index.html#lmflow.datasets.Dataset.type">type (lmflow.datasets.Dataset attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/datasets/dataset/index.html#lmflow.datasets.dataset.Dataset.type">(lmflow.datasets.dataset.Dataset attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.TemplateComponent.type">(lmflow.utils.conversation_template.base.TemplateComponent attribute)</a>
+</li>
+      </ul></li>
+  </ul></td>
+</tr></table>
+
+<h2 id="U">U</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/multimodal/index.html#lmflow.utils.multimodal.update_custom_config">update_custom_config() (in module lmflow.utils.multimodal)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/sophia/index.html#lmflow.optim.sophia.SophiaG.update_hessian">update_hessian() (lmflow.optim.sophia.SophiaG method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.use_accelerator">use_accelerator (lmflow.args.InferencerArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/hf_model_mixin/index.html#lmflow.models.hf_model_mixin.HFModelMixin.use_accelerator">(lmflow.models.hf_model_mixin.HFModelMixin attribute)</a>
+</li>
+      </ul></li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.use_accelerator_for_evaluator">use_accelerator_for_evaluator (lmflow.args.EvaluatorArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_apex">use_apex (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.use_beam_search">use_beam_search (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_cpu_amp">use_cpu_amp (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_cuda_amp">use_cuda_amp (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.use_customized_optim">use_customized_optim (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_dora">use_dora (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.use_dpo_data_collator">use_dpo_data_collator (lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_fast_tokenizer">use_fast_tokenizer (lmflow.args.ModelArguments attribute)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_flash_attention">use_flash_attention (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.MultiModalDatasetArguments.use_image_start_end">use_image_start_end (lmflow.args.MultiModalDatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_int8">use_int8 (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.FinetunerArguments.use_lisa">use_lisa (lmflow.args.FinetunerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_lora">use_lora (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.use_prompt_cache">use_prompt_cache (lmflow.args.VisModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_qlora">use_qlora (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.ModelArguments.use_ram_optimized_load">use_ram_optimized_load (lmflow.args.ModelArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_tune_checkpoints">use_tune_checkpoints (lmflow.pipeline.utils.raft_trainer.RaftTrainer attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.use_vllm">use_vllm (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.EvaluatorArguments.use_wandb">use_wandb (lmflow.args.EvaluatorArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/conversation_template/base/index.html#lmflow.utils.conversation_template.base.ConversationTemplate.user_formatter">user_formatter (lmflow.utils.conversation_template.base.ConversationTemplate attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ConversationTemplate.user_formatter">(lmflow.utils.conversation_template.ConversationTemplate attribute)</a>
+</li>
+      </ul></li>
+  </ul></td>
+</tr></table>
+
+<h2 id="V">V</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.validation_file">validation_file (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DatasetArguments.validation_split_percentage">validation_split_percentage (lmflow.args.DatasetArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#lmflow.utils.llava_conversation_lib.Conversation.version">version (lmflow.utils.llava_conversation_lib.Conversation attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.vision_feature_select">vision_feature_select() (lmflow.models.vision2seq_model.CustomAutoVision2SeqModel method)</a>
+</li>
+      <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.vision_model_from_pretrained">vision_model_from_pretrained() (lmflow.models.vision2seq_model.CustomAutoVision2SeqModel method)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.vision_select_layer">vision_select_layer (lmflow.args.VisModelArguments attribute)</a>
+</li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.vision_tower_name">vision_tower_name (lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments">VisModelArguments (class in lmflow.args)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.vllm_gpu_memory_utilization">vllm_gpu_memory_utilization (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.vllm_inference_batch_size">vllm_inference_batch_size (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.InferencerArguments.vllm_tensor_parallel_size">vllm_tensor_parallel_size (lmflow.args.InferencerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#lmflow.pipeline.vllm_inferencer.VLLMInferencer">VLLMInferencer (class in lmflow.pipeline.vllm_inferencer)</a>
+</li>
+      <li><a href="autoapi/lmflow/utils/data_utils/index.html#lmflow.utils.data_utils.VLLMInferenceResultWithInput">VLLMInferenceResultWithInput (class in lmflow.utils.data_utils)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="W">W</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.warmup_steps">warmup_steps (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.DPOAlignerArguments.weight_decay">weight_decay (lmflow.args.DPOAlignerArguments attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/optim/adabelief/index.html#lmflow.optim.adabelief.AdaBelief.weight_decouple">weight_decouple (lmflow.optim.adabelief.AdaBelief attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.VisModelArguments.with_qformer">with_qformer (lmflow.args.VisModelArguments attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/models/vision2seq_model/index.html#lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.with_qformer">(lmflow.models.vision2seq_model.CustomAutoVision2SeqModel attribute)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/constants/index.html#lmflow.utils.constants.WORKER_HEART_BEAT_INTERVAL">WORKER_HEART_BEAT_INTERVAL (in module lmflow.utils.constants)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.workspace_path">workspace_path (lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner attribute)</a>
+</li>
+      <li><a href="autoapi/lmflow/pipeline/evaluator/index.html#lmflow.pipeline.evaluator.Evaluator.world_size">world_size (lmflow.pipeline.evaluator.Evaluator attribute)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/pipeline/inferencer/index.html#lmflow.pipeline.inferencer.Inferencer.world_size">(lmflow.pipeline.inferencer.Inferencer attribute)</a>
+</li>
+        <li><a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#lmflow.pipeline.rm_inferencer.RewardModelInferencer.world_size">(lmflow.pipeline.rm_inferencer.RewardModelInferencer attribute)</a>
+</li>
+      </ul></li>
+  </ul></td>
+</tr></table>
+
+<h2 id="Y">Y</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.YI1_5_TEMPLATE">YI1_5_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/yi/index.html#lmflow.utils.conversation_template.yi.YI1_5_TEMPLATE">(in module lmflow.utils.conversation_template.yi)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/optim/yogi/index.html#lmflow.optim.yogi.Yogi">Yogi (class in lmflow.optim.yogi)</a>
+</li>
+      <li><a href="autoapi/lmflow/args/index.html#lmflow.args.OptimizerNames.YOGI">YOGI (lmflow.args.OptimizerNames attribute)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+<h2 id="Z">Z</h2>
+<table style="width: 100%" class="indextable genindextable"><tr>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/conversation_template/index.html#lmflow.utils.conversation_template.ZEPHYR_TEMPLATE">ZEPHYR_TEMPLATE (in module lmflow.utils.conversation_template)</a>
+
+      <ul>
+        <li><a href="autoapi/lmflow/utils/conversation_template/zephyr/index.html#lmflow.utils.conversation_template.zephyr.ZEPHYR_TEMPLATE">(in module lmflow.utils.conversation_template.zephyr)</a>
+</li>
+      </ul></li>
+  </ul></td>
+  <td style="width: 33%; vertical-align: top;"><ul>
+      <li><a href="autoapi/lmflow/utils/conversation_template/zephyr/index.html#lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate">ZephyrConversationTemplate (class in lmflow.utils.conversation_template.zephyr)</a>
+</li>
+  </ul></td>
+</tr></table>
+
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/index.html b/index.html
new file mode 100644
index 000000000..60aed65b8
--- /dev/null
+++ b/index.html
@@ -0,0 +1,794 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="./" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
+
+    <title>LMFlow &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="_static/doctools.js?v=9bcbadda"></script>
+    <script src="_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'index';</script>
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="search.html" />
+    <link rel="next" title="Blogs" href="blogs/index.html" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="#">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
+      <span class="fa-solid fa-outdent"></span>
+    </button>
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none"></div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+  <img src="_static/logo.png" alt="LMFlow" style="width: 100%; min-width: 300px; display: block; margin: auto; background-color: transparent;">
+<section id="lmflow">
+<h1>LMFlow<a class="headerlink" href="#lmflow" title="Link to this heading">#</a></h1>
+<p>An extensible, convenient, and efficient toolbox for finetuning large machine learning models, designed to be user-friendly, speedy and reliable, and accessible to the entire community.</p>
+<div class="sd-container-fluid sd-sphinx-override sd-mb-4 docutils">
+<div class="sd-row sd-row-cols-1 sd-row-cols-xs-1 sd-row-cols-sm-2 sd-row-cols-md-3 sd-row-cols-lg-4 docutils">
+<div class="sd-col sd-d-flex-row docutils">
+<div class="sd-card sd-sphinx-override sd-w-100 sd-shadow-sm docutils">
+<div class="sd-card-body docutils">
+<div class="sd-card-title sd-font-weight-bold docutils">
+Extensible</div>
+<p class="sd-card-text">Support common backbones (LLaMA, Galactica, GPT-2, etc.)</p>
+</div>
+</div>
+</div>
+<div class="sd-col sd-d-flex-row docutils">
+<div class="sd-card sd-sphinx-override sd-w-100 sd-shadow-sm docutils">
+<div class="sd-card-body docutils">
+<div class="sd-card-title sd-font-weight-bold docutils">
+Light-Weight</div>
+<p class="sd-card-text">Extremely few parameters with LoRA  (LLaMA 33B: only 25MB storage)</p>
+</div>
+</div>
+</div>
+<div class="sd-col sd-d-flex-row docutils">
+<div class="sd-card sd-sphinx-override sd-w-100 sd-shadow-sm docutils">
+<div class="sd-card-body docutils">
+<div class="sd-card-title sd-font-weight-bold docutils">
+Task-Oriented</div>
+<p class="sd-card-text">Comparable with ChatGPT on 7B/33B models.</p>
+</div>
+</div>
+</div>
+<div class="sd-col sd-d-flex-row docutils">
+<div class="sd-card sd-sphinx-override sd-w-100 sd-shadow-sm docutils">
+<div class="sd-card-body docutils">
+<div class="sd-card-title sd-font-weight-bold docutils">
+Open</div>
+<p class="sd-card-text">The whole pipeline (data, models, tuning, inference) is open-source.</p>
+</div>
+</div>
+</div>
+</div>
+</div>
+<section id="introduction">
+<h2>Introduction<a class="headerlink" href="#introduction" title="Link to this heading">#</a></h2>
+<p>The remarkable achievements of large foundation models, such as the expansive language models, have demonstrated their exceptional capacity to attain human-like intelligence that surpasses conventional methods. Despite their growing accessibility, these models still require fine-tuning to cater to specific tasks while maintaining their overall AI competency. We are pleased to introduce our lightweight toolkit, which features thoughtfully designed and easily scalable APIs. This tool simplifies the process of fine-tuning and inference of publicly available foundation models to maximize their effectiveness.</p>
+<p>We have thoroughly tested this toolkit and are pleased to make it available under <a class="reference external" href="https://github.com/OptimalScale/LMFlow">Github</a>.</p>
+</section>
+<section id="features">
+<h2>Features<a class="headerlink" href="#features" title="Link to this heading">#</a></h2>
+<section id="task-tuning">
+<h3>Task Tuning<a class="headerlink" href="#task-tuning" title="Link to this heading">#</a></h3>
+<p>The goal of Task Tuning is to enhance a language model’s proficiency in a particular field, such as medicine or mathematics. By doing so, the model acquires domain-specific information, allowing it to adapt better to the target subject matter.</p>
+<p>For instance, if a medical dataset is used for task tuning, the language model can gain medical knowledge that can be applied to other medical datasets.</p>
+<p>To emphasize its significance, we applied task tuning to LLaMA models on PubMedQA and MedMCQA datasets and evaluated their performance. We observed significant improvements both in-domain (PubMedQA, MedMCQA) and out-of-domain (MedQA-USMLE) dataset.</p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head text-center"><p></p></th>
+<th class="head text-center"><p>PubMedQA</p></th>
+<th class="head text-center"><p>MedQA-USMLE</p></th>
+<th class="head text-center"><p>MedMCQA</p></th>
+<th class="head text-center"><p>Average</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td class="text-center"><p>Human (pass)</p></td>
+<td class="text-center"><p></p></td>
+<td class="text-center"><p>60.0</p></td>
+<td class="text-center"><p>50.0</p></td>
+<td class="text-center"><p></p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p>Human (expert)</p></td>
+<td class="text-center"><p>78.0</p></td>
+<td class="text-center"><p>87.0</p></td>
+<td class="text-center"><p>90.0</p></td>
+<td class="text-center"><p>85.0</p></td>
+</tr>
+<tr class="row-even"><td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p>InstructGPT 175B</p></td>
+<td class="text-center"><p>73.2</p></td>
+<td class="text-center"><p>46.0</p></td>
+<td class="text-center"><p>44.0</p></td>
+<td class="text-center"><p>54.4</p></td>
+</tr>
+<tr class="row-even"><td class="text-center"><p>ChatGPT</p></td>
+<td class="text-center"><p>63.9</p></td>
+<td class="text-center"><p><strong>57.0</strong></p></td>
+<td class="text-center"><p>44.7</p></td>
+<td class="text-center"><p>55.2</p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p>LLaMA 7B</p></td>
+<td class="text-center"><p>5.2</p></td>
+<td class="text-center"><p>27.1</p></td>
+<td class="text-center"><p>24.3</p></td>
+<td class="text-center"><p>18.9</p></td>
+</tr>
+<tr class="row-even"><td class="text-center"><p>LLaMA 33B</p></td>
+<td class="text-center"><p>1.8</p></td>
+<td class="text-center"><p>43.4</p></td>
+<td class="text-center"><p>30.3</p></td>
+<td class="text-center"><p>25.2</p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+</tr>
+<tr class="row-even"><td class="text-center"><p>Task-tuned LLaMA 7B (Full)</p></td>
+<td class="text-center"><p><strong>75.1</strong></p></td>
+<td class="text-center"><p>44.5</p></td>
+<td class="text-center"><p>49.9</p></td>
+<td class="text-center"><p>56.5</p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p>Task-tuned LLaMA 33B (LoRA)</p></td>
+<td class="text-center"><p>74</p></td>
+<td class="text-center"><p>51.3</p></td>
+<td class="text-center"><p><strong>50.2</strong></p></td>
+<td class="text-center"><p><strong>58.5</strong></p></td>
+</tr>
+</tbody>
+</table>
+</div>
+<p>Moreover, we also test the MMLU performance to further verify the out-of-domain robustness of Task Tuning technique.</p>
+<div class="pst-scrollable-table-container"><table class="table">
+<thead>
+<tr class="row-odd"><th class="head text-center"><p>MMLU task</p></th>
+<th class="head text-center"><p>anatomy</p></th>
+<th class="head text-center"><p>clinical knowledge</p></th>
+<th class="head text-center"><p>college biology</p></th>
+<th class="head text-center"><p>college medicine</p></th>
+<th class="head text-center"><p>medical genetics</p></th>
+<th class="head text-center"><p>professional medicine</p></th>
+</tr>
+</thead>
+<tbody>
+<tr class="row-even"><td class="text-center"><p>LLaMA 33B</p></td>
+<td class="text-center"><p>39.2</p></td>
+<td class="text-center"><p>40.3</p></td>
+<td class="text-center"><p>44.4</p></td>
+<td class="text-center"><p>32.9</p></td>
+<td class="text-center"><p>36</p></td>
+<td class="text-center"><p>43.0</p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p>Galactica 30B</p></td>
+<td class="text-center"><p>32.5</p></td>
+<td class="text-center"><p>26</p></td>
+<td class="text-center"><p>30.5</p></td>
+<td class="text-center"><p>25.4</p></td>
+<td class="text-center"><p>39</p></td>
+<td class="text-center"><p>23.1</p></td>
+</tr>
+<tr class="row-even"><td class="text-center"><p>Galactica 120B</p></td>
+<td class="text-center"><p>58.5</p></td>
+<td class="text-center"><p>59.2</p></td>
+<td class="text-center"><p>68.7</p></td>
+<td class="text-center"><p>57.2</p></td>
+<td class="text-center"><p>68.0</p></td>
+<td class="text-center"><p>59.6</p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p>OPT 175B</p></td>
+<td class="text-center"><p>28.9</p></td>
+<td class="text-center"><p>21.9</p></td>
+<td class="text-center"><p>30.6</p></td>
+<td class="text-center"><p>-</p></td>
+<td class="text-center"><p>35.0</p></td>
+<td class="text-center"><p>27.9</p></td>
+</tr>
+<tr class="row-even"><td class="text-center"><p>BLOOM 176B</p></td>
+<td class="text-center"><p>37</p></td>
+<td class="text-center"><p>29.8</p></td>
+<td class="text-center"><p>28.5</p></td>
+<td class="text-center"><p>-</p></td>
+<td class="text-center"><p>36.0</p></td>
+<td class="text-center"><p>25.4</p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p>Gopher 280B</p></td>
+<td class="text-center"><p>56.3</p></td>
+<td class="text-center"><p>67.2</p></td>
+<td class="text-center"><p>70.8</p></td>
+<td class="text-center"><p>60.1</p></td>
+<td class="text-center"><p>69.0</p></td>
+<td class="text-center"><p>64.0</p></td>
+</tr>
+<tr class="row-even"><td class="text-center"><p>GPT3.5 175B</p></td>
+<td class="text-center"><p>56.3</p></td>
+<td class="text-center"><p>69.8</p></td>
+<td class="text-center"><p>72.2</p></td>
+<td class="text-center"><p>61.3</p></td>
+<td class="text-center"><p>70</p></td>
+<td class="text-center"><p>70.2</p></td>
+</tr>
+<tr class="row-odd"><td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+<td class="text-center"><p></p></td>
+</tr>
+<tr class="row-even"><td class="text-center"><p>Task-tuned LLaMA 33B (LoRA)</p></td>
+<td class="text-center"><p>51.8</p></td>
+<td class="text-center"><p>65.2</p></td>
+<td class="text-center"><p>70.1</p></td>
+<td class="text-center"><p>58.3</p></td>
+<td class="text-center"><p>65.6</p></td>
+<td class="text-center"><p>66.5</p></td>
+</tr>
+</tbody>
+</table>
+</div>
+</section>
+<section id="instruction-tuning">
+<h3>Instruction Tuning<a class="headerlink" href="#instruction-tuning" title="Link to this heading">#</a></h3>
+<p>Instruction Tuning is a technique used to improve the performance of language models by training them to follow natural language commands or instructions. This includes positive or negative examples, prompts, constraints, and other elements that are commonly found in human language.   The main goal of instruction-tuning is to improve the model’s ability to perform well on multiple tasks and to generalize more effectively to new or unseen tasks. This is achieved by teaching the model to understand and incorporate various language cues and constraints that are relevant to the task at hand.  Instruction-tuning is a powerful technique that is widely used in natural language processing, machine learning, and related areas. By improving the ability of language models to understand and follow natural language commands, this approach can help to unlock new levels of performance and productivity in a wide range of applications.</p>
+<p>We list some examples below. Full example list is saved as a <a class="reference external" href="https://github.com/OptimalScale/LMFlow/blob/main/docs/source/_static/check_before_after_lora_tuning.jsonl">Jsonl file</a>.</p>
+<p><img alt="Instruction Tuning Sample" src="_images/IT_sample1.png" /></p>
+<p><img alt="Instruction Tuning Sample" src="_images/IT_sample2.png" /></p>
+<p><img alt="Instruction Tuning Sample" src="_images/IT_sample3.png" /></p>
+<p><img alt="Instruction Tuning Sample" src="_images/IT_sample4.png" /></p>
+<p><img alt="Instruction Tuning Sample" src="_images/IT_sample5.png" /></p>
+<p><img alt="Instruction Tuning Sample" src="_images/IT_sample6.png" /></p>
+<p><img alt="Instruction Tuning Sample" src="_images/IT_sample7.png" /></p>
+</section>
+</section>
+<section id="installation">
+<h2>Installation<a class="headerlink" href="#installation" title="Link to this heading">#</a></h2>
+<p>This package can be be installed from sources with the following command:</p>
+<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>git<span class="w"> </span>clone<span class="w"> </span>https://github.com/OptimalScale/LMFlow.git
+
+conda<span class="w"> </span>create<span class="w"> </span>-n<span class="w"> </span>lmflow<span class="w"> </span><span class="nv">python</span><span class="o">=</span><span class="m">3</span>.9<span class="w"> </span>-y
+
+conda<span class="w"> </span>activate<span class="w"> </span>lmflow
+
+conda<span class="w"> </span>install<span class="w"> </span>mpi4py
+
+pip<span class="w"> </span>install<span class="w"> </span>-e<span class="w"> </span>.
+</pre></div>
+</div>
+</section>
+<section id="checkpoints">
+<h2>Checkpoints<a class="headerlink" href="#checkpoints" title="Link to this heading">#</a></h2>
+<p>We have prepared tuned LLaMA model (both task and instruction tuning).</p>
+<p>Refer to <a class="reference external" href="https://github.com/OptimalScale/LMFlow/blob/main/README.md">README</a>.</p>
+</section>
+<section id="content">
+<h2>Content<a class="headerlink" href="#content" title="Link to this heading">#</a></h2>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="blogs/index.html">Blogs</a></li>
+</ul>
+</div>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="examples/index.html">Examples</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="examples/index.html#data-preparation">Data preparation</a></li>
+<li class="toctree-l2"><a class="reference internal" href="examples/index.html#finetuning">Finetuning</a></li>
+<li class="toctree-l2"><a class="reference internal" href="examples/index.html#inference">Inference</a></li>
+<li class="toctree-l2"><a class="reference internal" href="examples/index.html#evaluation">Evaluation</a></li>
+</ul>
+</li>
+</ul>
+</div>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="autoapi/index.html">API Reference</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="autoapi/lmflow/index.html">lmflow</a></li>
+</ul>
+</li>
+</ul>
+</div>
+<div class="toctree-wrapper compound">
+<ul>
+<li class="toctree-l1"><a class="reference internal" href="about/index.html">About</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="about/changelog.html">Changelog</a></li>
+<li class="toctree-l2"><a class="reference internal" href="about/authors.html">Contributors</a></li>
+</ul>
+</li>
+</ul>
+</div>
+</section>
+<section id="vision">
+<h2>Vision<a class="headerlink" href="#vision" title="Link to this heading">#</a></h2>
+<p>Hello there! We are excited to announce the upcoming release of our code repository that includes a complete LLM training process, enabling users to quickly build their own language models and train them effectively.</p>
+<p>Our code repository is not just a simple model; it includes the complete training workflow, model optimization, and testing tools. You can use it to build various types of language models, including conversation models, question-answering models, and text generation models, among others.</p>
+<p>Moreover, we aim to create an open and democratic LLM sharing platform where people can share their checkpoints and experiences to collectively improve the skills of the community. We welcome anyone who is interested in LLM to participate and join us in building an open and friendly community!</p>
+<p>Whether you are a beginner or an expert, we believe that you can benefit from this platform. Let’s work together to build a vibrant and innovative LLM community!</p>
+<p><a class="reference external" href="https://discord.gg/u9VJNpzhvA"><img alt="Embark" src="https://img.shields.io/badge/discord-LMFlow-%237289da.svg?logo=discord" /></a>
+<a class="reference external" href="https://join.slack.com/t/lmflow/shared_invite/zt-1wju9nicy-woXbNtS~5MavHSAtiMxmxQ"><img alt="slack badge" src="https://img.shields.io/badge/Slack-join-blueviolet?logo=slack&amp;amp" /></a>
+<a class="reference external" href="https://s1.ax1x.com/2023/08/06/pPAQTPI.jpg"><img alt="WeChat badge" src="https://img.shields.io/badge/WeChat-Join-brightgreen?logo=wechat&amp;amp" /></a></p>
+</section>
+<section id="citation">
+<h2>Citation<a class="headerlink" href="#citation" title="Link to this heading">#</a></h2>
+<div class="highlight-bibtex notranslate"><div class="highlight"><pre><span></span><span class="nc">@misc</span><span class="p">{</span><span class="nl">lmflow</span><span class="p">,</span>
+<span class="w">  </span><span class="na">author</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="s">{Shizhe Diao and Rui Pan and Hanze Dong and KaShun Shum and Jipeng Zhang and Wei Xiong and Tong Zhang}</span><span class="p">,</span>
+<span class="w">  </span><span class="na">title</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="s">{LMFlow: An Extensible Toolkit for Finetuning and Inference of Large Foundation Models}</span><span class="p">,</span>
+<span class="w">  </span><span class="na">year</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="s">{2023}</span><span class="p">,</span>
+<span class="w">  </span><span class="na">publisher</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="s">{GitHub}</span><span class="p">,</span>
+<span class="w">  </span><span class="na">journal</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="s">{GitHub repository}</span><span class="p">,</span>
+<span class="w">  </span><span class="na">howpublished</span><span class="w"> </span><span class="p">=</span><span class="w"> </span><span class="s">{\url{https://optimalscale.github.io/LMFlow/}}</span><span class="p">,</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+</section>
+<section id="disclaimer">
+<h2>Disclaimer<a class="headerlink" href="#disclaimer" title="Link to this heading">#</a></h2>
+<p>This package aims to provide a streamlined and user-friendly pipeline for large model tuning. Its functionalities serve as a reference and are intended for use by the user. However, it is important to note that the responsibility for the preparation of the data and pretrained models lies solely with the user. This package does not guarantee the accuracy, completeness, applicability, or legality of the components from the user’s preparation. Users must be aware of and assume all risks and liabilities associated with the preparation of the models and data, and obtain legal, commercial, and technical advice before utilizing this package. The pipeline shall not be held responsible for any direct, indirect, special, incidental, or consequential damages resulting from the user’s improper preparation of the data and pretrained models.</p>
+<p>It is also crucial to highlight that the results generated by the model are based on probabilistic models and not directly related to this pipeline. The accuracy, reliability, applicability, and legality of the results are not guaranteed by this pipeline. Therefore, users must also be aware of the risks and liabilities associated with the results and seek legal, commercial, and technical advice before relying on the model-generated outcomes. This pipeline shall not be accountable for any direct, indirect, special, incidental, or consequential damages resulting from the user’s reliance on the model-generated results.</p>
+</section>
+<section id="support">
+<h2>Support<a class="headerlink" href="#support" title="Link to this heading">#</a></h2>
+<p>If you need any help, please submit a <a class="reference external" href="https://github.com/OptimalScale/LMFlow">Github</a> issue.</p>
+</section>
+<section id="indices-and-tables">
+<h2>Indices and tables<a class="headerlink" href="#indices-and-tables" title="Link to this heading">#</a></h2>
+<ul class="simple">
+<li><p><a class="reference internal" href="genindex.html"><span class="std std-ref">Index</span></a></p></li>
+<li><p><a class="reference internal" href="search.html"><span class="std std-ref">Search Page</span></a></p></li>
+</ul>
+</section>
+</section>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+    <a class="right-next"
+       href="blogs/index.html"
+       title="next page">
+      <div class="prev-next-info">
+        <p class="prev-next-subtitle">next</p>
+        <p class="prev-next-title">Blogs</p>
+      </div>
+      <i class="fa-solid fa-angle-right"></i>
+    </a>
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+                <dialog id="pst-secondary-sidebar-modal"></dialog>
+                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">
+
+
+  <div class="sidebar-secondary-item">
+<div
+    id="pst-page-navigation-heading-2"
+    class="page-toc tocsection onthispage">
+    <i class="fa-solid fa-list"></i> On this page
+  </div>
+  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
+    <ul class="visible nav section-nav flex-column">
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#introduction">Introduction</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#features">Features</a><ul class="nav section-nav flex-column">
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#task-tuning">Task Tuning</a></li>
+<li class="toc-h3 nav-item toc-entry"><a class="reference internal nav-link" href="#instruction-tuning">Instruction Tuning</a></li>
+</ul>
+</li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#installation">Installation</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#checkpoints">Checkpoints</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#content">Content</a><ul class="nav section-nav flex-column">
+</ul>
+</li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#vision">Vision</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#citation">Citation</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#disclaimer">Disclaimer</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#support">Support</a></li>
+<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#indices-and-tables">Indices and tables</a></li>
+</ul>
+  </nav></div>
+
+  <div class="sidebar-secondary-item">
+  <div role="note" aria-label="source link">
+    <h3>This Page</h3>
+    <ul class="this-page-menu">
+      <li><a href="_sources/index.md.txt"
+            rel="nofollow">Show Source</a></li>
+    </ul>
+   </div></div>
+
+</div></div>
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/objects.inv b/objects.inv
new file mode 100644
index 0000000000000000000000000000000000000000..a2c3b4eeb524ec3454add262e4f183a8ea301775
GIT binary patch
literal 12128
zcmV-mFQ3pOAX9K?X>NERX>N99Zgg*Qc_4OWa&u{KZXhxWBOp+6Z)#;@bUGkRO-5{Q
zcM2mQR%LQ?X>V>iAPOTORA^-&a%F8{X>Md?av*PJAarPHb0B7EY-J#6b0A}HZE$jB
zb8}^6Aa!$TZf78RY-wUH3V7PxeOr^`D7Nl<{t8E&^E&S7ntdX6>^E1YGt1G=OpooV
zp4u-2DPdbXbt{na(!YKKl5E*V5+D$o^U&2*7Od|B7K_Dg5k|W(d7q@8{}m;E9-_(L
zCos<vm<GRv{L^ni?4ytAA&bKQQzZG@|DF8pruaAo;0<9CBryO?h7U71`MpI#Ei1%(
z(uT{vw}Z13(;3Y*!}~l!amM9&AtdCQ#HF}aInzDLKpLbd3}OV>cTk9Bd{K5te06}Y
zDE1B!#IHfTH$hk^Z^~MLQ1UU15EBlw3<ocHhD-s35kPM+%wfh_omK`8yq-XJtLIr5
z>|=yY)Qdpc^aWA9G){8LDgsnBp2F;43xi}RB^HG5G4uoaHvl~^kMgi+RO@Ko0eP=!
z62uu*t-*jmP7uI|{*GasS%3`@#4!WHFhe%vCmz)WO8(V~@IBaJ^vjOuJ*yU5dKF?o
zXiKx<0fX!_wNdvG`~V?}_g4Bi6Q1HEN;A6{wi+NwvmgrovRC^_o~3yP`~X`Jo>JaE
zdJ8<nM>g;RGw%QhEstF=YKDT?E*eH@f-{gMHe~0q2`@t=n!q^7J`3w@Mf@AG))xpu
zz6G)G5J!1xjrBVU_6OR*kOx2ei+}rK)}OnUH>C~>qUN3sw7^T^T|ft~AVfB(8*&~6
zf(u&@QWV25`?Ld*5MszE3#8M@0Ug-s;0r@r0OYkC87;{-H>x*?H$1KjP^d%bClN4X
zcm(nZjM9+#Zz&8g@<BNP?g>@Qnv58T3sF-5(u8Cnpo2)UETKDPzmohyUGP^UgY~6k
zYQ)$|N-+V;5Up_obBHUDoMZ`e)O_GUI=Y&0s4g+8>2PV4O?U`R5H@9-n9X0w4hVb+
z5!mB|4&@p8$cSl(nrMTb<t0i$K%q+2;)iK~SA<f)*4|z)g0IGFiP`01U4@x8WJP}A
z-CF`ksI7R%0dsy>KQ=8srSk@yc!&^Y_TxMP?+JdT=E*b=$!PmV1{Q%X1!ss3046-b
zE5W>3?MbHA-;BR=DW0l(;1|P-4fZwhgnA3qfFkVCFL2)%YF?~1Cq^a{*TSYSChule
zE(F4TYRE|1-!l`4E?40ZFy@p9K!SLOFp51?8%@Az4Z}R3u{BPCY-W~Y^bWG*6~)AY
zd=r}?BDui__1LIU_5!xmW*)%U@|r+)Isc)SKKHU5TTmHu^Pd*shkUmSZIDMP$^vdT
z&8!0mFxYg0EtOcQeYUA`T#QLD*wOz`mCO#@J8hGfo}%xmbD8CamZ+|U)u&o~b=)}J
zD+Ew&sGEf`y~inl*p)v~@i;RtKBUtws-ju&Mn_8@gz&S!C))%Q#2Lb-c5ItKiV}d6
zWaBWULisQPZ1wYZ>j+<LAy}nvo%ug(3HS3T`m|d)CkSg*6ea9md1Onz!Cesn21XH`
z$AdhkW10`VB+R4Og2~{!T_KG3Ih~8K$>skbFYE6{3Xs^?Y=7KiD9jA@l)R{BHby)E
zfi!c=&?4c`_qo@|hngfU<}@JiiTP@o!ZZ!pZY<19txSthOih)=vJK2ug@OZ8%Jac#
zuFO`<f|z8G{=-It_({32XhAlegs?SgYV#5w)yjks$-~Sdpr|G-z=>8gEWj9B#}ElO
zx{x6RFuLL(LSS8_Ty`MnSB@<86mx3>Lg}U_?_v<v@ra4Jmq<K!4@wLR;?1o+h^5CQ
z!ED|L!w`kGIx`z!UoFW^Ey%rCld**Gf*|Wz5PJtp2ybB+SrE;{f_<8Uh<f1o6XaPC
z27kfKj-}&h1C2Ty%SNZjm?i~p$l$rIMww@O@XXcEY_Xp*-#nn>bU;&3=lR1Fkl7mz
zZDN@(=%)#Q?01Xw84evK(j=z+a*tx3GO>+Fyae~(9d1Q2w)fGBg3&v~K8UDK9a=F4
zI^d@E4IiKH=1Xw<{B*x~0JqE8W;5S3-v*PDw*9&4C4`mLMBh6XWNuD>A(suJCbpn2
z#J&W*ow^H&rp}FBiZymxOe~lkn8AUM4A%xTib;z23E2L2>PmRdCdR+={<oDbB%_2)
zCw$*HWomd5Of5?Q33$nOyZ#$37XV=b+42DR531RwC1XfO(ewZ<$n~j~(1FH+<ie;M
zm>Y_OV+%4f`i1&-#cf{0+EG-7u|wTL>C8j{?q1uGkI|cnH5Z|5^8-w2MW`3y8<xz?
zo(0C-UdP!h;0~+>*=b-yG+(BH`#nbcVi%kCC9?>8G0}oZelO7cz@8<{W%A7{NQU#+
zW9RC!1lTnwB@#a~X}teX;nY!OdyM>wlxgv7dB!)cgG}lUk_;R*$AWx15je2J6Z9A3
zc9sDIarU3)rG{22`<ETD7)GGF6wGWD?7RB@n1@+FfnaFx;!v~DX%Ml^a!69-WoFEI
z7uGHb{r>9Sf<i#VzDDe;Js5)6CDqT~TazsiypQoZBh;D90E+u+*K>7${fQ1!eKhJ^
z&~$b;`#N7P=J!VVak0<epYDuHy_+w$GviVp)-&@|AI(dg{mZ<}m3fhWg3ax>`Q7*B
z9Ne$xbK?L{W+mQze|-FDR^)Q__|>$;^~S8kC-XY`^!(%5d|z6dC%SpKTN#&VQ|mX+
zt8WX#hVSR|!=k6e)nY|QkEc19tslNW&Y!m6akg45o*wkq#n*6`8Qhr?iJN*!7#2qi
z*{u&cliK$^79oVOdo@W2$qubx`ho8kn%MsmqLe#Q#%7(ZAVF*&q}+cpuwo`e;Voip
zCrt>!4r)}TFm<nS&E@gQ)7I2Qlpg^xw)1r*tGsbiEQ<6je53GU@zWAUF>9bUgZ_+7
zuH;@rg&}-{eCFGKlch;|icxWC*w`k32&rsv%hi><v*QIxj(c0~YvENr@ykC~=iu!W
zsJ!-AY?oR=__a%LWLu5-vhxsn#wPXkIx`MX8Ww_>Z&F;LGLG>cWCu{TOHbh}W?~^k
z&(SL`;a67zcX?dg`B0Hl9>Z!4(p#lz57b+ws=@}ej7lPbEs~~yxi9TUF_cB}xO4o@
zBfHHS!OW!e7wElquUKhHm5j%tD&50~ZHVkJi}(_FM(cxDv&7X^9l4TMR+U9#>JFaW
z?=V#vR%)1Sx7W1Jg4nv+nG}Df8zu$ss8fb;ii6lV=W0XJl+45Kc>DS?ET$=l2&N{*
zovtRClW3>l%)2ouFH4MfUMjLW0c}hq_vLtmB<>{`mEFA^FDtTcQ@Xs5@-LEJNcorP
z-E2GC%(q=<P1{gf#mhT8dXkl#A6*DNXGjxG(K+geqUtPZqG>r#npn!tR1X|g=Smk!
z-PzJa8l;6(ow5EfO3qmiI5lUj2au}s)&ok~nbXAR<J@UN$~t?R;uV}fZG>SeO4>Qp
zl&|6}>O)97kGc>#&ZI7uf^(?|@yGlpSUhdEv!`1dt{@Y?xqV*G?TBU$ta-^@OOUE^
z&rYBL_y&WJMSf^spOl>~aLJ$$tuk{E<dd952y#wu`6kxY*)^FdKnIbY4Cqv0&q5WT
zNOHPhosRs>#WEFTpLqyA)yOacJ_^TEhoXeho(^j)Xm9<P_)v<(SllLI$U-aJCwP;m
z2)_jc`NQ}2E-JL2c`yuz?P#5(WDob7CUy#I4k)__7IX4)FFmAwjO0)>0G+qropGs+
z(VWY$L{3+R3cGW>m1L9x57WtFr<Jdls=i&yH^N&tMvD~fu0)Y$%5t6#ymSaWC03Np
zA2@xiP#Wx0D~O!*X?eS|EKP&WQ3QISZ7E$vr^HfKbV@$`O8+WZMPiU^=|vnhEKU9_
zy%hhm9MikvUy`M0M7xHmPEDy6NSIHmV8Wc!dGV`SIXg*QqqC(M#iCwjkt*e7rs=nP
zwX(ILD#KBS;Vk%GNB=dA&4wX}*%i7p&j^ntJ~BGz(VQG`a!jE3I5=$4fd_}>&w!&{
z2H452B+pLwB?qP9uDK(G_r=}QG8i%$oWiJVLf4|O70}8jwh!F3<fn_G>pV1Nl^t2d
zF2uKcEhj>TOkWmC_H_t05?5)4B&0G9*k@DUWtE-s3`Xqy{?J%hNv+roNlI}nu}?^M
z$hFT*e(<x;MQuzp$wz*S)I%*}&IBLDdSv$qS<4_hF4nTp`sFqHSOInpf(>{|Ky)Hg
ze&Q^tN5^jC$Tg-R{K~KNyAF9+LoSjO5egj=(D10O@!-b^${TYIY3NL{t&ri-M$46X
zCp7r%*P)6QVc;ARP;^+#pdVi=lXkt0;8w7X4TFzHLquu%V+J!RjNrhiq}IoVgVmfs
z8N*hcBN+pBVO8k2^R33=+WGj+g_0E?Bu@Ak#=RIihXcMGGlxT#J=tL#RyG!>VPLg~
zR!o3{+4t>p5hA}c7?5GS_EV<)!IsbK8Q9L&5A!Xch~Jm<%@AdjpU<lxWaxthzS(2y
zxSs_y>7LIEv&zowg_&tPZt3sq<InOd-koInvQXb~9)gn21kgUO(Gg<P)9T!q_$kC9
z1QZz~;qj8}z2j-0qx@R1eJ*Ni!6y00j~yDQ{E*@fR<|Cz4jNX8rACTV@@baG)in%x
zj53oZ)^vNqY#)KzA+NDCMqI6t_DU0!$5u66&~+#DeF?F`Yq=2`x-X5cG$sy11hAvZ
za#I1TgpRi)&Y4wVE6vwc5n}5LLBFK+^D>K+YF=h4?}@M@$Ullt8}v(3H`@sGLDq<8
zXqyrBBUr>qiMF>!p?o+Hx#gg~=4h3S%37c`B8R~DQ7jWc)i#r3h{iGWnR7D54V_as
zNtX@AUqZW#nCOa3Ii*U@Ztbvl+M*y1_G$JXF3DAaGDp;S$Dw&F4Oqtc90!?l%-kXV
zY<L;6a|DD&X9E=CjrO2Yonw-A=c)_}&Rmsc`jQ8sUwKVg@?O9;U;gU^?nLy$Ak{Bo
zJ-1GI>-nd*%f+f_<84)P8Sk}|OfMGG(?(Lt9=OBVr&~U!UHOslsrH>lLMM-+&&qg?
zhgR`>;U5rbqiLCx<`83>mEuIrGM$U5!8p_@AUwFf><OtsIB}O;ZAxjG5qq#CWiMkC
zU2Ysrcd@+@!WSxm?{b_!P&bAOEW^dutm#N}O(Bl`khJQKo6MuHHo2K`B_+7<moD2}
zN1mYFJLE~G&}W??K)Y^f$D~Sfr!yaZJt8zXmn{G?lBWQ$a2y#szu6jirZ0|<s`W&l
zJ=RP1U4ubiz+Lb`p?u`QPs>P{y-l*SkJ1qrv@Zb3P-;_taY2e-u*uWTH%C>`lDrJk
z+Sw6VO4)wM(()oy;yZicrSxvG8eLEWgvUS$F_Erz4Geh=k`Yb~d<LbOl&6$zQl6Gk
zK8o*Z(ktoRV&!W>O86c_Io)OnCF_ADWqcn>*<H18DY{Fhd=<z@j--k&J8VacWF%AS
zsFN5MC12QXf$&4Ftnfp&wD5zIEChF!_*`zgKy}llCSJZ;CB&w3^_t37Ysy)!q$)M-
zW>-jKG5ld1qmcT-=v}VtD#8;I#0Zce3POm3?9(^`pXAsfe|iSav}k!f>z4Tr2%W<7
zi?Tb6ket$8WRg|Ei)_3!e{Ra3?o0Ge<!k}hL!F|x_EiC!*z#>@9{YG1%s6-~rAM)(
z;>3)5ETbH8CowKceISq(Os72=#2KK!8<i`qCK3XRwJ?yOnC+@pt<}7Ep(L9c>^)Ty
z)57`AL$^}#WHMUC6Hf0csjYf-CA3YNWJ`6+3}N)rU0{|<ro1fObXkgQE$yYmW>H5^
zw{wD~Ad_;7Qa}xIj!cX8(Mte{X{mDUEg=vSndZhsuDLOpE@$hFL~~=(NsNn9G@1&6
zWvj45U@SNDie;x>JmK_7YVs9#PL^uY6l+pVwp?XO$V}c6xtV;GQmRhNw9_a07pcj-
zNP4$e)hgBlk*sSO;rvf_Qd7bToYKlvIhN}dShid6gwwTDx4@lhR+FaKW@@tKnyOnS
z+54(9rmHM6wS1Lnx_P)$nk%)!s^dw{l1!t@wVP>F$(fNdY$~NKsicCWjQ2`L5z<a#
zT$Ji^j*?{cwHy`kDhoL(a#dDw6y!>)b%($tJp>ZF4n;aCwc;tK8~(qCZr!>P+#ylX
z$g9hjtXXBbdFnIH2X<!(#zqi6lYP+Lb}W@l%dvFRpQXswM!S^QEK0IATLh*-ibC4)
z+sav`X)k4!s(2bgN$yz=9Tffs#i|C`P`}-XYEVU#@On`j+Q9_yYO$Ix7f<u<$QVY^
zp62tr4VeFHyPj=KL-K887zX>e{Q&k+UOj`PqSUkSXFg^pye>6qf>Xh%-cb&A#n<DF
z>n>F>0?vQ41E<u*&vK}=-0@$mR4QG(5wQ^}%azoSCs!#7&%n#E%~cfe3P={Arh4_O
z{!*Ht56b>+7{^Z&B{NY=FWEq~9Mikivx1Fiz>)`<f*=mEfPVN36#)R-?}M5|=I53r
zUWqn{DQ}SsVbOVqL<Bg4sIt#W8iZt%f5}Jf4Wdei*E5J&ygEEgQMq>q1B}l4Nb_cy
zCf?f@RpWNbA#dJJd02J6eNi)UVnj8ldT8A;*mLKr@#Z;rF!TUMBu!#=hLMK=bwh#O
z^x@A7l`KGtVi>lsyqG7$pKWJNQEGbNe-sy83xZ@^Y`_kZY*0!{<CQr?^-;?RwwJ_j
z2(!o(0?Rj84d{%ZMu4?OG$Y6uRZ%8gpu08#k0?s;=LYW3fMb&wbeIm`hCN)@?_jYD
zLc}f+9^6qG(h@XK7_?V5a2VV=tWg;_jE9?;;e8&VxQLV4u>jZuv@52f9tN8m<yDO4
zPVf4dTGF#Vrk~F1I9+B-2qP(f3d$dO1nR<rya+|zX}V0~`v|2ZMeRXROrytPI3!$}
z4gnXjOWXEwUc$tCl?nP~22Ls@y&))!SZgs8!s7%C&>@sZsa|N6LfS9a@wed$N7WLB
z0K@oo5IFTgwHMrbf<xa<uiMc-<aaQRsU?TGWBTcR9j9xU6T(Q!uR3xOPLo!d4~{46
zq*T8bxN?{7g?9XHILJLt@{|XswqKc4OR{1wDouho8>D>dJ-Zj2^sbi%inL!WE&CW3
z7)u`fsFF*Xi&KW_;tZ<GV2F{VAr$?L-e_8`MqgBA)6^GGD$urhG(2b_)g6z%i0T%q
zFCxjP!b-kwVhmK>=g~skq{z!dm<4ZW0Fxs`<435f!0B)3xS3(En#DAH3}xRRvj;zr
zc8e(tpy6WR6eB<IG7z&E`aBl$&<wcJ4PI{ICH%!yb|Cv@aPYy*+BQkT0j$e8%#?@E
zL?6f-s$kQ>t8yw6^1UCRv=^-<9kVB}QjkK-j&gxK+x@5HVG%7jlBa2cv(x@8s6x@2
zD_q9bli4_?di+k9Q+DPc$~#?Dzp9<h>m<^P!4$9^vaIH(cJ~y^g!sqml7E>miAO=r
zPpTs?g=e5>V(hakYwehHEW9$RT__u6tx;=7&XSF%Clu7g8aFnU;uana%hB`(ozoGP
zDXFi;*yW=!>#|Eo@8p7YQZ=(MB3|zaekCdN)K1+PBqi@7>tOPA0ToBGo<Y)otY?|7
z=|2?Q#U?lvoay4}Ubf+k-~lg0FqYrQF$3xc#A97ZVYR^waPfv)#vXkyCq8`P)+vK5
zNe9H&ye(t~T=~L~L0JB3!0_Pm?vFV*MJLAyu(B&-1XX<u)`pcH+>{V%>yDCxT0MvC
zs8-K2U32$^pUhNSjBJF~#GC^0+wXF{#XK431lS|<pn7G#Q0?M{c~^wq!Z31PEspWM
zBud$q90&vAs3FvCt7!(ikiw$aVI4Tu0?lIu#(AipGcZm;W6U*8LH$rb4+`=FzaobG
zdCMxX{CHm+{i@ZTbkGP9mJ*2kVK!wbm8OuAoM%Buq?&-Il&pWAw!EFfh#p@#RzMQu
z?u$NM;p$$_9sT__2}5=?_|G6atf23gM@_o)LyDeZUepz#W-n}yDdv`;$A3fPB^Z$E
zH8TF}k>C0Jn_Ig4bw7|ozj_ngA2r<wlGpMb?2+^QDB}f;o@V)~_`eC;au@F3fvw6}
z87B<W(`Ynyj#9zw-7@9&ldbm_%g!i7>@x0n2dawixW}gq=ds7n$MBUJhg_wdVk6k~
zO#>BAw#=vD!<IRVkQ3Vzyo#eLK-Yog@twki5ar%uioj|6h|vI)0}i|~&~eR#4!rdU
zNUD9&6pg$*(ZrYRywYwAOi0ThDB(ps&IBNE$4gs9Q&2Dp4HA^hLxYSS(I=$2sgqJ{
z6e}E}GVZ;^v0}q41RDwJdU7sATUE}*^l6x8LRw6EHIELL6zwvN0ll70sTJxOrz`r8
z4Sb`%)GZ9d7wEmp59&<}w4qt(xHpx>8}kg<)A4D-9Q4?CO=Jv}QH1IoyfTy>Ommop
z5?<VNw^nj~OhLW#*Rv^g{(8pg-TZ#`eYpj<v*q&Z?Dmh1wEb<uR|H<IpC8uq&F09U
zH-W(ayBvkZ25e>z^T*jA^EG%}fz|V73w{T`|9z-BcwTK6kBk4N=ufl9`B13r*p71N
zGKV<@bm9(3#LmWM2*-q7H{1sqb!QT+V(d&(+QaO?c9dfu#CenkyH8g{P<EWH6d}1V
z^)23z&wkCLm^kFcqf-|rMX^H~#Yr6C2QEnzgQDBYIHZ6eh^ePSP?o2TSOhUU=6H+*
z9VUr0CWPb&yhF^05I{dsV(iGNBR>4J>kA^~l^~Zaco0}Ef5teKWAG<#ZL4?PuuDpq
z>MegO%NYqa^Ch(-;fi~#Cdowb^93<Vg%hk4rOpDIVrEUhd9SJ)5k`5?F}OSlO5GJ7
zy#=1DDv7B@q5eSOvN+<9#O198NBHbG0`xFjc`r9lkSouqdJBPHYZwCuXBCl?!$=02
z!e<6~oD=7sGi{orh{S*N*IY44_<NA2)^SxOimO_tq222zXF+E%uHfkehOjs6vYZ-?
zvU{-S#bShf<jP&XV&rbLCq+ya_dTVUW)>T}DyOV1Fv;1o!lURt?--^or&MhH%ByXa
zIAEY~=HEOx1@c%VSQ0_Z<I}NA%XHol!@Vnqlpjj20&fJFI}%JMz=6(eLx-~!560B(
zpr$lFo{@3GBZ;U#ic}BkSYj{|O)VMKj_EjRn>fQs=nlV#xm)yE7K6Z1a}+exB9dEY
zF@~kFl9Ou|ImRf?0L|pCuk<)WW|BqyZ5B5S60paRX(kMnkA)lLq4P|t6%y_iMoV;d
z>e=0n3r$QZu;=F3Y=D|CN2C0X_^C5JIPQ!phL5*G<TQeKA>pIQ>$`OPIV+4Zw*NJT
zsyV$l?2^eL1u=}=&Erpd#%!ii9_n(#<TFJ$g_za`+oyH8GRmfm><9{9t93*nMIXoB
z>vC};VB3(Vy&Wg2K{*uR6{-*{)SxBKs-uW;;GRlx5tmsHcb;&V?A;Vz<F2G>biVGG
zI&{C|JOb#0A31U~pr;`VYFP$Eifh~8PU<5~QpfAC<uegZ!QD;)Y9vU8+>xlvGc`~_
zM!+t?&R5v@emL9q&0pM9OSuc~wu=;7MoAuL4j05Z35QM<IY8{QM;;u`Q;!GS9HFy^
zWvW%Y;afoSF1t|%Y}WNZK@}jLtr<9~3wH;%PLpt#OLxb^%I=0Nc0C!=IiIsSU<AB0
z2h^Q(%OUrna}xY;!^LTj!N_cGXUl=F0Bvr+&F{Vsjk;aW7Ecs&vz=|{!y_-A9>8q<
zuo)B?1cXl6d4&D4w*mMD13H%v(NG)$cEySfK6}2UhXHkt(XSkaAUnW#&>L?@Dj{>$
zEBN1sf>*p^-V=?y1}WGDn3SP85V46IX-7D`%&K;q^dfT^g|c^C!!}Z;*yuzv9E6u{
zaJWIp6`=q@l!sZ6h6p6PagtPzTs8=^@?chmVw#qWkf82V$qYAD6RnG?ap6AFyKWzy
zg%~L6Uj6=cq&yrnqH3SZ)8{aJsa=;?aL!LUoJ<c``q!$A7U0ff+sJjEi#(Ej=OS0_
zA#!G_o_Fl%iL67>!k`bd`fC8Th@gZh2j{iLMeTV2>n!jn8mN5tsCp3+BSM`g;<Sd;
z9}lp_gfGrJshmfCxp)J2=*M#TsP+brCFo)MPd5w9-bPk?$FL`-!JFcu=sli7GEBrJ
zxJztz*7AZ#`K)hJ#ChQdUD<gx7gPic2JedwEw%KlP!6Y@;+|erZ*XGk>DnFFddlfx
zSH0{|+jN$JxXjoCbNUh{-fQV0H%|qm;iYlCBz{Af6lc?qA<vHt24>TxIOY-QSKtJ7
z=!)id=`h-}H`%J_>q@FF6riT^LJ^v(-$_TC2Q}IOE)EWkf)4*ohc)*9>LEFo#7vqD
z^XybRvv~%E`ham))1^|4kfDZUw!&hlYUzAbP2w%JVujI5qNv?fDnu3O<R$W03almg
z91k#;mIX7~1t0CT(uCFoQ(k8Ek~BayK(ioSx}_>Fg0BdW0}6|Enj*H6CGII^H~3y)
zT&~ZXlU$xcwcE%s$zT<wr9(l>q;fi1CZ66tKW*2~OLk)B+kCd(g0J)07R>*(TC8W=
z#q(2lqYGf$@qWzL8!j8%&9<}6d<*X8o7?qbWgU+ev^bu#F%2Uq33!~XjKh?%r5BI0
zhdEfywzuEF_W6(blU*Ro5McgvXB}g+W!h~8<$n20`^@vx^5+m5y?A(fUeCee>2Cfn
z<FZf9Z3*Csw!_oy9NaJF%e#$@KC*nI?eQmAKCfqhw)q2XjmKwpCXl187Ecex5uYFK
z7HiW|ALozH>z`mVyPt!*)$@-pV78=fPdm#G@ME^7tpwII$&a~BV75+1HQ|1~p0ieS
ziG9C#T5P_7`{m4d1FmL^_52PDJvNoF+xfqkVJiMV0PK3c{l0#prM;sKx?Vr8!PoEi
zw3bcdu=0Evs!$@Z?QF5!*auoZKRuL1vsJRe1{V>SwQ4*E8loC$FWcNcSDkLT*bJtd
zRR<pm^Uvq?ALl;(;%Pfy|ClX}A=Q>_P?R2@rtX@hTxXM<v0P`I-hP{HAC`~*pjG?0
zTF$mI<54@l0`TK<c-Y#n+ghbJY$V$Up8}93B;&CO+fN%0!KHY8d>ky!JB3p^goU0B
zd{C{3J7mLO$VrhGum@K%h_m)^HG?xs($qa8LIQ4q*=6sDza`YMD?R|JQmN}X99I*q
z91j8VNoX?n^;0eOjj7F&7lQB`92+e@U{qlZ*bWeJtV!iUZpcIbqSSp2T@Zq}3eV`U
zlFoRztscPuG8RI|@xw78ZWYJW>8r0YLE?s>lO!P-Q~l(Gok$%GpZf%FlOl++v$;jd
zVH|G#1(}g}_Z2Ir-r3!JzS_*^e+)hwrf}Pz;OQA`R`c7%Y*}nvY#i|z8uelR_&6IJ
z_1n+KufwDAwX3J)<Cnpam&@7X%n|HAT*HF<XSUb#yc`OD^=&bHp;zlUTa2w>H$&n6
z^UwLy@S1gwI(Q5JT>K9H-{6J*@A>N6&-LK2FOX=jiL<==3*#)e$<^XZHhpQBSvNM%
zs-LTAi%)NF(*`jUtIEcC8C+ks%&pf-vw3bt*PM;hGdRC7&dty|w0T}8SEP-zGrcx#
zoS?Z?YU9+*tykNo)^Fw7I6D(-*yc$YTg|pet?#<Fd1~fXxQ%l&wbpH(n6XuF<J63;
zf1763ZzbFko3Tw)Os|gH;x`gM($2zle?|`Rqcr=xuXq1}JKH${_sbI(d<x6xf{4pT
z8CChG2S_IZ?46iH68diD?k6&J3+rH{Tm~d_sTpVZ4z*)LYd+q)tceQ-?k(3PA$N^a
zjsCYulqNBXGY8Fc#ld|8i72smhQ`B&WTPR&pH`1EM`re#PXx*9Xvk!rsjC4<HY9;T
zT<-w0eHiI(=#|;^_w-5urY^k*_$UyG!ans7h5f1_3eDWWx#l8Cf<IY~(5r4GRQdI*
zp^{+g_?3u}B$YjiBB)yvJhOf^@a#s;LR}G2W5E72xZ!SzjmZ^|eiZu{a9!lXXY``2
zs64a2qVnuUj?Y~YQDeaVG^nVS*qBsQU5WvF+A-E2@vHp$N>fQNcmVR6k|dWfg!)#V
zXV$L=p54q@%qu2+DD*r&XMoE;+*ijxMp7N@&o2fon=LwhH*ak5>bKKnnU&j0>*K8G
zBBzHy*V@vo`nHy4HgOL7f`cUZuXhydmYU_&FE-0<<Vf}<4SoEc=2Vl&r?%RC_RzY|
zz{t_?PaUQE3#ErotXt|Lw|>cs{AP}*Uvp6;!T)D)47Eq7?|{&kS)U<60-Ha<BpK%R
zlvYLyPpQkN#2#6o-Q!~jYyS2Dd&y=Ir+N35i5R>mDxLV0*quwucJPxtvq#|U@d^9&
z-3Dj1qFaHj)-a68JHh~cWEe_>qSl4skKcyDU&xMl{#*>VE>^@>hns^4?h)99aBl~P
zpbTIb`$+@{>Uu0d350Sc2Nrf2#a$1c6v8et$e7EEEbOfTT9DqzLNUhpOj#J}6k^1F
zib1t$*j4SyBsJOX63k5Sm?+WeVn=&G@s-!wJGE7(bg>(@`!xH~U)w}dbd*gb=cH@Z
zTPw#;+KSi3SC429wX;L0Qab+8l+Kx;3SyKP(K?YJO~n(5n)a!sE^gYS`stZ9S~YFs
z-wi5UZ_a#?tiQ$y#pvjpP>>Bv*U4CuY7ZT>MXA!i;;Sj`>(7?XJ~DO-v9I!p<>+Xi
zSkm;K{d-31>7Le6=R{GRo)KPjQsli>s@J|S<BDOeZ;X!MA3s*K{*XV`Ca&`vcT=A+
z9dncabL?&Y&wrmr|L53S9lX5z;1%Jxcu`Vyv>SvNNMd&ICItLl#SzkEZdkc^<tHGa
ze`F(MNwS2wim$+b8yUNk#|7%Wd7KP;c(OJLj<g%sE3O<)+V-cM5PN+~xs4t#!(St%
z$S)0L3HS@)WFWnx^Q$*_{z7XQymZfw#Q9`qL*krg_qsd1q;7N_*eOg?6#F2eBM;n5
zY&~0|;U(#(GjKofGH2j>lok6`#3=>peMG6dS}668NbR0l;*jii1Z6H6;3RZg1+uZ1
z?7@F=j8A>s&k#8!@fxHcMjuW!Q6c!7B?i<d*kkAi31C;+UJ`e<iT$k{p$DvtO%>T|
zQz_&??!h^5aUdRiaa#K6<7(@cz-l>LJh`Q?Sv)-~=WaP{|9l>p0Q;fbU-{BCYcY>X
zzq!OtLm0$Pkl+692o^+9Rvg<J9{hJluujTcdiw(+R)>Yjz_@RLm&XoNy22e$t$ehD
zd6-?@F*S@Hv6IH#3no07%=!pVIQ0(5dj*FiO7<v5LG~%L9(0A#AYATnyM#1ay#Y-I
zWy*g{D^&@{)KlogG%L*mr^Hssio~FbP|{g3Lhv?JemJ)rcreN^9ecpBrOm_Tsm~&c
zdO69Gi0R)h*c(Jj6B2M`<`;z1B!qHjAjJ7(8c>|GJUW7t6anJF5c#xH<5%<vQg+r_
zn^3!0Ykb+%$VV|j>m-Bta~{1Q-w)z_=LDs9G3@e}-&q}dK_d782>Jy=6ax=Zqv}h!
z>|-xhoV^0fB>aC-nl$(~EX?_FH-6I1E6XQixXl-%T$w4&PrDgCrIoRbr_@sruxR<-
zE4xxc$HpLpFQHU<#fXwn#5NXR^#7!WE(Wy)F7r6FNVRQ9%Y`bzUI)M_7boolo&22v
zz$!&Af2EB^#}P;}Yk(Mj4p6KN4^eq<VHnwlk(<YK%B^+zCbL;4X}6YTqA56P3~1&i
zkjO;Amn6@AqiGKbCKHnR|K%s%<Zm>4{R2-wq2p;WSkhn0jNSppdqnHyI_vRMnYsFB
znUnLl&UK<UnXw;ks$yKHuYV|WtQM0sDsPD1TvPzWd(tfBDp4V^tZDZ1r94*61ZAZ9
zOQ&pchBKThN%faTOlhoUj7Hhq=7GZKDVMD=v$&js<d?j<ps%D~3gFXd!c;i{wKawo
zkqtS{icqAPp3*eHvgn{{u$MVRu%|Q)u+^0kC1gP+5%^iM2K*wL0tQ+<S~83H2$MJP
zI_&f)iSvorT&8R$(yNZz^EVJvtf^x8D+)e>Sb^46t_Cu-bj=hp^pOFK_81X9!l~9#
zOa2C0!W<0*C3!T#HA<yZv>l!7DFB@35Ebo|rW-cPW9CI^z@O!8r!g(lHqg;+bos`X
zY~)@lsHlG`%J&wqBwyHalM*FyJ`tPCl&wT6<MNvq2I=u1N}O6mG}EinP3xKDnCg*}
z<a0?ON$E`TH|WC`6b5M52`)^32Fp`Sk13h~lRWkn@J`7a;8=o2G+GrIlt-&27o+h6
zjc9BoKpyOrT8zcMH$p}7I8hr&)QkW>6fmm94Jh!V87EbMNn26UW{`J4*aA46w1OQ*
zNC8rqvw^@x+D1CDJbv5;7>g17Jc>RgA!=y_82qakL{5BzV6n&$MlX_bPe}!=*~caX
zOr#MoX+gk5j1Vhy&bYOA#?9R`PTmp)V{@Kd1XU&wqwzn3zmRm*U1ks<o|1?$u|hTC
z)@sDf)kyZfmRV@XCjh@n*+?Xn$4=Ye8f+j*(nBCS64n!oQ21vt&S$a@BvDRD1sLq(
zB|;jc)VGLHu@-XtR~K4awln1F)ATH1BMZLl0mEP)OL%WBxx^gm>Dno%SA-y23kkW3
zpz|l?YN+LD!RM(O!TmH*Z-lE%V)$CJHvG3QS`03;HF7v5Y-EAnV3@Phx$+DyvWU@-
zi5ii2^bf_hzsMp+J|=2Jmd0yFX^~wn++!7(^E~2mol-S}XNVmMctb!-xm&U~P!eTm
zWQXA{QzvzuOU$L7u9-pvbQ(yIt${>6VH*oY#Xrx{h~2LCWfVzn7T7LD87)Ai>vxC{
z6Dwn1UT2bG$`27DO^#=oyV;e=J9sHF>B8<ni%^u{Pe9;~_p-{QtFmYn(^fML85LOb
zJ?&$+=tEp2NtXM0-VXXHqQ*p1^RDx@QLLx#AjK`5CaLSZZKUd{yGTX-Y>7C3E4fI?
ziZHXNS!a<#jYXaHb*_L4Y=jxbWG<3ka><fc6bFA5hvt;`5G8V-v8t!<+)um|uD4yG
zAf=V6)CR^Gm=d#g0yhmo^P@-t=Q{BjP8GwSeQv^uNd{wy(;33_LbNi)jn;FdG7P%k
zGo47jp?ahHSg0n=>jy_C@~@uWXuj&%gKHFOP>sSH?MF8?xe8wHoG3gXdZYL7q9)nP
zryD1-4@TbTJ@}_d^YY2SiR_DpH%gE3=@V3<cBB15pf<a!X9gz<&mG?AJ+Gxn_VR(Q
ziR`mrH#*N@X_7pDPHQ6fj@OOWYf+jEFP@v4NWOV>qw%_vCc{tFD*RNc!TPH{nhdWV
z2AW7eNOU9hkVYpj;maG7hNnGl`aXitNk{cTg^Bvp4>vud>MNI_BS8N|`5l3q{$bj6
z3egeteWDYj{zm@U7G=tk)NcBQ9akpS6~cU?9&G)lcQ`=hQgp@^oM=Qm6qiGeU<2}v
z^1c2}q|TrK6NzvHH(ht^J83jc`D-`(Z=^1XcO!{zl}{S(uHPu%9Ph;Mymmg3xWIX1
z`rf53ZRtCmlm6FIZ=`NOb(TbUWopuJ4eO@uW>X8)^$n=W`9-Q5I-u{#Z!akJ4iUt!
zby*jmitNkBdzyXnwfgnHCbM`lqo@(BJ8v>Cc5>NOuX_w5^qx?ey97_3QxyMgljPV#
zljZV}l)zU7sQp|3eq}%0pbtoWHAH^9n{8*C`F2y!bM>*vc~@SpIew-e{{Ro)7k6`E
z?$c*+Q6~?1;G>f`9zP^asO}J<yxRBbKG^YxniL+7BKQjdp>Had+@7aIs}fDl;=Wd|
zf1VhWd*JL=vk2_;xsZE*{P#)3RWqJ{DKgH>-%i;g<Oj5afod0CDB&{E8H-vm*gk|%
z0Q*p6TF>scwG8|f%VN2<sqhzRN-_M$$z%EBNg~P9G{G4)81ik|k*7f1REZkea*yIu
z6NKz=G{dNVEo3a%UvI8js+OWq_#0YSfw1)RZ`Fq)^<#05vk+VR+L3L=?|`4@CJ18h
WaGtUFtt@`=eFgXE=6?aGR=6Bf2yLnW

literal 0
HcmV?d00001

diff --git a/py-modindex.html b/py-modindex.html
new file mode 100644
index 000000000..3263fcb78
--- /dev/null
+++ b/py-modindex.html
@@ -0,0 +1,921 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="./" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Python Module Index &#8212; LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="_static/doctools.js?v=9bcbadda"></script>
+    <script src="_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'py-modindex';</script>
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="search.html" />
+
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+
+
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="search.html"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none"></div>
+              
+              
+              
+                
+<div id="searchbox"></div>
+                <article class="bd-article">
+                  
+
+   <h1>Python Module Index</h1>
+
+   <div class="modindex-jumpbox">
+   <a href="#cap-l"><strong>l</strong></a>
+   </div>
+
+   <table class="indextable modindextable">
+     <tr class="pcap"><td></td><td>&#160;</td><td></td></tr>
+     <tr class="cap" id="cap-l"><td></td><td>
+       <strong>l</strong></td><td></td></tr>
+     <tr>
+       <td><img src="_static/minus.png" class="toggler"
+              id="toggle-1" style="display: none" alt="-" /></td>
+       <td>
+       <a href="autoapi/lmflow/index.html#module-lmflow"><code class="xref">lmflow</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/args/index.html#module-lmflow.args"><code class="xref">lmflow.args</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/datasets/index.html#module-lmflow.datasets"><code class="xref">lmflow.datasets</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/datasets/dataset/index.html#module-lmflow.datasets.dataset"><code class="xref">lmflow.datasets.dataset</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/datasets/multi_modal_dataset/index.html#module-lmflow.datasets.multi_modal_dataset"><code class="xref">lmflow.datasets.multi_modal_dataset</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/index.html#module-lmflow.models"><code class="xref">lmflow.models</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/auto_model/index.html#module-lmflow.models.auto_model"><code class="xref">lmflow.models.auto_model</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/base_model/index.html#module-lmflow.models.base_model"><code class="xref">lmflow.models.base_model</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/decoder_model/index.html#module-lmflow.models.decoder_model"><code class="xref">lmflow.models.decoder_model</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/encoder_decoder_model/index.html#module-lmflow.models.encoder_decoder_model"><code class="xref">lmflow.models.encoder_decoder_model</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/hf_decoder_model/index.html#module-lmflow.models.hf_decoder_model"><code class="xref">lmflow.models.hf_decoder_model</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/hf_encoder_decoder_model/index.html#module-lmflow.models.hf_encoder_decoder_model"><code class="xref">lmflow.models.hf_encoder_decoder_model</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/hf_model_mixin/index.html#module-lmflow.models.hf_model_mixin"><code class="xref">lmflow.models.hf_model_mixin</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/hf_text_regression_model/index.html#module-lmflow.models.hf_text_regression_model"><code class="xref">lmflow.models.hf_text_regression_model</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/interfaces/index.html#module-lmflow.models.interfaces"><code class="xref">lmflow.models.interfaces</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/interfaces/tunable/index.html#module-lmflow.models.interfaces.tunable"><code class="xref">lmflow.models.interfaces.tunable</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/regression_model/index.html#module-lmflow.models.regression_model"><code class="xref">lmflow.models.regression_model</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/text_regression_model/index.html#module-lmflow.models.text_regression_model"><code class="xref">lmflow.models.text_regression_model</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/vision2seq_model/index.html#module-lmflow.models.vision2seq_model"><code class="xref">lmflow.models.vision2seq_model</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/vision_encoder/index.html#module-lmflow.models.vision_encoder"><code class="xref">lmflow.models.vision_encoder</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/models/vision_encoder/clip_encoder/index.html#module-lmflow.models.vision_encoder.clip_encoder"><code class="xref">lmflow.models.vision_encoder.clip_encoder</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/index.html#module-lmflow.optim"><code class="xref">lmflow.optim</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/adabelief/index.html#module-lmflow.optim.adabelief"><code class="xref">lmflow.optim.adabelief</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/adabound/index.html#module-lmflow.optim.adabound"><code class="xref">lmflow.optim.adabound</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/adadelta/index.html#module-lmflow.optim.adadelta"><code class="xref">lmflow.optim.adadelta</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/adagrad/index.html#module-lmflow.optim.adagrad"><code class="xref">lmflow.optim.adagrad</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/adam/index.html#module-lmflow.optim.adam"><code class="xref">lmflow.optim.adam</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/adamax/index.html#module-lmflow.optim.adamax"><code class="xref">lmflow.optim.adamax</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/adamp/index.html#module-lmflow.optim.adamp"><code class="xref">lmflow.optim.adamp</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/adamw_schedule_free/index.html#module-lmflow.optim.adamw_schedule_free"><code class="xref">lmflow.optim.adamw_schedule_free</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/adan/index.html#module-lmflow.optim.adan"><code class="xref">lmflow.optim.adan</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/dummy/index.html#module-lmflow.optim.dummy"><code class="xref">lmflow.optim.dummy</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/lamb/index.html#module-lmflow.optim.lamb"><code class="xref">lmflow.optim.lamb</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/lars/index.html#module-lmflow.optim.lars"><code class="xref">lmflow.optim.lars</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/nadam/index.html#module-lmflow.optim.nadam"><code class="xref">lmflow.optim.nadam</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/novograd/index.html#module-lmflow.optim.novograd"><code class="xref">lmflow.optim.novograd</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/optimizers/index.html#module-lmflow.optim.optimizers"><code class="xref">lmflow.optim.optimizers</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/radam/index.html#module-lmflow.optim.radam"><code class="xref">lmflow.optim.radam</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/sgd_schedule_free/index.html#module-lmflow.optim.sgd_schedule_free"><code class="xref">lmflow.optim.sgd_schedule_free</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/sgdp/index.html#module-lmflow.optim.sgdp"><code class="xref">lmflow.optim.sgdp</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/sophia/index.html#module-lmflow.optim.sophia"><code class="xref">lmflow.optim.sophia</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/optim/yogi/index.html#module-lmflow.optim.yogi"><code class="xref">lmflow.optim.yogi</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/index.html#module-lmflow.pipeline"><code class="xref">lmflow.pipeline</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/auto_pipeline/index.html#module-lmflow.pipeline.auto_pipeline"><code class="xref">lmflow.pipeline.auto_pipeline</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/base_aligner/index.html#module-lmflow.pipeline.base_aligner"><code class="xref">lmflow.pipeline.base_aligner</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/base_pipeline/index.html#module-lmflow.pipeline.base_pipeline"><code class="xref">lmflow.pipeline.base_pipeline</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/base_tuner/index.html#module-lmflow.pipeline.base_tuner"><code class="xref">lmflow.pipeline.base_tuner</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/dpo_aligner/index.html#module-lmflow.pipeline.dpo_aligner"><code class="xref">lmflow.pipeline.dpo_aligner</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/dpov2_aligner/index.html#module-lmflow.pipeline.dpov2_aligner"><code class="xref">lmflow.pipeline.dpov2_aligner</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/evaluator/index.html#module-lmflow.pipeline.evaluator"><code class="xref">lmflow.pipeline.evaluator</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/finetuner/index.html#module-lmflow.pipeline.finetuner"><code class="xref">lmflow.pipeline.finetuner</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/inferencer/index.html#module-lmflow.pipeline.inferencer"><code class="xref">lmflow.pipeline.inferencer</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/iterative_dpo_aligner/index.html#module-lmflow.pipeline.iterative_dpo_aligner"><code class="xref">lmflow.pipeline.iterative_dpo_aligner</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/raft_aligner/index.html#module-lmflow.pipeline.raft_aligner"><code class="xref">lmflow.pipeline.raft_aligner</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/rm_inferencer/index.html#module-lmflow.pipeline.rm_inferencer"><code class="xref">lmflow.pipeline.rm_inferencer</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/rm_tuner/index.html#module-lmflow.pipeline.rm_tuner"><code class="xref">lmflow.pipeline.rm_tuner</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/utils/index.html#module-lmflow.pipeline.utils"><code class="xref">lmflow.pipeline.utils</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.html#module-lmflow.pipeline.utils.dpov2_dataprocessor"><code class="xref">lmflow.pipeline.utils.dpov2_dataprocessor</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/utils/dpov2_trainer/index.html#module-lmflow.pipeline.utils.dpov2_trainer"><code class="xref">lmflow.pipeline.utils.dpov2_trainer</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.html#module-lmflow.pipeline.utils.memory_safe_dpov2_align"><code class="xref">lmflow.pipeline.utils.memory_safe_dpov2_align</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.html#module-lmflow.pipeline.utils.memory_safe_vllm_inference"><code class="xref">lmflow.pipeline.utils.memory_safe_vllm_inference</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/utils/peft_trainer/index.html#module-lmflow.pipeline.utils.peft_trainer"><code class="xref">lmflow.pipeline.utils.peft_trainer</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/utils/raft_trainer/index.html#module-lmflow.pipeline.utils.raft_trainer"><code class="xref">lmflow.pipeline.utils.raft_trainer</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.html#module-lmflow.pipeline.utils.rm_dataprocessor"><code class="xref">lmflow.pipeline.utils.rm_dataprocessor</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/utils/rm_trainer/index.html#module-lmflow.pipeline.utils.rm_trainer"><code class="xref">lmflow.pipeline.utils.rm_trainer</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/pipeline/vllm_inferencer/index.html#module-lmflow.pipeline.vllm_inferencer"><code class="xref">lmflow.pipeline.vllm_inferencer</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/tokenization/index.html#module-lmflow.tokenization"><code class="xref">lmflow.tokenization</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/tokenization/hf_decoder_model/index.html#module-lmflow.tokenization.hf_decoder_model"><code class="xref">lmflow.tokenization.hf_decoder_model</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/tokenization/hf_text_regression_model/index.html#module-lmflow.tokenization.hf_text_regression_model"><code class="xref">lmflow.tokenization.hf_text_regression_model</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/index.html#module-lmflow.utils"><code class="xref">lmflow.utils</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/common/index.html#module-lmflow.utils.common"><code class="xref">lmflow.utils.common</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/constants/index.html#module-lmflow.utils.constants"><code class="xref">lmflow.utils.constants</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/conversation_template/index.html#module-lmflow.utils.conversation_template"><code class="xref">lmflow.utils.conversation_template</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/conversation_template/base/index.html#module-lmflow.utils.conversation_template.base"><code class="xref">lmflow.utils.conversation_template.base</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/conversation_template/chatglm/index.html#module-lmflow.utils.conversation_template.chatglm"><code class="xref">lmflow.utils.conversation_template.chatglm</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/conversation_template/chatml/index.html#module-lmflow.utils.conversation_template.chatml"><code class="xref">lmflow.utils.conversation_template.chatml</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/conversation_template/deepseek/index.html#module-lmflow.utils.conversation_template.deepseek"><code class="xref">lmflow.utils.conversation_template.deepseek</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/conversation_template/gemma/index.html#module-lmflow.utils.conversation_template.gemma"><code class="xref">lmflow.utils.conversation_template.gemma</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/conversation_template/hymba/index.html#module-lmflow.utils.conversation_template.hymba"><code class="xref">lmflow.utils.conversation_template.hymba</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/conversation_template/internlm/index.html#module-lmflow.utils.conversation_template.internlm"><code class="xref">lmflow.utils.conversation_template.internlm</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/conversation_template/llama/index.html#module-lmflow.utils.conversation_template.llama"><code class="xref">lmflow.utils.conversation_template.llama</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/conversation_template/phi/index.html#module-lmflow.utils.conversation_template.phi"><code class="xref">lmflow.utils.conversation_template.phi</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/conversation_template/qwen/index.html#module-lmflow.utils.conversation_template.qwen"><code class="xref">lmflow.utils.conversation_template.qwen</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/conversation_template/yi/index.html#module-lmflow.utils.conversation_template.yi"><code class="xref">lmflow.utils.conversation_template.yi</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/conversation_template/zephyr/index.html#module-lmflow.utils.conversation_template.zephyr"><code class="xref">lmflow.utils.conversation_template.zephyr</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/data_utils/index.html#module-lmflow.utils.data_utils"><code class="xref">lmflow.utils.data_utils</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/flash_attention/index.html#module-lmflow.utils.flash_attention"><code class="xref">lmflow.utils.flash_attention</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.html#module-lmflow.utils.flash_attention.bloom_flash_attention"><code class="xref">lmflow.utils.flash_attention.bloom_flash_attention</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.html#module-lmflow.utils.flash_attention.gpt2_flash_attention"><code class="xref">lmflow.utils.flash_attention.gpt2_flash_attention</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.html#module-lmflow.utils.flash_attention.gpt_neo_flash_attention"><code class="xref">lmflow.utils.flash_attention.gpt_neo_flash_attention</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.html#module-lmflow.utils.flash_attention.llama_flash_attention"><code class="xref">lmflow.utils.flash_attention.llama_flash_attention</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.html#module-lmflow.utils.flash_attention.triton_flash_attention"><code class="xref">lmflow.utils.flash_attention.triton_flash_attention</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/llava_conversation_lib/index.html#module-lmflow.utils.llava_conversation_lib"><code class="xref">lmflow.utils.llava_conversation_lib</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/model/index.html#module-lmflow.utils.model"><code class="xref">lmflow.utils.model</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/multimodal/index.html#module-lmflow.utils.multimodal"><code class="xref">lmflow.utils.multimodal</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/position_interpolation/index.html#module-lmflow.utils.position_interpolation"><code class="xref">lmflow.utils.position_interpolation</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.html#module-lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch"><code class="xref">lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/utils/versioning/index.html#module-lmflow.utils.versioning"><code class="xref">lmflow.utils.versioning</code></a></td><td>
+       <em></em></td></tr>
+     <tr class="cg-1">
+       <td></td>
+       <td>&#160;&#160;&#160;
+       <a href="autoapi/lmflow/version/index.html#module-lmflow.version"><code class="xref">lmflow.version</code></a></td><td>
+       <em></em></td></tr>
+   </table>
+
+
+                </article>
+              
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/search.html b/search.html
new file mode 100644
index 000000000..3d89290ce
--- /dev/null
+++ b/search.html
@@ -0,0 +1,439 @@
+
+<!DOCTYPE html>
+
+
+<html lang="en" data-content_root="./" >
+
+  <head>
+    <meta charset="utf-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><title>Search - LMFlow  documentation</title>
+  
+  
+  
+  <script data-cfasync="false">
+    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
+    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
+  </script>
+  <!--
+    this give us a css class that will be invisible only if js is disabled
+  -->
+  <noscript>
+    <style>
+      .pst-js-only { display: none !important; }
+
+    </style>
+  </noscript>
+  
+  <!-- Loaded before other Sphinx assets -->
+  <link href="_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+<link href="_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
+
+    <link rel="stylesheet" type="text/css" href="_static/pygments.css?v=8f2a1f02" />
+    <link rel="stylesheet" type="text/css" href="_static/graphviz.css?v=4ae1632d" />
+    <link rel="stylesheet" type="text/css" href="_static/plot_directive.css" />
+    <link rel="stylesheet" type="text/css" href="_static/sphinx-design.min.css?v=95c83b7e" />
+  
+  <!-- So that users can add custom icons -->
+  <script src="_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
+  <!-- Pre-loaded scripts that we'll load fully later -->
+  <link rel="preload" as="script" href="_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
+<link rel="preload" as="script" href="_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
+
+    <script src="_static/documentation_options.js?v=5929fcd5"></script>
+    <script src="_static/doctools.js?v=9bcbadda"></script>
+    <script src="_static/sphinx_highlight.js?v=dc90522c"></script>
+    <script src="_static/design-tabs.js?v=f930bc37"></script>
+    <script>DOCUMENTATION_OPTIONS.pagename = 'search';</script>
+  <script src="_static/searchtools.js"></script>
+  <script src="_static/language_data.js"></script>
+  <script src="searchindex.js"></script>
+    <link rel="index" title="Index" href="genindex.html" />
+    <link rel="search" title="Search" href="#" />
+  <meta name="viewport" content="width=device-width, initial-scale=1"/>
+  <meta name="docsearch:language" content="en"/>
+  <meta name="docsearch:version" content="" />
+  </head>
+  
+  
+  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
+
+  
+  
+  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+  
+  <div id="pst-scroll-pixel-helper"></div>
+
+  
+  <dialog id="pst-search-dialog">
+    
+<form class="bd-search d-flex align-items-center"
+      action="#"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+  </dialog>
+
+  <div class="pst-async-banner-revealer d-none">
+  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
+</div>
+<aside class="bd-header-announcement" aria-label="Announcement">
+  <div class="bd-header-announcement__content">We've released our memory-efficient finetuning algorithm LISA, check out [<a href='https://arxiv.org/pdf/2403.17919.pdf'>Paper</a>][<a href='https://github.com/OptimalScale/LMFlow#finetuning-lisa'>User Guide</a>] for more details!</div>
+</aside>
+
+  
+    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
+<div class="bd-header__inner bd-page-width">
+  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
+    <span class="fa-solid fa-bars"></span>
+  </button>
+  
+  
+  <div class="col-lg-3 navbar-header-items__start">
+    
+      <div class="navbar-item">
+
+  
+    
+  
+
+<a class="navbar-brand logo" href="index.html">
+  
+  
+  
+  
+  
+  
+    <p class="title logo__title">LMFlow</p>
+  
+</a></div>
+    
+  </div>
+  
+  <div class="col-lg-9 navbar-header-items">
+    
+    <div class="me-auto navbar-header-items__center">
+      
+        <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+      
+    </div>
+    
+    
+    <div class="navbar-header-items__end">
+      
+        <div class="navbar-item navbar-persistent--container">
+          
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+        </div>
+      
+      
+        <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+      
+        <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+      
+    </div>
+    
+  </div>
+  
+  
+    <div class="navbar-persistent--mobile">
+
+<button class="btn search-button-field search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
+ <i class="fa-solid fa-magnifying-glass"></i>
+ <span class="search-button__default-text">Search</span>
+ <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd class="kbd-shortcut__modifier">K</kbd></span>
+</button>
+    </div>
+  
+
+  
+</div>
+
+    </header>
+  
+
+  <div class="bd-container">
+    <div class="bd-container__inner bd-page-width">
+      
+      
+      
+        
+      
+      <dialog id="pst-primary-sidebar-modal"></dialog>
+      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar hide-on-wide">
+        
+
+  
+  <div class="sidebar-header-items sidebar-primary__section">
+    
+    
+      <div class="sidebar-header-items__center">
+        
+          
+          
+            <div class="navbar-item">
+<nav>
+  <ul class="bd-navbar-elements navbar-nav">
+    
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="blogs/index.html">
+    Blogs
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="examples/index.html">
+    Examples
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="autoapi/index.html">
+    API Reference
+  </a>
+</li>
+
+
+<li class="nav-item ">
+  <a class="nav-link nav-internal" href="about/index.html">
+    About
+  </a>
+</li>
+
+  </ul>
+</nav></div>
+          
+        
+      </div>
+    
+    
+    
+      <div class="sidebar-header-items__end">
+        
+          <div class="navbar-item">
+
+<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
+  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
+  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
+  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
+</button></div>
+        
+          <div class="navbar-item"><ul class="navbar-icon-links"
+    aria-label="Icon Links">
+        <li class="nav-item">
+            
+          
+          
+          
+          
+          
+          
+          
+          
+          
+          <a href="https://github.com/OptimalScale/LMFlow" title="LMFlow" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><img src="_static/logo5.svg" class="icon-link-image" alt="LMFlow"/></a>
+        </li>
+</ul></div>
+        
+      </div>
+    
+  </div>
+  
+  
+  <div class="sidebar-primary-items__end sidebar-primary__section">
+      <div class="sidebar-primary-item">
+<div id="ethical-ad-placement"
+      class="flat"
+      data-ea-publisher="readthedocs"
+      data-ea-type="readthedocs-sidebar"
+      data-ea-manual="true">
+</div></div>
+  </div>
+
+
+      </div>
+      
+      <main id="main-content" class="bd-main" role="main">
+        
+        
+          <div class="bd-content">
+            <div class="bd-article-container">
+              
+              <div class="bd-header-article d-print-none"></div>
+              
+              
+  <div class="bd-search-container">
+    <h1>Search</h1>
+    <noscript>
+      <div class="admonition error">
+        <p class="admonition-title">Error</p>
+        <p>Please activate JavaScript to enable the search functionality.</p>
+      </div>
+    </noscript>
+    
+<form class="bd-search d-flex align-items-center"
+      action="#"
+      method="get">
+  <i class="fa-solid fa-magnifying-glass"></i>
+  <input type="search"
+         class="form-control"
+         name="q"
+         placeholder="Search the docs ..."
+         aria-label="Search the docs ..."
+         autocomplete="off"
+         autocorrect="off"
+         autocapitalize="off"
+         spellcheck="false"/>
+  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
+</form>
+    <div id="search-results"></div>
+  </div>
+  <script>
+    // Activate the search field on page load
+    let searchInput = document.querySelector("form.bd-search input");
+    if (searchInput) {
+        searchInput.focus();
+        searchInput.select();
+        console.log("[PST]: Set focus on search field.");
+    }
+  </script>
+
+              
+              
+              
+              
+                <footer class="prev-next-footer d-print-none">
+                  
+<div class="prev-next-area">
+</div>
+                </footer>
+              
+            </div>
+            
+            
+              
+            
+          </div>
+          <footer class="bd-footer-content">
+            
+          </footer>
+        
+      </main>
+    </div>
+  </div>
+  
+  <!-- Scripts loaded after <body> so the DOM is not blocked -->
+  <script defer src="_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
+<script defer src="_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
+
+  <footer class="bd-footer">
+<div class="bd-footer__inner bd-page-width">
+  
+    <div class="footer-items__start">
+      
+        <div class="footer-item">
+
+  <p class="copyright">
+    
+      © Copyright LMFlow 2024.
+      <br/>
+    
+  </p>
+</div>
+      
+        <div class="footer-item">
+
+  <p class="sphinx-version">
+    Created using <a href="https://www.sphinx-doc.org/">Sphinx</a> 8.1.3.
+    <br/>
+  </p>
+</div>
+      
+    </div>
+  
+  
+  
+    <div class="footer-items__end">
+      
+        <div class="footer-item">
+<p class="theme-version">
+  <!-- # L10n: Setting the PST URL as an argument as this does not need to be localized -->
+  Built with the <a href="https://pydata-sphinx-theme.readthedocs.io/en/stable/index.html">PyData Sphinx Theme</a> 0.16.1.
+</p></div>
+      
+    </div>
+  
+</div>
+
+  </footer>
+  </body>
+</html>
\ No newline at end of file
diff --git a/searchindex.js b/searchindex.js
new file mode 100644
index 000000000..8d116d047
--- /dev/null
+++ b/searchindex.js
@@ -0,0 +1 @@
+Search.setIndex({"alltitles": {"1 Introduction": [[112, "introduction"]], "1. Decompose your conversations": [[108, "decompose-your-conversations"]], "1. NLL Task Setting": [[106, "nll-task-setting"]], "1.1 Dataset description": [[112, "dataset-description"]], "2 Reward Modeling": [[112, "reward-modeling"]], "2. Choose proper Formatter": [[108, "choose-proper-formatter"]], "2. LM-Evaluation Task Setting": [[106, "lm-evaluation-task-setting"]], "2.1 Supervised Finetuning (SFT)": [[112, "supervised-finetuning-sft"]], "2.2 Reward Modeling": [[112, "id1"]], "2.3 LoRA Merge and Get Reward Model": [[112, "lora-merge-and-get-reward-model"]], "2023": [[104, "id1"]], "3 RAFT Alignment": [[112, "raft-alignment"]], "3. Build the template": [[108, "build-the-template"]], "3.1 Algorithms Overview": [[112, "algorithms-overview"]], "3.2 Hyper-parameters": [[112, "hyper-parameters"]], "3.3 Examples": [[112, "examples"]], "3.3.1 SFT": [[112, "sft"]], "3.3.2 RAFT Alignment": [[112, "id2"]], "3.3.3 End Note": [[112, "end-note"]], "4. Register your template": [[108, "register-your-template"]], "5. Use your template": [[108, "use-your-template"]], "API Reference": [[3, null]], "About": [[2, null]], "Attributes": [[4, "attributes"], [5, "attributes"], [8, "attributes"], [13, "attributes"], [14, "attributes"], [15, "attributes"], [16, "attributes"], [46, "attributes"], [51, "attributes"], [53, "attributes"], [55, "attributes"], [56, "attributes"], [57, "attributes"], [58, "attributes"], [59, "attributes"], [60, "attributes"], [61, "attributes"], [63, "attributes"], [64, "attributes"], [66, "attributes"], [67, "attributes"], [69, "attributes"], [70, "attributes"], [71, "attributes"], [73, "attributes"], [74, "attributes"], [75, "attributes"], [76, "attributes"], [77, "attributes"], [78, "attributes"], [79, "attributes"], [80, "attributes"], [81, "attributes"], [82, "attributes"], [83, "attributes"], [84, "attributes"], [85, "attributes"], [86, "attributes"], [87, "attributes"], [94, "attributes"], [96, "attributes"], [97, "attributes"], [101, "attributes"], [102, "attributes"]], "Blogs": [[104, null]], "Changelog": [[1, null]], "Chat Performance": [[103, "chat-performance"]], "ChatGLM-3": [[114, "chatglm-3"]], "ChatML": [[114, "chatml"]], "Checkpoints": [[107, null], [115, "checkpoints"]], "Citation": [[115, "citation"]], "Classes": [[4, "classes"], [5, "classes"], [6, "classes"], [7, "classes"], [9, "classes"], [10, "classes"], [11, "classes"], [12, "classes"], [13, "classes"], [14, "classes"], [15, "classes"], [16, "classes"], [19, "classes"], [20, "classes"], [21, "classes"], [22, "classes"], [23, "classes"], [25, "classes"], [26, "classes"], [27, "classes"], [28, "classes"], [29, "classes"], [30, "classes"], [31, "classes"], [32, "classes"], [33, "classes"], [34, "classes"], [36, "classes"], [37, "classes"], [38, "classes"], [39, "classes"], [41, "classes"], [42, "classes"], [43, "classes"], [44, "classes"], [45, "classes"], [46, "classes"], [47, "classes"], [48, "classes"], [49, "classes"], [50, "classes"], [51, "classes"], [52, "classes"], [53, "classes"], [55, "classes"], [56, "classes"], [57, "classes"], [58, "classes"], [59, "classes"], [60, "classes"], [61, "classes"], [65, "classes"], [66, "classes"], [67, "classes"], [68, "classes"], [69, "classes"], [75, "classes"], [79, "classes"], [80, "classes"], [81, "classes"], [83, "classes"], [87, "classes"], [88, "classes"], [94, "classes"], [96, "classes"], [100, "classes"]], "CommonSense Performance": [[103, "commonsense-performance"]], "Conclusion": [[103, "conclusion"]], "Content": [[115, "content"]], "Contributors": [[0, null]], "Conversation": [[105, "conversation"]], "Conversation Template": [[105, "conversation-template"], [109, null]], "Create Your Task Dataset File": [[106, "create-your-task-dataset-file"]], "Customize Conversation Template": [[105, "customize-conversation-template"], [108, null]], "Data Format": [[105, "data-format"]], "Data preparation": [[110, "data-preparation"]], "Dataset": [[105, null]], "Dataset Format in General": [[105, "dataset-format-in-general"]], "DeepSeek": [[114, "deepseek"]], "Disclaimer": [[115, "disclaimer"]], "Evaluation": [[110, "evaluation"]], "Examples": [[110, null], [113, "examples"]], "Features": [[115, "features"]], "Finetune": [[111, null]], "Finetuning": [[109, null], [110, "finetuning"]], "Formatted Dataset": [[105, null]], "Full Parameters": [[109, "full-parameters"]], "Functions": [[4, "functions"], [6, "functions"], [7, "functions"], [23, "functions"], [24, "functions"], [33, "functions"], [50, "functions"], [55, "functions"], [63, "functions"], [64, "functions"], [68, "functions"], [70, "functions"], [71, "functions"], [73, "functions"], [88, "functions"], [89, "functions"], [90, "functions"], [91, "functions"], [93, "functions"], [94, "functions"], [97, "functions"], [98, "functions"], [100, "functions"], [101, "functions"]], "Gemma": [[114, "gemma"]], "Hymba": [[114, "hymba"]], "Indices and tables": [[115, "indices-and-tables"]], "Inference": [[110, "inference"]], "Installation": [[115, "installation"]], "Instruction Following": [[103, "instruction-following"]], "Instruction Tuning": [[115, "instruction-tuning"]], "InternLM2": [[114, "internlm2"]], "Introduction": [[103, "introduction"], [113, "introduction"], [115, "introduction"]], "LLaMA Checkpoint": [[107, "llama-checkpoint"]], "LMFlow": [[115, null]], "LMFlow Benchmark Guide": [[106, null]], "LMFlow Benchmark: An Automatic Evaluation Framework for Open-Source LLMs": [[103, null]], "Layerwise Importance Sampled AdamW (LISA)": [[109, "layerwise-importance-sampled-adamw-lisa"]], "Llama-2": [[114, "llama-2"]], "Llama-3": [[114, "llama-3"]], "Low-Rank Adaptation (LoRA)": [[109, "low-rank-adaptation-lora"]], "Merge LoRA Weight": [[109, null]], "Metric": [[103, "metric"]], "Mixtral 8x22B": [[114, "mixtral-8x22b"]], "Mixtral 8x7B": [[114, "mixtral-8x7b"]], "Module Contents": [[4, "module-contents"], [5, "module-contents"], [7, "module-contents"], [9, "module-contents"], [10, "module-contents"], [11, "module-contents"], [12, "module-contents"], [13, "module-contents"], [14, "module-contents"], [15, "module-contents"], [16, "module-contents"], [19, "module-contents"], [20, "module-contents"], [21, "module-contents"], [22, "module-contents"], [23, "module-contents"], [25, "module-contents"], [26, "module-contents"], [27, "module-contents"], [28, "module-contents"], [29, "module-contents"], [30, "module-contents"], [31, "module-contents"], [32, "module-contents"], [33, "module-contents"], [34, "module-contents"], [36, "module-contents"], [37, "module-contents"], [38, "module-contents"], [39, "module-contents"], [41, "module-contents"], [42, "module-contents"], [43, "module-contents"], [44, "module-contents"], [45, "module-contents"], [46, "module-contents"], [47, "module-contents"], [48, "module-contents"], [49, "module-contents"], [50, "module-contents"], [51, "module-contents"], [52, "module-contents"], [53, "module-contents"], [55, "module-contents"], [56, "module-contents"], [57, "module-contents"], [58, "module-contents"], [59, "module-contents"], [60, "module-contents"], [61, "module-contents"], [63, "module-contents"], [64, "module-contents"], [65, "module-contents"], [66, "module-contents"], [67, "module-contents"], [68, "module-contents"], [69, "module-contents"], [70, "module-contents"], [71, "module-contents"], [73, "module-contents"], [74, "module-contents"], [75, "module-contents"], [76, "module-contents"], [77, "module-contents"], [78, "module-contents"], [79, "module-contents"], [80, "module-contents"], [82, "module-contents"], [83, "module-contents"], [84, "module-contents"], [85, "module-contents"], [86, "module-contents"], [87, "module-contents"], [88, "module-contents"], [89, "module-contents"], [90, "module-contents"], [91, "module-contents"], [93, "module-contents"], [94, "module-contents"], [96, "module-contents"], [97, "module-contents"], [98, "module-contents"], [100, "module-contents"], [101, "module-contents"], [102, "module-contents"]], "NOTICE": [[114, null], [114, null], [114, null], [114, null]], "Package Contents": [[6, "package-contents"], [8, "package-contents"], [24, "package-contents"], [81, "package-contents"]], "Paired Conversation": [[105, "paired-conversation"]], "Phi-3": [[114, "phi-3"]], "Qwen-2": [[114, "qwen-2"]], "RAFT": [[112, null]], "References": [[103, "references"]], "Returns": [[15, "returns"]], "Reward Modeling": [[113, null]], "Setup": [[106, "setup"]], "Step 1 Supervised Finetuning (SFT)": [[113, "step-1-supervised-finetuning-sft"]], "Step 2 Reward Modeling": [[113, "step-2-reward-modeling"]], "Submodules": [[6, "submodules"], [8, "submodules"], [17, "submodules"], [18, "submodules"], [24, "submodules"], [35, "submodules"], [54, "submodules"], [62, "submodules"], [72, "submodules"], [81, "submodules"], [92, "submodules"], [95, "submodules"], [99, "submodules"]], "Support": [[115, "support"]], "Supported Conversation Template": [[114, null]], "Supported Dataset and Detailed Formats": [[105, "supported-dataset-and-detailed-formats"]], "Task Registration": [[106, "task-registration"]], "Task Tuning": [[115, "task-tuning"]], "Text2Text": [[105, "text2text"]], "TextOnly": [[105, "textonly"]], "Version 0.0.1 (Mar 28, 2023)": [[1, "version-0-0-1-mar-28-2023"]], "Vision": [[115, "vision"]], "Work in Progress": [[105, null], [114, null], [114, null]], "Yi": [[114, "yi"]], "Yi-1.5": [[114, "yi-1-5"]], "Zephyr": [[114, "zephyr"]], "lmflow": [[8, null]], "lmflow.args": [[4, null]], "lmflow.datasets": [[6, null]], "lmflow.datasets.dataset": [[5, null]], "lmflow.datasets.multi_modal_dataset": [[7, null]], "lmflow.models": [[17, null]], "lmflow.models.auto_model": [[9, null]], "lmflow.models.base_model": [[10, null]], "lmflow.models.decoder_model": [[11, null]], "lmflow.models.encoder_decoder_model": [[12, null]], "lmflow.models.hf_decoder_model": [[13, null]], "lmflow.models.hf_encoder_decoder_model": [[14, null]], "lmflow.models.hf_model_mixin": [[15, null]], "lmflow.models.hf_text_regression_model": [[16, null]], "lmflow.models.interfaces": [[18, null]], "lmflow.models.interfaces.tunable": [[19, null]], "lmflow.models.regression_model": [[20, null]], "lmflow.models.text_regression_model": [[21, null]], "lmflow.models.vision2seq_model": [[22, null]], "lmflow.models.vision_encoder": [[24, null]], "lmflow.models.vision_encoder.clip_encoder": [[23, null]], "lmflow.optim": [[35, null]], "lmflow.optim.adabelief": [[25, null]], "lmflow.optim.adabound": [[26, null]], "lmflow.optim.adadelta": [[27, null]], "lmflow.optim.adagrad": [[28, null]], "lmflow.optim.adam": [[29, null]], "lmflow.optim.adamax": [[30, null]], "lmflow.optim.adamp": [[31, null]], "lmflow.optim.adamw_schedule_free": [[32, null]], "lmflow.optim.adan": [[33, null]], "lmflow.optim.dummy": [[34, null]], "lmflow.optim.lamb": [[36, null]], "lmflow.optim.lars": [[37, null]], "lmflow.optim.nadam": [[38, null]], "lmflow.optim.novograd": [[39, null]], "lmflow.optim.optimizers": [[40, null]], "lmflow.optim.radam": [[41, null]], "lmflow.optim.sgd_schedule_free": [[42, null]], "lmflow.optim.sgdp": [[43, null]], "lmflow.optim.sophia": [[44, null]], "lmflow.optim.yogi": [[45, null]], "lmflow.pipeline": [[54, null]], "lmflow.pipeline.auto_pipeline": [[46, null]], "lmflow.pipeline.base_aligner": [[47, null]], "lmflow.pipeline.base_pipeline": [[48, null]], "lmflow.pipeline.base_tuner": [[49, null]], "lmflow.pipeline.dpo_aligner": [[50, null]], "lmflow.pipeline.dpov2_aligner": [[51, null]], "lmflow.pipeline.evaluator": [[52, null]], "lmflow.pipeline.finetuner": [[53, null]], "lmflow.pipeline.inferencer": [[55, null]], "lmflow.pipeline.iterative_dpo_aligner": [[56, null]], "lmflow.pipeline.raft_aligner": [[57, null]], "lmflow.pipeline.rm_inferencer": [[58, null]], "lmflow.pipeline.rm_tuner": [[59, null]], "lmflow.pipeline.utils": [[62, null]], "lmflow.pipeline.utils.dpov2_dataprocessor": [[60, null]], "lmflow.pipeline.utils.dpov2_trainer": [[61, null]], "lmflow.pipeline.utils.memory_safe_dpov2_align": [[63, null]], "lmflow.pipeline.utils.memory_safe_vllm_inference": [[64, null]], "lmflow.pipeline.utils.peft_trainer": [[65, null]], "lmflow.pipeline.utils.raft_trainer": [[66, null]], "lmflow.pipeline.utils.rm_dataprocessor": [[67, null]], "lmflow.pipeline.utils.rm_trainer": [[68, null]], "lmflow.pipeline.vllm_inferencer": [[69, null]], "lmflow.tokenization": [[72, null]], "lmflow.tokenization.hf_decoder_model": [[70, null]], "lmflow.tokenization.hf_text_regression_model": [[71, null]], "lmflow.utils": [[95, null]], "lmflow.utils.common": [[73, null]], "lmflow.utils.constants": [[74, null]], "lmflow.utils.conversation_template": [[81, null]], "lmflow.utils.conversation_template.base": [[75, null]], "lmflow.utils.conversation_template.chatglm": [[76, null]], "lmflow.utils.conversation_template.chatml": [[77, null]], "lmflow.utils.conversation_template.deepseek": [[78, null]], "lmflow.utils.conversation_template.gemma": [[79, null]], "lmflow.utils.conversation_template.hymba": [[80, null]], "lmflow.utils.conversation_template.internlm": [[82, null]], "lmflow.utils.conversation_template.llama": [[83, null]], "lmflow.utils.conversation_template.phi": [[84, null]], "lmflow.utils.conversation_template.qwen": [[85, null]], "lmflow.utils.conversation_template.yi": [[86, null]], "lmflow.utils.conversation_template.zephyr": [[87, null]], "lmflow.utils.data_utils": [[88, null]], "lmflow.utils.flash_attention": [[92, null]], "lmflow.utils.flash_attention.bloom_flash_attention": [[89, null]], "lmflow.utils.flash_attention.gpt2_flash_attention": [[90, null]], "lmflow.utils.flash_attention.gpt_neo_flash_attention": [[91, null]], "lmflow.utils.flash_attention.llama_flash_attention": [[93, null]], "lmflow.utils.flash_attention.triton_flash_attention": [[94, null]], "lmflow.utils.llava_conversation_lib": [[96, null]], "lmflow.utils.model": [[97, null]], "lmflow.utils.multimodal": [[98, null]], "lmflow.utils.position_interpolation": [[99, null]], "lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch": [[100, null]], "lmflow.utils.versioning": [[101, null]], "lmflow.version": [[102, null]], "}": [[75, "id5"], [75, "id10"], [80, "id5"], [81, "id5"], [81, "id10"]]}, "docnames": ["about/authors", "about/changelog", "about/index", "autoapi/index", "autoapi/lmflow/args/index", "autoapi/lmflow/datasets/dataset/index", "autoapi/lmflow/datasets/index", "autoapi/lmflow/datasets/multi_modal_dataset/index", "autoapi/lmflow/index", "autoapi/lmflow/models/auto_model/index", "autoapi/lmflow/models/base_model/index", "autoapi/lmflow/models/decoder_model/index", "autoapi/lmflow/models/encoder_decoder_model/index", "autoapi/lmflow/models/hf_decoder_model/index", "autoapi/lmflow/models/hf_encoder_decoder_model/index", "autoapi/lmflow/models/hf_model_mixin/index", "autoapi/lmflow/models/hf_text_regression_model/index", "autoapi/lmflow/models/index", "autoapi/lmflow/models/interfaces/index", "autoapi/lmflow/models/interfaces/tunable/index", "autoapi/lmflow/models/regression_model/index", "autoapi/lmflow/models/text_regression_model/index", "autoapi/lmflow/models/vision2seq_model/index", "autoapi/lmflow/models/vision_encoder/clip_encoder/index", "autoapi/lmflow/models/vision_encoder/index", "autoapi/lmflow/optim/adabelief/index", "autoapi/lmflow/optim/adabound/index", "autoapi/lmflow/optim/adadelta/index", "autoapi/lmflow/optim/adagrad/index", "autoapi/lmflow/optim/adam/index", "autoapi/lmflow/optim/adamax/index", "autoapi/lmflow/optim/adamp/index", "autoapi/lmflow/optim/adamw_schedule_free/index", "autoapi/lmflow/optim/adan/index", "autoapi/lmflow/optim/dummy/index", "autoapi/lmflow/optim/index", "autoapi/lmflow/optim/lamb/index", "autoapi/lmflow/optim/lars/index", "autoapi/lmflow/optim/nadam/index", "autoapi/lmflow/optim/novograd/index", "autoapi/lmflow/optim/optimizers/index", "autoapi/lmflow/optim/radam/index", "autoapi/lmflow/optim/sgd_schedule_free/index", "autoapi/lmflow/optim/sgdp/index", "autoapi/lmflow/optim/sophia/index", "autoapi/lmflow/optim/yogi/index", "autoapi/lmflow/pipeline/auto_pipeline/index", "autoapi/lmflow/pipeline/base_aligner/index", "autoapi/lmflow/pipeline/base_pipeline/index", "autoapi/lmflow/pipeline/base_tuner/index", "autoapi/lmflow/pipeline/dpo_aligner/index", "autoapi/lmflow/pipeline/dpov2_aligner/index", "autoapi/lmflow/pipeline/evaluator/index", "autoapi/lmflow/pipeline/finetuner/index", "autoapi/lmflow/pipeline/index", "autoapi/lmflow/pipeline/inferencer/index", "autoapi/lmflow/pipeline/iterative_dpo_aligner/index", "autoapi/lmflow/pipeline/raft_aligner/index", "autoapi/lmflow/pipeline/rm_inferencer/index", "autoapi/lmflow/pipeline/rm_tuner/index", "autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index", "autoapi/lmflow/pipeline/utils/dpov2_trainer/index", "autoapi/lmflow/pipeline/utils/index", "autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index", "autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index", "autoapi/lmflow/pipeline/utils/peft_trainer/index", "autoapi/lmflow/pipeline/utils/raft_trainer/index", "autoapi/lmflow/pipeline/utils/rm_dataprocessor/index", "autoapi/lmflow/pipeline/utils/rm_trainer/index", "autoapi/lmflow/pipeline/vllm_inferencer/index", "autoapi/lmflow/tokenization/hf_decoder_model/index", "autoapi/lmflow/tokenization/hf_text_regression_model/index", "autoapi/lmflow/tokenization/index", "autoapi/lmflow/utils/common/index", "autoapi/lmflow/utils/constants/index", "autoapi/lmflow/utils/conversation_template/base/index", "autoapi/lmflow/utils/conversation_template/chatglm/index", "autoapi/lmflow/utils/conversation_template/chatml/index", "autoapi/lmflow/utils/conversation_template/deepseek/index", "autoapi/lmflow/utils/conversation_template/gemma/index", "autoapi/lmflow/utils/conversation_template/hymba/index", "autoapi/lmflow/utils/conversation_template/index", "autoapi/lmflow/utils/conversation_template/internlm/index", "autoapi/lmflow/utils/conversation_template/llama/index", "autoapi/lmflow/utils/conversation_template/phi/index", "autoapi/lmflow/utils/conversation_template/qwen/index", "autoapi/lmflow/utils/conversation_template/yi/index", "autoapi/lmflow/utils/conversation_template/zephyr/index", "autoapi/lmflow/utils/data_utils/index", "autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index", "autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index", "autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index", "autoapi/lmflow/utils/flash_attention/index", "autoapi/lmflow/utils/flash_attention/llama_flash_attention/index", "autoapi/lmflow/utils/flash_attention/triton_flash_attention/index", "autoapi/lmflow/utils/index", "autoapi/lmflow/utils/llava_conversation_lib/index", "autoapi/lmflow/utils/model/index", "autoapi/lmflow/utils/multimodal/index", "autoapi/lmflow/utils/position_interpolation/index", "autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index", "autoapi/lmflow/utils/versioning/index", "autoapi/lmflow/version/index", "blogs/benchmark", "blogs/index", "examples/DATASETS", "examples/TASK_GUIDE", "examples/checkpoints", "examples/customize_conversation_template", "examples/finetuning", "examples/index", "examples/medical_finetune", "examples/raft", "examples/reward_modeling", "examples/supported_conversation_template", "index"], "envversion": {"sphinx": 64, "sphinx.domains.c": 3, "sphinx.domains.changeset": 1, "sphinx.domains.citation": 1, "sphinx.domains.cpp": 9, "sphinx.domains.index": 1, "sphinx.domains.javascript": 3, "sphinx.domains.math": 2, "sphinx.domains.python": 4, "sphinx.domains.rst": 2, "sphinx.domains.std": 2, "sphinx.ext.todo": 2, "sphinx.ext.viewcode": 1}, "filenames": ["about/authors.md", "about/changelog.md", "about/index.md", "autoapi/index.rst", "autoapi/lmflow/args/index.rst", "autoapi/lmflow/datasets/dataset/index.rst", "autoapi/lmflow/datasets/index.rst", "autoapi/lmflow/datasets/multi_modal_dataset/index.rst", "autoapi/lmflow/index.rst", "autoapi/lmflow/models/auto_model/index.rst", "autoapi/lmflow/models/base_model/index.rst", "autoapi/lmflow/models/decoder_model/index.rst", "autoapi/lmflow/models/encoder_decoder_model/index.rst", "autoapi/lmflow/models/hf_decoder_model/index.rst", "autoapi/lmflow/models/hf_encoder_decoder_model/index.rst", "autoapi/lmflow/models/hf_model_mixin/index.rst", "autoapi/lmflow/models/hf_text_regression_model/index.rst", "autoapi/lmflow/models/index.rst", "autoapi/lmflow/models/interfaces/index.rst", "autoapi/lmflow/models/interfaces/tunable/index.rst", "autoapi/lmflow/models/regression_model/index.rst", "autoapi/lmflow/models/text_regression_model/index.rst", "autoapi/lmflow/models/vision2seq_model/index.rst", "autoapi/lmflow/models/vision_encoder/clip_encoder/index.rst", "autoapi/lmflow/models/vision_encoder/index.rst", "autoapi/lmflow/optim/adabelief/index.rst", "autoapi/lmflow/optim/adabound/index.rst", "autoapi/lmflow/optim/adadelta/index.rst", "autoapi/lmflow/optim/adagrad/index.rst", "autoapi/lmflow/optim/adam/index.rst", "autoapi/lmflow/optim/adamax/index.rst", "autoapi/lmflow/optim/adamp/index.rst", "autoapi/lmflow/optim/adamw_schedule_free/index.rst", "autoapi/lmflow/optim/adan/index.rst", "autoapi/lmflow/optim/dummy/index.rst", "autoapi/lmflow/optim/index.rst", "autoapi/lmflow/optim/lamb/index.rst", "autoapi/lmflow/optim/lars/index.rst", "autoapi/lmflow/optim/nadam/index.rst", "autoapi/lmflow/optim/novograd/index.rst", "autoapi/lmflow/optim/optimizers/index.rst", "autoapi/lmflow/optim/radam/index.rst", "autoapi/lmflow/optim/sgd_schedule_free/index.rst", "autoapi/lmflow/optim/sgdp/index.rst", "autoapi/lmflow/optim/sophia/index.rst", "autoapi/lmflow/optim/yogi/index.rst", "autoapi/lmflow/pipeline/auto_pipeline/index.rst", "autoapi/lmflow/pipeline/base_aligner/index.rst", "autoapi/lmflow/pipeline/base_pipeline/index.rst", "autoapi/lmflow/pipeline/base_tuner/index.rst", "autoapi/lmflow/pipeline/dpo_aligner/index.rst", "autoapi/lmflow/pipeline/dpov2_aligner/index.rst", "autoapi/lmflow/pipeline/evaluator/index.rst", "autoapi/lmflow/pipeline/finetuner/index.rst", "autoapi/lmflow/pipeline/index.rst", "autoapi/lmflow/pipeline/inferencer/index.rst", "autoapi/lmflow/pipeline/iterative_dpo_aligner/index.rst", "autoapi/lmflow/pipeline/raft_aligner/index.rst", "autoapi/lmflow/pipeline/rm_inferencer/index.rst", "autoapi/lmflow/pipeline/rm_tuner/index.rst", "autoapi/lmflow/pipeline/utils/dpov2_dataprocessor/index.rst", "autoapi/lmflow/pipeline/utils/dpov2_trainer/index.rst", "autoapi/lmflow/pipeline/utils/index.rst", "autoapi/lmflow/pipeline/utils/memory_safe_dpov2_align/index.rst", "autoapi/lmflow/pipeline/utils/memory_safe_vllm_inference/index.rst", "autoapi/lmflow/pipeline/utils/peft_trainer/index.rst", "autoapi/lmflow/pipeline/utils/raft_trainer/index.rst", "autoapi/lmflow/pipeline/utils/rm_dataprocessor/index.rst", "autoapi/lmflow/pipeline/utils/rm_trainer/index.rst", "autoapi/lmflow/pipeline/vllm_inferencer/index.rst", "autoapi/lmflow/tokenization/hf_decoder_model/index.rst", "autoapi/lmflow/tokenization/hf_text_regression_model/index.rst", "autoapi/lmflow/tokenization/index.rst", "autoapi/lmflow/utils/common/index.rst", "autoapi/lmflow/utils/constants/index.rst", "autoapi/lmflow/utils/conversation_template/base/index.rst", "autoapi/lmflow/utils/conversation_template/chatglm/index.rst", "autoapi/lmflow/utils/conversation_template/chatml/index.rst", "autoapi/lmflow/utils/conversation_template/deepseek/index.rst", "autoapi/lmflow/utils/conversation_template/gemma/index.rst", "autoapi/lmflow/utils/conversation_template/hymba/index.rst", "autoapi/lmflow/utils/conversation_template/index.rst", "autoapi/lmflow/utils/conversation_template/internlm/index.rst", "autoapi/lmflow/utils/conversation_template/llama/index.rst", "autoapi/lmflow/utils/conversation_template/phi/index.rst", "autoapi/lmflow/utils/conversation_template/qwen/index.rst", "autoapi/lmflow/utils/conversation_template/yi/index.rst", "autoapi/lmflow/utils/conversation_template/zephyr/index.rst", "autoapi/lmflow/utils/data_utils/index.rst", "autoapi/lmflow/utils/flash_attention/bloom_flash_attention/index.rst", "autoapi/lmflow/utils/flash_attention/gpt2_flash_attention/index.rst", "autoapi/lmflow/utils/flash_attention/gpt_neo_flash_attention/index.rst", "autoapi/lmflow/utils/flash_attention/index.rst", "autoapi/lmflow/utils/flash_attention/llama_flash_attention/index.rst", "autoapi/lmflow/utils/flash_attention/triton_flash_attention/index.rst", "autoapi/lmflow/utils/index.rst", "autoapi/lmflow/utils/llava_conversation_lib/index.rst", "autoapi/lmflow/utils/model/index.rst", "autoapi/lmflow/utils/multimodal/index.rst", "autoapi/lmflow/utils/position_interpolation/index.rst", "autoapi/lmflow/utils/position_interpolation/llama_rope_scaled_monkey_patch/index.rst", "autoapi/lmflow/utils/versioning/index.rst", "autoapi/lmflow/version/index.rst", "blogs/benchmark.md", "blogs/index.md", "examples/DATASETS.md", "examples/TASK_GUIDE.md", "examples/checkpoints.md", "examples/customize_conversation_template.md", "examples/finetuning.md", "examples/index.md", "examples/medical_finetune.md", "examples/raft.md", "examples/reward_modeling.md", "examples/supported_conversation_template.md", "index.md"], "indexentries": {"__call__() (lmflow.datasets.multi_modal_dataset.datacollatorforsuperviseddataset method)": [[7, "lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset.__call__", false]], "__call__() (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding method)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.__call__", false]], "__call__() (lmflow.pipeline.utils.rm_dataprocessor.rewarddatacollatorwithpadding method)": [[67, "lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.__call__", false]], "__distributed_inference() (lmflow.pipeline.rm_inferencer.rewardmodelinferencer method)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer.__distributed_inference", false]], "__filter_args() (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner method)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.__filter_args", false]], "__getitem__() (lmflow.datasets.multi_modal_dataset.custommultimodaldataset method)": [[7, "lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.__getitem__", false]], "__inference() (lmflow.models.hf_decoder_model.hfdecodermodel method)": [[13, "lmflow.models.hf_decoder_model.HFDecoderModel.__inference", false]], "__inference() (lmflow.models.hf_text_regression_model.hftextregressionmodel method)": [[16, "lmflow.models.hf_text_regression_model.HFTextRegressionModel.__inference", false]], "__inference() (lmflow.pipeline.rm_inferencer.rewardmodelinferencer method)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer.__inference", false]], "__len__() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.__len__", false]], "__len__() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.__len__", false]], "__len__() (lmflow.datasets.multi_modal_dataset.custommultimodaldataset method)": [[7, "lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.__len__", false]], "__model_module_inject() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.__model_module_inject", false]], "__post_init__() (lmflow.args.datasetarguments method)": [[4, "lmflow.args.DatasetArguments.__post_init__", false]], "__post_init__() (lmflow.args.inferencerarguments method)": [[4, "lmflow.args.InferencerArguments.__post_init__", false]], "__post_init__() (lmflow.args.modelarguments method)": [[4, "lmflow.args.ModelArguments.__post_init__", false]], "__post_init__() (lmflow.utils.conversation_template.base.conversationtemplate method)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.__post_init__", false]], "__post_init__() (lmflow.utils.conversation_template.base.emptyformatter method)": [[75, "lmflow.utils.conversation_template.base.EmptyFormatter.__post_init__", false]], "__post_init__() (lmflow.utils.conversation_template.base.stringformatter method)": [[75, "lmflow.utils.conversation_template.base.StringFormatter.__post_init__", false]], "__post_init__() (lmflow.utils.conversation_template.base.templatecomponent method)": [[75, "lmflow.utils.conversation_template.base.TemplateComponent.__post_init__", false]], "__post_init__() (lmflow.utils.conversation_template.conversationtemplate method)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.__post_init__", false]], "__post_process_model_output() (lmflow.pipeline.rm_inferencer.rewardmodelinferencer method)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer.__post_process_model_output", false]], "__prepare_dtype() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.__prepare_dtype", false]], "__prepare_inputs_for_inference() (lmflow.models.hf_decoder_model.hfdecodermodel method)": [[13, "lmflow.models.hf_decoder_model.HFDecoderModel.__prepare_inputs_for_inference", false]], "__prepare_inputs_for_vllm_inference() (lmflow.models.hf_decoder_model.hfdecodermodel method)": [[13, "lmflow.models.hf_decoder_model.HFDecoderModel.__prepare_inputs_for_vllm_inference", false]], "__prepare_model_config() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_config", false]], "__prepare_model_for_inference() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_inference", false]], "__prepare_model_for_training() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_training", false]], "__prepare_model_for_vllm_inference() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_for_vllm_inference", false]], "__prepare_model_post_process() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.__prepare_model_post_process", false]], "__prepare_peft_config() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.__prepare_peft_config", false]], "__prepare_quant_config() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.__prepare_quant_config", false]], "__prepare_tokenizer() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.__prepare_tokenizer", false]], "__prepare_training_args() (lmflow.pipeline.dpov2_aligner.dpov2aligner method)": [[51, "lmflow.pipeline.dpov2_aligner.DPOv2Aligner.__prepare_training_args", false]], "__repr__() (lmflow.utils.conversation_template.base.templatecomponent method)": [[75, "lmflow.utils.conversation_template.base.TemplateComponent.__repr__", false]], "__setstate__() (lmflow.optim.adabelief.adabelief method)": [[25, "lmflow.optim.adabelief.AdaBelief.__setstate__", false]], "__setstate__() (lmflow.optim.adabound.adabound method)": [[26, "lmflow.optim.adabound.AdaBound.__setstate__", false]], "__setstate__() (lmflow.optim.adamax.adamax method)": [[30, "lmflow.optim.adamax.Adamax.__setstate__", false]], "__setstate__() (lmflow.optim.adan.adan method)": [[33, "lmflow.optim.adan.Adan.__setstate__", false]], "__setstate__() (lmflow.optim.lars.lars method)": [[37, "lmflow.optim.lars.LARS.__setstate__", false]], "__setstate__() (lmflow.optim.nadam.nadam method)": [[38, "lmflow.optim.nadam.NAdam.__setstate__", false]], "__setstate__() (lmflow.optim.novograd.novograd method)": [[39, "lmflow.optim.novograd.NovoGrad.__setstate__", false]], "__setstate__() (lmflow.optim.radam.radam method)": [[41, "lmflow.optim.radam.RAdam.__setstate__", false]], "__setstate__() (lmflow.optim.sophia.sophiag method)": [[44, "lmflow.optim.sophia.SophiaG.__setstate__", false]], "__str__() (lmflow.utils.conversation_template.base.templatecomponent method)": [[75, "lmflow.utils.conversation_template.base.TemplateComponent.__str__", false]], "__version__ (in module lmflow)": [[8, "lmflow.__version__", false]], "__version__ (in module lmflow.version)": [[102, "lmflow.version.__version__", false]], "__vllm_inference() (lmflow.models.hf_decoder_model.hfdecodermodel method)": [[13, "lmflow.models.hf_decoder_model.HFDecoderModel.__vllm_inference", false]], "__vllm_inference() (lmflow.models.hf_text_regression_model.hftextregressionmodel method)": [[16, "lmflow.models.hf_text_regression_model.HFTextRegressionModel.__vllm_inference", false]], "__vllm_inference() (lmflow.pipeline.rm_inferencer.rewardmodelinferencer method)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer.__vllm_inference", false]], "_activated (lmflow.models.hf_model_mixin.hfmodelmixin attribute)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin._activated", false]], "_add_sm_patterns_to_gitignore() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._add_sm_patterns_to_gitignore", false]], "_align_single_iteration() (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner method)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._align_single_iteration", false]], "_attn() (in module lmflow.utils.flash_attention.gpt_neo_flash_attention)": [[91, "lmflow.utils.flash_attention.gpt_neo_flash_attention._attn", false]], "_bwd_kernel() (in module lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel", false]], "_bwd_kernel_one_col_block() (in module lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention._bwd_kernel_one_col_block", false]], "_bwd_preprocess_do_o_dot() (in module lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention._bwd_preprocess_do_o_dot", false]], "_bwd_store_dk_dv() (in module lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention._bwd_store_dk_dv", false]], "_calc_response_lengths() (lmflow.pipeline.dpov2_aligner.dpov2aligner method)": [[51, "lmflow.pipeline.dpov2_aligner.DPOv2Aligner._calc_response_lengths", false]], "_calc_reward_with_length_penalty() (lmflow.pipeline.dpov2_aligner.dpov2aligner method)": [[51, "lmflow.pipeline.dpov2_aligner.DPOv2Aligner._calc_reward_with_length_penalty", false]], "_channel_view() (lmflow.optim.adamp.adamp static method)": [[31, "lmflow.optim.adamp.AdamP._channel_view", false]], "_channel_view() (lmflow.optim.sgdp.sgdp static method)": [[43, "lmflow.optim.sgdp.SGDP._channel_view", false]], "_check_data_format() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset._check_data_format", false]], "_check_data_format() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset._check_data_format", false]], "_check_if_alignable() (lmflow.pipeline.base_aligner.basealigner method)": [[47, "lmflow.pipeline.base_aligner.BaseAligner._check_if_alignable", false]], "_check_if_tunable() (lmflow.pipeline.base_tuner.basetuner method)": [[49, "lmflow.pipeline.base_tuner.BaseTuner._check_if_tunable", false]], "_clean_text() (lmflow.pipeline.raft_aligner.raftaligner method)": [[57, "lmflow.pipeline.raft_aligner.RaftAligner._clean_text", false]], "_cosine_similarity() (lmflow.optim.adamp.adamp static method)": [[31, "lmflow.optim.adamp.AdamP._cosine_similarity", false]], "_cosine_similarity() (lmflow.optim.sgdp.sgdp static method)": [[43, "lmflow.optim.sgdp.SGDP._cosine_similarity", false]], "_discard_sample() (lmflow.pipeline.raft_aligner.raftaligner method)": [[57, "lmflow.pipeline.raft_aligner.RaftAligner._discard_sample", false]], "_distributed_inference() (lmflow.pipeline.vllm_inferencer.vllminferencer method)": [[69, "lmflow.pipeline.vllm_inferencer.VLLMInferencer._distributed_inference", false]], "_do_reward_model_inference() (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner method)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_reward_model_inference", false]], "_do_single_dpo_align() (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner method)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_single_dpo_align", false]], "_do_target_model_inference() (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner method)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._do_target_model_inference", false]], "_encode() (lmflow.utils.conversation_template.base.conversationtemplate method)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate._encode", false]], "_encode() (lmflow.utils.conversation_template.base.conversationtemplatefortool method)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplateForTool._encode", false]], "_encode() (lmflow.utils.conversation_template.conversationtemplate method)": [[81, "lmflow.utils.conversation_template.ConversationTemplate._encode", false]], "_encode() (lmflow.utils.conversation_template.conversationtemplatefortool method)": [[81, "lmflow.utils.conversation_template.ConversationTemplateForTool._encode", false]], "_encode() (lmflow.utils.conversation_template.llama.llama2conversationtemplate method)": [[83, "lmflow.utils.conversation_template.llama.Llama2ConversationTemplate._encode", false]], "_encode() (lmflow.utils.conversation_template.llama.llama2conversationtemplatefortool method)": [[83, "lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool._encode", false]], "_encode() (lmflow.utils.conversation_template.zephyr.zephyrconversationtemplate method)": [[87, "lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate._encode", false]], "_encode_template() (lmflow.utils.conversation_template.base.conversationtemplate method)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate._encode_template", false]], "_encode_template() (lmflow.utils.conversation_template.base.conversationtemplatefortool method)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplateForTool._encode_template", false]], "_encode_template() (lmflow.utils.conversation_template.conversationtemplate method)": [[81, "lmflow.utils.conversation_template.ConversationTemplate._encode_template", false]], "_encode_template() (lmflow.utils.conversation_template.conversationtemplatefortool method)": [[81, "lmflow.utils.conversation_template.ConversationTemplateForTool._encode_template", false]], "_ensure_id_list() (lmflow.utils.conversation_template.base.conversationtemplate method)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate._ensure_id_list", false]], "_ensure_id_list() (lmflow.utils.conversation_template.conversationtemplate method)": [[81, "lmflow.utils.conversation_template.ConversationTemplate._ensure_id_list", false]], "_evaluate_acc_with_accelerator() (lmflow.pipeline.evaluator.evaluator method)": [[52, "lmflow.pipeline.evaluator.Evaluator._evaluate_acc_with_accelerator", false]], "_evaluate_acc_with_deepspeed() (lmflow.pipeline.evaluator.evaluator method)": [[52, "lmflow.pipeline.evaluator.Evaluator._evaluate_acc_with_deepspeed", false]], "_evaluate_nll() (lmflow.pipeline.evaluator.evaluator method)": [[52, "lmflow.pipeline.evaluator.Evaluator._evaluate_nll", false]], "_evaluate_ppl() (lmflow.pipeline.evaluator.evaluator method)": [[52, "lmflow.pipeline.evaluator.Evaluator._evaluate_ppl", false]], "_flash_attn_backward() (in module lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention._flash_attn_backward", false]], "_flash_attn_forward() (in module lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention._flash_attn_forward", false]], "_fwd_kernel() (in module lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention._fwd_kernel", false]], "_gather_and_numpify() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._gather_and_numpify", false]], "_get_batch_dataset_local() (lmflow.pipeline.raft_aligner.raftaligner method)": [[57, "lmflow.pipeline.raft_aligner.RaftAligner._get_batch_dataset_local", false]], "_get_batch_dataset_top() (lmflow.pipeline.raft_aligner.raftaligner method)": [[57, "lmflow.pipeline.raft_aligner.RaftAligner._get_batch_dataset_top", false]], "_get_collator_with_removed_columns() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_collator_with_removed_columns", false]], "_get_eval_sampler() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_eval_sampler", false]], "_get_output_dir() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_output_dir", false]], "_get_train_sampler() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._get_train_sampler", false]], "_hp_search_setup() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._hp_search_setup", false]], "_inference() (lmflow.pipeline.rm_inferencer.rewardmodelinferencer method)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer._inference", false]], "_inference() (lmflow.pipeline.vllm_inferencer.vllminferencer method)": [[69, "lmflow.pipeline.vllm_inferencer.VLLMInferencer._inference", false]], "_initialize_trainer() (lmflow.pipeline.dpo_aligner.dpoaligner method)": [[50, "lmflow.pipeline.dpo_aligner.DPOAligner._initialize_trainer", false]], "_initialize_trainer() (lmflow.pipeline.raft_aligner.raftaligner method)": [[57, "lmflow.pipeline.raft_aligner.RaftAligner._initialize_trainer", false]], "_inner_training_loop() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._inner_training_loop", false]], "_is_native_cpu_amp_available (in module lmflow.pipeline.utils.raft_trainer)": [[66, "lmflow.pipeline.utils.raft_trainer._is_native_cpu_amp_available", false]], "_is_package_available() (in module lmflow.utils.versioning)": [[101, "lmflow.utils.versioning._is_package_available", false]], "_is_packages_available() (in module lmflow.utils.versioning)": [[101, "lmflow.utils.versioning._is_packages_available", false]], "_issue_warnings_after_load() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._issue_warnings_after_load", false]], "_layer_view() (lmflow.optim.adamp.adamp static method)": [[31, "lmflow.optim.adamp.AdamP._layer_view", false]], "_layer_view() (lmflow.optim.sgdp.sgdp static method)": [[43, "lmflow.optim.sgdp.SGDP._layer_view", false]], "_load_best_model() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_best_model", false]], "_load_dataset() (lmflow.pipeline.dpo_aligner.dpoaligner method)": [[50, "lmflow.pipeline.dpo_aligner.DPOAligner._load_dataset", false]], "_load_dataset() (lmflow.pipeline.raft_aligner.raftaligner method)": [[57, "lmflow.pipeline.raft_aligner.RaftAligner._load_dataset", false]], "_load_from_checkpoint() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_from_checkpoint", false]], "_load_input_dataset() (lmflow.pipeline.raft_aligner.raftaligner method)": [[57, "lmflow.pipeline.raft_aligner.RaftAligner._load_input_dataset", false]], "_load_optimizer_and_scheduler() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_optimizer_and_scheduler", false]], "_load_rng_state() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._load_rng_state", false]], "_loggers_initialized (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._loggers_initialized", false]], "_match() (lmflow.pipeline.evaluator.evaluator method)": [[52, "lmflow.pipeline.evaluator.Evaluator._match", false]], "_maybe_log_save_evaluate() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._maybe_log_save_evaluate", false]], "_memory_tracker (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._memory_tracker", false]], "_move_model_to_device() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._move_model_to_device", false]], "_multi_tensor_adan() (in module lmflow.optim.adan)": [[33, "lmflow.optim.adan._multi_tensor_adan", false]], "_nested_gather() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._nested_gather", false]], "_one_train() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._one_train", false]], "_pad_across_processes() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._pad_across_processes", false]], "_parse_dpo_aligner_args() (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner method)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_dpo_aligner_args", false]], "_parse_reward_model_inference_args() (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner method)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_reward_model_inference_args", false]], "_parse_target_model_inference_args() (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner method)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner._parse_target_model_inference_args", false]], "_prepare_attn_mask() (in module lmflow.utils.flash_attention.bloom_flash_attention)": [[89, "lmflow.utils.flash_attention.bloom_flash_attention._prepare_attn_mask", false]], "_prepare_decoder_attention_mask() (in module lmflow.utils.flash_attention.gpt2_flash_attention)": [[90, "lmflow.utils.flash_attention.gpt2_flash_attention._prepare_decoder_attention_mask", false]], "_prepare_decoder_attention_mask() (in module lmflow.utils.flash_attention.llama_flash_attention)": [[93, "lmflow.utils.flash_attention.llama_flash_attention._prepare_decoder_attention_mask", false]], "_prepare_input() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._prepare_input", false]], "_prepare_inputs() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._prepare_inputs", false]], "_projection() (lmflow.optim.adamp.adamp method)": [[31, "lmflow.optim.adamp.AdamP._projection", false]], "_projection() (lmflow.optim.sgdp.sgdp method)": [[43, "lmflow.optim.sgdp.SGDP._projection", false]], "_push_from_checkpoint() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._push_from_checkpoint", false]], "_remove_unused_columns() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._remove_unused_columns", false]], "_report_to_hp_search() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._report_to_hp_search", false]], "_rotate_checkpoints() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._rotate_checkpoints", false]], "_sampling_paired_idx_from_rewards() (lmflow.pipeline.dpov2_aligner.dpov2aligner method)": [[51, "lmflow.pipeline.dpov2_aligner.DPOv2Aligner._sampling_paired_idx_from_rewards", false]], "_sampling_paired_idx_from_rewards_fast() (lmflow.pipeline.dpov2_aligner.dpov2aligner method)": [[51, "lmflow.pipeline.dpov2_aligner.DPOv2Aligner._sampling_paired_idx_from_rewards_fast", false]], "_save() (lmflow.pipeline.utils.peft_trainer.peftsavingcallback method)": [[65, "lmflow.pipeline.utils.peft_trainer.PeftSavingCallback._save", false]], "_save() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._save", false]], "_save_checkpoint() (lmflow.pipeline.utils.peft_trainer.pefttrainer method)": [[65, "lmflow.pipeline.utils.peft_trainer.PeftTrainer._save_checkpoint", false]], "_save_checkpoint() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._save_checkpoint", false]], "_save_tpu() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._save_tpu", false]], "_set_signature_columns_if_needed() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._set_signature_columns_if_needed", false]], "_signature_columns (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._signature_columns", false]], "_single_tensor_adan() (in module lmflow.optim.adan)": [[33, "lmflow.optim.adan._single_tensor_adan", false]], "_sorted_checkpoints() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._sorted_checkpoints", false]], "_train_batch_size (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._train_batch_size", false]], "_tune_save_checkpoint() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._tune_save_checkpoint", false]], "_wrap_model() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer._wrap_model", false]], "accelerate_config_file (lmflow.args.dpov2alignerarguments attribute)": [[4, "lmflow.args.DPOv2AlignerArguments.accelerate_config_file", false]], "activate_model_for_inference() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.activate_model_for_inference", false]], "adabelief (class in lmflow.optim.adabelief)": [[25, "lmflow.optim.adabelief.AdaBelief", false]], "adabelief (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.ADABELIEF", false]], "adabound (class in lmflow.optim.adabound)": [[26, "lmflow.optim.adabound.AdaBound", false]], "adabound (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.ADABOUND", false]], "adadelta (class in lmflow.optim.adadelta)": [[27, "lmflow.optim.adadelta.Adadelta", false]], "adadelta (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.ADADELTA", false]], "adagrad (class in lmflow.optim.adagrad)": [[28, "lmflow.optim.adagrad.AdaGrad", false]], "adagrad (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.ADAGRAD", false]], "adam (class in lmflow.optim.adam)": [[29, "lmflow.optim.adam.Adam", false]], "adam (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.ADAM", false]], "adam (lmflow.optim.lamb.lamb attribute)": [[36, "lmflow.optim.lamb.Lamb.adam", false]], "adamax (class in lmflow.optim.adamax)": [[30, "lmflow.optim.adamax.Adamax", false]], "adamax (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.ADAMAX", false]], "adamp (class in lmflow.optim.adamp)": [[31, "lmflow.optim.adamp.AdamP", false]], "adamp (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.ADAMP", false]], "adamw_schedule_free (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.ADAMW_SCHEDULE_FREE", false]], "adamwschedulefree (class in lmflow.optim.adamw_schedule_free)": [[32, "lmflow.optim.adamw_schedule_free.AdamWScheduleFree", false]], "adan (class in lmflow.optim.adan)": [[33, "lmflow.optim.adan.Adan", false]], "adan (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.ADAN", false]], "adapt_llava_model_to_lmflow_type() (in module lmflow.utils.multimodal)": [[98, "lmflow.utils.multimodal.adapt_llava_model_to_lmflow_type", false]], "add_callback() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.add_callback", false]], "add_dataclass_attr_prefix() (in module lmflow.utils.common)": [[73, "lmflow.utils.common.add_dataclass_attr_prefix", false]], "add_special_starter() (lmflow.utils.conversation_template.base.conversationtemplate method)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.add_special_starter", false]], "add_special_starter() (lmflow.utils.conversation_template.conversationtemplate method)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.add_special_starter", false]], "add_special_stopper() (lmflow.utils.conversation_template.base.conversationtemplate method)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.add_special_stopper", false]], "add_special_stopper() (lmflow.utils.conversation_template.conversationtemplate method)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.add_special_stopper", false]], "additional_stop_token_ids (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.additional_stop_token_ids", false]], "align() (lmflow.pipeline.base_aligner.basealigner method)": [[47, "lmflow.pipeline.base_aligner.BaseAligner.align", false]], "align() (lmflow.pipeline.dpo_aligner.dpoaligner method)": [[50, "lmflow.pipeline.dpo_aligner.DPOAligner.align", false]], "align() (lmflow.pipeline.dpov2_aligner.dpov2aligner method)": [[51, "lmflow.pipeline.dpov2_aligner.DPOv2Aligner.align", false]], "align() (lmflow.pipeline.dpov2_aligner.memorysafedpov2aligner method)": [[51, "lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.align", false]], "align() (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner method)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.align", false]], "align() (lmflow.pipeline.raft_aligner.raftaligner method)": [[57, "lmflow.pipeline.raft_aligner.RaftAligner.align", false]], "aligner_args (lmflow.pipeline.dpo_aligner.dpoaligner attribute)": [[50, "lmflow.pipeline.dpo_aligner.DPOAligner.aligner_args", false]], "aligner_args (lmflow.pipeline.dpov2_aligner.dpov2aligner attribute)": [[51, "lmflow.pipeline.dpov2_aligner.DPOv2Aligner.aligner_args", false]], "aligner_args (lmflow.pipeline.dpov2_aligner.memorysafedpov2aligner attribute)": [[51, "lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.aligner_args", false]], "aligner_args (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner attribute)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.aligner_args", false]], "aligner_args (lmflow.pipeline.raft_aligner.raftaligner attribute)": [[57, "lmflow.pipeline.raft_aligner.RaftAligner.aligner_args", false]], "aligner_file_path (lmflow.pipeline.dpov2_aligner.memorysafedpov2aligner attribute)": [[51, "lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.aligner_file_path", false]], "answer_extraction() (in module lmflow.utils.data_utils)": [[88, "lmflow.utils.data_utils.answer_extraction", false]], "answer_type (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.answer_type", false]], "append_message() (lmflow.utils.llava_conversation_lib.conversation method)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.append_message", false]], "apply_chat_template (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.apply_chat_template", false]], "arch_type (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.arch_type", false]], "args (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.args", false]], "assistant_formatter (lmflow.utils.conversation_template.base.conversationtemplate attribute)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.assistant_formatter", false]], "assistant_formatter (lmflow.utils.conversation_template.conversationtemplate attribute)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.assistant_formatter", false]], "autoarguments (class in lmflow.args)": [[4, "lmflow.args.AutoArguments", false]], "autocast_smart_context_manager() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.autocast_smart_context_manager", false]], "automodel (class in lmflow.models.auto_model)": [[9, "lmflow.models.auto_model.AutoModel", false]], "autopipeline (class in lmflow.pipeline.auto_pipeline)": [[46, "lmflow.pipeline.auto_pipeline.AutoPipeline", false]], "autoregressive_sampling() (lmflow.pipeline.inferencer.speculativeinferencer method)": [[55, "lmflow.pipeline.inferencer.SpeculativeInferencer.autoregressive_sampling", false]], "backend (lmflow.datasets.dataset attribute)": [[6, "lmflow.datasets.Dataset.backend", false]], "backend (lmflow.datasets.dataset.dataset attribute)": [[5, "lmflow.datasets.dataset.Dataset.backend", false]], "backend_dataset (lmflow.datasets.dataset attribute)": [[6, "lmflow.datasets.Dataset.backend_dataset", false]], "backend_dataset (lmflow.datasets.dataset.dataset attribute)": [[5, "lmflow.datasets.dataset.Dataset.backend_dataset", false]], "backward() (lmflow.utils.flash_attention.triton_flash_attention.flashattnfunc static method)": [[94, "lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc.backward", false]], "backward() (lmflow.utils.flash_attention.triton_flash_attention.flashattnkvpackedfunc static method)": [[94, "lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc.backward", false]], "backward() (lmflow.utils.flash_attention.triton_flash_attention.flashattnqkvpackedfunc static method)": [[94, "lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc.backward", false]], "base_lrs (lmflow.optim.adabound.adabound attribute)": [[26, "lmflow.optim.adabound.AdaBound.base_lrs", false]], "basealigner (class in lmflow.pipeline.base_aligner)": [[47, "lmflow.pipeline.base_aligner.BaseAligner", false]], "basemodel (class in lmflow.models.base_model)": [[10, "lmflow.models.base_model.BaseModel", false]], "basepipeline (class in lmflow.pipeline.base_pipeline)": [[48, "lmflow.pipeline.base_pipeline.BasePipeline", false]], "basetuner (class in lmflow.pipeline.base_tuner)": [[49, "lmflow.pipeline.base_tuner.BaseTuner", false]], "batchlize() (in module lmflow.utils.data_utils)": [[88, "lmflow.utils.data_utils.batchlize", false]], "benchmarkingarguments (class in lmflow.args)": [[4, "lmflow.args.BenchmarkingArguments", false]], "beta (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.beta", false]], "beta (lmflow.args.dpov2alignerarguments attribute)": [[4, "lmflow.args.DPOv2AlignerArguments.beta", false]], "bits (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.bits", false]], "block_size (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.block_size", false]], "block_size (lmflow.pipeline.evaluator.evaluator attribute)": [[52, "lmflow.pipeline.evaluator.Evaluator.block_size", false]], "blocking() (in module lmflow.tokenization.hf_decoder_model)": [[70, "lmflow.tokenization.hf_decoder_model.blocking", false]], "blocking() (in module lmflow.tokenization.hf_text_regression_model)": [[71, "lmflow.tokenization.hf_text_regression_model.blocking", false]], "blocking_paired() (in module lmflow.tokenization.hf_text_regression_model)": [[71, "lmflow.tokenization.hf_text_regression_model.blocking_paired", false]], "blocking_text_to_textlist() (in module lmflow.tokenization.hf_text_regression_model)": [[71, "lmflow.tokenization.hf_text_regression_model.blocking_text_to_textlist", false]], "build_vision_tower() (in module lmflow.models.vision_encoder)": [[24, "lmflow.models.vision_encoder.build_vision_tower", false]], "build_vision_tower() (in module lmflow.models.vision_encoder.clip_encoder)": [[23, "lmflow.models.vision_encoder.clip_encoder.build_vision_tower", false]], "cache_dir (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.cache_dir", false]], "call_model_init() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.call_model_init", false]], "callback_handler (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.callback_handler", false]], "can_return_loss (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.can_return_loss", false]], "chatglm3_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.CHATGLM3_TEMPLATE", false]], "chatglm3_template (in module lmflow.utils.conversation_template.chatglm)": [[76, "lmflow.utils.conversation_template.chatglm.CHATGLM3_TEMPLATE", false]], "chatml_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.CHATML_TEMPLATE", false]], "chatml_template (in module lmflow.utils.conversation_template.chatml)": [[77, "lmflow.utils.conversation_template.chatml.CHATML_TEMPLATE", false]], "check_homogeneity() (in module lmflow.utils.model)": [[97, "lmflow.utils.model.check_homogeneity", false]], "clamp_value (lmflow.optim.lamb.lamb attribute)": [[36, "lmflow.optim.lamb.Lamb.clamp_value", false]], "clipvisiontower (class in lmflow.models.vision_encoder.clip_encoder)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower", false]], "code_exec() (lmflow.pipeline.inferencer.toolinferencer method)": [[55, "lmflow.pipeline.inferencer.ToolInferencer.code_exec", false]], "collate() (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding method)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.collate", false]], "collection_strategy (lmflow.args.raftalignerarguments attribute)": [[4, "lmflow.args.RaftAlignerArguments.collection_strategy", false]], "compress_list() (lmflow.pipeline.rm_inferencer.rewardmodelinferencer method)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer.compress_list", false]], "compute_loss() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_loss", false]], "compute_loss() (lmflow.pipeline.utils.rm_trainer.peftrewardtrainer method)": [[68, "lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer.compute_loss", false]], "compute_loss() (lmflow.pipeline.utils.rm_trainer.rewardtrainer method)": [[68, "lmflow.pipeline.utils.rm_trainer.RewardTrainer.compute_loss", false]], "compute_loss_context_manager() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_loss_context_manager", false]], "compute_metrics (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.compute_metrics", false]], "compute_metrics() (in module lmflow.pipeline.utils.rm_trainer)": [[68, "lmflow.pipeline.utils.rm_trainer.compute_metrics", false]], "condenserotaryembedding (class in lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch)": [[100, "lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding", false]], "config (lmflow.models.vision_encoder.clip_encoder.clipvisiontower property)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.config", false]], "config (lmflow.pipeline.evaluator.evaluator attribute)": [[52, "lmflow.pipeline.evaluator.Evaluator.config", false]], "config (lmflow.pipeline.inferencer.inferencer attribute)": [[55, "lmflow.pipeline.inferencer.Inferencer.config", false]], "config_name (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.config_name", false]], "config_overrides (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.config_overrides", false]], "content (lmflow.utils.conversation_template.base.templatecomponent attribute)": [[75, "lmflow.utils.conversation_template.base.TemplateComponent.content", false]], "control (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.control", false]], "controller_heart_beat_expiration (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.CONTROLLER_HEART_BEAT_EXPIRATION", false]], "conv_llama_2 (in module lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.conv_llama_2", false]], "conv_llava_llama_2 (in module lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.conv_llava_llama_2", false]], "conv_llava_plain (in module lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.conv_llava_plain", false]], "conv_llava_v0 (in module lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.conv_llava_v0", false]], "conv_llava_v0_mmtag (in module lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.conv_llava_v0_mmtag", false]], "conv_llava_v1 (in module lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.conv_llava_v1", false]], "conv_llava_v1_mmtag (in module lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.conv_llava_v1_mmtag", false]], "conv_mpt (in module lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.conv_mpt", false]], "conv_templates (in module lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.conv_templates", false]], "conv_vicuna_v0 (in module lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.conv_vicuna_v0", false]], "conv_vicuna_v1 (in module lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.conv_vicuna_v1", false]], "conversation (class in lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.Conversation", false]], "conversation_dataset_description (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.CONVERSATION_DATASET_DESCRIPTION", false]], "conversation_role_names (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.CONVERSATION_ROLE_NAMES", false]], "conversation_template (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.conversation_template", false]], "conversation_tokenize_function() (in module lmflow.tokenization.hf_decoder_model)": [[70, "lmflow.tokenization.hf_decoder_model.conversation_tokenize_function", false]], "conversation_tokenize_function() (in module lmflow.tokenization.hf_text_regression_model)": [[71, "lmflow.tokenization.hf_text_regression_model.conversation_tokenize_function", false]], "conversationtemplate (class in lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.ConversationTemplate", false]], "conversationtemplate (class in lmflow.utils.conversation_template.base)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate", false]], "conversationtemplatefortool (class in lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.ConversationTemplateForTool", false]], "conversationtemplatefortool (class in lmflow.utils.conversation_template.base)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplateForTool", false]], "convert_to_paired_dataset() (lmflow.pipeline.dpov2_aligner.dpov2aligner method)": [[51, "lmflow.pipeline.dpov2_aligner.DPOv2Aligner.convert_to_paired_dataset", false]], "copy() (lmflow.utils.llava_conversation_lib.conversation method)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.copy", false]], "create_copied_dataclass() (in module lmflow.utils.common)": [[73, "lmflow.utils.common.create_copied_dataclass", false]], "create_customized_optimizer() (lmflow.pipeline.finetuner.finetuner method)": [[53, "lmflow.pipeline.finetuner.Finetuner.create_customized_optimizer", false]], "create_dataloader() (lmflow.pipeline.evaluator.evaluator method)": [[52, "lmflow.pipeline.evaluator.Evaluator.create_dataloader", false]], "create_dataloader() (lmflow.pipeline.inferencer.inferencer method)": [[55, "lmflow.pipeline.inferencer.Inferencer.create_dataloader", false]], "create_from_dict() (lmflow.datasets.dataset class method)": [[6, "lmflow.datasets.Dataset.create_from_dict", false]], "create_from_dict() (lmflow.datasets.dataset.dataset class method)": [[5, "lmflow.datasets.dataset.Dataset.create_from_dict", false]], "create_model_card() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_model_card", false]], "create_optimizer() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_optimizer", false]], "create_optimizer_and_scheduler() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_optimizer_and_scheduler", false]], "create_scheduler() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.create_scheduler", false]], "current_flos (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.current_flos", false]], "custom_model (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.custom_model", false]], "custom_vision_model (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.custom_vision_model", false]], "custom_vision_model (lmflow.models.vision2seq_model.customautovision2seqmodel attribute)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.custom_vision_model", false]], "customautovision2seqmodel (class in lmflow.models.vision2seq_model)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel", false]], "customized_cache_dir (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.customized_cache_dir", false]], "customized_optim (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.customized_optim", false]], "customized_optim_args (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.customized_optim_args", false]], "custommultimodaldataset (class in lmflow.datasets.multi_modal_dataset)": [[7, "lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset", false]], "data_args (lmflow.datasets.dataset attribute)": [[6, "lmflow.datasets.Dataset.data_args", false]], "data_args (lmflow.datasets.dataset.dataset attribute)": [[5, "lmflow.datasets.dataset.Dataset.data_args", false]], "data_args (lmflow.datasets.multi_modal_dataset.custommultimodaldataset attribute)": [[7, "lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.data_args", false]], "data_args (lmflow.pipeline.dpo_aligner.dpoaligner attribute)": [[50, "lmflow.pipeline.dpo_aligner.DPOAligner.data_args", false]], "data_args (lmflow.pipeline.dpov2_aligner.dpov2aligner attribute)": [[51, "lmflow.pipeline.dpov2_aligner.DPOv2Aligner.data_args", false]], "data_args (lmflow.pipeline.dpov2_aligner.memorysafedpov2aligner attribute)": [[51, "lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.data_args", false]], "data_args (lmflow.pipeline.evaluator.evaluator attribute)": [[52, "lmflow.pipeline.evaluator.Evaluator.data_args", false]], "data_args (lmflow.pipeline.finetuner.finetuner attribute)": [[53, "lmflow.pipeline.finetuner.Finetuner.data_args", false]], "data_args (lmflow.pipeline.inferencer.inferencer attribute)": [[55, "lmflow.pipeline.inferencer.Inferencer.data_args", false]], "data_args (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner attribute)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.data_args", false]], "data_args (lmflow.pipeline.raft_aligner.raftaligner attribute)": [[57, "lmflow.pipeline.raft_aligner.RaftAligner.data_args", false]], "data_args (lmflow.pipeline.rm_inferencer.rewardmodelinferencer attribute)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer.data_args", false]], "data_args (lmflow.pipeline.vllm_inferencer.inferencerwithoffloading attribute)": [[69, "lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.data_args", false]], "data_collator (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.data_collator", false]], "data_dict (lmflow.datasets.multi_modal_dataset.custommultimodaldataset attribute)": [[7, "lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.data_dict", false]], "datacollatorforsuperviseddataset (class in lmflow.datasets.multi_modal_dataset)": [[7, "lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset", false]], "dataset (class in lmflow.datasets)": [[6, "lmflow.datasets.Dataset", false]], "dataset (class in lmflow.datasets.dataset)": [[5, "lmflow.datasets.dataset.Dataset", false]], "dataset_config_name (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.dataset_config_name", false]], "dataset_description_map (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.DATASET_DESCRIPTION_MAP", false]], "dataset_name (lmflow.args.benchmarkingarguments attribute)": [[4, "lmflow.args.BenchmarkingArguments.dataset_name", false]], "dataset_name (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.dataset_name", false]], "dataset_path (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.dataset_path", false]], "dataset_path (lmflow.datasets.dataset attribute)": [[6, "lmflow.datasets.Dataset.dataset_path", false]], "dataset_path (lmflow.datasets.dataset.dataset attribute)": [[5, "lmflow.datasets.dataset.Dataset.dataset_path", false]], "dataset_path_list (lmflow.args.iterativealignerarguments attribute)": [[4, "lmflow.args.IterativeAlignerArguments.dataset_path_list", false]], "dataset_types (in module lmflow.datasets.dataset)": [[5, "lmflow.datasets.dataset.DATASET_TYPES", false]], "datasetarguments (class in lmflow.args)": [[4, "lmflow.args.DatasetArguments", false]], "deactivate_model_for_inference() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.deactivate_model_for_inference", false]], "debias (lmflow.optim.lamb.lamb attribute)": [[36, "lmflow.optim.lamb.Lamb.debias", false]], "decode() (lmflow.models.hf_decoder_model.hfdecodermodel method)": [[13, "lmflow.models.hf_decoder_model.HFDecoderModel.decode", false]], "decode() (lmflow.models.hf_encoder_decoder_model.hfencoderdecodermodel method)": [[14, "lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.decode", false]], "decodermodel (class in lmflow.models.decoder_model)": [[11, "lmflow.models.decoder_model.DecoderModel", false]], "deepseek_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.DEEPSEEK_TEMPLATE", false]], "deepseek_template (in module lmflow.utils.conversation_template.deepseek)": [[78, "lmflow.utils.conversation_template.deepseek.DEEPSEEK_TEMPLATE", false]], "deepspeed (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.deepspeed", false]], "deepspeed (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.deepspeed", false]], "deepspeed (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.deepspeed", false]], "default_callbacks (in module lmflow.pipeline.utils.raft_trainer)": [[66, "lmflow.pipeline.utils.raft_trainer.DEFAULT_CALLBACKS", false]], "default_conversation (in module lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.default_conversation", false]], "default_im_end_token (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.DEFAULT_IM_END_TOKEN", false]], "default_im_start_token (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.DEFAULT_IM_START_TOKEN", false]], "default_image_patch_token (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.DEFAULT_IMAGE_PATCH_TOKEN", false]], "default_image_token (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.DEFAULT_IMAGE_TOKEN", false]], "default_progress_callback (in module lmflow.pipeline.utils.raft_trainer)": [[66, "id0", false], [66, "lmflow.pipeline.utils.raft_trainer.DEFAULT_PROGRESS_CALLBACK", false]], "degenerated_to_sgd (lmflow.optim.adabelief.adabelief attribute)": [[25, "lmflow.optim.adabelief.AdaBelief.degenerated_to_sgd", false]], "device (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.device", false]], "device (lmflow.models.hf_encoder_decoder_model.hfencoderdecodermodel attribute)": [[14, "lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.device", false]], "device (lmflow.models.hf_model_mixin.hfmodelmixin attribute)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.device", false]], "device (lmflow.models.vision_encoder.clip_encoder.clipvisiontower property)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.device", false]], "dict() (lmflow.utils.llava_conversation_lib.conversation method)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.dict", false]], "disable_group_texts (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.disable_group_texts", false]], "distributed_inference_num_instances (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.distributed_inference_num_instances", false]], "do_dpo_align (lmflow.args.iterativedpoalignerarguments attribute)": [[4, "lmflow.args.IterativeDPOAlignerArguments.do_dpo_align", false]], "do_grad_scaling (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.do_grad_scaling", false]], "do_response_generation (lmflow.args.iterativedpoalignerarguments attribute)": [[4, "lmflow.args.IterativeDPOAlignerArguments.do_response_generation", false]], "do_rope_scaling (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.do_rope_scaling", false]], "do_sample (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.do_sample", false]], "do_scoring (lmflow.args.iterativedpoalignerarguments attribute)": [[4, "lmflow.args.IterativeDPOAlignerArguments.do_scoring", false]], "do_train (lmflow.models.hf_model_mixin.hfmodelmixin attribute)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.do_train", false]], "double_quant (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.double_quant", false]], "dpo_loss() (lmflow.pipeline.utils.dpov2_trainer.dpov2trainer method)": [[61, "lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.dpo_loss", false]], "dpoaligner (class in lmflow.pipeline.dpo_aligner)": [[50, "lmflow.pipeline.dpo_aligner.DPOAligner", false]], "dpoalignerarguments (class in lmflow.args)": [[4, "lmflow.args.DPOAlignerArguments", false]], "dpov2aligner (class in lmflow.pipeline.dpov2_aligner)": [[51, "lmflow.pipeline.dpov2_aligner.DPOv2Aligner", false]], "dpov2alignerarguments (class in lmflow.args)": [[4, "lmflow.args.DPOv2AlignerArguments", false]], "dpov2trainer (class in lmflow.pipeline.utils.dpov2_trainer)": [[61, "lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer", false]], "draft_config (lmflow.pipeline.inferencer.speculativeinferencer attribute)": [[55, "lmflow.pipeline.inferencer.SpeculativeInferencer.draft_config", false]], "draft_model_args (lmflow.pipeline.inferencer.speculativeinferencer attribute)": [[55, "lmflow.pipeline.inferencer.SpeculativeInferencer.draft_model_args", false]], "drop_instances() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.drop_instances", false]], "drop_instances() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.drop_instances", false]], "ds_config (lmflow.models.hf_model_mixin.hfmodelmixin attribute)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.ds_config", false]], "dtype (lmflow.models.vision_encoder.clip_encoder.clipvisiontower property)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.dtype", false]], "dummy (class in lmflow.optim.dummy)": [[34, "lmflow.optim.dummy.Dummy", false]], "dummy (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.DUMMY", false]], "dummy_feature (lmflow.models.vision_encoder.clip_encoder.clipvisiontower property)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.dummy_feature", false]], "empty_no_special_tokens_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE", false]], "empty_no_special_tokens_template (in module lmflow.utils.conversation_template.base)": [[75, "lmflow.utils.conversation_template.base.EMPTY_NO_SPECIAL_TOKENS_TEMPLATE", false]], "empty_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.EMPTY_TEMPLATE", false]], "empty_template (in module lmflow.utils.conversation_template.base)": [[75, "lmflow.utils.conversation_template.base.EMPTY_TEMPLATE", false]], "emptyformatter (class in lmflow.utils.conversation_template.base)": [[75, "lmflow.utils.conversation_template.base.EmptyFormatter", false]], "enable_decode_inference_result (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.enable_decode_inference_result", false]], "enable_distributed_inference (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.enable_distributed_inference", false]], "encode() (lmflow.models.hf_decoder_model.hfdecodermodel method)": [[13, "lmflow.models.hf_decoder_model.HFDecoderModel.encode", false]], "encode() (lmflow.models.hf_encoder_decoder_model.hfencoderdecodermodel method)": [[14, "lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.encode", false]], "encode_conversation() (lmflow.utils.conversation_template.base.conversationtemplate method)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.encode_conversation", false]], "encode_conversation() (lmflow.utils.conversation_template.base.conversationtemplatefortool method)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplateForTool.encode_conversation", false]], "encode_conversation() (lmflow.utils.conversation_template.conversationtemplate method)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.encode_conversation", false]], "encode_conversation() (lmflow.utils.conversation_template.conversationtemplatefortool method)": [[81, "lmflow.utils.conversation_template.ConversationTemplateForTool.encode_conversation", false]], "encode_conversation() (lmflow.utils.conversation_template.gemma.gemmaconversationtemplate method)": [[79, "lmflow.utils.conversation_template.gemma.GemmaConversationTemplate.encode_conversation", false]], "encode_conversation() (lmflow.utils.conversation_template.hymba.hymbaconversationtemplate method)": [[80, "lmflow.utils.conversation_template.hymba.HymbaConversationTemplate.encode_conversation", false]], "encode_images() (lmflow.models.vision_encoder.clip_encoder.clipvisiontower method)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.encode_images", false]], "encoderdecodermodel (class in lmflow.models.encoder_decoder_model)": [[12, "lmflow.models.encoder_decoder_model.EncoderDecoderModel", false]], "eos_padding (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.eos_padding", false]], "eos_token_id (lmflow.pipeline.vllm_inferencer.inferencerwithoffloading attribute)": [[69, "lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.eos_token_id", false]], "eval() (lmflow.optim.adamw_schedule_free.adamwschedulefree method)": [[32, "lmflow.optim.adamw_schedule_free.AdamWScheduleFree.eval", false]], "eval() (lmflow.optim.sgd_schedule_free.sgdschedulefree method)": [[42, "lmflow.optim.sgd_schedule_free.SGDScheduleFree.eval", false]], "eval_dataset (lmflow.pipeline.dpo_aligner.dpoaligner attribute)": [[50, "lmflow.pipeline.dpo_aligner.DPOAligner.eval_dataset", false]], "eval_dataset (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.eval_dataset", false]], "eval_dataset_path (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.eval_dataset_path", false]], "eval_dataset_path (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.eval_dataset_path", false]], "eval_steps (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.eval_steps", false]], "evaluate() (lmflow.pipeline.evaluator.evaluator method)": [[52, "lmflow.pipeline.evaluator.Evaluator.evaluate", false]], "evaluate() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.evaluate", false]], "evaluate_block_size (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.evaluate_block_size", false]], "evaluation_loop() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.evaluation_loop", false]], "evaluator (class in lmflow.pipeline.evaluator)": [[52, "lmflow.pipeline.evaluator.Evaluator", false]], "evaluator_args (lmflow.pipeline.evaluator.evaluator attribute)": [[52, "lmflow.pipeline.evaluator.Evaluator.evaluator_args", false]], "evaluatorarguments (class in lmflow.args)": [[4, "lmflow.args.EvaluatorArguments", false]], "feature_select() (lmflow.models.vision_encoder.clip_encoder.clipvisiontower method)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.feature_select", false]], "finetune_part (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.finetune_part", false]], "finetuner (class in lmflow.pipeline.finetuner)": [[53, "lmflow.pipeline.finetuner.Finetuner", false]], "finetuner_args (lmflow.pipeline.finetuner.finetuner attribute)": [[53, "lmflow.pipeline.finetuner.Finetuner.finetuner_args", false]], "finetunerarguments (class in lmflow.args)": [[4, "lmflow.args.FinetunerArguments", false]], "fixed_decay (lmflow.optim.adabelief.adabelief attribute)": [[25, "lmflow.optim.adabelief.AdaBelief.fixed_decay", false]], "flash_attn_func (in module lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention.flash_attn_func", false]], "flash_attn_kvpacked_func (in module lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention.flash_attn_kvpacked_func", false]], "flash_attn_qkvpacked_func (in module lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention.flash_attn_qkvpacked_func", false]], "flashattnfunc (class in lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc", false]], "flashattnkvpackedfunc (class in lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc", false]], "flashattnqkvpackedfunc (class in lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc", false]], "flatten_list() (lmflow.pipeline.rm_inferencer.rewardmodelinferencer method)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer.flatten_list", false]], "float_only_dataset_description (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.FLOAT_ONLY_DATASET_DESCRIPTION", false]], "floating_point_ops() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.floating_point_ops", false]], "force_system (lmflow.utils.conversation_template.base.conversationtemplate attribute)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.force_system", false]], "force_system (lmflow.utils.conversation_template.conversationtemplate attribute)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.force_system", false]], "format() (lmflow.utils.conversation_template.base.emptyformatter method)": [[75, "lmflow.utils.conversation_template.base.EmptyFormatter.format", false]], "format() (lmflow.utils.conversation_template.base.formatter method)": [[75, "lmflow.utils.conversation_template.base.Formatter.format", false]], "format() (lmflow.utils.conversation_template.base.listformatter method)": [[75, "lmflow.utils.conversation_template.base.ListFormatter.format", false]], "format() (lmflow.utils.conversation_template.base.stringformatter method)": [[75, "lmflow.utils.conversation_template.base.StringFormatter.format", false]], "formatter (class in lmflow.utils.conversation_template.base)": [[75, "lmflow.utils.conversation_template.base.Formatter", false]], "forward() (in module lmflow.utils.flash_attention.bloom_flash_attention)": [[89, "lmflow.utils.flash_attention.bloom_flash_attention.forward", false]], "forward() (in module lmflow.utils.flash_attention.gpt2_flash_attention)": [[90, "lmflow.utils.flash_attention.gpt2_flash_attention.forward", false]], "forward() (in module lmflow.utils.flash_attention.gpt_neo_flash_attention)": [[91, "lmflow.utils.flash_attention.gpt_neo_flash_attention.forward", false]], "forward() (in module lmflow.utils.flash_attention.llama_flash_attention)": [[93, "lmflow.utils.flash_attention.llama_flash_attention.forward", false]], "forward() (lmflow.models.vision2seq_model.customautovision2seqmodel method)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.forward", false]], "forward() (lmflow.models.vision_encoder.clip_encoder.clipvisiontower method)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.forward", false]], "forward() (lmflow.utils.flash_attention.triton_flash_attention.flashattnfunc static method)": [[94, "lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc.forward", false]], "forward() (lmflow.utils.flash_attention.triton_flash_attention.flashattnkvpackedfunc static method)": [[94, "lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc.forward", false]], "forward() (lmflow.utils.flash_attention.triton_flash_attention.flashattnqkvpackedfunc static method)": [[94, "lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc.forward", false]], "forward() (lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.condenserotaryembedding method)": [[100, "lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.forward", false]], "from_dict() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.from_dict", false]], "from_dict() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.from_dict", false]], "fsdp (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.fsdp", false]], "function_formatter (lmflow.utils.conversation_template.base.conversationtemplate attribute)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.function_formatter", false]], "function_formatter (lmflow.utils.conversation_template.conversationtemplate attribute)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.function_formatter", false]], "gemma_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.GEMMA_TEMPLATE", false]], "gemma_template (in module lmflow.utils.conversation_template.gemma)": [[79, "lmflow.utils.conversation_template.gemma.GEMMA_TEMPLATE", false]], "gemmaconversationtemplate (class in lmflow.utils.conversation_template.gemma)": [[79, "lmflow.utils.conversation_template.gemma.GemmaConversationTemplate", false]], "generate() (lmflow.models.vision2seq_model.customautovision2seqmodel method)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.generate", false]], "get_backend() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.get_backend", false]], "get_backend() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.get_backend", false]], "get_backend_dataset() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.get_backend_dataset", false]], "get_backend_dataset() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.get_backend_dataset", false]], "get_backend_model() (lmflow.models.hf_encoder_decoder_model.hfencoderdecodermodel method)": [[14, "lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_backend_model", false]], "get_backend_model() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.get_backend_model", false]], "get_backend_model() (lmflow.models.vision2seq_model.customautovision2seqmodel method)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.get_backend_model", false]], "get_batch_loss_metrics() (lmflow.pipeline.utils.dpov2_trainer.dpov2trainer method)": [[61, "lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.get_batch_loss_metrics", false]], "get_batch_metrics() (lmflow.pipeline.utils.dpov2_trainer.dpov2trainer method)": [[61, "lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.get_batch_metrics", false]], "get_data_args() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.get_data_args", false]], "get_data_args() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.get_data_args", false]], "get_eval_dataloader() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_eval_dataloader", false]], "get_fingerprint() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.get_fingerprint", false]], "get_fingerprint() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.get_fingerprint", false]], "get_images() (lmflow.utils.llava_conversation_lib.conversation method)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.get_images", false]], "get_max_length() (lmflow.models.hf_encoder_decoder_model.hfencoderdecodermodel method)": [[14, "lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_max_length", false]], "get_max_length() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.get_max_length", false]], "get_model() (lmflow.models.auto_model.automodel class method)": [[9, "lmflow.models.auto_model.AutoModel.get_model", false]], "get_optimizer_cls_and_kwargs() (lmflow.pipeline.utils.raft_trainer.rafttrainer static method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_optimizer_cls_and_kwargs", false]], "get_paired_dataset() (in module lmflow.pipeline.dpo_aligner)": [[50, "lmflow.pipeline.dpo_aligner.get_paired_dataset", false]], "get_peft_without_qlora() (lmflow.models.hf_decoder_model.hfdecodermodel method)": [[13, "lmflow.models.hf_decoder_model.HFDecoderModel.get_peft_without_qlora", false]], "get_pipeline() (lmflow.pipeline.auto_pipeline.autopipeline class method)": [[46, "lmflow.pipeline.auto_pipeline.AutoPipeline.get_pipeline", false]], "get_pipeline_args_class() (lmflow.args.autoarguments method)": [[4, "lmflow.args.AutoArguments.get_pipeline_args_class", false]], "get_prompt() (lmflow.utils.llava_conversation_lib.conversation method)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.get_prompt", false]], "get_python_version() (in module lmflow.utils.versioning)": [[101, "lmflow.utils.versioning.get_python_version", false]], "get_test_dataloader() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_test_dataloader", false]], "get_tokenizer() (lmflow.models.hf_encoder_decoder_model.hfencoderdecodermodel method)": [[14, "lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.get_tokenizer", false]], "get_tokenizer() (lmflow.models.hf_model_mixin.hfmodelmixin method)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.get_tokenizer", false]], "get_tokenizer() (lmflow.models.vision2seq_model.customautovision2seqmodel method)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.get_tokenizer", false]], "get_train_dataloader() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.get_train_dataloader", false]], "get_type() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.get_type", false]], "get_type() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.get_type", false]], "gradient_accumulation_steps (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.gradient_accumulation_steps", false]], "gradient_checkpointing (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.gradient_checkpointing", false]], "gradient_checkpointing_use_reentrant (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.gradient_checkpointing_use_reentrant", false]], "group_text() (lmflow.pipeline.finetuner.finetuner method)": [[53, "lmflow.pipeline.finetuner.Finetuner.group_text", false]], "group_texts_batch_size (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.group_texts_batch_size", false]], "has_placeholder() (lmflow.utils.conversation_template.base.formatter method)": [[75, "lmflow.utils.conversation_template.base.Formatter.has_placeholder", false]], "hf_auto_model (lmflow.models.hf_model_mixin.hfmodelmixin attribute)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.hf_auto_model", false]], "hf_automodel_mapping (in module lmflow.models.hf_model_mixin)": [[15, "lmflow.models.hf_model_mixin.HF_AUTOMODEL_MAPPING", false]], "hf_automodel_type (in module lmflow.models.hf_model_mixin)": [[15, "lmflow.models.hf_model_mixin.HF_AUTOMODEL_TYPE", false]], "hf_dataset_sanity_check() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.hf_dataset_sanity_check", false]], "hf_dataset_sanity_check() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.hf_dataset_sanity_check", false]], "hf_model_config (lmflow.models.hf_model_mixin.hfmodelmixin attribute)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.hf_model_config", false]], "hfdecodermodel (class in lmflow.models.hf_decoder_model)": [[13, "lmflow.models.hf_decoder_model.HFDecoderModel", false]], "hfencoderdecodermodel (class in lmflow.models.hf_encoder_decoder_model)": [[14, "lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel", false]], "hfmodelmixin (class in lmflow.models.hf_model_mixin)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin", false]], "hftextregressionmodel (class in lmflow.models.hf_text_regression_model)": [[16, "lmflow.models.hf_text_regression_model.HFTextRegressionModel", false]], "hidden_size (lmflow.models.vision2seq_model.customautovision2seqmodel attribute)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.hidden_size", false]], "hidden_size (lmflow.models.vision_encoder.clip_encoder.clipvisiontower property)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.hidden_size", false]], "hp_name (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.hp_name", false]], "hp_search_backend (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.hp_search_backend", false]], "hymba_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.HYMBA_TEMPLATE", false]], "hymba_template (in module lmflow.utils.conversation_template.hymba)": [[80, "lmflow.utils.conversation_template.hymba.HYMBA_TEMPLATE", false]], "hymbaconversationtemplate (class in lmflow.utils.conversation_template.hymba)": [[80, "lmflow.utils.conversation_template.hymba.HymbaConversationTemplate", false]], "hyperparameter_search() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.hyperparameter_search", false]], "ignore_bias_buffers (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.ignore_bias_buffers", false]], "ignore_index (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.IGNORE_INDEX", false]], "image_aspect_ratio (lmflow.args.multimodaldatasetarguments attribute)": [[4, "lmflow.args.MultiModalDatasetArguments.image_aspect_ratio", false]], "image_encoder_name_or_path (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.image_encoder_name_or_path", false]], "image_folder (lmflow.args.multimodaldatasetarguments attribute)": [[4, "lmflow.args.MultiModalDatasetArguments.image_folder", false]], "image_folder (lmflow.datasets.multi_modal_dataset.custommultimodaldataset attribute)": [[7, "lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.image_folder", false]], "image_token_index (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.IMAGE_TOKEN_INDEX", false]], "inf (lmflow.pipeline.raft_aligner.raftaligner attribute)": [[57, "lmflow.pipeline.raft_aligner.RaftAligner.INF", false]], "inference() (lmflow.models.hf_decoder_model.hfdecodermodel method)": [[13, "lmflow.models.hf_decoder_model.HFDecoderModel.inference", false]], "inference() (lmflow.models.hf_encoder_decoder_model.hfencoderdecodermodel method)": [[14, "lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.inference", false]], "inference() (lmflow.models.hf_text_regression_model.hftextregressionmodel method)": [[16, "lmflow.models.hf_text_regression_model.HFTextRegressionModel.inference", false]], "inference() (lmflow.models.text_regression_model.textregressionmodel method)": [[21, "lmflow.models.text_regression_model.TextRegressionModel.inference", false]], "inference() (lmflow.pipeline.inferencer.inferencer method)": [[55, "lmflow.pipeline.inferencer.Inferencer.inference", false]], "inference() (lmflow.pipeline.inferencer.speculativeinferencer method)": [[55, "lmflow.pipeline.inferencer.SpeculativeInferencer.inference", false]], "inference() (lmflow.pipeline.inferencer.toolinferencer method)": [[55, "lmflow.pipeline.inferencer.ToolInferencer.inference", false]], "inference() (lmflow.pipeline.rm_inferencer.rewardmodelinferencer method)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer.inference", false]], "inference() (lmflow.pipeline.vllm_inferencer.inferencerwithoffloading method)": [[69, "lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.inference", false]], "inference() (lmflow.pipeline.vllm_inferencer.memorysafevllminferencer method)": [[69, "lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer.inference", false]], "inference() (lmflow.pipeline.vllm_inferencer.vllminferencer method)": [[69, "lmflow.pipeline.vllm_inferencer.VLLMInferencer.inference", false]], "inference_batch_size (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.inference_batch_size", false]], "inference_batch_size_per_device (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.inference_batch_size_per_device", false]], "inference_batch_size_per_device (lmflow.args.raftalignerarguments attribute)": [[4, "lmflow.args.RaftAlignerArguments.inference_batch_size_per_device", false]], "inference_func (lmflow.models.text_regression_model.textregressionmodel attribute)": [[21, "lmflow.models.text_regression_model.TextRegressionModel.inference_func", false]], "inferencer (class in lmflow.pipeline.inferencer)": [[55, "lmflow.pipeline.inferencer.Inferencer", false]], "inferencer_args (lmflow.pipeline.inferencer.inferencer attribute)": [[55, "lmflow.pipeline.inferencer.Inferencer.inferencer_args", false]], "inferencer_args (lmflow.pipeline.rm_inferencer.rewardmodelinferencer attribute)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer.inferencer_args", false]], "inferencer_args (lmflow.pipeline.vllm_inferencer.inferencerwithoffloading attribute)": [[69, "lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.inferencer_args", false]], "inferencer_file_path (lmflow.pipeline.vllm_inferencer.memorysafevllminferencer attribute)": [[69, "lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer.inferencer_file_path", false]], "inferencerarguments (class in lmflow.args)": [[4, "lmflow.args.InferencerArguments", false]], "inferencerwithoffloading (class in lmflow.pipeline.vllm_inferencer)": [[69, "lmflow.pipeline.vllm_inferencer.InferencerWithOffloading", false]], "init_git_repo() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.init_git_repo", false]], "init_to_zero() (in module lmflow.utils.flash_attention.triton_flash_attention)": [[94, "lmflow.utils.flash_attention.triton_flash_attention.init_to_zero", false]], "initial_iter_idx (lmflow.args.iterativealignerarguments attribute)": [[4, "lmflow.args.IterativeAlignerArguments.initial_iter_idx", false]], "input (lmflow.utils.data_utils.rewardmodelinferenceresultwithinput attribute)": [[88, "lmflow.utils.data_utils.RewardModelInferenceResultWithInput.input", false]], "input (lmflow.utils.data_utils.vllminferenceresultwithinput attribute)": [[88, "lmflow.utils.data_utils.VLLMInferenceResultWithInput.input", false]], "instance_fields_map (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.INSTANCE_FIELDS_MAP", false]], "internal_version (in module lmflow)": [[8, "lmflow.internal_version", false]], "internlm2_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.INTERNLM2_TEMPLATE", false]], "internlm2_template (in module lmflow.utils.conversation_template.internlm)": [[82, "lmflow.utils.conversation_template.internlm.INTERNLM2_TEMPLATE", false]], "ipex_optimize_model() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.ipex_optimize_model", false]], "is_custom_dataset (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.is_custom_dataset", false]], "is_encoder_decoder (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding attribute)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.is_encoder_decoder", false]], "is_flash_attn_available() (in module lmflow.utils.versioning)": [[101, "lmflow.utils.versioning.is_flash_attn_available", false]], "is_flask_available() (in module lmflow.utils.versioning)": [[101, "lmflow.utils.versioning.is_flask_available", false]], "is_gradio_available() (in module lmflow.utils.versioning)": [[101, "lmflow.utils.versioning.is_gradio_available", false]], "is_in_train (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_in_train", false]], "is_loaded (lmflow.models.vision_encoder.clip_encoder.clipvisiontower attribute)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.is_loaded", false]], "is_local_process_zero() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_local_process_zero", false]], "is_multimodal (lmflow.args.multimodaldatasetarguments attribute)": [[4, "lmflow.args.MultiModalDatasetArguments.is_multimodal", false]], "is_multimodal_available() (in module lmflow.datasets)": [[6, "lmflow.datasets.is_multimodal_available", false]], "is_multimodal_available() (in module lmflow.utils.versioning)": [[101, "lmflow.utils.versioning.is_multimodal_available", false]], "is_package_version_at_least() (in module lmflow.utils.versioning)": [[101, "lmflow.utils.versioning.is_package_version_at_least", false]], "is_ray_available() (in module lmflow.utils.versioning)": [[101, "lmflow.utils.versioning.is_ray_available", false]], "is_sagemaker_mp_post_1_10 (in module lmflow.pipeline.utils.raft_trainer)": [[66, "lmflow.pipeline.utils.raft_trainer.IS_SAGEMAKER_MP_POST_1_10", false]], "is_torch_greater_or_equal_than_1_10 (in module lmflow.pipeline.utils.raft_trainer)": [[66, "lmflow.pipeline.utils.raft_trainer.is_torch_greater_or_equal_than_1_10", false]], "is_torch_less_than_1_11 (in module lmflow.pipeline.utils.raft_trainer)": [[66, "lmflow.pipeline.utils.raft_trainer.is_torch_less_than_1_11", false]], "is_trl_available() (in module lmflow.utils.versioning)": [[101, "lmflow.utils.versioning.is_trl_available", false]], "is_vllm_available() (in module lmflow.utils.versioning)": [[101, "lmflow.utils.versioning.is_vllm_available", false]], "is_world_process_zero() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.is_world_process_zero", false]], "iterativealignerarguments (class in lmflow.args)": [[4, "lmflow.args.IterativeAlignerArguments", false]], "iterativedpoaligner (class in lmflow.pipeline.iterative_dpo_aligner)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner", false]], "iterativedpoalignerarguments (class in lmflow.args)": [[4, "lmflow.args.IterativeDPOAlignerArguments", false]], "keep_linebreaks (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.keep_linebreaks", false]], "key_instances (in module lmflow.datasets.dataset)": [[5, "lmflow.datasets.dataset.KEY_INSTANCES", false]], "key_score (in module lmflow.datasets.dataset)": [[5, "lmflow.datasets.dataset.KEY_SCORE", false]], "key_type (in module lmflow.datasets.dataset)": [[5, "lmflow.datasets.dataset.KEY_TYPE", false]], "label_names (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.label_names", false]], "label_pad_token_id (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding attribute)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.label_pad_token_id", false]], "lamb (class in lmflow.optim.lamb)": [[36, "lmflow.optim.lamb.Lamb", false]], "lamb (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.LAMB", false]], "language_model (lmflow.models.vision2seq_model.customautovision2seqmodel attribute)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.language_model", false]], "language_model_from_pretrained() (lmflow.models.vision2seq_model.customautovision2seqmodel method)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.language_model_from_pretrained", false]], "lars (class in lmflow.optim.lars)": [[37, "lmflow.optim.lars.LARS", false]], "lars (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.LARS", false]], "last_checkpoint (lmflow.pipeline.finetuner.finetuner attribute)": [[53, "lmflow.pipeline.finetuner.Finetuner.last_checkpoint", false]], "learning_rate (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.learning_rate", false]], "len_penalty (lmflow.pipeline.utils.dpov2_trainer.dpov2trainer attribute)": [[61, "lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.len_penalty", false]], "length_penalty (lmflow.args.dpov2alignerarguments attribute)": [[4, "lmflow.args.DPOv2AlignerArguments.length_penalty", false]], "lisa_activated_layers (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.lisa_activated_layers", false]], "lisa_interval_steps (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.lisa_interval_steps", false]], "lisa_layers_attribute (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.lisa_layers_attribute", false]], "listformatter (class in lmflow.utils.conversation_template.base)": [[75, "lmflow.utils.conversation_template.base.ListFormatter", false]], "llama2_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.LLAMA2_TEMPLATE", false]], "llama2_template (in module lmflow.utils.conversation_template.llama)": [[83, "lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE", false]], "llama2_template_for_tool (in module lmflow.utils.conversation_template.llama)": [[83, "lmflow.utils.conversation_template.llama.LLAMA2_TEMPLATE_FOR_TOOL", false]], "llama2conversationtemplate (class in lmflow.utils.conversation_template.llama)": [[83, "lmflow.utils.conversation_template.llama.Llama2ConversationTemplate", false]], "llama2conversationtemplatefortool (class in lmflow.utils.conversation_template.llama)": [[83, "lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool", false]], "llama3_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.LLAMA3_TEMPLATE", false]], "llama3_template (in module lmflow.utils.conversation_template.llama)": [[83, "lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE", false]], "llama3_template_for_tool (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.LLAMA3_TEMPLATE_FOR_TOOL", false]], "llama3_template_for_tool (in module lmflow.utils.conversation_template.llama)": [[83, "lmflow.utils.conversation_template.llama.LLAMA3_TEMPLATE_FOR_TOOL", false]], "llama_2 (lmflow.utils.llava_conversation_lib.separatorstyle attribute)": [[96, "lmflow.utils.llava_conversation_lib.SeparatorStyle.LLAMA_2", false]], "llava_loading (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.llava_loading", false]], "llava_pretrain_model_path (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.llava_pretrain_model_path", false]], "llm_model_name_or_path (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.llm_model_name_or_path", false]], "lm_evaluation_metric (lmflow.args.benchmarkingarguments attribute)": [[4, "lmflow.args.BenchmarkingArguments.lm_evaluation_metric", false]], "lmflow": [[8, "module-lmflow", false]], "lmflow.args": [[4, "module-lmflow.args", false]], "lmflow.datasets": [[6, "module-lmflow.datasets", false]], "lmflow.datasets.dataset": [[5, "module-lmflow.datasets.dataset", false]], "lmflow.datasets.multi_modal_dataset": [[7, "module-lmflow.datasets.multi_modal_dataset", false]], "lmflow.models": [[17, "module-lmflow.models", false]], "lmflow.models.auto_model": [[9, "module-lmflow.models.auto_model", false]], "lmflow.models.base_model": [[10, "module-lmflow.models.base_model", false]], "lmflow.models.decoder_model": [[11, "module-lmflow.models.decoder_model", false]], "lmflow.models.encoder_decoder_model": [[12, "module-lmflow.models.encoder_decoder_model", false]], "lmflow.models.hf_decoder_model": [[13, "module-lmflow.models.hf_decoder_model", false]], "lmflow.models.hf_encoder_decoder_model": [[14, "module-lmflow.models.hf_encoder_decoder_model", false]], "lmflow.models.hf_model_mixin": [[15, "module-lmflow.models.hf_model_mixin", false]], "lmflow.models.hf_text_regression_model": [[16, "module-lmflow.models.hf_text_regression_model", false]], "lmflow.models.interfaces": [[18, "module-lmflow.models.interfaces", false]], "lmflow.models.interfaces.tunable": [[19, "module-lmflow.models.interfaces.tunable", false]], "lmflow.models.regression_model": [[20, "module-lmflow.models.regression_model", false]], "lmflow.models.text_regression_model": [[21, "module-lmflow.models.text_regression_model", false]], "lmflow.models.vision2seq_model": [[22, "module-lmflow.models.vision2seq_model", false]], "lmflow.models.vision_encoder": [[24, "module-lmflow.models.vision_encoder", false]], "lmflow.models.vision_encoder.clip_encoder": [[23, "module-lmflow.models.vision_encoder.clip_encoder", false]], "lmflow.optim": [[35, "module-lmflow.optim", false]], "lmflow.optim.adabelief": [[25, "module-lmflow.optim.adabelief", false]], "lmflow.optim.adabound": [[26, "module-lmflow.optim.adabound", false]], "lmflow.optim.adadelta": [[27, "module-lmflow.optim.adadelta", false]], "lmflow.optim.adagrad": [[28, "module-lmflow.optim.adagrad", false]], "lmflow.optim.adam": [[29, "module-lmflow.optim.adam", false]], "lmflow.optim.adamax": [[30, "module-lmflow.optim.adamax", false]], "lmflow.optim.adamp": [[31, "module-lmflow.optim.adamp", false]], "lmflow.optim.adamw_schedule_free": [[32, "module-lmflow.optim.adamw_schedule_free", false]], "lmflow.optim.adan": [[33, "module-lmflow.optim.adan", false]], "lmflow.optim.dummy": [[34, "module-lmflow.optim.dummy", false]], "lmflow.optim.lamb": [[36, "module-lmflow.optim.lamb", false]], "lmflow.optim.lars": [[37, "module-lmflow.optim.lars", false]], "lmflow.optim.nadam": [[38, "module-lmflow.optim.nadam", false]], "lmflow.optim.novograd": [[39, "module-lmflow.optim.novograd", false]], "lmflow.optim.optimizers": [[40, "module-lmflow.optim.optimizers", false]], "lmflow.optim.radam": [[41, "module-lmflow.optim.radam", false]], "lmflow.optim.sgd_schedule_free": [[42, "module-lmflow.optim.sgd_schedule_free", false]], "lmflow.optim.sgdp": [[43, "module-lmflow.optim.sgdp", false]], "lmflow.optim.sophia": [[44, "module-lmflow.optim.sophia", false]], "lmflow.optim.yogi": [[45, "module-lmflow.optim.yogi", false]], "lmflow.pipeline": [[54, "module-lmflow.pipeline", false]], "lmflow.pipeline.auto_pipeline": [[46, "module-lmflow.pipeline.auto_pipeline", false]], "lmflow.pipeline.base_aligner": [[47, "module-lmflow.pipeline.base_aligner", false]], "lmflow.pipeline.base_pipeline": [[48, "module-lmflow.pipeline.base_pipeline", false]], "lmflow.pipeline.base_tuner": [[49, "module-lmflow.pipeline.base_tuner", false]], "lmflow.pipeline.dpo_aligner": [[50, "module-lmflow.pipeline.dpo_aligner", false]], "lmflow.pipeline.dpov2_aligner": [[51, "module-lmflow.pipeline.dpov2_aligner", false]], "lmflow.pipeline.evaluator": [[52, "module-lmflow.pipeline.evaluator", false]], "lmflow.pipeline.finetuner": [[53, "module-lmflow.pipeline.finetuner", false]], "lmflow.pipeline.inferencer": [[55, "module-lmflow.pipeline.inferencer", false]], "lmflow.pipeline.iterative_dpo_aligner": [[56, "module-lmflow.pipeline.iterative_dpo_aligner", false]], "lmflow.pipeline.raft_aligner": [[57, "module-lmflow.pipeline.raft_aligner", false]], "lmflow.pipeline.rm_inferencer": [[58, "module-lmflow.pipeline.rm_inferencer", false]], "lmflow.pipeline.rm_tuner": [[59, "module-lmflow.pipeline.rm_tuner", false]], "lmflow.pipeline.utils": [[62, "module-lmflow.pipeline.utils", false]], "lmflow.pipeline.utils.dpov2_dataprocessor": [[60, "module-lmflow.pipeline.utils.dpov2_dataprocessor", false]], "lmflow.pipeline.utils.dpov2_trainer": [[61, "module-lmflow.pipeline.utils.dpov2_trainer", false]], "lmflow.pipeline.utils.memory_safe_dpov2_align": [[63, "module-lmflow.pipeline.utils.memory_safe_dpov2_align", false]], "lmflow.pipeline.utils.memory_safe_vllm_inference": [[64, "module-lmflow.pipeline.utils.memory_safe_vllm_inference", false]], "lmflow.pipeline.utils.peft_trainer": [[65, "module-lmflow.pipeline.utils.peft_trainer", false]], "lmflow.pipeline.utils.raft_trainer": [[66, "module-lmflow.pipeline.utils.raft_trainer", false]], "lmflow.pipeline.utils.rm_dataprocessor": [[67, "module-lmflow.pipeline.utils.rm_dataprocessor", false]], "lmflow.pipeline.utils.rm_trainer": [[68, "module-lmflow.pipeline.utils.rm_trainer", false]], "lmflow.pipeline.vllm_inferencer": [[69, "module-lmflow.pipeline.vllm_inferencer", false]], "lmflow.tokenization": [[72, "module-lmflow.tokenization", false]], "lmflow.tokenization.hf_decoder_model": [[70, "module-lmflow.tokenization.hf_decoder_model", false]], "lmflow.tokenization.hf_text_regression_model": [[71, "module-lmflow.tokenization.hf_text_regression_model", false]], "lmflow.utils": [[95, "module-lmflow.utils", false]], "lmflow.utils.common": [[73, "module-lmflow.utils.common", false]], "lmflow.utils.constants": [[74, "module-lmflow.utils.constants", false]], "lmflow.utils.conversation_template": [[81, "module-lmflow.utils.conversation_template", false]], "lmflow.utils.conversation_template.base": [[75, "module-lmflow.utils.conversation_template.base", false]], "lmflow.utils.conversation_template.chatglm": [[76, "module-lmflow.utils.conversation_template.chatglm", false]], "lmflow.utils.conversation_template.chatml": [[77, "module-lmflow.utils.conversation_template.chatml", false]], "lmflow.utils.conversation_template.deepseek": [[78, "module-lmflow.utils.conversation_template.deepseek", false]], "lmflow.utils.conversation_template.gemma": [[79, "module-lmflow.utils.conversation_template.gemma", false]], "lmflow.utils.conversation_template.hymba": [[80, "module-lmflow.utils.conversation_template.hymba", false]], "lmflow.utils.conversation_template.internlm": [[82, "module-lmflow.utils.conversation_template.internlm", false]], "lmflow.utils.conversation_template.llama": [[83, "module-lmflow.utils.conversation_template.llama", false]], "lmflow.utils.conversation_template.phi": [[84, "module-lmflow.utils.conversation_template.phi", false]], "lmflow.utils.conversation_template.qwen": [[85, "module-lmflow.utils.conversation_template.qwen", false]], "lmflow.utils.conversation_template.yi": [[86, "module-lmflow.utils.conversation_template.yi", false]], "lmflow.utils.conversation_template.zephyr": [[87, "module-lmflow.utils.conversation_template.zephyr", false]], "lmflow.utils.data_utils": [[88, "module-lmflow.utils.data_utils", false]], "lmflow.utils.flash_attention": [[92, "module-lmflow.utils.flash_attention", false]], "lmflow.utils.flash_attention.bloom_flash_attention": [[89, "module-lmflow.utils.flash_attention.bloom_flash_attention", false]], "lmflow.utils.flash_attention.gpt2_flash_attention": [[90, "module-lmflow.utils.flash_attention.gpt2_flash_attention", false]], "lmflow.utils.flash_attention.gpt_neo_flash_attention": [[91, "module-lmflow.utils.flash_attention.gpt_neo_flash_attention", false]], "lmflow.utils.flash_attention.llama_flash_attention": [[93, "module-lmflow.utils.flash_attention.llama_flash_attention", false]], "lmflow.utils.flash_attention.triton_flash_attention": [[94, "module-lmflow.utils.flash_attention.triton_flash_attention", false]], "lmflow.utils.llava_conversation_lib": [[96, "module-lmflow.utils.llava_conversation_lib", false]], "lmflow.utils.model": [[97, "module-lmflow.utils.model", false]], "lmflow.utils.multimodal": [[98, "module-lmflow.utils.multimodal", false]], "lmflow.utils.position_interpolation": [[99, "module-lmflow.utils.position_interpolation", false]], "lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch": [[100, "module-lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch", false]], "lmflow.utils.versioning": [[101, "module-lmflow.utils.versioning", false]], "lmflow.version": [[102, "module-lmflow.version", false]], "lmflow_lora_target_modules_mapping (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.LMFLOW_LORA_TARGET_MODULES_MAPPING", false]], "load_data() (in module lmflow.utils.data_utils)": [[88, "lmflow.utils.data_utils.load_data", false]], "load_in_4bit (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.load_in_4bit", false]], "load_inference_results() (lmflow.pipeline.vllm_inferencer.inferencerwithoffloading method)": [[69, "lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.load_inference_results", false]], "load_inference_results() (lmflow.pipeline.vllm_inferencer.vllminferencer method)": [[69, "lmflow.pipeline.vllm_inferencer.VLLMInferencer.load_inference_results", false]], "load_llava_pretrain_model() (in module lmflow.utils.multimodal)": [[98, "lmflow.utils.multimodal.load_llava_pretrain_model", false]], "load_model() (lmflow.models.vision_encoder.clip_encoder.clipvisiontower method)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.load_model", false]], "load_prompt_cache() (lmflow.models.vision2seq_model.customautovision2seqmodel method)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.load_prompt_cache", false]], "local_rank (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.local_rank", false]], "local_rank (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.local_rank", false]], "local_rank (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.local_rank", false]], "local_rank (lmflow.pipeline.evaluator.evaluator attribute)": [[52, "lmflow.pipeline.evaluator.Evaluator.local_rank", false]], "local_rank (lmflow.pipeline.inferencer.inferencer attribute)": [[55, "lmflow.pipeline.inferencer.Inferencer.local_rank", false]], "local_rank (lmflow.pipeline.rm_inferencer.rewardmodelinferencer attribute)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer.local_rank", false]], "log() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.log", false]], "log_freq (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.log_freq", false]], "logdir (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.LOGDIR", false]], "logger (in module lmflow.args)": [[4, "lmflow.args.logger", false]], "logger (in module lmflow.datasets.dataset)": [[5, "lmflow.datasets.dataset.logger", false]], "logger (in module lmflow.models.hf_decoder_model)": [[13, "lmflow.models.hf_decoder_model.logger", false]], "logger (in module lmflow.models.hf_encoder_decoder_model)": [[14, "lmflow.models.hf_encoder_decoder_model.logger", false]], "logger (in module lmflow.models.hf_model_mixin)": [[15, "lmflow.models.hf_model_mixin.logger", false]], "logger (in module lmflow.models.hf_text_regression_model)": [[16, "lmflow.models.hf_text_regression_model.logger", false]], "logger (in module lmflow.pipeline.dpov2_aligner)": [[51, "lmflow.pipeline.dpov2_aligner.logger", false]], "logger (in module lmflow.pipeline.finetuner)": [[53, "lmflow.pipeline.finetuner.logger", false]], "logger (in module lmflow.pipeline.inferencer)": [[55, "lmflow.pipeline.inferencer.logger", false]], "logger (in module lmflow.pipeline.iterative_dpo_aligner)": [[56, "lmflow.pipeline.iterative_dpo_aligner.logger", false]], "logger (in module lmflow.pipeline.raft_aligner)": [[57, "lmflow.pipeline.raft_aligner.logger", false]], "logger (in module lmflow.pipeline.rm_inferencer)": [[58, "lmflow.pipeline.rm_inferencer.logger", false]], "logger (in module lmflow.pipeline.rm_tuner)": [[59, "lmflow.pipeline.rm_tuner.logger", false]], "logger (in module lmflow.pipeline.utils.dpov2_dataprocessor)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.logger", false]], "logger (in module lmflow.pipeline.utils.dpov2_trainer)": [[61, "lmflow.pipeline.utils.dpov2_trainer.logger", false]], "logger (in module lmflow.pipeline.utils.memory_safe_dpov2_align)": [[63, "lmflow.pipeline.utils.memory_safe_dpov2_align.logger", false]], "logger (in module lmflow.pipeline.utils.memory_safe_vllm_inference)": [[64, "lmflow.pipeline.utils.memory_safe_vllm_inference.logger", false]], "logger (in module lmflow.pipeline.utils.raft_trainer)": [[66, "lmflow.pipeline.utils.raft_trainer.logger", false]], "logger (in module lmflow.pipeline.utils.rm_dataprocessor)": [[67, "lmflow.pipeline.utils.rm_dataprocessor.logger", false]], "logger (in module lmflow.pipeline.vllm_inferencer)": [[69, "lmflow.pipeline.vllm_inferencer.logger", false]], "logger (in module lmflow.tokenization.hf_decoder_model)": [[70, "lmflow.tokenization.hf_decoder_model.logger", false]], "logger (in module lmflow.tokenization.hf_text_regression_model)": [[71, "lmflow.tokenization.hf_text_regression_model.logger", false]], "logger (in module lmflow.utils.common)": [[73, "lmflow.utils.common.logger", false]], "logger (in module lmflow.utils.conversation_template.base)": [[75, "lmflow.utils.conversation_template.base.logger", false]], "logger (in module lmflow.utils.conversation_template.gemma)": [[79, "lmflow.utils.conversation_template.gemma.logger", false]], "logger (in module lmflow.utils.conversation_template.llama)": [[83, "lmflow.utils.conversation_template.llama.logger", false]], "logger (in module lmflow.utils.conversation_template.zephyr)": [[87, "lmflow.utils.conversation_template.zephyr.logger", false]], "logger (in module lmflow.utils.model)": [[97, "lmflow.utils.model.logger", false]], "logger (in module lmflow.utils.versioning)": [[101, "lmflow.utils.versioning.logger", false]], "logging_steps (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.logging_steps", false]], "lora_alpha (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.lora_alpha", false]], "lora_dropout (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.lora_dropout", false]], "lora_model_path (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.lora_model_path", false]], "lora_r (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.lora_r", false]], "lora_target_modules (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.lora_target_modules", false]], "lora_target_modules_mapping (in module lmflow.models.hf_model_mixin)": [[15, "lmflow.models.hf_model_mixin.LORA_TARGET_MODULES_MAPPING", false]], "loss_type (lmflow.args.dpov2alignerarguments attribute)": [[4, "lmflow.args.DPOv2AlignerArguments.loss_type", false]], "low_resource (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.low_resource", false]], "lr_scheduler_type (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.lr_scheduler_type", false]], "main() (in module lmflow.pipeline.utils.memory_safe_dpov2_align)": [[63, "lmflow.pipeline.utils.memory_safe_dpov2_align.main", false]], "main() (in module lmflow.pipeline.utils.memory_safe_vllm_inference)": [[64, "lmflow.pipeline.utils.memory_safe_vllm_inference.main", false]], "make_shell_args_from_dataclass() (in module lmflow.utils.common)": [[73, "lmflow.utils.common.make_shell_args_from_dataclass", false]], "map() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.map", false]], "map() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.map", false]], "margin_scale (lmflow.args.dpov2alignerarguments attribute)": [[4, "lmflow.args.DPOv2AlignerArguments.margin_scale", false]], "mask (lmflow.utils.conversation_template.base.templatecomponent attribute)": [[75, "lmflow.utils.conversation_template.base.TemplateComponent.mask", false]], "mask_prompt (lmflow.args.dpov2alignerarguments attribute)": [[4, "lmflow.args.DPOv2AlignerArguments.mask_prompt", false]], "mask_prompt (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding attribute)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.mask_prompt", false]], "max_eval_samples (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.max_eval_samples", false]], "max_length (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.max_length", false]], "max_length (lmflow.args.dpov2alignerarguments attribute)": [[4, "lmflow.args.DPOv2AlignerArguments.max_length", false]], "max_length (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding attribute)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_length", false]], "max_length (lmflow.pipeline.utils.rm_dataprocessor.rewarddatacollatorwithpadding attribute)": [[67, "lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.max_length", false]], "max_new_tokens (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.max_new_tokens", false]], "max_new_tokens (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.max_new_tokens", false]], "max_prompt_length (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.max_prompt_length", false]], "max_prompt_length (lmflow.args.dpov2alignerarguments attribute)": [[4, "lmflow.args.DPOv2AlignerArguments.max_prompt_length", false]], "max_prompt_length (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding attribute)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_prompt_length", false]], "max_seq_len_cached (lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.condenserotaryembedding attribute)": [[100, "lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.max_seq_len_cached", false]], "max_steps (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.max_steps", false]], "max_target_length (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding attribute)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.max_target_length", false]], "max_train_samples (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.max_train_samples", false]], "memory_safe_dpov2_align_env_var_to_remove (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE", false]], "memory_safe_vllm_inference_env_var_to_remove (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE", false]], "memory_safe_vllm_inference_finish_flag (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG", false]], "memorysafedpov2aligner (class in lmflow.pipeline.dpov2_aligner)": [[51, "lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner", false]], "memorysafevllminferencer (class in lmflow.pipeline.vllm_inferencer)": [[69, "lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer", false]], "merge_lora_weights() (lmflow.models.hf_decoder_model.hfdecodermodel method)": [[13, "lmflow.models.hf_decoder_model.HFDecoderModel.merge_lora_weights", false]], "merge_lora_weights() (lmflow.models.hf_encoder_decoder_model.hfencoderdecodermodel method)": [[14, "lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.merge_lora_weights", false]], "messages (lmflow.utils.llava_conversation_lib.conversation attribute)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.messages", false]], "metric (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.metric", false]], "mixed_precision (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.mixed_precision", false]], "mixed_precision (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.mixed_precision", false]], "model (lmflow.pipeline.inferencer.toolinferencer attribute)": [[55, "lmflow.pipeline.inferencer.ToolInferencer.model", false]], "model (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding attribute)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.model", false]], "model (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.model", false]], "model_args (lmflow.models.hf_model_mixin.hfmodelmixin attribute)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.model_args", false]], "model_args (lmflow.pipeline.dpo_aligner.dpoaligner attribute)": [[50, "lmflow.pipeline.dpo_aligner.DPOAligner.model_args", false]], "model_args (lmflow.pipeline.dpov2_aligner.dpov2aligner attribute)": [[51, "lmflow.pipeline.dpov2_aligner.DPOv2Aligner.model_args", false]], "model_args (lmflow.pipeline.dpov2_aligner.memorysafedpov2aligner attribute)": [[51, "lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.model_args", false]], "model_args (lmflow.pipeline.evaluator.evaluator attribute)": [[52, "lmflow.pipeline.evaluator.Evaluator.model_args", false]], "model_args (lmflow.pipeline.finetuner.finetuner attribute)": [[53, "lmflow.pipeline.finetuner.Finetuner.model_args", false]], "model_args (lmflow.pipeline.inferencer.inferencer attribute)": [[55, "lmflow.pipeline.inferencer.Inferencer.model_args", false]], "model_args (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner attribute)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.model_args", false]], "model_args (lmflow.pipeline.raft_aligner.raftaligner attribute)": [[57, "lmflow.pipeline.raft_aligner.RaftAligner.model_args", false]], "model_args (lmflow.pipeline.rm_inferencer.rewardmodelinferencer attribute)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer.model_args", false]], "model_args (lmflow.pipeline.vllm_inferencer.inferencerwithoffloading attribute)": [[69, "lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.model_args", false]], "model_config_classes (in module lmflow.args)": [[4, "lmflow.args.MODEL_CONFIG_CLASSES", false]], "model_max_length (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.model_max_length", false]], "model_name_or_path (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.model_name_or_path", false]], "model_revision (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.model_revision", false]], "model_type (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.model_type", false]], "model_types (in module lmflow.args)": [[4, "lmflow.args.MODEL_TYPES", false]], "model_wrapped (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.model_wrapped", false]], "modelarguments (class in lmflow.args)": [[4, "lmflow.args.ModelArguments", false]], "module": [[4, "module-lmflow.args", false], [5, "module-lmflow.datasets.dataset", false], [6, "module-lmflow.datasets", false], [7, "module-lmflow.datasets.multi_modal_dataset", false], [8, "module-lmflow", false], [9, "module-lmflow.models.auto_model", false], [10, "module-lmflow.models.base_model", false], [11, "module-lmflow.models.decoder_model", false], [12, "module-lmflow.models.encoder_decoder_model", false], [13, "module-lmflow.models.hf_decoder_model", false], [14, "module-lmflow.models.hf_encoder_decoder_model", false], [15, "module-lmflow.models.hf_model_mixin", false], [16, "module-lmflow.models.hf_text_regression_model", false], [17, "module-lmflow.models", false], [18, "module-lmflow.models.interfaces", false], [19, "module-lmflow.models.interfaces.tunable", false], [20, "module-lmflow.models.regression_model", false], [21, "module-lmflow.models.text_regression_model", false], [22, "module-lmflow.models.vision2seq_model", false], [23, "module-lmflow.models.vision_encoder.clip_encoder", false], [24, "module-lmflow.models.vision_encoder", false], [25, "module-lmflow.optim.adabelief", false], [26, "module-lmflow.optim.adabound", false], [27, "module-lmflow.optim.adadelta", false], [28, "module-lmflow.optim.adagrad", false], [29, "module-lmflow.optim.adam", false], [30, "module-lmflow.optim.adamax", false], [31, "module-lmflow.optim.adamp", false], [32, "module-lmflow.optim.adamw_schedule_free", false], [33, "module-lmflow.optim.adan", false], [34, "module-lmflow.optim.dummy", false], [35, "module-lmflow.optim", false], [36, "module-lmflow.optim.lamb", false], [37, "module-lmflow.optim.lars", false], [38, "module-lmflow.optim.nadam", false], [39, "module-lmflow.optim.novograd", false], [40, "module-lmflow.optim.optimizers", false], [41, "module-lmflow.optim.radam", false], [42, "module-lmflow.optim.sgd_schedule_free", false], [43, "module-lmflow.optim.sgdp", false], [44, "module-lmflow.optim.sophia", false], [45, "module-lmflow.optim.yogi", false], [46, "module-lmflow.pipeline.auto_pipeline", false], [47, "module-lmflow.pipeline.base_aligner", false], [48, "module-lmflow.pipeline.base_pipeline", false], [49, "module-lmflow.pipeline.base_tuner", false], [50, "module-lmflow.pipeline.dpo_aligner", false], [51, "module-lmflow.pipeline.dpov2_aligner", false], [52, "module-lmflow.pipeline.evaluator", false], [53, "module-lmflow.pipeline.finetuner", false], [54, "module-lmflow.pipeline", false], [55, "module-lmflow.pipeline.inferencer", false], [56, "module-lmflow.pipeline.iterative_dpo_aligner", false], [57, "module-lmflow.pipeline.raft_aligner", false], [58, "module-lmflow.pipeline.rm_inferencer", false], [59, "module-lmflow.pipeline.rm_tuner", false], [60, "module-lmflow.pipeline.utils.dpov2_dataprocessor", false], [61, "module-lmflow.pipeline.utils.dpov2_trainer", false], [62, "module-lmflow.pipeline.utils", false], [63, "module-lmflow.pipeline.utils.memory_safe_dpov2_align", false], [64, "module-lmflow.pipeline.utils.memory_safe_vllm_inference", false], [65, "module-lmflow.pipeline.utils.peft_trainer", false], [66, "module-lmflow.pipeline.utils.raft_trainer", false], [67, "module-lmflow.pipeline.utils.rm_dataprocessor", false], [68, "module-lmflow.pipeline.utils.rm_trainer", false], [69, "module-lmflow.pipeline.vllm_inferencer", false], [70, "module-lmflow.tokenization.hf_decoder_model", false], [71, "module-lmflow.tokenization.hf_text_regression_model", false], [72, "module-lmflow.tokenization", false], [73, "module-lmflow.utils.common", false], [74, "module-lmflow.utils.constants", false], [75, "module-lmflow.utils.conversation_template.base", false], [76, "module-lmflow.utils.conversation_template.chatglm", false], [77, "module-lmflow.utils.conversation_template.chatml", false], [78, "module-lmflow.utils.conversation_template.deepseek", false], [79, "module-lmflow.utils.conversation_template.gemma", false], [80, "module-lmflow.utils.conversation_template.hymba", false], [81, "module-lmflow.utils.conversation_template", false], [82, "module-lmflow.utils.conversation_template.internlm", false], [83, "module-lmflow.utils.conversation_template.llama", false], [84, "module-lmflow.utils.conversation_template.phi", false], [85, "module-lmflow.utils.conversation_template.qwen", false], [86, "module-lmflow.utils.conversation_template.yi", false], [87, "module-lmflow.utils.conversation_template.zephyr", false], [88, "module-lmflow.utils.data_utils", false], [89, "module-lmflow.utils.flash_attention.bloom_flash_attention", false], [90, "module-lmflow.utils.flash_attention.gpt2_flash_attention", false], [91, "module-lmflow.utils.flash_attention.gpt_neo_flash_attention", false], [92, "module-lmflow.utils.flash_attention", false], [93, "module-lmflow.utils.flash_attention.llama_flash_attention", false], [94, "module-lmflow.utils.flash_attention.triton_flash_attention", false], [95, "module-lmflow.utils", false], [96, "module-lmflow.utils.llava_conversation_lib", false], [97, "module-lmflow.utils.model", false], [98, "module-lmflow.utils.multimodal", false], [99, "module-lmflow.utils.position_interpolation", false], [100, "module-lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch", false], [101, "module-lmflow.utils.versioning", false], [102, "module-lmflow.version", false]], "mpt (lmflow.utils.llava_conversation_lib.separatorstyle attribute)": [[96, "lmflow.utils.llava_conversation_lib.SeparatorStyle.MPT", false]], "multimodaldatasetarguments (class in lmflow.args)": [[4, "lmflow.args.MultiModalDatasetArguments", false]], "nadam (class in lmflow.optim.nadam)": [[38, "lmflow.optim.nadam.NAdam", false]], "nadam (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.NADAM", false]], "novograd (class in lmflow.optim.novograd)": [[39, "lmflow.optim.novograd.NovoGrad", false]], "novograd (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.NOVOGRAD", false]], "ntk_ratio (lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.condenserotaryembedding attribute)": [[100, "lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.ntk_ratio", false]], "num_examples() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.num_examples", false]], "num_output_sequences (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.num_output_sequences", false]], "num_patches (lmflow.models.vision_encoder.clip_encoder.clipvisiontower property)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.num_patches", false]], "num_raft_iteration (lmflow.args.raftalignerarguments attribute)": [[4, "lmflow.args.RaftAlignerArguments.num_raft_iteration", false]], "observation_formatter (lmflow.utils.conversation_template.base.conversationtemplate attribute)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.observation_formatter", false]], "observation_formatter (lmflow.utils.conversation_template.conversationtemplate attribute)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.observation_formatter", false]], "offset (lmflow.utils.llava_conversation_lib.conversation attribute)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.offset", false]], "on_epoch_end() (lmflow.pipeline.utils.peft_trainer.peftsavingcallback method)": [[65, "lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_epoch_end", false]], "on_save() (lmflow.pipeline.utils.peft_trainer.peftsavingcallback method)": [[65, "lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_save", false]], "on_train_end() (lmflow.pipeline.utils.peft_trainer.peftsavingcallback method)": [[65, "lmflow.pipeline.utils.peft_trainer.PeftSavingCallback.on_train_end", false]], "optim_adam_beta1 (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.optim_adam_beta1", false]], "optim_adam_beta2 (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.optim_adam_beta2", false]], "optim_beta1 (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.optim_beta1", false]], "optim_beta2 (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.optim_beta2", false]], "optim_beta3 (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.optim_beta3", false]], "optim_dummy_beta1 (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.optim_dummy_beta1", false]], "optim_dummy_beta2 (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.optim_dummy_beta2", false]], "optim_momentum (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.optim_momentum", false]], "optim_weight_decay (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.optim_weight_decay", false]], "optimizer_name (in module lmflow.pipeline.utils.raft_trainer)": [[66, "lmflow.pipeline.utils.raft_trainer.OPTIMIZER_NAME", false]], "optimizer_type (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.optimizer_type", false]], "optimizernames (class in lmflow.args)": [[4, "lmflow.args.OptimizerNames", false]], "output (lmflow.utils.data_utils.rewardmodelinferenceresultwithinput attribute)": [[88, "lmflow.utils.data_utils.RewardModelInferenceResultWithInput.output", false]], "output (lmflow.utils.data_utils.vllminferenceresultwithinput attribute)": [[88, "lmflow.utils.data_utils.VLLMInferenceResultWithInput.output", false]], "output_dir (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.output_dir", false]], "output_dir (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.output_dir", false]], "output_dir (lmflow.args.iterativedpoalignerarguments attribute)": [[4, "lmflow.args.IterativeDPOAlignerArguments.output_dir", false]], "output_max_length (lmflow.args.raftalignerarguments attribute)": [[4, "lmflow.args.RaftAlignerArguments.output_max_length", false]], "output_min_length (lmflow.args.raftalignerarguments attribute)": [[4, "lmflow.args.RaftAlignerArguments.output_min_length", false]], "output_reward_path (lmflow.args.raftalignerarguments attribute)": [[4, "lmflow.args.RaftAlignerArguments.output_reward_path", false]], "overwrite_cache (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.overwrite_cache", false]], "pad_to_multiple_of (lmflow.pipeline.utils.rm_dataprocessor.rewarddatacollatorwithpadding attribute)": [[67, "lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.pad_to_multiple_of", false]], "padding (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding attribute)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.padding", false]], "padding (lmflow.pipeline.utils.rm_dataprocessor.rewarddatacollatorwithpadding attribute)": [[67, "lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.padding", false]], "padding_side (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.padding_side", false]], "padding_value (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding attribute)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.padding_value", false]], "paired_conversation_dataset_description (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.PAIRED_CONVERSATION_DATASET_DESCRIPTION", false]], "paired_conversation_tokenize_function() (in module lmflow.tokenization.hf_text_regression_model)": [[71, "lmflow.tokenization.hf_text_regression_model.paired_conversation_tokenize_function", false]], "paired_text_to_text_dataset_description (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION", false]], "parse_to_sampling_params() (lmflow.pipeline.vllm_inferencer.vllminferencer method)": [[69, "lmflow.pipeline.vllm_inferencer.VLLMInferencer.parse_to_sampling_params", false]], "peft_config (lmflow.models.hf_model_mixin.hfmodelmixin attribute)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.peft_config", false]], "peftrewardtrainer (class in lmflow.pipeline.utils.rm_trainer)": [[68, "lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer", false]], "peftsavingcallback (class in lmflow.pipeline.utils.peft_trainer)": [[65, "lmflow.pipeline.utils.peft_trainer.PeftSavingCallback", false]], "pefttrainer (class in lmflow.pipeline.utils.peft_trainer)": [[65, "lmflow.pipeline.utils.peft_trainer.PeftTrainer", false]], "per_device_eval_batch_size (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.per_device_eval_batch_size", false]], "per_device_train_batch_size (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.per_device_train_batch_size", false]], "phi3_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.PHI3_TEMPLATE", false]], "phi3_template (in module lmflow.utils.conversation_template.phi)": [[84, "lmflow.utils.conversation_template.phi.PHI3_TEMPLATE", false]], "pi_ratio (lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.condenserotaryembedding attribute)": [[100, "lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding.pi_ratio", false]], "pipeline_argument_mapping (in module lmflow.args)": [[4, "lmflow.args.PIPELINE_ARGUMENT_MAPPING", false]], "pipeline_mapping (in module lmflow.pipeline.auto_pipeline)": [[46, "lmflow.pipeline.auto_pipeline.PIPELINE_MAPPING", false]], "pipeline_needs_extras (in module lmflow.pipeline.auto_pipeline)": [[46, "lmflow.pipeline.auto_pipeline.PIPELINE_NEEDS_EXTRAS", false]], "place_model_on_device (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.place_model_on_device", false]], "plain (lmflow.utils.llava_conversation_lib.separatorstyle attribute)": [[96, "lmflow.utils.llava_conversation_lib.SeparatorStyle.PLAIN", false]], "pop_callback() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.pop_callback", false]], "post_process_pairs() (lmflow.utils.conversation_template.base.conversationtemplate method)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.post_process_pairs", false]], "post_process_pairs() (lmflow.utils.conversation_template.conversationtemplate method)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.post_process_pairs", false]], "postprocess_distributed_inference_outputs() (lmflow.models.hf_text_regression_model.hftextregressionmodel static method)": [[16, "lmflow.models.hf_text_regression_model.HFTextRegressionModel.postprocess_distributed_inference_outputs", false]], "postprocess_inference_outputs() (lmflow.models.hf_text_regression_model.hftextregressionmodel static method)": [[16, "lmflow.models.hf_text_regression_model.HFTextRegressionModel.postprocess_inference_outputs", false]], "predict() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.predict", false]], "predict_next_token() (lmflow.pipeline.inferencer.speculativeinferencer static method)": [[55, "lmflow.pipeline.inferencer.SpeculativeInferencer.predict_next_token", false]], "prediction_loop() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.prediction_loop", false]], "prediction_step() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.prediction_step", false]], "preferencedatacollatorwithpadding (class in lmflow.pipeline.utils.dpov2_dataprocessor)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding", false]], "prepare_inputs_for_inference() (lmflow.models.hf_decoder_model.hfdecodermodel method)": [[13, "lmflow.models.hf_decoder_model.HFDecoderModel.prepare_inputs_for_inference", false]], "prepare_inputs_for_inference() (lmflow.models.hf_text_regression_model.hftextregressionmodel method)": [[16, "lmflow.models.hf_text_regression_model.HFTextRegressionModel.prepare_inputs_for_inference", false]], "prepare_inputs_labels_for_multimodal() (lmflow.models.vision_encoder.clip_encoder.clipvisiontower method)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.prepare_inputs_labels_for_multimodal", false]], "preprocess_llama_from_llava_plain() (in module lmflow.datasets.multi_modal_dataset)": [[7, "lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_plain", false]], "preprocess_llama_from_llava_v1() (in module lmflow.datasets.multi_modal_dataset)": [[7, "lmflow.datasets.multi_modal_dataset.preprocess_llama_from_llava_v1", false]], "preprocess_logits_for_metrics (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.preprocess_logits_for_metrics", false]], "preprocess_multimodal_llava() (in module lmflow.datasets.multi_modal_dataset)": [[7, "lmflow.datasets.multi_modal_dataset.preprocess_multimodal_llava", false]], "preprocessing_num_workers (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.preprocessing_num_workers", false]], "preset_templates (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.PRESET_TEMPLATES", false]], "pretrained_language_projection_path (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.pretrained_language_projection_path", false]], "print_banner() (in module lmflow.utils.common)": [[73, "lmflow.utils.common.print_banner", false]], "process_image_flag() (in module lmflow.utils.data_utils)": [[88, "lmflow.utils.data_utils.process_image_flag", false]], "processor_image_token_in_minigpt4() (lmflow.models.vision2seq_model.customautovision2seqmodel method)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.processor_image_token_in_minigpt4", false]], "prompt_cache_path (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.prompt_cache_path", false]], "prompt_structure (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.prompt_structure", false]], "push_to_hub() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.push_to_hub", false]], "qformer_from_pretrained() (lmflow.models.vision2seq_model.customautovision2seqmodel method)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.qformer_from_pretrained", false]], "qformer_name_or_path (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.qformer_name_or_path", false]], "quant_config (lmflow.models.hf_model_mixin.hfmodelmixin attribute)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.quant_config", false]], "quant_type (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.quant_type", false]], "qwen2_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.QWEN2_TEMPLATE", false]], "qwen2_template (in module lmflow.utils.conversation_template.qwen)": [[85, "lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE", false]], "qwen2_template_for_tool (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.QWEN2_TEMPLATE_FOR_TOOL", false]], "qwen2_template_for_tool (in module lmflow.utils.conversation_template.qwen)": [[85, "lmflow.utils.conversation_template.qwen.QWEN2_TEMPLATE_FOR_TOOL", false]], "radam (class in lmflow.optim.radam)": [[41, "lmflow.optim.radam.RAdam", false]], "radam (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.RADAM", false]], "raft_batch_size (lmflow.args.raftalignerarguments attribute)": [[4, "lmflow.args.RaftAlignerArguments.raft_batch_size", false]], "raftaligner (class in lmflow.pipeline.raft_aligner)": [[57, "lmflow.pipeline.raft_aligner.RaftAligner", false]], "raftalignerarguments (class in lmflow.args)": [[4, "lmflow.args.RaftAlignerArguments", false]], "rafttrainer (class in lmflow.pipeline.utils.raft_trainer)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer", false]], "random_seed (lmflow.args.dpov2alignerarguments attribute)": [[4, "lmflow.args.DPOv2AlignerArguments.random_seed", false]], "random_seed (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.random_seed", false]], "random_seed (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.random_seed", false]], "random_shuffle (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.random_shuffle", false]], "rectify (lmflow.optim.adabelief.adabelief attribute)": [[25, "lmflow.optim.adabelief.AdaBelief.rectify", false]], "ref_model_args (lmflow.pipeline.dpov2_aligner.dpov2aligner attribute)": [[51, "lmflow.pipeline.dpov2_aligner.DPOv2Aligner.ref_model_args", false]], "ref_model_args (lmflow.pipeline.dpov2_aligner.memorysafedpov2aligner attribute)": [[51, "lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner.ref_model_args", false]], "ref_model_args (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner attribute)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.ref_model_args", false]], "referencemodelarguments (in module lmflow.pipeline.dpov2_aligner)": [[51, "lmflow.pipeline.dpov2_aligner.ReferenceModelArguments", false]], "referencemodelarguments (in module lmflow.pipeline.utils.memory_safe_dpov2_align)": [[63, "lmflow.pipeline.utils.memory_safe_dpov2_align.ReferenceModelArguments", false]], "register_inference_function() (lmflow.models.text_regression_model.textregressionmodel method)": [[21, "lmflow.models.text_regression_model.TextRegressionModel.register_inference_function", false]], "register_prompt_cache() (lmflow.models.vision2seq_model.customautovision2seqmodel method)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.register_prompt_cache", false]], "register_tokenizer() (lmflow.datasets.multi_modal_dataset.custommultimodaldataset method)": [[7, "lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset.register_tokenizer", false]], "regressionmodel (class in lmflow.models.regression_model)": [[20, "lmflow.models.regression_model.RegressionModel", false]], "remove_callback() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.remove_callback", false]], "remove_dataclass_attr_prefix() (in module lmflow.utils.common)": [[73, "lmflow.utils.common.remove_dataclass_attr_prefix", false]], "remove_last_sep (lmflow.utils.conversation_template.base.conversationtemplate attribute)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.remove_last_sep", false]], "remove_last_sep (lmflow.utils.conversation_template.conversationtemplate attribute)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.remove_last_sep", false]], "remove_last_separator() (lmflow.utils.conversation_template.base.conversationtemplate method)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.remove_last_separator", false]], "remove_last_separator() (lmflow.utils.conversation_template.conversationtemplate method)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.remove_last_separator", false]], "remove_unused_columns (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.remove_unused_columns", false]], "repetition_penalty (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.repetition_penalty", false]], "repetition_penalty (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.repetition_penalty", false]], "replace_bloom_attn_with_flash_attn() (in module lmflow.utils.flash_attention.bloom_flash_attention)": [[89, "lmflow.utils.flash_attention.bloom_flash_attention.replace_bloom_attn_with_flash_attn", false]], "replace_gpt2_attn_with_flash_attn() (in module lmflow.utils.flash_attention.gpt2_flash_attention)": [[90, "lmflow.utils.flash_attention.gpt2_flash_attention.replace_gpt2_attn_with_flash_attn", false]], "replace_gpt_neo_attn_with_flash_attn() (in module lmflow.utils.flash_attention.gpt_neo_flash_attention)": [[91, "lmflow.utils.flash_attention.gpt_neo_flash_attention.replace_gpt_neo_attn_with_flash_attn", false]], "replace_llama_attn_with_flash_attn() (in module lmflow.utils.flash_attention.llama_flash_attention)": [[93, "lmflow.utils.flash_attention.llama_flash_attention.replace_llama_attn_with_flash_attn", false]], "replace_llama_with_condense() (in module lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch)": [[100, "lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.replace_llama_with_condense", false]], "report_to (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.report_to", false]], "reset() (lmflow.optim.adabelief.adabelief method)": [[25, "lmflow.optim.adabelief.AdaBelief.reset", false]], "restart_opt() (lmflow.optim.adan.adan method)": [[33, "lmflow.optim.adan.Adan.restart_opt", false]], "results_path (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.results_path", false]], "return_code_error_buffer (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.RETURN_CODE_ERROR_BUFFER", false]], "return_tensors (lmflow.pipeline.utils.rm_dataprocessor.rewarddatacollatorwithpadding attribute)": [[67, "lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.return_tensors", false]], "reward_model_args (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner attribute)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.reward_model_args", false]], "reward_model_inference_batch_size (lmflow.args.iterativedpoalignerarguments attribute)": [[4, "lmflow.args.IterativeDPOAlignerArguments.reward_model_inference_batch_size", false]], "reward_model_inference_block_size (lmflow.args.iterativedpoalignerarguments attribute)": [[4, "lmflow.args.IterativeDPOAlignerArguments.reward_model_inference_block_size", false]], "rewarddatacollatorwithpadding (class in lmflow.pipeline.utils.rm_dataprocessor)": [[67, "lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding", false]], "rewardmodelinferencer (class in lmflow.pipeline.rm_inferencer)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer", false]], "rewardmodelinferenceresultwithinput (class in lmflow.utils.data_utils)": [[88, "lmflow.utils.data_utils.RewardModelInferenceResultWithInput", false]], "rewardmodeltuner (class in lmflow.pipeline.rm_tuner)": [[59, "lmflow.pipeline.rm_tuner.RewardModelTuner", false]], "rewardmodeltunerarguments (class in lmflow.args)": [[4, "lmflow.args.RewardModelTunerArguments", false]], "rewardtrainer (class in lmflow.pipeline.utils.rm_trainer)": [[68, "lmflow.pipeline.utils.rm_trainer.RewardTrainer", false]], "rm_loss() (in module lmflow.pipeline.utils.rm_trainer)": [[68, "lmflow.pipeline.utils.rm_trainer.rm_loss", false]], "roles (lmflow.utils.llava_conversation_lib.conversation attribute)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.roles", false]], "rope_ntk_ratio (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.rope_ntk_ratio", false]], "rope_pi_ratio (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.rope_pi_ratio", false]], "rstrip_partial_utf8() (in module lmflow.pipeline.inferencer)": [[55, "lmflow.pipeline.inferencer.rstrip_partial_utf8", false]], "run_name (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.run_name", false]], "sample() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.sample", false]], "sample() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.sample", false]], "sample() (lmflow.pipeline.inferencer.speculativeinferencer static method)": [[55, "lmflow.pipeline.inferencer.SpeculativeInferencer.sample", false]], "sampling_paired_idx_from_rewards() (lmflow.pipeline.dpov2_aligner.dpov2aligner method)": [[51, "lmflow.pipeline.dpov2_aligner.DPOv2Aligner.sampling_paired_idx_from_rewards", false]], "sampling_paired_method (lmflow.args.dpov2alignerarguments attribute)": [[4, "lmflow.args.DPOv2AlignerArguments.sampling_paired_method", false]], "sampling_params (lmflow.pipeline.vllm_inferencer.vllminferencer attribute)": [[69, "lmflow.pipeline.vllm_inferencer.VLLMInferencer.sampling_params", false]], "sanity_check (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.sanity_check", false]], "sanity_check() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.sanity_check", false]], "sanity_check() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.sanity_check", false]], "save() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.save", false]], "save() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.save", false]], "save() (lmflow.models.hf_decoder_model.hfdecodermodel method)": [[13, "lmflow.models.hf_decoder_model.HFDecoderModel.save", false]], "save() (lmflow.models.hf_encoder_decoder_model.hfencoderdecodermodel method)": [[14, "lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.save", false]], "save() (lmflow.models.hf_text_regression_model.hftextregressionmodel method)": [[16, "lmflow.models.hf_text_regression_model.HFTextRegressionModel.save", false]], "save_aggregated_lora (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.save_aggregated_lora", false]], "save_counter (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.save_counter", false]], "save_inference_results() (lmflow.pipeline.vllm_inferencer.inferencerwithoffloading method)": [[69, "lmflow.pipeline.vllm_inferencer.InferencerWithOffloading.save_inference_results", false]], "save_inference_results() (lmflow.pipeline.vllm_inferencer.vllminferencer method)": [[69, "lmflow.pipeline.vllm_inferencer.VLLMInferencer.save_inference_results", false]], "save_language_projection (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.save_language_projection", false]], "save_model() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.save_model", false]], "save_pretrain_model_path (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.save_pretrain_model_path", false]], "save_prompt_cache() (lmflow.models.vision2seq_model.customautovision2seqmodel method)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.save_prompt_cache", false]], "save_results (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.save_results", false]], "save_steps (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.save_steps", false]], "scaler_name (in module lmflow.pipeline.utils.raft_trainer)": [[66, "lmflow.pipeline.utils.raft_trainer.SCALER_NAME", false]], "scheduler_name (in module lmflow.pipeline.utils.raft_trainer)": [[66, "lmflow.pipeline.utils.raft_trainer.SCHEDULER_NAME", false]], "score_to_prob() (lmflow.pipeline.inferencer.speculativeinferencer static method)": [[55, "lmflow.pipeline.inferencer.SpeculativeInferencer.score_to_prob", false]], "seed (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.seed", false]], "select_feature (lmflow.models.vision_encoder.clip_encoder.clipvisiontower attribute)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.select_feature", false]], "select_layer (lmflow.models.vision_encoder.clip_encoder.clipvisiontower attribute)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.select_layer", false]], "sep (lmflow.utils.llava_conversation_lib.conversation attribute)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.sep", false]], "sep2 (lmflow.utils.llava_conversation_lib.conversation attribute)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.sep2", false]], "sep_style (lmflow.args.multimodaldatasetarguments attribute)": [[4, "lmflow.args.MultiModalDatasetArguments.sep_style", false]], "sep_style (lmflow.utils.llava_conversation_lib.conversation attribute)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.sep_style", false]], "separator (lmflow.utils.conversation_template.base.conversationtemplate attribute)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.separator", false]], "separator (lmflow.utils.conversation_template.conversationtemplate attribute)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.separator", false]], "separatorstyle (class in lmflow.utils.llava_conversation_lib)": [[96, "lmflow.utils.llava_conversation_lib.SeparatorStyle", false]], "set_random_seed() (in module lmflow.utils.data_utils)": [[88, "lmflow.utils.data_utils.set_random_seed", false]], "sgd_schedule_free (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.SGD_SCHEDULE_FREE", false]], "sgdp (class in lmflow.optim.sgdp)": [[43, "lmflow.optim.sgdp.SGDP", false]], "sgdp (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.SGDP", false]], "sgdschedulefree (class in lmflow.optim.sgd_schedule_free)": [[42, "lmflow.optim.sgd_schedule_free.SGDScheduleFree", false]], "sharded_ddp (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.sharded_ddp", false]], "single (lmflow.utils.llava_conversation_lib.separatorstyle attribute)": [[96, "lmflow.utils.llava_conversation_lib.SeparatorStyle.SINGLE", false]], "skip_first_batches (in module lmflow.pipeline.utils.raft_trainer)": [[66, "lmflow.pipeline.utils.raft_trainer.skip_first_batches", false]], "skip_next (lmflow.utils.llava_conversation_lib.conversation attribute)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.skip_next", false]], "sophia (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.SOPHIA", false]], "sophiag (class in lmflow.optim.sophia)": [[44, "lmflow.optim.sophia.SophiaG", false]], "special_starter (lmflow.utils.conversation_template.base.conversationtemplate attribute)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.special_starter", false]], "special_starter (lmflow.utils.conversation_template.conversationtemplate attribute)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.special_starter", false]], "special_stopper (lmflow.utils.conversation_template.base.conversationtemplate attribute)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.special_stopper", false]], "special_stopper (lmflow.utils.conversation_template.conversationtemplate attribute)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.special_stopper", false]], "speculativeinferencer (class in lmflow.pipeline.inferencer)": [[55, "lmflow.pipeline.inferencer.SpeculativeInferencer", false]], "split_args() (in module lmflow.args)": [[4, "lmflow.args.split_args", false]], "state (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.state", false]], "step() (lmflow.optim.adabelief.adabelief method)": [[25, "lmflow.optim.adabelief.AdaBelief.step", false]], "step() (lmflow.optim.adabound.adabound method)": [[26, "lmflow.optim.adabound.AdaBound.step", false]], "step() (lmflow.optim.adadelta.adadelta method)": [[27, "lmflow.optim.adadelta.Adadelta.step", false]], "step() (lmflow.optim.adagrad.adagrad method)": [[28, "lmflow.optim.adagrad.AdaGrad.step", false]], "step() (lmflow.optim.adam.adam method)": [[29, "lmflow.optim.adam.Adam.step", false]], "step() (lmflow.optim.adamax.adamax method)": [[30, "lmflow.optim.adamax.Adamax.step", false]], "step() (lmflow.optim.adamp.adamp method)": [[31, "lmflow.optim.adamp.AdamP.step", false]], "step() (lmflow.optim.adamw_schedule_free.adamwschedulefree method)": [[32, "lmflow.optim.adamw_schedule_free.AdamWScheduleFree.step", false]], "step() (lmflow.optim.adan.adan method)": [[33, "lmflow.optim.adan.Adan.step", false]], "step() (lmflow.optim.dummy.dummy method)": [[34, "lmflow.optim.dummy.Dummy.step", false]], "step() (lmflow.optim.lamb.lamb method)": [[36, "lmflow.optim.lamb.Lamb.step", false]], "step() (lmflow.optim.lars.lars method)": [[37, "lmflow.optim.lars.LARS.step", false]], "step() (lmflow.optim.nadam.nadam method)": [[38, "lmflow.optim.nadam.NAdam.step", false]], "step() (lmflow.optim.novograd.novograd method)": [[39, "lmflow.optim.novograd.NovoGrad.step", false]], "step() (lmflow.optim.radam.radam method)": [[41, "lmflow.optim.radam.RAdam.step", false]], "step() (lmflow.optim.sgd_schedule_free.sgdschedulefree method)": [[42, "lmflow.optim.sgd_schedule_free.SGDScheduleFree.step", false]], "step() (lmflow.optim.sgdp.sgdp method)": [[43, "lmflow.optim.sgdp.SGDP.step", false]], "step() (lmflow.optim.sophia.sophiag method)": [[44, "lmflow.optim.sophia.SophiaG.step", false]], "step() (lmflow.optim.yogi.yogi method)": [[45, "lmflow.optim.yogi.Yogi.step", false]], "store_flos() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.store_flos", false]], "stream_inference() (lmflow.pipeline.inferencer.inferencer method)": [[55, "lmflow.pipeline.inferencer.Inferencer.stream_inference", false]], "stream_inference() (lmflow.pipeline.inferencer.speculativeinferencer method)": [[55, "lmflow.pipeline.inferencer.SpeculativeInferencer.stream_inference", false]], "streaming (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.streaming", false]], "stringformatter (class in lmflow.utils.conversation_template.base)": [[75, "lmflow.utils.conversation_template.base.StringFormatter", false]], "supported_dataset_type (in module lmflow.pipeline.inferencer)": [[55, "lmflow.pipeline.inferencer.supported_dataset_type", false]], "system (lmflow.utils.llava_conversation_lib.conversation attribute)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.system", false]], "system_formatter (lmflow.utils.conversation_template.base.conversationtemplate attribute)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.system_formatter", false]], "system_formatter (lmflow.utils.conversation_template.conversationtemplate attribute)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.system_formatter", false]], "temperature (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.temperature", false]], "temperature (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.temperature", false]], "template (lmflow.utils.conversation_template.base.formatter attribute)": [[75, "lmflow.utils.conversation_template.base.Formatter.template", false]], "template_name (lmflow.utils.conversation_template.base.conversationtemplate attribute)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.template_name", false]], "template_name (lmflow.utils.conversation_template.conversationtemplate attribute)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.template_name", false]], "templatecomponent (class in lmflow.utils.conversation_template.base)": [[75, "lmflow.utils.conversation_template.base.TemplateComponent", false]], "tensor_parallel_size (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.tensor_parallel_size", false]], "test_file (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.test_file", false]], "text2text_dataset_description (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.TEXT2TEXT_DATASET_DESCRIPTION", false]], "text2text_dataset_details (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.TEXT2TEXT_DATASET_DETAILS", false]], "text2text_dataset_long_descrition (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.TEXT2TEXT_DATASET_LONG_DESCRITION", false]], "text_only_dataset_description (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.TEXT_ONLY_DATASET_DESCRIPTION", false]], "text_only_dataset_details (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.TEXT_ONLY_DATASET_DETAILS", false]], "text_only_dataset_long_descrition (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.TEXT_ONLY_DATASET_LONG_DESCRITION", false]], "text_to_scored_textlist_dataset_description (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION", false]], "text_to_textlist_dataset_description (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.TEXT_TO_TEXTLIST_DATASET_DESCRIPTION", false]], "text_to_textlist_tokenize_function() (in module lmflow.tokenization.hf_text_regression_model)": [[71, "lmflow.tokenization.hf_text_regression_model.text_to_textlist_tokenize_function", false]], "textregressionmodel (class in lmflow.models.text_regression_model)": [[21, "lmflow.models.text_regression_model.TextRegressionModel", false]], "to_dict() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.to_dict", false]], "to_dict() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.to_dict", false]], "to_gradio_chatbot() (lmflow.utils.llava_conversation_lib.conversation method)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.to_gradio_chatbot", false]], "to_list() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.to_list", false]], "to_list() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.to_list", false]], "tok_logger (in module lmflow.tokenization.hf_decoder_model)": [[70, "lmflow.tokenization.hf_decoder_model.tok_logger", false]], "tok_logger (in module lmflow.tokenization.hf_text_regression_model)": [[71, "lmflow.tokenization.hf_text_regression_model.tok_logger", false]], "token (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.token", false]], "tokenize() (lmflow.models.hf_decoder_model.hfdecodermodel method)": [[13, "lmflow.models.hf_decoder_model.HFDecoderModel.tokenize", false]], "tokenize() (lmflow.models.hf_encoder_decoder_model.hfencoderdecodermodel method)": [[14, "lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel.tokenize", false]], "tokenize() (lmflow.models.hf_text_regression_model.hftextregressionmodel method)": [[16, "lmflow.models.hf_text_regression_model.HFTextRegressionModel.tokenize", false]], "tokenize_batch_element() (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding method)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.tokenize_batch_element", false]], "tokenize_function() (in module lmflow.tokenization.hf_decoder_model)": [[70, "lmflow.tokenization.hf_decoder_model.tokenize_function", false]], "tokenize_function() (in module lmflow.tokenization.hf_text_regression_model)": [[71, "lmflow.tokenization.hf_text_regression_model.tokenize_function", false]], "tokenizer (lmflow.datasets.multi_modal_dataset.datacollatorforsuperviseddataset attribute)": [[7, "lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset.tokenizer", false]], "tokenizer (lmflow.models.hf_model_mixin.hfmodelmixin attribute)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.tokenizer", false]], "tokenizer (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding attribute)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.tokenizer", false]], "tokenizer (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.tokenizer", false]], "tokenizer (lmflow.pipeline.utils.rm_dataprocessor.rewarddatacollatorwithpadding attribute)": [[67, "lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding.tokenizer", false]], "tokenizer_image_token() (in module lmflow.datasets.multi_modal_dataset)": [[7, "lmflow.datasets.multi_modal_dataset.tokenizer_image_token", false]], "tokenizer_name (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.tokenizer_name", false]], "toolinferencer (class in lmflow.pipeline.inferencer)": [[55, "lmflow.pipeline.inferencer.ToolInferencer", false]], "tools_formatter (lmflow.utils.conversation_template.base.conversationtemplate attribute)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.tools_formatter", false]], "tools_formatter (lmflow.utils.conversation_template.conversationtemplate attribute)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.tools_formatter", false]], "top_k (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.top_k", false]], "top_p (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.top_p", false]], "top_reward_percentage (lmflow.args.raftalignerarguments attribute)": [[4, "lmflow.args.RaftAlignerArguments.top_reward_percentage", false]], "torch_dtype (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.torch_dtype", false]], "torch_dtype (lmflow.models.hf_model_mixin.hfmodelmixin attribute)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.torch_dtype", false]], "torch_jit_model_eval() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.torch_jit_model_eval", false]], "train() (lmflow.optim.adamw_schedule_free.adamwschedulefree method)": [[32, "lmflow.optim.adamw_schedule_free.AdamWScheduleFree.train", false]], "train() (lmflow.optim.sgd_schedule_free.sgdschedulefree method)": [[42, "lmflow.optim.sgd_schedule_free.SGDScheduleFree.train", false]], "train() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.train", false]], "train_dataset (lmflow.pipeline.dpo_aligner.dpoaligner attribute)": [[50, "lmflow.pipeline.dpo_aligner.DPOAligner.train_dataset", false]], "train_dataset (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.train_dataset", false]], "train_file (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.train_file", false]], "train_on_prompt (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.train_on_prompt", false]], "train_test_split() (lmflow.datasets.dataset method)": [[6, "lmflow.datasets.Dataset.train_test_split", false]], "train_test_split() (lmflow.datasets.dataset.dataset method)": [[5, "lmflow.datasets.dataset.Dataset.train_test_split", false]], "trainer_state_name (in module lmflow.pipeline.utils.raft_trainer)": [[66, "lmflow.pipeline.utils.raft_trainer.TRAINER_STATE_NAME", false]], "training_args_name (in module lmflow.pipeline.utils.raft_trainer)": [[66, "lmflow.pipeline.utils.raft_trainer.TRAINING_ARGS_NAME", false]], "training_step() (lmflow.pipeline.utils.raft_trainer.rafttrainer method)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.training_step", false]], "truncate_to_model_max_length (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.truncate_to_model_max_length", false]], "truncation_mode (lmflow.pipeline.utils.dpov2_dataprocessor.preferencedatacollatorwithpadding attribute)": [[60, "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding.truncation_mode", false]], "truncation_side (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.truncation_side", false]], "trust_remote_code (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.trust_remote_code", false]], "tunable (class in lmflow.models.interfaces.tunable)": [[19, "lmflow.models.interfaces.tunable.Tunable", false]], "tune() (lmflow.pipeline.base_tuner.basetuner method)": [[49, "lmflow.pipeline.base_tuner.BaseTuner.tune", false]], "tune() (lmflow.pipeline.finetuner.finetuner method)": [[53, "lmflow.pipeline.finetuner.Finetuner.tune", false]], "tune() (lmflow.pipeline.rm_tuner.rewardmodeltuner method)": [[59, "lmflow.pipeline.rm_tuner.RewardModelTuner.tune", false]], "two (lmflow.utils.llava_conversation_lib.separatorstyle attribute)": [[96, "lmflow.utils.llava_conversation_lib.SeparatorStyle.TWO", false]], "type (lmflow.datasets.dataset attribute)": [[6, "lmflow.datasets.Dataset.type", false]], "type (lmflow.datasets.dataset.dataset attribute)": [[5, "lmflow.datasets.dataset.Dataset.type", false]], "type (lmflow.utils.conversation_template.base.templatecomponent attribute)": [[75, "lmflow.utils.conversation_template.base.TemplateComponent.type", false]], "update_custom_config() (in module lmflow.utils.multimodal)": [[98, "lmflow.utils.multimodal.update_custom_config", false]], "update_hessian() (lmflow.optim.sophia.sophiag method)": [[44, "lmflow.optim.sophia.SophiaG.update_hessian", false]], "use_accelerator (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.use_accelerator", false]], "use_accelerator (lmflow.models.hf_model_mixin.hfmodelmixin attribute)": [[15, "lmflow.models.hf_model_mixin.HFModelMixin.use_accelerator", false]], "use_accelerator_for_evaluator (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.use_accelerator_for_evaluator", false]], "use_apex (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_apex", false]], "use_beam_search (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.use_beam_search", false]], "use_cpu_amp (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_cpu_amp", false]], "use_cuda_amp (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_cuda_amp", false]], "use_customized_optim (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.use_customized_optim", false]], "use_dora (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.use_dora", false]], "use_dpo_data_collator (lmflow.pipeline.utils.dpov2_trainer.dpov2trainer attribute)": [[61, "lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer.use_dpo_data_collator", false]], "use_fast_tokenizer (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.use_fast_tokenizer", false]], "use_flash_attention (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.use_flash_attention", false]], "use_image_start_end (lmflow.args.multimodaldatasetarguments attribute)": [[4, "lmflow.args.MultiModalDatasetArguments.use_image_start_end", false]], "use_int8 (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.use_int8", false]], "use_lisa (lmflow.args.finetunerarguments attribute)": [[4, "lmflow.args.FinetunerArguments.use_lisa", false]], "use_lora (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.use_lora", false]], "use_prompt_cache (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.use_prompt_cache", false]], "use_qlora (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.use_qlora", false]], "use_ram_optimized_load (lmflow.args.modelarguments attribute)": [[4, "lmflow.args.ModelArguments.use_ram_optimized_load", false]], "use_tune_checkpoints (lmflow.pipeline.utils.raft_trainer.rafttrainer attribute)": [[66, "lmflow.pipeline.utils.raft_trainer.RaftTrainer.use_tune_checkpoints", false]], "use_vllm (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.use_vllm", false]], "use_wandb (lmflow.args.evaluatorarguments attribute)": [[4, "lmflow.args.EvaluatorArguments.use_wandb", false]], "user_formatter (lmflow.utils.conversation_template.base.conversationtemplate attribute)": [[75, "lmflow.utils.conversation_template.base.ConversationTemplate.user_formatter", false]], "user_formatter (lmflow.utils.conversation_template.conversationtemplate attribute)": [[81, "lmflow.utils.conversation_template.ConversationTemplate.user_formatter", false]], "validation_file (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.validation_file", false]], "validation_split_percentage (lmflow.args.datasetarguments attribute)": [[4, "lmflow.args.DatasetArguments.validation_split_percentage", false]], "version (lmflow.utils.llava_conversation_lib.conversation attribute)": [[96, "lmflow.utils.llava_conversation_lib.Conversation.version", false]], "vision_feature_select() (lmflow.models.vision2seq_model.customautovision2seqmodel method)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.vision_feature_select", false]], "vision_model_from_pretrained() (lmflow.models.vision2seq_model.customautovision2seqmodel method)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.vision_model_from_pretrained", false]], "vision_select_layer (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.vision_select_layer", false]], "vision_tower_name (lmflow.models.vision_encoder.clip_encoder.clipvisiontower attribute)": [[23, "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower.vision_tower_name", false]], "vismodelarguments (class in lmflow.args)": [[4, "lmflow.args.VisModelArguments", false]], "vllm_gpu_memory_utilization (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.vllm_gpu_memory_utilization", false]], "vllm_inference_batch_size (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.vllm_inference_batch_size", false]], "vllm_tensor_parallel_size (lmflow.args.inferencerarguments attribute)": [[4, "lmflow.args.InferencerArguments.vllm_tensor_parallel_size", false]], "vllminferencer (class in lmflow.pipeline.vllm_inferencer)": [[69, "lmflow.pipeline.vllm_inferencer.VLLMInferencer", false]], "vllminferenceresultwithinput (class in lmflow.utils.data_utils)": [[88, "lmflow.utils.data_utils.VLLMInferenceResultWithInput", false]], "warmup_steps (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.warmup_steps", false]], "weight_decay (lmflow.args.dpoalignerarguments attribute)": [[4, "lmflow.args.DPOAlignerArguments.weight_decay", false]], "weight_decouple (lmflow.optim.adabelief.adabelief attribute)": [[25, "lmflow.optim.adabelief.AdaBelief.weight_decouple", false]], "with_qformer (lmflow.args.vismodelarguments attribute)": [[4, "lmflow.args.VisModelArguments.with_qformer", false]], "with_qformer (lmflow.models.vision2seq_model.customautovision2seqmodel attribute)": [[22, "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel.with_qformer", false]], "worker_heart_beat_interval (in module lmflow.utils.constants)": [[74, "lmflow.utils.constants.WORKER_HEART_BEAT_INTERVAL", false]], "workspace_path (lmflow.pipeline.iterative_dpo_aligner.iterativedpoaligner attribute)": [[56, "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner.workspace_path", false]], "world_size (lmflow.pipeline.evaluator.evaluator attribute)": [[52, "lmflow.pipeline.evaluator.Evaluator.world_size", false]], "world_size (lmflow.pipeline.inferencer.inferencer attribute)": [[55, "lmflow.pipeline.inferencer.Inferencer.world_size", false]], "world_size (lmflow.pipeline.rm_inferencer.rewardmodelinferencer attribute)": [[58, "lmflow.pipeline.rm_inferencer.RewardModelInferencer.world_size", false]], "yi1_5_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.YI1_5_TEMPLATE", false]], "yi1_5_template (in module lmflow.utils.conversation_template.yi)": [[86, "lmflow.utils.conversation_template.yi.YI1_5_TEMPLATE", false]], "yogi (class in lmflow.optim.yogi)": [[45, "lmflow.optim.yogi.Yogi", false]], "yogi (lmflow.args.optimizernames attribute)": [[4, "lmflow.args.OptimizerNames.YOGI", false]], "zephyr_template (in module lmflow.utils.conversation_template)": [[81, "lmflow.utils.conversation_template.ZEPHYR_TEMPLATE", false]], "zephyr_template (in module lmflow.utils.conversation_template.zephyr)": [[87, "lmflow.utils.conversation_template.zephyr.ZEPHYR_TEMPLATE", false]], "zephyrconversationtemplate (class in lmflow.utils.conversation_template.zephyr)": [[87, "lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate", false]]}, "objects": {"": [[8, 0, 0, "-", "lmflow"]], "lmflow": [[8, 1, 1, "", "__version__"], [4, 0, 0, "-", "args"], [6, 0, 0, "-", "datasets"], [8, 1, 1, "", "internal_version"], [17, 0, 0, "-", "models"], [35, 0, 0, "-", "optim"], [54, 0, 0, "-", "pipeline"], [72, 0, 0, "-", "tokenization"], [95, 0, 0, "-", "utils"], [102, 0, 0, "-", "version"]], "lmflow.args": [[4, 2, 1, "", "AutoArguments"], [4, 2, 1, "", "BenchmarkingArguments"], [4, 2, 1, "", "DPOAlignerArguments"], [4, 2, 1, "", "DPOv2AlignerArguments"], [4, 2, 1, "", "DatasetArguments"], [4, 2, 1, "", "EvaluatorArguments"], [4, 2, 1, "", "FinetunerArguments"], [4, 2, 1, "", "InferencerArguments"], [4, 2, 1, "", "IterativeAlignerArguments"], [4, 2, 1, "", "IterativeDPOAlignerArguments"], [4, 1, 1, "", "MODEL_CONFIG_CLASSES"], [4, 1, 1, "", "MODEL_TYPES"], [4, 2, 1, "", "ModelArguments"], [4, 2, 1, "", "MultiModalDatasetArguments"], [4, 2, 1, "", "OptimizerNames"], [4, 1, 1, "", "PIPELINE_ARGUMENT_MAPPING"], [4, 2, 1, "", "RaftAlignerArguments"], [4, 2, 1, "", "RewardModelTunerArguments"], [4, 2, 1, "", "VisModelArguments"], [4, 1, 1, "", "logger"], [4, 5, 1, "", "split_args"]], "lmflow.args.AutoArguments": [[4, 3, 1, "", "get_pipeline_args_class"]], "lmflow.args.BenchmarkingArguments": [[4, 4, 1, "", "dataset_name"], [4, 4, 1, "", "lm_evaluation_metric"]], "lmflow.args.DPOAlignerArguments": [[4, 4, 1, "", "beta"], [4, 4, 1, "", "eval_dataset_path"], [4, 4, 1, "", "eval_steps"], [4, 4, 1, "", "gradient_accumulation_steps"], [4, 4, 1, "", "gradient_checkpointing"], [4, 4, 1, "", "gradient_checkpointing_use_reentrant"], [4, 4, 1, "", "learning_rate"], [4, 4, 1, "", "local_rank"], [4, 4, 1, "", "log_freq"], [4, 4, 1, "", "logging_steps"], [4, 4, 1, "", "lr_scheduler_type"], [4, 4, 1, "", "max_length"], [4, 4, 1, "", "max_prompt_length"], [4, 4, 1, "", "max_steps"], [4, 4, 1, "", "optimizer_type"], [4, 4, 1, "", "output_dir"], [4, 4, 1, "", "per_device_eval_batch_size"], [4, 4, 1, "", "per_device_train_batch_size"], [4, 4, 1, "", "report_to"], [4, 4, 1, "", "run_name"], [4, 4, 1, "", "sanity_check"], [4, 4, 1, "", "save_steps"], [4, 4, 1, "", "seed"], [4, 4, 1, "", "warmup_steps"], [4, 4, 1, "", "weight_decay"]], "lmflow.args.DPOv2AlignerArguments": [[4, 4, 1, "", "accelerate_config_file"], [4, 4, 1, "", "beta"], [4, 4, 1, "", "length_penalty"], [4, 4, 1, "", "loss_type"], [4, 4, 1, "", "margin_scale"], [4, 4, 1, "", "mask_prompt"], [4, 4, 1, "", "max_length"], [4, 4, 1, "", "max_prompt_length"], [4, 4, 1, "", "random_seed"], [4, 4, 1, "", "sampling_paired_method"]], "lmflow.args.DatasetArguments": [[4, 3, 1, "", "__post_init__"], [4, 4, 1, "", "block_size"], [4, 4, 1, "", "conversation_template"], [4, 4, 1, "", "customized_cache_dir"], [4, 4, 1, "", "dataset_config_name"], [4, 4, 1, "", "dataset_name"], [4, 4, 1, "", "dataset_path"], [4, 4, 1, "", "disable_group_texts"], [4, 4, 1, "", "group_texts_batch_size"], [4, 4, 1, "", "is_custom_dataset"], [4, 4, 1, "", "keep_linebreaks"], [4, 4, 1, "", "max_eval_samples"], [4, 4, 1, "", "max_train_samples"], [4, 4, 1, "", "overwrite_cache"], [4, 4, 1, "", "preprocessing_num_workers"], [4, 4, 1, "", "streaming"], [4, 4, 1, "", "test_file"], [4, 4, 1, "", "train_file"], [4, 4, 1, "", "train_on_prompt"], [4, 4, 1, "", "validation_file"], [4, 4, 1, "", "validation_split_percentage"]], "lmflow.args.EvaluatorArguments": [[4, 4, 1, "", "answer_type"], [4, 4, 1, "", "deepspeed"], [4, 4, 1, "", "evaluate_block_size"], [4, 4, 1, "", "inference_batch_size_per_device"], [4, 4, 1, "", "local_rank"], [4, 4, 1, "", "max_new_tokens"], [4, 4, 1, "", "metric"], [4, 4, 1, "", "mixed_precision"], [4, 4, 1, "", "output_dir"], [4, 4, 1, "", "prompt_structure"], [4, 4, 1, "", "random_seed"], [4, 4, 1, "", "random_shuffle"], [4, 4, 1, "", "repetition_penalty"], [4, 4, 1, "", "temperature"], [4, 4, 1, "", "use_accelerator_for_evaluator"], [4, 4, 1, "", "use_wandb"]], "lmflow.args.FinetunerArguments": [[4, 4, 1, "", "customized_optim"], [4, 4, 1, "", "customized_optim_args"], [4, 4, 1, "", "eval_dataset_path"], [4, 4, 1, "", "finetune_part"], [4, 4, 1, "", "lisa_activated_layers"], [4, 4, 1, "", "lisa_interval_steps"], [4, 4, 1, "", "lisa_layers_attribute"], [4, 4, 1, "", "optim_adam_beta1"], [4, 4, 1, "", "optim_adam_beta2"], [4, 4, 1, "", "optim_beta1"], [4, 4, 1, "", "optim_beta2"], [4, 4, 1, "", "optim_beta3"], [4, 4, 1, "", "optim_dummy_beta1"], [4, 4, 1, "", "optim_dummy_beta2"], [4, 4, 1, "", "optim_momentum"], [4, 4, 1, "", "optim_weight_decay"], [4, 4, 1, "", "remove_unused_columns"], [4, 4, 1, "", "save_language_projection"], [4, 4, 1, "", "use_customized_optim"], [4, 4, 1, "", "use_lisa"]], "lmflow.args.InferencerArguments": [[4, 3, 1, "", "__post_init__"], [4, 4, 1, "", "additional_stop_token_ids"], [4, 4, 1, "", "apply_chat_template"], [4, 4, 1, "", "deepspeed"], [4, 4, 1, "", "device"], [4, 4, 1, "", "distributed_inference_num_instances"], [4, 4, 1, "", "do_sample"], [4, 4, 1, "", "enable_decode_inference_result"], [4, 4, 1, "", "enable_distributed_inference"], [4, 4, 1, "", "inference_batch_size"], [4, 4, 1, "", "local_rank"], [4, 4, 1, "", "max_new_tokens"], [4, 4, 1, "", "mixed_precision"], [4, 4, 1, "", "num_output_sequences"], [4, 4, 1, "", "random_seed"], [4, 4, 1, "", "repetition_penalty"], [4, 4, 1, "", "results_path"], [4, 4, 1, "", "save_results"], [4, 4, 1, "", "temperature"], [4, 4, 1, "", "tensor_parallel_size"], [4, 4, 1, "", "top_k"], [4, 4, 1, "", "top_p"], [4, 4, 1, "", "use_accelerator"], [4, 4, 1, "", "use_beam_search"], [4, 4, 1, "", "use_vllm"], [4, 4, 1, "", "vllm_gpu_memory_utilization"], [4, 4, 1, "", "vllm_inference_batch_size"], [4, 4, 1, "", "vllm_tensor_parallel_size"]], "lmflow.args.IterativeAlignerArguments": [[4, 4, 1, "", "dataset_path_list"], [4, 4, 1, "", "initial_iter_idx"]], "lmflow.args.IterativeDPOAlignerArguments": [[4, 4, 1, "", "do_dpo_align"], [4, 4, 1, "", "do_response_generation"], [4, 4, 1, "", "do_scoring"], [4, 4, 1, "", "output_dir"], [4, 4, 1, "", "reward_model_inference_batch_size"], [4, 4, 1, "", "reward_model_inference_block_size"]], "lmflow.args.ModelArguments": [[4, 3, 1, "", "__post_init__"], [4, 4, 1, "", "arch_type"], [4, 4, 1, "", "bits"], [4, 4, 1, "", "cache_dir"], [4, 4, 1, "", "config_name"], [4, 4, 1, "", "config_overrides"], [4, 4, 1, "", "do_rope_scaling"], [4, 4, 1, "", "double_quant"], [4, 4, 1, "", "eos_padding"], [4, 4, 1, "", "ignore_bias_buffers"], [4, 4, 1, "", "load_in_4bit"], [4, 4, 1, "", "lora_alpha"], [4, 4, 1, "", "lora_dropout"], [4, 4, 1, "", "lora_model_path"], [4, 4, 1, "", "lora_r"], [4, 4, 1, "", "lora_target_modules"], [4, 4, 1, "", "model_max_length"], [4, 4, 1, "", "model_name_or_path"], [4, 4, 1, "", "model_revision"], [4, 4, 1, "", "model_type"], [4, 4, 1, "", "padding_side"], [4, 4, 1, "", "quant_type"], [4, 4, 1, "", "rope_ntk_ratio"], [4, 4, 1, "", "rope_pi_ratio"], [4, 4, 1, "", "save_aggregated_lora"], [4, 4, 1, "", "token"], [4, 4, 1, "", "tokenizer_name"], [4, 4, 1, "", "torch_dtype"], [4, 4, 1, "", "truncate_to_model_max_length"], [4, 4, 1, "", "truncation_side"], [4, 4, 1, "", "trust_remote_code"], [4, 4, 1, "", "use_dora"], [4, 4, 1, "", "use_fast_tokenizer"], [4, 4, 1, "", "use_flash_attention"], [4, 4, 1, "", "use_int8"], [4, 4, 1, "", "use_lora"], [4, 4, 1, "", "use_qlora"], [4, 4, 1, "", "use_ram_optimized_load"]], "lmflow.args.MultiModalDatasetArguments": [[4, 4, 1, "", "image_aspect_ratio"], [4, 4, 1, "", "image_folder"], [4, 4, 1, "", "is_multimodal"], [4, 4, 1, "", "sep_style"], [4, 4, 1, "", "use_image_start_end"]], "lmflow.args.OptimizerNames": [[4, 4, 1, "", "ADABELIEF"], [4, 4, 1, "", "ADABOUND"], [4, 4, 1, "", "ADADELTA"], [4, 4, 1, "", "ADAGRAD"], [4, 4, 1, "", "ADAM"], [4, 4, 1, "", "ADAMAX"], [4, 4, 1, "", "ADAMP"], [4, 4, 1, "", "ADAMW_SCHEDULE_FREE"], [4, 4, 1, "", "ADAN"], [4, 4, 1, "", "DUMMY"], [4, 4, 1, "", "LAMB"], [4, 4, 1, "", "LARS"], [4, 4, 1, "", "NADAM"], [4, 4, 1, "", "NOVOGRAD"], [4, 4, 1, "", "RADAM"], [4, 4, 1, "", "SGDP"], [4, 4, 1, "", "SGD_SCHEDULE_FREE"], [4, 4, 1, "", "SOPHIA"], [4, 4, 1, "", "YOGI"]], "lmflow.args.RaftAlignerArguments": [[4, 4, 1, "", "collection_strategy"], [4, 4, 1, "", "inference_batch_size_per_device"], [4, 4, 1, "", "num_raft_iteration"], [4, 4, 1, "", "output_max_length"], [4, 4, 1, "", "output_min_length"], [4, 4, 1, "", "output_reward_path"], [4, 4, 1, "", "raft_batch_size"], [4, 4, 1, "", "top_reward_percentage"]], "lmflow.args.VisModelArguments": [[4, 4, 1, "", "custom_model"], [4, 4, 1, "", "custom_vision_model"], [4, 4, 1, "", "image_encoder_name_or_path"], [4, 4, 1, "", "llava_loading"], [4, 4, 1, "", "llava_pretrain_model_path"], [4, 4, 1, "", "llm_model_name_or_path"], [4, 4, 1, "", "low_resource"], [4, 4, 1, "", "pretrained_language_projection_path"], [4, 4, 1, "", "prompt_cache_path"], [4, 4, 1, "", "qformer_name_or_path"], [4, 4, 1, "", "save_pretrain_model_path"], [4, 4, 1, "", "use_prompt_cache"], [4, 4, 1, "", "vision_select_layer"], [4, 4, 1, "", "with_qformer"]], "lmflow.datasets": [[6, 2, 1, "", "Dataset"], [5, 0, 0, "-", "dataset"], [6, 5, 1, "", "is_multimodal_available"], [7, 0, 0, "-", "multi_modal_dataset"]], "lmflow.datasets.Dataset": [[6, 3, 1, "", "__len__"], [6, 3, 1, "", "_check_data_format"], [6, 4, 1, "", "backend"], [6, 4, 1, "", "backend_dataset"], [6, 3, 1, "", "create_from_dict"], [6, 4, 1, "", "data_args"], [6, 4, 1, "", "dataset_path"], [6, 3, 1, "", "drop_instances"], [6, 3, 1, "", "from_dict"], [6, 3, 1, "", "get_backend"], [6, 3, 1, "", "get_backend_dataset"], [6, 3, 1, "", "get_data_args"], [6, 3, 1, "", "get_fingerprint"], [6, 3, 1, "", "get_type"], [6, 3, 1, "", "hf_dataset_sanity_check"], [6, 3, 1, "", "map"], [6, 3, 1, "", "sample"], [6, 3, 1, "", "sanity_check"], [6, 3, 1, "", "save"], [6, 3, 1, "", "to_dict"], [6, 3, 1, "", "to_list"], [6, 3, 1, "", "train_test_split"], [6, 4, 1, "", "type"]], "lmflow.datasets.dataset": [[5, 1, 1, "", "DATASET_TYPES"], [5, 2, 1, "", "Dataset"], [5, 1, 1, "", "KEY_INSTANCES"], [5, 1, 1, "", "KEY_SCORE"], [5, 1, 1, "", "KEY_TYPE"], [5, 1, 1, "", "logger"]], "lmflow.datasets.dataset.Dataset": [[5, 3, 1, "", "__len__"], [5, 3, 1, "", "_check_data_format"], [5, 4, 1, "", "backend"], [5, 4, 1, "", "backend_dataset"], [5, 3, 1, "", "create_from_dict"], [5, 4, 1, "", "data_args"], [5, 4, 1, "", "dataset_path"], [5, 3, 1, "", "drop_instances"], [5, 3, 1, "", "from_dict"], [5, 3, 1, "", "get_backend"], [5, 3, 1, "", "get_backend_dataset"], [5, 3, 1, "", "get_data_args"], [5, 3, 1, "", "get_fingerprint"], [5, 3, 1, "", "get_type"], [5, 3, 1, "", "hf_dataset_sanity_check"], [5, 3, 1, "", "map"], [5, 3, 1, "", "sample"], [5, 3, 1, "", "sanity_check"], [5, 3, 1, "", "save"], [5, 3, 1, "", "to_dict"], [5, 3, 1, "", "to_list"], [5, 3, 1, "", "train_test_split"], [5, 4, 1, "", "type"]], "lmflow.datasets.multi_modal_dataset": [[7, 2, 1, "", "CustomMultiModalDataset"], [7, 2, 1, "", "DataCollatorForSupervisedDataset"], [7, 5, 1, "", "preprocess_llama_from_llava_plain"], [7, 5, 1, "", "preprocess_llama_from_llava_v1"], [7, 5, 1, "", "preprocess_multimodal_llava"], [7, 5, 1, "", "tokenizer_image_token"]], "lmflow.datasets.multi_modal_dataset.CustomMultiModalDataset": [[7, 3, 1, "", "__getitem__"], [7, 3, 1, "", "__len__"], [7, 4, 1, "", "data_args"], [7, 4, 1, "", "data_dict"], [7, 4, 1, "", "image_folder"], [7, 3, 1, "", "register_tokenizer"]], "lmflow.datasets.multi_modal_dataset.DataCollatorForSupervisedDataset": [[7, 3, 1, "", "__call__"], [7, 4, 1, "", "tokenizer"]], "lmflow.models": [[9, 0, 0, "-", "auto_model"], [10, 0, 0, "-", "base_model"], [11, 0, 0, "-", "decoder_model"], [12, 0, 0, "-", "encoder_decoder_model"], [13, 0, 0, "-", "hf_decoder_model"], [14, 0, 0, "-", "hf_encoder_decoder_model"], [15, 0, 0, "-", "hf_model_mixin"], [16, 0, 0, "-", "hf_text_regression_model"], [18, 0, 0, "-", "interfaces"], [20, 0, 0, "-", "regression_model"], [21, 0, 0, "-", "text_regression_model"], [22, 0, 0, "-", "vision2seq_model"], [24, 0, 0, "-", "vision_encoder"]], "lmflow.models.auto_model": [[9, 2, 1, "", "AutoModel"]], "lmflow.models.auto_model.AutoModel": [[9, 3, 1, "", "get_model"]], "lmflow.models.base_model": [[10, 2, 1, "", "BaseModel"]], "lmflow.models.decoder_model": [[11, 2, 1, "", "DecoderModel"]], "lmflow.models.encoder_decoder_model": [[12, 2, 1, "", "EncoderDecoderModel"]], "lmflow.models.hf_decoder_model": [[13, 2, 1, "", "HFDecoderModel"], [13, 1, 1, "", "logger"]], "lmflow.models.hf_decoder_model.HFDecoderModel": [[13, 3, 1, "", "__inference"], [13, 3, 1, "", "__prepare_inputs_for_inference"], [13, 3, 1, "", "__prepare_inputs_for_vllm_inference"], [13, 3, 1, "", "__vllm_inference"], [13, 3, 1, "", "decode"], [13, 3, 1, "", "encode"], [13, 3, 1, "", "get_peft_without_qlora"], [13, 3, 1, "", "inference"], [13, 3, 1, "", "merge_lora_weights"], [13, 3, 1, "", "prepare_inputs_for_inference"], [13, 3, 1, "", "save"], [13, 3, 1, "", "tokenize"]], "lmflow.models.hf_encoder_decoder_model": [[14, 2, 1, "", "HFEncoderDecoderModel"], [14, 1, 1, "", "logger"]], "lmflow.models.hf_encoder_decoder_model.HFEncoderDecoderModel": [[14, 3, 1, "", "decode"], [14, 4, 1, "", "device"], [14, 3, 1, "", "encode"], [14, 3, 1, "", "get_backend_model"], [14, 3, 1, "", "get_max_length"], [14, 3, 1, "", "get_tokenizer"], [14, 3, 1, "", "inference"], [14, 3, 1, "", "merge_lora_weights"], [14, 3, 1, "", "save"], [14, 3, 1, "", "tokenize"]], "lmflow.models.hf_model_mixin": [[15, 2, 1, "", "HFModelMixin"], [15, 1, 1, "", "HF_AUTOMODEL_MAPPING"], [15, 1, 1, "", "HF_AUTOMODEL_TYPE"], [15, 1, 1, "", "LORA_TARGET_MODULES_MAPPING"], [15, 1, 1, "", "logger"]], "lmflow.models.hf_model_mixin.HFModelMixin": [[15, 3, 1, "", "__model_module_inject"], [15, 3, 1, "", "__prepare_dtype"], [15, 3, 1, "", "__prepare_model_config"], [15, 3, 1, "", "__prepare_model_for_inference"], [15, 3, 1, "", "__prepare_model_for_training"], [15, 3, 1, "", "__prepare_model_for_vllm_inference"], [15, 3, 1, "", "__prepare_model_post_process"], [15, 3, 1, "", "__prepare_peft_config"], [15, 3, 1, "", "__prepare_quant_config"], [15, 3, 1, "", "__prepare_tokenizer"], [15, 4, 1, "", "_activated"], [15, 3, 1, "", "activate_model_for_inference"], [15, 3, 1, "", "deactivate_model_for_inference"], [15, 4, 1, "", "device"], [15, 4, 1, "", "do_train"], [15, 4, 1, "", "ds_config"], [15, 3, 1, "", "get_backend_model"], [15, 3, 1, "", "get_max_length"], [15, 3, 1, "", "get_tokenizer"], [15, 4, 1, "", "hf_auto_model"], [15, 4, 1, "", "hf_model_config"], [15, 4, 1, "", "model_args"], [15, 4, 1, "", "peft_config"], [15, 4, 1, "", "quant_config"], [15, 4, 1, "", "tokenizer"], [15, 4, 1, "", "torch_dtype"], [15, 4, 1, "", "use_accelerator"]], "lmflow.models.hf_text_regression_model": [[16, 2, 1, "", "HFTextRegressionModel"], [16, 1, 1, "", "logger"]], "lmflow.models.hf_text_regression_model.HFTextRegressionModel": [[16, 3, 1, "", "__inference"], [16, 3, 1, "", "__vllm_inference"], [16, 3, 1, "", "inference"], [16, 3, 1, "", "postprocess_distributed_inference_outputs"], [16, 3, 1, "", "postprocess_inference_outputs"], [16, 3, 1, "", "prepare_inputs_for_inference"], [16, 3, 1, "", "save"], [16, 3, 1, "", "tokenize"]], "lmflow.models.interfaces": [[19, 0, 0, "-", "tunable"]], "lmflow.models.interfaces.tunable": [[19, 2, 1, "", "Tunable"]], "lmflow.models.regression_model": [[20, 2, 1, "", "RegressionModel"]], "lmflow.models.text_regression_model": [[21, 2, 1, "", "TextRegressionModel"]], "lmflow.models.text_regression_model.TextRegressionModel": [[21, 3, 1, "", "inference"], [21, 4, 1, "", "inference_func"], [21, 3, 1, "", "register_inference_function"]], "lmflow.models.vision2seq_model": [[22, 2, 1, "", "CustomAutoVision2SeqModel"]], "lmflow.models.vision2seq_model.CustomAutoVision2SeqModel": [[22, 4, 1, "", "custom_vision_model"], [22, 3, 1, "", "forward"], [22, 3, 1, "", "generate"], [22, 3, 1, "", "get_backend_model"], [22, 3, 1, "", "get_tokenizer"], [22, 4, 1, "", "hidden_size"], [22, 4, 1, "", "language_model"], [22, 3, 1, "", "language_model_from_pretrained"], [22, 3, 1, "", "load_prompt_cache"], [22, 3, 1, "", "processor_image_token_in_minigpt4"], [22, 3, 1, "", "qformer_from_pretrained"], [22, 3, 1, "", "register_prompt_cache"], [22, 3, 1, "", "save_prompt_cache"], [22, 3, 1, "", "vision_feature_select"], [22, 3, 1, "", "vision_model_from_pretrained"], [22, 4, 1, "", "with_qformer"]], "lmflow.models.vision_encoder": [[24, 5, 1, "", "build_vision_tower"], [23, 0, 0, "-", "clip_encoder"]], "lmflow.models.vision_encoder.clip_encoder": [[23, 2, 1, "", "CLIPVisionTower"], [23, 5, 1, "", "build_vision_tower"]], "lmflow.models.vision_encoder.clip_encoder.CLIPVisionTower": [[23, 6, 1, "", "config"], [23, 6, 1, "", "device"], [23, 6, 1, "", "dtype"], [23, 6, 1, "", "dummy_feature"], [23, 3, 1, "", "encode_images"], [23, 3, 1, "", "feature_select"], [23, 3, 1, "", "forward"], [23, 6, 1, "", "hidden_size"], [23, 4, 1, "", "is_loaded"], [23, 3, 1, "", "load_model"], [23, 6, 1, "", "num_patches"], [23, 3, 1, "", "prepare_inputs_labels_for_multimodal"], [23, 4, 1, "", "select_feature"], [23, 4, 1, "", "select_layer"], [23, 4, 1, "", "vision_tower_name"]], "lmflow.optim": [[25, 0, 0, "-", "adabelief"], [26, 0, 0, "-", "adabound"], [27, 0, 0, "-", "adadelta"], [28, 0, 0, "-", "adagrad"], [29, 0, 0, "-", "adam"], [30, 0, 0, "-", "adamax"], [31, 0, 0, "-", "adamp"], [32, 0, 0, "-", "adamw_schedule_free"], [33, 0, 0, "-", "adan"], [34, 0, 0, "-", "dummy"], [36, 0, 0, "-", "lamb"], [37, 0, 0, "-", "lars"], [38, 0, 0, "-", "nadam"], [39, 0, 0, "-", "novograd"], [40, 0, 0, "-", "optimizers"], [41, 0, 0, "-", "radam"], [42, 0, 0, "-", "sgd_schedule_free"], [43, 0, 0, "-", "sgdp"], [44, 0, 0, "-", "sophia"], [45, 0, 0, "-", "yogi"]], "lmflow.optim.adabelief": [[25, 2, 1, "", "AdaBelief"]], "lmflow.optim.adabelief.AdaBelief": [[25, 3, 1, "", "__setstate__"], [25, 4, 1, "", "degenerated_to_sgd"], [25, 4, 1, "", "fixed_decay"], [25, 4, 1, "", "rectify"], [25, 3, 1, "", "reset"], [25, 3, 1, "", "step"], [25, 4, 1, "", "weight_decouple"]], "lmflow.optim.adabound": [[26, 2, 1, "", "AdaBound"]], "lmflow.optim.adabound.AdaBound": [[26, 3, 1, "", "__setstate__"], [26, 4, 1, "", "base_lrs"], [26, 3, 1, "", "step"]], "lmflow.optim.adadelta": [[27, 2, 1, "", "Adadelta"]], "lmflow.optim.adadelta.Adadelta": [[27, 3, 1, "", "step"]], "lmflow.optim.adagrad": [[28, 2, 1, "", "AdaGrad"]], "lmflow.optim.adagrad.AdaGrad": [[28, 3, 1, "", "step"]], "lmflow.optim.adam": [[29, 2, 1, "", "Adam"]], "lmflow.optim.adam.Adam": [[29, 3, 1, "", "step"]], "lmflow.optim.adamax": [[30, 2, 1, "", "Adamax"]], "lmflow.optim.adamax.Adamax": [[30, 3, 1, "", "__setstate__"], [30, 3, 1, "", "step"]], "lmflow.optim.adamp": [[31, 2, 1, "", "AdamP"]], "lmflow.optim.adamp.AdamP": [[31, 3, 1, "", "_channel_view"], [31, 3, 1, "", "_cosine_similarity"], [31, 3, 1, "", "_layer_view"], [31, 3, 1, "", "_projection"], [31, 3, 1, "", "step"]], "lmflow.optim.adamw_schedule_free": [[32, 2, 1, "", "AdamWScheduleFree"]], "lmflow.optim.adamw_schedule_free.AdamWScheduleFree": [[32, 3, 1, "", "eval"], [32, 3, 1, "", "step"], [32, 3, 1, "", "train"]], "lmflow.optim.adan": [[33, 2, 1, "", "Adan"], [33, 5, 1, "", "_multi_tensor_adan"], [33, 5, 1, "", "_single_tensor_adan"]], "lmflow.optim.adan.Adan": [[33, 3, 1, "", "__setstate__"], [33, 3, 1, "", "restart_opt"], [33, 3, 1, "", "step"]], "lmflow.optim.dummy": [[34, 2, 1, "", "Dummy"]], "lmflow.optim.dummy.Dummy": [[34, 3, 1, "", "step"]], "lmflow.optim.lamb": [[36, 2, 1, "", "Lamb"]], "lmflow.optim.lamb.Lamb": [[36, 4, 1, "", "adam"], [36, 4, 1, "", "clamp_value"], [36, 4, 1, "", "debias"], [36, 3, 1, "", "step"]], "lmflow.optim.lars": [[37, 2, 1, "", "LARS"]], "lmflow.optim.lars.LARS": [[37, 3, 1, "", "__setstate__"], [37, 3, 1, "", "step"]], "lmflow.optim.nadam": [[38, 2, 1, "", "NAdam"]], "lmflow.optim.nadam.NAdam": [[38, 3, 1, "", "__setstate__"], [38, 3, 1, "", "step"]], "lmflow.optim.novograd": [[39, 2, 1, "", "NovoGrad"]], "lmflow.optim.novograd.NovoGrad": [[39, 3, 1, "", "__setstate__"], [39, 3, 1, "", "step"]], "lmflow.optim.radam": [[41, 2, 1, "", "RAdam"]], "lmflow.optim.radam.RAdam": [[41, 3, 1, "", "__setstate__"], [41, 3, 1, "", "step"]], "lmflow.optim.sgd_schedule_free": [[42, 2, 1, "", "SGDScheduleFree"]], "lmflow.optim.sgd_schedule_free.SGDScheduleFree": [[42, 3, 1, "", "eval"], [42, 3, 1, "", "step"], [42, 3, 1, "", "train"]], "lmflow.optim.sgdp": [[43, 2, 1, "", "SGDP"]], "lmflow.optim.sgdp.SGDP": [[43, 3, 1, "", "_channel_view"], [43, 3, 1, "", "_cosine_similarity"], [43, 3, 1, "", "_layer_view"], [43, 3, 1, "", "_projection"], [43, 3, 1, "", "step"]], "lmflow.optim.sophia": [[44, 2, 1, "", "SophiaG"]], "lmflow.optim.sophia.SophiaG": [[44, 3, 1, "", "__setstate__"], [44, 3, 1, "", "step"], [44, 3, 1, "", "update_hessian"]], "lmflow.optim.yogi": [[45, 2, 1, "", "Yogi"]], "lmflow.optim.yogi.Yogi": [[45, 3, 1, "", "step"]], "lmflow.pipeline": [[46, 0, 0, "-", "auto_pipeline"], [47, 0, 0, "-", "base_aligner"], [48, 0, 0, "-", "base_pipeline"], [49, 0, 0, "-", "base_tuner"], [50, 0, 0, "-", "dpo_aligner"], [51, 0, 0, "-", "dpov2_aligner"], [52, 0, 0, "-", "evaluator"], [53, 0, 0, "-", "finetuner"], [55, 0, 0, "-", "inferencer"], [56, 0, 0, "-", "iterative_dpo_aligner"], [57, 0, 0, "-", "raft_aligner"], [58, 0, 0, "-", "rm_inferencer"], [59, 0, 0, "-", "rm_tuner"], [62, 0, 0, "-", "utils"], [69, 0, 0, "-", "vllm_inferencer"]], "lmflow.pipeline.auto_pipeline": [[46, 2, 1, "", "AutoPipeline"], [46, 1, 1, "", "PIPELINE_MAPPING"], [46, 1, 1, "", "PIPELINE_NEEDS_EXTRAS"]], "lmflow.pipeline.auto_pipeline.AutoPipeline": [[46, 3, 1, "", "get_pipeline"]], "lmflow.pipeline.base_aligner": [[47, 2, 1, "", "BaseAligner"]], "lmflow.pipeline.base_aligner.BaseAligner": [[47, 3, 1, "", "_check_if_alignable"], [47, 3, 1, "", "align"]], "lmflow.pipeline.base_pipeline": [[48, 2, 1, "", "BasePipeline"]], "lmflow.pipeline.base_tuner": [[49, 2, 1, "", "BaseTuner"]], "lmflow.pipeline.base_tuner.BaseTuner": [[49, 3, 1, "", "_check_if_tunable"], [49, 3, 1, "", "tune"]], "lmflow.pipeline.dpo_aligner": [[50, 2, 1, "", "DPOAligner"], [50, 5, 1, "", "get_paired_dataset"]], "lmflow.pipeline.dpo_aligner.DPOAligner": [[50, 3, 1, "", "_initialize_trainer"], [50, 3, 1, "", "_load_dataset"], [50, 3, 1, "", "align"], [50, 4, 1, "", "aligner_args"], [50, 4, 1, "", "data_args"], [50, 4, 1, "", "eval_dataset"], [50, 4, 1, "", "model_args"], [50, 4, 1, "", "train_dataset"]], "lmflow.pipeline.dpov2_aligner": [[51, 2, 1, "", "DPOv2Aligner"], [51, 2, 1, "", "MemorySafeDPOv2Aligner"], [51, 1, 1, "", "ReferenceModelArguments"], [51, 1, 1, "", "logger"]], "lmflow.pipeline.dpov2_aligner.DPOv2Aligner": [[51, 3, 1, "", "__prepare_training_args"], [51, 3, 1, "", "_calc_response_lengths"], [51, 3, 1, "", "_calc_reward_with_length_penalty"], [51, 3, 1, "", "_sampling_paired_idx_from_rewards"], [51, 3, 1, "", "_sampling_paired_idx_from_rewards_fast"], [51, 3, 1, "", "align"], [51, 4, 1, "", "aligner_args"], [51, 3, 1, "", "convert_to_paired_dataset"], [51, 4, 1, "", "data_args"], [51, 4, 1, "", "model_args"], [51, 4, 1, "", "ref_model_args"], [51, 3, 1, "", "sampling_paired_idx_from_rewards"]], "lmflow.pipeline.dpov2_aligner.MemorySafeDPOv2Aligner": [[51, 3, 1, "", "align"], [51, 4, 1, "", "aligner_args"], [51, 4, 1, "", "aligner_file_path"], [51, 4, 1, "", "data_args"], [51, 4, 1, "", "model_args"], [51, 4, 1, "", "ref_model_args"]], "lmflow.pipeline.evaluator": [[52, 2, 1, "", "Evaluator"]], "lmflow.pipeline.evaluator.Evaluator": [[52, 3, 1, "", "_evaluate_acc_with_accelerator"], [52, 3, 1, "", "_evaluate_acc_with_deepspeed"], [52, 3, 1, "", "_evaluate_nll"], [52, 3, 1, "", "_evaluate_ppl"], [52, 3, 1, "", "_match"], [52, 4, 1, "", "block_size"], [52, 4, 1, "", "config"], [52, 3, 1, "", "create_dataloader"], [52, 4, 1, "", "data_args"], [52, 3, 1, "", "evaluate"], [52, 4, 1, "", "evaluator_args"], [52, 4, 1, "", "local_rank"], [52, 4, 1, "", "model_args"], [52, 4, 1, "", "world_size"]], "lmflow.pipeline.finetuner": [[53, 2, 1, "", "Finetuner"], [53, 1, 1, "", "logger"]], "lmflow.pipeline.finetuner.Finetuner": [[53, 3, 1, "", "create_customized_optimizer"], [53, 4, 1, "", "data_args"], [53, 4, 1, "", "finetuner_args"], [53, 3, 1, "", "group_text"], [53, 4, 1, "", "last_checkpoint"], [53, 4, 1, "", "model_args"], [53, 3, 1, "", "tune"]], "lmflow.pipeline.inferencer": [[55, 2, 1, "", "Inferencer"], [55, 2, 1, "", "SpeculativeInferencer"], [55, 2, 1, "", "ToolInferencer"], [55, 1, 1, "", "logger"], [55, 5, 1, "", "rstrip_partial_utf8"], [55, 1, 1, "", "supported_dataset_type"]], "lmflow.pipeline.inferencer.Inferencer": [[55, 4, 1, "", "config"], [55, 3, 1, "", "create_dataloader"], [55, 4, 1, "", "data_args"], [55, 3, 1, "", "inference"], [55, 4, 1, "", "inferencer_args"], [55, 4, 1, "", "local_rank"], [55, 4, 1, "", "model_args"], [55, 3, 1, "", "stream_inference"], [55, 4, 1, "", "world_size"]], "lmflow.pipeline.inferencer.SpeculativeInferencer": [[55, 3, 1, "", "autoregressive_sampling"], [55, 4, 1, "", "draft_config"], [55, 4, 1, "", "draft_model_args"], [55, 3, 1, "", "inference"], [55, 3, 1, "", "predict_next_token"], [55, 3, 1, "", "sample"], [55, 3, 1, "", "score_to_prob"], [55, 3, 1, "", "stream_inference"]], "lmflow.pipeline.inferencer.ToolInferencer": [[55, 3, 1, "", "code_exec"], [55, 3, 1, "", "inference"], [55, 4, 1, "", "model"]], "lmflow.pipeline.iterative_dpo_aligner": [[56, 2, 1, "", "IterativeDPOAligner"], [56, 1, 1, "", "logger"]], "lmflow.pipeline.iterative_dpo_aligner.IterativeDPOAligner": [[56, 3, 1, "", "__filter_args"], [56, 3, 1, "", "_align_single_iteration"], [56, 3, 1, "", "_do_reward_model_inference"], [56, 3, 1, "", "_do_single_dpo_align"], [56, 3, 1, "", "_do_target_model_inference"], [56, 3, 1, "", "_parse_dpo_aligner_args"], [56, 3, 1, "", "_parse_reward_model_inference_args"], [56, 3, 1, "", "_parse_target_model_inference_args"], [56, 3, 1, "", "align"], [56, 4, 1, "", "aligner_args"], [56, 4, 1, "", "data_args"], [56, 4, 1, "", "model_args"], [56, 4, 1, "", "ref_model_args"], [56, 4, 1, "", "reward_model_args"], [56, 4, 1, "", "workspace_path"]], "lmflow.pipeline.raft_aligner": [[57, 2, 1, "", "RaftAligner"], [57, 1, 1, "", "logger"]], "lmflow.pipeline.raft_aligner.RaftAligner": [[57, 4, 1, "", "INF"], [57, 3, 1, "", "_clean_text"], [57, 3, 1, "", "_discard_sample"], [57, 3, 1, "", "_get_batch_dataset_local"], [57, 3, 1, "", "_get_batch_dataset_top"], [57, 3, 1, "", "_initialize_trainer"], [57, 3, 1, "", "_load_dataset"], [57, 3, 1, "", "_load_input_dataset"], [57, 3, 1, "", "align"], [57, 4, 1, "", "aligner_args"], [57, 4, 1, "", "data_args"], [57, 4, 1, "", "model_args"]], "lmflow.pipeline.rm_inferencer": [[58, 2, 1, "", "RewardModelInferencer"], [58, 1, 1, "", "logger"]], "lmflow.pipeline.rm_inferencer.RewardModelInferencer": [[58, 3, 1, "", "__distributed_inference"], [58, 3, 1, "", "__inference"], [58, 3, 1, "", "__post_process_model_output"], [58, 3, 1, "", "__vllm_inference"], [58, 3, 1, "", "_inference"], [58, 3, 1, "", "compress_list"], [58, 4, 1, "", "data_args"], [58, 3, 1, "", "flatten_list"], [58, 3, 1, "", "inference"], [58, 4, 1, "", "inferencer_args"], [58, 4, 1, "", "local_rank"], [58, 4, 1, "", "model_args"], [58, 4, 1, "", "world_size"]], "lmflow.pipeline.rm_tuner": [[59, 2, 1, "", "RewardModelTuner"], [59, 1, 1, "", "logger"]], "lmflow.pipeline.rm_tuner.RewardModelTuner": [[59, 3, 1, "", "tune"]], "lmflow.pipeline.utils": [[60, 0, 0, "-", "dpov2_dataprocessor"], [61, 0, 0, "-", "dpov2_trainer"], [63, 0, 0, "-", "memory_safe_dpov2_align"], [64, 0, 0, "-", "memory_safe_vllm_inference"], [65, 0, 0, "-", "peft_trainer"], [66, 0, 0, "-", "raft_trainer"], [67, 0, 0, "-", "rm_dataprocessor"], [68, 0, 0, "-", "rm_trainer"]], "lmflow.pipeline.utils.dpov2_dataprocessor": [[60, 2, 1, "", "PreferenceDataCollatorWithPadding"], [60, 1, 1, "", "logger"]], "lmflow.pipeline.utils.dpov2_dataprocessor.PreferenceDataCollatorWithPadding": [[60, 3, 1, "", "__call__"], [60, 3, 1, "", "collate"], [60, 4, 1, "", "is_encoder_decoder"], [60, 4, 1, "", "label_pad_token_id"], [60, 4, 1, "", "mask_prompt"], [60, 4, 1, "", "max_length"], [60, 4, 1, "", "max_prompt_length"], [60, 4, 1, "", "max_target_length"], [60, 4, 1, "", "model"], [60, 4, 1, "", "padding"], [60, 4, 1, "", "padding_value"], [60, 3, 1, "", "tokenize_batch_element"], [60, 4, 1, "", "tokenizer"], [60, 4, 1, "", "truncation_mode"]], "lmflow.pipeline.utils.dpov2_trainer": [[61, 2, 1, "", "DPOv2Trainer"], [61, 1, 1, "", "logger"]], "lmflow.pipeline.utils.dpov2_trainer.DPOv2Trainer": [[61, 3, 1, "", "dpo_loss"], [61, 3, 1, "", "get_batch_loss_metrics"], [61, 3, 1, "", "get_batch_metrics"], [61, 4, 1, "", "len_penalty"], [61, 4, 1, "", "use_dpo_data_collator"]], "lmflow.pipeline.utils.memory_safe_dpov2_align": [[63, 1, 1, "", "ReferenceModelArguments"], [63, 1, 1, "", "logger"], [63, 5, 1, "", "main"]], "lmflow.pipeline.utils.memory_safe_vllm_inference": [[64, 1, 1, "", "logger"], [64, 5, 1, "", "main"]], "lmflow.pipeline.utils.peft_trainer": [[65, 2, 1, "", "PeftSavingCallback"], [65, 2, 1, "", "PeftTrainer"]], "lmflow.pipeline.utils.peft_trainer.PeftSavingCallback": [[65, 3, 1, "", "_save"], [65, 3, 1, "", "on_epoch_end"], [65, 3, 1, "", "on_save"], [65, 3, 1, "", "on_train_end"]], "lmflow.pipeline.utils.peft_trainer.PeftTrainer": [[65, 3, 1, "", "_save_checkpoint"]], "lmflow.pipeline.utils.raft_trainer": [[66, 1, 1, "", "DEFAULT_CALLBACKS"], [66, 1, 1, "id0", "DEFAULT_PROGRESS_CALLBACK"], [66, 1, 1, "", "IS_SAGEMAKER_MP_POST_1_10"], [66, 1, 1, "", "OPTIMIZER_NAME"], [66, 2, 1, "", "RaftTrainer"], [66, 1, 1, "", "SCALER_NAME"], [66, 1, 1, "", "SCHEDULER_NAME"], [66, 1, 1, "", "TRAINER_STATE_NAME"], [66, 1, 1, "", "TRAINING_ARGS_NAME"], [66, 1, 1, "", "_is_native_cpu_amp_available"], [66, 1, 1, "", "is_torch_greater_or_equal_than_1_10"], [66, 1, 1, "", "is_torch_less_than_1_11"], [66, 1, 1, "", "logger"], [66, 1, 1, "", "skip_first_batches"]], "lmflow.pipeline.utils.raft_trainer.RaftTrainer": [[66, 3, 1, "", "_add_sm_patterns_to_gitignore"], [66, 3, 1, "", "_gather_and_numpify"], [66, 3, 1, "", "_get_collator_with_removed_columns"], [66, 3, 1, "", "_get_eval_sampler"], [66, 3, 1, "", "_get_output_dir"], [66, 3, 1, "", "_get_train_sampler"], [66, 3, 1, "", "_hp_search_setup"], [66, 3, 1, "", "_inner_training_loop"], [66, 3, 1, "", "_issue_warnings_after_load"], [66, 3, 1, "", "_load_best_model"], [66, 3, 1, "", "_load_from_checkpoint"], [66, 3, 1, "", "_load_optimizer_and_scheduler"], [66, 3, 1, "", "_load_rng_state"], [66, 4, 1, "", "_loggers_initialized"], [66, 3, 1, "", "_maybe_log_save_evaluate"], [66, 4, 1, "", "_memory_tracker"], [66, 3, 1, "", "_move_model_to_device"], [66, 3, 1, "", "_nested_gather"], [66, 3, 1, "", "_one_train"], [66, 3, 1, "", "_pad_across_processes"], [66, 3, 1, "", "_prepare_input"], [66, 3, 1, "", "_prepare_inputs"], [66, 3, 1, "", "_push_from_checkpoint"], [66, 3, 1, "", "_remove_unused_columns"], [66, 3, 1, "", "_report_to_hp_search"], [66, 3, 1, "", "_rotate_checkpoints"], [66, 3, 1, "", "_save"], [66, 3, 1, "", "_save_checkpoint"], [66, 3, 1, "", "_save_tpu"], [66, 3, 1, "", "_set_signature_columns_if_needed"], [66, 4, 1, "", "_signature_columns"], [66, 3, 1, "", "_sorted_checkpoints"], [66, 4, 1, "", "_train_batch_size"], [66, 3, 1, "", "_tune_save_checkpoint"], [66, 3, 1, "", "_wrap_model"], [66, 3, 1, "", "add_callback"], [66, 4, 1, "", "args"], [66, 3, 1, "", "autocast_smart_context_manager"], [66, 3, 1, "", "call_model_init"], [66, 4, 1, "", "callback_handler"], [66, 4, 1, "", "can_return_loss"], [66, 3, 1, "", "compute_loss"], [66, 3, 1, "", "compute_loss_context_manager"], [66, 4, 1, "", "compute_metrics"], [66, 4, 1, "", "control"], [66, 3, 1, "", "create_model_card"], [66, 3, 1, "", "create_optimizer"], [66, 3, 1, "", "create_optimizer_and_scheduler"], [66, 3, 1, "", "create_scheduler"], [66, 4, 1, "", "current_flos"], [66, 4, 1, "", "data_collator"], [66, 4, 1, "", "deepspeed"], [66, 4, 1, "", "do_grad_scaling"], [66, 4, 1, "", "eval_dataset"], [66, 3, 1, "", "evaluate"], [66, 3, 1, "", "evaluation_loop"], [66, 3, 1, "", "floating_point_ops"], [66, 4, 1, "", "fsdp"], [66, 3, 1, "", "get_eval_dataloader"], [66, 3, 1, "", "get_optimizer_cls_and_kwargs"], [66, 3, 1, "", "get_test_dataloader"], [66, 3, 1, "", "get_train_dataloader"], [66, 4, 1, "", "hp_name"], [66, 4, 1, "", "hp_search_backend"], [66, 3, 1, "", "hyperparameter_search"], [66, 3, 1, "", "init_git_repo"], [66, 3, 1, "", "ipex_optimize_model"], [66, 4, 1, "", "is_in_train"], [66, 3, 1, "", "is_local_process_zero"], [66, 3, 1, "", "is_world_process_zero"], [66, 4, 1, "", "label_names"], [66, 3, 1, "", "log"], [66, 4, 1, "", "model"], [66, 4, 1, "", "model_wrapped"], [66, 3, 1, "", "num_examples"], [66, 4, 1, "", "place_model_on_device"], [66, 3, 1, "", "pop_callback"], [66, 3, 1, "", "predict"], [66, 3, 1, "", "prediction_loop"], [66, 3, 1, "", "prediction_step"], [66, 4, 1, "", "preprocess_logits_for_metrics"], [66, 3, 1, "", "push_to_hub"], [66, 3, 1, "", "remove_callback"], [66, 4, 1, "", "save_counter"], [66, 3, 1, "", "save_model"], [66, 4, 1, "", "sharded_ddp"], [66, 4, 1, "", "state"], [66, 3, 1, "", "store_flos"], [66, 4, 1, "", "tokenizer"], [66, 3, 1, "", "torch_jit_model_eval"], [66, 3, 1, "", "train"], [66, 4, 1, "", "train_dataset"], [66, 3, 1, "", "training_step"], [66, 4, 1, "", "use_apex"], [66, 4, 1, "", "use_cpu_amp"], [66, 4, 1, "", "use_cuda_amp"], [66, 4, 1, "", "use_tune_checkpoints"]], "lmflow.pipeline.utils.rm_dataprocessor": [[67, 2, 1, "", "RewardDataCollatorWithPadding"], [67, 1, 1, "", "logger"]], "lmflow.pipeline.utils.rm_dataprocessor.RewardDataCollatorWithPadding": [[67, 3, 1, "", "__call__"], [67, 4, 1, "", "max_length"], [67, 4, 1, "", "pad_to_multiple_of"], [67, 4, 1, "", "padding"], [67, 4, 1, "", "return_tensors"], [67, 4, 1, "", "tokenizer"]], "lmflow.pipeline.utils.rm_trainer": [[68, 2, 1, "", "PeftRewardTrainer"], [68, 2, 1, "", "RewardTrainer"], [68, 5, 1, "", "compute_metrics"], [68, 5, 1, "", "rm_loss"]], "lmflow.pipeline.utils.rm_trainer.PeftRewardTrainer": [[68, 3, 1, "", "compute_loss"]], "lmflow.pipeline.utils.rm_trainer.RewardTrainer": [[68, 3, 1, "", "compute_loss"]], "lmflow.pipeline.vllm_inferencer": [[69, 2, 1, "", "InferencerWithOffloading"], [69, 2, 1, "", "MemorySafeVLLMInferencer"], [69, 2, 1, "", "VLLMInferencer"], [69, 1, 1, "", "logger"]], "lmflow.pipeline.vllm_inferencer.InferencerWithOffloading": [[69, 4, 1, "", "data_args"], [69, 4, 1, "", "eos_token_id"], [69, 3, 1, "", "inference"], [69, 4, 1, "", "inferencer_args"], [69, 3, 1, "", "load_inference_results"], [69, 4, 1, "", "model_args"], [69, 3, 1, "", "save_inference_results"]], "lmflow.pipeline.vllm_inferencer.MemorySafeVLLMInferencer": [[69, 3, 1, "", "inference"], [69, 4, 1, "", "inferencer_file_path"]], "lmflow.pipeline.vllm_inferencer.VLLMInferencer": [[69, 3, 1, "", "_distributed_inference"], [69, 3, 1, "", "_inference"], [69, 3, 1, "", "inference"], [69, 3, 1, "", "load_inference_results"], [69, 3, 1, "", "parse_to_sampling_params"], [69, 4, 1, "", "sampling_params"], [69, 3, 1, "", "save_inference_results"]], "lmflow.tokenization": [[70, 0, 0, "-", "hf_decoder_model"], [71, 0, 0, "-", "hf_text_regression_model"]], "lmflow.tokenization.hf_decoder_model": [[70, 5, 1, "", "blocking"], [70, 5, 1, "", "conversation_tokenize_function"], [70, 1, 1, "", "logger"], [70, 1, 1, "", "tok_logger"], [70, 5, 1, "", "tokenize_function"]], "lmflow.tokenization.hf_text_regression_model": [[71, 5, 1, "", "blocking"], [71, 5, 1, "", "blocking_paired"], [71, 5, 1, "", "blocking_text_to_textlist"], [71, 5, 1, "", "conversation_tokenize_function"], [71, 1, 1, "", "logger"], [71, 5, 1, "", "paired_conversation_tokenize_function"], [71, 5, 1, "", "text_to_textlist_tokenize_function"], [71, 1, 1, "", "tok_logger"], [71, 5, 1, "", "tokenize_function"]], "lmflow.utils": [[73, 0, 0, "-", "common"], [74, 0, 0, "-", "constants"], [81, 0, 0, "-", "conversation_template"], [88, 0, 0, "-", "data_utils"], [92, 0, 0, "-", "flash_attention"], [96, 0, 0, "-", "llava_conversation_lib"], [97, 0, 0, "-", "model"], [98, 0, 0, "-", "multimodal"], [99, 0, 0, "-", "position_interpolation"], [101, 0, 0, "-", "versioning"]], "lmflow.utils.common": [[73, 5, 1, "", "add_dataclass_attr_prefix"], [73, 5, 1, "", "create_copied_dataclass"], [73, 1, 1, "", "logger"], [73, 5, 1, "", "make_shell_args_from_dataclass"], [73, 5, 1, "", "print_banner"], [73, 5, 1, "", "remove_dataclass_attr_prefix"]], "lmflow.utils.constants": [[74, 1, 1, "", "CONTROLLER_HEART_BEAT_EXPIRATION"], [74, 1, 1, "", "CONVERSATION_DATASET_DESCRIPTION"], [74, 1, 1, "", "CONVERSATION_ROLE_NAMES"], [74, 1, 1, "", "DATASET_DESCRIPTION_MAP"], [74, 1, 1, "", "DEFAULT_IMAGE_PATCH_TOKEN"], [74, 1, 1, "", "DEFAULT_IMAGE_TOKEN"], [74, 1, 1, "", "DEFAULT_IM_END_TOKEN"], [74, 1, 1, "", "DEFAULT_IM_START_TOKEN"], [74, 1, 1, "", "FLOAT_ONLY_DATASET_DESCRIPTION"], [74, 1, 1, "", "IGNORE_INDEX"], [74, 1, 1, "", "IMAGE_TOKEN_INDEX"], [74, 1, 1, "", "INSTANCE_FIELDS_MAP"], [74, 1, 1, "", "LMFLOW_LORA_TARGET_MODULES_MAPPING"], [74, 1, 1, "", "LOGDIR"], [74, 1, 1, "", "MEMORY_SAFE_DPOV2_ALIGN_ENV_VAR_TO_REMOVE"], [74, 1, 1, "", "MEMORY_SAFE_VLLM_INFERENCE_ENV_VAR_TO_REMOVE"], [74, 1, 1, "", "MEMORY_SAFE_VLLM_INFERENCE_FINISH_FLAG"], [74, 1, 1, "", "PAIRED_CONVERSATION_DATASET_DESCRIPTION"], [74, 1, 1, "", "PAIRED_TEXT_TO_TEXT_DATASET_DESCRIPTION"], [74, 1, 1, "", "RETURN_CODE_ERROR_BUFFER"], [74, 1, 1, "", "TEXT2TEXT_DATASET_DESCRIPTION"], [74, 1, 1, "", "TEXT2TEXT_DATASET_DETAILS"], [74, 1, 1, "", "TEXT2TEXT_DATASET_LONG_DESCRITION"], [74, 1, 1, "", "TEXT_ONLY_DATASET_DESCRIPTION"], [74, 1, 1, "", "TEXT_ONLY_DATASET_DETAILS"], [74, 1, 1, "", "TEXT_ONLY_DATASET_LONG_DESCRITION"], [74, 1, 1, "", "TEXT_TO_SCORED_TEXTLIST_DATASET_DESCRIPTION"], [74, 1, 1, "", "TEXT_TO_TEXTLIST_DATASET_DESCRIPTION"], [74, 1, 1, "", "WORKER_HEART_BEAT_INTERVAL"]], "lmflow.utils.conversation_template": [[81, 1, 1, "", "CHATGLM3_TEMPLATE"], [81, 1, 1, "", "CHATML_TEMPLATE"], [81, 2, 1, "", "ConversationTemplate"], [81, 2, 1, "", "ConversationTemplateForTool"], [81, 1, 1, "", "DEEPSEEK_TEMPLATE"], [81, 1, 1, "", "EMPTY_NO_SPECIAL_TOKENS_TEMPLATE"], [81, 1, 1, "", "EMPTY_TEMPLATE"], [81, 1, 1, "", "GEMMA_TEMPLATE"], [81, 1, 1, "", "HYMBA_TEMPLATE"], [81, 1, 1, "", "INTERNLM2_TEMPLATE"], [81, 1, 1, "", "LLAMA2_TEMPLATE"], [81, 1, 1, "", "LLAMA3_TEMPLATE"], [81, 1, 1, "", "LLAMA3_TEMPLATE_FOR_TOOL"], [81, 1, 1, "", "PHI3_TEMPLATE"], [81, 1, 1, "", "PRESET_TEMPLATES"], [81, 1, 1, "", "QWEN2_TEMPLATE"], [81, 1, 1, "", "QWEN2_TEMPLATE_FOR_TOOL"], [81, 1, 1, "", "YI1_5_TEMPLATE"], [81, 1, 1, "", "ZEPHYR_TEMPLATE"], [75, 0, 0, "-", "base"], [76, 0, 0, "-", "chatglm"], [77, 0, 0, "-", "chatml"], [78, 0, 0, "-", "deepseek"], [79, 0, 0, "-", "gemma"], [80, 0, 0, "-", "hymba"], [82, 0, 0, "-", "internlm"], [83, 0, 0, "-", "llama"], [84, 0, 0, "-", "phi"], [85, 0, 0, "-", "qwen"], [86, 0, 0, "-", "yi"], [87, 0, 0, "-", "zephyr"]], "lmflow.utils.conversation_template.ConversationTemplate": [[81, 3, 1, "", "__post_init__"], [81, 3, 1, "", "_encode"], [81, 3, 1, "", "_encode_template"], [81, 3, 1, "", "_ensure_id_list"], [81, 3, 1, "", "add_special_starter"], [81, 3, 1, "", "add_special_stopper"], [81, 4, 1, "", "assistant_formatter"], [81, 3, 1, "", "encode_conversation"], [81, 4, 1, "", "force_system"], [81, 4, 1, "", "function_formatter"], [81, 4, 1, "", "observation_formatter"], [81, 3, 1, "", "post_process_pairs"], [81, 4, 1, "", "remove_last_sep"], [81, 3, 1, "", "remove_last_separator"], [81, 4, 1, "", "separator"], [81, 4, 1, "", "special_starter"], [81, 4, 1, "", "special_stopper"], [81, 4, 1, "", "system_formatter"], [81, 4, 1, "", "template_name"], [81, 4, 1, "", "tools_formatter"], [81, 4, 1, "", "user_formatter"]], "lmflow.utils.conversation_template.ConversationTemplateForTool": [[81, 3, 1, "", "_encode"], [81, 3, 1, "", "_encode_template"], [81, 3, 1, "", "encode_conversation"]], "lmflow.utils.conversation_template.base": [[75, 2, 1, "", "ConversationTemplate"], [75, 2, 1, "", "ConversationTemplateForTool"], [75, 1, 1, "", "EMPTY_NO_SPECIAL_TOKENS_TEMPLATE"], [75, 1, 1, "", "EMPTY_TEMPLATE"], [75, 2, 1, "", "EmptyFormatter"], [75, 2, 1, "", "Formatter"], [75, 2, 1, "", "ListFormatter"], [75, 2, 1, "", "StringFormatter"], [75, 2, 1, "", "TemplateComponent"], [75, 1, 1, "", "logger"]], "lmflow.utils.conversation_template.base.ConversationTemplate": [[75, 3, 1, "", "__post_init__"], [75, 3, 1, "", "_encode"], [75, 3, 1, "", "_encode_template"], [75, 3, 1, "", "_ensure_id_list"], [75, 3, 1, "", "add_special_starter"], [75, 3, 1, "", "add_special_stopper"], [75, 4, 1, "", "assistant_formatter"], [75, 3, 1, "", "encode_conversation"], [75, 4, 1, "", "force_system"], [75, 4, 1, "", "function_formatter"], [75, 4, 1, "", "observation_formatter"], [75, 3, 1, "", "post_process_pairs"], [75, 4, 1, "", "remove_last_sep"], [75, 3, 1, "", "remove_last_separator"], [75, 4, 1, "", "separator"], [75, 4, 1, "", "special_starter"], [75, 4, 1, "", "special_stopper"], [75, 4, 1, "", "system_formatter"], [75, 4, 1, "", "template_name"], [75, 4, 1, "", "tools_formatter"], [75, 4, 1, "", "user_formatter"]], "lmflow.utils.conversation_template.base.ConversationTemplateForTool": [[75, 3, 1, "", "_encode"], [75, 3, 1, "", "_encode_template"], [75, 3, 1, "", "encode_conversation"]], "lmflow.utils.conversation_template.base.EmptyFormatter": [[75, 3, 1, "", "__post_init__"], [75, 3, 1, "", "format"]], "lmflow.utils.conversation_template.base.Formatter": [[75, 3, 1, "", "format"], [75, 3, 1, "", "has_placeholder"], [75, 4, 1, "", "template"]], "lmflow.utils.conversation_template.base.ListFormatter": [[75, 3, 1, "", "format"]], "lmflow.utils.conversation_template.base.StringFormatter": [[75, 3, 1, "", "__post_init__"], [75, 3, 1, "", "format"]], "lmflow.utils.conversation_template.base.TemplateComponent": [[75, 3, 1, "", "__post_init__"], [75, 3, 1, "", "__repr__"], [75, 3, 1, "", "__str__"], [75, 4, 1, "", "content"], [75, 4, 1, "", "mask"], [75, 4, 1, "", "type"]], "lmflow.utils.conversation_template.chatglm": [[76, 1, 1, "", "CHATGLM3_TEMPLATE"]], "lmflow.utils.conversation_template.chatml": [[77, 1, 1, "", "CHATML_TEMPLATE"]], "lmflow.utils.conversation_template.deepseek": [[78, 1, 1, "", "DEEPSEEK_TEMPLATE"]], "lmflow.utils.conversation_template.gemma": [[79, 1, 1, "", "GEMMA_TEMPLATE"], [79, 2, 1, "", "GemmaConversationTemplate"], [79, 1, 1, "", "logger"]], "lmflow.utils.conversation_template.gemma.GemmaConversationTemplate": [[79, 3, 1, "", "encode_conversation"]], "lmflow.utils.conversation_template.hymba": [[80, 1, 1, "", "HYMBA_TEMPLATE"], [80, 2, 1, "", "HymbaConversationTemplate"]], "lmflow.utils.conversation_template.hymba.HymbaConversationTemplate": [[80, 3, 1, "", "encode_conversation"]], "lmflow.utils.conversation_template.internlm": [[82, 1, 1, "", "INTERNLM2_TEMPLATE"]], "lmflow.utils.conversation_template.llama": [[83, 1, 1, "", "LLAMA2_TEMPLATE"], [83, 1, 1, "", "LLAMA2_TEMPLATE_FOR_TOOL"], [83, 1, 1, "", "LLAMA3_TEMPLATE"], [83, 1, 1, "", "LLAMA3_TEMPLATE_FOR_TOOL"], [83, 2, 1, "", "Llama2ConversationTemplate"], [83, 2, 1, "", "Llama2ConversationTemplateForTool"], [83, 1, 1, "", "logger"]], "lmflow.utils.conversation_template.llama.Llama2ConversationTemplate": [[83, 3, 1, "", "_encode"]], "lmflow.utils.conversation_template.llama.Llama2ConversationTemplateForTool": [[83, 3, 1, "", "_encode"]], "lmflow.utils.conversation_template.phi": [[84, 1, 1, "", "PHI3_TEMPLATE"]], "lmflow.utils.conversation_template.qwen": [[85, 1, 1, "", "QWEN2_TEMPLATE"], [85, 1, 1, "", "QWEN2_TEMPLATE_FOR_TOOL"]], "lmflow.utils.conversation_template.yi": [[86, 1, 1, "", "YI1_5_TEMPLATE"]], "lmflow.utils.conversation_template.zephyr": [[87, 1, 1, "", "ZEPHYR_TEMPLATE"], [87, 2, 1, "", "ZephyrConversationTemplate"], [87, 1, 1, "", "logger"]], "lmflow.utils.conversation_template.zephyr.ZephyrConversationTemplate": [[87, 3, 1, "", "_encode"]], "lmflow.utils.data_utils": [[88, 2, 1, "", "RewardModelInferenceResultWithInput"], [88, 2, 1, "", "VLLMInferenceResultWithInput"], [88, 5, 1, "", "answer_extraction"], [88, 5, 1, "", "batchlize"], [88, 5, 1, "", "load_data"], [88, 5, 1, "", "process_image_flag"], [88, 5, 1, "", "set_random_seed"]], "lmflow.utils.data_utils.RewardModelInferenceResultWithInput": [[88, 4, 1, "", "input"], [88, 4, 1, "", "output"]], "lmflow.utils.data_utils.VLLMInferenceResultWithInput": [[88, 4, 1, "", "input"], [88, 4, 1, "", "output"]], "lmflow.utils.flash_attention": [[89, 0, 0, "-", "bloom_flash_attention"], [90, 0, 0, "-", "gpt2_flash_attention"], [91, 0, 0, "-", "gpt_neo_flash_attention"], [93, 0, 0, "-", "llama_flash_attention"], [94, 0, 0, "-", "triton_flash_attention"]], "lmflow.utils.flash_attention.bloom_flash_attention": [[89, 5, 1, "", "_prepare_attn_mask"], [89, 5, 1, "", "forward"], [89, 5, 1, "", "replace_bloom_attn_with_flash_attn"]], "lmflow.utils.flash_attention.gpt2_flash_attention": [[90, 5, 1, "", "_prepare_decoder_attention_mask"], [90, 5, 1, "", "forward"], [90, 5, 1, "", "replace_gpt2_attn_with_flash_attn"]], "lmflow.utils.flash_attention.gpt_neo_flash_attention": [[91, 5, 1, "", "_attn"], [91, 5, 1, "", "forward"], [91, 5, 1, "", "replace_gpt_neo_attn_with_flash_attn"]], "lmflow.utils.flash_attention.llama_flash_attention": [[93, 5, 1, "", "_prepare_decoder_attention_mask"], [93, 5, 1, "", "forward"], [93, 5, 1, "", "replace_llama_attn_with_flash_attn"]], "lmflow.utils.flash_attention.triton_flash_attention": [[94, 2, 1, "", "FlashAttnFunc"], [94, 2, 1, "", "FlashAttnKVPackedFunc"], [94, 2, 1, "", "FlashAttnQKVPackedFunc"], [94, 5, 1, "", "_bwd_kernel"], [94, 5, 1, "", "_bwd_kernel_one_col_block"], [94, 5, 1, "", "_bwd_preprocess_do_o_dot"], [94, 5, 1, "", "_bwd_store_dk_dv"], [94, 5, 1, "", "_flash_attn_backward"], [94, 5, 1, "", "_flash_attn_forward"], [94, 5, 1, "", "_fwd_kernel"], [94, 1, 1, "", "flash_attn_func"], [94, 1, 1, "", "flash_attn_kvpacked_func"], [94, 1, 1, "", "flash_attn_qkvpacked_func"], [94, 5, 1, "", "init_to_zero"]], "lmflow.utils.flash_attention.triton_flash_attention.FlashAttnFunc": [[94, 3, 1, "", "backward"], [94, 3, 1, "", "forward"]], "lmflow.utils.flash_attention.triton_flash_attention.FlashAttnKVPackedFunc": [[94, 3, 1, "", "backward"], [94, 3, 1, "", "forward"]], "lmflow.utils.flash_attention.triton_flash_attention.FlashAttnQKVPackedFunc": [[94, 3, 1, "", "backward"], [94, 3, 1, "", "forward"]], "lmflow.utils.llava_conversation_lib": [[96, 2, 1, "", "Conversation"], [96, 2, 1, "", "SeparatorStyle"], [96, 1, 1, "", "conv_llama_2"], [96, 1, 1, "", "conv_llava_llama_2"], [96, 1, 1, "", "conv_llava_plain"], [96, 1, 1, "", "conv_llava_v0"], [96, 1, 1, "", "conv_llava_v0_mmtag"], [96, 1, 1, "", "conv_llava_v1"], [96, 1, 1, "", "conv_llava_v1_mmtag"], [96, 1, 1, "", "conv_mpt"], [96, 1, 1, "", "conv_templates"], [96, 1, 1, "", "conv_vicuna_v0"], [96, 1, 1, "", "conv_vicuna_v1"], [96, 1, 1, "", "default_conversation"]], "lmflow.utils.llava_conversation_lib.Conversation": [[96, 3, 1, "", "append_message"], [96, 3, 1, "", "copy"], [96, 3, 1, "", "dict"], [96, 3, 1, "", "get_images"], [96, 3, 1, "", "get_prompt"], [96, 4, 1, "", "messages"], [96, 4, 1, "", "offset"], [96, 4, 1, "", "roles"], [96, 4, 1, "", "sep"], [96, 4, 1, "", "sep2"], [96, 4, 1, "", "sep_style"], [96, 4, 1, "", "skip_next"], [96, 4, 1, "", "system"], [96, 3, 1, "", "to_gradio_chatbot"], [96, 4, 1, "", "version"]], "lmflow.utils.llava_conversation_lib.SeparatorStyle": [[96, 4, 1, "", "LLAMA_2"], [96, 4, 1, "", "MPT"], [96, 4, 1, "", "PLAIN"], [96, 4, 1, "", "SINGLE"], [96, 4, 1, "", "TWO"]], "lmflow.utils.model": [[97, 5, 1, "", "check_homogeneity"], [97, 1, 1, "", "logger"]], "lmflow.utils.multimodal": [[98, 5, 1, "", "adapt_llava_model_to_lmflow_type"], [98, 5, 1, "", "load_llava_pretrain_model"], [98, 5, 1, "", "update_custom_config"]], "lmflow.utils.position_interpolation": [[100, 0, 0, "-", "llama_rope_scaled_monkey_patch"]], "lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch": [[100, 2, 1, "", "CondenseRotaryEmbedding"], [100, 5, 1, "", "replace_llama_with_condense"]], "lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch.CondenseRotaryEmbedding": [[100, 3, 1, "", "forward"], [100, 4, 1, "", "max_seq_len_cached"], [100, 4, 1, "", "ntk_ratio"], [100, 4, 1, "", "pi_ratio"]], "lmflow.utils.versioning": [[101, 5, 1, "", "_is_package_available"], [101, 5, 1, "", "_is_packages_available"], [101, 5, 1, "", "get_python_version"], [101, 5, 1, "", "is_flash_attn_available"], [101, 5, 1, "", "is_flask_available"], [101, 5, 1, "", "is_gradio_available"], [101, 5, 1, "", "is_multimodal_available"], [101, 5, 1, "", "is_package_version_at_least"], [101, 5, 1, "", "is_ray_available"], [101, 5, 1, "", "is_trl_available"], [101, 5, 1, "", "is_vllm_available"], [101, 1, 1, "", "logger"]], "lmflow.version": [[102, 1, 1, "", "__version__"]]}, "objnames": {"0": ["py", "module", "Python module"], "1": ["py", "data", "Python data"], "2": ["py", "class", "Python class"], "3": ["py", "method", "Python method"], "4": ["py", "attribute", "Python attribute"], "5": ["py", "function", "Python function"], "6": ["py", "property", "Python property"]}, "objtypes": {"0": "py:module", "1": "py:data", "2": "py:class", "3": "py:method", "4": "py:attribute", "5": "py:function", "6": "py:property"}, "terms": {"": [4, 13, 16, 51, 66, 103, 105, 106, 108, 111, 112, 113, 114, 115], "0": [2, 4, 5, 6, 8, 13, 22, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45, 51, 52, 55, 57, 60, 61, 66, 94, 102, 103, 107, 112, 113, 114, 115], "000": 103, "0001": 44, "0005": 4, "001": [25, 26, 28, 29, 31, 33, 36, 41, 43, 45], "002": [30, 38], "0025": 32, "004": 38, "00962": 36, "01": [37, 39, 45], "02155": [112, 113], "03265": 41, "04": 44, "05": 4, "0501": 103, "06": [27, 36, 45], "06677": 33, "08": [26, 28, 29, 30, 31, 32, 33, 37, 38, 39, 41, 43], "08217": [31, 43], "09843": 26, "1": [2, 3, 4, 5, 6, 13, 22, 26, 27, 31, 37, 42, 43, 44, 51, 52, 55, 61, 66, 94, 103, 105, 109, 110, 111, 115], "10": [4, 36, 103, 112, 113], "100": [4, 55, 60, 61, 66, 74, 94, 113], "1000": 4, "10000": [100, 112, 113], "10000000000": 4, "101": [13, 103], "1010": 13, "102": 13, "1024": [4, 55, 112], "105": 103, "106": 103, "109": 103, "10k": 113, "112k": [112, 113], "113": 103, "119": 103, "12": [112, 113], "120b": 115, "121": 103, "123": 103, "124": 103, "125": [103, 112], "128": [4, 94, 113], "129": 103, "12k": 113, "130": 103, "134": [74, 103], "135": 103, "139": 103, "13b": [103, 113], "140": 103, "141": 103, "146": 103, "147": 103, "149": 103, "15": 74, "150": 103, "151": 103, "153": 103, "155": 103, "16": [25, 57, 94, 113], "160": 103, "163": 103, "164": 103, "165": 103, "1659": 112, "167": 103, "1692": 112, "170": 103, "17192": 55, "17192v2": 55, "172": 103, "173": 103, "175": 103, "175b": 115, "176b": 115, "18": 115, "180": 103, "181": 103, "183": 103, "184": 103, "188": 103, "1902": 26, "1904": 36, "1908": [15, 41], "1935": 112, "1938": 112, "198": 103, "1b": 103, "1e": [25, 26, 27, 28, 29, 30, 31, 32, 33, 36, 37, 38, 39, 41, 43, 45, 55], "2": [4, 5, 6, 32, 42, 55, 57, 75, 80, 81, 94, 103, 105, 109, 110, 111, 115], "20": [4, 66, 103, 109, 112], "200": [74, 103], "2005": 112, "2006": [31, 43], "2020": [25, 112], "2023": [2, 103, 115], "2048": [4, 100], "206": 103, "207": 103, "2088": 13, "21": 115, "211": 103, "213": 103, "214": 103, "215": 103, "219": 103, "22": 112, "220": 103, "2203": [112, 113], "2208": 33, "2211": 55, "222": 103, "224": 103, "228": 103, "23": 115, "237": 103, "24": [50, 112, 113, 115], "240": 103, "245": 103, "25": 115, "254": 103, "256": 112, "258": 103, "25mb": 115, "26": 115, "262": 103, "266": 103, "27": 115, "28": [2, 115], "280b": 115, "29": 115, "2e": 112, "2k": 112, "3": [94, 103, 105, 106, 109, 110, 115], "30": [74, 103, 112, 115], "30b": 115, "32": [4, 94, 113, 115], "33": 103, "33b": 115, "34": 103, "35": [103, 115], "36": 115, "37": [103, 115], "38": 103, "39": [103, 112, 115], "3b": [103, 109, 112, 113], "3e": 112, "4": [4, 69, 103, 105, 112, 115], "40": [94, 103, 115], "400": 112, "41": 103, "42": [4, 5, 6, 103], "43": 115, "44": [103, 115], "447": 103, "46": 115, "48": [57, 94], "49": 115, "4bit": 4, "4rtemi5": 45, "5": [4, 55, 61, 103, 109, 112, 113, 115], "50": 115, "51": 115, "512": [4, 113], "5120": 44, "52": [103, 112, 113], "53": 103, "54": 115, "55": [103, 113, 115], "56": [103, 115], "57": [103, 115], "58": [103, 112, 113, 115], "59": [103, 115], "5k": [112, 113], "6": [37, 55, 103, 112, 115], "60": [103, 115], "600mb": 103, "61": [103, 115], "62": 103, "63": [103, 115], "64": [4, 94, 103, 112, 113, 115], "65": [103, 112, 113, 115], "66": [103, 115], "67": [103, 115], "68": [103, 115], "69": [103, 112, 113, 115], "7": [103, 112, 115], "70": [103, 115], "71": [103, 112, 113], "72": [103, 115], "73": [103, 115], "74": [103, 115], "75": [103, 115], "7592": 13, "76": [36, 103], "767": 103, "77": 103, "78": [103, 115], "79": [103, 112, 113], "7b": [103, 105, 107, 109, 112, 113, 115], "7oemwynu": 113, "8": [4, 57, 103, 112, 115], "80": [94, 113], "800": 112, "81": [103, 113], "8186": 45, "82": 103, "82147": 112, "83": 103, "84": 113, "85": [103, 115], "86": 103, "87": 115, "88": [94, 103], "888888888": 57, "89": 103, "8b": 109, "8fc1rcf8": 113, "9": [4, 8, 25, 26, 29, 30, 31, 32, 34, 36, 38, 39, 41, 42, 45, 102, 103, 106, 115], "90": 115, "900mb": 103, "92": 33, "95": [4, 27, 103], "96": [94, 103], "965": 44, "97": 103, "98": 33, "99": [33, 44], "999": [4, 25, 26, 29, 30, 31, 32, 34, 36, 38, 39, 41, 45], "9b": 103, "A": [1, 5, 6, 11, 12, 13, 14, 16, 21, 22, 25, 26, 31, 32, 34, 36, 37, 41, 42, 43, 44, 45, 47, 49, 52, 57, 61, 66, 73, 88, 96, 103, 105, 112, 114], "And": 7, "As": [32, 42, 103, 105, 112, 113, 114], "At": 60, "But": 112, "By": [4, 66, 107, 112, 115], "For": [4, 66, 71, 75, 94, 103, 105, 108, 109, 110, 112, 113, 114, 115], "If": [4, 61, 66, 88, 94, 103, 105, 106, 111, 112, 113, 115], "In": [66, 103, 105, 106, 107, 112, 113], "It": [4, 13, 14, 26, 31, 36, 41, 43, 45, 52, 66, 103, 105, 108, 112, 115], "Its": [103, 105, 115], "NOT": 55, "No": 103, "ONE": 108, "Of": [13, 16], "On": [41, 103], "One": [66, 103], "Or": [112, 113], "That": 112, "The": [1, 4, 5, 6, 7, 11, 12, 13, 14, 16, 22, 32, 34, 37, 42, 46, 50, 52, 53, 55, 57, 61, 66, 73, 75, 88, 94, 103, 105, 112, 113, 114, 115], "Then": [52, 107, 112], "There": [103, 112, 113], "These": [103, 108], "To": [32, 42, 66, 105, 108, 112, 113, 115], "Will": [66, 69], "With": [107, 114], "_": 65, "__call__": [7, 60, 67], "__distributed_infer": 58, "__filter_arg": 56, "__getitem__": 7, "__infer": [13, 16, 58], "__init__": [13, 14, 108], "__len__": [5, 6, 7, 66], "__model_module_inject": 15, "__post_init__": [4, 75, 81], "__post_process_model_output": 58, "__prepare_dtyp": 15, "__prepare_inputs_for_infer": 13, "__prepare_inputs_for_vllm_infer": 13, "__prepare_model_config": 15, "__prepare_model_for_infer": 15, "__prepare_model_for_train": 15, "__prepare_model_for_vllm_infer": 15, "__prepare_model_post_process": 15, "__prepare_peft_config": 15, "__prepare_quant_config": 15, "__prepare_token": 15, "__prepare_training_arg": 51, "__repr__": 75, "__setstate__": [25, 26, 30, 33, 37, 38, 39, 41, 44], "__str__": 75, "__version__": [8, 102], "__vllm_infer": [13, 16, 58], "_activ": 15, "_add_sm_patterns_to_gitignor": 66, "_align_single_iter": 56, "_attn": 91, "_bwd_kernel": 94, "_bwd_kernel_one_col_block": 94, "_bwd_preprocess_do_o_dot": 94, "_bwd_store_dk_dv": 94, "_calc_response_length": 51, "_calc_reward_with_length_penalti": 51, "_channel_view": [31, 43], "_check_data_format": [5, 6], "_check_if_align": 47, "_check_if_tun": 49, "_clean_text": [57, 112], "_cosine_similar": [31, 43], "_discard_sampl": [57, 112], "_distributed_infer": 69, "_do_reward_model_infer": 56, "_do_single_dpo_align": 56, "_do_target_model_infer": 56, "_encod": [75, 81, 83, 87], "_encode_templ": [75, 81], "_ensure_id_list": [75, 81], "_evaluate_acc_with_acceler": 52, "_evaluate_acc_with_deepspe": 52, "_evaluate_nl": 52, "_evaluate_ppl": 52, "_flash_attn_backward": 94, "_flash_attn_forward": 94, "_foreach_mul_": [32, 42], "_fwd_kernel": 94, "_gather_and_numpifi": 66, "_get_batch_dataset_loc": 57, "_get_batch_dataset_top": 57, "_get_collator_with_removed_column": 66, "_get_eval_sampl": 66, "_get_output_dir": 66, "_get_train_sampl": 66, "_hp_search_setup": 66, "_infer": [58, 69], "_initialize_train": [50, 57], "_inner_training_loop": 66, "_internal_cal": 66, "_is_native_cpu_amp_avail": 66, "_is_package_avail": 101, "_is_packages_avail": 101, "_issue_warnings_after_load": 66, "_layer_view": [31, 43], "_load_best_model": 66, "_load_dataset": [50, 57], "_load_from_checkpoint": 66, "_load_input_dataset": 57, "_load_optimizer_and_schedul": 66, "_load_rng_stat": 66, "_loggers_initi": 66, "_match": 52, "_maybe_log_save_evalu": 66, "_memory_track": 66, "_move_model_to_devic": 66, "_multi_tensor_adan": 33, "_nested_gath": 66, "_one_train": 66, "_pad_across_process": 66, "_parse_dpo_aligner_arg": 56, "_parse_reward_model_inference_arg": 56, "_parse_target_model_inference_arg": 56, "_prepare_attn_mask": 89, "_prepare_decoder_attention_mask": [90, 93], "_prepare_input": 66, "_project": [31, 43], "_provided_": 61, "_push_from_checkpoint": 66, "_remove_unused_column": 66, "_report_to_hp_search": 66, "_rotate_checkpoint": 66, "_sampling_paired_idx_from_reward": 51, "_sampling_paired_idx_from_rewards_fast": 51, "_save": [65, 66], "_save_checkpoint": [65, 66], "_save_tpu": 66, "_set_signature_columns_if_need": 66, "_signature_column": 66, "_single_tensor_adan": 33, "_sorted_checkpoint": 66, "_train_batch_s": 66, "_tune_save_checkpoint": 66, "_wrap_model": 66, "a1": 103, "a100": 94, "a2": 103, "a3": 103, "ab": [26, 31, 33, 36, 41, 43, 55, 112, 113], "abc": [10, 11, 12, 15, 19, 20, 22, 48, 69, 75], "abil": [103, 115], "abl": [22, 66, 103, 112], "about": [4, 66, 103, 112, 113, 115], "abov": [103, 105, 108], "abspath": 111, "abstract": [13, 14, 16, 47, 49, 55, 58, 69, 75, 108], "accelerate_config_fil": 4, "accept": [14, 15, 21, 52, 66, 105, 112], "access": [4, 66, 103, 105, 107, 115], "accid": 112, "accomplish": 112, "accord": [37, 66, 103, 112], "account": 115, "accuraci": [4, 52, 103, 112, 113, 115], "achiev": [103, 112, 115], "acquir": [103, 115], "across": [75, 94], "activ": [106, 112, 115], "activate_model_for_infer": 15, "actual": [75, 103, 112], "ad": [109, 112, 113], "adabelief": [3, 4, 35], "adabound": [3, 4, 35], "adadelta": [3, 4, 35], "adagrad": [3, 4, 35], "adam": [3, 4, 25, 35, 36, 103], "adamax": [3, 4, 35], "adamp": [3, 4, 35, 43], "adamw": [32, 66, 110], "adamw_schedule_fre": [3, 4, 35], "adamwschedulefre": 32, "adan": [3, 4, 35], "adapt": [4, 13, 14, 25, 26, 33, 41, 45, 65, 66, 110, 112, 115], "adapt_llava_model_to_lmflow_typ": 98, "add": [7, 32, 42, 66, 73, 105, 106, 108, 112, 114], "add_callback": 66, "add_dataclass_attr_prefix": 73, "add_generation_prompt": 114, "add_special_start": [75, 81], "add_special_stopp": [75, 81], "add_special_token": [13, 16, 70, 71], "addit": [4, 7, 66], "addition": [103, 112, 113], "additional_stop_token_id": 4, "adjust": [55, 112, 113], "administ": 103, "administr": 112, "adobada": 103, "adopt": 112, "advanc": 112, "advantag": 103, "advers": 103, "advic": 115, "affect": 113, "afflict": 103, "afford": [103, 112], "after": [4, 7, 13, 16, 55, 69, 75, 103, 106, 108, 112, 114], "ag": [105, 108, 114], "again": 66, "ago": 113, "ai": [103, 105, 112, 113, 115], "aim": 115, "al": 103, "algorithm": [25, 26, 31, 33, 36, 41, 43, 45, 103, 109, 110], "alibi": [89, 94], "align": [4, 37, 47, 50, 51, 56, 57, 105, 110], "aligner_arg": [50, 51, 56, 57], "aligner_file_path": 51, "all": [15, 40, 61, 66, 94, 96, 103, 105, 107, 108, 109, 112, 113, 115], "allevi": 112, "allow": [13, 14, 66, 103, 109, 115], "almost": 108, "along": 66, "alpaca": [103, 105, 107, 109], "alpha": 57, "alreadi": [4, 66, 75, 103, 105, 112, 113], "alright": 113, "also": [4, 11, 12, 32, 42, 52, 60, 66, 103, 106, 108, 112, 113, 114, 115], "altern": [112, 114], "although": [103, 112, 113], "alwai": [66, 103, 112, 114], "am": [103, 105, 108, 112, 114], "america": [112, 113], "american": 103, "among": 115, "amount": [112, 113], "amsbound": 26, "amsgrad": [25, 39], "an": [4, 10, 11, 12, 15, 19, 20, 22, 34, 48, 66, 69, 75, 94, 104, 105, 108, 109, 112, 113, 115], "analog": 37, "anatomi": 115, "ani": [7, 13, 16, 60, 66, 67, 103, 105, 112, 113, 114, 115], "anim": 113, "announc": 115, "anoth": [66, 103, 112], "another_data": 105, "answer": [13, 14, 50, 88, 103, 105, 106, 112, 113, 115], "answer_extract": 88, "answer_neg": 113, "answer_posit": 113, "answer_typ": [4, 52, 88, 106, 107], "anthropologi": 112, "anyon": 115, "anyth": 112, "api": [1, 103, 115], "api_doc": 66, "app": 66, "appear": 113, "append": 75, "append_messag": 96, "appl": 103, "appli": [4, 69, 75, 103, 105, 112, 115], "applic": [37, 66, 103, 115], "apply_chat_templ": [4, 13, 69], "appreci": 112, "approach": 115, "appropri": [66, 103, 112, 113], "approxim": 55, "ar": [13, 14, 16, 50, 60, 66, 73, 75, 94, 103, 105, 106, 107, 108, 112, 113, 114, 115], "arbitrari": 94, "arc_": 103, "arc_c": 103, "arc_easi": 103, "arch_typ": 4, "architectur": [4, 66], "area": [103, 113, 115], "aren": 112, "arena": 103, "arg": [3, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 20, 21, 22, 23, 46, 47, 49, 51, 53, 55, 56, 57, 58, 59, 61, 63, 65, 66, 69, 70, 71, 79, 88, 96, 97, 111], "argmax": 55, "argument": [4, 5, 6, 13, 14, 15, 16, 21, 25, 26, 31, 32, 34, 36, 37, 41, 42, 43, 45, 52, 53, 55, 57, 58, 59, 66, 73, 75, 88, 105, 111], "argv": 111, "around": [13, 14, 103], "arr": 103, "arrai": [66, 103], "arriv": 103, "articl": 103, "arxiv": [26, 31, 33, 36, 41, 43, 55, 112, 113], "asada": 103, "ask": [108, 112], "aspect": 103, "assign": [4, 61, 112], "assist": [75, 80, 81, 103, 105, 106, 108, 112, 113, 114], "assistant_formatt": [75, 81, 108], "assistant_reply_0": 114, "assistant_reply_1": 114, "assistant_response_1": 105, "assistant_response_1_bad": 105, "assistant_response_1_good": 105, "assistant_response_2": 105, "assistant_response_2_bad": 105, "assistant_response_2_good": 105, "associ": 115, "assum": [113, 115], "at_init": 66, "atmospher": 112, "atomic_add": 94, "attack": 112, "attain": 115, "attempt": 103, "attend": 103, "attent": [13, 14, 22, 94], "attention_mask": [13, 22, 23, 89, 90, 91, 93, 113], "attn": 71, "auk": 113, "authent": 103, "author": [103, 115], "auto": [3, 4, 15], "auto_model": [3, 17], "auto_pipelin": [3, 54, 111], "autoapi": 3, "autoargu": [4, 111], "autocast": 66, "autocast_smart_context_manag": 66, "autograd": 94, "automat": [4, 9, 37, 46, 66, 104, 112], "automodel": 9, "automodelforsequenceclassif": 15, "autopipelin": [46, 111], "autoregressive_sampl": 55, "autotoken": 67, "avail": [66, 103, 105, 106, 115], "averag": [103, 115], "avoid": [22, 103, 112], "awar": [103, 115], "b": [44, 88, 106], "back": [4, 108, 112], "backbon": 115, "backend": [5, 6, 13, 14, 15, 16, 66, 94], "backend_dataset": [5, 6], "backward": [66, 94], "bad": 103, "bai": 103, "bake": [103, 112], "bar": [11, 12], "barbecu": 103, "barrel": 103, "base": [3, 4, 7, 10, 11, 12, 13, 14, 15, 16, 19, 20, 21, 22, 23, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 55, 57, 58, 59, 61, 65, 66, 68, 69, 79, 80, 81, 83, 87, 88, 94, 96, 100, 103, 108, 109, 112, 113, 115], "base_align": [3, 50, 51, 54, 57], "base_lr": 26, "base_model": [3, 11, 12, 15, 17, 20, 22], "base_pipelin": [3, 47, 49, 52, 54, 55, 58, 69], "base_trainer_class": 53, "base_tun": [3, 53, 54], "basealign": [47, 50, 51, 57], "basemodel": [10, 11, 12, 15, 20, 22, 57], "basepipelin": [47, 48, 49, 52, 55, 58, 69], "basetun": [47, 49, 53], "basic": [66, 113], "batch": [13, 22, 36, 37, 52, 60, 61, 66, 88, 94, 112, 113], "batch_input": 57, "batch_siz": [22, 58, 61, 66, 69, 88, 94, 112], "batchliz": [55, 88], "battl": 103, "bbq": 103, "beam": 4, "bean": 103, "beauti": 103, "becaus": [66, 103, 112], "been": [26, 31, 36, 41, 43, 45, 66, 94, 103, 112], "befor": [32, 42, 66, 105, 108, 112, 114, 115], "begin": [13, 16, 32, 37, 42, 105, 108, 112, 114], "begin_of_text": [105, 108, 114], "beginn": [108, 115], "behavior": 66, "behind": 103, "being": [66, 75, 80, 81, 103, 112, 113], "belief": [25, 112], "believ": 115, "below": [15, 105, 106, 108, 115], "ben": 103, "benchamrk": 106, "benchmark": [104, 110], "benchmarkingargu": 4, "benefit": 115, "bert": 36, "besid": 103, "best": [51, 65, 66, 103, 112, 113], "bestrun": 66, "beta": [4, 25, 26, 29, 30, 31, 32, 33, 34, 36, 37, 38, 39, 41, 44, 45, 61], "beta1": 33, "beta2": 33, "beta3": 33, "better": [103, 109, 112, 115], "between": [75, 94, 103, 108, 109], "beyond": 41, "bf16": 4, "bg677mxa": 113, "bia": [4, 94], "bias": 52, "bias_correction1": 33, "bias_correction2": 33, "bias_correction3_sqrt": 33, "bias_typ": 94, "bigger": 112, "bin": 66, "biologi": [103, 115], "bit": [4, 94, 112, 113], "blank": [11, 12], "blend": 103, "bleu": 66, "blip2config": 22, "blip2forconditionalgener": 22, "block": [4, 53, 66, 70, 71], "block_headdim": 94, "block_m": 94, "block_n": 94, "block_siz": [4, 52, 70, 71], "blocking_pair": 71, "blocking_text_to_textlist": 71, "blog": 115, "bloom": [103, 115], "bloom_flash_attent": [3, 92], "bo": [75, 105, 108, 114], "bold": 103, "bolt": 37, "book": 112, "bool": [4, 5, 6, 7, 13, 15, 16, 22, 26, 31, 33, 36, 37, 43, 44, 50, 51, 55, 58, 60, 61, 66, 67, 69, 73, 75, 81, 88, 89, 90, 93, 96, 97, 101], "boolean": 4, "boolq": 103, "booltensor": 89, "bos_token": [75, 108, 114], "both": [66, 94, 103, 112, 115], "bound": 26, "bowl": 103, "branch": [4, 106], "brand": 112, "break": [103, 113], "breakthrough": 103, "brief": [11, 12], "bring": [103, 112], "brisket": 103, "broadcast": 94, "brought": 103, "brown": 112, "buffer": 4, "bug": 112, "bui": 112, "build": [103, 112, 113, 115], "build_dataset": 113, "build_vision_tow": [23, 24], "built": 112, "burden": 112, "burrito": 103, "butter": 112, "byol": 37, "c": [88, 103], "cach": [4, 5, 6, 66], "cache_dir": [4, 50], "cache_en": 66, "cache_key_seqlen_k": 94, "cache_key_seqlen_q": 94, "cake": 103, "calam": 103, "call": [13, 14, 32, 42, 66, 103, 108, 112], "call_model_init": 66, "callabl": [25, 32, 34, 42, 61, 66], "callback": [61, 66], "callback_handl": 66, "calm": 103, "can": [4, 13, 14, 32, 42, 66, 73, 75, 103, 105, 106, 107, 108, 112, 113, 115], "can_return_loss": 66, "cannot": [15, 103], "cant": [112, 113], "capabl": 112, "capac": 115, "capit": 103, "capitol": 103, "caption": 22, "captur": 44, "card": 66, "care": 112, "carefulli": [103, 112], "caribbean": 112, "carn": 103, "carnita": 103, "carri": [103, 112], "carrier": 103, "casa": 103, "case": [60, 65, 66, 94, 103, 107, 112], "casual": 103, "cat": 112, "categori": 112, "cater": 115, "caus": [103, 112], "causal": 94, "causallmoutputwithpast": 22, "caution": 94, "cc": 45, "cd": [105, 106, 107, 109, 112], "cell": 112, "centuri": 112, "certain": [103, 112, 113], "cevich": 103, "chain": 103, "challeng": [103, 112, 113], "chang": [94, 106, 112], "changelog": [2, 115], "charact": 112, "chat": [4, 69, 105], "chatbot": [103, 105, 108, 114], "chatbot_typ": 55, "chatglm": [3, 81], "chatglm3": 105, "chatglm3_templ": [76, 81], "chatgpt": [103, 108, 115], "chatml": [3, 81, 105], "chatml_templ": [77, 81], "cheap": 103, "check": [5, 6, 66, 103, 105], "check_homogen": 97, "checkout": [106, 112], "checkpoint": [4, 32, 42, 65, 66, 103, 110, 112], "checkpoint_fold": 66, "checkpoint_path": 98, "checkpoint_prefix": 66, "chees": 103, "chemistri": 103, "chesapeak": 103, "chess": 103, "chessboard": 103, "chicken": 103, "child": 112, "chile": 103, "chili": 103, "chines": 103, "chip": 112, "chocol": [103, 112], "choic": [4, 88, 103, 112, 113], "choos": [4, 66, 103, 112], "chop": 112, "chosen": [50, 60, 61, 103, 105, 106, 112, 113], "chosen_attention_mask": 113, "chosen_input_id": 113, "chosen_reward": [61, 113], "circumst": [103, 112], "citi": 103, "civil": 112, "clamp_valu": 36, "class": [73, 108], "class_prefix": 73, "classfoo": [11, 12], "classic": 103, "classif": [13, 14, 66], "classmethod": [5, 6, 9, 46], "clean_text": 112, "clear": 112, "clearer": 113, "clearli": 112, "clinic": 115, "clip_encod": [3, 24], "clip_global_grad_norm": 33, "clipvisiontow": 23, "clone": [106, 115], "closur": [25, 26, 27, 28, 29, 30, 31, 32, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45], "clovaai": [31, 43], "co": [4, 112], "coal": 103, "coast": [103, 112], "cob": 103, "code": [5, 6, 7, 23, 26, 31, 36, 37, 41, 43, 44, 45, 55, 66, 103, 105, 114, 115], "code_exec": 55, "coleslaw": 103, "collard": 103, "collat": [7, 60, 66], "collect": [103, 112, 115], "collection_strategi": [4, 112], "colleg": 115, "coloni": 112, "columbia": 103, "column": 66, "column_nam": [70, 71], "com": [66, 103, 106, 112, 115], "comal": 103, "combin": 106, "come": [66, 103], "command": [106, 109, 112, 115], "comment": 108, "commerci": [112, 115], "commit": [4, 66], "commit_messag": 66, "common": [3, 95, 103, 105, 115], "commonli": [15, 74, 103, 105, 115], "commun": [103, 112, 115], "compani": 103, "compar": [103, 112, 113, 115], "comparison": [103, 112], "compet": [103, 115], "competit": 103, "compil": 94, "complet": [66, 112, 114, 115], "complex": [103, 113], "complic": 113, "compliment": 112, "compon": [75, 81, 108, 115], "comprehens": 103, "compress_list": 58, "comput": [61, 66], "compute_loss": [66, 68], "compute_loss_context_manag": 66, "compute_metr": [61, 66, 68], "compute_object": 66, "concaten": [66, 103, 105], "concept": [105, 108, 114], "concis": 113, "conda": [106, 115], "condens": 15, "condenserotaryembed": 100, "condit": [22, 94, 103, 105, 112], "confer": 103, "confid": 94, "config": [4, 15, 22, 23, 52, 55, 98, 111, 113], "config_nam": 4, "config_overrid": 4, "configu": [13, 14, 16], "configur": [4, 15, 112], "conflict": 103, "confus": 105, "conjectur": 103, "connot": 103, "consequ": 103, "consequenti": 115, "consist": [37, 112, 113], "consol": 52, "constant": [3, 95], "constexpr": 94, "constraint": 115, "constructor": 52, "contain": [3, 4, 5, 6, 11, 12, 13, 22, 52, 53, 55, 57, 58, 59, 61, 66, 69, 75, 103, 105, 112], "content": [105, 108, 113, 114], "context": [55, 66, 103, 112, 114], "context_window": 52, "contribut": 112, "contributor": [2, 115], "control": [4, 5, 6, 55, 65, 66], "controller_heart_beat_expir": 74, "conv_llama_2": 96, "conv_llava_llama_2": 96, "conv_llava_plain": 96, "conv_llava_v0": 96, "conv_llava_v0_mmtag": 96, "conv_llava_v1": 96, "conv_llava_v1_mmtag": 96, "conv_mpt": 96, "conv_templ": 96, "conv_vicuna_v0": 96, "conv_vicuna_v1": 96, "conveni": [13, 14, 75, 103, 112, 115], "convent": 115, "convers": [4, 5, 70, 71, 96, 103, 110, 112, 113, 115], "conversation_dataset_descript": 74, "conversation_id": [75, 80, 81, 105], "conversation_role_nam": 74, "conversation_templ": [3, 4, 70, 71, 95, 105, 108, 109], "conversation_tokenize_funct": [70, 71], "conversationtempl": [70, 71, 75, 79, 81, 83, 87, 108, 114], "conversationtemplatefortool": [75, 80, 81], "convert": [50, 51, 55, 60, 66, 75, 81, 88, 107], "convert_llama_weights_to_hf": 107, "convert_to_paired_dataset": 51, "convert_tokens_to_id": 75, "convienc": 105, "convolut": 37, "cook": [103, 112], "cooki": 112, "copi": [23, 73, 96, 112], "copyright": 107, "core": 66, "corn": 103, "corp": 112, "corpora": 103, "corpu": 103, "correct": [9, 105, 112, 113], "correctli": [65, 103, 112], "correl": 103, "correspond": [13, 69, 103, 105], "cosin": 4, "could": [103, 112], "count": [66, 103], "countri": 103, "court": 103, "cozi": 103, "crab": 103, "creat": [3, 5, 6, 10, 11, 12, 15, 19, 20, 22, 48, 52, 60, 65, 66, 69, 73, 75, 108, 110, 112, 113, 115], "create_copied_dataclass": 73, "create_customized_optim": 53, "create_dataload": [52, 55], "create_from_dict": [5, 6], "create_model_card": 66, "create_optim": 66, "create_optimizer_and_schedul": 66, "create_schedul": 66, "create_studi": 66, "creation": 112, "creator": 112, "crispi": 103, "criteria": 112, "critic": 103, "cross": 94, "cross_entropi": 61, "crowd": 103, "crucial": [113, 115], "ctx": 94, "cuda": [88, 94], "cuda_visible_devic": 107, "cue": 115, "cuisin": 103, "culinari": 103, "cultur": [103, 112], "cumul": 55, "cup": 112, "current": [4, 15, 66, 103, 105, 106, 109, 114], "current_flo": 66, "curv": 112, "custom": [4, 15, 66, 112, 113], "custom_model": [4, 14], "custom_vision_model": [4, 22], "customautovision2seqmodel": 22, "customized_cache_dir": 4, "customized_optim": 4, "customized_optim_arg": 4, "custommultimodaldataset": 7, "cybertronai": 36, "d": [88, 94, 103, 112, 113], "dahoa": [112, 113], "dai": 112, "damag": [103, 112, 115], "dampen": [37, 43], "danger": 103, "dark": 103, "data": [4, 5, 6, 7, 13, 16, 21, 52, 53, 57, 58, 66, 69, 75, 80, 81, 88, 103, 106, 107, 109, 112, 113, 115], "data_1": 105, "data_2": 105, "data_arg": [5, 6, 7, 46, 50, 51, 52, 53, 55, 56, 57, 58, 59, 69, 70, 71, 111], "data_col": [53, 59, 61, 66], "data_collect": 112, "data_dict": 7, "data_dir": 50, "data_fil": 113, "data_inst": 73, "data_root": 50, "data_util": [3, 13, 16, 58, 69, 95], "dataclass": [4, 73], "dataclass_object": 73, "datacol": [61, 66], "datacollatorforsuperviseddataset": 7, "datacollatorwithpad": 66, "dataload": [55, 57, 66, 88], "dataset": [1, 3, 4, 8, 13, 14, 16, 21, 47, 49, 50, 51, 52, 53, 55, 56, 57, 58, 59, 61, 66, 69, 70, 71, 75, 88, 103, 108, 109, 110, 111, 113, 115], "dataset_arg": 66, "dataset_config_nam": 4, "dataset_description_map": 74, "dataset_list": 56, "dataset_nam": [4, 106], "dataset_path": [4, 5, 6, 7, 105, 107, 108, 109, 112, 113], "dataset_path_list": 4, "dataset_s": 55, "dataset_tag": 66, "dataset_typ": [5, 51], "datasetargu": [4, 5, 6, 7, 51, 52, 53, 55, 56, 57, 58, 59, 69, 70, 71, 111], "date": 112, "ddp": 4, "de": 103, "deactiv": 15, "deactivate_model_for_infer": 15, "deadli": 103, "deal": 94, "death": 112, "debia": 36, "debug": 4, "decai": 37, "decid": [103, 106], "declin": 112, "decod": [13, 14, 52, 69, 88, 105], "decoder_model": [3, 13, 17], "decoder_onli": 4, "decodermodel": [11, 13], "decor": 4, "decreas": 113, "dedic": 112, "deep": [33, 36, 112], "deepseek": [3, 81, 105], "deepseek_templ": [78, 81], "deepspe": [4, 13, 14, 16, 66, 105, 106, 107, 111], "deepspeed_arg": 105, "def": [103, 111, 112, 113], "default": [4, 5, 6, 13, 14, 16, 34, 55, 66, 69, 73, 88, 112], "default_callback": 66, "default_compute_object": 66, "default_convers": 96, "default_data_col": 66, "default_hp_space_optuna": 66, "default_hp_space_rai": 66, "default_hp_space_sigopt": 66, "default_im_end_token": 74, "default_im_start_token": 74, "default_image_patch_token": 74, "default_image_token": 74, "default_progress_callback": 66, "defin": [1, 4, 5, 6, 7, 34, 37, 66, 105, 108, 114], "definit": 112, "degenerated_to_sgd": 25, "degrad": 103, "delay_load": 23, "delici": 103, "delta": [31, 43, 94], "demo": 103, "demo_example_in_prompt": 4, "democrat": 115, "demonstr": [103, 112, 115], "denot": 37, "depart": [103, 112], "depend": [4, 66, 103, 112], "deprec": [41, 66], "deriv": 4, "desc": 111, "describ": 103, "descript": [11, 12, 66, 105, 110, 114], "design": [46, 115], "desir": [66, 112, 113], "despit": 115, "detail": [66, 103, 106, 110, 112, 113, 114], "detect": 112, "determin": [66, 103], "detoken": [4, 13], "dev20221202": 94, "develop": [1, 103, 105, 108, 112, 114], "devic": [4, 13, 14, 15, 16, 23, 66, 100, 112], "diao": [0, 115], "dict": [4, 5, 6, 13, 15, 51, 55, 60, 61, 66, 67, 70, 71, 73, 75, 80, 81, 83, 87, 88, 96, 105, 108], "dict_obj": [5, 6], "dictionari": [5, 6, 34, 50, 52, 53, 66], "did": [112, 113], "didn": [112, 113], "diff": 107, "differ": [4, 5, 6, 13, 14, 51, 66, 75, 94, 96, 103, 107, 108, 112, 113, 114], "difficult": 103, "difficulti": 103, "dim": 100, "dimens": 94, "dinosaur": [112, 113], "diplomaci": 103, "diplomat": 112, "dir": [13, 14, 16, 106], "direct": [66, 105, 115], "directli": [75, 103, 106, 107, 115], "director": 112, "directori": [4, 13, 14, 16, 52, 66, 105, 112], "disabl": 112, "disable_dropout": 61, "disable_group_text": 4, "disast": [103, 112], "discard": 112, "discard_sampl": 112, "diseas": 103, "dish": [103, 112], "disk": 4, "distinct": 112, "distress": 103, "distribut": [4, 55, 66], "distributed_inference_num_inst": 4, "distributeddataparallel": 66, "district": 103, "divers": [4, 103, 112], "divis": 112, "dk": 94, "dk_ptr": 94, "do": [15, 66, 75, 94, 103, 105, 108, 112, 113, 114, 115], "do_dpo_align": 4, "do_grad_sc": 66, "do_response_gener": 4, "do_rope_sc": [4, 15], "do_sampl": 4, "do_scor": 4, "do_train": 15, "doc": 66, "docstr": [11, 12, 108], "document": [3, 66, 103], "doe": [34, 66, 103, 112, 113, 114, 115], "doesn": [15, 94, 103, 112], "dog": 103, "domain": [103, 115], "don": [7, 60, 65, 71, 105, 108, 112, 113, 114], "done": [94, 111], "dong": [0, 115], "donut": 103, "door": 103, "double_qu": 4, "down": [31, 43, 103], "download": [4, 103, 105, 106, 107, 109, 112], "dpo": [4, 51, 61, 105], "dpo_align": [3, 54], "dpo_loss": 61, "dpoalign": 50, "dpoalignerargu": 4, "dpotrain": 61, "dpov2": 4, "dpov2_align": [3, 54], "dpov2_dataprocessor": [3, 62], "dpov2_train": [3, 62], "dpov2align": 51, "dpov2alignerargu": [4, 51, 56], "dpov2train": 61, "dq": 94, "draft": [55, 66], "draft_config": 55, "draft_model": 55, "draft_model_arg": 55, "drink": 112, "drl": 112, "drop": [5, 6, 112], "drop_inst": [5, 6], "drop_invalid": [5, 6], "dropout": [66, 94], "ds_config": [4, 13, 14, 15, 16, 106, 107], "dtype": [4, 15, 23, 66], "due": [94, 103, 107], "dummi": [3, 4, 35], "dummy_featur": 23, "dure": [4, 66, 103, 113, 114], "dv": 94, "dv_ptr": 94, "dynam": [26, 66], "e": [4, 55, 57, 66, 88, 94, 105, 106, 112, 115], "each": [4, 13, 55, 61, 66, 69, 105, 108, 112, 113], "earth": 113, "easi": [103, 108, 112, 113], "easier": 66, "easili": [106, 113, 115], "east": 103, "econom": 103, "edit": [112, 113], "educ": 112, "effect": [103, 112, 113, 115], "effici": [1, 103, 109, 112, 115], "effortlessli": 103, "eg": 108, "egg": 112, "either": [66, 105, 112], "el": 103, "elabor": [112, 113], "element": [13, 60, 66, 69, 103, 115], "eleuth": [103, 106], "eleutherai": [103, 106, 112], "elif": 114, "elimin": 113, "ellipsi": 90, "els": [66, 111, 112, 113, 114], "embed": [15, 22], "emerg": 103, "emphas": 115, "employe": 112, "empti": [75, 105], "empty_no_special_token": 105, "empty_no_special_tokens_templ": [75, 81], "empty_templ": [75, 81], "emptyformatt": 75, "en": [66, 103], "en_multi_round_chat": 103, "en_singe_round_chat": 103, "enabl": [4, 115], "enable_decode_inference_result": [4, 69], "enable_distributed_infer": [4, 13, 16, 58, 69], "enchilada": 103, "encod": [13, 14, 52, 75, 81, 105, 114], "encode_convers": [75, 79, 80, 81], "encode_imag": 23, "encoded_pair": [75, 81], "encoder_attention_mask": 90, "encoder_decoder_model": [3, 14, 17], "encoder_hidden_st": 90, "encoderdecodermodel": [12, 14], "encourag": [105, 112], "end": [4, 13, 16, 37, 66, 105, 108, 114], "end_header_id": [105, 114], "end_of_turn": [105, 114], "end_str": 55, "endfor": 114, "endgam": 103, "endif": 114, "endoftext": [105, 114], "endpoint": 66, "endswith": 111, "enemi": 103, "engin": [103, 105], "england": 112, "english": 103, "enhanc": 115, "enjoi": [107, 112], "enough": [4, 113], "ensur": 37, "entail": 103, "enter": 103, "entir": [113, 115], "entri": 66, "enum": 96, "environ": 112, "eo": [4, 75, 105, 108], "eos_pad": 4, "eos_token": [75, 108, 114], "eos_token_id": 69, "eot_id": [105, 114], "ep": [25, 26, 27, 28, 29, 30, 31, 32, 33, 36, 37, 38, 39, 41, 43, 45], "epoch": [66, 103, 112, 113], "eq": 37, "equal": [60, 61, 66, 103], "equat": 37, "err": 106, "error": [66, 113], "escap": 103, "especi": 103, "essai": 103, "essenti": 103, "estim": [66, 113], "etc": [13, 14, 16, 21, 65, 66, 103, 105, 112, 115], "ethic": 103, "ethiopian": 103, "ethnic": 103, "evacu": 103, "eval": [32, 42, 61, 66, 112, 113], "eval_bleu": 66, "eval_dataset": [50, 51, 61, 66, 113], "eval_dataset_path": 4, "eval_pr": 68, "eval_step": [4, 112], "evalloopoutput": [61, 66], "evalpredict": 66, "evalu": [3, 4, 32, 42, 54, 66, 104, 105, 107, 112, 113, 115], "evaluate_block_s": 4, "evaluation_loop": 66, "evaluator_arg": 52, "evaluatorargu": [4, 52], "even": [103, 105, 113, 114], "even_headdim": 94, "even_m": 94, "even_n": 94, "event": 103, "eventu": 112, "ever": 112, "everett": 112, "everi": [57, 66, 105, 108], "everyon": 103, "exact": [106, 112], "exactli": [112, 113], "examin": 112, "exampl": [4, 7, 11, 12, 61, 66, 70, 71, 75, 80, 81, 88, 94, 103, 105, 106, 107, 108, 109, 114, 115], "example_dataset": 105, "excel": 112, "except": 115, "exchang": 112, "excit": 115, "exclud": 37, "execut": [66, 106], "exist": [66, 105, 108, 114], "exp_avg": 33, "exp_avg_diff": 33, "exp_avg_sq": 33, "expans": 115, "expect": [66, 103], "expens": 103, "experi": [66, 103, 112, 115], "experiment": [94, 103], "expert": 115, "explain": 112, "explan": 103, "explanation_in_prompt": 4, "exploit": 112, "explor": [112, 113], "explos": 103, "export": [11, 12], "extend": 37, "extens": [1, 112, 115], "extern": [66, 103], "extinct": 113, "extra": 112, "extra_id_0": [105, 114], "extra_id_1": [105, 114], "extract": [4, 88, 112], "extrem": 115, "face": [5, 6], "facebook": 109, "facebookresearch": 107, "facilit": 112, "factor": [103, 112], "factual": 103, "fail": 103, "falafel": 103, "fals": [4, 7, 13, 14, 15, 16, 22, 23, 25, 26, 31, 33, 36, 37, 39, 43, 44, 50, 51, 55, 58, 60, 61, 66, 68, 69, 75, 81, 89, 90, 91, 93, 94, 96, 101, 106, 112, 113, 114], "famou": 103, "far": 112, "fashion": [37, 66], "fast": 4, "faster": [33, 94, 103], "fate": 103, "featur": [60, 66, 67, 103], "feature_select": 23, "fed": [13, 14, 66], "feder": 103, "feed": [57, 66, 105], "feedback": [112, 113], "feel": [108, 112], "ferret": 103, "few": [103, 112, 115], "field": [4, 73, 105, 112, 113, 115], "field_prefix": 73, "fifth": 103, "fight": 112, "figur": [103, 112], "file": [4, 5, 6, 52, 66, 88, 105, 108, 110, 111, 112, 115], "file_nam": 88, "file_path": [5, 6], "fill": [103, 105, 114], "filter": [103, 113], "final": [65, 103], "final_lr": 26, "financi": 103, "find": [103, 112, 113], "fine": [7, 13, 14, 66, 112, 113, 115], "finetun": [1, 3, 54, 59, 103, 105, 107, 108, 115], "finetune_part": 4, "finetuned_from": 66, "finetuned_galactica_lora": 109, "finetuned_gpt2": 109, "finetuned_llama2_7b": 109, "finetuned_llama2_7b_lisa": 109, "finetuned_llama2_7b_lora": 109, "finetuned_llama3_8b": 109, "finetuner_arg": [53, 59], "finetunerargu": [4, 53], "fingerprint": [5, 6], "finish": [66, 106, 113], "fire": [103, 112], "firefight": 112, "fireplac": 112, "first": [1, 60, 66, 75, 80, 81, 103, 106, 107, 112, 113, 114], "fish": 103, "fit": [37, 103], "fix": [4, 94], "fixed_decai": 25, "flag": 22, "flame": 103, "flammabl": 103, "flash_attent": [3, 95], "flash_attn_func": 94, "flash_attn_kvpacked_func": 94, "flash_attn_qkvpacked_func": 94, "flashattent": 94, "flashattnfunc": 94, "flashattnkvpackedfunc": 94, "flashattnqkvpackedfunc": 94, "flatbread": 103, "flatten_list": 58, "flavor": 103, "flexibl": 108, "float": [4, 5, 6, 15, 16, 21, 26, 31, 33, 34, 36, 37, 41, 43, 45, 51, 52, 55, 58, 61, 66, 88, 112], "float32": 66, "float_onli": 5, "float_only_dataset_descript": 74, "floating_point_op": 66, "floattensor": [22, 61, 90], "florida": 112, "flour": 112, "folder": 65, "follow": [37, 50, 66, 105, 106, 108, 112, 113, 115], "foo": [11, 12], "food": [103, 112], "force_system": [75, 81], "foreach": [32, 33, 42], "foreign": 112, "forget": 103, "fork": 106, "form": [53, 66, 112], "format": [5, 6, 50, 55, 73, 75, 81, 108, 110, 114], "formatt": [75, 81], "former": [112, 113], "fort": [112, 113], "fortun": [112, 113], "forward": [22, 23, 66, 89, 90, 91, 93, 94, 100], "found": [66, 103, 112, 113, 115], "foundat": [103, 112, 115], "four": [105, 113], "fp16": 4, "framework": [1, 104, 112, 113], "free": [32, 42, 108], "frequent": 75, "fresh": 103, "fri": [103, 112], "friend": 103, "friendli": 115, "from": [4, 5, 6, 13, 14, 23, 25, 37, 44, 51, 52, 55, 66, 73, 88, 94, 103, 107, 108, 111, 112, 113, 114, 115], "from_dict": [5, 6], "from_pretrain": 66, "front": 7, "frustrat": 112, "fsdp": 66, "ft": 4, "fuel": 112, "full": [13, 14, 16, 65, 103, 110, 112, 113, 115], "fun": [112, 113], "function": [11, 12, 21, 22, 57, 66, 112, 113, 115], "function_formatt": [75, 81], "functionbar": [11, 12], "funtion": 88, "furnitur": [112, 113], "further": [4, 103, 112, 115], "futur": [15, 22, 112], "g": [4, 37, 66, 105, 112], "g_": 37, "ga": 103, "gain": [103, 112, 115], "galactica": [109, 115], "game": 112, "gamma": [26, 55], "gap": 103, "gather": [66, 112], "gear": 112, "gemma": [3, 81, 105], "gemma_templ": [79, 81], "gemmaconversationtempl": 79, "gener": [3, 4, 7, 13, 14, 16, 20, 22, 52, 55, 57, 66, 69, 88, 94, 103, 107, 110, 112, 113, 114, 115], "generate_during_ev": 61, "generate_kwarg": 22, "generate_qrcod": [105, 114], "generation_kwarg": 57, "genet": 115, "georg": 103, "get": [9, 21, 57, 66, 103, 106, 107, 110, 111], "get_backend": [5, 6], "get_backend_dataset": [5, 6], "get_backend_model": [14, 15, 22], "get_batch_loss_metr": 61, "get_batch_metr": 61, "get_data_arg": [5, 6], "get_eval_dataload": 66, "get_fingerprint": [5, 6], "get_imag": 96, "get_linear_schedule_with_warmup": 66, "get_max_length": [14, 15, 111], "get_model": 9, "get_optimizer_cls_and_kwarg": 66, "get_paired_dataset": 50, "get_peft_without_qlora": 13, "get_pipelin": [46, 111], "get_pipeline_args_class": [4, 111], "get_prompt": 96, "get_python_vers": 101, "get_reward_funct": 112, "get_test_dataload": 66, "get_token": [14, 15, 22], "get_train_dataload": 66, "get_typ": [5, 6], "giant": 113, "girl": 112, "girlfriend": 112, "git": [66, 106, 115], "github": [15, 103, 106, 112, 115], "gitignor": 66, "give": [103, 112, 113], "given": [4, 5, 6, 21, 52, 53, 55, 57, 58, 61, 66, 103, 105, 106, 112, 113, 114], "glad": [105, 112], "global": [66, 112], "gmask": [105, 114], "go": [66, 103, 107, 112], "goal": [112, 115], "golden": 112, "good": [103, 105, 107, 108, 112, 113], "goodi": 112, "gopher": 115, "govern": 103, "gpt": [103, 109, 112, 113, 115], "gpt2": [109, 112], "gpt2_flash_attent": [3, 92], "gpt3": 115, "gpt4_en_instruct": 103, "gpt_neo_flash_attent": [3, 92], "gpu": [4, 13, 14, 15, 16, 66, 69, 109, 112], "grad": [31, 33, 43], "grad_averag": 39, "grade": 112, "gradient": [25, 26, 37], "gradient_accumulation_step": 4, "gradient_checkpoint": 4, "gradient_checkpointing_use_reentr": 4, "grammar": 113, "great": [112, 113], "greater": [66, 103], "greatest": 112, "green": 103, "grill": 103, "groundtruth": 52, "group": [34, 53, 66, 111], "group_rank": 74, "group_text": [53, 111], "group_texts_batch_s": 4, "grow": 115, "guarante": [75, 80, 81, 115], "guess": [112, 113], "guest": 112, "guid": [110, 112, 113], "ha": [7, 13, 14, 26, 31, 36, 41, 43, 45, 52, 66, 75, 94, 103, 105, 108], "had": [108, 112, 113], "half": 112, "hallucin": 103, "hamburg": 103, "hand": [103, 112, 115], "handel": [70, 71], "handl": [60, 66, 75, 81, 112], "hanz": [0, 115], "happi": [103, 112], "har": [103, 106], "hard": [103, 112, 113], "hardship": 103, "hardwar": 112, "harm": 112, "has_imag": 7, "has_placehold": 75, "hasattr": [32, 42], "hasn": 66, "hate": 112, "have": [4, 15, 66, 94, 103, 105, 106, 107, 108, 112, 113, 114, 115], "hawaii": 103, "hazard": 112, "he": 112, "head": [94, 112], "head_mask": [89, 90, 91], "headdim": 94, "health": 103, "hear": 105, "heard": 112, "hearti": 103, "heat": [103, 112], "heater": 112, "heavi": 112, "heavili": 112, "height": [22, 103], "held": [103, 115], "hellaswag": 103, "hello": [13, 75, 80, 81, 115], "help": [4, 103, 105, 112, 114, 115], "helper": [10, 11, 12, 15, 19, 20, 22, 48, 66, 69, 75], "her": 112, "here": [52, 66, 71, 75, 80, 81, 103, 109, 112, 113], "hero": 112, "hf": [15, 105, 107, 109], "hf_auto_model": 15, "hf_auto_model_additional_arg": 15, "hf_automodel_map": 15, "hf_automodel_typ": 15, "hf_dataset_sanity_check": [5, 6], "hf_decoder_model": [3, 17, 51, 55, 56, 69, 72], "hf_encoder_decoder_model": [3, 17], "hf_model_config": 15, "hf_model_mixin": [3, 13, 16, 17], "hf_text_regression_model": [3, 17, 56, 58, 59, 72], "hfargumentpars": 111, "hfdecodermodel": [13, 14, 51, 52, 55, 56, 69], "hfencoderdecodermodel": 14, "hfmodelmixin": [13, 15, 16], "hftextregressionmodel": [16, 56, 58, 59], "hh": [112, 113], "hh_rlhf": [112, 113], "hh_rlhf_llama": 112, "hh_rlhf_raft_align": 112, "hh_rlhf_rm": 112, "hh_rlhf_rm_sft_gptneo_2_7b": 112, "hh_rlhf_rm_train": 112, "hh_rlhf_sft": 112, "hi": [75, 80, 81, 105, 112], "hidden": 108, "hidden_s": [22, 23], "hidden_st": [89, 90, 91, 93], "hide": [66, 112, 113], "hideo": 112, "high": 112, "higher": [55, 103, 112], "highlight": 115, "him": 112, "hing": 61, "hint": [4, 5, 6], "histor": 112, "histori": [96, 103, 112], "hit": 112, "hm": 112, "home": [103, 112, 113], "homeown": 103, "homework": 112, "homogen": 4, "honei": 103, "host": 112, "hostil": 103, "hot": [103, 112], "hous": [103, 112, 113], "how": [66, 103, 105, 108, 110, 112, 113, 114], "howev": [13, 14, 103, 107, 112, 115], "howpublish": 115, "hp": 66, "hp_name": 66, "hp_search_backend": 66, "hp_space": 66, "hpsearchbackend": 66, "html": [66, 103, 112], "http": [26, 31, 33, 36, 41, 43, 45, 55, 66, 103, 106, 112, 113, 115], "hub": [66, 103], "hub_model_id": 66, "hug": [5, 6, 105], "huge": [103, 113], "huggingfac": [4, 5, 6, 13, 14, 16, 107, 112, 113], "human": [103, 105, 106, 108, 112, 113, 114, 115], "hundr": 113, "hurrican": 112, "hymba": [3, 81, 105], "hymba_templ": [80, 81], "hymbaconversationtempl": 80, "hyper": [66, 110], "hyperparamet": 66, "hyperparameter_search": 66, "i": [4, 7, 13, 14, 15, 22, 32, 37, 42, 47, 49, 50, 52, 55, 57, 60, 66, 69, 75, 81, 94, 103, 105, 107, 108, 109, 112, 113, 114, 115], "ianz2020": 113, "icml": 103, "id": [4, 13, 14, 22, 69, 75, 81], "idea": [103, 112, 113], "ident": 66, "identifi": [66, 105, 113], "idx_gap": 113, "ignor": [61, 66], "ignore_bias_buff": 4, "ignore_index": 74, "ignore_kei": 66, "ignore_keys_for_ev": 66, "ignored_args_list": 73, "ill": 103, "illeg": 103, "illustr": [108, 113], "im_end": [74, 105, 114], "im_patch": 74, "im_start": [74, 105, 114], "imag": [7, 22, 23, 74], "image_aspect_ratio": 4, "image_encoder_name_or_path": [4, 22], "image_flag": 88, "image_fold": [4, 7], "image_forward_out": [22, 23], "image_processor": 7, "image_text": [5, 55], "image_token_index": [7, 22, 74], "imageher": 88, "imagin": [112, 113], "immedi": 103, "impact": [112, 113], "imperfect": 112, "implement": [13, 14, 15, 25, 26, 31, 33, 36, 41, 43, 45, 51, 66, 94, 103, 106, 112], "impli": 103, "implicitli": 61, "import": [4, 52, 66, 103, 108, 110, 111, 112, 115], "improp": 115, "improv": [103, 112, 113, 115], "incent": 112, "incentiv": 112, "incident": 115, "includ": [4, 5, 6, 51, 66, 88, 103, 112, 113, 115], "incompat": 66, "incomplet": 103, "inconveni": 103, "incorpor": 115, "incorrect": 103, "incorrectli": 112, "increas": [31, 43, 112, 113], "inde": 103, "index": [22, 66, 115], "index0": 114, "indian": 103, "indic": [4, 5, 6, 22, 103, 105], "indirect": 115, "induc": 103, "inf": 57, "infer": [1, 4, 13, 14, 16, 21, 22, 52, 55, 58, 69, 71, 105, 112, 115], "infer_batch_s": 57, "inferenc": [3, 4, 54, 58, 105], "inference_arg": 69, "inference_batch_s": 4, "inference_batch_size_per_devic": [4, 112], "inference_func": 21, "inference_result": 16, "inferencer_arg": [55, 58, 69], "inferencer_file_path": 69, "inferencerargu": [4, 55, 56, 58, 69], "inferencerwithoffload": 69, "influenc": 112, "info": 106, "inform": [4, 5, 6, 52, 66, 112, 115], "inher": 103, "inherit": [10, 11, 12, 15, 19, 20, 22, 48, 66, 69, 75, 108], "init": 66, "init_git_repo": 66, "init_to_zero": 94, "initi": [4, 5, 6, 13, 14, 16, 21, 52, 53, 55, 57, 58, 59, 66, 111, 112, 113], "initial_accumul": 45, "initial_iter_idx": 4, "inject": 66, "injera": 103, "injur": [103, 112], "injuri": 103, "injustic": 103, "inner": 66, "innov": 115, "inplac": 4, "input": [4, 7, 13, 14, 15, 16, 21, 22, 52, 55, 57, 61, 66, 68, 69, 71, 88, 103, 105, 106, 107, 108, 113], "input_dataset": 55, "input_dir": 107, "input_id": [7, 13, 22, 23, 55, 71, 113], "input_shap": [89, 90, 93], "inputs_emb": [22, 90, 93], "insert": 22, "inspir": 112, "inst": [105, 114], "instal": [66, 106], "instanc": [5, 6, 7, 13, 14, 16, 21, 51, 52, 53, 66, 73, 103, 105, 112, 113, 115], "instance_fields_map": 74, "instanti": 66, "instead": [4, 75, 94, 103], "institut": 103, "instruct": [1, 113], "instructgpt": [112, 113, 115], "int": [4, 5, 6, 13, 14, 15, 16, 51, 55, 58, 60, 61, 66, 67, 69, 70, 71, 75, 80, 81, 83, 87, 88, 89, 96, 112, 113], "int8": 4, "integ": [4, 75, 81], "integr": 106, "intellig": 115, "intend": 115, "intens": [103, 112], "interact": [108, 112], "interest": [112, 115], "interfac": [3, 13, 14, 16, 17], "intermedi": 65, "intern": [66, 103], "internal_vers": 8, "internlm": [3, 81], "internlm2": 105, "internlm2_templ": [81, 82], "interpol": 15, "interpret": 105, "interrupt": [65, 66], "intrins": 103, "introduc": [103, 112, 115], "introduct": 110, "invalid": 4, "invis": 108, "invok": 103, "involv": [103, 112, 113], "io": [66, 103, 112, 115], "ipex_optimize_model": 66, "irrelev": 103, "is_caus": 94, "is_custom_dataset": 4, "is_encoder_decod": [60, 61], "is_first_tim": 66, "is_flash_attn_avail": 101, "is_flask_avail": 101, "is_gradio_avail": 101, "is_in_train": 66, "is_load": 23, "is_local_process_zero": 66, "is_model_parallel": 66, "is_multimod": 4, "is_multimodal_avail": [6, 101], "is_package_version_at_least": 101, "is_ray_avail": 101, "is_sagemaker_mp_post_1_10": 66, "is_torch_greater_or_equal_than_1_10": 66, "is_torch_less_than_1_11": 66, "is_trl_avail": 101, "is_vllm_avail": 101, "is_world_process_zero": 66, "island": 103, "isn": 112, "issu": [4, 15, 103, 107, 112, 115], "item": [103, 106], "iter": [4, 34, 52, 55, 57, 66, 112], "iter_id": 57, "iterabledataset": 66, "iteration_nam": 56, "iterative_dpo_align": [3, 54], "iterativealignerargu": 4, "iterativedpoalign": 56, "iterativedpoalignerargu": [4, 56], "its": [46, 66, 103, 112, 115], "itself": 103, "j": [52, 112], "jack": 112, "japanes": 103, "jfk": 112, "jinja": 114, "jipeng": [0, 115], "job": 108, "join": 115, "joint": 103, "journal": 115, "journei": 112, "json": [4, 5, 6, 66, 75, 80, 81, 88, 105, 106, 107, 111, 112, 113], "json_fil": 111, "jsonl": 115, "judg": 112, "juic": 103, "just": [7, 15, 32, 42, 60, 94, 103, 105, 108, 112, 113, 115], "k": [57, 94, 103, 112], "ka": 0, "ka9v1ywd": 113, "kashun": 115, "katrina": 112, "kebab": 103, "keep": [96, 112, 113], "keep_end": [60, 61], "keep_linebreak": 4, "kei": [66, 91, 103, 105, 106, 112, 113], "keith": 112, "kennedi": 112, "key_1": [5, 6, 105], "key_2": [5, 6, 105], "key_3": 105, "key_4": 105, "key_inst": 5, "key_scor": 5, "key_typ": 5, "keyword": [5, 6, 13, 14, 16, 21, 53, 57, 59, 66, 75], "kid": [112, 113], "kind": [112, 113], "king": 103, "kitchen": 112, "kitfo": 103, "kl": 61, "knive": 112, "know": [112, 113], "knowledg": [103, 115], "known": 103, "kojima": 112, "konami": 112, "korean": 103, "kv": 94, "kwarg": [5, 6, 9, 10, 11, 12, 13, 14, 15, 16, 20, 21, 23, 24, 46, 47, 49, 53, 56, 57, 58, 59, 65, 66, 69, 75, 79, 80, 81, 83, 87], "kwd": 96, "l": 94, "label": [7, 22, 23, 60, 66, 71, 103, 112, 113], "label_column": [70, 71], "label_id": 66, "label_nam": 66, "label_pad_token_id": [60, 61], "lamb": [3, 4, 35], "lambda": 113, "lambdalr": [61, 66], "land": 113, "landmark": 103, "languag": [4, 13, 14, 44, 52, 53, 66, 94, 103, 109, 115], "language_model": [22, 23], "language_model_from_pretrain": 22, "language_model_input": 22, "language_model_name_or_path": 22, "language_project": [4, 23], "lar": [3, 4, 35], "larg": [36, 37, 103, 112, 115], "lars_lr": 37, "last": [66, 105, 112, 113, 114], "last_checkpoint": 53, "latest": [66, 109], "laugh": 113, "launch": 66, "layer": [4, 37, 66, 109], "layer_past": [89, 90, 91], "layerwis": 110, "lead": [13, 16, 103, 112], "leader": 112, "leadership": 112, "leak": 103, "learn": [26, 32, 34, 36, 41, 42, 66, 112, 113, 115], "learning_r": [4, 112], "leav": [11, 12, 103, 112], "led": 112, "left": [94, 103], "legal": [103, 115], "len": [88, 103, 111, 113], "len_penalti": 61, "length": [4, 14, 15, 22, 51, 52, 53, 55, 60, 66, 88, 94, 103, 105], "length_penalti": [4, 51], "less": 103, "let": [111, 112, 113, 115], "level": [103, 115], "li": 115, "liabil": 115, "librari": [4, 66], "licens": [66, 103, 112], "light": 115, "lighter": 103, "lightn": 37, "lightweight": 115, "like": [37, 66, 94, 103, 105, 106, 108, 112, 114, 115], "likelihood": [52, 103, 106], "lime": 103, "limit": 103, "line": [11, 12, 103], "link": 105, "lisa": [110, 113], "lisa_activated_lay": [4, 109], "lisa_interval_step": [4, 109], "lisa_layers_attribut": 4, "list": [4, 5, 6, 13, 14, 16, 22, 33, 50, 51, 56, 58, 60, 61, 66, 67, 69, 71, 73, 75, 80, 81, 83, 87, 88, 96, 97, 101, 105, 108, 112, 115], "list_of_list": 58, "list_to_compress": 58, "listformatt": 75, "liter": [61, 75], "littl": 103, "liuhong99": 44, "live": [112, 113], "liyuanlucasliu": 41, "ll": [94, 112, 113], "llama": [3, 15, 81, 103, 105, 108, 109, 110, 112, 113, 115], "llama2": [105, 109], "llama2_templ": [81, 83], "llama2_template_for_tool": 83, "llama2conversationtempl": 83, "llama2conversationtemplatefortool": 83, "llama3": [105, 109], "llama3_templ": [81, 83], "llama3_template_for_tool": [81, 83], "llama_2": 96, "llama_flash_attent": [3, 92], "llama_rope_scaled_monkey_patch": [3, 99], "llava": 23, "llava_conversation_lib": [3, 95], "llava_load": 4, "llava_pretrain_model_path": 4, "llm": [4, 104, 108, 112, 115], "llm_model_name_or_path": 4, "lm": [4, 103, 110], "lm_dataset": 111, "lm_eval_dataset_map": 106, "lm_evaluation_metr": 4, "lmflow": [3, 104, 105, 108, 110, 111, 112, 113, 114], "lmflow_lora_target_modules_map": 74, "lmsy": 103, "ln": 52, "lntwmcyd": 113, "lo": 103, "load": [4, 5, 6, 13, 14, 22, 50, 52, 53, 55, 57, 58, 59, 66, 88, 107], "load_data": 88, "load_dataset": [112, 113], "load_in_4bit": 4, "load_inference_result": 69, "load_llava_pretrain_model": 98, "load_model": 23, "load_prompt_cach": 22, "load_result": 66, "loader": 52, "local": [66, 103, 112], "local_datset_answertype_map": 106, "local_datset_group_map": 106, "local_datset_map": 106, "local_rank": [4, 52, 55, 57, 58, 74], "locat": [103, 108], "lock": 103, "locksmith": 103, "log": [52, 61, 66, 103, 106], "log_dir": 106, "log_freq": 4, "logdir": 74, "logger": [4, 5, 13, 14, 15, 16, 51, 53, 55, 56, 57, 58, 59, 60, 61, 63, 64, 66, 67, 69, 70, 71, 73, 75, 79, 83, 87, 97, 101], "logging_step": 4, "logit": 66, "logsigmoid": 113, "long": [60, 66, 103, 105, 112], "longer": [51, 103], "longtensor": [22, 61, 93], "look": [108, 112], "loop": [66, 114], "loop_messag": 114, "lora": [1, 13, 14, 110, 113, 115], "lora_alpha": 4, "lora_dropout": 4, "lora_merg": 109, "lora_model_path": [4, 107, 109], "lora_r": 4, "lora_target_modul": 4, "lora_target_modules_map": 15, "loss": [25, 26, 31, 32, 34, 36, 37, 41, 42, 43, 45, 61, 66, 113], "loss_typ": [4, 61], "lot": [112, 113], "low": [110, 113], "low_resourc": [4, 22], "lower": [55, 66, 103], "loyal": 103, "lr": [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45], "lr_schedul": [61, 66], "lr_scheduler_typ": 4, "lse": 94, "lst": 66, "luck": 103, "luolc": 26, "m": [94, 103, 105, 112, 113], "macaroni": 103, "machin": [66, 115], "made": [66, 103, 112, 113], "magic": 112, "magnitud": 103, "mai": [11, 12, 66, 103, 105, 106, 112, 113], "main": [4, 63, 64, 66, 106, 111, 112, 115], "main_process_first": 111, "mainstream": 112, "maintain": 115, "major": 103, "make": [55, 66, 75, 81, 94, 103, 105, 106, 108, 112, 113, 115], "make_shell_args_from_dataclass": 73, "malfunct": 112, "manag": 66, "mani": [94, 103, 112, 113], "manipul": [5, 6], "manual": [66, 103], "map": [4, 5, 6, 21, 111, 113], "mar": 2, "marathon": 113, "margarita": 103, "margin": 61, "margin_scal": [4, 51], "marin": 103, "maryland": 103, "mask": [4, 13, 14, 22, 71, 75, 94], "mask_prompt": [4, 60, 61], "master": 103, "match": [5, 6, 103], "materi": 112, "math": [37, 103], "mathemat": 115, "matter": 115, "max": [14, 15, 52], "max_eval_sampl": 4, "max_grad_norm": 33, "max_length": [4, 60, 61, 67], "max_max": 51, "max_min": 51, "max_new_token": [4, 55], "max_position_embed": 100, "max_prompt_length": [4, 60, 61], "max_random": [4, 51], "max_seq_len_cach": 100, "max_step": 4, "max_target_length": [60, 61], "max_train_sampl": 4, "maxim": [44, 66, 115], "maximum": [4, 52, 53, 55, 66], "mayb": [112, 113], "me": [112, 113], "mean": [66, 103, 112, 113], "meaning": 112, "meantim": 103, "measur": 113, "meat": [103, 112], "medic": 115, "medicin": 115, "medmcqa": 115, "medqa": 115, "melt": 112, "member": [66, 112], "memor": 103, "memori": [4, 109, 112], "memory_safe_dpov2_align": [3, 62], "memory_safe_dpov2_align_env_var_to_remov": 74, "memory_safe_vllm_infer": [3, 62], "memory_safe_vllm_inference_don": 74, "memory_safe_vllm_inference_env_var_to_remov": 74, "memory_safe_vllm_inference_finish_flag": 74, "memorysafedpov2align": 51, "memorysafevllminferenc": 69, "mention": 103, "menu": [103, 112], "merg": 110, "merge_lora": 112, "merge_lora_weight": [13, 14], "messag": [4, 5, 6, 66, 73, 75, 80, 81, 83, 87, 96, 105, 108, 113, 114], "meta": [105, 109], "metadata": [4, 66, 112], "metal": 112, "method": [5, 6, 13, 14, 15, 26, 45, 52, 66, 75, 103, 112, 115], "metric": [4, 52, 61, 65, 66, 106, 112], "metric_key_prefix": 66, "mexican": 103, "middl": 103, "might": [66, 112, 113], "mild": 103, "milk": 112, "million": 113, "min_vers": 101, "mini": 52, "mini_gpt": 55, "miniatur": 112, "minim": [66, 75], "minut": 36, "misc": 115, "misfortun": 103, "mistak": [103, 112], "mix": [4, 112], "mixed_arg": 56, "mixed_precis": 4, "mizuzu": 112, "mkdir": 106, "mlir": 94, "mmlu": 115, "modal": 7, "mode": [4, 32, 42, 66, 112], "model": [1, 3, 4, 7, 8, 25, 26, 31, 32, 33, 34, 36, 37, 41, 42, 43, 44, 45, 47, 49, 50, 51, 52, 53, 55, 56, 57, 58, 59, 60, 61, 65, 66, 68, 69, 95, 98, 103, 105, 107, 108, 109, 110, 111, 114, 115], "model_arg": [9, 13, 14, 15, 16, 21, 46, 50, 51, 52, 53, 55, 56, 57, 58, 59, 69, 98, 111], "model_args_list": 97, "model_config_class": 4, "model_for_causal_lm_map": 4, "model_init": [61, 66], "model_input": [58, 69], "model_max_length": [4, 53, 70, 71, 111], "model_nam": [66, 106], "model_name_or_path": [4, 105, 106, 107, 108, 109, 112], "model_output": 58, "model_revis": 4, "model_s": 107, "model_typ": 4, "model_wrap": 66, "modelargu": [4, 15, 16, 51, 52, 53, 55, 56, 57, 58, 59, 63, 69, 97, 111], "modelconfig": 15, "modeling_output": [16, 22, 58], "modeling_util": 66, "modern": 103, "modif": 66, "modifi": [25, 37, 112, 113], "mole": 103, "momentum": [31, 33, 37, 42, 43], "momentum_decai": 38, "monei": 103, "monitor": 103, "monument": 103, "more": [55, 66, 103, 105, 109, 112, 113, 114, 115], "moreov": [113, 115], "morgan": 103, "most": [66, 103, 105, 112, 113], "mostli": [75, 105], "motiv": 112, "mpi4pi": [106, 115], "mpt": 96, "mu": 37, "much": 94, "multi": [7, 103], "multi_modal_dataset": [3, 6], "multimod": [3, 95], "multimodaldatasetargu": 4, "multipl": [22, 51, 94, 103, 112, 113, 114, 115], "museum": 103, "must": [66, 111, 114, 115], "mutipl": 88, "mv": 106, "my": [103, 112, 113], "n": [5, 6, 13, 52, 69, 103, 105, 106, 108, 113, 114, 115], "n_trial": 66, "nadam": [3, 4, 35], "name": [4, 13, 14, 16, 21, 32, 42, 46, 66, 73, 88, 94, 103, 105, 106, 108, 112, 114], "namedtupl": 66, "nassist": 114, "nation": [103, 112], "natur": [103, 115], "ndarrai": 66, "necessari": [4, 50, 66, 103, 108, 113], "need": [7, 32, 42, 65, 66, 71, 75, 103, 106, 107, 108, 112, 115], "neg": [52, 103, 106, 112, 113, 115], "negro": 103, "neighborhood": 103, "neither": 103, "neo": [112, 113], "nest": 66, "nestedtensor": 94, "nesterov": [31, 33, 37, 43], "network": [37, 112], "neurip": 25, "nevertheless": 103, "new": [5, 6, 66, 73, 94, 103, 106, 112, 113, 115], "new_default": 73, "next": [55, 108, 112], "nf4": 4, "nhead": 94, "nhello": 105, "nhow": 114, "ni": 114, "nice": 112, "nif": 113, "nip": 45, "nll": [52, 103, 110], "nlp": [13, 14, 103], "nn": [23, 34, 61, 66, 100, 113], "no_prox": 33, "nogada": 103, "nois": [112, 113], "noisi": 112, "non": [94, 112], "nonconvex": 45, "none": [4, 5, 6, 7, 13, 14, 15, 16, 21, 22, 23, 25, 26, 27, 28, 29, 30, 31, 32, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45, 50, 52, 53, 57, 59, 60, 61, 65, 66, 67, 69, 73, 75, 80, 81, 83, 87, 88, 89, 90, 91, 93, 94, 96, 100, 113], "nonsens": 103, "noqa": 45, "nor": 103, "norm": [31, 43], "normal": [13, 14, 16, 52, 103], "north": 113, "northern": 103, "notat": 112, "note": [4, 15, 26, 31, 36, 37, 41, 43, 45, 66, 71, 103, 106, 115], "noth": 34, "notic": 106, "novograd": [3, 4, 35], "now": [103, 105, 107, 112, 113, 114], "np": 66, "ntk_ratio": 100, "num_capt": 22, "num_channel": 22, "num_exampl": 66, "num_inst": [58, 69], "num_label": 15, "num_new_token": 55, "num_output_sequ": 4, "num_patch": 23, "num_proc": 50, "num_raft_iter": [4, 112], "num_sampl": 55, "num_train_epoch": 112, "num_training_step": 66, "number": [4, 5, 6, 52, 55, 66, 109, 112, 113], "numpi": [66, 88], "nuser": 114, "nwhat": 105, "nwho": 114, "nyou": [105, 114], "o": [94, 103, 111], "oaxaca": 103, "obei": 103, "obj": [75, 81], "object": [5, 6, 7, 13, 16, 21, 52, 53, 55, 57, 58, 59, 66, 69, 73, 75, 81, 105, 114], "obqa": 103, "observ": [15, 25, 103, 112, 113, 115], "observation_formatt": [75, 81], "obstacl": 103, "obtain": [103, 107, 112, 113, 115], "occasion": 112, "occup": 112, "occupi": 113, "ocean": 112, "off": 112, "offer": [103, 112], "offici": [15, 107, 114], "offs_d": 94, "offs_n": 94, "offset": 96, "often": [103, 112], "oil": 112, "okai": 112, "old": [105, 108, 114], "omit": 103, "omp_num_thread": 74, "on_epoch_end": 65, "on_sav": 65, "on_train_end": 65, "onc": [66, 94], "one": [11, 12, 51, 66, 103, 105, 106, 109, 111, 112, 113], "one_sample_multiple_imag": 22, "ones": 113, "onli": [4, 21, 66, 94, 103, 105, 106, 108, 109, 111, 112, 113, 114, 115], "onlin": 103, "open": [104, 106, 115], "openai": 94, "oper": [4, 66], "opportun": 112, "oppress": 103, "opt": [103, 115], "optim": [3, 8, 61, 65, 66, 105, 112, 115], "optim_adam_beta1": 4, "optim_adam_beta2": 4, "optim_beta1": 4, "optim_beta2": 4, "optim_beta3": 4, "optim_dummy_beta1": 4, "optim_dummy_beta2": 4, "optim_momentum": 4, "optim_weight_decai": 4, "optimalscal": [103, 112, 115], "optimizer_kera": 45, "optimizer_nam": 66, "optimizer_typ": 4, "optimizernam": 4, "option": [4, 5, 6, 11, 12, 13, 14, 15, 16, 21, 22, 25, 32, 34, 42, 53, 55, 57, 59, 66, 69, 73, 94, 103, 105, 107, 108, 112], "optuna": 66, "order": [44, 103, 105], "org": [26, 31, 33, 36, 41, 43, 55, 103, 112, 113], "organ": [103, 112, 113], "orient": 115, "origin": [66, 73, 75, 103, 107, 112, 113], "original_dataclass": 73, "other": [52, 61, 66, 94, 103, 112, 113, 115], "otherwis": 66, "otherworld": 112, "ouptut": 103, "our": [15, 103, 105, 106, 107, 108, 109, 110, 111, 112, 115], "out": [66, 94, 103, 108, 112, 113, 115], "outag": 112, "outcom": 115, "output": [4, 13, 14, 16, 51, 52, 55, 57, 66, 69, 71, 88, 103, 105], "output_attent": [22, 89, 90, 91, 93], "output_dataset": 55, "output_dir": [4, 56, 66, 107, 112], "output_hidden_st": 22, "output_lora_path": 109, "output_max_length": [4, 57], "output_min_length": [4, 57], "output_model": [107, 108, 109, 112], "output_model_path": [108, 109], "output_reward_path": [4, 57], "over": [52, 103, 105, 112, 113], "overal": [11, 12, 13, 14, 94, 115], "overfit": 113, "overrid": [4, 15, 22, 66], "overridden": 66, "overse": 112, "overview": [110, 113], "overwrite_cach": 4, "overwrite_output_dir": 66, "overwritten": 112, "own": [66, 103, 105, 106, 108, 112, 113, 115], "oyster": 103, "p": [31, 37, 43, 52, 55], "p2ju3r1a": 113, "p_": [37, 52], "pacif": 103, "packag": [52, 101, 110, 115], "package_nam": 101, "pad": [4, 22, 60, 66, 67], "pad_index": 66, "pad_to_multiple_of": 67, "pad_token_id": [70, 71], "padding_sid": [4, 70, 71], "padding_valu": [60, 61], "paddingstrategi": 67, "page": [3, 103, 115], "paged_adamw_32bit": 4, "pain": 112, "pair": [51, 75, 80, 81, 103, 106, 108, 110], "paired_convers": [5, 105], "paired_conversation_dataset_descript": 74, "paired_conversation_tokenize_funct": 71, "paired_text_to_text_dataset_descript": 74, "pan": [0, 115], "panda": 113, "paper": [37, 45, 112, 113], "par": 103, "parallel": [4, 66, 94], "param": [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45], "paramet": [4, 5, 6, 13, 14, 15, 16, 21, 32, 34, 37, 42, 52, 53, 55, 57, 58, 59, 61, 66, 69, 73, 75, 81, 88, 105, 110, 113, 114, 115], "parent": 112, "pars": 111, "parse_args_into_dataclass": 111, "parse_json_fil": 111, "parse_to_sampling_param": 69, "parser": 111, "part": [37, 103], "partial": 103, "particip": 115, "particular": [112, 113, 115], "partit": 103, "pass": [4, 15, 66, 94, 103, 105, 111, 115], "past_key_valu": [22, 23, 93], "past_key_values_length": [89, 90, 93], "pasta": 103, "pastor": 103, "path": [4, 5, 6, 13, 14, 16, 21, 22, 66, 106, 107, 111, 112], "path_to_dataset": 105, "path_to_your_model": 108, "patient": 112, "pattern": 66, "peac": 112, "peakier": 55, "peanut": 112, "pecan": 112, "peft": 65, "peft_config": [15, 61], "peft_train": [3, 62, 68], "peftrewardtrain": 68, "peftsavingcallback": 65, "pefttrain": [65, 68], "penal": [4, 51], "peopl": [103, 112, 115], "pepper": 103, "per": [4, 112], "per_device_eval_batch_s": 4, "per_device_train_batch_s": [4, 112], "percentag": 113, "perfect": 112, "perform": [5, 6, 13, 14, 16, 22, 25, 26, 31, 32, 33, 34, 36, 37, 41, 42, 43, 45, 52, 53, 55, 57, 58, 59, 66, 69, 109, 112, 115], "period": [11, 12, 103], "permiss": 103, "permit": 103, "perplex": [4, 103], "persecut": [103, 112], "person": 112, "perturb": [31, 43], "pet": 103, "pete": 103, "phantom": 112, "phenomenon": [103, 112], "phi": [3, 81], "phi3": 105, "phi3_templ": [81, 84], "phil": 94, "phone": 112, "pi_ratio": 100, "pick": 66, "pie": 103, "piec": [105, 108, 114], "pile": [112, 113], "pioneer": 103, "pip": [106, 115], "pipelin": [3, 4, 8, 105, 111, 112, 115], "pipeline_arg": [14, 46, 111], "pipeline_argument_map": 4, "pipeline_map": 46, "pipeline_nam": [46, 111], "pipeline_needs_extra": 46, "pipelineargu": 111, "piqa": 103, "pit": 103, "pivot": 103, "pixel_valu": 22, "place": [32, 42, 66, 112, 113], "place_model_on_devic": 66, "placehold": 15, "plai": 103, "plain": [4, 88, 96], "plan": [94, 112], "plantain": 112, "platform": 115, "plausibl": 103, "pleas": [41, 75, 105, 106, 108, 109, 112, 113, 114, 115], "plot": 112, "poblano": 103, "pocket": 112, "point": [66, 94, 112], "polic": 103, "polici": [4, 61, 103, 112], "policy_chosen_logp": 61, "policy_rejected_logp": 61, "polish": 23, "poor": 112, "pop": 66, "pop_callback": 66, "popish": 112, "popul": [103, 113], "popular": [103, 112], "pork": 103, "portion": 103, "posit": [5, 6, 13, 14, 15, 16, 21, 53, 57, 59, 103, 112, 113, 115], "position_id": 93, "position_interpol": [3, 95], "possess": 103, "possibl": [103, 113], "post": [112, 113], "post_process_pair": [75, 81], "postprocess_distributed_inference_output": 16, "postprocess_inference_output": 16, "potenti": [66, 103], "powder": 112, "power": [112, 115], "ppl": 103, "ppo": 112, "practic": 103, "pre": 44, "pre_grad": 33, "precis": 4, "predict": [55, 66, 103, 108], "predict_next_token": 55, "predicted_answ": 52, "prediction_loop": 66, "prediction_loss_onli": 66, "prediction_step": 66, "predictionoutput": 66, "prefer": [105, 112, 113], "preferencedatacollatorwithpad": 60, "prefix": [66, 73, 113], "prefix_checkpoint_dir": 66, "prelat": 112, "prepar": [13, 15, 51, 57, 66, 106, 112, 113, 115], "prepare_inputs_for_infer": [13, 16], "prepare_inputs_labels_for_multimod": 23, "prepend": 66, "preprocess": [66, 113], "preprocess_llama_from_llava_plain": 7, "preprocess_llama_from_llava_v1": 7, "preprocess_logits_for_metr": [61, 66], "preprocess_multimodal_llava": 7, "preprocessing_num_work": [4, 61], "present": [66, 112], "preset": [15, 105, 108, 114], "preset_templ": [81, 108], "presid": 103, "pretrain": [4, 13, 14, 66, 103, 107, 115], "pretrained_language_projection_path": 4, "pretrained_path": 22, "pretrainedmodel": [60, 61, 66], "pretrainedtoken": [7, 15, 70, 71, 75, 80, 81, 83, 87], "pretrainedtokenizerbas": [60, 61, 66], "pretrainedtokenizerfast": [15, 70, 71], "pretti": 113, "previou": [66, 112, 113], "price": 103, "print_bann": 73, "print_change_log": 25, "priorit": 103, "privat": 4, "privileg": 112, "prob": 55, "probabilist": 115, "probabl": [55, 61, 66, 103, 112, 113], "problem": [103, 110, 112], "procedur": [103, 112, 113], "proceed": 103, "process": [7, 13, 14, 16, 22, 52, 53, 55, 57, 66, 103, 105, 110, 111, 112, 113, 115], "process_image_flag": 88, "processor_image_token_in_minigpt4": 22, "produc": 103, "product": [112, 115], "profession": 115, "profici": 115, "program": [11, 12, 88, 112], "progress": 66, "prohibit": 103, "project": [15, 103, 112, 113], "project_dir": 112, "prompt": [4, 7, 13, 14, 16, 22, 50, 55, 57, 60, 69, 75, 105, 112, 113, 115], "prompt_cache_path": 4, "prompt_id": 22, "prompt_keys_valu": 22, "prompt_structur": [4, 55, 106, 107], "proper": 103, "properli": [103, 112], "properti": [23, 103, 105, 114], "propon": 112, "proport": [4, 5, 6], "propos": [26, 31, 33, 36, 41, 43, 45], "protect": 112, "provid": [4, 10, 11, 12, 13, 14, 15, 19, 20, 22, 41, 48, 52, 55, 66, 69, 75, 103, 105, 106, 108, 110, 112, 114, 115], "pt": [66, 67], "public": [1, 112], "publicli": 115, "publish": 115, "pubmedqa": 115, "pull": 103, "punctuat": 113, "purpos": 108, "purs": 112, "push": 66, "push_to_hub": 66, "put": [7, 106], "py": [94, 105, 106, 107, 108, 112, 113], "python": [5, 6, 7, 75, 103, 106, 107, 115], "pytorch": [25, 33, 36, 37, 60, 66], "pytorch_": 41, "pytorchlightn": 37, "q": [94, 103], "q1": 103, "q2": 103, "q3": 103, "qa": [88, 103], "qformer_from_pretrain": 22, "qformer_name_or_path": [4, 22], "qkv": 94, "qr": [105, 114], "qualit": 103, "qualiti": [103, 112], "quant_config": 15, "quant_typ": 4, "quantit": 103, "quantiti": 66, "quantiz": 4, "queri": 91, "quesadilla": 103, "queso": 103, "question": [13, 14, 50, 57, 103, 105, 113, 115], "quick": 112, "quicker": 4, "quickli": [103, 115], "quicksort": 103, "quit": 94, "qwen": [3, 81, 109], "qwen1": 109, "qwen2": 105, "qwen2_templ": [81, 85], "qwen2_template_for_tool": [81, 85], "r": [32, 42], "r1": 103, "r2": 103, "r3": 103, "rabi": 103, "race": 94, "radam": [3, 4, 35], "raft": [4, 61, 110], "raft_align": [3, 4, 54, 112], "raft_aligner_arg": 57, "raft_batch_s": [4, 112], "raft_train": [3, 62], "raftalign": 57, "raftalignerargu": [4, 57], "rafttrain": 66, "raggedtensor": 94, "rai": [13, 16, 58, 66, 69], "rais": [5, 6, 66], "raise_except": 114, "random": [51, 55, 66, 88, 112], "random_se": 4, "random_shuffl": [4, 88], "randomli": [51, 109, 112], "rang": [61, 103, 113, 115], "rank": [74, 110, 112, 113], "rare": 113, "rarest": 113, "rate": [26, 32, 34, 41, 42, 66, 112, 113], "rather": [32, 42], "raw": [103, 105], "re": [60, 66, 112, 113], "read": [112, 113], "reader": 112, "readi": 112, "readm": 115, "readthedoc": 66, "real": [103, 105], "realli": [112, 113], "reason": [66, 112], "recal": 108, "receiv": 66, "recent": 103, "recip": 112, "recommend": [103, 112], "record": [112, 113], "rectifi": 25, "recurs": [66, 103], "reduc": [94, 112], "redund": 112, "reevalu": [25, 26, 31, 32, 34, 36, 37, 41, 42, 43, 45], "ref": 55, "ref_model": [51, 61], "ref_model_arg": [51, 56], "refer": [25, 26, 31, 36, 37, 41, 43, 45, 61, 66, 71, 105, 106, 108, 110, 112, 114, 115], "reference_chosen_logp": 61, "reference_fre": 61, "reference_rejected_logp": 61, "referencemodelargu": [51, 63], "reflect": [66, 103], "reform": 112, "reformul": 112, "regard": 103, "regist": [15, 21, 94], "register_inference_funct": 21, "register_prompt_cach": 22, "register_token": 7, "registr": 110, "regress": [20, 21], "regression_model": [3, 17, 21], "regressionmodel": [20, 21, 57], "reinforc": [112, 113], "reiniti": 66, "reject": [50, 51, 60, 61, 105, 112, 113], "rejected_attention_mask": 113, "rejected_input_id": 113, "rejected_reward": [61, 113], "rel": 113, "relat": [52, 112, 115], "releas": [13, 15, 16, 69, 112, 115], "release_gpu": [13, 16, 69], "relev": 115, "reli": 115, "reliabl": [103, 112, 115], "relianc": 115, "religion": 112, "relleno": 103, "reload": 66, "remain": [51, 103], "remark": [112, 113, 115], "rememb": [75, 106, 112], "remind": 112, "remov": [66, 73, 112], "remove_callback": 66, "remove_dataclass_attr_prefix": 73, "remove_image_flag": 55, "remove_last_sep": [75, 81], "remove_last_separ": [75, 81], "remove_unused_column": 4, "repetit": 4, "repetition_penalti": 4, "replac": [15, 75, 105, 112], "replace_bloom_attn_with_flash_attn": 89, "replace_gpt2_attn_with_flash_attn": 90, "replace_gpt_neo_attn_with_flash_attn": 91, "replace_llama_attn_with_flash_attn": 93, "replace_llama_with_condens": 100, "repli": 112, "repo": [66, 106, 112], "report_to": 4, "repositori": [66, 103, 115], "repres": [4, 5, 6, 13, 14, 16, 52, 75], "requir": [5, 6, 32, 42, 52, 53, 55, 57, 58, 59, 103, 105, 108, 112, 113, 114, 115], "require_vers": 4, "rerun": 66, "rescu": 112, "research": [103, 112], "reset": 25, "residu": 89, "resolv": 112, "resourc": [13, 15, 16, 69, 112], "respect": [32, 37, 42, 61, 112], "respond": 108, "respons": [60, 61, 66, 88, 103, 112, 113, 115], "rest": [11, 12], "restart": 112, "restart_opt": 33, "restaur": 103, "restor": 112, "result": [4, 21, 69, 103, 112, 113, 115], "result_cache_path": 56, "results_path": [4, 69], "resum": 66, "resume_from_checkpoint": 66, "retrain": 112, "retriev": [5, 6], "return": [4, 5, 6, 7, 13, 14, 16, 22, 25, 26, 31, 32, 34, 36, 37, 41, 42, 43, 45, 46, 52, 53, 55, 57, 61, 66, 69, 73, 75, 81, 88, 103, 112, 113], "return_code_error_buff": 74, "return_dict": 22, "return_output": [66, 68], "return_pil": 96, "return_tensor": [7, 67], "reus": [22, 66], "rev_kl": 61, "review": 112, "revis": [13, 14, 16, 21], "reward": [4, 15, 51, 57, 61, 105, 110], "reward_model": [47, 50, 57], "reward_model_arg": 56, "reward_model_inference_batch_s": 4, "reward_model_inference_block_s": 4, "reward_model_or_path": 112, "rewarddatacollatorwithpad": 67, "rewardmodelinferenc": 58, "rewardmodelinferenceresultwithinput": [16, 58, 88], "rewardmodeltun": 59, "rewardmodeltunerargu": [4, 59], "rewardtrain": 68, "rewrit": 113, "rho": [27, 44], "rib": 103, "rich": 103, "right": [4, 66, 70, 71, 103, 112, 114], "risk": 115, "rl": 112, "rlhf": [112, 113], "rlhf_prompt": 112, "rm": [71, 112, 113], "rm_dataprocessor": [3, 62], "rm_inferenc": [3, 54], "rm_loss": 68, "rm_trainer": [3, 62], "rm_tuner": [3, 54], "rng": 66, "robin": 103, "robust": [94, 115], "role": [75, 80, 81, 96, 103, 105, 108, 112, 113, 114], "role_nam": 74, "role_rank": 74, "room": [112, 113], "root": 112, "rope_ntk_ratio": 4, "rope_pi_ratio": 4, "roughli": 103, "round": [103, 105, 112, 114], "rstrip_partial_utf8": 55, "rui": [0, 115], "run": [4, 52, 53, 57, 66, 103, 105, 106, 107, 112, 113], "run_evaluation_with_lora": 107, "run_finetun": [105, 108, 109, 112], "run_finetune_with_lisa": 109, "run_finetune_with_lora": [109, 112], "run_merge_lora": 109, "run_nam": 4, "run_raft_align": 112, "run_reward_model": [112, 113], "run_summari": 66, "rwandan": 112, "safe": [66, 103, 112], "safeti": [103, 112], "sagemak": 66, "sai": [108, 112, 113], "said": [108, 112], "salam": 112, "salem": 112, "salmonellosi": 103, "salt": [103, 112], "same": [4, 66, 105, 112, 113], "samoa": 112, "sampl": [4, 5, 6, 13, 51, 52, 55, 66, 69, 103, 105, 110, 112, 113], "sample_dataset": [5, 6], "sample_input_1": 105, "sample_input_2": 105, "sample_input_3": 105, "sample_output_1": 105, "sample_output_2": 105, "sample_output_3": 105, "sample_text_1": 105, "sample_text_2": 105, "sample_text_3": 105, "sampler": 66, "sampling_paired_idx_from_reward": 51, "sampling_paired_method": [4, 51], "sampling_param": [13, 16, 69], "samplingparam": [13, 16, 69], "saniti": [5, 6], "sanity_check": [4, 5, 6, 50], "sauc": [103, 112], "save": [4, 5, 6, 13, 14, 16, 22, 32, 42, 65, 66, 69, 107, 112, 115], "save_aggregated_lora": 4, "save_count": 66, "save_file_path": 69, "save_full_model": [13, 14], "save_inference_result": 69, "save_language_project": 4, "save_model": 66, "save_pretrain_model_path": 4, "save_prompt_cach": 22, "save_result": [4, 69], "save_step": 4, "scalabl": [44, 103, 115], "scalar": 4, "scale": 37, "scaler": 66, "scaler_nam": 66, "scenario": 103, "scene": 103, "schedul": [32, 42, 66], "scheduler_nam": 66, "scienc": 112, "scope": 113, "score": [5, 16, 51, 55], "score_to_prob": 55, "scout": 112, "scratch": 4, "screen": 112, "script": [4, 66, 105, 107, 108, 109, 111, 112, 113], "seafood": 103, "search": [4, 66, 112, 115], "second": [44, 51, 66, 75, 80, 81, 107, 113], "section": [55, 103, 112], "see": [15, 66, 103, 108, 112, 113], "seed": [4, 5, 6, 66, 88], "seek": 115, "seem": [94, 112], "select": [51, 113], "select_featur": 23, "select_lay": 23, "selected_dataset": 57, "self": [5, 6, 66, 89, 90, 91, 93, 94, 112], "sens": [103, 105, 108, 114], "sentenc": [4, 13, 16, 103, 105, 113, 114], "sep": 96, "sep2": 96, "sep_styl": [4, 96], "separ": [75, 81, 96, 106], "separatorstyl": 96, "seq_len": 100, "seqlen": 94, "seqlen_k": 94, "seqlen_q": 94, "seqlen_q_round": 94, "sequenc": [4, 13, 14, 16, 22, 51, 66, 75, 80, 81, 83, 87, 94, 103], "sequence_length": 22, "sequence_parallel": 94, "sequenceclassifieroutputwithpast": [16, 58], "seri": 112, "serv": [66, 103, 112, 115], "servic": [103, 108], "session": [66, 108], "set": [4, 5, 6, 13, 14, 32, 37, 42, 66, 69, 75, 88, 103, 105, 107, 108, 110, 112, 113, 114], "set_epoch": 66, "set_random_se": 88, "setback": 103, "setup": [66, 110, 112], "seventeenth": 112, "sever": [4, 13, 14, 66, 88, 103, 105, 106, 110], "sft": [105, 110], "sgd": [37, 42], "sgd_schedule_fre": [3, 4, 35], "sgdp": [3, 4, 35], "sgdschedulefre": 42, "sh": [105, 106, 107, 108, 109, 112, 113], "shall": [105, 115], "shape": [22, 61, 94], "sharded_ddp": 66, "share": [66, 112, 115], "sharegpt": [4, 103, 105], "shaw": 103, "she": 112, "sheet": 103, "shell": [73, 105], "shizh": [0, 115], "shop": 103, "should": [4, 11, 12, 13, 15, 16, 23, 32, 42, 66, 75, 80, 81, 94, 103, 105, 112], "show": [103, 110, 112, 113], "shrimp": 103, "shuffl": [5, 6, 88], "shum": [0, 115], "shun": 0, "sick": 112, "side": [4, 103, 112], "sigmoid": [4, 61], "sign_sgd": 4, "signific": [112, 113, 115], "sigopt": 66, "simclr": 37, "similar": [103, 107, 108, 112], "similarli": 106, "simpl": [1, 66, 103, 108, 112, 113, 115], "simplest": [112, 113], "simpli": [106, 112], "simplifi": [1, 52, 53, 55, 57, 113, 115], "sinc": [75, 103, 105, 108, 112], "singl": [13, 25, 26, 31, 32, 33, 34, 36, 37, 41, 42, 43, 45, 60, 66, 96, 103, 105, 109, 112, 113], "situat": [66, 103, 112], "sixteenth": 112, "size": [4, 22, 52, 66, 94, 103, 112, 113], "skill": 115, "skillet": 112, "skip": [73, 112], "skip_default": 73, "skip_first_batch": 66, "skip_next": 96, "skippabl": 101, "slight": [103, 113], "slightli": [94, 107, 114], "slow": [31, 43, 103], "slower": 94, "small": [94, 103, 113], "smaller": [103, 112], "smell": 103, "smoke": 103, "smokehous": 103, "smoother": 108, "smoothli": 112, "so": [7, 15, 66, 103, 105, 108, 112, 113, 114, 115], "social": 103, "societi": 112, "soda": 112, "softmax": 55, "softmax_scal": 94, "softwar": [105, 108, 112, 114], "sole": 115, "solid": 112, "solv": 103, "some": [4, 15, 52, 66, 103, 106, 112, 113, 115], "someon": 112, "someth": [61, 66, 103, 112, 113], "sometim": [103, 112], "soon": [105, 114], "sop": [105, 114], "sophia": [3, 4, 35], "sophiag": 44, "sorri": [103, 105], "sort": [103, 112], "soul": 103, "sourc": [4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 55, 56, 57, 58, 59, 60, 61, 63, 64, 65, 66, 67, 68, 69, 70, 71, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 93, 94, 96, 97, 98, 100, 101, 102, 104, 105, 112, 115], "source_dataset": 51, "sourdough": 103, "southern": 103, "sovereign": 103, "space": [66, 105, 112], "special": [13, 15, 16, 75, 105, 108, 112, 114, 115], "special_start": [75, 81, 108], "special_stopp": [75, 81, 108], "specif": [4, 103, 105, 112, 115], "specifi": [105, 106, 108, 109, 112], "speculativeinferenc": 55, "speed": 94, "speedi": 115, "spell": 113, "spend": 112, "sphinx": 3, "spice": 103, "spici": [103, 112], "spill": 94, "split": [5, 6, 66, 112, 113], "split_arg": 4, "spot": 103, "src": [108, 112], "stabl": 66, "stage": [60, 103], "stai": [103, 105, 109, 112], "stand": 103, "standard": [10, 11, 12, 15, 19, 20, 22, 37, 48, 69, 75], "start": [52, 66, 94, 105, 112, 113], "start_header_id": [105, 114], "start_n": 94, "start_of_turn": [105, 114], "starter": 108, "state": [25, 26, 30, 33, 37, 38, 39, 41, 44, 65, 66, 103, 112], "state_dict": [66, 98], "static": [16, 31, 43, 55, 66, 94], "step": [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45, 66, 106, 108, 110, 112], "stepsiz": 25, "stew": 103, "stext": 112, "still": [60, 66, 94, 103, 112, 113, 115], "stochast": 44, "storag": 115, "store": [4, 94, 112], "store_flo": 66, "storm": 112, "str": [4, 5, 6, 7, 13, 14, 15, 16, 50, 51, 55, 56, 58, 60, 61, 66, 67, 69, 70, 71, 73, 75, 80, 81, 83, 87, 88, 96, 101, 112], "strand": 112, "strategi": [13, 14, 51, 112], "stream": 4, "stream_infer": 55, "streamlin": 115, "street": 103, "strictli": 103, "stride_bb": 94, "stride_bh": 94, "stride_bm": 94, "stride_dkb": 94, "stride_dkh": 94, "stride_dkn": 94, "stride_dob": 94, "stride_doh": 94, "stride_dom": 94, "stride_dqb": 94, "stride_dqh": 94, "stride_dqm": 94, "stride_dvb": 94, "stride_dvh": 94, "stride_dvn": 94, "stride_kb": 94, "stride_kh": 94, "stride_kn": 94, "stride_ob": 94, "stride_oh": 94, "stride_om": 94, "stride_qb": 94, "stride_qh": 94, "stride_qm": 94, "stride_vb": 94, "stride_vh": 94, "stride_vn": 94, "string": [4, 5, 6, 13, 14, 16, 22, 55, 66, 69, 73, 75, 88, 105, 108, 114], "stringformatt": [75, 108], "strip": [112, 114], "strong": 112, "stronger": 112, "strongest": 112, "strongli": 105, "structur": [5, 6, 50, 112, 113], "struggl": 103, "studi": [66, 112], "studio": 112, "stuf": 103, "stuff": [112, 113], "sturdier": [112, 113], "style": [96, 103], "sub": 103, "subarrai": 103, "subclass": [47, 49, 66], "subject": 115, "sublist_length": 58, "submit": 115, "subprocess": 73, "subsect": 112, "subtract": 51, "succe": 112, "success": 113, "suffer": 103, "sugar": 112, "suggest": [32, 42, 103, 112], "sum": [60, 66], "sum_": 52, "summar": 103, "summari": [11, 12, 66], "superior": 112, "superstit": 112, "supervis": [7, 110], "supervisor": 112, "support": [13, 14, 55, 75, 94, 106, 107, 110, 113], "supported_dataset_typ": 55, "suprem": 103, "sure": [75, 81, 94, 103, 105, 112, 113], "surfac": 113, "surpass": [103, 115], "surpris": 103, "surprisingli": 113, "surround": 112, "suspect": 112, "switch": [66, 112], "sy": [105, 111, 114], "synonym": 103, "sysinfo1": [75, 80, 81], "system": [75, 80, 81, 83, 87, 96, 105, 108, 114], "system_formatt": [75, 81, 108], "system_messag": 114, "system_propmt": 105, "t": [7, 15, 37, 60, 65, 66, 71, 94, 103, 105, 108, 112, 113, 114], "t3uwm8yp": 113, "tabl": [103, 106, 112], "taco": 103, "tag": [4, 66], "take": [13, 14, 52, 57, 66, 112], "talk": 112, "taller": [112, 113], "target": [7, 55, 66, 115], "target_cl": 56, "target_model_arg": [55, 56], "task": [1, 13, 14, 66, 103, 110, 112], "task_1": 106, "task_2": 106, "task_combin": 106, "task_guid": 103, "tast": 112, "teach": [112, 113, 115], "team": [105, 108, 114], "teaspoon": 112, "technic": 115, "techniqu": [105, 115], "technologi": 112, "tee": 106, "tell": [108, 112, 113], "temperatur": [4, 55, 61, 112], "templat": [4, 69, 75, 81], "template_nam": [75, 81, 108], "templatecompon": [75, 81, 108], "tend": [103, 112], "tensor": [4, 13, 16, 22, 33, 55, 60, 61, 66, 89, 90, 93], "tensor_parallel_s": 4, "term": [14, 15, 103, 112], "termin": [11, 12], "test": [5, 6, 52, 61, 66, 94, 103, 105, 107, 109, 112, 113, 115], "test_13": 105, "test_bleu": 66, "test_dataset": [5, 6, 66], "test_fil": 4, "test_flash_attn": 94, "test_flash_attn_triton_race_condit": 94, "test_siz": [5, 6], "text": [4, 7, 13, 14, 37, 52, 53, 55, 57, 88, 103, 105, 107, 108, 111, 112, 113, 114, 115], "text2text": [5, 70, 71, 106, 110], "text2text_dataset_descript": 74, "text2text_dataset_detail": 74, "text2text_dataset_long_descrit": 74, "text_onli": [5, 21, 55, 70, 71, 105, 106, 112], "text_only_dataset_descript": 74, "text_only_dataset_detail": 74, "text_only_dataset_long_descrit": 74, "text_regression_model": [3, 16, 17], "text_to_scored_textlist": 51, "text_to_scored_textlist_dataset_descript": 74, "text_to_textlist_dataset_descript": 74, "text_to_textlist_tokenize_funct": 71, "textonli": 110, "textregressionmodel": [16, 21], "th": 52, "than": [32, 42, 94, 103, 109, 112], "thank": 112, "thankfulli": 112, "thei": [4, 13, 16, 66, 103, 105, 108, 112, 113], "them": [66, 103, 105, 106, 107, 108, 112, 113, 115], "therefor": [112, 115], "theses": 112, "thi": [3, 4, 5, 6, 7, 11, 12, 13, 14, 15, 16, 32, 37, 42, 57, 60, 66, 75, 88, 94, 103, 105, 106, 108, 109, 112, 113, 114, 115], "thing": 112, "think": [112, 113], "thoma": 112, "thoroughli": 115, "those": [66, 103, 112], "thoughtfulli": 115, "three": [52, 61, 103, 105, 112], "threshold": 55, "thrive": 103, "through": [66, 103], "thu": [15, 103, 108], "tillet": 94, "time": [66, 103, 105, 112, 113], "tip": [66, 105], "titl": 115, "tmp": [4, 94], "tmp_trainer": 66, "to_dict": [5, 6], "to_gradio_chatbot": 96, "to_list": [5, 6], "todai": [108, 113], "todo": 111, "togeth": [53, 66, 112, 115], "tojson": 114, "tok_logg": [70, 71], "token": [3, 4, 7, 8, 13, 14, 15, 16, 22, 50, 52, 55, 57, 60, 61, 66, 67, 69, 75, 80, 81, 83, 87, 103, 105, 108, 111, 112, 113, 114], "token_dict": [70, 71], "token_id": 75, "token_per_step": 55, "token_type_id": 13, "tokenization_utils_bas": 66, "tokenize_batch_el": 60, "tokenize_funct": [70, 71], "tokenized_column_ord": [70, 71], "tokenized_dataset": [13, 14, 16, 53, 111], "tokenized_neg": 113, "tokenized_po": 113, "tokenizer_image_token": 7, "tokenizer_nam": 4, "told": 113, "tong": [0, 115], "too": [60, 103, 113], "tool": [75, 80, 81, 83, 87, 105, 108, 114, 115], "tool_1_desc": [75, 80, 81], "tool_description_1": 105, "tool_description_2": 105, "tool_description_3": 105, "tool_description_x": 105, "tool_info": 114, "toolbox": 115, "toolinferenc": 55, "toolkit": [103, 112, 115], "tools_formatt": [75, 81], "top": [4, 55, 103, 112, 113], "top_k": 4, "top_p": [4, 55], "top_reward_percentag": [4, 112], "torch": [7, 13, 15, 22, 23, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45, 55, 57, 61, 66, 88, 89, 90, 93, 94, 100], "torch_dtyp": [4, 15], "torch_jit_model_ev": 66, "tortilla": 103, "total": [103, 113], "touch": 112, "toxoplasmosi": 103, "tr_loss": 66, "track": [66, 105, 112], "tradeoff": 109, "tradit": [103, 105, 108, 112, 114], "trail": [13, 16], "train": [4, 5, 6, 13, 14, 32, 36, 37, 42, 44, 51, 53, 57, 59, 61, 65, 66, 88, 103, 105, 106, 109, 112, 113, 114, 115], "train_50": 105, "train_convers": 109, "train_dataset": [5, 6, 50, 51, 61, 66, 113], "train_ev": 61, "train_fil": 4, "train_on_prompt": 4, "train_test_split": [5, 6], "trainer": [57, 65, 66, 68], "trainer_callback": [61, 65, 66], "trainer_st": 66, "trainer_state_nam": 66, "trainer_util": [61, 66], "trainercallback": [61, 65, 66], "trainercontrol": 65, "trainerst": 65, "training_arg": [57, 65, 66], "training_args_nam": 66, "training_step": 66, "training_util": 66, "trainingargu": [4, 51, 61, 65, 66], "transform": [4, 7, 13, 14, 15, 16, 22, 51, 58, 60, 61, 65, 66, 67, 68, 70, 71, 75, 80, 81, 83, 87, 111], "transform_dataset_in_plac": [51, 53, 58, 59], "tri": 112, "trial": [65, 66, 103, 112], "tribul": 103, "trick": 108, "trim": [105, 114], "triton": 94, "triton_flash_attent": [3, 92], "trl": 61, "troubl": [103, 112], "truck": 103, "true": [4, 5, 6, 13, 14, 16, 25, 33, 51, 52, 53, 58, 59, 60, 61, 66, 67, 69, 73, 75, 88, 112, 113], "truncat": [4, 60, 113], "truncate_to_model_max_length": 4, "truncation_mod": [60, 61], "truncation_sid": [4, 70, 71], "trust_coeffici": 37, "trust_remote_cod": 4, "try": [75, 103, 112, 113], "tunabl": [3, 13, 14, 16, 18, 49], "tunable_model": 111, "tunablemodel": [52, 53, 55, 59, 111], "tune": [1, 7, 13, 14, 49, 53, 59, 66, 103, 105, 109, 111, 112, 113], "tune_strategi": [13, 14, 16], "tuned_model": 111, "tupl": [4, 22, 34, 51, 58, 61, 66, 75, 80, 81, 83, 87, 89, 90, 93, 101], "turn": [103, 112], "tweak": 66, "two": [51, 52, 66, 75, 96, 103, 106, 112, 113], "type": [4, 5, 6, 9, 21, 66, 75, 103, 105, 106, 108, 112, 113, 114, 115], "typeddict": 88, "typic": [11, 12, 61, 103, 112, 113], "u": [112, 115], "udpat": 22, "unabl": 103, "unclear": 113, "under": [4, 66, 103, 105, 106, 115], "understand": [103, 112, 115], "underw": 113, "unfair": 103, "unfinish": 112, "unfreez": 109, "uniform": 55, "union": [13, 16, 66, 73, 75], "unit": [75, 103, 112], "univers": 112, "unknown": 96, "unlock": 115, "unnecessari": 113, "unpack": 66, "unpleas": 103, "unseen": 115, "unsort": 103, "until": 112, "unus": 66, "unusu": 112, "up": [66, 94, 103, 112, 113], "upcom": 115, "updat": [66, 94, 105, 109, 112, 113, 114], "update_custom_config": 98, "update_hessian": 44, "upload": 66, "upscal": 103, "url": [66, 115], "us": [4, 5, 6, 10, 11, 12, 13, 14, 15, 16, 19, 20, 22, 32, 34, 41, 42, 48, 52, 55, 61, 66, 69, 73, 74, 75, 81, 88, 94, 103, 105, 106, 107, 109, 110, 112, 113, 114, 115], "usa": 103, "usag": [11, 12, 103], "use_acceler": [4, 13, 14, 15, 16], "use_accelerator_for_evalu": 4, "use_apex": 66, "use_beam_search": 4, "use_cach": [22, 89, 90, 91, 93], "use_cpu_amp": 66, "use_cuda_amp": 66, "use_customized_optim": 4, "use_dora": 4, "use_dpo_data_col": 61, "use_fast": 51, "use_fast_token": 4, "use_flash_attent": 4, "use_image_start_end": 4, "use_int8": 4, "use_lisa": 4, "use_lora": 4, "use_mtim": 66, "use_prompt_cach": [4, 22], "use_qlora": 4, "use_ram_optimized_load": [4, 106], "use_trunc": [70, 71], "use_tune_checkpoint": 66, "use_vllm": [4, 13, 15, 16, 58], "use_wandb": [4, 52], "user": [1, 75, 80, 81, 103, 105, 106, 108, 112, 113, 114, 115], "user_formatt": [75, 81, 108], "user_input_1": 105, "user_input_2": 105, "user_message_0": 114, "user_message_1": 114, "usernam": 106, "usml": 115, "usr_nam": 112, "usual": 103, "util": [3, 4, 7, 8, 13, 16, 54, 57, 58, 69, 70, 71, 103, 108, 113, 115], "v": [37, 51, 94, 108], "v_": 37, "valid": 66, "validation_fil": 4, "validation_split_percentag": [4, 112, 113], "valu": [4, 55, 66, 73, 75, 91, 106, 112], "value_1": [5, 6, 105], "value_2": [5, 6, 105], "value_3": 105, "varianc": 41, "variant": 33, "varieti": 103, "variou": [13, 14, 66, 105, 115], "ve": [94, 105, 112], "veget": [103, 112], "vegetarian": 103, "veloc": 37, "ventil": 103, "verbos": 52, "veri": [108, 112, 113], "verifi": [55, 115], "versa": 51, "version": [2, 3, 4, 8, 37, 41, 94, 95, 96, 103], "via": [4, 103, 112], "vibrant": 115, "vice": 51, "victorinox": 112, "vicuna": 103, "video": 112, "videogam": 112, "view_func": [31, 43], "vigil": 112, "virginia": 103, "viru": 103, "vision": 112, "vision2seq_model": [3, 17], "vision_encod": [3, 17], "vision_feature_select": 22, "vision_model_from_pretrain": 22, "vision_select_lay": 4, "vision_tow": 23, "vision_tower_cfg": [23, 24], "vision_tower_nam": 23, "visit": 103, "vismodelargu": 4, "visual": 112, "vllm": [4, 13, 15, 16, 69], "vllm_gpu_memory_util": [4, 15], "vllm_inferenc": [3, 54], "vllm_inference_batch_s": 4, "vllm_tensor_parallel_s": [4, 15], "vllminferenc": 69, "vllminferenceresultwithinput": [13, 69, 88], "vocabulari": 103, "w_": 52, "w_i": 52, "wa": [33, 103, 112, 113], "wai": [10, 11, 12, 15, 19, 20, 22, 48, 66, 69, 75, 103, 105, 106, 107, 112, 113], "wait": 103, "walnut": 103, "wandb": [4, 113], "want": [66, 106, 108, 112, 113], "warmup": [32, 42], "warmup_step": [4, 32, 42], "warn": 66, "washington": 103, "wat": 103, "watch": 66, "wd_ratio": [31, 43], "we": [51, 60, 61, 66, 75, 94, 103, 105, 106, 108, 110, 111, 112, 113, 114, 115], "weather": [105, 112], "wei": [0, 115], "weight": [4, 31, 37, 43, 52, 115], "weight_decai": [4, 25, 26, 28, 30, 31, 32, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45], "weight_decoupl": 25, "weight_lr_pow": [32, 42], "weixiong5237": 113, "welcom": [112, 115], "well": [66, 103, 105, 112, 115], "weqweasda": 112, "were": [112, 113], "what": [108, 112, 113], "when": [4, 13, 16, 32, 42, 51, 55, 66, 69, 73, 75, 103, 108, 112, 113], "whenev": 113, "where": [4, 5, 6, 13, 37, 52, 69, 103, 105, 112, 113, 115], "whether": [4, 7, 13, 14, 16, 66, 69, 73, 112, 115], "which": [4, 5, 6, 13, 14, 47, 49, 52, 60, 66, 75, 103, 105, 112, 113, 115], "while": [66, 75, 94, 103, 115], "white": 103, "who": [105, 108, 112, 114, 115], "whole": 115, "whose": 112, "why": [108, 112], "wide": [103, 115], "width": 22, "wild": 113, "willing": 112, "window": [103, 112], "window_length": 52, "winogrand": 103, "wipe": 66, "wise": 37, "wish": 66, "witch": 112, "witchcraft": 112, "with_deepspe": 14, "with_qform": [4, 22], "within": [55, 108], "without": [13, 16, 66, 103, 112, 113, 114], "won": 112, "word": [103, 108, 112, 113], "wordi": 113, "work": [66, 94, 103, 112, 113, 115], "worker_heart_beat_interv": 74, "workflow": 115, "workspace_path": 56, "world": [13, 103, 112, 113], "world_siz": [52, 55, 58], "wors": 112, "worst": 51, "would": [13, 69, 94, 103, 108, 112, 113], "wrap": 66, "wrapper": [13, 14, 66], "write": [52, 103, 113], "w\u00fcsthof": 112, "x": [31, 43, 100, 103, 112, 113], "xiong": [0, 115], "xiongwei": 112, "y": [31, 43, 106, 115], "ye": [103, 112, 113], "year": [103, 112, 113, 115], "yet": [13, 14, 60, 75, 94], "yi": [3, 81, 105], "yi1_5": 105, "yi1_5_templ": [81, 86], "yogi": [3, 4, 35], "you": [32, 42, 66, 75, 94, 103, 105, 106, 107, 108, 112, 113, 114, 115], "your": [66, 110, 112, 113, 114], "your_conversation_dataset": 108, "your_model": 108, "your_templ": 108, "your_template_fil": 108, "your_template_nam": 108, "yourself": 103, "zephyr": [3, 81, 105], "zephyr_templ": [81, 87], "zephyrconversationtempl": 87, "zero": 66, "zhang": [0, 115]}, "titles": ["Contributors", "Changelog", "About", "API Reference", "lmflow.args", "lmflow.datasets.dataset", "lmflow.datasets", "lmflow.datasets.multi_modal_dataset", "lmflow", "lmflow.models.auto_model", "lmflow.models.base_model", "lmflow.models.decoder_model", "lmflow.models.encoder_decoder_model", "lmflow.models.hf_decoder_model", "lmflow.models.hf_encoder_decoder_model", "lmflow.models.hf_model_mixin", "lmflow.models.hf_text_regression_model", "lmflow.models", "lmflow.models.interfaces", "lmflow.models.interfaces.tunable", "lmflow.models.regression_model", "lmflow.models.text_regression_model", "lmflow.models.vision2seq_model", "lmflow.models.vision_encoder.clip_encoder", "lmflow.models.vision_encoder", "lmflow.optim.adabelief", "lmflow.optim.adabound", "lmflow.optim.adadelta", "lmflow.optim.adagrad", "lmflow.optim.adam", "lmflow.optim.adamax", "lmflow.optim.adamp", "lmflow.optim.adamw_schedule_free", "lmflow.optim.adan", "lmflow.optim.dummy", "lmflow.optim", "lmflow.optim.lamb", "lmflow.optim.lars", "lmflow.optim.nadam", "lmflow.optim.novograd", "lmflow.optim.optimizers", "lmflow.optim.radam", "lmflow.optim.sgd_schedule_free", "lmflow.optim.sgdp", "lmflow.optim.sophia", "lmflow.optim.yogi", "lmflow.pipeline.auto_pipeline", "lmflow.pipeline.base_aligner", "lmflow.pipeline.base_pipeline", "lmflow.pipeline.base_tuner", "lmflow.pipeline.dpo_aligner", "lmflow.pipeline.dpov2_aligner", "lmflow.pipeline.evaluator", "lmflow.pipeline.finetuner", "lmflow.pipeline", "lmflow.pipeline.inferencer", "lmflow.pipeline.iterative_dpo_aligner", "lmflow.pipeline.raft_aligner", "lmflow.pipeline.rm_inferencer", "lmflow.pipeline.rm_tuner", "lmflow.pipeline.utils.dpov2_dataprocessor", "lmflow.pipeline.utils.dpov2_trainer", "lmflow.pipeline.utils", "lmflow.pipeline.utils.memory_safe_dpov2_align", "lmflow.pipeline.utils.memory_safe_vllm_inference", "lmflow.pipeline.utils.peft_trainer", "lmflow.pipeline.utils.raft_trainer", "lmflow.pipeline.utils.rm_dataprocessor", "lmflow.pipeline.utils.rm_trainer", "lmflow.pipeline.vllm_inferencer", "lmflow.tokenization.hf_decoder_model", "lmflow.tokenization.hf_text_regression_model", "lmflow.tokenization", "lmflow.utils.common", "lmflow.utils.constants", "lmflow.utils.conversation_template.base", "lmflow.utils.conversation_template.chatglm", "lmflow.utils.conversation_template.chatml", "lmflow.utils.conversation_template.deepseek", "lmflow.utils.conversation_template.gemma", "lmflow.utils.conversation_template.hymba", "lmflow.utils.conversation_template", "lmflow.utils.conversation_template.internlm", "lmflow.utils.conversation_template.llama", "lmflow.utils.conversation_template.phi", "lmflow.utils.conversation_template.qwen", "lmflow.utils.conversation_template.yi", "lmflow.utils.conversation_template.zephyr", "lmflow.utils.data_utils", "lmflow.utils.flash_attention.bloom_flash_attention", "lmflow.utils.flash_attention.gpt2_flash_attention", "lmflow.utils.flash_attention.gpt_neo_flash_attention", "lmflow.utils.flash_attention", "lmflow.utils.flash_attention.llama_flash_attention", "lmflow.utils.flash_attention.triton_flash_attention", "lmflow.utils", "lmflow.utils.llava_conversation_lib", "lmflow.utils.model", "lmflow.utils.multimodal", "lmflow.utils.position_interpolation", "lmflow.utils.position_interpolation.llama_rope_scaled_monkey_patch", "lmflow.utils.versioning", "lmflow.version", "LMFlow Benchmark: An Automatic Evaluation Framework for Open-Source LLMs", "Blogs", "Dataset", "LMFlow Benchmark Guide", "Checkpoints", "Customize Conversation Template", "Finetuning", "Examples", "Finetune", "RAFT", "Reward Modeling", "Supported Conversation Template", "LMFlow"], "titleterms": {"0": 1, "1": [1, 106, 108, 112, 113, 114], "2": [106, 108, 112, 113, 114], "2023": [1, 104], "28": 1, "3": [108, 112, 114], "4": 108, "5": [108, 114], "8x22b": 114, "8x7b": 114, "about": 2, "adabelief": 25, "adabound": 26, "adadelta": 27, "adagrad": 28, "adam": 29, "adamax": 30, "adamp": 31, "adamw": 109, "adamw_schedule_fre": 32, "adan": 33, "adapt": 109, "algorithm": 112, "align": 112, "an": 103, "api": 3, "arg": 4, "attribut": [4, 5, 8, 13, 14, 15, 16, 46, 51, 53, 55, 56, 57, 58, 59, 60, 61, 63, 64, 66, 67, 69, 70, 71, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 94, 96, 97, 101, 102], "auto_model": 9, "auto_pipelin": 46, "automat": 103, "base": 75, "base_align": 47, "base_model": 10, "base_pipelin": 48, "base_tun": 49, "benchmark": [103, 106], "blog": 104, "bloom_flash_attent": 89, "build": 108, "changelog": 1, "chat": 103, "chatglm": [76, 114], "chatml": [77, 114], "checkpoint": [107, 115], "choos": 108, "citat": 115, "class": [4, 5, 6, 7, 9, 10, 11, 12, 13, 14, 15, 16, 19, 20, 21, 22, 23, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 55, 56, 57, 58, 59, 60, 61, 65, 66, 67, 68, 69, 75, 79, 80, 81, 83, 87, 88, 94, 96, 100], "clip_encod": 23, "common": 73, "commonsens": 103, "conclus": 103, "constant": 74, "content": [4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 55, 56, 57, 58, 59, 60, 61, 63, 64, 65, 66, 67, 68, 69, 70, 71, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 93, 94, 96, 97, 98, 100, 101, 102, 115], "contributor": 0, "convers": [105, 108, 109, 114], "conversation_templ": [75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87], "creat": 106, "custom": [105, 108], "data": [105, 110], "data_util": 88, "dataset": [5, 6, 7, 105, 106, 112], "decoder_model": 11, "decompos": 108, "deepseek": [78, 114], "descript": 112, "detail": 105, "disclaim": 115, "dpo_align": 50, "dpov2_align": 51, "dpov2_dataprocessor": 60, "dpov2_train": 61, "dummi": 34, "encoder_decoder_model": 12, "end": 112, "evalu": [52, 103, 106, 110], "exampl": [110, 112, 113], "featur": 115, "file": 106, "finetun": [53, 109, 110, 111, 112, 113], "flash_attent": [89, 90, 91, 92, 93, 94], "follow": 103, "format": 105, "formatt": 108, "framework": 103, "full": 109, "function": [4, 6, 7, 23, 24, 33, 50, 55, 63, 64, 68, 70, 71, 73, 88, 89, 90, 91, 93, 94, 97, 98, 100, 101], "gemma": [79, 114], "gener": 105, "get": 112, "gpt2_flash_attent": 90, "gpt_neo_flash_attent": 91, "guid": 106, "hf_decoder_model": [13, 70], "hf_encoder_decoder_model": 14, "hf_model_mixin": 15, "hf_text_regression_model": [16, 71], "hymba": [80, 114], "hyper": 112, "import": 109, "indic": 115, "infer": 110, "inferenc": 55, "instal": 115, "instruct": [103, 115], "interfac": [18, 19], "internlm": 82, "internlm2": 114, "introduct": [103, 112, 113, 115], "iterative_dpo_align": 56, "lamb": 36, "lar": 37, "layerwis": 109, "lisa": 109, "llama": [83, 107, 114], "llama_flash_attent": 93, "llama_rope_scaled_monkey_patch": 100, "llava_conversation_lib": 96, "llm": 103, "lm": 106, "lmflow": [4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 106, 115], "lora": [109, 112], "low": 109, "mar": 1, "memory_safe_dpov2_align": 63, "memory_safe_vllm_infer": 64, "merg": [109, 112], "metric": 103, "mixtral": 114, "model": [9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 97, 112, 113], "modul": [4, 5, 7, 9, 10, 11, 12, 13, 14, 15, 16, 19, 20, 21, 22, 23, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 36, 37, 38, 39, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 55, 56, 57, 58, 59, 60, 61, 63, 64, 65, 66, 67, 68, 69, 70, 71, 73, 74, 75, 76, 77, 78, 79, 80, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 93, 94, 96, 97, 98, 100, 101, 102], "multi_modal_dataset": 7, "multimod": 98, "nadam": 38, "nll": 106, "note": 112, "notic": 114, "novograd": 39, "open": 103, "optim": [25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45], "overview": 112, "packag": [6, 8, 24, 81], "pair": 105, "paramet": [109, 112], "peft_train": 65, "perform": 103, "phi": [84, 114], "pipelin": [46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69], "position_interpol": [99, 100], "prepar": 110, "progress": [105, 114], "proper": 108, "qwen": [85, 114], "radam": 41, "raft": 112, "raft_align": 57, "raft_train": 66, "rank": 109, "refer": [3, 103], "regist": 108, "registr": 106, "regression_model": 20, "return": 15, "reward": [112, 113], "rm_dataprocessor": 67, "rm_inferenc": 58, "rm_trainer": 68, "rm_tuner": 59, "sampl": 109, "set": 106, "setup": 106, "sft": [112, 113], "sgd_schedule_fre": 42, "sgdp": 43, "sophia": 44, "sourc": 103, "step": 113, "submodul": [6, 8, 17, 18, 24, 35, 54, 62, 72, 81, 92, 95, 99], "supervis": [112, 113], "support": [105, 114, 115], "tabl": 115, "task": [106, 115], "templat": [105, 108, 109, 114], "text2text": 105, "text_regression_model": 21, "textonli": 105, "token": [70, 71, 72], "triton_flash_attent": 94, "tunabl": 19, "tune": 115, "us": 108, "util": [60, 61, 62, 63, 64, 65, 66, 67, 68, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101], "version": [1, 101, 102], "vision": 115, "vision2seq_model": 22, "vision_encod": [23, 24], "vllm_inferenc": 69, "weight": 109, "work": [105, 114], "yi": [86, 114], "yogi": 45, "your": [106, 108], "zephyr": [87, 114]}})
\ No newline at end of file